I Chemical Biology Approaches to Combat Parkinson's Disease By

Chemical Biology Approaches to Combat Parkinson’s Disease

Felix O. Nwogbo Jr

Department of Chemistry Duke University

Date: ______Approved:

______Dewey G. McCafferty, Supervisor

______Jennifer L. Roizen

______Jiyong Hong

______David M. Gooden

Dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Chemistry in the Graduate School of Duke University

2018

ABSTRACT

Chemical Biology Approaches to Combat Parkinson’s Disease

Felix O. Nwogbo Jr

Department of Chemistry Duke University

Date: ______Approved:

______Dewey G. McCafferty, Supervisor

______Jennifer L. Roizen

______Jiyong Hong

______David M. Gooden

An abstract of a dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Chemistry in the Graduate School of Duke University

2018

Abstract

Parkinson's disease (PD) is a debilitating neurodegenerative disease of the central nervous system characterized by loss of striatal dopaminergic projections from the substantia nigra. Although there is no cure for PD, dopamine (DA) replacement using

L-3,4-dihydroxyphenylalanine (L-DOPA) is the most common therapy used to manage

PD motor symptoms. L-DOPA is poorly absorbed into the brain and metabolized in the periphery causing its efficacy to wane over time. Additionally, within five years of use,

L-DOPA can induce its own severe motor dysfunction, such as dyskinesias, which can be irreversible. This underscores the need for the discovery and development of improved anti-Parkinsonian therapeutics. We have identified a class of conformationally-constrained phenylethylamines based on a tranylcypromine scaffold and demonstrated that many compounds in this structural class exhibited partial or full relief of akinesia in a DA-deficient/DA transporter knockout (DAT-KO) mouse model of

PD developed in the Caron laboratory. Two active arylcyclopropylamines from studies in DAT-KO mice were subsequently evaluated in a 6-hydroxydopamine-lesioned rat model to confirm their anti-Parkinsonian and anti-dyskinesia activities. In these rats, both compounds improved lesioned-induced motor deficits that mimic PD-associated akinesia. Target identification and activity assays suggest 5-HT2B and a2A as candidate targets to begin elucidation of novel non-dopaminergic pathways to combat PD.

Dedication

Simply and sincerely, I dedicate this dissertation to my family and friends who supported my journey to this point and undoubtedly will continue to do so for years to come. Your love and unwavering support have helped me through one of the most challenging experiences I have had—this would not be possible without each and every one of you.

Contents

Abstract ...... iv

List of Tables ...... x

List of Figures ...... xi

List of Schemes ...... xvi

List of Abbreviations ...... xvii

Acknowledgements ...... xxiv

1. Introduction ...... 1

1.1 Hallmarks of Parkinson’s Disease ...... 1

1.2 Dopamine Synthesis and Neurotransmission ...... 2

1.3 Nigrostriatal Pathway Responsible for Locomotion ...... 2

1.4 Loss of Striatal Dopaminergic Innervation Leads to Parkinson’s Disease ...... 4

1.5 Treatments for Parkinson’s Disease Symptoms are Limited and Problematic ...... 6

2. Chemical Genetic Mouse Model for Parkinson’s Disease ...... 11

2.1 Phenotypic-Screening of Compounds in a Parkinson’s Disease Mouse Model .... 11

2.2 Amphetamine Drug Class ...... 17

2.3 Aryl-Substituted (±)-trans-Arylcyclopropylamines Derivatives Alleviate Akinesia in Mouse Model ...... 21

2.4 Monoamine B Inhibition is not Responsible for Mechanism of Action ...... 29

2.6 Hypothesis ...... 33

3. Syntheses ...... 34

3.1 Original Synthesis of Arylcyclopropylamine-Derivative, Tranylcypromine ...... 34

3.2 Synthetic Strategies ...... 35

3.3 Preparative-Scale Synthesis ...... 35

3.3.1 Alternative Rearrangement Reactions ...... 41

3.3.2 Cross-Coupling Reactions ...... 43

3.4 Additional Chemical Probes of Interest ...... 45

3.4.1 Phenethylamines derivatives ...... 46

3.4.2 MDMA-Derivates, UWA-121 & UWA-122 ...... 46

3.5 Experimental ...... 48

3.5.1 ACPA derivatives ...... 49

3.5.2 UWA derivatives ...... 64

4.1 Unilateral 6-Hydroxydopamine-Lesion Rat ...... 66

4.2 Design and Validation of Animal Model ...... 67

4.3 Acute-Administration of ACPAs Accessed by Forelimb Adjusted Step Task ...... 72

4.4 Conclusions ...... 74

4.5 Experimental ...... 76

5. G-Protein Coupled Receptors ...... 80

5.1 Survey of the GPCRome ...... 81

5.2 G-Protein Coupled Receptor’s Structural Components ...... 82

5.3 G-Protein Coupled Receptor Biological Functions ...... 83

5.3.1 G-Protein Dependent Signaling ...... 83

5.3.2 G-Protein Independent Signaling ...... 86

5.4 G-Protein Coupled Receptor Assays ...... 88

vii

5.4.1 Affinity ...... 88

5.4.2 Efficacy ...... 89

6. Mechanistic Affinity Studies ...... 90

6.1 Experimental Design ...... 91

6.2 Library of Compounds Assessed ...... 92

6.2.1 Dopaminergic Targets ...... 93

6.2.2 Serotonergic Targets ...... 96

6.2.3. Adrenergic Targets ...... 102

6.2.4. Histaminergic Targets ...... 105

6.2.5 Non-Monoamine Targets ...... 106

6.3 Pharmacophoric Considerations ...... 109

6.3.1 Additional ACPAs Evaluated ...... 109

6.3.2 Phenethylamines Evaluated ...... 111

6.4 Conclusions ...... 114

6.5 Experimental ...... 117

7. Mechanistic Efficacy Studies ...... 119

7.1 G-Protein Activation Assays ...... 119

7.1.1 Calcium Flux Assay ...... 119

7.2 b-Arrestin Recruitment Assays ...... 124

7.2.2 Dopaminergic Targets ...... 131

7.2.2 Serotonergic Targets ...... 132

7.2.2 Adrenergic Targets ...... 132

viii

7.2.2 Non-Monoamine Targets ...... 133

7.3 Pharmacophoric Considerations ...... 134

7.4 Conclusions ...... 138

7.5 Experimental ...... 140

8. Conclusions ...... 142

8.1 Antagonism of Presynaptic Receptors Reinstates the Direct Pathway of Locomotion ...... 142

8.2 Future Directions ...... 144

8.2.1 Structure-Based Small Molecule Design ...... 144

8.2.2 Further Characterization of ACPA Biological Activity ...... 145

8.2.3 Animal Behavioral Studies ...... 146

Appendix A ...... 149

References ...... 177

List of Tables

Table 1. List of genetic and environmental factors known to lead to Parkinson’s disease. 5

Table 2. Select examples of effects of non-dopaminergic therapies in animal models...... 8

Table 3. Nondopaminergic drug treatments ineffective in restoring forward locomotion in DAT-KO mouse model for Parkinson’s disease...... 14

Table 4. Trace amines and amphetamines drug treatments ineffective in restoring forward locomotion in DAT-KO mouse model for Parkinson’s disease...... 15

Table 5. Summary of tranylcypromine in-vitro activity in relevant monoamine pathways. “≤” represents less than 2-fold difference, “>” and “<” represent about 2- to 100-fold difference...... 21

Table 6. Toxicity profile as a function of ACPA 4 or ACPA 9 at given concentrations in PD-like rat model...... 73

Table 7. GPCR profile for ACPA 66. Targets present in the all five ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*)...... 110

Table 8. GPCR profile for ACPA 67. Targets present in the all five ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*)...... 110

Table 9. GPCR profile for 42 (analog to ACPA 1). Targets present in the all five ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*)...... 111

Table 10. GPCR profile for 43 (analog to ACPA 2). Targets present in the all five ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*)...... 112

Table 11. GPCR profile for 44 (analog to ACPA 3). Targets present in the all five ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*)...... 112

Table 12. GPCR profile for 45 (analog to ACPA 4). Targets present in the all five ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*)...... 113

Table 13. GPCR profile for 46 (analog to ACPA 1). Targets present in the all five ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*) ...... 113

List of Figures

Figure 1. Nigrostriatal pathway (left) and coronal view of brain (right)...... 4

Figure 2. Visual schematic of a pre- and post-synaptic neuron depicting the DAT-KO mouse model for Parkinson’s disease...... 12

Figure 3. Chemical structures of ineffective nondopaminergic drug treatments...... 16

Figure 4. Chemical structures of trace amines and amphetamines. These compounds were ineffective at restoring full locomotion as well...... 17

Figure 5. Structural similarity between dopamine, b-phenethylamine, (±)-amphetamine. Each compound possesses an aryl ring separated from a primary amine by an ethylene bridge...... 18

Figure 6. Arylcyclopropylamine scaffold present in FDA approved drugs and drugs in clinical trials...... 19

Figure 7. Tranylcypromine enantiomers...... 19

Figure 8. Aryl-substituted arylcyclopropylamine derivatives...... 22

Figure 9. Results from unbiased screening of compounds 1 and 2 in DAT-KO mice (n = 4 for each)...... 23

Figure 10. Results from unbiased screening of compounds 3 and 5 in DAT-KO mice (n = 4 for each)...... 24

Figure 11. Results from unbiased screening of compounds 5 and 6 in DAT-KO mice (n = 4 for each)...... 25

Figure 12. Results from unbiased screening of compounds 7 and 8 in DAT-KO mice (n = 4 for each)...... 26

Figure 13. Results from unbiased screening of compounds 9 and 10 in DAT-KO mice (n = 4 for each)...... 27

Figure 14. Results from unbiased screening of compounds 11 in DAT-KO mice (n = 4 for each)...... 28

Figure 15. Results from unbiased screening of compounds 12–15 in DAT-KO mice (n = 4 for each)...... 30

Figure 16. Dose-dependent response of compounds on horizontal activity...... 32

Figure 17. Analogous PEA derivatives to the ACPA derivatives ...... 46

Figure 18. Behavioral observation of dyskinesias...... 70

Figure 19. Behavioral observation of dyskinesias (cont.) A: limb dyskinesia – uncontrolled limb movement; B. Forelimb adjustment step task...... 71

Figure 20. 6-OHDA rats assessed in forelimb adjustment step task post-administration of ACPA compound at indicated doses...... 73

Figure 21. Condensed summary a few of the multiple pathways GPCRs can transduce. 85

Figure 22. Visual representation depicting experimental design of radioligand binding competition assay...... 90

Figure 23. ACPAs active in the DAT-KO mice, reducing rigidity and promoting movement...... 93

Figure 24. Secondary binding data showing the amount of [3H]-SCH23390 (Kd = 0.89 ± 0.12 nM) bound to D1 as a function of the competing, investigational ligand...... 95

Figure 25. Secondary binding data showing the amount of [3H]-WIN35428 (Kd = 3.04 ± 0.21 nM) bound to DAT as a function of the competing, investigational ligand...... 95

Figure 26. Secondary binding data showing the amount of [3H]-WAY100635 (Kd = 0.43 ± 0.06 nM) bound to 5-HT1A as a function of the competing, investigational ligand...... 97

Figure 27. Secondary binding data showing the amount of [3H]5-CT (Kd = 1.91 ± 0.29 nM) bound to 5-HT1B as a function of the competing, investigational ligand...... 97

Figure 28. Secondary binding data showing the amount of [3H]5-CT (Kd = 1.47 ± 0.24 nM) bound to 5-HT1D as a function of the competing, investigational ligand...... 98

Figure 29. Secondary binding data showing the amount of [3H]5-HT (Kd = 3.36 ± 0.46 nM) bound to 5-HT1E as a function of the competing, investigational ligand...... 98

xii

Figure 30. Secondary binding data showing the amount of [3H]-Ketanserin (Kd = 1.57 ± 0.26 nM) bound to 5-HT2A as a function of the competing, investigational ligand...... 99

Figure 31. Secondary binding data showing the amount of [3H]-LSD (Kd = 1.04 ± 0.15 nM) bound to 5-HT2B as a function of the competing, investigational ligand...... 99

Figure 32. Secondary binding data showing the amount of [3H]-LSD (Kd = 2.92 ± 0.83 nM) bound to 5-HT2C as a function of the competing, investigational ligand...... 100

Figure 33. Secondary binding data showing the amount of [3H]-LSD (Kd = 3.97 ± 0.46 nM) bound to 5-HT6 as a function of the competing, investigational ligand...... 100

Figure 34. Secondary binding data showing the amount of [3H]-LSD (Kd = 5.30 ± 0.70 nM) bound to 5-HT7 as a function of the competing, investigational ligand...... 101

Figure 35. Secondary binding data showing the amount of [3H]-Citalopram (Kd = 3.8 ± 2.1 nM) bound to SERT as a function of the competing, investigational ligand...... 101

Figure 36. Secondary binding data showing the amount of [3H]-Prazosin (Kd = 0.89 ± 0.14 nM) bound to a1D as a function of the competing, investigational ligand...... 102

Figure 37. Secondary binding data showing the amount of [3H]-Rauwolscine (Kd = 0.5– 1.0 nM) bound to a2A as a function of the competing, investigational ligand...... 103

Figure 38. Secondary binding data showing the amount of [3H]-Rauwolscine (Kd = 1.36 ± 0.50 nM) bound to a2B as a function of the competing, investigational ligand...... 103

Figure 39. Secondary binding data showing the amount of [3H]-Rauwolscine (Kd = 0.73 ± 0.24 nM) bound to a2C as a function of the competing, investigational ligand...... 104

Figure 40. Secondary binding data showing the amount of [3H]-Nisoxetine (Kd = 4.9 ± 0.7 nM) bound to NET as a function of the competing, investigational ligand...... 104

Figure 41. Secondary binding data showing the amount of [3H]- Pyrilamine (Kd = 1.01 ± 0.13 nM) bound to H1 as a function of the competing, investigational ligand ...... 105

Figure 42. Secondary binding data showing the amount of [3H]-QNB (Kd = 0.6 ± 0.13 nM) bound to M5 as a function of the competing, investigational ligand...... 106

Figure 43. Secondary binding data showing the amount of [3H]-U69593 (Kd = 1.93 ± 0.45 nM) bound to KOR as a function of the competing, investigational ligand ...... 107

xiii

Figure 44. Secondary binding data showing the amount of [3H]-Pentazocine (Kd = 4.45 ± 0.55 nM) bound to s1 as a function of the competing, investigational ligand...... 108

Figure 45. Secondary binding data showing the amount of [3H]-DTG (Kd = 10.08 ± 1.38 nM) bound to s2 as a function of the competing, investigational ligand...... 108

Figure 46. trans-(N)-Alkylated arylcyclopropylamines 66 and 67...... 109

Figure 47. Phenethylamine derivatives assessed in addition to the ACPAs...... 111

Figure 48. Visual representation depicting experimental design and the reaction which Flur-4 AM is metabolized...... 121

Figure 49. Primary agonism data. Relative fluorescence units (RFU) is measures as a function of 5-HT2B activation by a given ACPA in stably-transfected CHO cells...... 123

Figure 50. Primary agonism data. Relative fluorescence units (RFU) is measures as a function of 5-HT2B activation by a given ACPA...... 123

Figure 51. Visual representation depicting experimental design and the reaction which luciferase catalyzes...... 125

Figure 52. Receptors profile for ACPA 1...... 127

Figure 53. Receptors profile for ACPA 2...... 128

Figure 54. Receptors profile for ACPA 3...... 129

Figure 55. Receptors profile for ACPA 4...... 130

Figure 56. Receptors profile for ACPA 9...... 131

Figure 57. Receptors profile for PEA 42...... 134

Figure 58. Receptors profile for PEA 43...... 135

Figure 59. Receptors profile for PEA 44...... 136

Figure 60. Receptors profile for PEA 45...... 137

Figure 61. Receptors profile for PEA 46...... 138

xiv

Figure 62. Visual representation of proposed mechanism of action of ACPAs in the DAT- KO mouse and 6-OHDA rat models for Parkinson’s disease...... 143

Figure 63. Crystal structure of the LSD-bound 5-HT2B receptor. PDB: 5TVN...... 145

List of Schemes

Scheme 1. Biosynthesis of dopamine in neurons within the brain...... 2

Scheme 2. Original synthesis of tranylcypromine performed by Burger and Yost in 1948...... 34

Scheme 3. Previous synthesis to access arylcyclopropylamine scaffold...... 36

Scheme 4. Modified synthetic route to access the arylcyclopropylamine scaffold...... 37

Scheme 5. Horner-Wadsworth-Emmons Olefinaton of commerically available aldehyes to yield the corresponding a,b-unsaturated esters...... 38

Scheme 6. One-pot Corey-Chaykovsky cyclopropanation and hydrolysis...... 39

Scheme 7. Curtius rearrangement of (3,4)-methylenedioxy- derivative-intermediate, carboxylic acid 30 to yield the corresponding carbamate 31 (Boc-protected amine)...... 41

Scheme 8. Curtius rearrangement of carboxylic acid to yield the corresponding carbamate (Boc-protected amine)...... 41

Scheme 9. Carboxylic acid 18 transformed into hydroxamic acid 32 for the subsequent Lossen rearrangement...... 42

Scheme 10. Mild-Curtius rearrangement of carboxylic acid 18 to the desired carbamate 34...... 42

Scheme 11. Suzuki-Miyaura method to access the cis- and trans-ACPA scaffold...... 43

Scheme 12. Attempt to access aminocyclopropylboronic ester precursor for Suzuki- Miyaura method...... 44

Scheme 13. Hiyama method access the cis- and trans-ACPA scaffold...... 45

Scheme 14. Attempt to access aminocyclopropyl organosilanes precursor for Hiyama method...... 45

Scheme 15. Synthesis of UWA enantiomers...... 47

xvi

List of Abbreviations

{Ir(OMe)(cod)}2 (1,5-cyclooctadiene) (methoxy) iridium (I) dimer 13C NMR carbon 13 nuclear magnetic resonance 1H NMR proton nuclear magnetic resonance

3,4,7,8-Me4phen 3,4,7,8-tetramethyl-1,10-phenanthroline 3H tritium 5-HT serotonin 6-OHDA 6-hydroxdopamine Å angstrom AACC aromatic L-amino acid decarboxylase AC adenylyl cyclase; adenylate cyclase ACh acetylcholine AcOH acetic acid ACPA arylcyclopropylamine AD Alzheimer disease

Ag2O silver oxide AIMs abnormal involuntary movement scale AMPH amphetamine ANOVA analysis of variance aq aqueous Ar-I aryl iodide ATP adenosine triphosphate BG basal ganglia Bn benzyl Boc tert-butyloxycarbonyl

Boc2O Boc anhydride Br bromo br broad peak BRET bioluminescence resonance energy transfer Ca2+ calcium ion

Ca3(PO)2 calcium phosphate Calcd calculated CAM cerium ammonium molybdate

xvii

cAMP cyclic adenosine monophosphate

CD3OD deuterated methanol

CDCl3 deuterated chloroform CDI carbonyldiimidazole

CF3 trifluoromethyl

CH2N2 diazomethane CNS central nervous system

CO2 carbon dioxide COMT catechol-O-methyltransferase cpm counts per minute

CsCO3 cesium carbonate Cu copper CuCN•2LiCl copper (I) cyanide di(lithium chloride) complex d doublet DA dopamine DAG diacylglycerol DAT dopamine transporter DCM dichloromethane dFBS dialyzed fetal bovine serum DMEM Dulbecco's Modified Eagle's Medium DMF dimethylformamide DMSO dimethyl sulfoxide DMSO-d6 deuterated dimethyl sulfoxide DNA deoxyribonucleic acid DOR delta opioidergic receptor DPPA diphenylphosphoryl azide

EC50 half maximal effective concentration eq equivalence

Et2O diethyl ether EtO ethoxy EtOAc ethyl acetate FBS fetal bovine serum FRET fluorescence resonance energy transfer G418 geneticin

xviii

GAP GTPase-accelerating protein GDP guanosine diphosphate GEF guanine nucleotide exchange factor GIRK G-protein coupled inwardly-rectifying potassium channel Glu glutamate GPCR G-protein coupled receptor GPe global pallidus external GPi global pallidus internal GRK G-protein regulatory kinase GTP guanosine triphosphate H hydrogen atom

H2O water HBpin Pinacolborane or 4,4,5,5-tetramethyl-1,3,2-dioxaborolane HBr hydrobromic acid HDO deuterated water HEK293 human embryonic kidney 293 Hex hexane HRMS high-resolution mass spectrometry hv light i.p. intraperitoneal IBCF isobutyl chloroformate

IC50 half maximal inhibition constant iCa2+ intracellular calcium ion IL intracellular helical (domain) Ins3PR inositol trisphosphate receptor

IP3 inositol trisphosphate iPrO isoproxy J coupling constant

K2CO3 potassium carbonate

Kd dissociation constant

KHSO4 potassium bisulfate

Ki inhibition constant KO knock-out koff rate of dissociation

xix

kon rate of association KOR kappa opioidergic receptor L-DOPA levodopa; L-3,4-dihydroxyphenylalanine

LD50 half maximal lethal dose LiCl lithium chloride LID levodopa-induced dyskinesia LiHMDS lithium hexamethyldisilazide LiOH lithium hydroxide LSD1 lysine-specific demethylase 1 M molar m multiplet m/z mass-to-charge ratio mAChR muscarinic acetylcholine receptor MAO monoamine oxidase MAPK mitogen-activated protein kinases MDMA 3,4-methylenedioxymethamphetamine; ecstasy Me methyl

Me3S(O)I trimethylsulfoxonium iodide MeCN acetonitrile

MeNH2 methylamine MeO methoxy MeOH methanol METH methamphetamine Mg magnesium

Mg2+ magnesium ion mmol micromole mol% mole percent MOM methyoxymethyl mRNA messenger ribonucleic acid

Na2SO4 sodium sulfate nAChR nicotinic acetylcholine receptor NaCl sodium chloride

NaCNBH3 sodium cyanoborohydride NaH sodium hydride

NaHCO3 sodium bicarbonate

NaN3 sodium azide NaOH sodium hydroxide nBu4NCl tetrabutylammonium chloride ND neurodegeneration NE norepinephrine NET norepinephrine transporter NF-�B nuclear factor kappa-light-chain-enhancer of activated B cells

NH4Cl ammonium chloride NMR nuclear magnetic resonance p p-value p53 cellular tumor antigen p53 PD Parkinson's disease Pd palladium

Pd(OAC)2 palladium acetate

PdCl2(dppf) [1,1'-bis(diphenylphosphino)ferrocene] palladium (II) dichloride PG pargyline

PhCH3 toluene PhO phenoxy

PIP2 phosphatidylinositol 4,5-bisphosphate Piv pivaloyl PKA protein kinase A PKC protein kinase C PLL poly-L-lysine Post-L post-lesion

PPi pyrophosphate ppm parts per million Pre-L pre-lesion q quartet R substituent

Rf retention factor RFU relative fluorescence units RGS regulatory of G-protein signaling ROS reactive oxidative species

xxi

rt room temperature s singlet S1P1 sphingosine-1-phosphate receptor 1 SDRI serotonin-dopamine reuptake inhibitor SEM standard error or the mean sep septuplet SERT serotonin transporter SN substania nigra SNc substania nigra par compacta STN subthalamic nucleus t triplet TBAF tetra-n-butylammonium fluoride TBDMS tert-butyldimethylsilyl TBDMSCl tert-butyldimethylsilyl chloride tBuOH tert-butanol TCP tranylcypromine; Parnate®

TD50 half maximal toxic dose TEA triethylamine TEV tobacco etch virus TH tyrosine hydroxylase THB tetrahydrobiopterin THF tetrahydrofuran TLC thin-layer chromatography TM transmembrane domain

TMSCHN2 trimethylsilyldiazomethane tTA tetracycline transactivator tx treatment UV ultraviolent Zn zinc

Zn(OTf)2 zinc triflate

ZnCl2 zinc chloride �MPT alpha-methyl-para-tyrosine �-arr beta-arrestin �-PEA beta-phenylethylamine; phenylethylamine; phenethylamine

xxii

� shift (ppm) � kappa � mu � sigma � residence time

xxiii

Acknowledgements

I would like to thank my advisor, Professor Dewey McCafferty, for his mentorship and unwavering support over the past five years. I have been able to and grow and learn in so many ways while at Duke. Your compassion and love for your students is unconditional and I could not have imagined a better advisor or being able to successfully complete my graduate studies elsewhere. I would also like to thank my committee members, Professor Jennifer Roizen, Professor Jiyong Hing, and Dr. David

Gooden, as well as former members Professor Emily Derbyshire and Professor Emeritus

Alvin Crumbliss for their feedback and support. I am also thankful for Dr. Claire

Siburt’s immense support, mentorship, and guidance in just the short amount of time we worked together.

I would like to thank the Caron, Napier, and Roth lab members for their collaboration, insight, and resources; Dr. George Dubay for his assistance in mass spectrometry issues; the Duke NMR Facility; as well as Christiana Gooden, Meg Avery,

Young Im Chang, and the rest of the staff in the administrative office. I am eternally grateful to my lab mates, especially Drs. Katherine Alser Kim and Jennifer Link

Schwabe, and Meghan Lawler for their endless wisdom, compassion, and support. I would like to note Dr. Bumki Kim as well for welcoming me into the department when I first arrived to Duke.

xxiv

I also would like to thank my other academic mentors who have continued to show support over the course of my career: Professor Laundette Jones and Professor

Katherine Seley-Radtke.

Last and above all, I thank my parents and sisters for not only their love, but for being examples of strength and discipline. I whole-heartedly could not have done this without you all.

xxv

1. Introduction

Neurodegeneration describes the progressive destruction of structure and function within the central or peripheral nervous system. These diseases represent one of the largest group of disorders with more than 600 defined cases, effecting more than

50 million people worldwide [1, 2]. Within the United States (U.S.) alone, $100 billion is lost each year due to direct health costs and lost opportunities, and as the older population (60+ years of age) continues to grow the number of affected Americans and people worldwide will increase significantly [2]. Great advances have been made by way of on-going research powered by increased awareness and strong funding efforts

(Michael J. Fox Foundation), however treatments for these disorders remain limited and ineffective. This is especially evident in late disease states due to many of the mechanisms underlying pathogenesis remaining unknown.

1.1 Hallmarks of Parkinson’s Disease

In spite of the hundreds of known diseases under the umbrella of NDs, only five have garnered significant attention: (1) Alzheimer’s disease (AD); Parkinson’s disease

(PD); (3) Huntington’s disease; (4) amyotrophic lateral sclerosis; and (5) multiple sclerosis. Of these five disorders, AD and PD represent the most common cases of NDs diagnosed each year [2]. Parkinson’s disease (PD) is a progressive nervous system disorder marked by its effect on movement and stability. PD becomes clinically apparent

when the loss of dopaminergic neurons in the substantia nigra pars compacta (SNc) reaches 60–70%, resulting in a 50% loss of dopamine (DA) in the striatum [3].

1.2 Dopamine Synthesis and Neurotransmission

During normal metabolism, DA is produced from L-tyrosine, which is converted to levodopa (L-DOPA) by tyrosine hydroxylase (TH) and then to DA by aromatic L- amino acid decarboxylase (AADC). L-DOPA restores DA levels since it can be converted within the brain to DA via AADC (Scheme 1) [6].

Scheme 1. Biosynthesis of dopamine in neurons within the brain.

1.3 Nigrostriatal Pathway Responsible for Locomotion

Typically, body movement is determined by interconnected neurons known as the basal ganglia (BG). Neurons send electrochemical signals (impulses) to a central area of the brain called the striatum. Neurons within the striatum exchange impulses with neurons of the substantia nigra (SN), which are relayed back and forth from the spinal

cord and BG of the brain. These exchanges ensure movement is fluid and uninterrupted.

Impulses from the spinal cord are finally relayed to skeletal muscles for movement.

Two pathways work in tandem to produce voluntary movement. These pathways are called the direct and indirect pathway. In the healthy-state, the substania nigra is sending dopaminergic (dopamine) impulses to dopaminergic receptors, D1

(direct pathway) and D2 (indirect pathways), in the striatum. In the direct pathway, an excitatory impulse (glutamate) from the motor cortex is sent to the striatum. The striatum sends an inhibitory impulse (γ-aminobutyric acid; GABA) to the global pallidus internal (GPi), which ceases the inhibitory signal the GPi is sending to the thalamus. The thalamus can then send an excitatory signal to the motor cortex to initiate movement. The indirect pathway opposes this process. The striatum simultaneously sends an inhibitory signal to the global pallidus external (GPe), which ceases the inhibitory signal on the subthalamic nucleus, and activates the GPi. GPi then sends an inhibitory signal in the thalamus, diminishing the excitatory output of the thalamus to the motor cortex. This process works to eliminate unproductive movement, in other words movement that interferes with voluntary movement. This demonstrates the necessity and complexity of both pathways for locomotion.

Major motor-deficits manifest due to DA loss in the SN as signaling becomes disrupted. Dopaminergic neurons in the striatum are not bound and activated, causing parts of the BG to become under- or over-stimulated. Under-stimulation causes the

subthalamic nucleus (STN) to become overactive and over-inhibit the GPi which leads to rigidity (i.e. “not enough movement”). While over-stimulation over-inhibits the thalamus, decrease thalamus output and causing tremors (i.e. “too much movement”)

(Figure 1) [4].

Figure 1. Nigrostriatal pathway (left) and coronal view of brain (right).

1.4 Loss of Striatal Dopaminergic Innervation Leads to Parkinson’s Disease

DA deficiency greatly impacts motor and non-motor functions especially as the disease progress into later stages. Patients suffer from resting tremors, rigidity, bradykinesia/akinesia, gait disturbance and postural instability, as well as mood disorders such as depression, anxiety, and irritability, and cognitive changes. Symptoms

initially are slight and localized, often appearing as a tremor in one hand, however symptoms worsen as the disease reaches its advanced stages [3].

Though a direct cause of PD is still unknown, several factors have been identified in pathogenesis and progression, with age being the most dominant risk factor. PD is in fact an age-related disorder, with incidence studies highlighting 1% of patients are older than 60 years and 3% being at least 80 years old. Other factors are mainly described as either genetic or environmental (Table 1) [3].

Table 1. List of genetic and environmental factors known to lead to Parkinson’s disease.

Genetic risk factors Environmental risk factors • SNCA (a-synuclein) • Pesticide exposure • PRKN (Parkin) • History of head injury • LRRK2 (Leucine-rich repeat • Non-smoker kinase 2) • PINK1 (PINK1) • Caffeine-free diets • PARK7 (DJ-1) • ATP13A2 (ATPase type 13A2)

While most cases are idiopathic (random, no genetic basis), 5–10% of cases are

Mendelian with autosomal dominant or recessive inheritance. To date, 28 distinct chromosomal regions have been identified with only six causing monogenic forms of

PD. Loss-of-function mutations in parkin, DJ-1, and pink1 cause mitochondrial dysfunction and the accumulation of reactive oxidative species (ROS), which is a known contributor to apoptosis (cell death). Missense mutations in α-synuclein and LRRK2 may

also affect protein degradation pathways, leading to protein aggregation and accumulation of Lewy bodies causing cellular dysfunction. Lastly, mutations in ATPase type 13A2 causes an atypical form of PD with dementia, named Kufor–Rakeb syndrome.

Mitochondrial dysfunction and protein aggregation in dopaminergic neurons are also responsible for their premature degeneration. Exposure to environmental and neurotoxins can also cause mitochondrial functional impairment and the release of ROS, leading to similar to that of the genetic mutations. These scenarios ultimately lead to a blockade of post-synaptic signaling, which results in neurodegeneration [5].

1.5 Treatments for Parkinson’s Disease Symptoms are Limited and Problematic

Currently, there is no known cure for PD however symptoms can be mitigated with therapeutic interventions for a period of time. One of the current PD treatments is

DA replacement therapy by administration of the DA precursor, L-DOPA.

Unfortunately, efficacy of this treatment wanes over time as long-term treatment is associated with psychotic behaviors and fluctuations in motor performance.

Dopaminergic neurons continue to diminish over the course of the disease-state, which leaves DA unbound and free to diffuse to other parts of the brain causing off-target activation. Elevated doses are able to compensate for the dislocation of DA, yet inevitably fail to stabilize motor function and even leads to dyskinesias induced by L-

DOPA (levodopa-induced dyskinesias [LIDs]) [6]. LIDs develop due to 6

overaccumulation of DA at peak-dose, which overstimulate GABAergic (inhibitory neurons) medium spin neurons of the striatum. LIDs often develop within five years after the beginning of treatment [6].

DA agonists have been utilized to directly or indirectly affect DA function, though with limited success. Monoamine oxidase (MAO) inhibitors and catechol-O- methyl transferase (COMT) inhibitors have served as beneficial treatments for PD patients in some cases [7]. However, they are marred by toxic side effects and are only most effective at early stages of PD or when utilized as adjunct medications with L-

DOPA.

Patients may opt for deep-brain stimulation. This is a surgical procedure in which an electrode is inserted into the brain where the STN is located. An electrical pulse from the probe generates an electric field that inhibits proximal neurons from firing. This action reduces the excessive inhibition of the thalamus to promote movement [3]. While effective, this procedure is quite invasive and expensive.

Additionally, the procedure may not be effective and does not seize the progression of degeneration of dopaminergic neurons.

Earlier success of treatment was founded on the understanding that PD is a DA deficiency disorder. Treatment of PD symptoms with L-DOPA began soon after the late

George Cotzias, MD first described L-DOPA potential as a therapy for PD in 1969 [8, 9].

Decades later, L-DOPA has remained the main treatment option for combating motor

systems of PD’s further emphasizing the need for new therapeutic strategies. This understanding has since evolved to development of treatments which restore or replace

DA signaling [3, 8, 10]. Numerous non-dopaminergic neurotransmitter systems have been implicated in the mechanisms contributing to the motor features of PD. These systems include the glutamatergic, cholinergic, adrenergic, serotonergic, histaminergic, and adenosinergic systems [7, 11]. Studies in rodents and non-human primates have redefined motor symptoms of PD as the result of an imbalance of inhibitory and excitatory drives in the direct and indirect pathways [7, 11]. A table of non- dopaminergic therapies in animal models is shown below (Table 2).

Table 2. Select examples of effects of non-dopaminergic therapies in animal models.

Drug Class Subclass Effect in animal models Adenosinergic A2A locomotion, survival of DA neurons antagonists antagonists ¯ catalepsy, behavioral sensitization, striatal DA nerve terminal loss NMDA receptor NMDA stepping w/ contralateral paw, contralateral circling, DA Antagonists antagonists neuron survival, DA levels, efficacy of L-DOPA ¯ haloperidol-induced catalepsy, akinesia, parkinsonian symptoms, dyskinesias Metabotropic Group I survival of DA neurons, striatal DA glutamate receptor metabotropic ¯ akinesia, LIDs antagonists glutamate receptor antagonists Serotonergic Agonists 5-HT1A ¯ LIDs, activity of raphe-striatal neurons and extracellular agonists DA release 5-HT1B ¯ LIDs, activity of raphe-striatal neurons and extracellular agonists DA release Adrenergic receptor a2 receptor ¯ L-DOPA-induced hyperkinesia, expression of AIMs, antagonists antagonists haloperidol-induced catalepsy, LIDs

A2A receptors have been the most implemented targets for non-dopaminergic interventions for PD systems. Antagonism (blocking endogenous activation) of the A2A receptors causes the co-localization of A2A and D2 receptors in the striatum. This heterodimerized receptor utilizes adenosine as its endogenous agonist, with the membrane of neurons, couples G-protein alpha subunit, Gs, to increase intracellular cyclic adenosine monophosphate (cAMP) and enhances GABA release in the GPe. This sequence of signaling mitigates hyperactivity of the indirect striatopallidal pathway.

The non-selective NMDA receptor antagonist amantadine, at the dose of 300 mg per day, was shown to be effective in the treatment of dyskinesia by blocking GABA binding. Metabotropic glutamate receptor (mGluR) modulation has been shown to be a promising strategy to manage LID. Two mGluR5 antagonists, mavoglurant and dipraglurant, were previously identified as potential anti-dyskinetic agents during initial clinical trials, however further investigation is needed due to mavoglurant failing to demonstrate anti-dyskinetic efficacy in two recent clinical studies. While a2 receptors have been highly implicated as targets, the role of these receptors is still poorly understood. Adrenergic receptors are present within the striatum and presumably modulate GABA release which results in hyperactivity of the indirect pathway, leading to dyskinesia. a2A antagonists were shown to have anti-dyskinesia properties and improve PD treatment. Lastly, agonism of 5-HT1A and 5-HT1B resulted in near-

amelioration of LIDs, implicating these receptors as additional targets to treat PD motor symptoms [7, 11].

In the McCafferty group, we are particularly interested in Parkinson’s disease from a chemical biology perspective. Current treatments, for the symptoms which arise due to the progressive loss of dopaminergic neurons, encompasses a myriad of compounds from differing chemical features, each with unique activities and thus benefits. This fact illustrates the key principle in chemical biology; that is a chemical

(structure) perturbing a biological system and eliciting a net effect greater of lesser than the basal activity (function). Interestingly, no single drug or therapy currently can address the full manifestations of symptoms of Parkinson’s disease. The aforementioned limitations of existing therapeutic approaches underscore the need for the development of more effective anti-Parkinsonian drugs and thus serves as inspiration for this project.

2. Chemical Genetic Mouse Model for Parkinson’s Disease

Several animal models of DA deficiency based on pharmacologic, neurotoxic, or genetic approaches have been developed to understand basic pathological processes leading to PD [17-24]. In rodents however, the prolonged absence of DA is not sustainable [25-27] and animals with chronic severe DA depletion are not readily available. Nonhuman primates are often used as well, as they can mimic PD-related deficits with better accuracy, however are unreadily available as well, require extensive training, and are very expensive due to the additional maintenance costs of this higher- ordered animal model [28]. As such, these details highlight another challenge in identifying new therapies.

2.1 Phenotypic-Screening of Compounds in a Parkinson’s Disease Mouse Model

The Caron Group developed a mouse model lacking a functional dopamine transporter [29] that displays remarkable alterations in the compartmentalization of DA

[30-32]. These mice demonstrate absolute dependence on the remaining DA produced from ongoing biosynthesis. Pharmacologic blockade/inhibition of DA synthesis with a- methyl-p-tyrosine (aMPT) in DAT-KO mice provides an effective approach to eliminate

DA acutely [30, 31], essentially working as a chemical genetic switch (Figure 2) [33].

Figure 2. Visual schematic of a pre- and post-synaptic neuron depicting the DAT-KO mouse model for Parkinson’s disease. Tyrosine hydroxylase begins DA biosynthesis. DA is then stored in the synaptic vesicles. When needed, the vesicular monoamine transporter (VMAT) shuttles the DA into the synapse. DA diffuses across the synapse to the post-synaptic neuron and binds to dopaminergic receptors to propagate signaling. Without a dopamine transporter, DA cannot be recycled into the pre-synaptic neuron, therefor these mice are strictly reliant upon DA biosynthesis to maintain voluntary motor function.

Primary studies in the DAT-KO mouse model yielded interesting, though complexing results. The loss of DA signaling that creates the motor symptoms of PD occurs upstream of many nondopaminergic pathways, suggesting that activation or inhibition of some of these downstream neuronal circuits could potentially reverse the motor deficits independent of restoration of DA activity. The Caron group therefore tested several nondopaminergic compounds that potentially could reverse the consequences of severe DA deficiency in DAT-KO mice (see Table 3–4, with chemical structures shown in Figures 3–4). Many of these compounds have been found to be effective in restoring some aspects of movement control in one or another experimental

animal models of PD and/or in PD patients [21,26,27,48,49], however in DAT-KO mice none of the drugs were effective in restoring the major aspects of movement control required for forward locomotion (distance traveled). Over 40 neuroactive compounds were assessed in the DAT-KO mice, however only amphetamine derivatives were effective in reducing manifestations of akinesia and rigidity in DAT-KO mice as detected in the catalepsy, grasping, and akinesia tests. More surprisingly, (+)-MDMA was the only compound effective in restoring movement control sufficiently to induce forward locomotion. This study supports the potential of amphetamine derivatives as novel non- dopaminergic therapies, and additionally suggests the possibility of DA-independent pharmacology for treat motor symptoms.

Table 3. Nondopaminergic drug treatments ineffective in restoring forward locomotion in DAT-KO mouse model for Parkinson’s disease.

Drug Class Drug NMDA receptor antagonist Glutamate Receptor Antagonists (+)-MK-801

(+)-MK801 + 5HT1B agonist RU24, 969 (+)-MK801 + Clonidine Muscarinic drugs Atropine Clonidine + Atropine MAO Inhibitors Deprenyl Pargyline A2 Adenosine Receptor Antagonists 8-(3-Chlorostyryl)caffeine Caffeine Serotonin releasers 5-Methoxy-N,N-dimethyltryptamine 5-Methyl-N,N-dimethyltryptamine α-Ethyltryptamine

5-HT1B receptor Serotonergic drugs Agonist RU24, 969 Methysergide Fluoxetine Adrenergic drugs NE precursor Droxidopa Desipramine Phentolamine Clonidine

Table 4. Trace amines and amphetamines drug treatments ineffective in restoring forward locomotion in DAT-KO mouse model for Parkinson’s disease.

Drug Class Drug Trace amines p-Tyramine Deprenyl + p-Tyramine m-Tyramine Pargyline + m-Tyramine Octopamine Deprenyl + Octopamine Pargyline + L-Tyrosine Tryptamine Pargyline + Tryptamine b-PEA Deprenyl + β-PEA 4-Methoxytyramine Metanephrine Normetanephrine 3,4-Dihydroxyphenylacetic acid Homovanillic acid 3-methoxytyramine Amphetamines (+)-Amphetamine Methamphetamine Methamphetamine + Methysergide 4-Chloro-amphetamine Phentermine Phentermine + Methysergide (±)-MDE (+)-MDE (-)-MDE (±)-MDA (±)-6-OH-MDA (±)-MDMA (+)-MDMA

Figure 3. Chemical structures of ineffective nondopaminergic drug treatments.

Figure 4. Chemical structures of trace amines and amphetamines. These compounds were ineffective at restoring full locomotion as well.

2.2 Amphetamine Drug Class

Amphetamines are substituted phenylethylamine derivatives that are structurally similar to DA and the endogenous trace amine, b-phenethylamine (Figure

5). This drug class represents a well-known group of compounds that potently affect psychomotor functions. Amphetamines are known to interact with plasma membrane monoamine transporters, including the DAT, the norepinephrine (NE) transporter

(NET), and the serotonin (5-HT) transporter (SERT). This complex interaction results in transporter-dependent efflux of monoamines, such as DA, into extracellular space from intraneuronal storages. It is commonly believed that DAT-mediated efflux of DA is

primarily responsible for the psychostimulant and hyperlocomotor actions of these drugs [13].

Figure 5. Structural similarity between dopamine, b-phenethylamine, (±)-amphetamine. Each compound possesses an aryl ring separated from a primary amine by an ethylene bridge.

Chemically similar, arylcyclopropylamines (ACPAs) are a class of molecules that have shown diverse biological activity [14]. While structurally simple, these molecules wield great bioactivity due to its set of chemical properties which can have significant outcomes in biological systems. The aryl ring increases hydrophobicity and can also enhance binding via pi-pi interactions. The hydrophilic, primary amine enhances solubility in aqueous solutions, serving as both a hydrogen-bond acceptor and donator.

Lastly, the cyclopropyl ring bridges the opposing functional groups and impose rigidity, while also conferring two stereocenters. ACPAs are currently used as antidepressants, antiretrovirals, LSD1 (Lysine-Specific Demethylase 1) inhibitors, and antiplatelets

(Figure 6) [14]. These molecules are structurally related to amphetamines but lack deleterious effects and, in some cases, are already FDA approved as they are generally well tolerated.

Figure 6. Arylcyclopropylamine scaffold present in FDA approved drugs and drugs in clinical trials.

trans-2-phenylcyclopropan-1-amine, also as known as Parnate or tranylcypromine (TCP), is the simplest arylcyclopropylamine. TCP is an FDA-approved racemic drug used to treat depression [15]. TCP was originally synthesized in 1948 by

Burger and Yost as an analog to amphetamine and exists as two enantiomers; (1S,2R)-

(+)-TCP and (1R,2S)-(-)-TCP (Figure 7). Burger et al hoped the introduction of the alicyclic cyclopropyl ring would enhance pharmaceutical properties of the amphetamine class of compounds[16].

Figure 7. Tranylcypromine enantiomers.

Since then, TCP’s pharmacology has been mostly elucidated (Table 5). TCP exhibits high activity as a monoamine oxidase inhibitor (MAO), medium activity as a monoamine transporter inhibition, and low activity as monoamine releasing-agent. Both enantiomers are active as irreversible, nonselective MAO inhibitors at low therapeutic doses of 20 mg/kg. (±)-TCP is more potent against MAO-A (IC50 = 76 nM) than MAO-B

(IC50 = 90 nM), with (+)-TCP being ten times more potent then (-)-TCP.

As stated, (±)-TCP’s pharmacology includes modulating levels of monoamine transmitters and monoamine reuptake inhibition. MAO-A preferentially oxidizes serotonin and norepinephrine, while MAO-B oxidizes dopamine, phenethylamine, and tyramine. Levels of these monoamines (which result from MAO activity) are increased

3–10 fold, while levels of neurotransmitter metabolites are by decreased by 30%. At higher doses (40–60 mg/kg) (±)-TCP has additional activity via norepinephrine re-uptake inhibition. (-)-TCP exhibits much greater activity than (+)-TCP; significantly releasing dopamine, minimally releasing serotonin and norepinephrine.

TCP is unique in that it has a short pharmacokinetic half-life (t1/2) of 2 hours, but an extended pharmacodynamic half-life of one week due to its ability to irreversibly inhibit MAO. TCP functions as a suicide-inhibitor, thus enzyme repair requires de nevo biosynthesis of new enzyme. Additionally, cytochrome P450 (CYP) 2A6 is the only drug- metabolizing CYP-enzyme inhibited by TCP at therapeutic doses [15]. This limits concerns of delivering an effective, and relatively low dose.

Table 5. Summary of tranylcypromine in-vitro activity in relevant monoamine pathways. “≤” represents less than 2-fold difference, “>” and “<” represent about 2- to 100-fold difference.

IC50 or EC50 Relative activities �M ng/mL Monoamine oxidase inhibition: high activity MAO-A 0.076 10 (+)-TCP > (–)-TCP MAO-B 0.09 12 (+)-TCP > (–)-TCP Monoamine transporter inhibition: low to medium activity NET 1 130 (+)-TCP < (–)-TCP SERT, DAT 10 1300 Monoamine release: low activity Dopamine 10 1300 (+)-TCP ≤ (–)-TCP Serotonin, norepinephrine >10 >1300

Taken altogether, these characteristics presents TCP’s arylcyclopropylamine scaffold as an appealing chemical probe that can be utilized to investigate the contribution of non-dopaminergic neurotransmission to motor functions.

2.3 Aryl-Substituted (±)-trans-Arylcyclopropylamines Derivatives Alleviate Akinesia in Mouse Model

Prior to my affiliation with McCafferty group alumni, Drs. Julie Polluck and

David Gooden synthesized a focus-library of aryl-substituted ACPAs (Figure 8) to be assessed for activity in the DAT-KO mouse model [34]. In collaboration with the Caron group here at Duke University, horizontal activity of mice was measured over time as a

function of three doses of a given ACPA derivative. Mice were placed in a locomotor box for 30 minutes to acclimate before administration of aMPT. After which, increasing concentrations of an ACPA derivative was administered every hour up to three hours

(Figure 9–14).

Figure 8. Aryl-substituted arylcyclopropylamine derivatives.

2500

2000

1500

(counts) 1000

Horizontal Activity 500

0 0 30 60 90 120 150 180 210 240 270 Time (min)

2500

2000

1500

(counts) 1000

Horizontal Activity 500

0 0 30 60 90 120 150 180 210 240 270 Time (min)

Figure 9. Results from unbiased screening of compounds 1 and 2 in DAT-KO mice (n = 4 for each). Horizontal activity after treatment of mouse at 90 minutes with 10 mg/kg, 150 minutes with 30 mg/kg, and 210 minutes 60 mg/kg.

2500

2000

1500

1000 (Counts)

Horizontal Activity 500

0 0 30 60 90 120 150 180 210 240 270 Time (min)

2500

2000

1500

(counts) 1000

Horizontal Activity 500

0 0 30 60 90 120 150 180 210 240 270 Time (min)

Figure 10. Results from unbiased screening of compounds 3 and 5 in DAT-KO mice (n = 4 for each). Horizontal activity after treatment of mouse at 90 minutes with 10 mg/kg, 150 minutes with 30 mg/kg, and 210 minutes 60 mg/kg.

2500

2000

1500

(counts) 1000

Horizontal Activity 500

0 0 30 60 90 120 150 180 210 240 270 Time (min)

2500

2000

1500

(counts) 1000

Horizontal Activity 500

0 0 30 60 90 120 150 180 210 240 270 Time (min)

Figure 11. Results from unbiased screening of compounds 5 and 6 in DAT-KO mice (n = 4 for each). Horizontal activity after treatment of mouse at 90 minutes with 10 mg/kg, 150 minutes with 30 mg/kg, and 210 minutes 60 mg/kg.

4000

3000

2000 (counts)

1000 Horizontal Activity

0 0 30 60 90 120 150 180 210 240 270 Time (min)

2500

2000

1500

(counts) 1000

Horizontal Activity 500

0 0 30 60 90 120 150 180 210 240 270 Time (min)

Figure 12. Results from unbiased screening of compounds 7 and 8 in DAT-KO mice (n = 4 for each). Horizontal activity after treatment of mouse at 90 minutes with 10 mg/kg, 150 minutes with 30 mg/kg, and 210 minutes 60 mg/kg.

3000

2000 (counts) 1000 Horizontal Activity

0 0 30 60 90 120 150 180 210 240 270 TIme (min)

3000

2500

2000

1500 (counts) 1000

Horizontal Activity 500

0 0 30 60 90 120 150 180 210 240 270 Time (min)

Figure 13. Results from unbiased screening of compounds 9 and 10 in DAT-KO mice (n = 4 for each). Horizontal activity after treatment of mouse at 90 minutes with 10 mg/kg, 150 minutes with 30 mg/kg, and 210 minutes 60 mg/kg.

2500

2000

1500

(counts) 1000

Horizontal Activity 500

0 0 30 60 90 120 150 180 210 240 270 Time (min)

Figure 14. Results from unbiased screening of compounds 11 in DAT-KO mice (n = 4 for each). Horizontal activity after treatment of mouse at 90 minutes with 10 mg/kg, 150 minutes with 30 mg/kg, and 210 minutes 60 mg/kg.

Shortly after administration of the compounds, mice exhibited marked increase in motor behavior compared to levodopa and carbidopa1. Over half of these compounds reduced rigidity and promoted movement in a dose-dependent manner within 50% of initial locomotion prior to administration of aMPT at a dose of 60 mg/kg. While the mice did not exhibit normal locomotion, this is significant in comparison to 100 mg/kg L-

DOPA reported by the Caron group [35].

1Dr. Julie Polluck, McCafferty group in collaboration w/ the Caron group, unpublished data. 28

2.4 Monoamine B Inhibition is not Responsible for Mechanism of Action

MAO inhibitors comprise of some of the earliest drugs evaluated in PD [3]. Non- selective MAO inhibitors have been limited because of their side-effect and adverse reaction profile [12]. On the other hand, MAO-B selective inhibitors such as selegiline and rasagiline have seen success in treating akinesia and motor fluctuations [36]. In order to confirm that the promising results are indeed not a manifestation of MAO-B inhibition, pargyline (PG) derivatives (Figure 15), including pargyline (compound 12), were evaluated in the same studies with DAT-KO mice.

A. PG 12 (Pargyline) B. PG 13 2500 2500

2000 2000

1500 1500

1000 1000

500 500

Horizontal Activity (Counts) 0 Horizontal Activity (Counts) 0 0 90 180 270 0 90 180 270 TIme (min) Time (min)

C. PG 14 D. PG 15 1500 2500

2000 1000 1500

1000 500 500

Horizontal Activity (counts) 0 Horizontal Activity (counts) 0 0 90 180 270 0 90 180 270 Time (min) Time (min)

Figure 15. Results from unbiased screening of compounds 12–15 in DAT-KO mice (n = 4 for each). Horizontal activity after treatment of mouse at 90 minutes with 10 mg/kg, 150 minutes with 30 mg/kg, and 210 minutes 60 mg/kg.

As seen in Figure 15, these pargyline derivatives do not elicit the same response as the ACPAs (Figure 9–14). A summary of the overall movement is presented in Figure

16.

In review, ACPAs elicited a restorative phenotype in DAT-KO mice, while pargylines and 40 neuroactive compounds could not elicit the same response. We conclude the outcomes of treatment of DAT-KO mice with ACPAs are irrespective of

MAO-B inhibition, despite the fact that these compounds are inhibitors of monoamine oxidases. Additionally, the rapid response and the anti-akinesia properties observed clearly indicate a unique mechanism of action.

20000 10 mg/kg 30 mg/kg 60 mg/kg 15000

10000

5000 Horizontal Activity (counts/1 h)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Compound

Figure 16. Dose-dependent response of compounds on horizontal activity. Horizontal activity of DAT-KO mice (n = 4) was measured against increasing concentrations of compounds 1–15 every hour, 30 minutes after administration of aMTP. Compounds were tested at 10, 30, 60 mg/kg.

2.6 Hypothesis

In summary, information stated thus far illustrates the complexities of

Parkinson’s disease and highlights the deficits of both current treatments and animal models used to identify novel therapies. Equipped with the current knowledge from the literature and recent discoveries of PD and application of chemical biology principles and techniques, we sought to further identify and describe additional mechanisms.

These mechanism may prove impactful on Parkinson’s disease as monotherapies or polytherapies with current and/or emerging therapies [12]. Upon my affiliation to the

McCafferty group, I was tasked to elucidate the mechanism of action of ACPAs. In doing so, I synthesized aryl-sub derivatives of the ACPA drug class and identified potential drug targets for non-dopaminergic therapies for PD in parallel mechanistic studies.

3. Syntheses

The following chapter encompass the synthetic routes used to access the arylcyclopropylamine scaffold, as well as the discussion of additional phenethylamine.

3.1 Original Synthesis of Arylcyclopropylamine-Derivative, Tranylcypromine

The original synthesis commenced as shown in Scheme 2 with the condensation of ethyl diazoacetate and styrene to give both the cis- and trans-isomers of carboxylate

17, as well as the enantiomers for each of the geometric isomers. The geometric racemates were separately hydrolyzed to the respective racemic carboxylic acids 18.

Acyl substitution using thionyl chloride activated carboxylic acid 18 for a subsequent acyl substitution with sodium azide to give the carbonyl azide. The azide is subjected to

Curtius degradation in hot toluene, ultimately giving the isocyanate which readily hydrolyzes to the racemic amine, 1 [16]. This route took had 5 steps with an overall yield

17%.

Scheme 2. Original synthesis of tranylcypromine performed by Burger and Yost in 1948.

3.2 Synthetic Strategies

When considering the best route to access a library of compounds with the

ACPA scaffold, a few options are available. As Burger first demonstrated, styrene as a starting material allows for the simultaneous transformation of the double bound to the cyclopropyl ring and the installation of the protected carboxylic acid (ester). A subsequent Curtius rearrangement affords the ACPA scaffold. This route however is not stereoselective for either the trans- or cis-isomers. Similarly, Drs. Dooden and Polluck have shown unsaturated acids, unsaturated esters, and aryl-substituted aldehydes can follow a similar route. Additionally, cross-coupling reactions with aryl iodides and cyclopropylamines conjugated with an organometallic handle could yield the ACPA scaffold as well.

3.3 Preparative-Scale Synthesis

As stated in Chapter 2, Group Alumni, Drs. Julie Polluck and David Gooden synthesized a library of 11 arylcyclopropylamines by the following route (Scheme 3)

[34]. This route began with either esterification of commercially available carboxylic acids, or by palladium catalyzed cross-coupling between aryl iodides and methyl acrylate. Initially, cyclopropanation was attempted with the Corey-Chaykovsky reagent.

Due to low yields, diazomethane with palladium (II) acetate (Kishner cyclopropane synthesis [41]) was used to access the arylcyclopropylmethylesters 22. Alkaline

hydrolysis yielded the corresponding carboxylic acids to be submitted to the Curtius rearrangement to afford the desired amine in a total of steps 4 steps in yields ranging from 41–56%.

Scheme 3. Previous synthesis to access arylcyclopropylamine scaffold.

Upon my affiliation to the McCafferty group and with guidance from alumna Dr.

Katherine Alser Kim, a second synthesis was established based on the previously established route with modifications. These changes were necessary to satisfy the amount of material necessitated by the rat studies that will be discussed in the following chapter. The former route was not amenable to scaling up, as the cyclopropanation step with diazomethane is limited to a 2 millimole scale to ensure safety [42]. Assuming

“complete conversion” (~ 95%) of the a,b-unsaturated ester, the Kishner method would only yield just under 300 milligrams of material to be subjected to two more reactions before affording the desired amine. Yields for this route were 41–56% over 4 steps. At the time, our collaborators requested 8 grams of each derivative for the rat studies, so we sought to modify the established synthesis to permit larger scales to yield more material efficiently and safely. The amended synthetic route used to access the ACPA scaffold is shown in Scheme 4.

Scheme 4. Modified synthetic route to access the arylcyclopropylamine scaffold.

The synthesis began with the esterification of commercially available aldehydes yielding a,b-unsaturated esters using the Horner-Wadsworth-Emmons (HWE) olefination (Scheme 5).

Scheme 5. Horner-Wadsworth-Emmons Olefinaton of commerically available aldehyes to yield the corresponding a,b-unsaturated esters.

The original synthesis established in lab used extreme conditions in the form of low temperatures and a strong base. Sodium hydride (NaH) was originally used as the base and require a lower temperature of -80 °C since it has a pKa of 35. The modified synthesis used milder conditions via lithium hexamethyldisilazane (LiHMDS). This amendment now permits the HWE olefination to be performed at a higher temperature of -20 °C. LiHMDS is a milder base by greater than ten orders of magnitude with a pKa of 23. As such, LiHMDS is better tolerated with a larger scope of aldehydes which may also possess base-labile functional groups. Additionally, the lithium counterion coordinates with the phosphonate oxygens, better stabilizing the phosphonate carbanion than sodium. The (E)-alkene is the kinetic product and predominate isomer yielded from the reaction, however the thermodynamic (Z)-alkene product is also observed at relatively higher temperatures. Initially, the HWE olefination with LiHMDS was performed at 0 °C with the (Z)-alkene appearing as a minor product. This was evident in the 1H NMR as the coupling constant (J value) for vicinal, cis-alkene protons peaks are typically 10 Hz, while the vicinal, trans-alkene protons are typically 16 Hz. Peaks corresponding to cis-alkene protons were distinguishably identified from the trans- alkene protons. Lowering the temperature to -20 °C proved to mitigate the formation of 38

the (Z)-alkene. Both olefination methods gave the product in high yields (up to 95%), however the reaction time for the method using LiHMDS is 3 hours, whereas the method using NaH requires 7 hours.

In any case, the (E)-alkene does not have to purified away from the (Z)-alkene since the Corey-Chaykovsky cyclopropanation is stereoselective for the trans- arylcyclopropanecarboxylic acids due to the allylic strain imposed by the bulky aryl ring and ylide-adduct preventing free rotation about the Ca-Cb bond. As already stated, cyclopropanation was formally accomplished with a palladium-mediated process with in situ-generated diazomethane (CH2N2). The reaction however, is limited to a 2.0 millimole scale. This method also required extensive preparation and optimization in order to capture appreciable amounts of the gaseous diazomethane in solvent, thus the

Corey-Chaykovsky cyclopropanation was revisited as an alternative cyclopropanation.

In this method, a substituted methylene is delivered by the trimethylsulfoxonium ylide

[Me3S(O)I] to react with a,b-unsaturated carbonyls in a conjugation-addition type fashion (Scheme 6).

Scheme 6. One-pot Corey-Chaykovsky cyclopropanation and hydrolysis.

Additionally, a one-pot cyclopropanation and hydrolysis procedure could be employed to simultaneously extract and hydrolyze the arylcyclopropanated ester 29 to the corresponding carboxylic acid 26 in good yield. However, some substrates required purification prior to the hydrolysis dependent on the side-products observed in the

Corey-Chaykovsky (Scheme 6). The combined reactions gave the product 26 in good yields up to 60% in one day.

These acids were transformed to the corresponding primary amines by Curtius rearrangements in yields through two methods. Both methods required prior activation of the acid to undergo acyl substitution to give the arylcyclopropyl-carbonyl-azide.

Thermal decomposition of the azides furnished the corresponding isocyanates which can be trapped by tert-butanol yielding the carbamates, or Boc-protection amine 27 & 31.

In the first method, acids were activated with isobutyl chloroformate in the presence of base in ethereal solvent to give the corresponding mixed anhydride.

Subsequent acyl substitution with sodium azide dissolved in minimal water yielded the required arylcyclopropyl-carbonyl-azide precursor for the Curtius rearrangement. These azides can be stably isolated and transferred to toluene. Toluene serves well as a solvent in the Curtius rearrangement as it has a boiling point of 110 °C, thus is a candidate choice of solvent when considering the isocyanate is formed at 100 °C. Once the presence of the azide was undetectable on TLC, tBu-OH was added to capture the isocyanate and yield the carbamate at 80 °C (Scheme 7).

Scheme 7. Curtius rearrangement of (3,4)-methylenedioxy- derivative-intermediate, carboxylic acid 30 to yield the corresponding carbamate 31 (Boc-protected amine).

In the second method, acids were activated with diphenylphosphoryl azide in the presence of base in anhydrous tBu-OH in a one-pot procedure. It was hypothesized the in-situ generation of the azide at high temperatures would yield the isocyanate to be immediately captured by dry tBu-OH would limit the several side-products seen in method 1, however no difference was seen between either method (Scheme 8).

Scheme 8. Curtius rearrangement of carboxylic acid to yield the corresponding carbamate (Boc-protected amine).

3.3.1 Alternative Rearrangement Reactions

Efforts to overcome poor yields observed by the Curtius rearrangement involved employment of 2 alternative synthetic strategies.

Developed by W. Lossen, the Lossen rearrangement utilizes the decomposition of a hydroxamic acid into the isocyanate [43]; similarly, to the Curtius but bypasses the acyl azide intermediate which becomes undesirable to work with on preparative scales.

trans-arylcyclopropanecarboxylic acid as a representative-model compound was transformed into the required trans-arylcyclopropanehydroxamic acid 18 pre-cursor with hydroxylamine hydrochloride in a carbonyldiimidazole-mediated process.

Unfortunately, attempts to access the desired isocyanate from the hydroxamic precursor

32 were unsuccessful as no conversion was observed evident by recovered starting material (Scheme 9).

Scheme 9. Carboxylic acid 18 transformed into hydroxamic acid 32 for the subsequent Lossen rearrangement.

A modified Curtius, developed by Dubé et al, produces the Boc-protected, primary amine in milder reaction conditions than the conventional Curtius (Scheme 10)

[44].

Scheme 10. Mild-Curtius rearrangement of carboxylic acid 18 to the desired carbamate 34.

The milder-Curtius uses tetrabutylammonium bromide and zinc (II) triflate which reduces the need for higher temperatures to achieve the arrangement. Attempts to access the desired isocyanate was again unsuccessful. The lower heat in the reaction conditions is suspected to be insufficient to permit rearrangement to the isocyanate substrates desired. Additionally, many side-products were observed, as also seen in the previously mentioned Curtius conditions. This is not unexpected when concerned with extreme conditions and reactive intermediates.

3.3.2 Cross-Coupling Reactions

Cross-coupling reactions have a demonstrated utility in both academic and industrial sectors of research. With readily available organic (alkyl or aryl) halides and an organometallic compound, novel carbon-carbon bonds can be formed [45]. In the case with ACPAs, aryl iodides could be coupled to boronic esters fashioned to protected cyclopropylamines (Scheme 11) [14]. This strategy not only possesses late-stage derivation step, but also lessen accessing the ACPA scaffold to 2 steps.

Scheme 11. Suzuki-Miyaura method to access the cis- and trans-ACPA scaffold.

Yamaguchi et al developed a Suzuki-Miyaura method to not only access the

ACPA scaffold with select aryl iodides and aminocyclopropylboronic esters 35 in the presence of palladium, but to produce and select for the cis- and trans-ACPA isomers based on the presence of oxygen. Efforts to obtain the isolate appreciable quantities of the necessary aminocyclopropylboronic ester precursor were unsuccessful, as the of the borylation was low-yielding and the boronic acid is not stable to purification by silica gel chromatography (Scheme 12).

Scheme 12. Attempt to access aminocyclopropylboronic ester precursor for Suzuki- Miyaura method.

Initial efforts followed reported conditions but failed to produce borylated cyclopropylamines. We suspect the moisture-sensitivity of the iridium catalyst, (1,5-

Cyclooctadiene)(methoxy)iridium(I) dimer ([Ir(OMe)(1,5-cod)]2), is too great, in spite of efforts to keep the reaction atmosphere moisture-free.

Aminocyclopropyl organosilanes were also considered to access the ACPA scaffold by a Hiyama coupling, similarly utilizing select aryl iodides and a palladium catalyst (Scheme 13).

Scheme 13. Hiyama method access the cis- and trans-ACPA scaffold.

This method did not prove to be more promising as the coupling precursor 39 was unable to synthesized (Scheme 14).

Scheme 14. Attempt to access aminocyclopropyl organosilanes precursor for Hiyama method.

3.4 Additional Chemical Probes of Interest

The Caron group reported that (+)-MDMA was the only amphetamine derivative which could elicit full locomotion in the DAT-KO mice. Though known for deleterious side effects [46], anecdotally, PD patients have noted relief from PD-associated motor systems after taking MDMA. This scaffold and additional phenethylamine derivatives may be of interest in subsequent in vitro assays.

3.4.1 Phenethylamines derivatives

Figure 17. Analogous PEA derivatives to the ACPA derivatives

Additionally, phenethylamine derivatives are of chemical interest to ascertain the need of the cyclopropyl ring for activity. If the cyclopropyl ring is not required for the restorative phenotype.

3.4.2 MDMA-Derivates, UWA-121 & UWA-122

UWA-121 (49) and UWA-122 (50) are the individual diastereomers which comprise of UWA-101. UWA-101 is a substituted phenethylamine invented by the

University of Western Australia as a potential drug for Parkinson’s disease. Similar to

MDMA, bares the 3,4-methylenedioxy aryl substitution and monomethylated amine.

Additionally, UWA-101 possesses an a-cyclopropyl ring instead of the a-methyl. This modification retained the anti-Parkinsonian properties while eliminating the cytotoxicity and adverse behavior in animals. UWA-121 is a dual serotonin-dopamine reuptake inhibitor (SDRI), while UWA-122 is a serotonin transporter reuptake inhibitor. 46

Scheme 15. Synthesis of UWA enantiomers.

The synthesis for UWA-101 followed as proposed in Scheme 15. Knochol’s organozinc chemistry provided a method to access the piperonyl cyclopropyl ketone for subsequent reductive amination. Ethanolic methylamine in the presence of acetic acid and sodium cyanoborohydride yielded both UWA enantiomers. While these compounds were not assessed in later assays, these compounds will certainly provide interesting data in affinity and efficacy assays, as well as interesting data in the DAT-KO

(i.e. genetic knockout animal models).

3.5 Experimental

Chemicals and Reagents. All reagents and compounds were purchased from

Sigma-Aldrich, Arcos, or Fisher and were used without further purification. All solvents were ACS grade or better and used as is.

Unless otherwise stated, all air- or moisture-sensitive reactions were ran under inert atmosphere using argon. All chromatographic purifications were conducted via flash chromatography using ultra-pure silica gel (230-400 mesh, 60 Å) purchased from

Silicycle as the stationary phase. Thin Layer Chromatography was performed with glass backed silica gel (60 Å) plates purchased from Fischer and visualized with 254 nm UV light and the ceric ammonium molybdate (CAM), ninhydrin, and curcumin stains All 1H spectra were recorded in CDCl3 unless otherwise noted, using a 400 MHz Varian NMR spectrometer and coupling constants are reported in units of hertz (Hz).

3.5.1 ACPA derivatives

Procedure for Methoxymethyl Protection of Alcohol (Phenol):

4-(methoxymethoxy)benzaldehyde (51). Diisopropylethylamine (2.5 eq) was added slowly dropwise to a stirring solution of 4-hydroxybenzaldehyde (1 eq) in dichloromethane (0.33 M). Chloromethyl methyl ether (1.5 eq) was added slowly producing a gas. The reaction was stirred at room temperature for 1.25 h. After which, the reaction was quenched with water. The organic products were extracted with three portions of ethyl acetate, dried with anhydrous Na2SO4 and concentrated to a golden oil under reduced pressure. The desired product was isolated by flash chromatography over silica. 2.42 g, 91%, yellow oil. Rf = 0.33 (1:4 EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 9.91 (s, 1H), 7.84 (d, J = 8.0 Hz, 2H), 7.15 (d, J = 8.0 Hz 2H), 5.26 (s, 2H) 3.50 (s, 3H).

Procedure for tert-Butyldimethylsilyl Protection of Alcohol Protection:

4-((tert-butyldimethylsilyl)oxy)benzaldehyde (52). An alcohol (1.0 eq), N- methylimidazole (3.0 eq) and iodine (2.5 eq) were dissolved in anhydrous dichloromethane (0.33 M). Silyl chloride, TBDMSCl, (1.1 eq) was added and the reaction mixture was stirred at room temperature until the complete disappearance of the starting material (TLC analysis). The solvent was evaporated, the residue dissolved in

EtOAc and washed with concentrated aqueous Na2S2O3. The organic phase was dried over anhydrous Na2SO4 and evaporated. The products were purified by silica gel column chromatography. 2.01 g, 88%, red oil. Rf = 0.3. 1H NMR (400 MHz, CDCl3): d 9.89

(s, 1H), 7.79 (d, J = 8.0 Hz, 2H), 6.95 (d, J = 8.0 Hz, 2H), 1.00 (s, 9H) 0.25 (s, 6H).

General Procedure for Horner-Emmons-Wadsworth Olefination:

Under inert atmospheric conditions, LiHMDS (1.2 eq) was slowly added to a stirring solution of triethyl phosphonoacetate (1.2 ep). The solution was allowed to stir for 15–20 minutes before it was slowly cannulated into piperonal (1.0 eq) dissolved in anhydrous THF (0.8 M) at 0 °C. The base mixture was allowed to stir and allowed to come to room temperature, after which the reaction was quenched with saturated ammonium chloride. The layers were separated, and the aqueous layer was extracted with two portions of dichloromethane. All organic phases were combined, dried with sodium sulfate, and condensed under reduced pressure to give the a,b-unsaturated ester. The residue was purified by flash column chromatography (1:4 EtOAc/Hex) to give pure product as a yellow oil or solid.

Ethyl (E)-3-phenylprop-2-enoate (53). 1.23 g, 89 %, yellow oil. Rf = 0.45 (1:4

EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 7.68 (d, d, J = 16 Hz, 1H), 7.49 (m, 2H), 7.37

(m, 3H), 6.42 (d, J = 16.0 Hz, 1H), 4.25 (q, J = 8.0 Hz, 2H), 1.31 (t, J = 8.0 Hz, 3H).

Ethyl (E)-3-(4-(trifluoromethyl)phenyl)acrylate (54). 1.42 g, 90 %, red-orange oil.

Rf = 0.46 (1:4 EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 7.75 (d, J = 16.0 Hz, 1H), 7.65 (d,

J = 8.0 Hz, 2H), 7.52 (d, J = 8.0 Hz, 2H), 6.56 (d, J = 16.0 Hz, 1H), 4.41 (q, J = 8.0 Hz, 2H),

1.48 (t, J = 8.0 Hz).

Ethyl (E)-3-(4-bromophenyl)acrylate (55). 1.13 g, 92 %, red-orange oil. Rf = 0.45

(1:4 EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 7.60 (d, J = 16.0 Hz, 1H), 7.50 (d, J = 8.0

Hz, 2H), 7.36 (d, J = 8.0 Hz, 2H), 6.41 (d, J = 16.0 Hz, 1H), 4.26 (q, J = 8.0 Hz, 2H), 1.33 (t, J

= 8.0 Hz, 3H).

Ethyl (E)-3-(4-methoxyphenyl)acrylate (56). 2.50 g, 87 %, yellow oil. Rf = 0.47 (1:4

EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 7.64 (d, J = 16.0 Hz, 1H), 7.47 (d, J = 8.0 Hz,

2H), 6.90 (d, J = 8.0 Hz, 2H), 6.31 (d, J = 16.0 Hz, 1H), 4.25 (q, J = 8.0 Hz, 2H), 3.00 (s, 3H),

1.33 (t, J = 8.0 Hz, 3H).

Ethyl (E)-3-(4-ethoxyphenyl)acrylate (57). 1.16 g, 86 %, red-orange oil. Rf = 51 (1:4

EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 7. 64 (d, J = 16.0 Hz, 1H), 7.46 (d, J = 8.0 Hz,

2H), 6.88 (d, J = 8.0 Hz, 2H), 6.30 (d, J = 16.0 Hz, 1H), 4.25 (q, J = 8.0 Hz, 2H), 4.06 (q, J =

7.0 Hz, 2H), 1.42 (t, 7.0 Hz, 3H), 1.27 (t, J = 8.0, 3H).

Ethyl (E)-3-(4-isopropoxyphenyl)acrylate (58). 1.27 g, 89 %, red-orange oil. Rf =

53 (1:4 EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 7.63 (d, J = 16.0 Hz, 1H), 7.45 (d, J = 8.0

Hz, 2H), 6.87 (d, J = 8.0 Hz, 2H), 6.30 (d, J = 16.0 Hz, 1H), 4.58 (sep, 1H), 4.25 (q, J = 8.0

Hz, 2H), 1.33 (apparent m, 9H).

Ethyl (E)-3-(4-(methoxymethoxy)phenyl)acrylate (59). 10.4 g, 90 %, red-orange oil. Rf = 0.60 (1:4 EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 7.64 (d, J = 16 Hz, 1H), 7.46

(d, J = 8 Hz, 2H), 7.03 (d, J = 8 Hz, 2H), 6.32 (d, J = 16 Hz, 1H), 5.19 (s, 2H), 4.25 (q, J = 8

Hz, 2H), 3.47 (s, 3H), 1.33 (t, J = 8 Hz, 3H).

Ethyl (E)-3-(4-phenoxyphenyl)acrylate (60). 1.72 g, 92 %, red-orange oil. Rf = 0.6

(1:4 EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 7.64 (d, J = 16.0 Hz, 1H), 7.48 (d, J = 8.0

Hz, 2H), 7.37 (apparent t, 2H), 7.15 (t, J = 8.0 Hz, 1H), 7.04 (d, J = 8.0 Hz, 2H), 6.97 (d, J =

8.0 Hz, 2H), 6.34 (d, J = 16.0 Hz, 1H), 4.26 (q, J = 8.0 Hz, 2H), 1.33 (t, J = 8.0 Hz, 3H).

Ethyl (E)-3-(benzo[d][1,3]dioxol-5-yl)acrylate (61). 5.6 g, 86 %, yellow solid. Rf =

0.51 (1:4 EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 7.57 (d, J = 16.0 Hz, 1H), 7.01 (s, 1H)

6.97 (d, J = 8.0 Hz, 1H), 6.78 (d, J = 8.0 Hz, 1H), 6.24 (d, J = 16.0 Hz, 1H), 5.98 (s, 2H), 4.24

(q, J = 8.0 Hz, 2H), 1.32 (t, J = 8.0 Hz, 3H).

General Procedure for One-Pot Corey Chaykovsky Cyclopropanation & Alkaline

Hydrolysis:

Under inert conditions, trimethylsulfoxonium iodide (10 g) was dried under reduced pressure. Neat sodium hydride (1.09 g) was added to the trimethylsulfoxonium iodide and thoroughly stirred, yielding the dry ylide, (dimethyl(oxo)-λ6- sulfanylidene)methane (11.09 g, 8.2 mmol).

Dried a,b-unsaturated ester (1 eq), dissolved in anhydrous equal parts THF and

DMSO (0.2 M) was added to dry ylide (1.5 eq) and stirred. Once the starting material is consumed, lithium hydroxide hydrate (20 eq in [equal volume of reaction] DI water) was added to hydrolyze the cyclopropyl-ester intermediate. Once complete, the reaction mixture was acidified with HCl to a pH below 3 and left to stir overnight. Product was extracted with three portions of isopropyl acetate and organic layers were combined and washed with 50% NaCl solution, dried over sodium sulfate, and condensed under reduced pressure. The residue was purified by flash column chromatography (1:4

EtOAc/Hex) to give pure a pale yellow or brown solid.

(1R,2R)-2-phenylcyclopropane-1-carboxylic acid (18). 0.6 g, 51 %, dark yellow solid. Rf = 0.4 (1:2 EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 11.21 (br, 1H), 7.31–7.20 (m,

5H), 2.41 (apparent m, 1H), 1.79 (apparent m, 1H), 1.49 (apparent m, 1H), 1.27 (apparent m, 1H).

(1R,2R)-2-(4-(trifluoromethyl)phenyl)cyclopropane-1-carboxylic acid (62). 0.55 g, 59 %, dark yellow solid. Rf = 0.43 (1:2 EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 7.40

(d, J = 8.0 Hz, 2H), 6.98 (d, J = 8.0 Hz, 2H), 2.54 (apparent m, 1H), 1.86 (apparent m, 1H),

1.66 (apparent m, 1H), 1.36 (apparent m, 1H).

(1R,2R)-2-(4-bromophenyl)cyclopropane-1-carboxylic acid (63). 0.51 g, 55 %, dark yellow solid. Rf = 0.45 (1:2 EtOAc:Hex). 1H NMR (400 MHz, CDCl3): 11.67 (br, 1H),

7.09 (d, J = 8.0 Hz, 2H), 6.88 (d, J = 8.0 Hz, 2H), 2.62 (apparent m, 1H), 1.87 (apparent m,

1H), 1.66 (apparent m, 1H), 1.36 (apparent m, 1H).

(1R,2R)-2-(4-methoxyphenyl)cyclopropane-1-carboxylic acid (64). 0.54 g, 48 %, dark yellow solid. Rf = 0.47 (1:2 EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 11.61 (br, s), 56

7.03 (d, J = 8.0 Hz, 2H), 6.82 (d, J = 8.0 Hz, 2H), 3.83 (s, 3H), 2.56 (apparent m, 1H), 1.81

(apparent m, 1H), 1.61 (apparent m, 1H), 1.35 (apparent, 1H).

(1R,2R)-2-(4-(methoxymethoxy)phenyl)cyclopropane-1-carboxylic acid (65). 5.1 g, 55 %, dark-yellow solid. Rf = 0.40 (1:2 EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 7.02

(d, J = 8.0 Hz, 2H), 6.95 (d, J = 8.0 Hz, 2H), 5.13 (s, 2H), 3.45 (s, 3H), 2.48 (apparent m, 1H),

1.83 (apparent m, 1H), 1.55 (apparent m, 1H), 1.29–1.22 (apparent m, 1H).

(1R,2R)-2-(4-phenoxyphenyl)cyclopropane-1-carboxylic acid (66). 0.48 g, 53 %, yellow oil oil. Rf = 0.51 (1:2 EtOAc:Hex). 1H NMR (400 MHz, CDCl3): d 7.33 (t, J = 8.0 Hz,

2H, 7.09 (apparent t, J = 8 Hz, 3H), 6.99 (d, J = 8 Hz, 2H), 6.94 (d, J = 8.0 Hz, 2H), 2.59

(apparent m, 1H), 1.87 (apparent m, 1H), 1.66 (apparent m, 1H), 1.39 (apparent m, 1H).

(1R,2R)-2-(benzo[d][1,3]dioxol-5-yl)cyclopropane-1-carboxylic acid (30). 0.6 g,

60 %, yellow solid. Rf = 0.52. 1H NMR (400 MHz, CDCl3): δ 6.72 (d, J = 8.0 Hz, 1H), 6.61 57

(d, J = 8.0 Hz, 1H), 6.57 (s, 1H), 5.93 (s, 2H), 2.55 (apparent m, 1H), 1.81 (apparent m, 1H),

1.61 (apparent m, 1H), 1.34 (apparent m, 1H).

General Procedure for Curtius Rearrangement & Acidic Deprotection:

Diphenylphosphorazidate (1.1 eq) and anhydrous triethylamine (1.4 eq) were added sequentially to a room temperature 0.5 M solution of the carboxylic acid in anhydrous tert-butanol. The reaction was heated to 90 °C with an oil bath for 16 h, cooled to room temperature and concentrated to dryness under reduced pressure. The resulting residue was partitioned between ethyl acetate (15 mL) and 10% aqueous K2CO3

(10 mL). The organic products were extracted with ethyl acetate (2 x 15 mL), dried over

Na2SO4, filtered, and concentrated in vacuo.

Under inert conditions dried a dried carbamate (1 eq) is dissolved in dioxane (0.1

M) stirring. 4 N HCl (10 eq) is then added slowly to reaction mixture and then allowed to stir overnight. Once complete, the reaction is diluted and solvent is removed under reduced pressure. The remaining substance is washed and filtered with cold dioxane to give a yellow or brown solid. This solid is dissolved in water, filtered through a syringe, and then lyophilized to ensure purity in animal studies.

(1R,2S)-2-phenylcyclopropan-1-aminium chloride (1). 0.126 g, 32 %, white solid.

1H NMR (400 MHz, CD3OD): d 7.31–7.15 (m, 5H), 4.95 (br, 3H), 2.83 (apparent m, 1H),

2.42 (apparent m, 1H), 1.45 (apparent m, 1H), 1.30 (apparent m, 1H).

(1R,2S)-2-(4-(trifluoromethyl)phenyl)cyclopropan-1-aminium chloride (2). 0.120 g, 29 %, dark yellow solid. 1H NMR (400 MHz, CD3OD): d 7.61 (d, J = 8.0 Hz, 2H) 7.36 (d,

J = 8.0 Hz, 2H), 4.88 (br, 3H), 2.94 (apparent m, 1H), 2.47 (apparent m, 1H), 1.51

(apparent m, 1H), 1.43 (apparent m, 1H).

(1R,2S)-2-(4-bromophenyl)cyclopropan-1-aminium chloride (3). 0.191 g, 27 %, dark yellow solid. 1H NMR (400 MHz, CD3OD): 7.45 (d, J = 8.0 Hz, 2H) 7.10 (d, J = 8.0 Hz,

2H), 4.87 (br, 3H), 2.84 (apparent m, 1H), 2.33 (apparent m, 1H), 1.41 (apparent m, 1H),

1.34 (apparent m, 1H).

(1R,2S)-2-(4-methoxyphenyl)cyclopropan-1-aminium chloride (4). 0.169 g, 35 %, dark yellow solid. 1H NMR (400 MHz, CD3OD): d 7.09 (d, J = 8.0 Hz, 2H), 6.86 (d, J = 8.0

Hz, 2H), 4.88 (br, 3H), 3.76 (s, 3H), 2.76 (apparent m, 1H), 2.32 (apparent m, 1H), 1.36

(apparent m, 1H), 1.27 (apparent m, 1H).

(1R,2S)-2-(benzo[d][1,3]dioxol-5-yl)cyclopropan-1-aminium chloride (9). 0.126 g, 40 %, pale-yellow solid. 1H NMR (400 MHz, CD3OD): δ 6.76–6.66 (m, 3H), 5.90 (s, 2H),

4.89 (br, 3H), 2.76 (apparent m, 1H), 2.34 (apparent m, 1H), 1.38 (apparent m, 1H), 1.26

(apparent m, 1H). HRMS Calcd for (C10H11NO2Na) [M+Na]+: (178.0790 m/z); Found

(178.0).

Procedure for Acyl Substitution:

(1R,2R)-N-hydroxy-2-phenylcyclopropane-1-carboxamide. 0.364 g, Under inert conditions dried 28 (500 mg, 3.1 mmol) is dissolved in THF (5.18 mL) at room temperature. Carbonyldiimidazole (750.5 mg, 4.65 mmol) is then added and stirred until starting material is consumed. Hydroxylamine hydrochloride (1.29 g, 18.5 mmol) is then added to reaction mixture and stirred. Once complete, reaction is diluted with 5%

KHSO4 (31 mL) and extracted with EtOAc (2 x 31 mL). The combined organic phase was washed with brine and dried over sodium sulfate. The extract was filtered and condensed under reduced pressure. Rf = 0.43 (1:4 EtOAc/Hex). 1H NMR (400 MHz,

DMSO-d6): δ 10.58 (s, 1H), 8.80 (s, 1H), 7.29–7.10 (m, 5H), 2.26–2.21 (m, 1H), 1.68–1.64 (m,

1H), 1.37–1.32 (m, 1H), 1.25–1.20 (m, 1H). HRMS Calcd or (C10H11NO2Na) [M+Na]+:

(177.0790 m/z); Found (178.0865).

Procedure for Pivaloyl Protection:

N-cyclopropylpivalamide. To a solution of cyclopropylamine (1.0 eq) and N,N- diisopropylethylamine (1.1 eq) in dichloromethane (0.93 M), pivaloyl chloride (1.05 eq) was slowly added at 0 °C. The reaction mixture was stirred at room temperature overnight. Saturated aqueous solution of NaHCO3 was added to the reaction mixture and extracted with two portions of CH2Cl2. The combined organic layers were washed with brine, dried over Na2SO4, filtered and concentrated in vacuo. The resultant residue was purified by silica gel flash column chromatography (1:2, EtOAc:Hex). 1H NMR (400

MHz, CDCl3) δ 5.85 (br, 1H), 2.73–2.68 (m, 1H), 1.17 (s, 9H), 0.78–0.75 (m, 2H), 0.49–0.45

(m, 2H).

3.5.2 UWA derivatives

2-(benzo[d][1,3]dioxol-5-yl)-1-cyclopropylethan-1-one. A flame-dried and argon-flushed round-bottom flash, equipped with a magnetic stirrer and a septum, was charged with magnesium turnings (304 mg, 12.5 mmol). LiCl (12.5 mL, 0.5 M in THF,

6.25 mmol) and ZnCl2 (11.0 mL, 0.5 M in THF, 5.5 mmol) were added. 3,4- methylenedioxybenzyl chloride (1.71 mL, 50% in DCM, 6.67 mmol) was added in one portion at room temperature. The reaction mixture was stirred for 2 h. CuCN·2LiCl (25.0

µL, 1.0 M in THF, 0.03 mmol) was added at -20 °C. After stirring for 15 min, cyclopropanecarbonyl chloride (306 µL, 3.5 mmol) was added and the mixture was stirred for 1.5 h at 0 °C followed by 30 min at 25 °C. The reaction mixture was quenched with sat. aq. NH4Cl solution (20 mL) followed by 25% aq. NH3 solution (10 mL) and extracted with Et2O (3x 20 mL). The combined organic layers were dried over Na2SO4 and concentrated in vacuo. Flash column chromatography (EtOAc/Hex = 1:9) furnished

2-(benzo[d][1,3]dioxol-5-yl)-1-cyclopropylethan-1-one as a yellow oil (362 mg, 27 %). Rf =

0.42 (1:9 EtOAc/Hex). 1H NMR (400 MHz, CDCl3): d 6.76 (d, J = 8.0 Hz, 1H), 6.70-6.66 (m,

2H), 5.93 (s, 2H), 3.73 (s, 2H), 1.99-1.93 (m, 1H), 1.04–1.00 (m, 2H), 0.87–0.82 (m, 2H).

Procedure for Reductive Amination:

(±)-2-(benzo[d][1,3]dioxol-5-yl)-1-cyclopropyl-N-methylethan-1-amine. 0.3 g, 77

%, colorless oil, Rf = 0.25 (1:1:99 NEt3/MeOH/DCM). Glacial acetic acid (0.97 mL, 17 mmol) was added dropwise to a stirred mixture of _ (398 mg, 1.66 mmol), 33% ethanolic methylamine (2.06 mL, 16.6 mmol) and powdered 3A molecular sieves (0.45 g) in dry

THF (2 mL) at 0 °C under argon. After 10 min, NaCNBH3 (104 mg, 1.66 mmol) was added and the reaction mixture was stirred at 50 °C for 24 h. On cooling the reaction mixture was quenched with 1 M HCl, vacuum filtered through a pad of Celite and washed through with H2O, MeOH and Et2O. The filtrate was basified with 1 M NaOH

(50 mL) then extracted with Et2O (2 × 40 mL). The extract was washed with water (3 × 40 mL) and brine (40 mL), dried and evaporated to afford a viscous yellow/brown oil, which was purified by rapid silica filtration. 1H NMR (400 MHz, CD3OD): δ 6.81–6.75

(m, 3H), 5.94 (s, 2H), 3.07–2.94 (m, 2H), 2.75 (s, 3H), 2.66—2.62 (m, 1H), 0.9 (m, 1H), 0.72

(m, 1H), 0.59 (m, 1H), 0.06 (m, 1H).

4. Behavioral Studies Evaluation of Arylcyclopropylamines in 6-hydroxydopamine-lesion Rat Model for Parkinson’s disease

As mentioned in the previous chapter, experimental models of PD utilizing neurotoxicants, mutations in genes linked to PD, or at times a combination of both, have been generated to study PD to identify new treatments. The rate of success to translate successful treatments is heavily dependent on these models accurately recapitulating hallmarks of the disease state [28].

After ACPAs 4 and 9 were synthesized and purified, approximately 1 gram of each derivative1 was provided to the Napier group as part of a collaboration to compare activity observed in the DAT-KO mouse model. The following study was performed at

Rush University2.

4.1 Unilateral 6-Hydroxydopamine-Lesion Rat

The phenotypic-rescue observed in the DAT-KO mouse model was promising, though at the time of studies the model was had not yet been recognized as a valid model for PD. Subsequent studies have since demonstrated to utility of the model [37], however we sought out to determine if ACPAs could reproduce similar results in the 6- hydroxydopamine (6-OHDA)-lesioned rat model, a well-established animal model for

1 With help from Lindsey Olivere, Duke University. 2 Behavioral studies were performed by Dr. Brinda Bradaric, Rush University, unpublished data. 66

PD [38]. Two representative derivatives were selected from the focused-library to illustrate opposite phenotypes observed in the mouse model; ACPA 4 representing a well performing derivative and ACPA 9 representing a poor performing derivative.

4.2 Design and Validation of Animal Model

To determine the capacity of the ACPA-class of compounds to provide motor benefits (i.e. anti-parkinsonian activity) the two beforementioned compounds were assessed by the forelimb adjusted step task. Before studies could begin, each rat had to be bred, raised, and trained to perform tasks which can be interpreted to display the appropriate responses from which we could ascertain the effects of the ACPAs. Once training was complete, the rats underwent surgery where they were unilaterally lesioned with 5-(2-aminoethyl)benzene-1,2,4-triol, formally known as 6-OHDA. 6-

OHDA is a neurotoxin that destroys catecholaminergic (dopaminergic and noradrenergic) nerve terminals. To specifically target the dopaminergic terminals, a norepinephrine reuptake inhibitor was injected intraperitoneally (i.p.) into the medial forebrain bundle 30 minutes before the 6-OHDA injection into one side of the brain.

Such a lesion produces PD-like symptoms on the contralateral side within 3 days following the 6-OHDA injection, including deficits in initiating movement and postural adjustment steps. This emulates aspects of the akinesia and postural instability seen in

PD. Briefly describing the test, when using the forelimb adjust step task (step-test), 3 of

the 4 paws are held, while the 4th paw is allowed to adjust step as the animals is moved across a platform (0.9 meters in 5 seconds). A normal, un-lesioned rat would be able to adjust step immediately as they are being moved, completing approximately 12–15 steps in the length of the test on each paw. However, in a lesioned rat, only the unaffected paw, that is the same side as the 6-OHDA injection, would be able to adjust step; whereas the affected paw, opposite side to the 6-OHDA injection, would drag along the length of the board. For this study, the rats were step-tested throughout the experimental paradigm (before surgery, after surgery, before L-DOPA treatment and during L-DOPA treatment) per published protocols [39].

The experimental time line was as follows: Following recovery from surgery and stability of the lesion (approximately 20 days post-surgery), animals were given a daily dose of L-DOPA, and stepping was assessed. With chronic L-DOPA treatment, animals

(as well as PD patients) develop dyskinesias, or abnormal involuntary movements [40].

Following administration of L-DOPA, rats are observed for dyskinetic behavior

(uncontrolled limb movement [limb dyskinesia], distortion or twisting of the body [axial dystonia], increased locomotion (spinning in one direction)) in a Plexiglas cylinder every

20 minutes for a total of 3 hours. These dyskinesias can be given a severity score between 0 and 4 according to the following scale: 0 = no dyskinesia, 1 = dyskinesia < 50

% of observation time, 2 = dyskinesia > 50% of observation time, 3 = continuous dyskinesia interrupted by strong stimulus, 4 = continuous, uninterrupted dyskinesia.

Animals were given daily injection of L-DOPA for up to 4 weeks or until stepping and dyskinesia scores appeared to stabilize.

Animals (n = 4–6) were administered either a L-DOPA vehicle (saline; 1 ml/kg, i.p.), L-DOPA (3 mg/kg, i.p.) or L-DOPA (6 mg/kg, i.p.) once per day, and dyskinesias were scored by a treatment-blinded observer. The scores for each time point were added to give a total score per session (the maximum possible score is 36) (Figures 11–

12).

A. Locomotor Score 25

20 Vehicle (n = 5) 15

10 3 mg/kg L-DOPA (n = 4) 5 6 mg/kg L-DOPA Tottal Score per Session (n = 4) 0 0 10 20 Number of Injections

B. Axial Dystonia Score 25

20 Vehicle (n = 5) 15

10 3 mg/kg L-DOPA (n = 4) 5 6 mg/kg L-DOPA Total Score per Session (n = 4) 0 0 10 20 Number of Injections

Figure 18. Behavioral observation of dyskinesias. A: locomotor - circular, rotating movement (usually in one direction); B: axial dystonia score – distortion or twisting of the body.

A. Limb Dyskinesia Score 25

20 Vehicle (n = 5) 15

10 3 mg/kg L-DOPA (n = 4) 5 6 mg/kg L-DOPA Total Score per Session (n = 4) 0 0 10 20 Number of Injections

B. Forelimb Adjustment Step Task 18

15 L-DOPA Tx: Vehicle 12 (n = 2) 9 L-DOPA Tx: 3 6 mg/kg (n = 4) 3 L-DOPA Tx: 6 0 mg/kg -10 0 10 20 30 40 50 60 70 (n = 4)

Average Steps per Session per Day Time (days)

Figure 19. Behavioral observation of dyskinesias (cont.) A: limb dyskinesia – uncontrolled limb movement; B. Forelimb adjustment step task.

Rats with 6-OHDA-induced lesions rats develop dyskinesias with chronic l L-

DOPA treatment. Following either the 3 or 6 mg/kg of L-DOPA, rats exhibit axial dystonia and limb dyskinesias that persist throughout the dosing paradigm (day 55).

The higher dose of L-DOPA (6 mg/kg), as anticipated, produced a greater dyskinesia severity as well as increased motor activity as compared to the 3 mg/kg L-DOPA dose. 6-

OHDA lesioned rats that were treated with vehicle showed no dyskinetic behavior and normal exploratory locomotion (*p < 0.5 using 2-way repeated measure ANOVA with a

Bonferroni post hoc). It was noted a stepping deficit is present one day after the surgical procedures used to inject 6-OHDA into the brain (Day 0), persisted until the start of L-

DOPA treatment (day 33; tx). One week following either the 3 or 6 mg/kg dose of L-

DOPA, rats completely recover of the stepping deficit. The stepping recovery remains stable and persists throughout the L-DOPA treatment period (i.e., until Day 54). Once the therapeutic index of L-DOPA for rats in this study, ACPA 4 and ACPA 9 were assessed in the forelimb adjustment task as just described.

4.3 Acute-Administration of ACPAs Accessed by Forelimb Adjusted Step Task

Dosing profiles for each compound were established to determine at which dose each compound becomes toxic. The toxic dose (TD50) is defined as animal hypersensitivity to sensory stimuli, tremors, or seizures at 50% of a given dose and the lethal dose (LD50) is defined as death at 50% of a given dose. ACPAs were acutely administered as shown in Table 4.

Table 6. Toxicity profile as a function of ACPA 4 or ACPA 9 at given concentrations in PD-like rat model.

Adverse Events Dose Tested n (# of rats showing TD50 LD50 (mg/kg) behavior) 3 5 None

6 5 None > 6 mg/kg Not < 9 mg/kg defined 9 5 Tremor (5) 12 4 Tremor (4)

6 8 None Tremor (2), seizures (1), > 9 mg/kg 9 4 9 mg/kg catalepsy (1), death < 60 mg/kg (1) Tremor (4), seizure 60 4 (4), death (4)

ACPA 4 ACPA 9 20 20

15 15

10 10

5 5 Adjusted Steps Adjusted Steps Average Number of Average Number of 0 0

Pre-LPost-L Pre-L Post-L 3 mg/kg6 mg/kg9 mg/kg12 mg/kg 6 mg/kg9 mg/kg60 mg/kg

Figure 20. 6-OHDA rats assessed in forelimb adjustment step task post-administration of ACPA compound at indicated doses.

4.4 Conclusions

As previously mentioned, two representative derivatives were selected from the focused-library to illustrate opposite phenotypes observed in the mouse model; ACPA 4 representing a well performing derivative and ACPA 9 representing a poor performing derivative. Each compound was administered at the indicated doses and were assessed in the forelimb adjustment step task. Post 6-OHDA-lesion, rats were unable to adjust step in the task on the affected limb (Figure 20). However, following administration of

ACPA 4 at all doses tested (3–12 mg/kg), rats significantly increased the number of steps taken compared to post-lesion baseline. At higher doses, rats had almost complete restoration of number of adjusted steps. While, ACPA 4 was well tolerated at lower doses (3 and 6 mg/kg), and at higher doses (9 and 12 mg/kg), all rats exhibited adverse pharmacology, such as tremors. Even so, the therapeutic range for compound 4 was determined to be 3 mg/kg < TD 50 < 9 mg/kg. Following administration of ACPA 9, lesioned rats showed improvement in adjusted steps with increasing dose. However, at higher doses, rats exhibited greater adverse pharmacology, including tremors, seizures and death. The therapeutic range for ACPA 9 was determined to be TD50 ≤ 6 mg/kg. Due to the larger therapeutic range and less severe adverse pharmacology, further studies were continued with lower doses of ACPA 4, which improved postural instability without adverse side effects. It is important to note as well, a structure-activity

relationship is apparent with only two ACPA derivatives. Subsequent studies with an expanded library will certainly yield more interesting results.

4.5 Experimental

Chemicals and Reagents. Some reagents and compounds were purchased from

Sigma-Aldrich, Arcos, or Fisher and were used without further purification. All solvents were ACS grade or better and used as is.

Animals. Male Sprague-Dawley rats (Harlan Laboratories, Indianapolis, IN), weighing 300–400 grams prior to 6-OHDA lesion were used for forelimb adjust step task and adverse pharmacology. All rats were handled in accordance with the procedures established in the Guide for the Care and Use of Laboratory Animals (National Research

Council, Washington DC). Protocols were approved by the Rush University Medical

Center Institutional Animal Care and Use Committee.

6-OHDA-induced Lesion: All surgeries were performed as in previous studies

(Muma, et al. 2001). Prior to surgery, animals were injected with desipramine (25 mg/kg, intraperitoneal; a norepinephrine re-uptake inhibitor), allowing for specific dopaminergic lesion. The rats were then anesthetized with isoflurane (2–3%) and placed in a small stereotaxic apparatus (David Kopf, Tujunga, CA) with the nose piece set at 3.3 mm below the horizontal. Following shaving and application of topical antiseptic solution, a midline incision was made, and two burr holes drilled over the medial forebrain bundle (MFB) at the following coordinates (in mm, from Lambda): AP + 4.2,

ML ± 1.6, DV -8.5 from the surface of the skull. Infusion of 6-OHDA HBr (12 µg in 4.0

µL of 0.2% ascorbic acid with 0.9% normal saline, pH = 5) was administered at a rate of

0.5 µL/min in one hemisphere using a 33-gauge injector connected to an infusion pump.

Vehicle injections (4.0 µL of 0.2% ascorbic acid with 0.9% normal saline, pH = 5) were used for the injections in the opposite hemisphere. Injectors were left in place for 2 minutes to allow diffusion away from the tip. The burr holes were covered with bone wax and the incision sutured. Rats were then returned to the home cage following full recovery from the anesthesia and had free access to food and water. Lesioned rats were assessed in behavior tests approximately 2 weeks following surgery to allow for full recovery and lesion extent.

Forelimb Adjusted Step Task: Pre- and post-lesion baseline and following daily escalating dose of compound (3–60 mg/kg, i.p.), rats were subjected to the adjusted step task as previously described (Olsson et al., 1995; Rokosik & Napier 2012). In brief, the experimenter firmly suspended the rear legs and one forelimb of the rat while the weight of the rat was supported on the unrestrained limb. The rat is then moved horizontally along a table (0.9 meters in 5 seconds) in three trials. Both the right and left forelimb were assessed. For each session, the average steps for each forelimb was calculated by averaging the number of adjusted steps made during the three trials on each forelimb.

Adverse Pharmacology. Arylcyclopropylamine compounds were screen in vivo for presence of adverse pharmacology. [PMID: 26091637] Adverse pharmacology included: hypersensitivity to sensory stimuli, weight loss greater than 10 % overnight,

tremors, seizures, or death. Qualitative abnormal involuntary movements, such as increased rotational behavior, limb dyskinesia or axial dystonia were also noted.

Locomotor activity in 6-OHDA lesioned rats. Observations were video recorded by an experimenter that was blinded to the treatment groups. For each observation time point, rats were placed in a clear Perspex cylinder (22 cm x 34 cm x 20 cm) for 1 minute for assessment. Prior to as well as following administration of either vehicle saline, L-DOPA (3 mg/kg, i.p.) or L-DOPA (3 mg/kg, i.p.) and ACPA 4 (1.5 mg/kg, i.p.), rats were observed for 1 minute every 20 minutes for 3 hours. Rats were given a score from 1–4 based on amount of horizontal and/or vertical movement and rotational behavior in the cylinder.

0 = absent 3 = present throughout the min but 1 = present for less than 30 s suppressed by stimuli 2 = present for more than 30 s 4 = present throughout the min but not suppressible by stimuli

Borderline scores (i.e. 0.5) were allowed in order to increase the sensitivity of the rating scale. Enhanced manifestations of normal behavior, such as sniffing, grooming and gnawing were not included in the scoring, but were noted.

Data analysis. The data are presented as mean ± SEM and analyzed using a two- tailed Student’s t-test and one-way analysis of variance (ANOVA) followed by Dunnet’s multiple comparison test or a two-tailed Mann-Whitney U test when appropriate. To

determine changes in forelimb adjusted step task following each dose, data were compared using a two-way repeated measures ANOVA (GB-Stat, Silver Spring, MD) for time vs dose, followed by Newman–Keuls test to determine between and within group differences. Significance was set a priori at p < 0.05. Data are presented as the mean ±

SEM.

5. G-Protein Coupled Receptors

Thus far, the ACPA-drug class has demonstrated anti-akinesia activity comparable to the L-DOPA within two non-primate animal models for PD. This phenocopy is greatly significant because of the difference in design of each animal model. The 6-(OHDA)-lesioned rat model mimics motor deficits contralaterally to the lesioned hemisphere of the rat brain. Similarly, the DAT-KO DA-deficient mouse model is a chemical-genetics animal model which acutely mimics PD-like motor deficits when

TH inhibitor, a-methyl-p-tyrosine, is administered. However, the mouse model assessment conditions are devoid of DA, unlike the rat model which require supplemental DA (chronically administered L-DOPA) to be viable. Once lesioned, these rats can longer store DA as synaptic vesicles are no longer intact. This suggests a mechanism independent of DA, meaning dopaminergic signaling is not contributing to the rescue phenotype, and thus begins to eliminate biological targets while identifying potential drug targets.

The on-set of the ACPA activity also provides insight into target identification as well. Both animals displayed sustainable improvement from motor deficits within minutes of administration. Such kinetics suggest ACPAs mechanism of action is not mediated by one or multiple intracellular enzymes (i.e. transferases, hydrolases, lyases).

Rather, one or more membrane-bound proteins at the cell surface (i.e. receptors, transporters, ion channels). Conventional pharmacodynamics understands drug

responses rates which involve membrane proteins (sec-1 or min-1) are rapider than rates involving intracellular proteins (h-1 or days-1), making these proteins lead candidate targets to assay for affinity and activity [47, 48].

5.1 Survey of the GPCRome

Understandably, G-protein coupled receptors (GPCRs) represent the largest and most versatile group of cell surface receptors. This superfamily is classically divided in three main classes which have no detectable shared sequence homology. Class A

(rhodopsin-like) is the largest group consisting of 85% of GPCR genes and includes the

GPCRs responsible for mediating neurochemical activity in brain. Unsurprisingly,

GPCRs consist of over one-third of drug targets for medicinal drugs currently on the market. To date, there over are 700 known GPCRs [49]. This success would not have been possible without the foundational work of Robert J. Lefkowitz, MD (Duke, HHMI) and Brian K. Kobilka, MD (Stanford) in which they began elucidating molecular properties and regulatory mechanisms controlling GPCR function [50-57]. GPCRs are seven-transmembrane domain receptors that activate intracellular signal transduction in the presence of a ligand, eliciting a number of cellular responses which vastly differ based on receptor type and localization within the body. These types of receptors are found only in eukaryotes, including yeasts and other unicellular organism, and are able to detect an array of chemical molecules (ligands) with appreciable selectivity.

5.2 G-Protein Coupled Receptor’s Structural Components

The binding event of ligand to receptor defined as the ligand-receptor complex for a given ligand-receptor pair [58]. The nature of a complex is measured respective to the ligand’s affinity (probability of the drug occupying a defined molecular region of the receptor) and intrinsic efficacy (activity—degree which ligand activates the receptor).

These two features agree with the structure-function dogma of biochemistry. The ligand’s chemical structure defines the molecular recognition (i.e. hydrogen bonding, hydrophobic interactions, pi-pi interactions, halogen bonding), and these additive forces stabilize the ligand-receptor complex, and in part, defines the ligand’s function in the body (pharmacology).

Prior to a binding event, an un-liganded receptor exists in equilibrium of a number of possible (biologically relevant) conformational states. The binding of a ligand induces dynamic, conformational changes shifting the equilibrium to a new conformational state, described by the rate of association (kon). In the active complex

(ligand-receptor complex), rearrangement of the receptor’s transmembrane (TM) helices, particularly helix III and VI, and a consequent opening of the receptor’s cytosolic face, allows it to interact with its native G-protein and trigger downstream signaling events.

In the reverse process, described as rate of dissociation (koff), the ligand is unbounded, returning the receptor back to an inactive state [47, 48, 59-61].

The scope of molecules that are able to bind and activate GPCRs is vast and diverse, ranging from amino acids or neurotransmitters to peptides, and even large proteins. While GPCRs can discriminate chemical structures and scaffolds, most still possess a ligand profile consisting of a multiple chemically-distinct classes of compounds, and thus a single GPCR can elicit more than one biological response based on effectors recruited by conformationally-distinct ligand-receptor complexes. This presents a challenge when identifying candidate GPCR targets, as techniques must be scalable, robust, and applicable against the increasing number of identified and cloned

GPCRs.

5.3 G-Protein Coupled Receptor Biological Functions

The classical role of GPCRs is to couple the binding of agonists (ligands which activate the receptor) to the activation of specific heterotrimeric G-proteins, leading to the modulation of downstream effector proteins.

5.3.1 G-Protein Dependent Signaling

When an agonist binds, dynamic, conformational changes occur which reveal amino-acid residues of the intracellular helices (IL) and transmembrane domains (TM).

These domains can be further described as a guanine nucleotide exchange factor (GEF).

GEFs catalyze the release of G-protein bound GDP (G•GDP) which allows GTP to bind,

activating the G-protein (G•GTP). In some cases, the G-protein is pre-coupled to the

GEF of a GPCR, but most bind once the necessitated GEF is exposed. The catalysis is critical to signal transductions pathways, as G•GDP is tightly bound, thus exhibiting very slow rate kinetics as the rate-limiting step of signal transduction. The G-proteins which couple receptors are heterotrimeric (large) GTPases composed of three subunits—

Ga, Gb, and Gg. Hydrolysis of GTP to GDP (and phosphate) is enacted by the Ga subunit, thus is the site which GDP and GTP “exchange”. 20 (18 in human) Ga subunits have been identified and are assigned into four different families: (1) Gas (Gs) subunits trigger cAMP-dependent transduction by activating adenylyl cyclase (AC) which converts ATP into cAMP, increasing intracellular concentrations; while (2) Gai (Gi) subunits inactivates adenylate cyclase, decreasing intracellular cAMP.; (3) Gaq (Gq) subunits trigger inositol triphosphate (IP3) transduction by activating phospholipase C (PLC). PLC hydrolyzes phosphatidylinositol 4,5-bisphosphate (PIP2) into IP3 and diacyl glycerol (DAG). DAG remains membrane associated and activates protein kinase C (PKC), while IP3 diffuses through cytosol and stimulates the release of Ca2+ from the smooth endoplasmic reticulum.; And (4) G12/13 which activate monomeric (small) GTPases of the Rho Family, subfamily of the Ras superfamily, and modulate cellular trafficking and cell cycling.

The beta and gamma units form the beta-gamma (Gbg) complex. This complex is regulatory in that is binds (sequesters) Ga•GDP until Ga is active (Ga•GTP) and binds once again after catalysis (Ga•GDP). Similar to DAG, the Gbg subunits remain bound to

cytosolic face of the membrane and transduces other downstream signals, effecting cellular responses as well (Figure 21) [48, 60, 78-80].

Figure 21. Condensed summary a few of the multiple pathways GPCRs can transduce.

With vast function, comes vast regulation. A GPCR activated for an extended period of time will become eventually desensitized. Homologous desensitization of a

GPCR occurs to active and inactive forms of the receptor, while heterologous desensitization of a GPCR occurs to active and inactive forms on the receptor and inactive forms of other receptors types as well. Termination of a cellular response occurs at two stages early in the signal transduction. In feedback mechanisms, serine/threonine

residues within IL-3 and C-terminal tail of an activated GPCR are phosphorylated by either cAMP-dependent protein kinases (PKA or PKC) or G-protein coupled receptor kinases (GRK). This phosphorylation enhances the binding affinity for b-arrestin, which blocks coupling of an agonist-bound receptor to its G-protein. b-arrestin also function as a scaffolding protein by recruiting endocytic machinery, promoting internalization of the receptor. The phosphorylation state of the receptor dictates if the receptor is either recycled back to the plasma membrane or degraded by lysosomal vesicles. While G- proteins intrinsically are able to self-terminate, this happens very slowing. Regulators of

G-protein signaling (RGS) bind Ga subunit with their GTPase-accelerating protein (GAP) domain and accelerates hydrolysis 1500-fold. This action inactivates the G-protein, returning it to the Gbg-complex (Figure 21) [37, 47-49, 55, 61, 68, 78].

5.3.2 G-Protein Independent Signaling

In the classic paradigm, GPCRs were believed to function as a binary switch that could be activated by agonist binding or inhibited by antagonist blockade of agonist binding. In the mid-1990s, it became clear that GPCR signaling was more complex as G- protein independent signal transduction were discovered. This complexity of GPCR signaling was further uncovered when different receptor conformations, receptor phosphorylation patterns, receptor binding sites, and even receptor independent bias was shown to activate unique downstream responses. While most endogenous agonists

(“balanced agonists”) activate pathways dependent and independent of G-protein, some agonists can selectivity activate one pathway over another yielding distinct cellular responses. Functional selectivity refers to the process by which GPCR ligands differentially modulate canonical pathways involving G-proteins and non-canonical pathways involving other signaling proteins, mainly β-arrestins. Bias mechanisms have also been described between two different G-proteins, between β-arrestin-1 and -2, and between different states of the same receptor bound to different ligands as well.

However, the most well described examples of biased ligands refer to selective G- protein signaling versus β-arrestin-mediated signaling (Figure 21) [37, 47-49, 55, 61, 68,

78].

Believed to have solely regulate GPCR activation, b-arrestin as an early-stage transducer of GPCRs has received much attention. Studies investigating b-arrestin signal transduction suggests delineation of the G-protein and b-arrestin pathways could enhance the therapeutic index of treatments. Biased agonism of opioidergic receptors, µ, d, and k, for G-protein pathways could still elicit signaling that leads to a therapeutic response without activation b-arrestin pathways which lead to the side effects of balanced agonists. However, in the case of the sphingosine 1-phosphate Type 1 (S1P1) receptor where biased agonism for b-arrestin signaling led to a therapeutic outcome as b-arrestin targets activated S1P1 and rapidly traffics the receptor into degradation pathways by promoting a unique ubiquitination profile [37, 47-49, 55, 61, 68, 78].

As such, predicating biological activity is challenging and requires multiple in vivo assays performed in tandem.

5.4 G-Protein Coupled Receptor Assays

As already stated, GPCRs are valuable molecular targets for drug discovery. An important component of the early drug discovery process is the design and implementation of high-throughput GPCR functional assays that allow the cost-effective screening of large compound libraries to identify novel drug candidates. Several assays based on radioactivity, fluorescence, and/or chemiluminescence detection are commercially available for convenient screen development. Additionally, new GPCR biosensors and imaging technologies have been developed for functional screen in live cells.

5.4.1 Affinity

A receptor affinity assay can be used to characterize the interaction between a receptor and an investigational ligand. The assay provides can provide detailed information such as the intrinsic affinity of ligands to the receptor, association/dissociation rates, and the density of receptor in tissues or cells. Receptor affinity assay is a cell-free method suitable for any GPCR screening without involving downstream signaling from the receptor. This type of assay can also obtain determine

agonism and antagonism in one experiment but cannot discern between either type of activation. Additionally, the low availability of labeled ligands greatly limits the application of this assay.

5.4.2 Efficacy

While affinity assays are useful to identify new compounds and/or GPCR targets, further characterization of the biological responses after binding will elucidate the full characteristics of an investigational compound. Upon ligand binding, GPCRs change their conformation and activate coupled G-proteins, which subsequently promote second messenger production via downstream effectors. G-protein dependent assays measure either G-protein activation or G-protein mediated events, including second messenger generation and reporter activation. Assays have also been developed to characterize G-protein independent pathways and are more intensive but have become more readily robust.

6. Mechanistic Affinity Studies

In this chapter I will discuss the power of radioligand assays and generation of

GPCR affinity profiles. Competition binding assays are used to determine the selectivity for a particular ligand for receptor sub-types. Competition curves are obtained by plotting specific binding, which is the percentage of the total binding, against the log concentration of the competing ligand. A steep competition curve is usually indicative of binding to a single population of receptors, whereas a shallow curve, or a curve with clear inflection points, is indicative of multiple populations of binding sites. Figure 22 depicts a representation of the assay and how affinity is measured as a function of the investigational ligand.

Figure 22. Visual representation depicting experimental design of radioligand binding competition assay. A GPCR (green) of interest is preincubated with the appropriate radioligand (orange). An investigational ligand (purple) is then added to assess how well it can compete for the same binding site, which is measured by the amount of displaced radioligand.

6.1 Experimental Design

In primary binding assays, nonspecific binding is defined as 100% inhibition in the presence of 10 µM of the appropriate reference compound, while total binding is defined as 0% inhibition in the absence of test compound or reference compound. The radioactivity in the presence of test compound is calculated with the following equation and expressed as a percent inhibition.

�� − �� ℎ�� = 100 − × 100 �� − ��

Compounds that show a minimum of 50% inhibition are assayed again in a secondary binding assay to determine equilibrium binding affinity at specific targets. In the secondary binding assays, selected compounds are tested at 11 concentrations (0.1,

0.3, 1, 3, 10, 30, 100, 300 nM, 1, 3, 10 µM) and in triplicate. Counts (cpm/well) are pooled and fitted to a three-parameter logistic function for competition binding determine IC50 values where Y is total binding in the presence of corresponding concentration of test drug (X). Total and nonspecific binding are as described before.

(�� − �� ) � = �� + 1 + 10()

The concentration at which the competing ligand displaces 50% or the radioligand gives the IC50 which is transformed to an absolute inhibition constant, Ki, using the Cheng-Prusoff equation where L is the radioligand concentration used in the competition binding assay and Kd is the radioligand equilibrium binding affinity determined in saturation binding assays (obtained from NIMH Psychoactive Drug

Screening Program) [62].

�� = � 1 + �

6.2 Library of Compounds Assessed

Studies began with assaying select ACPAs in competitive binding experiments against a comprehensive screen of GPCRs. These compounds were active in the DAT-

KO mice, reducing rigidity and promoting movement. In this assay, an investigational compound is measured by the amount of a pre-bound, tritiated (3H)-labeled ligand it can displace from the GPCR (Figure 23).

20000 10 mg/kg

30 mg/kg

16000 60 mg/kg

12000

8000

4000 Horizontal Activity (counts/1 h)

0 1 2 3 4 9 Compounds

Figure 23. ACPAs active in the DAT-KO mice, reducing rigidity and promoting movement. Horizontal activity of DAT-KO mice was measured against increasing doses of ACPAs 1, 2, 3, 4, and 9. Compounds were tested at 10, 30, and 60 mg/kg.

The following binding curves represent the target profile of the ACPAs with displayed affinity separated by receptor type. Inhibition constants (Ki) are listed next to the reference or test compound in parentheses in each figure:

6.2.1 Dopaminergic Targets

The relationship between dopaminergic neurotransmission in the striatum and normal locomotion functions is well established, and progressive degeneration of DA

neurons within the substania nigra is known to cause PD. DA is a catecholamine which also belongs to the phenethylamine chemical family. DA regulates a variety of functions such as locomotion, feeding, emotion, and reward. The receptors within the dopaminergic system are either D1-like (D1R and D5R) and D2-like (D2R, D3R, and D4R)

DA receptors [63]. As previously stated, DA levels in the synapse are modulated by the

DAT. Target profiles are shown in Figures 24–25.

D1 2500

2000

1500

1000 (+)-Butaclamol

500 ACPA 4 SCH23390 Binding - CPM CPM Remaining 0 H] 3

[ -12 -10 -8 -6 -4 Log [ligand]

Figure 24. Secondary binding data showing the amount of [3H]-SCH23390 (Kd = 0.89 ± 0.12 nM) bound to D1 as a function of the competing, investigational ligand concentration: (+)-Butaclamol (4.8 nM); ACPA 4 (> 10,000 nM).

DAT 2000

1500

1000 GBR12909 ACPA 3 500 ACPA 9 WIN35428 Binding - CPM CPM Remaining 0 H] 3

[ -12 -10 -8 -6 -4 Log [ligand]

Figure 25. Secondary binding data showing the amount of [3H]-WIN35428 (Kd = 3.04 ± 0.21 nM) bound to DAT as a function of the competing, investigational ligand concentration: GBR12909 (3.04 ± 0.21 nM); ACPA 3 (1281 nM); ACPA 9 (1181 nM).

6.2.2 Serotonergic Targets

Similar to DA, 5-HT is also a monoamine derived from an amino acid. L-

Tryptophan is used to biosynthesis 5-HT rather than L-tyrosine, thus bares the aromatic indole. 5-HT is produced in the central nervous system, specifically in the Raphe nuclei located in the brainstem. The Raphe nuclei functions mainly to release 5-HT to the rest of the brain, and neuronal projections from this site terminate in the striatum. 5-HT receptors mediate 5-HT activity and can modulate many biological processes which effect cognition (memory and learning), mood (anxiety, depression, and aggression), consciousness, appetite, neurotransmitter release of itself and others (5-HT, DA, NE, and

Glu), and even cardiovascular systems (pulmonary hypertension and vasoconstriction).

There are seven receptors within in system and are as follows with the receptor isoforms in parentheses; 5-HT1 (5-HT1A, 5-HT1B, 5-HT1D, 5-HT1E, and 5-HT1F), 5-HT2 (5-HT2A, 5-

HT2B, and 5-HT2C), 5-HT3, 5-HT4, 5-HT5 (5-HT5A and 5-HT5B), 5-HT6, and 5-HT7. The serotonin transporter (SERT) is another monoamine transporter which functions similarly to the DAT [7, 64, 65]. Target profiles are shown in Figures 26–35.

5-HT1A

1500 8-OH-DPAT ACPA 1 1000 ACPA 3 500 ACPA 4 ACPA 9 0 WAY100635 Binding CPM CPM Remaining - -13 -11 -9 -7 -5 H] 3 [ Log [ligand]

Figure 26. Secondary binding data showing the amount of [3H]-WAY100635 (Kd = 0.43 ± 0.06 nM) bound to 5-HT1A as a function of the competing, investigational ligand concentration: 8-OH-DAT (0.6 ± 0.10 nM); ACPA 1 (574 nM); ACPA 3 (2035 nM); ACPA 4 (836.5 nM); ACPA 9 (190.67 nM).

5-HT1B 500

Ergotamine 250 ACPA 2 CT Binding - ACPA 3 H]5 3 CPM CPM Remaining

[ ACPA 4 0 ACPA 9 -14 -12 -10 -8 -6 -4 Log [ligand]

Figure 27. Secondary binding data showing the amount of [3H]5-CT (Kd = 1.91 ± 0.29 nM) bound to 5-HT1B as a function of the competing, investigational ligand concentration: Ergotamine (4.75 ± 0.51 nM); ACPA 2 (663 nM); ACPA 3 (466 nM); ACPA 4 (489 nM); ACPA 9 (726.5 nM).

5-HT1D

1500

Ergotamine 1000 ACPA 2 CT Binding - 500 ACPA 3 H]5 3 CPM CPM Remaining

[ ACPA 4 0 ACPA 9 -12 -10 -8 -6 -4 Log [ligand]

Figure 28. Secondary binding data showing the amount of [3H]5-CT (Kd = 1.47 ± 0.24 nM) bound to 5-HT1D as a function of the competing, investigational ligand concentration: Ergotamine (5.04 ± 0.47 nM); ACPA 2 (1,008 nM); ACPA 3 (2123 nM); ACPA 4 (1255 nM); ACPA 9 (1125.5 nM).

5-HT1E

1500

1000

HT Binding 5-HT - 500 ACPA 3 H]5 3 CPM CPM Remaining [ 0 -12 -10 -8 -6 -4 Log [ligand]

Figure 29. Secondary binding data showing the amount of [3H]5-HT (Kd = 3.36 ± 0.46 nM) bound to 5-HT1E as a function of the competing, investigational ligand concentration: 5-HT (11.8 ± 1.2 nM); ACPA 3 (4467 nM).

5-HT2A 1000

500 Clozapine ACPA 2 ACPA 3 Ketanserin Binding -

CPM CPM Remaining 0 H]

3 -12 -10 -8 -6 -4 [ Log [ligand]

Figure 30. Secondary binding data showing the amount of [3H]-Ketanserin (Kd = 1.57 ± 0.26 nM) bound to 5-HT2A as a function of the competing, investigational ligand concentration: Clozapine (9.8 ± 1.03 nM); ACPA 2 (1585 nM); ACPA 3 (1350 nM).

5-HT2B 2000

1500 SB 206553 ACPA 1 1000 ACPA 2 LSD LSD Binding - 500 ACPA 3 H] 3 CPM CPM Remaining [ 0 ACPA 4 -12 -10 -8 -6 -4 ACPA 9 Log [ligand]

Figure 31. Secondary binding data showing the amount of [3H]-LSD (Kd = 1.04 ± 0.15 nM) bound to 5-HT2B as a function of the competing, investigational ligand concentration: SB 206553 (18.2 ± 2.3 nM); ACPA 1 (1,800 nM); ACPA 2 (351 nM); ACPA 3 (535 nM); ACPA 4 (1098.67 nM); ACPA 9 (131.67 nM).

5-HT2C 3500 3000 2500 Ritanserin 2000 ACPA 2 1500 1000 ACPA 3 500 ACPA 4 0 Mesulergine Binding CPM CPM Remaining

- ACPA 9 -12 -10 -8 -6 -4 H] 3 [ Log [ligand]

Figure 32. Secondary binding data showing the amount of [3H]-LSD (Kd = 2.92 ± 0.83 nM) bound to 5-HT2C as a function of the competing, investigational ligand concentration: Ritanserin (2.12 ± 0.22 nM); ACPA 2 (923 nM); ACPA 3 (224 nM); ACPA 4 (2528 nM); ACPA 9 (498 nM).

5-HT6 600

400 Clozapine ACPA 2

LSD LSD Binding 200 - ACPA 3 H] 3 CPM CPM Remaining [ 0 ACPA 9 -12 -10 -8 -6 -4 Log [ligand]

Figure 33. Secondary binding data showing the amount of [3H]-LSD (Kd = 3.97 ± 0.46 nM) bound to 5-HT6 as a function of the competing, investigational ligand concentration: Clozapine (4.2 nM); ACPA 2 (441 nM); ACPA 3 (1538 nM); ACPA 9 (3581 nM).

100

5-HT7 1500

Clozapine 1000 ACPA 1 ACPA 2

LSD LSD Binding 500 - ACPA 3 H] CPU Remaining 3 [ 0 ACPA 4 -12 -10 -8 -6 -4 ACPA 9 Log [ligand]

Figure 34. Secondary binding data showing the amount of [3H]-LSD (Kd = 5.30 ± 0.70 nM) bound to 5-HT7 as a function of the competing, investigational ligand concentration: Clozapine (25 nM); ACPA 1 (809 nM); ACPA 2 (1001 nM); ACPA 3 (774 nM); ACPA 4 (423 nM); ACPA 9 (151.67 nM).

SERT

1500

1000 Amitriptyline ACPA 1 500 ACPA 4

Citalopram Binding ACPA 9 - CPM Remaining 0

H] -12 -10 -8 -6 -4 3 [ Log [ligand]

Figure 35. Secondary binding data showing the amount of [3H]-Citalopram (Kd = 3.8 ± 2.1 nM) bound to SERT as a function of the competing, investigational ligand concentration: Amitriptyline (4.46 ± 0.55 nM); ACPA 1 (> 10,000 nM); ACPA 4 (4575 nM); ACPA 9 (2529 nM).

101

6.2.3. Adrenergic Targets

Norepinephrine (NE) has the same chemical structure as DA with an additional b-hydroxyl substituent. This added chemical feature yields unique biological responses such as alertness, arousal, and readiness. The locus coeruleus is the primary source of

NE in the brain with many projections, including the thalamus which is responsible for relaying motor signals. PD-patients experience tremors due to the over-inhibition on the thalamus caused by the death of dopaminergic neurons in the SN. These receptors are divided in two main groups, a or b. The a group is further categorized as a1 or a2, each with three subtypes: a1A, a1B, a1D, a2A, a2B, and a2C, and the b group also has three subtypes: b1, b2, and b3. The norepinephrine transporter (NET) serves as the monoamine transporter for NE [7, 11, 66-68]. Target profiles are shown in Figures 36–40.

a1D 3000 2500 2000 1500 Prazosin 1000 ACPA 3

Prazosin Binding 500 - CPM CPM Remaining

H} 0 3 [ -13 -11 -9 -7 -5 Log [ligand]

Figure 36. Secondary binding data showing the amount of [3H]-Prazosin (Kd = 0.89 ± 0.14 nM) bound to a1D as a function of the competing, investigational ligand concentration: Prazosin (1.04 ± 0.10 nM); ACPA 3 (> 10,000 nM).

102

a2A 2000

1600 Oxymetazoline 1200 ACPA 1 800 ACPA 2

400 ACPA 3

0 ACPA 4 CPM CPM Remaining Rauwolscine Binding - -12 -10 -8 -6 -4 ACPA 9 H] 3 [ Log [ligand]

Figure 37. Secondary binding data showing the amount of [3H]-Rauwolscine (Kd = 0.5– 1.0 nM) bound to a2A as a function of the competing, investigational ligand concentration: Oxymetazoline (5.15 ± 0.21 nM); ACPA 1 (65 nM); ACPA 2 (757 nM); ACPA 3 (833 nM); ACPA 4 (696.5 nM); ACPA 9 (362 nM).

a2B 875

700 Yohimbine 525 ACPA 1 350 ACPA 2

175 ACPA 3

0 ACPA 4 CPM CPM Remaining Rauwolscine Binding - -12 -10 -8 -6 -4 ACPA 9 H] 3 [ Log [ligand]

Figure 38. Secondary binding data showing the amount of [3H]-Rauwolscine (Kd = 1.36 ± 0.50 nM) bound to a2B as a function of the competing, investigational ligand concentration: Yohimbine (12 nM), ACPA 1 (82 nM); ACPA 2 (1,171 nM); ACPA 3 (680 nM); ACPA 4 (248 nM); ACPA 9 863.67 nM).

103

a2C 600

450 Oxymetazoline 300 ACPA 2 ACPA 3 150 ACPA 4 0 CPM CPM Remaining Rauwolscine Binding

- ACPA 9 -12 -10 -8 -6 -4 H] 3 [ Log [ligand]

Figure 39. Secondary binding data showing the amount of [3H]-Rauwolscine (Kd = 0.73 ± 0.24 nM) bound to a2C as a function of the competing, investigational ligand concentration: Oxymetazoline (0.73 ± 0.24 nM); ACPA 2 (376 nM); ACPA 3 (229 nM); ACPA 4 (3717.5 nM); ACPA 9 (1427 nM).

NET 700

525 Desipramine ACPA 1 350 ACPA 2 175 ACPA 3 Nisoxetine Binding - CPM CPM Remaining 0 ACPA 4 H]

3 -12 -10 -8 -6 -4

[ ACPA 9 Log [ligand]

Figure 40. Secondary binding data showing the amount of [3H]-Nisoxetine (Kd = 4.9 ± 0.7 nM) bound to NET as a function of the competing, investigational ligand concentration: Desipramine (3.16 ± 0.28 nM), ACPA 1 (306 nM); ACPA 2 (5966 nM); ACPA 3 (316 nM); ACPA 4 (254 nM); ACPA 9 (1394 nM).

104

6.2.4. Histaminergic Targets

Histamine is the last of the classical monoamine neurotransmitters, and as similarly biosynthesized from an amino acid (L-histidine). The tuberomammillary nucleus is a histaminergic cell bundle location in the hypothalamus and in the sole source of histamine in the brain. Projections from this nucleus reach the cerebral cortex, hippocampus, striatum, nucleus accumbens, and the amygdala. There are four histamine receptors (H1, H2, H3, and H4), however H4 is not expressed in the brain.

Unlike the aforementioned monoamines, histamine lacks a specialized transporter [7, 11,

69]. Target profiles are shown in Figures 41.

H1 1500

1000

Chlorpheniramine 500 ACPA 3 Pyrilamine Binding - CPM CPM Remaining 0 H]

3 -12 -10 -8 -6 -4 [ Log [ligand]

Figure 41. Secondary binding data showing the amount of [3H]- Pyrilamine (Kd = 1.01 ± 0.13 nM) bound to H1 as a function of the competing, investigational ligand concentration: Chlorpheniramine (3.22 ± 0.33 nM), ACPA 3 (> 10,000 nM).

105

6.2.5 Non-Monoamine Targets

Acetylcholine (ACh) is the ester of acetic acid and choline, and is the only neurotransmitter derived from a vitamin (acetyl-COA). Nicotinic acetylcholine receptors

(nAChRs) are ionotropic (ligand-gated ion channels) and respond to nicotine, while muscarinic acetylcholine receptors (mAChRs) are metabotropic (act through a second messenger) and respond to muscarine. 17 subunits have been identified. nAChRs are pentamers formed by these subunits and modify the membrane potential of neurons by moving positively charged ions across its channel. mAChRs are GPCRs and consist of the sub-types M1, M2, M3, M4, and M5. [7, 11, 70]. Target profiles are shown in Figures 42.

M5 1500

1000

Atropine

QNB Binding 500 - ACPA 4 H] CPM CPM Remaining 3 [ 0 -12 -10 -8 -6 -4 Log [ligand]

Figure 42. Secondary binding data showing the amount of [3H]-QNB (Kd = 0.6 ± 0.13 nM) bound to M5 as a function of the competing, investigational ligand concentration. Atropine (0.88 ± 0.08 nM), ACPA 4 (> 10,000 nM).

106

The k-opioid receptor (KOR) is one of four opioidergic receptors which bind opioid-like compounds modulating mood, nociception, consciousness, and motor control [70-73]. Target profiles are shown in Figures 43.

1000 27) - 07 -

500 Salvinorin A

Binding ACPA 4 U69593 (2007 - CPM CPM Remaining 0 H]

3 -12 -10 -8 -6 -4 [ Log [ligand]

Figure 43. Secondary binding data showing the amount of [3H]-U69593 (Kd = 1.93 ± 0.45 nM) bound to KOR as a function of the competing, investigational ligand concentration. Salvinorin A (1.93 ± 0.45 nM), ACPA 4 (> 10,000 nM).

Sigma (s1 and s2) receptors were once believed to be part of the opioiodergic class of receptors since the opioid drug benzomorphan also has activity against these receptors. Once the s1 receptor was isolated and cloned, it was evident sigma receptors bare no structural similarity to opioid receptors [74]. Target profiles are shown in

Figures 44–45.

107

500

Haloperidol ACPA 2 ACPA 4

Pentazocine Binding 0 CPM CPM Remaining - -12 -10 -8 -6 -4 H] 3 [ Log [ligand]

Figure 44. Secondary binding data showing the amount of [3H]-Pentazocine (Kd = 4.45 ± 0.55 nM) bound to s1 as a function of the competing, investigational ligand concentration: Haloperdiol (3.29 ± 0.50 nM); ACPA 2 (453 nM); ACPA 4 (1865 nM).

1000

Haloperidol 500

DTG Binding ACPA 2 - H] CPM CPM Remaining 3 ACPA 9* [ 0 -12 -10 -8 -6 -4 Log [ligand]

Figure 45. Secondary binding data showing the amount of [3H]-DTG (Kd = 10.08 ± 1.38 nM) bound to s2 as a function of the competing, investigational ligand concentration: Haloperdiol (24.7 ± 1.8 nM); ACPA 2 (385.5 nM); ACPA 9 (1838 nM).

108

6.3 Pharmacophoric Considerations

As previously mentioned, additional phenethylamine derivatives may yield additional insight to the mechanism of action for the rescuing phenotype. While the paradigm of structure dictating function is well established, it is not necessarily always understood in a given system. As such, additional derivatives of the phenethylamine class of molecules will aid identification of pharmacophoric elements already present within arylcyclopropylamines, and which elements can be modified to enhance activity.

These compounds have yet to be evaluated in the DAT-KO mouse phenotypic screen, however comparison of the GPCR profiles of ACPAs active in the mice will serve as the basis of potential phenotypes.

6.3.1 Additional ACPAs Evaluated

Two additional ACPAs were submitted to the competitive binding assay as well

(Figure 46). These ACPAs are N-alkylated and will provide insight into activity as a function of amine structure (i.e. substituents of the amine nitrogen).

Figure 46. trans-(N)-Alkylated arylcyclopropylamines 66 and 67.

109

ACPA 66 is the monomethylated variant of tranylcypromine, also analogous to amphetamine that is conformationally-constrained. ACPA 67, also known as GSK-LSD1, bares an N-piperidine, and is a potent inhibitor of lysine specific demethylase 1 (LSD1).

Previous work by our Group and others has well characterized LSD1, and ACPAs’ potent inhibitory action as an irreversible conduct is formed upon binding of an ACPA and subsequent opening the cyclopropyl ring [75-77]. GPCR affinity profiles for these

ACPAs are listed in Tables 7–8.

Table 7. GPCR profile for ACPA 66. Targets in common with ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*).

Target Ki *5-HT2B 1527 nM *5-HT7 4672 nM SERT 1226 nM *a2A 915 nM a2C 550 nM *NET 4,539 nM

Table 8. GPCR profile for ACPA 67. Targets in common with ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*).

Target Ki 5-HT2A 8,675 nM SERT 999 nM H1 899 nM

110

6.3.2 Phenethylamines Evaluated

Similarly, phenethylamine derivatives were assayed as well to ascertain the importance of the cyclopropyl ring (Figure 39). GPCR affinity profiles for these ACPAs are listed in Tables 9–13.

Figure 47. Phenethylamine derivatives assessed in addition to the ACPAs.

Table 9. GPCR profile for 42 (analog to ACPA 1). Targets in common with ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*).

Target Ki 5-HT1A 232 nM 5-HT1D 2091 nM *5-HT2B 1579 nM *5-HT7 295 nM *a2A 85 nM a2B 170 nM

111

Table 10. GPCR profile for 43 (analog to ACPA 2). Targets in common with ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*).

Target Ki DAT 2525 nM 5-HT1B 400 nM 5-HT1D 1196 nM *5-HT2B 134.5 nM 5-HT2C 1008 nM *5-HT7 317 nM *a2A 442 nM a2B 291 nM a2C 147 nM

Table 11. GPCR profile for 44 (analog to ACPA 3). Targets in common with ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*).

Target Ki D4 2,408 nM DAT 1162 nM 5-HT1A 803 nM 5-HT1B 82 nM 5-HT1D 550 nM 5-HT1E 673 nM *5-HT2B 109 nM 5-HT2C 412 nM *5-HT7 55 nM *a2A 178 nM a2B 284 nM a2C 73 nM H1 312 nM

112

Table 12. GPCR profile for 45 (analog to ACPA 4). Targets in common with ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*).

Target Ki *5-HT2B 1,159.50 nM *5-HT7 2049 nM *a2A 902 nM a2B 2800 nM

Table 13. GPCR profile for 46 (analog to ACPA 1). Targets in common with ACPAs 1–4, and 9 (see Figure 23) are indicated with an asterisks (*)

Target Ki a1D > 10,000 nM a2B 3649 nM H1 1003 nM

113

6.4 Conclusions

Efforts to identify candidates targets of the ACPAs active in the DAT-KO mouse study using the radioligand binding competition assay yielded complex profiles of receptors belonging to many subtypes of the GPCRome. Since the DAT-KO mice lack the DAT and functional levels of DA, a mechanism of action that is not mediated by dopaminergic receptors or the dopaminergic transporter would be expected. Indeed, we only observed that ACPA 4 had an affinity towards a dopaminergic receptor and that

ACPA 3 and 9 bound to the dopamine transporter; all of which exhibited poor affinities in the micromolar range. When considering serotonergic targets, only the 5-HT1A and 5-

HT2B subtypes of the and 5-HT1 and 5-HT2 receptors have been previously implicated as therapeutic targets for PD in treating motor and nonmotor (mood) symptoms [18-19].

Though, the number of serotonergic receptors present in each ACPA’s profile observed may explain toxicity seen in the rats. While ACPA 9 was effective in the 6-OHDA- lesioned rats, toxicity was observed at higher doses (> 9 mg/kg). We suspect the toxicity is caused by an overaccumulation of serotonin (serotonin syndrome). This event is known to occur by a number of mechanisms, most notably by inhibition of serotonin reuptake or activation of serotonergic receptors, particularly 5-HT6 [31]. 5-HT6 is likely contributing to toxicity and an undesired off-target. Previous studies also implicate 5-

HT7, like 5-HT1A aforementioned, activation helps treat mood disorders associated with

PD.

114

Previous studies have demonstrated ACPA activity against the norepinephrine transporter, so it was unsurprising to observe affinities for this target. Although this effect is seen at higher does, it is interesting to note that the NET was the only transporter all ACPAs tested had in common in their affinity profile [32]. Additionally,

ACPAs bound a2 receptors but had little to no affinity to any subtype of the a1 receptors.

Interestingly, a2A and a2c are implicated as targets eliciting anti-Parkinsonian activity [21,

33]. Poor affinities (> 10,000 nM = > 10 µM) were also observed at receptors H1, M5, and k-opioid. As such, these can be eliminated with confidence as targets which elicit the restorative phenotype. Of the sigma receptors, sigma-1 activation is noted to be neuroprotective in animal models of Alzheimer's disease (AD), amyotrophic lateral sclerosis (ALS), and stroke. These effects have been attributed to an attenuation of excitotoxicity and apoptosis by multiple mechanisms, such as up-regulation of protective genes, reduced microglial activation, reduced generation of reactive oxygen species. Since mitochondrial dysfunction and nitrosative stress are critical mediators of nigrostriatal dopamine damage, and neuroinflammation is known to contribute to neurodegeneration in parkinsonian disorders. Sigma-1 may have potential as a new target for disease-modifying therapy in PD [30], but not as an ACPA target.

When comparing the affinities of the ACPA 66 and 67, and the PEA derivatives

42–46, additional information can be gathered for subsequent library design. ACPA 1

(tranylcypromine) and 42 (phenethylamine) exhibited very similar profiles, only

115

differing in sub-types of receptor classes. NET is noted to be missing from 42’s profile as it is with the other four PEAs (43–46) as well. This indicates the cyclopropyl ring is required for activity, as well as its positioning in the ACPA scaffold, as phenethylamine

46’s cyclopropyl substituent is location at the b-position of the ethylene bridge. Another important note is the number of GPCRs within both ACPA 3 and PEA 44 which both possess a para-bromo substituent. A massive GPCR profile suggests numerous off-target leading to undesirable side-effects. Subsequent libraries should not select bromo and likely other para-halogenated derivatives.

Taken altogether, this assay provided 5-HT2B, a2A, and the NET as candidate targets for the ACPA-drug class. While the radiolabel binding competition assay generated a profile of receptors for the select ACPAs (good activity in DAT-KO mice), it can only provide affinity measured as an inhibition constant (Ki), that is how well it can inhibit binding of the radioligand in equilibrium with the receptor. Compounds which exhibit long residence time (t), particularly slow off-rates (koff) can skew results if assay conditions are not adjusted for kinetics of the radioligand. Fortunately, many of the rate constants and residence times have been experimentally calculated and can be considered in advanced stages of generations of lead-compound optimization.

Functional assays to deduce efficacy of the ACPAs will further illuminate these compounds mode of action and will be discussed in the following.

116

6.5 Experimental

Drugs and chemicals. (±)-Arylcyclopropylamines derivatives were synthesized as previous described. All reference chemicals and radiolabeled chemicals were graciously provided by the National Institute of Mental Health's Psychoactive Drug

Screening Program, Contract # HHSN-271-2013-00017-C (NIMH PDSP).

Membrane preparation and binding assay. To make membrane fractions from stably transfected cell lines, cells are subcultured in 15-cm dishes and grown to 90% confluency. For 5-HT receptors, cells are incubated overnight in either serum-free medium or medium containing 1% dialyzed FBS before harvesting. The next day, the cells are rinsed with PBS, scraped off into 50 ml conical tubes, and pelleted by centrifugation (1000 x g, 10 min at 4 °C). The cell pellet is resuspended in chilled (4 °C) lysis buffer (50 mM Tris HCl buffer, pH 7.4) and triturated gently for hypotonic lysis.

The suspension is then centrifuged at 21,000 x g for 20 min at 4 °C to obtain a crude membrane fraction pellet. The fresh membrane pellet is then resuspended in 3 volumes of cold lysis buffer and is immediately subjected to the Bradford protein assay to determine protein concentration, followed by a saturation binding assay (see following section for detail) to determine he receptor expression level of the receptor and the affinity of the selected radioligand. Based on the receptor expression level and the Kd value, the fresh membrane suspension is stored at -80 °C freezer in small aliquots, such

117

that one aliquot is sufficient for one 96-well plate to have at least 500 cpm/well when assayed at 0.5–1.0x Kd value of the appropriate radioligand.

118

7. Mechanistic Efficacy Studies

GPCR ligands with distinct functional selectivity patterns will be extremely useful tools for elucidating the key signaling pathways essential for both the therapeutic actions and the side effects of GPCR targeted drugs. G-protein functional assays seek to measure a ligand’s efficacy for a G-protein mediated pathway, often taking advantage of the measurable changes in levels of second messengers resulting from activation of the receptor (i.e. cAMP, IP3, DAG, Ca2+). A comparable secondary messenger for b-arrestin has not been identified, thus requires generation of expression fusion proteins which are expressed as a function receptor activation and b-arrestin is recruitment. The catalysis of the exogeneous reporter-substrate produces a detectable and quantifiable chemiluminescent signal.

7.1 G-Protein Activation Assays

Various functional assays to measure G-protein activation are well established for each Ga subunit and measure changes in intracellular cAMP (Gs and Gi) and intracellular Ca2+ (Gq) in various well formats.

7.1.1 Calcium Flux Assay

Experiments began with assaying 5-HT2B in Chinese hamster ovary (CHO) cells.

Assays began once the receptor was stably-transfected using a lipofectamine protocol. 5-

119

HT2B was of greater interest due to localization and function in the brain [111]. However, with such as large scope of potential GPCR target to assay, an assay in which a GPCR can be tested regardless of the G-protein(s) it natively binds is advantageous to streamline productivity, lower costs, and reduce risk. G15 (& G16) are G-proteins found outside of the brain which can promiscuously bind to most currently known and isolated GPCRs and transduce signaling similarly as Gq. In the calcium flux assay, an activated GPCR binds G15 and increases intracellular Ca2+ (iCa2+) just as Gq with native its receptors. The release of iCa2+ can be detected by Flur-4 which produces a fluorescent signal upon chelation of iCa2+ (Figure 48) [112].

120

Figure 48. Visual representation depicting experimental design and the reaction which Flur-4 AM is metabolized by esterases after diffusing the through the plasma membrane into the cytosol. Flur-4 is then able to chelate iCa2+ which is detectable by a fluorescent plate reader.

The calcium flux assay has power to determine if a compound is an agonist

(activates the receptor) or an antagonist (precludes agonist binding, effectively inactivating the receptor. To determine the mode of binding, agonism and antagonism experiments performed with again select ACPAs 1–4, and 9 (Chapter 6, Figure 23).

HEK-293 cells are cultured in DMEM medium supplemented with 10% dialyzed

FBS and 1% pen/strep. HEK-293 cells stably expressing 5-HT2B were generated using lipofectamine 3000 and passaged at low-densities in the presence of G418 for selection.

121

HEK-293 cells stably-expressing 5-HT2B are plated into Poly-L-Lysine (PLL) coated 96- well black clear bottom cell culture plates with DMEM supplemented with 0% FBS and at density of 30–50,000 cells in 100 µl per well. The plates are cultured overnight before assays. On the day of the assay, media is removed, and cells washed then are loaded 100

µl/well of Fluo-4 Direct Calcium dye (Molecular Devices). In antagonism experiments, the investigational ligand is added as well as at a fixed concertation per well. Plates are incubated for 60 minutes at 37 °C, followed by a 10 minutes incubation at room temperature in the dark, and then loaded in the plate reader (Spectramax i3x equipped with an injector). The plate reader is programmed to take 10 readings (1 read per second) first as a baseline before addition of 100 µl of 2X drug solutions. The fluorescence intensity is recorded for 2 minutes after drug addition (first addition for agonist activity). Results from agonism and antagonism experiments are shown in

Figures 49 and 50, respectively.

122

5-HT2B

500000

375000

250000 RFU

125000

0 5-HT DMSO 1 2 3 4 9 Compound (10 uM)

Figure 49. Primary agonism data. Relative fluorescence units (RFU) is measures as a function of 5-HT2B activation by a given ACPA in stably-transfected CHO cells.

5-HT2B

500000

375000

250000 RFU

125000

0 1 2 3 4 9 Compounds (10 uM)

Figure 50. Primary agonism data. Relative fluorescence units (RFU) is measures as a function of 5-HT2B activation by a given ACPA in stably-transfected CHO cells. ACPA performance was compared to the 5-HT control in the agonism experiment (Figure 49).

123

7.2 b-Arrestin Recruitment Assays

Functional assays to measure b-arrestin activation are well established. The

PRESTO-Tango assay is a recently developed functional assay developed by the Roth group [81]. This assay similar to the Tango assay, allows assessment of test compounds independent of G-protein signaling in parallel against the available GPCRome (> 300 receptors). Upon activation of the GPCR by a ligand, a β-arrestin-TEV protease fusion protein is recruited to the c-terminus of the receptor. This is followed by cleavage by the fusion protein at the TEV protease–cleavage site. Cleavage results in the release of the tetracycline transactivator (tTA) which diffuses to the nucleus and activates transcription of the luciferase reporter gene. Luciferase then catalyzes exogenously introduced, D-luciferin which is detectable by a luminescence plate reader (Figure 51).

124

Figure 51. Visual representation depicting experimental design and the reaction which luciferase catalyzes.

Efforts to perform secondary assays in which luminescence is measured as a function of ligand concentration proved difficult. Endogenous agonist assays against their native receptors were not able to produce a typical dose-response curve as expected, therefore assays could not be performed with ACPAs with confidence. We suspected this issue is due to a low transfection efficiency. HTLA cells were transfected using the Ca3(PO4)2 protocol with 20, 40, and 80 ng of DNA (Tango plasmid with a

GPCR insert) per well of 30,000 cells in a 96-well plate and yielded the same result. This

125

differs from the PRESTO-Tango protocol in 20 ng of DNA per well in 384-well plates of

15,000 cells. While the DNA, cells, and plating volume were scaled to appropriates amounts conducive for a 96-well plate, the surface area (area in which cells are exposed to media) differences between plate sizes and can affect transfection efficiency.

Nonetheless, scaling the assay has proven to not be amendable to scaling of any sort in our hands currently. For immediate experimentation, we will have to continue to use the

384-well format, but optimization will continue to enhance the efficiency of the assay for a smaller scale of receptor targets. Stable transfection may resolve this issue as the receptor would be constituently expressed at the membrane. These stably-transfected cells would have to be evaluated to ensure receptors are functional and do not perform differently than transiently-transfected cells.

126

Receptors that displayed a significant fold increase in activation (≥ 2) over baseline-control measurements by ACPAs are considered significant. Targets without data shown did not display significant activation. This visualization better depicts the selectivity of ACPAs target as a function of the aryl substituent. ACPAs 1–4, and 9 were assessed in comparison as done previously (Chapter 6, Figure 23). The following charts represent the target profile of the ACPAs with displayed efficacy separated by receptor type (shown as their gene name) (Figures 52–56).

1 8

2 Fold Change Raio 0

DRD2 GRPR CHRM4 CXCR1 HTR1D OPRM1 ADRA1BADRA2BADRA2C

Figure 52. Receptors profile for ACPA 1. GPCRs that displayed a significant fold increase in activation (≥ 2) over baseline-control measurements by ACPA 1 are shown in blue. Targets without data shown did not display significant activation.

127

2 6

2 Fold Change Ratio 0

DRD2 GRPR CHRM4 CXCR1 HTR1D OPRM1 ADRA1BADRA2BADRA2C

Figure 53. Receptors profile for ACPA 2. GPCRs that displayed a significant fold increase in activation (≥ 2) over baseline-control measurements by ACPA 2 are shown in green. Targets without data shown did not display significant activation.

128

3 6

2 Fold Change Ratio 0

DRD2 GRPR CHRM4 CXCR1 HTR1D OPRM1 ADRA1BADRA2BADRA2C

Figure 54. Receptors profile for ACPA 3. GPCRs that displayed a significant fold increase in activation (≥ 2) over baseline-control measurements by ACPA 3 are shown in orange. Targets without data shown did not display significant activation.

129

4 6

2 Fold Change Ratio

DRD2 GRPR CHRM4 CXCR1 HTR1D OPRM1 ADRA1B ADRA2B ADRA2C

Figure 55. Receptors profile for ACPA 4. GPCRs that displayed a significant fold increase in activation (≥ 2) over baseline-control measurements by ACPA 4 are shown in purple. Targets without data shown did not display significant activation.

130

9 6

2 Fold Change Ratio

DRD2 GRPR CHRM4 CXCR1 HTR1D OPRM1 ADRA1BADRA2BADRA2C

Figure 56. Receptors profile for ACPA 9. GPCRs that displayed a significant fold increase in activation (≥ 2) over baseline-control measurements by ACPA 9 are shown in red. Targets without data shown did not display significant activation.

7.2.2 Dopaminergic Targets

D2 activation was seen with TCP (ACPA 1) and ACPA 9. Within the context of

PD functions as an autoreceptor, providing negative feedback to neurons in the striatum is critical. In the disease-sate. D2 receptors in the SN are lost and coupling to Gai is enhanced, super-sensitizing remaining D2 receptors. It is now understood β-arrestins regulate physiology and behaviors independently of G-protein signaling through their ability to scaffold multiple intracellular signaling molecules such as kinases and phosphatases [82-85]. Studies have shown that through both dopamine D1 and D2 receptors, βarr2-mediated signaling plays a major role in DA-dependent locomotion [86-

131

88], while studies in animal models of PD suggest that dyskinesias are associated with enhanced G-protein mediated signaling at dopamine receptors [89-95] potentially leading to changes in gene expression and uncontrolled neuronal excitability [93, 96-

100]. This super-sensitization is the cause for L-DOPA induced dyskinesias treated.

7.2.2 Serotonergic Targets

The ACPA scaffold showed moderate binding affinity against 5-HT1 and 5-HT2 receptors in the previous competition assay, in particularly 5-HT1A and 5-HT2B. Within the b-arrestin assay, we only observed inconsistent activity at 5-HT1 receptor and no activity at the 5-HT2 receptors. This suggests our compounds may be functionally selective and do not mediate by b-arr-dependent signaling for serotonergic receptors

(Figure 44).

7.2.2 Adrenergic Targets

Similarly, the ACPA scaffold showed moderate binding affinity against a2 receptors. However, within the b-arrestin assay, we observed inconsistent activity at these receptors, suggesting our compounds are not mediated by b-arrestin-dependent signaling either (Figure 45).

132

7.2.2 Non-Monoamine Targets

Receptors from other classes observed affinity and activation as well. Of the muscarinic receptors (Figure 46), only M4 has been implicated as target for PD [101]. M4 inhibitory output to the striatum, inhibits DA action of D1, which may prove effective in reducing over-stimulation of the direct pathway. At this time, it is not clear if these receptors are significant in their contribution to the phenotype of alleviating akinesia or off-targets. Compounds 1, 4, and 9 activated gastrin-releasing peptide receptor (BB2,

Figure 47). This receptor will be of interest in further studies as previous work has shown antagonism of this receptor blocks D-amphetamine-induced hyperlocomotion

[102]. Compounds 1, 2, and 3 activated the µ-opioid receptor as well. The receptor comes from the opioid peptide receptor class which consists of DOR and KOR as well. This receptor is known primarily for its analgesic and sedative properties but has also been in implicated in PD’s. More specifically, COMT gene rs6267 allele “T” has been associated with pain from dystonia in PD patients [103]. Potent striosomal opioid signaling within a dopamine-depleted disease state has also been observed in 6-OHDA-lesioned rat model for PD and malfunction within the opioid signaling contributes to the pathophysiology od levodopa-induced dyskinesias [73, 104, 105]. CXCR7, Y5, and SST5 are likely off-targets do not contribute to the phenotypic alleviation of akinesia observed in the DDD-mice and 6-OHDA-lesioned rats based on known functions and expression in the brain [106-110].

133

7.3 Pharmacophoric Considerations

As previously described (Chapter 6.3), ascertaining pharmacophoric elements of

ACPAs from inconsequential or deleterious elements will inform subsequent libraries

(Figure 57–61). ACPAs 66 and 67 were assayed as well, but ACPA 67 was able to significantly activate a receptor: a1B (2.1-fold change ratio).

42 25

5 Fold Change Ratio 0

DRD3 DRD4 HTR5 HTR1A ADRA2A ADRA2B ADRA2C

Figure 57. Receptors profile for PEA 42. GPCRs that displayed a significant fold increase in activation (≥ 2) over baseline-control measurements by PEA 42 are shown in blue.

134

43 25

Fold Change Ratio 5

0 DRD3 SSTR5

Figure 58. Receptors profile for PEA 43. GPCRs that displayed a significant fold increase in activation (≥ 2) over baseline-control measurements by PEA 43 are shown in green.

135

44 25

5 Fold Change Ratio 0

DRD2 DRD3 DRD4 HTR1B HTR1D HTR1E HTR2A ADRA2BADRA2C HTR2C VNV

Figure 59. Receptors profile for PEA 44. GPCRs that displayed a significant fold increase in activation (≥ 2) over baseline-control measurements by PEA 44 are shown in orange.

136

45 25

Fold Change Ratio 5

0 HTR2A HTR2C VNV

Figure 60. Receptors profile for PEA 45. GPCRs that displayed a significant fold increase in activation (≥ 2) over baseline-control measurements by PEA 45 are shown in purple.

137

46 25

5 Fold Change Ratio 0 ADRA1A ADRA1B ADRA2C DRD2 HTR1E HTR6

Figure 61. Receptors profile for PEA 46. GPCRs that displayed a significant fold increase in activation (≥ 2) over baseline-control measurements by PEA 46 are shown in grey.

7.4 Conclusions

The receptor profiles for ACPAs differed from binding and functional assays in that receptors which were shown to bind receptors in the competition assay were not activated in the functional b-arrestin assay. This suggests (1) ACPAs do not bind GCPRs to a manner which recruits b-arrestin, or (2) ACPAs are binding to an allosteric site on the receptors. Allosteric binding sites are located either proximally next to the orthosteric (canonical binding site) site or distally located elsewhere on the receptor binding curves. Binding and functional assays described thus far do have the power to measure allosteric modulation and further experiments will want to include experimental design to assess any potential activity which may arise from a mode of binding such as an allosteric binding event.

138

Additionally, the phenethylamine derivatives in these efficacy assays exhibited profiles which differed from their ACPA analogs; certainly, differing more than differences noted within the binding assays. This may be attributed to the lack of the cyclopropyl ring no longer imposing rigidity, as the phenethylamines are able to freely rotate about the a and b carbons (ethylene bridge) and can adopt a greater number of confirmations and reducing selectivity.

To date, 5-HT2B has solely been assayed with ACPAs in the calcium flux assay.

Primary assays demonstrated ACPAs do not activate this receptor to the effect 5-HT does, which may point to other forms of binding or activation. Agonism for this receptor has proven lethal within the heart and liver, while antagonism has proven to be protective. ACPAs as ligands for 5-HT2B may be advantageous in the CNS, but not in the periphery. Co-administration of 5-HT2B antagonists or designing ligands which bias signaling to solely therapeutics pathways, and may prove to be a viable strategy for treating PD-associated symptoms [46].

139

7.5 Experimental

Drugs and chemicals. (±)-Arylcyclopropylamines derivatives were synthesized as previous described. All other chemicals and reagents were purchased from Sigma.

PRESTO-Tango Assay. HTLA (an HEK293 cell line stably expressing a tTA- dependent luciferase reporter and a b-arrestin-TEV fusion gene) cells were graciously gifted by the Roth group. Cells are maintained in DMEM supplemented with 10% FBS and antibiotics (2 µg/ml puromycin and 100 µg/ml hygromycin B). GPCR constructs were purchased from Addgene provided by the Roth group. HTLA cells are transfected with GPCR tango constructs overnight and are plated in Poly-L-Lys (PLL) coated 384- well white clear-bottom cell culture plates at a density of 15,000 cells in 40 µl per well of

DMEM with 1% dFBS. The cells are incubated overnight for them to recover before receiving a drug. Drug stimulation solutions are prepared in filtered Tango assay buffer at 5x and added to cells (10 µl per well) for overnight. On the day of measurement, medium and drug solutions are removed and 20 µl per well of BrightGlo reagent

(diluted by 20-fold with Tango assay buffer) is added.

Calcium Flux Assay. HEK-293 cells are cultured in DMEM medium supplemented with 10% dialyzed FBS and 1% pen/strep. HEK-293 cells stably expressing 5-HT2B were generated using lipofectamine 3000 and passaged at low- densities in the presence of G418 for selection. HEK-293 cells stably-expressing 5-HT2B are plated into Poly-L-Lysine (PLL) coated 96-well black clear bottom cell culture plates

140

with DMEM supplemented with 0% FBS and at density of 30–50,000 cells in 100 µl per well. The plates are cultured overnight before the assay. The following day of the assay, media is removed, and cells are washed then are loaded with 100 µl/well of Fluo-4

Direct Calcium dye (Molecular Devices). In antagonism experiments, the investigational ligand is added as well as at a fixed concertation per well. Plates are incubated for 60 minutes at 37 °C, followed by a 10 minutes incubation at room temperature in the dark, and then loaded in the plate reader (Spectramax i3x equipped with an injector). The plate reader is programmed to take 10 readings (1 read per second) first as a baseline before addition of 100 µl of 2X drug solutions. The fluorescence intensity is recorded for

2 minutes after drug addition (first addition for agonist activity).

141

8. Conclusions

Presently, Parkinson’s disease continues to lessen the quality of life for all patients diagnosed. Aging populations and inevitable-to-fail treatments underscore the importance for new strategies and therapeutics.

8.1 Antagonism of Presynaptic Receptors Reinstates the Direct Pathway of Locomotion

Our top-down approach began with the phenotypic screen in mice. Select arylcyclopropylamines rapidly reduced rigidity and akinesia in the absence of dopamine, which phenocopied into the rat model for PD as well. This suggest (1)

ACPAs activity is independent of dopamine, and (2) these compounds are working within the direct pathway. As previously stated in Chapter 1, the direct pathway starts with an input to the motor cortex. An excitatory signal is sent to the striatum, which relays inhibitory signals to the SN and GPi which allows the thalamus to send an excitatory signal to the motor cortex to promote movement. Another note to consider is the raphe nuclei innervates to the SN but does not reside within the basal ganglia. And while the locus coeruleus innervates similarly, it does so to a lesser extent. The two receptors which correspond with neurotransmitter release of their respective ligand in the brain are 5-HT2B and a2A. Both of these receptors were identified in the competitive binding assay and displayed the strongest affinities. These two receptors are both presynaptic, autoreceptor which function in negative feedback mechanisms to stop the 142

over-release, and thus over-stimulation, of their respective neurotransmitter. This suggests ACPAs are antagonistic, blocking the binding and activation of these receptors by endogenous ligands, 5-HT and NE respectively. Once these receptors are inactivated, the presynaptic neuron is unable to inhibit more 5-HT or NE from being released into the synapse, prolonging action of the post-synaptic neuron (i.e. neurotransmission).

Anatomically speaking, the Raphe nuclei activates the SN which can then activate the striatum in the direct pathway while the locus coeruleus. The LC inhibits the subthalamic nucleus which stops inhibitory output to the GPi, stopping inhibitory output to the thalamus, and allowing excitatory output to the motor cortex (Figure 63).

Preliminary efficacy assays have demonstrated ACPAs are unable to activate 5-HT2B but can block action of 5-HT when an ACPA is pre-bound.

Figure 62. Visual representation of proposed mechanism of action of ACPAs in the DAT- KO mouse and 6-OHDA rat models for Parkinson’s disease.

143

8.2 Future Directions

Assays to address the role of the NET is still required. Each enantiomer of tranylcypromine is known to modulate monoamine transport to varying extents. ACPAs in this study will also be assayed as racemates initially, as they were done in the animal models. At this time, enantiomeric consequences of arylcyclopropylamines are not clear.

Chiral resolution either using chiral column chromatography or crystallization with a chiral acid will afford both enantiomers for individual assessment.

8.2.1 Structure-Based Small Molecule Design

With lead biological targets now identified, additional drug design principles can be employed to enhance the beneficially effects of ACPAs by incorporation of select chemical moieties while lessening the harmful side-effects. Crystal structures for a2A and

NET already exist and will be helpful to further characterize the binding event of

ACPAs to these targets. 5-HT2B has recently been crystallized lending more insight into the binding interactions and the prerequisite change in confirmation and structure which dictate G-protein coupling and function (Figure 64).

144

Figure 63. Crystal structure of the LSD-bound 5-HT2B receptor. PDB: 5TVN.

Ligand-based drug design will able be helpful in identifying chemical components of characterized ligands which we may or may not want incorporated into the next generation of compounds. If the expansion of the ACPA library reaches an appreciable number of compounds, quantitative structure-activity relationships will also provide valuable information, further describing the binding effect and the chemical features important for binding.

8.2.2 Further Characterization of ACPA Biological Activity

More efficacy assays will be necessary to fully characterize the effects of 5-HT2B and a2A activation by ACPAs. cAMP accumulation assays will be necessary for receptors which canonically couple to Gs and Gi. While the calcium flux assay can be used when using the G15/16 initially, measuring the cAMP accumulation as a function of ACPA concentration rather than intracellular calcium better models ACPA bioactivity since the 145

receptor is using its native Ga subunit. Receptor trafficking and internalization assays will provide information as to whether activation of target GPCRs will cause the internalization and possibly the destruction of the receptor by b-arrestin. The degree of activation correlates to the phosphorylation state of the receptor, and in-turn whether the receptor is (1) recycled back to the cellular membrane, or (2) targeted for lysosomal degradation. This will be critical information as either scenario could be advantageous.

If receptor agonism is mechanism of action which elicits a therapeutic effect, then receptor-recycling permits the availability of your target. If receptor antagonism is mechanism of action which elicits a therapeutic effect, then receptor degradation will remove the target, preventing its activation as well as lowering its expression level.

Bioluminescence resonance energy transfer (BRET) and fluorescence resonance energy transfer (FRET) assays to assess receptor dimerization could also be employed, however no data we gathered thus far indicates such a mechanism is taking place.

8.2.3 Animal Behavioral Studies

Once more characterization of ACPA’s bioactivity has been elucidated, additional behavioral studies in animal models for Parkinson’s disease will resume. The

DAT-KO mouse model will continue to serve as a phenotypic screen to identify derivatives with anti-akinetic activity and evaluate toxicity. Since the initial study done with the DAT-KO mice (see Chapter 2), this animal model has demonstrated its utility as

146

a model for PD. The Caron group has since presented evidence demonstrating the DAT-

KO mouse model can accurately mimic hallmarks of PD symptoms and even demonstrated genetic manipulation of β-arr2 affords a separation of the biology of the therapeutic and dyskinesia effect of L-DOPA. The DAT-KO model, which has been shown to have reproducible, non-variable, bilateral DA depletion, is also amenable to bilateral genetic manipulations and most importantly, can be subjected to automated robust behavioral screens to test therapeutic efficacy. Since mice are cheaper than rats to maintain, subsequent animal studies will be performed first in the DAT-KO mouse model before studies can be performed with higher-ordered animal models of

Parkinson’s disease. Pharmacokinetic and pharmacodynamic modeling will eventually be necessary to understand how a lead therapy is tolerated in organism

(pharmacokinetics) and the overall and/or additive effect(s) of a lead therapy

(pharmacodynamics).

Forthcoming generations of ACPAs will be assessed in the horizontal activity assay to identify compounds with anti-akinetic properties. While these results are promising, much work is left as the current library of ACPAs are yet isoform selective for any receptor and work to identify biased signaling is imperative to mitigate unwanted side-effects through activating off-target GPCRs. A larger library scope and parallel in vitro functional assays will certainly expand our current findings and will

147

better inform the scientific community, as well as the greater population, on how we can better treat Parkinson’s disease symptoms and improve the quality of life for patients.

148

Appendix A

1H and 13C NMR spectra for compounds listed in chapter 3.

149

150

151

152

153

154

155

156

157

158

159

160

161

162

163

164

165

166

167

168

169

170

171

172

173

174

175

176

References

1. Lawler JMHC. Neurodegenerative Diseases 2018 [updated August 6. 2018; cited 2018 August 28]. Available from: https://www.niehs.nih.gov/research/supported/health/neurodegenerative/index.cfm.

2. Brown RC, Lockwood AH, Sonawane BR. Neurodegenerative Diseases: An Overview of Environmental Risk Factors. Environmental Health Perspectives. 2005;113(9):1250-6. doi: 10.1289/ehp.7567. PubMed PMID: PMC1280411.

3. Massano J, Bhatia KP. Clinical Approach to Parkinson's Disease: Features, Diagnosis, and Principles of Management. Cold Spring Harbor Perspectives in Medicine. 2012;2(6):a008870. doi: 10.1101/cshperspect.a008870. PubMed PMID: PMC3367535.

4. Blandini F, Nappi G, Tassorelli C, Martignoni E. Functional changes of the basal ganglia circuitry in Parkinson's disease. Progress in Neurobiology. 2000;62(1):63-88. doi: https://doi.org/10.1016/S0301-0082(99)00067-2.

5. Antony PMA, Diederich NJ, Krüger R, Balling R. The hallmarks of Parkinson's disease. The FEBS Journal. 2013;280(23):5981-93. doi: doi:10.1111/febs.12335.

6. Santini E, Valjent E, Fisone G. Parkinson's disease: Levodopa-induced dyskinesia and signal transduction. The FEBS Journal. 2008;275(7):1392-9. doi: doi:10.1111/j.1742- 4658.2008.06296.x.

7. Fox SH. Non-dopaminergic Treatments for Motor Control in Parkinson’s Disease. Drugs. 2013;73(13):1405-15. doi: 10.1007/s40265-013-0105-4.

8. Ovallath S, Sulthana B. Levodopa: History and Therapeutic Applications. Annals of Indian Academy of Neurology. 2017;20(3):185-9. doi: 10.4103/aian.AIAN_241_17. PubMed PMID: PMC5586109.

9. Lees AJ, Tolosa E, Olanow CW. Four pioneers of L-dopa treatment: Arvid Carlsson, Oleh Hornykiewicz, George Cotzias, and Melvin Yahr. Movement Disorders. 2014;30(1):19-36. doi: 10.1002/mds.26120.

10. Oertel WH. Recent advances in treating Parkinson’s disease. F1000Research. 2017;6:260. doi: 10.12688/f1000research.10100.1. PubMed PMID: PMC5357034.

177

11. Ivanenkov YA, Veselov MS, Chufarova NV, Majouga AG, Kudryavceva AA, Ivachtchenko AV. Non-dopamine receptor ligands for the treatment of Parkinson's disease. Insight into the related chemical/property space. Molecular Diversity. 2016;20(1):345-65. doi: http://dx.doi.org/10.1007/s11030-015-9598-y. PubMed PMID: 1756907893.

12. Van der Schyf CJ, Geldenhuys WJ. Multimodal drugs and their future for Alzheimer's and Parkinson's disease. In: Youdim MBH, Douce P, editors. International Review of Neurobiology. 100: Academic Press; 2011. p. 107-25.

13. Sulzer D, Sonders MS, Poulsen NW, Galli A. Mechanisms of neurotransmitter release by amphetamines: A review. Progress in Neurobiology. 2005;75(6):406-33. doi: https://doi.org/10.1016/j.pneurobio.2005.04.003.

14. Miyamura S, Araki M, Suzuki T, Yamaguchi J, Itami K. Stereodivergent Synthesis of Arylcyclopropylamines by Sequential CH Borylation and Suzuki–Miyaura Coupling. Angewandte Chemie International Edition. 2014;54(3):846-51. doi: 10.1002/anie.201409186.

15. Ulrich S, Ricken R, Adli M. Tranylcypromine in mind (Part I): Review of pharmacology. European Neuropsychopharmacology. 2017;27(8):697-713. doi: https://doi.org/10.1016/j.euroneuro.2017.05.007.

16. Burger A, Yost WL. Arylcycloalkylamines. I. 2-Phenylcyclopropylamine. Journal of the American Chemical Society. 1948;70(6):2198-201. doi: 10.1021/ja01186a062.

17. W S. Depletion of dopamine in the striatum as an experimental model of Parkinsonism: direct effects and adaptive mechanisms. Prog Neurobiol. 1982;18:121-66.

18. Zigmond MJ SE. Parkinson's disease: Studies with an animal model. Life Sciences. 1984;35:5-18.

19. Langston JW LE, Irwin I. MPTP-induced parkinsonism in human and nonhuman primates—clinical and experimental aspects. Acta Neurol Scand Suppl. 1984;100:49-54.

20. Carlsson A. Development of new pharmacological approaches in Parkinson's disease. Adv Neurol. 1987;45:513-8.

178

21. Gerlach M RP. Animal models of Parkinson's disease: an empirical comparison with the phenomenology of the disease in man. J Neural Transm. 1996;103:987-1041.

22. Dawson TM. New animal models for Parkinson's disease. Cell. 2000;101:115-8.

23. Dauer W PS. Parkinson's disease: Mechanisms and models. Neuron. 2003;39:889- 909.

24. Levine MS CC, Hickey MA, Fleming SM, Chesselet MF. Genetic mouse models of Huntington's and Parkinson's diseases: illuminating but imperfect. Trends Neurosci. 2004;27:691-7.

25. Zhou Q-Y, Palmiter RD. Dopamine-deficient mice are severely hypoactive, adipsic, and aphagic. Cell. 83(7):1197-209. doi: 10.1016/0092-8674(95)90145-0.

26. Fon EAP, E.N.; Sun, B.C.; Killeen, N.; Sulzer, D. Vesicular transport regulates monoamine storage and release but is not essential for amphetamine action. Neuron. 1997;19:1271-83.

27. Wang Y-M, Gainetdinov RR, Fumagalli F, Xu F, Jones SR, Bock CB, et al. Knockout of the Vesicular Monoamine Transporter 2 Gene Results in Neonatal Death and Supersensitivity to Cocaine and Amphetamine. Neuron. 19(6):1285-96. doi: 10.1016/S0896-6273(00)80419-5.

28. Tieu K. A Guide to Neurotoxic Animal Models of Parkinson’s Disease. Cold Spring Harbor Perspectives in Medicine. 2011;1(1).

29. Giros B, Jaber M, Jones SR, Wightman RM, Caron MG. Hyperlocomotion and indifference to cocaine and amphetamine in mice lacking the dopamine transporter. Nature. 1996;379(6566):606-12.

30. Jones SR, Gainetdinov RR, Jaber M, Giros B, Wightman RM, Caron MG. Profound neuronal plasticity in response to inactivation of the dopamine transporter. Proceedings of the National Academy of Sciences of the United States of America. 1998;95(7):4029-34. PubMed PMID: PMC19957.

31. Gainetdinov RRC, M.G. Monoamine transporters: from genes to behavior. Annu Rev Pharmacol Toxicol. 2003;43:261-84. doi: 10.1146/annurev.pharmtox.43.050802.112309. PubMed Central PMCID: PMC12359863.

179

32. Benoit-Marand M JM, Gonon F. Release and elimination of dopamine in vivo in mice lacking the dopamine transporter: functional consequences. Eur J Neurosci 2000;12:2985-92.

33. Sotnikova TD, Beaulieu J-M, Barak LS, Wetsel WC, Caron MG, Gainetdinov RR. Dopamine-Independent Locomotor Actions of Amphetamines in a Novel Acute Mouse Model of Parkinson Disease. PLOS Biology. 2005;3(8):e271. doi: 10.1371/journal.pbio.0030271.

34. Gooden DM, Schmidt DMZ, Pollock JA, Kabadi AM, McCafferty DG. Facile synthesis of substituted trans-2-arylcyclopropylamine inhibitors of the human histone demethylase LSD1 and monoamine oxidases A and B. Bioorganic & medicinal chemistry letters. 2008;18(10):3047-51. doi: 10.1016/j.bmcl.2008.01.003. PubMed PMID: PMC2661354.

35. Sotnikova TD, Beaulieu J-M, Barak LS, Wetsel WC, Caron MG, Gainetdinov RR. Dopamine-Independent Locomotor Actions of Amphetamines in a Novel Acute Mouse Model of Parkinson Disease. PLoS Biol. 2005;3(8):e271. doi: 10.1371/journal.pbio.0030271.

36. Peretz C, Segev H, Rozani V, Gurevich T, El-Ad B, Tsamir J, et al. Comparison of Selegiline and Rasagiline Therapies in Parkinson Disease: A Real-life Study. Clinical Neuropharmacology. 2016;39(5):227-31. doi: 10.1097/WNF.0000000000000167. PubMed PMID: PMC5028154.

37. Urs NM, Bido S, Peterson SM, Daigle TL, Bass CE, Gainetdinov RR, et al. Targeting β-arrestin2 in the treatment of l-DOPA–induced dyskinesia in Parkinson’s disease. Proceedings of the National Academy of Sciences. 2015;112(19):E2517.

38. Deumens R, Blokland A, Prickaerts J. Modeling Parkinson's Disease in Rats: An Evaluation of 6-OHDA Lesions of the Nigrostriatal Pathway. Experimental Neurology. 2002;175(2):303-17. doi: https://doi.org/10.1006/exnr.2002.7891.

39. Riddle JL, Rokosik SL, Napier TC. Pramipexole- and methamphetamine-induced reward-mediated behavior in a rodent model of Parkinson's disease and controls. Behavioural Brain Research. 2012;233(1):15-23. doi: https://doi.org/10.1016/j.bbr.2012.04.027.

180

40. Cenci MA. L-DOPA-induced dyskinesia: cellular mechanisms and approaches to treatment. Parkinsonism & Related Disorders. 2007;13:S263-S7. doi: https://doi.org/10.1016/S1353-8020(08)70014-2.

41. Kishner Decomposition. Comprehensive Organic Name Reactions and Reagents. 2010. doi: doi:10.1002/9780470638859.conrr360 10.1002/9780470638859.conrr360.

42. Lewis DE. Disability, Despotism, Deoxygenation—From Exile to Academy Member: Nikolai Matveevich Kizhner. Angewandte Chemie International Edition. 2013;52(45):11704-12. doi: 10.1002/anie.201303165.

43. Dubé P, Nathel NFF, Vetelino M, Couturier M, Aboussafy CL, Pichette S, et al. Carbonyldiimidazole-Mediated Lossen Rearrangement. Organic Letters. 2009;11(24):5622-5. doi: 10.1021/ol9023387.

44. Lebel H, Leogane O. Boc-Protected Amines via a Mild and Efficient One-Pot Curtius Rearrangement. Organic Letters. 2005;7(19):4107-10. doi: 10.1021/ol051428b.

45. Roy D, Uozumi Y. Recent Advances in Palladium-Catalyzed Cross-Coupling Reactions at ppm to ppb Molar Catalyst Loadings. Advanced Synthesis & Catalysis. 2017;360(4):602-25. doi: 10.1002/adsc.201700810.

46. Kish SJ. What is the evidence that Ecstasy (MDMA) can cause Parkinson's disease? Movement Disorders. 2003;18(11):1219-23. doi: 10.1002/mds.10643.

47. Lane JR, May LT, Parton RG, Sexton PM, Christopoulos A. A kinetic view of GPCR allostery and biased agonism. Nature Chemical Biology. 2017;13:929. doi: 10.1038/nchembio.2431.

48. Mahoney JP, Sunahara RK. Mechanistic insights into GPCR–G protein interactions. Current Opinion in Structural Biology. 2016;41:247-54. doi: https://doi.org/10.1016/j.sbi.2016.11.005.

49. Roth BL, Irwin JJ, Shoichet BK. Discovery of new GPCR ligands to illuminate new biology. Nature Chemical Biology. 2017;13:1143. doi: 10.1038/nchembio.2490.

50. Lavin TN, Heald SL, Jeffs PW, Shorr RG, Lefkowitz RJ, Caron MG. Photoaffinity labeling of the beta-adrenergic receptor. Journal of Biological Chemistry. 1981;256(22):11944-50.

181

51. Lomasney JW, Leeb-Lundberg LM, Cotecchia S, Regan JW, DeBernardis JF, Caron MG, et al. Mammalian alpha 1-adrenergic receptor. Purification and characterization of the native receptor ligand binding subunit. Journal of Biological Chemistry. 1986;261(17):7710-6.

52. Kobilka BK, MacGregor C, Daniel K, Kobilka TS, Caron MG, Lefkowitz RJ. Functional activity and regulation of human beta 2-adrenergic receptors expressed in Xenopus oocytes. Journal of Biological Chemistry. 1987;262(32):15796-802.

53. Leeb-Lundberg LM, Cotecchia S, DeBlasi A, Caron MG, Lefkowitz RJ. Regulation of adrenergic receptor function by phosphorylation. I. Agonist-promoted desensitization and phosphorylation of alpha 1-adrenergic receptors coupled to inositol phospholipid metabolism in DDT1 MF-2 smooth muscle cells. Journal of Biological Chemistry. 1987;262(7):3098-105.

54. Lefkowitz RJ, Hausdorff WP, Caron MG. Role of phosphorylation in desensitization of the β-adrenoceptor. Trends in Pharmacological Sciences. 1990;11(5):190-4. doi: https://doi.org/10.1016/0165-6147(90)90113-M.

55. Koch WJ, Inglese J, Stone WC, Lefkowitz RJ. The binding site for the beta gamma subunits of heterotrimeric G proteins on the beta-adrenergic receptor kinase. Journal of Biological Chemistry. 1993;268(11):8256-60.

56. Lefkowitz RJ, Cotecchia S, Samama P, Costa T. Constitutive activity of receptors coupled to guanine nucleotide regulatory proteins. Trends in Pharmacological Sciences. 1993;14(8):303-7. doi: https://doi.org/10.1016/0165-6147(93)90048-O.

57. Hawes BE, Luttrell LM, Exum ST, Lefkowitz RJ. Inhibition of G protein-coupled receptor signaling by expression of cytoplasmic domains of the receptor. Journal of Biological Chemistry. 1994;269(22):15776-85.

58. Burgen ASV. The drug-receptor complex. Journal of Pharmacy and Pharmacology. 1966;18(3):137-49. doi: 10.1111/j.2042-7158.1966.tb07840.x.

59. Burke RE. Kinase signaling pathways: potential therapeutic targets in Parkinson's disease. Future Neurology. 2007;2(1):39-49. doi: http://dx.doi.org/10.2217/14796708.2.1.39. PubMed PMID: 874066507.

182

60. Lohse MJ, Maiellaro I, Calebiro D. Kinetics and mechanism of G protein-coupled receptor activation. Current Opinion in Cell Biology. 2014;27:87-93. doi: https://doi.org/10.1016/j.ceb.2013.11.009.

61. Strasser A, Wittmann H-J, Seifert R. Binding Kinetics and Pathways of Ligands to GPCRs. Trends in Pharmacological Sciences. 2017;38(8):717-32. doi: https://doi.org/10.1016/j.tips.2017.05.005.

62. Besnard J, Ruda GF, Setola V, Abecassis K, Rodriguiz RM, Huang X-P, et al. Automated design of ligands to polypharmacological profiles. Nature. 2012;492:215. doi: 10.1038/nature11691 https://www.nature.com/articles/nature11691#supplementary-information.

63. Sidhu A, Niznik HB. Coupling of dopamine receptor subtypes to multiple and diverse G proteins. International Journal of Developmental Neuroscience. 2000;18(7):669-77. doi: https://doi.org/10.1016/S0736-5748(00)00033-2.

64. Haleem DJ. 5-HT1A receptor-dependent control of nigrostriatal dopamine neurotransmission in the pharmacotherapy of Parkinson’s disease and schizophrenia. Behavioural Pharmacology. 2015;26(1 and 2 - Special Issue).

65. Burke MV, Nocjar C, Sonneborn AJ, McCreary AC, Pehek EA. Striatal serotonin 2C receptors decrease nigrostriatal dopamine release by increasing GABA-A receptor tone in the substantia nigra. Journal of Neurochemistry. 2014;131(4):432-43. doi: 10.1111/jnc.12842.

66. Cash R, Ruberg M, Raisman R, Agid Y. Adrenergic receptors in Parkinson's disease. Brain Research. 1984;322(2):269-75. doi: https://doi.org/10.1016/0006- 8993(84)90117-3.

67. Delaville C, De Deurwaerdère P, Benazzouz A. Noradrenaline and Parkinson's Disease. Frontiers in Systems Neuroscience. 2011;5:31.

68. Goswami P, Joshi N, Singh S. Neurodegenerative signaling factors and mechanisms in Parkinson's pathology. Toxicology in Vitro. 2017;43:104-12. doi: https://doi.org/10.1016/j.tiv.2017.06.008.

69. Anichtchik OV, Rinne JO, Kalimo H, Panula P. An Altered Histaminergic Innervation of the Substantia Nigra in Parkinson's Disease. Experimental Neurology. 2000;163(1):20-30. doi: https://doi.org/10.1006/exnr.2000.7362.

183

70. Dencker D, Thomsen M, Wörtwein G, Weikop P, Cui Y, Jeon J, et al. Muscarinic Acetylcholine Receptor Subtypes as Potential Drug Targets for the Treatment of Schizophrenia, Drug Abuse, and Parkinson’s Disease. ACS Chemical Neuroscience. 2012;3(2):80-9. doi: 10.1021/cn200110q.

71. Henry B, Fox SH, Crossman AR, Brotchie JM. µ- and δ-Opioid Receptor Antagonists Reduce Levodopa-Induced Dyskinesia in the MPTP-Lesioned Primate Model of Parkinson's Disease. Experimental Neurology. 2001;171(1):139-46. doi: https://doi.org/10.1006/exnr.2001.7727.

72. Samadi P, Bédard PJ, Rouillard C. Opioids and motor complications in Parkinson's disease. Trends in Pharmacological Sciences. 2006;27(10):512-7. doi: 10.1016/j.tips.2006.08.002.

73. Pan J, Cai H. Opioid system in L-DOPA-induced dyskinesia. Translational Neurodegeneration. 2017;6(1):1. doi: 10.1186/s40035-017-0071-y.

74. Francardo V. Sigma-1 receptor: a potential new target for Parkinson's disease? Neural Regeneration Research. 2014;9(21):1882-3. doi: 10.4103/1673-5374.145351.

75. Schmidt DMZ, McCafferty DG. trans-2-Phenylcyclopropylamine Is a Mechanism-Based Inactivator of the Histone Demethylase LSD1. Biochemistry. 2007;46(14):4408-16. doi: 10.1021/bi0618621.

76. Burg JM, Makhoul AT, Pemble CW, Link JE, Heller FJ, McCafferty DG. A rationally-designed chimeric KDM1A/KDM1B histone demethylase tower domain deletion mutant retaining enzymatic activity. FEBS Letters. 2015;589(18):2340-6. doi: https://doi.org/10.1016/j.febslet.2015.07.028.

77. Burg JM, Gonzalez JJ, Maksimchuk KR, McCafferty DG. Lysine-Specific Demethylase 1A (KDM1A/LSD1): Product Recognition and Kinetic Analysis of Full- Length Histones. Biochemistry. 2016;55(11):1652-62. doi: 10.1021/acs.biochem.5b01135. PubMed PMID: PMC4803528.

78. Milligan G, Kostenis E. Heterotrimeric G-proteins: a short history. British Journal of Pharmacology. 2009;147(S1):S46-S55. doi: 10.1038/sj.bjp.0706405.

184

79. Rasmussen SGF, DeVree BT, Zou Y, Kruse AC, Chung KY, Kobilka TS, et al. Crystal structure of the β2 adrenergic receptor–Gs protein complex. Nature. 2011;477:549. doi: 10.1038/nature10361 https://www.nature.com/articles/nature10361#supplementary-information.

80. Koehl A, Hu H, Maeda S, Zhang Y, Qu Q, Paggi JM, et al. Structure of the µ- opioid receptor–Gi protein complex. Nature. 2018;558(7711):547-52. doi: 10.1038/s41586- 018-0219-7.

81. Kroeze WK, Sassano MF, Huang X-P, Lansu K, McCorvy JD, Giguère PM, et al. PRESTO-Tango as an open-source resource for interrogation of the druggable human GPCRome. Nature Structural &Amp; Molecular Biology. 2015;22:362. doi: 10.1038/nsmb.3014 https://www.nature.com/articles/nsmb.3014#supplementary-information.

82. Lefkowitz RJ SS. Transduction of receptor signals by beta-arrestins. Science. 2005;308(5721):512-7.

83. Beaulieu JM GR, Caron MG. The Akt-GSK-3 signaling cascade in the actions of dopamine. Trends Pharmacol Sci. 2007;28(4):166-72.

84. Luttrell LM ea. Beta-arrestin-dependent formation of beta2 adrenergic receptor- Src protein kinase complexes. Science. 1999;283(5402):655-61.

85. Xiao K ea. Functional specialization of beta-arrestin interactions revealed by proteomic analysis. 2007. 2007;104(29):12011-6.

86. Beaulieu JM ea. An Akt/beta-arrestin 2/PP2A signaling complex mediates dopaminergic neurotransmission and behavior. Cell. 2005;122(2):261-73.

87. Beaulieu JM ea. Regulation of Akt signaling by D2 and D3 dopamine receptors in vivo. J Neurosci. 2007;27(4):881-5.

88. Urs NM DT, Caron MG. A dopamine D1 receptor-dependent β-arrestin signaling complex potentially regulates morphine-induced psychomotor activation but not reward in mice. Neuropsychopharmacology. 2011;36(3):551-8.

89. Kovoor A ea. D2 dopamine receptors colocalize regulator of G-protein signaling 9-2 (RGS9-2) via the RGS9 DEP domain, and RGS9 knock-out mice develop dyskinesias associated with dopamine pathways. J Neurosci. 2005;25(8):2157-65.

185

90. Gold SJ ea. RGS9-2 negatively modulates L-3,4-dihydroxyphenylalanine-induced dyskinesia in experimental Parkinson’s disease. J Neurosci. 2007;27(52):14338-48.

91. Gerfen CR PR, Worley P. Differences between dorsal and ventral striatum in Drd1a dopamine receptor coupling of dopamine- and cAMP-regulated phosphoprotein- 32 to activation of extracellular signal-regulated kinase. J Neurosci. 2008;28(28):7113-20.

92. Pavón N MA, Mendialdua A, Moratalla R. ERK phosphorylation and FosB expression are associated with L-DOPA-induced dyskinesia in hemiparkinsonian mice. Biol Psychiatry. 2006;59(1):64-74.

93. Santini E ea. Critical involvement of cAMP/DARPP-32 and extracellular signal- regulated protein kinase signaling in L-DOPA-induced dyskinesia. J Neurosci. 2007;27(26):6995-7005.

94. Fiorentini C SP, Savoldi D, Barbon A, Missale C. Persistent activation of the D1R/Shp-2/Erk1/2 pathway in l-DOPA-induced dyskinesia in the 6-hydroxy-dopamine rat model of Parkinson’s disease. Neurobiol Dis. 2013;53:339-48.

95. Santini E HM, Greengard P, Valjent E, Fisone G. Inhibition of mTOR signaling in Parkinson’s disease prevents L-DOPA-induced dyskinesia. Sci Signal. 2009;2(80):36.

96. Hallett PJ ea. Alterations of striatal NMDA receptor subunits associated with the development of dyskinesia in the MPTP-lesioned primate model of Parkinson’s disease. Neuropharmacology. 2005;48(4):503-16.

97. Silverdale MA ea. Synaptic recruitment of AMPA glutamate receptor subunits in levodopa-induced dyskinesia in the MPTP-lesioned nonhuman primate. Synapse. 2010;64(2):177-80.

98. Bezard E BJ, Gross CE. Pathophysiology of levodopa-induced dyskinesia: Potential for new therapies. Nat Rev Neurosci. 2001;2(8):577-88.

99. P J. Molecular mechanisms of L-DOPA-induced dyskinesia. Nat Rev Neurosci. 2008;9(9):665-77.

100. Picconi B ea. Loss of bidirectional striatal synaptic plasticity in L-DOPA-induced dyskinesia. Nat Neurosci. 2003;6(5):501-6.

186

101. Moehle MS, Pancani T, Byun N, Yohn SE, Wilson GH, III, Dickerson JW, et al. Cholinergic Projections to the Substantia Nigra Pars Reticulata Inhibit Dopamine Modulation of Basal Ganglia through the M₄ Muscarinic Receptor. Neuron. 2017;96(6):1358-72.e4. doi: 10.1016/j.neuron.2017.12.008.

102. Kauer-Sant’Anna M, Andreazza AC, Valvassori SS, Martins MR, Barbosa LM, Schwartsmann G, et al. A gastrin-releasing peptide receptor antagonist blocks d- amphetamine-induced hyperlocomotion and increases hippocampal NGF and BDNF levels in rats. Peptides. 2007;28(7):1447-52. doi: http://dx.doi.org/10.1016/j.peptides.2007.06.010.

103. Lin C-H, Chaudhuri KR, Fan J-Y, Ko C-I, Rizos A, Chang C-W, et al. Depression and Catechol-O-methyltransferase (COMT) genetic variants are associated with pain in Parkinson’s disease. Scientific Reports. 2017;7(1):6306. doi: 10.1038/s41598-017-06782-z.

104. Samadi P, Bedard PJ, Rouillard C. Opioids and motor complications in Parkinson’s disease. Trends Pharmacol Sci. 2006;27. doi: 10.1016/j.tips.2006.08.002.

105. Koizumi H, Morigaki R, Okita S, Nagahiro S, Kaji R, Nakagawa M, et al. Response of striosomal opioid signaling to dopamine depletion in 6-hydroxydopamine- lesioned rat model of Parkinson's disease: a potential compensatory role. Frontiers in Cellular Neuroscience. 2013;7:74.

106. Verleyen D, Luyten FP, Tylzanowski P. Orphan G-Protein Coupled Receptor 22 (Gpr22) Regulates Cilia Length and Structure in the Zebrafish Kupffer’s Vesicle. PLOS ONE. 2014;9(10):e110484. doi: 10.1371/journal.pone.0110484.

107. Takeda J, Fernald AA, Yamagata K, Le Beau MM, Bell GI. Localization of human somatostatin receptor 5 gene (SSTR5) to chromosome band 16p13.3 by fluorescence in situ hybridization. Genomics. 1995;26(3):638-9. doi: http://dx.doi.org/10.1016/0888- 7543(95)80195-R.

108. Reichmann F, Wegerer V, Jain P, Mayerhofer R, Hassan AM, Fröhlich EE, et al. Environmental enrichment induces behavioural disturbances in neuropeptide Y knockout mice. Scientific Reports. 2016;6:28182. doi: 10.1038/srep28182. PubMed PMID: PMC4910086.

109. Higuchi S, Ohtsu H, Suzuki H, Shirai H, Frank Gerald D, Eguchi S. Angiotensin II signal transduction through the AT1 receptor: novel insights into mechanisms and pathophysiology. Clinical Science. 2007;112(8):417.

187

110. Grymula K, Tarnowski M, Wysoczynski M, Drukala J, Barr FG, Ratajczak J, et al. Overlapping and Distinct Role of CXCR7-SDF-1/ITAC and CXCR4-SDF-1 Axes in Regulating Metastatic Behavior of Human Rhabdomyosarcomas. International journal of cancer Journal international du cancer. 2010;127(11):2554-68. doi: 10.1002/ijc.25245. PubMed PMID: PMC2907445.

111. Doly S, Valjent E, Setola V, Callebert J, Hervé D, Launay J-M, et al. Serotonin 5- HT2B Receptors Are Required for 3,4- Methylenedioxymethamphetamine-Induced Hyperlocomotion and 5-HT Release In Vivo and In Vitro. The Journal of Neuroscience. 2008;28(11):2933.

112. Caers J, Peymen K, Suetens N, Temmerman L, Janssen T, Schoofs L, et al. Characterization of G Protein-coupled Receptors by a Fluorescence-based Calcium Mobilization Assay. Journal of Visualized Experiments : JoVE. 2014;(89):51516. doi: 10.3791/51516. PubMed PMID: PMC4457351.

113. Weinreb O, Amit T, Riederer P, Youdim MBH, Mandel SA. Neuroprotective profile of the multitarget drug rasagiline in Parkinson's disease. In: Youdim MBH, Douce P, editors. International Review of Neurobiology. 100: Academic Press; 2011. p. 127-49.

188

Biography

Felix O. Nwogbo Jr was born June 4, 1991 to parents, Antonia and Felix Nwogbo

Sr. Raised in Charlotte, NC until the age of 8, he and his family moved to Maryland where he Howard High School. In the fall of 2009, he started his undergraduate education at the University of Maryland, Baltimore County, where he was awarded his

Bachelor of Science in Biochemistry and Molecular Biology in the spring of 2013. The following fall he began his graduate studies at Duke University with mentorship from

Dr. Dewey McCafferty and completed the program and received his Doctor of

Philosophy in chemistry in September 2018.

Honors & Awards:

Graduate Assistance in Areas of National Need (GAANN) Fellow | 2017–2018

Pharmacological Sciences Training Program Participant | 2013–2015

BioCoRE Scholar | 2013–2018

Conferences & Symposia:

F. Nwogbo, K. Alser, B.D. Bradaric, L. Olivere, D.G. McCafferty. “Chemical Biology

Approaches to Combat Parkinson’s disease.” Graduate Research Symposium, October 9,

2017, French Family Science Center, Atrium, Durham, NC.

189

F. Nwogbo, K. Alser, B.D. Bradaric, L. Olivere, D.G. McCafferty. “Chemical Biology

Approaches to Combat Parkinson’s disease.” NC Biosciences Collaborative Symposium,

July 28, 2016, Trent Semans Center, Great Hall, Durham, NC.

F. Nwogbo, K. Alser, B.D. Bradaric, L. Olivere, D.G. McCafferty. “Chemical Biology

Approaches to Combat Parkinson’s disease.” ACS National Meeting & Exposition,

March 16, 2016, Marriott Marquis San Diego Marina, Marina Salon E, San Diego, CA.

190