"Science, Faculty of"@en . "Chemistry, Department of"@en . "DSpace"@en . "UBCV"@en . "Ferreira, Cara Lee"@en . "2010-01-16T20:46:05Z"@en . "2006"@en . "Doctor of Philosophy - PhD"@en . "University of British Columbia"@en . "This thesis examines the use of carbohydrates to prepare targeted metal complexes as potential diagnostic and therapeutic agents. Carbohydrates are important molecules in living systems, thus highly developed pathways exist for their transport and utilization, and carbohydrate conjugates that take advantage of these pathways have a range of potential medical applications. Carbohydrate conjugates of relevant radioisotopes have value as diagnostic imaging and radiotherapy agents. Utilizing a number of radioisotopes, metal carbohydrate conjugates were prepared and evaluated as potential radiopharmaceuticals. Bidentate, tridentate and cyclopentadienyl ligands with pendant carbohydrates were studied with the [M(CO)\u00E2\u0082\u0083]\u00E2\u0081\u00BA (M = Re, \u00E2\u0081\u00B9\u00E2\u0081\u00B9mTc, \u00C2\u00B9\u00E2\u0081\u00B8\u00E2\u0081\u00B6Re) core; \u00E2\u0081\u00B9\u00E2\u0081\u00B9mTc is an optimal imaging radioisotope and \u00C2\u00B9\u00E2\u0081\u00B8\u00E2\u0081\u00B6Re is of interest for radiotherapy. The complexes of the tridentate ligands were typically more stable than were the complexes of the bidentate ligands. Biodistribution and planar scintigraphy imaging studies with the tridentate complexes showed tumour uptake, suggesting they may have potential as tumour imaging agents. One complex, a glucosamine conjugate with a cyclopentadienyl ligand, retained some biological recognition, as it was shown to be a competitive inhibitor of hexokinase, the first enzyme in glucose metabolism. Ga, In, Cu, and Co all have radioisotopes that can be used in nuclear medicine. Carbohydrate bearing bidentate ligands formed tris(ligand) complexes with Ga(III) and In(III), and bis(ligand) complexes with Cu(II) and Co(II). \u00E2\u0081\u00B6\u00E2\u0081\u00B7Ga and \u00E2\u0081\u00B6\u00E2\u0081\u00B4Cu were used in high yield radiolabelling procedures using these bidentate ligands, but the resulting complexes proved to be susceptible to ligand exchange with amino acids. Targeting the high glucose consumption of the parasite that causes malaria is a novel approach for developing anti-malarials. As well, ferrocene based complexes have shown efficacy in resistant strains of the parasite. Thus ferrocene-carbohydrate conjugates were prepared and examined as potential malaria therapeutics. Several of these complexes had moderate anti-malarial activity in in vitro studies, but their proposed ability to target the high glucose consumption of infected erythrocytes may result in higher efficacy in vivo."@en . "https://circle.library.ubc.ca/rest/handle/2429/18469?expand=metadata"@en . "M E T A L CARBOHYDRATE CONJUGATES FOR DIAGNOSTIC AND THERAPEUTIC APPLICATIONS by C A R A L E E FERREIRA B. Sc., The University of Victoria, 2001 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF T H E REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE F A C U L T Y OF G R A D U A T E STUDIES (Chemistry) T H E UNIVERSITY OF BRITISH COLUMBIA August 2006 \u00C2\u00A9 Cara Lee Ferreira, 2006 A B S T R A C T This thesis examines the use of carbohydrates to prepare targeted metal complexes as potential diagnostic and therapeutic agents. Carbohydrates are important molecules in living systems, thus highly developed pathways exist for their transport and utilization, and carbohydrate conjugates that take advantage of these pathways have a range of potential medical applications. Carbohydrate conjugates of relevant radioisotopes have value as diagnostic imaging and radiotherapy agents. Utilizing a . number of radioisotopes, metal carbohydrate conjugates were prepared and evaluated as potential radiopharmaceuticals. Bidentate, tridentate and cyclopentadienyl ligands with pendant carbohydrates were studied with the [M(CO) 3] + (M = Re, 9 9 m T c , 1 8 6Re) core; 9 9 m T c is an optimal imaging radioisotope and 1 8 6 Re is of interest for radiotherapy. The complexes of the tridentate ligands were typically more stable than were the complexes of the bidentate ligands. Biodistribution and planar scintigraphy imaging studies with the tridentate complexes showed tumour uptake, suggesting they may have potential as tumour imaging agents. One complex, a glucosamine conjugate with a cyclopentadienyl ligand, retained some biological recognition, as it was shown to be a competitive inhibitor of hexokinase, the first enzyme in glucose metabolism. Ga, In, Cu, and Co all have radioisotopes that can be used in nuclear medicine. Carbohydrate bearing bidentate ligands formed tris(ligand) complexes with Ga(III) and In(III), and bis(ligand) complexes with Cu(II) and Co(II). 6 7 G a and 6 4 C u were used in high yield radiolabelling procedures using these bidentate ligands, but the resulting complexes proved to be susceptible to ligand exchange with amino acids. Targeting the high glucose consumption of the ii parasite that causes malaria is a novel approach for developing anti-malarials. As well, ferrocene based complexes have shown efficacy in resistant strains of the parasite. Thus ferrocene-carbohydrate conjugates were prepared and examined as potential malaria therapeutics. Several of these complexes had moderate anti-malarial activity in in vitro studies, but their proposed ability to target the high glucose consumption of infected erythrocytes may result in higher efficacy in vivo. iii T A B L E OF C O N T E N T S Abstract ii Table of Contents iv List of Figures x List of Schemes xv List of Tables xvii List of Abbreviations xix Acknowledgements xxiv Chapter 1 Introduction 1.1 Metals in Medicine 1 1.2 Carbohydrates 2 1.3 Nuclear Medicine 7 1.4 Ferrocene Therapeutics 19 1.5 Malaria 20 1.6 Thesis Overview 22 1.7 References 23 Chapter 2 Carbohydrate-appended 3-Oxy-4-pyridinone Complexes of the [M(CO) 3] + Core (M = Re, Tc) 2.1 Introduction 34 2.2 Experimental 38 2.2.1 Materials and Methods 38 2.2.2 Synthesis of H L 5 40 2.2.3 General Synthesis of Re(L1\"5)(CO)3(H20) 42 2.2.4 Synthesis ofRe(L1-5)(CO)3(H20) by Method C 43 2.2.5 Solid State Structure Determination of H L 5 48 2.2.6 Hexokinase Substrate Assay 49 iv 2.2.7 Hexokinase Inhibition Assay 49 2.2.8 9 9 m T c Radiolabelling 50 2.2.9 1 8 6 Re Radiolabelling 50 2.2.10 Cysteine and Histidine Stability Challenges 51 2.3 Results and Discussion 51 2.3.1 Synthesis and Characterization of H L 5 51 2.3.2 Synthesis and Characterization of ReL 1\" 5(CO) 3(H 20) 54 2.3.3 Evaluation of Compounds as Hexokinase Substrates or 57 Inhibitors 2.3.4 Radiolabelling with 9 9 m T c and 1 8 6 Re 59 2.3.5 Cysteine and Histidine Stability Challenges 62 2.4 Conclusions 64 2.5 References 65 Chapter 3 Rhenium and Technetium Complexes of Tridentate Ligands with Pendant Carbohydrates 3.1 Introduction 71 3.2 Experimental 73 3.2.1 Materials and Methods 73 3.2.2 Synthesis ofReL 6 (CO) 3 and of [ReL 7(CO) 3]Br 74 3.2.3 9 9 m T c Radiolabelling 76 3.2.4 1 8 6 Re Radiolabelling 76 3.2.5 Cysteine and Histidine Stability Challenges 77 3.2.6 Biodistribution Studies 77 3.2.7 Planar Scintigraphy Imaging 78 3.3 Results and Discussion 79 3.3.1 Synthesis and Characterization of ReL 6 (CO) 3 79 3.3.2 Synthesis and Characterization of [ReL 7(CO) 3]Br 82 3.3.3 9 9 m T c and 1 8 6 Re Radiolabelling 87 v 3.3.4 Cysteine and Histidine Stability Challenges 89 3.3.5 Biodistribufion Studies 90 3.3.6 Planar Scintigraphy 95 3.4 Conclusions 97 3.5 References 98 Chapter 4 Glucosamine Conjugates of the Cyclopentadienyl tricarbonylrhenium(I) and Technetium(I) Cores 4.1 Introduction : 102 4.2 Experimental 104 4.2.1 Materials and Methods 104 4.2.2 Modification of the DLT 105 4.2.3 SLT Using Alternative Rhenium Starting Materials 106 4.2.4 Synthesis of Glucosamine Conjugates of the Tricarbonyl 107 cyclopentadienylrhenium(I) Core 4.2.5 Solid State Structure Determination of 2a 109 4.2.6 Hexokinase Substrate Assay 110 4.2.7 Hexokinase Inhibition Assay 110 4.2.8 9 9 m T c Radiolabelling 113 4.3 Results and Discussion 113 4.3.1 Modifications to the DLT and SLT 113 4.3.2 Synthesis and Characterization of Glucosamine Conjugates of the Tricarbonylcyclopentadienylrhenium(I) Core 4.3.3 Solid State Structure of 2a 120 4.3.4 Evaluation of Compounds as Hexokinase Substrates or 122 Inhibitors 4.3.5 9 9 m T c Radiolabelling 123 4.4 Conclusions 126 vi 4.5 References 126 Chapter 5 Gallium and Indium Complexes of 3-Hydroxy-4-pyridinone Pro-ligands Bearing Pendant Carbohydrates 5.1 Introduction 129 5.2 Experimental 132 5.2.1 Materials and Methods 132 5.2.2 Synthesis of Ga(L 5) 3 and In(L5)3 133 5.2.3 Hexokinase Inhibition Assay 135 5.2.4 6 7 G a Radiolabelling 135 5.2.5 Cysteine and Histidine Stability Challenges 135 5.3 Results and Discussion 136 5.3.1 Synthesis and Characterization of Ga(L 5) 3 and In(L5)3 136 5.3.2 Hexokinase Inhibition Studies 140 5.3.3 6 7 G a Radiolabelling 140 5.3.4 Cysteine and Histidine Stability Challenges 142 5.4 Conclusions 144 5.5 References 144 Chapter 6 Copper Complexes of 3-Hydroxy-4-pyridinone and Tetrahydrosalen Pro-ligands with Pendant Carbohydrates 6.1 Introduction 147 6.2 Experimental 153 6.2.1 Materials and Methods 153 6.2.2 Synthesis of Cu(II) Complexes 153 6.2.3 Hexokinase Inhibition Assay 156 6.2.4 6 4 C u Radiolabelling 156 6.2.5 Cysteine and Histidine Stability Challenges 156 6.2.6 PET Imaging Studies 157 vii 6.3 Results and Discussion 157 6.3.1 Synthesis and Characterization of Copper(II) Complexes ... 157 6.3.2 Hexokinase Inhibition Studies 160 6.3.3 6 4 C u Radiolabelling 160 6.3.4 Cysteine and Histidine Stability Challenges 163 6.3.5 PET Imaging Studies 164 6.4 Conclusions 166 6.5 References 166 Chapter 7 Cobalt(II) Complexes of 3-Hydroxy-4-pyridinone Pro-Ligands Bearing Pendant Carbohydrates 7.1 Introduction 171 7.2 Experimental 172 7.2.1 Materials and Methods 172 7.2.2 Synthesis of Co(II) Complexes 173 7.2.3 Hexokinase Inhibition Assays 175 7.3 Results and Discussion 175 7.3.1 Synthesis and Characterization Cobalt(II) Complexes 175 7.3.2 Hexokinase Inhibition Assays 178 7.4 Conclusions 178 7.5 References 179 Chapter 8 Synthesis, Structure and Biological Activity of Ferrocenyl-carbohydrate Conjugates 8.1 Introduction 181 8.2 Experimental 183 8.2.1 Materials and Methods 183 8.2.2 Synthesis of Ferrocenyl-carbohydrate Conjugates 184 8.2.3 Solid State Structure Determination of 1,2 and 12 187 8.2.4 In Vitro Anti-plasmodial Activity Studies 188 viii 8.2.5 Cell Viability Assays 189 8.3 Results and Discussion 191 8.3.1 Synthesis and Characterization of Ferrocenyl-carbohydrate \u00E2\u0080\u009E \u00E2\u0080\u00A2 191 Conjugates 8.3.2 Solid State Structure o f l , 2, and 12 197 8.3.3 Biological Studies 200 8.4 Conclusions 204 8.5 References 204 Chapter 9 Conclusions and Future Work 9.1 Re and Tc Carbohydrate Conjugates for Molecular Imaging 209 and Radiotherapy 9.1.1 Macroscopic Scale Synthesis of Re Carbohydrate Conjugates 209 9.1.2 Radiolabelling and Stability 210 9.1.3 Biological Studies 212 9.2 Carbohydrate Conjugates of Other Isotopes for Use in 214 Nuclear Medicine 9.3 Other Potential Applications in Nuclear Medicine 215 9.4 Ferrocenyl-carbohydrate Conjugates as Potential Anti-malarials 215 9.5 References 216 Appendix 219 ix LIST OF FIGURES Figure 1.1 Examples of metal containing drugs. 1 Figure 1.2 Common hexoses in biological systems including the two anomers of D-glucopyranose. 3 Figure 1.3 Simplified diagram showing glucose uptake, utilization, and 5 efflux in cells. Figure 1.4 Illustration of PET and SPECT detection modalities (not to 12 scale). Figure 1.5 Examples of 9 9 m T c based radiopharmaceuticals. 14 Figure 1.6 F-2-Deoxy-2-fluoro-D-glucopyranose. 15 Figure 1.7 Diagram of FDG uptake and trapping in cells. 16 Figure 1.8 Ferrocene. 19 Figure 2.1 Structures of FDG, [Tc(CO) 3(H 20) 3] +, and a 3-hydroxy-4- 36 pyridinone Figure 2.2 Carbohydrate appended 3-hydroxy-4-pyridinone pro-ligands. 38 Figure 2.3 Structure of H L 5 showing atom labelling scheme (50% thermal 53 probability ellipsoids). Figure 2.4 *H NMR spectra (400 MHz, C D 3 O D : D 2 0 1:1) of (a) 57 ReL 3(CO) 3(H 20) and (b) H L 3 . x Figure 2.6 Figure 2.5 HPLC results for radiolabelling of H L S (a) 1 8 6Re(L 5)(CO) 3(H 20) 61 radiation trace, (b) 9 9 m Tc(L s )(CO) 3 (H 2 0) radiation trace, (c) Re(L 5)(CO) 3(H 20) UV trace. HPLC radiation traces of 9 9 m Tc(L 3 )(CO) 3 (H 2 0) incubated in excess (a) cysteine or (b) histidine after 1, 4, and 24 hours. 63 Figure 3.1 Tridentate pro-ligand and ligands with pendant carbohydrates. 71 Figure 3.2 1 3 C NMR spectra (75 MHz, (CD 3) 2SO) of (a) ReL 6 (CO) 3 and (b) 81 H L 6 . Figure 3.3 'H NMR spectra (400 MHz, C D 3 C N : D 2 0 1:1) of (a) 84 [ReL7(CO)3]Br and (b) L 7 . Figure 3.4 Region of 'H NMR spectrum (400 MHz, C D 3 C N : D 2 0 1:1) of 86 [ReL7(CO)3]Br showing coupling pattern for H8. Figure 3.5 Tumounblood and tumour:muscle ratios at 15, 60 and 120 min 94 in tumour bearing mice (n = 6) for (a) [ 9 9 m TcL 7 (CO) 3 ] + , (b) [ 9 9 m TcL 8 (CO) 3 ] + and (c) [ 9 9 m TcL 9 (CO) 3 ] + . Figure 3.5 Activity (counts/pixel) vs. time curves from planar scintigraphy 96 images using [ 9 9 m TcL 9 (CO) 3 ] + . Figure 3.7 Planar Scintigaphy image using [ 9 9 m TcL 9 (CO) 3 ] + with the visible 96 tumour circled in red. Figure 3.8 Activity (counts/pixel) vs. time curves from planar scintigraphy 97 images using [ 9 9 m TcL 9 (CO) 3 ] + . (Blood activity was estimated as activity observed in head). Figure 4.1 Structures of compounds prepared in Chapter 4. 104 xi Figure 4.2 Lineweaver-Burke plot (double reciprocal plot of 1/initial 111 reaction rate vs. 1/substrate concentration at different FDG inhibitor concentrations) for inhibition of hexokinase by FDG. Figure 4.3 Lineweaver-Burke plot (double reciprocal plot of 1 /initial 112 reaction rate vs. 1/substrate concentration at different 2a inhibitor concentrations) for inhibition of hexokinase by 2a. Figure 4.4 Ki determination for 2a and FDG (Lineweaver-Burke plot slope 112 vs. inhibitor concentration). Figure 4.5 TOCSY ' H N M R spectra (400 MHz, C D 3 O D : D 2 0 1:1) of 2a for 118 P (a) and a (b) anomers and overall 'H NMR spectrum (c) for the region of glucose hydrogen atom resonances. Figure 4.6 lU N M R spectrum (300 MHz, CDC13) of la showing 4 119 resonances (stars) associated with the inequivalent hydrogen atoms on the cyclopentadienyl ring. Figure 4.7 1 3 C N M R spectrum (75 MHz, C D 3 O D : D 2 0 1:1) of 2a with 5 120 major (a anomer, stars) and 5 minor (p anomer, circles) resonances due to in the inequivalent carbon atoms on the cyclopentadienyl ring. Figure 4.8 ORTEP view of 2a showing the atom labelling scheme (50% 121 thermal ellipsoids). Figure 4.9 HPLC traces from radiation detector (top) and U V detector 125 (bottom) for co-injection of cold standard (2a) and reaction mixture containing radiolabeled complex (2b) (both peaks labelled with a star). Figure 5.1 Ligands used in Ga and In-radiopharmaceuticals. 130 xii Figure 5.2 Carbohydrate bearing 3-hydroxy-4-pyridinone pro-ligands. 132 Figure 5.3 'H NMR spectra (400 MHz, CD 3 OD:D 2 0 1:1) of (a) In(L5)3, (b) 138 Ga(L 5) 3 and (c) H L 5 . Figure 5.4 1 3 C NMR spectra (100 MHz, CD 3 OD:D 2 0 l:5)of (a) In(L s) 3, (b) 139 Ga(L 5) 3 and (c) H L 5 . Figure 6.1 Structure of Cu-bis(thiosemicarbazone) complexes. 149 Figure 6.2 Macrocyclic bifunctional chelators for Cu labelling. 150 Figure 6.3 3-Hydroxy-4-pyridinone and tetrahydrosalen pro-ligands bearing 152 pendant carbohydrates. Figure 6.4 HPLC Radiation (top) and UV (bottom) traces for a) Cu(L 3 ) 2 162 and b) C u L 1 1 . Figure 7.1 3-Hydroxy-4-pyridinone pro-ligands bearing pendant 172 carbohydrates. Figure 8.1 Examples of metalloantimalarials 182 Figure 8.2 Structure of 1 showing atom labelling scheme (50% thermal 199 ellipsoids). Figure 8.3 Structure of 2 showing atom labelling scheme (50% thermal 199 ellipsoids). Figure 8.4 Structure of 12 showing atom labelling scheme and the disorder 200 in the unsubstituted Cp ring (50% thermal ellipsoids). xiii Figure 9.1 2+1 Mixed ligand complexes of the [M(CO)3]+ core with a 211 bidentate pyridinone ligand and examples of monodentate ligands. xiv LIST OF S C H E M E S Scheme 1.1 Examples of radioisotopes and decay routes. 8 Scheme 2.1 Synthesis of H L 5 . 52 Scheme 2.2 Preparation and structures of M(L 1\" 5)(CO) 3(H 20) (M = Re, 55 9 9 m T c , 1 8 6Re), and M = Re structures showing numbering scheme for NMR studies. Scheme 3.1 Synthesis of ReL 6(CO) 3 . 79 Scheme 3.2 Synthesis of [ML 7 (CO) 3 ] + , M = Re, 9 9 m T c , 1 8 6 Re. 83 Scheme 3.3 Synthesis and solution structures of ReL (CO) 3 and' 89 [ 1 8 6ReL 7- 9(CO) 3] + . Scheme 4.1 General double ligand transfer (DLT) and single ligand 103 transfer (SLT) reactions. Scheme 4.2 Enzymatic reaction used in hexokinase substrate and 111 inhibition assays. Scheme 4.3 Synthesis of la and 2a. 115 Scheme 4.4 Preparation of 9 9 m T c complexes lb and 2b. 124 Scheme 5.1 Synthesis of M(L S ) 3 , M = Ga, In. 136 Scheme 5.2 Synthesis of 6 7 Ga(L l ' 3 - 5 ) 3 . 141 Scheme 6.1 Synthesis of Cu(II) complexes, Cu(L 1 , 3\" 5) 2. 158 Scheme 6.2 Radiolabelling of tetrahydrosalen pro-ligands. 161 xv Scheme 7.1 Synthesis of Co(II) complexes, Co(L 1 , 3\" 5) 2. 176 Scheme 8.1 Synthesis of compounds 1-5. 193 Scheme 8.2 Synthesis of compounds 6-10. 194 Scheme 8.3 Synthesis of compounds 11 and 12. 195 Scheme 8.4 Preparation of sugar derivative 13. 195 Scheme 9.1 Alternative synthesis of conjugates of the CpM(CO)3 core 212 (Cp = cyclopentadienyl, M = Re, Tc). xv i LIST OF TABLES Table 1.1 Tissue and organ specificity, and natural substrates of some of 4 the glucose transporters. Table 1.2 Examples of cyclotron and reactor produced radioisotopes. 9 Table 1.3 Examples of generator radioisotope combinations. 10 Table 2.1 Selected bond lengths (A) and angles (\u00C2\u00B0) in H L 5 . 53 Table 2.2 Identification methods, retention times and radiochemical yields 59 for the preparation of M(L 1- 5)(CO) 3(H 20) (M = 9 9 m T c , 1 8 6Re). Table 2.3 Percentage 9 9 mTc(L ,\" 5)(CO)3(H 20) remaining intact determined 62 by HPLC after 1, 4, and 24 h in 10\"3 M solution of cysteine or histidine at 37 \u00C2\u00B0C. Table 3.1 Retention times and radiochemical yields for M(L 6)(CO)3 and [M(L 7\" 9)(CO) 3]+ (M = 9 9 m T c , 1 8 6Re). Table 3.2 Percentage 9 9 m Tc(L 6 ) (CO) 3 and [ 9 9 m TcL 7 (CO) 3 ] + remaining intact determined by HPLC after 1, 4, and 24 h in 10\"3 and 10\"4 M solutions of cysteine or histidine, respectively, at 37 \u00C2\u00B0C. 88 90 Table 3.3 Biodistribution results in tumour bearing mice (n = 6) for 92 [ 9 9 m TcL 7 \" 9 (CO )3] + expressed in % total injected dose / weight in grams of wet tissue. Table 4.1 Selected Bond Lengths (A) and Angles (deg) in 2a. 121 Table 5.1 HPLC retention times and radiochemical yields for 142 6 7 Ga(L 1 , 3 - 5 ) 3 . xvii Table 6.1 Cu radioisotopes of interest in nuclear medicine. 147 Table 6.2 Comparison of pyridinone ring IR spectra vCO and UV-visible 159 spectra Amax and for free pro-ligands and Cu-complexes. Table 6.3 HPLC retention times and radiolabelling yields for 163 \"CuOL 1 \" 3 -^ a n d 6 4 C u L 1 0 - n . Table 6.4 Comparison of tumour activity / unit area (10-7 Bq/pixel). 165 Table 7.1 Comparison of pyridinone ring IR spectra vCO for free pro- 177 ligands and Co-complexes. Table 8.1 Selected Bond Lengths (A) and Angles (deg) in 1, 2 and 12. 198 Table 8.2 Cytotoxicity and anti-plasmodial activity of ferrocenoyl- 202 carbohydrate complexes. Table A l Crystallographic data for H L 5 (Chapter 2) and 2a (Chapter 4). 219 Table A2 Crystallographic data for 1, 2, and 12 (Chapter 8). 220 xviii LIST O F ABBREVIATIONS ~ approximate a alpha or alpha particle A angstrom P beta P+ positron P\" beta particle s extinction coeficient (UV-visible) in L mol\"1 cm\"1 y gamma ray X. wavelength Xmax wavelength of maximum absorption (UV-visible) p. micro v frequency 5 chemical shift in parts per million (ppm) from a standard (NMR) Ac acetyl ADEPT antibody directed enzyme prodrug therapy ADP adenosine diphosphate Anal. Analytical atm atmosphere ATP adenosine triphosphate BBB blood brain barrier xix Bn benzyl BNCT boron neutron capture therapy br broad (IR) Bq Becquerels \u00C2\u00B0C degrees Celsius Calcd. calculated Ci Curie CIS coordination induced shift cm\"1 wavenumber COSY correlation spectroscopy (NMR) Cp cyclopentadiene CT computed tomography d doublet (NMR) or days 2D two dimensional DCC dicyclohexylcarbodiimide DLT double ligand transfer DMF dimethylformamide DMSO dimethylsulfoxide DOTA 1,4,7,10-tetraazacyclododecane-N,N',N\",N'\"- tetraacetic acid DTPA diethylenetriaminepentaacetic acid EA elemental analysis EC50 effective concentration (compound concentration at which 50% of the wanted effect is observed compared to control) ECD N-(3-dimethylaminopropyl)-N'-ethylcarbodiimide EDTA ethylenediaminetetraacetic acid ES-MS electron impact mass spectrometry EtOH ethanol EtOAc ethyl acetate eV electron volt fac facial FDA Food and Drug Administration (USA) FDG 2-deoxy-2-fluoro-D-glucose g gram G6P glucose-6-phosphate GLUT glucose transporter h hour(s) HIV human immunodeficiency virus HK hexokinase HMBC heteronuclear multiple bond coherence (NMR) HMIT H + / myo-inositol co-transporter HMQC heteronuclear multiple quantum coherence (NMR) HOBt 1 -hydroxybenzyltriazole HPLC high performance liquid chromatography Hz hertz (s\"1) IC50 inhibitory concentration (compound concentration at which 50% of growth is inhibited relative to the control) xxi IR infrared J coupling constant (NMR) k kilo Ki inhibition constant LEAPT lectin directed enzyme activated prodrug monotherapy LSIMS liquid secondary ion mass spectrometry m metre or milli or mutliplet (NMR) or medium (IR) M molar (moles / litre) or mega MeOH methanol min minutes mol mole MRI magnetic resonance imaging MS mass spectrometry MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide m/z mass per unit charge N A D H nicotinamide adenine dinucleotide n nano NCT neutron capture therapy NHS N-hydrxoysuccinimide NMR nuclear magnetic resonance ORTEP Oak Ridge Thermal Ellipsoid Plot PBS phosphate buffer saline Pd/C palladium on carbon xxii PET positron emission tomography pH -log[H 30] + PMT prodrug monotherapy ppm parts per million q quartet (NMR) rf retention factor rt retention time s singlet (NMR) or strong (IR) SOD superoxide dismutase SLT single ligand transfer SPECT single photon emission computed tomography t triplet (NMR) ti/2 half-life duration T E T A 1,4,8,1 l-tetraazacyclotetradecane-N,N',N\",N'\"-tetraacetic acid THF tetrahydrofuran T L C thin layer chromatography TOCSY total correlated spectroscopy UBC University of British Columbia UV ultraviolet xxiii A C K N O W L E D G E M E N T S I would like to thank my best friend and husband, Tony, for his support and encouragement (and keeping me well-fed). Thanks to my parents for all their help and support over the years. My brothers, Dean and Jon, should be thanked for all their negative reinforcement, nothing makes me more driven to succeed than proving my brothers wrong. I could not have had better mentors than my supervisors Dr. Mike Adam and Dr. Chris Orvig. Thank you for your confidence in me, your enthusiasm for my research, and the independence you gave me to pursue so many projects. All the opportunities you gave me to travel and present this research in the scientific community are greatly appreciated. The work environment in the Orvig group was outstanding, thanks to Tim, Dave, Cheri, Neil, Simon, Vishakha, Fabio, Michael, Jen, Meryn, Kathie, Chuck, Lauren, Kathy, Karen, George, Jessica, Shirley, Khosro, Barry, and anyone else I forgot. I would also like to thank the PET group at TRIUMF for their interest and help, Suzy, Tom, Jen, Andre, Salma, Ken, Wade, Paul, Julia, Milan, Yulia and James. Thanks to Dr. Shigenobu Yano, Dr. Yuji Mikata, and Yuko Sugai for their continued collaboration. Dr. Don Yapp, Dr. Anna Celler, Dr. Don Lyster and Fabio Marques are acknowledged for their assistance with animal studies. Acknowledgements for assistance in biological studies go to Dr. Elena Polishchuk, Candice Martins, Dr. Vanessa Yardley, Susan Little, and Dr. John Moss. Finally, I would like to recognize the support staff: Dr. Brian Patrick (X-ray), Marrieta Austria (NMR), Liane Darge (NMR), Dr. Nick Burlinson (NMR), Dr. Yun Ling (MS), and the office staff. xxiv Chapter 1 Introduction 1.1 Metals in Medicine Metals have been used in medicine for over a century and continue to play an increasing role in our health care. Metal compounds containing bismuth,1 gold2 and lithium have historically been used as medications for numerous disorders, but their modes of action were not well understood. Yet, some of the most successful drugs contain metals, such as auranofin (rheumatoid arthritis),4 cisplatin (cancer),5 and Cardiolite\u00E2\u0084\u00A2 (heart imaging) (Figure 1.1).6 Although today there is greater understanding of the mechanism of action of inorganic drugs, many of the first metal containing drugs, such as cisplatin and lithium carbonate (bipolar disorder), were OAc AcO OAc S-Au-PEt 3 Auranofin CI NH-j Cisplatin R I N III C R III % N I R R = CH 2C(CH 3)20CH3 Cardiolite Figure 1.1. Examples of metal containing drugs. 1 discovered serendipitously, later paving the way for the field of medicinal inorganic chemistry. The development of metal containing drugs has become increasingly design based, as our understanding of bioinorganic chemistry and the fate of administered drugs increases. Metal based drugs can be optimized to better fit their intended application by, for example, tuning the metal oxidation state and ligand properties to produce a better balance of the toxic and beneficial properties of the metal or to improve bioavailability.7 The attachment of biomolecules, such as antibodies, has increased the ability of inorganic compounds to target specific functions in the body.8\"10 This thesis describes the use of metal conjugates of carbohydrates in the development of targeted metal compounds for potential diagnostic and therapeutic applications. 1.2 Carbohydrates Carbohydrates are the most abundant class of organic molecules found in living organisms, and have a wide range of functions.11'12 The general formula for carbohydrates is C H 2 O , hence the name's suggestion of hydrated carbon. A more general definition also includes derivatives containing heteroatoms such as S, N, and P. Most biologically important carbohydrates are monomers, dimers or polymers of hexoses, which have the basic formula [CeHnOejn. The most ubiquitous hexoses in living systems are glucose, galactose and fructose.13 These hexoses are isomers that are found in both open chain and cyclic forms, and usually as five or six membered rings. In the cyclic form, two isomers are possible, known as a and p anomers, which can 2 interconvert via mutorotation (Figure 1.2). In biological systems smaller carbohydrates such as monosacchrides (n =1) and disaccharides (n = 2) are basic energy sources for 12 cells, while larger polymeric carbohydrates, such as glycogen and cellulose, have a myriad of biological roles, including energy storage and as cellular structural material.13 More complex biomolecules, where carbohydrates are chemically attached to other fragments, are known as glycoconjugates, such as glycoproteins and glycolipids, and are important in cellular recognition and communication between cells, in addition to functioning as cellular structural material.13 oc-D-galactose OH| a-D-glucose ^ H HO' O H 0 -OH OH CH2OH (3-D-fructose OH :HCjjr> OH OH p-D-glucose Figure 1.2. Common hexoses in biological systems including the two anomers of D-glucose. Simple carbohydrates are the major energy source for the human body, for which highly developed transport and utilization pathways exist.14\"16 Because carbohydrates are hydrophilic, they cannot penetrate the lipid bilayer of cells and thus require facilitated transport via specific proteins known as glucose transporters, GLUT-1-12 and HMIT (a F T / myo-inositol co-transporter).15'16 Glucose transporters are large proteins that span 3 the cell membrane, and many are relatively substrate and tissue specific.1 6'1 7 Examples of some glucose transporters, their locations in the body and some of their natural substrates are given in Table 1.1. Glucose is the most important monosaccharide Table 1.1. Tissue and organ specificity, and natural substrates of some of the glucose transporters.15,17\"20 Glucose transporter GLUT-1 GLUT-2 GLUT-3 GLUT-4 GLUT-5 Associated organ or tissue Selected natural substrate(s) erythrocytes, blood brain barrier liver, pancreas, kidney brain (neuronal), nerve cells, liver, heart muscle, heart intestine, testis, muscle glucose, glucosamine, galactose glucose, glucosamine, fructose, galactose, fructose glucose, galactose glucose, glucosamine fructose for energy in the body, especially for the function of the heart and brain. Glucose is a substrate for most glucose transporters except (GLUT-5 and HMIT which are specific for fructose) and is readily converted to energy in almost any cell type.16 Although the brain comprises only 2% of total body mass, it accounts for up to 30% of the body's utilization of glucose. This is mainly due to glucose being the only energy source for neurons, except in extreme conditions such as starvation. Because of the importance of glucose for brain function, there is a high concentration of GLUT-1 transporters at the blood 4 brain barrier (BBB) to facilitate uptake into the brain. ' The heart is also a large consumer of glucose, as well as other energy supplying biomolecules such as fatty acids. Other areas of high glucose consumption include muscle tissue and the liver; the latter is where glucose is converted to glycogen for energy storage. Once transported into cells, glucose is converted to glucose-6-phosphate (G6P) by hexokinase or glucokinase (liver), stopping efflux back into the blood stream. G6P is then utilized in one of several pathways, the Krebs cycle for energy production being the major route of metabolism. Blood plasma Cell glucose Further reactions Glucose-6-phosphate +ADP Glucose ^ transporter hexokinase Glucose +ATP Figure 1.3. Simplified diagram showing glucose uptake, utilization, and efflux in cells. Many current and potential drugs utilize carbohydrates as either an active part of the drug, or as a drug conjugate for targeting, lowering toxicity, or increasing solubility. Auranofin is an orally active gold based therapeutic containing an acetylated 1-deoxy-l-thiol-D-glucose moiety; it is used in the treatment of rheumatoid arthritis.4 Carbohydrate vaccines are used for infectious diseases23 and are being considered for cancer.2 4'2 5 5 Carbohydrates are abundant in the structure of pathogens, like bacteria and viruses, hence glycoconjugates that resemble a portion of the pathogen are used in vaccines to stimulate appropriate antibody production. So successful are carbohydrate based vaccines that the majority of vaccines in use contain a carbohydrate antigen.13 Cancer cells have different surface carbohydrates than do normal cells; incorporating these carbohydrate motifs into a vaccine is being investigated for the induction of immune responses against cancer cells. 2 4' 2 5 Carbohydrates have been conjugated to active pharmaceuticals, such as anti-\u00E2\u0080\u00A2y/r 77 78 70 \"lf\ convulsants, analgesics, ' neurotransmitter derivatives, anti-cancer agents, and 31 HIV drugs'' to increase delivery to the brain and central nervous system via facilitated transport by GLUT-1 across the BBB. Tumours can also be targeted by carbohydrate conjugates due to the increased expression of GLUT-1 transporter on the surface of tumours and higher hexokinase activity within many tumour cells. \" Targeting aids in lowering toxicity, by reducing the dose to non-target cells, and increasing efficacy of the drug. More complex targeting mechanisms also use carbohydrates to mask the potential active drugs, such as anti-cancer agents;37'38 once the drug reaches the target, it is unmasked and thereby activated, through enzymatic cleavage of the carbohydrate by a glycosidase. The masked potential drug is called a prodrug. When the enzyme that unmasks the prodrug is naturally present at the target, this method is called prodrug T O mono-therapy (PMT). Alternatively, an exogenous enzyme can be directed to the target by antibodies or lectins prior to the introduction of the prodrug, in a technique \u00E2\u0080\u00A2 ^ 7 known as antibody directed enzyme prodrug therapy (ADEPT) and lectin directed enzyme activated prodrug therapy (LEAPT). 3 9 In LEAPT, the carbohydrate is bound to both the site directed enzyme and the prodrug to increase cell type selectivity of both via specific carbohydrate-lectin interactions.39 These targeting methods greatly increase 6 selectivity of the potential drug uptake, improving their efficacy. Clearly, carbohydrates are useful tools in designing novel medicines. 1.3 Nuclear Medicine The field of nuclear medicine utilizes radiopharmaceuticals; pharmaceuticals that incorporate radioisotopes. Depending on the choice of radioisotope, the radiopharmaceutical can be used for either diagnostic imaging or radiotherapy applications. In the case of imaging, the radiopharmaceutical incorporates either a positron ( P + ) (positively charged electron or electron anti-matter) or a gamma-emitting (y) isotope and is designed to relay information about biological function, such as metabolism or receptor activity. The ability to image biological function is a complement to other common imaging techniques, such as X-ray, computed tomography (CT), and magnetic resonance imaging (MRI), which typically only give images of the physical anatomy such as skeletal or soft tissue. In the case of therapy, the radiopharmaceutical incorporates a radioisotope that emits a particle that kills cells, such as an alpha particle (a) (helium nucleus), beta particle (P\" ) (high energy electron), or Auger electron. The radiopharmaceutical is designed to localize at a particular site, such as in tumours, to kill selectively only the unwanted cells and not harm healthy cells. Metal radioisotopes are of particular interest for radiopharmaceutical development because they offer a wide range of physical (nuclear) and chemical properties that can be selected to match the application. 7 Radioisotopes are isotopes, elements with the same number of protons (same atomic number) but with a different numbers of neutrons, that have an unstable nucleus.40 An unstable nucleus can arise from an unfavorable combination of protons and neutrons or excess energy in the nucleus. To become a stable nucleus the radioisotope will decay to produce another isotope which can be stable or unstable; unstable nuclides will continue to decay until a stable nucleus is achieved. Decay can occur by several different routes, some examples of which are shown in Scheme 1.1, and it is common for radioisotopes to decay simultaneously by one or more of these routes.40 Positron decay Electron capture decay Beta particle decay Isomer transition Alpha decay decay routes. The production and half-life (time for half of the radionuclides to decay) of a radioisotope influences their availability and utility. Three methods are used to prepare medically relevant radioisotopes, a reactor,41 a particle accelerator42 or a generator. For both reactor and cyclotron produced radioisotopes (Table 1.2), availability can be an issue. If the half-life of the radioisotope is short, it can only be acquired and used at locations near the production site, of which there are few. Longer 8 1 8 F \u00E2\u0080\u00A2 1 8 0 + (3+ 6 7 G a \u00E2\u0080\u00A2 6 7 Z n + y i 8 8 R e ^ 1 8 8 0 s + p\" H3m I n ^ 1 1 3 In + y 2 i i A t ^ 2 0 7 B i + a Scheme 1.1. Examples of radioisotopes and Table 1.2. Examples of cyclotron and reactor produced radioisotopes.4 Radioisotope Production reaction Production method 5 5 Co 56Fe[p,2n] Cyclotron 6 4 C u 67Zn[p,a] Cyclotron 94m T c 94Mo[p,n] Cyclotron 1 1'In mCd[p,n] Cyclotron 6 7 C u 67Zn[n,p] Reactor 1 8 6 Re . 185Re[n,y] Reactor i 6 9 Y b 1 6 8Yb[n,y] Reactor \"Mo 98Mo[n,y] Reactor lived isotopes can and are shipped greater distances, resulting in greater practical availability. The target material can affect the purity and specific activity of the radioisotope. Specific activity is a measure of radioactivity per weight or unit mass of material, for example mCi activity/mmol. Many target materials, even if a single element, contain numerous isotopes which, when irradiated, give numerous products. As well, different nuclear reactions will also yield additional products. If these products are different elements they can be separated, but if they are different isotopes of the same element they typically are not separated, contaminating the desired radioisotope. It is common to use enriched target materials that are mainly one isotope and to optimize the reaction to limit contaminants. Sometimes the target material is the same element as the product and is not separated, such as in the reactor production of Re-186 from Re-185; 9 thus the specific activity is reduced. A generator utilizes a parent radionuclide, which decays to the radioisotope of interest. The parent radionuclide is produced either by a reactor or a cyclotron, and is then mounted on a column material. The decay product, or daughter radionuclide, can be eluted from the column periodically. Generator produced radioisotopes are ideal because the generator can be shipped longer distances; the daughter radioisotope can then be eluted when needed at the site of radiopharmaceutical use in high purity and high specific activity. Medically relevant radioisotopes pairs that are amenable to a generator are given in Table 1.3. Table 1.3. Examples of generator radioisotope combinations. Parent radionuclide Half-life of parent Daughter radionuclide \"Mo 66 h 9 9 m T c 1 8 8 W 69 d 1 8 8 Re 6 2 Z n 9.2 h 6 2 C u 6 8 Ge 271 d 6 8 G a 8 2 Sr 25.6 d 8 2 Rb Most radioisotopes require synthetic modification of the chemical form in which they are acquired to obtain a radiopharmaceutical with the appropriate chemical and biological properties. Radiochemistry differs from typical synthetic chemistry in several ways44 and the synthetic routes chosen are controlled by several factors: synthesis must start from the chemical form in which the radioisotope is available (generally very limited), the radioisotope is in usually very low concentration (< 10\"6 M) (thus other 10 reagents are usually in excess), the radioisotope is not normally isolated as a solid, but in an aqueous solution (so chemistry should be compatible with aqueous systems), and the number and complexity of synthetic steps is limited by the half-life of the isotope (total synthesis time should be less than one half-life). For commercialization, the chemistry must be simple so it can be performed by a radiopharmacy technician who is preparing several radiopharmaceutical agents at once, often very early in the morning. Purification must also be simple and fast, the final product must have high radiochemical yield and be in an injectable solution, such as saline or saline buffer. The optimal radiopharmaceutical preparation is an injectable saline solution of the radioisotope in a single vial containing all the reagents, with product formation in high yield without need for purification. The typical characterization techniques used in synthetic chemistry are inapplicable to most radiopharmaceutical chemistry, due to the low concentrations and limiting half-lives of the radioisotopes. Fast, informative identification of the product, as well as verification of radiochemical yield and purity, is done by comparison of the radiochemical product with a macroscopic non-radioactive standard using either thin layer chromatography (TLC) or high performance liquid chromatography (HPLC) coupled with a radiation detector. Two techniques utilize radiopharmaceuticals for molecular imaging in nuclear medicine, positron emission tomography (PET) and single photon emission computed tomography (SPECT). These two techniques use (3+ and y emitting radioisotopes, respectively. Due to the different types of emissions used, PET and SPECT have different detection methods, which give different resolution and sensitivity (Figure 1.4). In PET, the emitted P + travels away from the radiopharmaceutical where it annihilates 11 with an electron. The distance traveled by the P + is usually only a few millimeters and is dependent on the kinetic energy of the particle. The annihilation reaction yields two coincident y-rays with the same energy (511 keV). Detectors encircling the patient detect both y-rays and are used to determine the site of annihilation. Because it is the annihilation that is detected and not the originally emitted positron, the resolution is = radiopharmaceutical accumulation Figure 1.4. Illustration of P E T and S P E C T detection modalities (not to scale). limited to the average distance the P + travels from the radiopharmaceutical. The majority of the y-rays are detected and used in the computing of a three dimensional image, resulting in high sensitivity. In S P E C T , a single y is emitted directly from the radioisotope and travels in any direction. Theoretically, the emission of a y directly from a radiopharmaceutical can yield optimal resolution. In order to form a three dimensional structure however, the direction from which the y arrives at the detector must be determined, and so a collimator is placed in front of the detectors. A collimator is a lead 12 plate covered in very small holes; only the y perpendicular to the collimator can pass through the perforations to the detector. This allows the trajectory of the arriving y to be determined, while all other y are absorbed by the collimator and remain undetected. Consequently, only a small fraction of the emissions are detected and used to create the three dimensional image, lowering sensitivity. Many other factors beyond the scope of this discussion influence the image quality produced from PET and SPECT imaging, but currently PET imaging is favoured for resolution and sensitivity. Numerous radiopharmaceuticals have been approved by the U.S. Food and Drug Administration (FDA) and are in clinical use for SPECT imaging.6'4 5 One of the first SPECT agents approved was [ I]NaI for thyroid imaging. Some radiopharmaceuticals are quickly metabolized and the radioisotope is then incorporated into a biological substance that is responsible for its biodistribution in vivo. Examples include 67Ga-citrate, from which 6 7 G a is incorporated into transferrin and used in tumour and infection imaging, 4 6 and 1 1 ' in -oxine, which is used to incorporate 1 1 ' in into white blood cells for inflammation and infection imaging.6 The vast majority of SPECT agents utilize 9 9 m T c , due to its ideal nuclear properties (ti/2 \u00E2\u0080\u0094 6 h, y = 141 kEv, 89%) and inexpensive, convenient availability from a 9 9 M o / \" m T c generator system.47 Greater than 85% of imaging that utilizes a radioactive substance incorporate Tc. Some Tc agents clinically used are shown in Figure 1.5. The majority of agents in use can be referred to as radioisotope essential, meaning that the radioisotope plays a crucial role in the in vivo 13 R = CH 2 CH 2 OCH 2 CH3 Ceretec Neurolite Myoview Figure 1.5. Examples of 9 9 m T c based radiopharmaceuticals.47 properties of the radiopharmaceutical.47 The biodistribution is controlled by factors such as charge, lipophilicity and size. Some examples include Myoview\u00E2\u0084\u00A2 and Cardiolite\u00E2\u0084\u00A2, lipophilic cationic species, used for myocardial perfusion, and Ceretec\u00E2\u0084\u00A2 and Neurolite\u00E2\u0084\u00A2, neutral species, used in cerebral perfusion.45 9 9 m T c - H M D P (HMDP = hydroxymethylenediphosphonate), known as Qsteoscan\u00E2\u0084\u00A2, and several other similar agents are used for bone imaging due to the high affinity of the phosphate and/or phosphonate ligands for bridging to the calcium hydroxyapatite bone matrix.6 The future of radiopharmaceutical development is focused on tagging biomolecules, such as peptides,8'48-50 antibodies33'51 and receptor antagonists47'48'52 to facilitate better targeting of specific tissue type and function. Recently approved, In-DTPA octreotide (Octreoscan\u00E2\u0084\u00A2), 5 3 a peptide conjugate used for imaging neuroendocrine tumours, is an example of this next generation of radiopharmaceuticals. In contrast to the multitude of FDA-approved radiopharmaceuticals utilized in SPECT, only one agent has been FDA approved for clinical PET imaging. I8F-2-Deoxy-2-fluoro-D-glucopyranose (FDG) (Figure 1.6) is used to image glucose metabolism.54\"57 FDG, which is structurally similar to glucose differing only in the substituent on C2, is 14 transported through the body in the same, manner as glucose (Figure 1.7). It is transferred into cells by the same glucose transporter proteins, where it is phosphoryiated Figure 1.6. 1 8F-2-Deoxy-2-fluoro-D-glucopyranose (FDG). by hexokinase. F D G is not a substrate for any of the subsequent enzymatic processes in the metabolism or utilization of glucose, including glucose-6-phosphate isomerase, the enzyme responsible for de-phosphorylation. Because the phosphoryiated carbohydrate cannot transverse back across the cell membrane, F D G becomes trapped intracellularly thus accumulating in the cell resulting in highly sensitive imaging. A n y of the OH Blood plasma C Cell e m e m b a FDG-6-phosphate +ADP n e hexokinase F D G Glucose transporter F D G +ATP Figure 1.7. Diagram of F D G uptake and trapping in cells. 15 radiotracer that is not taken up into cells is excreted through the urinary system. Again, FDG is not a substrate for the enzyme responsible for transporting glucose from the urinary system back into blood plasma, so FDG shows fast blood clearance, giving good target to background ratios.57 Because glucose is a major energy source for the body, especially for the heart and brain, FDG has found utility in a myriad of areas of medicine, such as cardiology55'58 and neurology.54'55 In cardiology, FDG is used to differentiate between scar tissue and compromised myocardial tissue that is still viable and will benefit from revascularization. In neurology, glucose consumption is closely related to neuron activity and the different patterns of cerebral FDG localization are used in focal epilepsy to plan surgical procedures, and in the diagnosis of dementia sources, such as Alzheimer's disease or infarction.55'59 As well, FDG is used in oncology as a tumor marker, due to the elevated level of glucose metabolism in many tumor types, 3 2' 6 0 - 6 2 with the rate of glucose metabolism being proportional to the level of malignancy.36'63 FDG is used to locate tumour and metastatic tissue, to aid in tumour grading, to evaluate tumour response to treatment, and to detect tumour recurrence by differentiating tumour tissue from necrosis or scar tissue. 5 5' 6 0 - 6 2 Even with the myriad of applications of PET imaging with FDG already being employed, further applications in areas such as psychiatric disorders55 are currently being investigated. Radiotherapy is the use of radiation to damage and kill unwanted cells, such as cancer cells. 6 4' 6 5 As energetic charged particles, such as those emitted by radioisotopes, travel through matter they lose energy by interacting with and transferring energy to the orbital electrons of atoms. The atoms encountered are ionized by the ejection of the electron with increased energy or are promoted to an excited energy state. Damage to the cell occurs either directly by the emitted electron breaking chemical bonds, or indirectly 16 by the formation of radical species, commonly hydroxyl radicals. Radiation damage either kills the cells immediately or interferes with cell reproduction. Cells have repair mechanisms allowing them to recover from radiation damage, but normal cells recover more easily than do those that are malignant. There are three types of radiotherapy: external radiation, internal radiation with a sealed source (also called brachytherapy) and internal radiation with an unsealed source. The majority of clinical radiotherapy uses external radiation, where a highly focused beam of radiation is directed at the edge of a tumour, destroying cells and preventing growth or regrowth.64 The radiation beam is commonly high energy x-rays, electrons or gamma rays from a 6 0 Co source. Some of the limitations of external beam radiotherapy include required a priori knowledge of the tumour location, significant damage to the normal cells in proximity of the tumour and inability to treat secondary or distant metastasis. A potential method to overcome these limitations is neutron capture therapy (NCT), often using 1 0 B and referred to as boron neutron capture therapy (BCNT). 6 6 ' 6 7 BNCT requires first the selective delivery of 1 0 B to the target cells. Once the 1 0 B is localized, external beam radiation of the area is done using innocuous low energy neutrons, initiating the fission of 1 0 B to 7 L i and an a particle which causes localized cellular damage. If the boron delivery is selective it is not necessary to know the exact location of the tumour and there is less damage to surrounding normal cells. The effectiveness of BNCT depends on the selectivity of localization and the number of boron atoms that accumulate in the targeted cells; to ensure cell death approximately 109 boron atoms are required per cell. Hence, BNCT research is focused on the developing agents with tumour targeting moieties that incorporate many boron atoms in one entity.66\"69 17 Internal radiation therapy uses radioactive material positioned in the body near or at the site of interest, minimizing the irradiation of non-targeted tissue compared to external radiation therapy. In brachytherapy, a sealed radioactive source, sealed in vehicles such as microspheres or colloids,70 is introduced in close proximity to the tumour, by such means as a catheter or implantation. The radioactive materials used are either beta particle or Auger electron emitting radioisotopes. p~ particles are high energy electrons, with a range of energies particular to the radioisotope. These high energy electrons travel away from their source producing ionizing damage leading to cell death. The range of the electrons needs to be appropriate for the size of the tumour to limit the amount of radiation deposited to the surrounding normal tissue. Auger electrons have a short range requiring close proximity to the target tissue, but minimizing the dose to non-target tissue. The final type of radiation therapy uses an unsealed radioactive source, such as a small molecule containing a radioisotope, internalized in the body. 5 1 ' 6 5 ' 7 0 A solution containing a radiopharmaceutical is injected into the patient, and localizes at the target. Different types of radioisotopes that could be utilized for internal radiation therapy include P\" particles, a particles and Auger electrons.70 Currently, most radiopharmaceuticals used for unsealed internal radiation therapy incorporate beta particle emitting radioisotopes, likely due to their wider availability and effectiveness. Radiopharmaceuticals of this type have been used in cancer and bone palliation therapy 70 for decades. Alpha particles, which are also produced in BNCT, are high energy helium nuclides that produce damaging ionization in high density over a short range.70 Auger electrons deposit their ionizing radiation at the subcellular level, and need to be 18 taken up into the c e l l nucleus to cause effective c e l l d a m a g e . 7 0 A l p h a particles are opt imal for s m a l l tumours due to the more l o c a l i z e d nature o f their damage. A u g e r electrons are even more l o c a l i z e d , and must target every c e l l to be c o m p l e t e l y effective. 7 W h i l e the h i g h l y l o c a l i z e d nature o f a particles and A u g e r electrons is benef ic ia l for m i n i m i z i n g the p r o b l e m s related to healthy tissue damage, such as bone m a r r o w suppression, further w o r k is required to obtain the h i g h l e v e l o f c e l l targeting needed for effective therapy. 1.4 Ferrocene Therapeutics Since ferrocene was first prepared and characterized h a l f a century a g o , 7 1 - 7 4 its chemistry and appl icat ions have been extensively s t u d i e d . 3 7 ' 7 5 - 1 0 9 Ferrocene has a r i c h chemistry s t e m m i n g f r o m the large number o f derivatives and their interesting e lectrochemical properties, w h i c h has led to their use i n numerous industries, i n c l u d i n g 75 76 petroleum, plast ics , and metal lurgy. Current areas o f interest i n ferrocene chemistry include use i n c a t a l y s i s , 7 7 \" 8 4 as sensors and as immunoassay reagents . 8 5 \" 9 2 Fe Figure 1.8. Ferrocene. 19 Ferrocene has several properties which have facilitated its investigation for potential biological applications.93'94 Typically, organometallic compounds are sensitive to moisture and air, but ferrocene belongs to a unique group of organometallics which are stable under both aqueous and aerobic conditions. The small size, lipophilicity, easy chemical modification, and the biologically accessible one electron oxidation potential of ferrocene make it an attractive reporter moiety 9 5 - 1 0 1 and an intriguing pharmaceutical vector.3 7'1 0 2\"1 0 9 By exploiting these properties, many ferrocene derivatives, including amino acid,9 5 peptide,9 6\"1 0 0'1 1 0 protein96\"100 and D N A 1 0 1 ' 1 1 1 conjugates, have been used for studying electron transfer processes and in the development of potential biosensors and immunoassay reagents.85\"92 Recently, the redox activity of ferrocene lipid conjugates has been used to modulate cell transfection.\"2 Ferrocene itself has been used as both a cytotoxic and an anti-anemic agent. Similarly, ferrocene derivatives have also shown anti-tumour103\"107 and anti-anemic activity. 3 7' 1 0 8' 1 0 9 Drug conjugates which have been * \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 113 structurally varied to incorporate ferrocene have displayed a wide range of activities, ' 1 1 4 examples of which include conjugates with antibiotics (penicillin) and cancer drugs (tamoxifen).113'114 1.5 Malaria Malaria is a widespread parasitic disease that affects an enormous population. ' 1 1 5 Four species of the human infecting parasite are known, Plasmodium falciparum being the most lethal of these. It is estimated that over 2 billion people have been exposed to the parasite; yearly, 300-500 million new cases are reported and between 1 and 2.7 million people die from the infection.3 7'1 1 5 Historically, malaria is found in 20 tropical areas, including the poorest countries; although the majority of cases originate in Africa, Southeast Asia, India, and parts of South America, the disease is now threatening to spread into more temperate zones of the world. Malaria is caused by a protozoan organism which has a highly complex life cycle.1 1 6 When a mosquito infected with the parasite feeds on a human host, a small amount of the parasite, in the form of sporozoites, is transferred from the saliva of the mosquito to the bloodstream of the human. These sporozoites travel through the blood stream to the liver and mature to merozoites, which burst out of the liver cells and invade red blood cells, where the parasite undergoes several stages of its life cycle. In the final stage, the infected red blood cells burst releasing more merozoites which propagate the cycle. In the true nature of a parasite, the malaria parasite feeds off its host, sequestering nutrients such as hemoglobin, amino acids and glucose. Specific high affinity pathways are used by the parasite to accumulate these nutrients in preference over non-infected cells. 1 1 6' 1 1 7 Notably, the parasite is highly dependent on a large and continuous supply of hexose acquired exclusively from its host.118 Since 1942, chloroquine has been widely used as an effective, inexpensive anti-malarial agent, but parasite resistance to chloroquine and pyrimethamine, another inexpensive anti-malarial, has become problematic for treating malaria, especially in poorer countries.119 In Southeast Asia there is resistance to all the widely used anti-malarials, except combination therapies using artemisinins.119 Parasite mutations have 21 o Chloroquine Artemisinin Figure 1.9. Currently used anti-malarial drugs. been implicated in increased drug resistance. The need has risen for new anti-malarials with different uptake mechanisms or modes of action, which are active against the resistant strains, and alternative methods to eradicate the parasite. Significant effort has focused on finding a vaccine for malaria, but barriers such as the complexity of the parasite life cycle and the ability of the parasite to undergo antigenic variation have slowed research in this area. Novel therapies for malaria include synthesis of derivatives of currently used anti-malarials,121 metal conjugates of known anti-malarials,109' 1 2 2 \" 1 3 4 and the investigation of inhibitors of enzymes crucial to parasite 121 135 137 growth. ' \" To lower transmission of the disease, insecticides and mosquito nets are being used in areas with high infection rates. Unfortunately, scant funding has led to insufficient research for novel drugs and treatments for malaria, and the disease still threatens almost half the world population. 1.6 Thesis Overview This introduction provides a basic background for the experimental work in this thesis. The investigation of carbohydrate conjugates of radioisotopes for potential use in 22 nuclear medicine is discussed in Chapters 2 through 7. FDG has shown great versatility in molecular imaging, but its use has been limited by its availability and its lack of therapeutic application. Thus, the development of novel radiopharmaceuticals that could be made more readily available or have therapeutic potential, and that utilize the biological properties of carbohydrates, is a worthwhile goal. Metal carbohydrate conjugates were prepared and evaluated as potential diagnostic and therapeutic agents. A range of radionuclides, each with their own advantages, were investigated. Metal-carbohydrate conjugates were prepared with non-radioactive metals on the macroscopic scale, fully characterized, and used as standards for radiolabelling. The radiochemistry was developed, and the complexes evaluated as potential radiopharmaceuticals in vitro and/or in vivo. The second and third row transition metal congeners, Tc and Re, for both diagnostic ( 9 9 mTc) and therapeutic ( 1 8 6' 1 8 8Re) applications, are the focus of Chapters 2 through 4. Carbohydrate conjugates of some other radioisotopes of interest to nuclear medicinal chemistry are covered in Chapters 5 through 7, 6 7 ' 6 8 G a / m I n , 6 4 ' 6 7 C u , and 5 5 Co, respectively. The use of carbohydrates as targeting vectors also has medicinal applications outside of nuclear medicine. In Chapter 8, ferrocenyl-carbohydrate conjugates are examined as potential antimalarials, taking advantage of the higher glucose consumption of infected erythrocytes.117 Several ferrocenyl-carbohydrate conjugates were synthesized and evaluated in vitro. Anti-plasmodial studies were done with both chloroquine sensitive and resistant strains of the parasite to assess their potential as anti-malarials. Cytotoxicity studies were also done to determine if the compounds would be toxic to the host and to confirm that anti-malarial activity did not correlate with general toxicity. Finally, Chapter 9 makes suggestions for future studies stemming from the findings of the work in this thesis. 23 1.7 References 1. Briand, G. G.; Burford, N., Chem. Rev. 1999, 99, 2601-2658. 2. Shaw, C. F., Chem. Rev. 1999, 99, 2589-2600. 3. Birch, N. J., Chem. Rev. 1999, 99, 2659-2682. 4. Sutton, B. M . ; Hempel, J. C ; Hill, D. T., Curr. Clin. Pract. 1983, 7, 6-16. 5. Maxwell, G.; Hollander, S., J. Med. 1993, 24, 4-5. 6. Jurrison, S.; Berning, D.; Jia, W.; Ma, D., Chem. Rev. 1993, 93, 1137-1156. 7. Orvig, C.; Abrams, M . J., Chem. Rev. 1999, 99. 8. Liu, S.; Edwards, D. S., Ozem. itev. 1999, 99, 2235-2268. 9. Caravan, P.; Ellison, J. J.; McMurry, T. J.; Lauffer, R. B., Chem. Rev. 1999, 99, 2293-2352. 10. Shiue, C.-Y.; Shmitz, A.; Schirrmacher, R.; Shiue, G. G.; Alavi, A., Curr. Med. Chem. 2004, 4, 271-280. 11. Nelson, D. L.; Cox, M . M . , Chapter 9. In Lehninger Principles of Biochemistry. 3rd ed.; Worth: New York, 2000. 12. Ege, S. N., Carbohydrates. In Organic Chemistry Structure and Activity. 3rd ed.; D. C. Heath and Company: Lexington, 1994. 13. Stick, R. V. , Carbohydrates the sweet molecules of life. Academic Press: San Diego, CA, 2001. 14. McAllister, M . S.; Krizanac-Bengez, L.; Macchia, F.; Naftalin, R. J.; Pedley, K. C ; Mayberg, M . R.; Marroni, M. ; Leaman, S.; Stanness, K. A.; Janigro, D., Brain Res. 2001, 409, 20-30. 15. Joost, H.-G.; Thorens, B., Mol. Membrane Biol. 2001,18, 247-256. 24 16. Gould, G. W., Molecular Biology Intelligence Unit. Facilitative Glucose Transporters. Chapman and Hall: Georgetown, TX, 1997. 17. Pardridge, W. M . ; Boado, R. J.; Farrell, C. R., J. Biol. Chem. 1990, 265, 18035-18040. 18. Uldry, M . ; Ibberson, M . ; Hosokawa, M. ; Thorens, B., FEBSLett. 2002, 524, 199-203. 19. Arbuckle, M ; Kane, S.; Porter, L. M.; Seatter, M . J.; Gould, G. W., Biochemistry 1996, 55, 16519-16527. 20. Gould, G. W.; Thomas, H. M. ; Jess, T. J.; Bell, G. I., Biochemistry 1991, 30, 5139-5145. 21. La Manna, J. C ; Harik, S. I., Brain Res. 1985, 326, 299-305. 22. Depre, C ; Rider, M . H.; Hue, L., Eur. J. Biochem. 1998, 258, 277-290. 23. Lucas, A. H.; Apicella, M . A.; Taylor, C. E. , Clin. Infect. Dis. 2005, 41, 705-712. 24. Slovin, S. F.; Keding, S. J.; Ragupathi, G., Immunol. Cell Biol. 2005, 83, 418-428. 25. Dube, D. D.; Bertozzi, C. R., Nature Rev. Drug Disc. 2005, 4, 477-488. 26. Battaglia, G.; Russa, M . L.; Arenare, B. L.; Ippolito, R.; Copani, A.; Bonina, F.; Nicoletti, F., Brain Res. 2000, 860, 149-156. 27. Polt, R.; F, P.; Szabo, L. Z.; Bilsky, E. J.; Davis, P.; Abbrusca, T. J.; Davis, T. P.; Horcath, R.; Yamamura, H.; Hruby, V., Proc. Natl. Acad. Sci. USA 1994, 91, 7114-7118. 28. Bilsky, E. J.; Engleton, R. D.; Mitchell, S. A.; Palian, M . M . ; Davis, P.; Huber, J. D.; Jones, H.; Yamamura, H. I.; Janders, J.; Davis, T. P.; Porreca, F.; Hruby, V. J.; Polt, R., J. Med. Chem. 2000, 43, 2586-2590. 25 29. Fernandez, C ; Nieto, O.; Rivas, E.; Montengro, G.; Fontenla, J. A.; Fernanadez-Mayoralas, A., Carbohydrate Res. 2000, 327, 353-365. 30. Halmos, T.; Santarromana, M . ; Antonakis, K.; Scherman, D., Eur. J. Pharmacol. 1996, 318, 4774S4. 31. Bonina, F.; Puglia, C.; Rimoli, M . G.; Avallone, L. ; Abignente, E.; Boatto, G.; Nieddu, M ; Meli, R.; Amorna, M . ; Caprariis, P. d., Eur. J. Pharm. Sci. 2002,16, 167-174. 32. Smith, T. A., Br. J. Biomed. Sci. 1999, 56, 285-292. 33. Zhang, M . ; Zhang, Z.; Blessington, D.; Li , H.; Busch, T. M . ; Madrak, V.; Miles, J.; Chance, B.; Glickson, J. D.; Zheng, G., Bioconjugate Chem. 2003,14, 709-714. 34. Muhlhausen, M . ; Schirrmacher, R.; Piel, M. ; Lecher, B.; Briefert, M . ; Piee-Staffa, A.; Kaina, B.; Rosch, F., J. Med. Chem. 2006, 49, 263-272. 35. Vaidyanathan, G.; Affleck, D. J.; Schottelius, M. ; Wester, H.; Friedman, H. S.; Zalutsky, M . R., Bioconjugate Chem. 2006,17, 195-203. 36. Pedersen, P. L.; Mathupala, S.; Rempel, A.; Geschwind, J. F.; Ko, Y. H., Biochim. Biophys. Acta 2002, 7555, 14-20. 37. Cheng, H.; Cao, X.; Xian, M . ; Fang, L.; Cai, T. B.; Ji, J. J.; Tunac, J. B.; Sun, D.; Wang, P. G., J. Med. Chem. 2005, 48, 645-652. 38. Groot, F. M . d.; Damen, E. W.; Scheeren, H. W., Curr. Med. Chem. 2001, 8, 1093-1122. 39. Robinson, M . A.; Charlton, S. T.; Gamier, P.; Wang, X.-T.; Davis, S. S.; Perkins, A. C ; Frier, M . ; Duncan, R.; Savage, T. J.; Wyatt, D. A.; Watson, S. A.; Davis, B. G., Proc. Natl. Acad. Sci. USA 2004,101, 14527-14532. 26 40. Giancoli, D. C., Nuclear physics and radioactivity. In Physics Principles with Applications Prentice Hall: Englewood Cliffs, NJ, 1995; pp 857-881. 41. Mausner, L. F.; Mirzadeh, S., Reactor production of radionulcides. In Handbook of Radipharmaceuticals Radiochemistry and Applications Welch, M . J.; Redvanly, C. S., Eds. John Wiley and Sons Ltd.: West Sussex, England, 2003; pp 87-118. 42. Ruth, T. J., Accelerators available for isotope production. In ref.41, 2003; pp 71-86. 43. Schyler, D. J., Production of radionuclides in accelerators. In ref. 41, 2003; pp 1-70. 44. Fowler, J. S.; Wolf, A. P., Acc. Chem. Res. 1997, 30, 181-188. 45. Dilworth, J. R.; Parrott, S. J., Chem. Soc. Rev. 1998, 27, 43-55. 46. Luttropp, C. A.; Jackson, J. A.; Jones, B. J.; Sohn, M.-H.; Lynch, R. E.; Morton, K. A., J. Nucl. Med. 1998, 39, 1405-1411. 47. Jurisson, S. S.; Lydon, J. D., Chem. Rev. 1999, 99, 2205-2218. 48. Roy, R. K.; Katzenellenbogen, J. A., Nucl. Med. Biol. 1997, 24, 485-498. 49. Liu, S.; Edwards, D. S.; Barret, J. A., Bioconjugate Chem. 1997, 8, 664-672. 50. Knight, L. C , Radiolabeled peptides for tumor imaging. In ref. 41, 2003; pp 643-684. 51. Zalutsky, M . R.; Lewis, J. S., Radiolabeled antibodies for tumor imaging and therapy. In reference 41 2003; pp 685-714. 52. Katzenellenbogen, J. A., Receptor imaging of tumors (non-peptide). In ref. 41 2003. 27 53. de Jong, M ; Breeman, W. A.; Bakker, W. H.; Kooij, P. P.; Bernard, B. F.; Hofland, L. J.; Visser, T. J.; Srinivasan, A.; Schmidt, M . A.; Erion, J. L.; Bugaj, J. E.; Macke, H. R.; Krenning, E. P., Cancer Res. 1998, 58, 4 3 7 - 4 4 1 . 54. Krohn, K. A.; Mankoff, D. A.; Muzi, M. ; Link, J. M . ; Spence, A. M . , Nucl. Med. Biol. 2005,52,663-671. 55. Beuthien-Baumann, B.; Hamacher, K.; Oberdorfer, F.; Steinbach, J., Carbohydrate Res. 2000, 327, 107-118. 56. Adam, M . J., J. Label. Compd. Radiopharm. 2002, 45, 167-180. 57. Fowler, J. S.; Ido, T., Design and synthesis of 2-deoxy-2-[18F]fluoro-D-gluocse ( 1 8FDG). In ref. 41, 2003; pp 307-322. 58. Segall, G., Nucl. Med. Comm. 2002, 23, 323-330. 59. Barrington, S. F., Nucl. Med. Comm. 2000, 21, 237-240. 60. Czernin, J., Acta Medica Austrica 2002, 5, 162-170. 61. Scott, A. M . , PET imaging in oncology. In Positron Emission Tomography Basic Science. Bailey, D. L.; Townsend, D. W.; Valk, P. E.; Marsey, M . N. , Eds. Springer-Verlag: London, 2005; pp 311-326. 62. Hustinx, R.; Alavi, A., Tumor imaging. In ref. 41, 2003; pp 629-641. 63. Chiro, D. I.; Brooks, R. A., J. Nucl. Med. 1988, 29, 421-423. 64. Elshaikh, M. ; Ljungman, M . ; Haken, R. T.; Lichter, A. S., Annu. Rev. Med. 2006, 57, 19-31. 65. Kassis, A. I.; Adelstein, S. J., Consideration in the selection of radionuclides for cancer therapy. In ref. 41, 2003. 66. Soloway, A. H.; Tjarks, W.; Barnum, B. A.; Rong, F.-G.; Barth, F. F.; Codogni, I. M. ; Wilson, J. G., Chem. Rev. 1998, 98, 1515-1562. 28 67. Hawthorne, M . F.; Maderna, A., Chem. Rev. 1999, 99, 3421-3434. 68. Parrot, M . C ; Marchington, E. B.; Valliant, J. F.; Andronov, A., J. Am. Chem. Soc. 2005, 727, 12081-12089. 69. Valliant, J. F.; Guenther, K. J.; King, A. S.; Morel, P.; Schaffer, P.; Sogbein, O. O.; Stephenson, K. A., Coord. Chem. Rev. 2002, 232, 173-230. 70. Volkert, W. A.; Hoffman, T. J., Chem. Rev. 1999, 99, 2269-2292. 71. Kealy, T . J.; Paulson, P. L. , Nature 1951,168, 1039-1040. 72. Fischer, E. O.; Pfab, W., Z Naturforch. 1952, 7b, 377-379. 73. Eiland, P. F.; Pepinsky, R., J. Am. Chem. Soc. 1952, 74, 4971. 74. Dunitz, J. D.; Orgel, L. E. , Nature 1953,171, 121-122. 75. Hermann, W. A., Ferrocene as a gasoline and fuel additive. In Applied Homogenous catalysis with organometallic compounds 2nd ed.; Wiley-VCH Verlag: Weinheim, Germany, 2002; pp 586-590. 76. Manners, I., Chem. Comm. 1999, 857-865. 77. Albinati, A.; Pregosin, P. S.; Wick, K., Organometallics 1996,15, 2419-2421. 78. Dai, L. X.; You, S. L.; Deng, W. P.; Hou, W. L., Acc. Chem. Res. 2003, 36, 659-667. 79. Nazarov, A. A.; Christian G. Hartinger; Arion, V. B.; Giester, G.; Keppler, B. K., Tetrahedron 2002, 58, 8489-8492; 80. Seitzberg, J. G.; Dissing, C ; Sotofte, I.; Norrby, P. O.; Johannsen, M . , J. Org. Chem. 2005, 70, 8332-8337. 81. Togni, A., Ferrocenes: Homogeneous Catalysis, Organic Synthesis, Materials Science. V C H : Weinheim, 1995. 82. Togni, A., Angew. Chem. Int. Ed. 1996, 35, 1475-1477. 29 83. Cabera, S.; Mancheno, O. G.; Arrayas, R. G.; Alonso, I.; Mauleon, P.; Carretero, J. C, Pure Appl. Chem. 2006, 78, 257-265. 84. Atkinson, R. C. J.; Gibson, V. C ; Long, N. J., Chem. Soc. Rev. 2004, 33, 313-328. 85. Choi, S.-J.; Choi, B.-G.; Park, S.-M., Anal. Chem. 2002, 74, 1998-2002. 86. Forrow, N. J.; Foulds, N. C ; Frew, J. E.; Law, J. T., Bioconjugate Chem. 2004, 15, 134-144. 87. Immoos, C. E.; Lee, S. J.; Grinstaff, M . W., J. Am. Chem. Soc. 2004,126, 10814-10815. 88. Liu, J.; Tian, S.; Tiefenauer, L.; Nilesen, P. E.; Knoll, W., Anal. Chem. 2005, 77, 2756-2761. 89. Tarraga, W.; Molina, P.; Lopez, J. L.; Velasco, M . D.; Bautisla, D.; Jones, P. G., Organometallics 2002, 21, 2055-2065. 90. Tompatanaget, F.; Tuntulani, T.; Chailapakul, O., Org. Lett. 2003, 5, 1539-1542. 91. Willner, I.; Katz, E. , Angew. Chem. Int. Ed. 2000, 39, 1180-1218. 92. Yu, C. J.; Wang, H.; Yowanta, Y.; Kim, J. C ; Donilon, L. H.; Tao, C ; Strong, M. ; Chong, Y., J. Org. Chem. 2001, 66, 2937-2942. 93. Fisher, R. H.; Jaouen, G., Organometallics 2003, 22, 2166-2177. 94. Staveren, D. R. V.; Metzler-Nolte, N., Chem. Rev. 2004,104, 5931-5985. 95. Kraatz, H.-B.; Lusztyk, J.; Enright, G. D., Inorg. Chem. 1997, 36, 2400-2405. 96. Kraatz, H.-B.; Leek, D. M. ; Houmam, A.; Enright, G. A.; Lusztyk, J.; Wayner, D. M. , J. Organomet. Chem. 1999, 598, 38-49. 97. Moriuchi, T.; Yoshida, K.; Hirao, T., J. Organomet. Chem. 2003, 690, 1750-1755. 30 98. Saweczko, P.; Enright, G. D.; Kraatz, H.-B., Inorg. Chem. 2001, 40, 4409-4419. 99. Xu, Y.; Kraatz, H.-B., Tetrahedron Lett. 2001, 42, 2601-2603. 100. Xu, Y.; Saweczko, P.; Kraatz, H.-B., J. Organomet. Chem. 2001, 637-639, 335-342. 101. Bucci, E.; Napoli, L. D.; Fabio, G. D.; Messere, A.; Montesarchio, D.; Romanelli, A.; Piccialli, G.; Varra, M . , Tetrahedron 1999, 55, 14435-14450. 102. Kopf-Maier, P.; Kopf, H., Chem. Rev. 1987, 87, 1137. 103. Duivenvoorden, W. C. M ; Liu, Y.-N.; Schatte, G.; Kraatz, H.-B., Inorg. Chim. Acta 2005, 358, 3183-3189. 104. Osella, D.; Ferrali, M. ; Zanello, P.; Laschi, F.; Fontani, M . ; Nervi, C.; Cavigiolio, G., Inorg. Chim. Acta 2000, 306, 42-48. 105. Vessieres, A.; Top, S.; Pigeon, P.; Hillard, E.; Boubeker, L.; Spera, D.; Jaouen, G.,J. Med. Chem. 2005, 48, 3937-3940. 106. Hartinger, C. G.; Nazarov, A. A.; Arton, V. B.; Giester, G.; Jakupec, M . ; Galanski, M. ; Keppler, B. K., New J. Chem. 2002, 26, 671-673. 107. Snegur, L. V.; Simenel, A. A.; Nekrasov, Y. S.; Morzova, E. A.; Starikova, Z. A.; Peregudova, S. M. ; Kuzmenko, Y. V.; Babin, V. N.; Ostrovskaya, L. A., J. Organomet. Chem. 2004, 689, 2473-2479. 108. Chen, Y.; Liu, H.-Y.; Fu, S.-L., Yingyong Huaxue 2002,19, 603-605. 109. Wasi, N.; Singh, H. B.; Gajanana, A.; Raichowdhary, A. N., Inorg. Chim. Acta 1987,135, 133-137. 110. Kraatz, H.-B.; Galka, M . , Met. Ions Biol. Syst. 2001, 38, 385-409. 111. Beilstein, A. E.; Grinstaff, M . W., J. Organomet. Chem. 2001, 398-406. 31 112. Abbott, N. L.; Jewell, C. M . ; Hays, M . E.; Knodo, Y.; Lynn, D. M . , J. Am. Chem. Soc. 2005,727, 11576-11577. 113. Top, S.; Tang, J.; Vessieres, A.; Carrez, C ; Provot, C ; Jaouen, G., Chem. Comm. 1996, 955-956. 114. Top, S.; Vessieres, A.; Leclerq, G.; Quivy, J.; Tang, J.; Vaissermann, J.; Huche, M. ; Jaouen, G., Chem. Eur. J. 2003, 9, 5223-5236. 115. Korenromp, E. L. Malaria Incidence Estimates at Country Level for the Year 2004-proposed estimates and draft report; World Health Organization Roll Back Malaria Monitoring and Evaluation Reference Group: 2004. 116. Sherman, I. W., The Biochemistry of Malaria-An Overview. Taylor and Francis: London, 1991. 117. Roth, E. , Blood Cells 1990,16, 453-466. 118. Krishna, S.; Eckstein-Ludwig, U.; Joet, T.; Uhlemann, A . - C ; Morin, C ; Webb, R.; Woodrow, C. K., J. F. J.; Kremsner, P. G., Int. J. Parasitol. 2002, 32, 1567-1573. 119. Hyde, J. E. , Trends Parasitol. 2005, 21, 494-498. 120. Waters, A., Cell 2006,124, 689-693. 121. Ridley, G.R. , Nature 2002, 415, 686-693. 122. Atteke, C ; Ndong, J. M . M. ; Aubouy, A.; Maciejewski, L.; Brocard, J.; Lebibi, J.; Deloron, P., J. Antimicrob. Chem. 2003, 51, 1021-1024. 123. Beagely, P.; Blackie, M . A. L.; Chibale, K.; Clarkson, C ; Meijboom, R.; Moss, J. R.; Smith, P. J.; Su, H., Dalton Trans. 2003, 3046-3051. 124. Beagley, P.; Blackie, M . A. L.; Chibale, K.; Clarkson, C ; Moss, J. R.; Smith, P. J., J. Chem. Soc, Dalton Trans. 2002, 4426-4433. 32 125. Biot, C ; Delhaes, L.; Abessolo, H.; Dormarle, O.; Maciejewski, L. A.; Mortuaire, M.; Delcourt, P.; Deloron, P.; Camus, D.; Dive, D.; Brocard, J. S., J. Organomet. Chem. 1999,559,59-65. . 126. Biot, C ; Delhaes, L.; Maciejewski, L. A.; Mortuaire, M . ; Camus, D.; Dive, D.; Brocard, J. S., Eur. J. Med. Chem. 2000, 35, 707-714. 127. Biot, C ; Dessolin, J.; Ricard, I.; Dive, D., J. Organomet. Chem. 2004, 689, 4678-4682. 128. Biot, C ; Glorian, G.; Maciejewski, L. A.; Brocard, J. S., J. Med. Chem. 1997, 40, 3715-3718. 129. Delhaes, L.; Abessolo, H.; Biot, C ; Berry, L.; Delcourt, P.; Maciejewski, L. ; Brocard, J.; Camus, D.; Dive, D., Parasitol. Res. 2001, 87, 239-244. 130. Harpstrite, S. A.; Beatty, A. A.; Collins, S. D.; Oksman, A.; Goldberg, D. E.; Sharma, V., Inorg. Chem. 2003, 42, 2294-2300. 131. Navarro, M . ; Perez, H.; Sanchez-Delgado, R. A., J. Med. Chem. 1997, 40, 1937-1939. 132. Navarro, M . ; Vasquez, F.; Sanchez-Delgado, R. A.; Perez, H.; Sinou, V.; Schrevel, J., J. Med. Chem. 2004, 47, 5204-5209. 133. Sanchez-Delgado, R. A.; Navarro, M. ; Perez, H.; Urbina, J. A., J. Med. Chem. 1996, 39, 1095-1099. 134. Sharma, V.; Piwnica-Worms, D., Chem. Rev. 1999, 99, 2545-2560. 135. Werbovetz, K. A., Curr. Med. Chem. 2000, 7, 835-860. 136. Joet, T.; Eckstein-Ludwig, U.; Morin, C ; Krishna, S., Proc. Natl. Acad. Sci. USA 2003,100, 7476-7479. 137. Joet, T.; Krishna, S., Acta Tropica 2004, 89, 371-374. 33 Chapter 2 Carbohydrate-appended 3-Hydroxy-4-pyridinone Complexes of the [M(CO)3] + Core (M = Re, Tc)* 2.1 Introduction Because glucose is a major energy source for the body, especially for the heart and brain, the ability to monitor cell glucose metabolism has found utility in a myriad of areas of medicine.1 Utilizing positron emission tomography, [18F]-2-deoxy-2-fluoro-D-glucose (FDG) is the preeminent radiotracer used to image glucose metabolism2 and is an important tool in cardiology and neurology. ' As well, FDG is used in oncology as a tumor marker, due to the elevated level of glucose metabolism in many tumor types.1'4 Unfortunately, the radioisotope used in FDG, 1 8 F , is cyclotron produced and has a short half-life (t\a = 1.8 h), significantly limiting the availability of the radiotracer. Hence, there is much interest in developing FDG-like radiopharmaceuticals that are less expensive and more accessible. As well, no analog of FDG that can be used in radiotherapy is currently available. In developing a more accessible radiotracer, 9 9 m T c is an optimal radioisotope to consider; 9 9 m T c is already used in over 85% of nuclear medicine scans and is generator produced making it inexpensive and widely available.5 The physical properties of 9 9 m T c * A version of this chapter has been accepted for publication. Ferreira, C. L., Bayly, S. R., Green, D. E., Storr, T., Barta, C. A., Steele, J. S., Adam, M. J., Orvig, C. Carbohydrate appended 3-hydroxy-4-pyridinone complexes of the M(CO)3+ core (M = Re, 9 9 mTc, 186Re). Bioconjugate Chem. 2006, In press. 34 (ti/2 = 6.01 h, y = 142.7 keV) are ideal for imaging with single photon emission computed tomography (SPECT), the infrastructure for which is already available in most major hospitals. In addition, the third row congener of Tc, Re, has relatively similar chemistry to that of Tc, and has two particle emitting radioisotopes of interest for therapy: 1 8 6 Re (ti/2 = 3.68 d, p\" - 1.07 MeV, y = 137 keV) and 1 8 8 Re (t1/2 = 16.98 h, p - = 2.12 MeV, y = 155 keV). Clearly, the development of an FDG analog which incorporates 9 9 m T c could have a significant impact on health care cost and availability, while using the same methods to develop a Re-FDG analog has potential for targeted radiotherapy. Research to combine the accessibility and ideal nuclear properties of 9 9 m T c with the biological properties and applications of FDG has led several groups to develop 9 9 m T c glucose conjugates.6\"16 The initial investigations of these conjugates suffered from poor characterization,8'12 or formed undesirable bonding interactions between the carbohydrate and metal. 7 , 1 7' 1 8 More recently, well characterized ligands and complexes have been be prepared, the ligands radiolabeled in high yields, and the complexes shown to be highly stable. 9'\"' 1 4 - 1 6 Also, in vivo studies of a few of these compounds have shown promising biodistribution results and tumour uptake,8'12 although the mechanism of uptake does not appear to mimic that of FDG or glucose. While non-radioactive Re compounds are commonly used as a surrogate for 9 9 m T c on the macroscopic scale and as a standard for identifying the analogous 9 9 m T c complexes, only two 1 8 6 Re glucose conjugates have been reported.14'17 Further research is needed to develop glucose conjugates of 9 9 m T c , 1 8 6 Re and 1 8 8 Re with high stability and optimal biological properties. 35 Figure 2.1. Structures of FDG, [Tc(CO) 3(H 20) 3] +, and a 3-hydroxy-4-pyridinone. In the development of 9 9 m T c and 1 8 6 / 1 8 8 R e glucose conjugates, the [M(CO) 3] + core is of interest due to its stability, small size, and well-studied chemistry.19\"27 The [M(CO) 3] + core is kinetically inert due to the metal ion +1 oxidation state and low spin d 6 electron configuration, which is further stabilized by the facial and strongly back bonding CO ligands. The compact core is advantageous in retaining biological activity when labelling small molecules such as glucose. Alberto and co-workers have pioneered the chemistry and radiochemistry of this core and have developed a commercially available kit preparation of the [ 9 9 m Tc(CO) 3 (H 2 0) 3 ] + species. The three aqua ligands, each trans to a CO ligand, are labilized for substitution by various chelates functionalized with the biomolecule of interest.19\"21'24'26 The [ 1 8 8Re(CO) 3(H 20) 3] + analog has also been described.22'23 Different chelating derivatives have been studied with the [M(CO) 3] + core, to optimize stability, labelling, and the physical properties of the products. Both bidentate and tridentate ligands have been investigated with a broad range of hard and/or soft bases as binding groups, including amine, 2 0' 2 1' 2 6' 2 8\" 3 3 thiol/thioether,2 8'3 0'3 1'3 4'3 5 thiourea,36 carboxylate,9'30'37 imine, 3 8' 3 9 phosphine,40 carborane13'41 and cyclopentadienyl42\"45 ligands. Most of these ligands can be efficiently labeled via the [ 9 9 m Tc(CO) 3 (H 2 0) 3 ] + precursor and form highly stable complexes of varying charge and lipophilicity. 36 Aromatic amines and borderline bases are considered among the best binding functionalities for this core, 2 1' 3 3 but do not neutralize the 1+ charge of the metal center. A highly useful class of chelators that has not been studied with the [M(CO)3] + system are the 3-hydroxy-4-pyridinones, small, monoanionic, bidentate 0,0- chelators with a high affinity for a variety of metal ions. These chelators can be easily functionalized via the ring nitrogen. 3-Hydroxy-4-pyridinones have similar basicity to aromatic amines, but can neutralize the charge of a 1+ metal center, such as Re(I) and Tc(I). Stable complexes with Re/Tc(V) 4 6 and (IV) 4 7 with 3-hydrOxy-4-pyridinones have been reported. 3-Hydroxy-4-pyridinones have also been used to chelate a large number of other metals, both hard and soft acids, including V , 4 8 Fe , 4 8 ' 4 9 Co , 5 0 C u , 5 1 ' 5 2 M n , 5 3 Zn, 5 4 Pb, 5 4 Rh, 5 5 R u , 5 5 ' 5 6 Ir,5 5 A l , 4 9 ' 5 7 ' 5 8 Ga, 4 9 ' 5 7 \" 6 0 In, 5 7 ' 5 9 U , 6 1 and G d . 6 2 Previously in the Orvig group, several pyridinone bearing pendant glucose derivatives have been prepared (Figure 2.2).59 A 3-hydroxy-4-pyridinone with a pendant glucosamine moiety, Feralex-G (Figure 2.2), has also been reported,63 and an improved synthesis of Feralex-G is described here. With the [M(C0 )3] + core, the pyridinone pro-ligand is expected to form a stable neutral complex. When labelling a small neutral biomolecule such as glucose, a neutral complex may better retain the biological properties of the biomolecule compared to a charged complex. In this work, we investigate the utility of the aforementioned pyridinone pro-ligands bearing pendant glucose derivatives as chelates for the [M(C0)3] + core. The resulting neutral complexes were fully characterized on the macroscopic scale, as well as prepared on the tracer level with both 9 9 m T c and 1 8 6Re. Preliminary evaluation of the complexes as potential radiopharmaceuticals is also reported. 37 Figure 2.2. Carbohydrate appended 3-hydroxy-4-pyridinone pro-ligands. 2.2 Experimental 2.2.1 Materials and Methods [NEt4]2[Re(CO)3Br3], 4 [Re(CO)3(H20)3]Br , 3-benzyloxy-l-(carboxymethyl)-2-methyl-4-(l//)-pyridinone (l) 5 9 and H L 1 \" 4 5 9 ' 6 3 were prepared as reported in the literature. Re(CO)5Br was purchased from Strem Chemicals (Newburyport, MA). The Isolink boranocarbonate kits were a gift from Mallinckrodt Inc. Saline solutions of Na 9 9 mTcC>4 186 and Na ReC>4 (>500 mCi/mg specific activity) were obtained from Vancouver Coastal Health-UBC Hospital and MDS Nordion Inc., respectively. All other chemicals were purchased from Sigma Aldrich and used without further purification. Water was deionized, purified (Barnstead D9802 and D9804 cartridges) and distilled with a Corning MP-1 Mega-Pure Still. All solvents were HPLC grade and purchased from Fisher. N 2 38 gas was acquired from Praxair. Thin layer chromatography (TLC) was performed using silica T L C plates with aluminum backing (Merck), and silica column chromatography was performed using silica from Silicycle (Quebec City, PQ). All NMR solvents were purchased from Cambridge Isotope Laboratories. ' J H NMR spectra were recorded on Bruker AV-300 or AV-400 instruments at 300.13, or 400.21 MHz, respectively. The NMR spectra were calibrated with the deuterated solvent used in each case, and ' H - ' H COSY and ' H - ^ C H M Q C and HMBC 2D NMR spectra were used to aid in the characterization of the compounds. Infrared spectra were recorded on an ATI Mattson Galaxy\u00E2\u0084\u00A2 Series FTIR 5000 spectrophotometer as thin films on NaCl plates or as KBr discs. Mass spectra were obtained on either a Kratos Concept II H32Q instrument (Cs+, LSIMS) or a Macromass L C T (electrospray, ES-MS). Elemental analysis was performed at the University of British Columbia Chemistry Department by Mr. M . Lakha (Carlo Erba analytical instrument). UV-visible spectra and hexokinase inhibition assays were run on a Hewlett-Packard model HP8453 diode array spectrophotometer, equipped with a kinetics software package. Purification by semi-preparative HPLC was done on a Water Xterra RP CI8 column (7 urn, 19 x 300 mm) using a Waters WE 600 HPLC system equipped with a 2478 dual wavelength absorbance U V detector and the Empower software package. Verification of radiolabeled products was done by HPLC analysis on a Phenomenex Hydro-Synergi CI8 RP column using a Wellchrom Knauer K1001 system equipped with a K-2501 U V absorption detector, a Kapintek radiometric well counter and Peak simple software. 39 2.2.2 Synthesis of H L 5 1 - {N-[2-Amino-2-deoxy-D-glucopy ranose] ethanamide}-2-methy 1-3-benzyloxy-4(l//)-pyridinone (2) hydrochloride salt (728 mg, 3.4 mmol), dicyclohexylcarbodiimide (DCC) (744 mg, 3.6 mmol), and N-hydroxysuccinamide (NHS) (377 mg, 3.3 mmol). Dimethylformamide (10 mL) was added via a syringe and the mixture stirred into a slurry. Diisopropylethylamine (1.3 mL, 7.5 mmol) was added to the stirring mixture and the reaction was heated at 80\u00C2\u00B0C for 16 hours. The solvent and excess base were removed on a rotary evaporator, and the resulting dark residue was redissolved in water. Insoluble impurities were removed by filtration and the filtrate was reduced to a thick orange syrup, which was purified by silica chromatography eluting with 2:2:1 EtOAc:CH3CN:MeOH. Fractions containing the product were recrystallized in MeOH to give a white solid (2) (675 mg, 54%). 'H NMR (CD 3 OD:D 2 0 1:1, 400 MHz) a anomer: 6 2.00 (s, 3H, COC//3), 3.45 (m, 1H, H3) 3.72-3.86 (m, 4H, H4, H5, H6) 3.89 (d, 1H, #14, 3J 1 4,i5 = 7.6Hz) (dd, 1H, HI, J= 11.27 Hz, J= 1.52 Hz) 4.81 (s, 2H, H8), 5.02 (s, 2H, H\5), 5.16 (d, 1H, HI, 3Jh2 = 3.96 Hz) 6.56 (d, 1H, H12, V 1 3 , i 2 = 7.30 Hz) 7.4 (m, 5H, Aryl) 7.65 (d, 1H, #13, 3J 1 2,i3 = 7.31 Hz); p anomer: 5 2.06 (s, 3H, COC//3), 3.42 (m, 2H, H3, H4) 3.55 (m, 1H, H5) 3.68 (dd, 1H, HI, 3 J = 8.22 Hz) 3.72-3.87 (m, 2H, H6) O H This reaction was done under dry conditions using pyridinone (1) (769 mg, 2.8 mmol), glucosamine benzyloxy-l-(carboxymethyl)-2-methyl-4-(l//)-Schlenck line techniques. To a 100 mL RB was added 3-40 4.70 (d, 1H, HI, 3Ji,2 = 8.22 Hz) 4.81 (s, 2H, H8) 5.02 (s, 2H, #15) 6.56 (d, 2H, H12, V,2,i3 = 7-30 Hz) 7.4 (m, 5H, Aryl) 7.65 (d, 1H, #13, V.3,12 = 7.31 Hz). nC{lH) NMR (CD 3 OD:D 2 0 1:1, 100 MHz) for the a anomer: 5 13.15 (C14), 55.40 (C2), 56.92 (C8), 61.73 (C6), 71.30 (C3), 71.80 (CA), 72.62 (C5), 75.13 (CI5), 91.8 (CI), 116.98 (C12), 129.59, 129.70, 129.73, 130.40, 130.53 (C17-C21), 137.07 (C9) 143.01 (C13), 145.81 (CI6), 146.70 (CIO), 169.09 (C7), 174.90 (Cll); p anomer: 5 13.10 (CI4), 57.17 (C8), 58.11 (CI), 61.84 (C6), 71.09 (C3), 74.80 (C5), 77.11 (C4), 75.10 (C15), 95.8 (CI), 116.98 (C12), 129.59, 129.70, 129.73, 130.40, 130.53 (C17-C21), 137.14 (C9) 142.98 (C13), 145.81 (C16), 146.70 (CIO), 169.43 (C7), 174.87 (Cll) . MS (ESI+): m/z (relative intensity) = 457 ([M+Na]+, 100), 435 ([M+H]+, 10). Anal. Calcd. for C 2 i H 2 6 N 2 0 8 : C, 58.06; H, 6.03; N, 6.45. Found: C, 57.70; H, 6.08; N, 6.40. l-{N-[2-Amino-2-deoxy-D-glucopyranose]ethanamide}-2-methyl-3-hydroxyl-4(l#)-pyridinone hydrate ( H L 5 H 2 0 , Feralex-GH20) The starting material 2 (450 mg, 1.0 mmol) was dissolved in MeOH (30 mL) and water (30 mL) in a 250 mL round bottom flask. 10% Pd/C (115 mg, 0.11 mmol) was wetted with water (2 mL) and added to the flask, which was then evacuated: The round bottom flask was fitted with an adapter which was opened to a connected hydrogen gas filled balloon. The reaction was stirred rapidly under the hydrogen atmosphere for 12 h. The catalyst was removed by filtration and the filtrate solvent removed on a rotary evaporator (40\u00C2\u00B0C) to give a quantitative white crystalline product (358 mg, 100%). *H NMR (CD 3 OD:D 2 0 1:1, 400 MHz) a anomer: 8 41 2.33 (s, 3H, COC# 3), 3.44 (m, 1H #3) 3.74-3.86 (m, 4H, #4, H5, H6) 3.91 (dd, 1H, HI, 2J= 12.35 Hz) 4.90 (s, 2H, #8), 5.19 (d, 1H, HI, 3JU2 = 3.50 Hz) 6.49 (d, 1H, #12, 3 y 1 2 ; l 3 = 7.16 Hz) 7.59 (d, 1H, #13, Vi3,i2 = 7.14 Hz); p anomer: 8 2.34 (s, 3H, COC# 3), 3.41 (m, 2H, #3, #4) 3.55 (m, 1H, #5) 3.67 (dd, 1H, #2, 2J= 8.40 Hz) 3.72-3.87 (m, 2H, #6) 4.71 (d, 1H, #1, V l j 2 = 8.38 Hz) 4.88 (s, 2H, #8) 6.49 (d, 2H, #12, 3 J n ,n = 7.16 Hz) 7.61 (d, lH,#13 , 3 Ji 3 , i2 = 7.16Hz). 1 3C{'H} NMR (CD 3 OD:D 2 0 1:1, 100 MHz) a anomer: 5 12.04 (C14), 55.21 (C2), 56.66 (CS), 61.53 (C6), 71.13 (C3), 71.63 (CA), 72.39 (C5), 91.54 (CI), 112.81 (C12), 134.72 (C9) 140.48 (C13), 145.63 (CIO), 169.33 (C7), 170.57 (Cll); (3 anomer: 5 11.98 (C14), 56.96 (C8), 57.96 (C2), 61.64 (C6), 70.92 (C3), 74.62 (C5), 76.93 (C4), 95.64 (CI), 112.81 (C12), 134.78 (C9) 140.42 (C13), 145.63 (CIO), 169.04 (C7), 170.54 (CI 1). MS (ESI+): m/z (relative intensity) = 345 ([M+H]+, 100). Anal. Calcd. for C i 4 H 2 o N 2 0 8 H 2 0 : C, 46.41; H, 6.12; N, 7.73. Found: C, 46.44; H, 6.01; N, 7.45. 2.2.3 General Synthesis of RefL 1 5)(CO) 3(H 20) Method A: A 1:1 stoichiometric mixture of Re(CO)sBr and HL 1 \" 5 was dissolved in borate buffer (10-20 mL, 20 mM, pH 9), stirred and heated at 75\u00C2\u00B0C for 30 min. The solvent was removed using a rotary evaporator to yield a white solid. The reaction was quantitative by ] H N M R spectroscopy. 42 Method B: A 1:1 stoichiometric mixture of [Et,N]2[Re(CO)3Br3] and HL 1 \" 5 was dissolved in methanol (10-20 mL) and refluxed for 2 h. The solvent was removed using a rotary evaporator to yield a white solid. Method C: A 1:1 stoichiometric mixture of [Re(CO)3(H20)3]Br and HL 1 \" 5 was dissolved in methanol (10-20 mL) and refluxed for 2 h. The solvent was removed using a rotary evaporator and pure product was obtained by recrystallization, column chromatography, or semi-preparative HPLC. 2.2.4 Synthesis of Re(L1'5)(CO)3(H20) by Method C Fac-(aqua)tri(carbonyl)(l-{N-[p-benzyI [3-D-glucopyranoside]ethanamide}-2-methyl-3-oxy-4(l#)-pyridinato)rhenium(I), (R eL 1(CO) 3(H 20)) OH The title compound was prepared 0.294 mmol) in methanol (20 mL). The product was purified by semi-preparative HPLC (30 mL/min gradient method 95% water 5% methanol to 100% methanol over 30 min.) with the product collected between 11.5 and 12.5 min. Solvent was removed by rotary evaporation and the resulting white solid dried in vacuo overnight (90 mg, 44%). 'H NMR (CD 3 OD:D 2 0 1:1, 400 MHz): 8 2.48 (s, 3H, C13-C7/3), 3.35-3.57 (m, 4H, HI, H3, HA, H5), 3.72 (dd, 1H, //6a, 3 J 5 j 6 = 5.63 Hz, 2J6^6b = 12.2 Hz), 3.89 (d, 1H, //6b, J= 10.51 Hz), 4.98 (d, 1H, HI, V i , 2 = 7.22 Hz), 5.12 (s, 2H, H12), 6.68 (d, 1H, #16, 3 Ji 6 , i7 = OH2 ' s | ,\u00C2\u00A30 from H L (127 mg, 0.291 mmol) \u00E2\u0080\u00A2T I 0 0 and[R (CO)3(H20)3]Br(119mg, 43 6.55 Hz), 7.12 (d, 2H, HB, HM', % 9 = 8.87 Hz), 7.46 (d, 2H, H9, H9', % 9 = 9.05 Hz), 7.59 (d, l H , / / 1 7 , 3 J 1 6 , i 7 = 6.54Hz). 13C{H} NMR (CD 3 OD:D 2 0 1:1, 75 MHz): 8 11.68 (C13-CH3), 56.2 (C12), 60.69, (C6), 69.69 (CA), 73.21 (C2), 76.58 (C3), 76.99 (C5), 100.72 (CI), 109.15 (C16), 116.57 (C8, C8'). 120.49 (C9,C9'), 132.56 (CIO), 134.55 (C13), 135.09 (C17), 158.65 (C14), 164.34 (CI 1), 174.06 (C15), 199.28, 199.99, 200.32 (3 CO). IR (KBr disc, cm\"1): 3393 (br) (vOH), 2013 (vs), 1884 (vs) (v(/ac-Re(CO)3)), 1686 (m), 1545 (s), (v(CONH)), 1605 (s), (v(CO)), 1508 (s), 1478 (s), (aryl), 1358 (s), 1285 (m), 1227 (m, br) (8(OH)), 1073 (s, br), (v(COH)), 835 (m), (para-substituted benzene). UV-visible spectrum (H 20): A,max/nm ( E m a x / L M\" 1 cm -1) = 3 1 9 (6.2 x 103), 276 (1.2 x 104). MS (ESI+): m/z (relative intensity) = 729, 727 ([M-H20+Na]+, 100). MS (ESI-): m/z (relative intensity)= 706, 704 ([M-H 20]\ 75). Anal. Calcd. for C 2 3 H 2 5 N 2 O i 3 R e : C, 38.18; H, 3.50; N, 3.87. Found: C, 38.00; H, 3.90; N, 4.20. Fac-(aqua)tri(carbonyl)(l - {N- [p-benzy 1 (2,3,4,6-tetra-O-acety l-(3-D-glucopyranoside)]ethanamide}-2-methyl-3-oxy-4(l/0-py\u00C2\u00BB\"idinato)rhenium(I), (ReL2(CO)3(H20)CH3OH) The title compound was prepared by 1 6 \u00C2\u00AB o ?H 2 refluxing H L 2 (61 mg, 0.10 mmol) ^N^NNC^O^NO and [Re(CO)3(H20)3]Br (40 mg, H 12 | 1J CO CH3 0.10 mmol) in methanol (8 mL). The product was purified by alumina column chromatography eluting with 100% methanol. The fractions containing product were reduced on a rotary evaporator. To remove any alumina present from the column, the residual solid was dissolved in ethyl 44 acetate and passed through a 0.22 urn syringe. The solvent was removed using a rotary evaporator to yield the product as a white solid, which was dried under reduced pressure overnight (57 mg, 65%). ' H N M R (CD 3 OD:CD 3 CN 1:1, 400 MHz): 8 2.01, 2.02, 2.05, 2.06 (4s, 12H, COOC# 3 ), 2.42 (s, 3H, C13-C#3), 4.17 (m, 1H, #5), 4.29 (d, 1H, #6a, 2 J 6 a , 6 b = 12.32 Hz), 4.42 (dd, 1H, //6b, 3 y 5 , 6 b = 5.16 Hz, 2J6a,6b = 12.32 Hz), 4.99 (s, 2H, #12), 5.06-5.22 (m, 2H, #2, HA), 5.29 (d, 1H, HI, 3 y u = 7.89 Hz), 5.37 (t, 1H, #3, 3 J 3 ; 4 = 9.45 Hz), 6.62 (d, 1H, # 1 6 , z J m i = 6.55 Hz), 7.04 (d, 2H, #8, #8', 3 J 8 , 9 = 8.38 Hz), 7.50 (d, 1H, #17, 3 Ji 6 ; i7 = 6.70 Hz), 7.53 (d, 2H, #9, #9', 3 J 8 ; 9 = 8.38 Hz). 13C{H} (CD 3 OD:CD 3 CN 1:1, 75 MHz): 8 12.30 (C9CH 3), 20.75 (4 COOCH 3 ) , 57.7 (C12), 62.86 (C6), 69.39 (CA), 72.20 (CI), 72.82 (C5), 73.54 (C3), 99.80 (CI), 111.05 (C16), 118.26 (C8), 122.50 (C9), 134.43 (C10), 136.44 (C17), 137.36 (C13), 154.70 (CI), 159.44 (C14), 165.77 (Cll) , 170.93, 171.06, 171.41, 171.92 (4 COOCH 3 ) , 174.1 (C15), 199.1, 199.2, 199.4 (3 CO). IR (NaCl plate, cm\"1): 2012, 1883 (v(/ac-Re(CO)3), 1754 (v(COOCH3)), 1692 (v(CONH)). MS (ESI+): m/z (relative intensity) = 897, 895 ([M-H 2 0 +Na]+, 80). MS (ESI-): m/z (relative intensity) = 873, 871 ([M-H 2 0 -1]\", 100). Anal. Calcd. for C 3 i H 3 3 N 2 O i 7 R e C H 3 O H : C, 42.43; H, 3.89; N, 3.09. Found: C, 42.26; H, 4.05; N, 3.27. Fac-(aqua)tri(carbonyl)(l-{p-benzyl P-D-glucopyranoside}-2-methyl-3-oxy-4(l#)-pyridinato)rhenium(I) hemihydrate (ReL3(CO)3(H2O)0.5H2O) O H The title compound was prepared from HO' ....-co [Re(CO)3(H20)3]Br (164 mg, 0.406 'co H L 3 (154 mg, 0.406 mmol) and 45 mmol) in methanol (20 mL). The product was purified by recrystallization from hot methanol, yielding a white solid that was dried in vacuo overnight (95 mg, 36%). 'H NMR (CD 3 OD:D 2 0 1:1, 400 MHz): 8 2.16 (s, 3H, Cl l -C# 3 ) , 3.48 (t, 1H, #4, 3 J 3 , 4 = 9.14 Hz), 3.57-3.60 (m, 3H, HI, H3, H5), 3.76 (dd, 1H, H6a, 2J6a>6b = 12.34 Hz, V5,6 = 5.79 Hz), 3.92 (d, 1H, H6b, J= 10.81), 5.12 (d, 1H, HI, 3 J i , 2 = 7.16 Hz), 6.73 (d, 1H, #14, Vi4, l 5 = 6.85 Hz), 7.29 (d, 2H, #8, #8', 3 J 8 ; 9 = 8.68 Hz), 7.36 (2H, H9, H9', %9 = 8.99 Hz), 7.58 (d, 1H, #15, V 1 4,i 5 = 6.35 Hz). I 3C{H} NMR (CD 3 OD:D 2 0, 75 MHz): 8 14.05 (CH 3), 62.49 (C6), 71.31 (CA), 74.82 (C2), 77.94 (C3), 78.34 (C5), 102.14 (CI), 111.21 (C14), 118.69 (C8, C8'), 133.21 (C9, C9'), 135.78 (Cl l ) , 135.86 (C15), 137.68 (C10), 159.26 (C7), 159.86 (C12), 175.61 (C13), 199.06, 199.45, 199.82 (3 CO). IR (KBr disc, cm -1): 331 l(br) (vOH), 2025 (vs), 1907 (vs) (v(/ac-Re(CO)3)), 1609 (s), (v(CO)), 1538 (s), 1505 (s), 1478 (s) (aryl), 1362 (s), 1306 (m), 1240 (m, br) (8(OH)), 1044 (s, br), 1072 (s), 1018 (s), (v(COH)), 817 (s), (para-substituted benzene). I n -visible spectrum (H 20): XmJnm (emJ L M\"1 cm\"1) = 214 (3.3 x 104), 318 (1.1 x 104). MS (ESI +): m/z (relative intensity) = 672, 670 ([M-H 20 +Na]+, 100), 650, 648 ([M-H 20 +1]+). Anal. Calcd. for C 2 l H 2 2 N O l 2 R e 0 . 5 H 2 O : C, 37.33; H, 3.43; N, 2.07. Found: C, 37.41; H, 3.50; N, 2.38. Fac-(aqua)tri(carbonyi)(l-{N-[methyl 6-amino-6-deoxy-a-D-glucopyranoside] ethanamide}-2-methyl-3-oxy-4(l/T)-pyridinato)rhenium(I) (ReL4(CO)3(OH2)) OH2 The title compounds was prepared using H L 4 (123 mg, | , ^ c o a N ^UP ^ - R f \ 0.343 mmol) and [Re(CO)3(H20)3]Br ( 139 mg, 0.344 HI\T ^cr I co CO H 0 - v x n 3 mmol) in methanol (10 mL), and the product was HO- 46 purified by semi-preparative HPLC. 'H NMR (CD 3 OD:D 2 0 1:1, 400 MHz): 5 2.39 (s, 3H, C9-C// 3 ), 3.22 (t, 1H, #4, 3JiA = 9.25 Hz), 3.41 (s, 3H, OC# 3), 3.44 (m, 1H, //6a), 3.48 (dd, 1H, HI, 3 J 1 > 2 = 3.66 Hz), 3.59-3.71 (m, 3H, #3, H5, //6b), 4.7 (d, 1H, HI, 3 J U = 3.66 Hz), 4.99 (s, 2H, Hi), 6.68 (d, 1H, #12, 3J i 2,i 3 = 6.74 Hz), 7.56 (d, 1H, #13, 3 J 1 2 J 3 = 6.93 Hz). 1 3C{H} NMR (CD 3 OD:D 2 0 1:1, 75 Mz): 5 12.53 (C9-CH 3), 41.37 (C6), 55.94 (OCH 3), 57.79 (CS), 70.86 (C5), 70.90 (CA), 70.96 (CI), 74.12 (C3), 100.50 (CI), 111.81 (C12), 137.1 (C13), 138.5 (C9), 158.00 (CIO), 169.1 (C7), 174.23 (Cll) , 199.06, 198.80, 198.30 (3 CO). IR (KBr disc, cm -1): 3393 (br) (vOH), 2016 (vs), 1891 (vs) (v(/ac-Re(CO)3)), 1676 (m), 1546 (s) (v(CONH)), 1610 (s) (v(CO)), 1343 (m), 1284 (m) (8(OH)), 1058 (s, br) (v(COH)). UV-visible spectrum (H 20): X m a x /nm (sm a x/ L M\"1 cm\"1) 318 (6.1 x 103), 273 (9.0 x 103). MS (ESI+): m/z (relative intensity) = 651, 649 ([M-H 20 +Na]+, 100), 629, 627 ([M-H 20 +1]+, 20). Anal. Calcd. for C i 8 H 2 3 N 2 0 l 2 R e : C, 33.49; H, 3.59; N, 4.34. Found: C, 33.12; H, 3.50; N, 3.90. Fac-(aqua)tri(carbonyl)(l-{N-[2-amino-2-deoxy-D-glucopyranose]ethanamide}-2-methyl-3-oxy-4(l#)-pyridinato)rhenium(I) hydrate, (Re L 5 (CO) 3 (OH 2 )H 2 0) OH The title compound was prepared from H L 5 (167 by one of two methods. The product was either allowed to precipitate from methanol (<2 mL), filtered out and washed with cold methanol or eluted from a silica column with 4:1 dichloromethane:methanol (Rf = -0.2), and obtained by reducing the volume of product-HO' 'co mg, 0.485 mmol) in a mixture of methanol (10 mg, 0.485 mmol) and [Re(CO)3(H20)3]Br (196 mL) and water (1 mL). The product was purified 47 containing fractions on a rotary evaporator. Both methods yielded a white solid that was dried in vacuo overnight (98 mg, 32%). 'H NMR (CD 3 OD:D 2 0 1:1, 400 MHz): 5 2.39 (s, 3H, C O O C / / 3 a), 2.40 (s, 3H, COOC / /3 P) , 3.40-3.46 (m, 3H, H3, //4p), 3.55 (m, 1H, //5p), 3.67 (m, 1H, //2p), 3.70-3.88 (m, 6H, HAa, H5a, H6), 3.90 (dd, 1H, H2a, \ 2 = 3.6 Hz, 3 J 2 ,3 = 6.9 Hz), 4.71 (d, 1H, / / ip , 3 J 1 ; 2 = 8.4 Hz), 4.99 (s, 2H, //8p), 5.01 (s, 2H, //8a), 5.18 (d, 1H, / / l a , 3 J 1 ; 2 = 3.66 Hz), 6.67 (d, 2H, H12, *Jl2A3 = 6.86 Hz), 7.56 (2 overlapping dd, 2H, H13, V i 2 J 3 = 6.85 Hz). 13C{H} NMR (CD 3 OD:D 2 0 1:1, 75 MHz): 8 12.48 (C9CH 3 P), 12.54 (C9CH 3), 55.72 (C2), 57.83 (CS), 58.11 (C2p), 58.37 (C8P), 62.03 (C6), 62.15 (C6P), 71.42 (C3p), 71.66 (C3), 72.11 (C5), 72.84 (CA), 75.14 (C5P), 77.44 (C4p), 92.00 (CI), 96.12 (Clp), 111.52 (C13), 137.11 (C12P), 137.18 (C12), 144.35 (C9), 158.19 (C10), 174.40 (CI 1), 198.98,199.15, 199.38 (3 CO). IR (KBr disc, cm\"1): 3394 (br) (vOH), 2016 (vs), 1891 (vs) (v(/ac-Re(CO)3)), 1677 (m), 1546 (s) (v(CONH)), 1610 (s) (v(CO)), 1344 (m), 1284 (m) (8(OH)), 1058 (s, br) (v(COH)). UV-visible spectrum (H 20): X m a x /nm (sm a x/ L M\" 1 cm\"1) = 318 (8.3 x 103), 263 (9.9 x 103). MS (ESI+): m/z (relative intensity) = 637, 635 ([M-H 20 +Na]+, 100). Anal. Calc. for Ci 7 H 2 iN 2 0i 2 Re-H 2 0: C, 31.43; H, 3.57; N, 4.31. Found: C, 31.23; H, 3.91; N, 4.14. 2.2.5 Solid State Structure Determination of H L 5 Needle crystals of H L 5 were grown by slow evaporation of a methanol solution, mounted on glass fibers and cooled to 173 K. The data were collected by Cheri A. Barta, and solved by myself with the assistance of Dr. Michael Merkel. Data sets were collected on a Bruker X8 APEX diffractometer using graphite-monochromated Mo K a 48 radiation (k = 0.71073 A). Data were collected and integrated using the Bruker SAINT 6 5 software package and corrected for Lorentz and polarization effects as well as for absorption (SADABS). 6 6 The structure was solved by direct methods (SIR92).67 Non-hydrogen atoms were refined anisotropically, while hydrogen atoms were added but not refined. Final refinement was completed using SHELXL-97. 6 8 2.2.6 Hexokinase Substrate Assay The compounds HL 1 \" 5 were qualitatively tested as substrates for hexokinase. Each compound (10 u.L, IO\"2 M) in 30 mM triethanolamine buffer (pH = 6) was added to 1 mL of the assay mixture, which contained 100 U/L hexokinase, 1 mM ATP, and 4 mM MgCb all dissolved in 30 mM triethanolamine buffer (pH = 6). After incubating for 10 min or 24 h, the assay mixture was analyzed by analytical HPLC, using 30 mM KH2PO4 buffer as the solvent. The compound was determined to be a substrate of hexokinase if the HPLC chromatogram showed the disappearance of ATP (retention time 7.0 min) and the appearance of the ADP (retention time 9.8 min). 2.2.7 Hexokinase Inhibition Assay The pro-ligands and their complexes were tested for their capacity to inhibit glucose phosphorylation by hexokinase. In a cuvet, 500 uL glucose (HK) assay kit (Sigma) was pipetted into a solution (750 uL) containing the pro-ligand or complex (final concentration 10\"3-10\"4 M), and glucose, or just glucose alone (final concentration 10-4 49 M). The final total volume was 1250 uL. The solution was monitored by UV-visible spectroscopy at 340 nm every 30 seconds for 10 min to determine the reaction rate. The reaction rates with and without the pro-ligand or complex were compared to determine if each compound could inhibit the phosphorylation of glucose by hexokinase. 2.2.8 9 9 m T c Radiolabelling [ 9 9 m Tc(CO) 3 (H 2 0 )3] + was prepared using Isolink kits and 200 MBq of [ 9 9 mTc0 4]\" in 1 mL saline. The [ 9 9 m Tc(CO) 3 (H 2 0 )3] + solution (100 uL) and 500 uL of a 10'3 M solution of HL 1 \" 5 in PBS (phosphate buffered saline pH = 7.4) were combined in a sealed vial purged with N 2 gas. The vial was heated to 75 \u00C2\u00B0C for 30 min. The radiochemical yield of 9 9 mTc(L'- 5)(CO) 3(H 20) was determined by HPLC using Re(L 1 _ 5 ) (CO)3 (H 2 0) as a standard. 2.2.9 1 8 6Re Radiolabelling A sealed vial containing 3.5 mg of B H 3 N H 3 was purged with CO gas for 10 min. A solution containing 400 uL of 10\"3 M solution of HL 1 \" 5 in PBS, 100 uL of 100 MBq of [186Re04]\" in saline, and 4.5 uL of 85% H 3 P 0 4 was added to the vial. A syringe to compensate for any gas evolution was attached to the vial, and the vial was heated at 60 \u00C2\u00B0C for 15 min. Radiochemical yield of 1 8 6Re(L ,\" 5)(CO) 3(H 20) was determined by HPLC using Re(L 1 _ 5)(CO) 3(H 20) as a standard. 50 2.2.10 Cysteine and Histidine Stability Challenges A solution of 9 9 mTc(L 1\" 5)(CO) 3(H 20) (100 uL) was added to a 900 uL solution of either 10\"3 M histidine, or 10\"3 M cysteine. The solutions were incubated at 37 \u00C2\u00B0C for 24 h, with samples analyzed by HPLC at 1, 4 and 24 h. 2.3 Results and Discussion 2.3.1 Synthesis and Characterization of H L 5 The 3-hydroxy-4-pyridinone moiety is a bifunctional chelate that deprotonates and binds as a monoanionic bidentate ligand, forming a stable 5-membered ring with the metal centre. Simple functionalization of the nitrogen of the pyridinone ring gives a remarkable spectrum of ligand designs. In this work we examined the binding of five pyridinone based pro-ligands bearing pendant carbohydrates.59 The glucose derivatives are tethered to the pyridinone binding moiety through the CI, C2, or C6 position with linkers of varying length. The pyridinone pro-ligands, HL 1 \" 4 have been previously prepared59 by Dr. David E. Green of the Orvig research group. A synthesis of H L 5 , also known as Feralex-G, has been previously reported.63 In the literature preparation, the non-benzylated analogue of 1, l-carboxymethyl-2-methyl-4(lH)-pyridinone, is coupled to glucosamine.63 Although the reported synthesis gives good yields, purification of the product posed difficulties, thus an alternate route was utilized (Scheme 2.1). More efficeint purification was afforded by preparing the 51 benzyloxy protected compound 2. The greater lipophilicity of 2 facilitated its silica gel purification and a protected pyridinone also prevented adventitious metal chelation from the silica. Debenzylation of 2 with H 2 and 10% Pd/C afforded high purity H L 5 . Scheme 2.1. Synthesis of H L 5 . a) glucosamine-HCl, DCC, NHS, diisopropylethylamine, DMF, 80\u00C2\u00B0C, 16 h. b) H 2 , Pd/C, MeOH, H 2 0 , The 'H and 1 3 C NMR spectra of H L 5 have been previously reported, but only for 63 1 13 the a anomer. In water both anomers are present in a 7:3 ct:P ratio, and the H and C N M R spectra of each was fully assigned with the aid of 2D N M R techniques. Verification of the anomeric nature of this ligand was important; it may play a role in the ability of the ligand to utilize the glucose metabolic pathways, and it contrasts starkly with the other five ligands, which were all prepared and isolated as single anomers. The solid state structure of H L 5 is reported and confirms several salient points (Figure 2.3, Table 2.1). The structure diagram clearly shows the distance of the glucose derivative, in this case glucosamine, from the pyridinone moiety, suggesting that the carbohydrate will not interact with the metal center upon complexation. As well, the glucosamine moiety is in the common chair conformation. To retain biological activity, minimal hindrance or modification of the glucose derivative is ideal. As H L 5 has one of 52 the shortest tether lengths between the glucose and binding moieties, it can by hypothesized that none of the ligands used in this study will have direct metal-carbohydrate interaction. Although both anomers of the glucosamine are seen in the NMR spectra,59 only one anomer is observed in the crystal lattice, due to preferential crystallization. N1-C2 1.457(4) N1-C7 1.327(4) 06- C7 1.222(4) 07- C11 1.368(4) 08- C12 1.260(4) C2-N1-C7 123.3(3) N1-C7-06 125.9(3) N1-C7-C8 113.3(3) 07- C11-C12 120.7(4) 08- C12-C11 120.4(3) 53 2.3.2 Synthesis and Characterization of ReL 1 5(CO) 3(H 20) The ReL 1- 5(CO) 3(H 20) complexes were prepared in high yield by three different methods. Preparation from Re(CO)5Br required a basic buffer system and heating with the pro-ligand, while preparation from [NEt4]2[Re(CO)3Br3] and [Re(CO)3(H20)3]Br, which have labile Br - and H 2 0 ligands, respectively, only required heating. All methods were stoichiometric and quantitative as monitored by lH NMR spectroscopy. Preparation from [Re(CO)3(H20)3]Br resulted in minimal side-products and was consequently the method of choice, as purification posed difficulties due to the polar nature of the complexes (Scheme 2.2). In cases where recrystallization was not successful and solvent limitations excluded silica and alumina chromatography, semi-preparative reverse phase HPLC was used. After purification, yields were 30-65%. All complexes were thoroughly characterized by the mass spectrometry, elemental analysis, IR and NMR spectroscopy. Electrospray mass spectrometry gave characteristic pairs of peaks with the correct isotopic pattern for Re ( 1 8 7Re 62.6%, 1 8 5 Re 37.4%). In all cases the H 2 0 ligand proposed to occupy the sixth coordination site was not observed in the MS. The mass-to-charge ratios observed corresponded to [M-H 2 0+H] + or [M-H 20+Na] + for positive ES-MS, and [M-H 20-H]' for negative ES-MS. Other peaks corresponding to the free ligand were observed, but are due to fragmentation, as no free pro-ligand is observed in the NMR spectra of the crude or purified complexes. Elemental analysis confirmed the bulk composition of the complexes, and included in every case the one water molecule proposed to occupy the sixth coordination site. Most complexes had more than just one water molecule 54 R I OH H L 1-5 a) [Re(CO)3(H20)3]Br, MeOH, reflux, 2 h b) [99mTc(CO)3(H20)3r, 75\u00C2\u00B0C, 30 min c) r^ReCy\", BH 3NH 3, CO ( g ) , 60\u00C2\u00B0C, 15 R I .CH, ,*OH 2 w \u00E2\u0080\u0094 - M oc*^ I ^co CO ML , \" 5 (CO) 3 (H 2 0) M = Re, 9 9 m T c , 1 8 6 R e HO HO OH 1 - \u00C2\u00B0 ^ ^ 9 C H 3 R = H, ReL1(CO)3(H20) R = acetyl, ReL2(CO)3(H20) Rel/(CO)3(H20) O H 2 >vvCO C H 3 C 0 O C H 3 ReL ,(CO)3(H20) ReL'(CO)3(H20) Scheme 2.2. Preparation and structures of M(L I\" s)(CO) 3(H 20) (M = Re, 9 9 m T c , 1 8 6Re), and M = Re structures showing numbering scheme for NMR studies. 5 5 associated with the solid product, common for carbohydrate containing compounds and also observed in the solid state structure of HL 5 (Figure 2.3). IR spectroscopy contained similar vibrational frequencies as the pro-ligands, with additional peaks diagnostic of the carbonyl stretches. Three peaks for the carbonyl stretches are expected due to the low symmetry of each compound. Two peaks were observed, one at 2025-2013 cm - 1 and another at 1907-1883 cm - 1, with the broader latter peak attributable to two peaks overlapping. 1 1 3 Both H and C NMR spectroscopy confirmed the solution structures of the complexes, of particular relevance as radiolabelling is done exclusively in solution. The pendant nature of the carbohydrate groups after complexation was confirmed by the small shifts in carbohydrate-associated 'H and 1 3 C NMR resonances, < 0.02 ppm and 1 ppm, respectively (Figure 2.4). Large coordination induced shifts (CIS) would be expected if any of the carbohydrate hydroxyl groups directly interacted with the metal ion. Shifts in the *H NMR spectra were observed for the pyridinone moiety, confirming binding. For all the complexes, the resonance associated with the hydrogen on the carbon atom adjacent to the binding ketone functionality shifted downfield 0.17-0.19 ppm (Figure 2.4). In the 1 3 C NMR spectra, the CISs were clearly observed for the carbon atoms bound to the coordinating oxygen atoms, as expected for bidentate chelation. For the ketone carbon, a downfield shift of 3.3-4.7 ppm confirms the dative bond, while the large downfield shift of 12-14.7 ppm for the adjacent hydroxyl carbon verifies the covalent metal oxygen bond. The three CO ligands were also observed in the 1 3 C NMR spectra, but the water molecule associated with the sixth coordination site was not observed in the 'H N M R spectra, possibly due to exchange with the deuterated solvent (water and/or methanol). 56 a b 7 . 5 7 . 0 6 . S G . O S . S 5 . 0 - 4 . 6 * * . 0 3 . 5 3 . 0 2 . 6 2 . 0 Figure 2.4. 'H NMR spectra (400 MHz, CD 3 OD:D 2 0 1:1) of (a) ReL 3(CO) 3(H 20) and (b) H L 3 . 2.3.3 Evaluation of Compounds as Hexokinase Substrates or Inhibitors The optimal imaging properties of FDG partially stem from its unique interaction with hexokinase (HK), the first enzyme in glucose metabolism. FDG is a substrate for H K , 6 9 which phosphorylates the sugar at the 6 position, but is not a substrate for the subsequent enzyme in the metabolism pathway. Instead, FDG becomes trapped in the 57 cell and accumulates, resulting in higher resolution imaging than if FDG perfectly mimicked glucose.1 Hence, one of the potential targets for the Tc and Re-glucose conjugates reported herein, is the hexokinase enzyme. Each of the pro-ligands and/or Re-complexes was tested as a potential substrate or inhibitor of HK. The pro-ligands were evaluated as substrates of HK by monitoring the conversion of ATP to ADP in the presence of pro-ligand and HK; none of the pro-ligands were substrates. To determine if other interactions between H K and the pro-ligands could be observed, inhibition studies were completed by monitoring the affect of each pro-ligand on the rate of glucose (the natural HK substrate) conversion. The rate was monitored using a secondary enzyme conversion, coupled to the conversion of N A D + to N A D H which can be quantitatively monitored by U V spectroscopy. At the concentrations tested, no inhibitory activity was observed. Other groups have noted inhibitory activity of similar Re-carbohydrate complexes, even when no activity was observed for the ligand.11 Thus, the complexes ReL , ' 5(CO) 3(H 20) were also tested for inhibition of HK, but no activity was observed at the concentration tested for the complexes either. At higher concentrations absorptions of the ligand chromophore interfere with the assay, limiting accuracy at the higher concentrations for which the pro-ligands or complexes may show activity. Further studies with other potential targets, such as glucose transporters, may show different results. 58 2.3.4 Radiolabelling with 9 9 m T c and 1 8 ( ,Re 186T Radiolabelling of H L 1 - 5 with 9 9 m T c and 1 8 6 Re was simple, fast and high yielding. For 9 9 m T c , the [ 9 9 m Tc(CO) 3 (H 2 0) 3 ] + precursor was heated with excess pro-ligand in PBS to give >95% yield for all complexes except 9 9 m TcL 2 (CO) 3 (H 2 0) (Table 2.2, Figure 2.5). Table 2.2. Identification methods, retention times and radiochemical yields for the preparation of M(L 1\" 5)(CO) 3(H 20) (M = y y m T c , l i S 0Re) \u00E2\u0080\u0094 9 9 m T - \u00E2\u0080\u009E 186T Ligand HPLC Method3 UV (254 nm) detected RT of standard / min Radiation detected RT / min (radiochemical yield) Tc Re H L 1 70% A 30% B H L 2 T L C method\" H L 3 50% A 50% B H L 4 70% A 30% B H L 5 50% A 50% B 9.5 0.7b 7.4 9.9 7.0 9.2 (97.7\u00C2\u00B10.5) 9 .5(99\u00C2\u00B10.1) 0.7b (87) 7.0 (96+3) 0.7b(70) 7.3 (89\u00C2\u00B13) 9 .8 (95\u00C2\u00B12) 9.9(99.1\u00C2\u00B10.3) 6.9 (96.3\u00C2\u00B12) 6.9 (91\u00C2\u00B12) a HPLC run on Synergi 4 urn C- l 8 Hydro RP analytical column, isocratic 1 mL/min, solvent A: 0.1% w/w trifluoracetic acid in water, solvent B: methanol, b T L C run on silica plates, developing with 80% ethyl acetate and 20% methanol, analyzed with UV lamp or RadioTLC, RT given is retention factor. 59 The lower yield when using H L is due to the lower water solubility of the pro-ligand. The Isolink kits are found to be unsuitable for the preparation of the analogous 1 8 6 Re precursor, although alternative methods have been used to prepare [188Re(CO)3(H20)3]+ in good yields (80-90%).23 Although Tc and Re have somewhat similar chemistry, Re is easier to oxidize, and thus re-oxidation to perrhenate is facile, especially in basic media with extended heating. Preparing [186Re(CO)3(H20)3]+ as described in the literature,14'17 followed by incubation at 75 \u00C2\u00B0C for 10-30 min gave radiochemical yields < 85%, with perrhenate as the major side product. To obtain the same high labelling yields for 1 8 6 Re as were achieved for 9 9 m T c , several modifications were necessary. The best method was a one pot preparation wherein the Re precursor is formed in the presence of the pro-ligand, facilitating quick chelation of the precursor and shorter heating times to minimize re-oxidation. The one pot preparation is optimal for radiopharmaceutical synthesis, and gave unprecedented radiochemical yields, for the [186Re(CO)3+] core, as high as 99% (Table 2.2, Figure 2.5). Again, H L 2 labelled at lower yields due to low ligand solubility. For both 9 9 m T c and 1 8 6 Re, the radiolabelled complexes were identified and radiochemical yields determined by either HPLC or T L C using the non-radioactive Re complexes for comparison. At tracer level concentrations of the radiolabelled complexes other characterization methods are precluded. 60 a 0 200 400 600 800 1000 b i < 200 400 600 800 1000 ro .n < i i i i 1 1 0 200 400 600 800 1000 Time/s Figure 2.5. HPLC results for radiolabelling of H L 5 (a) 1 8 6Re(L 5)(CO) 3(H 20) radiation trace, (b) 9 9 m Tc(L 5 ) (CO)3 (H 2 0) radiation trace, (c) Re(L 5)(CO) 3(H 20) U V trace. 61 2.3.5 Cysteine and Histidine Stability Challenges To assess the stability of the radiolabeled compounds, the 9 9 m T c labelled species were challenged with an excess of cysteine or histidine, potential metal binding amino acids ubiquitous in vivo. Al l complexes were stable to 24 hours, suggesting that they would be appropriate for radiopharmaceutical applications (Table 2.3, Figure 2.6). At the 1 and 4 hour time points little decomposition was observed in all cases, with >95% of the complex remaining intact. At 24 hours some decomposition was observed, mainly due to substitution of the amino acid for the ligand, but >70% of the complex was still intact regardless of which complex was challenged. High stability under these conditions has been reported for other [ 9 9 m Tc(CO) 3 ] + complexes of bidentate ligands,15 but does not obviate the possibility of unfavorable binding of the complex to plasma proteins.19 Table 2.3. Percentage 9 9 mTc(L 1\" 5)(CO) 3(H 20) remaining intact determined by HPLC after 1, 4, and 24 h in IO - 3 M solution of cysteine or histidine at 37 \u00C2\u00B0C. Ligand Cysteine Histidine Th 4h 24h Th 4h 24h 95\u00C2\u00B17 95\u00C2\u00B13 71+8 99\u00C2\u00B13 97\u00C2\u00B12 77\u00C2\u00B16 99\u00C2\u00B12 99\u00C2\u00B14 75\u00C2\u00B16 98+1 96\u00C2\u00B13 93\u00C2\u00B15 62 H L 1 H L 3 H L 4 H L 5 99\u00C2\u00B17 99\u00C2\u00B11 99\u00C2\u00B11 99\u00C2\u00B11 99\u00C2\u00B16 97+2 99\u00C2\u00B11 99\u00C2\u00B11 97\u00C2\u00B13 79+1 95\u00C2\u00B15 81\u00C2\u00B18 a b 24 hours c < 0 200 4 hours 400 600 800 1000 1200 c r> \u00E2\u0080\u00A2 E < 24 hours c E < 0 200 400 600 800 1000 1200 4 hours c E < 0 200 400 600 800 1000 1200 1 hour c E \u00E2\u0080\u00A2G < 0 200 400 600 800 1000 1200 1 hour D i < 0 200 400 600 800 1000 1200 Time / s 0 200 400 600 800 1000 1200 Time / s Figure 2.6. HPLC radiation traces of 9 9 m Tc(L 3 )(CO) 3 (H 2 0) incubated in excess (a) cysteine or (b) histidine after 1, 4, and 24 hours. 63 2.4 Conclusions This work adds to the type of bifunctional chelates that have been studied with the popular [M(CO)3] + core (M = Re, Tc), and broadens the scope of Tc-glucose conjugates. The complexes are well characterized, and confirm the pendant nature of the carbohydrate, the bidentate chelation of the pyridinone, and the presence of a water molecule in the sixth coordination site. High yield labelling of the pyridinone based ligands with both 9 9 m T c and 1 8 6 Re was achieved. The one vial, one step labeling of 1 8 6 Re at such high yields is unprecedented and could be easily developed into a kit preparation. Only limited progress of Re labeling with the [Re(CO)3]+ core has been reported to date, especially with the 1 8 6 Re isotope, despite its radiotherapeutic potential. All 9 9 m T c complexes proved to be reasonably stable up to 24 hours when challenged with cysteine or histidine solutions, although further stability challenges in plasma and with the 1 8 6 Re complexes are needed to confirm that the complexes will be stable in vivo. Finally, the pro-ligands and complexes were evaluated as potential substrates or inhibitors of the glucose metabolism enzyme hexokinase. Unfortunately, no substrate or inhibitory activity was observed at the concentrations tested. These compounds may still have potential as imaging and therapy agents, if they interact with glucose transporters (some of which are highly expressed in certain tumor types) or if they can accumulate in tumors via the another mechanism as suggested for other promising Tc-glucose conjugates.8 64 2.5 References 1. Beuthien-Baumann, B.; Hamacher, K.; Oberdorfer, F.; Steinbach, J., Carbohydrate Res. 2000, 327, 1107-1118. 2. Krohn, K. A.; Mankoff, D. A.; Muzi, M. ; Link, J. M. ; Spence, A. M . , Nucl. Med. Biol. 2005, 32, 663-671. 3. Segall, G., Nucl. Med. Comm. 2002, 23, 323-330. 4. Czernin, J., Acta Medica Austriaca 2002, 5, 162-170. 5. Dilworth, J. R.; Parrott, S. J., Chem. Soc. Rev. 1998, 27, 43-55. 6. Banerjee, S. R.; Babich, J. W.; Zubieta, J., Inorg. Chim. Acta 2006, 359, 1603-1612. 7. Duatti, A.; Marchi, A.; Magon, L.; Deutsch, E.; Bertolasi, V. , Inorg. Chem. 1987, 2(5,2182-2186. 8. Yang, D. J.; Kim, C.-G.; Schechter, N. R.; Azhdarinia, A.; Yu, D.-F.; Oh, C.-S.; Bryant, J. L.; Won, J.-J.; Kim, E. E.; Podoloff, D. A., Radiology 2003, 226, 465-473. 9. Petrig, J.; Schibli, R.; Dumas, C.; Alberto, R.; Schubiger, P. A., Chem. Eur. J. 2001, 7, 1868-1873. 10. Dumas, C ; Petrig, J.; Frei, L.; Spingler, B.; Schibli, R., Bioconjugate Chem. 2005,16, 421-428. 11. Schibli, R.; Dumas, C ; Petrig, J.; Spadola, L.; Scapozza, L.; Garcia-Garayoa, E.; Schubiger, P. A., Bioconjugate Chem. 2005,16, 105-112. 12. Ozker, K.; Collier, B. D.; Lindner, D. J.; Kabasakal, L.; Liu, Y.; Krasnow, A. Z.; Hellman, R. S.; Edwards, S. D.; Bourque, C. R.; Crane, P. D., Nuclear Med. Comm. 1999, 20, 1055-1058. 65 13. Sobgbein, O. O.; Green, A. E. C.; Schaffer, P.; Chankalai, R.; Lee, E.; Healy, B. D.; Morel, P.; Valliant, J. F., Inorg. Chem. 2005, 44, 9574-9584. 14. Storr, T.; Fisher, C. L.; Mikata, Y.; Yano, S.; Adam, M . J.; Orvig, C., Dalton Trans. 2005, 654-655. 15. Storr, T.; Obata, M . ; Fisher, C. L.; Bayly, S. R.; Green, D. E.; Brudzinska, I.; Mikata, Y.; Patrick, B. O.; Adam, M . J.; Yano, S.; Orvig, C., Chem. Eur. J. 2005, 11, 195-203. 16. Storr, T.; Sugai, Y.; Barta, C. A.; Mikata, Y.; Adam, M . J.; Yano, S.; Orvig, C., Inorg. Chem. 2005, 44, 2698-2075. 17. Bayly, S. R.; Fisher, C. L.; Storr, T.; Adam, M . J.; Orvig, C., Bioconjugate Chem. 2004, 75, 923-926. 18. Klufers, P.; Krotz, O.; OBberger, M . , Eur. J. Inorg. Chem. 2002, 1919-1923. 19. Schibli, R.; Bella, R. L.; Alberto, R.; Garcia-Garayoa, E.; Ortner, K.; Abram, U.; Schubiger, P. A., Bioconjugate Chem. 2000, 77, 345-351. 20. Hafliger, P.; Mundwiler, S.; Ortner, K.; Spingler, B.; Alberto, R., Synth. React. Inorg. Metal-Org. Nano-Metal Chem. 2005, 35, 27-34. 21. Rattat, D.; Eraets, K.; Cleynhens, B.; Knight, H.; Fonge, H.; Verbruggen, A., Tetrahedron Lett. 2004, 45, 2531-2534. 22. Schibli, R.; Schwarzbach, R.; Alberto, R.; Ortner, K.; Schmalle, H.; Dumas, C.; Egli, A.; Schubiger, P. A., Bioconjugate Chem. 2002,13, 750-756. 23. Park, S. H.; Seifert, S.; Pietzsch, H.-J., Bioconjugate Chem. 2006, 77, 223-225. 24. Alberto, R.; Pak, J. K.; Staveren, D. v.; Mundwiler, S.; Benny, P., Biopolymers 2004, 76, 324-333. 66 25. Alberto, R.; Schibli, R.; Egli, A.; Schubiger, P. A., J. Am. Chem. Soc. 1998, 120, 7987-7988. 26. Alberto, R.; Schibli, R.; Waibel, R.; Abram, U.; Schubiger, A. P., Coord. Chem. Rev. 1999, 99, 901-919. 27. Lazarova, N.; James, S.; Babich, J.; Zubieta, J., Inorg. Chem. Comm. 2004, 7, 1023-1026. 28. Banerjee, S. R.; Levadala, M . K.; Lazarova, N.; Wei, L.; Valliant, J. F.; Stephenson, K. A.; Babich, J. W.; Maresca, K. P.; Zubieta, J., Inorg. Chem. 2002, 41, 6417-6425. 29. Pak, J. K.; Benny, P.; Spingler, B.; Ortner, K.; Alberto, R., Chem. Eur. J. 2003, 9, 2053-2061. 30. Staveren, D. E. v.; Benny, P. D.; Waibel, R.; Kurz, P.; Pak, J. K.; Alberto, R., Helv. Chim. Acta 2005, 88, 447-459. 31. Vitor, R. F.; Alves, S.; Correia, J. D. G.; Paulo, A.; Santos, I., J. Organomet. Chem. 2004, 689, 4764-4774. 32. Hamzavi, R.; Happ, T.; Weitershaus, K.; Metzler-Nolte, N. , J. Organomet. Chem. 2004, 689, 4745-4750. 33. Alberto, R.; Schibli, R.; Schubiger, P. A., J. Am. Chem. Soc. 1999, 121, 6076-6077. 34. He, H.; Lipowska, M . ; Xu, X.; Taylor, A. T.; Carlone, M . ; Marzilli, L. G., Inorg. Chem. 2005, 44, 5437-5446. 35. Lazarova, N.; Babich, J.; Valliant, J.; Schaffer, P.; James, S.; Zubieta, J., Inorg. Chem. 2005, 44, 6763-6770. 36. Braband, H.; Abram, U., J. Organomet. Chem. 2004, 689, 2066-2072. 67 37. Schibli, R.; Netter, M . ; Scapozza, L.; Birringer, M . ; Schelling, P.; Dumas, C ; Schoch, J.; Schubiger, P. A., J. Organomet. Chem. 2003, 668, 67-74. 38. Frantz, S.; Fiedler, J.; Hartenbach, I.; Schleid, T.; Kaim, W., J. Organomet. Chem. 2004, 689, 3031-3039. 39. Grewe, J.; Hagenbach, A.; Stromburg, B.; Alberto, R.; Vazquel-Lopez, E.; Abram, U., Z. Anorg. Allg. Chem. 2002, 629, 303-311. 40. Palma, E.; Correia, J. D. G.; Domingos, A.; Santos, I.; Alberto, R.; Spies, H., J. Organomet. Chem. 2004, 689, 4811-4819. 41. Songbein, O. O.; Merdy, P.; Morel, P.; Valliant, J. F., Inorg. Chem. 2004, 43, 3032-3034. 42. Bernard, J.; Ortner, K.; Spingler, B.; Pietzsch, H.-J.; Alberto, R., Inorg. Chem. 2003, 42, 1014-1022. 43. Masi, S.; Top, S.; Boubekeur, L.; Jaouen, G.; Mundwiler, S.; Spingler, B.; Alberto, R., Eur. J. Inorg. Chem. 2004, 2013-2017. 44. Saidi, M . ; Seifert, S.; Kretzschmar, M. ; Bergmann, R.; Pietzsch, H.-J., J. Organomet. Chem. 2004, 689, 4739-4744. 45. Spradau, T. S.; Katzenellenbogen, J. A., Organometallics 1998, 77, 2009-2017. 46. Luo, H.; Rettig, S. J.; Orvig, C, Inorg. Chem. 1993, 32, 4491-4497. 47. Edwards, D. S.; Liu, S.; Poirier, M . J.; Zhang, Z.; Webb, G. A.; Orvig, C., Inorg. Chem. 1994, 33, 5607-5609. 48. Katoh, A.; Taguchi, K.; Saito, R.; Fujisawa, Y.; Takino, T.; Sakurai, H., Heterocycles 2003, 60, 1147-1159. 49. Sanots, M . A.; Gil, M. ; Gano, L.; Chaves, S., J. Biol. Inorg. Chem. 2005,10, 564-580. 68 50. Burgess, J.; Fawcett, J.; Llewellyn, M . A.; Parsons, S. A.; Russell, D. R., Transition Met. Chem. 2000, 25, 541-546. 51. Tanaka, K.; Tengeiji, A.; Kato, T.; Toyama, N.; Shiro, M . ; Shionoya, M . , J. Am. Chem. Soc. 2002,124, 12494-12498. 52. El-Jammal, A.; Howell, P. L.; Turner, M . A.; Lin, N.; Templeton, D. M . , J. Med. Chem. 1994,57,461-466. 53. Hsieh, W.-Y.; Liu, S., Inorg. Chem. 2004, 44, 2031-2038. 54. Ahmed, S. I.; Burgess, J.; Fawcett, J.; Parsons, S. A.; Russell, D. R.; Laurie, S. H., Polyhedron 2000, 19, 129-135. 55. Lang, R.; Palborn, K.; Severin, T.; Severin, K., Inorg. Chim. Acta 1999, 294, 62-67. 56. Kennedy, D. C ; Wu, A.; Patrick, B. O.; James, B. R., Inorg. Chem. 2005, 44, 6529-6535. 57. Nelson, W. O.; Rettig, S. J.; Orvig, C , Inorg. Chem. 1988, 30, 509-515. 58. Nelson, W. O.; Karpishin, T. B.; Rettig, S. J.; Orvig, C , Inorg. Chem. 1988, 27, 1045-1051. 59. Green, D. E.; Ferreira, C. L.; Stick, R. V.; Patrick, B. O.; Adam, M . J.; Orvig, C., Bioconjugate Chem. 2005,16, 1597-1609. 60. Zhang, Z.; Lyster, D.; Webb, G. A.; Orvig, C , Nucl. Med. Biol. 1992, 19, 327-335. 61. Drouza, C.; Gramlich, V.; Sigalas, M . P.; Pashalidis, I.; Keramidas, A. D., Inorg. Chem. 2004, 43, 8336-8345. 62. Xu, J.; Churchill, D. G.; Botta, M. ; Raymond, K. N., Inorg. Chem. 2004, 43, 5492-5494. 69 Chapter 3 Rhenium and Technetium Complexes of Tridentate Ligands with Pendant Carbohydrates* 3.1 Introduction Tridentate ligands are well suited to binding the [M(CO)3] + (M = Tc, Re) core, as they form more stable complexes than do bidentate ligands. The saturation of the [M(CO)3] + coordination sphere by tridentate ligands invokes a higher level of stability both due to the larger chelate effect and by limiting the accessibility of the metal centre to protein binding in vivo.1'2 In the search for the optimal tridentate chelate for this core, several studies have compared different donor atom sets, and found aromatic amines to be superior binding groups. \" Tridentate ligands containing sulphur, \" carboxylate ' \"' 10 12 18 2 6 8 16 18 ' _ and alkyl amine ' ' \" ' donors have also been shown to form stable complexes, often in concert with the preferred aromatic amine donor. 2 ' 5 ' 9 ' 1 1 - 1 8 Derivatives of the dipicolylamine chelate moiety, which contains two aromatic amine donors, have been shown to bind the [M(CO)3] + core with high affinity to give chemically robust complexes. 5' 9' 1 2' 1 7' 1 9' 2 0 The utility of these bifunctional chelates (ligands with both a chelating moiety and a position for biomolecule attachement) to attach biomolecules, *A version of this chapter has been published. Storr, T., Obata, M., Fisher, C. L. Bayly, S. R., Green D. E., Brudzifiska, I., Mikata, Y., Patrick, B. O., Adam, M. J., Yano, S., Orvig C. Novel carbohydrate-appended metal complexes for potential use in molecular imaging. Chem. Eur. J. 2005,11, 195-203, and Bayly, S. R., Fisher, C. L. Stor, T., Adam, M. J. Orvig, C. 99mTc(I) and 186Re(I) tricarbonyl complexes of N-(2'-hydroxybenzyl)-2-amino-2-deoxy-D-glucose. Bioconjugate Chem. 2004,15, 923-926. 71 63. Kruck, T. P. A.; Burrow, T. E. , J. Inorg. Biochem. 2002, 88, 19-24. 64. Alberto, R.; Egli, A.; Abram, U.; Hegetschweiler, K.; Gramlich, V.; Schubiger, P. A., J. Chem. Soc. Dalton Trans. 1994, 2815-2820. 65. SAINT, 1999, Bruker AXS Inc.: Madison WI. 66. SADABS, 1999, Bruker AXS Inc.: Madison, WI. 67. Altomare, A.; Burla, M . C ; Camalli, M. ; Cascarano, G. L.; Giocavazzo, C.; Guagliardi, A.; Moliterni, A. G. G.; Polidori, G.; Spagna, R., J. Appl. Crystallogr. 1999, 52,115-119. 68. SHELXL, 1997, Bruker AXS Inc.: Madison, WI. 69. Bessell, E. M. ; Foster, A. B.; wood, J. H. W., Biochem. J. 1972,128, 199-204. 70 such as peptides,17'20 proteins,17 nucleosides19 and carbohydrates,21'22 to the [M(CO) 3] + core has been documented. Previous work in the Orvig group, in collaboration with Prof. Shigenobu Yano and Prof. Yuji Mikata, has examined the [M(CO )3 ] + core with dipicolylamine ligands bearing pendant carbohydrates (mannose, xylose, glucose and glucosamine).21'22 Several complexes were prepared in reasonable yields and thoroughly characterized. As well, the ligands were labelled with 9 9 m T c in high radiochemical yields and the radiolabeled complexes evaluated for stability. The complexes were found to be stable in solutions containing 100 fold excess of either cysteine or histidine over 24 h. The high yield labelling and excellent stability of these complexes warrants further investigation into their potential as radiopharmaceuticals. This chapter examines the binding of an unexpected tridentate N-(2'-hydroxybenzyl)-2-amino-2-deoxy-D-glucose pro-ligand (HL 6) and continues the study of dipicolylamine ligands bearing pendant carbohydrates, focusing on the glucose and glucosamine conjugates (L 7' 9) (Figure 3.1). H L 6 was designed to be a bidentate ligand, but extensive NMR studies, described herein, show that the ligand binds [Re(CO)3J + core in an unexpected tridentate fashion. The 1-deoxy-l-thiol-D-glucopyranoside conjugate of dipicolylamine, L , has not been previously studied with the [Re(CO)3] core, thus complexation and characterization are reported. Radiolabelling both H L 6 and L 7 with 9 9 m T c and assessing stability was also undertaken, and compared to previous work on the dipicolylamine carbohydrate conjugates. Radiolabelling with the potentially therapeutic 186 6 7 9 isotope Re was completed for H L and L \" . Finally, further biological studies with the 9 9 m T c complexes of the three dipicolylamine ligands, L 7 \" 9 are reported, to better assess their potential as radiopharmaceuticals. 72 Figure 3.1. Tridentate pro-ligand and ligands with pendant carbohydrates. 3.2 Experimental 3.2.1 Materials and Methods Most relevant information can be found in section 2 .2 .1 . The pro-ligand H L 6 was prepared by Dr. Simon B a y l y , 2 3 the ligands L 7 \" 8 were gifts from Prof. Yu j i Mikata and Prof. Shigenobu Yano, Nara Women's University, Japan; 2 4 L 9 and [ReL 8\" 9(CO)3JBr were 71 77 prepared by Dr. T i m Storr. ' 73 3.2.2 Synthesis of ReL 6 (CO) 3 and of [ReL 7(CO) 3]Br Fac-tri(carbonyl)(N-(2'-oxybenzyl)-2-amino-2-deoxy-D-glucopyranose)rhenium(I) monohydrate ( R e L 6 ( C O ) 3 H 2 0 ) . OH The title compound was originally prepared by Dr. Simon HO Q 5 / ^ 1 7 2 3 ^ 8 13 Bayly, while subsequent preparations and N M R studies , - ^ 0 \u00E2\u0080\u0094 \ ^ 1 2 were done by myself. For completeness, the synthesis and ^ 0 \u00C2\u00B0 \u00C2\u00B0 full characterization are reported here. [NEt 4]2[Re(CO) 3Br3] (200 mg, 0.26 mmol), H L 6 (74 mg, 0.26 mmol) and sodium acetate trihydrate (40 mg, 0.32 mmol) were dissolved in H 2 O (7 mL) and heated with stirring at 50\u00C2\u00B0C for 2 h. The solvent was then removed in vacuo and the residue dissolved in CH2CI2 (10 mL) for 30 min. On standing, a brown residue was recovered by decanting off the solvent. The brown residue was purified by silica column chromatography eluting with 5:1 C H 2 C l 2 : M e O H yielding an off-white powder (58 mg, 40%). 'H N M R ((CD 3) 2SO, 400 MHz) 5 2.37 (1H, HI), 3.20 (1H, HA), 3.40 (1H, //6a), 3.43 (1H, H5), 3.60 (1H, //6b), 3.66 (1H, H3), 3.85, 4.30 (2H, HT), 5.22 (d, 1H, HI), 6.35 (1H, #10), 6.45 (1H, H12), 6.80 (1H, HI 1), 6.95 (1H, #13). 1 3C{H} N M R ((CD3)SO, 100 MHz) 5 51.1 (CT), 58.0 (CI), 59.8 (C6), 70.6 (CA), 71.8 (CS), 79.8 (C3), 87.5 (CI), 114.1 (C12), 119.4 (C8), 120.3 (C10), 129.1 (CI 1), 130.6 (C13), 163.2 (C9). MS (ESI+) m/z (relative intensity): 556, 554 ([M+H] +, 100), 578, 576 ([M+Na] + , 55). Ana l . Calcd. for C 1 6 H i 8 N 0 9 R e - H 2 0 : C, 33.57; H, 3.52; N , 2.45. Found: C, 33.55; H, 3.53; N , 2.75. 74 Fac-{(2-bis(2-pyridinyImethyl)amino)ethyl 1-deoxy-l-thio-P-D-glucopyranoside) tricarbonylrhenium(I) bromide tetrahydrate ([ReL7(CO)3]Br-4H20). evaporation yielded a yellow solid, which was pure by H N M R spectroscopy. Further purification was achieved by silica column chromatography eluting with 1:1 ethyl acetate:methanol (rf = ~0.6). After removing the solvent on a rotary evaporator any residual silica from the column was removed by passing the product down a size exclusion Sephadex G10 column, eluting with water. The water was removed by rotary evaporation yielding a faintly yellow solid which was vacuum dried overnight (66 mg, 87%). ' H N M R ( C D 3 C N : D 2 0 1:1, 400 M H z ) 5 3.10 (ddd, 1H, H7a, 2 J 7 a , 7 b = 12.79 Hz , 3 J 7 a , 8 b = 12.03 Hz , 3JlaM = 4.72 Hz), 3.2 (ddd, 1H, H7b, 2 J 7 a , 7 b = 12.79 H z , V 7 b , 8 a = 12.03 Hz , 3 J 7 b ; 8 b = 4.72 Hz), 3.21-3.32 (m, 2H, H2, H4), 3.33-3.4 (m, 2 H , H3, H5), 3.56 (dd, 1H, HI, V 6 a , 6 b = 12.0 H z , 3JSfi = 5.6 Hz), 3.75 (d, 1H, H6, 2J6a>6b = 12.0 Hz) , 3.94 (ddd, 1H, H8a, 2 J 8 a , 8 b = 12.94 Hz , 3 J l h M = 12.03 Hz , 3 J 7 a , 8 a = A.12 Hz) , 4.04 (ddd, 1H, H8b, 2 J 8 a , 8 b = 12.94 Hz , 3 J 7 a , 8 b = 12.03 Hz , V 7 b ; 8 b - 4.72 Hz) , 4.54 (d, 1H, HI, 3 J i , 2 = 9.6 Hz), 4.67-4.79 (m, 4 H , H9, H9', 2 J 9 a , 9 b = 16.8 Hz , J= 13.6 Hz) , 7.26 (t, 2 H , H13, H13\3J= 6.4 Hz), 7.42 (d, 2H , Hll, HIT, 3 J 1 U 2 = 7.6 Hz), 7.83 (t, 2 H , H12, HI2', 3J= 8.0 Hz), 8.71 (d, 2H , HI4, H14', 3 J 1 3 j 4 = 5.2 Hz). 1 3 C { H } N M R ( C D 3 C N : D 2 0 1:1, 400 M H z ) 5 25.89 ( C 7 ) , 61.89 (Cd), 68.10, 68.13 (CP, C9% 70.34 (C4), 70.82 (C8), 73.23 (C2), 78.08 (C3), 80.89 (CS), 86.33 (CI), 124.20, 124.24 (Cll, Cll'), 126.51 (CI3, CI3% h\ (45 mg, 0.11 mmol), and HO Br [Re(CO) 3 (H 2 0) 3 ]Br (44 mg, 0.11 mmol) were refluxed in methanol (8 mL) for 2 h. Removal of the solvent by rotary 75 141.36 (C12, C12% 152.66 (CI4, C14'), 160.88, 160.90 (C10,CW), 196.27, 197.12, 198.10 (3 CO). IR (KBr disc, cm-1): 3395(br) (vOH), 2019 (vs), 1930 (vs) (v(fac-Re(CO)3)), 1611 (s) (aryl), 1384 (m, br) (5(OH)) 769 (m) (aryl). MS (ESI+): m/z = 692.2 ([M-Br]+, 100). Anal. Calcd. for C 2 3 H 2 7 BrN 3 0 8 ReS-4H 2 0 : C, 32.74; H, 4.18; N, 4.98. Found: C, 32.86; H, 3.77; N, 4.97. 3.2.3 9 9 m T c Radiolabelling 9 9 m T c radiolabelling of H L 6 and L 8 \" 9 were done by Dr. Simon Bayly 2 3 and Dr. Tim Storr, 2 1' 2 2 respectively, in a similar manner to the method described here for L 7 . [ 9 9 m Tc(CO) 3 (H 2 0 )3] + was prepared using Isolink kits and 200 MBq of [ 9 9 m Tc0 4 ] _ in 1 mL saline. The [ 9 9 m Tc(CO) 3 (H 2 0 )3] + solution (100 uL) and 500 uL of a 10-3 M solution of L 7 in PBS (phosphate buffered saline pH = 7.4) were combined in a sealed vial purged with N 2 gas. The vial was heated to 75 \u00C2\u00B0C for 30 min. The radiochemical yield of [ 9 9 mTc(L 7)(CO) 3] + was determined by HPLC using [Re(L7)(CO)3]Br as a standard. 3.2.4 1 8 6Re Radiolabelling The [ Re(CO) 3(H 20) 3] precursor was prepared in a similar manner to that 1 88 + 2^ reported for the preparation [ Re(CO) 3(H 20) 3] . Briefly, borane ammonia complex (3.5 mg) was weighed into a 3 mL vial, the vial sealed and purged for 10 min with CO 76 gas. A solution of 300 M B q [ 1 8 6 Re0 4 ] \" in 0.5 m L saline and 85% H 3 P 0 4 (4 uL) was added to the vial , and the vial heated at 60\u00C2\u00B0C for 15 min. The [ 1 8 6 R e ( C O ) 3 ( H 2 0 ) 3 ] + precursor (100 uL) was added to a sealed vial purged with N 2 gas containing 500 u L of 10\"3 M H L 6 or L 7 \" 9 in P B S . For the labelling of H L 6 the reaction was heated at 75\u00C2\u00B0C for 30 min. While for the labelling of L 7 \" 9 optimal yields were obtained by heating the reaction at 60\u00C2\u00B0C for 15 min. A s above, H P L C was used determine the radiochemical yields of 1 8 6 R e L 6 ( C O ) 3 and [ 1 8 6 R e L 7 \" 9 ( C O ) 3 ] + using R e L 6 ( C O ) 3 and [ReL 7 \" 9 (CO) 3 ]Br as standards, respectively. 3.2.5 Cysteine and Histidine Stability Challenges Histidine and cysteine stability challenges were carried out on 9 9 m T c L 8 - 9 ( C O ) 3 by Dr. T i m Storr in a similar manner to the method described here . 2 1 ' 2 2 A solution of 9 9 m T c ( L 6 ) ( C O ) 3 (100 uL) was added to a 900 u L solution of either 10\"3 M histidine or cysteine, while a solution of [ 9 9 m T c ( L 7 ) ( C O ) 3 ] + (100 uL) was added to a 900 uL solution of 10\" M histidine or cysteine. The solutions were incubated at 37 \u00C2\u00B0C for 24 h, with samples analyzed by H P L C at 1, 4 and 24 h. 3.2.6 Biodistribution Studies r Biodistribution studies were carried out for [ 9 9 m T c ( L 7 \" 9 ) ( C O ) 3 ] + by Fabio Marques at the Faculty of Medicine, University of Sao Paulo, Brazi l . C576 mice bearing murine melanoma tumours were injected with 0.111-0.185 M B q of the radiolabelled 77 compound. After a set period of time the mice were anesthetised using pentobarbital, and blood obtained by cardiac punch. Then the mice were sacrificed and the organs were excised, weighed and their radioactivity measured. The results are the average of a minimum of 3 mice at each time point (15, 60 and 120 min) and are expressed as % of total injected activity / g weight of tissue. 3.2.7 Planar Scintigraphy Imaging Planar scintigraphy imaging studies, using [ 9 9 m Tc(L 7 _ 9 ) (CO )3] + , were undertaken with the assistance of Dr. Don Yapp (BC Cancer Agency) and Prof. Anna Celler (Head of Medical Imaging Research Group and Dept. of Radiology, UBC) in DD/S mice bearing the Shionogi or SC-115 tumour line (derived from a mammary gland of a mouse sensitive to androgen). The mice were sedated with an appropriate amount of ketamine, positioned on the bed of a Siemens Ecam SPECT camera with a ultrahigh resolution low energy collimator and injected with 12-20 MBq of [ 9 9 mTc(L 7 _ 9)(CO) 3] + . Planar images were acquired from only one mouse, which was injected with [ 9 9 m Tc(L 9 ) (CO )3] + , due to mortality issues. Images were acquired over 2.5 h, and the liver, kidney, bladder and tumour activity estimated from activity / pixel. Blood activity was estimated by measuring the activity in the head, as no brain uptake is expected and other tissue uptake is minimal. 78 3.3 Results and Discussion 3.3.1 Synthesis and Characterization of ReL6(CO)3 The reaction of H L 6 with [NEt4]2[Re(CO)3Br3] and NaOAc in H 2 0 produced the compound ReL 6 (CO)3 in 40% yield after column chromatographic purification (Scheme 3.1). The molecular ion was identified as [ReL6(CO)3+H]+ by ESIMS, and the Scheme 3.1. Synthesis of ReL 6(CO) 3 . formulation of the bulk sample was confirmed by elemental analysis. The 'H NMR spectrum of the complex is highly convoluted, but the shifting and broadening out of the aromatic resonances compared to those of H L 6 signify that the phenol \"arm\" participates, as desired, in binding of the [Re(CO)3]+ core. The splitting of the methylene (H7) proton signal into two doublets for each anomer indicates the methylene proton inequivalence on formation of the complex. Binding of the ligand N and O donor atoms incorporates the methylene in a ring, rigidly holding its two protons in diastereotopic chemical environments. Signals due to the HI protons were shifted downfield in both anomers 79 compared to those of H L 6 . Peaks due to the sugar H2 protons are also well-resolved and, compared to those of H L 6 , are also shifted slightly downfield, due to their close proximity to the binding amine. Small extraneous peaks in the spectrum also indicate that at least one other minor species is present. When kept overnight in MeOH or DMSO solution, samples of the complex become visibly brown and the relative intensities of these peaks increase, indicating that they arise from decomposition products. The signals do not correlate with the chemical shifts of uncomplexed H L 6 . Minor species are also detected in the HPLC UV-visible trace of the complex and become more significant over time. 13 The C NMR spectrum of the complex, run in DMSO due to solubility issues, was fully assigned for the a anomer, and confirms the expected bidentate binding mode of the ligand, but also suggests a third interaction between the Re centre and the sugar (Figure 3.2). The Re carbonyls show three sharp resonances at 196-198 ppm as expected due to the low symmetry. Peaks due to the phenol CO and the C H 2 linker are shifted significantly downfield, compared to their values in the NMR of H L 6 , giving a clear indication that the Re is bound by both the phenol O and the glucosamine N . 2 6 The CI and C2 signals are shifted upfield on complexation, presumably reflecting some slight conformation change in the hexose skeleton. Unexpectedly, the C3 signal has shifted downfield 7.4 ppm, suggesting that the C3 glucosamine hydroxyl is binding the Re centre in place of the predicted solvent molecule;27-29 the anticipated bidentate ligand shows tridentate binding to the metal. Although tridentate binding is preferable, with respect to stability, as it saturates the coordination sphere, the direct interaction between 80 a Figure 3.2. 1 3 C N M R spectra (75 M H z , ( C D 3 ) 2 S O ) of (a) R e L 6 ( C O ) 3 and (b) H L 6 . the carbohydrate and the metal is problematic for retaining biological function. Unfortunately, many of the resonances of the (3 anomer, including C 3 , could not be assigned, due to the lower concentration of the anomer in the D M S O solution. Less polar than water or methanol, D M S O is unable to stabilize the unfavourable dipole moments present in the (3 anomer. 3 0 It is unlikely the stereochemistry at C I can have an 81 effect on the geometry-dependent propensity of the C3 hydroxyl to coordinate Re, thus both anomers are predicited to bind Re in a similar manner. 3.3.2 Synthesis and Characterization of [ReL7(CO)3]Br [ReL7(CO)3]Br was prepared by refluxing the ligand, L 7 , and [Re(CO)3(H20)3]Br in methanol for 2 hours (Scheme 3.2). The complex was purified by silica chromatography using a polar eluent (1:1 ethyl acetate: methanol), hence further purification by size exclusion chromatography was necessary to remove any residual silica. Although the 'H NMR spectrum of the crude product indicated quantitative product formation, the yield after purification was 87%. Characterization of [ReL7(CO)3]Br was done by elemental analysis, mass spectrometry, IR and N M R spectroscopy. Elemental analysis confirmed the bulk composition of the sample, including a bromide counter anion and 4 waters of hydration. Waters of hydration are commonly seen in complexes containing carbohydrates. The parent peak in the mass spectrum corresponded to the cation, [ReL 7(CO) 3] +, with 100% relative intensity and the correct isotope distribution for Re ( 1 8 7Re 62.6%, 1 8 5 Re 37.4%). IR spectroscopy contained the diagnostic metal carbonyl stretches at 2019 and 1930 cm\"1 and other peaks indicative of the ligand. 82 a) [Re(CO)3(H20)3]Br, MeOH reflux, 2h b) [ 9 9 mTc(CO) 3(H 20) 3] +, PBS 75\u00C2\u00B0C, 30 min c) [ 1 8 6Re(CO) 3(H 20) 3]+, PBS 60\u00C2\u00B0C, 15 min [ML 7(CO) 3] + M = Re, 9 9 m T c , 1 8 6Re Scheme 3.2. Synthesis of [ML 7 (CO) 3 ] + , M = Re, 9 9 m T c , 1 8 6Re. H and C N M R spectra were fully assigned using 2D N M R experiments and confirmed several salient points (Figure 3.3). On binding of the ligand, numerous coordination induced shifts (CIS) were observed. Each of the pyridyl hydrogen atom resonances shifted upfield between 0.05 and 0.37 ppm, except HI 1 which shifted downfield by 0.07 ppm. Similarly, the pyridyl carbon atom resonances shifted downfield between 1.5 and 3.6 ppm, except CI 1 which shifted upfield by 0.3 ppm. Larger CIS were observed for the methylene hydrogens on carbon atoms adjacent to the coordinating tertiary nitrogen. The hydrogen atom resonances shifted downfield 1.2 ppm and 1.0 ppm for H8 and H9, respectively, while the carbon atom resonances shifted downfield 15.8 ppm and 8.0 ppm for C8 and C9, respectively. Taken together these shifts clearly 83 84 indicate the tridentate binding of the ligand and are consistent with other results for this binding moiety chelated to the [Re(CO)3]+ core.22 In contrast, resonances attributed to the carbohydrate moiety and the methylene adjacent to the sulphur atom show only small shifts upon ligand coordination, confirming the pendant nature of the carbohydrate and the non-coordinating nature of the thioether. Although the chelating moiety is symmetrical, the stereocentres in the attached carbohydrate generate an asymmetric ligand and complex, which is supported by the 'H and 1 3 C NMR spectra. In the ligand, no asymmetry is observed in the binding moiety, as only one set of resonances are observed for the pyridyl and the adjacent methylene groups. Because the H7 and H8 resonances overlap and are not well resolved from one another in the ! H NMR spectrum of the ligand, it cannot be determined if these hydrogen atoms are rendered inequivalent by the carbohydrate. Upon complexation the far reaching effect of the carbohydrate chirality is observed. In the 1 3 C N M R spectrum of the complex, two closely related resonances are observed for each CIO and CI 1 of the pyridyl rings and C9, due to the inequivalence of these atoms. As well as the C9 carbon atoms being inequivalent, the two hydrogen atoms on each of these methylenes are also rendered inequivalent on binding, due to the formation of a rigid ring with the metal centre. A complicated pattern of resonances is observed, similar to those observed in 1O 01 1 other asymmetric complexes with this binding moiety. ' ' The H NMR resonances for the H7 and H8 hydrogen atoms are well resolved in the complex 'H NMR spectrum and show an interesting coupling pattern. Again, due to the close proximity of the chiral carbohydrate, the hydrogen atoms are all inequivalent. For each hydrogen atom geminal, trans and cis couplings are observed. Each hydrogen atom is expected to generate a doublet of doublets of doublets, but due to overlapping of these resonances caused by the 85 similar geminal and trans coupling constants, each hydrogen atom is assigned a triplet of doublets (Figure 3.4). Finally, in the 1 3 C NMR spectrum, three resonances for the 4.J2 4.JO 4.08 4.06 4.04 4.02 4.O0 3.98 3.96 3.94 3.92 3.90 3.88 3.86 (ppm) F i g u r e 3.4. Region of 'H NMR spectrum (400 MHz, C D 3 C N : D 2 0 1:1) of [ReL7(CO)3]Br showing coupling pattern for H8. three carbonyl ligands are observed between 196 and 198 ppm, again illustrating the long range effect of the chiral carbohydrate reducing the symmetry of the complex, thus rendering all three carbonyl ligands inequivalent. In contrast, symmetric ligands with the same binding moiety exhibit mirror symmetry along the aliphatic amine and the trans carbonyl, which is indicated in NMR by only one set of resonances being observed for the two pyridyl groups and the adjacent methylenes, and two resonances in a 2:1 ratio observed for the three carbonyl ligands.5 , 1 7 86 3.3.3 w m T c and , 8 b Re Radiolabelling [ 9 9 m TcL 7 (CO) 3 ] + was prepared by adding 100 uL of the precursor [ 9 9 m Tc(CO) 3 (H 2 0 )3] + ( 20 MBq in saline) to 500 uX of a IO\"3 M solution of L 7 in PBS (phosphate buffer saline, pH 7.4) and heating at 75\u00C2\u00B0C for 30 min (Scheme 3.2, p. 83). Identification of the product and determination of radiochemical yield were done by HPLC, co-injecting the reaction solution with [ReL7(CO)3]Br and comparing the radiation and U V chromatograms, as other methods of characterization are precluded by the low concentrations involved. The retention time of the cationic [ 9 9 m TcL 7 (CO) 3 ] + species is similar to these of the other cationic complexes, [ 9 9 m TcL 8 ~ 9 (CO) 3 ] + , while shorter (less retained on the CI 8 column) than that of the neutral complex 9 9 m T c L 6 ( C O ) 3 . The high radiochemical yield was consistent with the 9 9 m T c labelling of other similar ligands (Table 3.1).2 1'2 2 Labelling H L 6 and L 7 \" 9 with 1 8 6 Re posed several challenges, as the Isolink kit 186 + preparation is not applicable to preparing the [ Re(CO) 3(H 20) 3] precursor. Instead the 186 A- 25 [ Re(CO) 3(H 20) 3] precursor was prepared in 80-90% yield by an alternative method adapted from the literature using B H 3 N H 3 as a reducing agent in the presence of CO gas (Scheme 3.3). The complexes 1 8 6 ReL 6 (CO) 3 and [ 1 8 6 ReL 7 - 9 (CO) 3 ] + were then prepared and characterized similarly to the 9 9 m T c analogues. For the preparation of 1 8 6 ReL 6 (CO) 3 , heating the [ 1 8 6Re(CO) 3(H 20) 3] + precursor in the presence of 10\"3 M H L 6 in PBS at 75\u00C2\u00B0C for 30 min gave optimal yields, while the preparation of [ 1 8 6ReL 7\" 9(CO) 3] +, using the same ligand concentration and buffer required lower temperatures (60\u00C2\u00B0C) and shorter 87 Table 3.1 Retention times3 and radiochemical yields for M(L 6)(CO)3 and [M(L7\" 9)(CO) 3] + (M = 9 9 m T c , 1 8 6 R e ) . Ligand U V (254 nm) Radiation detected RT / min (radiochemical yield) detected RT of _ _ * J J / \u00E2\u0080\u00A2 y y m T : mv7-standard / min 1 L K e H L 6 17.9 17 .9 (95\u00C2\u00B12) b 18.2 (94\u00C2\u00B13) L 7 13.3 13.5 (98\u00C2\u00B11) 13.5 (82\u00C2\u00B13) L 8 13.7 13.9 ( 9 9 \u00C2\u00B1 l ) c 13.8 (82\u00C2\u00B13) L 9 13.7 13.9 ( 9 9 \u00C2\u00B1 l ) c 13.9 (85\u00C2\u00B12) a HPLC run on Synergi 4um C- l 8 Hydro RP analytical column, isocratic 1 mL/min, solvent A: 0.1% w/w trifluoracetic acid in water, solvent B: acetonitrile, b ' c Results from Dr. S. Bayly and Dr. T. Storr, respectively included here for comparison.21\"23 reaction times (15 min) to limit the re-oxidation of the precursor back to perrhenate. Longer reaction times or in situ preparation of the precursor in the presence of the ligands, in a method similar to that used for 1 8 6 Re labelling in Chapter 2 (p. 50), gave unacceptably low yields and significant perrhenate formation. Identification of the product and determination of the radiochemical yields was done by HPLC, as for the 9 9 m T c complexes (Table 3.1). The radiochemical yield for 1 8 6 ReL 6 (CO) 3 was 94\u00C2\u00B13%, while only 82-85% for the dipicolylamine complexes, [ 1 8 6ReL 7\" 9(CO) 3] +. In all cases, the main impurity had a retention time consistent with perrhenate. The radiochemical 88 ['86Re04]-H34pCsaS)e \u00C2\u00BB [ , 8 6Re(CO) 3(H 20) 3] + ^ 6 o r ^ 9 \u00E2\u0080\u009E . 8 6 ^ . ^ Q r r \u00C2\u00AB W ^ C O f e ] * [ 1 8 6 ReL 8 (CO) 3 ] + [ 1 8 6 ReL 9 (CO) 3 ] + Scheme 3.3. Synthesis and solution structures of 1 8 6 ReL 6 (CO) 3 and [ 1 8 6ReL 7\" 9(CO) 3] +. yield for 1 8 6 ReL 6 (CO) 3 is actually higher than the yield for the precursor formation, suggesting further precursor formation is taking place during the preparation of 1 8 6 ReL 6 (CO) 3 . In contrast, the radiochemical yields for [ 1 8 6ReL 7\" 9(CO) 3] + are consistent with the 80-90% precursor formation followed by similar labelling yields to those with 9 9 m T c to give overall yields of between 82-85%. 3.3.4 Cysteine and Histidine Stability Challenges To assess the stability of the 9 9 m T c complexes, challenge experiments were done with cysteine and histidine, potential metal binding amino acids ubiquitous in vivo. The complexes 9 9 m T c L 6 ( C O ) 3 and [ 9 9 m TcL 7 (CO) 3 ] + were incubated in solutions containing an 89 excess of either cysteine or histidine, and the amount of complex remaining intact determined at different times points by HPLC analysis (Table 3.2). In the case of y y m T c L \u00C2\u00B0 ( C O ) 3 a 10 fold excess of the amino acid compared to the ligand was used. HPLC analysis showed the complex to be stable in either histidine or cysteine solution, but only in the short term; by 4 hours less than 30% of the complex remained intact. Histidine labelled [ 9 9 m Tc(CO) 3 (H 2 0) 3 ] + was determined to be the major decomposition product of the histidine challenge experiments. The complex instability may be due to the relatively weak binding ability of the donor atoms, especially the carbohydrate hydroxyl. Since the dipicolylamine ligand has been previously shown to be stable in cysteine and histidine solutions, a 100 fold excess of amino acid compared to the ligand was used in the challenge experiments for [ 9 9 m TcL 7 (CO) 3 ] + . Consistent with the findings for other similar compounds21'22 [ 9 9 m TcL 7 (CO) 3 ] + showed no decomposition (>95% complex intact) even after incubating for 24 h, indicating that this complex should have high stability in vivo. Table 2.3. Percentage 9 9 m Tc(L 6 ) (CO) 3 and [ 9 9 m TcL 7 (CO) 3 ] + remaining intact determined by HPLC after 1, 4, and 24 h in 10\"3 and 10\"4 M solutions of cysteine or histidine, respectively, at 37 \u00C2\u00B0C. Ligand Cysteine Histidine Th 4h 24h Th 4h 24h H L 6 88 28 0 50 24 4 L 7 99+1 99\u00C2\u00B11 99\u00C2\u00B11 99\u00C2\u00B11 99\u00C2\u00B12 95+3 90 3.3.5 Biodistribution Studies Biodistribution studies were done in tumour bearing mice with the radiolabeled compounds [ 9 9 m TcL 7 - 9 (CO) 3 ] + , to assess their potential as radiopharmaceuticals. A minimum of three mice were used for each time point, the results averaged and the standard deviations calculated (Table 3.3). The biodistribution of each of the tested radiolabeled compounds showed similarities. All compounds showed high initial activity in the liver, kidneys and small intestines. [ 9 9 m TcL 8 (CO) 3 ] + and [ 9 9 m TcL 9 (CO) 3 ] + also showed significant activity in the stomach; for [ 9 9 m TcL 9 (CO) 3 ] + , this activity was still retained in the stomach at 2 h. Otherwise, the majority of the activity cleared the kidney, liver and stomach by the 2 h time point, at which time most activity was measured in the large and small intestines. This suggests the compounds are cleared through the urinary and hepatobiliary pathways, although measurements at longer time points are needed to confirm further clearance from these tissues. If the biodistribution of the analogous 1 8 6 ' 1 8 8 Re complexes is similar, radiotherapeutic applications of these complexes would be limited by the large dose delivered to non-target tissues such as the liver, kidneys and intestines. All other organs measured had lower levels of activity which cleared over time, albeit at different rates. The lowest amount of radioactivity was measured in brain, suggesting these cationic complexes are not transported across the blood brain barrier. 91 Table 3.3. Biodistribution results in tumour bearing mice (n = 3-5) for [ 9 9 m T c L 7 ' 9 (CO)3] + expressed in % total injected dose / weight in grams of wet tissue. Time / min Organ 15 60 120 [ 9 9 m TcL 7 (CO) 3 ] + Blood 9+4 1+1 0.18+0.02 Spleen 1.4\u00C2\u00B10.4 0.6+0.3 0.6+0.1 Liver 30\u00C2\u00B12 30+10 14+3 Kidney 20\u00C2\u00B18 14+14 5.2+0.6 Lung 3\u00C2\u00B12 0.9+0.8 0.36+0.03 Stomach 4\u00C2\u00B14 3+2 2.3+0.7 Large intestines 2.4\u00C2\u00B10.3 3+2 29+11 Small intestines 24\u00C2\u00B112 52+12 40+6 Heart 2\u00C2\u00B11 0.7+0.6 0.47+0.06 Brain 0.12+0.08 0.04+0.03 0.03+0.01 Thyroid 3\u00C2\u00B13 0.9+0.9 0.5+0.2 Tumour 1.8+0.8 1+1 0.8+0.3 Muscle 1.3+0.5 0.4+0.4 2+2 [ 9 9 m TcL 8 (CO) 3 ] + Blood 4+2 0.41+0.08 0.8+0.8 Spleen 1.09+0.08 0.4+0.2 0.5+0.1 Liver 35+6 8.6+0.8 5+2 Kidney 20\u00C2\u00B13 8+2 4.3+0.3 . Lung 2.1+0.5 0.6+0.2 0.51+0.08 Stomach 12+10 5.5+0.9 1.7+0.6 Large intestines 2+2 6+4 41+21 92 Organ 15 Time / min 60 120 Small intestines 27+2 47\u00C2\u00B110 24+10 Heart 1.3\u00C2\u00B10.3 0.6+0.3 0.56+0.04 Brain 0.10\u00C2\u00B10.02 0.070.02 0.05+0.01 Thyroid 2+1 0.9+0.3 0.5+0.2 Tumour 1.9\u00C2\u00B10.6 0.5+0.1 0.4+0.1 Muscle 0.8\u00C2\u00B10.2 0.3+0.1 0.4+0.2 [ 9 9 m TcL 9 (CO) 3 ] + Blood 2.0\u00C2\u00B10.5 3+3 1+1 Spleen 0.910.2 0.6+0.2 0.6+0.2 Liver 44\u00C2\u00B115 26+4 18+2 Kidney 15\u00C2\u00B11 7+2 3.2+0.5 Lung 2.0\u00C2\u00B10.6 0.7+0.1 \" 0.6+0.2 Stomach 20\u00C2\u00B120 12+9 20+1 Large intestines 1.7+0.5 2.0+0.3 73+3 Small intestines 24\u00C2\u00B114 11+9 45+9 Heart 1.1+0.2 0.5+0.1 0.38+0.08 Brain 0.13+0.02 0.09+0.05 0.04+0.01 Thyroid 1.9+0.5 0.9+0.2 0.44+0.08 Tumour 1.44\u00C2\u00B10.08 0.38\u00C2\u00B10.04 0.26\u00C2\u00B10.03 Muscle 0.7\u00C2\u00B10.2 0.4+0.1 0.22+0.09 To better evaluate the potential for tumour imaging with the compounds [ 9 9 m T c L 7 \" 9 ( C O )3 ] + , the ratios of tumour tissue activity compared to surrounding tissue activity were considered (Figure 3.5). Tumour:blood and tumour:muscle ratios were 93 Tumour: Blood Ratio Tumour: Muscle Ratio 15 60 Time / min 120 15 60 120 Time / min Figure 3.5. Tumounblood and tumounmuscle ratios at 15, 60 and 120 min in tumour bearing mice (n = 6) for (a) [ 9 9 m TcL 7 (CO) 3 ] + , (b) [ 9 9 m TcL 8 (CO) 3 ] + and (c) [ 9 9 m TcL 9 (CO) 3 ] + . determined for each mouse, the results averaged and the standard deviations calculated (this method is more representative than calculating ratios from already averaged data). Over time, the activity in both the blood and tumour decreases. For [ 9 9 m TcL 7 (CO) 3 ] + and [ TcL (CO)3] the activity clears faster from the blood than from the tumour, thus the tumounblood ratio increases, peaking at the last time point measured (2 h) with values of 2.6 and 1.2, respectively. The higher activity in the tumour compared to in the blood at the 2 h time point suggests that the tumour will be observable via imaging. For [ 9 9 m TcL 9 (CO) 3 ] + , the tumounblood ratio remains between 0.5 and 0.7 throughout, 9 4 indicating that the blood and tumour activity are clearing at the same rate. Because the ratio is less than 1, meaning more activity is present in the blood, [ 9 9 m TcL 9 (CO)3] + will not likely allow differentiation of the blood and the tumour tissue. As well, the parallel clearance rates suggest that the activity observed in the tumour may be due to blood flow in the tumour vasculature and not actual tumour uptake. The tumour.muscle ratios were above 1 at all time points for all compounds, indicating the tumour and muscle tissue will be differentiated from one another. In general the tumour:muscle ratios decreased over time with the lowest ratio values, between 1.2 and 1.4, observed at the 2 h time point. Since the highest tumounblood ratios are not well matched with the highest tumour:muscle ratios, these ratios need to be optimized together for tumour imaging. 3.3.6 Planar Scintigraphy Acquisition of planar scintigraphic images of tumour bearing mice using [ 9 9 m TcL 7 \" 9 (CO)3] + was undertaken. Due to mortality issues, only one set of images for one compound, [ 9 9 m TcL 9 (CO) 3 ] H was obtained. Many aspects of the imaging experiment correlated with the findings of the biodistribution studies. High initial activity was observed in the kidney, liver and bladder. The liver and kidney activity decreased slightly over time, while the bladder activity increased (Figure 3.6). While biodistribution studies suggested that it would not be possible to differentiate the tumour from blood, the tumour can clearly be seen in the image (Figure 3.7). A closer look at the activity in the blood (estimated from the head activity) compared to the tumour shows 95 M o u s e Act iv i ty Levels - S m o o t h e d T ime (s) Figure 3 . 6 . Activity (counts/pixel) vs. time curves from planar scintigraphy images using [ 9 9 m TcL 9 (CO) 3 ] + . Figure 3 .7 . Planar scintigaphy image of a tumour bearing mouse 2.5 h after injection of [ 9 9 m TcL 9 (CO) 3 with the visible tumour circled in red. (Residual activity in and around the tail due to problems with tail vein injection). 96 that the tumour activity is higher, but that they clear at the same rate (Figure 3.8). The parallel clearance agrees with the biodistribution results, and suggests that the tumour activity is due to the larger blood volume in the increased vasculature around the tumour, In the biodistribution studies, a portion of this vasculature, that is not within the tumour mass, would not be measured, explaining the discrepancy in the low tumounblood ratio compared to the unexpectedly discernable tumour images. An alternative explanation is the different tumour types used in the biodistribution and planar scintigraphy studies have different levels of vasculature. Mouse Activity Levels - Smoothed i i i 1 1 1 1 r 0,6 -0 I I I I l I I I I I I 0 200 400 600 800 1000 1200 1400 1600 1800 2000 Time (s) Figure 3.8. Activity (counts/pixel) vs. time curves from planar scintigraphy images using [ 9 9 m TcL 9 (CO) 3 ] + . (Blood activity was estimated as activity observed in head). 3.4 Conclusions This chapter continued our work on pendant carbohydrate bearing tridentate ligands for the [M(CO)3] + core. The expected bidentate pro-ligand HL 6 , was shown by 97 NMR spectroscopy to bind the [M(CO)3] + core in a tridentate fashion, albeit due to the carbohydrate binding to the metal centre. In contrast, the purposely tridentate L 7 ligand formed a cationic asymmetric complex with the carbohydrate remaining pendant. Although radiolabelling with H L 6 with both 9 9 m T c and 1 8 6 Re gave high radiochemical yields, cysteine and histidine challenge experiments suggested low stability for the 9 9 m T c L 6 ( C O ) 3 species. In this and previous work, the dipicolylamine based ligands gave high radiochemical yields with 9 9 m T c and slightly lower yields with 1 8 6 Re, but the 9 9 m T c complexes were highly stable in the cysteine and histidine challenge experiments. Biodistribution of [ 9 9 m TcL 7 - 9 (CO) 3 ] + and planar scintigraphy with [ 9 9 m TcL 9 (CO) 3 ] + suggested the complexes were being cleared through the excretory organs, the liver and kidneys. The increase of the tumounblood ratios over time suggested that [ 9 9 m TcL 7 (CO) 3 ] + and [ 9 9 m TcL 8 (CO) 3 ] + may be useful tumour imaging agents. Despite the ability to visualize the tumour using [ 9 9 m TcL 9 (CO) 3 ] + in the planar scinitgraphy study, the parallel tumour and blood clearance in both the imaging and biodistribution results for this compound suggested that tumour vasculature was observed rather than true tumour uptake. 3.5 References 1. Alberto, R.; Pak, J. K.; Staveren, D. v.; Mundwiler, S.; Benny, P., Biopolymers 2004, 76, 324-333. 2. Schibli, R.; Bella, R. L.; Alberto, R.; Garcia-Garayoa, E.; Ortner, K.; Abram, U.; Schubiger, P. A., Bioconjugate Chem. 2000,11, 345-351. 98 3. Alberto, R.; Schibli, R.; Schubiger, P. A., J. Am. Chem. Soc. 1999,121, 6076-6077. 4. Alberto, R.; Schibli, R.; Waibel, R.; Abram, T J . ; Schubiger, A. P., Coord. Chem. Rev. 1999, 99, 901-919. 5. Banerjee, S. R.; Levadala, M . K.; Lazarova, N.; Wei, L.; Valliant, J. F.; Stephenson, K. A.; Babich, J. W.; Maresca, K. P.; Zubieta, J., Inorg. Chem. 2002, 41, 6417-6425. 6. Rattat, D.; Eraets, K.; Cleynhens, B.; Knight, H.; Fonge, H.; Verbruggen, A., Tetrahedron Lett. 2004, 45, 2531-2534. 7. Luyt, L. G.; Bigott, H. M . ; Welch, M . J.; Katzenellenbogen, J. A., Bioorg. & Med. Chem. 2003,11, 4977-4989. 8. van Steveren, D. R.; Benny, P. D.; Waibel, R.; Kurz, P.; Pak, J. K.; Alberto, R., Helv. Chim. Acta 2005, 88, 447-459. 9. Lazarova, N.; Babich, J.; Valliant, J.; Schaffer, P.; James, S.; Zubieta, J., Inorg. Chem. 2005, 44, 6763-6770. 10. He, H.; Lipowska, M . ; Xu, X.; Taylor, A. T.; Carlone, M . ; Marzilli, L. G., Inorg. Chem. 2005, 44, 5437-5446. 11. Santos, I. G.; Abram, U.; Alberto, R.; Lopez, E. V.; Sanchez, A., Inorg. Chem. 2004,45,1834-1836. 12. Wei, L.; Banerjee, S. R.; Levadala, M . K.; Babich, J.; Zubieta, J., Inorg. Chem. Comm. 2003, 6, 1099-1103. 13. Pak, J. K.; Benny, P.; Spingler, B.; Ortner, K.; Alberto, R., Chem. Eur. J. 2003, 9, 2053-2061. 99 14. van Steveren, D. R.; Mundwiler, S.; Hoffmanns, U.; Pak, J. K.; Spingler, B.; Metzler-Nolte, N.; Alberto, R., Org. Biomol. Chem. 2004, 2, 2593-2603. 15. van Steveren, D. R.; Waibel, R.; Mundwiler, S.; Schubiger, P. A.; Alberto, R., J. Organomet. Chem. 2004, 689, 4803-4810. 16. Waibel, R.; Alberto, R.; Willuda, J.; Finnern, R.; Schibli, R.; Stichelberger, A.; Egli, A.; Abram, U.; Mach, J.-P.; Pluckthun, A.; Schubiger, P. A., Nature Biotech. 1999, 17. 897-901. 17. Banerjee, S. R.; Schaffer, P.; Babich, J.; Valliant, J.; Zubieta, J., Dalton Trans. 2005, 3886-3897. 18. Wei, L.; Banerjee, S. R.; Levadala, M . K.; Babich, J.; Zubieta, J., Inorg. Chim. ActalWA, 357, 1499-1516. 19. Wei, L.; Babich, J.; Eckelman, W. C ; Zubieta, J., Inorg. Chem. 2005, 44, 2198-2209. 20. Stephenson, K. A.; Zubieta, J.; Banerjee, S. R.; Levadala, M . K.; Taggart, L.; Ryan, L.; McFarlane, N.; Boreham, D. R.; Maresca, K. P.; Babich, J.; Valliant, J., Bioconjugate Chem. 2004,15, 128-136. 21. Storr, T.; Fisher, C. L.; Mikata, Y.; Yano, S.; Adam, M . J.; Orvig, C , Dalton Trans. 2005, 654-655. 22. Storr, T.; Obata, M . ; Fisher, C. L.; Bayly, S. R.; Green, D. E.; Brudzifiska, I.; Mikata, Y.; Patrick, B. O.; Adam, M . J.; Yano, S.; Orvig, C., Chem. Eur. J. 2005,11, 195-203. 23. Bayly, S. R.; Fisher, C. L.; Storr, T.; Adam, M . J.; Orvig, C., Bioconjugate Chem. 2004,15, 923-926. 24. Mikata, Y.; Sugai, Y.; Yano, S., Inorg. Chem. 2004, 43, 4778-4780. 100 25. Schibli, R.; Schwarzbach, R.; Alberto, R.; Ortner, K.; Schmalle, H.; Dumas, C ; Egli, A.; Schubiger, P. A., Bioconjugate Chem. 2002,13, 750-756. 26. Klufers, P.; Kunte, T.,Agnew, Chem. Int. Ed. 2001, 40, 4210-4212. 27. Andrews, M . A.; Voss, E. J.; Gould, G. L.; Klooster, W. T.; Koetze, T. F., J. Am. Chem. Soc. 1994, 116, 5730-5740. 28. Klufers, P.; Krotz, O.; OBberger, M . , Eur. J. Inorg. Chem. 2002, 1919-1923. 29. Klufers, P.; Kunte, T., Chem. Eur. J. 2003, 9, 2013-2018. 30. Prayly, J. P.; Lemieux, R. U., Can. J. Chem. 1987, 65, 213-223. 101 Chapter 4 Glucosamine Conjugates of the Cyclopentadienyltricarbonylrhenium(I) and Technetium(I) Cores* 4.1 Introduction The 'piano stool' organometallic core (CpM(CO)3 Cp = cyclopentadienyl, M = Re(I), Tc(I)) combines stability, small size and accessibility for preparing 9 9 m T c complexes.1\"6 This core is a neutral 18 electron species, with high stability due to the low spin d 6 electron configuration further stabilized by the Cp and CO ligands. Although many highly stable chelate systems have been developed for 9 9 m T c , 7 ' 8 the small size of the 'piano stool' core should be advantageous for retaining biological activity, particularly when labelling small molecules such as glucose. Several routes to CpM(CO)3 derivatives have been reported,1\"6 including one step radiolabelling procedures.9 Clearly, the CpM(CO)3 core is a very attractive candidate for preparing 9 9 m T c complexes containing glucose. Two related synthetic routes, the double ligand transfer (DLT) 4\" 6 and the single ligand transfer (SLT), 9 ' 1 0 have been reported for synthesizing CpM(CO)3 containing complexes on both the macroscopic and tracer scales (Scheme 4.1). The D L T route was * A version of this chapter has been accepted for publication. Ferreira, C. L. , Bayly, S. R., Patrick, B. O., Steele, J. S., Adam, M . J., Orvig, C. Glucosamine Conjugates of the Cyclopentadienyltricarbonylrhenium(I) and Technetium(I) Cores Inorg. Chem. 2006, In press. 102 first reported by Wenzel,5'6 and has since been improved by Katzenellenbogen.4 Advantages of this route include the short reaction time, important when using radiotracers with short half-lives, and the use of [MO4]\" (M = Re, Tc) as starting material, R = substituent M = Re Tc R' = H or R Scheme 4.1. General double ligand transfer (DLT) 4 - 6 and single ligand transfer (SLT) reactions:9' 1 0 For DLT, [ M 0 4 ] \ CrCl 3 , Cr(CO) 6, MeOH, 1 h, 160\u00C2\u00B0C; for SLT, [M(CO) 3 (H 2 0) 3 ] + , DMSO /H2O, 4 h , 95\u00C2\u00B0C. which is the form in which 9 9 m T c is acquired from hospital generators. More recently, the SLT route has been reported,9'10 which uses the [M(CO ) 3 (H20) 3 ] + core developed by Alberto and co-workers.11 The [ 9 9 m Tc(CO ) 3 (H20) 3 ] + complex can be prepared in high yield via a commercially available kit, and removes the need for the carbonyl ligand donor and reducing agent used in the DLT. Both routes require a ferrocene derivative, from which the substituted Cp ring originates. By functionalizing the Cp ring with a biomolecule, the biomolecule can be incorporated into the final 'piano stool' type product. Because numerous ferrocenoyl glucose conjugates are known in the literature,12\"14 both the DLT and SLT are potential routes to 9 9 m T c glucose containing complexes. In this work, we describe the modification of the DLT and SLT towards obtaining glucose derivative containing compounds of the CpM(CO) 3 core. Synthesis of 103 glucosamine-CpM(CO)3 complexes on the macroscopic (M = Re) scale and radiotracer (M = 9 9 m Tc) scale is described (Figure 4.1), the solid state structure of 2a was determined, and the biological activity related to the glucose metabolism enzyme hexokinase was evaluated and compared to FDG to determine radiopharmaceutical potential. Figure 4.1. Structures of compounds prepared in Chapter 4. 4.2 Experimental 4.2.1 Materials and Methods Most related information is contained in section 2.2.1. Tricarbonyl(cyclopentadienyl carboxylic acid)rhenium,15'16 [NEt4]2[Re(CO)3Br3],17 1,1'-bis(methoxycarbonyl)ferrocene4 and 2-/V-(l,3,4,6-tetra-0-acetyl-2-amino-2-deoxy-D-glucopyranose)-l-ferrocene carboxamide12 were prepared as reported in the literature. Potassium perrhenate was purchased from Strem Chemicals (Newburyport, MA). For high temperature and high pressure reactions, a cylindrical aluminum reaction block was R O R = acetyl, M = Re (1a), M = y 9 m T c (1b) R = H, M = Re (2a), M = 9 9 m T c (2b) 104 used, with holes fitted for high pressure thick wall glass tubes (Ace Glass) and an IKA probe, and with an aluminum cover that could be secured with screws. 4.2.2 Modification of the D L T The D L T was modified from the published methods4 by varying the solvent and/or temperature (Scheme 4.1). In brief, KRe0 4 (29 mg, 0.10 mmol), CrCl 3 (80 mg, 0.50 mmol), Cr(CO)6 (140 mg, 0.63 mmol), l,r-bis(methoxycarbonyl)ferrocene (90 mg, 0.30 mmol) and a small stir bar were added to a 4 mL pressure tube, which was then purged with N 2 gas. Solvent (1 mL) was added, and the pressure tube was sealed with a teflon cap. The pressure tube was placed in an aluminum reaction block filled partially with high temperature heating oil, and an aluminum cover screwed on top of the block. The block was placed on a heating/stirring plate, stirring was started, and an IKA probe was inserted into an appropriate hole in the block, partially filled with high temperature heating oil. Using the IKA probe, the reaction temperature was set. After heating for one hour at the desired reaction temperature, heating was stopped and the block was allowed to cool. Once the block was well below 100 \u00C2\u00B0C, the aluminum cover was carefully removed, the pressure tubes were gently transferred to a dry ice/ methanol bath for 10 min. and then to an ice bath for 10 min. The pressure tubes were opened and the contents transferred to a round bottom flask rinsing with methanol. The solvent volume was reduced by rotary evaporation to a dark viscous residue. The residue was dissolved in dichloromethane and the solution clarified by filtration. The filtrate solvent volume was reduced by rotary evaporation, and the reaction yield was estimated by comparing 105 the integration of the starting material and product resonances in the H NMR spectrum of the crude reaction mixture. 4.2.3 SLT Using Alternative Rhenium Starting Materials The SLT reaction was modified from that in the literature (Scheme 4.1).10 A 4 mL pressure tube containing acetyl ferrocene (33 mg, 0.15 mmol) and rhenium starting material, either Re(CO)5Br (19 mg, 0.047 mmol) or [NEt4]2[Re(CO)3Br3] (37 mg, 0.048 mmol), was purged with N 2 gas. Anhydrous methanol was added (1 mL), and the tube sealed with a teflon cap before being placed in the aluminum heating block. The reaction was stirred and heated at 160 \u00C2\u00B0C for 1 h. After the aluminum block had cooled to less than 100 \u00C2\u00B0C, the pressure tube was removed and cooled in an ice bath for 10 min. Yields were estimated from the 'H N M R spectra of the crude reaction mixtures by comparing the integration of resonances due to acetyl ferrocene and the expected product. Estimated yields were 40% and 16% for the reactions using Re(CO)5Br and [NEt4]2[Re(CO)3Br3], respectively. 106 4.2.4 Synthesis of Glucosamine Conjugates of the Tricarbonyl cyclopentadienylrhenium(I) Core Tricarbonyl{N-(l,3,4,6-tetra-0-acetyl-2-amino-2-deoxy-P-D-glucopyranose)cyclopentadienyl carboxamide}rhenium(I) (la). 1 -Hydroxybenzyltriazole (20 mg, 0.15 mmol), N-(3-AcO Ij^o dimethylaminopropyl)-N'-ethylcarbodiimide (29 mg, 0.15 mmol) AcO' - ^ O A c A C \u00C2\u00B0 3 J^L]G=4\ a n < ^ tricarbonyl(cyclopentadienyl carboxylic acid)rhenium(I) (40 oc' j^co mg> 0-10 mmol) were weighed into a flask. The flask was purged with nitrogen gas, freshly distilled dichloromethane (15 mL) was added, and the reaction mixture was stirred at RT. After 3 h, 1,3,4,6-tetra-O-acetyl-P-D-glucosamine hydrochloride salt (44 mg, 0.12 mmol) and N,N-diisopropylethylamine (0.2 mL, 1.2 mmol) were added and the reaction was refluxed overnight. The dark mixture was reduced on a rotary evaporator to a red oil, which was redissolved in dichloromethane (20 mL) and washed with water (3 x 30 mL). The organic layer was separated, dried with MgSC\u00C2\u00BB4, and the solvent removed with a rotary evaporator. The product was obtained as an off-white solid after purification by silica chromatography eluting with 7:3 ethyl acetate to hexanes (rf ~ 0.5), and drying in vacuo for 24 h (31 mg, 44 %). 'H NMR (CDC13, 300 MHz): 5 2.04, 2.05, 2.07, 2.10 (4 s, 12H, C//3COO), 3.85 (dd, 1H, H5, \ 5 = 9.74 Hz, 3J5,6a = 2.59 Hz),4.12 (dd, 1H, #6a, 2J6a,6b = 12.64 Hz), 4.24 (dd, 1H, H6b, 3 J 5 > 6 b = 4.87 Hz, 3 J 6 a ; 6 b = 12.54 Hz), 4.40 (dd, 1H, HI, 3Jh2= 9.14 Hz, 3 J 2 ( 3 = 10.35 Hz), 5.13 (t, 1H, H4, V 3 >4 = 9.59 Hz, 3 J 4 , 5 = 9.75 Hz), 5.23 (t, 1H, H3,3J2j3 = 10.51 Hz, 3J 3 >4 = 9.59 Hz), 5.30, 5.36 (2d, 2H, H10, #10', 3 J 9 , i 0 = 1.59 Hz), 5.65 (d, 1H, HI, 3 J 1 ; 2 = 8.84 107 Hz), 5.86, 5.94 (2d, 2H, H9, H9', V 9 , i o = 1.37 Hz), 6.32 (d, 1H, N i / , 3J NH ,2 = 9.29 Hz). 1 3 C NMR (CDC13, 75 MHz): 5 20.54, 20.70, 20.78, 20.83 ( C H 3 C O O ) 52.94 (CT), 61.70 (C6), 67.88 (C3), 72.58 (CA), 72.90 (CS), 84.11, 85.61, 85.88, 87.40 (C9, C9', CIO, CIO'), 92.65 (CI), 92.86 (CS), 162.32 (CT), 169.28, 169.58, 170.66, 171.97 (CH3COO), 192.13 (3 CO). IR (cm-1, NaCl plate): 2027 (s), 1931 (s, br), (v(/ac-Re(CO)3), 1752 (s), (v(CH3CO)), 1625 (s) (v(CONH)). MS (ESI+): m/z (relative intensity) = 732, 730 ([M+Na]+, 100). MS (ESI-): m/z (relative intensity) = 708, 706 ([M-l]\", 28). Anal. Calcd. for C23H 24N0 1 3Re: C, 38.98; H, 3.41; N, 1.97. Found: C, 39.11; H, 3.30; N, 2.29. Tricarbonyl{N-(2-amino-2-deoxy-D-glucopyranose)cyclopentadienyl carboxamide}rhenium(I) trihydrate (2a). H O 1-Hydroxybenzyltriazole (20 mg, 0.15 mmol), N,N'-H ^ O - V ^ J H ^ dicyclohexylcarbodiimide (23 mg, 0.11 mmol) and tricarbonyl(cyclopentadienyl carboxylic acid)rhenium(I) (40 mg, O C ' j ^ C O 0.10 mmol) were weighed into a flask. After purging the flask with nitrogen, freshly distilled dichloromethane (15 mL) was added and the reaction mixture stirred. After 3 h, glucosamine hydrochloride (23 mg, 0.11 mmol), N,N-diisopropylethylamine (0.2 mL, 1.2 mmol) and dimethylformamide (15 mL) were added to the reaction mixture, and stirring was continued overnight. The volume was reduced on a rotary evaporator and the residue vacuum dried to obtain a dark thick oil, which was dissolved in water (20 mL), filtered, and washed with water (10 mL). The aqueous filtrate was reduced in volume on a rotary evaporator and vaccum dried, before purification by silica column chromatography eluting with 8:2 ethyl acetate:methanol (rf 108 ~ 0.6). The off-white solid obtained was dried in vacuo for 24 h (32 mg, 66%). *H NMR (CD 3 OD:D 2 0 1:1, 400 MHz): 5 3.4-3.46 (m, 1.3H, H5a, H6&), 3.60 (m, 0.3H, //5P), 3.7-3.8 (m, 2.6H, //2p, //4p, H6a), 3.8-3.9 (m, 1.7H, H3, //4a), 3.95 (dd, 0.7H, H2a, 3Ji, 2 = 3.2 Hz, V 2 > 3 = 10.73 Hz), 4.70 (d, 0.3H, / / lp , 3 J 1 > 2 = 8.37 Hz), 5.19 (d, 1H, / / l a , 3 J l j 2 = 3.2 Hz), 5.59 (s, 2H, H10, HW), 6.20, 6.25 (2s, 2H, H9, H9'). 1 3 C NMR (CD 3 OD:D 2 0 1:1, 75 MHz): 5 55.97 (C2a), 58.54 (C2p), 62.46 (C6a), 62.53 (C6p), 71.85 (C3a), 72.01 (C3p), 72.17 (C5a), 72.98 (C4a), 75.33 (C5P), 77.68 (C4p), 86.29, 86.53, 87.84, 88.46 (C9p, ClOp, C9'p, ClO'p), 86.42, 86.73, 87.68, 88.58 (C9a, ClOa, C9'a, ClO'a), 92.29 (Cla), 95.26 (C8a), 95.57 (C8p), 96.61 (Cip), 165.46 (CI), 194.10 (3CO). IR (cm\"1, NaCl plate): 2028 (s), 1931 (s, br), (v(/ac-Re(CO)3), 1625 (s) (v(CONH)). MS (ESI+): m/z (relative intensity) = 564, 562 ([M+Na]+, 100), 542, 540 ([M+l]+, 45). MS (ESI-): m/z (relative intensity)= 541 ([M]\", 100). Anal. Calcd. for Ci 5 Hi 6 N0 9 Re-3H 2 0: C, 30.30; H, 3.73; N, 2.36. Found: C, 30.61; H, 3.30; N, 2.49. 4.2.5 Solid State Structure Determination of 2a Crystals were grown by slow evaporation of a MeOH/H 2 0 solution. The data was collected and the structure solved by Dr. Brian O. Patrick using methods described in section 2.2.5. The material was a mixture of anomers, with 92% of the (R) configuration and 8% of the (S) configuration at CI. The atoms of the minor fragment were refined isotropically, all other non-hydrogen atoms were refined anisotropically and the hydrogen atoms were included in calculated positions. 109 4.2.6 Hexokinase Substrate Assay Compound 2a was qualitatively tested as a substrate for hexokinase (Scheme 4.2) as described in section 2.2.6. 4.2.7 Hexokinase Inhibition Assay Compound 2a and 2-deoxy-2-fluoro-D-glucopyranose (FDG) were tested for their capacity to inhibit glucose phosphorylation by hexokinase, and their inhibition constants were determined (Scheme 4.2). In a cuvet, 500 pL glucose (HK) assay kit (Sigma) was pipetted into a solution containing either 2a or FDG, and glucose (total volume 750 pL). The kit contains HK, ATP, from which the phosphate is acquired, 6-phosphate-glucosedehydrogenase and N A D + . The reaction rate of glucose phosphorylation is measured via the second enzymatic reaction and the conversion of N A D to NADH. The solution was monitored by UV spectroscopy at 340 nm every 30 seconds for 10 minutes, to determine the reaction rate via the formation of NADH. The assay was repeated at several concentrations of 2a or FDG (0-2850 mM) and glucose (20-140 mM). Using Microsoft Excel linear regression, Kj values were determined by first plotting the double reciprocal plot of 1/reaction rate versus 1/[substrate] (Lineweaver-Burke plot) for each concentration of 2 or FDG, and then plotting the slope of each line versus the [inhibitor], and calculating Kj values as the -slope/intercept for the line (Figures 4.2-4.4). Error was determined using Microsoft Excel statistical methods for least squares approximations. 110 OH HO HO OPO,2\" O Hexokinase OH r \ * HO HO ATP ADP OP032\" Q 6GP-dehydrogenase _ L Q OH f ^ * H F H O V ^ ' NAD+ NADH Scheme 4.2. Enzymatic reaction used in hexokinase substrate and inhibition assays. Figure 4.2. Lineweaver-Burke plot (double reciprocal plot of 1/initial reaction rate vs. 1/substrate concentration at different FDG inhibitor concentrations) for inhibition of hexokinase by FDG. I l l 6 0 0 0 0 -I , , , , , 0 0 . 0 0 5 0.01 0 . 0 1 5 0 .02 0 . 0 2 5 1/[S] (1/micromolar) Figure 4.3. Lineweaver-Burke plot (double reciprocal plot of 1/initial reaction rate vs. 1/substrate concentration at different 2a inhibitor concentrations) for inhibition of hexokinase by 2a. 3 0 0 0 0 0 N -1200 -700 -200 300 8 0 0 1300 1800 2 3 0 0 2 8 0 0 3300 [I] / (micromolar) Figure 4.4. Ki determination for 2a and FDG (Lineweaver-Burke plot slope vs. inhibitor concentration). 112 4.2.8 w m T c Radiolabelling Complexes were prepared using a modified SLT reaction.10 [ 9 9 m Tc(CO )3 (H 2 0 )3] + was prepared using Isolink kits and 200 MBq of [99mTcC>4]\" in 1 mL saline. A solution of 2-A -^(l ,3,4,6-tetra-0-acetyl-2-amino-2-deoxy-D-glucopyranose)-l -ferrocene carboxamide (0.5 mL, 10\"2 M in DMSO) was sealed in a 3 mL vial and purged with N 2 gas; 100 uL of the [ 9 9 m Tc(CO) 3 (H 2 0 )3] + solution was then added. The vial was heated to 60-80 \u00C2\u00B0C for 1-3 h, cooled to room temperature, and analyzed by HPLC to verify the presence of lb. To prepare 2b, 600 uL of 0.1 M sodium methoxide in methanol was added to the reaction vial, followed by 0.1 M HCI to neutralize. The radiochemical yield was determined by HPLC (gradient 9:1 0.1% trifluoroacetic acid w/w in watenacetonitirile to 100% acetonitrile over 30 min.) using 2a as a standard. 4.3 Results and Discussion 4.3.1 Modifications to the DLT and SLT Initial attempts to use the DLT reaction with ferrocenyl carbohydrate conjugates encountered several difficulties. Methanol was determined to be the best solvent for the reaction by Katzenellenbogen and coworkers,4 but in our system, when using ferrocenyl-carbohydrate conjugates, the carbohydrate functional groups, either hydroxyls or protecting groups such as acetyls, were replaced by methoxy groups under the high 113 temperature and pressure conditions. As well, other decomposition of the ferrocenyl-carbohydrate conjugates was observed. To overcome these difficulties, the D L T reaction conditions were modified to lower the reaction temperature and incorporate non-nucleophilic solvents. Trial reactions were done using l,l'-bis(methoxy carbonyl)ferrocene, which has been successfully used in the D L T previously.4 DMF, THF and C H 3 C N have been used as solvents in the D L T reaction with acetyl ferrocene, but gave no yield of the desired product in this system. 4 In our work, using 1,1'-bis(methoxy carbonyl)ferrocene at 165\u00C2\u00B0C, reasonable yields were obtained of 35% and 10% using anhydrous DMF and THF, respectively, but no yield using C H 3 C N or any wet solvent system. Good yields of 65% were also obtained using D M F at only 130\u00C2\u00B0C. Unfortunately, even at lower temperatures and using DMF, the ferrocenyl-carbohydrate complexes yielded no desired product. Because the SLT reaction, which also utilizes ferrocenyl compounds, uses a [M(CO)3+]-containing starting material, it requires neither a carbonyl donor nor a reducing agent, which may reduce the ferrocenyl carbohydrate conjugate decomposition observed in the D L T reaction. The reported Re starting material used in the SLT, [Re(CO)6][BF4], is not commercially available, so we investigated the potential of other Re(I)(CO)3 starting materials.10'18 The SLT reactions using acetyl ferrocene and Re(CO)sBr, which is commercially available, or [NEt4]2[Re(CO)3Br3], which is easily prepared,17 gave comparable yields to the reported [Re(CO)6][BF4] starting material.9'10' 18 Again, reactions with the ferrocenyl-carbohydrate conjugates did not yield the desired product due to decomposition of the starting material. 114 4.3.2 Synthesis and Characterization of Glucosamine Conjugates of the TricarbonylcycIopentadienylrhenium(I) Core Because the ferrocenyl-carbohydrate conjugates were not stable under the DLT or SLT reaction conditions, an alternative route of synthesis was considered (Scheme 4.3). An indirect D L T method has been reported for peptide labelling,1 5'1 6 where the DLT reaction is used to prepare tricarbonyl(cyclopentadienyl methoxycarbonyl)rhenium(I), the methyl ester is converted to a carboxylic acid and then attached to the peptide by RO \u00C2\u00B0 C C O ^ C O N H 2 I oc'Epo C O Scheme 4.3. Synthesis of la and 2a. Reaction conditions for R = Ac (la), HOBt, EDC, diisopropylethylamine, CH 2 C1 2 , reflux overnight, for R = H (2a), HOBt, DCC, diisopropylethylamine, CH 2 C1 2 , DMF. amide coupling. This method is attractive for the synthesis of aminoglucose conjugates, especially of glucosamine which is known to have similar biochemistry to glucose and has been utilized in other promising Tc-conjugates.19'20 Using the D L T reaction, tricarbonyl(cyclopentadienyl carboxylic acid)rhenium(I) was prepared by a literature method.1 5'1 6 An activated ester was prepared in situ using HOBt with either DCC or EDC as activating agents. DCC (organic soluble) was used when preparing the non-115 protected conjugate 2a (water soluble) and EDC (water soluble) for the protected conjugate la (organic soluble) to facilitate easy removal by extraction of the byproduct urea. Either glucosamine or l,3,4,6-tetra-0-acetyl-(3-D-glucopyranose was added with base to the reaction to form 2a and la, respectively. The coupling reaction to form la required heating to obtain acceptable yields, which may be attributed to the larger steric hindrance of the adjacent acetyl groups around the amine. After purification by column chromatography, la and 2a were isolated as off-white solids in 44% and 66% yields, respectively. Both la and 2a were fully characterized by the usual methods. Positive electrospray mass spectra contained the mass peaks [M+Na]+ at 100% relative intensity with the correct isotopic pattern for Re (62.6% 1 8 7Re, 37.4% 1 8 5Re). Elemental analysis confirmed the purity and bulk composition of the products. The elemental analysis of 2a showed three water molecules associated with the compound; not surprisingly, free carbohydrate moieties often have associated water molecules, even after considerable drying. The IR spectra of the compounds confirmed both the presence of CO ligands and the formation of amide bonds. Two peaks diagnostic of G=0 stretches were found at 2027 and 1931 cm\"1; up to three peaks could be expected, but two peaks suggest three CO ligands with Civ local symmetry at the metal centre (yielding two metal CO stretching frequencies, where one is doubly degenerate) or with two peaks overlapping to give the slightly broader peak at 1931 cm\"1. The IR spectrum of la has peaks from both ester (1727 cm\"1) and amide (1625 cm\"1) functionalities due to the protecting acetyl groups and the amide link to the metal core, respectively. The IR spectrum of 2a has only a peak due to the amide functionality, confirming the formation of an amide bond at 116 the C2 position of glucosamine and not an ester bond with the hydroxyl functional groups. The preference for amide bond formation was expected due to the higher nucleophilicity of primary amines compared to alcohols. 'H and 1 3 C NMR spectra were fully assigned for both compounds, confirm the pendant nature of the carbohydrate and show the chirality induced by the carbohydrate. All expected resonances were observed, including one broad 1 3 C resonance near 200 ppm assignable to the CO ligands. For 2a the free hydroxyl group at the CI position of the glucosamine is free to mutarotate between the a and P anomers. As the J H resonances of the two anomers are overlapping and hard to differentiate, ID TOCSY N M R was utilized, irradiating the CI hydrogen atoms of each anomer in turn and acquiring the resonances associated within the same anomer ring system (Figure 4.5). Together with the 2D COSY spectrum, the complete assignment of hydrogen resonances for both anomers was accomplished. The coordination of the glucosamine moieties through the amine functionality is confirmed in the ! H NMR spectra of each compound, as the C2 hydrogen resonances are significantly shifted, up to 0.5 ppm downfield, from the resonances observed for free glucosamine. All other carbohydrate associated resonances are only slightly shifted, suggesting that the glucosamine is pendant and not interacting with the metal. The chirality of the glucosamine moiety induces splitting of the cyclopentadienyl (Cp) ring hydrogen and carbon resonances. In a substituted Cp ring 117 a 5.4 ' S . O ' 4 . 6 ' ' 4 . 2 ' ' 3 . 8 ' ' ' 3.4 Figure 4.5. TOCSY ' H N M R spectra (400 MHz, C D 3 O D : D 2 0 1:1) of 2a for p (a) and a (b) anomers and overall 'H NMR spectrum (c) for the region of glucose hydrogen atom resonances. 118 where the substituent is achiral there are two sets of equivalent C H groups, and only two sets of hydrogen and carbon atom resonances are observed. In la and 2a, the substituent, glucosamine, contains many chiral centers. Compound la is a single isomer, while two anomers are observed for 2a. This chirality renders all the hydrogen and carbon atoms in the Cp rings inequivalent. For la four resonances are observed for the C H groups of the Cp ring in both the *H and 1 3 C NMR spectra (Figure 4.6). For 2a, the spectra are further complicated by the two anomers and ten resonances are observed for the carbon atoms of the Cp ring (Figure 4.7). The 'H NMR spectrum for 2a does not resolve all eight hydrogen atoms of the Cp ring; three broad resonances are observed. \u00E2\u0080\u0094] i i i i i i i i > i i i i i i i \u00E2\u0080\u0094 ' \u00E2\u0080\u0094 i \u00E2\u0080\u0094 i \u00E2\u0080\u0094 i \u00E2\u0080\u0094 i \u00E2\u0080\u0094 i \u00E2\u0080\u0094 i \u00E2\u0080\u0094 i \u00E2\u0080\u0094 ' \u00E2\u0080\u0094 | \u00E2\u0080\u0094 6.6 6.4 6.2 6.0 5.8 5.6 5.4 5.2 5.0 4.8 4.6 4.4 4.2 4.0 Figure 4.6. *H N M R spectrum (300 MHz, CDCI3) of la showing 4 resonances (stars) associated with the inequivalent hydrogen atoms on the cyclopentadienyl ring. 119 C1o 98 96 94 92 I 1 I 90 88 86 ppm . J 200 180 160 140 120 100 80 60 40 ppm Figure 4.7. l 3 C N M R spectrum (75 MHz, CD 3 OD:D 2 0 1:1) of 2a with 5 major (a anomer, stars) and 5 minor (P anomer, circles) resonances due to in the inequivalent carbon atoms on the cyclopentadienyl ring. 4.3.3 Solid State Structure of 2a The solid state structure of 2a shows the glucosamine moiety to be remote from the metal centre, as well as the relatively small size of the core (Figure 4.8, Table 4.1). The metal centre has an octahedral geometry, with the three carbonyl ligands facially coordinated at approximately 90\u00C2\u00B0 to one another. The bond lengths between the CO ligands and the metal complex were typical for Re(I)CO, as was the almost linear angle 120 formed by each R e C O . 1 , 2 ' 2 2 The planar Cp ligand occupied the other facial plane of the octahedron. The structure is consistent with the many 'piano stool' type complexes j 1 21 22 reported. ' ' As expected from the spectroscopic data, the glucosamine moiety is Figure 4.8. ORTEP view of 2a showing the atom labelling scheme (50% thermal ellipsoids). Table 4.1. Selected Bond Lengths (A) and Angles (deg) in 2a. Re-C(0) a v e r a ge 1.903 C(0)-Re-C(0) a v e ra ge 89.6 C-Oaverage 1.161 Rc-C*Oaverage 177.5 C(Cp)-C(Cp) a v e r a ge 1.420 C(8)-C(7)-N(l) 116.39(4) C(7)-0(6) 1.236(6) C(8)-C(7)-0(6) 120.8(4) C(7)-N(l) 1.341(5) Q(6)-C(7)-N(l) 122.9(4) covalently bonded to the Cp ring of the metal core via an amide bond. The carbohydrate is positioned above the 'piano stool' core, well removed from the metal centre, obviating any possibility of interaction. The structure of the glucosamine moiety is relatively 121 undistorted from its typical boat conformation. While two anomers in similar amounts were observed in the solution state NMR spectra, the (3 anomer crystalized preferentially, with only 7% of the a anomer present in the solid state. Since the anomer concentrations are different, other differences may also be present in the solution and solid state structures. To verify important structural aspects, such as the pendant nature of the carbohydrate moiety in solution, the solid state structure is not enough. Full characterization by non-ambiguous NMR spectral assignments was also carried out (vide supra). 4.3.4 Evaluation of Hexokinase Substrate and Inhibition Activity The interaction of 2a with hexokinase (HK) was examined to evaluate the retention of glucosamine biological activity. Two assays were used to determine how 2a interacts with HK. Compound 2a was first tested as a substrate for H K by monitoring the co-reaction of ATP to ADP, but no substrate activity was observed. In the second assay the inhibitory ability of 2a was determined by competition with glucose. The rate of glucose phosphorylation was measured alone or in the presence of 2a at different concentrations; 2a was determined to be a competitive inhibitor with K, = 330 \u00C2\u00B1 70 uM, suggesting that 2a strongly binds the active site of the enzyme. The competitive inhibition constant for FDG was determined using the same method and found to be Kj = 1060 \u00C2\u00B1 80 uM, suggesting that 2a is a better inhibitor than is FDG and binds to the enzyme active site either more strongly or more easily. Although FDG is a substrate for HK, it competes with glucose for the enzyme active site, and is therefore also an 122 inhibitor. The inhibition of H K by FDG could be measured because the phosphorylation of glucose was monitored by a secondary reaction for which FDG and its phosphoryiated product are not substrates. Although 2a does not act like FDG and is not a substrate for HK, the high affinity competitive inhibition by 2a suggests it retains some glucose-like biological activity. Compound 2a may be a substrate for certain glucose transporters (which may be less sensitive to substrate modifications). 4.3.5 y y m T c Radiolabelling As 2a may have biological properties of interest, development of the radiochemistry was undertaken. The indirect D L T route, which was used to prepare la and 2a on the macroscopic scale, is not ideal for radiopharmaceutical preparation, as it requires several steps, several purifications, and gives only moderate yields. In comparison, the SLT route requires fewer manipulations, only two steps with no purification in between, and may give higher radiochemical yields and purity in a shorter time. Although the SLT route was not successful using ferrocenyl carbohydrate complexes on the macroscopic scale, the reaction may be feasible on the tracer scale as milder temperatures have been reported for the 9 9 m T c S L T . 1 0 Attempts to prepare 9 9 m T c analogs, lb and 2b, from the corresponding ferrocenyl carbohydrates gave mixed results. For the preparation of 2b from N-(2-deoxy-2-amino-D-glucopyranose)ferrocenyl carboxamide, no product formation was observed. For the preparation of lb, the product formed in 45-80% yield, depending on time and temperature (Scheme 4.4). The highest yields were obtained at temperatures of 60-80\u00C2\u00B0C, and reaction times between 2.5 and 3.5 hours. The complex was identified and radiochemical yields determined by HPLC using 123 the non-radioactive Re complexes for comparison. At the tracer level concentrations of the radiolabeled complexes other methods of characterization are precluded. Scheme 4.4. Preparation of m T c complexes lb and 2b. As 2b could not be prepared directly, it was acquired by removing the acetyl protecting groups from lb (Scheme 4.4). In the HPLC trace of lb, numerous impurities were present with similar retention times to that of la, possibly complexes wherein the four acetyl protecting groups were only partially removed. After complete deacetylation, these impurities disappeared and the radiochemical yield of the 2b increased proportionally to 78%. Although in the co-injection HPLC trace the UV-visible peak attributed to 2a agrees well with the radiation peak attributed to 2b, the UV-visible trace of 2a has two peaks due to the two anomers present, while only one product peak is observed in the radiation trace of 2b (Figure 4.9) suggesting that only one anomer is present for the 9 9 m T c complex. Because the acetylated compound is anomerically pure, initially only one anomer is formed upon.deacetylating. If the equilibration of anomers via mutarotation is slow, only the initially formed anomer may be observed for the 9 9 m T c complex. 124 c D \u00C2\u00A3\u00E2\u0080\u00A2 re J 2 < 200 , 400 600 800 time / s 1000 1200 0) o c re JQ L _ o (fl n < 200 400 600 time / s 800 1000 1200 Figure 4.9. HPLC traces from radiation detector (top) and U V detector (bottom) for co-injection of cold standard (2a) and reaction mixture containing radiolabelled complex (2b) (both peaks labelled with a star). Similar chemistry was attempted with Re, a potential therapeutic radioisotope. [ 1 8 6Re(CO) 3(H 20) 3] + was prepared via literature methods,23'24 and then used in the SLT reaction as described above. Unfortunately, Re is more susceptible to re-oxidation than is Tc, and after the extended period of heating required for the SLT reaction no product had formed, while a significant amount of [ ReCu]\" was present. The more timely and 186 difficult method used on the macroscopic scale may be necessary to prepare the Re complex. 125 4.4 Conclusions This work demonstrates the utility of the CpM(CO)3 core in the labeling of highly functionalized and sensitive biomolecules, such as glucosamine. The glucosamine conjugates la and 2a were prepared and fully characterized. The solid state structure of 2a shows the small size of the CpM(CO)3 core, and pendant nature of the glucosamine moiety, remote from the metal. Preliminary in vitro evaluation of 2a shows that it retains some biological activity, as it is a high affinity competitive inhibitor of hexokinase, with a Kj value approximately three times smaller than that for FDG. It was necessary to use different methods for macroscopic (Re, indirect DLT) and tracer (Tc, SLT) syntheses; however, 9 9 m T c analogs lb and 2b were prepared in significant radiochemical yields. 9 9 mTc-labelled 2b is expected to be highly stable in vivo and may be useful as a tumour marker or glucose transporter imaging agent. 4.5 References 1. Bernard, J.; Ortner, K.; Spingler, B.; Pietzsch, H.-J.; Alberto, R., Inorg. Chem. 2003, 42, 1014-1022. 2. Wald, J.; Alberto, R.; Ortner, K.; Candreia, L. , Angew. Chem. Int. Ed. 2001, 40, 3062-3066. 126 3. Salmain, M . ; Gunn, M . ; Gorfti, A.; Top, S.; Jaouen, G., Bioconjugate Chem. 1993, 4, 425-433. 4. Spradau, T. S.; Katzenellenbogen, J. A., Organometallics 1998, 17, 2009-2017. 5. Wenzel, M . , J. Labelled Compds. Radiopharm. 1992, 31, 641-649. 6. Wenzel, M. ; Saidi, M . , J. Labelled Compds. Radiopharm. 1993, 33, 77-80. 7. Rattat, D.; Eraets, K.; Cleynhens, B.; Knight, H.; Fonge, H.; Verbruggen, A., Tetrahedron Lett. 2004, 45, 2531-2534. 8. Schibli, R.; Bella, R. L.; Alberto, R.; Garcia-Garayoa, E.; Ortner, K.; Abram, U.; Schubiger, P. A., Bioconjugate Chem. 2000, i i , 345-351. 9. Saidi, M . ; Seifert, S.; Kretzschmar, M. ; Bergmann, R.; Pietzsch, H.-J., J. Organomet. Chem. 2004, 689, 4739-4744. 10. Masi, S.; Top, S.; Boubekeur, L.; Jaouen, G.; Mundwiler, S.; Spingler, B.; Alberto, R., Eur. J. Inorg. Chem. 2004, 2013-2017. 11. Alberto, R.; Schibli, R.; Egli, A.; Schubiger, P. A., J. Am. Chem. Soc. 1998, 120, 7987-7988. 12. Adam, M . J.; Hall, L. D., Can. J. Chem. 1980, 55, 1188-1197. 13. Casas-Solvas, J. M. ; Vargas-Berenguel, A.; Capitan-Vallvey, L. F.; Santoyo-Gonzalez, F., Org. Lett. 2004, 6, 3687-3690. 14. Chahma, M . ; Lee, J. S.; Kraatz, H.-B., J. Organomet. Chem. 2002, 648, 81-86. 15. Spradau, T. W.; Edwards, W. B.; Anderson, C. J.; Welch, M . J.; Katzenellenbogen, J. A., Nucl. Med. Biol. 1999, 26, 1-7. 16. Spradau, T. W.; Katzenellenbogen, J. A., Bioconjugate Chem. 1998, 9, 765-772. 17. Alberto, R.; Egli, A.; Abram, U.; Hegetschweiler, K.; Gramlich, V.; Schubiger, P. A., J. Chem. Soc. Dalton Trans. 1994, 2815-2820. 127 18. Top, S.; Masi, S.; Jaouen, G., Eur. J. Inorg. Chem. 2002, 1848-1853. 19. Storr, T.; Fisher, C. L.; Mikata, Y.; Yano, S.; Adam, M . J.; Orvig, C., Dalton Trans. 2005, 654-655. 20. Yang, D. J.; Kim, C.-G.; Schechter, N. R.; Azhdarinia, A.; Yu, D.-F.; Oh, C.-S.; Bryant, J. L.; Won, J.-J.; Kim, E. E.; Podoloff, D. A., Radiology 2003, 226, 465-473. 21. Bolm, C ; Kesselgruber, M. ; Hermanns, N.; Hildebrand, J. P.; Raabe, G., Angew. Chem. Int. Ed. 2001, 40, 1488-1490. 22. Top, S.; Hafa, H. E.; Vessieres, A.; Quivy, J.; Vaissermann, J.; Hughes, D.; Mcglinchey, M . J.; Mornon, J.-P.; Thoreau, E.; Gerard, J., J. Am. Chem. Soc. 1995, 117, 8372-8380. 23. Bayly, S. R.; Fisher, C. L.; Storr, T.; Adam, M . J.; Orvig, C., Bioconjugate Chem. 2004,15, 923-926. 24. Park, S. H.; Seifert, S.; Pietzsch, H.-J., Bioconjugate Chem. 2006,17, 223-225. 128 Chapter 5 Gallium and Indium Complexes of 3-Hydroxy-4-pyridinone Pro-ligands Bearing Pendant Carbohydrates* 5.1 Introduction Several isotopes of Ga and In can be used in radiopharmaceuticals for diagnostic imaging, of particular current interest are 6 7 Ga, 6 8 Ga, and '\"in. 1 6 7 G a and l u I n are both cyclotron produced, but have sufficiently long half-lives, 78 h and 68 h respectively, so that they can be distributed significant distances from the production location.1 Both isotopes are y emitting radioisotopes which can be used in SPECT imaging, while 6 8 G a is a p + emitting radioisotope which can be used in PET imaging. Unlike the majority of p + emitting radioisotopes, which are cyclotron produced and thus of limited availability, 6 8 G a can be acquired from a generator with the parent radionuclide 6 8 Ge. 1 As well, because 6 8 Ge has a long half-life, 275 days, a 6 8 Ge/ 6 8 Ga generator system could be used for 1 or 2 years before requiring replacement. Radiopharmaceuticals incorporating 6 8 G a have the potential to greatly increase the accessibility of PET imaging. Several Ga and In radiopharmaceuticals are in current use (Figure 5.1).2 The most frequently used Ga imaging agent is 67Ga-citrate employed in tumour imaging.2'3 1 A version of this chapter has been published. Green, D. E., Ferreira, C. L., Stick, R. V., Patrick, B. O., Adam, M. J., Orvig, C. Carbohydrate-bearing 3-hydroxy-4-pyridinato complexes of gallium(III) and indium(III). Bioconjugate Chem. 2005, 16, 1597-1609. 129 Ga-citrate is unstable in vivo and its exact structure is not known. Once injected into the blood stream, 6 7 G a is quickly incorporated into transferrin where Ga can take the place of O , H O . N O H O H O 8-Hydroxyquinoline (Oxine) Citric Acid O H H O . o Diethylenetriaminepentaacetic acid (DTPA) Figure 5.1. Ligands used in Ga and In-radiopharmaceuticals. Fe due to similar size and charge.4 Uptake into tumours is then facilitated by the over expression of tranferrin receptors on cancer cells. 1 1 ' in is used to label biological substances in a slightly different manner. inIn-oxine (inIn(8-oxyquinolinate)3) is a lipophilic species that can cross cell membranes.2'5 Upon entering the cell, 1 1 ' in becomes protein bound and the freed oxine ligands diffuse back out of the cell. This property has been utilized for labelling white blood cells5 for imaging inflammation. For the labelling of peptides and monoclonal antibodies, a more stable chelate system is necessary. Diethylenetriaminepentaacetic acid (DTPA) has a larger stability constant on complexation with In (log KI\u00E2\u0080\u009E-DTPA = 29.0) compared to potential biological ligands such as transferrin (log Kin.tranferrin = 18.3).6 Hence, several In-DTPA conjugates of 130 monoclonal antibodies7\"9 and one of a small peptide10 have been approved as imaging agents for clinical use.2 The Orvig group has had much interest in the 3-oxy-4-pyridinone complexes of Ga and In over the past two decades.11-15 3-Hydroxy-4-pyridinone pro-ligands form neutral, soluble, octahedral tris-ligand complexes with Ga and In. Radiopharmaceuticals need to be stable, both kinetically and thermodynamically. Ga and In-pyridinone complexes are inert towards hydrolysis (a possible decomposition pathway in physiological systems) and have thermodynamic stability constants for the tris-ligand complexes significantly greater than the stability constants for Ga and In-transferrin complexes;16 transferrin is a major competing ligand for Ga and In in vivo. Biodistribution studies in mice with 6 7 G a complexes of 3-hydroxy-4-pyridinone pro-ligands showed faster blood clearance and lower liver uptake than 67Ga-citrate, suggesting that these compounds do not bind transferrin in the manner of 67Ga-citrate. The carbohydrate bearing 3-hydroxy-4-pyridinone pro-ligands first mentioned in Chapter 2 (Figure 5.2) were originally designed for chelating Ga and In to form potential radiopharmaceuticals targeting the biological pathways of carbohydrates. This chapter expands on previous work1 7 with the synthesis and characterization of the tris-ligand Ga and In complexes of the pyridinone pro-ligand with a pendant glucosamine (HL 5), the 6 7 G a radiolabelling of the series of carbohydrate bearing 3-hydroxy-4-pyridinones (HL 1' 3\" 5), and preliminary evaluation of these complexes as potential radiopharmaceuticals. 131 Feralex-G Figure 5.2. Carbohydrate bearing 3-hydroxy-4-pyridinone pro-ligands. 5.2 Experimental 5.2.1 Materials and Methods Most information relating to this section can be found in section 2.2.1. In(NC>3)3-3H20 and Ga(NC\u00C2\u00BB3)3-6H20 was purchased from Strem Fine Chemicals and Sigma Aldrich, respectively. 6 7GaCl3 was obtained as a 0.1 N HCI solution from MDS Nordion Inc. Ga(L U - 4 ) 3 and In(L''3\"4) were prepared, characterized and supplied by Dr. David E. Green.17 132 5.2.2 Synthesis of Ga(L 5) 3 and In(L5)3 Tris{(l-{N-[2-amino-2-deoxy-D-gIucopyranose]ethanamide}-2-methyI-3-oxy-4(lH)-pyridinato)}gallium(III) pentahydrate (Ga(L5)3-5H20). , OH v Gallium nitrate hexahydrate (11 mg, 0.03 mmol) / H O ^ - \u00C2\u00B0 V \ ' X 1 V \u00E2\u0080\u00A2 \u00E2\u0080\u009EOH 12 ^ \ was dissolved in 3 mL water and added to HL (32 H O - ^ \u00E2\u0080\u0094 \" ^ J ^ 1 * ^ Qs . 3 NH 13(^i if ' o - 7 \ / N ^ , 0 - o -^Ga mg, 0.093 mmol) dissolved in 3 mL water. The pH C H 3 / 3 of the resulting solution was adjusted to ~7 with 0.1 Ga(L 5) 3 M NaOH. The solutions was reduced on a rotary evaporator (30\u00C2\u00B0C) and purified on a Sephadex G10 column. Precipitation with acetone from a minimal amount of water gave a white solid, which was dried for 48 hours in a vacuum desiccator (30 mg, 91% yield). 'H NMR (CD 3 OD:D 2 0 1:1, 400 MHz) a anomer: 5 2.36 (s, 3H, C# 3), 3.43 (m, 1H, #3), 3.72-3.86 (m, 4H, H4-H6), 3.91 (m, 1H, #2), 4.96 (s, 2H, #8), 5.18 (d, 1H, HI, 3 J U = 2.81 Hz), 6.62 (d, 1H, #12, 3Ji 2,i 3 = 5.80 Hz), 7.55 (br s, 1H, #13); p anomer 8 2.38 (s, 3H, C#3), 3.40 (m, 2H, #3, #4), 3.54 (m, 1H, #5), 3.68 (m, 1H, H2), 3.8 (m, 2H, #6), 4.71 (d, overlapped with water peak, 1H, #1), 4.95 (s, 2H, #8), 6.62 (d, 2H, #12, 3J 1 2,i 3 = 5.80 Hz), 7.55 (br s, 1H, #13). 13C{lH} NMR (D 2 0:CD 3 OD 5:1, 100 MHz) a anomer: 5 13.04 (CH 3), 55.28 (CT), 58.13 (CS), 61.55 (C6), 71.09 (C3), 71.67 (CA), 72.52 (C5), 91.62 (CI), 109.2 (C12), 135.26 (C9), 137.77 (C13), 151.85 (C10), 168.14 (C7), 169.69 (CI 1); p anomer: 8 12.98 (CH 3), 58.44 (C8), 57.90 (C2), 61.69 (C6), 70.94 (C3), 74.54 (C5), 76.90 (C4), 96.0 (CI), 109.2 (C12), 135.26 (C9), 137.77 (C13), 151.85 (C10), 168.09 (C7), 170.00 (Cll) . MS (+ESI): 133 m/z (relative intensity) 757, 755 ([GaL 2]+, 100). Anal. Calcd. for C 4 2 H 5 7N 6 0 2 4 Ga-5H 2 0: C, 42.40; H, 5.68; N, 7.06. Found: C, 42.51; H, 5.73; N, 7.10. Tris{(l-{n-[2-amino-2-deoxy-D-glucopyranose]ethanamide}-2-methyl-3-oxy-4(lH)-pyridinato}indium(III) tetrahydrate (In(OG2GP)3-4H20). This reaction was carried out in a similar fashion as , O H . / H O ' ^ v i ^ T 0 \ \ f \u00C2\u00B0 r the analogous gallium complex except that \u00C2\u00A5 H 1 3 f i i ] / i n indium nitrate trihydrate was used (20 mg, 0.045 \ C H 3 / 3 mmol) with H L 5 (59 mg, 0.17 mmol). The product I n (L 5 ) 3 was a white solid (45 mg, 70% yield). lH NMR (CD 3 OD:D 2 0 1:1, 400 MHz) a anomer: 8 2.38 (s, 3H, C/fc), 3.46 (m, 1H#3), 3.72-3.84 (m, 4H, #4, #5, #6), 3.93 (m 1H, #2), 5.03 (s, 2H, #8), 5.24 (d, 1H, HI, 3Jh2 = 3.27 Hz), 6.73 (d, 1H, #12), 7.60 (br s, 1H, #13); p anomer: 5 2.38 (s, 3H, C# 3), 3.43 (m, 2H, #3, #4), 3.55 (m, 1H, #5), 3.63 (m, 1H, #2), 3.8 (m, 2H, #6), 4.72 (d, overlapping with water peak, 1H, #1), 5.03 (s, 2H, #8), 6.73 (d, 2H, #12, 3 J 1 2 ; i 3 = 5.80 Hz), 7.60 (br s, 1H, #13). l3C{lK} NMR (D 2 0:CD 3 OD 5:1, 100 MHz) a anomer: 8 13.23 (CH 3), 55.29 (CI), 58.20 (CS), 61.55 (C6), 71.09 (C3), 71.69 (CA), 72.52 (CS), 91.64 (CI), 110.86 (C12), 137 (C9), 137 (C13), 151.82 (C10), 168.34 (CI), 169.61 (CI 1); p anomer 8 13.18 (CH 3), 57.91 (C2), 58.25 (C8), 61.69 (C6), 70.93 (C3), 74.57 (CS), 76.91 (C4), 95.64 (CI), 110.86 (C12), 137 (C9), 137 (C13), 151.82 (C10), 168.34 (CI), 169.90 (Cll) . MS (+ESI): m/z (relative intensity) 801 ([InL2]+, 100). Anal. Calcd. for C 4 2 H57N 6 0 2 4ln4H 2 0: C, 41.32; H, 5.45; N, 6.91. Found: C, 41.46; H, 5.38; N, 7.32. 134 5.2.3 Hexokinase Inhibition Assay For a description of the hexokinase inhibition assays refer to section 2.2.7. 5.2.4 6 7 Ga Radiolabelling The pro-ligands were dissolved in water (HL1'3\"5). 6 7 G a C l 3 (300 uCi) in 0.1 N HC1 (300 uL) was added to each pro-ligand solution (500 uL) to give a final pro-ligand \u00E2\u0080\u00A2y concentration of 10\" M . To neutralize the acid, 0.1 N NaOH was added, until the pH was between 6 and 8. The labelled complexes were identified and radiochemical yield determined by comparison of the HPLC trace with that of the respective non-radioactive complex. Full description of HPLC system and detectors used is given in section 2.2.1. 5.2.5 Cysteine and Histidine Stability Challenges A solution of each Ga-complex (100 uL) was added to a 900 uL solution of either 10\"2 M histidine or 10\"2 M cysteine. The solutions were incubated at 37 \u00C2\u00B0C for 1 h and analyzed by HPLC. 135 5.3 Results and Discussion 5.3.1 Synthesis and Characterization of Ga (L 5 )3 and In(L 5)3 The neutral tris-ligand Ga(III) and In(III) complexes of the pro-ligand H L 5 were prepared in 91% and 70% yield, respectively (Scheme 5.1). Ga or In nitrate solutions were added to a solution containing three equivalents of H L 5 . The initially acidic solutions were neutralized with N a O H to facilitate complexation and the complexes were purified by size exclusion column chromatography. OH Scheme 5.1. Synthesis of M ( L 5 ) 3 , M = Ga, In. The complexes were fully characterized by elemental analysis, mass spectrometry, and N M R spectroscopy. Elemental analyses were consistent with hydrates of the expected 3:1 ligand to metal complex. These waters of hydration are consistent with other reported Ga and In-pyridinone complexes 1 4 ' 1 5 ' 1 8 and for complexes containing carbohydrates. The most prominent peak in the mass spectra was attributed to the complex with one ligand removed ( [ M L 2 ] + ) , and displayed the expected isotopic distributions ( 6 9 G a 60.1%, 7 1 G a 39.9%). 136 Extensive NMR experiments were done to confirm the binding of both oxygen atoms of the pyridinones and the pendant nature of the carbohydrate (Figures 5.3 and 5.4). In general the N M R spectra of the complexes had broader, less resolved resonances than did the NMR spectra of the non-complexed ligand, due to the numerous possible isomers. The ligand, H L 5 , is a mixture of two anomers which have different resonances. The complexes can have any combination of these two anomers, yielding a mixture of species with slightly different chemical environments. The NMR spectra resonances of each species are very similar and cannot be resolved from one another, but instead produce the broadened resonances observed. Most notably in the ! H N M R spectra, the expected doublet of HI 3 is only observed as a broad singlet and the two singlets, one for each anomer, cannot be resolved from one another for either the methyl or the H8 hydrogen atom resonances. 'H NMR spectra indicated that the pyridinone ring hydrogen atoms had the most prominent changes in chemical shift upon metal complexation. The largest coordination induced shift (CIS) was observed for HI2, where metal complexation resulted in downfield shifts of 0.13 and 0.24 ppm for the Ga and In complexes, respectively. These are close to the 0.19-0.32 ppm downfield shifts observed for other (crystallographically characterized) Ga and In-pyridinone complexes.18 All of the pyridinone ring carbon atoms shifted significantly, indicating the pyridinone moiety is complexing Ga and In. The largest CIS occurs for the CIO atoms, which are the carbon atoms of the binding alkoxy functionality, where a downfield shift of ~6 ppm occurs upon complex formation. Since Ga(III) and In(III) have been shown to bind to the alkoxy groups from citric acid ' or D-gluconic acid, it was important to 137 JC . 5 \u00C2\u00A9 . O S . S S . O - 4 . 5 4 . 0 3 . 5 3 - 0 2 . 5 2 . 0 I\u00E2\u0080\u0094\u00E2\u0080\u00A2\u00E2\u0080\u0094\u00E2\u0080\u00A2\u00E2\u0080\u0094\u00E2\u0080\u00A2\u00E2\u0080\u0094\u00E2\u0080\u00A2\u00E2\u0080\u0094i\u00E2\u0080\u0094 7 . 5 7 . 0 e . 5 e . O 5 - 5 5 . 0 - 4 . 5 4,0 3 . 5 3 . 0 2 . 5 2 . 0 7 . 5 7 . 0 e . 5 \u00C2\u00A9 . O 5 . 5 5 . 0 4 . 5 - 4 . 0 3 . 5 3 . 0 2 . 5 2 . 0 p p m Figure 5.3. *H NMR spectra (400 MHz, CD 3 OD:D 2 0 1:1) of (a) In(L 5) 3, (b) Ga(L 5) 3 and (c) H L 138 a Li 1 4 0 j IA1 11\u00E2\u0080\u0094 Figure 5.4. I 3 C NMR spectra (100 MHz, CD 3 OD:D 2 0 1:5)of (a) In(L s) 3, (b) Ga(L 5) 3 and (c) H L determine if the carbohydrate hydroxyl groups had binding interactions with the metal centres. This type of interaction could be problematic for retaining biological recognition of the carbohydrates. The overlap of the glucose hydrogen atom resonances 139 in the 'H NMR spectra, further complicated by the presence of the two anomers, made it difficult to ascertain whether there were any interactions between the hydroxyl groups of the glucose and the metal centres. Hence, 1 3 C NMR spectra were imperative in determining whether any glucose-metal direct interactions were present. There are very small chemical shift differences for the glucose carbon atoms of the free pro-ligand and the complexes (all less than 0.1 ppm shifts) indicating clearly that the hydroxyl groups of glucose do not coordinate to the Ga or In metal centres; more dramatic shifts would be expected if the glucose hydroxyl groups coordinated. 5.3.2 Hexokinase Inhibition Studies Each of the Ga and In complexes, M(L 1 , 3 \" 5 )3 (M = Ga or In), were tested for inhibition of glucose phosphorylation by hexokinase, the first enzyme in glucose metabolism. None of the complexes inhibited this enzymatic process, suggesting that they do not interact with the active site of the enzyme. 5.3.3 6 7 Ga Radiolabelling Each of the pro-ligands (HL W \" S ) were labelled with 6 7 G a in high radiochemical yields (Scheme 5.2). 6 7 G a in a 0.1 N HC1 solution was added to a 10\"2 M solution of each pro-ligand, followed by neutralization with 0.1 N NaOH. Using lower pro-ligand concentrations gave significantly lower yields, even when heated or allowed to react for longer periods of time. These findings are consistent with other reported 6 7 G a labelling 140 S c h e m e 5.2. Synthesis of 6 7 G a ( L 1 , 3 \" 5 ) 3 . 141 of 3-hydroxy-4-pyrdinone derivatives. HPLC identification of the radiolabeled product in each case was done by co-injecting the radiolabelling reaction with the respective cold complex and comparing the radiation detector trace to the U V detector trace. Radiochemical yields and retention times of the standards and radiolabeled products are given in Table 5.1. Table 5.1. HPLC retention times and radiochemical yields for 6 7 Ga(L 1 , 3 \" 5 ) 3 . Pro- U V (254 nm) detected RT Radiation detected ligand of standard / mina RT / mina % Radiochemical Yield H L 1 10.1 10.9 99\u00C2\u00B11 H L 3 11 11.5 99\u00C2\u00B11 H L 4 6.7 6.8 94\u00C2\u00B12 H L 5 4.0 4.0 99\u00C2\u00B11 a HPLC run on Synergi 4pm C-l8 Hydro RP analytical column, 1 mL/min, gradient 100% A to 100% B, solvent A: 0.1% w/w trifluoroacetic acid in water, solvent B: methanol. 5.3.4 Cysteine and Histidine Stability Challenges To assess the potential stability of the 67Ga-pyridinone complexes ( 6 7 Ga(L 1 , 3 \" 5 )3) in vivo, challenge experiments were done with excess cysteine and histidine, amino acids 142 found in the body with the ability to chelate metals. The complexes were incubated in the presence of a 10 fold excess (compared to the pro-ligand) of either cysteine or histidine. After 1 h the challenge experiments were analyzed by HPLC to determine the 67 amount of Ga-pyridinone complex still intact. Irrespective of the complex or amino acid used, a number of large broad peaks between 3 and 9 min were observed in the radiation trace. For complexes 6 7 Ga(L 1 ) 3 and 6 7 Ga(L 3 ) 3 these peaks do not correspond to the 6 7 G a labelled pyridinone complexes, and indicate that the complexes are not stable under these conditions. The peaks with retention times consistent with the complexes 67Ga(V)3 and 6 7 Ga(L 3 ) 3 were only 10-15% of total radiation, suggesting only a small amount (<15%) of the complexes remain intact under these conditions. The complexes 6 7 Ga(L 4 ) 3 and 6 7 Ga(L 5 ) 3 have retention times that overlap with the broad range of peaks between 3 and 9 min and thus the amount of complex remaining intact could not be determined. It is expected that all of the complexes would have similar stability, thus it is unlikely that any significant amounts of 6 7 Ga(L 4 ) 3 and 6 7 Ga(L 5 ) 3 remain, considering the lack of stability of 6 7 Ga(L ! ) 3 and 6 7 Ga(L 3 ) 3 . Incubation of 6 7 G a C l 3 with cysteine or histidine gave a similar radiation trace of broad peaks between 3 and 9 min, suggesting that numerous species are formed between Ga and these amino acids, and that the Ga-pyridinone complexes form these same species, or similar species, in the presence of excess cysteine or histidine. It was not determined if all three of the pyridinone ligands 67 were substituted in the Ga-complexes, or if mixed ligand species, containing both pyridinones and amino acids, were formed. 143 5.4 Conclusions In this chapter the Ga(III) and In(III) tris ligand complexes of the pro-ligand HL 5 were prepared and characterized. Characterization confirmed the bidentate binding of three pyridinone ligands to form a neutral complex. NMR studies were used to evince the pendant nature of the carbohydrate moiety. The complexes were tested as inhibitors of hexokinase to assess the retention of biological properties of the carbohydrate moieties. Although none of the complexes showed inhibitory activity, they may still be substrates for glucose transporters or other biological pathways. Labelling of the pro-1 3 5 67 ligands HL '\" with Ga produced the expected radiolabeled complexes in high radiochemical yield (>94%). Stability challenges showed that the complexes were not stable in the presence of histidine or cysteine, and are therefore unlikely to remain intact in vivo. Further studies are needed to determine i f 1 1 ' in can be used to label these ligands in high radiochemical yields, and if these complexes are more stable than their 6 7 G a analogues. 5.5 References 1. Anderson, C. J.; Welch, M . J., Chem. Rev. 1999, 99, 2219-2234. 2. Weiner, R. E.; Thakur, M . L., Chemistry of gallium and indium radiopharmaceuticals. In Handbook of Radiopharmaceuticals Radiochemistry and Applications., Welch, M . ; Redvanly, C. S., Eds. John Wiley and Sons Ltd.: West Sussex, England, 2003. 144 3. Alazraki, N. P., Gallium-67 imaging in infection. In Principles and Practice of Nuclear Medicine, Early, P. J.; Sodee, D. B., Eds. Mosby: St. Louis, 1995; pp 702-713. 4. Weiner, R. E. , Nucl. Med. Biol. 1996, 23, 745-751. 5. Thakur, M . L.; Segal, A. W.; Louis, L.; Welch, M . J.; Hopkins, J.; Peters, T. J., J. Nucl. Med. 1977,18, 1022-1027. 6. Harris, R. L.; Chen, Y. C ; Wein, K., Inorg. Chem. 1995, 33, 4991-4998. 7. Quintana, J. C ; Blend, M . J., Clin. Nucl. Med. 2000, 25, 33-40. 8. Pinkas, L.; Robins, P. D.; Forstrom, L. A.; Mahoney, D. W.; Mullan, B. P., Nucl. Med. Comm. 1999, 20, 689-696. 9. Carrio, I.; Lopez-Pousa, A.; Estorch, M. ; Duncker, D.; Berna, L.; Torres, G.; de Andres, L., J. Nucl. Med. 1993, 34, 1503-1507. 10. Krenning, E. P.; Kwekkeboom, D. J.; Bakker, W. H.; Breeman, W. A. P.; Kooij, P. P. M. ; Oei, H. Y.; van Hagen, M. ; Poterma, P. T. E.; de Jong, F. H.; Reubi, J. C ; Visser, T. J.; Reijs, A. E. M . ; Hofland, L. J.; Koper, J. W.; Lamberts, S. W. J., Eur. J. Nucl. Med. 1993, 20, 716-731. 11. Clevette, D. J.; Lyster, D. M. ; Nelson, W. O.; Rihela, T.; Webb, G. A.; Orvig, C , Inorg. Chem. 1990, 29, 661-612. 12. Nelson, W. O.; Karpishin, T. B.; Rettig, S. J.; Orvig, C , Inorg. Chem. 1988, 27, 1045-1051. 13. Nelson, W. O.; Karpishin, T. B.; Rettig, S. J.; Orvig, C , Can. J. Chem. 1988, 66, 123-131. 14. Nelson, W. O.; Rettig, S. J.; Orvig, C , Inorg. Chem. 1989, 28, 3153-3157. 15. Matsuba, C. A.; Nelson, W. O.; Rettig, S. J.; Orvig, C , Inorg. Chem. 1988, 27, 3935-3939. 145 16. Clarke, E. T.; Martell, A. E. , Inorg. Chim. Acta 1992,191, 57-63. 17. Green, D. E. Carbohydrate-Bearing Ligands for Biologically Active Metal Ions. Ph.D. Thesis. University of British Columbia, Vancouver, BC, 2004. 18. Zhang, Z.; Rettig, S. J.; Orvig, C , Inorg. Chem. 1991, 30, 509-515. 19. Chang, C. H. F.; Pitner, T. P.; Lenkinski, R. E.; Glickson, J. D., Inorg. Chem. 1977, 99, 5858-5863. 20. Hawkes, G. E.; O'Brien, P.; Salacinski, H.; Motevalli, M . ; Abrahams, I., Eur. J. Inorg. Chem. 2001,1005-1011. 21. Escandar, G. M . ; Olivieri, A. C ; Gonzalez-Sierra, M . ; Frutos, A. A.; Sala, L. F., Dalton Trans. 1995, 1393-1393. 146 Chapter 6 Copper Complexes of 3-Hydroxy-4-pyridinone and Tetrahydrosalen Pro-ligands with Pendant Carbohydrates* 6.1 Introduction Cu has a range of radionuclides that are of interest for diagnostic imaging and/or radiotherapy (Table 6.1). Most Cu radioisotopes relevant to nuclear medicine are Table 6.1. Cu radioisotopes of interest in nuclear medicine. Radioisotope Half-life Major emission Production Method(s) 6 0 C u 23 min P+ Accelerator 6 , C u 3.3 h P +,Y Accelerator 6 2 C u 9.8 min P+ Generator 6 4 C u 12.8 h P+, P\", Y Reactor, Accelerator 6 7 C u 62 h P\",Y Accelerator produced by accelerated particle bombardment,1\"3 with the exception of 6 2 C u which can be acquired from a generator.4 The short half-lives of 6 0 \" 6 2 Cu limit their utility to perfusion imaging, while the longer half-lives of 6 4 C u and 6 7 C u are more applicable to 1 A version of this chapter will be submitted for publication. Ferreira, C. L., T. Storr, D. E. Green, J. Steele, S. Lapi, D. Yapp, A. Celler, M. J. Adam, C. Orvig. Copper Complexes of 3-Hydroxy-4-pyridinone and Tetrahydrosalen Pro-ligands with Pendant Carbohydrates. 147 the slower kinetics of biomolecule targeting. Availability of the two longer lived radioisotopes, 6 4 C u and 6 7 C u , has been limited. Two factors have increased 6 4 C u availability in recent years: the production of 6 4 C u using a small biomedical cyclotron2 and the isolation of 6 4 C u as a by-product of 6 7 G a production,5 which is the method used for commercially available 6 4 C u from MDS Nordion Inc. 6 7 C u availability remains limited, as it requires a high energy accelerator for production.3 Both 6 4 C u and 6 7 C u have P\" emission with radiotherapeutic potential, and 6 4 C u has a P + emission that can be utilized in PET imaging. Theoretically, 6 4 C u can be used for both PET imaging and radiotherapy, albeit using different amounts of activity. The ability to image the biodistribution of a 64Cu-radiopharmacuetical before administrating a therapeutic dose is advantageous for obtaining information about the selectivity of the radiopharmaceutical and the potential damage to non-targeted organs. Most therapeutic radiopharmaceuticals require a radioisotope of a different element to be used in imaging which may not be consistent with the biological properties of the actual radiotherapeutic. Presently, no Cu based radiopharmaceuticals have been clinically approved, but the range of isotopes and the diverse coordination chemistry of Cu make it a good candidate for radiopharmaceutical development. The short half-lives of the positron emitting 6 0 C u and 6 2 C u radioisotopes limit their use to perfusion imaging, while the longer half-lives of 6 4 C u and 6 7 C u are better suited for the labelling of antibodies and peptides.6 Cu has two common oxidation states (I, II); of these, Cu(II) is the form in which radioisotopes are typically acquired. Complexes of Cu(II), a d 9 metal centre, have large Jahn-Teller distortions, resulting in a preferred square planar geometry rather than octahedral, although octahedral complexes with significant axial elongation or equatorial 148 compression are known. Current research has focused on two types of Cu complexes, 8 15 1 2\"^ bis(thiosemicarbazone) \" and biomolecule tagged tetraazamacrocycle complexes. The former can be considered Cu radioisotope essential, and are being investigated as blood flow9 and hypoxic tissue imaging agents, and as radiotherapeutics.10\"12'15 The neutral lipophilic compounds cross the cell membrane, after which event Cu(II) is reduced to Cu(I), leading to dissociation of Cu from the ligand and intracellular trapping \u00E2\u0080\u00A2 \u00E2\u0080\u00A2 8 12 13 of the radioisotope. ' ' The high first pass tissue extraction of these agents makes them promising candidates for tissue blood flow, potential hypoxic tumour imaging and therapeutic agents due to the higher expression of reductive enzymes in hypoxic tumours.24 Figure 6.1 shows the structure of two compounds of this class, Cu-PTSM and Cu-ATSM investigated for imaging blood flow, and hypoxic tissue, respectively. R= H Cu-PTSM R = C H 3 Cu-ATSM Figure 6 .1 . Structure of Cu-bis(thiosemicarbazone) complexes. The labelling of tumour targeting biomolecules, such as monoclonal antibodies and receptor directed peptides, requires a more stable chelate system. To this end, -tetraazamacrocyclic ligands have been examined as stable bifunctional chelates for Cu radiolabelling (Figure 6.2).25\"29 Macrocyclic chelators, such as 1,4,7,10-tetraazacyclododecane-N,N',N\",N\"'- tetraacetic acid (DOTA) and 1,4,8,11-149 tetraazacyclotetradecane-N,N',N\",N'\"-tetraacetic acid (TETA), have been shown to have higher in vivo stability than do acyclic multidentate ligands such diethylenetriaminepentaacetic acid (DTPA), despite similar copper complex Figure 6.2. Macrocyclic bifunctional chelators for Cu labelling. thermodynamic stability constants (log K C U - D O T A = 22.7, log K C U - D T P A = 21.4 ' ). The increased in vivo stability of the macrocyclic Cu complexes is an effect of the kinetic inertness of the complex, rather than just thermodynamic stability. Kinetics is an important factor in vivo due to the much larger concentration of competing endogenous ligands compared to the radiolabelled complex. Further increased in vivo stability has been observed in the copper complexes of cross-bridged analogues of D O T A and T E T A (Figure 6.2). ' Even relatively kinetically inert Cu-macrocyclic complexes have been shown to undergo loss of Cu with its incorporation into superoxide dismutase (SOD), but at much slower rates. These macrocyclic complexes have been tagged with DOTA T E T A H 3 COOC \ / Cross-bridged DOTA Cross-bridged T E T A 150 biomolecules, such as peptides \" and monoclonal antibodies, ' and evaluated as targeted imaging and therapy agents. Only a handful of Cu complexes containing a carbohydrate derivative have been reported. The majority of these complexes examine direct binding interactions between the carbohydrate and the Cu atom to better understand their biological implications.32\"38 Cu complexes of ligands with pendant carbohydrates have also been investigated,39\"42 such as dipicolylamine, Schiff base and bis(thiosemicarbazone) Cu complexes with pendant sugar derivatives. Despite the success of FDG as an imaging agent and the potential of Cu radioisotopes for both imaging and therapy, no Cu radiolabelled carbohydrate conjugates have been reported. Both 3-hydroxy-4-pyridinone and tetrahydrosalen based ligands have been previously studied with Cu. 4 3 \" 4 8 Thermodynamic stability studies of representative Cu-pyridinone complexes have been reported, with log K Q U L 2 = 2 1 . 7 for bis(l,2-dimethyl-3-oxy-4-pyridinato)copper(II)43 (the major species expected to be present under physiological conditions). Similarly, the thermodynamic stability constant for the simplest copper tetrahydrosalen complex (N,N'-bis(o-oxybenzyl)ethylenediaminocopper(II)) has been reported (log K C U L = 2 0 . 5 ) . 4 4 Derivatives of these binding moieties with pendant carbohydrates, such as the 3-hydroxy-4-pyridinone ligands used in Chapter 2 , have been previously synthesized. As well glycosylated tetrahydrosalen ligands have been complexed to copper and studied previously in the Orvig group 4 9 The availability of these carbohydrate pendant ligand derivatives coupled with the binding moieties reported thermodynamic stability upon binding to Cu make these systems intriguing for the development of carbohydrate based Cu radiopharmaceuticals. 151 In this chapter, copper complexes o f the 3-hydroxy-4-pyridinone ( H L 1 ' 3 \" 5 ) and tetrahydrosalen pro- l igands ( H 2 L 1 0 \" 1 1 ) (Figure 6.3) bearing pendant carbohydrate derivatives are investigated as potential i m a g i n g and radiotherapeutic agents. B o t h pro-l igand sets y i e l d neutral , t h e r m o d y n a m i c a l l y stable Cu(II) c o m p l e x e s . 4 9 T h e complexes were prepared and characterized o n the macroscopic scale w i t h non-radioact ive C u , 4 9 and prepared o n the tracer scale us ing 6 4 C u . E v a l u a t i o n o f the complexes, as potential radiopharmaceuticals was carried out both in vitro and in vivo. R = t-butyl H 2 L U Figure 6 . 3 . 3-Hydroxy-4-pyridinone and tetrahydrosalen pro- l igands bear ing pendant carbohydrates. 152 6.2 Experimental 6.2.1 Materials and Methods Most information related to this section is contained in section 2.2.1. 6 4 C u C l 2 was provided by MDS Nordion Inc. as a solution in 0.1 N HCI, and contained -10% 6 7 C u at the time of activity measurement. H 2 L 1 0 \" 1 1 and CuL 1 0 \" 1 1 were prepared, characterized and supplied by Dr. Tim Storr.49 6.2.2 Synthesis of Cu(II) Complexes General preparation of bis(3-oxy-4-pyridinato)copper(II) complexes. Stoichiometric amounts of copper(II) acetate monohydrate (one equivalent) and the respective pyridinone pro-ligand H L 1 ' 3 - 5 (two equivalents) were dissolved separately in 1:1 mixtures of methanol and water. The copper solution was added to the pro-ligand solution, resulting in a clear green solution with an initial pH of ~3. The pH was adjusted to ~7 with 0.1 M NaOH resulting in the formation of a green precipitate. The flask was cooled overnight, and the green precipitate filtered out and washed with cold water, before drying over P2O5 in a vacuum dessicator. 153 Bis(l-{N-[p-benzyl P-D-glucopyranoside]ethanamide}-2-methyl-3-oxy-4(lH)-pyridinato)copper(II) dihydrate (Cu(L1)2 -2H20). / 9 H \ Copper(II) acetate monohydrate (12 mg, X^'t^^4 \" 0 - 0 6 m m o l ) a n d H L l ( 5 3 m g ' 0 - 1 2 OH \ H < L H J2 mmol) yielded a green solid (52 mg, 93%). IR (KBr disc, cm\"1): 3337 (br) (vOH), 1671 (s) (v(CONH)), 1605 (s) (v(CO)), 1548 (s), 1511 (s), (aryl), 1359 (s), 1301 (m), 1229 (m) (6(OH)), 829 (m) (para-substituted benzene). UV-visible spectrum (H2O): A.max/nm (emax/ L M\" 1 cm\"1) = 305 (2.2 x 104). MS (ESI+): m/z = 956 ([M+Na]+, 100). Anal. Calcd. for C4oH46CuN4Oi8-2H20: C, 49.51; H, 5.19; N, 5.77. Found: C, 49.81; H, 5.36; N, 5.95. Bis(l-{p-benzyl P-D-glucopyranoside}-2-methyl-3-oxy-4(lH)-pyridinato)copper(II) pentahydrate (Cu(L3)2 -5H20). ^ H . Copper(II) acetate monohydrate (21 mg, 0.10 9 \" 3 ^ mmol) and H L 3 (80 mg, 0.21 mmol) yielded a HO' , 0 H \u00C2\u00B0 OH sTCu green solid (66 mg, 80%). IR (KBr disc, cm\" \ \u00C2\u00B0 / 2 '): 3300 (w, br) (vOH), 1606 (s), (v(CO)), 1542 (s), 1503 (s), 1469 (s) (aryl), 1350 (s), \u00E2\u0080\u00A2 1293 (m), 1233 (m) (5(OH)), 1060 (s), 1023 (s) (v(COH)), 820 (s), (para-substituted benzene). UV-visible spectrum (H 20): A,m a x/nm (sm a x/ L M\" 1 cm\"1) = 307 (1.1 x 104). MS (ESI+): m/z = 843 ([M+Na]+, 100). Anal. Calcd. for C36H 4 oCuN 2 0 1 6 -5H 2 0: C, 47.50; H, 5.54; N, 3.08. Found: C, 47.37; H, 5.32; N, 3.13. 154 Bis(l-{N-[methyl 6-amino-6-deoxy-a-D-glucopyranoside] ethanamide}-2-methyl-3-oxy-4(lH)-pyridinato)copper(II) dihydrate (Cu(L4)2 -2H20). 0 r^^Y^^Q Copper(II) acetate monohydrate (12 mg, 0.06 mmol) and \" T ^ ^ ^ L \u00C2\u00B0 | H L 4 (43 mg, 0.12 mmol) yielded a light green solid (42 mg, 90%). IR (KBr disc, cm\"1): 3330 (br) (vOH), 1672 2 (s) (v(CONH)), 1605 (s) (v(CO)), 1548 (m), 1512 (s), 1487 (s) (aryl), 1361 (m), 1288 (m) (8(OH)), 1049 (br) (v(COH)). UV-visible spectrum (H 20): XmJnm (emJ L M\"1 cm\"1) 305 (2.0 x 104). MS (ESI+): m/z = 800 ([M+Na]+, 100). MS (ESI-): m/z = 776 ([M-H]\ 100). Anal. Calcd. for C 3 0 H4 2 CuN4Oi6-5H 2 O: C, 41.50; H, 6.04; N, 6.45. Found: C, 41.57; H, 6.14; N, 6.67. Bis(l-{N-[2-amino-2-deoxy-D-glucopyranose]ethanamide}-2-methyl-3-oxy-4(lH)-pyridinato)copper(II) trishydrate (Cu(L5)2 '3H20). . Copper(II) acetate monohydrate (12 mg, 0.06 OH \ H O - ^ ^ - ' O ' \ mmol) and H L 5 (41 mg, 0.12 mmol) yielded a light H O A - - - - - \" ^ A ^ O H ^ \ / C u green solid (38 mg, 85%). IR (KBr disc, cm\"1): NH > ^ CH 3 3362 (br) (vOH), 1677 (s) (v(CONH)), 1605 (s) 2 (v(CO)), 1549 (m), 1509 (m), 1486 (m) (aryl), 1360 (m), 1287 (m) (8(OH)). UV-visible spectrum (H 20): XmJnm (smJ L M\" 1 cm\"1) = 3 1 8 (8.3 x 103), 263 (9.9 x 103). MS (ESI+): m/z = 772 ([M+Na]+, 100). Anal. Calcd. for C 2 8 H 3 8CuN 4 Oi6-3H 2 0: C, 41.82; H, 5.51 ; N , 6.97. Found: C, 42.12; H, 5.52; N, 7.04. 155 6.2.3 Hexokinase Inhibition Assay Each copper complex was tested as a potential inhibitor of glucose phosphorylation by hexokinase as described in section 2.2.7. 6.2.4 6 4 Cu Radiolabelling The pro-ligands were dissolved in water (HL 1' 4' 5) or ethanol (HL 3 , H 2 L 1 0 \" 1 1 ) . 6 4 C u C l 2 (500 uCi) in 0.1 N HCI (500 uL) was added to each ligand solution (500 uL) to give a final ligand concentration of 10\"2 M . To neutralize the acid, 0.1 N NaOH was added, until the pH was between 6 and 8. The labelled complexes were identified and radiochemical yield determined by comparison of the HPLC trace with that of the respective non-radioactive complex. Full description of HPLC system and detectors used is given in section 2.2.1. 6.2.5 Cysteine and Histidine Stability Challenges A solution of each 64Cu-complex (100 uL) was added to a 900 uL solution of 2 2 either 10\" M histidine or 10\" M cysteine. The solutions were incubated at 37 \u00C2\u00B0C for 1 h and analyzed by HPLC. 156 6.2.6 PET Imaging Studies PET imaging studies were carried out for each of the Cu complexes in DD/S mice bearing the Shionogi or SC-115 tumour line using a Concorde Focus microPET camera. The mice were injected in the tail vein with -20 MBq of the respective 6 4 C u complex as a saline solution or a 9% ethanol saline solution. For the complexes 6 4 Cu(L x ' 3 ' 5 ) 2 and 6 4 C u ( L 1 0 ' n ) a static scan was taken 3 h after injection. For 6 4 C u ( L 4 ) 2 and a free 6 4 C u 2 + control in saline, dynamic scans were carried' out over 3 h. 6.3 Results and Discussion 6.3.1 Synthesis and Characterization of Bis(pyridinato)copper(II) Complexes Bis-ligand copper complexes of the 3-hydroxy-4-pyridinone pro-ligands were prepared in high yield, >80%, and fully characterized by mass spectrometry, elemental analysis, UV-visible and infrared spectroscopy. Complex formation was achieved by dissolving stoichiometric amounts of the ligand (two equivalents) and copper(II) acetate in a methanolic solution, then neutralizing the pH with aqueous NaOH (Scheme 6.1). A green precipitate readily formed which was purified by washing with cold water and/or methanol to remove any residual ligand or excess copper(II) acetate. Elemental analyses confirmed the composition of the bulk samples to be hydrated bis-ligand Cu complexes. Each complex had a stoichiometric number of water molecules (2-5) associated with the vacuum dried sample; these were not removed even after extended drying. The water molecules may be associated with the carbohydrate derivatives, or may reside in the 157 a) Cu(OAc) 2, 0.1 N NaOH b) 6 4 C u C l 2 , 0.1 N NaOH : \u00E2\u0080\u00A2 H L 1,3-5 Cu(L' ' 3 \" 5 ) 2 weakly binding axial positions of the Jahn-Teller distorted octahedral geometry, typically observed as four coordinate square planar. 7 The major peak observed in the mass spectrum of each complex was assigned to the [M+Na] + ion with 1 0 0 % relative intensity. 158 Infrared spectra were similar to the free ligand spectra, except for the shift of the resonance assigned to the pyridinone ring ketone carbonyl stretch (Table 6.2). The shift to a lower wavenumber is consistent with back bonding from the Cu d-orbitals to the TI C=0 anti-bonding orbitals. A shift was also observed upon coordination in the UV-visible spectral peak attributed to the ligand chromophore (Table 6.2). Cu(II) compounds have a d 9 electron configuration which has a characteristic d-d electron transition in the visible region. The symmetry forbidden d-d transitions were not observed for all compounds due to limited solubility of the complexes and the weakness of the transitions, but the characteristic green colour (650-700 nm) of the compounds confirms a d 9 Cu centre.7 Table 6.2. Comparison of pyridinone ring IR spectra vCO and UV-visible spectra Xmaii and for free pro-ligands and Cu-complexes. Pro- vCO free vCO Cu- A vCO A, m a x free A. m a x Cu- A A. m a x / ligand ligand (HL) / complex / cm\"1 ligand / complex / nm cm\"1 (CuL 2) / cm\"1 nm nm H L 1 1631 1605 -26 280 305 +25 H L 3 1631 1606 -25 281 307 +26 H L 4 1632 1605 -27 282 305 +23 H L 5 1638 1605 -33 286 304 +22 159 6.3.2 Hexokinase Inhibition Studies The rationale for studying the interaction of complexes with the glucose metabolism enzyme hexokinase has been discussed in previous chapters. None of Cu-complexes showed inhibitory action towards the phosphorylation of glucose by hexokinase at the concentrations tested. The highest concentration tested was ~3 mM, as the intensity of the ligand chromophore complicated testing at higher concentrations. 6.3.3 M C u Radiolabelling Both the pyridinone (HL1'3\"5) and tetrahydrosalen pro-ligands ( H 2 L 1 0 \" 1 1 ) were labeled with 6 4 C u in high yield (Schemes 6.land 6.2). Addition of 6 4 C u 2 + in 0.1 N HCI to an aqueous or ethanolic pro-ligand solution, depending on pro-ligand solubility, followed by neutralization with NaOH facilitated complex formation. No increase in yield was observed upon heating nor upon extending the reaction time. Pro-ligand concentrations of a least 10\" M were required to obtain high yields; radiolabelling of other ligands, such as tetrazamacrocycles, have also used 10\"2 M or greater ligand solution concentrations.26' 29 HPLC analysis, with co-injection of the non-radioactive standard, was used to identify the radiolabelled product (Figure 6.4), and radiochemical yields and retention times of the complexes (radiation detection) and standards (UV detection) are given in Table 6.3. 160 OH ^ ^ V ^ O H HO o y=< O R R = H 10 R = t-butyl H 2 L 11 64 CuCl 2 0.1 N N a O H R = H 6 4 C u L 1 0 R = t-butyl 6 4 C u L n HO Scheme 6.2: Radiolabelling of tetrahydrosalen pro-ligands. 161 a b c E < 500 Time /s 1000 500 Time Is 1000 LA 500 Time /s 1000 a> o c ro \u00E2\u0080\u00A2e o to < 500 Time/s 1000 Figure 6.4. HPLC Radiation (top) and U V (bottom) traces for a) Cu(L 3 ) 2 and b) C u L 1 1 . 162 Table 6.3. HPLC retention times and radiolabelling yields for 6 4 Cu(L 1 , 3 \" 5 ) 2 and 6 4 C u L 1 0 - n Pro- U V (254 nm) detected RT Radiation detected ligand of standard / mina R T / m i n a % Radiochemical Yield H L 1 9.3 9.3 95+2 H L 3 12.0 12.3 98+2 H L 4 13.0 13.2 99+1 H L 5 6.5 6.7 99+1 H 2 L'\u00C2\u00B0 4.7 4.7 96+2 H 2 L n 13.3b 13.7b 97+3 a HPLC run on Synergi 4um C-l8 Hydro RP analytical column, 1 mL/min, gradient 100% A to 100% B, solvent A: 30 mM K H 2 P 0 4 , solvent B: methanol, b Isocratic 40% A : 60% B. 6.3.4 Cysteine and Histidine Stability Challenges To assess the in vivo stability of the 64Cu-complexes, challenge experiments were carried out in solutions of excess cysteine and histidine, ubiquitous metal binding amino acids found in vivo. After 1 h, HPLC analysis was used to determine the amount of 6 4 C u -complex remaining intact. For the 3-oxy-4-pyridnone complexes, (Cu(L1 , 3\"5)2), no complex remained irrespective of the complex or the amino acid used in the challenge. 163 For the tetrahydrosalen complexes(CuL \" ), no complex remained in the histidine challenge, and less than 20% remained in the cysteine challenge. The slightly higher stability of the tetrahydrosalen complexes over the 3-oxy-4-pyridinone complexes is most likely due to the higher denticity of the ligand (increasing the probability of re-coordinating if one or two of the coordinating atoms is substituted by cysteine or histidine) and the greater steric bulk of the glucose derivatives closer to the Cu centre (protecting the Cu centre from incoming competing amino acid ligands). The higher stability of the tetrahydrosalen complexes in the cysteine challenge experiments over the histidine challenge experiments suggests the higher affinity of Cu for histidine. Both the 3-oxy-4-pyridinone and tetrahydrosalen Cu complexes were not stable in the presence of excess histidine or cysteine, despite the noted thermodynamic stability of related complexes with the same binding moiety.4 3'4 4 Once again, the importance of kinetic inertness in the development of radiopharmaceuticals is illustrated. Despite the observed lability, the complexes may still have potential as radiopharmaceuticals if they are uptaken into target tissue before transchelating to cysteine and histidine residues in proteins. Furthermore, the lability may produce a mechanism for intracellular trapping, whereby the 6 4 C u becomes protein bound once inside the target cells. 6.3.5 P E T Imaging Studies Imaging studies with each of the 6 4 C u complexes were carried out using a small animal PET camera. Preliminary results suggest that some of the complexes may have potential as imaging agents, despite their noted instability. The dynamic scans of 6 4 Cu(L 4 ) 2 were similar to those of the free 6 4 C u 2 + ion with both appearing to clear 164 through the bladder. One apparent difference in the scans for 6 4 C u ( L 4 ) 2 , compared to 6 4 C u 2 + , was the higher uptake in the lung, which did not clear by the 3 h time point. Tumour uptake was observed in the static scans 3 h after injection for 6 4 C u ( L 1 , 3 ' 5 ) 2 and 6 4 C u L 1 0 \" u , and was compared to the tumour activity at 3 h for 6 4 Cu(L 4 ) 2 and 6 4 C u 2 + (Table 6.4). While 6 4Cu(L 4) 2 , in which the carbohydrate is functionalized at the C6 Table 6.4. Comparison of tumour activity / unit area (10\"7Bq/pixel). Species Tumour activity / unit area 6 4 C u 2 + 6 4 Cu(L 1 ) 2 6 4 Cu(L 3 ) 2 6 4 Cu(L 4 ) 2 6 4 Cu(L 5 ) 2 6 4 C u ( L 1 0 ) 2 6 4 C u ( L n ) 2 3.7+0.4 3.4+0.2 4.48+0.05 5.7+0.6 4.10+0.05 4.4+0.1 4.7+0.3 position, had comparable tumour activity to that of Cu , the other complexes, functionalized at either the CI or C2 position of the carbohydrate, showed higher tumour activity. Since the imaging results for the complexes differ from that of free 6 4 C u 2 + , the ligands must influence the biodistribution of the complexes to some extent, even if they are substituted by protein amino acid residues at some time point. As these are preliminary results, further evaluations are needed. 165 6.4 Conclusions This chapter examines the first reported glucose conjugates of 6 4 C u as potential radiopharmaceuticals. Using 3-hydroxy-4-pyridinone pro-ligands bearing pendant carbohydrate derivatives, a series of Cu-complexes were prepared and characterized by various techniques. High radiochemical yields were obtained radiolabelling both the 3-hydroxy-4-pyridinone and tetrahydrosalen pro-ligands with 6 4 C u . The 64Cu-complexes proved to be unstable in solutions of excess histidine or cysteine, although the tetrahydrosalen complexes were slightly less labile than the bis(3-oxy-4-pyridinato)Cu(II) complexes. Despite the instability of the 64Cu-complexes, preliminary PET imaging results showed differences in the images from the 64Cu-complexes compared to the images of the free ion 6 4 C u 2 + , including higher tumour uptake. Further evaluation of these preliminary results, including comparison of the activity in other tissues, will give a better indication of the potential of these complexes as radiopharmaceuticals. 6.5 References 1. McCarthy, D. W.; Bass, L. A.; Cutler, P. D.; Shefer, R. E.; Klinkowstein, R. E.; Herrero, P.; Lewis, J. S.; Cutler, C. S.; Anderson, C. J.; Welch, M . J., Nucl. Med. Biol. 1999, 26, 351-358. 2. McCarthy, D. W.; E. , S. R.; Klinkowstein, R. E.; Bass, L. A.; Margenau, W. H.; Cutler, C. S.; J., A. C ; Welch, M . J., Nucl. Med. Biol. 1997, 24, 35-43. 166 3. Mirzadeh, S.; Mausner, L. F.; Srivastava, S. C , Appl. Rad. hot. 1986, 37, 29-36. 4. Haynes, N. G.; Lacy, J. L.; Nayak, N.; Martin, C. S.; Dai, D.; Mathias, C. J.; Green, M . A., J. Nucl. Med. 2000, 41, 309-314. 5. Smith, S. V.; Waters, D. J.; Di Bartolo, N., Radiochim. Acta 1996, 75, 65-68. 6. Williams, H. A.; Robinson, S.; Julyan, P.; Zweit, J.; Hastings, D., Eur. J. Nucl. Med. Biol. 2005, 32, 1473-1480. 7. Miessler, G. L.; Tarr, D. A., Inorganic Chemistry. 2nd ed.; Prentice Hall: Upper Saddle River, NJ, 1999. 8. Maurer, R. I.; Blower, P. J.; Dilworth, J. R.; Reynolds, C. A.; Zheng, Y.; Mullen, G. E. D., J. Med. Chem. 2002, 45, 1420-1431. 9. Mathias, C. J.; Bergmann, S. R.; Green, M . A., Nucl. Med. Biol. 1993, 20, 343-349. ^ \u00E2\u0080\u009E 10. Fubiyashi, Y.; Taniuchi, H.; Yonekura, Y.; Ohtani, H.; Konishi, J.; Yokoyama, A., J. Nucl. Med. 1997, 40, 1155-1160. 11. Lewis, J. S.; McCarthy, D. W.; McCarthy, T. J.; Fujibayashi, Y.; Welch, M . J., J. Nucl. Med. 1999, 40. 12. Dearling, J. L.; Lewis, J. S.; Mullen, G. E. D.; Welch, M . J.; Blower, P. J., J. Biol. Inorg. Chem. 2002, 7, 249-259. 13. Holland, J. P.; Green, J. C ; Dilworth, J. R., Dalton Trans. 2006, 783-794. 14. Cowley, A. R.; Dilworth, J. R.; Donnelly, P. S.; Gee, A. D.; Heslop, J. M . , Dalton Trans. 2004, 2404-2412. 15. Obata, A.; Kasamatsu, S.; Lewis, J. S.; Furukawa, T.; Takamatsu, S.; Toyohara, J.; Asa, T.; Welch, M . J.; Adams, S. G.; Hideo, S.; Yonekura, Y.; Fujibayashi, Y., Nucl. Med. Biol. 2005, 52,21-28. 167 16. Anderson, C. J.; Green, M . A.; Fujibayashi, Y., Chemistry of copper radionuclides and radiopharmaceutical products. In Handbook of Radiopharmaceuticals Radiochemistry and Applications., Welch, M . J.; Redvanly, C. S., Eds. John Wiley and Sons Ltd.: West Sussex, England, 2003; pp 401-422. 17. Li , M . ; Meares, C. F., Bioconjugate Chem. 1993, 4, 275-283. 18. McQuade, P.; Knight, L. C ; Welch, M . J., Bioconjugate Chem. 2004,15, 988-996. 19. Wu, Y.; Zhang, X.; Xiong, A.; Cheng, Z.; Fisher, D. R.; Liu, S.; Gambhir, S. S.; Chen, X.,J.Nucl. Med. 2005, 46, 1707-1718. 20. Sun, X.; Fang, H.; Li , X.; Rossin, R.; Welch, M . J.; Taylor, J.-S., Bioconjugate Chem. 2005,16, 294-305. 21. McQuade, P.; Yubin, M . ; Yoo, J.; Quinn, T. P.; Welch, M . J.; Lewis, J. S., J. Med. Chem. 2005, 48, 2985-2992. 22. Chen, X.; Park, R.; Huo, Y.; Tohme, M. ; Shahinian, A. H.; Bading, J. R.; Conti, P. S.,J. Nucl. Med. 2004, 45, 1390-1397. 23. Moi, M . K.; Meares, C. F.; McCall, M . J.; Cole, W. C ; DeNardo, S. J., Anal. Biochem. 1985,148, 249-253. 24. Obata, A.; Fujibayashi, Y.; Miono, Y.; Waki, A.; Yonekura, Y.; Welch, M . J., J. Labelled Compd. Radiopharm. 2001, 42, S273-S275. 25. Yoo, J.; Reichert, D. E.; Welch, M . J., J. Med. Chem. 2004, 47, 6625-6637. 26. Sun, X.; Wuest, M. ; Weismann, G. R.; Wong, E. H.; Reed, D. P.; Boswell, C. A.; Motekaitis, R.; Martell, A. E.; Welch, M . J.; Anderson, C. J., J. Med. Chem. 2002, 45, 469-477. 168 27. Boswell, C. A.; Sun,.X.; Niu, W.; Weismann, G. R.; Wong, E. H.; Rheingold, A. L.; Anderson, C. J., J. Med. Chem. 2004, 47, 1465-1474. 28. Boswell, C. A.; McQuade, P.; Weismann, G. R.; Wong, E. H.; Anderson, C. J., Nucl. Med. Biol. 2005, 32, 29-38. 29. Jones-Wilson, T. M . ; Deal, K. A.; Anderson, C. J.; McCarthy, D. W.; Kovacs, Z.; Motekaitis, R.; Sherry, D. A.; Martell, A. E.; Welch, M . J., Nucl. Med. Biol. 1998, 25, 523-530. 30. Clarke, E. T.; Martell, A. E. , Inorg. Chim. Acta 1991,190, 27-36. 31. Bryan, J. N.; Jia, F.; Mohsin, H.; Sivaguru, G.; Miller, W. H.; Anderson, C. J.; Henry, C. J.; Lewis, M . R., Nucl. Med. Biol. 2005, 32, 851-858. 32. Cerchiaro, G.; Sant'Ana, A. C ; Temperini, M . L. A.; Ferreira, A. M . , Carbohydrate Res. 2005, 340, 2352-2359. 33. Striegler, S.; Dittel, M . , J. Am. Chem. Soc. 2003,125, 11518-11524. 34. Tanase, T.; Inukai, H.; Onaka, T.; Kato, M. ; Shigenobu, Y.; Lippard, S. J., Inorg Chem. 2001, 40, 3943-3953. 35. Jezowska-Bojczuk, M . ; Chruscinska, E.; Trnka, T.; Micera, G., J. Inorg. Biochem. 1996, 63, 231-239. 36. Kato, M . ; Tanase, T., Inorg. Chem. 2005, 44, 8-10. 37. Kato, M . ; Tanase, T.; Mikuriya, M . , Inorg. Chem. 2006, 45, 2925-2941. 38. Zhang, W.; Jiang, T.; Ren, S.; Zhang, Z.; Guan, H.; Yu, J., Carbohydrate Res. 2004, 339, 2139-2143. 39. Sah, A. K.; Tanase, T.; Mikuriya, M. , Inorg. Chem. 2006, 45, 2083-2092. 40. Mikata, Y.; Sugai, Y.; Yano, S., Inorg. Chem. 2004, 43, 4778-4780. 169 41. Mikata, Y.; Sugai, Y.; Obata, M. ; Harada, M. ; Yano, S., Inorg. Chem. 2006, 45, 1543-1551. 42. Diaz, A.; Garcia, I.; Cao, R.; Beraldo, H.; Salberg, M . M . ; West, D. X.; Gonzalez, L.; Ochoa, E. , Polyhedron 1997,16, 3549-3555. 43. El-Jammal, A.; Howell, P. L.; Turner, M . A.; Li , N.; Templeton, D. M . , J. Med. Chem. 1994,57,461-466. 44. Gnuenwedel, D. W., Inorg. Chem. 1968, 7, 495-501. 45. Boettcher, A.; Elias, H.; Jager, E. G.; Langfelderova, H.; Mazur, M . ; Mueller, L.; Paulus, H.; Pelikan, P.; Rudolph, M. ; Valko, M. , Inorg. Chem. 1993, 32, 4131-4138. 46. Soto-Garrido, G.; Salas-Reyes, V., Trans. Met. Chem. 2000, 25, 192-195. 47. Klement, R.; Stock, F.; Elias, H.; Paulus, H.; Pelikan, P.; Valko, M . ; Mazur, M. , Polyhedron 1999,18, 3617-3628. 48. Valko, M . ; Klement, R.; Pelikan, P.; Boca, R.; Dlhan, L.; Boettcher, A., J. Phys. Chem. 1995, PP, 137-143. 49. Storr, T. J. Designed Ligands for Diagnosis and Therapy. Ph.D. Thesis. University of British Columbia, Vancouver, BC, 2005. 170 Chapter 7 Cobalt(II) Complexes of 3-Hydroxy-4-pyridinone Pro-Ligands Bearing Pendant Carbohydrates1 7.1 Introduction Two isotopes of Co that have properties relevant to nuclear medicine applications 57 55 are Co and Co. Both isotopes are accelerator produced, but have sufficiently long half-lives to allow transportation to hospitals some distance from production. The only clinically used Co radiopharmaceutical is 57Co-cyanocobalamin (vitamin B 1 2 ) , used in evaluating vitamin B12 absorption by monitoring 5 7 Co activity in the patient's urine.1 5 7 Co has been considered for use in SPECT imaging,2\"4 but the long half-life (271 d) results in a high radiation dose to the patient making it a poor imaging radioisotope.1 5 5 Co has a shorter half-life of 17.6 h and can be used in PET imaging, but relatively'few Co complexes have been investigated as potential PET imaging agents. The free ion 5 5 C o 2 + has been investigated for imaging cellular C a 2 + influx in ischemic brain tissue.5\"8 One complex, 5 5 C o - E D T A (ethylenediaminetetraacetic acid), has been examined as a potential renal imaging agent.9 * A version of this chapter will be submitted for publication. Ferreira, C. L., Lapi, S., Steele, J., Green, D . E., Adam, M. J., Orvig, C. Cobalt(II) Complexes of 3-Hydroxy-4-pyridinone Pro-Ligands Bearing Pendant Carbohydrates. 171 In this chapter Co-complexes of the carbohydrate bearing 3-hydroxy-4-pyridinone pro-ligands (Figure 7.1) were prepared as potential imaging agents, adding to the small number of Co-complexes studied to date for nuclear imaging. F e r a l e x - G Figure 7.1. 3-Hydroxy-4-pyridinone pro-ligands bearing pendant carbohydrates. 7.2 Experimental 7.2.1 Materials and Methods Information related to this section can be found in section 2.2.1. 172 7.2.2 Synthesis of Co(II) Complexes General preparation of bispyridinatocobalt(II) complexes: Stoichiometric amounts of cobalt(II) chloride hexahydrate (one equivalent) and the respective pyridinone ligand HL 1 ' 3 \" 5 (two equivalents) were dissolved separately in 1:1 mixtures of methanol and water. The cobalt solution was added to the ligand solution, resulting in a clear pink solution with an initial pH of ~3. The pH was adjusted to ~7 with 0.1 M NaOH resulting in the formation of a pink precipitate. The flask was cooled overnight, and the pink precipitate filtered out and washed with cold water, before drying over P2O5 in a vacuum dessicator. Bis(l-{N-[p-benzyl p-D-glucopyranoside]ethanamide}-2-methyl-3-oxy-4(lH)-pyridinato)cobalt(II) heptahydrate (Co(L1)2 -7H20). Cobalt(II) chloride hexahydrate (24 mg, O H \" S S ^ o ^ o 0-10 mmol) and H L 1 (87 mg, 0.20 \ ^ N / ^ N Y U / / mmol) yielded a pink solid (79 mg, \ H C H 3 A 85%). IR (KBr disc, cm\"1): 3415 (br) (vOH), 1640 (s) (v(CONH)), 1605 (s) (v(CO)), 1546 (s), 1508 (s), 1485 (s), (aryl), 1382 (s), 1285 (m) (8(OH)), 836 (m) (para-substituted benzene). UV-visible spectrum (H 20): XmJnm (sm a x/ L M\" 1 cm\"1) = 285 (2.2 x 104), 311 (2.2 x 104). MS (ESI+): m/z = 952 ([M+Na]+, 100). Anal. Calcd. for C 4 oH 4 6 CoN 4 Oi 8 -7H 2 0: C, 45.50; H, 5.73; N, 5.31. Found: C, 45.87; H, 5.96; N, 5.18. 173 Bis(l-{p-phenyl P-D-glucopyranoside}-2-methyl-3-oxy-4(lH)-pyridinato)cobalt(II) pentahydrate (Co(L 3) 2 -5H20). 0 H Cobalt(II) chloride hexahydrate (26 mg, 0.11 H \u00C2\u00B0 ^ ^ ^ 2 ^ 0 - ^ ^ J H , \ mmol) and H L 3 (83 mg, 0.22 mmol) yielded a ^ tTjC >^ co p i n k sol id (74 m g ' 7 5 % ) ' I R ( K B r disc' cm\"I): 2 1594 (s), (v(CO)), 1542 (s), 1510 (s), 1469 (s) (aryl), 1370 (s), 1311 (m), 1235 (m) (8(OH)), 819 (s) (para-substituted benzene). UV-visible spectrum (H 20): XmJnm (emJ L M\" 1 cm\"1) = 294 (2.7 x 104), 312 (2.5 x 104). MS (ESI+): m/z = 815 ([M+H]+, 100). Anal. Calc. for C 3 6 H4oCuN 2 Oi 6 -5H 2 0: C, 47.74; H, 5.56; N, 3.09. Found: C, 47.62; H, 5.73; N, 3.37. Bis(l-{N-[methyI 6-amino-6-deoxy-a-D-glucopyranoside] ethanamide}-2-methyl-3-oxy-4(lH)-pyridinato)cobalt(II) trishydrate (Co(L4)2 -3H20). Cobalt(II) chloride hexahydrate (24 mg, 0.10 mmol) and X^M^^i C o H I j 4 ( 7 2 m \u00C2\u00A7 ' 0 2 0 m m o 1 ) y i e l d e d a P i n k s o l i d ( 6 8 mg> 88%). IR (KBr disc, cm\"1): 3472 (br) (vOH), 1659 (s) HN HO' \ H \u00C2\u00B0 \" OH \ OCH 3 / 2 (v(CONH)), 1602 (s) (v(CO)), 1547 (m), 1485 (s) (aryl), 1384 (m), 1288 (m) (8(OH)). UV-visible spectrum (H 20): XmJnm (emJ L M\" 1 cm\"1) 286 (1.7 x 104), 311 (1.4 x 104). MS (ESI+): m/z = 796 ([M+Na]+, 100), 774 ([M+H]+, 34). Anal. Calcd. for C 3oH4 2CoN 4Oi6-3H 20: C, 43.54; H, 5.85; N, 6.77. Found: C, 43.54; H, 5.72; N, 6.70. 174 Bis(l-{N-[2 -amino-2-deoxy-D-glucopyranose]ethanamide}-2-methyl-3-oxy-4(lH)-pyridinato)cobalt(II) pentahydrate(Co(L5)2 -5H20). 0 H Cobalt(II) chloride hexahydrate (19 mg, 0.08 mmol) H H O ^ ^ \ - O H ^ a n d h l 5 ( 5 4 m g ' \u00C2\u00B0 - 1 6 m m o l ) y i e l d e d a P i n k s o l i d N H i ^ V ^ \" n C o o ^ ^ / N ^ k Q A (45 mg, 76%). IR (KBr disc, cm -1): 3297 (br) c \" 3 / 2 (vOH), 1659 (s) (v(CONH)), 1603 (s) (v(CO)), 1546 (m), 1489 (m) (aryl), 1385 (m), 1289 (m) (8(OH)). UV-visible spectrum (H 20): Xm a x/nm (sm a x/ L M\" 1 cm\"1) = 283 (1.8 x 104), 308 (9.5 x 103). MS (ESI+): m/z = 768 ([M+Na]+, 100). Anal. Calcd. for C 2 8 H 3 8CoN 4 0 1 6 -5H 2 0: C, 42.79; H, 5.73; N, 7.13. Found: C, 42.52; H, 5.83; N, 7.31. 7.2.2 Hexokinase Inhibition Assays Each of the complexes was tested as a potential inhibitor of glucose phosphorylation by hexokinase as described in section 2.2.7. 7.3 Results and Discussion 7.3.1 Synthesis and Characterization of Bis(3-oxy-4-pyridinato)cobaIt(II) Complexes Bis-ligand Co(II) complexes of the 3-hydroxy-4-pyridinone pro-ligands were prepared in good yields (75%-88%) (Scheme 7.1). Complexes were synthesized by dissolving stoichiometric amounts of the ligand (two equivalents) and cobalt(II) 175 Co(L 4) 2 Co(L5)2 Scheme 7.1. Synthesis of Co(II) complexes, Co(Ll,3'5)2 chloride hexahydrate (one equivalent) in a methanolic solution, then neutralizing the pH with aqueous NaOH. A light pink precipitate readily formed and was purified by 176 washing with cold water and/or methanol to remove any residual ligand or uncomplexed Co. Attempts to prepare tris-ligand octahedral complexes with Co(II), either by using three equivalents or larger excesses of the pro-ligand, resulted only in the bis-ligand complexes. Each complex was characterized by mass spectrometry, elemental analysis, UV-visible and infrared spectroscopy. The major peak observed in the mass spectrum of each complex was assigned to either the [M+Na]+ or [M+H]+ ion with 100% relative intensity. Infrared spectra were similar to the free ligand spectra, except for the shift of the resonances assigned to the pyridinone ring ketone carbonyl stretches (Table 7.1), Table 7.1. Comparison of pyridinone ring IR spectra vCO for free pro-ligands and Co-complexes. Pro-ligand vCO free ligand vCO Co-complex A vCO / cm\"1 (HL) / cm\"1 (CoL 2) / cm\"1 H L 1 1631 1605 -26 H L 3 1631 1594 -37 H L 4 1632 1602 -30 H L 5 1638 1603 -35 confirming binding of the pyridinone moiety. Back bonding from the Co d-orbitals to the TI C=0 anti-bonding orbitals of the pyridinone would result in a shift of the C=0 stretching frequency to a lower frequency. The UV- visible spectra peaks were attributed to the ligand chromophore. Co(II) compounds have a d 7 electron configuration which has 177 characteristic d-d electron transitions in the visible region. Tetrahedral Co(II) complexes have strong d-d electron transitions, while octahedral Co(II) complexes have weak, symmetry forbidden d-d transitions. Since the d-d transitions were not observed for any of the complexes (even at the highest concentrations before solubility becomes an issue) it can be assumed that the complexes are octahedral; this is also supported by the light pink colour observed which is associated with octahedral Co(II) complexes.10 The inability to coordinate three ligands to the Co(II) centre suggests that the complexes may be distorted towards a square planar geometry, due to Jahn Teller effects, which can be expected for low spin d 7 metal centers. Elemental analysis confirmed the composition of the bulk samples to be hydrated bis-ligand Co complexes, with each complex having a stoichiometric number of water molecules (3-7) associated with the vacuum dried samples. The water molecules may be associated with the carbohydrate derivatives, or may reside in the coordination sphere around the metal centre. 7.3.2 Hexokinase Inhibition Assays None of Co-complexes showed inhibitory action towards the phosphorylation of glucose by hexokinase. 7.4 Conclusion The Co(II) complexes, Co(L 1 , 3\" s)2, were prepared in good yield and characterized by several methods. Although these complexes do not interact with hexokinase, the first 178 enzyme in glucose metabolism, further investigation of these complexes as diagnostic imaging agents is warranted. Since few 5 5 Co complexes have been studied, basic investigation of the radiochemistry, stability, and biological properties of bis(pyridinato)cobalt(II) complexes is of general interest to the development of 5 5 Co PET agents. Thus, radiolabelling with 5 5 Co and studies with the radiolabelled complexes are planned. 7.5 References 1. Jurisson, S.; Berning, D.; Jia, W.; Ma, D., Chem. Rev. 1993, 93, 1137-1156. 2. Front, D.; Israel, O.; Iosilevsky, G.; Even-Sapir, E.; Frenkel, A.; Peleg, H.; Steiner, M. ; Kuten, A.; Kolodny, G. M . , Radiology 1987,165, 129-133. 3. Stevens, H.; Knollema, S.; Piers, D. A.; Van De Wiele, C.; Jansen, H. M . L.; De Jager, A. E. J.; De Reuck, J.; Dierckx, R.; Korf, J., Nucl. Med. Comm. 1998, 19, 573-580. 4. Joosten, A. A.; Jansen, H. M . L.; Piers, D. A.; Minderhoud, J. M . ; Korf, J., Nucl. Med. Comm. 1995,16, 703-705. 5. De Reuck, J.; Paemeleire, K.; Santens, P.; Strickmans, K.; Lemahieu, I., Clin. Neurol. Neurosurg. 2004,106, 77-81. 6. De Reuck, J.; Santens, P.; Strickmans, K.; Lemahieu, I., J. Neurol. Sci. 2001,193, 1-6. 7. De Reuck, J.; Vonck, K.; Santens, P.; Boon, P.; De Bleecker, J.; Strickmans, K.; Lemahieu, I., J. Neurol. Sci. 2000,181, 13-18. 179 8. Jansen, H. M . L.; Pruim, J.; Vliet, A. M. ; Hew, J. M . ; Fanssen, E. J.; de Jong, B. M. ; Kosterink, J. G.; Haaxma, R.; Korf, J., J. Nucl. Med. 1994, 35, 456-460. 9. Goethals, P.; Volkaert, A.; Vandewielle, C ; Dierckx, R.; Lamiere, N., Nucl. Med. Biol. 2000,27, 77-81. 10. Cotton, F. A.; Wilkinson, G.; Gaus, P. L., Basic Inorganic Chemistry. John Wiley & Sons, Inc.: New York, 1995. 180 Chapter 8 Synthesis, Structure and Biological Activity of Ferrocenyl-carbohydrate Conjugates1 8.1 Introduction An attractive area of research for new malaria treatments is metal conjugates of known anti-malarial drugs.1\"16 Previous research in this area includes complexes containing mefloquine, quinine, artemisinin, and chloroquine with metal ions such as Fe(II/III),''8 Ga(III),9 Ru(II),1 0'1 1 Rh, 1 1 Au(I), 1 2' 1 3 Pd(II)14 and Pt(II)14. Metal complexes with ligands containing chloroquine, or quinine type structures, have been shown to 11 i in i i i i retain efficacy in chloroquine resistant strains ' ' ' ' ' including an Au(I) complex with a carbohydrate, 1-deoxy-l-thio-P-D-glucopyranoside, co-ligand (Fi gure8.1).13 The most potent metalloantimalarials are chloroquine based metallocenes of Fe(II)2 and Ru(II) (Figure 8.1).10 Anti-malarial activity has also been observed in metal complexes that do not incorporate a known anti-malarial drug,16\"19 including amine phenol Ga(III) and Fe(III) complexes17'18, and ferrocenyl-carbohydrate derivatives.19 The amine phenol Ga(III) and A version of the chapter has been accepted for publication. Ferreira, C. L., Ewart, C. B., Barta, C. A., Little, S, Yardley, V , Martins, C , Polishchuk, E., Smith, P. J., Moss, J. R., Adam, M. J , Orvig, C. Synthesis, Structure and Biological Activity of Ferrocenyl-carbohydrate Conjugates. Inorg. Chem. 2006, In press. 181 Fe(III) complexes were found to be inhibitors of hemozoin formation with strong anti-plasmodial activity, suggesting a similar mode of action to chloroquine. Interestingly, one of these complexes showed efficacy against a chloroquine resistant strain only.18 Ferrocenyl ellagitannin derivatives have shown moderate to high anti-plasmodial activity, while the ellagitannin derivative without the ferrocene moiety had no activity.19 The metalloantimalarial mechanism of enhanced activity in resistant strains is unclear, but quite pronounced. For quinine type anti-malarials, such as chloroquine, the drug accumulates in the parasite food vacuoles where it forms a complex with ferric heme (ferriprotoporphyrin IX); a product of the parasite sequestering and consumption of hemoglobin. The drug-ferric heme complex halts the detoxification of ferric heme to hemozoin, resulting in parasite death. It has been suggested that the resistant parasite accumulates less of the drug in its food vacuoles.20 Varying the anti-malarial drug structure by incorporation of a lipophilic, redox-active metal center should increase membrane permeability, and possibly aid in the accumulation of drug in the resistant parasite food vacuoles, thereby increasing efficacy. A secondary effect might be the toxicity of the metal, or toxicity induced by intracellular oxidation. The metal complex may also have improved affinity for the drug target. 182 Ferrocenoyl-carbohydrate conjugates have potential as metalloantimalarials. Combining the ferrocene moiety with a glucose derivative is a novel approach for developing targeted therapy. 1 - 8 The ferrocene moiety has proven to be a successful addition to known malaria therapeutics, increasing efficacy towards chloroquine resistant strains of the parasite.1'2'5\"7'15 As well, glucose uptake and metabolism in infected erythrocytes is elevated at all stages of the parasite's life cycle; 2 1 ' 2 2 and glucose consumption has been a target in anti-malarial research.23 The hypothesis of this work is that ferrocene-carbohydrate conjugates have the potential to retain activity in chloroquine resistant parasite strains, and to have increased efficacy by targeting infected cells. This work studies the cytotoxicity and anti-plasmodial activity of several ferrocene-carbohydrate conjugates, and includes the synthesis and characterization of three new ferrocene-carbohydrate conjugates. 8.2 Experimental 8.2.1 Materials and Methods Most information pertaining to this section can be found in section 2.2.1. The starting materials ferrocene carbonyl chloride, l,l'-bis(carbonyl chloride) ferrocene24, ferrocenyl benzotriazolate (FcCOOBt) 2 5, methyl 2,3,4-tetra-0-acetyl-6-azido-6-deoxy-a-Ofi 77 D-glucopyranoside , and methyl 6-amino-6-deoxy-a-D-glucopyranoside were prepared using previously published methods. The previously published ferrocenoyl 78 78 78 78 70 carbohydrate compounds 1, 2, 6, 7 , and 11 were prepared as described in the literature. Compounds 4, 5, 9, and 10 were prepared in our research group by Charles 183 Ewart as part of his honours research project under my supervision. The cell viability assay culture medium reagents were acquired from Gibco products (Grand Island, NY), and well plates were acquired from Beckton Dickinson and Co. (Franklin Lakes, NJ). 8.2.2 Synthesis of Ferrocenyl-carbohydrate Conjugates Methyl 2,3,4-tri-0-acetyl-6-amino-6-deoxy-a-D-gIucopyranoside hydrochloride salt (13) Methyl 2,3,4-tri-0-acetyl-6-azido-6-deoxy-o>D-glucopyranoside (760 mg, 2.04 mmol) was dissolved in a mixture of ethyl acetate O C H 3 (20 mL) and methanol (20 mL), and the mixture was transferred into a Parr reactor apparatus fitted with a thick wall glass insert. To the reaction, palladium black (23 mg, 0.20 mmol) wetted with water was added, followed by hydrochloric acid (2 M , 1 mL, 0.2 mmol). The Parr reactor was sealed and pressurized with H2 gas (100 atm); after stirring the reaction solution rigorously for 24 h, the H2 gas pressure was released and the reaction mixture was filtered through a medium frit to remove the palladium catalyst, which was rinsed with methanol. Rotary evaporation of the filtrate solvent afforded an off white solid. The solid was dissolved in methanol and precipitated with the addition of ethyl acetate. After cooling the flask overnight, the precipitate was filtered out, and dried in vacuo to yield a white solid which was used without further purification (394 mg, 54%). } H NMR (CDCI3, 400 MHz) 8 1.99, 2.03, 2.05 (3s, 9H, C//3COO), 3.04 (dd, 1H, H6a, 3J5M = 8.4 Hz, 2J6^h = 13.4 Hz), 3.19 (dd, 1H, H6b, 3 / 5 > 6 b = 2.88 Hz, 2 J 6 a ,6b = 13.4 Hz), 3.47 (s, 3H, OC/ / 3 ) , 4.05 (ddd, 1H, H5,3J5,6b 184 = 2.88 Hz, 3 J 5 ,6a = 8.36 Hz, V 4 ; 5 = 10.63 Hz), 4.86 (dd, 1H, HI, 3 J 1 > 2 = 3.66 Hz, 3 J 2 ; 3 = 10.05 Hz), 4.91 (t, 1H, HA,3J3,4 = 9.75 Hz, 3 J 4 , 5 = 9.79 Hz), 5.03 (d, 1H, HI, Vi, 2 = 3.5 Hz), 5.45 (t, 1H, Hi, V 3,4 = 9.74 Hz, V 2 j 3 = 10.1 Hz). IR (NaCl, cm\"1): 3128-2761 (br, vNH), 1746 (vCO), 1608, 1500 (s, vNH). General method for the preparation of 1-10 One or two equivalents of the appropriate sugar were combined with one equivalent of either ferrocene carbonyl chloride or l,l'-bis(carbonyl chloride) ferrocene. The reagents were dissolved with stirring in dry dichloromethane (10 mL) and placed under nitrogen. Either one or two equivalents of pyridine were added via a syringe to the reaction mixture, and stirring continued. For coupling to thio (2, 7) and amino sugars (1, 3, 6, 8) the reaction was quenched with water after stirring at room temperature for 30 min. For coupling involving hydroxyl groups (4-5, 9-10) the reaction was stirred while refluxing for 30 h. The reaction was quenched with water (15 mL), washed with 5% bicarbonate (15 mL x 2) and then washed again with water (15 mL x 2). The organic layer was dried with MgS0 4 , clarified by filtration, and reduced in volume by rotary evaporation. The crude product was purified using column chromatography and/or recrystallization. 6-N-(Methyl 2,3,4-tri-0-acetyl-6-amino-6-deoxy-a-D-glucopyranoside)}-l-ferrrocene carboxamide (3) Ferrocene carbonyl chloride (129 mg, 0.518 mmol), 13 (164 mg, 0.461 mmol) and anhydrous pyridine (0.08 mL, 1.0 185 mmol) gave an orange oil. Purification by recrystallization from hot ethanol afforded orange needle crystals (98 mg, 35%). 'H NMR (CDCI3, 300 MHz): 8 1.70, 1.81, 1.86 (3s, 9H, C//3COO), 3.04 (s, 3H, OC/ / 3 ) , 3.44 (dt, 2H, #6, 3 J 5 , 6 = 5.79 Hz, 2 J 6 a > 6 b = 15.03 Hz), 3.88 (m, 2H, #5, #6), 4.06 (t, 2H, #10, #10', l/9,io = 1.93 Hz), 4.20 (s, 5H, #Cp), 4.65, 4.79 (2d, 2H, #9, #9', 3J9,io = 1-55 Hz), 5.03 (d, 1H, #1, 3J h 2 = 3.66 Hz), 5.19 (dd, 1H, #2, 3Ji, 2 = 3.66 Hz, 3 y 2 ,3 = 10.21 Hz), 5.30 (t, 1H, #4, 3J 3,4 = 9.83 Hz), 5.96 (t, 1H, #3, 3J 3,4 = 9.83 Hz), 6.04 (t, 1H, N#, 3 y 2 , N H = 6.10 Hz). MS (LSIMS+): m/z 531 ([M+H]+, 100). Anal. Calcd. for C 1 7 H 2 9FeN0 9 : C, 54.26; H, 5.50; N, 2.64. Found: C, 54.60; H, 5.45; N, 2.64. Bis{6-N-(methyl 2, 3, 4-tri-0-acetyl-6-amino-6-deoxy-a-D-glucopyranoside)}-l,l'-ferrrocenecarboxamide (8) 1,1'- Bis(carbonyl chloride) ferrocene (84 mg, 0.27 mmol), 13 (200 mg, 0.56 mmol) and anhydrous pyridine (0.10 mL, 1.2 mmol) yielded a red oil (125 mg, 53%). 'H N M R A C O ^ ^ \u00E2\u0080\u0094 \u00E2\u0080\u0094 \u00C2\u00B0 \ (CDCI3, 300 MHz): 8 1.74, 1.85, 1.96 (3s, 18H, C# 3COO), AcoX\u00E2\u0080\u0094-\"\"T-A. O A r 1 0CH3 3.18 (s, 6H, OC# 3), 3.73 (m, 2H, #6), 3.94 (m, 2H, #6), 4.08 (m, 2H, #5), 4.15, 4.21 (2 broad s, 4H, #10, #10') 4.80, 4.84 (2 broad s, 4H, #9, #9'), 4.82 (d, 2H, #1, V u = 3.27 Hz), 5.34 (dd, 2H, #2, 3 J , ; 2 = 3.28 Hz, 3 J 2 > 3 = 10.21 Hz), 5.49 (t, 2H, #4, 3y 3,4 = 9.83 Hz), 6.01 (t, 2H, #3, 3J 3,4 = 9.83 Hz), 7.16 (broad t, 2H, N#). MS (LSIMS+): m/z 877 ([M+H]+, 100). Anal Calcd. for C 3 8 H 4 8 F e N 2 0 1 8 : C, 52.06; H, 5.52; N, 3.20. Found: C, 52.00; H, 5.26; N, 3.18. 186 {6-N-(Methyl-6-amino-6-deoxy-a-D-glucopyranoside)}-l-ferrocene carboxamide monohydrate (12) borate buffer solution with stirring. The resulting solution was adjusted to pH 9 using 0.3 N NaOH and stirred overnight. The solvent was removed using a rotary evaporator, and the product was purified by silica column chromatography eluting with 8:2 ethyl acetate: methanol. Fractions containing the product (Rf = 0.5) were reduced in volume on a rotary evaporator and the residue vacuum dried to yield an orange solid (83 mg, 56%). 'H NMR (CD 3 OD, 300 MHz): 5 3.27 (t, 1H, HA, 3J= 9.06 Hz), 3.41 (s, 3H, OC7/3), 3.51 (dd, 1H, HI, 3JU2 = 3.66 Hz, V 2 > 3 = 9.72 Hz), 3.53 (d, 1H, H6b,J= 3.53 Hz) 3.66 (t, 1H, H3,3J= 9.24 Hz), 3.70 (m, 1H, H5,3J= 5.97 Hz, 3J= 4.02 Hz), 3.75 (d, 1H, H6a, J = 2.49 Hz), 4.26 (s, 5H, //Cp), 4.48 (t, 2H, //10, //10', 3J9,io = 1-92 Hz), 4.69 (d, 1H, HI, 3Jh2 = 3.45 Hz), 4.80 (t, 2H, H9, H9', 3J9A0 = 1-92 Hz). MS (LSIMS+): 406 ([M+H]+, 100). Anal. Calc. for C i s ^ F e N C V ^ O : C, 51.09; H, 5.65; N, 3.31. Found: C, 51.00; H, 5.95; N, 3.63. 8.2.3 Solid State Structure Determination of 1, 2 and 12 Orange crystals were grown after several days by cooling concentrated ethanol solutions of the respective compounds. Further information related to this section can be Methyl 6-amino-6-deoxy-a-D-glucopyranoside (71 mg, 0.37 0 mmol) was dissolved in 5 mL of borate buffer (20 mM, pH 9). Fc(COOBt) (190 mg, 0.55 mmol) was dissolved in 5 mL tetrahydrofuran and the solution was slowly added to the 187 found in section 2.2.5. Data were collected by Cheri Barta, and structures were solved by Cheri Barta, Dr. Michael Merkel or myself. 8.2.4 In Vitro Anti-plasmodial Activity Studies These studies were performed in collaboration with Prof. John Moss (Chemistry Department, University of Cape Town, South Africa), and Prof. Peter Smith (Pharmacology Department, University of Cape Town, South Africa), as well as Susan Little and Dr. Vanessa Yardley (Infectious Disease Department, London School of Hygiene and Tropical Medicine). All samples were tested in duplicate on a single occasion against the DIO strain. Continuous in vitro cultures of asexual erythrocyte stages of P. falciparum were maintained using a modified method of Trager and Jensen.31 Quantitative assessment of anti-plasmodial activity in vitro was determined via the parasite lactate dehydrogenase (pLDH) assay using a modified method described by Makler.3 2 Al l compounds were dissolved 2 mg/mL in 10% methanol/DMSO. The samples were then diluted with water to 25, 12.5 and 6.25 ug/mL, and stored at 20\u00C2\u00B0C until use. Chloroquine was used as the positive control in all experiments, and was tested at concentrations of 30, 15 and 7.5 ng/mL. All samples were tested in triplicate against the 3D7 and K l strains. The cultures were maintained in continuous log phase growth in RPMI1640 medium supplemented with 5% wash human A+ erythrocytes, 25 mM Hepes, 32 nM NaHC0 3 , and AlbuMAXII (lipid-rich bovine serum albumin). All cultures and assays were conducted at 37\u00C2\u00B0C 188 under an atmosphere of 5% C 0 2 and 5% 0 2 , and 90% N 2 . Stock compound solutions were prepared in 100% DMSO at 5 mg/mL. The compounds were further diluted using complete medium RPMI1640 supplemented with cold hypoxanthine and AlbuMAXII. Assays were performed in sterile 96-well microtitre plates; each plate contained 100 uL of parasite culture (1% parasitemia, 2.5% hemacrit). Each compound was tested at 30, 10, 3, 1, 0.3, and 0.1 L i g / m L . After 24 h of incubation at 37\u00C2\u00B0C, 3.7 Bq of [ H]hypoxanthine was added to each well. Cultures were incubated for a further 24 h before they were harvested onto glass-fiber filter mats and the radioactivity was counted using a Wallac Microbeta\u00E2\u0084\u00A2 1450 scintillation counter.33 8.2.5 Cell Viability Assays These studies were done in collaboration with Candice Martins and Dr. Elena Polishchuk from the Biological Services Laboratory, Chemistry Department of UBC. HTB-129 cells were detached from culture flasks with 0.25% trypsin and 0.03% EDTA (Sigma) and resuspended in fresh culture medium (90% Leibovitz's L-15 medium with 2 mM L-glutamine, 0.01 mg/mL bovine insulin and 10% fetal bovine serum) at a density of 1 x 105cells/ml. Using a Falcon 96-well, flat bottom plate, 100(0.1 of cell suspension were added to wells. Sterile, distilled water (200pl) was added to the plate's outer wells to prevent evaporation. The cells were incubated for 24 hours at 37\u00C2\u00B0C under 5% C 0 2 ( g ) . Using another Falcon 96-well plate, lOOpl L1210 cell suspension in RPMI1640 medium (200ul) supplemented with 10% heat-inactivated horse serum, 50 IU/ml 189 penicillin, 50 p.g/mL streptomycin, 0.5 mM sodium pyruvate, 0.05% (w/v) pluronic acid F68, and 0.01M HEPES (1 x 105cells/ml) was added to wells. The cells were incubated for 24 hours at 37\u00C2\u00B0C under 5% C02(g). For both cell lines, compounds were tested in triplicate. The compounds were dissolved in DMSO and diluted in culture medium to concentrations of 200, 100, 50, 25, 12.5, 6.25, and 3.13ug/ml. Aliquots (lOOpl) were then transferred to the cells in plate columns 5 through 11, making final experimental compound concentrations of 200, 100, 50, 25, 12.5, 6.25, 3.13, and 1.56 p.g/ml. A DMSO control was also done adding lOOul 2% DMSO. The cells were then incubated for 72 hours under the same conditions described above. Measurement of cell viability was done with 3-(4, 5-dimethylthiazol-2-yl)-2, 5-diphenyltetrazolium bromide (MTT) solution (50ul, 2.5mg/ml) which was added to each experimental well of the plate and the cells were incubated for a further 3 hours. The plates were then spun on an IEC HN-SII centrifuge for 5 minutes at low speed. The supernatant was carefully aspirated from each experimental well and DMSO (150uL/well) was added to each experimental well to dissolve the tetrazolium salts. The plates were read on a Beckman-Coulter DTX800 plate reader (Fullerton, CA) at 570nm. 190 8.3 Results and Discussion 8.3.1 Synthesis and Characterization of Ferrocenyl-carbohydrate Conjugates Both novel and previously reported ferrocenyl-carbohydrate conjugates, with slight variations in structure, were prepared (Schemes 8.1-8.3). Some compounds had only one cyclopentadienyl (Cp) ring substituted (1-5,11-12) with various glucose derivatives, while others had both Cp rings substituted with identical derivatives (6-10). The carbohydrate moieties in these compounds were attached to the ferrocene via amide (1, 3, 6, 8,11,12), ester (4-5, 9-10) or thioester (2, 7) linkages. The ester or amide bond was linked at the C- l (2, 4, 7, 9), C-2 (1, 6,11) or C-6 (3, 5, 8,10,12) position of the sugar. Protecting groups, including methoxy, benzyl, and acetyl groups, were present on the other positions of the sugar. Several known compounds were prepared for study. Compounds 1, 2, 6, 7, and 11, have been previously reported28'29 and were prepared by those literature methods (Schemes 8.1-8.3). Following purification, their structures were verified by 'H NMR spectroscopy. Compounds 4 ,5, 9, and 10 were prepared30 by similar methods and fully characterized by Charles Ewart. The sugar derivative 13 is closely related to known compounds,26'27 and was prepared using similar methods (Scheme 8.4). Conversion of the azide in methyl 2,3,4-tris-0-acetyl-6-azido-6-deoxy glucopyranoside to the amino compound 13 was achieved by palladium metal catalyzed hydrogenation. A small amount of hydrochloric acid was added to the hydrogenation to form a hydrochloride salt with the newly formed amine, thereby obviating intermolecular transamidation. After precipitation as a white solid, 191 compound 13 was verified by H NMR and infrared spectroscopy, and then used without further purification; the ! H NMR spectrum did confirm that the product was pure. The three acetyl groups were confirmed by three singlets between 1.99 and 2.05 ppm, and the presence of only one anomer was confirmed by the one doublet observed at 5.03 ppm. The infrared spectrum confirmed that the azido group had been converted to an amine group and that the amine group was present as an ammonium salt. The N H stretch at 3128-2761 cm\"1 and the N H bend at 1608 cm\"1 and 1500 cm\"1 were at lower frequencies than the analogous free primary amine values, as expected for an ammonium salt34. Compounds 3 and 8 were all prepared by methods similar to those for compounds 1, 2, 6, and 7 (Scheme 8.1 and 8.2). As expected, due to the better nucleophilicity of nitrogen compared to that of oxygen, the formation of the amide bond in the synthesis of 3 and 8 proceeded rapidly, with yields similar to those for other ferrocenyl carbohydrate conjugates of this nature19'28. The complexes were purified by either recrystallization from ethanol, or column chromatography, followed by recrystallization. Compound 12 was prepared in a manner similar to that for 11 (Scheme 8.3). The appropriate sugar was reacted with the FcCOOBt in a buffered mixed solvent system. Despite the aqueous reaction conditions, the activated ester was not converted back to the carboxylic acid, but selectively reacted to form an amide in moderate yields; 12 was purified by column chromatography. 192 193 OAc Scheme 8.2. Synthesis of compounds 6-10. 194 O C H 3 Scheme 8.3. Synthesis of compounds 11 and 12. Scheme 8.4. Preparation of sugar derivative 13. All new compounds were fully characterized by elemental analysis, mass spectrometry, and NMR spectroscopy. Elemental analyses of the bulk samples were consistent with the proposed molecular formulae and evince the purity of the compounds. In the case of 12, which was synthesized under aqueous conditions, one water molecule was associated with the compound, typical of unprotected carbohydrate derivatives, even when dried in vacuo for extended periods of time, with or without heating. The mass spectra of the compounds showed the parent ion, [M+] or [M+Na]+, with typically 100% 195 relative intensity. Other peaks in the mass spectra included fragments with protected groups or complete carbohydrate moieties removed from the ferrocene core. Assignments in the 'FI NMR spectra were facilitated by 2D 'if- 1!! COSY experiments, and support the proposed structures. The 'H NMR resonances assigned to the sugar moieties were typically shifted downfield relative to the free sugar 'H NMR resonances, with additional hydrogen resonances observed for the hydrogen atoms associated with the amide groups. The largest shifts were observed for the hydrogen atoms on the sugar carbon directly connected to the amide or ester linkage. For example, in the 'H NMR spectrum of 12, the resonances of the sugar C-6 hydrogen atoms were shifted downfield by 0.41 ppm and 0.52 ppm, while the other hydrogen atoms associated with the sugar shifted < 0.1 ppm. Although the amine hydrogen atoms in 13 were not observed, the amide hydrogen atoms were observed for 3 and 8, likely due to the slower deuterium exchange for the amide versus the primary amine. In contrast, the amide hydrogen atoms in 12 were not observed, as the 'H NMR spectrum of 12 was run in methanol-^, a protic solvent where deuterium easily exchanges with hydrogen atoms such as those in amine or amide groups. Conversely, the 'H NMR spectra of 3 and 8 were run in aprotic solvents; deuterium exchange with the amide protons did not occur (or was significantly slower), and the amide hydrogen resonances were observed. The cyclopentadienyl ring hydrogen 'IT NMR resonances are indicative of both the structure and the symmetry of the compounds. For 3, which is substituted on one Cp ring only, three different resonances are observed. The unsubstituted Cp ring resonance was a singlet between 4.0 ppm and 4.3 ppm. The substituted Cp had resonances for the hydrogen atoms on the carbons adjacent, or a, (between 4.80 ppm and 4.95 ppm) and the 196 hydrogen atoms on the carbons p (between 4.00 ppm and 4.48 ppm) to the carbonyl linker. It is expected in a symmetric compound that the two a hydrogen atoms should give one doublet, and the two P hydrogen atoms should give a doublet of doublets. Due to the stereocentres of the glucose moiety, all hydrogen atoms on the substituted Cp ring are inequivalent, and two sets of the above described resonances are expected. For example in the 'H NMR spectrum of 12, the a hydrogen atoms appeared as two doublets, as predicted for an asymmetric structure. The P hydrogen atoms appeared as a triplet, which may have been an overlapped doublet of doublets. For 8, where both Cp rings are substituted, only one set of resonances for each of the sugar hydrogens and for each of the four hydrogen atoms on the substituted Cp rings was observed. Each of the substituted Cp rings and the respective linked sugar moieties are equivalent. Like the monosubstituted compounds the two a protons of the Cp ring are rendered inequivalent by the stereocenters of the sugar, and appeared as two doublets instead of one. 8.3.2 Solid State Structure of 1, 2, and 12 Solid state structures of the monosubstituted ferrocenoyl-carbohydrate conjugates 1, 2 and 12 were determined by X-ray crystallography and bear many structural similarities (Figures 8.2-8.4 and Table 8.1). (Few crystal structures of ferrocene carbohydrate conjugates have been reported). \" In each of the structures reported herein, the unit cell is monoclinic, containing two molecules in the asymmetric unit. In the structures of 1 and 2, the two cyclopentadienyl rings are eclipsed and very close to parallel. In the structure of 12, the non-substituted cyclopentadienyl ring is rotationally disordered, equally occupying the eclipsed and the staggered orientations, relative to the other 197 cyclopentadienyl ring. The Fe-C bond lengths are similar to those reported elsewhere for ferrocene derivatives 3 5' 3 6' 3 9' 4 0. Slightly shorter Fe-C bond lengths are observed in the non-substituted cyclopentadienyl ring than in the substituted ring, while 2 and 12 have shorter Fe-C bond lengths than does 1. The C-C bonds of the Cp ring are shorter in the non-substituted ring and in the Cp rings of 2. The slight differences in Fe-C and Cp C-C bond lengths can be attributed to the electron withdrawing nature of the amide and thioether linkages, reducing the electron density in the substituted ring with concomitant longer bond lengths. Many ferrocene derviatives substituted by an amide have been Table 8.1. Selected Bond Lengths (A) and Angles (deg) in 1,2 and 12. 1 2 12 Fe-Crnean 2.046 2.039 Fe-Cmean (substituted) 2.050 2.040 2.039 C(7)-C(8) 1.492(3) 1.460(5) 1.473(9) C(7)-0(9)/0(6)/0(l) 1.226(2) 1.217(4) 1.241(7) C(7)-N(l)/S(l) 1.361(2) 1.797(4) 1.353(8) C(8)-C(7)-0(9)/0(6)/0(l) 122.49(18) 124.5(3) 121.0(6) C(8)-C(7)-N(l)/S(l) 114.20(16) 113.2(3) 116.9(5) 0(9)/0(6)/0(l)-C(7)-N(l)/ S(l) 123.32(18) 122.3(3) 122.0(6) Tilt between Cp rings 0.56\u00C2\u00B0 2.76\u00C2\u00B0 198 Figure 8.3. Structure of 2 showing atom labelling scheme (50% thermal ellipsoids). 199 Figure 8.4. Structure of 12 showing atom labelling scheme and the disorder in the unsubstituted Cp ring (50% thermal ellipsoids). structurally analyzed, and the bond lengths and angles are similar to those reported here 3 9' 4 1' 4 2 . The geometry around the carbonyl carbon is distorted trigonal planar, with the C(8)-C(7)-N(l)/S(l) angle being slightly less than, and the other two angles slightly greater than, 120\u00C2\u00B0. As expected, only the |3 anomers were isolated for 1 and 2 and only the a anomer was isolated for 12, which agrees with the single anomeric forms observed in the respective 'H NMR spectra. In the structure of 2, 010 has a relatively large anisotropic thermal ellipsoid, due to the significant motion consistent with an alkyl chain. All structures confirm that the respective carbohydrates are pendant, and do not interact with the iron metal centre. 200 8.3 .3 Biological Studies To show that the compounds would not be toxic to the infected host, the cytotoxicity of each of the ferrocenoyl carbohydrate complexes was studied in two cell lines (Table 8.2). Cytotoxicity was determined via the M T T (3-(4, 5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide) assay. 4 3 A l l compounds showed similar toxicity values to one another, within each cell line. In the human cell line, only two compounds, 2 and 5, had IC50 values in the range of concentrations tested. Other ferrocenyl-carbohydrates have also been reported to show only low levels of toxicity, even at high concentrations. 1 9 ' 3 6 Anti-malarial activity was assessed in three strains of Plasmodium falciparum, two chloroquine sensitive strains and one chloroquine-resistant strain. Several of the compounds showed promising anti-malarial activity (Table 8.2). In the DIO strain, two compounds, 5 and 10, showed moderate to potentially high activity respectively. Compound 5 had an EC50 value of 12.5 u M . Compound 10, which is the disubstituted analog of 5, displayed 78% inhibition at the lowest concentration tested, 6.7 u M . In the 3D7 and K l (chloroquine resistant) strains, 5 showed no activity while 10 showed only moderate activity with EC50 values of 16.4 u M and 15.5 u M in the two strains, respectively. Compound 7, which did not show activity in the DIO strains, was moderately active in the 3D7 and K l strains with EC50 values of 4.9 u M and 6.1 u M , respectively. The difference in the anti-plasmodial activities between the two non-resistant strains, although mainly due to the differences in the strains themselves, may be partially attributed to the differences in the assay methods. For the non-resistant DIO strain, parasite viability was determined by measuring the enzymatic activity of parasite 201 lactate dehydrogenase (pLDH). For the 3D7 and the chloroquine resistant K l strains, incorporation of tritium labeled hypoxanthine was employed to measure parasite growth. The latter is considered to be the more sensitive of the two assays.44 Table 8.2. Cytotoxicity and anti-plasmodial activity of ferrocenoyl-carbohydrate complexes. Compound Cytotoxicity IC50 (pM) Anti-plasmodial activity EC50 (pM) Mouse Lymphoma (L1210) Human Breast Cancer (HTB-129) P. Falciparum Strain D10 3D7 K l (chloroquine resistant) 1 17 >321 32 33 >54 2 25 87 >43 33 40 3 55 >325 >47 >56 52 4 ND ND >33 ND ND 5 ND 180 12.5 >54 >54 6 15.5 >115 ND >32 >32 7 13.5 ND >26 4.9 6.1 9 ND ND >25 ND ND 10 ND >121 <6.7 16.4 15.5 11 67 >468 56 >77 61 12 ND >474 ND >74 >74 ND = not determined 202 A s most of the compounds showed similar toxicity and anti-plasmodial activity, only a minimal structure activity relationship can be determined. Notably, for the active compounds, those substituted on both Cp rings were more active than their monosubstituted analogs. For example, compound 7 was more than 6 fold more active than its monosubstituted analogue, 2, in both the 3D7 and K l strains. The most active compounds all contained acetyl protecting groups on the carbohydrate moiety, and were linked to the ferrocene via either a thioester or an ester, rather than an amide functional group. Although none of the compounds proved to be highly effective inhibitors of P. falciparum in vitro, carbohydrate functionalization is an attractive approach to novel anti-malarials. Addit ion of the carbohydrate substituent(s) yielded ferrocene compounds with low toxicity, even for the compounds with the best anti-malarial activity. This suggests that the cytotoxic and anti-malarial activities are not correlated, and that the concentrations needed for anti-malarial activity would likely be non-toxic towards the host. A s well , the anti-malarial activity is similar in both chloroquine resistant and non-resistant strains. A s the malaria parasite sequesters many nutrients, such as hemoglobin and glucose, it has developed a higher affinity for these substrates than the host. 2 1 Ce l l permeability for carbohydrates is increased and the rate of glucose consumption is elevated 50 to 100 fold in infected erythrocytes. 2 1 ' 2 2 This higher affinity for glucose may result in targeting of the parasite by glucose derivatives, such as ferrocenyl-carbohydrate conjugates, in vivo. Targeting would result in lower dose amounts and higher efficacy than suggested by the in vitro studies. 203 8.4 Conclusions Ferrocenyl -carbohydrate conjugates were easi ly synthesized v i a amide or a thioester l inkages between ferrocene carbony l ch lor ide and a respect ive amine or th io l carbohydrate der ivat ive. The ester i f icat ion o f a carbohydrate h y d r o x y l required more aggressive condi t ions. The structures o f the compounds were character ized by the usual methods, and showed l o w levels o f symmetry due to the carbohydrate stereocentres. Three o f the compounds were also analyzed in the so l id state by X - r a y crysta l lography, w h i c h con f i rmed the pendant nature o f the carbohydrate substituents, and the anomer ic conformat ion o f the carbohydrates. The compounds were non- tox ic at h i gh levels i n the human breast cancer H T B 1 2 9 ce l l l ine, wh i le some o f the compounds showed moderate ant i -malar ia l act iv i ty in vitro. Further exper iments are needed to determine i f these complexes have h igher in vivo act iv i ty attributable to targeting o f the parasites elevated leve l o f g lucose consumpt ion . 8.5 References 1. At teke, C ; N d o n g , J . M . M . ; A u b o u y , A . ; M a c i e j e w s k i , L . ; B roca rd , J . ; L e b i b i , J . ; De lo ron , ?.,J. Antimicrob. Chem. 2003, 57 , 1021-1024. 2. Beage ly , P . ; B l a c k i e , M . A . L . ; Ch iba le , K . ; C l a r k s o n , C ; M e i j b o o m , R. ; M o s s , J . R.; Smi th , P. J . ; S u , H . , Dalton Trans. 2003, 3046-3051. 204 3. Biot, C ; Delhaes, L.; Abessolo, H.; Dormarle, O.; Maciejewski, L. A.; Mortuaire, M.; Delcourt, P.; Deloron, P.; Camus, D.; Dive, D.; Brocard, J. S., J. Organomet. Chem. 1999, 589, 59-65. 4. Biot, C ; Delhaes, L.; Maciejewski, L. A.; Mortuaire, M . ; Camus, D.; Dive, D.; Brocard, J. S., Eur. J. Med. Chem. 2000, 35, 707-714. 5. Biot, C ; Dessolin, J.; Ricard, I.; Dive, D., J. Organomet. Chem. 2004, 689, 4678-4682. 6. Biot, C ; Glorian, G.; Maciejewski, L. A.; Brocard, J. S., J. Med. Chem. 1997, 40, 3715-3718. 7. Delhaes, L.; Abessolo, H.; Biot, C ; Berry, L.; Delcourt, P.; Maciejewski, L.; Brocard, J.; Camus, D.; Dive, D., Parasitol. Res. 2001, 87, 239-244. 8. Delhaes, L.; Biot, C ; Berry, L.; Maciejewski, L. A.; Camus, D.; Brocard, J. S.; Dive, D., Bioorg. Med. Chem. 2000, 8, 2739-2745. 9. Ocheskey, J. A.; Harpstrite, S. E.; Oksman, A.; Goldberg, D. E.; Sharma, V., Chem. Comm. 2005, 1622-1624. 10. Beagley, P.; Blackie, M . A. L.; Chibale, K.; Clarkson, C ; Moss, J. R.; Smith, P. J., J. Chem. Soc., Dalton Trans. 2002, 4426-4433. 11. Sanchez-Delgado, R. A.; Navarro, M. ; Perez, H.; Urbina, J. A., J. Med. Chem. 1996, 39, 1095-1099. 12. Navarro, M. ; Perez, H.; Sanchez-Delgado, R. A., J. Med. Chem. 1997, 40, 1937-1939. 13. Navarro, M . ; Vasquez, F.; Sanchez-Delgado, R. A.; Perez, H.; Sinou, V.; Schrevel, J., J. Med. Chem. 2004, 47, 5204-5209. 205 14. Hubel, R.; Polborn, K.; Knizek, J.; Noth, H.; Beck, W., Z Anorg. Allg. Chem. 2000, 626, 1701-1708. 15. Domarle, O.; Blampain, G.; Agnaniet, H.; Nzadiyabe, T.; Lebibi, J.; Brocard, J.; Maciejewski, L.; Biot, C.; Georges, A. J.; Millet, P., Antimicrob. Agents Chemother. 1998, 42, 540-544. 16. Sharma, V.; Piwnica-Worms, D., Chem. Rev. 1999, 99, 2545-2560. 17. Harpstrite, S. A.; Beatty, A. A.; Collins, S. D.; Oksman, A.; Goldberg, D. E.; Sharma, V., Inorg. Chem. 2003, 42, 2294-2300. 18. Sharma, V.; Beatty, A.; Goldberg, D. E.; Piwnica-Worms, D., J. Chem. Soc, Chem. Comm. 1997, 2223-2224. 19. Itoh, T.; Shirakami, S.; Ishida, N.; Yamashita, Y.; Yoshida, T.; Kim, H.-S.; Wataya, Y., Bioorg. Med. Chem. Lett. 2000,10, 1657-1659. 20. Fitch, C. D., Life Sci. 2004, 74, 1957-1972. 21. Roth, E. , Blood Cells 1990,16, 453-466. 22. Sherman, I. W., The Biochemistry of Malaria-An Overview. Taylor and Francis: London, 1991. 23. Joet, T.; Eckstein-Ludwig, U.; Morin, C ; Krishna, S., Proc. Natl. Acad. Sci. USA 2003,100, 7476-7479. 24. Paulson, P. L.; Watts, W. E. , J. Chem. Soc. 1963, 2990-2995. 25. Kraatz, H.-B.; Lusztyk, J.; Enright, G. D., Inorg. Chem. 1997, 36, 2400-2405. 26. Cramer, F. D., Methods Carbohydrate Chem. 1963, 2, 243. 27. Green, D. E.; Ferreira, C. L.; Stick, R. V.; Patrick, B. O.; Adam, M . J.; Orvig, C , Bioconjugate Chem. 2005,16, 1597-1609. 28. Adam, M . J.; Hall, L. D., Can. J. Chem. 1980, 58, 1188-1197. 206 29. Chahma, M . ; Lee, J. S.; Kraatz, H.-B., J. Organomet. Chem. 2002, 648, 81-86. 30. Ewart, C. B. University of British Columbia, Vancouver, BC, 2003. 31. Trager, W.; Jensen, J. B., Science 1976,193, 635-637. 32. Makler, M . T.; Ries, J. M. ; Williams, J. A.; Bancroft, J. E.; Piper, R. C ; Gibbins, B. L.; Hinrichs, D. J., Am. J. Trop. Med. Hyg. 1993, 48, 739-741. 33. Desjardins, R. E.; Canfield, C. J.; Haynes, J. D.; Chulay, J. D., Antimicrob. Agents Chemother. 1979,16, 710-718. 34. Williams, D. H.; Fleming, I., Spectroscopic Methods in Organic Chemistry. McGraw-Hill: London, 1995. 35. Albinati, A.; Pregosin, P. S.; Wick, K., Organometallics 1996,15, 2419-2421. 36. Hartinger, C. G.; Nazarov, A. A.; Arton, V. B.; Giester, G.; Jakupec, M. ; Galanski, M. ; Keppler, B. K., New J. Chem. 2002, 26, 671-673. 37. Kerr, J. L.; Landells, J. S.; Larsen, D. S.; Robinson, B. H.; Simpson, J., J. Chem. Soc, Dalton Trans. 2000, 1411-1417. 38. Nazarov, A. A.; Christian G. Hartinger; Arion, V. B.; Giester, G.; Keppler, B. K., Tetrahedron 2002, 58, 8489-8492. 39. Snegur, L. V.; Simenel, A. A.; Nekrasov, Y. S.; Morzova, E. A.; Starikova, Z. A.; Peregudova, S. M . ; Kuzmenko, Y. V.; Babin, V. N.; Ostrovskaya, L. A., J. Organomet. Chem. 2004, 689, 2473-2479. 40. Xu, Y.; Saweczko, P.; Kraatz, H.-B., J. Organomet. Chem. 2001, 637-639, 335-342. 41. Appoh, F. E.; Sutherland, T. C ; Kraatz, H.-B., J. Organomet. Chem. 2004, 689, 4669-4677. 42. Saweczko, P.; Enright, G. D.; Kraatz, H.-B., Inorg. Chem. 2001, 40, 4409-4419. 207 43. Mosmann, T., Immunol. Methods 1983, 65, 55-63. 44. Noedl, H . ; Wongrechanalai, C ; Wemsdorfer, W . H . , Trends Parasitol. 2003,19, 175-181. 208 Chapter 9 Conclusions and Future Work 9.1 Re and Tc Carbohydrate Conjugates for Molecular Imaging and Radiotherapy 9.1.1 Macroscopic Scale Synthesis of Re Carbohydrate Conjugates In Chapter 2-4 carbohydrate conjugates of the [M(CO)3] + core were investigated as potential radiopharmaceuticals, utilizing bidentate, tridentate, and cyclopentadienyl ligands, respectively. Each ligand type had different advantages. The bidentate 3-hydroxy-4-pyridinone pro-ligands afforded neutral complexes, while the tridentate dipicolylamine ligands, gave cationic complexes but with a saturated coordination sphere. Neutral complexes may show better biological properties (mimicking a neutral molecule with a neutral complex), while coordinatively saturated complexes will have higher stability in vivo. The cyclopentadienyl ligands both saturate the coordination sphere and balance the metal charge. As well, they are compact and low molecular weight; all of these properties are beneficial towards maintaining the biological recognition of a small molecule like glucose. The compounds were prepared on the macroscopic scale with Re, and were all thoroughly characterized. Each compound chelated the metal in the expected fashion, while the carbohydrate remained pendant, except ReL 6(CO)3 (L 6 = N-(2'-oxybenzyl)-2-209 amino-2-deoxy-D-glucopyranose), which was fortuitously tridentate due to the carbohydrate binding to the metal centre. In order for the carbohydrate to be recognized by enzymes and proteins in vivo, the amount of modification to the carbohydrate should be minimized, thus ReL 6(CO)3 is the least promising candidate for imaging glucose metabolism. 9.1.2 Radiolabelling and Stability The tridentate and bidentate ligands and pro-ligands were labelled with 9 9 m T c and Re in high radiochemical yields via a simple, fast process applicable to radiopharmacy preparation. Notably, the 3-hydroxy-4-pyridinone pro-ligands were labelled with 1 8 6 Re in a one vial preparation with unprecedented radiochemical yields for the [ 1 8 6Re(CO) 3] + core. The 9 9 m T c complexes of the tridentate dipicolylamine ligands showed higher stability than the bidentate ligands in challenge experiments with potential metal binding 1 2 amino acids. ' As well, it has been shown that bidentate ligand complexes of the [M(CO)3]+ core become protein bound in vivo, most likely due to the open coordination site.3'4 The higher stability of the dipicolylamine ligand complexes suggested that these were the most promising candidates for further evaluation. The utility of the bidentate pyridinone ligand complexes warrants further investigation, but higher stability is necessary. To increase the stability of these complexes the sixth coordination site must be masked. A potential answer to this problem is forming 2+1 mixed ligand complexes,4\"6 where a second monodentate ligand is added to saturate the coordination sphere (Figure 9.1); this second ligand would also 210 OC I T O CO M = Re, Tc L = monodentate ligand Examples of monodentate ligands H - N \ // \" N R C = N R \u00E2\u0080\u0094 N H 2 R - O H R\u00E2\u0080\u0094SH Figure 9.1. 2+1 M i x e d ligand complexes of the [ M ( C O ) 3 ] + core with a bidentate pyridinone ligand and examples of monodentate ligands. allow for systematic modification of the properties of the complex. Possible monodentate ligands include imidazole derivatives, phosphines, amines or isocyanides to give neutral complexes, and thiols or alcohols to give negatively charged complexes. The monodentate ligand can be functionalized to moderate the lipophilicity of the complex, which could be used to optimize biodistribution properties, such as blood and excretory organ clearance. Preparation of the glucosamine conjugate of 9 9 m T c C p ( C O ) 3 (Cp = cyclopentadienyl) required a more complicated synthesis. The complex was prepared using a ferrocenyl-acetylated glucosamine conjugate in the single ligand transfer (SLT) reaction, followed by the removal of the acetyl protect groups. The ferrocene based starting material and other ferrocene containing side products need to be removed by purification before the complex can be used in animals or humans. Because the synthesis and purification are not optimal for radiopharmaceutical preparation or animal/human administration, an 211 alternative synthesis could be considered, whereby a cyclopentadienyl based ligand is used to directly coordinate the [M(CO) 3] + core (Scheme 9.1).7 For this route to be M = Re, Tc Scheme 9.1. Alternative synthesis of conjugates of the CpM(CO) 3 core (Cp = cyclopentadienyl, M = Re, Tc). feasible, the cyclopentdienyl derivative would need to be stable under aqueous conditions. It has been shown that this is possible by using cyclopentadiene derivatives substituted with a carbonyl functionality; upon deprotonation the keto-enol tautomerisation stabilizes the negative charge.7 The main obstacle of this route is the synthesis of a sodium cyclopentadienyl glucosamine conjugate. This alternative route to the glucosamine conjugate of MCp(CO) 3 may also be applicable to 1 8 6 / 1 8 8 R e labelling, which was not possible with the SLT route. 9.1.3 Biological Studies Each of the Re complexes was tested as a substrate and/or inhibitor of hexokinase; an enzyme instrumental in the imaging properties of F D G . 8 While none of the compounds were substrates, one of the compounds, glucosamine conjugated cyclopentadienyl tricarbonylrhenium(I) (2a), was a high affinity competitive inhibitor. The possible COv 212 interaction of this compound with the active site of the enzyme is promising, but substrate activity is needed to imitate the high uptake and retention of FDG in target tissues. Hexokinase has an open conformation when glucose or F D G initially bind to the enzyme active site, which changes to a closed conformation before phosphorylation of the substrate.9'10 It is likely that 2a binds the active site of the enzyme in the open conformation, but due to the bulk of the cyclopentadienyltricarbonylrhenium(I) core the enzyme conformation change necessary for phosphorylation of the glucosamine is impeded.10 Increasing the distance between the carbohydrate moiety and the metal core with a non-bulky alkyl chain could alleviate the impediment of the conformation change, and produce a complex that is a substrate rather than just an inhibitor. Molecular docking studies of hexokinase with C2 derivatized carbohydrate conjugates of the [M(CO)3]+ core, suggest that an alkyl chain of 7 carbons would be sufficient to allow closure of the enzyme.11 The insertion of a minimum 7 carbon alkyl chain between the glucose derivative and the metal chelate in the ligands and complexes discussed in this thesis (especially those with C2 functionalization of the glucose) may result in substrate interactions with hexokinase. In vivo biological evaluation of the dipicolylamine tridentate ligand 9 9 m T c complexes, [ 9 9 mTcL 7\" 9(CO)3] +, included biodistribution and planar imaging studies. The biodistribution studies suggested that all the compounds were being cleared through the excretion organs, such as the kidney and liver. Further biodistribution studies at longer time points would confirm the route of clearance of the compounds, and determine the time required for complete clearance. Studies with the pendant glucosamine complex, [ 9 9 m TcL 9 (CO)3] + , suggested activity observed in the tumour was due to blood in the 213 tumour vasculature, and not actual uptake into the tumour cells. The biodistribution of the CI linked glucose derivatives, [ 9 9 m TcL 7 (CO) 3 ] + and [ 9 9 m TcL 8 (CO) 3 ] + , showed higher uptake in the tumour than the blood after 2 h. Again, biodistribution studies at longer time points would show whether the tumounblood ratios continue to increase. Finally, imaging studies with [ 9 9 m TcL 7 (CO) 3 ] + and [ 9 9 m TcL 8 (CO) 3 ] + are needed to confirm that the tumour can be visualized with these compounds. Providing the results are positive, further studies to fully delineate the mechanism of uptake and retention will be needed. 9.2 Carbohydrate Conjugates of Other Isotopes for Use in Nuclear Medicine Carbohydrate conjugates of several other isotopes of relevance to molecular imaging and radiotherapy were examined in Chapters 5-7. The carbohydrate bearing 3-hydroxy-4-pyridinone pro-ligands formed neutral bisligand complexes with Cu and Co, and trisligand complexes with Ga and In. These complexes formed readily both on the macroscopic scale and in radiolabelling with 6 7 G a and 6 4 C u . As well, tetrahydrosalen pro-ligands with pendant carbohydrates were readily labelled with 6 4 C u in high radiochemical yields. Further radiolabelling with 1 1 ' in and 5 5 Co is planned. Stability challenges with the 6 7 G a and 6 4 C u complexes showed them to be unstable in the presence of excess cysteine and histidine. It will be interesting to determine if the l u I n and 5 5 Co complexes show similar instability. The goal of making useful carbohydrate conjugates of these isotopes requires complexes that have higher kinetic inertness than the ones studied in this work. To prepare more inert complexes cross bridged cyclam ligands could be functionalized with glucose derivatives for chelating Cu, while DTPA glucose 214 conjugates could be used to chelate Ga and In. Since the instability observed in the complexes is kinetic in nature, d 6 octahedral Co(III) species may be more stable, and thus of greater interest, than the d 7 Co(II) species.13 9.3 Other Potential Applications in Nuclear Medicine Although developing Re and Tc analogues of FDG was a central goal to this work, there are greater applications for conjugates of carbohydrates. Using other carbohydrates, polymeric carbohydrates, and glycoconjugates with the systems studied in this thesis opens a larger field of potential imaging agents. These types of complexes may find utility in the imaging of tumour specific lectins or glycopeptide receptors, or in investigating other biological processes where carbohydrates and carbohydrate conjugates play a central role.14\"17 9.4 Ferrocenyl-carbohydrate Conjugates as Potential Anti-malarials Targeting the high glucose utilization of the parasite that causes malaria is a novel approach to developing anti-malarials.18'19 Other ferrocene compounds have been previously researched, and show promising anti-malarial activity, especially in resistant strains of the parasite.20\"23 Numerous ferrocenyl-carbohydrate conjugates were prepared and examined as potential malaria therapeutics. Several of the complexes showed moderate anti-malarial activity, and no difference in activity was observed between the chloroquine resistant and non-resistant strains. Cytotoxicity studies showed the 2 1 5 compounds to be non-toxic in a human cancer cell line, suggesting that the complexes would not be harmful to the host and that the observed anti-malarial activity was not due to general toxicity of the compounds. Although none of compounds were as active as chloroquine, or other currently used anti-malarials, in vitro, the hypothesis that ferrocenyl-carbohydrate conjugates will selectively target infected over normal erythrocytes could result in higher efficacy in vivo, thus testing of the complexes in animal malaria models is warranted. Conjugation of other active anti-malarials, such as chloroquine, .to the ferrocenyl-carbohydrate conjugates could produce compounds with higher activity and efficacy. 9.5 References 1. Storr, T.; Fisher, C. L.; Mikata, Y.; Yano, S.; Adam, M . J.; Orvig, C , Dalton Trans. 2005, 654-655. 2. Storr, T.; Obata, M . ; Fisher, C. L.; Bayly, S. R.; Green, D. E.; Brudzinska, I.; Mikata, Y.; Patrick, B. O.; Adam, M . J.; Yano, S.; Orvig, C , Chem. Eur. J. 2005,11, 195-203. 3. Schibli, R.; La Bella, R.; Alberto, R.; Garcia-Garayoa, E.; Ortner, K.; Abram, U.; Schubiger, P. A., Bioconjugate Chem. 2000,11, 345-351. 4. Albinati, A.; Pregosin, P. S.; Wick, K., Organometallics 1996,15, 2419-2421. 5. Mundwiler, S.; Kiindig, M. ; Ortner, K.; Alberto, R., Dalton Trans. 2004, 1320-1328. 216 6. Gorshkov, N. I.; Lumpov, A. A.; Miroslavov, A. E.; Suglobov, D. N., Radiochemistry 2 0 0 5 , 47, 45-49. 7. Wald, J.; Alberto, R.; Ortner, K.; Candreia, L. , Angew. Chem. Int. Ed. 2 0 0 1 , 40, 3062-3066. 8. Beuthien-Baumann, B.; Hamacher, K.; Oberdorfer, F.; Steinbach, J., Carbohydrate Res. 2 0 0 0 , 327, 1107-1118. 9. Kuser, P. R.; Krauchenco, S.; Antunes, O. A. C ; Polikarpov, I., J. Biol. Chem. 1999,275,20814-20821. 10. Ohning, G. V.; Neet, K. E. , Biochemistry 1 9 8 3 , 22, 2986-2995. 11. Schibli, R.; Dumas, C ; Petrig, J.; Spadola, L. ; Scapozza; Garcia-Garayoa, E. ; Schubiger, P. A., Bioconjugate Chem. 2005,16, 105-112. 12. Boswell, C. A.; Sun, X.; Niu, W.; Weismann, G. R.; Wong, E. H.; Rheingold, A. L.; Anderson, C. J., J. Med. Chem. 2 0 0 4 , 47, 1465-1474. 13. Miessler, G. L.; Tarr, D. L. , Inorganic Chemistry. 2nd ed.; Prentice Hall, Inc.: Upper Saddle River, NJ, 1999. 14. Galanina, O.; Feofanov, A.; Tuzikov, A. B.; Rapoport, E.; Crocker, P. R.; Grichine, A.; Egret-Charlier, M. ; Vigny, P.; Le Pendu, J.; Bovin, N. V. , Spectrochim. Acta Part A 2 0 0 1 , 57, 2285-2296. 15. Andre, J. P.; Geraldes, C. F. G. C ; NMartins, J. A.; Merbach, A. E.; Prata, M . I. M.; Santos, A. C ; de Lima, J. J. P.; Toth, E. , Chem. Eur. J. 2004,10 , 5804-5816. 16. Haubner, R.; Werster, H.-J.; Weber, W. A.; Mang, C ; Ziegler, S. I.; Goodman, S. L.; Senekowitsch-Schmidtke, R.; Kessler, H.; Schwaiger, M . , Cancer Res. 2 0 0 1 , 61, 1781-1785. 217 17. Paschkunova-Martic, I.; Kremser, C ; Mistlberger, K.; Shcherbakova, N.; Dietrich, H.; Talasz, H.; Zou, Y.; Hugl, B.; Galanski, M . ; Solder, E.; Pfaller, K.; Holiner, I.; Buchberger, W.; Keppler, B.; Debbage, P., Histochem. Cell Biol 2005,123, 283-301. 18. Krishna, S.; Eckstein-Ludwig, U.; Joet, T.; Uhlemann, A . - C ; Morin, C ; Webb, R.; Woodrow, C. K., J. F. J.; Kremsner, P. G., Int. J. Parasitol. 2002, 32, 1567-1573. 19. Roth, E. , Blood Cells 1990,16, 453-466. 20. Beagely, P.; Blackie, M . A. L.; Chibale, K.; Clarkson, C ; Meijboom, R.; Moss, J. R.; Smith, P. J.; Su, H. , Dalton Trans. 2003, 3046-3051. 21. Biot, C ; Delhaes, L.; Abessolo, H.; Dormarle, O.; Maciejewski, L. A.; Mortuaire, M. ; Delcourt, P.; Deloron, P.; Camus, D.; Dive, D.; Brocard, J. S., J. Organomet. Chem. 1999, 589, 59-65. 22. Delhaes, L.; Abessolo, H.; Biot, C ; Berry, L.; Delcourt, P.; Maciejewski, L.; Brocard, J.; Camus, D.; Dive, D., Parasitol. Res. 2001, 87, 239-244. 23. Itoh, T.; Shirakami, S.; Ishida, N.; Yamashita, Y.; Yoshida, T.; Kim, H.-S.; Wataya, Y., Bioorg. Med. Chem. Lett. 2000, 10, 1657-1659. 218 APPENDIX Table A l . Crystallographic data for H L 5 (Chapter 2) and 2a (Chapter 4). Crystal Data H L 2a Formula Formula Weight Crystal System Crystal Size [mm] Space Group Lattice Parameters V[A3] z Dcaicd.tg/cm3] u(MoKa) [cm-1] Rl(Fo) wR2(F02) Goodness of Fit Indicator C1.4H24N2O10 380.35 monoclinic 0.250x0.100x0.020 P2, a = 4.6973(2) A b = 25.5366(18) A c = 7.1749(5) A a = 90\u00C2\u00B0 P = 93.473(2)\u00C2\u00B0 y = 90\u00C2\u00B0 859.07(9) 2 1.470 1.26 0.0441 0.0894 1.007 C 1 5 H 1 6 N 0 9 R e 540.49 monoclinic 0.10x0.05x0.05 P2i a = 7.110(l) A b = 8.926(2) A c= 13.823(2) A a = 90\u00C2\u00B0 P = 93.57(1)\u00C2\u00B0 y = 90\u00C2\u00B0 575.5(3) 2 2.050 69.89 0.023 0.037 0.77 219 Table A2. Crystallographic data for 1, 2, and 12 (Chapter 8). Crystal Data 1 2 12 Formula C25H29NOioFe C25H28SOioFe C i 9 H 2 7 N 0 7 F e Formula Weight 559.34 576.38 437.27 Crystal System monoclinic monoclinic monoclinic Space Group P2, P2i PI Lattice Parameters a= 10.6731(19) A a= 10.7128(8) A a = 6.1030(10) A b= 10.55961(16) A b = 7.8560(5) A b = 8.473(2) A c= 11.425(2) A c= 15.7821(11) A c= 19.341(5) A a = 90\u00C2\u00B0 a = 90\u00C2\u00B0 a = 90\u00C2\u00B0 p = 100.666(6)\u00C2\u00B0 P = 97.001(3)\u00C2\u00B0 P = 97.24(2)\u00C2\u00B0 y = 90\u00C2\u00B0 y = 90\u00C2\u00B0 Y = 90\u00C2\u00B0 V[A3] 1265.4(4) 1318.31(16) 992.2(4) z 2 2 2 Dcaicd.[g/cm3] 1.468 1.452 1.464 p(MoKa) [cm\"1] 6.54 7.06 8.01 Rl(Fo) . 0.0313 0.0709 0.0554 wR2(F02) 0.0674 0.0809 0.1053 Goodness of Fit 1.003 1.012 0.984 Indicator 220 "@en . "Thesis/Dissertation"@en . "2006-11"@en . "10.14288/1.0059324"@en . "eng"@en . "Chemistry"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use."@en . "Graduate"@en . "Metal carbohydrate conjugates for diagnostic and therapeutic applications"@en . "Text"@en . "http://hdl.handle.net/2429/18469"@en .