Pdf4 Complex I and the Aryl Palladium Precursor II Underwent Sequential Single Electron Abstraction from Aryl Pd(II) Complex

Design, synthesis, methodology development, and evaluation of PET imaging agents targeting cancer and CNS disorders

By Gengyang Yuan

B.S. in Chemical Engineering and Technology, Zhejiang University of Technology M.S. in Pharmaceutical Engineering, Zhejiang University

A dissertation submitted to

The Faculty of the College of Science of Northeastern University in partial fulfillment of the requirements for the degree of Doctor of Philosophy

April 21, 2017

Dissertation directed by

Michael P. Pollastri Associate Professor and Chair of Chemistry and Chemical Biology

Co-directed by

Neil Vasdev Adjunct Associate Professor of Chemsitry and Chemical Biology Associate Professor of Radiology, Massachusetts General Hospital and Harvard Medical School

Dedication

To my parents Zhijun and Yongmian and my wife Ran and daughter Isabella

Acknowledgements

This dissertation would not have been possible without the support, guidance and encouragement of numerous people who have helped me along the way.

First and foremost, I would like to thank Northeastern University and the Department of

Chemistry and Chemical Biology for supporting me to pursue my doctoral study. I would like to especially thank my current advisor Professor Michael Pollastri for helping me out when I needed it the most. I appreciate you for taking me into your group and giving me full support to finish my thesis projects. I also especially thank my co-advisor Professor Neil Vasdev for taking me into his group at Mass. General Hospital & Harvard Medical School and teaching me the

PET radiochemistry and PET imaging. I could not image how I could accomplish this work without your help. I also got a lot of help from Dr. Lori Ferrins and enjoyed the stay with other group members in Pollastri’s laboratory. I want to further extend my gratitude to my thesis committee: Professor Mary Jo Ondrechen and Professor Ke Zhang. I appreciate your time commitment and value your expertise.

I would also like to thank my former advisor Professor Graham Jones for initially taking me to his group and providing me the opportunity to collaborate with Professor Neil Vasdev’s group.

Thank you for encouraging me to grow as an independent researcher. I recognize the help I got from Dr. Sara Sadler, Dr. Nadeesha Ranasinghe and Dr. Chiara Chapman at the beginning of my research as well as Dr. Enrico M. Mongeau, Dr. Meaghan Fallano, Katie Hargrove, Tanner

Jankins, Chris Patrick and Nick Gedeon for their support along my work at Jones’ laboratory. I am also indebted to Professor Michail Sitkovsky and his students Dr. Stephen Hatfield and

Phaethon Philbrook for performing the immunoassays for me.

iii

I would like to especially thank Professor Mary Jo Ondrechen for allowing me to explore the field of molecular modeling and encouraging me to achieve my goals. I also want to thank

Jenifer Winters and Timothy Coulther for teaching me how to use YASARA and Glide to carry out my research.

I would like to acknowledge other people who also gave me tremendous help during my research at MGH, besides Professor Neil Vasdev. I would like to thank Professor Steven H. Liang for guiding me through in this process. I also want to give special thanks to Dr. Benjamin Rotstein for his encouragement and support as well as Dr. Lu Wang, Dr. Lee Collier and Ran Cheng for their support.

Last, but certainly not least, I would like to thank my family and friends for their support during this long journey. I could not have done it without their help and encouragement. I especially want to thank my parents, wife and daughter for their endless love and support.

Abstract of Dissertation

As a non-invasive imaging technique, positron emission tomography (PET) is capable of in vivo quantification of biochemical and pharmacological progress via radiolabeled molecular probes.

This dissertation highlights the development of a novel PET radiotracer for imaging α-amino-3- hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptor, 18F labeling methodology for

18 [ F]arylCF2H functionality, and design and synthesis of PET tracers targeting the endocanabinoid system in the brain (i.e. fatty acid amide hydrolase (FAAH) and monoacylglycerol lipase (MGL)) and the adenosine A2A receptor (A2AR) in the immune system.

Chapter 2 describes the radiosynthesis of 2-(1-(3-[18F]fluorophenyl)-2-oxo-5-(pyrimidin-2-yl)-

1,2-dihydropyridin-3yl)benzonitrile ([18F]2-11), which shows similar biodistribution results in mice as that of its [11C]2-11 isotopologue. In combination with the longer half-life of fluorine-18,

[18F]2-11 is beneficial to the PET imaging studies when translated to higher species.

Chapter 3 introduces a metal-free benzylic C-H bond activation enabled 18F labeling of

18 [ F]arylCF2H functionality. This methodology features a superior specific activity compared with those reported in literature as well as a diverse substrate scope.

Chapter 4 highlights the molecular modeling-assisted elucidation of the severe adverse event brought by the FAAH inhibitor BIA 10-2474 in phase 1 clinical trial and the development of series of novel covalent and non-covalent MGL inhibitors as potential PET radiotracers.

Chapter 5 describes the design and synthesis of series of compounds targeting A2AR as potential cancer immunotheraputics, following the hypothesized hypoxia-adenosinergic pathway. This project results in one promising compound 5-34 with satisfactory results in cAMP and IFN- gamma immunoassays.

Table of Contents

Dedication ...... ii

Acknowledgements ...... iii

Abstract of Dissertation ...... v

Table of Contents ...... vi

List of Figures ...... x

List of Schemes ...... xiv

List of Abbreviations...... xxi

Chapter 1: PET imaging and synthesis of PET radiotracers ...... 1

1.1 Background ...... 1

1.2 Application of fluorine in medicinal chemistry ...... 9

1.3 Fluorine-18 labeling chemistry ...... 12

1.3.1 [18F]-fluorine production ...... 12

1.3.2 Nucleophilic 18F-fluorination of alky groups ...... 14

1.3.3 Aromatic 18F-fluorination via [18F]F- fluoride ...... 16

1.3.4 Aromatic electrophilic 18F-Fluorination ...... 30

1.3.5 18F-trifluorination and difluorination of (hetero)arenes...... 33

18 18 1.3.6 F-labeling of aryl-SCF3, OCF3 and OCHF2 with [ F]fluoride ...... 37

1.4 Carbon-11 labeling chemistry ...... 38

1.5 Summary ...... 44

References ...... 46

Chapter 2: Radiosynthesis and preliminary PET evaluation of [18F]2-11 as AMPA radiotracer ...... 53

2.1 Background ...... 53

2.2 AMPA receptor as a drug discovery target...... 58

2.3 AMPA PET radiotracers ...... 63

2.4 Radiosynthesis of [18F]2-11 as AMPA radiotracer ...... 70

2.5 Preliminary PET imaging evaluation of [18F]2-11 ...... 83

2.6 Radiosynthesis of the para-analogue [18F]2-36 as potential AMPA PET tracer ...... 88

2.7 Summary ...... 90

Experimental Section ...... 91

References ...... 102

18 Chapter 3: Metal-free F labeling of aryl-CF2H via nucleophilic radiofluorination and oxidative C-H activation ...... 107

3.1 Background ...... 107

3.2 Synthesis of the (hetero)aryl CF2H group ...... 108

18 3.3 F-labeling of the aryl-CF2H group ...... 117

18 3.4 F-labeling of the aryl-CF2H via oxidative benzylic C-H bond activation ...... 121

3.5 Investigation of the substrate scope for the benzylic C-H activation radiosynthesis ...... 127

3.6 Specific activity measurement ...... 131

vii

3.7 Summary ...... 135

Experimental Section ...... 136

References ...... 165

Chapter 4: Molecular modeling of FAAH and MGL targeting the cannabinoid system ... 169

4.1 Background ...... 169

4.2 Investigation of BIA 10-2474 as irreversible FAAH inhibitor ...... 173

4.2.1 FAAH as a drug discovery target ...... 173

4.2.2 Investigation of the adverse side effects of BIA 10-2474 ...... 178

4.2.3 Molecular docking of BIA 10-2474 and PF-04457845 ...... 180

4.3 Development of novel radiotracers for the PET imaging of MGL ...... 186

4.3.1 MGL as a drug discovery target ...... 186

4.3.2 Development of PET radiotracers for MGL ...... 192

4.4 Molecular modeling assisted drug design for novel MGL PET tracers ...... 193

4.4.1 Investigation of the binding interaction of ZYH in MGL ...... 193

4.4.2 Development of novel covalent MGL antagonists based on compound 4-1 ...... 195

4.4.3 Noncovalent MGL antagonists of PAD and its analogs ...... 203

4.5 Summary ...... 209

Experimental Section ...... 210

References ...... 225

viii

Chapter 5: Fluorinated adenosine A2A receptor (A2AR) antagonists as potential cancer immunotherapeutics ...... 235

5.1 Background ...... 235

5.1.1 Development of A2AR agonists ...... 235

5.1.2 Development of A2AR antagonists ...... 239

5.2 Structure and binding pocket of A2AR...... 246

5.3 A2AR antagonists as cancer immunotherapeutics ...... 249

5.4 Design of novel A2AR antagonists as potential cancer immunotherapeutics ...... 255

5.5 Synthesis of the PEGylated derivatives ...... 257

5.5.1 Optimization and scale up of KW6002 and KW-PEG ...... 257

5.5.2 Synthesis of preladenant and tozadenant and their PEGylated derivatives ...... 260

5.6 Immunoassay results ...... 269

5.7 Summary ...... 275

Experimental Section ...... 276

References ...... 288

Chapter 6: Conclusions and Future Directions...... 293

Apendix ...... 295

List of Figures

Chapter 1: PET imaging and synthesis of PET radiotracers………………………...……….1

Figure 1-1. Schematic view of PET isotope production by cyclotron...... 4

Figure 1-2. Mechanism of PET molecular imaging...... 5

Chapter 2: Radiosynthesis and preliminary PET evaluation of [18F]2-11 as AMPA radiotracer………………………………………………………………………………………53

Figure 2-1. Overview of glutamate receptor family...... 53

Figure 2-2. Structures of iGluRs natural binding ligands...... 54

Figure 2-3. Schematic excitatory synapse and functions of iGluRs...... 55

Figure 2-4. Chemical structures of NMDA receptor antagonists...... 57

Figure 2-5. Chemical structures of kainate receptor selective binding ligands...... 58

Figure 2-6. The architecture of rat GluA2 receptor in a “broad” view...... 59

Figure 2-7. Illustration of AMPA receptor architecture via domain structures...... 60

Figure 2-8. Mechanism of epileptogenesis and potential therapeutic intervention via AMPA antagonists...... 61

Figure 2-9. Chemical structures of competitive AMPA receptor antagonists...... 62

Figure 2-10. Chemical structures of non-competitive AMPA receptor antagonists...... 63

Figure 2-11. PET baseline scan results of a rhesus monkey with [11C]2-8 to [11C]2-11...... 69

Figure 2-12. HPLC analysis of reformulated [18F]2-11 in 10% EtOH/saline (v/v)...... 84

Figure 2-13. Co-injection of [18F]2-11 with its unlabeled isotopologue...... 84

Figure 2-14. Whole brain biodistribution of [18F]2-11...... 85

Figure 2-15. Whole body biodistribution of [18F]2-11...... 87

Figure 2-16. Chromatograph of the second step radiosynthesis of [18F]2-36...... 90

18 Chapter 3: Metal-free F labeling of aryl-CF2H via nucleophilic radiofluorination and oxidative C-H activation………………………………………………………………………107

Figure 3-1. Aryl-CF2H containing bioactive molecules...... 108

Figure 3-2. 18F-labeling of bioactive molecules via palladium catalyzed α-arylation...... 121

Chapter 4: Molecular modeling of FAAH and MGL targeting the cannabinoid system...169

Figure 4-1. Chemical structures of 9-THC, 2-AG and AEA...... 169

Figure 4-2. Schematic overview of eCB system and 2-AG and AEA metabolism...... 171

Figure 4-3. Structures of URB597 and OL-135 as FAAH inhibitors...... 172

Figure 4-4. FAAH in complex with URB 597 and the location of MAC, ABP and CP...... 176

Figure 4-5. Chemical structures of typical FAAH inhibitors...... 177

Figure 4-6. Structure of BIA 10-2474...... 178

Figure 4-7. Structure of [18F]DOPP...... 179

Figure 4-8. Structure of α-ketooxadiazole...... 180

Figure 4-9. Noncovalent docking result of PF-04457845 in FAAH (PDB ID 3PPM)...... 181

Figure 4-10. Noncovalent docking result of BIA 10-2474 in FAAH (PDB ID: 3PPM)...... 182

Figure 4-11. Covalent docking results of PF-04457845 in FAAH...... 183

Figure 4-12. Covalent docking results of BIA 10-2474 in FAAH...... 185

Figure 4-13. Key structural features of MGL-crystallized structure of 3HJU...... 186

Figure 4-14. Chemical structures of SAR629, SAR127303 and ZYH...... 188

Figure 4-15. Crystallized structure of MGL in complex with ZYH...... 189

Figure 4-16. Chemical structures of typical MGL inhibitors...... 191

Figure 4-17. Urea and carbamate based MGL PET tracers...... 192

Figure 4-18. Comparisons of the docking results of ZYH from SP and XP methods, superimposed with the crystal structure position...... 194

Figure 4-19. XP docking result of ZYH...... 195

Figure 4-20. Compound 4-1 and its analogs bearing different leaving groups...... 196

Figure 4-21. Binding interaction of compound 4-1 in the MGL binding pocket...... 197

Figure 4-22. Covalent binding interactions of compound 4-1 in the binding site of MGL...... 198

Figure 4-23. Analogs of compound 4-1 with modifications on its core structure...... 199

Figure 4-24. Binding interaction of compound 4-7 in the MGL binding pocket...... 201

Figure 4-25. Covalent binding interactions of compound 4-7 in the binding site of MGL...... 202

Figure 4-26. Structures of PAD, FEPAD and FPPAD...... 203

Figure 4-27. (A) Overview of MGL in complex with PAD; (B) XP docking result of PAD. .... 204

Figure 4- 28. XP docking result of FEPAD in the binding site of MGL...... 205

Figure 4-29. XP docking result of FPPAD in the binding site of MGL...... 206

Figure 4-30. Superimposition of the binding poses of ZYH, PDA, FEPAD and FPPAD in the active site of MGL...... 207

Figure 4-31. Representative whole body PET images (0-90 min) of [11C]PAD, [18F]FEPAD and

[18F]FPPAD...... 208

Chapter 5: Fluorinated adenosine A2A receptor (A2AR) antagonists as potential cancer immunotherapeutics…………………………………………………………………………..235

Figure 5-1. Typical adenosine A2AR agonists...... 237

Figure 5-2. Structures of A2AR agonists as therapeutics...... 238

Figure 5-3. Structures of Caffeine, Theophylline, DMPX, CSC and DMPTX...... 240

Figure 5-4. Structures of p-SS-DMPX, MSX-2, MSX-3, and MSX-4...... 241

xii

Figure 5-5. Structures of KF-17837 and KW-6002...... 242

Figure 5-6. Structures of A2AR antagonists CGS15943 and Preladenant...... 244

Figure 5-7. Structures of ZM241385 and Tozadenant...... 245

Figure 5-8. Examples of monocyclic 1,2,4-triazine and tri-substituted pyrimidine as A2AR antagonists...... 245

Figure 5-9. Crystal structure of the human A2AR in complex with ZM241385...... 247

Figure 5-10. Key binding interactions of ZM241385 in A2AR...... 248

Figure 5-11. Homology model of A2AR based on the crystal structure 3EML...... 249

Figure 5-12. Hypoxia-A2AR–mediated mechanism of tissue protection...... 251

Figure 5-13. Validations of the hypoxia-adenosinergic pathway...... 253

Figure 5-14. Structure of KW-PEG...... 254

Figure 5-15. Preliminary in vivo liver damage results from ConA mouse model...... 255

Figure 5-16. Glide XP docking results of preladenant and tozadenant...... 256

Figure 5-17. Cyclic-AMP results from lymphocytes...... 270

Figure 5-18. Docking results of compounds 5-32 to 5-34 via Glide XP method...... 272

Figure 5-19. The IFN-gamma assay results from splenocytes...... 273

xiii

List of Schemes

Chapter 1: PET imaging and synthesis of PET radiotracers……………...………………….1

Scheme 1-1. Structures of fludrocortisone and 5-flurouracil...... 9

Scheme 1-2. Fluorohydrin analogs of HIV-1 protease inhibitor indinavir...... 10

Scheme 1-3. Development of KSP inhibitor MK-0731...... 10

Scheme 1-4. Fluoro thrombin inhibitors based on RWJ-445167...... 11

Scheme 1-5. Structures of HCV NS5B inhibitors...... 12

18 18 Scheme 1-6. Synthesis of [ F]FDG and [ F]florbetapir via SN2 nucleophilic fluorination...... 14

Scheme 1-7. Synthesis of [18F](S)-THK-5105 and [18F]FMISO via epoxide ring opening...... 15

Scheme 1-8. Direct insertion of fluorine-18 onto benzylic position...... 16

Scheme 1-9. Fluorodeamination reaction of trimethylanilinium triflates...... 17

Scheme 1-10. Radiosynthesis of [18F]fluoroarenes via Balz-Schiemann reaction...... 17

Scheme 1-11. Radiosynthesis of [18F]spiperone via Wallach methodology...... 18

Scheme 1-12. Synthesis of 4-[18F]fluoroanisole via diaryliodonium salts...... 19

Scheme 1-13. Radiosynthesis of [18F]DAA1106, [18F]FMZ, and [18F]FIMX via asymmetrical diaryliodonium salts...... 20

Scheme 1-14. Radiosynthesis of 3-[18F]FPPMP and 4-[18F]FPPMP via their iodonium ylide precursors...... 21

Scheme 1-15. 18F-fluorination via spirocyclic iodonium ylides...... 22

Scheme 1-16. Radiosynthesis of drug-like molecules via sulfonium salts...... 23

Scheme 1-17. Radiosynthesis of 3- and 4-[18F]fluoronitrobenzene via diaryl sulfoxides...... 24

Scheme 1-18. Radiosynthesis of 2-bromo-4-[18F]fluorophenol via oxidative 18F-fluorination. ... 24

xiv

Scheme 1-19. Radiosynthesis of bis-Boc protected 4-[18F]fluorocatechol via electrochemical 18F- fluorination...... 25

Scheme 1-20. Radiosynthesis of [18F]paroxetine via palladium-mediated fluorination...... 27

Scheme 1-21. Radiosynthesis of [18F]MDL100907 via Ni-mediated fluorination...... 29

Scheme 1-22. Radiosynthesis of 6-[18F]fluoro-L-DOPA via copper-mediated fluorination...... 30

Scheme 1-23. Radiosynthesis of 6-[18F]fluoro-L-DOPA from [18F]AcOF...... 31

18 18 Scheme 1-24. Radiosynthesis of 6-[ F]fluoro-L-DOPA from [ F]F2 via radiofluorodemetalation...... 32

Scheme 1-25. Radiosynthesis of 6-[18F]fluoro-L-DOPA via [18F]Selectfluor fluorination...... 33

Scheme 1-26. Radiosynthesis of [18F]α,α,α-trifluorotoluene via isotopic and halogen exchange.34

Scheme 1-27. Radiosynthesis of [18F]celecoxib via halogen exchange...... 34

Scheme 1-28. Radiosynthesis of 4-([18F]trifluoromethyl)-1,1′-biphenyl and 4-

([18F]difluoromethyl)-1,1′-biphenyl via 18F-fluorodecarboxylation...... 35

18 18 Scheme 1-29. (A) In situ generation of [ F]CuCF3. (B) Radiosynthesis of [ F]fluoxetine and

[18F]flutamide...... 36

18 18 Scheme 1-30. Radiosynthesis of [ F](hetero)arylCF3 via [ F]CuCF3...... 37

18 Scheme 1-31. Radiosynthesis of [ F]aryl-SCF3, OCF3 and OCHF2 via silver-catalyzed halogen exchange...... 38

11 Scheme 1-32. Carbon-11 synthons derived from the [ C]CO2...... 39

Scheme 1-33. Radiosynthesis of [N-methyl-11C]flumazenil, [O-methyl-11C]raclopride, L-[S- methyl-11C]methionine and [11C]MNQP via 11C-methylation...... 40

Scheme 1-34. Radiosynthesis of [11C]1-6 via 11C-cyanation...... 41

Scheme 1-35. Radiosynthesis of [11C]zolmitriptan via Se-mediated 11C-carbonylation...... 41

11 11 Scheme 1-36. Radiosynthesis of [ C]MFTC via [ C]COCl2...... 42

11 11 11 Scheme 1-37. Radiosynthesis of [ C]PF-04457845 and [ C]AR-A014418 via [ C]CO2 fixation...... 44

Chapter 2: Radiosynthesis and preliminary PET evaluation of [18F]2-11 as AMPA radiotracer………………………………………………………………………………………53

Scheme 2-1. Radiosynthesis of the isoquinoline derivatives as potential PET tracers...... 65

Scheme 2-2. Radiosynthesis of [11C]perampanel using biaryl phosphine Pd(0) complex...... 67

Scheme 2-3. Novel PET radiotracers based on perampanel derivatives...... 68

Scheme 2-4. Synthesis of compound 2-12...... 71

Scheme 2-5. Synthesis of the final compound 2-11...... 72

Scheme 2-6. Attempted radiosynthesis of [18F]2-11...... 72

Scheme 2-7. Proposed synthesis of [18F]2-11 from its iodine(III) precursor 2-20...... 73

Scheme 2-8. Synthesis of the aryl-iodide analogue 2-21...... 74

Scheme 2-9. Proposed two-step radiosynthesis of [18F]2-11...... 75

Scheme 2-10. Synthesis of the iodine(III) precursor 2-28...... 75

Scheme 2-11. Synthesis of compound 2-30...... 76

Scheme 2-12. Novel method for the synthesis of compound 2-11...... 76

Scheme 2-13. Radiosynthesis of [18F]2-29...... 77

Scheme 2-14. Radiosynthesis of compound [18F]2-11...... 78

Scheme 2-15. Synthesis of the iodine(III) precursor 2-32...... 88

Scheme 2-16. Radiosynthesis of the para-analog [18F]2-36...... 89

18 Chapter 3: Metal-free F labeling of aryl-CF2H via nucleophilic radiofluorination and oxidative C-H activation………………………………………………………...…………….107

xvi

Scheme 3-1. Difluorination of aldehydes or ketones with aminosulfur trifluorides...... 109

Scheme 3-2. Different difluoromethylation rendered novel methodologies for the synthesis of

(hetero)aryl-CF2H functionality...... 110

Scheme 3-3. Formation and decomposition of the difluoromethyl copper complex...... 111

Scheme 3-4. Copper-mediated direct difluoromethylation...... 112

Scheme 3-5. Copper-absent methodologies for the synthesis of (hetero)aryl-CF2H...... 115

Scheme 3-6. Radical methodologies for the synthesis of aryl-CF2H...... 116

Scheme 3-7. Development of the [18F]Selectfluor bis(triflate)-rendered radiosynthesis of [18F] aryl-CF2H...... 118

18 Scheme 3-8. Silver (I)-mediated halogen exchange for radiosynthesis of [ F] aryl-CF2H...... 119

18 Scheme 3-9. Radiosynthesis of [ F] aryl-CF2H via benzoyl auxiliary and its application...... 120

18 Scheme 3-10. Radiosynthesis of [ F] aryl-CF2H via benzylic C-H bond activation...... 122

Scheme 3-11. Radiosynthesis of 4-([18F]difluoromethyl)-1,1'-biphenyl ([18F]3-3a) ...... 122

Scheme 3-12. Workflow for the radiosynthesis of [18F]3-2a and [18F]3-3a...... 126

Scheme 3-13. Radiosynthesis of [18F]3-2b to [18F]3-2i...... 128

Scheme 3-14. Radiosynthesis of [18F]3-3b to [18F]3-3i...... 130

18 Scheme 3-15. Proposed mechanism for the radiosynthesis of [ F]aryl-CF2H...... 131

Scheme 3-16. Specific activity determination for 4-([18F]difluoromethyl)-1,1'-biphenyl ([18F]3-

3a)...... 132

Chapter 4: Molecular modeling of FAAH and MGL targeting the cannabinoid system...169

Scheme 4-1. Catalytic reaction mechanism of AEA with FAAH...... 174

Scheme 4-2. Covalent binding mechanism of PF-04457845 with FAAH...... 184

Scheme 4-3. Radiosynthesis of [11C]PAD, [18F]FEPAD and [18F]FPPAD...... 208

xvii

Chapter 5: Fluorinated adenosine A2A receptor (A2AR) antagonists as potential cancer immunotherapeutics…………………………………………………………………………..235

Scheme 5-1. Photo- dimerization and isomerization of KW6002...... 243

Scheme 5-2. Synthesis of KW6002...... 258

Scheme 5-3. Synthesis of KW-PEG...... 260

Scheme 5-4. Synthesis of intermediate 5-18 for preladenant and its PEGylated analogs...... 261

Scheme 5-5. Synthesis of preladenant and its PEGylated analogs...... 262

Scheme 5-6. Synthesis of Preladenant with method I...... 263

Scheme 5-7. Proposed mechanism for the synthesis of compound 5-39...... 264

Scheme 5-8. Alternative method for the synthesis of preladenant...... 266

Scheme 5-9. Reaction of compound 5-15 and hydroxylethyl hydrazine...... 267

Scheme 5-10. Synthesis of tozadenant and its demethylated and PEGylated analogs...... 268

Scheme 5-11. Synthesis of 5-56...... 269

xviii

List of Tables

Chapter 1: PET imaging and synthesis of PET radiotracers……………...………………….1

Table 1-1. Properties of common positron emission radionuclides for PET...... 6

Chapter 2: Radiosynthesis and preliminary PET evaluation of [18F]2-11 as AMPA radiotracer………………………………………………………………………………………53

Table 2-1. Optimizations for the radiosynthesis of [18F]2-29...... 77

Table 2-2. Optimizations for the one-pot radiosynthesis of [18F]2-11...... 80

Table 2-3. Scale up for the radiosynthesis of [18F]2-29...... 82

Table 2-4. Scale up for the radiosynthesis of [18F]2-11...... 82

18 Chapter 3: Metal-free F labeling of aryl-CF2H via nucleophilic radiofluorination and oxidative C-H activation………………………………………………………………………107

Table 3-1. Optimization for the radiosynthesis of 4-([18F]fluoromethyl)-1,1'-biphenyl ([18F]3-2a)

...... 123

Table 3-2. Optimization for the radiosynthesis of 4-([18F]difluoromethyl)-1,1'-biphenyl ([18F]3-

3a)...... 125

18 18 Table 3-3. Summary of the specific activity for the current F-labeling of [ F]aryl-CF2H...... 134

Chapter 4: Molecular modeling of FAAH and MGL targeting the cannabinoid system...169

Table 4-1. Noncovalent docking results of compounds 4-1 to 4-6 and their binding potency. .. 199

Table 4-2. Noncovalent docking results of compounds 4-7 to 4-12 and their binding potency. 200

Table 4-3. Final energy of the covalent binding complex for compounds 4-6 to 4-12...... 201

Chapter 5: Fluorinated adenosine A2A receptor (A2AR) antagonists as potential cancer immunotherapeutics…………………………………………………………………………..235

xix

Table 5-1. Physicochemical properties and docking results of compounds 5-32 to 5-34...... 274

List of Abbreviations

ABP acyl-chain binding pocket

Abh4 α/β-hydrolase 4

AC adenylate cyclase

AChE acetylcholinesterase

ADME absorption, distribution, metabolism, and excretion

AEA N-arachidonolethanolamine

AMPA α-amino-3-hydroxy-5-methyl-4-isoxazolepropionicacid

AIBN azobisisobutyronitrile

A1R Adenosine A1 receptor

A2AR Adenosine A2A receptor

A2BR Adenosine A2B receptor

A3R Adenosine A3 receptor

BBB brain-blood barrier

BDMS bromodimethylsulfonium bromide

BEMP 2-tert-butylimino-2-diethylamino-1,3-dimethylper-

hydro-1,3,2-diazaphosphorine

BSA N,O-bis(trimethylsilyl)acetamide cAMP cyclic adenosine monophosphate xxi

CB1R cannabinoid receptor 1

CB2R cannabinoid receptor 2

CGS CGS21680

CNS central nervous system

Cod cyclo-1,5-octadiene

ConA concanavalin A

CP cytosolic port

DAA1106 N-(2,5-Dimethoxybenzyl)-N-(5-[18F]-fluoro-

2-phenoxyphenyl)acetamide

DAGLα diacylglycerol lipase-α

DAST diethylaminosulfur trifluoride

DFT density functional theory

DMA N,N-dimethylacetamide

DMEDA N,N′-dimethylethylenediamine

DMF N,N-dimethylmethanamide

DMFS zinc difluoromethanesulfinate

DMPU 1,3-dimethyl-3,4,5,6-tetrahydro-2(1H)-pyrimidinone

Dppf 1,1’-bis(diphenylphosphanyl)-ferrocene

xxii eCBs endocannabinoids

ECS endocannabinoid system

EDCI carbodiimide(3-dimethyl-aminopropyl)-

ethylcarbodiimidehydrochloride

EL extracellular loop

EOB end of bombardment

EOS end of synthesis

FAA fatty acid amides

FAAH fatty acid amide hydrolase

FDG fluorodeoxyglucose

FIMX 4-fluoro-N-[4-[6-(isopropylamino)pyrimidin-4-yl]-

1,3- thiazol-2-yl]-N-methylbenzamide

FMZ flumazenil

F-TEDA-OTf 1-chloromethyl-4-fluoro-1,4-diazoniabicyclo[2.2.2]

octane bis(trifluoromethane sulfonate)

F-TEDA-PF6 chloromethyl-4-fluoro-1,4-diazoniabicyclo [2.2.2]

octane bis(hexafluorophosphate)

GABA gamma-aminobutyric acid

xxiii

GABA-A Gamma Amino Butyric Acid-A

GSK-3b glycogen synthase kinase 3b

HCV hepatitis C virus

HMDS hexamethyldisilazane

IFN-γ interferon gamma iGluR ionotropic glutamate receptor

IL intracellular loop

IL-10 interleukin 10

IL-2 interleukin-2 kainate 2-carboxy-3-carboxymethyl-4-

isopropenylpyrrolidine

K222 Kryptofix 2.2.2, 2.2.2-Cryptand

KSP kinesin spindle protein

MAC membrane access channel

MAP methoxy arachidonyl fluorophosphonate

MGL monoacylglycerol lipase mGluR metabotropic receptor

MPI myocardial perfusion imaging

xxiv

MPO multiparameter optimization

MRI magnetic resonance imaging

MW molecular weight

NAE N-acyl ethanolamines

NAM N-arachidonylmaleimide

NAPE-PLD N-acyl phosphatidylethanolamine phospholipase D

NAT N-acyl taurine

NBQX 2,3-dihydroxy-6-nitro-7-sulfamoyl-

benzo[f]quinoxaline-2, 3-dione

NBS N-bromosuccinimide

NFSI N-fluorobenzenesulfonimide

NMDA N-methyl-D-aspartate

NMP N-methylpyrrolidone

OEA N-oleoyl ethanolamine

PDB protein data bank

PD Parkinson’s disease

PEA N-palmitoyl ethanolamine

PEG polyethylene glycol

xxv

PET Positron emission tomography

PIDA bis(acetoxy)iodobenzene

PLC phospholipase C

PPB plasma protein binding

PPTS pyridium p-toluenesulfonate

QMA quaternary ammonium chloride polymer

RCC radiochemical conversion

RCY radiochemical yield

SA specific activity

SAR structure-activity relationship

SNR signal to noise ratio

SPIAD (1r,3r,5r,7r)-spiro[adamantane-2,2'

-[1,3]dioxane]-4',6'-dione

SP standard precision

SUV standardized uptake value

TBOH tetra-n-butylammonium hydroxide

TCR T cell receptor

TEAB tetraethylammonium bicarbonate

xxvi

TLC thin-layer chromatography

TM transmembrane

TMEDA N,N,N′,N′-tetramethylethylenediamine

TNF-α tumor-necrosis factor alpha

TSSC Temporary Specialist Scientific Committee

T4L T4 lysozyme

XP extra-precision

2-AG 2-arachidonoylglycerol

3-[18F]FPPMP 4-((3-[18F]fluorophenoxy)-

phenylmethyl)piperidine

4-[18F]FPPMP 4-((4-[18F]fluorophenoxy)-

phenylmethyl)piperidine

9-THC (-)-trans-Δ9-tetrahydrocannabinol

xxvii

Chapter 1: PET imaging and synthesis of PET radiotracers

1.1 Background

Strategies for early decision making on new drug candidates are crucial to avoid extensive studies on molecules that are destined to fail in clinical trials.1-3 In drug development, successful candidates need to demonstrate their target engagement (e.g. exposure at the site of target and target binding) and pharmacodynamic effects induced from the binding event.4 Positron emission tomography (PET) can validate target engagement in preclinical assays (e.g. non-human primate), reducing the likelihood of mechanism failures at early stages. As a noninvasive nuclear medicine imaging technique, PET provides a moderate-resolution, sensitive image of the biodistribution and pharmacokinetics of a radiotracer in the body.3 In addition, PET prompts the development of novel drugs with reduced side effects by providing insights into the off-target binding events.5

Biodistribution imaging provides information about the distribution of a radiotracer and its routes of clearance. This information guides the safety dosimetry of new PET tracers during the drug receptor occupancy studies, especially the radiation exposure to organs where drugs are metabolized.6 The radioligands should have sufficiently high specific-to-non-specific binding ratio (i.e. signal to noise, SNR > 1.4) in the target enriched regions compared to the target lacking tissues to generate useful PET imaging for the analysis of target occupancy.1-3

Quantifications of target availability and direct assessment of drug-target binding interaction require the development of target-specific radioligands. As reported in the literature, many lead compounds rejected from the drug discovery program can translate to successful PET tracers.3, 7

This is due to the different property requirements for being a PET tracer versus being a marketable drug. For example, PET tracers are normally injected into subjects intravenously (IV)

1 without being optimized to be orally bioavailable before use. On the other hand, successful central nervous system (CNS) PET tracers demand the properties of fast brain penetration, no brain permeable radiolabled metabolites and physicochemical properties suitable for minimized non-specific binding to maximize SNR sensitivity.3,7 The tissue target concentration/tracer receptor affinity ratio (Bmax/Kd) is defined to characterize the specific binding of PET radiotracers with an ideal value of around 10 for successful PET radiotracers.3

PET has been successfully used to assess target engagement and support the development of cost effective drugs, in particular for CNS indications.3,7 Currently, a number of radioligands has been developed for various CNS targets, including metabotropic glutamate mGluR5 receptors, 5-

HT1B receptors, D2/D3 receptors, fatty acid amino hydrolase (FAAH) enzyme and adenosine A2A receptors.3, 8-9 Successful PET imaging for ion channel receptors are only reported for those formed by the inhibitory Gamma Amino Butyric Acid-A (GABA-A) receptors and the ionotropic glutamate receptors (i.e. AMPA).3, 10 The difficulties of developing novel PET tracers for the ion channels attribute to their low expressions in the brain and the low binding affinity of the radiolabled molecules.3 Noteworthy, most of the successful neuronal PET tracers originate from the orthosteric antagonists since they have a clear relationship between drug occupancy and efficacy. However, radiolabled agonists or partial agonists can be hard to track due to their low fractional occupancies. Allosteric modulators, either positive or negative, may have binding sites that are unique to specific chemical series, leading to detected signals irrespective to their binding location.3

Measurement of target availability/occupancy at different doses of drugs, starting from the baseline, allows the comprehensive evaluation of the time course of target occupancy changes followed by drug plasma concentration increase.11-14 Compared with an initial single time point

2 target occupancy study, 15 the complexity of this time-dose-occupancy relationship study has significantly increased. However, this comprehensive evaluation not only provides brain penetration information for CNS, but also defines the therapeutic safety window and ensures the optimized dose selection for Phase II and III trials.11-14 Moreover, it could also be estimated from single dose occupancy experiments via modeling and simulation, assuming the absolute target density is constant during the measurement. As a result, the PET occupancy studies can be carried out at the early stages and give the optimized drug dose range for the early stages first-in- human safety and tolerability studies, supporting the development of cost effective drugs.3

PET imaging is capable of assessing pharmacodynamics or functional effects besides target

15 18 engagement. For example, [ O]H2O and [ F]fluorodeoxyglucose (FDG) have been used to characterize neuronal activity and reflect brain function through measuring brain flow and glucose metabolism respectively.3,16 However, combination of PET with other molecular imaging technologies, such as X-ray and magnetic resonance imaging (MRI), provides both anatomic and functional information for biological processes at the molecular level, making PET as a new kind of “precision pharmacology”.17-18

Development of suitable, efficient and easily prepared radiopharmaceuticals/radiotracers has become the most important component for the PET imaging studies as this confers PET imaging the ability to quantify in vivo biochemical and pharmacological progresses.17-18 The information gathered from the PET studies can enable biodistribution, metabolism, target engagement and time-dose-occupancy relationships studies, allowing more effective experiments design for clinical trials.7

PET radioligands are prepared by incorporating the short-lived radionuclides, typically 15O,

13N, 11C and 18F, onto bioactive molecules. Typical cyclotron production of the PET isotopes is

3 illustrated in Figure 1-1. The protons generated by the ion source enter the hollow metal electrodes (dees), which are positioned between the north and south of a large electromagnet and separated by a narrow gap. Under the high-frequency alternating voltage (typically 20 kV, 25-30

MHz), the charged particles are accelerated, following a circular path until they exit the cyclotron and are deflected onto the target.19-20

Figure 1-1. Schematic view of PET isotope production by cyclotron.

Source: see website in reference 20, and copyright of 2014 Encyclopedia Britannica, inc.20

Decay of these radioisotopes emits a positron (β+) that collides with an electron after traveling a short distance in an event described as annihilation (Figure 1-2).21 Consequently, the two photons generated from annihilation travel collinearly in nearly opposite directions with an individual energy of 511 keV. This gamma radiation is able to penetrate body tissue and be

4 detected by PET cameras, providing well-defined PET images in high resolution to reflect the biodistribution of a PET tracer as a function of time.7, 21

Figure 1-2. Mechanism of PET molecular imaging.

Source: see website in reference 21.21

The use of radionuclides 15O and 13N are hampered by their short half-lives (i.e. 2.037 min and 9.965 min respectively, Table 1-1), making the multistep radiosynthesis of complex radiotracers very difficult.22 However, these radionuclides can be incorporated into water or ammonia to track a physiological process (e.g. tissue blood flow) due to their low radiation exposures.21 On the other hand, compounds labeled with 11C or 18F radionuclides are more

5 commonly used for the target engagement studies due to their half-lives of 20.39 min and 109.8 min, respectively (Table 1-1).3, 7, 22 Introduction of the isotopes ((e.g. 15O, 13N, 11C and 18F)) does not change the physiochemical or pharmacological characteristics of the drug molecules since their stable isotopes are also present in the drug molecules themselves.22 In addition, the trace/negligible amount of radiolabled molecules (normally in the nano-gram range) used in PET imaging also confers PET the capability of evaluating drug candidates in early drug discovery stage with limited toxicology testing.3, 7

Table 1-1. Properties of common positron emission radionuclides for PET. Maximum Range (mm) in Nuclides T1/2 (min) Production Decay product energy (MeV) H2O 11C 20.39 14N(p,α)11C 0.96 4.1 11B 13N 9.965 16O(p,α)13N 1.19 5.4 13C 15O 2.037 14N(d,n)15O 1.72 8.2 15N 18O(p,n)18F 18F 109.8 0.635 2.4 18O 20Ne(d,α)18F

Among the above mentioned nuclides, [18F]fluorine has the most ideal half-life for labeling pharmaceuticals.3, 7 This half-life allows multi-step synthesis of 18F-labeled complex molecules, same-day imaging with the same batch of radiolabeled pharmaceuticals, convenient distribution to different locations without the need of on-site cyclotron as well as multi-center trials. In addition, [18F]fluorine also has the advantages of low positron energy and shorter travel distance before annihilation, offering lower overall radiation exposure and better spatial resolutions

(Table 1-1).7, 22 Fluroine-18 has been successfully incorporated into small molecule drugs, peptides, aptamers and proteins.23

[18F]Fluorodeoxyglucose (FDG) is the most widely used PET tracer clinically.24-25 It is being produced and distributed on a daily basis for oncology studies. [18F]FDG mimics the behavior of glucose and is transformed into [18F]FDG-6-phophate in the cell. Whereas, the presence of fluorine atom prevents [18F]FDG-6-phophate from being metabolized, making it accumulate at a proportional speed to glucose need.24-25 Therefore, abnormal glucose metabolism, such as in tumor/cancer cells, can be detected by monitoring the accumulation of [18F]FDG-6-phophate.

However, [18F]FDG is not a universal tumor biomarker, especially when it comes to the cerebral tumors where the ratio of tumor metabolism on surrounding tissue is underestimated by the neuronal glucose consumption. Moreover, other biological events, such as inflammation, may also lead to the enhanced metabolism.25

In general, design of effective PET radioligands is based on or starting from the known molecules with properties suitable for imaging their biological targets in vivo, regarding to affinity (pico- to nanomolar range), selectivity (> 100-fold), permeability of biological barriers, etc.26 The CNS PET tracers demand good brain-blood barrier (BBB) penetration and low nonspecific binding as mentioned earlier, which correlate to the lipophilicity of the ligands

(ideally logD = 1-3).26 In addition, the PET radioligand should have adequate metabolic stability, since it is impossible to distinguish between labeled metabolites and the parent radioligand in vivo.7

To facilitate the incorporation of radionuclides, the ligand should also have functionality that is amenable to labeling with selective and robust methods.26-35 The radiosyntheses are typically designed as late-stage introduction, considering the short half-lives of PET radionuclides, though radiolabeled prosthetic groups are also frequently used.26-35

The radiochemical yield (RCY) of the radiosynthesis and the specific activity (SA) of the radiolabled products are also important.7, 36-37 RCY is the ratio between isolated amount of radioactivity for the purified labeled product and the amount of radioactivity present in the original source at the beginning of the synthesis.7 Obtaining high RCY is crucial for an efficient radiosynthesis but also present significant synthetic challenges. In radiochemistry, the radionuclides or labeled prosthetic groups are usually the limiting reagents with their concentrations in pico- to nanomolar ranges, while the labeling or coupling precursors are in large excess. This unfavorable stoichiometry compared with the corresponding “cold” fluorine-

19 model attributes to the difficulties of applying the “cold” chemistry to the radiochemistry.7

“Specific Activity” refers to the ratio of the amount of isolated radioactivity of radiolabled compound to the total moles of all radioactive and nonradioactive species present in the isolated tracer.36-37 SA normally has the units of mCi/mmol or GBq/µmol. Each isotope has a theoretical maximum SA. For instance, fluorine-18 has a high theoretical maximum SA of 1.71 x 106

Ci/mmol.22 Under this SA condition, only negligible amount of labeled mass is present, namely,

1 Ci fluorine-18 only has a mass of 10.5 ng.22 However, such high SA can never be reached due to the inevitable fluorine-19 contamination at various points of radiosynthesis, such as fluorine- containing tubing, the carrier gas F2 as well as the product decay. Normally, the amount of reactivity is determined to a fixed range to produce high quality images.37-38 Obtaining high SA is also required when applied to receptor/enzyme with a low density. Otherwise, the nonradioactive version of PET tracer will saturate the binding sites and compromise the sensitivity of PET imaging.38

1.2 Application of fluorine in medicinal chemistry

Fluorine is one of the leading constituents of the pharmaceuticals, present in about 20% of the currently approved pharmaceuticals in a wide variety of therapeutic areas.7, 39 Recent analysis shows the increasing frequency of the new fluorinated drugs since the development of the first fluorine-containing fluorcortisones and 5-flurouracil in 1950s (Scheme 1-1).39 Fluorine atom is the smallest halogen atom with a van der Waals radius of 1.47 Å, close to the hydrogen atom

(1.2 Å), but is the most electronegative element (3.98 on the Pauling electronegativity scale).

Integration of fluorine atom into a molecule is known to affect its conformation, pKa, lipophilicity and electrostatic interactions, therefore, dramatically influence its intrinsic potency, selectivity and pharmacokinetic properties including absorption, distribution, metabolism, and excretion (ADME).7

Scheme 1-1. Structures of fludrocortisone and 5-flurouracil.

The conformational controlling effect of fluorine comes from its high electronegativity, resulting in highly polarized C-F bond with strong dipole moment (µ C-F = 1.41 D) and low lying C-F σ* orbital.7 In general, the gauche alignment between C-F bond and its vicinal

+ functionality (C-X, X = F, OH, NH2, NH3 , OAc and NHAc, etc) is preferred to provide favorable hyperconjugation and electrostatic interactions.7, 40-42 This principal was applied in the developments of novel drugs and organocatalysts. For example, among the newly designed

9 analogs of HIV-1 protease inhibitor indinavir (Scheme 1-2),43 fluorine atom that adopted a gauche relationship with the adjacent OH group (i.e. indinavir-a and indinavir-b) would produce similar crystallographic conformations and parallel inhibitory ability toward HIV-1 protease.

Scheme 1-2. Fluorohydrin analogs of HIV-1 protease inhibitor indinavir.

The strong electron withdrawing properties of fluorine is also utilized to modulate the pKa of proximal functionality to affect the solubility, permeability and binding affinities through lipophilicity changes.7 This is exemplified by the drug design of kinesin spindle protein (KSP) inhibitor MK-0731 (Scheme 1-3), which was a clinical candidate for the treatment of taxane- refractory solid tumors. Introduction of the fluorine substituent on the piperidine ring of compound 1-1 reduced the basicity of the molecule to a pKa range of 6.5-8.0, where the limited efficacy caused by P-glycoprotein-mediated efflux can be minimized.44

Scheme 1-3. Development of KSP inhibitor MK-0731.

A similar tactic was utilized in the design of thrombin inhibitors based on RWJ-445167, which entered clinical evaluations, but suffered poor bioavailability.45 To solve this problem, the fluorophenyl derivatives 1-2 and 1-3 were designed as more lipophilic inhibitors (Scheme 1-4).

46-47 The aryl fluorine atoms in compounds 1-2 and 1-3 were found in close contact with Gly216 as the pyridone C=O group in RWJ-445167. Meanwhile, the gem-difluoroethyl moiety mimicked the shape and dipole of the sulfonamide group in RWJ-445167, but reduced the pKa of aniline N-H by 3 log units, which strengthened a key interaction with Gly216. Both compounds

1-2 and 1-3 were potent thrombin inhibitors. Compound 1-3 showed robust efficacy in canine anticoagulation and thrombosis models following oral administration.46-47

Scheme 1-4. Fluoro thrombin inhibitors based on RWJ-445167.

In addition, fluorine can also be considered as a hydrogen bond acceptor involved in drug- target interactions.7 For example, during the drug design in Scheme 1-5, based on an indole-2- carboxylic acid phenotype 1-4, as hepatitis C virus (HCV) NS5B polymerase inhibitors,48 the phenyl fluorine atom was found to engage in a FH-N bonding interaction with the backbone of residue Tyr448 at a distance of 2.6 Å. Inspired by this observation, replacement of this fluorine atom with a powerful H-bond donating carbonyl group in compound 1-5 significantly increased the binding potency of compound 1-4, supporting the authors’ hyphothesis.48

Scheme 1-5. Structures of HCV NS5B inhibitors.

1.3 Fluorine-18 labeling chemistry

Fluorine-containing drugs can be easily re-envisioned as PET tracers by simply replacing fluorine-19 with fluorine-18. The physicochemical properties of the radiolabeled compounds will stay unchanged compared with their fluorine-19 counterparts. On the other hand, direct introduction of fluorine-18 onto the drug molecules is also a frequently used strategy in PET radiotracer design. The challenge is finding robust methods for installation of fluorine-18 atoms.

Herein, the production of fluorine-18 nuclide and the development of fluorine-18 incorporation methodologies are discussed.

1.3.1 [18F]-fluorine production

[18F]-fluorine can be produced by both 20Ne(d,α)18F and 18O(p,n)18F reactions in a cyclotron.

In the 20Ne(d,α)18F reaction, deuterons are used as charged particles to collide with the high-

49 18 18 pressure neon-20 gas. The fluorine-18 is obtained in the form of [ F]-fluorine gas ([ F]-F2) as a highly reactive electrophilic fluorinating reagent. However, it contains fluorine-19 as a carrier,

25, 50 18 leading to significantly reduced SA. Alternatively, [ F]-F2 could be generated from the

18 18 18 51 nuclear reaction of O(p,n) F by adding 0.2% F2 to an enriched [ O]O2 gas target. This method also requires the addition of carrier, resulting in a much reduced SA of 1 GBq/µmol. A

18 higher SA of 555 GBq/ µmol for [ F]-F2 is witnessed by using a “post-target” method using

18 [ F]-fluoromethane, which is derived from the nucleophilic replacement of CH3I via

[18F]fluoride.52

[18F]fluoride is the most commonly used reagent for the synthesis of [18F]-labeled PET radioligands.7, 23 It is generated through the proton bombardment of 18O-enriched water (18O-

18 18 51 18 18 H2O) in O(p,n) F nuclear reaction. The resulting [ F]fluoride is obtained in the form of [ -

F]F– ions in an aqueous solution, making it a poor nucleophile because of its high solvation.

18 18 Isolation of [ F]fluoride from O-H2O is achieved by using an anion exchange resin cartridge, such as quaternary ammonium chloride polymer (QMA) or Chromafix PS-HCO3 polymer. The

18 [ F]fluoride is typically eluted from the cartridge with K2CO3 (or another alkali base) and a phase transfer catalyst (e.g. Kryptofix 2.2.2 (K222) or crown ether) in MeCN/H2O (typically, v/v

= 7:3). Then, azeotropic drying is performed to remove any residual water to generate the

18 K[ F]/K222 complex. Alternatively, tetrabutylammonium hydroxide (TBOH) or tetraethylammonium bicarbonate (TEAB), as both a base and a phase transfer reagent, could be used to give the [18F]TEAF complex.7, 53 To improve this process, Seo et al. reported an eluting method of using of ionic liquid (tetrabutylammonium methanesulfonate or 1-butyl-3- methylimidazolium triflate) in MeOH.54 The subsequent azeotropic drying requires only 1 min with a 10% increase of the recovered radioactivity from this process. In addition, Wessmann et

18 - al. reported the elution of [ F]fluoride with an anhydrous solution of [K222]OH in MeCN, where the elution mixture can be directly used in their nucleophilic fluorination reactions without azeotropic drying process.55 Notably, [18F]fluoride can get a SA as high as 5500 GBq/µmol because no carrier is added during it production.56

1.3.2 Nucleophilic 18F-fluorination of alky groups

Traditional nucleophilic displacement of a aliphatic (pseudo)halide atom by [18F]fluoride has been successful for introducing [18F]fluorine onto a molecule.53 For example, the radiosynthesis of [18F]FDG is typically obtained from its OTf precursor in a polar aprotic solvent (MeCN) following the removal of the acetyl groups (Scheme 1-6).24 Similarly, [18F]florbetapir was prepared in a two-step manner in DMSO as the first clinically approved PET tracer for the quantification of amyloid plaque burden in human.57

18 18 Scheme 1-6. Synthesis of [ F]FDG and [ F]florbetapir via SN2 nucleophilic fluorination.

The need for 18F-fluorinated ligands also stimulates the development of new methodology.53

Doyle et al. reported the synthesis of enantioselective [18F]fluorohydrins under mild reaction conditions via epoxide ring-opening reaction with chiral transition metal fluoride catalysts

18 18 58 [ F](R,R)-(salen)CoF and [ F](R,R,R,R)-(linked salen)Co2OTsF (Scheme 1-7). These catalysts

14 were easily synthesized from their OTf precursors by elution of [18F]fluoride from QMA cartridge under air without using rigorously dried solvents or glassware. This approach was successfully applied to the preparation of several clinically validated PET tracers, including a previous tracer [18F](S)-THK-5105 for tau pathology studies and [18F]FMISO for quantifying tumor hypoxia in vivo. The presence of base-sensitive functional groups, epimerizable stereocenters and nitrogen rich moitifs in these two PET radioligands was well tolerated due to the exceptionally mild reaction conditions.59-61

Scheme 1-7. Synthesis of [18F](S)-THK-5105 and [18F]FMISO via epoxide ring opening.

Similarly, Groves et al. described a late-stage insertion of fluorine-18 onto the benzylic position using no-carrier-added [18F]fluoride and F-transfer catalyst Mn(salen)OTs (Scheme 1-

8).61 This method had been applied to the radiolabeling of various organic molecules and known drug, including 18F-ibuprofen ester as a COX inhibitor, 18F-celecoxib analog as a selective COX-

2 inhibitor as well as 18F-N-TFA-rasagiline as a MAO-B inhibitor with RCY ranging from 20%

15 to 72%. In addition, this method avoided the laborious azeotropic drying process prior to reaction and the [18F]fluoride could be directly eluted from an ion exchange cartridge with the catalyst.61

Scheme 1-8. Direct insertion of fluorine-18 onto benzylic position.

1.3.3 Aromatic 18F-fluorination via [18F]F- fluoride

Incorporation of [18F]fluorine into [18F]fluoroarenes has traditionally been accomplished via

62 SNAr reaction with electron-deficient aryl precursors. The arenes are activated by strong electron-withdrawing groups positioned ortho or para relative to the displaced group. The

16 commonly used activating groups are ranked in an order of their reactivity as p-NO2 > p-CF3 ≈ p-CN > p-CHO > p-Ac > m-NO2. Nitro and trimethylammonium groups are frequently used as leaving groups in the SNAr reactions, whereas halogen exchange (F > Cl > Br > I) with

[18F]fluoride is generally much less effective.63-65 However, this methodology suffers poor reaction scope, harsh reaction conditions, and low to moderate RCYs. It normally requires high reaction temperature in polar aprotic solvent. In the case of aryltrimethylammonium salts rendered fluorination, the competing side reaction led to the formation of [18F]-fluoromethane as a side-product (Scheme 1-9).66

Scheme 1-9. Fluorodeamination reaction of trimethylanilinium triflates.

On the other hand, electron-rich and neutral [18F]fluoroarenes were originally synthesized with [18F]fluoride via Balz-Schiemann reaction followed by thermal fluorodediazonation of aryldiazonium [18F]tetrafluoroborates (Scheme 1-10).53 The 18F-19F isotopic exchange with the

18 - unlabeled fluoride in the [ F]BF4 counterion, however, makes this reaction a maximum

- theoretical RCY of 25% and a lower SA. Knochel et al. replaced the BF4 counterion with diazonium anions that wouldn’t undergo isotopic exchange, such as 2,4,6- triisopropylbenzenesulfonate, resulting in an improved RCY (decay-corrected) of 39% for the synthesis of 4-[18F]fluorotoluene.67

Scheme 1-10. Radiosynthesis of [18F]fluoroarenes via Balz-Schiemann reaction.

An alternative strategy takes advantage of the Wallach reaction, where the stable triazenes are used to generate aryldiazonium salts in situ after protonation (Scheme 1-11).68-69 Subsequent thermal decomposition of the aryldiazonium salts in the presence of [18F]fluoride yields the

[18F]fluoroarenes. Wallach radiolabeling is found to give better RCYs when halogenated solvents are used to inhibit the protodediazoniation. This is exemplified in the radiosynthesis of

18 [ F]spiperone, where the orginal solvent (MeCN) was replaced by Cl3CCN to improve the RCY from 0.5% to 4%.70

Scheme 1-11. Radiosynthesis of [18F]spiperone via Wallach methodology.

Recent development of novel SNAr methodologies has promoted the radiolabeling of unactivated and electron-rich aromatic system. These methodologies often resort to exogenous catalysts or precursors that bear activating auxiliaries.53

Di(hetero)aryliodonium salts are highly labile leaving groups that have been reported as air- and moisture-stable compounds since 1894.71 Preparation of these compounds starts from oxidation of the aryl iodide to an iodide(III) species under acidic conditions followed by anion

18 exchange with an arene or an organometallic reagent. SNAr reaction of di(hetero)aryliodonium salts with [18F]fluoride leads to the formation of [18F]fluoroarenes and iodoarenes.72 This methodology applies to both electron-deficient and electron-rich substrates. In cases of the unsymmetrical diaryliodonium salts, 18F-fluorination occurs at the more electron-deficient aromatic ring.73 Therefore, electron-rich substrates can be selectively radiolabled by utilizing a more electron-rich leaving groups in the asymmetrical diaryliodonium salts. For example, 4-

[18F]fluoroanisole was synthesized from (4-methoxyphenyl)(2-thienyl)iodonium bromide with a

RCY of 29 ± 3% (Scheme 1-12).73

Scheme 1-12. Synthesis of 4-[18F]fluoroanisole via diaryliodonium salts.

This method had been successfully applied to the radiolabeling of various PET radioligands and drug molecules, including N-(2,5-Dimethoxybenzyl)-N-(5-[18F]-fluoro-2- phenoxyphenyl)acetamide ([18F] DAA1106),74 [18F]flumazenil ([18F]FMZ),75 and 4-[18F]-fluoro-

N-[4-[6-(isopropylamino)pyrimidin-4-yl]-1,3-thiazol-2-yl]-N-methylbenzamide ([18F]FIMX)76

(Scheme 1-13). [18F]DAA1106 is a isotopologue of the [11C]DAA1106, which is a PET ligand for the imaging of peripheral benzodiazepine receptors in the brain. With the electron-rich 4- anisole group as a counterpart, [18F]DAA1106 was synthesized with a RCY of 65% and a ratio of 71:29 to the 4-[18F]fluoroanisole side product.74 [18F]FMZ is an effective PET radiotracer for the central benzodiazepine receptor. The utilization of diaryliodonium methodology with 4- methylphenyl-mazenil iodonium tosylate led to a decay-corrected RCY of 67% compared to that

18 75 obtained from the classical SNAr reaction of its nitro-analog with [ F]KF/K222 (26%).

[18F]FIMX is an effective PET radiotracer for imaging metabotropic glutamate receptor 1

(mGluR1) in monkey brain. It had been radiolabled by the SNAr reaction of a diaryliodonium precursor with [18F]fluoride in microwave reactor with the addition of TEMPO as radical scavenger. The resulting product had a decay-corrected RCY of 20% and a SA of 105

GBq/µmol.76

Scheme 1-13. Radiosynthesis of [18F]DAA1106, [18F]FMZ, and [18F]FIMX via asymmetrical diaryliodonium salts.

18F-fluorination of iodonium ylides has been successfully applied for the synthesis of PET radioligands, in particular the recent publications on spirocyclic hypervalent idodine(III) methodologies.53 Iodinium ylides are prepared from the condensation reaction between

(diacetoxyiodo)arene and various dicarbonyl compounds under basic conditions. Subsequent radiofluorination with nucleophilic [18F]fluoride is similar to that of diaryliodonium salts to get the [18F]fluoroarenes.77 Based on density functional theory (DFT) calculation, iodonium ylides derived from Meldrum acid and barbituric acid have better chemoselectivity than the diaryliodonium salts.78 Moreover, Meldrum’s acid-based iodonium ylides exhibited suitable stability and were employed for the synthesis of electron-rich 4-((3-[18F]fluorophenoxy)- phenylmethyl)piperidine (3-[18F]FPPMP) and its regioisomer 4-((4-

[18F]fluorophenoxy)phenylmethyl)piperidine (4-[18F]FPPMP) as potential PET tracers for serotonin and norepinephrine transporters receptors (Scheme 1-14).79

Scheme 1-14. Radiosynthesis of 3-[18F]FPPMP and 4-[18F]FPPMP via their iodonium ylide precursors.

In order to improve the stability of iodonium ylides, the dicarbonyl auxiliary were further optimized.80 Spirocyclic iodonium ylides based on Meldrum’s acid demonstrated enhanced

18 80 stability and improved RCYs for the SNAr reaction nucleophilic [ F]fluoride. For example, synthesis of 4-[18F]fluorobiphenyl was achieved with better radiochemical conversions (RCCs)

21 ranging from 53% to 85% when spirocyclic ylides were used. In addition, Meldrum’s acid derived spirocyclic iodonium ylides are crystalline solids, whereas analogues of barbituric acid suffered poor stability or were viscous oils.80 The cyclopentyl ring bearing spirocyclic iodonium ylides had been successfully applied to the radiosynthesis of [18F]fluoroestrone, 5-

[18F]fluorouracil and N,O-protected 4-[18F]fluorophenylalanine (Scheme 1-15).80

Scheme 1-15. 18F-fluorination via spirocyclic iodonium ylides.

Triarylsulfonium salts can be used as precursors in nucleophilic SNAr reaction to yield electron-deficient or electron-neutral [18F]fluoroarenes.81 The [18F]fluorination occurred at the

Cipso of the most electron-deficient aromatic rings. Utilization of electron-donating ancillary rings facilitated the labeling of the nonactivated arenes or functionalized heteroarenes. For example,

22 sulfonium salts bearing two anisole groups led to moderate to excellent RCYs for the 18F labeling of compounds shown in Scheme 1-16.81

Scheme 1-16. Radiosynthesis of drug-like molecules via sulfonium salts.

Similarly, diaryl sulfoxides were also used as labeling precursors under the standard

18 82 [ F]KF/K222 conditions, but this was only applicable to the electron-deficient arenes. It also required the activation of electron-withdrawing group in the para-position, whereas a meta- positioned electron-withdrawing substituent or electron-rich substrate failed to yield the

[18F]fluoroarene. For example, 4-[18F]fluoronitrobenzene was synthesized in 91% RCY from its

18 bi(4-NO2-phenyl)-sulfoxide compared to the negligible 2% RCY for 3-[ F]fluoronitrobenzene from its corresponding diarylsulfoxide precursor (Scheme 1-17).82

Scheme 1-17. Radiosynthesis of 3- and 4-[18F]fluoronitrobenzene via diaryl sulfoxides.

Electron rich arenes bearing 4-fluorophenol motif can be radiolabeled with [18F]fluoride by an oxidative metal-free umpolung-based methodology from the para-tertbutyl positioned phenol susbtrates.83 The reaction started from initial oxidation of the phenol substrate to form a hypovalent phenyloxenium ion, followed by nucleophilic attack by the [18F]fluoride and subsequent rearomatization to give the resulting 4-[18F]fluorophenol. The para-substituted tertbutyl group served as a directing and leaving group. In the radiosynthesis of 2-bromo-4-

[18F]fluorophenol, bis(acetoxy)iodobenzene (PIDA) was used as an oxidant and the resulting product was obtained with a RCY of 16% (Scheme 1-18).83

Scheme 1-18. Radiosynthesis of 2-bromo-4-[18F]fluorophenol via oxidative 18F- fluorination.

An analogous strategy was reported as electrochemical 18F-fluorination of electron-rich arenes.84-85 This reaction progressed by single electron oxidation of the arene to form an aryl radical cation, which was attacked by [18F]fluoride to form an aryl radical, followed by a second single electron oxidation and rearomatization to give the 18F-fluorinated product. Introduction of tert-butyl group was also found to favor this nucleophilic fluorination.84-85 In the radiosynthesis of bis-Boc-4-[18F]fluorocatechol, the oxidation potentials needed for the formations of radical and radical cation intermediates were lowered in the presence of tert-butyl group. At the end, bis-Boc protected 4-[18F]fluorocatechol was obtained in a decay-corrected RCY of 8.9% and a

SA of 0.043 GBq/µmol (Scheme 1-19).86

Scheme 1-19. Radiosynthesis of bis-Boc protected 4-[18F]fluorocatechol via electrochemical 18F-fluorination.

Palladium catalyzed C-F formation has been adapted to the 18F-fluorination under the mechanism involving a Pd(II/IV) catalytic cycle.87-88 According to Ritter et al., 18F-fluorination of a highly electrophilic Pd(IV) complex with [18F]KF/18-crown-6 led to the formation of high- valent [18F]Pd(IV)F complex with high RCY of 90% (complex I, Scheme 1-20).89-90 This complex (I) was stabilized by benzo[h]quinolyl and tetrapyrazole borate ligands and served as an

25 oxidant and electrophilic 18F-fluorinating reagent with high specific activity. On the other hand, the labeling precursors were prepared in the form of an aryl Pd(II) complexes from a benzo[h]quinolyl-sulfonamide Pd(II) complex and aryl boronic acids (complex II, Scheme 1-

20).89-90 These precursors were heat (≤ 100 °C) and moisture stable and could be isolated via

18 flash chromatography. Reaction between the [ F]PdF4 complex I and the aryl palladium precursor II underwent sequential single electron abstraction from aryl Pd(II) complex,

[18F]fluoride transfer from [18F]Pd(IV)F complex and a second single electron transfer from aryl palladium precursor to generate a second high-valent pyridyl-sulfonamide-stablized palladium

[18F]fluoride complex III. This complex III can undergo reductive elimination to form

[18F]fluoroarenes.89-90 This process had been exemplified in the radiosynthesis of [18F]paroxetine for imaging serotonin receptors in nonhuman primate (Scheme 1-20).91

Scheme 1-20. Radiosynthesis of [18F]paroxetine via palladium-mediated fluorination.

Ritter et al. also reported the preformed organometallic Ni(II)-aryl complexes as 18F labeling precursors.92 These precursors could be synthesized from aryl bromides and subsequent ligand exchange with a silver sulfonamide salt. Compared to the above palladium-mediated methodology, nickel-mediated [18F]fluorination only required one-step synthesis in presence of an oxidant and aqueous [18F]fluoride without the necessity to prepare an electrophilic fluorinating reagent.92 The typical azeotropic drying procedures could be avoided in small scale nickel-mediated [18F]fluorination starting with 3.7-18.5 MBq radioactivity, but the volumetric ratio of aqueous [18F]fluoride (normally, 2-5µL) to MeCN needed to be less than 1%. An excess amount of water would, however, degrade the Ni(II)-aryl complex and oxidant.92 Removal of water in the case of scale-up radiosynthesis required a controlled pH, otherwise the reaction would be too basic to enable the Ni-mediated fluorination.92 Utilization of pyridium p- toluenesulfonate (PPTS) as a buffering acid after the azeotropical drying with TEAB provided an in-process pH adjustment protocol to ensuring the success of this labeling process.93 This

18 strategy was applied to the radiosynthesis of [ F]MDL100907 for studying 5HT2A dysregulation in the brain of nonhuman primates (Scheme 1-21).93

Scheme 1-21. Radiosynthesis of [18F]MDL100907 via Ni-mediated fluorination.

Gouverneur et al. adapted the copper mediated Chan-Lam cross coupling reaction to the

18 18 94 18 [ F]fluorination of aryl boronic esters via [ F]KF/K222. This noncarrier added [ F]carbon- fluorine bond formation was typically carried out in DMF at 110 °C for 20 min in the presence of [Cu(OTf)2(py)4]complex. This strategy was applicable to the electron-rich arenes and ortho-

94 18 substituted substrates. The radioligand 6-[ F]fluoro-L-DOPA had been synthesized via this method to study the dopaminergic system in the brain (Scheme 1-22).95 The protected 6-

18 18 [ F]fluoro-L-DOPA could be obtained with a RCY of 55% starting from 3GBq of [ F]KF/K222.

18 Subseqeunt deprotection with HI (aq) gave 6-[ F]fluoro-L-DOPA with highly maintained enantioselectivity (ee > 99%).95

Scheme 1-22. Radiosynthesis of 6-[18F]fluoro-L-DOPA via copper-mediated fluorination.

1.3.4 Aromatic electrophilic 18F-Fluorination

18 Direct electrophilic fluorination of arenes using [ F]F2(g) yields many side products due to

96 its high oxidizing strength. It also suffers low specific activity because of the added F2 carrier

18 gas during its production. Nonetheless, [ F]F2(g) diluted with neon gas had been used for the

18 97 radiosynthesis of 6-[ F]fluoro-L-DOPA at low temperature (-65 °C) in superacidic media.

18 18 Milder electrophilic fluorinating reagents such as [ F]XeF2 and [ F]AcOF also led to the

18 radiosynthesis of this radioligand, however, the 2- and 5-[ F]fluoro-L-DOPA isomers were observed in these reactions (Scheme 1-23).98

Scheme 1-23. Radiosynthesis of 6-[18F]fluoro-L-DOPA from [18F]AcOF.

Electrophilic fluorination with the above fluorinating reagents could be achieved in a

99 18 regioselective manner by radiofluorodemetalation. Radiosynthesis of 6-[ F]fluoro-L-DOPA was accomplished regioselectively with [18F]AcOF via fluorodemercuration with a decay

100 18 corrected RCY of 11%. Alternatively, [ F]F2(g) could be employed with arylstannanes, arylsilanes and arylpentafluorosilicates as precursors, among which the arylstannane substrate

18 yielded the desired 6-[ F]fluoro-L-DOPA with an improved RCY of 33% and a low SA of 0.01

GBq/µmol (Scheme 1-24).101

18 18 Scheme 1-24. Radiosynthesis of 6-[ F]fluoro-L-DOPA from [ F]F2 via radiofluorodemetalation.

Selectfluor has proven to be a safe, nontoxic, and easy to handle fluorinating reagent.102 It has been used to prepare the fluoroarenes based on aryl stannane, aryl boronic ester substrates in the presence of silver or copper complexes.102 Preparation of [18F]Selectfluor bis(triflate) from

18 18 [ F]F2(g) provided an efficient and mild electrophilic fluorination reagent. [ F]Selectfluor had been utilized for the 18F-fluorination of electron-rich arylstannes in the presence of silver

103 18 triflate. Synthesis of 6-[ F]fluoro-L-DOPA with strategy based on a neopentyl glycol boronate substrates gave a RCY of 19% and an improved SA of 2.6 GBq/µmol (Scheme 1-25).104

Scheme 1-25. Radiosynthesis of 6-[18F]fluoro-L-DOPA via [18F]Selectfluor fluorination.

1.3.5 18F-trifluorination and difluorination of (hetero)arenes

18 F-fluorination of aryl-CF3 group was initially carried out via isotopic exchange reaction with a 13% RCY for the model substrate [18F]α,α,α-trifluorotoluene (Scheme 1-26).105 Halogen exchange reaction with the α-chloro-α,α-difluorotoluene substrate gave an improved RCY of

48% but required much elevated temperature (220 °C).105 Alternatively, benzotrichloride was

18 subjected to the halogen exchange reaction with [ F]HF/Sb2O3 mixture followed by isotopic

105 exchange reaction with SbF3 led to the desired product with 50% RCY.

Scheme 1-26. Radiosynthesis of [18F]α,α,α-trifluorotoluene via isotopic and halogen exchange.

It was also found the α-bromo-α,α-difluorotoluene motifs can undergo halogen exchange reaction under relatively mild conditions.106 This strategy had been applied to the radiosynthesis of [18F]celecoxib as a PET tracer for COX-2 receptor (Scheme 1-27).

Scheme 1-27. Radiosynthesis of [18F]celecoxib via halogen exchange.

Fluorodecarboxylation of the α,α-difluoro and α-fluoro-aryl acetic acids with the above radiofluorinating reagent [18F]Selectfluor bis(triflate) (Scheme 1-25) led to the formation of 18F- labeled trifluoro- and difluoromethylated arenes (Scheme 1-28).107 This reaction was catalyzed via silver nitrate and was applicable to the electron-neutral and electron-rich substrates. The exemplified radiosynthesis of 4-([18F]trifluoromethyl)-1,1′-biphenyl and 4-

([18F]difluoromethyl)-1,1′-biphenyl from their corresponding carboxylic acid precursors was achieved under mild reaction conditions with moderate RCYs and moderate SAs (Scheme 1-

107 18 28). Other methods for the radiosynthesis of [ F](hetero)aryl CF2H will be discussed in detail in Chapter 3.

Scheme 1-28. Radiosynthesis of 4-([18F]trifluoromethyl)-1,1′-biphenyl and 4- ([18F]difluoromethyl)-1,1′-biphenyl via 18F-fluorodecarboxylation.

18 Generation of [ F]CuCF3 from difluorocarbene source has been implemented in the

18 108-111 radiosynthesis of F-labled aryl-CF3 bearing molecules. Gouverneur et al. reported the

18 [ F]trifluoromethylation of (hetero)aryl iodides in the presence of CuI, the CF2: containing methylchlorodifluoroacetate, the chelating reagent N,N,N’,N’-tetramethylethylenediamine

18 18 108 (TMEDA) and nucleophilic [ F]fluoride in the form of [ F]KF/K222. Upon heating, the methylchlorodifluoroacetate underwent CuI-mediated deprotection and subsequent decarboxylation to generated difluorocarbene, which further reacted with [18F]fluoride to yield

18 108 [ F]CuCF3 (Scheme 1-29A). This species was then coupled with (hetero)aryl iodides to form

18 the [ F](hetero)arylCF3. This strategy had been applied to radiosynthesis of the antidepressant drug [18F]fluoxetine (Prozac) and of [18F]flutamide (Eulexin) for a prostate cancer study

(Scheme 1-29B).108

18 18 Scheme 1-29. (A) In situ generation of [ F]CuCF3. (B) Radiosynthesis of [ F]fluoxetine and [18F]flutamide.

Alternatively, Vugts et al. disclosed the copper-mediated deprotonation of

18 18 109 [ F]trifluormethane for the formation of an active [ F]CuCF3 species. The

[18F]trifluormethane was obtained via the displacement of iodide from difluoroiodomethane with

18 t [ F]KF/K222 with a RCY of 60 ± 15%. KO Bu was used as a strong base to deprotonate

36 trifluormethane, while Et3N3HF was employed to remove the excess base to stabilize the

18 110 [ F]CuCF3 species. The subsequent cross-coupling reaction with aryl iodides or aryl boronic

18 acid precursors led to the [ F](hetero)arylCF3. The same strategy was also reported by

Pannecoucke et al., except that they utilized S-(difluoromethyl)diarylsulfonium salts as precursors to access the [18F]trifluormethane (Scheme 1-30).111 [18F]trifluoromethylation of the aryl boronic acid precursors required the purging of air to the reaction mixture to form

18 18 [ F]Cu(II)CF3 species. This method could provide the [ F](hetero)arylCF3 at room temperature within 1 min.111

18 18 Scheme 1-30. Radiosynthesis of [ F](hetero)arylCF3 via [ F]CuCF3.

18 18 1.3.6 F-labeling of aryl-SCF3, OCF3 and OCHF2 with [ F]fluoride

18 53 Radiosynthesis of [ F]aryl-SCF3, OCF3 and OCHF2 has not been well explored. The current methods available for their 18F-labeling relied on silver-catalyzed halogen exchange

18 112 reaction with [ F]KF/K222 (Scheme 1-31). Their corresponding bromides or chlorides served

37 as precursors. The order of the precursors’ reactivity followed ArOCHFCl > ArCF2Br ≈

ArCHFCl > ArSCF2Br > ArOCF2Br. In general, the SA lay in the range of 0.1 GBq/µmol and no obvious 19F-18F isotope exchange occurred on the 18F-labeled products.112

18 Scheme 1-31. Radiosynthesis of [ F]aryl-SCF3, OCF3 and OCHF2 via silver-catalyzed halogen exchange.

1.4 Carbon-11 labeling chemistry

Incorporation of carbon-11 onto bioactive molecules or drugs has been widely used for developing PET radiotracers.1, 3, 113-114 The short half-life of carbon-11 (20.38 min) allows same day multiple PET imaging studies.3, 113 The theoretical maximum SA for 11C-radiolabeled molecules is 3.4 × 105 GBq/μmol,22, 114 which however can never be reached due to the presence of impurities, such as CO, CO2, CH4 and other sources of carbon. In human CNS PET imaging

38 studies, a radiotracer with SA > 74 GBq/μmol is required to provide PET images for the target engagement studies.115

The 14N(p,a)11C nuclear reaction via proton bombardment of nitrogen-14 is most frequently

11 114 11 used to produce the [ C]CO2. In the routine radiolabeling, [ C]CO2 is generally converted to various more reactive radiolabeling precursors/synthons as shown in Scheme 1-32.

11 Scheme 1-32. Carbon-11 synthons derived from the [ C]CO2.

11 11 [ C]CH3I and [ C]CH3OTf have been extensively applied to the PET radiotracer synthesis

11 3, 114 11 via various C-methylations. The initial radiosynthesis of [ C]CH3I adopted a so-called

11 11 ‘wet’ method, where the [ C]CO2 was reduced to [ C]CH3OH via LiAlH4 followed iodination via hydroiodic acid.116 Though this method was robust and provided high RCYs, the use of

11 LiAlH4 significantly reduced the SA of the [ C]CH3I and the following products since LiAlH4

116 was a source of non-radioactive CO2. Alternatively, a ‘dry’ or ‘gas phase’method was

11 11 developed to form [ C]CH3I through nickel catalyzed hydrogenation of [ C]CO2 and

11 117-118 11 subsequent iodination of [ C]CH4 with iodide vapor. This method led to highest SA C-

117-118 11 tracer of 4,700 GBq/µmol (Scheme 1-32). It is noteworthy that CH4 targets can also be employed for this process. As a result, the ‘dry’ method almost superseded the ‘wet’ method,

3, 114 11 especially when high SA was required for imaging the targets in the CNS. The [ C]CH3OTf

11 is a more reactive methylation reagent and is synthesized from [ C]CH3I with silver triflate at

119 11 11 elevated temperatures. Moreover, compared to [ C]CH3I, [ C]CH3OTf generally leads to

119 11 11 higher RCYs in reduced reaction time and at lower temperatures. [ C]CH3I and [ C]CH3OTf rendered 11C-methylations perform as N-, O-,and S-methylations. Recently, it had also been extended to the Stille, Suzuki, or Sonogashira cross-coupling reactions to form a 11C-C bond.

Examples are shown in Scheme 1-33 for the routine radiosynthesis of [N-methyl-11C]flumazenil,

11 11 11 [O-methyl- C]raclopride, L-[S-methyl- C]methionine as well as synthesis of 5-[ C]methyl-6- nitroquipazine) via Stille reaction.120

Scheme 1-33. Radiosynthesis of [N-methyl-11C]flumazenil, [O-methyl-11C]raclopride, L-[S- methyl-11C]methionine and [11C]MNQP via 11C-methylation.

[11C]HCN and [11C]CO are also important 11C-labeling precursors.114, 121 Reaction of

[11C]HCN with copper sulfate leads to the formation of [11C]CuCN, which can be applied to the

Rosenmund-von Braun reaction to yield [11C]aryl-CN.122 Radiosynthesis of compound 1-6 was achieved using this method with the corresponding aryl-bromide as a substrate. [11C]1-6 had been demonstrated as a translational PET radiotracer for imaging ionotropic glutamate receptor in nonhuman primate brain (Scheme 1-34).122

Scheme 1-34. Radiosynthesis of [11C]1-6 via 11C-cyanation.

[11C]CO has been applied in various palladium- and selenium catalyzed carbonylation reactions of organic halides to form carbon-11 labeled functional groups, such as ureas, carbamates, amides, ketones etc.123-124 These methods are generally performed in one-step manner and tolerable to most functional groups. The resulting carbon-11 products also feature high specific activity due to the low concentration of [11C]CO. Zolmitriptan is a serotonin 5-

123-124 11 HT1B/1D receptor agonist and is used for the treatment of migraines. [ C]zolmitriptan had been prepared by Se-mediated 11C-carbonylation (Scheme 1-35). [11C]zolmitriptan was

123-124 synthesized to map the 5-HT1B/1D binding sites in rhesus monkey brain.

Scheme 1-35. Radiosynthesis of [11C]zolmitriptan via Se-mediated 11C-carbonylation.

11 11 125 [ C]COCl2 is another alternative reagent for the C-carbonylation. It has advantageous high reactivity and good solubility in various solvents, but is limited by its rather complicated

11 production and requirements of specialized apparatus. In the radiotracer synthesis, [ C]COCl2 could be further converted to other 11C-labeled synthons, including isocyanates and carbamoyl chlorides from amines to get [11C]ureas and [11C]carbamates, as well as chloroformates from alcohols to yield alkyl [11C]carbonates.126 Zhang et al. reported the synthesis of carbamate

[11C]MFTC as a FAAH radiotracer.126 The symmetric diaryl 11C-carbonate intermediate was

11 synthesized to temper the high reactivity of [ C]COCl2. Subsequent transamidation with the secondary amine precursor led to the desired [11C]MFTC (Scheme 1-36).

11 11 Scheme 1-36. Radiosynthesis of [ C]MFTC via [ C]COCl2.

11 In addition, direct incorporation of [ C]CO2 from cyclotron onto bioactive molecules is also

127 11 an attractive strategy because of its high specific activity. The [ C]CO2 obtained from cyclotron generally undergoes purification before the reaction to remove oxygen and the side

127 11 products of nitric oxides. In the cryogenic purification, [ C]CO2 in target gas is trapped in a vessel cooled by liquid nitrogen to remove non-condensable gases, while condensable impurities 42

(e.g. NOx species) will be eliminated by chemical traps.127 Once the undesired impurities are

11 11 isolated, the [ C]CO2 is thermally released into the reactor. Fixation of [ C]CO2 with basic organometallic reagents is widely used to access various functional groups, including carboxylic

11 acids, carbamates, oxazolidinones and ureas. For example, [ C]CO2 was transformed to the

11 active acylation intermediate [ C]acetate in the presence of Grignard reagents CH3MgBr or

127 11 CH3MgCl. [ C]CO2 fixation had been recently applied to the radiosynthesis of many bioactive molecules or radiopharmaceuticals, such as the clinical evaluating fatty acid amide hydrolase (FAAH) inhibitor [11C]PF-04457845 and the selective glycogen synthase kinase 3β

(GSK-3β) inhibitor [11C]AR-A014418 (Scheme 1-37).128-129 These reactions were all carried out under mild reaction conditions in a short synthesis time. The 2-tert-butylimino-2-diethylamino-

1,3-dimethylper-hydro-1,3,2-diazaphosphorine (BEMP) was used as an efficient trapping reagent

11 128-129 for [ C]CO2 to replace the highly basic organometallic reagents.

11 11 11 Scheme 1-37. Radiosynthesis of [ C]PF-04457845 and [ C]AR-A014418 via [ C]CO2 fixation.

1.5 Summary

As a non-invasive imaging technology, PET is capable of in vivo quantification of biochemical and pharmacological progress via radiolabeled molecular probes. Quantification of biological targets of interest by PET would enable investigations of their functions under normal and disease conditions, assessment of their distribution in the brain and periphery, and target engagement for validation of promising drug candidates in clinical trials.

Recent developments of the 18F-fluorination methodologies have significantly broad the scope and synthetic routes in radiolabeling various fluorine-containing drugs without only resorting to the traditional SN2 displacement based on activated electron-deficient substrates.

18 18 Preparations of various key F-motifs, such as [ F]CFH2, CF2H, CF3, OCF3, SCF3 and OCHF2,

44 have been readily achieved on nonactivated or electron rich substrates. The continuous development of fluorine-bearing drugs in turn provides a strong impetus to keep investigating novel 18F-labeling methodologies to satisfy the PET imaging studies.

11 11 Meanwhile, the C-radiolabeling methodologies have advanced from the classic [ C]CH3I

11 methylation to ample radiolabeling synthons derived from [ C]CO2. The direct radiolabeling via

11 [ C]CO2 also gains more attentions and applies to the radiolabeling of clinical candidates. These progresses enable the access to various carbon-11 labeled motifs, allowing the radiolabeling of structurally complex radiopharmaceuticals and precise insertions of carbon-11. It is anticipated that future developments for radiolabeling methodologies will continuously prompt the PET imaging studies.

References

1. Frank, R.; Hargreaves, R. Clinical biomarkers in drug discovery and development. Nat. Rev. Drug Discov. 2003, 2, 566-580. 2. Hargreaves, R. J. The role of molecular imaging in drug discovery and development. Clin. Pharmacol. Ther. 2008, 83, 349-353. 3. Hargreaves, R. J.; Rabiner, E. A. Translational PET imaging research. Neurobiol. Dis. 2014, 61, 32-38. 4. Morgan, P.; Van der Graaf, P. H.; Arrowsmith, J.; Feltner, D. E.; Drummond, K. S.; Wegner, C. D.; Street, S. D. A. Can the flow of medicines be improved? Fundamental pharmacokinetic and pharmacological principles toward improving Phase II survival. Drug Discov. Today 2012, 17, 419-424. 5. Jacobson, O.; Chen, X. Y. PET Designated Flouride-18 Production and Chemistry. Curr. Top. Med. Chem. 2010, 10, 1048-1059. 6. Sprague, D. R.; Chin, F. T.; Liow, J. S.; Fujita, M.; Burns, H. D.; Hargreaves, R.; Stubbs, J. B.; Pike, V. W.; Innis, R. B.; Mozley, P. D. Human biodistribution and radiation dosimetry of the tachykinin NK1 antagonist radioligand [F-18]SPA-RQ: Comparison of thin-slice, bisected, and 2-dimensional planar image analysis. J. Nucl. Med. 2007, 48, 100-107. 7. Gillis, E. P.; Eastman, K. J.; Hill, M. D.; Donnelly, D. J.; Meanwell, N. A. Applications of Fluorine in Medicinal Chemistry. J. Med. Chem. 2015, 58, 8315-8359. 8. Jones, T.; Rabiner, E. A. The development, past achievements, and future directions of brain PET. J. Cerebr. Blood F. Met. 2012, 32, 1426-1454. 9. Barret, O.; Hannestad, J.; Vala, C.; Alagille, D.; Tavares, A.; Laruelle, M.; Jennings, D.; Marek, K.; Russell, D.; Seibyl, J.; Tamagnan, G. Characterization in humans of 18F-MNI-444, a PET radiotracer for brain adenosine 2A receptors. J. Nucl. Med. 2015, 56, 586-591. 10. Oi, N.; Tokunaga, M.; Suzuki, M.; Nagai, Y.; Nakatani, Y.; Yamamoto, N.; Maeda, J.; Minamimoto, T.; Zhang, M. R.; Suhara, T.; Higuchi, M. Development of Novel PET Probes for Central 2-Amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic Acid Receptors. J. Med. Chem. 2015, 58, 8444- 8462. 11. Addy, C.; Wright, H.; Van Laere, K.; Gantz, I.; Erondu, N.; Musser, B. J.; Lu, K.; Yuan, J.; Sanabria-Bohorquez, S. M.; Stoch, A.; Stevens, C.; Fong, T. M.; De Lepeleire, I.; Cilissen, C.; Cote, J.; Rosko, K.; Gendrano, I. N.; Nguyen, A. M.; Gumbiner, B.; Rothenberg, P.; de Hoon, J.; Bormans, G.; Depre, M.; Eng, W. S.; Ravussin, E.; Klein, S.; Blundell, J.; Herman, G. A.; Burns, H. D.; Hargreaves, R. J.; Wagner, J.; Gottesdiener, K.; Amatruda, J. M.; Heymsfield, S. B. The acyclic CB1R inverse agonist taranabant mediates weight loss by increasing energy expenditure and decreasing caloric intake. Cell Metab. 2008, 7, 68-78. 12. Bergstrom, M.; Hargreaves, R. J.; Burns, H. D.; Goldberg, M. R.; Sciberras, D.; Reines, S. A.; Petty, K. J.; Ogren, M.; Antoni, G.; Langstrom, B.; Eskola, O.; Scheinin, M.; Solin, O.; Majumdar, A. K.; Constanzer, M. L.; Battisti, W. P.; Bradstreet, T. E.; Gargano, C.; Hietala, J. Human positron emission tomography studies of brain neurokinin 1 receptor occupancy by aprepitant. Biol. Psychiat. 2004, 55, 1007-1012. 13. Erondu, N.; Gantz, I.; Musser, B.; Suryawanshi, S.; Mallick, M.; Addy, C.; Cote, J.; Bray, G.; Fujioka, K.; Bays, H.; Hollander, P.; Sanabria-Bohorquez, S. M.; Eng, W.; Langstrom, B.; Hargreaves, R. J.; Burns, H. D.; Kanatani, A.; Fukami, T.; MacNeil, D. J.; Gottesdiener, K. M.; Amatruda, J. M.; Kaufman, K. D.; Heymsfield, S. B. Neuropeptide Y5 receptor antagonism does not induce clinically meaningful weight loss in overweight and obese adults. Cell Metab. 2006, 4, 275-282. 14. Searle, G.; Beaver, J. D.; Comley, R. A.; Bani, M.; Tziortzi, A.; Slifstein, M.; Mugnaini, M.; Griffante, C.; Wilson, A. A.; Merlo-Pich, E.; Houle, S.; Gunn, R.; Rabiner, E. A.; Laruelle, M. Imaging Dopamine D-3 Receptors in the Human Brain with Positron Emission Tomography, [C-11]PHNO, and a Selective D-3 Receptor Antagonist. Biol. Psychiat. 2010, 68, 392-399.

15. Bench, C. J.; Lammertsma, A. A.; Dolan, R. J.; Grasby, P. M.; Warrington, S. J.; Gunn, K.; Cuddigan, M.; Turton, D. J.; Osman, S.; Frackowiak, R. S. J. Dose-Dependent Occupancy of Central Dopamine-D(2) Receptors by the Novel Neuroleptic Cp-88,059-01 - a Study Using Positron Emission Tomography and C-11 Raclopride. Psychopharmacology 1993, 112, 308-314. 16. Yu, S. Review of F-FDG Synthesis and Quality Control. Biomed. Imaging Interv. J. 2006, 2, e57. 17. Matthews, P. M.; Rabiner, E. A.; Passchier, J.; Gunn, R. N. Positron emission tomography molecular imaging for drug development. Brit. J. Clin. Pharmaco. 2012, 73, 175-186. 18. Cook, D.; Brown, D.; Alexander, R.; March, R.; Morgan, P.; Satterthwaite, G.; Pangalos, M. N. Lessons learned from the fate of AstraZeneca's drug pipeline: a five-dimensional framework. Nat. Rev. Drug Discov. 2014, 13, 419-431. 19. Cherry, S. R.; Sorenson, J. A.; Phelps, M. E. (2012). Physics in Nuclear Medicine (4th ed.). Philadelphia, PA: Saunders. 20. https://www.britannica.com/technology/particle-accelerator/Constant-voltage-accelerators. 21. Miller, P. W.; Long, N. J.; Vilar, R.; Gee, A. D. Synthesis of C-11, F-18, O-15, and N-13 Radiolabels for Positron Emission Tomography. Angew. Chem. Int. Ed. 2008, 47, 8998-9033. 22. Fowler, J. S.; Wolf, A. P. Working against time: Rapid radiotracer synthesis and imaging the human brain. Acc. Chem. Res. 1997, 30, 181-188. 23. Jacobson, O.; Kiesewetter, D. O.; Chen, X. Y. Fluorine-18 Radiochemistry, Labeling Strategies and Synthetic Routes. Bioconjug. Chem. 2015, 26, 1-18. 24. Hamacher, K.; Coenen, H. H.; Stocklin, G. Efficient Stereospecific Synthesis of No-Carrier- Added 2-[F-18]-Fluoro-2-Deoxy-D-Glucose Using Aminopolyether Supported Nucleophilic-Substitution. J. Nucl. Med. 1986, 27, 235-238. 25. Le Bars, D. Fluorine-18 and medical imaging: Radiopharmaceuticals for positron emission tomography. J. Fluorine Chem. 2006, 127, 1488-1493. 26. Van de Bittner, G. C.; Ricq, E. L.; Hooker, J. M. A Philosophy for CNS Radiotracer Design. Acc. Chem. Res. 2014, 47, 3127-3134. 27. Li, Z.; Conti, P. S. Radiopharmaceutical chemistry for positron emission tomography. Adv. Drug Deliv. Rev. 2010, 62, 1031-51. 28. Kim, D. W.; Jeong, H. J.; Lim, S. T.; Sohn, M. H. Recent Trends in the Nucleophilic [(18)F]- radiolabeling Method with No-carrier-added [(18)F]fluoride. Nucl. Med. Mol. Imaging 2010, 44, 25-32. 29. Cole, E. L.; Stewart, M. N.; Littich, R.; Hoareau, R.; Scott, P. J. H. Radiosyntheses using Fluorine-18: The Art and Science of Late Stage Fluorination. Curr. Top. Med. Chem. 2014, 14, 875-900. 30. Brooks, A. F.; Topczewski, J. J.; Ichiishi, N.; Sanford, M. S.; Scott, P. J. H. Late-stage [F- 18]fluorination: new solutions to old problems. Chem. Sci. 2014, 5, 4545-4553. 31. Littich, R.; Scott, P. J. H. Novel Strategies for Fluorine-18 Radiochemistry. Angew. Chem. Int. Ed. 2012, 51, 1106-1109. 32. Way, J.; Bouvet, V.; Wuest, F. Synthesis of 4-[F-18]fluorohalobenzenes and Palladium-mediated Cross-coupling Reactions for the Synthesis of F-18-labeled Radiotracers. Curr. Org. Chem. 2013, 17, 2138-2152. 33. Pretze, M.; Pietzsch, D.; Mamat, C. Recent Trends in Bioorthogonal Click-Radiolabeling Reactions Using Fluorine-18. Molecules 2013, 18, 8618-8665. 34. Tredwell, M.; Gouverneur, V. 18F Labeling of Arenes. Angew. Chem. Int. Ed. 2012, 51, 11426- 11437. 35. Liang, T.; Neumann, C. N.; Ritter, T. Introduction of Fluorine and Fluorine-Containing Functional Groups. Angew. Chem. Int. Ed. 2013, 52, 8214-8264. 36. Bonardi, M. L.; Birattari, C.; Groppi, F.; Gini, L.; Mainardi, H. S. C. Cyclotron production and quality control of "high specific activity" radionuclides in "no-carrier-added" form for radioanalytical applications in the life sciences. J. Radioanal. Nucl. Chem. 2004, 259, 415-419. 37. de Goeij, J. J. M.; Bonardi, M. L. How do we define the concepts specific activity, radioactive concentration, carrier, carrier-free and no-carrier-added? J. Radioanal. Nucl. Chem. 2005, 263, 13-18.

38. Lapi, S. E.; Welch, M. J. A historical perspective on the specific activity of radiopharmaceuticals: What have we learned in the 35 years of the ISRC? Nucl. Med. Biol. 2013, 40 (3), 314-320. 39. Ilardi, E. A.; Vitaku, E.; Njardarson, J. T. Data-mining for sulfur and fluorine: an evaluation of pharmaceuticals to reveal opportunities for drug design and discovery. J. Med. Chem. 2014, 57, 2832-42. 40. Zimmer, L. E.; Sparr, C.; Gilmour, R. Fluorine Conformational Effects in Organocatalysis: An Emerging Strategy for Molecular Design. Angew. Chem. Int. Ed. 2011, 50, 11860-11871. 41. Buissonneaud, D. Y.; van Mourik, T.; O'Hagan, D. A. DFT study on the origin of the fluorine gauche effect in substituted fluoroethanes. Tetrahedron 2010, 66, 2196-2202. 42. Briggs, C. R. S.; Allen, M. J.; O'Hagan, D.; Tozer, D. J.; Slawin, A. M. Z.; Goeta, A. E.; Howard, J. A. K. The observation of a large gauche preference when 2-fluoroethylamine and 2-fluoroethanol become protonated. Org. Biomol. Chem. 2004, 2, 732-740. 43. Myers, A. G.; Barbay, J. K.; Zhong, B. Y. Asymmetric synthesis of chiral organofluorine compounds: Use of nonracemic fluoroiodoacetic acid as a practical electrophile and its application to the synthesis of monofluoro hydroxyethylene dipeptide isosteres within a novel series of HIV protease inhibitors. J. Am. Chem. Soc. 2001, 123, 7207-7219. 44. Cox, C. D.; Coleman, P. J.; Breslin, M. J.; Whitman, D. B.; Garbaccio, R. M.; Fraley, M. E.; Buser, C. A.; Walsh, E. S.; Hamilton, K.; Schaber, M. D.; Lobell, R. B.; Tao, W.; Davide, J. P.; Diehl, R. E.; Abrams, M. T.; South, V. J.; Huber, H. E.; Torrent, M.; Prueksaritanont, T.; Li, C.; Slaughter, D. E.; Mahan, E.; Fernandez-Metzler, C.; Yan, Y.; Kuo, L. C.; Kohl, N. E.; Hartman, G. D. Kinesin spindle protein (KSP) inhibitors. 9. Discovery of (2S)-4-(2,5-difluorophenyl)-N-[(3R,4S)-3-fluoro-1- methylpiperidin-4-yl]-2-(hydroxymethyl)-N-methyl-2-phenyl-2,5-dihydro-1H-pyrrole-1-carboxamide (MK-0731) for the treatment of taxane-refractory cancer. J. Med. Chem. 2008, 51, 4239-4252. 45. Maryanoff, B. E.; McComsey, D. F.; Costanzo, M. J.; Yabut, S. C.; Lu, T. B.; Player, M. R.; Giardino, E. C.; Damiano, B. P. Exploration of potential prodrugs of RWJ-445167, an oxyguanidine- based dual inhibitor of thrombin and factor Xa. Chem. Biol. Drug Des. 2006, 68, 29-36. 46. Kreutter, K. D.; Lu, T. B.; Lee, L.; Giardino, E. C.; Patel, S.; Huang, H.; Xu, G. Z.; Fitzgerald, M.; Haertlein, B. J.; Mohan, V.; Crysler, C.; Eisennagel, S.; Dasgupta, M.; McMillan, M.; Spurlino, J. C.; Huebert, N. D.; Maryanoff, B. E.; Tomczuk, B. E.; Damiano, B. P.; Player, M. R. Orally efficacious thrombin inhibitors with cyanofluorophenylacetamide as the P2 motif. Bioorg. Med. Chem. Lett. 2008, 18, 2865-2870. 47. Lee, L.; Kreutter, K. D.; Pan, W.; Crysler, C.; Spurlino, J.; Player, M. R.; Tomczuk, B.; Lu, T. 2- (2-Chloro-6-fluorophenyl)acetamides as potent thrombin inhibitors. Bioorg. Med. Chem. Lett. 2007, 17, 6266-6269. 48. Anilkumar, G. N.; Lesburg, C. A.; Selyutin, O.; Rosenblum, S. B.; Zeng, Q. B.; Jiang, Y. H.; Chan, T. Y.; Pu, H. Y.; Vaccaro, H.; Wang, L.; Bennett, F.; Chen, K. X.; Duca, J.; Gavalas, S.; Huang, Y. H.; Pinto, P.; Sannigrahi, M.; Velazquez, F.; Venkatraman, S.; Vibulbhan, B.; Agrawal, S.; Butkiewicz, N.; Feld, B.; Ferrari, E.; He, Z. Q.; Jiang, C. K.; Palermo, R. E.; Mcmonagle, P.; Huang, H. C.; Shih, N. Y.; Njoroge, G.; Kozlowski, J. A. I. Novel HCV NS5B polymerase inhibitors: Discovery of indole 2- carboxylic acids with C3-heterocycles. Bioorg. Med. Chem. Lett. 2011, 21, 5336-5341. 49. Cole, E. L.; Stewart, M. N.; Littich, R.; Hoareau, R.; Scott, P. J. Radiosyntheses using fluorine-18: the art and science of late stage fluorination. Curr. Top. Med. Chem. 2014, 14, 875-900. 50. Guillaume, M.; Luxen, A.; Nebeling, B.; Argentini, M.; Clark, J. C.; Pike, V. W. Recommendations for F-18 Production. Appl. Radiat. Isot. 1991, 42, 749-762. 51. Kilbourn, M. R. Fluorine-18 Labeling of Radiopharmaceuticals; National Academy Press: Washington, DC, 1990. 52. Bergman, J.; Solin, O. Fluorine-18-labeled fluorine gas for synthesis of tracer molecules. Nucl. Med. Biol. 1997, 24, 677-683. 53. Preshlock, S.; Tredwell, M.; Gouverneur, V. F-18-Labeling of Arenes and Heteroarenes for Applications in Positron Emission Tomography. Chem. Rev. 2016, 116, 719-766.

54. Seo, J. W.; Lee, B. S.; Lee, S. J.; Oh, S. J.; Chi, D. Y. Fast and Easy Drying Method for the Preparation of Activated [F-18]Fluoride Using Polymer Cartridge. Bull. Korean Chem. Soc. 2011, 32, 71- 76. 55. Wessmann, S. H.; Henriksen, G.; Wester, H. J. Cryptate mediated nucleophilic F-18-fluorination without azeotropic drying. Nuklearmedizin. 2012, 51, 1-8. 56. Bailey, D. L.; Townsend, D. W.; Valk, P. E.; Maisey, M. N. Eds. Positron Emission Tomography: Basic Sciences; Springer: New York, 2005. 57. Hayashi, K.; Tachibana, A.; Tazawa, S.; Mizukawa, Y.; Osaki, K.; Morimoto, Y.; Zochi, R.; Kurahashi, M.; Aki, H.; Takahashi, K. Preparation and stability of ethanol-free solution of [18F]florbetapir ([18F]AV-45) for positron emission tomography amyloid imaging. J. Labelled Comp. Rad. 2013, 56, 295-300. 58. Graham, T. J. A.; Lambert, R. F.; Ploessl, K.; Kung, H. F.; Doyle, A. G. Enantioselective Radiosynthesis of Positron Emission Tomography (PET) Tracers Containing [F-18]Fluorohydrins. J. Am. Chem. Soc. 2014, 136, 5291-5294. 59. Yang, D. J.; Wallace, S.; Cherif, A.; Li, C.; Gretzer, M. B.; Kim, E. E.; Podoloff, D. A. Development of F-18 Labeled Fluoroerythronitroimidizole as a Pet Agent for Imaging Tumor Hypoxia. Radiology 1995, 194, 795-800. 60. Kurihara, H.; Honda, N.; Kono, Y.; Arai, Y. Radiolabelled Agents for PET Imaging of Tumor Hypoxia. Curr. Med. Chem. 2012, 19, 3282-3289. 61. Huang, X. Y.; Liu, W.; Ren, H.; Neelamegam, R.; Hooker, J. M.; Groves, J. T. Late Stage Benzylic C-H Fluorination with [F-18]Fluoride for PET Imaging. J. Am. Chem. Soc. 2014, 136, 6842- 6845. 62. Adams, D. J.; Clark, J. H. Nucleophilic routes to selectively fluorinated aromatics. Chem. Soc. Rev. 1999, 28, 225-231. 63. Irie, T.; Fukushi, K.; Inoue, O.; Yamasaki, T.; Ido, T.; Nozaki, T. Preparation of F-18-Labeled 6- Fluoro-9-Benzylpurine and 2-Fluoro-9-Benzylpurine as a Potential Brain-Scanning Agent. Int. J. Appl. Radiat. Isot. 1982, 33, 633-636. 64. Lemaire, C.; Guillaume, M.; Christiaens, L.; Palmer, A. J.; Cantineau, R. A New Route for the Synthesis of [F-18] Fluoroaromatic Substituted Amino-Acids - No Carrier Added L-P-[F- 18]Fluorophenylalanine. Appl. Radiat. Isot. 1987, 38, 1033-1038. 65. Shen, B.; Loffler, D.; Zeller, K. P.; Ubele, M.; Reischl, G.; Machulla, H. J. Decarbonylation of multi-substituted [F-18]benzaldehydes for modelling syntheses of F-18-labelled aromatic amino acids. Appl. Radiat. Isot. 2007, 65, 1227-1231. 66. Sun, H. R.; DiMagno, S. G. Competitive demethylation and substitution in N,N,N- trimethylanilinium fluorides. J. Fluorine Chem. 2007, 128, 806-812. 67. Knochel, A.; Zwernemann, O. Development of a no-carrier-added method for F-18-labelling of aromatic compounds by fluorodediazonation. J. Labelled Comp. Rad. 1996, 38, 325-336. 68. Tewson, T. J.; Welch, M. J. Preparation of F-18 Aryl Fluorides - Piperidyl Triazenes as a Source of Diazonium Salts. J. Chem. Soc. Chem. Commun. 1979, 1149-1150. 69. Satyamurthy, N.; Barrio, J. R.; Schmidt, D. G.; Kammerer, C.; Bida, G. T.; Phelps, M. E. Acid- Catalyzed Thermal-Decomposition of 1-Aryl-3,3-Dialkyltriazenes in the Presence of Nucleophiles. J. Org. Chem. 1990, 55, 4560-4564. 70. Barrio, J. R.; Satyamurthy, N.; Ku, H.; Phelps, M. E. The Acid Decomposition of 1-Aryl-3,3- Dialkyltriazenes - Mechanistic Changes as a Function of Aromatic-Substitution, Nucleophile Strength, and Solvent. J. Chem. Soc. Chem. Commun. 1983, 443-444. 71. Merritt, E. A.; Olofsson, B. Diaryliodonium Salts: A Journey from Obscurity to Fame. Angew. Chem. Int. Ed. 2009, 48 (48), 9052-9070. 72. Pike, V. W.; Aigbirhio, F. I. Reactions of Cyclotron-Produced [F-18] Fluoride with Diaryliodonium-Salts - a Novel Single-Step Route to No-Carrier-Added [(18)]Fluoroarenes. J. Chem. Soc. Chem. Commun. 1995, 2215-2216.

73. Shah, A.; Pike, V. W.; Widdowson, D. A. The synthesis of [F-18]fluoroarenes from the reaction of cyclotron-produced [F-18]fluoride ion with diaryliodonium salts. J. Chem. Soc. Perk. T.1 1998, 2043- 2046. 74. Zhang, M. R.; Kumata, K.; Suzuki, K. A practical route for synthesizing a PET ligand containing [F-18]fluorobenzene using reaction of diphenyliodonium salt with [F-18]F-. Tetrahedron Lett. 2007, 48, 8632-8635. 75. Moon, B. S.; Kil, H. S.; Park, J. H.; Kim, J. S.; Park, J.; Chi, D. Y.; Lee, B. C.; Kim, S. E. Facile aromatic radiofluorination of [18F]flumazenil from diaryliodonium salts with evaluation of their stability and selectivity. Org. Biomol. Chem. 2011, 9, 8346-8355. 76. Xu, R.; Zanotti-Fregonara, P.; Zoghbi, S. S.; Gladding, R. L.; Woock, A. E.; Innis, R. B.; Pike, V. W. Synthesis and evaluation in monkey of [(18)F]4-fluoro-N-methyl-N-(4-(6-(methylamino)pyrimidin-4- yl)thiazol-2-yl)benzami de ([(18)F]FIMX): a promising radioligand for PET imaging of brain metabotropic glutamate receptor 1 (mGluR1). J. Med. Chem. 2013, 56, 9146-55. 77. Satyamurthy, N.; Barrio, J. R. WO2010/117435 A2, 2010. 78. Hill, D. E.; Holland, J. P. Computational studies on hypervalent iodonium(III) compounds as activated precursors for F-18 radiofluorination of electron-rich arenes. Comput. Theor. Chem. 2015, 1066, 34-46. 79. Cardinale, J.; Ermert, J.; Humpert, S.; Coenen, H. H. Iodonium ylides for one-step, no-carrier- added radiofluorination of electron rich arenes, exemplified with 4-(([F-18] fluorophenoxy)-phenylmethyl) piperidine NET and SERT ligands. Rsc. Adv. 2014, 4, 17293-17299. 80. Rotstein, B. H.; Stephenson, N. A.; Vasdev, N.; Liang, S. H. Spirocyclic hypervalent iodine(III)- mediated radiofluorination of non-activated and hindered aromatics. Nat. Commun. 2014, 5, 4365-4369. 81. Sander, K.; Gendron, T.; Yiannaki, E.; Cybulska, K.; Kalber, T. L.; Lythgoe, M. F.; Arstad, E. Sulfonium Salts as Leaving Groups for Aromatic Labelling of Drug-like Small Molecules with Fluorine- 18. Sci. Rep. 2015, 5, 9941-9945 82. Chun, J. H.; Morse, C. L.; Chin, F. T.; Pike, V. W. No-carrier-added [18F]fluoroarenes from the radiofluorination of diaryl sulfoxides. Chem. Commun. 2013, 49, 2151-2153. 83. Gao, Z. H.; Lim, Y. H.; Tredwell, M.; Li, L.; Verhoog, S.; Hopkinson, M.; Kaluza, W.; Collier, T. L.; Passchier, J.; Huiban, M.; Gouverneur, V. Metal-Free Oxidative Fluorination of Phenols with [F- 18]Fluoride. Angew. Chem. Int. Ed. 2012, 51, 6733-6737. 84. Reischl, G.; Kienzle, G. J.; Machulla, H. J. Electrochemical radiofluorination: Labeling of benzene with [F-18]fluoride by nucleophilic substitution. J. Radioanal. Nucl. Chem. 2002, 254, 409-411. 85. Reischl, G.; Kienzle, G. J.; Machulla, H. J. Electrochemical radiofluorination. Part 2. Anodic monofluorination of substituted benzenes using [F-18]fluoride. Appl. Radiat. Isot. 2003, 58, 679-683. 86. He, Q.; Wang, Y.; Alfeazi, I.; Sadeghi, S. Electrochemical nucleophilic synthesis of di-tert-butyl- (4-[18F]fluoro-1,2-phenylene)-dicarbonate. Appl. Radiat. Isot. 2014, 92, 52-57. 87. Lee, E.; Kamlet, A. S.; Powers, D. C.; Neumann, C. N.; Boursalian, G. B.; Furuya, T.; Choi, D. C.; Hooker, J. M.; Ritter, T. A Fluoride-Derived Electrophilic Late-Stage Fluorination Reagent for PET Imaging. Science 2011, 334, 639-642. 88. Furuya, T.; Ritter, T. Carbon-fluorine reductive elimination from a high-valent palladium fluoride. J. Am. Chem. Soc. 2008, 130, 10060-10061. 89. Furuya, T.; Benitez, D.; Tkatchouk, E.; Strom, A. E.; Tang, P. P.; Goddard, W. A.; Ritter, T. Mechanism of C-F Reductive Elimination from Palladium(IV) Fluorides. J. Am. Chem. Soc. 2010, 132, 3793-3807. 90. Brandt, J. R.; Lee, E.; Boursalian, G. B.; Ritter, T. Mechanism of electrophilic fluorination with Pd(IV): fluoride capture and subsequent oxidative fluoride transfer. Chem. Sci. 2014, 5, 169-179. 91. Kamlet, A. S.; Neumann, C. N.; Lee, E.; Carlin, S. M.; Moseley, C. K.; Stephenson, N.; Hooker, J. M.; Ritter, T. Application of palladium-mediated (18)F-fluorination to PET radiotracer development: overcoming hurdles to translation. PLoS One 2013, 8, e59187-e59196. 92. Lee, E.; Hooker, J. M.; Ritter, T. Nickel-Mediated Oxidative Fluorination for PET with Aqueous [F-18] Fluoride. J. Am. Chem. Soc. 2012, 134, 17456-17458.

93. Ren, H.; Wey, H. Y.; Strebl, M.; Neelamegam, R.; Ritter, T.; Hooker, J. M. Synthesis and Imaging Validation of [F-18]MDL100907 Enabled by Ni-Mediated Fluorination. ACS Chem. Neurosci. 2014, 5, 611-615. 94. Tredwell, M.; Preshlock, S. M.; Taylor, N. J.; Gruber, S.; Huiban, M.; Passchier, J.; Mercier, J.; Genicot, C.; Gouverneur, V. A General Copper-Mediated Nucleophilic F-18 Fluorination of Arenes. Angew. Chem. Int. Ed. 2014, 53, 7751-7755. 95. Zlatopolskiy, B. D.; Zischler, J.; Krapf, P.; Zarrad, F.; Urusova, E. A.; Kordys, E.; Endepols, H.; Neumaier, B. Copper-Mediated Aromatic Radiofluorination Revisited: Efficient Production of PET Tracers on a Preparative Scale. Chem. Eur. J. 2015, 21, 5972-5979. 96. Cacace, F.; Wolf, A. P. Substrate Selectivity and Orientation in Aromatic-Substitution by Molecular Fluorine. J. Am. Chem. Soc. 1978, 100, 3639-3641. 97. Firnau, G.; Chirakal, R.; Garnett, E. S. Aromatic radiofluorination with [18F]fluorine gas: 6- [18F]fluoro-L-dopa. J. Nucl. Med. 1984, 25, 1228-33. 98. Chirakal, R.; Firnau, G.; Couse, J.; Garnett, E. S. Radiofluorination with F-18-Labeled Acetyl Hypofluorite - [F-18] L-6-Fluorodopa. Int. J. Appl. Radiat. Isot. 1984, 35, 651-653. 99. Visser, G. W. M.; Vonhalteren, B. W.; Herscheid, J. D. M.; Brinkman, G. A.; Hoekstra, A. Reaction of Acetyl Hypofluorite with Aromatic Mercury-Compounds - a New Selective Fluorination Method. J. Chem. Soc. Chem. Commun. 1984, 655-656. 100. Luxen, A.; Perlmutter, M.; Bida, G. T.; Vanmoffaert, G.; Cook, J. S.; Satyamurthy, N.; Phelps, M. E.; Barrio, J. R. Remote, Semiautomated Production of 6-[F-18]Fluoro-L-Dopa for Human Studies with Pet. Appl. Radiat. Isot. 1990, 41, 275-281. 101. de Vries, E. F. J.; Luurtsema, G.; Brussermann, M.; Elsinga, P. H.; Vaalburg, W. Fully automated synthesis module for the high yield one-pot preparation of 6-[F-18]fluoro-L-DOPA. Appl. Radiat. Isot. 1999, 51, 389-394. 102. Furuya, T.; Strom, A. E.; Ritter, T. Silver-mediated fluorination of functionalized aryl stannanes. J. Am. Chem. Soc. 2009, 131, 1662-1663. 103. Teare, H.; Robins, E. G.; Kirjavainen, A.; Forsback, S.; Sandford, G.; Solin, O.; Luthra, S. K.; Gouverneur, V. Radiosynthesis and Evaluation of [F-18]Selectfluor bis(triflate). Angew. Chem. Int. Ed. 2010, 49, 6821-6824. 104. Stenhagen, I. S. R.; Kirjavainen, A. K.; Forsback, S. J.; Jorgensen, C. G.; Robins, E. G.; Luthra, S. K.; Solin, O.; Gouverneur, V. [F-18]Fluorination of an arylboronic ester using [F-18]selectfluor bis(triflate): application to 6-[F-18]fluoro-L-DOPA. Chem. Commun. 2013, 49, 1386-1388. 105. Ido, T.; Irie, T.; Kasida, Y. Isotope Exchange with F-18 on Super-Conjugate System. J. Labelled Comp. Rad. 1979, 16, 153-154. 106. Prabhakaran, J.; Underwood, M. D.; Parsey, R. V.; Arango, V.; Majo, V. J.; Simpson, N. R.; Van Heertum, R.; Mann, J. J.; Kumar, J. S. D. Synthesis and in vivo evaluation of [F-18]-4-[5-(4- methylphenyl)-3-(trifluoromethyl)-1H-pyrazol-1-yl]benzenesulfonamide as a PET imaging probe for COX-2 expression. Bioorg. Med. Chem. 2007, 15, 1802-1807. 107. Verhoog, S.; Pfeifer, L.; Khotavivattana, T.; Calderwood, S.; Collier, T. L.; Wheelhouse, K.; Tredwell, M.; Gouverneur, V. Silver-Mediated 18F-Labeling of Aryl-CF3 and Aryl-CHF2 with 18F- Fluoride. Synlett. 2016, 27, 25-28. 108. Huiban, M.; Tredwell, M.; Mizuta, S.; Wan, Z. H.; Zhang, X. M.; Collier, T. L.; Gouverneur, V.; Passchier, J. A broadly applicable [F-18]trifluoromethylation of aryl and heteroaryl iodides for PET imaging. Nat. Chem. 2013, 5, 941-944. 109. van der Born, D.; Herscheid, J. D. M.; Orru, R. V. A.; Vugts, D. J. Efficient synthesis of [F-18] trifluoromethane and its application in the synthesis of PET tracers. Chem. Commun. 2013, 49, 4018-4020. 110. Zanardi, A.; Novikov, M. A.; Martin, E.; Benet-Buchholz, J.; Grushin, V. V. Direct Cupration of Fluoroform. J. Am. Chem. Soc. 2011, 133, 20901-20913. 111. Ivashkin, P.; Lemonnier, G.; Cousin, J.; Gregoire, V.; Labar, D.; Jubault, P.; Pannecoucke, X. [F- 18]CuCF3: A [F-18]Trifluoromethylating Agent for Arylboronic Acids and Aryl Iodides. Chem. Eur. J. 2014, 20, 9514-9518.

112. Khotavivattana, T.; Verhoog, S.; Tredwell, M.; Pfeifer, L.; Calderwood, S.; Wheelhouse, K.; Collier, T. L.; Gouverneur, V. F-18-Labeling of Aryl-SCF3, -OCF3 and -OCHF2 with [F-18]Fluoride. Angew. Chem. Int. Ed. 2015, 54, 9991-9995. 113. Wadsak, W.; Mitterhauser, M. Basics and principles of radiopharmaceuticals for PET/CT. Eur. J. Radiol. 2010, 73, 461-469. 114. Wolf, A. P.; Redvanly, C. S. C-11 and Radiopharmaceuticals/Cals. Int. J. Appl. Radiat. Isot. 1977, 28, 29-48. 115. Ametamey, S. M.; Honer, M.; Schubiger, P. A. Molecular imaging with PET. Chem. Rev. 2008, 108, 1501-1516. 116. Comar, D.; Maziere, M.; Crouzel, M. Synthesis and metabolism of 11C-chlorpromazine methiodide. J. Labelled Comp. Radiopharm.. 1973, 461–469. 117. Larsen, P.; Ulin, J.; Dahlstrom, K.; Jensen, M. Synthesis of [C-11]iodomethane by iodination of [C-11]methane. Appl. Radiat. Isot. 1997, 48, 153-157. 118. Link, J. M.; Krohn, K. A.; Clark, J. C. Production of [C-11]CH3I by single pass reaction of [C- 11]CH4 with I-2. Nucl. Med. Biol. 1997, 24, 93-97. 119. Jewett, D. M. A Simple Synthesis of [C-11] Methyl Triflate. Appl. Radiat. Isot. 1992, 43, 1383- 1385. 120. Sandell, J.; Halldin, C.; Sovago, J.; Chou, Y. H.; Gulyas, B.; Yu, M. X.; Emond, P.; Nagren, K.; Guilloteau, D.; Farde, L. PET examination of [C-11]5-methyl-6-nitroquipazine, a radioligand for visualization of the serotonin transporter. Nucl. Med. Biol. 2002, 29, 651-656. 121. Iwata, R.; Ido, T.; Takahashi, T.; Nakanishi, H.; Iida, S. Optimization of [11C]HCN production and no-carrier-added [1-11C]amino acid synthesis. Int. J. Rad. Appl. Instrum. A. 1987, 38, 97-102. 122. Oi, N.; Tokunaga, M.; Suzuki, M.; Nagai, Y.; Nakatani, Y.; Yamamoto, N.; Maeda, J.; Minamimoto, T.; Zhang, M. R.; Suhara, T.; Higuchi, M. Development of Novel PET Probes for Central 2-Amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic Acid Receptors. J. Med. Chem. 2015, 58, 8444- 8462. 123. Lindhe, O.; Almqvist, P.; Kagedal, M.; Gustafsson, S. A.; Bergstrom, M.; Nilsson, D.; Antoni, G. Autoradiographic Mapping of 5-HT(1B/1D) Binding Sites in the Rhesus Monkey Brain Using [carbonyl- C]zolmitriptan. Int. J. Mol. Imaging 2011, 2011, 694179-694184. 124. Almqvist, P.; Lindhe, O.; Kagedal, M.; Gustavsson, S. A.; Bergstrom, M.; Nilsson, D.; Antoni, G. Autoradiographic mapping of [C-11] zolmitriptan binding sites in Rhesus monkey brain. Neurology 2007, 68, A210. 125. Roeda, D.; Dolle, F. [C-11]Phosgene: A Versatile Reagent for Radioactive Carbonyl Insertion Into Medicinal Radiotracers for Positron Emission Tomography. Curr. Top. Med. Chem. 2010, 10, 1680- 1700. 126. Wang, L.; Yui, J.; Wang, Q. F.; Zhang, Y. D.; Mori, W.; Shimoda, Y.; Fujinaga, M.; Kumata, K.; Yamasaki, T.; Hatori, A.; Rotstein, B. H.; Collier, T. L.; Ran, C. Z.; Vasdev, N.; Zhang, M. R.; Liang, S. H. Synthesis and Preliminary PET Imaging Studies of a FAAH Radiotracer ([C-11]MPPO) Based on alpha-Ketoheterocyclic Scaffold. ACS Chem. Neurosci. 2016, 7, 109-118. 127. Rotstein, B. H.; Liang, S. H.; Holland, J. P.; Collier, T. L.; Hooker, J. M.; Wilson, A. A.; Vasdev, N. (CO2)-C-11 fixation: a renaissance in PET radiochemistry. Chem. Commun. 2013, 49, 5621-5629. 128. Hicks, J. W.; Parkes, J.; Sadovski, O.; Tong, J. C.; Houle, S.; Vasdev, N.; Wilson, A. A. Synthesis and preclinical evaluation of [C-11-carbonyl]PF-04457845 for neuroimaging of fatty acid amide hydrolase. Nucl. Med. Biol. 2013, 40, 740-746. 129. Hicks, J. W.; Wilson, A. A.; Rubie, E. A.; Woodgett, J. R.; Houle, S.; Vasdev, N. Towards the preparation of radiolabeled 1-aryl-3-benzyl ureas: Radiosynthesis of [(11)C-carbonyl] AR-A014418 by [(11)C]CO(2) fixation. Bioorg. Med. Chem. Lett. 2012, 22, 2099-2101.

Chapter 2: Radiosynthesis and preliminary PET evaluation of [18F]2-11 as AMPA radiotracer

2.1 Background

As a principle excitatory neurotransmitter in the brain, glutamate has been recognized since late 1970s.1 Its physiologic effects mainly function through the two major families of receptor proteins: metabotropic and ionotropic glutamate receptors (mGluRs and iGluRs, respectively).2

Figure 2-1 provides an overview of the glutamate receptor family tree, including mGluRs and iGluRs, as well as their known subunits and signal transduction mechanisms.3

Figure 2-1. Overview of glutamate receptor family.

Reproduced with permission from Springer3

Glutamate regulates the activation of phospholipase C (PLC) and inhibition of adenylate cyclase (AC) activity through cation-permeability controlling iGluRs and G-protein coupled

53 mGluRs.3 The mGluRs belong to the group C family of G-protein-coupled receptors, and were found to regulate glutamate-rendered cell excitatory through indirect metabotropic processes in

1980s, either by releasing secondary messengers in the cytoplasm or by releasing G protein subunits within the membrane to affect the ion channels.4-6

Structurally, mGluRs share common morphology as that of other G protein-linked metabotropic receptors with seven transmembrane domains that span the cell membrane, an extracellular N-terminal and intracellular COOH terminal.4, 7 They are divided into three groups with eight subtypes (mGluR 1-8) based on their structure, pharmacology and signal transduction mechanism. The mGluRs are distributed in pre- and postsynaptic neurons in the brain and in peripheral tissues.8-10 The sensitivity of different receptors to glutamate is related to their relative locations on the cell membrane to the synaptic cleft.8-10 Abnormalities in mGluRs expression contribute to various diseases, such as Parkinson’s disease, generalized anxiety disorder and autism etc.11-13

The iGluRs were named after their preferred agonists (Figure 2-2),3, 14-16 namely, N-methyl-

D-aspartate (NMDA) receptors, 2-carboxy-3-carboxymethyl-4-isopropenylpyrrolidine (kainate) receptors, and α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid (AMPA) receptors.

Figure 2-2. Structures of iGluRs natural binding ligands.

Unlike mGluRs, iGluRs structurally incorporate ligand-gated tetrametric ion channels that are permeable to cations to allow fast synaptic responses (millisecond time scale) to the glutamate, though their Na+ and Ca++ permeability varies, depending on individual family and subtype composition (Figure 2-3).15, 17 For example, AMPA receptors that lack the GluR2 (also known as GluRB) subunit, which is important in determining the Ca++ permeability of the heteromeric receptors, are Ca++ permeable and display inward rectification due to a voltage- dependent block of the ion channel by intracellular polyamines.18

Figure 2-3. Schematic excitatory synapse and functions of iGluRs.

Reproduced from reference 17, copyright of Wiley.

The iGluRs mediate most of the excitatory postsynaptic neurotransmission in CNS and regulate synaptic plasticity and neurodevelopment, playing key roles in memory, learning, cognition and the differentiation and growth of the nervous system.15-16 Their diverse functional properties also come from the fact that their intracellular carboxyl terminal is capable of

55 interacting with various intracellular proteins, which involve not only in the spatial and functional organization of postsynaptic densities but also signal transduction.1, 15 For example,

AMPA could activate the mitogen-activated protein kinase pathway via a protein tyrosine kinase,

Lyn.19

Abnormal activation of iGluRs by endogenous glutamate causes excessive intracellular calcium influx and ultimately leads to the neuronal cell damage and death in a process called excitotoxicity.20-21 Therefore, antagonists of these receptors are considered as therapeutic agents for the treatment of psychiatric disorders and neurodegenerative diseases, such as epilepsy, cerebral ischemia, stroke, Parkinson’s disease and amyotrophic lateral sclerosis.22-24

Explorations of NMDA receptor antagonists have, however, been hampered by the severe psychotomimetic adverse effects that occur before these drugs reach a sufficient concentration to block the pathological actions in the brain. 25-27 For example, blockade of the ion channel by the noncompetitive NMDA antagonists, aptiganel and phencyclidine (Figure 2-4), and competitive inhibition of the glutamate binding site via midafotel and selfotel led to significant psychiatric adverse effects such as sedation, panronid and hallucinations in patients with stroke.25-26 The noncompetitive open-channel blocking agent memantine (Figure 2-4) was approved by the U.S.

FDA to treat moderate-to-severe Alzheimer's disease,27 but now has been recommended by the

UK's National Institute for Health and Care Excellence in 2011 to only use for patients with severe Alzheimer’s disease or who are intolerant of acetylcholinesterase (AChE) inhibitors due to its adverse effects, such as agitation and hallucinations.25

Figure 2-4. Chemical structures of NMDA receptor antagonists.

Kinate receptors have both presynaptic and postsynaptic physiology/pathophysiology effects: presynaptic kainate receptors inhibit neurotransmission via the inhibitory neurotransmitter gamma-aminobutyric acid (GABA); postsynaptic kainate receptors stimulate excitatory neurotransmission.28-29 Development of kinate receptors binding ligands were not successful due to the lack of the selective ligands and its more limited distributions in the brain than NMDA and AMPA receptors.30 Among the subtypes of kinate receptors, GluR5 receptor selective agonists and antagonists had been identified, such as (S)-5-iodowillardiine as GluR5 agonsit and LY382884 and NS3763 as GluR5 non-competitive antagonists (Figure 2-5), but had not shown promising results in clinical use.31 Development of selective binding ligands for kainate receptors is ongoing and may enable the understanding of the receptor and the kainate receptor-mediated pharmacological activity. Currently, no kainate antagonist has reached the market yet.

Figure 2-5. Chemical structures of kainate receptor selective binding ligands.

2.2 AMPA receptor as a drug discovery target

The crystal structure of the AMPA receptor, rat GluA2 receptor, was elucidated in 2009 as a homotetramer in complex with a competitive antagonist ZK 200775 with a resolution of 3.6 Å.32

The disclosed AMPA crystal structure had a two-fold symmetry at the extracellular domains with pairs of local dimers, a four-fold symmetry at the ion channel domain, and a mismatched symmetry between the extracellular and ion channel domains that renders two pairs of diagonally related and conformationally distinct subunits, A (green color)/C (blue color) and B (red color)/D

(yellow color), to couple the ion channel gate in different manners (Figure 2-6).17, 32 This architecture was also found similar to GluN1 and GluN2 NMDA receptors.32

Figure 2-6. The architecture of rat GluA2 receptor in a “broad” view.

Reproduced with permission from Springer Nature.32 The structure is perpendicular to the overall two-fold of axis of molecular symmetry. Each unit is in a different color.

The binding sites and action mechanism of competitive and noncompetitive AMPA antagonists are elucidated in Figure 2-7 through a domain structure of a single AMPA receptor.17, 32 Besides the amino- and carboxy-terminal domains, domains 1 and 2 (D1 and D2) are composed of the competitive antagonists binding core; Strands 1 and 2 (S1 and S2)

59 connecting to the membrane-spanning regions 1 and 4 (M1 and M4) separately (i.e. S1-M1 and

S2-M4) form the linker segments as noncompetitive antagonists binding site; in the transmembrane region, M1, M2 and M3 build up an ion channel and the M3 segment lines the pore loop M2. Activation of AMPA receptor via either glutamate or other agonists at the competitive binding core starts from the ligand binding to D1 domain, followed by the movement of the flexible D2 domain toward D1 to pull apart the linker segments toward the transmembrane channel and lead to the opening of the ion channel.17 Alternatively, binding of noncompetitive AMPA antagonists at the linker domains would stabilized the linkers in a configuration to prevent the agonist induced gating of the ion channel.17

Figure 2-7. Illustration of AMPA receptor architecture via domain structures.

Reproduced from reference 17, copyright of Wiley.

As therapeutic targets, AMPA receptors had been proven to involve in excitotoxic conditions. For example, Figure 2-8 shows the mechanism of epileptogenesis and potential therapeutic intervention via AMPA antagonists.33 After a seizure-inducing event happens, such as a brain trauma, the resulting seizure leads to incorporation of GluA2-lacking AMPA receptors to the synapse, which will enhance the synaptic strength and show a significant Ca++ conductance. If no further synaptic changes occur after this process, it means there will be no epileptogenesis in the brain. One the other hand, in cases where further incorporation of GluA2- lacking AMPA receptors occurs, it will lead to epileptogenesis. Pharmacological intervention via

AMPA receptors antagonists is hypothesized to prevent or delay the disease progression at this point.33

Figure 2-8. Mechanism of epileptogenesis and potential therapeutic intervention via AMPA antagonists.

Reproduced from reference 33, copyright of Dove Press.

Development of AMPA receptor binding ligands have been the focus of drug discovery targeting the iGluRs since the first reported competitive AMPA receptor antagonist 2,3- dihydroxy-6-nitro-7-sulfamoyl-benzo[f]quinoxaline-2, 3-dione (NBQX) in 1991 (Figure 2-9).34

Subsequent efforts in improving the aqueous solubility and BBB penetration of such quinoxalinedione derivatives led to fanapanel (ZK200775),35 becampanel (AMP397),36 tezampanel (LY293558)37 and selurampanel (BGG492).38 In this category, becampanel

(AMP397) and selurampanel (BGG492) entered the clinical trials for the treatment of epilepsy by Novartis Co., but no competitive AMPA antagonist has yet reached the market due to their side effects.39-40

Figure 2-9. Chemical structures of competitive AMPA receptor antagonists.

Meanwhile, noncompetitive AMPA receptor antagonists are expected to lead more successful drugs and avoid the side effects due to the fact that they occupy the less polar allosteric binding site and are not supposed to influence significantly the normal glutamatergic

62 activity.41-42 In this category, the 2,3-benzodiazepine derivatives were quite well-known with

GYKI-53655 and talampanel (GYKI-537773) as representative compounds (Figure 2-10).41, 43

Currently, talampanel was studied in a Phase II clinical trial as an antiepileptic agent. Analogs in this class of compounds with anticonvulsant activity also include CP465022,44 YM928,45 irampanel (BIIR561)46 and perampanel (Fycompa®) (Figure 2-10).47-48 Irampanel was evaluated in Phase I/IIa clinical trials for the treatment of stroke.46 Perampanel has been approved by the

FDA and other regulatory agencies for treating refractory partial onset seizures in 2012, 49 and as an adjunctive treatment for primary generalized tonic-clonic seizures in 2015.50

Figure 2-10. Chemical structures of non-competitive AMPA receptor antagonists.

2.3 AMPA PET radiotracers

Non-invasive PET imaging of AMPA in the living brain would enable mechanistic investigations of the AMPA system under normal and disease conditions, assessment of AMPA

63 distribution in the brain and periphery, and target engagement for validation of promising drug candidates in clinical trials. However, PET imaging for AMPA has remained largely unexplored due to the lack of a selective PET imaging probe.51-54 Currently, only a handful of AMPA PET radiotracers, most of which are 11C-labeled molecules, have been reported (Figure 2-11). An array of carbon-11 and fluorine-18 labeled N-acetyl-1-aryl-6,7-dimethoxy-1,2,3,4- tetrahydroisoquinoline derivatives [11C]2-1 to 2-6 and [18F]2-7 were synthesized by Gao et al. as potential AMPA receptor PET tracers (Scheme 2-1).51

Scheme 2-1. Radiosynthesis of the isoquinoline derivatives as potential PET tracers.

11 11 The carbon-11 labeled products [ C]2-1 to 2-6 were synthesized by [ C]CH3OTf methylation based on their corresponding phenolic precursors.52 The only fluorine-18 labeled

18 18 compound [ F]2-7 was achieved via SNAr nucleophilic reaction with [ F]KF/K222 with its NO2 analogue as precursor.52 The carbon-11 radiolabeled [11C]2-1 to 2-6 were synthesized with 30-

11 45% RCC based on [ C]CO2 from end of bombardment (EOB), and > 37 GBq/μmol specific

65 activity at the end of synthesis (EOS, synthetic time 15-20 min). The fluorine-18 labeled [18F]2-7 was obtained in 15-25% RCC at EOB and 37-44.4 GBq/μmol specific activity at EOS.

Compounds 2-1 to 2-3 and 2-6 were all noncompetitive AMPA receptor antagonists, of which compound 2-1 was reported to be a more potent noncompetitive AMPA antagonist than talampanel.55 Therefore, [11C]2-1 was evaluated in an in vivo PET imaging study, however, this radioligand turned out not to be a suitable PET tracer for AMPA receptors due to its rapid clearance from rat brain and poor selective binding at AMPA receptors.52

Recently, the carbon-11 labeled perampanel was achieved based on its aryl-bromide precursor through an in situ prepared LPd(Ar)X complex (L = biaryl phosphine) based on aryl

(pseudo)halides to allow insertion of [11C]CN- and subsequent reductive elimination (Scheme 2-

2).53 The [11C]perampanel could be furnished in 1 min at room temperature with a RCC of 32 ±

5% (n = 3). However, the specific activity and in vivo performance of this radiotracer has not been determined.

Scheme 2-2. Radiosynthesis of [11C]perampanel using biaryl phosphine Pd(0) complex.

After the development of perampanel, Eisai Co. further explored a novel class of potent noncompetitive AMPA receptor antagonists.54 Compounds 2-8 to 2-12 in Scheme 2-3 exemplified the antagonists in this class and were radiolabeled and evaluated by in vivo PET imaging in rats and monkeys to determine their potential as PET tracers for AMPA receptors.54

[11C]2-8, [11C]2-10 and [11C]2-11, were radiosynthesized via [11C]CuCN cyanation with the aryl- halide substrates and subsequent TFA deprotection in the case of BOC protected substrates.

11 11 18 [ C]2-9 was obtained via [ C]CH3OTf rendered methylation. [ F]2-12 was furnished through

18 54 SN2 nucleophilic reaction with [ F]KF/K222 based on its OTs activated precursor.

Scheme 2-3. Novel PET radiotracers based on perampanel derivatives.

[11C]2-8 and [11C]2-10 exhibited very low brain penetration in rhesus monkey (i.e.

18 standardized uptake value (SUV)max ≤0.5, Figure 2-11), and [ F]2-12 showed inconsistent distribution with the localization of AMPA receptors in rat brain, probably due to the decomposition of this tracer, according to the author. Whereas, [11C]2-9 displayed good radioactivity uptakes in both rat and monkey brains (i.e. SUVmax = 1.6) and consistent binding patterns to the AMPA localization as well as target specific binding features. The fluorine bearing AMPA radiotracer [11C]2-11 displayed even higher radioactivity uptake in rhesus

11 54 monkey brain (i.e. SUVmax = 2.7) as that of [ C]2-9.

Figure 2-11. PET baseline scan results of a rhesus monkey with [11C]2-8 to [11C]2-11.

[11C]2-8 [11C]2-9 [11C]2-10 [11C]2-11

Reproduced with permission from American Chemical Society.54

In addition, both compounds 2-9 and 2-11 are potent noncompetitive AMPA antagonists (i.e.

Ki = 10 nM and 20 nM respectively). They produced high contrast in vitro autoradiography images in rat brain slices at the cortex and hippocampal regions compared with the baseline.

These two carbon-11 labeled tracers also showed consistent radioactivity distribution as that of the localization of AMPA receptors in both rats and rhesus monkey brains.54 In combination with the advantages of fluorine-18 isotope that are mainly attributed to its longer half-life (t1/2 = 109.7 min) than carbon-11 (t1/2 = 20.4 min) as mentioned in the introduction chapter, the goal of the present work is to develop a practical method to radiolabel the aryl-fluoride in compound 2-11 to give the 18F-labeled isotopologue [18F]2-11 as a potential PET tracer for imaging AMPA receptors.

2.4 Radiosynthesis of [18F]2-11 as AMPA radiotracer

As discussed in the first chapter, conventional methods for the introduction of [18F]fluorine onto arenes and heteroarenes require strong electron-withdrawing group ortho or para to the displacing position. Though the meta-position of the electron-withdrawing group is not favorable, compound 2-12 bearing a strong electron-withdrawing group (-NO2) was still synthesized and initially used as a precursor for the radiosynthesis of [18F]2-11. Synthesis of the

1,3,5-trisubstituted 2(1H)-pyridone derivative 2-12 was achieved following the reported method as shown in Scheme 2-4.54, 56 Suzuki−Miyaura coupling reaction between 2-13 and 2-14 led to compound 2-15, which was then demethylated under acidic reaction conditions to get compound

2-16. Iodination of 2-16 allowed the following coupling reaction of 2-17 with (3- nitrophenyl)boronic acid under modified Ullmann conditions to afford compound 2-18. A second

Suzuki−Miyaura coupling of compound 2-18 with boronate ester (i.e. 2-(1,3,2-dioxaborinan-2- yl)benzonitrile) gave the desired product 2-12.

Scheme 2-4. Synthesis of compound 2-12.

Meanwhile, synthesis of compound 2-11 was completed to provide a standard reference compound in the radiosynthesis (Scheme 2-5).54 Following a similar synthetic strategy, the meta- nitrophenyl boronic acid was replaced by meta-fluorophenyl boronic acid (i.e. (3- fluorophenyl)boronic acid) to give 2-19, and its subsequent coupling with 2-17 led to the desired product 2-11.

Scheme 2-5. Synthesis of the final compound 2-11.

The obtained nitro-analogue 2-12 was subsequently applied to the radiosynthesis as shown

18 18 in Scheme 2-6. The typical [ F]fluoride nucleophile is in the forms of [ F]KF/K222 and

[18F]TEAF for this reaction. Though various reaction conditions were attempted with varying combinations of temperature (80°C, 100°C, 120°C and 150°C), solvents (MeCN, DMF and

DMSO), and reagent stoichiometry, no desired product [18F]2-11 was formed.

Scheme 2-6. Attempted radiosynthesis of [18F]2-11.

These results led us to use our recently developed strategy for the radiofluorination of non- activated arenes based on the spirocyclic hypervalent iodine(III)-mediated radiofluorination, which was also introduced in Chapter 1.57-61 It requires the synthesis of the spirocyclic hypervalent iodine(III) intermediates as radiofluorination precursors, which normally could be obtained through the oxidation of corresponding (hetero)aryl iodides and subsequent condensation with a suitable spirocyclic auxiliary, such as (1r,3r,5r,7r)-spiro[adamantane-2,2'-

[1,3]dioxane]-4',6'-dione (SPIAD).57-61 This method has been proven to be efficient for radiolabelling both non-activated and hindered aromatics and has been applied to a diverse range of substrates. Following this strategy, the method in Scheme 2-7 was proposed, and radiosynthesis of [18F]2-11 could thus be achieved in one step from the iodine(III) precursor 2-

20.

Scheme 2-7. Proposed synthesis of [18F]2-11 from its iodine(III) precursor 2-20.

The iodide derivative 2-21 could be furnished via sequential diazotization-iodination reactions based on the corresponding aryl-NH2 analog of 2-11. Initial hydrogenation of compound 2-12 via Pd/C and hydrogen gas, however, failed to provide the aryl-iodide 2-21, leaving only the starting material 2-12. Alternatively, the aryl-NHBoc analog of 2-11 was synthesized (i.e. 2-24) and subsequent Boc-deprotection yielded compound 2-25, allowing the diazotization-iodination reactions to get the desired aryl-iodide analog 2-21 (Scheme 2-8).

Scheme 2-8. Synthesis of the aryl-iodide analogue 2-21.

Once the aryl-idodide 2-21 was obtained, it was subjected to various oxidative reaction conditions, such as those reported in the original paper,58 to produce the spirocyclic hypervalent iodine(III) precursor 2-20. Unfortunately, no desired product 2-20 was formed even at the mildest oxidation reaction conditions (i.e. Selectfluor/TMSOAc), and the aryl-idodide 2-21 decomposed to numerous side products. This is probably due to the presence of pyridine, piperazine and cyano groups in 2-21, which are quite labile under oxidative reaction conditions.

To realize the radiosynthesis of [18F]2-11, a two-step synthetic method was proposed as shown in Scheme 2-9, where the simple and relatively inert 1-bromo-3-iodobenzene 2-27 was first oxidized with suitable oxidants and then condensed with SPIAD to give the hypervalent iodine(III) species 2-29. The spirocyclic hypervalent iodine(III) auxiliary was finally replaced by the nucleophilic [18F]fluoride to give [18F]2-30. Subsequent coupling of [18F]2-30 with the core structure of 2-31 would give the final radiolabeled product [18F]2-11.

Scheme 2-9. Proposed two-step radiosynthesis of [18F]2-11.

To prepare the precursor 2-28 for radiofluorination, aryl iodide 2-26 was oxidized with

NaBO34H2O in glacial acetic acid to give the intermediate 2-27, followed with condensation with SPIAD in basic conditions to afford the desired hypervalent spirocyclic iodine(III) 2-28

(Scheme 2-10).58

Scheme 2-10. Synthesis of the iodine(III) precursor 2-28.

The coupling counterpart 2-30 was synthesized via the coupling reactions between N-Boc- protected 2-31, derived from compound 2-17, and 2-cyanophenylboronic acid 1, 3-propanediol ester under modified Ullmann reactions (Scheme 2-11).

Scheme 2-11. Synthesis of compound 2-30.

Before carrying out the actual radiolabeling, this method was first tested in a ‘cold’ coupling reaction between 1-bromo-3-fluorobenzene 2-29 and the coupling counterpart 2-30 under copper(I) iodide-mediated reaction conditions (Scheme 2-12). The reaction was carried out on a scale to mimic the actual amounts that would be used in the radiolabeling settings, excepting that of compound 2-29 (86.4 μmol), which otherwise would be in the pico- or nanomolar scale in radiosynthesis. In the reaction conditions of compound 2-30 (72 μmol), CuI (94.6 μmol), K3PO4

(216 μmol), trans-N, N'-dimethylcyclohexane-1, 2-diamine (72 μmol) and DMF (2 mL), the product 2-11 could be achieved with a yield of 40%.

Scheme 2-12. Novel method for the synthesis of compound 2-11.

With this encouraging result, the radiosynthesis of [18F]2-11 commenced from the optimization of the first step radiosynthesis of [18F]2-29 (Scheme 2-13).

Scheme 2-13. Radiosynthesis of [18F]2-29.

Optimization of the radiofluorination of the iodine(III) precursor 2-28, was carried out to

18 18 18 yield [ F]2-29 (Table 2-1). Herein, the typical agents [ F]KF/K222 and [ F]TEAF were used to

18 provide nucleophilic [ F]fluoride for the SNAr replacement.

Table 2-1. Optimizations for the radiosynthesis of [18F]2-29.

a Entry 2-28 (mg) Base (mg) K222 (mg) T (°C) t (min) DMF (mL) RCC (%) 1 (n = 3) 2 TEAB (1.0) N 80 10 0.2 25 ± 2 2 (n = 3) 2 TEAB (1.0) N 100 10 0.2 41 ± 3 3 (n = 3) 2 TEAB (1.0) N 120 10 0.2 69 ± 2

4 (n > 10) 2 K2CO3 (1.0) 5.0 120 10 0.2 72 ± 3

5 (n = 3) 2 K2CO3 (1.5) 7.5 120 10 0.2 46 ± 2

6 (n = 3) 2 K2CO3 (2.0) 10.0 120 10 0.2 15 ± 3

7 (n = 3) 2 K2CO3 (1.0) 5.0 120 5 0.2 50 ± 2

8 (n = 3) 2 K2CO3 (1.0) 5.0 120 15 0.2 45 ± 2

9 (n = 3) 3 K2CO3 (1.0) 5.0 120 10 0.2 60 ± 1

10 (n = 3) 4 K2CO3 (1.0) 5.0 120 10 0.2 48 ± 2

aDetermined by radio-TLC integration of product peak relative to [18F]fluoride and 18F-side products if observed.

Optimization of the reaction conditions for TEAB (temperature) and K2CO3/K222 (reaction time, stoichiometries of compound 2-28 and K2CO3/K222) systems led to the desired compound

[18F]15 with a optimal RCC of 72 ± 3% (n > 10, entry 4), where compound 13 (3.9 μmol),

K2CO3 (7.2 μmol), and K222 (13.3 μmol) were heated for 10 min at 120 °C in N,N-

77 dimethylmethanamide (DMF, 0.2 mL). Similarly, under optimal reaction conditions, the TEAF system also gave a comparable RCC of 69 ± 2% (n = 3, entry 3).

Considering the decay of fluorine-18 radioisotope and the safety of handling minimal starting activity, we desired to accomplish the radiosynthesis of [18F]2-11 via a one-pot manner from 2-28. During the radiosynthesis, aliquots of the first step reaction mixture under the optimal reaction condition (Table 2-1, entry 4) were subjected to the reactions and optimizations of the second step radiosynthesis for [18F]2-11. The second step radiolabeling reaction is shown in

Scheme 2-14.

Scheme 2-14. Radiosynthesis of compound [18F]2-11.

Screening of other amines (Table 2-2, entry 1-4), as base and copper chelating reagents, including N,N,N′,N′-tetramethylethylenediamine (TMEDA), N,N′-dimethylethylenediamine

(DMEDA) failed to produce [18F]2-11. It was also found that inclusion of TEAB from the first

18 reaction step led to no desired [ F]2-11 (Table 2-2, entry 2). Therefore, the K2CO3/K222 system was used for the first step radiosynthesis of [18F]2-29 and the trans-N,N′-dimethylcyclohexane-1,

2-diamine was utilized in the following coupling reaction to get [18F]2-11. Further optimization of reaction temperature (Table 2-2, entries 5-6), stoichiometry of the coupling counterpart 2-30

(Table 2-2, entries 7-9), copper(I) iodide and its chelating amine (Table 2, entries 10-12), as well

78 as the amount of K3PO4 additive (Table 2-2, entries 13-15) were carried out, resulting in the optimal RCC of 72 ± 5% (n = 5, Table 2-2, entry 12).

Table 2-2. Optimizations for the one-pot radiosynthesis of [18F]2-11.

a 18 c Entry 2-30 (mg) CuI (mg) Amine (µL) K3PO4 (mg) T(°C) Time (min) DMF (mL) [ F]2-11 (%) 1 (n = 2) 2 2 1.2b 4.8 110 10 0.4 8 ± 4 2 (n = 2) 2 2 TEAB (2.0 mg) 4.8 110 10 0.4 NR 3 (n = 2) 2 2 TMEDA (1.2) 4.8 110 10 0.4 NR 4 (n = 2) 2 2 DMEDA (1.2) 4.8 110 10 0.4 NR 5 (n = 2) 2 2 1.2 4.8 130 10 0.4 16 ± 3 6 (n = 2) 2 2 1.2 4.8 150 10 0.4 NR 7 (n = 2) 4 2 1.2 4.8 130 10 0.4 20 ± 3 8 (n = 2) 8 2 1.2 4.8 130 10 0.4 10 ± 2 9 (n = 2) 16 2 1.2 4.8 130 10 0.4 6 ± 3 10 (n = 2) 4 4 2.4 4.8 130 10 0.4 48 ± 5 11 (n = 2) 4 8 4.8 4.8 130 10 0.4 70 ± 3 12 (n = 5) 4 12 7.2 4.8 130 10 0.4 72 ± 5 13 (n = 2) 4 12 7.2 2 130 10 0.4 60 ± 3 14 (n = 2) 4 12 7.2 10 130 10 0.4 60 ± 3 15 (n = 2) 4 12 7.2 15 130 10 0.4 20 ± 1 aThe first step reaction mixture was from Table 3-1, entry 4. bAmine refers to trans-N, N′-dimethylcyclohexane-1, 2-diamine. cDetermined by radio-HPLC integration of product peaks, relative to [18F]fluoride and 18F-side products if observed.

To our surprise, when the above optimized reaction conditions were employed for the scale- up synthesis of [18F]2-11 starting with 0.74 Gbq 18F-fluoride, the resulting [18F]2-11 had a much lower isolated radiochemical yield of RCY < 3% (non-decay-corrected) than the theoretical RCY of 50%. This was attributed to the small amounts of K2CO3/K222 (1 mg/5 mg) used in the first step.

In the scaled-up synthesis of [18F]2-11, the entire first step reaction mixture, instead of aliquots (1/4) of the mixture, would be dosed to the second step reaction, since it is not appropriate to start with large amount of activity but only take a portion of it for the subsequent reaction. The previous optimal reaction conditions (Table 2-1, entry 4, K2CO3/K222 (1 mg/5 mg))

18 18 led to inefficient extraction of [ F]fluoride from H2 O via a QMA cartridge (i.e. 60% recovery yield compared with the normal yield of > 95%) as well as the incomplete formation of

18 [ F]KF/K222 complex during azeotropic drying (i.e. only 20% of the activity was transferred to the first step reaction vial compared to the normal percentage of > 50%). To solve these problems, several reported methods were explored, including using smaller anion exchange resins, eluting with K2CO3/MeOH, and enhancing the concentration of base by reducing the amount of eluting solvent, but none of them worked well in this case.62

To improve the RCY of [18F]2-11, the amounts of reagents used in the first step were optimized again. As shown in Table 2-3, scale up of all the reagents by three-fold without changing their concentrations would keep the RCC (68 ± 3%, n > 5, Table 3, entry 3) similar to the previous one (RCC = 72 ± 3%, n > 10, Table 1, entry 4). Meanwhile, the overall recovery

18 18 yield from the [ F]fluoride to the [ F]KF/K222 transferred to the first reaction mixture was increased from around 12% to roughly 50%, thus, reducing the starting activity needed for the in vivo animal study.

Table 2-3. Scale up for the radiosynthesis of [18F]2-29.

a Entry 2-28 (mg) K2CO3 (mg) K222 (mg) T (°C) Time (min) DMF (mL) RCC (%) 1 (n = 3) 3 1.5 7.5 120 10 0.3 45 ± 3 2 (n = 3) 4 2.0 10.0 120 10 0.4 55 ± 3 3 (n > 5) 6 3.0 15.0 120 10 0.6 68 ± 3

aDetermined by radio-TLC integration of product peak, relative to [18F]fluoride and 18F-side products if observed.

Subsequent optimization of the second step, based on the addition of the entire reaction mixture from table 2-3, entry 3, also required a 3-fold scale up for the reagents as well as a prolonged reaction time of 20 min. to keep similar RCCs (65 ± 10%, Table 2-4, entry 3). As a result, the corresponding isolated RCY of [18F]2-11 relative to starting 18F-fluoride (0.74 Gbq) was significantly improved to 10 ± 2% (non-decay corrected, n = 5) after a 60 min synthesis time.

Table 2-4. Scale up for the radiosynthesis of [18F]2-11.

18 2-30 CuI Amine K3PO4 Time DMF [ F]2-11 Entry T(°C) (mg) (mg) (µL)a (mg) (min) (mL) (%)b 1 (n = 2) 4 12 7.2 4.8 130 10 1.0 30 ± 5 2 (n = 2) 12 36 21.6 14.4 130 10 1.0 45 ± 2 3 (n = 5) 12 36 21.6 14.4 130 20 1.0 65 ± 10

aAmine refers to trans-N, N′-dimethylcyclohexane-1, 2-diamine. bDetermined by radio-HPLC integration of product peak, relative to [18F]fluoride and 18F-side products if observed.

2.5 Preliminary PET imaging evaluation of [18F]2-11

To determine the in vivo distribution and brain permeability of [18F]2-11, preliminary PET imaging studies were performed in mice. Based on the optimal reaction conditions mentioned above, the one-pot, two-step radiosynthesis of [18F]2-11 was carried out as follows: Fluorine-18

(0.93-1.1 GBq) was separated from 18O-enriched water using ion exchange cartridges and subsequently released into a V-shaped vial via a solution of K2CO3/K222 (K2CO3 (3 mg) and K222

(15 mg) in a 1 mL solution of MeCN/H2O, (v/v 7:3)). The solution was azeotropically dried 3 cycles with addition of anhydrous MeCN (1 mL each cycle) at 95 °C and resolubilized in anhydrous DMF (0.6 mL). This solution was transferred into a vial containing compound 2-28 (6 mg). The reaction was heated to 120 °C for 10 min, after which K3PO4 (14.4 mg) and a solution of compound 2-30 (12 mg), CuI (36 mg) and trans-N,N′-dimethylcyclohexane-1, 2-diamine

(21.6 μL) in DMF (0.4 mL) were added at room temperature. The reaction was allowed to react for 20 min at 130 °C. Then, the reaction mixture was quenched with water (10 mL) and filtered through a syringe filter connected to a t-C18 plus Sep-Pak® cartridge, (Waters; pre-activated with

18 10 mL EtOH followed by 10 mL H2O) to remove the precipitates and trap the [ F]2-11.

Subsequently, 1.5 mL MeCN was pushed through the cartridge to elute off the [18F]2-11, to which NaOAc (3 mL 0.01 M) aqueous solution was added to reformulated it into 4.5 mL 33%

MeCN in 0.01 M NaOAc solution before semi-preparative HPLC purification with 40% MeCN in 0.01 M NaOAc to get the [18F]2-11 in 24 min.

The obtained [18F]2-11 was reformulated into 10% ethanolic saline before administrating to mice. The 0.015 GBq activity, at the time of injection, featured excellent chemical and radiochemical purity (> 99%), and specific activity of 29.6 ± 7.4 GBq/μmol (n = 3) (Figure 2-12

& 2-13).

Figure 2-12. HPLC analysis of reformulated [18F]2-11 in 10% EtOH/saline (v/v).

Impurities from 10% EtOH/saline Compound 2-11

254 nm 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 Time (min)

Figure 2-13. Co-injection of [18F]2-11 with its unlabeled isotopologue.

[18F]2-11

Compound 2-11 3-11

3-11

γ 254 nm 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 Time (min)

Administration of [18F]2-11 in 10% ethanolic saline was coincident with initiation of 60 minute dynamic brain PET scans with a SOFIE Biosciences G4 Genisys PET/X-Ray (Culver

City, CA, USA) to determine its biodistribution. [18F]2-11 rapidly penetrated the blood-brain barrier (BBB) and reached a maximum whole brain activity of 2.3 ± 0.1 SUV at approximately

70 seconds post injection (Figure 2-14). This distribution of [18F]2-11 is similar to its

11 54 isotopologue [ C]2-11, confirming its brain permeability. The logP7.4 of compound 2-11 was also measured to be 1.5 by an established HPLC method,63 which is similar to reported value of

1.7.54

Figure 2-14. Whole brain biodistribution of [18F]2-11.

Reproduced with permission from Elsevier.64

Further whole body biodistribution analysis revealed that [18F]2-11 rapidly accumulated in the blood, heart and lung at 1 min post injection (SUV > 1), but was quickly washed out within 3 min (Figure 2-15). This tracer also had low uptake in muscle and bladder over the entire analysis.

In addition, the activity was rapidly distributed in kidney and liver and slowly declined in kidney and liver over 60 min, indicating the biliary secretion of this drug. Based on these results, [18F]2-

11 is believed to be a promising PET tracer that could be applied to further translational research in higher species.

Figure 2-15. Whole body biodistribution of [18F]2-11.

Reproduced with permission from Elsevier.64

2.6 Radiosynthesis of the para-analogue [18F]2-36 as potential AMPA PET tracer

Encouraged by this successful radiolabeling method as well as the consistent binding profile of [18F]2-11 with its 11C-isotopologue obtained via this method, the ortho- and para-

[18F]fluorinated derivatives of compound 2-11 were attempted to study their structure-activity relationship and more importantly their biodistribution in the brain. However, only the para- positioned spirocyclic hypervalent iodine(III) precursor 2-32 was synthesized (Scheme 2-15), whereas, the ortho-spirocyclic hypervalent iodine(III) precursor was not obtained probably due to the steric hindrance imposed from the SPIAD auxiliary. Precursor 2-32 was synthesized from

1-bromo-4-iodobenzene 2-33 under the oxidation conditions of Oxone and TFA followed the condensation with SPIAD in basic conditions. The overall yield for 2-32 was 10%. Other oxidative reactions/oxidants, however, led to no desired product.

Scheme 2-15. Synthesis of the iodine(III) precursor 2-32.

The subsequent radiolabeling of the para-iodine(III) precursor 2-32 and copper(I) iodide mediated cross coupling reaction with the compound 2-30 were carried out following the one-pot optimal reaction conditions as mentioned for the meta-analogue [18F]2-11 (Scheme 2-16).

Scheme 2-16. Radiosynthesis of the para-analog [18F]2-36.

18 The first step SNAr reaction gave [ F]2-35 a RCC of 75 ± 3% (n = 5) based on radio-TLC integration and the final compound [18F]2-36 was obtained with a radio-HPLC determined RCC of 28 ± 2% (n = 3) in the cross-coupling step. Compared with the meta-analogue [18F]2-11, the para-derivative [18F]2-36 was obtained with a decreased RCC due to the presence of side product peaks as shown in the chromatographs in Figure 2-16. Further investigation of [18F]2-36 as a potential AMPA receptor PET tracer is ongoing.

Figure 2-16. Chromatograph of the second step radiosynthesis of [18F]2-36.

[18F]2-36

2-36 γ 254 nm

0 2 4 6 8 10 12 14 16 18 20 Time (min)

2.7 Summary

In summary, we have described a novel one-pot two-step radiosynthesis of [18F]2-11, enabled by a spirocyclic hypervalent iodine(III)-activated SNAr radiofluorination of the non- activated arene, followed by a copper(I) iodide-mediated cross coupling with compound 2-30.

Preliminary PET imaging studies of [18F]2-11 in mice demonstrated consistent high brain uptake as its isotopologue [11C]2-11. [18F]2-11 also showed reasonable uptake and clearance of activity in main organs, thereby representing a promising fluorine-18 labeled radiotracer for further evaluation in higher species as a potential AMPA receptor imaging agent. The radiosynthesis of its para-analog [18F]2-36 has been achieved, following the same reaction conditions as that of

[18F]2-11. Evaluation of [18F]2-36 as a potential AMPA receptor PET tracer will be reported in due course.

Experimental Section

General methods for radiosynthesis

General methods for radioisotope preparation

(1 mL, v/v 7 : 3) was added to an aliquot of target water (≤ 1 mL) containing the appropriate amount of [18F]fluoride in a V-shaped vial sealed with a teflon-lined septum. The vial was heated to 110 °C (for TEAB) or 95 °C (for K2CO3/Kryptofix®) while nitrogen gas was passed through a

P2O5-Drierite™ column followed by the vented vial. When no liquid was visible in the vial, it was removed from heat, anhydrous acetonitrile (1 mL) was added, and the heating was resumed until dryness. This step was repeated an additional three times. The vial was then cooled at room temperature under nitrogen pressure. The contents were resolubilized in the suitable solvents.

General methods for analysis of radiofluorination reactions

Radiochemical incorporation yields were determined by radioTLC. EMD TLC Silica gel 60 plates (10 x 2 cm) were spotted with an aliquot (1-5 μL) of crude reaction mixture approximately

1.5 cm from the bottom of the plate (baseline). TLC plates were developed in a chamber containing ethyl acetate (EtOAc) until within 2 cm of the top of the plate (front). Analysis was performed using a Bioscan AR-2000 radio-TLC imaging scanner and WinScan software.

Radiochemical identity and purity were determined by radioHPLC. A Phenomenex Luna C18,

250 x 4.6 mm, 10 µm HPLC column was used for the analytical analysis with a Waters 1515

Isocratic HPLC Pump equipped with a Waters 2487 Dual λ Absorbance Detector, a Bioscan

Flow-Count equipped with a NaI crystal, and Breeze software. The mobile phases for analytical

HPLC analysis included: 60% MeCN, 40% 0.1 M NH4·HCO2 (aq); 40% MeCN, 60% 0.01 M

NaOAc(aq). The flow rate was 1mL/min. whereas, the semi-preparative purifications were performed on a Phenomenex Luna C18, 250 x 100 mm, 10 µm HPLC column with 40% 0.1 M

NH4·HCO2(aq); 40% MeCN, 60% 0.01 M NaOAc(aq) as mobile phase and the flow rate was

7mL/min. All radiochemical yields are non-decay corrected. Each radiochemical labeling was conducted at least two times (n ≥ 2).

Synthesis of related intermediates and products.

Synthesis of tert-butyl 3-iodo-2-oxo-5-(pyrimidin-2-yl)pyridine-1(2H)-carboxylate (2-31):

To and ice cold solution of 3-iodo-5-(pyrimidin-2-yl)pyridin-2(1H)-one 2-17 (2.0 g, 6.68 mmol) in THF (30 mL) was added 4-dimethylaminopyridine (DMAP) (0.08 g, 0.067 mmol), followed by dropwise addition of di-tert-butyl dicarbonate (1.6 g, 7.35 mmol). The reaction mixture was allowed to warm to room temperature and stirred for another 1h. After completion of reaction, water was added to quench the reaction, followed by extraction with ethyl acetate (3 x 25 mL).

The organic layer was dried with MgSO4, filtered and evaporated in vacuo to give dark brown solid. The residue was purified with flash column chromatography (Hexanes/EtOAc = 3/1, v/v) to afford tert-butyl 3-iodo-2-oxo-5-(pyrimidin-2-yl)pyridine-1(2H)-carboxylate 2-31 (1.66 g,

1 yield 62.2%) as a yellow solid. H NMR (300 MHz, CDCl3) δ 9.05 (d, J = 2.4 Hz, 1H), 8.80 (d, J

= 2.3 Hz, 1H), 8.72 (d, J = 4.9 Hz, 2H), 7.71 (t, J = 4.9 Hz, 1H), 1.65 (s, 3H); 13C NMR (75

MHz, CDCl3) δ 160.3, 157.3, 157.2, 150.3, 148.2, 135.8, 119.1, 118.3, 94.0, 87.7, 27.6; Mass-

+ Spectrometry: HRMS-ESI (m/z): Calcd for C14H14IN3O3Na [M+Na] ,421.9972. Found,

421.9979.

Synthesis of 2-(2-oxo-5-(pyrimidin-2-yl)-1,2-dihydropyridin-3-yl)benzonitrile (2-30):

A mixture of tert-butyl 3-iodo-2-oxo-5-(pyrimidin-2-yl)pyridine-1(2H)-carboxylate 2-31 (0.65 g,

1.65 mmol), 2-(1,3,2-dioxaborinan-2-yl)benzonitrile (0.95 g, 4.95 mmol), cesium carbonate

(0.81 g, 2.48 mmol) and tetrakis(triphenylphosphine) palladium (0) (0.19 g, 0.16 mmol) in DMF

(35 mL) was stirred at 110 °C overnight under nitrogen atmosphere. The reaction mixture was diluted with water, was extracted with ethyl acetate and dried over anhydrous magnesium sulfate. The organic layer was evaporated in vacuo and the residue was purified by silica gel column chromatography (Hexanes/EtOAc = 1/1 to 100% EtOAC v/v) to give the title compound

2-(2-oxo-5-(pyrimidin-2-yl)-1,2-dihydropyridin-3-yl)benzonitrile 2-30 (0.38 g, yield 41%) as a

1 pale yellow powder. H NMR (300 MHz, DMSO-d6) δ 12.22 (brs, 1H), 8.67-8.73 (m, 4H), 7.79

(d, J = 7.7 Hz, 1H), 7.67-7.71 (m, 2H), 7.45-7.51 (m, 1H), 7.14 (t, J = 4.9 Hz, 1H); 13C NMR (75

MHz, DMSO-d6) δ 161.4, 161.0, 158.1, 140.6, 139.2, 137.9, 133.44, 133.42, 131.2, 129.0, 128.7,

119.6, 118.6, 116.2, 112.3; Mass-Spectrometry: HRMS-ESI (m/z): Calcd for C16H11N4O

[M+H]+,275.0927. Found, 275.0930.

Synthesis of Spirocyclic hypervalent iodine(III) precursor (2-28)58:

Sodium perborate tetrahydrate (11.55 g, 75 mmol) was added in portions to a 0.15 M solution of

1-bromo-3-iodobenzene 2-26 (0.96 mL, 7.5 mmol) in glacial acetic acid (50.1 mL) heated to

50°C. The reaction mixture was stirred at this temperature overnight, and then it was cooled to room temperature, diluted with water, and extracted three times with dichloromethane. The combined organic layers were dried with anhydrous magnesium sulfate, filtered, and

93 concentrated. The brown yellow crude product was suspended in ethanol (2 mL) and dichloromethane (5 mL) and SPIAD (1.77 g, 7.5 mmol) was added followed by 10% Na2CO3

(aq) (w/v). The pH of the reaction mixture was tested and adjusted with Na2CO3 until the reaction pH>10. The reaction mixture was stirred overnight, then it was diluted with water and extracted with chloroform. The chloroform extracts were combined and washed with water and brine. The organic layer was dried with anhydrous magnesium sulfate, filtered, and concentrated. The residue was purified by silica gel column chromatography (Hexanes/EtOAc = 1/1 to pure EtOAC v/v) to give the title compound 2-28 (0.77 g, yield 20 %) as a white powder. 1H NMR (300 MHz,

DMSO-d6) δ 7.91-7.92 (m, 1H), 7.73-7.76 (m, 2H), 7.41 (t, J = 8.0 Hz, 1H), 2.33 (brs, 2H), 1.95

13 (brs, 2H), 1.91 (brs, 2H), 1.79 (brs, 2H), 1.63-1.67 (m, 6H); C NMR (75 MHz, DMSO-d6) δ

163.0, 134.4, 134.0, 133.2, 131.4, 123.2, 117.3, 105.7, 58.4, 36.9, 35.3, 33.6, 26.4; Mass-

+ Spectrometry: HRMS-ESI (m/z): Calcd for C19H19BrIO4 [M+H] ,516.9485. Found, 516.9475.

De novo synthesis of standard compound 2-11 (Scheme 2-12):

Under nitrogen atmosphere, 1-bromo-3-fluorobenzene 2-29 (10.0 μL, 86.4 μmol) was added to a mixture of 2-(2-oxo-5-(pyrimidin-2-yl)-1,2-dihydropyridin-3-yl)benzonitrile 2-30 (20.0 mg, 72.0

μmol), CuI (18.0 mg, 94.6 μmol), K3PO4 (48.0 mg, 216 μmol) and trans-N,N'- dimethylcyclohexane-1,2-diamine (12.0 μL, 72.0 μmol) in DMF (2 mL). The reaction mixture was protected from light by wrapping aluminum foil around the reaction vessel. The reaction was stirred at 110°C for 2 h. The reaction mixture was diluted with water, was extracted with ethyl acetate and dried over anhydrous magnesium sulfate. The organic layer was evaporated in vacuo and the residue was purified by silica gel column chromatography (Hexanes/EtOAc = 2: 1 to 1: 1) to give the title compound 2-11 (11.0 mg, yield 40%) as greenish yellow powder. The 1H

NMR and 13C NMR characterizations match the literature description.

Optimizations of the 1st step radiosynthesis of [18F]2-29

(1) Method:

Tetraethyl ammonium bicarbonate (TEAB) as the base: 7.0 mg TEAB was used to dry [18F] fluoride. DMF (1.4 mL) was used to solubilize the [18F]TEAF; precursor (2.0 mg) was added into the vial. The resulting activity solution was divided into several aliquots (0.2 mL) into separate vials. Each reaction was tested at indicated reaction conditions, and then quenched with mobile phase (0.4 mL 40% MeCN, 60% 0.01 M NaOAc (aq)).

18 Potassium carbonate as the base: 5.0 mg K2CO3 and 15mg K222 were used to dry [ F]fluoride.

18 DMF (1.0 mL) was used to solubilize the [ F]KF/K222; precursor (2.0 mg) was added into the vial. The resulting activity solution was divided into several aliquots (0.2 mL) into separate vials.

Each reaction was tested at indicated reaction conditions, and then quenched with mobile phase

(0.4 mL 40% MeCN, 60% 0.01 M NaOAc (aq)).

(3) RadioTLC chromatogram of 1st step reaction mixture:

1 2 3 4 5 mean Standard deviation

RCC (%) 64 68 73 74 63 68 5

(4) RadioHPLC chromatogram:

Column: Luna 10u C18 100 Å 250 × 4.6 mm

Mobile phase: 40% MeCN, 60% 0.01 M NaOAc (aq)

Flow rate: 1 mL/min

Optimizations of the 2nd step radiosynthesis of [18F]2-11

(1) Method:

K3PO4 and a solution of compound 2-30, copper(I) iodide and trans-N,N′-dimethylcyclohexane-

1,2-diamine in DMF were added into the 1st step reaction mixture at room temperature. The reaction was allowed to react for 10 at 130 °C. Then, the reaction mixture was quenched with water (10 mL) and filtered through a syringe filter in connection with the t-C18 plus Sep-Pak® cartridge, (Waters; pre-activated with 10 mL EtOH followed by 10 mL H2O) to remove the precipitates and trap the [18F]2-11. Subsequently, 1.5 mL MeCN was pushed through the cartridge to elute off the [18F]2-11, to which 3 mL 0.01 M NaOAc aqueous solution was added to reformulated it into 4.5 mL 33% MeCN in 0.01 M NaOAc solution before semi-preparative

HPLC purification with 40% MeCN in 0.01 M NaOAc to get the [18F]2-11.

(3) Specific Activity (SA) determination for [18F]2-11:

1800000

1600000 y = 6E+14x 1400000 R² = 0.9868 1200000 1000000 800000 600000 400000 200000 0 0 5E-10 1E-09 1.5E-09 2E-09 2.5E-09 3E-09

Figure S2-1. Standard curve for the specific activity determination of [18F]2-11.

The specific activity of [18F]2-11 was determined to be 0.8 ± 0.2 Ci/μmol before injections to the mice.

Radiosynthesis of the 18F-labeled para-aryl analog 2-36 (Scheme 2-16).

(1) Method: The optimal reaction conditions (Table 2-1, entry 4) for the synthesis of [18F]2-29

18 18 were applied to the radiosynthesis of its [ F]2-35. Namely, azeotropically dried [ F]KF/K222 (1

18 mg K2CO3, 5.0 mg K222, 1.0-3.0 mCi [ F]fluoride), resolubilized in DMF (0.2 mL), was added to a vial containing compound 2-32 (2.0 mg). The reaction was heated to 120 °C for 10 min, after which the 1st step RCC was determined by rTLC with ethyl acetate as developing solvent.

Then, K3PO4 (4.8 mg) and a solution of compound 2-30 (4.0 mg), CuI (12.0 mg) and trans-N,

N′-dimethylcyclohexane-1, 2-diamine (7.2 μL) in DMF (0.2 mL) were added at room temperature. The reaction was allowed to react for 10 min at 130 °C. Then, the reaction mixture was quenched with water (0.6 mL) and filtered through a syringe filter to remove the precipitates

97 before HPLC analysis. Finally, the 1st step radio-TLC gave a RCC= 75 ± 3% (n = 5), and the 2nd step RCC was determined by using HPLC as 28 ± 2% (n = 3).

(2) RadioTLC chromatogram of 1st step reaction mixture

1 2 3 4 5 mean standard deviation

RCC (%) 80 75 76 73 72 75 3

(3) RadioHPLC chromatogram:

Column: Luna 10 u C18 100 Å 250 × 4.6 mm

Mobile phase: 60% CH3CN, 40% 0.1 M NH4·HCO2 (aq)

Flow rate: 1 mL/min

γ 254 nm

0 2 4 6 8 10 12 14 Time (min)

Figure S2-2. 1st step co-injection of the 18F-labeled compound 2-35 with its unlabeled analog.

γ 254 nm

0 2 4 6 8 10 12 14 16 18 20 Time (min)

Figure 2-16. 2nd step co-injection of the 18F-labeled compound 2-36 with its unlabeled analog.

LogP Measurement

Based on an established method in literature, the LogP of compound 2-11 was measured via

HPLC.64 A standard curve value was plotted with the reference compounds for their log- corrected retention time (relative to Catechol according to the literature) against their actual

LogP data, as shown in Table S2-1 and Figure S2-3. The HPLC analysis were carried out by using a Phenomenex Luna C18, 250 x 4.6 mm, 10 µm HPLC column and MeOH/10 mM aqueous phosphate buffer (pH = 7.4) = 85/15 v/v as the mobile phase with a flow rate of 1 mL/min.

Compound Name r.t. Corrected r.t. Log (r.t.) LogP

Catechol 3.34 0.00 0.00 0.95

Benzene 5.00 1.66 0.22 2.13

Bromobenzene 5.97 2.63 0.42 2.99

Ethylbenzene 6.85 3.51 0.55 3.13

Hexachlorobenzene 33.82 30.48 1.48 6.53

Compound 2-11 4.58 1.24 0.09 1.50

Table S2-1. Measurements of the retention time for the studied compounds by HPLC.

Log(r.t.) versus LogP 7.00 y = -0.5513x2 + 4.5098x + 1.0392 6.00 R² = 0.9948 5.00

4.00

3.00

2.00

1.00

0.00 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 1.60

Figure S2-3. Standard curve of Log (r.t.) versus the corresponding LogP values.

The retention time of compound obtained under the above mentioned HPLC condition was 4.58 min, therefore, the resulting LogP was 1.5, in consistent with the reported value (1.7). It

100 demonstrates the feasibility of this HPLC method in the predictions of LogP values for the studied compounds.

POSITRON EMISSION TOMOGRAPHY IMAGING

All animal procedures were performed in accordance with the National Institutes of Health

Guide for the Care and Use of Laboratory Animals and were approved by the Massachusetts

General Hospital Institutional Animal Care and Use Facility ((PROTOCOL NUMBER??). The mice were serially imaged using a microPET. For all imaging experiments, mice were anesthetized using 2% isoflurane in O2 at a flow rate of 1.5 L/min, positioned in a prone position along the long axis of the SOFIE Biosciences G4 Genisys PET/X-Ray (Culver City, CA, USA) and imaged. A 60 min-dynamic PET image acquisition was initiated followed by administration of 0.1 mL of the radiotracer in a homogenous solution of 10% ethanol and 90% isotonic saline

(tail iv), and 0.1 mL saline to flush the radiotracer left in the syringe into the mice. Mouse 1 and

2 were injected with 44 µCi and 40 µCi activity separately. Images were dynamically acquired for 10 frames as: 2 x 30 s, 5 x 60 s, 2 x 720 s and 1 x 1800s, and then reconstructed using a filtered back projection reconstruction algorithm. For image analysis, cylindrical regions of interest (ROIs) were manually drawn from three-dimensional filtered back projection (FBP) reconstructed PET images using AMIDE software. Regional radioactivity was expressed as the standardized uptake value (SUV). Two- and three-dimensional visualizations were produced using the DICOM viewer OsiriX (© Pixmeo SARL, 2003-2014).

101

References

1. Meldrum, B. S. Glutamate as a neurotransmitter in the brain: review of physiology and pathology. J. Nutr. 2000, 130, 1007S-15S. 2. Niswender, C. M.; Conn, P. J. Metabotropic Glutamate Receptors: Physiology, Pharmacology, and Disease. Annu. Rev. Pharmacol. Toxicol. 2010, 50, 295-322. 3. Kew, J. N. C.; Kemp, J. A. Ionotropic and metabotropic glutamate receptor structure and pharmacology. Psychopharmacology 2005, 179, 4-29. 4. Conn, P. J.; Pin, J. P. Pharmacology and functions of metabotropic glutamate receptors. Annu. Rev. Pharmacol. 1997, 37, 205-237. 5. Pin, J. P.; Duvoisin, R. The Metabotropic Glutamate Receptors - Structure and Functions. Neuropharmacology 1995, 34, 1-26. 6. Schoepp, D. D.; Conn, P. J. Metabotropic Glutamate Receptors in Brain-Function and Pathology. Trends Pharmacol. Sci. 1993, 14, 13-20. 7. Nakanishi, S. Metabotropic Glutamate Receptors - Synaptic Transmission, Modulation, and Plasticity. Neuron 1994, 13, 1031-1037. 8. Lujan, R.; Nusser, Z.; Roberts, J. D. B.; Shigemoto, R.; Somogyi, P. Perisynaptic location of metabotropic glutamate receptors mGluR1 and mGluR5 on dendrites and dendritic spines in the rat hippocampus. Eur. J. Neurosci. 1996, 8, 1488-1500. 9. Lujan, R.; Roberts, J. D. B.; Shigemoto, R.; Ohishi, H.; Somogyi, P. Differential plasma membrane distribution of metabotropic glutamate receptors mGluR1 alpha, mGluR2 and mGluR5, relative to neurotransmitter release sites. J. Chem. Neuroanat. 1997, 13, 219-241. 10. Shigemoto, R.; Kulik, A.; Roberts, J. D. B.; Ohishi, H.; Nusser, Z.; Kaneko, T.; Somogyi, P. Target-cell-specific concentration of a metabotropic glutamate receptor In the presynaptic active zone. Nature 1996, 381, 523-525. 11. Dunayevich, E.; Erickson, J.; Levine, L.; Landbloom, R.; Schoepp, D. D.; Tollefson, G. D. Efficacy and tolerability of an mGlu2/3 agonist in the treatment of generalized anxiety disorder. Neuropsychopharmacology 2008, 33, 1603-1610. 12. Marino, M. J.; Williams, D. L.; O'Brien, J. A.; Valenti, O.; McDonald, T. P.; Clements, M. K.; Wang, R. P.; DiLella, A. G.; Hess, J. F.; Kinney, G. G.; Conn, P. J. Allosteric modulation of group III metabotropic glutamate receptor 4: A potential approach to Parkinson's disease treatment. Proc. Natl. Acad. Sci. U.S.A. 2003, 100, 13668-13673. 13. Dolen, G.; Osterweil, E.; Rao, B. S. S.; Smith, G. B.; Auerbach, B. D.; Chattarji, S.; Bear, M. F. Correction of fragile X syndrome in mice. Neuron 2007, 56, 955-962. 14. Mayer, M. L.; Armstrong, N. Structure and function of glutamate receptor ion channels. Annu. Rev. Physiol. 2004, 66, 161-181. 15. Traynelis, S. F.; Wollmuth, L. P.; McBain, C. J.; Menniti, F. S.; Vance, K. M.; Ogden, K. K.; Hansen, K. B.; Yuan, H. J.; Myers, S. J.; Dingledine, R. Glutamate Receptor Ion Channels: Structure, Regulation, and Function. Pharmacol. Rev. 2010, 62, 405-496. 16. Watkins, J. C.; Jane, D. E. The glutamate story. Brit. J. Pharmacol. 2006, 147, S100-S108. 17. Rogawski, M. A. Revisiting AMPA receptors as an antiepileptic drug target. Epilepsy Curr. 2011, 11, 56-63. 18. Bowie, D.; Mayer, M. L. Inward Rectification of Both Ampa and Kainate Subtype Glutamate Receptors Generated by Polyamine-Mediated Ion-Channel Block. Neuron 1995, 15, 453-462. 19. Hayashi, T.; Umemori, H.; Mishina, M.; Yamamoto, T. The AMPA receptor interacts with and signals through the protein tyrosine kinase Lyn. Nature 1999, 397, 72-76. 20. Olney, J. W.; Gubareff, T. D. Glutamate Neurotoxicity and Huntingtons-Chorea. Nature 1978, 271, 557-559.

102

21. Choi, D. W. Glutamate Neurotoxicity and Diseases of the Nervous-System. Neuron 1988, 1, 623- 634. 22. Gitto, R.; De Luca, L.; De Grazia, S.; Chimirri, A. Glutamatergic Neurotransmission As Molecular Target of New Anticonvulsants. Curr. Top. Med. Chem. 2012, 12, 971-993. 23. Chang, P. K. Y.; Verbich, D.; McKinney, R. A. AMPA receptors as drug targets in neurological disease - advantages, caveats, and future outlook. Eur. J. Neurosci. 2012, 35, 1908-1916. 24. Calabresi, P.; Cupini, L. M.; Centonze, D.; Pisani, F.; Bernardi, G. Antiepileptic drugs as a possible neuroprotective strategy in brain ischemia. Ann. Neurol. 2003, 53, 693-702. 25. Lees, K. R. Cerestat and other NMDA antagonists in ischemic stroke. Neurology 1997, 49, S66- S69. 26. Muir, K. W.; Lees, K. R. Clinical-Experience with Excitatory Amino-Acid Antagonist Drugs. Stroke 1995, 26, 503-513. 27. Mount, C.; Downton, C. Alzheimer disease: progress or profit? Nat. Med. 2006, 12 (7), 780-784. 28. Contractor, A.; Mulle, C.; Swanson, G. T. Kainate receptors coming of age: milestones of two decades of research. Trends Neurosci. 2011, 34, 154-163. 29. Huettner, J. E. Kainate receptors and synaptic transmission. Prog Neurobiol 2003, 70, 387-407. 30. Lerma, J.; Paternain, A. V.; Rodriguez-Moreno, A.; Lopez-Garcia, J. C. Molecular physiology of kainate receptors. Physiol. Rev. 2001, 81, 971-998. 31. Alt, A.; Weiss, B.; Ogden, A. M.; Knauss, J. L.; Oler, J.; Ho, K.; Large, T. H.; Bleakman, D. Pharmacological characterization of glutamatergic agonists and antagonists at recombinant human homomeric and heteromeric kainate receptors in vitro. Neuropharmacology 2004, 46 (6), 793-806. 32. Sobolevsky, A. I.; Rosconi, M. P.; Gouaux, E. X-ray structure, symmetry and mechanism of an AMPA-subtype glutamate receptor. Nature 2009, 462, 745-756. 33. Hanada, T. The AMPA receptor as a therapeutic target in epilepsy: preclinical and clinical evidence. J. Recept. Ligand Channel Res. 2014, 7, 39-50. 34. Meldrum, B. S.; Rogawski, M. A. Molecular targets for antiepileptic drug development. Neurotherapeutics 2007, 4, 18-61. 35. Turski, L.; Huth, A.; Sheardown, M.; McDonald, F.; Neuhaus, R.; Schneider, H. H.; Dirnagl, U.; Wiegand, F.; Jacobsen, P.; Ottow, E. ZK200775: A phosphonate quinoxalinedione AMPA antagonist for neuroprotection in stroke and trauma. Proc. Natl. Acad. Sci. U.S.A. 1998, 95, 10960-10965. 36. Ornstein, P. L.; Arnold, M. B.; Augenstein, N. K.; Lodge, D.; Leander, J. D.; Schoepp, D. D. (3sr,4ars,6rs,8ars)-6-[2-(1h-Tetrazol-5-Yl)Ethyl]Decahydroisoquinoline-3-Carboxylic Acid - a Structurally Novel, Systemically Active, Competitive Ampa Receptor Antagonist. J. Med. Chem. 1993, 36, 2046-2048. 37. Gilron, I.; Max, M. B.; Lee, G.; Booher, S. L.; Sang, C. N.; Chappell, A. S.; Dionne, R. A. Effects of the 2-amino-3-hydroxy-5-methyl-4-isoxazole-proprionic acid/kainate antagonist LY293558 on spontaneous and evoked postoperative pain. Clin. Pharmacol. Ther. 2000, 68, 320-327. 38. Faught, E. BGG492 (selurampanel), an AMPA/kainate receptor antagonist drug for epilepsy. Expert Opin. Inv. Drug 2014, 23, 107-113. 39. Gomez-Mancilla, B.; Brand, R.; Jurgens, T. P.; Gobel, H.; Sommer, C.; Straube, A.; Evers, S.; Sommer, M.; Campos, V.; Kalkman, H. O.; Hariry, S.; Pezous, N.; Johns, D.; Diener, H. C.; Grp, B. S. Randomized, multicenter trial to assess the efficacy, safety and tolerability of a single dose of a novel AMPA receptor antagonist BGG492 for the treatment of acute migraine attacks. Cephalalgia 2014, 34, 103-113. 40. Citraro, R.; Aiello, R.; Franco, V.; De Sarro, G.; Russo, E. Targeting alpha-amino-3-hydroxyl-5- methyl-4-isoxazole-propionate receptors in epilepsy. Expert Opin. Ther. Tar. 2014, 18, 319-334. 41. Donevan, S. D.; Yamaguchi, S.; Rogawski, M. A. Non-N-Methyl-D-Aspartate Receptor Antagonism by 3-N-Substituted 2,3-Benzodiazepines - Relationship to Anticonvulsant Activity. J. Pharmacol. Exp. Ther. 1994, 271, 25-29.

103

42. Barreca, M. L.; Gitto, R.; Quartarone, S.; De Luca, L.; De Sarro, G.; Chimirri, A. Pharmacophore modeling as an efficient tool in the discovery of novel noncompetitive AMPA receptor antagonists. J. Chem. Inf. Comp. Sci. 2003, 43, 651-655. 43. Chappell, A. S.; Sander, J. W.; Brodie, M. J.; Chadwick, D.; Lledo, A.; Zhang, D.; Bjerke, J.; Kiesler, G. M.; Arroyo, S. A crossover, add-on trial of talampanel in patients with refractory partial seizures. Neurology 2002, 58, 1680-1682. 44. Menniti, F. S.; Buchan, A. M.; Chenard, B. L.; Critchett, D. J.; Ganong, A. H.; Guanowsky, V.; Seymour, P. A.; Welch, W. M. CP-465,022, a selective noncompetitive AMPA receptor antagonist, blocks, AMPA receptors but is not neuroprotective in vivo. Stroke 2003, 34, 171-176. 45. Yamashita, H.; Ohno, K.; Amada, Y.; Hattori, H.; Ozawa-Funatsu, Y.; Toya, T.; Inami, H.; Shishikura, J.; Sakamoto, S.; Okada, M.; Yamaguchi, T. Effects of 2-[N-(4-chlorophenyl)-N- methylamino]-4H-pyrido[3.2-e]-1,3-thiazin-4-one (YM928), an orally active alpha-amino-3-hydroxy-5- methyl-4-isoxazolepropionic acid receptor antagonist, in models of generalized epileptic seizure in mice and rats. J. Pharmacol. Exp. Ther. 2004, 308, 127-133. 46. Feigin, V. Irampanel Boehringer Ingelheim. Curr. Opin. Investig. Drugs 2002, 3, 908-910. 47. Hanada, T.; Hashizume, Y.; Tokuhara, N.; Takenaka, O.; Kohmura, N.; Ogasawara, A.; Hatakeyama, S.; Ohgoh, M.; Ueno, M.; Nishizawa, Y. Perampanel: A novel, orally active, noncompetitive AMPA-receptor antagonist that reduces seizure activity in rodent models of epilepsy. Epilepsia 2011, 52, 1331-1340. 48. Hibi, S.; Ueno, K.; Nagato, S.; Kawano, K.; Ito, K.; Norimine, Y.; Takenaka, O.; Hanada, T.; Yonaga, M. Discovery of 2-(2-Oxo-1-phenyl-5-pyridin-2-yl-1,2-dihydropyridin-3-yl)benzonitrile (Perampanel): A Novel, Noncompetitive alpha-Amino-3-hydroxy-5-methyl-4-isoxazolepropanoic Acid (AMPA) Receptor Antagonist. J. Med. Chem. 2012, 55, 10584-10600. 49. French, J. A.; Krauss, G. L.; Steinhoff, B. J.; Squillacote, D.; Yang, H. C.; Kumar, D.; Laurenza, A. Evaluation of adjunctive perampanel in patients with refractory partial-onset seizures: Results of randomized global phase III study 305. Epilepsia 2013, 54, 117-125. 50. French, J. A.; Krauss, G. L.; Wechsler, R. T.; Wang, X. F.; DiVentura, B.; Brandt, C.; Trinka, E.; O'Brien, T. J.; Laurenza, A.; Patten, A.; Bibbiani, F. Perampanel for tonic-clonic seizures in idiopathic generalized epilepsy A randomized trial. Neurology 2015, 85, 950-957. 51. Gao, M.; Kong, D.; Clearfield, A.; Zheng, Q. H. Synthesis of carbon-11 and fluorine-18 labeled N-acetyl-1-aryl-6,7-dimethoxy-1,2,3,4-tetrahydroisoquinoline derivatives as new potential PET AMPA receptor ligands. Bioorg. Med. Chem. Lett. 2006, 16, 2229-2233. 52. Arstad, E.; Gitto, R.; Chimirri, A.; Caruso, R.; Constanti, A.; Turton, D.; Hume, S. P.; Ahmad, R.; Pilowsky, L. S.; Luthra, S. K. Closing in on the AMPA receptor: Synthesis and evaluation of 2-acetyl-1- (4 '-chlorophenyl)-6-methoxy-7-[(11)C]methoxy-1,2,3,4-tetrahydroisoquinoline as a potential PET tracer. Bioorg. Med. Chem. 2006, 14, 4712-4717. 53. Lee, H. G.; Milner, P. J.; Placzek, M. S.; Buchwald, S. L.; Hooker, J. M. Virtually Instantaneous, Room-Temperature [C-11]-Cyanation Using Biaryl Phosphine Pd(0) Complexes. J. Am. Chem. Soc. 2015, 137, 648-651. 54. Oi, N.; Tokunaga, M.; Suzuki, M.; Nagai, Y.; Nakatani, Y.; Yamamoto, N.; Maeda, J.; Minamimoto, T.; Zhang, M. R.; Suhara, T.; Higuchi, M. Development of Novel PET Probes for Central 2-Amino-3-(3-hydroxy-5-methyl-4-isoxazolyl)propionic Acid Receptors. J. Med. Chem. 2015, 58, 8444- 8462. 55. Gitto, R.; Barreca, M. L.; De Luca, L.; De Sarro, G.; Ferreri, G.; Quartarone, S.; Russo, E.; Constanti, A.; Chimirri, A. Discovery of a novel and highly potent noncompetitive AMPA receptor antagonist. J. Med. Chem. 2003, 46, 197-200. 56. Terence, S. Preparation of 2-pyridinone AMPA receptor antagonists for the treatment of demyelinating disorders and neurodegenerative diseases. WO03047577A2, 2003. 57. Jacobson, O.; Weiss, I. D.; Wang, L.; Wang, Z.; Yang, X.; Dewhurst, A.; Ma, Y.; Zhu, G.; Niu, G.; Kiesewetter, D. O.; Vasdev, N.; Liang, S. H.; Chen, X. 18F-Labeled Single-Stranded DNA Aptamer for PET Imaging of Protein Tyrosine Kinase-7 Expression. J. Nucl. Med. 2015, 56, 1780-1785.

104

58. Rotstein, B. H.; Stephenson, N. A.; Vasdev, N.; Liang, S. H. Spirocyclic hypervalent iodine(III)- mediated radiofluorination of non-activated and hindered aromatics. Nat. Commun. 2014, 5, 4365-4369. 59. Rotstein, B. H.; Wang, L.; Liu, R. Y.; Patteson, J.; Kwan, E. E.; Vasdev, N.; Liang, S. H. Mechanistic studies and radiofluorination of structurally diverse pharmaceuticals with spirocyclic iodonium(III) ylides. Chem. Sci. 2016, 7, 4407-4417. 60. Stephenson, N. A.; Holland, J. P.; Kassenbrock, A.; Yokell, D. L.; Livni, E.; Liang, S. H.; Vasdev, N. Iodonium ylide-mediated radiofluorination of 18F-FPEB and validation for human use. J. Nucl. Med. 2015, 56, 489-492. 61. Wang, L.; Jacobson, O.; Avdic, D.; Rotstein, B. H.; Weiss, I. D.; Collier, L.; Chen, X.; Vasdev, N.; Liang, S. H. Ortho-Stabilized (18) F-Azido Click Agents and their Application in PET Imaging with Single-Stranded DNA Aptamers. Angew. Chem. Int. Ed. 2015, 54, 12777-12781. 62. Zlatopolskiy, B. D.; Zischler, J.; Krapf, P.; Zarrad, F.; Urusova, E. A.; Kordys, E.; Endepols, H.; Neumaier, B. Copper-Mediated Aromatic Radiofluorination Revisited: Efficient Production of PET Tracers on a Preparative Scale. Chem. Eur. J. 2015, 21, 5972-5979. 63. Webster, G. R. B.; Friesen, K. J.; Sarna, L. P.; Muir, D. C. G. Environmental fate modelling of chlorodioxins: Determination of physical constants. Chemosphere 1985, 14, 609-622. 64. Yuan, G.; Jones, G. B.; Vasdev, N.; Liang, S. H. Radiosynthesis and preliminary PET evaluation of 18F-labeled 2-(1-(3-fluorophenyl)-2-oxo-5-(pyrimidin-2-yl)-1, 2-dihydropyridin-3-yl)benzonitrile for imaging AMPA receptors. Bioorg. Med. Chem. Lett. 2016, 26, 4857-4860.

105

106

18 Chapter 3: Metal-free F labeling of aryl-CF2H via nucleophilic radiofluorination and oxidative C-H activation

3.1 Background

Fluorine-bearing functionalities are found in a wide range of materials and biologically active molecules as discussed in the introductory chapter.1-5 Methods of incorporating the fluorine-18/19 atom into bioactive molecules to facilitate the application of noninvasive PET imaging technique in quantifying biochemical and pharmacological progress were also illustrated earlier in Chapter 1.6-11

The aryl-CF2H functionality has been widely used in medicinal chemistry to serve as an isostere of the hydroxyl group.5, 12-16 Isosteristic replacement of phenolic hydroxyl groups with

(hetero)arylCF2H has been successfully applied to the discovery of bioactive molecules, including thiazopyr, fluxapyroxad, deracoxib, pantoprazole, ZSTK474 and a potent kidney urea transporter inhibitor (Figure 3-1).5, 12-17

107

Figure 3-1. Aryl-CF2H containing bioactive molecules.

Reproduced from reference 17, copyright of Royal Society of Chemistry.

The resulting beneficial properties are attributed to the fact that CF2H functionality shares

18 similar acidic and isopolar characteristics with that of the hydroxyl group. Moreover, the CF2H group has a hydrogen-bond-donating capability that is comparable to that of OH and NH groups, but with much reduced hydrophilicity.

3.2 Synthesis of the (hetero)aryl CF2H group

Traditional methods for the synthesis of the difluoromethyl arenes rely on the deoxo-gem- difluoromethylation of adehydes or ketones with the fluorinating reagents sulfur tetrafluoride

(SF4) or aminosulfur trifluorides (e.g. diethylaminosulfur trifluoride (DAST) and bis(2- methoxyethyl)aminosulfur trifluoride (deoxofluor)) (Scheme 3-1).19-20 However, these methods

108 often require harsh reaction conditions and thus are not compatible with a diverse range of functional groups.

Scheme 3-1. Difluorination of aldehydes or ketones with aminosulfur trifluorides.

Methodologies that allow late-stage installation of difluoromethyl functionality onto bioactive molecules have therefore been extensively explored. According to Amii et al., the difluoromethyl arenes could be prepared via the hydrolysis and subsequent decarboxylation of the aryldifluoroacetate intermediates, which were obtained through coupling reactions between

α-silyldifluoroacetates and aryl iodides (Scheme 3-2A).21-22

109

Scheme 3-2. Different difluoromethylation rendered novel methodologies for the synthesis of (hetero)aryl-CF2H functionality.

The other easily-handled difluormethylation reagent, difluoromethyl phenyl ketone, can couple with aryl bromide or chloride substrates via Pd-catalyzed (e.g. palladium complex 1) α- arylation, and subsequent hydrolysis of the ketone via KOH/H2O leading to difluoromethyl arenes (Scheme 3-2B).23 This reaction was achieved in a one-pot, two –step manner. However, both cleavage steps demand harsh reaction conditions. The alternative fluorinating reagent, zinc

110 difluoromethanesulfinate (Zn(SO2CF2H)2, DMFS), was reported by Baran et al. (Scheme 3-

2C).24 This reagent allowed the innate difluomethylation of heteroarenes, conjugated π-systems and thiols under mild reaction conditions. However, this method suffered limited substrate scope and undesired regioisomer formation.

Inspired by the success of the analogous aryl-trifluoromethylation via trifluoromethyl copper

25 complexes [CuCF3], the difluoromethyl copper complexes are expected to react with aryl electrophiles to realize the difluoromethylation (Scheme 3-3). The analogous difluoromethyl source Me3Si-CF2H (TMS-CF2H) could be easily obtained via reduction of the Ruppert-Prakash

26-27 reagent TMS-CF3 with NaBH4. However, TMS-CF2H proved substantially less reactive than

28 TMS-CF3 due to the increased bond strength of Si-CF2H. The resulting Cu-CF2H complex was easily decomposed to the 1,1,2,2-tetrafluoroethane and cis-difluoroethylene side products.

Scheme 3-3. Formation and decomposition of the difluoromethyl copper complex.

Despite these difficulties using difluoromethyl copper complexes, Hartwig et al. have reported a direct nucleophilic difluoromethylation of electron-rich aryl and vinyl iodides with excess Me3SiCF2H (5 equiv) as the difluoromethyl source and stoichiometric amount of copper iodide as catalyst under elevated temperature in N-methylpyrrolidone (NMP) (Scheme 3-4A).29

111

Scheme 3-4. Copper-mediated direct difluoromethylation.

112

Qing et al. have improved this methodology by using phenanthroline as a ligand and t-

BuOK as the base to replace CsF (Scheme 3-4B).30 As a result, this reaction could be realized at room temperature with 2.5 equiv. of TMS-CF2H and was also extended to electron-deficient substrates and heteroarenes. Goossen et al. disclosed a Sandmeyer-type difluoromethylation of

(hetero)arenediazonium salts, derived from (hetero)aromatic amines, to form difluoromethyl

(hetero)arenes under mild reaction conditions (Scheme 3-4C).31 The difluoromethyl copper complex was generated in situ from copper thiocyanate and TMS-CF2H. The disadvantage of these methods was attributed to the use of TMS-CF2H, which requires elevated temperature or

32 addition of strong nucleophiles to cleave the Si-CF2H bond. Complementary to this work,

Prakash et al. utilized tin reagents (e.g. nBu3SnCF2H, Scheme 3-4D) as the difluoromethyl source to replace the previous silicon based reagents, allowing the cleavage of Sn-CF2H at

33 decreased temperature or with milder nucleophiles, (e.g. KF). The nBu3SnCF2H was synthesized with a yield of 86% from the reaction between TMSCF3 and nBu3SnH at 45°C with

CaI2 as initiator in N,N-dimethylacetamide (DMA). This method was successfully applied to iodoarenes, iodoheteroarenes as well as β-styryl halide substrates. It also tolerated the CHO,

COR and COOEt functional groups. In addition, they also found the selected solvent DMF

33 served to stabilize the reactive Cu-CF2H species.

All the above methods utilized difluoromethyl copper complexes. Unlike the well-known ligated [CuCF3] species, which are thermally stable and highly reactive, the [CuCF2H] has never been isolated in solid state.34 As a result, it hampered fundamental studies on well-defined

I [LCu CF2H] complexes. Hartwig et al. suggested the reactive [CuCF2H] species for the reaction

- depicted in Scheme 3-4A came from the [Cu(CF2H)2] intermediates, which explained the compatibility of the reaction with high temperature.29 In addition, when aryl bromides are used

113 for the difluoromethylation, the copper reagents would require elevated temperatures to activate these substrates.35

To avoid the problematic copper reagents, other transition metal catalyzed difluoromethylation reactions were explored. Shen et al. disclosed a cooperative dual palladium/silver catalyzed direct transformation of aryl bromides and iodides to difluormethylated arenes under mild reaction conditions (Scheme 3-5A).36 However, this method required the addition of silver co-catalysts and the use of TMS-CF2H as difluoromethyl source and tert-butoxide as activator.

114

Scheme 3-5. Copper-absent methodologies for the synthesis of (hetero)aryl-CF2H.

Recently, Vicic et al. reported a nickel catalyzed ((dppf)Ni(COD),37 a known precatalyst, dppf=1,1’-bis(diphenylphosphanyl)-ferrocene, cod=cyclo-1,5-octadiene) difluoromethylation of aryl (pseudo)halides (i.e. I, Br, OTf) with a stable, isolable and easily prepared difluoromethyl

38 zinc reagent (Scheme 3-5B). The (DMPU)2Zn(CF2H) reagent was prepared in high isolated yield by reacting ICF2H with diethyl zinc in the presence of 1,3-dimethyl-3,4,5,6-tetrahydro-

2(1H)-pyrimidinone (DMPU). The difluorination reaction could be achieved under room temperature with moderate to excellent yields (50%-85%).

115

Another class of (hetero)aryl-CF2H synthetic methodologies utilizes the fluorination of benzylic C-H bonds, following radical reaction mechanisms (Scheme 3-6). Besides the previous mentioned radical reaction with DMFS (Scheme 3-2C), the methods listed below are examples of recently published work.

Scheme 3-6. Radical methodologies for the synthesis of aryl-CF2H.

Chen et al. reported a direct transformation of benzylic C-H bonds to C-F bonds.39 Visible light was found to activate diarylketones to selectively abstract a benzylic hydrogen atom from the substrates and subsequent insertion of fluorine gave the benzylic fluoride. In the case of C-H gem-difluorination, xanthone was used as a catalyst and Selectfluor II was employed as the fluorine radical donor to promote the formation of aryl-CF2H (Scheme 3-6A). Tang et al. disclosed a silver-catalyzed oxidative activation of benzylic C-H bonds of aryl-CH3 for the

40 synthesis of aryl-CF2H (Scheme 3-6B). Sodium persulfate was used as an oxidant and

116

Selectfluor was used as a fluorinating reagent. This reaction was carried out under degassed conditions. This methodology was also proven to work well under transition-metal-free conditions according to Cai et al.41

18 3.3 F labeling of the aryl-CF2H group

Compared with the tremendous progress made in the difluoromethylation of aryl halides, the

18 radiosynthesis of [ F]Ar–CF2H functionality has not been fully exploited. The fluorodecarboxylation of carboxylic acids was initially realized by F2 and XeF2 (Scheme 3-

42-45 7A). Mechanistic studies of the reaction between XeF2 and the secondary and tertiary carboxylic acids revealed that the C-F bond was formed through the reaction of XeF2 with carboxylic acid generated alkyl radicals.44 Sammis et al. then disclosed the alkyl radicals could be generated from a milder t-butyl peroxyester reagent via thermal decarboxylation, and the fluorinating reagent could be N-fluorobenzenesulfonimide (NFSI) or Selectfluor.46 A similar strategy was also revealed by the same group for the photo-fluorodecarboxylation of 2-aryloxy and 2-aryl carboxylic acid with Selectfluor as fluorinating reagent (Scheme 3-7B).47 Li et al. reported that the fluorodecarboxylation of aliphatic carboxylic acids could be realized with

48-49 Selecfluor under AgNO3 catalysis. This reaction involved the formation of a putative alkyl radical by Ag(III)-mediated single-electron transfer and subsequent fluorine atom transfer, whereas, this method led to no reaction for the aryl carboxylic acids. Gouverneur et al. then proposed the incorporation of the CF2H unit between an aryl group and carboxylic acid would lead a new class of reactive species since the α, α- difluorobenzyl radical is known to be stable and adopt an all planar geometry.49 With the success of this method, they further developed the silver(I)-mediated fluorodecarboxylation of α-fluorarylacetic acids by [18F]Selectfluor

18 49-50 bis(triflate) to get the [ F]aryl-CF2H (Scheme 3-7D).

117

Scheme 3-7. Development of the [18F]Selectfluor bis(triflate)-rendered radiosynthesis of 18 [ F] aryl-CF2H.

The use of [18F]Selectfluor bis(triflate) as a fluorinating reagent avoids the poor yields

18 especially for electron-rich substrates and side reactions brought by the highly reactive [ F]F2.

18 18 The [ F]Selectfluor bis(triflate) could be rapidly synthesized from [ F]F2 by the method shown in Scheme 3-7D.50 The exemplified substrate in Scheme 3-7D gave a radiochemical yield (RCY) of 9 ± 3%.49

Gouverneur et al. also reported the silver(I)-mediated halogen exchange for the

18 51 radiosynthesis of [ F] aryl-CF2H (Scheme 3-8).

118

18 Scheme 3-8. Silver (I)-mediated halogen exchange for radiosynthesis of [ F] aryl-CF2H.

18 Condition A: [ F]KF/K222, AgOTf (1.0 eq.), CH2Cl2, r.t., 20 min. 18 51 Condition B: [ F]KF/K222, AgOTf (2.0 eq.), 1,2-dichloroethane, 60 °C, 20 min.

This method allowed the exchange of aryl-halides with the readily available non-carrier- added 18F-fluoride under mild reaction conditions. However, this method had limited substrate scope and low to moderate RCCs. In addition, the Aryl-CHFCl precursors require multi-step synthesis: in a general procedure, the aryl(fluoro)acetic acid first reacts with (COCl)2 to form the acid chloride. Then, Na-N-hydroxy-2-thiopyridone and DMAP in BrCCl3 is used to transform this intermediate into a mixture of bromo and chlorodecarboxylated product which is subsequently converted into the desired chlorinated product using tetrabutylammonium chloride.51

Our recent work disclosed a more practical method, derived from the method shown in

Scheme 3-2B (Scheme 3-9).23, 52 Starting from the easily accessed aryl (pseudo)halides, the

119 palladium catalyzed α-arylation with 2-fluoro-1-phenylethan-1-one and subsequent bromination

18 18 at the benzylic position gave the F-labeling precursors. Radiosynthesis of [ F]Ar–CF2H was achieved by nucleophilic replacement of bromide via TEA[18F] and subsequent hydrolysis to cleave the benzoyl group.

18 Scheme 3-9. Radiosynthesis of [ F] aryl-CF2H via benzoyl auxiliary and its application.

This method featured excellent substrate scope (Figure 3-2).52 Seven 18F-labeled difluoromethylarene containing drug derivatives and radiopharmaceuticals were synthesized via this method, including DAA1106 (potent and selective benzodiazepine receptor agonist),

Fenofibrate (cholesterol scavenger and a drug for cardiovascular disease), Claritin (treatment for allergies), SC-58125 (COX2 inhibitor), Fluoxetine (antidepressant) and Estrone (estrogenic hormone). The RCYs of these products range from 11%-39%. This transformation could be

18 achieved via one-pot starting from the labeling precursors to get the final F-labeled aryl-CF2H products.

120

Figure 3-2. 18F-labeling of bioactive molecules via palladium catalyzed α-arylation.

Reproduced with permission from Wiley.52

18 3.4 F-labeling of the aryl-CF2H via oxidative benzylic C-H bond activation

The purpose of our current work was to explore novel 18F-labeling methods for the synthesis

18 of [ F]aryl-CF2H. The urgent need of developing novel methodologies also comes from the fact that the unsatisfactory [low to moderate] specific activity available through current methods limits their application in PET imaging studies, particularly for low density targets.17 Inspired by the benzylic C-H bond activation and Selectfluor rendered fluorination methodologies as discussed in Scheme 3-6B,40-41 we hypothesized this protocol would lead to a new approach, where the difluoromethylation could be realized via the nucleophilic radiofluorination of aryl-

121

(pseudo)halides with nucleophilic [18F]fluoride followed by benzylic C-H bond activation to insert the second fluorine atom with Selectfluor (Scheme 3-10).

18 Scheme 3-10. Radiosynthesis of [ F] aryl-CF2H via benzylic C-H bond activation.

Reproduced from reference 17, copyright of Royal Society of Chemistry.

To test the feasibility of the above method, the radiosynthesis of 4-([18F]difluoromethyl)-

1,1'-biphenyl ([18F]3-3a) was carried out, starting from the 4-(bromomethyl)-1,1'-biphenyl (3-1a)

(Scheme 3-11).

Scheme 3-11. Radiosynthesis of 4-([18F]difluoromethyl)-1,1'-biphenyl ([18F]3-3a)

The nucleophilic [18F]fluoride in the first step was initially prepared in the form of

[18F]TBAF with the base of tetra-n-butylammonium hydroxide (TBAH), according to a reported method.53 Optimization of the reaction conditions with regard to the amount of the base and the reaction time gave the highest RCC of 49 ± 2% (n = 2). Changing the base from TBAH to

TEAB, however, gave much enhanced RCCs. Near quantitative RCC (98 ± 2%, n > 10) was achieved after optimization of the reaction conditions, considering the amounts of the base, the reaction time and temperature. The optimal reaction conditions for the first step was 3-1a (8.09

μmol), TEAB (10.4 μmol) in MeCN (0.2 mL) at 130 °C for 10 min (Table 3-1, entry 6).

122

Table 3-1. Optimization for the radiosynthesis of 4-([18F]fluoromethyl)-1,1'-biphenyl ([18F]3-2a)

Entry 3-1a (mg) Base (mg) Time (min) Temp. (°C) [18F]3-2a RCC (%)a 1 2.0 TBAH (1.56) 5 130 40 ± 2 (n = 2) 2 2.0 TBAH (3.12) 5 130 43 ± 1 (n = 2) 3 2.0 TBAH (4.69) 5 130 39 ± 1 (n = 2) 4 2.0 TBAH (3.12) 10 130 49 ± 2 (n = 2) 4 2.0 TEAB (0.5) 10 130 50 ± 2 (n = 2) 5 2.0 TEAB (1.0) 10 130 85 ± 1 (n = 2) 6 2.0 TEAB (2.0) 10 130 98 ± 2% (n > 10) 7 2.0 TEAB (2.5) 10 130 88 ± 1 (n = 2) 8 2.0 TEAB (2.0) 10 100 90 ± 2 (n = 3) 9 2.0 TEAB (2.0) 10 80 85 ± 3 (n = 3)

aRCC was determined based on rTLC integration.

Initially, the [18F]3-3a was proposed to be synthesized in a two-step, one-pot manner, however, application of the second step reaction conditions with the first step reaction mixture led to no desired product. Since the first step starting material would consume the oxidant

Na2S2O8 and fluorine source Selectfluor as well as TEAB as a base may affect the second step reaction, the [18F]3-2a was isolated from the first step reaction mixture. The collected eluent from semi-preparative HPLC that contains [18F]3-2a was further purified using a C18 cartridge to remove the ammounium formate in the mobile phase before being applied to the second step radiosynthesis. The eluent passed through the C18 cartridge to trap the activity, which was then eluted out by 1.0-1.5 mL MeCN or EtOH.

123

The second step oxidative benzylic C-H activation was quite challenging. Initial efforts focused on optimizing the reaction temperature and stoichiometry of reagents including AgNO3,

Na2S2O8, and Selectfluor. However, the freeze-pump-thaw-degas cycle had to be applied to the reaction mixture under elevated temperature (Table 3-2, entries 1-3), as that described in the original method, and produced only negligible amounts of product under these conditions (Table

3-2, entry 3, 4 ± 1%). The HPLC chromatographs of the failed experiments only showed the

[18F]fluoride peak, indicating that [18F]3-2a was quite unstable under the second step oxidative reaction condition. Surprisingly, under the same reaction conditions, when tiny amounts of EtOH

(i.e. MeCN: H2O: EtOH = 10: 10: 1, total volume 0.4 mL) from the C18 purification step were accidently mixed into the second step reaction, it significantly increased the second step RCCs

(Table 3-2, entries 4-5, 44 ± 7% (entry 5, n = 2)). Absence of the EtOH in the reaction solvent led to no reaction under the same conditions (Table 3-2, entry 4 versus entry 6). The reaction was tested without degassing operations under the same reaction conditions as that of entry 4, and a similar result was obtained (Table 3-2, entry 7, 14 ± 2% (n = 2)). Further optimization of the amount of EtOH used in this system revealed that 0.1 volumetric equiv of EtOH (20 µL) was ideal for this reaction. More EtOH (e.g. 100 µL, entry 8) led to reaction. The HPLC analysis of the second step reaction, in this case, would only show the [18F]3-2a peak.

124

Table 3-2. Optimization for the radiosynthesis of 4-([18F]difluoromethyl)-1,1'-biphenyl ([18F]3-3a)

18 a Entry Degas EtOH (µL) AgNO3 (mg) Na2S2O8 (mg) Selectfluor (mg) Temp. (°C) Time (min) [ F]3-3a RCC (%) 1 Y N 2 4 20 80 10 0 (n = 1) 2 Y N 2 4 20 100 10 0 (n = 1) 3 Y N 2 4 20 120 10 4 ± 1 (n = 1) 4 Y 20 2 4 20 120 10 20 (n = 1) 5 Y 20 4 4 20 120 10 36 ± 7 (n = 2) 6 Y N 2 4 20 120 10 0 (n = 1) 7 N 20 2 4 20 120 10 14 ± 2 (n = 2) 8 N 100 2 4 20 120 10 0 (n = 2) 9 N 20 4 4 20 120 10 38 ± 5 (n = 2) 10 N 20 8 4 20 120 10 20 ± 2 (n = 2) 11 N 20 12 4 20 120 10 22 ± 3 (n = 2) 12 N 20 4 8 20 120 10 41 ± 6 (n = 2) 13 N 20 4 12 20 120 10 23 ± 3 (n = 2) 14 N 20 4 16 20 120 10 20 ± 4 (n = 2) 15 N 20 4 8 5 120 10 0 (n = 2) 16 N 20 4 8 10 120 10 9 ± 1 (n = 2) 17 N 20 4 8 15 120 10 33 ± 3 (n = 2) 18 N 20 0 8 20 120 10 50 ± 5 (n = 3) 19 N N 0 8 20 120 10 0 (n = 3) aRCC was determined based on rHPLC integration.

125

Once the importance of the addition of EtOH was identified, the reaction conditions were optimized by adjusting the amount of AgNO3 (entries 9-11), Na2S2O8 (entries 12-14), and

40 Selectfluor (entries 15-17). Inspired by the work of Cai et al., we questioned if the AgNO3 could be removed from the reaction mixture. To our delight, under the initial optimal reaction conditions (Table 3-2, entry 12, RCC = 41 ± 6%), removal of AgNO3 gave superior RCC (50 ±

5%, entry 18), making the HPLC analysis and subsequent semi-preparative purification of the second step straightforward. A further test of the removal of EtOH under the transition-metal- free conditions, however, led to no desired [18F]3-3a (Table 3-2, entry 19). Finally, compound

18 [ F]3-3a was obtained in 50 ± 5% RCC (n = 3) with Na2S2O8 (33.6 μmol), Selectfluor (56.5

μmol) at 120 °C for 10 min (Table 3-2, entry 18). A schematic workflow for the radiosynthesis of [18F]3-2a and [18F]3-3a are depicted below (Scheme 3-12).17

Scheme 3-12. Workflow for the radiosynthesis of [18F]3-2a and [18F]3-3a.

Reproduced from reference 17, copyright of Royal Society of Chemistry..

126

Synthesis of [18F]3-2a started from (a) trapping and releasing 18F[fluoride] into a 5 mL V- vial via a solution of TEAB in 1 mL MeCN/H2O (v/v 7:3). (b) Azeotropic drying with anhydrous

MeCN (3*1 mL, 3 min for each cycle) at 110 °C to give the nucleophilic complex [18F]TEAF.

The first step reaction of [18F]TEAF with substrate 3-1a (c) in MeCN at130°C in 10 min gave the

[18F]3-2a. Dilution and subsequent semi-preparative HPLC purification of the first step reaction mixture (d) were completed within 12 min. Concentration of the eluent containing [18F]3-2a via

C18 cartridge (e) and subsequent release (f) via 1 mL MeCN transferred [18F]3-2a into a MeCN solution suitable for the second step of the reaction. Synthesis of [18F]3-3a was achieved by

18 adding [ F]3-2a in MeCN into a mixture of Na2S2O8, Selectfluor, H2O and EtOH (g) followed by heating at 120°C for 10 min (h). Once the reaction was completed the second step reaction mixture was diluted and purified to get the desired [18F]3-3a (i).

3.5 Investigation of the substrate scope for the benzylic C-H activation radiosynthesis

The scope of this benzylic C-H bond activation methodology was then explored with more challenging substrates (Scheme 3-13).17 Under the optimized first reaction conditions for the synthesis of [18F]3-2a, all [18F]monofluorinated products were synthesized quantitatively, except

[18F]3-2e (85 ± 5%, n = 3) and [18F]3-2f (70 ± 3%, n = 3).

127

Scheme 3-13. Radiosynthesis of [18F]3-2b to [18F]3-2i.

The isolated [18F]monofluorinated compounds ([18F]3-2b to [18F]3-2i) were then subjected to the second step under optimized reaction conditions as described for the radiosynthesis of

[18F]3-3a. It was found that oxidative benzylic C-H bond activation of these substrates, however, did not require the addition of EtOH.

128

As shown in Scheme 3-14, the para-substituted electron-deficient ketone substrates [18F]3-

2b and [18F]3-2c gave moderate RCCs of the corresponding products (10% and 15% respectively). Para-substituted ester substrate [18F]3-2d also yielded similar RCC of 12%.

Substrate [18F]3-2e is an example of multi-substituted, sterically hindered compound, which gave a better RCC of 28% for [18F]3-3e. The bromide functionality of [18F]3-3e could facilitate further coupling reactions with various counterparts to synthesize more complicated bioactive molecules. Noteworthy, this method had the advantage of allowing the direct transformation of the primary amide bearing substrate [18F]3-2f with a good RCC of 45%, since it is well known that radiofluorination of such a motif without a protecting group is challenging. The 2- arylpyridine substrate [18F]3-2g demonstrated the capability of this methodology in the application of the N-heteroatom containing compounds. Noteworthy as well is substrate [18F]3-

2h, which shows the selectivity of this strategy in specifically activating the benzylic C-H bond, followed by the insertion of the second fluorine atom to the benzylic tertiary carbon.

Furthermore the (1,1-difluoroethyl)benzene fragment of product [18F]3-3h is a key structural motif in a series of potent kidney urea transporter inhibitors,11 analogous to the last compound shown in Scheme 3-1, indicating the potential use of the method to radiosynthesize molecules in this class.54 The electron-rich arene [18F]3-2i, however, immediately decomposed to [18F]fluoride under the standard oxidative conditions (See experimental section for detailed optimizations of each substrate). Alternative oxidation/fluorination reagents, for example, 1-chloromethyl-4- fluoro-1,4-diazoniabicyclo [2.2.2] octane bis(hexafluorophosphate) (F-TEDA-PF6), a fluorinating reagent reported by Ritter et al.,55 failed to produce the desired [18F]3-3i. It is possible that the electron-rich substrates are more vulnerable to the fluorinating reagents and/or

129 radical defluorination due to their increased electron density brought by the electron donating groups.

Scheme 3-14. Radiosynthesis of [18F]3-3b to [18F]3-3i.

A plausible reaction mechanism for this protocol is proposed as shown in Scheme 3-15.

Thermal homolytic decomposition of a peroxydisulfate anion generates two sulfate radical

130

18 anions, which abstracts an electron from the benzylic C-H bond of [ F]aryl-CFH2 (A) to give a benzylic radical (B) and a bisulfate ion. Then, fluorinating reagent Selectfluor reacts through a radical mechanism,56-57 providing a fluorine radical for the C-F bond formation at the benzylic

18 radical (B) to give the desired [ F]aryl-CF2H (C).

18 Scheme 3-15. Proposed mechanism for the radiosynthesis of [ F]aryl-CF2H.

Reproduced from reference 17, copyright of Royal Society of Chemistry..

3.6 Specific activity measurement

Specific activity (SA), as discussed in the introduction Chapter 1, is a critical parameter in animal/human PET imaging that should be taken into consideration at the beginning of developing novel radiolabeling methods.58-59 Obtaining high specific activity is an ongoing challenge for many 18F-labeling methodologies, especially for the radiosynthesis of multi-

18 19 18 fluorinated functional groups, such as [ F]aryl-CF2H due to the F- F isotopic exchange.

To determine the SA and its potential loss in the current method, we measured the SAs for both [18F]3-2a and [18F]3-3a (See scheme 3-16 and experimental section for details). The first

18 step radiosynthesis of [ F]3-2a was carried out and isolated by a GE TRACERlab™ FXFN (see experimental section for details), starting from 1.7 GBq of [18F]fluoride. The [18F]3-2a was

131 obtained in a 61% RCY (non-decay corrected) with chemical and radiochemical purities > 99% and a specific activity of 51.8 GBq/µmol at the end-of-synthesis, which was determined by the analytical HPCL. The radiosynthesis took around 60 min starting from the [18F]fluoride. The isolated intermediate [18F]3-2a was further purified as the procedures indicated in Scheme 3-12, namely, it was trapped in a C-18 cartridge and then eluted out to the reaction vial via 1.0 mL of

MeCN. The 1.0 mL [18F]3-2a in MeCN solution was diluted with 1.0 mL of sterile water, from which 0.4 mL of the mixture was taken for the second step. Under the optimized conditions mentioned above, the mixture was heated for 10 min, cooled and diluted with mobile phase. The reaction mixture was then injected into a semi-preparative HPLC for purification. The purified

[18F]3-3a was isolated in 38% RCY (non-decay corrected) based on the added activity of [18F]3-

2a. [18F]3-3a had a SA of 22.2 GBq/µmol at the end of synthesis. The second step reaction took around 25 min synthesis time from the addition of [18F]3-2a to the end of the purification. The analytical HPLC determined chemical and radiochemical purities of [18F]3-3a were more than

99%.

Scheme 3-16. Specific activity determination for 4-([18F]difluoromethyl)-1,1'-biphenyl ([18F]3-3a).

Reproduced from reference 17, copyright of Royal Society of Chemistry..

18 The current method gives superior SA among other reported synthesis of [ F]aryl-CF2H as shown in Table 3-3.49, 51-52 The electrophilc, [18F]Selectfluor bis(triflate) rendered radiofluorination gave a decay-corrected specific activity of 2.5 GBq/µmol.49 The halogen

132 exchange method suffered the poorest specific activity, which was only 0.03 GBq/µmol.51 The recently developed method in our lab also gave a moderate SA of 3.0 GBq/µmol.52 The current method significantly improves the specific activity of the final product [18F]3-3a with a specific activity of 22.2 GBq/µmol. Further efforts in improving the specific activity were tried by employing Selectfluor with different counterions to reduce/eliminate potential isotopic exchange,

- - for example, PF6 (F-TEDA-PF6) and OTf (1-chloromethyl-4-fluoro-1,4- diazoniabicyclo[2.2.2]octane bis(trifluoromethane sulfonate) ((F-TEDA-OTf), see Experiment section for its detailed preparation). However, none of these fluorinating reagents led to the

18 - desired product [ F]3-3a. On the other hand, Scott et al. had indicated that the counterion BF4 was able to give comparably high specific activity as that of OTf- in their radiosynthesis of aryl-

[18F]fluoride from (mesityl)(aryl)iodonium salt precursors.60 Therefore, the reasonably high specific activity of current protocol may attribute to the fact that the 19F-18F isotopic exchange in

CF2H labeling reactions either did not occur or were on a negligible scale.

133

18 18 Table 3-3. Summary of the specific activity for the current F-labeling of [ F]aryl-CF2H.

Reproduced from reference 17, copyright of Royal Society of Chemistry..

134

3.7 Summary

In summary, we have investigated a transition-metal-free oxidative benzylic C-H activation

18 18 strategy for the F-labeling of [ F]aryl-CF2H functionality with moderate to good radiochemical yields. This method resulted in reasonably high specific activity and was superior to the other reported methods. The current method was successfully applied to various mono- or multi-substituted electron-neutral and electron-deficient substrates. It also demonstrated compatibility toward the unprotected primary amide group, N-containing heterocycles, and the tertiary benzylic C-H bond. Further ongoing research resides in extending this method to electron-rich substituted (hetero)arenes and its application toward the synthesis of complex molecules.

135

Experimental Section

ORGANIC CHEMISTRY

1) General procedure for the synthesis of standard aryl-CFH2 compounds:

Compounds 3-2a to 3-2i were synthesized from 3-1a to 3-1i via tetra-n-butylammonium fluoride (TBAF) fluorination. Briefly, in a dried 3-neck round bottom flask, 0.36 mmol 3-1a to 3-1i were dissolved in 1.0 mL tetrahydrofuran (THF), and then 1.08 mL TBAF (1.0 M in THF solution) was added under nitrogen atmosphere. The reaction mixture was heated to reflux for 2 h or overnight. The resulting mixture was quenched with water, extracted with ethyl acetate (3 x 10 mL), and dried with anhydrous sodium sulfate. Solvent was removed under vacuum, and the crude product was purified via silica gel chromatography.

2) General procedure for the synthesis of standard aryl-CF2H compounds:

The aryl-CF2H products were synthesized following our previous method with the aryl-CH3

40-41 substrates : In general, under a nitrogen atmosphere, to the corresponding aryl-CH3 (1.00 mmol, 1.00 equiv), Na2S2O8 (5.00 mmol, 5.00 equiv) and 1-chloromethyl-4-fluoro-1,4- diazoniabicyclo[2.2.2]octane bis(hexafluorophosphate) (Selectfluor, 3.00 mmol, 3.00 equiv) in a

Schlenk tube were added MeCN (4.0 mL) and H2O (4.0 mL). The mixture was cooled by liquid nitrogen, then AgNO3 (0.10 mmol, 0.10 equiv) was added. The reaction mixture was degassed three times by freeze-pump-thaw cycles and then was stirred for 3.0 h or overnight at 80 °C. After cooling to room temperature, the mixture was filtered through a pad of celite, eluting with ethyl acetate. The filtrate was washed with saturated NaHCO3 solution (10 mL), sat. NaCl, and dried over Na2SO4. The filtrate was concentrated in vacuo and the residue was purified by chromatography on silica gel, eluting with hexane/ethyl acetate to afford the 3-3a to 3-3h.

1 Synthesis of methyl 5-bromo-2-(fluoromethyl)benzoate (3-2e). H NMR (300 MHz, CDCl3) δ 8.16 (t, J = 1.7 Hz, 1H), 7.72 (dd, J = 8.5Hz, 2.1 Hz, 1H), 7.55 (d, J = 8.5 Hz, 1H), 5.78 (d, J =

13 48.3 Hz, 2H), 3.90 (s, 3H); C NMR (75 MHz, CDCl3) δ 165.6, 138.9 (d, J = 17.7Hz), 135.8 (d, 136

J = 2.1Hz), 133.4, 128.1 (d, J = 3.5Hz), 127.5 (d, J = 17.0Hz), 121.3 (d, J = 2.1Hz), 82.3 (d, J =

19 169.1Hz), 52.4; F NMR (282 MHz, CDCl3) δ -214.9 (t, J = 48.2Hz). Mass-Spectrometry:

+ HRMS-ESI (m/z): Calcd for C9H8BrFO2H [M+H] , 246.9764. Found, 246.9759.

Synthesis of 1-chloromethyl-4-fluoro-1,4-diazoniabicyclo [2.2.2] octane

61 bis(hexafluorophosphate) (F-TEDA-PF6) was achieved by using a reported method.

Synthesis of 1-chloromethyl-4-fluoro-1,4-diazoniabicyclo [2.2.2] octane bis(trifluoromethane sulfonate) (F-TEDA-OTf. In a 15 mL round bottom flask, Selectfluor (0.10 g, 0.28 mmol),

LiOTf (0.45 g, 2.8 mmol) and 2.5 mL H2O were added. The resulting mixture was stored in fridge (2-5 °C) overnight. The product precipitated out as white crystal, which was washed with

1 2*3 mL water to get the pure product (0.12 g, 90%). H NMR (300 MHz, D2O) δ 5.47 (s, 2H),

13 4.94 (dd, J = 14.7 Hz, 7.0 Hz, 6H), 4.48 (t, J = 7.3 Hz, 6H); C NMR (75 MHz, D2O) δ 112.9,

19 121.7, 117.5, 113.3, 69.0, 57.4, 57.2, 53.7 (t, J = 3.3 Hz); F NMR (282 MHz, D2O) δ -74.9. LC-MS found 329.2 [M-OTf]+.

137

RADIOCHEMISTRY

General methods for radioisotope preparation

A GE PETtrace 16.5 MeV cyclotron was used for [18F]fluoride production by the 18O(p,n)18F nuclear reaction to irradiate 18O-enriched water. [18F]fluoride was delivered to a lead- shielded hot cell in 18O-enriched water by nitrogen gas pressure. [18F]fluoride was prepared for radiofluorination of aromatics by the following method: a solution of base (tetraethylammonium bicarbonate (TEAB)) or potassium carbonate/K222 (Kryptofix®) in acetonitrile and water (1 mL, v/v 7 : 3) was added to an aliquot of target water (≤ 1 mL) containing the appropriate amount of [18F]fluoride in a V-shaped vial sealed with a teflon-lined septum. The vial was heated to 110 °C

(for TEAB) or 95 °C (for K2CO3/Kryptofix®) while nitrogen gas was passed through a P2O5- Drierite™ column followed by the vented vial. When no liquid was visible in the vial, it was removed from heat, anhydrous acetonitrile (1 mL) was added, and the heating was resumed until dryness. This step was repeated an additional three times. The vial was then cooled at room temperature under nitrogen pressure. The contents were resolubilized in the suitable solvents.

General methods for analysis of radiofluorination reactions

Radiochemical incorporation yields were determined by radioTLC. EMD TLC Silica gel 60 plates (10 x 2 cm) were spotted with an aliquot (1-5 μL) of crude reaction mixture approximately 1.5 cm from the bottom of the plate (baseline). TLC plates were developed in a chamber containing ethyl acetate (EtOAc) until within 2 cm of the top of the plate (front). Analysis was performed using a Bioscan AR-2000 radio-TLC imaging scanner and WinScan software. Radiochemical identity and purity were determined by radioHPLC. A Phenomenex Luna C18, 250 x 4.6 mm, 5 µm HPLC column was used for the analytical analysis with a Waters 1515 Isocratic HPLC Pump equipped with a Waters 2487 Dual λ Absorbance Detector, a Bioscan Flow-Count equipped with a NaI crystal, and Breeze software. The mobile phase for analytical

HPLC analysis was MeCN in 0.1 M NH4·HCO2 (aq); The flow rate was 1mL/min. The semi- preparative purifications were performed on a Phenomenex Luna C18, 250 x 100 mm, 5 µm

138

HPLC column with MeCN in 0.1 M NH4·HCO2 (aq) and the flow rate was 5 mL/min. All radiochemical yields are non-decay corrected. Each radiochemical labeling was conducted at least two times (n ≥ 2).

Optimization of radiofluorination conditions

Optimizations of the 1st step radiosynthesis of [18F]3-2a

(1) Method:

5 mg tetraethylammonium bicarbonate (TEAB) was used as a base to dry [18F]fluoride. 0.5 mL MeCN was used to solubilize the [18F]TEAF; 2.0 mg precursor (3-1a) was added into the reaction vial, to which the indicated amount of [18F]TEAF in MeCN was added. The resulting reaction mixture was tested at different reaction conditions, and the resulting RCCs were determined by RadioTLC chromatography.

(2) RadioTLC chromatogram of 1st step reaction mixture:

1 2 3 4 5 mean standard deviation

RCC (%) 99 98 95 96 99 98 2

139

(3) RadioHPLC chromatogram: a) Analytical analysis:

Column: Phenomenex Luna, C18, 250 x 4.6 mm, 5 µm

Mobile phase: 70% MeCN, 30% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

γ 254 nm

0 2 4 6 8 10 12 14 16 18 20 Time (min) b) Preparative HPLC purification:

Column: Phenomenex Luna C18, 250 x 100 mm, 5 µm

Mobile phase: 65% MeCN, 35% 0.1 M NH4·HCO2(aq);

Flow rate: 5 mL/min;

Optimizations of the 2nd step radiosynthesis of [18F]3-3a

(1) Method:

18 ® The isolated [ F]3-2a was diluted with 30 mL H2O, concentrated with t-C18 plus Sep-Pak cartridge (Waters; pre-activated with 10 mL EtOH, 10 mL air, followed by 10 mL H2O), and

140 then eluted with 1 mL MeCN. This solution was divided into several aliquots (0.2 mL each) into separate vials, containing the indicated amounts of AgNO3, Na2S2O8 and Selectfluor. Then, 0.2 mLH2O and EtOH was added to each vial, and the reaction was allowed to react at specified conditions. At the end, each reaction was quenched with 0.4 mL mobile (70% MeCN, 30% 0.1

M NH4·HCO2(aq)), and the resulting RCCs were determined by RadioHPLC chromatography.

(2) RadioHPLC chromatogram:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm Mobile phase: 70% MeCN, 30% 0.1 M

NH4·HCO2(aq);

Flow rate: 1 mL/min;

γ 254 nm

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time (min)

Application of the optimal reaction condition of [18F]3-3a toward [18F]3-3b to [18F]3-3i.

1st step radiosynthesis of [18F]3-2b

(1) Method:

The same optimal reaction conditions as that of [18F]3-2a were used with 2.0 mg compound 3- 1b. 141

(2) RadioTLC chromatogram of [18F]3-2b:

1 2 3 4 5 mean standard deviation

RCC (%) 99 98 95 96 99 98 2

(3) RadioHPLC chromatogram: a) Analytical analysis:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

Mobile phase: 50% MeCN, 50% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

γ 254 nm

0 2 4 6 8 10 12 14 16 18 20 Time (min)

b) Preparative HPLC purification:

Column: Phenomenex Luna C18, 250 x 100 mm, 5 µm

142

Mobile phase: 50% MeCN, 50% 0.1 M NH4·HCO2(aq);

Flow rate: 5 mL/min;

2nd step radiosynthesis of [18F]3-3b

(1) Method:

The same optimal reaction conditions as that of [18F]3-3a were used for the synthesis of [18F]3- 3b, but no EtOH was added.

18 Entry EtOH (µL) AgNO3 (mg) Na2S2O8 (mg) Selectfluor (mg) [ F]3-3b RCC (%) 1 20.0 0.0 8.0 20.0 0 (n = 2) 2 0.0 4.0 8.0 20.0 10 ± 2% (n = 2) 3 0.0 0.0 8.0 20.0 10 ± 2% (n = 3)

(2) RadioHPLC chromatogram:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

Mobile phase: 50% MeCN, 50% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

γ 254 nm

0 2 4 6 8 10 12 14 16 18 20 Time (min)

143

1st step radiosynthesis of [18F]3-2c

(1) Method:

The same optimal reaction conditions as that of [18F]3-2a were used with 2.0 mg compound 1c.

(2) RadioTLC chromatogram of [18F]3-2c:

1 2 3 4 5 mean standard deviation

RCC (%) 99 98 97 94 99 98 2

(3) RadioHPLC chromatogram: a) Analytical analysis:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

Mobile phase: 60% MeCN, 40% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

144

γ 254 nm

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time (min) b) Preparative HPLC purification:

Column: Phenomenex Luna C18, 250 x 100 mm, 5 µm

Mobile phase: 60% MeCN, 40% 0.1 M NH4·HCO2(aq);

Flow rate: 5 mL/min;

2nd step radiosynthesis of [18F]3-3c

(1) Method:

The same optimal reaction conditions as that of [18F]3-3a were used for the synthesis of [18F]3- 3c, but no EtOH was added.

18 Entry EtOH (µL) AgNO3 (mg) Na2S2O8 (mg) Selectfluor (mg) [ F]3-3c RCC (%) 1 20.0 0.0 8.0 20.0 0 (n = 2) 2 0.0 4.0 8.0 20.0 12 ± 2% (n = 2) 3 0.0 0.0 8.0 20.0 15 ± 3% (n = 3)

(2) RadioHPLC chromatogram:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

Mobile phase: 60% MeCN, 40% 0.1 M NH4·HCO2(aq); 145

Flow rate: 1 mL/min;

γ 254 nm

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time (min)

1st step radiosynthesis of [18F]3-2d

(1) Method:

The same optimal reaction conditions as that of [18F]3-2a were used with 2.0 mg compound 3- 1d.

(2) RadioTLC chromatogram of [18F]3-2d:

146

1 2 3 4 5 mean standard deviation

RCC (%) 99 98 97 94 98 98 2

(3) RadioHPLC chromatogram: a) Analytical analysis:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

Mobile phase: 50% MeCN, 50% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

γ 254 nm

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time (min)

b) Preparative HPLC purification:

Column: Phenomenex Luna C18, 250 x 100 mm, 5 µm

Mobile phase: 50% MeCN, 50% 0.1 M NH4·HCO2(aq);

Flow rate: 5 mL/min;

2nd step radiosynthesis of [18F]3-3d

147

(1) Method:

The same optimal reaction conditions as that of [18F]3-3a were used for the synthesis of [18F]3- 3d, but no EtOH was added.

18 Entry EtOH (µL) AgNO3 (mg) Na2S2O8 (mg) Selectfluor (mg) [ F]3-3d RCC (%) 1 20.0 0.0 8.0 20.0 0 (n = 2) 2 0.0 4.0 8.0 20.0 8 ± 1% (n = 2) 3 0.0 0.0 8.0 20.0 12 ± 2% (n = 3) 4 0.0 4.0 8.0 40.0 8 ± 1% (n = 2) 5 0.0 4.0 16.0 20.0 5 ± 2% (n = 2) 6 0.0 8.0 16.0 20.0 5 ± 2% (n = 2)

(2) RadioHPLC chromatogram:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

Mobile phase: 50% MeCN, 50% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

γ 254 nm

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time (min)

1st step radiosynthesis of [18F]3-2e

148

(1) Method:

The same optimal reaction conditions as that of [18F]3-2a were used with 2.0 mg compound 3- 1e.

(2) RadioTLC chromatogram of [18F]3-2e:

1 2 3 4 5 mean standard deviation

RCC (%) 80 92 85 82 86 85 5

(3) RadioHPLC chromatogram: a) Analytical analysis:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

Mobile phase: 60% MeCN, 40% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

149

γ 254 nm

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time (min)

b) Preparative HPLC purification:

Column: Phenomenex Luna C18, 250 x 100 mm, 5 µm

Mobile phase: 60% MeCN, 40% 0.1 M NH4·HCO2(aq);

Flow rate: 5 mL/min;

2nd step radiosynthesis of [18F]3-3e

(1) Method:

The same optimal reaction conditions as that of [18F]3-3a were used for the synthesis of [18F]3- 3e, but no EtOH was added.

18 Entry EtOH (µL) AgNO3 (mg) Na2S2O8 (mg) Selectfluor (mg) [ F]3-3e RCC (%) 1 20.0 0.0 8.0 20.0 0 (n = 2) 2 0.0 4.0 8.0 20.0 12 ± 2% (n = 2) 3 0.0 0.0 8.0 20.0 28 ± 3% (n = 3)

(2) RadioHPLC chromatogram:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

150

Mobile phase: 60% MeCN, 40% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

γ 254 nm

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time (min)

1st step radiosynthesis of [18F]3-2f

(1) Method:

The same optimal reaction conditions as that of [18F]3-2a were used with 2.0 mg compound 3-1f.

(2) RadioTLC chromatogram of [18F]3-2f:

151

1 2 3 4 5 mean standard deviation

RCC (%) 67 75 73 68 70 70 3

(3) RadioHPLC chromatogram: a) Analytical analysis:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

Mobile phase: 25% MeCN, 75% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

3-2f

γ 254 nm

0 2 4 6 8 10 12 14 16 18 20 Time (min)

b) Preparative HPLC purification:

Column: Phenomenex Luna C18, 250 x 100 mm, 5 µm

Mobile phase: 10% MeCN, 90% 0.1 M NH4·HCO2(aq);

Flow rate: 5 mL/min;

2nd step radiosynthesis of [18F]3-3f

(1) Method: 152

The same optimal reaction conditions as that of [18F]3-3a were used for the synthesis of [18F]3- 3f, but no EtOH was added, and 40.0 mg Selectfluor was used.

18 Entry EtOH (µL) AgNO3 (mg) Na2S2O8 (mg) Selectfluor (mg) [ F]3-3f RCC (%) 1 20.0 0.0 8.0 20.0 0 (n = 2) 2 0.0 0.0 8.0 20.0 18 ± 3% (n = 2) 3 0.0 4.0 8.0 40.0 20 ± 2% (n = 2) 4 0.0 0.0 8.0 40.0 45 ± 5% (n = 3)

(2) RadioHPLC chromatogram:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

Mobile phase: 25% MeCN, 75% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

γ 254 nm

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Time (min)

1st step radiosynthesis of [18F]3-2g

(1) Method:

The same optimal reaction conditions as that of [18F]3-2a were used with 2.0 mg compound 3- 1g.

153

(2) RadioTLC chromatogram of [18F]3-2g:

1 2 3 4 5 mean standard deviation

RCC (%) 98 98 97 96 99 98 2

(3) RadioHPLC chromatogram: a) Analytical analysis:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

Mobile phase: 50% MeCN, 50% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

γ 254 nm

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time (min)

b) Preparative HPLC purification:

Column: Phenomenex Luna C18, 250 x 100 mm, 5 µm

154

Mobile phase: 50% MeCN, 50% 0.1 M NH4·HCO2(aq);

Flow rate: 5 mL/min;

2nd step radiosynthesis of [18F]3-3g

(1) Method:

The same optimal reaction conditions as that of [18F]3-3a were used for the synthesis of [18F]3- 3g, but no EtOH was added.

18 Entry EtOH (µL) AgNO3 (mg) Na2S2O8 (mg) Selectfluor (mg) [ F]3-3g RCC (%) 1 20.0 0.0 8.0 20.0 0 (n = 2) 2 0.0 4.0 8.0 20.0 18 ± 2% (n = 3) 3 0.0 0.0 8.0 20.0 20 ± 3% (n = 3)

(2) RadioHPLC chromatogram:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

Mobile phase: 50% MeCN, 50% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

155

γ 254 nm

0 2 4 6 8 10 12 14 16 18 Time (min)

1st step radiosynthesis of [18F]3-2h

(1) Method:

The same optimal reaction conditions as that of [18F]3-2a were used with 2.0 mg compound 3- 1h.

(2) RadioTLC chromatogram of [18F]3-2h:

1 2 3 4 5 mean standard deviation

RCC (%) 99 98 97 94 99 98 2

(3) RadioHPLC chromatogram:

156 a) Analytical analysis:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

Mobile phase: 50% MeCN, 50% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

γ 254 nm

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time (min) b) Preparative HPLC purification:

Column: Phenomenex Luna C18, 250 x 100 mm, 10 µm

Mobile phase: 50% MeCN, 50% 0.1 M NH4·HCO2(aq);

Flow rate: 5 mL/min;

2nd step radiosynthesis of [18F]3-3h

(1) Method:

The same optimal reaction conditions as that of [18F]3-3a were used for the synthesis of [18F]3- 3h, but no EtOH was added.

157

18 Entry EtOH (µL) AgNO3 (mg) Na2S2O8 (mg) Selectfluor (mg) [ F]3-3h RCC (%) 1 20.0 0.0 8.0 20.0 0 (n = 2) 2 0.0 4.0 8.0 20.0 10 ± 3% (n = 2) 3 0.0 0.0 8.0 20.0 15 ± 2% (n = 3)

(2) RadioHPLC chromatogram:

Column: Phenomenex Luna C18, 250 x 4.6 mm, 5 µm

Mobile phase: 50% MeCN, 50% 0.1 M NH4·HCO2(aq);

Flow rate: 1 mL/min;

γ 254 nm

0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Time (min)

1st step radiosynthesis of [18F]3-2i

(1) Method:

The same optimal reaction conditions as that of [18F]3-2a were used with 2.0 mg compound 3-1i.

(2) RadioTLC chromatogram of [18F]3-2i:

158

1 2 3 4 5 mean standard deviation

RCC (%) 98 97 97 94 99 98 2

2nd step radiosynthesis of [18F]3-3i

(1) Method:

The same optimal reaction conditions as that of [18F]3-3a were used for the synthesis of [18F]3- 3i, but no EtOH was added.

18 Entry EtOH (µL) AgNO3 (mg) Na2S2O8 (mg) Selectfluor (mg) [ F]3-3i RCC (%) 1 20.0 0.0 8.0 20.0 0 (n = 2) 2 0.0 4.0 8.0 20.0 0 (n = 2) 3 0.0 0.0 8.0 20.0 0 (n = 2) 4 0.0 0.0 0.0 4.0 0 (n = 2) 5 0.0 0.0 8.0 20.0 (F-TEDA-PF6) 0 (n = 2) 6 0.0 0.0 0.0 4.0 (F-TEDA-PF6) 0 (n = 2)

Result: the analytic HPLC showed only a peak at 2-3 min, indicating the desired product [18F]3- 3i was not formed and the intermediate [18F]3-2i was decomposed.

159

SA determinations of the 4-([18F]fluoromethyl)-1,1'-biphenyl([18F]3-2a) and 4- ([18F]difluoromethyl)-1,1'-biphenyl([18F]3-3a)

Step 1.

18 (1) Automated synthesis of [ F]3-2a by GE TracerLab FXFN method

Following completion of bombardment, the [18F]fluoride was transferred to the GE

TRACERlab™ FXFN radiosynthesis module via helium gas overpressure. A schematic diagram of the GE medical systems commercial TRACERlab™ FXFN radiosynthesis module used for the synthesis of [18F]3-2a is shown in Figure S1.

Figure S1. Schematic of the GE TRACERlab™ FXFN radiosynthesis module automated synthesis manifold for [18F]3-2a.

160

Automated synthesis involves the following: (1) azeotropic drying of [18F]fluoride; (2) [18F]fluorination; and (3) HPLC purification, followed by solid-phase formulation of the final product. The synthesis module was operated in the following sequences with numerical references to Figure S1.

[18F]Fluoride was produced by the 18O(p,n)18F nuclear reaction using a GE cyclotron and delivered to the radiosynthesis module via 10. The [18F]fluoride was quantitatively trapped on a QMA carbonate ion exchange solid phase extraction (SPE) light cartridge

(Waters; activated with 6 mL of trace grade H2O). Automated synthesis began with the elution of resin-bound [18F]fluoride using a solution

of tetraethylammonium bicarbonate (6 mg in 300μL H2O and 700μL CH3CN), pre-loaded into 1 and delivered to the reactor (12). The reaction mixture (12) was dried azeotropically by addition of 1 mL anhydrous

CH3CN, preloaded into 5, at 85 °C under N2 flow and vacuum over 8 min, then at 110 °C

under N2 flow and vacuum for 4 min.

Precursor 3-1a (6 mg in 0.6 mL CH3CN) pre-loaded into 3 was added to 12. The reactor was sealed via the closure of valve V13, V20 and V24 and the reaction mixture was heated to 130 ºC and this temperature was maintained for 10 min. The reaction mixture was then cooled to 40 °C, vented via valves V24 and V25, and

diluted with 1:1 CH3CN/ H2O (3 mL), pre-loaded into 6. The crude reaction mixture was eluted into 14 and the contents of 14 were transferred to

the HPLC loop via N2 pressure via a fluid detector, injected onto a semi-preparative column (Luna C18 semi-preparative, 250 × 10.00 mm, 5µ), and eluted with 65:35

CH3CN/ 0.1M ammonium formate by volume at a flow rate of 5 mL/min. The eluent was monitored by UV (λ = 254 nm) and radiochemical detectors connected in series. The fraction containing the major radiochemical product was collected, via valve 18, into a large dilution vessel (15), which was preloaded with 23 mL of sterile water for injection (United States Pharmacopeia (USP); Hospira). The diluted HPLC fraction was then loaded onto a C18 SPE cartridge (16) (Waters;

preactivated with 5 mL EtOH followed by 10 mL H2O).

161

Cartridge 16 was washed with 10 mL sterile water for injection, USP, preloaded into 7, to remove traces of salts, HPLC mobile phase, and [18F]fluoride. Then 16 was eluted with

2.0 mL CH3CN preloaded into 8, into collection vial 17. We next took 0.4 mL of product solution and carried out the step 2. Analyses of radioactive mixtures were performed by HPLC with an in-line UV (λ = 254 nm) detector in series with a CsI PIN diode radioactivity detector. Uncorrected radiochemical yields of [18F]3-2a were 61% (non-decay corrected) relative to starting [18F]fluoride.

(2) RadioHPLC chromatogram from GE TRACERlab™ FXFN radiosynthesis module

(3) Specific Activity (SA) determination for [18F]3-2a:

162

3000000 y = 5E+14x + 6252.7 2500000 R² = 0.9998 2000000

1500000

1000000

500000

0 0 1E-09 2E-09 3E-09 4E-09 5E-09 6E-09

Standard curve for the specific activity determination of [18F]3-2a.

Step 2. Manual Labeling of [18F]3-3a

(1) Method:

18 The isolated [ F]3-2a in 2.0 mL CH3CN from FXFN was diluted with 2.0 mL sterile water, then

0.4 mL above solution was added to a reaction vial containing 8.0 mg Na2S2O8 and 20.0 mg Selectfluor and 20 µL EtOH was added (Attention: the vial cap should be closed tightly to avoid solvent evaporation.) The reaction was allowed to react at 120 °C for 10 min, after which it was cooled immediately by an ice-water bath, and quenched with 1.6 mL mobile phase (MeCN/ 0.1

M NH4·HCO2(aq), v/v 6.5:3.5). Preparative HPLC purification (Phenomenex Luna C18, 250 x 100 mm, 5 µm, 5 mL/min) led to the desired [18F]3-3a (1.5 mL mobile phase). The chemical & radiochemical purities and SA of [18F]3-3a was determined via analytical HPLC (Phenomenex

Luna C18, 250 x 4.6 mm, 5 µm, 70% MeCN, 30% 0.1 M NH4·HCO2(aq), 1 mL/min).

(2) Quality Control of [18F]3-3a: a) Chemical & radiochemical purity

163

Impurities from mobile phase

b) Co-injection with standard 3-3a

(3) Specific Activity (SA) determination for [18F]3-3a:

600000 y = 5E+14x 500000 R² = 0.9928

400000

300000

200000

100000

0 0.00E+00 2.00E-10 4.00E-10 6.00E-10 8.00E-10 1.00E-09 1.20E-09

Standard curve for the specific activity determination of [18F]3-3a.

164

References

1. Hagmann, W. K. The many roles for fluorine in medicinal chemistry. J. Med. Chem. 2008, 51, 4359-69. 2. Jeschke, P. The unique role of fluorine in the design of active ingredients for modern crop protection. Chembiochem. 2004, 5, 571-589. 3. Muller, K.; Faeh, C.; Diederich, F. Fluorine in pharmaceuticals: looking beyond intuition. Science 2007, 317, 1881-1886. 4. Purser, S.; Moore, P. R.; Swallow, S.; Gouverneur, V. Fluorine in medicinal chemistry. Chem. Soc. Rev. 2008, 37, 320-330. 5. Wang, J.; Sanchez-Rosello, M.; Acena, J. L.; del Pozo, C.; Sorochinsky, A. E.; Fustero, S.; Soloshonok, V. A.; Liu, H. Fluorine in pharmaceutical industry: fluorine-containing drugs introduced to the market in the last decade (2001-2011). Chem. Rev. 2014, 114, 2432-506. 6. Brooks, A. F.; Topczewski, J. J.; Ichiishi, N.; Sanford, M. S.; Scott, P. J. Late-stage [18F]Fluorination: New Solutions to Old Problems. Chem. Sci. 2014, 5, 4545-4553. 7. Buckingham, F.; Gouverneur, V. Asymmetric 18F-fluorination for applications in positron emission tomography. Chem. Rev. 2016, 7, 1645-1652. 8. Campbell, M. G.; Ritter, T. Modern carbon-fluorine bond forming reactions for aryl fluoride synthesis. Chem. Rev. 2015, 115, 612-633. 9. Jacobson, O.; Kiesewetter, D. O.; Chen, X. Fluorine-18 radiochemistry, labeling strategies and synthetic routes. Bioconjug. Chem. 2015, 26, 1-18. 10. Preshlock, S.; Tredwell, M.; Gouverneur, V. (18)F-Labeling of Arenes and Heteroarenes for Applications in Positron Emission Tomography. Chem. Rev. 2016, 116, 719-766. 11. Tredwell, M.; Gouverneur, V. 18F labeling of arenes. Angewandte Chemie 2012, 51, 11426-37. 12. Rewcastle, G. W.; Gamage, S. A.; Flanagan, J. U.; Frederick, R.; Denny, W. A.; Baguley, B. C.; Kestell, P.; Singh, R.; Kendall, J. D.; Marshall, E. S.; Lill, C. L.; Lee, W. J.; Kolekar, S.; Buchanan, C. M.; Jamieson, S. M.; Shepherd, P. R. Synthesis and biological evaluation of novel analogues of the pan class I phosphatidylinositol 3-kinase (PI3K) inhibitor 2-(difluoromethyl)-1-[4,6-di(4-morpholinyl)-1,3,5-triazin- 2-yl]-1H-benzimidazole (ZSTK474). J. Med. Chem. 2011, 54, 7105-7126. 13. Meanwell, N. A. Synopsis of some recent tactical application of bioisosteres in drug design. J. Med. Chem. 2011, 54, 2529-2591. 14. Kaneko, S.; Yamazaki, T.; Kitazume, T. A remarkably simple route to versatile difluoromethylated molecules. J. Org. Chem. 1993, 58, 2302–2312. 15. Goure, W. F.; Leschinsky, K. L.; Wratten, S. J.; Chupp, J. P. Synthesis and herbicidal activity of N-substituted 2,6-bis(polyfluoromethyl)dihydropyridine-3,5-dicarboxylates. J. Agric. Food Chem. 1991, 39, 981–986. 16. Anderson, M. O.; Zhang, J.; Liu, Y.; Yao, C.; Phuan, P. W.; Verkman, A. S. Nanomolar potency and metabolically stable inhibitors of kidney urea transporter UT-B. J. Med. Chem. 2012, 55, 5942-50. 17. Yuan, G.; Wang, F.; Stephenson, N. A.; Wang, L.; Rotstein, B. H.; Vasdev, N.; Tang, P.; Liang, S. H. Metal-free 18F-labeling of aryl-CF2H via nucleophilic radiofluorination and oxidative C-H activation. Chem. Commun. 2016, 53, 126-129. 18. Erickson, J. A.; McLoughlin, J. I. Hydrogen Bond Donor Properties of the Difluoromethyl Group. J. Org. Chem. 1995, 60, 1626–1631. 19. Middleton, W. J. New Fluorinating Reagents - Dialkylaminosulfur Fluorides. J. Org. Chem. 1975, 40, 574-578. 20. Markovskij, L. N.; Pashinnik, V. E.; Kirsanov, A. V. 1973. 21. Fujikawa, K.; Fujioka, Y.; Kobayashi, A.; Amii, H. A New Method for Aromatic Difluoromethylation: Copper-Catalyzed Cross-Coupling and Decarboxylation Sequence from Aryl Iodides. Org. Lett. 2011, 13, 5560-5563.

165

22. Fujikawa, K.; Kobayashi, A.; Amii, H. An Efficient Route to Difluoromethylated Pyridines. Synthesis (Stuttg) 2012, 44, 3015-3018. 23. Ge, S. Z.; Chaladaj, W.; Hartwig, J. F. Pd-Catalyzed alpha-Arylation of alpha,alpha- Difluoroketones with Aryl Bromides and Chlorides. A Route to Difluoromethylarenes. J. Am. Chem. Soc. 2014, 136, 4149-4152. 24. Fujiwara, Y.; Dixon, J. A.; Rodriguez, R. A.; Baxter, R. D.; Dixon, D. D.; Collins, M. R.; Blackmond, D. G.; Baran, P. S. A New Reagent for Direct Difluoromethylation. J. Am. Chem. Soc. 2012, 134, 1494-1497. 25. Liu, X.; Xu, C.; Wang, M.; Liu, Q. Trifluoromethyltrimethylsilane: nucleophilic trifluoromethylation and beyond. Chem. Rev. 2015, 115, 683-730. 26. Eujen, R.; Hoge, B.; Brauer, D. J. Synthesis and properties of donor-free bis(difluoromethyl)cadmium, (CF2H)(2)Cd - NMR spectroscopic detection and structure of tetrakis(difluoromethyl)cuprate(III) and related compounds. J. Organomet. Chem. 1996, 519, 7-20. 27. Burton, D. J.; Hartgraves, G. A. The preparation of HCF2CdX and HCF2ZnX via direct insertion into the carbon halogen bond of CF2HY (Y = Br, I). J Fluorine Chem 2007, 128, 1198-1215. 28. Hagiwara, T.; Fuchikami, T. Difluoroalkylation of Carbonyl-Compounds with (1,1- Difluoroalkyl)Silane Derivatives. Synlett. 1995, 717-718. 29. Fier, P. S.; Hartwig, J. F. Copper-Mediated Difluoromethylation of Aryl and Vinyl Iodides. J. Am. Chem. Soc. 2012, 134, 5524-5527. 30. Jiang, X. L.; Chen, Z. H.; Xu, X. H.; Qing, F. L. Copper-mediated difluoromethylation of electron-poor aryl iodides at room temperature. Org. Chem. Front. 2014, 1, 774-776. 31. Matheis, C.; Jouvin, K.; Goossen, L. J. Sandmeyer Difluoromethylation of (Hetero- )Arenediazonium Salts. Org. Lett. 2014, 16, 5984-5987. 32. Igoumnov, S. M.; Boyko, V. E.; Tyutyunov, A. A. Difluoromethylation of Carbonyl Compounds with (Difluoromethyl)trimethylsilane. Fluorine Notes 2011, 5. 33. Prakash, G. K. S.; Ganesh, S. K.; Jones, J. P.; Kulkarni, A.; Masood, K.; Swabeck, J. K.; Olah, G. A. Copper-Mediated Difluoromethylation of (Hetero)aryl Iodides and beta-Styryl Halides with Tributyl(difluoromethyl)stannane. Angew. Chem. Int. Ed. 2012, 51, 12090-12094. 34. Wang, H.; Vicic, D. A. Organometallic Aspects of Fluoroalkylation Reactions with Copper and Nickel. Synlett. 2013, 24, 1887-1898. 35. Dubinina, G. G.; Ogikubo, J.; Vicic, D. A. Structure of Bis(trifluoromethyl)cuprate and Its Role in Trifluoromethylation Reactions. Organometallics 2008, 27, 6233-6235. 36. Gu, Y.; Leng, X. B.; Shen, Q. L. Cooperative dual palladium/silver catalyst for direct difluoromethylation of aryl bromides and iodides. Nat. Commun. 2014, 5, 5405-5411. 37. Yin, G. Y.; Kalvet, I.; Englert, U.; Schoenebeck, F. Fundamental Studies and Development of Nickel-Catalyzed Trifluoromethylthiolation of Aryl Chlorides: Active Catalytic Species and Key Roles of Ligand and Traceless MeCN Additive Revealed. J. Am. Chem. Soc. 2015, 137, 4164-4172. 38. Xu, L.; Vicic, D. A. Direct Difluoromethylation of Aryl Halides via Base Metal Catalysis at Room Temperature. J. Am. Chem. Soc. 2016, 138, 2536-2539. 39. Xia, J. B.; Zhu, C.; Chen, C. Visible Light-Promoted Metal-Free C-H Activation: Diarylketone- Catalyzed Selective Benzylic Mono- and Difluorination. J. Am. Chem. Soc. 2013, 135, 17494-17500. 40. Xu, P.; Guo, S.; Wang, L.; Tang, P. Silver-catalyzed oxidative activation of benzylic C-H bonds for the synthesis of difluoromethylated arenes. Angew. Chem. Int. Ed. 2014, 53, 5955-5958. 41. Ma, J. J.; Yi, W. B.; Lu, G. P.; Cai, C. Transition-metal-free C-H oxidative activation: persulfate- promoted selective benzylic mono- and difluorination. Org. Biomol. Chem. 2015, 13, 2890-2894. 42. Patrick, T. B.; Johri, K. K.; White, D. H. Fluorodecarboxylation with Xenon Difluoride. J. Org. Chem. 1983, 48, 4158-4159. 43. Patrick, T. B.; Johri, K. K.; White, D. H.; Bertrand, W. S.; Mokhtar, R.; Kilbourn, M. R.; Welch, M. J. Replacement of the Carboxylic-Acid Function with Fluorine. Can. J. Chem. 1986, 64, 138-141. 44. Patrick, T. B.; Khazaeli, S.; Nadji, S.; Heringsmith, K.; Reif, D. Mechanistic Studies of Fluorodecarboxylation with Xenon Difluoride. J. Org. Chem. 1993, 58, 705-708.

166

45. Ramsden, C. A.; Shaw, M. M. Fluorodecarboxylation, rearrangement and cyclisation: the influence of structure and environment on the reactions of carboxylic acids with xenon difluoride. Tetrahedron Lett. 2009, 50, 3321-3324. 46. Rueda-Becerril, M.; Sazepin, C. C.; Leung, J. C. T.; Okbinoglu, T.; Kennepohl, P.; Paquin, J. F.; Sammis, G. M. Fluorine Transfer to Alkyl Radicals. J. Am. Chem. Soc. 2012, 134, 4026-4029. 47. Leung, J. C. T.; Chatalova-Sazepin, C.; West, J. G.; Rueda-Becerril, M.; Paquin, J. F.; Sammis, G. M. Photo-fluorodecarboxylation of 2-Aryloxy and 2-Aryl Carboxylic Acids. Angew. Chem. Int. Ed. 2012, 51, 10804-10807. 48. Yin, F.; Wang, Z. T.; Li, Z. D.; Li, C. Z. Silver-Catalyzed Decarboxylative Fluorination of Aliphatic Carboxylic Acids in Aqueous Solution. J. Am. Chem. Soc. 2012, 134, 10401-10404. 49. Mizuta, S.; Stenhagen, I. S.; O'Duill, M.; Wolstenhulme, J.; Kirjavainen, A. K.; Forsback, S. J.; Tredwell, M.; Sandford, G.; Moore, P. R.; Huiban, M.; Luthra, S. K.; Passchier, J.; Solin, O.; Gouverneur, V. Catalytic decarboxylative fluorination for the synthesis of tri- and difluoromethyl arenes. Org. Lett. 2013, 15, 2648-2651. 50. Teare, H.; Robins, E. G.; Kirjavainen, A.; Forsback, S.; Sandford, G.; Solin, O.; Luthra, S. K.; Gouverneur, V. Radiosynthesis and evaluation of [18F]Selectfluor bis(triflate). Angew. Chem. Int. Ed. 2010, 49, 6821-6824. 51. Verhoog, S.; Pfeifer, L.; Khotavivattana, T.; Calderwood, S.; Collier, T. L.; Wheelhouse, K.; Tredwell, M.; Gouverneur, V. Silver-Mediated 18F-Labeling of Aryl-CF3 and Aryl-CHF2 with 18F- Fluoride. Synlett. 2016, 27, 25-28. 52. Shi, H.; Braun, A.; Wang, L.; Liang, S. H.; Vasdev, N.; Ritter, T. Synthesis of F-18- Difluoromethylarenes from Aryl (Pseudo) Halides. Angew. Chem. Int. Edit. 2016, 55, 10786-10790. 53. Lee, K. C.; Lee, S.-Y.; Choe, Y. S.; Chi, D. Y. Metabolic stability of [18F]fluoroalkylbiphenyls. Bull. Korean Chem. Soc. 2004, 25, 1225–1230. 54. Liu, Y.; Esteva-Font, C.; Yao, C.; Phuan, P. W.; Verkman, A. S.; Anderson, M. O. 1,1- Difluoroethyl-substituted triazolothienopyrimidines as inhibitors of a human urea transport protein (UT- B): new analogs and binding model. Bioorg. Med. Chem. Lett. 2013, 23, 3338-3341. 55. Furuya, T.; Strom, A. E.; Ritter, T. Silver-Mediated Fluorination of Functionalized Aryl Stannanes. J. Am. Chem. Soc. 2009, 131, 1662-1663. 56. Serguchev, Y. A.; Ponomarenko, M. V.; Lourie, L. F.; Fokin, A. A. Transannular additions of selectfluor and xenon difluoride: regioselectivity and mechanism. J. Phys. Org. Chem. 2011, 24, 407-413. 57. Nyffeler, P. T.; Duron, S. G.; Burkart, M. D.; Vincent, S. P.; Wong, C. H. Selectfluor: mechanistic insight and applications. Angew. Chem. Int. Ed. 2004, 44, 192-212. 58. Lapi, S. E.; Welch, M. J. A historical perspective on the specific activity of radiopharmaceuticals: What have we learned in the 35 years of the ISRC? Nucl. Med. Biol. 2013, 40, 314-320. 59. de Goeij, J. J. M.; Bonardi, M. L. How do we define the concepts specific activity, radioactive concentration, carrier, carrier-free and no-carrier-added? J. Radioanal. Nucl. Chem. 2005, 263, 13-18. 60. Ichiishi, N.; Brooks, A. F.; Topczewski, J. J.; Rodnick, M. E.; Sanford, M. S.; Scott, P. J. H. Copper-Catalyzed [F-18]Fluorination of (Mesityl)(aryl)iodonium Salts. Org. Lett. 2014, 16, 3224-3227. 61. Furuya, T.; Strom, A. E.; Ritter, T. Silver-mediated fluorination of functionalized aryl stannanes. J. Am. Chem. Soc. 2009, 131, 1662-1663.

167

168

Chapter 4: Molecular modeling of FAAH and MGL targeting the cannabinoid system

4.1 Background

Cannabis sativa has long been known for its various medical properties by many cultures all over the world dating back to 1000 to 1500 B.C.1-2 The major psychoactive substrate underlying the pharmacological effects of cannabis is (-)-trans-Δ9-tetrahydrocannabinol (9-THC, Figure 4-

1), which has been identified and isolated since the 1960s.3-4 The neurobiological principal mechanisms of action of the cannabinoid drugs were gradually elucidated with the observations of their adenylate cyclase inhibiting activity in neuroblastoma cells5, followed by the discovery

6-7 of the two distinct G protein-coupled receptors, cannabinoid receptor 1 (CB1) and cannabinoid

8 receptor 2 (CB2) , as well as the two endocannabinoids (eCBs), 2-arachidonoylglycerol (2-AG, an analog of arachidonic acid)9-10 and N-arachidonolethanolamine (anandamide; AEA, Figure 4-

1)11.

Figure 4-1. Chemical structures of 9-THC, 2-AG and AEA.

Endocannabinoids, as important retrograde messenger lipids, exert their biological actions through the activations of CB1 and CB2 receptors, which are also the molecular targets for the psychoactive component of marijuana, 9-THC.12 Unlike hydrophilic neurotransmitters, 2-AG,

AEA and other N-acyl ethanolamines (NAEs) were synthesized by neurons on demand through

169 activity-dependent cleavage of membrane lipid precursors.13-14 The biological activity of these agonists are local and transient in the central nervous system and in peripheral tissues due to the rapid enzymatic degradations, instead of being absorbed and stored into synaptic vesicles.13-14

The Gi/o-coupled protein CB1 receptor (CB1R) is enriched in the central nervous system

15-17 (CNS), whereas the Gi -coupled protein CB2 receptor (CB2R) is abundant in particular in the

8, 18-19 immune cells and in the peripheral tissues. In neurons, CB1 receptor is typically located at the presynaptic terminal. Activation of CB1R by 2-AG or AEA suppresses the release of neurotransmitter (e.g. glutamate or GABA) from the presynaptic terminal.20-21

2-AG is postsynaptically biosynthesized by phospholipase C (PLC) and diacylglycerol lipase-α (DAGLα) from the phospholipid precursors, once the depolarization occurs in the postsynaptic terminal, such as activation of metabotropic glutamate receptor 1.22-23 2-AG then serves to tune-down /counteract the synaptic transmission by activating on the presynaptic CB1R in a retrograde manner. Monoacylglycerol lipase (MGL) is the key hydrolytic enzyme for 2-AG degradation, which accounts for 85% of 2-AG hydrolysis and presents at the presynaptic terminal. On the other hand, the other two α/β-hydrolase domain containing serine hydrolases 6 and 12 (i.e. ABHD6 and ABHD12) contribute approximately 4%, and 9% respectively.24-25

ABHD6 and ABHD12 can limit 2-AG activity at the site of synthesis.

N-acyl phosphatidylethanolamine phospholipase D (NAPE-PLD) and later identified enzymes of α/β-hydrolase 4 (Abh4) and phospholipase C (PLC) are responsible for the biosynthesis of AEA.26-27 NAPE-PLD mainly presents in the presynaptic terminal, though it might be also found in the postsynaptic region. Fatty acid amide hydrolase (FAAH) has been identified as the principal enzyme responsible for AEA degradation to arachidonic acid and ethanolamine.26-27 It also capable of hydrolyzing the noncannabinoid fatty acid amides (FAAs),

170 such as the anti-inflammatory factor N-palmitoyl ethanolamine (PEA), the satiating signal N- oleoyl ethanolamine (OEA), the sleep-inducing compound oleamide and the anti-proliferative agent N-acyl taurines (NATs).28-29 FAAH is expressed at the postsynaptic terminal, thus AEA can modulate the synaptic transmission in both autocrine and retrograde manner.28-29

The endocannabinoids, cannabinoid receptors, specific transporters and the enzymes responsible for synthesizing and degrading the eCBs form the architecture of the endocannabinoid system (ECS, Figure 4-2).30-32 The ECS is involved in a wide range of biological functions and the cannabinoid receptors have been the therapeutic targets for controlling pain,33 obesity,34 epilepsy35, anxiety,36 and schizophrenia.37

Figure 4-2. Schematic overview of eCB system and 2-AG and AEA metabolism.

Reproduced with permission from Elsevier.32

171

Direct activation of CBs by 9-THC and other agonists has been known to produce medicinally beneficial effects, such as the antinociception.38-39 However, the undesirable side effects, such as cognition impairments and motor dysfunctions, have limited their therapeutic applications. In addition, the cannabimimetic effects induced by 2-AG and AEA are weak and transient due to their rapid degradations by FAAH and MGL respectively.38-39

On the other hand, blockade of the activity of hydrolytic enzymes (i.e. FAAH and MGL) has been an attractive approach to maintain or elevate the levels of eCBs to achieve the beneficial effects with lesser untoward side effects than global CBs activation.40 Because inhibition of either FAAH or MGL only triggers one signaling pathway and the resulting increased signaling eCBs only act at their sites of biosynthesis, this provides a temporal and spatial pharmacological regulation that is not available to the CB agonists. Pharmacological inhibition by irreversible

(e.g. URB597, Figure 4-3) and reversible (e.g. OL-135, Figure 4-3) inhibitors of FAAH, or genetic knockout (FAAH (-/-)), in mice has displayed analgesia, anti-inflammation, anxiolysis and anti-depression effects without the side effects of hypothermia, cognition impairments and mobility dysfunction.40-45

Figure 4-3. Structures of URB597 and OL-135 as FAAH inhibitors.

MGL inhibitors also indicated their potential for the treatment of cancer and neuroflammatory diseases.46 For example, the arachidonic acid generated from the catabolism of

172

2-AG by MGL serves as a substrate for the biosynthesis of neuroinflammatory prostaglandins via cyclooxygenases. Therefore, inhibition of MGL may help to regulate the level of prostaglandin and to reduce the inflammation in neurodegenerative diseases.46

In this section, the development of pharmacological inhibitors targeting FAAH and MGL is introduced. The structures of the FAAH and MGL enzymes are also illustrated to facilitate the following molecular modeling assisted drug design. The covalent FAAH inhibitors BIA 10-2474 and PF-04457845 were investigated and compared via both molecular modeling and PET imaging to interrogate the adverse effects of BIA 10-2474 in clinical trials. Novel covalent and noncovalent inhibitors targeting MGL were developed and radiolabeled via either fluorine-18 or carbon-11 to allow the PET imaging studies.

4.2 Investigation of BIA 10-2474 as irreversible FAAH inhibitor

Both FAAH and MGL are serine hydrolases and act as homodimers. The scope of viable substrates of FAAH and MGL are also partially overlapped, considering that FAAH is also able to hydrolyze esters.47 However, they are structurally different enzymes. FAAH is a large integral enzyme with a molecular weight of 63 kD and it belongs to the amidase signature superfamily.48

Whereas, MGL has a smaller molecular weight of 33 kD and is categorized as a member of α/β- hydrolase superfamily due to sequence similarity. Though MGL is a membrane-associated enzyme, cytosolic MGL has also been reported.25

4.2.1 FAAH as a drug discovery target

FAAH is the only well-characterized amidase signature class of serine hydrolases with an unusual Ser-Ser-Lys catalytic triad (i.e. Ser241-Ser217-Lys142). The catalytic mechanism of

173

AEA metabolism is illustrated in Scheme 4-1.47-51 It starts from the nucleophilic attack of the catalytic Ser241 onto the carbonyl group of the substrate (S1) to form the tetrahedral intermediate (S2), followed by the collapse of this intermediate to release the amine and furnish the covalently bound acyl-enzyme intermediate (S3). Hydrolysis of the acyl-enzyme intermediate (S3) by activated water (S4) regenerates the enzyme and releases the free fatty acid

(S5). In this mechanism, Lys142 acts as a general base to deprotonate Ser241 and a general acid to protonate the substrate leaving group, both of which are mediated through the hydroxyl group of Ser217.47-51 FAAH hydrolyzes primary amides two-fold faster than ethanolamides with a preference for the arachidonoyl or oleoyl substrates.52-53

Scheme 4-1. Catalytic reaction mechanism of AEA with FAAH.

174

The binding pocket of FAAH consists of a series of channels and cavities, including the acyl-chain binding pocket (ABP) that is adjacent to the catalytic site, the membrane access channel (MAC) leading the active site to the membrane and the cytosolic port (CP) connecting the active site to the cytoplasm.54-56 The ABP and MAC parts of FAAH have hydrophobic surfaces to interact with the substrate acyl chain. On the other hand, the CP is a hydrophilic cavity that normally interacts with the head group of the substrates and provides an exit for the hydrophilic products from the active site to the cytoplasm. For example, the covalently bound

FAAH (dimer) in complex with URB59 (PDB ID: 3LJ7) is shown in Figure 4-4.57

175

Figure 4-4. FAAH in complex with URB 597 and the location of MAC, ABP and CP.

Reproduced with permission from Elsevier.57

To date, various classes of FAAH inhibitors have been developed due to the therapeutic potential of FAAH inhibitors in the treatment of pain,42, 44, 58 inflammation59 and sleep disorders60-61. Among them, three classes of inhibitors have shown potential in therapeutic intervention, including the irreversible carbamates (e.g. URB59743 and URB69462, Figure 4-5) and urea derivatives (e.g. PF-75063, LY218324064 and PF-0445784565) that covalently bind to the catalytic Ser241 residue in FAAH as well as the reversible α-ketoheterocycle-based

176 inhibitors (e.g. OL-13566) that bind to FAAH by reversible hemiketal intermediate. URB597 is

43 one of the earliest drug-like FAAH inhibitors with an IC50 of 4.0 nM. It showed efficacy against neuropathic pain in a mouse model.67 LY-2183240 demonstrated both analgesic and anxiolytic effects in animal models, however it was not used in clinical trials since it lacks

68 selectivity. PF-04457845 is a potent (i.e. IC50 = 7.2 nM) and extraordinarily selective inhibitor toward FAAH.65 It showed both analgesic and anti-inflammatory effects in animal models.69-70

As a drug for the treatment of osteoarthritis, it completed Phase II clinical trials, but failed to show efficacy.71 Inhibition of FAAH by OL-135 induces analgesia or anti-inflammatory activity at lower doses compared with other medications without significant offsite target activity.45, 72-74

Figure 4-5. Chemical structures of typical FAAH inhibitors.

177

4.2.2 Investigation of the adverse side effects of BIA 10-2474

In a recent phase 1 clinical trial, the putative Figure 4-6. Structure of BIA 10-2474. FAAH inhibitor BIA 10-2474 (Figure 4-6),

developed by Bial-Portela & Ca. SA, caused severe

adverse events. Acute and rapid toxic cerebral

syndrome developed among three of the four participants after four days’ drug administration, including headache, memory impairment, altered consciousness, and amnesia.75 One patient became brain dead; one of the other two patients had residue memory impairment and the other one had a residue cerebellar syndrome.

The precise mechanism of the toxic effects is still unknown. The Temporary Specialist Scientific

Committee (TSSC), responsible for the investigation of this tragedy, believed the adverse effects came from the direct toxicity of BIA 10-2474,75 which showed nonlinear pharmacokinetics at doses higher than 40-100 mg, however, the highest cumulative dose of BIA 10-2474 was 250-

300 mg in this clinical trial.76 They also indicated that the off-target effects due to the poor selectivity of BIA 10-2474 toward FAAH or a metabolite might be the reasons of the observed toxic cerebral syndrome.76 The complexity also lies in the fact that there are no severe toxic effects in CNS ever reported because of the increased level of eCBs.77 In addition, the distribution of the brain lesions is not exactly consistent with the localizations of ECS, such as the pons region. For example, the lesions also occurred on pons where endocannabinoid receptors and FAAH are much less expressed than in hippocampi.77

However, limited information about BIA 10-2474 was disclosed by Bial Co. before their

Phase 1 clinical trial. In their patent, they reported this compound a 22% inhibition at 100 nM in rat brain homogenates and a 98% inhibition ex vivo in mice treated with 3 mg/kg BIA 10-2474.78

178

The TSSC reported BIA 10-2474 with an IC of Figure 4-7. Structure of [18F]DOPP. 50 79 1.1-1.7 µM in rat. However, in a recent

publication, Wilson et al. determined BIA 10-2474

as a potent FAAH inhibitor with IC50s of 50-70

µg/kg.80 These results were obtained by ex and in vivo experiments in rat brain using a selective FAAH radiotracer [18F]DOPP (Figure 4-7).81 The

IC50 of BIA 10-2474 was calculated from a sigmoidal dose-response curve, i.e. the SUV of

[18F]DOPP binding to FAAH corresponded to the pretreatment doses of BIA ranging from 1-500

µg/kg. BIA 10-2474 was found to display comparable inhibition effects as other reported potent

FAAH inhibitors, such as URB597 (0.15 mg/kg),82 URB694 (40 µg/kg),83 and PF-04457845 (>

30 but <100 µg/kg).69 However, this method might result in errors due to the variations/changes of radiotracer delivery under blockade conditions, such as the flow limitations.80

Herein, the binding potency of BIA 10-2474 against FAAH was investigated in comparison with the well-documented potent FAAH inhibitor PF-04457845 via molecular docking studies in both AUTODOCK / YASARA and Glide.84-85 Meanwhile, the experiments were set up to determine the binding potency and selectivity of BIA 10-2474 as a FAAH inhibitor. More importantly, the target engagement studies via PET imaging of the either carbon-11 or fluorine-

18 labeled BIA 10-2474 and its metabolite/analogue are also ongoing to elucidate the mechanism underlying BIA 10-2474 induced cerebral toxicity.

179

4.2.3 Molecular docking of BIA 10-2474 and PF-04457845

Series of crystal structures of FAAH55-57, 86- Figure 4-8. Structure of α-ketooxadiazole. 92 have been reported since the first disclosure of

a 2.8 Å crystal structure of FAAH in complex

with the irreversible inhibitor methoxy

54 arachidonyl fluorophosphonate (MAP).

Notably, the high resolution crystal structures of humanized variant of rat FAAH in complex with the exceptionally potent covalent binder α-ketooxadiazole (Ki = 290 pM, Figure 4-8) were not only available in its covalent binding pose (PDB ID: 3PRO, 2.2 Å) but also the form of noncovalent binding (PDB ID: 3PPM, 1.78 Å) via the inclusion of fluoride ions.93 Considering the facts that both the BIA 10-2474 and PF-04457845 are covalent FAAH inhibitors and the molecular modeling experiments would start from the noncovalent binding study to investigate the accommodations of the parent compounds in the FAAH binding pocket before the subsequent covalent binding predictions, the crystal structure of 3PPM was selected for the molecular docking studies.93

In Glide docking, the centroid of the docking grid box was set up as the center of the catalytic Ser241 residue and the ligand diameter midpoint box was extended from the default 10

Å to 20 Å to accommodate the larger size of PF-04457845. Both Glide standard precision (SP) and extra-precision (XP) were used to predict the binding interactions.

The noncovalent binding interaction of PF-04457845 in FAAH is shown in Figure 4-9. PF-

04457845 projects deeply into the hydrophobic channel (membrane access channel, MAC), thus forming one extra π-π stacking interaction with Trp531, besides the key π-π stacking interactions with Phe192, Phe381, and Phe432. In the acyl-chain binding pocket (ABP), the carbonyl oxygen

180 atom forms an H-bond with Ile238, which is also hydrogen bonding with one of the pyridazien nitrogens. In addition, the carbamate NH group forms a hydrogen bond with Met191, though only this H-bonding interaction was shown in the YASARA rendered image (Figure 4-9). The

Glide XP docking gives PF-04457845 an XP docking score of -12.7 kcal/mol.

Figure 4-9. Noncovalent docking result of PF-04457845 in FAAH (PDB ID 3PPM).

Picture is rendered from YASARA. Interactions are shown in cyan dotted lines for H-bonds, blue for cation-π stacking, and red for π-π stacking. The red circles indicate the binding regions of PF-04457845 in FAAH.

On the other hand, BIA 10-2474 mainly occupies the ABP and CP regions of the binding pocket. It has the π-π stacking interaction with Phe192 and H-bonding to Lys142. The Glide XP docking gives BIA 10-2474 a XP docking score of -6.9 kcal/mol (Figure 4-10).

181

Figure 4-10. Noncovalent docking result of BIA 10-2474 in FAAH (PDB ID: 3PPM).

Picture is rendered from YASARA. Interactions are shown in red for π-π stacking.

The covalent docking poses of both BIA 10-2474 and PF-04457845 were generated via both

AUTODOCK as embedded in YASARA and Glide covalent docking predictions.84-85 YASARA covalent docking method demonstrated more favorable binding interactions than those from

Glide covalent docking. The AUTODOCK / YASARA docking pose (Figure 4-11A) showed one extra H-bond of the carbonyl group with Gly239 and extra π-π stacking interactions with

Phe381 and Trp531 with a final energy of -198622.9 kcal/mol. One the other hand, the Glide covalent docking (Figure 4-11B) gave a prime energy of -23720.4 kcal/mol and a ΔG of -94.90 kcal/mol.

182

Figure 4-11. Covalent docking results of PF-04457845 in FAAH. (A)

(B)

Pictures are rendered from YASARA. Results were obtained from YASARA covalent docking (A), and Glide covalent docking (B). Interactions are shown in cyan dotted lines for H-bond, blue for cation-π stacking, and red for π-π stacking.

183

The irreversible binding mechanism of PF-04457845 is proposed as shown in Scheme 4-2, referring to the covalent binding behavior of URB597.57 Nucleophilic attack of the catalytic

Ser241 onto the carbonyl group of PF-04457845 forms the tetrahedral intermediate, which was then collapsed to release the covalently bounded PF-04457845-FAAH complex and pyridazin-3- amine.

Scheme 4-2. Covalent binding mechanism of PF-04457845 with FAAH.

YASARA covalent docking of BIA 10-2474 resulted in a different binding pose from that generated from the Glide covalent docking. In the YASARA rendered pose (Figure 4-12A), the

N-oxide atom forms two H-bonds with Cys144 and Thr236 separately and π-π stacking interaction with Phe192, which gave a final energy of -175034.3 kcal/mol. On the other hand,

Glide covalent docking (Figure 4-12B) gave π-π stacking interactions with Phe244, Tyr194, and

Phe192. The carbonyl group also forms two H-bonds with Ile238 and Gly239. The Glide

184 covalent binding pose gave a prime energy of -23675.7 kcal/mol and a ΔG of -34.4 kcal/mol.

Figure 4-12. Covalent docking results of BIA 10-2474 in FAAH. (A) (B)

In summary, the molecular modeling studies of PF-04457845 and BIA 10-2474 indicate that

PF-0457845 is a potent and selective hFAAH covalent binder, which is consistent with its

65 reported potency of Ki = 7.2 nM. However, molecular modeling predicts BIA 10-2474 to be a much weaker binder and poor selective reagent toward hMGL, which also matches its experimentally determined µM binding potency and poor selectivity toward hMGL.79 The experimental binding results of BIA 10-2474 are expected soon.

185

4.3 Development of novel radiotracers for the PET imaging of MGL

4.3.1 MGL as a drug discovery target

MGL shares common structural features as other /-hydrolases with a highly conserved - sheet core and a more variable lid or cap domain.94-95 The -sheet core consists of seven parallel strands and one antiparallel strand and is surrounded by -helices. The lid domain hides the core region of the enzyme, regulating the access of the substrates to the catalytic site. The crystallizations of human MGL have provided more insights into the architecture of this enzyme and its catalytic sites. In 2010, Bertrand et al. and Labar et al. independently disclosed the crystal structures of MGL in their apo states as asymmetric dimers (PDB ID: 3JW8 and 3HJU respectively, Figure 4-13).96-97

Figure 4-13. Key structural features of MGL-crystallized structure of 3HJU.

Picture is rendered from YASARA.

186

According to them, MGL has two additional -helices (4 and 5), besides the other commonly shared six helices (1-3 and 6-8). Helices 4, 5, 6 and the loops between them comprise the lid domain with residues 151-225, where a wide U-shaped structure was delineated by 4 and the loop connecting 4 to 5 to gate the substrate access to the catalytic sites.32, 96-97

The catalytic triad Ser122-Asp239-His269 is buried at the bottom of a long channel, which is delimited by hydrophobic residues to accommodate the lipophilic chain of 2-AG. The environment around the catalytic triad, however, becomes more hydrophilic. The three regulatory cysteine residues, Cys201, Cys208 and Cys242, are situated within the loop connecting α5 and α6 helices, the top of helix 6 and the vicinity of catalytic triad respectively.

The nucleophilic residue Ser122 locates on a turn between helix 3 and 5, also called the

“nucleophilic elbow”. A hydrogen-bond network is formed among residues Asp239, His269, and

Ser122 to maximize the catalytic activity of Ser122.32, 96-97

On the loop that connects 1 and 3, an oxyanion hole is constituted by the backbone NH groups of Ala51 and Met123 to stabilize the anionic transition state of the hydrolysis reaction. A polar cleft was also formed by the backbone of Ala51, the backbone NH group of Met123, the hydroxyl group of Tyr58, the imidazolic nitrogen atoms of His121 and 272, the guanidinium group of Arg57 and the carboxylate group of Glu53. At the entrance of this polar cleft, an alcohol-binding pocket, consisting of the carbonyl group of Ala51, the carboxylate group of

Glu53 and the hydroxyl group of Tyr94, is proposed to accommodate the polar head of 2-AG in the crystal structure of 3HJU (Figure 4-13).32, 96-97

187

The MGL crystal structures in complex with different inhibitors further elucidate the enzymatic active site. Bertrand et al. also reported the covalently bounded MGL crystal structure with the carbamate-based inhibitor SAR629 (PDB ID 3JWE).96 This inhibitor mimics the pathway of 2-AG cleavage by making a stable carbamate adduct with Ser132 (i.e. Ser122 in

3HJU). SAR629 adopts a Y-shaped conformation in MGL with two fluorophenyl groups spanning into the lipophilic portion within the access channel and the carbonyl oxygen atom projecting into the polar oxyanion hole (Figure 4-14). Griebel et al. disclosed a crystal structure of MGL in complex with the covalent inhibitor SAR127303 (PDB ID: 4UUQ).98 This inhibitor also formed a carbamate covalent bond with MGL but its binding modes in the two monomers are different due to the structural variants in chain A and chain B, especially in the flexible lid domain. A high-resolution crystal structure of MGL in complex with the noncovalent reversible inhibitor ZYH has been resolved by Schalk-Hihi et al. in 2011 (1.35 Å, PDB ID: 3PE6).99 ZYH occupies the entire binding region of MGL. The amide carbonyl group projects into the oxyanion hole to form polar interactions with Met123. The 3-(4-pyrimidin-2-yl)piperazin-1-yl)azetidinyl part of ZYH situates in the alcohol-binding pocket as indicated in 3HJU. The 2- cyclohexylbenzoxazol-6-yl moiety spans into the lipophilic tunnel and forms van der Waals interactions with the enzyme.

Figure 4-14. Chemical structures of SAR629, SAR127303 and ZYH.

188

Noteworthy, the crystal structure 3PE6 disclosed a new closed conformation of the lid domain and the substantial conformational changes of helix 4 during the catalytic cycle (Figure

4-15, green).32, 99

Figure 4-15. Crystallized structure of MGL in complex with ZYH.

Picture is rendered from YASARA. MGL is colored grey except that α4 is highlighted with green color. Compound ZYH is colored cyan for its carbon atoms, red for its oxygen atoms and blue for its nitrogen atoms.

The crystal structures of 3HJU and 3JW8 adopt an open conformation in the lid domain, which might be attributed to the stabilizing effects of detergents and crystal soaking adjuvant used in the crystallization.96-97 However, 3PE6 was obtained in a more soluble form without using detergents, eliminating the affects from these reagents.97 It clearly defines that the movement of the 4 helix over the active site leads to the partially-closed conformation of the lid domain. The resulting hydrophobicity and electrostatic changes in the lid domain are

189 hypothesized for the dissociation of MGL from the membrane. After enzymatic hydrolysis of the substrate, the enzyme would resume the open conformation through rearrangements of the lid domain to re-associate to the membrane.97

As a primary enzyme to degrade the endogenous cannabinoid, 2-AG, MGL regulates the 2-

AG-mediated-signaling in a wide range of physiological and pathological processes, such as its regulation of pain,100 neural inflammation,101 and anxiety,102 and in its roles in obesity and metabolic syndromes103-104 and in emotional memory.102 Abnormal expression of MGL has also been observed in various human cancer cells and primary high-grade tumors. Pharmacological inhibition and genetic ablation of MGL have been reported to reduce cancer cells migration and aggressiveness as well as tumor growth.105-107

The identified selective and potent MGL inhibitors can be divided into three classes (Figure

4-16). The first class of inhibitors are noncompetitive and partially reversible, among which the

O-biphenylcarbamate URB602 enhances CB1 receptor-mediated stress-induced analgesia and shows antinociceptive effects in various models of pain.108-110 The second class of inhibitors targets the catalytic Ser122, such as the irreversible piperidine-carbamate JZL184111-112 and its analogs of JZL195,113 KML29,114 JW651,115 and SAR12730398 as well as the reversible inhibitors of CL6a116 and ZYH.99 SAR127303 displays beneficial effects including antiepileptic and antinociceptive effects in inflammatory and visceral pain models.98 The third class is the cysteine-targeting inhibitors, such as the irreversible N-arachidonylmaleimide (NAM), which inhibits the MGL via covalently bonding to Cys 201 and Cys242 residue.117-118 In addition, the isothiazolone and benzisothiazolinone derivatives have also been reported by King et al. as potent MGL inhibitors via a partially reversible mechanism involving Cys208 residue, such as octhilinone (Figure 4-13).119 Although blockade of MGL via these inhibitors exerts beneficial

190 effects, concerns about using MGL inhibitor arose due to observed adverse neuropsychiatric effects, such as anxiety and depression.120 In addition, the irreversible MGL inhibitors, such as

SAR127303, are found to have negative influences toward learning performances and memory function due to the long-term potentiation of CB1 synaptic transmission and acetylcholine levels.98

Figure 4-16. Chemical structures of typical MGL inhibitors.

191

4.3.2 Development of PET radiotracers for MGL

Understanding the target engagement of MGL inhibitors and its relationship to inhibitor dosing is essential to prompt MGL-targeted therapeutic discovery and to guide the dose selection for patients. As mentioned earlier, PET is a noninvasive, sensitive and selective molecular imaging tool to interrogate the drug-target interaction and can be translated to human brain imaging. However, application of this technique has been hampered by the lack of MGL inhibitors entering into clinical evaluations and by the absence of potent and selective PET imaging probes.121-122 Initially, five [11C-carbonyl]carbamate- and urea-based PET tracers, such as [11C]KML29, [11C]JW642, [11C]ML30 and [11C]JJKK-0048 (Figure 4-17), were attempted by

Wilson et al., however, these radioligands showed poor brain penetration (0.2-0.8 SUVs), precluding them as suitable for PET brain neuroimaging.121

Figure 4-17. Urea and carbamate based MGL PET tracers.

192

Another disclosed radioligand [11C]TZPU also suffered the same issue with a SUV of 0.5.122

The recently disclosed [11C]SAR127303, developed by two independent research groups, turned out to be the only successful PET probe for MGL and has the potential for translational research in human CNS MGL imaging (Figure 4-17).122-123 The [11C]SAR127303 has been evaluated in preliminary nonhuman primate studies, resulting in a good brain permeability (SUV > 1.5), good selectivity and binding patterns consistent with the distribution of MGL.122

Herein, efforts are made to design and synthesize novel PET probes targeting MGLs via molecular modeling both in CNS for neurological and psychiatric applications and peripheral system for MGL implications in cancer, obesity and metabolic syndromes.

4.4 Molecular modeling assisted drug design for novel MGL PET tracers

4.4.1 Investigation of the binding interaction of ZYH in MGL

The high resolution of MGL crystal structure in complex with reversible inhibitor ZYH

(PDB ID 3PE6) was used for the molecular modeling experiments.99 The protein was prepared via the protein preparation wizard in Glide before use. Both the standard precision (SP) and extra precision (XP) noncovalent docking methods were tried and compared during the docking experiments.124 The reliability of the docking protocol was tested at the beginning by docking the innate ligand ZYH into this crystal structure, followed by the comparisons of the resulting poses with the original binding pattern of ZYH in its crystal structure.

As shown in Figure 4-18, the docking pose of innate ligand ZYH in crystal structure overlaps well with these generated from either SP or XP docking experiments, except only small variations in the flexible cyclohexyl part. A closer examination shows XP gives a more

193 consistent binding pattern than from SP docking, which is also the case for most of the ligands tested in this chapter, especially the PAD series of compounds.

Figure 4-18. Comparisons of the docking results of ZYH from SP and XP methods, superimposed with the crystal structure position.

Picture is rendered from YASARA. The carbon atoms of ZYH are colored pale pink in SP docking result, canary yellow in XP docking result, and pale cyan in the crystal structure.

A detailed binding interaction of ZYH in the MGL binding pocket is shown in Figure 4-19.

The catalytic triad of MGL, including Ser122, Asp239, and His269, is positioned in the center of the active site and colored magenta. The structurally conserved network of hydrogen-bond donors, consisting of Gly50, Ala51, Met123 and Gly124, forms an oxyanion hole and serves to stabilize the anionic transition state of the catalytic reaction. The amide carbonyl group of ZYH points into the oxyanion hole and forms two critical hydrogen bonds with the backbone amide nitrogen atoms of Met123 and Ala51 separately. The azetidine-piperazine-pyrimidine part of

ZYH projects almost entirely into a narrow amphiphilic pocket of the active site. One of the piperazine nitrogen atoms and the pyrimidine nitrogen atoms form two hydrogen bonds with

Arg57 and Try194 separately through a water network involving two buried water molecules.

The pyrimidine ring also interacts with Tyr194 via face-to-face π–π stacking in this area. On the other hand, the benzoxazole-cyclohexane part of ZYH occupies a hydrophobic and more

194 spacious void, where the binding ligand is accessible to solvent and mostly forming van der

Waals interactions with the protein. ZYH gave a XP docking score of -11.39, corresponding to nanomolar binding potency.

Figure 4-19. XP docking result of ZYH.

Picture is rendered from YASARA. The MGL structure is colored grey and ZYH is colored canary yellow for carbon, red for oxygen, blue for nitrogen and yellow for sulfur. The H- bond is represented as cyan dotted line, and the three catalytic triad residues are colored magenta.

The covalent docking experiments were also carried out for the covalent binders targeting the Ser122 by either Glide covalent docking125 or an iterative energy minimization protocol in

YASARA (see experimental sections 1.4 & 1.5 for details).

4.4.2 Development of novel covalent MGL antagonists based on compound 4-1

Exploration of irreversible covalently bound antagonists to MGL in the CNS was based on a carbamate chemical structure of compound 4-1, which is disclosed in a conference as a potent, irreversible, selective and serine-active MGL inhibitor (Figure 4-20). Analogs of this compound were designed by varying the leaving group, which would affect the covalent binding behavior

195 of the irreversible inhibitors. The influence of the leaving group is dependent on their physicochemical properties besides the steric hindrance effect.126 For example, the binding potency of a class of β-lactam-based irreversible MGL inhibitors are affected by the pKa of the

127 leaving group’s conjugated acid with the ideal pKa identified to be around 10.

Figure 4-20. Compound 4-1 and its analogs bearing different leaving groups.

The noncovalent binding pose of compound 4-1 is demonstrated in Figure 4-21. The carbonyl group of compound 4-1 projects into the oxyanion hole and forms two hydrogen bonds with residues of Ala51 and Met123. Another hydrogen bond is formed between the oxygen atom of 1,2,4-oxadiazole and a water molecule. At the bottom of the binding pocket, His121 and

Tyr194 form π-π stacking interactions with 1,2,4-oxadiazole heterocylic ring and phenyl ring respectively.

196

Figure 4-21. Binding interaction of compound 4-1 in the MGL binding pocket.

Picture is rendered from Glide. The H-bond is represented as pink arrows, and the π-π stacking is shown in green line.

Compounds 4-2 to 4-6 give similar noncovalent docking poses and docking interactions as that of compound 4-1 (see experimental section 1.3 for details). The docking scores range from -

8.1 to -9.2 kcal/mol. Removal of the leaving groups of this class compounds after the hydrolytic enzymatic reaction would lead to the same covalent complex as that from compound 4-1 as shown in Figure 4-22. The covalent binding behavior of compound 4-1 in MGL was predicted by

AUTODOCK / YASARA and the resulting final energy was -109680 kcal/mol and was set up as a reference to evaluate the covalent binding potential of it analogs. The covalent bond is formed with the catalytic Ser122. The key binding interactions observed in the noncovalent binding mode is preserved.

197

Figure 4-22. Covalent binding interactions of compound 4-1 in the binding site of MGL.

Picture is rendered from YASARA. The carbon atoms are colored cyan for MGL residues, and green for compound 4-1. The H-bond is represented as cyan dotted line, and the name of the interacting residues are colored magenta.

Since the covalently bound complex for this class of compounds will be the same, the resulting noncovalent docking score could attribute to the observed binding potency of these compounds. As shown in Table 4-1, all the compounds give consistent binding potency results with their docking scores.

198

Table 4-1. Noncovalent docking results of compounds 4-1 to 4-6 and their binding potency. Compound Docking score (kcal/mol) Potency (nM) 4-1 -9.0 15.4 4-2 -8.1 NDa 4-3 -8.3 49 4-4 -9.2 79 4-5 -8.2 145 4-6 -8.8 307

a ND: not determined

The second class of analogs was designed based on the feasibility and convenience for late- stage installations of radioactive isotopes of fluorine-18 or carbon-11 for the following PET imaging studies (Figure 4-23).

Figure 4-23. Analogs of compound 4-1 with modifications on its core structure.

199

Noncovalent docking scores of the first class of compounds vary significantly, ranging from

-6.0 to -9.5 kcal/mol (see experimental section 1.3 for details). However, these compounds were experimentally determined as potent MGL inhibitors as compound 4-1 (Table 4-2).

Table 4-2. Noncovalent docking results of compounds 4-7 to 4-12 and their binding potency. Compound Docking score (kcal/mol) Potency (nM) 4-7 -6.0 10.5 4-8 -6.8 13.0 4-9 -7.1 10.3 4-10 -9.5 18.6 4-11 -6.5 14.1 4-12 -6.1 11.4

On the other hand, compounds with low docking scores also share similar key binding interactions as that of compound 4-1. For example, compound 4-7 has a docking score of -6.0 kcal/mol, but its key H-bonding interactions and π-π stacking interactions with residues Ala51,

Met123, His121 and Tyr194 are preserved (Figure 4-24). Therefore, the low scores are probably the result of lack of covalent bond as discussed below.

200

Figure 4-24. Binding interaction of compound 4-7 in the MGL binding pocket.

Picture is rendered from Glide. The H-bond is represented as pink arrows, and the π-π stacking is shown in green line.

As covalent MGL inhibitors, the covalent binding prediction of these compounds by

YASARA turned out to give better explanations for the observed binding potency. As shown in

Table 4-3, the final energies of these compounds, obtained from an iterative energy minimization protocol (see experimental section 1.4 for details), were similar or even lower than that of compound 4-1, which is consistent with their observed binding potency.

Table 4-3. Final energy of the covalent binding complex for compounds 4-6 to 4-12.

Compound 4-6 4-7 4-8 4-9 4-10 4-11 4-12 Final Energy -109700 -110800 -111500 -112200 -110500 -110000 -109600 (kcal/mol)

201

The AUTODOCK / YASARA predicted binding pose of these compounds would also enhance the binding interactions of these compounds compared with their Glide predicted noncovalent binding patterns. As shown in Figure 4-25, compound 4-7 forms the covalent bond with catalytic residue Ser122, meanwhile, its key interactions are preserved with extra H- bonding with a water molecule. The steric clashes of this compound with MGL are also reduced.

Figure 4-25. Covalent binding interactions of compound 4-7 in the binding site of MGL.

Picture is rendered from YASARA. The carbon atoms are colored cyan for MGL residues, and green for compound 4-7. The H-bond is represented as cyan dotted line, and the name of the interacting residues are colored magenta.

Radiosynthesis of these PET radioligands and subsequent evaluation of their potential as

MGL PET probes in the primate brain are ongoing.

202

4.4.3 Noncovalent MGL antagonists of PAD and its analogs

A novel 4-piperidinyl azetidine diamide scaffold PAD and its two analogs FEPAD and

FPPAD were used to explore reversible MGL inhibitors and PET radiotracers for the non-CNS indications of MGL (Figure 4-26).128

Figure 4-26. Structures of PAD, FEPAD and FPPAD.

The previously reported irreversible carbamate or urea based PET probes, such as

[11C]JJKK-0048 and [11C]SAR127303 (Figure 4-17), had the radiolabeling placed on the serine acylating moiety with the non-expelling portion. This new reversible chemotype, however, could allow the radioactive labeling on the sites other than the acylating carbonyl carbon atom.

The binding interactions of selected lead compound PAD and its fluoride derivatives (i.e.

FEPAD and FPPAD) were investigated by docking studies via Glide.129-130 XP docking was employed for the docking experiments. As shown in Figure 4-27, PAD is buried in an extended active site and stretches across almost the entire space of the binding pocket. The azetidine amide carbonyl group points into the oxyanion hole and establishes two H-bonds with the backbone amide nitrogen atoms of Met123 and Ala51 separately. The azetidine-piperazine-pyrimidine part of PAD projects almost entirely into a narrow amphiphilic pocket of the active site. One of the piperazine nitrogen atoms and the piperazine amide carbonyl group forms two hydrogen bonds with Arg57 and Try194 through a water network involving two buried water molecules. The pi electrons of the pyrimidine and thiazole rings are partially overlapped with those of Tyr194 for a

203

π–π stacking interaction. On the other hand, the substituted biphenyl part of PAD occupies a hydrophobic and more spacious part of the pocket, where the binding ligand is accessible to solvent and mostly forms van der Waals interactions with the protein. Moreover, the chlorine atom forms a halogen bond with Ser176.

Figure 4-27. (A) Overview of MGL in complex with PAD; (B) XP docking result of PAD.

Picture is rendered from YASARA. The MGL structure is colored grey and PAD is colored brown yellow for carbon, red for oxygen, blue for nitrogen, yellow for sulfur, light green for chlorine and green for fluorine. The H-bond is represented as cyan dotted line, the halogen bond is indicated as black dotted line, and the three catalytic triad residues are colored magenta for carbons.

Subsequent docking studies for FEPAD shows the key interactions of FEPAD are similar to that of PDA, except the fluoroethyl function group flips toward the lid of the binding pocket to accommodate itself (Figure 4-28). FEPAD had a binding potency of 23.8 nM and the resulting

XP docking score was -12.75 kcal/mol.

204

Figure 4- 28. XP docking result of FEPAD in the binding site of MGL.

Picture is rendered from YASARA. FEPAD is colored purple for carbon, and other element colorations are the same as Figure 4-27.

As indicated in Figure 4-29, the key interactions of FPPAD are also similar to that of PDA, except the fluoropropyl function group flips toward the lid of the binding pocket to accommodate itself, and the sulfur atom of thiazole lies syn to the piperazine amide carbonyl group. FPPAD featured a binding affinity of 13.2 nM, and the XP docking score was -13.04 kcal/mol.

205

Figure 4-29. XP docking result of FPPAD in the binding site of MGL.

Picture is rendered from YASARA. FPPAD is colored dark green for carbon, and other element colorations are the same as Figure 4-27.

An overlap of the resulting binding poses of ZYH, PAD, FEPAD, and FPPAD is shown in

Figure 4-30 for comparison. The piperazine amide carbonyl group in PAD and its derivatives replace the pyrimidine nitrogen in ZYH to H-bond to Arg57 through the water network. The decreasing overlap of the pi electrons of the pyrimidine and thiazole rings of the PAD, FEPAD and FPPAD series with Tyr194 reduces the corresponding π–π stacking interaction. In addition, because of the increased length of the alkoxyl substitutions in FEPAD and FPPAD, their halogen bonding with Ser176 is eliminated and the alkoxyl chains are flipped to project toward the lid domain, thus accommodating the increasing size of aliphatic substituents and providing more interaction energy through hydrophobic interactions with the surrounding residues. The sulfur atom in the thiazole group of FPPAD is also changed to the syn position relative to the piperazine amide carbonyl group. The predicted docking scores of PAD and its derivatives are consistent with the reported range and trend for their binding potencies, following an order of

206

FPPAD (13.2 nM, docking score -13.04 kcal/mol) > FEPAD (23.8 nM, docking score -12.75 kcal/mol) > PAD (26.4 nM, docking score -11.90 kcal/mol).

Figure 4-30. Superimposition of the binding poses of ZYH, PDA, FEPAD and FPPAD in the active site of MGL.

The carbon atoms are colored canary yellow for ZYH, brown for PAD, pink for FEPAD and dark green for FPPAD. Picture is rendered from YASARA.

The radiolabeling of these compounds was achieved at the phenol locus via either [11C]CH I 3 alkylation ([11C]PAD) or coupling reactions with 18F labeled short alkyl chains ([18F]FEPAD and

[18F]FPPAD) by our collaborators in National Institute of Radiological Science in Japan (Scheme

4-3). Subsequent PET imaging and ex vivo distribution studies validated the specificity of these

PET radioligands.

207

Scheme 4-3. Radiosynthesis of [11C]PAD, [18F]FEPAD and [18F]FPPAD.

The best candidate [18F]FEPAD was further utilized to perform an in vivo assessment between MGL activity and different thermogenesis conditions in brown adipose tissue, indicating its potential as a biomarker to image the lipolysis in metabolic disease (Figure 4-31).

Figure 4-31. Representative whole body PET images (0-90 min) of [11C]PAD, [18F]FEPAD and [18F]FPPAD.

208

4.5 Summary

Molecular modeling provides an efficient way to investigate the noncovalent and covalent binding behaviors of the newly designed compounds targeting FAAH and MGL. It provided guidance, assistance, and explanations for the drug design for both FAAH and MGL inhibitors, promoting their further studies in PET imaging. The molecular modeling of BIA 10-2474 revealed its poor binding affinity toward FAAH. The novel irreversible MGL inhibitors based on compound 4-1 and reversible inhibitors based on the core structure of PAD have led to successful lead compounds as both MGL inhibitors and potential PET imaging tracers. The PAD series, especially [18F]FEPAD, turned out to be useful PET tracers for imaging MGL in the peripheral system. On the other hand, most of the novel ligands based on compound 4-1 have proved to be potent MGL inhibitors; further studies of their radiolabled isotopologs are currently being evaluated in nonhuman primate models.

209

Experimental Section

1. Computational studies

All calculations performed in this work were carried out on two Cooler Master Centurion 5 (Intel

Core-i7 Quad CPU Q6600 @ 3.33GHz) operating system running Maestro 10.4 (Schrödinger,

LLC, New York, NY, 2015), ChemBioDraw® Ultra 14.0 (PerkinElmer) and YASARA (Yet

Another Scientific Artificial Reality Application).84 All pictures presented in this study were generated by Maestro and YASARA.

1.1 Molecule preparation

The ligands were prepared in ChemBio3D Ultra 14.0. The three-dimensional structures were built by importing the SMILES files of the ligands into LigPrep (LigPrep, version 3.6,

Schrödinger, LLC, New York, NY, 2015), implemented in Maestro 10.4. The LigPrep program employs the Potentials for Liquid Simulations-all atom (OPLS-AA) force field 2005 for energy minimizations and a cellular pH value 7.0 to generate the most probable ionization state of the ligand.131

1.2 Enzyme preparation

The FAAH (PDB ID: 3PPM) and MGL (PDB ID: 3PE6) enzymes were prepared using Maestro

10.4 Protein Preparation Wizard (Schrödinger, LLC, New York, NY, 2015) before docking.

Bond orders were assigned and the orientation of hydroxyl groups, amide groups of the side chains of Asn and Gln, and the protonation state of His residues were optimized. A restrained refinement of the protein structure was performed using the default constraint of 0.3 Å RMSD and the OPLS 2005 force field. The docking studies were performed by Glide (Grid-Based

210

Ligand Docking with Energetics) (Glide, version 10.4, Schrödinger, LLC, New York, NY,

2015).

1.3 Noncovalent docking via Glide

Both standard precision (SP) and extra precision (XP) docking protocols were carried out and the binding poses were examined. Water molecules were not included in these docking studies due to the creation of the homology model.132

Figure S4-1. Noncovalent binding interaction of compound 4-2 in the MGL binding pocket. SP docking score -8.07 kcal/mol. Picture is rendered from Glide. The H-bond is represented as pink arrows, and the π-π stacking is shown in green line.

211

Figure S4-2. Noncovalent binding interaction of compound 4-3 in the MGL binding pocket. SP docking score -8.34 kcal/mol.

Figure S4-3. Noncovalent binding interaction of compound 4-4 in the MGL binding pocket. SP docking score -9.18 kcal/mol.

212

Figure S4-4. Noncovalent binding interaction of compound 4-5 in the MGL binding pocket. SP docking score -8.16 kcal/mol.

Figure S4-5. Noncovalent binding interaction of compound 4-6 in the MGL binding pocket. SP docking score -9.17 kcal/mol.

213

Compounds 4-2 to 4-6 have different leaving groups, but with the same core structure. In general, the SP gives better docking score, which represent the lowest binding energy and favorable binding pose. According to the docking scores, these compounds could fit into the

MGL binding pocket very well.

Figure S4-6. Noncovalent binding interaction of compound 4-7 in the MGL binding pocket. SP docking score -6.0 kcal/mol.

214

Figure S4-7. Noncovalent binding interaction of compound 4-8 in the MGL binding pocket. SP docking score -6.8 kcal/mol.

215

Figure S4-8. Noncovalent binding interaction of compound 4-9 in the MGL binding pocket. SP docking score -7.1 kcal/mol.

Figure S4-9. Noncovalent binding interaction of compound 4-10 in the MGL binding pocket. SP docking score -9.5 kcal/mol.

Figure S4-10. Noncovalent binding interaction of compound 4-11 in the MGL binding pocket.

SP docking score -6.5 kcal/mol.

216

Figure S4-11. Noncovalent binding interaction of compound 4-12 in the MGL binding pocket.

SP docking score -6.1 kcal/mol.

The noncovalent docking scores of the newly designed compounds 4-7 to 4-12 were significantly reduced. However, the in vitro binding assay showed that these compounds had comparable or even better binding potency (Table 4-2). Therefore, the covalent docking behaviors of these compounds were further tested to justify the observed experimental binding results. Moreover, these compounds were designed as covalent binders of MGL.

217

1.4 Covalent docking via YASARA

The covalent binding interactions of these compounds were evaluated by using YASARA, following a protocol of iterative energy minimization of enzyme & ligand covalent binding complex:

Preparation of the enzyme & ligand covalently bound complex:

Load the PDB file of enzyme & ligand complex from Glide noncovalent docking result

into YASARA.

Manually build the covalent bond between the enzyme and binding ligand: Editdelete

atoms from O2 to F6, then Editadd bond between C1 and Serine122 OG.

Iterative energy minimization:

Select Force Field: SimulationForce Field. Yamber3. Set the speed to Normal. Then

click “OK”.133

Define the simulation cell: SimulationDefine simulation cell. Set size automatically:

Extend 5.0A. Then, click “around all atoms”.

Cell neutralization: OptionsChoose Experiment Cell neutralization and pka

prediction.

Default settings for the parameters.

Minimization: Options Choose ExperimentEnergy minimization.

Record the final energy and repeat the above process until the minimum energy is

obtained.

In the end, the resulting final energy from these covalent docking experiments were compared with the known covalent binder compound 4-1 to predict and explain their propensity and binding strength toward MGL.

218

This strategy turned out to be more reliable/reasonable in predicting and explaining their relative binding affinities toward MGL. As shown in Table 4-3, the final energies of these new derivatives were parallel or higher than that of compound 4-1. The resulting covalent docking poses for compounds 4-7 to 4-12 also demonstrated enhanced binding interactions in the MGL binding pocket after the above iterative energy minimizations. The detailed binding poses for each compound are shown below (Figure S4-12 to S4-17):

Figure S4-12. Covalent binding interaction of compound 4-7 in the binding site of MGL.Picture is rendered from YASARA. The carbon atoms are colored cyan for MGL residues, and green for compound 4-1. The H-bond is represented as cyan dotted line and the names of the interacting residues are colored magenta.

219

Figure S4-13. Covalent binding interaction of compound 4-8 in the binding site of MGL. Picture is rendered from YASARA.

Figure S4-14. Covalent binding interaction of compound 4-9 in the binding site of MGL. Picture is rendered from YASARA.

220

Figure S4-15. Covalent binding interaction of compound 4-10 in the binding site of MGL.

Picture is rendered from YASARA.

Figure S4-16. Covalent binding interaction of compound 4-11 in the binding site of MGL.

Picture is rendered from YASARA.

221

Figure S4-17. Covalent binding interaction of compound 4-12 in the binding site of MGL.

Picture is rendered from YASARA.

222

1.5 Covalent docking via Glide

Covalent docking method in Glide125 was used for the covalent binding predictions of BIA

10-2475 and PF-04457845 besides the above mentioned YASARA covalent docking. The protocol was introduced in Glide User Manual Chapter 5.

In Maestro 10.4, the covalent docking interaction of BIA 10-2474 can be realized as follows:

Load the PDB file of enzyme & ligand complex from Glide noncovalent docking result.

Application Glide Covalent docking.

Define the receptor: Pick reactive residue Ser241; Enclosing box: centroid of the

workspace ligand, box size: auto;

Reaction type: Custom the reaction type in a file as follows to build up the covalent bond

between Ser241 and BIA 10-2474:

LIGAND_SMARTS_PATTERN 2,O=CN(C)C

RECEPTOR_SMARTS_PATTERN 3,CCO

CUSTOM_CHEMISTRY ("<2>N(C)C1CCCCC1",("delete",[2,3,4,5,6,7,8,9]))

CUSTOM_CHEMISTRY ("<1>|<2>",("bond",1,(1,2)))

No constraints

Docking:

Docking mode: Pose prediction (through);

Refinement: Cutoff to retain poses for further refinement: 2.5 kcal/mol. Max number of

poses to retain for further refinement: 200.

Scoring: Calculate affinity score using Glide; Perform MM-GBSA scoring.

223

Output: 1 output pose per ligand reaction site, max number of top-scoring ligands to

report.

Name the job.

Run.

Similarly, Glide covalent docking result for PF-04457845 was obtained. The customized reaction type was as follows:

LIGAND_SMARTS_PATTERN 2,O=CNcn

RECEPTOR_SMARTS_PATTERN 3,CCO

CUSTOM_CHEMISTRY ("<2>Nc1cccnn1",("delete",[2,3,4,5,6,7,8]))

CUSTOM_CHEMISTRY ("<1>|<2>",("bond",1,(1,2)))

224

References

1. Adams, R. Marihuana: Harvey Lecture, February 19, 1942. Bull. N. Y. Acad. Med. 1942, 18, 705- 730. 2. Todd, A. R. Hashish. Experientia 1946, 2, 55-60. 3. Budzikiewicz, H.; Alpin, R. T.; Lightner, D. A.; Djerassi, C.; Mechoulam, R.; Gaoni, Y. [Mass spectroscopy and its application to structural and stereochemical problems. 68. Mass spectroscopic studies of constituents of hashish]. Tetrahedron 1965, 21, 1881-1888. 4. Gaoni, Y.; Mechoulam, R. The isolation and structure of delta-1-tetrahydrocannabinol and other neutral cannabinoids from hashish. J. Am. Chem. Soc.1971, 93, 217-24. 5. Howlett, A. C. Inhibition of Neuro-Blastoma Adenylate-Cyclase by Cannabinoid and Nantradol Compounds. Life Sci. 1984, 35, 1803-1810. 6. Devane, W. A.; Dysarz, F. A.; Johnson, M. R.; Melvin, L. S.; Howlett, A. C. Determination and Characterization of a Cannabinoid Receptor in Rat-Brain. Mol. Pharmacol. 1988, 34, 605-613. 7. Matsuda, L. A.; Lolait, S. J.; Brownstein, M. J.; Young, A. C.; Bonner, T. I. Structure of a Cannabinoid Receptor and Functional Expression of the Cloned Cdna. Nature 1990, 346, 561-564. 8. Munro, S.; Thomas, K. L.; Abushaar, M. Molecular Characterization of a Peripheral Receptor for Cannabinoids. Nature 1993, 365, 61-65. 9. Mechoulam, R.; Benshabat, S.; Hanus, L.; Ligumsky, M.; Kaminski, N. E.; Schatz, A. R.; Gopher, A.; Almog, S.; Martin, B. R.; Compton, D. R.; Pertwee, R. G.; Griffin, G.; Bayewitch, M.; Barg, J.; Vogel, Z. Identification of an Endogenous 2-Monoglyceride, Present in Canine Gut, That Binds to Cannabinoid Receptors. Biochem. Pharmacol. 1995, 50, 83-90. 10. Stella, N.; Schweitzer, P.; Piomelli, D. A second endogenous cannabinoid that modulates long- term potentiation. Nature 1997, 388, 773-778. 11. Devane, W. A.; Hanus, L.; Breuer, A.; Pertwee, R. G.; Stevenson, L. A.; Griffin, G.; Gibson, D.; Mandelbaum, A.; Etinger, A.; Mechoulam, R. Isolation and Structure of a Brain Constituent That Binds to the Cannabinoid Receptor. Science 1992, 258, 1946-1949. 12. Mackie, K. Cannabinoid receptors as therapeutic targets. Annu. Rev. Pharmacol. Toxicol. 2006, 46, 101-22. 13. Di Marzo, V. Targeting the endocannabinoid system: to enhance or reduce? Nat. Rev. Drug Discov. 2008, 7, 438-455. 14. Piomelli, D. The molecular logic of endocannabinoid signalling. Nat. Rev. Neurosci.2003, 4, 873- 884. 15. Hohmann, A. G.; Herkenham, M. Cannabinoid receptors undergo axonal flow in sensory nerves. Neuroscience 1999, 92, 1171-1175. 16. Walczak, J. S.; Pichette, V.; Leblond, F.; Desbiens, K.; Beaulieu, P. Behavioral, pharmacological and molecular characterization of the saphenous nerve partial ligation: A new model of neuropathic pain. Neuroscience 2005, 132, 1093-1102. 17. Farquhar-Smith, W. P.; Egertova, M.; Bradbury, E. J.; McMahon, S. B.; Rice, A. S. C.; Elphick, M. R. Cannabinoid CB1 receptor expression in rat spinal cord. Mol. Cell Neurosci. 2000, 15, 510-521.

225

18. Galiegue, S.; Mary, S.; Marchand, J.; Dussossoy, D.; Carriere, D.; Carayon, P.; Bouaboula, M.; Shire, D.; Lefur, G.; Casellas, P. Expression of Central and Peripheral Cannabinoid Receptors in Human Immune Tissues and Leukocyte Subpopulations. Eur. J. Biochem. 1995, 232, 54-61. 19. Atwood, B. K.; Mackie, K. CB2: a cannabinoid receptor with an identity crisis. Br. J. Pharmacol. 2010, 160, 467-479. 20. Castillo, P. E.; Younts, T. J.; Chavez, A. E.; Hashimotodani, Y. Endocannabinoid Signaling and Synaptic Function. Neuron 2012, 76, 70-81. 21. Ohno-Shosaku, T.; Kano, M. Endocannabinoid-mediated retrograde modulation of synaptic transmission. Curr. Opin. Neurobiol. 2014, 29, 1-8. 22. Katona, I.; Urban, G. M.; Wallace, M.; Ledent, C.; Jung, K. M.; Piomelli, D.; Mackie, K.; Freund, T. F. Molecular composition of the endocannabinoid system at glutamatergic synapses. Journal of Neuroscience 2006, 26, 5628-5637. 23. Suarez, J.; Ortiz, O.; Puente, N.; Bermudez-Silva, F. J.; Blanco, E.; Fernandez-Llebrez, P.; Grandes, P.; de Fonseca, F. R.; Moratalla, R. Distribution of Diacylglycerol Lipase Alpha, an Endocannabinoid Synthesizing Enzyme, in the Rat Forebrain. Neuroscience 2011, 192, 112-131. 24. Ueda, N. Endocannabinoid hydrolases. Prostaglandins Other Lipid Mediat. 2002, 68, 521-534. 25. Blankman, J. L.; Simon, G. M.; Cravatt, B. F. A comprehensive profile of brain enzymes that hydrolyze the endocannabinoid 2-arachidonoylglycerol. Chem. Biol. 2007, 14, 1347-1356. 26. Cravatt, B. F.; Giang, D. K.; Mayfield, S. P.; Boger, D. L.; Lerner, R. A.; Gilula, N. B. Molecular characterization of an enzyme that degrades neuromodulatory fatty-acid amides. Nature 1996, 384, 83-87. 27. McKinney, M. K.; Cravatt, B. F. Structure and function of fatty acid amide hydrolase. Annu. Rev. Biochem. 2005, 74, 411-432. 28. Schlosburg, J. E.; Kinsey, S. G.; Lichtman, A. H. Targeting fatty acid amide hydrolase (FAAH) to treat pain and inflammation. AAPS J. 2009, 11, 39-44. 29. Chatzakos, V.; Slatis, K.; Djureinovic, T.; Helleday, T.; Hunt, M. C. N-acyl taurines are anti- proliferative in prostate cancer cells. Lipids 2012, 47, 355-361. 30. Lutz, B.; Marsicano, G.; Maldonado, R.; Hillard, C. J. The endocannabinoid system in guarding against fear, anxiety and stress. Nat. Rev. Neurosci. 2015, 16, 705-718. 31. Lu, H. C.; Mackie, K. An Introduction to the Endogenous Cannabinoid System. Biol. Psychiat. 2016, 79, 516-525. 32. Scalvini, L.; Piomelli, D.; Mor, M. Monoglyceride lipase: Structure and inhibitors. Chem. Phys. Lipids 2016, 197, 13-24. 33. Lynch, M. E.; Ware, M. A. Cannabinoids for the Treatment of Chronic Non-Cancer Pain: An Updated Systematic Review of Randomized Controlled Trials. J. Neuroimmune Pharm. 2015, 10, 293- 301. 34. Kim, J.; Li, Y.; Watkins, B. A. Endocannabinoid signaling and energy metabolism: A target for dietary intervention. Nutrition 2011, 27, 624-632. 35. Reddy, D. S.; Golub, V. M. The Pharmacological Basis of Cannabis Therapy for Epilepsy. J. Pharm. Exp. Ther. 2016, 357, 45-55.

226

36. Marco, E. M.; Rapino, C.; Caprioli, A.; Borsini, F.; Laviola, G.; Maccarrone, M. Potential Therapeutic Value of a Novel FAAH Inhibitor for the Treatment of Anxiety. Plos One 2015, 10, e0137034. 37. Rohleder, C.; Muller, J. K.; Lange, B.; Leweke, F. M. Cannabidiol as a Potential New Type of an Antipsychotic. A Critical Review of the Evidence. Front. Pharmacol. 2016, 7, 422. 38. Hohmann, A. G. Spinal and peripheral mechanisms of cannabinoid antinociception: behavioral, neurophysiological and neuroanatomical perspectives. Chem. Phys. Lipids 2002, 121, 173-190. 39. Di Marzo, V. Targeting the endocannabinoid system: to enhance or reduce? Nat. Rev. Drug Discov. 2008, 7, 438-455. 40. Saario, S. M.; Laitinen, J. T. Therapeutic potential of endocannabinoid-hydrolysing enzyme inhibitors. Basic Clin. Pharmacol. Toxicol. 2007, 101, 287-293. 41. Wiskerke, J.; Irimia, C.; Cravatt, B. F.; De Vries, T. J.; Schoffelmeer, A. N.; Pattij, T.; Parsons, L. H. Characterization of the effects of reuptake and hydrolysis inhibition on interstitial endocannabinoid levels in the brain: an in vivo microdialysis study. ACS Chem. Neurosci. 2012, 3, 407-417. 42. Cravatt, B. F.; Demarest, K.; Patricelli, M. P.; Bracey, M. H.; Giang, D. K.; Martin, B. R.; Lichtman, A. H. Supersensitivity to anandamide and enhanced endogenous cannabinoid signaling in mice lacking fatty acid amide hydrolase. Proc. Natl. Acad. Sci. U.S.A. 2001, 98, 9371-9376. 43. Kathuria, S.; Gaetani, S.; Fegley, D.; Valino, F.; Duranti, A.; Tontini, A.; Mor, M.; Tarzia, G.; La Rana, G.; Calignano, A.; Giustino, A.; Tattoli, M.; Palmery, M.; Cuomo, V.; Piomelli, D. Modulation of anxiety through blockade of anandamide hydrolysis. Nat. Med. 2003, 9, 76-81. 44. Lichtman, A. H.; Shelton, C. C.; Advani, T.; Cravatt, B. F. Mice lacking fatty acid amide hydrolase exhibit a cannabinoid receptor-mediated phenotypic hypoalgesia. Pain 2004, 109, 319-327. 45. Lichtman, A. H.; Leung, D.; Shelton, C. C.; Saghatelian, A.; Hardouin, C.; Boger, D. L.; Cravatt, B. F. Reversible inhibitors of fatty acid amide hydrolase that promote analgesia: evidence for an unprecedented combination of potency and selectivity. J. Pharm. Exp. Ther. 2004, 311, 441-448. 46. Nomura, D. K.; Morrison, B. E.; Blankman, J. L.; Long, J. Z.; Kinsey, S. G.; Marcondes, M. C.; Ward, A. M.; Hahn, Y. K.; Lichtman, A. H.; Conti, B.; Cravatt, B. F. Endocannabinoid hydrolysis generates brain prostaglandins that promote neuroinflammation. Science 2011, 334 , 809-813. 47. Patricelli, M. P.; Cravatt, B. F. Fatty acid amide hydrolase competitively degrades bioactive amides and esters through a nonconventional catalytic mechanism. Biochemistry 1999, 38, 14125-14130. 48. Patricelli, M. P.; Lovato, M. A.; Cravatt, B. F. Chemical and mutagenic investigations of fatty acid amide hydrolase: evidence for a family of serine hydrolases with distinct catalytic properties. Biochemistry 1999, 38, 9804-9812. 49. McKinney, M. K.; Cravatt, B. F. Structure and function of fatty acid amide hydrolase. Annu. Rev. Biochem. 2005, 74, 411-432. 50. McKinney, M. K.; Cravatt, B. F. Evidence for distinct roles in catalysis for residues of the serine- serine-lysine catalytic triad of fatty acid amide hydrolase. J. Biol. Chem. 2003, 278, 37393-37399. 51. Patricelli, M. P.; Cravatt, B. F. Clarifying the catalytic roles of conserved residues in the amidase signature family. J. Biol. Chem. 2000, 275, 19177-19184. 52. Boger, D. L.; Fecik, R. A.; Patterson, J. E.; Miyauchi, H.; Patricelli, M. P.; Cravatt, B. F. Fatty acid amide hydrolase substrate specificity. Bioorg. Med. Chem. Lett. 2000, 10, 2613-2616.

227

53. Patricelli, M. P.; Cravatt, B. F. Characterization and manipulation of the acyl chain selectivity of fatty acid amide hydrolase. Biochemistry 2001, 40, 6107-6115. 54. Bracey, M. H.; Hanson, M. A.; Masuda, K. R.; Stevens, R. C.; Cravatt, B. F. Structural adaptations in a membrane enzyme that terminates endocannabinoid signaling. Science 2002, 298, 1793- 1796. 55. Mileni, M.; Garfunkle, J.; DeMartino, J. K.; Cravatt, B. F.; Boger, D. L.; Stevens, R. C. Binding and inactivation mechanism of a humanized fatty acid amide hydrolase by alpha-ketoheterocycle inhibitors revealed from cocrystal structures. J. Am. Chem. Soc.2009, 131, 10497-506. 56. Mileni, M.; Johnson, D. S.; Wang, Z.; Everdeen, D. S.; Liimatta, M.; Pabst, B.; Bhattacharya, K.; Nugent, R. A.; Kamtekar, S.; Cravatt, B. F.; Ahn, K.; Stevens, R. C. Structure-guided inhibitor design for human FAAH by interspecies active site conversion. Proc. Natl. Acad. Sci. U.S.A. 2008, 105, 12820- 12824. 57. Mileni, M.; Kamtekar, S.; Wood, D. C.; Benson, T. E.; Cravatt, B. F.; Stevens, R. C. Crystal structure of fatty acid amide hydrolase bound to the carbamate inhibitor URB597: discovery of a deacylating water molecule and insight into enzyme inactivation. J. Mol. Biol. 2010, 400, 743-754. 58. Cravatt, B. F.; Saghatelian, A.; Hawkins, E. G.; Clement, A. B.; Bracey, M. H.; Lichtman, A. H. Functional disassociation of the central and peripheral fatty acid amide signaling systems. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 10821-10826. 59. Karsak, M.; Gaffal, E.; Date, R.; Wang-Eckhardt, L.; Rehnelt, J.; Petrosino, S.; Starowicz, K.; Steuder, R.; Schlicker, E.; Cravatt, B.; Mechoulam, R.; Buettner, R.; Werner, S.; Di Marzo, V.; Tuting, T.; Zimmer, A. Attenuation of allergic contact dermatitis through the endocannabinoid system. Science 2007, 316, 1494-1497. 60. Huitron-Resendiz, S.; Gombart, L.; Cravatt, B. F.; Henriksen, S. J. Effect of oleamide on sleep and its relationship to blood pressure, body temperature, and locomotor activity in rats. Exp. Neurol. 2001, 172, 235-243. 61. Huitron-Resendiz, S.; Sanchez-Alavez, M.; Wills, D. N.; Cravatt, B. F.; Henriksen, S. J. Characterization of the sleep-wake patterns in mice lacking fatty acid amide hydrolase. Sleep 2004, 27, 857-865. 62. Mor, M.; Rivara, S.; Lodola, A.; Plazzi, P. V.; Tarzia, G.; Duranti, A.; Tontini, A.; Piersanti, G.; Kathuria, S.; Piomelli, D. Cyclohexylcarbamic acid 3 '- or 4 '-substituted biphenyl-3-yl esters as fatty acid amide hydrolase inhibitors: Synthesis, quantitative structure-activity relationships, and molecular modeling studies. J. Med. Chem. 2004, 47, 4998-5008. 63. Ahn, K.; Johnson, D. S.; Fitzgerald, L. R.; Liimatta, M.; Arendse, A.; Stevenson, T.; Lund, E. T.; Nugent, R. A.; Nomanbhoy, T. K.; Alexander, J. P.; Cravatt, B. F. Novel mechanistic class of fatty acid amide hydrolase inhibitors with remarkable selectivity. Biochemistry 2007, 46, 13019-13030. 64. Alexander, J. P.; Cravatt, B. F. The putative endocannabinoid transport blocker LY2183240 is a potent inhibitor of FAAH and several other brain serine hydrolases. J. Am. Chem. Soc. 2006, 128, 9699- 9704. 65. Johnson, D. S.; Stiff, C.; Lazerwith, S. E.; Kesten, S. R.; Fay, L. K.; Morris, M.; Beidler, D.; Liimatta, M. B.; Smith, S. E.; Dudley, D. T.; Sadagopan, N.; Bhattachar, S. N.; Kesten, S. J.; Nomanbhoy,

228

T. K.; Cravatt, B. F.; Ahn, K. Discovery of PF-04457845: A Highly Potent, Orally Bioavailable, and Selective Urea FAAH Inhibitor. ACS Med. Chem. Lett. 2011, 2, 91-96. 66. Boger, D. L.; Miyauchi, H.; Du, W.; Hardouin, C.; Fecik, R. A.; Cheng, H.; Hwang, I.; Hedrick, M. P.; Leung, D.; Acevedo, O.; Guimaraes, C. R.; Jorgensen, W. L.; Cravatt, B. F. Discovery of a potent, selective, and efficacious class of reversible alpha-ketoheterocycle inhibitors of fatty acid amide hydrolase effective as analgesics. J. Med. Chem.2005, 48, 1849-1856. 67. Russo, R.; Loverme, J.; La Rana, G.; Compton, T. R.; Parrott, J.; Duranti, A.; Tontini, A.; Mor, M.; Tarzia, G.; Calignano, A.; Piomelli, D. The fatty acid amide hydrolase inhibitor URB597 (cyclohexylcarbamic acid 3'-carbamoylbiphenyl-3-yl ester) reduces neuropathic pain after oral administration in mice. J. Pharm. Exp. Ther. 2007, 322, 236-242. 68. Powers, M. S.; Barrenha, G. D.; Mlinac, N. S.; Barker, E. L.; Chester, J. A. Effects of the novel endocannabinoid uptake inhibitor, LY2183240, on fear-potentiated startle and alcohol-seeking behaviors in mice selectively bred for high alcohol preference. Psychopharmacology (Berl) 2010, 212, 571-583. 69. Ahn, K.; Smith, S. E.; Liimatta, M. B.; Beidler, D.; Sadagopan, N.; Dudley, D. T.; Young, T.; Wren, P.; Zhang, Y.; Swaney, S.; Van Becelaere, K.; Blankman, J. L.; Nomura, D. K.; Bhattachar, S. N.; Stiff, C.; Nomanbhoy, T. K.; Weerapana, E.; Johnson, D. S.; Cravatt, B. F. Mechanistic and pharmacological characterization of PF-04457845: a highly potent and selective fatty acid amide hydrolase inhibitor that reduces inflammatory and noninflammatory pain. J. Pharm. Exp. Ther. 2011, 338, 114-124. 70. Li, G. L.; Winter, H.; Arends, R.; Jay, G. W.; Le, V.; Young, T.; Huggins, J. P. Assessment of the pharmacology and tolerability of PF-04457845, an irreversible inhibitor of fatty acid amide hydrolase-1, in healthy subjects. Br. J. Clin. Pharmacol. 2012, 73, 706-716. 71. Huggins, J. P.; Smart, T. S.; Langman, S.; Taylor, L.; Young, T. An efficient randomised, placebo-controlled clinical trial with the irreversible fatty acid amide hydrolase-1 inhibitor PF-04457845, which modulates endocannabinoids but fails to induce effective analgesia in patients with pain due to osteoarthritis of the knee. Pain 2012, 153, 1837-1846. 72. Schlosburg, J. E.; Boger, D. L.; Cravatt, B. F.; Lichtman, A. H. Endocannabinoid modulation of scratching response in an acute allergenic model: a new prospective neural therapeutic target for pruritus. J. Pharm. Exp. Ther. 2009, 329, 314-323. 73. Booker, L.; Kinsey, S. G.; Abdullah, R. A.; Blankman, J. L.; Long, J. Z.; Ezzili, C.; Boger, D. L.; Cravatt, B. F.; Lichtman, A. H. The fatty acid amide hydrolase (FAAH) inhibitor PF-3845 acts in the nervous system to reverse LPS-induced tactile allodynia in mice. Br. J. Pharmacol. 2012, 165, 2485-2496. 74. Kinsey, S. G.; Long, J. Z.; O'Neal, S. T.; Abdullah, R. A.; Poklis, J. L.; Boger, D. L.; Cravatt, B. F.; Lichtman, A. H. Blockade of endocannabinoid-degrading enzymes attenuates neuropathic pain. J. Pharm. Exp. Ther. 2009, 330, 902-910. 75. Sharma, P.; Eesa, M.; Scott, J. N. Toxic and acquired metabolic encephalopathies: MRI appearance. AJR. Am. J. Roentgenol. 2009, 193, 879-886. 76. Report by the Temporary Specialist Scientific Committee (TSSC), “FAAH (Fatty Acid Amide Hydrolase)”, on the causes of the accident during a Phase 1 clinical trial in Rennes in January 2016. 77. Hu, S. S.; Mackie, K. Distribution of the Endocannabinoid System in the Central Nervous System. Handb. Exp. Pharmacol. 2015, 231, 59-93.

229

78. Kiss, L. E.; Learmonth, D. A.; Rosa, C. P. C.; GUSMÃO, D. N. R.; Palma, P. N. L.; Manuel, D. S. P.; Soares, V. A.; Beliaev, A. Pharmaceutical compounds. 2010. 79. ANSM, Report by the Temporary Specialist Scientific Committee (TSSC), ‘‘FAAH (Fatty Acid Amide Hydrolase),’’ on the causes of the accident during a Phase 1 clinical trial in Rennes in January 2016. 2016. 80. Tong, J.; Mizrahi, R.; Houle, S.; Kish, S. J.; Boileau, I.; Nobrega, J.; Rusjan, P. M.; Wilson, A. A. Inhibition of fatty acid amide hydrolase by BIA 10-2474 in rat brain. J. Cereb. Blood Flow Metab. 2016, 0, 1-5. 81. Rotstein, B. H.; Wey, H. Y.; Shoup, T. M.; Wilson, A. A.; Liang, S. H.; Hooker, J. M.; Vasdev, N. PET Imaging of Fatty Acid Amide Hydrolase with [F-18]DOPP in Nonhuman Primates. Mol. Pharmaceut. 2014, 11, 3832-3838. 82. Piomelli, D.; Tarzia, G.; Duranti, A.; Tontini, A.; Mor, M.; Compton, T. R.; Dasse, O.; Monaghan, E. P.; Parrott, J. A.; Putman, D. Pharmacological profile of the selective FAAH inhibitor KDS-4103 (URB597). CNS Drug Rev. 2006, 12, 21-38. 83. Wilson, A. A.; Garcia, A.; Parkes, J.; Houle, S.; Tong, J.; Vasdev, N. [11C]CURB: Evaluation of a novel radiotracer for imaging fatty acid amide hydrolase by positron emission tomography. Nucl. Med. Biol. 2011, 38, 247-253. 84. Krieger, E.; Vriend, G. YASARA View-molecular graphics for all devices-from smartphones to workstations. Bioinformatics 2014, 30, 2981-2982. 85. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart, W. E.; Belew, R. K.; Olson, A. J. Automated docking using a Lamarckian genetic algorithm and an empirical binding free energy function. J. Comput. Chem. 1998, 19, 1639-1662. 86. Otrubova, K.; Brown, M.; McCormick, M. S.; Han, G. W.; O'Neal, S. T.; Cravatt, B. F.; Stevens, R. C.; Lichtman, A. H.; Boger, D. L. Rational design of fatty acid amide hydrolase inhibitors that act by covalently bonding to two active site residues. J. Am. Chem. Soc.2013, 135, 6289-99. 87. Kono, M.; Matsumoto, T.; Kawamura, T.; Nishimura, A.; Kiyota, Y.; Oki, H.; Miyazaki, J.; Igaki, S.; Behnke, C. A.; Shimojo, M.; Kori, M. Synthesis, SAR study, and biological evaluation of a series of piperazine ureas as fatty acid amide hydrolase (FAAH) inhibitors. Bioorg. Med. Chem. 2013, 21, 28-41. 88. Bertolacci, L.; Romeo, E.; Veronesi, M.; Magotti, P.; Albani, C.; Dionisi, M.; Lambruschini, C.; Scarpelli, R.; Cavalli, A.; De Vivo, M.; Piomelli, D.; Garau, G. A binding site for nonsteroidal anti- inflammatory drugs in fatty acid amide hydrolase. J. Am. Chem. Soc.2013, 135, 22-25. 89. Min, X.; Thibault, S. T.; Porter, A. C.; Gustin, D. J.; Carlson, T. J.; Xu, H.; Lindstrom, M.; Xu, G.; Uyeda, C.; Ma, Z.; Li, Y.; Kayser, F.; Walker, N. P.; Wang, Z. Discovery and molecular basis of potent noncovalent inhibitors of fatty acid amide hydrolase (FAAH). Proc. Natl. Acad. Sci. U.S.A. 2011, 108, 7379-7384. 90. Gustin, D. J.; Ma, Z.; Min, X.; Li, Y.; Hedberg, C.; Guimaraes, C.; Porter, A. C.; Lindstrom, M.; Lester-Zeiner, D.; Xu, G.; Carlson, T. J.; Xiao, S.; Meleza, C.; Connors, R.; Wang, Z.; Kayser, F. Identification of potent, noncovalent fatty acid amide hydrolase (FAAH) inhibitors. Bioorg. Med. Chem. Lett. 2011, 21, 2492-2496.

230

91. Mileni, M.; Garfunkle, J.; Ezzili, C.; Kimball, F. S.; Cravatt, B. F.; Stevens, R. C.; Boger, D. L. X-ray crystallographic analysis of alpha-ketoheterocycle inhibitors bound to a humanized variant of fatty acid amide hydrolase. J. Med. Chem. 2010, 53, 230-240. 92. Ahn, K.; Johnson, D. S.; Mileni, M.; Beidler, D.; Long, J. Z.; McKinney, M. K.; Weerapana, E.; Sadagopan, N.; Liimatta, M.; Smith, S. E.; Lazerwith, S.; Stiff, C.; Kamtekar, S.; Bhattacharya, K.; Zhang, Y.; Swaney, S.; Van Becelaere, K.; Stevens, R. C.; Cravatt, B. F. Discovery and characterization of a highly selective FAAH inhibitor that reduces inflammatory pain. Chem. Biol. 2009, 16, 411-420. 93. Mileni, M.; Garfunkle, J.; Ezzili, C.; Cravatt, B. F.; Stevens, R. C.; Boger, D. L. Fluoride- mediated capture of a noncovalent bound state of a reversible covalent enzyme inhibitor: X-ray crystallographic analysis of an exceptionally potent alpha-ketoheterocycle inhibitor of fatty acid amide hydrolase. J. Am. Chem. Soc.2011, 133, 4092-4100. 94. Karlsson, M.; Contreras, J. A.; Hellman, U.; Tornqvist, H.; Holm, C. cDNA cloning, tissue distribution, and identification of the catalytic triad of monoglyceride lipase. Evolutionary relationship to esterases, lysophospholipases, and haloperoxidases. J. Biol. Chem. 1997, 272, 27218-27223. 95. King, A. R.; Lodola, A.; Carmi, C.; Fu, J.; Mor, M.; Piomelli, D. A critical cysteine residue in monoacylglycerol lipase is targeted by a new class of isothiazolinone-based enzyme inhibitors. Br. J. Pharmacol. 2009, 157, 974-983. 96. Bertrand, T.; Auge, F.; Houtmann, J.; Rak, A.; Vallee, F.; Mikol, V.; Berne, P. F.; Michot, N.; Cheuret, D.; Hoornaert, C.; Mathieu, M. Structural basis for human monoglyceride lipase inhibition. J. Mol. Biol. 2010, 396, 663-673. 97. Labar, G.; Bauvois, C.; Borel, F.; Ferrer, J. L.; Wouters, J.; Lambert, D. M. Crystal structure of the human monoacylglycerol lipase, a key actor in endocannabinoid signaling. Chembiochem. 2010, 11, 218-227. 98. Griebel, G.; Pichat, P.; Beeske, S.; Leroy, T.; Redon, N.; Jacquet, A.; Francon, D.; Bert, L.; Even, L.; Lopez-Grancha, M.; Tolstykh, T.; Sun, F.; Yu, Q.; Brittain, S.; Arlt, H.; He, T.; Zhang, B.; Wiederschain, D.; Bertrand, T.; Houtmann, J.; Rak, A.; Vallee, F.; Michot, N.; Auge, F.; Menet, V.; Bergis, O. E.; George, P.; Avenet, P.; Mikol, V.; Didier, M.; Escoubet, J. Selective blockade of the hydrolysis of the endocannabinoid 2-arachidonoylglycerol impairs learning and memory performance while producing antinociceptive activity in rodents. Sci. Rep. 2015, 5, 7642-7657. 99. Schalk-Hihi, C.; Schubert, C.; Alexander, R.; Bayoumy, S.; Clemente, J. C.; Deckman, I.; DesJarlais, R. L.; Dzordzorme, K. C.; Flores, C. M.; Grasberger, B.; Kranz, J. K.; Lewandowski, F.; Liu, L.; Ma, H.; Maguire, D.; Macielag, M. J.; McDonnell, M. E.; Mezzasalma Haarlander, T.; Miller, R.; Milligan, C.; Reynolds, C.; Kuo, L. C. Crystal structure of a soluble form of human monoglyceride lipase in complex with an inhibitor at 1.35 A resolution. Protein Sci. 2011, 20, 670-683. 100. Hohmann, A. G.; Suplita, R. L. Endocannabinoid mechanisms of pain modulation. AAPS J. 2006, 8, E693-E708. 101. Xu, J. Y.; Chen, C. Endocannabinoids in synaptic plasticity and neuroprotection. Neuroscientist 2015, 21, 152-168. 102. Shonesy, B. C.; Bluett, R. J.; Ramikie, T. S.; Baldi, R.; Hermanson, D. J.; Kingsley, P. J.; Marnett, L. J.; Winder, D. G.; Colbran, R. J.; Patel, S. Genetic Disruption of 2-Arachidonoylglycerol Synthesis Reveals a Key Role for Endocannabinoid Signaling in Anxiety Modulation. Cell Rep. 2014, 9, 1644-1653.

231

103. Matias, I.; Di Marzo, V. Endocannabinoids and the control of energy balance. Trends Endocrin. Met. 2007, 18, 27-37. 104. Pagotto, U.; Marsicano, G.; Cota, D.; Lutz, B.; Pasquali, R. The emerging role of the endocannabinoid system in endocrine regulation and energy balance. Endocr. Rev. 2006, 27, 73-100. 105. Nomura, D. K.; Long, J. Z.; Niessen, S.; Hoover, H. S.; Ng, S. W.; Cravatt, B. F. Monoacylglycerol Lipase Regulates a Fatty Acid Network that Promotes Cancer Pathogenesis. Cell 2010, 140, 49-61. 106. Nomura, D. K.; Morrison, B. E.; Blankman, J. L.; Long, J. Z.; Kinsey, S. G.; Marcondes, M. C. G.; Ward, A. M.; Hahn, Y. K.; Lichtman, A. H.; Conti, B.; Cravatt, B. F. Endocannabinoid Hydrolysis Generates Brain Prostaglandins That Promote Neuroinflammation. Science 2011, 334, 809-813. 107. Nomura, D. K.; Lombardi, D. P.; Chang, J. W.; Niessen, S.; Ward, A. M.; Long, J. Z.; Hoover, H. H.; Cravatt, B. F. Monoacylglycerol Lipase Exerts Dual Control over Endocannabinoid and Fatty Acid Pathways to Support Prostate Cancer. Chem. Biol. 2011, 18, 846-856. 108. Hohmann, A. G.; Suplita, R. L.; Bolton, N. M.; Neely, M. H.; Fegley, D.; Mangieri, R.; Krey, J. F.; Walker, J. M.; Holmes, P. V.; Crystal, J. D.; Duranti, A.; Tontini, A.; Mor, M.; Tarzia, G.; Piomelli, D. An endocannabinoid mechanism for stress-induced analgesia. Nature 2005, 435, 1108-1112. 109. Comelli, F.; Giagnoni, G.; Bettoni, I.; Colleoni, M.; Costa, B. The inhibition of monoacylglycerol lipase by URB602 showed an anti-inflammatory and anti-nociceptive effect in a murine model of acute inflammation. Br. J. Pharmacol. 2007, 152, 787-794. 110. Desroches, J.; Guindon, J.; Lambert, C.; Beaulieu, P. Modulation of the anti-nociceptive effects of 2-arachidonoyl glycerol by peripherally administered FAAH and MGL inhibitors in a neuropathic pain model. Br. J. Pharmacol. 2008, 155, 913-24. 111. Long, J. Z.; Nomura, D. K.; Cravatt, B. F. Characterization of Monoacylglycerol Lipase Inhibition Reveals Differences in Central and Peripheral Endocannabinoid Metabolism. Chem. Biol. 2009, 16, 744-753. 112. Long, J. Z.; Li, W. W.; Booker, L.; Burston, J. J.; Kinsey, S. G.; Schlosburg, J. E.; Pavon, F. J.; Serrano, A. M.; Selley, D. E.; Parsons, L. H.; Lichtman, A. H.; Cravatt, B. F. Selective blockade of 2- arachidonoylglycerol hydrolysis produces cannabinoid behavioral effects. Nat. Chem. Biol. 2009, 5, 37- 44. 113. Long, J. Z.; Nomura, D. K.; Vann, R. E.; Walentiny, D. M.; Booker, L.; Jin, X.; Burston, J. J.; Sim-Selley, L. J.; Lichtman, A. H.; Wiley, J. L.; Cravatt, B. F. Dual blockade of FAAH and MAGL identifies behavioral processes regulated by endocannabinoid crosstalk in vivo. Proc. Natl. Acad. Sci. U.S.A. 2009, 106, 20270-20275. 114. Chang, J. W.; Niphakis, M. J.; Lum, K. M.; Cognetta, A. B.; Wang, C.; Matthews, M. L.; Niessen, S.; Buczynski, M. W.; Parsons, L. H.; Cravatt, B. F. Highly Selective Inhibitors of Monoacylglycerol Lipase Bearing a Reactive Group that Is Bioisosteric with Endocannabinoid Substrates. Chem. Biol. 2012, 19, 579-588. 115. Chang, J. W.; Cognetta, A. B.; Niphakis, M. J.; Cravatt, B. F. Proteome-Wide Reactivity Profiling Identifies Diverse Carbamate Chemotypes Tuned for Serine Hydrolase Inhibition. ACS Chem Biol 2013, 8, 1590-1599.

232

116. Tuccinardi, T.; Granchi, C.; Rizzolio, F.; Caligiuri, I.; Battistello, V.; Toffoli, G.; Minutolo, F.; Macchia, M.; Martinelli, A. Identification and characterization of a new reversible MAGL inhibitor. Bioorg. Med. Chem. 2014, 22, 3285-3291. 117. Matuszak, N.; Muccioli, G. G.; Labar, G.; Lambert, D. M. Synthesis and in Vitro Evaluation of N-Substituted Maleimide Derivatives as Selective Monoglyceride Lipase Inhibitors. J. Med. Chem. 2009, 52, 7410-7420. 118. Laitinen, T.; Navia-Paldanius, D.; Rytilahti, R.; Marjamaa, J. J.; Karizkova, J.; Parkkari, T.; Pantsar, T.; Poso, A.; Laitinen, J. T.; Savinainen, J. R. Mutation of Cys242 of human monoacylglycerol lipase disrupts balanced hydrolysis of 1- and 2-monoacylglycerols and selectively impairs inhibitor potency. Mol. Pharmacol.2014, 85, 510-519. 119. King, A. R.; Lodola, A.; Carmi, C.; Fu, J.; Mor, M.; Piomelli, D. A critical cysteine residue in monoacylglycerol lipase is targeted by a new class of isothiazolinone-based enzyme inhibitors. Br. J. Pharmacol. 2009, 157, 974-983. 120. Schlosburg, J. E.; Blankman, J. L.; Long, J. Z.; Nomura, D. K.; Pan, B.; Kinsey, S. G.; Nguyen, P. T.; Ramesh, D.; Booker, L.; Burston, J. J.; Thomas, E. A.; Selley, D. E.; Sim-Selley, L. J.; Liu, Q. S.; Lichtman, A. H.; Cravatt, B. F. Chronic monoacylglycerol lipase blockade causes functional antagonism of the endocannabinoid system. Nat. Neurosci. 2010, 13, 1113-1119. 121. Hicks, J. W.; Parkes, J.; Tong, J. C.; Houle, S.; Vasdev, N.; Wilson, A. A. Radiosynthesis and ex vivo evaluation of [C-11-carbonyl]carbamate- and urea-based monoacylglycerol lipase inhibitors. Nucl. Med. Biol. 2014, 41, 688-694. 122. Wang, L.; Mori, W.; Cheng, R.; Yui, J. J.; Hatori, A.; Ma, L. L.; Zhang, Y. D.; Rotstein, B. H.; Fujinaga, M.; Shimoda, Y.; Yamasaki, T.; Xie, L.; Nagai, Y.; Minamimoto, T.; Higuchi, M.; Vasdev, N.; Zhang, M. R.; Liang, S. H. Synthesis and Preclinical Evaluation of Sulfonamido-based [C-11-Carbonyl]- Carbamates and Ureas for Imaging Monoacylglycerol Lipase. Theranostics 2016, 6, 1145-1159. 123. Wang, C.; Placzek, M. S.; Van de Bittner, G. C.; Schroeder, F. A.; Hooker, J. M. A Novel Radiotracer for Imaging Monoacylglycerol Lipase in the Brain Using Positron Emission Tomography. ACS Chem. Neurosci. 2016, 7, 484-9. 124. Friesner, R. A.; Murphy, R. B.; Repasky, M. P.; Frye, L. L.; Greenwood, J. R.; Halgren, T. A.; Sanschagrin, P. C.; Mainz, D. T. Extra precision glide: docking and scoring incorporating a model of hydrophobic enclosure for protein-ligand complexes. J. Med. Chem. 2006, 49, 6177-96. 125. Zhu, K.; Bonelli, K. W.; Greenwood, J. R.; Day, T.; Abel, R.; Farid, R. S.; Harder, E. Docking Covalent Inhibitors: A Parameter Free Approach To Pose Prediction and Scoring. J. Chem. Inf. Model 2014, 54, 1932-1940. 126. Aaltonen, N.; Savinainen, J. R.; Ribas, C. R.; Ronkko, J.; Kuusisto, A.; Korhonen, J.; Navia- Paldanius, D.; Hayrinen, J.; Takabe, P.; Kasnanen, H.; Pantsar, T.; Laitinen, T.; Lehtonen, M.; Pasonen- Seppanen, S.; Poso, A.; Nevalainen, T.; Laitinen, J. T. Piperazine and Piperidine Triazole Ureas as Ultrapotent and Highly Selective Inhibitors of Monoacylglycerol Lipase. Chem. Biol. 2013, 20, 379-390. 127. Brindisi, M.; Maramai, S.; Gemma, S.; Brogi, S.; Grillo, A.; Mannelli, L. D.; Gabellieri, E.; Lamponi, S.; Saponara, S.; Gorelli, B.; Tedesco, D.; Bonfiglio, T.; Landry, C.; Jung, K. M.; Armirotti, A.; Luongo, L.; Ligresti, A.; Piscitelli, F.; Bertucci, C.; Dehouck, M. P.; Campiani, G.; Malone, S.; Ghelardini, C.; Pittaluga, A.; Piomelli, D.; Di Marzo, V.; Butini, S. Development and Pharmacological

233

Characterization of Selective Blockers of 2-Arachidonoyl Glycerol Degradation with Efficacy in Rodent Models of Multiple Sclerosis and Pain. J. Med. Chem.2016, 59, 2612-2632. 128. Bian, H.; Chevalier, K.; Clemente, J.; Connolly, P. J.; Flores, C.; Lin, S.-C.; Liu, L.; Mabus, J.; Macielag, M.; McDonnell, M. M.; Pitis, P. M.; Zhang, Y.-M.; Zhang, S.-P.; Zhu, B. Azetidinyl diamides as monoacylglycerol lipase inhibitors, US2010032049, 2014. 129. Halgren, T. A.; Murphy, R. B.; Friesner, R. A.; Beard, H. S.; Frye, L. L.; Pollard, W. T.; Banks, J. L. Glide: A new approach for rapid, accurate docking and scoring. 2. Enrichment factors in database screening. J. Med. Chem. 2004, 47, 1750-1759. 130. Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.; Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.; Perry, J. K.; Shaw, D. E.; Francis, P.; Shenkin, P. S. Glide: A new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy. J. Med. Chem. 2004, 47, 1739-1749. 131. Shivakumar, D.; Williams, J.; Wu, Y.; Damm, W.; Shelley, J.; Sherman, W. Prediction of Absolute Solvation Free Energies using Molecular Dynamics Free Energy Perturbation and the OPLS Force Field. J. Chem. Theory Comput. 2010, 6, 1509-1519. 132. Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.; Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.; Perry, J. K.; Shaw, D. E.; Francis, P.; Shenkin, P. S. Glide: a new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy. J. Med. Chem. 2004, 47, 1739-1749. 133. http://yasara.org/autosmiles.htm.

234

Chapter 5: Fluorinated adenosine A2A receptor (A2AR) antagonists as potential cancer immunotherapeutics

5.1 Background

The adenosine A2A receptor (A2AR) is a member of purinergic G protein-coupled receptors

(GPCRs). Other subtypes of the adenosine receptors include adenosine A1 receptor (A1R), adenosine A2B receptor (A2BR), and adenosine A3 receptor (A3R) and are differentiated based on

1-2 their signal transduction pathways. A1Rs and A3Rs are activated by Gi proteins to inhibit the production of adenylate cyclase (AC). Activation of A2ARs and A2BRs stimulate AC formation, thus regulating the level of intracellular cyclic adenosine monophosphate (cAMP), which is a

3-5 second messenger known to trigger a complex sequence of cellular events. A2ARs are highly enriched in the striatum and the shell regions of nucleus accumbens, but are less distributed in the olfactory tubercle, hypothalamus, hippocampus and other parts in the brain.6-7 In the

periphery system, A2ARs are found in platelets, immune tissues, neutrophils, endothelium and

6-7 vascular smooth muscle. The implications of A2ARs in a broad spectrum of diseases have been well established, such as neurodegenerative disorders (e.g. Parkinson’s disease (PD)), cardiac ischemia, sickle cell disease, diabetic nephropathy, inflammation and cancer.1-2, 5, 8-9

5.1.1 Development of A2AR agonists

10 Development of A2ARs agonists started from the natural ligand of adenosine (Figure 5-1).

Previous work has shown that introduction of a chlorine at the 2-position (2-Cl-Ado, hA2A Ki =

180 nM) or substitution with the ethylamide group at the 5’-position of adenosine (NECA, hA2A

235

Ki = 20 nM) improved the A2AR binding potency, but these modifications did not sufficiently bias their binding toward the other three subtypes.11 The 2-hexynyl substituted NECA (i.e.

HENECA) exhibited high potency (hA2A Ki = 6.4 nM) and a 10-fold selectivity over

12 recombinant hA1R (Ki = 60 nM), and further modifications led to the 2-adenine substituted

13 analogue CGS21680 (CGS) as a potent A2AR agonist (hA2A Ki = 27 nM). It showed moderate selectivity over A3R (hA3 Ki = 67 nM), but it had 10-fold selectivity over A1R (hA2A Ki = 290

13 nM) as well as high selectivity toward A2BR (hA2B EC50 = 361 µM). However, modifications toward the 2’- and 3’-hydroxy groups of ribose14-15 and C-8 and N-6 positions of purine16-17 either led to a decrease or complete loss of the A2AR potency.

236

Figure 5-1. Typical adenosine A2AR agonists.

A2AR agonists are mainly used as vasodilating agents in myocardial perfusion imaging

(MPI), for the treatments of inflammatory diseases and neuropathic pain.17 Adenosine is a pharmacological stress agent used to induce coronary arterial vasodilation through its activation

18 of A2ARs on arteriolar vascular smooth muscle cells. However, its nonselective nature could cause several side effects, such as bronchospasms, dyspnea and high-grade atrioventricular block, though these side effects are minimized due to the extremely short half-life of adenosine brought by its sugar moiety.19 Efforts in improving the selectivity of adenosine for MPI led to the promising agents of binodenoson, apadenoson and regadenoson (Figure 5-2).

237

Figure 5-2. Structures of A2AR agonists as therapeutics.

Binodenoson is conformationally constrained by a hydrozone double bond at the C-2

position. It has moderate A2AR binding affinity but good selectivity toward adenosine subtypes

20-21 (hA2A Ki = 270 nM, hA2B EC50 > 100 μM, hA1 Ki = 48 μM, and hA3 Ki = 903 nM).

Apadenoson employs a C-2 substitution at adenine with the propynylcyclohexanemethyl ester

group and displays a high A2AR binding affinity (hA2A Ki = 0.5 nM) and good selectivity (150- fold and 90-fold selectivity toward A1R and A3R respectively). It entered phase III clinical trials in 2009 for MPI, but was discontinued in 2012 without clinical data disclosed.22 Regadenoson is

a derivative with a constrained N-pyrazole moiety at the C-2 position. It has moderate A2A

238 binding affinity (hA2A Ki = 290 nM) with 30-fold selectivity toward A2BR and A3R and 13-fold selectivity versus A1R. Regadenoson was approved by the FDA in 2008 for MPI and marketed under the name of Lexiscan.23-24

Activation of the A2AR can also regulate the activity of various inflammatory cells, such as neutrophils, macrophages, lymphocytes, fibroblasts, monocytes etc., thus, adjusting the innate and adaptive immune responses and participating in terminating inflammations.25 In this aspect, apadenoson was shown to reduce joint destruction brought by septic arthrosis, and CGS21680 was found to regulate the immune response caused by the HIV-1 transactivating regulatory protein.26 Agonist GW328267, developed by GlaxoSmithKline, was used for the treatment of

27 allergic rhinitis and asthma (Figure 5-2). In addition, activation of A2AR was shown to relieve neuropathic pain by suppressing the glial proinflammatory cytokine release, an important contributor to nuropathic pain, and simultaneously increasing the anti-inflammatory cytokine interleukin 10 (IL-10) level.

5.1.2 Development of A2AR antagonists

In general, the A2AR antagonists possess a mono-, bi-, or tricyclic core structure that is similar to the adenine part of adenosine, but lack the conserved sugar moiety among A2AR agonists. They are classified into xanthines and non-xanthines.17, 28-29 Development of potent and

selective A2AR antagonists commenced from the naturally occurring nonselective xanthine A2AR antagonists, caffeine and theophylline, which have only micromolar range binding affinities

30 toward A2AR (Figure 5-3). 239

Figure 5-3. Structures of Caffeine, Theophylline, DMPX, CSC and DMPTX.

Chemical modifications also aimed to improve the poor aqueous solubility of xanthines.31

The propargyl substitution at the 1-position of caffeine (DMPX) improved its aqueous solubility and bioavailability;31 but simultaneous improvements of the binding potency and selectivity required the 8-trans-styryl substitution, such as in CSC and DMPTX.32-33

To improve the aqueous solubility of the xanthine-like antagonists, bioisosteric replacements with polar functional groups were also employed. Introduction of highly polar / hydrophilic

substituents, such as sulfonic or carboxylic acid groups, however, resulted in reduced A2AR binding affinity. For example, p-SS-DMPX, a sulfonate salt, showed 5-30-fold decrease in A2AR binding potency compared with other 8-trans-styryl substituted DMPX analogs according to

240

Muller et al. (Figure 5-4).34 On the other hand, the MSX series of prodrugs developed by Muller et al. turned out to be more successful (Figure 5-4).35-38 The hydroxyl group of the potent and selective A2AR antagonist MSX-2 was found to be a viable locus for the attachment of a prodrug moiety. The disodium phosphate salt prodrug MSX-3 had aqueous solubility of 9 mg/mL (17

mM) and the L-valine chimera prodrug MSX-4 exhibited aqueous solubility of 7.3 mg/mL (13.8 mM). These prodrugs were stable in solution but could be rapidly degraded in vivo by enzymes

(i.e. phosphatases for MSX-3 and esterases for MSX-4) to release the active compound.35-38

MSX-4 also showed positive results in motivational and motor function-related animal models, indicating its potential in treating Parkinson’s disease (PD) and other motor disorders.39

Figure 5-4. Structures of p-SS-DMPX, MSX-2, MSX-3, and MSX-4.

Further modifications on the 8-trans-styryl substituted xanthines led to the discovery of KF-

17837 by Suzuki et al. (Figure 5-5).40 Replacement of the 1,3-dipropyl moieties on KF-17837 with ethyl groups yielded istradefylline (KW6002), which was approved in Japan in 2013 under the name of Nouriast® for the treatment of PD.41-43

241

Figure 5-5. Structures of KF-17837 and KW-6002.

Though the development of xanthine-like A2AR antagonists has led to the success of

KW6002, many issues still need to be addressed for this class of compounds, such as inherent poor aqueous solubility and photo-instability of the 8-trans-styryl double bond (Scheme 5-1).44

When exposed to the daylight or irradiated by UV light, the 8-trans-styrylxanthine double bond of KW6002 dimerizes to form the cyclobutane side product, resulting in significantly reduced

binding affinity against A2AR and poor selectivity toward A1R. The photo-induced (E)/(Z)- isomerization of the styryl double bond of KW6002 also occurred rapidly in daylight to form the less potent and nonselective Z-isomer with an unfavorable ratio (E)/(Z) = 19:81. This was also observed for the antagonists CSC and KF-17837.44 Attempts to introduce conformationally restrained motifs (e.g. propargyl-, cyclopropyl- or diazo-group) to replace the unstable styryl

double bond, however, resulted in the loss of the binding affinity against A2AR and selectivity

45 over A1R.

242

Scheme 5-1. Photo- dimerization and isomerization of KW6002.

The above-mentioned limitations of the xanthine derivatives initiated the exploration of a

different structural class of compounds, non-xanthine A2AR antagonists based on mono-, bi-, and tri-heterocycles. The first non-xanthine A2AR antagonist, CGS15943 featured a tri-cyclic triazoloquinazoline core structure, which was reported by Williams et al. in 1987 (Figure 5-6).46

It showed good binding affinity against A2AR, but its high potency toward A1R led to undesired cardiovascular effects and disruption of the activity of anti-PD agents. Further optimization of this class of compounds led to the discovery of preladenant, which showed positive results in

Phase II clinical trials for PD,47 but failed to prove its efficacy during Phase III trials, and so was

243 discontinued in 2013. In general, this series of compounds suffered from poor aqueous solubility and were difficult to synthesize due to the presence of the fused tri-cyclic heterocycles.

Figure 5-6. Structures of A2AR antagonists CGS15943 and Preladenant.

To circumvent the above-mentioned problems, Zeneca Co. reported the first potent bi-

cyclic-non-xanthine A2AR antagonist ZM241385 (ZM, hA2A Ki = 0.8 nM, Figure 5-7) with

48 favorable aqueous solubility, but poor selectivity over A2BR and limited brain penetration. In addition, a series of novel bi-cyclic heterocycles bearing benzoxazole or thiazolopyridine core scaffolds were also developed, among which tozadenant (SYN115) entered the Phase III clinical trials in 2015 for the treatment of PD.49

244

Figure 5-7. Structures of ZM241385 and Tozadenant.

Monocyclic non-xanthine A2AR antagonists have also been explored to improve the aqueous solubility and simplify the synthetic accessibility. Typical compounds in this category include the 1,2,4-triazine derivatives developed by Heptares Therapeutics Co. (Figure 5-8),50-51 and the tri-substituted pyrimidine derivative reported by Neurocrine Biosciences52 and Yang et al.53

However, none of these compounds have been evaluated in clinical trials yet.

Figure 5-8. Examples of monocyclic 1,2,4-triazine and tri-substituted pyrimidine as A2AR antagonists.

245

5.2 Structure and binding pocket of A2AR

As a transmembrane protein, A2AR is very difficult to crystalize due to its high flexibility and poor solubility when isolated from the cell membrane environment. In 2008, the first crystal

structure of human A2AR in complex with antagonist ZM241385 was deciphered (PDB ID:

3EML), providing more insights into the structure of A2AR and the key binding events in the

54 active site. The T4 lysozyme (T4L) was grafted onto the A2AR to stabilize its conformation

55 during crystallization. As shown in Figure 5-9, A2AR shares similar structural features to other

GPCRs: it incorporates extracellular N-terminus and intracellular C-terminals connected via seven transmembrane α-helices (TM1-7), each of which is linked by three intracellular loops

(IL1-3) and three extracellular loops (EL1-3). Furthermore, a short transmembrane α-helix

(TM8) runs parallel to the cytoplasmic surface of the membrane and eight cysteine residues form

four disulfide bridges across the receptor. A unique feature of A2AR is that its ligand binding pocket is nearly perpendicular to the membrane plane.

246

Figure 5-9. Crystal structure of the human A2AR in complex with ZM241385.

Reproduced with permission from Elsevier.55

28 The key binding interaction of ZM241385 in A2AR is depicted in Figure 5-10. The bicyclic triazolotriazine core of ZM241385 forms a face-to-face π-π stacking interaction with Phe168 as well as aliphatic hydrophobic interactions with residue Ile274. The exocyclic amino group of

ZM241385 serves as hydrogen bond donors in the interactions with both Glu169 and Asn253.

One of the triazole nitrogen atoms and the furan oxygen atom form hydrogen bonds with Asn253

247 as hydrogen bond acceptors. The phenolic hydroxyl group forms a hydrogen bond with an ordered water molecule and the electrons of the phenyl group are overlapped with Leu267 and

Met270 to form π-π stacking interactions. The phenylethylamine group directs into the

extracelluar region (EL2 and EL3) of A2AR. The furan ring sits deeply at the bottom of the ligand-binding cavity, where it forms hydrophobic interactions with His250 and Leu249.

Figure 5-10. Key binding interactions of ZM241385 in A2AR.

Reproduced from reference 28, copyright of Taylor & Francis.28

Picture is adopted from Yuan, G. et al. Curr. Med. Chem. 2014, 21(34), 3918. EL2 is reported as an important ‘cap’ at the top of the binding cavity during receptor activation and deactivation. However, this important substructure is missing in the crystal structure 3EML due to its high flexibility and low experimental electron density in this region.

Direct docking experiments in the absence of EL2 led to a docking result of ZM241385 that is inconsistent with its original docking pattern in the crystal structure. To facilitate the molecular modeling assisted drug discovery of novel A2AR antagonists, the missing EL2 in the crystal structure (Figure 5-11, blue) was fixed via YASARA by Dr. Joslynn Lee in Ondrechen’s group

55 at Northeastern to give a complete homology model of A2AR (Figure 5-11, purple). The

248 resulting homology model was successfully applied to the design of novel A2AR antagonists based on KW6002.55

Figure 5-11. Homology model of A2AR based on the crystal structure 3EML.

Reproduced from Dr. Joslynn Lee’s thesis at Northeastern.

5.3 A2AR antagonists as cancer immunotherapeutics

Cancer immunotherapy is complementary to traditional chemotherapy, radiotherapy and surgery. It aims to stimulate/improve the patients’ immune system to reject the malignant tumor cells that are responsible for the cancerous disease. One strategy is to adoptively transfer the activated T cells or vaccines to the patients.56 For example, PROVENGE® (Sipuleucel-T) is the first FDA approved vaccine for the treatment of prostate cancer, which is furnished by culturing 249 a patient's own immune cells with a prostate specific recombinant antigen. Alternatively, the immune response is endogenously improved or sustained by blocking the inhibitory mechanisms.57 For this purpose, the monoclonal anti-body Yervoy® (Ipilimumab) was approved by the FDA to treat melanoma. Yervoy® activates the inhibitory cytotoxic T lymphocyte by blocking the negative immune checkpoint CTLA-4. Although these two drugs suffer short life extension issues, their approvals are encouraging, emphasizing the importance of understanding the tumor protection mechanisms.

Antagonism of A2ARs on the immune cells in the hypoxia tumor microenvironment represents a potential and novel pathway for cancer immunotherapy as reported by Sitkovsky et al.58-67 The disclosed hypoxia-adenosinergic pathway was found responsible for tumor protection, wherein hypoxia-driven accumulation of extracellular adenosine induced immune

suppression via A2AR activation on the surface of immune cells. Under normal conditions, lymphocytes, as part of the cell-mediated immune response, recognize and kill foreign antigens and infected cells. During this process, adenosine and adenosine receptors on T cells, especially

A2AR and A2BR, are involved in the regulations of the immune response (Figure 5-12). The hypoxia microenvironment and accumulation of extracellular adenosine caused by inflammation increase the level of intracellular cAMP, which would trigger downstream intracellular signaling pathways to suppress immune response, such as the reduction of proinflammatory cytokines tumor-necrosis factor alpha (TNF-α), interferon gamma (IFN-γ) and interleukin-2 (IL-2).68

250

Figure 5-12. Hypoxia-A2AR–mediated mechanism of tissue protection.

Reproduced with permission from Elsevier.68

However, this immune suppressive mechanism is misused by tumor cells to protect themselves from T cells due to the easily formed hypoxia microenvironment in tumor-infected tissues. In advanced tumors, the formation of hypoxia tissue areas is quite common and it occurs in various tumor affected tissue, such as brain tumors, soft tissue sarcomas, malignant melanomas, metastatic liver tumors, etc.69 This hypoxia microenvironment is caused by the angiogenesis, structural and functional abnormalities of vasculature, limited blood flow, increased diffusion distances due to tumor growth, and low levels of hemoglobin in the blood.70-

The hypoxia-adenosinergic pathway was further validated by genetic and pharmacological interventions. In the Concanavalin A (ConA) mouse model, a model used for the liver

inflammation study, administration of potent and selective A2AR agonist CGS241385 prevented

251 liver damage from uncontrolled inflammation and decreased the accumulation of cytokine TNF-

α. Whereas, the A2AR gene knock-out (A2AR-/-) mice and those without the administration of

CGS2160 showed severe liver damage and death of two out of four mice, illustrating the

62 protective role of A2AR agonists. On the other hand, blockage of tumor protection via the hypoxia-adenosinergic pathway and the resulting rejection to tumor cells were studied using

CL8-1 malignant melanoma and RMA T lymphoma tumor models in C57BL/6 mice.66 60%

A2AR-/- mice rejected the tumor cells and survived, whereas, the wide-type mice died after 20 days (Figure 5-13A).63 Administration of caffeine and ZM241385 were also seen to restore the production of proinflammatory cytokine IFN-γ, an indication of the rescue of immune response

55, 63 by A2AR antagonists (Figure 5-13B). In addition, antagonism of A2ARs on the endogenously developed antitumor CD8+ cells via caffeine and ZM241385 delayed the growth of CL8-1 melanoma, however, the rejection of tumors was not observed due to the short half-life of such

59-61, 64-66 antagonists (ZM241385, t1/2 ~30 min).

252

Figure 5-13. Validations of the hypoxia-adenosinergic pathway.

(A) (B)

Reproduced from reference 63, copyright of National Academy of Sciences.

253

Unlike the criteria of A2AR antagonists as Figure 5-14. Structure of KW-PEG.

anti-PD drugs, development of novel A2AR

antagonists as potential cancer

immunotherapeutics emphasizes the

requirements of diminished brain-blood barrier

(BBB) penetrations to target the peripheral immune cells, improved aqueous solubility (a

common issue of the current A2AR antagonists), prolonged half-life or stability and synthetic accessibility. Previous modifications based on KW6002 showed the introduction of polyethylene glycol and its derivatives suitable for the above considerations. The lead compound KW-PEG

(Figure 5-14),55 an octaethyleneglycol monomethyl ether derivative of KW6002, demonstrated enhanced chemical stability (no E/Z isomerization after 2 weeks in solid state), 2-fold increase in hydrophilicity based on logP calculation, and parallel results in two immune-assays as KW6002

(i.e. cAMP suppression and cytokine IFN-gamma restoration assays).

Spurred on the success of KW-PEG and the lack of efficient cancer immunotherapeutics

derived from A2AR antagonists, there is an urgent need to design enhanced derivatives based on other classes of A2AR antagonists for immunotherapy application. Another consideration is to install functionality that allows the diagnostic imaging with PET imaging tracers – specifically

18 introduction of F-isotope (t1/2 = 109.8 min).

254

5.4 Design of novel A2AR antagonists as potential cancer immunotherapeutics

To select lead compounds, an in vivo immunoassay was performed using the ConA mouse model, where KW6002, KW-PEG and preladenant were screened to compare their capabilities of effecting ConA-induced liver damage.62 As shown in Figure 5-15, preladenant caused the most severe liver damage among these tested compounds, thus it was employed as a starting lead compound for the subsequent chemical modifications. Meanwhile, the excellent binding profile of tozadenant as well as its continuous promising progress in clinical trials made it another ideal lead compound for our novel immunotherapeutic development. Fluorinated analogs were synthesized to potentially serve as leads to be labeled at the distal position with fluorine-18. A series of fluorinated PEG groups with increasing chain lengths were designed to map the structure-activity relationship (SAR): such modifications would increase both hydrophilicity and molecular weight (MW), and potentially to reduce BBB penetration according to the central nervous system multiparameter optimization (CNS-MPO) score reported by Wager et al.72

Figure 5-15. Preliminary in vivo liver damage results from ConA mouse model.

The Glide docking experiments based on the A2AR homology model mentioned above were carried out to identify the ideal position for chemical modifications on preladenant and tozadenant (See experimental sections 3.1-3.4 for details). As shown in Figure 5-16A,

255 preladenant has a relatively large geometry/size and it fits into the entire binding pocket of A2AR and bears similar key binding interactions as that of ZM241385. The hetero-tricyclic core structure forms H-bonding with Glu169 and Asn253 and face-to-face π-π stacking with Phe168.

The resulting Glide XP docking score value is -12.0 kcal/mol. Notably, the methoxyethoxyl ether group reaches the cytosolic solution, where the oxygen atom forms an additional H-bond with

Pro266 at the solvent-exposed surface. This binding pose indicates the possibility of introducing hydrophilic and fluorinated PEG groups at this phenolic hydroxyl group without affecting the current key binding interactions of the core structure.

Figure 5-16. Glide XP docking results of preladenant and tozadenant.

(A) (B)

Pictures are rendered from YASARA. The residues of A2AR are colored grey and the H-bond is represented by the cyan dotted line. (A) Preladenant is colored green for carbon, red for oxygen, blue for nitrogen. (B) Tozadenant is colored brown for carbon.

256

Docking studies of tozadenant (Figure 5-16B) revealed that the piperidine quaternary alkyl

group also sits near the top of the A2AR binding pocket, where its tertiary alcohol group is hydrogen-bonded to Glu169, besides the hydrophobic interactions with Leu267 and His264. The benzyothiazole heterocycle part of tozadenant also forms face-to-face π-π stacking with Phe168 and the carbamate carbonyl group is hydrogen bonding to the backbone nitrogen atom of Phe168

(not shown). The phenolic methyl ether oxygen atom forms a hydrogen bond with the key residue Asn253. Considering the immunotherapeutic applications, the new analogs based on this core structure would require more drastic structural additions than preladenant in order to potentially diminish its BBB penetration due to its much smaller size and lower molecular weight. Herein, the octaethylene glycol monomethyl ether group, a beneficial substituent of KW-

PEG, combined with the phenyl-piperazine linker presented in preladenant is employed to replace the piperidine moiety of tozadenant. This substructure is expected to direct itself into the cytosolic medium, thus would not disturb the key binding events of tozadenant.

5.5 Synthesis of the PEGylated derivatives

5.5.1 Optimization and scale up of KW6002 and KW-PEG

Following a reported method,55 synthesis of KW6002 commenced with a reaction between cyanoacetic acid and dialkyl urea to form 6-aminouracil-1,3-diethyl-1h-pyrimidine-2,4-dione 5-1

(Scheme 5-2). Installation of the nitroso group was achieved via sodium nitrate and subsequent reduction of the nitroso group using sodium dithionate gave diaminouracil 5-3. Coupling of 5-3 with (E)-3,4-dimethoxycinnamic acid 5-4 in the water soluble carbodiimide(3-dimethyl-

257 aminopropyl) ethylcarbodiimidehydrochloride (EDCI) solution led to arylamide compound 5-5.

Cyclization of compound 5-5 under basic conditions, and the subsequent methylation with iodomethane furnished the final compound KW6002. The recrystallization procedures were optimized for the key intermediates 5-3, 5-5 and 5-6, resulting in an improvement of the corresponding yields for these compounds. Finally, six grams of KW6002 was made with a 6- fold increase of the yield over the entire synthesis (i.e. from 6%55 to 24%).

Scheme 5-2. Synthesis of KW6002.

Preparation of KW-PEG also followed our previously reported method (Scheme 5-3),55 with exception to the azobisisobutyronitrile (AIBN)/NBS (N-bromosuccinimide) promoted formation

258 of the 8-substituted xanthine scaffold 5-11.73 Heck coupling of acrolein 5-8 to iodobenzoate 5-7 under phosphine-free conditions led to compound 5-9, which was then coupled to diaminouracil

5-10 in the presence of AIBN/NBS to afford compound 5-11. Methylation of compound 5-11 and subsequent ester saponification yielded compound 5-13, which was conjugated with octaethylene glycol monomethyl ether to give KW-PEG.

Synthesis of 5-11 adopted a radical mechanism to achieve the cyclization reaction under room temperature in 2.5 h with a yield of 56%. In comparison, the previous method utilized the bromodimethylsulfonium bromide (BDMS) salt, which required the preparation from dimethyl sulfide and bromine before use.55 The BDMS promoted cyclization reaction only led to a highest yield of 22% after overnight reaction at room temperature. Finally, three grams of KW-PEG was furnished following this optimized protocol.

259

Scheme 5-3. Synthesis of KW-PEG.

5.5.2 Synthesis of preladenant and tozadenant and their PEGylated derivatives

Different synthetic methods were attempted for the synthesis of preladenant before the robust method in Scheme 5-4 was selected.74-76 The key intermediate 5-18 for the final coupling step was obtained as follows: 5-14 was reacted with Vilsmeier reagent to get the halogenated and formylated intermediate 5-15, which was condensed consecutively with 2-furoic acid hydrazide and 2-hydroxyethyl hydrazine to get intermediate 5-17. The following Dimroth rearrangement of compound 5-17 with POCl3/ZnBr2 under elevated temperature allowed the formation of the triazole ring and bromination of the primary alcohol to get the intermediate 5-18.

260

Scheme 5-4. Synthesis of intermediate 5-18 for preladenant and its PEGylated analogs.

The coupling counterparts of 5-26 to 5-31 were prepared starting from either commercially available coupling fragment 5-19 or fluorination/activation products of the mono-or di-tosylated

PEG chains (5-20 to 5-24), and subsequent coupling reaction with 1-(4-(4-hydroxyphenyl) piperazin-1-yl) ethan-1-one 5-25, followed by deacetylation prior to the final coupling reaction with intermediate 5-18. This method was therefore selective to the desired N-7 position and led to the easy to preparation of the designed analogs. Preladenant and the desired analogs 5-32 to 5-

36 could be achieved under basic reaction conditions.74, 77 Further optimizations of the final coupling step, regarding to the reaction temperature and stoichiometires of the involving reagents, resulted in the desired analogs with yields ranging from 25% to 55%. The lengthier fragments led to lower yields in this coupling reaction (Scheme 5-5), leaving more of the unreacted starting materials.

261

Scheme 5-5. Synthesis of preladenant and its PEGylated analogs.

Another two attempted methods, however, suffered from regioselectivity and reproducibility problems. In method I (Scheme 5-6), the key intermediate 5-37 reacted with hydrazine monohydrate to get 5-38, and a subsequent Dimroth rearrangement in the presence of hexamethyldisilazane (HMDS) and N,O-bis(trimethylsilyl)acetamide (BSA) at 120°C overnight yielded 5-39.75 However, the reaction conditions for the Dimroth rearrangement step were harsh and difficult to reproduce. The workup was also problematic due to the significant amount of acetamide from the BSA along with other salts. The high polarity and poor solubility in organic solvents of compound 5-39 also made it hard for the extraction and subsequent purification via either recrystallization or column chromatography. According to the literature,76 the carbazide

262 starting material 5-38 and the product 5-39 of this reaction are both insoluble in silicon containing reagents, an obstacle that complicates the process control and the separation/purification of the product. In addition, silicon containing reagents are very sensitive to hydrolysis and thus the reaction demands special protection from moisture. Moreover, synthesis of compound 5-40 via this method suffered poor yield (23% for a mixture of N-7 and

N-8 isomers) because of the double coupling nature of the ditosylated ethylene glycol reactant and the poor selectivity over the N-7 and N-8 positions for this coupling reaction. According to

Baraldi et al., purification of the N-7 and N-8 regioisomers was extremely difficult via column chromatography due to their similar polarities.78

Scheme 5-6. Synthesis of Preladenant with method I.

The mechanism of the Dimroth rearrangement for the formation of the triazole ring in 5-39 was not disclosed in literature. A reaction mechanism, according to the synthesis of similar core 263 structure scaffolds,79-81 is proposed herein for the synthesis of 5-39 from 5-38 via base-catalyzed tandem ring opening and ring closure reactions, as shown in Scheme 5-7. The occurrence of such rearrangement is attributed to the fact that the [1,2,4]triazolo[1,5-c]pyrimidine ring system is thermodynamically more stable than its isomer [1,2,4]triazolo[4,3-c]pyrimidine.82

Scheme 5-7. Proposed mechanism for the synthesis of compound 5-39.

The other alternative method is shown in Scheme 5-8 to address the problems mentioned above.76 In this method, the pyrazole ring was constructed first via the coupling reaction between intermediates 5-42 and 5-15 to access intermediate 5-43, thus avoiding the N-7, N-8 selectivity 264 issue. The triazole ring and the installation of furan group were achieved through the condensation of hydrazine bearing compound 5-44 and furfural 5-45, followed by (diacetoxyiodo) benzene promoted ring formation reaction. However, the hydrazine derivative 5-42 was found to be very unstable, and this intermediate decomposed when it was dried under vacuum overnight, thus its yield was not determined. The mono-Boc-protected hydrazine was also tried to replace the hydrazine monohydrate to make it a potentially more stable intermediate. However, the protected hydrazine led to very poor yield for the Boc-protected intermediate 5-42.

265

Scheme 5-8. Alternative method for the synthesis of preladenant.

Meanwhile, to probe the coupling reaction between compound 5-42 and 5-15, the reaction between compound 5-15 and hydroxyethyl hydrazine was first investigated. However, it was found that the side product 5-48 predominated in the reaction mixture with a ration of 1: 9

(Scheme 5-9). Similar problem might exist in the reaction of 5-42 and 5-15.

266

Scheme 5-9. Reaction of compound 5-15 and hydroxylethyl hydrazine.

Another concern is that the (diacetoxyiodo) benzene may not be able to promote the cyclization of 5-46 to access the triazole ring since this reagent was not able to realize the

Dimroth rearrangement for the intermediate 5-38 (Scheme 5-6). In addition, the methoxyethoxy group was introduced at the beginning of the synthesis, thus it was not convenient for the synthesis of the other PEGylated derivatives. Therefore, none of the above two methods were used for the synthesis of preladenant and its analogs.

5.6.3 Synthesis of tozadenant and its derivatives

Synthesis of tozadenant is illustrated in Scheme 5-10. Following a patent method,83 compound 5-49 was coupled with morpholine under palladium catalyzed conditions to get 5-50.

The nitro-group of 5-50 was reduced with tin powder to give the aniline derivative 5-51.

Subsequent condensation with benzoyl isothiocyanate led to intermediate 5-52, which was cyclized by bromine to form the bi-cyclic benzo-thiazole ring in compound 5-53. Installation of the piperidine ring and PEGylated phenyl-piperazine fragments through intermediate 5-54 afforded the final product of tozadenant and new derivative 5-55 respectively.

267

Scheme 5-10. Synthesis of tozadenant and its demethylated and PEGylated analogs.

The demethylation reaction at the phenolic methyl ether position of tozadenant was also investigated. Preparation of a demethylated product, such as 5-56 (Scheme 5-11), could facilitate

11 11 potential synthesis of radiotracers, such as [ C]tozadenant based on its [ C]CH3I methylation and 18F-labeled tracers through coupling with 18F-labeled short alkyl chains. Initially, direct

268 demethylation of tozadenant was tried with BBr3, however, this method did not lead to the desired product 5-56, but resulted in the decomposition of tozadenant and bromination of its

84 tertiary alcohol. Though L-selectride, as an alternative demethylation reagent, led to the desired product 5-56, the conversion was negligible (5%).85

This transformation was finally realized via a two-step synthesis through the phenyl carbamated intermediate 5-54 (Scheme 5-11). Due to the presence of this phenyl carbamate protecting group, the demethylation reaction proceeded with a yield of 52%. Subsequent replacement of the phenyl carbamate group of 5-57 with 4-methylpiperidin-4-ol hydrochloride afforded the desired product 5-56.84

Scheme 5-11. Synthesis of 5-56.

5.6 Immunoassay results

Two immunoassays were carried out for preladenant, tozadenant and their analogs in

comparison with the previous KW6002 and KW-PEG to evaluate A2AR binding-dependent

55 signaling through A2ARs on the surface of immune cells. The A2AR agonist, CGS21680 (CGS), was used to activate the A2AR. The first assay serves to screen compounds based on their extent

269 of inhibition toward A2AR-induced intracellular cAMP accumulation in A2AR expressing lymphocytes (Figure 5-17).58, 63

Figure 5-17. Cyclic-AMP results from lymphocytes.

The cAMP level was tested after incubation with vehicle, 1 µM CGS, 1 µM CGS plus 1 mM/mLof preladenant, compounds 5-32 to 5-36, tozadenant, 5-55, KW6002, and KW-PEG. The intracellular cAMP levels were determined 15 min following stimulation using quantitative cAMP ELISA and are expressed as fmol/million cells. Data shown represent mean ± SEM of triplicate samples.

As shown in Figure 5-17, all of the tested compounds, except the PEGylated tozadenant

derivative 5-55, are able to prevent CGS21680-activated A2AR signaling by blocking A2AR 270 activation rendered cAMP accumulation. The preladenant-based analogs 5-32 to 5-34 exhibited stronger antagonism than the previously evaluated KW6002 and KW-PEG in this functional assay. Longer PEG chain-bearing derivatives of preladenant (i.e. 5-35 & 5-36) resulted in decreased suppression of intracellular cAMP level. Surprisingly, the more potent A2AR antagonist tozadenant showed inferior inhibition for intracellular cAMP than that of KW-6002 and KW-PEG, and its PEGylated derivative 5-55 even exhibited no inhibition toward intracellular cAMP.

Encouraged by these results, a re-examination of the positive hits 5-32 to 5-34 in the cAMP assay were carried out in silico by Glide docking to predict their binding affinity in the A2AR.

The docking results confirmed the initial assumption for such analog design (Figure 5-18). As expected, all the core structures of 5-32 to 5-34 anchored in similar positions to preladenant, forming identical key binding interactions with Asn253, Glu169 and Phe168. The installed PEG chain tails project into the protein-cytosolic solution surface and interact with the residues at the

edge of A2AR via hydrophobic and H-bonding interactions. The disclosed high A2AR binding affinities of compounds 5-32 and 5-33 (i.e. 2.8 nM and 1.8 nM respectively)86 and their high

Glide XP docking scores of -12.2 kcal/mol and -11.8 kcal/mol respectively further validated the reliability of this molecular modeling methodology. Compound 5-34 also gave a high Glide XP docking score of -12.3 kcal/mol, indicating strong binding affinity with the A2AR binding pocket.

Biological assays to determine the experimental A2AR binding potency are ongoing.

271

Figure 5-18. Docking results of compounds 5-32 to 5-34 via Glide XP method.

Pictures are rendered from YASARA. The interacting residues of A2AR are colored grey and the H-bond is represented as a cyan dotted line. Carbon atoms of compounds 5-32 to 5-34 are colored as purple for compound 5-32 cyan for compound 5-33, and pink for compound 5-34.

272

With the help of Professor Sitkovsky’s group at Northeastern university, 5-34, KW-6002,

KW-PEG and preladenant were tested for their capability in restoring the secretion of the IFN- gamma, a proinflammatory cytokine to potentiate and prolong the immune response, which is

55 sensitive to the A2AR signaling pathway (Figure 5-19).

Figure 5-19. The IFN-gamma assay results from splenocytes.

The IFN-gamma production by splenocytes was tested after activation with 0.1 µg/mL anti- CD3 and when treated with vehicle, 1 µM CGS, 1 µM CGS plus either 1 µM Compound 5- 34, KW6002, KW-PEG, or preladenant. The IFN-gamma levels were determined in the supernatant one day following stimulation using quantitative ELISA and are expressed as pg/mL. Data shown represent mean ± SEM of triplicate samples.

In these assays, the C57BL/6 mice splenocytes T cells were incubated with A2AR agonist

CGS21680 (CGS) to inhibit IFN-gamma secretion during T cell receptor (TCR) activation by the

273

CD3 ligand. The inhibition of IFN-gamma cytokine came from A2AR activation induced elevation of intracellular cAMP. In this assay, effective A2AR antagonists/immunotherapeutics would block the CGS-activated A2AR signal, thus restoring the secretion of IFN-gamma to rescue the immune response. As illustrated in Figure 5-18, the lead compound 5-34 from the cAMP assay showed similar restoration of IFN-gamma levels to that of the potent and selective

A2AR antagonist, preladenant. Compared with the previously evaluated KW6002 and KW-PEG, compound 5-34 and preladenant displayed better restoration effects for the IFN-gamma secretion.

Given the promising results of compounds 5-32 to 5-34 in functional assays and molecular modeling studies, the physicochemical properties of these compounds were determined,

including their LogD7.4, aqueous solubility, human plasma protein binding (PPB) and metabolic stability (i.e. human liver microsome and rat hepatocyte clearance) as shown in Table 5-1.

Broadly similar results were obtained for these compounds, however, compared with the recently

87 evaluated compound 5-32 as a successful PET tracer for A2AR, compound 5-34, however, showed poor aqueous solubility and fast intrinsic clearance.

Table 5-1. Physicochemical properties and docking results of compounds 5-32 to 5-34.

Glide Aqueous Rat Hepatocyte Human HLM CLint Comp. score LogD Solubility CLint 7.4 PPB (%) (µL/min/mg) (kcal/mol) (µM) (µL/min/106) 5-32 -12.2 2.8 74 99 16.5 29.9

5-33 -11.8 2.5 2 98.7 71.9 12.8

5-34 -12.3 2.3 10 98.3 66.1 72.9

274

5.7 Summary

We have designed and synthesized a family of PEGylated analogs of preladenant and tozadenant assisted by molecular modeling techniques. The successful lead compound 5-34, a fluorinated tri-ethylene glycol derivative of preladenant, gave promising results in cAMP and

IFN-gamma functional immunoassays. Future work will focus on studies of the detailed mechanistic mode of action of 5-34 and investigation of its feasibility as a potential cancer immunotherapeutic.

275

Experimental Section

1. Chemical Synthesis

All solvents were of reagent or anhydrous grade quality and purchased from Sigma-Aldrich, Alfa

Aesar, or Fisher Scientific. All reagents were purchased from Sigma-Aldrich, Alfa Aesar, Fisher

Scientific, or Oakwood Chemical, unless otherwise stated. All deuterated solvents were purchased from Cambridge isotopes. Analytical thin-layer chromatography (TLC) was

performed on pre-coated glass-backed plates (EMD TLC Silica gel 60 F254) and visualized using a UV lamp (254 nm), potassium permanganate stain. Silica gel for manual flash chromatography was high purity grade 40-63 μm pore size and purchased from Sigma-Aldrich. Yields refer to purified and spectroscopically pure compounds. NMR spectra were obtained with Bruker or

Varian, operating at 300 or 400 MHz separately for 1H acquisitions as noted. LCMS analysis was performed using a Waters Alliance reverse-phase HPLC, with single wavelength UV−visible detector and LCT Premier time-of-flight mass spectrometer (electrospray ionization). All newly synthesized compounds that were submitted for biological testing were deemed >95% pure by

LCMS analysis (UV and ESI-MS detection) prior to submission for biological testing.

Synthesis of KW6002 and KW-PEG were achieved followed by our previous report. All the compounds involved matched the reported characterizations.55 The novel method applied for the synthesis of compound 5-11 is illustrated as below: 73

Synthesis of methyl (E)-4-(2-(1,3-diethyl-2,6-dioxo-2,3,6,7-tetrahydro-1H-purin-8-yl)vinyl)-2- methoxybenzoate (5-11). In a 500 mL round-bottom flask, the solution of (E)-methyl 2-methoxy-

276

4-(3-oxoprop-1-en-1-yl) benzoate 5-9 (4.37 g, 19.84 mmol) and 1,3-diethyl-5,6-aminouracil 5-10

(3.96 g, 19.84 mmol) in acetonitrile / water (78 mL in a ratio of 9:1) was stirred for 15 min. To this mixture, catalytic amount of AIBN (0.06 g, 2% mmol) was added and after stirring for 5 min, NBS was added in two portions (1.78 g and 0.71 g after 30 min (2.49 g, 13.97 mmol in total)). The reaction was stirred at room temperature for another 2.5 h to complete. The resulting precipitate was filtered under vacuum and purified by flash column chromatography using a 50% ethyl acetate in hexane to yield 5-11 as a yellow solid (4.42 g, 56%). The spectra were consistent with the previous report.55

Synthesis of 1-(4-(2-(2-(2-fluoroethoxy)ethoxy)ethoxy)phenyl)piperazine (5-28).74 The phenol 5-

25 (0.78 g, 3.56 mmol) was added to a solution of NaH (0.13 g, 5.34 mmol) in DMF (6.85 mL), followed by the addition of compound 5-21 (1.20 g, 3.92 mmol). The reaction was stirred at room temperature for 18 h, which was then concentrated and participated between ethyl acetate and 5% citric acid. The organic layer was washed with 1M NaOH solution and brine before it

was dried over MgSO4, and reduced under vacuum to obtain the alkylated product as sticky yellow oil. This yellow oil was then heated in 6M HCl (4.3 mL) in reflux for 1 h. The reaction was allowed to cool to room temperature and was basified to pH = 10 with aq. NaOH solution.

The CH2Cl2 was used to extract the product. The resulting organic layer was washed with brine and water and dried with MgSO4 before it was concentrated to give dark red oil. The flash column chromatography purification with a gradient of 1-5% MeOH in CH2Cl2 yielded the

1 desired product 5-28 as yellow solid (0.62 g, 56%). HNMR (400 MHz, CDCl3) δ 6.84-6.90 (m,

277

4H), 5.30 (brs, 1H), 4.57 (dt, 2H, J= 47.6 Hz, 4.4 Hz), 4.09 (t, 2H, J= 5.1 Hz), 3.84 (t, 2H, J= 5.1

13 Hz), 3.80 (t, 1H, J= 4.4 Hz), 3.69-3.77 (m, 5H), 3.03-3.05 (m, 8H). CNMR (100 MHz, CDCl3)

δ 153.1, 146.6, 118.3, 115.5, 83.3(d, J= 168.6 Hz), 71.0, 70.7, 70.5, 70.1, 68.0, 51.9, 46.4.

LCMS found 313.2 [M + H] +.

Synthesis of 1-(4-((17-fluoro-3,6,9,12,15-pentaoxaheptadecyl)oxy)phenyl)piperazine (5-

29).Following the same method for the synthesis of compound 5-28, compound 5-29 was

1 obtained as yellow oil (0.11 g, 35%). HNMR (400 MHz, CDCl3) δ 6.83-6.88 (m, 4H), 5.30 (brs,

1H), 4.54 (dt, 2H, J= 48.4 Hz, 4.4 Hz), 4.07 (t, 2H, J= 5.9 Hz), 3.81 (t, 2H, J= 5.1 Hz), 3.76 (t,

13 1H, J= 3.7 Hz), 3.61-3.72 (m, 17H), 3.01-3.03 (m, 8H). CNMR (100 MHz, CDCl3) δ 151.8,

145.3, 117.0, 114.2, 82.0(d, J= 169.4 Hz), 69.7, 69.6, 69.5, 69.4, 69.3, 69.2, 68.8, 66.8, 50.7,

45.2. LCMS found 445.3 [M + H] +.

Synthesis of 1-(4-((2,5,8,11,14,17,20,23-octaoxapentacosan-25-yl)oxy)phenyl)piperazine (5-

31).Following the same method for the synthesis of compound 5-28, compound 5-31 was

1 obtained as yellow oil (0.78 g, 70%). HNMR (400 MHz, CDCl3) δ 6.83-6.88 (m, 4H), 5.30 (brs,

1H), 4.08 (t, 2H, J= 3.7 Hz), 3.83 (t, 2H, J= 5.1 Hz), 3.70-3.73 (m, 2H), 3.62-3.67 (m, 23H),

13 3.52-3.56 (m, 3H), 3.38 (s, 3H), 3.0-3.05 (m, 8H). CNMR (100 MHz, CDCl3) δ 153.0, 146.5,

118.2, 115.4, 72.0, 70.9, 70.7, 70.6, 70.0, 68.0, 59.1, 53.8, 51.9, 46.4. LCMS found 545.3[M +

H] +.

Synthesis of 7-(2-(4-(4-(2-(2-(2-fluoroethoxy)ethoxy)ethoxy)phenyl)piperazin-1-yl)ethyl)-2-

(furan-2-yl)-7H-pyrazolo[4,3-e][1,2,4]triazolo[1,5-c]pyrimidin-5-amine (5-34).74 In a flame

278 dried 10 mL round-bottom flask, compound 5-18 (0.20 g, 0.58 mmol) and 5-28 (0.15 g, 0.48 mmol) were dissolved in 1.5 mL DMF, followed by the addition of DIPEA (0.38 mL). The reaction mixture was heated to 80 °C for 10 h. After the reaction was completed, it was diluted with water (5 mL), extracted with ethyl acetate and washed with water and then brine. The

organic layer was then dried with MgSO4 and concentrated under vacuum to give the dark brown oil, which was purified via flash column chromatography using a gradient of 1-10% MeOH in

CH2Cl2 to yield the desired product 5-34 as yellow powder (91 mg, 33%). Mp: 171-172°C.

1 HNMR (300 MHz, CDCl3) δ 8.20 (s, 1H), 7.59-7.61 (m, 1H), 7.24 (d, 1H, J=3.5 Hz), 6.84-6.86

(m, 4H), 6.57-6.60 (m, 1H), 6.11 (brs, 2H), 4.62-4.65 (m, 1H), 4.47-4.55 (m, 3H), 4.08 (t, 2H,

J=4.8 Hz), 3.79-3.85 (m, 3H), 3.69-3.75 (m, 5H), 3.06-3.09 (t, 4H, J=4.7 Hz), 2.97 (t, 2H, J=7.0

13 Hz), 2.73 (t, 4H, J=4.8 Hz). CNMR (75 MHz, CDCl3) δ 156.6, 152.9, 149.2, 148.1, 145.8,

145.6, 145.0, 144.5, 132.1, 117.9, 115.4, 112.5, 111.9, 97.3, 83.1 (d, J=169.2 Hz), 70.8, 70.5,

70.3, 69.9, 67.9, 56.9, 53.2, 50.4, 45.0. LCMS found 580.3 [M + H] +.

Synthesis of 7-(2-(4-(4-((17-fluoro-3,6,9,12,15-pentaoxaheptadecyl)oxy)phenyl)piperazin-1- yl)ethyl)-2-(furan-2-yl)-7H-pyrazolo[4,3-e][1,2,4]triazolo[1,5-c]pyrimidin-5-amine (5-35).

Following the same method for the synthesis of compound 5-34, compound 5-35 was obtained as

1 greasy yellow solid (42 mg, 25%). HNMR (400 MHz, d6-DMSO) δ 8.16 (s, 1H), 8.05 (brs, 2H),

7.93 (s, 1H), 7.21 (d, 1H, J=2.9 Hz), 6.90 (d, 1H, J=8.8 Hz ), 6.76-6.84 (m, 4H), 6.70-6.73 (m,

1H), 4.53-4.56 (m, 1H), 4.38-4.41 (m, 3H), 3.95-4.0 (m, 3H), 3.63-3.70 (m, 4H), 3.45-3.60 (m,

15H), 3.0 (t, 1H, J=5.1 Hz), 2.90-2.95 (m, 4H), 2.81 (t, 2H, J=5.9 Hz), 2.55-2.29 (m, 3H).

279

13 CNMR (75 MHz, d6-DMSO) δ 156.6, 152.9, 149.2, 148.1, 145.8, 145.6, 145.0, 144.5, 132.1,

117.9, 115.4, 112.5,111.9, 97.3, 83.5 (d, J=165.8 Hz), 70.8, 70.5, 70.3, 69.9, 67.9, 56.9, 53.2,

50.4, 45.0. LCMS found 712.3 [M + H] +.

Synthesis of 7-(2-(4-(4-((2,5,8,11,14,17,20,23-octaoxapentacosan-25-yl)oxy)phenyl)piperazin-1- yl)ethyl)-2-(furan-2-yl)-7H-pyrazolo[4,3-e][1,2,4]triazolo[1,5-c]pyrimidin-5-amine (5-36).

Following the same method for the synthesis of compound 5-34, compound 5-36 was obtained as

1 greasy yellow solid (0.15 g, 25%). HNMR (300 MHz, CDCl3) δ 8.16 (s, 1H), 7.54-7.55 (m, 1H),

7.20 (d, 1H, J= 3.5 Hz), 6.78-6.86 (m, 4H), 6.67 (brs, 2H), 6.56 (dd, 1H, J= 3.4, 1.8 Hz), 4.64 (t,

2H, J= 6.6 Hz), 4.06 (t, 2H, J= 4.8 Hz), 3.81 (t, 2H, J=4.8 Hz), 3.63-3.64 (m, 28H), 3.52-3.55

13 (m, 2H), 3.37 (s, 3H), 3.19-3.22 (m, 4H), 2.91-2.94 (m, 4H). CNMR (75 MHz, CDCl3) δ 156.5,

153.4, 149.0, 148.3, 145.6, 145.5, 144.8, 144.5, 132.4, 118.4, 115.4, 112.5, 111.9, 97.2, 71.8,

70.7, 70.6, 70.5, 70.4, 70.3, 69.8, 67.8, 59.0, 56.3, 54.0, 52.9, 49.6, 44.0, 42.2. LCMS found

812.4 [M + H] +.

Synthesis of 4-(4-((2,5,8,11,14,17,20,23-octaoxapentacosan-25-yl)oxy)phenyl)-N-(4-methoxy-7- morpholinobenzo[d]thiazol-2-yl)piperazine-1-carboxamide (5-55).88 To a solution of compound

5-54 (35.40 mg, 91.80 μmol) and N,N-Diisopropylethylamine (DIPEA, 48.70 μL, 276.70 μmol) in chloroform (0.55 mL) was added a solution of compound 5-31 (50.00 mg, 91.8 μmol) in chloroform (0.33 mL) and tetrahydrofuran (0.33 mL). The resulting mixture was heated to reflux for 1 h. The reaction mixture was then cooled to ambient temperature and extracted with saturated aqueous sodium carbonate (0.17 mL) and water (2*0.10 mL). Final drying with

280 magnesium sulfate and evaporation of the solvent and flash column chromatography purification using a gradient of 1-3% MeOH in dichloromethane yielded compound 5-55 as greasy pale

1 yellow solid (21.00 mg, 28%). HNMR (400 MHz, CDCl3) δ 6.76-6.86 (m, 6H), 4.06 (t, 2H, J=

4.4 Hz), 3.90 (s, 3H), 3.86 (t, 4H, J= 4.4 Hz), 3.81 (t, 2H, J= 4.4 Hz), 3.67-3.75 (m, 6H), 3.56-

3.66 (m, 24H), 3.53 (t, 2H, J= 4.4 Hz), 3.36 (s, 3H), 3.02-3.11 (m, 8H). 13CNMR (100 MHz,

CDCl3) δ 153.8, 145.5, 140.7, 126.4, 119.0, 115.6, 112.5, 107.4, 72.1, 71.0, 70.9, 70.8, 70.7,

70.0, 68.0, 67.6, 59.3, 56.2, 52.0, 50.9, 44.3, 29.9. LCMS found 836.4 [M + H] +.

Synthesis of phenyl (4-hydroxy-7-morpholinobenzo[d]thiazol-2-yl)carbamate (5-57).84 To a solution of compound 5-54 (0.60 g, 1.48 mmol) in dichloromethane (31.2 mL) was added BBr3

(10.6 mL, 1 M solution in dichloromethane) dropwise at -78 °C in a liquid nitrogen-ethanol bath.

The mixture was allowed to warm to room temperature and was stirred overnight. Water (20 mL)

was then added to quench the reaction. After neutralization with NH4OH until pH = 7.0, the reaction mixture was extracted with CH2Cl2 (3 x 20 mL). After removing the solvent under vacuum and flash column chromatography purification with 1% methanol in CH2Cl2, compound

5-57 was obtained as a pale yellow solid (0.29 g, 52%). Mp: decomposed at 143 °C. 1HNMR

(400 MHz, CDCl3) δ 7.43 (t, 2H, J= 8.1 Hz), 7.30 (t, 1H, J= 7.3 Hz), 7.24 (t, 2H, J= 8.1 Hz),

6.99 (d, 1H, J= 8.1 Hz), 6.89 (d, 1H, J= 8.8 Hz), 3.86 (t, 4H, J= 5.1 Hz), 3.08 (t, 4H, J= 4.4 Hz).

13 CNMR (100 MHz, CDCl3) δ 175.3, 159.5, 152.0, 150.2, 144.6, 140.2, 137.9, 129.9, 126.8,

121.6, 114.7, 112.6, 67.6, 52.1. LCMS found 371.9 [M + H] +.

281

Synthesis of 4-hydroxy-N-(4-hydroxy-7-morpholinobenzo[d]thiazol-2-yl)-4-methylpiperidine-1- carboxamide (5-56).88 Following the same method for the synthesis of compound 5-55, compound 5-56 was obtained as pale yellow solid (65.00 mg, 62%). Mp: decomposed at 155 °C.

1 H NMR (400 MHz, d6-DMSO) δ 11.14 (brs, 1H), 9.36 (brs, 1H), 6.68-6.70 (m, 2H), 4.36 (brs,

1H), 3.83 (m, 2H), 3.73 (t, 4H, J= 4.41 Hz), 3.20-3.28 (m, 2H), 2.93 (t, 4H, J= 4.4 Hz), 1.36-1.48

13 (m, 4H), 1.12 (s, 3H). CNMR (75 MHz, d6-DMSO) δ 173.0, 155.0, 151.7, 139.0, 137.7, 112.7,

111.7, 108.0, 67.1, 66.5, 53.6, 52.0, 38.7, 30.2. LCMS found 393.0 [M + H] +.

2. Bioassay procedures

2.1 Induction of liver injury62

Assay was performed by Professor Sitkovsky’s group at Northeastern University: Mice were injected intravenously with Con A (20 mg kg-1) in sterile PBS, and serum samples were taken or mice were killed at indicated time points. Some mice were coinjected intraperitoneally with

CGS21680 (2 mg kg-1), KW6002 (2 mg kg-1), KW-PEG (2 mg kg-1) and preladenant (2 mg kg-1) separately just before treatment with Con A. The magnitude or liver damage was evaluated by serum alanine aminotransferase (ALT) levels and liver tissue histology.

55 2.2 Measuring functionality of A2AR antagonism by cAMP assay

Assay was performed by Professor Sitkovsky’s group at Northeastern University: Stimulation of intracellular cAMP production and measurement of cAMP levels were performed as described previously.58 Lymphocytes were isolated from the spleen of C57/BL6 mice and suspended at a

282 concentration of 400,000 cells/well, and then treated with 1 μM CGS 21680 (A2AR-specific agonist; from Tocris, Ellisville, MO) with or without KW6002, KW-PEG, preladenant, tozadenant and their PEGylated analogs 5-32 to 5-36 and 5-55 (at a concentration of 1 mM/mL).

The cells were incubated for 15 min at 37 °C, and the reaction was stopped by addition of 1 N hydrochloric acid. Levels of cAMP were determined by ELISA (Amersham Biosciences,

Buckinghamshire, UK). All treatment groups were done in triplicate.

283

2.3 Cytokine release assay55

Assay was performed by Professor Sitkovsky’s group at Northeastern University: Splenocytes (2 x 106 /mL) were isolated from the spleen of C57/BL6 mice and activated with 0.1 μg/mL CD3 mAb to induce production of IFN-gamma. Immediately following the addition of mAb-CD3, the cells were treated with or without 1 μM CGS 21680 agonist and 1 uM KW6002, KW-PEG, preladenant and compound 5-34. Supernatants were collected after 24 h and levels of IFN- gamma were assayed by ELISA (Amersham Biosciences, Buckinghamshire, UK) using paired mAb and standard purchased from BD Pharmingen. All treatment groups were done in triplicate.

3. Molecular modeling procedures

3.1 Homology modeling

The homology model from previous report was used for the docking study (Also see Dr. Joslynn

Lee’s thesis).55 It was constructed via YASARA (Yet Another Scientific Artificial Reality

Application) based on the crystal structure of A2AR in complex with ZM241385 (PDB 3EML)54 to fix the second extracellular loop (ECL2) residues Gln148 to Ser156. Residues were missing due to weak experimental electron density in that region. The quality of this homology model was examined by PROCHECK and was found to be of sufficiently good quality (Also see Dr.

Joslynn Lee’s thesis).

3.2 Computational studies

All calculations performed in this work were carried out on two Cooler Master Centurion 5 (Intel

Core-i7 Quad CPU Q6600 @ 3.33GHz) operating system running Maestro 10.4 (Schrödinger,

LLC, New York, NY, 2015), ChemBioDraw® Ultra 14.0 (PerkinElmer) and YASARA (Yet

284

Another Scientific Artificial Reality Application). All pictures presented in this study were generated by Maestro and YASARA.

3.3 Molecule preparation

The ligands were prepared in ChemBio3D Ultra 14.0. The three-dimensional structures were built by importing the SMILES files of the ligands into LigPrep (LigPrep, version 3.6,

Schrödinger, LLC, New York, NY, 2015), implemented in Maestro 10.4. The LigPrep employs the Potentials for Liquid Simulations-all atom (OPLS-AA) force field 2005 for energy minimizations and a cellular pH value 7.0 to generate the most probably ionization state of the ligand.89

3.4 Docking studies

The homology model of A2AR was prepared using Maestro 10.4 Protein Preparation Wizard

(Schrödinger, LLC, New York, NY, 2015) before docking, bond orders were assigned and the orientation of hydroxyl groups, amide groups of the side chains of Asn and Gln, and the protonation state of His residues were optimized. A restrained refinement of the protein structure was performed using the default constraint of 0.3 Å RMSD and the OPLS 2005 force field. The docking studies were performed by Glide (Grid-Based Ligand Docking with Energetics) (Glide, version 10.4, Schrödinger, LLC, New York, NY, 2015). The enclosing docking box was set up with the centroid of four selected atoms, e.g. Glu169 (OE1), His250 (NE2), Asn253 (ND2) and

His278 (NE2). The ligand diameter midpoint box was set up to be cubic with a box length of

20 Å. No other constrains were applied. Both standard precision (SP) and extra precision (XP) docking protocols were carried out and the binding poses were examined. Water molecules were not included in these docking studies due to the creation of the homology model.90

285

4. Determination of Aqueous Solubility, Log D7.4, Human Plasma Protein Binding (PPB)

91 and Human Mics CLint

Aqueous pH 7.4 Solubility

Compounds are dried down from 10 mM DMSO solutions using centrifugal evaporation technique. Phosphate buffer (0.1 M pH 7.4) added and StirStix inserted in the glass vials, shaking is then performed at a constant temperature of 25°C for 20-24 hours. This step is followed by double centrifugation with a tip wash in between, to ensure that no residues of the dried compound are interfering. The solutions are diluted before analysis and quantification using LC/MS/MS is performed.

91 Log D7.4

Shake-flask octanol-water distribution coefficient at pH 7.4 (Log D7.4). The aqueous solution used is 10 mM sodium phosphate pH 7.4 buffer. The method has been validated for Log D7.4 ranging from -2 to 5.0.

Human Plasma Protein Binding (PPB)

PPB is determined using equilibrium dialysis (RED device) to separate free from bound compound. The amount of compound in plasma (10 µM initial concentration) and in dialysis buffer (pH 7.4 phosphate buffer) is measured by LC-MS/MS after equilibration at 37°C in a dialysis chamber. The fraction unbound (fu) is reported.

Human Mics CLint

286

In vitro intrinsic clearance determined from human liver microsomes using a standard approach.

Following incubation and preparation, the samples are analyzed using LC/MS/MS. Refined data are uploaded to IBIS and are displayed as CLint (intrinsic clearance) in uL/min/mg.

Rat Heps CLint

In vitro intrinsic clearance determined from rat hepatocytes using a standard approach.

Following incubation and preparation, the samples are analyzed using LC/MS/MS. Refined data are uploaded to IBIS and are displayed as CLint (intrinsic clearance) uL/min/1 million cells.

287

References

1. Fredholm, B. B.; AP, I. J.; Jacobson, K. A.; Klotz, K. N.; Linden, J. International Union of Pharmacology. XXV. Nomenclature and classification of adenosine receptors. Pharmacol. Rev. 2001, 53, 527-552. 2. Trincavelli, M. L.; Daniele, S.; Martini, C. Adenosine receptors: what we know and what we are learning. Curr. Top. Med. Chem. 2010, 10, 860-877. 3. Jacobson, K. A.; Gao, Z. G. Adenosine receptors as therapeutic targets. Nat. Rev. Drug. Discov. 2006, 5, 247-264. 4. Moro, S.; Gao, Z. G.; Jacobson, K. A.; Spalluto, G. Progress in the pursuit of therapeutic adenosine receptor antagonists. Med. Res. Rev. 2006, 26, 131-159. 5. Congreve, M.; Langmead, C. J.; Mason, J. S.; Marshall, F. H. Progress in structure based drug design for G protein-coupled receptors. J. Med. Chem. 2011, 54, 4283-4311. 6. Ralevic, V.; Burnstock, G. Receptors for purines and pyrimidines. Pharmacol. Rev. 1998, 50, 413-492. 7. Peterfreund, R. A.; MacCollin, M.; Gusella, J.; Fink, J. S. Characterization and expression of the human A2a adenosine receptor gene. J. Neurochem. 1996, 66, 362-368. 8. Bilkei-Gorzo, A.; Abo-Salem, O. M.; Hayallah, A. M.; Michel, K.; Muller, C. E.; Zimmer, A. Adenosine receptor subtype-selective antagonists in inflammation and hyperalgesia. Naunyn Schmiedebergs Arch. Pharmacol. 2008, 377, 65-76. 9. Jaakola, V. P.; Ijzerman, A. P. The crystallographic structure of the human adenosine A2A receptor in a high-affinity antagonist-bound state: implications for GPCR drug screening and design. Curr. Opin. Struct. Biol. 2010, 20, 401-414. 10. Valls, M. D.; Cronstein, B. N.; Montesinos, M. C. Adenosine receptor agonists for promotion of dermal wound healing. Biochem. Pharmacol. 2009, 77, 1117-1124. 11. Monopoli, A.; Conti, A.; Dionisotti, S.; Casati, C.; Camaioni, E.; Cristalli, G.; Ongini, E. Pharmacology of the highly selective A1 adenosine receptor agonist 2-chloro-N6-cyclopentyladenosine. Arzneimittelforschung 1994, 44, 1305-1312. 12. Monopoli, A.; Conti, A.; Dionisotti, S.; Casati, C.; Camaioni, E.; Cristalli, G.; Ongini, E. Pharmacology of the Highly Selective a(1) Adenosine Receptor Agonist 2-Chloro-N6- Cyclopentyladenosine. Arzneimittel-Forsch 1994, 44, 1305-1312. 13. Liu, X.; Smith, B. J.; Chen, C.; Callegari, E.; Becker, S. L.; Chen, X.; Cianfrogna, J.; Doran, A. C.; Doran, S. D.; Gibbs, J. P.; Hosea, N.; Liu, J.; Nelson, F. R.; Szewc, M. A.; Van Deusen, J. Evaluation of cerebrospinal fluid concentration and plasma free concentration as a surrogate measurement for brain free concentration. Drug Metab. Dispos. 2006, 34, 1443-1447. 14. Siddiqi, S. M.; Jacobson, K. A.; Esker, J. L.; Olah, M. E.; Ji, X. D.; Melman, N.; Tiwari, K. N.; Secrist, J. A. 3rd; Schneller, S. W.; Cristalli, G.; et al. Search for new purine- and ribose-modified adenosine analogues as selective agonists and antagonists at adenosine receptors. J. Med. Chem.1995, 38, 1174-1188. 15. van Tilburg, E. W.; Gremmen, M.; von Frijtag Drabbe Kunzel, J.; de Groote, M.; AP, I. J. 2,8- Disubstituted adenosine derivatives as partial agonists for the adenosine A2A receptor. Bioorg. Med. Chem. 2003, 11, 2183-2192. 16. Volpini, R.; Costanzi, S.; Lambertucci, C.; Taffi, S.; Vittori, S.; Klotz, K. N.; Cristalli, G. N(6)- alkyl-2-alkynyl derivatives of adenosine as potent and selective agonists at the human adenosine A(3) receptor and a starting point for searching A(2B) ligands. J. Med. Chem. 2002, 45, 3271-3279. 17. de Lera Ruiz, M.; Lim, Y. H.; Zheng, J. Adenosine A2A receptor as a drug discovery target. J. Med. Chem. 2014, 57, 3623-3650. 18. Hendel, R. C.; Wackers, F. J. T.; Berman, D. S.; Ficaro, E.; DePuey, E. G.; Klein, L.; Cerqueira, M. American society of nuclear cardiology consensus statement: Reporting of radionuclide myocardial perfusion imaging studies. J. Nucl. Cardiol. 2003, 10, 705-708.

288

19. Cerqueira, M. D. Advances in pharmacologic agents in imaging: new A2A receptor agonists. Curr. Cardiol. Rep. 2006, 8, 119-122. 20. Niiya, K.; Olsson, R. A.; Thompson, R. D.; Silvia, S. K.; Ueeda, M. 2-(N'- alkylidenehydrazino)adenosines: potent and selective coronary vasodilators. J. Med. Chem. 1992, 35, 4557-4561. 21. Niiya, K.; Thompson, R. D.; Silvia, S. K.; Olsson, R. A. 2-(N'- Aralkylidenehydrazino)Adenosines - Potent and Selective Coronary Vasodilators. J. Med. Chem. 1992, 35, 4562-4566. 22. Murphree, L. J.; Marshall, M. A.; Rieger, J. M.; MacDonald, T. L.; Linden, J. Human A(2A) adenosine receptors: high-affinity agonist binding to receptor-G protein complexes containing Gbeta(4). Mol. Pharmacol. 2002, 61, 455-462. 23. Garnock-Jones, K. P.; Curran, M. P. Regadenoson. Am. J. Cardiovasc. Drugs 2010, 10, 65-71. 24. Palle, V. P.; Elzein, E. O.; Gothe, S. A.; Li, Z.; Gao, Z.; Meyer, S.; Blackburn, B.; Zablocki, J. A. Structure-affinity relationships of the affinity of 2-pyrazolyl adenosine analogues for the adenosine A2A receptor. Bioorg. Med. Chem. Lett. 2002, 12, 2935-2939. 25. Lappas, C. M.; Sullivan, G. W.; Linden, J. Adenosine A2A agonists in development for the treatment of inflammation. Expert Opin. Investig. Drugs 2005, 14, 797-806. 26. Cohen, S. B.; Gill, S. S.; Baer, G. S.; Leo, B. M.; Scheld, W. M.; Diduch, D. R. Reducing joint destruction due to septic arthrosis using an adenosine(2A) receptor agonist. J. Orthopaed. Res. 2004, 22, 427-435. 27. Folkesson, H. G.; Kuzenko, S. R.; Lipson, D. A.; Matthay, M. A.; Simmons, M. A. The adenosine 2A receptor agonist GW328267C improves lung function after acute lung injury in rats. Am. J. Physiol. Lung Cell Mol. Physiol. 2012, 303, L259-271. 28. Yuan, G.; Jones, G. B. Towards next generation adenosine A(2A) receptor antagonists. Curr. Med. Chem. 2014, 21, 3918-3935. 29. Yuan, G.; Gedeon, N. G.; Jankins, T. C.; Jones, G. B. Novel approaches for targeting the adenosine A2A receptor. Expert. Opin. Drug. Discov. 2015, 10, 63-80. 30. Klotz, K. N.; Hessling, J.; Hegler, J.; Owman, C.; Kull, B.; Fredholm, B. B.; Lohse, M. J. Comparative pharmacology of human adenosine receptor subtypes - characterization of stably transfected receptors in CHO cells. Naunyn Schmiedebergs Arch. Pharmacol. 1998, 357, 1-9. 31. Daly, J. W.; Padgett, W. L.; Shamim, M. T. Analogs of Caffeine and Theophylline - Effect of Structural Alterations on Affinity at Adenosine Receptors. J. Med. Chem. 1986, 29, 1305-1308. 32. Shamim, M. T.; Ukena, D.; Padgett, W. L.; Daly, J. W. Effects of 8-phenyl and 8-cycloalkyl substituents on the activity of mono-, di-, and trisubstituted alkylxanthines with substitution at the 1-, 3-, and 7-positions. J. Med. Chem. 1989, 32, 1231-1237. 33. Erickson, R. H.; Hiner, R. N.; Feeney, S. W.; Blake, P. R.; Rzeszotarski, W. J.; Hicks, R. P.; Costello, D. G.; Abreu, M. E. 1,3,8-Trisubstituted Xanthines - Effects of Substitution Pattern Upon Adenosine Receptor-A1/A2 Affinity. J. Med. Chem. 1991, 34, 1431-1435. 34. Muller, C. E.; Sandoval-Ramirez, J.; Schobert, U.; Geis, U.; Frobenius, W.; Klotz, K. N. 8- (sulfostyryl)xanthines: Water-soluble A(2A)-selective adenosine receptor antagonists. Bioorg. Med. Chem. 1998, 6, 707-719. 35. Muller, C. E.; Sauer, P.; Maurinsh, Y.; Huertas, R.; Fulle, F.; Klotz, K. N.; Nagel, J.; Hauber, W. A(2A)-selective adenosine receptor antagonists: Development of water-soluble prodrugs and a new tritiated radioligand. Drug Develop Res. 1998, 45, 190-197. 36. Sauer, R.; Maurinsh, J.; Reith, U.; Fulle, F.; Klotz, K. N.; Muller, C. E. Water-soluble phosphate prodrugs of 1-propargyl-8-styrylxanthine derivatives, A(2A)-selective adenosine receptor antagonists. J. Med. Chem. 2000, 43, 440-448. 37. Hauber, W.; Nagel, J.; Sauer, R.; Muller, C. E. Motor effects induced by a blockade of adenosine A2A receptors in the caudate-putamen. Neuroreport. 1998, 9, 1803-1806. 38. Vollmann, K.; Qurishi, R.; Hockemeyer, J.; Muller, C. E. Synthesis and properties of a new water-soluble prodrug of the adenosine A 2A receptor antagonist MSX-2. Molecules 2008, 13, 348-359.

289

39. Santerre, J. L.; Nunes, E. J.; Kovner, R.; Leser, C. E.; Randall, P. A.; Collins-Praino, L. E.; Lopez Cruz, L.; Correa, M.; Baqi, Y.; Muller, C. E.; Salamone, J. D. The novel adenosine A(2A) antagonist prodrug MSX-4 is effective in animal models related to motivational and motor functions. Pharmacol. Biochem. Behav. 2012, 102, 477-487. 40. Shimada, J.; Suzuki, F.; Nonaka, H.; Ishii, A.; Ichikawa, S. (E)-1,3-dialkyl-7-methyl-8-(3,4,5- trimethoxystyryl)xanthines: potent and selective adenosine A2 antagonists. J. Med. Chem. 1992, 35, 2342-2345. 41. Dungo, R.; Deeks, E. D. Istradefylline: First Global Approval. Drugs 2013, 73, 875-882. 42. Chen, W. Q.; Wang, H. Q.; Wei, H. T.; Gu, S. L.; Wei, H. P. Istradefylline, an adenosine A(2A) receptor antagonist, for patients with Parkinson's Disease: A meta-analysis. J. Neurol. Sci. 2013, 324, 21- 28. 43. Jenner, P. Istradefylline, a novel adenosine A(2A) receptor antagonist, for the treatment of Parkinson's disease. Expert Opin. Investig. Drugs 2005, 14, 729-738. 44. Hockemeyer, J.; Burbiel, J. C.; Muller, C. E. Multigram-scale syntheses, stability, and photoreactions of A2A adenosine receptor antagonists with 8-styrylxanthine structure: potential drugs for Parkinson's disease. J. Org. Chem. 2004, 69, 3308-3318. 45. Cacciari, B.; Pastorin, G.; Spalluto, G. Medicinal chemistry of A2A adenosine receptor antagonists. Curr. Top. Med. Chem. 2003, 3, 403-411. 46. Francis, J. E.; Cash, W. D.; Psychoyos, S.; Ghai, G.; Wenk, P.; Friedmann, R. C.; Atkins, C.; Warren, V.; Furness, P.; Hyun, J. L.; et al. Structure-activity profile of a series of novel triazoloquinazoline adenosine antagonists. J. Med. Chem. 1988, 31, 1014-1020. 47. Silverman, L. S.; Caldwell, J. P.; Greenlee, W. J.; Kiselgof, E.; Matasi, J. J.; Tulshian, D. B.; Arik, L.; Foster, C.; Bertorelli, R.; Monopoli, A.; Ongini, E. 3H-[1,2,4]-Triazolo[5,1-i]purin-5-amine derivatives as adenosine A2A antagonists. Bioorg. Med. Chem. Lett. 2007, 17, 1659-1662. 48. Caulkett, P. W. R.; Jones, G.; Mcpartlin, M.; Renshaw, N. D.; Stewart, S. K.; Wright, B. Adenine Isosteres with Bridgehead Nitrogen .1. 2 Independent Syntheses of the [1,2,4]Triazolo[1,5- a][1,3,5]Triazine Ring-System Leading to a Range of Substituents in the 2-Position, 5-Position and 7- Position. J. Chem. Soc. Perk. T. 1 1995, 7, 801-808. 49. Biotie Therapies: Biotie’s Tozadenant (SYN115) Meets Primary and Multiple Secondary Endpoints in Phase 2b Study in Parkinson’s Disease. Press Release, December 11, 2012. 50. Langmead, C. J.; Andrews, S. P.; Congreve, M.; Errey, J. C.; Hurrell, E.; Marshall, F. H.; Mason, J. S.; Richardson, C. M.; Robertson, N.; Zhukov, A.; Weir, M. Identification of novel adenosine A(2A) receptor antagonists by virtual screening. J. Med. Chem. 2012, 55, 1904-1909. 51. Congreve, M.; Andrews, S. P.; Dore, A. S.; Hollenstein, K.; Hurrell, E.; Langmead, C. J.; Mason, J. S.; Ng, I. W.; Tehan, B.; Zhukov, A.; Weir, M.; Marshall, F. H. Discovery of 1,2,4-triazine derivatives as adenosine A(2A) antagonists using structure based drug design. J. Med. Chem. 2012, 55, 1898-1903. 52. Slee, D. H.; Zhang, X.; Moorjani, M.; Lin, E.; Lanier, M. C.; Chen, Y.; Rueter, J. K.; Lechner, S. M.; Markison, S.; Malany, S.; Joswig, T.; Santos, M.; Gross, R. S.; Williams, J. P.; Castro-Palomino, J. C.; Crespo, M. I.; Prat, M.; Gual, S.; Diaz, J. L.; Wen, J.; O'Brien, Z.; Saunders, J. Identification of novel, water-soluble, 2-amino-N-pyrimidin-4-yl acetamides as A2A receptor antagonists with in vivo efficacy. J. Med. Chem. 2008, 51, 400-406. 53. Lanier, M. C.; Moorjani, M.; Luo, Z. Y.; Chen, Y. S.; Lin, E.; Tellew, J. E.; Zhang, X. H.; Williams, J. P.; Gross, R. S.; Lechner, S. M.; Markison, S.; Joswig, T.; Kargo, W.; Piercey, J.; Santos, M.; Malany, S.; Zhao, M.; Petroski, R.; Crespo, M. I.; Diaz, J. L.; Saunders, J.; Wen, J.; O'Brien, Z.; Jalali, K.; Madan, A.; Slee, D. H. N-[6-Amino-2-(heteroaryl)pyrimidin-4-yl]acetamides as A(2A) Receptor Antagonists with Improved Drug Like Properties and in Vivo Efficacy. J. Med. Chem. 2009, 52, 709-717. 54. Jaakola, V. P.; Griffith, M. T.; Hanson, M. A.; Cherezov, V.; Chien, E. Y.; Lane, J. R.; Ijzerman, A. P.; Stevens, R. C. The 2.6 angstrom crystal structure of a human A2A adenosine receptor bound to an antagonist. Science 2008, 322, 1211-1217.

290

55. Thomas, R.; Lee, J.; Chevalier, V.; Sadler, S.; Selesniemi, K.; Hatfield, S.; Sitkovsky, M.; Ondrechen, M. J.; Jones, G. B. Design and evaluation of xanthine based adenosine receptor antagonists: potential hypoxia targeted immunotherapies. Bioorg. Med. Chem. 2013, 21, 7453-7464. 56. Plosker, G. L. Sipuleucel-T In Metastatic Castration-Resistant Prostate Cancer. Drugs 2011, 71, 101-108. 57. Andrew, P. New Class of Drugs Shows More Promise in Treating Cancer, New York Times. 58. Apasov, S.; Chen, J. F.; Smith, P.; Sitkovsky, M. A(2A) receptor dependent and A(2A) receptor independent effects of extracellular adenosine on murine thymocytes in conditions of adenosine deaminase deficiency. Blood 2000, 95, 3859-3867. 59. Sitkovsky, M. V.; Kjaergaard, J.; Lukashev, D.; Ohta, A. Hypoxia-adenosinergic immunosuppression: tumor protection by T regulatory cells and cancerous tissue hypoxia. Clin. Cancer Res. 2008, 14, 5947-5952. 60. Sitkovsky, M. V. T regulatory cells: hypoxia-adenosinergic suppression and re-direction of the immune response. Trends Immunol. 2009, 30, 102-108. 61. Sitkovsky, M.; Lukashev, D.; Deaglio, S.; Dwyer, K.; Robson, S. C.; Ohta, A. Adenosine A2A receptor antagonists: blockade of adenosinergic effects and T regulatory cells. Br. J. Pharmacol. 2008, 153, S457-464. 62. Ohta, A.; Sitkovsky, M. Role of G-protein-coupled adenosine receptors in downregulation of inflammation and protection from tissue damage. Nature 2001, 414, 916-920. 63. Ohta, A.; Gorelik, E.; Prasad, S. J.; Ronchese, F.; Lukashev, D.; Wong, M. K.; Huang, X.; Caldwell, S.; Liu, K.; Smith, P.; Chen, J. F.; Jackson, E. K.; Apasov, S.; Abrams, S.; Sitkovsky, M. A2A adenosine receptor protects tumors from antitumor T cells. Proc. Natl. Acad. Sci. 2006, 103, 13132-13137. 64. Lukashev, D.; Sitkovsky, M.; Ohta, A. From "Hellstrom Paradox" to anti-adenosinergic cancer immunotherapy. Purinergic Signal. 2007, 3, 129-134. 65. Lukashev, D.; Ohta, A.; Sitkovsky, M. Hypoxia-dependent anti-inflammatory pathways in protection of cancerous tissues. Cancer Metastasis Rev. 2007, 26, 273-279. 66. Chouker, A.; Thiel, M.; Lukashev, D.; Ward, J. M.; Kaufmann, I.; Apasov, S.; Sitkovsky, M. V.; Ohta, A. Critical role of hypoxia and A2A adenosine receptors in liver tissue-protecting physiological anti-inflammatory pathway. Mol. Med. 2008, 14, 116-123. 67. Lukashev, D.; Ohta, A.; Apasov, S.; Chen, J. F.; Sitkovsky, M. Cutting edge: Physiologic attenuation of proinflammatory transcription by the Gs protein-coupled A2A adenosine receptor in vivo. J. Immunol. 2004, 173, 21-24. 68. Lukashev, D.; Ohta, A.; Sitkovsky, M. Targeting hypoxia--A(2A) adenosine receptor-mediated mechanisms of tissue protection. Drug Discov. Today 2004, 9, 403-409. 69. Vaupel, P.; Mayer, A. Hypoxia in cancer: significance and impact on clinical outcome. Cancer Metast. Rev. 2007, 26, 225-239. 70. Vaupel, P.; Harrison, L. Tumor hypoxia: Causative factors, compensatory mechanisms, and cellular response. Oncologist 2004, 9, 4-9. 71. Carmeliet, P.; Jain, R. K. Principles and mechanisms of vessel normalization for cancer and other angiogenic diseases. Nat. Rev. Drug Discov. 2011, 10, 417-427. 72. Wager, T. T.; Hou, X.; Verhoest, P. R.; Villalobos, A. Moving beyond rules: the development of a central nervous system multiparameter optimization (CNS MPO) approach to enable alignment of druglike properties. ACS Chem. Neurosci. 2010, 1, 435-49. 73. Bandyopadhyay, P.; Agrawal, S. K.; Sathe, M.; Sharma, P.; Kaushik, M. P. A facile and rapid one-step synthesis of 8-substituted xanthine derivatives via tandem ring closure at room temperature. Tetrahedron 2012, 68, 3822-3827. 74. Kuo, S. C.; Tran, L. T.; Zhang, P. Process for preparing substituted 5-amino-pyrazolo-[4,3-e]- 1,2,4-triazolo[1,5-c]pyrimidines. WO2005054245 A3, 2005. 75. Neustadt, B. R.; Lindo, N. A.; Greenlee, W. J.; Tulshian, D.; Silverman, L. S.; Xia, Y.; Boyle, C. D.; Chackalamannil, S. Adenosine a2a receptor antagonists. Adenosine a2a receptor antagonists, WO2001092264 A1, 2001.

291

76. Marom, E.; Mizhiritskii, M.; Rubnov, S. Process and intermediates for the preparation of preladenant and related compounds. Process and intermediates for the preparation of preladenant and related compounds. WO 2012127472 A1, 2012. 77. Marom, E.; Mizhiritskii, M.; Rubnov, S. Process and intermediates for the preparation of preladenant and related compounds. WO2012127472 A1, 2012. 78. Baraldi, P. G.; Bovero, A.; Fruttarolo, F.; Romagnoli, R.; Tabrizi, M. A.; Preti, D.; Varani, K.; Borea, P. A.; Moorman, A. R. New strategies for the synthesis of A(3) adenosine receptor antagonists. Bioorg. Med. Chem. 2003, 11, 4161-4169. 79. Shawali, A. S.; Hassaneen, H. M.; Shurrab, N. K. A new strategy for the synthesis of pyrazolo[4,3-e][1,2,4]triazolo[4,3-c]pyrimidines and pyrazolo[4,3-e][1,2,4]triazolo[1,5-c]pyrimidines. Tetrahedron 2008, 64, 10339-10343. 80. Rashad, A. E.; Heikal, O. A.; El-Nezhawy, A. O. H.; Abdel-Megeid, F. M. E. Synthesis and isomerization of thienotriazolopyrimidine and thienotetrazolopyrimidine derivatives with potential anti- inflammatory activity. Heteroatom Chem. 2005, 16, 226-234. 81. Abdel-Fattah, B.; Kandeel, M. M.; Abdel-Hakeem, M.; Fahmy, Z. M. Synthesis of certain fused thienopyrimidines of biological interest. J. Chin. Chem. Soc-Taip. 2006, 53, 403-412. 82. Fischer, G. 1,2,4-Triazolo[1,5-Alpha]Pyrimidines. Adv Heterocycl Chem 1993, 57, 81-138. 83. Halgren, T. A.; Murphy, R. B.; Friesner, R. A.; Beard, H. S.; Frye, L. L.; Pollard, W. T.; Banks, J. L. Glide: a new approach for rapid, accurate docking and scoring. 2. Enrichment factors in database screening. J. Med. Chem. 2004, 47, 1750-1759. 84. Cui, M. C.; Ono, M.; Kimura, H.; Ueda, M.; Nakamoto, Y.; Togashi, K.; Okamoto, Y.; Ihara, M.; Takahashi, R.; Liu, B. L.; Saji, H. Novel F-18-Labeled Benzoxazole Derivatives as Potential Positron Emission Tomography Probes for Imaging of Cerebral beta-Amyloid Plaques in Alzheimer's Disease. J. Med. Chem. 2012, 55, 9136-9145. 85. Coop, A.; Rice, K. C. Methods of O-demethylation and N-deprotection. US6395900 B1, 2002. 86. Vala, C.; Morley, T. J.; Zhang, X.; Papin, C.; Tavares, A. A.; Lee, H. S.; Constantinescu, C.; Barret, O.; Carroll, V. M.; Baldwin, R. M.; Tamagnan, G. D.; Alagille, D. Synthesis and in vivo Evaluation of Fluorine-18 and Iodine-123 Pyrazolo[4,3-e]-1,2,4-triazolo[1,5-c]pyrimidine Derivatives as PET and SPECT Radiotracers for Mapping A2A Receptors. Chem. Med. Chem. 2016, 11, 1936-1943 87. Barret, O.; Hannestad, J.; Vala, C.; Alagille, D.; Tavares, A.; Laruelle, M.; Jennings, D.; Marek, K.; Russell, D.; Seibyl, J.; Tamagnan, G. Characterization in humans of 18F-MNI-444, a PET radiotracer for brain adenosine 2A receptors. J. Nucl. Med. 2015, 56, 586-591. 88. Flohr, A.; Moreau, J. L.; Poli, S.; Riemer, C.; Steward, L. 4-Hydroxy-4-methyl-piperidine-1- carboxylic acid (4-methoxy-7-morpholin-4-yl-benzothiazol-2-yl)-amide. US20050261289 A1, 2005. 89. Shivakumar, D.; Williams, J.; Wu, Y.; Damm, W.; Shelley, J.; Sherman, W. Prediction of Absolute Solvation Free Energies using Molecular Dynamics Free Energy Perturbation and the OPLS Force Field. J. Chem. Theory Comput. 2010, 6, 1509-1519. 90. Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.; Klicic, J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.; Perry, J. K.; Shaw, D. E.; Francis, P.; Shenkin, P. S. Glide: a new approach for rapid, accurate docking and scoring. 1. Method and assessment of docking accuracy. J. Med. Chem. 2004, 47, 1739-1749. 91. Chiang, P. C.; Hu, Y. D. Simultaneous Determination of LogD, LogP, and pK(a) of Drugs by Using a Reverse Phase HPLC Coupled with a 96-Well Plate Auto Injector. Comb. Chem. High T. Scr. 2009, 12, 250-257.

292

Chapter 6: Conclusions and Future Directions

As a non-invasive imaging technology, PET has been used to quantify the biochemical and pharmacological progress via selective radiolabeled molecular probes. Quantification of biological targets of interest by PET enables investigations of their functions under normal and disease conditions, assessment of their distribution in the brain and periphery, and target engagement for validation of promising drug candidates in clinical trials. The development of novel radiosynthetic methodologies has significantly broadended the scope of functional groups that could be subjected to radiolabeling, especially for fluorine-containing functionalities.

Chapter 2 describes a novel one-pot two-step radiosynthesis of [18F]2-11 relying on a recently developed spirocyclic hypervalent iodine(III) species as a radiolabeling precursor and subsequent copper(I) iodide-mediated cross coupling with compound 2-30. The resulting [18F]2-

11 features a satisfactory RCY, excellent chemical & radiochemical purities as well as a reasonably high specific activity. The brain permeability and biodistribution of [18F]2-11 were further determined via a preliminary PET imaging studies of [18F]2-11 in mice, which demonstrates a promising profile, indicating its potential to translate to higher species as a

AMPA PET imaging tracer. Similarly, the para-analog [18F]2-36 has been radiosynthesized following the same reaction conditions as that of [18F]2-11. The performance of this analogue as

AMPA PET radiotracer needs to be further evaluated in vivo in the near future.

Chapter 3 reports a novel transition-metal-free oxidative benzylic C-H activation

18 methodology for the radiosynthesis of [ F]aryl-CF2H functionality. This methodology features moderate to good radiochemical yields with a diverse range of substrates. Noteworthy, this method gives a superior specific activity compared with those reported in literature. Further

293 ongoing research resides in extending this method to electron-rich substituted (hetero)arenes and applying it to the radiosynthesis of more complex molecules or drugs.

Molecular modeling could be used to investigate the non-covalent and covalent binding behaviors of the newly designed compounds in the biological targets at molecular level and guide the following drug design. Chapter 4 illustrates the success of utilizing molecular modeling technique in PET radiotracer design targeting FAAH and MGL enzymes, which are involved in the endocannabinoid system. The investigation of BIA 10-2474 through molecular modeling reveals its low binding affinity and poor selectivity as FAAH inhibitor, which may be attributed to its severe side effects in phase I clinical trials. Molecular modeling is also employed to assist the design of novel reversible MGL inhibitors based on the core structure of PAD, among which

[18F]FEPAD proves to be a useful PET tracer for imaging MGL in the peripheral system. On the other hand, a serie of irreversible MGL inhibitors was also designed based on lead compound 4-

1, most of which have been proved to be potent MGL inhibitors. Further evaluations of these 4-

1-based analogs are being carried out in nonhuman primate models.

In Chapter 5, the design and synthesis of series of PEGylated analogs, based on the core structures of preladenant and tozadenant, have been described to yield potential cancer immunotherapeutics. The successful preladenant analogue 5-34 bears a fluorinated tri-ethylene glycol group, which gives promising results in the cAMP and IFN-gamma functional immunoassays. Future work must focus on studies of the detailed mechanistic mode of action of

5-34 and investigation of its feasibility as a potential cancer immunotherapeutic in vivo.

294

Apendix

Chapter 3: 1H NMR, 13C NMR and 19F NMR spectra of new compound 3-2e:

295

296

297

298

299

300

301 302 303 304 305 306 307 308 309 310