Open Collections will be undergoing maintenance Monday June 8th, 2020 11:00 – 13:00 PT. No downtime is expected, but site performance may be temporarily impacted.

Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Multifunctional pro-ligands as potential Alzheimer’s disease therapeutics Scott, Lauren 2009

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

Item Metadata

Download

Media
24-ubc_2009_fall_scott_lauren.pdf [ 2.15MB ]
Metadata
JSON: 24-1.0061398.json
JSON-LD: 24-1.0061398-ld.json
RDF/XML (Pretty): 24-1.0061398-rdf.xml
RDF/JSON: 24-1.0061398-rdf.json
Turtle: 24-1.0061398-turtle.txt
N-Triples: 24-1.0061398-rdf-ntriples.txt
Original Record: 24-1.0061398-source.json
Full Text
24-1.0061398-fulltext.txt
Citation
24-1.0061398.ris

Full Text

MULTIFUNCTIONAL PRO-LIGANDS AS POTENTIAL ALZHEIMER’S DISEASE THERAPEUTICS  by LAUREN SCOTT Hons. B.Sc., McMaster University, 2004  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY  in  THE FACULTY OF GRADUATE STUDIES (Chemistry)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) July 2009  © Lauren Scott, 2009 ii   ABSTRACT  Alzheimer’s disease is the most common form of dementia, affecting more than 24 million individuals worldwide. Although the exact causes of disease development and progression are unknown, the amyloid hypothesis links the observed pathologies of elevated metal ion levels (Cu2+, Fe3+, Zn2+), deposition of amyloid peptide in senile plaques, oxidative stress and neurodegeneration in a cohesive manner. As part of a possible intervention for this process, a series of multifunctional pyridinone pro-ligands were designed and synthesised. 3- Hydroxy-4-pyridinones display a high affinity for metal ions - particularly Fe3+ and Cu2+ - and are readily functionalised by variation of the N-substituent on the heterocyclic ring. The α- hydroxyketone functionality serves not only to bind metal ions, but as an antioxidant via phenolic hydrogen donation; in addition, these activities may be masked by glycosylation at the 3-hydroxy position. Seven pyridinone pro-ligands were synthesised, each containing a pyridinone moiety and a second aromatic ring. Five of these pro-ligands incorporate structural features of amyloid imaging agents: 2-methyl-3-hydroxy-1-(4-dimethylaminophenyl)-4(1H)-pyridinone (Hdapp), 2- methyl-3-hydroxy-1-(4-methylaminophenyl)-4(1H)-pyridinone (Hsapp), 1-(4-aminophenyl)-3- hydroxy-2-methyl-4(1H)-pyridinone (Hzapp), 1-(6-benzothiazolyl)-3-hydroxy-2-methyl-4(1H)- pyridinone (Hbt6p) and 1-(2-benzothiazolyl)-3-hydroxy-2-methyl-4(1H)-pyridinone (Hbt2p). The final two compounds, 3-hydroxy-2-methyl-1-phenyl-4(1H)-pyridinone (Hppp) and 1- benzyl-3-hydroxy-2-methyl-4(1H)-pyridinone (Hnbp), were synthesised and compared to probe the impact of linker length modification between the two aromatic rings. In addition to pro- ligand synthesis, their activities were assessed using a number of in vitro assays. Ability to iii   interfere with metal ion-induced amyloid peptide aggregation in solution, antioxidant activity, cytotoxicity, coordination of Cu2+ and binding to amyloid fibrils were all assayed on this series. This was done as a preliminary screen to identify promising lead compounds for further development. The compounds displayed marked ability to resolubilise metal ion-aggregated Aβ, excellent antioxidant activity comparable to that of α-tocopherol and acceptable cytotoxicity levels. Furthermore, the ligands coordinate Cu2+ in the bis, square planar, tetracoordinate fashion typical of 3-hydroxy-4-pyridinones, and their binding to Aβ fibrils was found to be dependent on ring structure. This work incorporates the first examples of rationally-designed small-molecule Alzheimer’s therapeutics incorporating such multifunctionality, and it is expected that the combination will promote more effective Alzheimer’s intervention than current metal ion- binding therapeutics such as clioquinol and other 8-hydroxyquinoline derivatives in development. iv   TABLE OF CONTENTS  ABSTRACT  ............................................................................................................................... ii TABLE OF CONTENTS .............................................................................................................. iv LIST OF TABLES ........................................................................................................................ vii LIST OF FIGURES ..................................................................................................................... viii LIST OF CHARTS ........................................................................................................................ xi LIST OF SCHEMES .................................................................................................................... xii LIST OF ABBREVIATIONS ..................................................................................................... xiii ACKNOWLEDGEMENTS ......................................................................................................... xix CHAPTER 1 Introduction ........................................................................................................... 1 1.1 Metals in Medicine .......................................................................................................... 1 1.1.1 Introduction of Metal Ions into the Biological System: Imaging Agents, Therapeutics, Biomolecule Mimetics ............................................................................................................ 1 1.1.2 Removal of Metal Ions from the System: Considerations for the Development of Therapeutic Metal Binding Agents ......................................................................................... 5 1.2 Alzheimer’s Disease ........................................................................................................ 7 1.2.1 Introduction to Alzheimer’s Disease and the Amyloid Hypothesis .............................. 7 1.2.2 The Aetiology of Alzheimer’s Disease and Current Focus of Treatment ..................... 8 1.2.3 The Involvement of Metals in the Pathology of Alzheimer’s Disease ........................ 11 1.2.4 Amyloid-beta (Aβ) Peptide as Metalloprotein ............................................................ 12 1.2.5 The Role of Aluminium in Alzheimer’s Disease ........................................................ 16 1.3 Metal Ion Passivation as a Therapeutic Strategy for Alzheimer’s Disease ................... 17 1.3.1 The Use of Metal Chelators to Attenuate Aβ-Mediated Toxicity ............................... 17 1.3.2 Testing Biological Activity/Applicability of Putative Alzheimer’s Disease Therapeutics .......................................................................................................................... 19 v   1.3.3 Compounds Tested and/or Designed as Alzheimer’s Disease Therapeutics ............... 22 1.4 Imaging Alzheimer’s Disease ........................................................................................ 36 1.5 The Multi-Functionality Approach to Chelator Design ................................................. 40 1.6 Thesis Overview ............................................................................................................ 41 CHAPTER 2 Multifunctional Pyridinone Chelators – Synthesis and Characterisation ........... 42 2.1 Introduction .................................................................................................................... 42 2.2 Experimental .................................................................................................................. 46 2.2.1 Materials ...................................................................................................................... 46 2.2.2 Instrumentation ............................................................................................................ 47 2.2.3 Nomenclature Conventions, Abbreviations for 3-Hydroxy-4-Pyridinones ................. 48 2.2.4 Pro-Ligand Synthesis ................................................................................................... 49 2.2.5 Glycosylated Compound (Prodrug) Synthesis ............................................................ 61 2.3 Results and Discussion .................................................................................................. 67 2.3.1 Pro-Ligand Preparation ................................................................................................ 67 2.3.2 X-Ray Diffraction Structural Characterisation of Pro-Ligands ................................... 70 2.3.3 Glycosylated Prodrug Preparation ............................................................................... 73 2.3.4 1H and 13C NMR of Prodrugs ...................................................................................... 76 2.4 Conclusions .................................................................................................................... 78 CHAPTER 3 In Vitro Characterisation of Multifunctional Pro-Ligands as Alzheimer’s Disease Therapeutics…. ............................................................................................................................. 80 3.1 Introduction .................................................................................................................... 80 3.2 Experimental .................................................................................................................. 82 3.2.1 Materials ...................................................................................................................... 82 3.2.2 Instrumentation ............................................................................................................ 83 3.2.3 Copper(II) Complexes ................................................................................................. 84 3.2.4 Trolox Equivalent Antioxidant Capacity (TEAC) Assay ............................................ 88 3.2.5 Cytotoxicity (MTT) Assay .......................................................................................... 89 vi   3.2.6 Turbidity Assay ........................................................................................................... 90 3.2.7 Enzyme-Linked Immunosorbent Assay (ELISA) ....................................................... 92 3.3 Results and Discussion .................................................................................................. 94 3.3.1 Synthesis and Characterisation of Bis(pyridonato)copper(II) Complexes .................. 94 3.3.2 Antioxidant Capacity of Pro-Ligands ........................................................................ 103 3.3.3 Cytotoxicity of Pro-Ligands ...................................................................................... 106 3.3.4 Interaction of Pro-Ligands with Aβ Fibrils ............................................................... 108 3.3.5 Interaction of Pro-Ligand with Metal Ions and Aβ Peptide in Solution .................... 112 3.4 Conclusions .................................................................................................................. 115 CHAPTER 4 Future Work ...................................................................................................... 117 4.1 Future In Vitro and In Vivo Assays for Drug Development ....................................... 117 4.1.1 Optimisation of Amyloid Fibril Binding Assay (ELISA) ......................................... 118 4.1.2 Expansion of In Vitro Aβ Fibrillisation Assays: Fibrillisation Inhibition and Prevention of Aβ-Induced Cytotoxicity ............................................................................. 119 4.1.3 Expansion of Turbidity, TEAC and MTT Assays to Include Glycosylated Pro-Ligands (Prodrugs) ........................................................................................................................... 120 4.2 Radioactive Analogues of Pyridinone Pro-Ligands ..................................................... 121 4.2.1 Gastrointestinal Crossing Assays .............................................................................. 122 4.2.2 Blood-Brain Barrier Crossing and Biodistribution Assays ....................................... 123 4.2.3 Probing In Vitro Aβ Targeting................................................................................... 123 4.3 Other Masking Groups for the 3-Hydroxyl Functionality ........................................... 125 References…...  ........................................................................................................................... 126 Appendix…….  ........................................................................................................................... 137 Crystallographic Data for 3-Hydroxy-4-Pyridinone Pro-Ligands .......................................... 137 Crystallographic Data for Bis(pyridinato)copper(II) Complexes ........................................... 146  vii   LIST OF TABLES  Table 1.1. Metal binding compounds designed and/or tested for AD therapy………………….23  Table 3.1. Selected infrared absorption bands and their assignments. .................................... 95  Table 3.2. Trolox equivalent antioxidant capacity (TEAC) values for all tested compounds, compared to standard antioxidants α-tocopherol (vitamin E) and butylated hydroxytoluene (BHT). .......................................................................................................................................... 105  Table 3.3. IC50 values of tested compounds, compared to cisplatin, used here as a reference compound. ................................................................................................................................... 108  Table 3.4. Percent efficacy of pro-ligand attenuation of metal ion-mediated Aβ40 aggregation in solution. “M” refers to either Zn2+ or Cu2+. ................................................................................. 114   viii   LIST OF FIGURES  Figure 1.1. (a) First-generation imaging agents 99mTc-sestamibi (Cardiolite, left) and 99mTc- bicisate (Neurolite, right) are targeted to the tissue of interest by the chemical and physical properties of the overall compound.5 (b) Cisplatin6 and auranofin7 for cancer and rheumatoid arthritis treatment, respectively. ..................................................................................................... 2  Figure 1.2. Vanadium complexes as insulin-enhancing agents: bis(ethylmaltolato)oxo- vanadium(IV) (BEOV), bis(metformin)14 and bis(thiazolidinedione)15 complexes. ...................... 5  Figure 1.3. Successive cleavage of amyloid precursor protein (APP) by β- and γ-secretases, producing amyloidogenic peptide fragments Aβ40 and Aβ42 in addition to other peptide fragments including soluble APPβ (sAPPβ). ................................................................................... 9  Figure 1.4. “Crossover compounds” developed for other applications yet tested for AD therapy: (a) copper overload therapeutics D-penicillamine115 and triethylenetetraamine (TETA);133 (b) NAC and NACA (AD4).122 .......................................................................................................... 24  Figure 1.5. Multidentate pro-ligands used for treatment of iron-overload and other conditions: DFO, L1.139 ................................................................................................................................... 26  Figure 1.6. (a) “Hybrid” metal binding and antioxidant ligands in prodrug (non-chelating) form incorporating structural features of deferiprone (metal binder) and BHT (antioxidant); R = any of a series of acyl groups.150 (b) Selected 3-hydroxy-4-pyridinone compounds showing activity in various AD-relevant assays: all pendant carbohydrate-bearing compounds (upper) are able to inhibit metal-induced Aβ1-40 aggregation in vitro.144 Feralex (lower), a glucose-bearing deferiprone derivative, was developed for therapeutic metal ion manipulation.145 ...................... 28  Figure 1.7. Compounds developed for therapeutic metal ion manipulation in Alzheimer’s disease, prodrug compound DP-109,148 putatively Aβ-associating chelator XH1,97 and bicyclam JKL 169.131 .................................................................................................................................. 30  Figure 1.8. Tetrahydrosalen pro-ligands (a) with pendant glucose molecules or (b, c) with glucose linkage installed as part of the prodrug concept, masking metal ion binding until activation by enzymatic deglycosylation. ..................................................................................... 31  Figure 1.9. (a) Chloroquine (R = H) and hydroxychloroquine (R = OH), metal chelating antimalarial agents tested for AD application, (b) clioquinol (5-chloro-7-iodo-8- hydroxyquinoline), and tetradentate quinoline-based chelators (c) bis(3-hydroxyquinoline)164 and (d) bis(8-aminoquinoline).165 ........................................................................................................ 33   ix   Figure 1.10. Structural similarity between a pyridinone pro-ligand and an established PET imaging agent for amyloid. Current amyloid imaging agent Pittsburgh Compound B (6-OH- BTA-1, * indicates location of 11C radiolabel);172 (a) 3-hydroxy-4-pyridinone compounds Hdapp (R,R′ = CH3,), Hsapp (R = H, R′ = CH3) and Hzapp (R,R′ = H); (b) two 3-hydroxy-4- pyridinone compounds differing in linker length between aromatic rings: Hppp (n = 0) and Hnbp (n = 1); (c) benzothiazole-bearing 3-hydroxy-4-pyridinones Hbt2p (upper) and Hbt6p (lower)........................................................................................................................................... 39  Figure 1.11. Rational design of multifunctional masked pyridinone prodrugs for neurological disease therapy. ............................................................................................................................. 41  Figure 2.1. (a) 3-Hydroxy-2-methyl-4-pyranone, maltol; (b) tautomers of a generalised 3- hydroxy-2-methyl-4-pyridinone. ...................................................................................... 42  Figure 2.2. Diagram of prodrug approach: systemic circulation of glycosylated (inactive) pyridinone pro-ligand (prodrug), facilitated diffusion across biological membranes (BBB) into the brain, and β-glucosidase-mediated hydrolysis yielding the active pyridinone metal ion chelator. ......................................................................................................................................... 45  Figure 2.3. Ellipsoid plots (50 % probability; for clarity, H-atoms not shown) of Hdapp, Hsapp, Hzapp, Hbt6p, Hbt2p and Hnbp. ................................................................................................ 71  Figure 2.4. 1H NMR (MeOD-d4, 400.13 MHz, RT) spectra of Hsapp (upper) and β-Gsapp (lower), from 9.0 – 2.0 ppm; s indicates solvent peaks. ................................................................ 77  Figure 3.1. Structures of reagents used in Chapter 3. ................................................................... 81  Figure 3.2. IR spectra for Hsapp pro-ligand (upper) and corresponding Cu2+ complex Cu(sapp)2 (lower), 4000-400 cm-1. ............................................................................................... 97  Figure 3.3. Ellipsoid plot (50 % probability; for clarity, H-atoms not shown) of Cu(dapp)2; side-on view of square planar copper(II) coordination centre. ...................................................... 98  Figure 3.4. Ellipsoid plots (50 % probability; for clarity, H-atoms not shown) of Cu(dapp)2, Cu(zapp)2, Cu(ppp)2 and Cu(nbp)2. ............................................................................................ 99  Figure 3.5. Experimental and simulated EPR spectra of Cu(dapp)2 (130 K). ........................... 102  Figure 3.6. Trolox equivalent antioxidant capacity (TEAC) of hydroxypyridinones compared to standard antioxidants α-tocopherol (α-toc, vitamin E) and butylated hydroxytoluene (BHT). Shown are means of three trials; error bars represent ± one standard deviation. ........................ 104  Figure 3.7. Sample survival plot of HepG2 cells exposed to varying concentrations of Hppp for 72 h, as monitored by MTT assay. n = 4, error bars indicate ± one standard deviation. ............. 107  Figure 3.8. 1-Carboxyethyl-3-hydroxy-2-methyl-4(1H)-pyridinone, Hcep. .............................. 110 x    Figure 3.9. Positive result for thioflavin T association with Aβ42 fibrils by ELISA; non- statistically significant evidence for Hzapp fibril binding up to the solubility limit (4 mM).  ... 111  Figure 3.10. Pro-ligand disaggregation of metal-induced Aβ40 aggregates in vitro. Bars indicate mean resolubilisation efficiency of each pro-ligand based on solution absorbance (405 nm, n ≥ 3); error bars indicate ± one standard deviation. ......................................................................... 113  Figure 4.1. Schematic diagram of a Caco-2 cell monolayer cultured on a porous membrane yielding two accessible sides of the model GI endothelium, both apical and basolateral, to be analysed for test compound concentration. ................................................................................. 122   xi   LIST OF CHARTS  Chart 1.1. Redox chemistry involving metal ion cycling and the Aβ peptide: (a) reduction of metal ion by amyloid beta peptide (Aβ) (b) redox cycling of metal ion (i.e. Fe3+, Cu2+), (c) production of H2O2, (d) Fenton and (e) Haber-Weiss chemistry to produce the hydroxyl radical. ........................................................................................................................................... 15  Chart 1.2. Structures of selected Aβ imaging agents as derived from starting pharmacophores (a) tissue stain Congo Red, (b) tissue stain thioflavin T and (c) the aminonaphthyl scaffold. Site of radiolabelling is indicated by *. ................................................................................................ 37   xii   LIST OF SCHEMES Scheme 2.1. Mechanism of aminolysis reaction converting a generalised pyranone (I) to its pyridinone analogue (VII) in aqueous acidic conditions. R = methyl or benzyl group; R′ = a variously substituted aryl group. ............................................................................................................................................. 43  Scheme 2.2. (Upper) Generalised structures and names of all pyridinone pro-ligands (left) and glycosylated prodrugs (right) discussed in this chapter. (Lower) complete structures of all pyridinone pro-ligands. .................................................................................................................. 48  Scheme 2.3. Synthesis of Hpnp, Hzapp from parent compound Bnpnp. ................................... 53  Scheme 2.4. Synthesis of Hnbp. ................................................................................................... 59  Scheme 2.5. General synthetic route to glycosylated pyridinone prodrugs; a mixture of α- and β- anomers is formed.......................................................................................................................... 61  Scheme 2.6. Synthesis of β-Gzapp via β-Gpnp, circumventing production of multiply- glycosylated biproducts. ................................................................................................................ 63  Scheme 2.7. General synthetic route for 3-hydroxy-2-methyl-4-pyridinone pro-ligands. For preparation of ether-protected pyranones: For preparation of Hsapp: R = Me, b = 2:3 H2O:MeOH, 100 °C, 30 h; c = BBr3, CH2Cl2, -78 to  -12 °C, 8 h. For preparation of Hdapp, Hbt6p, Hbt2p, Hppp: R = Bn, b = 2:1 H2O:MeOH, 110-130 °C, 96 h; c = HBr, acetic acid, 78 °C, 35 min. ................................................................................................................................ 68  Scheme 2.8. (a) Average structure and (b) average bond distances derived from six available data sets of pyridinone solid state structures; all distances labelled in angstroms (Å). ................. 72  Scheme 3.1 General procedure for preparation of bis(pyridonato)copper(II) complexes. ........... 84  Scheme 4.1. Preparation of a radiolabelled, iodinated analogue [123I]Hbt2p. R = Me or Bn .... 121  Scheme 4.2. Boronate-masked pyridinone compounds BL would be ineffective for metal ion binding; only conversion by H2O2 produces the pro-ligand HL which may coordinate metal ions (Mn+) such as Cu2+ or Zn2+ in areas of oxidative stress. .............................................................. 125  xiii   LIST OF ABBREVIATIONS  α alpha α-tocopherol (2R)-2,5,7,8-tetramethyl-2-[(4R,8R)-4,8,12-trimethyltridecyl]-3,4-dihydro- 2H-chromen-6-ol Å angstrom, 1 x 10-10 metre β beta γ gamma δ delta; chemical shift; in parts per million (ppm) vs. a standard (NMR) εmax epsilon-max; extinction coefficient (spectrophotometry); in L•mol-1•cm-1 λmax lambda-max; wavelength at which maximal absorbance occurs (spectrophotometry) µ micro (10-6) ν nu; frequency in cm-1 (IR) 6-OH-BTA-1 2-[4′-(methylamino)phenyl]-6-hydroxy-benzothiazole, PIB a atto (10-18) Aβ beta-amyloid Abg Agrobacterium faecalis β-glucosidase ABTS 2,2’-azinobis-(3-ethylbenzothiazoline-6-sulphonic acid) AD Alzheimer’s disease AFM atomic force microscopy ALS amyotrophic lateral sclerosis Anal analytical ANOVA analysis of variance APP amyloid precursor protein Ar aromatic (NMR) AU arbitrary units β-Gbt2p 1-(2-benzothiazolyl)-3-(β-D-glucopyranosyloxy)-2-methyl-4(1H)- pyridinone xiv   β-Gbt6p 1-(6-benzothiazolyl)-3-(β-D-glucopyranosyloxy)-2-methyl-4(1H)- pyridinone β-Gdapp 3-(β-D-glucopyranosyloxy)-2-methyl-1-(4-dimethylaminophenyl)-4(1H)- pyridinone β-Gpnp 3-(β-D-glucopyranosyloxy)-2-methyl-1-(4-nitrophenyl)-4(1H)-pyridinone β-Gsapp 3-(β-D-glucopyranosyloxy)-2-methyl-1-(4-methylaminophenyl)-4(1H)- pyridinone β-Gzapp  1-(4-aminophenyl)-3-(β-D-glucopyranosyloxy)-2-methyl-4(1H)-pyridinone BBB blood-brain barrier BCA bicinchoninic acid; 4,4'-dicarboxy-2,2'-biquinoline BHT butylated hydroxytoluene Bn benzyl br broad (IR) BSA bovine serum albumin BTA-1 2-[4′-(methylamino)phenyl]benzothiazole ° C degrees Celsius Calcd calculated CCDC Cambridge crystallographic data centre cisplatin cis-diamminedichloroplatinum(II) cm-1  wavenumber (reciprocal centimetre) CNS central nervous system CQ 5-chloro-7-iodo-8-hydroxyquinoline, clioquinol, PBT1 CSF cerebrospinal fluid d doublet (NMR) D-pen D-penicillamine, (2S)-2-amino-3-methyl-3-sulphanyl-butanoic acid dd doublet of doublets (NMR) Da Dalton; one atomic mass unit DFO N'-[5-(acetyl-hydroxy-amino)pentyl]-N-[5-[3-(5-aminopentyl-hydroxy- carbamoyl)propanoylamino]pentyl] -N-hydroxy-butane diamide, desferrioxamine, desferrioxamine-B, deferoxamine, desferal, DFO-B DFT density functional theory xv   DMF dimethylformamide DMS dimethylsulphate DMSO dimethylsulphoxide DNA deoxyribonucleic acid DP-109 1,2-bis(2-aminophenyloxy)ethane-N,N,N',N'-tetraacetic acid, N,N′-bis(2- octadecyloxyethyl)ester DTPA diethylenetriaminepentaacetic acid EA elemental analysis EDTA ethylenediaminetetraacetic acid ELISA enzyme-linked immunosorbent assay EPR electron paramagnetic resonance ESI-MS electrospray ionisation mass spectrometry Et2O diethyl ether EtOAc ethyl acetate EtOH ethanol FBS fetal bovine serum FDA Food and Drug Administration (USA) FDDNP 2-(1-{6-[(2-fluoroethyl(methyl)amino]-2-naphthyl}ethylidene)malononitrile FT-IR Fourier transform infrared spectroscopy g gram GI gastrointestinal GL glycosylated pro-ligand, or prodrug gluc glucose (NMR) GLUT a protein belonging to the glucose transport facilitator family h hour(s) Hbt2p 1-(2-benzothiazolyl)-3-hydroxy-2-methyl-4(1H)-pyridinone Hbt6p 1-(6-benzothiazolyl)-3-hydroxy-2-methyl-4(1H)-pyridinone Hcep 1-carboxyethyl-3-hydroxy-2-methyl-4(1H)-pyridinone Hdapp 3-hydroxy-2-methyl-1-(4-dimethylaminophenyl)-4(1H)-pyridinone HepG2 human hepatocellular liver carcinoma cell line HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulphonic acid xvi   HF hyperfine HIV human immunodeficiency virus HL free (unbound) pro-ligand Hnbp 1-benzyl-3-hydroxy-2-methyl-4(1H)-pyridinone HPLC high-performance liquid chromatography Hpnp 3-hydroxy-2-methyl-1-(4-nitrophenyl)-4(1H)-pyridinone Hppp 3-hydroxy-2-methyl-1-phenyl-4(1H)-pyridinone HR-MS high resolution-mass spectrometry Hsapp 3-hydroxy-2-methyl-1-(4-methylaminophenyl)-4(1H)-pyridinone Hz Hertz (s-1) Hzapp 1-(4-aminophenyl)-3-hydroxy-2-methyl-4(1H)-pyridinone IC50 concentration at which 50 % of test subjects are viable relative to the control IMPY 6-iodo-2-[4′-N-dimethylamino]phenylimidazo[1,2-a]pyridine IR infrared iPrOH 2-propanol J coupling constant (NMR) JKL 169 1,1′-xylyl bis-1,4,8,11 tetraaza cyclotetradecane Kd dissociation constant L litre L-DOPA 3-(3,4-dihydroxyphenyl)-L-alanine L-mimosine (2S)-β-[N-(3-hydroxy-4-pyridone)]-α-aminopropionic acid L1 3-hydroxy-1,2-dimethyl-4-pyridinone; Hdpp m metre, milli (10-3), multiplet (NMR) M molarity (moles•cm-1 ) or metal Ma maltol; 3-hydroxy-2-methyl-4-pyrone MALDI-TOF matrix assisted laser desorption/ionisation time-of-flight Me methyl (CH3 functional group) MeOH methanol Met35 methionine-35 (residue at position 35 of the peptide) MHz megahertz min minute(s) xvii   mL millilitre(s) mol mole(s) mp melting point MPAC metal-protein attenuating compound MRI magnetic resonance imaging MS mass spectrometry MTT 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide mW milliwatt m/z mass per unit charge n nano (10-9) NAC (R)-2-acetamido-3-sulphanyl-propanoic acid, N-acetyl cysteine NACA N-acetyl cysteine amide, AD4 NEt3 triethylamine NFT neurofibrillary tangle NMR nuclear magnetic resonance NP nanoparticle ORTEP Oak Ridge thermal ellipsoid plot PBS phosphate-buffered saline PBT2 dichloro- analogue of clioquinol (structure unadvertised) PBu3 tributylphosphine Pd/C palladium on carbon PD Parkinson’s disease PET positron emission tomography pH -log[H3O+] PIB Pittsburgh Compound B, 2-[4′-(methylamino)phenyl]-6-hydroxy- benzothiazole, 6-OH-BTA-1 ppm parts per million PSA polar surface area PTFE polytetrafluoroethylene pyrid pyridinone (NMR) q quartet (NMR) xviii   RBF round-bottomed flask Rf retention factor (TLC) ROS reactive oxygen species RT room temperature s second(s), singlet (NMR) SD standard deviation SDS-PAGE sodium dodecyl sulphate-polyacrylamide gel electrophoresis SEM standard error of the mean SIMS secondary ion mass spectrometry SP senile plaque SPECT single photon emission computed tomography t triplet (NMR) t1/2 half-life TEAC Trolox equivalent antioxidant capacity TEM transmission electron microscope/microscopy TETA triethylenetetraamine TFA trifluoroacetic acid Tg transgenic theor theoretical THF tetrahydrofuran TLC thin layer chromatography Trolox 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid UBC University of British Columbia (Vancouver) UV ultraviolet vis visible X34 1,4-bis(3-carboxy-4-hydroxyphenylethenyl)benzene XAS X-ray absorption spectroscopy XH1 [(4-benzothiazol-2-yl-phenylcarbamoyl)-methyl]-{2-[(2-{[(4-benzothiazol- 2-yl-phenylcarbamoyl)methyl]-carboxymethyl-amino}-ethyl)- carboxymethyl-amino]-ethyl}-amino)-acetic acid xix   ACKNOWLEDGEMENTS  Thank you to all members of the Orvig lab, past and present, for their support both in and out of the lab; particularly Drs. Cheri Barta and Meryn Bowen. For all their help in the early days of my grad school career, I thank Drs. Tim Storr, Michael Merkel and Neil Lim for showing me how good research is done, and how to properly set up and manage a lab space that works. Thanks to the best summer student a grad student could have, Mr. Brent Page. You were a pleasure to supervise, and you know how much you helped me in completing some of the sections of this work. I won a good coin-toss there. Thanks to all the lab group members (Yasmin, Paloma, Eszter) who helped me out especially when I was recovering from my multiple snow-related injuries; I know you guys moved a bunch of extra boxes, and hauled a few extra Dewars of liquid nitrogen because of me, and I thank you for it. Thanks to my group members outside the group, including JDT and Brian for general synthesis discussions and troubleshooting. Of course, special thanks go to my best biochemistry lab partner, reviewer and editor, Jonathan. This project couldn’t have happened without the support of the people mentioned above, and wouldn’t have happened without the support and encouragement of my supervisor, Chris Orvig. Thanks, Chris, for giving me the freedom to determine the direction and focus of my project, and for always being my advocate. The majority of the material in Chapter 1 has been submitted for publication: Scott, L. E. and Orvig, C. Medicinal inorganic chemistry approaches to passivation and removal of aberrant metal ions in disease. Chemical Reviews, 2009. Manuscript ID CR-2009-000176. I was xx   responsible for researching and writing this review article, preparing figures and formatting the manuscript; editing was performed by myself and Dr. Orvig. Portions of the content of Chapters 2 and 3 pertaining to compounds Hppp and Hnbp compounds have been published: Scott, L. E.; Page, B. D. G.; Patrick, B. O.; Orvig, C. Altering pyridinone N-substituents to optimise activity as potential prodrugs for Alzheimer’s disease. Dalton Transactions, 2008, 6364-6367. I was responsible for the outline of the project and the design of ligand Hnbp; I directly supervised Mr. Brent Page in the chemical synthesis and analysis of all compounds. Turbidity and TEAC assays were done both by myself, and by Mr. Page under my supervision; MTT assays were done by Mr. Page. Mr. Page was responsible for figure preparation, and we were jointly responsible for manuscript writing and editing. Portions of the content in Chapters 2 and 3 pertaining to the compounds comprising amyloid-targeting structures (Hdapp, Hsapp, Hzapp, Hbt6p, Hbt2p) will be submitted for publication: Scott, L. E.; Merkel, M.; Page, B. D. G.; Bowen, M. L.; Adam, M. J.; Orvig, C. Synthesis and characterisation of N-aryl-substituted 3-(β- D-glucopyranosyloxy)-2-methyl-4(1H)-pyridinones as brain-directed chelators for Alzheimer’s therapy. Dr. Michael Merkel was responsible for the original design and synthesis of pyridinone Hbt2p; I was responsible for the design of the remaining pyridinones, the synthesis of all five compounds, and the preparation and analysis of all other forms of the compounds including their glycosylated prodrug forms and copper coordination complexes. I was solely responsible for the implementation and optimisation of all assays with the exclusion of the MTT assay, which Mr. Page performed. All solid state structures in this thesis were obtained by X-ray diffraction by Dr. Brian Patrick, UBC X-ray services.   1   CHAPTER 1 Introduction  1.1 Metals in Medicine  The field of medicinal inorganic chemistry has evolved with three conceptual aims: the introduction of metal ions to the biological system, manipulation and redistribution of metal ions within the system, and removal of metal ions from the system. This project arises from the latter two goals: binding of metal ions for redistribution or removal. Metal ions play a pivotal role in the development and pathology of a range of conditions, and in some cases are implicated in redox chemistry leading to oxidative stress. Among these conditions are Parkinson’s disease,1 Friedreich’s ataxia,2 transfusion-related iron overload,3 Wilson’s disease4 and Alzheimer’s disease.1 All of these conditions involve elevated levels of metal ions in particular tissues or cell compartments of the body, and present challenges in the field of medicinal inorganic chemistry for development of new chelators for therapeutic application.  1.1.1 Introduction of Metal Ions into the Biological System: Imaging Agents, Therapeutics, Biomolecule Mimetics  Metal complexes are often introduced into the biological system as imaging agents for the diagnosis of disease. These complexes generally incorporate gamma-emitting radionuclides for use in single-photon emission computed tomography (SPECT) or positron-emitting isotopes 2   for positron emission tomography (PET).5 So-called “first-generation” complexes are targeted to the tissue or organ of interest solely by the chemical and physical properties of the complex. For example, the technetium-based imaging agent 99mTc-sestamibi (Cardiolite, Figure 1.1a, left) is lipophilic and monocationic; it is taken up by the sodium/potassium pump in hard-working heart tissue for cardiac imaging, whereas 99mTc-bicisate (Neurolite, Figure 1.1a, right) is an uncharged complex and thus is capable of permeating the blood-brain barrier (BBB) for measurement of cerebral blood flow.  + Tc NN S S EtOOC COOEtO Tc CC C C N N NN C C N N OMe OMe MeO MeO OMe OMe a b O AcOAcO OAc S OAc Au P H3N Pt Cl ClH3N   Figure 1.1. (a) First-generation imaging agents 99mTc-sestamibi (Cardiolite, left) and 99mTc- bicisate (Neurolite, right) are targeted to the tissue of interest by the chemical and physical properties of the overall compound.5 (b) Cisplatin6 and auranofin7 for cancer and rheumatoid arthritis treatment, respectively.  In contrast, “second-generation” imaging agents use an incorporated biomolecule to interact with a specific receptor within the body resulting in preferential uptake of the complex 3   in a certain type of organ, tissue or cell. This bioconjugate approach requires a biomolecule, a linker and a metal ion-binding moiety within the pro-ligand. Examples of second-generation complexes include carbohydrate-linked 99mTc agents for SPECT-based cancer imaging8 and antibody- or peptide-linked 99mTc complexes such as 99mTc-apcitide (AcuTect) or 99mTc- Arcitumomab (CEA-Scan) for selective imaging of deep-vein thromboses and colorectal cancer, respectively. In addition to their use as diagnostic agents, metal complexes may be introduced into the body for therapeutic use. Therapeutic radiopharmaceuticals generally incorporate beta particle-emitting radiometals such as 90Y or other lanthanides, and like their diagnostic counterparts, these complexes may be targeted to their preferred site of action by complexation with bifunctional chelators (with metal ion-binding and biological activity). Examples include the non-Hodgkin’s lymphoma radiotherapeutic 90Y-ibritumomab tiuxetan (Zevalin), targeted to cancer cells by an incorporated monoclonal antibody.9 In addition to radiotherapeutic applications, stable isotopes of transition metals are administered in complexes designed for treatment of various conditions including cancer (e.g., platinum in cisplatin) and rheumatoid arthritis (e.g., gold complexes such as auranofin, both Figure 1.1b). The intended localisation and activity of the metal ion is achieved via the ligands imparting various physical and chemical characteristics to the complexes; cisplatin (cis-diamminedichloroplatinum(II), Figure 1.1b) hydrolyses within cells, yielding a positively charged complex which is trapped within the cell, binds to DNA and, through cross-linking adducts, effects antiproliferative activity (recently reviewed in context of new platinum compound development).6 Auranofin ((2,3,4,6-tetra-O- acetyl-1-thio-β-D-glycopyranosato-S)(triethylphosphine)gold, Figure 1.1b) is formulated for oral availability, unlike its injectible predecessors such as gold sodium thiomalate. Though the mechanisms for therapeutic (and toxicological) effects are still unclear, auranofin is thought to 4   effect immunosuppression by a number of different actions.7 In these and other metal complexes administered for therapy, the metal ion is integral to drug activity which is targeted and modulated by ligands. Metal compounds may also be introduced to the biological system as biomolecule mimetics. For example, the Meggers group has developed protein kinase inhibitors using organometallic moieties as structural scaffolds for the design of biologically active compounds.10 In contrast to the metal-dependent activities of radionuclide complexes or drugs such as cisplatin and auranofin, in this construct the metal ion performs no direct action in the biological system. Instead, it is incorporated into a kinetically inert coordination complex such that the metal plays the role of “innocent bystander” yet organises the organic ligands to mimic enzyme substrates and become very high-affinity inhibitors for enzymes (for example, glycogen synthase kinase 3).11 A similar use of biologically active inorganic complexes uses vanadium compounds for enhancement (often misnamed mimicry) of the effect of a larger biomolecule (insulin) for therapy of diabetes mellitus. Phosphate ([PO4]3-) and vanadate ([VO4]3-) are chemically similar. Thus, vanadate can enter into cell signalling cycles and replicate the overall effect of insulin, increasing glucose transport and oxidation, stimulating glycogen synthesis in the liver and inhibiting glucose synthesis.12 Development of these vanadium insulin therapeutics has involved complexation of vanadyl with maltol or ethylmaltol to improve bioavailability13 or with small molecules that enhance the activity of insulin such as biguanides14 (metformin) or thiazolidinediones15 to try to get synergistic effects (Figure 1.2).  5   V O O O O V H N N H O NH NH2 NH2 2 2 V O O O O 2 O NH S O O  Figure 1.2. Vanadium complexes as insulin-enhancing agents: bis(ethylmaltolato)oxo- vanadium(IV) (BEOV), bis(metformin)14 and bis(thiazolidinedione)15 complexes.  1.1.2 Removal of Metal Ions from the System: Considerations for the Development of Therapeutic Metal Binding Agents  The use of chelating agents to adjust metal ion/metalloid toxicity began in the early 1900s with researchers such as Alfred Werner, Paul Ehrlich, and Carl Voegtlin aiming to reduce toxicity of arsenic- and antimony-containing drugs for such parasitic diseases as syphilis, trypanosomiasis and schistosomiasis; small-molecule chelators were applied to relieve the effects of heavy metal and metalloid overload. The use of small-molecule chelators to relieve accidental overexposure to metal ions began in 1941 with the (questionably appropriate) use of citrate for lead intoxication.16 Manipulating the distribution of metal ions in biological systems in a specific way is a complicated process, and it is exceedingly difficult to effectively model the expected pharmacokinetics of both the free pro-ligand and the complex, and to take into account all of the biological contributions to every chemical reaction - and all biochemical implications of such reactions. Thus the efficacy and activity of metal ion chelators are generally not well quantified or theoretically modelled, but probed by and discussed in context of empirical evidence from biological experiments. Model systems are used to characterise the action of new 6   therapeutics; these include metal-loaded cultured cells, ex vivo tissue and cells, animal models (including genetic mimics of disease states) and finally clinical observations and formal clinical trials. Metal ion abnormalities are implicated in a number of neurological disorders (such as AD,1 PD1 and Friedreich’s ataxia2), and putative drug therapies for these conditions must be developed taking into account what is known as the blood-brain barrier. The blood-brain barrier, or BBB, is formed by the endothelium of the brain blood vessels, the basal membrane and the neuroglial cells, and separates the brain interstitial fluid from the circulating blood to insulate the brain from fluctuations in blood levels of metal ions and small molecule metabolites. Its extreme selectivity means that from a drug design perspective, physicochemical properties such as lipophilicity and molecular weight must be considered if the drug is to permeate to the central nervous system (CNS). For passive BBB penetration, drugs or prodrugs must be uncharged at physiological pH, relatively lipophilic (octanol/water partition coefficient, logP > 1.5), compact (small polar surface area, PSA), and of low molecular weight (less than about 500 g/mol). Functional group modification can be used to increase passive BBB uptake; generally this would focus on increasing drug lipophilicity, for instance by esterification of carboxyl groups. Other transport mechanisms exist in the BBB, however, such as hexose, amino acid and neuropeptide transporters, and these can be utilised to impart brain uptake to other therapeutics not meeting the above criteria for passive BBB permeation. The glucose transporter (GLUT) family of membrane transport proteins can be utilised by conjugation of the drug to a glucose molecule; this has been used to increase brain uptake of HIV,17 Alzheimer’s disease (AD),18 and Parkinson’s disease (PD)19 therapeutics, among others. The amino acid transporter can be exploited by creation of a “pseudonutrient,” modifying the structure of the drug to mimic 7   nutrient structure (e.g., using L-DOPA as substrate for amino acid transporter for effective brain delivery of dopamine20). The “Trojan horse” approach uses peptides such as insulin or transferrin bound to the drug to exploit receptor-mediated transcytosis mechanisms in the BBB.21 Similarly, nanoparticles are the newest vector to be suggested for BBB permeation of CNS drugs.22 A more exhaustive summary of approaches for increased BBB penetration has recently been presented elsewhere.23  1.2 Alzheimer’s Disease 1.2.1 Introduction to Alzheimer’s Disease and the Amyloid Hypothesis  A few of the erroneous assumptions existing today about Alzheimer’s disease are that the disease is not fatal, it is a natural part of the aging process, and it only affects the elderly. Although the aetiology of AD is still poorly understood, its pathology is well known and has been characterised for over one hundred years since Alois Alzheimer first described “a peculiar disease of the cerebral cortex”.24 Now, AD affects more than 24 million people worldwide, with this number expected to reach over 81 million by 2040.25 AD patients experience multiple cognitive deficits including memory loss and disorientation linked with the breakdown of neuronal function and neuron death. This section will outline the prevailing understanding of the biochemical causes of AD, with a special focus on the role of metal ions and their interactions with Alzheimer’s-associated proteins (including redox chemistry). Current therapeutic approaches will be mentioned, followed by an introduction to metal binding molecules as applied to AD treatment, with considerations for their design. Next, an overview will be given of 8   compounds tested and developed for AD intervention, and finally, recent developments will be discussed in the theory of metal ion passivation for AD therapy.  1.2.2 The Aetiology of Alzheimer’s Disease and Current Focus of Treatment  There are two types of AD currently recognised: early-onset, in which symptoms appear prior to age 65, and late-onset which manifests after age 65, with the latter comprising 95 % of all diagnoses. While genetic factors have been identified in the development of early-onset AD, increased age is the major risk factor for late-onset AD.26 In both early- and late-onset forms of the condition, the pathology leading to a positive diagnosis is the same: the presence of extracellular plaques formed from a peptide called β-amyloid (Aβ), and intracellular deposits of a peptide called tau. These tissue markers of AD are generally accompanied by high levels of oxidative stress, inflammation in the brain and neurodegeneration. Linking the noted pathologies to a causative agent, the amyloid cascade hypothesis has defined the fibrillisation of Aβ into amyloid deposits as a toxic “gain-of-function.”27 Indeed, the genetic, biochemical and neuropathological evidence strongly suggest that Aβ amyloidogenesis is central to AD pathogenesis, with age-related increases in metal ion concentration associated with Aβ plaque deposition, redox reactions and oxidative damage in brain tissue.28 Aβ peptide is cleaved by secretase enzymes from membrane-bound amyloid precursor protein, APP (Figure 1.3).  9     Figure 1.3. Successive cleavage of amyloid precursor protein (APP) by β- and γ-secretases, producing amyloidogenic peptide fragments Aβ40 and Aβ42 in addition to other peptide fragments including soluble APPβ (sAPPβ).  Although the function of APP is unknown, recent evidence suggests that it functions in the maintenance of copper homeostasis.29 While Aβ is a natural product and is present in the brain (and the cerebrospinal fluid, CSF) normally throughout life, a particular self-association occurs in AD to form Aβ plaques. These plaques are extracellular fibrillised deposits of amyloid beta peptide (40-43 amino acid residues long)30,31 and first deposit in the glutamatergic synapse in the cortex and hippocampus, which is important for formation of the physical substrate of memory. Notably, this is the only place in the body where exchangeable copper and zinc are found together; Zn2+ is thought to be released into the extracellular space in either a free or exchangeable form32 and copper has been shown to be released in ionic form by post-synaptic APP sAPPβ β-secretase cell exterior cell interior Aβ40 Aβ42 γ-secretase γ-secretase 10   neurons.33,34 These Aβ plaques are known to be toxic, and aggregated Aβ is likely more neurotoxic than the native peptide.35 More recent evidence points to oligomers being particularly toxic;36 however, it is generally conceded that many forms of polymerised Aβ peptide, from small oligomers to large fibrils, are harmful.37 Recent progress in delineating the mechanism of Aβ toxicity has been reviewed by Cappai and Barnham.38 Another myth surrounding Alzheimer’s disease is that there are treatments available to halt the progression of the disease. In fact, current therapies are not able to stop disease progression, but offer only symptomatic relief and can, in the best case, slow cognitive decline. Generally these therapies attempt to address neurotransmitter defects, bolstering neuronal activity by enhancing the amount of acetylcholine neurotransmitter in the synaptic cleft (through acetylcholinesterase inhibition), by protecting neurons from further damage (via glutamate blockers/NMDA receptor inhibitors) or by restoring nerve activity (through supplementation of nerve growth factor). In addition, statins are thought to show promise in slowing neurodegeneration. Other therapies aim to alleviate the associated inflammation and oxidative stress in brain tissue (i.e. administration of non-steroidal anti-inflammatory drugs or antioxidants such as vitamin E, respectively); current AD therapies have recently been reviewed in detail.39 These treatments target only the symptoms and as it stands, new therapies are needed to target the underlying pathology of AD.   11   1.2.3 The Involvement of Metals in the Pathology of Alzheimer’s Disease  Increased age is the major risk factor for neurodegenerative disease and it is known that brain metal concentration increases as a result of normal aging.40-42 It is also clear that metal ions mediate the oxidative stress mechanism of Aβ toxicity.1,43 Copper in the AD brain appears to be miscompartmentalised rather than universally elevated; it is concentrated within Aβ plaques44 with observed levels of up to 400 µM,45 approximately two orders of magnitude higher than the normal brain extracellular level,46-48 and higher even than the levels reached in the synaptic cleft (250 µM on average).46 This elevation, however, is not matched by results from bulk brain studies which show no change in overall copper levels,49,50 or even a decrease in copper concentration vs. age-matched controls.51 Serum copper levels are elevated in AD patients which suggests that this may serve as a peripheral diagnostic tool for AD.52 Zinc has been shown to co-localise with dense (but not diffuse) Aβ plaques,53 showing levels of up to approximately 1 mM in plaques,45 but zinc levels in bulk brain are more difficult to quantify. There is evidence for elevated zinc levels in AD tissue vs. control (non-AD) tissue in various brain regions;45,51 other reports however, show a decrease in zinc levels in the AD brain.50,54 More focussed studies on the Aβ plaques of the AD brain show co-localisation of zinc within the plaques44,55 and zinc is also elevated in the CSF of AD patients.56 The status of iron in the Alzheimer’s brain is somewhat complicated as well. Iron imbalance in the AD brain was first reported in 1953 with Prussian blue staining of iron deposits in the cytoplasm of neurons containing neurofibrillary tangles (NFT) and in senile plaques.57 Separate analyses of bulk AD brain matter indicate iron elevation in a number of AD brain 12   regions vs. controls,58,59 elevation in NFT-bearing neurons at the cellular level,60 and in senile plaques.45 Unlike copper and zinc, iron does not co-purify with plaque-extracted Aβ;61 instead, it appears that iron is elevated in the neurons surrounding and extending within the plaques, but may not interact with the plaque peptide itself.62 Clearly there is evidence of dishomeostasis and overall miscompartmentalisation of metals such as copper, zinc and iron in the AD brain, with accumulation of copper and zinc in amyloid deposits and iron in plaque-associated neurons.  1.2.4 Amyloid-beta (Aβ) Peptide as Metalloprotein  The amyloid precursor protein, APP, is a ubiquitous transmembrane protein of unknown biological function. It has specific and saturable binding sites for zinc and copper ions with dissociation constants Kd of 764 nM63 and 10 nM,64 respectively (reviewed by Kong et al.65). Because these binding sites seem to be conserved across the APP superfamily of proteins, it seems that zinc and copper binding may play an important role in APP function and metabolism;66 putative functions of APP include regulation of cell growth and adhesion, and metal ion homeostasis, among others.67 The amyloid peptide fragment Aβ1-40 specifically and saturably binds Zn2+; early solution studies showed one high-affinity binding site (Kd = 107 nM) with 1:1 stoichiometry, and one low-affinity binding site (Kd = 5.2 µM) exhibiting 2:1 zinc:Aβ stoichiometry.68 Variable pH trials have pointed to this binding being histidine-mediated, as it was inhibited by low pH and by chemical alteration of histidine residues.69 In vitro, low micromolar concentrations of Zn2+ rapidly precipitate soluble Aβ into amyloid aggregates68,70 and intermolecular His(Nτ)-Zn2+–His(Nτ) bridges are thought to mediate this reaction.71 At 13   physiological pH (7.4), Zn2+ is the only relevant biometal able to precipitate Aβ,68 while at slightly acidic conditions Cu2+ and even Fe3+ are known to induce Aβ aggregation.69 Like for Zn2+, Aβ1-40 displays both high- and low-affinity binding sites for Cu2+ with Kd = 0.05 nM (estimated) and 13 nM, respectively.72 The other major peptide fragment Aβ1-42 displays higher Cu2+ affinity with Kd = 7 aM and 5 nM for the high- and low-affinity Cu2+ binding sites, respectively.72 Nuclear magnetic resonance (NMR) studies have characterised the structure of the Aβ metal binding site and implicate three histidine residues at positions 6, 13 and 14 on the peptide.73 In the same study, electron paramagnetic resonance (EPR) experiments indicated a square planar N3O coordination site, with N donors from the imidazole rings of the three histidine residues, and proposed oxygen donation from the position-10 tyrosine side chain hydroxyl group.73,74 Conversely, other studies have postulated that the oxygen donor is the carboxylate group of Glu5 or the N-terminal aspartate residue,74 the peptide amino terminus itself,74,75 or some other exogenous ligand such as water.71 Most recently, X-ray absorption spectroscopy (XAS) combined with density functional theoretical (DFT) analysis points to a N3O3 distorted six-coordinate binding mode for the high-affinity Cu2+-binding site of Aβ consisting of three histidine N-donors, glutamic and/or aspartic acid and axial water.76 While the main body of study into Aβ-metal binding has certainly focussed on copper and zinc, there is some evidence for an Aβ1-42-Fe complex with calculated Kd of 36 µM; furthermore, it seems that fibrillised Aβ1-42 binds Fe2+ much more tightly (Kd = 0.2 µM) than does the monomeric form.77 The conformation and kinetics of Aβ1-40 aggregation upon pH and metal ion challenge has been examined with a variety of methods. Atomic force microscopy (AFM) has identified two different types of Aβ1-40 aggregates formed depending on pH and the presence of metal ions (Fe2+, Cu2+, Zn2+); furthermore, these two aggregate types were found to differ in cytotoxicity by 14   in vitro assay.78 Surface plasmon resonance bioimaging has been employed to monitor the kinetics of Aβ1-40 aggregation upon exposure to a number of metal ions including Cu2+, Zn2+, Fe2+, Fe3+ and Ca2+.79 A range of biophysical techniques such as AFM, Fourier transform infrared spectroscopy (FT-IR), thioflavin T fluorescence assay and secondary ion mass spectrometry (SIMS) have been used to examine Aβ1-40 and Aβ1-42 deposition induced by metal ions; aggregate structure varied with metal ion used.80 Electron microscopy, gel electrophoresis, thioflavin T and light scattering methods were used to demonstrate the dependence of Aβ1-42 aggregation state on Cu2+:Aβ peptide ratio; sub-equimolar ratios led to thioflavin T-reactive amyloid and supra-equimolar ratios led to larger oligomers and amorphous aggregates.81 Beyond simple conformational/fibrillisation state change, the reaction of Aβ with various metal ions in vitro has been probed. As mentioned above, the metal ions of interest in amyloid- mediated Alzheimer’s pathology are zinc, copper and iron. Of these three, only copper and iron are redox-active under physiological conditions. It was noted in 1999 that Aβ peptide can directly produce H2O2 through Cu2+ and Fe3+ reduction,82 and that this catalytic production of H2O2 in the presence of biological reductants, Aβ, and Cu2+ was inhibited by the addition of copper chelators.61 In fact, Aβ  can rapidly reduce Cu2+ and Fe3+ 83 and promote redox cycling at near-physiological buffered conditions.84 This is important as the generated H2O2 can then react with the reduced metal ions via Fenton chemistry (Chart 1.1) to produce the hydroxyl radical (HO•). The hydroxyl radical is particularly reactive, abstracting hydrogen atoms from organic molecules extremely quickly and unselectively causing immediate oxidative damage in the vicinity of its production.85  15     Chart 1.1. Redox chemistry involving metal ion cycling and the Aβ peptide: (a) reduction of metal ion by amyloid beta peptide (Aβ) (b) redox cycling of metal ion (i.e. Fe3+, Cu2+), (c) production of H2O2, (d) Fenton and (e) Haber-Weiss chemistry to produce the hydroxyl radical.  Oxidative conditions can influence the mono/oligomeric state of Aβ as free radical attack on the tyrosine residue at position 10 can lead to stable dityrosine cross-linked dimers.86 The biological implications of these Aβ-associated redox reactions are observed in cell system experiments wherein redox-active Cu2+ and Fe3+ increase Aβ toxicity, but Zn2+ attenuates Aβ toxicity in vitro.87 There are conflicting data implicating the methionine-35 (Met35) residue of Aβ in the redox-mediated toxicity of Aβ peptide. Free radicals could oxidise the sulphur- containing side chain of Met35 to form a radical cation which then can abstract H from surrounding lipids to initiate lipid peroxidation, or from proteins to initiate protein oxidation. On the other hand, Met35 could donate an electron for metal ion reduction which could participate in Fenton chemistry. Methionine is at least important to the redox chemistry of the Aβ peptide, and an in-depth discussion of this residue’s role in Aβ redox chemistry can be found elsewhere.43  (a)  Aβ + M(n+1)+  Aβ+• + Mn+ (b) Mn+ + O2  M(n+1)+ + O2- (c)  O2- + O2- + 2H+  H2O2 + O2 (d)  Mn+ + H2O2  M(n+1)+ + HO- + HO• (e)  O2- + H2O2  HO- + O2 + HO• 16   1.2.5 The Role of Aluminium in Alzheimer’s Disease  Some researchers have linked aluminium to the aetiology of AD due to evidence of its elevation in NFT-bearing neurons. Specifically, X-ray spectrometric evidence showing aluminium accumulation in NFT-bearing neurons was given in 1980,88 laser microprobe techniques were used to demonstrate a small increase in aluminium level within AD neurons89 and some epidemiological studies have made an association between Al in drinking water and AD occurrence.90 Indeed, Al was the only metal monitored at all in participants in the desferrioxamine (DFO) AD clinical trial.91 While convincing evidence is still pending (after decades) to link this metal with the pathology of Alzheimer’s, its role is still a controversial issue in many circles and research continues today to prove this connection. Some researchers posit that aluminium is linked to amyloid deposition both in vitro and in vivo.92 A new staining method was recently used to demonstrate elevated Al3+ levels in ex vivo AD neurons,93 and a recent study on the AD transgenic mouse demonstrated impaired cognition and increased amounts of Aβ fragments in the brain with oral Al3+ supplementation.94 Despite these findings, it is generally accepted that although Al3+ is neurotoxic and can cause Alzheimer’s-like lesions in the brain, there is no convincing causal link between aluminium and AD initiation or progression.95    17   1.3 Metal Ion Passivation as a Therapeutic Strategy for Alzheimer’s Disease 1.3.1 The Use of Metal Chelators to Attenuate Aβ-Mediated Toxicity  Because of the detrimental interactions of metal ions with Alzheimer’s Aβ peptide, considerable focus has been placed on developing novel therapeutic approaches to modulate the metal–protein interactions.66,96 The term metal-protein attenuating compound (MPAC) was coined to describe the approach of chelator introduction to disrupt specific, abnormal metal- protein interactions,29 and it is distinct from the process of chelation and excretion of bulk metal ions, as is the case in copper removal in Wilson’s disease. The two approaches differ conceptually in the localisation (targeted vs. systemic) of chelator activity and the affinity with which the chelator binds metal ions. While use of MPACs is meant to repartition and normalise metal ion distribution, traditional chelation sequesters and clears metal ions from the body. With the MPAC concept in mind, some groups have attempted to rationally design metal ion binding agents to target the metal ions associated with Alzheimer’s disease Aβ. Targeting of chelator activity to the amyloid plaques may be attempted via structural manipulation, as in the case of the chelating AD therapeutic XH1 (vide infra) 97 and others.98 Because in AD therapy the desired site of action is in the brain, the BBB permeation of the chelator must be considered. Diffusion through the BBB requires a high level of hydrophobicity that is generally not seen in hard base- containing metal chelators, but other methods can be employed to imbue the chelator with the ability to permeate the BBB. For example, pendant glucose molecules have been attached to metal binding drugs to increase their CNS uptake, attempting to take advantage of the many GLUT1 hexose transporter proteins localised within the BBB.99 The glycosylation strategy has been applied to dopamine,19 3-hydroxy-4-pyridinone18 and tetrahydrosalen pro-ligands.100,101 18   Another strategy involves the use of nanoparticles (NP) to carry putative brain drugs across the BBB via the low-density lipoprotein receptor-mediated transport system.102 One advantage of this approach is that the lipophilicity of the chelator itself no longer needs to be considered, neither in the structural design of chelator nor regarding expected drug toxicity. The use of NPs to transport drugs also obviates molecular weight concerns generally accompanying brain drug design.103 Nanoparticle binding will likely also significantly change how the complexes exit the brain after chelation. It has been postulated that if the nanoparticles are not biodegradable they may be able to leave the brain via the apolipoprotein carrier-mediated transport system.104 A number of metal ion binding agents have been prepared based on the 3- hydroxy-4-pyridinone scaffold and proposed for conjugation to NPs for therapy of CNS diseases such as AD or PD; in addition to imparting greater BBB permeation, it is thought that NP linkage of many bidentate chelators will increase the effective ligand denticity and impart greater complex stability to complexes.102 In a similar bid to improve CNS uptake of metal chelators, the Mumper group has covalently linked D-penicillamine to NPs for brain delivery.105   19   1.3.2 Testing Biological Activity/Applicability of Putative Alzheimer’s Disease Therapeutics  Turbidity Assay for Inhibition of Metal Ion-Mediated Aβ Aggregation In aqueous medium, Zn2+ and Cu2+ promote the aggregation of synthetic human Aβ peptide, which may be reversed by the addition of metal chelating agents;18 the process is most readily observed by a light scattering-based “turbidity” assay and allows the comparison of different metal chelators based on efficacy of their interaction with, and attenuation of, metal ion-promoted amyloid aggregation. However, the turbidity assay posits no structural characterisation of the fibrillisation state of Aβ. The metal-induced aggregation of human Aβ1-40 was first observed in vitro by Bush et al. in 1994 wherein Zn2+ exposure reduced the recovery of Aβ1-40 from solutions passed through a 0.2-µM filter.68 The aggregation of Aβ1-40 at low pH was then visualised by simple light scattering (“turbidity”) and Congo Red binding in 1996.106 Since then, turbidity assays have marked Cu2+-induced Aβ1-40 aggregation at decreased pH; a process that is reversible with treatment with metal chelators such as EDTA.69 Various metal chelators have been challenged in the turbidity assay and attenuated Aβ1-40 aggregation induced by both Zn2+ and Cu2+ at pH 7.4 and 6.6, respectively. These are outlined in the following sections and include XH1,97 a series of multifunctional 3-hydroxy-4-pyridinones,18,107 and a number of tetrahydrosalen multifunctional ligands,100,101 in addition to representative multidentate chelators such as EDTA and DTPA.   20   Enzyme-Linked Immunosorbent Assay (ELISA) for Aβ1-42 Fibril Binding The Yang group has developed an enzyme-linked immunosorbent assay (ELISA) technique to screen for the association of small molecules with insoluble deposits of aggregated Aβ peptides.108 Synthetic Aβ1-42 peptide is fibrillised by incubation in distilled water, adsorbed onto multi-well spectrophotometry plates and then exposed to putative binding agents. After overnight incubation, visualisation is performed using antibodies specific to the Aβ1-42 peptide; any small molecule binding the fibrils and obscuring the site of antibody binding will give a positive result by this assay.  Solution Fibrillisation Assays Monitored by Fluorescence Conformation-specific ligands to probe peptide aggregation states are in constant development.109 One in vitro method available relies on the binding of thioflavin T to fibrillised Aβ1-42 to screen small molecules for development into Alzheimer’s therapeutics.110 The test compound is added to a solution of Aβ1-42 peptide; after incubation, the extent of Aβ1-42 fibrillisation (or inhibition thereof) is monitored via thioflavin T binding and characteristic changes in fluorescence. This approach has been used recently to test various natural products,111 N-methylated peptides,112 and bis(styrylpyridine)/ bis(styrylbenzene) derivatives113 for inhibition of Aβ1-42 fibrillisation. A similar assay has been put forward using 4,4′-dianilino-1,1′-binaphthyl- 5,5′-disulphonate to probe the aggregation state of Aβ1-40 in solution upon exposure to various small molecules.114   21   In Vitro Studies on Brain-Deposited Amyloid Plaque: Ex Vivo Tissue Assays In this type of experiment, Alzheimer’s-affected post-mortem brain tissue sections are exposed to test chelators which can effect solubilisation of Aβ plaques115 and inhibit Aβ- mediated redox activity.116 Both human AD-affected and transgenic AD-model mouse tissue have been analysed in this way with a number of different chelators.117  In Vitro Neuron Studies Cell study is the most basic way to probe the efficacy of a putative Alzheimer’s therapeutic chelator in a biological environment and is the first line of testing beyond in vitro peptide/metal binding studies. Pre-treatment of synthetic Aβ1-42 fibrils with a chelator before exposure to neurons attenuates their toxicity vs. non-pretreated fibrils.118 This supports the concept that redox-active metal ions are the mediators of amyloid toxicity.  Model Animal Studies Transgenic mouse models have been developed based on the knowledge of genes involved in familial AD and via manipulation of those genes to alter expression of protein deposited in plaques (and NFTs); these models have been recently reviewed.119 Typically the mouse models exhibit the behavioural, biochemical and pathological abnormalities reminiscent of AD. For example, the Tg2576 mouse model displays increased levels of Aβ1-40, Aβ1-42 and Aβ plaques,120 and upon zinc supplementation, shows increased and preferential localisation of zinc within the plaques.53  22   1.3.3 Compounds Tested and/or Designed as Alzheimer’s Disease Therapeutics  A wide range of metal-binding compounds have been designed and tested for use as AD therapeutics based on the metal ion-linked amyloid hypothesis. Some of these were originally developed for treatment of other metal-associated disease conditions and are discussed in the “crossover compound” section, while others were rationally designed as brain-permeating metal binders to target AD pathology and are discussed in the subsequent section; all compounds are summarised in Table 1.1.  23   Table 1.1. Metal binding compounds designed and/or tested for AD therapy. Compound Name Fig State of Development for AD Selected Other Recorded Uses N-Acetylcysteine amide (NACA, AD4) 1.4b In vitro study: rat121 and human cells122 exposed to AD-related biochemical challenge Derivative of NAC, a mucolytic and antidote for acetaminophen overdose Clioquinol 1.9b Clinical trial123 Antifungal, antibacterial Deferiprone (L1) 1.5 In vitro mouse neuron study124 Iron overload therapy Desferrioxamine (DFO) 1.5 Clinical trial91 Iron overload therapy DP-109 1.7 Transgenic mouse study125 Feralex 1.6b Removal of Al3+ from pre-loaded human brain cell nuclei,126 ex vivo AD tissue study127  Glucose-bearing Pyridinones 1.6b Aβ turbidity study, mouse brain uptake study18,107  Glucose-bearing Tetrahydrosalens 1.8 Aβ turbidity studies100,101 Hydroxychloroquine 1.9a Clinical trial128 Anti-malarial, anti- inflammatory bis(8- Hydroxyquinoline) 1.9c In vitro Aβ resolubilisation129 and inhibition of Aβ-Cu2+ redox chemistry130  JKL 169 1.7 Rat study131 Anti-HIV and others PBT2  Clinical trial132 D-Penicillamine (D- Pen) 1.4a Ex vivo AD tissue study,115 nanoparticle conjugation105 Copper overload therapy Triethylenetetraamine (TETA) 1.4a Transgenic mouse study133 Copper overload therapy XH1 1.7 Transgenic mouse study97 24   Crossover Compounds A number of “crossover compounds” have been put forward as possible Alzheimer’s therapeutics, based on their proven utility for therapy of other conditions such as genetic copper overload and thalassemia-related iron overload. D-Penicillamine ((2S)-2-amino-3-methyl-3-sulphanyl-butanoic acid, D-pen, Figure 1.4a), has been used in ex vivo AD plaque resolubilisation studies and can markedly enhance the solubilisation of Aβ.115 In addition, it has been conjugated to nanoparticles in a bid to increase its delivery to the brain;105 however, the applicability of this conjugate has not yet been tested. Triethylenetetraamine (N,N'-bis(2-aminoethyl)ethane-1,2-diamine, TETA, Figure 1.4a) has been used since 1982 for the treatment of genetic copper overload and has been tested for AD application in the transgenic mouse model. Treatment with TETA over twelve weeks had no significant effect on Aβ deposition and was likely too hydrophilic to penetrate the brain and interact with brain metals or plaques directly; at the same time, higher doses of TETA were found to cause significant toxicity in wild-type mice.133  R = OH,     NAC R = NH2,   NACA (AD4) NHO SH R Oa b D-Penicillamine C C C OH O H NH2 HS Triethylenetetraamine H2N H N N H NH2   Figure 1.4. “Crossover compounds” developed for other applications yet tested for AD therapy: (a) copper overload therapeutics D-penicillamine115 and triethylenetetraamine (TETA);133 (b) NAC and NACA (AD4).122 25   N-Acetylcysteine ((R)-2-acetamido-3-sulphanyl-propanoic acid, NAC, Figure 1.4b) is FDA-approved for medicinal use mainly as a mucolytic agent and for treatment for acetaminophen overdose. Being a small thiol-containing compound it is also an effective copper chelator and has antioxidant activity. Its radical scavenging abilities and pharmacokinetics have recently been reviewed.134 N-Acetylcysteine has relatively low bioavailability due to its carboxylate group (negatively charged at physiological conditions) and does not readily permeate membranes by diffusion. However, the analogue N-acetylcysteine amide (NACA, or AD4) has the carboxyl group replaced by an amide group such that the molecule remains uncharged and is thus more able to permeate biological membranes, even penetrating the BBB after oral administration in animals.135 In vitro examination of the antioxidant activities of NAC, NACA and many known antioxidants has been carried out by six different tests with NACA demonstrating superior antioxidant activity vs. NAC.136 Ex vivo studies on human β-thalassemic blood cells have demonstrated the ability of NAC and NACA to attenuate oxidative stress in a biological setting, with NACA being slightly more effective.137 N-Acetylcysteine amide and a few other thiol-containing amide derivatives of oligopeptides prevented protein oxidation, protected rat neurons from in vitro  Aβ1-42 toxicity,121 and NACA successfully relieved the indicators of oxidative stress present in AD fibroblasts.122 Desferrioxamine, (N'-[5-(acetyl-hydroxy-amino)pentyl]-N-[5-[3-(5-aminopentyl- hydroxy-carbamoyl) propanoylamino]pentyl]-N-hydroxy-butane diamide also known as deferoxamine or DFO, Figure 1.5) is a potentially hexadentate hydroxamate trivalent metal ion chelator and is the most widely-used chelator for treatment of iron overload. To investigate its applicability to AD therapy, DFO has been used in Aβ solution studies, neuron culture studies and ex vivo brain tissue experiments. Early ex vivo tissue studies with DFO demonstrated the co- 26   localisation of redox-active iron with plaques.138 Desferrioxamine can provide protection for neurons from metal ion-mediated toxicity of synthetic Aβ,118 inhibit Aβ-mediated redox activity,116 remove Fe3+ from hyperphosphorylated tau,127 and, used in concert with Feralex (vide infra), efficiently remove Al3+ from neurons.126  L1 H NO O N N H O O N OH NH3+ N OH O OH N O OH DFO   Figure 1.5. Multidentate pro-ligands used for treatment of iron-overload and other conditions: DFO, L1.139  Desferrioxamine was the focus of the first clinical attempt to target metal ions for amelioration of neurodegenerative disease; a 1991 study used sustained slow doses of DFO to see if the clinical progression of AD-related dementia could be slowed.91 The drug was able to slow the progression of Alzheimer’s-related dementia and at the time, it was hypothesised that DFO targeted iron or aluminium in the patients; however, neither blood nor CSF metal concentrations were monitored over the course of the trial, and only urinary Al levels were monitored, but not reported. No conclusion was made at the time on DFO’s metal binding action in AD patients, and because DFO has significant affinity for metal ions beside Al3+ and Fe3+ (DFO stability constants, logK, for Fe3+, Al3+, Cu2+, and Zn2+ are 30.6, 22.0, 14.1, and 11.1 respectively)140 it is possible that the drug acted by targeting Fe3+, Cu2+ or Zn2+. Furthermore, it 27   has been suggested (Cuajungco et al.)66 that the study authors verbally reported a decrease in postmortem brain zinc and iron levels after study completion. Deferiprone (3-hydroxy-1,2-dimethyl-4(1H)-pyridinone, L1, Figure 1.5) is approved for use as therapeutic iron chelator for iron overload conditions in Europe and in India.141 Its relative lipophilicity gives it high oral activity and BBB penetration, it is as effective as DFO for iron removal, and while deferiprone appears to mobilise iron primarily from the serum pool, it is an effective binder of other metal ions in the order:141 Fe3+ > Cu2+ > Al3+ > Zn2+ at pH 7.4. Deferiprone has recently been shown to protect mouse cortical neurons exposed to AD-related insults Fe3+, H2O2, Aβ1-40 and to prevent neuronal death.124  Rationally-Designed Multifunctional Molecules Because of deferiprone’s precedent for clinical application, its formation from cheap precursors (i.e. 3-hydroxy-2-methyl-4-pyranone (maltol), an FDA-approved food additive) and its ease of derivatisation, many such derivatives have been made for therapeutic metal ion binding, and some of these have been applied to the treatment of AD. Bebbington et al. imbued the basic structure of deferiprone with antioxidant activity by incorporating butylated hydroxytoluene (BHT, a known antioxidant) into the structure to make “hybrid” metal binding and antioxidant compounds.142 To improve solubility, prodrugs were formed by esterifying the 3-hydroxyl group of the pyridinone moiety with various groups including amino acids (Figure 1.6a).143 Select functionalisations significantly increased the water solubility of hybrid molecules and in vitro ester hydrolysis testing showed good stability in buffer, with limited to complete hydrolysis in rat plasma, demonstrating promise for the prodrug strategy.143 The suitability of 28   these deferiprone analogues for metal binding in an Alzheimer’s disease model has not been assayed.  N O HO NH O O HO HO OOH N O HO O O HO HO OH OH O HO HO NH OH N O HO OH HOG6GP HOGBPP HOG2GP O N HO O O HN O OHOH HO OH Feralex OH N OR O a b   Figure 1.6. (a) “Hybrid” metal binding and antioxidant ligands in prodrug (non-chelating) form incorporating structural features of deferiprone (metal binder) and BHT (antioxidant); R = any of a series of acyl groups.150 (b) Selected 3-hydroxy-4-pyridinone compounds showing activity in various AD-relevant assays: all pendant carbohydrate-bearing compounds (upper) are able to inhibit metal-induced Aβ1-40 aggregation in vitro.144 Feralex (lower), a glucose-bearing deferiprone derivative, was developed for therapeutic metal ion manipulation.145  Other variations on the deferiprone structure have been constructed by incorporating secondary rings and carbohydrate moieties. A number of pyridinone compounds bearing pendant carbohydrate groups were synthesised (Figure 1.6b, upper),146 and though originally designed for 29   Ga3+ and Al3+ complexation, the compounds retain the same metal binding moiety and have in fact shown effective inhibition of Zn2+- and Cu2+-mediated Aβ1-40 aggregation.147 Feralex, also known as Feralex-G (2-deoxy-2-(N-carbamoylmethyl-[N'-2'-methyl-3'-hydroxypyrid-4'-one])-D- glucopyranose, Figure 1.6b, lower) is a glucose-bearing deferiprone derivative.145 In AD brain tissue experiments, Feralex was comparable to DFO in Fe3+ removal from AD-related neurofibrillary tangles.127 In human brain cell cultures exposed to Al3+, Feralex exerted a co- operative effect with DFO for Al3+ removal, possibly by participating in molecular shuttle chelation in which Feralex released Al3+ from within the cells to DFO acting as an extracellular high-affinity metal ion sink.126 DP-109 (1,2-bis(2-aminophenyloxy)ethane-N,N,N',N'-tetraacetic acid, N,N′-bis(2- octadecyloxyethyl)ester; Figure 1.7) is the more lipophilic diester derivative of BAPTA (1,2- bis(2-aminophenyloxy)ethane-N,N,N',N'-tetraacetic acid), a known calcium chelator. DP-109 is the prodrug form of an hexadentate chelator and is designed for oral administration, greater brain penetration, increased residence time in the brain and selective chelation of Zn2+, Cu2+ and Fe3+ within membrane compartments; it demonstrates better chelating efficacy for Cu2+ and Zn2+ than for other divalent metal ions.148 A trial on the Alzheimer’s transgenic mouse model has demonstrated DP-109’s ability to reduce the level of aggregated insoluble Aβ while increasing the level of soluble Aβ forms; tissue staining showed a  reduction in plaque number, density, and percent area in cortical sections as well as reduced zinc content in AD tissue samples.125 Finally, the authors report that the effect of a lower daily dosage of DP-109 was comparable to that of clioquinol (vide infra).125 No cognitive testing was reported, and further study must demonstrate a positive link between these physiological effects and positive neurological effects for further development of DP-109 as a viable candidate for AD therapy. 30    DP-109 OO NNHOOC COOH C18H37O(H2C)2O O O(CH2)2OC18H37O N S H N O N COOH N SH N O N COOH N XH1 COOH NH NH HN N N NH HN HN JKL 169   Figure 1.7. Compounds developed for therapeutic metal ion manipulation in Alzheimer’s disease, prodrug compound DP-109,148 putatively Aβ-associating chelator XH1,97 and bicyclam JKL 169.131  XH1 [(4-benzothiazol-2-yl-phenylcarbamoyl)-methyl]-{2-[(2-{[(4-benzothiazol-2-yl- phenylcarbam-oyl)methyl]-carboxymethyl-amino}-ethyl)-carboxymethyl-amino]-ethyl}-amino)- acetic acid (Figure 1.7) was developed to target metal binding activity to Aβ by covalent linkage of an amyloid-binding functionality (benzothiazole) with a DTPA-like metal ion binding core.149 The result is a relatively lipophilic molecule which, in computations, shows putative binding to the Aβ1-40 peptide and is able to effectively inhibit Zn2+-induced Aβ1-40 aggregation in solution.97 In addition XH1 reduces APP protein expression in human neurons and attenuates amyloid pathology in the brains of APP transgenic mice.97 After the fortuitous discovery of a high-potency HIV inhibitor JM1657,150 bicyclams have been developed for potential HIV therapy and other applications such as stem cell 31   mobilisation. The bicyclam JKL 169 (1,1′-xylyl bis-1,4,8,11 tetraaza cyclotetradecane, Figure 1.7) has been directly compared to clioquinol (below) in rat studies, with both compounds decreasing CSF copper concentration, slightly decreasing serum copper concentration, and JKL 169 significantly increasing copper levels in the brain cortex.131 Thus, JKL 169 is capable of affecting body distribution of copper and may be a candidate for further development into a viable AD therapeutic. A number of tetrahydrosalen compounds have been designed, prepared and evaluated for potential use in AD therapy.100,101 A series has been produced bearing pendant glucose molecules which demonstrate significant antioxidant activity, Zn2+ and Cu2+ coordinating ability and inhibition of metal ion-induced Aβ1-40 aggregation in solution (Figure 1.8a).100  N N HOOH R R OO O OHO HO OH OH HO HO OH OH H2GL1-2 N N RO OR R'R' R'' R'' HN NH RO RO H2L2      R = H GL2       R = glucose H2L1/GL1     R = H/glucose,  R' = H,           R'' = H H2L3/GL3     R = H/glucose,  R' = methyl,   R'' = H H2L4/GL4     R = H/glucose,  R' = benzyl,   R'' = H H2L5/GL5     R = H/glucose,  R' = H,           R'' = t-butyl a b c   Figure 1.8. Tetrahydrosalen pro-ligands (a) with pendant glucose molecules or (b, c) with glucose linkage installed as part of the prodrug concept, masking metal ion binding until activation by enzymatic deglycosylation.  32   A second series of tetrahydrosalen compounds was synthesised utilising the prodrug approach with glucose-masking of the metal binding site, to potentially provide for more brain-specific metal binding (Figure 1.8b, c).101 Antioxidant activity, facile enzymatic deglycosylation and inhibition of metal-ion mediated Aβ1-40 aggregation were demonstrated,101 paving the way for further biological investigation of these compounds for application to AD therapy.  Hydroxyquinoline Derivatives Both chloroquine (N'-(7-chloroquinolin-4-yl)-N,N-diethyl-pentane-1,4-diamine) and its analogue hydroxychloroquine (2-[(4-[(7-chloroquinolin-4-yl)amino]pentyl)- (ethyl)amino]ethanol, both Figure 1.9a) have been used as anti-malarial agents and anti- inflammatory agents. Metal binding agent chloroquine has been shown to inhibit iron uptake into cultured cells151 and into rat tissue;152 however, a double-blind trial of hydroxychloroquine on patients with minimal or mild AD indicated no significant advantage with the drug vs. the control on the rate of cognitive decline or quality of life.128 Clioquinol (5-chloro-7-iodo-8-hydroxyquinoline, CQ, also known as PBT1, Figure 1.9b) was initially approved decades ago for use as an antibiotic. It is an 8-hydroxyquinoline that binds Zn2+ and Cu2+ (with greater affinity than it binds Ca2+ and Mg2+), is hydrophobic and freely crosses the blood-brain barrier.153 Clioquinol has been shown to reduce or prevent the formation of amyloid plaques in the transgenic AD mouse model, and the effect is correlated with improved cognitive activity, with the proposed mechanism of action involving removal of metals from brain amyloid plaques.133 33   N HO N OH N H2N N N NH2 N Cl OH I NCl HN N CH2Ra b c d   Figure 1.9. (a) Chloroquine (R = H) and hydroxychloroquine (R = OH), metal chelating antimalarial agents tested for AD application, (b) clioquinol (5-chloro-7-iodo-8- hydroxyquinoline), and tetradentate quinoline-based chelators (c) bis(3-hydroxyquinoline)164 and (d) bis(8-aminoquinoline).165  In another strain of APP transgenic mouse, CQ supplementation significantly reduced plasma levels of copper, zinc and iron whereas supplementation with Cu2+ and CQ increased cerebral copper; it is thought that CQ’s role as an intracellular Cu2+ transporter is responsible for its effects.154 Although promising, the same authors report reduced survival of another strain of APP transgenic mouse with CQ supplementation in food.154 Radiolabelling experiments with CQ have shown that in vitro, it saturably binds synthetic Aβ precipitated by Zn2+ (Kd=0.45 and 1.40 nm for Aβ1-42 and Aβ1-40, respectively) and localises to the Aβ- and Zn2+-enriched fraction of human ex vivo AD brain homogenates.155 The distribution of CQ differs in the APP transgenic mouse and AD human brain vs. their respective controls; [125I]CQ retention is higher in the AD mouse model and its uptake is significantly more rapid into AD patient brains.155 A 2003 clinical trial showed that CQ can significantly affect Aβ metabolism in AD patients; in cases of severe 34   AD, CQ slows cognitive decline with a concurrent reduction of plasma Aβ1-42 and increase in plasma [Zn2+] and no effect on plasma [Cu2+].123 The drug was overall well-tolerated enough to leave the possibility of human use in the future. Based on these data, CQ’s mechanism of action is thought to be: a) inhibition of Aβ-metal interaction and prevention of Aβ aggregation and associated generation of ROS;133  b) scavenging of redox-active metal ions like Cu2+; c) lowering of cellular Aβ production by increasing cellular levels of metal ions, and elevating levels of matrix metalloproteinase activity to degrade Aβ.156 Because of difficulties encountered in scale-up of CQ synthesis for clinical trials (cited as the presence of a di-iodo-8-hydroxyquinoline impurity), further studies pursuing CQ clinical use have been postponed. Instead, researchers close to the CQ project are now focusing on other analogues, the most promising among them being a compound of undisclosed structure, PBT2.157 This 8-hydroxyquinoline derivative contains no iodine, and thus is not capable of forming the di-iodo impurity. The results of a number of in vitro and in vivo tests on both CQ and PBT2 were recently published and explain the current focus on PBT2 for further clinical testing and development by Prana Biotechnology.158 While both compounds performed well in in vitro assays such as Zn2+-mediated turbidity assays and inhibition of Aβ:Cu redox chemistry, PBT2 exhibited greater membrane permeability than did CQ for increased cellular and brain permeation, and effected no change in tissue levels of metals including copper, zinc, iron or manganese.158 Finally, PBT2 showed better performance than CQ in reduction of soluble brain Aβ and improvement of cognitive functioning in the AD mouse model.158 Furthermore, a phase IIa clinical trial was performed for PBT2 similar to that done on CQ in 2001; PBT2 caused no serious adverse toxicity events, had no effect on serum copper or zinc levels, but significantly reduced the level of the AD biomarker Aβ1-42 in cerebrospinal fluid.132 35   Compound clioquinol analogues have been synthesised with varying linker lengths between and attachment points on the two quinoline moieties.129 This forms a tetradentate pro- ligand which chelates Cu2+ and Zn2+ in a 1:1 manner with much higher apparent affinity than clioquinol,130 and which is significantly more effective than clioquinol in restoring Aβ1-42 solubility after metal ion-induced precipitation, particularly at low metal ion concentrations.129 The bis(8-hydroxyquinoline) compound (Figure 1.9c) is also able to inhibit Aβ1-42:Cu-mediated H2O2 production in vitro.129 Further derivations have generated a bis(8-aminoquinoline) version (Figure 1.9d) which displays extra selectivity for Cu2+, and may thus be particularly useful in mitigating the oxidative stress observed in the AD brain.159 The results from the latest PBT2 trials and others have given rise to the newest theory on metal ion distribution in AD and its treatment: that extracellular Aβ peptide interacts with metal ions (Cu2+, Zn2+) to form oligomers and aggregates and in addition to the related ROS production and oxidative damage to brain tissue, this leads to depletion of intracellular metal ion reserves. It is hypothesised that metal-binding compounds such as CQ or PBT2 bind metal ions from extracellular Aβ aggregates to dissolve the aggregates, then redistribute metals via ionophore action (carriage of charged ions across the cell membrane) to restore depleted intracellular concentrations.158 This can up-regulate matrix metalloprotein expression, which then degrades and clears aggregated extracellular Aβ.156 Based on this hypothesis, the focus of new AD therapeutic development is on small, relatively lipophilic chelators which can enter the brain and capture metal ions from oligomerised and precipitated interstitial Aβ, ideally forming redox-neutral complexes (for reduced neuronal damage) and dissolving oligomeric and aggregated Aβ to facilitate brain clearance. In addition, the complexes should permeate cell membranes to increase cellular copper and zinc concentration. 36   1.4 Imaging Alzheimer’s Disease  A molecular imaging agent is a probe used to visualise, characterise and measure a biological process in a living system. Applied to AD, molecular imaging of Aβ deposition would enable monitoring of disease activity and progression such that the effect of Aβ-modulating drugs (including secretase inhibitors and MPACs) can be quantified. Real-time and non- invasive, the ability to image Aβ in the AD brain is an incredible improvement over current AD diagnostic techniques based on pathological tissue analysis (post-mortem) or at the least CSF analysis (requiring a lumbar puncture, and arguably not conclusively diagnostic of AD progression). Positron emission tomography (PET) using radiolabelled compounds with high affinity for aggregated forms of Aβ found in AD-associated senile plaques is an attractive alternative diagnostic method. Although PET is relatively low-resolution by nature, and so would not afford visualisation of individual amyloid plaques, the technique could serve to quantify regions of the brain showing increased ligand retention and therefore elevated plaque population. The ideal Aβ radiotracer is lipid soluble for BBB penetration, associates with plaques with high affinity and selectivity, displays slow dissociation from the Aβ binding site, is not readily metabolised but rapidly cleared from the blood, and thus able to provide quantitative and reproducible information about the Aβ burden in the brain; finally, the compound must be easily labelled with radiolabels such as 18F, 11C or 123I.160 Since the early 1990s a great many plaque binding agents have been synthesised including larger polyaromatic compounds based on Congo Red,161 benzothiazole compounds based on thioflavin T162-164 (both tissue stains for Aβ) and variations thereof such as benzofurans165 and flavones.166 Other compounds are based on a stilbene structure,167 tissue 37   stains such as acridine orange168 or are themselves novel dyes such as 1,1-dicyano-2-[6- (dimethylamino)naphthalene-2-yl]propene (DDNP).169 Structural modifications were made in the development of new imaging agents to enhance brain uptake by removing ionisable or charged functional groups (Chart 1.2).    Chart 1.2. Structures of selected Aβ imaging agents as derived from starting pharmacophores (a) tissue stain Congo Red,170 (b) tissue stain thioflavin T162,171 and (c) the aminonaphthyl scaffold.169 Site of radiolabelling is indicated by *.  For example, removal of the methyl substituent at the 3-position of benzothiazole in thioflavin T gave rise to 2-[4′-(methylamino)phenyl]benzothiazole (BTA-1) and 2-[4′-(methylamino)phenyl]- a Congo Red b Thioflavin T c Aminonaphthyl S N+ N N N N N N NH2 H2N -O3S SO3- S N NH HO * N I N N * S N NH * HN OH * N F NC CN * FDDNP PIB 6-OH-BTA-1 BTA-1 IMPY SB13 38   6-hydroxy-benzothiazole (6-OH-BTA-1, PIB);171 this approach also directed the development of 6-iodo-2-[4′-N-dimethylamino]phenylimidazo[1,2-a]pyridine (IMPY).162 Chart 1.2 indicates the structures of original tissue stains and chemical scaffolds followed by examples of imaging agents derived from each type. Of all the compounds made only a few have progressed into clinical evaluation involving AD patients; namely, [18F]-FDDNP,169 [11C]-PIB172 and [11C]-SB13170 and IMPY.173 Trials with [18F]-FDDNP were the first to successfully image amyloid in the brains of AD patients; the radiotracer showed greater accumulation and slower clearance from amyloid plaque and NFT- dense regions of the brain; residence time in the AD brain was significantly longer than in control brains.169 Results with Pittsburgh Compound B (PIB, also called 6-OH-BTA-1)172 and SB13170 were similar, with marked retention of radiotracer in regions of the brain known to contain large amounts of amyloid deposits in AD brain. Human trials with IMPY are ongoing, with preliminary data said to support the feasibility of using this imaging agent to distinguish between AD vs. normal pathology.173 In the development of new Aβ imaging agents over the last decade it seems the focus is on smaller benzothiazole or thioflavin derivatives. These compounds are of molecular weight in the range of 300 ± 60 g/mol, are uncharged, contain at least two aromatic rings connected by rotatable single bonds and are functionalised with primary, secondary and tertiary amino groups (displaying varying degrees of methylation). Based on these criteria, the pyridinone compounds in this work were designed to incorporate two aromatic rings joined by a single rotatable bond, and para-amino groups with varied degrees of methylation (Hdapp, Hsapp, Hzapp, Figure 1.10a). The structure of these compounds gives an additional synthetic advantage in the fact that the NH2- and NHCH3-terminated compounds (Hzapp, Hsapp) are amenable to direct C-11 labelling using [11C]CH3OTf  or [11C]CH3I (to produce 39   [11C]-Hsapp and [11C]-Hdapp respectively for biodistribution studies/PET imaging).161 Further structural variation is achieved via alteration of linker length between the two aromatic rings (exemplified by Hppp vs. Hnbp, Figure 1.10b). Finally, two pyridinones were synthesised containing benzothiazole groups in two different orientations (joined by the 6- and the 2- position; Hbt6p and Hbt2p, Figure 1.10c) in a bid to directly mimic the structures of imaging agents such as the BTA-1/PIB series of compounds. It is hoped that this mimicry of amyloid- targeting structure will target the pyridinone functionality (metal ion binding, antioxidant activity) to the site of interest for AD biochemical intervention, namely the amyloid plaques.  S N NH HO NO HO N R' * R NO HO (CH2)n PIB a b NO HO NO HO S N N S c   Figure 1.10. Structural similarity between a pyridinone pro-ligand and an established PET imaging agent for amyloid. Current amyloid imaging agent Pittsburgh Compound B (6-OH- BTA-1, * indicates location of 11C radiolabel);172 (a) 3-hydroxy-4-pyridinone compounds Hdapp (R,R′ = CH3,), Hsapp (R = H, R′ = CH3) and Hzapp (R,R′ = H); (b) two 3-hydroxy-4- pyridinone compounds differing in linker length between aromatic rings: Hppp (n = 0) and Hnbp (n = 1); (c) benzothiazole-bearing 3-hydroxy-4-pyridinones Hbt2p (upper) and Hbt6p (lower). 40   1.5 The Multi-Functionality Approach to Chelator Design  The traditional “one molecule, one target” paradigm for drug discovery has been unsuccessful so far in providing a disease-modifying treatment for AD. The multifactorial nature of the disease and the current lack of a unifying theory for its progression have spurred a growing movement in the field focussing on the development of single compounds able to interact with several molecular targets involved in the neurotoxic cascade of AD. This approach is centred around multifunctional compounds variously labelled “designed multiple ligands”174 or “multi-target-directed ligands”.175 Such multifunctionality (i.e. “one molecule, multiple targets”) offers better hope than a combination of different single-function drugs (i.e. a “cocktail” of different drugs, perhaps on varying schedules) because of obvious issues of patient compliance. This approach is preferable also to multi-compound drug formulations (i.e. “single- pill drug combinations”) because the optimisation of the pharmacokinetics is performed at earlier (and cheaper) stages of drug development, and there is lower risk of unwanted drug interactions in the final multifunctional product vs. mixtures of drugs. Figure 1.11 outlines the multifunctionality built in to these pyridinone pro-ligands which includes (1) glycosylation as part of the prodrug approach to improve BBB penetration of the compound and access to the brain; once the free pro-ligand is exposed by glycosidic bond hydrolysis the active drug includes activities such as (2) metal ion complexation and (3) phenolic H-donation for antioxidant activity; finally, (4) Aβ imaging agent structural mimicry is designed to localise the compound to the site of Aβ deposition.  41     Figure 1.11. Rational design of multifunctional masked pyridinone prodrugs for neurological disease therapy.  1.6 Thesis Overview  Chapter 2 is focussed on organic synthesis: the preparation and characterisation of novel 3-hydroxy-4-pyridinone pro-ligands will be discussed followed by a report on the preparation of the corresponding glycosylated derivatives (i.e. prodrug preparation). The subsequent chapter will focus on the testing of pro-ligand activity in vitro by a number of methods. First, pro-ligand metal ion complexation will be examined via discussion of bis(pyridonato)copper(II) complexes; next, pro-ligand antioxidant capacity, cytotoxicity, ability to interfere with metal ion-induced Aβ aggregation in solution, and binding to synthetic Aβ fibrils will be discussed. The final chapter is devoted to outlining potential avenues for progression of this project.  NO HO O HO OH O OH R protecting group masked metal binding site / antioxidant functionality prodrug linker functionalised aromatic ring system amyloid-targeting structural mimic  42 CHAPTER 2 Multifunctional Pyridinone Chelators – Synthesis and Characterisation  2.1 Introduction  3-Hydroxy-4-pyranones and their equivalent pyridinones are compounds based on a six- membered ring structure and incorporate one heteroatom in the ring (oxygen and nitrogen, respectively); they are further functionalised by hydroxyl and carbonyl groups at the 3- and 4- positions of the ring, respectively (Figure 2.1). Molecules containing 3-hydroxy-4-pyridinones possess antibacterial,176 antimalarial,177 antineoplastic,178 anti-inflammatory,179 analgesic179,180 and oral metal chelating181 activity. The hydroxypyranones are of relatively low toxicity;182,183 for example, 3-hydroxy-2-methyl-4-pyranone (maltol, Figure 2.1a) occurs naturally in products such as cocoa, strawberries, coffee and milk, and is an approved food additive in many countries.  N N+ O OH R O- OH R O O OH a b  Figure 2.1. (a) 3-Hydroxy-2-methyl-4-pyranone, maltol; (b) tautomers of a generalised 3- hydroxy-2-methyl-4-pyridinone.   43 The benzoid mesomer of 3-hydroxy-4-pyridinone (Figure 2.1b) places a high charge density on the carbonyl oxygen, giving these ligands good metal-chelating properties.184 The pyridinones are prepared from the corresponding 3-hydroxy-4-pyranones by reaction of a protected (3-methyloxy- or 3-benzyloxy-) pyranone with any primary amine of choice (R′NH2, wherein R′ is the eventual N-substituent of the pyridinone, Scheme 2.1). The pyranone- to-pyridinone aminolysis reaction involves a Michael-type nucleophilic addition of the primary amine to the heterocyclic ring, followed by a second nucleophilic attack on the other α,β- unsaturated group of the ring and dehydration to produce the unsaturated nitrogen-containing heterocycle (Scheme 2.1, structure VII).  N O R' OR O O OR I O OH OR + NH2 R' O OH OR +NH2 R' O+ OH OR NH R'H N+ OH R' OH OR H N +OH2 R' O OR H II III IV V VI VII -H+ -H2O +H+ -H+   Scheme 2.1. Mechanism of aminolysis reaction converting a generalised pyranone (I) to its pyridinone analogue (VII) in aqueous acidic conditions. R = methyl or benzyl group; R′ = a variously substituted aryl group.   44 The nature of this reaction is such that it may be catalysed under either basic or acidic conditions: resonance or protonation of the carbonyl group can cause the β-carbon in the α,β- unsaturated heterocyclic ring to be electron deficient and therefore more susceptible to nucleophilic attack by the amine (Scheme 2.1, structure II) or basic conditions can render the attacking primary amine more nucleophilic. This synthetic method can be extended to produce a range of pyridinone pro-ligands by altering the primary amine used in the synthesis; in this case, primary amines were chosen so as to imbue the final pyridinones with structures mimicking those of amyloid imaging agents such as PIB.172 It is thought that the incorporation of structural features common to many amyloid imaging agents could help localise the pyridinones to their site of action, the amyloid plaques in the Alzheimer’s affected brain; however, this feature must be tested both in vitro and in vivo to verify its feasibility. Structural features include two aromatic rings joined by a single bond, para-amino-, methylamino- and dimethylamino- substituted phenyl rings, and benzothiazole groups.165,185 Another tenet of the multifunctionality approach to pyridinone design (Section 1.5) is glucose linkage. Glycosylation at the 3-hydroxy position of the pyridinone ring masks metal ion chelating activity as well as antioxidant activity to create a precursor to the active drug, or a prodrug. Glycosylation can increase prodrug solubility186 and may improve its pharmacokinetics preventing its metabolism and excretion; furthermore, it could mitigate the undesired chelation of systemic metal ions. The glycoconjugate approach has been used by various groups to impart elevated brain uptake by exploitation of the glucose transport system in the blood-brain barrier, with glycosylated compounds exhibiting improved uptake into the brain.19,187,188 Once the glycosylated prodrug is enzymatically cleaved by the β-glucosidase enzymes (present in the  45 brain and elsewhere in the body) the desired active form of the drug is released (Figure 2.2).189,190    Figure 2.2. Diagram of prodrug approach: systemic circulation of glycosylated (inactive) pyridinone pro-ligand (prodrug), facilitated diffusion across biological membranes (BBB) into the brain, and β-glucosidase-mediated hydrolysis yielding the active pyridinone metal ion chelator.  brain blood-brain barrier enzymatic hydrolysis β-glycosidase active drug metal binding, anti-oxidant free carbohydrate hexose transporter (i.e., GLUT) inactive prodrug amyloid targeting moiety carbohydrate pyridinone BBB transporter binding circulatory system  46 There are a number of enzymes that cleave β-glycosides in humans,191 including glucocere- brosidase (EC 3.2.1.45) and cytosolic β-glucosidase (EC 3.2.1.21).192 It has previously been established by members of our research group that the β-glycosidic bond between glucose and the 3-hydroxy group of a pyridinone is active toward hydrolysis by the appropriate enzyme, β- glucosidase.193 Specifically, a broad-specificity bacterial enzyme, Agrobacterium faecalis β- glucosidase (Abg) was added to buffered solutions of pyridinone glycosides and the reaction monitored by TLC.193 This assay has been performed on a number of pyridinone compounds including the N-phenyl-bearing pyridinone (Gppp) and others; although the test was preliminary and qualitative, in all cases a significant amount of cleavage was observed after 20-30 minutes at room temperature.193  2.2 Experimental 2.2.1 Materials  All solvents and chemicals (Aldrich, Fisher, Alfa Aesar) were reagent grade and used without further purification. Water was purified using an Elgastat Maxima-HPLC system incorporating reverse osmosis followed by ion-exchange and photooxidation processes (Elga, Bucks, England).    47 2.2.2 Instrumentation   1H NMR and 13C{1H} NMR spectra were recorded at room temperature using a Bruker AV-300 or AV-400 spectrometer, as indicated, at 300.13 (75.48 for 13C NMR) or 400.13 (100.62 for 13C NMR) MHz respectively; spectra were calibrated using residual solvent peaks. Low- resolution mass spectra were obtained using a Bruker Esquire Ion Trap ESI-MS spectrometer. High-resolution mass spectrometry (Micromass LCT instrumentation) and elemental analysis (EA, Fisons EA 1108 instrumentation) for C, H and N were completed by Mr. David Wong at the UBC Chemistry Mass Spectrometry and Microanalysis Services. Melting points were collected using a Mel-Temp apparatus (Electrothermal, Fisher Scientific) and are uncorrected. Semi-preparative high-performance liquid chromatography (HPLC) was performed on a system outfitted with a Waters W600 solvent gradient controller and a Waters 2487 dual wavelength absorbance detector set to monitor absorbance at 254 nm. Sample volumes of 0.3-1.0 mL were run through a Phenomenex precolumn (LUNA 10u C18(2); 60 x 21.20 mm, 10 µm; part number 398332G) before the main XTerraRP column (187µm. 19 x 300 mm, part number 186000630). Solvent flow rate was 10 mL/min with a 45-minute gradient from 100 % H2O (containing 0.1 % trifluoroacetic acid, TFA) to 100 % CH3CN, unless otherwise indicated. X-ray data were collected and processed by Dr. B.O. Patrick at UBC using a Bruker X8 APEX II diffractometer with graphite monochromated Mo-Kα radiation; the structure was solved using the Bruker SAINT software package (SAINT. Version 7.46A. Bruker AXS Inc., Madison, Wisconsin, USA. 1997-2007). All hydrogen atoms in all structures were placed in calculated positions; only the hydroxyl hydrogen in Hnbp was treated differently: in this case the hydrogen was located in a difference map and refined isotropically.  48 2.2.3 Nomenclature Conventions, Abbreviations for 3-Hydroxy-4-Pyridinones  The different forms of the compounds (prodrug, active drug, coordinated to metal ion) are treated separately in this work. O-Glycosylation effectively masks the chelator, thereby preventing metal ion binding. The molecule in this form is referred to as a prodrug because the primary functionalities (metal binding or antioxidant activity) are masked; once the carbohydrate moiety is removed the molecule is referred to as a free pro-ligand, able to bind metals in solution. Only when the pyridinone is coordinating a metal ion is it termed a “ligand”.  Hdapp  R = 4-(N,N-dimethylamino)-                     phenyl Hsapp  R = 4-(N-methylamino)phenyl Hzapp   R = 4-aminophenyl Hbt6p   R = 6-benzothiazolyl Hbt2p   R = 2-benzothiazolyl Hppp    R = phenyl Hnbp    R = benzyl Gdapp  R = 4-(N,N-dimethylamino)-                     phenyl Gsapp  R = 4-(N-methylamino)phenyl Gzapp   R = 4-aminophenyl Gbt6p   R = 6-benzothiazolyl Gbt2p   R = 2-benzothiazolyl Gppp    R = phenyl Gnbp    R = benzyl N O HO R N O O R OHO HO OH OH  N O HO NH2 N O HO N N O HO NH N O HO N S N O HO N S N O HO N O HO Hdapp         Hsapp         Hzapp          Hbt6p          Hbt2p                Hppp           Hnbp   Scheme 2.2. (Upper) Generalised structures and names of all pyridinone pro-ligands (left) and glycosylated prodrugs (right) discussed in this chapter. (Lower) complete structures of all pyridinone pro-ligands.   49 The pyridinones are named according to their state (free pro-ligand vs. glycoconjugate vs. coordinating a metal ion) and based on their particular N-substituent. For example, Hdapp represents the compound in its pro-ligand form (non-glycosylated) and refers to the compound with a dimethylamino phenyl substituent on the pyridinone: Hdapp. Accordingly, the Gbt6p compound is the glycosylated version of the 6-benzothiazolyl-bearing pyridinone, and Cu(nbp)2 is the copper complex consisting of two molecules of deprotonated N-benzyl-substituted pyridinone and a copper(II) ion. All organic compounds are presented in Scheme 2.2.  2.2.4 Pro-Ligand Synthesis  2-Methyl-3-methoxy-4-pyranone, MeMa Based on a previously published method194 maltol (3-hydroxy-2-methyl-4-pyranone, 20.0 g, 0.159 mol) was suspended in 1:1 H2O:MeOH (140 mL); KOH (12.2 g, 0.218 mol) was added, and the mixture kept cool in an ice/water bath. Dimethylsulphate (15 mL, 0.158 mol) was added slowly over a period of 30 min, and after stirring for 24 h the mixture was extracted with CHCl3 (3 x 75 mL); the organic phase was dried and the solvent removed under vacuum to yield MeMa as a brown oil (17.0 g, 0.121 mol, 76 %). 1H NMR (CDCl3, 300.13 MHz): δ 8.04 (d, 3J6,5 = 5.6 Hz, 1H; H6), 6.33 (d, 3J5,6 = 5.7 Hz, 1H; H5), 3.71 (s, 3H; C3-OCH3), 2.27 (s, 3H; C2-CH3).   O O H3CO  50 3-Benzyloxy-2-methyl-4-pyranone, BnMa In a similar manner to that previously published,195 benzyl maltol was prepared by dissolving maltol (3-hydroxy-2-methyl-4-pyranone; 30.3 g, 0.240 mol) in MeOH (200 mL); NaOH (10.08 g, 0.252 mol in 24 mL H2O) was added and the mixture was stirred. Benzyl chloride (BnCl; 32.0 mL, 0.278 mol) was added slowly and the reaction mixture was refluxed (78 °C) for 10 h and monitored via TLC (9:1 EtOAc:MeOH). All solvent was removed under reduced pressure yielding a yellow oil, which was recrystallised from minimal EtOH to yield large colourless crystals (32.68 g, 63 %). 1H NMR (CDCl3, 300.13 MHz): δ 7.53 (d, 3J6,5 = 5.7 Hz, 1H; H6), 7.29 (m, 5H; ArH), 6.30 (d, 3J5,6 = 5.7 Hz, 1H; H5), 5.10 (s, 2H; Bn-CH2), 2.03 (s, 3H; C2-CH3).  3-Benzyloxy-2-methyl-1-(4-dimethylaminophenyl)-4(1H)-pyridinone, Bndapp BnMa (1.00 g, 4.63 mmol) and N,N-dimethyl-1,4-phenylenediamine (0.55 g, 4.0 mmol) were combined with 2:1 H2O:MeOH (15 mL) in a 20-mL thick-walled glass tube with o-ring-sealing polytetrafluoroethylene (PTFE) stopper. The tube was placed in a metal reactor consisting of a solid metal cylindrical block with a solid lid, both approximately 12 cm in diameter. In both cylinders were bored cylindrical wells large enough to accommodate the glass tubes with approximately 1 mL heating bath oil. Both cylinders also contained four holes for screws reaching from the top of the outer lid down approximately four cm into the bottom piece. The bottom and top were fixed together with four screws such that when assembled, the tubes were held in place and the apparatus was explosion proof. This allowed heating of the entire apparatus to high temperatures (> 165 °C). In this preparation, the O O BnO N O BnO N  51 tubes and reactor were assembled and heated to 120-130 °C for 96 h. After cooling, the clear orange supernatant was removed and discarded while the darker brown bottom phase was concentrated to yield a brown solid. Column chromatography (silica, 9:1 acetone:MeOH) afforded the product Bndapp (1.74 g, 51 %). 1H NMR (DMSO-d6, 300.13 MHz): δ 7.51 (d, 3J6,5 = 7.3 Hz, 1H; H6), 7.38 (m, 5H; Bn-ArH),  7.15 (d, 3J  = 8.9 Hz, 2H, ArH), 6.77 (d, 3J  = 8.9 Hz, 2H; ArH), 6.20 (d, 3J6,5 = 7.5 Hz, 1H; H5), 5.07 (s, 2H, Bn-CH2), 2.95 (s, 6H; N(CH3)2), 1.85 (s, 3H; C2-CH3).  3-Hydroxy-2-methyl-1-(4-dimethylaminophenyl)-4(1H)-pyridinone, Hdapp Hydrobromic acid (HBr, 33% in acetic acid; 18 mL) was added to Bndapp (1.3 g, 3.9 mmol) in a round-bottomed flask (RBF, 50 mL). The residue dissolved upon heating to reflux; the mixture was heated and stirred for 35 min during which time a light- coloured precipitate was formed. Ethyl acetate (20 mL) was added upon cooling and the mixture filtered on a frit. The beige solid was redissolved in dilute HCl (pH 2, 300 mL) and brought to pH 5 with dropwise addition of NaOH (1 M). After further cooling to 4 °C, the solid was isolated by filtration and dried to yield Hdapp (0.77 g, 82 % from Bndapp, 40 % from N,N- dimethyl-1,4-benzenediamine). Crystals suitable for X-ray diffraction study were obtained by slow evaporation at 4 °C of a 95 % MeOH/5 % CH2Cl2 solution of Hdapp, mp 252-254 °C. 1H NMR (DMSO-d6, 300.13 MHz): δ 8.18 (d, 3J6,5 = 7.0 Hz, 1H; H6), 7.38 (d, 3J = 8.9 Hz, 2H; ArH),  7.26 (d, 3J5,6 = 7.0 Hz, 1H; H5), 6.92 (d, 3J = 8.9 Hz, 2H; ArH), 3.00 (s, 6H; N(CH3)2), 2.21 (s, 3H; C2-CH3). 13C NMR (DMSO-d6, 75.48 MHz): δ 169.37 (C4), 150.24 (C6), 144.91 (C3), 138.39 (C2), 130.44, 129.30 (ArC), 127.25, 112.08 (ArCH), 110.55 (C5), 39.99 (N(CH3)2), N O HO N  52 13.28 (C2-CH3). Anal. Calcd (found) for C14H16N2O2: C, 68.83 (68.89); H, 6.60 (6.60); N, 11.65 (11.47). Infrared spectrum (cm-1, total reflectance): 3153 (br), 1628, 1583, 1522, 1488, 1295.  3-Methoxy-2-methyl-1-(4-methylaminophenyl)-4(1H)-pyridinone, Mesapp MeMa (3.00 g, 21.4 mmol) and N-methyl-1,4-phenylenediamine dichloride (3.60 g, 18.4 mmol) were combined in 2:3 H2O:MeOH (34 mL) in an RBF (100 mL); after adjusting the pH to 7, the mixture was refluxed with stirring for approximately 30 h. The solvent was removed in vacuo and the residue taken up in CH2Cl2 (100 mL) for extraction with pH 5 H2O (3 x 75 mL). The organic phase was retained and all solvent removed to yield Mesapp as a dark purple residue (3.80 g, 85 %). 1H NMR (DMSO-d6, 300.13 MHz): δ 7.47 (d, 3J6,5 = 7.5 Hz, 1H; H6), 7.10 (d,3 J = 8.8 Hz, 2H; ArH), 6.59 (d,3 J = 8.8 Hz, 2H; ArH), 6.13 (d, 3J5,6 = 7.5 Hz, H5), 6.10 (q, 1H; NHCH3), 3.73 (s, 3H; C3-OCH3), 2.70 (d, 3J = 4.9 Hz, 3H; NHCH3), 1.95 (s, 3H; C2-CH3).  3-Hydroxy-2-methyl-1-(4-methylaminophenyl)-4(1H)-pyridinone, Hsapp Mesapp (3.80 g, 15.6 mmol) was dissolved in CH2Cl2 (90 mL) and stirred at -78 °C (acetone/CO2(s)). Boron tribromide (BBr3, 6.0 mL, 63 mmol) was added dropwise and the resultant mixture was stirred for 45 min after which time the flask was transferred to a -12 °C cooling bath (ethylene glycol/CO2(s)) and stirred an additional 3 h. The mixture was allowed to warm overnight to RT, after which the reaction was quenched by Et2O addition (10 mL) and all solvents were removed under reduced pressure. The resulting purple solid was dissolved in warm pH 1.5 H2O (HCl, 250 mL), affording a purple solution; N O MeO NH N O HO NH  53 neutralisation (to pH 7, dropwise addition of 5 M NaOH) caused black precipitate to form. The solid Hsapp was collected by filtration (3.54 g, 83 % from N-methyl-1,4-phenylenediamine dichloride). Crystals suitable for X-ray diffraction study were obtained by slow evaporation of a 99 % MeOH/1 % H2O solution of Hsapp (RT), mp 198-200 °C. 1H NMR (DMSO-d6, 300.13 MHz): δ 7.45 (d, 3J6,5 = 7.3 Hz, 1H; H6), 7.10 (d, 3J = 8.7 Hz, 2H; ArH), 6.60 (d, 3J = 8.8 Hz, 2H; ArH), 6.16 (d, 3J5,6 = 7.2 Hz, 1H; H5), 6.08 (q, 1H; NHCH3), 2.71 (d, 3J = 5.0 Hz, 3H; NHCH3), 1.96 (s, 3H; C2-CH3). 13C NMR (DMSO-d6, 75.48 MHz): δ 169.31 (C4), 150.09 (C6), 144.87 (C3), 138.44 (C2), 129.99, 129.37 (ArC), 127.31, 111.41 (ArCH), 110.47 (C5), 29.57 (NHCH3), 13.24 (C2-CH3). Anal. Calcd (found) for C13H14N2O2: C, 67.81 (68.10); H, 6.13 (6.14); N, 12.17 (12.17). Infrared spectrum (cm-1, total reflectance): 3417, 3046 (br), 1623, 1573, 1513, 1489, 1297.  NH2 BnMa NO2 Bnpnp Hpnp 5:1 (3 % HCl) :MeOH reflux 10 % Pd/C, H2(g) MeOH    RT 73 h Hzapp O O BnO N O BnO NO2 N O HO NO2 N O HO NH2  Scheme 2.3. Synthesis of Hpnp, Hzapp from parent compound Bnpnp.   54 3-Benzyloxy-2-methyl-1-(4-nitrophenyl)-4(1H)-pyridinone, Bnpnp In a manner adapted from that already published,196 BnMa (2.57 g, 11.9 mmol) was combined with p-nitroaniline (2.50 g, 18.1 mmol) in 5:1 (3% HCl):MeOH (60 mL) and gently refluxed for 73 h. While still hot, the aqueous phase was decanted. The oil-like phase was dried in vacuo and purified by column chromatography (40 % CH3CN/60 % CH2Cl2) to yield Bnpnp (0.78 g, 19 %). This was used in the next step without further purification. ESI-MS(+): 337 (M+H+), 359 (M+Na+).   1-(4-Aminophenyl)-3-hydroxy-2-methyl-4(1H)-pyridinone, Hzapp Bnpnp (0.78 g, 2.3 mmol) was combined with MeOH and H2O (19.6 and 5.0 mL, respectively). The hydrogenation catalyst (10 % Pd on C; 0.4 g) was wetted (0.5 mL H2O), added to the mixture, and a small amount of acid was added (0.06 M HCl, 1 mL). After sealing the flask with a new rubber septum, an H2(g) balloon was attached to provide positive H2(g) pressure for the 22-h reaction time. The reaction was monitored via ESI- MS and once deemed complete, the mixture was filtered through Celite and concentrated in vacuo to yield Hzapp as a pink solid. Recrystallisation from 4:1 MeOH:H2O (200 mL) yielded clear beige crystalline Hzapp (0.24 g, 48 %). Crystals suitable for X-ray diffraction study were obtained by slow evaporation at 4 °C of a 4:1 MeOH:H2O solution of Hzapp, mp 185-187 °C. 1H NMR (DMSO-d6, 300.13 MHz): δ 7.46 (d, 3J6,5 = 6.4 Hz, 1H; H6), 7.01 (d, J = 7.4 Hz, 2H; ArH), 6.63 (d, J = 7.2 Hz, 2H; ArH), 6.17 (d, 3J5,6 = 6.0 Hz, 1H; H5), 3.71 (s, 2H; NH2), 1.95 (s, 3H; C2-CH3). 13C NMR (DMSO-d6, 75.48 MHz): δ 169.34 (C4), 149.34 (C6), 144.91 (C3), 138.42 (C2), 130.12, 129.45 (ArC), 127.30, 113.67 (ArCH), 110.52 (C5), 13.26 (C2-CH3). Anal. N O HO NH2 N O BnO NO2  55 Calcd (found) for Hzapp•0.5H2O•0.5MeOH; C25H30N4O6: C, 62.23 (62.66); H, 6.27 (6.35); N, 11.61 (11.94). Infrared spectrum (cm-1, total reflectance): 3313, 3204 (br), 1622, 1555, 1512, 1487, 1292.  3-Hydroxy-2-methyl-1-(4-nitrophenyl)-4(1H)-pyridinone, Hpnp The title compound was prepared as previously published from BnMa and p- nitroaniline.196 BnMa (6.00 g, 27.8 mmol) was combined with p-nitroaniline (5.83 g, 42.2 mmol) in 5:1 (3% HCl):MeOH (90 mL) and vigorously refluxed for 74 h. While still hot, the aqueous phase was decanted and cooled, while the oily immiscible phase was discarded. Filtration yielded bright yellow solid Hpnp (1.64 g, 24 %). 1H NMR (DMSO-d6, 300.13 MHz): δ 8.39 (d, 3J = 9.1 Hz, 2H; ArH), 7.79 (d, 3J = 9.1 Hz, 2H; ArH), 7.62 (d, 3J6,5 = 7.5 Hz, 1H; H6), 6.26 (d, 3J5,6 = 7.3 Hz, 1H; H5), 2.00 (s, 3H; C2-CH3).  1-(6-Benzothiazolyl)-3-benzyloxy-2-methyl-4(1H)-pyridinone, Bnbt6p BnMa (0.75 g, 3.5 mmol) and 6-aminobenzothiazole (0.45 g, 3.0 mmol) were combined with 5 mL MeOH and 7 mL H2O in a 20-mL thick-walled glass tube with o-ring-sealing PTFE stopper. Heating to 110-126 °C for 96 h afforded a clear yellow supernatant over an orange viscous liquid. The solvent was removed in vacuo and the resultant orange solid dissolved in minimal warm 9:1 acetone:MeOH. A fine white precipitate formed after 2 h at RT; this was isolated by vacuum filtration to yield Bnbt6p (0.24 g, 23 %), mp 193-194 °C. 1H NMR (DMSO-d6, 300.13 MHz): δ 9.55 (s, 1H; benzothiazole-H2), 8.34 (s, 1H; benzothiazole-H7), 8.22 (d, 3J7,8 = 8.2 Hz, 1H; benzothiazole- N O HO NO2 N O BnO N S  56 H5), 7.70 (d, 3J6,5 = 7.5 Hz, 1H; pyrid-H6), 7.58 (d, 3J8,7 = 8.7 Hz, 1H; benzothiazole-H4), 7.38 (m, 5H; Bn-ArH), 7.53 (d, 3J5,6 = 7.5 Hz, 1H; pyrid-H5), 1.90 (s, 3H; C2-CH3). Anal. Calcd (found) for C20H16N2O2S: C, 68.94 (68.83); H, 4.63 (4.86); N, 8.04 (8.03).  1-(6-Benzothiazolyl)-3-hydroxy-2-methyl-4(1H)-pyridinone, Hbt6p Hydrobromic acid (HBr, 33% in acetic acid; 25 mL) was added to Bnbt6p (2.56 g, 7.36 mmol) in an RBF (50 mL). The residue dissolved upon heating to reflux; the mixture was heated and stirred for 35 min during which time a light-coloured precipitate was formed. Ethyl acetate (10 mL) was added upon cooling and the mixture filtered on a frit. The beige solid was redissolved in dilute HCl (pH 1.5, 100 mL) and brought to pH 6 (dropwise addition of 1 M NaOH). After further cooling to 4 °C, the solid was isolated by filtration and dried to yield Hbt6p (1.63 g, 86 %). Crystals suitable for X-ray diffraction were obtained by slow evaporation (RT) of a methanolic solution of Hbt6p/Gbt6p, mp 280-282 °C. 1H NMR (DMSO-d6, 300.13 MHz): δ 9.55 (s, 1H; benzothiazole-H2), 8.36 (s, 1H; benzothiazole-H7), 8.23 (d, 3J7,8 = 8.2 Hz, 1H; benzothiazole-H5), 7.65 (d, 3J6,5 = 7.8 Hz, 1H; pyrid-H6), 7.62 (m, 1H; benzothiazole-H4),  6.24 (d, 3J5,6 = 7.3 Hz, 1H; pyrid-H5), 1.99 (s, 3H; C2-CH3). 13C NMR (DMSO-d6, 75.48 MHz): δ 161.59 (pyrid-C4), 159.49 (C11), 153.54 (pyrid-C6), 149.10 (pyrid-C3), 140.14, 138.94 (benzothiazole-ArC/ArCH), 138.13 (pyrid-C2), 134.54, 124.69, 123.96, 121.14 (benzothiazole-ArC/ArCH), 110.59 (pyrid-C5), 14.48 (C2-CH3). Anal. Calcd (found) for C13H10N2O2S: C, 60.45 (60.70); H, 3.90 (4.04); N, 10.85 (10.78). Infrared spectrum (cm-1, total reflectance): 3043 (br), 1624, 1588, 1556, 1493, 1297.   N O HO N S  57 1-(2-Benzothiazolyl)-3-benzyloxy-2-methyl-4(1H)-pyridinone, Bnbt2p BnMa (1.64 g, 7.58 mmol) and 2-aminobenzothiazole (1.25 g, 8.32 mmol) were combined with 5 mL MeOH and 7 mL H2O in a 20-mL thick-walled glass tube with o-ring-sealing PTFE stopper. Heating to 110-126 °C for 96 h afforded a clear, colourless supernatant over a brown viscous liquid. The entire contents of the tube were transferred to a RBF (100 mL) with MeOH (50 mL), concentrated in vacuo and extracted with CH2Cl2 (3 x 25 mL). The organic fraction was dried (MgSO4) and evaporated to dryness whereupon the mixture was separated by column chromatography (silica, 100 % EtOAc to 9:2 EtOAc:MeOH). White solid Bnbt2p was obtained (0.22 g, 9 %). 1H NMR (DMSO-d6, 300.13 MHz): δ 8.22 (m, 1H; benzothiazole-H7), 8.08 (m, 1H; benzothiazole-H4), 8.01 (d, 3J6,5 = 7.5 Hz, 1H; pyrid-H6), 7.62 (m, 2H; benzothiazole-H5,H6), 7.40 (m, 5H; Bn-ArH), 6.36 (d, 3J6,5 = 7.8 Hz, 1H; pyrid-H5), 5.10 (s, 2H; Bn-CH2), 2.15 (s, 3H; C2-CH3). Anal. Calcd (found) for C20H16N2O2S: C, 68.94 (68.83); H, 4.63 (4.86); N, 8.04 (8.03).  1-(2-Benzothiazolyl)-3-hydroxy-2-methyl-4(1H)-pyridinone, Hbt2p Bnbt2p (0.22 g, 0.65 mmol) was deprotected by reflux with HBr (33% in acetic acid; 2.2 mL) for 35 min during which time a light-coloured precipitate was formed. Ethyl acetate (2 mL) was added upon cooling and the mixture filtered on a frit. The light solid was redissolved in aqueous NaOH (pH 13, 10 mL); HCl (6 M) was added dropwise until the solution reached pH 7, whereupon it was cooled and filtered to yield beige solid Hbt2p (0.12 g; 74 % yield from Bnbt2p, 6 % from 2-aminobenzothiazole). X-ray quality crystals were obtained by slow evaporation of a solution of Hbt2p (2:1 MeOH:CH2Cl2, RT), mp N O BnO N S N O HO N S  58 245-247°C. 1H NMR (DMSO-d6, 300.13 MHz): δ 8.23 (m, 1H; benzothiazole-H7), 8.09 (m, 1H; benzothiazole-H4), 7.98 (d, 3J6,5 = 7.5 Hz, 1H; pyrid-H6), 7.61 (m, 2H; benzothiazole-H5,H6), 6.33 (d, 3J5,6 = 7.5 Hz, 1H; pyrid-H5), 2.26 (s, 3H; C2-CH3). 13C NMR (DMSO-d6, 75.48 MHz): δ 170.90 (pyrid-C4), 160.25 (benzothiazole-C2), 149.15 (pyrid-C6), 145.09 (pyrid-C3), 137.72 (pyrid-C2), 135.01, 127.95 (benzothiazole-ArC), 127.28, 126.57, 123.48, 122.79 (benzothiazole- ArCH), 112.25 (pyrid-C5), 13.04 (C2-CH3). Anal. Calcd (found) for C13H10N2O2S: C, 60.45 (60.34); H, 3.90 (3.88); N, 10.85 (10.82). Infrared spectrum (cm-1, total reflectance): 3091 (br), 1624, 1576, 1508, 1483, 1302.  3-Hydroxy-2-methyl-1-phenyl-4(1H)-pyridinone, Hppp Hppp was prepared by a previously published method.196 1H NMR (DMSO-d6, 300.13 MHz): δ 7.53 (m, 5H; Ph), 7.43 (d, 3J6,5 = 9 Hz, 1H; H6), 6.18 (d, 3J5,6 = 9 Hz, 1H; H5), 1.94 (s, 3H; C2-CH3). 13C NMR (DMSO-d6, 75.48 MHz): δ 171.68 (C4), 143.34 (C6), 139.44 (C3), 133.06 (C2), 131.21, 131.00, 128.10, 112.69 (ArC/ArCH), 104.49 (C5), 13.92 (C2-CH3). Anal. Calcd (found) for C12H11NO2: C, 71.63 (71.66); H, 5.51 (5.56); N, 6.96 (6.96). Infrared spectrum (cm-1, total reflectance): 3200 (br), 1626, 1579, 1537, 1485, 1296.  N O HO  59 O O HO Cl O O O N H O O N H O O N O O N O HO Cl NaOH MeOH 65 °C 16 h NH4OH EtOH 20 °C 16 h NaH THF 45 °C 16 h HBr acetic acid 78 °C 48 h maltol BnMa Bnmpp Bnmpp Bnnbp Hnbp   Scheme 2.4. Synthesis of Hnbp.  3-Benzyloxy-2-methyl-4(1H)-pyridinone, Bnmpp The title compound was prepared by a previously published method.197 1H NMR (DMSO-d6, 300.13 MHz): δ 11.2 (s, 1H; NH), 7.35 (m, 6H; Bn-ArH, H6), 6.10 (d, 1H; H5), 5.04 (s, 2H; Bn-CH2), 2.04 (s, 3H; C2-CH3).  1-Benzyl-3-benzyloxy-2-methyl-4(1H)-pyridinone, Bnnbp Based on a previously published procedure,198 sodium hydride (0.29 g, 1.2 mmol) was added to a suspension of Bnmpp (2.00 g, 9.29 mmol) in THF (30 mL). Benzyl chloride (1.39 mL, 12.1 mmol) was added dropwise and the reaction was stirred for 16 h at 45 °C. Reaction progress was monitored with TLC (9:1 EtOAc:MeOH). Upon completion, the reaction mixture was diluted with EtOAc (20 mL) and washed with 5 % NaCl N O BnO N H O BnO  60 solution and with H2O (25 mL each). The organic phase was dried over MgSO4, solvents were removed under reduced pressure and the product was crystallised from H2O (35 mL). Crystals of Bnnbp  (1.54 g, 54 %) were collected via filtration, washed with cold methanol and dried in vacuo.  1H NMR (DMSO-d6, 300.13 MHz): δ, 7.74 (d, 3J6,5 = 7.4 Hz, 1H; H6), 7.32 (m, 8H; O- CH2ArH, N-CH2Ph[Hyz]), 6.94 (d, 2H; N-CH2Ph[Hx]), 6.22 (d, 3J5,6 = 7.4 Hz, 1H; H5), 5.17 (s, 2H; N-CH2Ph), 5.05 (s, 2H; O-CH2Ph), 1.95 (s, 3H; C2-CH3).  1-Benzyl-3-hydroxy-2-methyl-4(1H)-pyridinone, Hnbp Bnnbp (0.50 g, 1.6 mmol) was deprotected with HBr (33 % in acetic acid; 5 mL). The solution was refluxed for 48 h, allowed to cool and quenched with Et2O (5 mL). Solvent and excess HBr and acetic acid were removed under reduced pressure; the resultant solid was dissolved in H2O (50 mL), brought to pH 12 (dropwise addition of 5 M NaOH), and washed with EtOAc (25 mL). The pH was then adjusted to 6 (dropwise addition of 6 M HCl) such that the product precipitated. Colourless Hnbp (0.28 g, 81 %) was collected via filtration and recrystallised from minimal CHCl3; crystals suitable for X-ray diffraction were obtained from slow evaporation of CHCl3 solution. 1H NMR (DMSO-d6, 300.13 MHz): δ 7.75 (d, 3J6,5 = 7.3 Hz, 1H; H6), 7.33 (m, 3H; N-CH2Ph[Hyz]), 7.07 (d, 2H; N- CH2Ph[Hx]), 6.19 (d, 3J5,6 = 7.2 Hz, 1H; H5), 5.24 (s, 2H; N-CH2Ph), 2.12 (s, 3H; C2-CH3). 13C NMR (DMSO-d6, 100.62 MHz): δ 171.17 (C4), 140.00 (C6), 137.52 (C3), 133.34 (C2), 130.35, 129.36, 127.38 (ArCH/ArC), 112.80 (C5), 104.49 (ArCH/ArC), 58.61 (CH2Ph), 12.30 (C2-CH3). Anal. Calcd (found) for C13H13NO2: C, 72.54 (72.16); H, 6.09 (6.08); N, 6.51 (6.43). Infrared spectrum (cm-1, total reflectance): 3181 (br), 1628, 1576, 1527, 1476, 1262. N O HO  61 2.2.5 Glycosylated Compound (Prodrug) Synthesis  All glycosylated versions of 3-hydroxy-4-pyridinone pro-ligands were synthesised via the substitution reaction in basic conditions as per Scheme 2.5.  HL GL N O HO R N O O R OHO HO OH OH CH2Cl2, KOH/MeOH RT 24 - 72 h OAcO AcO OAc OAc Br   Scheme 2.5. General synthetic route to glycosylated pyridinone prodrugs; a mixture of α- and β- anomers is formed.  3-(β-D-Glucopyranosyloxy)-2-methyl-1-(4-dimethylaminophenyl)-4(1H)-pyridinone, β-Gdapp 2,3,4,6-Tetra-O-acetyl-α-D-glucopyranosyl bromide (0.73 g, 1.8 mmol) was dissolved in dry CH2Cl2 (14 mL). Hdapp (0.34 g, 1.4 mmol) was dissolved in anhydrous KOH/MeOH solution (0.44 M KOH, 7.0 mL MeOH); the pro-ligand solution was added dropwise to the glucose solution. The reaction mixture was stirred (RT, 65 h) and monitored by TLC (alumina, 1:1 iPrOH:H2O). The resultant mixture was neutralised (dropwise addition of 12 M HCl), evaporated to dryness, and the residue taken up in dry MeOH (8 mL) for filtration. Column chromatography N O O N OHO HO OH OH  62 (alumina, 15 % H2O in iPrOH) yielded β-Gdapp (0.10 g, 0.25 mmol, 18 % from Hdapp). 1H NMR (MeOD-d4, 300.13 MHz): δ 8.30 (d, 3J6,5 = 7.1 Hz, 1H; pyrid-H6), 7.31 (overlapping doublets, 3H; ArH, pyrid-H5), 6.95 (d, 3J  = 9.1 Hz, 2H; ArH),  4.99 (d, 3J  = 7.5 Hz, 1H; gluc-H1), 3.86 (dd, 2J6a,6b = 12.0 Hz, 3J6a,5 = 1.5 Hz, 1H; gluc-H6a), 3.68 (dd, 2J6b,6a = 11.7 Hz, 3J6b,5 = 5.1 Hz, 1H; gluc-H6b), 3.47 (m, 4H; gluc-H2,H3,H4,H5), 3.07 (s, 6H; N(CH3)2), 2.47 (s, 3H; C2- CH3). 13C NMR (MeOD-d4, 75.48 MHz): δ 166.41 (pyrid-C4), 153.87 (pyrid-C6), 152.80 (ArC), 145.19 (pyrid-C3), 143.18 (pyrid-C2), 131.75 (ArC), 127.73 (ArCH), 114.32 (pyrid-C5), 114.09 (ArCH), 106.25 (gluc-C1), 78.98, 77.81 (gluc-C3,C5), 75.52 (gluc-C2), 71.18 (gluc-C4), 62.50 (gluc-C6), 40.96 (pyrid-N(CH3)2), 16.64 (pyrid-C2-CH3). HR-ESIMS m/z for C20H26N2O7Na (M+Na+) calcd (found): 429.1638 (429.1640).  3-(β-D-Glucopyranosyloxy)-2-methyl-1-(4-methylaminophenyl)-4(1H)-pyridinone, β-Gsapp 2,3,4,6-Tetra-O-acetyl-α-D-glucopyranosyl bromide (0.36 g, 0.88 mmol) was dissolved in dry CH2Cl2 (7 mL). Hsapp (0.17 g, 0.72 mmol) was dissolved in anhydrous KOH/MeOH solution (0.44 M KOH, 3.5 mL MeOH); the pro-ligand solution was added dropwise to the stirred glucose solution. The reaction mixture was stirred (RT, 70 h) and monitored by TLC (7:2:1 iPrOH:H2O:NH3). The resultant mixture was neutralised (dropwise addition of 12 M HCl), evaporated to dryness, and the residue taken up in dry MeOH (5 mL) for filtration. Semi- preparative HPLC allowed collection of β-Gsapp (0.08 g, 0.2 mmol, 28 % from Hsapp). 1H NMR (MeOD-d4, 400.13 MHz): δ 8.29 (d, 3J6,5 = 7.2 Hz, 1H; pyrid-H6), 7.23 (overlapping doublets, 3H; ArH, pyrid-H5), 6.77 (d, 3J  = 8.9 Hz, 2H; ArH),  4.96 (d, 3J  = 7.9 Hz, 1H; gluc-H1), N O O NH OHO HO OH OH  63 3.86 (dd, 2J6a,6b = 12.0 Hz, 3J6a,5 = 2.1 Hz, 1H; gluc-H6a), 3.67 (dd, 2J6b,6a = 12.0 Hz, 3J6b,5 = 5.5 Hz, 1H; gluc-H6b), 3.57, 3.46, 3.38, 3.34 (m, 4H; gluc-H2,H3,H4,H5), 2.84 (s, 3H; NHCH3), 2.47 (s, 3H; C2-CH3). 13C NMR (DMSO-d6, 100.62 MHz): δ 167.92 (pyrid-C4), 150.67 (pyrid- C6), 148.12 (pyrid-C3), 142.76 (ArC),  142.48 (pyrid-C2), 129.45 (ArCH), 126.92 (ArC), 113.82 (pyrid-C5), 111.57 (ArCH), 105.00 (gluc-C1), 77.52, 76.56 (gluc-C3,C5), 74.04 (gluc-C2), 69.63 (gluc-C4), 60.98 (gluc-C6), 29.60 (pyrid-N(CH3)2), 15.46 (pyrid-C2-CH3). HR-ESIMS m/z for C19H24N2O7Na (M+Na+) calcd (found): 415.1481 (415.1474).  N O HO NO2 Hpnp Gpnp N O O NO2 OHO HO OH OH CH2Cl2, KOH/MeOH     RT  44 h OAcO AcO OAc OAc Br N O O NH2 OHO HO OH OH 10 % Pd/C, H2(g) MeOH    RT 9 h β-Gzappβ-Gpnp N O O NO2 OHO HO OH OH   Scheme 2.6. Synthesis of β-Gzapp via β-Gpnp, circumventing production of multiply- glycosylated biproducts.    64 3-(β-D-Glucopyranosyloxy)-2-methyl-1-(4-nitrophenyl)-4(1H)-pyridinone, β-Gpnp 2,3,4,6-Tetra-O-acetyl-α-D-glucopyranosyl bromide (1.6 g, 3.8 mmol) was dissolved in dry CH2Cl2 (28 mL). Hpnp (0.76 g, 3.1 mmol) was dissolved in anhydrous KOH/MeOH solution (0.44 M KOH, 14 mL); the pro-ligand solution was added dropwise to the stirred glucose solution. The reaction mixture was stirred (RT, 44 h) and monitored by TLC (9:1 EtOAc:MeOH). The resultant mixture was neutralised (dropwise addition of 12 M HCl), evaporated to dryness, and the residue taken up in dry MeOH (5 mL) for filtration. Column chromatography (alumina, 15 % H2O in iPrOH) yielded light yellow solid β-Gpnp (0.26 g, 0.65 mmol, 21 % from Hpnp). 1H NMR (MeOD-d4, 300.13 MHz): δ 8.46 (d, 3J = 9.1 Hz, 2H; ArH), 7.81 (d, 3J6,5 = 7.5 Hz, 1H; pyrid-H6), 7.76 (d, 3J  = 9.0 Hz, 2H; ArH), 6.60 (d, 3J6,5 = 7.5 Hz, 1H; pyrid-H5), 4.77 (d, 3J = 7.5 Hz, 1H; gluc-H1), 3.86 (2 overlapping dd, 2H; gluc-H6a,b), 3.69, 3.65, 3.45 (m, 4H; gluc- H2,H3,H4,H5), 2.29 (s, 3H; C2-CH3). 13C NMR (MeOD-d4, 75.48 MHz): δ 175.38 (pyrid-C4), 149.98 (pyrid-C6), 147.70 (pyrid-C3), 146.22 (pyrid-C2), 141.90 (ArC), 129.94, 126.48 (ArCH), 117.39 (pyrid-C5), 106.95 (ArC), 105.56 (gluc-C1), 78.56, 78.10 (gluc-C3,C5), 75.66 (gluc-C2), 71.25 (gluc-C4), 62.88 (gluc-C6), 15.90 (pyrid-C2-CH3). HR-ESIMS m/z for C18H20N2O9Na (M+Na+) calcd (found): 431.1067 (431.1077).   N O O NO2 OHO HO OH OH  65 1-(4-Aminophenyl)-3-(β-D-glucopyranosyloxy)-2-methyl-4(1H)-pyridinone, β-Gzapp β-Gpnp (0.02 g, 0.05 mmol) was dissolved in MeOH (1.2 mL). Pd (10 % on activated charcoal, 0.05 g) was first wetted (H2O, 0.5 mL) and added to the reaction flask. The reaction mixture was stirred under positive H2(g) pressure for 9 h, whereupon the reaction mixture was filtered and evaporated to dryness in vacuo. β-Gzapp was obtained as a clear beige glass (0.010 g, 54 % from β-Gpnp). 1H NMR (DMSO-d6, 400.13 MHz): δ 7.81 (d, 3J6,5 = 7.5 Hz, 1H; pyrid-H6), 7.10 (d, 3J = 8.2 Hz, 2H; ArH), 6.68 (d, 3J = 8.5 Hz, 2H; ArH), 6.53 (d, 3J5,6 = 7.5 Hz, 1H; pyrid-H5), 4.56 (d, 3J  = 7.2 Hz, 1H; gluc-H1), 3.65 (d, 2J6a,6b = 10.2 Hz, 1H; gluc-H6a), 3.44 (dd, 2J6b,6a = 12.0 Hz, 3J6b,5 = 5.5 Hz, 1H; gluc-H6b), 3.20, 3.12 (m, 4H; gluc-H2,H3,H4,H5), 2.17 (s, 3H; C2- CH3). 13C NMR (DMSO-d6, 100.62 MHz): δ 170.69 (pyrid-C4), 145.50 (pyrid-C6), 143.66 (pyrid-C3), 141.95 (pyrid-C2), 130.02, 127.15, 125.68, (2ArC, 1ArCH), 114.73 (pyrid-C5), 114.06 (ArCH), 105.96 (gluc-C1), 77.46, 76.78 (gluc-C3,C5), 74.03 (gluc-C2), 69.53 (gluc-C4), 60.99 (gluc-C6), 15.03 (pyrid-C2-CH3). HR-ESIMS m/z for C18H22N2O7Na (M+Na+) calcd (found): 401.1325 (401.1318).  1-(6-Benzothiazolyl)-3-(β-D-glucopyranosyloxy)-2-methyl-4(1H)-pyridinone, β-Gbt6p 2,3,4,6-Tetra-O-acetyl-α-D-glucopyranosyl bromide (0.72 g, 1.7 mmol) was dissolved in dry CH2Cl2 (14 mL). Hbt6p (0.37 g, 1.4 mmol) was dissolved in anhydrous KOH/MeOH solution (0.22 M KOH, 6.8 mL); the pro-ligand solution was added dropwise to the stirred glucose solution. The reaction mixture was stirred (RT, 72 h) and monitored by TLC (10 % MeOH in N O O NH2 OHO HO OH OH N O O OHO HO OH OH N S  66 acetone). The resultant mixture was neutralised (dropwise addition of 12 M HCl), evaporated to dryness, the residue taken up in dry MeOH (5 mL) and filtered. Semi-preparative HPLC (100 % H2O to 100 % ACN over 45 min) allowed collection of β-Gbt6p (0.07 g, 12 % from Hbt6p). 1H NMR (DMSO-d6, 400.13 MHz): δ 9.59 (s, 1H; benzothiazole-H2), 8.43 (s, 1H; benzothiazole- H7), 8.27 (d, 3J = 8.9 Hz, 1H; benzothiazole-H5), 8.04 (d, 3J6,5 = 7.2 Hz, 1H; pyrid-H6), 7.70 (d, 3J = 8.8 Hz, 1H; benzothiazole-H4), 6.66 (d, 3J5,6 = 7.2 Hz, 1H; pyrid-H5), 4.65 (d, 3J = 7.2 Hz, 1H; gluc-H1), 3.66 (d, 2J6a,6b = 11.3 Hz, 1H; gluc-H6a), 3.46 (dd, 2J6b,6a = 11.3 Hz, 3J6b,5 = 5.4 Hz, 1H; gluc-H6b), 3.24, 3.15 (m, 4H; gluc-H2,H3,H4,H5), 2.21 (s, 3H; C2-CH3). 13C NMR (DMSO-d6, 100.62 MHz): δ 170.89 (pyrid-C4), 159.21 (benzothiazole-C2), 153.28 (pyrid-C6), 145.35 (pyrid-C3), 143.59 (pyrid-C2), 141.71, 138.31, 134.57, 125.26, 123.89, 121.59 (benzothiazole-ArC/ArCH), 114.88 (pyrid-C5), 105.76 (gluc-C1), 77.52, 76.76 (gluc-C3,C5), 74.09 (gluc-C2), 69.56 (gluc-C4), 61.00 (gluc-C6), 15.26 (pyrid-C2-CH3). HR-ESIMS m/z for C19H20N2O7SNa (M+Na+) calcd (found): 443.0889 (443.0895).  1-(2-Benzothiazolyl)-3-(β-D-glucopyranosyloxy)-2-methyl-4(1H)-pyridinone, β-Gbt2p 2,3,4,6-Tetra-O-acetyl-α-D-glucopyranosyl bromide (0.78 g, 1.9 mmol) was dissolved in dry CH2Cl2 (14 mL). Hbt2p (0.40 g, 1.5 mmol) was dissolved in an anhydrous KOH/MeOH solution (0.44 M KOH, 6.7 mL); the pro-ligand solution was added dropwise to the stirred glucose solution. The reaction mixture was stirred (RT, 24 h) and monitored by TLC (9:1 EtOAc:MeOH). The resultant mixture was neutralised (dropwise addition of 12 M HCl), evaporated to dryness and the residue taken up in dry MeOH (5 mL) for filtration. Semi- preparative HPLC (100 % H2O to 100 % ACN over 45 min) allowed collection of β-Gbt2p N O O OHO HO OH OH N S  67 (0.15 g, 24 % from Hbt2p). 1H NMR (MeOD-d4, 400.13 MHz): δ 8.28 (d, 3J6,5 = 7.5 Hz, 1H; pyrid-H6), 8.10 (overlapping doublets, 2H; benzothiazole-H4,H7), 7.62 (m, 2H; benzothiazole- H5,H6), 6.88 (d, 3J5,6 = 7.5 Hz, 1H; pyrid-H5), 4.90 (d, 3J = 7.5 Hz, 1H; gluc-H1), 3.86 (dd, 2J6a,6b = 12.0 Hz, 3J6a,5 = 2.1 Hz, 1H; gluc-H6a), 3.69 (dd, 2J6b,6a = 12.0 Hz, 3J6b,5 = 5.5 Hz, 1H; gluc-H6b), 3.50, 3.39, 3.34 (m, 4H; gluc-H2,H3,H4,H5), 2.54 (s, 3H; C2-CH3). 13C NMR (DMSO-d6, 100.62 MHz): δ 173.34 (pyrid-C4), 159.64 (benzothiazole-C2), 148.97 (pyrid-C6), 144.07 (pyrid-C3), 142.87 (pyrid-C2), 140.24, 135.38, 127.43, 126.91, 123.69, 122.94 (benzothiazole-ArC/ArCH), 116.13 (pyrid-C5), 105.64 (gluc-C1), 77.50, 76.74 (gluc-C3,C5), 74.10 (gluc-C2), 69.60 (gluc-C4), 61.05 (gluc-C6), 14.50 (pyrid-C2-CH3). HR-ESIMS m/z for C19H20N2O7SNa (M+Na+) calcd (found): 443.0889 (443.0897).  2.3 Results and Discussion  2.3.1 Pro-Ligand Preparation  In general, pyridinone pro-ligands are prepared in three steps: first the 3-hydroxy functional group is protected as either a methyl- or benzyl-ether generating 3-methoxy- or 3- benzyloxy-2-methyl-4-pyranone (Scheme 2.7); this protected pyranone is reacted with the appropriate primary amine and finally the protecting group is removed to yield the free pro- ligand target molecule (Scheme 2.7).   68 O O RO NH2 R' N O RO R' N O HO R' R = Me, MeL R = Bn, BnL HL R = Me, MeMa R = Bn, BnMa cba O O HO maltol   Scheme 2.7. General synthetic route for 3-hydroxy-2-methyl-4-pyridinone pro-ligands. For preparation of ether-protected pyranones: For preparation of Hsapp: R = Me, b = 2:3 H2O:MeOH, 100 °C, 30 h; c = BBr3, CH2Cl2, -78 to  -12 °C, 8 h. For preparation of Hdapp, Hbt6p, Hbt2p, Hppp: R = Bn, b = 2:1 H2O:MeOH, 110-130 °C, 96 h; c = HBr, acetic acid, 78 °C, 35 min.  All 3-hydroxy-2-methyl-4-pyridinone pro-ligands were synthesised via the 3-benzyloxy- or 3- methoxy-ether protected pyranone (benzyl maltol, BnMa or methyl maltol, MeMa, respectively); the syntheses of these precursors were based on, but are variations of, previously published methods.197,199 The use of two different protecting groups was due in part to solubility differences between the methyl vs. benzyl pyridinone products, wherein the 3-benzyloxy-2- methyl-4-pyridinone was in some cases more soluble in organic solvents and thus easier to purify by extraction. In the case of the Hzapp synthesis, the benzyl ether route was preferable, as hydrogenation in the presence of palladium on carbon was a planned part of the synthesis of this molecule to reduce the p-nitrophenyl N-substituent to a p-aminophenyl group after glycosylation. Because the benzyl ether is also removed under such reducing conditions, this obviated a separate deprotection step (with BBr3) as would be required for removal of the methyl ether group. In some cases, however, the methyl protection group proved to be more convenient, as in  69 the preparation of Hsapp. 3-Methoxy-2-methyl-1-(4-methylaminophenyl)-4(1H)-pyridinone (Mesapp) was readily prepared by reflux in a round-bottomed flask which was more convenient than preparation in the high temperature/high pressure “bomb” apparatus. The reaction could be performed on a larger scale, and unlike the benzyl protecting group, the 3-methoxy group was stable under these conditions, allowing easier isolation of the ether-protected pyridinone from the reaction mixture. Although in theory the conversion of pyranone to pyridinone can be achieved without any protection at the 3-hydroxyl group,200 this reaction tends to be extremely low-yielding. Purification of HL compounds was achieved using various methods; usually column chromatography was applied when compounds were in the ether-protected form. The increased lipophilicity simplified silica gel purification in terms of more optimal elution time, better resolution from other mixture components and prevention of adventitious metal ion chelation from the silica. After deprotection the HL compounds were isolated by repeated crystallisation/reprecipitation from aqueous solution. This allowed characterisation of all HL compounds by NMR and EA. Infrared spectroscopic characterisation was performed and these data were used primarily in comparison with those from the corresponding copper complexes (Chapter 3). Pyridinone 1H- and 13C NMR spectra were readily assigned by comparison to previous work done on similar compounds146,196 and by comparison to spectra of starting materials or simpler structural analogues. Spectra consist of both aromatic and alkyl signals in discrete regions; these arise from the pyridinone and other aryl groups (between 6-8 ppm) and, separately, methyl groups on both the pyridinone rings and N-substituents (at approximately 2.1 ± 0.2 ppm). A representative 1H NMR spectrum is provided (vide infra, in comparison of  70 Hsapp, Gsapp spectra, Figure 2.4), and a representative 13C NMR spectrum is provided (for Hsapp, Figure A.12 in the Appendix).  2.3.2 X-Ray Diffraction Structural Characterisation of Pro-Ligands  X-ray quality crystals of six of seven pro-ligands (all except for Hppp) were grown by slow evaporation of rather dilute solutions of the compounds, sometimes in the presence of significant amounts of other components such as starting materials and reaction side products. The ORTEP representations of each pro-ligand solid-state structure are given below (Figure 2.3) followed by a general discussion of the average representation of this set of compounds. The appendix contains a brief outline of the crystallisation procedures and nature of the crystal for each compound. These are followed by the crystal data and, separately, selected bond lengths and angles in tabulated form (Tables A.1, A.2-3, respectively; Appendix).   71   Figure 2.3. Ellipsoid plots (50 % probability; for clarity, H-atoms not shown) of Hdapp, Hsapp, Hzapp, Hbt6p, Hbt2p and Hnbp.  The observed bond lengths and angles for the six compounds Hdapp, Hsapp, Hzapp, Hbt6p, Hbt2p and Hnbp (Appendix, Tables A.2, A.3) are typical of 3-hydroxy-4-pyridinones for which solid-state structures have been reported by our group and others.146,201-203 A set of average pyridinone ring bond lengths was calculated from the six sets of structural data (illustrated in Scheme 2.8). The C(4)-O(1) distance is shorter at 1.27 Å and falls within the expected range for a carbon-oxygen double bond, while the C(3)-O(2) distance is longer at an average 1.36 Å and is Hdapp Hsapp Hzapp Hbt6p Hbt2p Hnbp  72 more typical of a carbon-oxygen single bond. Furthermore, it is apparent from the shorter C(2)- C(3) and C(5)-C(6) distances (1.37 and 1.36 Å respectively) compared to the other longer C-C distances in the pyridinone ring (up to 1.44 Å) that the typical placement of unsaturation between C(2)-C(3) and C(5)-C(6) is represented in these compounds. These parameters are in line with the fact that these are the free pyridinone pro-ligands, and are not in the benzoid mesomer as would be expected if the carbonyl group were protonated or if the ligand were bound to a metal ion.     Atoms Distance (Å) Standard Deviation C(4)-O(1) 1.27 0.01 C(3)-O(2) 1.36 0.01 N(1)-C(2) 1.39 0.01 C(2)-C(3) 1.37 0.01 C(3)-C(4) 1.44 0.01 C(4)-C(5) 1.42 0.01 C(5)-C(6) 1.36 0.01                      a  b Scheme 2.8. (a) Average structure and (b) average bond distances derived from six available data sets of pyridinone solid state structures; all distances labelled in angstroms (Å).  C5 C6 N1 C2 C3 C4 O1 O2 C1 Ar 1.36 1.37 1.27 1.36  73 All rings in all compounds are very nearly planar, and all dihedral angles between the pyridinone ring and the phenyl ring (for Hdapp, Hsapp, Hzapp) are between 15 and 20 degrees off of perpendicular. The solid state structure of Hbt6p is similar, with a dihedral angle between the pyridinone and benzothiazole ring falling within the aforementioned range (approximately 73 °); in contrast, the dihedral angle in Hbt2p is closer to 124 degrees. Because the dihedral angle is representative only of the solid-state structure and this information is not applicable to the pro- ligand orientation in solution, the X-ray diffraction data serve primarily as unambiguous structural, but not functional, characterisation of the pro-ligands.  2.3.3 Glycosylated Prodrug Preparation  Although some work has been done with pyridinone formation from O-glycosylated pyranone precursors,204 there is no precedent for O-glycosylation of 3-hydroxy-4-pyridinones outside of the Orvig group. Preliminary work by Dr. David Green resulted in O-glycosylated analogues of four different pyridinones.193 There are a few established methods to perform this functionalisation: the first, and rather well-known method is the Mitsunobu reaction.205 This method utilises a trialkylphosphine and a diesterazo compound (such as 1,1'-(azodicarbonyl)- dipiperidine, ADDP) to catalyse the substitution of an hydroxyl group via an SN2 mechanism. The reaction is driven by formation of phosphorous-oxygen bonds and resulting in inversion of the hydroxyl-bearing stereocentre, if present. This process has been successfully used by other groups for N- and O-carbohydrate linkage of molecules such as amino acids and aryl alcohols,206,207 and was attempted in this work with Hdapp. Tributylphosphine (PBu3), ADDP  74 and 2,3,4,6-tetra-O-acetyl-α-D-glucopyranosyl bromide were combined under dry conditions in CH2Cl2. These attempts resulted in poor yields and, because the Mitsunobu method requires additional reagents and synthesis steps vs. other available methods, it was not employed. Another method to effect glycosylation uses Koenigs-Knorr reaction conditions, requiring the addition of silver or mercury salts or boron trifluoride.208 This reaction has been attempted by other investigators for glycosylation of 3-hydroxy-4-pyranones, however was met with little success.193,208 Because of the metal ion-chelating activity of 3-hydroxy-4-pyridinones, the use of such metal salts or boron reagents is not ideal. In fact, stable pyridinone complexes of boron have been previously prepared (in the Orvig group);209 thus, this method was not employed. A further method available for glycosylation of pyridinones is a biphasic reaction involving CHCl3 and aqueous NaOH; this has been used in the past (in the Orvig group) with some success,18 but was met with solubility difficulties when applied to the current library of pyridinones, and so was not pursued. The method used in this work required significant variation from previous (Orvig group) work, and involved deprotonation of the pyridinone in dry methanolic KOH and addition to 2,3,4,6-tetra-O-acetyl-α-D-glucopyranosyl bromide. This method yielded consistent results for O-glycosylation of all pyridinones; in addition, the basic conditions served to remove the carbohydrate acetyl protection groups, obviating one step in the synthesis. The general synthetic scheme was modified for the preparation of 1-(4-aminophenyl)- 3-(β-D-glucopyranosyloxy)-2-methyl-4(1H)-pyridinone (Gzapp). In this case, a precursor, 3- hydroxy-2-methyl-1-(4-nitrophenyl)-4(1H)-pyridinone (Hpnp) was glycosylated instead; the presence of a nitrosyl functional group allowed glycosylation to occur at the 3-hydroxyl position  75 of the pyridinone without addition of a glycosyl group on the primary amino group of Hzapp. This modified route is outlined in Scheme 2.6. All five of the pyridinone compounds directly involved in the multifunctional Alzheimer’s therapeutic development project were glycosylated in this work: Hdapp, Hsapp, Hzapp, Hbt6p and Hbt2p. The remaining two compounds Hppp and Hnbp were characterised as part of a smaller side-project within the Alzheimer’s therapeutic development programme to probe the effect of inter-ring linker length on in vitro activity. While Hppp, (N-phenyl- substituted) was previously glycosylated and characterised by Dr. David Green,18 its partner compound Hnbp (N-benzyl-substituted) was not as part of this project. The progression of glycosylation reactions was easily monitored by TLC. While the free pyridinone pro-ligands (HL) tend to exhibit extremely low retention factors (Rf), the glycosylated analogues tend to elute more readily (higher Rf). Fortunately, the two compounds are readily distinguishable on TLC with the use of charring. A developed TLC plate is dipped in a solution of 5 % H2SO4 in EtOH. After drying, heat is applied to the plate and any compounds containing carbohydrate moieties are visible as black areas on the silica. Purification of the glycosylated pyridinone prodrugs was significantly more difficult than was the purification of the free pyridinones or their precursors. Due to the similar retention times of the pyridinone pro- ligand (HL), the glycosylated product (GL) and free glucose side-product, column chromatography was ineffective for full separation of compounds. In particular, separation of GL and glucose side product (hydrolysed and deprotected) was incomplete by column chromatography; however, the smaller scale of the reactions allowed purification of the mixtures by semi-preparative HPLC. A representative HPLC trace (of β-Gbt2p purification) is included in the appendix as Figure A.7.  76 2.3.4 1H and 13C NMR of Prodrugs  The glycosylation of the pyridinone compounds is monitored both by TLC and low- resolution mass spectrometry. Only once the compounds had been purified by either HPLC alone or a combination of column chromatography/HPLC were they characterised by 1H NMR. The presence of multiple carbohydrate peaks in the 2.8-5.2 ppm-region of any 1H NMR spectrum prevents full assignment of a mixture. Although the glycosylation reaction produces a mixture of both β- and α-anomeric products,210 it is typical for the β-anomer to elute first from columns,146 such that any sample pure enough for 1H NMR in this project contained only the β-anomer. This is readily observable on the 1H NMR spectra for the glycosylated products: the doublet signal assigned to glucose H1 (anomeric protons) appears at approximately 4.6-5.0 ppm with a coupling constant of 7.5 Hz (characteristic of β-anomeric protons), while the corresponding α- anomeric signal was absent where expected about 0.5 ppm further downfield, with a coupling constant of less than 4 Hz.145,193   77   Figure 2.4. 1H NMR (MeOD-d4, 400.13 MHz, RT) spectra of Hsapp (upper) and β-Gsapp (lower), from 9.0 – 2.0 ppm; s indicates solvent peaks.  GSAPP MEOD_001000FID 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 Chemical Shift (ppm) 0 0.005 0.010 0.015 0.020 0.025 0.030 0.035 0.040 0.045 0.050 0.055 0.060 0.065 0.070 0.075 0.080 0.085 0.090 0.095 0.100 0.105 No rm a liz e d In te n si ty 2.53.01.80.91.00.90.91.12.03.01.0 8. 29 8. 28 7. 24 7. 23 7. 21 6. 78 6. 76 4. 97 4. 95 4. 93 3. 87 3. 87 3. 84 3. 70 3. 55 3. 53 3. 46 3. 37 3. 35 3. 32 3. 31 3. 31 3. 31 3. 30 2. 84 2. 47 2. 38 090205-HSAPP MEOD_001000FID 8.5 8.0 7.5 7.0 6.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 Chemical Shift (ppm) 0 0.01 0.02 0.03 0.04 0.05 0.06 0.07 0.08 0.09 0.10 0.11 0.12 0.13 0.14 0.15 0.16 No rm a liz e d In te n sit y 3.03.01.02.02.01.0 7. 55 7. 54 7. 09 7. 07 6. 70 6. 68 6. 45 6. 43 4. 86 3. 32 3. 31 3. 31 3. 31 3. 30 2. 81 2. 12 9.0 2.0 Chemical Shift (ppm) N O HO NH N O O NH OHO HO OH OH s s s s  78 2.4 Conclusions  Pyridinones are some of the leading small molecule candidates for therapeutic metal ion manipulation in the treatment of conditions such as thalassaemia-related iron overload, malaria, and cancer.211-213 This diversity in application of the 3-hydroxy-4-pyridinone family of compounds is not a coincidence; the pyridinone scaffold is an excellent starting point for rational drug development. The structure is amenable to functionalisation by N-substituent variation, which can alter the physical properties of the drug without significant alteration of metal ion binding.214 Lipophilicity can be modified for alteration of drug bioavailability, toxicity or biodistribution. This structural freedom may be exploited by incorporating known pharmacophores, as in the case of the attachment of benzothiazole or p-aminophenyl groups for amyloid targeting. The prodrug approach may be applied to alter or target drug activity to the tissue of interest. Both of these opportunities are exploited in this work: a series of pyridinone compounds was prepared with N-substituents chosen such that the overall pyridinone structure mimics that of known amyloid imaging agents. A total of six new pro-ligands were prepared and fully characterised. Purification of all HL compounds was achieved using silica gel chromatography, crystallisation, and/or pH-trituration from aqueous solution, and all structures were confirmed with both 1H and 13C NMR, IR, EA and single crystal X-ray crystallography (for all except Hppp). To demonstrate the feasibility of the prodrug approach, the glycosylated analogues of five of the pro-ligands were synthesised and characterised. Structures of these 1-aryl-3-(β-D-  79 glucopyranosyloxy)-2-methyl-4(1H)-pyridinones were confirmed with 1H and 13C NMR and high-resolution mass spectrometry. The target drug structures are attainable; the next phase in this applied medicinal chemistry project is to assay compound activity in Alzheimer’s-relevant in vitro assays, and determine which of these variations imbue the compounds with the intended biological activity.   80 CHAPTER 3 In Vitro Characterisation of Multifunctional Pro- Ligands as Alzheimer’s Disease Therapeutics  3.1 Introduction  A series of new 3-hydroxy-4-pyridinone pro-ligands has been prepared under the guiding principle of producing multifunctional Alzheimer’s disease (AD) therapeutics; the next step in compound development is testing each aspect of their function in vitro. Metal ion binding may be verified and characterised by preparation of coordination complexes. Bis(pyridonato)- copper(II) complexes can be prepared and structurally characterised to verify that the pyridinones chelate biologically relevant metal ions as expected. Antioxidant activity is readily examined using a colourimetric test, the Trolox equivalent antioxidant capacity (TEAC) assay.215 The pyridinone moiety can donate phenolic hydrogen atoms in the same manner as does α- tocopherol to effect radical quenching;216 relative values for antioxidant activity can be gathered and normalised to that of Trolox (Figure 3.1).215 Cytotoxicity can be determined in vitro using cultured human cells,217 and although the cell-based cytotoxicity assay cannot assay overall pharmacokinetics of the test compounds including bioavailability/absorption, distribution, metabolism or excretion by a whole organism, it is a cost-effective and relatively easy way to enable compound screening at an early stage of drug development. Finally, after the functions of metal binding, antioxidant activity and low toxicity have been confirmed, the next step is to characterise the applicability of the pyridinones to AD; this can be done in vitro using conditions  81 mimicking those of AD. The intended location for action of these compounds is the amyloid plaque in the AD brain. An enzyme-linked immunosorbent assay (ELISA) can be utilised to compare the tendency of the compounds to associate with fibrils with that of thioflavin T, an amyloid plaque histochemical stain.108  OH O HO COOH O HO BHT Trolox α-tocopherol (vitamin E) formazan (purple) MTT tetrazolium salt (yellow) ABTS, diammonium salt EDTA DTPA N S-O3S N N S N SO3-+NH4 NH4+ N HO O OH O N OH O OH O N OH O HO ON O HO NO OH O OH N N N N S N Br - N N H N N S N thioflavin T S N+ N   Figure 3.1. Structures of reagents used in Chapter 3.   82 Because Aβ fibrils are thought to have a different structure than simple aggregates,78 it is useful to probe their interaction with Aβ aggregates in addition to their association with fibrils. Specifically, the aggregation of Aβ peptide in the AD-affected brain is considered to be central to disease progression, and associated with elevated concentrations of metal ions such as Cu2+ and Zn2+.218 A turbidity assay uses human synthetic Aβ40 aggregated by exposure to these metal ions, and each test chelator is examined for its effect on aggregation.97,107 Although the metal ion-coordinating functional groups of the pyridinones is conserved across the series, it is possible that variation in other structural features may affect compound interaction with Aβ and disaggregation efficacy. This chapter focuses solely on the functional characterisation of the series of pyridinone pro-ligands to determine which of the intended functionalities are present, and, where assays are possible on a range of pro-ligands, to identify which structural features generate increased efficacy.  3.2 Experimental 3.2.1 Materials  All solvents and chemicals (Aldrich, Fisher, Alfa Aesar) were reagent grade and used without further purification. Water was purified using an Elgastat Maxima HPLC reverse osmosis and deionisation system (Elga, Bucks, England). Synthetic Aβ1-40 peptide was obtained from Bachem (Torrance, CA, USA). Synthetic Aβ1-42 peptide was obtained from EZBiolabs (Carmel, IN, USA), primary antibody was purchased from Covance Signet Antibodies  83 (Princeton, NJ, USA) and secondary antibody from Abcam (Cambridge, MA, USA). Atomic absorption standards Cu(NO3)2 and ZnCl2 were purchased from Aldrich. Human hepatocellular liver carcinoma cells, HepG2, were provided by UBC Biological Services.  3.2.2 Instrumentation   Elemental analyses for C, H and N (Fisons EA 1108 instrumentation) were completed by Mr. David Wong at the UBC Chemistry Mass Spectrometry and Microanalysis Services. UV- visible absorbance data were collected on samples in individual cuvettes using a Hewlett- Packard 8543 diode array spectrometer outfitted with a Fisher Scientific 1016D Isotemp thermostatted water cooling system; absorbance data for multi-well plates were collected using a Labsystems iEMS or a Molecular Devices Thermomax microplate reader. Matrix-assisted laser desorption/ionisation time-of-flight (MALDI-TOF) mass spectrometry experiments were performed by Mr. Marshall Lapawa (UBC Chemistry Mass Spec Services) on a Bruker Biflex IV instrument. X-ray data were collected and processed by Dr. B.O. Patrick at UBC using a Bruker X8 APEX II diffractometer with graphite monochromated Mo-Kα radiation; structures were solved using Bruker SAINT software package (SAINT. Version 7.46A. Bruker AXS Inc., Madison, Wisconsin, USA. 1997-2007). X-band powder EPR spectra were recorded on a Bruker ECS-106 EPR spectrometer in quartz tubes (4 mm diameter). The temperature (130 K) was maintained by liquid nitrogen flowing through a cryostat combined with a Eurotherm B-VT- 2000 variable-temperature controller. The microwave frequency (9.4311 x 10-9 Hz) and magnetic field were calibrated with an EIP 625A microwave frequency counter and a Varian  84 E500 gaussmeter, respectively. 2048 points were taken at 0.2 mW; 2 G amplitude modulation, 40 kHz modulation frequency, 41.9 s sweep time, 5.12 ms time constant, 20.48 ms conversion time. Simulations of the EPR spectra were performed using the WinEPR Simfonia software package. Infrared absorption spectra (4000 – 400 cm-1) were collected using a Thermo Scientific Nicolet 6700 attenuated total reflectance FTIR with Smart Orbit attachment.  3.2.3 Copper(II) Complexes  N O OR NO O Cu R1:9 CH2Cl2:MeOH     NEt3 RT Cu(ClO4)2.6H2O N O OH R   Scheme 3.1 General procedure for preparation of bis(pyridonato)copper(II) complexes.  Bis(1-(4-dimethylaminophenyl)-2-methyl-3-oxy-4-pyridonato)copper(II), Cu(dapp)2 Hdapp (0.062 g, 0.25 mmol) was dissolved in 1:9 CH2Cl2:MeOH solution (13 mL) to which was added NEt3 (33.8 µL, 0.240 mmol). Copper(II) perchlorate hexahydrate (0.046 g, 0.12 mmol) was dissolved in 1:9 CH2Cl2:MeOH (4 mL) and the pro-ligand solution was added dropwise to the copper salt solution. After addition of the majority of pro-ligand solution, the reaction mixture formed a fine N O O N N O O Cu N  85 light green precipitate; filtration yielded a fine green solid which was soluble only in CHCl3 (0.050 g, 74 %). Liquid-liquid diffusion between CHCl3 and Et2O afforded dark green blade- shaped crystals suitable for X-ray diffraction. Anal. Calcd (found) for C28H30CuN4O4: C, 61.13 (60.79); H, 5.50 (5.38); N, 10.18 (10.08). Infrared spectrum (cm-1, total reflectance): 1607, 1519, 1504, 1463, 1296, 726, 574, 537. EPR (130 K, powder): A┴ = 150 x 10-3 cm-1, g┴ = 2.063, A|| = 202 x 10-4 cm-1, g|| = 2.263.  Bis(1-(4-methylaminophenyl)-2-methyl-3-oxy-4-pyridonato)copper(II), Cu(sapp)2 Hsapp (0.057 g, 0.25 mmol) was dissolved in 1:9 CH2Cl2:MeOH solution (13 mL) to which was added NEt3 (33.8 µL, 0.240 mmol). Copper(II) perchlorate hexahydrate (0.045 g, 0.12 mmol) was dissolved in 1:9 CH2Cl2:MeOH (4 mL) and the pro-ligand solution was added dropwise to the copper salt solution. After addition of the majority of pro-ligand solution, the reaction mixture formed a fine light green precipitate. Filtration yielded a green fine solid Cu(sapp)2 (0.040 g, 64 %). Anal. Calcd (found) for C26H26CuN4O4: C, 59.82 (59.49); H, 5.02 (5.16); N, 10.73 (10.60). Infrared spectrum (cm-1, total reflectance): 3337, 1608, 1504, 1459, 1293, 725, 583, 538.   N O O N H N O O Cu H N  86 Bis(1-(4-aminophenyl)-2-methyl-3-oxy-4-pyridonato)copper(II), Cu(zapp)2 Hzapp (0.055 g, 0.25 mmol) was dissolved in 1:9 CH2Cl2:MeOH solution (13 mL) to which was added NEt3 (33.8 µL, 0.240 mmol). Copper(II) perchlorate hexahydrate (0.045 g, 0.12 mmol) was dissolved in 1:9 CH2Cl2:MeOH (4 mL) and the pro- ligand solution was added dropwise to the copper salt solution. After addition of the majority of pro-ligand solution, the reaction mixture formed a fine light green precipitate. Filtration yielded a green fine solid Cu(zapp)2 (0.042 g, 70 %). Green plate crystals suitable for X-ray analysis were grown from slow evaporation of a solution of 3:1 MeOH:DMSO. Anal. Calcd (found) for C24H22CuN4O4: C, 58.35 (58.02); H, 4.49 (4.52); N, 11.34 (11.21). Infrared spectrum (cm-1, total reflectance): 3314, 3213, 1607, 1538, 1502, 1462, 1281, 737, 575, 539.  Bis(1-(6-benzothiazolyl)-2-methyl-3-oxy-4-pyridonato)copper(II), Cu(bt6p)2 Hbt6p (0.072 g, 0.28 mmol) was dissolved in 1:9 CH2Cl2:MeOH solution (10 mL) to which was added NEt3 (39.0 µL, 0.280 mmol). Copper(II) perchlorate hexahydrate (0.050 g, 0.14 mmol) was dissolved in 1:9 CH2Cl2:MeOH (4 mL) and the pro- ligand solution was added dropwise to the copper salt solution. After addition of the majority of pro-ligand solution, the reaction mixture formed a fine light green precipitate. Filtration yielded a green fine solid Cu(bt6p)2 (0.039 g, 50 %). Anal. Calcd (found) for C26H18CuN4O4S2: C, 54.02 (53.67); H, 3.14 (3.17); N, 9.69 (9.58). Infrared spectrum (cm-1, total reflectance): 1586, 1536, 1507, 1460, 1288, 723, 573, 533. N O O H2N NO O Cu NH2 N O O NO O Cu S N S N  87 Bis(1-(2-benzothiazolyl)-2-methyl-3-oxy-4-pyridonato)copper(II), Cu(bt2p)2 Hbt2p (0.018 g, 0.070 mmol) was dissolved in 1:9 CH2Cl2:MeOH solution (3.6 mL) to which was added NEt3 (9.4 µL, 0.067 mmol). Copper(II) perchlorate hexahydrate (0.013 g, 0.035 mmol) was dissolved in 1:9 CH2Cl2:MeOH (1.1 mL) and the pro- ligand solution was added dropwise to the copper salt solution. After addition of the majority of pro-ligand solution, the reaction mixture formed a fine light green precipitate. Filtration yielded a green solid (0.014 g, 35 %). Anal. Calcd (found) for C26H18CuN4O4S2•MeOH: C, 53.15 (53.00); H, 3.63 (3.28); N, 9.18 (9.33). Infrared spectrum (cm-1, total reflectance): 1586, 1545, 1500, 1456, 1299, 727, 564, 543.  Bis(1-phenyl-2-methyl-3-oxy-4(1H)-pyridinato)copper(II), Cu(ppp)2 Hppp (0.056 g, 0.28 mmol) was dissolved in 1:9 CH2Cl2:MeOH solution (13 mL) to which was added NEt3 (37.1 µL, 0.270 mmol). Copper(II) perchlorate hexahydrate (0.050 g, 0.14 mmol) was dissolved in 1:9 CH2Cl2:MeOH (4 mL) and the pro-ligand solution was added dropwise to the copper salt solution. After addition of the majority of pro-ligand solution, the reaction mixture formed a fine light green precipitate. Filtration yielded a green solid (0.042 g, 68 %). Anal. Calcd (found) for C24H20CuN2O4: C, 62.13 (61.96); H, 4.43 (4.36); N, 6.04 (6.02). Infrared spectrum (cm-1, total reflectance): 1586, 1535, 1505, 1450, 1306, 698, 586, 530.  N O O N O O CuS N S N N O O N O O Cu  88 Bis(1-benzyl-2-methyl-3-oxy-4(1H)-pyridinato)copper(II), Cu(nbp)2 Hnbp (0.059 g, 0.28 mmol) was dissolved in 1:9 CH2Cl2:MeOH (13 mL) to which was added NEt3 (37.1 µL, 0.270 mmol). This solution was added (dropwise) to Cu(ClO4)2.6H2O (0.050 g, 0.14 mmol) in 1:9 MeOH:CH2Cl2 (4 mL). The reaction mixture was stirred (RT, 6 h) yielding a green solid Cu(nbp)2, which was isolated via filtration (0.0361 g, 53 %); crystals suitable for X-ray diffraction were obtained via liquid-liquid diffusion between CHCl3 and Et2O. Anal. Calcd (found) for C26H24CuN2O4: C, 63.30 (63.30); H, 4.92 (4.91); N, 5.69 (5.69). Infrared spectrum (cm-1, total reflectance): 1594, 1536, 1502, 1476, 1284, 736, 571, 523.  3.2.4 Trolox Equivalent Antioxidant Capacity (TEAC) Assay   The antioxidant capacity of all 3-hydroxy-4-pyridinone pro-ligands were determined by a TEAC assay using 6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox) as a standard and α-tocopherol (vitamin E) and butylated hydroxytoluene (BHT) as reference compounds. TEAC values were calculated according to an improved 2,2’-azinobis-(3- ethylbenzothiazoline-6-sulphonic acid diammonium salt (ABTS) radical cation decolourisation assay.215 The radical was generated by dissolving ABTS (0.0387 g, 7.05 µmol, 7 mM)  in water (10 mL) and exposing the solution to potassium persulphate (0.0066 g, 24.4 µmol, 2.45 mM). After incubation (dark, RT, 16 h), the resultant solution was diluted with HPLC-grade methanol such that the absorbance of the solution at 745 nm was 0.7 ± 0.2 (i.e. 2.5 mL diluted to 200 mL). N O O N O O Cu  89 Both the ABTS•+ solution and the test solutions were maintained at 30 °C using a water bath. ABTS•+ solution (2 mL) was loaded into a cuvette and the absorbance was recorded at 745 nm prior to addition of the test compounds and then at times 1, 3 and 6 min following addition. Test solutions were prepared such that the addition of 20 µL to the radical would provide a reduction in absorbance by 20-80 %; this resulted in solutions ranging in concentration from 0.25-16 µM. Upon addition of 20 µL of test solution to the cuvette, the contents were mixed thoroughly by vigorously pipetting the mixture for twenty seconds. A new blank spectrum (MeOH) was collected before each sample run. Solution absorbance was plotted vs. test compound concentration; each resultant slope was normalised with respect to that obtained for Trolox to give the Trolox-equivalence (TEAC) value for each time point (1, 3, 6 min). The error in each slope was calculated using linear regression techniques, and then statistically carried through calculation to the final values for the TEAC. Controls for test compound and standard absorption at 745 nm were performed: separate solutions were prepared and their absorption (745 nm) monitored; all compounds displayed insignificant absorption at this wavelength.  3.2.5 Cytotoxicity (MTT) Assay  Human hepatocytes (cell line: HepG2) were cultured according to standard procedures.219 Cells were harvested for use by first removing excess culture media and adding trypsin (~ 4 mL) to the culture flask to release the cells from the wall of the dish. The cells were incubated (37 °C, 5 min), after which 10 % fetal bovine serum (FBS) media (~ 4 mL) was added to quench the  90 trypsin. The cells were suspended in solution, centrifuged (800 rpm, 3 min), and diluted to a concentration of 104 cells per 100 µL media. Cells were allotted to the wells of a 96-well plate (104 cells or 100 µL per well) and incubated (37 °C, 24 h). Test compounds were dissolved in 10 % FBS media (some requiring 1 % DMSO due to low solubility) and filtered through a 0.2 µM “Milex” polyethersulphone syringe filter for sterilisation. The initial solutions were diluted with 10 % FBS media (with 1 % DMSO for the less soluble compounds) to extend across a range of concentrations from 1 to 1000 µM. Cisplatin was used as a positive control for cell death. Cells were incubated with test compounds for 72 h; after incubation 3-(4,5- dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT; 50 µL of 2.5 g/L in PBS) was added and cells were incubated for an additional 3 h. Supernatant was removed and the cells were dissolved in DMSO (150 µL). Absorbance was measured at 577 nm and referenced to the control and blank wells to find the relative cell viabilities for each assay condition (n ≥ 4). Survival curves were plotted (cell viability as a function of test compound concentration) and fifty-percent inhibitory concentration (IC50) values calculated from linear regression on the appropriate regions of the plots.  3.2.6 Turbidity Assay  4-(2-Hydroxyethyl)-1-piperazineethanesulphonic acid (HEPES) buffer was prepared and Chelex resin was added to the solution to rid the buffer of any residual transition metal ions; the pH was adjusted to 6.6 for Cu2+ and to 7.4 for Zn2+ studies. After removal of Chelex via filtration (Millipore 0.22-µm acetate filter), these buffer solutions were used in the subsequent preparation  91 of metal, ligand and test solutions. Lyophilised synthetic human Aβ1-40 amyloid peptide was prepared as a ~ 200-µM solution in distilled, deionised water. Intermittent sonication (1 min on, 30 s off) was applied for 3 minutes through a water bath to achieve full peptide dissolution. Next, the peptide solution was filtered through a 0.2-µm nylon syringe filter (Whatman; Kent, UK) to remove any microparticulate protein matter. A bicinchoninic acid (BCA, 4,4'-dicarboxy- 2,2'-biquinoline) assay was performed to assess the concentration of the Aβ peptide, and was used to ensure that the concentration of Aβ peptide stayed consistent over many different assays.220 Stock solutions of Cu2+ and Zn2+ were prepared from ACS standards to final concentrations of 200 µM using pH 6.6 and 7.4 buffers respectively, and solutions of the free ligands were prepared in HEPES buffer of the appropriate pH. Turbidity assays were performed in flat-bottomed 96-well microtitre plates (Falcon) with 200 µL of assay solution per well. Generally, to evaluate the effects of the pyridinone pro- ligands, Aβ1-40 peptide (~ 25 µM) was co-incubated with either Zn2+ or Cu2+ (25 µM) for two minutes before the pro-ligand of interest was added (150 µM). Similar trials were performed with DTPA (50 µM) as a reference chelator with a high affinity for Zn2+ and Cu2+. After a 45- minute incubation at 22 ˚C, the 405-nm absorbances of all test solutions were measured using a Labsystems iEMS or Molecular Devices Thermomax microplate reader. Wells containing Aβ peptide and the respective test ligand in the appropriate pH HEPES buffer were used as blanks for spectrophotometric analysis, and MALDI-TOF spectroscopy was used to confirm the presence of intact Aβ peptide before and after the metal/chelator trials (for a representative MALDI-TOF spectrum see Appendix Figure A.13). Finally, an analysis of variance (ANOVA) between groups was performed to determine if chelator efficacy is dependent on structure.   92 3.2.7 Enzyme-Linked Immunosorbent Assay (ELISA)  Preparation of Aβ Fibrils Aβ42 fibrils were grown from synthetic human amyloid peptide (1-42) by dissolving Aβ peptide in H2O (0.105 mg in 315 µL, pH 5). The suspension was sonicated (3 x 5 min) to minimise the amount of particulate matter, mixed and placed in an incubator (37 °C, 72 h). The fibril solutions were stored (RT) for assay use no longer than 2 weeks.  Preparation of Reagent Solutions Phosphate-buffered saline was prepared (PBS, sodium phosphate (monobasic, 1.20 g, 10 mM), sodium phosphate (dibasic, 1.42 g, 10 mM), sodium chloride (8.06 g, 138 mM) and potassium chloride (0.20 g, 2.7 mM, pH 7). The chemiluminescence substrate solution was prepared containing p-nitrophenyl phosphate (0.100 g, 2.7 mM), diethanolamine (1.051 g, 0.1 M) and magnesium chloride (0.010 g, 0.5 mM), and the pH adjusted to 9.8. Antibody solutions were prepared no more than 1 h before use by diluting the purchased solution of antibodies in 1 % (w/v) BSA/PBS. For primary Ab, dilution of 1:6000 was used (i.e. 3.34 µL in 20.00 mL BSA/PBS); for secondary Ab, dilution of 1:1000 was used (i.e. 8.00 µL in 8.000 mL BSA/PBS).    93 Fibril Binding Assay Aβ42 fibrils were diluted (PBS) to make the working solution (5.8 µg/mL). Each test well of a 96-well plate (clear polystyrene, flat-bottomed, Nalge-Nunc, Rochester, NY) was exposed to fibril solution for ELISA plate preparation by peptide adsorption (50 µL, 3 h, RT). After removal of the excess sample, each well was blocked with 1 % BSA/PBS (300 µL, 30 min, RT). The BSA/PBS solution was removed and each well washed twice with PBS (2 x 300 µL). The test compounds were added to each well (50 µL, in PBS, 10 nm -10 mM), the plates covered with Parafilm and incubated (RT, 12 h). After incubation, the excess solution was removed from the wells and the wells were again blocked with 1 % BSA/PBS (300 µL, 30 min, RT). After rinsing twice with PBS (2 x 300 µL) the primary antibody solution was added (anti-Aβ IgG, clone 6E10, monoclonal, mouse; dilution 1:6000 in 1 % BSA/PBS) and the plates incubated (1 h, RT). The wells were washed (2 x 300 µL PBS) prior to secondary antibody (anti-mouse IgG H+L, polyclonal, rabbit; dilution 1:1000 in 1 % BSA/PBS) was applied for incubation (1 h, RT). After wells were again washed (2 x 300 µL PBS) the previously prepared visualisation solution was applied (50 µL). The absorbance (405 nm) was monitored during the period of incubation such that readings were obtained ranging from 0.2 to 1.0 AU.    94 3.3 Results and Discussion 3.3.1 Synthesis and Characterisation of Bis(pyridonato)copper(II) Complexes   All copper complexes were synthesised by deprotonation of pyridinone pro-ligand with triethylamine and reaction with copper(II) perchlorate hexahydrate; reagents were soluble in the solvent system used, but neutral coordination complexes produced were not and formed green precipitates (Scheme 3.1). Zinc(II) complexation was attempted by various approaches including literature preparations,221 but no complexes were isolated. All copper(II) complexes were characterised by elemental analysis (EA) and IR spectroscopy, but unlike the pro-ligands were not characterised by nuclear magnetic resonance (NMR) spectroscopy. Complicated spectra are expected due to coupling between the resonating nucleus and the spin of the unpaired electron of the paramagnetic copper(II) centre (3d9). Comparison with similar pyridinone complexes (of diamagnetic Ga(III) and In(III) ions), however, predicts significant downfield shifts for pyridinone ring protons.146 Percent elemental compositions by EA are consistent with the expected values for all Cu(L)2 complexes except Cu(bt2p)2, for which EA data are consistent with one molecule of MeOH included per Cu2+ centre; MeOH may be co-precipitated with the complex given that it is included in the reaction solvent mixture. After EA confirmation of composition, all complexes were analysed by IR spectroscopy and compared to their respective pro-ligands. The pyridinone moieties and the substituents on the pyridinone compounds (additional aromatic systems) are expected to vibrate relatively independently of each other, and the observed IR absorption spectra are consistent with this principle. It is impossible to identify discretely the νC=O and higher energy νring stretches, as they  95 are strongly coupled and there is no mode which is localised solely on the carbonyl bond.222 There is, however, a four-band IR spectral pattern observable at 1600-1400 cm-1 which is characteristic of pyridinone compounds.201,223  Table 3.1. Selected infrared absorption bands and their assignments. compound selected IR data (cm-1) νNH a  νOH νC=O + νring νCO νMO Hdapp - 3153 1628 1583 1522 1488 1295 - - - Cu(dapp)2 - - 1607 1519 1504 1463 1296 726 574 537 Hsapp 3417 3046 1623 1573 1513 1489 1297 - - - Cu(sapp)2 3337 - 1608    -  b 1504 1459 1293 725 583 538 Hzapp 3313,    - b  3204 1622 1555 1512 1487 1292 - - - Cu(zapp)2  3314, 3213 - 1607 1538 1502 1462 1281 737 575 539 Hbt6p - 3043 1624 1588 1556 1493 1297 - - - Cu(bt6p)2 - - 1586 1536 1507 1460 1288 723 573 533 Hbt2p - 3091 1624 1576 1508 1483 1302 - - - Cu(bt2p)2 - - 1586 1545 1500 1456 1299 727 564 543 Hppp - 3200 1626 1579 1537 1485 1296 - - - Cu(ppp)2 - - 1586 1535 1505 1450 1306 698 586 530 Hnbp - 3181 1628 1576 1527 1494 1262 - - - Cu(nbp)2 - - 1594 1536 1502 1476 1284 736 571 523 a  asymmetric and symmetric, respectively; b obscured by an adjacent band.   96 On complexation with metal ions, these ligand bands generally undergo a bathochromic shift (of approximately 20 cm-1) and there may also be a re-ordering of bands. Thus all four bands (attributable to νC=O, νring, νC=C) are listed as mixed in Table 3.1. The highest wavenumber band at approximately 1630  cm-1 in the pro-ligand spectra may be attributable to the carbonyl stretching frequency, νC=O, usually observed at about 1700 cm-1. The lower frequency is due in part to the aromatic resonance structure (Figure 2.1b), placing more single-bond character on this carbonyl group, shifting the stretching frequency down (toward the phenol C-O stretching frequency expected at 1200 cm-1).224 Because only the α-hydroxyketone group coordinates the ion directly, the IR spectra of the pro-ligands and their corresponding copper(II) complexes are expected to be very similar. There are, however, some diagnostic differences, the most obvious being the loss of νOH at 3200-3000 cm-1 upon complexation. The aforementioned bathochromic shifts are observed for almost all identifiable peaks upon metal ion complexation, even in such distal functional groups as the 4-aminophenyl νN-H bands (Table 3.1). A number of new bands are observed below 800 cm-1 in the Cu(L)2 spectra which are identified as νM-O bands, corresponding to the O-Cu2+ dative bonds in the complexes; these are tentatively assigned as they are also likely coupled to ring deformation modes.225 An additional spectral feature is apparent in the 3300-3000-wavenumber region in the IR spectra of Hzapp, Hsapp and their respective complexes. Arising from the aromatic primary and secondary amino groups respectively, these are weaker and sharper than νO-H bands.224 Because Hzapp contains a primary amino group, two N-H stretches are expected corresponding to  the asymmetric N-H stretch (appearing at higher wavenumber, at 3313 cm-1), and the symmetric N-H stretch (appearing at lower wavenumber, here obscured by the broad, strong νOH band around 3204 cm-1, but visible in the Cu(zapp)2 spectrum at 3213 cm-1). In the spectra of both Hsapp and Cu(sapp)2 (Figure 3.2),  97 the N-H stretching band is observed at 3417 cm-1 and 3337 cm-1, respectively. As expected, these IR spectral data delineate the fine electronic structural changes occurring on metal ion complexation by bidentate pyridinone ligands. Additional structural characterisation of complexes is provided by X-ray diffraction and EPR studies.    Figure 3.2. IR spectra for Hsapp pro-ligand (upper) and corresponding Cu2+ complex Cu(sapp)2 (lower), 4000-400 cm-1.  A number of copper complexes produced X-ray quality crystals upon crystallisation by the liquid-liquid layered diffusion method (Cu(dapp)2, Cu(ppp)2, Cu(nbp)2),226 or by slow 51 1. 7 53 3. 5 57 3. 9 64 1. 1 66 0. 9 74 7. 2 76 2. 2 82 6. 9 88 0. 694 9. 7 98 1. 5 10 35 . 5 10 56 . 4 11 01 . 9 11 58 . 0 11 76 . 5 12 04 . 512 33 . 212 54 . 7 12 97 . 2 13 93 . 5 14 64 . 6 14 88 . 5 15 13 . 0 15 72 . 61 62 2. 7 30 45 . 934 17 . 3 LS 080514-2 Hsapp 5-074  50  55  60  65  70  75  80  85  90  95 45 1. 0 53 7. 8 58 2. 859 6. 1 64 9. 5 72 5. 2 73 5. 5 76 5. 2 89 5. 6 97 7. 5 10 55 . 9 11 06 . 7 11 56 . 4 11 77 . 41 20 9. 4 12 50 . 3 12 73 . 2 12 92 . 6 13 39 . 6 14 31 . 3 14 58 . 6 15 04 . 2 16 08 . 1 33 37 . 0 LS 080514-4 Cusapp2 5-120  50  55  60  65  70  75  80  85  90  95  500    1000   1500   2000   2500   3000   3500   4000 Wavenumbers (cm-1) % Tr a n sm itt a nc e Cu(sapp)2 Hsapp 82 0. 5 34 17 33 37  98 evaporation of a methanol/DMSO solution (Cu(zapp)2). All four of the neutral bis(pyridonato)- copper(II) complexes co-crystallise with a one molecule of solvent (CHCl3 or DMSO) per asymmetric unit, or two solvent molecules per Cu2+ centre. The Cu2+ ion resides on a centre of inversion resulting in a symmetric molecule around it. All complexes display nearly square planar coordination spheres about the metal ion (Figure 3.3) and trans- ligand geometries (Figure 3.4). The appendix contains a brief outline of the crystallisation procedures and nature of the crystal for each complex. These are followed by the crystallographic data and, separately, selected bond lengths and angles in tabulated form (Tables A.4, A.5, respectively).    Figure 3.3. Ellipsoid plot (50 % probability; for clarity, H-atoms not shown) of Cu(dapp)2; side-on view of square planar copper(II) coordination centre.  99  Figure 3.4. Ellipsoid plots (50 % probability; for clarity, H-atoms not shown) of Cu(dapp)2, Cu(zapp)2, Cu(ppp)2 and Cu(nbp)2. Cu(ppp)2 Cu(nbp)2 Cu(zapp)2 Cu(dapp)2  100 Although X-ray diffraction (XRD) data for a number of pyridinone complexes of Fe(III), Al(III), Ga(III) and In(III) have been reported,184,196,203,227 only a few examples of Cu(II) complexes are available.228 Data for both the pro-ligand, HL, and the copper complex, Cu(L)2, must be available for direct analysis of bond changes upon complexation. In this case, it is possible to observe changes in bond distances within the pyridinones Hdapp, Hzapp and Hnbp upon copper(II) complexation. First, C-O bond distance changes may be examined. In the case of Hdapp Cu2+ complexation, the carbonyl C-O bond C(4)-O(1) increases in length from 1.2721(16) to 1.295(3) Å while the hydroxyl C-O(H) bond C(3)-O(2) decreases in length from 1.3543(16) to 1.321(3) Å. Similarly, Cu2+ complexation by Hzapp is accompanied by an increase in carbonyl bond length (C(4)-O(1)) from 1.281(2) to 1.298(2) Å and a decrease in hydroxyl C-O(H) bond length (C(3)-O(2)) from 1.358(2) to 1.328(2) Å. Copper(II) complexation by Hnbp also involves lengthening of the carbonyl C-O bond from 1.266(2) to 1.283(5) Å and shortening of the hydroxyl C-O(H) bond from 1.350(2) to 1.344(5) Å. These changes follow the same trends as those reported over Fe(III) complexation by pyridinones.203 In addition to changes in the C-O bond distances, changes may be expected in the C-C bond distances within the pyridinone ring. Previous evaluations of corresponding HL and Fe(L)3 pairs report lengthening of the C(2)-C(3) distance by 0.037 Å and shortening of the C(3)-C(4) distance by 0.021 Å.203 The lengthening of the C(2)-C(3) distance is mirrored here in the cases of all three HL/Cu(L)2 pairs, Hdapp, Hzapp and Hnbp, with changes of + 0.011, 0.017 and 0.033 Å, respectively. However, significant shortening of the C(3)-C(4) distance is only observed in the Hnbp/Cu(nbp)2 pair with a change of - 0.021 Å over Cu2+ complexation. This conservation of C(3)-C(4) bond length over Cu2+ complexation, in contrast to the shortening observed over Fe3+ complexation, is consistent with structural data for ligand complexation (L1, maltol) of  101 divalent metal ions (Zn2+, Pb2+).221 These observations of pyranone/pyridinone metal ion bond length changes over complexation indicate firstly: the changes in pyridinone ring electronic structure are slight, and secondly: the changes are dependent on the valence of the coordinated metal ion. Overall bond length changes are exhibited more often and more consistently in the carbon-oxygen bond distances of the metal-coordinating functional groups. In Cu(nbp)2 the benzyl substituent is disordered and was modelled in two orientations with equivalent populations. The additional freedom of ring orientations seen in solid state Cu(nbp)2 and presumably in Hnbp itself, compared to that in its analogues Hppp/Cu(ppp)2 may allow the compound to better infiltrate Aβ aggregates and permit improved access to the problematic metal species involved in AD. It is unlikely that either of these pyridinones remain in a fixed-ring orientation while in physiological conditions; however, the solid state structures hint that there is a smaller energy barrier to rotation with Hnbp which may prove to be advantageous. Further investigation is required to determine the magnitude of this effect in vivo. Electron paramagnetic resonance was used to confirm the speciation of copper in the bis(pyridinato) complexes. Because all seven complexes were prepared in the same manner, and because solid-state structural (XRD) data had confirmed a similar structure for a number of complexes, only one complex (Cu(dapp)2) was analysed by EPR as it is believed to be representative of all complexes prepared in this work. The α-hydroxyketone moiety of the pyridinone ligand is known to coordinate Cu2+ in solution even in the presence of other possible donor atoms such as those in amino or carboxylate groups.229 Previous studies on bis(maltolato)copper(II) complexes and bis(L-mimosinato) complexes (both containing the α- hydroxyketo-chelating moiety) were characterised by EPR and shown to contain Cu2+ centres in axial coordination environments, as expected for square-planar bis-bidentate ligand  102 complexes.229 The EPR spectrum obtained (shown with modelled spectrum, Figure 3.5.) contains a g┴-value typical of a Cu2+ centre with a (CO, O-) donor set at 2.063 (compared to that of maltol, L-mimosine Cu2+ complexes at 2.304).229 The expected hyperfine (HF) splitting peaks are observed, arising from coupling of the lone electron with the Cu2+ centre (d9, nuclear spin = 3/2). It should be noted that although four peaks are expected, only three are clearly discernable on the spectrum, as one HF peak overlaps the main spectral line. This EPR experiment does confirm the speciation of the copper ion in these bis(pyridinato) complexes as Cu2+.    Figure 3.5. Experimental and simulated EPR spectra of Cu(dapp)2 (130 K). 2200 2400 2600 2800 3000 3200 3400 3600 3800 4000 4200 Magnetic Field (G) experimental simulated g┴ = 2.063 gII = 2.263 AII = 202 x 10-4 cm-1  103 3.3.2 Antioxidant Capacity of Pro-Ligands  Oxidative stress is implicated not only in AD but in other neurodegenerative diseases such as Parkinson’s disease and amyotrophic lateral sclerosis, effecting tissue damage and leading to neuronal dysfunction and death. As mentioned in Chapter 1, redox cycling of metal ions associated with the Aβ peptide can produce reactive oxygen species (ROS) such as the extremely reactive hydroxyl radical, HO•. This ROS can abstract hydrogen atoms from biological molecules in the vicinity of its production and cause oxidative damage to lipids, proteins, DNA, etc. Phenols are a common type of chain-breaking antioxidant and serve to inhibit oxidative processes by donating the phenolic hydrogen atom to a radical, breaking the chain of radical propagation. The best-known examples of phenolic antioxidants are α- tocopherol and BHT, of which the latter is used commercially as a preservative in food packaging. 3-Hydroxy-4-pyridinones are able to donate hydrogen atoms in a similar manner to that of α-tocopherol to quench free radicals, and this ability can be measured by the ABTS radical assay. The assay compares the ability of the test compounds to quench the ABTS•+ radical cation, a reaction which is readily observed via the disappearance of the ABTS•+ absorption band in a UV-visible spectrum over time. Because the outcome of this assay is particularly sensitive to variations in experimental conditions, the results are normalised to the antioxidant capacity of a reference antioxidant such as Trolox, a hydrophilic analogue of α-tocopherol. All the pyridinone pro-ligands tested (except Hbt2p) showed comparable antioxidant capacity to α-tocopherol (Figure 3.6, Table 3.2.).   104   Figure 3.6. Trolox equivalent antioxidant capacity (TEAC) of hydroxypyridinones compared to standard antioxidants α-tocopherol (α-toc, vitamin E) and butylated hydroxytoluene (BHT). Shown are means of three trials; error bars represent ± one standard deviation.  The clear exception, with a lower TEAC value than α-tocopherol, is Hbt2p. This result is suspect, as in all applications Hbt2p was consistently the least soluble of the series of hydroxypyridinones. Although every attempt was made to find a solvent system for the TEAC assay to accommodate all pro-ligands with varying solubility, and though Hbt2p seemed to dissolve completely when the assay was performed, this is a possible explanation for the lower α-toc BHT Hdpp Hdapp Hsapp Hzapp Hbt6p Hbt2p Hppp Hnbp TE AC 1 min 3 min 6 min 1.80 1.60 1.40 1.20 0.60 0.40 0.20 0.00 1.00 0.80  105 TEAC value of the compound. It should be noted that because the 3-hydroxyl functional group responsible for pyridinone antioxidant activity is the same in all members of the series of pyridinone pro-ligands, similar TEAC values are to be expected across the series.  Table 3.2. Trolox equivalent antioxidant capacity (TEAC) values for all tested compounds, compared to standard antioxidants α-tocopherol (vitamin E) and butylated hydroxytoluene (BHT). compound TEAC value 1 min SD 3 min SD 6 min SD α-tocopherol 1.12 0.03 1.11 0.03 1.11 0.03 BHT 0.05 0.09 0.10 0.07 0.15 0.08 Hdpp 0.82 0.08 0.99 0.09 1.04 0.08 Hdapp 1.18 0.09 1.39 0.07 1.45 0.07 Hsapp 0.97 0.09 1.13 0.06 1.17 0.05 Hzapp 0.74 0.04 0.96 0.03 1.03 0.03 Hbt6p 0.6 0.1 0.8 0.1 0.89 0.09 Hbt2p 0.12 0.03 0.14 0.04 0.15 0.04 Hppp 0.94 0.08 1.1 0.1 1.22 0.06 Hnbp 1.0 0.1 1.2 0.1 1.25 0.09  As expected, the TEAC values for Hppp and Hnbp are statistically equivalent at all time points, demonstrating that variation in linker length between the two rings of the compound has no effect on antioxidant capacity. Overall, pyridinone TEAC values are comparable to that of the reference antioxidant α-tocopherol, and in fact substantially exceed that of BHT, a compound  106 exploited commercially for its antioxidant properties. This indicates that the antioxidant tenet of the multifunctionality design of these pro-ligands is achieved.  3.3.3 Cytotoxicity of Pro-Ligands  In order for the pyridinone pro-ligands to be considered for medicinal application, they must exhibit acceptably low toxicities. Human hepatocellular liver carcinoma cells (line HepG2) were used, as these are readily cultured are a microcosm for studying drug effect on the liver, the organ responsible for drug metabolism and detoxification.230 The MTT (3-(4,5-dimethylthiazol- 2-yl)-2,5-diphenyltetrazolium bromide) assay217 is a colourimetric test of the viability of cultured cells after exposure to a challenge compound, and is used as a method of screening drugs in early stages of development. The assay quantifies the amount of yellow MTT reduced to purple formazan in the mitochondria of living cells; the amount of purple reagent is directly proportional to the percent viability of the cell population.217 Cisplatin (cis-diammine- dichloroplatinum(II)) was used as a positive control for cell death, to verify that the assay was working. Due to their low solubility, some of the pro-ligands (Hbt6p, Hnbp) required the addition of DMSO to the test solutions. This addition was carried through to the appropriate control wells, and DMSO concentrations of up to 1 % in cell culture media were well tolerated by the cells. Unfortunately a few of the pyridinone pro-ligands (Hdapp, Hbt2p) were unsuitable for MTT assay as their aqueous solubilities were too low to afford the high concentrations required of the stock solutions, even with the addition of 1 % DMSO.  107 All of the tested compounds lowered the viability of the test cell cultures, with increasing detrimental effect with increasing compound concentration, as expected. A representative survival plot is given in Figure 3.7. To compare the effect of a test condition (e.g., a drug) on a population of subjects, the fifty-percent inhibitory concentration (IC50) is used; this refers to the concentration at which fifty percent of the population are affected (here, at which fifty percent are nonviable). The IC50 values of all tested compounds were calculated (Table 3.3).    Figure 3.7. Sample survival plot of HepG2 cells exposed to varying concentrations of Hppp for 72 h, as monitored by MTT assay. n = 4, error bars indicate ± one standard deviation. Hppp y = -0.373ln(x) + 1.48 R² = 0.999 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 0.1 1 10 100 1000 10000 Pe rc en t V ia bi lit y Concentration of Test Compound (µM)  108 Table 3.3. IC50 values of tested compounds, compared to cisplatin, used here as a reference compound. compound IC50 (µM) error (±) cisplatin 15 3 Hzapp 92 18 Hbt6p 19 5 Hppp 16 6 Hnbp 18 4  Variation of linker length between the two rings seems to have no effect on pyridinone cytotoxicity, as Hppp and Hnbp have comparable IC50 values (16 ± 6 and 18 ± 4 µM, respectively); however, it seems that hydrophilicity does, as more hydrophilic N-substituents (such as that in Hzapp) tend to imbue pyridinones with higher IC50 values. All of the assayed pro-ligands show acceptably low cytotoxicity, such that their application as metal chelating drugs is feasible when low concentrations are used.  3.3.4 Interaction of Pro-Ligands with Aβ Fibrils  The Yang group has developed an ELISA technique to screen for the association of small molecules with insoluble deposits of aggregated Aβ peptides.108 Synthetic Aβ42 peptide is fibrillised by incubation in distilled water, adsorbed onto multi-well spectrophotometry plates and then exposed to putative binding agents. After overnight incubation, visualisation is  109 performed using an antibody specific to the Aβ42 peptide; any small molecule that has bound the Aβ fibrils and obstructs the site of antibody binding will inhibit absorbance upon visualisation to give a positive result for Aβ binding by this assay. Before test compounds were examined in this assay, a number of pre-tests were performed to verify that the assay was functioning properly. A specificity control was performed to rule out indiscriminate primary antibody binding to the well. Aβ42 fibrils were omitted and the rest of the assay performed with primary/secondary antibody as described; another specificity control was performed by omitting the primary and using only the secondary Ab. In both of these controls, the lack of absorbance signal confirmed that neither primary nor secondary antibody showed significant indiscriminate binding to the wells. Another series of controls were performed to verify the specificity of antibody binding to the fibrils. Wells were coated with Aβ42 fibrils but application of primary antibody was omitted from the procedure. The absence of absorbance signal ruled out nonspecific binding of secondary antibody to the fibrils. Having verified that the assay was functioning properly, test compounds were then examined using the ELISA, with positive and negative controls for fibril binding run in parallel with any test compounds (n ≥ 5). Thioflavin T was used as the positive control for fibril binding, as it is known to bind fibrillised amyloid peptide. Buffer (PBS) was used as the negative control for fibril binding yielding high absorbance upon visualisation of the assay. Although consistent results were obtained, they are not comparable (within error) to those published elsewhere.108 Specifically, the IC50 values obtained for thioflavin T in this work are 7.6 ± 0.9 mM; the value reported by Inbar et al. is three orders of magnitude lower at 5 µM.108 This could possibly be explained by the use of different batches of Aβ. It is known (Kaan Biron,  110 Jeffries Lab (UBC), personal communication) that even within the same product from the same supplier, different lots of peptide will yield different results in fibrillisation trials. Preliminary results indicate that while the smaller, monocyclic pyridinones 1,2-dimethyl- 3-hydroxy-4(1H)-pyridinone (Hdpp) and 1-carboxyethyl-3-hydroxy-2-methyl-4(1H)-pyridinone (Hcep, Figure 3.8) do not associate with Aβ42 fibrils, pro-ligands incorporating a second aromatic ring joined by a flexible methylene linker (Hnbp) or one bearing a primary amine (Hzapp) may show fibril association (Figure 3.9). Due to solubility limitations, quantitative data were not obtained; however, the ELISA method shows promise for testing the suitability of the novel pro-ligands for their intended application to AD pathology.  NO HO OH O   Figure 3.8. 1-Carboxyethyl-3-hydroxy-2-methyl-4(1H)-pyridinone, Hcep.231  The fibril-binding ELISA will only be suitable for compound characterisation in this project if it is more sensitive to interference by Aβ42-binding molecules. It is possible that manipulation of Aβ fibril concentration, or fibrillisation state nature of Aβ fibrils will be able to lower the IC50 values such that they are more consistent with those observed by the Yang group.108   111   Figure 3.9. Positive result for thioflavin T association with Aβ42 fibrils by ELISA; non- statistically significant evidence for Hzapp fibril binding up to the solubility limit (4 mM).  The low sensitivity of this assay as it is now limits its value for this project; the maximum concentration achievable for a stable solution of thioflavin T is 1 mM (above this value, compound precipitates out of solution). The maximum achievable concentrations (PBS, without DMSO) for Hnbp is 3.25 mM and for Hzapp, 4.03 mM, for Hsapp (5 % DMSO in PBS), 1 mM. Although it has been reported that DMSO concentrations of up to 5 % in solution are tolerated in this assay,108 this was not observed to be so in these trials. Both Hsapp and Hzapp were dissolved in 5 % v/v DMSO/PBS for testing in the ELISA; however, results were inconsistent with those performed without DMSO. Absorbance values obtained with DMSO were scattered such that no clear result was observable. It is possible that the procedure may be modified to allow reproducible results to be gathered in the presence of 5 % DMSO, or using -10% 0% 10% 20% 30% 40% 50% 60% 70% 1.0E-09 1.0E-07 1.0E-05 1.0E-03 1.0E-01 %  In hi bi tio n  o f A bs , 40 5 n m [Test Compound] /M Hzapp Thioflavin T S N+ N NO HO NH2  112 lower amounts of DMSO may allow higher concentrations of lower-solubility compounds to be tested.  3.3.5 Interaction of Pro-Ligand with Metal Ions and Aβ Peptide in Solution  Zinc(II) and copper(II) are known to cause Aβ aggregation in solution.69,97 A turbidity assay was performed to investigate the ability of the pyridinone compounds to dissolve metal ion-aggregated Aβ species in vitro. Synthetic human Aβ1-40 was dissolved in a buffered aqueous solution to which was added metal ions (Cu2+ or Zn2+, in different trials). Copper and zinc were used in this assay as they are found in elevated concentrations in Alzheimer’s amyloid plaques.44,55 Metal ion addition causes Aβ aggregation, which, through formation of a turbid solution, is measureable by absorption spectroscopy. Suspended solids (such as aggregated Aβ) scatter light and give an increase in apparent absorbance of the solution. Addition of metal binding pro-ligands reduces this apparent absorption and thus a comparison of Aβ disaggregating ability can be made between pro-ligands; lower absorbance indicates greater efficacy. Diethylenetriaminepentaacetic acid was used as a positive control; it is a multidentate ligand known to bind metals strongly and dissolve metal-containing Aβ aggregates.18,232 Because activity in this assay is dependent upon metal ion chelation, the free pro-ligands were used, as opposed to their glycosylated pro-drug forms; a number of pyridinone compounds were tested including Hdpp, Hzapp, Hbt6p, Hppp and Hnbp. Pro-ligand effect on metal-induced Aβ1-40 aggregation was assessed at physiological pH (7.4) for Zn2+ and at lower pH (6.6) for Cu2+. The  113 interaction of the Aβ peptide with Cu2+ is known to be pH-dependent with no aggregation apparent at physiological pH.69    Figure 3.10. Pro-ligand disaggregation of metal-induced Aβ40 aggregates in vitro. Bars indicate mean resolubilisation efficiency of each pro-ligand based on solution absorbance (405 nm, n ≥ 3); error bars indicate ± one standard deviation.  Analysis of variance was performed to establish what effect, if any, the variation of pyridinone N-substituent had on peptide resolubilisation efficacy. All pyridinones significantly reduce the amount of aggregated Aβ compared to the “Aβ and Metal” negative control for both the copper and zinc trials (Figure 3.10, Table 3.4). In Cu2+ trials all pyridinones demonstrated 0% 20% 40% 60% 80% 100% 120% Aβ+M Aβ+M+ DTPA Aβ+M+ Hdpp Aβ+M+ Hzapp Aβ+M+ Hbt6p Aβ+M+ Hppp Aβ+M+ Hnbp R e la tiv e  R e s o lu bi lis a tio n  Ef fic ie n c y Zn(II), pH 7.4 Cu(II), pH 6.6  114 significant Aβ resolubilisation efficacy equal to that of DTPA, and this efficacy was statistically equivalent across the series of pyridinones. In Zn2+ trials, DTPA was significantly more effective for reduction of Aβ aggregates than were any of the pyridinone pro-ligands. Furthermore, it is apparent that the structure of the pro-ligand does have some bearing on resolubilisation efficacy in the case of Zn2+-aggregated Aβ40; Hdpp was significantly more effective than the other pyridinones for Aβ disaggregation.  Table 3.4. Percent efficacy of pro-ligand attenuation of metal ion-mediated Aβ40 aggregation in solution. “M” refers to either Zn2+ or Cu2+.  Percent Efficacy of Pro-Ligand  Zn2+ (pH 7.4) SD Cu2+ (pH 6.6) SD Aβ blank 93 3 87 5 Aβ + M 0 13 0 9 Aβ + M + DTPA 100 10 81 5 Aβ + M + Hdpp 80 3 78 1 Aβ + M + Hzapp 30 10 74 1 Aβ + M + Hbt6p 33 4 71 3 Aβ + M + Hppp 40 6 72 3 Aβ + M + Hnbp 32 5 77 4  The other pyridinones Hzapp, Hbt6p, Hppp and Hnbp displayed similar abilities to reduce Aβ aggregates in solution, possibly indicating that while a smaller (single-ring) pyridinone structure may be advantageous in this application, the structural nature of any N-substituent aryl group does not seem to affect ligand propensity for interaction with metal-Aβ binding site(s) or reduction of metal ion-induced Aβ aggregation. Further experiments will be required to  115 understand the dominant processes involved in the interaction of pyridinones with metal ion- aggregated Aβ peptide, but the current results highlight the potential benefit of molecular structural variation to affect pyridinone biological function.  3.4 Conclusions  A series of 3-hydroxy-4-pyridinone compounds were characterised by a variety of physical and biochemical methods. Copper(II) complexes of each compound were synthesised and characterised by EA, IR, XRD and EPR, confirming consistent ion coordination, as expected, via the α-hydroxyketone functional group of the pyridinone ring. The pro-ligands displayed significant antioxidant activity in an ABTS•+ quenching assay, for the most part in excess of that of α-tocopherol. All pro-ligands sufficiently soluble in cell culture media were tested by MTT assay and found to be of acceptably low toxicity. Pro-ligand propensity for Aβ42 fibril binding was probed with an ELISA procedure. The intention of this assay was to test the value of incorporating amyloid-targeting structural features into the pyridinone compounds; structural mimicry of a compound known to bind fibrils, such as thioflavin T, was used in a bid to promote localisation of these chelators to the amyloid fibrils and thus to the site of elevated metal ion concentration. Although the ELISA was meant specifically to test the ‘targeted’ pyridinones such as Hdapp, Hsapp, Hzapp, Hbt6p and Hbt2p, it was not possible to test the range of compounds due to solubility limits. It is hoped that future optimisation of this assay will enable this range of compounds to be tested. Finally, in turbidity experiments, a number of  116 representative pro-ligands demonstrated significant ability to inhibit both Zn2+- and Cu2+- mediated aggregation of Aβ peptide in vitro. Regarding the results gained for the pair of compounds Hppp and Hnbp, the performed assays have confirmed that the in vitro toxicity, antioxidant capacity and ability to dissolve Aβ aggregates are not significantly affected by the addition of a methylene spacer to our drug model. Aβ42 fibril binding ability remains to be tested for Hnbp for comparison to the negative result for fibril binding gathered for Hppp in this work. There is evidence from the solid-state structures of Cu(ppp)2 and Cu(nbp)2 that the addition of the methylene unit may allow more rotational freedom which may improve the compound’s ability to infiltrate large Aβ aggregates in vivo; further investigation into this and other structural modifications is called for in the development of new multifunctional metal ion chelators for Alzheimer’s disease therapy.   117 CHAPTER 4 Future Work  A series of new 3-hydroxy-4-pyridinone pro-ligands have been synthesised along with their corresponding glycosylated derivatives (i.e. prodrugs). Pyridinones were rationally designed to incorporate a number of activities including metal ion binding, antioxidant activity and interaction with synthetic Aβ peptide. The pyridinone compounds were tested for activity in vitro. The compounds coordinate copper(II) in the expected trans fashion forming square planar complexes, show good antioxidant activity and acceptably low cytotoxicity to human cells. Preliminary experiments indicate that single-ring and simple-ring pyridinone structures do not interact with Aβ42 fibrils; however, it was not possible to test the full range of pro-ligands. Those pyridinones which dissolved adequately in buffered aqueous solution showed good resolubilisation activity for Zn2+, and particularly Cu2+-aggregated Aβ40. The final chapter is devoted to outlining potential avenues for progression of this project.  4.1 Future In Vitro and In Vivo Assays for Drug Development  Generally the progression of Aβ imaging agent development from the research lab to the clinic follows from (i) in vitro screening using Aβ fibrils and calculation of Kd or Ki values and (ii) visualisation of radiolabelled compound localisation to Aβ by ex vivo AD brain section (human or mouse model) autoradiography; to (iii) a biodistribution study in a small animal before (iv) assessment in patient cohorts.233 The further development of this series of multi-  118 functional metal-binding, amyloid-targeted pyridinone pro-ligands may follow a similar path. This indicates that the next step for assessment of the multi-functional pyridinone compounds of this project should focus on the full characterisation of compound binding in vitro to Aβ fibrils.  4.1.1 Optimisation of Amyloid Fibril Binding Assay (ELISA)  The ELISA method is already in use in our lab, putting this assay at the forefront for screening compounds for fibril binding. This would serve as groundwork for any evaluation of binding constants of chelators to fibrils. Specifically, the next task in the ELISA optimisation is to confirm that up to 5 % DMSO in solution is tolerable in this assay; that is, that DMSO has no significant effect on Aβ fibrils or compound binding to those fibrils in solution.108 The developers of this assay have stated that DMSO levels of up to 5 % are tolerated in this assay;108 however, trials on 3-hydroxy-4-pyridinones using DMSO-free and 5 % DMSO-containing solutions of pro-ligands have yielded inconsistent results. Further trials with DMSO, possibly at lower concentrations, must be performed in order to demonstrate that consistent and comparable results may be obtained with this solvent. If this is achieved, DMSO can be used to solubilise compounds such as Hdapp, Hsapp, Hbt6p, and Hbt2p and enable testing of the full range of pyridinone pro-ligands.   119 4.1.2 Expansion of In Vitro Aβ Fibrillisation Assays: Fibrillisation Inhibition and Prevention of Aβ-Induced Cytotoxicity  Beyond probing simple binding of test compounds to previously-generated Aβ fibrils, the ability of test compounds to interfere with the process of Aβ fibrillisation can be examined. A number of research groups have developed assays to probe the capacities of compounds such as N-methylated peptides,112 bis(styrylbenzene) and bis(styrylpyridine) derivatives,113 and certain mulberry plant extracts111 to inhibit in vitro Aβ42 fibrillisation. Generally, monomeric Aβ42 is dissolved in H2O and exposed to varying concentrations of the putative fibrillisation inhibitor.111,112 After periods of incubation  ranging from one to four days (37 °C), samples of test solution are removed and exposed to thioflavin T. Thioflavin T is used as a fibrillisation probe, as its fluorescence profile (excitation, emission wavelengths) alters upon Aβ fibril binding.234 This assay could be readily applied to pyridinone pro-ligands in the Orvig research lab, as it utilises the same instrumentation and many of the same techniques as the ELISA. Transmission electron microscopy (TEM) and atomic force microscopy (AFM) have been used in parallel with the thioflavin T-monitored fibrillisation assays to structurally characterise Aβ fibrils produced in the presence and absence of inhibitors; generally the presence of inhibitor induces much more diffuse and/or incomplete fibrillisation of Aβ.111,112 These results correlate very well with those from the thioflavin T-monitored assays, such that thioflavin T reporting is sufficient to monitor fibrillisation inhibitory activity of test compounds.112 If thioflavin T assays indicated inhibition of Aβ fibrillisation by the 3-hydroxy-4-pyridinones, it would be interesting to confirm this via TEM or AFM fibril characterisation.  120 Pyridinone interference with Aβ fibrillisation processes in vitro can be further investigated by their attenuation of Aβ-induced neurotoxicity. Neuronal cytotoxicity studies have been performed using N-methylated peptide fragments112 and certain plant extracts.111 These test compounds, previously screened for inhibition of Aβ fibrillisation via the thioflavin T assay, are added to cell cultures along with Aβ42 peptide and can significantly inhibit Aβ-induced cell death.111,112 This cell-based assay, combined with thioflavin T- and TEM/AFM-monitored fibrillisation inhibition assays, can supplement the existing ELISA and turbidity assays to screen pyridinones for applicability to Alzheimer’s disease therapy.  4.1.3 Expansion of Turbidity, TEAC and MTT Assays to Include Glycosylated Pro- Ligands (Prodrugs)  One of the next tasks in this project would be to expand the appropriate in vitro assays to include the compounds in their glycosylated forms. The glycosylated compounds are prodrugs, and are without the α-hydroxyketone (chelating) functionality and the phenol (antioxidant) functional group. As such, they should show no disaggregation efficacy in the turbidity assay and no antioxidant activity in the TEAC assay. It would be interesting to evaluate the cytotoxicity of the glycosylated ligands, as glucose linkage increases compound molecular weight by approximately 70 %, and is expected to significantly increase their solubility. These changes in physical characteristics will likely affect the cell exposure to the compound through altered compound partitioning between the cell and the surrounding media, and altered distribution within the cell if it is taken up.  121 4.2 Radioactive Analogues of Pyridinone Pro-Ligands  Other interesting in vitro assays to screen for appropriate biological activity of these compounds would include gastrointestinal (GI) cell penetration assays, blood-brain barrier (BBB) crossing assays, and lastly biodistribution assays. These assays are most easily monitored using radiolabelled pyridinone compounds. Radiolabelling can be achieved in a variety of different ways: 11C-Methylation of the Hzapp and Hsapp compounds would yield [11C]Hsapp and [11C]Hdapp, respectively. Iodine destannylation of precursors could provide [123I/125I]- labelled analogues of Hppp, Hnbp, Hbt6p or Hbt2p, in the same manner as has been previously reported by our research group (Scheme 4.1).18  Br NH2 NH4SCN Br2 O RO O N RO O N S Br MeOH/H2O ~ 120 °C ~ 100 h N HO O N S SnBu3 S N NH2 Br N RO O N S I * Sn2Bu6 Pd(PPh3)4 toluene [123I]NaI Chloramine-T N RO O N S SnBu3 deprotectionN RO O N S Br   Scheme 4.1. Preparation of a radiolabelled, iodinated analogue [123I]Hbt2p.18 R = Me or Bn.   122 4.2.1 Gastrointestinal Crossing Assays  A bidirectional GI epithelial cell crossing assay may be performed to gauge compound oral bioavailability. The experiments involve culturing human GI epithelial cells (e.g., Caco-2 cells) in a specialised culture well, creating a single layer of cells which acts as a barrier between two chambers (Figure 4.1).    Figure 4.1. Schematic diagram of a Caco-2 cell monolayer cultured on a porous membrane yielding two accessible sides of the model GI endothelium, both apical and basolateral, to be analysed for test compound concentration.197,235  After exposure of one side of the cell layer to the test compound, the amount of compound present on both sides of the cell layer barrier as well as within the cells themselves is measured by HPLC to quantify the amount of diffusion and/or transport across the cell membrane(s).197,235 This assay will be performed on the glycosylated prodrug compounds, as any drug will be administered in this form. Because this assay’s relevance depends on the maintenance of the β- glycosidic bond in the GI system, the GI crossing assay must be complemented by an acid apical side basolateral side cell monolayer porous support  123 stability assay. Oral dosing exposes compounds to pH 1 to 2 in the stomach, pH 4.5 in the beginning of the small intestine, pH 6.6 on average in the small intestine and pH 5-9 in the colon.236 The prodrug compounds will be incubated in solutions mimicking stomach conditions (pH 1.2 HCl buffer, 37 °C, 75 min) or small intestine conditions (pH 6.8 phosphate buffer, 37 °C) and the amount of intact prodrug will be monitored over time by HPLC.236 If significant amounts of glycosylated compound remain in solution, the prodrug may be chemically stable during the transit of the GI tract. 4.2.2 Blood-Brain Barrier Crossing and Biodistribution Assays  Radiolabelled analogues of pyridinones are also amenable to BBB uptake assays in the mouse system. A perfusion of test compound into the carotid artery of a mouse, followed by a short incubation, sacrifice of the animal and scintillation counts of brain tissue can be used to determine BBB permeation of the compound.18 Full biodistribution assays are an extension of this procedure, requiring altered routes of administration (oral or intravenous) and that scintillation counts be performed on all sections (tissues) of interest in the animal.  4.2.3 Probing In Vitro Aβ Targeting  Further to the progression from lab space to clinic outlined at the beginning of this chapter, the Aβ plaque-targeting activity of the pyridinones should be evaluated. One method uses radiolabelled compounds and ex vivo Aβ plaque-containing brain sections either from a patient affected with AD or from a transgenic mouse exhibiting senile plaque deposition. After  124 exposure of tissue sections to radiolabelled test compound, autoradiography can outline the localisation of the compound. Taken together with a view of the histochemically-stained section to visualise the amyloid plaques, the co-localisation of Aβ plaques and pyridinone compound could be qualitatively observed. A more quantitative experiment can be performed to obtain the affinity constant for association of a test compound with ex vivo Aβ fibrils. Human AD brain tissue is homogenised and incubated in wells of a multiwell plate with the radiolabelled test compound.237 A radiolabelled reference compound known to bind Aβ fibrils, such as radiolabelled Pittsburgh Compound B ([3H]PIB),237 is added to the tubes before incubation; nonradioactive PIB or test compound is then added to test wells in a range of concentrations. After mixing and incubation the tube contents are separated with a cell harvester to extract from the mixture and rinse any Aβ fibrils/plaques; the filters are placed in a scintillation counter to determine radioactivity.237 More radioactivity indicates strong association of the radiolabelled test compound with the fibrils/plaques, while less (or no) radioactivity indicates that the radiolabelled compound was either weakly associated with the fibrils/plaques such that it was displaced by the competitor compound, or not at all associated and thus rinsed away. This procedure can yield Ki values for any radiolabelled pyridinone pro-ligand.237    125 4.3 Other Masking Groups for the 3-Hydroxyl Functionality  Other functional groups beyond hexoses could be used to mask the 3-hydroxyl metal ion- binding/antioxidant functionality of the pyridinone pro-ligands. For example, the hydroxyl group may be converted to a boronic ester which is expected to show little interaction with metal ions such as Fe3+, Cu2+ or Zn2+; however, the boronic ester is readily hydrolysed upon exposure to H2O2 (as is present in the highly oxidative conditions of an AD-affected brain)43 to produce the active pyridinone chelator (Scheme 4.2).238   N B R OO O H2O2 N HO R O Mn+ M N O R O n   Scheme 4.2. Boronate-masked pyridinone compounds BL would be ineffective for metal ion binding; only conversion by H2O2 produces the pro-ligand HL which may coordinate metal ions (Mn+) such as Cu2+ or Zn2+ in areas of oxidative stress.  Although this differs from the glycosylation strategy exploited in this work, this type of masking may have interesting effects on prodrug cell permeation, BBB crossing by passive diffusion and localisation of chelating effect to the site of oxidative stress.  126  References  (1) Barnham, K. J.; Bush, A. I. Curr. Op. Chem. Biol. 2008, 12, 222. (2) Pandolfo, M. Arch. Neurol. 2008, 65, 1296. (3) Hershko, C.; Link, G.; Konijn, A. M.; Cabantchik, Z. I. Ann. N. Y. Acad. Sci. 2005, 1054, 124. (4) Ala, A.; Walker, A. P.; Ashkan, K.; Dooley, J. S.; Schilsky, M. L. Lancet 2007, 369, 397. (5) Liu, S. Chem. Soc. Rev. 2004, 33, 445 (6) Galanski, M.; Jakupec, M. A.; Keppler, B. K. Curr. Med. Chem. 2005, 12, 2075. (7) Kean, W.; Kean, I. InflammoPharmacology 2008, 16, 112. (8) Bowen, M. L.; Orvig, C. Chem. Comm. 2008, 41, 5077. (9) Chapuy, B.; Hohloch, K.; Trümper, L. Biotechnol. J. 2007, 2, 1435. (10) Maksimoska, J.; Feng, L.; Harms, K.; Yi, C.; Kissil, J.; Marmorstein, R.; Meggers, E. J. Am. Chem. Soc. 2008, 130, 15764. (11) Atilla-Gokcumen, G. E.; Williams, D. S.; Bregman, H.; Pagano, N.; Meggers, E. ChemBioChem 2006, 7, 1443. (12) Thompson, K. H.; Orvig, C. Met. Ions Biol. Syst. 2003, 41, 221. (13) Caravan, P.; Gelmini, L.; Glover, N.; Herring, F. G.; Li, H.; McNeill, J. H.; Rettig, S. J.; Setyawati, I. A.; Shuter, E.; Sun, Y.; Tracey, A. S.; Yuen, V. G.; Orvig, C. J. Am. Chem. Soc. 1995, 117, 12759. (14) Woo, L. C. Y.; Yuen, V. G.; Thompson, K. H.; McNeill, J. H.; Orvig, C. J. Inorg. Biochem. 1999, 76, 251. (15) Storr, T.; Mitchell, D.; Buglyo, P.; Thompson, K. H.; Yuen, V. G.; McNeill, J. H.; Orvig, C. Bioconjugate Chem. 2003, 14, 212. (16) Kety, S. S.; Letonoff, T. V. Proc. Soc. Exp. Biol. Med. 1941, 46, 476. (17) Parang, K.; Wiebe, L. I.; Knaus, E. E. Curr. Med. Chem. 2000, 7, 995. (18) Schugar, H.; Green, D. E.; Bowen, M. L.; Scott, L. E.; Storr, T.; Böhmerle, K.; Thomas, F.; Allen, D. D.; Lockman, P. R.; Merkel, M.; Thompson, K. H.; Orvig, C. Angew. Chem., Int. Ed. 2007, 46, 1716. (19) Fernandez, C.; Nieto, O.; Fontenla, J. A.; Rivas, E.; de Ceballos, M. L.; Fernandez- Mayoralas, A. Org. Biomol. Chem. 2003, 5, 767. (20) Di Stefano, A.; Sozio, P.; Cerasa, L. Molecules 2008, 13, 46. (21) Pardridge, W. M. Curr. Op. Pharmacol. 2006, 6, 494. (22) Ulbrich, K.; Hekmatara, T.; Herbert, E.; Kreuter, J. Eur. J. Pharm. Biopharm. 2009, 71, 251. 127  (23) Pavan, B.; Dalpiaz, A.; Ciliberti, N.; Biondi, C.; Manfredini, S.; Vertuani, S. Molecules 2008, 13, 1035. (24) Alzheimer, A.; Stelzmann, R. A.; Schnitzlein, H. N.; Murtagh, F. R. Clin. Anat. 1995, 8, 429. (25) Ferri, C. P.; Prince, M.; Brayne, C.; Brodaty, H.; Fratiglioni, L.; Ganguli, M.; Hall, K.; Hasegawa, K.; Hendrie, H.; Huang, Y.; Jorm, A.; Mathers, C.; Menezes, P. R.; Rimmer, E.; Scazufca, M. Lancet 2005, 366, 2112. (26) Walsh, D. M.; Selkoe, D. J. J. Neurochem. 2007, 101, 1172. (27) Selkoe, D. J. J. Neuropathol. Exp. Neurol. 1994, 53, 438. (28) Selkoe, D. J. Physiol. Rev. 2001, 81, 741. (29) Barnham, K. J.; Masters, C. L. Nat. Rev. Drug Discov. 2004, 3, 205. (30) Glenner, G. G.; Wong, C. W. Biochem. Biophys. Res. Commun. 1984, 120, 885. (31) Masters, C. L.; Simms, G.; Weinman, N. A.; Multhaup, G.; McDonald, B. L.; Beyreuther, K. Proc. Natl. Acad. Sci. U. S. A. 1985, 82, 4245. (32) Frederickson, C. J.; Giblin, L. J.; Rengarajan, B.; Masalha, R.; Frederickson, C. J.; Zeng, Y.; Lopez, E. V.; Koh, J.-Y.; Chorin, U.; Besser, L.; Hershfinkel, M.; Li, Y.; Thompson, R. B.; Krezel, A. J. Neurosci. Meth. 2006, 154, 19. (33) Schlief, M. L.; Craig, A. M.; Gitlin, J. D. J. Neurosci. 2005, 25, 239. (34) Schlief, M. L.; West, T.; Craig, A. M.; Holtzman, D. M.; Gitlin, J. D. Proc. Natl. Acad. Sci. 2006, 103, 14919. (35) Pike, C. L.; Walencewicz, A. J.; Glabe, C. G.; Cotman, C. W. Eur. J. Pharmacol. 1991, 207, 367. (36) Rauk, A. Dalton Trans. 2008, 1273 (37) Dahlgren, K. N.; Manelli, A. M.; Stine, W. B., Jr.; Baker, L. K.; Krafft, G. A.; LaDu, M. J. J. Biol. Chem. 2002, 277, 32046. (38) Cappai, R.; Barnham, K. Neurochem. Res. 2008, 33, 526. (39) Shah, R. S.; Lee, H.-G.; Xiongwei, Z.; Perry, G.; Smith, M. A.; Castellani, R. J. Biomed. Pharmacother. 2008, 62, 199. (40) Morita, A.; Kimura, M.; Itokawa, Y. Biol. Trace Elem. Res. 1994, 42, 165. (41) Takahashi, S.; Takahashi, I.; Sato, H.; Kubota, Y.; Yoshida, S.; Muramatsu, Y. Biol. Trace Elem. Res. 2001, 80, 145. (42) Maynard, C. J.; Cappai, R.; Volitakis, I.; Cherny, R. A.; White, A. R.; Beyreuther, K.; Masters, C. L.; Bush, A. I.; Li, Q.-X. J. Biol. Chem. 2002, 277, 44670. (43) Smith, D. G.; Cappai, R.; Barnham, K. J. Biochim. Biophys. Acta, Biomembr. 2007, 1768, 1976. (44) Miller, L. M.; Wang, Q.; Telivala, T. P.; Smith, R. J.; Lanzirotti, A.; Miklossy, J. J. Struct. Biol. 2006, 155, 30. 128  (45) Lovell, M. A.; Robertson, J. D.; Teesdale, W. J.; Campbell, J. L.; Markesbery, W. R. J. Neurol. Sci. 1998, 158, 47. (46) Linder, M. C.; Hazegh-Azam, M. Am. J. Clin. Nutr. 1996, 63, 797. (47) Stuerenburg, H. J. J. Neural Transm. 2000, 107, 321. (48) Schumann, K.; Classen, H. G.; Dieter, H. H.; Konig, J.; Multhaup, G.; Rukgauer, M.; Summer, K. H.; Bernhardt, J.; Biesalski, H. K. Eur. J. Clin. Nutr. 2002, 56, 469. (49) Tandon, L.; Ni, B.; Ding, X.; Ehmann, W.; Kasarskis, E.; Markesbery, W. J. Radioanal. Nucl. Chem. 1994, 179, 331. (50) Ward, N.; Mason, J. J. Radioanal. Nucl. Chem. 1987, 113, 515. (51) Deibel, M. A.; Ehmann, W. D.; Markesbery, W. R. J. Neurol. Sci. 1996, 143, 137. (52) 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. (53) Lee, J.-Y.; Mook-Jung, I.; Koh, J.-Y. J. Neurosci. 1999, 19, RC10 1. (54) Corrigan, F. M.; Reynolds, G. P.; Ward, N. I. BioMetals 1993, 6, 149. (55) Stoltenberg, M.; Bruhn, M.; Sondergaard, C.; Doering, P.; West, M.; Larsen, A.; Troncoso, J. C.; Danscher, G. Histochem. Cell Biol. 2005, 123, 605. (56) Hershey, C. O.; Hershey, L. A.; Varnes, A.; Vibhakar, S. D.; Lavin, P.; Strain, W. H. Neurology 1983, 33, 1350. (57) Goodman, L. J. Nerv. Ment. Dis. 1953, 117, 97. (58) Samudralwar, D. L.; Diprete, C. C.; Ni, B. F.; Ehmann, W. D.; Markesbery, W. R. J. Neurol. Sci. 1995, 130, 139. (59) Thompson, C. M.; Markesbery, W. R.; Ehmann, W. D.; Mao, Y.-X.; Vance, D. E. NeuroToxicology 1988, 9, 1. (60) Good, P. F.; Perl, D. P.; Bierer, L. M.; Schmeidler, J. Ann. Neurol. 1992, 31, 286. (61) Opazo, C.; Huang, X.; Cherny, R. A.; Moir, R. D.; Roher, A. E.; White, A. R.; Cappai, R.; Masters, C. L.; Tanzi, R. E.; Inestrosa, N. C.; Bush, A. I. J. Biol. Chem. 2002, 277, 40302. (62) Grundke-Iqbal, I.; Fleming, J.; Tung, Y.-C.; Lassmann, H.; Iqbal, K.; Joshi, J. G. Acta Neuropathol. 1990, 81, 105. (63) Bush, A. I.; Multhaup, G.; Moir, R. D.; Williamson, T. G.; Small, D. H.; Rumble, B.; Pollwein, P.; Beyreuther, K.; Masters, C. L. J. Biol. Chem. 1993, 268, 16109. (64) Hesse, L.; Beher, D.; Masters, C. L.; Multhaup, G. FEBS Lett. 1994, 349, 109. (65) Kong, G.; Miles, L.; Crespi, G.; Morton, C.; Ng, H.; Barnham, K.; McKinstry, W.; Cappai, R.; Parker, M. Eur. Biophys. J. 2008, 37, 269. (66) Cuajungco, M. P.; Faget, K. Y.; Huang, X.; Tanzi, R. E.; Bush, A. I. Ann. N.Y. Acad. Sci. 2000, 920, 292. (67) Wolfe, M. S.; Guenette, S. Y. J. Cell Sci. 2007, 120, 3157. 129  (68) Bush, A. I.; Pettingell, W. H.; Multhaup, G.; de Paradis, M.; Vonsattel, J.-P.; Gusella, J. F.; Beyreuther, K.; Masters, C. L.; Tanzi, R. E. Science 1994, 265, 1464. (69) Atwood, C. S.; Moir, R. D.; Huang, X.; Scarpa, R. C.; Bacarra, M. E.; Romano, D. M.; Hartshorn, M. A.; Tanzi, R. E.; Bush, A. I. J. Biol. Chem. 1998, 273, 12817. (70) Bush, A. I.; Pettingell, W. H.; de Paradis, M.; Tanzi, R. E. J. Biol. Chem. 1994, 269, 12152. (71) Miura, T.; Suzuki, K.; Kohata, N.; Takeuchi, H. Biochemistry 2000, 39, 7024. (72) 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. (73) Curtain, C. C.; Ali, F.; Volitakis, I.; Cherny, R.; Norton, R. S.; Beyreuther, K.; Barrow, C. J.; Masters, C. L.; Bush, A. I.; Barnham, K. J. J. Biol. Chem. 2001, 276, 20466. (74) Karr, J. W.; Akintoye, H.; Kaupp, L. J.; Szalai, V. A. Biochemistry 2005, 44, 5478. (75) Syme, C. D.; Nadal, R. C.; Rigby, S. E. J.; Viles, J. H. J. Biol. Chem. 2004, 279, 18169. (76) Streltsov, V. A.; Titmuss, S. J.; Epa, V. C.; Barnham, K. J.; Masters, C. L.; Varghese, J. N. Biophys. J. 2008, 95, 3447. (77) Garzon-Rodriguez, W.; Yatsimirsky, A. K.; Glabe, C. G. Bioorg. Med. Chem. Lett. 1999, 9, 2243. (78) Klug, G. M. J. A.; Losic, D.; Subasinghe, S. S.; Aguilar, M.-I.; Martin, L. L.; Small, D. H. Eur. J. Biochem. 2003, 270, 4282. (79) Hu, W.-P.; Chang, G.-L.; Chen, S.-J.; Kuo, Y.-M. J. Neurosci. Meth. 2006, 154, 190. (80) Ha, C.; Ryu, J.; Park, C. B. Biochemistry 2007, 46, 6118. (81) Smith, D. P.; Ciccotosto, G. D.; Tew, D. J.; Fodero-Tavoletti, M. T.; Johanssen, T.; Masters, C. L.; Barnham, K. J.; Cappai, R. Biochemistry 2007, 46, 2881. (82) 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. (83) 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.; Glabe, C.; Bowden, E. F.; Masters, C. L.; Fairlie, D. P.; Tanzi, R. E.; Bush, A. I. J. Biol. Chem. 1999, 274, 37111. (84) Khan, A.; Dobson, J. P.; Exley, C. Free Radic. Biol. Med. 2006, 40, 557. (85) Anbar, M.; Neta, P. Int. J. Appl. Radiat. Isot. 1967, 18, 493. (86) Atwood, C. S.; Perry, G.; Zeng, H.; Kato, Y.; Jones, W. D.; Ling, K. Q.; Huang, X.; Moir, R. D.; Wang, D.; Sayre, L. M.; Smith, M. A.; Chen, S. G.; Bush, A. I. Biochemistry 2004, 43, 560. (87) Cuajungco, M. P.; Goldstein, L. E.; Nunomura, A.; Smith, M. A.; Lim, J. T.; Atwood, C. S.; Huang, X.; Farrag, Y. W.; Perry, G.; Bush, A. I. J. Biol. Chem. 2000, 275, 19439. (88) Perl, D. P.; Brody, A. R. Science 1980, 208, 297. 130  (89) Lovell, M. A.; Ehmann, W. D.; Markesbery, W. R. Ann. Neurol. 1993, 33, 36. (90) Martyn, C. N.; Barker, D. J. P.; Osmond, C.; Harris, E. C.; Edwardson, J. A.; Lacey, R. F. Lancet 1989, 1, 59. (91) 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. (92) Kawahara, M. J. Alzheimers Dis. 2005, 8, 171. (93) Walton, J. R. NeuroToxicology 2006, 27, 385. (94) Ribes, D.; Colomina, M. T.; Vicens, P.; Domingo, J. L. Exp. Neurol. 2008, 214, 293. (95) Savory, J.; Herman, M. M.; Ghribi, O. J. Alzheimers Dis. 2006, 10, 135. (96) Bush, A. I.; Tanzi, R. E. Neurotherapeutics 2008, 5, 421. (97) Dedeoglu, A.; Cormier, K.; Payton, S.; Tseitlin, K. A.; Kremsky, J. N.; Lai, L.; Li, X.; Moir, R. D.; Tanzi, R. E.; Bush, A. I. Exp. Gerontol. 2004, 39, 1641. (98) Rodriguez-Rodriguez, C.; Sanchez de Groot, N.; Rimola, A.; Alvarez-Larena, A.; Lloveras, V.; Vidal-Gancedo, J.; Ventura, S.; Vendrell, J.; Sodupe, M.; Gonzalez-Duarte, P. J. Am. Chem. Soc. 2009, 131, 1436. (99) Pardridge, W. M.; Boado, R. J.; Farrell, C. R. J. Biol. Chem. 1990, 265, 18035. (100) Storr, T.; Merkel, M.; Song-Zhao, G. X.; Scott, L. E.; Green, D. E.; Bowen, M. L.; Thompson, K. H.; Patrick, B. O.; Schugar, H. J.; Orvig, C. J. Am. Chem. Soc. 2007, 129, 7453. (101) Storr, T.; Scott, L. E.; Bowen, M. L.; Green, D. E.; Thompson, K. H.; Schugar, H. J.; Orvig, C. Dalton Trans. 2009, 3034. (102) Liu, G.; Garrett, M. R.; Men, P.; Zhu, X.; Perry, G.; Smith, M. A. Biochim. Biophys. Acta, Mol. Basis Dis. 2005, 1741, 246. (103) Schroeder, U.; Sommerfeld, P.; Ulrich, S.; Sabel, B. A. J. Pharm. Sci. 1998, 87, 1305. (104) Kreuter, J.; Shamenkov, D.; Petrov, V.; Ramge, P.; Cychutek, K.; Koch-Brandt, C.; Alyautdin, R. J. Drug Target. 2002, 10, 317 (105) Cui, Z.; Lockman, P. R.; Atwood, C. S.; Hsu, C.-H.; Gupte, A.; Allen, D. D.; Mumper, R. J. Eur. J. Pharm. Biopharm. 2005, 59, 263. (106) Wood, S. J.; Maleeff, B.; Hart, T.; Wetzela, R. J. Mol. Biol. 1996, 256, 870. (107) Scott, L. E.; Page, B. D. G.; Patrick, B. O.; Orvig, C. Dalton Trans. 2008, 6364. (108) Inbar, P.; Bautista, M. R.; Takayama, S. A.; Yang, J. Anal. Chem. 2008, 80, 3502. (109) Leliveld, S. R.; Korth, C. J. Neurosci. Res. 2007, 85, 2285. (110) LeVine, H. Methods Enzymol. 1999, 309, 274. (111) Niidome, T.; Takahashi, K.; Goto, Y.; Goh, S.; Tanaka, N.; Kamei, K.; Ichida, M.; Hara, S.; Akaike, A.; Kihara, T.; Sugimoto, H. NeuroReport 2007, 18, 813. (112) Kokkoni, N.; Stott, K.; Amijee, H.; Mason, J. M.; Doig, A. J. Biochemistry 2006, 45, 9906. 131  (113) Byeon, S. R.; Lee, J. H.; Sohn, J.-H.; Kim, D. C.; Shin, K. J.; Yoo, K. H.; Mook-Jung, I.; Lee, W. K.; Kim, D. J. Bioorg. Med. Chem. Lett. 2007, 17, 1466. (114) LeVine, H. Arch. Biochem. Biophys. 2002, 404, 106. (115) Cherny, R. A.; Barnham, K. J.; Lynch, T.; Volitakis, I.; Li, Q.-X.; McLean, C. A.; Multhaup, G.; Beyreuther, K.; Tanzi, R. E.; Masters, C. L.; Bush, A. I. J. Struct. Biol. 2000, 130, 209. (116) Sayre, L. M.; Perry, G.; Harris, P. L. R.; Liu, Y.; Schubert, K. A.; Smith, M. A. J. Neurochem. 2000, 74, 270. (117) Fonte, J.; Miklossy, J.; Atwood, C.; Martins, R. J. Alzheimers Dis. 2001, 3, 209. (118) 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. (119) Gotz, J.; Ittner, L. M. Nat. Rev. Neurosci. 2008, 9, 532. (120) Hsiao, K.; Chapman, P.; Nilsen, S.; Eckman, C.; Harigaya, Y.; Younkin, S.; Yang, F.; Cole, G. Science 1996, 274, 99. (121) Bartov, O.; Sultana, R.; Butterfield, D. A.; Atlas, D. Brain Res. 2006, 1069, 198. (122) Moreira, P. I.; Harris, P. L. R.; Zhu, X.; Santos, M. S.; Oliveira, C. R.; Smith, M. A.; Perry, G. J. Alzheimers Dis. 2007, 12, 195. (123) 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. (124) Molina-Holgado, F.; Gaeta, A.; Francis, P. T.; Williams, R. J.; Hider, R. C. J. Neurochem. 2008, 105, 2466. (125) Lee, J.-Y.; Friedman, J. E.; Angel, I.; Kozak, A.; Koh, J.-Y. Neurobiol. Aging 2004, 25, 1315. (126) Kruck, T. P.; Cui, J.-G.; Percy, M. E.; Lukiw, W. J. Cell Mol. Neurobiol. 2004, 24, 443. (127) Shin, R.-W.; Kruck, T. P. A.; Murayama, H.; Kitamoto, T. Brain Res. 2003, 961, 139. (128) Van Gool, W. A.; Weinstein, H. C.; Scheltens, P. K.; Walstra, G. J. M. Lancet 2001, 358, 455. (129) Deraeve, C.; Pitié, M.; Mazarguil, H.; Meunier, B. New J. Chem. 2007, 31, 193. (130) Deraeve, C.; Boldron, C.; Maraval, A.; Mazarguil, H.; Gornitzka, H.; Vendier, L.; Pitié, M.; Meunier, B. Chem. Eur. J. 2008, 14, 682. (131) Moret, V.; Laras, Y.; Pietrancosta, N.; Garino, C.; Quéléver, G.; Rolland, A.; Mallet, B.; Norreel, J.-C.; Kraus, J.-L. Bioorg. Med. Chem. Lett. 2006, 16, 3298. (132) Lannfelt, L.; Blennow, K.; Zetterberg, H.; Batsman, S.; Ames, D.; Harrison, J.; Masters, C. L.; Targum, S.; Bush, A. I.; Murdoch, R.; Wilson, J.; Ritchie, C. W. Lancet Neurol. 2008, 7, 779. 132  (133) 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.; 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. (134) Arakawa, M.; Ito, Y. Cerebellum 2007, 6, 308. (135) Offen, D.; Gilgun-Sherki, Y.; Barhum, Y.; Benhar, M.; Grinberg, L.; Reich, R.; Melamed, E.; Atlas, D. J. Neurochem. 2004, 89, 1241. (136) Ates, B.; Abraham, L.; Ercal, N. Free Radic. Res. 2008, 42, 372 (137) Amer, J.; Atlas, D.; Fibach, E. Biochim. Biophys. Acta, Gen. Subj. 2008, 1780, 249. (138) Smith, M. A.; Harris, P. L. R.; Sayre, L. M.; Perry, G. Proc. Natl. Acad. Sci. 1997, 94, 9866. (139) Daar, S.; Pathare, A. Ann. Hematol. 2006, 85, 315. (140) Hider, R. C.; Hall, A. D. Prog. Med. Chem. 1991, 28, 41. (141) Kontoghiorghes, G. J. Analyst 1995, 120, 845. (142) 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. (143) Bebbington, D.; Dawson, C. E.; Gaur, S.; Spencer, J. Bioorg. Med. Chem. Lett. 2002, 12, 3297. (144) Scott, L. E.; Orvig, C.; unpublished work., 2007. (145) Kruck, T. P. A.; Burrow, T. E. J. Inorg. Biochem. 2002, 88, 19. (146) Green, D. E.; Ferreira, C. L.; Stick, R. V.; Patrick, B. O.; Adam, M. J.; Orvig, C. Bioconjugate Chem. 2005, 16, 1597. (147) Lopresti, B. J.; Klunk, W. E.; Mathis, C. A.; Hoge, J. A.; Ziolko, S. K.; Lu, X.; Meltzer, C. C.; Schimmel, K.; Tsopelas, N. D.; DeKosky, S. T.; Price, J. C. J. Nucl. Med. 2005, 46, 1959. (148) Kozak, A.; Shapiro, I. Lipophilic diesters of chelating agents. World Patent 1999/16741, Apr. 8, 1999. (149) Zhuang, Z. P.; Kung, M. P.; Kung, H. F. J. Med. Chem. 2006, 49, 2841. (150) De Clercq, E. Nat. Rev. Drug Discov. 2003, 2, 581. (151) Swaiman, K. F.; Machen, V. L. J. Neurochem. 1986, 46, 652. (152) Legssyer, R.; Ward, R. J.; Crichton, R. R.; Boelaert, J. R. Biochem. Pharmacol. 1999, 57, 907. (153) Padmanabhan, G.; Becue, I.; Smith, J. A. In Analytical Profiles of Drug Substances; Florey, K., Ed.; Academic Press: New York, 1989, 57-90. (154) Schafer, S.; Pajonk, F.-G.; Multhaup, G.; Bayer, T. J. Mol. Med. 2007, 85, 405. (155) Opazo, C.; Luza, S.; Villemagne, V. L.; Volitakis, I.; Rowe, C.; Barnham, K. J.; Strozyk, D.; Masters, C. L.; Cherny, R. A.; Bush, A. I. Aging Cell 2006, 5, 69. 133  (156) White, A. R.; Du, T.; Laughton, K. M.; Volitakis, I.; Sharples, R. A.; Xilinas, M. E.; Hoke, D. E.; Holsinger, R. M. D.; Evin, G.; Cherny, R. A.; Hill, A. F.; Barnham, K. J.; Li, Q.-X.; Bush, A. I.; Masters, C. L. J. Biol. Chem. 2006, 281, 17670. (157) Barnham, K. J.; Gautier, E. C. L.; Kok, G. B.; Krippner, G. 8-Hydroxyquinoline derivatives. World Patent 2004/007461, Jan. 22, 2004. (158) Adlard, P. A.; Cherny, R. A.; Finkelstein, D. I.; Gautier, E.; Robb, E.; Cortes, M.; Volitakis, I.; Liu, X.; Smith, J. P.; Perez, K.; Laughton, K.; Li, Q.-X.; Charman, S. A.; Nicolazzo, J. A.; Wilkins, S.; Deleva, K.; Lynch, T.; Kok, G.; Ritchie, C. W.; Tanzi, R. E.; Cappai, R.; Masters, C. L.; Barnham, K. J.; Bush, A. I. Neuron 2008, 59, 43. (159) Deraeve, C.; Maraval, A.; Vendier, L.; Faugeroux, V.; Pitié, M.; Meunier, B. Eur. J. Inorg. Chem. 2008, 5622. (160) Villemagne, V.; Fodero-Tavoletti, M.; Pike, K.; Cappai, R.; Masters, C.; Rowe, C. Mol. Neurobiol. 2008, 38, 1. (161) Klunk, W. E.; Debnath, M. L.; Pettegrew, J. W. Neurobiol. Aging 1994, 15, 691. (162) Kung, M.-P.; Hou, C.; Zhuang, Z.-P.; Zhang, B.; Skovronsky, D.; Trojanowski, J. Q.; Lee, V. M. Y.; Kung, H. F. Brain Res. 2002, 956, 202. (163) Klunk, W. E.; Wang, Y.; Huang, G.; Debnath, M. L.; Holt, D. P.; Mathis, C. A. Life Sci. 2001, 69, 1471. (164) Henriksen, G.; Hauser, A. I.; Westwell, A. D.; Behrroz, H. Y.; Schwaiger, M.; Drzezga, A.; Wester, H.-J. J. Med. Chem. 2007, 50, 1087. (165) Ono, M.; Kawashima, H.; Nonaka, A.; Kawai, T.; Haratake, M.; Mori, H.; Kung, M. P.; Kung, H. F.; Saji, H.; Nakayama, M. J. Med. Chem. 2006, 49, 2725. (166) Ono, M.; Yoshida, N.; Ishibashi, K.; Haratake, M.; Arano, Y.; Mori, H.; Nakayama, M. J. Med. Chem. 2005, 48, 7253. (167) Ono, M.; Wilson, A.; Nobrega, J.; Westaway, D.; Verhoeff, P.; Zhuang, Z.-P.; Kung, M.- P.; Kung, H. F. Nucl. Med. Biol. 2003, 30, 565. (168) Suemoto, T.; Okamura, N.; Shiomitsu, T.; Suzuki, M.; Shimadzu, H.; Akatsu, H.; Yamamoto, T.; Kudo, Y.; Sawada, T. Neurosci. Res. 2004, 48, 65. (169) Shoghi-Jadid, K.; Small, G. W.; Agdeppa, E. D.; Kepe, V.; Ercoli, L. M.; Siddarth, P.; Read, S.; Satyamurthy, N.; Petric, A.; Huang, S.-C.; Barrio, J. R. Am. J. Geriatr. Psychiatr. 2002, 10, 24. (170) Verhoeff, N.; Wilson, A.; Takeshita, S.; Trop, L.; Hussey, D.; Singh, K.; Kung, H.; Kung, M.-P.; Houle, S. Am. J. Geriatr. Psychiatr. 2004, 12, 584. (171) Mathis, C. A.; Wang, Y.; Holt, D. P.; Huang, G. F.; Debnath, M. L.; Klunk, W. E. J. Med. Chem. 2003, 46, 2740. (172) Klunk, W. E.; Engler, H.; Nordberg, A.; Wang, Y.; Blomqvist, G.; Holt, D. P.; Bergström, M.; Savitcheva, I.; Huang, G.-F.; Estrada, S.; Ausén, B.; Debnath, M. L.; Barletta, J.; Price, J. C.; Sandell, J.; Lopresti, B. J.; Wall, A.; Koivisto, P.; Antoni, G.; Mathis, C. A.; Långström, B. Ann. Neurol. 2004, 55, 306. 134  (173) Seibyl, J.; Jennings, D.; Koren, A.; Skovronsky, D.; Tamagnan, G.; Marek, K. J. Nucl. Med. 2007, 48, 57P. (174) Morphy, R.; Rankovic, Z. J. Med. Chem. 2005, 48, 6523. (175) Cavalli, A.; Bolognesi, M. L.; Minarini, A.; Rosini, M.; Tumiatti, V.; Recanatini, M.; Melchiorre, C. J. Med. Chem. 2008. (176) Feng, M.-H.; van der Does, L.; Bantjes, A. J. Med. Chem. 1993, 36, 2822. (177) Hershko, C.; Theanacho, E. N.; Spira, D. T.; Peter, H. H.; Dobbin, P.; Hider, R. C. Blood 1991, 77, 637. (178) Timeus, F.; Valle, P.; Crescenzio, N.; Ruggieri, L.; Rosso, P.; Luca Pagliardi, G.; Cordero di Montezemolo, L.; Gabutti, V.; Ramenghi, U. Am. J. Hematol. 1994, 47, 183. (179) Ozturk, G.; Erol, D. D.; Aytemir, M. D.; Uzbay, T. Eur. J. Med. Chem. 2002, 37, 829. (180) Aytemir, M. D.; Uzbay, T.; Erol, D. D. Drug Res. 1999, 49, 250. (181) Kontoghiorghes, G. J.; Aldouri, M. A.; Sheppard, L.; Hoffbrand, A. V. Lancet 1987, 1, 1294. (182) Gralla, E. J.; Stebbins, R. B.; Coleman, G. L.; Delahunt, C. S. Toxicol. Appl. Pharmacol. 1969, 15, 604. (183) Rennhard, H. H. J. Agric. Food Chem. 1971, 19, 152. (184) Ellis, B. L.; Duhme, A. K.; Hider, R. C.; Hossain, M. B.; Rizvi, S.; van der Helm, D. J. Med. Chem. 1996, 39, 3659. (185) Mathis, C. A.; Wang, Y.; Klunk, W. E. Curr. Pharmaceut. Des. 2004, 10, 1469. (186) Gyurcsik, B.; Nagy, L. Coord. Chem. Rev. 2000, 203, 81. (187) Battaglia, G.; La Russa, M.; Bruno, V.; Arenare, L.; Ippolito, R.; Copani, A.; Bonina, F.; Nicoletti, F. Brain Res. 2000, 860, 149. (188) 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. (189) Cheng, H.; Cao, X.; Xian, M.; Fang, L.; Cai, T. B.; Ji, J. J.; Tunac, J. B.; Sun, D.; Wang, P. G. J. Med. Chem. 2005, 48, 645. (190) Matern, H.; Boermans, H.; Lottspeich, F.; Matern, S. J. Biol. Chem. 2001, 276, 37929. (191) Coutinho, P. M.; Henrissat, B. Carbohydrate-Active Enzymes server at http://www.cazy.org/. Accessed February 2009. (192) Henrissat, B.; Callebaut, I.; Fabrega, S.; Lehn, P.; Mornon, J. P.; Davies, G. Proc. Natl. Acad. Sci. U. S. A. 1996, 93, 5674. (193) Green, D. E., Ph.D. thesis, The University of British Columbia, 2004. (194) Campbell, K. N.; Ackerman, J. F.; Campbell, B. K. J. Org. Chem. 1950, 15, 221. (195) Harris, R. L. N. Aust. J. Chem. 1976, 29, 1329. (196) Zhang, Z.; Rettig, S. J.; Orvig, C. Inorg. Chem. 1991, 30, 509. 135  (197) Barta, C. A.; Sachs-Barrable, K.; Jia, J.; Thompson, K. H.; Wasan, K. M.; Orvig, C. Dalton Trans. 2007, 5019 (198) Kitagawa, H.; Kumura, K.; Takahata, S.; Iida, M.; Atsumi, K. Bioorg. Med. Chem. 2007, 15, 1106. (199) Ma, Y.; Luo, W.; Quinn, P. J.; Liu, Z.; Hider, R. C. J. Med. Chem. 2004, 47, 6349. (200) Imafuku, K.; Takahashi, K.; Matsumura, H. Bull. Chem. Soc. Jpn. 1979, 52, 111. (201) Nelson, W. O.; Karpishin, T. B.; Rettig, S. J.; Orvig, C. Can. J. Chem. 1988, 66, 123. (202) Zhang, Z.; Rettig, S. J.; Orvig, C. Can. J. Chem. 1992, 70, 763. (203) Clarke, E. T.; Martell, A. E.; Reibenspies, J. Inorg. Chim. Acta 1992, 196, 177. (204) Yasue, M.; Kawamura, N.; Sakakibara, J. Journal of the Pharmaceutical Society of Japan 1970, 90, 1222. (205) Mitsunobu, O.; Yamada, M. Bull. Chem. Soc. Jpn. 1967, 40, 2380. (206) Turner, J. J.; Filippov, D. V.; Overhand, M.; van der Marel, G. A.; van Boom, J. H. Tetrahedron Lett. 2001, 42, 5763. (207) Chida, N.; Ohtsuka, M.; Nakazawa, K.; Ogawa, S. The Journal of Organic Chemistry 1991, 56, 2976. (208) Kroger, L.; Thiem, J. J. Carbohydr. Chem. 2003, 22, 9. (209) Nelson, W. O.; Orvig, C.; Rettig, S. J.; Trotter, J. Can. J. Chem. 1988, 66, 132. (210) Stick, R. V. Carbohydrates: the sweet molecules of life; Academic Press: New York, 2001. (211) Kontoghiorghes, G. J. Hemoglobin 2006, 30, 329. (212) Dehkordi, L. S.; Liu, Z. D.; Hider, R. C. Eur. J. Med. Chem. 2008, 43, 1035. (213) Kalinowski, D. S.; Richardson, D. R. Pharmacol. Rev. 2005, 57, 547. (214) Dobbin, P. S.; Hider, R. C.; Hall, A. D.; Taylor, P. D.; Sarpong, P.; Porter, J. B.; Xiao, G.; van der Helm, D. J. Med. Chem. 1993, 36, 2448. (215) Re, R.; Pellegrini, N.; Proteggente, A.; Pannala, A.; Yang, M.; Rice-Evans, C. Free Radic. Biol. Med. 1999, 26, 1231. (216) Wijtmans, M.; Pratt, D. A.; Valgimigli, L.; DiLabio, G. A.; Pedulli, G. F.; Porter, N. A. Angew. Chem., Int. Ed. 2003, 42, 4370. (217) Mosmann, T. J. Immunol. Methods 1983, 65, 55. (218) Bush, A. I. Trends Neurosci. 2003, 26, 207. (219) Freshney, R. I. Culture of animal cells: a manual of basic technique; 4th ed.; Wiley-Liss: New York, 2000. (220) Walker, J. M. In The Protein Protocols Handbook; 2nd ed.; Walker, J. M., Ed.; Humana Press: New Jersey, 2002. (221) Ahmed, S. I.; Burgess, J.; Fawcett, J.; Parsons, S. A.; Russell, D. R.; Laurie, S. H. Polyhedron 2000, 19, 129. 136  (222) Katritzky, A. R. Physical methods in heterocyclic chemistry; Academic Press: New York, 1971. (223) Katritzky, A. R.; Jones, R. A. J. Chem. Soc. 1960, 2947. (224) Coates, J. In Encyclopedia of Analytical Chemistry; Meyers, R. A., Ed.; John Wiley and Sons: Chichester, 2000. (225) Nakamoto, K. Infrared and Raman Spectra of Inorganic and Organic Compounds; 3rd ed.; Wiley-Interscience: New York, 1978. (226) Orvig, C. J. Chem. Educ. 1985, 62, 84. (227) Charalambous, J.; Dodd, A.; McPartlin, M.; Matondo, S. O. C.; Pathirana, N. D.; Powell, H. R. Polyhedron 1988, 7, 2235. (228) El-Jammal, A.; Howell, P. L.; Turner, M. A.; Li, N.; Templeton, D. M. J. Med. Chem. 1994, 37, 461. (229) Chruscinska, E.; Garribba, E.; Micera, G.; Panzanelli, A. J. Inorg. Biochem. 1999, 75, 225. (230) Javitt, N. B. FASEB J. 1990, 4, 161. (231) Hider, R. C.; Kontoghiorghes, G.; Silver, J. Pharmaceutical compositions. U.K. Patent 1984/2136807, Sept. 26, 1984. (232) Huang, X.; Atwood, C. S.; Moir, R. D.; Hartshorn, M. A.; Vonsattel, J.-P.; Tanzi, R. E.; Bush, A. I. J. Biol. Chem. 1997, 272, 26464. (233) Lockhart, A. Drug Discov. Today 2006, 11, 1093. (234) Voropai, E. S.; Samtsov, M. P.; Kaplevskii, K. N.; Maskevich, A. A.; Stepuro, V. I.; Povarova, O. I.; Kunznetsova, I. M.; Turoverov, K. K.; Fink, A. L.; Uverskii, V. N. J. Appl. Spectros. 2003, 70, 868. (235) Hu, M.; Ling, J.; Lin, H.; Chen, J. In Optimization in Drug Discovery, 2004. (236) Kerns, E. H.; Di, L. Drug-like properties: concepts, structure design and methods: from ADME to toxicity optimization; Academic Press: New York, 2008. (237) Cai, L.; Liow, J.-S.; Zoghbi, S. S.; Cuevas, J.; Baetas, C.; Hong, J.; Shetty, H. U.; Seneca, N. M.; Brown, A. K.; Gladding, R.; Temme, S. S.; Herman, M. M.; Innis, R. B.; Pike, V. W. J. Med. Chem. 2008, 51, 148. (238) Charkoudian, L. K.; Pham, D. M.; Kwon, A. M.; Vangeloff, A. D.; Franz, K. J. Dalton Trans. 2007, 5031.    137  Appendix  Crystallographic Data for 3-Hydroxy-4-Pyridinone Pro-Ligands  All crystal data were collected and solved by Dr. Brian O. Patrick, Structural Chemistry Facility, UBC. For each compound a brief outline is provided of the crystallisation procedures and nature of the crystal. These are followed by the crystal data and, separately, selected bond lengths and angles in tabulated form (Tables A.1, A.2-3, respectively).  Hdapp  Figure A.1. Ellipsoid plot (50 % probability; for clarity, H-atoms not shown) of Hdapp.  X-ray quality crystals of Hdapp were obtained by slow evaporation (4 °C) of the entire contents of the reaction flask (MeOH:CH2Cl2). Forming large, colourless plates, the pro-ligand co- crystallises with one molecule of MeOH. The average torsion angle formed between the two ring systems is 72.6 °; selected bond lengths and angles are given in Table A.2. 138  Hsapp  Figure A.2. Ellipsoid plot (50 % probability; for clarity, H-atoms not shown) of Hsapp.  Pink plate-like X-ray quality crystals of Hsapp were obtained by evaporation of a methanolic solution (99:1 MeOH:H2O, RT). The unit cell includes two molecules of pro-ligand co- crystallised with two molecules of H2O. Hydrogen bonding is observed with donation from the p-NHCH3 groups to the H2O molecules, and between the hydroxyl groups to the carbonyl groups in both intra- and inter-molecular fashions. The H2O molecules show further hydrogen bonding within themselves and to the carbonyl and hydroxyl groups of the pyridinone. The average torsion angle formed between the two ring systems of the pyridinone is similar to that in Hdapp at 75.6 °. Selected bond lengths and angles in the pro-ligand are given in Table A.2.  139  Hzapp  Figure A.3. Ellipsoid plot (50 % probability; for clarity, H-atoms not shown) of Hzapp.  Colourless plate crystals of Hzapp were formed from slow evaporation of a 4:1 MeOH:H2O solution of the compound at RT. The unit cell includes two molecules of pro-ligand co- crystallised in a “head-to-toe” fashion with one molecule each of MeOH and H2O included. Hydrogen bonding is observed between the primary amino group (para-substituent on the phenyl ring) and, separately, the carbonyl oxygen and these solvent molecules. Both intramolecular (to the α-keto group) and intermolecular hydrogen bonding is observed between the hydroxyl groups and the carbonyl groups. Consistent with those found in the above structures for Hdapp and Hsapp, the average torsion angle between the two ring systems of Hzapp is 67.3 °. Selected bond lengths and angles in the pyridinone pro-ligand are given in Table A.2.   140  Hbt6p  Figure A.4. Ellipsoid plot (50 % probability; for clarity, H-atoms not shown) of Hbt6p.  Slow evaporation of a mixture of Hbt6p and Gbt6p (MeOH, RT) yielded colourless prism crystals of Hbt6p. Like Hdapp, the pro-ligand co-crystallises with one molecule of MeOH. The average torsion angle formed between the two ring systems is similar to those of the pyridinones discussed above at 72.7 °. Selected bond lengths and angles are given in Table A.3.  Hbt2p  Figure A.5. Ellipsoid plot (50 % probability; for clarity, H-atoms not shown) of Hbt2p.  141  Slow evaporation of a solution of Hbt2p (2:1 MeOH:CH2Cl2, RT) yielded orange needle-like X- ray quality crystals. The torsion angle formed between the two ring systems is the furthest from perpendicular of all the pro-ligands, with an average value of 55.8 °. Selected bond lengths and angles are given in Table A.3.  Hnbp  Figure A.6. Ellipsoid plot (50 % probability; for clarity, H-atoms not shown) of Hnbp.  Colourless plate crystals of Hnbp were obtained from CHCl3 by slow evaporation at 4 °C. The pro-ligand co-crystallises with one molecule of CHCl3. Selected bond lengths are given in Table A.3; because of the extra methylene linker between the pyridinone and phenyl rings, dihedral angles formed across C(6)/C(2)-N(1)-C(7)/C(12) are not present and therefore not reported, unlike for all other pro-ligand structures. Cambridge crystallographic data centre (CCDC) entry 697418.   142  Table A.1. Crystallographic data for Hdapp, Hsapp, Hzapp, Hbt6p, Hbt2p and Hnbp. crystal data Hdapp • MeOH Hsapp • MeOH 2Hzapp• MeOH•H2O Hbt6p • MeOH Hbt2p Hnbp • CHCl3 empirical formula C15H20N2O3 C13H16N2O3 C25H30N4O6 C14H14N2O3S C13H10N2O2S C13H13NO2 .CHCl3 fw 276.15 248.28 482.52 290.33 258.29 334.61 crystal system,  space group triclinic, P -1 (#2) monoclinic, P 21 (#4) monoclinic, P 21/n  (#14) triclinic, P -1 (#2) monoclinic, P 21/c (#14) monoclinic, P 21/c (#14) a (Å) 7.5438(8)  8.2835(4) 12.9650(17) 11.368(2) 10.3006(12) 11.7981(13) b (Å) 9.0878(9) 13.9790(7) 14.3780(18) 11.457(2) 7.8737(10) 10.8093(13) c (Å) 11.3719(12) 10.5581(5) 12.9795(16) 12.809(2) 13.7999(16) 12.5786(15) α (deg) 85.485(6) 90.0 90.0 67.68(1) 90.0 90.0 β (deg) 78.375(6) 90.255(2) 90.445(8) 76.79(1) 93.837(4) 107.760(6) γ (deg) 69.760(5) 90.0 90.0 74.17(1) 90.0 90.0 V [Å3] 716.44(13) 1222.56(10) 2419.4(5) 1470.0(5) 1116.7(2) 1527.7(3) Z, Dcalcd (g/cm3) 2, 1.281 4, 1.349 4, 1.325 4, 1.312 4, 1.536 4, 1.455 µ(Mo Kα), (cm-1) 0.90 0.97 0.96 2.28 2.84 5.99 F000 296.00 528.00 1024.00 608.00 536.00 688.00 temp. (K) 173(2) 173(2) 173(2) 173(2) 173(2) 173(2) reflns collcd / unique 14 829/3490 12 253/5314 20 588/4455 26 978/6723 16 448/2541 16 003/3665  (Rint = 0.032) (Rint = 0.023) (Rint = 0.044) (Rint = 0.036) (Rint = 0.022) (Rint = 0.025) residuals (F2, all data) wR2 = 0.128 wR2 = 0.093 wR2 = 0.112 wR2 = 0.095 wR2 = 0.088 wR2 = 0.110 residuals (F, I > 2σ(I)) R1 = 0.046 R1 = 0.037 R1 = 0.043 R1 = 0.039 R1 = 0.033 R1 = 0.045   143  Table A.2. Selected bond lengths (Å) and angles (°) in Hdapp, Hsapp and Hzapp. Hdapp • MeOH Hsapp • MeOH 2Hzapp• MeOH•H2O Atoms Distance Atoms Distance Atoms Distance C(3)-O(2) 1.3544(16) C(3)-O(2) 1.360(2) C(3)-O(2) 1.358(2) C(4)-O(1) 1.2721(16) C(4)-O(1) 1.270(2) C(4)-O(1) 1.281(2) N(1)-C(2) 1.3788(17) N(1)-C(2) 1.383(2) N(1)-C(2) 1.378(2) C(2)-C(3) 1.3644(19) C(2)-C(3) 1.375(2) C(2)-C(3) 1.363(3) C(3)-C(4) 1.431(2) C(3)-C(4) 1.440(3) C(3)-C(4) 1.419(3) C(4)-C(5) 1.410(2) C(4)-C(5) 1.413(3) C(4)-C(5) 1.410(3) C(5)-C(6) 1.358(2) C(5)-C(6) 1.355(3) C(5)-C(6) 1.353(3) C(6)-N(1) 1.3518(18) C(6)-N(1) 1.356(3) C(6)-N(1) 1.355(2)  Atoms Angle Atoms Angle Atoms Angle C(8)-C(7)-N(1)-C(6) -71.69(18) C(8)-C(7)-N(1)-C(6) -74.2(2) C(8)-C(7)-N(1)-C(6) -113.4(2) C(12)-C(7)-N(1)-C(6) 110.09(16) C(12)-C(7)-N(1)-C(6) 103.6(2) C(12)-C(7)-N(1)-C(6) 66.3(2) C(8)-C(7)-N(1)-C(2) 104.81(15) C(8)-C(7)-N(1)-C(2) 105.2(2) C(8)-C(7)-N(1)-C(2) 68.2(2) C(12)-C(7)-N(1)-C(2) -73.41(17) C(12)-C(7)-N(1)-C(2) -76.9(2) C(12)-C(7)-N(1)-C(2) -112.1(2)    144  Table A.3. Selected bond lengths (Å) and angles (°) in Hbt6p, Hbt2p and Hnbp. Hbt6p • MeOH Hbt2p Hnbp • CHCl3 Atoms Distance Atoms Distance Atoms Distance C(3)-O(2) 1.3794(17) C(3)-O(2) 1.3536(17) C(3)-O(2) 1.350(2) C(4)-O(1) 1.2993(18) C(4)-O(1) 1.2585(17) C(4)-O(1) 1.266(2) N(1)-C(2) 1.4065(18) N(1)-C(2) 1.3944(17) N(1)-C(2) 1.371(3) C(2)-C(3) 1.398(2) C(2)-C(3) 1.3706(19) C(2)-C(3) 1.362(3) C(3)-C(4) 1.460(2) C(3)-C(4) 1.4462(19) C(3)-C(4) 1.425(3) C(4)-C(5) 1.443(2) C(4)-C(5) 1.429(2) C(4)-C(5) 1.411(3) C(5)-C(6) 1.386(2) C(5)-C(6) 1.352(2) C(5)-C(6) 1.353(3) C(6)-N(1) 1.382(2) C(6)-N(1) 1.3686(18) C(6)-N(1) 1.345(3)  Atoms Angle Atoms Angle C(8)-C(7)-N(1)-C(6) -108.98(16) N(2)-C(7)-N(1)-C(6) 124.06(15)- C(13)-C(7)-N(1)-C(6) 71.98(18) S(1)-C(7)-N(1)-C(6) 55.93(16) C(8)-C(7)-N(1)-C(2) 73.44(18) N(2)-C(7)-N(1)-C(2) -55.57(19) C(13)-C(7)-N(1)-C(2) -105.59(16) S(1)-C(7)-N(1)-C(2) 124.44(12)    145     Figure A.7. Representative semi-preparative HPLC trace for purification of β-Gbt2p. Solvent flow rate: 10 mL/min; 45-minute gradient from 100 % H2O (containing 0.1 % trifluoroacetic acid, TFA) to 100 % acetonitrile.  β-Gbt2p (collected) Hbt2p 146  Crystallographic Data for Bis(pyridinato)copper(II) Complexes  All crystal data were collected and solved by Dr. Brian O. Patrick, Structural Chemistry Facility, UBC. For each complex a brief outline is provided of the crystallisation procedures and nature of the crystal. These are followed by the crystal data and, separately, selected bond lengths and angles in tabulated form (Tables A.4, A.5, respectively).  Cu(dapp)2  Figure A.8. Ellipsoid plot (50 % probability; for clarity, H-atoms not shown) of Cu(dapp)2.  Green blade crystals of Cu(dapp)2 were obtained from CHCl3 by liquid/liquid diffusion with Et2O at RT and co-crystallises with one molecule of CHCl3. The Cu(1)-O(1) and O(1)-C(4) bond lengths (1.9191(16) and 1.295(3) Å, respectively) are reflective of the relatively keto- character of this O-donor group, while the Cu(1)-O(2) and O(2)-C(3) bond lengths (1.9085(14) 147  and 1.321(3) Å, respectively) are reflective of the anionic nature of this O-donor. The limited bite angle of the pyridinone leads to a slightly narrowed O(1)-Cu(1)-O(2) angle of 86.76(6) ° instead of the 90 °-angle expected for an ideal square planar complex. Because of this, the O(1)- Cu(1)-O(2*) angle is slightly widened from 90 ° to 93.24(6) °. Within the ligand, the phenyl ring is oriented at an average angle of 64.5 ° relative to the pyridinone ring. Selected bond lengths and angles are given in Table A.5.  Cu(zapp)2  Figure A.9. Ellipsoid plot (50 % probability; for clarity, H-atoms not shown) of Cu(zapp)2.  Green prism crystals of Cu(zapp)2 were obtained from a mixture of approximately 3:1 MeOH:DMSO by slow evaporation at RT and co-crystallises with two molecules of DMSO. The Cu(1)-O(1) and O(1)-C(4) bond lengths (1.9311(12) and 1.298(2) Å respectively) are reflective of the relatively keto-character of this O-donor group, while the Cu(1)-O(2) and O(2)-C(3) bond lengths (1.9113(12) and 1.328(2) Å, respectively) are reflective of the anionic nature of this O- donor. The limited bite angle of the pyridinone leads to a slightly narrowed O(1)-Cu(1)-O(2) 148  angle of 86.03(5) ° instead of the 90 °-angle expected for an ideal square planar complex. Because of this, the O(1)-Cu(1)-O(2*) angle is slightly widened from 90 ° to 93.97(5) °. Within the ligand, the phenyl ring is oriented at an average angle of 72.8 ° relative to the pyridinone ring. Selected bond lengths and angles are given in Table A.5.  Cu(ppp)2  Figure A.10. Ellipsoid plot (50 % probability; for clarity, H-atoms not shown) of Cu(ppp)2.  Green prism crystals of Cu(ppp)2 were obtained using the same crystallisation conditions as those used for Cu(dapp)2, and, like Cu(dapp)2, co-crystallises with one molecule of CHCl3. In a similar manner as that in Cu(dapp)2, the O(1) donor group exhibits more keto character while the O(2) donor group exhibits more anionic character. Due to the limited bite angle of the pyridinone the O(1)-Cu(1)-O(2) angle is slightly narrowed to 86.36(8) ° instead of the 90 °-angle expected for an ideal square planar complex; this causes the O(1)-Cu(1)-O(2*) angle to exceed 90 ° with a value of 93.62(6) °. Within the ligand, the phenyl ring is oriented at an average angle of 63.1 ° relative to the pyridinone ring. Bond lengths between the Cu2+ ion and the donor atoms 149  of the pyridinone, as well as within the pyridinone functional groups, may be compared in Table A.5. CCDC entry 697417.  Cu(nbp)2  Figure A.11. Ellipsoid plot (50 % probability; for clarity, H-atoms not shown) of Cu(nbp)2.  Green prism crystals of Cu(nbp)2 were obtained from CHCl3 by liquid/liquid diffusion with Et2O at RT. The coordination complex co-crystallises with one molecule of CHCl3. In a similar manner to that of the other three copper complexes, this complex contains donor groups on the pyridinone ligands displaying both keto character (O(1)) and anionic character (O(2)). As before, the O(1)-Cu(1)-O(2) angle is slightly narrowed to 86.82(12) °, making the O(1)-Cu(1)- O(2*) angle slightly widened from 90 ° to 93.18(12) °. Due to the extra methylene linker between the two ring systems, the N-substituent ring of the pyridinone ligand shows significant disorder in the crystal lattice; two orientations were modelled with equivalent populations. The solved structure for Cu(nbp)2 and the structure shown in Figure A.11 is representative of one of these orientations. CCDC entry 697416. 150  Table A.4. Crystallographic data for Cu(dapp)2, Cu(zapp)2, Cu(ppp)2 and Cu(nbp)2. crystal data Cu(dapp)2 • 2CHCl3 Cu(zapp)2 • 2DMSO Cu(ppp)2 • 2CHCl3 Cu(nbp)2 • 2CHCl3 formula C28H30N4O4Cu .2CHCl3 C24H22N4O4Cu .2DMSO C24H20N2O4Cu .2CHCl3 C26H24N2O4Cu .2CHCl3 fw 788.84 650.25 702.70 730.75 crystal system, space group monoclinic, P 21/c (#14) monoclinic, P 21/c (#14) orthorhombic, P bca (#61) monoclinic, P 21/n (#14) a (Å) 12.9970(12) 7.7458(9) 11.5466(11) 6.3813(3) b (Å) 15.3280(17) 16.433(2) 11.6975(11) 21.0370(13) c (Å) 9.0063(8) 12.0999(14) 20.9936(18) 11.7337(7) α (deg) 90.0 90.0 90.0 90.0 β (deg) 110.181(4) 103.182(4) 90.0 96.787(2) γ (deg) 90.0 90.0 90.0 90.0 V [Å3] 1684.1(3) 1499.6(3) 2835.5(5) 1562.53(15) Z, Dcalcd (g/cm3) 2, 1.556 2, 1.440 4, 1.646 2, 1.553 µ(Mo Kα), (cm-1) 11.66 9.15 13.73 12.49 F000 806.00 678.00 1420.00 742.00 temp. (K) 173(2) 173(2) 173(2) 173(2) reflns collcd / unique 17 601/4105 17 254/3618 35 879/2789  13 415/3545  (Rint = 0.035) (Rint = 0.045) (Rint = 0.046) (Rint = 0.033) residuals (F2, all data) wR2 = 0.100 wR2 =0.098 wR2 = 0.071 wR2 = 0.163 residuals (F, I > 2σ(I)) R1 = 0.038 R1 = 0.035 R1 = 0.028 R1 = 0.053   151  Table A.5. Selected bond lengths (Å) and angles (°) in Cu(dapp)2, Cu(zapp)2, Cu(ppp)2 and Cu(nbp)2. Cu(dapp)2 • 2CHCl3 Cu(zapp)2 • 2DMSO Cu(ppp)2 • 2CHCl3 Cu(nbp)2 • 2CHCl3 Atoms Distance Atoms Distance Atoms Distance Atoms Distance C(3)-O(2) 1.321(3) C(3)-O(2) 1.328(2) C(3)-O(2) 1.322(3) C(3)-O(2) 1.345(5) C(4)-O(1) 1.295(3) C(4)-O(1) 1.298(2) C(4)-O(1) 1.308(2) C(4)-O(1) 1.284(5) N(1)-C(2) 1.369(3) N(1)-C(2) 1.384(2) N(1)-C(2) 1.379(3) N(1)-C(2) 1.369(6) C(2)-C(3) 1.376(3) C(2)-C(3) 1.380(2) C(2)-C(3) 1.381(3) C(2)-C(3) 1.395(6) C(3)-C(4) 1.431(3) C(3)-C(4) 1.427(2) C(3)-C(4) 1.428(3) C(3)-C(4) 1.404(6) C(4)-C(5) 1.389(3) C(4)-C(5) 1.403(2) C(4)-C(5) 1.395(3) C(4)-C(5) 1.431(6) C(5)-C(6) 1.359(3) C(5)-C(6) 1.364(3) C(5)-C(6) 1.362(3) C(5)-C(6) 1.353(7) C(6)-N(1) 1.352(3) C(6)-N(1) 1.356(2) C(6)-N(1) 1.352(3) C(6)-N(1) 1.337(7) Cu(1)-O(2) 1.9085(14) Cu(1)-O(2) 1.9113(12) Cu(1)-O(2) 1.9096(14) Cu(1)-O(2) 1.904(3) Cu(1)-O(1) 1.9191(16) Cu(1)-O(1) 1.9311(12) Cu(1)-O(1) 1.9275(15) Cu(1)-O(1) 1.927(3)   Atoms Angle Atoms Angle Atoms Angle Atoms Angle O(1)-Cu(1)-O(2) 87.76(6) O(1)-Cu(1)-O(2) 86.03(5) O(1)-Cu(1)-O(2) 86.36(8) O(1)-Cu(1)- O(2) 86.82(12) O(1)-Cu(1)- O(2*) 93.24(6) O(1)-Cu(1)- O(2*) 93.97(5) O(1)-Cu(1)- O(2*) 93.62(6) O(1)-Cu(1)- O(2*) 93.18(12) C(8)-C(7)- N(1)-C(6) -65.3(3) C(8)-C(7)- N(1)-C(6) 109.8(2) C(8)-C(7)- N(1)-C(6) 62.7(3)  C(12)-C(7)- N(1)-C(6) 113.3(3) C(12)-C(7)- N(1)-C(6) -71.9(2) C(12)-C(7)- N(1)-C(6) -116.0(2)  C(8)-C(7)- N(1)-C(2) 117.6(3) C(8)-C(7)- N(1)-C(2) -73.6(2) C(8)-C(7)- N(1)-C(2) -117.8(2)  C(12)-C(7)- N(1)-C(2) -63.7(3) C(12)-C(7)- N(1)-C(2) 104.7(2) C(12)-C(7)- N(1)-C(2) 63.5(3)  152     Figure A.12. Representative 13C NMR spectrum (DMSO-d6, 75.48 MHz, RT) for pyridinone pro-ligand Hsapp. bm-nib-1_002000fid 176 168 160 152 144 136 128 120 112 104 96 88 80 72 64 56 48 40 32 24 16 8 0 Chemical Shift (ppm) 0 0.05 0.10 0.15 0.20 N o r m a l i z e d  I n t e n s i t y 1 6 9 . 2 9 1 5 0 . 0 7 1 4 4 . 8 5 1 3 8 . 4 2 1 2 9 . 9 7 1 2 9 . 3 5 1 2 7 . 2 9 1 1 1 . 3 9 1 1 0 . 4 5 4 0 . 3 5 4 0 . 0 7 3 9 . 7 9 3 9 . 5 1 3 9 . 2 3 3 8 . 9 6 3 8 . 6 8 2 9 . 5 6 1 3 . 2 3 N O OH HN 12 3 4 5 78 9 10 11 6 1 2 3 5 6 4 8,9 11 7,10 1 3 . 2 4 2 9 . 5 7DMSO 1 1 0 . 4 7 1 1 1 . 4 1 1 2 7 . 3 1 1 2 9 . 3 7 1 2 9 . 9 9 1 5 0 . 0 9 1 3 8 . 4 4 1 4 4 . 8 7 1 6 9 . 3 1 153     Figure A.13. Representative MALDI-TOF mass spectrum confirming the presence of Aβ40 (MW = 4329.9 g/mol) in solution after turbidity assay at pH 7.4; Zn2+ addition was followed by Hdpp addition. 4331.4 3810.3

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0061398/manifest

Comment

Related Items