Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Fluorinated nitroimidazoles and their ruthenium complexes : potential hypoxia-imaging agents Baird, Ian Robert 1999

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

Item Metadata

Download

Media
831-ubc_1999-463141.pdf [ 16.82MB ]
Metadata
JSON: 831-1.0228854.json
JSON-LD: 831-1.0228854-ld.json
RDF/XML (Pretty): 831-1.0228854-rdf.xml
RDF/JSON: 831-1.0228854-rdf.json
Turtle: 831-1.0228854-turtle.txt
N-Triples: 831-1.0228854-rdf-ntriples.txt
Original Record: 831-1.0228854-source.json
Full Text
831-1.0228854-fulltext.txt
Citation
831-1.0228854.ris

Full Text

FLUORINATED N I T R O P A I I D A Z O L E S A N D THEIR R U T H E N I U M C O M P L E X E S : POTENTIAL HYPOXIA-IMAGING AGENTS By IAN ROBERT BAIRD B.Sc , University of British Columbia, 1994 A THESIS SUBMITTED IN PARTIAL F U L F I L L M E N T OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH C O L U M B I A March 1999 © Ian Robert Baird, 1999 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of The University of British Columbia Vancouver, Canada DE-6 (2/88) Abstract With the goal of synthesizing nitroimidazole-based hypoxia imaging agents for the detection of cancerous tumours, the chemistry of halogen incorporation, especially fluorine, into the N I side-chain of nitroimidazoles was investigated. The coordination of these nitroimidazoles and other methyl (Me)-substituted imidazoles to Ru(II) and Ru(III) centres was also investigated. The general objective was to combine the hypoxia-selective nature of the nitroimidazole moiety and the DNA-binding properties of Ru to deliver the active species selectively to its target. The nitroimidazole compounds and Ru complexes were characterized in general by a combination of NMR, IR and UV-Visible spectroscopies, as well as mass spectrometry, cyclic voltammetry, conductivity and elemental analysis; six nitroimidazoles and eleven Ru complexes were also characterized by X-ray crystallography. The halogenated nitroimidazoles were synthesized using a standard amide coupling reaction; this was used for the synthesis of EF5 [2-(2-nitro-l-H-imidazol-l-yl)-N-(2,2,3,3,3-pentafluoropropyl) acetamide], a known, sensitive probe for quantifying the amount of hypoxia within cells. Treatment of a desired acid with N-methylmorpholine and zso-butylchloroformate, followed by addition of the appropriate amine (either H2NCH2CX3 or H 2 NCH 2 CX 2 CX3, where X= H , F, CI and/or Br) led to formation of the halogenated nitroimidazole; 2-nitro- (2N0 2), 2-methyl-5-nitro- (2Me5N0 2) and 2-methyl-4-nitro- (2Me4N0 2) imidazole compounds were isolated in yields of 15 to 83 %. The synthesis of EF5 (typical yield 45 %) was improved by coupling iodoacetic acid with H 2 N C H 2 C F 2 C F 3 to give ICH 2 C(0)NHCH 2 CF 2 CF 3 (TF5) which was subsequently reacted with 2N0 2 Im and C s 2 C 0 3 to yield the final product (78 %); this 2-step method circumvented the previously required 5-step synthesis. Other analogous compounds were less stable than IF5, especially those species containing a C-Br moiety, and they often decomposed in the presence of heat and/or light. Alteration of the associated side-chain on EF5 by 'reversal' of its amide linkage was performed; however, the biological activity of the derivative was significantly less than that of EF5. A nitroimidazole (N0 2lm) containing side-chain with a highly reactive terminal group, with the potential to exchange fluoride, was also sought. Attempts to isolate a ii tosylated derivative of SR2508 [2-(2-mtro-l-H-irnidazol-l-yl)-N-(2-hydroxyethyl) acetamide] proved to be futile. The reaction with tosyl chloride led to the formation of a monochloro analogue of SR2508, presumably via the reactive tosylate intermediate, while reaction with triflic anhydride led to an intramolecular cyclization of the side-chain to give a compound whose X-ray structure was determined. The focus was then shifted to F-for-Br exchange; however, attempts to perform such a task revealed that the C - N 0 2 bond of the Im ring was more labile than the C-Br bond, and this resulted in the formation of a 2-fluoroimidazole species. Further efforts to incorporate an F-atom into a nitroimidazole side-chain were unsuccessful. [Ru(II)(L)6]2 + complexes were synthesized from [Ru(DMF) 6][CF3S03]3; D M F = dimethylformamide, L = imidazole (Im), N-methylimidazole (NMelm) and 5-methyl-imidazole (5MeIm). The 2-methylimidazole complex fram-[Ru(CO)(DMF)(2MeIm)4] [CF3S03]2 was synthesized via a reaction involving abstraction of CO from DMF; this complex loses CO reversibly at ambient temperature to form [Ru(DMF)(2MeIm)4] [CF 3 S0 3 ] 2 , and the DMF can be removed to generate [Ru(CF3S03)x(2MeIm)4][CF3S03]y (x = 2, y = 0, or x = 1 = y). The X-ray structures of [Ru(Im) 6][CF 3S0 3] 2, [Ru(NMeIm) 6][CF 3S0 3] 2 and [Ru(5MeIm) 6][CF 3S0 3] 2 were obtained. Analogous [Ru(II)(L)x]2 + nitroimidazole complexes (L = 2N0 2Im, x = 6; L = 4N0 2Im, x = 6; L = 2Me5N0 2Im, x = 5) were isolated from reaction with [Ru(DMF) 6][CF 3S03] 3 . The reaction with EF5 and SR2508 in EtOH yielded the bis-substituted complexes [Ru(DMF) 2(EF5) 2(EtOH) 2][CF 3S03] 3 and [Ru(DMF)4(SR2508)2][CF3SO3]3, respectively. Ru(III) complexes of composition RuCl 3 L 3 (L = 2N0 2Im, 4N0 2Im, 2Me5N0 2Im, metro) and RuCl 3L 2(EtOH) (L= EF5, SR2508) were synthesized directly from RuCl 3»3H 20. Their *H N M R spectra were typically broad and sometimes signals were not observed; however, their paramagnetic, d 5 low-spin composition was confirmed using the Evans method. EF5 SR2508 metro iii Some new Ru(II) and Ru(III) bis-P-diketonate (acac = acetylacetonate; hfac = 1,1,1,5,5,5-hexafluoroacetylacetonate) Im and N 0 2 l m complexes were synthesized. Reaction of two equiv. of an imidazole with c/s-[Ru(acac)2(MeCN)2][CF3S03] (synthesized from Ru(acac)3 and C F 3 S 0 3 H in MeCN) yielded Ru(III) complexes with composition [Ru(acac)2(L)2][CF3S03] (L = Im, NMelm, 2MeIm, 5MeIm, 2N0 2Im, metro, EF5 and SR2508), the first four being structurally characterized by X-ray crystallography. The analogous Ru(II) hfac complexes were isolated from a reaction of cz'5-Ru(hfac)2(MeCN)2 (71) (synthesized from either Ru(hfac)3 or Na[Ru(hfac)3] and C F 3 S 0 3 H in MeCN) with two equiv. of imidazole to yield Ru(hfac)2(L)2 (L = Im, NMelm, 2MeIm, 4(5)MeIm, 2N0 2Im, EF5 and SR2508). X-ray structures of 71 and Ru(hfac)3 were obtained. Reaction of 71 with neat NMelm gave [Ru(hfac)CNMeIm)4][hfac]. The mixed ligand complex c/5-Ru(hfac)(acac)(MeCN)2 (X-ray) was synthesized from the new species Ru(hfac)2(acac), which was isolated from a reaction of 71 with Hacac. In vitro assays for toxicity and monoclonal antibody (MoAb) binding with SCCVLT cells were used to evaluate the potential of selected nitroimidazoles as hypoxia-selective imaging agents. The toxicity, MoAb binding, cell accumulation and DNA-binding assays were used to test the utility of selected Ru complexes for transporting and/or localizing the coordinated nitroimidazoles within the cell. A preliminary radiosensitization study was also performed with two Ru complexes. In general, the nitroimidazoles (N0 2lms) and Ru complexes were non-toxic under both oxic and hypoxic conditions. The accumulation of the N0 2 lms within the cell required hypoxic conditions, while the amount of N 0 2 l m bound within the hypoxic cells correlated with its one-electron reduction potential, the N0 2 lms with the more positive reduction potentials giving the higher concentrations. The interaction of the fluorescently labeled MoAbs (ELK3-51 and ELK5-A8) with the N0 2 lms depended on the length, size and composition of the halogenated side-chains. The Ru-N0 2 lm complexes displayed relatively high cell accumulation and DNA-binding levels when compared to literature data for other Ru complexes. The most interesting result came from the MoAb assay in which the cells treated with [Ru(acac)2(EF5)2] [CF 3 S0 3 ] afforded a fluorescence signal four times greater than that seen for EF5 itself. iv Table of Contents Abstract ii Table of Contents v List of Figures xv List of Schemes xx List of Tables xxi List of Abbreviations xxvi Key to Chemical Compound Numbers and Abbreviations xxix Acknowledgements xxxv Chapter 1 - Introduction 1.1 Introduction 1 1.2 Hypoxia 2 1.2.1 What is it? 2 1.2.1.1 Resistance to Treatment: Radioresistance 3 1.2.2 Hypoxia: The Aggressor? 5 1.3 Role of Nitroimidazoles in Cancer Therapy 6 1.3.1 Hypoxic Radiosensitizers 6 1.3.2 Chemistry of Nitroimidazoles in Biological Systems 7 1.4 Imaging of Hypoxic Tumours 10 1.4.1 Radiolabelling of Hypoxic Cells 11 1.4.2 Nuclear Medicine Techniques 11 1.4.3 Fluorescent Probes 12 1.4.4 Magnetic Resonance Imaging (MRI) 14 1.4.5 Positron Emission Tomography (PET) 14 1.5 Role of Metals in Cancer Therapy 16 1.6 Ruthenium Complexes 18 1.6.1 Chemical Properties Relevant to Tumour Treatment 18 1.6.2 D N A Binding 20 v 1.6.3 Tumouricidal Effects of Ru Complexes 22 1.7 Thesis Overview 24 1.8 References 26 Chapter 2 - General Experimental 2.1 Materials 33 2.1.1 Chemicals 33 2.1.1.1 Imidazoles 33 2.1.1.2 Amines 33 2.1.1.3 Miscellaneous Reagents 33 2.1.2 Solvents 34 2.2 Analytical Techniques 34 2.2.1 Nuclear Magnetic Resonance Spectroscopy 34 2.2.2 Infrared Spectroscopy 37 2.2.3 UV-Visible Spectroscopy 37 2.2.4 Mass Spectrometry 37 2.2.5 Gas Chromatography 38 2.2.6 Cyclic Voltammetry 38 2.2.7 Conductivity 39 2.2.8 X-Ray Analysis 40 2.2.9 Elemental Analysis 40 2.3 Compound Purification Techniques 40 2.3.1 Column Chromatography 40 2.3.2 Preparative Thin Layer Chromatography 41 2.3.3 Chromatotron 41 2.4 General Methodologies 41 2.4.1 Amide Coupling Reaction 41 2.5 Ruthenium Precursors 42 2.5.1 RuCl 3 ' 3H 2 0 42 2.5.2 [Ru(DMF) 6][CF 3S0 3]3 42 vi 2.5.3 [Ru(DMF)6][CF3S03]2 43 '2.5.4 c/s-RuCl2(DMS0)(DMSO)3 43 2.5.5 trans-RuCl2(pMSO)4 44 2.5.6 cw-RuCl2(TMSO)4 44 2.5.7 /we/--RuCl3(DMSO)3 44 2.5.8 [RuCl2(COD)]x 45 2.5.9 [RuCl2(dppb)]2(u-dppb) 45 References 46 Chapter 3 - Synthesis and Characterization of New 2-, 4- and 5-Nitroimidazoles with Halogenated Side-Chains 3.1 Introduction 48 3.2 Experimental Section 49 3.2.1 Synthesis of Side-Chains 49 3.2.1.1 2-Iodo-/V-(2,2,3;3,3-pentafluoropropyl)acetamide [IF5] 49 3.2.1.2 2-Iodo-7V-(3-bromopropyl)acetamide [IBr] 49 3.2.1.3 2-Chloro-/V-(2,2,3,3,3-pentafluoropropyl)acetamide [C1F5] .... 51 3.2.1.4 Reduction of IF5 to ICH 2CH 2NHCH 2CF 2CF 3 (3) 52 3.2.1.5 3-Fluoropropylamine Hydrochloride (5) 52 3.2.2 2-Nitroimidazole Compounds 54 3.2.2.1 2-(2-Nitro-l-H-imidazol-l-yl)acetic acid (7) 54 3.2.2.2 2-(2-Nitro-1 -H-imidazol-1 -yl)-N-(2,2,3,3,3-pentafluoropropyl) acetamide [EF5] 56 3.2.2.3 2-(Imidazol-l-yl)-N-(2,2,3,3,3-pentafluoropropyl)acetamide [ImF5] 57 3.2.2.4 N-(2-nitro-l-H-imidazol-1-ethyl) pentafluoropropionamide [RevEF5] 58 3.2.2.5 2-(2-Nitro-l-H-imidazol-l-yl)-N-(3-bromo-2,2,3,3-tetrafluoropropyl) acetamide [EF4Br] 61 vii 3.2.2.6 2-(2-Nitro-l-H-imidazol-l-yl)-N-(3,3,3-trifluoropropyl)acetamide [EF3] 62 3.2.2.7 2-(2-Nitro-1 -H-imidazol-1 -yl)-N-(2,2,2-trifluoroethyl)acetamide [EF3(-1>] 63 3.2.2.8 2-(2-Nitro-1 -H-imidazol-1 -yl)-N-(3 -bromo-3,3 -difluoropropyl) acetamide [EF2Br] 63 3.2.2.9 2-(2-Nitro-l-H-imidazol-l-yl)-N-(3,3-difluoropropylene)acetamide [E=F2] 64 3.2.2.10 2-(2-Nitro-l-H-imidazol-l-yl)-N-(3-fluoropropyl)acetamide [EF1] 65 3.2.2.11 2-(2-Nitro-l-H-irmdazol-l-yl)-N-(2-fluoroethyl)acetamide [EFl(-l)] 67 3.2.2.12 2-(2-Nitro-1 -H-imidazol-1 -yl)-N-(3 -chloropropyl)acetamide [ECU] 68 3.2.2.13 2-(2-Nitro-1 -H-imidazol-1 -yl)-N-(2-chloroethyl)acetamide [ECU(-l)] 69 3.2.2.14 2-(2-Nitro-1 -H-imidazol-1 -yl)-N-(3 -bromopropyl)acetamide [EBrl] 70 3.2.2.15 2-(2-Nitro-1 -H-imidazol-1 -yl)-N-(2-bromoethyl)acetamide [EBrl(-l)] 71 3.2.2.16 2-(2-Nitro-l-H-imidazol-l-yl)-N-(propyl)acetamide [EPrA].. 72 3.2.2.17 2-(2-Nitro-l-H-imidazol-l-yl)-N-(/'50-amyl)acetamide [EIAA] 73 3.2.3 2-Methyl-5-Nitroimidazole Compounds 74 3.2.3.1 2-(2-Methyl-5-nitro-IB-imidazol-l-yl)acetic acid (13) 74 3.2.3.2 2-(2-Methyl-5-nitro-l/f-imidazol-l-yl)-N-(2,2,3,3,3) pentafluoropropyl) acetamide [MF5] 74 3.2.3.3 2-(2-Methyl-5-nitro-l#-imidazol-1 -yl)-N-(2,2,2-trifluoroethyl) acetamide [MF3(-1)] 76 viii 3.2.3.4 2-(2-Methyl-5-mtro-l^-irnidazol-l-yl)-N-(2-fluoroethyl)acetamide [MFl(-l)] 76 3.2.3.5 2-(2-Methyl-5 -nitro- 177-imidazol-1 -yl)-N-(2-chloroethyl)acetamide [MCIl(-l)] 77 3.2.3.6 2-(2-Methyl-5-nitro- l#-imidazol- l-yl)-N-(2-bromoethyl)acetamide [MBrl(-l)] 79 3.2.3.7 (2-Methyl-5-mtro-l/f-irmdazol-l-yl)-N-(2-chloroethane) (16).. 80 3.2.4 2-Methyl-4-Nitroimidazole Compounds 81 3.2.4.1 2-(2-Methyl-4-nitro-l-H-imidazol-l-yl)propionic acid (17) 81 3.2.4.2 3-(2-Methyl-4-nitro-l-H-imidazol-l-yl)-N-(2,2,3,3,3-pentafluoropropyl)propionamide [2M4NF5] 81 3.2.4.3 3-(2-Methyl-4-mtro-l-H-imidazol-l-yl)-N-(2,2,2-trifluoroethyl) propionamide [2M4NF3(-1)] 82 3.2.4.4 3-(2-Methyl-4-nitro-l-H-imidazol-l-yl)-N-(2-fluoroethyl) propionamide [2M4NF1(-1>] 83 3.2.4.5 3-(2-methyl-4-nitro-1 -H-imidazol-1 -yl)-N-(2-chloroethyl)propionamide [2M4NC11(-1)] 84 3.2.4.6 2-(2-methyl-4-nitro-l-H-imidazol-l-yl)-N-(2-bromoethyl)propionamide [2M4NBrl(-l)] 85 3.2.5 Reactions of SR2508 with Tf 2 0 85 3.2.5.1 2-(2-Nitro-l-H-imidazol-l-yl)-N-(ethylformate)acetamide (18) 85 3.2.5.2 CycF3 (19) 86 3.2.6 Reactions with RSU1111 87 3.2.6.1 Synthesis of (20), a pentafluorinated derivative ofRSUllll.... 87 3.2.6.2 Synthesis of (21), a dibromo derivative ofRSUllll 88 3.2.6.3 Synthesis of (22), a terminal nitrile derivative ofRSUllll 88 3.2.7 Reactions of Nitroimidazoles with Bu 4NF»H 20 90 3.2.7.1 Reaction of E B r l with Bu 4NF«H 20 90 3.2.7.2 Reaction of E B r l with Bu 4NF«H 20 and C H 3 C 0 2 H 91 3.2.7.3 Reaction of SR2508 with Bu 4NF«H 20 91 ix 3.2.7.4 Reaction of «N0 2 Im (n = 2, 4, 5) with Bu 4 NF«H 2 0 92 3.2.8 Cyclic Voltammetry of Nitroimidazoles 94 3.3 Results and Discussion 96 3.3.1 Synthesis of the Nitroimidazole Side-Chains 96 3.3.2 Nitroimidazoles '. 101 3.3.3 Electrochemistry of Nitroimidazoles 117 3.3.4 Incorporation of Fluorine into an Existing Side-chain 122 3.4 References 129 Chapter 4 - Synthesis and Characterization of Ru(II) and Ru(III) Imidazole Complexes 4.1 Introduction 134 4.2 Experimental Section 135 4.2.1 Complexes Synthesized from [Ru(DMF) 6][CF 3S03]3 135 4.2.1.1 [Ru(Im)6][CF3S03]2 (28) 135 4.2.1.2 [Ru(NMeIm) 6][CF 3S0 3] 2 (29) 135 4.2.1.3 [Ru(5MeIm)6][CF3S03]2 (30) 136 4.2.1.4 fra»5-[Ru(CO)(DMF)(2MeIm)4][CF3S03]2 (31), [Ru(DMF)(2MeIm)4][CF 3S0 3] 2 (32) and [Ru(CF 3 S0 3 ) x (2MeIm)4][CF3S03]y (x=2, y=0; or x=l=y) (33) 137 4.2.1.5 [Ru(2N02Im)6][CF3S03]2 (34) 138 4.2.1.6 [Ru(5N02Im)6][CF3S03]2 (35) 138 4.2.1.7 [Ru(2Me5N0 2Im) 5][CF3S0 3] 2 (36) 139 4.2.1.8 [Ru(DMF) 4(SR2508) 2][CF 3SO 3] 3 (37) 140 4.2.1.9 [Ru(DMF) 2(EF5) 2(EtOH) 2][CF 3S0 3] 3 (38) 140 4.2.2 Complexes Synthesized from RuCl 3*3H 20 141 4.2.2.1 /wer-RuCl3(2N02Im)3 (39) 141 4.2.2.2 fac- and /wer-RuCl3(5N02Im)3 (40) 142 4.2.2.3 "RuCl 3(2Me5N0 2Im) 3*3C0 2" (41) 143 x 4.2.2.4 RuCl3(metro)3 (42) 143 4.2.2.5 RuCl3(SR2508)2(EtOH) (43) 144 4.2.2.6 RuCl 3(EF5) 2(EtOH) (44) 145 4.2.2.7 RuCl2(metro)4 (45) 146 4.2.3 Complexes Synthesized from RuCl 2 (DMSO) 4 146 4.2.3.1 c/.v-RuCl2(DMSO)2(en) (46) 146 4.2.3.2 /ram-RuCl2(DMSO)2(en) (47) 147 4.2.3.3 RuCl2(DMSO)2(EF5)(acetone) (48) 148 4.2.4 Miscellaneous Complexes 148 4.2.4.1 [RuCl(dppb)(EF5)]2(u-Cl)2 (49) 148 4.2.4.2 ds-RuCl 2(MeCN) 4 (51) and roer-RuCl3(MeCN)3 (52) 149 4.2.4.3 Other Attempted Reactions 150 4.3 Results and Discussion 153 4.3.1 Complexes Synthesized from [Ru(DMF) 6][CF 3S0 3] 3 153 4.3.1.1 Hexakis(imidazole) Ru(II) Complexes 153 4.3.1.2 Reaction of [Ru(DMF) 6][CF 3S0 3] 3 with 2MeIm 158 4.3.1.3 Hexakis(nitroimidazole) Ru(II) Complexes, and some Ru(III) mixed DMF-SR2508 and -EF5 Complexes 160 4.3.2 Complexes Synthesized from RuCl 3»3H 20 165 4.3.3 Complexes Synthesized from c/V*ram-RuCl 2(DMSO) 4 173 4.3.4 Miscellaneous Ru Complexes 175 4.4 References 182 Chapter 5 - Synthesis and Characterization of Ru(II) and Ru(III) (3-Diketonate Imidazole Complexes 5.1 Introduction 187 5.2 Experimental 188 5.2.1 Ruthenium(III) Acetylacetonato Complexes 188 5.2.1.1 Ru(acac)3 (53) 188 xi 5.2.1.2 c/5-[Ru(acac)2(MeCN)2][CF3S03] (54) 189 5.2.1.3 cw-[Ru(acac)2(Im)2][CF3S03] (55) 189 5.2.1.4 c/5-[Ru(acac)2(NMeIm)2][CF3S03] (58) 190 5.2.1.5 c/5-[Ru(acac)2(2MeIm)2][CF3S03] (60) 191 5.2.1.6 cw-[Ru(acac)2(5MeIm)2][CF3S03] (62) 192 5.2.1.7 [Ru(acac)2(2N02Im)2][CF3S03] (64) 193 5.2.1.8 [Ru(acac)2(SR2508)2][CF3SO3] (65) 193 5.2.1.9 [Ru(acac)2(EF5)2][CF3S03] (66) 194 5.2.1.10 [Ru(acac)2(metro)2][CF3S03] (67) 195 5.2.1.11 Reaction of c/5-[Ru(acac)2(MeCN)2][CF3S03] (54) with EtOH 195 5.2.1.12 Reaction of c/5-[Ru(acac)2(MeCN)2][CF3S03] (54) with H 20 196 5.2.2 Ruthenium(II) 1,1,1,5,5,5-Hexafluoroacetylacetonate Complexes.. 197 5.2.2.1 [Na][Ru(hfac)3] (68) and [Ru(hfac)(EtOH)4][hfac] (69) 197 5.2.2.2 Ru(hfac)3 (70) 198 5.2.2.3 C75-Ru(hfac)2(MeCN)2 (71) 198 5.2.2.4 c/5-Ru(hfac)2(Im)2 (72) 199 5.2.2.5 c/5-Ru(hfac)2(NMeIm)2 (74) 200 5.2.2.6 [Ru(hfac)(NMeIm)4][PF6] (76) 201 5.2.2.7 cz5-Ru(hfac)2(2MeIm)2 (77) 202 5.2.2.8 c«-Ru(hfac)2(4MeIm)(5MeIm) (80) 203 5.2.2.9 Ru(hfac)2(2N02Im)2 (82) 203 5.2.2.10 Ru(hfac)2(EF5)2 (83) 204 5.2.2.11 Attempted Synthesis of Ru(hfac)2(SR2508)2 (84) 205 5.2.3 Ruthenium(II) and (III) acac/hfac Complexes 205 5.2.3.1 Ru(acac)2(hfac) (85) 205 5.2.3.2 [Na][Ru(hfac)2(acac)] (86) 206 5.2.3.3 Ru(hfac)2(acac) (87) 206 5.2.3.4 m-Ru(acac)(hfac)(MeCN)2 (88) 207 xii 5.2.3.5 c/.v-Ru(acac)(hfac)(Im)2 (89) 208 5.2.3.6 c/s-Ru(acac)(hfac)(NMeIm)2 (90) 208 5.2.4 Cyclic Voltammetry of Ru(II) and Ru(III) p-diketonato Complexes 209 5.3 Results and Discussion 211 5.3.1 Ruthenium tris(P-diketonato) complexes 211 5.3.2 Bis-MeCN Complexes 219 5.3.3 Bis-imidazole Complexes 227 5.3.3.1 Bis(acetylacetonato)bis(imidazole)ruthenium(III) complexes .. 227 5.3.3.2 Bis( 1,1,1,5,5,5 -hexafluoroacetylacetonato)bis(imidazole) njthenium(II) complexes 238 5.3.3.3 Cyclic Voltammetric Results for the Bis-imidazole Complexes 245 5.4 References 248 Chapter 6 - In Vitro Evaluation of Selected Nitroimidazoles and Ruthenium Nitroimidazole Complexes 6.1 Introduction 252 6.2 Experimental 253 6.2.1 Materials and Methods 253 6.2.1.1 Media 253 6.2.1.2 General Solutions 253 6.2.1.3 Cell Handling 255 6.2.1.4 Preparation of Compound Solutions 255 6.2.1.5 Cell Incubation Procedures 256 6.2.2 Instrumentation 257 6.2.2.1 Atomic Absorption Spectroscopy (AAS) 257 6.2.2.2 UV-Vis Spectroscopy 258 6.2.2.3 Flow Cytometry (FCM) 258 6.2.2.4 Image Cytometry (ICM) 258 xiii 6.2.3 Biological Assays 260 6.2.3.1 Toxicity Assays 260 6.2.3.2 Cell Accumulation (Uptake) Assay 260 6.2.3.3 DNA-binding Assay (Isolation of D N A from SCCVII Cells)... 261 6.2.3.4 Radiosensitizing Assays 261 6.2.3.5 Monoclonal Antibody Assay 263 6.3 Results and Discussion 265 6.3.1 Compound Solubility 265 6.3.2 Toxicity 266 6.3.2.1 Nitroimidazoles 266 6.3.2.2 Ruthenium Complexes 268 6.3.3 Cellular Accumulation of Ru Complexes 269 6.3.4 DNA-binding by Ru Complexes 272 6.3.5 Radio sensitization 274 6.3.6 Nitroimidazole Adduct Recognition using Monoclonal Antibodies. 276 6.3.6.1 Recognition of N 0 2 l m Adducts by ELK3-51 278 6.3.6.2 Recognition of N 0 2 l m Adducts by ELK5-A8 284 6.4 Refereneces 287 Chapter 7 - Conclusions and Recommendations for Future W o r k 7.1 General Remarks 291 7.2 Halogenated Nitroimidazoles (Chapter 3) 291 7.3 Ruthenium Imidazole Complexes (Chapter 4) 295 7.4 Ruthenium P-Diketonate Imidazole Complexes (Chapter 5) 297 7.5 In Vitro Evaluation of Nitroimidazoles and their Ru Complexes 298 7.6 References 301 xiv Appendix I - X-ray data Appendix 1-1 X-ray data for EF5 302 Appendix 1-2 X-ray data for EF2Br»0 5 H 2 0 304 Appendix 1-3 X-ray data for MF5 306 Appendix 1-4 X-ray data for 2M4NF5 308 Appendix 1-5 X-ray data for 2M4NF1(-1) 311 Appendix 1-6 X-ray data for cycF3 (19) 313 Appendix 1-7 X-ray data for [Ru(Im)6][CF3S03]2 316 Appendix 1-8 X-ray data for [Ru(NMeIm)6][CF3S03]2 318 Appendix 1-9 X-ray data for [Ru(5MeIm) 6][CF 3S0 3] 2 320 Appendix 1-10 X-ray data for 7wer-RuCl3(MeCN)3»CHCl3 322 Appendix 1-11 X-ray data for Ru(hfac)3 324 Appendix 1-12 X-ray data for Ru(hfac)2(MeCN)2 327 Appendix 1-13 X-ray data for Ru(acac)(hfac)(MeCN)2 330 Appendix 1-14 X-ray data for [Ru(acac) 2(Im) 2][CF 3S0 3]»benzene 333 Appendix 1-15 X-ray data for [Ru(acac)2(NMeIm)2][CF3S03] 335 Appendix 1-16 X-ray data for [Ru(acac)2(2MeIm)2][CF3SO3]«0.5 hexane... 338 Appendix 1-17 X-ray data for [Ru(acac)2(5MeIm)2][CF3S03] 340 Appendix II - Miscellaneous Biological Data 343 List of Figures Figure i: Imidazole position numbering xxviii Figure 1-1: Oxygen gradient observed in a typical tumour (adapted from ref. 6) 2 Figure 1-2: Typical results for X-ray irradiation of Chinese Hamster Ovary cells under aerobic (0 2) or hypoxic (N 2) conditions (adapted from ref. 6) 3 Figure 1-3: Structures of metronidazole, misonidazole and etanidazole 7 Figure 1-4: Nitroimidazole reduction scheme (adapted from ref. 30) 8 Figure 1-5: Ranges of reduction potentials for nitroimidazoles vs. N H E at pH 7 8 xv Figure 1-6: Nucleophilic addition of water to l-methyl-2-nitrosoimidazole 9 Figure 1-7: Proposed reactions of 2-hydroxyaminoimidazole within the cell 10 Figure 1-8: Iodoazomycin arabinoside (IAZA) 12 Figure 1-9: Indolizine-linked 2-nitroimidazole, fluorescent hypoxia probe 12 Figure 1-10: Structures of the hypoxia-selective fluorinated 2-nitroimidazoles CCI-103F and EF5 13 Figure 1-11: y-ray production from positron emitting isotope [1 8F] 15 Figure 1-12: Chemical structures of antitumour agents cisplatin, carboplatin, JM216 and AMD473 17 Figure 1-13: Cisplatin/DNA interaction to form a G G intrastrand cross-link 21 Figure 2-1: Diagramatic representation of a capillary (containing a solution of the complex) within a N M R tube (containing solvent only) used for the Evans method 36 Figure 2-2: Cyclic voltammetry cell (A) containing a Pt reference electrode 39 Figure 2-3: Cyclic voltammetry cell (B) containing an Ag reference electrode 39 Figure 3-1: TLC analysis of products obtained from 2-(2-Nitro-l-H-imidazol- l-yl)acetic acid (7) synthesis 55 Figure 3-2: T L C analysis of products obtained from EF1 synthesis 66 Figure 3-3: T L C analysis of MXl(-l) compounds (X =F, CI, Br) 79 Figure 3-4: T i N M R spectrum of EF1 in d6-acetone 111 Figure 3-5: * H N M R spectrum of EFl(-l) in d6-acetone I l l Figure 3-6: Typical IR spectrum of a 2-nitroimidazole, EF1 113 Figure 3-7: ORTEP view of EF5; 33 % probability thermal ellipsoids are shown 114 Figure 3-8: ORTEP view of MF5; 33 % probability thermal ellipsoids are shown 115 Figure 3-9: ORTEP view of 2M4NF5; 33 % probability thermal ellipsoids are shown.. 116 Figure 3-10: ORTEP view of 2M4NF1(-1); 33 % probability thermal ellipsoids are shown 117 Figure 3-11: A lowering of pH levels in hypoxic cells due to increased [C0 2 ] 118 Figure 3-12: Quasi-reversible C V plot for EF1 referenced to FeCp 2 120 Figure 3-13: C V plot for MCIl(-l) referenced to FeCp 2 121 xvi Figure 3-14: ORTEP view of 19; 33 %'probability thermal ellipsoids are shown 123 Figure 3-15: ORTEP view of EF2Br; 33 % probability thermal ellipsoids are shown....128 Figure 4-1: TLC analysis of products obtained from synthesis of 52 150 Figure 4-2: ORTEP view of Ru(Im) 6 2 + (28); 50% probability thermal ellipsoids are shown 154 Figure 4-3: ORTEP view of Ru(NMeIm) 6 2 + (29); 33% probability thermal ellipsoids are shown 155 Figure 4-4: ORTEP view of Ru(5MeIm) 6 2 + (30); 50% probability thermal ellipsoids are shown 156 Figure 4-5: Proposed stepwise D M F displacement by 2N02lm for the synthesis of [Ru(2N0 2Im) 6][CF,S03]2 (34) 161 Figure 4-6: In situ T i N M R spectrum in CD 3 OD during the synthesis of 34 162 Figure 4-7: T i N M R spectrum for 39 in d6-dmso 167 Figure 4-8: Geometrical isomerization of complex 40 in d6-dmso 168 Figure 4-9: Possible configurations of a dimer formed from 2Me5N0 2Im 168 Figure 4-10: IR spectrum for the isolated, orange complex 41 169 Figure 4-11: *H N M R spectra for complex 41 in CD 3 OD (initial and after 6 d) 170 Figure 4-12: Successive aquation of 43 (L = SR2508) and 44 (L = EF5) 172 Figure 4-13: TI N M R (200 MHz) spectrum (Evans Method) of 43 (0.0032 g / mL D 2 0 , in a 0.1 mm capillary tube) referenced to the residual proton peak, HOD. 172 Figure 4-14: Isomerization of c/'s-RuCl2 (DMSO)2(en) from 46a to 46b 174 Figure 4-15: 3 1P{ 1H} N M R spectrum for complex 49, with proposed structure 176 Figure 4-16: 3 1P{TI} N M R spectrum for the proposed [RuCl2(dppb)(EF5)]2(u>dppb) (50) 177 Figure 4-17: [RuCl2(diop)]2(u.-diop) and the proposed structure for 50 178 Figure 4-18: ORTEP view of /wer-RuCl3(MeCN)3 (52); 50% probability thermal ellipsoids are shown 180 Figure 5-1: TLC analysis for the synthesis of 88 from 85 207 Figure 5-2: Resonance structures for acetylacetonate (acac") 211 xvii Figure 5-3: ORTEP view of Ru(hfac)3 (70); 50 % probability thermal ellipsoids are shown 213 Figure 5-4: TT N M R spectrum of Ru(hfac)2(acac) (87) in CDC1 3 at r.t 217 Figure 5-5: Relationship between E i / 2 and I om in 0.1 M TBAP at 25°C. — • —, this work (CV data in MeCN); — • —, Patterson and Holm (polarographic data in DMF) . 1 5 218 Figure 5-6: Proposed reaction mechanism for acid catalyzed displacement of acac (adapted from ref. 19) 220 Figure 5-7: ORTEP view of c/5-Ru(hfac)2(MeCN)2 (71); 50 % probability thermal ellipsoids are shown 222 Figure 5-8: ORTEP view of cw-Ru(acac)(hfac)(MeCN)2 (88); 50 % probability thermal ellipsoids 227 Figure 5-9: lH N M R spectrum of czs-[Ru(acac)2(Im)2][CF3S03] (55) in d6-acetone 230 Figure 5-10: *H N M R spectrum of cis-[Ru(acac)2(NMeIm)2][CF3S03] (58) in d6-acetone 232 Figure 5-11: ORTEP view of the cation of czs-[Ru(acac)2(Im)2][Tf] (55); 50 % probability thermal ellipsoids are shown 234 Figure 5-12: ORTEP view of cz5-[Ru(acac)2(NMeIm)2][Tf] (58); 50 % probability thermal ellipsoids 235 Figure 5-13: ORTEP view of the cation of czs-[Ru(acac)2(2MeIm)2][Tf] (60); 50 % probability thermal ellipsoids 236 Figure 5-14: ORTEP view of the cation of czs-[Ru(acac)2(5MeIm)2][Tf] (62); 50 % probability thermal ellipsoids 237 Figure 5-15: TT N M R spectra for complexes 72 and 73 in d6-acetone 241 Figure 5-16: Proton signals for the coordinated NMelm ligands of 74 in d6-acetone.... 242 Figure 5-17: *H N M R spectrum of [Ru(hfac)(NMeIm)4][PF6] (76) in d6-acetone 242 Figure 5-18: TT N M R spectra of cis- and zra«s-Ru(lrfac)2(2MeIm)2 (77 and 79) in d6-acetone 243 Figure 5-19: TT N M R of the mixed-ligand complex Ru(hfac)2(4MeIm)(5MeIm) in d6-acetone 244 xviii Figure 5-20: Cyclic voltarnmograms for cis- and rra«s-Ru(hfac)2(2MeIm)2 in 0.1 M Bu 4 NC10 4 in MeCN 247 Figure 6-1: Diagram of 'tox' vessel used for the accumulation and toxicity experiments 257 Figure 6-2: Different types of SCCVII cells observed on microscope slides 260 Figure 6-3: Setup used for irradiation of the "ducks" in the radiosensitzation assay 263 Figure 6-4: Pictorial summary of the in vitro MoAb assay 264 Figure 6-5: Comparison of toxicity for selected compounds (at 100 u M except for TF5) in oxic and hypoxic SCCVII cells (incubated for 3 h); average of 2 or 3 experiments. (The plating efficiencies for the other 2N0 2Ims, which also show little oxic/hypoxic variability, reported in Table II-1, Appendix II.)... 267 Figure 6-6: Accumulation data for RuCl3(SR2508)2(EtOH) in SCCVII cells after a 3 h incubation; reported as ng Ru/106 cells (± 5 %) 272 Figure 6-7: DNA-binding data for RuCl3(SR2508)2(EtOH) in SCCVII cells after a 3 h incubation; reported as ng Ru/mg D N A (± 5 %) 274 Figure 6-8: Radio sensitization by SR2508 (600 uM) and EF5 (150 uM) and lack thereof by their Ru complexes RuCl3(SR2508)2(EtOH) (308 uM) and RuCl 3(EF5) 2 (EtOH) (60.6 uM); after 3 h pre-incubation with hypoxic SCCVII cells. ... 275 Figure 6-9: The indocarbocyanine fluorescent dye Cy3 276 Figure 6-10: Direct comparison of relative median fluorescence intensity and radioactive drug uptake for 9L cells incubated with 100 u M EF5 for 3 h at indicated oxygen partial pressures (adapted from ref. 13) 277 Figure 6-11: Relative mean fluorescence intensity for SCCVII cells incubated with lOOuM drug for 3 h under N 2 and then treated with ELK3-51 (Cy3); determined using flow cytometry (average of 2 or 3 expts.; ± 10 %). (The oxic data are reported in Table II-2, Appendix II.) 279 Figure 6-12: Log of mean fluorescence signal determined by flow cytometry versus the compound's reduction potential (determined by CV; chapter 3, Section 3.2.8). See Table II-3, Appendix II 280 xix Figure 6-13: Relative mean fluorescence intensity for SCCVTI cells incubated with lOOuM drug for 3 h under N 2 and then treated with ELK3-51; determined using image cytometry (data from 2 expts.; ± 1 0 %). (The oxic data are reported in Table II-4, Appendix II.) 281 Figure 6-14: Relative mean fluorescence intensity for SCCVTT cells incubated with drug for 3 h under N 2 and treated with ELK3-51; determined using flow cytometry (data for Ru complexes from 2 experiments; ± 1 0 %). Al l values are normalized for EF5 content (100 uM). The m-PtCl 2(NH 3)(EF5)(PtEF5) results were obtained from ref. 13 283 Figure 6-15: Relative mean fluorescence intensity for SCCVII cells incubated with lOOuM drug for 3 h under N 2 and treated with ELK5-A8; determined using flow cytometry. (The oxic data are reported in Table II-5, Appendix II.) . 285 Figure 6-16: Proposed recognition sites for the ELK5-A8 MoAb 286 Figure 7-1: Proposed coordination geometry of [RuCl(PPh3)2(2M4NF3(-l))]Cl 296 List of Schemes Scheme 2-1: Standard amide bond forming reaction 42 Scheme 3-1: Synthesis of petafluoropropylamine from perfluoropropionic acid.3 5 96 Scheme 3-2: Synthesis of IF5, the precursor to EF5 97 Scheme 3-3: Formation of aziridine ring upon exposure of 3 to U V light 98 Scheme 3-4: Cyclization of IBr in the presence of heat and/or UV-radiation 98 Scheme 3-5: Synthesis of 3-fluoropropylamine reported by Pattison et al.45 99 Scheme 3-6: Reaction of 3-hydroxypropylamine with DAST 100 Scheme 3-7: High yield Gabriel synthesis of 3-fluoropropylarnine hydrochloride 101 Scheme 3-8: Reaction mechanism for addition of side-chain via an amide linkage 102 Scheme 3-9: Synthesis of 7, via 8 from SR2508 102 Scheme 3-10: Oxidation of metronidazole with Jone's Reagent to yield the carboxylic acid derivative, 13 103 Scheme 3-11: Acid hydrolysis of 2-methyl-4-nitro-l-imidazolepropionitrile (abbreviated as xx RCN) 104 Scheme 3-12: New synthetic route to EF5 104 Scheme 3-13: Formation of the isobutylester (11) in the presence of H 2 0 106 Scheme 3-14: The 3-step synthesis of RevF5 from 2-bromoethylphthalamide 107 Scheme 3-15: Stepwise reduction of 2-nitroimidazole to 2-aminoimidazole 118 Scheme 3-16: Synthesis of ECll(-l) via the tosylate intermediate 124 Scheme 3-17: Synthesis of fluoroetanidazole (EFl(-l)) via a tosylate intermediate.77... 125 Scheme 3-18: Mechanism for reaction of SR2508 with D M F to yield the formate ester (18) 126 Scheme 4-1: Conversion of 31 -> 32 —> 33, with suggested formulations (L= 2MeIm). 158 Scheme 5-1: Synthesis ofNa[Ru(hfac)3] (68) from RuCl 3 ' 3H 2 0 212 Scheme 5-2: Synthesis of c/5-Ru(acac)(hfac)(MeCN)2 (88) from Ru(acac)2(hfac) (85)..223 Scheme 5-3: Attempted synthesis of c/'s-Ru(acac)(hfac)(MeCN)2 (88) from Ru(hfac)2(acac) (87) 224 Scheme 7-1: Proposed reduction of bis-CN compound (24) with BH3»THF 292 Scheme 7-2: Proposed synthesis of ligands containing two or three - C H 2 C F 2 C F 3 units..294 Scheme 7-3: Proposed synthesis of [18F]-EF1 similar to that used by Tewson to synthesize [18F]-fluoroetanidazole 295 Scheme 7-4: Possible F 2 addition to fluoroalkenes to yield EF4 and EF5 296 List of Tables Table 3.1: Summary of the T i NMR, 19F{lH} N M R and UV-Vis data for «N0 2 Im compounds and for their reaction with Bu4NF»H20 93 Table 3.2: Summary of reduction potentials for the 2-nitroimidazoles vs. SCE 94 Table 3.3: Summary of reduction potentials for the 2-methyl-5-nitroimidazoles vs. SCE.95 Table 3.4: Summary of reduction potentials for the 2-methyl-4-nitroimidazoles vs. SCE 95 Table 4.1: Conductivity measurements versus time for 44 in H 2 0 146 Table 4.2: Summary of attempted reactions for which no pure product was isolated or characterized 151 Table 5.1: Summary of Ru(III/II) reduction potentials vs. SCE 210 xxi Table 5.2 Summary of UV-Vis data for complexes 53, 70, 85 and 87 in MeOH [ U (exlO' 3 )] 215 Table 5.3 ! H and 1 9 F{ J H} N M R data for the Ru(III) tris(p-diketonato) complexes 216 Table 5.4 N M R data for the Im and Melm complexes [Ru(acac) 2(L) 2][CF 3S03] in d6-acetone at r.t 231 Table 5.5 N M R data for the free imidazole ligands and their complexes Ru(hfac)2(L)2 in d6-acetone at r.t 240 Table 6.1 Actual drug concentrations used for in vitro experiments 265 Table 6.2 Aerobic and hypoxic toxicity data (PE at 3 h) of Ru complexes in SCCVII cells. (See Table 6.1 for complex concentrations.) 269 Table 6.3 Accumulation of Ru complexes (normalized to 100 uM) in SCCVTI cells incubated for 3 h; expressed as ng Ru/106 cells (± 5 %) 270 Table 6.4 Amount of Ru bound to D N A extracted from SCCVII cells after 3 h exposure to complex (normalized to 100 uM; ng Ru/mg D N A (± 5 %)) 273 Table 7-1 : Summary of most significant in vitro assay results obtained during this thesis 300 Table 1-1 1 Experimental details for EF5 A l Table 1-1 2 Atomic coordinates and B i S 0 / B e q for EF5 A2 Table 1-1 3 Bond lengths (A) for EF5 A2 Table I-1 4 Bond angles (°) for EF5 A2 Table 1-2 1 Experimental details for EF2Bi-0 5 H 2 0 A3 Table 1-2 2 Atomic coordinates and B i 8 0 / B e q for EF2Br«0.5 H 2 0 A4 Table 1-2 3 Bond lengths (A) for EF2BH3.5 H 2 0 A4 Table 1-2 4 Bond angles (°) for EF2Br«0 5 H 2 0 A4 Table 1-3 1 Experimental details for MF5 A5 Table 1-3 2 Atomic coordinates and B i S 0 / B e q for MF5 A6 Table 1-3 3 Bond lengths (A) for MF5 A6 Table 1-3 4 Bond angles (°)for MF5 A6 Table 1-4 1 Experimental details for 2M4NF5 A7 Table 1-4.2 Atomic coordinates and B i s 0 / B e q for 2M4NF5 A8 Table 1-4.3 Bond lengths (A) for 2M4NF5 A8 Table 1-4.4 Bond angles (°)for 2M4NF5 A9 Table 1-5.1 Experimental details for 2M4NF1(-1) A10 Table 1-5.2 Atomic coordinates and B i s o / B e q for 2M4NF1(-1) A l 1 Table 1-5.3 Bond lengths (A) for 2M4NF1(-1) A l 1 Table 1-5.4 Bond angles (°)for 2M4NF1(-1) A l 1 Table 1-6.1 Experimental details for cycF3 A12 Table 1-6.2 Atomic coordinates and B i s o / B e q for cycF3 A13 Table 1-6.3 Bond lengths (A) for cycF3 A13 Table 1-6.4 Bond angles (°) for cycF3 A14 Table 1-7.1 Experimental details for [Ru(Im)6][CF3S03]2 A l 5 Table 1-7.2 Atomic coordinates and B i s 0 / B e q for [Ru(Im) 6][CF 3S0 3] 2 A16 Table 1-7.3 Bond lengths (A) for [Ru(Im) 6][CF 3S0 3] 2 A16 Table 1-7.4 Bond angles (°)for [Ru(Im) 6][CF 3S0 3] 2 A16 Table 1-8.1 Experimental details for [Ru(NMeIm) 6][CF 3S0 3] 2 A17 Table 1-8.2 Atomic coordinates and B i s o / B e q for [Ru(NMeIm) 6][CF 3S0 3] 2 A18 Table 1-8.3 Bond lengths (A) for [Ru(NMeIm) 6][CF 3S0 3] 2 A18 Table 1-8.4 Bond angles (°)for [Ru(NMeIm) 6][CF 3S0 3] 2 A18 Table 1-9.1 Experimental details for [Ru(5MeIm) 6][CF 3S0 3] 2 A19 Table 1-9.2 Atomic coordinates and B i s o / B e q for [Ru(5MeIm) 6][CF 3S0 3] 2 A20 Table 1-9.3 Bond lengths (A) for [Ru(5MeIm) 6][CF 3S0 3] 2 A20 Table 1-9.4 Bond angles (°)for [Ru(5MeIm) 6][CF 3S0 3] 2 A20 Table 1-10.1 Experimental details for /«e/--RuCl3(MeCN)3'CHCl3 A21 Table 1-10.2 Atomic coordinates and B i s 0 / B e q for /wer-RuCl 3(MeCN) 3 'CHCl 3 A22 Table 1-10.3 Bond lengths (A) for /w^-RuCl 3 (MeCNyCHCl 3 A22 Table 1-10.4 Bond angles (°)for we/--RuCl3(MeCN)3»CHCl3 A22 Table 1-11.1 Experimental details for Ru(hfac)3 A23 Table 1-11.2 Atomic coordinates and B i s o / B e q for Ru(hfac)3 A24 xxiii Table 1-11.3 Bond lengths (A) for Ru(hfac)3 A24 Table 1-11.4 Bond angles (°)for Ru(hfac)3 A25 Table 1-12.1 Experimental details for Ru(hfac)2(MeCN)2 A26 Table 1-12.2 Atomic coordinates and B i s o / B e q for Ru(hfac)2(MeCN)2 A27 Table 1-12.3 Bond lengths (A) for Ru(hfac)2(MeCN)2 A29 Table 1-13.2 Atomic coordinates and B i s o / B e q for Ru(acac)(hfac)(MeCN)2 A30 Table 1-13.3 Bond lengths (A) for Ru(acac)(hfac)(MeCN)2 A30 Table 1-13.4 Bond angles (°) for Ru(acac)(hfac)(MeCN)2 A31 Table 1-14.1 Experimental details for [Ru(acac) 2(Im) 2][CF 3S0 3]»benzene A32 Table 1-14.2 Atomic coordinates and B i s 0 / B e q for [Ru(acac)2(Im)2][CF3S03]'benzene..A33 Table 1-14.3 Bond lengths (A) for [Ru(acac) 2(Im) 2][CF 3S0 3]«benzene A33 Table 1-14.4 Bond angles (°) for [Ru(acac) 2(Im) 2][CF 3S0 3]»benzene A33 Table 1-15.1 Experimental details for [Ru(acac)2(NMeIm)2][CF3S03] A34 Table 1-15.2 Atomic coordinates and B i s o / B e q for [Ru(acac)2(NMeIm)2][CF3S03] A35 Table 1-15.3 Bond lengths (A) for [Ru(acac)2(NMeIm)2][CF3S03] A35 Table 1-15.4 Bond angles (°)for [Ru(acac)2(NMeIm)2][CF3S03] A36 Table 1-16.1 Experimental details for [Ru(acac)2(2MeIm)2][CF3SO3]«0.5 hexane A37 Table 1-16.2 Atomic coordinates and B i s o / B e q for [Ru(acac)2(2MeIm)2][CF3SO3]»0.5 hexane A3 8 Table 1-16.3 Bond lengths (A) for [Ru(acac)2(2MeIm)2][CF3SO3]»0.5 hexane A38 Table 1-16.4 Bond angles (°) for [Ru(acac)2(2MeIm)2][CF3SO3]»0.5 hexane A38 Table 1-17.1 Experimental details for [Ru(acac)2(5MeIm)2][CF3S03] A39 Table 1-17.2 Atomic coordinates and B i s o / B e q for [Ru(acac)2(5MeIm)2][CF3S03] A40 Table 1-17.3 Bond lengths (A) for [Ru(acac)2(5MeIm)2][CF3S03] A41 Table 1-17.4 Bond angles (°) for [Ru(acac)2(5MeIm)2][CF3S03] A41 Table II-1: Toxicity (PE ± 10 %) for selected compounds (at 100 uM) in oxic and hypoxic SCCVII cells (incubated for 3 h) A42 xxiv Table II-2: Relative mean fluorescence intensity for SCCVII cells incubated with l O O u M drug for 3 h under N 2 and then treated with ELK3-51 (Cy3); determined using flow cytometry (average of 2 or 3 expts.; ± 10 %) A42 Table II-3: Log of mean fluorescence signal determined by flow cytometry and the compound reduction potential (determined by C V vs. SCE) A42 Table II-4: Relative mean fluorescence intensity for SCCVII cells incubated with l O O u M drug for 3 h under N 2 and then treated with ELK3-51; determined using image cytometry (data from 2 expts.; ± 10 %) A43 Table II-5: Relative mean fluorescence intensity for SCCVII cells incubated with l O O u M drug for 3 h under N 2 and treated with ELK5-A8; determined using flow cytometry A43 X X V List of Abbreviations Abbreviation Meaning Sr relative permitivity 2D 2-dimensional A A S atomic absorption spectroscopy Ab antibody acacH acetylacetone B . M . Bohr magneton b.p. boiling point B C C R C British Columbia Cancer Research Centre B S A bovine serum albumin BSO buthionine sulphoximine CCD charge coupled device CHO Chinese hamster ovary (cell line) CNS central nervous system COSY correlated spectroscopy C T R O N chromatotron C V cyclic voltammetry d day DAST diethylaminosulfonytrifluoride DCC dicylcohexylcarbodiimide D C U dicyclohexylurea dd doubly distilled D E M diethylmaleate D M A D dimethyladenine D M F N,N-dimethylformamide dmso dimethylsulfoxide (solvent) D M S O dimethylsulfoxide coordinated through O-atom DMSO dimethylsulfoxide coordinated through S-atom D N A deoxyribonucleic acid dppb 1,4 - bis(diphenylphosphino)butane ELISA enzyme-linked immunosorbent assay E L K Edith Lord Koch en 1,2-diaminoethane (ethylenediamine) EPR electron paramagnetic resonance equiv. equivalent F A B fast atom bombardment F C M flow cytometry F D G fluorodeoxyglucose GC gas chromatography GI gastrointestinal GSH glutathione Gy gray xxvi List of Abbreviations (cont.) HEPES N-2-hydroxylethylpiperazine-N'-2-ethane sulfonic acid hfacH 1,1,1,6,6,6-hexafluoroacetylacetone HR-MS high resolution mass spectrometry HSC hypoxia selective cytotoxin iBuClFrm z'so-butylchloroformate I C M image cytometry Im imidazole IOI integrated optical intensity K a acid dissociation constant L A H lithium aluminum hydride L E T linear energy transfer lit. literature reference L M C T ligand-to-metal charge transfer LR-MS low resolution mass spectrometry LSJJVIS low energy secondary ionization mass spectrometry m.p. melting point M A L D I matrix assisted laser desorption ionization M L C T metal-to-ligand charge transfer M O molecular orbital MoAb monoclonal antibody M R I magnetic resonance imaging M W C O molecular weight cutoff NBS N-bromo succinimide N E M N-ethyl maleimide N H E normal hydrogen electrode NHS N-hydroxysuccinimide N M M N-methylmorpholine N 0 2 nitro N 0 2 l m nitroimidazole NVIm N-vinylimidazole OD optical density ORTEP Oakridge Thermal Ellipsoid Program P partition coefficient PBS phosphate buffered saline PE plating efficiency PET positron emission tomography pF paraformaldehyde r.t. room temperature RBF round-bottom flask RF radio-frequency Rf retention frequency R N A ribonucleic acid RS radiosensitizing xxvii List of Abbreviations (cont.) SCCVII squamous cell carcinoma VII (cell line) SCE saturated calomel electrode SDS sodium dodecylsulphate SER sensitivity enhancement ratio SF surviving fraction SR2508 etanidazole SR4233 tirapazamine (1,2,4-benzotriazin-3-amine-1,4-di-N-oxide) S T M U ' B F 4 0-(N-succinimidyl)-N, N , N ' , N'-tetramethyl-uronium tetrafluoroborate T/C treated/control TBAP tetrabutylammonium perchlorate TE tris(hydroxymethyl)aminomethane + ethylenediamine-tetraacetic acid Tf triflate (CF3SO3") Tf 2 0 trifluoro sulfonic anhydride TF5 3 -N-pentafluoropropylamido-1,2,4-benzotriazine-1,4-di-N-oxide TLC thin layer chromatography TMSO tetramethylenesulfoxide TNE T E + 150 mM NaCl TsCl p-toluenesulfonyl chloride V vinyl xxviii Key to Chemical Compound Numbers and Abbreviations The imidazole ring is numbered as shown in Figure i . The 1-position is typically referred to as the N-position at which various types of side-chains were added. For this work the 2-, 4- and 5-positions were usually substituted with either Me or N 0 2 groups to give the corresponding imidazole compounds (Ri = R 2 = R 4 - R 5 - H , Im; R 2 = R 4 = R 5 = H and R i = Me, NMelm; R x = R 4 = R 5 = H and R 2 = Me, 2MeIm; R x = R 2 = R4/5 = H and R4/5 = Me, 4(5)MeIm; R i = R 4 = R 5 = H and R 2 = N 0 2 , 2N02Im; R i = R 2 = R4/5 = H and R4/5 = N 0 2 , 4(5)N02Im; Rj = R 4 = H and R 2 = Me and R 5 = N 0 2 , 2Me5N02Im; R x = R 5 = H and R 2 = Me and R 4 = N 0 2 , 2Me4N02Im). Numerous compounds were synthesized and used during the course of this thesis. The nitroimidazoles synthesized in chapter 3 were typically given names corresponding to the derivative from which they were synthesized (E = etanidazole (SR2508), M = metronidazole, 2M4N = 2-methyl-4-nitro-) followed by the type and number of halogen atoms on the terminal propyl group attached to the amide N-atom of the N I side-chain (e.g. EF5 has a terminal -CF 2 -CF 3 group, 5 F-atoms). A (-1) included at the end of the compound's name (e.g. EFl(-l)) corresponds to a terminal ethyl group (i.e. one - C H 2 -group shorter). Other starting materials, side-products and incompletely characterized compounds are numbered 1 thru 27 in chapter 3. The Ru complexes reported in chapters 4 and 5 are numbered in the sequence in which they appear. It is suggested that the Quick Reference sheets on the following pages are used when reading the specific chapters. R4 R5 Figure i: Imidazole position numbering. xxix Chapter 3 Quick Reference Compound Library 2N0 2 Im compounds R Compound N Y N N 0 2 R O C H 2 C F 2 C F 3 C H 2 C F 2 C F 2 B r C H 2 C H 2 C F 3 C H 2 C F 3 C H 2 C H 2 C F 2 B r CH 2 CH=CF 2 C H 2 C H 2 C H 2 F C H 2 C H 2 F CH 2 CH 2 CH 2 C1 CH 2 CH 2 C1 C H 2 C H 2 C H 2 B r C H 2 C H 2 B r C H 2 C H 2 C H 3 CH 2 CH(CH 3 ) 2 EF5 EF4Br EF3 EF3(-1) EF2Br E=F2 EF1 EFl(-l) E C U ECll(-l) EBrl EBrl(-l) EPrA EIAA 5N0 2Im compounds N 0 2 Me R O C H 2 C F 2 C F 3 C H 2 C F 3 C H 2 C H 2 F CH 2 CH 2 C1 C H 2 C H 2 B r MF5 MF3(-1) MFl(-l) MCll(-l) MBrl(-l) 4N0 2 Im compounds 0 2 N Me 0 N ' I H ,R C H 2 C F 2 C F 3 C H 2 C F 3 C H 2 C H 2 F CH 2 CH 2 C1 C H 2 C H 2 B r 2M4NF5 2M4NF3(-1) 2M4NF1(-1) 2M4NC11(-1) 2M4NBrl(-l) R-Misc. Compounds ^CF 3 i o R = I, IF5 = CI, C1F5 IBr RevEF5 ImF5 XXX H H 1 -X0 2 H 1 . ~ N . C F 2 — ^ C F 3 3 4 H a N ^ ^ ^ ^ F .HC1 5 0 6 N 0 2 7 T o N 0 2 8 ° 9 / \ JSTH 3 CI NYN N 0 2 10 11 T F F 12 N 0 2 f \ . O H Me 13 N 0 2 VT Me 14 r K . 0 . Js. Me 15 N 0 2 Me 16 o2N 0 N y N ^ 7 OH Me 17 H 0 N Y N T 0 H N 0 2 ° 18 SO, N V N > NO, O—l 19 O 20 O N ^ N ^ V ^ ^ I j r ^ CBr2H 21 Io2 0 H H 22 N ^ V I 2 T 0 . ^ NOj \ / ^ C \ 23 24 T 0 F 25 NO, 26 H T 0 F 27 xxxi Chapter 4 Quick Reference Compound Library Compound Number 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 Composition [Ru(Im)6][CF3S03]2 [Ru(NMeIm) 6][CF 3S0 3] 2 [Ru(5MeIm) 6][CF 3S0 3] 2 rra/75-[Ru(CO)(DMF)(2MeIm)4][CF3S03]2 [Ru(DMF)(2MeIm)4] [CF 3 S 0 3 ] 2 [Ru(CF3 S03)x(2MeIm)4] [CF 3 S0 3 ] y (x=2, y=0, or x=l=y) [Ru(2N0 2Im) 6][CF 3S0 3] 2 [Ru(5N0 2Im) 6][CF 3S0 3] 2 [Ru(2Me5N0 2Im) 5][CF 3S0 3] 2 [Ru(DMF)4(SR2508)2] [CF 3 S0 3 ] 3 [Ru(DMF) 2(EF5) 2(EtOH) 2][CF 3S0 3] 3 /wer-RuCl3(2N02Im)3 fac- and /wer-RuCl3(5N02Im)3 "RuCl 3(2Me5N0 2Im) 3-3 C 0 2 " RuCl3(metro)3 RuCl3(SR2508)2(EtOH) RuCl 3(EF5) 2(EtOH) RuCl2(metro)4 cw-RuCl2(DMSO)2(en) *ra«s-RuCl2(DMSO)2(en) RuCl2(DMSO)2(EF5)(acetone) [RuCl2(dppb)(EF5)]2(u-Cl)2 [RuCl2(dppb)EF5]2(u-dppb) cw-RuCl 2(MeCN)4 /wer-RuCl3(MeCN)3 xxxii Chapter 5 Quick Reference Compound Library Compound Number 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 Composition Ru(acac)3 c/'5-[Ru(acac)2(MeCN)2][CF3S03] czs-[Ru(acac)2(Im)2][CF3S03] toms-[Ru(acac)2(Im)2][CF3S03] c/5-[Ru(acac)2(MeCN)(Im)][CF3S03] C75-[Ru(acac)2(NMeIm)2][CF3S03] cz5-[Ru(acac)2(MeCN)(NMeIm)][CF3S03] cw-[Ru(acac)2(2MeIm)2][CF3S03] fraws-[Ru(acac)2(2MeIm)][CF3S03] c/5-[Ru(acac)2(5MeIm)2][CF3S03] c/5-[Ru(acac)2(4MeIm)(5MeIm)] [CF 3 S0 3 ] [Ru(acac)2(2N02Im)2] [CF 3 S0 3 ] [Ru(acac)2(SR2508)2][CF3SO3] [Ru(acac)2(EF5)2][CF3S03] [Ru(acac)2(metro)2][CF3S03] [Na][Ru(hfac)3] [Ru(hfac)(EtOH)4] [hfac] Ru(hfac)3 c/5-Ru(hfac)2(MeCN)2 cz'5-Ru(hfac)2(Im)2 cw-Ru(hfac)2(Im)(MeCN) c/'5-Ru(hfac)2(NMeIm)2 [Ru(hfac)(NMeIm)4] [hfac] [Ru(hfac)(NMeIm)4] [PF6] c/s-Ru(hfac)2(2MeIm)2 ds-Ru(hfac)2(2MeIm)(MeCN) xxxiii Compound Number Composition 79 fr-a«5-Ru(hfac)2(2MeIm)2 80 c/5-Ru(rrfac)2(4MeIm)(5MeIm) 81 fra«5-Ru(rrfac)2(4MeIm)(5MeIm) 82 Ru(hfac)2(2N02Im)2 83 Ru(hfac)2(EF5)2 84 Ru(hfac)2(SR2508)2 85 Ru(acac)2(hfac) 86 [Na][Ru(hfac)2(acac)] 87 Ru(hfac)2(acac) 88 c/5-Ru(acac)(hfac)(MeCN)2 89 cz'5-Ru(acac)(hfac)(Im)2 90 cz5-Ru(acac)(hfac)(NMeIm)2 91 cw-Ru(acac)(hfac)(NMe!m)(MeCN) EF5 SR2508 xxxiv Acknowledgements I would like to thank my supervisors Brian James and Kirsten Skov for their guidance, encouragement, and support throughout the duration of this work. The low-key role they played was essential in helping me become a better scientist. I would also like to thank the departmental services, especially Lianne Diarge, Marietta Austria, Peter Borda, Steve Rak, the late Steve Rettig, and the people at mass spectrometry for their support. From the B C Cancer Research Centre I am indebted to Hans Adomat, "the all-knowing-one," for help with essentially everything associated with the in vitro experiments. The help provided by Haibo Zhou and Jennifer Decker (thanks for staying late with me so many times) was also greatly appreciated. I must also give special thanks to Cameron Koch and Alex Kachur at the University of Pennsylvania and Mike Adam at TRTUMF for their collaboration on this project. From BRJ's lab a special thanks must go to Jeff Posakony (Mr. Late Night) for being a great lab partner, running buddy, workout guru and diving mate. Never did I think that my second year T A would become my lab companion; thanks for waiting around Jeffy. Matt LePage (Pooch-Boy) must be commended for being the weirdest person in the lab B Y FAR! Thanks for making sure things did not get too serious Matt. Other members of the James group past and present: Ken MacFarlane, Terrance Wong, Patrie Meessen, Paul Cyr, Nathan Jones, Erin Ma, Elizabeth Cheu, Dave Kennedy (and all the rest) deserve honorable mention. Thanks to all for being my friend and also helping with the proofreading of parts of this thesis. I must also thank Jeff and Katja for the fishing trips to rainbow country and Paul (my hunting buddy) and Helen for the many fantastic moose hunting trips up north. I am sure that it was these trips away from the city which helped me keep my sanity in an otherwise chaotic world. Most importantly I must thank my beautiful wife, Kerry, for being the most caring, understanding and giving person I know. Without her encouragement throughout these past five years I surely would not have achieved this goal. I would also like to thank my parents and the rest of my family for all their love and support over the years. Deluxe! Asta-lavista baby! xxxv Chapter 1 Introduction 1.1 Introduction Cancer is a complex family of diseases that is characterized by abnormal gene expression, resulting in uncontrolled growth and spread of abnormal cells. If the spread (metastases) of the cancer cells is not controlled it can ultimately lead to death. Approximately one in three people in North America will develop some form of cancer in their lifetime, and of these people approximately two-thirds will die of the disease. The most common cancers in the western world are lung, breast and colorectal cancer, which account for approximately 50 % of all cancer related deaths.1 Although it is now possible to successfully treat ("cure") testicular cancer with cisplatin, most other cancers have limited response to chemotherapy and have only prolonged survival at best.2 Hence, a major goal is to develop a system of cancer treatment which cures patients of their tumours while avoiding harmful side-effects to normal tissues. Our increased understanding of tumour biology and the cellular processes determining the behavior of cancer cells has allowed treatments to be directed specifically towards the tumour.3 The differences in tumour versus normal tissue stem from the rapid growth of cells within tumours which leads to formation of a tumour microenvironment. Typically, solid tumours contain areas of low extracellular pH and regions of low oxygen concentration (hypoxia). Exploitation of these properties, especially hypoxia, has been advantageous for the treatment of cancer using both organic compounds (e.g. radiosensitizers, hypoxic-selective cytotoxins) and metal complexes (e.g. antineoplastic and antimetastatic agents, see Section 1.6.3). However, the early detection of cancer using known imaging procedures (i.e. MRI, PET), prior to metastasis, is essential to ensuring patient survival. With early detection there would be approximately a 30 % increase in patient survival post treatment.1 1 References on page 26 Chapter 1 This chapter introduces the concept of hypoxia, how it develops in tumours, how it is measured/imaged and the postulation that hypoxia may make tumours more aggressive. The use of metal complexes in cancer therapy, focussing on Ru, is also briefly reviewed. These two topics are the major focus of this thesis and are reflected in the design and synthesis of the new compounds reported. 1.2 Hypoxia 1.2.1 What is it? Hypoxic cells are those that are deficient in oxygen, yet remain viable, and are often found in tumours. Such cells may also exist in other living tissue; for example, both heart tissue damaged by cardiac arrest or a stroke and retinal tissue in the eye contain hypoxic cells. The tissue in malignant tumours often outgrows its supply of oxygen and nutrients as the tumour enlarges. This is due to an inadequate production of a network of functioning blood vessels and capillaries. This poorly organized vasculature accounts for many of the environmental features unique to tumours.4 Although cells deprived of oxygen and nutrients ultimately die, at any given time a tumour may possess viable hypoxic cells. There are currently two accepted mechanisms by which hypoxia may develop in tumours. First is the classical "chronic" diffusion-limited model of hypoxia.5 With chronic hypoxia, as oxygen diffuses from the blood, a gradient is formed where oxygen concentration decreases with increased distance from the vasculature (Figure 1-1). Necrotic Hypoxic Figure 1-1: Oxygen gradient observed in a typical tumour (adapted from ref. 6). 2 References on page 26 Chapter 1 Studies have determined the mean distance from the vasculature to the necrotic regions of a tumour to be 150 urn, consistent with the estimated diffusion range of oxygen.5 The cells just inside this distance, a layer of one to two cells thick, were proposed to be hypoxic. The second mechanism is illustrated by the "acute" form of hypoxia which results from a localized pinching-off of blood vessels7 (i.e. when blood flow within the tumour vasculature fluctuates, resulting in a temporary cessation due to the transient opening and closing of vessels8'9). The cells affected from acute hypoxia will, however, only be resistant to radiotherapy if the hypoxic environment lasts throughout the treatment (Section 1.2.1.1). The presence of viable hypoxic cells (both chronic and acute) has been observed in both animal9'10 and human tumours.11 1.2.1.1 Resistance to Treatment: Radioresistance Ionizing radiation is commonly used to kill tumour cells, however, it is not equally effective on each cell within a tumour. The fate of cells after radiation can be attributed to the biological effect of ionizing radiation which is greatly affected by the oxygen partial pressure within the cells (the "oxygen effect"). The ratio of hypoxic to oxic doses needed to achieve the same biological effect is referred to as the oxygen enhancement ratio (OER). For sparsely ionizing radiations, such as X - or y-rays, the OER at high doses has a value between 2.5 and 3 (Figure 1-2). For densely ionizing radiation T 1 1 1 1 r I i_i L. i i> • O S «S 15 20 25 30 Dose (Gy) Figure 1-2: Typical results for X-ray irradiation of Chinese Hamster Ovary cells under aerobic (O2) or hypoxic (N2) conditions (adapted from ref. 6). 3 References on page 26 Chapter 1 there is no oxygen effect. For the oxygen effect to be observed, oxygen must be present during the irradiation because the lifetime of the radicals produced by radiation via the indirect effect of low LET radiation is very short. The free radicals formed (typically from the hydrolysis of water) result in cleavage of chemical bonds leading to lesions which initiate the chain of events that results in the final expression of biological damage. The chemical damage produced by free radicals in D N A can be sometimes repaired under hypoxia, but the damage is "fixed", that is, made permanent and not infrequently irrepairable if molecular oxygen is present. The sensitivity of living cells to damage induced by radiation is due to a competition between oxygen and endogenous thiols (e.g. glutathione) for radiation induced radicals on the D N A . 1 2 The electron affinic oxygen reacts with the unpaired electrons produced on the D N A resulting in lesions which the cell has difficulty repairing, particularly the double-strand break.12 At the same time, hydrogen atom donation to the unpaired electron from molecules such as glutathione yield lesions which are more readily repaired and it is this process which predominates in the absence of oxygen. To combat this effect various chemicals (DEM, N E M , diamide) which oxidize thiols have been tested against D N A repair with limited success. However, buthionine sulphoximine (BSO), a compound which selectively inhibits y-glutamyl cysteine synthetase (the enzyme responsible for glutathione synthesis), exhibits some radiosensitization and has been shown to increase the effectiveness of other radiosensitizers (e.g. nitroimidazoles).13'14 As a result of the radioresistance of hypoxic cells, a number of different methods have been employed to treat many tumours which might contain hypoxic regions. 1) Fractionated radiation treatment (2-2.5 Gy administered 5 days a week for 6 weeks) ensures that the treated cells could not accumulate and repair sublethal damage and the effect of increased cellular proliferation (repopulation) would be nullified. Moreover, the variation in the radio sensitivity of cells in different phases of the cell cycle and the ability of hypoxic cells to undergo reoxygenation would both be accounted for; 2) hyperbaric oxygen could be used to increase oxygen concentration within the blood in hopes of increasing the oxygen concentration in the tumour; 3) hypoxic-selective radiosensitizers (Section 1.3.1), many of which are oxygen mimetic, are useful. Like oxygen, these species 4 References on page 26 Chapter 1 are highly electron affinic, but unlike oxygen they are not metabolized as they diffuse through the tissue and so actually reach and hence accumulate within the hypoxic cells; and 4) the use of bioreductive cytotoxins, which selectively target and directly kill hypoxic cells, can be beneficial. Methods 2-4 (all containing selectivity for hypoxia) provide a therapeutic advantage over other treatments because hypoxic cells are rare in normal tissue while they occur frequently in tumour tissue. 1.2.2 Hypoxia: The Aggressor? Over the past 60 years there has been increasing evidence that hypoxia is prevalent in certain tumours and it poses a major problem in the treatment of cancer with either radiotherapy or chemotherapy.5'15'16 A meta analysis reporting all hypoxic sensitizer studies (>10,000 patients), including those agents now known to be ineffective, showed only a 14 % improvement in outcome.17'18 There is clearly an urgent need to find better methods of detection, and to select those patients who would benefit from adjuvant therapy such as a sensitizer or hypoxic cytotoxin. There is compelling evidence that hypoxia in the tumour has far greater implications than the response to treatment of the primary tumour itself. Direct evidence for this comes from the assessment of hypoxia in patients containing carcinomas of the uterine cervix, using an Eppendorf electrode to determine oxygen concentration prior to treatment.19'20 The treatment of patients with radiation during this study showed a direct correlation between tumour inhibition and low hypoxic fractions. More importantly though, is that hypoxia was prognostic for outcome (curability) after surgery. For patients containing either cervix carcinomas21 or soft tissue sarcomas,22 the result was significantly better in those whose tumours were relatively well oxygenated and were treated with surgery, implying that tumours with hypoxia are more aggressive. Indirect evidence for hypoxia as an aggressor includes reports that hypoxia promotes genomic instability, a potential factor in tumour heterogeneity and progression.23 The report that the expression of the p53 gene (which encodes a tumour supressor protein important in chemical and viral carcinogenesis, apoptosis and cell cycle drug resistance) is induced by hypoxia and that cells exposed to hypoxia die by p53-mediated apoptosis is essential to our 5 References on page 26 Chapter 1 understanding of hypoxia. Hypoxic regions were shown to correlate spatially with highly apoptotic regions of tumours.24 However, cells lacking intact p53 were resistant to low oxygen concentrations; hence hypoxia selects for mutant p53, and these cells survive. Hypoxia was also shown to be present in metastases which affected the cell's response to radiation; this may potentially increase the metastatic potential of a tumour. The above findings which suggest that hypoxia makes a tumour more "aggressive" (or vice-versa), greatly increase the importance of determining which patients have hypoxia-containing tumours (not all tumours contain hypoxia). Finding treatments which will not only eradicate the hypoxic cells in the primary tumour, but also be directed towards the metastatic disease accompanying the aggressiveness caused by hypoxia, is also imperative. 1.3 Role of Nitroimidazoles in Cancer Therapy 1.3.1 Hypoxic Radiosensitizers In the early 1970s nitroimidazoles were first proposed as radiosensitizers because of the high electron affinity of the nitro group. The 5-nitroimidazole, metronidazole (Figure 1-3), demonstrated radiosensitization in a variety of murine tumour systems25 and its low toxicity and long biological half-life made it a good candidate for clinical testing. The drug at high doses, however, resulted in severe gastrointestinal (GI) intolerance which led to the search for a more effective, less toxic radiosensitizer. A number of 2-nitroimidazoles (2NC«2lrn) were investigated due to their higher electron affinity (more positive E1/2 value) and were shown to be more active than their 5-nitroimidazole analogues in vitro. The leading compound of this series was misonidazole (Figure 1-3) which showed marked activity in many different tumour models in vivo using several different assays.25 Initial clinical trials with misonidazole using single doses appeared promising, with a decrease in GI intolerance from that of metronidazole. Notwithstanding, when multiple doses were administered patients developed neurological complications,26 27 limiting the doses required to produce significant hypoxic cell sensitization. 6 References on page 26 Chapter 1 K Me H N. OH l=\ OCH 3 OH Metronidazole (flagyl) Misonidazole (Ro 07-0582) Etanidazole ( S R 2 5 0 8 ) Figure 1-3: Structures of metronidazole, misonidazole and etanidazole. Investigations of the mechanism of neural toxicity observed for misonidazole and other related compounds have revealed that toxicity is related to lipophilicity of the compounds.28 This study revealed that lowering the partition coefficient (lipid:water) of a 2-nitroimidazole, via alteration of side-chain substituents on N l of the imidazole ring, decreased the penetration of drug across the blood-brain barrier. The reduction of lipophilicity also led to poorer absorption of the drug into the blood from the peritoneal cavity. This problem was overcome by using intravenous injection. These studies also revealed that the partition coefficient of a 2N02lm could be lower than that of misonidazole by a factor of up to 20 without compromising tumour penetration or the radio sensitizing effectiveness. The 2N02lm, etanidazole (Figure 1-3), exhibited a greatly reduced lipophilicity compared to misonidazole (0.046 compared to 0.43) resulting in more rapid plasma clearance and less penetration into neural tissue, while maintaining the potency of misonidazole in vivo29 Clinical trials of etanidazole demonstrated that patients could tolerate a total dose of three times that of misonidazole without significant side-effects; however, etanidazole did result in nausea, vomiting and peripheral neuropathies which limited its use.29 1.3.2 Chemistry of Nitroimidazoles in Biological Systems The importance of using nitro-containing compounds (e.g. nitroimidazoles) for treatment and/or detection of hypoxia stems from the reducibility of the nitro group at physiological pH. A mechanism involving the reduction of the nitro group has been proposed (Figure 1-4).30 Under hypoxic conditions the nitro group is reduced (it has been suggested that cytochrome P450 reductase is the dominant enzyme for the reduction of 2N0 2Ims 3 1) through a number of reactive intermediates, one or more of which is 7 References on page 26 Chapter 1 presumed to be toxic to the cells.32 The selectivity towards hypoxic cells is because the reverse oxidation to the parent nitroimidazole is effectively reduced in the presence of low oxygen concentrations. In aerobic cells the reverse oxidation results in futile cycling of the nitro group with concomitant production of the superoxide radical and hydrogen peroxide.33 A direct correlation between the sensitizing efficiency and electron affinity has been supported by various pulse radiolysis studies on one-electron transfer reactions between known radiosensitizers.34 r=\ .NR N O , Nitroreductase n • F u t f l e n W 2 Redox Cycling W 2 N ^ NR N 0 2 ' Radical Anion e',2H -H,0 r\ e-NO Nitroso N y M " N H O H Hydroxylamine e- r\ » N . .NR N H , Amine Figure 1-4: Nitroimidazole reduction scheme (adapted from ref. 30). The position of the nitro group on the imidazole ring is also of marked importance to the effectiveness of the nitroimidazole as a sensitizer. The electron affinity of the nitro group is measured by the one-electron reduction potential of the nitroimidazole; the more positive the Em value the easier the reduction to the radical nitro anion. The reduction potentials for 2-, 4-, and 5-nitroimidazoles fall in typical ranges (Figure 1-5), with those of the 2NC«2lms being the most positive.32 N O , 2-nitroimidazole N O , 0 , N N ^ N - R 5-rritroimidazole 4-nitroimidazole E i / 2 (mV) -360 to -400 -510 to -545 -540 to -685 Figure 1-5: Ranges of reduction potentials for nitroimidazoles vs. N H E at pH 7. 8 References on page 26 Chapter 1 In the nitroimidazole reduction scheme, focussing on 2-nitroimidazoles, three products are expected from reduction in two-electron jumps. 2-Aminoimidazoles can be prepared as stable compounds by chemical methods and have been detected as metabolites of 2-nitroimidazoles.35 However, the amines lack significant biological activity, leading to the consensus that one of the intermediate reaction products, either the 2-hydroxylamino-imidazole or the 2-nitrosoimidazole, is involved in adduct formation within the cell. 3 6 The nitrosoimidazole, l-methyl-2-nitrosoimidazole, has been isolated in the solid state, but was found to be unstable in aqueous solution with a maximum half-life of 10 h at neutral pH. The product in acidic conditions, typical of hypoxic regions, has been identified, and results from the nucleophilic addition of water to the imidazole ring (Figure 1-6).37 Products at neutral pH are more complex, but appear to arise from ring opening reactions of the hydrolyzed species. Reactions have also been observed with phosphines and aromatic amines, but the products have not yet been identified. 2-Nitrosoimidazole is more electron affinic than oxygen, can cause strand breaks in D N A and its half-life in cells with [GSH] > 0.1 mM is estimated to be less than one second. The rapid redox reaction with GSH is thought to be the major mode of activity for the nitrosoimidazole, leading to the formation of hydroxylaminoimidazole.37 H H H O . I I ^OH / \ 2 H 2 0 , H+ N s ^ .N—Me ** H - N o . .N—Me y © y N O N H O H Figure 1-6: Nucleophilic addition of water to l-methyl-2-nitrosoimidazole. The 2-hydroxylaminoimidazoles have a longer half-life (-3-5 min) than the 2-nitrosoimidazoles and form stable species in acid. 3 7 Products of further reaction have been identified as the dihydrodihydroxyimidazolium ions resulting from the N-0 bond heterolysis of the hydroxylamine that yields the nitrenium ion, which subsequently reacts with two equivalents of water. In the presence of other nucleophiles, the chemistry is modified such that the nucleophile competes with water for the nitrenium ion intermediate, and nucleophile-incorporated products are formed (Figure 1-7). These products include 9 References on page 26 Chapter 1 covalent binding with GSH and other protein thiols, guanine (DNA), phosphates (DNA) and other biological amines.37'39 H-N(T~}NR © y NHOH 2 -hydroxylarninoirnidazolium ion r\ N^NR W ^OH^ 0 H H e03 P O vj_L^OH © y NH, f=\ NH © nitreniurn ion H20 H HPO, ^0 H H PhNH I^ ^OH H-N^ .NR © y NH, N \ / N R n NH RSH H H H-N^ -NR © y NH, N. .NR Y® NH 1 I H H j N^R i ». I V r o H-N*. .NR©y' NH, Figure 1-7: Proposed reactions of 2-hydroxyaminoimidazole within the cell. 1.4 Imaging of Hypoxic Tumours The ideal marker of tissue oxygenation should: 1) have a large ratio of absolute binding between hypoxic tissue and aerobic tissue (large dynamic range); 2) exhibit an oxygen dependency that is similar to that of cellular radiosensitivity; 3) have a partition coefficient which facilitates its rapid diffusion into both well-perfused and poorly-perfused tissues; 4) contain a probe which can be accurately measured by invasive and/or non-invasive procedures; and 5) have no toxicity at its optimum concentration for this diagnostic use.30 Many techniques have been used in an attempt to meet these requirements, a few of which are summarized in the following subsections. 10 References on page 26 Chapter 1 AAA Radiolabelling of Hypoxic Cells By taking advantage of the selective bioreduction and subsequent binding of nitroimidazoles under conditions of low oxygen concentration, it was proposed that incorporation of 3 H or 1 4 C into the imidazole would allow for detection of hypoxia by radioactivity. For 14C-misonidazole, for example, it was determined using this technique that the binding to cellular molecules within hypoxia was proportional to the square root of the drug concentration.40 Detection methods are liquid scintillation techniques, immunochemical techniques or through autoradiography of histological sections depending on the label attached to the nitroimidazole. This last mentioned technique was used for patients with different metastatic lesions and, although there was a large tumour to plasma ratio of radiolabeled nitroimidazole, the high dose of radiolabels required made this approach unsuitable as a predictive assay of tumour hypoxia.41 1.4.2 Nuclear Medicine Techniques Similar to the technique described above, this approach uses high energy y-emitting radionuclides as labels on nitroimidazoles. The 82Br-labelled 4-bromo-misonidazole compound was one of the first compounds developed in an attempt to quantify the amount of hypoxia in a tumour.42 Biodistribution and blood clearance studies revealed that the label was retained on the compound in vitro and in vivo and that the tumour and normal tissue distribution kinetics were similar or superior, from the perspective of imaging, to those of the parent compound, misonidazole. There was also higher uptake in the tumour for the bromo derivative, likely resulting from the significantly higher partition coefficient when compared to that of misonidazole (2.9 vs. 0.43). Moreover, this property also led to increased toxicity, presumably due to CNS toxicity. Hence, the usefulness of this compound as a hypoxia imaging agent was abandoned. More recently, a series of 2-nitroimidazoles linked to iodinated sugars has been developed. These 123I-radioiodinated azomycin (2N02im) nucleosides show promise for use in the detection of tumour hypoxia while avoiding the toxicity of previously developed compounds. The leading compound in this series is iodoazomycin arabinoside (IAZA, Figure 1-8) and was shown to undergo hypoxia-dependent binding in murine tumour 11 References on page 26 Chapter 1 models in vitro."'1' Tumour to blood ratios of 5-10 at 8 h post injection were observed indicating areas of hypoxia in vivo for EMT-6 tumours in BALB/c mice. These promising results led to the clinical study investigating 1 2 3 I - IAZA as a potential non-invasive marker of hypoxia which is currently under consideration.44 Figure 1-8: Iodoazomycin arabinoside (IAZA). 1.4.3 Fluorescent Probes Analogous to the methods used to develop radiolabeled compounds for the detection of hypoxia, the addition of a fluorophore chemically linked to the side-chain of the 2-nitroimidazole moiety was pursued in hopes of utilizing the selective binding of the nitro species in hypoxia. Provided that the chain linking the 2-nitroimidazole moiety does not suffer metabolic cleavage from the fluorophore, the fluorescence should be "fixed" within the hypoxic cells. The best compound developed using this technique was an indolizine-linked 2-nitroimidazole species (Figure 1-9) which yielded a maximum fluorescence differential between oxic and hypoxic cells of 12.5:1 when tested in vitro. Unfortunately, when this probe was tested in vivo there was no observable difference between oxic and hypoxic cells, presumably due to an insufficient concentration of compound remaining in the cells.45 CM \ ,CH2CH(OH)CH2—N CONH(CH2)2N \ •(CH2)2OH Figure 1-9: Indolizine-linked 2-nitroimidazole, fluorescent hypoxia probe. 12 References on page 26 Chapter 1 Fluorescent markers for the detection of hypoxia have also been used.46 The first study of this kind involved raising polyclonal antibodies against the protein adduct of reductively-activated CCI-103F. The antibodies were coupled with a fluorescent dye and when introduced to hypoxic cells pretreated with CCI-103F, the antibodies remained intact within the cells, as determined by fluorescence. Studies with murine tumour models in vivo have shown that the antibody binds to the drug adduct in tumour sections with patterns similar to those revealed by autoradiographic studies of 3Ff-labelled CO-1COF. 4 6 Figure 1 - 1 0 : Structures of the hypoxia-selective fluorinated 2-nitroimidazoles CCI-103F More recently, a pentafluorinated derivative of etanidazole ( E F 5 ) was synthesized and a monoclonal antibody (ELK3-51), significantly more specific than the previously used polyclonal antibodies for E F 5 and EF5-protein adducts, was isolated. The dynamic range (between nitrogen and air) of binding by E F 5 , as measured by radioactive drug uptake, was a factor of at least 50 in human cells and 100 in rodent cells. 4 7 ' 4 8 Consequently, it is important that the drug adduct-detecting antibodies have very high sensitivity and specificity. This was the case for ELK3-51, which revealed, by flow cytometric analysis, the same dynamic range and oxygen dependence as the radioactivity assay. The advantage of the immunological method is that the signal difference between oxic and hypoxic cells can be detected on an individual cell basis which allows for measurement of inter- and intra-tumour heterogeneity. The development of solid state microscopy techniques now enables quantitative assessment of the hypoxia distribution within a tumour.49 This technique is currently the most powerful method available for hypoxia imaging and quantification. H CCI-103F EF5 and E F 5 . 13 References on page 26 Chapter 1 1.4.4 Magnetic Resonance Imaging (MRI) MRI is a biological multi-slice imaging technique utilizing the effect of atoms on an applied magnetic field, and is able to access specific areas of the body. The application of a magnetic field results in the orientation of nuclear spins, usually parallel to the magnetic field. For MRI, the patient is subjected to energy in the form of radiowaves via an RF (radio-frequency) pulse. This energy causes some of the low energy atoms (P, H , or F) to flip to their high energy (anti-parallel) state. A two-dimensional Fourier transform then manipulates the signal into an output which allows a computer to transform it into a signal strength versus intensity picture. M R I has been used as a non-invasive diagnostic method for detecting tumour hypoxia for molecules containing 3 1 P, *H and 1 9 F. 3 1 P-MRI makes use of the fact that compounds associated with cellular energetics contain phosphorus atoms (especially ATP). In animal tumours these spectra were shown to correlate with changes in oxygen concentration as measured by microelectrodes, but there was too much scatter in the data to be clinically useful.50 ^ - M R I makes use of the high rate of glycolytic metabolism found in tumours resulting from lactate production. However, interference from other cellular signals make conclusive measurement difficult.51 The incorporation of 1 9 F atoms into known hypoxia-selective agents like 2-nitroimidazoles has also been investigated.52'53 The hexafluorinated 2-nitroimidazole CCI-103F was shown to lose the hexafluoropropyl substituent from its side-chain in vivo, but the trifluorinated hypoxia probe SR4554 was shown to remain intact upon enzymatic reduction in vivo, displaying selective uptake into the tumour in preliminary M R I studies.53 M R I studies with EF5, which revealed a dynamic range of at least 50 using immunohistochemical techniques (as above), are currently underway in hopes of producing a clinically useful MRI hypoxia probe.54 1.4.5 Positron Emission Tomography (PET) PET is a biological imaging technique that allows one to measure in microscopic detail the function and/or metabolism of a labelled compound in specific areas throughout the body. This technique uses short-lived, neutron-poor, positron-emitting isotopes 14 References on page 26 Chapter 1 produced by a cyclotron such as 1 5 0 (half-life of 2 min), 1 3 N (10 min), n C (20 min) and 1 8 F (110 min) which are incorporated into biological substrates, substrate analogues or drugs. PET, like MRI, allows for multi-slice imaging. The most attractive candidate radionuclide is 1 8 F due to its sufficiently long half-life which allows for complex or multistep organic synthesis. The decay of 1 8 F is predominantly by positron emission (97 %) and the positron is of relatively low energy (0.635 M e V maximum) and thus has a short mean range (2.39 mm in water).55 When the positron meets an electron they annihilate each other. Al l that remains is energy in the form of two y-rays, of equal intensity, going in opposite directions (Figure 1-11). The y-rays are then registered by detectors within the PET camera to generate a useful image. Figure 1-11: y-ray production from positron-emitting isotope [ F]. Fluorine is an interesting substituent in medicinal chemistry with a long and successful history of F-incorporation into organic pharmaceuticals. Fluorine is small (atomic radius, 0.57 A) and exhibits high C-F bond energies (~ 106 kcal/mol). Fluorine is also extremely electronegative (x = 3.98) and its substitution into a molecule can often produce large changes in physiochemical and biological properties. A good example of this is the differences observed between SR2508 (no F-atoms) and EF5 (5 F-atoms). The most widely used PET agent to date, fluorodeoxyglucose (FDG), incorporates an 1 8 F atom into the ring of D-glucose and is used for imaging in both cancerous and non-cancerous tissues.56 15 References on page 26 Chapter 1 In order to provide a non-invasive detection method for hypoxia, F was incorporated into a proven, hypoxia-selective 2-nitroimidazole. The first and most investigated compound of this type was [18F]-fluoromisonidazole;57 however, this compound is not ideal due to its hypoxia-independent tissue retention.58 More recently, syntheses of [18F]-fluoroetanidazole59 and [ 1 8F]-EF1 6 0 were reported, both showing promising hypoxia imaging results in cells and small animals. These findings indicate that the introduction of a non-invasive hypoxia imaging agent into the clinic is not far off. 1.5 Role of Metals in Cancer Therapy Initially the use of metals and metal complexes as antitumour agents was not considered by most scientists due to the fact that numerous cases of poisoning, especially with heavy metals, had been documented. Researchers did not assume that such metals could be used to synthesize effective therapeutic agents when introduced to different ligand environments.61 Systematic studies into the relationship between chemical structure and efficacy or toxicity of metal compounds began only about sixty-five years ago. Collier and Krauss in 1931 descibed the first metal complex tumour-inhibiting study and developed a theory, which remains valid: "the effect of a heavy metal on experimental murine cancer is not only due to the metal alone, but also the structure of the compounds and the type of compound."62 Nevertheless, it was not until 34 years later (1965) when the first metal-containing anticancer drug was discovered, somewhat serendipitously by Rosenberg et al.,63 that the field of metal complexes in cancer therapy escalated. Rosenberg's new platinum "wonder-drug" was initially generated as an electrolysis product formed by the association of dissolved platinum (from a Pt electrode) with the ammonium cation from a culture medium containing E. coli bacteria.63'64 The resultant product was identified as (NFL^MPtCle] which converted to c/s-[PtCl4CNH3)2] via a photochemical reaction. Similar neutral, cz's-amine Pt species were synthesized and the series was tested for possible antitumour activity against the mouse tumour, Sarcoma 180. Al l of the compounds reduced tumour growth, but the most promising complex was cis-[PtCl2(NH3)2] (now known as cisplatin, Figure 1-12),65 which entered clinical trials in 1971 and was approved by the US Food and Drug Administration in 1978 for treatment of 16 References on page 26 Chapter 1 testicular and ovarian cancer.2 Due to the severe toxic side-effects associated with cisplatin therapy, in particular nausea and vomiting, neuropathy, ototoxicity and most significantly, nephrotoxicity, the search for metal complexes having similar activity but with decreased side-effects was initiated. cisplatin carboplatin JM216 AMD473 Figure 1-12: Chemical structures of antitumour agents cisplatin, carboplatin, JM216 and AMD473. A less toxic analogue of cisplatin, carboplatin (Figure 1-12), was introduced into clinical practice in 1981, but disappointingly the two drugs were effective against the same population of tumours and thus shared cross-resistance with one another.66'67 A novel class of platinum (IV) dicarboxylates which demonstrated promising oral activity against a number of murine tumour models was discovered within the past ten years.6 8 In preclinical studies, one such compound (bis-acetatoammine(dichloro)cyclohexylamineplatinum (IV)) (JM216, Figure 1-12) provided a structural lead to platinum complexes which may circumvent transport-determined resistance to cisplatin;69'70 this complex contains toxicological properties reminiscent of carboplatin rather than cisplatin.71 More recently, the sterically hindered Pt(II) complex AMD473 has entered clinical trials because of its circumvention of cisplatin resistance.72 Since the discovery of cisplatin, a myriad of different metal complexes (Ti, V , Mo, Tc, Fe, Ru, Co, Rh, Ni , Pd, Cu, Au, Ga, Ge, Sn and Bi) have been synthesized and tested for their antitumour activity both in vitro and in vivo in hopes of discovering drugs which have activity against Pt resistant tumours. A few examples are Au(I) auranofin which displays irreversible inhibition of D N A synthesis in HeLa cells73; Rh(II) carboxylates which act to inhibit murine ascitic tumours74; and Ti(IV) titanocene dichloride which 17 References on page 26 Chapter 1 demonstrates a reduced growth of solid animal tumour systems and in some cases this effect was more pronounced than that of cisplatin.75 For the most part these complexes do not have the chemical properties required for an antitumour agent (see below). Most metal complexes have dose-limiting side-effects and are not tumour specific, restricting their possible use in the clinic. For example, Ga(III) nitrate was shown to inhibit tumour growth by greater than 90 % in 6 of 8 solid tumours transplanted subcutaneously,76 but the gallium concentration was 130 times higher in the kidney than in the tumour.77 The lack of selectivity likely explains this drug's poor clinical effect. It is important, when considering the utility of a potential metal-containing antitumour agent, to take into account its kinetic characterisitics. If too labile, the complex is likely to interact with physiological nucleophiles before reaching its site of action in the tumour and, if too inert, the complex may not interact with its biomolecular target as required to produce the antitumour effect.78 In this regard, many metals form complexes that would be deemed too labile. However, the Pt group metals (Ru, Os, Rh, Ir, Pd, Pt) are capable of forming many relatively inert complexes whose kinetic properties can be tailored by variation of the ligands about the metal centre. Although the main focus over the past 30 years has been on Pt antitumour complexes, the biological effects of Ru complexes, with the possibility of producing Ru-based drugs active in the treatment of human malignancies, have also been widely investigated.79"85 The Ru complexes differ from cisplatin, particularly when tumour sensitivity and the effects of tumour reduction on host survival time are concerned. This suggests that Ru complexes (the focus of this thesis work) have a select place in cancer therapy, opening further directions in the development of a new generation of anticancer agents.79 1.6 Ruthenium Complexes 1.6.1 Chemical Properties Relevant to Tumour Treatment As antitumour drugs are generally cytotoxic, the development of drugs which selectively target the tumour to ensure a selective cell kill is essential. The use of Ru over Pt is one approach to avoid the problems associated with Pt drug therapy. 18 References on page 26 Chapter 1 There are two significant differences between Ru- and most of the Pt-based antitumour drugs which make their prospective mechanism of action advantageous: 1) the usually six-coordinate, octahedral geometry of Ru complexes (versus square planar Pt (II)) may allow for new modes of binding to nucleic acids, and 2) the ease with which electron-transfer takes place for the Ru1™ couple because of a small activation barrier (see below).8 3 The most stable complexes in aqueous solution are generally Ru 1 1 and R u m and these are usually readily interconverted via oxidation/reduction within biological systems. The Ru 1 1 1 ion (d5, low-spin) usually accepts 7i-electron density into its partially filled dn-orbital and thus has a strong attraction for halides and anionic oxygen ligands; a low-spin, d 6 Ru 1 1 ion has little affinity for 7i-donors, but readily binds 7i-acceptor ligands. This is due to further extended rL. orbitals which lead to better 7r-orbital overlap.86 The R u m reduction potentials can be tuned by changing the type of coordinated ligand.8 7 For example, the reduction potential for [L(NH3)5Rum] varies from -0.08 V (L = OH) to 1.1 V (L = N 2 ) . In general, anionic a-donor ligands lower the reduction potential, while neutral or cationic, 7c-acceptor ligands raise it. As the Franck-Condon barrier to electron transfer is low for Ru1™ couples, owing to small bond distortions between these two ions, redox reactions are often rapid.8 8 The reduction of R u m to Ru 1 1 allows for the use of Ru 1 1 1 complexes as "prodrugs" for Ru 1 1 species which have an elevated propensity to coordinate to biomolecules. In particular, the binding of R u n with the imine N of imidazole rings on histidine and purines89'90 predominates where reducing power is relatively high and oxygen, which can reoxidize Ru 1 1 back to R u m , is virtually non-existent. Such an environment occurs in many tumours 9 1 , 9 2 which are hypoxic due to their higher metabolism and diffusion limited oxygen supply.9 3'9 4 Tumours also typically have a lower pH which favours pH-dependent reduction of some complexes.95 Consequently, the Ru n /Ru m ratio should be higher in most types of tumours than in other tissues leading to increased binding and hence localized cytotoxicity. And so, the treatment of tumours with Ru 1 1 complexes (potentially toxic for all tissues) should be avoided in favour of the more inert R u m prodrugs. 19 References on page 26 Chapter 1 The accumulation of Ru ions within tumour tissues is not only due to reduction of R u m , but is also significantly increased by coupling with transferrin.96'97 This phenomenon is due to the similarities between Ru f f l and Fe m , which both have a high affinity for phenolate ligands9 8'9 9 which are integral in complexing F e m in transferrin for transport through the blood to the tissues. Release of R u m from transferrin (like Fe m ) may be facilitated by lower pH or reduction to Ru 1 1 , 1 0 0 which has little affinity for phenolate. 1.6.2 DNA Binding Antitumour agents are expected to block the D N A activity of tumour cells in order to be effective, and accordingly D N A is thought to be the major target for most antitumour drugs. Ru complexes are no exception. The bulk of evidence suggests that D N A binding is possible, and in some instances is as frequent and effective as with cisplatin, for which a D N A interaction has been well documented using X-ray crystallography.101 Cisplatin (czs-DDP) is believed to result mainly from the 1,2 (GG) and 1,3 (GXG) intrastrand cross-links within D N A which trans-DDP cannot form. The 1,2 cross-link involves binding of Pt to N7 of adjacent deoxyguanosines along the same strand of D N A to form a 17-membered chelate ring (Figure 1-13).102 This intrastrand adduct can be removed by nucleotide excision repair (an entire oligomer containing the adduct is excised103), but some results suggest that the repair of a G G intrastrand cross-link is significantly impaired when compared with that for monofunctional adducts.104 The intrastrand cross-link also results in a bending of the D N A double helix (a 33° kink) which promotes binding of a specific class of damage-recognition proteins.105 It has been postulated that these proteins may be important in cisplatin toxicity by blocking access of repair enzymes to damaged D N A . 1 0 6 The trans isomer of cisplatin (transplatin) was determined to be much less active likely due to its steric inability to form the major intrastrand cross-links. Ru complexes, and more importantly Ru 1 1 complexes, were shown to be capable of strong interaction with oligonucleotides (e.g. with calf thymus D N A ) 8 9 for which preferential N - N adduct formation between two adjacent guanines, identical to that of 20 References on page 26 Chapter 1 cisplatin, was observed. However, Clarke reports that the G G cross-link formation with cr H Figure 1-13: Cisplatin/DNA interaction to form a G G intrastrand cross-link. D N A is not essential for antitumour activity, with, for example, pentaammine-Ru complexes as it is for the platinum drugs.83 The N7 of guanine is relatively exposed in the major groove of B - D N A and the metal complexes, typically positively charged species (e.g. [Ru(NH 3) 5(H 20)] 2 +), undergo a fairly strong electrostatic attraction for the polyanionic DNA. Once bound to the D N A the Ru complex is also thought to induce localized denaturation of the D N A which makes the interior sites (N6 of adenine and N4 of cytosine) available for Ru 1 1 interaction.107 Interestingly, a comparison of D N A binding for cis- and zraws-RuCl2(DMSO)4 reveals that the trans isomer covalently binds at a markedly higher reaction rate (behaviour opposite to that of cis- and trans-DDP) and protects G-rich sequences from cutting by specific restriction endonucleases, indicating a preferential interaction with adjacent guanines.90 DNA-bound Ru was also found after in vivo treatment with antitumour doses of either Ru 1 1 or Ru 1 1 1 complexes; this was measured by atomic absorption spectroscopy, in samples of D N A extracted from tumour cells and from cells of lung, internal mucous, liver 21 References on page 26 Chapter 1 and spleen tissue. These data displayed unequal distribution of DNA-bound Ru, which was more concentrated (selective localization) in the lung and tumour tissue than in the spleen or liver. 1 0 8 1.6.3 Tumouricidal Effects of Ru Complexes The selectivity of Ru complexes for tumours makes them highly appropriate for investigation in antitumour research. As none of the existing antitumour drugs are devoid of severe and dose-limiting side-effects, the search for novel non-toxic compounds continues. Preliminary studies have revealed that "Ru complexes may help to overcome the dose-limiting complication, and it would be hypothesized that new derivatives could find a place in therapy as new-generation target-specific anticancer drugs."79 There is a myriad of newly synthesized Ru complexes reported in the literature each year, but those which exhibit antitumour activity can be grouped into three categories. The first is the Ru-ammine complexes, the first series of Ru complexes to undergo screening for antitumour activity. The compound, /ac-RuCl3(NH3)3 exhibited excellent activity in several tumour screens (up to T/C of 190 %); however, its poor solubility impedes it from adequate formulation as a drug.80 Keppler and co-workers synthesized a number of similar complexes, yielding species which were more soluble in aqueous media due to their anionic charge. For example, cw-[RuCl2(NH3)4]Cl (with T/C values up to 160 %) is active against P388 leukemia, a highly sensitive platinum tumour model used to compare Ru drug performance to cisplatin. 1 0 9 ' 1 1 0 The second group of complexes is that characterized by the presence of DMSO, which, as a ligand, facilitates transport through, and penetration into membranes, increases complex water solubility and is significantly labile to undergo displacement by stronger binding nitrogenous D N A bases (especially N7 of guanine). Both cis- and trans-RuCl 2 (DMSO )4 exhibit interesting antitumour properties, with the trans isomer exhibiting significantly higher toxicity than the cis isomer. More importantly, both exhibit antimetastatic activity greater than cisplatin.82 The third group (the focus of this thesis) is comprised of Ru complexes containing one or two heterocyclic ligands such as imidazole, pyrazole, indazole and their 22 References on page 26 Chapter 1 methyl substituted derivatives. Generally, these complexes exhibit significant antitumour activity.81 For example, the water soluble, bis-imidazole complex [ImH][RuCl4(Im)2] showed a T/C value of about 200 % against the P388 leukemia (cf. cisplatin, T/C 175 % ) . 1 U The complex also possessed potent and selective antitumour activity in a carcinogen-induced colorectal tumour of the rat. Assessment of cytotoxicity with all three classes of Ru antitumour complexes generally revealed a low in vitro effectiveness.79 This is not the case for all Ru complexes, but it is clearly evident that most of them, despite being capable of reducing in vivo tumours and/or increasing the lifespan of tumour-bearing animals, are essentially void of any capacity to reduce cell viability when tested in vitro19 This implies that if the new Ru complexes are initially screened using in vitro assays, the result might be misleading and hence fail to unmask the real effectiveness of these complexes against cancer growth. The development of Ru complexes by examining the general effect of tumour growth, or by examining tumour size variation with solid tumours, may also limit the scope of these complexes as potential antitumour agents. It is clear that each Ru complex must be used at a maximum tolerated dose to significantly reduce primary tumours, while the reduction of solid tumour metastases is achievable at doses with undetectable cytotoxic side-effects for the hosts.112 The development of antimetastatic drugs for the treatment of solid tumours is essential in combating cancer which can readily disseminate throughout the host. Tumour metastases behave quite differently from their primary neoplasms due to differences in drug sensitivity, antigenicity and clonogenic capacity.113 The drugs available for the management of human tumours are usually only active on primary lesions. Therefore, if the tumour has metastasized, the antimetastatic drug becomes equally as important in prevention of tumour growth. The first compound capable of selective activity against metastatic tumours (frara-Na[RuCl4(DMSO)(Im)]) was developed by Sava et al. and showed a significant advantage for the postsurgical prognosis when associated with ablation of primary tumours.114 The mechanism of action of this complex is, however, poorly understood and so obtaining a full understanding of this process is essential to the 23 References on page 26 Chapter 1 preparation of complexes which will circumvent the actual limits of trans-Na[RuCl4(DMSO)(Im)]. 1.7 Thesis Overview At the onset of this thesis work the major goal was to synthesize, characterize and assess in mamalian cells in vitro new Ru nitroimidazole complexes for use as potential non-toxic radiosensitizers. However, because of the development of E F 5 (a pentafluorinated 2N02lm) and a highly selective monoclonal antibody to E F 5 (ELK3-51), 4 7 ' 4 9 the focus shifted to the synthesis and characterization of new 2-, 4- and 5-nitroimidazoles with halogenated side-chains (Chapter 3). Alteration of the side-chain and position of the nitro group on the imidazole ring provides valuable information on the interaction of the monoclonal antibody with the drug (Chapter 6). This thesis consists of seven chapters plus appendices of 17 X-ray structures. Chapter 2 describes the general experimental procedures used in the characterization of the newly synthesized complexes, including the synthetic procedures for all the Ru starting complexes used in Chapter 4. Each of Chapters 3, 4 and 5 contains its own experimental and results and discussion sections. Chapter 3 contains the synthesis and complete characterization of nearly 30 new 2-, 4-, and 5-nitroimidazoles, containing N l -halogenated side-chains, including their reduction potentials as determined by cyclic voltammetry (CV). Chapter 4 describes the synthesis and characterization of many Ru complexes obtained from the reaction of imidazoles and nitroimidazoles with different Ru precursors (e.g. [Ru(DMF) 6][CF 3S03]3, RuCl 3*3H 20, cis/trans-RuC\2(DMSO)4). Many of these complexes were paramagnetic Ru(IIT) species which made their characterization by N M R more difficult. Chapter 5 is comprised of the synthesis and characterization of Ru(II) and Ru(III) bis-acetylacetonate (acac), bis-l,l,l,6,6,6-hexafluoroacetylacetonate (hfac) and mixed acac/hfac complexes with imidazole and nitroimidazole ligands. A comprehensive study of the reduction potentials of these complexes, as determined by C V , suggests the capability of synthesizing a complex with a reduction potential in the desired biological range by altering the substituents on the chelating P-diketonato ligands. Chapter 6 focuses on the in vitro testing of selected nitroimidazoles and Ru 24 References on page 26 Chapter 1 imidazole/nitroimidazole complexes. The toxicity and monoclonal antibody recognition (using both M o Abs ELK3-51 and ELK5-A8) of some of the nitroimidazoles from Chapter 3 are evaluated in SCCVII cells. The Ru complexes are tested for their toxicities and abilities to accumulate within the cell and to coordinate to DNA, and a preliminary radiosensitization study of two of these complexes is included. Conclusions and suggestions for future experiments, including a report on the studies that are currently underway to answer some of the chemical and biological questions generated from this thesis work, are reported in Chapter 7. 25 References on page 26 Chapter 1 1.8 References 1 Cook, A. R. The New Cancer Source Book, Health Reference Series, Vol. 12; Omnigraphics, Inc., Detroit, 1996, p. 3. 2 Flicker, S. P. In Metal Compounds in Cancer Therapy; S. P. Flicker, Ed.; Chapman and Hall, London, 1994, p.3. 3 Ruddon, R. W. Cancer Biology; Oxford University Press, New York, 1995, p. 3. 4 Jain, R. K. J. Natl. Cancer Inst. 1989, 81, 570. 5 Thomlinson, R. H. ; Gray, L. H . Br. J. Cancer 1955, 9, 539. " 6 Farrell, N . In Transition Metal Complexes as Drugs and Chemotherapeutic Agents; R. Ugo and B. R. James, Eds.; Kluwer Academic, Dordrecht, 1989, p. 183. 7 Brown, J. M . Br. J. Radiol. 1979, 52, 650. 8 Dudar, T. E.; Jain, R. K. Cancer Res. 1984, 44, 605. 9 Chaplin, D. J.; Durand, R. E.; Olive, P. L . Int. J. Radial Oncol. Biol. Phys. 1986, 12, 1279. 10 Moulder, J. E.; Rockwell, S. Int. J. Radiat. Oncol. Biol. Phys. 1984,10, 695. 11 Vaupel, P.; Kallinowski, F.; Okunieff, P. Cancer Res. 1989, 49, 6449. 12 Ward, J. F. In Radioprotectors and Anticarcinogens; O. F. Nygaard and M . G. Simic, Eds.; Academic Press, New York, 1983, p. 73. 13 Bump, E. A ; Yu, N . Y. ; Brown, J. M . Pharmac. Ther. 1990, 47, 117. 14 Koch, C. J.; Skov, K. A. Int. J. Radiat. Oncol. Biol. Phys. 1994, 29, 345. 15 Bush, R. S.; Jenkin, R. D. T.; Allt, W. E. C ; Beale, F. A ; Bean, H . A ; Dembo, A. J.; Pringle, J. F. Br. J. Cancer 1978, 37, 255. 16 Teicher, B. A ; Laxo, J. S.; Sartorelli, A. C. Cancer Res. 1981, 41, 73. 17 Coleman, C. N . Br. J. Cancer 1996, 74, S297. 18 Overgaard, J. Conference Summary - clinical. 9th Int. Conf. on Chem. Modifiers of Cancer Treatment, Oxford, U . K., Aug. 22-26, 1995. 19 Vaupel, P.; Schlenger, K. ; Knoop, C ; Hdckel, M . Cancer Res. 1991, 57, 3316. 20 Hockel, M . ; Schlenger, K. ; Knoop, C ; Vaupel, P. Cancer Res. 1991, 51, 6098. 26 Chapter 1 21 Hockel, M . ; Schlenger, K. ; Aral, B.; Mitze, M . ; Schaffer, U . ; Vaupel, P. Cancer Res. 1996, 56, 4509. 22 Brizel, D. M . ; Scully, S. P.; Harrelson, J. M . ; Layfield, L. J.; Bean, J. M . ; Prosnitz, L . R.; Dewhirst, M . W. Cancer Res. 1996, 56, 941. 23 Paquette, B. ; Little, J. B. Cancer Res. 1994, 54, 3173. 24 Graeber, T. G.; Osmanian, C ; Jacks, T.; Housman, D. E.; Koch, C. J.; Lowe, S. W.; Giaccia, A. J. Nature 1996, 379, 88. 25 Fowler, J. F.; Denekamp, J. Pharmacol. Therapeut. 1979, 7, 413. 26 Dische, S. Int. J. Radiat. Oncol. Biol. Phys. 1991, 20, 147. 27 Brown, J. M . Int. J. Radiat. Oncol. Biol. Phys. 1984, 52, 650. 28 Brown, J. M . ; Workman, P. Radiat. Res. 1980, 82, 171. 29 Brown, J. M . ; Yu, N . Y . ; Brown, D. M . ; Lee, W. W. Int. J. Radiat. Oncol. Biol. Phys. 1981, 7, 695. 30 Chapman, J. D.; Lee, J.; Meeker, B. E. In Selective Activation of Drugs by Redox Processes; G. E. Adams, A. Breccia, E. M . Fielden, P. Wardman, Eds.; Plenum Press, New York, 1990, p. 316. 31 Joseph, P.; Jaiswal, A. K. ; Stobbe, C. C ; Chapman, J. D. Int. J. Radiat. Oncol. Biol. Phys. 1994, 29, 351. 32 Adams, G. E. ; Flockhart, I. R.; Smither, C. E.; Stratford, I. J.; Wardman, P.; Watts, M . E. Radiat. Res. 1976, 67, 9. 33 Biaglow, J.; Varaes, M ; Roizen-Towle, L . ; Clarke, E.; Epp, E.; Astor, M . ; Hall, E. Biochem. Pharmacol. 1986, 35, 77. 34 Adams, G. E.; Cooke, M . S. Int. J. Radiat. Biol. 1969, 75, 457. 35 Walton, M . I.; Workman, P. Biochem. Pharmacol. 1987, 36, 887. 36 Palcic, B. ; Josephy, D.; Skov, K ; Skarsgard, L . D. Proceedings of a Conference Sponsored by the Radiosensitizer/Radioprotector Working Group DCT/NCI, 1979, p. 16. 27 Chapter 1 37 McClelland, R. A. In Selective Activation of Drugs by Redox Processes; G. E . Adams, A. Breccia, E. M . Fielden, P. Wardman, Eds.; Plenum Press, New York, 1990, p. 125. 38 Raleigh, J. A. ; Koch, C. J. Biochem. Pharmacol. 1990, 40, 2457. 39 McClelland, R. A. ; Panicucci, R ; Rauth, A. M . J. Am. Chem. Soc. 1987,109, 4308. 40 Koch, C. J.; Stobbe, C. C ; Baer, K. A. Int. J. Radiat. Oncol. Biol. Phys. 1984, 10, mi. 41 Urtasun, R. C ; Chapman, J. D.; Raleigh, J. A ; Franko, A. J.; Koch, C. J. Int. J. Radiat. Oncol. Biol. Phys. 1986,12, 1263. 42 Rasey, J. S.; Krohn, K. A.; FreaufF, S. Radiat. Res. 1982, 91, 555. 43 Mannan, R. H. ; Somayaji, V. V. ; Lee, L ; Mercer, J. R.; Chapman, J. D.; Wiebe, L . I. J. Nucl. Med. 1991, 32, 1764. 44 Parliament, M . B.; Chapman, J. D.; Urtasun, R. C ; McEwan, A. J.; Goldberg, L . ; Mercer, J. R ; Mannan, R. H. ; Wiebe, L. I. Br. J. Cancer 1992, 65, 90. 45 Parrick, J.; Hodgkiss, R. J.; Jones, G. W.; Middleton, R. W.; Rami, H . K. ; Wardman, P. In Selective Activation of Drugs by Redox Processes; G. E. Adams, A. Breccia, E. M . Fielden, P. Wardman, Eds.; Plenum Press, New York, 1990, p. 249. 46 Raleigh, J. A ; Miller, G. G.; Franko, A. J.; Koch, C. I ; Fuciarelli, A. F.; Kelly, D. A. Br. J. Cancer 1987, 56, 395. 47 Koch, C. J.; Evans, S. M . ; Lord, E. M . Br. J. Cancer 1995, 72, 869. 48 Matthews, J.; Adomat, H . ; Farrell, N . ; King, P.; Koch, C ; Lord, E.; Palcic, B. ; Poulin, N . ; Sangulin, I ; Skov, K. Br. J. Cancer 1996, 74, S200. 49 Evans, S. M . ; Joiner, B. ; Jenkins, W. T.; Laughlin, K. M . ; Lord, E. M . ; Koch, C. J. Br. J. Cancer 1995, 72, 875. 50 Fu, K. K. ; Wendland, M . F.; Iyer, S. B.; Lam, K. N . ; Eugeseth, H . ; James, T. L . Int. J. Radiat. Oncol. Biol. Phys. 1990,18, 1341. 51 Stone, H. B. ; Brown, J. M . ; Philips, T. L. ; Sutherland, R. M . Radiat. Res. 1993, 136, 422. 28 Chapter 1 52 Raleigh, J. A. ; Franko, A. J.; Treiber, E. O.; Lunt, J. A. ; Allen, P. S. Int. J. Radiat. Oncol. Biol. Phys. 1986, 72, 1243. 53 Aboagye, E. O.; Kelson, A. B. ; Workman, P. Anti-cancer Drug Design 1998, 13, 703. 54 Siemann, D. W.; Inglis, B. A. ; Cranston, B. A.; Lord, E. M . ; Koch, C. J. 10th International Conference on Chemical Modifiers of Cancer Treatment, Jan. 28-31, 1998, Clearwater, Florida, Abstract p. 43. 55 Kilbourn, M . R. Nucl. Med. (Nuclear Science Series); National Academy Press, Washington, D. C , 1990, p. 1. 56 Brennan, M . B. Chemical and Engineering News 1996, 2, 26. 57 Grierson, J. R.; Link, J. M . ; Mathis, C. A. ; Rasey, J. S.; Krohn, K. A . J. Nucl. Med. 1989, 30, 343. 58 Rasey, J. S.; Nelson, N . J.; Chin, L. ; Evans, M . L. ; Grunbaum, Z. Radiat. Res. 1990, 122, 301. 59 Tewson, T. J. Nucl. Med. Biol. 1997, 24, 755. 60 Kachur, A. V. ; Evans, S. M . ; Shiue, C.-Y.; Dolbier, Jr., W. R.; L i , A.-R.; Roche, A. ; Skov, K. A. ; Baird, I. R.; James, B. R.; Koch, C. J. Appl. Rod. hot. (submitted Oct., 1998). 61 Keppler, B. K. In Metal Complexes In Cancer Chemotherapy; B . K . Keppler, Ed.; V C H , Weinheim, 1993, p.3. 62 Source of information, ref. 61. 63 Rosenberg, B.; Van Camp, L. ; Krigas, T. Nature 1965, 205, 698. 64 Rosenberg, B. ; Van Camp, L . ; Grimely, E. B. ; Thomson, A. J. J. Biol. Chem. 1967, 242, 1347. 65 Rosenberg, B.; Van Camp, L. ; Trosko, J. E.; Mansour, V . H . Nature 1969, 222, 385. 66 Gore, M . ; Fryatt, I.; Wiltshaw, E.; Dawson, T.; Robinson, B. A. ; Calvert, A . H . Br. J. Cancer 1989, 60, 767. 29 Chapter 1 67 Eisenhauer, E.; Swerton, K. ; Sturgeon, J. et al. In Carboplatin: Current Perspectives and Future Directions; P. Burn, R. Caretta, R. Ozols, M . Rozencweig, Eds.; W. B . Saunders Company, Philadelphia, 1990, p. 133. 68 Harrap, K. R.; Murrer, B. A.; Giandomenico, C. In Platinum and Other Metal Coordination Compounds in Cancer Chemotherapy; S B . Howell, Ed.; Plenum Press, New York, 1991, p. 391. 69 Kelland, L . R ; Mistry, P.; Abel, G.; Freidlos, F.; Loh, S. Y . ; Roberts, J. J.; Harrap, K. R. Cancer Res. 1992, 52, 1710. 70 Kelland, L . R.; Jones, M . ; Abel, G.; Harrap, K. R. Cancer Chemother. Pharmacol. 1992, 30, 43. 71 McKeage, K. J.; Morgan, S. E.; Boxall, F. E.; Murrer, B. A.; Hard, G. C ; Harrap, K. R. Proc. Am. Assoc. Cancer Res. 1992, 33, A3197. 72 Holford, J.; Sharp, S. Y . ; Murrer, B. A ; Abrams, M . ; Kelland, L . R. Br. J. Cancer 1998, 77, 366. 73 Simon, T. M . ; Kunishima, D. H. ; Vilbert, G. J.; Lorber, A. Cancer 1979, 44, 1965. 74 Howard, R. A ; Sherwood, E.; Erck, A ; Kimball, A. P.; Bear, J. L . J. Med. Chem. 1911, 20, 943. 75 Kbpf-Maier, P.; Kopf, H . Arzneim.-Forsch./DrugRes. 1987, 70, 103. 76 Adamson, R. H . ; Canellos, G. P.; Sieber, S. M . Cancer Chemother. Rep. 1975, 59, 599. 77 Hall, S. W.; Yeung, K. ; Benjamin, R. S. Clin. Pharmacol. Ther. 1979, 25, 82. 78 Buckley, R. G. In Metal Compounds in Cancer Therapy; S. P. Flicker, Ed.; Chapman and Hall, London, 1994, p. 92. 79 Sava, G. In Metal Compounds in Cancer Therapy; S. P. Flicker, Ed.; Chapman and Hall, London, 1994, p. 65. 80 Clarke, M . J. In Progress in Clinical Biochemistry and Medicine; Springer-Verlag, Berlin, 1989, Vol. 10, p. 26. 30 Chapter 1 81 Keppler, B . K. ; Hern, M. ; Juhl, U . M. ; Berger, M . R ; Niebl, R ; Wagner, F. E. In Progress in Clinical Biochemistry and Medicine; Springer-Verlag, Berlin, 1989, Vol. 10, p. 41. 82 Mestroni, G.; Alessio, E.; Calligaris, M . ; Attia, W. M . ; Quadrifoglio, F.; Cauci, S.; Sava, G.; Zorzet, S.; Pacor, S.; Monti-Bragadin, C ; Tamaro, M . ; Dolzani, L . In Progress in Clinical Biochemistry and Medicine; Springer-Verlag, Berlin, 1989, Vol . 10, p. 71. 83 Clarke, M . J. In Metal Complexes in Cancer Chemotherapy; B. K. Keppler, Ed.; V C H , Weinheim, 1993, p. 129. 84 Mestroni, G.; Alessio, E.; Sava, G.; Pacor, S.; Coluccia, M . In Metal Complexes in Cancer Chemotherapy; B. K. Keppler, Ed.; V C H , Weinheim, 1993, p. 157. 85 Keppler, B. K. ; Lipponer, K . -G. ; Stenzel, B.; Kratz, F. In Metal Complexes in Cancer Chemotherapy; B. K. Keppler, Ed.; V C H , Weinheim, 1993, p. 188. 86 Wishart, J. F.; Bino, A ; Taube, H . Inorg. Chem. 1986, 25, 3318. 87 See Chapter 5, Table 5.1. 88 Chou, M . ; Creutz, C ; Sutin, N . J. J. Am. Chem. Soc. 1977, 99, 5615. 89 Kelly, J. M . ; Feeney, M . M . ; Tosi, A. B.; Lecomte, J. P.; Kirsch-De Mesmaeker, A. Anticancer Drug Des. 1990, 5, 69. 90 Loseto, F.; Mestroni, G.; Lacidogna, G.; Nassi, A ; Giordano, D.; Coluccia, M . Anticancer Res. 1991, 77,1549. 91 Palmer, B. D.; Wilson, W. R ; Pullen, S. M . J. Med. Chem. 1990, 33, 112. 92 Luk, C. K. ; Veinot-Drebot, L . ; Tjan, E. J. Natl. Cane. Inst. 1990, 82, 684. 93 Vaupel, P.; Kallinowski, F.; Okunleff, P. Cancer Res. 1989, 49, 6449. 94 Sasai, K. ; Ono, K. ; Hiraoko, M . Int. J. Radiat. Oncol. Biol. Phys. 1989,16, 1477. 95 Herman, T. S.; Teicher, B. A ; Holden, S. A. Cancer Res. 1989, 49, 3338. 96 Som, P.; Oster, Z. H . ; Matsui, K. ; Gugliemi, G.; Persson, B. R ; Pellettieri, M . L . ; Srivastava, S. C ; Richards, P.; Atkins, H. L. ; Brill, A. B. Eur. J. Med. 1983, 8, 491. 31 Chapter 1 97 Srivastava, S. C ; Mausner, L. F.; Clarke, M . J. In Ruthenium and other Non-Platinum Metal Complexes in Cancer Chemotherapy; M . J. Clarke, Ed.; Springer-Verlag, Heidelberg, 1989, Vol.10, p. 11. 98 Snetsinger, P. A. ; Chasteen, N . D.; van Willigen, H. J. J. Am. Chem. Soc. 1990, 112, 8155. 99 Pell, S. J.; Salmousen, R.; Abelliera, A. ; Clarke, M . J. Inorg. Chem. 1984, 23, 385. 100 Bomford, A. ; Young, S. P.; Williams, R. Biochem. 1985, 24, 3472. 101 Takahara, P. M . ; Frederick, C. A.; Lippard, S. J. J. Am. Chem. Soc. 1996, 118, 12309. 102 Fichtinger-Schepman, A. M . J.; van der Veer, J. L. ; den Hartog, J. H . J.; Lohman, P. H. M . ; Reedijk, J. Biochem. 1985, 24, 707. 103 Beck, D. J.; Popoff, S.; Sancar, A.; Rupp, W. D. Nucleic Acid Res. 1985, 13, 7395. 104 Page, J. D.; Husain, I.; Sancar, A.; Chaney, S. G. Biochem. 1990, 29, 1016. 105 Hughes, E. N . ; Engelsberg, B. N . ; Billings, P. C. J. Biol. Chem. 1992, 267, 13520. 106 Donahue, B. A. ; Augot, M . ; Bellon, S. F.; Treiber, D. K.; Toney, J. H . ; Lippard, S. J.; Essigmann, J. M . Biochem. 1990, 29, 5872. 107 Clarke, M . J.; Buchbinder, M . ; Kelman, A. D. Inorg. Chim. Acta 1978, 27, L87. 108 Sava, G. Unpublished results, from ref. 18. 109 Keppler, B. K. ; Wehe, D.; Endres, H. ; Rupp, W. Inorg. Chem. 1987, 26, 844. 110 Keppler, B. K. ; Wehe, D.; Endres, H. ; Rupp, W. Inorg. Chem. 1987, 26, 4366. 111 Keppler, B. K. ; Rupp, W.; Juhl, U . M . ; Endres, H. ; Niebl, R.; Balzer, W. Inorg. Chem. 1987, 26, 4366. 112 Sava, G ; Pacor, S.; Mestroni, G ; Alessio, E. Anti-Cancer Drugs 1992, 3, 25. 113 Talmadge, J. E. Cancer Metast. Rev. 1983, 2, 25. 114 Sava, G.; Pacor, S.; Coluccia, M . ; Mariggio, M . ; Cocchietto, M . ; Alessio, E. ; Mestroni, G Drug Invest. 1994, 8, 150. 32 Chapter 2 General Experimental 2.1 Materials 2.1.1 Chemicals 2.1.1.1 Imidazoles Im, 2MeIm and 4MeIm (from Aldrich) and NMelm (from Merck) were used as received. 2N02lm was synthesized according to literature procedures1 or purchased from Aldrich. Other nitroimidazoles purchased from Aldrich included: 4(5)N02lm, 2Me5N0 2Im, 2-methyl-4-nitro-l-imidazolepropionitrile and 2-methyl-4-nitro-l-imidazole-propionic acid. SR2508 was kindly donated by Dr. C. Koch (Univ. of Perm.), while metronidazole and RSU1111 were donated from the laboratory of Dr. M . Naylor (MRC -Chilton). 2.1.1.2 Amines The following amines were synthesized and kindly donated from Prof. W. Dolbier's group (Univ. of Florida): H 2NCH 2CF 2CF 3«HC1, H 2 N C H 2 C F 2 C F 2 B r ' H C l , H 2 NCH 2 CH 2 CF 3 'HC1 , H 2 N C H 2 C H 2 C F 2 B P H C 1 , H 2NCH 2CH 2CH 2F«HF. A l l other amines used throughout this work were purchased from Aldrich and generally used without further purification. E t 3 N was distilled prior to use. 2.1.1.3 Miscellaneous Reagents RuCl 3 *3H 2 0 was supplied on loan from both Johnson Matthey Ltd. and Colonial Metals, Inc. with a Ru composition of 39 to 41 %. N-methylmorpholine (NMM) and iso-butylchloroformate (iBuClFrm) were purchased from Aldrich and were vacuum distilled prior to use. Al l other commercial chemicals were used as purchased, unless otherwise indicated, from Aldrich Chemical Co., Fisher Scientific Co., Strem, Sigma Chemical Co., B D H Chemicals, Merck and Co., Inc. or Mallinckrodt Chemical Works. NBS and TsCl 33 References on page 46 Chapter 2 were recrystallized before use. Al l gases (N 2 , H 2 and CO; minimum purity 99.99 %) were purchased from Linde (Union Carbide, Inc.) and used as received. 2.1.2 Solvents Reagent grade D M F (Fisher) and dmso (Fisher) were distilled and dried over molecular sieves, and M e C N and EtOAc (anhydrous, Aldrich) were used as received. Other solvents were dried using the appropriate drying agent and distilled prior to use: MeOH (Mg), EtOH (Mg), THF (Na-K alloy), E t 2 0 (Na), hexanes (Na), pentane (CaH 2), acetone (K2CO3) and CH 2 C1 2 (CaH2). Deuterated solvents (d6-acetone, d6-dmso, CD 3 OD, CDCI3, C D 3 C N , D 2 0 ) were supplied by MSD Isotopes or Isotech Inc. and used immediately upon opening of an ampule (except for D 2 0 ) to ensure dryness. Please note that throughout this thesis dmso is used to indicate solvent while DMSO is used to indicate a coordinated ligand. 2.2 Analytical Techniques 2.2.1 Nuclear Magnetic Resonance Spectroscopy TI (200 MHz), 2 H (46.1 MHz), 1 3C{TI} (50.3 MHz) and 1 9F{TI} (188.2 MHz) N M R spectra were recorded on a Bruker AC200F spectrometer. 1 9 F N M R (282.2 MHz) spectra requiring larger sweep widths (SW up to 100,000) were recorded on a Varian XL300 spectrometer, while large sweep width *H N M R spectra were recorded on either the Varian XL300 (300.0 MHz) or the Bruker WH400 (400.0 MHz) spectrometers. 3 1 P{ 1 H} (121.5 MHz) and 1 3C{TI} (75.4 MHz) N M R spectra were recorded on a Varian XL300 spectrometer. The TI N M R chemical shifts were referenced to the residual proton signal of the internal deuterated solvent (5 7.24 for CDCI3, 4.78 for D 2 0 , 3.30 for CD3OD, 2.50 for (CD 3) 2SO, 2.20 for (CD 3 ) 2 C0; all are reported relative to the external standard of tetramethylsilane (8 0.00)) while the 1 3 C{ J H} N M R chemical shifts were also referenced to the shift from the solvent (5 29.8 for Me of acetone). The 1 9 F{Ti} N M R chemical shifts were referenced to trifluoroacetic acid in D 2 0 (external reference) and the 3 1 P{ 1 H} N M R chemical shifts were reported relative to 85 % H 3 P 0 4 (external reference) with the downfield shifts observed as positive. Al l spectra were recorded at r.t. 34 References on page 46 Chapter 2 The peak multiplicities are reported as observed in a given spectrum (s = singlet; d = doublet; t = triplet; q = quartet; p = pentet; sex = sextet; spt = septet; br = broad, vbr = very broad, including a combination of these symbols, for example, qt = quartet of triplets). For those signals where peak coincidences have altered the peak multiplicity for a specific set of protons, the expected multiplicity is reported in brackets (e.g. for - N H -CH2-CH2- one would expect to observe a doublet of triplets; however, a quartet is observed and so is recorded as q (dt)). The J values are reported for those nitroimidazole compounds containing F atom(s) on their side-chain and those Ru complexes containing more than two P-bound ligands. The computer program MestRe-C (Magnetic Resonance Companion, Version 1.5.0) was used to expand various spectra obtained on the AC200F N M R spectrometer. This program was especially useful for observing the F-F coupling in the ^ { T T } N M R spectra which could not be observed with the resolution of the standard HP plotter. Determination of the p,efr and number of unpaired electrons for selected paramagnetic Ru(III) complexes (in chapter 4) was performed at room temperature (r.t.) using the Evans method.2 Two resonance lines were obtained from the residual solvent protons in two solutions owing to the difference in their volume susceptibilities, with the line from the more paramagnetic solution lying at higher frequencies. Typically the solution of the complex (of known concentration) is outside and the solvent only is placed within the capillary tube; however, because of problems with overlap of peaks of the complex with those of the residual solvent this was reversed (Figure 2-1). The A8 values were determined using both the 200 and 300 M H z N M R machines and the u.eff calculated for each case. The \xe& values agreed within ± 0.1. Of note, the Evans method measurements were made for a compound dissolved in the same deuterated solvent as reported for the *H N M R unless otherwise stated. 35 References on page 46 Chapter 2 • , complex dissolved in deuterated solvent (~ 2-3 mgin I O O J I L ) ,deuterated solvent (-0.5 mL) Figure 2-1: Diagramatic representation of a capillary (containing a solution of the complex) within a N M R tube (containing solvent only) used for the Evans method. The following equations used to calculate peff were obtained from various literature sources.2"4 The equations are listed in the order in which they were used. 3A/ r = — - — i - y m = concentration of sample (g/mL) Af = frequency separation between two lines (Hz) f = frequency at which proton resonances are studied (Hz) Xo = mass susceptibility of solvent (cm3/g) X M X M 7 ' X L Xn M = compound molecular weight (g/mol) XM ' = molar susceptibility of metal ion (cm3/mol) XL = diamagnetic correction for ligand (cm3/mol) X m = diamagnetic correction for metal (cm3/mol) jueff - 2&3ylxM,T = yjn(n + 2) T= absolute temperature (K) n = number of unpaired electrons on metal ion 36 References on page 46 Chapter 2 The diamagnetic correction values for the ligands (XL) were calculated from a table of Pascal's constants5 (XL= 34, 34, 46, 74, 91 and 115 (x 10"6 cmVmol) for 2N0 2Im, 4N0 2Im, 2Me5N0 2Im, metronidazole, SR2508 and EF5, respectively). 2.2.2 Infrared Spectroscopy IR spectra were recorded on an ATI Mattson Genesis FTIR spectrophotometer using either K B r (by compression to 17,000 psi of a mixture of finely ground K B r and the compound, ~ 100:3 ratio) or a thin film on a KBr disc (from slow evaporation of a solution of a compound on the surface of a 0.5 mm KBr disc). Only the major bands in the spectra (transmittance <70%) are reported. Only select bands are assigned (stretching frequencies unless otherwise indicated) according to references from standard IR texts.6"8 Values are given in cm"1 (± 4 cm"1). 2.2.3 UV-Visible Spectroscopy UV-Vis absorption spectra were recorded on an HP 8452A diode array spectrophotometer and are given as AmaX (± 2 nm), [Smax x lO^fJvf^cm"1)], sh=shoulder. 2.2.4 Mass Spectrometry The mass spectra for selected compounds were obtained from the Mass Spectrometry (MS) lab at U B C (under the supervision of Dr. G. Eigendorf). The following spectrometers were used: KRATOS MS80 (DCI), KRATOS Concept IIHQ (LSEVIS) and KRATOS MS50 (EI). For the most part, the most valuable MS technique used for determination of the composition of the nitroimidazoles was DCI+ MS. For those nitroimidazoles which did not yield acceptable elemental analyses, a high resolution determination of the exact elemental composition of the parent peak was performed. The Ru complexes were all run using +LSLMS, but unfortunately only a small number of the Ru-nitroimidazole complexes displayed acceptable MS spectra using this technique. For the nitroimidazole complexes other MS techniques were used in addition to LSEVIS; these 37 References on page 46 Chapter 2 included M A L D I , Electrospray and EI mass spectrometries; however, none of these techniques yielded acceptable/useful M S data. 2.2.5 Gas Chromatography GC Analysis was done on an HP 5890A instrument using a 30 kPa He flow at 75 °C with an HP-17 50% phenyl, 50% methyl, polysiloxane intermediate polarity column, using a flame ionization detector and an HP 3 3 92A Integrator. 2.2.6 Cyclic Voltammetry Electrochemistry was performed on 10"3 M solutions of selected compounds with 0.1 M TBAP in dry, degassed MeCN under Ar or N 2 in the cells illustrated in Figures 2.2 and 2.3; the potential was scanned down at 100 mV/s. The potentiometer was a Pine Biopotentiostat Model AFCBP1 attached to a 80MHz computer using Pine Chem (Version 2.00) cyclic voltammetry software. The working and counter electrodes for cell A (Figure 2-2) were Pt plates (6 mm2) and the reference electrode was a Pt wire, while the working and counter electrodes for cell B (Figure 2-3) were Pt wire, and the reference electrode was Ag wire. For each sample, an initial run was performed to determine the potential range required to observe all of the expected reductions; a fresh solution of the compound and electrolyte was then prepared and the scan performed again. If necessary, further solutions were prepared and analyzed until the values were in good agreement (AEi/2 < ± 5 mV). The reduction potentials (Ei/2) were obtained from the average of the potentials of the reduction and oxidation peaks [(Ep.c.+Ep.a.)/2] and the reported E1/2 value (+ 3 mV) is the average of the combined runs. Because of the irreproducibility of the reduction potential of the ferrocene/ferrocenium couple for both of the cells, ferrocene was added as an internal standard to the solution of each compound measured. The ferrocene/ferrocenium couple in MeCN vs. the SCE is reported as 424 mV; 9 hence the conversion factor used to report the selected compound's E i / 2 vs. the SCE was 424-Ei/2(FeCp2). 38 References on page 46 Chapter 2 -* >• 15 cm Figure 2-2: Cyclic voltammetry cell (A) containing a Pt reference electrode. Pt Working Electrode 15 cm Figure 2-3: Cyclic voltammetry cell (B) containing an Ag reference electrode. 2.2.7 Conductivity Conductivity measurements were made on a 10"3 M solution of the selected complex at room temperature (r.t.) using a Serfass conductance bridge model RCM15B1 (Arthur H . Thomas Co. Ltd.) connected to a Yellow Springs Instrument Co. 3404 cell (cell constant =1.016), and are reported as A M (+ 0.5 ohm^mol^cm2). 39 References on page 46 Chapter 2 2.2.8 X-Ray Analysis All single crystal X-ray diffraction studies were performed by the late Dr. S. J. Rettig of the U B C chemistry department on a Rigaku/ADSC CCD area detector or a Rigaku AFC6S diffractometer, both of which use graphite monochromated Cu-Ka radiation. 2.2.9 Elemental Analysis Elemental analyses were performed in the U B C chemistry department by Mr. P. Borda using either a Carlo Erba (Model 1106) or a Fisons Instruments (Model E A 1108) CFfN-0 elemental analyzer. The results reported have an absolute accuracy of ± 0.3%. 2.3 Compound Purification Techniques In general, the visualization of nitroimidazoles on TLC plates was achieved with the aid of a U V lamp (Mineralight® Lamp Model UVG-54 short wave UV-254 nm), but imidazoles and other non-UV absorbing compounds were visualized by exposing the developed TLCs to I 2 or Br 2 . For each chromatographic technique used, the solvent combination, including the ratio, follows in brackets (e.g. solvent A : solvent B, 10:1) and in specific cases where the eluent strength was altered an arrow is included (e.g. 10:1 —» 5:1). 2.3.1 Column Chromatography Column chromatography was performed using Merck silica gel (230-400 mesh). The eluent was added to the silica gel in a large Erlenmeyer flask; the contents were stirred with a glass rod for 1 min and then poured into the column on a 45° angle to ensure that no air-bubbles were introduced. (If air-bubbles were present the column was tapped gently to agitate them to the surface of the silica gel.) The silica gel was packed using a pressurized air-flow adapter and then eluted to the edge of the silica gel. The compound was introduced onto the silica gel two ways: 1) by dissolving the compound in a minimum volume of the eluent and then loading onto the silica gel, making sure not to 40 References on page 46 Chapter 2 disturb the top of the gel layer; 2) by adding silica gel to a solution of the mixture to be purified, followed by removal of the solvent under vacuum and then addition of the silica gel (with the compound) to the top of the column. For this technique the solvent must be slightly above the gel layer. The solvent (containing a mixture of compounds) was eluted to the gel front, sea sand (~2 cm) was then added, and the column eluted dropwise, unless otherwise stated. The Rf values of the pure compounds were measured on Merck silica gel 60F 2 5 4 TLC A l sheets. 2.3.2 Preparative Thin Layer Chromatography Preparative TLCs of the Ru complexes were performed with Merck silica gel 60, 0.5 mm plates (without indicator) and for nitroimidazoles and other organic compounds, preparative TLCs were performed with silica gel 60 F254, 1.0 mm plates. A l l plates used for preparative T L C had dimensions of 20cm x 20cm. 2.3.3 Chromatotron Some products were purified on a Harrison Research model 7924T Chromatotron (a rotating disc TLC) using silica as the stationary phase and an eluent flow rate of ~3 mL/min. The thickness of the plate used (1.0 mm or 2.0 mm) was dependent on the amount of compound used for purification (up to 250 mg or 750 mg, respectively). This technique was extremely useful for compounds which had similar Rf values as the band separation could be observed during the elution using a U V lamp, and hence the amount of compound overlap in the collected fractions could be reduced to a minimum. 2.4 General Methodologies 2.4.1 Amide Coupling Reaction The synthesis of most of the nitroimidazole compounds was performed using a standard amide coupling technique (Scheme 2-1) 1 0 ' 1 1 in which a carboxylic acid and an amine are coupled to form an amide linkage. The N-methylmorpholine (TSMM) deprotonates the acid group, making it a better nucleophile, which attacks the acyl carbonyl group of /'so-butylchloroformate (iBuClFrm); this displaces the chlorine as CI" 41 References on page 46 Chapter 2 and gives formation of the mixed anhydride species. Addition of R-NH 2«IIX and another equivalent of N M M (to mop up HX) leads to attack at the activated carbonyl nearest the nitroimidazole ring by the amine N-atom to yield an amide bond. In the presence of any H 2 0 , the iso-butylcarbonate hydrolyzes to form CO2 and wo-butanol, which can then react with the activated anhydride; hence extremely anhydrous conditions are pertinent to the success of this reaction. R—C—N—R' I H + u HO O 1 M e R'-NH 2 *HX Scheme 2-1: Standard amide bond forming reaction. 2.5 Ruthenium Precursors 2.5.1 RuCI3«3H20 This starting material, for which the oxidation state of the metal (possibly a mixture of R u m and R u w oxidation states) and degree of hydration of the compound are ill-defined,12'13 was kindly donated from both Johnson Matthey Ltd. and Colonial Metals, Inc. The formulation of the chloride for the purpose of estimating a known number of millimoles was taken as RuCl3 ,3H 20. 2.5.2 [Ru(DMF)6][CF3S03]314 In a 2-neck flask, RuCl 3«3H 20 (2.00 g, 8.13 mmol) was dissolved in D M F (120 mL) and then Sn metal (7.00 g, 58.9 mmol) was added. The mixture was stirred at r.t. for 1 h, the colour changing from orange-brown to green to blue. To the blue solution, Pb(CF 3 S0 3 ) 2 (6.51 g, 12.9 mmol; synthesized from Pb(C0 3) and reagent grade C F 3 S 0 3 H , 42 References on page 46 Chapter 2 1:2) were added and the mixture was heated to 50 °C for 2 h (the colour changed to red-brown). The mixture was gravity filtered and then the volume was reduced to -25 mL before the addition of CH 2 C1 2 (300 mL). The mixture instantly formed a dark brown slurry which was stirred for 12 h at r.t. The brown solid was filtered off and the filtrate's volume was reduced to -30 mL. The addition of 1,2-dichloroethane (100 mL) to the solution at 0 °C led to the formation of a yellow precipitate. After 1 h at 0 °C the yellow solid was collected via suction filtration and washed with 1,2-dichloroethane (3 x 10 mL). This crude solid was recrystallized from CH2Cl2:pentane to yield a yellow microcrystalline solid (4.85 g, 56%). Anal, calc for C^H^NsO.sSjFgRu: C, 25.56; H, 4.29; N, 8.52; found: C, 25.43; H, 4.30; 8.54. UV-Vis (DMF): 254 (1.72), 273 (6.95), 342 (5.80). IR (KBr): 1642 ( C = C w DMF) (cf. 1673 for C=O F R E E D MF) . X H N M R (300 MHz, d6-dmso): 6 22.50 (br s, -Mej), 19.80 (br s, -Me2); the lower-field Me resonance has been assigned to the Me group cis to the carbonyl oxygen atom. These data agree well with those reported in the literature.15 2.5.3 [Ru(DMF)6][CF3S03]214 In a Schlenk tube, [Ru(DMF) 6][CF 3S03]3 (0.237 g, 0.240 mmol) was dissolved in 5 mL D M F (purged with N 2 for 10 min) to give a yellow solution. A trace amount (<1 mg) of Pt black (Adam's catalyst) was added and the mixture was stirred at 50 °C for 1.5 h under H 2 bubbling. The Pt black was filtered off and the red-orange filtrate's volume was reduced to -1 mL prior to the addition of n-pentanol (5 mL). Vigorous stirring for 20 min at 0 °C yielded an orange solid that was washed with E t 2 0 ( 3 x 5 mL) and dried in vacuo (0.207 g, 87 %). UV-Vis (DMF): 494 (0.182). IR (KBr): 1635 (C=O c o o r d. DMF). X H N M R (200 MHz, d6-dmso): 5 7.50 (s, -CM)) , 3.10 (s, 3H, -Me,), 2.89 (s, 3H, -Me2). These data agree well with those reported in the literature.15 2.5.4 c/s-RuCI2(DMSO)(DMSO)316 In a 50 mL flask RuCl 3»3H 20 (1.00 g, 4.06 mmol) was dissolved in 10 mL dmso, and the mixture was refluxed for 30 min in air. The reaction mixture was cooled to r.t. and addition of acetone (20 mL) led to the accumulation of a yellow precipitate. The yellow 43 References on page 46 Chapter 2 solid was collected, washed with acetone (3x10 mL) and dried in vacuo at 78 °C for 24 h (1.32 g, 84 %). Anal, calc for C 8 H 2 4 Cl 2 0 4 S4Ru: C, 19.38; H , 4.99; found: C, 19.23; H , 4.81. UV-Vis (CHCI3): 305 (0.74), 355 (1.05). IR (KBr): 1116 (S=0), 963 (S=<9). X H N M R (200 MHz, CDC13): 8 3.55, 3.51, 3.45, 3.44 (S-bonded DMSO); 2.76 (O-bonded DMSO); 2.59 (free DMSO). The spectroscopic data agree well with those previously reported.16'17 2.5.5 fra/?s-RuCI2(DMSO)418 In a 25 mL flask, RuCl 3 ' 3H 2 0 (0.500 g, 2.03 mmol) was dissolved in 3 mL dmso, and the mixture was heated to 70 °C and stirred for 15 min in air. The resulting red solution was cooled to room temperature and acetone (15 mL) was added slowly with continuous stirring. A red solid precipitated which was collected, washed with acetone (3 x 10 mL) and dried in vacuo at 78 °C for 24 h (0.525 g, 67 %). Anal, calc for C8H24Cl204S4Ru: C, 19.38; H , 4.99; found: C, 19.53; H , 5.16. UV-Vis (CHC13): 440 (0.22). IR (KBr): 1086 (S=0). lH N M R (200 MHz, CDC13): 8 3.41 (s, S-bonded CH 3 ) , 2.59 (free DMSO). The spectroscopic data agree well with those previously reported.17'18 2.5.6 cis 2(TMSO)419 In a 100 mL RBF, RuCl 3*3H 20 (2.00 g, 8.13 mmol) was refluxed in MeOH (40 mL) under H 2 for 4 h when the colour of the solution turned from brown-orange through a green intermediate to finally become dark blue. TMSO (8 mL, 89.1 mmol) was added and refluxing under H 2 was continued for an additional 4 h, generating a yellow-green precipitate. The solid was collected, washed with E t 2 0 (3x10 mL) and dried in vacuo at 78 °C for 24 h (4.02 g, 84 %). Anal, calc for Ci6H 3 2Cl 20 4S4Ru: C, 32.60; H , 5.44; found: C, 32.29; H , 5.19. UV-Vis (CHC13): 355 (1.07), 300 (0.58). IR (KBr): 1121, 1086 ( S O ) . ! H N M R (200 MHz, CDC13): 8 4.15, 3.47 (m, 2H each, -CH2-S(0)-CH2-), 2.26 (m, 4H, -CH2-CH2-). The spectroscopic data agree well with those previously reported.19 2.5.7 /ner-RuCI3(DMSO)320 In a large RBF, [(DMSO) 2H][Ru(DMSO) 2Cl 4] (1.12 g, 2.13 mmol; kindly donated by E. Cheu of this group) was suspended in acetone (80 mL), and then dmso (0.5 44 References on page 46 Chapter 2 mL, 7 mmol) was added. The mixture was refluxed with vigorous stirring. To the refluxing mixture, a AgBF 4 solution (0.400 g, 2.05 mmol, dissolved in 20 mL acetone) was added dropwise to give a bright orange solution which becomes turbid over time (due to formation of AgCl). The AgCl was filtered off (whilst the mixture was hot) and then the volume of the filtrate was reduced to ~ 10 mL before the addition of 5 mL Et 2 0. Storage at 4 °C for 12 h yielded a fine, microcrystalline red solid that was collected, washed with E t 2 0 and dried in vacuo at 78 °C for 24 h (0.297 g, 34 %). UV-Vis (CH2C12): 446 (1.28), 382 (3.54), 268 (sh) (6.43). IR (KBr): 1123, 1105 (SK)); 910 (S=0). *H N M R (300 MHz, CDC13): 6 9.45 (br s, DMSO), 2.62 (free DMSO), -15.35 (br s, DMSO). The spectroscopic data agree well with those previously reported.20 2.5.8 [RuCI2(COD)]x 21 In a Schlenk tube RuCl 3«3H 20 (2.0 g, 8.13 mmol) was dissolved in EtOH (80 mL), and 1,5-cyclooctadiene (7.5 mL, 61.0 mmol) was added under N 2 to the resulting solution. The orange mixture was heated at reflux for 3 d, at which point the solution was pale yellow after a brown product had precipitated. The brown solid was collected, washed with EtOH (3 x 10 mL) and dried in vacuo (2.01 g, 86 %). Anal, calc for [C 8Hi 2 Cl 2 Ru] x : C, 34.30; H, 4.32; found: C, 34.49; H, 4.58. 2.5.9 [RuCI2(dppb)]2(n-dppb)22'23 This complex was kindly donated by Dr. K. S. MacFarlane, formerly of this group. The physical and spectroscopic data of this precursor agree with those previously reported.2 2 , 2 3 45 References on page 46 Chapter 2 2.6 References 1 Davis, D. P.; Kirk, K . L . ; Cohen, L . A. J. Heterocyclic Chem. 1982,19, 253. 2 Evans, D. F. J. Chem. Soc. 1959, 2003. 3 Live, D. H. ; Chan, S. I. Anal. Chem. 1970, 42, 791. 4 Crawford, T. H . ; Swanson, J. J. Chem. Educ. 1971, 48, 382. 5 Figgis, B. N . ; Lewis, J. In Modern Coordination Chemistry; J. Lewis and R. G. Wilkins, Eds.; Interscience Publishers, Inc., London, 1960, p. 402. 6 Silverstein, R. M . ; Bassler, G. C ; Morrill, T. C. Spectrometric Identification of Organic Compounds; John Wiley and Sons, Inc., Toronto, 1991. 7 Nakamoto, K. ; McCarthy, P. J. Spectroscopy and Structure of Metal Chelate Compounds; John Wiley and Sons, Inc., New York, 1968. 8 Nakamoto, K. Infrared Spectra of Inorganic and Coordination Compounds, 2nd Ed.; Wiley-Interscience, Toronto, 1970. 9 Connelly, N . G ; Geiger, W. E. Chem. Rev. 1996, 96, 877. 10 Bock, M . G.; DiPardo, R. M . ; Evans, B. E.; Rittle, K. E.; Veber, D. F.; Freidinger, R. M . Tetrahedron Lett. 1987, 28, 939. 11 Holzapfel, C. W.; Pettit, G. R. J. Org. Chem. 1985, 50, 2323. 12 Seddon, E. A. ; Seddon, K. R. The Chemistry of Ruthenium; Elsevier, Amsterdam, 1984 13 Hui, B. C ; James, B. R. Can. J. Chem. 1974, 52, 348. 14 Cao, R. M.Sc. Thesis, University of Bern, 1992. 15 Judd, R. J.; Cao, R.; Biner, M . ; Armbruster, T.; Burgi, H.-B.; Merbach, A. E.; Ludi, A. Inorg. Chem. 1995, 34, 5080. 16 Evans, I. P.; Spencer, A. ; Wilkinson, G. J. Chem. Soc, Dalton Trans. 1973, 204. 17 Alessio, E.; Milani, B. ; Mestroni, G ; Calligaris, M . ; Faleschini, P.; Attia, W. M . Inorg, Chim. Acta 1990, 177, 255. 18 Jaswal, J.; Rettig, S. J.; James, B. R. Can. J. Chem. 1990, 68, 1808. 19 Yapp, D. T. T. Ph. D. Thesis, University of British Columbia, 1993. 46 Chapter 2 20 Alessio, E.; Balducci, G.; Calligaris, M . ; Costa, G.; Attia, W. M . Inorg. Chem. 1991, 30, 609. 21 Genet, J. P.; Pinel, C ; Ratovelomanana-Vidal, V. ; Mallart, S.; Pfister, X . ; Cafio De Andrade, M . C ; Laffitte, J. A. Tetrahedron: Asymmetry 1994, 5, 665. 22 Joshi, A. M . Ph. D. Thesis, University of British Columbia, 1990. 23 Bressan, M . ; Rigo, P. Inorg. Chem. 1975,14, 2286. 47 Chapter 3 Synthesis and Characterization of New 2-, 4- and 5-Nitroimidazoles with Halogenated Side-Chains 3.1 Introduction The use of nitroimidazoles provides a sensitive technique for the detection of hypoxia in human cancers; as a result of hypoxia-dependent bioreduction of nitro-imidazoles by cellular nitroreductases, they can be used to determine the variation of oxygen concentration within a tumour based on the amount of reduction product that forms adducts with cellular macromolecules.1 Recently, a monoclonal antibody (ELK3-51) was raised against the bioreduction adduct of a fluorine-containing analogue of etanidazole, named EF5 [2-(2-nitro-l-H-imidazol-l-yl)-N-(2,2,3,3,3-pentafluoropropyl) acetamide],2'3 and the amount of binding in tumour cells was determined based on the fluorescence of a fluorophore called Cy3 4 that was chemically coupled to the antibody.5 This technique has been used to assess hypoxia in vitro and in various animal tumour models.6'7 The detection and quantification of hypoxia in tumours should provide better insight into which patients would most likely benefit from hypoxia-targeted therapies. The National Cancer Institute (US) has approved clinical trials on EF5 through support of large scale synthesis, formulation and toxicology. In this chapter we report the sythesis and characterization of various halogenated nitroimidazole compounds, including their one-electron reduction potentials. The major goal was to expand this area of chemistry in hopes of developing a nitroimidazole compound that was more biologically active than EF5. The preliminary studies for the development of a new PET imaging agent are also reported. 48 References on page 129 Chapter 3 3.2 Experimental Section 3.2.1 Synthesis of Side-Chains 3.2.1.1 2-Iodo-7V-(2,2,3,3,3-pentaf luoropropyl)acetamide [EF5] Iodoacetic acid (0.224 g, 1.20 mmol) was dissolved in THF (15 mL) at 0°C (ice-bath) under a constant N 2-flow. N M M (140 p.L, 1.28 mmol) was added, and the mixture was stirred for 10 min after which iBuClFrm (195 uL, 1.50 mmol) was added. The mixture was further stirred for 30 min during which a slurry was formed. The ice-bath was removed, and H 2NCH 2CF 2CF 3«HC1 (0.249 g, 1.45 mmol) and N M M (165 uL, 1.51 mmol) were added. Stirring was continued for 20 h at r.t. after which the resulting yellow-orange precipitate was filtered off and washed with dry THF. The dark orange washings were combined and eluted through a silica gel column (CH 2 Cl 2 :MeOH, 10:1); a purple band (I2) appeared, followed by a yellow band that was analyzed using TLC and found to contain two products. The yellow band was eluted off and, after removal of the solvents, the residue was placed in a sublimation apparatus and heated to 110 °C under vacuum, when pure I F 5 sublimed as a pale yellow solid (0.202 g, 53%); mp 90-93 °C. Anal. calc. for C 5 H 5 NOF 5 I : C, 18.94; H , 1.59; N , 4.42; found: C, 18.76; H , 1.67; N , 4.61. IR (v, cm"1): 3305 (N-H); 3088 (C-Hm); 2963, 2858 (C-H); 1658 (C=0); 1204, 1185, 1141, 1067, 1036. XH N M R (300 MHz, d6-acetone): 5 8.06 (s, IH, N-/ / ) , 4.05 (td, 2H, 3JHF = 6.5 Hz, 3JHH = 0.9 Hz, -Cr7 2-CF 2-), 3.81 (s, 2H, l-CH2-). 19VCH} N M R (188 MHz, de-acetone): 8 -8.18 (t, -CF 3 ) , -45.34 (q, -CF2-). H O IF5 49 References on page 129 Chapter 3 3.2.1.2 2-Iodo-A /-(3-bromopropyl)acetamide [IBr] A similar procedure to that for the synthesis of EF5 was used (see above). Iodoacetic acid (0.979 g, 5.27 mmol) and N M M (576 uL, 5.27 mmol) were dissolved in THF (20 mL), and iBuClFrm (752 uL, 5.78 mmol) was added. After 40 min, H 2 N C H 2 C H 2 C H 2 B r ' H B r (1.26 g, 5.79 mmol) and N M M (639 uL, 5.85 mmol) were then added, the mixture giving a yellow slurry. Stirring was continued for 4 h at r.t. during which the colour changed to a red/brown and a yellow precipitate was formed. This was filtered off and washed with THF (3 x 15 mL). The washings were combined and the solvent was removed by rotoevaporation yielding a dark red oil; subsequent TLC analysis revealed several products. Purification of the desired product was achieved via column chromatography (CH 2 Cl 2 :MeOH, 20:1). The first band (yellow, low concentration) was followed by a minor red band after which the product band (bright orange) emerged; T L C analysis of the orange band revealed a single species. Subsequent removal of solvent afforded an orange oil (0.978 g, 61 %). LR-MS [DCI(+)]: 306 (TVf), 255 (I2), 227 (M" - Br), 180 Qs/T - I), 169 (TVf - NHCH 2 CH 2 CH 2 Br) , 128 (I). Tf N M R (300 MHz, de-acetone): 5 7.90 (s, IH, N-//) , 3.77 (s, 2H, I-CrY2-), 3.52 (t, 2H, -Gr72-Br), 3.36 (dt, 2H, -NH-C# 2-), 2.07 (m, 2H, -NH-CH 2-Cr7 2-, partially overlapped with residual acetone peak). (These chemical shifts are comparable to those seen for compounds of similar composition, e.g. IF5 and E B r l (Section 3.2.2.14)). Unlike with IF5, purification of IBr by sublimation (at 70°C) was unsuccessful due to thermal instability as evidenced by N M R spectroscopy (see Discussion, p. 98). Of note, mass spectrometric analysis of IBr after such a sublimation procedure revealed the presence of species attributable to compounds containing cyclic rings, such as 1 and 2 of the same molecular weight (see Scheme 3-4, section 3.3.1, p. 98). LR-MS [DCI(+)]: 226 fJVf), 128 (I), 99 (TVf - I). T i NMR, including 2D-cosy (200 MHz, d6-acetone): (1) 5 4.81 (t, 2H, 0-CH2-), 4.12 (s, IH, l-CH=), 3.74 (t, 2H, -NH-CfY 2-), 2.51 (s, IH, -N#-), 2.37 (p, 2H, -CH 2 -Cr7 2 -CH 2 -); (2) 5 4.82 (s, 2H, I-C#2-), 3.55 (t, 2H, -N-CH2-), 3.32 (t, 2H, -N-Cr72-), 2.03 (m, 2H, - C H 2 -C/ / 2 -CH 2 - ) . Of note, partial formation of 1 and 2 was observed when neat IBr was exposed to U V radiation (TLC lamp) at r.t. for 2 d. 50 References on page 129 Chapter 3 IBr 1 2 3.2.1.3 2-Chloro-iV-(2,2,3,3,3-pentafluoropropyl)acetamide [C1F5] Chloroacetic acid (0.101 g, 1.08 mmol) was dissolved in THF (10 mL) under 1 atm N 2 . N M M (120 uL, 1.08 mmol) was then added, and the reaction mixture was stirred at 0 °C for 10 min. Afterwards, iBuClFrm (151 pL, 1.16 mmol) was added and a white precipitate was formed instantaneously. Stirring was continued for 30 min after which the ice-bath was removed, and H 2NCH 2CF 2CF 3»HC1 (0.189 g, 1.10 mmol) and N M M (128 pL, 1.17 mmol) were added. The final mixture was stirred at r.t. for 8 h before the white solid was collected and washed with THF. The solvent was subsequently removed giving a pale yellow oil whose TLC analysis revealed two products, one visible under U V light and the other appearing after development using LVhexanes. The mixture was sublimed at 60 °C under vacuum, and a crystalline white solid was obtained (the remaining non-sublimed product was a dark yellow oil). TLC analysis revealed a single species, not visible under U V light (0.148 g, 61 %); mp 44-45 °C. LR-MS [DCI(+)]: 225 (M+), 190 (M+ -CI), 176 (TVf - CH2C1). HR-MS [DCI(+)] calc. for C 5 H 5 N0 3 5 C1F 5 : 224.99799; found 224.99782. JR (v, cm"1): 3324 (N-H); 3106 ( C - H M ) ; 2966, 2872 (C-H); 1653 (C=0); 1206, 1180, 1146, 1067, 1037. : H N M R (300 MHz, d6-acetone): 8 8.12 (s, IH, N-H), 4.24 (s, 2H, Cl-Gr72-), 4.17 (td, 2H, 3 J H F = 6.7 Hz, 3 J H H = 2.2 Hz, -C/f 2 -CF 2 -) . 1 9 F{ J H} N M R (188 MHz, d6-acetone): S -8.21 (t, -CF3), -45.26 (q, -CF 2 -). H C1F5 51 References on page 129 Chapter 3 3.2.1.4 Reduction of IF5 to ICH2CH2NHCH2CF2CF3 (3) IF5 (0.040 g, 0.126 mmol) was placed in a 25 mL two-neck flask and THF (5 mL) was added. To the resulting pale yellow solution was added BH 3 *THF (1.0 M , 290 uL, 0.290 mmol), when the reaction mixture turned bright yellow within 1 min. After 15 min the colour reverted to pale yellow; additional BH3«THF (50 uL) was added and the solution became colourless. A condenser was attached and the solution was refluxed at 80 °C for 1 h to ensure completion of the reaction. The reaction mixture was then cooled to r.t. and MeOH (10 mL) was added to decompose unreacted B H 3 . The solvent was then removed to give a yellow oil; TLC revealed the presence of one major and several minor products. The major product was purified via preparative TLC (CH 2 Cl 2 :MeOH, 50:1) as a yellow oil, 3 (0.028 g, 73 %). Note: the product spot on the TLC plate, on exposure to U V light, turned yellow, a phenomenon not seen for 1F5 T i N M R (300 MHz, dg-acetone): 5 7.75 (br s, IH, -NH-), 4.02 (td, 2H, 3 J H F = 10 Hz, 3 J H H = 2 H Z , -Cr7 2-CF 2), 3.59 (t, 2H, l-CHr-), 3.31 (dt, 2H, -C772-NH-). ^ { T T } N M R (188 MHz, d6-acetone): 5 -8.13 (t, -CF 3 ) , -45.27 (q, -CF2-). After exposure to U V light the solution became dark red/brown and a new species (4) formed. *H N M R (300 MHz, d6-acetone): 5 4.02 (t, 2H, 3 J H F = 10 Hz, 3 J H H = 1.6 Hz, -Cr7 2-CF 2), 3.62 (m, 2H, -C7feq-), 2.05 (m, 2H, -C/7 a x-). 19¥{lH} N M R (188 MHz, d6-acetone): 5 -8.16 (t, -CF 3 ) , -45.19 (q, -CF 2 -) . Two of the minor bands, one yellow and one red, were determined by UV-Vis spectophotometric analysis to be I3" and I 2, respectively.8 UV-Vis (CH2C12): [band 2] 292, 362; [band 3] 502. 3.2.1.5 3-Fluoropropylamine Hydrochloride (5) The parent amine for this compound was previously synthesized by Pattison et al. from 3-fluoropropionic acid (see Scheme 3-5, p. 99),9 but the yield was low. Here other possible routes to the HC1 salt of 3-fluoropropylamine were investigated.10 4 52 References on page 129 Chapter 3 Synthesis 1 (Scheme 3-6, p. 100) The use of dialkylaminosulfur fluorides as effective fluorinating agents has been reported previously; the hydroxyl group in alcohols can be replaced with fluorine atoms, and high yields are obtained for the non-rearranged fluoro compound.11'12 In a 25 mL two-neck flask at 0°C under a constant flow of N 2 was added DAST (1 mL) to 5 mL CH 2 C1 2 to give a pale yellow solution. 3-Hydroxypropylamine (500 uL) was then added dropwise, when the mixture turned orange and then quickly brown. The mixture was stirred for 1 h at 0 °C, and then frozen using liquid N 2 . The volatile components were then vacuum transferred to another flask as a pale yellow solution through which HCl(g) was bubbled for 5 min. A white solid was formed, and E t 2 0 (10 mL) was added to complete the precipitation before the solid was isolated via suction filtration. ( N B . This compound is extremely hygroscopic and is best handled in an inert atmosphere.) X H N M R spectroscopic analysis of the solid revealed that 5 was formed, but only as a minor product. The major product was Et2NH«HCl and according to spectral integration was present in a six-fold excess over 5. X H N M R (300 MHz, d6-dmso): (5) 8 8.09 (br s, 2H, -NH-), 4.52 (dt, 2H, 2 J H F = 44 Hz, 3 J H H = 4.8 Hz, -C# 2F), 2.91 (peak overlaps with H 2 0 peak, -NH-Cr7 2-), 2.00 (dp (dtt), 2H, 3 J H F = 20 Hz, 3 J H H = 4.7 Hz, -C# 2-CH 2F); (Et 2 NH-HCl) 8 8.80 (br s, IH, -NH), 2.91 (m, 2H, NH-C# 2), 1.20 (t, 3H, C/f 3 -CH 2 -) . 1 9 F N M R (282 M H z , d6-dmso): (5) 8 -144.94 (spt (tt), 2 J H F = 47.4 Hz, 3 J H F = 24.0 Hz, -CH 2 F). Separation of these two products was extremely difficult and was not further pursued. Synthesis 2 (Scheme 3-7, p. 101) This procedure was adapted from that reported by Gibson and Bradshaw on the Gabriel synthesis of primary amines.13 In a Schlenk tube under 1 atm N 2 was added 10 mL D M F to potassium phthalamide (0.500 g, 2.70 mmol) to produce a white slurry. 1-Bromo-3-fluoropropane (225 | iL , 2.45 mmol) was then added, and the reaction mixture was stirred at 70 °C for 24 h. Afterwards, the mixture was filtered and the filtrate was chromatographed using preparative TLC (CH2C12:acetone, 20:1); the major band (Rf = 0.85) was isolated as a colourless oil which when cooled yielded a white crystalline solid 53 References on page 129 Chapter 3 (0.334 g, 66 %), N-3-fluoropropylphthalamide (6). LR-MS [DCI(+)]: 207 (M*), 187 (TVT - F), 160 (IVF - CH 2 CH 2 F) . HR-MS [DCI(+)]: calc. for C n H 1 0 N Q 2 F 207.06955; found 207.06938. *H N M R (300 MHz, de-acetone): 5 7.85 (s, 4H, -CHBz-), 4.56 (dt, 2H, 2 J H F = 51 Hz, 3 J H H = 6.2 Hz, -C# 2F), 3.80 (t, 2H, 3 J H H = 6.8 Hz, -N-CH2-\ 2.10 (dp (dtt), 2H, 3 J H F = 27 Hz, 3 J H H = 5.6 Hz, -Gr7 2-CH 2F). 1 9 F N M R (282 MHz, de-acetone): -142.08 (spt (tt), 2 J H F = 49.4 Hz, 3 J H F = 24.0 Hz, -CH 2 F). The hydrazinolysis of 6 was adapted from a procedure described by Sheehan and Ryan. 1 4 6 (0.120 g, 0.579 mmol) was suspended in 10 mL EtOH and the mixture was heated to 50 °C to facilitate dissolution. Hydrazine monohydrate (40 uL) was then added and the mixture was refluxed for 2.5 h and then was stirred for 16 h at r.t.; the initially colourless solution became a white slurry. The volume was reduced to ~5 mL, 6 M HC1 (10 mL) added, and the mixture stirred at 80 °C for 1 h. The resulting white precipitate was filtered and extracted with warm H 2 0 (3 x 10 mL). The solvent volume was then reduced resulting in the re-formation of the white precipitate which was filtered off. The rotoevaporation/filtration procedure was then repeated to obtain additional product. The final filtrate was taken to dryness to recover the remaining product (0.060 g, 91 %). The *H N M R spectrum revealed only the presence of 5. Analysis calc. for C 3H 9NC1F: C, 31.72; H , 7.93; N , 12.33; found C, 31.56; H , 8.20; N , 11.83.15 LR-MS [DCI(+)]: 78 (M+), 59 -F), 30 (H 2 NCH 2 -) . HR-MS [DCI(+)]: calc. for C3H9NF 78.07190; found 78.07158. 3.2.2 2-Nitroimidazole Compounds 3.2.2.1 2-(2-Nitro-l-H-imidazol-l-yl)acetic acid (7) (see Scheme 3-9, p. 102) In a 250 mL RBF, SR2508 (3.70 g, 0.0173 moi) was dissolved in MeOH (100 mL) to give a clear, colourless solution. Dowex 50-X8(Fr) (40.3g, previously washed with equiv. amounts (50 mL) of 0.1M HC1, dd H 2 0 and MeOH) was then added followed by an additional 50 mL MeOH. The mixture was refluxed for 24 h when amide bond cleavage occurs to form the intermediate methyl ester, 8, as shown by TLC analysis; the reaction mixture was then filtered and the filtrate was rotary evaporated to dryness 54 References on page 129 Chapter 3 yielding an orange oil (the accompanying orange colour comes from the cation exchange resin and, if the methyl ester is purified via column chromatography at this stage, 8 can be isolated as a white solid). The oil was dissolved in 100 mL 0.1M NaOH and the mixture stirred at 100 °C for 45 min to hydrolyze the ester (monitored by TLC). The mixture was cooled to r.t. and 2 M HC1 added dropwise until the pH was ~ 4. Saturated NaCl solution (50 mL) was added and the product was extracted with EtOAc (10 x 100 mL). The organic extracts were combined and dried with MgS0 4 ; the mixture was filtered and the solvent removed to give a pale yellow solid (0.805 g, 27%). UV-Vis (MeOH): (7) 316 (7.7); (8) 316 (7.7). X H N M R (200 MHz, d6-dmso): (7) 8 13.42 (s, IH, -OH), 7.63 (s, IH, lm-H5), 7.22 (s, IH, lm-H4), 5.20 (s, 2H, -CH2-); (8) 8 7.64 (s, IH, lm-H5), 7.22 (s, IH, Im-H4), 5.33 (s, 2H, -CH2-), 3.72 (s, 3H, -CH3). Data are in agreement with those previously reported.16 OH N 0 2 o OH I H NO, NO, OMe SR2508 7 8 The aqueous layers were also combined and rotovapped to dryness. The resulting yellow solid was then dissolved in MeOH (40 mL) and the solution was filtered; removal of the solvent from the filtrate gave a bright yellow solid (1.35 g). T L C analysis (shown below) showed the presence of large amounts of 7 that was not extracted with EtOAc. Eluent (20: L CH 2 Cl 2 :MeOH) (E = SR2508) Figure 3-1: TLC analysis of products obtained from 2-(2-Nitro-1 -H-imidazol-1 • yl)acetic acid (7) synthesis. 55 References on page 129 Chapter 3 5.2.2.2 2-(2-Nitro-l-H-imidazol-l-yl)-N-(2,2,3,3,3-pentafluoropropyl) acetamide [EF5] Synthesis 1 (cf. Scheme 3-8, p. 102) 7 (0.200 g, 1.16 mmol) was dissolved in THF (25 mL) at 0°C under N 2 . N M M (130 uL, 1.18 mmol) was then added and the colourless mixture was stirred for 10 min. iBuClFrm (167 pL, 1.28 mmol) was added and the mixture was stirred for 30 min during which a white slurry was formed. The ice-bath was removed, and H 2NCH 2CF 2CF 3«HC1 (0.241 g, 1.41 mmol) and N M M (142 uL, 1.30 mmol) were added. The reaction slurry was stirred at r.t. for 4.5 h and the resulting white precipitate was filtered off and washed with THF (3x5 mL). The yellow filtrate was rotovapped to dryness yielding a pale yellow oil that was purified via column chromatography (CH2C12:acetone, 10:1 —» 5:2). The major band (colourless) followed a minor yellow band and the combined fractions were rotovapped to dryness to yield EF5. Recrystallization from EtOAc/hexanes yielded a white microcrystalline solid (0.116 g, 33%). Synthesis 2 1 7 2N0 2 Im (0.0257 g, 0.227 mmol) was dissolved in DMF (10 mL) at r.t. under N 2 , and powdered C s 2 C 0 3 (0.0720 g, 0.221 mmol) was added to give a white slurry. The mixture was heated to 50 °C when a pale yellow solution resulted; after 30 min a white precipitate was formed. The slurry was stirred for 2 h and IF5 (0.0724 g, 0.229 mmol) was then added, and the mixture stirred for 3 h at 50 °C. The resulting pale yellow precipitate was filtered off and washed with D M F (3x5 mL); the filtrate was then worked up as described in Synthesis 1 (0.0535 g, 78%); mp 136-137 °C. X-ray quality crystals were obtained by slow evaporation of a concentrated MeOH solution of EF5. Anal. calc. for C 8 H 7 N40 3 F 5 : C, 31.80; H , 2.34; N , 18.54; found: C, 31.96; H , 2.24; N , 18.37. L R - M S [DCI(+)]: 303 O v T ) , 256 (M+ -N0 2 ) . IR ( v , cm"1): 3317 (N-H); 3086 (C-H^); 2921, 2850 (C-H); 1689 (C=0); 1490 (N-O a s y m ); 1367 (N-O s y m). UV-Vis (H 2 0): 323 (6.8). TI N M R (obtained in riVdmso and in CD 3 OD to show the effect that H/D exchange of the N - H proton has on the -Gr7 2-CF 2- signal): (300 MHz, CD 3 OD) 5 7.47 (s, IH, lm-H5), 7.20 (s, 56 References on page 129 Chapter 3 1H, Im-H4\ 5.25 (s, 2H, -Gr72-CO-), 4.01 (t, 2H, 3 J H F = 15 Hz, -Gr7 2-CF 2-); (200 MHz, d6-dmso) 6 9.05 (t, IH, N-#), 7.66 (s, IH, Im-H5), 7.21 (s, IH, lm-H4), 5.24 (s, 2H, -CH2-CO-), 4.06 (td, 2H, 3 J H F = 15 Hz, 3 J H H = 1.5 Hz, -C# 2-CF 2-). 1 9F{TI} N M R (188 MHz, CD3OD): 5 -10.66 (t, -CF 3 ) , -47.24 (q, -CF2-); (188 MHz, d6-dmso) 6 -4.34 (t, -CF 3 ) , -41.31 ( q , - C i v ) . EF5 3.2.2.3 2-(Imidazol-l-yl)-N-(2,2,3,3,3-pentafluoropropyl)acetamide [ImF5] Sodium imidazolate (0.0312 g, 0.344 mmol) was added in D M F (5 mL) at r.t. under N 2 to give a yellow slurry. TF5 (0.117 g, 0.366 mmol) was then added and the mixture was stirred for 4 h at 70 °C. The volume was reduced under vacuum to yield a yellow oil that was subsequently loaded onto a preparative TLC plate (CH 2 Cl 2 :MeOH, 20:1). (Two major bands with similar Rf values were visible only after development using I2.) The isolated brown solid from the two bands was heated to 120 °C under vacuum in a sublimation apparatus when a yellow oil deposited onto a cold finger maintained at 5 °C (0.0365 g, 41%). Of note, exposure of the oil to air caused a colour change to dark purple while re-dissolution and exposure to air in acetone generated a bright yellow colour; removal of the acetone regenerated the dark purple colour. LR-MS [EI]: 257 OvT), 81 OVf - CONHCH 2 CF 2 CF 3 ) . HR-MS [EI] calc. for C 8 H 8 N 3 O F 5 : 257.05875; found 257.05903. TI N M R (300 MHz, d6-acetone): 5 7.96 (br. s, IH, N-#), 7.65 (s, IH, Im-H2), 7.16 (s, IH, Im-H5), 6.97 (s, IH, lm-H4), 4.92 (s, 2H, -Cr72-CO-), 4.10 (td, 2H, 3 J H F = 17 Hz, 3 J H H = 6.4 Hz, -Gr7 2-CF 2-). 1 9F{TI} N M R (188 MHz, d6-acetone): 8 -8.16 (t , -CF 3 ) , -44.87 (q,-CF 2-). H '2 57 References on page 129 Chapter 3 H O I m F 5 3.2.2.4 N-(2-nitro-l-H-imidazol-l-ethyl) pentafluoropropionamide [RevEF5] Synthesis 1 (see Scheme 3-14, p. 107) Step 1 In a 250 mL 2-neck flask, 2N0 2Im (0.870 g, 7.70 mmol) was dissolved in 50 mL D M F under a constant flow of N 2 . C s 2 C 0 3 (2.75 g, 8.47 mmol) was then added and the resulting yellow slurry was stirred at 90 °C for 3 h when dissolution was completed. Bromoethylphthalamide (3.96 g, 15.6 mmol) was added and the resulting yellow slurry was stirred for 15 h during which a white solid appeared. The solid was filtered off and the filtrate was reduced in volume to 2 mL before MeOH (20 mL) was added to precipitate a white solid (9). The solid was filtered, washed with hot MeOH (3 x 25 mL) and dried in vacuo at 80 °C for 3 d (1.95 g, 99 %). LR-MS [DCI(+)]: 287 fJVT + H), 240 (TVf - N 0 2 ) , 160 (Wt - 2N0 2 ImCH 2 ). HR-MS [DCI(+)] calc. for C13H11N4O4: 287.07803; found 287.07810. TI N M R (300 MHz, d6-acetone): 5 7.78 & 7.73 (overlapping multiplets, 4H, Bz-H), 7.01 (s, IH, Im-H5), 6.89 (s, IH, Im-H4\ 4.70 (t, 2H, -C#2-Phth), 4.18 (t, 2H, -C# 2-CH 2-Phth). Step 2 In a 25 mL RBF, 9 (0.0705 g, 0.273 mmol) was suspended in 10 mL EtOH. H 2 NNH 2 »H 2 0 (35 uL, 0.721 mmol) was then added via syringe and the mixture was stirred at 80 °C for 18 h during which a clear, colourless solution was formed; T L C analysis revealed that 9 had completely reacted. 6 M HC1 (10 mL) was added to the reaction mixture and the EtOH was then removed to yield a white slurry. The solid [LR-MS DCI(+): 162 (M*), 132 (IVf - N 2 H 2 ) , 104 (TVT - CO - N 2 H 2 ) , 76 (M+ - 2CO - N 2 H 2 ) .] was isolated and washed with copious amounts of H 2 0 . The solvent was removed from the pale yellow filtrate to yield an extremely hygroscopic, cream coloured 58 References on page 129 Chapter 3 solid (10) (0.0565 g, 90 %). LR-MS [DCI(+)]: 157 ( M + + H), 110 (IVf - N 0 2 ) . [Only the parent cation is observed with DCI (+), so it is unclear whether 1 or 2 moi of HC1 is associated with the compound.] ! H N M R (300 MHz, D 2 0) : 6 7.48 (d, IH, lm-H5), 7.18 (d, IH, lm-H4), 4.75 (t, 2H, Im-Cr72-), 3.51 (t, 2H, -C# 2-NH 2). Step 3a In a 3-neck 50 mL flask, DCC (0.182 g, 0.880 mmol) and NHS (0.101 g, 0.880 mmol) were dissolved in 4 mL DMF under N 2 to give a clear, colourless solution. C F 3 C F 2 C 0 2 H (92 pL, 0.870 mmol) was added and a white microcrystalline precipitate was immediately formed. The mixture was stirred for 1 h at r.t. before a D M F solution (3 mL) of 10 (0.206 g, 0.900 mmol) and N M M (198 uL, 1.80 mmol), was added. Stirring was continued for 4 h after which TLC analysis revealed that 10 had completely reacted. The white precipitate [DCU according to LR-MS [DCI(+)]: 225 (TVf), 143 (Tyf - C 6 H 5 ) , 99 (CeH5-NH-)] was isolated and washed with DMF; the filtrate was reduced in volume to ~ 1 mL before being loaded onto a preparative TLC plate (CH 2 Cl 2 :MeOH, 18:1). The product band (Rf = 0.45) was isolated to yield RevEF5 as an off-white solid (0.0757 g, 29 %). Step 3b , This reaction was continued from Step 2 prior to the addition of HC1 (same amount of 9 used for this synthesis). The colourless EtOH solution was taken to dryness to yield an oily, white film which was subsequently taken up in C H C I 3 (10 mL) to give a white slurry. N M M (1 equiv.) was added and the mixture stirred for 30 min at r.t. Upon addition of (CF 3 CF 2 CO) 2 0 (1 equiv.) the mixture became a clear solution which slowly deposited a white residue on the walls of the flask over 5 h. Addition of EtOH led to the formation of a white precipitate which was filtered off and washed with EtOH ( 3 x 5 mL). TLC analysis of the filtrate revealed two major products which were separated using preparative TLC (CH 2 Cl 2 :MeOH, 20:1). The band corresponding to RevEF5 was identified by comparison of the Rf value with that of an authentic sample of the compound. The isolated RevEF5 was reprecipitated from an EtOAc/hexanes mixture to yield a white solid (0.0265 g, 32%). 59 References on page 129 Chapter 3 H N, Y N, .N. T CF3 N 0 2 Y O N 0 2 O 9 10 RevEF5 Synthesis 2 Reaction conditions similar to those used for synthesis 1 of EF5 (section 3.2.2.2) were used: C F 3 C F 2 C 0 2 H (25 uL, 0.238 mmol) was dissolved in THF (10 mL) after which N M M (25 uL, 0.238 mmol) and iBuClFrm (36 uL, 0.276 mmol) were added giving a white slurry. 10 (0.0605 g, 0.262 mmol) and N M M (60 pL, 0.571 mmol) were then added and the mixture was stirred at r.t. for 4 h. The white precipitate that formed was filtered off and washed with THF ( 3 x 5 mL); the crude product (filtrate) was then purified using preparative TLC (CH 2 Cl 2 :MeOH, 20:1). Four bands were observed and the minor band with R f = 0.37 was isolated to yield RevEF5 (0.0115 g, 16 %). Anal. calc. for C8H7N4O3F5: C, 31.80; H, 2.34; N , 18.54; found: C, 31.85; H , 2.33; N , 18.59. L R - M S [DCI(+)]: 320 (Tvf + N H / ) , 303 (M* + H), 287 (Vf - O), 273 (Jvf - NO), 256 (JVf - N0 2 ) . HR-MS [DCI(+)] calc. for C 8 H 8 N 4 0 3 F 5 : 303.05164; found 303.05137. IR (v, cm-1): 3327 (N-H); 2965, 2932, 2866 (C-H); 1613 (C=0); 1457 (N-O a s y m ) ; 1383 (N-O s y m ). UV-Vis (MeOH): 316 (6.2). J H N M R (300 MHz, d6-acetone): 5 8.89 (br s, IH, -NH-), 7.41 (s, IH, 1m-Hs), 7.05 (s, IH, Im-H4), 4.68 (t, 2H, Im-Cr72-), 3.86 (q (dt), 2H, -C// 2 , -NH-) . 1 9 F{ J H} N M R (188 MHz, d6-acetone): 5 -6.79 (t, -CF 3 ) , -46.43 (q, -CF 2 -) . Synthesis 3 Based on published peptide-coupling techniques,18'19 C F 3 C F 2 C 0 2 H (30 pX, 0.286 mmol) and E t 3 N (1 equiv.) were dissolved in DMF (2 mL) and STMT>BF 4 (0.0941 g, 0.313 mmol), a reagent used to produce the -NHS esters was then added. The clear mixture was then stirred for 18 h at r.t. after which TLC analysis (CH 2 Cl 2 :MeOH, 20:1) indicated the presence of a new product in high yield, presumed to be the activated succinimidyl ester. 7 (0.0652 g, 0.288 mmol) and E t 3 N (2 equiv.) were subsequently added, and the mixture, now cloudy, was stirred for 24 h when TLC analysis indicated the 60 References on page 129 Chapter 3 presence of two major products. The solvent was removed under reduced pressure, and the residue was chromatographed using preparative TLC (CH 2 Cl 2 :MeOH, 25:1) to give two pale yellow oils; comparison of TLC data with those of an authentic sample of RevEF5 revealed that neither oil contained the desired product. A minor band with the same Rf as RevEF5 was observed but the product was not isolated due to low concentrations (<1%). 3.2.2.5 2-(2-Nitro-l-H-imidazoI-l-yI)-N-(3-bromo-2,2,3,3-tetrafluoropropyI) acetamide [EF4Br] Reaction conditions similar to those used for the synthesis of EF5 (synthesis 1) were employed: 7 (0.0760 g, 0.442 mmol) was dissolved in THF (15 mL) and N M M (50 pL, 0.458 mmol) and iBuClFrm (63 pL, 0.483 mmol) were added. To the resulting pale yellow slurry, H 2NCH 2CF 2CF 2Br«HCl (0.120 g, 0.487 mmol) and N M M (53 pL, 0.485 mmol) were added, and the mixture was then stirred at r.t. for 5 h. The white precipitate that formed was filtered off and washed with THF ( 3 x 5 mL). The filtrate was taken to dryness to give a yellow oil that was purified via preparative TLC (CH 2 Cl 2 :MeOH, 20:1). The isolated product band appeared to have two species present (with almost identical Rf values) and so a second preparative TLC was performed (CH 2 Cl 2 :MeOH, 12:1). The major band separated from the minor band; however, when isolated and eluted with acetone the eluate was taken to dryness to give back the original oil. The oil was dissolved in 1 mL EtOAc followed by addition of hexanes (15 mL) which resulted in formation of a white precipitate. The fine white solid was isolated via suction filtration, washed with E t 2 0 ( 3 x 5 mL) and dried in vacuo (0.0425 g, 27%). Anal. calc. for CgH7N4O3F4Br»0.15 E t 2 0 : C, 27.61; H , 2.29; N , 14.97; found C, 27.76; H , 2.13; N , 14.94. LR-MS [DCI(+)]: 365, 363 (Wt+ H), 335, 333 (Wt-NO), 318, 316 (M^ - N0 2 ) , 283 (Wt- Br). HR-MS [DCI(+)] calc. for CgHgOsNdVTir (CgHg03N4F47 9Br): 364.96954 (362.97163); found 364.96834 (362.97100). IR (v, cm - 1): 3310 (N-H); 3076 (C-H^); 2960, 2851 (C-H); 1697 (C=0); 1493 (N-O a s y m ) ; 1370 (N-O s y m). UV-Vis (MeOH): 316 (6.9), 232 (3.5). TI N M R (300 MHz, de-acetone): 8 8.18 (br s, IH, -NH-), 7.49 (s, IH, lm-Hs), 7.20 (s, IH, lm-H4), 5.31 61 References on page 129 Chapter 3 (s, 2H, -C#2-C0-), 4.10 (td, 2H, 3 J H - F = 15.6 Hz, -NH-C/^ , -CF 2 - ) . ^Ff/H} N M R (188 MHz, de-acetone): 5 10.04 (t, -CF 2 Br), -38.75 (t, -CU2-CF2-). 3.2.2.6 2-(2-Nitro-l-H-imidazol-l-yl)-N-(3,3,3-trinuoropropyI)acetamide [EF3] Compound 7 (0.150 g, 0.873 mmol) was dissolved in THF (20 mL) at 0°C under N 2 . N M M (96 pL, 0.873 mmol) was then added and the clear, colourless mixture was stirred for 10 min. iBuClFrm (125 uL, 0.961 mmol) was added and the solution was stirred for 30 min during which the mixture became a white slurry. The ice-bath was removed and H 2 NCH 2 CH 2 CF 3 *HC1 (0.199 g, 0.961 mmol) and N M M (105 uL, 0.961 mmol) were added. Stirring was continued for 5.5 h at r.t. before the white precipitate that formed was filtered off and washed with THF ( 3 x 5 mL); the yellow filtrate was taken to dryness to yield a yellow solid that was subsequently recrystallized from EtOAc/hexanes to give white microcrystalline needles (0.193 g, 83 %). Anal. calc. for C8H9N4O3F3: C, 36.10; H , 3.41; N , 21.05; found: C, 36.35; H , 3.35; N , 21.05. IR (v, cm"1): 3312 (N-H), 3026, 2923, 2851 (C-H), 1669 (C=0), 1493 (N-O a 9 y r a), 1370 (N-O s y m ). UV-Vis (dmso): 328 (9.4); (MeCN) 322 (4940). *H N M R (300 MHz, d6-acetone): 5 7.81 (br s, IH, -NH-), 7.50 (s, IH, Im-H5), 7.15 (s, IH, Im-H4), 5.25 (s, 2H, -CH2-CO-), 3.50 (q, 2H, -NH-Cr7 2-), 2.50 (qt, 2H, 3 J H - F = 11 H Z , -CH2CF3), ' ^ { ' H ) N M R (188 MHz, de-acetone): 5 10.84 (s, -CF3). H H •N, NO, 62 References on page 129 Chapter 3 3.2.2.7 2-(2-Nitro-l-H-imidazol-l-yl)-N-(2,2,2-trifluoroethyl)acetamide [EF3(-1)] Reaction conditions similar to those used for the synthesis of EF3 were employed: 7 (0.172 g, 0.998 mmol) was dissolved in THF (17 mL) and N M M (109 uL, 0.996 mmol) and iBuClFrm (142 pL, 1.09 mmol) were added. To the resulting pale yellow slurry, H 2NCH 2CF 3«HC1 (0.149 g, 1.10 mmol) and N M M (120 uL, 1.10 mmol) were added, and the mixture was then stirred at r.t. for 4 h. The white precipitate that formed was filtered off and washed with Et 20 (3x5 mL); the filtrate was taken to dryness to give a yellow oil that was purified using a C T R O N (THF:acetone, 10:1). The fourth, most concentrated band was isolated to yield a white solid (0.208 g, 83%). Anal. calc. for C 7 H 7 N 4 O 3 F 3 ' 0 .15 acetone: C, 34.40; H , 3.08; N , 21.39; found C, 34.25; H , 2.86; N , 21.31. IR (v, cm"1): 3299 (N-H); 3117 (C-H^); 2923, 2851 (C-H); 1685 (C=0); 1489 (N-O a s y m ) ; 1374 (N-O s y m). UV-Vis (MeOH): 316 (8.5). TI N M R (300 MHz, d6-acetone): 6 8.21 (br s, IH, -NH-), 7.53 (s, IH, Im-H5\ 7.15 (s, IH, lm-H4), 5.36 (s, 2H, -CH2-CO-), 4.06 (qd, 2H, 3 J H - F = 9.4 Hz, -CH2-CF3). 19F{lH} N M R (188 MHz, de-acetone): 5 4.30 (s, -CF 3 ) . 3.2.2.8 2-(2-Nitro-l-H-imidazol-l-yl)-N-(3-bromo-3,3-difluoropropyl) acetamide Reaction conditions similar to those in the synthesis of EF3 were used: 7 (0.206 g, 1.20 mmol) was dissolved in THF (20 mL) and N M M (131 uL, 1.20 mmol) and iBuClFrm (172 uL, 1.32 mmol) were added. To the resulting pale orange slurry were added H 2 N C H 2 C H 2 C F 2 B r ' H C l (0.273 g, 1.30 mmol) and N M M (142 uL, 1.30 mmol). The mixture was then stirred at r.t. for 6 h. The white precipitate that formed was filtered off and washed with THF ( 3 x 5 mL). The filtrate was taken to dryness to give an orange oil that was purified via CTRON (Et20:acetone, 10:0 -» 10:3). An orange band eluted H [EF2Br] 63 References on page 129 Chapter 3 from the plate first using Et20; the product band was then eluted using acetone which when taken to dryness yielded an off-white solid (0.242 g, 62%). Crystals suitable for X-ray analysis were obtained from slow evaporation of a concentrated MeOH solution. Anal. calc. for C 8 HoN 4 0 3 F 2 Br: C, 29.38; H , 2.77; N , 17.13; found C, 29.35; H , 2.87; N , 17.22. L R - M S [DCI(+)]: 346, 344 (tvf + N H / ) , 329, 327 (TVT+ H), 299, 297 (Ivf - NO), 282, 280 OVT - N02), 247 ( M + - Br). HR-MS [DCI(+)] calc. for C 8 H i 0 O 3 N 4 8 1 B r (C 8 Hio0 3 N4 7 9 Br): 328.98838 (326.99047); found 328.98733 (326.99038). IR (v, cm"1): 3287 (N-H); 3120 (C-Hm,); 3062, 2963 (C-H); 1671 (C=0); 1494 (N-O a s y m ) ; 1371 (N-O s y m ). UV-Vis (MeCN): 328 (6.6). TT N M R (300 MHz, d6-acetone): 5 7.86 (br s, IH, -NH-), 7.50 (s, IH, Im-H5), 7.14 (s, IH, Im-H4\ 5.25 (s, 2H, -CH2-CO-), 3.55 (q, 2H, -NH-Cr7 2-), 2.73 (tt, 2H, 3 J H - F = 17 Hz, -C# 2-CF 2Br). 1 9F{TI} N M R (188 MHz, de-acetone): 5 32.71 (s, -GF 2Br). 3.2.2.9 2-(2-Nitro-l-H-imidazol-l-yl)-N-(3,3-difluoropropylene)acetamide [E=F2] An attempt to synthesize EF3 from EF2Br using a standard halogen exchange procedure (Br for F) led mainly to the "-HBr elimination product" E=F2; no EF3 was observed according to ^ { / H } NMR. Addition of EF2Br (0.0401 g, 0.122 mmol) to Bu 4 NF»H 2 0 (0.0684 g, 0.245 mmol) in 2 mL MeCN led to formation of a bright yellow solution. Stirring was continued at r.t. for 10 h before the solvent was removed and the major product isolated as a yellow oil via preparative TLC (CH 2 Cl 2 :Me0H, 25:1). The oil was dissolved in EtOAc, and hexanes were added to precipitate a white solid, which was filtered off and dried in vacuo (0.0132 g, 44 %). Anal. calc. for C 8 H 8 N 4 0 3 F 2 : C, 39.03; H , 3.28; N , 22.76; found C, 39.12; H , 3.20; N , 22.55. IR (v, cm"1): 3311 (N-H); 3104(0-^) ; 2923, 2851 (C-H); 1668 (C=0); 1487 (N-O a s y m ); 1375 (N-O s y m ). UV-Vis (MeCN): 320 (3.9). X H N M R (300 MHz, d6-acetone): 5 7.75 (br s, IH, -NH-), 7.50 (s, IH, Im-Hi), 7.12 (s, IH, Im-H4), 5.22 (s, 2H, -CH2-CO-), 4.55 (dtd, IH, ^(trans) = 25 H 64 References on page 129 Chapter 3 Hz, 3 J H H = 7.7 Hz, 3 J H F ( C I S ) = 2.2 Hz, -CH=CF2), 3.85 (dddd, 2H, 2 J H H = 9.2 Hz, 3 J H H = 7.1 Hz, Vctrans) = 2.1 Hz, Vcds) = 1.4 Hz, -Cft-CH=). ^ { T I } N M R (188 MHz, ds-acetone): 5 -12.78 (d, 2 J F F = 26.3 Hz, =C-F c i s), -13.82 (d, 2 J F F = 26.3 Hz, ^ - i w ) . 3.2.2.10 2-(2-Nitro-l-H-imidazol-l-yl)-N-(3-fluoropropyl)acetamide [EF1] Reaction conditions similar to those in the synthesis of EF3 were used: 7 (0.124 g, 0.722 mmol) was dissolved in THF (15 mL) and N M M (79 uL, 0.717 mmol) and iBuClFrm (103 pL, 0.789 mmol) were added. To the resulting white slurry were added H 2 N C H 2 C H 2 C H 2 F ' H F (0.241 g, 1.41 mmol) and N M M (142 pL, 1.30 mmol). The mixture was stirred at r.t. for 4.5 h before the white precipitate that formed was filtered off and washed with THF ( 3 x 5 mL). The filtrate was taken to dryness to give a yellow oil that was purified using preparative TLC (CH 2 Cl 2 :MeOH, 20:1); the major band isolated yielded a creamy white solid (0.0146 g, 9 %). The hydrochloride salt of 3-fluoropropylamine (5) was used in order to eliminate problems caused by the presence of HF in the hydrofluoride salt. Identical reaction conditions were used as described above, but using 0.264 g (1.41 mmol) of the hydrochloride, and a white solid was isolated (0.0249 g, 15 %). Anal. calc. for C8H11N4O3F: C, 41.74; H , 4.82; N , 24.34; found C, 40.69; H , 4.62; N , 21.34. L R - M S [DCI(+)]: 231 (M++H), 184 (M+-N02). HR-MS [DCI(+)J calc. for CgHjzOsN^: 231.08933; found 231.08881. IR (v, on 1 ) : 3284 (N-H); 3124 (C-Hfa); 2926, 2851 (C-H); 1660 (C=0); 1492 (N-O a 8 y m); 1369 (N-O s y m). UV-Vis (MeOH): 316 (6.7), 210 (5.2). TI N M R (300 MHz, d6-acetone): 8 7.64 (br s, IH, -NH-), 7.52 (s, IH, lm-H5), 7.18 (s, IH, lm-H4), 5.22 (s, 2H, -CH2-CO-\ 4.51 (dt, 2H, 2 J H . F = 66 Hz, 3 J H . H = 9.4 Hz, -CH2F), 3.38 (q (dt), 2H, -NH-Cr7 2,-), 1.91 (dp (dtt), 2H, 3 J H . F = 42 Hz, 3 J H . H = 9.7 Hz, -C# 2 -CH 2 F). 65 References on page 129 Chapter 3 1 9 F N M R (282 MHz, d6-acetone): 5 -142.50 (spt (tt), 2 J H . F = 49.1 Hz, 3 J H - F = 25.1 Hz, -CH 2 F). T o N02 Comparison of the TLCs obtained from the above reactions shows that both gave two other major products having higher Rf values than that of EF1; a small amount of unreacted 7 (R f = 0) was also observed (Figure 3-2). The difference in yield of EF1 (9% vs. 15%) between the two reactions probably indicates that the presence of HF does not limit the reaction yield. Band 1 = 2N02Im-methylester (8) Band 2 = 2N02Im-isobutylester (11) Band 3 = EF1 Band 4 = 2N02Im-acetic acid (7) N Y N N 0 2 R= OH (7) OMe (8) 0"Bu (11) Figure 3-2: TLC analysis of products obtained from EF1 synthesis. Bands 1 and 2 were also isolated and characterized in order to determine the identities of the side-products responsible for the low yields of EF1. The pale yellow solid isolated from band 1 was identified as 8 as evidenced by X H N M R spectroscopy (p. 55). This species is likely formed from reaction of the acyl carbonate with MeOH during preparative TLC separation. The species isolated from band 2 was more difficult to identify; the *H N M R spectrum shows signals corresponding to the isobutyl protons and 66 References on page 129 Chapter 3 JK revealed the presence of a Vco band. MS analysis definitely showed that the species is the isobutylester (11) (see Scheme 3-13, p. 106). LR-MS [DCI(+)]: 228 (Tvf), 182 (JVT -N0 2 ) . IR (v, cm -1): 3125 (C-H^); 2964, 2876 (C-H); 1751 (C=0); 1490 (N-O a s y m ) ; 1369 (N-O s y m ) . *H N M R (200 MHz, d6-acetone): 5 7.59 (s, IH, Im-H5), 7.19 (s, IH, lm-H4), 5.40 (s, 2H, -C#2-CO-), 4.01 (d, 2H, O-C/^-CH-), 1.92 (m, IH, -Ci7-(CH 3) 2), 0.87 (d, 6H, -CH-(C# 3) 2). 3.2.2.11 2-(2-Nitro-l-H-imidazoI-l-yl)-N-(2-nuoroethyl)acetamide [EFl(-l)] Synthesis 1 Reaction conditions similar to those in the synthesis of EF3 were used: 7 (0.154 g, 0.891 mmol) was dissolved in THF (10 mL), and N M M (105 pL, 0.953 mmol) and iBuClFrm (140 uL, 1.07 mmol) were added. To the resulting pale yellow slurry were added H 2 NCH 2 CH 2 F-HC1 (0.132 g, 1.33 mmol) and N M M (120 pL, 1.10 mmol). The mixture was stirred at r.t. for 6 h before the white precipitate that formed was filtered off and washed with THF ( 3 x 5 mL). The filtrate and washings were taken to dryness to give a yellow oil that was chromatographed using CTRON (CH 2C1 2:THF, 10:3). A white solid was isolated from the fourth band and was recrystallized from acetone/Et20 (1:1) (0.124 g, 64%). Anal. calc. for C 7 H 9 N 4 0 3 F : C, 38.89; H , 4.20; N , 25.92; found C, 38.88; H , 4.25; N , 25.77. IR (v, cm"1): 3296 (N-H); 3120 (C-HmO; 2926, 2851 (C-H); 1669 (C=0); 1488 (N-O a s y m ) ; 1372 (N-O s y m). UV-Vis (MeOH): 316 (6.2), 212 (4.7). *H N M R (300 MHz, de-acetone): 5 7.87 (br s, IH, -N/7-), 7.50 (s, IH, Jm-H5), 7.14 (s, IH, Jm-H4), 5.26 (s, 2H, -CH2-CO-), 4.48 (dt, 2H, 2 J H . F = 48 Hz, 3 J H . H = 6.5 Hz, -C# 2F), 3.55 (dq (dtd), 2H, 3 J H . F = 30 Hz, 3 JHH = 6.2 Hz, -NH-Cf^-). 1 9 F N M R (282 MHz, de-acetone): 6 -144.94 (spt (tt), 2 J H -F = 49.1 Hz, 3 J H - F = 26.2 Hz, -CH 2 F). 1 3 C N M R (50 MHz, d6-acetone): 5 166.89 (-CO-), 130.92 and 129.92 (Jm-C4and5), 84.70 (d, = 166.5 Hz, -CH 2 F), 54.33 (-CH2-CO-), 42.89 (d, 2 J C F = 21 Hz, -NH-CH 2 ,-) . H 67 References on page 129 Chapter 3 Synthesis 2 In a N2-flushed 50 mL, 2-neck flask at -78 °C, CH 2 C1 2 (10 mL) was added to SR2508 (0.100 g, 0.466 mmol) to give a white slurry; DAST (80 uL, 0.606 mmol) was then added dropwise and the mixture was stirred for 3 h at r.t. The resulting yellow/orange solution was chromatographed using CTRON (Et 2 0 —> THF), but neither of the two products was the desired EFl(-l). A wafer-like, crystalline, white solid was isolated from the second band (R f = 0.40 with THF) (~ 10 mg). *H N M R (300 MHz, d6-acetone): 8 7.59 (s, IH, lm-H5), 7.38 (dd, IH, -CF 2-N#-), 7.13 (s, IH, Im-H4), 5.34 (q, 2H, -C/f 2 -CF 2 -) , 4.38 (dt, 2H, -NH-C# 2), 3.71 (t, 2H, -C# 2OH). ^F l /H} N M R (188 MHz, d6-acetone): 8 -6.22 (s, -GF 2-NH-). N M R data indicate fluorine substitution at the carbonyl group (12) rather than at the OH group, and so this reaction procedure was pursued no further. 3.2.2.12 2-(2-Nitro-l-H-imidazol-l-yl)-N-(3-chloropropyl)acetamide [ECU] Reaction conditions similar to those for the synthesis of EF3 were used: 7 (0.103 g, 0.596 mmol) was dissolved in THF (15 mL) and N M M (64 uL, 0.581 mmol) and iBuClFrm (83 pL, 0.636 mmol) were added. To the resulting pale yellow slurry were added H2NCH2CH2CH2C1«HC1 (0.0871 g, 0.700 mmol) and N M M (71 pX, 0.650 mmol). The mixture was then stirred at r.t. for 3.5 h after which the yellow precipitate that formed was filtered off and washed with THF ( 3 x 5 mL). The filtrate and washings were taken to dryness and the resulting yellow oil was chromatographed using preparative T L C (CH 2 Cl 2 :MeOH, 20:1). The major band yielded a white solid (0.0945 g, 64%). Anal. calc. for CgHnN^sCl : C, 38.96; H , 4.50; N , 22.71; found: C, 39.36; H , 4.64; N , 22.70. IR (v, cm - 1): 3287 (N-H); 3099 (C-HfoO; 2946, 2868 (C-H); 1657 (C=0); 1484 (N-O a s y m ) ; 1368 (N-O s y m ). UV-VIS (MeOH): 316 (6.1), 216 (3.4). *H N M R (300 MHz, d6-acetone): 8 H 12 68 References on page 129 Chapter 3 7.70 (br s, 1H, -NH-), 7 4 9 (s> 1 H> Im-#j), 7.14 (s, IH, Im-/^), 5.23 (s, 2H, -CH2-CO-\ 3.66 (t, 2H, -CftCl) , 3.39 (q (dt), 2H, -NH-Cft-) , 1 9 8 (P (tt)> 2 H > -CH 2-C# 2-). 3.2.2.13 2-(2-Nitro-l-H-imidazol-l-yl)-N-(2-chloroethyl)acetamide [ECU(-l)] Synthesis 1 Reaction conditions similar to those for the synthesis of EF3 were used: 7 (0.100 g, 0.585 mmol) dissolved in THF (15 mL) and N M M (65 uL, 0.590 mmol) and iBuClFrm (84 pL, 0.644 mmol) were added. To the resulting yellow slurry were added H2NCH2CH2C1»HC1 (0.0746 g, 0.644 mmol) and N M M (72 uL, 0.659 mmol). The mixture was then stirred at r.t. for 4.5 h before the white precipitate that formed was filtered off and washed with THF ( 3 x 5 mL). The filtrate and washings were taken to dryness to give a yellow oil which was chromatographed using preparative T L C (CH 2 Cl 2 :MeOH, 20:1); the major band yielded a white solid (0.0939 g, 69%). Anal. calc. for C7H9N4O3CI: C, 36.14; H , 3.90; N , 24.08; found: C, 36.17; H , 3.92; N , 24.04. L R - M S [EI(+)]: 233 (M 4 ), 196 (M* - CI), 186 (TS/t - N0 2 ) . IR (v, cm"1): 3292 (N-H); 3117 (C-Hfa); 2926, 2851 (C-H); 1665 (C=0); 1489 (N-O a s y m ); 1373 (N-O s y m). UV-Vis (MeOH): 316 (5.8), 218 (3.7). *H N M R (300 MHz, d6-dmso) 8 8.64 (br t, IH, -NH-), 7.53 (s, IH, lm-H5X 7.20 (s, IH, lm-H4), 5.15 (s, 2H, -CH2-CO-), 3.61 (t, 2H, -CH2C\), 3.44 (q (dt), 2H, -NH-C# 2-). 1 3 C N M R (50 MHz, d6-acetone): 8 166.17 (-CO-), 128.90 and 127.58 (Im-C4and5), 51.63 (-CH 2-CO-), 43.46 (-OLC1), 40.90 (-NH-CH 2 )-). H a H 69 References on page 129 Chapter 3 Synthesis 2 (see Scheme 3-16, p. 124) In a flask equipped with a reflux condenser, CHC1 3 (10 mL) was added to SR2508 (0.403 g, 1.88 mmol) at 0 °C to yield a white slurry. D M F (3 mL) was then added, resulting in a clear, colourless solution. TsCl (0.553 g, 2.90 mmol) and py (350 pL, 4.34 mmol) were added (together), and the reaction mixture was stirred for 0.5 h. The ice-bath was then removed and the reaction mixture was allowed to warm to r.t. E t 2 0 was then added dropwise ( -30 mL) and the white precipitate that formed was isolated via suction filtration. (From *H NMR, this solid was found to be a mixture of pyridinium hydrochloride and pyridinium tosylate.) The filtrate was taken to dryness and the resulting residue was dissolved in minimal acetone before being loaded onto a column. The column was then eluted with 5% acetone in CH2CI2 and two major bands were isolated. The first band yielded unreacted TsCl, while the second band gave ECll(-l) (0.207 g, 47 %). 3.2.2.14 2-(2-Nitro-l-H-imidazol-l-yl)-N-(3-bromopropyl)acetamide [EBrl] Reaction conditions similar to those for the synthesis of EF3 were used: 7 (0.100 g, 0.585 mmol) was dissolved in THF (15 mL), and N M M (65 pL, 0.590 mmol) and iBuClFrm (84 pL, 0.644 mmol) were added. To the resulting pale yellow slurry were added H 2NCH 2CH 2CH 2Br»HBr (0.141 g, 0.644 mmol) and N M M (72 pL, 0.659 mmol). The mixture was then stirred at r.t. for 4 h before the white precipitate that formed was filtered off and washed with THF ( 3 x 5 mL). The filtrate and washings were reduced in volume and subsequently loaded onto a column and eluted dropwise (CH 2C1 2:THF, 5:1). The first 200 mL of eluate was discarded and the first band was collected in 3 x 50 mL fractions. The eluent strength was increased (using 50% THF) and after 150 mL of eluate, the second band emerged containing the desired product. Eight 50 mL tractions were collected (the purity of the product being confirmed by TLC analysis) and combined before being taken to dryness to give a pale yellow solid (0.113 g, 67%). Anal. calc. for C 8 H n N 4 0 3 B r : C, 33.04 ; H , 3.83; N , 19.22; found: C, 32.19; H , 3.68; N , 18.57. L R - M S [DCI(+)]: 293, 291 (Wt + H), 246, 244 (M+ - N0 2 ) . HR-MS [DCI(+)] calc. for C 8 H i 2 0 3 N 4 8 1 B r ( C 8 H 1 2 0 3 N 4 7 9 B r ) : 293.00723 (291.00932); found 293.00732 (291.00932). 70 References on page 129 Chapter 3 IR (v, cm-1): 3286 (N-H); 3100 (C-Hfa); 2929 (C-H); 1656 ( C O ) ; 1484 (N-0a s y m); 1366 (N-0 sym). UV-Vis (MeOH): 316 (6.2), 212 (3.7). X H N M R (300 MHz, d6-acetone): 6 7.70 (br s, IH, -NH-), 7.51 (s, IH, lm-H5), 7.08 (s, IH, Jm-H4), 5.21 (s, 2H, -CH2-C0-), 3.50 (t, 2H, -CH2Br), 3.39 (q (dt), 2H, - N H - C ^ - ) , 2.01 (m (tt), 2H, -C# 2-CH 2Br). 3.2.2.15 2-(2-Nitro-l-H-imidazol-l-yl)-N-(2-bromoethyl)acetamide [EBrl(-l)] Synthesis 1 Reaction conditions similar to those for the synthesis of EF3 were used: 7 (0.237 g, 1.37 mmol) was dissolved in THF (25 mL), and N M M (154 uL, 1.40 mmol) and iBuClFrm (196 uL, 1.50 mmol) were added. To the resulting pale yellow slurry were added H 2 NCH 2 CH 2 Br«HBr (0.312 g, 1.53 mmol) and N M M (167 uL, 1.53 mmol). The mixture was then stirred at r.t. for 4 h before the white precipitate was filtered off and washed with Et 20 ( 3x5 mL). The pale yellow filtrate and washings were taken to dryness and the residue was chromatographed on a silica gel column. The desired product was collected in the first 3 fractions as its R f was higher than those of the side-products and unreacted starting material. The purity of these fractions was confirmed by T L C analysis; the combined fractions were taken to dryness to give a pale yellow solid (0.269 g, 71%). Anal. calc. for C v H o N ^ B r ' f O . S THF): C, 35.90 ; H , 3.63 ; N , 17.53; found: C, 36.04 ; H , 3.92 ; N , 17.60. IR (v, cm'1): 3293 (N-H); 3108 (C-H^); 2923, 2854 (C-H); 1668 ( C O ) ; 1488 (N-OaSym); 1370 (N-O s y m). UV-Vis (MeOH): 316 (6.4), 214 (3.5). *H N M R (300 MHz, CD 3 OD): 8 7.41 (s, IH, lm-H5), 7.15 (s, IH, lm-H4), 5.13 (s, 2H, -C# 2-CO-), 3.58 (t, 2H, -NH-CH2-), 3.43 (t, 2H, -C# 2Br); (300 MHz, d6-acetone) 7.99 (br s, IH, -N#-CH 2-), 7.53 (s, IH, Im-Hs), 7.17 (s, IH, lm-H4), 5.28 (s, 2H, -CH2-CO-\ 3.64 (q (dt), 2H, -NH-Cr7 2-), 3.52 (t, 2H, -CH2Br). [The analysis in d6-acetone permitted identification of the triplet -NH-CH2- (methylene) signals found in the CD 3 OD spectrum.] H .Br 71 References on page 129 Chapter 3 1 3 C N M R (50 MHz, d6-acetone): 5 172.75 (-CO-), 134.84 and 134.26 (Im-C4 and5), 58.60 (-CH2-CO-), 48.24 (-NH-CH2,-), 37.88 (-CH 2Br). Synthesis 2 In a Schlenk tube under a flow of N 2 , SR2508 (0.114 g, 0.533 mmol) was dissolved in D M F (6 mL). The solution was then acidified using HBr( g ) (pH acidity tests were accomplished using pH paper, the DMF sample first being added to H 20; the HBr ( g) line was flushed with N 2 prior to use as this reduces fuming that results from the presence of H20.) The N 2-flow was resumed and PBr 3 (62.5 uL, 0.659 mmol) was added via syringe. The reaction was monitored by TLC and, after 3 h MeOH (10 mL) was added to quench unreacted PBr 3 . The mixture was then stirred for 5 min after which E t 3 N was added until the mixture was strongly basic. Excess E t 3 N and D M F were removed under vacuum and the remaining residue was chromatographed (CH 2 Cl 2 :MeOH, 10:1). The eluate was collected in fractions, each analyzed using TLC; the fractions containing the product were combined and taken to dryness to yield a pale yellow solid (0.0177 g, 12 %). 3.2.2.16 2-(2-Nitro-l-H-imidazol-l-yl)-N-(propyl)acetamide [EPrA] Reaction conditions similar to those for the synthesis of EF3 were used: 7 (0.0808 g, 0.469 mmol) was dissolved in THF (10 mL), and N M M (50 uL, 0.467 mmol) and iBuClFrm (70 pL, 0.536 mmol) were added. To the resulting yellow slurry were added H 2NCH 2CH 2CH 3»HC1 (0.0466 g, 0.488 mmol) and N M M (56 uL, 0.513 mmol). The mixture was then stirred at r.t. for 4.5 h before the white precipitate was filtered off and washed with THF (3x5 mL). The filtrate and washings were taken to dryness and the desired product was isolated using preparative TLC (CH 2 Cl 2 :MeOH, 10:1) to give a white solid (0.0301 g, 30 %). Anal. calc. for C g H i 2 N 4 0 3 : C, 45.28 ; H , 5.70 ; N , 26.40; found: C, 45.21 ; H , 5.58 ; N , 26.27. IR (v, cm -1): 3290 (N-H); 3113 (C-Hm,); 2969, 2878 (C-H) 1662 ( C O ) ; 1490 (N-O a s y m ); 1361 (N-O s y m). UV-Vis (MeOH): 316 (5.6), 228 (3.1). TI H r=\ •N. N02 72 References on page 129 Chapter 3 N M R (300 MHz, d6-acetone): 5 7.52 (br s, IH, -Nr7-CH 2-, overlapping with peak at 7.50), 7.50 (s, IH, 1m-H5), 7.13 (s, IH, Im-H4), 5.20 (s, 2H, -C# 2-CO-), 3 20 (q (dt), 2H, -NH-Cr7 2-), 1.52 (sex (qt), IH, -NHCH 2 C# 2 -) , 0.89 (t, 3H, -CH 2-C# 3). 3.2.2.17 2-(2-Nitro-l-H-imidazol-l-yl)-N-(wo-amyl)acetamide [EIAA] Reaction conditions similar to those for the synthesis of EF3 were used: 7 (0.0812 g, 0.472 mmol) dissolved in THF (10 mL), and N M M (50 pL, 0.467 mmol) and iBuClFrm (70 pL, 0.536 mmol) were added. To the resulting yellow slurry was added H 2 N C H 2 C H 2 C H ( C H 3 ) 2 (59 pL, 0.508 mmol). The mixture was then stirred at r.t. for 4 h before the white precipitate was filtered off and washed with THF ( 3 x 5 mL). The filtrate and washings were taken to dryness and the desired product was isolated using preparative TLC (CH 2 Cl 2 :MeOH, 20:1) to give a white solid (0.0773 g, 68%). Anal. calc. for C i 0 H 1 6 N 4 O 3 : C, 49.99 ; H , 6.71 ; N , 23.32; found: C, 49.64 ; H , 6.56 ; N , 22.98. IR (v, cm -1): 3267 (N-H); 3127 (C-H^); 2958, 2875 (C-H); 1656 ( C O ) ; 1490 (N-O a s y m ); 1374 (N-O s y m). UV-Vis (MeOH): 316 (7.1), 228 (4.0). X H N M R (300 MHz, d6-acetone): 5 7.52 (br s, IH, -Nr7-CH 2-, overlapping with peak at 7.50), 7.50 (s, IH, lm-Hs), 7.11 (s, IH, Im-H4), 5.20 (s, 2H, -C# 2-CO-), 3.27 (q (dt), 2H, -NH-C# 2-), 1.65 (m, IH, -C#-(CH 3) 2), 1.38 (q (dt), 2H, -NHCH 2 C# 2 -) , 0.88 (d, 6H, -CH(C# 3) 2). H r=\ •N. H 73 References on page 129 Chapter 3 3.2.3 2-Methyl-5-Nitroimidazole Compounds 3.2.3.1 2-(2-Methyl-5-nitro-lfl-imidazol-l-yl)acetic acid (13) A slightly modified literature synthetic procedure was used.20 In a 3-neck flask, metronidazole (3.33 g, 19.5 mmol) was dissolved in acetone (120 mL) under N 2 . To the resulting white slurry was added Jones' Reagent (20 mL of (7g C r 0 3 + 50 mL H 2 0 + 6.1 mL H2SO4}) dropwise via a pressure-equalized addition funnel, when a clear, dark orange solution resulted; this was then stirred at r.t. for 24 h to ensure complete oxidation of the alcohol to the corresponding acid. [After 3 h, TLC analysis revealed the presence of three compounds, the unreacted metronidazole, trace aldehyde and the acid (Rflaidehyde) > Rf i^cohoi) > Rf(acid)).] The final mixture was now dark green and a green, oily deposit was observed. The mixture was filtered twice through a plug of Celite to remove the green oil, and the clear, pale yellow filtrate was rotovapped to dryness. The resulting residue was then dissolved in H 2 0 (50 mL) and the solution was transferred to a separatory funnel and washed with EtOAc (5 x 80 mL). The organic fractions were combined and dried over MgS04 (10 min) before filtration and rotary evaporation yielded a pale yellow solid (3.61g, 55 %); mp 176-178 °C (lit.2 1 mp 179-180 °C). TI N M R (200 MHz, D 2 0 ) : 5 8.00 (s, IH, Im-H4), 4.85 (s, 2H, -CH2-\ 2.40 (s, 3H, Im-Gr73). The N M R data are in agreement with previously reported values.20 Metronidazole 13 3.2.3.2 2-(2-Methyl-5-nitro-l#-imidazol-l-yl)-N-(2,2,3,3,3) pentafluoropropyl) acetamide [MF5] Compound 13 (0.130 g, 0.70 mmol) was dissolved in a 5:1 mixture of THF (25 mL) and D M F (5 mL) at 0°C under N 2 [13 is sparingly soluble in neat THF]. 74 References on page 129 Chapter 3 N M M (80 u.L, 0.73 mmol) was then added and the clear, colourless mixture was stirred for 10 min before iBuClFrm (130 uL, 1.00 mmol) was added. Stirring was continued for 30 min during which the reaction mixture became a pale yellow slurry. The ice-bath was removed, and H 2NCH 2CF 2CF 3«HC1 (0.149 g, 0.87 mmol) and N M M (95 u,L, 0.87 mmol) were added. The reaction slurry was then stirred at r.t. for 3 h when TLC analysis confirmed the presence of a single product. The creamy white precipitate that formed was filtered off and washed with THF ( 3 x 5 mL). The pale yellow filtrate was taken to dryness to give a yellow oil that was purified via column chromatography using THF as the eluent. Two bands were observed: the first, a pale yellow band and the second, a colourless one containing MF5 . The second band was collected in several fractions and these were combined and taken to dryness to yield a white solid (0.117 g, 53 %). X-ray quality crystals were grown by slow evaporation of a concentrated MeOH solution of MF5 . Anal. calc. for C9H9N4O3F5: C, 34.19; H , 2.87; N , 17.53; found: C, 34.25; H , 2.99; N , 17.53. IR (v, cm"1): 3212 (N-H); 3028 ( C - H M ) ; 2921, 2851 (C-H); 1670 (C=0); 1473 (N-O a s y m ); 1368 (N-O s y m ). UV-Vis (MeOH): 310 (8.1), 230 (3.7), 212 (3.9). J H N M R (300 MHz, CD 3 OD): 8 8.07 (s, IH, lm-H4), 5.24 (s, 2H, -CH2-CO-), 4.08 (t, 2H, -CH2CF2-), 2.54 (s, 3H, Im-Cflj); (300 MHz, CDCI3) 8 7.93 (s, IH, lm-H4), 6.28 (br s, IH, -NH-), 4.93 (s, 2H, -CH2-CO-), 3.97 (dt, 2H, -CH2CF2), 2.47 (s, 3H, \m-CH3). ^{TJ.} N M R (188 MHz, CD3OD): 8 -7.56 (t, -CF 3 ) , -44.63 (q, -CF 2 -); (188 MHz, CDCh) 8 -8.17 (t, -CF 3 ) , -45.79 (q, -CF2-). 75 References on page 129 Chapter 3 3.2.3.3 2-(2-Methyl-5-nitro-l/?-imidazol-l-yl)-N-(2,2,2-trifluoroethyl) acetamide [MF3(-1)] Reaction conditions similar to those for the synthesis of MF5 were employed: 13 (0.252 g, 1.359 mmol) was added to THF (15 mL) and N M M (150 uL, 1.361 mmol) and iBuClFrm (192 pL, 1.472 mmol) were added (13 is not completely soluble in THF but the subsequent addition of base increases solubility). The resulting off-white slurry was then stirred at 0°C before H2NCH2CF3«HC1 (0.202 g, 1.490 mmol) and N M M (164 pX, 1.489 mmol) were added. The mixture was then warmed to r.t. and stirred for 4.5 h. The precipitate that formed was filtered off and the filtrate taken to dryness. Purification of the desired product from the residue was achieved via column chromatography (THF). It was imperative to elute dropwise (slowly) to ensure the separation of the desired product from a number of side-products (as seen from TLC analysis). MF3(-1) was the third and most concentrated band to elute from the column. The fractions from this band were combined and taken to dryness to give an off-white solid (0.291 g, 80 %). Anal. calc. for C g H o N ^ F s : C, 36.10; H, 3.41; N , 21.05; found: C, 36.24; H , 3.25; N , 20.01. IR (v, cm - 1): 3304 (N-H); 3117 (C-H^); 2960, 2871 (C-H); 1678 (C=0); 1471 (N-O a s ym); 1373 (N-O s y m ). UV-Vis (MeOH): 310 (8.6), 230 (3.5), 210 (4.0). X H N M R (300 MHz, de-acetone): 5 8.26 (br s, IH, -NH-), 7.92 (s, IH, lm-H4), 5.25 (s, 2H, -C/^-CO-) , 4.05 (qd, 2H, -Gr7 2CF 3), 2.43 (s, 3H, lm-CH3). l9E{lU) N M R (188 MHz, d6-acetone): 6 4.76 3.2.3.4 2-(2-Methyl-5-nitro-lfl-imidazol-l-yl)-N-(2-fluoroethyl)acetamide Reaction conditions similar to those for the synthesis of MF3(-1) were used: 13 (0.251 g, 1.359 mmol) was dissolved in THF (15 mL), and N M M (150 pX, 1.361 mmol) (s, -CF,). Me [ M F l ( - l ) ] 76 References on page 129 Chapter 3 and iBuClFrm (192 pL, 1.472 mmol) were added. To the resulting white slurry at 0°C were added H 2NCH 2CH 2F«HC1 (0.155 g, 1.560 mmol) and N M M (164 pL, 1.489 mmol). The mixture was warmed to r.t. and stirred for 4.5 h before the precipitate formed was filtered off. The filtrate was taken to dryness and the residue purified via column chromatography (THF); dropwise (slow) elution yielded MFl(-l) (fourth and most concentrated band) as several fractions from the column. These fractions were combined and taken to dryness to yield a pale yellow solid (0.207 g, 66 %). Anal. calc. for CgHnNtOsF: C, 41.74; H , 4.82; N , 24.34; found: C, 41.85; H , 4.80; N , 24.11. IR (v, cm"1): 3288 (N-H); 3110 ( C - H j m ) ; 2956, 2874 (C-H); 1668 (C=0); 1463 (N-O a s y m); 1367 (N-O s y m ). UV-Vis (MeOH): 310 (9.2), 226 (4.7), 210 (5.4). TI N M R (300 MHz, dg-acetone) 8 7.80 (br s, 2H, -NH- + lm-H4), 5.16 (s, 2H, -Gr72-CO-), 4.50 (dt, 2H, -CH2F, 2 J H F = 42 Hz), 3.55 (dq (ddt), 2H, -NH-C# 2-, 3 J H F = 26 Hz), 2.45 (s, 3H, \m-CH3). 1 9 F N M R (282 MHz, d6-acetone) 8 -144.91 (spt (tt), 2 J H F = 50.5 Hz, 3 J H F = 25.7 Hz, -CH2F). 3.2.3.5 2-(2-Methyl-5-nitro-lfl-imidazol-l-yl)-N-(2-chloroethyl)acetamide Reaction conditions similar to those for the synthesis of MFl(-l) were used: 13 (0.301 g, 1.62 mmol) was dissolved in THF (20 mL), and N M M (165 uL, 1.62 mmol) and iBuClFrm (220 uL, 1.73 mmol) were added. To the resulting white slurry at 0°C were added H 2 NCH 2 CH 2 CKHC1 (0.200 g, 1.73 mmol) and N M M (184 uL, 1.73 mmol). The mixture was warmed to r.t. and stirred for 5.5 h. The precipitate formed was removed via filtration and the pink filtrate was taken to dryness to yield an orange-pink oil that was purified using preparative TLC (CH 2 Cl 2 :MeOH, 10:1). The major band (lowest Rf) was isolated to give a white, microcrystalline solid (0.190 g, 47 %). Anal. calc. for [MCU(-l)] 77 References on page 129 Chapter 3 C8H11N4O3CI: C, 38.96; H , 4.50; N , 22.71; found: C, 37.28; H , 4.68; N , 20.89. L R - M S [DCI(+)]: 249, 247 (Tvf + H), 202, 200 (TVf - N0 2 ) . HR-MS [DCI(+)] calc. for C 8 H i i N 4 0 3 3 7 C l (C 8 HiiN 4 03 3 5 Cl): 249.05684 (247.05980); found 249.05725 (247.05896). IR (v, cm - 1): 3312 (N-H); 3028 (C-H t a ) ; 2921, 2851 (C-H); 1670 (C=0); 1473 (N-O a s y m ) ; 1368 (N-O s y m ). UV-Vis (MeOH): 310 (6.9), 232 (2.9), 210 (3.7). *H N M R (300 MHz, de-acetone) 5 7.94 (br s, 2H, -NH- + lm-H4), 5.16 (s, 2H, -CH2-CO-\ 3.69 (t, 2H, -CH2CI), 3.61 (q (dt), 2H, -NH-CH2-), 2.46 (s, 3H, lm-CH3). Of note, this reaction was not as clean when compared to those for the preparation of MF3(-1) and MFl(-l). TLC analysis reveals the presence of two other major products having Rf values higher than that of MCll(-l); a small amount of unreacted 13 (R f = 0) was also observed. Comparison of the TLC for authentic samples of MFl(-l) and MBrl(-l) with the reaction mixture facilitated identification of the desired product (Figure 3-3). *H N M R spectroscopic analysis of band 1 revealed the presence of the methyl ester (14), probably formed from the reaction of acyl carbonate with MeOH on the preparative TLC plate [TI N M R (300 MHz, de-acetone): 5 7.99 (s, IH, Jm-H4), 5.11 (s, 2H, -CH2-X 3.72 (s, 3H, -C#3), 2.42 (s, 3H, Im-C#3)]. For band 2, the isolated white crystalline solid (in relatively high concentration) was determined to be the isobutyl ester side-product (15) (analogous to 11, Scheme 3-13, p. 106). This finding suggests that H 2 0 was present during the reaction, likely coming from the extremely hygroscopic amine hydrochloride. XH N M R (300 MHz, de-acetone): 8 7.96 (s, IH, lm-H4), 5.26 (s, 2H, -CH2-CO-), 3.97 (d, 2H, O-CH2-), 2.49 (s, 3H, lm-CH3\ 1.93 (m, IH, -C#-(CH 3) 2), 0.91 (d, 6H, -CH-(C# 3) 2). 78 References on page 129 Chapter 3 N O 2 Band 1 = 2Me5N02Im-methyl ester (14) Band 2 = 2Me5N02Im-isobutyl ester (15) Band 3 = MCIl(-l) Band 4 = 13 J. o A = MFl(-l); B = MCll(-l) rxn. mixture; C = MBrl(-l) Figure 3-3: TLC analysis of MXl(-l) compounds (X =F, CI, Br). 3.2.3.6 2-(2-Methyl-5-nitro-lfl-imidazol-l-yl)-N-(2-bromoethyl)acetamide [MBrl(-l)] Reaction conditions similar to those for the synthesis of MFl(-l) were used: 13 (0.108 g, 0.586 mmol) was dissolved in THF (10 mL) and N M M (65 uL, 0.590 mmol) and iBuClFrm (83 pL, 0.636 mmol) were added. To the resulting white slurry at 0°C were added H 2NCH 2CH 2Br«HBr (0.135 g, 0.660 mmol) and N M M (73 uL, 0.663 mmol). The mixture was warmed to r.t. and stirred for 5 h. The precipitate was filtered off and the filtrate was taken to dryness. The resulting residue was chromatographed using preparative TLC (CH 2 Cl 2 :MeOH, 20:1), and the major band was isolated to yield an off-white solid (0.137 g, 80 %). Anal. calc. for C 8 H n N 4 0 3 B r : C, 33.01; H , 3.81; N , 19.25 [with 0.1 moi acetone C, 33.58; H , 3.94; N 18.87]; found: C, 33.69; H , 3.75; N , 18.66. IR(v, cm - 1): 3295 (N-H); 3100 (C-H^); 2926, 2854 (C-H); 1661 ( C O ) ; 1458 (N-O a s y m ) ; 1372 (N-O s y m ). UV-Vis (MeOH): 310 (8.9), 228 (3.7), 210 (4.8). TT N M R (300 MHz, de-acetone) 5 7.91 (br s, 2H, -NH- + lm-H4), 5.16 (s, 2H, -C# 2-CO-), 3.65 (q (dt), 2H, -NH-C# 2-), 3.53 (t, 2H, -C# 2Br), 2.45 (s, 3H, lm-CH3). 79 References on page 129 Chapter 3 N. Br 3.2.3.7 (2-Methyl-5-nitro-lfl-imidazol-l-yl)-N-(2-chloroethane)(16) Reaction conditions similar to those described for the synthesis of ECU(-l) were used (see synthesis 2, p. 70). In a flask equipped with a reflux condenser, CHCI3 (10 mL) was added to metronidazole (0.434 g, 2.54 mmol) at 0 °C to give a white slurry. TsCl (0.672 g, 3.53 mmol) and py (380 pL, 4.71 mmol) were then added simultaneously, and the reaction mixture was stirred for 0.5 h. The ice-bath was removed and the mixture was permitted to warm to r.t. over ~2 h. In attempts to isolate both the tosylated intermediate and the chlorinated metronidazole at this stage, the mixture was taken to dryness and the residue was column chromatographed. CH 2 C1 2 was first used to elute unreacted TsCl. Afterwards, the eluent strength was then increased to 5% MeOH and a pale yellow band emerged followed by a colourless band. These were identified to contain the desired product (16) and the unreacted metronidazole, respectively. No evidence for the presence of the tosylate was seen. The yellow fractions were combined and taken to dryness. A small amount of py present, as evidenced by TLC analysis, was removed under vacuum. The final product was isolated as a yellow, microcrystalline solid (0.125 g, 26%). Anal. calc. for C 6 H 8 N 3 0 2 C 1 : C, 38.01; H , 4.25; N , 22.16; found: C, 37.79; H , 3.96; N , 22.56. L R - M S [DCI(+)]: 190 Of ) . HR-MS [DCI(+)] calc. for C 6 H 8 N 3 0 2 3 7 C 1 (C 6 H 8 N 3 0 2 3 5 C1) : 192.03539 (190.03833); found 192.03610 (190.03792). *H N M R (300 MHz, d6-acetone) 8 7.96 (s, IH, Im-H4\ 4.80 (t, 2H, -CH2-C\), 4.09 (t, 2H, lm-CH2-), 2.60 (s, 3H, Im-CH3). •CI Me 16 80 References on page 129 Chapter 3 3.2.4 2-Methyl-4-Nitroimidazole Compounds 3.2.4.1 2-(2-Methyl-4-nitro-l-H-imidazol-l-yl)propionic acid (17) This synthesis was adapted from the nitrile hydrolysis procedure reported by Mann and Porter.22 2-Methyl-4-nitro-l-imidazolepropionitrile (2.709 g, 15.0 mmol) and N a N 0 2 (1.384 g, 20.1 mmol) were refluxed in aqueous 40% H 2 S 0 4 solution (25 mL). T L C analysis revealed that the reaction was complete after 2.5 d. The solution was neutralized with 6 M NaOH and then the pH adjusted to 3.23 using 0.1 M HC1 and 0.1 M NaOH. The resulting 17 precipitate was collected, washed with H 2 0 (2 x 25 mL) and dried in vacuo to yield a white powder (2.287g, 76 %). The purity of the compound was confirmed by melting point (mp: found 223-225 °C, reported 221-225°C (Aldrich)) and X H N M R (200 MHz, d6-dmso) 5 8.32 (s, IH, lm-Hs), 4.18 (t, 2H, -C# 2-CO-), 2.84 (t, 2H, Im-C#2-), 2.39 (s, 3H, 1m-CH3). Me 17 3.2.4.2 3-(2-Methyl-4-nitro-l-H-imidazol-l-yl)-N-(2,2,3,3,3-pentafluoro-propyl)propionamide [2M4NF5] 17 (0.305 g, 1.53 mmol) was placed in a 25 mL two-neck flask under N 2 , and THF (10 mL) was added. To the resulting white slurry was added D M F dropwise (~3 mL) until a clear, colourless solution was obtained. N M M (169 uL, 1.53 mmol) was then added and the mixture stirred for 10 min at 0°C before iBuClFrm (215 uL, 1.65 mmol) was next added. Stirring was continued for 30 min during which a pale yellow slurry formed. The ice-bath was removed, H 2NCH 2CF 2CF 3«HC1 (0.311 g, 1.82 mmol) and N M M (200 pL, 1.83 mmol) were added and the slurry was stirred at r.t. for 4 h. The white precipitate that formed was filtered off and washed with dry THF ( 3 x 5 mL). The filtrate was reduced in volume to ~5 mL before E t 2 0 was added slowly. A resulting pale yellow 81 References on page 129 Chapter 3 precipitate was isolated via filtration, but TLC analysis revealed later that this precipitate was not 2M4NF5. The filtrate was stored at 0°C overnight, resulting in the formation of clear, pale yellow crystals which were isolated via filtration (0.403 g, 83%). X-ray quality crystals were obtained from slow evaporation of a concentrated MeOH solution of 2M4NF5. Anal. calc. for C i o H n N ^ F s : C, 36.37; H , 3.36; N , 16.97; found: C, 36.67; H , 3.45; N , 16.69. JR (v, cm -1): 3260 (N-H); 3128 ( C - H n O ; 3079, 2921 (C-H); 1674 (C=0); 1506 (N-O a s y m ) ; 1400 (N-O s y m). UV-Vis (MeOH): 300 (6.2), 224 (3.6), 210 (4.2). *H N M R (300 MHz, d6-dmso) 5 8.69 (t, IH, -NH-), 8.22 (s, IH, Jm-H5), 4.23 (t, 2H, -CH2-CO-), 3.96 (td, 2H, 3 J H F = 16.5 Hz, -CH2-CF2-), 2.78 (t, 2H, Im-C#2-), 2.38 (s, 3H, Jm-CH3).  19FCH} N M R (188 MHz, d6-dmso) 5 -7.06 (t, -GF 3), -43.99 (q, -CF2-). 3.2.4.3 3-(2-Methyl-4-nitro-l-H-imidazol-l-yl)-N-(2,2,2-trifluoroethyl) propionamide [2M4NF3(-1)] Reaction conditions similar to those for the synthesis of 2M4NF5 were used: 17 (0.308 g, 1.55 mmol) was dissolved in THF (10 mL) and D M F (~3 mL), and N M M (170 pL, 1.55 mmol) and iBuClFrm (220 pL, 1.68 mmol) were added. To the resulting yellow/pink slurry were added H 2NCH 2CF 3«HC1 (0.234 g, 1.73 mmol) and N M M (190 pX, 1.74 mmol). The mixture was stirred at r.t. for 4 h before the white solid was filtered off and washed with THF ( 3 x 5 mL). The THF volume was reduced to ~ 5 mL under vacuum and then Et 2 0 was added until a cloudy white mixture persisted. This mixture was stored at 0°C for 16 h to yield a white, microcrystalline solid (0.377 g, 87%). Anal. calc. for C9HHN4O3F3: C, 38.58; H , 3.96; N , 19.99; found: C, 38.79; H , 3.93; N , 19.78. IR (v, cm - 1): 3263 (N-H); 3122 (C-Hm,); 3076, 2946 (C-H); 1675 (C=0); 1506 (N-O a s y m ) ; 1400 (N-O s y m ). UV-Vis (MeOH): 300 (6.3), 224 (3.8), 212 (3.9). *H N M R (300 MHz, Me 82 References on page 129 Chapter 3 ds-dmso) 8 8.70 (t, IH, -NH-), 8.21 (s, IH, Im-Hs), 4.20 (t, 2H, -C# 2-CO-), 3.90 (sex (qd), 2H, -C# 2-CF 3), 2.73 (t, 2H, Im-C#2-), 2.37 (s, 3H, lm-CH3). 19F{lB.} N M R (188 MHz, de-dmso) 8 5.63 (s, -CF 3 ) . 3.2.4.4 3-(2-MethyI-4-nitro-l-H-imidazol-l-yl)-N-(2-fluoroethyl) propionamide [2M4NF1(-1)] Reaction conditions similar to those for the synthesis of 2M4NF5 were used: 17 (0.308 g, 1.55 mmol) was dissolved in THF (10 mL) and D M F (~3 mL), and N M M (170 uL, 1.55 mmol) and iBuClFrm (220 pL, 1.68 mmol) were added. To the resulting yellow/pink slurry was added H 2 NCH 2 CH 2 F-HC1 (0.174 g, 1.74 mmol) and N M M (190 uL, 1.74 mmol). The mixture was stirred at r.t. for 5 h before the white solid was filtered off and washed with THF ( 3 x 5 mL). The volume of the pink filtrate was reduced to ~3 mL under vacuum, followed by the addition of E t 2 0 until a cloudy white mixture persisted. Sonication of the mixture for 5 min at 5 °C and then filtration yielded a white microcrystalline solid (0.290 g, 77%). X-ray quality crystals were grown from slow evaporation of a concentrated MeOH solution of 2M4NF1(-1). Anal. calc. for C 9 H i 3 N 4 0 3 F : C, 44.25; H , 5.37; N , 22.94; found: C, 44.38; H , 5.34; N , 22.75. IR (v, cm"1): 3298 (N-H); 3110 (C-H^); 3073, 2947 (C-H); 1668 ( C O ) ; 1504 (N-O a 8 y m ); 1396 (N-Osym). UV-Vis (MeOH): 300 (9.1), 210 (7.0). *H N M R (300 MHz, d6-dmso) 8 8.28 (t, IH, -N#-), 8.23 (s, IH, lm-H5), 4.37 (dt, 2H, -C# 2F, 2 J H F = 56 Hz), 4.19 (t, 2H, -Cr7 2-CO-), 3.31 (dq (ddt), 2H, -C# 2 -CH 2 F, 3 J H F = 28 Hz), 2.69 (t, 2H, Im-Cr72-), 2.38 (s, 3H, Im-CHs). 1 9 F N M R (282 MHz, d6-dmso) 8 -142.64 (spt (tt), 2 J H F = 49.2 Hz, 3 J H F = 24.2 Hz). H Me 83 References on page 129 Chapter 3 3.2.4.5 3-(2-methyl-4-nitro-l-H-imidazol-l-yl)-N-(2-chIoroethyI)propionamide [2M4NC11(-1)] Reaction conditions similar to those for the synthesis of 2M4NF5 were used: 17 (0.306 g, 1.54 mmol) was dissolved in THF (10 mL) and D M F (~3 mL), and N M M (169 pL, 1.54 mmol) and iBuClFrm (220 uL, 1.68 mmol) were added. To the resulting yellow/pink slurry were added H2NCH2CH2C1»HC1 (0.195 g, 1.69 mmol) and N M M (188 pL, 1.69 mmol). The mixture was stirred at r.t. for 4 h before the white solid was filtered off and washed with THF ( 3 x 5 mL). The colourless filtrate was taken to dryness to give an oily white residue; TLC analysis indicated several side-products. The residue was chromatographed using preparative TLC (CH 2 Cl 2 :MeOH, 20:1); the major band yielded a colourless oil. Addition of a small volume (~3 mL) of MeOH to the oil, followed by addition of E t 2 0 (15 mL) led to a cloudy white mixture. The white slurry was stored at r.t. for 2 h, before sonication (5 min at 5 °C) and then filtration gave a white, microcrystalline solid (0.315 g, 79%). Anal. calc. for C9H13N4O3CI: C, 41.47; H , 5.03; N , 21.49; found: C, 41.83; H , 4.81; N , 20.64. IR (v, cm -1): 3258 (N-H); 3119 ( C - H ^ ; 3079, 2923 (C-H); 1666 (C=0); 1499 (N-O a s y m ); 1384 (N-O s y m). UV-Vis (MeOH): 302 (6.7), 226 (3.9), 212 (4.3). *H N M R (300 MHz, d6-acetone) 8 8.03 (s, IH, lm-H5), 7.60 (br s, IH, -NH-), 4.38 (t, 2H, -CH2-CO-) 3.60 (t, 2H, -C#2C1), 3.51(q (dt), 2H, -C/ / 2 -CH 2 Cl) , 2.84 (t, 2H, Im-C// 2-), 2.44 (s, 3H, Im-CH3). 84 References on page 129 Chapter 3 3.2.4.6 2-(2-methyl-4-nitro-l-H-imidazol-l-yl)-N-(2-bromoethyl)propionamide [2M4NBrl(-l)] Reaction conditions similar to those for the synthesis of 2M4NF5 were used: 17 (0.307 g, 1.54 mmol) was dissolved in THF (10 mL) and D M F (~3 mL), and N M M (169 pL, 1.54 mmol) and iBuClFrm (220 pX, 1.68 mmol) were added. To the resulting pale yellow slurry were added H 2NCH 2CH 2Br«HBr (0.347 g, 1.70 mmol) and N M M (189 pL, 1.70 mmol). The mixture was stirred at r.t. for 6 h. A workup identical to that used for 2M4NC11(-1) yielded clear, colourless crystals (0.098 g, 21%). Anal. calc. for C9H13N403Br: C, 35.43; H , 4.29; N , 18.36; found: C, 35.62; H , 4.28; N , 18.65. IR (v, cm'1): 3259 (N-H); 3115 (C-Hm,); 3072, 2918 (C-H); 1660 (C=0); 1499 (N-O a s y m ) ; 1399 (N-O s y m ). UV-Vis (MeOH): 300 (7.4), 226 (4.4), 210 (5.6). *H N M R (300 MHz, ds-acetone) 5 8.03 (s, IH, lm-Hs), 7.67 (br s, IH, -N/7-), 4.36 (t, 2H, -C# 2-CO-), 3.56 (q (dt), 2H, -C# 2-CH 2Br), 3.44 (t, 2H, -C# 2Br), 2.83 (t, 2H, Im-C#2-), 2.42 (s, 3H, Im-CHs). 3.2.5 Reactions of SR2508 with Tf20 3.2.5.1 2-(2-Nitro-l-H-imidazol-l-yl)-N-(ethylformate)acetamide (18) This reaction was carried out in an attempt to synthesize the triflate analogue of SR2508; however, the presumed triflate was too reactive as demonstrated by the results from the attempted synthesis. In a Schlenk tube under 1 atm N 2 , CH 2 C1 2 (15 mL) and D M F (3 mL) were added to SR2508 (0.303 g, 1.14 mmol) to give a clear, colourless solution; this was then stirred at 0 °C for 30 min before py (320 pL, 3.95 mmol) and subsequently (CF 3 S0 2 ) 2 0 (342 pL, 3.95 mmol) were added dropwise. The ice-bath was removed and the dark orange reaction mixture was stirred for 3 h at r.t. TLC analysis 85 References on page 129 Chapter 3 (using THF) revealed the presence of three products characterized by a bright yellow band (R f = 0.95, 19), a colourless band (R f = 0.65, 18) and another colourless band (R f = 0.10, unidentified). The solvent was removed (under high vacuum) and 18 was isolated using column chromatography (THF); further purification via sublimation (at 130 °C under vacuum) yielded a white solid (0.703 g, 67 %). Anal. calc. for C8H10N4O5: C, 39.67; H , 4.16; N , 23.13; found: C, 39.90; H , 4.15; N , 23.03. LR-MS [EI(+)]: 243 OVT), 196 OVf - N0 2 ) . IR (v, cm"1): 3275 (N-H); 3096 (C-H^); 2940 (C-H); 1716 (C=Oaid); 1655 ( 0 = 0 ^ ) ; 1487 (N-O a s y m ); 1362 (N-O s y m). TI N M R (300 MHz, d6-acetone): 5 8.13 (s, IH, -CHO), 7.80 (br. s, IH, -NH-), 7.50 (s, IH, lm-H5), 7.14 (s, IH, lm-H4), 5.25 (s, 2H, -C# 2-CO-), 4.21 (t, 2H, -CH2-0), 3.55 (q (dt), 2H, -NH-Cr7 2-). 1 3 C N M R (75 MHz, de-acetone): 5 166.7 (-CO-NH-), 161.8 (-CHO), 128.7 and 128.2 (lm-C4 and J), 62.8 (-CH 2-CO-), 52.5 (-NH-CH 2-), 39.0 (-CH 2-0). 3.2.5.2 CycF3 (19) Conditions similar to those for the synthesis of 18 were used, except that D M F was not used. Following addition of (CF 3 S0 2 ) 2 0, the slurry became a clear, bright yellow solution; stirring was continued for 1 h at 0 °C and 2 h at r.t. during which a yellow slurry was formed. The mixture was filtered and the filtrate was reduced in volume to ~5 mL. before E t 2 0 (25 mL) was added to yield a white precipitate. The solid was filtered off and the yellow filtrate was reduced in volume to ~2 mL before being column chromatographed (CH 2C1 2 -> CH2C12:acetone, 4:1). The first eluted band (R f = 0.85, yellow) yielded 19, a brilliant yellow powder (0.135 g, 36 %). Crystals suitable for X-ray analysis were obtained from slow evaporation of a solution of the compound in a 1:1 mixture of CH 2C1 2 : acetone. Anal. calc. for C8H7N4O5SF3: C, 29.28; H , 2.15; N , 17.07; found: C, 29.15; H , 2.22; N , 18 86 References on page 129 Chapter 3 17.42. L R - M S [DCI(+)]: 329 (TVf + H). IR (v, cm 1): 3155 ( C T ^ ) ; 2980 (C-H t a ) ; 2918 (C-H); 1710 (C-O); 1485 (N-O a s y m ); 1362 (N-O s y m); 1203 (S=0). UV-Vis (MeOH): 314 (4.3). X H N M R (300 MHz, d6-acetone): 5 7.58 (d, IH, lm-H5), 7.17 (d, IH, Im-H4), 6.78 (s, -CH=C-), 4.64 (t, 2H, -CH2-0-\ 4.36 (t, 2H, -N-Cff 2-)- 1 3 C N M R (75 MHz, de-acetone): 5 146.5 (-CH=C-), 128.6 and 127.8 (Im-C,wj), 120.9 (q, -CF 3 ), 89.0 (-CH=C-), 68.8 (-CH2-O-), 49.9 (-N-CH2-). 1 9F{TI} N M R (188 MHz, dg-acetone): 5 0.88 (s, -3.2.6 Reactions with RSU1111 3.2.6.1 Synthesis of (20), a pentafluorinated derivative of RSU1111 The same reaction procedure to synthesize RevEF5 (synthesis 3) was used to synthesize 20. C F 3 C F 2 C 0 2 H (105 uL, 1.00 mmol) was dissolved in THF (15 mL) and N M M (110 uL, 1.05 mmol) and iBuClFrm (150 uL, 1.15 mmol) were added. To the resulting white slurry were added RSU1111 (0.206 g, 1.07 mmol) and N M M (125 pL, 1.19 mmol). Stirring was continued at r.t. for 20 h before the yellow precipitate that formed was filtered off, washed with THF ( 3 x 5 mL) and the filtrate purified via column chromatography (CH2CI2:acetone, 20:1). Two bands eluted, and the minor band with Rf = 0.60 was isolated to yield 20 as a colourless oil (0.0155 g, 5 %). LR-MS [DCI(+)]: 350 (TVf + NH4), 333 O T + H). HR-MS [DCI(+)] calc. for CgHjoOaN^: 333.06221; found 333.06198. TI N M R (300 MHz, d6-acetone): 8 8.76 (br. s, IH, -NH-), 7.54 (s, IH, lm-H5), 7.09 (s, IH, Im-H4), 4.90 (d, IH, -OH), 4.78 (dd, IH, Im-CMH)-), 4.38 (dd, IH, lm-CH(H)-), 4.22 (m, IH, -Cr7(0H)), 3.50 (m, 2H, -Gr72-NH-). l9F{xB} N M R (188 MHz, de-acetone): 5 -6.77 (t, -CF3), -46.35 (q, -C7v). 19 87 References on page 129 Chapter 3 O e e N H 3 C 1 N. r=\ 'N' OH OH N02 H RSU1111 20 3.2.6.2 Synthesis of (21), a dibromo derivative ofRSUllll Reaction conditions including stoichiometrics identical to those for the synthesis of compound 20 were used except CBr 3 C02H was used instead of CF3CF2CO2H. The white precipitate that formed was filtered off, washed with THF (3x5 mL) and the filtrate was purified using the CTRON (Et20:acetone, 1:1). Two bands eluted and the second band (Rf = 0.65) gave a yellow oil. Addition of E t 2 0 yielded a white precipitate that was isolated via filtration and dried in vacuo (0.0184 g, 5 %). Anal. calc. for CgHiolSUO^^: C, 24.89; H , 2.61; N , 14.51; found: C, 25.60; H , 2.65; N , 13.54. LR-MS [EI(+)]: 387 (M+l), 340 O f -N0 2 ) , 156 Ovf -CH 2 NHCOCBr 2 H), 114 (2N02Im), 80 (Br). HR-MS [EI(+)] calc. for C 8 Hio0 4 N 4 8 1 Br2 (C 8 Hi 0 O 4 N 4 8 1 Br 7 9 Br) [ C 8 H 1 0 O 4 N 4 7 9 B r 2 ] : 388.91061 (386.91266) [384.91470]; found 388.91604 (386.91227) [384.91401]. *H N M R (300 MHz, CD3OD): 5 7.46 (s, IH, lm-H5), 7.13 (s, IH, lm-H4), 6.21 (s, IH, -CBr 2#), 4.67 (dd, IH, Im-Ci7(H)-), 4.29 (dd, IH, Im-CH(7i)-), 4.05 (m, IH, -C#(OH)-), 3.38 (dd, 2H, -C# 2-NH-). 1 3 C N M R ( 7 5 MHz, CD 3 OD): 5 151.7 (-C=0), 129.1 and 128.2 ( Im-C 4 wj) , 66.9 (-CH(OH)-), 54.3 (-CH 2-NH-), 44.8 (Im-CH2-), 37.2 (-CBr 2H). 3.2.6.3 Synthesis of (22), a terminal nitrile derivative ofRSUllll This procedure was adapted from a synthesis reported by Fortier and McAuley describing the synthesis of a chelating ligand from an amine and an acrylonitrile.23'24 In a O 21 88 References on page 129 Chapter 3 50 mL flask, CH 2 C1 2 (5 mL) was added to RSU1111 (0.102 g, 0.531 mmol) to give a slurry. Acrylonitrile (5 mL, excess) and K 2 C 0 3 (0.150 g) were then added and the reaction mixture was refluxed for 24 h. TLC analysis revealed the presence of numerous products (~ 19). The two major bands observed were isolated via preparative TLC (band 1 (22), Rf = 0.33; band 2 (23), R f = 0.55 for CH 2 Cl 2 :MeOH, 20:1) to yield colourless oils (0.0136 g and 0.0334 g, respectively). IR (22, v, cm'1): 3376 (OH); 3139 (C-Hjm); 2929, 2847 (C-H); 2247 (CsN); 1631 (C-OH); 1487 (N-O a s y m ); 1363 (N-O s y m). X H N M R (200 MHz, ds-acetone, 22): 5 7.49 (s, IH, Im-i^), 7.09 (s, IH, Im-H4), 4.73 (dd, IH, Im-C#(H)-), 4.46 (dd, IH, Im-CH(//)-), 4.10 (m, IH, -C#(OH)-), 2.93 (td, 2H, -NH-Gr7 2-), 2.80 (dd, IH, -C#(H)-NH-), 2.71 (dd, IH, -CH(//)-NH-), 2.64 (t, 2H, -C# 2-CN); integrations support the addition of only 1 moi acrylonitrile per RSU1111, while peak assignments were made with the aid of 2D COSY; (d6-acetone, 23): 5 7.52 (s, IH, lm-Hs), 7.11 (s, IH, lm-H4), 4.83 (dd, IH, Im-Gr7(H)-), 4.60 (dd, IH, Im-CH(#)-), 3.98 (m, IH, -CH(0-CH 2-)-), 3.81 (dt, IH, -0-Cr7(H)-), 3.60 (dt, IH, -0-CH(i/)-), 2.80 (dd, IH, -C#(H)-NH 2 - ) , 2.76 (dd, IH, -CH(#)-NH 2-), 2.64 (dd, 2H, -Gr7 2-CN). 1 3 C N M R (75 MHz, ds-acetone, 22): 6 128.5 and 127.9 (Im-C4 a„d s), 100.6 (-CN), 69.7 (-CH(OH)-), 53.9 (-CH 2-NH-), 52.8 (-NH-CH 2-), 46.0 (Im-CH2-), 18.8 (-CH 2-CN). The reaction was repeated as described above, except E t 3 N (165 pL) was substituted for K 2 C 0 3 and the reaction was performed in neat acrylonitrile; the resulting white slurry became an orange solution after 24 h at reflux. The solvent was then removed and acetone was added to the residual oil to generate a tan-coloured, microcrystalline solid that was filtered off. The filtrate was taken to dryness to give a yellow-brown oil; TLC revealed 2 major bands, one of which was 22, that was purified via column chromatography (CH 2 Cl 2 :Me0H, 20:1). Two colourless oils corresponding to 22 (0.0409 g) and a new product (24) (0.0196 g) were isolated. LR-MS [DCI(+)]: 283 ( M + NH4 - CN), 267 ( M + H - CN), 240 (M - 2CN). *H N M R (300 MHz, d6-acetone): 6 7.50 (s, IH, Im-H5), 7.09 (s, IH, Im-H4), 4.81 (dd, IH, Im-C#(H)-), 4.44 (dd, IH, Im-CH(77», 4.19 (m, IH, -C#(OH)-), 3.00 (t, 4H, -N-C# 2-), 2.86 (m, IH, -C# 2-N-), 2.71 (t, 4H, -Cr7 2-CN); integrations support the addition of 2 moi acrylonitrile per 89 References on page 129 Chapter 3 RSU1111, while peak assignments made with the aid of 2D COSY. 1 3 C N M R (75 MHz, de-acetone): 5 128.4 and 127.9 (Im-C4 a„d s), 100.7 (-CN), 69.2 (-CH(OH)-), 58.0 (-CH 2-N-), 54.1 (-N-CH2-), 50.7 (Im-CH2-), 16.6 (-CH 2-CN). 3.2.7 Reactions of Nitroimidazoles with Bu4NF»H20 The syntheses of fluorinated nitroimidazoles via exchange of F" with other halogens on the side-chain were attempted, with the goal being to develop a useful procedure for obtaining fluorinated PET precursors. 3.2.7.1 Reaction of EBrl with Bu4NF»H20 EBrl (10 mg) and Bu 4 NF*H 2 0 (40 mg, 4.2 equiv.; this compound must be handled quickly in air as it is extremely hygroscopic) were added to a Schlenk tube and the system was evacuated at r.t. (~1 h). Dry d6-dmso (1 mL) was added via syringe and the mixture was stirred for 3 h; within 1 min the initially colourless solution immediately changed to blue, then green, green-brown, orange and finally to yellow-orange. T L C analysis revealed the formation of a new product that was isolated via preparative T L C (CH 2 Cl 2 :MeOH, 20:1) as a colourless oil (25). UV-Vis (dmso): 372, 264 nm. T i N M R (200 MHz, de-dmso): 7.58 (s, lm-H5), 7.01 (s, Im-#4), 5.11 (s, -C# 2-CO-), 4.15 (t, -CH2BT), 3.46 (t, -NH-C#2-), 2.11 (m, -C# 2-CH 2Br). 1 9 F N M R (188 MHz, d6-dmso): 5 -69.85 (s). Identical observations were made when the nitroimidazole used was EBrl(-l); however, when the fluoride source used was KF/2,2,2-Kryptofix with either EBrl or EBrl(-l) there was no blue intermediate observed. 90 References on page 129 Chapter 3 H F 25 3.2.7.2 Reaction of EBrl with Bu4NF«H20 and CH 3 C0 2 H The above procedure (see 3.2.7.1) was repeated except acetic acid (1.2 equiv.) was also used, its function being to perhaps stabilize the nitroimidazole ring and hence promote fluoride exchange on the side-chain (C-Br bond) 2 5 Again, the final solution was yellow, and TLC analysis revealed two bands that were isolated via preparative T L C (CH 2 Cl 2 :MeOH, 20:1). Band 1 (26, R f = 0.15) was isolated to yield a white solid, while band 2 (Rf = 0.55) was a colourless oil, its analytical data matching those of compound 25. Data for 26: L R - M S [EI]: 229 (IVf + H), 211 (TVf - OH), 182 (M* - N0 2 ) . TI N M R (200 MHz, de-acetone): 6 7.62 (br s, IH, -NH-), 7.40 (s, IH, lm-Hs), 7.00 (s, IH, Im-H4), 5.11 (s, 2H, -C7f2-CO-), 3.60 (t, 1H, -OH), 3.46 (q, 2H, -C# 2OH), 3.23 (q (dt), 2H, -NH-CrY 2-), 1.54 (p, 2H, -C# 2 -CH 2 OH); peak assignments made with the aid of 2D COSY. These data suggest that the presence of acetic acid leads to hydrolysis of the C-Br bond. H N02 26 3.2.7.3 Reaction of SR2508 with Bu4NF»H20 Reaction conditions identical to those for the reaction of fluoride with EBrl (section 3.2.7.1) were used, except the solvent was MeCN. The major band observed for TLC analysis was isolated to give a yellow oil that was purified using the C T R O N (CH 2 Cl 2 :MeOH, 10:1) to yield a colourless oil (27) (0.0252 g, 27 %). LR-MS[EI(+)]: 185 91 References on page 129 Chapter 3 (IVf - H), 167 (TVT - F), 137 (Ivf - F - CH 2 OH), 123 (TVT - F - CH 2 CH 2 OH), 108 (Im-CH 2CO-). IR (v, cm-1): 3403 (N-H); 2924, 2851 (C-H); 1735 (C=0); 1588, 1377, 1331, 1059; no v No bands were observed. UV-Vis (dmso): 370 (s was -70 % of that for SR2508). xrl N M R (200 MHz, d6-acetone): 5 7.00 (d, IH, lm-H5), 6.73 (s, IH, lm-H4), 4.77 (br s, IH, -NH-), 4.56 (s, 2H, -Cr7 2-CO-), 3.83 (m, 4H, -Cr7 2 OH + -NH-C# 2-), 3.52 (t, IH, -OH). 1 9 F N M R (188 MHz, d6-acetone): 5 -74.72 (s). 3.2.7.4 Reaction of « N 0 2 I m (« = 2, 4, 5) with Bu 4 NF«H 2 0 The title reaction was carried out in d6-dmso and the analytical data are summarized in Table 3.1. The results from this study were all obtained from the in situ reaction mixtures, and are used for the qualitative comparison of the data obtained from the reactions with fluoride in the previous sections (3.2.7.1-3). For the reaction with 2Me5N0 2Im, the major band from the TLC analysis was isolated [IR (v, cm"1): 3406 (N-H); 2959, 2872 (C-H); 1523, 1460, 1376, 1254, 1206, 1028, 878; no v N 0 bands at 1501 and 1380 cm - 1 were observed.] H F 27 92 References on page 129 Chapter 3 Table 3.1: Summary of the Tf NMR, 1 9 F{Ti} N M R and UV-Vis data for «N0 2 Im compounds and for their reaction with Bu4NF«H20 Sample T i N M R (5) 1 9F{TI} N M R a (6) UV-Vis (nm) Imidazole Positions l b 2 4 5 2N0 2 Im 14.40 ~ 7.35 7.35 ~ 328 2N0 2 Im + F ~ ~ 6.70 6.70 -68.95 374 4N0 2 Im 13.25 8.32 ~ 7.84 ~ 302 4N0 2 Im + F" ~ 7.70 — 7.04 -67.94 364 2Me5N0 2Im 12.95 2.32 8.20 ~ — 314 2Me5N0 2Im + F - — 2.15 7.60 — -67.80 376 a The 1 9 F signal observed in each spectrum is rather broad. b The N-H signal is not observed in the *H NMR spectrum of the reaction mixture because of the addition of H 2 0 with the fluoride source 0Bu4NF«H2O) which leads to rapid proton exchange with the N-H proton. 93 References on page 129 Chapter 3 3.2.8 Cyclic Voltammetry of Nitroimidazoles The cyclic voltammograms for the nitroimidazoles were obtained using the procedure described in Chapter 2 (Section 2.2.6) using the cell depicted in Figure 2-2 which contains the Ag reference electrode. The reduction potentials of the one-electron reduction of the nitro group are reported in Tables 3.2-3.4 versus the SCE, the standard reference used to report these values in the literature. The listed conversion factor is detailed in Chapter 2, Section 2.2.6. Table 3.2: Summary of reduction potentials for the 2-nitroimidazoles vs. SCE Compound FeCp2 Em Ep„ E p a : Ei/2 (avg.) Conversion Factor26 E l * vs. SCE (avg.) vs. Ag vs. Ag (424 - El/2 FeCp2) (mV) SR2508 479 -953,-1088:-1021 -56 -1077 EF5 408 -933,-1053:-993 +16 -977 EF3 336 -1060, -1163; -1112 +88 -1024 EF3(-1) 474 -910, -1020; -965 -50 -1015 EF2Br 471 -905,-1045;-975 -47 -1022 E=F2 520 -843,-1027;-935 -96 -1031 EF1 507 -899, -1009; -954 -83 -1037 EFl(-l) 620 -857, -981; -919 -196 -1115 ECU 565 -836,-954;-895 -141 -1036 ECll(-l) 595 -883,-1013;-948 -171 -1119 EBrl 489 -893,-1053;-973 -65 -1038 EBrl(-l) 650 -815, -965; -890 -226 -1116 94 References on page 129 Chapter 3 Table 3.3: Summary of reduction potentials for the 2-methyl-5-nitroimidazoles vs. SCE Compound FeCp2 E 1 / 2 E p c , E p a : E i / 2 (avg.) Conversion Factor26 E i / 2 vs. SCE (avg.) vs. Ag vs. Ag (424 - El/2 FeCp2) (mV) Metronidazole 393 -1039,-1207;-1123 +31 -1092 MF5 363 -1031,-1225;-1128 +60 -1068 MF3(-1) 379 -1026,-1212;-1119 +45 -1074 MFl(-l) 338 -1139, -1331; -1235 +85 -1150 MCll(-l) 504 -999, -1167; -1083 -80 -1163 MBrl(-l) 649 -856,-1008;-932 -225 -1157 Table 3.4: Summary of reduction potentials for the 2-methyl-4-nitroimidazoles vs. SCE Compound FeCp2 Ei/2 E p c , Epa: Ei/2 (avg.) Conversion Factor26 Ei/2 vs. SCE (avg.) vs. Ag vs. Ag (424-E 1/2FeCp2) (mV) 2M4NF5 575 -1078,-1220;-1149 -151 -1300 2M4NF3(-1) 628 -1060,-1220;-1140 -204 -1344 2M4NF1(-1) 629 -1134, -1286; -1210 -205 -1415 2M4NC11(-1) 660 -1090,-1216;-1153 -236 -1389 2M4NBrl(-l) 707 -1029,-1201;-1115 -283 -1398 95 References on page 129 Chapter 3 3.3 Results and Discussion 3.3.1 Synthesis of the Nitroimidazole Side-Chains Nitroimidazoles themselves have been shown to exhibit some activity in biological systems (e.g. 2N0 2 Im or azomycin is a natural antibiotic and antiprotozoal agent).27 However, because of their relatively low solubility in aqueous systems their use is limited. Various groups have reported the addition of numerous side-chains onto the N l position of the nitroimidazole ring 1 6 ' 2 8" 3 1 and the efficacy of such derivatives as hypoxia-selective cytotoxins or radiosensitizers seems to depend on the partition coefficient (P); as P increases the lipophilicity increases and thus the compound's ability to cross the cell membrane also increases.32 The major problem with most of these nitroimidazole compounds is that addition of the side-chain leads to increased cytotoxicity, specifically neurotoxicity which is the major factor that limits clinical dose.3 3'3 4 A relatively new compound with a pentafluorinated side-chain (EF5) seems to be one of the best candidates for use as a hypoxia-selective species.2'5 The original synthesis of EF5 (Section 3.2.2.2, synthesis 1) required a number of reaction steps that resulted in a relatively low yield (33 %), so a new method was developed in which the side-chain could be added in a single, high-yield step. The perfluoropropylamine side-chain for EF5 (section 3.2.2.2, synthesis 1) was originally synthesized via a 3-step reaction from perfluoropropionic acid (Scheme 3-1). 3 5 ' 3 6 The synthesis of compound 7 required two steps (see Scheme 3-9, p. 102) while the coupling of 7 and pentafluoropropylamine to form the amide linkage took place in the final Scheme 3-1: Synthesis of petafluoropropylamine from perfluoropropionic acid. step (total of six steps). In the present work, linking the amine with iodoacetic acid to form IF5 (Scheme 3-2), two of the steps are eliminated which gives a higher overall yield 96 References on page 129 Chapter 3 of E F 5 . 1 7 IF5 is easily purified via sublimation, however the C-I bond is mildly reactive H o r O N - M e I OH / \ / C F a W ^ \ ^CF 2 > + H 2 N ' ^ F f «HC1 5 " 1^  ' ' ~"CF 3 II ^A. ^ II IF5 Scheme 3-2: Synthesis of IF5, the precursor to EF5. and if the compound is not stored in the absence of light, it will slowly decompose. Changing the halogen from iodine to chlorine produces a more stable side-chain (C1F5), as the C-Cl bond is more resistant to cleavage by U V radiation.37 An interesting feature of C1F5 is that it is only visible on the TLC plate in the presence of U V light, behaviour opposite to that of EF5. The use of CIF5, although stable upon exposure to U V light (as monitored by *H N M R spectroscopy) did not, however, give a higher yield for the synthesis of EF5. Therefore, from a practical sense, C1F5 may be more useful than IF5. Kagiya and others28"30 have investigated reduction of the amide side-chain to the amine using the relatively mild reducing agent borane-THF which selectively reduces the amide3 8'3 9 without affecting the nitro moiety. The reduction of IF5 using BH 3»THF was successful in forming the corresponding amine (3) (see Scheme 3-3), but its stability was much lower than that of IF5. An N M R sample of the initially pale yellow solution of 3, stored at r.t. under ambient light, turned dark red-brown over a period of several days. TI N M R spectra recorded every 2 d revealed a decrease in the intensities of the signals for 3 and the appearance of new signals. Chromatographic separation of a 10-day old N M R sample (via preparative TLC) gave three bands with the first one assigned as compound 4; this was initially colourless but became yellow on the TLC plate upon exposure to U V light. Analysis of the *H N M R data indicated that 3 underwent cyclization to form compound 4 (Scheme 3-3). The instability of 3 is in accordance with the weak C-I bond; the loss of iodine is confirmed from the UV-Vis data which show that both I 2 and I3" are formed (presumably by the photolysis of HI). The liberation of I 2 from HI in the presence of U V light is well known. 8 ' 3 7 Of note, 4 has an aziridine ring analogous to that in R S U 97 References on page 129 Chapter 3 1069 (formed from the bromoethylamino derivative RB 6145). R S U 1069 is used as a D N A alkylating agent, suggesting that 4 may also serve the same purpose; however, because there is no nitroimidazole moiety attached 4 will not be hypoxia-selective. Scheme 3-3: Formation of aziridine ring upon exposure of 3 to U V light. Because of the interest in using halogen exchange reactions to form radiolabelled species for use in PET imaging of hypoxia,41 a monobromo derivate of IF5 was synthesized (IBr) (see Scheme 3-4). This compound is extremely reactive as it contains two reactive C - X (X = Br, I) bonds. IBr was only stable at < 10 °C in the dark, while exposure to heat and/or U V light resulted in decomposition to two new species (Scheme 3-4). The decompostion was monitored by X H N M R spectroscopy. The peak at 5 7.90 for IBr, typical of N H amides, disappears and a new peak appears at 5 2.51, in the range typical for cyclic amine N H protons.42 The other signals at 8 4.81 (a methylene triplet shifted slightly downfield due to the proximity of the C H 2 to an O atom), 8 3.74 (a + HI H 3 4 H IBr 1 2 Scheme 3-4: Cyclization of IBr in the presence of heat and/or UV-radiation. 98 References on page 129 Chapter 3 triplet adjacent to the amine N H , showing no coupling to the N H proton) and 8 2.37 (a pentet, presumably resulting from an overlapping triplet of triplets) together support the formation of a cyclic species with the proposed structure 1. Analogous cyclization products have been suggested in the literature, but no spectroscopic data has been reported to support these hypotheses.43'44 For the reaction of SR2508 with Tf 2 0 a similar cyclized product was observed in this thesis work, as confirmed by an X-ray structure (Figure 3-14, p. 123). The H-atoms adjacent to an amide linkage are more acidic (activated) because the amide exists in two tautomeric forms, while removal of electron density from the C-H bonds also increases their acidity. Although the bond energy of C-I is lower than that of C-Br, reactivity of the C-I appears to be limited by its proximity to the amide moiety. This prevents the formation of any cyclization products resulting from the cleavage of the C-I bond. The formation of a species corresponding to 2 (Scheme 3-4) is also supported by the J H N M R spectrum. Because the four-membered ring is torsionally strained this species is short-lived and eventually rearranges to 1. Most of the "side-chain compounds" were either purchased from Aldrich or received as kind donations from Prof. W. Dolbier at the University of Florida. The only fluorinated amine side-chain synthesized was 3-fluoropropylamine hydrochloride (5), the precursor for the synthesis of EF1. Obtaining 5 was imperative for comparison of the reaction products from the halogen exchange reaction between E B r l and F", and the authentic EF1 compound. A previously reported synthesis required four steps to obtain the parent amine of 5, and the yield was necessarily small (Scheme 3-5). 4 5' 4 6 For this thesis \ ^ / 0 H SOC12 F V Y C 1 NH 3 / x ^ ^ N H * II 77% * J 52% 0 , N H 2 < P t0 2 /H 2 F \ ^ a j _ P 20 5 18% 7 1 o / o Scheme 3-5: Synthesis of 3-fluoropropylamine reported by Pattison et al.45 99 References on page 129 Chapter 3 work, two different syntheses of 5 were attempted. Synthesis 1 used the fluorinating agent diethylamino-sulfur trifiuoride (Et 2NSF 3, DAST), a useful reagent for converting alcohols to their corresponding fluorides in high yield. 4 7 Previous results have suggested that DAST has less carbonium-type rearrangements and less dehydration side-reactions than occur with other fluorinating agents (e.g. H F 4 8 , SeF^pyridine49). Reaction of DAST with 3-hydroxy-propylamine did indeed yield the desired fluorinated product, 3-fluoropropyl-amine; however, other less volatile side-products formed which were not vacuum transferred from the reaction mixture. 3-Fluoropropylamine and diethylamine (as their HC1 salts) were the only two species detected in the T i N M R spectrum of the isolated solid. The presence of diethylamine (a side-product of the DAST reaction, Scheme 3-6) as the major product (Et 2 NH : 5 « 6) suggested that all of the DAST had reacted, but not DAST H 9 M HJSL O - F + HF + Et2NSF H 9 0 Et2NH»HF + SO, Scheme 3-6: Reaction of 3-hydroxypropylamine with DAST. selectively to form 5. An explanation for the low yield is that the HF formed during the synthesis of 5 produced the hydrofluoride salt of the fluorinated product, which remained in the solid residue after vacuum transfer of the volatile components. Unfortunately, the X H N M R spectrum of this residue was not obtained. The second method to synthesize 5 proved to be much more successful. Incorporation of a primary amino group onto an alkyl chain was easily done using the Gabriel synthesis;13 reaction of potassium phthalamide with the dihaloalkyl species, l-bromo-3-fluoropropane, produced the 3-fluoropropyl-phthlamide (6) by S N 2 substitution at the reactive C-Br bond (Scheme 3-7). The major advantage of the use of the phthaloyl group for protecting the nitrogen atom is its susceptibility to facile removal by means of hydrazine.14 Another advantage of this 100 References on page 129 Chapter 3 reaction is that the amine can be easily separated by extraction with H C 1 from the phthaloyl hydrazide formed during the hydrazinolysis. In this case, the hydrochloride salt was actually the desired product, 5 (Scheme 3-7). One other synthesis of 5 exists in which the reaction intermediate is the azide species,10 although the yield is much lower than via the chemistry illustrated in Scheme 3-7. O Scheme 3-7: High yield Gabriel synthesis of 3-fluoropropylamine hydrochloride. 3.3.2 Nitroimidazoles Nitroimidazoles are useful compounds for targeting the hypoxic regions of cancerous tumours; the nitro moiety is readily reduced in the presence of nitroreductases within the cells to form macromolecular adducts. Various nitroimidazoles (2, 4 and 5 N 0 2 derivatives) have been synthesized previously for use as radiosensitizers50"52 and their efficacy was up to three times greater in areas of hypoxia. Because the nitroimidazole, EF5, was found to be non-toxic when compared to other compounds of similar composition,2 the focus of this thesis work was on variation of the fluorinated side-chain and the position of the nitro group about the imidazole ring. Most of the nitroimidazoles were synthesized via the mixed anhydride species generated from reaction of an acid moiety with isobutylchloroformate (Scheme 3-8). The carbonyl group nearest the imidazole is the more reactive of the two, the isobutyl carbonate being the better leaving group on reaction with an appropriate R i N H 2 amine, 101 References on page 129 Chapter 3 this leading to formation of an amide bond. N-Methylmorpholine (NMM) was used to "mop up" free ET in solution (Scheme 3-8). /R5 , O H Y R2 H ( X ^ O . Y o + R2=N02, R4=R5=H R5=N02, R2=Me, R4=H R4=N02, R2=Me, R5=H H 2 NRi -HCI yR5 R 2 o o MR1 (Ri= halogenated side-chain) 2M4NR-I Scheme 3-8: Reaction mechanism for addition of side-chain via an amide linkage. Synthesis of the nitroimidazole acetic acid derivatives was achieved by chemically altering existing nitroimidazoles. For 2-nitroimidazole derivatives, the hypoxia selective radio sensitizer SR2508, obtainable in large amounts from Dr. C. Koch (Univ. of Perm.), was treated with FfTMeOH. This treatment cleaved the amide bond to yield the methyl ester (8) which was subsequently worked up with NaOH/HCl to yield the corresponding acid (7) (Scheme 3-9). The major problem with this procedure is that OH Dowex (H+) MeOH 1 N O 2 0 OMe N . NaOH HCI N 0 2 ° OH SR2508 Scheme 3-9: Synthesis of 7, via 8 from SR2508. 102 References on page 129 Chapter 3 increasing the acidity of the solution (with HC1) to a pH of ~4, to ensure the formation of 7, results in concomitant protonation of the imidazole ring at N3. Subsequent extraction with EtOAc thus produced 7 in significantly low yield because most of it remained in the aqueous layer as the hydrochloride salt (as judged by TLC and T i N M R spectroscopy). The protonated species proved to be useable for the synthesis of the desired halogenated nitroimidazoles, the procedure (Scheme 3-8) requiring an additional equivalent of N M M ; however, the yields were typically half of those obtained when the pure acid was used. For the 2Me5N02lm compounds, the hypoxia-selective cytotoxin metrodinazole, which is used in the treatment of anaerobic bacterial infections,53 was oxidized with Jones' reagent converting the alcohol to the acid functional group (Scheme 3-10). The yield obtained was slightly higher than reported20 and, according to TLC, some of the metronidazole was only partially oxidized to the aldehyde species. Of particular note, when additional Jones' reagent was added to convert the remaining aldehyde to the acid, a lower yield was obtained. NO 2 OH Me N N 0 2 H Me N O , •Tone's Reagent / \ OH acetone, r.t Me Metronidazole 13 Scheme 3-10: Oxidation of metronidazole with Jones Reagent to yield the carboxylic acid derivative, 13. For the 2Me4N02lm compounds, the acid species had an extra methylene in the link between the imidazole ring and the acid functional group. Treatment of 2-methyl-4-nitro-l-imidazolepropionitrile (Aldrich) with sulfuric acid resulted in hydrolysis of the nitrile to the carboxylic acid after several days at high temperature. Formation of the amide is facile, nevertheless, further hydrolysis to the acid requires relatively harsh conditions, mainly because of the extra resonance capability of the nitrogen lone-pair (Scheme 3-11). 103 References on page 129 Chapter 3 R — C = N H 3 0 R' NH X N H O H 2 © Tautomer of amide R N H 2 X OH H,0 R-O - N H 2 O H 2 © • H N H 4 + O x R " " ^ O H carboxylic acid © OH OH OH N H 3 + R ^ ^ O H R-© -NH3 OH + H R- - N H 2 OH Scheme 3-11: Acid hydrolysis of 2-methyl-4-mtro-l-imidazolepropionitrile (abbreviated as RCN). The title compound (EF5) on which most of this work was based on has been approved for clinical trials by the National Cancer Institute (US) through support of large scale synthesis, formulation and toxicological studies. The improved synthesis (Scheme 3-12)1 7 of EF5 from IF5 will be useful, especially if the compound is approved for use in NO, O I F 5 CF, Cs 2 CQ 3 DMF, 5CPC CF-2^  T N 0 2 CF, O E F 5 Scheme 3-12: New synthetic route to EF5. humans as either a radiosensitizer or a hypoxia-selective imaging agent. Other 2-nitroimidazole compounds in this series (EF4Br, EF3, EF3(-1), EF2Br, EFl(-l), ECU, ECll(-l), EBrl, EBrl(-l), EIAA, EPrA) were all isolated in relatively high yield (up to 83 %) and could 'be synthesized via the alternate route (i.e. synthesis of the complete side-chain via reaction of the appropriate amine with iodoacetic acid, followed by coupling with 2N02lm, see Scheme 3-2 and Scheme 3-12) with the exception of the bromo derivatives which are too reactive (see Section 3.2.1.2). The high reactivity of the bromo derivatives is also indicated by their poor elemental analyses as the C-Br bond is readily hydrolyzed by H2O (in the air). 104 References on page 129 Chapter 3 The monofluorinated species (EF1, see Scheme 3-13) was the only 2-nitroimidazole compound in the series that could not be isolated in a yield higher than 15 %. The required monofluoropropylamine hydrochloride (or hydrofluoride) reagent was extremely hygroscopic which may explain why 2-nitro-l-H-imidazol-1-yl-ethyl isobutanoate (11) is the major product (Scheme 3-13). The presence of H 2 0 in this reaction could result in the formation of isobutylcarbonate via "reaction with isobutylchloroformate. The isobutylcarbonate would also be formed as a co-product of the reaction between 3-fluoropropylamine and isobutylmethylimidazoylcarbonate. The isobutylcarbonate could lose C 0 2 to form isobutanol, which is the next strongest nucleophile present after the amine. Because of the formation of HCI the amine would remain as its hydrochloride salt, thus decreasing its nucleophilicity and therefore allowing the isobutanol to react with the acyl carbonate group to form 11. The synthesis of ECll(-l) via reaction of TsCl with SR2508 was unexpected. The desired product was the tosylate species that could be used as a reactive intermediate for the formation of EFl(-l) (Section 3.2.7). Unfortunately, the tosylate could not be isolated because the chloride ions readily react to form the monochloro species ECll(-l). The reaction of phenylmethanesulfonyl fluoride using the same reaction conditions did not yield the sulfonate intermediate and consequently the desired EFl(-l) species was not formed. The highly reactive tosylate intermediate was also exploited for the synthesis of compound 16, the chlorinated derivative of metronidazole (Section 3.2.3.7). The ability to alter the biological activity of a compound by changing the composition of an associated side-chain prompted the attempted synthesis of a pentafluoro species analogous to EF5, but with the amide linkage reversed (Scheme 3-14). Because of the difficulty in forming the reverse amide linkage for RevEF5, three different syntheses were attempted. Synthesis 2 uses conditions identical to those of synthesis 1 for EF5 (Scheme 3-8), the major difference being that the pentafluoro group is introduced as pentafluoropropionic acid. This acid species is highly reactive with H 2 0 from the air, as apparent from substantial fuming when the sample bottle is opened, but does readily react 105 References on page 129 Chapter 3 E F l Scheme 3-13: Formation of the isobutylester (11) in the presence of F£20. with /'so-butylchloroformate to form the mixed anhydride. The problem with this reaction arises when 10 (the 2-nitroimidazole ethylamine product from synthesis 1, step 2) is added to the reaction mixture. The amine preferentially reacts with the carbonyl group associated with the isobutylformate to yield a species similar to 11 as the major product (as shown by X H NMR), analogous to that formed during the synthesis of E F l (see Scheme 3-13), although a different mechanism must be involved. This result can be explained accordingly: the electron-withdrawing fluorine atoms stabilize the negative charge of the pentafluoropropionate, making it a better leaving group than isobutylcarbonate. To avoid the problem encountered in synthesis 2, the peptide coupling agent STMl>BF 4 was reacted with pentafluoropropionic acid (synthesis 3). Addition of base (Et 3N) in the first step leads to formation of tetramethylurea and (N-succinimidyl)-pentafluoropropionate (according to TLC data). Once again, addition of 10 results in a reaction; however, neither of the major products was RevEF5, which was observed in low 106 References on page 129 Chapter 3 concentration via TLC. Synthesis 1 allowed the isolation of RevEF5 in high yield using two different synthetic approaches for the third step of the reaction (steps 3 a and 3b). The first route, a well known procedure for the formation of an amide bond using D C C (synthesis 1, step 3 a) was employed, leading to the formation of a reaction intermediate analogous to that in synthesis 3. This reaction (29 % yield), however, was more successful than synthesis 3 (< 1% yield), and may be attributed to the greater stability of the side-product formed, dicyclohexylurea (DCU). The presence of BF 4 " in synthesis 3 may also hinder the reaction although it is not obvious how. An advantage of synthesis 1 (step 3 a) is that the D C U (because of its low solubility in DMF) can be readily separated from the product mixture via filtration. For the second route (step 3b), the amine on the side-chain of 10 (nucleophilic) was reacted with the highly reactive perfluoroanhydride (CF 3 CF 2 CO) 2 0 to give RevEF5 in a somewhat higher yield (32 %). P S ^ 1 P F = \ s t e P 2 = © 0 / \ / \ / \ . N H 3 CI T 2)6MHC1 Y N 0 2 N 0 2 10 Step 3 a H 1) DCC, NHS, C F 3 C F 2 C 0 2 H I \ x ^ \ ^ N . ^.CFx 2) NMM y Y CF3 NO O 1)NMM 2 RevEF5 ' 2) (CF 3CF 2CO) 20 Step 3b Scheme 3-14: The 3-step synthesis of RevF5 from 2-bromoethylphthalamide. Another 2-nitroimidazole species with a side-chain terminating with an amino group is RSU1111. This compound contains a side-chain one methylene unit longer than in 10, and also contains a secondary hydroxyl group similar to that of the radiosensitizer misonidazole.54 Analogous to synthesis 2 of RevEF5, the major product formed was the isobutylformate-linked amide species, while 20 was isolated in low yield (5 %). The incorporation of the pentafluoropropyl acyl group is confirmed by HR-MS and more importantly, by 19F{JH} N M R spectroscopy, which gives signals at 8 -6.77 and -46.35 107 References on page 129 Chapter 3 corresponding to the CF3 and CF 2 groups, respectively (cf. RevEF5: 8 -6.79 and -46.43). In the *H N M R spectrum the protons on the methylene groups adjacent to the chiral -CH(OH)- centre are all magnetically inequivalent (diastereotopic protons). An interesting aspect of this spectrum is that the -CH2-NH- protons have essentially the same chemical shift while the Im-CH2- proton chemical shifts are well separated (AS 0.40). This may suggest that the O H group on the adjacent carbon is interacting with one of the protons of the methylene group nearest the nitroimidazole. The same chemical shift distribution is seen in the TI N M R spectrum of 21; the lm-CH2- proton chemical shifts have nearly the same separation as above (A5 0.38). The observation of a signal at 8 6.21 is in accordance with the HR-MS results which suggest that one of the Br-atoms of the C B r 3 C 0 2 H precursor has been displaced by a hydrogen. The anticipated product was the tribromo species, but presumably because of the bulkiness of the Br-atoms and the relatively weak C-Br bond, the dibromo species is isolated. The 1 3 C N M R spectrum of 21, compared with that of RSU1111, has extra signals at 8 151.7 and 37.2, confirming the presence of -C=0 and - C B r 2 H units. The coordination chemistry of nitroimidazoles is also of interest because these do not typically form strong interactions with the metal centres of interest (e.g. Pt, Ru), unlike imidazoles which form more stable species.55'56 The metal complexes may be used as nitroimidazole delivery agents to the D N A by coordinating to the D N A s nitrogen bases, hence bringing the nitroimidazoles in close proximity to the site of action.57 Synthesis of a nitroimidazole with a side-chain containing a chelating moiety would ensure stronger interaction with the metal centre, while maintaining the functionality of the hypoxia active nitroimidazole. Reaction of acrylonitrile with an amine is a useful method for adding two C H 2 C H 2 C N groups to the existing amine, followed by reduction of the nitriles to yield the tridentate species R N ( C H 2 C H 2 C H 2 N H 2 ) 2 . 2 3 ' 2 4 Addition of acrylonitrile to RSU1111 proved to be much more difficult than anticipated. Even when the reaction was done in neat acrylonitrile, the major product was the mono-substituted species 22, while the bis-substituted species (24) was formed only half as much. A small quantity of a compound corresponding to the additon of acrylonitrile at the hydroxyl group (23) was 108 References on page 129 Chapter 3 also isolated and characterized by TI N M R spectroscopy. An interesting property of compounds 22 and 23 is that not only do they exhibit inequivalent methylene *H signals for the C H 2 bound to the 2N02lm moiety, but they also have inequivalent methylene proton signals for a C H 2 bound to the N H and N H 2 group, respectively. This is not observed for compound 24. The difference between the chemical shifts of the two protons is not as pronounced for the methylene group bound to either the N H or N H 2 groups (e.g. 5 2.80 and 2.71 for 22 vs. 5 4.73 and 4.46 for the 2N0 2Im-Gr7 2- protons), suggesting a weaker intramolecular interaction. This finding is puzzling and tends to suggest that the newly added propylnitrile arm in 22 has a conformation which interacts with one of the methylene protons nearest to the N H group, consequently inducing magnetic inequivalence. The presence of the two pendant arms in 24 possibly makes the species bulky enough to prevent such an interaction. Reduction of the nitrile groups of 24 was unsuccessful using a number of different reducing agents (e.g. L A H , NaH, H 2 ) including B 2 H 6 , which has been reported to reduce C N groups in the presence of an N 0 2 group.2 9 In all cases, the nitro group was reduced, either partially or completely, making further pursuit of this chelating compound futile. Comparison of the lH N M R spectra for the 2-nitroimidazole compounds with variation in number and type of halogen atoms present reveals some interesting results. For the most part, for each of these compounds the chemical shifts of the protons of the 2N0 2 Im-CH 2 - entity fall in the same range, displaying that the composition of the side-chain has little effect on this component of the molecule (lm-H5: 5 7.49 —» 7.66; Im-H*. 8 7.08 -> 7.21; Gr72CO: 8 5.15 5.36). There is a larger variation for the chemical shift of the N H proton (8 7.52 -» 9.05) as its position shifts with changes in concentration of H 2 0 in the N M R solvent due to proton exchange between the two species.42 Removal of the N0 2-group, however, does effect the chemical shifts of the imidazole protons and C H 2 bound to the imidazole at N l . For ImF5, the compound identical to EF5 less the nitro group, the chemical shifts are significantly upfield of those for EF5 (e.g. Im-H^. 8 7.66 (EF5) -> 7.16 (ImF5)). The electron-withdrawing N0 2-group decreases electron density within the imidazole ring. The *H N M R signals for the C H 2 adjacent to the CF 2CF3 group 109 References on page 129 Chapter 3 for these two compounds are not effected by the presence or absence of the N0 2 -group on the imidazole ring. The chemical shift(s) of the C H 2 group(s) in the side-chain bound to the amide N H are dependent on the atoms on either side of them. For instance, the chemical shift for the C H 2 of E F 5 (bound to N H and CF 2 ) is 4.06 ppm, which matches exactly that of EF3(-1) whose only difference is that the CF 2 group is now CF 3 . Comparison of the 8 values for E F 5 versus E F 3 and EFl shows an upheld shift for the -NH-C# 2 - moiety (8 4.06 —» 3.50 —» 3.38, respectively), exhibiting the effect of the proximity and number of electonegative F-atoms. The absence of halogen atoms on the side-chain (e.g. as in EPrA) results in a 8 of 3.20, even further upfield. The most pronounced effect of the fluorine atom can be seen for EFl (Figure 3-4) and EFl(-l) (Figure 3-5) as the protons for CH 2 F are shifted downfield to 8 ~ 4.50. This group is also easily identified in the *H N M R spectrum as a doublet of triplets due to the large 2JHF value (66 Hz and 48 Hz, respectively). The adjacent C H 2 also has a strong interaction with the fluorine atom, as indicated by 3JHF values (42 Hz and 30 Hz, respectively). The N H signal for EFl is not observable at the magnification presented in Figure 3-4; however, the signal is present at 8 7.64 if the vertical scale is expanded. The N H signal is more pronounced for the EFl(-l) spectrum at 8 7.87 (Figure 3-5) and may be related to the increased amount of H 2 0 evident in this spectrum; the broad signal is a result of slow exchange of the N H proton, and the electrical quadrupole moment of the nitrogen nucleus that induces moderately efficient spin relaxation.42 Comparison of the compounds with identical composition, except for variation in the halogen atom terminating the side-chain, demonstrates the direct correlation between electronegativity and chemical shift. The 8 values for the C H 2 X protons (EFl: 8 4.51; ECU: 8 3.66; EBrl: 8 3.50) decrease almost proportionally to the decrease in electronegativity (EN) of the X atom (EN for F, CI and Br is 4.0, 3.0 and 2.8, respectively); the trend is as expected. The same effect is observed for the EXl(-l) compounds. Analysis of the 1 3 C N M R spectra for these compounds for the C atom 110 References on page 129 Chapter 3 directly bound to X also exhibits the dependence of 6 on E N ( E F l ( - l ) : 5 84.70; E C l l ( - l ) : 5 43.46; E B r l ( - l ) : 5 37.88). N 0 2 b a .. .. < acel f d 11 1 , one e ' 1 1 1 1 1 1 ' 1 1 1 1 1 1 1 1 1 1 ' 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 ' ' 1 . . . 1 . J 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 3.0 8.5 6.0 7.5 7.0 G.5 6.0 5.5 5.0 4.5 4.0 3.5 3.0 2.5 2.0 1.5 1.0 0.5 ppm Figure 3-4: *H N M R spectrum of E F l in d6-acetone . d N 0 2 b a d C H; f e 11J „ i U l J ace 4 , J one . . . | | . . . . | . . . . | . . . . | . . . . | . . | . . . . | . . . . . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | . . . . | .. . 3.0 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 1.5 1.0 0.5 ppm Figure 3-5: ^ N M R spectrum of E F l ( - l ) in d6-acetone. Analysis of the 1 9 F N M R spectra for the fluorinated compounds reveals some interesting and unpredictable results. For example, comparison of the chemical shifts of 111 References on page 129 Chapter 3 the CF 3 groups of EF5 and EF3, 5 -4.34 and 10.84 respectively, shows results opposite to those expected. The signal for EF5, which has an electron-withdrawing CF2-group adjacent to the CF 3 , is further upheld than for EF3 which has a C H 2 bound to the CF 3 . Also surprising is the downfield shift of the fluorine signal for EF2Br (5 32.71) when compared to that for EF3. The two compounds are identical with the exception of substitution of one of the F atoms by Br. The lower electronegativity of Br should have shifted the signal further upfield; however, the opposite is true. The C F 2 of EF5 comes at 8-41.31 while the CH 2 Fhas a signal at 8 -142.50 showing that as the number of F-atoms on a single carbon decreases the signal is shifted further upfield, following the expected electronegativity trend. Examination of the X H and 1 9 F N M R spectra for the 4-nitro- and 5-nitroimidazole compounds reveals the same trends observed for the 2-nitroimidazole species. In the T i N M R spectra, the closer proximity of the nitro group to the imidazole proton leads to a downfield shift of the signal (e.g. MF5: Im-H4 8 7.93 vs. EF5: Im-H4 8 7.21). For both MXl(-l) and 2M4NX1(-1) (where X= F, CI and Br), chemical shift differences are observed analogous to those for the EXl(-l) compounds, following the trend that increased E N leads to a downfield shift in the C H 2 X proton signal. Other analytical techniques used to characterize these compounds included mass spectrometry, and UV-Vis and IR spectroscopies. The UV-Vis spectra did not reveal much information other than the presence of a nitro group on the imidazole ring (the energy band of the major chromophore was very destinctive of the nitro group position; 316, 310 and 300 nm for 2N0 2 , 5N0 2 , and 4N0 2 , respectively). Mass spectrometry was used to characterize the compounds for which acceptable elemental analysis results were not obtainable. For the most part, the compounds with the brominated side-chains were problematic due to their highly reactive C-Br bond. The MS spectra typically contained a very weak parent peak and a base peak which corresponded to Wf - N 0 2 , suggesting that the bond between the imidazole ring and the nitro group is easily cleaved. IR spectroscopy, next to N M R spectroscopy, was the most diagnostic tool for characterizing the nitroimidazole compounds. The major bands of interest were the stretching bands for 112 References on page 129 Chapter 3 the N - H and C=0 groups of the side-chain and the N - 0 of the N0 2-group associated with the imidazole ring. A comparison of the IR data for these compounds shows that there is some electronic influence, albeit weak, from the substituents of the side-chain on the N - H and C=0 stretching frequencies. For the most part the N0 2-group was too far removed from the substituents on the end of the side-chain to be affected. The asymmetric and symmetric VNO values were two of the more dominant bands observed in the spectrum of each nitroimidazole and typically spanned the ranges, VNOasym (1484 - 1506 cm"1) and VNOsym (1361 - 1400 cm"1). The 4 N 0 2 group had bands at slightly higher wavenumbers than those of the 2 N 0 2 group, while the values of the 5 N 0 2 group were lower. These findings may result from the proximity of the N0 2-group to the side-chain. The 4 N 0 2 group is the furthest removed and therefore the possibility of interaction with the side-chain via H-bonding is limited. A comparison of the 2N0 2 Im compounds EF1, ECU and EBrl (each with -C(0)-NH-(CH 2 ) 2 CH 2 X moieties) reveals virtually no change in VNH or Vco with variation in X . When two CH2-groups (vs. three) separate the halogen from the amide functionality, VNH and vco both are shifted to higher values (e.g. EFl(-l): vco 1669 cm"1 vs. EF1: v c o 1660 cm"1; see Figure 3-6). Figure 3-6: Typical IR spectrum of a 2-nitroimidazole, EF1. For the three pentafluorinated compounds EF5, MF5 and 2M4NF5, X-ray structures were obtained (Figures 3-7 to 3-9, respectively). These structures may prove useful for modelling the active binding site of the MoAb used for the imaging of hypoxia 113 References on page 129 Chapter 3 (see Chapter 6). E F 5 crystallized in the primitive monoclinic space group Pc, and all bond lengths and bond angles within the structure are in the normally expected ranges (Appendix I - l ) . 5 8 The packing of the E F 5 molecules within the unit cell results in a layered effect with inter- and intramolecular H-bonding stabilizing the lattice. The intermolecular interactions include H-bonding between H( l ) and 0(1), and N(2) with either F(5) or H(7), alternating every second molecule. The intramolecular interactions are stonger than the intermolecular interactions (-2.8 A vs. 3.0-3.3 A), and are observed between H(5) and 0(2), involving the nitro group, and H(6) and 0(1). Figure 3-7: ORTEP view of E F 5 ; 33 % probability thermal ellipsoids are shown. M F 5 crystallized in the primitive monoclinic space group P2i/c and the bond lengths and bond angles (Appendix 1-3) were similar to those observed for E F 5 . Also, 114 References on page 129 Chapter 3 similar to EF5, intramolecular H-bonding interactions are observed between H(6) and 0(2) and H(9) and 0(1) (2.49 and 2.39 A, respectively). The intermolecular interactions, however, are much stronger and are observed at different points between the molecules, H( l ) and N(3) and H(7) and 0(1) (2.11 and 2.46 A, respectively). H(3) N(3) Figure 3-8: ORTEP view of M F 5 ; 33 % probability thermal ellipsoids are shown. 2M4NF5 crystallized in the primitive monoclinic space group P2i/n, and the bond lengths and bond angles (Appendix 1-4) were also comparable to those of E F 5 5 8 Disorder was observed for the CF 2 and CF 3 groups. For this molecule only a single intramolecular interaction is observed between H(10) and 0(1), similar to those found in both EF5 and M F 5 . The nitro group is at the 4-position on the imidazole ring and is far enough removed from the side-chain that no H-bonding interaction is observed for the 115 References on page 129 Chapter 3 Figure 3-9: ORTEP view of 2M4NF5; 33 % probability thermal ellipsoids are shown. nitro moiety. Intermolecular H-bonding is observed once again between H( l ) and N(2), and a new interaction between H(3) of the methyl group and F(l) is integral in the arrangement of the 2M4NF5 molecules within the lattice. 2M4NF1(-1) (Figure 3-10) crystallized in the same space group as 2M4NF5. The F( l ) atom for 2M4NF1(-1) falls in the same position as F(l) of the C F 2 group of 2M4NF5 and again undergoes intermolecular H-bonding with the methyl group on the imidazole ring (Appendix 1-5). 116 References on page 129 Chapter 3 Figure 3-10: ORTEP view of 2M4NF1(-1); 33 % probability thermal ellipsoids are shown. 3.3.3 Electrochemistry of Nitroimidazoles The investigation of the use of nitroimidazoles as radiosensitizers and hypoxia-selective cytotoxins has clearly verified the correlation between the efficacy of these compounds and the one-electron reduction potential of the nitro moiety.59"62 The nitro group undergoes a series of one-electron reduction steps to finally yield the amine (Scheme 3-15). The extent to which the nitro group is reduced within a given biological system will be determined by the nature of its enzymatic activation.63 The initial steps in 117 References on page 129 Chapter 3 hv '2* °2 Radical Anion Nitroso Amine Scheme 3-15: Stepwise reduction of 2-nitroimidazole to 2-aminoimidazole. the reductive metabolism of 2-nitroimidazoles are mainly catalyzed by cytochrome P450 reductase and to a lesser extent by cytochrome P 4 5 0 , both of which are found in the endoplasmic reticulum.64'65 The stability of the radical anion depends on the prototropic properties of the compound and pH; as the pH level decreases below physiological values the amount of D N A damage observed during radiation increases.66 Because the hypoxic cells are low in oxygen concentration, the pH is lowered slightly because of the increased concentration of C 0 2 from cellular respiration (Figure 3-11). H2O + co 2 WOs - Hccv +[iP| Figure 3-11: A lowering of pH levels in hypoxic cells due to increased [C0 2 ]. Nitroimidazoles can enter most cells of a living organism where the nitro group is reduced enzymatically. In oxic cells (high [O2]), the reduced radical anion species is readily oxidized back to the N 0 2 moiety as the reduction potential of 0 2 is higher (more positive) than that of the nitroimidazole, this resulting in a futile cycle (at least for effective radiosensitization).67 Hypoxic cells, however, have significantly reduced [0 2] and lead to binding of the reduced species to macromolecules within the vicinity of the metabolism site. Importantly, there is evidence to suggest that the reactive metabolites do not migrate out of the cells.68 118 References on page 129 Chapter 3 Prediction of the efficacy of nitroimidazoles as radiosensitizers or as HSCs is done in part by measuring the one-electron reduction potential, the so-called "E 7 " , of the nitro moiety by pulse radiolysis in neutral, buffered, aqueous media.69 This initial reduction step activates the molecule and is the key to initiation of its action within hypoxia, although the active species may be one of the subsequently reduced species.62 For nitroimidazoles, the ideal reduction potential for the first reduction step falls in the range E 7 -300 to -450 mV vs. N H E . 6 9 The reduction potentials measured by cyclic voltammetry in M e C N (E 1 / 2) cannot be compared directly to E 7 values, but a linear correlation has been observed between the two and can be used to approximately convert from Ey2 to E 7 and vice-versa (E 7 « El/2 - 241 mV). 7 0 ' 7 1 The electrode system used to measure the one-electron reduction potential of the nitroimidazoles from this thesis work included a Pt working and counter electrode and Ag wire (Ag/AgO) as the reference electrode. The amount of oxide deposited on the Ag wire varies from one run to the other as demonstrated by the inconsistent measurements for the ferrocene/ferrocenium redox couple (range: 330 to 660 mV vs Ag, Tables 3.2 to 3.4). To compensate for this problem, FeCp 2 was added as an internal standard for the measurement of the cyclic voltammogram of each nitroimidazole (there is a large separation between the waveforms for FeCp 2 and N 0 2 l m ; see Figure 3-12). The measured E1/2 value was then converted to a potential versus the SCE. Analysis of three separate solutions of the same nitroimidazole produced Ei/ 2 values within ± 3 mV, confirming the accuracy of this technique. Of note, the reduction potentials are measured in M e C N and so do not correlate with those reported in aqueous solution. Values can be compared to those of the known 2-N0 2Im, SR2508, whose reduction potential is known in aqueous solution (E 7 = -388 mV, vs. N H E ) 6 9 and in MeCN (Ei/2 = -1077 mV vs. SCE, this thesis work, Table 3.2). There is significant variation between the C V plots for the different nitroimidazoles. For the most part, the nitroimidazole redox couple is quasi-reversible as the peaks shift further apart with increasing scan rate. The cathodic peak current is 119 References on page 129 Chapter 3 typically not equal to the anodic peak current (Figure 3-12), a requirement of electrochemical reversibility, implying that the electron transfer to the nitroimidazole and 3.00E-04 -2.00E-04 -5- E F l 1.00E-04 -c , SI / O 1 \f- 0.00E 1 00 FeCp 2 -2.00 _ ^ - a o o y 0. / -1.00E-O4 --2.00E-04 -Voltage (V) )0 Figure 3-12: Quasi-reversible C V plot for E F l referenced to FeCp 2. back to the electrode is not completely reversible. The potential difference between the cathodic (EpC) and anodic (Ep a) peaks was considerably higher than 59 mV, also suggesting the absence of reversibility. The 4N0 2 Im and 5N0 2Im compounds exhibit even less reversible behaviour as the Ep C and E p a peaks are further separated from one another (although a slower scan rate has little effect on the separation) and the difference between the peak currents is even more pronounced than in the 2N0 2 Im systems (Figure 3-13). Analysis of the reduction potentials of the nitroimidazoles (Tables 3.2 to 3.4) leads to a number of conclusions. Most importantly, the E i / 2 value decreases noticeably as the position of the nitro group is changed from 2 N 0 2 (-977 to -1116 mV) to 5 N 0 2 (-1068 to -1163 mV) to 4 N 0 2 (-1300 to -1415 mV) on the imidazole ring, that is, the electron affinity decreases as the nitro group changes from the 2 —» 5 —• 4 positions. This trend is, for example, readily apparent for the pentafluorinated species EF5, M F 5 and 2M4NF5 which have E i / 2 values of -977, -1068 and -1300 mV, respectively. A literature example of 120 References on page 129 Chapter 3 6 . 0 0 E - M T FeCp 2 < -2.00 MCll(-l) 1.00 -4.00E-O4 - 1 V o l t a g e ( V ) Figure 3-13: C V plot for MCU(-l) referenced to FeCp 2. this effect is seen with the comparison of the reduction potentials of the isomeric nitroimidazoles RGW-601 (2N0 2ImR), nimorazole (5N0 2ImR) and RGW-611 (4N0 2ImR, where R= -CH 2 CH 2 N(CH 2 CH 2 ) 2 0) ( E 7 1 values of-390, -457 and -554 mV, respectively).52 Adams et al. reported that a comparison of E i / 2 values within a given group reveals that the constitution of the side-chain has only a small effect on the reduction potential of the nitro group; the electron-withdrawing groups are sufficiently distant from the imidazole ring that little inductive effect is incurred on the nitro group.5 2 However, for EF5 and the other nitroimidazoles containing more than one F-atom in their side-chains this suggestion appears to be false as there appears to be an influence on the E i / 2 values (cf. E i / 2 for SR2508, -1077 mV). EF5 has the highest reduction potential at -977 mV, as expected, while successive removal of fluorine atoms leads to more negative E i / 2 values (e.g. EF3, -1024 mV). Surprisingly, for the monohalogenated compounds (EX1 and EXl(-l), where X = F, CI and Br) the reduction potential is significantly decreased (by ~ 80 mV) when one methylene is removed. The C V data imply that pursuit of 2N0 2 Im compounds with longer side-chains and more fluorine atoms (i.e. EF5) for use as hypoxia-selective bioreductive agents will be most advantageous. 121 References on page 129 Chapter 3 3.3.4 Incorporation of Fluorine into an Existing Side-chain The imaging and, more importantly, quantification of hypoxia using nitroimidazoles have been demonstrated recently using a fluorescently labeled monoclonal antibody which recognizes a bioreduction adduct of a F-containing nitroimidazole, in this case EF5, within hypoxic cells.5'7 2 The problem with this technique is that it is invasive, requiring a biopsy to obtain samples of the cancerous tissues. Therefore, utilization of a technique for which hypoxia can be imaged non-invasively would be extremely advantageous. A non-invasive technique, known as positron emission tomography (PET), does exist and is used to study, for example, metabolism of dopamine (labeled with the positron-emitting isotope 1 8 F) within the heart.73 Consequently, if the chemistry can be developed to incorporate 1 8 F into one of the existing nitroimidazoles synthesized in this thesis work, the new species could be used to image areas of hypoxia non-invasively. The limiting factor for this chemistry is the time taken to introduce the F-atom into the nitroimidazole and subsequently isolate the product, because the half-life of 1 8 F is only 109.8 min. There are many potential precursor (2N02lm) compounds presented in this thesis, but the most obvious choice was SR2508 as it contains a primary hydroxyl group. The conversion of an alcohol to a sulfonate ester is an excellent method for labeling compounds with fluorine,74'75 provided that these esters can be isolated. The most reactive of the sulfonate esters are the triflates which are used in the synthesis of 18F-labeled sugars; the best example is 3-[ 1 8F]FDG, 7 6 the most frequently used radiolabeled imaging compound since its discovery in 1975. The attempted synthesis of the triflate analogue of SR2508 by its reaction with triflic anhydride was not successful (Section 3.2.5). The X-ray structure of the isolated yellow product (19, Figure 3-14, see Appendix 1-6) revealed surprisingly a 2N02lm species containing a "cyclic" side-chain; the C(4)-C(5) bond length of 1.312 A is consistent with the exocyclic methylene functionality. The ^{ /F f} N M R spectrum displays a singlet at 5 0.88 that results from the triflate bonded at the side-chain N-atom. The ^ N M R spectrum is consistent with the solid state structure, and shows shifts 122 References on page 129 Chapter 3 downfield from those of the precursor SR2508 side-chain. The signal for the methylene group between the carbonyl and the imidazole ring of SR2508 (5 5.21) is replaced by one Figure 3-14: ORTEP view of 19; 33 % probability thermal ellipsoids are shown. at 5 6.78 corresponding to a single, olefinic proton. The H-bonding interactions, involving the oxygens of the N 0 2 group (H(3)-0(2) 2.35 A) and the S 0 2 C F 3 group (H(3)-0(4a) 2.50 A) (both electron-withdrawing groups), also have a marked effect on 8 value of the C H proton. The *H N M R signals corresponding to the two other methylene groups on the SR2508 side-chain are also shifted significantly downfield (8 3.59 to 4.36, and 3.34 to 4.64, for N - C H 2 - and -CH 2 -0 , respectively) due to the intramolecular cyclization. The shift observed for the N - C H 2 - protons may be because of their proximity to, and H-bonding interaction (H(5)-0(5a) 2.47 A) with, the S0 2 CF 3 group. This intramolecular 123 References on page 129 Chapter 3 cyclization has been previously suggested,44'75 but no evidence was presented in either report. The synthesis of the tosylate derivative of SR2508 was also attempted using TsCl, but the isolated product was ECll(-l) formed via a presumed tosylate intermediate (Scheme 3-16). These results suggest that formation of a sulfonate ester from SR2508 is not a viable route to the incorporation of fluorine into the side-chain via introduction of an active F-containing species. Lim and Berridge reported the isolation of the tosylated N 0 2 SR2508 TsClpy T o N 0 2 OTs c r ECll(-l) Scheme 3-16: Synthesis of ECll(-l) via the tosylate intermediate. SR2508 compound via an alternate route (Scheme 3-17) and subsequent reaction with fluoride gave an acceptable yield (40 %) of EFl(-l).7 7 However, Tewson later reported that reaction of fluoride ( 1 8F or 1 9 F) with the isolated tosylate gave no sign of any fluorinated product.44 Johnstrom and Stone-Elander have reported the synthesis of the fluorinated alkylating agent 2,2,2-trifluoroethyl triflate78 which in theory could be coupled to a nitroimidazole containing the composition required (for example, the 2-nitroimidazolacetamide in Scheme 3-17) to yield a desired product (in this case, EF3(-1)). This coupling would avoid unwanted side-reactions (such as nitro goup displacement (to be discussed below) and cyclization chemistry) incurred by doing chemistry on the nitroimidazole itself. 124 References on page 129 Chapter 3 N0 2 I  DMF O N, TsO' OTs H F F' N. KOH OTs EtOH Scheme 3-17: Synthesis of fluoroetanidazole (EFl(- l )) via a tosylate intermediate.77 An interesting and unexpected finding during the attempted synthesis of the triflated SR2508 was the isolation of the formate ester, compound 18 (Scheme 3-18). Originally, D M F was added to the reaction mixture to solubilize the SR2508 in CH 2 C1 2 , and reaction of triflic anhydride under basic conditions yielded compound 19 as a minor product, but mainly 18. The 1 9F{TI} N M R spectrum indicated the absence of triflate, while TI N M R data unveiled a new proton signal at 8 8.13 and a shifted signal for the original CH2-0 protons of SR2508 (from 8 3.34 to 4.21). These results indicated the presence of an aldehydic proton on the terminal end of the side-chain. The possibility of the new signal corresponding to a carboxylic acid functionality was nullified when the compound did not react with a fluorinated amine (using N M M and iBuClFrm). Further evidence supporting formation of an aldehyde is observed in the IR spectrum which displays two vco bands at 1716 and 1655 cm"1 corresponding to COaidehyde and COamide, respectively. Interestingly, formation of the aldehyde implies that the SR2508 alcohol preferentially reacts with D M F over the triflic anhydride. The carbonyl C-atom of D M F is apparently more electrophilic than the S-atoms of the Tf 2 0, and this leads to formation of the formate (18) (Scheme 3-18). This result is contrary to the expected reaction of the alcohol with Tf 2 0 which should be more active because of the presence of the strongly electron-withdrawing CF 3 group.79 125 References on page 129 Chapter 3 Scheme 3-18: Mechanism for reaction of SR2508 with D M F to yield the formate ester (18). The unsuccessful formation of a sulfonate ester of SR2508 led to a different approach to incorporate fluorine into the side-chain. The halide substitution reaction is widely used for the synthesis of 18F-containing compounds.76 In general, the yields of substitution products are higher with fluorine displacements of Br and I, although there is one successful synthesis which utilized 1 8F-for-Cl exchange.80 According to other reactions investigated in this work (e.g. reactivity of 3 (Section 3.2.1.4, also see Scheme 3-3) and IBr (Section 3.2.1.2, also see Scheme 3-4)) the terminal C-I and C-Br bonds are quite reactive and lead to formation of intramolecular cyclization products analogous to 19. The synthesis of an E I l ( - l ) species was unsuccessful as a cyclized product forms instantaneously, and so the E B r l ( - l ) species was used as a precursor in the exchange reactions. The introduction of fluorine as a nucleophile requires the use of the fluoride ion; however, fluoride salts (like KF) have low solubility in organic solvents. The enhancement of K + solubility in organic solvents is obtained through use of crown ethers (e.g. 18-crown-6) or aminopolyethers (e.g. 2,2,2-Kryptofix) which in turn solubilize the fluoride anion. Tetraalklyammonium fluoride salts can also be used as the cation is readily soluble in dipolar, aprotic solvents (e.g. MeCN, DMF and dmso) which are the preferred solvents for the exchange reaction. Because the nucleophilicity of the fluoride ion is decreased in 126 References on page 129 Chapter 3 the presence of H 2 0 (truly anhydrous fluoride has never been achieved), the use of dry solvents and a dry atmosphere is essential. The reaction of Bu 4NF»H 20 with EBrl (section 3.2.7.1) did not yield the desired EFl product, but data suggest that the fluoride was incorporated into the isolated product. UV-Vis analysis revealed a split in the major chromophore of the precursor EBrl at 328 nm, two bands at 264 and 372 nm being seen. IR analysis revealed the disappearance of the VNO bands of EBrl at 1484 and 1386 cm"1 which indicates nucleophilic displacement of the N0 2-group. 8 1 ' 8 2 The *!! N M R spectrum revealed no change in the splitting pattern for the methylene groups on the side-chain, while incorporation of a F-atom would lead to two doublet of triplets. The 1 9F{TI} N M R spectrum, however, does show that F is present (8 -69.85). As no H-F coupling is observed in the ! H N M R spectrum, the F atom is assumed to be at the 2-position of the imidazole ring and the product is formulated as 25; no peak is observed at 5 -142.50 for EFl. The nucleophilic displacement of the N0 2-group appears to be quite facile,83 and occurs independently of the position of the N0 2-group in the imidazole ring (see Section 3.2.7.4, p.92). This reaction is actually used as a method for incorporation of 1 8 F into N-methylspiroperidol, a PET agent used for investigating dompamine receptors.84 Investigation of the reaction of EBrl with an 1 8 F fluoride source gave up to a 2 % radiochemical yield of [18F]-EF110, implying that y-ray detection is a more sensitive technique than N M R for detection of the fluorinated nitroimidazole. The synthesis of EF2Br (Section 3.2.2.8) was intended to increase the electrophilicity of the C-Br carbon to the point where fluoride would preferentially react at the C-Br to displace the bromine (and generate EF3), rather than at the 2-position of the imidazole ring to displace the N 0 2 group. The X-ray structure of EF2Br (Figure 3-15, see Appendix 1-2) shows that the C(8)-Br(l) bond (1.936(3) A) is much longer than either of the C(8)-F(l or 2) bonds (average = 1.353(4) A). Unfortunately, the fluorine atoms on EF2Br apparently increase the acidity of the protons on the C H 2 group adjacent to CF 2 Br, and reaction of EF2Br with Bu4NF'H20 resulted in the elimination of HBr to form the difluoroalkene, E=F2. 1 9F{TI} N M R analysis of the reaction mixture before E=F2 was 127 References on page 129 Chapter 3 Figure 3-15: ORTEP view of EF2Br; 33 % probability thermal ellipsoids are shown. isolated indicated that there was some N0 2-group displacement, but no signals corresponding to EF3 were observed. In order to prevent the elimination of HBr, EF4Br was synthesized (Section 3.2.2.5) that in which the P-H atoms of EF2Br were replaced by F-atoms. EF4Br, however, was totally inert and did not react with fluoride, even at high [F"] and high temperatures (and a small amount of the nitro displacement product was detected using ^ { / H } N M R ) . 8 5 These results suggest that the chemistry of the side-chain needs to be investigated further before a high yield exchange reaction can be accomplished. 128 References on page 129 Chapter 3 3.4 References 1 Urtasun, R. C ; Chapman, J. D.; Raleigh, J. A. ; Franko, A . J; Koch, C. J. Int. J. Radiat. Oncol. Biol. Phys. 1986, 12, 1263. 2 Lord, E. M . ; Harwell, L . ; Koch, C. J. Cancer Res. 1993, 53, 5271. 3 Woods, M . L . ; Koch, C. J.; Lord, E. M . Int. J. Rad. Oncol. Biol. Phys. 1996, 34,93. 4 Southwick, P. L . ; Ernst, L. A. ; Tauriello, E. W.; Parker, S R . ; Mujumdar, R. B. ; Mujumdar, S. R.; Clever, H . A.; Waggoner, A. S. Cytometry 1990,11, 418. 5 Koch, C. J.; Evans, S. M . ; Lord, E. M . Br. J. Cancer 1995, 72, 869. 6 Evans, S. M . ; Jenkins, W. T.; Joiner, B. ; Lord, E. M . ; Koch, C. J. Cancer Res. 1996, 56, 405. 7 Evans, S. M . ; Joiner, B. ; Jenkins, W. T.; Laughlin, K , M . ; Lord, E . M . ; Koch, C. J. Br. J. Cancer 1995, 72, 875. 8 Wong, T. Y . H . Ph.D. Thesis, University of British Columbia, 1996. 9 Pattison, F. L . M . ; Howell, W. C ; White, R. W. J. Am. Chem. Soc. 1956, 78, 3487. 10 Kachur, A. V. ; Evans, S. M . ; Shiue, C. -Y . ; Dolbier, Jr., W. R.; L i A. -R.; Roche, A. ; Skov, K. A . ; Baird, I. R.; James, B. R.; Koch, C. J. Appl. Rad. hot. submitted. 11 Middleton, W. J. J. Org. Chem. 1975, 40, 574. 12 Markovskij, L . N . ; Pashinnik, V . E.; Kirsanov, A. V . Synthesis 1973, 787. 13 Gibson, M . S.; Bradshaw, R. W. Angew. Chem. Int. Ed. Engl. 1968, 7, 919. 14 Sheehan, J. C ; Ryan, J. J. J. Am. Chem. Soc. 1951, 73, 1204. 15 Elemental analysis data obtained in collaboration with Dr. A. Kachur (Univ. of Penn.). 16 Beaman, A. G ; Duschinsky, R.; Tautz, W. P. U.S. Patent #3679698 (1972). 17 This work, performed during the course of my Ph. D., was recently published: Baird, I. R.; Rettig, S. J.; James, B. R.; Skov, K. A. Synth. Commun. 1998, 28, 3701. 18 Knorr, R.; Trzeciak, A. ; Bannwarth, W.; Gillessen, D. Tetrahedron Lett. 1989, 30, 1927. 19 Bannwarth, W.; Schmidt, D.; Stallard, R. L. ; Hormung, C ; Knorr, R.; Muller, F. Helv. Chim. Acta 1988, 71, 2085. 129 Chapter 3 20 Hay, M . P.; Wilson, W. R.; Moselen, J. W.; Palmer, B. D.; Denny, W. A. J. Med. Chem. 1994, 57, 381. 21 Kajfez, F.; Sunjic, V. ; Kolbah, D.; Fajdiga, T.; Oklodziga, M . J. Med. Chem. 1968, 77, 167. 22 Mann, F. G ; Porter, J. W. G. J. Chem. Soc. 1945, 751. 23 Fortier, D. G ; McAuley, A. Inorg. Chem. 1989, 28, 655. 24 Fortier, D. G ; McAuley, A. J. Am. Chem. Soc. 1990,112, 2640. 25 Koch, C. Private communication, Univ. of Perm. 26 Connelly, N . G ; Geiger, W. E. Chem. Rev. 1996, 96, 877. 27 Jenkins, T. C. In The Chemistry of Antitumour Agents; D. E. V . Wilman, Ed.; Blackie and Son Ltd., Glasgow, 1990, p. 354. 28 Kagiya, T.; Abe, M . ; Nishimoto, S.; Shibamoto, Y . ; Shimokawa, K. ; Hisanaga, Y . ; Nakada, T.; Yoshizawa, T. U.S. Patent #4927941 (1990). 29 Kagiya, T.; Abe, M . ; Nishimoto, S.; Shibamoto, Y . ; Otomo, K ; Tanami, T.; Shimokawa, K.; Yoshizawa; Hisanaga, Y . U.S. Patent #4977273 (1990). 30 Kagiya, T.; Abe, M . ; Nishimoto, S.; Shibamoto, Y . ; Shimokawa, K. ; Hisanaga, Y . ; Nakada, T.; Yoshizawa, T. U.S. Patent #5304654 (1994). 31 Tracy, M . ; Kelson, A. B. ; Workman, P.; Lewis, A. D.; Aboagye, E.O. International Patent: PCT/US95/09611 (1996). 32 Adams, G. E. ; Clarke, E. D.; Flockhart, I. R.; Jacobs, R. S.; Sehmi, D. S.; Stratford, I. J.; Wardman, P.; Watts, M . E. Int. J. Radiat. Biol. 1979, 35, 133. 33 Urtasan, R. C ; Band, P.; Chapman, J. D.; Rabin, H . R.; Wilson, A. F.; Fryer, C. G. Radiology 1977,122, 801. 34 Coleman, C. N . ; Urtasun, R. C ; Wasserman, T. H ; Hancock, S.; Harris, J. W.; Halsey, J.; Hirst, K. V . Int. J. Radiat. Oncol. Biol. Phys. 1984,10, 1749. 35 Husted, D. R.; Ahlbrecht, A. H . J. Am. Chem. Soc. 1953, 75, 1605. 36 Haszeldine, R. N . ; Leedham, K. J. Chem. Soc. 1953, 1548. 130 Chapter 3 37 Calvert, J. G ; Pitts, Jr., J. N . Photochemistry; John Wiley and Sons, New York, 1966, p. 522. 38 Threadgill, M . D.; Webb, P. Synth. Commun. 1990, 20, 2319. 39 Brown, H . C ; Heim, P. J. Org. Chem. 1973, 38, 912. 40 Jenkins, T. C ; Naylor, M . A.; O'Neill, P.; Threadgill, M . D.; Cole, S.; Stratford, I. J.; Adams, G. E.; Fielded, M . ; Suto, M . J.; Stier, M . A. J. Med. Chem. 1990, 33, 2603. 41 Koh, W.-J.; Rasey, J. S.; Evans, M . L . ; Grierson, J. R.; Lewellen, T. K. ; Graham, M . M . ; Krohn, K. A ; Griffin, T. W. Int. J. Radiat. Oncol. Biol. Phys. 1992, 22, 199. 42 Silverstein, R. M . ; Bassler, G. C ; Morrill, T. C. Spectrometric Identification of Organic Compounds, 5 t h edn.; John Wiley and Sons, Toronto, 1991, p. 185. 43 Chi, D. Y . ; Kilbourn, M . R ; Katzenellenbogen, J. A. ; Welch, M . J. J. Org. Chem. 1987, 52, 658. 44 Tewson, T. J. Nucl. Med. Biol. 1997, 24, 755. 45 Pattison, F. L . M . ; Cott, W. J.; Howell, W. C ; White, R. W. J. Am. Chem. Soc. 1956, 78, 3484. 46 Pattison, F. L . M . ; Howell, W. C ; White, R. W. J. Am. Chem. Soc. 1956, 78, 3487. 47 Middleton, W. J. J. Org. Chem. 1975, 40, 574. 48 Olah, G. A ; Nojima, M . ; Kerekes, I. Synthesis 1973, 786. 49 Olah, G. A. ; Nojima, M . ; Kerekes, I. J. Am. Chem. Soc. 1974, 96, 925. 50 Wilson, R. L . ; Cramp, W. A ; Ings, R. M . J. Int. J. Radiat. Biol. 1914, 26, 557. 51 Denny, W. A ; Roberts, P. B. ; Anderson, R. F.; Brown, J. M . ; Wilson, W. R. Int. J. Radiat. Biol. Oncol. Phys. 1992, 22, 553. 52 Adams, G. E.; Clarke, E. D.; Flockhart, I. R.; Jacobs, R. S.; Sehmi„D. S.; Stratford, I. I ; Wardman, P.; Watts, M . E.; Parrick, J.; Wallace, R. G ; Smithen, C. E. Int. J. Radiat. Biol. 1979, 35, 133. 53 Edwards, D. I.; Dye, M . ; Carne, H. J. Gen. Microbiol. 1973, 76, 135. 54 Franko, A. J. Int. J. Radiat. Oncol. Biol. Phys. 1986,12, 1195. 55 Anderson, C ; Beauchamp, A. L. Inorg. Chem. 1995, 34, 6065. 131 Chapter 3 56 Anderson, C ; Beauchamp, A. L . Inorg. Chim. Acta 1995, 233, 33. 57 Clarke, M . J. In Progress in Clinical Biochemistry; Springer-Verlag, Berlin, 1989, Vol. 10, p.25. 58 Orpen, G. A. ; Brammer, L. ; Allen, F. H. ; Kennard, O.; Watson, D. G ; Taylor, R. J. Chem. Soc, Dalton Trans. 1989, S31. 59 Adams, G. E.; Stratford, I. T; Wallace, R. G ; Wardman, P.; Watts, M . E. J. Natl. Cancer Inst. 1980, 64, 555. 60 Adams, G E.; Flockhart, I. R.; Smithen, C. E.; Stratford, I. T; Wardman, P.; Watts, M . E . Radiat. Res. 1976, 67, 9. 61 Denny, W. A.; Wilson, W. R. J. Med. Chem. 1986, 29, 879. 62 Denny, W. A.; Wilson, W. R.; Hay, M . P. Br. J. Cancer 1996, 74, S32. 63 Joseph, P.; Jaiswal, A. K. ; Stobbe, C. C ; Chapman, J. D. Int. J. Radiat. Oncol. Biol. Phys. 1994, 29, 351. 64 McManus, M . E.; Lang, M . A. ; Stuart, K. ; Strong, J. Biochem. Pharmacol. 1982, 31, 547. 65 Walton, M . I.; Workman, P. Biochem. Pharmacol. 1987, 36, 887. 66 Edwards, D. T; Knight, R. C ; Zahoor, A. Int. J. Radiat. Oncol. Biol. Phys. 1986, 12, 1207. 67 Wardman, P.; Clarke, E. D. Biochem. Biophys. Res. Commun. 1976, 69, 942. 68 Aboagye, E. O.; Lewis, A. D.; Johnson, A.; Workman, P.; Tracy, M . ; Huxham, I. M . Br. J. Cancer 1995, 73, 312. 69 Wardman, P. Environ. Health Per spec. 1985, 64, 309. 70 Breccia, A. ; Berrilli, G. Roffia, S. Int. J. Radiat. Biol. 1979, 36, 85. 71 Tercel, M . ; Wilson, W. R.; Denny, W. A. J. Med. Chem. 1993, 36, 2578. 72 Evans, S. M . ; Joiner, B. ; Jenkins, W. T.; Laughlin, K. M . ; Lord, E. M . ; Koch, C. J. Br. J. Cancer 1995, 72, 875. 73 Brennan, M . B. Chemical and Engineering News 1996, 2, 26. 132 Chapter 3 74 Kiesewetter, D. O.; Kilbourn, M . R.; Landvatter, S. W.; Heiman, D. F.; Katzenellen-bogen, J. A. ; Welch, M . J. J. Nucl. Med. 1984, 25, 1212. 75 Chi, D. Y . ; Kilbourn, M . R ; Katzenellenbogen, J. A. ; Welch, M . J. J. Org. Chem. 1987, 52, 658. 76 Kilbourn, M . R. Fluorine-18 Labeling of Radiopharmaceuticals, Nucl. Med. (Nuclear Science Series), National Academy Press, Washington, D. C , 1990, p. 49. 77 Lim, J. -L . ; Berridge, M . S. J. Nucl. Med. 41st Annual Meeting, 1994, section 6P, No.13. 78 Johnstrom, P.; Stone-Elander, S. J. LabelledCompd. Radiopharm. 1995, 36, 537. 79 McBee, E. T.; Battershell, R. D.; Braendlin, H . P. J. Chem. Soc. 1962, 3157. 80 Luxen, A ; Barrio, J. R.; Satyamurthy, N . ; Bida, G. T.; Phelps, M . E. J. Fluorine Chem. 1987, 36, 83. 81 Beck, J. R. Tetrahedron 1978, 34, 2057. 82 Bartoli, G ; Todesco, P. E. Acc. Chem. Res. 1977,10, 125. 83 Beck, J. R. Tetrahedron 1978, 34, 2057. 84 Arnett, C. D.; Wolf, A. P.; Shiue, C. - Y . ; Fowler, J. S.; MacGregor, R. R ; Christman, D. R ; Smith, M . R. J. Nucl. Med. 1986, 27, 1878. 85 Kachur, A. ; Koch, C. Unpublished results (Univ. of Penn.). 133 Chapter 4 Synthesis and Characterization of Ru(II) and Ru(III) Imidazole Complexes 4.1 Introduction Ruthenium imidazole complexes as chemotherapeutic agents show less cytotoxicity than their platinum counterparts.1'2 These Ru complexes are of interest for their antitumour activity,3 for their ability to ligate radiosensitizing agents to D N A , 4 as models for the bonding of Ru to nucleic acids (occurring most frequently on the imidazole ring of guanine),5'6 and as a new class of compounds that, in the low nanomolar range, specifically inhibit the proliferation of human T-lymphocytes in vitro2'1 The interaction of Ru complexes with biomolecules has been studied extensively,8'9 and the form of the complex surviving to coordinate the biomolecules usually differs from that originally introduced into the organism. This is especially true for Ru(III) complexes which are transformed by an in vivo reduction to Ru(II), a more active species toward D N A binding, as reduction accelerates substitution reactions at the metal center.10 This chapter reports the synthesis of some new Ru(II) and Ru(III) imidazole and nitroimidazole complexes which may have potential biological implications. Several different starting materials (see Chapter 2) were used, and the product complexes were characterized using NMR, UV-Vis , IR, MS, elemental analysis, conductivity and magnetic moment (Evans method) measurements, while X-ray analysis was only performed on the Ru(II) hexakis(imidazole) complexes and /wer-RuCl3(MeCN)3. 134 References on page 182 Chapter 4 4.2 Experimental Section 4.2.1 Complexes Synthesized from [Ru(DMF)6][CF3S03]3 4.2.1.1 [Ru ( Im) 6 ] [CF 3 S0 3 ] 2 (28) In a 100 mL Schlenk tube, [Ru(DMF) 6][CF 3S0 3] 3 (0.362 g, 0.367 mmol) was dissolved in MeOH (20 mL) under N 2 to yield a bright yellow solution. Finely ground Im (0.189 g, 2.78 mmol), dissolved in hot MeOH (5 mL), was then added. The mixture was stirred at 70 °C for 5 h over which time the solution became orange and then dark green; however, the reaction remained incomplete according to TLC (CH 2 Cl 2 :MeOH, 20:1). The solution was thus stirred at 40 °C for an additional 8 h. The MeOH was removed in vacuo leaving a dark green gum that contained DMF. Acetone (5 mL) was then added, followed by CH 2 C1 2 (-30 mL) which resulted in a green, microcrystalline precipitate. This was collected, washed with warm CH 2 C1 2 ( 6 x 5 mL, to ensure removal of excess Im) and dried in vacuo at 80 °C for 2 days (yield 0.236 g, 80%). Dark green, X-ray quality crystals resulted from slow evaporation of a MeOH solution of 28. Anal. calc. for C 2 oH 2 4 Ni 2 F 6 S 2 0 6 Ru: C, 29.74; H , 3.00; N , 20.81; found C, 29.87; H , 2.87; N , 20.45. IR (v, cm - 1): 3329 (N-H); 3145, 2966, 2864 (C-H); 1253, 1226(sh), 1173, 1068, 1027, 766, 636. UV-Vis (MeOH): 218 (12.3), 288 (7.6), 642 (0.033); for free Im, 216 (1.6). TI N M R (200 MHz, d6-dmso): 5 12.47 (s, IH, Im-Zfj), 7.23 (s, IH, Im-#5), 6.88 (br s, IH, Im-#2), 6.22 (s, IH, Im-#4). 1 9F{TI} N M R (188 MHz, d6-dmso): 5 -1.20 (s, CF 3 S0 3 "). E 1 / 2 = 246 mV. 4.2.1.2 [Ru (NMeIm) 6 ] [