@prefix vivo: . @prefix edm: . @prefix ns0: . @prefix dcterms: . @prefix skos: . vivo:departmentOrSchool "Science, Faculty of"@en, "Chemistry, Department of"@en ; edm:dataProvider "DSpace"@en ; ns0:degreeCampus "UBCV"@en ; dcterms:creator "Wang, Xiaozhu"@en ; dcterms:issued "2019-04-29T15:40:29Z"@en, "2019"@en ; vivo:relatedDegree "Doctor of Philosophy - PhD"@en ; ns0:degreeGrantor "University of British Columbia"@en ; dcterms:description "This thesis presents studies on a new family of oxine (8-hydroxyquinoline) based acyclic chelators for application to radiometals (e.g., ⁶⁴Cu, ⁶⁷/⁶⁸Ga, ¹¹¹In) in nuclear medicine. Picolinic acid-based chelators (“pa” family) are reported as excellent radiometal chelators by our group’s previous study. Further development leads to the next generation chelators – the “ox” family (8-hydroxyquinoline). H₂hox showed a marked improvement from its “pa” counterpart, H₂dedpa, including an easy preparation, single complex species in a broad pH range (1-11) and high log KML (34.4) and pM (28.3) values. H₂hox showed fast (5 minutes) and quantitative ⁶⁸Ga labelling at room temperature with a concentration as low as 10-⁷ M and obtained a high molar activity. Excellent in vitro and in vivo stability was confirmed with plasma challenge experiments and dynamic PET imaging. The chelation enhanced fluorescence emission property was used directly to investigate the cellular distribution of [Ga(hox)]⁺ and showed the potential for dual channel imaging. Expanding from the lead chelator, H₂CHXhox was then synthesized by incorporating a cyclohexane (CHX) in the backbone to pre-organise the chelator and showed a great improvement on the kinetic inertness and thermodynamic stability. H₂C3hox was prepared by adding one more carbon to the backbone of H₂hox to obtain a 6-membered chelate ring, in order to investigate the preference of ring size on metal ion radii. H₂C3hox showed a decrease in solution stability, thought to be due to adopting a less stable conformation in the 6-membered chelate ring. H₄octox was designed with increased denticity (N₄O₄) vs H₂hox (N₄O₂) for larger metal ions and showed fast and stable chelation with metal ions (Y³⁺, In³⁺, La³⁺, Lu³⁺and Gd³⁺) in solution. Its in vitro stability with In³⁺ and Y³⁺ was proved using plasma and Fe³⁺ challenge experiments, and in vivo stability was confirmed with ¹¹¹In SPECT imaging. The 60-fold fluorescence emission increase when complexed with Y³⁺ can also be applied in probe design or bi-modal imaging. These studies have indicated that the new “ox” family of chelators is an excellent and useful platform in the development of radiometal-based pharmaceuticals."@en ; edm:aggregatedCHO "https://circle.library.ubc.ca/rest/handle/2429/69977?expand=metadata"@en ; skos:note " New Chelators for Radiopharmaceutical Chemistry by Xiaozhu Wang B.S., Capital Medical University, 2010 M.S., King Abdullah University of Science and Technology, 2011 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Chemistry) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) April 2019 © Xiaozhu Wang, 2019 ii The following individuals certify that they have read, and recommend to the Faculty of Graduate and Postdoctoral Studies for acceptance, the dissertation entitled: New Chelators for Radiopharmaceutical Chemistry submitted by Xiaozhu Wang in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY in Chemistry Examining Committee: Chris Orvig Supervisor Urs Häfeli Supervisory Committee Member Supervisory Committee Member Pierre Kennepohl University Examiner Adam Frankel University Examiner Additional Supervisory Committee Members: Supervisory Committee Member Supervisory Committee Member iii Abstract This thesis presents studies on a new family of oxine (8-hydroxyquinoline) based acyclic chelators for application to radiometals (e.g., 64Cu, 67/68Ga, 111In) in nuclear medicine. Picolinic acid-based chelators (“pa” family) are reported as excellent radiometal chelators by our group’s previous study. Further development leads to the next generation chelators – the “ox” family (8-hydroxyquinoline). H2hox showed a marked improvement from its “pa” counterpart, H2dedpa, including an easy preparation, single complex species in a broad pH range (1-11) and high log KML (34.4) and pM (28.3) values. H2hox showed fast (5 minutes) and quantitative 68Ga labelling at room temperature with a concentration as low as 10-7 M and obtained a high molar activity. Excellent in vitro and in vivo stability was confirmed with plasma challenge experiments and dynamic PET imaging. The chelation enhanced fluorescence emission property was used directly to investigate the cellular distribution of [Ga(hox)]+ and showed the potential for dual channel imaging. Expanding from the lead chelator, H2CHXhox was then synthesized by incorporating a cyclohexane (CHX) in the backbone to pre-organise the chelator and showed a great improvement on the kinetic inertness and thermodynamic stability. H2C3hox was prepared by adding one more carbon to the backbone of H2hox to obtain a 6-membered chelate ring, in order to investigate the preference of ring size on metal ion radii. H2C3hox showed a decrease in solution stability, thought to be due to adopting a less stable conformation in the 6-membered chelate ring. H4octox was designed with increased denticity (N4O4) vs H2hox (N4O2) for larger metal ions and showed fast and stable chelation with metal ions (Y3+, In3+, La3+, Lu3+and Gd3+) in solution. Its in vitro stability with In3+ and Y3+ was proved using plasma and Fe3+ challenge experiments, and in vivo stability iv was confirmed with 111In SPECT imaging. The 60-fold fluorescence emission increase when complexed with Y3+ can also be applied in probe design or bi-modal imaging. These studies have indicated that the new “ox” family of chelators is an excellent and useful platform in the development of radiometal-based pharmaceuticals. v Lay Summary Nuclear medicine is an important field of medicine, applying radioactive substances in the diagnostic imaging and therapy. Chelators are integral parts of radiometal based pharmaceuticals. In this thesis, a new group of acyclic bimodal chelators based on “oxine” arms has been explored and showed a marked improvement from its “pa” counterpart and potential for toolkit radiopharmaceuticals (“shake and shoot”) in clinical application. vi Preface Chapter 2 contains an adaptation of published work and is reproduced in part from Jaraquemada-Pelaez, M. d. G.; Wang, X.; Clough, T. J.; Cao, Y.; Choudhary, N.; Emler, K.; Patrick, B.O. and Orvig, C. H4octapa: synthesis, solution equilibria and complexes with useful radiopharmaceutical metal ions. Dalton Trans. 2017, 46, 14647-14658. Copyright 2017 The Royal Society of Chemistry. Xiaozhu Wang performed the synthesis with assistance from Thomas J. Clough and Kirsten Emler. Solution studies were performed by Dr. María de Guadalupe Jaraquemada-Peláez; DFT calculations were performed by Dr. Yang Cao; X-ray crystallography was solved by Neha Choudhary and Dr. Brian O. Patrick. This project was supervised by Dr. Chris Orvig. The manuscript was written by Xiaozhu Wang and Dr. María de Guadalupe Jaraquemada-Peláez. Chapter 3 is an adaptation of published work, and is reproduced in part from Wang, X.; Jaraquemada-Pelaez, M. d. G.; Cao, Y.; Pan, J.; Lin K. S.; Patrick, B. O.; Orvig, C. H2hox: Dual Channel Oxine-Derived Acyclic Hexadentate Chelating Ligand for 68Ga Radiopharmaceuticals. Inorg. Chem. 2019, 58, 2275-2285. Copyright 2018 American Chemical Society. Xiaozhu Wang designed the ligand and performed the synthesis, characterization and cell imaging studies. Solution studies were performed by Dr. María de Guadalupe Jaraquemada-Peláez; DFT calculations were performed by Dr. Yang Cao; radiolabeling and PET/CT imaging studies were performed by Dr. Jinhe Pan (under the supervision of Dr. Kuoshyan Lin) at the BC vii Cancer Research Centre; X-ray crystallography was solved by Dr. Brian O. Patrick. This project was supervised by Dr. Chris Orvig and the manuscript was written by Xiaozhu Wang. Chapter 4 is an adaptation of a manuscript in preparation: Wang, X.; Jaraquemada-Pelaez, M. d. G.; Kostelnik, T. I.; Cao, Y.; Rodríguez-Rodríguez, C.; Pan, J.; Lin, K. S.; Patrick, B. O.; Saatchi, K.; Häfeli, U. O.; Orvig, C. H2CHXhox: cyclohexane reinforced chelating Ligand for Ga3+. Xiaozhu Wang designed the ligand and performed the synthesis, characterization and cell imaging studies. Solution studies were performed by Dr. María de Guadalupe Jaraquemada-Peláez; DFT calculations were performed by Dr. Yang Cao; radiolabeling and PET/CT imaging studies were performed by Dr. Jinhe Pan (under the supervision of Dr. Kuoshyan Lin) at the BC Cancer Research Centre; X-ray crystallography was solved by Dr. Brian O. Patrick; SPECT/CT imaging studies were performed by Dr. Cristina Rodríguez-Rodríguez and Dr. Katayoun Saatchi at the Centre for Comparative Medicine. This project was supervised by Dr. Chris Orvig and the manuscript was written by Xiaozhu Wang. Chapter 5 is an adaptation of a manuscript in preparation: Wang, X.; Jaraquemada-Pelaez, M. d. G.; Kostelnik, T. I.; Rodríguez-Rodríguez, Choudhary, N; C. Häfeli, U. O.; Orvig, C. H2C3hox: conformation of chelate ring makes a big difference. Xiaozhu Wang designed the ligand and performed the synthesis, characterization and DFT calculation; Solution studies were performed by Dr. María de Guadalupe Jaraquemada-Peláez; SPECT/CT imaging studies were performed by Dr. Cristina Rodríguez-Rodríguez at the Centre for Comparative Medicine with constructive viii advice from Dr. Urs Häfeli; X-ray crystallography was solved by Neha Choudhary. This project was supervised by Dr. Chris Orvig and the manuscript was written by Xiaozhu Wang. Chapter 6 is an adaptation of published work, and is reproduced in part from Wang, X.; Jaraquemada-Pelaez, M. d. G.; Rodríguez-Rodríguez, C.; Cao, Y.; Buchwalder, C.; Choudhary, N; Jermilova, N.; Ramogida, C. F.; Saatchi, K.; Häfeli, U. O.; Patrick, B. O.; Orvig, C. H4octox: Versatile Bimodal Octadentate Acyclic Chelating Ligand for Medicinal Inorganic Chemistry. J. Am. Chem. Soc. 2018, 140, 15487-15500. Copyright 2018 American Chemical Society. Xiaozhu Wang designed the ligand and performed the synthesis, characterization and fluorescence-based studies. Solution studies were performed by Dr. María de Guadalupe Jaraquemada-Peláez; DFT calculations were performed by Dr. Yang Cao; radiolabeling and SPECT/CT imaging studies were performed by Dr. Cristina Rodríguez-Rodríguez, Christian Buchwalder and Dr. Katayoun Saatchi at the Centre for Comparative Medicine and UBC Faculty of Pharmaceutical Sciences; X-ray crystallography was solved by Dr. Brian O. Patrick and Neha Choudhary; In vitro stability studies were performed by Una Jermilova (under the supervision of Caterina Ramogida) at TRIUMF. This project was supervised by Dr. Chris Orvig and the manuscript was written by Xiaozhu Wang. Chapter 7 is an adaptation of published work, and is reproduced in part from Cao, Y; Wang, X.; Shi, X.; Clee, S. M.; McGeer, P. L.; Wolf, M. O.; Orvig, C. Biological Imaging with Medium-Sensitive Bichromatic Flexible Fluorescent (FlexFluor) Dyes, Angew. Chem. Int. Ed, 2017, 56, 15603. Copyright 1999-2018 John Wiley & Sons, Inc. Xiaozhu Wang performed the fluorescence ix microscope imaging studies. Dyes synthesis was performed by Dr. Yang Cao; Biological sample preparation was performed by Xiaozhu Wang with assistance from Dr. Xiaolei Shi and constructive suggestion from Dr. Susanne Clee. and Pat McGeer. This project was supervised by Dr. Chris Orvig and Dr. Michael O. Wolf. Manuscript written by Xiaozhu Wang and Dr. Yang Cao. x Table of Contents Abstract ......................................................................................................................................... iii Lay Summary .................................................................................................................................v Preface ........................................................................................................................................... vi Table of Contents ...........................................................................................................................x List of Tables .............................................................................................................................. xix List of Figures ........................................................................................................................... xxiii List of Schemes .........................................................................................................................xxxv List of Symbols and Abbreviations ..................................................................................... xxxvii Acknowledgements ..................................................................................................................... xli Chapter 1: Introduction ........................................................................................................ 1 1.1 Medicinal Inorganic Chemistry .............................................................................. 1 1.2 Medicinal Imaging .................................................................................................. 2 1.2.1 Magnetic Resonance Imaging (MRI) and Contrast Agents ............................ 3 1.2.2 Ultrasound (US) and Contrast Agents ............................................................ 5 1.2.3 X-ray and CT .................................................................................................. 6 1.3 Nuclear Medicine .................................................................................................... 6 xi 1.3.1 Nuclear Imaging.............................................................................................. 7 1.3.2 Nuclear Medicine Therapy ............................................................................. 8 1.3.3 Radiometal Isotopes ........................................................................................ 9 1.4 Radiometal Chelators ............................................................................................ 11 1.4.1 Macrocyclic Chelators .................................................................................. 12 1.4.2 Acyclic Chelators .......................................................................................... 13 1.5 Biovectors ............................................................................................................. 14 1.6 Bioconjugation ...................................................................................................... 15 1.7 Thesis Overview ................................................................................................... 15 Chapter 2: H4octapa, New Synthetic Route and Derivatives ............................................ 18 2.1 Introduction ........................................................................................................... 18 2.2 Experimental ......................................................................................................... 21 2.2.1 Materials and Methods .................................................................................. 21 2.2.2 Ligand Synthesis and Characterization. ........................................................ 22 2.2.3 X-ray Crystallography .................................................................................. 26 2.2.4 DFT Calculations .......................................................................................... 27 2.2.5 Solution Thermodynamics ............................................................................ 28 xii 2.3 Results and Discussion ......................................................................................... 30 2.3.1 Synthesis of H4octapa ................................................................................... 30 2.3.2 Metal Complexes and X-ray Crystallography .............................................. 31 2.3.3 Ligand Protonation Constants ....................................................................... 33 2.3.4 Sm(III), Dy(III) and Yb(III) Complex Formation Equilibria with H4octapa 36 2.3.5 DFT Calculations and Molecular Electrostatic Potential Mapping .............. 39 2.4 Conclusions ........................................................................................................... 41 Chapter 3: H2hox: Dual Channel Oxine-Derived Acyclic Hexadentate Chelating Ligand for 68Ga Radiopharmaceuticals ......................................................................................... 43 3.1 Introduction ........................................................................................................... 43 3.2 Experimental ......................................................................................................... 46 3.2.1 Materials and Methods .................................................................................. 46 3.2.2 Ligand Synthesis and Characterization ......................................................... 47 3.2.3 Synthesis of Hox-Metal complex.................................................................. 48 3.2.4 X-ray Crystallography .................................................................................. 49 3.2.5 DFT and TDDFT Calculations ..................................................................... 49 3.2.6 Solution Thermodynamics ............................................................................ 50 3.2.7 [68Ga(hox)]+ Labeling Procedure. ................................................................. 51 xiii 3.2.8 Log D7.4 Measurements ................................................................................. 52 3.2.9 Stability in Mouse Plasma ............................................................................ 52 3.2.10 PET/CT Imaging Studies ............................................................................. 53 3.2.11 Fluorescence Spectra .................................................................................... 53 3.2.12 Fluorescence Microscopy ............................................................................ 54 3.3 Results and Discussion ......................................................................................... 54 3.3.1 Synthesis and Characterization of Ligands ................................................... 55 3.3.2 Preparation of the Metal Complex ................................................................ 56 3.3.3 X-ray Crystallography and DFT Calculations .............................................. 57 3.3.4 Solution Thermodynamics ............................................................................ 61 3.3.5 Dissociation Kinetics of Ga3+ Complex ........................................................ 69 3.3.6 68Ga Labelling Experiments .......................................................................... 70 3.3.7 Mouse Plasma Competition Experiments ..................................................... 71 3.3.8 Dynamic PET/CT imaging ........................................................................... 72 3.3.9 [Ga(hox)]+ Fluorescence and Cell Imaging Studies ..................................... 74 3.4 Conclusions ........................................................................................................... 77 Chapter 4: H2CHXhox, Cyclohexane Reinforced Chelating Ligand for Ga3+ and Cu2+ ... 79 xiv 4.1 Introduction ........................................................................................................... 79 4.2 Experimental ......................................................................................................... 83 4.2.1 Materials and Methods .................................................................................. 83 4.2.2 Ligand Synthesis and Characterization ......................................................... 84 4.2.3 Synthesis of CHXhox-Metal Complex.......................................................... 86 4.2.4 X-ray Crystallography .................................................................................. 87 4.2.5 Solution Thermodynamics ............................................................................ 88 4.2.6 PET/CT Imaging Studies .............................................................................. 90 4.2.7 SPECT/CT Imaging Studies ......................................................................... 91 4.3 Results and Discussion ......................................................................................... 91 4.3.1 Synthesis and Characterization of Ligands ................................................... 92 4.3.2 Preparation and Characterization of Metal Complexes ................................ 95 4.3.3 Solution Study ............................................................................................. 100 4.3.4 In vivo Imaging ........................................................................................... 111 4.4 Conclusions ......................................................................................................... 117 Chapter 5: H2C3hox, the Conformation Makes a Big Difference ................................... 118 5.1 Introduction ......................................................................................................... 118 xv 5.2 Experimental ....................................................................................................... 121 5.2.1 Materials and Methods ................................................................................ 121 5.2.2 Ligand Synthesis and Characterization ....................................................... 122 5.2.3 Synthesis of H2C3hox-Ga Complex ........................................................... 123 5.2.4 X-ray Crystallography ................................................................................ 123 5.2.5 Solution Thermodynamics .......................................................................... 124 5.2.6 SPECT/CT Imaging Studies ....................................................................... 126 5.3 Results and Discussion ....................................................................................... 127 5.3.1 Synthesis and Characterization of Ligand .................................................. 127 5.3.2 Preparation and Characterization of the Metal Complex ............................ 128 5.3.3 X-ray Crystallography ................................................................................ 130 5.3.4 Solution Studies .......................................................................................... 132 5.3.4.1 Protonation Constants of H2C3hox ....................................................... 132 5.3.4.2 Thermodynamic Stability ...................................................................... 135 5.3.5 SPECT/CT Imaging Studies. ...................................................................... 139 5.4 Conclusions ......................................................................................................... 141 Chapter 6: H4octox: Versatile Bimodal Octadentate Acyclic Chelating Ligand for Medicinal Inorganic Chemistry ...................................................................................... 142 xvi 6.1 Introduction ......................................................................................................... 142 6.2 Experimental ....................................................................................................... 145 6.2.1 Materials and Methods ................................................................................ 145 6.2.2 Synthesis and Characterization ................................................................... 146 6.2.3 Synthesis of Octox-Metal complex ............................................................. 147 6.2.4 X-Ray Crystallography ............................................................................... 148 6.2.5 DFT Calculations ........................................................................................ 149 6.2.6 Solution Thermodynamics .......................................................................... 149 6.2.7 pH- and Temperature-dependent [In(octox)]- Labeling Procedure. ........... 152 6.2.8 Serum Stability............................................................................................ 152 6.2.9 111In Radiolabeling of H4octox for in vivo studies ...................................... 153 6.2.10 In vivo SPECT/CT imaging ....................................................................... 153 6.2.11 Biodistribution of [111In(octox)]- ................................................................ 154 6.2.12 Fluorescence Spectra .................................................................................. 155 6.3 Results and Discussion ....................................................................................... 155 6.3.1 Synthesis and Characterization ................................................................... 155 6.3.2 X-ray Crystallography ................................................................................ 159 xvii 6.3.3 DFT Simulations and Molecular Electrostatic Potential Maps ................... 162 6.3.4 Solution Thermodynamics .......................................................................... 163 6.3.5 Chelation Enhanced Fluorescence Emission of H4octox ............................ 176 6.3.6 pH and Temperature Dependent [111In(octox)]- Labeling .......................... 179 6.3.7 Serum Stability............................................................................................ 180 6.3.8 In vivo Imaging of [111In(octox)]- ............................................................... 181 6.4 Conclusions ......................................................................................................... 186 Chapter 7: Biological Imaging with Medium-Sensitive Bichromatic Flexible Fluorescent (FlexFluor) Dyes ............................................................................................................. 188 7.1 Introduction ......................................................................................................... 188 7.2 Experimental ....................................................................................................... 190 7.2.1 Materials and Methods ................................................................................ 190 7.2.2 Dye Synthesis.............................................................................................. 190 7.2.3 Solution Studies .......................................................................................... 191 7.2.4 Biological Imaging...................................................................................... 191 7.3 Results and Discussion ....................................................................................... 192 7.4 Conclusion .......................................................................................................... 201 Chapter 8: Other Work and Future Studies ..................................................................... 202 xviii 8.1 Further Enhancing the H2dedpa and H4octapa Technology ............................... 202 8.2 Further Enhancing the H2hox Technology ......................................................... 203 8.3 Further Enhancing the H4octox Technology ....................................................... 204 8.4 Other Interesting Chelators ................................................................................. 205 Reference ....................................................................................................................................207 Appendices ..................................................................................................................................224 Appendix A : Supplementary Data for Chapter 2 .................................................................. 224 Appendix B : Supplementary Data for Chapter 3 .................................................................. 229 Appendix C : Supplementary Data for Chapter 6 .................................................................. 230 xix List of Tables Table 1.1. Comparison of various imaging modalities.a ................................................................ 2 Table 1.2. Ultrasound contrast agents that have/had been clinically approved.a ............................ 5 Table 1.3. Widely used radioisotopes for SPECT & PET imaging. a ............................................. 8 Table 1.4. Widely used radioisotopes for therapy.18 ....................................................................... 9 Table 1.5. Properties of some popular radiometal isotopes.20 ...................................................... 10 Table 1.6. Widely used bioconjugation strategies. ....................................................................... 16 Table 2.1. Selected average bond distances and angles for the [La(octapa)(H2O)2]- and [Gd(octapa)(H2O)2]- complexes.88 ................................................................................................ 33 Table 2.2. Protonation constants of the H4octapa ligand obtained at 25 °C, I = 0.16 M NaCl, calculated by using the HypSpec program.86 ................................................................................ 35 Table 2.3. Complex formation constants of Sm3+, Dy3+ and Yb3+ and H4octapa at 0.16 M NaCl and 25 °C. ............................................................................................................................................ 37 Table 2.4. Selected bond parameters of DFT calculated structures of [Ln(octapa)(H2O)]-2H2O anions. ........................................................................................................................................... 40 Table 3.1. Selected bond distances in the cations [Ga(hox)]+ and [Ga(dedpa)]+. ......................... 59 Table 3.2. Selected bond lengths of [Ga(hox)]+ in the solid state and calculated from DFT in solution. ......................................................................................................................................... 61 xx Table 3.3. Selected bond lengths of [Ga(dedpa)]+ in the solid state and calculated from DFT in solution. ......................................................................................................................................... 61 Table 3.4. Protonation Constants of H2hox at 25 °C. ................................................................... 62 Table 3.5. Molar absorptivity of selected wavelengths in different species of oxine and H2hox. 65 Table 3.6. Formation constants of Ga3+ complexes and pMk values. ........................................... 66 Table 3.7. Mouse plasma stability of 68Ga-hox. ........................................................................... 71 Table 4.1. Stability of [67Ga(dedpa)]+and [67Ga(CHXdedpa)]+ in human serum at 37 °C.55 ........ 80 Table 4.2. The bond lengths of the four Cu complex structures. .................................................. 99 Table 4.3. Protonation constants of H2CHXhox and H2hox at 25 °C. ........................................ 102 Table 4.4. Stepwise stability constants (log K) of H2CHXhox and H2hox complexes with Cu2+, and pM values a. ................................................................................................................................. 104 Table 4.5. Formation constants of Cu2+ complexes and pM values.187 ...................................... 104 Table 4.6. Stepwise stability constants (log K) of H2CHXhox and H2hox complexes with Ga3+, and pM valuesa. .................................................................................................................................. 108 Table 5.1. Selected bond distances and bond angles in the cations [Ga(hox)]+168 and [Ga(C3hox)]+...................................................................................................................................................... 130 Table 5.2. Protonation constants of H2C3hox and H2hox at 25 °C. ........................................... 133 xxi Table 5.3. Stability constants (log K) of H2C3hox, H2CHXhox and H2hox complexes with Ga3+...................................................................................................................................................... 135 Table 5.4. Formation constants of Ga3+ complexes and pMk values. ......................................... 139 Table 6.1. Selected average bond angles and bond distances in the [La(octox)]- complex of Figure 6.5................................................................................................................................................ 161 Table 6.2. DFT optimized structures and electrostatic potential (ESP) mapping of [In(octapa)]- and In(octox)]-. .................................................................................................................................. 163 Table 6.3. Protonation constants of H4octox (H4L) at 25 °C. ..................................................... 167 Table 6.4. Stepwise stability constants (log K) of H4octox complexes with Y3+, La3+, Gd3+, Lu3+ and In3+. ....................................................................................................................................... 175 Table 6.5. pMa values of the most relevant In3+ chelators. ......................................................... 176 Table 6.6. Radiochemical yields (RCY in %) of the various labeling reactions performed for the 111In-H4octox system. .................................................................................................................. 179 Table 6.7. Mouse serum stability challenges performed at ambient temperature.a .................... 181 Table A.1. 1H NMR chemical shifts δ (ppm) and variations of chemical shifts Δδ (ppm) of a set of solutions at the same ligand concentration [H4octapa] = 1.03 x 10-3 M and different acidities, at 25 °C, I = 0.16 M NaCl. .............................................................................................................. 224 Table A.2. Crystallographic data for H4octapa and its La(III) complex. .................................... 225 Table B.1. Crystallographic Data for the H2hox and [Ga(hox)][ClO4] Structures……………...229 xxii Table C.1. Crystallographic data for the La-octox structure………………………...……….230 xxiii List of Figures Figure 1.1. The author’s view of the subject of Medicinal Inorganic Chemistry. .......................... 1 Figure 1.2. Drugs targeting metal ions: dimercaptosuccinic acid, a fluorescent calcium probe. ... 1 Figure 1.3. Metal-based drugs and imaging agents: cisplatin, lithium carbonate and 99mTc sestamibi. ........................................................................................................................................ 2 Figure 1.4. Gd(III)-based MRI contrast agents.5 ............................................................................ 3 Figure 1.5. Mn(II)-based MRI contrast agents.6, 10-11 ...................................................................... 4 Figure 1.6. Samples of X-ray and CT contrast agents (adapted from the literature13). .................. 6 Figure 1.7. Depiction of SPECT, PET and X-ray.14 ....................................................................... 7 Figure 1.8. Depiction of a chelating agent functionalized with a monoclonal antibody.21........... 11 Figure 1.9. Macrocyclic chelating ligands: DOTA, NOTA, CB-DO2A and PCTA. ................... 12 Figure 1.10. Acyclic chelating ligands: EDTA, DFO, DTPA, CHX-A\"-DTPA, H2dedpa, H4octapa........................................................................................................................................................ 13 Figure 1.11. Classes of biovectors. ............................................................................................... 14 Figure 2.1. ORTEP diagram of H8(octapa)Cl4·2H2O (H4octapa·4HCl·2H2O) (–x, 1-y, 1-z); associated crystallographic data are presented in Appendix (Table A.1). .................................... 31 xxiv Figure 2.2. ORTEP diagram of the two La-octapa complexes in the asymmetric unit with hydrogen atoms omitted for clarity; associated crystallographic data are presented in Appendix Table A.2........................................................................................................................................................ 33 Figure 2.3. Representative absorption spectra of a potentiometric-spectrophotometric titration of H4octapa; [H4octapa] = 6.61  10-4 M, at I = 0.16 M NaCl, 25 °C; path length = 0.2 cm. .......... 35 Figure 2.4. a) b) and c) Representative spectra of the in batch UV spectrophotometric titration of the Yb3+-H4octapa system, [H4octapa] = 1.31 × 10-4 M, [Yb3+] = 1.32 × 10-4 M, path length = 1 cm. d) Set of spectra from the potentiometric-spectrophotometric titration of the Yb3+-H4octapa system, [H4octapa] = [Yb3+] = 6.70 × 10-4 M, path length = 0.2 cm. ........................................... 38 Figure 2.5. Speciation plots for the H4octapa complexes with: a) Sm3+, [Sm3+] = [H4octapa] = 5.52  10-4 M; b) Dy3+, [Dy3+] = [H4octapa] = 6.65  10-4 M; c) Yb3+, [Yb3+] = [H4octapa] = 6.70  10-4 M, at 25 °C and I = 0.16 M NaCl. .............................................................................................. 38 Figure 2.6. a) Optimized structure of the [Sm(octapa)(H2O)]-2H2O anion with labels on selected atoms, O5 indicates the oxygen atom on the metal-bound water molecule while O6 and O7 are the second-sphere waters; hydrogen atoms have been omitted for clarity. b) Zoomed-in view of the anion from a different angle, showing the hydrogen bonds between the water molecules and carboxylate arms. .......................................................................................................................... 40 Figure 2.7. MEP mapping of the [Sm(octapa)(H2O)]-2H2O anion, red = negative, blue = positive, representing a maximum potential of 0.200 au and a minimum of -0.200 au, mapped onto electron density isosurface of 0.002 Å-3. All hydrogen atoms have been omitted for clarity. .................... 41 xxv Figure 3.1. 1H NMR spectra of H2hox in MeOD (400 MHz, 25 °C) (upper) and [Ga(hox)][ClO4] in MeOD (300 MHz, 25 °C) (lower). ........................................................................................... 57 Figure 3.2. ORTEP-style image of H2hox, including symmetry equivalencies. (i) 2-x, -y,1-z, (ii) 1-x,1-y,1-z. .................................................................................................................................... 58 Figure 3.3. ORTEP-style picture of [Ga(hox)][ClO4] complexes. The two crystallographically distinct cation structures and perchlorate anions shown. Solvent acetonitrile molecules not shown, for clarity ....................................................................................................................................... 59 Figure 3.4. DFT optimized structures of the (a) [Ga(hox)]+ and (b) [Ga(dedpa)]+ cations with hexacoordinated metal centers, and the electrostatic potentials of (c) [Ga(hox)]+ and (d) [Ga(dedpa)]+ mapped onto their electron density. The MEP represents a maximum potential of 0.25 au, and a minimum of −0.02 au mapped onto electron density isosurface (0.002 e Å–3, red to blue = negative to positive). .......................................................................................................... 60 Figure 3.5. Spectra of the in batch UV spectrophotometric titration of 2.72 × 10-5 M solution of H2hox. ........................................................................................................................................... 64 Figure 3.6. pH-dependent absorption spectra (in molar absorptivity) of the 7 light absorbing species of H2hox. .......................................................................................................................... 64 Figure 3.7. Representative spectra of the in batch UV spectrophotometric acidic titration for the Ga3+-H2hox system, 1:1 M:L molar ratio, [H2hox] = 2.82 × 10-5 M, (l = 1 cm) at 0.16 M NaCl and 25 °C. ............................................................................................................................................ 66 xxvi Figure 3.8. (a) Distribution diagram of H2hox calculated using the protonation constants in Table 3.5, at ligand concentration of 2.72 × 10-5 M; (b) Distribution diagram of the Ga-hox system, [Ga3+] = [H2hox] = 2.82 × 10-5 M; 25 °C; I = 0.16 M NaCl. (c) and (d) representative spectra in the batch UV spectrophotometric titration of the Ga3+-H2hox system, 1:1 M:L molar ratio, [H2hox] = 2.82 × 10-5 M, (l = 1 cm) at 0.16 M NaCl and 25 °C. .............................................................................. 68 Figure 3.9. Time course of the UV spectra of the [Ga(hox)]+ complex (2.82 × 10-5 M) at pH 1 and 25 °C (spectra obtained at 15 min intervals). The inset shows the absorbance at 243 nm vs. time, and its fitting to the observed first-order rate constant. ................................................................ 70 Figure 3.10. HPLC traces of radiation and UV absorption of the mixtures of [68Ga(hox)]+ ([H2hox)] = 1  10-7 M) and [Ga(hox)]+ non-radioactive complex ([Ga(hox)]+ = 5  10-5 M). ................... 71 Figure 3.11. (a) PET/CT dynamic imaging and (b) biodistribution of [68Ga(hox)]+ in male mice during (b) 0 min – 3 min and (c) 3 min – 53 min. ........................................................................ 74 Figure 3.12. Fluorescence spectra of H2hox and its Ga3+ complex in PBS (λexc = 365 nm, [H2hox] =[[Ga(hox)]+] = 1.7  10-5 M) (upper). Time dependent fluorescence microscopy images from Hela cells treated with 150 M [Ga(hox)] [ClO4] (scale = 20 m) (lower). ................................ 76 Figure 3.13. Bright field images from Hela cells incubated with [Ga(hox)][ClO4] (scale = 20 m)........................................................................................................................................................ 77 Figure 4.1. 1H NMR spectrum of H2CHXhox in MeOD (300 MHz, 25 °C) ................................ 93 Figure 4.2. 1H NMR spectrum of [Ga(CHXhox)][ClO4] in MeOD (300 MHz, 25 °C). ............... 96 Figure 4.3. ORTEP-style picture of [Cu(hox)]. ........................................................................... 98 xxvii Figure 4.4. ORTEP-style picture of [Cu(CHXhox)]4. ................................................................... 98 Figure 4.5. Comparison of solid structure of In(oxine)3190 and [Cu(CHXhox)]. .......................... 99 Figure 4.6. Representative spectra of the in batch UV spectrophotometric titration of 2.75  10-5 M H2O solution of H2CHXhox, l = 1 cm at 25 °C. ..................................................................... 101 Figure 4.7. Speciation plots of H2CHXhox (black line) and H2hox (blue dash line) at 2.75  10-5 M ligand concentration and fit for the ligand batch titration at 25 °C (upper). Change of absorbance at 258 nm (blue dots are original data points and the red line represents the fit versus data points (lower left). Obtained (blue) and fitted (red) spectra for a pH = 9.12 specific data point (lower right). ........................................................................................................................................... 102 Figure 4.8. Representative spectra of the UV-potentiometric titrations of the Cu(II)-H2CHXhox system, showing the different transformations as the pH is raised a) and b) from -0.47 to 0.53, c) and d) from 0.53 to 1.37, e) and f) from 1.37 to 4.0, g) and h) from 4.73 to 10.78. [H2CHXhox] = [Cu2+] = 2.72  10-5 M, path length = 1 cm, at 25 °C. ................................................................ 105 Figure 4.9. Representative spectra of the UV-potentiometric titrations of the Cu(II)-H2hox system, showing the different transformations as the pH is raised a) and b) from 0.27 to 1.14, c) and d) from 1.14 to 1.60, e) and f) from 1.72 to 4.5, g) and h) from 5.05 to 11.51. [H2hox] = 2.90  10-5 M; [Cu2+] = 2.76  10-5 M, path length = 1 cm, at 25 °C. .......................................................... 106 Figure 4.10. Representative spectra of the UV-potentiometric titrations of the Ga(III)-H2CHXhox system, showing the different transformations as the pH is raised a) and b) from -0.44 to 0.60, c) xxviii and d) from 0.6 to 10.59. [H2CHXhox] = [Ga3+] = 2.72  10-5 M, path length = 1 cm, at 25 °C...................................................................................................................................................... 108 Figure 4.11. Fit for the batch titration of the Ga(III)-H2CHXhox system at 25 °C, [H2CHXhox] = [Ga3+] = 2.72  10-5 M. Left: Change of absorbance at 258 nm (blue dots are original data points and the red line represents the fit versus data points. Right: Obtained (blue) and fitted (red) spectra for a pH = 0.91 specific data point. ............................................................................................. 109 Figure 4.12. UV spectra of dissociation kinetics of [Ga(CHXhox)]+. ........................................ 110 Figure 4.13. Dissociation kinetics of [Ga(CHXhox)]+ complex. ................................................ 111 Figure 4.14. PET/CT dynamic imaging of [68Ga(CHXhox)]+. ................................................... 112 Figure 4.15 SPECT/CT imaging of [67Ga(CHXhox)]+. .............................................................. 113 Figure 4.16. Biodistribution of [68Ga(hox)]+ and [68Ga(CHXhox)]+ in male mice. .................... 114 Figure 4.17. Heart/liver uptake ratio of [68Ga(hox)]+, [68Ga(CHXhox)]+, [67Ga(dedpa-D8)]+ and [67Ga(dedpa-D9)]+.52 ................................................................................................................... 115 Figure 4.18. SPECT/CT scans imaging of [67Ga(Br4hox)]+. ...................................................... 116 Figure 5.1. a) The low strain chair form conformer of cyclohexane, and b) how this relates to minimum strain energy forms of the chelate ring with respect to the M-N bond length and N-M-N bond angle, for 5/6-membered chelate rings.195 .......................................................................... 119 Figure 5.2. Plot of change in formation constant, log K from 5-membered chelate ring to 6-membered chelate ring versus metal ions with various ionic radii.195 ........................................ 119 xxix Figure 5.3. 1H NMR spectra of H2C3hox in MeOD (300 MHz, 25 °C) (upper) and [Ga(C3hox)][ClO4] in MeOD (300 MHz, 25 °C) (lower). ......................................................... 129 Figure 5.4. ORTEP-style picture of [Ga(C3hox)][ClO4] (C23H22ClGaN4O6). ........................... 130 Figure 5.5. Side-on view of N(pn)-Ga-N(pn) bond angle in [Ga(C3hox)][ClO4] (left) and N(en)-Ga-N(en) angle (right) in [Ga(hox)][ClO4]. ................................................................................ 132 Figure 5.6. pH-dependent electronic absorption spectra (in molar absorptivity) of the seven light absorbing species of H2C3hox. ................................................................................................... 133 Figure 5.7. Representative spectra of the in-batch UV spectrophotometric titrations of 3.16  10-5 M H2O solution of H2C3hox, l = 1 cm at 25 °C. ........................................................................ 134 Figure 5.8. Representative spectra of the UV-potentiometric titration of the Ga(III)-H2C3hox system, showing the different transformations as the pH is raised a) and b) from 2.46 to 8.15, c) and d) from 8.15 to 10.91. [H2C3hox] = [Ga3+] = 3.12  10-5 M, path length = 1 cm, at 25 °C. 136 Figure 5.9. Speciation diagrams for the Ga(III)-C3hox system calculated from values given in Tables 5.2 (a) and 5.3 (b). [Ga3+] = [H2C3hox] = 3.16  10-5 M, at 25 °C. ................................ 136 Figure 5.10. Conformations of 5 or 6 membered chelate ring formed by backbone nitrogen atoms and Ga3+ in [Ga(hox)][ClO4] (left) and [Ga(C3hox)][ClO4] (right). .......................................... 138 Figure 5.11. SPECT/CT imaging of [67Ga(C3hox)]+.................................................................. 140 xxx Figure 6.1. a) Portion of the 1H NMR spectrum of the [In(octox)]- complex showing the aromatic protons (pH = 8, 400 MHz, 25 °C); b) VT 1H NMR experiment for [In(octox)]- (pH = 8, 400 MHz). A and B represent the two different species in solution of the [In(octox)]- complex. ................ 157 Figure 6.2. 1H NMR spectrum of Na[In(octox)] in D2O (400 MHz, 25 °C). ............................. 158 Figure 6.3. The 2D NOESY spectrum of Na[In(octox)] in D2O (400 MHz, 25 °C). ................. 158 Figure 6.4. ORTEP diagram of the [La3(octox)4]7-•7H2O in the asymmetric with 50% probability ellipsoids; associated crystallographic data are presented in Appendix Table C.1. ................... 160 Figure 6.5. ORTEP diagram of 10 coordinated La[(octox)]- in the asymmetric unit (50% probability ellipsoids). The solvent molecules, cations, and hydrogens are omitted for clarity. 161 Figure 6.6. Ball-and-stick presentation of the two isomeric species simulated using DFT calculations. ................................................................................................................................ 162 Figure 6.7. Representative spectra of the in batch UV spectrophotometric titration of 3.06  10-5 M solution of H4octox in H2O, l = 1 cm at 25 °C. ...................................................................... 165 Figure 6.8. Representative spectra of the simultaneous UV-potentiometric titration of 9.55  10-4 M solution of H4octox, l = 0.2 cm at I = 0.16 M NaCl and 25 °C. ............................................. 166 Figure 6.9. Speciation plot of H4octox (H4L) calculated using protonation constants in Table 6.1, at ligand concentration 1  10-3 M. ............................................................................................. 166 Figure 6.10. Representative spectra of the UV-potentiometric titration of the Y3+-H4octox system, showing the different transformations as the pH is raised a) from 1.68 to 2.35, b) from 2.35 to 3.13, xxxi c) from 3.55 to 4.53, d) from 4.53 to 7.61 and e) from 6.58 to 10.74. [H4octox] = [Y]3+ = 8.57  10-4 M, path length = 0.2 cm, I = 0.16 M (NaCl) at 25 °C. f) Distribution curves for Y3+-H4octox complexes; [Y3+] = [H4octox] = 8.5  10-4 M, dash line represents physiological pH (7.4). ..... 170 Figure 6.11. Representative spectra of the UV-potentiometric titration of the La3+-H4octox system, showing the different transformations as the pH is raised a) from 1.89 to 3.48, b) from 3.48 to 5.86, c) from 5.86 to 9.20, d) from 9.98 to 11.43. [H4octox] = [La3+] = 7.65  10-4 M, path length = 0.2 cm, I = 0.16 M (NaCl) at 25 °C. e) Distribution curves for La3+-H4octox complexes; [La3+] = [H4octox] = 8.5  10-4 M, dash line represents physiological pH (7.4). ..................................... 171 Figure 6.12. Representative spectra of the UV-potentiometric titration of the Gd3+-H4octox system, showing the different transformations as the pH is raised a) from 1.77 to 2.15, b) from 2.15 to 3.22, c) from 3.26 to 4.96, d) from 4.96 to 7.93. [H4octox] = [Gd3+] = 7.03  10-4 M, path length = 0.2 cm, I = 0.16 M (NaCl) at 25 °C. e) Distribution curves for Gd3+-H4octox complexes; [Gd3+] = [H4octox] = 8.5  10-4 M, dash line represents physiological pH (7.4). ..................................... 172 Figure 6.13. Representative spectra of the UV-potentiometric titration of the Lu3+-H4octox system, showing the different transformations as the pH is raised a) from 2.44 to 3.0, b) from 3.24 to 4.02, c) from 4.02 to 6.51, d) from 6.51 to 11.44. [H4octox] = [Lu3+] = 6.93  10-4 M, path length = 0.2 cm, I = 0.16 M (NaCl) at 25 °C. e) Distribution curves for Lu3+-H4octox complexes; [Lu3+] = [H4octox] = 8.5  10-4 M, dash line represents physiological pH (7.4). ..................................... 173 Figure 6.14. a) Representative spectra of the in batch UV spectrophotometric titration of the In3+-H4octox system, showing the complex formation as the pH is raised from 0.74 to 1.65. [H4octox] = [In3+] = 3.04  10-5 M, path length = 1 cm, I = 0.16 M (NaCl) at 25 °C. Representative spectra xxxii of the UV-potentiometric titration of the In3+-H4octox system, showing the different transformations as the pH is raised b) from 1.59 to 2.24, c) from 2.23 to 3.33, d) from 3.43 to 9.94, e) from 9.51 to 11.40. [H4octox] = [In3+] = 8.13  10-4 M, path length = 0.2 cm, I = 0.16 M (NaCl) at 25 °C. ...................................................................................................................................... 174 Figure 6.15. a) pM values vs ionic radius for M3+- H4octox complexes (CN = 8);196 b) Distribution curves for In3+- H4octox complexes, [In3+] = [H4octox] = 8.5  10-4 M. Dashed line in b at physiological pH (7.4). ............................................................................................................... 175 Figure 6.16. Fluorescence spectra of H4octox and its Y3+, La3+, Lu3+, In3+ complexes at pH = 8 (λexc = 350 nm, [H4octox] = [M(octox)]- = 2  10-4M). ...................................................................... 177 Figure 6.17 a) Fluorescence spectra of YCl3(5  10-5M), transferrin (1 mg/mL), [Y(octox)]- (5  10-5 M), transferrin (1 mg/mL) + YCl3 (5  10-5 M) and Octox-Y complex (5  10-5 M) + transferrin (1 mg/mL) in PBS (pH 7.4) solution. b) Octox-Y complex stability study against transferrin. . 178 Figure 6.18. The [111In(octox)]- radiocomplex (top, blue) elutes with nearly the same radio-HPLC retention time (tR = 6.1min) as the non-radioactive [In(octox)]- complex (bottom, blue). H4octox (bottom, brown) elutes 1.4 min later and 111InCl3 (top, green) elutes with the mobile phase front...................................................................................................................................................... 182 Figure 6.19. Top view maximum intensity projections (MIPs) on total body SPECT/CT scans at various time points after administration of [111In(octox)]- complex. .......................................... 184 Figure 6.20. Representative SUV time-activity curves for [111In(octox)]- in mice (n = 3). Biodistribution of [111In(octox)]-................................................................................................. 185 xxxiii Figure 6.21. Biodistribution of [111In(octox)]- after 24 hours post-injection, n = 3. ................... 185 Figure 7.1. Bichromatic medium-sensitive fluorescent dyes. ..................................................... 189 Figure 7.2. Normalized excitation and emission spectra of (a) Dye A, b) FlexFluor-C6-I, (c) Dye B and (d) Dye C in ethyl octanoate and methanol (1 x 10-5 M). (For Dye A, Dye B and FlexFluorC6-I, lEX (MeOH) = 330 nm, lEM (MeOH) = 540 nm, lEX (EO) = 360 nm, lEM (EO) = 450 nm; for Dye C, lEX (MeOH) = 350 nm, lEM (MeOH) = 660 nm, lEX (EO) = 450 nm, lEM (EO) = 560 nm; methanol used as the strong HB donor solvent due to poor solubility of Dye A in water). 193 Figure 7.3. Ethyl octanoate (EO)/water partitioning experiment for (a) Dye A and (b) Dye B. A solution of dye (10 mL, 1 x 10-5 M) in water saturated EO was mixed with EO saturated water (10 mL) for five minutes. The EO/water phases were then separated and the emission profile measured using a fluorimeter. ..................................................................................................................... 195 Figure 7.4. (a) Photograph of emission from the organic and water phases of a 1:1 (v/v) equilibrated mixture of ethyl octanoate and water containing Dye A (left) and Dye B (right). (b) Cartoon illustration of adipocyte cells. Optical fluorescence microscopy image of adipose tissue treated with (c) Dye A and (d) Dye B. (1 ms exposure, scale bar = 100 mm.) Images of mice adipose tissue with co-existing Dye A and Dye B in (e) bright field full color and (f) color-separated red channel. (1 ms exposure, scale bar = 100 mm). ........................................................................................ 196 Figure 7.5. Fluorescence images of mice adipose tissue with co-existing Dye A and Dye B in (a) bright field full color and (b) color-separated blue channel (c) color-separated green channel and (d) color-separated red channel. (1 ms exposure, scale bar = 100 um). ..................................... 197 xxxiv Figure 7.6. (a) Cartoon illustration showing a section of axon with myelin sheath, and fluorescence image of brain tissue with Dye A showing (b) region with no observable neuron, scale bar = 100 mm; (c) area in inferior olivary nucleus rich in neurons, scale bar = 50 mm; and (d) zoomed-in image, scale bar = 25 mm. .......................................................................................................... 199 Figure 7.7. Fluorescence image of brain tissue with Dye A showing (a) region rich in axons and (b) lipid-rich region with few axons. .......................................................................................... 200 Figure A.1. Representative spectra of the UV-Vis spectrophotometric titration of [H4octapa] = 1.28 x 10-4 M at 25 °C, path length = 1 cm. ........................................................................................ 226 Figure A.2. Representative spectra of the Sm3+-H4octapa system. a) In batch UV spectrophotometric experiments, [H4octapa] = [Sm3+] = 1.51 x 10-4 M, path length = 1cm; b) and c) Combined potentiometric-spectrophotometric titrations, [H4octapa] = 5.56 x 10-4 M, [Sm3+] = 5.52 x 10-4 M, path length = 0.2 cm, at 25 °C and I = 0.16 M NaCl. ......................................... 227 Figure A.3. a) and b) Representative spectra of the in batch UV spectrophotometric experiments for the Dy3+-H4octapa system, [H4octapa] = 1.38 x 10-4 M, [Dy3+] = 1.37 x 10-4 M, path length = 1 cm, at 25 °C and I = 0.16 M NaCl. .......................................................................................... 227 Figure A.4. Optimized structure of (a) the [Dy(octapa)(H2O)]-2H2O anion and (b) the [Yb(octapa)(H2O)]-2H2O with labels on selected atoms. .......................................................... 228 xxxv List of Schemes Scheme 2.1. Some of the pyridinecarboxylate scaffold (pa) chelators synthesized previously, and the EDTMP ligand discussed in this work. ................................................................................... 21 Scheme 2.2. Improved synthesis of the ligand H4octapa. a) SOCl2, MeOH, reflux overnight, 99%; b) Na2CO3, THF, 50 °C, 24 h, 83%; c) HCl (6 M), 63% (cumulative yield of ∼ 50% over three steps). ............................................................................................................................................ 31 Scheme 3.1. Examples of chelating ligands for Ga3+ from the literature. .................................... 45 Scheme 3.2. Design paradigm for H2hox...................................................................................... 55 Scheme 3.3. Synthesis of H2hox. a) en, CH3CH2OH, at 60 °C, 4 hr; b) CH3CH2OH, NaBH4 (5 equiv), overnight. .......................................................................................................................... 56 Scheme 4.1. DOTA, NOTA, DTPA, H2dedpa and cyclohexane reinforced chelators. ................ 80 Scheme 4.2. Schiff-base amine chelators and log D values of their 68Ga complexes.154, 173 ........ 82 Scheme 4.3. H2dedpa derivatives and log P values of their 67Ga complexes.52, 127 ...................... 83 Scheme 4.4. Design of H2CHXhox. .............................................................................................. 92 Scheme 4.5. Synthesis of H2CHXhox. a) en, CH3OH, 60 °C, 4 hr; b) CH3OH, NaBH4 (5 equiv), overnight. ...................................................................................................................................... 93 Scheme 4.6. Preparation of H2Br4hox and H2CHXBr4hox. .......................................................... 94 Scheme 5.1. Examples of ligands with 5/6 membered chelate rings. ......................................... 120 xxxvi Scheme 5.2. Design of H2C3hox. ............................................................................................... 121 Scheme 5.3. Synthesis of H2C3hox. .......................................................................................... 127 Scheme 5.4. Potential bifunctional derivatives of H2hox and H2C3hox. .................................... 139 Scheme 6.1. Representative chelating ligands for In3+, Y3+, Lu3+, Gd3+ in the literature. .......... 144 Scheme 6.2 Design paradigm of H4octox. .................................................................................. 145 Scheme 6.3. Synthesis of H4octox. Reagents and conditions: a) CH3CN, tert-butyl bromoacetate, Na2CO3 (excess), 50 °C, 24 h, 84%; b) HCl (6 M), 60 °C, overnight, 97%. .............................. 155 Scheme 7.1. Synthetic route to three flexfluor dyes. .................................................................. 189 Scheme 7.2. Synthesis of Precursor FlexFluor-C6-Cl. ............................................................... 191 Scheme 8.1. H2dedpa and H4octapa derivatives. ........................................................................ 202 Scheme 8.2. p-NO2-hox and p-NCS-hox. ................................................................................... 203 Scheme 8.3. Other possible bifunctional derivatives of H2hox and H2C3hox............................ 204 Scheme 8.4. p-NCS-octox preparation. ..................................................................................... 205 Scheme 8.5. Other interesting chelators. .................................................................................... 205 xxxvii List of Symbols and Abbreviations %ID/g percentage of injected radioactive dose per gram of tissue ~ approximate 2D two dimensional 3D three dimensional α alpha particle or alpha position on molecule Å angstrom, 1·10-10 m β- beta particle β+ positron γ gamma ray δ delta of chemical shift in parts per million (NMR) Δ heat or reflux λ wavelength in nm μ micro (10-6) μM micromolar (10-6 M) AAS atomic absorption standard Ab antibody Ac acetate ATSM diacetyl-bis(N4-methylthiosemicarbazone) BFC bifunctional chelate, or bifunctional ligand Bn benzyl °C degrees Celsius calcd. calculated CB-TE2A 4,11-bis(carboxymethyl)-1,4,8,11-tetraazabicyclo-[6.6.2]-hexadecane CHX cyclohexane/cyclohexyl CHX-A”-DTPA cyclohexyldiethylenetriaminepentaacetic acid CHEF chelation enhanced fluorescence Ci Curie CN coordination number COSY correlation spectroscopy (1H-1H NMR) CT computed tomography d day(s) or doublet (NMR) D distribution coefficient (octanol/aqueous at specified pH) xxxviii Da dalton DCM dichloromethane DFT density functional theory (in silico calculations) DFO desferrioxamine B DIPEA diisopropylethylamine DMF dimethylformamide DMSO dimethylsulfoxide DOTA 1,4,7,10-tetraazacyclododecane-N,N’,N”,N’’’-tetraacetic acid DTPA diethylenetetraaminepentaacetic acid E reduction or oxidation or half potential (for electrochemistry) EA elemental analysis EC electron capture EDDA ethylenediamine diacetate EDTA ethylenediaminetetraacetic acid en pn ethylenediamine propylenediamine EPR enhanced permeability and retention equiv equivalent(s) ESI-MS electrospray ionization mass spectrometry EtOAc ethyl acetate EtOH ethanol eV electronvolt(s) FDA Food and Drug Administration (USA) FDG 2-deoxy-2-[18F]fluoro-D-glucose FITC fluorescein isothiocyanate g gram h hour(s) H2C3hox 2,2'-((propane-1,3-diylbis(azanediyl))bis(methylene))bis(quinolin-8-ol) H2CHXdedpa cyclohexyl-H2dedpa H2CHXhox cyclohexyl- H2hox H4CHXoctapa cyclohexyl-H4octapa H4CHXoctox cyclohexyl- H4octox H2dedpa 1,2-[[6-carboxy-pyridin-2-yl]-methylamino]ethane H2hox 2,2'-((ethane-1,2-diylbis(azanediyl))bis(methylene))bis(quinolin-8-ol) H4octapa N,N’-bis(6-carboxy-2-pyridylmethyl)-ethylenediamine-N,N’-diacetic acid xxxix H4octox 2,2'-(ethane-1,2-diylbis(((8-hydroxyquinolin-2-yl)methyl)azanediyl))diacetic acid HPLC high performance liquid chromatography HR high resolution HSQC heteronuclear single bond correlation/coherence (1H-13C NMR) Hz Hertz I ionic strength Ig immunoglobulin J coupling constant (NMR) k kilo KML thermodynamic complex stability constant L litre or ligand LET linear energy transfer m milli- or multiplet (NMR) mAb monoclonal antibody M molar (moles/litre) or mega (106) or metal MeCN acetonitrile MeOH methanol MEP molecular electrostatic potential min minute(s) mol mole MRI magnetic resonance imaging MS mass spectrometry m/z mass per unit charge n nano (10-9) or number of unit or neutron NBS N-bromosuccinimide NHE normal hydrogen electrode NHL non-Hodgkin’s lymphoma NHS N-hydroxysuccinimide NI nitroimidazole NMR nuclear magnetic resonance nM nanomolar (10-9 M) Nosyl 2-nitrobenzenesulfonamide (protecting group) NOTA 1,4,7-triazayclononane-1,4,7-triacetic acid NRU National Research Universal (reactor) ORTEP Oak Ridge Thermal Ellipsoid Plot Program P partition coefficient (octanol/water) p proton p para substituent p.i. post injection PBS phosphate buffered saline xl PCTA 3,6,9,15-tetraazabicyclo[9.3.1]pentadeca-1(15),11,13-triene-3,6,9-triacetic acid PET positron emission tomography pH -log[H3O+] pKa protonation constant pm picometer (10-12 m) pM -log[free metal], or picomolar (10-12 M) ppm parts per million pyr pyridyl/pyridine q quartet (NMR) ® trademark RCY radiochemical yield RGD Arg-Gly-Asp cyclic peptide RIT radioimmunotherapy RP reverse phase (column chromatography) rpm rotations per minute RT room temperature s singlet (NMR) or second(s) S.A. specific activity (radioactivity per unit mass) SPECT single photon emission computed tomography t triplet (NMR) or time tR retention time (HPLC) t1/2 half-life TAC time-activity curve TBAP tetrabutylammonium perchlorate TCMC 1,4,7,10-tetraaza-1,4,7,10-tetra-(2-carbamoyl methyl)-cyclododecane TETA 1,4,8,11-tetraazacyclotetradecane-1,4,8,11-tetraacetic acid TFA trifluoroacetic acid THF tetrahydrofuran TLC thin layer chromatography TM Trade Market TOF time-of-flight TRAP 1,4,7-triazacyclononane-1,4,7-tris[methyl(2-carboxyethyl)phosphinic acid UV ultraviolet V volt VEGF Vascular Endothelial Growth Factor VT-NMR variable temperature NMR y year xli Acknowledgements First and foremost, I must acknowledge my supervisor Professor Chris Orvig for his guidance and support in the past 6 years. Thank you for all the encouragement, from my research to English and presentation, for the freedom to explore my “broad” scientific interests, for all the patience with my “abnormal” work time. I would also like to thank every member in this wonderful group. It is a great please to work with you. Thank Eric (Dr. Eric Price) and Caterina (Dr. Caterina F. Ramogida) for the guidance in my first year, Maria (Dr. Maria A. Telpoukhovskaia), Katja (Dr. Katja Dralle Mjos), Karen (Karen Arane), and Sara (Dr. Sarah Spreckelmeyer) for all the lab tips. Tom (Tom Kostelnik) for all the good discussion and being a table tennis and squash partner. Chirag (Chirag Apte), Jim (Xiaodong Zhao), and Anna (Anna Schnörr) for all the hard work and eagerness to learn. Aidan (Aidan Ingham) for all the shared late nights in the lab with mutual growth, Neha (Neha Choudhary) for all the beautiful “crystal” work, Lily S (Lily Southcott) for the scientific trust and inspiring talk, Lily L (Lily Li) for the effective communication, Jeff (Jefferey Jang) for the bright mood every day, Luke (Luke Wharton) for the “weekend” workmate. In particular, I’d like to thank Lupe (Dr. María de Guadalupe Jaraquemada-Peláez) for all the precious understanding, support, encouragement and help, and Yang (Dr. Yang Cao) for all the patient guidance. I learned so much from you two. We are always a great team together. I would also like to acknowledge all the shops and services at UBC Chemistry, Brian (Dr. Brian O. Patrick) for excellent work on crystallography; Elena (Dr. Elena Polishchuk) and Jessie (Dr. Jessie Chen) for all the convenience and countless smiles in the bioservices; Paul (Dr. xlii Zhicheng Xia) and Maria (Dr. Maria Ezhova) for help on NMR data; Yun (Dr. Yun Ling) and the MS labs; the electronic shop; and the mechanical shop. All my collaborators have contributed a lot to my work. I’d like to thank Professor Kuoshyan Lin, Professor Mike Wolf and Professor Urs Hafeli for all the kind guidance and advice; Dr. Jinhe Pan at the BC Cancer Agency for all of his PET/CT imaging help. Cristina (Dr. Cristina Rodríguez-Rodrígue), Christian (Christian Buchwalder), Kathy (Dr. Katayoun Saatchi), for the SPECT imaging; Feng (Dr. Feng Gao), Una (Una Jermilova) and Valery (Dr. Valery Radchenko) at TRIUMF for the radiolabeling help. Si (Dr. Si Zhang), Xiaolei (Dr. Xiaolei Shi) for biological sample preparation, Mike (Mike Tran) for microscope imaging. Thank you to my PhD committee members Professor Russ Algar, Professor Urs Hafeli and Professor Peter Legzdins for their time to read my thesis and report and attend my committee meetings. Thanks to Aidan, Lily S and James (James Cudmore) for proof reading of my report and thesis. Thank you to my friend Leixing, for being a huge support in my life and always keeping me company. Finally, I must thank to my parents for giving me infinite love. I am lucky and proud to be your son. 1 Chapter 1: Introduction 1.1 Medicinal Inorganic Chemistry Figure 1.1. The author’s view of the subject of Medicinal Inorganic Chemistry. Medicinal inorganic chemistry is an interdisciplinary research field, bridging medicinal chemistry and bioinorganic chemistry, which dates back to the fifth century B.C and shows increasing significance in both therapeutic and diagnostic medicine today.1 While metalloenzyme research, the other important area of bioinorganic chemistry, focuses on investigation of the role of metal ions, metalloproteins, and metal ions homeostasis in nature, medicinal inorganic chemistry mainly explores how to use and apply those metals and metal ions, metal ion chelators, as well as metal complexes in medicine. It could be further divided into two main categories: ligands as drugs targeting metal ions in some form, such as those chelators in chelation therapy and metal ion probes (Figure 1.2), and metal-based drugs and imaging agents such as cisplatin for cancer therapy, lithium carbonate and 99mTc sestamibi for imaging.2 (Figure 1.3) Dimercaptosuccinic acid (Pb and Hg poisoning) BAPTA based fluorescent Ca2+ probe Figure 1.2. Drugs targeting metal ions: dimercaptosuccinic acid, a fluorescent calcium probe. Medicinal Chemistry Bioinorganic Chemistry Medicinal Inorganic Chemistry Metal-based Drugs Drug Targeting Metal Ions Metalloproteins Medicinal Organic Chemistry 2 Cisplatin (anti-cancer) Lithium carbonate (bipolar disorder) 99mTc sestamibi (myocardial imaging) Figure 1.3. Metal-based drugs and imaging agents: cisplatin, lithium car`bonate and 99mTc sestamibi. 1.2 Medicinal Imaging In medicine, medical imaging is the process by which physicians evaluate an area of the patients’ body that is not normally visible, using magnetic resonance imaging, ultrasound, X-ray radiographs, computed tomography, and other techniques. As summarized in Table 1.1, each imaging technique has its own advantages and limitations.3 Table 1.1. Comparison of various imaging modalities.a a adapted from the literature3 S.R. = spatial resolution MRI Ultrasound X-ray CT SPECT PET Spatial Resolution (mm) 0.01-0.1 (small animal) 0.5-1.5 (clinical) 0.04-0.1 (small animal) 0.1-1 (clinical) 0.001-1 0.00035-0.05 (micro) 0.2-1 (clinical) 0.5-2 (micro) 7-15 (clinical) 1-2 (micro) 6-10 (clinical) Advantages High S.R.; superb soft tissue discrimination; variable thickness; any plane Widely available; easy to use; portable; real-time Fast imaging; easy method; less motion artefact High S.R.; No superposition of images of overlapping structures High sensitivity; quantitative; tracer amount of probe; physiological change High sensitivity; quantitative; tracer amount of probe; free of background; physiological change Disadvantages Strong magnetic fields disturb activated implants; toxicity from high concentration of contrast agent. Operator dependent; difficult image of bone & lungs Superposition of structures; Ionizing radiation High dose per examination Low S.R.; blurring effect; Ionizing radiation Low S.R.; blurring effect; Ionizing radiation Cost High Low-Medium Low High Medium-High High Radiation source (Energy used) Radio frequency waves (non-ionizing) High frequency sound waves (non-ionizing) X-ray (ionizing) X-ray (ionizing) Gamma-ray (ionizing) Annihilation photos (ionizing) + 3 1.2.1 Magnetic Resonance Imaging (MRI) and Contrast Agents After being introduced into clinical diagnostic imaging by Lauterbur and Mansfield4 in 1973, magnetic resonance imaging (MRI) developed as a fast technique in the past forty years and has become one of the most important clinical imaging tools today. By detecting the interaction of water proton magnetic moments with an applied magnetic field, MRI can generate sectional images and identify different tissues, especially soft tissue with high special resolution without using ionizing radiation. Paramagnetic contrast agents are usually applied to enhance the contrast and differentiation of tissues. Most of the commonly used contrast agents are Gd(III)-based agents, since Gd(III) has a large magnetic moment due to its 4f7 electronic configuration, with all spins parallel. Free Gd(III) ions are highly toxic; however, thus chelating ligands are used to form thermodynamically stable complexes and increase the biocompatibility5. Structures of some of some commonly used Gd complexes with both macrocyclic ligands such as Gd-DOTA, Gd-BT-DO3A and Gd-HP-DO3A, and linear ligands including Gd-DTPA, Gd-BOPTA, Gd-DTPA-BMA are displayed in Figure 1.4. Figure 1.4. Gd(III)-based MRI contrast agents.5 4 Even though those Gd complexes are safer than free Gd3+ ions, a high concentration (>0.1 mM) is required so that NMR signal changes can be detected, and there is still a safety risk in clinical application especially for patients with chronic kidney disease.6 In addition, the requirement of high concentration (low sensitivity) makes them hard to develop into specific targeting imaging agents, for example, as an antibody based imaging agent. Recently, some other alternatives have continually attracted growing interest. One good example is superparamagnetic iron oxide nanoparticles that have been developed as T2 contrast agents with high sensitivity and contrast; however, this application is limited by their size, and the long-term toxicity of nanoparticles is still controversial.7 Another alternative group of agents is manganese (Mn)-based complexes, since Mn(II) (3d5) has relatively similar physical properties with Gd(III) and is eliminated via biliary excretion rather than renal clearance.8-9 Some potential Mn(II) contrast agents are displayed in Figure 1.5.6, 10-11 The challenge of designing better Mn-based contrast agents is to increase the in vivo stability and kinetic inertness of the compound while maintaining the magnetic properties and fast water exchange kinetics.6 Figure 1.5. Mn(II)-based MRI contrast agents.6, 10-11 5 1.2.2 Ultrasound (US) and Contrast Agents Ultrasound (US) imaging is a widely used cheap, non-invasive imaging technique. It can achieve real-time imaging without using ionizing radiation. Like MRI, it has relatively high soft tissue contrast compared with X-ray and computed tomography (CT). One of the most important advantages of ultrasound over other techniques is that it can be developed into a portable device which is quite useful in point of care application. The contrast of ultrasound could also be enhanced by using bubbles as contrast agents; various bubbles with different shells made from hard shell such as polymers, proteins and silica to soft shell such as phospholipid and surfactant, filled with various gases, are developed as contrast agents. Clinically approved contrast agents are summarized below.12 Table 1.2. Ultrasound contrast agents that have/had been clinically approved.a Name First approved Shell material Gas Application (Sample) Producer Optison 1998 Cross-linked serum albumin Octafluoropropane Left ventricular opacification GE Healthcare Sonazoid 2007 Phospholipid Perfluorobutane Myocardial perfusion GE Healthcare Lumason/SonoVue 2001/2004 Phospholipid Sulphur hexafluoride Left ventricular opacification, microvascular enhancement Bracco Diagnositics Definity/Luminity 2001/2006 Phospholipid Octafluoropropane Echoardiography, Liver/kidney imaging (Canada) Lantheus Medical Imaging Imagent/Imavist 2002, withdrawn Phospholipid Perfluorohexane Echoardiography, heart perfusion, tumor/blood flow anomalies Schering AG Echovist 1991, withdrawn Galactose microparticles Air Right heart imaging Schering AG Albunex 1993, withdrawn Sonicated serum albumin Air Transpulmonary imaging Molecular Biosysterms a adapted from the literature12 6 1.2.3 X-ray and CT X-ray radiography is an old and well established technique; it was first used in medical application by John Hall-Edwards in 1896, one year after the discovery of X-rays by W. C. Roentgen. X-ray computed tomography (CT) is relatively young, however, and was invented in 1972 by G. N. Hounsfield. Compared with other imaging techniques, CT can acquire and collect 3D isotropic images with high resolution within several minutes. It is applicable and convenient for imaging bone injuries, lung and chest disease.13 Contrast agents that contain elements of higher atomic numbers are also commonly applied in X-ray and CT to enhance the image quality by increasing the attenuation difference between the target tissue and surrounding tissue and fluids. Iodine, lanthanide and gold nanoparticle based-contrast agents are the main categories studied and used today. Corresponding representatives are displayed in Figure 1.6. Figure 1.6. Samples of X-ray and CT contrast agents (adapted from the literature13). 1.3 Nuclear Medicine Nuclear medicine, a specialty applying radioactive substances in the diagnostic imaging and treatment of disease, is a powerful tool in many fields of medicine such as heart disease, cancer, and neurodegenerative disorders. It is sometimes referred as \"endo-radiology\" since it records radiation emitting from within the body rather than radiation generated by external sources like X-7 rays (Figure 1.7).14 By using radiopharmaceuticals that are specific for a particular disease process or receptor, nuclear medicine has the capacity of imaging the extent of a disease based on the functional and physiological changes at the molecular level rather than anatomical physical changes. Thus, in some diseases nuclear medicine studies can diagnose medical problems at an earlier stage compared with other diagnostic tests.15-17 Figure 1.7. Depiction of SPECT, PET and X-ray.14 1.3.1 Nuclear Imaging Single photon emission computed tomography (SPECT) and positron emission tomography (PET) are the two major diagnostic imaging techniques used in nuclear medicine. As shown in Figure 1.7, the common SPECT system uses a typical gamma camera to detect the gamma rays emitted by γ-emitting radionuclides such as 99mTc, 123I, 67Ga, and 111In, and convert them into electrical signals which are then processed by an online computer to form a 3D image. In the PET system, a cylinder of detectors is used to detect the coincidence events of the two 511-keV photons emitted in opposite directions after annihilation of a positron from a β+ emitter meeting an electron in the medium. The commonly used PET radionuclides are 11C, 13N, 15O, 18F, 64Cu, 68Ga, and 82Rb. Some of the widely used radioisotopes for SPECT and PET imaging are summarized below in Table 1.3. 8 Table 1.3. Widely used radioisotopes for SPECT & PET imaging.a Radionuclide Daughter isotopes t 1/2 Imaging modality Clinical use 11C 11B 20.4 min PET (β+) yes 18F 18O 109.7 min PET (β+) yes 64Cu 64Ni 12.8 h PET (β+) yes 67Ga 67Zn 78.2 h SPECT (γ, 184 keV) yes 68Ga 68Zn 68 min PET (β+) yes 82Rb 82Kr 1.273 min PET (β+) yes 99mTc 99Tc 6.01 h SPECT (γ, 140.5 keV) yes 111In 111Cd 67.4 h SPECT (γ, 171 keV) yes 123I 123Te 13.22 h SPECT (γ, 157 keV) yes a adapted from the literature18 While SPECT only provides one piece of spatial information per decay event, PET imaging provides two pieces of spatial information. Therefore, generally, clinical PET has a higher spatial resolution and shorter scan time, but SPECT is currently more widely used due to the SPECT modality being more established in hospitals and the availability of abundant and cheap SPECT radio-tracers. 1.3.2 Nuclear Medicine Therapy Radioisotope therapy can be accomplished by using high energy β- emitters (e.g. 67Cu, 131I, 90Y, 153Sm, 177Lu, 186Re, 188Re, 198Au) and α-emitters (e.g. 223Ra, 224Ra, 225Ac). Low energy Auger electrons (e.g. 111In) have also been studied due to their ionizing damage to cancer cells. In general, the therapeutic effects of radiopharmaceuticals rely on the tissue-destructive power of ionizing radiation.19 Some widely used radioisotopes for therapy are summarized in Table 1.4. 9 Table 1.4. Widely used radioisotopes for therapy.18 1.3.3 Radiometal Isotopes Among all the radioisotopes used to form radiopharmaceuticals, those of radiometals have drawn great interest since they have many advantages such as longer half-lives which allow for use with the larger proteins and peptides, combined imaging and therapeutic capacity, as well as convenient sources and easier labeling methods.19 99mTc is the most widely used and commercially available radiometal; however, the recent worldwide shortage of 99Mo/99mTc due to the frequent shutdown of two of the world’s major 99Mo producing reactors suggests the need for the use of other radioisotopes such as 67Ga, 111In, 86Y, and 177Lu, and the development of new efficient corresponding chelators for those metals. A summary of some popular radiometal isotopes is listed in Table 1.5.20 Radionuclide Daughter isotopes t 1/2 Emitted particles Avg. energy Range (mm) 64Cu 64Zn 12.8 h β- 573 keV ~ 2.1 67Cu 67Zn 61.9 h β- 141 keV 0.26 90Y 90Zr 64.1 h β- 932 keV 4.0 177Lu 177Hf 6.7 d β- 133 keV 0.23 186Re 186Os 3.8 d β- 349 keV 1.1 188Re 188Os 17 h β- 764 keV 3.1 224Ra 220Rn 3.6 d α 5.45 MeV ~ 0.06 225Ac 207Bi 7.2 h α 6.79 MeV 0.06 10 Table 1.5. Properties of some popular radiometal isotopes.20 Isotope T1/2 (h) Decay mode Energy (keV) Production Method 44Sc 3.9 β+ (94%) EC (6%) β+, 1474 γ, 1157 44Ti/44Sc generator 47Sc 80.2 β- (100%) β-, 441, 600 γ, 159 47Ti(n,p)47Sc 60Cu 0.4 β+ (93%) EC (7%) β+, 3920, 3000, 2000 Cyclotron, 60Ni(p,n)60Cu 61Cu 3.3 β+ (62%) EC (38%) β+, 1220, 1150, 2000 Cyclotron, 60Ni(p,n)60Cu 62Cu 0.16 β+ (98%) EC (2%) β+, 2910 62Zn/62Cu generator 64Cu 12.7 β+ (19%) EC (41%) β- (40%) β+, 656 Cyclotron, 64Ni(p,n)64Cu 67Ga 78.2 EC (100%) γ, 93, 184, 300 Cyclotron, 68Zn(p,2n) 67Ga 68Ga 1.1 β+ (90%) EC (10%) β+, 1880 68Ge/68Ga generator 86Y 14.7 β+ (33%) EC (66%) β+, 1221 Cyclotron, 86Sr(p,n) 86Y 90Y 64.1 β- (100%) β-, 2280 90Zr(n,p) 90Y 89Zr 78.5 β+ (23%) EC (77%) β+, 897 89Y (p,n) 89Zr 111In 67.2 EC (100%) γ, 245, 172 Cyclotron, 111Cd(p,n) 111In 114mIn 49.5d EC (100%) γ, 190 Cyclotron, 114mCd(p,n) 114In 116mCd(p,3n) 114In 177Lu 159.4 β- (100%) β-, 177, 385, 498 γ, 112, 208 Cyclotron, 176Lu(n,γ) 177Lu 212Bi 1.1 α (36%) β- (64%) α, 6050 β-, 6089 228Pb/212Pb generator 213Bi 0.76 α (2.2%) β- (97.8%) α, 5549 β-, 5869 228Th/212Pb generator 225Ac 240 α (100%) α, 5600-5830 226Ra(p,2n) 225Ac 11 Figure 1.8. Depiction of a chelating agent functionalized with a monoclonal antibody.21 1.4 Radiometal Chelators As shown in Figure 1.8, most of the radiometal chelators used in nuclear medicine are “bifunctional chelating agents” which contain two functional moieties: a strong metal binding unit to sequester the radionuclide and a chemically reactive functional group which is ready for bioconjugation to a targeting biovector.21-22 Generally, a good radiometal chelate should have the following properties:16, 22 a) Thermodynamic stability and kinetic inertness: the complex should be as stable as possible, with kinetic inertness more important than thermodynamic stability. Low in vivo inertness will lead to decomposition of the complex and release of the radiometal ion which then results in an unnecessary radiation dose, decreased imaging quality or therapeutic effect, and potential toxicity. b) Quick complexation under mild conditions: complexes should be formed quickly at low temperatures and low concentration of the ligand. This is especially important for short half-life isotopes and environmentally sensitive biovectors. 12 c) Bioconjugation position or capacity: bifunctional chelators should have a reactive functional group (e.g. activated esters, isothiocyanates, terminal alkynes or azides (for click chemistry), or free primary amines) that could bind to the targeting vector (e.g. biomolecule) or carrier (e.g. nanoparticles) without affecting the complexation behaviour of the chelating unit. d) Synthetic accessibility: synthesis of the chelate should be quick, easy, and cheap on a practical scale. Figure 1.9. Macrocyclic chelating ligands: DOTA, NOTA, CB-DO2A and PCTA. 1.4.1 Macrocyclic Chelators The metal chelators could be divided into two categories based on the structure, macrocyclic chelators and acyclic chelators. 1,4,7,10-Tetraazacyclododecanetetraacetic acid (DOTA) is the most widely used macrocyclic chelator (Figure 1.9) in radiopharmaceuticals and is the current ‘‘gold standard’’ for several different isotopes, including 111In, 177Lu, 86/90Y, and 44/47Sc.20, 23 It was designed with closed chains and usually has higher thermodynamic stability in metal-ion-binding compared to acyclic chelators, because of the macrocycle effect.13 The size of the chelate ring can be adjusted to match the size of metal ions, for example, 1,4,7-triazacyclononanetriacetic acid (NOTA) is more stable for 68Ga labeling because of its smaller ring size. Sometimes, the closed chain can be further reinforced by an extra bridge such as in the structure of CB-DO2A, or by 13 adding a more rigid moiety such as pyridine in the structure of PCTA to enhance the stability. This constrained geometry results, however, in slower complexation kinetics and often requires long labeling times and high reaction temperatures that are not suitable for isotopes with a short half-life time and thermally sensitive biovectors. Figure 1.10. Acyclic chelating ligands: EDTA, DFO, DTPA, CHX-A\"-DTPA, H2dedpa, H4octapa. 1.4.2 Acyclic Chelators Acyclic (non-macrocyclic) chelators (Figure 1.10), generally have faster complexation kinetics of only a few minutes, and do not require high temperature, and thus DTPA and DFO have been used widely in radiometal labeling. The disadvantage of open chain structures, in most cases, is a decreased thermodynamic stability compared with macrocyclic chelators. One strategy to overcome this problem is increasing the rigidity. For example, CHX-A\"-DTPA has a rigid cyclohexane modification and showed increased stability. Our group’s recent work used the chelation strengths of the picolinate motif and proved that those carefully designed acyclic chelators (H2dedpa, H4octapa) can achieve fast complexation kinetics at room temperature while keeping thermodynamic stability and in vivo kinetic inertness comparable to macrocyclic chelates.24-26 14 1.5 Biovectors Biovectors are molecules that have a high specific affinity for a biological target of interest, for example a surface receptor more highly expressed on a cancer cell, and thus they may accomplish site-specific delivery of the radioactive complex. Typical biovectors include small bioactive molecules (e.g. folate, biotin),27-28 peptides and proteins (e.g. RDG, monoclonal antibody),29 oligonucleic acid (e.g. DNA or RNA aptamer),30 and virus vectors (e.g. HVJ-Es)31 etc. (Figure 1.11) Folic acid RGD peptide Aptamer Antibody Viral vector Figure 1.11. Classes of biovectors. 15 1.6 Bioconjugation Bioconjugation is the chemical strategy used to attach the bifunctional chelate or complex to biovectors in nuclear medicine. It is of great importance to the function and quality of the final imaging agents. Several principles to be considered are listed below: a) The reaction should occur rapidly at mild conditions to decrease the risk of biovector degradation. b) The reaction should be specific to the available group on the biovector and should not change important functions of biovectors. c) The conjugation should not decrease the binding affinity of the chelate to metal ions and the stability of the formed complex. d) The established links should be robust and have a good in vivo stability. Some of the commonly used bioconjugation strategies are summarized in Table 1.6. 1.7 Thesis Overview This thesis presents studies on the design and synthesis of acyclic (non-macrocyclic) chelators for the application of radiometals, (e.g. 64Cu, 67/68Ga 111In, 90Y) in nuclear medicine. These studies follow our group’s previous development and studies on picolinic acid-based chelators (“pa” family). Chapter 2 focuses on refining the synthesis of several “pa” chelators and developing novel possible bifunctional sites to expand their application. Chapter 3 explores the synthesis of H2hox, an “ox” (8-hydroxyquinoline) based acyclic chelator and its application in 68Ga labeling. 16 Table 1.6. Widely used bioconjugation strategies. 17 Chapter 4 describes the synthesis, characterization, solution study of H2CHXhox, a cyclohexane reinforced derivative of the lead chelator H2hox, with Ga3+ an Cu2+. In vivo imaging studies are also included in this Chapter and lead to the preparation of more lipophilic halogenated derivatives, H2Br4hox and H2CHXBr4hox, for potential myocardial perfusion imaging. In Chapter 5, H2C3hox is prepared by adding one more carbon to the backbone of H2hox to obtain a 6-membered chelate ring, and the subsequent effect on the stability of the formed metal complex is explored. Chapter 6 reports H4octox, a novel versatile bimodal octadentate acyclic chelator for large metal ions (Y3+, In3+, La3+, Lu3+and Gd3+) in medicinal inorganic chemistry. Chapter 7 investigates the biological application of a new family of medium-sensitive bichromatic flexible fluorescent dyes (FlexFluor) in the collaborative work with Dr. Yang Cao. Chapter 8 summarizes some other studies not fully reported in this thesis and gives some perspectives for future work with the “ox” family. 18 Chapter 2: H4octapa, New Synthetic Route and Derivatives This chapter contains an adaptation of published work, and is reproduced in part from María de Guadalupe Jaraquemada-Peláez, Xiaozhu Wang, Thomas J. Clough, Yang Cao, Neha Choudhary, Kirsten Emler, Brian O. Patrick and Chris Orvig. H4octapa: synthesis, solution equilibria and complexes with useful radiopharmaceutical metal ions. Dalton Trans. 2017, 46, 14647-14658. 2.1 Introduction Radiometal isotopes are used extensively in diagnosis, imaging and treatment, in which the targeted delivery of radiation to a specific disease site is desired.32-33 Ideally, a radiopharmaceutical will accumulate at a disease site, such as a cancerous tumour, with high specificity, allowing imaging or therapy of the target region to be achieved whilst minimising radiation dose to normal surrounding tissues.33-34 Bifunctional chelating ligands (BFC) that have the ability to effectively radiolabel a variety of different radiometals are highly desirable - changing the radiometal bound to the chelator allows the resulting radiopharmaceutical to be useful either for imaging or therapy.20 When developing a chelator for radiopharmaceuticals, complexes formed must be kinetic and thermodynamically stable in vivo (i.e. in an aqueous environment at physiological pH), and retain their chemical integrity on exposure to the vast array of biological processes present in biological systems.20, 35-36 Kinetic inertness is of primary importance because in a biological environment the metal chelate is in low concentration and therefore susceptible to different competition reactions: transmetallation by endogenous metals or transchelation by native chelators.20, 34, 36-37 Metal complexes with macrocyclic chelating ligands exhibit both high thermodynamic stability and kinetic inertness, and have been used extensively in clinical medicine. The macrocyclic chelators 19 1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid (DOTA) and 1,4,7-triazacyclononane-1,4,7-triacetic acid (NOTA) exploited by Maecke,38-39 are currently the “gold standards” for chelation of radiometals including 64Cu, 68Ga, 111In, 86/90Y and 177Lu.20 Their constrained geometries and preorganized coordination sites confer a generally higher kinetic stability than that of acyclic chelators;40-43 however, the requirement for long reaction times and heating (40-90 °C, 30-150 min) limit their application to isotopes with short half-lives and thermally sensitive biovectors.20, 36, 41-46 An ideal chelator would combine the high thermodynamic stability and kinetic inertness of a macrocyclic chelator with the rapid and quantitative incorporation of radiometals under mild conditions exhibited by acyclic (open-chain) chelators.20, 36-37 Therefore, many efforts have been taken to increase the stability of such acyclic chelators to overcome the drawbacks of both binding modalities. In the past few years, our research group has focused on the development and derivatisation of a family of acyclic chelators based on a pyridinecarboxylate scaffold (pa) (Scheme 2.1). The ligand system is extremely versatile and has been shown to coordinate an array of metals with medical applications.23-26, 37, 45, 47-53 H4octapa, first reported in 2004 has been shown to form extremely stable coordination complexes with a variety of metals, including In(III), Y(III), Lu(III), Gd(III), La(III) and Cu(II). Effective radio-labelling has been demonstrated under exceptionally mild conditions (RT, < 15 min) and target complexes explored for their potential applications in SPECT and MRI imaging.20, 24, 37, 47-48, 50, 54-55 The success of previous studies indicates that it is extremely versatile and piques interest for potential applications with other metal ions of medical importance such as Sm(II), Dy(III) and Yb(III). Dy-DTPA-BMA has been investigated as an MRI contrast agent for delineation of myocardial infarction.56-60,153Sm(III)-EDTMP (Quadramet®), is one example of a U.S. FDA-approved radiopharmaceutical for routine use in humans - it is used 20 for bone pain palliation in skeletal metastatic cancer and its γ decay allows external imaging of the radioisotope distribution.33, 61-63 153Sm has three β-emissions (30% 0.64 MeV, 50% 0.71 MeV and 20% 0.81 MeV) and γ-emission (28% 103 keV with a half-life of 1.96 days). Attention to lower-energy β-emitting radionuclides is of particular interest in order to maximize the radiation dose to tumor cells and to reduce the side-effects of irradiation to normal tissues, e.g. minimize the degree of bone marrow suppression.64-65 175Yb is a radionuclide with potential for the development of radiotherapeutic compounds; 175Yb undergoes β-decay (Emax = 480 keV) to stable 175Lu with a half-life of 4.2 days. Studies of 175Yb complexes incorporating polyaminomethylene phosphonic acid ligands have been shown to have potential to be used in palliative treatment of painful bone metastases.66 In this work we have evaluated the possibility of the use of Sm(III), Dy(III) and Yb(III) metal ions with H4octapa for development of radiopharmaceuticals in terms of thermodynamic stability due to the highly thermodynamic stability and in vivo and in vitro kinetic inertness of the previously reported H4octapa and H4octapa analogues complexed with the radionuclides 111In and 177Lu.24, 37, 45, 54 Herein, an improved synthetic route to H4octapa was reported, with substantially higher cumulative yield compared to that of previous methods, increasing the accessibility to this ligand. A crystal structure of the La-octapa complex was obtained and showed that the metal ion La3+ is ten coordinated by the octadentate [octapa]4- ligand and two water molecules. The thermodynamic stability of complexes formed between H4octapa and Sm(III), Dy(III), and Yb(III) metal ions was evaluated through solution equilibria studies. Density Functional Theory (DFT) simulations of the [Ln(octapa)(H2O)]- coordination structures and their molecular electrostatic potential (MEP) gave further insight into the speciation of the metal complexes in solution. 21 Scheme 2.1. Some of the pyridinecarboxylate scaffold (pa) chelators synthesized previously, and the EDTMP ligand discussed in this work. 2.2 Experimental 2.2.1 Materials and Methods All solvents and reagents were purchased from commercial suppliers (TCI America, Sigma Aldrich) and were used as received unless otherwise noted. 1H and 13C NMR spectra were recorded at room temperature on Bruker AV300 instruments; the NMR spectra are expressed in the δ (ppm) scale and were referenced by the residual solvent signal of the deuterated solvent. Assignments of peaks in the NMR spectra are approximate. Low-resolution mass spectrometry was performed on a Waters ZQ spectrometer equipped with an electrospray and chemical ionization (ESCI) source; high resolution spectra were collected on a micromass LCT instrument, both at the Department of Chemistry, University of British Columbia. The HPLC system used for purification of ligands and 22 precursors consisted of a Waters 600 controller equipped with a Waters 2487 dual λ absorbance detector connected to either a Phenomenex Jupiter C 18 100 mm × 4.6 mm 5 micron analytical column, or a Phenomenex Synergi Hydro-RP 80A 250 mm × 4.60 mm 4 micron analytical column. 2.2.2 Ligand Synthesis and Characterization. 2.2.2.1 New Synthetic Route to H4octapa Ethylenediaminediacetic acid dimethyl ester dihydrochloride (2.1) Compound 2.1 was synthesized with slight variation from the literature.67 To a solution of ethylenediamine-N,N'- diacetic acid (2.64 g, 15 mmol) in methanol (100 mL), SOCl2 (4.35 mL, 60 mmol) was added dropwise slowly at 0 °C. The mixture was then refluxed at 80 °C for 16 h. Solvent then was removed by rotary evaporation and the residue was dried under vacuum to obtain the product as a white crystalline powder (4.12 g, 99%). 1H NMR (300 MHz, D2O, 25 °C) δ: 4.09 (s, 4H), 3.86 (s, 6H), 3.52 (s, 4H). 13C NMR (75 MHz, DMSO, 25 °C) δ: 167.4, 53.2, 47.2, 43.1. HR-ESI-MS calcd. for [C8H16N2O4 + H]+: 259.1188; found: 259.1190. Methyl 6-(bromomethyl) picolinate (2.2) Compound 2.2 was obtained as a white, crystalline solid following our previous literature24 and characterized by 1H NMR. 1H NMR (300 MHz, CDCl3, 25 °C) δ: 8.07 (d, 1H, J = 7.8 Hz), 7.88 (t,1H, J = 7.8 Hz), 7.70 (d,1H, J = 7.8 Hz), 4.64 (s, 2H), 4.02 (s, 3H).13C NMR (100 MHz, CDCl3, 25 °C) δ: 165.3, 157.4, 147.6, 138.2, 127.0, 123.8, 53.1, 33.1. LR-ESI-MS calcd. for [C8H981BrNO2 + H]+: 232.0; found: 232.2. 23 Dimethyl-6,6'-((3,10-dioxo-2,11-dioxa-5,8-diazadodecane-5,8-diyl)bis(methylene))-dipicolinate (2.3) To a solution of methyl 6-(bromomethyl)picolinate (2.2) (460 mg, 2.0 mmol) in dry tetrahydrofuran (40 mL) was added sodium carbonate (∼ 1.5 g) and compound 2.1 (277 mg, 1.0 mmol). The reaction mixture was stirred at 50 °C overnight. Sodium carbonate was removed by filtration, and the crude reaction mixture was concentrated by rotary evaporation. The crude oil was purified by column chromatography (CombiFlash® Rf automated column system; 24 g HP silica; A: dichloromethane, B: methanol, 100% A to 30% B gradient) to afford compound 2.3 as colorless oil (417 mg, 83%). 1H NMR (300 MHz, CDCl3, 25 °C) δ: 7.94 (d, 2H), 7.88 (t, 2H), 7.59 (t, 2H) 3.91 (s, 6H), 3.75 (s, 4H), 3.47 (s, 6H), 3.05 (s, 4H), 2.58 (s, 4H). LR-ESI-MS calcd. for [C24H30N4O8 + Na]+: 525.2; found: 525.2. H4octapa, N,N′-bis(6-carboxy-2-pyridylmethyl)ethylenediamine-N,N′-diacetic Acid (2.4) The mixture containing dimethyl-6,6'-((3,10-dioxo-2,11-dioxa-5,8-diazadodecane-5,8-diyl)-bis(methylene))-dipicolinate (2.3) (1.00 g, 2.00 mmol) was dissolved in aqueous HCl (6 M, 10 mL) and left to stir under reflux at 140 °C. After 20 h, the solvent was reduced by half in vacuo and a white precipitate appeared. The solid was filtered out and washed with HCl (6 M, 10 mL) and ice-cold diethyl ether (10 mL) to give the hydrated HCl salt of 2.4 as a white solid (0.83 g, 1.25 mmol, 63%). 1H NMR (300 MHz, D2O, 25 °C) δ: 8.16 (t, 2H, J = 7.7 Hz), 8.10 (dd, 2H, J = 7.7, 1.4 Hz), 7.76 (dd, 2H, J = 7.7, 1.4 Hz), 4.53 (s, 4H), 3.92 (s, 4H), 3.57 (s, 4H).13C NMR (100 MHz, D2O, 25 °C) δ: 171.2, 165.5, 151.4, 142.7, 128.3, 125.9, 57.6, 55.3, 51.3. HR-ESI-MS calcd. for [C20H22N4O8 + H]+: 447.1516, found: 447.1519. Elemental analysis: 24 calcd. % for H4octapa·4HCl·3H2O (C20H22N4O8·4HCl·3H2O): C 37.20, H 4.99, N 8.66; found: C 37.62, H 4.86, N 8.61. Data agree with the literature.24 2.2.2.2 Synthesis of Na[La(octapa)] Na[La(octapa)] (2.5) H4octapa·4HCl·3H2O (0.02 g, 0.03 mmol) was dissolved in HCl (0.1 M, 1 mL). La(NO3)3·6H2O (0.03 g, 0.07 mmol) was added. The reaction mixture was brought to pH 4 through dropwise addition of NaOH solution (0.1 M) and left to stir at room temperature. After 1 h, MS indicated the formation of the anion so the solvent was removed in vacuo, giving a white solid. The complex was only used in NMR and crystallography to illustrate the connection of the ligand to the metal ion. 1H NMR (400 MHz, D2O, 25 °C) δ: 7.97 (m, 2H), 7.56 (d, 1H), 4.35 (d, 1H), 3.75 (d, 1H), 3.40 (d, 1H), 3.18 (d, 1H), 3.09 (d, 1H), 2.61 (d, 1H). 13C NMR (300 MHz, D2O, 25°C) δ: 179.7, 173.6, 157.2, 151.0, 140.4, 126.4, 124.3, 61.2, 58.2, 53.6. HR-ESI-MS calcd. for [C20H18LaN4O8]-: 581.0188; found: 581.0184. 2.2.2.3 Synthesis of H4octapa Derivatives with Bromine Substituted Pyridyl Group Dimethyl 4-bromopyridine-2,6-dicarboxylate (2.6) Compound 2.6 was synthesized with slight variation from the literature.68 Chelidamic acid (4.2 g, 23 mmol) and PBr5 (37 g, 86 mmol) were mixed and heated to 80 °C for 5 hours. The mixture was then cooled to room temperature, and chloroform (100 mL) was added to obtain a deep red solution which was then filtered. The filtrate was cooled on ice and methanol (100 mL) was added slowly. Crystallization was induced with scratching and the white solid product was collected by 25 filtration. The filtrate was concentrated in vacuo, cooled on ice, and white crystals were collected by filtration again (5.97 g, 95%). 1H NMR (300 MHz, CDCl3, 25 °C) δ 8.44 (s, 2H), 4.01 (s, 6H). Methyl 4-bromo-6-(bromomethyl)picolinate (2.7) Compound 2.7 was synthesized with slight variation from the literature.68 Compound 2.6 (4.91 g, 18 mmol) was dissolved in a mixture of MeOH and DCM (4:1, 200 mL total) and cooled to 0 C. To this, sodium borohydride (1.02 g, 27 mmol, 1.5 eq) was slowly added over 1 h in portions. The reaction was monitored using TLC (silica, 100% EtOAc) and quenched with a saturated aqueous solution of NaHCO3 (100 mL). Dichloromethane (80 mL) was added and the organic layer was separated from the mixture. Methanol in the aqueous phase was removed via rotary evaporation, and the remaining solution was extracted with dichloromethane (4  50 mL). All organic phases were combined, dried with MgSO4, filtered and the solvent was removed in vacuo, giving the crude product 2.7 (4.62 g). All of the crude product was suspended in dry CHCl3 (220 mL) at 0 °C, and PBr3 (2.84 mL, 30 mmol) was slowly added. The reaction mixture was then allowed to warm to room temperature and stirred for a further 3.5 h. After that, the mixture was cooled to 0 °C, quenched with an aqueous solution of K2CO3, and extracted with dichloromethane (4  50 mL). The organic phase was dried with MgSO4, filtered and the solvent was removed in vacuo. The crude product was purified using column chromatography (CombiFlash Rf automated column system; 40 g HP silica; A: hexane, B: EtOAc, 100% A to 50% B gradient) and the pure product was obtained as a white solid (4.06 g, 73%). 1H NMR (300 MHz, CDCl3, 25 °C) δ 8.17 (d, J= 1.5 Hz, 1H), 7.82 (d, J= 1.5 Hz, 1H), 4.56 (s, 2H), 3.98 (s, 3H). 13C NMR (75 MHz, CDCl3, 25 °C) δ 163.7, 158.1, 148.0, 134.1, 129.7, 127.4, 26 77.0, 76.6, 76.1, 52.9, 31.6. HR-ESI-MS calcd. for [C8H7Br2O2 + Na]+: 331.8721; found: 331.8682. Dimethyl,6'-((3,10-dioxo-2,11-dioxa-5,8-diazadodecane-5,8-diyl)bis(methylene))bis(4-bromo-picolinate) (2.8) To a solution of methyl 4-bromo-6-(bromomethyl)-picolinate (2.8) (618 mg, 2.0 mmol) in dry tetrahydrofuran (40 mL) was added sodium carbonate (1.5 g) and compound 2.1 (277 mg, 1.0 mmol). The reaction mixture was stirred at 50 °C overnight. Sodium carbonate was removed by filtration, and the crude reaction mixture was concentrated by rotary evaporation. The crude oil was purified by column chromatography (CombiFlash Rf automated column system; 24 g HP silica; A: hexane, B: EtOAc, 100% A to 50% B gradient) to afford compound 3 (580 mg, 88%). 1H NMR (300 MHz, CDCl3, 25 °C) δ8.09 (d, J = 1.7 Hz, 1H), 7.92 (d, J = 1.7 Hz, 1H), 3.95 (s, 2H), 3.94 (s, 3H), 3.64 (s, 3H), 3.40 (s, 2H), 2.79 (s, 2H). 3.94 (s, 3H), 3.64 (s, 3H), 3.40 (s, 2H), 2.79 (s, 2H). 13C NMR (75 MHz, CDCl3, 25 °C) δ 171.4, 164.6, 162.1, 148.2, 134.5, 129.1, 127.1, 60.0, 55.0, 53.1, 52.2, 51.5. HR-ESI-MS calcd. for [C24H28Br2N4O8 + H]+: 661.0332; found: 661.0345. 2.2.3 X-ray Crystallography Crystals of ligands were obtained in its fully protonated form by slow evaporation of an aqueous solution containing H4octapa after addition of HCl (3 M). Crystals of La(III) complex was obtained by slow evaporation of a mixture containing La(NO3)3 and H4octapa in 1.6:1 metal to ligand molar ratio at pH ~ 4 in H2O. A crystal suitable for single crystal X-ray crystallography having approximate dimensions of 0.03 x 0.14 x 0.15 mm was mounted on a glass fiber. All measurements were made on a Bruker APEX DUO diffractometer equipped with a TRIUMPH 27 curved-crystal monochromated Mo-Kα radiation. For the La-octapa structure, data were integrated for both components, including both overlapped and non-overlapped reflections. The material crystallizes as a two-component twin with components one and two related by a 179.7º rotation about the (0 0 1) real axis. Data were integrated for both components, including both overlapped and non-overlapped reflections. In total 28327 reflections were integrated (1330 from component one only, 1307 from component two only, 25690 overlapped). Data were collected and integrated using the Bruker SAINT69 software packages. Corrections for absorption effects were made by using the multi-scan technique (TWINABS).70 The structure was solved by direct methods71 using non-overlapped data from the major twin component. Subsequent refinements were carried out using an HKLF 5 format data set containing complete data from the major twin component and overlaps from the minor twin component. There are at least 5 sites within the asymmetric unit occupied by solvent water molecules. There was residual electron density that would not be reasonably modelled within the asymmetric unit, and thus the PLATON/SQUEEZE72 program was used to generate a data set ‘free’ of solvent at that site. All non-hydrogen atoms were refined anisotropically. All hydrogen atoms, including O-H hydrogen atoms, were placed in calculated positions. All refinements were performed using the SHELXL-201672 via the OLEX273 interface. Further details are available in the Appendix (Table A.1). 2.2.4 DFT Calculations All calculations were performed using the Gaussian 09 package (Revision D.01).74 Full geometry optimizations of the [Ln(octapa)(H2O)]-2H2O (where Ln = La, Sm, Dy and Yb) were performed with the TPSSh exchange-correlation functional75 in aqueous solution using the polarizable continuum model (PCE).76 Geometry optimizations were carried out using the 6-31G(d,p) basis set on first and second row elements, and the large-core quasirelativistic effective 28 core potential (LCRECP) and corresponding [5s4p3d]-GTO valence basis sets for the lanthanide atoms.77-78 The input coordinate of atoms were adapted from a previous reported data of the Gd analogues,79 and no constraints on symmetry were imposed during the geometry optimization. The resulting geometries showed no imaginary frequencies and thus were confirmed to be minima on the potential energy surfaces. 2.2.5 Solution Thermodynamics Potentiometric titrations were carried out using a Metrohm Titrando 809 equipped with a Ross combined electrode and a Metrohm Dosino 800. The titration apparatus consisted of a 20 mL and 25 ºC thermostated glass cell and an inlet-outlet tube for nitrogen gas (purified through a 10% NaOH solution) to exclude any CO2 prior to and during the course of the titration. The electrode was calibrated daily for hydrogen ion concentrations using a standard HCl as described before80 and the results were analysed with the Gran procedure81. Solutions were titrated with carbonate-free NaOH (0.16 M) that was standardized against freshly recrystallized potassium hydrogen phthalate. Protonation equilibria of the ligand were studied by titrations of a solution containing H4octapa 6.61  10-4 M at 25 ºC and 0.16 M NaCl ionic strength using a joined potentiometric-spectrophotometric procedure.82-83 Spectra were recorded in the 200–450 nm spectral range with a 0.2 cm path length optic dip probe connected to a Varian Cary 60 UV/Vis spectrophotometer. The last two ligand protonation equilibria were studied via UV-Vis spectrophotometry on a set of solutions at the same ligand concentration ([H4octapa] = 1.28  10-4 M) containing different amounts of HCl. 1H NMR measurements on a set of H4octapa solutions at 1 mM ligand concentration in a mixture of H2O-D2O (90%-10%) at 25 °C in 5 mm NMR tubes were carried out in order to verify the deprotonation events on the pyridine nitrogen atoms. The equilibrium H+ concentration in the UV and 1H NMR in batch titration procedure at low pH solutions (0 ≤ pH ≤ 29 2) was calculated from solution stoichiometry, not measured with a glass electrode. For the solutions of high acidity, the correct acidity scale H0 was used.84 In the study of complex formation equilibria, the determination of the stability constants of Ln(Hoctapa) species was carried out by two different approaches. The first approach used UV-vis spectrophotometric measurements on a set of solutions containing 1:1 metal to ligand molar ratio ([H4octapa] = [Ln]3+ ~ 1.51 x 10-4 M) and different amounts of HCl in the spectral range 200-400 nm at 25 °C and 1 cm path length. The molar absorptivities of all the protonated species of H4octapa were included in the calculations. The second approach used competition pH-potentiometric titrations with ttha6- as the ligand competitor, where the composition of the solutions was [Ln] ~ 1.14 x 10 -3 M, [H4octapa] ~ 3.84 x 10 -4 M and [ttha6-] ~ 6.19 x 10 -4 M at 25 °C and I = 0.16 M NaCl. Protonation constants of the ttha6- ligand and metal stability constants of the complexes formed with Sm(III), Dy(III) and Dy(III) metal ions were determined by pH-potentiometric titration. The metal to ligand molar ratios were 1:1 and 2:1 with [ttha6-] = 1  10 -3 M. Direct pH-potentiometric titrations of the Ln3+/H4octapa systems were also carried out. Metal solutions were prepared by adding the atomic absorption (AA) standard metal ion solutions to a H4octapa solution of known concentration in the 1:1 metal to ligand molar ratio. The exact amount of acid present in the samarium, dysprosium and ytterbium standards was determined by Gran’s method81 titrating equimolar solutions of either Sm(III), Dy(III) or Yb(III) and Na2H2-EDTA. Ligand and metal concentrations were in the range of 0.6-0.8 mM. Each titration consisted of 100-150 equilibrium points in the pH range 1.8-11.5, equilibration times for titrations were 2 min for pKa titrations and up to 5 min for metal complex titrations. At least two replicate titrations were performed for each individual system. Relying on the stability constants obtained by the two different methods for the species Ln(Hoctapa), the fitting of the direct potentiometric titrations was possible and, in addition to the stability constant 30 of the [Ln(octapa)]- species, the stability constant of a [Ln(octapa)(OH)]2- species was also evaluated. All the potentiometric measurements were processed using the Hyperquad2013 software85 while the obtained spectrophotometric data were processed with the HypSpec86 program. Proton dissociation constants corresponding to hydrolysis of Sm(III), Dy(III) and Yb(III) aqueous ions included in the calculations were taken from Baes and Mesmer.87 The overall equilibrium (formation) constants log β referred to the overall equilibria: pM + qH + rL ⇆ MpHpLr (the charges are omitted), where p might also be 0 in the case of protonation equilibria and q can be negative for hydroxide species. Stepwise equilibrium constants log K correspond to the difference in log units between the overall constants of sequentially protonated (or hydroxide) species. 2.3 Results and Discussion 2.3.1 Synthesis of H4octapa The first synthesis of H4octapa was published by Platas-Iglesias et al. in 2004 50 and, because of the importance of this ligand, it has been improved by many alternative routes ever since.24, 47-48 Currently, the two most successful routes utilize a similar synthetic strategy; the precursor 6,6’-((ethane-1,2-diylbis(azanediyl))bis-(methylene))dipicolinic acid is synthesized first either by reductive amination or through involvement of 2-nitrobenzenesulfonamide (nosyl) protecting groups, before N-alkylation with tert-butyl-2-bromoacetate. Both strategies achieve 30 to 40% cumulative yield over five steps. Herein, the new vastly improved three step synthetic route shown in Scheme 2.2 was reported. Ethylenediamine-N,N'-diacetic acid (EDDA) is first protected in the form of a methyl ester group, using thionyl chloride in methanol, to give EDDA dimethyl ester as a hydrogen chloride salt, 2.1. The backbone is then conjugated with the methyl 6-(bromomethyl) picolinate arm 2.2 to form compound 2.3. Deprotection of methyl ester with HCl (6 M) generates 31 the final product, H4octapa 2.4. The cumulative yield for the new route is 50%, over three steps. Scheme 2.2. Improved synthesis of the ligand H4octapa. a) SOCl2, MeOH, reflux overnight, 99%; b) Na2CO3, THF, 50 °C, 24 h, 83%; c) HCl (6 M), 63% (cumulative yield of ∼ 50% over three steps). 2.3.2 Metal Complexes and X-ray Crystallography Crystals of the ligand were obtained in its fully protonated form H8(octapa)Cl4·2H2O, by slow evaporation of an aqueous solution of H4octapa after addition of HCl (3 M). The crystal structure is shown in Figure 2.1. From the structure, it can be seen that all four carboxylates are protonated and the hydrogen atoms are hydrogen bonded to two water molecules and two chloride ions. The pyridine nitrogen atoms are protonated and the protons are hydrogen bonded to two chloride ions. Figure 2.1. ORTEP diagram of H8(octapa)Cl4·2H2O (H4octapa·4HCl·2H2O) (–x, 1-y, 1-z); associated crystallographic data are presented in Appendix (Table A.1). 32 Complexation studies were performed by dissolving H4octapa (2.4) with an excess of the desired metal salts LnCl3·6H2O or Ln(NO3)3·6H2O (Ln = La, Sm, Dy, Yb) in either HCl or H2O. This solution was then adjusted to a pH between 3 and 8 through addition of aqueous NaOH. Metal complexes were characterized using HR-ESI-MS, ATR-IR and X-ray diffraction. H4octapa and its complexes have been thoroughly studied in previous investigations; however, crystal structures have been very difficult to obtain - there is only one example given in the literature, with Gd(III).88 A crystal structure with La(III) was obtained by slow evaporation of a mixture containing La(NO3)3 and H4octapa in 1.6:1 metal to ligand molar ratio at pH ~ 4 in H2O which yielded colorless blade crystals. The material crystallizes as a two-component twin with the chemical formula [La(octapa)(H2O)2]2[La(H2O)6.15]1/2Na1/2 as shown in Figure 2.2 and in Appendix (Table A.2). In addition there appears to be a La(H2O)x moiety bridging symmetry equivalent La-octapa complexes in different asymmetric units, as well as one half Na+ atom, disordered over two sites. This bridging La(III) is disordered over two positions in the asymmetric unit, or over four positions about an inversion center. Each one of the two La-octapa complexes is ten coordinated by H4octapa and two water molecules. Although the quality of the crystal structure is not perfect, the coordination of La(III) by H4octapa is clearly and unambiguously delineated. The average bond distances and angles are summarized in Table 2.1 and compared with those of the Gd(III) complex.88 The longer distances obtained for the complex with La(III) vs. Gd(III) are consistent with the lanthanide contraction. The coordination sphere is highly similar between the two structures, both having two water molecules coordinated to the metal which indicates structural consistency within the lanthanide H4octapa complexes. 33 Figure 2.2. ORTEP diagram of the two La-octapa complexes in the asymmetric unit with hydrogen atoms omitted for clarity; associated crystallographic data are presented in Appendix Table A.2. Table 2.1. Selected average bond distances and angles for the [La(octapa)(H2O)2]- and [Gd(octapa)(H2O)2]- complexes.88 [La(octapa)(H2O)2]- [Gd(octapa)(H2O)2]-88 N(pyr)-M (Å) 2.71(2) 2.61(2) N(en)-M (Å) 2.81(3) 2.69(3) O(pyr-COO)-M (Å) 2.53(1) 2.52(2) O(COO)-M (Å) 2.56(2) 2.46(3) O(aqua)-M (Å) 2.64(2) 2.43(4) N(pyr)-M-N(pyr) angle (deg) 171.1(1) 172(2) 2.3.3 Ligand Protonation Constants Although protonation constants of the H4octapa ligand were first reported by Chatterton et al.,88 subsequently by our group,24 and more recently by Platas-Iglesias et al.,48 in this work for completeness, the full set of protonation constants for all eight protonation sites available was report for the first time. These were determined by simultaneous potentiometric-spectrophotometric titrations following the spectral changes in the absorption band of the 34 picolinate chromophore for the first six protonation equilibria, and by UV spectrophotometric and 1H NMR studies of a batch of solutions at the same ligand concentration containing different amounts of HCl for the two most acidic protonation equilibria. The different absorption features indicating the transformations of the corresponding protonation equilibria of H4octapa are presented in Figure 2.3 and Appendix (Figure A.1). In Table 2.2 are presented the protonation constants obtained by refinement of the experimental data using the HypSpec program.86 The values for the first six protonation constants are in good agreement with those previously reported in the literature.24, 48, 88 The seventh and eighth protonation equilibria studied in strongly acidic solutions are not evaluated at constant 0.16 M NaCl ionic strength due to the high acid concentration used. The crystal structure obtained for the octa-protonated ligand, H4octapa·4HCl·2H2O, together with the hexa-protonated form of H4octapa obtained by Chatterton et al.88 (by adjusting the pH to ~ 0.5 and showing that all the carboxylates were protonated), allowed the attribution of the two lowest protonation steps to the pyridine nitrogen atoms. 1H NMR spectra of very acidic solutions of H4octapa support the conclusion that the more acidic proton residues are those of the pyridine nitrogen atoms. No significant variations in the chemical shifts of either of the methylene protons or the aromatic nucleus between pH 0.68 and 1.5 are observed; however, high field proton shifts associated with the aromatic region are observed below pH 0.68 (Appendix Figure A.1 and Table A.1), which is indicative of protonation of the pyridine rings. Comparison of the protonation equilibria corresponding to the pyridine nitrogen atoms of H4octapa with those obtained for the EDAMP analogue (Scheme 2.1)89 suggests that the increase in acidity is due the presence of the electron-withdrawing carboxylic acid substituents. 35 Table 2.2. Protonation constants of the H4octapa ligand obtained at 25 °C, I = 0.16 M NaCl, calculated by using the HypSpec program.86 equilibrium reaction [octapa]4-a [octapa]4- L + H+ ⇆ HL 8.58(1) 8.52(1)c 8.5(1)d 8.59(4)e HL + H+ ⇆ H2L 5.43(2) 5.40(1)c 5.2(2)d 5.59(6)e H2L + H+ ⇆ H3L 3.75(1) 3.65(1)c 3.5(1)d 3.77(2)e H3L + H+ ⇆ H4L 3.08(1) 2.97(1)c 2.9(1)d 2.77(4)e H4L + H+ ⇆ H5L 2.21(2) 1.66(1)c 2.79(4)e H5L + H+ ⇆ H6L 1.61(2) H6L + H+ ⇆ H7L 0.12(4)b H7L + H+ ⇆ H8L -0.46(3)b a This work, using UV-potentiometric titrations, I = 0.16 M NaCl; b This work UV spectrophotometric in batch experiments. c I = 0.15 M NaCl from ref. 48. d I = 0.1 M KCl from ref.88. eI = 0.15 M NaCl form ref. 24. Charges are omitted for simplicity. Figure 2.3. Representative absorption spectra of a potentiometric-spectrophotometric titration of H4octapa; [H4octapa] = 6.61  10-4 M, at I = 0.16 M NaCl, 25 °C; path length = 0.2 cm. 200 225 250 275 3000.00.51.01.52.0pH 3.41AbsorbanceWavelength (nm)pH 2.32200 225 250 275 3000.00.51.01.52.0pH 6.35AbsorbanceWavelength (nm)pH 4.47200 225 250 275 3000.00.51.01.52.0pH 11.02AbsorbanceWavelength (nm)pH 6.35200 225 250 275 3000.00.51.01.52.0pH 4.47AbsorbanceWavelength (nm)pH 3.4136 2.3.4 Sm(III), Dy(III) and Yb(III) Complex Formation Equilibria with H4octapa Complex formation equilibria of H4octapa in the presence of Sm3+, Dy3+ and Yb3+ metal ions were evaluated by different methods, as the complexes were quantitatively formed at low pH ~ 1.8 and thus direct potentiometric titration is not suitable to determine the first formation constant. The ligand-ligand potentiometric competition method involving H6ttha as a known competitor, allowed the stability constants of the different species in solution to be determined. For all three metal ions, mono-protonated species (MHL) were found. This species deprotonates to the ML species which can further deprotonate to form the monohydroxo complex ML(OH). The stability constants obtained are presented in Table 2.3. The log βML values for the ML complexes are in good agreement with those for the [Gd(octapa)]- and the [La(octapa)]- complexes published by Kalman et al.,48 although no protonation or deprotonation events were reported. Their kinetic studies48 obtained a protonation constant of 2.6 for the Gd(Hoctapa) species, but they were not successful fitting the constant with the pH-potentiometric and relaxometric data. In our competition experiments with the ttha6- ligand, at pH 2, ~ 30% of the MHL complex is present and cannot be ignored. Further UV spectrophotometric competition experiments were performed on sets of solutions containing the complex with variable molar concentrations of hydrochloric acid. Although the variably protonated species of H4octapa showed little difference in terms of UV-vis spectroscopic features due to small degree of pi-conjugation, in conjunction with the potentiometric data a monoprotonated complex species MHL was identified, and its stability constant was determined (Table 2.3) using the HypSpec program.86 The related UV spectra are presented in Figures 2.4, A.2 and A.3 (Appendix). Direct potentiometric titrations showed the presence of three different species in solution, with the predominant species at pH ~ 2 being the MHL complex (~ 97%). The stability 37 constants of the ML and ML(OH) species were then calculated based upon the formation constant for the MHL species obtained from UV proton competition experiments. The stability constants and the related speciation plots are presented in Table 2.3 and Figure 2.5 respectively. As noted by Kalman et al.,48 direct potentiometric titrations of the Gd-H4octapa system performed by Chatterton et al.88 in the pH range 2.5-8.5 returned an underestimated stability constant. Besides the lower stability constant, within such pH range it was not possible to evaluate the presence of the ML(OH) species that we calculated in this work for other lanthanide metal ions. The values of the stability constants for the ML species for all three metal ions obtained in this work, using two different methods, are similar to those reported for La3+, Gd3+ and Lu3+ metal ions37, 48 and therefore it can be inferred that the stability constants remain approximately the same along the lanthanide series which is in agreement with the results reported by Kalman et al..48 Table 2.3. Complex formation constants of Sm3+, Dy3+ and Yb3+ and H4octapa at 0.16 M NaCl and 25 °C. Sm3+ Dy3+ Yb3+ log KML 20.20(2)a 20.10(2)c 20.35(1)a 20.14(3)c 20.10(2)a 19.90(1)c log KMHL 3.62(3)a 3.44(4)b 3.55(2)a 3.47(4)b 3.14 (3)a 3.26(4)b log KML(OH) 9.98(3)a 10.13(2)c 10.49(2)a 10.48(3)c 10.32(2)a 10.12(3)c pM 19.8 19.9 19.6 a Obtained by ligand-ligand competition titrations using H6ttha as a ligand competitor. b Obtained from UV spectrophoto-metric titrations. c Obtained from direct potentiometric titrations. 38 Figure 2.4. a) b) and c) Representative spectra of the in batch UV spectrophotometric titration of the Yb3+-H4octapa system, [H4octapa] = 1.31 × 10-4 M, [Yb3+] = 1.32 × 10-4 M, path length = 1 cm. d) Set of spectra from the potentiometric-spectrophotometric titration of the Yb3+-H4octapa system, [H4octapa] = [Yb3+] = 6.70 × 10-4 M, path length = 0.2 cm. Figure 2.5. Speciation plots for the H4octapa complexes with: a) Sm3+, [Sm3+] = [H4octapa] = 5.52  10-4 M; b) Dy3+, [Dy3+] = [H4octapa] = 6.65  10-4 M; c) Yb3+, [Yb3+] = [H4octapa] = 6.70  10-4 M, at 25 °C and I = 0.16 M NaCl. Although Sm-EDTMP is an FDA-approved radiopharmaceutical, the various Ln-EDTMP stability constants reported show considerable differences, due to the poor solubility of protonated complexes at pH < 5.63, 90-91 The high thermodynamic stability of the [Sm(octapa)]- complex log 240 250 260 270 280 290 3000.00.51.01.5pH 11.43pH 6.23AbsorbanceWavelength (nm)240 250 260 270 280 290 3000.00.51.01.5pH 6.23AbsorbanceWavelength (nm)pH 1.92240 250 260 270 280 290 3000.00.51.01.5AbsorbanceWavelength (nm)H0 -0.17pH 1.92a) b) c) d) 240 250 260 270 280 290 3000.00.51.01.5pH 11.33AbsorbanceWavelength (nm)pH 9.062 4 6 8 10 12020406080100[SmL(OH)]2-[SmL]-% Formation relative to Sm3+pHSm(HL)2 4 6 8 10 120255075100[DyL(OH)]2-[DyL]-% Formation relative to Dy3+pHDy(HL)b) 2 4 6 8 10 120255075100[YbL(OH)]2-Yb(HL)% Formation relative to Yb3+pH[YbL]-a) c) 39 βML = 20.10(2), is comparable to that of the clinically used Sm-EDTMP log βML = 20.71(8).63 More interesting is the pM value which allows for the comparison of the ability of different ligands to coordinate a metal ion at physiological pH, 7.4 and [L] = 10-5 M and [M] = 10-6 M.92 The pSm for H4octapa was found to be 19.8, 4.5 units higher than that calculated for Sm(EDTMP) complexes with the stability constants reported by Kalman et al..63 Although log βML of the Sm(EDTMP) species is 0.6 units higher than that for [Sm(octapa)]- species, the lower basicity of H4octapa translates into a better complexing agent for Sm3+. 2.3.5 DFT Calculations and Molecular Electrostatic Potential Mapping To better understand the speciation of the metal complexes of H4octapa (Figure 2.5), DFT calculations were carried out on the complexes in the form of [Ln(octapa)(H2O)]-2H2O anions (where Ln = Sm, Dy and Yb). Even though the La(III)-octapa solid state structure indicates the presence of two metal-coordinated water molecules, La(III) is the largest Ln(III) ion and only one deprotonation of the metal complex in solution was seen. The solution structure of the Gd-octapa complex reported by Platas-Iglesias et al..50 is in agreement with the presence of one metal-coordinated water molecule. Therefore one metal-coordinated, as well as two hydrogen-bound second-sphere water molecules were included in the structure, as suggested by earlier experimental computational results on Gd(III)79 and other lanthanide-octapa systems.48 The same TPSSh hybrid functional with appropriate effective core potentials (ECPs) was employed, and obtained optimized geometry of these three anions in the aqueous media (Figures 2.6 and Appendix Figure A.4). The geometries of the Sm, Dy and Yb complexes were found to be very similar to one another. As summarized in Table 2.4, the bond distances of the coordination sphere were found to follow the similar trend reported by Kalman et al.48 – a consistent decrease with higher atomic number except those between ethylenediamine backbone nitrogen atoms and the metal center. 40 Figure 2.6. a) Optimized structure of the [Sm(octapa)(H2O)]-2H2O anion with labels on selected atoms, O5 indicates the oxygen atom on the metal-bound water molecule while O6 and O7 are the second-sphere waters; hydrogen atoms have been omitted for clarity. b) Zoomed-in view of the anion from a different angle, showing the hydrogen bonds between the water molecules and carboxylate arms. Table 2.4. Selected bond parameters of DFT calculated structures of [Ln(octapa)(H2O)]-2H2O anions. Bond Parameters Ln = Sm Ln = Dy Ln = Yb N1(pyr)-M (Å) 2.702 2.672 2.654 N2(pyr)-M (Å) 2.698 2.665 2.647 N3(en)-M (Å) 2.818 2.798 2.803 N4(en)-M (Å) 2.775 2.778 2.774 O1a(pyr-COO)-M (Å) 2.433 2.386 2.342 O2a(pyr-COO)-M (Å) 2.468 2.410 2.364 O3a(Ac-COO)-M (Å) 2.437 2.376 2.322 O4a(Ac-COO)-M (Å) 2.400 2.330 2.270 O5(aqua) -M (Å) 2.553 2.504 2.457 N(pyr)-M-N(pyr) angle (deg) 171.3 170.5 170.0 a) b) 41 The molecular electrostatic potential (MEP) mapping was also obtained based on these fully optimized structures, as shown in Figure 2.7 and Appendix Figure A.5. A clear region of negatively charged oxygen atoms (Table A.2) indicates possible sites of protonation on either O2a of the acetic carboxylate arm (Ac-COO) or O4a of the pyridyl carboxylate group (pyr-COO), to form the protonated metal species MHL observed in our solution studies at lower pH values. Similarly, the formation of the ML(OH) species detected at higher pH values can be explained by the deprotonation of the metal-bound water in the optimized structure that was calculated. Figure 2.7. MEP mapping of the [Sm(octapa)(H2O)]-2H2O anion, red = negative, blue = positive, representing a maximum potential of 0.200 au and a minimum of -0.200 au, mapped onto electron density isosurface of 0.002 Å-3. All hydrogen atoms have been omitted for clarity. 2.4 Conclusions H4octapa, a highly desirable ligand for potential applications in nuclear medicine, was synthesized using a highly improved synthetic route in only three steps with a cumulative yield of 50%. This synthetic approach, involving the protection of EDDA as the methyl ester, avoids other lengthy preparative methods and allows milder reaction conditions. Such improvement is of 42 relevant interest since this pyridinecarboxylate ligand possesses high affinity for nearly all medicinally relevant trivalent radiometals. Complete characterization of the ligand by X-ray crystallography, pH-potentiometry, UV spectrophotometry and 1H NMR spectroscopy allowed for determination of the pKa values for all eight protonation sites available for the first time, representing the most complete characterisation yet for this important ligand. Complex formation equilibria of H4octapa with the lanthanides Sm(III), Dy(III) and Yb(III) were evaluated by pH-potentiometric and UV spectrophotometric methods and three different species - MHL, ML and ML(OH) were found in solution. A rough X-ray crystal structure of [La(octapa)]- confirms the atomic connectivity in the anion, while DFT calculations and MEP mapping of the [Ln(octapa)(H2O)]- structures were employed to rationalize the protonation and deprotonation of the ML species. The high thermodynamic stability of the [Sm(octapa)]- species log βML = 20.1, and more importantly the significantly higher pM value compared to the FDA approved 153Sm(EDTMP) radiopharmaceutical, along with the high thermodynamic and in vivo stability demonstrated for the [Lu(octapa)]- complexes, make H4octapa a highly desirable candidate for further studies with the isotope 153Sm in pharmaceutical applications, as well as with many other radiometal ions of the lanthanides. In addition, a H4octapa derivative with bromine substituted pyridyl groups are also reported. The use of a halogenated pyridyl precursor will enable functionalisation via Pd catalyzed cross-coupling including Suzuki and Sonogashira coupling, or via nucleophilic substitution by piperazine and sodium azide that can be the future work to prepare bifunctional H4octapa. 43 Chapter 3: H2hox: Dual Channel Oxine-Derived Acyclic Hexadentate Chelating Ligand for 68Ga Radiopharmaceuticals This chapter is an adaptation of published work, and is reproduced in part from Xiaozhu Wang, María de Guadalupe Jaraquemada-Pelaez, Yang Cao, Jinhe Pan, Kuo-Shyan Lin, Brian O. Patrick, and Chris Orvig. H2hox: Dual Channel Oxine-Derived Acyclic Hexadentate Chelating Ligand for 68Ga Radiopharmaceuticals. Inorg. Chem. 2019, 58, 2275-2285. Cover article for first issue of 2019. 3.1 Introduction The 99Mo/99mTc generator has been the most successful and widely used clinical isotope generator for decades. It is an important and convenient source of the 99mTc radionuclide, especially for most hospitals or institutes that do not have a nearby cyclotron. There is a great need for alternatives since the supply of the parent isotope 99Mo will be badly disrupted after the shut down of Canada’s National Research Universal (NRU) reactor (the main source of 99Mo in North America) in 2018. The 68Ge/68Ga generator system is a most attractive alternative that could be utilized for an extended period due to the 271-day half-life of the parent isotope 68Ge. The daughter isotope 68Ga has a high positron decay yield (89%, 1.899 keV) and a relatively short half-life time (t1/2 = 68 min), and is becoming increasingly important as a versatile and easily available positron emission tomography (PET) imaging tracer.93-95 Because of the 68Ga short half-life, an ideal chelating ligand should rapidly achieve efficient and reproducible radiolabelling under mild conditions (room temperature and near neutral pH), yielding a stable radiopharmaceutical with high molar activity (molar activity - the measured 44 radioactivity per total amount in mole of compound) at low concentration that could be used in routine clinical practice without further purification. Thus, such a radiopharmaceutical ligand would be ideal for the development of a toolkit for convenient clinical use.96-99 The macrocyclic chelator DOTA (Scheme 3.1) is the most commonly used chelator for 68Ga labelling.100-104 The pre-organized closed chain structure guarantees a high kinetic inertness and thermodynamic stability; however, concomitant is longer labelling time and high temperature which is completely incompatible with the labelling of thermally sensitive bio-vectors. Moreover, the synthesis of those macrocyclic chelators and corresponding bifunctional derivatives is also challenging and expensive involving post-synthetic purification steps and work-up for the removal of impurities.105 Over the past decade, several Ga3+ chelators have been developed to overcome these drawbacks including NOTA,106-107 NOTP,108 TRAP,22, 109 DATA,110-112 PCTA,113 HBED114-116 and H2dedpa 20, 25, 39, 93, 117-120 (Scheme 3.1). NOTA (1,4,7-triazacyclononane-1,4,7-triacetic acid) radiolabels Ga3+ at RT in 10 min at pH 3-5.5.106 NOTP and TRAP, the tri-phosphate analogs of NOTA, could be used in a broader pH range, especially at lower pH.108, 121 HBED and its bioconjugate showed a high thermodynamic stability and serum stability.114 THP achieved a 97% radiochemical yield at 0.5 μM and near neutral pH,99 and finally H2dedpa, can obtain a quantitative radiochemical yield at concentrations as low as 0.1 μM.25 There are still limitations among those chelators. For example, the labelling performance of NOTA and TRAP at near-physiological pH is not as good as at lower pH, and requires higher concentrations; HBED forms multiple species in solution with Ga3+ and thus, it is not ideal for quick and easy purification and kit-based application.99 The synthesis of THP and its derivative are non-trivial.110 Synthetic accessibility is also a challenge for H2dedpa and most of the cyclic chelators as well.25 45 Scheme 3.1. Examples of chelating ligands for Ga3+ from the literature. In this work, a tightly binding acyclic hexadentate chelator is reported that will certainly be applicable to 68Ga PET imaging: H2hox with rigid “bis-ox” (8-hydroxyquinoline) arms. The synthesis of H2hox is easy and fast (two steps), and solution studies reveal the presence of a single complex species in a broad pH range (1-11) with higher log KML (34.4) and pM value (28.3) than for most of Ga3+ chelators. H2hox showed fast and quantitative 68Ga complexation at mild conditions (5 minutes, RT) with a concentration as low as 10-7 M and an unprecedented high molar activity without purification. Plasma challenge experiments and dynamic PET imaging confirm excellent in vitro and in vivo stability. These characteristics suggest strong relevance to toolkit radiopharmaceuticals (“shake and shoot”); to complement these ideal properties, the fluorescence 46 of the chelating ligand is turned on by complexation to Ga3+, and this was used to investigate the cellular distribution of [Ga(hox)]+ directly and showed the potential for dual channel imaging. 3.2 Experimental 3.2.1 Materials and Methods All solvents and reagents were purchased from commercial sources (TCI America, Sigma Aldrich, Fisher Scientific) and used as received unless otherwise indicated. The analytical thin-layer chromatography (TLC) plates used were aluminum-backed ultrapure silica gel 60 Å, 250 μm thickness; 1 H and 13C NMR spectra were recorded at ambient temperature on Bruker Avance 300 and Avance 400 spectrometers; the 1H NMR spectra were calibrated against residual protio-solvent peak, and the 13C NMR spectra were referenced to the deuterated solvent. Low-resolution mass spectrometry was performed on a Waters ZG spectrometer with an ESCI (electrospray/chemical-ionization) source, and high-resolution electrospray ionization mass spectrometry (ESI-MS) were performed on a Micromass LCT time-of-flight (TOF) instrument. Microanalyses for C, H, and N were performed on a Carlo Erba Elemental Analyzer EA 1108. Purification and quality control of [68Ga(hox)]+ were performed on an Agilent HPLC system equipped with a model 1200 quaternary pump, a model 1200 UV absorbance detector, and a Bioscan (Washington, DC) NaI scintillation detector. The radiodetector was connected to a Bioscan B‐FC‐1000 Flow‐count system, and the output from the Bioscan Flow‐count system was fed into an Agilent 35900E Interface which converted the analog signal to digital signal. The operation of the Agilent HPLC system was controlled using the Agilent ChemStation software. The HPLC columns used were a semi‐preparative column (Phenomenex C18, 5 µ, 250 × 10 mm) and an analytical column (Phenomenex C18, 5 µ, 250 × 4.6 mm). The HPLC solvents were A: H2O containing 0.1% TFA, and B: CH3CN containing 0.1% TFA. 68Ga was obtained from an Eckert & Ziegler (Berlin, Germany) IGG100 47 68Ga generator, and was purified according to the previously published procedures122 using a DGA resin column. Radioactivity of [68Ga(hox)]+ was measured using a Capintec (Ramsey, NJ) CRC® ‐25R/W dose calibrator. PET imaging experiments were conducted using a Siemens (Erlangen, Germany) Inveon microPET/CT scanner. 3.2.2 Ligand Synthesis and Characterization 8-Hydroxyquinoline-2-carboxaldehyde (3.1) Compound 3.1 could be purchased from a commercial supplier or synthesized by the following procedure. A mixture of 8-hydroxy-2- methylquinoline (5.0 g, 31 mmol) and selenium dioxide (3.9 g, 35 mmol) was stirred in 250 mL 1,4-dioxane at 80 °C overnight. Celite (5g) was added after cooling and then filtered out. The subsequent filtrate was evaporated to a crude solid and was purified by silica-gel column chromatography (hexane: ethyl acetate, 10:90 to 50:50, v/v) to obtain pure yellow needle crystals (4.4 g, 25 mmol), yield = 82%. 1H NMR (300 MHz, CDCl3, 25 °C) δ 10.21 (d, J = 0.9 Hz, 1H), 8.32 (dd, J = 8.5, 0.8 Hz, 1H), 8.15 (s, 1H), 8.05 (d, J = 8.5 Hz, 1H), 7.62 (t, J = 8.0 Hz, 1H), 7.42 (dd, J = 8.3, 1.2 Hz, 1H), 7.28 (dd, J = 7.7, 1.3 Hz, 1H). 13C NMR (75 MHz, CDCl3, 25 °C) δ 192.9, 153.1, 150.3, 138.0, 137.6, 131.1, 130.6, 118.2, 118.1, 111.4, 77.2. LR-ESI-MS calcd. for [C10H7NO2 + Na]+: 196.0; found: 196.3 H2hox (3.2) 8-Hydroxyquinoline-2-carboxaldehyde (3.1) (2 g, 11.6 mmol) was dissolved in 50 mL ethanol, and ethylenediamine (387 μL, 5.8 mmol) dissolved in 5 ml of ethanol was added dropwise; the reaction 48 mixture was stirred at 60 °C for 4 h. A light-yellow precipitate formed, was collected and resuspended in 50 mL ethanol. Five equivalents of NaBH4 were added in portions and the reaction mixture was stirred at room temperature overnight. HCl (20 mL, 6 M) was added and the reaction mixture stirred for 4 h to quench. The pH of the reaction mixture was then readjusted to neutral using NaOH (2 M), and the white precipitate was filtered and dried as crude product. The crude product was further washed with water and methanol to obtain the pure product (1.97 g, 5.2 mmol), yield = 87%. 1H NMR (300 MHz, MeOD, 25 °C) δ 8.21 (d, J = 8.5 Hz, 1H), 7.50 – 7.38 (m, 2H), 7.36 (dd, J = 8.3, 1.6 Hz, 1H), 7.10 (dd, J = 7.3, 1.6 Hz, 1H), 4.20 (s, 2H), 3.03 (s, 2H). 13C NMR (75 MHz, MeOD, 25 °C) δ 157.4, 154.1, 139.2, 138.1, 129.5, 128.4, 122.0, 119.0, 112.3, 54.7, 49.0, 48.8. HR-ESI-MS calcd. for [C22H22N4O2 + H]+: 375.1821; found: 375.1819. H2hox·2HCl used for titration was synthesized by mixing a solution of H2hox (2) in THF and 6 M HCl and drying in vacuo to afford a white powder with elemental analysis: H2hox·2HCl·0.5 H2O, calcd. % for C 57.90, H 5.52, N 12.28; found: C 58.24, H 5.30, N 11.90. 3.2.3 Synthesis of Hox-Metal complex [Ga(hox)][ClO4](3.3) H2hox (40 mg, 0.107 mmol) was dissolved in acetonitrile. Ga(ClO4)3·6H2O (55 mg, 0.116 mmol) was added, and the pH was adjusted to ~ 5 using 0.1 M NaOH. The reaction mixture was stirred for 1 h at 50 °C, CH2Cl2 was added to extract the product and a yellow crystal suitable for X-ray diffraction formed from layer diffusion of diethyl ether into the CH2Cl2 extraction layer.1H NMR (300 MHz, MeOD, 25 °C) δ 8.81 (d, J = 8.5 Hz, 1H), 7.87 (d, J = 8.5 Hz, 1H), 7.76 – 7.53 (m, 1H), 7.41 (d, J = 8.3 Hz, 1H), 7.00 (d, J = 7.7 Hz, 1H), 4.76 (d, J = 17.8 Hz, 1H), 4.38 (d, J = 17.9 Hz, 1H), 3.11 (d, J = 10.0 Hz, 1H), 2.63 (d, J = 9.9 Hz, 1H). 13C NMR (75 MHz, MeOD, 25 °C) δ 49 155.9, 150.2, 142.7, 134.9, 131.2, 128.6, 120.2, 113.1, 112.7, 49.7. HR-ESI-MS calcd. for [C22H2269GaN4O2]+: 441.0842; found 441.0841. Elemental analysis for [Ga(hox)][ClO4]·H2O, calcd. % C 47.22, H 3.96, N 10.01; found C47.19, H 3.84, N 9.72. 3.2.4 X-ray Crystallography Colorless tablet-shaped crystals of H2hox were obtained by recrystallization from methanol. X-ray diffraction data for a suitable crystal were collected using a Bruker APEX II area detector diffractometer with Mo-Kα radiation. The structure was solved in the monoclinic P21/c space group. Yellow blade crystals of [Ga(hox)][ClO4] suitable for X-ray diffraction were obtained by layer diffusion of diethyl ether into a CH2Cl2 solution of the complex. X-ray diffraction data of a suitable crystal were collected on a Bruker APEX DUO diffractometer using Mo-Kα radiation. The structure was solved in the monoclinic C2/c space group. All non-hydrogen atoms were refined anisotropically. All N-H hydrogen atoms were located in difference maps and refined isotropically. All other hydrogen atoms were placed in calculated positions. Further structural refinement details are available in the Appendix (Table B.1). 3.2.5 DFT and TDDFT Calculations All calculations were performed using the Gaussian 09 package (Revision D.01). Full geometry optimizations of the [Ga(hox)]+ cation were performed with the CAM-B3LYP hybrid exchange–correlation functional123 in aqueous solution using the polarizable continuum model (PCE).124 Geometry optimizations were carried out using the 6-311+G(d,p) basis set on first and second row elements, and the Los Alamos effect core potential (ECP) and valence basis set of double zeta quality (LANL2DZ) on the Ga atom.125 The input coordinates of atoms were adapted from the crystal structure of the [Ga(hox)][ClO4] complex and no constraint on symmetry was 50 imposed during the geometry optimization. The resulting geometries showed no imaginary frequencies and thus were confirmed to be minima on the potential energy surfaces. The PBE0 hybrid functional126 and the same basis set and ECP was employed to simulate the UV-Vis absorption features of the fully optimized structure, and generate its ground state molecular electrostatic potential (MEP) mapping. 3.2.6 Solution Thermodynamics Protonation constants and metal stability constants were calculated from UV spectrophotometric titration data obtained using a Cary 60 UV-vis spectrophotometer in the spectral range of 200-450 nm. The path length was 1 cm for all samples. Individual samples (5 mL) containing the ligand (H2hox, 2.72  10-5 M) or the corresponding gallium complex, ([Ga(hox)][ClO4], 2.82  10-5 M) in water were prepared by adjusting the pH with different amounts of standardized HCl or NaOH solutions, and NaCl was added to maintain a constant 0.16 M ionic strength in the pH range 0.8-11.85. A Ross combination pH electrode was daily calibrated for hydrogen ion concentrations using HCl as described before80 and the results were analyzed by the Gran81 procedure. pH was measured in ligand and metal-ligand samples between the pH range 2.0-11.5. With samples at pH < 0.8 it was not possible to maintain a constant ionic strength since that depends on the HCl content; the equilibrium H+ concentration was calculated from solution stoichiometry, not measured with a glass electrode. For the solutions of high acidity, the correct acidity scale H0 was used.84 For the ligand protonation equilibria study up to 2 min was required at 25 °C to reach equilibrium before measuring the pH and the UV absorption spectrum. For the samples containing the metal-ligand complex, the measurements were performed only after 24 hours, at equilibrium. The protonation constants for H2hox, and the Ga(III) complex stability constants, were calculated from the experimental data using the HypSpec201486 program. Proton 51 dissociation constants corresponding to hydrolysis of Ga(III) aqueous ions included in the calculations were taken from Baes and Mesmer.87 The species formed in the studied systems are characterized by the general equilibrium: pM + qH + rL = MpHqLr (charges omitted). For convention, a complex containing a metal ion, M, proton, H and ligand, L, has the general formula MpHqLr. The stoichiometric indices p, might also be 0 in the case of protonation equilibria and negative values of q refers to proton removal or hydroxide ion addition during formation of the complex. The overall equilibrium constant for the formation of the complexes MpHqLr from its components is designated as log β. Stepwise equilibrium constants log K correspond to the difference in log units between the overall constants of sequentially protonated (or hydroxide) species. pM is defined as (-log[Mn+]free) and is always calculated at [Mn+] = 1 µM, [Lx-] = 10 µM, pH 7.4 and 25 °C.92 3.2.7 [68Ga(hox)]+ Labeling Procedure. 68Ga was eluted from an iThemba Labs (Somerset West, South Africa) generator, and purified according to the previously published procedures using a DGA resin column from Eichrom Technologies LLC (Lisle, IL).The 68Ga generator was eluted with a total of 4 mL of 0.6 mol/L HCl. The elution that contained the activity was mixed with 2.5 mL concentrated HCl. The mixture was passed through a DGA resin column and the column was washed by 3 mL 5 mol/L HCl. After the column was dried by passage of air, 68Ga was eluted off with 0.5 mL water. [68Ga(hox)]+ was obtained by adding 0.2 mCi (4-11 L) purified 68Ga to 200 µL of a 10-7 M solution of H2hox in 0.1 M NaOAc solution (pH 8.5) and left for 5 min at RT. The reaction progress was monitored by analytical HPLC eluted with 84/16 phosphate buffer (pH 7.4)/CH3CN at a flow rate of 2 mL/min. The retention time of [68Ga(hox)]+ was 8.6 min. 52 3.2.8 Log D7.4 Measurements Aliquots (2 μL) of [68Ga(hox)]+ were added to a vial containing 3 mL of octanol and 3 mL of 0.1 M phosphate buffer (pH 7.4). The mixture was vortexed for 1 min and then centrifuged for 10 min. Samples of the octanol (1 mL) and buffer (1 mL) layers were taken and counted. Log D7.4 was calculated using equation 3.1 Log D7.4 = log10[(counts in octanol phase)/(counts in buffer phase)]. 3.1 3.2.9 Stability in Mouse Plasma Purified 68Ga in 0.5 mL water was added into a 4‐mL glass vial preloaded with 0.7 mL of HEPES buffer (2 M, pH 5.0) and 25 nmol H2hox. The radiolabeling reaction was carried out under microwave heating for 1 min. The reaction mixture was purified by HPLC using the semi‐preparative column eluted with 83/17 A/B (A: H2O containing 0.1% TFA, and B: CH3CN containing 0.1% TFA) at a flow rate of 4.5 mL/min. The retention time of [68Ga(hox)]+ was 13.8 min. The eluate fraction containing the radiolabeled product was collected, diluted with water (50 mL), and passed through a C18 Sep-Pak cartridge that was pre-washed with ethanol (10 mL) and water (10 mL). After washing the C18 Sep-Pak cartridge with water (10 mL) the 68Ga-labeled product was eluted off the cartridge with ethanol (0.4 mL), dried by helium, flow and re-dissolved with saline (0.5 mL) for plasma stability and imaging studies. Aliquots (20 μL) of [68Ga(hox)]+ were incubated with 80 μL of mouse plasma for 5, 15, 30, and 60 minutes at 37 °C. At the end of each incubation period, samples were quenched with 100 μL 70% CH3CN and centrifuged for 20 min. After samples were quenched with 100 μL 70% CH3CN and centrifuged for 20 minutes, the suspension was loaded onto the HPLC with a C-18 semi-preparative column (Luna C18, 5 μ, 250 × 10 mm). Neither the suspension nor the precipitate was counted before HPLC analysis. 53 3.2.10 PET/CT Imaging Studies PET/CT imaging studies were conducted in accordance with the guidelines established by the Canadian Council on Animal Care and approved by the Animal Ethics Committee of the University of British Columbia. Male NOD.Cg-Rag1tm1MomIl2rgtm1Wjl/SzJ (NRG) mice were purchased from in-house colonies at the Animal Research Centre, BC Cancer Research Centre, Vancouver, Canada. PET/CT imaging experiments were conducted using a Siemens (Knoxville, TN, USA) Inveon microPET/CT scanner. Mice were sedated with 2% isoflurane in oxygen inhalation and positioned in the scanner. A baseline CT scan was obtained for localization and attenuation correction before radiotracer injection, using 80 kV X-rays at 500 mA, three sequential bed positions with 34% overlap, and 180-degree continuous rotation. The mice were kept warm by a heating pad during acquisition. The dynamic acquisition of 60 min was started at the time of intravenous injection with ~ 3.4-4.0 MBq of [68Ga(hox)]+. The list mode data were rebinned into time intervals (12 × 10 sec, 8 × 60 sec, 7 × 300 sec, 1 × 900 sec) to obtain tissue time-activity curves. Images were reconstructed using iterative 3-dimensional ordered subset expectation maximization (OSEM3D, 2 iterations) using maximum a priori with shifted poisson distribution (SP-MAP, 18 iterations). 3.2.11 Fluorescence Spectra H2hox and [Ga(hox)][ClO4] were each dissolved in PBS (7.4) buffer at 1.7  10-5 M. Fluorescence emission spectra were measured at 365 nm excitation wavelength using an Agilent Cary Eclipse Fluorescence Spectrophotometer. 54 3.2.12 Fluorescence Microscopy HeLa cells were purchased from the American Type Culture Collection (ATCC). Cells were grown in Eagle's Minimal Essential Medium (MEM) supplemented with heat-inactivated 10% fetal bovine serum, 1 mM sodium pyruvate, 4 mM L-glutamine, and 1% nonessential amino acids in a humidified incubator at 37 °C and 5% CO2. Cells were seeded 8-well culture slips 24 hr prior to treatment. The [Ga(hox)][ClO4] working solution for fluorescence microscopy was prepared from a PBS stock solution. No precipitation of the compound was observed in the working solution under this condition. Cells were exposed to 150 uM [Ga(hox)][ClO4] for 2h and 24 h, washed with phosphate buffered saline (PBS) and imaging was done using a fluorescence microscope BX40, an U-MWU filter cube, an F-View CCD Camera (all Olympus), Cell-F fluorescence imaging software (Olympus) and a 60 × magnification oil immersion objective lens. 3.3 Results and Discussion Our group has screened many acyclic chelates for gallium labelling in the past 20 years and H2dedpa was the most successful, showing fast chelation and high stability, properties close to those of DOTA and NOTA, until we discovered H2hox.25, 49, 52, 55, 80, 127 The high stability of H2dedpa was attributed to a near perfect size fit and geometric arrangement of coordination atom and bonds. Oxine (8-hydroquinoline), the bidentate ligand in the anticancer compound KP46, possesses a very high stability constant for its 3:1 complex with gallium (log 𝛽𝑀𝐿3 = 36.4).128 In the design of our second generation of acyclic chelates for gallium, a combination of the respective advantages of H2dedpa and oxine was sought, leading to the structure of H2hox (Scheme 3.2). 55 Scheme 3.2. Design paradigm for H2hox. 3.3.1 Synthesis and Characterization of Ligands The only other report of H2hox was in 1972 by Hata and Uno 129-130 as an analytical reagent (BHQED) for divalent metals and it has not been further explored in the intervening near-half-century; however, Hata and Uno’s synthetic protocol was difficult to reproduce, resulting in low yield and purity. Thus, a modified synthetic route was developed (Scheme 3.3). 8-Hydroxyquinoline-2-aldehyde is a cheap starting material purchased from commercial supplier or could be synthesized from 2-methyl-8-hydroxyquinoline in one step without protection. H2hox was synthesized by reductive amination of 8-hydroxyquinoline-2-aldehyde and ethylenediamine in one step and purified by recrystallization to obtain a 87% yield without the need of column separation. In the Hata and Uno synthetic protocol, most of the product precursor was filtered out and discarded before the reaction was quenched with acetic acid, giving a low yield. Therefore, in our new protocol the reaction was quenched by HCl directly and the product precipitated out in high yield after the pH was adjusted to neutral. The new route takes only one or two steps (depending on starting material), quickly and economically yielding grams of ligand. This is particularly notable when comparing with most of the previously reported ligands such as NOTA, DOTA, TRAP, NOTP, DATA, THP and H2dedpa. Moreover, it potentially should be transformed 56 conveniently into a bifunctional version by using a functionalized diamine backbone (for example, (4-nitrobenzyl) ethylenediamine). The bifunctional tracer and the hydrophilicity tuning could also be achieved by direct modification on the aromatic ring; this has been widely used in the 8-hydroquinoline-based pharmaceuticals and in organic light emitting diode (OLED) agents.131-135 The tedious and challenging synthesis and purification of most of the previous chelators is a significant barrier to commercial applicability. Thus, we think H2hox can overcome this barrier and is ideally suited for real clinical research and wide application. Scheme 3.3. Synthesis of H2hox. a) en, CH3CH2OH, at 60 °C, 4 hr; b) CH3CH2OH, NaBH4 (5 equiv), overnight. 3.3.2 Preparation of the Metal Complex H2hox and Ga(ClO4)3 was mixed in 1:1 molar ratio in methanol and the pH was adjusted to neutral using NaOH (1 M); the [Ga(hox)][ClO4] was then extracted by DCM. The formation of metal complex was confirmed by the the conversion from homotopic proton on the free ligand to diastereotopic protons in the complex as shown in Figure 3.1. 57 Figure 3.1. 1H NMR spectra of H2hox in MeOD (400 MHz, 25 °C) (upper) and [Ga(hox)][ClO4] in MeOD (300 MHz, 25 °C) (lower). 3.3.3 X-ray Crystallography and DFT Calculations The structure of H2hox was confirmed by X-ray diffraction data using a suitable crystal obtained by recrystallization from methanol. Two crystallographically independent structures, both with Ci symmetry (Figure 3.2), have been identified featuring two 8-hydroxyquinoline moieties 58 connected to an ethylenediamine backbone giving rise to six potential coordination sites (Figure 3.2). [Ga(hox)][ClO4] crystals were obtained by layer diffusion of diethyl ether into a dichloromethane solution of the complex, and the material crystallizes with two Ci symmetric structures (Figure 3.3). Selected bond parameters of the mer-[Ga(hox)]+ cation are summarized in Table 3.1, and compared to those of mer-[Ga(dedpa)]+. The meridional arrangement of the six-coordinating atoms in [Ga(hox)]+ is similar to that in [Ga(dedpa)]+; however, in [Ga(hox)]+, the Ga-O and Ga-N(ox) bonds are slightly shorter while Ga-N(en) bonds are longer. The evenly distributed array of bond lengths suggests an excellent fit of hox2- with the Ga3+ cation, and a stability possibly comparable to that of dedpa2- Figure 3.2. ORTEP-style image of H2hox, including symmetry equivalencies. (i) 2-x, -y,1-z, (ii) 1-x,1-y,1-z. 59 Figure 3.3. ORTEP-style picture of [Ga(hox)][ClO4] complexes. The two crystallographically distinct cation structures and perchlorate anions shown. Solvent acetonitrile molecules not shown, for clarity. Table 3.1. Selected bond distances in the cations [Ga(hox)]+ and [Ga(dedpa)]+. [Ga(hox)]+ [Ga(dedpa)]+a Atom Atom Length (Å) Atom Atom Length (Å) Ga1 O1(ox) 1.959(2) Ga O1(pyr-COO) 1.971(1) Ga2 O2(ox) 1.952(2) Ga O2(pyr-COO) 1.983(1) Ga1 N1(ox) 1.982(2) Ga N1(pyr) 1.987(2) Ga2 N3(ox) 1.981(2) Ga N2(pyr) 1.990(2) Ga1 N2(en) 2.190(3) Ga N3(en) 2.112(2) Ga2 N4(en) 2.193(3) Ga N4(en) 2.113(2) a From Ref.25 pyr = pyridine, and en = ethylenediamine. 60 The coordination geometries of the [Ga(dedpa)]+ and [Ga(hox)]+ cations in aqueous solution were simulated using density functional theory (DFT), Figure 3.4 a-b. The calculated bond parameters of these two cations are summarized in Tables 3.2 and 3.3 and compared with their solid-state structures. Similar geometries and bond lengths are observed in the simulated solution structures of both cations. The corresponding molecular electrostatic potential mapping (MEP) of these geometries is shown in Figure 3.4 c-d. In the [Ga(hox)]+ cation, less prominent electronegative potential (indicated by red areas) is observed, which could result in a higher stability and a slower rate of hydrolysis of [Ga(hox)]+ compared to that of [Ga(dedpa)]+ under acidic conditions. Moreover, a more evenly distributed surface charge would translate into a higher lipophilicity for [Ga(hox)]+ vs. [Ga(dedpa)]+. Figure 3.4. DFT optimized structures of the (a) [Ga(hox)]+ and (b) [Ga(dedpa)]+ cations with hexacoordinated metal centers, and the electrostatic potentials of (c) [Ga(hox)]+ and (d) [Ga(dedpa)]+ mapped onto their electron density. The MEP represents a maximum potential of 0.25 au, and a minimum of −0.02 au mapped onto electron density isosurface (0.002 e Å–3, red to blue = negative to positive). 61 Table 3.2. Selected bond lengths of [Ga(hox)]+ in the solid state and calculated from DFT in solution. Solid State Solution Atom Atom Length (Å) Atom Atom Length (Å) Ga1 O1(ox) 1.959(2) Ga O2(ox) 1.943 Ga2 O2(ox) 1.952(2) Ga O32(ox) 1.943 Ga1 N1(ox) 1.982(2) Ga N5(ox) 1.990 Ga2 N3(ox) 1.981(2) Ga N29(ox) 1.990 Ga1 N2(en) 2.190(3) Ga N19(en) 2.229 Ga2 N4(en) 2.193(3) Ga N25(en) 2.229 Table 3.3. Selected bond lengths of [Ga(dedpa)]+ in the solid state and calculated from DFT in solution. Solid State Solution Atom Atom Length (Å) a Atom Atom Length (Å) Ga O1(pyr-COO) 1.971(1) Ga O5(pyr-COO) 1.956 Ga O2(pyr-COO) 1.983(1) Ga O37(pyr-COO) 1.956 Ga N1(pyr) 1.987(2) Ga N1(pyr) 2.005 Ga N2(pyr) 1.990(2) Ga N2(pyr) 2.005 Ga N3(en) 2.112(2) Ga N3(en) 2.165 Ga N4(en) 2.113(2) Ga N4(en) 2.169 a From Ref.25 pyr = pyridine, and en = ethylenediamine. 3.3.4 Solution Thermodynamics High thermodynamic stability (log KML) and kinetic inertness are required for in vivo application of radiopharmaceutical metal complexes, in order to minimize transchelation and transmetalation reactions caused by endogenous ligands and metals. The stepwise protonation constants of H2hox included in the calculation of the log KML were determined by UV in batch spectrophotometric titrations as the pH-potentiometric method was unsuitable because of the 62 insufficient solubility of H2hox. The spectral data were refined using the HypSpec2014 program (Table 3.4).86 Table 3.4 Protonation Constants of H2hox at 25 °C. Equilibrium Reaction log β log K log Kc L + H+ ⇆ HL 10.88(1) a 10.88 a 11.35 HL + H+ ⇆ H2L 20.69(1) a 9.81 a 10.81 H2L + H+ ⇆ H3L 29.08(1) a 8.39 a 8.15 H3L + H+ ⇆ H4L 35.14(2) a 6.06 a 5.12 H4L + H+ ⇆ H5L 35.78(6) b 0.64 b ND H5L + H+ ⇆ H6L 36.02(8) b 0.24 b ND a This work, I = 0.16 M NaCl. b This work, not evaluated at constant I = 0.16 M NaCl. c From ref. 129, 50 v/v% aqueous dioxane solution, I = 0.1 M KCl. Charges are omitted for clarity, ND = not determined. H2hox in its neutral form is denoted as H2L, whereas its fully protonated form in very acidic solution is H6L4+. The latter possesses six potential protonation sites: two phenolate oxygen atoms, two pyridyl nitrogen atoms and the two secondary amine nitrogen atoms on the ethylenediamine backbone. Hata and Uno determined and assigned the first four protonation constants of H2hox by pH-potentiometric titrations in a H2O-dioxane solvent mixture (50% v/v, µ = 0.1 M KCl, 25°C) due to its insufficient solubility in water.129-130 In this work, we were able to analyse all six protonation events in water by exhaustive UV spectrophotometric titrations, taking advantage of the high and different molar absorptivities of each protonated species of H2hox. These titrations were carried out from very acidic solutions until pH 11.80, showing well-defined isosbestic points (Figure 3.5 a-f) indicative of consecutive deprotonation steps of the ligand. Seven absorbing species were identified (Figure 3.6); six protonation constants were determined using the HypSpec2014 program86 and summarized in Table 3.4. The distribution diagram is presented in Figure 3.5 a. 63 From the analysis of the spectra of H2hox solutions between pH 0-3.39 (Figures 3.5 a and 3.5 b), the ligand at its fully protonated form H6L4+ presents similar spectroscopic features to those of fully protonated oxine (H2Lox+) reported by both Choppin136 and Enyedy,128 with almost doubled molar absorptivities for λmax = 260 and 378 nm consistent with two protonated quinolinium-NH+ chromophores (Table 3.5). The H6L4+ species transforms into the H5L3+ species with the appearance of isobestic points at 249 and 346 nm which have been reported also for the deprotonation of H2Lox+ (Table 3.5). The molar absorptivities for the H5L3+ species are of the same order of magnitude as those of the fully protonated oxine (H2Lox+), consistent with one protonated quinolinium -NH+. This allows for the assignment of the two most acidic pKa values to the dissociation of the protons from the two quinoline nitrogen atoms (pK1 = 0.24(8) and pK2 = 0.64(6)). In addition, the spectra of the species HL- and L2- in most basic conditions show spectroscopic evolution similar to that observed for the quinolinate species of oxine (Lox-), with the appearance of isosbestic points at 249 and 320 nm (Figures 3.5 e and 3.5 f, Table 3.5).128, 136 This indicates the deprotonation of phenol-OH groups in both steps (pK5 = 9.81(1) and pK6 = 10.88(1)). The equilibria involving the species H4L3+, H3L2+ and H2L show much smaller spectroscopic differences than those for the previously mentioned processes (Figures 3.5 c and 3.5 d, Figure 3.6). These species present molar absorptivities doubled vs. the neutral HLox species (Table 3.5) which accounts for the loss of the protons from the secondary amine nitrogen atoms on the backbone (pK3 = 6.06(2) and pK4 = 8.39(1)). 64 Figure 3.5. Spectra of the in batch UV spectrophotometric titration of 2.72 × 10-5 M solution of H2hox. Figure 3.6. pH-dependent absorption spectra (in molar absorptivity) of the 7 light absorbing species of H2hox. 200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0pH 3.391.901.421.160.750.890.610.520.450.390.28 0.10AbsorbanceWavelength (nm)Ho-0.01 = 249 nma)275 300 325 350 375 400 425 4500.000.050.100.150.20pH 3.391.900.890.610.39 0.10AbsorbanceWavelength (nm)Ho-0.01 = 346 nmb)200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.08.918.568.036.50AbsorbanceWavelength (nm)pH  = 249 nm 3.39c)275 300 325 350 375 400 425 4500.000.050.100.150.208.91AbsorbanceWavelength (nm)pH 3.39d)200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.011.9711.1510.5810.3310.199.899.38AbsorbanceWavelength (nm)pH  = 249 nm8.91e)275 300 325 350 375 400 425 4500.000.050.100.150.2011.9711.1510.5810.239.899.38AbsorbanceWavelength (nm)pH  = 320 nm8.91f)200 225 250 275 300020000400006000080000100000Absorptivity (M-1 cm-1)Wavelength (nm) HL- H2L H3L+ H4L2+ H5L3+ H6L4+ L2-275 300 325 350 375 400 425 45001000200030004000500060007000Absorptivity (M-1 cm-1)Wavelength (nm) HL- H2L H3L+ H4L2+ H5L3+ H6L4+ L2-65 Table 3.5. Molar absorptivity of selected wavelengths in different species of oxine and H2hox. Ligand Species λ (nm) ε (M-1 cm-1) εhox / εoxine λ isosbestic (nm) oxine H2Lox+ 253 3.75  104 a 245; 334 358 1.68  103 b H2hox H6L4+ 260 8.58  104 2.3 249; 346 378 3.39  103 2.0 H5L3+ 260 4.30  104 1.1 378 1.92  103 1.1 Ligand Species λ (nm) ε (M-1 cm-1) εhox / εoxine λ isosbestic (nm) oxine HLox 236 3.09  104 a 245; 334 306 2.38  103 b H2hox H4L2+ 243 6.57  104 2.1 249 308 5.04  103 2.1 H3L+ 243 6.43  104 2.1 308 5.06  103 2.1 H2L 243 5.74  104 1.9 308 4.50  103 1.9 Ligand Species λ (nm) ε (M-1 cm-1) εhox / εoxine λ isosbestic (nm) oxine Lox- 253 2.87  104 a 245; 320 334 2.82  103 b 354 2.74  103 b H2hox L2- 257 6.93  104 2.4 249; 320 337 5.90  103 2.1 354 5.41  103 2.0 a Calculated from spectroscopic data in ref.136 b Calculated from spectroscopic data in ref.128. 66 Table 3.6. Formation constants of Ga3+ complexes and pMk values. log K log KML pGa H2hox a 34.35(1) 28.3 H2dedpa b 28.11(8) 27.4 Oxine (KP46) c 13.13(8) *36.41(1) 21 NOTA d 30.98 27.9 DOTA e,f 21.33 e; 26.05 f 18.5; 19.50 TRAP g 26.24 23.1 HBED h,i 37.73 g; 39.57 h 27.7 g; 29.4 h DFO j 28.65 21.2 a This work (0.16 M NaCl at 25 °C). b From ref.25 (0.16 M NaCl at 25 °C). c From ref.128 (*) log K Ga(Lox)3 (0.20 M KCl at 25 °C). d From ref.137 (0.10 M KCl at 25 °C). e From ref.138 (0.1 M KCl at 25 °C). f From ref.101 (0.1 M (NMe4)Cl at 25 °C). g From ref.22. h From ref. 116. i From ref.139. j From ref.140. k Calculated at specific conditions ([Ga3+] = 1 µM, [Lx-] = 10 µM, pH 7.4 and 25 °C. 200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.00.950.610.520.750.330.270.19pHAbsorbanceWavelength (nm) = 249 nma)275 300 325 350 375 400 425 4500.000.050.100.150.200.950.610.520.330.19 pH AbsorbanceWavelength (nm) = 346 nmb) Figure 3.7. Representative spectra of the in batch UV spectrophotometric acidic titration for the Ga3+-H2hox system, 1:1 M:L molar ratio, [H2hox] = 2.82 × 10-5 M, (l = 1 cm) at 0.16 M NaCl and 25 °C. 67 Determination of the complex formation constant of [Ga(hox)]+ by direct pH-potentiometric method was not possible due to both the insufficient solubility of the Ga-H2hox mixtures and the great stability of the metal complex; UV spectrophotometry was applied to individual samples following the spectral changes on the ligand absorption bands in the pH range 0.19 to 11 (Figure 3.5). Samples containing the complex and different amounts of standardized HCl or NaOH were prepared and, due to the high stability of the complex even at very acidic pH, the samples were prepared and measured after 24 h, when equilibrium was achieved (see acid-assisted dissociation study in 3.3.5 dissociation kinetics of Ga3+ complex, Figure 3.9). The spectra of the Ga-hox complex collected from pH 0.19 to pH 0.95 showed identical features to those of the free ligand at the same pH values (Figure 3.7) suggesting the presence of the H6L4+ and H5L3+ species that deprotonate to the H4L2+ species (λmax = 245 nm and λmax = 308 nm) with two well-defined isosbestic points at λ = 249 nm and λ = 346 nm, respectively (Figures 3.5 a, 3.5 b, 3.7 a, 3.7 b). There is no complex formation evidence until pH > 1, when a new band appears at λ = 368 nm as well as two isosbestic points at λ = 251 and 335 nm and the shift of the band of the free ligand at λ = 243 nm to lower energies at 260 nm due to the complex formation (Figure 3.8 c-d). There is no further transformation along the pH range 2 to 10.5. These data, together with the molar absorption coefficients of the different absorbing species of the free ligand (Figure 3.6), allowed the determination of the stability constant of the Ga-hox complex using the HypSpec program86 (Table 3.6). 68 Figure 3.8. (a) Distribution diagram of H2hox calculated using the protonation constants in Table 3.5, at ligand concentration of 2.72 × 10-5 M; (b) Distribution diagram of the Ga-hox system, [Ga3+] = [H2hox] = 2.82 × 10-5 M; 25 °C; I = 0.16 M NaCl. (c) and (d) representative spectra in the batch UV spectrophotometric titration of the Ga3+-H2hox system, 1:1 M:L molar ratio, [H2hox] = 2.82 × 10-5 M, (l = 1 cm) at 0.16 M NaCl and 25 °C. In Figure 3.8 b is shown the speciation plot of the Ga-hox system calculated from the stability constant from Table 3.6. The absorption band with maxima at 368 nm and ε = 4468 M-1 cm-1 in the [Ga(hox)]+ complex, presents similarities with that reported by Enyedy128 for the Ga(Lox)3 species in water at I = 0.20 M KCl and 25 °C. The molar absorptivity Ga(Lox)3 (ε = 6627 M-1 cm-1, calculated from their spectra) is 1.5 times the obtained value for [Ga(hox)]+, which agrees with our results considering that in H2hox there are two hydroxyquinoline chromophores. The high log KML = 34.35(1) value of the [Ga(hox)]+ ion characterizes H2hox as a very strong chelator indeed for Ga3+. Even more interesting are the conditional stability constants or pM values 69 (defined as -log [Mfree] at [L] = 10 M and [M] = 1 M at pH = 7.4)92 which predict the stability of the complexes in vivo, and allow for the most suitable comparison of the relative ability of different ligands with different basicities to sequester a specific metal ion.141-142 In Table 3.6 are summarized the stability constants and calculated pM values for some of the most important 68Ga chelators. The Ga-hox pM value of 28.3, among the current reported 68Ga chelators pM values, falls in between the two literature reported values for HBED which has an even higher pM value for Ga3+. This fact, together with the presence in solution of a single species ([Ga(hox)]+) in the pH range 1-11 and considering that chelators such as 8-hydroxyquinoline128 or NOTA143 start to hydrolyze in the pH range 7-9, represent a distinct advantage for H2hox and strongly suggests that H2hox as a promising ligand for gallium radiopharmaceutical compounds in a toolkit application. 3.3.5 Dissociation Kinetics of Ga3+ Complex The acid-assisted dissociation kinetic study of the Ga(III)-hox complex was performed by UV spectrophotometry at 25 °C and 0.1 M HCl because of the high stability of the Ga-hox complex found even at very acidic pH. Concentrated standardized HCl was added to a Ga-hox stock solution ([Ga(hox)][ClO4] = 2.82 × 10-5 M) to achieve pH 1. The reaction was followed by registering the spectra at 15 min intervals at 25 °C. A first-order rate constant was found and the half-life determined in those conditions was 73.3 min. From the UV spectra, it is clear that there are no protonated complex species in the dissociation pathway by the presence of the well-defined isosbestics points previously depicted in the complex formation equilibria studies, and the complex dissociation leads to the ligand in its H4L2+ species. The absorbance of the samples for the in batch UV spectrophotometric titrations was measured after 24 hr when the equilibrium was completely achieved. 70 Figure 3.9. Time course of the UV spectra of the [Ga(hox)]+ complex (2.82 × 10-5 M) at pH 1 and 25 °C (spectra obtained at 15 min intervals). The inset shows the absorbance at 243 nm vs. time, and its fitting to the observed first-order rate constant. 3.3.6 68Ga Labelling Experiments Encouraged by the promising solution studies, 68Ga labeling studies were performed in order to investigate the coordination kinetics and radiolabeling efficiency, crucial properties for an ideal PET radiopharmaceutical based on 68Ga. These studies show that H2hox coordinates 68Ga quantitatively within 5 minutes at room temperature (faster than do H2dedpa, NOTA and DOTA). DOTA, however, requires a high temperature (≥ 90 °C) for quantitative yields and is thus incompatible with the labelling of thermally sensitive biomolecules.144 In our experiments, quantitative conversion (99% purity) (Figure 3.10) was achieved with ligand concentrations as low as 10-7 M and the molar activity obtained was as high as 11 ± 1 mCi/nmol without any purification. The molar activity for H2hox with 68Ga is even higher than that of H2dedpa (9.8 ± 1 mCi/nmol), which was reported to be the highest obtained of all 68Ga chelators with neither heating nor purification.25, 104, 145 Log D7.4 measurements showed [68Ga(hox)]+ was still a reasonably 200 225 250 275 300 325 350 375 400 425 4500.00.51.01.52.0275 300 325 350 375 400 425 4500.000.050.100.150.20 = 274 nmAbsorbanceWavelength (nm) = 335 nm0 5000 10000 15000 200000.51.01.52.0y = - 0.0001577 x + 0.212R² = 0.9996time (s)Abs at 243 nm-3-2-10 ln(Absmax- A) = 225 nm = 251 nm = 274 nmAbsorbanceWavelength (nm) = 335 nm71 hydrophilic complex, with an average log D7.4 of -0.47 ± 0.01 (n=4), even though it was more lipophilic than NOTA, DOTA and H2dedpa. Figure 3.10. HPLC traces of radiation and UV absorption of the mixtures of [68Ga(hox)]+ ([H2hox)] = 1  10-7 M) and [Ga(hox)]+ non-radioactive complex ([Ga(hox)]+ = 5  10-5 M). All of these chemistry advantages characterize H2hox as a superbly promising ligand for one-step kit-based labelling and led to in vitro and in vivo experiments. Knowing that a suitably substituted ethylenediamine precursor is available25, 49 to form an easily functionalizable (hox)2- ligand gave even stronger impetus to these studies. 3.3.7 Mouse Plasma Competition Experiments To investigate the in vitro stability of the 68Ga-hox system, a mouse plasma competition experiment was carried out. [68Ga(hox)]+ was incubated with mouse plasma for 5, 15, 30, and 60 minutes at 37 °C and analysed by radio HPLC. As shown in Table 3.7, the complex was completely intact (> 99%) at all time points, confirming its excellent in vitro stability of a whole half-life. Table 3.7. Mouse plasma stability of 68Ga-hox. 5 min 15 min 30 min 1 hr 68Ga-hox > 99% > 99% > 99% > 99% 72 3.3.8 Dynamic PET/CT imaging The high in vitro stability showed by [Ga(hox)]+, suggests high in vivo stability. A dynamic PET/CT imaging study in mice was therefore used to investigate the in vivo stability and biodistribution of the [68Ga(hox)]+ cation. As shown in Figure 3.11, dynamic PET/CT imaging showed no leakage of free 68Ga which would accumulate significantly in bone.146 There is no accumulation in muscle as well. The complex showed high heart signal intensity in the first two minutes followed by fast clearance via both hepatobiliary (liver, then gut) and renal (kidney, then bladder) pathways which could be explained by its small size and log D7.4 value (-0.47 ± 0.01). Even though the “naked” [Ga(hox)]+ is still a hydrophilic complex based on its negative log D value, the increased lipophilicity compared with most high polar multi-armed carboxylate based chelators such as NOTA and DOTA, may elicit higher liver uptake.20 However, recent studies with PSMA (prostate-specific membrane antigen) showed that the increased lipophilicity of the 68Ga-PSMA tracer using a more lipophilic HBED-CC chelator, translated into reduced unspecific binding and increased specific tumor cell uptake and imaging quality dramatically, compared with the conjugate using the hydrophilic DOTA chelator.114, 147 In research reported recently, Yoo and co-workers conjugated five different chelators with the same RGD peptide and investigated the effect of lipophilicity of the chelators on the biological behaviour of the bioconjugates.148 Their research also revealed that bioconjugates with more lipophilic chelators have higher tumor uptake as well as tumor/organ ratios even though the liver uptake was increased versus those bioconjugates with polar hydrophilic chelators. Another pertinent example comes from our own work; 111In-octapa-trastuzumab, using a more lipophilic picolinate-based chelator, showed a markedly higher tumor uptake than did 111In-DOTA-trastuzumab, which suggests that even for large bio-vectors, the effect of chelating moieties on the pharmacokinetic properties may not be 73 negligible.37 Therefore, we postulate that H2hox is a good complementary choice for the currently used chelator library in tuning the pharmacokinetics of the bioconjugates, especially for the small high polar targeting vectors like the Glu-urea-Lys PSMA inhibitors. The quick initial heart signal also suggests that H2hox could potentically be a useful scaffold in designing lipophilic cations for heart imaging, even though it can be difficult to distinguish between myocardial and blood pool activity in the heart in the early time point. Log P values between 0.8-1.2 have been hypothesized to be optimal for good heart imaging contrast tracers;149-150 however, all the chelators currently used are too hydrophilic for that application. We have tried to design a series of lipophilic cations based on the H2dedpa scaffold, but it is hard to increase the logP value to the targeted region, without increasing the molecular size too much.52,127 Radiotracers with appropriate log P have also been reported using a Schiff base ligand; however, complex stability was sacrificed, because the Schiff base complex was not metabolically stable.151-154 Therefore, we think, with this novel lipophilic H2hox scaffold, a simple modification on the aromatic ring will increase the lipophilicity and log P value, and would provide novel 68Ga-based candidates for heart imaging applications. 74 Figure 3.11. (a) PET/CT dynamic imaging and (b) biodistribution of [68Ga(hox)]+ in male mice during (b) 0 min – 3 min and (c) 3 min – 53 min. 3.3.9 [Ga(hox)]+ Fluorescence and Cell Imaging Studies H2hox shows chelation-enhanced fluorescence properties. As shown in Figure 3.12, the peak emission wavelength of H2hox in PBS buffer (pH = 7.4) was ~ 460 nm, and shifted to 560 nm once complexed with Ga3+, with a fourfold increase in intensity. This property of H2hox could encompass two major advantages over the other reported chelators. Firstly, it could significantly inform non-radioactive metal ion complexation by monitoring fluorescence changes and concomitant dissociation of metal ion in stability studies of bifunctional chelators, and therefore provide useful information without the need of radioisotope experiments. Secondly, the intrinsic 75 fluorescence could enable a bi-modal tracer and direct fluorescence imaging of labelled bioconjugates without an extraneous fluorescence tag that could alter biological behaviour by mutual disturbance of the two moieties and/or affect the pharmacokinetics of the radiopharmaceutical. One important advantage of a bimodal agent is that it can combine complementary information obtained from separate experiments to obtain a comprehensive synergistic analysis. For example, in studies of a somatostatin receptor imaging agent, the unexpected results from in vivo PET/SPECT imaging were finally explained by an optical fluorescence cellular imaging study of endocytotic uptake, benefitting from its high spatial resolution which cannot obtained by just PET/SPECT imaging.155 For clinical use, fluorescence techniques could also be useful through endoscopy or the surgical excision in fluorescence directed surgery.156 Most of the previously reported bimodal (optical imaging and PET/SPECT) agents require one fluorophore and one radioactive moiety.157-162 The extraneous fluorescence tag, however, may affect the metal sequestering capacity of the chelate moiety161 or require an extra spot for bioconjugation on the biovectors157, 162 and may change the pharmacokinetics of the whole tracer as well. Therefore, H2hox is a significant discovery to obviate this dual-modality-dual-probe problem. To prove this concept, we investigated the subcellular distribution and stability of [Ga(hox)]+ in living HeLa cells using fluorescence microscopy. HeLa cells were incubated with 150 μM [Ga(hox)][ClO4] for 2 or 24 h. Bright field images of treated cells (Figure 3.13) taken prior to fluorescence imaging verified the cells as viable. The morphology of HeLa cells appeared normal and suggested a low cellular toxicity of [Ga(hox)][ClO4]. The fluorescence imaging was taken with a 520 nm emission filter which allowed the detection of fluorescence at wavelengths > 520 nm (Figure 3.12). The complex was found to accumulate in cytoplasm, as Enyedy et al.128 noted 76 for the anticancer Ga(Lox)3 complex. No obvious decomposition of the [Ga(hox)][ClO4] complex was observed within the 24 hr cellular environment, as the fluorescence intensity was constant and the free ligand would otherwise exhibit markedly lower fluorescence intensity at wavelengths above 520 nm (Figure 3.12). This proof-of-concept study shows that this chelation-enhanced fluorescence property could be used directly in intracellular distribution and stability studies. The intracellular distribution could provide important preparatory information for many other studies; for instance, Auger electron-based therapy requires localization of the radionuclides in the nucleus, or hypoxia imaging tracers which could also be evaluated using fluorescent imaging with a multi-cellular spheroid tumor model before radioactive in vivo study.161 Figure 3.12. Fluorescence spectra of H2hox and its Ga3+ complex in PBS (λexc = 365 nm, [H2hox] =[[Ga(hox)]+] = 1.7  10-5 M) (upper). Time dependent fluorescence microscopy images from Hela cells treated with 150 M [Ga(hox)] [ClO4] (scale = 20 m) (lower). 77 Figure 3.13. Bright field images from Hela cells incubated with [Ga(hox)][ClO4] (scale = 20 m). 3.4 Conclusions In summary, H2hox, as a next generation bimodal acyclic chelating ligand for Ga3+ complexation, displays an unprecedented array of properties. The synthesis of H2hox is easier and more straightforward than that for any previously reported analogous chelator. The higher thermodynamic stability of the [Ga(hox)]+ complex (log KML = 34.4) compared to that of most relevant Ga3+ chelators, and most importantly, the largest pM value of 28.3 among those ligands proves the affinity of H2hox towards Ga3+. Moreover, H2hox showed fast and quantitative 68Ga radiolabelling at room temperature and low concentration (10-7 M) yielding a remarkably high molar activity of 11 ± 1 mCi/nmol without any purification. Most importantly, mouse serum stability experiments and dynamic PET imaging studies have shown high in vitro and in vivo stabilities that correlate with the high thermodynamic stability found in the solution studies. [68Ga(hox)]+ is also quickly cleared from the mouse via hepatobiliary and renal pathways. The lipophilicity of H2hox provides a choice complementary to the current library of chelators, to tune the pharmacokinetics of bioconjugates and design lipophilic tracers for heart imaging or cell 78 labelling. The intrinsic fluorescence of [Ga(hox)]+ imaged HeLa cancer cells showing accumulation in the cytoplasm and suggests strongly that this compound could serve in dual channel (bimodal) imaging or fluorescence directed surgery. The fluorescence emission red shift and intensity increase (chelation enhanced fluorescence) upon the complexation of H2hox with Ga3+ could be directly used to study the stability of a non-radioactive bioconjugate in vitro. The high affinity of H2hox for Ga3+, confirmed in solution as well as in vitro and in vivo, together with the fast quantitative radiolabelling at low concentrations and mild conditions encourage the use of H2hox for development of a convenient toolkit radiopharmaceutical compound. Current efforts are focused on bifunctional analogues of the H2hox scaffold. 79 Chapter 4: H2CHXhox, Cyclohexane Reinforced Chelating Ligand for Ga3+ and Cu2+ 4.1 Introduction The application of many acyclic chelators in nuclear medicine, for example DTPA, has been limited by their poor kinetic inertness, even though they showed fast and quantitative radiometal labeling compared with macrocyclic chelators such as DOTA and NOTA.163-164 As discussed in Chapter 3, one important strategy to overcome this limitation is to “preorganize” the flexible open chain structure by incorporating a rigid backbone group, decreasing the entropy penalty for wrapping it around the metal ion. A good example is CHX-A”-DTPA which is derived from DTPA by adding a 1,2-trans-cyclohexanediamine backbone. This modification resulted in a marked increased kinetic inertness while still keeping the acyclic character.164-167 In our group’s previous work, H2CHXdedpa was synthesized by substitution of a chiral 1R,2R-transcyclohexanediamine in place of the ethylenediamine backbone of H2dedpa; this led to better in vitro kinetic inertness and thermodynamic stability.55 More than 91% of [67Ga(CHXdedpa)]+ was intact in a 2 h serum challenge experiment, improved from 78% of [67Ga(dedpa)]+ (Table 4.1). A new generation of acyclic bimodal chelators containing the oxine moiety was reported in Chapter 3. The development of this generation was led by H2hox, which itself was inspired by the Ga(oxine)3 complex. H2hox possesses promising characteristics to develop kit-based pharmaceuticals. Its synthesis is easy and fast. Solution phase thermodynamic studies reveal the presence of a single complex species in a broad pH range (1-11) with higher log KML (34.4) and pM values (28.3) than for most Ga3+ chelators. H2hox showed fast and quantitative 68Ga 80 complexation at mild conditions (5 min, RT) with a concentration as low as 10-7 M and an unprecedented high molar activity without purification.168 In this Chapter, further improvement in kinetic inertness and thermodynamic stability was attempted based on this platform with H2CHXhox, incorporating a 1S,2S-transcyclohexanediamine backbone, following the same strategy discussed above. Scheme 4.1. DOTA, NOTA, DTPA, H2dedpa and cyclohexane reinforced chelators. Table 4.1. Stability of [67Ga(dedpa)]+and [67Ga(CHXdedpa)]+ in human serum at 37 °C.55 Complex 1 h (%) 2 h (%) [67Ga(dedpa)]+ 94.8 ± 3.4 77.8 ± 1.5 [67Ga(CHXdedpa)]+ 95.7 ± 0.7 90.5 ± 4.4 Additionally, the increased lipophilicity results in higher heart signal in the PET/CT imaging studies compared with H2hox, in agreement with the previous finding that the lipophilic cation could be useful for myocardial perfusion imaging.151, 169-171 Nowadays 99mTc-sestamibi is routinely 81 used as a SPECT tracer for myocardial damage imaging; however, as described in the Chapter 3, there is a risk of interrupted supply of 99Mo, the parent isotope of 99mTc, leading to a need for alternative radioisotopes. Moreover, PET could offer a higher spatial resolution, accurate correction of photon attenuation and possible regional myocardial blood flow (MBF) quantification in absolute terms compared with SPECT.172-173 For these reasons, myocardial perfusion imaging tracers based on 68Ga, a generator-produced PET emitter, have drawn a lot of interest and effort in the past few decades.52, 151, 172-179 No successful 68Ga radiopharmaceuticals suitable for myocardial perfusion imaging (MPI) have been reported yet. Some 68Ga-labeled compounds have shown high myocardial uptake, but have rapid clearance.172 Based on the experience from 99mTc radiotracers, a good heart imaging tracer should be a small and lipophilic cation with a log P value in the range of 0.8-1.2 in order to achieve a high heart uptake and long retention.149 These criteria are hard for Ga3+-based tracer design. Ga3+ is a small hard metal ion, and a stable chelator usually contains several polar oxygens, for example carboxylate oxygens and thus is hydrophilic. Generally, two strategies have been applied to achieve these criteria. The first one is to use Schiff-base chelators which are quite hydrophobic (as summarized in Scheme 4.2173); however, Schiff-base linkages are vulnerable to hydrolysis in the physiological environment and therefore the tracers are not stable. The second strategy is to increase the lipophilicity of the stable chelators, for example H2dedpa,52, 179 by adding hydrophobic functional groups. Most of the previous reported stable chelators for Ga3+ are hydrophilic, making it hard to adjust the log P value to the right range using small hydrophobic functional groups. For example, in our group’s previous study, the H2dedpa derivatized tracers have been designed as summarized in Scheme 4.3, and showed that it was hard to increase the log P to a proper value without using large functional 82 groups which may interfere with the chelate stability, even though underivatized H2dedpa is already more hydrophobic than some other chelators such as NOTA and DOTA.52, 179 Inspired by the in vivo imaging result of H2hox and H2CHXhox reported in Chapter 3 and this Chapter, a new group of halogenated derivatives with expanded lipophilicity, H2Br4hox and H2Br4CHXhox are developed in this Chapter, as is their potential application in myocardial perfusion imaging. Scheme 4.2. Schiff-base amine chelators and log D values of their 68Ga complexes.154, 173 83 Scheme 4.3. H2dedpa derivatives and log P values of their 67Ga complexes.52, 127 4.2 Experimental 4.2.1 Materials and Methods All solvents and reagents were purchased from commercial sources (TCI America, Sigma Aldrich, Fisher Scientific) and were used as received unless otherwise indicated. The analytical thin-layer chromatography (TLC) plates used were aluminum-backed ultrapure silica gel 60 Å, 250 μm thickness; 1H and 13C NMR spectra were recorded at ambient temperature on Bruker Avance 300 and Avance 400 spectrometers. The 1H NMR spectra were calibrated against respective residual protio-solvent peaks, and the 13C NMR spectra were referenced to the deuterated solvent. Low-resolution mass spectrometry was performed on a Waters ZG spectrometer with an ESCI (electrospray/chemical-ionization) source, and high-resolution 84 electrospray ionization mass spectrometry (ESI-MS) was performed on a Micromass LCT time-of-flight (TOF) instrument. Microanalysis for C, H, and N was performed on a Carlo Erba Elemental Analyzer EA 1108. Purification and quality control of [68Ga(CHXhox)]+ were performed on an Agilent HPLC system equipped with a model 1200 quaternary pump, a model 1200 UV absorbance detector, and a Bioscan (Washington, DC) NaI scintillation detector. The radiodetector was connected to a Bioscan B‐FC‐1000 Flow‐count system, and the output from the Bioscan Flow‐count system was fed into an Agilent 35900E Interface which converted the analog signal to digital signal. The operation of the Agilent HPLC system was controlled using the Agilent ChemStation software. The HPLC columns used were a semi‐preparative column (Phenomenex C18, 5 µ, 250 × 10 mm) and an analytical column (Phenomenex C18, 5 µ, 250 × 4.6 mm). The HPLC solvents were A: H2O containing 0.1% TFA, and B: CH3CN containing 0.1% TFA. 4.2.2 Ligand Synthesis and Characterization H2CHXhox (4.1) 8-Hydroxyquinoline-2-carboxaldehyde (2.00 g, 11.6 mmol) was dissolved in 50 mL methanol and 1S,2S-transcyclohexanediamine (696 µL, 5.8 mmol) dissolved in 5 ml of methanol was added dropwise; the reaction mixture was stirred at 60 °C for 4 h. A light-yellow precipitate formed, was collected and resuspended in 50 mL methanol. Eight equivalents of NaBH4 (3.50 g, 92.8 mmol) were added in portions and the reaction mixture was stirred at room temperature overnight. HCl (20 mL, 6 M) was added and the reaction mixture stirred for 4 h. The pH of the reaction mixture was then readjusted to neutral using NaOH (2 M) and the off-white precipitate was collected by filtration and dried as crude product. The crude product was further washed with water and methanol to 85 obtain the pure product (2.05 g, 4.8 mmol), yield = 83%. 1H NMR (300 MHz, MeOD, 25 °C) δ 8.15 (d, J = 8.5 Hz, 1H), 7.40 (d, J = 8.5 Hz, 1H), 7.36 – 7.21 (m, 2H), 6.96 (dd, J = 6.9, 2.0 Hz, 1H), 4.58 – 4.30 (m, 2H), 2.81 – 2.60 (m, 1H), 2.23 (d, J = 10.7 Hz, 1H), 1.74 (d, J = 6.8 Hz, 0H), 1.28 (q, J = 9.3, 7.8 Hz, 2H). 13C NMR (75 MHz, MeOD, 25 °C) δ 157.1, 153.0, 137.9, 136.8, 128.3, 127.0, 120.9, 117.6, 111.0, 60.6, 50.4, 30.6, 24.7. LR-ESI-MS calcd. for [C26H28N4O2 + H]+:429.2; found 429.2 H2Br4hox (4.2) H2hox (30 mg, 0.08 mmol) was dissolved in 30 mL methanol at 60 °C and cooled to 30 °C, and bromine (68 µL, 1.28 mmol) in 5 mL methanol was added dropwise. The mixture was stirred at 30 °C for one hour and then Na2SO3 (630 mg, 5 mmol) was added. After 10 min, 35 mL of water was added and the precipitate was filtered out and washed with water and methanol, and dried then to obtain the pure product (52 mg, 0.075 mmol), yield = 94%.1H NMR (300 MHz, D2O, 25 °C) δ 8.12 (d, J = 8.6 Hz, 1H), 7.85 (s, 1H), 7.26 (d, J = 8.6 Hz, 1H), 3.87 (s, 2H), 2.74 (s, 2H). LR-ESI-MS calcd. for [C22H18Br4N4O2 + H]+: 690.8; found 690.8. H2CHXBr4hox (4.3) H2CHXhox (30 mg, 0.07 mmol) was dissolved in 30 ml methanol at 60 °C and cooled to 30 °C, and bromine (60 µL, 1.16 mmol) in 5 mL methanol was added dropwise. The mixture was stirred at 30 °C for one hour and then Na2SO3 (630 mg, 5 mmol) was added. After 10 min, 35 mL of water was added and the precipitate was filtered out, washed with water and methanol, then dried to obtain the pure product (50.6 mg, 0.068 mmol), yield = 97%. 1H NMR (300 MHz, MeOD, 86 25 °C) δ 8.32 (d, J = 8.7 Hz, 1H), 7.77 (s, 1H), 7.45 (d, J = 8.7 Hz, 1H), 4.53 – 4.26 (m, 2H), 2.86 (d, J = 8.1 Hz, 1H), 2.38 – 2.12 (m, 1H), 1.84 (d, J = 7.2 Hz, 1H), 1.53 – 1.12 (m, 2H). LR-ESI-MS calcd. for [C26H24Br4N4O2 + H]+: 744.9; found 744.8. 4.2.3 Synthesis of CHXhox-Metal Complex [Ga(CHXhox)][ClO4](4.4) H2CHXhox (40 mg, 0.11 mmol) was dispersed in 5 mL methanol. Ga(ClO4)3·6H2O (55 mg, 0.11 mmol) was added as a solid, and the pH was adjusted to ~ 6 using 0.1 M NaOH. The reaction mixture was stirred for 1 h at 50 °C and 5 mL of H2O was added after the solution had cooled to room temperature. CH2Cl2 (3 × 5 mL) was then added to extract the product and the organic phase was dried in vacuo to obtain a yellow crystalline powder as the final product. 1H NMR (300 MHz, MeOD, 25 °C) δ 8.80 (d, J = 8.5 Hz, 1H), 7.89 (d, J = 8.5 Hz, 1H), 7.59 (t, J = 8.1 Hz, 1H), 7.44 – 7.31 (dd, J = 7.8, 0.8 Hz, 1H), 6.96 (dd, J = 7.8, 0.8 Hz, 1H), 4.73 – 4.46 (m, 2H), 2.44 (d, J = 10.4 Hz, 1H), 2.18 (d, J = 7.7 Hz, 1H), 1.81 (s, 1H), 1.27 (m, J = 12.1 Hz, 2H). LR-ESI-MS calcd. for [C26H2669GaN4O2]+: 495.1; found 495.2. [Cu(CHXhox)](4.5) H2CHXhox (40 mg, 0.11 mmol) was dissolved in 5 mL methanol. CuSO4·5H2O (29 mg, 0.11 mmol) was added as a solid and the pH was adjusted to ~ 8 using 0.1 M NaOH. The reaction mixture was stirred for 1 h at 50 °C and 5 mL of H2O was added after the solution was cooled to room temperature. CH2Cl2 (3 × 5 mL) was added to extract the product and a red needle crystal suitable for X-ray diffraction formed from layer diffusion of diethyl ether into the CH2Cl2 extraction layer. 87 4.2.4 X-ray Crystallography A red needle crystal of C26H26CuN4O2, approximately 0.08 × 0.14 × 0.24 mm, was mounted on a cryo-loop. All measurements were made on a Bruker APEX DUO diffractometer with a TRIUMPH curved-crystal monochromator using Mo-Kα radiation. The data were collected at a T = -183.0 ± 0.1 °C to a maximum 2 value of 60.2°. Data were collected in a series of  and  scans in 0.5° oscillations using 15.0-second exposures. The crystal-to-detector distance was 40.16 mm. Of the 104329 reflections that were collected, 30586 were unique (Rint = 0.041); equivalent reflections were merged. Data were collected and integrated using the Bruker SAINT69 software package. The linear absorption coefficient, , for Mo-K radiation is 8.65 cm-1. Data were corrected for absorption effects using the multi-scan technique (SADABS180), with minimum and maximum transmission coefficients of 0.843 and 0.933, respectively. The data were corrected for Lorentz and polarization effects. The structure was solved by direct methods.71 The material crystallizes as a two-component pseudo-merohedral twin. Refinements were carried out using the following twin law (-1 0 0 0 -1 0 0 0 1) in conjunction with a BASF of ~ 0.36. The material also crystallizes with four crystallographically independent molecules in the asymmetric unit. Additionally, there are sites in the asymmetric unit occupied by solvent (either CH2Cl2 or Et2O). The solvent molecules undergo significant thermal motion, and in some cases the sites are partially occupied by both solvents. The solvent regions could not be reasonably modeled, therefore the PLATON/SQUEEZE72 program was employed to generate ‘solvent-free’ data. All non-hydrogen atoms were refined anisotropically. All N-H hydrogen atoms were located in difference maps and refined isotropically; however, some required constraints on their isotropic displacement parameters to maintain reasonable geometries. All other hydrogen atoms were placed in calculated positions. The 88 absolute configurations at C11, C16, C37, C42, C63, C68, C89 and C90 were confirmed on the basis of the refined Flack x-parameter [0.001(3)]. The final cycle of full-matrix least-squares refinement on F72 was based on 30586 reflections and 1217 variable parameters and converged (largest parameter shift was 0.00 times its esd) with unweighted and weighted agreement factors of: R1 (I>2.00(I)) =  ||Fo| - |Fc|| /  |Fo| = 0.038 4.1 wR2 (all data) = [  ( w (Fo2 - Fc2)2 )/  w(Fo2)2]1/2 = 0.083 4.2 The standard deviation of an observation of unit weight was 1.00. The weighting scheme was based on counting statistics. The maximum and minimum peaks on the final difference Fourier map corresponded to 0.73 and -0.53 e-/Å3, respectively. Neutral atom scattering factors were taken from Cromer and Waber.181 Anomalous dispersion effects were included in Fcalc;182 the values for f' and f\" were those of Creagh and McAuley.183 The values for the mass attenuation coefficients are those of Creagh and Hubbell.184 All refinements were performed using the SHELXL-2016185 via the OLEX273 interface. 4.2.5 Solution Thermodynamics Protonation equilibria of the ligand and complex formation equilibria were studied by UV spectrophotometric batch experiments as described before168 using a Cary 60 UV-vis spectrophotometer in the spectral range 200-450 nm, at 25 °C and 1 cm path length (l). A set of 49 solutions containing the ligand (H2CHXhox, 2.75  10-5 M) in water were prepared and different amounts of standardized HCl or NaOH were added to cover a range from H0 -0.52 to pH 11.51. The ionic strength of each sample was adjusted (when possible) to 0.16 M by addition of different 89 amounts of NaCl. In the most acidic samples (below pH 0.6), it was not possible to maintain constant ionic strength since that depends on the HCl content, and for those solutions the correct acidity scale H0 was used.84 In the samples < pH 2, the equilibrium H+ concentration was calculated from solution stoichiometry and for the rest of the samples, pH was measured with a Ross combination glass electrode that was calibrated daily for hydrogen ion concentrations using HCl as described before80 and the results were analyzed by the Gran81 procedure. For the complex formation equilibria, the set of solutions were prepared in the same way as described above. For the Ga(III)-H2CHXhox and Cu(II)-H2CHXhox systems, the set of samples were prepared by solving the corresponding [Ga(CHXhox)][ClO4] or [Cu(CHXhox)] complexes, while in the case of the Cu(II)-H2hox system, the set of ligand-metal solutions were prepared by adding the atomic absorption (AA) standard metal ion solution to a H2hox solution of known concentration in the metal-to-ligand molar ratios 1:1. The exact amount of acid present in the copper standard was determined by Gran’s method81 titrating equimolar solutions of Cu(II) and Na2H2EDTA. Equilibration time of 2 min for the ligand protonation equilibria study was allowed before measuring the pH and the UV absorption spectrum. For the complex formation equilibria with Cu(II), the equilibration time was up to 5 min, while for the samples containing the Ga(III) complexes, the measurements were performed only after three weeks because of the longer time to equilibrate the most acidic samples. The spectral data were analyzed using the HypSpec2014 program.86 Proton dissociation constants corresponding to hydrolysis of Ga(III) and Cu(II) aqueous ions included in the calculations were taken from Baes and Mesmer.87 The species formed in the studied systems are characterized by the general equilibrium: pM + qH + rL = MpHqLr (charges omitted). For convention, a complex containing a metal ion, M, proton, H and ligand, L, 90 has the general formula MpHqLr. The stoichiometric indices p, might also be 0 in the case of protonation equilibria and negative values of q refers to proton removal or hydroxide ion addition during formation of the aqua complex. The overall equilibrium constant for the formation of the complexes MpHqLr from its components is designated as log β. Stepwise equilibrium constants log K correspond to the difference in log units between the overall constants of sequentially protonated (or hydroxide) species. A more straightforward comparison of the ability of different ligands to coordinate a specific metal ion than the thermodynamic stability constants alone, is the use of the pM values, defined as (-log[Mn+]free) and is calculated at specific conditions ([Mn+] = 1 µM, [Lx-] = 10 µM, pH 7.4 and 25 °C), taking into consideration metal-ligand association and ligand basicity.92 4.2.6 PET/CT Imaging Studies PET/CT imaging studies were conducted at BC Cancer in accordance with the guidelines established by the Canadian Council on Animal Care and approved by the Animal Ethics Committee of the University of British Columbia. Male NOD.Cg-Rag1tm1MomIl2rgtm1Wjl/SzJ (NRG) mice were purchased from in-house colonies at the Animal Research Centre, BC Cancer Research Centre, Vancouver, Canada. PET/CT imaging experiments were conducted using a Siemens (Knoxville, TN, USA) Inveon microPET/CT scanner. Mice were sedated with 2% isoflurane in oxygen inhalation and positioned in the scanner. A baseline CT scan was obtained for localization and attenuation correction before radiotracer injection, using 80 kV X-rays at 500 mA, three sequential bed positions with 34% overlap, and 180-degree continuous rotation. The mice were kept warm by a heating pad during acquisition. The dynamic acquisition of 60 min was started at the time of intravenous injection with ~ 3.4-4.0 MBq of [68GaCHXhox]+. The list mode data were rebinned into time intervals (12 × 10 sec, 8 × 60 sec, 7 × 300 sec, 1 × 900 sec) to obtain 91 tissue time-activity curves. Images were reconstructed using iterative 3-dimensional ordered subset expectation maximization (OSEM3D, 2 iterations) using maximum a priori with shifted poisson distribution (SP-MAP, 18 iterations). 4.2.7 SPECT/CT Imaging Studies This was performed at the Center for Comparative Medicine, UBC, in accordance with the Canadian Council on Animal Care (CCAC) and protocol approved by the Animal Care Committee (ACC) of the University of British Columbia (A16-0150). Three healthy C57Bl/6 female mice (~ 16 g) were anaesthetized using isoflurane on a precision vaporizer (5% in oxygen for induction, between 1.5 and 2.5% in oxygen for maintenance) and received a subcutaneous injection of lactated Ringer’s solution (0.5 mL) for hydration prior to each imaging scan. Dynamic whole-body images were acquired during 60 min using a multimodal SPECT/CT scanner (VECTor/CT, MILabs, the Netherlands) equipped with a XUHS-2 mm mouse pinhole collimator. Six frames of 10 min were acquired for the first hour scan. Following each SPECT acquisition, a whole-body CT scan was acquired to obtain anatomical information and both images were registered. For the SPECT images, 16 subsets, 10 iterations and an isotropic 0.4 mm voxel grid were used. The images were decay-corrected and after CT registration, attenuation correction was applied. For visual representation, the reconstructed volumes of SPECT scans were post-filtered with a 3D Gaussian filter. CT scans were acquired with a tube setting of 55 kV and 615 µA. 4.3 Results and Discussion H2hox is the most successful acyclic chelator for gallium labeling in our group’s past 10 years study. It quickly and quantitively chelates Ga3+ at mild conditions (5 minutes, RT) with high thermodynamic stability (log KML = 34.4 and pM = 28.3). Compared with H2dedpa, the previous 92 generation,168 it is more rigid and the preorganized structure enhances its kinetic inertness. The dissociation half-life time at pH = 1 is 73.1 minutes. While this is an encouraging result, it still cannot compare to macrocyclic chelators such as NOTA and DOTA that have even higher kinetic inertness due to the more fixed structure macrocycles innately possess. In this work, H2CHXhox, a more preorganized ligand, was designed by incorporating a 1S,2S-transcyclohexanediamine backbone to H2hox as shown in Scheme 4.4. H2CHXhox is quite a preorganized structure, with only 2 carbons as flexible joints between the two rigid 8-hydroxyquinoline arms and the “fixed” backbone diamine. This structure is expected to mix the advantages of both acyclic and macrocyclic chelators, achieving the fast labeling and excellent kinetic inertness at the same time. Scheme 4.4. Design of H2CHXhox. 4.3.1 Synthesis and Characterization of Ligands The racemic mixture of H2CHXhox has been tested as a chelator for the mobilization of iron from cells;186 however, the single enantiomer has not been prepared and its chelation properties have not been investigated until this report. Herein, as in the modified synthetic route of H2hox reported in Chapter 3, enantiomerically pure H2CHXhox was prepared by reductive amination of 8-hydroxyquinoline-2-aldehyde and 1S,2S-transcyclohexanediamine in one step (Scheme 4.5). The crude product precipitated out of solution after the reaction was quenched with HCl (6 M) and neutralized with NaOH (2 M). After washing with H2O and methanol, a pure product was obtained 93 with a total yield of 83%. Both starting materials were cheaply purchased from a commercial supplier. Grams of ligand are prepared quickly and economically, which easily enables further investigation and future application of H2CHXhox. As shown in Figure 4.1, the 1H NMR spectrum of H2CHXhox revealed the expected C2 symmetry with half-integrations of the resonances present, the incorporation of the chiral cyclohexane ring results in a typical diastereotopic splitting of protons (Figure 4.1, protons H and I) observed α to the chiral center. This also agrees with the previously reported 1H NMR spectrum of H2CHXdedpa.55 Scheme 4.5. Synthesis of H2CHXhox. a) en, CH3OH, 60 °C, 4 hr; b) CH3OH, NaBH4 (5 equiv), overnight. Figure 4.1. 1H NMR spectrum of H2CHXhox in MeOD (300 MHz, 25 °C) 94 Scheme 4.6. Preparation of H2Br4hox and H2CHXBr4hox. To further explore the potential application of this new generation of chelators in myocardial perfusion imaging, two more hydrophobic chelators, H2Br4hox and H2Br4CHXhox, which are each tetrabromine derivatives of H2hox and H2CHXhox, were then prepared following a modified literature procedure as shown in Scheme 4.4.187 The starting materials, H2hox and H2CHXhox, which generally have poor solubility in most solvents, are be dissolved in hot methanol at 60 °C with the help of ultrasonication first. The solution was cooled down to 30 °C and liquid Br2 in methanol was then added dropwise in 15 min. The mixture was stirred for one hour at 30 °C and the product was precipitated after quenching with Na2SO3 and H2O.188 The reaction was quite fast and almost quantitive conversion was achieved. No further purification is required after a thorough washing with H2O and methanol. As expected, both of the two products were found to be more hydrophobic than their parent compounds, showing decreased solubility in H2O and methanol and increased solubility in CH2Cl2. Br2, MeOH 95 4.3.2 Preparation and Characterization of Metal Complexes H2CHXhox and Ga(ClO4)3 were mixed in a 1:1 molar ratio in methanol and the pH was adjusted to around 6 using NaOH (0.1 M), the mixture was stirred for 1 h at 50 °C to ensure a thorough reaction. [Ga(CHXhox)][ClO4] was then extracted by CH2Cl2 and dried in vacuo to obtain yellow crystalline powder as the final product. The 1H NMR spectrum confirmed the metal complex formation. As shown in Figures 4.1 and 4.2, homotopic proton (F) on the free ligand (Figure 4.1) is converted to the diastereotopic protons (F1 and F2) in the [Ga(CHXhox)][ClO4] complex (Figure 4.2). The J-coupling value of the two protons is 17.4 Hz, which is consist with the previously reported [GaCHXdedpa][ClO4] complex.55 This indicates the formation of the chelate rings which split the orientation of the two protons. The aromatic protons are shifted downfield in the metal complex (Figure 4.2) due to the decreased electron density on the hydroxyquinoline ring, which also confirms the binding of the two arms to the Ga3+. Meanwhile proton G on the chiral carbon is shifted upfield by about 0.3 ppm because of the increase in shielding from the coordination bond Cu-N(en). This suggests the two backbone nitrogen atoms are also binding to the central Ga3+. Recystallization of this powder from a CH2Cl2/Et2O system yielded very thin needle shaped crystals, Regrettably, no single crystal was good enough for X-ray crystallography. [Cu(CHXhox)] was prepared by mixing CuSO4·5H2O and H2CHXhox in a 1:1 molar ratio in methanol. The reaction mixture was stirred for 1 h at 50 °C, and the pH was adjusted to around 8 using 0.1 M NaOH. A mixture of H2O and CH2Cl2 was then added to extract the product. The dark blue complex entered into the CH2Cl2 layer while the aqueous layer showed a very light blue color because of the trace amount of remaining copper salt. Unlike the cationic [GaCHXdedpa]+ 96 complex, the [Cu(CHXhox)] complex is neutral with no net charge at this pH; therefore, it is more hydrophobic, as was observed by the fast and effective transfer into the CH2Cl2 phase. Figure 4.2. 1H NMR spectrum of [Ga(CHXhox)][ClO4] in MeOD (300 MHz, 25 °C). Red needle crystals of [Cu(CHXhox)] suitable for X-ray diffraction were obtained by layer diffusion of diethyl ether into the CH2Cl2 extraction layer. To make a comparison, orange flake shape crystals of [Cu(hox)] were prepared following the same procedure. The solid state structures of [Cu(hox)] and [Cu(CHXhox)] as determined by X-ray diffraction are shown as ORTEP diagrams in Figures 4.3 and 4.4, respectively, and the coordination bond lengths are summarized in comparison with [Cu(dedpa)] and [Cu(CHXdedpa-N,N’-propyl-2-NI)] in Table 4.2.55, 189 The absence of a counterion in the crystallographic unit confirmed that both are neutral complexes. All the donor atoms (N4O2) were fully coordinated and form distorted octahedral complexes, as in [Cu(dedpa)] and [Cu(CHXdedpa-N,N’-propyl-2-NI)] reported before.55, 189 97 In the [Cu(hox)] crystals (Figure 4.3) only one crystallographically independent structure was found, and the complex molecule kept the C2 symmetry of the ligand with the Cu atom residing on a two-fold rotation axis. The bonds formed between the Cu2+ and the two ethylenediamine nitrogen atoms (Cu-Nen) were elongated. This is quite different from the [Cu(dedpa)] and [Cu(CHXdedpa-N,N’-propyl-2-NI)] structures (Table 4.2), in which one of the Cu-Nen and one of the Cu-Ocoo bonds were elongated due to Jahn-Teller distortion, changing the symmetry slightly.55, 189 This could be explained by the high rigidity of the two 8-hydroxyquinoline arms and relative flexibility of the two ethylenediamine nitrogen atoms in H2hox, while in the picolinic acid-based chelators the carboxylate oxygen is still relatively flexible to adapt to the elongation. In [Cu(CHXhox)], as shown in Figure 4.4, four crystallographically independent molecules structures have been identified. The cyclohexane backbone takes a stable chair conformation in all four. The C2 symmetry seems to be broken and the bond lengths are slightly asymmetric (Table 4.2) due to the high rigidity of the whole chelator. As shown in Table 4.2, almost all (5 out of 6) of the coordination bond lengths are shorter in [Cu(CHXhox)] than those in [Cu(hox)]. When compared with [Cu(CHXdedpa-N,N’-propyl-2-NI)], which also has a cyclohexane backbone reinforcement, the bond lengths in [Cu(CHXhox)] are more evenly distributed and 4 out 6 of the coordination bonds are shorter, especially the Cu-Oox bond. This analysis suggests a higher stability of [Cu(CHXhox)] compared with the other three. 98 Figure 4.3. ORTEP-style picture of [Cu(hox)]. Figure 4.4. ORTEP-style picture of [Cu(CHXhox)]4. 99 Table 4.2. The bond lengths of the four Cu complex structures. [Cu(hox)] [Cu(CHXhox)] [Cu(dedpa)]55, 189 [Cu(CHXdedpa-N,N’-propyl-2-NI)]187 Bonds Length/Å Length/Å Bonds Length/Å Length/Å Cu-O(ox) 2.1427 2.132 Cu-O(COO) 2.3014 2.110 Cu-O(ox) 2.1427 2.159 Cu-O(COO) 2.0430 2.230 Cu-N(ox) 1.9740 1.945 Cu-N(pyr) 2.0008 1.937 Cu-N(ox) 1.9740 1.9432 Cu-N(pyr) 1.9386 1.968 Cu-N(en) 2.3275 2.287 Cu-N(en) 2.3171 2.311 Cu-N(en) 2.3275 2.3165 Cu-N(en) 2.1364 2.342 In(oxine)3·EtOH [Cu(CHXhox)] Figure 4.5. Comparison of solid structure of In(oxine)3190 and [Cu(CHXhox)]. Interestingly, 111In(oxine)3 has been used in clinical blood cell SPECT imaging ever since the mid-1970s.191 It is used as a nonspecific blood cell labelling agent since it is neutral and a highly hydrophobic complex. As shown in Figure 4.5, the solid structure of [Cu(CHXhox)] is quite similar 100 to that of In(oxine)3190 and more importantly, [Cu(CHXhox)] is a neutral and hydrophobic complex as well. Therefore, it may be useful as an intermediate time blood cell PET tracer, considering the 12.7 h half-life time of 64Cu. However, to make this possible, it will require kinetic instability which should be investigated in the future. 4.3.3 Solution Study 4.3.3.1 Protonation constants of H2CHXhox Due to the low solubility of H2CHXhox, the pH potentiometric technique was not suitable for the determination of the protonation constants. UV-in-batch spectrophotometric titration was carried out as in the case of H2hox.168 The spectra collected during the titration (Figure 4.6) show the same spectral evolutions as H2hox marked by the appearance of different isosbestic points. Analysis of the spectrophotometric data with the HypSpec2014 software86 allowed the determination of the six protonation states and generated the corresponding speciation plots (Table 4.3 and Figure 4.7). The largest variances in pKa values of H2CHXhox versus H2hox reside on the Nox atoms of the quinoline units and a slight difference on the Nen atoms in the backbone. The increased rigidity in H2CHXhox might favor the deprotonation of the Nox atom that could be stabilized by hydrogen bonding with the protonated Nen in the backbone. This may explain the lower pKa values obtained for the H6L4+ and H5L3+ species: log K6 = -0.36(3) and log K5 = 0.15(2). The first deprotonation of the Nen atom, species H4L2+ (log K4 = 5.95(1)) in H2CHXhox is fairly similar to that in H2hox, while the second Nen deprotonation, species H3L+ (log K3 = 8.75(1)) is 0.36 units higher than the correspondent in H2hox. The last deprotonation events are assigned to the phenol-OH groups, species H2L and HL- (log K2 = 9.84(1) and log K1 = 10.87(1)) are in good agreement with those of H2hox. 101 Figure 4.6. Representative spectra of the in batch UV spectrophotometric titration of 2.75  10-5 M H2O solution of H2CHXhox, l = 1 cm at 25 °C. 200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0pH 3.52AbsorbanceWavelength (nm) -0.52 -0.46 -0.40 -0.33 -0.29 -0.23 -0.13 -0.04 0.07 0.25 0.43 0.60 0.93 1.59 2.81 2.93 3.52iso = 250 nmH -0.52a)275 300 325 350 375 400 425 4500.000.050.100.150.20b)pH 3.52AbsorbanceWavelength (nm) -0.52 -0.46 -0.40 -0.33 -0.29 -0.23 -0.13 -0.04 0.07 0.25 0.43 0.60 0.93 1.59 2.81 2.93 3.52iso = 343 nmH -0.52200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0c)pH 7.35AbsorbanceWavelength (nm) 3.52 4.07 5.56 6.50 7.35 4.74iso = 257 nmpH 3.52275 300 325 350 375 400 425 4500.000.050.100.150.20d)pH 7.35AbsorbanceWavelength (nm) 3.52 4.07 5.56 6.50 7.35 4.74iso = 296 nmpH 3.52200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0e)pH 9.12AbsorbanceWavelength (nm) 6.50 7.35 8.47 9.12pH 6.50iso = 250 nm275 300 325 350 375 400 425 4500.000.050.100.150.20pH 9.12AbsorbanceWavelength (nm) 6.50 7.35 8.47 9.12pH 6.50iso = 325 nmf)200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0pH 11.80AbsorbanceWavelength (nm) 9.12 9.59 9.91 10.14 10.29 10.39 10.51 10.59 10.66 10.79 10.87 11.14 11.51 11.80iso= 249 nmpH 9.12g)275 300 325 350 375 400 425 4500.000.050.100.150.20h)pH 11.80AbsorbanceWavelength (nm) 9.12 9.59 9.91 10.14 10.29 10.39 10.51 10.59 10.66 10.79 10.87 11.14 11.51 11.80iso = 320 nmpH 9.12102 Table 4.3. Protonation constants of H2CHXhox and H2hox at 25 °C. Equilibrium reaction H2hoxa H2CHXhox L + H+ ⇆ HL 10.88(1) 10.87(1) HL + H+ ⇆ H2L 9.81(1) 9.84(1) H2L + H+ ⇆ H3L 8.39(1) 8.75(1) H3L + H+ ⇆ H4L 6.06(2) 5.95(1) H4L + H+ ⇆ H5L 0.64(6)b 0.15(2)b H5L + H+ ⇆ H6L 0.24(8)b -0.36(3)b a from ref.168; b in-batch-UV spectrophotometric titrations, not evaluated at constant I = 0.16 M NaCl. Charges are omitted for clarity. Figure 4.7. Speciation plots of H2CHXhox (black line) and H2hox (blue dash line) at 2.75  10-5 M ligand concentration and fit for the ligand batch titration at 25 °C. 0 2 4 6 8 10 120255075100 H2hox H2CHX-hox % Formation relative to LigandpH103 4.3.3.2 Complex Formation Equilibria with Cu2+ The Cu(II)-H2CHXhox and Cu(II)-H2hox formation constants were determined by direct UV-vis batch proton competitions, as the complexation reaction was complete when the reagents were mixed at pH ~ 2. For each of the ligands, samples containing the complex and different amounts of standardized HCl or NaOH were prepared and the spectra were measured after an equilibration time of ~ 24 h. The spectra for the Cu(II)-H2hox and Cu(II)-H2CHXhox systems show the same evolution as the pH is raised from very acidic solutions, with the appearance of different isosbestic points indicating the presence of different species in equilibrium (Figures 4.8 and 4.9). Analysis of spectra using the HypSpec201486 program allowed for the determination of the formation constants (Table 4.4). In both systems three different species, [Cu(H2L)]2+, [Cu(HL)]+ and CuL were found. While the first formation constant was not determined at a constant ionic strength (I = 0.16 M), as the ionic strength is affected by the concentration of HCl in the most acidic samples, the other two consecutive formation constants were determined at constant ionic strength (I = 0.16 M) by addition of different amounts of NaCl. In both systems, the formation of bi-protonated species [Cu(H2L)]2+ suggests coordination around the Cu2+ by the four nitrogen atoms, Nen and Nox. This species would lose two protons with pKa values of ~ 2 and ~ 6.8, presumably the phenolic-OH, to form the neutral species CuL in both systems. Comparing the stability constants of the Cu(II)-H2CHXhox and Cu(II)-H2hox systems, it seems that the preorganised ligand H2CHXhox favors Cu(II)-complexation as the complexes form from a lower pH. Also, this increased stability translates into the 1.1 higher unit pCu value which would be important for medicinal applications. As summarized in Table 4.5, compared with H2dedpa, H2CHXhox showed a 10 unit increase in logK and 5.4 unit increase in pCu value. This is a surprising improvement 104 considering the similarity of the two chelators. All these data suggest that H2CHXhox can be useful as a chelating agent in 64Cu radiotracers. Table 4.4. Stepwise stability constants (log K) of H2CHXhox and H2hox complexes with Cu2+, and pM values a. Equilibrium reaction H2CHXhox H2hoxc M2+ + L ⇆ ML 29.78(1) 28.64(1) ML + H+ ⇆ MHL 6.83(1) 6.86 (1) MHL + H+ ⇆ MH2L 2.37(1)b 2.00(1)b pCu 23.9 22.8 a pM is defined as -log [M]free at [L] = 10 M, [M] = 1 M and pH = 7.4; b In-batch acidic spectrophotometric competition at 25 °C, not evaluated at constant I = 0.16 M (NaCl); c from ref.168 Charges are omitted for clarity. Table 4.5. Formation constants of Cu2+ complexes and pM values.187 H2CHXhox H2hox H2dedpa DOTA NOTA H4octapa log KML 29.8 28.6 19.2 22.7 21.2 22.1 pCu 23.9 22.8 18.5 NA NA NA 105 Figure 4.8. Representative spectra of the UV-potentiometric titrations of the Cu(II)-H2CHXhox system, showing the different transformations as the pH is raised a) and b) from -0.47 to 0.53, c) and d) from 0.53 to 1.37, e) and f) from 1.37 to 4.0, g) and h) from 4.73 to 10.78. [H2CHXhox] = [Cu2+] = 2.72  10-5 M, path length = 1 cm, at 25 °C. 200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0H -0.47AbsorbanceWavelength (nm) -0.47 -0.37 -0.28 -0.19 -0.12 -0.03 0.03 0.14 0.29 0.45 0.53pH 0.53iso = 251 nma)275 300 325 350 375 400 425 4500.000.050.100.150.20H -0.47AbsorbanceWavelength (nm) -0.47 -0.37 -0.28 -0.19 -0.12 -0.03 0.03 0.14 0.29 0.45 0.53pH 0.53 = 347 nmb)200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0pH 1.37pH 0.53AbsorbanceWavelength (nm) 1.37 1.16 0.92 0.67 0.53iso = 256 nmc)275 300 325 350 375 400 425 4500.000.050.100.150.20pH 1.37pH 0.53AbsorbanceWavelength (nm) 1.37 1.16 0.92 0.67 0.53d)200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0iso = 230 nmpH 1.37pH 4.0AbsorbanceWavelength (nm) 4.0 3.19 2.8 2.58 2.24 2.03 1.82 1.37iso = 251 nme)275 300 325 350 375 400 425 4500.000.050.100.150.20pH 1.37pH 4.0AbsorbanceWavelength (nm) 4.0 3.19 2.78 2.58 2.24 2.03 1.82 1.37iso = 346 nmf)200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0pH 4.73pH 10.78AbsorbanceWavelength (nm) 10.78 8.90 7.72 7.40 6.99 6.46 5.84 4.73iso = 255 nmg)275 300 325 350 375 400 425 4500.000.050.100.150.20pH 4.73pH 10.78AbsorbanceWavelength (nm) 10.78 8.90 7.72 7.40 6.99 6.46 5.84 4.73iso = 347 nmh)106 Figure 4.9. Representative spectra of the UV-potentiometric titrations of the Cu(II)-H2hox system, showing the different transformations as the pH is raised a) and b) from 0.27 to 1.14, c) and d) from 1.14 to 1.60, e) and f) from 1.72 to 4.5, g) and h) from 5.05 to 11.51. [H2hox] = 2.90  10-5 M; [Cu2+] = 2.76  10-5 M, path length = 1 cm, at 25 °C. 200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0 pH 1.14AbsorbanceWavelength (nm) 0.27 0.63 0.71 1.12 1.14H 0.27a)275 300 325 350 375 400 425 4500.000.050.100.150.20 pH 1.14AbsorbanceWavelength (nm) 0.27 0.63 0.71 1.12 1.14H 0.27b)200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0 pH 1.60AbsorbanceWavelength (nm) 1.60 1.34 1.14pH 1.14c)275 300 325 350 375 400 425 4500.000.050.100.150.20 pH 1.60AbsorbanceWavelength (nm) 1.60 1.34 1.14pH 1.14d)200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0 pH 4.5 pH 1.72AbsorbanceWavelength (nm) 4.5 3.72 3.35 3.31 3.01 2.78 2.59 2.28 2.05 1.86 1.72d)275 300 325 350 375 400 425 4500.000.050.100.150.20 pH 4.5 pH 1.72AbsorbanceWavelength (nm) 4.5 3.72 3.35 3.31 3.01 2.78 2.59 2.28 2.05 1.86 1.72e)200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0 pH 11.51 pH 5.05AbsorbanceWavelength (nm) 11.51 8.76 7.52 6.9 6.65 6.45 6.15 5.05iso = 255 nmg)275 300 325 350 375 400 425 4500.000.050.100.150.20 pH 11.51 pH 5.05AbsorbanceWavelength (nm) 11.51 8.76 7.52 6.9 6.65 6.45 6.15 5.05iso = 347 nmh)107 4.3.3.3 Complex Formation Equilibria with Ga3+ H2CHXhox complexation with Ga3+ metal ions was studied by following the spectral changes in the ligand absorption bands as the pH was raised from 0.19 to 11 by UV in batch spectrophotometry (Figure 4.10). The same studies were performed with H2hox previously168 and an equilibriation time of 24 hours was given to each the sample before measurements were carried out. Except in the experiments with H2CHXhox, the most acidic samples took three weeks to equilibrate until the spectra could be collected (Figure 4.10). The increased rigidity in H2CHXhox might have played an important role in decreasing the acid-assisted dissociation rate of the [Ga(CHXhox)]+ species. The spectra of the Ga(III)-H2CHXhox system collected from very acidic samples Hº -0.44 to pH 0.60 showed identical features to those of the free ligand at the same pH values (FigureS 4.6 a-b, 4.10 a-b). Complex formation starts from pH = 0.6, when a new band appears at λ = 368 nm as well as two isosbestic points at λ = 252 and 340 nm and the shift of the band of the free ligand at λ = 243 nm to 260 nm (Figure 4.10 c-d). Any further transformation occured as the pH was raised from 1.73 to 10.59. Analysis of the spectroscopic data, together with the molar absorption coefficients of the different absorbing species of the free ligand, allows the determination of the stability constant of the [Ga(CHXhox)]+ complex via the use of the HypSpec program86 (Figure 4.11 and Table 4.6). Same as Cu(II)-complex studies, the preorganised ligand H2CHXhox strongly complexes Ga(III) and an increased stability was observed for the [Ga(CHXhox)]+ complex (1.56 units higher than the corresponding value for H2hox). This is a quite encouraging improvement considering that H2hox is already one quite good Ga3+ chelators. 108 Figure 4.10. Representative spectra of the UV-potentiometric titrations of the Ga(III)-H2CHXhox system, showing the different transformations as the pH is raised a) and b) from -0.44 to 0.60, c) and d) from 0.6 to 10.59. [H2CHXhox] = [Ga3+] = 2.72  10-5 M, path length = 1 cm, at 25 °C. Table 4.6. Stepwise stability constants (log K) of H2CHXhox and H2hox complexes with Ga3+, and pM valuesa. Equilibrium reaction H2CHXhox H2hox M3+ + L ⇆ ML 35.91(1)b 34.35(1)c pGa3+ 28.6 28.3 a pM is defined as -log [M]free at [L] = 10 M, [M] = 1 M, pH = 7.4; b In-batch acidic spectrophotometric competition at 25 °C and I = 0.16 M (NaCl); c from ref.168 Charges are omitted for clarity. 200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.02.22.4pH 0.6 -0.44 -0.1 0.03 0.11 0.39 0.49 0.60AbsorbanceWavelength (nm)H -0.44iso = 250 nma)275 300 325 350 375 400 425 4500.000.050.100.150.20pH 0.60H -0.44 -0.44 -0.1 0.03 0.11 0.39 0.49 0.60AbsorbanceWavelength (nm)iso = 344 nmb)200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.02.22.4pH 0.6 10.59 8.94 4.96 1.73 1.61 1.34 1.19 1.04 0.91 0.6AbsorbanceWavelength (nm)pH 10.59 iso = 252 nmc)275 300 325 350 375 400 425 4500.000.050.100.150.20pH 0.6 pH 10.59 10.59 8.94 4.96 1.73 1.61 1.34 1.19 1.04 0.91 0.6AbsorbanceWavelength (nm)iso = 340 nmd)109 Figure 4.11. Fit for the batch titration of the Ga(III)-H2CHXhox system at 25 °C, [H2CHXhox] = [Ga3+] = 2.72  10-5 M. Left: Change of absorbance at 258 nm (blue dots are original data points and the red line represents the fit versus data points. Right: Obtained (blue) and fitted (red) spectra for a pH = 0.91 specific data point. 4.3.3.4 Acid-assisted dissociation kinetic study of Ga(III)-CHXhox The acid-assisted dissociation kinetic study of the [Ga(III)-CHXhox] complex was performed by UV spectrophotometry at 25 °C and 0.1 M HCl due to the high stability of the Ga-hox complex, even at very acidic pH. Standardized concentrated HCl was added to a Ga-CHXhox stock solution ([Ga(CHXhox)][ClO4] = 2.72 × 10-5 M) to achieve pH 1. The reaction was followed by registering the spectra at 15 min intervals at 25 °C. The absorbance of the samples for the in-batch UV spectrophotometric titrations were measured after 3 weeks when the equilibrium had been completely achieved. As in the H2hox study in Chapter 3, the UV spectra clearly show that there are no protonated complex species in the dissociation pathway by the presence of the well-defined isosbestic points : Batch data at 2610.40.81.21.62.0Intensity0102030405060708090100% formation relative to GaObs-Calc intensity (unweighted)0 4 8pH (obs)-0.0500.05value: point 8 pH=0.91000.40.81.2Intensityignored 200 300 400wavelength0value110 previously depicted in the complex formation equilibria studies, and the complex dissociation leads to the ligand in its H4L2+ species. (Figure 4.12) A first-order rate constant was found as shown in Figure 4.13 and the half-life determined in those conditions was 57.8 h. This is around a 50 fold increase compared to the 73 min half-life of Ga(III)-hox measured at the same conditions, confirming that the incorporation of a cyclohexane ring into the backbone not only increases the thermodynamic stability, but also greatly improves the kinetic inertness of the metal-ligand complex. The most acidic in vivo environment is in the stomach with a pH around 1.5-3.5, while most of the other in vivo environments have a pH between 4.5-8.5. Therefore, the long half-life of [Ga(III)-CHXhox] measured at pH = 1 actually suggests an excellent in vivo kinetic inertness. Thermodynamic solution studies and acid-decomplexation studies of [Ga(III)-CHXhox] encourage serum challenge assays with radiolabeled complexes, which is a more relevant indicator of in vivo stability and kinetic inertness. Regrettably, at the time of these studies, Ga isotopes (68Ga or 67Ga) were not available, and concentration- and pH-dependent radiolabeling of the chelator, as well as serum stability assays are still in progress. Figure 4.12. UV spectra of dissociation kinetics of [Ga(CHXhox)]+. 200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0day 6day 5day 4at equilibriumthree weeks afterAbsorbanceWavelength (nm)day 1 = 252 nmAbsorbance275 300 325 350 375 400 425 4500.000.050.100.150.20day 6day 4day 5AbsorbanceWavelength (nm)day 1 = 340 nmAbsorbanceat equilibrium three weeks after111 Figure 4.13. Dissociation kinetics of [Ga(CHXhox)]+ complex. 4.3.4 In vivo Imaging The high thermodynamic stability constant determined from solution studies, in combination with the excellent kinetic inertness found for the [Ga(CHXhox)]+ complex, suggested it may have high stability within an in vivo environment. A dynamic PET/CT imaging study in mice was used to investigate the in vivo stability and biodistribution of the [68Ga(CHXhox)]+ cation. As shown in Figure 4.14, similar with [68Ga(hox)]+, dynamic PET/CT imaging showed quick heart uptake in the first two minutes followed by clearance via both hepatobiliary (liver, then gut) and renal (kidney, then bladder) pathways, which could be explained by the amphiphilic character of the cation. The clearance from the renal pathway is really fast; activity in the kidney showed up in just around 2.5 min post-administration. This is not surprising given the small size of the complexes. The clearance from the hepatobiliary pathway is a little bit slower. The activity in the liver increased in the first 10 min and then started to excrete into the gastrointestinal tract after around y = -0.000208x - 0.028783R² = 0.999842-0.35-0.3-0.25-0.2-0.15-0.1-0.0500 200 400 600 800 1000 1200 1400Ln(Abs(t)-Abs(0))t (min)Dissociation kinetics of [Ga(CHXhox)]+ complex112 20 min. Limited by the short life time of 68Ga, the dynamic PET imaging was recorded for only one hour, which is not long enough to see the full hepatobiliary clearance profile. Therefore, [67Ga(CHXhox)]+ SPECT/CT imaging was taken also to monitor a longer time slot since the liver activity was still high after 1 h in the dynamic PET/CT imaging of [68Ga(CHXhox)]+. As shown in [67Ga(CHXhox)]+ SPECT/CT imaging (Figure 4.15), most of the activity entered into the intestinal tract after 90 min and showed good clearance from the liver. In both of the [68Ga(CHXhox)]+ PET/CT imaging and [67Ga(CHXhox)]+ SPECT/CT imaging, there is no leakage of free 68Ga or 67Ga which would otherwise accumulate significantly in bone.146 No accumulation in muscle, lungs and brain was observed. These results suggest [67/68Ga(CHXhox)]+ has excellent in vivo stability and fast clearance, worth developing further for use as a bifunctional tracer. Figure 4.14. PET/CT dynamic imaging of [68Ga(CHXhox)]+. 113 Figure 4.15 SPECT/CT imaging of [67Ga(CHXhox)]+. The activities found in different organs were then extracted from the PET/CT imaging data and were compared with the data from the [68Ga(hox)]+ PET/CT imaging reported in Chapter 3168. As shown in Figure 4.16, the kidney uptake of [68Ga(hox)]+ is higher than the liver uptake, while in [68Ga(CHXhox)]+, the opposite trend was found with the activity in the liver being 1.5 fold higher than that found in the kidneys. This is not surprising considering the lipophilicity increases from [68Ga(hox)]+ to [68Ga(CHXhox)]+. More importantly the heart uptake and retention also showed an expected increase, agreeing with the previous finding that the lipophilic cations show better heart uptake and retention. [68Ga(CHXhox)]+ may therefore be useful for myocardial perfusion imaging as discussed in the introduction of this Chapter.52, 173, 177-178 Generally, high liver uptake is not a good characteristic for an ideal myocardial perfusion tracer because the liver activity may disturb the interpretation of the heart activity in the inferior and left ventricular wall.149 Therefore, to make a better comparison and evaluation, the heart/liver ratio was calculated as well and compared with H2dedpa derivatives reported.52, 127 As shown in Figure 4.17, [68Ga(CHXhox)]+ showed a higher heart/liver ratio in the first 4 mins than 114 [68Ga(hox)]+, while at the 30 min and 60 min time points, there are no obvious differences between them because the heart and liver uptake increment is almost equal. When compared with [67Ga(dedpa-D8)]+ (log P = 0.66) and [67Ga(dedpa-D9)]+ (log P = 1.10) (Figure 4.17)52, the best two H2dedpa derivatives for myocardial imaging, [68Ga(hox)]+ and [68Ga(CHXhox)]+ showed a great improvement with a 4-6 fold increase in heart/liver ratio, even though the log D value of [68Ga(hox)]+ is still negetive. This may be due to the smaller size of [68Ga(hox)]+ and [68Ga(CHXhox)]+ compared to [67Ga(dedpa-D8)]+ and [67Ga(dedpa-D9)]+, and therefore, it is easier for these smaller structures to undergo transmembrane diffusion into the heart muscle cells or mitochondria if passive diffusion is the major pathway for the cellular uptake of the four complexes. The other hypothesis is that the organic cation transport protein on the cell membrane may have a higher selectivity or affinity for [68Ga(hox)]+ and [68Ga(CHXhox)]+ because of the rigid hydroxyquinoline arms or its 3D molecular surface, assuming the transporter is the major pathway for these four complexes entering into heart cells. Figure 4.16. Biodistribution of [68Ga(hox)]+ and [68Ga(CHXhox)]+ in male mice. -5.00E+000.00E+005.00E+001.00E+011.50E+012.00E+012.50E+013.00E+010 20 40 60% ID/gtime (min)68Ga-hoxbrain - % ID/g heart - % ID/gbone - % ID/g muscle - % ID/gkidney - % ID/g liver - % ID/g0.00E+005.00E+001.00E+011.50E+012.00E+012.50E+013.00E+010 20 40 60ID/gtime (min)68Ga-CHXhoxbrain - % ID/g heart - % ID/gbone - % ID/g muscle - % ID/gkidney - % ID/g liver - % ID/g115 Figure 4.17. Heart/liver uptake ratio of [68Ga(hox)]+, [68Ga(CHXhox)]+, [67Ga(dedpa-D8)]+ and [67Ga(dedpa-D9)]+.52 The analysis from these data indicates that H2hox and H2CHXhox could be a good platform for the development of myocardial perfusion imaging tracers, even though it is really difficult to distinguish between myocardial activity and blood pool activity in the heart in the early time point. But encouraged by this result, two more derivatives H2Br4hox and H2Br4CHXhox were synthesized by adding four bromine atoms to the oxine arms to further increase the lipophilicity. A comprehensive study including solution studies and characterization with the new derivatives is still in progress. This is quite challenging due to the even lower solubility of H2Br4hox and H2Br4CHXhox in aqueous solution compared with H2hox and H2CHXhox. Meanwhile, a preliminary SPECT/CT imaging study with H2Br4hox was carried out and the result is shown in Figure 4.18. The heart uptake was further improved in [67Ga(Br4hox)]+ SPECT/CT imaging, while the in vivo distribution and clearance showed a big difference. There is some brain uptake suggesting it may be able to cross the blood brain barrier. A small amount of [67Ga(Br4hox)]+ was cleared from the kidneys, while most of it was trapped in the liver, with very litter activity excreted into the gastrointestinal track. This can be explained by its high lipophilicity. Some activity was found to remain in the lung after 90 min as well, similar to some previously reported Schiff-base 00.511.522.533.540 1 2 3 4heart/liver ID/g ratio Time (min)Heart/liver uptake ratioHox-ga CHXhox-Gatime (min) 2 30 60 [68Ga(hox)]+ 0.588 0.146 0.124 [68Ga(CHXhox)]+ 0.790 0.143 0.119 [67Ga(dedpa-D8)]+ 0.174 0.034 0.023 [67Ga(dedpa-D9)]+ 0.16 0.11 0.07 116 amine chelators.172 No obvious bone and muscle uptake was found, suggesting the in vivo stability is still high; however, there is a slightly dispersed activity compared with H2hox and H2CHXhox.168 This suggests that [67Ga(Br4hox)]+ may combine with some proteins, for example albumin, in the blood and therefore has a longer circulation time. Another possibility is that the stability may be sacrificed and there is some decomposition and transchelation. There are two hypotheses for possible decreased stability. Firstly, the substitution of two strong electron withdrawal bromine atoms on the aromatic ring decreases the electron donation capacity of the hydroxy group of the oxine arms and thus decreases the bond strength between the hydroxy oxygen and Ga3+ ions. Secondly, the substitution of bromine atom on position 7 of quinoline ring leads to an increased steric hindrance affecting the formation of a stable coordination complex. A more detailed study is in progress to explore a more concrete explanation. Figure 4.18. SPECT/CT scans imaging of [67Ga(Br4hox)]+. 117 4.4 Conclusions H2CHXhox, a cyclohexane reinforced derivative of H2hox, was designed, synthesized, and studied in solution showing that the thermodynamic stability of its complexes with both Ga3+ and Cu2+ was improved compared to corresponding H2hox metal complexes (with logKML values increased by more than 1 unit). More importantly, in acid-assisted Ga3+ complexes dissociation kinetic studies, the dissociation half-life time was increased 50-fold from 73 min to 58 h, proving excellent kinetic inertness. When compared with H2dedpa and H2CHXdedpa, the stability constant and pM value of H2hox and H2CHXhox with Cu2+ exhibited a surprisingly high improvement suggesting their application in 64Cu labelling as well. The dynamic PET imaging studies have shown high in vivo stabilities. [68Ga(CHXhox)]+ is quickly cleared from the mouse via hepatobiliary and renal pathways. Compared to [68Ga(hox)]+, [68Ga(CHXhox)]+ showed increased heart and liver uptake and decreased kidney clearance due to the increased lipophilicity. Both [68Ga(hox)]+ and [68Ga(CHXhox)]+ exhibit a greatly improved heart/liver ratio compared with 67Ga(dedpa-D8) and 67Ga(dedpa-D9). Inspired by this result, two more derivatives H2Br4hox and H2CHXBr4hox were synthesized and the preliminary [67Ga(Br4hox)]+ SPECT/CT imaging showed further enhanced heart uptake, meanwhile, brain and lung retention were observed as well and clearance is much slower. A more detailed study for comprehensive explanation and further improvement is still in progress. 118 Chapter 5: H2C3hox, the Conformation Makes a Big Difference 5.1 Introduction As discussed in previous Chapters, radiometal-based radiopharmaceuticals have drawn increasing interest for their potential in nuclear medicine.16, 20, 35, 192 The development of new chelators is an important field with an increasing number of routinely produced radiometals. In the design of new chelators or modification of old ones, even subtle changes to the chelators including changes to the donor atoms, chelator denticity, structure rigidity, bioconjugation site, cavity and chelate-ring size, can cause drastic changes to their radiolabeling properties and stability of the radiometal complex.16, 20, 35, 54, 192 For example, changing from secondary amine to tertiary amine N atoms in the backbone of H2dedpa caused a marked decrease of stability with Ga3+.49 Therefore, the effects of these changes should be studied comprehensively to match the nature of each specific metal ion. The importance of the coordination atoms and rigidity of the chelators have been discussed in Chapters 3 and 4. Additionally, the cavity and chelate-ring size are crucial factors as well and will be discussed in this Chapter. As shown in Figure 5.1, 5-membered and 6-membered chelate rings, the two most common chelate rings, prefer different metal ionic radii, based on the minimum strain energy modeling of their low strain chair form.193-194 The optimal M-N (Metal-Nitrogen) bond length and N-M-N (Nitrogen-Metal-Nitrogen) bond angle in a 5-membered ring are 2.5 Å and 69 ̊ respectively, while a 6-membered ring prefers a 1.6 Å M-N bond length and 109 ̊ N-M-N bond angle based on the MM (Molecular Mechanics) calculation. Therefore, 5-membered rings favor larger metal ions and higher dentate coordination and 6-membered rings favor smaller metal ions and lower coordination number. 119 Figure 5.1. a) The low strain chair form conformer of cyclohexane, and b) how this relates to minimum strain energy forms of the chelate ring with respect to the M-N bond length and N-M-N bond angle, for 5/6-membered chelate rings.195 Figure 5.2. Plot of change in formation constant, log K from 5-membered chelate ring to 6-membered chelate ring versus metal ions with various ionic radii. 193, 195 Cu2+Ni2+(S=1)Co2+Zn2+Fe2+Cd2+Ca2+Sr2+Pb2+Ba2+Ni2+(S=2)Cu2+Ni2+(S=1)Zn2+Cd2+Pb2+-5-4-3-2-1012340.3 0.8 1.3ΔlogKIONIC RADIUSLogK(TMDTA)-LogK(EDTA)LogK(13-ane-N4)-LogK(12-ane-N4)a b 120 This explanation was used widely in Cu2+-based studies, which found that 6-membered chelate rings work better than 5-membered chelate rings, supported by the comparison of cyclam-based (6-membered chelate rings) and cyclen-based (5-membered chelate rings) chelator studies (Scheme 5.1).195-198 Previous work has summarized the stability of 5- or 6-membered chelate ring complexes of bivalent metal ions with various ionic radii in Figure 5.2.193, 195 This study found that generally metal ions with an ionic radius larger than Cu2+ (e.g. Zn2+, Cd2+ and Ca2+) showed higher stability with a 5-membered ring (cyclen and EDTA) over a 6-membered ring (cyclam and 1,3-PDTA), while smaller metal ions (e.g. Ni2+) have an opposite predilection. A previous study in our group54 investigated differences between 5- and 6- membered chelate rings in the picolinate series of ligands, with larger metal ions including In3+ and Lu3+. This study confirmed that the preference change from 5-membered rings in H4octapa (Scheme 5.1) to 6-membered rings H4C3octapa by adding one more carbon decreases the stability of corresponding metal complexes by around 102. Whether this trend could apply to Ga3+ which has a smaller ionic radius than Cu2+ was not really clear. Scheme 5.1. Examples of ligands with 5/6 membered chelate rings. In this Chapter, H2C3hox was designed and synthesized by adding one more carbon to the diamine backbone of H2hox (Scheme 5.2). The new chelator formed a 6-membered ring between the backbone diamine and the central metal ions, and stability of its metal complex has been explored in solution studies and compared with H2hox. The in vivo stability of the Ga complex was also studied by SPECT/CT imaging. 121 Scheme 5.2. Design of H2C3hox. 5.2 Experimental 5.2.1 Materials and Methods All solvents and reagents were purchased from commercial sources (TCI America, Sigma Aldrich, Fisher Scientific) and used as received unless otherwise indicated. The analytical thin-layer chromatography (TLC) plates used were aluminum-backed ultrapure silica gel 60 Å, 250 μm thickness. 1 H and 13C NMR spectra were recorded at ambient temperature on Bruker Avance 300 and Avance 400 spectrometers; the 1H NMR spectra were calibrated against residual protio-solvent peak and the 13C NMR spectra were referenced to the deuterated solvent. Low-resolution mass spectrometry was performed on a Waters ZG spectrometer with an ESCI (electrospray/chemical-ionization) source, and high-resolution electrospray ionization mass spectrometry (ESI-MS) was performed on a Micromass LCT time-of-flight (TOF) instrument. Microanalysis for C, H, and N was performed on a Carlo Erba Elemental Analyzer EA 1108. Purification and quality control of [67Ga(C3hox)]+ for SPECT/CT imaging were performed on an Agilent HPLC system equipped with a model 1200 quaternary pump, a model 1200 UV absorbance detector, and a Bioscan (Washington, DC) NaI scintillation detector. The radio‐detector was connected to a Bioscan B‐FC‐1000 Flow‐count system, and the output from the Bioscan Flow‐count system was fed into an Agilent 35900E Interface which converted the analog signal to digital signal. The operation of the Agilent HPLC system was controlled using the Agilent ChemStation software. The HPLC columns used were a semi‐preparative column (Phenomenex C18, 5 µ, 250 × 10 mm) and an analytical 122 column (Phenomenex C18, 5 µ, 250 × 4.6 mm). The HPLC solvents were A: H2O containing 0.1% TFA, and B: CH3CN containing 0.1% TFA. Radioactivity of [67Ga(C3hox)]+ was measured using a Capintec (Ramsey, NJ) CRC® ‐25R/W dose calibrator. SPECT imaging experiments were conducted using a Siemens (Erlangen, Germany) Inveon microPET/CT scanner. Human serum and Sephadex G-25 PD10 Desalting columns were purchased from Sigma-Aldrich or GE Healthcare. 5.2.2 Ligand Synthesis and Characterization H2C3hox (5.1) 8-Hydroxyquinoline-2-carboxaldehyde (2.0 g, 11.6 mmol) was dissolved in 50 mL ethanol, and 1,3-diaminopropane (488 µL, 5.8 mmol) dissolved in 5 mL of ethanol was added dropwise; the reaction mixture was stirred at 60 °C for 4 h. Five equivalents of NaBH4 were then added in portions and the reaction mixture was stirred at RT overnight. HCl (20 mL, 6 M) was added and the reaction mixture stirred for 4 h. The pH of the reaction mixture was then readjusted to 7 using NaOH (2 M), and the off-white precipitate was filtered out and dried as crude product. The crude product was further washed with water and methanol to obtain the pure product (1.59 g, 4.1 mmol), yield = 71%. 1H NMR (300 MHz, MeOD, 25 °C) δ 8.19 (d, J = 8.5 Hz, 1H), 7.53 – 7.26 (m, 3H), 7.10 (dd, J = 7.4, 1.4 Hz, 1H), 4.14 (s, 2H), 2.86 (t, J = 7.1 Hz, 2H), 1.94 (p, J = 7.2 Hz, 1H). 13C NMR (75 MHz, MeOD, 25 °C) δ 157.0, 153.7, 138.4, 136.5, 128.2, 127.0, 120.8, 117.1, 111.0, 54.0, 28.6, 25.6. LR-ESI-MS calcd. for [C23H24N4O2 + H]+:389.2; found 389.2 123 5.2.3 Synthesis of H2C3hox-Ga Complex [Ga(C3hox)][ClO4](5.2) H2C3hox (40 mg, 0.11 mmol) was dissolved in acetonitrile. Ga(ClO4)3·6H2O (55 mg, 0.11 mmol) was added, and the pH was adjusted to ~ 5 using 0.1 M NaOH. The reaction mixture was stirred for 1 h at 50 °C; CH2Cl2 (5 mL × 3) was added to extract the product and yellow needle crystals suitable for X-ray diffraction formed from layer diffusion of diethyl ether into the CH2Cl2 extraction. 1H NMR (300 MHz, MeOD, 25 °C) δ 8.82 (s, 1H), 7.87 (d, J = 8.5 Hz, 2H), 7.60 (d, J = 8.0 Hz, 2H), 7.39 (dd, J = 8.4, 0.9 Hz, 2H), 6.98 (dd, J = 7.8, 0.9 Hz, 2H), 4.76 (d, J = 18.2 Hz, 2H), 4.43 (d, J = 18.2 Hz, 2H), 3.02 (s, 1H), 2.93 (s, 1H), 1.96 – 1.78 (m, 2H). 13C NMR (75 MHz, MeOD, 25 °C) δ 155.8, 154.6, 150.9, 149.8, 142.8, 131.2, 128.5, 119.9, 112.7, 53.5, 49.3, 20.5. LR-ESI-MS calcd. for [C23H2269GaN4O2]+:455.2; found 455.2 5.2.4 X-ray Crystallography A suitable crystal 0.07×0.05×0.01 mm3 was selected and mounted on a suitable support for a Bruker APEX-II CCD diffractometer. The crystal was kept at a steady T = 96(2) K during data collection. The structure was solved with the ShelXT185 structure solution program using the Intrinsic Phasing solution method and by using Olex273 as the graphical interface. The model was refined with version 2018/1 of ShelXL185 using Least Squares minimisation. Data were measured using f and w scans using MoKa radiation. The maximum resolution that was achieved was Q = 29.865° (0.71 Å). The diffraction pattern was indexed, and the unit cell was refined using SAINT69 on 4177 reflections, 18% of the observed reflections. Data reduction, 124 scaling and absorption corrections were performed using SAINT.69 The final completeness is 100.00% out to 29.865° in Q. A multi-scan absorption correction was performed using SADABS-2016/2180 was used for absorption correction. wR2(int) was 0.0670 before and 0.0410 after correction. The ratio of minimum to maximum transmission is 0.8785. The l/2 correction factor is not present. The absorption coefficient m of this material is 1.435 mm-1 at this wavelength (λ = 0.711Å) and the minimum and maximum transmissions are 0.871 and 0.992. 5.2.5 Solution Thermodynamics Protonation and complex formation equilibria were studied by UV spectrophotometric batch experiments as previously described168 using a Cary 60 UV-Vis spectrophotometer in the spectral range 200-450 nm, at 25 °C and 1 cm path length (l). A set of 60 solutions containing the ligand (H2C3hox, 3.16  10-5 M) in pure water were prepared and different amounts of standardized HCl or NaOH were added to cover a range from H0 -0.64 to pH 11.75. The ionic strength of each sample was adjusted (when possible) to 0.16 M by addition of different amounts of NaCl. In the most acidic samples (below pH 0.6), it was not possible to maintain constant ionic strength since that depends on the HCl content and for those solutions, the correct acidity scale H0 was used.84 In the samples below pH 2, the equilibrium H+ concentration was calculated from solution stoichiometry and for the rest of the samples, pH was measured with a Ross combination glass electrode that was calibrated daily for hydrogen ion concentrations using HCl as previously described80 and the results were analyzed by the Gran81 procedure. For the complex formation equilibria, the set of solutions were prepared in the same way as described above. For the GaIII-H2C3hox system, the set of samples were prepared by adding the atomic absorption (AA) standard metal ion solution into a H2C3hox solution of known 125 concentration in the metal to ligand molar ratios 1:1. The exact amount of acid present in the gallium(III) standard was determined by Gran’s method81 titrating equimolar solutions of Ga(III) and Na2H2-EDTA. Equilibration time of 2 min for the ligand protonation equilibria study was allowed before measuring the pH and the UV absorption spectrum. For the complex formation equilibria with Ga(III), the samples were prepared and measured after 24 h when equilibrium was reached. The spectral data were analyzed using the HypSpec201486 program. Proton dissociation constants corresponding to hydrolysis of Ga(III) aqueous ions included in the calculations were taken from Baes and Mesmer.87 The species formed in the studied systems are characterized by the general equilibrium: pM + qH + rL = MpHqLr (charges omitted). For convention, a complex containing a metal ion, M, proton, H and ligand, L, has the general formula MpHqLr. The stoichiometric indices p, might also be 0 in the case of protonation equilibria and negative values of q refers to proton removal or hydroxide ion addition during formation of the complex. The overall equilibrium constant for the formation of the complexes MpHqLr from its components is designated as log β. Stepwise equilibrium constants log K correspond to the difference in log units between the overall constants of sequentially protonated (or hydroxide) species. A more straightforward comparison of the ability of different ligands to coordinate a specific metal ion than the thermodynamic stability constants alone, is the use of the pM values; pM is defined as (-log[Mn+]free) and is calculated at specific conditions ([Mn+] = 1 µM, [Lx-] = 10 µM, pH 7.4 and 25 °C, taking into consideration metal-ligand association and ligand basicity.92 126 5.2.6 SPECT/CT Imaging Studies Animal imaging studies were performed in accordance with the Canadian Council on Animal Care (CCAC) and protocol approved by the Animal Care Committee (ACC) of the University of British Columbia (A16-0150). Three healthy C57Bl/6 female mice (~ 16 g) were anaesthetized using isoflurane on a precision vaporizer (5% in oxygen for induction, between 1.5 and 2.5% in oxygen for maintenance) and received a subcutaneous injection of lactated Ringer’s solution (0.5 mL) for hydration prior to each imaging scan. Immediately after injection of [67Ga(C3hox)]+, dynamic whole-body images were acquired during 90 min using a multimodal SPECT/CT scanner (VECTor/CT, MILabs, The Netherlands) equipped with a XUHS-2 mm mouse pinhole collimator. Throughout the entire scanning procedure, the mouse was kept under isoflurane anesthesia and constant body temperature was maintained using a heating pad. Following each SPECT acquisition, a whole-body CT scan was acquired to obtain anatomical information and both images were registered. For quantitative analysis, SPECT image reconstructions were carried out with a pixel-ordered subset expectation maximization (POSEM) algorithm that included resolution recovery and compensation for distance-dependent pinhole sensitivity. For the SPECT images, 16 subsets, 10 iterations and an isotropic 0.4 mm voxel grid were used. The images were decay-corrected and after CT registration, attenuation correction was applied. For visual representation, the reconstructed volumes of SPECT scans were post-filtered with a 3D Gaussian filter. CT scans were acquired with a tube setting of 55 kV and 615 µA. In total 2 frames of 180 projections over 360 degrees were acquired in step and shoot rotation mode. The acquired projection data was reconstructed using SkyScan NRecon software to generate a 3D CT image on 0.169 mm3 voxel size. Volumes of interest (VOIs) were manually defined using AMIDE (v.1.0.5) to determine the 127 time activity pattern per target organ. Thus, the delineated regions were liver, bladder, bone and gallbladder. 5.3 Results and Discussion Scheme 5.3. Synthesis of H2C3hox. 5.3.1 Synthesis and Characterization of Ligand Generally, as shown in Scheme 5.3, H2C3hox was prepared by sodium borohydride reductive amination, similar to the synthesis of H2hox and H2CHXhox. The difference is that in the preparation of H2hox and H2CHXhox, a crystalline Schiff base intermediate precipitated out and can be separated and collected before the addition of NaBH4. However, in the synthesis of H2C3hox, the Schiff base intermediate has a better solubility in the solvent and thus the reaction progressed by direct addition of the reductant NaBH4.. Similar with H2hox and H2CHXhox, the product’s solubility in water is quite low and thus could be purified by washing with a mixture of H2O and methanol without further column separation. The product was characterized using 1H, 13C NMR and MS spectra, and as shown in Figure 5.3 (upper), the 1H spectra of H2C3hox revealed a C2 symmetry same as the H2hox and H2CHXhox reported before.168 128 5.3.2 Preparation and Characterization of the Metal Complex H2C3hox and Ga(ClO4)3 was mixed in 1:1 molar ratio in methanol and the pH was adjusted to neutral using NaOH (0.1M), [Ga(C3hox)][ClO4] was then extracted by DCM. The formation of fully coordinated metal complex was confirmed by 1H NMR spectra. As shown in Figure 5.3, the homotopic proton (A and B) on the free ligand (upper) to diastereotopic protons (A1, A2 and B1, B2) in the complex (lower), and this indicates the binding of the 4 nitrogens and the formation of chelate rings. In addition, the aromatic protons on the hydroxyquinoline ring are shifted downfield in the metal complex due to the decreased electron density after the complexation, confirming the binding of the two arms to Ga3+. All these change are generally the same as found with H2hox and H2CHXhox. Interestingly, the protons B shifted downfield for around 0.2 unit. This is quite different from the protons at the same position in H2hox and H2CHXhox, which are shifted upfield. That different change upon complexation indicates that the orientation of protons B in the diamine backbone and its interaction with coordination bond in 6-membered ring is quite different from 5-membered ring. 129 Figure 5.3. 1H NMR spectra of H2C3hox in MeOD (300 MHz, 25 °C) (upper) and [Ga(C3hox)][ClO4] in MeOD (300 MHz, 25 °C) (lower). 130 5.3.3 X-ray Crystallography Figure 5.4. ORTEP-style picture of [Ga(C3hox)][ClO4] (C23H22ClGaN4O6). Table 5.1. Selected bond distances and bond angles in the cations [Ga(hox)]+168 and [Ga(C3hox)]+. [Ga(hox)][ClO4] [Ga(C3hox)][ClO4] Atom Atom Length (Å) Atom Atom Length (Å) Ga O1(ox) 1.959(2) Ga O1(ox) 1.9651(17) Ga O2(ox) 1.952(2) Ga O2(ox) 1.9618(17) Ga N1(ox) 1.982(2) Ga N1(ox) 1.987(2) Ga N3(ox) 1.981(2) Ga N4(ox) 1.981(2) Ga N2(en) 2.190(3) Ga N2(pn) 2.174(2) Ga N4(en) 2.193(3) Ga N3(pn) 2.154(2) Angle degree [deg] Angle degree [deg] N(en)-Ga-N(en) 81.0 N(pn)-Ga-N(pn) 89.5 N(en)-Ga-N(ox) 77.3 N(pn)-Ga-N(ox) 77.1 N(ox)-Ga-O(ox) 82.0 N(ox)-Ga-O(ox) 82.0 ox = oxine, pn = propylenediamine, and en = ethylenediamine 131 Yellow needle shaped [Ga(C3hox)][ClO4] crystals were obtained by layer diffusion of diethyl ether into a dichloromethane solution of the complex. The solid state structure was studied by X-ray diffraction and presented in Figure 5.4. The crystals are monoclinic and there is a single molecule in the asymmetric unit, which is represented by the reported sum formula C23H22ClGaN4O6. As shown in Figure 5.4, the Ga3+ ion is bound to all 6 coordinating atoms in the ligand, agreeing with the NMR analysis. Similar to [Ga(hox)][ClO4], the complex structure is distorted octahedral with a slight elongation of the two Ga-N(pn) bonds. The C2 symmetry of the chelator was kept with Ga3+ residing on a two-fold rotation axis. Selected bond length and angle parameters [Ga(C3hox)][ClO4] are summarized compared with previously reported [Ga(hox)][ClO4] 168 in Table 5.1. The meridional arrangement of the six-coordinating atoms as well as all the coordination bond lengths in [Ga(C3hox)][ClO4] are almost the same to those in [Ga(hox)][ClO4] with the Ga-O bonds slightly longer and Ga-N(pn) bonds slighter shorter than Ga-N(en) in [Ga(C3hox)][ClO4]. When comparing the bond angles, most of the bond angles are mostly the same, while the major difference is in the backbone chelate ring. The N(pn)-Ga-N(pn) angle is 89.5° in [Ga(C3hox)][ClO4] and it is larger than the 80.8° N(en)-Ga(1)-N(en) angle in [Ga(hox)]+ as shown in Table 5.1 and Figure 5.5. This agreed with what we discussed in the introduction that a 6-membered ring favors larger a N-M-N angle. The larger angle results in a bit more even distribution of the coordination bonds and slightly less distorted octahedral geometry as illustrated in Figure 5.5. To summarize, the crystal structure suggests a similar fit of [C3hox]2- and [hox]2- with the Ga3+ cation, and a possibly similar stability, suggesting solution studies would be useful. 132 Figure 5.5. Side-on view of N(pn)-Ga-N(pn) bond angle in [Ga(C3hox)][ClO4] (left) and N(en)-Ga-N(en) angle (right) in [Ga(hox)][ClO4]. 5.3.4 Solution Studies 5.3.4.1 Protonation Constants of H2C3hox The acidity of ligand donor groups can affect metal coordination and radiolabeling properties, and to further explore this, protonation constants of H2C3hox were determined by UV spectrophotometric titrations following the same method as described for H2hox168 and for H2CHXhox (Chapter 4). H2C3hox presents the same spectral evolutions as H2hox168 and H2CHXhox (Figure 5.7). Spectrophotometric data were analysed with the HypSpec2014 software86 and allowed the determination of the six protonation states and generation of the corresponding speciation plots (Table 5.2 and Figure 5.9). Although these ligands present the same denticity and coordinating groups, the lower pKa for functional groups can allow for metal coordination at lower pH values. The largest variances in pKa values of H2C3hox versus H2hox reside on the Nox atoms of the quinoline units and on the Nen atoms in the backbone. The two Npn atoms in the backbone are chemically identical but deprotonate with different pKa values as a result of the charge repulsion of the H4L2+ species when the two Npn atoms are protonated. Despite H2C3hox only 89.5 ° 81.0 ° 133 differing with H2hox in a single backbone carbon, its H4L2+ species, having a three-carbon propylene bridge instead of a two-carbon ethylene bridge, deprotonates at a higher pH because of the lower charge repulsion (6.88(1) vs 6.06(2)). Table 5.2. Protonation constants of H2C3hox and H2hox at 25 °C. Equilibrium reaction H2hoxa H2C3hox L + H+ ⇆ HL 10.88(1) 10.74(1) HL + H+ ⇆ H2L 9.81(1) 9.88(1) H2L + H+ ⇆ H3L 8.39(1) 9.00(1) H3L + H+ ⇆ H4L 6.06(2) 6.88(1) H4L + H+ ⇆ H5L 0.64(6)b 0.75(2)b H5L + H+ ⇆ H6L 0.24(8)b 0.60(3)b a from ref.168; b in-batch UV spectrophotometric titrations, not evaluated at constant I = 0.16 M NaCl. Charges are omitted for clarity. Figure 5.6. pH-dependent electronic absorption spectra (in molar absorptivity) of the seven light absorbing species of H2C3hox. 200 225 250 275 300020000400006000080000 L2- HL- H2L H3L+ H4L2+ H5L3+ H6L4+Molar absorptivity (M-1cm-1)Wavelength (nm)a)275 300 325 350 375 400 425 4500100020003000400050006000 L2- HL- H2L H3L+ H4L2+ H5L3+ H6L4+Molar absorptivity (M-1cm-1)Wavelength (nm)b)134 Figure 5.7. Representative spectra of the in-batch UV spectrophotometric titrations of 3.16  10-5 M H2O solution of H2C3hox, l = 1 cm at 25 °C. 200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.02.22.4pH 5.93 -0.12 0.24 0.37 0.52 0.66 0.81 0.92 1.10 1.57 1.72 2.47 3.22 5.93AbsorbanceWavelength (nm)Ho -0.12iso = 249 nma)275 300 325 350 375 400 425 4500.000.050.100.150.20b)pH 5.93 -0.12 0.24 0.37 0.52 0.66 0.81 0.92 1.10 1.57 1.72 2.47 5.93AbsorbanceWavelength (nm)Ho -0.12iso = 343 nm200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.02.22.4pH 5.93pH 8.12 5.93 6.92 7.02 8.12AbsorbanceWavelength (nm)c)275 300 325 350 375 400 425 4500.000.050.100.150.20pH 5.93pH 8.12 5.93 6.92 7.02 8.12AbsorbanceWavelength (nm)d)200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.02.22.4pH 8.12pH 9.28 9.28 9.14 8.93 8.12AbsorbanceWavelength (nm)e)275 300 325 350 375 400 425 4500.000.050.100.150.20pH 8.12pH 9.28 9.28 9.14 8.93 8.12AbsorbanceWavelength (nm)f)200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.02.22.4pH 9.28pH 11.83 11.83 11.69 11.02 10.59 10.30 10.16 10.07 9.94 9.52 9.33 9.28AbsorbanceWavelength (nm)g)iso = 249 nm275 300 325 350 375 400 425 4500.000.050.100.150.20pH 9.28pH 11.83 11.83 11.69 11.02 10.59 10.30 10.16 10.07 9.94 9.52 9.33 9.28AbsorbanceWavelength (nm)h)iso = 320 nm135 5.3.4.2 Thermodynamic Stability The thermodynamic stability of the metal complex formed by H2C3hox and Ga(III) was determined by acidic competitive UV-spectrophotometric titrations, under the same experimental conditions as those previously used for H2hox.168 The thermodynamic stability constant of the [Ga(C3hox)]+ species was experimentally determined to be 29.85(2). A gallium(III) hydroxocomplex was observed for the Ga(III)-H2C3hox system (log KM(OH)L = 9.38(2)) (Table 5.3 and Figure 5.9). In this hydroxospecies, the hydroxide anion probably replaces one of the coordinated hydroxyquinolinate arms of H2C3hox. The different behavior with respect to H2hox and H2CHXhox, can be ascribed to the lower thermodynamic stability of the ML species with H2C3hox (4.5 and 6.06 units lower respectively), resulting from the additional carbon in the backbone. Table 5.3. Stability constants (log K) of H2C3hox, H2CHXhox and H2hox complexes with Ga3+. Equilibrium reaction H2C3hox H2CHXhox H2hox M3+ + L ⇆ ML 29.85(2) 35.91(1)b 34.35(1)c M(OH)L + H+ ⇆ ML + H2O 9.38(2) - - pMa 23.2 28.6 28.3 a pM is defined as -log [M]free at [L] = 10 M, [M] = 1 M and pH = 7.4; b In-batch acidic spectrophotometric competition at 25 °C and I = 0.16 M (NaCl); c from ref.168 Charges are omitted for clarity. 136 Figure 5.8. Representative spectra of the UV-potentiometric titration of the Ga(III)-H2C3hox system, showing the different transformations as the pH is raised a) and b) from 2.46 to 8.15, c) and d) from 8.15 to 10.91. [H2C3hox] = [Ga3+] = 3.12  10-5 M, path length = 1 cm, at 25 °C. Figure 5.9. Speciation diagrams for the Ga(III)-C3hox system calculated from values given in Tables 5.2 (a) and 5.3 (b). [Ga3+] = [H2C3hox] = 3.16  10-5 M, at 25 °C. 200 225 250 275 3000.000.250.500.751.001.251.501.752.00pH 2.46 8.15 3.79 3.35 3.22 3.20 3.07 2.98 2.92 2.82 2.72 2.71 2.63 2.56 2.46AbsorbanceWavelength (nm)pH 8.15iso = 250 nma)275 300 325 350 375 400 425 4500.000.050.100.150.20pH 2.46 8.15 3.79 3.22 3.20 3.07 2.98 2.92 2.82 2.72 2.71 2.63 2.56 2.46AbsorbanceWavelength (nm)pH 8.15iso = 250 nmb)200 225 250 275 3000.00.20.40.60.81.01.21.41.61.82.0iso = 272 nm pH 10.91 8.15 9.46 10.32 10.44 10.63 10.91AbsorbanceWavelength (nm)pH 8.15iso = 256 nm c)275 300 325 350 375 400 425 4500.000.050.100.150.20pH 10.91 8.15 9.46 10.32 10.44 10.63 10.91AbsorbanceWavelength (nm)pH 8.15iso = 368 nm d)0 2 4 6 8 10 120255075100 H2hox H2C3hox % Formation relative to ligandpHa)2 4 6 8 10 120255075100Ga3+[Ga(OH)2]+Ga(OH)3[Ga(OH)4]-Ga(OH)(C3hox)% Formation relative to Ga3+pHb)[Ga(C3hox)]+137 As reported above (Table 5.3), H2C3hox showed a drastically decreased thermodynamic stability and pM value with Ga3+ compared to H2hox and H2CHXhox. This is a huge change considering the similarity between H2C3hox and H2hox, and greatly differed from our expectation of metal ions smaller than Cu2+ preferring a 6-membered chelate ring than a 5-membered chelate ring following the previously reported trend. To explain this phenomenon, we compared the crystal structure of [Ga(hox)]+ and [Ga(C3hox)]+ carefully, and found that the conformation of the chelate ring actually makes a big difference. As shown in Figure 5.10, in [Ga(hox)]+, the 5-membered chelate ring formed by the two backbone nitrogen atoms and Ga3+ took a half-chair conformation which is the minimum energy ring form; however, in [Ga(C3hox)]+, a twist boat conformation was found which is not the lowest energy conformation for 6-membered ring and introduces more torsion strain (Figure 5.9). Additionally, although bond lengths of Ga-N(pn) in [Ga(C3hox)]+ are slightly shorter than Ga-N(en) in [Ga(hox)]+, they are still around 2.16 Å and closer to the 2.5 Å favored by a 5-membered ring rather than 1.6 Å favored by a 6-membered ring. Also when comparing the bond angle, even though the 89.5° N(pn)-Ga-N(pn) angle is larger, it is still around 20° off from the perfect 109° angle in 6-membered ring, while the 81.0° N(en)-Ga-N(en) angle in [Ga(hox)]+ has only around 12° difference from the 69° angle favored in 5-membered ring. All the analysis together explains why Ga3+ is more stable with 5-membered rather than 6-membered rings. This case also proves that, the conformation of the formed chelate ring is also an important factor and must be carefully considered in addition to the ring size when designing a chelator. Actually, the conformation of the complexes can be predicted using MM (Molecular Mechanics) or DFT (Density Functional Theory) simulation, and those chelators form energy unfavorable rings can be excluded from the design. 138 half-chair twist boat Figure 5.10. Conformations of 5 or 6 membered chelate ring formed by backbone nitrogen atoms and Ga3+ in [Ga(hox)][ClO4] (left) and [Ga(C3hox)][ClO4] (right). Another lesson learned from this study is that perhaps, Cu2+ is not a good reference for Ga3+, since it has two stable configurations, square planar and octahedral. The previously reported trend was summarized from bivalent metal ions, thus future studies should explore ionic radii and ring size selectivity of trivalent metal ions.195 Although H2C3hox is not as good a ligand as H2hox and H2CHXhox, its thermodynamic stability is still comparable to other ligands like TRAP and NOTA (Table 5.4) (structures shown in Figure 3.1). From the view of synthetic accessibility, the added carbon provides an easy access bifunctionalization site, compared to the bifunctionalization of H2hox, where a chiral centre will be introduced, making the synthetic route and characterization much more difficult (Scheme 5.4). Thus, H2C3hox will still be a useful chelator to explore, especially with an incorporated bifunctional handle. 139 Table 5.4. Formation constants of Ga3+ complexes and pMk values. log K log KML pGa H2C3hox a 29.85(2) 23.2 H2dedpa b 28.11(8) 27.4 Oxine (KP46) c 13.13(8) *36.41(1) 21 NOTA d 30.98 27.9 DOTA e,f 21.33 e; 26.05 f 18.5; 19.50 TRAP g 26.24 23.1 HBED h,i 37.73 g; 39.57 h 27.7 g; 29.4 h DFO j 28.65 21.2 a This work (0.16 M NaCl at 25 °C). b From ref.25 (0.16 M NaCl at 25 °C). c From ref.128 (*) log K Ga(Lox)3 (0.20 M KCl at 25 °C). d From ref.137 (0.10 M KCl at 25 °C). e From ref.138 (0.1 M KCl at 25 °C). f From ref.101 (0.1 M (NMe4)Cl at 25 °C). g From ref.22. h From ref.116. i From ref.139. j From ref.140. k Calculated at specific conditions ([Ga3+] = 1 µM, [Lx-] = 10 µM, pH 7.4 and 25 °C). Scheme 5.4. Potential bifunctional derivatives of H2hox and H2C3hox. 5.3.5 SPECT/CT Imaging Studies. A 67Ga SPECT/CT imaging study in mice was used to investigate the in vivo stability and biodistribution of the [67Ga(C3hox)]+ cation. As shown in Figure 5.11, the biodistribution of [67Ga(C3hox)]+ is quite similar to that of [68Ga(hox)]+ which is quite reasonable considering the 140 similarity of the two structures. There is no leakage of free 67Ga during the imaging, which otherwise would accumulate in bone. No accumulation or retention was found in muscle and lungs as well. This confirmed that [67Ga(C3hox)]+ has a good in vivo stability similar to [68Ga(hox)]+. No activity was found in brain, suggesting that [67Ga(C3hox)]+ cannot pass the blood-brain barrier (BBB). The complex was cleared via both hepatobiliary (liver, then gut) and renal (kidney, then bladder) pathways which can be explained by its amphiphilic nature. The clearance is fast like most small molecules, and after 90 min, most of the activity was in the intestinal tract and showed good clearance from the liver. The imaging studies confirmed our prediction that even though H2C3hox does not have a very high stability constant and pM value with Ga3+, as do H2hox and H2CHXhox in solution studis, its in vivo stability is still good enough for application as an imaging tracer. The good in vivo stability and clearance encouraged the preparation of bifunctional derivatives, which has been designed and is under investigation now. Figure 5.11. SPECT/CT imaging of [67Ga(C3hox)]+. 141 5.4 Conclusions In summary, H2C3hox was designed and synthesized by adding one more carbon to the ethylenediamine backbone of H2hox, to investigate the prediction of the previously reported trend that smaller metal ions will prefer 6-membered chelate rings over 5-membered rings. However, solution studies showed that the thermodynamic stability of its complexes with Ga3+ was actually lower than H2hox. A comparison of the two crystal structures of the respective Ga3+ complexes revealed that the 5-membered chelate ring formed by the two backbone nitrogen atoms and Ga3+ in [Ga(hox)]+ took a half-chair conformation which is the minimum energy ring form. In [Ga(C3hox)]+, a twist boat conformation was found, which is not the lowest energy conformation for a 6-membered ring and introduces more torsion strain leading to a large ( 5 orders of magnitude) decrease of stability constant. This study gives a good perspective to the discussion of 5 or 6-membered chelate ring preferences with metal ionic radius, and highlights the importance of careful ligand design and radiometal-ligand matching for radiopharmaceutical applications. Even though H2C3hox is not as strong as H2hox and H2CHXhox in term of thermodynamic stability, it is still a good chelator for Ga3+, comparable to other ligands like TRAP and NOTA and this was confirmed by the result of in vivo SPECT/CT imaging. Thus, a biofunctional derivative was designed and is being explored now. 142 Chapter 6: H4octox: Versatile Bimodal Octadentate Acyclic Chelating Ligand for Medicinal Inorganic Chemistry This chapter is an adaptation of published work, and is reproduced in part from Xiaozhu Wang, María de Guadalupe Jaraquemada-Peláez, Cristina Rodríguez-Rodríguez, Yang Cao, Christian Buchwalder, Neha Choudhary, Una Jermilova, Caterina F. Ramogida, Katayoun Saatchi, Urs O. Häfeli, Brian O. Patrick, Chris Orvig. H4octox: Versatile Bimodal Octadentate Acyclic Chelating Ligand for Medicinal Inorganic Chemistry, J. Am. Chem. Soc., 2018, 140, 15487-15500. 6.1 Introduction 111In is an important cyclotron-produced isotope in nuclear medicine. The long half-life of 111In (2.8 days) matches biovectors with a long biological half-life, such as antibodies.20 It decays by electron capture with emission of γ rays (245 and 172 keV) that can be used for single photon emission computed tomography (SPECT) imaging and Auger electrons which can be used for therapy.199 110mIn (t1/2 = 69 min) has also attracted great interest because of its potential application in positron emission tomography (PET) imaging.200 177Lu (t1/2 = 6.6 days, 134.17 keV) and 90Y (t1/2 = 2.67 days, 933.61 keV) are two other important β− emission radioisotopes used in therapy. 90Y only emits β− particles, and is therefore usually used in combination with 111In as a congener for imaging and dosimetry, based on the same charge and similar size of their metal ions.199, 201-203 Gd3+ is the most commonly used metal ion for MRI imaging. Gadolinium-based MRI contrast agents can sometimes be repurposed with other metal radioisotopes of similar chemical properties.204-205 The application of all these metal isotopes mentioned above and their combined or interchanged use requires a good chelator that can bind all these metal ions with high 143 thermodynamic stability and kinetic inertness. DTPA (diethylenetriamine tetraacetic acid) and DOTA (1,4,7,10-tetraazacyclododecane-1,4,7,10-tetraacetic acid) are the most commonly used chelators in recent years.20 DTPA is an acyclic chelator which can radiolabel these metal ions fast (within 15 min) at mild conditions (ambient temperature) and is used in currently available FDA approved 111In-based radiopharmaceuticals; however, it shows inferior in vivo stability compared to the macrocyclic chelator DOTA. DOTA, on the other hand, often requires elevated temperatures (60-95 °C) and extended reaction times (30-120 min) to achieve quantitative radiolabeling yields and is not favored for the use of thermally-sensitive carriers like antibodies.24, 206 Significant effort has been made in the past decade to overcome these limitations. A summary of the most representative chelators is presented in Scheme 6.1.24, 165, 206-210 For macrocyclic chelators, one major strategy to accelerate the complexation kinetics is to create a hybrid chelator by introducing an acyclic moiety, for example, both AAZTA and NETA added one iminodiacetic acid group to the macrocyclic backbone. CHX-A\"-DTPA has a rigid cyclohexane modification and showed increased stability.164-167 H4octapa, H4neunpa and H2bispa2 exploit the picolinic acid motif, which decreases the flexibility of the chelator.24, 209 There remain limitations. Most of the hybrid chelators showed decreased stability compared to DOTA. The reinforced acyclic chelators CHX-A\"-DTPA still showed clear dissociation in vivo. The syntheses of these mentioned above are still quite complex and time consuming and thus, with limited application. Most of them are also quite hydrophilic and will accelerate the clearance of the pharmaceuticals especially when the carrier is small. Herein, we report a new octadentate chelator H4octox which was designed from the inspiration of H4octapa and In(oxine)3 which has been used in white blood cell labeling for 30 years (Scheme 6.2).211 The synthesis of this new chelator is significantly easier and cheaper than those for most 144 previously reported chelators. H4octox showed fast chelation with metal ions (Y3+, In3+, La3+, Lu3+ and Gd3+) and formed highly stable complexes in solution. The in vivo stability was confirmed with 111In SPECT imaging. Moreover, H4octox, compared with the multi-armed carboxylic or picolinic acid based chelators in Scheme 6.1, is more lipophilic and therefore showed a longer circulation time and more liver clearance. Another distinguishing property is its enhanced fluorescence once chelated with Y3+ and Lu3+ which could potentially be useful for bi-modal imaging. Scheme 6.1. Representative chelating ligands for In3+, Y3+, Lu3+, Gd3+ in the literature. 145 Scheme 6.2 Design paradigm of H4octox. 6.2 Experimental 6.2.1 Materials and Methods All solvents and reagents were purchased from commercial sources (TCI America, Sigma Aldrich, Fisher Scientific) and used as received unless otherwise indicated. The analytical thin-layer chromatography (TLC) plates used were aluminum-backed ultrapure silica gel 60 Å, 250 μm thickness; 1 H and 13C NMR spectra were recorded at ambient temperature, unless otherwise noted, on Bruker AV300 and AV400 spectrometers; The 1H NMR spectra were calibrated against residual protio-solvent peak, and the 13C NMR spectra were referenced to the deuterated solvent. NOESY experiments were recorded on a Bruker AV400 spectrophotometer at 400.13 MHz. Low-resolution mass spectrometry was performed on a Waters ZG spectrometer with an ESCI (electrospray/chemical-ionization) source, and high-resolution electrospray ionization mass spectrometry (ESI-MS) was performed on a Micromass LCT time-of-flight (TOF) instrument. Microanalyses for C, H, and N were performed on a Carlo Erba Elemental Analyzer EA 1108. Purification and quality control of [111In(octox)]- for SPECT/CT imaging were performed on an Agilent HPLC system equipped with a model 1200 quaternary pump, a model 1200 UV absorbance detector, and a Bioscan (Washington, DC) NaI scintillation detector. The radio‐detector was connected to a Bioscan B‐FC‐1000 Flow‐count system, and the output from the Bioscan Flow‐146 count system was fed into an Agilent 35900E Interface which converted the analog signal to digital signal. The operation of the Agilent HPLC system was controlled using the Agilent ChemStation software. The HPLC columns used were a semi‐preparative column (Phenomenex C18, 5 µ, 250 × 10 mm) and an analytical column (Phenomenex C18, 5 µ, 250 × 4.6 mm). The HPLC solvents were A: H2O containing 0.1% TFA, and B: CH3CN containing 0.1% TFA. Radioactivity of [111In(octox)]- was measured using a Capintec (Ramsey, NJ) CRC® ‐25R/W dose calibrator. SPECT imaging experiments were conducted using a Siemens (Erlangen, Germany) Inveon microPET/CT scanner. Human serum and Sephadex G-25 PD10 Desalting columns were purchased from Sigma-Aldrich and GE Healthcare, respectively. 111In was obtained from Nordion Inc. (Vancouver, BC, Canada) as 111InCl3 in dilute HCl. pH- and temperature-dependent labeling studies of 111In(octox)]- were performed on a Waters Alliance e2696 separations module coupled to a Waters 2489 UV/Vis-detector (λ = 254 nm) and a LabLogic Scan-RAM radio-detector operated at 950 V. The column was a reversed-phase C18 Waters Atlantis T3, 100 Å, 5 μm particle size (4.6 × 150 mm), supported by a C18 guard cartridge and was operated in an oven at 40 °C. The column was eluted with following gradient: A = 0.1% trifluoroacetic acid (TFA) in water; B = acetonitrile; flow rate = 1 mL/min; 0-20 min 95% A; 20-23 min 100% B. 6.2.2 Synthesis and Characterization Di-tert-butyl 2,2'-(ethane-1,2-diylbis(((8-hydroxyquinolin-2-yl)methyl)azanediyl))-diacetate (6.1) H2hox (312 mg, 0.83 mmol) and NaHCO3 (530 mg, 5 mmol, 6 equiv) were suspended in dry CH3CN and tert-butyl bromoacetate (1.84 mmol, 27 μL) was added dropwise. The mixture was stirred at 50 °C for 24 h and then filtered. The filtrate was collected and purified using column chromatography to obtain light 147 yellow powder (421 mg, 0.7 mmol, yield = 84%). (CombiFlash Rf automated column system; 40 g HP silica; A: hexane, B: EtOAc, 100% A to 80% B gradient) 1H NMR (400 MHz, CD2Cl2, 25 °C) δ 8.05 (d, J = 8.5 Hz, 1H), 7.66 (d, J = 8.5 Hz, 1H), 7.43 (dd, J = 8.2, 7.6 Hz, 1H), 7.31 (dd, J = 8.3, 1.3 Hz, 1H), 7.16 (dd, J = 7.6, 1.3 Hz, 1H), 4.10 (s, 2H), 3.41 (s, 2H), 2.91 (s, 2H), 1.47 (s, 9H). 13C NMR (101 MHz, CD2Cl2, 25 °C) δ 170.9, 158.9, 152.5, 137.7, 136.7, 130.0, 127.5, 122.5, 118.0, 110.2, 81.1, 61.0, 56.5, 52.5, 28.3. 13C NMR (75 MHz, D2O, 25 °C) δ 173.0, 151.7, 147.9, 143.6, 131.4, 129.7, 128.3, 120.8, 119.3, 114.7, 56.3, 55.9, 51.4. H4octox 4HCl·1.8H2O (6.2) Compound (6.1) (103 mg, 0.171 mmol) was suspended in 6 M HCl, stirred at 60 °C overnight and then vacuum dried to a yellow powder as the final product (111 mg, 0.166 mmol, yield = 97%). 1H NMR (300 MHz, D2O, 25 °C) δ 8.31 (d, J = 8.5 Hz, 1H), 7.49 (d, J = 8.5 Hz, 1H), 7.34 (t, J = 8.0 Hz, 1H), 7.24 (d, J = 8.2 Hz, 1H), 6.87 (d, J = 7.6 Hz, 1H), 4.46 (s, 2H), 3.85 (s, 2H), 3.51 (s, 2H). 13C NMR (75 MHz, D2O, 25 °C) δ 173.0, 151.7, 147.9, 143.6, 131.4, 129.7, 128.3, 120.8, 119.3, 114.7, 56.3, 55.9, 51.4. HR-ESI-MS calcd. for [C26H27N4O6 + H]+: 491.1931; found: 491.1930. Elemental analysis: H4octox.·4HCl·1.8 H2O, calcd. % for C 46.69, H 5.06, N 8.38; found: C 46.63, H 5.05, N 8.56. 6.2.3 Synthesis of Octox-Metal complex Na[In(octox)] (6.3) Compound (6.2) (9.8 mg, 0.02 mmol) and In(ClO4)3·6H2O (10.5 mg, 0.02 mmol) were mixed in H2O, and NaOH (0.1M) was used to adjust the pH to 8. The mixture was stirred at 50 °C overnight, and then vacuum dried to obtain the product as a yellow powder. 13C NMR (75 MHz, MeOD, 25 °C) δ 170.5, 153.0, 148.9, 144.7, 131.2, 130.0, 129.3, 122.0, 118.8, 148 115.1, 56.8, 55.1, 52.4. HR-ESI-MS calcd. for [C26H22InN4O6 + H + Na]+: 625.0554; found: 625.0551. 6.2.4 X-Ray Crystallography Crystals of La(III) octox complex was obtained by slow evaporation of a mixture containing La(NO3)3 and H4octapa in 1:1 metal to ligand molar ratio at pH ~ 7 in H2O. An irregular yellow crystal of C104H101La3N16O32 (La3octox4) having approximate dimensions of 0.08  0.10  0.25 mm was mounted on a cryo-loop. All measurements were made on a Bruker APEX DUO diffractometer with a TRIUMPH curved-crystal monochromator with Mo-Kα radiation. The data were collected at a temperature of -183.0 + 0.1 °C to a maximum 2θ value of 44.94°, in a series of ϕ and ω scans in 0.5° oscillations using 15.0 second exposures. The crystal-to-detector distance was 40.16 mm. Of the 78144 reflections, collected, 17280 were unique (Rint = 0.0734); equivalent reflections were merged. Data were collected and integrated using the Bruker SAINT69 software package. The linear absorption coefficient, , for Mo-Kα radiation is 10.16 cm-1. Data were corrected for absorption effects using the multi-scan technique (SADABS),180 with minimum and maximum transmission coefficients of 0.8011 and 0.9219, respectively. The data were corrected for Lorentz and polarization effects. The structure was solved by direct methods.180 The solvent regions could not be reasonably modeled, therefore, the PLATON/SQUEEZE212 program was employed to generate ‘solvent-free’ data. All non-hydrogen atoms were refined anisotropically while all hydrogen atoms were refined isotropically. The standard deviation of an observation of unit weight was 1.00. The weighting scheme was based on counting statistics. The maximum and minimum peaks on the final difference Fourier map corresponded to 2.71 and –1.03 e-/Å3, respectively. Neutral atom scattering factors were taken from Cromer and Waber.213 Anomalous dispersion effects were included in Fcalc;182 the values for Δf' and Δf\" were those of Creagh and 149 McAuley.214 The values for the mass attenuation coefficients are those of Creagh and Hubbell.184 All refinements were performed using the SHELXL-2016185 via the OLEX273 interface. 6.2.5 DFT Calculations All calculations were performed using the Gaussian 09 package (Revision D.01). Full geometry optimizations of the [In(octox)]- and [In(octapa)]- anions were performed using the B3LYP hybrid functional215-216 in aqueous solution using the polarizable continuum model (PCM).217 Geometry optimizations were carried out using the 6-311+G(d,p) basis set on first and second row elements, and the Los Alamos effect core potential (ECP) and valence basis set of double zeta quality (LANL2DZ)125, 218-219 on the indium atom. The input coordinates of atoms were adapted from the crystal structure of the [La3(octox)4] complex and no constraints on symmetry was imposed during the geometry optimization. The resulting geometries showed no imaginary frequencies thus were confirmed to be minima on the potential energy surfaces. The same functional, basis set and ECP were employed to generate its ground state molecular electrostatic potential (MEP) mapping. The MEP was mapped onto the calculated electron density surface. 6.2.6 Solution Thermodynamics Protonation constants and metal stability constants were obtained by combined UV-Potentiometric titrations as described before,209, 220-221 using a Metrohm Titrando 809 equipped with a Ross combined electrode, a Metrohm Dosino 800 and a Varian Cary 60 UV/Vis spectrophotometer (200-450 nm spectral range) connected to a 0.2 cm path length optic dip probe immersed in the titration cell. The titration apparatus consisted of a 20 mL and 25 °C thermostated glass cell and an inlet-outlet tube for nitrogen gas (purified through a 10% NaOH solution) to exclude any CO2 prior to and during the course of the titration. The electrode was calibrated daily 150 for hydrogen ion concentrations using a standard HCl as described before80 and the results were analysed with the procedure of Gran.81 Solutions were titrated with carbonate-free NaOH (0.16 M) that was standardized against freshly recrystallized potassium hydrogen phthalate. Protonation equilibria of the ligand were studied by joined potentiometric-spectrophotometric titrations of a solution containing H4octox (9.55  10-4 M) at 25 °C, l = 0.2 cm and 0.16 M NaCl ionic strength. Electromotive Force values and spectra were recorded after each NaOH or HCl addition and both apparatuses were synchronized in order to have constant delays between each titrant addition and sufficient time to equilibrium. The last two ligand protonation equilibria, outside the electrode threshold, were studied via in batch UV-Vis spectrophotometry on a set of solutions at the same ligand concentration ([H4octox] = 3.06  10-5 M, l = 1 cm) containing different amounts of HCl. The equilibrium H+ concentration in the UV in batch titration procedure at low pH solutions (pH ≤ 2) was calculated from solution stoichiometry, not measured with a glass electrode. For the solutions of high acidity, the correct acidity scale H0 was used.84 For metal complexation studies with In3+, due to the strength of the metal complexes, two different methods were used; direct proton competition experiments and ligand-ligand competition with the known competitor EDTA. The first method used in batch UV-potentiometric measurements on a set of solutions containing 1:1 metal to ligand molar ratio ([H4octox] = [In3+] = 3.06  10-5 M) and different amounts of HCl or NaOH in the spectral range 200-450 nm at 25 °C and l = 1 cm path length, and automated UV-potentiometric titrations - H4octox] = [In3+] = 8.13  10-4 M, 25 °C, I = 0.16 M NaCl and l = 0.2 cm path length. The second method used ligand-ligand pH-potentiometric competition titrations with EDTA at [H4octox] = 9.86  10-4 M, [In3+] = 3.12  10-4 M and [EDTA] = 2.91  10-4 M, 25 °C, I = 0.16 M NaCl. 151 Complex formation equilibria with Y(III), La(III) or Gd(III) were studied by direct UV-potentiometric titrations of 1:1 metal to ligand molar ratio ([H4octox] = [Mn+] = 9.86  10-4 M) at 25 °C, I = 0.16 M and 0.2 cm path length. Metal solutions were prepared by adding the atomic absorption (AA) standard metal ion solutions to a H4octox solution of known concentration in the 1:1 metal to ligand molar ratio. The exact amount of acid present in the indium, lanthanum, gadolinium, yttrium and lutetium standards was determined by Gran’s method81 titrating equimolar solutions of either metal ions and EDTA. All the potentiometric measurements were processed using the Hyperquad2013 software85 while the obtained spectrophotometric data were processed with the HypSpec86 program. The molar absorptivities of all protonated species of H4octox were included in the metal stability calculations. Protonation constants of EDTA and its In3+ metal stability constants were taken from the literature222 while proton dissociation constants corresponding to hydrolysis of In(III), Y(III), La(III), Gd(III) and Lu(III) aqueous ions included in the calculations were taken from Baes and Mesmer.87 The species formed in the studied systems are characterized by the general equilibrium: pM + qH + rL = MpHqLr (charges omitted). For convention, a complex containing a metal ion, M, proton, H and ligand, L, has the general formula MpHqLr. The stoichiometric indices p, might also be 0 in the case of protonation equilibria, and negative values of q refers to proton removal from coordinated water, equivalent to hydroxide ion addition during formation of the complex. The overall equilibrium constant for the formation of the complexes MpHqLr from its components is designated as log β. Stepwise equilibrium constants log K correspond to the difference in log units between the overall constants of sequentially protonated (or hydroxide) species. pM is defined as (- log[Mn+]free) and is always calculated at [Mn+] = 1 µM, [Lx-] = 10 µM, pH 7.4 and 25 °C.92 152 6.2.7 pH- and Temperature-dependent [In(octox)]- Labeling Procedure. All labelling reactions were over a 15-minute period. The reaction mixture was kept at 1 mL with NaOAc buffer (10 mM, pH = 5.5) or NH4OAc buffer (0.15 M, pH = 4), using approximately 80-100 µCi of 111In3+, and 100 µL of the appropriate H4octox solution. The H4octox solutions were prepared by dilution of a stock 1  10-3 M solution (2 mg of H4octox in 3 mL of MQ H2O). Two reaction conditions were varied in these experiments; pH (4 and 5.5) and temperature (RT and 60 °C). The reaction progress was monitored using HPLC and a 4-micron Synergy Hydro-RP Analytical column. A linear gradient of 0% → 100% B (A, 0.1 % trifluoroacetic acid (TFA) in water; B, acetonitrile) over 20 minutes was used to separate the product. Radioactive product was detected with a Raytest GABI* HPLC Gamma Spectrometer. 6.2.8 Serum Stability PD-10 Desalting column method: A quantitatively labelled reaction with ~ 3 mCi of activity is divided into replicates and each replicate is diluted to 1 mL. An equal volume (1 mL) of human serum was added to a quantitatively radiolabeled [111In(octox)]- complex and the mixture was incubated at 37 °C. At time points of 1 hour, 1 day and 5 days, a 400 uL aliquot (or 800 µL on day 5) was diluted to 2.5 mL and the activity was measured with the CRC55tR dose calibrator. The dilution was then loaded onto an equilibrated PD-10 desalting column, and the activity of the empty vial was measured to determine the “residual activity”. Once the entire mixture has been adsorbed onto the column, the proteins (MW > 5000 Da) are eluted with 3.5 ml of PBS. The activity of the elution is then measured and the stability of the complex in % is calculated as: % stable = 1 - (activity of elution) / ((activity of load) - (activity residual)). 153 6.2.9 111In Radiolabeling of H4octox for in vivo studies 111InCl3 (3.22 mCi, 10 µL, in dilute HCl, Nordion Inc., Canada) was added to a fresh aqueous solution of H4octox (100 µL, 2 mM), the pH was raised to pH ~ 7 by addition of Na2CO3 (5 μL, 0.1 M), and the mixture was agitated (350 rpm) at room temperature for 15 min. An aliquot was analyzed by radio-HPLC confirming 98% radiolabeling. Radio-HPLC [111In(octox)]-: tR = 6.1 min. Doses for animal injections were prepared by diluting 78 μL of the above solution with PBS (530 μL, pH 7.4). 6.2.10 In vivo SPECT/CT imaging This was performed in accordance with the Canadian Council on Animal Care (CCAC) and protocol approved by the Animal Care Committee (ACC) of the University of British Columbia (A16-0150). Three healthy C57Bl/6 female mice (~ 16 g) were anaesthetized using isoflurane on a precision vaporizer (5% in oxygen for induction, between 1.5 and 2.5% in oxygen for maintenance) and received a subcutaneous injection of lactated Ringer’s solution (0.5 mL) for hydration prior to each imaging scan. After the induction of anesthesia, an injection containing 120 µL of [111In(octox)]- complex in PBS was administered via tail vein. Average injected activities were 310 µCi. Immediately after injection, dynamic whole-body images were acquired during 60 min using a multimodal SPECT/CT scanner (VECTor/CT, MILabs, The Netherlands) equipped with a XUHS-2 mm mouse pinhole collimator. Six frames of 10 min were acquired for the first hour scan. Thereafter, acquisitions were done at 5 and 24 h post-radiotracer injection using a single frame of 40 and 60 min, respectively. Throughout the entire scanning procedure, the mouse was kept under isoflurane anesthesia and constant body temperature was maintained using a heating 154 pad. Following each SPECT acquisition, a whole-body CT scan was acquired to obtain anatomical information and both images were registered. The 111In photopeak window was centered at 171 keV with a 20% energy window width. For quantitative analysis, SPECT image reconstructions were carried out with a pixel-ordered subset expectation maximization (POSEM) algorithm that included resolution recovery and compensation for distance-dependent pinhole sensitivity. For the SPECT images, 16 subsets, 10 iterations and an isotropic 0.4 mm voxel grid were used. The images were decay corrected and after CT registration, attenuation correction was applied. For visual representation, the reconstructed volumes of SPECT scans were post-filtered with a 3D Gaussian filter. CT scans were acquired with a tube setting of 55 kV and 615 µA. In total 2 frames of 180 projections over 360 degrees were acquired in step and shoot rotation mode. The acquired projection data was reconstructed using SkyScan NRecon software to generate a 3D CT image on 0.169 mm3 voxel size. Volumes of interest (VOIs) were manually defined using AMIDE (v.1.0.5) to determine the time activity pattern per target organ. Thus, the delineated regions were liver, bladder, bone and gallbladder. The average organ activity per volume was obtained from the SPECT images and the Standardized Uptake Value (SUVs) was extracted from each organ. In order to relate the scanner units (counts/voxel) to radioactivity concentration, a calibration factor was determined scanning a source with a known concentration of 111In. Mice were sacrificed for ex vivo biodistribution and the radioactivity in diverse organs was determined by γ-counting. 6.2.11 Biodistribution of [111In(octox)]- A full biodistribution was conducted (blood, heart, liver, kidneys, lungs, small intestine, brain, bladder, muscle, spleen, stomach, bone, tumor, pancreas and feces) following the last scan at 24 h post-injection. Organs were cleaned from blood, weighed and the activity determined using a γ-155 counter (Packard Cobra II auto-gamma counter, Perkin Elmer, Waltham, MA, USA). The calibration factor for 37 kBq of 111In was 78,395 cpm (instrument specific). Total organ weights were used for the calculations of injected dose per gram of tissue (% ID/g organ) except for blood, liver, muscle, bone and pancreas, where average literature values were used. 6.2.12 Fluorescence Spectra Solutions of H4octox and its respective metal complexes with Y3+, Lu3+, La3+ and In3+ ([H4octox] = 2  10-4 M and 1:1 metal to ligand molar ratios) were prepared in Milli-Q water and the pH was adjusted to 8 by adding NaOH(0.1M). Fluorescence emission spectra were measured at excitation wavelength λexc = 350 nm using an Agilent Cary Eclipse Fluorescence Spectrophotometer. 6.3 Results and Discussion Scheme 6.3. Synthesis of H4octox. Reagents and conditions: a) CH3CN, tert-butyl bromoacetate, Na2CO3 (excess), 50 °C, 24 h, 84%; b) HCl (6 M), 60 °C, overnight, 97%. 6.3.1 Synthesis and Characterization The simple bidentate ligand 8-hydroxyquinoline (oxine) has proven to have a high affinity for In3+; 111In(oxine)3 has been used in white blood cell labeling for 30 years.211 Therefore 8-hydroxyquinoline could be a really good motif upon which to build larger multidentate chelator platforms. H4octapa24 has shown excellent in vitro and in vivo stability with In3+ and its 156 bifunctional derivative showed even better performance than DOTA and DTPA. By combining the geometry of H4octapa and the rigid 8-hydroxyquinoline motif, we designed H4octox as shown in Scheme 6.2. In the synthesis of H4octox (Scheme 6.3), the precursor H2hox, was synthesized first as we reported before,168 and then conjugated with tert-butyl bromoacetate. The protection of the phenol group was not needed in this step, since the nucleophilicity of the secondary amine backbone is higher than that of the phenol oxygen in the 8-hydroxyquinoline. The HCl salt of H4octox was finally obtained by deprotection of tert-butyl ester groups using 6 M HCl. The final product was characterized by 1H and 13C NMR spectroscopy and purity was confirmed by elemental analysis. The whole synthetic route is straightforward and the cumulative yield (74%) is higher than that for any of H4octapa, H4neunpa and most of the other chelators in Scheme 6.1. The use of multiple synthons is also an advantage for a straightforward preparation of bifunctional derivatives using a functionalized diamine backbone (for example, (4-nitrobenzyl)-ethylenediamine) compared with the tedious synthesis of certain bifunctional macrocyclic chelators. The good synthetic accessibility together with the low cost of the starting materials easily available from commercial suppliers is of great importance for wide uptake, application and clinical use. [In(octox)]− was prepared by directly mixing the HCl salt of H4octox with In(ClO4)3 in H2O in equimolar ratio and adjusting the pH to 8 with NaOH. Metal complexation of H4octox with In3+ was confirmed by 1H NMR spectroscopy (Figure 6.1), and a fully 8-coordinated solution structure of [In(octox)]− is suggested from the sharp resolved diastereotopic splitting of the protons associated with all the six methylene –CH2 protons in H4octox (Figure 6.2). Two distinct symmetrical isomeric species are present in solution at pH 8 based on the integration of the well resolved aromatic proton signals (6.1A and 6.1B) in Figure 6.1a. The existence and exchange of 157 the two possible isomers was further evidenced by VT-NMR and NOESY experiments (Figures 6.1b and 6.3). At room temperature, the exchange is slow in the NMR time scale, so two different sets of signals are observed, and when the temperature is raised (Figure 6.1b), the exchange between the two isomers is faster and two peaks (6.1A and 6.1B) merged into one broad peak (6.1(A+B)); similar changes can be observed for the rest of signals. NOESY experiments at room temperature showed exchange between the two sets of peaks within both the aromatic and aliphatic protons. The expected NOE within aromatic protons between protons 1 and 2, 2 and 3 as well as 4 and 5 for both isomers were observed, however no NOE was observed between aliphatic protons due to a small T1 for those protons and the relatively low concentration (Figure 6.3). Figure 6.1. a) Portion of the 1H NMR spectrum of the [In(octox)]- complex showing the aromatic protons (pH = 8, 400 MHz, 25 °C); b) VT 1H NMR experiment for [In(octox)]- (pH = 8, 400 MHz). A and B represent the two different species in solution of the [In(octox)]- complex. 158 Figure 6.2. 1H NMR spectrum of Na[In(octox)] in D2O (400 MHz, 25 °C). Figure 6.3. The 2D NOESY spectrum of Na[In(octox)] in D2O (400 MHz, 25 °C). 159 6.3.2 X-ray Crystallography Yellow colored irregular crystals suitable for solid-state X-ray analysis were obtained of the complex. The material crystallizes with four crystallographically independent molecules in the asymmetric unit, three having lanthanum as metal centre while one lacks a metal centre (Figure 6.4). Additionally, there are sites in the asymmetric unit occupied by cations and solvent molecules (either CH2Cl2 or CH3OH). The solvent molecules undergo significant thermal motion, and in some cases the sites are partially occupied by both solvents. Out of the three crystallographically independent molecules of [La3(octox)4]7- 7H2O, La2 and La3 are ten-coordinated (N4O6) having the same coordination environment while La1 is nine-coordinated (N4O5). La1 forms bonds with the eight inherent donor atoms (N4O4) from the binding cavity of octox4-, and one O atom from – COO- group from another octox4- molecule binding to La2. La2 is ten-coordinated, with eight sites similarly occupied by donor atoms from the octox4- scaffold, O atom from –COO- group from the octox4- molecule binding to La3 and one molecule of water. La3 is ten-coordinated as well, forming bonds with octox4- binding cavity (N4O4), one O atom from –COO- group from octox4- molecule, and one molecule of water. The fourth crystallographically independent molecule is that of the octox4- ligand itself. One of the carboxylate O atoms is bound to La3, and the aromatic rings are π stacked together. The structure of the [La(octox)]- complex is not of very high quality due to the poor quality of the crystals, but the coordination environment of the La(III) ion is perfectly clear. In spite of several attempts using different disorder models, the refinement of the co-crystallized solvent molecules and cations remained unstable. The solvent regions could not be reasonably modeled, therefore the PLATON/SQUEEZE72 program was employed to generate a ‘solvent-free’ data set. The structure details can be found in Figure 6.4. Selected average bond distances and angles are shown in Table 160 6.1. In Figure 6.5, it can be seen that La3 is ten-coordinated, the coordination sphere consists of eight donor atoms (N4O4) from octox4- binding cavity, one O atom (O18) from the –COO- group from another octox4- molecule, and one molecule of water. There is a lack of a clear and well-defined geometry around the lanthanum center which is often the case with non-rigid lanthanide complexes. Nevertheless, the crystal structure for [La(octox)]- provides a visually compelling insight into the nature of the interaction between a trivalent metal ion and the octox4- motif. Figure 6.4. ORTEP diagram of the [La3(octox)4]7-•7H2O in the asymmetric with 50% probability ellipsoids; associated crystallographic data are presented in Appendix Table C.1. 161 Figure 6.5. ORTEP diagram of 10 coordinated La[(octox)]- in the asymmetric unit (50% probability ellipsoids). The solvent molecules, cations, and hydrogens are omitted for clarity. Table 6.1. Selected average bond angles and bond distances in the [La(octox)]- complex of Figure 6.5. [La(octox)]- complex N(ox)-M (Å) 2.708(7) N(en)-M (Å) 2.726(7) O(ox)-M (Å) 2.585(6) O(COO)-M (Å) 2.556(6) O(aqua)-M (Å) 2.560(6) N(ox)-M-N(ox) angle (deg) 172.0(2) 162 Figure 6.6. Ball-and-stick presentation of the two isomeric species simulated using DFT calculations. 6.3.3 DFT Simulations and Molecular Electrostatic Potential Maps DFT calculations were carried out to explore the behavior of the anion [In(octox)]- in solution. Two most stable isomeric structures A and B were identified featuring different bond conformations on the ethylenediamine backbone (Figure 6.6). Both structures have 8-coordinated metal centers and exhibit similar C2v–like symmetry and calculated dipole moments. Structure A with equatorial hydroxyquinoline arms is only 8.7 kJ/mol less stable than structure B with axial hydroquinoline groups. These results are consistent with the observations in the 1H NMR solution study of [In(octox)]- in which two coexisting isomers were found to have different coupling patterns in the aliphatic region. Similar isomers have also been identified in the [In(octapa)]- 24 anion and the corresponding structure A is 6.4 kJ/mol less stable than structure B. These isomer were not observed in [In(octapa)]- possibly due to a much faster equilibrium between isomers 163 resulting from the less rigid ligand structure in octapa4-, therefore only the-averaged signal was observed.24 [In(octox)]- and [In(octapa)]- have been compared by examining the calculated structures and electrostatic potentials. As shown in Table 6.2, [In(octapa)]- shows areas of much higher electronegative potential around the picolinic acid group compared to the region of the hydroxyquinoline groups in [In(octox)]-. This difference may result in a lower propensity toward protonation and slower decomposition under acidic conditions for [In(octox)]-. Table 6.2. DFT optimized structures and electrostatic potential (ESP) mapping of [In(octapa)]- and In(octox)]-. Anions Top-down Side-view [In(octox)]- [In(octapa)]- *The ESP mappings represent a maximum potential of 0.03 au, and a minimum of -0.20 au, mapped onto electron density isosurfaces of 0.002 e·Å-3 (red to blue = negative to positive). 6.3.4 Solution Thermodynamics When evaluating metal chelating ligands for radiopharmaceutical applications, knowledge of their acid-base properties (protonation constants) and the thermodynamic stability of their metal 164 complexes is tantamount, and in particular, the extent to which a metal complex is formed in solution at physiologically relevant conditions ca. pH 7.4. The most effective thermodynamic comparison of the vast array of metal chelators that have been developed for radiopharmaceutical purposes is via conditional stability constants. pM (the metal-ion-scavenging ability) was introduced by Raymond92 (defined as -log [Mfree] at [L] = 10 M, [M] = 1 M and pH = 7.4) and can give an insight of the stability of the complexes in vivo, allowing for the most suitable comparison of the ability of different ligands, with diverse basicities and protonation states, to sequester specific metal ions. pM values are linearly correlated to the conditional stability constant of metal complexes but most importantly depend on the ligand basicity and/or denticities and difference in metal-ligand stoichiometries and thus on the competition of the metal and proton for the same coordinating groups.141-142, 223 Recently our group has reported an exhaustive UV spectrophotometric analysis (see Chapter 3) of the protonation equilibria of H2hox, a new promising hexadentate Ga3+ chelator for PET imaging.168 In this chapter are reported the eight protonation constants of the octadentate derivative version of H2hox, H4octox. The two additional carboxylic acid substituents in H4octox confer a higher aqueous solubility than that of H2hox, allowing the use of the simultaneous UV-potentiometric titrations for the determination of protonation constants. The dissociations of the most acidic protons in H4octox (H4L), species H8L2+ and H7L+, respectively, were determined by in-batch UV spectrophotometric titrations, as the pH was lower than the electrode threshold. As in the case of H2hox, the spectral features of H4octox in the spectral range 200-450 along the titration allowed for the determination and assignment of the protonation events to the different functionalities (Figures 6.7, 6.8). The protonation constants were calculated using the 165 HypSpec201486 program and are summarized in Table 6.3. Solution speciation diagrams were calculated using the acidity constants given in Table 6.3 and are shown in Figure 6.9. Figure 6.7. Representative spectra of the in batch UV spectrophotometric titration of 3.06  10-5 M solution of H4octox in H2O, l = 1 cm at 25 °C. 200 225 250 275 3000.00.51.01.52.0pH 6.56AbsorbanceWavelength (nm) 6.56 7.03 7.74 8.18 8.73 9.18 9.37pH 9.36 = 252 nm e)200 225 250 275 3000.00.51.01.52.0pH 6.56AbsorbanceWavelength (nm) 4.08 4.45 5.08 5.50 5.80 6.08 6.56pH 4.08 = 250 nmd)200 225 250 275 3000.00.51.01.52.0pH 4.08AbsorbanceWavelength (nm) 1.74 1.96 2.53 3.54 3.91 4.08pH 1.74 = 249 nmc)275 300 325 350 375 400 425 4500.000.020.040.060.080.100.120.140.160.180.20pH 1.71AbsorbanceWavelength (nm) Ho -0.92 Ho -0.59 Ho -0.16 Ho 0.11 Ho 0.24 pH 0.62 pH 1.09 pH 1.36 pH 1.71H -0.91b)200 225 250 275 3000.00.51.01.52.0pH 1.71AbsorbanceWavelength (nm) Ho -0.92 Ho -0.59 Ho -0.16 Ho 0.11 Ho 0.24 pH 0.62 pH 1.09 pH 1.36 pH 1.71H -0.91 = 249 nm a)200 225 250 275 3000.00.51.01.52.0pH 9.49AbsorbanceWavelength (nm) 9.49 9.85 10.10 10.21 10.38 10.61 10.77 11.08pH 11.08  = 249 nmf)166 Figure 6.8. Representative spectra of the simultaneous UV-potentiometric titration of 9.55  10-4 M solution of H4octox, l = 0.2 cm at I = 0.16 M NaCl and 25 °C. Figure 6.9. Speciation plot of H4octox (H4L) calculated using protonation constants in Table 6.1, at ligand concentration 1  10-3 M. 275 300 325 350 375 400 425 4500.000.250.500.751.00pH 3.96AbsorbanceWavelength (nm) 2.47 2.63 2.71 2.99 3.44 3.96 = 310 nm pH 2.47a)275 300 325 350 375 400 425 4500.000.250.500.751.00pH 3.96AbsorbanceWavelength (nm) 3.96 4.16 4.44 4.73 4.99 5.24 5.50 5.85 6.50 7.34 = 323nm pH 7.34b)275 300 325 350 375 400 425 4500.000.250.500.751.00c)pH 9.32AbsorbanceWavelength (nm) 7.93 8.55 8.79 8.98 9.15 9.32 = 295nm pH 7.93275 300 325 350 375 400 425 4500.000.250.500.751.00d)pH 11.36AbsorbanceWavelength (nm) 9.48 9.69 9.81 9.92 10.02 10.11 10.20 10.27 10.34 10.41 10.47 10.53 10.59 10.66 10.75 10.87 11.04 11.15 11.36 = 321nm pH 9.480 2 4 6 8 10 12020406080100L4-HL3-H2L2-H4LH5L+H6L2+H7L3+H8L4+% Formation relative to ligandpHH3L-167 Table 6.3. Protonation constants of H4octox (H4L) at 25 °C. Equilibrium Reaction log β log K L + H+ ⇆ HL 10.65(1) (a) 10.65 HL + H+ ⇆ H2L 20.67(1) (a) 10.02 H2L + H+ ⇆ H3L 29.60(1) (a) 9.03 H3L + H+ ⇆ H4L 34.78(1) (a) 5.18 H4L + H+ ⇆ H5L 37.83(1) (a) 3.05 H5L + H+ ⇆ H6L 39.86(2) (a) 2.03 H6L + H+ ⇆ H7L 39.55(8) (b) -0.31 H7L + H+ ⇆ H8L 38.88(7) (b) -0.67 a Joined UV-potentiometric titrations at I = 0.16 M NaCl; b in-batch UV spectrophotometric titrations, not evaluated at constant I = 0.16 M NaCl. Charges are omitted for clarity. From the analysis of the UV spectra of the most acidic solutions of H4octox (Hº -0.91 to pH 1.71), (Figure 6.7 a-b), the ligand presents similar spectroscopic features to those of H2hox with two quinoline chromophores protonated (H6hox4+ and H5hox3+), sharing the bands at λmax = 260 and 378 nm and an isosbestic point at λ = 249 nm. This allows for the assignment of the two most acidic pKa values to the dissociation of the protons from the two quinoline nitrogen atoms, H8octox4+ and H7octox3+ (pK1 = -0.67(7) and pK2 = -0.31(8)). Smaller spectroscopic variations between pH 1.71-4 (Figure 6.7 c and Figure 6.8 a), with λmax = 246, 260 and 324 nm and isosbestic points at λ = 249 and 310 nm, are due to the carboxylic acid deprotontation (pK3 = 2.03(2) and pK4 = 3.05(1)), corresponding to H6octox2+ and H5octox+. The dissociation of the protonated tertiary ammonium in the backbone in the pH range 4-9.3, involving the species H4octox and H3octox-, is characterized by two different spectroscopic evolutions; initially the neutral species H4octox deprotonates with pK5 = 5.18(1), and the band at λmax = 246 nm shifts to λmax = 243 nm with an isosbestic at 250 nm (Figure 6.7 d), meanwhile a new band appears at 308 nm and a new isosbestic 168 point at 323 nm (Figure 6.8 b). Secondly, H3octox- deprotonates with pK6 = 9.03(1) and the band at λmax = 243 nm decreases in intensity and a new band at 260 nm appears with a shift of isosbestic point to 252 nm (Figure 6.7 e), while at higher wavelengths in the spectra, a new band appears at 370 nm and a new isosbestic point at 295 nm is observed (Figure 6.8 c). The pK5 = 5.18(1) attributed to one Nen in the backbone if compared to the one for the H4hox2+ species (pK3 = 6.06(2)), is more acidic due to the presence of the electron-withdrawing –COOH substituent. The higher basicity of the H3octox- species (pK6 = 9.03(1)) compared to that of H3hox+ (pK4 = 8.39(1)) can be explained by the anionic charge. The last two dissociation equilibria involve the phenol–OH of the quinoline moieties (pK7 = 10.02(1) and pK8 = 10.65(1)). The species H2octox2- and Hoctox3- deprotonate and the fully deprotonated octox4- is formed. These last two dissociation steps are characterized as in the case of H2hox ligand, by the disappearance of the band at 243 nm and the increase of the band at 260 nm with the isosbestic point at 249 nm and at higher wavelengths the decrease of the band at 308 nm while an increase of two bands at 337 and 354 nm are observed with a new isosbestic point at 321 nm (Figures 6.7 f and 6.8 d). In order to evaluate the affinity of H4octox towards metal ions of clinical interest - In(III), Y(III), La(III), Gd(III) and Lu(III) - the thermodynamic stability constants of each metal-octox complex have been evaluated by combined UV-potentiometric titrations and, in the case of In(III) due to the high affinity, additional acidic spectrophotometric competitions and ligand-ligand competition potentiometric titrations with the known competitor EDTA were required. In Table 6. 4 are summarized the related stability constants that were refined by using the HypSpec program and in Figures 6.10 – 6.13 and Figure 6.15 b the related speciation plots. Combined potentiometric-spectrophotometric titrations of H4octox with each of Y(III), La(III), Gd(III) and Lu(III) showed the start of the complex formation from pH ~ 2, based on the 169 distinct features of the spectra compared with the electronic spectra of H4octox (6.10 – 6.13), and three consecutive transformations with increasing the pH due to the loss of protons from the species [M(H2L)]+, M(HL) and [M(L)]- (L4- = octox4-), respectively. All of these metal complex deprotonations are marked by the appearance of well-defined isosbestic points (6.10 – 6.13). The deprotonations of the neutral species M(HL) may be attributed to the last protonated OH-quinoline moiety, based on the similarity of spectroscopic features to those of [Ga(hox)]+ and Ga(oxine)3 and on its higher basicity compared to that of the –COOH substituents, while the last deprotonation is presumably due to a coordinated water molecule with higher pKa (10.41-10.81), as previously observed in similar systems.209, 220 H4octox with In(III) yielded a complete reaction even at pH ~ 1.6 (spectra in Figures 6.14 a and b), and since the determination of thermodynamic stability constants requires the values of the concentration of free and bound metal ions at equilibrium, the direct potentiometric-spectrophotometric method was insufficient and competitive methods were applied to this system, for which the stability constants were determined (Table 6.4). The system, as with previous metal ions, presented different transformations with pH and [In(H2L)]+, In(HL), [In(L)]- and [In(OH)(L)]2- species were identified. A particularly high log K = 31.38(1) for the species [In(L)]- was determined (Table 6.4). The analysis of Table 6.5 and Figure 6.15, and in particular, the pM values, shows that the size of the metal cation within the cavity of octox4- is best fitted for the smaller and harder metal In3+, and the trend in metal selectivity of this new octadentate acyclic chelator by means of pM value is In3+ > Lu3+ > Y3+ > Gd3+ > La3+. 170 Figure 6.10. Representative spectra of the UV-potentiometric titration of the Y3+-H4octox system, showing the different transformations as the pH is raised a) from 1.68 to 2.35, b) from 2.35 to 3.13, c) from 3.55 to 4.53, d) from 4.53 to 7.61 and e) from 6.58 to 10.74. [H4octox] = [Y]3+ = 8.57  10-4 M, path length = 0.2 cm, I = 0.16 M (NaCl) at 25 °C. f) Distribution curves for Y3+-H4octox complexes; [Y3+] = [H4octox] = 8.5  10-4 M, dash line represents physiological pH (7.4). 275 300 325 350 375 400 425 4500.000.250.500.751.00pH 2.35 1.68 1.75 1.83 1.95 2.05 2.18 2.35AbsorbanceWavelength (nm)pH 1.68a)275 300 325 350 375 400 425 4500.000.250.500.751.00 = 286 nmpH 2.35 2.35 2.53 2.60 2.66 2.70 2.75 2.79 2.84 2.89 2.94 3.01 3.06 3.13AbsorbanceWavelength (nm)pH 3.13 = 353 nmb)275 300 325 350 375 400 425 4500.000.250.500.751.00pH 4.53 3.55 3.64 3.75 3.94 4.03 4.22 4.53AbsorbanceWavelength (nm)pH 3.55 = 340 nmc)275 300 325 350 375 400 425 4500.000.250.500.751.00 = 282 nmpH 4.53 4.53 4.95 5.36 5.69 6.09 6.58 7.61AbsorbanceWavelength (nm)pH 7.61 = 340 nmd)2 4 6 8 10 120255075100[Y(OH)L]2-[Y(L)]-Y(HL)[Y(H2L)]+% Formation relative to Y3+pHY3+f)275 300 325 350 375 400 425 4500.000.250.500.751.00pH 6.58 6.58 7.61 9.32 10.36 10.59 10.74AbsorbanceWavelength (nm)pH 10.74e)171 Figure 6.11. Representative spectra of the UV-potentiometric titration of the La3+-H4octox system, showing the different transformations as the pH is raised a) from 1.89 to 3.48, b) from 3.48 to 5.86, c) from 5.86 to 9.20, d) from 9.98 to 11.43. [H4octox] = [La3+] = 7.65  10-4 M, path length = 0.2 cm, I = 0.16 M (NaCl) at 25 °C. e) Distribution curves for La3+-H4octox complexes; [La3+] = [H4octox] = 8.5  10-4 M, dash line represents physiological pH (7.4). 275 300 325 350 375 400 425 4500.000.250.500.751.00 = 283 nm pH 1.89 1.89 2.03 2.17 2.26 2.39 2.47 2.53 2.64 2.75 2.97 3.48AbsorbanceWavelength (nm)pH 3.48 = 341 nm a)275 300 325 350 375 400 425 4500.000.250.500.751.00 = 279 nm pH 5.86 3.48 3.71 4.02 4.41 4.61 4.80 5.02 5.33 5.57 5.86AbsorbanceWavelength (nm)pH 3.48 = 336 nm b)275 300 325 350 375 400 425 4500.000.250.500.751.00 = 279 nm pH 5.86 5.86 6.13 6.37 6.60 6.84 7.12 7.52 9.20AbsorbanceWavelength (nm)pH 9.20 = 333 nm c)275 300 325 350 375 400 425 4500.000.250.500.751.00pH 9.98 9.98 10.64 10.98 11.12 11.37 11.43AbsorbanceWavelength (nm)pH 11.43d)2 4 6 8 10 120255075100[La(OH)L]2-[La(L)]-La(HL)[La(H2L)]+% Formation relative to La3+pHLa3+e)172 Figure 6.12. Representative spectra of the UV-potentiometric titration of the Gd3+-H4octox system, showing the different transformations as the pH is raised a) from 1.77 to 2.15, b) from 2.15 to 3.22, c) from 3.26 to 4.96, d) from 4.96 to 7.93. [H4octox] = [Gd3+] = 7.03  10-4 M, path length = 0.2 cm, I = 0.16 M (NaCl) at 25 °C. e) Distribution curves for Gd3+-H4octox complexes; [Gd3+] = [H4octox] = 8.5  10-4 M, dash line represents physiological pH (7.4). 275 300 325 350 375 400 425 4500.000.250.500.751.00pH 2.15 1.77 1.81 1.98 2.09 2.15AbsorbanceWavelength (nm)pH 1.77a)275 300 325 350 375 400 425 4500.000.250.500.751.00 = 346 nmpH 2.15 2.15 2.26 2.38 2.46 2.51 2.57 2.68 2.77 2.85 3.22AbsorbanceWavelength (nm)pH 3.22 = 284 nmb)275 300 325 350 375 400 425 4500.000.250.500.751.00 = 339 nmpH 4.96 3.26 3.43 3.68 3.84 4.03 4.26 4.56 4.96 AbsorbanceWavelength (nm)pH 3.26 = 280 nmc)275 300 325 350 375 400 425 4500.000.250.500.751.00 = 339 nmpH 4.96 4.96 5.17 5.38 5.60 5.81 6.02 6.53 6.90 7.93AbsorbanceWavelength (nm)pH 7.93 = 282 nmd)2 4 6 8 10 120255075100[Gd(OH)L]2-[Gd(L)]-Gd(HL)[Gd(H2L)]+% Formation relative to Gd3+pHGd3+e)173 Figure 6.13. Representative spectra of the UV-potentiometric titration of the Lu3+-H4octox system, showing the different transformations as the pH is raised a) from 2.44 to 3.0, b) from 3.24 to 4.02, c) from 4.02 to 6.51, d) from 6.51 to 11.44. [H4octox] = [Lu3+] = 6.93  10-4 M, path length = 0.2 cm, I = 0.16 M (NaCl) at 25 °C. e) Distribution curves for Lu3+-H4octox complexes; [Lu3+] = [H4octox] = 8.5  10-4 M, dash line represents physiological pH (7.4). 275 300 325 350 375 400 425 4500.000.250.500.751.00 = 362 nmpH 2.44 2.44 2.53 2.61 2.71 3.0 AbsorbanceWavelength (nm)pH 3.0 = 288 nma)275 300 325 350 375 400 425 4500.000.250.500.751.00 = 343 nmpH 4.02 3.24 3.44 3.54 3.68 4.02 AbsorbanceWavelength (nm)pH 3.24 = 279 nmb)275 300 325 350 375 400 425 4500.000.250.500.751.00 = 343 nmpH 4.02 4.02 4.15 4.28 4.43 4.58 4.75 4.96 5.20 5.61 6.51AbsorbanceWavelength (nm)pH 6.51  = 280 nmc)275 300 325 350 375 400 425 4500.000.250.500.751.00pH 6.51 6.51 9.31 10.28 11.38 11.44AbsorbanceWavelength (nm)pH 11.44d)2 4 6 8 10 120255075100[Lu(OH)L]2-[Lu(L)]-Lu(HL)[Lu(H2L)]+% Formation relative to Lu3+pHLu3+e)174 Figure 6.14. a) Representative spectra of the in batch UV spectrophotometric titration of the In3+-H4octox system, showing the complex formation as the pH is raised from 0.74 to 1.65. [H4octox] = [In3+] = 3.04  10-5 M, path length = 1 cm, I = 0.16 M (NaCl) at 25 °C. Representative spectra of the UV-potentiometric titration of the In3+-H4octox system, showing the different transformations as the pH is raised b) from 1.59 to 2.24, c) from 2.23 to 3.33, d) from 3.43 to 9.94, e) from 9.51 to 11.40. [H4octox] = [In3+] = 8.13  10-4 M, path length = 0.2 cm, I = 0.16 M (NaCl) at 25 °C. 275 300 325 350 375 400 425 4500.000.050.100.150.20pH 0.74 pH 1.65 AbsorbanceWavelength (nm) 0.74 0.91 1.14 1.28 1.45 1.65 = 374 nma)275 300 325 350 375 400 425 4500.000.250.500.751.00pH 2.24AbsorbanceWavelength (nm) 1.59 1.66 1.74 1.81 1.92 2.0 2.14 2.23 = 374 nmpH 1.59b)275 300 325 350 375 400 425 4500.000.250.500.751.00 = 284 nmpH 3.33AbsorbanceWavelength (nm) 2.23 2.39 2.48 2.58 2.67 2.80 2.99 3.33 = 363 nmpH 2.23c)275 300 325 350 375 400 425 4500.000.250.500.751.00 = 290 nmpH 3.43AbsorbanceWavelength (nm) 3.43 3.76 4.67 5.56 6.22 6.63 6.98 9.94 = 359 nmpH 9.94d)275 300 325 350 375 400 425 4500.000.250.500.751.00pH 11.40AbsorbanceWavelength (nm) 9.51 10.73 11.12 11.19 11.40 = 382 nmpH 9.51e)175 Table 6.4. Stepwise stability constants (log K) of H4octox complexes with Y3+, La3+, Gd3+, Lu3+ and In3+. Equilibrium reaction Y3+ a La3+ a Gd3+ a Lu3+ a In3+ M3+ + L ⇆ ML 23.78(3) 21.91(2) 23.54(2) 24.66(1) 31.38(1)a; 31.64(3)c ML + H+ ⇆ MHL 5.49(3) 6.78(1) 5.96(2) 4.84(2) 6.76(1)a; 7.08(6)c MHL + H+ ⇆ MH2L 3.71(5) 4.85(2) 4.05(3) 3.51(2) 2.66(1)b M(OH)L + H+ ⇆ ML 10.61(4) 10.81(3) 10.61(4) 10.41(2) 10.69(5)a; 10.54(4)c pMd 17.3 15.5 17.1 18.2 25.0 a combined UV-potentiometric titrations at I = 0.16 M (NaCl) and 25 °C; b In-batch acidic spectrophotometric competition at 25 °C, not evaluated at constant I = 0.16 M (NaCl); c ligand-ligand potentiometric competition with EDTA at I = 0.16 M (NaCl) and 25 °C. d pM is defined as -log [Mfree] at [L] = 10 M, [M] = 1 M and pH = 7.4. Charges are omitted for clarity. Figure 6.15. a) pM values vs ionic radius for M3+- H4octox complexes (CN = 8);196 b) Distribution curves for In3+- H4octox complexes, [In3+] = [H4octox] = 8.5  10-4 M. Dashed line in b at physiological pH (7.4). In Table 6.5, the In3+ scavenging ability of H4octox in terms of pIn, has been compared to those of the most relevant In3+ chelators. The high pIn = 25.0 for the In-octox system, being 6.2 units higher than that of one of the current gold standards for 111In-based radiopharmaceuticals, DOTA, and 0.7 units lower than the value for DTPA, characterizes H4octox as a promising scaffold for In3+ complexation in the development of 111In-BFCs (Bifunctional Chelators). In addition, 176 when comparing the pIn values in Table 6.5, of the acylic chelators developed in the past years by our group, only H4octapa exceeds the value of H4octox (by 1.5 units because of the higher basicity of H4octox vs. H4octapa.) Even though the stability constant for [In(octox)]- (log KML = 31.38) is higher than that for [In(octapa)]- (log KML = 26.8(1)) the competition between the protons and metal for the basic sites in the ligand is higher in H4octox. Nevertheless, the thermodynamic stability, while a very important parameter to consider for the development of metal-based drugs, cannot be considered in isolation, as it does not always correlate well with serum stability or in vivo behavior – for [In(octapa)]- or [In(DTPA)]2-, despite higher pIn values vs. [In(neunpa)]-, it was the last showing higher serum stability after one day.209 Table 6.5. pMa values of the most relevant In3+ chelators. pMetal H4octox H4octapa H4neunpa H5decapa DTPA DOTA Bispa pIn3+ 25.0 26.5b 23.6c 23.1c 25.7 b 18.8 b 25.0 d a pM = -log [Mfree] at [L] = 10 M, [M] = 1 M and pH = 7.4. b from ref.24 c from ref.209 d from ref.208 6.3.5 Chelation Enhanced Fluorescence Emission of H4octox 8-Hydroxyquinoline shows very weak fluorescence in aqueous solution due to intermolecular photoinduced proton transfer (PPT) between the hydroxyl group and the nearby quinoline nitrogen as well as intermolecular photoinduced electron transfer (PET) from the amine to the hydroxyquinoline moiety. Once complexed with a metal ion (e.g., Zn2+, Mg2+), both PPT and PET are inhibited and the fluorescence intensifies.224-229 Chelation Enhanced Fluorescence Emission (CHEF) of the 8-hydroxyquinoline chromophore has been extensively studied and used for the design of fluorescent probes;74, 230-233 however, it has not yet been reported in radiometal chelating 177 ligands. Herein, we have studied the CHEF of H4octox in the presence of Y3+, La3+, Lu3+and In3+. As shown in Figure 6.16, at pH 8 in aqueous solution, H4octox presents a really weak fluorescence, as does its complexes with In3+ at the same pH; however, when H4octox is chelated to Y3+, Lu3+, and La3+ the fluorescence is notably enhanced, in particular for [Y(octox)]- whose intensity was a 60-fold increase. This property of H4octox could be useful in the development of fluorescent chemosensors for tracing the intracellular distribution of cold metal bioconjugates without the need for further fluorescence tags, such as fluorescein, and potentially, for the development of bimodal imaging contrast agents. Figure 6.16. Fluorescence spectra of H4octox and its Y3+, La3+, Lu3+, In3+ complexes at pH = 8 (λexc = 350 nm, [H4octox] = [M(octox)]- = 2  10-4M). 178 Figure 6.17 a) Fluorescence spectra of YCl3(5  10-5M), transferrin (1 mg/mL), [Y(octox)]- (5  10-5 M), transferrin (1 mg/mL) + YCl3 (5  10-5 M) and Octox-Y complex (5  10-5 M) + transferrin (1 mg/mL) in PBS (pH 7.4) solution. b) Octox-Y complex stability study against transferrin. Monitoring Chelation Enhanced Fluorescence Emission (CHEF) of H4octox, an in vitro stability study of [Y(octox)]- against transferrin was undertaken using non-radioactive metal ions. The fluorescence emission spectra of YCl3 (5 10-5 M), transferrin (1 mg/mL) and [Y(octox)]- (5  10-5 M) in PBS (pH 7.4) solutions were registered, respectively at first. As shown in Figure 6.17 a, while YCl3 had no emission, transferrin showed very weak emission over the 500-600 nm range (around 550 nm) compared with [Y(octox)]-. When YCl3 was then added to transferrin solution (final [Y]- = 5  10-5 M), the emission spectrum showed little change and was still weak in at 500-600 nm; therefore, when [Y(octox)]- was incubated with transferrin, the intensity of fluorescence emission would decrease if the Y3+ was released or transchelated to transferrin. As presented in Figure 6.17 b, the fluorescence emission of [Y(octox)]- (5  10-5 M) in transferrin (1 mg/mL) was recorded for 5 days. The intensity of the peak emission decreased ~ 2% over 5 days, confirming excellent in vitro stability of [Y(octox)]- against transferrin and suggesting 179 H4octox could be used as a bimodal chelator for 86/90Y labeling. Meanwhile, this experiment also proved that the CHEF of H4octox was useful in the stability study using non-radioactive isotopes and provides more information than that available for previous chelators. 6.3.6 pH and Temperature Dependent [111In(octox)]- Labeling The high thermodynamic stability of [In(octox)]- in solution suggested 111In labeling studies. The radiolabeling was conducted at two different pH values and temperatures, and the results are summarized in Table 6.6. As shown there, pH 5.5 is better to ensure quantitative labeling at lower concentrations vs. pH 4; this can be explained by the solution studies: the presence of protonated species of In-octox up to pH 5, where the species [In(Hoctox)] starts to lose the last proton from one of the –OH phenol groups. Quantitative labeling was achieved at RT within 15 min with the concentration of ligand as low as 10-6 - 10-7 M. This result supports the suitability for a kit-based application. Table 6.6. Radiochemical yields (RCY in %) of the various labeling reactions performed for the 111In-H4octox system. Temperature Ambient temperature 60 °C Ligand concentration pH = 4 pH = 5.5 pH = 4 pH = 5.5 10-4 M 96 97 98 100 10-5 M 91 98 92 100 10-6 M 5 97 92 99 10-7 M 0 80 12 75 10-8 M - 8 - 53 180 6.3.7 Serum Stability A serum competition study was performed to further explore the in vitro stability of [111In(octox)]-; the results are presented and compared with previously reported chelators in Table 6.7. [111In(octox)]- presents a high stability (96.2 ± 0.4%) after 1 h incubation in human serum. This exceeds the 1 h competition values for H4octapa, H2dedpa, DOTA, DTPA, p-NH2−Bn−CHX-A”−DTPA and Bispa chelators. As to the 24 h stability, it is still higher than for most. Although no data are available for many of the In-chelators after 5 days, [111In(octox)]- was 83.6% intact, with stability higher than that of DTPA, and close to that of Bispa, p-NH2−Bn−CHX-A”−DTPA and lower than that of p-NO2−Bn−neunpa. It is noteworthy that serum stability data alone do not prove the superiority of one chelator over the others, as that depends also on kinetics, thermodynamics and the experimental conditions used. For instance, to be noted in this serum competition study, the non-covalent binding of intact 111In complex to the serum protein cannot be discerned and was counted as transchelation. The introduced error may be small and negligable when evaluating the previous hydrophilic chelators. Considering that [111In(octox)]- is a hydrophobic anion and the tris(8-hydroxyquinolinato) indium complex was reported to have a significant noncovalent binding to human serum albumin (HSA),234-235 the error from the non-covalent binding of [111In(octox)]- to HSA might be significant, only underestimating the stability data. 181 Table 6.7. Mouse serum stability challenges performed at ambient temperature.a Complex 1 h stability 24 h stability 5 days stability [111In(octox)]− 96.2 ± 0.4 91.4 ±0.6 83.6 ± 1.4 [111In(dedpa)]+ b 96.1 ± 0.1 19.7 ±1.5 NA [111In(octapa)]− b 93.8 ± 3.6 92.3 ±0.04 NA [111In(DOTA)]− b 89.6 ± 2.1 89.4 ±2.2 NA [111In(DTPA)]2− b 86.5 ± 2.2 88.3 ± 2.2 < 60% e [111In(p-NO2−Bn−neunpa)]− c 97.9 ± 0.3 97.8 ± 0.1 97.8 ± 0.7 [111In(p-NH2−Bn−CHX-A”−DTPA)]2− c 91.8 ± 1.8 89.9 ± 0.6 90.1 ± 0.9 [111In(Bispa)]+ d 88.4 ± 1.2 87.4 ± 0.6 87.4 ± 1.5 a Stability shown as the percentage of intact 111In complex; b from Ref.24 c from Ref.209 d from Ref.208 e from Ref236 6.3.8 In vivo Imaging of [111In(octox)]- The 5-day [111In(octox)]- in vitro stability suggested in vivo SPECT/CT imaging and biodistribution studies in mice. The addition of 111InCl3 solution (3.22 mCi, 10 μL) to H4octox in water (100 μL, 2 mM) produced [111In(octox)]- at a radiochemical yield of 98% within 15 min at room temperature as confirmed by radio-HPLC. The radiocomplex [111In(octox)]- eluted with nearly the same retention time (tR = 6.1 min) as the non-radioactive In-octox complex (Figure 6.18) with only 5 s between the UV- and gamma-traces, a consequence of the sequential arrangement of the detectors. In contrast, the free ligand H4octox alone eluted 1.4 min later (tR = 7.5 min) indicating the more polar nature of the negatively charged complex compared to the ligand. Uncomplexed 111InCl3 alone eluted with the mobile phase front (tR = 1.9 min). From the HPLC of each, radiometal and nonradioactive complex, only one single sharp peak was observed (Figure 6.18), indicating that the two isomers found in solution by NMR have the same polarity, and affect neither the purification nor the in vivo biodistribution. 182 Figure 6.18. The [111In(octox)]- radiocomplex (top, blue) elutes with nearly the same radio-HPLC retention time (tR = 6.1min) as the non-radioactive [In(octox)]- complex (bottom, blue). H4octox (bottom, brown) elutes 1.4 min later and 111InCl3 (top, green) elutes with the mobile phase front. Following intravenous administration, [111In(octox)]- distributed in liver, bile ducts, and gallbladder, ultimately being excreted into the intestinal tract and urinary tract/bladder. Figure 6.19 illustrates longitudinal changes in the complex biodistribution over time. Assessment of the organ-specific time-activity curves for [111In(octox)]- displayed differences in organ uptake and kinetics (Figure 6.20). Standardized uptake values (SUVs) of [111In(octox)]- as a function of time (n=3) showed high liver and gallbladder activity within the first 5 min (Figure 6.20). About 26.0 ± 1.5% of the total [111In(octox)]- complex was rapidly excreted within 60 min through the renal system, as shown by a very quick and high activity increase in the bladder very early on (Figures 6.20 and 6.21). The kidneys, however, did not show measurable activity which could be an advantage over many previously reported chelators. In many peptide conjugates of DOTA, CHX-A\"-DTPA and particularly in DTPA, kidney uptake is observed and it can decrease the imaging quality and minimize nephrotoxicity when using therapeutic isotopes such as 90Y or 177Lu. 163-164, 192, 203, 237 Similarly, 32 ± 2.5% was eliminated through the liver and secreted into the bile (gallbladder) 183 within the first hour post-injection. In the same time, the data show that 40 ± 3.1% of the initial [111In(octox)]- was excreted into the intestinal tract from where it was excreted in the feces within 24 h post-administration (Figures 6.20 and 6.21). This is consistent with the higher lipophilicity of H4octox, unlike most previously reported multi-armed carboxylate or picolinate based chelators which were cleared through the kidneys.20, 24 This could provide a complementary choice in the tracer design and may have a positive impact on the pharmacokinetics or binding properties, especially for small molecules. For instance, studies with a PSMA inhibitor-based 68Ga tracer revealed that using a more lipophilic chelator HBED-CC increased tumor uptake and reduced unspecific binding compared with the DOTA conjugate.114, 147 Similarly, in a very recent study, five different chelators were conjugated to the same RGD peptide and investigated to evaluate the effect of lipophilicity of the chelators on the pharmacokinetics of the bioconjugates.148 This research also reported that those tracers using more lipophilic chelators show higher tumor uptake and tumor/organ ratios despite the increased liver uptake versus those using polar hydrophilic chelators. In our own work, an 111In-labeled trastuzumab bioconjugate incorporating a more lipophilic picolinate based chelator H4octapa showed a markedly higher tumor uptake than did 111In-DOTA-trastuzumab, indicating that even for large bio-vectors such as an antibody, the moderating effect of the chelating moiety on the pharmacokinetic properties may not be negligible.37 Therefore, H4octox provides an important more hydrophobic complement for the currently used chelator library in manipulating the pharmacokinetics of the bioconjugates, considering that almost all of the currently used large trivalent metal ion chelators are polar resulting in fast renal clearance, obviating tumor uptake. Negligible activity accumulated in the bone marrow consistent with minimal dissociation and free 111In3+ (Figure 6.19 and 6.20). In fact, the small amount of activity in the bone measured early 184 is likely just the radioactivity of the radiolabeled complex in the blood stream, based on its fast disappearance. Moreover, no liver accumulation is shown after 5 h, indicating that it has a good metabolic stability in liver with no metal release. Analysis of the latest imaging time points (5 and 24 h) indicates that clearance of this compound occurs mainly through the gastrointestinal tract. This is again typical for lipophilic substances of this size, as for example shown with 99mTc-hepatobiliary iminodiacetic acid (HIDA).238 The final passage through the intestine into the colon 24 h post-injection is well observed in the images (Figure 6.19). Figure 6.19. Top view maximum intensity projections (MIPs) on total body SPECT/CT scans at various time points after administration of [111In(octox)]- complex. 185 0.01 0.1 1 10 100024681012SUV (g/mL)Time (h) Gallbladder Liver Bladder BoneMarrow Figure 6.20. Representative SUV time-activity curves for [111In(octox)]- in mice (n = 3). Biodistribution of [111In(octox)]- The radiolabeled complex is processed through enterohepatic circulation. This means that the complex enters the portal circulation, is taken up in the liver and potentially metabolized, and then returns to the gastrointestinal tract via biliary excretion. The complex leaves the body by excretion through the feces, as it is not reabsorbed in the intestine. This process depends heavily on the physicochemical properties (lipophilicity) of the compound and its alteration by the metabolism. The SPECT images clearly corroborate this process (Figure 6.19). This trend was further confirmed by the results obtained from the biodistribution study (Figure 6.21). Figure 6.21. Biodistribution of [111In(octox)]- after 24 hours post-injection, n = 3. bloodheartliverkidneys lungssmall intestinebrainbladdermusclespleenstomachbonepancreasfeces051015202530%ID / organbloodheartliverkidneys lungssmall intestinebrainbladdermusclespleenstomachbonepancreasfeces0204010055010001450%ID / g tissue(a) (b)186 6.4 Conclusions H4octox incorporates the rigid oxine arm motif into an octadentate ligand for large metal ions and shows great potential for radiopharmaceuticals. The facile economical synthesis of this new chelator allows large scale preparation and wide use. It shows both fast chelation kinetics with metal ions (Y3+, In3+, La3+, Gd3+, Lu3+) as an acyclic chelator, and high thermodynamic stability of the formed complexes resulting from the excellent geometry arrangement of the coordination atoms and rigidity of the formed chelate. The cavity of this preorganized chelator fits best In3+> Lu3+ > Y3+ > Gd3+ > La3+, following a trend of decreasing ionic radii, carrying a major advantage for the development of a versatile radiopharmaceutical or MRI contrast agent. In the 111In radiolabeling experiments, H4octox achieves quantitative radiolabeling within 15 min at ambient conditions (pH 5.5 and room temperature) with ligand concentrations as low as 10-6 M to 10-7 M, and this is significant for developing kit-based radiopharmaceuticals, especially for thermally sensitive biovectors such as antibodies. The in vitro serum stability study together with in vivo SPECT/CT imaging and biodistribution data confirmed the good stability of radiometal-octox-complexes with no persistent kidney uptake or liver accumulation in vivo. Moreover, the ligand is more lipophilic than most of the multi-armed carboxylate or picolinate chelators, therefore it shows gastrointestinal clearance and provides a complementary choice for the design of metal based pharmaceuticals and tuning of their pharmacokinetic properties. The Chelation Enhanced Fluorescence Property was used to confirm an excellent in vitro stability of [Y(octox)]- against transferrin, a unique advantage over the previously reported chelators in “cold” metal based study. The intrinsic fluorescence of [Y(octox)]- and [Lu(octox)]- could be potentially useful in intracellular fluorescence or a bimodal imaging tracer as well. 187 The high thermodynamic stability in solution, the turn-on fluorescence of H4octox with Y3+and Lu3+ and the great [111In(octox)]- stability in vivo, stimulate future studies with 86/90Y and 177Lu. A bifunctional chelator of this promising novel scaffold is currently being synthesized and further in vivo stability and biodistribution studies with 111In, 86/90Y and 177Lu will be reported. 188 Chapter 7: Biological Imaging with Medium-Sensitive Bichromatic Flexible Fluorescent (FlexFluor) Dyes This chapter contains an adaptation of published work, and is reproduced in part from Yang Cao, Xiaozhu Wang, Xiaolei Shi, Suzanne M Clee, Patrick L McGeer, Michael O Wolf, Chris Orvig. Biological Imaging with Medium-Sensitive Bichromatic Flexible Fluorescent (FlexFluor) Dyes, Angew. Chem. Int. Ed, 2017, 56, 15603-15606. 7.1 Introduction Detection and visualization of lipid droplets (LDs) and membrane lipids in an aqueous context has become increasingly biomedically relevant. LDs are dynamic organelles known to play important roles in the storage, transportation and metabolism of triglycerides.239-240 Abnormalities in LDs are associated with diseases such as fatty liver disease (FLD),241 type 2 diabetes,242 atherosclerosis243 and even tumorigenesis and tumor progression.244 Membrane lipids, as essential cellular structural components, are also of great importance in various activities such as cellular signaling and regulation.245 Changes and defects in lipid membranes are known to be associated with many neurological diseases such as Alzheimer’s and multiple sclerosis (MS).246-248 Early detection of abnormalities in LDs and membrane lipids can help with diagnosis and treatment of corresponding diseases. Dynamic fluorescence tracking249 of the movement and accumulation of lipid content in living tissue is therefore important to the understanding of these diseases and development of treatments. In order to visualize lipid contents, numerous fluorescent dyes have been developed with lipid-water contrast, for example BODIPY250 and Nile Red251 dyes have been widely used and commercially available for many years. Most currently used dyes, however, are monochromatic, 189 offering little to no emission color change that can be effectively used to resolve the difference between aqueous and lipid media. Additionally, BODIPY analogues usually have a rather small Stokes shift which gives rise to problematic interference from the photo-excitation source. Herein, a new family of strongly emissive dyes is reported and offer good lipid/water contrast by fluorescing with two different color channels, both of which possess Stokes shifts larger than 100 nm. These bichromatic dyes feature highly tunable chemical structures and fluorescence colors, and can therefore easily be tailored to have different degrees of hydrophilicity and to enable coupling with biological vectors and other functionalities towards specific applications. Figure 7.1. Bichromatic medium-sensitive fluorescent dyes. Scheme 7.1. Synthetic route to three flexfluor dyes. 190 7.2 Experimental 7.2.1 Materials and Methods 1H NMR, 13C NMR, and 31P NMR spectra were recorded on a Bruker Avance 400 (400 MHz) spectrometer. 31P NMR chemical shifts are referenced to external 85% H3PO4, and other chemical shifts are referenced to the residual solvent signals. ESI mass spectra were measured in-house using a Waters LC-MS ESI mass spectrometer. All syntheses were carried out under a dry N2 atmosphere using standard Schlenk techniques. Dry toluene and tetrahydrofuran (THF) were purchased from Aldrich and purified with solvent purification system. The synthesis of Dye A was reported previously.252 Br4T2 (Scheme 7.2) is made according to a known protocol.253 P-Chlorodiphenylphosphine (99%, Sigma-Aldrich), n-butyllithium (1.6 M in hexanes, Sigma-Aldrich), dimesitylfluoroborane (98%, TCI America), zinc bromide (98%, Sigma-Aldrich), 1-chloro-6-iodohexane (96%, Sigma-Aldrich), hydrogen peroxide solution (30% (w/w) in H2O, Sigma-Aldrich), trimethylamine solution (~ 45 wt.% in H2O, Sigma-Aldrich), 4-bromo-N,N-dimethylaniline (97%, Sigma-Aldrich), tetrakis-(triphenylphosphine)palladium(0) (99%, Strem Chemicals) and ethyl octanoate (99%, Sigma-Aldrich) were purchased and used without further purification. 7.2.2 Dye Synthesis All of the dyes listed were designed together with collaborator Dr. Cao Yang, and synthesized by Dr. Cao Yang. Please refer to the published paper254 for the details of synthesis and characterization.) 191 Scheme 7.2. Synthesis of Precursor FlexFluor-C6-Cl. 7.2.3 Solution Studies All dyes were evaluated in ethyl octanoate (EO, non-hydrogen bonding) and methanol (MeOH, strongly hydrogen bonding) at a concentration of 1 x 10-5 M. Methanol was used due to poor solubility of Dye A and Dye C in water. Photo excitation and emission measurements were performed on a PTI QuantaMaster 50 fluorimeter. Water/lipid partition experiments with Dyes A and B were performed using EO (water saturated) as the lipid phase and water (EO saturated) as the aqueous phase. Standard solutions of the dyes (1 x 10-5 M) in both solvents were prepared. The dye solution in EO (1 x 10-5 M) was gently mixed with water (EO saturated) for 5 min and separated using a separation funnel. The EO and water phases were separated and their fluorescence spectra were collected and compared with those of 1 x 10-5 M standard solutions of the dyes in EO and water. 7.2.4 Biological Imaging General: Fluorescence microscopy images were taken with Olympus IX70 Inverted Fluorescence Microscope with an Olympus DP80 dual CCD 12.7-megapixel color camera. A longpass filter (cut-on wavelength of 400 nm, Thorlabs, Inc.) was used. 192 Adipocyte Imaging: Mouse inguinal adipose tissue was dissected, placed overnight in several volumes of 4% (v/v) paraformaldehyde at 4 °C, then overnight in 30% (w/v) sucrose at 4 °C, frozen in Neg50® substrate on dry ice, and kept at -80 °C until sectioning. The sample was then immersed and frozen in VWR Clear Frozen Section Compound® and made into 100 µm thick sections on a Leica CM3050S cryostat at -20 °C and mounted on MillenniaTM 2.0 adhesion (glass) slides. Prior to fluorescent imaging, samples were treated with 1 mg/mL Dye solution in DMSO, incubated for 30 min, and rinsed thoroughly with deionized water. Brain Slice Imaging: Human brain tissues (fixed with 4% paraformaldehyde) of a specific region of human brain with the inferior olivary nucleus were selected from the UBC Brain Bank, and transferred to a 15% buffered sucrose maintenance solution. After 3 days, sections were cut into 30 μm thickness and stored in the maintenance solution before use. Prior to fluorescent imaging, similar to mice adipose tissue, the brain slice samples were mounted on glass slides, treated with 1 mg/mL Dye solution in DMSO, incubated for 1 hour, and rinsed thoroughly with deionized water. 7.3 Results and Discussion The bichromatic fluorescent dye system is based on the shape-shifting structure shown in Figure 7.1. The fluorescent bithiophene core is in equilibrium between the closed and open forms, which under irradiation with UV light give rise to bright blue and yellow emission, respectively.252 The open form is predominant in strong hydrogen bond (HB) donating environments such as water while the closed form is favored in weak HB donors such as lipids. The emissive core as well as the side-chain functionalities (R groups) can be modified to yield different dyes tailored for various 193 applications. Three FlexFluor dyes (Dyes A-C) are shown with hydrophilic or ionic side chains (Scheme 7.1), and two proof-of-concept case studies profiling fluorescent imaging of LDs and neurological structures containing phospholipids. Dye A (Flexfluor-C4), Dye C (Flexfluor-DMA) and Flexfluor-C6-Cl (the precursor to Dye B) can be prepared in one pot from corresponding starting materials. Dye B is synthesized from FlexFluor-C6-I via simple substitution of the reactive iodo group with nucleophilic trimethylamine. FlexFluor-C6-I may also be useful in coupling the stimuli-responsive core with alternate functional groups such as phosphonium-based mitochondrial biological vectors via reaction with phosphines,255-256 taking advantage of the reactivity of the iodo leaving group towards substitution. Figure 7.2. Normalized excitation and emission spectra of (a) Dye A, b) FlexFluor-C6-I, (c) Dye B and (d) Dye C in ethyl octanoate and methanol (1 x 10-5 M). (For Dye A, Dye B and FlexFluorC6-I, lEX (MeOH) = 330 nm, lEM (MeOH) = 540 nm, lEX (EO) = 360 nm, lEM (EO) = 450 nm; for Dye C, lEX (MeOH) = 350 nm, lEM (MeOH) = 660 nm, lEX (EO) = 450 nm, lEM (EO) = 560 nm; methanol used as the strong HB donor solvent due to poor solubility of Dye A in water). 194 Dye A, Dye B and FlexFluor-C6-I all contain the same bithienyl-based stimuli-responsive core, and exhibit similar emission colors in ethyl octanoate (EO) and methanol, representative of a non-HB donating lipid and a strongly HB donating aqueous environment, respectively (Figure 7.2). Dyes A and B, however, are hydrophobic (lipophilic) and hydrophilic, respectively, due to the presence of alkyl and alkylammonium side-chains. These dyes require higher energy photo-excitation and emit lower energy light in methanol compared to in EO. This difference is the result of the equilibrium between open and closed forms of the stimuli-responsive bithienyl core which undergoes structural changes in different environments (Figure 7.1). Strong HB donors such as water and methanol stabilize the open structure, which is yellow emissive, by hydrogen-bonding with the phosphine-oxide moiety and disrupting the Lewis acid-base interaction; weak or non-HB donors such as EO and lipids lack this HB interaction, therefore the blue emissive closed structure, an intramolecular Lewis adduct, is preferred in these environments. Success in red-shifting the emission in water and lipid came from extending the conjugation of the core in Dye C relative to Dye A. Dye C shows analogous environment-dependent two-color emission behaviour (Figure 7.2 d), with a decrease in excitation and emission energy of both the open and closed forms relative to Dye A. This is significant for applications in fluorescence imaging, as it allows deeper penetration of excitation and emission light into tissue which is especially beneficial for in vivo applications. All dyes emit only in two dramatically different color channels corresponding to the open and closed forms, and emission is primarily sensitive to the HB donor strength of the surrounding environment. In contrast, the emission color of most organic dyes with charge transfer (CT) excited states shifts incrementally as the solvent polarity changes. This emission behaviour of FlexFluor dyes can result in excellent image resolution since the emission from lipid regions can be 195 conveniently separated from that of the aqueous background. Moreover, due to the involvement of CT excited states,252 both emission channels have a large difference between excitation and emission maxima, more than 100 nm for the closed forms, and ~ 200 nm for the open forms. This is significant in that many common fluorescent dyes suffer from small Stokes shifts (less than 20 nm difference in wavelength for common commercially available BODIPY dyes),22 which leads to detrimental inner-filter effects and self-quenching even at very low concentrations.257-259 Figure 7.3. Ethyl octanoate (EO)/water partitioning experiment for (a) Dye A and (b) Dye B. A solution of dye (10 mL, 1 x 10-5 M) in water saturated EO was mixed with EO saturated water (10 mL) for five minutes. The EO/water phases were then separated and the emission profile measured using a fluorimeter. Water/lipid partition experiments with Dyes A and B were performed in which the dye was extracted from an EO solution (water saturated) with water (EO saturated). The EO and water phases were separated and their fluorescence spectra were collected and compared with those of 1 x 10-5 M standard solutions of the dyes in EO and water (Figure 7.3). It is evident that Dye A prefers to stay in the lipid-like EO phase (Figure 7.3 a) while Dye B shows stronger affinity 196 toward water (Figure 7.3 b). The observed emission color of Dyes A and B in different media is clearly visually different (Figure 7.4 a). Figure 7.4. (a) Photograph of emission from the organic and water phases of a 1:1 (v/v) equilibrated mixture of ethyl octanoate and water containing Dye A (left) and Dye B (right). (b) Cartoon illustration of adipocyte cells. Optical fluorescence microscopy image of adipose tissue treated with (c) Dye A and (d) Dye B. (1 ms exposure, scale bar = 100 mm.) Images of mice adipose tissue with co-existing Dye A and Dye B in (e) bright field full color and (f) color-separated red channel. (1 ms exposure, scale bar = 100 mm). Based on the encouraging outcomes from the solution studies, proof-of-concept fluorescence imaging experiments of LDs were carried out using inguinal mice adipose tissue. Adipose tissue is mostly composed of adipocytes, cells with sufficiently large LDs to be visible under an optical microscope (Figure 7.4 b). Beyond energy storage, adipocytes play an important role in stabilizing blood glucose levels via protein and hormone secretion. In Type II diabetes, the loss of capacity 197 to store fat in adipocytes results in abnormal accumulation of fat in the muscles, pancreas and liver which leads to further damage to the body. This has the motivation to provide another set of tools to visualize and monitor LDs with clear distinction from other hydrophilic entities. Figure 7.5. Fluorescence images of mice adipose tissue with co-existing Dye A and Dye B in (a) bright field full color and (b) color-separated blue channel (c) color-separated green channel and (d) color-separated red channel. (1 ms exposure, scale bar = 100 um). Adipocyte sections were incubated with 1 mg/mL Dye A in DMSO solution for 30 min, followed by extensive rinsing with deionized water. Images of these cells obtained under a fluorescence microscope show that only the lipid globules (LDs) are blue-emitting under UV excitation (Figure 7.4 c). This indicates that lipophilic Dye A preferentially stays in the lipid 198 droplet with blue emission color, indicative of the presence of predominantly closed form. Adipocyte sections incubated with 1 mg/mL hydrophilic Dye B in DMSO following the same protocol show a different distribution and yellow-green emission color of the dye - the fluorescence image in Figure 7.4 d indicates that Dye B stays in the cytoplasm (periphery of the cell) corresponding to the open form. To show that it is possible to highlight the lipophilic and hydrophilic environment simultaneously using the same excitation light source, samples treated with both Dye A and Dye B were tested (Figure 7.4 e). Not only are the lipid and aqueous environments clearly distinguishable with different colours, they can also be easily seen by separating the color channels using ImageJ software (Red channel shown in Figure 7.4 f, blue and green channels shown in Figure 7.5). Similarly, one should also be able to identify the lipid and aqueous environments using suitable blue bandpass and red long-pass filters, respectively. It is also worth noting that these measurements were performed using the same light source regardless of the emission color, which allows for continuous observation of dynamic behavior with potential applications such as tracking of the interaction of relevant molecules with LDs which is of great importance but still poorly understood.260-261 This approach can also be potentially used for fluorescent tagging to report on distribution information of biomolecules, nanoparticles or other objects of interest and their surrounding environments, which is difficult with conventional dyes. These proof-of-concept imaging experiments show that the fluorophore core is able to indicate hydrophobic or hydrophilic environments by showing drastically different emission colors for different regions. Next aim was to take advantage of the bichromatic nature of the FlexFluor dyes for multicolor fluorescence imaging of biological tissues with a single dye. 199 Figure 7.6. (a) Cartoon illustration showing a section of axon with myelin sheath, and fluorescence image of brain tissue with Dye A showing (b) region with no observable neuron, scale bar = 100 m; (c) area in inferior olivary nucleus rich in neurons, scale bar = 50 m; and (d) zoomed-in image, scale bar = 25 m. Phospholipid-rich neurological tissues were selected to image with the FlexFluor dyes. Abnormalities in phospholipid-containing structures such as cell membranes and the myelin sheath of neurons in nerve tissues are relevant to the diagnosis and study of aging,262 neurodegenerative diseases such as Alzheimer’s and multiple sclerosis (MS).263 The myelin sheath consists of alternating layers of phospholipid membrane and cytoplasm around the axon forming an electrically insulating layer (Figure 7.6 a), and is of great importance for nervous system function. It is proposed that these bichromatic fluorescent dyes can serve as imaging dyes for the structure of myelin sheaths and provide good contrast between the lipid membrane and the aqueous layer as well as the cytoplasm of neurons. Lipophilic Dye A and Dye C were chosen for the fluorescent imaging experiments of the brain samples provided by the Kinsmen Laboratory of Neurological Research. These samples contain a specific part of the medulla oblongata called the inferior olivary nucleus which is rich in neuronal myelin sheath content as the major source of signal input to the 200 cerebellum. They were sliced thinly perpendicular to the axons, exposing the cross section containing the myelin sheath and the cytoplasm within (Figure 7.6 a). Even though rich in lipid, the brain tissue is rather complex and does not offer the same clear structural distinction between lipid and water as in the adipose tissue. For instance, the lipid-rich region with few axons stained with Dye A (Figure 7.6 b) shows a blue-green emission color rather than the blue color in Figure 7.4 c, possibly due to coexisting aqueous and lipid content. Longer wavelength yellow-orange emission is observed in the predominantly aqueous environment of the axon cytoplasm, while brighter yellow-green emission is observed from the surrounding myelin sheath which has greater lipid content (Figure 7.6 c-d). Dye C also shows outstanding fluorescence intensity in the myelin sheath region, even the multiple membrane layer structure of the myelin sheath becomes clearly visible under a standard optical fluorescence microscope (Figure 7.7). Highly specific recognition of the myelin sheath without the use of immunofluorescent tags, and demonstrate good color and intensity contrast between the myelin sheaths of axons vs. internal cytoplasm and external lipid contents was acheived. A significant advantage of the FlexFluor dyes over conventional dyes is that typically more than one fluorescent protein or dye is needed to piece together a complete picture of this area of brain, using advanced and sophisticated imaging techniques such as confocal microscopy.264 (a) (b) Figure 7.7. Fluorescence image of brain tissue with Dye A showing (a) region rich in axons and (b) lipid-rich region with few axons. 201 7.4 Conclusion In summary, a new class of easy-to-synthesize bichromatic fluorescence dyes are reported; they can indicate the hydrophilic/hydrophobic property of the environment by their emission colors. The emission color and hydrophilicity of the dye can be customized according to specific application with various conjugation length of the core and different side-groups, respectively. The environment dependent fluorescence of these dyes was investigated in solution and applied in fluorescence imaging of adipocytes and myelin sheath structures. The full potential of FlexFluor dyes are being explored and anticipated to be powerful tools in multiple applications such as dynamic tracking of lipid content and visualization of cross-membrane molecule transport. Based on the high selectivity of myelin sheath coloration and good lipid/water contrast, these dyes can be potentially used to visualize degenerative changes such as loss of neurons in MS diseased tissues. These FlexFluor dyes are now being tested on MS diseased spinal cord samples which show a pathological demyelination and decreased lipid/water ratio. These results will be reported in due course. 202 Chapter 8: Other Work and Future Studies 8.1 Further Enhancing the H2dedpa and H4octapa Technology Scheme 8.1. H2dedpa and H4octapa derivatives. Several other bifunctional derivatives of H2dedpa and H4octapa have also been synthesized and require further investigation. These are shown in Scheme 8.1. Since earlier studies showed the stability of the Ga3+ complexes formed with H2dedpa derivatives decreased when the backbone ethylenediamine nitrogen atoms were converted from secondary to tertiary amines, an interesting question is to what extent was this effect due to the conversion of sp2 nitrogen to sp3 nitrogen versus the change in sterics. Thus, a comprehensive study with (OH)2dedpa and (NH2)2dedpa (Scheme 8.1) will help to answer this question as they are less bulky than the previously reported benzyl amine functionalized derivatives. In addition, both of these suggested chelators could be conjugated to biovectors for further application research. N3octapa, a H4octapa derivative with an azide group, was synthesized to conjugate a Cu (Ⅰ) metal ion to alkyne-functionalized biotin via a Cu-catalyzed clicked chemistry reaction; however, 203 this approach failed because N3octapa is a good chelator for copper ions. This barrier can be avoided by carrying out the “click” reaction after the radiolabeling procedure, meaning N3octapa may be more useful when applied using the “chase” method; an in situ reaction involving pre-targeted antibodies. This could be tested in the future. 8.2 Further Enhancing the H2hox Technology Scheme 8.2. p-NO2-hox and p-NCS-hox. For the new H2hox platform, an NO2 derivative, p-NO2-hox, has been synthesized that will be converted to an NCS derivative, p-NCS-hox, for bioconjugation (Scheme 8.2). Similarly, other possible bifunctional derivatives of H2hox and H2C3hox have been summarized in Scheme 8.3. While H2hox is more stable than is H2dedpa with Ga3+, the functionalization of the backbone amine should still be carried out with care to avoid any reduction in on chelate stability. 204 Scheme 8.3. Other possible bifunctional derivatives of H2hox and H2C3hox. The major challenge in the synthesis currently is the purification of the product considering the poor solubility of this new group of chelators. Adding a protecting group on the hydroxyquinoline arm might be helpful to solve the problem. Acetyl and benzenesulfonyl groups have been tested and some new chelators have been successfully synthesized; however, there are still problems. Acetyl protection on phenol groups is really sensitive and not really durable in many reaction conditions, while benzenesulfonyl is too robust to remove without using hard conditions. 8.3 Further Enhancing the H4octox Technology For bifunctional H4octox preparation, p-NO2-hox can be used as a starting material as shown in Scheme 8.4. The synthesis of p-NCS-octox may be easier than p-NCS-hox, since the solubility of H4octox derivatives are better than H2hox derivatives. Other bioconjugation groups, for example -NHS, -NH2, could also be prepared to expand the application. The clickable derivatives such as azide or alkyne-functionalized H4octox could be useful as well in the future especially in the “Chasing” or “pretargeting method”. Generally, the functional groups would be appropriately 205 adapted to the nature of the biovectors of interest and a proper spacer could be incorporated if necessary to manipulate the pharmacokinetics of the tracer. Scheme 8.4. p-NCS-octox preparation. 8.4 Other Interesting Chelators Scheme 8.5. Other interesting chelators. A few other novel chelators have been prepared or are currently being synthesized, summarised in Scheme 8.5. Properties such as the change of donor atoms, spatial arrangement, rigidity, and the denticity of the chelators are adapted in attempts to make strong coordination 206 complexes with relevant metal ions that best suit the chelator coordination properties and vice versa. Therefore, comparative studies on these chelators will be informative for the design of new chelators in the future. 207 Reference 1. Lippard, S. J., J. Am. Chem. Soc. 2010, 132, 14689-14693. 2. Storr, T.; Thompson, K. H.; Orvig, C., Chem. Soc. Rev. 2006, 35, 534-544. 3. Xing, Y.; Zhao, J.; Conti, P. S.; Chen, K., Theranostics 2014, 4 (3), 290-306. 4. Lauterbur, P. C., Nature 1973, 242, 190-191. 5. Zhou, Z.; Lu, Z., Wires. Nanomed. Nanobi. 2013, 5, 1-18. 6. Gale, E. M.; Atanasova, I. P.; Blasi, F.; Ay, I.; Caravan, P., J. Am. Chem. Soc. 2015, 137, 15548-15557. 7. Li, L.; Jiang, W.; Luo, K.; Song, H.; Lan, F.; Wu, Y.; Gu, Z., Theranostics 2013, 3, 595-615. 8. Boros, E.; Gale, E. M.; Caravan, P., Dalton Trans. 2015, 44, 4804-4818. 9. Lauffer, R. B., Chem. Rev. 1987, 87, 901-927. 10. Vogl, T. J.; Hamm, B.; Schnell, B.; McMahon, C.; Branding, G.; Lissner, J.; Wolf, K. J., J. Magn. Reson. Imaging 1993, 3, 51-58. 11. Loving, G. S.; Mukherjee, S.; Caravan, P., J. Am. Chem. Soc. 2013, 135, 4620-4623. 12. Paefgen, V.; Doleschel, D.; Kiessling, F., Front Pharmacol. 2015, 6, 197. 13. Lusic, H.; Grinstaff, M. W., Chem. Rev. 2012, 113, 1641-1666. 14. Månsson, J. http://www.med.lu.se/bioimaging_center/modalities/pet_spect_ct/basic_prin-ciples (accessed May 12th, 2018). 15. Bandoli, G.; Barreca, D.; Gasparotto, A.; Seraglia, R.; Tondello, E.; Devi, A.; Fischer, R. A.; Winter, M.; Fois, E.; Gamba, A., Phys. Chem. Chem. Phys. 2009, 11, 5998-6007. 16. Bartholoma, M. D.; Louie, A. S.; Valliant, J. F.; Zubieta, J., Chem. Rev. 2010, 110 (5), 2903-2920. 17. Wikipedia contributors, \"Nuclear medicine,\" Wikipedia, The Free Encyclopedia, https://en.wikipedia.org/w/index.php?title=Nuclear_medicine&oldid=889631214 (accessed April 22, 2019). 208 18. Welch, M. J.; Redvanly, C. S., Handbook of Radiopharmaceuticals: Radiochemistry and Applications. NY: J. Wiley: New York, 2003. 19. Brechbiel, M. W., Q. J. Nucl. Med. Mol. Im. 2008, 52, 166-173. 20. Price, E. W.; Orvig, C., Chem. Soc. Rev. 2014, 43, 260-290. 21. Milenic, D. E.; Brady, E. D.; Brechbiel, M. W., Nat. Rev. Drug Discov. 2004, 3, 488-499. 22. Notni, J.; Hermann, P.; Havlickova, J.; Kotek, J.; Kubicek, V.; Plutnar, J.; Loktionova, N.; Riss, P. J.; Rosch, F.; Lukes, I., Chem. Eur. J. 2010, 16, 7174-7185. 23. Bailey, G. A.; Price, E. W.; Zeglis, B. M.; Ferreira, C. L.; Boros, E.; Lacasse, M. J.; Patrick, B. O.; Lewis, J. S.; Adam, M. J.; Orvig, C., Inorg. Chem. 2012, 51, 12575-12589. 24. Price, E. W.; Cawthray, J. F.; Bailey, G. A.; Ferreira, C. L.; Boros, E.; Adam, M. J.; Orvig, C., J. Am. Chem. Soc. 2012, 134, 8670-8683. 25. Boros, E.; Ferreira, C. L.; Cawthray, J. F.; Price, E. W.; Patrick, B. O.; Wester, D. W.; Adam, M. J.; Orvig, C., J. Am. Chem. Soc. 2010, 132, 15726-15733. 26. Boros, E.; Ferreira, C. L.; Cawthray, J. F.; Price, E. W.; Wester, D. W.; Adam, M. J.; Orvig, C., J. Labelled. Compd. Rad. 2011, 54, S321-S321. 27. Mathias, C. J.; Wang, S.; Lee, R. J.; Waters, D. J.; Low, P. S.; Green, M. A., J. Nucl. Med. 1996, 37, 1003-1008. 28. Yao, Z.; Zhang, M.; Axworthy, D. B.; Wong, K. J.; Garmestani, K.; Park, L.; Park, C. W.; Mallett, R. W.; Theodore, L. J.; Yau, E. K.; Waldmann, T. A.; Brechbiel, M. W.; Paik, C. H.; Pastan, I.; Carrasquillo, J. A., Cancer Res. 2002, 62, 5755-5760. 29. Boros, E.; Ferreira, C. L.; Yapp, D. T. T.; Gill, R. K.; Price, E. W.; Adam, M. J.; Orvig, C., Nucl. Med. Biol. 2012, 39, 785-794. 30. Hicke, B. J.; Stephens, A. W.; Gould, T.; Chang, Y. F.; Lynott, C. K.; Heil, J.; Borkowski, S.; Hilger, C. S.; Cook, G.; Warren, S.; Schmidt, P. G., J. Nucl. Med. 2006, 47, 668-678. 31. Lucignani, G., Eur. J. Nucl. Med. Mol. Imaging 2009, 36, 869-874. 209 32. Burns, H. D.; Gibson, R. E.; Eckelman, W. C., Nuclear Imaging in Drug Discovery, Development and Approval. Birkhauser Boston: Cambridge, MA, USA, 1993; p 334. 33. Volkert, W. A.; Hoffman, T. J., Chem. Rev. 1999, 99, 2269-2292. 34. Liu, S., Chem. Soc. Rev. 2004, 33, 445-461. 35. Ramogida, C. F.; Orvig, C., Chem. Commun. 2013, 49, 4720-4739. 36. Boros, E.; Marquez, B. V.; Ikotun, O. F.; Lapi, S. E.; Ferreira, C. L., Coordination Chemistry And Ligand Design In The Development Of Metal Based Radiopharmaceuticals in Ligand Design. In Medicinal Inorganic Chemistry, Storr, T., Ed. John Wiley & Sons Ltd.: Chichester, U.K., 2014; pp 47-79. 37. Price, E. W.; Zeglis, B. M.; Cawthray, J. F.; Ramogida, C. F.; Ramos, N.; Lewis, J. S.; Adam, M. J.; Orvig, C., J. Am. Chem. Soc. 2013, 135, 12707-12721. 38. Ruser, G.; Ritter, W.; Maecke, H. R., Bioconjug. Chem. 1990, 1, 345-9. 39. Andre, J. P.; Maecke, H. R.; Zehnder, M.; Macko, L.; Akyel, K. G., Chem. Commun. 1998, 12, 1301-1302. 40. Stimmel, J. B.; Kull, F. C., Jr., Nucl. Med. Biol. 1998, 25, 117-125. 41. Milenic, D. E.; Garmestani, K.; Chappell, L. L.; Dadachova, E.; Yordanov, A.; Ma, D.; Schlom, J.; Brechbiel, M. W., Nucl. Med. Biol. 2002, 29, 431-442. 42. Camera, L.; Kinuya, S.; Garmestani, K.; Brechbiel, M. W.; Wu, C. C.; Pai, L. H.; Mcmurry, T. J.; Gansow, O. A.; Pastan, I.; Paik, C. H.; Carrasquillo, J. A., Eur. J. Nucl. Med. 1994, 21, 640-646. 43. Cooper, M. S.; Ma, M. T.; Sunassee, K.; Shaw, K. P.; Williams, J. D.; Paul, R. L.; Donnelly, P. S.; Blower, P. J., Bioconjug. Chem. 2012, 23, 1029-1239. 44. Schneider, D. W.; Heitner, T.; Alicke, B.; Light, D. R.; McLean, K.; Satozawa, N.; Parry, G.; Yoo, J.; Lewis, J. S.; Parry, R., J. Nucl. Med. 2009, 50, 435-443. 45. Price, E. W.; Zeglis, B. M.; Lewis, J. S.; Adam, M. J.; Orvig, C., Dalton Trans. 2014, 43, 119-31. 46. Cabbiness, D. K.; Margerum, D. W., J. Am. Chem. Soc. 1970, 92, 2151-2153. 47. Price, E. W.; Cawthray, J. F.; Adam, M. J.; Orvig, C., Dalton Trans. 2014, 43, 7176-7190. 210 48. Kalman, F. K.; Vegh, A.; Regueiro-Figueroa, M.; Toth, E.; Platas-Iglesias, C.; Tircso, G., Inorg. Chem. 2015, 54, 2345-2356. 49. Boros, E.; Ferreira, C. L.; Yapp, D. T. T.; Gill, R. K.; Price, E. W.; Adam, M. J.; Orvig, C., Nucl. Med. Biol. 2012, 39, 785-794. 50. Platas-Iglesias, C.; Mato-Iglesias, M.; Djanashvili, K.; Muller, R. N.; Elst, L. V.; Peters, J. A.; de Blas, A.; Rodriguez-Blas, T., Chem. Eur. J. 2004, 10, 3579-3590. 51. Boros, E.; Cawthray, J. F.; Ferreira, C. L.; Patrick, B. O.; Adam, M. J.; Orvig, C., Inorg. Chem. 2012, 51, 6279-6284. 52. Boros, E.; Ferreira, C. L.; Patrick, B. O.; Adam, M. J.; Orvig, C., Nucl. Med. Biol. 2011, 38, 1165-1174. 53. Ferreiros-Martinez, R.; Esteban-Gomez, D.; Platas-Iglesias, C.; de Blas, A.; Rodriguez-Blas, T., Dalton Trans. 2008, 0, 5754-5765. 54. Price, E. W.; Zeglis, B. M.; Cawthray, J. F.; Lewis, J. S.; Adam, M. J.; Orvig, C., Inorg. Chem. 2014, 53, 10412-10431. 55. Ramogida, C. F.; Cawthray, J. F.; Boros, E.; Ferreira, C. L.; Patrick, B. O.; Adam, M. J.; Orvig, C., Inorg. Chem. 2015, 54, 2017-2031. 56. Saeed, M.; Wendland, M. F.; Tomei, E.; Rocklage, S. M.; Quay, S. C.; Moseley, M. E.; Wolfe, C.; Higgins, C. B., Radiology 1989, 173, 763-767. 57. Wikström, M., Acta Radiologica. Supplementum 1991, 379, 1-30. 58. Nilsson, S.; Wikstrom, G.; Ericsson, A.; Wikstrom, M.; Oksendal, A.; Waldenstrom, A.; Hemmingsson, A., Acta Radiol. 1996, 37, 27-35. 59. Dupas, B.; Bach-Gansmo, T.; Meflah, K.; Muller, R. N., Invest. Radiol. 1998, 33, 268-272. 60. Vander Elst, L.; Roch, A.; Gillis, P.; Laurent, S.; Botteman, F.; Bulte, J. W.; Muller, R. N., Magn. Reson. Med. 2002, 47, 1121-1130. 61. Forsterova, M.; Jandurova, Z.; Marques, F.; Gano, L.; Lubal, P.; Vanek, J.; Hermann, P.; Santos, I., J. Inorg. Biochem. 2008, 102, 1531-1540. 211 62. Amoroso, A. J.; Fallis, I. A.; Pope, S. J. A., Coordin. Chem. Rev. 2017, 340, 198-219. 63. Kalman, F. K.; Kiraly, R.; Brucher, E., Eur. J. Inorg. Chem. 2008, 2008, 4719-4727. 64. Bishayee, A.; Rao, D. V.; Srivastava, S. C.; Bouchet, L. G.; Bolch, W. E.; Howell, R. W., J. Nucl. Med. 2000, 41, 2043-2050. 65. Chakraborty, S.; Unni, P. R.; Venkatesh, M.; Pillai, M. R. A., Appl. Radiat. Isot. 2002, 57, 295-301. 66. Mathew, B.; Chakraborty, S.; Das, T.; Sarma, H. D.; Banerjee, S.; Samuel, G.; Venkatesh, M.; Pillai, M. R., Appl. Radiat. Isot. 2004, 60, 635-642. 67. Lee, T. M.; Cheng, T. H.; Ou, M. H.; Chang, C. A.; Liu, G. C.; Wang, Y. M., Magn. Reson. Chem. 2004, 42, 329-336. 68. Lundgren, S.; Lutsenko, S.; Jonsson, C.; Moberg, C., Org. Lett. 2003, 5, 3663-3665. 69. SAINT. Version 8.34A. Bruker AXS Inc., M., Wisconsin, USA. SAINT. Version 8.34A. Bruker AXS Inc., Madison, Wisconsin, USA. (1997-2013), (1997-2013). 70. TWINABS.V.2012/1 Bruker AXS Inc., M., Wisconsin (2012) TWINABS.V.2012/1 Bruker AXS Inc., Madison, Wisconsin (2012). 71. Sheldrick, G. M., Acta Crystallogr A Found Adv 2015, 71, 3-8. 72. Spek, A. L., Acta Crystallogr C Struct Chem 2015, 71, 9-18. 73. Dolomanov, O. V.; Bourhis, L. J.; Gildea, R. J.; Howard, J. A. K.; Puschmann, H., J. Appl. Crystallogr. 2009, 42, 339-341. 74. Zhang, H.; Han, L. F.; Zachariasse, K. A.; Jiang, Y. B., Org. Lett. 2005, 7, 4217-4220. 75. Tao, J.; Perdew, J. P.; Staroverov, V. N.; Scuseria, G. E., Phys. Rev. Lett. 2003, 91, 146401. 76. Tomasi, J.; Mennucci, B.; Cammi, R., Chem. Rev. 2005, 105, 2999-3093. 77. Dolg, M.; Stoll, H.; Savin, A.; Preuss, H., Theoretica Chimica Acta 1989, 75, 173-194. 78. Dolg, M.; Stoll, H.; Preuss, H., Theor. Chem. Acc. 1993, 85, 441-450. 79. Esteban‐Gómez, D.; De Blas, A.; Rodríguez‐Blas, T.; Helm, L.; Platas‐Iglesias, C., ChemPhysChem 2012, 13, 3640-3650. 212 80. Weekes, D. M.; Ramogida, C. F.; Jaraquemada-Pelaez, M. G.; Patrick, B. O.; Apte, C.; Kostelnik, T. I.; Cawthray, J. F.; Murphy, L.; Orvig, C., Inorg. Chem. 2016, 55, 12544-12558. 81. Gran, G., Analyst 1952, 77, 661-671. 82. Nurchi, V. M.; Crisponi, G.; Crespo-Alonso, M.; Lachowicz, J. I.; Szewczuk, Z.; Cooper, G. J., Dalton Trans. 2013, 42, 6161-6170. 83. Nurchi, V. M.; Lachowicz, J. I.; Crisponi, G.; Murgia, S.; Arca, M.; Pintus, A.; Gans, P.; Niclos-Gutierrez, J.; Dominguez-Martin, A.; Castineiras, A.; Remelli, M.; Szewczuk, Z.; Lis, T., Dalton Trans. 2011, 40, 5984-5998. 84. Paul, M. A.; Long, F. A., Chem. Rev. 1957, 57, 1-45. 85. Gans, P.; Sabatini, A.; Vacca, A., Talanta 1996, 43, 1739-1753. 86. Gans, P.; Sabatini, A.; Vacca, A., Ann. Chim. 1999, 89, 45-49. 87. Baes, C. F. Jr.; Mesmer, R. E., The hydrolysis of cations. In The Hydrolysis of Cations, John Wiley & Sons, New York: 1976. 88. Chatterton, N.; Gateau, C.; Mazzanti, M.; Pecaut, J.; Borel, A.; Helm, L.; Merbach, A., Dalton Trans. 2005, 0, 1129-1135. 89. Lacoste, R. G.; Marttel, A. E., Inorg. Chem. 1964, 3, 881-884. 90. Ernestova, M.; Jedinakova-Krizova, W.; Vanura, P., Croat. Chem. Acta 2004, 77, 633-637. 91. Popov, K.; Ronkkomaki, H.; Lajunen, L. H. J., Pure Appl. Chem. 2001, 73, 1641-1677. 92. Harris, W. R.; Carrano, C. J.; Raymond, K. N., J. Am. Chem. Soc. 1979, 101, 2213-2214. 93. Green, M. A., J. Nucl. Med. 1990, 31, 1641-1645. 94. Blower, P. J., Dalton Trans. 2015, 44, 4819-4844. 95. Baranyai, Z.; Reich, D.; Vagner, A.; Weineisen, M.; Toth, I.; Wester, H. J.; Notni, J., Dalton Trans. 2015, 44, 11137-11146. 96. Ferreira, C. L.; Lamsa, E.; Woods, M.; Duan, Y.; Fernando, P.; Bensimon, C.; Kordos, M.; Guenther, K.; Jurek, P.; Kiefer, G. E., Bioconjug. Chem. 2010, 21, 531-536. 213 97. Notni, J.; Reich, D.; Maltsev, O. V.; Kapp, T. G.; Steiger, K.; Hoffmann, F.; Esposito, I.; Weichert, W.; Kessler, H.; Wester, H.-J., J. Nucl. Med. 2017, 58, 671-677. 98. Young, J. D.; Abbate, V.; Imberti, C.; Meszaros, L. K.; Ma, M. T.; Terry, S. Y. A.; Hider, R. C.; Mullen, G. E.; Blower, P. J., J. Nucl. Med. 2017, 58 (8), 1270-1277. 99. Tsionou, M. I.; Knapp, C. E.; Foley, C. A.; Munteanu, C. R.; Cakebread, A.; Imberti, C.; Eykyn, T. R.; Young, J. D.; Paterson, B. M.; Blower, P. J.; Ma, M. T., RSC Adv. 2017, 7, 49586-49599. 100. Bianchi, A.; Calabi, L.; Giorgi, C.; Losi, P.; Mariani, P.; Paoli, P.; Rossi, P.; Valtancoli, B.; Virtuani, M., J. Chem. Soc. Dalton. 2000, 0, 697-705. 101. Kubicek, V.; Havlickova, J.; Kotek, J.; Tircso, G.; Hermann, P.; Toth, E.; Lukes, I., Inorg. Chem. 2010, 49, 10960-10969. 102. Zhernosekov, K. P.; Filosofov, D. V.; Baum, R. P.; Aschoff, P.; Bihl, H.; Razbash, A. A.; Jahn, M.; Jennewein, M.; Rosch, F., J. Nucl. Med. 2007, 48, 1741-1748. 103. Eppard, E.; Wuttke, M.; Nicodemus, P. L.; Rosch, F., J. Nucl. Med. 2014, 55, 1023-1028. 104. Velikyan, I.; Beyer, G. J.; Langstrom, B., Bioconjug. Chem. 2004, 15, 554-560. 105. Ma, M. T.; Blower, P. J., CHAPTER 8 Chelators for Diagnostic Molecular Imaging with Radioisotopes of Copper, Gallium and Zirconium. In Metal Chelation in Medicine, The Royal Society of Chemistry: 2017; pp 260-312. 106. Velikyan, I.; Maecke, H.; Langstrom, B., Bioconjug. Chem. 2008, 19, 569-573. 107. de Sa, A.; Matias, A. A.; Prata, M. I.; Geraldes, C. F.; Ferreira, P. M.; Andre, J. P., Bioorg. Med. Chem. Lett. 2010, 20, 7345-7348. 108. Prata, M. I.; Santos, A. C.; Geraldes, C. F.; de Lima, J. J., J. Inorg. Biochem. 2000, 79, 359-363. 109. Simecek, J.; Schulz, M.; Notni, J.; Plutnar, J.; Kubicek, V.; Havlickova, J.; Hermann, P., Inorg. Chem. 2012, 51, 577-590. 110. Waldron, B. P.; Parker, D.; Burchardt, C.; Yufit, D. S.; Zimny, M.; Roesch, F., Chem. Commun. 2013, 49, 579-581. 214 111. Nock, B. A.; Kaloudi, A.; Nagel, J.; Sinnes, J. P.; Roesch, F.; Maina, T., Dalton Trans. 2017, 46, 14584-14590. 112. Parker, D.; Waldron, B. P.; Yufit, D. S., Dalton Trans. 2013, 42, 8001-8008. 113. Ferreira, C. L.; Yapp, D. T.; Mandel, D.; Gill, R. K.; Boros, E.; Wong, M. Q.; Jurek, P.; Kiefer, G. E., Bioconjug. Chem. 2012, 23, 2239-2246. 114. Eder, M.; Schafer, M.; Bauder-Wust, U.; Hull, W. E.; Wangler, C.; Mier, W.; Haberkorn, U.; Eisenhut, M., Bioconjug. Chem. 2012, 23, 688-697. 115. Afshar-Oromieh, A.; Malcher, A.; Eder, M.; Eisenhut, M.; Linhart, H. G.; Hadaschik, B. A.; Holland-Letz, T.; Giesel, F. L.; Kratochwil, C.; Haufe, S.; Haberkorn, U.; Zechmann, C. M., Eur. J. Nucl. Med. Mol. Imaging 2013, 40, 486-495. 116. Motekaitis, R. J.; Sun, Y.; Martell, A. E.; Welch, M. J., Inorg. Chem. 1991, 30, 2737-2740. 117. Bollinger, J. E.; Mague, J. T.; Oconnor, C. J.; Banks, W. A.; Roundhill, D. M., J. Chem. Soc. Dalton. 1995, 0, 1677-1688. 118. Farkas, E.; Nagel, J.; Waldron, B. P.; Parker, D.; Toth, I.; Brucher, E.; Rosch, F.; Baranyai, Z., Chem. Eur. J. 2017, 23, 10358-10371. 119. Baranyai, Z.; Uggeri, F.; Maiocchi, A.; Giovenzana, G. B.; Cavallotti, C.; Takacs, A.; Toth, I.; Banyai, I.; Benyei, A.; Brucher, E.; Aime, S., Eur. J. Inorg. Chem. 2013, 2013, 147-162. 120. Vágner, A.; D'Alessandria, C.; Gambino, G.; Schwaiger, M.; Aime, S.; Maiocchi, A.; Tóth, I.; Baranyai, Z.; Tei , L., ChemistrySelect 2016, 1, 163-171. 121. Fellner, M.; Riss, P.; Loktionova, N.; Zhernosekov, K.; Thews, O.; Geraldes, C. F. G. C.; Kovacs, Z.; Lukes, I.; Rosch, F., Radiochimica. Acta 2011, 99, 43-51. 122. Lin, K.-S.; Pan, J.; Amouroux, G.; Turashvili, G.; Mesak, F.; Hundal-Jabal, N.; Pourghiasian, M.; Lau, J.; Jenni, S.; Aparicio, S.; Bénard, F., Cancer Res. 2015, 75, 387-393. 123. Yanai, T.; Tew, D. P.; Handy, N. C., Chem. Phys. Lett. 2004, 393, 51-57. 124. Scalmani, G.; Frisch, M. J., J. Chem. Phys. 2010, 132, 114110. 125. Wadt, W. R.; Hay, P. J., J. Chem. Phys. 1985, 82, 284-298. 215 126. Adamo, C.; Barone, V., J. Chem. Phys. 1999, 110, 6158-6170. 127. Ramogida, C. F.; Schindler, D.; Schneider, C.; Tan, Y. L. K.; Huh, S.; Ferreira, C. L.; Adam, M. J.; Orvig, C., RSC Adv. 2016, 6, 103763-103773. 128. Enyedy, E. A.; Domotor, O.; Varga, E.; Kiss, T.; Trondl, R.; Hartinger, C. G.; Keppler, B. K., J. Inorg. Biochem. 2012, 117, 189-97. 129. Hata, T.; Uno, T., Bull. Chem. Soc. Jpn. 1972, 45, 477-481. 130. Hata, T.; Uno, T., Bull. Chem. Soc. Jpn. 1972, 45, 2497-2503. 131. Li, L. H.; Xu, B., Tetrahedron 2008, 64, 10986-10995. 132. Mishra, A.; Periasamy, N.; Patankar, M. P.; Narasimhan, K. L., Dyes Pigments 2005, 66, 89-97. 133. Prachayasittikul, V.; Prachayasittikul, S.; Ruchirawat, S.; Prachayasittikul, V., Drug. Des. Devel. Ther. 2013, 7, 1157-1178. 134. Gonzalez-Vera, J. A.; Lukovic, E.; Imperiali, B., J. Org. Chem. 2009, 74, 7309-7314. 135. Xie, J. T.; Ning, Z. J.; Tian, H., Tetrahedron Lett. 2005, 46, 8559-8562. 136. Xia, Y. X.; Chen, J. F.; Choppin, G. R., Talanta 1996, 43, 2073-2081. 137. Heppeler, A.; Froidevaux, S.; Macke, H. R.; Jermann, E.; Behe, M.; Powell, P.; Hennig, M., Chem. Eur. J. 1999, 5, 1974-1981. 138. Clarke, E. T.; Martell, A. E., Inorg. Chim. Acta 1991, 190, 37-46. 139. Motekaitis, R. J.; Martell, A. E.; Welch, M. J., Inorg. Chem. 1990, 29, 1463-1467. 140. Evers, A.; Hancock, R. D.; Martell, A. E.; Motekaitis, R. J., Inorg. Chem. 1989, 28, 2189-2195. 141. Liu, Z. D.; Khodr, H. H.; Liu, D. Y.; Lu, S. L.; Hider, R. C., J. Med. Chem. 1999, 42, 4814-4823. 142. Crisponi, G.; Nurchi, V. M., Chapter 2 - Chelating Agents as Therapeutic Compounds—Basic Principles. In Chelation Therapy in the Treatment of Metal Intoxication, Academic Press: Boston, 2016; pp 35-61. 143. Clarke, E. T.; Martell, A. E., Inorg. Chim. Acta 1991, 181, 273-280. 144. Meyer, G. J.; Macke, H.; Schuhmacher, J.; Knapp, W. H.; Hofmann, M., Eur. J. Nucl. Med. Mol. Imaging 2004, 31, 1097-1104. 216 145. Harris, W. R.; Pecoraro, V. L., Biochemistry 1983, 22, 292-299. 146. Sephton, R. G.; Hodgson, G. S.; De Abrew, S.; Harris, A. W., J. Nucl. Med. 1978, 19, 930-935. 147. Eder, M.; Neels, O.; Muller, M.; Bauder-Wust, U.; Remde, Y.; Schafer, M.; Hennrich, U.; Eisenhut, M.; Afshar-Oromieh, A.; Haberkorn, U.; Kopka, K., Pharmaceuticals (Basel) 2014, 7, 779-796. 148. Sarkar, S.; Bhatt, N.; Ha, Y. S.; Huynh, P. T.; Soni, N.; Lee, W.; Lee, Y. J.; Kim, J. Y.; Pandya, D. N.; An, G. I.; Lee, K. C.; Chang, Y.; Yoo, J., J. Med. Chem. 2018, 61, 385-395. 149. Kim, Y. S.; Wang, F.; Liu, S., World J. Hepatol. 2010, 2, 21-31. 150. Liu, S., Dalton Trans. 2007, 0, 1183-1193. 151. Tsang, B. W.; Mathias, C. J.; Fanwick, P. E.; Green, M. A., J. Med. Chem. 1994, 37, 4400-4406. 152. Thews, O.; Zimny, M.; Eppard, E.; Piel, M.; Bausbacher, N.; Nagel, V.; Rasch, F., Mol. Imaging. Biol. 2014, 16, 802-812. 153. Tarkia, M.; Saraste, A.; Saanijoki, T.; Oikonen, V.; Vahasilta, T.; Strandberg, M.; Stark, C.; Tolvanen, T.; Teras, M.; Savunen, T.; Green, M. A.; Knuuti, J.; Roivainen, A., Nucl. Med. Biol. 2012, 39, 715-723. 154. Hsiao, Y. M.; Mathias, C. J.; Wey, S. P.; Fanwick, P. E.; Green, M. A., Nucl. Med. Biol. 2009, 36, 39-45. 155. van Tilborg, G. A.; Vucic, E.; Strijkers, G. J.; Cormode, D. P.; Mani, V.; Skajaa, T.; Reutelingsperger, C. P.; Fayad, Z. A.; Mulder, W. J.; Nicolay, K., Bioconjug. Chem. 2010, 21, 1794-1803. 156. Thorp-Greenwood, F. L.; Coogan, M. P., Dalton Trans. 2011, 40, 6129-6143. 157. Bullok, K. E.; Dyszlewski, M.; Prior, J. L.; Pica, C. M.; Sharma, V.; Piwnica-Worms, D., Bioconjug. Chem. 2002, 13, 1226-1237. 158. Xu, H.; Baidoo, K.; Gunn, A. J.; Boswell, C. A.; Milenic, D. E.; Choyke, P. L.; Brechbiel, M. W., J. Med. Chem. 2007, 50, 4759-4765. 159. Esteves, T.; Xavier, C.; Gama, S.; Mendes, F.; Raposinho, P. D.; Marques, F.; Paulo, A.; Pessoa, J. C.; Rino, J.; Viola, G.; Santos, I., Org. Biomol. Chem. 2010, 8, 4104-4116. 217 160. Bhushan, K. R.; Misra, P.; Liu, F.; Mathur, S.; Lenkinski, R. E.; Frangioni, J. V., J. Am. Chem. Soc. 2008, 130, 17648-17649. 161. Ramogida, C. F.; Murphy, L.; Cawthray, J. F.; Ross, J. D.; Adam, M. J.; Orvig, C., J. Inorg. Biochem. 2016, 162, 253-262. 162. Uppal, R.; Ciesienski, K. L.; Chonde, D. B.; Loving, G. S.; Caravan, P., J. Am. Chem. Soc. 2012, 134, 10799-10802. 163. Forster, G. J.; Engelbach, M. J.; Brockmann, J. J.; Reber, H. J.; Buchholz, H. G.; Macke, H. R.; Rosch, F. R.; Herzog, H. R.; Bartenstein, P. R., Eur. J. Nucl. Med. 2001, 28, 1743-1750. 164. Camera, L.; Kinuya, S.; Garmestani, K.; Wu, C.; Brechbiel, M. W.; Pai, L. H.; McMurry, T. J.; Gansow, O. A.; Pastan, I.; Paik, C. H.; et al., J. Nucl. Med. 1994, 35, 882-889. 165. Chong, H. S.; Garmestani, K.; Ma, D.; Milenic, D. E.; Overstreet, T.; Brechbiel, M. W., J. Med. Chem. 2002, 45, 3458-3464. 166. Brechbiel, M. W.; Gansow, O. A.; Pippin, C. G.; Rogers, R. D.; Planalp, R. P., Inorg. Chem. 1996, 35, 6343-6348. 167. Stimmel, J. B.; Stockstill, M. E.; Kull, F. C., Bioconjugate. Chem. 2002, 6, 219-225. 168. Wang, X.; Jaraquemada-Pelaez, M. G.; Cao, Y.; Pan, J.; Lin, K. S.; Patrick, B. O.; Orvig, C. Inorg. Chem. 2019, 58, 2275-2285. 169. De Puey, E. G., J. Nucl. Med. 1994, 35, 17N-20N. 170. Kiat, H.; Berman, D. S.; Maddahi, J., Radiol. Clin. North. Am. 1993, 31, 795-815. 171. Nunn, A. D., Semin. Nucl. Med. 1990, 20, 111-118. 172. Tarkia, M.; Saraste, A.; Saanijoki, T.; Oikonen, V.; Vahasilta, T.; Strandberg, M.; Stark, C.; Tolvanen, T.; Teras, M.; Savunen, T.; Green, M. A.; Knuuti, J.; Roivainen, A., Nucl. Med. Biol. 2012, 39, 715-723. 173. Thews, O.; Zimny, M.; Eppard, E.; Piel, M.; Bausbacher, N.; Nagel, V.; Rosch, F., Mol. Imaging. Biol. 2014, 16, 802-812. 218 174. Green, M. A.; Welch, M. J.; Mathias, C. J.; Fox, K. A.; Knabb, R. M.; Huffman, J. C., J. Nucl. Med. 1985, 26, 170-180. 175. Green, M. A.; Mathias, C. J.; Neumann, W. L.; Fanwick, P. E.; Janik, M.; Deutsch, E. A., J. Nucl. Med. 1993, 34, 228-233. 176. Kung, H. F.; Liu, B. L.; Mankoff, D.; Kung, M. P.; Billings, J. J.; Francesconi, L.; Alavi, A., J. Nucl. Med. 1990, 31, 1635-1640. 177. Plossl, K.; Chandra, R.; Qu, W.; Lieberman, B. P.; Kung, M. P.; Zhou, R.; Huang, B.; Kung, H. F., Nucl. Med. Biol. 2008, 35, 83-90. 178. Hsiao, Y. M.; Mathias, C. J.; Wey, S. P.; Fanwick, P. E.; Green, M. A., Nucl. Med. Biol. 2009, 36, 39-45. 179. Ramogida, C. F.; Schindler, D.; Schneider, C.; Tan, Y. L. K.; Huh, S.; Ferreira, C. L.; Adam, M. J.; Orvig, C., RSC Adv. 2016, 6 (105), 103763-103773. 180. Krause, L.; Herbst-Irmer, R.; Sheldrick, G. M.; Stalke, D., J. Appl. Crystallogr. 2015, 48, 3-10. 181. Mighell, A. D.; Santoro, A.; Donnay, J. D. H., Acta Crystallogr. Sect. B 1971, B 27, 1837. 182. Ibers, J. A.; Hamilton, W. C., Acta Crystallogr. 1964, 17, 781. 183. Janssen, T., Acta Crystallogr. 1996, A52, C573. 184. Creagh, D. C. H., J. H., International Tables for Crystallography. Kluwer Academic Publishers: Boston, 1992; Vol. C, Table 4.2.4.3, 200-206. 185. Sheldrick, G. M., Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 2015, 71, 3-8. 186. Mouralian, C.; Buss, J. L.; Stranix, B.; Chin, J.; Ponka, P., Biochem. Pharmacol. 2005, 71, 214-222. 187. Vázquez, K.; Espinosa-Bustos, C.; Soto-Delgado, J.; Tapia, R. A.; Varela, J.; Birriel, E.; Segura, R.; Pizarro, J.; Cerecetto, H.; González, M.; Paulino, M.; Salas, C. O., RSC Adv. 2015, 5, 65153-65166. 188. Hussain, H.; Specht, S.; Sarite, S. R.; Hoerauf, A.; Krohn, K., Eur. J. Med. Chem. 2012, 54, 936-942. 219 189. Ramogida, C. F.; Boros, E.; Patrick, B. O.; Zeisler, S. K.; Kumlin, J.; Adam, M. J.; Schaffer, P.; Orvig, C., Dalton Trans. 2016, 45, 13082-13090. 190. Green, M. A.; Huffman, J. C., J. Nucl. Med. 1988, 29, 417-420. 191. Roca, M.; de Vries, E. F.; Jamar, F.; Israel, O.; Signore, A., Eur. J. Nucl. Med. Mol. Imaging 2010, 37, 835-841. 192. Wadas, T. J.; Wong, E. H.; Weisman, G. R.; Anderson, C. J., Chem. Rev. 2010, 110, 2858-2902. 193. Hancock, R. D., J. Chem. Educ. 1992, 69, 615-621. 194. Lippard, S. J., Current research topics in bioinorganic chemistry. Wiley: New York, 1973. 195. Hancock, R. D., Acc. Chem. Res. 1990, 23, 253-257. 196. Shannon, R. D., Acta Crystallogr. Sect. A 1976, 32, 751-767. 197. Dale, A. V.; Pandya, D. N.; Kim, J. Y.; Lee, H.; Ha, Y. S.; Bhatt, N.; Kim, J.; Seo, J. J.; Lee, W.; Kim, S. H.; Yoon, Y. R.; An, G. I.; Yoo, J., ACS Med. Chem. Lett. 2013, 4, 927-931. 198. Boswell, C. A.; Sun, X.; Niu, W.; Weisman, G. R.; Wong, E. H.; Rheingold, A. L.; Anderson, C. J., J. Med. Chem. 2004, 47, 1465-1474. 199. Holland, J. P.; Williamson, M. J.; Lewis, J. S., Mol. Imaging. 2010, 9, 1-20. 200. Lubberink, M.; Tolmachev, V.; Widstrom, C.; Bruskin, A.; Lundqvist, H.; Westlin, J. E., J. Nucl. Med. 2002, 43, 1391-1397. 201. Pauwels, S.; Barone, R.; Walrand, S.; Borson-Chazot, F.; Valkema, R.; Kvols, L. K.; Krenning, E. P.; Jamar, F., J. Nucl. Med. 2005, 46 Suppl 1, 92S-8S. 202. Banerjee, S.; Pillai, M. R.; Knapp, F. F., Chem. Rev. 2015, 115, 2934-2974. 203. Carrasquillo, J. A.; White, J. D.; Paik, C. H.; Raubitschek, A.; Le, N.; Rotman, M.; Brechbiel, M. W.; Gansow, O. A.; Top, L. E.; Perentesis, P.; Reynolds, J. C.; Nelson, D. L.; Waldmann, T. A., J. Nucl. Med. 1999, 40, 268-276. 204. Port, M.; Idee, J. M.; Medina, C.; Robic, C.; Sabatou, M.; Corot, C., Biometals 2008, 21, 469-490. 205. Zhang, L.; Liu, R.; Peng, H.; Li, P.; Xu, Z.; Whittaker, A. K., Nanoscale 2016, 8, 10491-10510. 220 206. Tircso, G.; Benyo, E. T.; Suh, E. H.; Jurek, P.; Kiefer, G. E.; Sherry, A. D.; Kovacs, Z., Bioconjugate. Chem. 2009, 20, 565-575. 207. Baranyai, Z.; Uggeri, F.; Giovenzana, G. B.; Benyei, A.; Brucher, E.; Aime, S., Chem. Eur. J. 2009, 15, 1696-1705. 208. Comba, P.; Jermilova, U.; Orvig, C.; Patrick, B. O.; Ramogida, C. F.; Ruck, K.; Schneider, C.; Starke, M., Chem. Eur. J. 2017, 23, 15945-15956. 209. Spreckelmeyer, S.; Ramogida, C. F.; Rousseau, J.; Arane, K.; Bratanovic, I.; Colpo, N.; Jermilova, U.; Dias, G. M.; Dude, I.; Jaraquemada-Pelaez, M. G.; Benard, F.; Schaffer, P.; Orvig, C., Bioconjugate. Chem. 2017, 28, 2145-2159. 210. Tircso, G.; Kovacs, Z.; Sherry, A. D., Inorg. Chem. 2006, 45, 9269-9280. 211. Van Nostrand, D.; Abreu, S. H.; Callaghan, J. J.; Atkins, F. B.; Stoops, H. C.; Savory, C. G., Radiology 1988, 167, 495-498. 212. Spek, A. L., Acta Crystallogr., Sect. C: Cryst. Struct. Commun. 2015, 71, 9-18. 213. Cromer, D. T. W., J. T., \"International Tables for X-ray Crystallography\". 1974; Vol. IV. 214. Creagh, D. C. M., W. J., International Tables for Crystallography. Kluwer Academic Publishers: Boston, 1992; Vol. Vol C, Table 4.2.6.8, 219-222. 215. Stephens, P. J.; Devlin, F. J.; Chabalowski, C. F.; Frisch, M. J., J. Phys. Chem. 1994, 98, 11623-11627. 216. Becke, A. D., J. Chem. Phys. 1993, 98, 5648-5652. 217. Caricato, M., J. Chem. Theory Comput. 2012, 8, 5081-5091. 218. Hay, P. J.; Wadt, W. R., J. Chem. Phys. 1985, 82, 299-310. 219. Hay, P. J.; Wadt, W. R., J. Chem. Phys. 1985, 82, 270-283. 220. Jaraquemada-Pelaez, M. G.; Wang, X.; Clough, T. J.; Cao, Y.; Choudhary, N.; Emler, K.; Patrick, B. O.; Orvig, C., Dalton Trans. 2017, 46, 14647-14658. 221 221. Nurchi, V. M. M.; Jaraquemada-Pelaez, M. d. G.; Lachowicz, J. I. I.; Zoroddu, M. A.; Peana, M. F.; Dominguez-Martin, A.; Choquesillo-Lazarte, D.; Remelli, M.; Szewczuk, Z.; Crisponi, G., New J. Chem. 2018, 42, 8021-8034. 222. Schwarzenbach, G.; Gut, R.; Anderegg, G., Helv. Chim. Acta 1954, 37, 937-957. 223. Tabata, M., Conditional Constants, The IUPAC Stability Constants Database, Academic Software. 1993. 224. Bardez, E.; Devol, I.; Larrey, B.; Valeur, B., J. Phys. Chem. B 1997, 101, 7786-7793. 225. Goldman, M.; Wehry, E. L., Anal. Chem. 1970, 42, 1178. 226. Prodi, L.; Bargossi, C.; Montalti, M.; Zaccheroni, N.; Su, N.; Bradshaw, J. S.; Izatt, R. M.; Savage, P. B., J. Am. Chem. Soc. 2000, 122, 6769-6770. 227. Prodi, L.; Montalti, M.; Zaccheroni, N.; Bradshaw, J. S.; Izatt, R. M.; Savage, P. B., Tetrahedron Lett. 2001, 42, 2941-2944. 228. Segura, M.; Bricoli, B.; Casnati, A.; Munoz, E. M.; Sansone, F.; Ungaro, R.; Vicent, C., J. Org. Chem. 2003, 68, 6296-6303. 229. Bronson, R. T.; Montalti, M.; Prodi, L.; Zaccheroni, N.; Lamb, R. D.; Dalley, N. K.; Izatt, R. M.; Bradshaw, J. S.; Savage, P. B., Tetrahedron 2004, 60, 11139-11144. 230. Rahier, R.; Noiriel, A.; Abousalham, A., Anal. Chem. 2016, 88, 666-674. 231. Jiang, X. H.; Wang, B. D.; Yang, Z. Y.; Liu, Y. C.; Li, T. R.; Liu, Z. C., Inorg. Chem. Commun. 2011, 14, 1224-1227. 232. Moon, S. Y.; Cha, N. R.; Kim, Y. H.; Chang, S. K., J. Nucl. Med. 2004, 69 (1), 181-183. 233. Farruggia, G.; Iotti, S.; Prodi, L.; Montalti, M.; Zaccheroni, N.; Savage, P. B.; Trapani, V.; Sale, P.; Wolf, F. I., J. Am. Chem. Soc. 2006, 128, 344-350. 234. Enyedy, E. A.; Domotor, O.; Bali, K.; Hetenyi, A.; Tuccinardi, T.; Keppler, B. K., J. Biol. Inorg. Chem. 2015, 20, 77-88. 235. Abou-Zied, O. K.; Al-Lawatia, N.; Elstner, M.; Steinbrecher, T. B., J. Phys. Chem. B 2013, 117 (4), 1062-1074. 222 236. Paik, C. H.; Herman, D. E.; Eckelman, W. C.; Reba, R. C., J. Radioanal. Chem. 1980, 57 (2), 553-564. 237. Wei, L.; Zhang, X.; Gallazzi, F.; Miao, Y.; Jin, X.; Brechbiel, M. W.; Xu, H.; Clifford, T.; Welch, M. J.; Lewis, J. S.; Quinn, T. P., Nucl. Med. Biol. 2009, 36, 345-354. 238. Lambie, H.; Cook, A. M.; Scarsbrook, A. F.; Lodge, J. P.; Robinson, P. J.; Chowdhury, F. U., Clin. Radiol. 2011, 66, 1094-1105. 239. Martin, S.; Parton, R. G., Nat. Rev. Mol. Cell. Biol. 2006, 7, 373-378. 240. Zehmer, J. K.; Huang, Y.; Peng, G.; Pu, J.; Anderson, R. G.; Liu, P., Proteomics 2009, 9, 914-921. 241. Fon Tacer, K.; Rozman, D., J. Lipids 2011, 2011, 783976. 242. Guilherme, A.; Virbasius, J. V.; Puri, V.; Czech, M. P., Nat. Rev. Mol. Cell. Biol. 2008, 9, 367-377. 243. Maxfield, F. R.; Tabas, I., Nature 2005, 438, 612-621. 244. Rohrig, F.; Schulze, A., Nat. Rev. Cancer 2016, 16, 732-749. 245. Hannun, Y. A.; Obeid, L. M., Nat. Rev. Mol. Cell. Biol. 2008, 9, 139-150. 246. Lauwers, E.; Goodchild, R.; Verstreken, P., Neuron 2016, 90, 11-25. 247. Shaharabani, R.; Ram-On, M.; Avinery, R.; Aharoni, R.; Arnon, R.; Talmon, Y.; Beck, R., J. Am. Chem. Soc. 2016, 138 (37), 12159-12165. 248. Aufschnaiter, A.; Kohler, V.; Diessl, J.; Peselj, C.; Carmona-Gutierrez, D.; Keller, W.; Buttner, S., Cell Tissue Res. 2017, 367, 125-140. 249. Yoon, Y.; Lee, P. J.; Kurilova, S.; Cho, W., Nat. Chem. 2011, 3, 868-874. 250. Karolin, J.; Johansson, L. B. A.; Strandberg, L.; Ny, T., J. Am. Chem. Soc. 1994, 116, 7801-7806. 251. Greenspan, P.; Mayer, E. P.; Fowler, S. D., J. Cell Biol. 1985, 100, 965-973. 252. Cao, Y.; Nagle, J. K.; Wolf, M. O.; Patrick, B. O., J. Am. Chem. Soc. 2015, 137 (15), 4888-4891. 253. Beaujuge, P. M.; Tsao, H. N.; Hansen, M. R.; Amb, C. M.; Risko, C.; Subbiah, J.; Choudhury, K. R.; Mavrinskiy, A.; Pisula, W.; Bredas, J. L.; So, F.; Mullen, K.; Reynolds, J. R., J. Am. Chem. Soc. 2012, 134, 8944-8957. 223 254. Cao, Y.; Wang, X.; Shi, X.; Clee, S. M.; McGeer, P. L.; Wolf, M. O.; Orvig, C., Angew. Chem. Int. Ed. 2017, 56, 15603-15606. 255. Fulda, S.; Galluzzi, L.; Kroemer, G., Nat. Rev. Drug Discov. 2010, 9, 447-464. 256. Murphy, M. P., Biochim. Biophys. Acta 2008, 1777, 1028-1031. 257. Moriarty, R. D.; Martin, A.; Adamson, K.; O'Reilly, E.; Mollard, P.; Forster, R. J.; Keyes, T. E., J. Microsc. 2014, 253, 204-218. 258. Ziessel, R.; Ulrich, G.; Harriman, A., New J. Chem. 2007, 31, 496-501. 259. Loudet, A.; Burgess, K., Chem. Rev. 2007, 107, 4891-932. 260. Schrul, B.; Kopito, R. R., Nat. Cell. Biol. 2016, 18, 740-751. 261. Saliba, A. E.; Vonkova, I.; Gavin, A. C., Nat. Rev. Mol. Cell. Biol. 2015, 16, 753-761. 262. Safaiyan, S.; Kannaiyan, N.; Snaidero, N.; Brioschi, S.; Biber, K.; Yona, S.; Edinger, A. L.; Jung, S.; Rossner, M. J.; Simons, M., Nat. Neurosci. 2016, 19, 995-998. 263. Romanelli, E.; Merkler, D.; Mezydlo, A.; Weil, M. T.; Weber, M. S.; Nikic, I.; Potz, S.; Meinl, E.; Matznick, F. E.; Kreutzfeldt, M.; Ghanem, A.; Conzelmann, K. K.; Metz, I.; Bruck, W.; Routh, M.; Simons, M.; Bishop, D.; Misgeld, T.; Kerschensteiner, M., Nat. Commun. 2016, 7, 13275. 264. Micu, I.; Ridsdale, A.; Zhang, L.; Woulfe, J.; McClintock, J.; Brantner, C. A.; Andrews, S. B.; Stys, P. K., Nat. Med. 2007, 13, 874-879. 224 Appendices Appendix A : Supplementary Data for Chapter 2 Table A.1. 1H NMR chemical shifts δ (ppm) and variations of chemical shifts Δδ (ppm) of a set of solutions at the same ligand concentration [H4octapa] = 1.03 x 10-3 M and different acidities, at 25 °C, I = 0.16 M NaCl. H0 or pH H10 H8 H7 H3 H4 H5 -0.11 3.46 3.87 4.44 7.98 8.02 7.64 0.08 3.51 3.92 4.49 8.04 8.08 7.7 0.33 3.59 3.98 4.57 8.13 8.18 7.79 0.69 3.6 3.98 4.58 8.14 8.19 7.8 0.83 3.6 3.98 4.58 8.15 8.21 7.81 0.92 3.62 3.99 4.6 8.17 8.23 7.84 1.02 3.6 3.97 4.58 8.16 8.22 7.82 1.13 3.6 3.96 4.58 8.17 8.23 7.83 1.29 3.59 3.94 4.57 8.17 8.24 7.84 1.45 3.58 3.92 4.57 8.17 8.24 7.84 1.62 3.58 3.9 4.56 8.16 8.24 7.84 1.81 3.57 3.88 4.55 8.17 8.25 7.85 2.00 3.56 3.86 4.55 8.16 8.25 7.85 H0 or pH ΔδH10 ΔδH8 ΔδH7 ΔδH3 ΔδH4 ΔδH5 0.08 0.05 0.05 0.05 0.06 0.06 0.06 0.33 0.08 0.06 0.08 0.09 0.1 0.09 0.69 0.01 0 0.01 0.01 0.01 0.01 0.83 0 0 0 0.01 0.02 0.01 0.92 0.02 0.01 0.02 0.02 0.02 0.03 1.02 -0.02 -0.02 -0.02 -0.01 -0.01 -0.02 1.13 0 -0.01 0 0.01 0.01 0.01 1.29 -0.01 -0.02 -0.01 0 0.01 0.01 1.45 -0.01 -0.02 0 0 0 0 1.62 0 -0.02 -0.01 -0.01 0 0 1.81 -0.01 -0.02 -0.01 0.01 0.01 0.01 2.00 -0.01 -0.02 0 -0.01 0 0 225 Table A.2. Crystallographic data for H4octapa and its La(III) complex. H4octapa·4HCl·2H2O [La(octapa)(H2O)2]- Formula C20H18N4O10Cl4 C40H56.3La2.5N8Na0.5O26.15 Formula weight 628.28 1426.39 Crystal system Monoclinic Triclinic Space group P 21/c (#14) P -1 (#2) Lattice type primitive primitive Lattice parameters a = 9.8590(11) Å b = 6.1687(8) Å c = 22.747(2) Å α = 90° β = 101.898(4)° γ = 90° a = 8.703(3) Å b = 16.076(6) Å c = 20.536(7) Å α = 82.406(5)o β = 84.822(5)o γ = 90.092(4)o Lattice volume 1353.7(3) Å3 2835.9(17) Å3 Z 2 2 Dcalc’d 1.541 g/cm3 1.670 g/cm3 F(000) 652.00 1419.00 μ (MoKα) 4.97 cm-1 19.44 cm-1 Diffractometer Bruker X8 APEX II Bruker APEX DUO Radiation MoK ( = 0.71073 Å) graphite monochromated Mo-K ( = 0.71073 Å) Data Images 1912 exposures @ 30.0 seconds 1827 exposures @ 30.0 seconds Detector Position 40.19 mm 39.97 mm 2max 55.7o 50.6o Reflections collected Total: 14239 Total: 28327 Corrections Unique: 3187 (Rint = 0.042) Absorption (Tmin = 0.836, Tmax= 0.990) Lorentz-polarization Unique: 8861 (Rint = 0.040) Absorption (Tmin = 0.726, Tmax= 0.943) Lorentz-polarization Structure Solution Direct Methods (XT) Direct Methods (XT) 226 Refinement Full-matrix least-squares on F2 Full-matrix least-squares on F2 Function Minimized w (Fo2 - Fc2)2  w (Fo2 - Fc2)2 Least Squares Weights w=1/(2(Fo2)+(0.0590P)2 + 0.7697P) w=1/(2(Fo2)+(0.1002P)2+ 69.4212P) Anomalous Dispersion All non-hydrogen atoms All non-hydrogen atoms No.Observations (I>0.00 (I)) 3187 8861 No. Variables 196 734 Reflection/Parameter Ratio 16.26 12.07 Residuals (refined on F2, all data): R1; wR2 0.056; 0.111 0.091; 0.213 Goodness of Fit Indicator 1.03 1.08 No. Observations (I>2.00 (I)) 2561 7402 Residuals (refined on F2): R1; wR2 0.042; 0.104 0.079; 0.206 Max Shift/Error in Final Cycle 0.00 0.00 Maximum peak in Final Diff. Map 0.59 e-/Å3 2.96 e-/Å3 Minimum peak in Final Diff. Map -0.28 e-/Å3 -2.28 e-/Å3 Figure A.1. Representative spectra of the UV-Vis spectrophotometric titration of [H4octapa] = 1.28 x 10-4 M at 25 °C, path length = 1 cm. 200 225 250 275 3000.00.51.01.52.0H -0.21AbsorbanceWavelength (nm)H -1.24200 225 250 275 3000.00.51.01.52.0H -0.21AbsorbanceWavelength (nm)pH 0.65200 225 250 275 3000.00.51.01.52.0pH 0.65AbsorbanceWavelength (nm)pH 1.88227 Figure A.2. Representative spectra of the Sm3+-H4octapa system. a) In batch UV spectrophotometric experiments, [H4octapa] = [Sm3+] = 1.51 x 10-4 M, path length = 1cm; b) and c) Combined potentiometric-spectrophotometric titrations, [H4octapa] = 5.56 x 10-4 M, [Sm3+] = 5.52 x 10-4 M, path length = 0.2 cm, at 25 °C and I = 0.16 M NaCl. Figure A.3. a) and b) Representative spectra of the in batch UV spectrophotometric experiments for the Dy3+-H4octapa system, [H4octapa] = 1.38 x 10-4 M, [Dy3+] = 1.37 x 10-4 M, path length = 1 cm, at 25 °C and I = 0.16 M NaCl. 240 250 260 270 280 290 3000.00.51.01.52.0AbsorbanceWavelength (nm)H0 0.06pH 1.47240 260 280 3000.00.51.01.5pH 4.84AbsorbanceWavelength (nm)pH 2.10240 260 280 3000.00.51.01.5pH 11.16AbsorbanceWavelength (nm)pH 4.17240 250 260 270 280 290 3000.00.51.01.5pH 1.34AbsorbanceWavelength (nm)pH 0.21240 250 260 270 280 290 3000.00.51.01.5pH 4.05AbsorbanceWavelength (nm)pH 1.34a) b) c) a) b) 228 (a) (b) Figure A.4. Optimized structure of (a) the [Dy(octapa)(H2O)]-2H2O anion and (b) the [Yb(octapa)(H2O)]-2H2O with labels on selected atoms. (a) (b) Figure A.5. MEP mapping of (a) [Dy(octapa)(H2O)]-2H2O anion and (b) [Yb(octapa)(H2O)]-2H2O anion, where red = negative, blue = positive, representing a maximum potential of 0.200 au and a minimum of -0.200 au, mapped onto electron density isosurface of 0.002 Å-3. All hydrogen atoms have been omitted for clarity. 229 Appendix B : Supplementary Data for Chapter 3 Table B.1. Crystallographic Data for the H2hox and [Ga(hox)][ClO4] Structures. H2hox [Ga(hox)][ClO4] formula C22H22N4O2 C24H23N5O6GaCl formula weight 374.43 582.64 crystal system monoclinic monoclinic crystal dimensions (mm) 0.32×0.22×0.07 0.10×0.15×0.39 space group P21/c C 2/c (#15) a/ Å 24.582(2) 12.8572(7) b/ Å 8.1598(7) 33.0580(17) c/ Å 9.2725(8) 12.7024(7) α 90° 90° β 93.629(5)° 116.307(2)° γ 90° 90° lattice volume 1856.2(3) 4839.8(5) Å3 Z 4 8 Dcalc’d 1.340 g/cm3 1.599 g/cm3 μ (Mo-Kα) 0.088 cm-1 12.99 cm-1 residuals (refined on F2): R1[a]; wR2[b] 0.0587; 0.1355 0.064; 0.122 [a] 𝑹𝟏 = [∑||𝑭𝒐| − |𝑭𝒄||]/ ∑|𝑭𝒐|. [b] 𝒘𝑹𝟐 = [∑ 𝒘(𝑭𝒐𝟐 − 𝑭𝒄𝟐)𝟐/ ∑ 𝒘𝑭𝒐𝟒]𝟏𝟐⁄ 230 Appendix C : Supplementary Data for Chapter 6 Table C.1. Crystallographic data for the La-octox structure. C104H92La3N16O26 .6(H2O) Formula C104H104La3N16O32 Formula weight 2506.76 Crystal system Monoclinic Crystal dimensions (mm) 0.25 × 0.1 × 0.08 Space group P 21/c a/ Å 14.5237(8) b/ Å 23.9029(14) c/ Å 38.344(2) α 90° β 90.844(4)° γ 90° Lattice volume 13310.1(13) Å3 Z 4 Dcalc’d 1.251 g/cm3 μ (Mo-Kα) 10.16 cm-1 Residuals (refined on F2): R1a; wR2b 0.0661; 0.1742 a) R1 = ∑ ||Fo| - |Fc|| / ∑ |Fo|, b) wR2 = [∑ ( w (Fo 2 – Fc 2)2 )/ ∑ w(Fo2)2 ] 1/2 "@en ; edm:hasType "Thesis/Dissertation"@en ; vivo:dateIssued "2019-09"@en ; edm:isShownAt "10.14288/1.0378502"@en ; dcterms:language "eng"@en ; ns0:degreeDiscipline "Chemistry"@en ; edm:provider "Vancouver : University of British Columbia Library"@en ; dcterms:publisher "University of British Columbia"@en ; dcterms:rights "Attribution-NonCommercial-NoDerivatives 4.0 International"@* ; ns0:rightsURI "http://creativecommons.org/licenses/by-nc-nd/4.0/"@* ; ns0:scholarLevel "Graduate"@en ; dcterms:title "New chelators for radiopharmaceutical chemistry"@en ; dcterms:type "Text"@en ; ns0:identifierURI "http://hdl.handle.net/2429/69977"@en .