UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Ruthenium nitroimidazole complexes as radiosensitizers Chan, Peter Ka-Lin 1988

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

Item Metadata


831-UBC_1988_A1 C44.pdf [ 7.91MB ]
JSON: 831-1.0060318.json
JSON-LD: 831-1.0060318-ld.json
RDF/XML (Pretty): 831-1.0060318-rdf.xml
RDF/JSON: 831-1.0060318-rdf.json
Turtle: 831-1.0060318-turtle.txt
N-Triples: 831-1.0060318-rdf-ntriples.txt
Original Record: 831-1.0060318-source.json
Full Text

Full Text

RUTHENIUM NITROIMIDAZOLE COMPLEXES AS RADIOSENSITIZERS by PETER K A - L I N CHAN B.Sc. (Co-op), University of Victoria, 1982 M.Sc, University of British Columbia, 1984 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE STUDIES (Department of Chemistry) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA AUGUST 1988 ©Copyright Peter Ka-Lin Chan, 1988 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 CHEmtSTR.V The University of British Columbia 1956 Main Mall Vancouver, Canada V6T 1Y3 Date DE-6G/81) i i ABSTRACT Local control of tumours by radiotherapy may fail due to the presence of regions of hypoxic cells. Radiosensitizers, such as nitroimidazoles, enhance killing of the resistant cells by ionizing radiation. However, dose limiting side-effects have prevented the attainment of maximum sensitization. The successful chemotherapeutic drug, cz's-diamminedichloroplatinum(II) (m-DDP), and analogues show moderate radiosensitizing effects, possibly because of binding to DNA. A rationale is then to use the DNA binding property of a metal to carry a sensitizer to the target of radiation damage, DNA, thereby improving the radiosensitizing effect while reducing the toxic side-effects of nitroimidazoles. The complex c/^-RuC^Cdmso)^ was used as a precursor for synthesis of Ru(II)-nitroimidazole complexes because of its anti-tumour and DNA binding activities. A series of Ru(II) complexes of formulation RuC^dmso^Ljj, where dmso is S-bonded dimethyl sulphoxide, L = a nitroimidazole, and n=l or 2, has been synthesized and characterized, and their toxicities and radiosensitizing abilities examined in vitro. When L = 2-nitroimidazole or a substituted-2-nitroimidazole, n = 2, but the nitroimidazole ligands dissociate in aqueous medium. With L = the 5-nitroimidazole, metronidazole, n=2, the sensitizing ability of the six-coordinate cis complex was disappointing with sensitizer enhancement ratio (SER) of 1.2 in hypoxic Chinese hamster ovary (CHO) cells. A series of 4-nitroimidazoles ligands was then studied. With L = 4-nitroimidazole (4-N02~Im), 1-(1' -aziridinyl-2' -propanol)-2-methyl-4-nitroimidazole (RSU-1170), 2-(l,2-dimethyl-4-nitroimidazolyl)-2-aminoethanol (RSU-3083), and 1-methyl-4-nitro-5-phenoxyimidazole (RSU-3100), n=2 and the six coordinate complexes appear to be of all cis geometry. The NMe-4-N0 2-Im ligand (n=l) chelates through the imidazole-N and the oxygen of N 0 2 group as evidenced from spectroscopic data. Coordination via the nitrito group is uncommon and other examples involving nitroimidazole ligands have not been reported. For the l-methyl-5-(2'-thioimidazolyl)-4-nitroimidazole (RSU-3159) ligand (n=l), binding to Ru occurs through the thioether and chelation may occur through the imidazole-NCH^. In this series of Ru(II)-4-nitroimidazole complexes studied, RuC^Cdmso^^-N02-Im)2, 5, was the most effective radiosensitizer (SER = 1.6 at 200 ,wM) and is better than the clinically used misonidazole (SER = 1.3 at 200 ^M). In addition, 5 did not sensitize oxic CHO cells. Other Ru-N-substituted-4-nitroimidazole complexes gave SER values of 1.1-1.4 at 100-200 jM. Complex 5 also produced a dose-dependent increase in genotoxic activity (as measured by the in vitro induction of chromosome aberrations in CHO cells), which is similar to that of misonidazole but much less than that of c/s-DDP. Two changes in ancillary ligands and geometry of complexes were also examined: replacement of (i) dmso by tmso (tetramethylene sulphoxide), (ii) CI" by Br". The Ru-nitroimidazole complexes were synthesized from the precursors RuCl2(tmso)4 and /raHj-RuB^dmso)^ In this series of complexes , only Ru02(tmso)2(4-N02-Im)2, i l , and RuCl2(tmso)2(SR-2508), i i , have significantly higher SER values (1.6 and 1.5, respectively) than their corresponding nitroimidazole ligands. The tmso complexes of 2-N02"Im derivatives were more stable than the dmso series in aqueous solution with respect to the dissociation of the nitroimidazole ligands, which might be due to the improved lipophilicity of tmso complexes. Complex 18. is suggested to be penta-coordinated from XPS and ir data. The RuBr2(dmso)2(4-N02-Im)2 was a less effective sensitizer (SER = 1.3 at 200 _#M) than the dichloro analogue which may result from different geometrical structures or different behaviour in aqueous solution chemistry. The enhanced radiosensitizing effect over the corresponding free nitroimidazole ligand observed for complexes 5, J l and 18 may depend on: (a) the metal's ability to target the sensitizer to DNA; complex 5 does bind to DNA, dissociation of CI" perhaps i v fac i l i tat ing the reaction; (b) the increase in reduction potential or (c) an increase in l ipoph i l i c i ty of the ni troimidazole l igand on coordinat ion. However , the enhanced radiosensit izat ion does not result f rom depletion of non-prote in thiols. In the present study, the R u complexes are less toxic than their corresponding ni t ro imiazole ligands in vitro. The radiosensitization and toxic i ty of the complexes 5, 15 and 18 are better than those of the free nitroimidazole ligands and the c l in ica l ly used radiosensit izer, misonidazole. The data encourage further investigations of the use of transit ion metal complexes as radiosensitizers to combat the hypoxic tumour cells. L igand Structures NO, R' / V (3)N NR(1) <3)N. NR(1) NO, R" = 2-nitroimidazole OH I DesmetJiylmisonidazole (De-miso)-Etanidazole (SR-2508) R = H, R' = H R" = H : 4-nitroimidazole (4-N02-Im> R =CH3> R' = H, R" = H : N-methy1-4-nitroimidazole R =CH2CH(OH)CH2N^] , R' = H , R" = CH : RSU-1170 H I R =CH3< R' = N-CH -CH -OH, R" = C H , : RSU-3083 OH I = Hisonidazole NO, / CH2CH2OH CH„ , R" = H : RSU-3159 metronidazole V TABLE OF CONTENTS PAGE Abstract i i Table of Contents v List of Tables x i List of Figures x i i i List of Abbreviations xv Acknowledgements xv i i CHAPTER ONE: INTRODUCTION 1.1 Radiation 1 1.2 Radiation and Cell Survival 2 1.3 Radiation and Oxygen Effect 3 1.4 The Proposed Mechanisms for the Oxygen Effect 6 1.5 Cancer and Radiation Therapy 8 1.6 Therapies Which May Overcome the Radioresistance of Hypoxic Cells 10 1.7 General Properties of Radiosensitizers 13 1.8 Radiosensitizing and Hypoxic Toxic Effects of Nitroimidazoles 14 1.9 Sulphydryl Binding Compounds as Radiosensitizers 17 1.10 Platinum Complexes as Chemotherapeutic Agents and Radiosensitizers 20 1.11 Metal Complexes as Radiosensitizers 22 1.12 Rationale for Using Metal Radio-sensitizer Complexes 25 1.13 Metal-Nitroimidazole Complexes as Radiosensitizers 2 7 v i TABLE OF CONTENTS PAGE 1.14 Ruthenium Complexes as Anticancer Agents 28 1.15 Nature of Sulphoxides as Ligands 29 1.16 Biological Activity of Metal-dimethyl-sulphoxide Complexes 31 1.17 Short-term in vitro Assays 33 1.18 Objectives 33 CHAPTER TWO: MATERIALS AND METHODS 2.1 Chemicals 35 2.2 Physical Techniques and Methods 3 8 2.2.1 X-ray Photoelectron Spectroscopy 38 2.2.2 Electrochemical Methods 40 (i) Polarographic Measurements 40 (ii) Cyclic Voltammetric Measurements 40 2.3 Preparation of N-Methyl-4-nitroimidazole 41 2.4 Preparation of Complexes 41 2.4.1 c/s-RuCl2(dmso)2, I 41 2.4.2 RuCl2(tmso)4, 2 42 2.4.3 "RuCl2(mpso)2", 3 4 2 2.4.4 //Yz«s-RuBr 2(dmso) 4, 4 4 3 2.4.5 RuCl2(dmso)2(4-N02-Im)2, £ 4 3 2.4.6 RuCl2(dmso)2(NMe-4-N02-Im), 6 4 4 2.4.7 RuCl2(dmso)2(RSU-1170)2, 7 4 4 2.4.8 RuCl2(dmso)2(RSU-3083)2,8 4 4 2.4.9 RuCl2(dmso)2(RSU-3100)2, 9 4 5 v i i TABLE OF CONTENTS PAGE 2.4.10 RuCl2(dmso)2(RSU-3159), 10 45 2.4.11 RuCl2(dmso)2(miso)2, i i 45 2.4.12 RuCl2(dmso)2(De-miso)2, 12, 46 2.4.13 RuCl2(dmso)2(2-N02-Im)2, i l 46 2.4.14 RuCl2(dmso)2(metro)2, i4 46 2.4.15 RuCl2(tmso)2(4-N02-Im)2, 15 4 7 2.4.16 RuCl2(tmso)2(NMe-4-N02-Im)2, i6 4 7 2.4.17 RuCl2(tmso)2(De-miso)2,il 4 7 2.4.18 RuCl2(tmso)2(SR-2508), i £ 4 8 2.4.19 RuCl2(tmso)3(CMNI), 19 4 8 2.4.20 RuCl2(tmso)(RSU-3159),20 4 8 2.4.21 RuBr2(dmso)2(4-N02-Im)2, 21 4 9 2.4.22 RuBr2(dmso)2(NMe-4-N02-Im), 22 4 9 2.4.23 RuCl2(dmso)2(l,10-phenanthroline), 23 4 9 2.4.24 RuCl2(dmso)2(5-Cl-l,10-phenanthroline), 24 5 0 2.4.25 RuCl2(dmso)2(5-NO2-l,10-phenanthroline), 25 50 2.4.26 RuCl2(dmso)2(NMe-Im)2, 26 50 2.4.27 RuCl2(dmso)2(2Me-Im)2, 27 51 2.4.28 [Ru(NH 3) 5Cl]Cl 2, 28 51 2.4.29 [Ru(NH3)5(NMe-Im)]Cl3, 29 52 2.5 Cell Culture Procedures 52 2.6 Irradiation 53 2.7 Radiosensitization - Cell Survival Experiments 53 2.8 Toxicity Studies - Cell Survival Experiments 56 2.9 Assay for Chromosome Aberrations 59 v i i i TABLE OF CONTENTS PAGE 2.10 Measurement of Non-protein Thiols 60 2.11 Inhibition of Restriction Enzymes 6 0 2.12 Measurement of Partition Coefficients 62 CHAPTER THREE: CHARACTERIZATION OF COMPLEXES 3.1 XPS Data 66 3.2 Infrared Spectral Data 72 3.2.1 Infrared Data of RuCl2(dmso)2Ln Complexes (n = 1 or 2) 72 3.2.2 Infrared Data of RuCl 2(tmso)mL n Complexes (m = 1, 2 or 3; n = 1 or 2) 75 3.2.3 Infrared Data of RuB^dmso^Ljj Complexes (n = 1 or 2) 78 3.2.4 Infrared Data of "RuCl2(mpso)2" 79 3.3 Nuclear Magnetic Resonance Spectral Data 79 3.3.1 Nuclear Magnetic Resonance Data of RuC^dmso^Ljj Complexes (n = 1 or 2) 80 3.3.2 *H nmr Data of RuCl 2(tmso)mL n Complexes (m = 1, 2 or 3; n = 1 or 2) 8 7 3.3.3 *H nmr Data of RuBr 2(dmso) 2Ln Complexes (n = 1 or 2) 9 3 3.4 4-Nitroimidazole vs. 5-Nitroimidazole Formulation 95 CHAPTER FOUR: AQUEOUS SOLUTION CHEMISTRY OF THE RUTHENIUM NITROIMIDAZOLE COMPLEXES 4.1 Solution Stability of the Substituted-2-nitroimidazole Complexes 99 i x TABLE OF CONTENTS 4.2 Aqueous Solution Chemistry of the RuCl 2 (R 2 SO) 2 L n Complexes; R 2SO = dimethyl-or tetramethylene-sulphoxide; L = 4-N0 2-Im (n = 1) or Substituted-4-N02-Im (n = 1 or 2) 4.3 Aqueous Solution Chemistry of the RuBr 2(dmso) 2Ln Complexes (L = 4-N0 2-Im, n=2; L = NMe-4-N0 2-Im, n=l) CHAPTER FIVE: CYTOTOXICITY AND RADIOSENSITIZATION  IN VITRO 5.1 Toxicity of the Ru Complexes and Their Nitroimidazole Ligands 5.2 Chromosomal-Damaging Activity of Complex 5 5.3 Radiation Sensitization of Aerobic and Hypoxic CHO Cells 5.3.1 Radiosensitization by the Metal Complexes CHAPTER SIX: FACTORS RELATED TO RADIOSENSITIZATION 6.1 DNA Binding 6.1.1 Assessment by Inhibition of Restriction Enzymes 6.1.2 DNA Binding by the Ru Complexes 6.2 Partition Coefficient (P) 6.3 Non-protein Thiol Depletion 6.4 Electron Affinities (Reduction Potentials) 6.5 Feasibility in IN VIVO System X TABLE OF CONTENTS PAGE CHAPTER SEVEN: CONCLUSIONS and RECOMMENDATIONS for FUTURE WORK 149 BIBLIOGRAPHY 152 APPENDICES I Preparation of Stock Solutions 161 II Preparation of Buffers 162 III Chromatography Through Sephadex G-50 1 6 3 IV Preparation of Agarose Gel 1 6 4 V Linearization of pSV2-gpt Plasmid 1 6 5 VI Preparation of Rat Liver S9 Fraction 1 6 7 VII Procedures for in vivo Study 1 6 8 x i LIST OF TABLES Table I Table II Table III Table IV Table V Table VI Table VII Table VIII Table IX Table X Table XI Table XII Table XIII Table XIV Metal compounds with radiosensitization ability XPS data for Ru(II) compounds and free ligands Selected ir spectral data for 4-nitroimidazole complexes of Ru and the free nitroimidazole ligands Selected ir spectral data for 2-nitroimidazoles, metronidazole and 1,10-phenanthrolines, and their Ru complexes Selected ir spectral data for RuCl2(tmso)mLn and RuBr2(dmso)2Ln complexes; m = 1, 2, or 3, n = 1 or 2 Selected *H nmr spectral data for RuC^dmso^Lp complexes and the free nitroimidazole ligands Selected nmr spectral data for RuCl2(tmso)mLn complexes and the free nitroimidazole ligands Selected ] H nmr and 1 3C{ JH} chemical shifts for RuBr2(dmso)2(4-N02-Im)2 and RuBr2(dmso)2(NMe-4-N02-Im) Molar conductivity data for some Ru complexes in water at RT Plating efficiency of the 4-nitromidazole ligands and their Ru(II) complexes in hypoxic CHO cells (4 h at 37°C) Hypoxic toxicity (PE) of the 2-nitroimidazoles and metronidazole and their Ru(II) complexes in CHO cells (4 h at 37°C) Clastogenic activity of RuCUCdmso^^-N0 2 -Im) 2 and ds-DDP on CHO cells Clastogenic activity of m-RuCi2(dmso)4 and 4-N0 2-Im on CHO cells The hypoxic SER values of 4-nitroimidazoles ligands and their Ru(II) complexes x i i LIST OF TABLES Table XV Table XVI Table XVII Table XVIII Table XIX The SER values of 2-nitroimidazoles and metronidazole and their Ru(II) complexes Relative inhibition of BamHI activity by some Ru complexes, c/5-DDP and trans-DDP Partition coefficients of nitroimidazole ligands and their ruthenium complexes NPSH depletion in hypoxia by some Ru(II) complexes Half-wave reduction potentials of Ru(II) compounds and their nitroimidazole ligands PAGE 12 7 132 136 139 142 x i i i LIST OF FIGURES PAGE Figure 1 A typical dose-response survival curve 4 Figure 2 The effect of hypoxia on survival curves 5 Figure 3 The cross-section of a tumour illustrating the oxic and hypoxic regions 9 Figure 4 A scheme showing the direct action mechanism of a radiosensitizer on a target molecule 12 Figure 5 Structural formulae of a nitrofuran, metronidazole and misonidazole 15 Figure 6 The mechanism of oxygen "inhibition" of nitroreduction 18 Figure 7 Proposed mechanism of sulphydryl radioprotection 19 Figure 8 Possible types of DNA interactions with m-DDP 21 Figure 9 Rationale for using metal-radiosensitizer complexes 26 Figure 10 The three resonance structures for the S-O bond of sulphoxides 30 Figure 11 Structural formulae of the 2-nitroimidazole ligands 36 Figure 12 Structural formulae of the 4-nitroimidazole ligands 37 Figure 13 Schematic representation of an anaerobic spectral cell 39 Figure 14 Irradiation set-up 54 Figure 15 The glass vessel for cell suspension toxicity tests 57 Figure 16 Outline of binding assay using plasmid DNA 61 Figure 17 XPS spectra N(ls) of RuCl2(dmso)2(4-N02-Im)2 and 4-N0 2-Im 70 Figure 18 nmr spectrum of c/s-RuCl2(dmso)4 in CDCl^ 81 Figure 19 nmr spectrum of Ru0 2(dmso) 2(4-N0 2-Im) 2 and RuCl2(dmso)2(NMe-4-N02-Im) in CDCI3 85 Figure 20 l3C{lH] nmr spectra of RuCl2(dmso)2(4-N02-Im)2 and RuCl2(dmso)2(NMe-4-N02-Im) in CDCI3 88 Figure 21 *H nmr spectra of RuCl2(tmso)4 and free tmso in CDCK 90 xiv LIST OF FIGURES PAGE Figure 22 Figure 23 Figure 24 Figure 25 Figure 26 Figure 27 Figure 28 Figure 29 Figure 30 Figure 31 Figure 32 Figure 33 Figure 34 Figure 35 H nmr spectra of RuCl2(tmso)2(4-N02-Im)2 and RuCl2(tmso)2(SR-2508) in C D C I 3 nmr spectra of RuBr2(dmso)2(4-N02-Im)2 and RuBr2(dmso)2(NMe-4N02-Im) in C D C I 3 pH titration curves Toxicity of RuCl^(dmso)2(4-N02-Im)2 and 4-N0 2-Im in hypoxic and 0x1c CHO cells Chromosome damaging activity of RuCl2(dmso)2(4-N0 2 -Im) 2 , misonidazole and 4-N0 2-Im in CHO cells in the absence of S9 Radiosensitization of CHO cells by RuCl2(dmso)2(4-N0 2 -Im) 2 , 5, or 4-N0 2-Im Radiosensitization of CHO cells by RuCl2(tmso)2(4-N0 2 -Im) 2 , 15, or 4-N0 2-Im Radiosensitization of CHO cells by RuBr2(dmso)2(4-N0 2 -Im) 2 , 2i, or 4-N0 2-Im Radiosensitization of CHO cells by RuCl2(dmso)2(NMe-4-N02-Im), 6, or NMe-4-N0 2-Im Radiosensitization of CHO cells by RuCWtmsoWSR-2508), J8, or SR-2508 Inhibition of endonuclease activities of BamHI and EcoRI Titration of BamHI site with RuCl2(dmso)2(4-N02-Im)2, 4-N0 2-Im and ds-DDP The E w 2 reduction potentials of RuCl2(dmso)2(4-N0 2 -Im) 2 and 4-N0 2-Im The effect on tumour cells survival of the time interval between RuCl2(tmso)2(SR-2508) administration and irradiation with 10 Gy of X-rays 92 9.6 102 108 117 121 122 123 124 126 130 131 141 147 XV LIST OF ABBREVIATIONS BE binding energy as-DDP czs-diamminedichloroplatinum(II), c-DDP CMNI 5-chloro-1 -methyl-4-nitroimidazole CHO Chinese hamster ovary De-miso (2-nitro-1 -imidazolyl)-3-hydroxypropanol (desmethylmisonidazole), Ro-07-9963 dmso dimethyl sulphoxide Ej/2 half-wave reduction potential EDTA disodium ethylenediaminetetraacetate dihydrate Ellman's reagent 5,5' -dithiobis-(2-nitrobenzoic acid) esr electron spin resonance FCS Fetal Calf Serum Gy Gray = 100 Rads. Im imidazole ir infrared LET Linear Energy Transfer M E M Minimal Essential Medium metro l-(2' -hydroxymethyl)-2-methyl-5-nitroimidazole (metronidazole) miso 1 -(2-nitro-1 -imidazolyl)-3-methoxypropanol (misonidazole), Ro-07-0582 mpso methylphenyl sulphoxide NMe-4-N02"Im N-methyl-4-nitroimidazole nmr nuclear magnetic resonance NO->-Im nitroimidazole xvi LIST OF ABBREVIATIONS NPSH non-protein thiols OER oxygen enhancement ratio PBS phosphate-buffered saline PE plating efficiency RSU-1170 l-(l'-aziridinyl-2'-propanol)-2-methyl-4-nitroimidazole RSU-3083 2-(l,2-dimethyl-4-nitroimidazolyl)-2-aminoethanol RSU-3100 l-methyl-4-nitro-5-phenoxyimidazole RSU-3159 1 -methyl-5-(2'-thioimidazolyl)-4-nitroimidazole RT room temperature S survival S9 liver microsomal preparations SER sensitizer enhancement ratio, usually ratio of radiation doses in hypoxic cells to give 1% survival SH sulphydryl, thiol SR-2508 1 -{N-(2-hydroxyethyl)-acetamido}-2-nitroimidazole tmso tetramethylene sulphoxide trans-DDT fra«5-diamminedichloroplatinum(II) XPS X-ray Photoelectron Spectroscopy x v i i ACKNOWLEDGEMENTS I am grateful to Professor B.R. James and Dr. K.A. Skov for their guidance, support and encouragement throughout the course of this work. I thank them for their patience and open-minded approach to a major clinical problem from both chemical and biological point of views. I also thank members of the "James' Gang" and the staff of Medical Biophysics Unit at the B.C. Cancer Research Centre, H. Adomat, I, Harrison, D. Wurst and Dr. P. Olive, for their support and co-operation. Early discussions with Dr. N.P. Farrell on the content and problems of this project were particularly rewarding. I am indebted to my brother, Paul, for helping me on the XPS work and Dr. D.C. Frost for allowing me to use his XPS instrument. The assistance of Mr. P. Borda on micro-analyses, Dr. O.S. Chan on nmr, and the electrical and mechanical services is also gratefully acknowledged. I would like to thank UBC for awarding me a University Graduate Fellowship and the Faculty of Medicine of UBC for awarding Summer Research Studentships during this programme. The granting agencies, NSERC, MRC and B.C. Cancer Foundation are also acknowledged for support of this project. Johnson Matthey Ltd. provided the loan of RuCl3 -3H 20; Dr. T.C. Jenkins is thanked for providing CMNI, RSU-3100, RSU-1170 and Dr. I.J. Stratford for supplying RSU-3083. Roche Products Ltd. kindly donated Ro-05-9963 (desmethylmisonidazole), and NCI of the United States, the SR-2508. I would like to extend a special thanks to my parents for allowing me to achieve what I set out to do. Finally, I dedicate this thesis to my wife, May, for giving me the strength, and going through the ups and downs with me; and also to my new born baby, Emilie, for providing the "Energy after mid-night" to complete this thesis. x v i i i We have not succeeded in answering all your problems. The answers we have found only serve to raise a whole set of new questions. In some ways we feel we are as confused as ever, but we believe we are confused on a higher level and about more important things. from Hickory Plaque Inc. Van Nuys, California, 1972. 1 CHAPTER ONE INTRODUCTION 1.1 Radiation Radiation capable of ionizing molecules is appropriately called ionizing radiation. Ionizing radiation, such as X-rays, causes damage by imparting sufficient energy to electrons within the irradiated matter to allow them to be released with concomitant ionization of molecules. This means that the electrons are removed from atoms, and this can cause chemical bonds to be broken and/or changed: AB ~ v ^ A - + B- — • P [1.1] AB — » A - + B- £L AR- + BR- [1.2] The electrons can also be removed from molecules of water, creating reactive ions which can in turn cause additional damage/ 2H20 - ^ r ^ c / + e ~ a + r^o" [1.3] + Rp — H 3 0 + + OH- [1.4] r^O* — - H- + OH- [1.5] OH- + DNA — * DNA- + r^O [1.6] The electrons released by ionization can also impart some of their energy to other electrons, thus creating a "chain reaction" and causing more ionizations, freeing other less energetic electrons. This process continues until all of the energy available is expended. These electron reactions can cause "permanent" chemical alterations, or they can terminate in harmless recombination processes/ The most dramatic effects of ionizing radiation on biological systems - mutation and death - are the consequence of the formation of molecular ions which can directly or indirectly cause damage to cellular biomolecules. If the DNA of the cell is damaged, this can result in mutation and chromosomal aberrations which can eventually transform 2 the cell or render the cell unable to divide. However, some of the long-lived chemical changes occurring may be subject to repair by chemical or enzymatic processes. The term "X-ray dose", as used in this thesis, refers to the absorbed dose, i.e. the energy imparted (absorbed) per unit mass. The Systeme International (SI) unit is the Gray (Gy) which corresponds to an energy absorption of 1 Joule (J) per kilogram (kg) of irradiated material. Previously, the unit of dose was called the Rad, defined as an energy absorption of 100 ergs/gram. Since IJ = 10 ergs, the Gray is equivalent to 100 Rads. 1.2 Radiation and Cell Survival In order to study the effects of radiation on biological systems, radiobiologists have chosen to study various biological endpoints such as DNA strand breaks, cell mutation, loss of cell proliferative capacity (e.g. in bacterial or mammalian cell system) or loss of organ function.'* An "endpoint" is defined as the effect that is measured at the conclusion of the experiment. One of the endpoints used in this thesis is the loss of proliferative capacity or viability of mammalian cells. For normally dividing cell populations, proliferative capacity is defined as the capacity of a cell to generate a clone of similar cells, called a colony. In this context, the "lethal" effect of radiation is defined as that which induces loss of proliferative capacity. Conversely, a "survivor" is defined as a cell which, after irradiation, has retained its ability to generate a clone of like cells. Techniques for scoring viable mammalian cells were developed in the mid-1950's,^  which allow for dose-response relationships to be obtained for mammalian cells. Survival curves for mammalian cells, with the surviving fraction plotted 3 logarithmically and the dose linearly, are usually described as having a "shoulder" (Figure 1). The shoulder region is generally followed by a region of approximately exponentially decreasing survival. Therefore, survival curves for mammalian cells at low doses of radiation show proportionally less killing per unit dose than at higher doses. There are many different models and theories to describe mammalian cell survival. Most of them are based on some hypothesis as to the mechanism of radiation damage, e.g. the presence of multiple targets per cell, or a single target requiring multiple hits, or a limited cellular repair capacity, could all account for shouldered survival curves (see 5 and references therein). The use of these expressions allows for comparisons to be made between parameters describing different cell lines, or between parameters describing different radiation conditions for the same cell lines. 1.3 Radiation and Oxygen Effect The radiosensitizing action of molecular oxygen (dioxygen)- was observed in the 1920's,^  but its significance was not appreciated. The effect of oxygen on cellular radiation response is illustrated in Figure 2, which shows the response of Chinese Hamster Ovary (CHO) cells to X-rays, first for irradiation in an air-saturated medium, and secondly for a medium from which all oxygen has been removed by purging with purified nitrogen. The absence of oxygen has a dramatic radioprotective effect; the oxic and hypoxic survival curves are of similar shape, but the dose required to produce a given level of cell kill is very different in the two cases. This difference is measured by the oxygen enhancement ratio (OER), defined as: Dose i n nitrogen OER = , at equal survival [1.7] Dose i n a i r a) The term "oxygen" will be used throughout this thesis to represent molecular 0 2 (dioxygen). 4 Figure 1: A t y p i c a l dose-response s u r v i v a l curve (taken from r e f . 1). (adapted from r e f . 1). 5 o < cc 1.0 0.1 I 4 \ x — r r C H O C E L L S > 0.01 > cc 3 0.001 CO \ o 2 — 1 ' 1 1 • 0 5 10 15 20 25 30 D O S E ( G r a y ) Figure 2: The e f f e c t of hypoxia on s u r v i v a l curves, oxic c e l l s ( • ) , hypoxic cells ( o ) (adapted from r e f . 7). 6 If oxygen is "dose-modifying", that is, if the aerobic and hypoxic survival curves can be superimposed by a change of the dose scale, then the OER will be independent of the value of S chosen for comparison; in practice, S = 0.01 is commonly chosen and is used in this thesis. For irradiation at this dose range the OER for mammalian cells is about 3 7 1.4 The Proposed Mechanisms for the Oxygen Effect The absorption of high energy radiation in matter causes ionizations and excitations of atoms and molecules (see Section 1.1). Regardless of the type of incident radiation or particle, electrons are principally responsible for these effects. Electron impact creates free radicals by breaking covalent bonds. This "physical stage" of energy deposition is followed by a "chemical stage" lasting a few microseconds.^  In this latter stage, the free radicals created by the radiation interact with undamaged molecules and with one another. These reactions can cause "permanent" (or at least, long-lived) chemical alterations, or they can terminate in harmless recombination processes. The biological effect of radiation will be changed by agents that modify the stage of chemical interactions. Presumably, the lethal and mutagenic effects of radiation are due to damage in macromolecules (especially DNA) which is not repaired, or which is misrepaired. Primary radiation events may be divided into two classes: direct action and indirect action.^ Direct action is the ionization of a macromolecule by a primary radiation event; indirect action is a secondary reaction caused by a reactive chemical species, which is formed by the absorption of radiation energy in small molecules (such as water). These species may then attack macromolecules. 7 Water is the most common molecule in a living cell, and the action of radiation on water has been studied in detail. The chemistry of water radiolysis was studied by Weiss^ who initially proposed that the process was the production of H - and OH" radicals, but it is now known that hydrated electrons are the principal reducing species.^0 These species react with each other to produce even-electron-numbered molecules: e" + e" 2 V H? + 2 0H~ [1.8] aq aq — 2 OH' + OH" — H 2 0 2 [1.9] OH" + e" — O H " [1.10] aq However, in the presence of oxygen, other reactions can occur, altering the yields of the primary radiolysis produces, and creating particularly superoxide,^ which can be dangerous when reacting with proteins, lipids, polysaccharides and nucleic acids.^ (aq) 2 2 H- + 0 2 — - 0~ + H + t 1 - 1 2 ! The hydroxyl radical is the most reactive oxidizing species, and thus damaging, of the water radiolysis products. Indirect damage from this species can then occur either by H-abstraction or OH' addition, subsequent chemical rearrangement "fixing" the lesion, e.g.: OH- + RH —Uf) + R- [1.13] The process yields products similar to those formed by direct damage: X-ray + RH —»- RH+ + e~ — • R- + H + + e" [1.14] Oxygen may act as a radiosensitizer in a number of ways. Alper^ proposed that the mechanism involves electron acceptance from target radicals generated by the radiation: R- + 0 2 —*- R0 2- — R + + 0 2 [1.15] 8 These R' radicals have a very short lifetime and may undergo chemical changes; reaction of a product with oxygen to give an organic peroxide is non-reversible and results in biological damage by fixation of a lethal lesion on the target, i.e. DNA. In the absence of oxygen, fewer peroxides are formed and thus more lesions are repaired. A second model of Adams^ suggested that electron affinic compounds, such as oxygen, might act by enhancing the effects of direct damage. After a target molecule has been ionized, the electron was proposed to localize at some electron affinic site on the molecule. Electron transfer to a sensitizer molecule of greater electron affinity would reduce the chance of recombination, and increase the chance of decay of the positive target ion to a free radical and subsequent damage fixation which could lead to DNA strand breaks (see Section 1.6). 1.5 Cancer and Radiation Therapy Radiotherapy forms at least part of the treatment for more than half of the new cancer patients in North America.^-5 The unlimited cell proliferation associated with malignant tumour growth often leads to a reduced oxygen concentration within the tumour. This in turn leads to a relative radioresistance not encountered in normally healthy tissue. The full relevance of this physico-chemical effect in therapy was demonstrated in the study by Gray and co-workers^ who showed that many tumours contain necrotic areas usually separated some distance (150-200 j*m) from the vascular system. This situation is depicted in Figure 3 and is explained by the fact that the disordered nature of tumour cell replication and the subsequent requirements of oxygen metabolism produce an oxygen gradient, rendering the cells furthest from the capillaries hypoxic. The clinical relevance is that because of the oxygen effect, these cells, which may represent up to 30% of the total tumour mass will be radioresistant, and upon 7 8 termination of treatment may become aerobic and tumour growth will recommence. One of the most direct pieces of evidence for the importance of the oxygen effect on The cross-section of a tumour i l l u s t r a t i n g the oxic and hypoxic regions (oxygen gradient) (taken from ref. 17). (adapted from r e f . 17). 10 radiotherapy is an analysis by Bush et al}^ of the influence of blood hemoglobin levels on cure rates of patients treated for cervical carcinoma. Those patients with greater hemoglobin levels had better cure rates than those with lower levels. Further, Chaplin 20 et al. have demonstrated that intermittent blood flow can occur in the tumour vasculature of murine squamous carcinoma, and this loss of flow has been shown to result in acute hypoxia in areas close to blood vessels. As a result of the oxygen effect in tumour cells, the radiation dose which can be delivered to a tumour is limited by the dose which the surrounding healthy tissue can tolerate. Any means of effectively reducing the relative radioresistance of the hypoxic tumour cells will increase the effectiveness of the radiotherapy, and thus should improve the cure rate. 1.6 Therapies Which May Overcome the Radioresistance of Hypoxic Cells Various approaches to the modification of radiotherapy, which may reduce the radioresistance of hypoxic cells, have been proposed. These include the use of hyperbaric oxygen, various forms of radiation, and chemical radiosensitization/-' The most apparent approach to overcome the hypoxic cell problem is hyperbaric oxygen therapy/ With this technique, patients are irradiated in a pressurized, oxygen-rich environment. The rationale for this approach is straightforward: an increase in blood oxygenation may reduce the size of the hypoxic fraction in a tumour. Nonetheless, results with hyperbaric oxygen have not been very encouraging, the reasons probably being two-fold/ First, the increase in blood oxygen-carrying capacity is far smaller than the increase in oxygen partial pressure. Secondly, oxygen, because of rapid depletion by metabolism in the periphery, is unable to penetrate into the hypoxic region of the tumour. In addition to these problems, hyperbaric oxygen is 11 toxic, and the procedure as a whole is dangerous and difficult to perform on a routine basis. While this thesis is concerned with X-ray irradiation, there are other forms of ionizing radiation which have been used to overcome the oxygen effect. The biological effectiveness of a particular form of radiation such as by X-rays or neutrons is related to its ionization density. This may be expressed in terms of the linear energy transfer (LET) of the radiation. This quantity describes the rate at which the particle deposits energy on travelling through matter. The value of OER decreases with increasing LET;21 for example, for X-ray irradiation, the OER is about 3 while for neutron irradiation, the OER is less than 2. It seems clear that the use of high LET irradiation in cancer therapy should reduce the limiting effect of tumour hypoxia. Additional benefits may be if, in addition to giving a lower OER, the depth-dose characteristics of the radiation permit better localization of dose in the tumour. However, prior to application, the development of these new radiation modalities have technical difficulties to be overcome. Further, relative cost is higher. One of the most promising approaches for the control of hypoxic cells is the use of a chemical radiosensitizer, that is, a drug which mimics the radiation-sensitizing properties of oxygen. Adams and Dewey observed that the compounds known to be sensitizers shared the property of being efficient electron acceptors as measured by reduction potentials. In 1969, Adams and Cooke^ suggested that the sensitizer increases the damage resulting from a direct action mechanism. In this case, direct ionization of the target molecule, e.g. DNA (see Figure 4), produces a positive ion and an electron, the latter migrating to some electron-affinic site on the molecule. If there is a compound of higher electron affinity available, either as a metal complex or as a free molecule, electron transfer from the ionized target molecule to the electron-affinic 12 Figure 4: A scheme showing the d i r e c t a c t i o n mechanism of a r a d i o s e n s i t i z e r on a target molecule. Process (a) would lead to DNA s e l f - h e a l i n g , while process (b) would r e s u l t i n DNA damage 13 molecule could occur. This process (b) would compete with charge recombination (a) in the target molecule which would contribute to self-healing. Electron-transfer to the sensitizer would reduce the chance of recombination and favour the decay of the ionized molecule to a free radical (Figure 4). In support of this mechanism experiments using electron spin resonance have shown directly that a charge transfer complex is formed in irradiated solid mixtures of DNA. In addition, Adams et al. have demonstrated rapid one-electron transfer between nucleotides in solution, implying that intramolecular electron transfer along a polynucleotide should readily occur. This group also demonstrated that known radiosensitizers can accommodate electrons that are rapidly and quantitatively transferred from one-electron adducts of simple pyrimidine and purine bases. These results have guided the search for effective radiosensitizers, using a variety of in vitro test systems. Several compounds were found to be effective in vitro, including stable nitroxyl free radicals'2** and aromatic nitro compounds/7 1.7 General Properties of Radiosensitizers For classification as an "effective radiosensitizer", the compound must possess several properties. First, the compound must be effective at a clinically achievable drug concentration, although it does not have to be as effective as oxygen. That is, if a "sensitizer enhancement ratio", SER, is defined as the ratio of radiation doses required to produce a given effect in the absence and presence of the drug, then the SER need not necessarily be as large as the OER. A large SER value is desirable, but an SER as low as 1.3, for example, could still be of significant clinical benefit. Secondly, the sensitizer should not increase the radiosensitivity of normal, aerobic cells to the same extent, or the benefits will be nullified. Thirdly, the sensitizer must be able to penetrate into the centre of the tumour; this implies ability to cross the barrier 14 presented by the cell membranes of the surrounding tumour tissue. Generally, this means that the drug must have lipophilic properties; however, water-solubility is also required for administration. Fourthly, the sensitizer must be metabolically stable, otherwise the drug will be consumed before reaching the hypoxic target cells. Finally, the toxicity and side-effects of the drug must be kept within a tolerable limit. This last criterion has proven to be the most difficult to satisfy, as it is only assessed in the clinical situation. 1.8 Radiosensitizing and Hypoxic Toxic Effects of Nitroimidazoles In 1973, Foster and Willson^ suggested that available drugs be screened for structures with radiosensitizing potential. There were two clinically used, antibacterial drugs that were a nitrofuran derivative-*0 and the 5-nitroimidazole drug, l-(2' -hydroxymethyl)-2-methyl-5-nitroimidazole (flagyl or metronidazole) (Figure 5), which appeared to be good radiosensitizers in vitro. The electron-affinity of these compounds had been measured. ' Another class of nitroaromatic compounds which showed more promise was the 2-nitroimidazole derivatives/-* Of this class, l-(2-nitro-l-imidazolyl)-3-methoxy-propanol (misonidazole) (Figure 5) has received by far the most attention both in laboratories and in clinics/-* The SER value for misonidazole has been reported to be 1.6 at 300 ,uM in vitro?* Within a nitroimidazole series the in vitro radiosensitizing ability correlated with electron-affinity/^ Anticancer drugs, in many cases, must be delivered systematically (i.e. orally, intravenously), rather than directly at the tumour. Thus, toxic side-effects on normal tissues are frequently encountered. It is critical, then, that drugs of outstandingly low systemic toxicity be selected. For example patients receiving misonidazole may experience neurotoxicity/-* Usually, the neuropathic side-effects have been temporary, and they can be avoided by limiting the total drug dose. However, such limitations 15 Figure 5: S t r u c t u r a l formulae of a ni t r o f u r a n , metronidazole and misonidazole 16 reduce the l ike l ihood of achieving effect ive radiosensit ization. Fur ther , the 2 -ni t roimidazoles are relat ively tox ic, especially to hypoxic c e l l s / ' * Th is property might at f i rst be considered an advantage for eradication of hypox ic cells, and indeed many studies are being directed at the interaction between toxic i ty and radiosensit izing properties. There was a suggestion that the radiosensit izing abi l i ty and hypoxic and oxic toxic i ty of some nitroimidazoles were consequences of their e lect ron-af f in i ty , i.e. the ease wi th wh ich they could be reduced to the nitro anion radical , R - N O ^ The metabolic reduct ion of aromatic nitro compounds by mammal ian systems v ia nitroreductases was f irst studied by Fouts and Brodie in 1957.^° They made several important observations: enzyme act iv i ty was found in both the aqueous soluble fract ion and the microsomal f ract ion of l iver; there were large dif ferences in act iv i ty of species; the enzyme system showed few r ig id structural requirements for substrates. Fur thermore, the mammal ian nitroreductase act iv i ty was strongly inhib i ted by air. The mechanism of oxygen inh ib i t ion of the enzymatic ni troreduct ion was elucidated by the work of Mason and H o l t z m a n , ^ esr spectroscopy being used to demonstrate the format ion of the nitro anion radical , dur ing the hypoxic microsomal reduct ion of ni trobenzoic a c i d . ^ This showed that the in i t ia l step in the reduction process was a one-electron transfer f rom the enzyme to the nitro compound, and suggested the possibi l i ty of a free radical mechanism in the oxygen inh ib i t ion of the reduct ion. In addi t ion, these workers suggested that the in i t ia l one-electron step in the ni t roreduct ion cont inued in the presence of oxygen; however, the nitro anion radical was considered to be re -ox id ized rapidly to the parent compound. Thus no net disappearance of the nitro compound was observed, and no esr signal f rom the radical was detected under aerobic condit ions. Oxygen was reduced to superoxide, result ing in 17 oxygen depletion and the build-up of superoxide. The toxic superoxide product was thought to be rendered harmless by the action of superoxide dismutase and catalase (Figure 6). However, the actual toxic product from the nitroimidazoles has remained elusive, and its mode of action (e.g. interaction with DNA, proteins or membranes) unknown. 1.9 Sulphvdrvl Binding Compounds as Radiosensitizers A hydrogen donation model was first proposed in 1954 to explain the reduced radiosensitivity found under hypoxia in the presence of sulphydryl (SH) compounds. It was postulated that a free radical is produced in the target molecule by radiation, and that subsequent reactions of this free radical lead to permanent biological damage. In the presence of a sulphydryl compound a hydrogen atom is transferred to the target radical before it has time to decompose, and thus the chemical lesion is repaired (Figure 7). The thiol radical RS" then recombines to form the oxidized dithiol species (RSSR). On the other hand, Loman et al^ have proposed an alternative mechanism of the SH protection that could complement the H-donor model. During irradiation, organic free radicals arise in the cell by both direct and indirect mechanisms. The model of Loman et al. proposes that these organic free radicals are eliminated by SH compounds through a process of hydrogen donation. But an essential difference between these two models is that in the first process, repair of the target molecule has occurred, whereas in the second process, free radical intermediates are eliminated and indirect damage to the target molecule is prevented (Figure 7). One of the mechanisms proposed to explain the sensitizing effect of certain compounds assumes that the sensitizer acts by binding free, non-protein sulphydryls or thiols (NPSH) groups which normally have a protective effect in irradiated cells/'* One 18 (a) Reductase R-N02 < (superoxide) / (reduced) j V / \/ +e ( -e ^ Reductase^ R-NO, (oxidized) (hypoxia) (reduced products) L SOD (b) H + + 0 2 > h (0 2 + H202) CATALASE H2°2 > \ 0 2 + H20 NET: H + + e" + 0 2 —•* H+ + o~ 3/4 02 + \ Figure 6: Mechanism of oxygen 'inhibition' of nitroreduction (a) The i n i t i a l product of nitroreduction i s the nitro anion radical. Under hypoxia, this species i s further reduced to unidentified products, including perhaps, toxic and mutagenic species. Under aerobic conditions, the radical anion i s re-oxidized quickly to the parent nitro compound. (b) the resulting superoxide i s detoxified by superoxide dismutase (SOD) and catalase. 19 i r r a d i a t i o n i r r a d i a t i o n damage to target + RSH + AH + RS-1 1/2 RSSR Proposed mechanisms of sulphydryl radioprotection: (a) The hydrogen donation model and (b) the Loman model, AH = organic molecules i n the c e l l . 20 such sulphydryl depleting radiosensitizer is N-ethylmaleimide (NEM). Experiments with E. coli have yielded results which are consistent with an SH binding mechanism.^ Another example is the use of buthionine sulphoximine (BSO), which on administration to mice or its incorporation in tissue culture media, results °in rapid depletion of intracellular glutathione concentration by inhibiting glutathione synthesis and enhanced radiosensitivity/^ Nitroimidazoles, such as misonidazole and metronidazole, may sensitize partly because of removal of endogenous NPSH. • This appears to be particularly true under conditions of prolonged incubation such as those used by Hall et al48 1.10 Platinum Complexes as Chemotherapeutic Agents and Radiosensitizers In 1969, Rosenberg et al. described several complexes of platinum which have significant anti-tumour activity/^ Since that time, cis-diamrninedichloroplatinum(II) (c/s-DDP) has rapidly attained clinical recognition for use alone and in combination with other chemotherapeutic agents/0. Chemists have put considerable effort into a search for related complexes with improved therapeutic properties/0 Platinum complexes bind to DNA, - 5 - 2 - 5 - * and this presumably is the means by which they kill the rapidly dividing tumour cells. However, the exact nature of the lethal lesion remains the subject of much research and controversy. The ris-isomer is much more effective for killing tumour then the corresponding trans-isomer. One plausible explanation is that c/s-DDP forms a lethal intrastrand crosslink (in addition to the known interstrand crosslink which both isomers can form), which is more difficult to be recognized and requires a longer time to repair this type of damage^ (Figure 8). The DNA binding action of platinum complexes suggested a rationale for combining the use of the Pt drugs with radiotherapy, and indeed interaction between Figure 8: Possible types of DNA in t e r a c t i o n s with eis-DDP (adapted from r e f . 55). 22 radiation and c/s-DDP was reported by Richmond and Powers in 1976, who showed that a concentration (50 JJM) of cis-DDP sensitized the spores of Bacillus megaterium especially under hypoxic conditions/6 In 1978, Douple and Richmond demonstrated that radiation sensitization of hypoxic mammalian cells was obtained with the use of c/s-DDP, trans-DDP and ds-dichlorobis(aziridine)platinum(II). Since this time, many Pt complexes have been examined in various systems as radiosensitizers/0'^'•^'-'*>,<*0 Platinum complexes possibly act as radiosensitizers by interfering with the cell's repair mechanisms in some way by binding to DNA. Some groups have been studying sublethal and potentially lethal radiation damage, and find that repair of such damage is inhibited by Pt complexes/^ Other groups have suggested that some Pt complexes, such as cis-DDP and cis, trans, c;'s-dichlorodihydroxybis(isopropylamine)platinum(IV), show enhanced radiosensitizing effects by reacting with the protecting NPSH compound;6'2 however, the reasons are still not yet understood/0 1.11 Metal Complexes as Radiosensitizers The early studies on metal salts, chosen for their oxidizing ability, mostly used bacterial systems and no real mechanistic pointers emerged from these results because no uniform parameter of activity was adopted. However, the examples do demonstrate the potential for study and the need for systematic evaluation of closely related complexes (Table I). The radiosensitizing effects of copper salts in mammalian cells have been demonstrated. Whereas Cu(II) salts had little or no toxicity in normal oxygenated cells, incubation under hypoxic conditions and the reduction to Cu(I), or the introduction directly of cuprous solutions in the oxygen-free medium, resulted in enhanced sensitivity to radiation, with a dose modifying factor of 1.5 being obtained in the 23 Table I Metal Compounds with Radiosensitization A b i l i t y Metal Biological Test Systems Examples References Ft Bacterial Mammalian ois -DDP Pt(nitroimidazoles) 50, 57 70 Cu Fe Co Ni Mn Rh Bacterial Mammalian Mammalian Bacterial Mammalian Bacterial Mammalian Bacterial Mammalian CuCl. [Fe(CN)5N0] 2-[Fe(CN)6] 3-vitamin B. 12 3+ Co(NH3)6 Nidapachol^ MnO, Rh(II) carboxylates 63 64 65 64 66 69 70 71 Lapachol i s 2-hydroxy-3-isoprenyl-l,4 - naphthoquinone. 24 mammalian system. This example of reduction-enhanced cytotoxicity is of particular relevance because of the role of copper as an essential trace element and its function in many intracellular redox processes. The nitroprusside salt, [Fe(CN)5NO] , was an effective sensitizer in V-79 cells at 10" M. This effect was attributed to toxic ligand (CN~) release. Ferricyanide [Fe(CN)g] , is a sensitizer of bacterial cells, the activity being attributed to its thiol binding capacity6"^ and no sensitization of oxic cells by this compound was seen.6^ A further apparent example of reduction enhanced cytotoxicity with metal complexes is that of the hexamminecobalt(III) ion, [Co(NH 3) 6] 3 + , when reduction to Co(II) also resulted in enhanced sensitivity to radiation.66 Cobalt chelates such as [Co(2,2'-bipy)3]3+, however, did not sensitize cells to radiation.67 The more recent results on the interaction of some inert Co(III) complexes with D N A 6 * may well be relevant to these biological results especially as a 1,10-phenanthroline chelate, such as c/5-[Co( 1,10-phenanthroline)2(NO2)2]+-A nickel chelate of lapachol, 2-hydroxy-3-isoprenyl-l,4-naphthoquinone, has also been studied for its radiosensitizing ability.6 9 The chelation is through the 2-OH and adjacent keto group. The results show that the Ni chelate is a moderate sensitizer (hypoxic CHO cells, SER=1.3 at 100 JJM), and is better than the Cu or Zn complex. This Ni chelate also enhances radiation-induced breaks in DNA of hypoxic cells.69 In vitro sensitization of rhodium complexes containing carboxylates has been 7 / reported. The Rh carboxylate complexes sensitize in both air and nitrogen, with the greater effect generally occurring in nitrogen. Radiation chemical experiments show the Rh complexes have substantially low reduction potentials and do not undergo electron 25 transfer reactions, adduct formation, or interaction with radicals derived from DNA bases. The increase in radiation sensitivity of cells treated with Rh(II) carboxylates was attributed to the ability of the Rh compounds to deplete intracellular thiols. 1.12 Rationale for Using Metal-Radiosensitizer Complexes The dose limiting toxic side-effects of nitroimidazoles have prevented the attainment of maximum sensitization in clinical use.-^ A desirable aim is to decrease the overall concentration of radiosensitizer, while retaining a high drug concentration at the target of ionizing radiation damage, the DNA. If this could be achieved with good radiosensitizers such as misonidazole, one might be able to modify or decrease the toxic side-effects which currently limit the use of these drugs. Studies on DNA as a radiation target may well lead to the development of new species having radiosensitizing and DNA-binding properties. In this respect, attachment of a radiosensitizer to transition metals that are known to bind to DNA may be of value. As a consequence of binding to DNA, the metal could act as a carrier, and increase the local concentration of radiosensitizer at the target and thus decrease the deleterious side-effects;7,2 further, the metal itself may exert a radiosensitizing effect (see section 1.11), The hypothesis is that use of a complex (M-S) between a sensitizer (S) and a metal template (M), which binds to DNA, may result in a radiosensitizer with a better therapeutic index than either "parent" drug because of improved radiosensitization - the primary sensitizer S would be located at the target DNA, by virtue of the binding properties of the metal. In addition, the template moiety itself could be contributing to the radiosensitization. Decreased toxicity might be expected because the overall concentration of S would be lowered (Figure 9). Extensive structure-activity relationship using free nitroimidazoles have established a correlation between the in vitro radiosensitizing efficiency and the one-S = S e n s i t i z e r + = M e t a l Complex +S =Metal M o l e c u l e D i s t r i b u t e d C o n c e n t r a t e d a t DNA S e n s i t i z e r Throughout C e l l M F i g u r e 9: R a t i o n a l e f o r u s i n g m e t a l - r a d i o s e n s l t i z e r complexes. 27 electron reduction potential of the nitro group. Metal binding renders this potential more positive because of the inductive effect resulting from bond formation of the lone-pair nitrogen of the imidazole ring to the metal. ' In principle, this allows for "fine-tuning" of the reduction potential of the nitro group, and more efficient sensitization. If attachment of a metal to a sensitizer moiety does not diminish radiosensitization drastically or increase toxicity, this proposed design will have advantages over straightforward combination of the two types of drugs.7"* However, a combination of chemical stability, solubility, and pharmacological factors (e.g. partition coefficients, distribution) will also need to be optimized.-*'* 1.13 Metal-Nitroimidazole Complexes as Radiosensitisers Platinum is one candidate for such a carrier because of its known binding 52 53 properties to DNA. ' J Some Pt complexes are already used as chemotherapeutic agents'*9"50"5^-5-* and some also have moderate radiosensitizing ability." 5 7 , 6 0' 6^' 7 3 In addition, Pt drugs exhibit their chemotherapeutic and radiosensitizing ability at micromolar levels, whereas much higher, overall cellular concentrations of misonidazole are required to achieve optimum radiosensitization. Initial results on m-tPtC^metronidazole^] (FLAP) were promising with a reported SER of 2.4 and little toxicity,7^ but later studies7'*'7""' failed to corroborate these findings and the complex was at best a weak to moderate radiosensitizer. A general problem with the bis(nitroimidazole) complexes is their lack of water solubility. Methods to circumvent this problem include the syntheses of complexes with only one nitroimidazole and also containing an ammine ligand. One example is [PtC^CNH^X^-NO2-I1T1)] which showed DNA binding property as well as having better 28 radiosensitizing enhancement than the bis(nitroimidazole) complexes.76 Palladium has been used in an attempt to target nitroimidazoles but these 77 complexes generally dissociated in solution. However, a postirradiation-sensitizing effect was observed in vivo (KHT sarcoma), possibly because of Pd ions exerting an effect on the repair of potentially lethal damage. 75 Chibber et al. have been assessing the interaction of Rh(II) species with 7 5 70 radiation, and have used Rh(II) to target misonidazole and analogs. ' A Rh complex was better than RSU-1111 (a 2-nitroimidazole ligand) alone, or the corresponding platinum complex (RSU-1113), in sensitizing hypoxic V79 cells.7^ 1.14 Ruthenium Complexes as Anticancer Agents The metal, ruthenium, is used in the present study for the development of metal-based radiosensitizers. Approaches using ruthenium as antineoplastic agents have been related to the mechanistic action of the clinically employed c/s-DDP, namely, via o n DNA binding. In both their initial DNA binding site and in the oncological consequences of their DNA interactions, ruthenium ammine complexes such as [Ru(NH3)jH20]^+ resemble m-DDP. Their activity is thought to be exerted, at least initially, by coordination to guanine N7 sites on cellular nucleic acids. In vitro biochemical studies demonstrate that Ru(II) and Ru(III) compounds are active in inhibiting DNA synthesis and possess mutagenic activity in the Ames Salmonella and related assays. It has been suggested that ruthenium complexes with bipyridine ligands intercalate into DNA prior to covalently binding to guanine residues. Preliminary studies of the binding of ruthenium cations to DNA indicate that advantage can be taken of their inertness to ligand substitution once coordination 29 has been established at a DNA site. Once coordinated to a nitrogen base, Ru(II) and o c Ru(III) usually remain bound to the same base. Additional advantage can be taken from the ready availability of both Ru(II) and Ru(III) oxidation states under physiological conditions, whereby the drug is inactive in oxic cells as Ru(III), but which becomes active as a radiosensitizer upon reduction to Ru(II) in hypoxic cells. 1.15 Nature of Sulphoxides as Ligands The sulphoxides, of general formula R2SO, are examples of ambidentate ligands which may bind to metal ions through either the sulphur or oxygen. Both atoms have lone pairs of electron, enabling the sulphoxide to act as a Lewis base. The three resonance structures for the S-O bond are shown in Figure 10; structure II is R7 dominant. Both steric and electronic effects have been shown to dictate the choice of donor atom and these effects may be summarized** as follows: (a) Steric Effects: These may be manifested by the sulphoxide, and other ligands in the coordination sphere. (i) effect of sulphoxide ligand Both the [PdCl2(dmso)2] and [PdCl2(mbso)2] (mbso = methyl-3-methylbutyl sulphoxide) complexes have both sulphoxides bound through sulphur; however, the op sterically more demanding mbso ligand results in a trans configuration, as opposed to this ds-geometry of the dmso analogue. (ii) other ligands This is demonstrated in complexes of the type [M(dppe)(dmso)Cl]+ (where dppe = l,2-bis(diphenylphosphino)ethane, M=Pd, Pt). Bonding through sulphur is disfavoured because of interaction of the methyl groups with the phenyl groups of the 30 >s-g s^=6= >s^ o. i n i n gure 10: The three resonance structures f or the S-0 bond of sulphoxides. 31 bulky phosphine ligand, and thus bonding at the sterically-unrestricted oxygen is preferred/ 0 (iii) size of central metal ion Within the same column of the Periodic Table, the complex trans-i OJ [FeCl2(dmso)4] is all O-bonded, whereas c/s-[RuCl2(dmso)4] has three S- and one OJ O-bonded dmso ligand. (b) Electronic Effects: A graphic demonstration of how purely electronic effects control the mode of coordination of the sulphoxide ligand is demonstrated in dmso adducts of rhodium(II) carboxylates, [Rh2(02CR)4(dmso)2]. When R=CH 3 or C 2 H 5 , S-coordination of dmso occurs, but when R=CF 3 the highly electronegative C F 3 substituents force preference for O-bonding/'' 1.16 Biological Activity of Metal-dimethyl sulphoxide Complexes Intensive research has been carried out to study the mechanism of action of cis-DDP and to develop suitable analogues with better pharmacological properties by improvement of lipophilicity, lower toxicity and higher water-solubility/^ Metal complexes with dmso as ligands are generally water-soluble species, and the lipophilicity of dmso and its low toxicity have been of interest in applications of transition metal chemotherapy/'* The most comprehensively studied complex is m-RuCl 2 (dmso) 4 / 7 , 9 ^$$$6 I n a series of studies comparing its biological activity with that of c/s-DDP, the Ru species 97 was shown to produce filamentous growth in E. coli and induce alpha prophage. The Ru complex is mutagenic in strain S. typhimurium, carrying the His mutation, specific QO for base pair substitution mutagens; it has in vivo anti-tumour activity at doses of 32 4 0 0 mg/kg, which indicates a high L D ^ Q for this complex. Other metal-sulphoxide complexes such as [Ru(dmso)6]2+, [RhC^dmso^], [RhCl(dmso)5]2+ and [Pt(dmso)4]2+ have similar behaviour but none of the complexes show more anti-tumour activity than c/5-RuCl2(dmso)4." Apart from the understanding of the mechanism of anti-tumour action of cis-DDP, a major interest in the area of nucleic acid-metal ion interactions is the possibility of sequence determination, using specific binding of metal complexes. Reactions using [PtCl^dmso)]" with DNA and RNA were found, in which the complex could be used as a differential staining agent for adenine and guanine residues in both polynucleotides/00 The differential staining is made possible by the binding of two Pt atoms per adenine residue at both pH 6.0 and 7 . 5 , while with guanine at the lower pH, one Pt, and at higher pH, two Pt atoms are bound/ 0 0 An interesting difference emerges between the octahedral Ru(II) and Rh (III) systems and the square planar Pt(II) cases; with adenine, only monodentate binding via N 7 is found for both Ru and Rh systems, whereas with the Pt system, binding can be via N 7 of adenine or guanine/0^ Hydrogen-bonding between the exocyclic 6 - N H 2 group of adenine and the dmso oxygen atoms or chloride atoms (both H-bond 102 acceptors) in the octahedral sphere may dictate specificity, these interactions being minimized in a square-planar complex/0"^ Dimethyl sulphoxide, as a ligand, has been shown to possess many of the properties desirable for metal-based chemotherapeutic agents. Complexes containing 92 93 103 dmso, rather than ammine, result in more water-soluble, less toxic derivatives. ' * In chemical studies, liganded dmso may be displaced by pyrimidine and purine bases, 33 the products and binding sites being dictated by the nature of the starting complexes/-*'70-* 1.17 Short-term In-Vitro Assays Numerous short-term in vitro test systems have been developed to demonstrate whether or not a chemical is a genotoxic substance. Some of the genetic damages employed for this purpose include mutation, DNA damage, chromosome aberration, and gene conversion/0** These sensitive, rapid, and economical tests for genotoxicity are also applied to predict the carcinogenic properties of individual compounds70^ or mixtures of chemicals/ 0 6 The usefulness of the tests is based mainly on a fairly good correlation between results of in vivo carcinogenicity tests on various rodents and in vitro data from genotoxicity assays/ 0 7 Before a metal-based chemical compound can be introduced as an effective radiosensitizer, it would be useful to know whether such a complex exerts a genotoxic effect on normal cells. The genotoxic effect of some Pt coordination compounds has been studied by these short-term in vitro assays, which have shown to induce sister chromatid exchange, mutation, micronuclei induction, and chromosomal aberrations/ 0*" 7 7 7 In addition, some Rh(I) and Ru(II) complexes which have anti-tumor activity have also been shown to raise the number of revertants of S. typhimurium over the background level (i.e. the complexes possess mutagenic activity for bacteria)/ 7 2 1.18 Objectives The purposes of this project were to synthesize and characterize certain Ru-nitroimidazole complexes with potential as radiosensitizers as adjuncts to radiotherapy, and to study the factors which may contribute to the sensitizing process. Based on the literature data available as described above, and expertise available through earlier 3 4 research work in the UBC chemistry department on Ru/sulphoxide systems, • ' the compounds initially selected for the present work were dihalidebis(sulphoxide) complexes of ruthenium(II), containing also known radiosensitizing nitroimidazole ligands. The potential of such Ru-radiosensitizer complexes as adjuncts to radiation is studied largely in vitro for different sulphoxides and halides complexes in mammalian cells. Certain studies addressing mechanisms of radiosensitization and some preliminary in vivo data are presented. 35 CHAPTER TWO MATERIALS AND METHODS 2.1 Chemicals The 4-nitroimidazole, 2-nitroimidazole, Ellman's reagent (5,5' -dithiobis-(2-nitrobenzoic acid)), tris-(hydroxymethyl)aminomethane (Tris), [tris-(hydroxymethyl) aminomethane hydrochloride] (Tris-Cl), and disodium ethylenediaminetetraacetate dihydrate (EDTA) were purchased from Sigma Co. Ltd. (St. Louis, MO), while tetramethylene sulphoxide (tmso), methylphenyl sulphoxide (mpso), N-methylimidazole, 2-methylimidazole, metaphosphoric acid, ds-DDP, 1,10-phenanthroline, 5-NO 2-l,10-phenanthroline and 5-C1- 1,10-phenanthroline were from Aldrich Chemical Company (Milwaukee, WI). Metronidazole (metro) was obtained from Poulenc Ltd. (Montreal, Quebec) and Eagle's Minimal Essential Medium (MEM), penicillin, streptomycin sulphate, kanamycin and fungizone were obtained from Gibco Canada Inc. (Burlington, Ontario). Fetal calf serum was a Bocknek (Ontario) product. The restriction enzymes PvuII, BamHI and EcoRI were purchased from Bethesda Research Laboratories Life Technologies Inc. (Gaithersburg, MD). The R u C l 3 3 H 2 0 (35% Ru) was received on loan from Johnson Matthey Ltd. and was used as provided. Misonidazole (miso) and Desmethylmisonidazole (De-miso) (Figure 11) were donated by Roche Products Limited (Welwyn, U.K.). Of the 4-nitroimidazole ligands, RSU-1170, RSU-3100 and RSU-3159 were kindly supplied by Dr. T.C. Jenkins,^ and RSU-3083 by Dr. I.J. Stratford^ (Figure 12). The compound SR-2508-. was a gift from the National Cancer Institute of the United States. These chemicals were used without further purification. All other common reagents used were at least reagent grade. Methanol was dried by distilling over Mg metal turnings for at least four hours before being collected and used. The b The RSU terminology is used by Drs. T.C. Jenkins and I.J. Stratford (Medical Research Council, Harwell, Didcot, England), the suppliers of these ligands whose structures are given in Figure 12; see for example ref. 114. c SR = Stanford Research. 36 R = H OH R = CH2-CH-CH2~OH 2-nitroimidazole (2-NO -Im) Desmethylmisonidazole (De-miso) 0 II R = CH -C-N-CH -CH OH = Etanidazole (SR-2508) 2 | 2 2 H OH I R = CH -CHCHOCH = Misonidazole Figure 11: St r u c t u r a l formulae of the 2-nitroimidazole ligands. 37 R = H, R' = H R" = H : 4-nitroimidazole (4-N02~Im) R =CH , R' = H, R" = H : N-methyl-4-nitroimidazole (NMe-4-NO -Im) R =CH2CH(OH)CH2N^] , R' = H, R" = CH 3 : RSU-1170 R =CH3, R' = N-CH2-CH2-OH, R" = CH 3 : RSU-3083 R =CH3, R' = OPh, R" = H : RSU-3100 ;- = \\ ... • R =CH , R' u / , R" = H : RSU-3159 3 f—NH S R =CH3, R' =C1, R" = H : CMNI Figure 12: Str u c t u r a l formulae of the 4-nitroimidazole ligands. 38 procedures for preparing stock solutions, buffers, rat liver S9 fractions, Sephadex G-50 columns and linearized pSV2-gpt plasmid are given in Appendices I-VI. 2.2 Physical Techniques and Methods Infrared spectra (KI discs or Nujol mulls) were recorded on a Nicolet DX FTIR spectrometer; *H and 13C{*H} nmr spectra (in CDCI3 vs. TMS) were recorded on a Varian XL-300 or Bruker WH-400 instrument operating in the Fourier transform mode. Conductivity data were measured in water at RT under N 2 with a conductivity cell connected to a Beckman Serfass Conductance Bridge. The pH-titrations were carried out in aqueous media using a Corning 12 pH-meter and a combination glass-electrode: typically, 25 mL of aqueous HC1 (2 mM), in the absence and presence of added complex (2 mM), were titrated with standard 4 mM NaOH solution; differences between the two titration curves allow for a simple determination of p K a values.77-5 In each experiment, about fifteen measurements were made, and care was taken to ensure that each pH reading recorded represented equilibrium conditions. Carbon dioxide was excluded by maintaining a flow of nitrogen through the titration cell fitted with a cap. Optical spectra were recorded on a Perkin Elmer 552A spectrometer with 1 cm path length quartz cells in air. For recording UV/VIS spectra under anaerobic conditions, one type of anaerobic cell as shown in Figure 13 was employed. A solid compound was put in the L tube; a solvent was placed in the 1 cm path length quartz cell, and the two mixed under N 2 . Microanalyses were performed by Mr. P. Borda of the department of Chemistry, University of British Columbia. 2.2.1 X-rav Photoelectron Spectroscopy The XPS data were taken on a Varian IEE-15 spectrometer at a background pressure < 10 torr, using MgK radiation (1253.60 eV) at 280 W and an analyzer pass energy of 100 eV. Resolution was 1.6 eV at full width, half maximum on a gold standard, Au(4f 7 / 2). Each sample was lightly dusted onto 39 V i Figure 13: Schematic representation of an anaerobic spectral 40 single-sided 3M scotch tape and introduced into the spectrometer on a rapid insertion probe cooled by liquid N 2 . All binding energies (BE) were calibrated with reference to the C Is peak (284.0 eV) of graphite. There was no evidence for any sample deterioration during irradiation. 2.2.2 Electrochemical Methods (i) Polarographic Measurements The reduction potentials of the compounds were determined using a Princeton Applied Research Model 364 polarographic analyzer with a dropping mercury electrode. In a typical experiment, the compounds were dissolved in 0.01 M phosphate-buffered saline (PBS) solution at concentration of 500 JJM solution. One mL of this solution was then added to the measurement vessel which contained 9 mL of a 0.01 M PBS solution previously purged with N 2 . The half-wave reduction potential (Ej^2) of the compound was measured in the vessel at RT, under N 2 , in mV versus a saturated calomel electrode. (ii) Cyclic Voltammetric Measurements To obtain cyclic voltammograms a standard three compartment H-cell, fitted with fine frits, was used in conjunction with an E G and G PARC Model 175 Universal Programmer (controlling the range of sweep and the scan speed) linked to a model 173 PAR Potentiostat (controlling the potential of the auxiliary electrode). The traces were photographed by a Tektronics oscilloscope camera (C-12) using Polaroid film. In a typical experiment the compounds were dissolved in 0.01M PBS solution and the solution purged with argon via a teflon lead for about 10 min. The teflon lead was then raised above the level of the liquid to keep a layer of argon above the solution. The solution was periodically stirred and recharged with argon although before a cyclic voltammogram was 41 run, the agitation of the solution had to be stopped. The scan speed used was 500 mV/s. 2.3 Preparation of N-methvl-4-nitroimidazole The compound N-methyl-4-nitroimidazole (NMe-4-NC>2-Im) was prepared according to a literature procedure.776 Concentrated HNO^ (25 mL) was added slowly to N-methylimidazole (10 mL, 0.12 mol) and the reaction mixture cooled to 4°C on an ice-bath, when a vigorous reaction took place. Concentrated H 2 S 0 4 (25 mL) was then added slowly to the reaction mixture, which was then boiled gently until all nitrous fumes were gone (approximately 4 h). The mixture was then poured into 50 mL of ice and distilled water (1:1) and made basic with concentrated ammonia (approximately 50 mL). The mixture was then extracted 3 times with CHCI3 (150 mL total), the non-aqueous layer was collected, and the solvent removed completely by rotovap to give a white solid (3.88 g, 25% yield). Anal, calcd. for C 4 H 5 N 3 0 2 : C 37.76, H 3.93, N 33.04; found: C 37.60, H 3.92, N 33.00. 2.4 Preparation of Complexes The spectroscopic data (UV/VIS, IR, nmr, XPS) for the ligands and complexes are discussed later in sections 3.1-3.3. 2.4.1 m-RuCl 2(dmso) 4. 1 (i) The precursor complex ds-RuCl2(dmso)4, which contains one O-bonded and three S-bonded sulphoxides^ was prepared by a literature procedure/6 Some RuCl 3 "H 2 0 (3 g, 11 mmol) was refluxed together with dmso (8 mL) for 2 h under N 2 . The resulting dark orange solution was then cooled to RT, and acetone (40 mL) was added slowly to precipitate out a yellow product which was filtered off in air, washed with acetone and diethylether, and dried in 42 vacuo at RT (1.18 g, 63% yield). Anal, calcd. for CoH 2 4 0 4 S 4 C l 2 R u : C 19.83, H 4.96, Cl 14.65; found: C 19.78, H 4.83, Cl 14.69. Am a xnm(log e), H 2 0: 350 (2.69), 300 (2.45). (ii) Another route for making 1 was used, according to that reported by James et al.95 Some RuClj^f^O (3g, 11 mmol) was added to previously dried methanol (50 mL), in a two-necked flask equipped with gas-inlet tube, reflux condenser and magnetic stirrer. The solution was refluxed under H 2 (1 atm) overnight (approximately 16 h) to give a deep blue color. To this Ru blue methanolic solution was added 4 mL of dmso. Refluxing under H 2 was continued for 16 h to yield a red solution. No reaction was observed when the refluxing was carried out under N 2 or Ar. The solution was cooled to yield 1.0 g (59%) of yellow cubes, which were filtered off in air. Found: C 19.67, H 5.04, Cl 14.39. 2.4.2 RuCl2(tmso)4. 2 To the Ru blue solution, as mentioned above, was added 3 mL of tmso. Refluxing under H 2 was continued for 6 h yielding an orange solution, which was cooled to RT when a yellow precipitate formed. Again no reaction was observed when the refluxing was carried out under N 2 or Ar. The solid was filtered off in air, washed with acetone and dried in vacuo at RT (1.62g, 70% yield). Anal, calcd. for C , 6 H 3 2 0 4 S 4 C l 2 R u : C 32.61, H 5.48, S 21.74; found: C 32.49, H 5.50, S 21.67. Am a xnm(log C), H 2 0: 343 (2.54), 303 (2.37). 2.4.3 "RuCl2(mpso)2". 3 The complex "RuCl2(mpso)2" was prepared according to a literature procedure.^"* To the blue Ru methanolic solution prepared as described in section 2.4.1(H) was added mpso (1.3 g, 9.3 mmol), and the solution refluxed 43 overnight under H 2 . The solution turned brownish green and a yellowish orange precipitate was obtained when the solution was cooled to RT. The precipitate was filtered off in air, washed with ethanol and acetone and dried in vacuo at RT (2.9 g, 69% yield). Anal, calcd. for C I 4 H 1 6 0 2 S 2 C l 2 R u : C 37.16, H 3.57, CI 15.90; found: C 36.81, H 3.68, CI 15.68. 2.4.4 frans-RuBr2(dmso)4.4 Complex 1 (483 mg, 1 mmol) dissolved in CH 2 C1 2 (75 mL), and NaBr (205 mg, 2 mmol) dissolved in a minimum amount of water (approximately 3 mL), were mixed together, and sufficient methanol (approximately 40 mL) then added to make the two layers miscible. This mixture was refluxed overnight (approximately 16 h) under N 2 to give an orange solution, which was roto-evaporated to dryness. The resulting pale orange solid was dissolved in dmso (3 mL) and the solution heated for 1 h. Acetone (approximately 25 mL) was added to precipitate a pale orange solid, which was filtered off in air, washed with acetone/water mixture (2:1) and acetone, and then dried in vacuo at RT (0.34 g, 60% yield). Anal, calcd. for C g H 2 4 0 4 S 4 B r 2 R u : C 16.76, H 4.22, S 22.36; found: C 17.01, H 4.24, S 22.46. A m a x n m (log C), H 2 0: 465 (2.32), 314 (2.17). 2.4.5 RuCl2(dmsoW4-NQ2Im)2. 5 Dry methanol (15 mL) was added to RuCl2(dmso)4 (483 mg, 1 mmol) and 4-N0 2-Im (282 mg, 2.5 mmol), and the mixture refluxed for 4 h under N 2 . The resulting brown solution was cooled to RT and diethylether (40 mL) added slowly to precipitate out a brown product, which was filtered off in air, washed with acetone and diethylether, and dried in vacuo at RT (0.36 g, 65% yield). Anal, calcd. for C 1 0 H l g p 6 N 6 S 2 C l 2 R u : C 21.66, H 3.25, N 15.16; found: C 22.00, H 3.06, N 15.38. UV/VIS - see text, section 4.2. 44 2.4.6 RuCl2(dmsoWNMe-4-N02Im). 6 The complex RuCl2(dmso)4 (483 mg, 1 mmol) was suspended in an isopropanol solution (20 mL) containing NMe-4-N0 2-Im (254 mg, 2 mmol), and the mixture was refluxed for 4 h under N 2 . The resulting blue solution was cooled to RT, when a blue precipitate formed. This was collected, dissolved in C H 2C1 2 (10 mL), reprecipitated with diethylether (20 mL), filtered off in air and dried in vacuo at RT (0.36 g, 8 0 % yield). Anal, calcd. for C o H 1 70 4 N 3 S 2 C l 2 R u : C 21.06, H 3.73, N 9.22; found: C 20.99, H 3.93, N 9.25. A m a x nm(log £), H 20: 303 (4.32), 427 (3.06). 2.4.7 RuCWdmso)2(RSU-l 170) 2. 2 The nitroimidazole RSU-1170 (226 mg, 1 mmol) was mixed with RuCl2(dmso)4 (242 mg, 0.5 mmol) in dry methanol (10 mL), and the mixture refluxed for 4 h under N 2 . The resulting green solution was cooled to RT, and diethylether (30 mL) added slowly to precipitate out a green product, which was filtered off in air, washed with acetone and dried in vacuo at RT (0.25 g, 6 5 % yield). Anal, calcd. for C 2 2 H 4 Q O O N O S 2 C 1 2 R U : C 33.85, H 5.13, N 14.35; found: C 33.51, H 5.39, N 13.91. A m a x n m (log e) H 20: 310 (4.39). , 2.4.8 RuCl2(dmsoWRSU-3083)2. 8 The compound RSU-3083 (200 mg, 1 mmol) was added to RuCl2(dmso)4 ((242 mg, 0.5 mmol) in ethanol (10 mL), and the mixture refluxed for 4 h under N->. The resulting red solution was cooled to RT and filtered; slow addition of diethylether (30 mL) yielded a red solid, that was filtered off in air, washed with ethanol and acetone, and dried in vacuo at RT (0.25 g, 70% yield). 45 Anal, calcd. for C 1 6H 3 6OoN 8S 2Cl 2Ru: C 29.6, H 4.98, N 15.37; found : C 29.34, H 4.64, N 15.66. Am a x nm(log £ ), H 2 0: 385 (4.40), 450 (4.12). 2.4.9 RuCl2(dmso)2(RSU-3100)2. 9 The complex RuCl2(dmso)4 (242 mg, 0.5 mmol) was suspended in an isopropanol (10 mL) solution containing RSU-3100 (219 mg, 1 mmol) and the mixture refluxed for 6 h under N 2 . The resulting brown solution yielded a brown product as in the synthesis of 5. (0.25 g, 65% yield). Anal, calcd. for C 2 2 H 3 0 O 8 N 6 S 2 C l 2 R u : C 37.58, H 3.92, N 10.96; found C 37.72, H 4.34, N 10.68. A m a x n m (log e), H 2 0: 350 (4.13), 396 (4.09). 2.4.10 RuCl2(dmso)2(RSU-3159). 10 The complex RuCl2(dmso)4 (242 mg, 0.5 mmol) was added to RSU-3159 (225 mg, 1 mmol) in methanol (15 mL), and the mixture refluxed for 3 h under N2. The resulting dark green solution was filtered hot, and diethylether (30 mL) added slowly to the cooled filtrate to precipitate a green solid; this was filtered off in air, washed with acetone and diethylether, and dried in vacuo at RT (0.18 g, 75% yield). Anal, calcd. for Cj jHjgC^N^C^Ru: C 23.87, H 3.46, N 12.65; found: C 24.13, H 3.79, N 12.53. Am a xnm(log C ) H 2 0: 325 (4.22). 2.4.11 RuCl2(dmso)2(miso)2. 11 The complex RuCl2(dmso)4 (483 mg, 1 mmol) was added to miso (402 mg, 2 mmol) dissolved in toluene (10 mL), and the mixture refluxed for 6 h under N2. The resulting blue solution was reduced in volume, and hexanes (30 mL) added slowly to give a blue oily product. The oily product was cooled in liquid N2 to yield a blue precipitate, which was filtered in air, washed with diethylether and dried in vacuo at RT (0.36 g, 50% yield). Anal, calcd. for 46 C 1 8 H 3 4 O 1 0 N 6 S 2 C l 2 R u : C 29.63, H 4.69, N 11.51; found: C 29.81, H 4.95, N 11.86. A m a x nm(log t) EtOH: 325 (3.82), 638 (1.84). 2.4.12 RuClofdmsoWDe-miso^. 1 2 Dry methanol (15 mL) was added to RuCl2(dmso)4 (483 mg, 1 mmol) and De-miso (374 mg, 2 mmol), and the solution refluxed for 4 h under N 2 . The resulting blue solution was filtered hot, and the filtrate cooled to RT to give a blue precipitate. This was filtered off in air, washed with diethylether (20 mL) and dried in vacuo at RT (0.55 g, 78% yield). Anal, calcd. for C , 6 H 3 0 O 1 0 N 6 S 2 C l 2 R u : C 27.35, H 4.30, N 11.96; found: C 27.15, H 4.27, N 12.07. A m a x nm(log e) EtOH: 332 (3.57), 584 (2.01). 2.4.13 RuCl 2(dmsoW2-NQ 2-ImU. 13 The ligand 2-N0 2-Im (282 mg, 2.5 mmol) was added to RuCl2(dmso)4 (483 mg, 1 mmol) in dry methanol (15 mL), and the solution refluxed for 6 h under N2. The resulting blue solution was cooled to RT, when slow addition of diethylether (40 mL) gave a blue precipitate, which was filtered off in air, washed with hexanes and diethylether, and dried in vacuo at RT (0.36 g, 65% yield). Anal, calcd. for C 1 0 H ] 8 O 6N 6 S 2 C l 2 R u : C 21.66, H 3.25, N 15.16; found: C 21.32, H 3.37, N 15.31. A m a x nm(log Z ), EtOH: 332 (3.72), 420 (2.34). 2.4.14 RuCl 2( dmsoW metroW 14 The metronidazole (342 mg, 2 mmol) was mixed with RuCl2(dmso)4 (483 mg, 1 mmol) in dry methanol (15 mL), and the solution refluxed for 4 h under N2. The resulting brown solution was filtered, and diethylether (40 mL) added slowly to the filtrate to precipitate a brownish yellow solid; this was filtered off in air, washed with acetone and diethylether, and dried in vacuo at 47 RT (0.45 g, 67% yield). Anal, calcd. for C J ^ Q O O N ^ C ^ R U : C 28.66, H 4.51, N 12.53; found: C 28.81, H 4.58, N 12.26. Am a xnm(log Z ), H 2 0: 320(3.82). 2.4.15 RuCl2(tmsoW4-NQ2Im)2. 15 Dry methanol (15 mL) was added to RuCl2(tmso)4 (588 mg, 1 mmol) and 4-N0 2-Im (282 mg, 2.5 mmol), and the solution refluxed for 4 h under N 2 . The resulting brown solution was roto-evaporated to a reduced volume ( 3 mL) and diethylether/hexane (1:1) mixture (30 mL) added slowly to precipitate out a brown product, which was filtered off in air, washed with acetone and diethylether, and dried in vacuo at RT (0.33 g, 55% yield). Anal, calcd. for C 1 4 H 2 2 0 6 N 6 S 2 C l 2 R u : C 27.72, H 3.65, N 13.86; found: C 27.84, H 3.48, N 13.95. A m a x n m (log e), EtOH: 330 (3.88). 2.4.16 RuCl2(tmso)2(NMe-4-NQ2-Im)2. 16 The complex RuCl2(tmso)4 (588 mg, lmmol) was suspended in a methanol solution (20 mL) of NMe-4~N02-Im(254 mg, 2 mmol), and the mixture refluxed for 6 h under N 2 . The resulting blue solution was filtered hot and cooled to RT; slow addition of diethylether (30 mL) yielded a blue solid, which was filtered off in air, washed with acetone, and diethylether, and dried in vacuo at RT (0.44 g, 69% yield). Anal, calcd. for C 1 6 H 2 6 0 6 N 6 S 2 C l 2 R u : C 30.28, H 4.13, N 13.25; found: C 30.45, H 4.26, N 13.59. A m a x n m (log C) , H 2 0: 309 (3.92), 413 (2.87). 2.4.17 RuCl2(tmso)2(De-miso)2. 17 The De-miso ligand (374 mg, 2 mmol) was added to RuCl2(tmso)4 (588 mg, 1 mmol) in dry methanol (10 mL), and the solution refluxed for 6 h under N 2. The resulting blue solution was cooled to RT and filtered; slow 48 addition of diethylether (30 mL) to the filtrate yielded a blue solid, which was filtered off in air, washed with acetone and dried in vacuo at RT (0.54 g, 72% yield). Anal, calcd. for C 2 0 H 3 4 O 1 0 N 6 S 2 C l 2 R u : C 31.83, H 4.54, N 11.13; found: C 32.22, H 4.84, N 11.29. A m a x n m (log C), H 2 0: 330 (3.24), 532 (1.93). 2.4.18 RuCi2(tmsoWSR-2508). 18 The SR-2508 ligand (428 mg, 2 mmol) was added to RuCl2(tmso)4 (588 mg, 1 mmol) in dry methanol (10 mL), and the solution refluxed for 4 h under N 2 . The resulting blue solution was filtered hot and diethylether (30 mL) added slowly to the cooled filtrate to precipitate a blue solid; this was filtered off in air, washed with acetone and diethylether, and dried in vacuo at RT (0.44 g, 74% yield). Anal, calcd. for C 1 5 H 2 6 0 6 N 4 S 2 C l 2 R u : C 30.30, H 4.41, N 9.43; found: C 30.38, H 4.73, N 9.04. Am a xnm(log e), H 2 0: 341(3.39), 495 (1.82). 2.4.19 RuCl2(tmso)3(CMNI). 19 The compound CMNI (322 mg, 2 mmol) was mixed with RuCl2(tmso)4 (588 mg, 1 mmol) in dry methanol (10 mL), and the solution refluxed for 6 h under N 2 . The resulting green solution was cooled to RT, and diethylether (30 mL) was added slowly to precipitate out a green product. This was filtered off, washed with acetone and dried in vacuo at RT (70% yield). Anal, calcd. for C 1 6 H 2 8 ° 5 N 3 S 3 C 1 3 R u : C 2 9 ' 7 4 ' H 4 - 3 7 ' N 6 - 5 0 ; f o u n d : C 3 0 0 1 » H 4 - 4 2 » N 6 - 7 5 -Aj^nm (logC), H 2 0: 320 (3.71), 381 (2.71). 2.4.20 RuCl2(tmso)(RSU-3159). 20 The complex RuCl2(tmso)4 (588 mg, 1 mmol) was added to RSU-3159 (450 mg, 2 mmol) in methanol (15 mL), and the mixture refluxed for 6 h under N 2 . The resulting dark green solution was filtered hot, and diethylether (40 mL) 49 added slowly to the cooled filtrate to precipitate a green solid; this was filtered off in air, washed with acetone and diethylether, and dried in vacuo at RT (0.31g, 62% yield). Anal, calcd.for C{ J H J J O J N ^ C ^ R U : C 26.34, H 3.01, N 13.97; found;C 26.33, H 2.93, N 13.74. A m a x n m (log £ ) H 2 0: 330 (3.89). 2.4.21 RuBr 2(dmsoW4-NQ 2-Im) 2 . 21 Dry methanol (15 mL) was added to RuBr2(dmso)4 (573 mg, 1 mmol) and 4-N0 2-Im (282 mg, 2.5 mmol), and the solution refluxed for 4 h under N 2 . The resulting brown solution was cooled to RT and diethylether (40 mL) added slowly to precipitate out a brown product, which was filtered off in air, washed with acetone and diethylether, and dried in vacuo at RT (0.45 g, 70% yield). Anal, calcd. for C 1 0 H l g O 6 N 6 S 2 B r 2 R u : C 18.67, H 2.82, N 13.06; found:C 18.43, H 2.71, N 12.78. A m a x nm(log £ ), EtOH: 347 (4.17). 2.4.22 RuBr 2(dmsoWNMe-4-N0 2-im). 22 The complex RuBr2(dmso)4 (572 mg, 1 mmol) was suspended in a methanol solution (15 mL) of NMe-4-N0 2-Im (254 mg, 2 mmol), and the mixture refluxed for 4 h under N 2 . The resulting blue solution was cooled to RT, when a blue precipitate formed. This was collected, washed with diethylether (20 mL), and dried in vacuo at RT (0.38 g, 70% yield). Anal, calcd. for C 8 H 1 7 0 4 N 3 S 2 B r 2 R u : C 17.65, H 3.15, N 7.72; found: C 17.72, H 3.31, N 7.84. A m a x n m (log £ ) , H 20: 306 (4.28), 440 (3.01). 2.4.23 RuCWdmsoW 1.10-phenanthroline). 23 The complex RuCl2(dmso)4 (483 mg, 1 mmol) was added to 1,10-phenanthroline monohydrate (0.396 mg, 2 mmol) in methanol (15 mL), and the solution refluxed for 6 h under N 2 . The resulting brown solution was cooled to 50 RT, when a brownish yellow precipitate formed. This was filtered off in air, washed with acetone and diethylether, and dried in vacuo at RT (0.40 g, 78% yield). Anal, calcd. for C 1 6 H2 0 O2N2S 2 Ci2Ru: C 37.78, H 3.94, N 5.51; found: C 37.52, H 3.83, N 5.33. A m a x n m (log £ ) , H 2 0: 265 (3.98), 332 (2.42). 2.4.24 RuCl2(dmso)2(5-Cl-1.10-phenanthroline). 24 The 5-C1-1,10-phenanthroline (0.321 g, 1.5 mmol) was mixed with RuCl2(dmso)4 (483 mg, 1 mmol) in dry methanol (15 mL), and the mixture refluxed for 4 h under N 2 . The resulting brown solution was filtered hot, and a brown precipitate formed when the filtrate was cooled. This was filtered off in air, washed with acetone and diethylether, and dried in vacuo at RT (0.38 g, 70% yield). Anal, calcd. for C 1 6 H 1 9 02N 2 S 2 Cl2Ru: C 35.41, H 3.50, N 5.16; found: C 35.62, H 3.72, N 5.02. Am a xnm(log e), H 2 0: 267(3.84), 328 (2.57). 2.4.25 RuCl2(dmsoW5-NO2-1.10-Phenanthroline^. 25 The 5-N02-1,10-phenanthroline (0.338 mg, 1.5 mmol) was mixed with RuCl2(dmso)4 (483 mg, 1 mmol) in dry methanol (15 mL), and the solution refluxed for 6 h under N2. The resulting brown solution was filtered hot, and a brown precipitate formed when the filtrate was cooled. This was filtered off in air, washed with acetone and diethylether, and dried in vacuo at RT (0.42 g, 76% yield). Anal, calcd. for C 1 6 H 1 9 0 4 N 3 S 2 C l 2 R u : C 34.72, H 3.43, N 7.60; found: C 34.33, H 3.78, N 7.32. ^ m a x n m (log C ) , H 2 0: 267 (4.06) 330 (2.81). 2.4.26 RuCUdmsoWNMe-ImK. 26 The compound NMe-Im (0.5 mL, 6.0 mmole) was added to RuCl2(dmso)4 (483 mg, 1 mmol) in dry methanol (10 mL), and the solution refluxed for 4 h under N2. The resulting yellow solution was cooled to RT and 51 diethylether (30 mL) added slowly to precipitate out a yellow product, which was filtered off in air. This product was redissolved in CH2CI2 (10 mL), reprecipiated with diethylether (30 mL), filtered off in air, and dried in vacuo at RT (0.36 g, 73% yield). Anal, calcd. for C 1 2 H 2 4 0 2 N 4 S 2 C l 2 R u : C 29.24, H 4.87, N 11.37; found: C 29.37, H 4.62, N 10.98. A m a x nm(log C), H 2 0: 330 (3.67). 2.4.27 RuCl2(dmso)2(2Me-Im)2. 27 The compound 2Me-Im (0.164 g, 2 mmol) was added to RuCl2(dmso)4 (483 mg, 1 mmol), in dry methanol (15 mL), and the solution refluxed for 4 h under N 2 . The resulting yellow solution was cooled to RT and diethylether (40 mL) added slowly to precipitate out a yellow product, which was filtered off in air. This product was redissolved in CH 2 C1 2 (10 mL), reprecipitated with diethylether (30 mL), filtered off in air, and dried in vacuo at RT (0.38 g, 78% yield). Anal, calcd. for C , 2 H 2 4 0 2 N 4 S 2 C l 2 R u : C 29.24, H 4.87, N 11.37; found: C 29.52, H 4.63, N 11.58. k „ nm (logC), H 9 0: 333 (3.82). rn SIX ^ 2.4.28 r R u ( N H 3) 5CnCl 2. 28 The compound [Ru(NH3)^Cl]Cl 2 was prepared according to a literature /17 procedure. Some RuClyt^O (3 g, 11 mmol) was dissolved in H 2 0 (10 mL) in a 200 mL beaker, and hydrazine hydrate (85%, 20 mL) was then added to the well-stirred solution. The solution was stirred for about 12 h and filtered by gravity. To the filtered solution, concentrated HC1 was carefully added until the pH was 2 when vigorous gas evolution took place. The solution was boiled, with stirring and became yellow, slowly precipitating the yellow product. When no further precipitation was observed, the mixture was cooled to R T , and the crude product collected by filtration. The product was reprecipitated by heating an aqueous slurry (3 g in 10 mL H 2 0) to 60°C and concentrated N H 3 was added 52 dropwise until the yellow complex dissolved to give a wine-red solution, which was filtered hot and cooled in an ice-bath. To the cooled solution, concentrated HC1 was added dropwise to reprecipitate the yellow product which was filtered, washed with 6M HC1, water, alcohol and acetone (1.1 g, 39% yield). Anal, calcd. for H 1 5 N 5 C l 3 R u : H 5.19, N 23.98, Cl 36.35; found: H 5.30, N 24.73, Cl 36.12. 2.4.29 rRu(NH 3) 5(NMe-Im)lCl 3. 29 The compound [Ru(NH3)5(NMe-Im)]Cl3 was prepared according to a literature procedure. The compound NMe-Im (1.0 mL, 12.0 mmol) was added to 40 mL of 0.05 M HC1. Solid [Ru(NH 3) 5Cl]Cl 2 (145 mg, 0.50 mmol) was added and the resulting suspension was reduced over zinc amalgam for a period of 4 h, during which time the [Ru(NH3)^Cl]Cl2 dissolved completely. A stream of argon was bubbled through the mixture throughout the course of the reaction. At the end of the reaction period the solution was removed from the zinc, diluted with 60 mL of water and subjected to an air stream for 2 h. The pH of the resulting solution was adjusted to 1-2 with dilute HC1. Ion exchange was carried out using the H + form of AG50W-X4 resin supplied by Bio-Rad Laboratories. The column size was approximately 2.5 x 10 cm. The major product, eluted by 3M HC1, was isolated by roto-evaporation to dryness and reprecipitation from water-ethanol; the product was filtered off in air and dried in vacuo at RT (0.24 g, 65% yield). Anal, calcd. for C 4 H 2 1 N 7 C l 3 R u : C 12.82, H 5.65, N 26.17; found: C 12.94, H 5.78, N 25.87. A m a x nm(logE), H 2 0: 320 (3.42). 2.5 Cell Culture Procedures The in vitro biological experiments reported here were performed on a Chinese hamster ovary (CHO) cell line grown in tissue culture. The CHO cell line is used widely 5 3 because it grows rapidly, can be cloned with high plating efficiency, and can be grown in monolayer, suspension, and soft agar cultures. The CHO cells were grown in suspension culture in an alpha-medium, supplemented with 10% fetal calf serum (FCS), antibiotics, and bicarbonate buffer (see Appendix I). The flask spinning cultures were grown in an incubator (Incu-cover, Associated Biomedic Systems) at 37°C, with pH regulated to 7.4 by a continuous gassing of the cultures with commercially supplied 5% C 0 2 in air (Canadian Liquid Air Co. Ltd.). The cell culture was diluted daily to about 7xl0 4 cells/mL, maintaining asynchronous exponential growth with a doubling time of about 12 h. 2.6 Irradiation Irradiation was performed using an X-ray source (Picker, 280 kVp, HVL 1.7 mm Cu or Philips RT 250, 250 kV, 0.5 mm Cu). Samples were placed in special glass irradiation vessels, the "arms" of which were taped to the sides of the plexiglass container in which the vessel sat in to prevent it from tilting. Cold water was added to about 600 mL of crushed ice to bring the total volume to 820 mL, which was poured around the irradiation vessel to keep the temperature at 4°C. A stirrer motor was suspended above the vessel which contained a teflon coated magnetic bar, while the X -ray source was placed below the vessel (Figure 14). 2.7 Radiosensitization-Cell Survival Experiments Radiosensitization experimental procedures were as described by Moore et al.^ except that cells were exposed to drugs for 1 h at 37°C before irradiation at 4°C. Typically, sterile Petri dishes (60x15 mm, Falcon) were prepared the day before the radiosensitization experiment, and 5 mL alpha-medium supplemented with 10% FCS and 10^  "feeder cells" were added. Feeder cells are cells of the same line, sterilized by heavy irradiation, which supply nutrients and growth factors to increase the overall plating 54 Figure 14: I r r a d i a t i o n set—up The c e l l suspension was put into a s p e c i a l glass i r r a d i a t i o n vessel (with a magnetic s t i r bar) which was placed i n a p l e x i g l a s s container containing i c e water; a s t i r motor was suspended above the vessel. To obtain a dose rate of 1.6 Gy/min the p l e x i g l a s s container was placed d i r e c t l y on top of the 20 x 20 cm collimator. 55 efficiency of the cloning procedure, and to overcome the variation in plating efficiency which may be observed when widely differing inocula are used. These Petri dishes were placed in the incubator, and had reached pH and temperature equilibrium by the following day. An asynchronous population of CHO cells was grown in a spinner flask. Cells were then put in the special glass irradiation vessels in a growth medium, with or without the test compounds, in a total volume of 20 mL at approximately 1.25x10^  cells/mL. The cells were made hypoxic by flowing purified containing less than 5 ppm of 0 2 (Canadian Liquid Air Co. Ltd.), over stirred cell suspensions for 1 h at 37°C (the time being based on toxicity data, see Section 5.1) prior to the start of irradiation. Aerobic cells were treated identically except that the flow of N 2 was not employed. Aliquots of 1 mL were withdrawn by auto pipette from the irradiation vessel after the required doses and pipetted in plastic dilution tubes filled with 9 mL of chilled (4°C) alpha-medium (lacking NaHC0 3 and 10% FCS). Cells were plated from the dilution tubes into the Petri dishes previously prepared. The cell concentration in the dilution tubes was determined using a Coulter counter (Coulter Electronics Inc., Hialech, Florida). The cells were plated from the dilution tubes so that approximately 100 colonies per Petri dish would emerge after 7 days. After the plating procedure, the dishes were returned to the incubator for 7 days. The medium was then carefully poured out of the Petri dishes and 4 mL of methylene blue stain was added to each dish. After approximately 6 min, the stain was poured off and the Petri dishes rinsed gently in cold running water. The stained colonies were counted after the dishes had dried. A clone of 50 or more cells was assumed to represent a survivor. 56 Surviving fractions are defined as plating efficiency (PE) of treated cells divided by PE of controls. The PE is defined as: number of colonies PE = [2.1] number of cells plated In general, experiments were repeated at least three times. Values of log (surviving fraction) were averaged for each X-ray dose, and standard errors of the mean were calculated. Results are presented as log S vs. dose of X-ray, where number of colonies S = / PE of Controls [2.2] number of cells plated Data curves were drawn by using the linear-quadratic fit of the Oi, (3 model/' 7 2.8 Toxicity Studies - Cell Survival Experiments The toxic effect of a compound was measured by exposing cells to the test compounds for various time intervals, then washing the cells free of the compound with alpha-medium and plating aliquots of the cells into Petri dishes to determine colony Of. forming ability. Solutions containing the test compound in medium were prepared before each experiment in alpha-medium lacking bicarbonate buffer, and sterility was assured by filtering through a Nalgene 0.22 ^m filter unit. From this stock solution, 19,0 mL was added to a 120 mL wide-mouth, glass centrifuge vessel, this being sealed with a rubber stopper fitted with two syringe needles to permit gas flow, and a stoppered central port for removal of samples (Figure 15). In a typical experiment, such vessels, including controls and various complexes, were set up in a large water-bath, maintained at 37°C and placed on top of a multi-magnetic stirrer unit. The entire apparatus was kept in a 37°C warm room. The appropriate gas phase (N 2 or air), supplied from a cylinder, and humidified in a glass bubbler filled with sterile 57 •CELL SUSPENSION MAGNETIC STIR BAR Figure 15: The glass vessel f o r c e l l suspension t o x i c i t y t e s t s . Nitrogen (5% CO^ was flowed over a s t i r r e d 20 mL suspension of 2.0 x 10 cells/mL contained i n a glass vessel. Samples were obtained by removing the small stopper b r i e f l y and lowering a pipet down the glass tubing into the suspension. Experiments were performed at 37 c. 58 (autoclaved) water, was admitted through a short tube fitted with a plastic syringe tip. The solutions of test compounds in the medium were pre-gassed before addition of cells for 1 h to achieve temperature and gas phase equilibria. Cells were harvested from culture by centrifugation for 8 minutes at 600 rpm using a SORVALL RC-3 centrifuge, and resuspended in compound-free medium at 4x10^ cells/mL, and they remained at this concentration for about 15 min, to ensure metabolic depletion of any remaining 0 2 in the medium. At zero time, 1 mL of cell suspension was added to give a volume of 20 mL, a cell concentration of 2x10^ cells/mL, and a test compound concentration as desired. Aliquots (1 mL) were removed at zero time (immediately after adding cells) and at regular intervals thereafter, before being immediately diluted in 9 mL fresh medium at 4°C. The samples were spun to harvest the cells; the supernatant medium was decanted and replaced with fresh medium, and the cells were resuspended by vigorous pipetting and vortexing. Samples of this suspension were then plated into Petri dishes, previously prepared as described in section 2.7, using a micropipette. Typically, \0 jiL of cell suspension were plated into a Petri dish. If significant toxicity were anticipated, larger samples were plated (up to 2 mL); triplicate plates were plated at several inoculum volumes. Subsequent to the plating procedure, the cell suspensions were again vortexed and a sample (2 mL) was removed, diluted with PBS, and cell concentration determined with the Coulter counter. The Petri dishes were then incubated at 37°C in a tray incubator (National Inc.) with 5% C 0 2 (Canadian Liquid Air Co. Ltd.) flow for 7 days. The medium was then carefully decanted and replaced with methylene blue stain. Finally, the stain was washed off, and the colonies per Petri dish were determined. According to convention,5 a clone of 50 or more cells was assumed to represent a survivor. 59 2.9 Assay for Chromosome Aberrations The method employed was that previously described by Stich et alJ19 with some modifications, such as the cells being seeded 2-3 days before the experiment; the cells were exposed to the test compounds for 3 h. Typically, approximately 140,000 CHO 2 cells were seeded on 22 mm coverlips in 3.5 cm plastic dishes (Falcon) and kept in 2 mL M E M with 10% FCS, antibiotics (Streptomycin sulphate, 29.6 ^g/mL; penicillin, 125 ^g/mL; kanamycin, 100 ^g/mL, and fungizone, 2.5 ^g/mL) and sodium bicarbonate (1 mg/mL) (10% MEM) at 37°C for 2-3 days. When cells were 40%-60% confluent, the tissue culture medium was removed and replaced with 0.5 mL 10% M E M and 0.5 mL of test compounds at various concentrations. After 3 h incubation at 37°C, the test compounds and the 10% M E M were removed by suction and the cells washed twice with MEM. The cells were then incubated for 16 h at 37°C in 2.0 mL 10% MEM. For tests involving the inclusion of liver microsomal preparations (S9) (see Appendix VI), 0.5 mL aliquots of this activation mixture containing 150 /Ug S9 were added to each Petri dish instead of 0.5 mL 10% MEM. The S9 was prepared from livers of Aroclor 1254-pretreated Wistar male rats as described by Ames et al.120 The activity of the S9 mixture was verified by its capacity to activate the precarcinogen aflatoxin Bl to induce chromosome aberrations in CHO cells. Aflatoxin Bl without the S9 mixture did not elevate the frequency of metaphases with chromosome aberrations over that found for non-treated control cell cultures. For an estimation of the frequency of chromosome aberrations, 5 ^g/mL of colchicine was added to the test compounds at 16 h post-exposure and left for 3 h to arrest dividing cells at metaphase. Cells were then treated with 1% sodium citrate hypotonic solution for 20 min, this being followed immediately with fixation in ethanol/glacial acetic acid (3:1, V/V) for 20 min. Air-dried slides were stained with 2% orcein in glacial acetic acid/water (1:1 V/V), dehydrated and mounted. For each 60 sample, 100 metaphases were analyzed for chromatid exchanges and chromatid breaks in each of three replicate cultures. 2.10 Measurement of Non-protein Thiols Measurement of NPSH was carried out essentially as previously described by Tietze 7 2 7 but cells were pre-treated with test compounds before NPSH determination. Typically, a metaphosphoric acid solution (0.5 mL of 10%) was added to 1 mL of CHO cell suspension (3x10^ cells), which had been pre-treated with the test compounds (200 }iM final concentration) under hypoxia at 37°C for 1 h, in 0.2 M PBS. The cell mixture was sonicated for 10 s and allowed to cool for 10 min in ice, before being centrifuged at 4000 rpm for 10 min. Some PBS (1 mL of 0.5 M), which contained 1.7 g NaOH and 0.2 g EDTA/lOOmL, was added to 1 mL of supernatant, and 0.05 mL of Ellman's reagent was added immediately and the mixture left for 10 min to attain equilibrium at 25°C. Optical absorption of samples were then read at 412 nm on a UV/VIS spectrometer (Aminco DW-2). 2.11 Inhibition of Restriction Enzymes The plasmid, pSV2-gpt, was obtained as described by Skov et al.122 After growth in E. coli, the DNA (5200 base pairs) was extracted and linearized using PvuII. Figure 16 shows the scheme used: cleavage by BamHI or EcoRI at the defined sites results in only 2 fragments per enzyme because purified linear plasmid is used. The DNA (20 yL, 20 ,wg/mL) in Tris-Cl (100 mM)/EDTA (1 mM) at pH 8.0, was exposed to the test compound (80 ,wL of desired concentration) for 1 h at 37°C, after which unbound drug was removed using a spin-column of G-50 Sephadex (Pharmacia). Buffer (3.5 uL) (containing 0.33 M Tris (pH 7.9), 0.66 M potassium acetate, 0.10 M magnesium acetate, 0.005 M dithiothreitol and 1 mg/mL bovine serum albumin) and 2 pL of enzyme were added to the eluted DNA (30 yL). A solution of EDTA (1 jiL, 0.5 61 p S V 2 - g p t I Pvull •^XXJOOODOOOOOOOOC^ B a m H I / \ EcoRI 2.0kb _ 3.1kb X)Ooooa:cTAG xxx>ocoooooc<  OATC:>oooooocccc AATTDoooccc 3.2kb 2.1kb Figure 16: Outline of binding assay using plasmid DNA. Binding of complex at or near r e s t r i c t i o n s i t e i n h i b i t s r e s t r i c t i o n endonuclease a c t i v i t y on l i n e a r i z e d plasmid; kb=kilobases, number of bases i n the DNA strand (taken from ref. 120). 62 M as the disodium salt) was used to stop enzyme activity after 30 min at 37 C. A tracking dye (7 yL: 0.25% bromophenol blue, 0.25% xylene cyanol, in 30% glycerol water) was added, and the resulting solution (20 piL) was loaded onto a 1% agarose gel, made with E-buffer (0.04 M Tris., 0.005 M sodium acetate, 0.001 M EDTA as the disodium salt, pH 7.8 with glacial acetic acid), and subjected to horizontal electrophoresis for 16 h at lV/cm at RT. The slab gel was stained for 15 min with aqueous ethidium bromide solution (500 mL, 1 mg/mL) and photographed under UV light with an MP-4 Polaroid camera. Inhibition of the restriction enzymes BamHI and EcoRI was assessed qualitatively by noting the relative proportions of bands (cleaved vs. uncleaved). More quantitatively, the proportion of uncleaved DNA was determined from a densitometric (Kapp and Zonen DD2) scan of the negatives of the recorded pictures. Analysis was by integration of the scanned peaks, with the fraction in the uncut band over the total area of the three bands per lane being used to calculate the uncleaved (inhibited) fraction. 2.12 Measurement of Partition Coefficient The experimental procedure for determining the partition coefficients of the Ru complexes was as outlined by Fujita et al.124 The analyses of the concentrations of the partitioned subtances were made using a Perkin Elmer 552A UV/VIS spectrometer. The anhydrous 1-octanol was used as obtained without further purification. For the partitioning, octanol saturated with distilled water and distilled water saturated with octanol were used. The volume ratio of the two phases and the amount of sample were chosen so that, in most cases, the absorbance of a sample from the aqueous layer after partitioning had a value between 0.2 and 0.9 absorbance unit using a 1 cm cell. Only the concentration of the sample in the aqueous layer was determined, and that in the octanol was obtained by difference. The partition coefficient (P) was calculated as the 63 ratio of the concentration of complex in the octanol phase to that in the aqueous phase: Concentration i n octanol Partition Coefficient = [2.3] Concentration i n water For each complex, an average of three measurements were made, and care was taken to ensure that each reading recorded represented equilibrium conditions. Measurements were done with both aerobic and hypoxic conditions. 64 CHAPTER THREE CHARACTERIZATION OF COMPLEXES Complexes of formulation RuCl2(dmso)2L2 (L=4-N02-Im, RSU-1170, RSU-3083, RSU-3100, 2-N0 2-Im, miso, De-miso, metro, NMe-Im and 2Me-Im) are readily synthesized from the precursor complex c/s-RuCl2(dmso)4, of known structure, 9^ - 9 6 by substitution of two sulphoxide ligands. The ir and *H nmr data (see sections 3.2.1 and 3.3.1) are consistent with the diamagnetic products (5, 7-9, 11-14. 26 and 27) having a purely cis structure (i.e. cis, cis, cis) with both dmso ligands being S-bonded. The syntheses proceed via reaction [3.1]: 0 L [3.1] (O or S implies oxygen or sulphur-bonded dmso, respectively). Complexes 6 and 23-25 of formulation RuCl2(dmso)2L (L=NMe-4-N02Im, 1,10-phenanthroline, 5-C1-1,10-phenanthroline and 5-N02-1,10-phenanthroline) are again hexacoordinate. When L=NMe-4-N0 2-Im, the L binding is bidentate via an imidazole nitrogen and an oxygen of the N 0 2 group; the phenanthrolines bind in their common bidentate mode ; the two S-bonded dmso ligands are probably trans. Complex 10. RuCl2(dmso)2(RSU-3159), has S-bonded dmso ligands and a coordinated thioether of RSU-3159, this ligand probably being chelated via the nitrogen of the NMe group (see section 3.3.1). 65 Complexes of formulation RuCl 2(tmso) 2L 2 (L=4-N02-Im, NMe-4-N0 2-Im and De-miso) 15-12, RuCl2(tmso)2L (L=SR-2508) 18, RuCl2(tmso)3(CMNI) 19, and RuCl2(tmso)(RSU-3159) 20, are synthesized from the precursor complex RuCl2(tmso)4 of unknown structure, by substitution of two of the all S-bonded tmso ligands (see section 3.1). The ir data of 15-20 indicate that the tmso ligands are S-bonded but the nmr data are very complex (see section 3.3.2) and geometric arrangements of the ligands have not been determined. Complexes of formulation RuBr2(dmso)2(4-N02-Im)2, 21, and RuBr2(dmsp)2(NMe-4-N02-Im), 22, are synthesized from the precursor trans-RuBr^dmso)^ 2- 5 by substitution of two dmso ligands via reaction [3.2]: Br Ru-Br -S + 2L (L') -2. S L Br Ru-Br 21 L -S or S-Br .Ru Br 22 L 1 / [3.2] L=4-N02-Im; L'=NMe-4-N02-Im; S=S-bonded dmso The J H nmr data (see section 3.3.3) suggest a purely trans structure with both dmso ligands being S-bonded for complex 21_. Complex 22 is also hexacoordinate with L again binding via an imidazole nitrogen and oxygen of the N 0 2 group, similar to 6, thus having a trans, cis, cis structure. The attempted preparations of Ru(III)-nitroimidazole complexes from the compound [Ru(NH3)^Cl]Cl2,known to bind DNA and have some anti-tumour on oj BA activity, ' ' were unsuccessful. Only the previously reported complex, 66 [Ru(NH3)5(NMe-Im)]Cl3, 29, ' containing the more basic NMe-Im could be made (see section 2.4.29). Considerable efforts to grow suitable crystals of any of the nitroimidazole complexes proved frustratingly fruitless. 3.1 XPS Data XPS can be used for determining bonding configurations because the technique detects changes in the electronic environment of an atom, as evidenced by shifts in the BE of core electrons. XPS data for N, O, Cl and S atoms within complexes 5, 6, 10, 11, 16, 18, 22, m-RuCl2(dmso)4 and RuCl2(tmso)4, and for some of the free ligands, are summarized in Table II. The XPS N Is spectra of 5 and its free 4-N0 2-Im ligand are shown in Figure 17. Coordination of imidazole ligands via the tertiary nitrogen N(3) is well established for metal ions, including RuQl),117'126'127 and the XPS data are consistent with this. The XPS data show that there is a shift to higher BE for all the N atoms present in 5, 6, 15, 16 and 22 compared to the energies within the free ligands, indicating derealization of electrons within the systems. The double intensity peaks centred at 399.6 and 398.9 eV (Figure 17) must each result from the addition of two peaks (considered to be of equal shape) from N(l) and N(3); this allows for a simple deconvolution into the two component peaks as shown. Thus, the deconvolution method assumes the same intensity and shape for each particular type of atom; the error in these particular assignments could be as high as +0.3 eV. In the free ligands 4-NC»2-Im and NMe-4-NC>2-Im, the N(3), atoms adjacent to the electron withdrawing NC>2 group are considered to have a higher BE (399.70 and 399.80 eV, respectively) than those of N(l) (398.10 and 397.90 eV, respectively). In the complexes, the coordinated N(3) might be expected to have an even higher BE than those observed (for 5, 400.20 vs. 399.70 eV; 6, 400.60 vs. 399.80 eV, 15, 400.10 vs. 399.70 eV; 16, 400.40 vs. 399.80 6 7 Table II XPS data for Ru(II) compounds and free ligands 3 Atom Compound Group BE Width0 5, Ru(4-N02-Im)2 -N02 405.80(405.35) 2.6(2.6) N(3) 400.20(399.70)° 2.4(2.6) N(l) 399.00(398.10)d 2.4(2.5) 6, Ru(NMe-4-N02-Im) -ONO 405.60(404.90) 2.5(2.3) N(3) 400.60(399.80)d 2.3(2.4) N(l) 399.30(397.90) 2.3(2.4) H> Ru1 (4-N02-Im)2 -N02 405.50(405.35) 2.5(2.6) N(3) 400.10(399.70) 2.4(2.5) N(l) 398.70(398.10) 2.4(2.5) 16, Ru'(NMe-4-N02-Im)2 -N02 405.20(404.90) 2.6(2.4) N(3) 400.40(399.80) 2.5(2.4) N(l) 398.90(397.90) 2.5(2.4) 18, Ru'(SR-2508) -N02 404.60(404.40) 2.4(2.3) 22, Ru;MNMe-4-N02-Im) -ONO 405.30(404.90) 2.5(2.4) N(3) 400.60(399.80) 2.4(2.3) N(l) 399.00(397.90) 2.5(2.3) 1, cis -RuCl 2 (dmso) ^ -S(0)Me2 532.00d 3.7 -OSMe2 533.70a 3.7 2, RuCl2(tmso)^ -S(O) 530.90 3.6 5, Ra(4-N02-Im) N02 531.70(531.70) 2.9(2.9) -S(0)Me2 532.00 3.6 6, Ru(NMe^N02-Im) -ONO 531.70d(530.70) 3.8(4.0) 530.80d 3.8 -S(0)Me2 532.00 3.8 10, Ru(RSU-3159) N02 530.10(530.10) 3.6(3.6) -S(0)Me9 532.01 3.8 continued continued 68 Atom Compound*3 Group BE° Peak Width0 0 (Is) 15, Ru '(4-N0 2-Im) 2 -N0 2 -S(0) 531.90(531.70) 531.20 3 .6(3 .3) 3.8 16, Ru' (NMe^r-N02-Im)2 -N0 2 531.90(531.70) 3 .6(3 .8) -S(0) 531.20 3.8 18, Ru'(SR-2508) -N0 2 532.00(531.90) 3 .6(3 .5) -C=0 531.50(531.45) 3.6 -S(0) 531.20 3.8 -OH 530.40(530.40) 3.4 19, Ru'(tmso)(CMNI) -N0 2 -S(0) 532.00(531.95) 531.20 3 .5(3 .4) 22, Ru"(NMe-4-N02-Im) -ONO -S(0)Me2 531.80(530.70) 530.90 531.70 ( 3 . 6 M 3 . 5 Cl 1, cis -RuCl2(dmso)^ Cl 197.50 3.0 ( 2 P V 2 , 3 / 2 ) 5, 6 , and 10 Cl 197.50-197.60 3.1 s ( 2 p l / 2 , 3 / 2 ) 1, ais -RuCl 2 (dmso)^ 2 , RuCl 2(tmso) 4 SO -S(0) 1 6 6 . 3 e 165.10 2.8 2.7 5 and 6 -S(0)Me2 166.40-166.50 2.9 10, Ru(RSU-3159) -S(0)Me2 166.30 2.7 -S 164.20(163.00) 2 .7 (2 .7 ) 15, Ru '(4-N0 2-Im) 2 -S(0) 165.10 2.9 16, Ru'(NMe-4-N02-Im)2 -S(0) 165.20 2.9 18, Ru'(SR-2508) -S(0) 164.70 2.8 20, Ru'(RSU-3159) -S(0) 165.10 2.8 -S 163.90(163.00) 2 .8 (2 .7 ) 22, Ru"(4-N02-Im)2 -S(0) 165.80 2.9 continued continued 69 Binding energy and peak width given i n eV; estimated maximum error +0.15 eV. Ru, Ru' and Ru" represent RuCl^dmso^, RuC^Ctmso)^ and RuB^Cdmso^ respectively, except 20 has only one tmso ligand, a l l the sulphoxide ligands being S-bondedT Number i n parentheses gives data for corresponding group i n the free ligand; the ranges of BE values for atoms are given i n ref. 128. The deconvolution method assumes the same intensity and shape for each particular type of atom; the error i n these particular assignments could be as high as +0.3 eV. The XPS peaks for sulphurs of the one 0-bonded and three S-bonded ligands are not resolved. 70 71 eV); perhaps a relatively low value results because d , low spin Ru(II) has a quite 117 remarkable donor ability toward unsaturated N-atom donors. The greater shift to higher BE of N(3) of 6 (0.8 eV) compared to those in 5 (0.5 eV), 15 (0.4 eV) and 16 (0.60 eV) is consistent with chelation, with further loss of electron density via an oxygen of N0 2 - Stronger evidence for coordinated nitrito is provided by the doublet seen for the O Is of the N 0 2 group of £ at 531.70 and 530.80 eV; similarly this O Is doublet is also observed for 20. This O Is doublet is not seen for 5 and 16 where the corresponding O Is BE is similar to, or the same as, that of the free ligand at 531.70 eV. The N Is BE for the N 0 2 of 18 is similar to that of the free ligand, SR-2508, suggesting no coordination via the N 0 2 group and a five- coordinate geometry The 531.70 eV and 530.80 eV BE's for O in 6 are assigned to the coordinated and uncoordinated oxygens, respectively. Chelating ligands utilizing a nitrito moiety are 729 uncommon, and we are unware of other examples involving nitroimidazole ligands. On comparison of the N Is and O Is BE's of 4-N0 2-Im and NMe-4-N0 2-Im, the lower values for the latter reflect the electon-donating properties of the methyl group which likely promote the chelation.7-*0 The BE's of CI 2p 1 / / 2 3 / / 2 for complexes 5_, 6 and 10, are 197.5-197.6 eV, which are essentially identical to that of c/.s-RuCI2(dmso)4. Therefore, the possibility of 6 being a chloride-bridged dimer with monodentate imidazole ligands is ruled out. The complex 1, c's-RuCl2(dmso)4 shows two O Is energies (532.0 and 533.7 eV) in an intensity ratio of about 3:1, entirely consistent with structural data for the complex/"* (see eq. [4.1]). However, complex 2, RuCl2(tmso)4, shows only one O Is (530.90 eV), the value indicating only S-bonded tmso. The corresponding O Is and S 2 p l / 2 3/2 d a t a a v a n a ° l e f ° r 5, 6, 10, 15, 16, J_8 and 22, on comparison with those for cis-RuCl2(dmso)4 and RuCl2(tmso)4, reveal the presence of only S-bonded dmso ligands in the nitroimidazole complexes. Previous work using XPS to distinguish O-and S-bonded dmso ligands has generally been less definitive. 72 The BE for the S 2pj / 2 3 / 2 l e v e l o t" t h e thioether sulphur atom of RSU-3159 (163.0 eV) is increased by 1.2 eV and 0.9 eV within complexes 10 (164.2 eV) and 20 (163.9 eV) respectively, strongly implying coordination via this sulphur. Models suggest that chelation of RSU-3159 via the lone pair of N(l), the methylated nitrogen, is plausible. However, the N Is data for IQ and 20 are insufficiently resolved to give information about binding to a nitrogen; *H nmr data (see section 3.3.1) provide some support for the chelation. Further, data for the S 2 p t / 2 3 / 2 L E V E L O F T N E T M S 0 complexes 2, 1_5_, 16 and J l reveal that the BE for 18 (164.70 eV) is much less than those of the other complexes. This indicates a weaker interaction between Ru and the S in 18, and may be consistent with a five- coordinate complex which is trigonal bipyramidal with the tmso ligands in the axial positions, giving a longer bond and weaker interaction with the Ru centre. 3.2 Infrared Spectral Data Infrared spectroscopy is one of the tools available to help elucidate the nature of the bonding of sulphoxide ligand. The v S Q for free dmso is seen at 1050 cm'*,95'131 while coordination through oxygen atom decreases this by enhancing the contribution of resonance structure(I) (Figure 10), because of the withdrawal of electrons from oxygen. This results in a lowering of the SO stretching frequency by approximately 100 cm"1; coordination through the sulphur atom increases the SO stretching frequency by about 70 cm"1, by enhancing the contribution of resonance structure(III) (Figure 10).^.^3 3.2.1 Infrared Data for RuCl 2(dmso) 2Ln Complexes (n=l or 2) The ir spectrum of the yellow compound 1 c/s-RuCl2(dmso)4, in Nujol is virtually identical to that reported previously 9 5 , 9 6' 7 7'* and, based on the band assignments, can be used to assign some bands for other complexes 73 described below. Bands at 1120 and 1091 cm show the presence of S-bonded dmso ligands, while a band at 920 cm"' indicates O-coordinated dmso. Selected ir spectral data for the 4-nitroimidazoles complexes, RuCl2(dmso)2Ln (n=l or 2), are summarized in Table III. The ir data for all the complexes £-J0 reveal bands in both the 1085-1091 and 1112-1167 c m - 1 regions, consistent with S-bonded dmso ligands. Oxygen-bonded sulphoxide is known to be labile relative to an S-bonded one,*7 and this has been demonstrated for cis-RuCtydmso)^.95,96'113 Thus, substitution of O-bonded sulphoxide by the nitroimidazole ligand would occur first and leave only S-bonded complex, as revealed by ir data. All the free 4-nitroimidazole ligands reveal ir bands in the region at 1557-1619 and 1518-1592 cm"1, except RSU-3159, which has ir bands at 1641 and 1673 cm"', the two values for each ligand are being attributable to the asymmetric (higher value) and symmetric N O 2 stretches. ^  Values within complex 6, 1543 and 1523 cm"1, are both some 20 cm"1 lower than those of the free ligand NMe-4-N02"Im (1565 and 1545 cm"1), presumably reflecting the coordination of the coordinated nitrito group. Changes in the on coordination of the other nitroimidazoles are smaller. The 2815 cm"1 band of free NMe-4 -N0 2 - Im, attributed to v ^ N C H _ H ^, is found at 2800 cm"1 with 6, while corresponding bands at 2724, 2730 and 2724 cm"1 for complexes 8-.10, respectively, are 6-18 cm"1 lower than that found in the free ligands. These data provide no strong support for chelation of RSU-3159 via N(l) within complex 10, although the coordination shift is greatest (11 cm"1) for this system. 74 Table III Selected i r spectral data for 4-nitroirnidazole complexes of Ru and the free nitroimidazole ligands Cc>mpoundsc v, SO _1 IR bands (Nujol, cm ) yN0„ b "N-Cr^-H yRu-Cl _ l s e£s-RuCl2(drnso)^  5, Ru(4-N02-Im)2 6, Ra(NMe^-N02-Im) 7, Ru(RSU-1170)2 8, Ru(RSU-3083)2 9, Ru(RSU-3100)2 10, Ru(RSU-3159) 1120,1091 920 1127 1090 1112 1085 1123 1087 1167 1086 1137 1091 1125 1085 1559(1560)asym. 1535(1550)sym. 1543(1565) 1523(1545) 1560(1557) 1522(1518) 1627(1619) 1599(1592) 1614(1611) 1538(1577) 1684(1673) 1653(1641) 2800(2815) 2724(2730) 2730(2738) 2724(2742) 328 342 332 354 325 349 304 325 330 347 326 351 Ru represents RuCl2(dmso)2; a l l sulphoxides ligands are-S-bonded. Values i n parentheses are for the free nitroimidazole ligands. These bands are tentatively assigned because of uncertainty i n the complex f a r - i r region. 75 The i r bands for the 2-ni troimidazoles and metronidazole, and the corresponding complexes are summarized in Table IV. The i r data for complexes 11-14 reveal v S Q bands in both the 1078-1094 and 1100-1162 c m - 1 regions, again suggesting S-bonded dmso ligands. A l l the complexes have ir bands at about 1500-1548 and 1520-1560 c m " 1 attributable to symmetr ic and asymmetr ic N 0 2 , respectively, the values being sl ightly lower (by up to 12 c m " than those found in the free ligands. The i r data for the 1,10-phenanthrolines and mefhyl imidazoles, and complexes (23.-27) (Table IV ) , reveal bands in both the 1076-1098 and 1095-1134 c m " 1 regions, again consistent wi th S-bonded dmso l igands. A 2791 c m " 1 band, attr ibuted to ^ N C H J - H ) * s ^ o u n d w i th in 26, wh ich is again lower than the value for the free l igand N M e - I m (2804 c m " 1 ) . The v R u _ c i b a n u S ° f a n < the complexes are d i f f i cu l t to assign because of uncertainty in the mu l t i -band fa r - i r region. Even for 1, only one band at 344 c m " 1 has been def in i te ly a s s i g n e d , 9 5 ' 9 6 and indeed the presence of only one intense R u - C l band in this region indicated that the product was the trans-i s o m e r . 7 ^ 7 However , this was shown later by structural data to be incorrect, and the complex is i n fact a c i s - i somer . 9 ^ Nevertheless, the structures are generally thought to contain cr's-chlorides (see section 3.3.1) and the two bands l isted for V ^ U _ Q in Tables III and IV seem appropriate for such a geometry. 3.2.2 Infrared Data for R u C l 2 ( t m s o ) m L n Complexes (m=l . 2 or 3: n=l or 2) Selected i r spectral data for the tmso complexes are summarized in Table V . The ir bands at 1133 and 1094 c m " 1 for complex 2 show the presence 76 Table IV Selected i r spectral data for 2-nitroimidazoles, metronidazole and 1,10-phenanthrolines, and their Ru complexes /~1 1 3- IR bands (Nujol, cm ) Compounds yS0 *N0 2 b yCH2"H yRu-Cl0 11, Ru(miso)2 1089 1100 1520(1528)asym. 1500(1512)sym. 335 350 12, Ru(De-miso)2 1078 1105 1532(1538) 1515(1527) 328 348 13, Ru(2-N02-Im)2 1084 1112 1560(1567) 1548(1559) 336 347 14, Ru(metro)2 1094 1162 1548(1552) 1540(1546) 331 352 23, Ru(phen) 1098 1134 324 338 24, Ru(5-Cl-phen) 1079 1095 329 342 25, Ru(5-N02-phen) 1093 1107 1536(1533) 1515(1518) 326 340 26, Ru(NMe-Im)2 1076 1114 2791(2804) 330 342 27, Ru(2Me-Im)2 1078 1099 2799(2811) 338 351 Ru represents RuC^dmso^; the sulphoxides ligands are S-bonded. Values i n parentheses are for the free ligands. These bands are tentatively assigned because of uncertainty i n the f a r - i r region. 77 Table V Selected i r spectral data for RuCl9(tmso) L and RuBr9(dmso)9L complexes; m=l,2 or 3, n=l or 2 1 m n 1 1 1 IR bands (Nujol, cm ) Compounds3 r U S 0 D ^N0 2 C yN-CH2-H yRu-Cl' 2, RuCl 2(tmso) 4 1094 1133 339 4, trans -RuBr2( dmso )^ 1080 15, Ru1(4-N02-Im)2 1075 1102 1557(1560)asym. 1537(1550)sym. 348 16, Ru1 (NMe^r-N02-Im)2 1088 1115 1554(1565) 1538(1545) 2809(2815) 343 17, Ru'(De-miso)2 1078 1114 1534(1538) 1515(1527) 337 18, Ru'(SR-2508) 1050 1093 1532(1538) 1508(1517) 342 19, Ru'(CMNI) 1057 1094 1593(1587) 1574(1568) 336 20, Ru'(RSU-3159) 1079 1106 1681(1673) 1652(1641) 339 21, Ru"(4-N02-Im)2 1084 1552(1560) 1539(1550) 22, Ru"(NMe^-N02-Im) 1073 1092 1542(1565) 1521(1545) 2806(2815) Ru' represents RuCl 2(tmso) m, m = 1(20), m = 2(15-18) and m = 3(19); Ru" represents RuBr2(dmso)2. k AH bands represent S-bonded sulphoxides. Values i n parentheses are for the free nitroimidazole ligands. ^ These bands are tentatively assigned because of uncertainty i n the f a r - l r region. 78 of S-bonded tmso ligands, while bands at 1075, 1035, 952 and 875 c m - 1 are probably due to v(T-ingy132'13^ particularly as the XPS data suggest only S-bonded tmso. The V ^ Q for free tmso is seen at 950 cm"'.''^ The ir spectra of complexes 15-20 reveal sulphur-bonded tmso by characteristic V^Q bands in the 1050-1115 c m - 1 region. The V N Q values assigned to complex 15 are similar to those observed for 5j the changes in V^Q^ within 1£ (a six-coordinate compound) on coordination of the ligand are smaller than the corresponding differences for the chelated complex 6 and its free ligand. The V N Q ^ bands for complexes 17 and !§. are in the regions 1508-1515 and 1532-1534 c m - 1 , whereas those v N 0 ^ bands for 19 are at about 1574-1595 c m - 1 , but all are similar to those of the free ligands. The small change in V J ^ Q ^ between complexes 1_8_ and 12, and their corresponding free SR-2508 and CMNI ligands, respectively, suggests that there is no coordination via the nitrito group (see Section 3.1). Bands due to V R U _ Q in the far-ir region are again difficult to assign and the geometric configuration of complexes 11-20 (cis or trans chlorides) cannot be determined from the ir spectrum. 3.2.3 Infrared Data of RuB^dmsoUI^ Complexes (n = 1 or 2) The ir spectrum of the orange product, 4, in Nujol is essentially identical to that reported previously (prepared from a different method) with a /ra/is-configuration.725 A strong ir band at 1080 c m - 1 can be assigned as V ^ Q (S-bonded) as previously reported. No band due to O-dmso is observed, which agrees with data reported previously. 9 5' 7 2 5 The ir spectra of complexes 21 and 22 possess v S Q bands in the 1070-1090 cm"1 region, again consistent with S-bonded sulphoxide ligands. The v N Q ^ values assigned in 2J. are similar to those observed for 5; replacement of CI by 79 Br within 5 thus affects V N Q ^ very little. Values within 22, 1542 and 1521 cm"1 are both approximately 20 cm"1 lower than the values for the free ligand; a similar trend is seen for 6. The data are consistent with coordination to Ru via the nitrito group and the imidazole ring N occurring in 22 as in 6. The Ru-Br stretch in the far-ir spectrum is difficult to assign, thus making it impossible to determine unambigously the structure of these complexes on the basis of the IR evidence alone. 3.2.4 Infrared Data for "RuCUmpsoW. 3 Compound 3 is very slightly soluble in acetone, CH2Cl2 and ethanol. Elemental analysis confirmed the product to be of the stoichiometry shown as demonstrated previously.77-* The ir spectrum contains a strong band centred at 1128 cm"1, assignable to V S Q (S-bonded); the free sulphoxide v S Q is at 1045 cm"1. Bands are also seen at 940 and 960 cm"1 but a band due to V < J Q (O-bonded) cannot be assigned because the spectrum in the 920-990 cm"1 region is complicated by the possible presence of the rocking modes of C H 3 7 - * 7 which in the free ligand are seen at 987, 972 and 957 cm'1.132 The low solubility of 2 (presumably resulting from its polymeric nature77-*) prevents any chemical reaction with the nitroimidazole ligands even in a heterogenous phase and, therefore, the chemical properties of this compound were not explored any further. Nuclear Magnetic Resonance Spectral Data Nuclear magnetic resonance is one of the most important spectroscopic methods 80 available for structure determination of an unknown molecule. While infrared spectroscopy reveals information as to the types of functional groups present in a molecule, nmr gives information about the number of each type of nuclei in such groups. It also gives information regarding the nature of the immediate environment of each of these types of nuclei. In * H nmr spectroscopy, coordination of sulphoxides to metal centres causes the adjacent proton nuclei to resonate at lower field positions. Methyl protons of dmso, other methyl alkyl, and methyl aryl sulphoxides, are deshielded by approximately 1 ppm on coordination to Ru(II) through sulphur and approximately 0.1 ppm on coordination through oxygen. 5 7 , 9 6 The ^ Q ' H } nmr spectral data for dmso and imidazole ligands show that the methyl groups of the sulphoxide are detected at 35-45 ppm, whereas the C resonances of the imidazole ring are at 1 1 0 - 1 2 0 ppm/ -** 3.3.1 Nuclear Magnetic Resonance Data of RuCWdmso^Ln Complexes (n  = 1 or 2 ) The proton nmr spectrum of compound i in CDCI3 was identical to that reported and discussed previously.9"^'96"^'^6 Four singlets due to the methyl protons of the S-bonded dmso ligands occur at 53 . 3 3 , 3.44, 3.50 and 3.53 ppm. The equivalent methyl protons of O-coordinated and and small amounts of free dmso are observed at 62 . 7 3 and 2.60 ppm, respectively (Figure 18). The integration ratio of the downfield singlets to the upfield pair indicates that 1 contains three S-bonded dmso ligands and one O-bonded dmso ligand (eq. [3.1]). The presence of the four singlets for the S-bonded ligands is readily rationalized, because the S-bonded ligands are trans to O-bonded dmso, S-bonded dmso or Cl, and this leads to three different chemical shifts for the methyl resonances of these molecules. Extra methyl resonances result from the 81 i t i i j i i I ~ I I i i i • i -p i i ; i | i i i i j i i i i | i i i i | i 5 4 3 2 T J I I I I I I I I I | I I I I j 1 PPM o Figure 18: H nmr spectrum of cis-RuCl (dmso) , 1, i n CDC1 . 82 formation of isomers from dissolution of the complex, , J the presence of some degree of methyl inequivalence could also lead to the complex spectrum observed. 9 6 - 7 ^ 7 * 5 Table VI lists selected ' H nmr spectral data for the RuCl 2(dmso) 2L n (n = 1 or 2) complexes. Complexes 5-10 show only S-bonded dmso, because no methyl resonance is observed around 5 2.7 ppm (the O-bonded region); the findings agree with the ir data. There is no dissociation of dmso ligand in CDCI3 solution (Figure 19a), and even in D 2 0 , complex 5 generates no free dmso . Oxygen-bonded dmso is known to be labile, thus, as seen in equation [3.1], substitution of O-bonded dmso and a S-bonded dmso cis to it (necessarily so for the chelating ligand NMe-4-N0 2-Im) would lead to the c/5-chloride cis-sulphoxide geometry for the product. For complexes 5, 7-2, the presence of four major, equal intensity resonances, occurring as partly resolved multiplets centred at the shifts listed in Table VI, reveals inequivalent methyl groups for each of the two inequivalent dmso ligands. Only the low symmetry all cis-structure of 6-coordinate complexes can give rise to such an nmr pattern.7-^'7-*6 A model of 5 shows that free rotation about the Ru-S bonds for cis-dmso ligands is restricted because of steric interaction between methyl groups, but this is not a requirement for the magnetic inequivalence of the methyls.7"*6 Proton decoupling experiments did simplify the observed multiplet patterns of the methyls, but these provided no further information on the structures. For the chelated complex 6_, there are two methyl resonances again of equal intensity (Figure 19b). These are consistent with trans dmso ligands (and thus ds-chlorides).7-*-5,7'*6 Therefore, rearrangement of the dmso ligands has occurred in this reaction, different from the pathway shown in eq. 3.1. The 83 Table VT Selected H nmr spectral data for RuCl9(dmso)9L complexes and the free nitroimidazole l igands 3 n Compounds^ H-2C H-4c(s) H-5C(s) N-CH3(s) or 2-CH 3 CH3(m) (dmso) 1, RuCl 2(dmso) 4 2.60 2.73 3.33 3.44 3.50 3.53 5, Ru(4-N02-Im)2 8.35(8.25)d'e 7.87(7.80)d'e 3.37 3.38 3.40 3.47 6, Ru(NMe-4-N02-Im) 7.96(7.54)d 8.78(7.87)d 3.93(3.90) 3.48 3.5^ 7, Ru(RSU-1170)2 g 7.91(7.90) 3.20 3.25 3.33 3.35 8, Ru(RSU-3083)2h 3.60(3.60) 3.37 3.42 3.46 3.52 9, Ru(RSU-3100)2 7.18(7.16) 3.62(3.60) 3.20 3.30 3.39 3.50 JOjRutRsu-sisg)1 7.64(7.92) 3.62(3.49) 3.18 3.20 3.25 3.28 H,Ru(miso) 2^ 7.20(7.19) 7.15(7.11) 3.39 3.48 3.50 3.52 12,Ru(De^Tiiso) 2 k 7.53(7.45) 7.24(7.19) 3.48 3.53 3.61 3.62 T3,Ri(2-N02-Im) 2 8.39(8.37) 8.03(7.99) 3.41 3.46 3.50 3.53 14,Ru (metro ) 2 8.05(8.00) 2.55(2.45) 3.69 3.77 3.85 3.87 23,Ru(phen) 3.45 3.4^ %,Ru(5-Cl-phen) 3.48 3.5 / 25,Ru(5-N02-phen) 3.52 3.55f 26,Ru(NMe-Im)2 7.49(7.47)d 7.13(7.08)d 6.91(6.88)d 3.72(3.70) 3.25 3.31 3.42 3.48 27,Ru(2Me-Im)2 7.11(7.04)d 7.05(7.O4)d 3.76(3.73) 3.25 3.33 3.47 3.51 continued .. continued 84 a 2 Oin ppm wrt TMS, i n CDCl^ at room temperature. In each case the integrations are consistent with the assignment. k Ru represents RuCl^Cdmso^; both sulphoxides ligands are S-bonded. Shift values i n parentheses are for the free nitroimidazole ligands. d Values are i n good agreement with literature data (ref. 137). The H-2 and H-5 singlets each becomes a doublet when dissolved i n CDC1„ with 57o of trifluoroacetic acid. ^ The CH^ resonances are singlets. ^ Shift values for ^ N-CH2 a n c^ HC-CIL multiplets are centred at 54.00 ppm and are the same for tEe free RSU-I170 ligand. n Shift values for HN-CH^t), HO-CH^t) and 2-CH3(s) are 53.89, 3.59 and 2.35 ppm,respectively,and are the same for the free RSU-3083 ligand. 1 Shift values for 4' and 5' protons are 7.05 and 7.02 ppm, respectively. j Shift values for the ^ N-O^-CH(OH) (m), -CH_2-OCH3(m), (HO)CH and 0-CrUs) are at 54.66, 4.42, 4.15 and 3.41 ppm,respectively, and are the same for the free miso ligand. k Shift values for the ^ N-O -^CmOH) (m), (HO)CH(m) and (H0)CH2(m) are at 54.39, 4.10 and 3.66 ppm, respectively, and are the same for the free De-miso ligand. 85 a r-^ ' • 1 >—» •- r - -» ^ -^- | 4 2 P P m Figure 19: "Si nmr spectra of (a): RuCl 2(dmso) 2(4-N0 2~Im) 2, 5_, i n CDCl.,; and (b) RuCl_(dmso)_(NMe-4-N0 -Im), 6, i n CDCl . 86 presence of four methyl resonances within 10 is again consistent with a chelated 6-coordinate complex with cis arrangements for both chloride and dmso ligands (Table VI). On comparison with data for the free 4-nitroimidazole ligands (Table VI), the H-2 and H-5 resonances in the complexes are generally shifted slightly to lower field, although the effects are more marked for 6, presumably resulting from the chelation as discussed in section 3.1. Similarly, the H-4 and H-5 resonances in the complexes 26 and 27 are also shifted slightly downfield on comparison with the free methylimidazole (Table VI), as have been observed in complexes of, for example, Ru(II) 7 7 7 and Pt(II).7 i S The larger downfield shift for the NMe protons of RSU-3159 on coordination, and a surprising upfield shift of 0.28 ppm for H-2, provide evidence for interaction of the Ru(II) centre with the nitroimidazole ring, presumably via the nitrogen of the NMe. Thus, together with the XPS data (see section 3.1), the nmr data suggest formation of a five-membered chelate ring through the sulphur and nitrogen. Of the 1,10-phenanthroline complexes, 23-25 show only S-bonded dmso (Table VI). As discussed above, substitution of O-bonded dmso and a S-bonded dmso cis to it, is expected for the chelating 1,10-phenanthrolines. There are two methyl resonances again of equal intensity similar to those of complex 6, suggesting rearrangement of the dmso ligands to trans positions, thus giving the /ra/ts-sulphoxides, c/s-chlorides, cis- 1,10-phenanthroline geometric configuration (as for 6 ). The 1,10-phenanthroline complexes show five sets of multiplets in the downfield aromatic region between 57.8-9.2 ppm, and the 5-C1 and 5-NO2 substituents do not markedly affect the position of these resonances, although the total integration on the protons is decreased by one atom. 87 The C{ H) nmr spectrum of 5 shows four peaks for the inequivalent methyls (535.40, 36.63, 44.77, 45.13 ppm) and imidazole carbons at 112.69 (C-2), 116.48 (C-5) and 120.49 (C-4) (Figure 20a). For 6, consistent with the H nmr, two J C shifts were seen at 38.13 and 45.02 ppm for the methyls of the two inequivalent dmso ligands; other assigned peaks were 112.41 (C-2), 116.18 (C-5), 119.94 (C-4) and 122.24 (NMe) ppm (Figure 20b). The *H nmr data for complexes 11-13 (Table VI) again show only S-bonded dmso in CDCl^ solution. Dissociation of the substituted-2-nitroimidazole ligands in non-dried CDCI3 or D-,0 solution, as evidenced by UV/VIS spectroscopy and aqueous solution chemistry (see section 4.1), results in the formation of even more complex multiplets for the methyl resonances but no free dmso methyl resonance is observed. However, when spectra of these two complexes were recorded in dried CDCI3, relevant *H nmr data could be obtained. Complexes 11-13 have the H-4 and H-5 resonances, and complex 14 has the CH^-2 and H-4 resonances shifted slightly downfield, as compared with the free ligand values, similar to findings for the other complexes 5, 7-10. The ' H nmr spectra of complexes 26 and 27 reveal only S-bonded dmso with four partly resolved multiplets, again indicating inequivalent methyl groups of two inequivalent dmso ligands with the complexes having an all c i s -geometric configuration. 3.3.2 - H nmr Data of RuCl2(tmso)mLn Complexes (m=l. 2 or 3: n = 1 or The ' H nmr spectrum of free tmso in CDCI3 reveals three sets of multiplets centred at 52.60 (S-C-CH 2 -CH 2 ) , 2.19 (S-CH 2) and 1.75 (S-CH 2) 88 a Ji Solvent Peaks l • ' i . | 1 I T T'l ' ' ' ' | ' ' ' i i i i i i | i i i i | i i i i | i i i i i i i i i | i i i \ | i i i i [ i i i i i '20 100 80 60 40 20 PPM Solvent Peaks 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 1 1 1 1 120 100 80 60 40 20 PPM Figure 20: 1 3 _ i l C J H } nmr spectra i n CDC13 of (a):' RuCl 2 (dmso) 2 (4-N02~Im) 2 , e-4-NO -] are shown i n the 576-78 ppm region. 5_, and (b) : RuCl 2 (dmso) (NMe-4-N02"Im) , 6_. The solvent peaks 89 ppm with an integration ratio of 2:1:1, respectively, in agreement with the literature.7-*-* The two S - C H 2 resonances are shifted downfield, while the S-C-CH.2-CH2 resonances become shifted to higher field by approximately 0.30 ppm, when the tmso coordinates to Ru(II) (Figure 21, Table VII). Three sets of multiplets centred at 5 4.00 (S-CH 2), 3.42 (S-CH2) and 2.30 (S-C-CH 2 -CH 2 ) ppm are observed, having a relative intensity of 1:1:2, respectively. The large downfield shift of these S-CH2 multiplets strongly suggests the presence of only S-bonded sulphoxide (Figure 2 l ) . 8 8 ' 1 3 2 - 1 3 3 Indeed, the XPS data for RuCl2(tmso)4 and all the other tmso-containing complexes indicate the presence of only S-bonded sulphoxides. The ' H nmr spectra for complexes J_5 and 16 in CDCI3 (Table VII) show five sets of multiplets for the CH2 resonances of tmso (see below), centred at 54.00, 3.65, 3.15, 2.45 and 2.20 ppm, and 5 4.00, 3.51, 3.45, 2.40 and 2.23 ppm, with an integration ratio of 2:1:1:2:2, respectively (Figure 22a). On comparison with data for the free imidazole ligands, the H-2 and H-5 resonances in the complexes are shifted to lower field, the effects being more marked for 15 than 16. The smaller shift in H-5 in J_6 (vs. 6) again supports chelation via the nitrito group in 6. The geometric arrangements of ligands within 15 and 16 cannot be deduced from these spectra. The ' H nmr spectrum for complex J_7 in CDCI3 (Table VII) again shows five sets of multiplets of CH2 resonances centred at 5 3.80, 3.27, 3.12, 2.50 and 2.20 ppm with an integration ratio of 2:1:1:2:2. The ^H nmr data for 18 (Figure 22b, Table VII) also reveal five sets of CH2 multiplets with the same relative intensity as found for 12, centred at 5 3.77, 3.31, 3.22, 2.51 and 2.21. Similarly, ' H nmr data for 19 and 20 also reveal corresponding patterns centred 90 91 Table VTI Selected "^H nmr spectral data for RuQ^tmso) L complexes and the free nitroimidazole ligands. Compounds13 H-2C(s) H-4c(s) H-5C(s) N-CH3(s) Cr^Cm) (tmso) 2, RuCl9(tmso), 2.30 3.42 ~ 4 4.00 15, Ru(4~N09-Im)9 8.92(8.25) 8.21(7.80) 2.20 2.45 ~ ~ 3.15 3.65 4.00 16, Ru(NMe^-N0o-Im)9 7.91(7.54) 8.25(7.87) 3.91(3.90) 2.23 2.40 ~ — Z 3.45 3.57 4.00 17, Ru(De^niso) d 7.48(7.45) 7.38(7.19) 2.20 2.50 ~~ ~~ 3.12 3.27 3.80 18, Ru( SR-2508 ) e 7.51(7.45) 7.48(7.24) 2.21 2.51 ~ 3.22 3.31 3.77 19, Ru'(CMNI) 7.01(7.00) 3.56(3.49) 2.00 2.13 ~~ ~~ 3.15 3.35 3.75 20, Ru"(RSU-3159) 7.73(7.92) 2.21 2.51 ~~ ~ 3.21 3.31 3.75 a 5 i n ppm wrt IMS, i n CDCl^ at room temperature. In each case, the integrations are consistent with the assignment. k 5i> a n c^ ^i" represent RuCl^ttmso^j RuCl 9(tmso) 3 and RuC^tmso), respectively; a l l sulphoxides ligands are S-Donded. shift \ ligand. Sh values given i n parentheses are for the free nitroimidazole d Shift values for the ^N-O^-CWOH) (m), (HO)CH(m) and (HOJCH^m) are 54.39, 4.10 and 3.66 ppm, respectively, and are the same for the free De-miso. 6 Shift values for the ^N-CH^-CtO) (s), H-N-CH^t) and HD-O^(t) are 55.23, 3.66 and 3.39 ppm, respectively, and are tKe same for the free SR-2508. 9 3 at §3.75, 3.35, 3.15, 2.13 and 2.00 ppm and 3.75, 3.31, 3.21, 2.51 and 2.21 ppm, respectively. These relative intensities suggest that the two protons of one of the S - C H 2 firouPs °f the coordinated tmso ligand are magnetically inequivalent. This perhaps results from the fixed orientation of the S=0 when the tmso ligand is coordinated, so that one of the S - C H 2 protons will be in the same "plane" as the O-S-C atoms and the other one will be out of the "plane". If this explanation is correct, one would expect to observe a similar effect by the other S-CH_2 groups giving rise to four sets of S - C H 2 and two sets of S - C - C H 2 - C H 2 multiplets with an integration ratio of 1:1:1:1:2:2. However, this is not observed; perhaps the coupling between one set of S - C H . 2 protons is smaller and the multiplet signal overlaps to give the observed 2:1:1:2:2 integration ratio. On comparison with data for the free ligands De-miso and SR-2508, the H-4 and H-5 resonances in the complexes i l and .18 are shifted slightly to lower field at 57.48 and 7.38, and 5 7.51 and 7.48, respectively. The lH spectrum of J l does not suggest chelation by the SR-2508 ligand. 3.3.3 - H nmr Data of RuB^fdmsoULp Complexes (n=l or 2) The ' H nmr spectrum of 4 in C D C I 3 is complex at room temperature (Table VIII), and consists of a complex pattern from 5 3.3 to 3.6 in the S -bonded region and a singlet at 52.6 for the free dmso, with relative intensity of the S-bonded to free dmso being 3:1 (Table VIII). This *H nmr spectrum is very similar to that observed previously for / raMs -RuB^dmso)^ 7 2 5 and is consistent with dissociation of dmso in C D C I 3 solution to give a mixture of five- and six-coordinate species.725 No O-bonded dmso resonance is observed in the 2.7 ppm region. In the solid state, /rans-RuB^dmso^ contains all S -bonded dmso. 7 2 5 94-Table VIII Selected H nmr and c{ H } chemical shifts for RuBro(dmso)o(4-N0o-Im)o and RuBro(dmso)o(NMe^+-N09-Im) Compounds^ H-2 C-2 H-5 C-5 N-CH3 CH3 (dmso) 4,trans- RuBr2(dmso)4 3.35 3.41 3.46 3.57 21, Ru"(4-N02-Im)2 8.91(8.250° 112.35a 8.20(7.800° 116.13a 3.59 3.70 45.21 47.22 22, Ru"(NMe-4-N02-Im) 8.03(7.54)° 108.62d 8.80(7.87)° 112.39d 3.95(3.90)° 119.93d 3.49 3.53 3.61 3.65 44.91 45.16 47.12 47.58 a S i n ppm wrt IMS, i n CDC13 at room temperature. k Ru" represents RuBr2(dmso)2; both sulphoxides ligands are S-bonded. c 1 H nmr shift values i n parentheses are for the free nitroimidazole ligand. d 13C|1H} values for 21 and 22; C-4 for imidazole carbons of 21 and 22 appear at 119.89 and 116.16 ppm, respectively. 95 The 1 H nmr data for 21 and 22 (Table VIII), like those for 4, show only S-bonded dmso, in agreement with the ir data (see section 3.2.3). The *H nmr spectra for these dibromo complexes 21 and 22 are markedly different from those of the corresponding dichloro complexes 5 and 6. The 4-N02-Im complex 2\ has two sets of quartets centred at 5 3.59 and 3.70 ppm, revealing inequivalent methyl groups for each of the two equivalent dmso ligands (Figure 23a). These data suggest an all trans configuration for 2L As seen in equation [3.2], substitution of an S-bonded dmso and another one trans to it would lead to the all trans geometry of the product. For the NMe-4-N0 2-Im complex 22, there are four quartets of equal intensity centred at 6 3.65, 3.61, 3.53 and 3.49 ppm, which indicate inequivalent methyl groups for each of the two inequivalent dmso ligands, consistent with a /rans-bromides, c/s-sulphoxides, c/s-NMe-4-N0 2-Im configuration (Figure 23b). This geometric configuration would result from 4 by the substitution of one S-bonded dmso and another dmso cis to it. On comparison with data for the free nitroimidazole ligands, the H-2 and H-5 resonances in the complexes 21 and 22 are shifted to lower field (Table VIII). The effect for H-5 is more marked for the NMe-4-N02-Im complex 22, again presumably resulting from the chelation of the NMe-4-N02"Im via the nitrito group, as observed for 6 (see XPS data, section 3.1). This chelation fits the trans, cis, cis configuration predicted from the ir and *H nmr data of 22. Further, the ^C^H} spectra for complexes 2± and 22 (Table VIII) are compatible with the corresponding ^H nmr spectra, showing two and four different types of dmso methyl groups respectively. 4-Nitroimidazole vs. 5-Nitroimidazole Formulation 13< The 4-N0 9-Im and 5-N0 9-Im ligands are tautomers in solution. Laviron 96 I I I I I i I I { 1 I 1 I I 4 3 p p m J. Br .,NMe-4-N0 -Im Z j 2 Br ~ i — i — [ — i i . i | i i i i i * ' i • | • 4 3 p p m Figure 23: \ nmr spectra i n CDC^of (a): RuBr 2 (dmso) 2 (4-NO -Im) 2 , 21, and (b) : RuBr (dmso) (NMe-4-N0 -Im) , 22_. ^ 2 2 9 7 and Grimison et al. have shown that in aqueous solution the 4-NC»2-Im is the predominant form, being approximately 400 times the concentration of the 5-NC»2-Im: However, complexes with 4-NC>2-Im and 5-N0 2-Im as ligands of Pt have been prepared, depending on the experimental conditions:7'*7 at ambient conditions, the white c/s-PtCi2(5-N02-Im)2 complex (as characterized by *H nmr and ir spectroscopies) was precipitated out initially on treating K 2 P t C l 4 with 4(5)-N02-Im ligand in ethanol; use of large excess of 4(5)-N02-Im ligand and refluxing in dimethylfuran for 2 h gave the corresponding bright yellow 4-NC>2-Im complex, cis-PtCl2(4-N02-Im)2, as confirmed by X-ray crystallography. In the present study, complex 5 is considered to contain a 4-N02-Im ligand. Comparison with the data for metronidazole (a substituted 5-NC>2-Im) reveals that the coordination shift in symmetric VJ^Q^ for this ligand (complex J_4) is 6 c m - 1 , whereas for the 4-NC>2-Im (complexes 5 and 15) and 2-N02~Im (complex 13), the shifts are greater (15, 13 and 11 cm"1 respectively, Tables III, IV, V). These trends are consistent with a "closer" N 0 2 substituent to the metal in 5 and 15 (similarly for 13). i.e. a 4- rather than 5-N0 2 substituent. The *H nmr data do not show significant differences between the 4-NC>2-Im (5 and 15) and metronidazole (J4.) complexes, except that the H-2 resonance peak for the 98 4-NC>2-Im complexes is sharp and gives a larger downfield shift compared with the value for the free 4-NC»2-Im ligand. For the 5-NC>2-Im system (14). the H-4 resonance peak is broad and is slightly downfield shifted compared to data for the free metronidazole ligand. Similar differences in shapes and shifts of the resonance peaks are also observed between the c/s-PtC^^-NC^-im^ and m-PtC^S-NC^-Im^ systems.7^7 Use of the NMe-4NC*2-Im ligand, which can exist in only this one form, seemed likely to help resolve the tautomer problem; however, this ligand coordinates in a bidentate fashion through the oxygen of the N 0 2 group, resulting in a greater coordination shifts in the symmetric V N Q (~20 cm"1) for this ligand (see section 3.2.1 and Table III). 99 CHAPTER FOUR AQUEOUS SOLUTION CHEMISTRY OF THE RUTHENIUM NITROIMIDAZOLE COMPLEXES 4.1 Solution Stability of the Substituted-2-nitroimidazole Complexes When the substituted-2-nitroimidazole complexes, JJ_ and 12, are dissolved in water, PBS, or water/PBS containing a 10-fold excess Cl~ or dmso, under N 2 , the blue colour of the solutions disappears in less than 5 min. The visible bands at 638 and 584 nm for complexes JJ. and 12, respectively, as observed in dry ethanol solution or in the solid state, disappear. However, if the complexes are dissolved in water containing a 10-fold excess amount of the respective imidazole ligands, the colour of the solutions remains blue and the spectra are unchanged. These results together with the *H nmr data in D 2 0 (see section 3.3.1) indicate that the disappearance of the blue colour is due to the dissociation of the nitroimidazole ligand(s) and that the presence of the following aquation/anation equilibrium exists [4.1]: RuCl 2(dmso) 2(H 20) x(miso) y + x miso [4.1] Colourless (x + y = 2) Further, the aqueous solution of these complexes are non-acidic and non-conducting, and no free CI" is detected using AgNO^; therefore, the possibility of a chloride dissociation is ruled out. The aquation phenomenon [4.1] is not observed for the tmso complexes ]_7 and 18, and the colour of aqueous solutions of these remained blue for at least up to 8 h. Complexes 12 and 17 are the same except for the sulphoxide ligands, and therefore the difference in the aquation phenomenon perhaps resulting from the lipophilicity of tmso as a ligand (section 6.3). In water, complex 13, which contains the RuCl 2(dmso) 2(miso) 2 + H 20 —+• 11 Blue 100 unsubstituted 2-N02-Im ligand, does not dissociate the NC^-Im and the colour of aqueous solution remained blue for at least 8 h (see section 3.3.1); presumably the hydrophilic nature of the hydroxy-containing substituents of the miso and De-miso systems favours dissociation of the free ligands in the aqueous media. 4.2 Aqueous Solution Chemistry of the RuCl 2 (R 2 S0) 2 L n Complexes: R2SO =  dimethyl- or tetramethvlene-sulphoxide: L = 4-N0 2-Im (n=2) or substituted-4- NQ 2-Im (n=l or 2) On dissolution in water under N 2 , complex 5 instantly gives a molar conductivity (100.8 ohm"1 mol - 1 cm 2, Table IX) corresponding to a 1:1 electrolyte/'*2 which results from loss of a single chloride ligand as estimated by AgN0 3 titration ; no further dissociation of chloride occurs over 2 days. The UV/VIS spectrum of 5 in the solid state shows a m a x identical to that obtained in a solution containing excess CI" ( Am a x=345 nm). These data together with those obtained from the XPS (CI 2p^ 2 3/2) show that 5 is not the ionic compound, RuCl(dmso)2(4-N02-Im)2+Cr. Further, the resulting aqueous solution is weakly acidic (pH=3.9 for a 2 mM solution of 5), implying partial loss of a proton from coordinated water (-7% dissociation based on the pH 3.9 value, or ~38% dissociation based on an experimentally determined p K a value), and therefore the molar conductivity corresponding to 1:1 electrolyte must also result partially from this H + loss. Standard pH-titration experiments775 yield data that analyze well for a p K a value of 4.10 + 0.10 (Figure 24): RuCl2(R2SO)2(4-N02-Irn)2 + H20 —[RuCl(H 20) (R^O^^-NC^-Im)^ + Cl ~ 5, 15 5b, 15b 1! [ 4- 2 1 RgSO - dmso, 5 R U C 1 ( 0 H } ( R S 0 } { 4 _ N 0 _ l m ) + H + + C l ~ R^ SO = tmso, 15 i l i L 5c, 15c The p K a value seems reasonable for H 2 0 coordinated at a Ru(II) center.7'*-* A second proton loss measured at higher pH (pK' a = 8.90 + 0.10) (Figure 24) is attributed to loss of the pyrrolic-N hydrogen to form coordinated 4-nitroimidazolate; a 101 Table IX Molar conductivity (A^) for some Ru complexes i n water, at 25°C and 2mM. Complex3 A M -1 -1 2 ohm mol cm 1, c£a-RuCl2 (dmso) ^ 12.4 2, RuCl^ltmso)^ 13.7 5, Ru(4-N02-Im)2 100.8 6, Ru(NMe^+-NO?-Im) 15.5 7, Ru(RSU-1170)2 35.2 h Ru(RSU-3083)2 50.1 9, Ru(RSU-3100)2 37.4 10, Ru(RSU-3159) 30.7 Ru(miso)2 14.1 12, Ru(De-miso)2 16.8 Ru(2-N02-Im)2 27.1 Ru(metro)2 22.3 15, Ru'(4-N02-Im)2 108.2 16, Ru'(NMe-4-N02-Im)2 18.4 Ru'(De-miso)2 29.1 18, Ru'(SR-2508) 25.4 21, Ru"(4-N02-Im)2 118.7 22, Ru"(NMe-4-N02-Im) 20.7 Ru, Ru', and Ru" represent RuCl2(dmso)2, RuCl 2(tmso) 2, and RuBr2(dmso) respectively. 102 1 1 1 r 10 20 30 40 Vol of NaOH (mL) Figure 24: pH T i t r a t i o n curves obtained using 4 mM NaOH. 103 corresponding value for imidazole itself within Ru(NH3)5(imidazole)2+ is also 8.90.777 A UV/VIS spectrum recorded for 5_ at pH 7.0, a pale brown solution ( A m a x 338 nm, log E = 4.63) thus refers to species 5£ minor differences are noted at pH 3.0 (species gb, A___ 343 nm) and at similar acidity in the presence of added CI" (0.1 M) (species I l ia A, 5_, A m a x 345 nm). Based on the equilibria shown in equation [4.2], the results suggest 144 that the high CI concentration in the blood ([CI ] ~90 mM) would prevent the hydrolysis of complex 5_, and thus species 5_ would dominate in such a medium. However, hydrolysis can presumably occur inside the cells ([CI"] ~1 mM)^^ after penetration through the cell membrane. The main species present under physiological conditions (pH 7.4) in the presence of low Cl~ concentration would be predominantly species 5_c_. The clinically used drug cis-DDP has been shown to undergo similar type of equilibria reactions under physiological conditions (see eq. 4.3).^1,144 The results from the studies on complex 1_5_, the tmso analogue, show very similar behaviour in aqueous media (Table IX) giving a A =^108.2 ohm"1 mol"'cm2 and a pKa=4.4, resulting from the loss of a chloride and the H + from the coordinated water. A second proton loss measured at higher pH (pK'a = 9.01 + 0.10) is again attributed to loss of the pyrrolic-N hydrogen to form coordinated 4-nitroimidazolate. cis-PtCl 2(NH 3) 2 cis-[PtCKH 20)(NH 3) 2] + ^ [F t C N H ^ C r ^ O ) ^ -H+ |J-H+ pKa = 5.6 1^ 0 cis-[PtCKOH)(NH 3) 2] [Pt(OH)(R20)(NH3)2] 7.3 (Taken from ref. 50) [Pt(OH) 2(NH 3) 2] Titration with base of the other complexes 6-10, J4, 16-ii and 23-27 shows no titratable protons are present at least up to pH 11 and, consistent with this, molar 104 conductivity values measured within 10 min of dissolution are lower (15.5-50.1), showing little replacement of Cl" ligand by H 2 0 . The lack of Cl" dissociation may result from the electron donating effect of the substituents in the substituted-4-nitroimidazoles, and the greater basicity of the phenanthroline ligands (vs. 4-N02"Im). The UV/VIS spectra of all the complexes (see sections on Preparation of Complexes) are dominated by charge transfer bands in the 300-450 nm region, and offer little insight into the coordination geometry. 4.3 Aqueous Solution Chemistry of the RuBr2(dmso)2Ln Complexes (L^-NCs-Im. n=2: L=NMe-4-NQ 2-Im. n=l) On dissolution in water under N 2 , complex 2 1 , RuBr2(dmso)2(4-N02-Im)2, instantly gives a molar conductivity of 118.7 ohm 'mol cnr4 (Table IX), again corresponding to 1:1 electrolyte/^ which results from loss of a single bromide ligand as evidenced by AgBr formation; no further bromide loss occurs over 2 days under N 2 . As with 5_ and J_5, hydration of 21 occurs giving species 21b. However, unlike the aquo derivatives (5_b and 15b) of 5 and 15 , 21b does not deprotonate at the coordinated H 2 0 or coordinated 4-N0 2-Im, at least up to pH 11. Thus, the higher molar conductivity for 21 as compared with 5 and 15 is surprising because there is no H + contributions. Based on these results, the following solution equilibrium [eq. 4.4] is suggested: RuBr2(dmso)2(4-N02-Im)2 + ^ 0 — [RuBrd^O) (dmso)2(4-N02-Im)2]+ + Br~ 21 21b [4.4] Thus the more basic bromide system, 21, (compared to the chloride system, 5) has 7 7 coordinated H 2 0 and imidazole ligands that are at least 10 and 10 times, respectively, less acidic than the corresponding ligands in 5. The lower acidities are consistent with the presence of more basic bromide ligands, but a more important factor almost certainly arises because of differences in the geometrical isomers present. 105 Complex 5 (and 15) has all cis geometry [eq. 4.5] and the chloride replaced is likely to be the one trans to the 4-NC>2-Im (even though dmso has a strong trans effect ), because only complexes with 4-N02~Im (5 and ]_5), but not N-substituted-4-NC^-Im, show the dissociation of a chloride (eq. 4.2 and 4.5): Ru-Cl 5 Cl -Cl" L Ru 01 ,H2 5b Cl [4.5] For 21, bromide loss must occur according to eq. 4.6: Br- Ru L 21 -Br -Br" Br L •Ru L 21b [4 .e; For ligands trans to water, it has been argued that the higher the trans effect, the weaker the metal-oxygen bond and the stronger the oxygen-hydrogen bond, i.e. the weaker the acidity of the coordinated I^O.7^"* However, this suggestion was based on data for just cis and /ra«.s-Pt(NH 3)2(H20)2^ + complexes (cis, p K a = 5.6; trans, p K a = 4.3). For the data: trans-?t(C2Hd)(U20)1C\2 (pK a = 5); cis-Pt(NH 3)(H20)Cl 2 (pK a ~7), the correlation is clearly not valid because in term of trans effects, C2H4 » Cl". Here, the strong TT -acid properties of the C2H4 were invoked to account for stronger 106 acidity of the water within the ethylene complex. The strong trans effect - weak acidity argument clearly does not hold for 21b vs. 5t> because in the latter, the water is probably trans to the 4-N02"Im, which is likely to have a strong trans effect judging from its unsaturated nature. Thus 4-N02"Im is considered to have a stronger trans effect than Br", and yet the coordinated water in 5b is much more acidic than that in 21b. The strong acidity could be rationalized in terms of the presence of the strong trans IT -acceptor 4-NC>2-Im. Thus, if the structures of 5 and 2i are as shown (the evidence being nmr data, sections 3.3.1 and 3.3.3), then the monoaquo species are probably the isomers shown in 5b and 21b (eq. 4.5 and 4.6). The discussion on the differences between the aqueous solution chemistry of 21 with 5 (and 15J is clearly very speculative, but at least it mentions the factors that have been used in the literature to account for the acidities of coordinated water. Species 21b would predominate in aqueous solution and presumably under physiological conditions. A UV/VIS spectrum recorded for 21 at pH 7.0 ( A m a x 342 nm, log£= 4.02) thus refers to species 21b: minor differences are noted in the presence of added Br" (0.1 M) (species 21, A _ o v 347 nm, log£= 4.13). Titration of complex 22 with base shows no titratable protons up to pH 11 and the molar conductivity value measured under the same conditions is low (Table IX), showing little replacement of Br" ligand by water. Clearly, the aqueous solution chemistry of the complexes is germane for an understanding of their biological properties. Of the complexes synthesized, only 5, 15 and 21 demonstrate high lability according to equations [4.2] and [4.4]. 107 CHAPTER FIVE CYTOTOXICITY AND RADIOSENSITIZATION IN VITRO An important objective in investigating the sensitizing properties of new compounds is to identify a drug that will be superior to misonidazole in the clinic. In addition, these compounds should have beneficial properties, e.g.: the ability to penetrate into the centre of the tumour; metabolic stability and have low side-effects in vivo, but these studies are beyond the scope of the present work. However, the more fundamental beneficial properties of the compounds such as larger SER value in hypoxic cells but not normal oxic ones, and low cytotoxicity, can be assessed in vitro and are discussed below. 5.1 Toxicity of the Ru Complexes and their Nitroimidazole Ligands Various in vitro studies with Chinese hamster cells have been used to study hypoxic cytotoxicity of nitroimidazoles-^'-^"*6*7^ and platinum complexes.-57'60'7'5 The toxicity of the chemicals is measured as plating efficiency (PE) at various times of drug treatment. In the present experiments, t=0 denotes the time when the cells are added to the compound dissolved in growth media, immediately removed, washed and plated for survival. The data given represent the average of three resulting PE values. The higher the PE value, the lower the toxicity. The PE values for CHO cells plotted as a function of incubation time at 37°C in solutions containing 4-N02~Im and the Ru complex 5, under hypoxic and oxic conditions, are shown in Figure 25. For each point, three aliquots of cell suspension were processed and plated independently on the same day. The data show that the hypoxic toxicity of the complex at 200 juM is significantly less than that of the free ligand at equivalent concentration (400 ^«M). There is no toxicity for either compound at the corresponding concentration in aerobic cells. Note that there has been no 108 U J 1 -01 T I M E C h o u r s ) Figure 25: T o x i c i t y of RuCl (dmso) (4-NO -Im) , 5, and 4-NO -Im i n 2 2 2 2 — 2 hypoxic and oxic CHO c e l l s . Medium containing the indicated concentration of chemical was degassed f or 1 h at 37°c before c e l l s were added (t=0). Hypoxic control (•), N 2 + 400 pM 4-N02~Im (T) , N 2 + 200 \iH of 5^  (A) , and 0 2 + 200 uM of 5 (•). 109 normalization of PE against the drug free control. The hypoxic toxicity data for the other 4-NC>2-Ini complexes 15 and 21 are summarized in Table X. Their toxicities are similar to the value obtained for 5. The toxicity data obtained when CHO cells are exposed to the various N -substituted-4-nitroimidazoles and their Ru complexes (6-K), 16 20 and 22), under hypoxic conditions after 4 h of incubation, are shown in Table X. The hypoxic toxicity of the complexes at lOO^M is again significantly less than that of the corresponding free ligand at equivalent concentration. In addition, no toxicity is seen in an oxic environment at these times and concentrations. The complex concentrations used for comparison with the ligands NMe-4-N0 2-Im and RSU-3159 were 200 and 100 MM, respectively, as the complexes contained only one such ligand per metal. No significant difference in PE is observed between the dmso and tmso complexes (6 vs. 16) or the Cl" and Br" complexes (6 vs. 22). These data show that the substitution of dmso by tmso and Cl" by Br" ligands does not alter the toxicity of the corresponding complexes. The hypoxic toxicities of complex 12 and that of its free CMNI ligand were very high, with PE values at 50 JJM of about 0.0007 and 0.0004 respectively, and therefore their radiosensitizing abilities were not studied. The PE values of CHO cells after 4 h incubation at 37°C with various 2-N02~Im derivatives, metronidazole and their complexes under hypoxic conditions are shown in Table XI. For complexes JJL, 12, J_4, and 18., the hypoxic toxicities of these complexes at 200 MM are similar to those of the free ligands at 400 MM. The hypoxic toxicity data for complexes 13. and 12 are significantly lower than those of their corresponding free ligands at equivalent concentrations. Again, no toxicity is seen in aerobic cells. 110 Table X Plating efficiency (PE) of the 4-nitroimidazole ligands and their Ru(II) complexes i n hypoxic CHO cells (4 h at 37 C) PE (cone. |jM) a Complex „ ,. , , r Free ligand complex 5, Ru(4-N02-Im)2 0.052 (400) 0.10 (200) 6, Ra(NMe-4-N02-Im)b 0.49 (200) 0.62 (200) 7, Ru(RSU-1170)2 0.037 (100) 0.15 (200) 8, Ru(RSU-3083)2 0.036 (100) 0.12 (200) 9, Ru(RSU-3100)2 0.0042 (100) 0.12 (200) io, RutRsu-aisg)^ 0.0027 (100) 0.16 (100) 15, Ru"(4-N02-Im)2 0.058 (400) 0.13 (200) 16, Ru'(NMe-4-N02-Im)2 0.49 (200) 0.53 (200) 19, Ru'(tmsoMCMNI) 0.0004 (50) 0.0007 (50) 20, Ru'(RSU-3159)b 0.0027 (100) 0.11 (100) 21, Ru"(4-N02-Im)2 0.058 (400) 0.32 (200) 22, Ru'"(NMe^+-N02-Im)b 0.49 (200) 0.62 (200) a respectively b Ru, Ru', and Ru" represent RuClo(dmso)9, RuCl(tmso)^, and RuBr2(dmso)9, , except 2(] contains only 4me tmso ligarid. z L The concentrations used for the free ligands NMe-4-N02-Im and RSU-3159 (both presumably bidentate) were 200 JJM and 100 J J M, respectively. I l l Table XI Hypoxic toxicity (PE) of the 2-nitroimidazoles and metronidazole and their Ru(II) complexes i n CHO cells (4 h at 37°C)a b Complex Free ligand (400 pM) PE Complex (200 MM) 11, Ru(miso)2 0.75 0.77 12, Ru(De-miso)2 0.12 0.14 13, Ru(2-N02-Im)? 0.0031 0.52 14, Ru(metro)2 0.80 0.79 17, Ru'(De-miso)2 0.13 0.42 18, Ru'(SR-2508)c 0.073 0.099 Ligand and complex concentrations used were 400 and 200pM, respectively, unless stated otherwise. k Ru and Ru' represent RuC^dmso^ and RuC^tmso^s respectively. C The concentration used for 18 and the ligand SR-2508 were 200 MM, because the complex contains only one such ligand per metal. 112 The hypoxic toxicities of the uncoordinated 1,10-phenanthrolines are very severe, with PE values of about 0.001 at 20 MM after 3 h incubation. However, the corresponding Ru complexes (23-25) show drastically reduced hypoxic toxicity with PE values of 0.05 at 200 uM after 3 h incubation. Only a small degree of toxicities is observed in aerobic cells for both free 1-10-phenanthroline ligands and the associated Ru complexes. The cytotoxicity of nitroimidazole radiosensitizers is believed to result from metabolic reduction of the nitro group to derivatives which cause damage in cells."*'*'5^'7-*5 This reduction only proceeds under hypoxic conditions or at low oxygen concentrations and is known to involve the initial addition of a single electron. The subsequent addition of electrons proceeds via formation of the nitroso, hydroxylamine and amine derivatives:7-*5 R-N0 2 R N 0 2 T — RN=0 — - R-N-0 — - RNHOH — RNR^ [5 .1] In the absence of oxygen, a toxic species is produced which interacts with crucial cellular molecules and cell death occurs.-*'*'-^ This reduction process is prevented by the presence of oxygen which reverses the first stage of nitroreduction, the formation of a radical anion (see section 1.8). This explains why oxic toxicity is not as high as hypoxic toxicity. A change in biological interactions with the reduction enzymes, and/or the depletion of reducing equivalents in hypoxic cells^"* 9 , 5 9 , 7 -* 5 may explain why the Ru complexes are always less toxic than the free ligands. 5.2 Chromosomal-Damaging Activity of Complex 5 The most dramatic effect on biological systems is DNA damage; this can result in some genotoxicities such as chromosomal aberrations,70-* which can eventually transform 113 the cell or render the cell unable to divide, leading to cell death. Because complex 5, RuCl2(dmso)2(4-NC>2-Im)2, is a good radiosensitizer, less toxic than the free 4-N02~Im ligand and has been shown to bind to DNA (see section 6.1.1), it was examined for a genotoxic effect on normal cells, as measured by the in vitro induction of chromosome aberrations in CHO cells. The frequencies of chromosome aberrations in CHO cells exposed for 3 h to complex 5 are shown in Table XII. Complex 5 showed a dose-dependent increase in chromosome damaging activity in CHO cells. The extent of clastogenic activities is expressed as the percentage of metaphases plates with at least one chromatid break or one chromatid exchange. The percentage of metaphases with chromatid aberrations following exposure to 5 is much less than that elicited by exposure to c/s-DDP. There is a significant increase in chromosomal breakage at 10 ,uM c/s-DDP and the drug becomes toxic, leading to cells death in this experiment after this dose (Table XII). This difference in effects of 5 and c/s-DDP might be the result of differential repair of the lesions by the cells.'5"'* Several chemicals can be activated by the mixed function oxidases present in the S9 mixture to yield electrophilic species, such as in the case of nitroso species or alkylating agents/^ which bind to DNA, RNA or proteins.7**7 In this study, addition of the S9 mixture did not enhance the clastogenic activity of 5 and c/s-DDP, indicating that the complexes were acting directly. In a similar study,7'** S9 had no effect on the mutagenicity of Pt(II) and Pt(IV) coordination complexes as measured in the Ames Salmonella mutagenic assay. Binding of Pt complexes to DNA is presumably responsible for their genotoxic effects;^7 ruthenium complexes have anti-tumor activity*^'^ and also bind to D N A 8 0 ' 8 4 , 1 5 2 and thus the lack of activation by the S9 mixture seems reasonable. 114 Table XII Clastogenic activity of RuCl9(dmso)9(4-N09-Im)9 (5) and cis-DDP on CHO cells ' z z Z Average 70 metaphase plates with chromatid aberrations' RuClo(dmso)o(4-N0o-Im)o ets-DDP Cone. (mM) +S9 Cone. (JJM) +S9 10.0 31-3(0.18) 29±3(0.14) 50.0 Tb T 6.0 26±2(0.10) 22±2(0.09) 30.0 T MI 2.0 19±1(0.04) 19±2(0.05) 10.0 MI C 80-10(3.66) 1.0 14±2(0.02) 15±3(0.03) 8.0 61-5(2.54) 67±5 (2.51) 0.4 9±2(0.01) 8±2(0.02) 4.0 45±4(1.32) 40±5 (1.28) 0.2 4±1(0) 5±1(0) 2.0 30±5(0.76) 28-2 (0.85) a. Average "L of three plates - s.d.; the figures i n parentheses show the average number of chromatid exchanges per metaphase. b. T = Toxic, no metaphase was observed. c. MI = Mitotic inhibition: fewer than 40 diploid metaphases per plate observed. 115 For purposes of comparison, the clastogenic activities of the precursor, cis-RuCl2(dmso)4 complex, and the 4-NC*2-Im ligand, were also studied (Table XIII). Similar frequencies of chromatid aberrations are induced by the precursor and by the ligand at the same concentrations. Further, both the precursor and the 4-NC>2-Im ligand produce fewer chromosome aberrations than 5_ at an equivalent molar concentration. The enhanced clastogenic activity of 5, however, appears to result from the combined effect of the metal and the 4-N0 2-Im ligand (Tables XII, XIII), suggesting that both components of 5 are contributing to the chromatid aberrations. Incubation with activating enzymes (the S9 mixture) does not increase the extent of damage by the precursor, i or 4-N0 2-Im (Table XIII). At these concentrations, dmso ligand does not show any genotoxic effect. The clastogenic activity of complex 5 is also compared to that of the clinically used nitroimidazole radiosensitizer, miso. The frequencies of metaphase plates with chromatid aberrations following exposure to 5, 4-NC»2-Im, or miso, are shown in Figure 26. Complex 5 and miso produce a similar yield of such metaphases with chromatid aberrations that is higher than that of 4-N0 2-Im at the same concentration. The results indicate that 5 might have some advantage as a radiosensitizer over miso, because the complex produces a similar percent metaphases with chromatid aberrations but yields a higher SER value (see section 5.3). In addition, 5 is much less genotoxic than m-DDP, which produces more chromatid multiple-exchanges7^ and which also is used in conjunction with radiation. ' Despite the lack of understanding of the mechanism of chromosome aberrations, accumulated data have suggested the role of chromosome damaging activity in 116 Table XIII Clastogenic activity of cts-RuCl9(dmso)/ and 4-N00-Im on CHO cells Z 4 1 Average "L metaphase plates with chromatid aberrations cis -RuCl 0 (DMSO), 4-nitroimidazole Cone (mM) +S9 +S9 10.0 15±2(0.07) 13±2(0.07) 16±3(0.06) 17-3(0.05) 6.0 12±2(0.04) 10±1(0.05) 13-2(0.04) 14±2(0.03) 2.0 7±2(0.02) 6±2(0.03) 10±2(0.02) 11±2(0.01) 1.0 4±1(0.01) 3-1(0.01) 7±2(0) 6±1(0) 0.5 2±1(0) 1-1(0) 4±1(0) 3±1(0) a. Average % of three plates - s.d.; the figures i n parentheses show the average number of chromatid exchanges per metaphase. 117 C D R U G ) C m M ) Figure 26: Chromosome damaging a c t i v i t y of 5_ (•), miso (A) and 4 ~ N ° 2 ~ I m ( - ) i n C H O c e l l s following 3 h exposure (in the absence of S9). 118 carcinogenesis. ' Evaluation of the genotoxicity of agents with pharmaceutical potential is important. 5.3 Radiation Sensitization of Aerobic and Hypoxic CHO Cells In this study, the sensitizing enhancement ratio (SER) value is calculated as the ratio of X-ray doses for cells in medium with and without the test compound required to produce a survival fraction of 0.01. The linear-quadratic CY, (3 model'2"5 was used to fit curves to the data (see section 2.7). Each data point represents the average of three individual experiments. No radiosensitizing enhancement or protection is observed at 400 dmso or tmso, or with 200 m-RuCl 2(dmso) 4, RuCl2(tmso)4 or trans-RuB^dmso^ in control experiments. In these experiments, the chemical is added to the cell suspension and incubated for 1 h at 37°C prior to the start of irradiation at 4°C. In no instance was there measurable toxicity from exposure to the compounds under these conditions: the PE values for zero X-ray dose control samples taken immediately prior to irradiation are the same as for cells not exposed to the chemical. 5.3.1 Radiosensitization by the Metal Complexes It is well established that c/s-DDP, trans-DDP, and a variety of other Pt(II) and Pt(IV), complexes are radiation sensitizers for hypoxic cells, although the detailed mechanisms are not known. •50,57,57,72,76,705 s o m e //-arcs-complexes of Pt are also radiosensitizers and are generally much less cytotoxic than cis-complexes.7^'7^-* C/s-dichlorobis(metronidazole)platinum(II) (FLAP) 7 , 2 was the first reported metal-sensitizer complex studied but its "initial" activity could not be substantiated by later reports.74>7$J03 i n contrast, the present study demonstrates that an enhancement of X-ray damage in hypoxic CHO cells in vitro can indeed be achieved with a range of different Ru(II) nitroimidazole complexes. 119 All of the Ru(II)-nitroimidazole complexes sensitize to a greater extent in hypoxic than oxic cells. The SER values for the free 4-N0 2-Im ligands and respective complexes in hypoxic conditions are summarized in Table XIV. The survival curves obtained with complex 5 and free 4-N0 2-Im ligand in CHO cells are shown in Figure 27 (Table XIV). The SER value with 200 MM of 5 is 1.6 in hypoxia, while there is no sensitization of oxic CHO cells to X-ray by this complex. The SER value for 4-N0 2-Im at equivalent concentration (400 ^MM) is only 1.2 and again no sensitization of oxic CHO cells is observed. The oxygen enhancement ratio (OER) in this experiment is 3.0 which is similar to the value reported by others for CHO cells.6'* The SER value for the tmso analogue of 5, complex 15, is also 1.6 at 200 juM in hypoxia (Figure 28 and Table XIV). There is a measurable increase in SER value over ligand alone for the dibromo complex 21 at 200 MM concentration (Figure 29). Thus, RuBr2(dmso)2(4-N02-Im)2, complex is not as good a radiosensitizer as the dichloride analogue; this is certainly related to the different isomer forms and aqueous solution behaviour (see sections 4.1 and 4.2). Figure 30 shows the survival curves obtained when complex 6 or NMe-4-N0 2-Im is present during irradiation of hypoxic CHO cells. The SER values obtained for 6 and NMe-4-N0 2-Im at 200 pM are 1.3 and 1.2, respectively. The SER values for the other N-substituted-4-N02-Im ligands and their Ru(II) complexes, 7-10, 16 and 20, are summarized in Table XIV. The SER values for these complexes in hypoxia are between 1.1 and 1.3 and again no radiosensitization of oxic CHO cells by any of these complexes is observed. In general, the SER values of the N-substituted-4-N02-Im complexes are lower than those of 5 and 15 containing unsubstituted 4-N0 2-Im ligands. 120 Table XIV The hypoxic SER values of the 4-nitroimidazole ligands and their Ru(II) complexes Complex3, SER values d Free ligand complex (cone. pM) (cone. |jM) 5, Ru(4-N02-Im)2 1.2 (400) 1.6 (200) 6, Ru(NMe-4-N02-Im)b 1.2 (200) 1.3 (200) 7, Ru(RSU-1170)2 1.1 (200) 1.3 (100) 1> Ru(RSU-3083)? 1.2 (200) 1.2 (100) 9, Ra(RSU-3100)2 1.2 (200) 1.1 (100) 10, (RSU-3159)b 1.3 (100) 1.3 (100) 15, Ru'(4-N02-Im)2 1.2 (400) 1.6 (200) 16, Ru' (MMe^N02-Im)b 1.2 (200) 1.4 (400) 19, Ru'(tmso)(CMNI)C 1.2 (50) 1.3 (50) 20, Ru'(RSU-3159)b 1.3 (100) 1.3 (100) 21, Ru"(4-N02-Im)2 1.2 (400) 1.3 (200) 22, Ru"(NMe^+-N02-Im) 1.2 (200) 1.3 (200) c i Ru, Ru', and Ru" represent RuCl2(dmso)2, RuCl 2(tmso) 2, and RuBr2(dmso)2, respectively, except 2!0 contains only one tmso ligand. b The concentrations used for comparision with the ligands NMe-4N02-Im and RSU-3159 (presumably bidentate) were 200 (jM and 100 pM, respectively. The concentrations used for this complex and i t s ligand were 50 |JM because of their toxicity. d Estimated maximum error + 0.05. 121 i 1 1 T 1 r Z j i i I I I S 18 15 20 25 30 D O S E ( G r a y ) Figure 27: Radiosensitization of CHO c e l l s by RuCl (dmso) (4-NO -Im) 2 g 2 2 5_, or 4-NO -Im. The c e l l s were incubated at 37 C f o r 1 h. Hypoxic control (•), N + 400 pM 4-NO -Im (•), N + 200 pM 2 2 2 5_ (A) , oxic control (•) , and O + 200 pM 5 (•) . 122 i i 1 1 1 r Z D O S E ( G r a y ) Figure 28: Radiosensitization of CHO c e l l s by RuCl (tmso) (4-NO -Im) 2 2 . 2 15, or 4-NO -Im. The c e l l s were pre-incubated at 37 c f o r 1 h. Hypoxic control (•) , N + 400 \iM 4-NO -Im (A) and N 2 2 2 + 200 pM 15 (•) . 123 IS 28 25 38 D O S E ( G r a y ) Figure 29: Radiosensitization of CHO c e l l s by RuBr^(dmso) (4-NO^-Im)^» 21, or 4-N02~Im. The c e l l s were pre-incubated at 37°C f o r 1 h. Hypoxic control (•); N 2 + 400 pM 4-N02~Im (A) and N 2 + 200 pM 21 (•). 124 T 1 1 1 r i _! i i 1 — i J 0 5 I e 15 20 25 30 D O S E C G r a v > Figure 30: Radiosensitization of CHO cells by RuCl (dmso) (NMe-4-N0 -Im), 2 2 2 6, or NMe-4-N02~Im. The cells were pre-incubated at 37°C for 1 h. Hypoxic control (•), N 2 + 200 M  NMe-4-N02~Im (A), and N2 + 200 pM 6_ (•) . 125 The survival curves obtained with SR-2508 and RuC^tmso^SR-2508), 18, in CHO cells under hypoxia are shown in Figure 31. The SER values for the ligand SR-2508 (200 jM), and the relevant complex 18 (200 ^M), are 1.3 and 1.5, respectively. Complex J_8 has the highest SER value of all the 2-nitroimidazole complexes being studied probably because it is stable in solution with respect to loss of the nitroimidazole ligand unlike the other substituted-2-nitroimidazole species. The free ligand, SR-2508, has been shown to yield a good SER value (~ 1.6 at 1.5 mM). 7 * 9 The SER values for the other 2-nitroimidazole and metronidazole complexes (11-14 and 12) in hypoxia are between 1.2-1.4 (Table XV). There are no significant increases in SER values between the dmso complexes ( H and 12) compared with those for their free nitroimidazole ligands probably because these dissociate from the complexes in solution. As with all the other systems, no oxic sensitization was observed. For the other complexes containing no nitroimidazole ligands, such as 23-29, no radiosensitization effect was observed. 126 D O S E ( G r a y ) Figure 31: Radiosensitization of CHO c e l l s by RuCl (tmso) (SR-2508), 18_, or SR-2508. The c e l l s were pre-incubated at 37°C f o r 1 h. Hypoxic control (•) , N 2 + 200 pM SR-2508 (A) and N 2 + 200 \iM 18 (•). 127 Table XV The SER values of 2-rdtroimidazoles, metronidazole, and their Ru(II) complexes SER values 0 Complex Free ligand (400 pM) Complex (200 pM) 11, Ru(miso)2 1.3 1.4 12, Ru(De-Miso)2 1.3 1.4 13, Ru(2-N02-Im)2 1.3 1.4 14, Ru(metro)2 1.1 1.2 17, Ru'(De-miso)2 1.3 1.4 18, Ru'(SR-2508)b 1.3 1.5 Ru and Ru' represent RuCl 2 k The concentration used for complex contains only one (dmso)2 and RuCl2(tmso) the ligand SR-2508 was such ligand per metal. 2, respectively. 200 MM, because the Estimated maximum error + 0.05. 128 CHAPTER SIX FACTORS RELATED TO RADIOSENSITIZATION Chemicals in general can modify biological response to radiation in a variety of ways / 5 0 for example: binding to DNA, interaction with radio-protective thiols, and enhancing radiation damage by electron affinity of the chemicals In most cases where compounds have been shown to radiosensitize, one or several mechanisms have been proposed to explain the observation. The mechanisms of interaction listed below are not necessarily independent of each other. 6.1 DNA Binding With the intent of improving the radiosensitizing properties of radiosensitizers by using DNA-binding metals to carry the sensitizer to the DNA, questions as to whether the desired goal of targeting is achieved, i.e. whether these complexes are in fact binding to DNA, need to be addressed. 6.1.1 Assessment by Inhibition of Restriction Enzymes The method described here for comparing the binding of complexes to isolated DNA is a variation of that used previously to study cis and trans-D D P , 7 5 7 and has been modified and expanded by Skov et al}22 Plasmid was linearized in order to simplify the interpretation of gel pattern, and titration (inhibition vs. concentration) was used to provide a means of comparing the DNA binding agents. The assay provides a quick comparison of binding by various compounds. It is a qualitative assay at present - the numbers obtained are meaningful only when compared with others obtained under the same experimental conditions such as temperature, incubation time and pH. 129 The binding site for cis and trans-DDP has been known for some time to be on G-rich areas of D N A . 5 2 , 7 5 7 Restriction enzymes which cut DNA in G-rich areas are more inhibited than those with fewer G residues in their recognition sequence.757 The enzymes BamHI (G-rich) and EcoRI (G-poor) were chosen to see whether this specificity was affected by attachment of ligands and complexes. There is no intent to imply that the binding of Ru complexes occurs specifically at the BamHI site; this location is used only because of its guanine (G) content, which is expected to have a relatively high attraction for platinum metals. 8 0 ' 5 ^' 7 2 2 ' 7 5 7 6.1.2 DNA Binding by the Ru Complexes The complexes were assessed qualitatively, i.e., whether there is noticeable inhibition at 200 uM, after 1 h incubation. Of all the complexes studied, only 5, 23-25, and 28, show inhibition of the endonuclease activity of BamHI, but not of EcoRI, on plasmid pSV2-gpt (Figure 32). These findings are assumed to be related to the binding of the complex at or near the GGATCC sequence, and suggest that these complexes preferentially bind to G -rich regions similar to those reported previously for other ruthenium complexes. 0U<0J>° •°° A range of concentrations of 5 was also examined and compared with data for c/s-DDP as shown in Figure 33. Both 5 and c/s-DDP inhibited BamHI but c/s-DDP also inhibited EcoRI activity. The concentration for 10% inhibition of BamHI by 5 was 27 MM which was higher than that of c/s-DDP (12 MM) but lower than that for c/s-DDP inhibition of EcoRI (30 MM for 10%). Table XVI summarizes the data (concentration to achieve 10% inhibition) for several of the complexes examined. It should be noted that in control experiments with free ligands, the only inhibition observed in the concentration range under study for the complexes was that for the 4-N02"Im 130 a b c d e f g h i j k l m n Figure 32: I n h i b i t i o n of endonuclease a c t i v i t y of BamHI and EcoRI. (a) DNA alone; (b) DNA + 5_ + EcoRI, (c) DNA + 5 + BamHI; (d) DNA + 15_; (e) DNA + 15 + EcoRI; (f) DNA + 15 + BamHI; (g) DNA + 23_; (h) DNA + _23 + EcoRI; (i) DNA + 2_3 + BamHI; (j) DNA + 24 + BamHI; (k) DNA + 5-C1-1,10-phenanthroline; (1) DNA + 5-NO -1,10-phenanthroline; (m) DNA + ais-DDV + BamHI; (n) DNA + BamHI. 131 Figure 33: T i t r a t i o n of BamHI s i t e with RuCl (dmso) (4-NO 5_, 4-NO -Im, and c-DDP (Cts-DDP). 2 2 Table XVT Relative inhibition of BamHI activity by some Ru complexes, cis -DDP and trans -DDP Approximate concentration Relative inhibition Complex3 for 10% inhibition (|jM) / i n i t i a l slope,, { e.g. Fig. 33 ; cis -DDP 12 0.85 trans -DDP 12 0.85 cis -RuCl2(dmso)^, 1 b 0.06 Ru(4-N02-Im)2, 5 27 0.38 Ru(phen), 23 18 0.49 Ru-(5-Cl-phen), 24 16 0.52 Ru-(5-N02-phen), 25 18 0.47 4-N02-Im b 0.07 Ru represents RuCl2(dmso)2. k 107o inhibition not achieved. 133 (Figure 33). This suggests that targeting of the ligand via coordination to Ru (within 5) is occurring and would explain why 5 is a better sensitizer than the free 4-NC>2-Im ligand. However, the analogous tmso complex 15 which has aqueous solution behaviour and SER data similar to those of 5, did not show inhibition of activity of restriction enzymes up to the limit of solubility (500 jiM) under the same experimental conditions. This suggests that DNA binding is not the factor giving rise to the improved sensitization for this system. More data are required before any definitive conclusions can be deduced. Within the dichlorobis(dimethyl sulphoxide)-4-nitroimidazole complexes, only 5 binds to DNA, the N-substituted ones do not (i.e. inhibit restriction enzyme activity). One explanation is the importance of the presence of the pyrrolic proton of 4-N02-Im, an unsubstituted N-H(l), in enhancing the binding, as observed in the PtCl2(NH3)(4-N02Im) series.76 Hydrogen bonding by the N -H to the phosphate group of the DNA backbone was suggested by van Kralinger et al.138 and was later shown for some Pt(II) systems.-5-2 The lack of DNA binding may also result from the stability of the complexes in solution; in the case of 5, aquation to the neutral hydroxy species may generate an active reagent as demonstrated for the c/s-DDP systems.5^ Further, none of the Ru complexes inhibited EcoRI in this study, a behaviour unlike that of the m-DDP and PtC^NH^XL), L=miso or metro, compounds which also inhibit 7 ?2 the activity of this restriction enzyme. The bromide complex 21, RuBr2(dmso)2(4-N02-Im)2, does not show any inhibition of enzyme activity. When 21 is dissolved in aqueous media, no acidic proton is detected after dissociation of a bromide ion, and the charged species 21b. RuBr(H20)(dmso)2(4-N02-Im)2+ is formed (see section 4.3). This might be expected to bind to electron-rich DNA, as shown for 134 [Ru(NH 3) 5H 2Or + by Clarke et alP* The non-inhibition of activity of restriction enzymes perhaps results from the way 21 binds to DNA. For the /raws-bromide complex, binding may not be greatly distorting the DNA structure, or may be weak,5'* and therefore the restriction enzymes could still cut the plasmid DNA. General conclusions from a comparison of binding ability of the Ru complexes with enhancement ratio from radiosensitizing experiments cannot be drawn, because other factors may be involved. Complex 5 binds to DNA but 15 and 18 do not according to our assay, even though they all give good SER values. Platinum complexes, such as cis and trans-DDP, radiosensitize even without attachment of a radiosensitizer ligand, 5 7 by mechanisms not understood though presumably related to DNA binding. Conversely, electron-affinic compounds such as nitroimidazoles sensitize without binding to DNA. However, for a series of P t C ^ N ^ X L ) vs. PtCl 2(L) 2 , L=miso, metro, or 4-N0 2-Im, the complex which binds to DNA is the better sensitizer76 a behaviour pattern that 5 appears to follow. Thus, it appears that metals may be used in some cases to target sensitizers to DNA to improve efficacy by their proximity to the target. 6.2 Partition Coefficient (P) In general, the lipophilicity of pharmacological agents (usually measured by their octanolrwater partition coefficients) has been found to have a major influence on their activity, distribution and toxicity.75^'75'* For nitroimidazoles, one factor considered was that their neurotoxicity would be related to their uptake into nervous tissue. Because the amounts of drugs entering neural tissues depend on their ability to cross the blood-brain barriers, this being related to the drugs' lipid solubility, it is 135 expected that drugs of lipophilicity lower than that of miso (P=0.43) would penetrate the barriers less efficiently than miso, and hence be less neurotoxic. Further, the influence of the octanolwater partition coefficient, P, on the radiosensitizing efficiency of a large number of nitroaromatic and nitroheterocyclic compounds has been investigated in vitro. Although Adams et al.33 showed that P values over a wide range (0.05-240) were relatively unimportant in determining the radiosensitizing efficiency in mammalian cells, Anderson et alJ^ demonstrated some effect of variation in P using E. coli. In addition, Brown et al.156 measured intracellular uptake of neutral 2-nitroimidazoles with P values in the range 0.01 - 0.43 and found that the rate and extent of uptake correlated with radiosensitization: that is the compound with a higher P value has higher radiosensitizing efficiency. Butler et J 57 al. also found that the partition coefficient was the most important factor determining the activity range of 4- and 5-nitroimidazoles administered orally to mice against an infection of Trichomonas foetus. Thus, in the expectation that the degree of lipophilicity influences the distribution and intracellular concentration of the Ru radiosensitizers, the partition coefficients of the compounds synthesized were examined, and the results are summarized in Table XVII. Note that at the equilibrium measurement for complexes 5, 1 5 and 2 1 , the hydroxy species 5c and 15c. and the mono aquo species 21b are present in solution (see sections 4.2 and 4.3). The tmso complexes have higher P values than the corresponding dmso analogues. Except for the metro ligand and complex 14, 15c has the highest P value of all the other compounds studied here, and it also has a higher SER value at the same concentration. This observed result for 15c agrees with those previously reported by Brown et al.J49 whereby the more hydrophilic compounds in a series of 2- and 4-nitroimidazoles required higher drug concentration to achieve equal radiosensitization 136 Table XVTI Partition coefficients of nitroimidazole ligands and their ruthenium complexes Partition Coefficient, P [ ]Octanol/[ ]Aqueous N02-Im Ligand Complex 5, RuCl 2 [dmso)2(4-N02~Im)2 0.24 (1.2) a 0.42b (1.6) 6, RuCl 2 [dmso)2(NMe-4-N02-Im) 0.16 (1.2) 0.18 (1.3) 7, RuCl 2 (dmso)2(RSU-1170)2 0.22 (1.1) 0.28 (1.3) 8, RuCl 2 (dmso)2(RSU-3083)2 0.38 (1.2) 0.40 (1.2) 9, RuCl 2 (dmso)2(RSU-3100)2 0.33 (1.2) 0.38 (1.1) 10, RuCl 2 (dmso)2(RSU-3159) 0.26 (1.3) 0.29 (1.3) 11, RuCl 2 (dmso)2(miso)2 0.43 (1.3) 0.12° (1.4) 12, RuCl 2 (dmso)2(De-miso)2 0.13 (1.3) 0.04° (1.4) 14, RuCl 2 (dmso)2(metro)2 0.96 (1.1) 0.77 (1.2) 15, RuCl 2 (tmso)2(4-N02-Im)2 0.24 (1.2) 0.64b (1.6) 16, RuCl 2 (tmso)2(NMe-4-N02-Im)2 0.16 (1.2) 0.20 (1.4) 17, RuCl 2 (tmso)2(De-miso)2 0.13 (1.3) 0.22 (1.4) 18, RuCl 2 (tmso)2(SR-2508) 0.046(1.3) 0.21 (1.5) 21, RuBr2 (dmso)2(4-N02~Im)2 0.24 (1.2) 0.07b (1.3) 22, RuBr2 (dmso)2(NMe-4-N02-Im) 0.16 (1.2) 0.14 (1.3) a Number i n parentheses gives the corresponding SER value from Tables XIV and XV. b These values refer to species 5c, RuCKOH) (dmso)9(4-N09-Im)9, 15c, RuCl (OH) (tmso) 2 (4-N02-Im) 2, ancT~19b, [ RuBr (r^O) (dmso) ^ ^NO^ImTJ ] . c These values were taken i n 4 min after mixing, because of dissociation of nitroimidazole ligands; a l l other values were recorded at equilibrium conditions (± 0.05). 137 level . F o r the other compounds studied, there is no correlat ion between S E R and P values. The free metro l igand has the highest P value (0.96) but only a poor S E R value of 1.1, whereas SR-2508 has the smallest P value but an S E R value of 1.3 s imi lar to that observed for miso (P=0.43). Complexes 5 and 15. have simi lar S E R values (approximately 1.6), but the P values for 5c and ! 5 c are quite d i f ferent (0.42 vs. 0.64, respectively). Thus , the data are in agreement wi th Adams et al33'34 who suggested that the part i t ion coef f ic ient is not the dominant property af fect ing the abi l i ty o f some nitroimidazoles to act as hypoxic cel l sensitizers (see also section 1.8); these workers consider electron a f f in i ty to be the key factor (see section 6.4) O f the 4 - N 0 2 - I m complexes, 2J. has the lowest P value (0.07) wh ich is almost certainly due to the charged species 21b being formed (equation [4.4]) making 19 apparently more hydroph i l i c . Complexes H . and 12. have low P values (data measured in the 600 nm region) presumably associated wi th the dissociation of ni t roimidazole l igands. 149 Brown et al. suggested that a decrease in radiosensit izing ef f ic iency and cytotoxic i ty of compounds wi th P values below 0.04 correlated wi th a decreased abi l i ty of the compounds to enter the cells. In this study, no compound has a P value of less than 0.04. 6.3 Non -P ro te i n T h i o l Deplet ion The su lphydry l containing compounds have been shown to protect cells f rom i r radiat ion (section 1.9). M a n y metals b ind strongly to various sulphur donor ligands and therefore might be expected to radiosensitize by deplet ing intracel lular th io ls ,** ' * " 5 such as glutathione, wh ich is known to protect organisms against radiat ion (e.g. by H atom donation to repair chemical ly the damaged t a r g e t / " 5 9 see section 1.9). Th io l deplet ion has been proposed to explain c / s - D D P s e n s i t i z a t i o n / ' 5 9 but this does not 138 appear to be generally accepted. However, depletion of non-protein thiols (NPSH) by Rh(II) carboxylates correlated with in vitro radiosensitizing ability. Nitroimidazoles, such as misonidazole,*6** have been shown to deplete thiols. It has been established that various ortho-substituted-4-nitroimidazoles are much more efficient radiosensitizers than would be predicted from their electron affinity.59 It has been shown that these compounds react readily with thiols,- 5 9 ' 7 6 0 ' 7 ' 5 7 , 7 ' 5 7 which reportedly contributes to their abnormally high sensitizing efficiency. In the present work, cells were analyzed for NPSH after 1 h of incubation at 37°C in oxic and hypoxic conditions with test compounds. Reactions of the complexes or their free ligands with NPSH were monitored by the extent of formation of an adduct ( A m a x 412 nm) between the SH content and Ellman's reagent. Of all the Ru(II) complexes described in this thesis, only complexsJJ. and 12 show some NPSH depletion but this must result partly from the dissociated miso and De-miso ligands, respectively, because the precursor complex m-RuC^dmso^ does not deplete NPSH. Of the ligands, the 2-.nitroimidazoles (2-N02-Im, miso, De-miso and SR-2508) and 4-N0 2-Im depleted thiol levels, while the N-substituted-4-nitroimidazoles and metro did not (Table XVIII). The results also show that thiol depletion occurs only under hypoxic and not aerobic conditions. There have been reports that hypoxic tissues lose intracellular free thiols by membrane leakage into surrounding fluid.76-* However in control experiments, cells incubated under the experimental conditions did not exhibit thiol loss even after 2 h of hypoxia. In addition, several other compounds such as ortho-substituted-4- or 5-nitroimidazoles are reported to show increased sensitizing efficiency with increasing contact time with cells. 7 6 0 For the longer contact time (~2 h), cellular thiols decreased by one third relative to control values (-30 min). 7 6 0 In the present study, a 139 Table XVTII NPSH depletion i n hypoxia by some Ru(II) complexes3' 7o NPSH depletion Complex0 N0?-Im free ligand (400 pM) Ru complex (200 MM) 5, Ru(4-N02-Im)2 23 0 6, Ru(NMe-4-N02-Im) 0 0 7, Ru(RSU-1170)2 6 1 8, Ru(RSU-3083)2 5 1 9, Ru(RSU-3100)2 4 1 10, Ru(RSU-3159) 11 2 11, Ru(miso)2 15 25 12, Ru(De-miso)2 22 20 13, Rj(2-N02-Im)2 38 0 14, Ru(metro)2 0 0 15, Ru*(4-N02-Im)2 23 2 16, Ru'(NMe-4-N02-Im)2 0 0 17, Ru'(De-miso)2 22 3 18, Ru'(SR-2508) 23 1 21, Ra"(4-N02-Im)2 23 0 22, Ru"(NMe-4-N02-Im) 0 0 No depletion of NPSH was observed under oxic conditions; estimated maximum error Jl 5"L. k No depletion of NPSH was observed for the 1,10-phenanthrolines or their complexes. Ru, Ru', and Ru" represent RuCl2(dmso)2, RuCl 2(tmso) 2, and RuBr2(dmso) respectively. 140 one hour incubation time was chosen to be relevant to the radiosensitizing experiments. The compound, miso, which is known to deplete NPSH,** gave no further NPSH depletion upon longer incubation times. This result indicated that thiol depletion (to 15 %) which had occurred within the 1 h contact period at 37°C was close to maximum. Stratford et alJ^4 have shown that for some 2- and 4-nitroimidazole systems there is a dependence of thiol reactivity on reduction potential, namely a trend showing an increase in thiol reactivity with increase in reduction potential. However, in the present study, this trend is not followed in that none of the Ru(II) complexes, except IJ. and J_2, react with NPSH, although their reduction potentials (see next section) are higher than those of the free ligands. Enhancement of radiosensitization by metallation of the nitroimidazole ligands to Ru does not result from depletion of NPSH. 6.4 Electron Affinities (Reduction Potentials) Various in vitro studies with mammalian cells have shown that hypoxic cytotoxicity correlates to some extent with electron affinity of the compounds.^"'6* The relevance of E j y 2 or the one-electron reduction potential to toxicity by nitroimidazoles has often been discussed.^" 5*" 5 9 , 1 6 2- 1 6 4 The half-wave reduction potentials of the nitroimidazoles and their complexes as measured by polarography (see e.g. Figure 34) are shown in Table XIX. These reduction potential values are in good agreement with values obtained by cyclic voltammetry; the differences between the two values were within +10 mV, except in the case of complex J_0. The compounds are divided into different groups depending on the substituents of the complexes and the positions of the nitro group. Because free nitroimidazoles showed almost a hundred times greater toxicity toward hypoxic cell than oxic cells following a one-141 -685 -400 -600 ' -800 mV Figure 34; The reduction p o t e n t i a l s of RuCl 2(dmso)^ (4-NO -Im) 5, and 4-NO -Im 142 Table XIX Half-wave reduction potentials of Ru(II) compounds and their nitroimidazole ligands Complex5 E1 (mV)a NG^ -Im free ligands Ru complexes 5, Ra(4-N02-Im)2 -685 [-690f (1.2) d -615e [-620]° (1.6f 6, Ru(NMe-4-N02-Im) -535 [-545] (1.2) -518 [-525] (1.3) 7, Ru(RSU-1170)2 -410 [-420] (1.1) -390 [-400] (1.3) 8, Ru(RSU-3083)2 -560 [-550] (1.2) -540 [-535] (1.2) 9, Ru(RSU-3100)2 -470 [-460] (1.2) -455 [-450] (1.1) 10, Ru(RSU-3159) -370 [-360] (1.3) -360 [-345] (1.3) 11, Ru'(miso)2 -445 (1.3) -435 (1.4) 12, Ru(De-miso)2 -389 (1.3) -367 (1.4) 13, Ru(2-N02-Im)2 -400 (1.3) -385e (1.4) 14, Ru(metro)2 -520 (1.1) -490 (1.2) 15, Ru'(4-N02-Im)2 -685 (1.2) -605 (1.6) 16, RJ'(NMe-4-N02-Im)2 -535 (1.2) -500 (1.4) 20, Ru'(RSU-3159) -370 (1.3) -345 (1.3) 17, Ru'(De-miso)2 -389 (1.3) -355 (1.4) 18, Ru'(SR-2508) -388 (1.3) -345 (1.5) 21, Ifo"(4-N02-Im)2 -685 (1.2) -645 (1.3) 22, Ru"(NMe-4-N02-Im) -535 (1.2) -515 (1.3) continued ... continued 143 Values obtained i n PBS w.r.t. SCE; estimated maximum error ± 5 mV. Ru, Ru', and Ru" represent RuCl^ldmso^j RuCL-^tmso^, and RuBr (dmso^j respectively, complexes 5, 15 and 21 are the hydroxy and mono aquo species (see sections 4/2" and" 4.3) The number i n "square brackets" gives Ej. values obtained from cyclic voltammetry. 2 Number i n parentheses gives corresponding SER value. The 4-N0.2-Im and 2-N02-Im complexes give additional Ej. values at -355 mV and -290 mV, respectively, i n the polarographic measurements, but these values were not observed i n cyclic voltammetric measurements. 144 electron reduction process,"^"*6'7*'"' one assumes that the enhanced hypoxic toxicity in the complexes also results from the reduction of the nitroimidazole within the complex. An electron is expected to reduce the organic moiety rather than the metal in these Ru complexes, as observed for the c/s-PtCl2(metro)2,5* and PtCl2(NH3)(4(5)-N0 2-Im) systems. The measured half-wave potentials (Ejy2 a t pH 7) for the free nitroimidazole ligands are in good agreement with the one-electron reduction 3 1 3 1 C P CQ potentials for these ligands as reported previously. The attachment of a nitroimidazole to the Ru moiety increases the reduction potential with respect to that of the free ligand; that is, the ligand centres are more easily reduced when coordinated to Ru. This observation is also known to occur upon coordination of nitroimidazoles to platinum(II). ' The toxicity of the Ru complexes (Tables X and XI) toward CHO cells in vitro, compared to that of the respective nitroimidazole ligand, does not follow the trend of increasing toxicity with increasing electron affinity, although such a trend has been observed for unbound nitroheterocyclic compounds."*'*'7'5* In the present work, the ligands are always more toxic than their corresponding complexes which are more electron affinic. This emphasizes that other factors must also be considered. Studies with vegetative bacteria, bacterial spores, and mammalian cells have shown that there is a general relationship between the efficiency of radiosensitization and the electron affinity of the compounds, and this has aided considerably the search for new compounds.3^'32 The general relationship between sensitizing efficiency and electron affinity has been supported by studies on one-electron transfer reactions between known sensitizers, such as the 2-nitroimidazoles, 5-nitrofurans, nitrobenzenes and 4-nitroimidazoles,^'-^"** and the results have suggested that the sensitizing efficiency increases with increasing electron affinity or reduction potential. ' Thus, comparison of the reduction potentials (at pH 7.0 in aqueous medium) for 145 accepting a single electron shows that oxygen (the "best" radiosensitizer, SER=3.0) is more electron affinic than the nitroimidazoles: 0 2 + e" —»• 0 2", = "155 mV* 2 L + e" — * L ~ , E 1 / 2 = -370 to -685 mV In the present study, complexes 5 and 15 have the best SER value (approximately 1.6) and, indeed, the radiosensitizing ability of the 4-nitroimidazole ligands increases with an increase in the reduction potential when the ligand is bound to Ru. For 5 and 1_5, this increase in potential upon metallation is approximately 70 mV which is the largest "coordination shift" among all the systems studied. Complex 18 also shows an increase in the SER value compared to the value for the nitroimidazole free ligand; the increase in reduction potential is -40 mV. The SER values are 1.3 and 1.5 at 200^X1 for SR-2508 ligand and its complex, 18, respectively. The SER values for the other Ru nitroimidazole complexes do not show significant enhancement over those of the free ligands alone at an equivalent molar concentration. This may be a consequence of the relatively small increase in one-electron reduction potential between the free and coordinated ligand, the "coordination shift" being <30 mV (Table XIX). Electron-donor substituents on the nitroimidazole (e.g. the N-CH3, and the amino and thioether substituents of 8 and 10, respectively) lower the reduction potentials. A higher reduction potential will certainly favour thermodynamically abstraction of an electron from irradiated DNA. Surprisingly, 5, and 15 have the lowest electronic affinities ( E j / 2 —610 mV) of all the complexes studied (except one, 21) and yet have the highest SER values. High SER values with low electronic affinities have been realized also within a series of ortho-substituted-4-nitroimidazoles,"^ but thiol depletion was important in these systems (section 6.3), and is not so in the metal systems. Clearly, factors other than "absolute" electron affinities (higher reduction potentials) also contribute to radiosensitization. Data on the Ru 146 systems would tend to indicate increased radiosensitizing property with increasing coordination shift of the ligand reduction potential. 6.5 Feasibility in IN VIVO System The in vitro results in the present study have led to a testing of the radiosensitizing efficiency of several of the Ru-nitroimidazole complexes in experimental animal tumours; the experimental work was performed by Ms. D. Ruta under the guidance of Dr. D. Chaplin at the B. C. Cancer Research Centre. The complex RuCl2(tmso)2(SR-2508) is of particular interest, with a relatively high SER value in vitro (1.5 at 200 uM) in view of the presence of only one coordinated nitroimidazole. Tumour cell survival was determined after a single X-ray dose of 10 Gy delivered to the mice at various times following intraperitoneal injection of RuCl2(tmso)2(SR-2508), 18, at 300 mg/kg tumour. The results (Figure 35) show that the maximum sensitization occurred when complex 18 was administered ~1 h before the start of the irradiation. The interval between drug administration and irradiation, that is required for maximum radiosensitization, is defined as the optimum time. The cell survival at this optimum time with J_8 and X-ray irradiation was 0.04%, whereas the irradiation alone gave a survival of 0.104%. Further, complex 18 alone did not show any toxicity to the mice at 300 mg/kg. The degree of radiosensitization achieved by a drug may therefore depend on the time-elapsed since dosing. The importance of the time interval between drug administration and irradiation for hypoxic cell radiosensitization has been shown by several authors 7 6 5 ' 7 6 6 ' 7 6 7 who showed that there is an optimum time to irradiate after drug administration in order to achieve maximum radiosensitization. If irradiation 147 Hours Before I r r a d i a t i o n (h) Figure 35: The e f f e c t on tumour c e l l s s u r v i v a l on the time i n t e r v a l between RuCl 2(tmso) (SR-2508) (300 mg/kg tumour) administration and i r r a d i a t i o n with 10 Gy of X-ray 148 follows too early there may be inadequate time for diffusion of the drug to the hypoxic cells; conversely, if the interval is too long the drug concentration may drop as a result of excretion and/or metabolism. Further studies will include the use of higher dose of 18 and dose response survival as compared to the free SR-2508 ligand. In addition, the other 4-N0 2-Im complexes, 5 and 15, which show high SER values in vitro will be examined. 149 CHAPTER SEVEN CONCLUSIONS and RECOMMENDATIONS for FUTURE WORK Metal complexes can alter the effects of radiation within radiotherapy in many ways. Several studies suggest that the combination of metal complexes and radiation 70 may radiosensitize by: (a) enhancement of initial DNA damage, (b) inhibition of recovery from radiation damage, (c) interaction with thiols, and (d) properties that 71 7 ? depend on electron affinity. ' Therefore, a single metal complex could radiosensitize by several mechanisms. This work was carried out to improve the radiosensitizing ability and reduce the toxicity of nitroimidazoles (currently used clinically), with a view to gaining an understanding of the effects of transition metal (Ru)/nitroimidazole complexes on cellular radiation response. The data in the present work show that some of the Ru-complexes have greater sensitizing efficiency, lower toxicity, and a lower reaction rate with NPSH, than the corresponding free nitroimidazole ligand. One of the better complexes synthesized, RuCl2(dmso)2(4-N02-Im)2, 5, also binds to DNA as assessed by inhibition of restriction enzyme activity. At present, there is no definite explanation why 15 (in contrast to 5) does not bind to DNA but steric effects of tmso vs. dmso may play a role. The data show that complexes 5 and RuCl2(tmso)2(4-N02-Im)2, i l , are better radiosensitizers than the free 4-N0 2-Im, and than a range of other Ru/nitroimidazole/sulphoxide/halide complexes. When examined at equivalent concentrations, they are also better in vitro than clinically used misonidazole. Replacement of the dmso ligand generally by tmso does not markedly affect the redox properties and radiosensitizing ability of related complexes. However, such replacement does affect the stability in aqueous solution of the substituted-2-N02"Im complexes. The substitution of Cl~ by Br" (5 vs. 21) reduces the radiosensitizing ability 150 of the Ru-4-nitroimidazole complexes, but this may result from the presence of different forms of geometrical isomers: the all cis geometry for 5 and the trans-bromide, c/s-sulphoxide, ds-nitroimidazole for 21_; the less electronegative Br" causes a smaller change in "coordination shift" reduction potential and different behaviour in aqueous solution chemistries (equations [4.2] and [4.4]). Complex 5 gives a hydroxy species, RuCl(OH)(dmso)2(4-N02-Im)2, while 21 gives a mono aquo species, [RuBr(H20)(dmso)2(4-N02-Im)2]+. The better radiosensitizing ability of 5 and 15 compared to other complexes is likely to depend on some or all of the following factors: (a) increased reduction potential of the 4-NC>2-Im ligand when it becomes coordinated, (b) hydration to a species capable of binding to DNA, (c) better lipophilicity, (d) the presence of the pyrrolic proton of imidazole which could be involved in H-bonding that stabilizes a DNA-Ru interaction - as demonstrated for Pt(II) systems.5'2'7-** Thiol depletion is probably not involved. As regards to (a), a higher reduction potential will certainly favour thermodynamically abstraction of an electron from irradiation-damaged DNA; electron-donor substituents on the nitroimidazole (e.g. the N - C H 3 , and the amino and thioether substituents of 8 and 10, respectively) lower the reduction potentials. Enhancement ratios produced by metal complexes as radiosensitizers have often been small using cells in tissue culture.60 However, a relatively low effect still results in therapeutic gain in the clinic, as seen for c/s-DDP. 7 6 9 It is clearly not possible to assess the full potential of metal-based radiosensitizers at this stage. Experimental investigations must be pursued if the interactions between radiation, metal complexes, and cells are to be optimized to give cell inactivation. There is a need to gain a more complete basic understanding of the mechanisms of radiation sensitization so that guidelines can be established for designing and selecting the most effective metal 151 complexes. For example, the effects of systematic variation of the ligands on a known radiosensitizing metal complex, or within a given ligand system, variations of the metal on radiosensitizing ability would be instructive. In addition, studies on the mechanisms and factors involved, such as electron affinity, thiol depletion, lipophilicity, DNA binding and repair, must be included. Further studies using Ru(III) as a metal centre should be included to determine if these complexes radiosensitize and have different reduction process(es) as compared with the Ru(II) systems. The metal-based radiosensitizers must continue to be actively tested in tissue culture and animal tumor systems that are relevant to the selection of metal complexes for application in the radiation therapy of clinical cancer. The individual responses of patients to therapeutic agents are invaluable in the evaluation of cancer treatments. Ideally, for the clinical situation, one would like to find a non-toxic radiosensitizer. The beneficial effects of sensitizing agents must be balanced against detrimental toxic effects. The data in the present study encourage further investigations of the use of transition metal complexes as radiosensitizers to combat the hypoxic tumour cells. 152 BIBLIOGRAPHY 1. B.H.J. Bielski and J.M. Gebicki, in: "Free Radicals in Biology", W.A. Pryor (Ed.), Academic Press, New York, Vol. Ill, pp. 2-48 (1977). 2. K.H. Chadwick and H.P. Leenhouts, "The Molecular Theory of Radiation Biology", Springer-Verlag, New York, p. 1 (1981). 3. M.M. Elkind and G.F. Whitmore, "The Radiobiology of Cultured Mammalian Cells", Gordon and Breach, New York (1963). 4. T.T. Puck and P.L. Marcus, J. Exp. Med.. 103. 653 (1956). 5. T. Alper, "Cellular Radiobiology", Cambridge University Press, Cambridge (1979). 6. E. Petry, Biochem. Z.. 135. 353 (1923). 7. B. Palcic, J. Brosing and L.D. Skarsgard, Br. J. Cancer. 46, 980 (1982). 8. J.D. Chapman and C.J. Gillespie, Radiat. Biol.. 9, 143 (1981). 9. Z.M. Bacq and P. Alexander, "Fundamentals of Radiobiology", 2nd ed. Pergamon Press, Oxford (1961). 10. J. Weiss, Nature. 153. 748 (1944). 10a. C. von Sonntag, "The Chemical Basis of Radiation Biology", Taylor and Francis Ltd., London (1987). 11. L.H. Gray, Radiat. Res.. 1, 189 (1954). 12. I. Fridovich, in: "Free Radicals in Biology", W.A. Pryor (Ed.), Academic Press, New York. Vol. I, pp. 241-247 (1976). 13. T. Alper, Radiat. Res.. 5, 573 (1956). 14. G.E. Adams, in: "Radiation Protection and Sensitization", H. Moroson and M.L. Quintiliani (Eds.), Taylor and Francis, London, pp. 1-14 (1970). 15. E.J. Hall, in: "Cancer a Comprehensive Treatise", F.Becker (Ed.), Plenum, New York, Vol. 6, pp. 281-312 (1977). 16. L.H. Gray, A.D. Conger, M. Ebert, S. Hornsey and O.C.A. Scott, Brit. J.  Radiol.. 26, (1953). 17. N.P. Farrell, "Transition Metal as Drugs and Chemotherapeutic Agents in Catalysis by Metal Complexes", B.R. James and R. Ugo (Eds.), Reidel-Kluwer, Dordrecht (in press). 18. R.H. Thomlinson, Europ. J. Cancer. 7, 139 (1971). 153 19. R.S. Bush, R.D.T. Jenkin, W.E.C. Alet, F.A. Beale, H. Bean, A.J. Dembo and J.F. Pringle, Br. J. Cancer. 3J7 (Suppl. Ill), 302 (1978). 20. D.J. Chaplin, P.L. Olive and R.E. Durand, Cancer Research. 47, 597 (1987). 21. G.W. Barendsen, C.J. Koot, G.R. van Kersen, D.K. Bewley, S.B. Field and C.J. Parnell, Int. J. Radiat. Biol.. 10, 317 (1966). 22. G.E. Adams and D.L. Dewey, Biochem. Biophvs. Res. Comm.. 12, 473 (1963). 23. G.E. Adams and M.S. Cooke, Int. J. Radiat. Biol., i l , 457 (1969). 24. G.E. Adams and B.D. Michael, in: "Energetics and Mechanisms in Radiation biology", G.O. Phillips (Ed.), Academic Press, New York, p. 333 (1968). 25. G.E. Adams, C.L. Greenstock, J.J. van Hemmen and R.L. Wilson, Radiat. Res.. 42, 85 (1972). 26. L. Parker, L.D. Skarsgard and P. Emmerson, Radiat. Res.. 38, 493 (1969). 27. G.E. Adams, J.C. Asquith, D.L. Dewey, J.L. Foster, B.D. Michael and R.L. Wilson, Int. J. Radiat. Biol.. 19, 575 (1971). 28. "Advances in Chemical Radiosensitization: Proceedings of a Panel", Stockholm, June 25-29, 1973, International Atomic Energy Agency, Vienna (1974). 29. J.L. Foster and R.L. Willson, Br. J. Radiol.. 46, 234 (1973). 30. A.P. Reuvers, J.D. Chapman and J. Borsa, Nature. 237. 402 (1972). 31. T. Saski, Pharm. Bull.. 2, 104 (1954). 32. J.D. Chapman, A.P. Reuvers, J. Borsa, J.S. Henderson and R.D. Migliora, Cancer Chemother. Reports. 5.8, 559 (1974). 33. G.E. Adams, E.D. Clarke, I.R. Flockhart, R.S. Jacobs, D.S. Sehmi, I.J. Stratford, P. Wardman, M.E. Watts, J. Parrick, R.G. Wallace and C.E. Smithen, Int. J.  Radiat. Biol.. 35, 133 (1979). 34. G.E. Adams, I.R. Flockhart, C.E. Smithen, I.J. Stratford, P. Wardman and M.E. Watts, Radiat. Res.. 67, 9 (1976). 35. R.C. Urtasun, P. Band and J.D. Chapman, Radiat. Res.. 70, 704 (1977). 36. B.A. Moore, B. Palcic and L.D. Skarsgard, Radiat. Res.. 67, 459 (1976). 37. M. Korbelik, B. Palcic and L.D. Skarsgard, Radiat. Res.. 8J5, 343 (1981). 38. P. Wardman, Curr. Topics in Radiat. Res. Quart.. J l , 347 (1977). 39. R.A. McClelland, R. Panicucci and A.M. Rauth, J. Am. Chem. Soc. 109, 4308 (1987). 40. J.F. Fouts and B.B. Brodie, J. Pharm. Exo. Ther.. 1J9, 197 (1957). 41. R.P. Mason and J.L. Holtzman, Biochem.. 14, 1626 (1975). 154 42. P. Alexander and A. Gharlesby, in: "Physico-chemical Methods of Protection Against Ionizing Radiation. Radiobiology Symposium", Z.M. Bacq and P. Alexander (Eds.), Butterworths Scientific Publication, London, pp. 49-60 (1954). 43. H. Loman, S. Voogd and J. Blok, Radiat. Res.. 42, 437 (1970). 44. B.A. Bridges, Nature. 188. 415 (1960). 45. B.A. Bridges and R.J. Munson, Int. J. Radiat. Biol.. 1 3 , 179 (1967). 46. O.W. Griffith, J. Biol. Chem.. 257. 13704 (1982). 47. D. Bahnemann, H. Basaga, J.R. Dunlop, A.J.F. Searle and R.L. Willson, Br. J.  Cancer. 37 (Suppl. Ill), 16 (1978). 48. E.J. Hall and J. Biaglow, Int. J. Radiat. Oncol. Biol. Phvs.. 2, 521 (1977). 49. B. Rosenberg, L. Van Camp, J.E. Trosko and V.H. Mansour, Nature. 222. 385 (1969). 50. "Cis-platin: Current Status and New Developments", A.W. Prestayko, S.T. Crooke and S.K. Carter (Eds.), Academic Press, New York (1980). 51. "Proceedings of the Third International Symposium on Platinum Coordination Complexes in Cancer Chemotherapy. J. Clin. Hemotol. Oncol.. 7 (1977). 52. S.E. Sherman, D. Gibson, A.H.J. Wang and S.J. Lippard, Science. 230. 412 (1985). 53. J.J. Roberts, in: "Metal Ions in Genetic Information Transfer", G.L. Gchhorn and L.G. Marzilli (Eds.), Elsevier Scientific Publishers Ireland Ltd., Ireland, Ch.10, pp.273 (1981). 54. R.B. Ciccarelli, M.J. Solomon, A. Varshavsky and S.J. Lippard, Biochemistry. 24, 7533 (1985). 55. L.A. Zwelling and K.W. Kohn, Cancer Treat. Rep.. 63, 1439 (1979) 56. R.C. Richmond and E.L. Powers, Radiat. Res.. 68, 251 (1976). 57. E.B. Douple and R.C. Richmond, Br. J. Cancer. 37 (Suppl. Ill), 98 (1978). 58. J. Butler, B.M. Hoey and A.J. Swallow, Radiat. Res.. 102. (1985). 59. I.J. Stratford, S. Hoe, G.E. Adams, C. Hardy and C. Williamson, Int. J. Radiat.  Biol.. 43, 31 (1983) 60. A.H.W. Nias, Int. J. Radiat. Biol.. 48, 297 (1985). 61. K. Lindquist, E.B. Douple and R.C. Richmond, Radiat. Res.. 9_L, 408 (1982). 62. E. Smith and W.D.B. Johnson, Int. J. Radiat. Oncol. Biol. Phvs.. 10, 1803 (1984). 155 63. I.P. Hesselwood, W.A. Cramp, D.C.H. McBrien, P. Williamson and K.A.K. Lott, Br. J. Cancer. 37 (Suppl. Ill), 95 (1978). 64. E.B. Douple, C.J. Green and M.G. Simic, Int. J. Radiat. Oncol. Biol. Phvs.. 6, 1545 (1980). 65. H. Moroson and D. Tenney, Experienta. 24, 1041 (1968). 66. R.C. Richmond, M. Simic and E.L. Powers, Radiat. Res.. 63, 140 (1975). 67. C H . Chang and C F . Meares, Biochemistry. 23, 2268 (1984). 68. J.K. Barton and A.L. Raphael, J. Am. Chem. Soc. 106. 2466 (1984). 69. K.A. Skov, H. Adomat and N.P. Farrell, in: "Proceedings of the Fifth International Symposium on Platinum and Other Metal Coordination Compounds in Cancer Chemotherapy", M. Nicolini (Ed.), Martinus Nijhoff Publishing, Boston, p. 733 (1988). 70. K.A. Skov, Radiat. Res.. 112. 217 (1987) and references therein. 71. R. Chibber, I.J. Stratford, P. O'Neill, P.W. Sheldon, I Ahmed and B. Lee, Int. J.  Radiat. Biol.. 48, 513 (1985). 72. J.R. Bales, P.J. Sadler, C J . Coulson, M. Laverick and A.H.W. Nias, Br. J.  Cancer. 46, 701 (1982). 73. N.P. Farrell and K.A. Skov, Radiat. Res.. 91., 378 (1982). 74. E. Smith, O.C.A. Scott, A.H.W. Nias and A.P. Brock, Br. J. Radiol.. 60, 601 (1987). 75. R. Chibber, I.J. Stratford, I. Ahmed, A.B. Robbins, D.M.L. Goodgame and B. Lee, Int. J. Radiat. Oncol. Biol. Phvs.. 10, 1213 (1984). 76. K .A. Skov, N.P. Farrell and H. Adomat, Radiat. Res.. 112. 273 (1987) and references therein. 77. K.A. Skov, L. MacDonald, D.J. Chaplin and N.P. Farrell, Presented at the 32nd Annual Scientific Meeting of the Radiation Research Society, Orlando, Florida, March 25-29 (1984). 78. K.A. Skov, and D.J. Chaplin, Personal Communication. 79. D.M.L. Goodgame, A.S. Lawrence, A.M.Z. Slawin, D.J. Williams and I.J. Stratford, Inore. Chim. Acta. 125. 143 (1986). 80. M.J. Clarke, in: "Inorganic Chemistry in Biology and Medicine", A.J. Martel (Ed.), ACS Symp. 140, p. 157 (1983). 81. M.J. Clarke, in: "Platinum, Gold and Other Metal Chemotherapeutic Agents", S.J. Lippard (Ed.), ACS. Symp. 209, p.335 (1983). .156 82. R.E. Yasbin, C R . Matthews and M.J. Clarke, Chem.-Biol. Interactions. 30, 355 (1980). 83. J.K. Barton and E. Lolis, J. Am. Chem. Soc. 107. 708 (1985). 84. M.J. Clarke, M. Buchbinder and A.D. Kelman, Inorg. Chim. Acta. 27, L27 (1978) . 85. R. Shepherd and H. Taube, Inorg. Chem.. J2, 1392 (1973). 86. M.J. Clarke, Met. Ions Bol. Svst.. J l , 231 (1980) and references therein. 87. J.A. Davies, Adv. Inorg. Chem. Radiochem.. 24, 115 (1981) and references therein. 88. W.L. Reynolds, Prog. Inorg. Chem.. 12, 1 (1970). 89. J.H. Price, H.N. Williamson, R.F. Schramm and B.B. Wayland, Inorg. Chem.. J l , 1280 (1972). 90. J.A. Davies, F.R. Hartley and S.G. Murray, J. Chem. Soc. Dalton Trans.. 1705 (1979) . 91. F.A. Cotton and T.R. Felthouse, Inorg. Chem.. 19, 323 (1980). 92. "Metal Complexes as Anticancer Agents", H. Sigel (Ed.), Vol. II, Marcel Dekker, New York (1980). 93. N.P. Farrell, J. Chem. Soc. Chem. Commun.. 331 (1982). 94. A. Mercer and J. Trotter, J. Chem. Soc. Dalton Trans.. 2480 (1975). 95. B.R. James, E. Ochiai and G.L. Rempel, Inorg. Nucl. Chem. Lett.. 7, 781 (1971). 96. I.P. Evans, A. Spencer and G. Wilkinson, J. Chem. Soc. Dalton Trans.. 204 (1973). 97. C. Monti-Bragadin, L. Ramani and L. Samer, Antimicr. Ag. Chemother.. 7, 825 (1975). 98. C. Monti-Bragadin, M. Tamaro and E. Banti, Chem.-Biol. Interactions. 11, 469 (1975). 99. T. Giraldi, G. Sava, B. Dertoli, G. Mestion and G. Zassinovich, Cancer Res.. 31, 2662 (1977). 100. R.F. Whiting and F.B. Ottensmeyer, Biochim. Bioohvs. Acta. 434, 334 (1977). 101. N.P. Farrell and N.G. de Oliveira, Inorg. Chim. Acta. 66, L61 (1982). 102. S.N. Gamage, B.R. James, S.J. Rettig and J. Trotter, Can. J. Chem.. 66, 1123 (1988). 157 103. "Platinum Coordination Complexes in Cancer Chemotherapy", M.P. Hacker, E.B. Douple and I.J. Krakoff (Eds.), Martinus Nijhoff Publishing, Boston (1984). 104. "Short-term Tests for Chemical Carcinogens", H.F. Stich and R.H.C. San (Eds.), Springer-Verlag, New York (1981). 105. M.P. Rosin and H.F. Stich, Int. J. Cancer. 23, 722 (1979). 106. H.F. Stich, P.K.L. Chan and M.P. Rosin, Int. J. Cancer. 30, 719 (1982). 107. M . Meselson and K. Russell, in: "Origins of Human Cancer", H.H. Hiatt, J.D. Watson and J.A. Winster (Eds.), Cold Spring Harbor Laboratory, New York, p. 1473 (1977). 108. E. Bocian, M. Laverick and A.M.W. Nias, Br. J. Cancer. 47, 503 (1983). 109. I. Pleskova, M. Blakso and J. Siracky, Neoolasma. 6, 655 (1984). 110. D. Turnbull, N.C. Popescu, J.A. Dipaolo and B.C. Myhr. Mutat. Res.. 6J>, 267 (1979). 111. E. Bocian, M. Laverick and A.H.W. Nias, Br. J. Cancer. 48, 803 (1983). 112. C. Monti-Bragadin, M. Giacca, L. Dolzani and M. Tamaro, Inorg. Chim. Acta. 137. 31 (1987). 113. B.R. James, R.S. McMillan, R.H. Morris and D.K.W. Wang, in: "Adv. in Chem. Series", R. Bau (Ed.), ACS Symp. 167, p. 122 (1978). 114. I.J. Stratford, G.E. Adams, C. Hardy, S. Hoe, P. O'Neill and P.W. Sheldon, Int.  J. Radiat. Biol.. 46, 731 (1984). 115. B.R. James and R.J.P. Williams, J. Chem. Soc. 2007 (1961). 116. R.G. Fargher and F.L. Pyman, J. Chem. Soc. Trans.. 115. 217 (1919). 117. R.J. Sundberg, R.F. Bryan, I.F. Taylor, Jr. and H. Taube, J. Am. Chem. Soc. 96, 381 (1974). 118. L.H. Thompson, in: "Methods in Enzymology, Vol. LVIIP. Cell Culture." W.B. Jakoby and LH. Pastan (Eds.), Academic Press, New York (1979). 119. H.F. Stich, L. Wei and R.F. Whiting, Cancer Res.. 39, 4145 (1979). 120. B.N. Ames, J. McCann and E. Yamasaki, Mutat. Res.. 31., 347 (1975). 121. F. Tietze, Anal. Biochem.. 27, 501 (1969). 122. K.A. Skov, H. Adomat, D.C. Konway and N.P. Farrell, Chem.-Biol.  Interactions. 62, 117 (1987). 123. H.C. Birnboim and J. Doly, Nucleic Acids Res.. 7, 1513 (1979). 124. T. Fujita, J. Iwasa and C. Hansch, J. Am. Chem. Soc. 86, 5175 (1964). 158 125. J.D. Oliver and D.P. Riley, Inorg. Chem.. 23, 156 (1984). 126. T.G. Fawcett, E.E. Bernarducci, K. Krogh-Jespersen and H.J. Schugar, J. Am. Chem. Soc. 102. 2599 (1980). 127. C R . Johnson, R.E. Shepherd, B. Marr, S. O'Donnell and W. Dressick, J. Am.  Chem. Soc. 102. 6227 (1980). 128. T.A. Carlson, "Photoelectron and Auger Spectroscopy", Plenum Press, New York, p. 349 (1975). 129. R.G. Little, Acta Crvstallogr. Sec B.. 35, 2398 (1979). 130. R.S. Berman and J.K. Kochi, Inorg. Chem.. 19, 248 (1980). 131. K. Nakamoto, "Infrared Spectra of Inorganic and Coordination Compounds", 1st ed. J. Wiley and Sons, Inc., New York, p. 160 (1970). 132. B.R. James and R.H. Morris, Can. J. Chem.. 58, 399 (1980). 133. F.D. Rochon, P.C. Kong and L. Girard, Can. J. Chem.. 64, 1897 (1986). 134. D.L. Pavia, G.M. Lampman and G.S. Kriz Jr., "Introduction to Spectroscopy", Saunders College Publishing, Philadelphia, p. 167 (1979). 135. J.R. Barnes and R.J. Goodfellow, J. Chem. Res. Svnop.. 350 (1979). 136. J.R. Barnes, P.L. Goggin and R.J. Goodfellow, J. Chem. Res. Svnop.. 118 (1979). 137. G.B. Barlin and T.J. Batterham, J. Chem. Soc. (B). 516 (1967) 138. C.G. van Kralingen, J.K. de Ridder and J. Reedijk, Inorg. Chim. Acta. 36, 69 (1979) . 139. E. Laviron, Bull. Soc Chim.. 12, 2840 (1963). 140. A. Grimison, J.H. Ridd and B.V. Smith, J. Chem. Soc. 1352 (1960). 141. N.P. Farrell, S. Fonseca and K.A. Skov, (to be submitted to Inorg. Chem.). 142. W.J. Geary, Coord. Chem. Rev.. 7, 81 (1971). 143. Z. Harzion and G. Navon, Inorg. Chem.. 19, 2236 (1980) and references therein. 144. N.P. Johnson, J.D. Hoeschele and R.O. Rahn, Chem.-Biol. Interactions. 30, 151 (1980) . 145. A.M. Rauth, Int. J. Radiat. Oncol. Biol. Phvs.. 10, 1293 (1984). 146. L.W. Wattenberg, W.D. Loub, L.K. Lam and J.L. Speier, Fed. Proc. 35, 1327 (1976). 159 147. J.A. Miller and E.C. Miller, in: "Origins of Human Cancers", H.H. Hiatt, J.D. Watson and J.A. Winsten (Eds.), Cold Spring Harbor Laboratory, New York, p. 605 (1977). 148. I.E. Mattern, L. Cocchiarella, C.G. van Kralingen and P.H.M. Lohman, Mut.  Res.. 95., 79 (1982). 149. D.M. Brown, E. Parker and J.M. Brown, Radiat. Res.. 90, 98 (1982). 150. "Chemical Modifiers of Cancer Treatment", J.M. Brown (Ed.), Int. J. Radiat.  Oncol. Biol. Phvs.. 12, 1019 (1986) and references therein. 151. W.M. Scovell, L.R. Kroos and V.J. Capponi, in: "Metal Chemotherapeutic Agents", S.J. Lippard (Ed.), ACS Symp. 209, p. 101 (1983). 152. M.J. Clarke, B. Jansen, K.A. Marx and R. Kruger, Inorg. Chim. Acta. 124. 13 (1986). 153. "Absorption and Distribution of Drugs", T.B. Binns (Ed.), Livingstone Inc., London (1964). 154. S.H. Curry, "Drug Disposition and Pharmacokinetics", Blackwell, Oxford (1977). 155. R.F. Anderson, K.B. Patel and D.S. Sehmi, Radiat. Res.. 85, 496 (1981). 156. D.M. Brown, R. Gonzalez-Mendez and J.M. Brown, Radiat. Res.. 93, 492 (1983). 157. K. Butler, H.L. Howes, J.E. Lynch and D.K. Pirie, J. Med. Chem.. H), 891 (1967). 158. P.C. Jocelyn, "Biochemistry of the SH group", Academic Press, New York, Chapter 15, p. 323 (1972). 159. M.V. Alvarez, G. Cobreros, A. Heras and M.C. Lopez Zumel, Br. J. Cancer. 37 (Suppl. Ill), 68 (1978). 160. M. Astor, E.J. Hall, J. Martin, M. Flynn, J. Biaglow and J.C. Parham, Int. J.  Radiat. Oncol. Biol. Phvs.. 8, 409 (1982). 161. I.J. Stratford, G.E. Adams, C. Hardy, S. Hoe, P. O'Neill and P.W. Sheldon, Int.  J. Radiat. Biol.. 46, 731 (1984). 162. M.E. Watts, M.F. Dennis, M.R.L. Stratford and P. Wardman, Int. J. Radiat.  Oncol. Biol. Phvs.. J2, 1135 (1986). 163. C. Guarnieri, F. Flamigni and C. Rossoni-Caldarera, Biochem. Biophys. Res.  Commun.. 89, 678 (1979). 164. G.E. Adams, I.J. Stratford, R.G. Wallace, P. Wardman and M.E. Watts, J. Natl.  Cancer Inst.. 64, 555 (1980). 165. P.W. Sheldon and S.A. Hill, Br. J. Cancer. 36, 198 (1977). 160 166. N.J. McNally, J. Denekamp, P.W. Sheldon, I.R. Flockhart and F.A. Stewart, Radiat. Res.. 73, 568 (1978). 167. J.M. Brown and N.Y. Fu, Br. J. Radiol.. ^3, 915 (1980). 168. Y.L. Ho and S.K. Ho, Radiat. Res.. 6J_, 230 (1975). 169. C T . Coughlin and R.C. Richmond, Int. J. Radiat. Oncol. Biol. Phvs.. i i , 915 (1984). 170. D.J. Chaplin, Ph. D. Dissertation, University of London, 1982. 171. P.W. Sheldon and S.A. Hill, Br. J. Cancer. M , 795 (1977). 172. V.D. Courtenay, Br. J. Cancer. 34, 39 (1976). 161 APPENDIX I PREPARATION OF STOCK SOLUTIONS (a) Oi—medium One package of alpha M E M medium powder (Gibco, Catalogue number 410-2000), 4 vials (20 mL) of penicillin and streptomycin were mixed in 9 L of distilled water. Fetal-calf serum (1 L) and NaHCO-j (20 g) were added to the above mixture and stirred. The solution was sterilized by filtration through a 0.22 micron filter and was stored at 4°C. (b) PBS NaCl (160 g), KC1 (4 g), Na 2 HP0 4 (23 g) and K H 2 P 0 4 (4 g) were dissolved in distilled water (20 L). The solution was sterilized by filtration through a 0.22 micron filter and stored at 4°C. (c) Methylene-blue Methylene-blue (2 g) was dissolved in distilled water (1 L) at 37°C. The solution was allowed to stand for 1 h and filtered. (d) Ethidium bromide Ethidium bromide (1 g) was added to distilled water (100 mL). The solution was stirred for several hours until complete dissolution and was stored in a container wrapped in aluminum foil at 4°C. 162 APPENDIX II PREPARATION OF BUFFERS (a) Electrophoresis buffer (20X) Tris base (387.5 g), sodium acetate (54.4 g) and EDTA (29.8 g) were dissolved in distilled water (3.8 L). The pH of the mixture was adjusted to 7.8 with approximately 125 mL glacial acetic acid and made up to 4 L with distilled water. (b) Loading buffer (6X) Bromophenol blue 0.25% (W/W), xylene cyanol 0.25% (W/W) and glycerol 30% (W/W) were mixed in distilled water (100 mL) and the solution was stored at 4°C. (c) Tris-EDTA (TE) buffer EDTA (1 mM) was mixed with Tris-chloride solution (10 mM) at a 1:1 ratio and the pH was adjusted to 8.0. (d) Tris-acetate (TA) buffer (10X1 The pH of the 10X TA buffer which consisted of: Tris-acetate (0.33 M), potassium acetate (0.66 M), magnesium acetate (0.10 M) and dithiothreitol (0.005 M), was adjusted to 8.0 and this solution was stored frozen in aliquots at -20°C. 163 APPENDIX III CHROMATOGRAPHY THROUGH SEPHADEX G-50 Sephadex G-50 was used to separate high-molecular weight molecules (e.g. DNA) from smaller molecules (such as salts, chemicals) by gel filtration. A. PREPARATION OF SEPHADEX G-50 Sephadex G-50 (Pharmacia, 30 g) was added slowly to TE (250 mL) in a 500 mL beaker. The solution was allowed to stand overnight at room temperature, and the supernatant was decanted and replaced with an equal volume of TE. The mixture was stored at 4°C in a screw-capped bottle. B. SPIN COLUMN PROCEDURE 1. A quick-sep disposable column with sintered disc (5 mL) was packed with Sephadex G-50 (3.5 mL). 2. The column was spun in a 10 mL polypropylene test tube for 5 min at 1000 rpm and 5 min at 3000 rpm to pack the column and spin out the buffer solution. 3. The spun-out buffer solution was decanted and an Eppendorf vial with its cap removed was placed in the bottom of the polypropylene test tube. The spun column was then placed back in a test tube. 4. DNA sample (0.1 mL) was added to the above prepared column, and the column was spun for 5 min at 3000 rpm; the chemicals remained in the column and the DNA was collected into a decapped Eppendorf vial. 164 APPENDIX IV A. PREPARATION OF AGAROSE GEL The agarose (type I-A, IG) was dissolved in electrophoresis buffer (100 mL). This agarose solution was poured on a plate with a "comb" in place, whose teeth would provide a gel slot that had 2-3 times the volume of the DNA sample to be loaded. This precaution eliminated the DNA from streaming out of the gel at the top of the slot. B. RUNNING GEL 1. The gel was submerged in IX E-buffer (made from 20X E-buffer). 2. The digest-dye solution was loaded, so that the slot was only one-half full. 3. The electrodes were connected, and the gel was electrophoresed for desired time and voltage. 4. The gel was submerged in an ethidium bromide solution (1:10,000 dilution of 10 mg/mL) and was stained for 15 min. 5. The gel was carefully placed on a UV-box and the gel was observed with UV-protective glasses and a picture was taken with Polaroid film. 165 APPENDIX V LINEARIZATION OF pSV2-gpt PLASMID (a) Tris-acetate buffer (50 yL) and PvuII restriction enzymes (15 )iL) were added to 100 ^g/mL pSV2-gpt (450 »\) . (b) The above mixture was incubated at 37°C for 3-4 h. (c) After 3 h, TE (12j<L), 0.5 M EDTA (1 jAS) and loading buffer (3 ^L) were added to the above mixture (3 JJL)-(d) An electrophoresis on this mixture and the uncut control was carried out and the DNA bands were compared. (e) Once it was ensured that all DNA was linearized, the enzyme activity of PvuII was stopped with 0.5 M EDTA (lO^wL). (f) Phenol extraction and ethanol precipitation of the DNA: (1) The buffered phenol (0.5 mL) was added to the DNA, and the mixture was shaken and centrifuged for 2 min. (2) The top aqueous layer was transferred to a fresh Eppendorf vial and the organic phase was discarded. Steps 1 and 2 were repeated several times. (3) The residual phenol was extracted with ether (0.5 mL) and the mixture was spun for 10 s; the top ether layer was removed and this procedure was repeated several times; the mixture was warmed at 37°C for approximately 5 min to remove residual ether. 166 (4) NaCl (10 of 5M solution) was then added followed by 95% ethanol (1 mL), and the mixture was chilled at -20°C for 1 h (or overnight) (6) The mixture was centrifuged for 10 min to spin down the DNA; the supernatant was discarded very carefully, and the DNA was partially dried by an airflow. The DNA pellet was then redissolved in TE (0.5 mL) and stored at 4°C. 167 APPENDIX VI PREPARATION OF RAT LIVER S9 FRACTION Day 1: Induction of rat liver enzymes for carcinogen activation 1. Aroclor 1254 was dissolved in corn oil (lg/5mL) by warming up to 37°C (Aroclor is a polychlorinated biphenyl mixture). 2. The rats were injected with Aroclor (200 g rat - 0.5 mL (IP)). 3. The rats were given drinking water and Purina Laboratory Chow until day 4. Day 4: The food was removed. Day 5: The rats were sacrificed by a blow to the head and then decapitated. 1. KC1 (0.15 M) was added to the weighed liver (3 mL/g) in a Erlenmyer flask and the liver was homogenized and put into 4 plastic bottles. 2. The mixture was centrifuged for 10 min at 9000 g, decanted and the supernatant, S9 fraction was saved. The S9 fraction was frozen quickly after distributing into small plastic tubes and was stored at -80°C in a Revco freezer. 168 APPENDIX VII PROCEDURES for IN VIVO STUDY The tumour cell line, K H T sarcoma, was maintained in liquid nitrogen, and the tumour transplantation was carried out as described by Chaplin. 7 7 0 On attaining a weight of 450 to 800 mg, the tumours were irradiated in a manner developed from that 171 described by Sheldon and Hill. The procedure for the preparation of the cell 170 suspensions is described by Chaplin, and the assay system used for cell survival is the 172 soft agar clonogenic assay of Courtenay. The relative cell forming unit per gram of tumour (CFU/g) is calculated as followed: , _ Survival fraction of c e l l yield (chemically treated) ^ chemically treated cells ,, . , . J c e l l yield (control) 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items