Copyright

by

Douglas Campbell Jones

2004

The Dissertation Committee for Douglas Campbell Jones certifies that this is the approved version of the following dissertation

THE ROLE OF ALPHA-METHYLDOPAMINE THIOETHERS IN THE

SEROTONERGIC NEUROTOXICITY OF MDA AND MDMA

Committee:

______Christine L. Duvauchelle, Supervisor

______Terrence J. Monks, Co-Supervisor

______Serrine S. Lau

______Richard E. Wilcox

______John H. Richburg

______Francisco Gonzalez-Lima

THE ROLE OF ALPHA-METHYLDOPAMINE THIOETHERS IN THE

SEROTONERGIC NEUROTOXICITY OF MDA AND MDMA

by

Douglas Campbell Jones, M.S., B.A.

DISSERTATION Presented to the Faculty of the Graduate School of The University of Texas at Austin In Partial Fulfillment of the Requirements for the Degree of

DOCTOR OF PHILOSOPHY

The University of Texas at Austin May, 2004

Dedication:

To my son, Zachary Hunter

If my words did glow with the gold of sunshine And my tunes were played on the harp unstrung Would you hear my voice come through the music Would you hold it near as it were your own

- Robert Hunter

Acknowledgements

I would like to thank my parents for their understanding, constant support, and encouragement. I would also like to thank Dr. Monks for inviting me into his laboratory and introducing me to an exciting and interesting research project. Thank you for your insight, patience and guidance throughout my graduate career. A special thank you to Dr. Christine Duvauchelle for supporting me during my last year and for allowing me to work in her laboratory. Thank you to my dissertation committee, Drs. Duvauchelle, Serrine Lau, John Richburg, Richard Wilcox, and Francisco Gonzalez-Lima. I would also like to thank Drs. Wilcox, Gary Miller, Stony Lo, and Raphael de la Torre for their assistance with a variety of experiments. Thank you to Fengju Bai for her initial guidance on this project. I would like to thank the members of Drs. Monks, Lau, and Duvauchelle laboratories for their friendship and support. I would also like to acknowledge the financial support provide by Drs. Monks and Duvauchelle. Finally, I must express the deepest gratitude to my wife, Rayna, for her unparallel understanding, constant support, and never ending love, without which, the completion of this degree would not have been possible. Thank you.

v THE ROLE OF ALPHA-METHYLDOPAMINE THIOETHERS IN THE

SEROTONERGIC NEUROTOXICITY OF MDA AND MDMA

Publication No.______

Douglas Campbell Jones, Ph.D.

The University of Texas at Austin, 2004

Supervisors: Christine Duvauchelle and Terrence Monks

3,4-Methylenedioxyamphetamine (MDA) and 3,4-methylenedioxymethamphetamine (MDMA, ecstasy) are widely abused amphetamine derivatives that target the serotonin (5-HT) system. The direct administration of MDA/MDMA into the brain fails to reproduce the “serotonin syndrome” that defines the neurotoxicity of MDA/MDMA; thus, the neurotoxicity of MDA and MDMA appears dependent on their systemic metabolism. 5-(Glutathion-S-yl)-α- methyldopamine and 2,5-bis(glutathion-S-yl)-α-methyldopamine, metabolites of MDA/MDMA, are potent serotonergic neurotoxicants, and produce behavioral and neurochemical changes similar to those observed with MDA/MDMA. Therefore, we investigated the transport of α- MeDA thioethers into the brain, and the biochemical mechanisms underling the development of neurotoxicity. Collection of extracellular fluid samples with microdialysis and subsequent analysis by HPLC and LC-MS/MS lead to the identification of thioether metabolites of N-Me-α- MeDA in the brain following peripheral administration of MDMA. GSH conjugate concentrations increased rapidly prior to a rapid decrease, whereas brain concentrations of the N- acetylcysteine conjugates increased slowly. Correlations exist between the concentration of N- Me-α-MeDA in the brain and decreases in brain 5-HT and 5-HIAA induced by MDMA. MDMA

vi is demethylenated to N-Me-α-MeDA; we therefore, examined the potential neurotoxicity of 5- (NAC)-N-Me-α-MeDA. Following intrastriatal injections, 5-(NAC)-N-Me-α-MeDA produced decreases in 5-HT, 5-HIAA, and dopamine (DA). We subsequently utilized three different cell models to investigate the potential mechanisms by which 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)- α-MeDA induce neurotoxicity. 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA are more potent than MDA/MDMA at inhibiting 5-HT transport into i) SK-N-MC cells transfected with the human serotonin transporter (SERT), ii) JAR cells, and iii) primary rat hippocampal cells. 5- (GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA induced an increase in (DA) transport into all three cell models, an effect attenuated by fluoxetine, indicating that DA transport was SERT- dependent. 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA increased reactive oxygen species (ROS) in each cell model. Fluoxetine attenuated the increase in ROS generation in hSERT- expressing cells. Finally, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA produced loss of cell viability and apoptosis in SERT-transfected SK-N-MC cells, JAR cells and hippocampal cells. These results are consistent with the view that the serotonergic neurotoxicity of MDA/MDMA requires i) the systemic metabolism to α-MeDA and N-Me-α-MeDA and conjugation to GSH, and ii) is likely mediated through ROS generation and the stimulation of DA transport into serotonergic neurons.

vii Table of Contents

List of Tables ...... xiv

List of Figures ...... xv

List of Abbreviations...... xx

CHAPTER 1: INTRODUCTION

I. INTRODUCTORY COMMENTS...... 1

A. SPECIFIC AIM 1 ...... 1 B. SPECIFIC AIM 2...... 2 C. SPECIFIC AIM 3...... 2

II. NEUROTOXICITY...... 2

III. 3,4-METHYLENEDIOXYAMPHETAMINE (MDA) AND 3,4-METHYLENE- DIOXYMETHAMPHETAMINE (MDMA; ECSTASY) ...... 6

A. ADVERSE CONSEQUENCES OF MDA AND MDMA...... 8

IV. SEROTONERGIC NEUROTOXICITY OF MDA AND MDMA...... 8

A. THE SEROTONIN SYSTEM...... 8 B. EFFECTS OF MDA AND MDMA ON THE SEROTONIN SYSTEM...... 12

V. ROLE OF METABOLISM IN MDA AND MDMA NEUROTOXICITY...... 13

A. METABOLISM OF MDA AND MDMA ...... 14 B. METABOLISM OF α-MEDA THIOETHERS ...... 17

viii

VI. THE NEUROTOXICITY OF QUINONE-THIOETHERS...... 19

A. QUINONE CHEMISTRY...... 19 B. REDOX CYCLING AND FORMATION OF ROS...... 21 C. GLUTATHIONE (GSH)...... 23 D. CHEMISTRY AND TOXICITY OF QUINONE-THIOETHERS ...... 25

VII. FACTORS AFFECTING QUINONE THIOETHER NEUROTOXICITY...... 27

A. EXPRESSION AND DISTRIBUTION OF CYTOCHROME P450s AND GLUTATHIONE-S-TRANSFERASE ...... 27 B. EXPRESSION AND DISTRIBUTION OF γ-GT ...... 29 C. EXPRESSION AND DISTRIBUTION OF SOD, CATALASE AND GSH ...... 30 D. EXPRESSION AND DISTRIBUTION TPH AND SERT PROTEINS...... 31

VIII. ROLE OF 5-(GSYL)-α-MEDA AND 2,5-BIS(GSYL)-α-MEDA IN MDA AND MDMA SEROTONERGIC NEUROTOXICITY...... 32

IX. TRANSPORT OF α-MEDA THIOETHERS INTO THE BRAIN40

A. THE BLOOD-BRAIN BARRIER (BBB)...... 34 B. TRANSPORT MECHANISMS ACROSS THE BBB...... 36 C. TRANSPORT OF α-MEDA THIOETHERS ACROSS THE BBB...... 37

VIII. MECHANISMS INVOLVED IN MDA, MDMA 5-(GSYL)-α-MEDA AND 2,5-BIS(GSYL)-α-MEDA-INDUCED NEUROTOXICITY...... 39

A. CELLULAR OXIDATIVE STRESS IN MDA AND MDMA-INDUCED SEROTONERGIC TOXICITY...... 40 B. THE INVOLVEMENT OF DOPAMINE IN MDA AND MDMA-INDUCED SEROTONERGIC NEUROTOXICITY...... 41 C. THE INVOLVEMENT OF THE SEROTONIN TRANSPORTER IN MDA AND MDMA-INDUCED NEUROTOXICITY...... 42 D. CYTOTOXICITY AND APOPTOSIS...... 43

ix IX. SIGNIFICANCE OF RESEARCH AND DISSERTATION RATIONALE ...... 44

CHAPTER 2: MATERIALS AND METHODS

I. CHEMICALS ...... 46

II. IN-VIVO EXPERIMENTS

A. MICRODIALYSIS ...... 47 1. Animals ...... 47 2. Surgical canula implantation ...... 47 .. 3. In-vitro recovery calibration...... 47 4. Probe implantation and assay of dialysate ...... 49 .. 5. High Performance Liquid Chromatography ...... 50 6. Liquid Chromatography-Tandem Mass Spectrometry (LC-MS/MS)...... 50

B. IN-VIVO CHANGES IN BRAIN MONOAMINE CONCENTRATIONS ...... 52 1. Cannula implantation surgery and intrastriatal administration of 5-(GSyl)-N-Me-α-MeDA...... 52 2. Brain dissection and tissue preparation...... 52

C. DATA ANALYSIS AND STATISTICS ...... 53

II. IN-VITRO EXPERIMENTS

A. CELL CULTURES...... 53 1. JAR cells ...... 53 2. Primary rat hippocampal and striatal cells ...... 54 3. SK-N-MC cells ...... 54

B. TRANSIENT TRANSFECTION OF SK-N-MC CELLS...... 54

C. CELLULAR UPTAKE EXPERIMENTS...... 56

D. IMMUNOCYTOCHEMISTRY ...... 57

E. EVALUATION OF ROS GENERATION...... 57

F. TUNEL STAINING ...... 58

G. STATISTICS...... 58

x CHAPTER 3: ACIVICIN INCREASES THE CONCENTRATION OF α- MeDA THIOETHERS IN THE BRAIN AND POTENTIATES SEROTONERGIC NEUROTOXICITY FOLLOWING PERIPHERAL ADMINISTRATION OF MDMA

I. INTRODUCTION AND RATIONALE ...... 59

II. RESULTS...... 62

A. ACIVICIN POTENTIATES MDMA-INDUCED LONG-TERM DECREASES IN BRAIN 5-HT AND 5-HIAA CONCENTRATIONS ...... 62

B. LC-MS/MS IDENTIFICATION OF THIOETHER METABOLITES IN STRIATAL DIALYSATE SAMPLES FOLLOWING PERIPHERAL ADMINISTRATION OF MDMA WITH LC-MS/MS ...... 65

C. HPLC-CEAS QUANTIFICATION OF THIOETHER METABOLITES IN STRIATAL DIALYSATE SAMPLES FOLLOWING PERIPHERAL ADMINISTRATION OF MDMA...... 81

D. PEARSON CORRELATION BETWEEN STRIATAL METABOLITE CONCENTRATION AND DEGREE OF NEUROTOXICITY...... 84

III. DISCUSSION ...... 91

CHAPTER 4: IN-VIVO SEROTONERGIC NEUROTOXICITY OF 5-(NAC)- N-ME-α-MEDA FOLLOWING INTRASTRIATAL ADMINISTRATION

I. INTRODUCTION AND RATIONALE ...... 97

II. RESULTS...... 100

A. BEHAVIORAL PROFILE FOLLOWING ADMINISTRATION OF MDMA AND 5-(NAC)-N-ME-α-MEDA...... 100 B. DECREASES IN 5-HT AND 5-HIAA FOLLOWING INTRASTRIATAL ADMINISTRATION OF 5-(NAC)-N-ME-α-MEDA...... 100 C. 5-(NAC)-N-ME-α-MEDA-INDUCED DECREASE IN DA FOLLOWING INTRASTRIATAL ADMINISTRATION...... 104

III. DISCUSSION ...... 106

xi CHAPTER 5: EFFECT OF 5-(GSYL)-α-MEDA, AND 2,5-BIS(GSYL)-α-MEDA ON CELLULAR UPTAKE OF NEUROTRANSMITTERS IN PRIMARY, JAR, AND hSERT-TRANSFECTED SK-N-MC CELLS

I. INTRODUCTION RATIONALE...... 112

II. RESULTS...... 115

A. MDA, MDMA, 5-(GSYL)-α-MEDA, AND 2,5-BIS(GSYL)-α-MEDA INHIBIT 5-HT TRANSPORT INTO MULTIPLE CELL MODELS...... 115

1. hSERT-TRANSFECTED SK-N-MC CELLS...... 115 2. SEROTONERGIC JAR CELLS...... 120 3 PRIMARY HIPPOCAMPAL AND STRIATAL CELLS ...... 120

B. MDA, MDMA, 5-(GSYL)-α-MEDA, AND 2,5-BIS(GSYL)-α-MEDA STIMULATE DA TRANSPORT INTO MULTIPLE SERT-EXPRESSING CELL MODELS...... 125

1. hSERT-TRANSFECTED SK-N-MC CELLS...... 125 2. SEROTONERGIC JAR CELLS...... 129 3 PRIMARY HIPPOCAMPAL AND STRIATAL CELLS ...... 129

III. DISCUSSION ...... 136

CHAPTER 6: MDA, MDMA, 5-(GSYL)-α-MEDA, AND 2,5-BIS(GSyl)-α- MeDA-INDUCE ROS GENERATION IN PRIMARY, JAR AND hESRT- TRANSFECTED SK-N-MC CELLS

I. INTRODUCTION AND RATIONALE ...... 144

II. RESULTS...... 147

A. hSERT-TRANSFECTED SK-N-MC CELLS...... 147 B. SEROTONERGIC JAR CELLS...... 147 C. PRIMARY HIPPOCAMPAL AND STRIATAL CELLS ...... 154

III. DISCUSSION ...... 160

xii CHAPTER 7: MDA, MDMA, 5-(GSYL)-α-MEDA, AND 2,5-BIS(GSyl)-α- MeDA-INDUCED LOSS OF CELL VIABILITY AND APOPTOSIS IN PRIMARY, JAR, AND hSERT-TRANSFECTED CELLS

I. INTRODUCTION...... 165

II. RESULTS...... 167

A. APOPTOSIS IN hSERT-TRANSFECTED SK-N-MC CELLS ...... 167 B. APOPTOSIS IN SEROTONERGIC JAR CELLS ...... 167 C. CYTOTOXICITY AND APOPTOSIS IN PRIMARY STRIATAL AND HIPPOCAMPAL CELLS ...... 171

III. DISCUSSION ...... 178

CHAPTER 8: CONCLUDING REMARKS

I. CONCLUSIONS...... 184

II. FUTURE STUDIES...... 188

REFERENCES...... 190

VITA...... 221

xiii List of Tables

Table 2.1: Molecular weights and mass/change ratio of α–MeDA and N-Me- α-MeDA thioethers used for LC-MS/MS analysis...... 51

Table 3.1: Summary of the number of animals demonstrating evidence for α- MeDA and N-Me-α-MeDA thioethers in striatal dialysate samples identified by LC-MS/MS ...... 68

Table 4.1: Behavioral profile for 5-(NAC)-N-Me-α-MeDA...... 101

xiv List of Figures

Figure 1.1: Schematic representation illustrating the hypothesis of metabolism- dependent neurotoxicity of MDA and MDMA ...... 3

Figure 1.2: Chemical structures of selected amphetamines...... 7

Figure 1.3: Serotonergic projections in the brain...... 10

Figure 1.4: Metabolism of MDA and MDMA...... 15

Figure 1.5: Mercapturic metabolism of α-MeDA and N-Me-α-MeDA thioethers . 18

Figure 1.6: Selected examples of quinone structures...... 20

Figure 1.7: Redox cycling of α-MeDA...... 22

Figure 1.8: Chemical equations for the production and elimination of reactive oxygen species...... 24

Figure 1.9: Proposed transport mechanisms across the BBB ...... 38

Figure 2.1: Sterotaxic representation of microdialysis probe placement ...... 48

Figure 2.2: Confirmation of protein expression by western analysis and cellular uptake assays ...... 55

Figure 3.1: Effect of acivicin (18 mg/kg, i.p.) pretreatment on 5-HT concentrations...... 63

Figure 3.2: Effect of acivicin (18 mg/kg, i.p.) pretreatment on 5-HIAA Concentrations ...... 64

Figure 3.3: Effects of MDA, MDMA, and MDMA + acivicin on rectal body Temperature...... 66

Figure 3.4: LC-MS/MS scans of MDMA standard in ACSF ...... 69

Figure 3.5: LC-MS/MS scans of MDMA in brain dialysate samples...... 71

Figure 3.6: LC-MS/MS scans of standard 5-(GSyl)-N-Me-a-MeDA in ACSF...... 73

xv

Figure 3.7: LC-MS/MS scans of 5-(GSyl)-N-Me-α-MeDA in brain dialysate sample...... 75

Figure 3.8: LC-MS/MS scans of standard 5-(GSyl)-N-Me-α-MeDA in ACSF.... 77

Figure 3.9: LC-MS/MS scans of 5-(GSyl)-N-Me-α-MeDA in brain dialysate samples ...... 79

Figure 3.10: HPLC identification of 5-(GSyl)-N-Me-α-MeDA, 2,5bis-(GSyl)-N- Me-α-MeDA, 5-(NAC)-N-Me-α-MeDA, and 2,5bis-(NAC)-N-Me- α-MeDA in brain dialysate samples following MDMA administration ...... 82

Figure 3.11: Effect of acivicin on the concentration of 5-(GSyl)-N-Me-α-MeDA and 2,5bis-(GSyl)-N-Me-α-MeDA in brain dialysate samples following MDMA ...... 83

Figure 3.12: Effect of acivicin on the concentration of 5-(NAC)-N-Me-α-MeDA and 2,5bis-(NAC)-N-Me-α-MeDA in brain dialysate samples following MDMA ...... 86

Figure 3.13: Significant correlations between concentrations of 5-(NAC)-N-Me- α-MeDA (pmol) and MDMA and MDMA + acivicin and the long- term depletions in striatal 5-HT and 5-HIAA...... 87

Figure 3.14: Significant correlations between concentrations of 5-(NAC)-N-Me- α-MeDA (pmol) and MDMA and MDMA + acivicin and the long- term depletions in cortical 5-HT and 5-HIAA...... 88

Figure 3.15: Significant correlations between concentrations of 5-(NAC)-N-Me- α-MeDA (pmol) and MDMA and MDMA + acivicin and the long- term depletions in hippocampal 5-HT and 5-HIAA ...... 89

Figure 3.16: Significant correlations between concentrations of 5-(NAC)-N-Me- α-MeDA (pmol) and MDMA and MDMA + acivcin and the long- term depletions in hypothalmic 5-HT and 5-HIAA...... 90

Figure 4.1: Effect of MDMA (sc) and 5-(NAC)-N-Me-α-MeDA (intrastriatal) on 5-HT concentrations in the striatum, cortex, hippocampus and hypothalamus seven days following multiple injections...... 102

xvi

Figure 4.2: Effect of MDMA (sc) and 5-(NAC)-N-Me-α-MeDA (intrastriatal) on 5-HIAA concentrations in the striatum, cortex, hippocampus and hypothalamus seven days following multiple injections...... 103

Figure 4.3: Effect of MDMA (sc) and 5-(NAC)-N-Me-α-MeDA (intrastriatal) on DA concentrations in the striatum, cortex, hippocampus and hypothalamus seven days following multiple injections...... 105

Figure 5.1: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA inhibit 5-HT uptake into hSERT-transfected cells...... 116

Figure 5.2. MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA inhibition of 5-HT uptake into hSERT-transfected SK-N-MC cells; effect of fluoxetine...... 117

Figure 5.3: Kinetics of MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α- MeDA-mediated inhibition of 5-HT uptake in hSERT-transfected SK-N-MC cells ...... 118

Figure 5.4: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA inhibit 5-HT uptake into JAR cells ...... 121

Figure 5.5: Time course for MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA inhibition of 5-HT uptake into JAR cells...... 122

Figure 5.6: Kinetics of MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)- α-MeDA-mediated inhibition of 5-HT uptake in JAR cells ...... 123

Figure 5.7: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA inhibit 5-HT uptake into hippocampal cells ...... 126

Figure 5.8: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA inhibit 5-HT uptake into striatal cells...... 127

Figure 5.9: MDA, MDMA, 5-(GSyl)-a-MeDA and 2,5-bis(GSyl)-a-MeDA induce DA uptake into SERT-expressing SK-N-MC cells ...... 128

Figure 5.10: Fluoxetine, but not nomifensine attenuates MDA, MDMA, 5- (GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA stimulation of DA uptake into SERT-expressing SK-N-MC cells ...... 131

xvii

Figure 5.11: Fluoxetine, but not nomifensine attenuates MDA, MDMA, 5- (GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA stimulation of DA uptake into JAR cells...... 132

Figure 5.12: Time course for MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA stimulation of uptake into JAR cells...... 133

Figure 5.13: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA inhibit DA uptake into striatal cells ...... 134

Figure 5.14: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA Stimulate DA uptake into hippocampal cells ...... 135

Figure 6.1: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA induce ROS generation in hSERT-transfected cells ...... 148

Figure 6.2: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA induce ROS generation in hDAT-transfected cells...... 149

Figure 6.3: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA- induced ROS generation is hSERT-transfected SK-N-MC cells transfected is concentration-dependent ...... 150

Figure 6.4: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA- induced ROS generation in hDAT-transfected SK-N-MC cells is concentration-dependent ...... 151

Figure 6.5: Fluoxetine attenuates MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA induced ROS generation in hSERT-transfected SK-N-MC cells ...... 152

Figure 6.6: Fluoxetine attenuates MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA induced ROS generation in hDAT-transfected SK-N-MC cells ...... 153

Figure 6.7: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA- induced ROS generation JAR cells is concentration-dependent ...... 155

xviii Figure 6.8: Time course for MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA-induced ROS generation in JAR cells ...... 156

Figure 6.9: Fluoxetine attenuates MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA induced ROS generation in JAR cells ...... 157

Figure 6.10: Fluoxetine attenuates MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α -MeDA induced ROS generation in hippocampal cells .. 158

Figure 6.11: Fluoxetine attenuates MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA induced ROS generation in striatal cells...... 159

Figure 7.1: MDA, MDMA 5-(GSyl)-α-MeDA, and 2,5-(GSyl)-α-MeDA- induced apoptosis in hSERT-transfected SK-N-MC cells ...... 168

Figure 7.2: Lack of MDA, MDMA 5-(GSyl)-α-MeDA, and 2,5-(GSyl)-α- MeDA-induced apoptosis in hDAT-transfected SK-N-MC cells...... 169

Figure 7.3: MDA, MDMA 5-(GSyl)-α-MeDA, and 2,5-(GSyl)-α-MeDA-induced apoptosis in JAR cells ...... 170

Figure 7.4: Tyrosine hydroxylase (TH) and tryptophan hydroxylase (TPH) immunoreactivity in hippocampal and striatal cells ...... 173

Figure 7.5: Effect of MDA, MDMA, 5-(GSyl)-α-MeDA and 5-(GSyl)-α-MeDA on TH and TPH immunoreactivity in primary hippocampal cells...... 174

Figure 7.6: Effect of MDA, MDMA, 5-(GSyl)-α-MeDA and 5-(GSyl)-α-MeDA on TPH immunoreactivity in primary striatal cells...... 175

Figure 7.7: MDA, MDMA 5-(GSyl)-α-MeDA, and 2,5-(GSyl)-α-MeDA-induced apoptosis in primary hippocampal cells...... 176

Figure 7.8: MDA, MDMA 5-(GSyl)-α-MeDA, and 2,5-(GSyl)-α-MeDA-induced apoptosis in primary striatal cells ...... 177

xix List of Abbreviations

α-MeDA Alpha-methyldopamine

ACSF Artificial cerebral spinal fluid

BBB Blood-brain-barrier

CNS Central nervous system

CSF Cerebral spinal fluid

CYP450 Cytochrome P450

5-(CYS)-α-MeDA 5-(cystein-S-yl)-α-MeDA

DA Dopamine

DAT Dopamine transporter

DCF-DA 2´7´-dichlorofluorescin diacetate

DOPAC 3,4-dihydroxyphenylacetic acid

ECF Extracellular fluid

γ-GT Gamma-glutamyltranspeptidase

GSH Glutathione

GST Glutathione-S-transferase α 5-(GSyl)-α-MeDA 5-(glutathion-S-yl)- -MeDA α 2,5-bis(GSyl)-α-MeDA 2,5-bis(glutathion-S-yl)- -MeDA α 5-(GSyl)-N-Me-α-MeDA 5-(glutathion-S-yl)-N- -MeDA α 2,5-bis(GSyl)-N-Me-α-MeDA 2,5-bis(glutathion-S-yl)-N- -MeDA

HPLC-CEAS HPLC-coulometric electrode array

xx icv Intracerebroventricular

ip Iptraperitoneal

LC-MS/MS Liquid chromatography tandem mass spectrometry

l-dopa l-dihydroxyphenylalanine

MAO Monoamine oxidase

MDA 3,4-Methylenedioxyamphetamine

MDMA 3,4-methylenedioxymethamphetamine

5-(NAC)-α-MeDA 5-(N-acetylcysteine-S-yl)-α-MeDA

2,5-bis(NAC)-α-MeDA 2,5-bis(N-acetylcysteine-S-yl)-α-MeDA

5-(NAC)-N-Me-α-MeDA 5-( N-acetylcysteine-S-yl)-N-α-MeDA

2,5-bis(NAC)-N-Me-α-MeDA 2,5-bis(N-acetylcysteine-S-yl)-N-α- MeDA

NSE Non-specific enolase

ROS Reactive oxygen species sc Subcutaneous

SERT Serotonin transporter

SOD Superoxide dismutase

TH Tyrosine hydroxylase

TPH Tryptophan hydroxylase

xxi CHAPTER 1

INTRODUCTION

I. INTRODUCTORY COMMENTS The bioactivation of inherently innocuous compounds to toxic metabolites has been recognized and studied for decades. In this scenario, the metabolism of xenobiotics not only targets the compound for elimination from the body but it also creates, as byproducts, reactive metabolites with potentially toxic properties. The extent of the toxic insult is often a reflection of the balance between the deleterious effect of the toxic metabolite and the body’s protective abilities. If the toxic insult induced by a xenobiotic is greater than the body’s protective mechanisms a toxic response may be elicited, the toxicological effects may not be overcome and the target organ and/or tissue may be irreversibly damaged. Hence, because the toxic insult is dependent on the concentration of the metabolite, it follows that the toxic response is also dependent on the dose of the xenobiotics and the frequency and duration of exposure. Consequently, a toxic response is ultimately dependent on i) the rates of metabolite formation and elimination, ii) the rate of transport to the target tissue, iii) the dose of the xenobiotics, and iv) the frequency and duration of exposure. The chapters that follow represent a compilation of studies examining the bioactivation-dependent serotonergic neurotoxicity of methylenedioxyamphetamine (MDA) and methylenedioxymethamphetamine (MDMA; ecstasy). Two major areas of focus are the transport of the metabolites into the brain and the mechanisms involved in the serotonergic neurotoxicity of the thioether metabolites of MDA and MDMA. A schematic illustrating the hypothesis of this dissertation work is provided in Figure 1.1.

A. SPECIFIC AIM 1. Studies described in Chapter 3 were designed to examine the hypothesis that the systemically formed thioether metabolites of α-MeDA and N-Me-α- MeDA, 5-(glutathion-S-yl)-α-MeDA [5-(GSyl)-α-MeDA], 2,5-bis(glutathion-S-yl)-α-

1 MeDA [2,5-bis(GSyl)-α-MeDA], 5-(glutathion-S-yl)-N-Me-α-MeDA [5-(GSyl)-N-Me- α-MeDA] and 2,5-bis(glutathion-S-yl)-N-Me-α-MeDA [2,5-bis(GSyl)-N-Me-α-MeDA] are transported into the brain via glutathione (GSH) transporters located on the blood- brain-barrier (BBB) following the peripheral administration of MDMA.

B. SPECIFIC AIM 2. The direct injection of α-MeDA thioethers into the brain produces long-term decreases in brain serotonin (5-HT) concentrations (Bai et al., 1999). Therefore, because N-Me-α-MeDA thioethers have similar structural and chemical properties, the studies in Chapter 4 were designed to test the hypothesis that the mercapturic acid of N-Me-α-MeDA, 5-(N-acetylcysteine-S-yl)-N-Me-α-MeDA (5- (NAC)-N-Me-α-MeDA) produces long-term decreases in brain 5-HT concentrations.

C. SPECIFIC AIM 3. The serotonin transporter (SERT) serves as a prime molecular target for MDA and MDMA (Ricaurte et al., 1988) and the oxidation of DA within the serotonergic nerve terminal has been implicated in the neurotoxicity of MDMA (Sprague and Nichols, 1995a; Sprague et al., 1998). Finally, the generation of reactive oxygen species (ROS) has been demonstrated for MDMA (Shankaran et al., 2000). Therefore, because the thioether metabolites of α-MeDA posses strong redox potentials and hence, the high capability to redox cycle (Miller et al., 1995), studies described in Chapters 5 and 6 were designed to examine the hypothesis that the SERT is a molecular target for 5- (GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA and the neurotoxic effect is mediated via the generation of ROS.

II. NEUROTOXICITY The central nervous system (CNS) regulates how humans feel, think, and behave. Therefore, drugs and chemicals that target the CNS, or the brain in particular, are of significant interest to us as humans because they manipulate natural biochemical processes to achieve the desired neuropharmacological effect.

2

Figure 1.1: Schematic representation illustrating the hypothesis of metabolism-dependent neurotoxicity of MDA and MDMA. A) MDA and MDMA are metabolized in the liver and conjugated to glutathione (GSH). B) The thioethers metabolites of α-MeDA and N-Me-α-MeDA are transported via the blood stream to the blood-brain-barrier (BBB) and transported into the brain by GSH transporters lining the BBB. C) Inside the brain, the thioether metabolites are quickly converted to their corresponding o-quinone species. D) Quinones can act as i) oxidants, leading to the generation of reactive oxygen species (ROS) or ii) electrophiles which may covalently modify the 5-HT transporter (SERT). E) The extracellular loops of the SERT contain cysteinyl residues which are prime targets for covalent modification and ROS- induced oxidation. F) Dopamine (DA) may enter the 5-HT nerve terminal and undergo oxidation generating ROS and damaging the cell.

3 A. α GSH 5-(GSyl)- -MeDA MDA & MDMA & (Liver) 2,5-bis(GSyl)-α-MeDA

GSH transporter

GSyl and NAC-α-MeDA C. B. quinones

D. BBB ROS Electrophile DA

Cys E. F. 200 Cys 209

SERT

ROS

4 Understanding the mechanisms underlying the chemical- and xenobiotic-induced changes in biochemical processes or neuronal morphology provides significant insight into the normal and abnormal functioning of the brain, and contributes to the development of therapeutic strategies and disease specific drugs. Research in neurotoxicity has identified hundreds of compounds that manipulate biochemical processes or neuronal morphology to produce a toxic response. Deliberate or unintentional exposure to these neurotoxicants may occur via substance abuse or environmental and work-related health hazards. Defining a chemical as a neurotoxicant can involve numerous criteria and is rather subjective in nature. Although there is no single definition of neurotoxicity, the U.S. Environmental Protection Agency (EPA) defines neurotoxicity as the capacity of chemical, biological, or physical agents to cause adverse functional (neurochemical) or structural (neuroanatomical) changes in the peripheral or CNS (Tilson, 1996). The Office of Technology Assesment (OTA, 1990), the Interagency Committee on Neurotoxicity (ICON, 1990) and the Consumer Product Safety Commission (CPSC, 1992) all provide similar definitions of neurotoxicity. Functional and structural changes caused by chemical, biological, or physical insults may affect the cell body (neuropathy) or specifically target the axonal projections (axonopathy). For instance, quinolinic acid lesions the cell body while sparing the axons (Beal et al., 1986) whereas, in contrast, substituted amphetamine derivatives selectively target serotonergic axonal projections while the cell body appears to remain intact (Ricaurte et al., 1988; Colado et al., 1993). In general, the current criteria employed for identifying a chemical as a long-term neurotoxicant are i) long-term depletions in brain neurotransmitter concentrations, ii) inhibition and loss of neurotransmitter transporters, and iii) alterations in neuronal morphology (Klevin and Seiden, 1992). Therefore, these criteria were considered in this study and were used to examine the serotonergic neurotoxicity of MDA, MDMA, and α- MeDA thioethers.

5 III. 3,4-METHYLENEDIOXYAMPHETAMINE (MDA) AND 3,4- METHYLENEDIOXYMETHAMPHETAMINE (MDMA; ECSTASY)

Amphetamines, first synthesized in 1887 (reviewed in Caldwell, 1980), are the prototypical psychostimulant, releasing intracellular stores of catecholamines and producing stimulatory and euphoric effects on the CNS. Amphetamine abuse is associated with extreme changes in both mood and social behavior, from intense acts of aggression to depressive social withdrawal. Of the numerous amphetamine derivatives, MDA, and especially MDMA, have justifiably attracted the attention of the public, the media, the government, as well as numerous scientific researchers. Once extremely popular in the dance clubs during the late 1970’s and early 1980’s, the popularity of the drug decreased in the late 80’s (Dowling and McDonough, 1987). In fact, although used and revered as adjuncts to psychotherapy (Shulgin, 1990), the abuse potential and neurotoxicity observed in various animal models lead the Drug Enforcement Agency (DEA) to place both MDA and MDMA on Schedule I in 1985. However, the recreational use of these amphetamine derivatives has increased significantly over the past decade (Johnston et al., 2000; Christophersen, 2000). MDA, synthesized in 1910 (reviewed in Thiessen and Cook, 1973) and MDMA, synthesized in 1914 (reviewed in Steele et al., 1994) are ring-substituted amphetamine derivatives (Figure 1.2) with stimulant and hallucinogenic properties (Ricaurte et al., 1985; Commins et al., 1987) that are typically ingested during dance parties, or “raves”. Ingestion of MDA and MDMA, which does not cause perceptual distortions or hallucinations, produces multiple “desired” effects on the CNS resulting in an overwhelming feeling of euphoria, a heightened sense of well being, increased tactile sensations, and a strong desire to interact with others (Hegadoren et al., 1999). However, there appears to be a price one must pay in return for achieving these desired effects as MDA and MDMA have demonstrated both peripheral and central neurotoxic properties.

6

NH2 A. CH3

NHCH3 B. CH3

O NH2 C. CH O 3

O NHCH3 D. CH O 3

O NHCH2CH3

E. CH O 3

Figure 1.2: Chemical structures of amphetamine (A), methamphetamine (B), MDA (C), MDMA (D), and methylenedioxyethylamphetamine (MDE), (E).

7 A. ADVERSE CONSEQUENCES OF MDA AND MDMA

Multiple sympathomimetic effects, or effects on the peripheral nervous system have been documented, including tremors and tachycardia (Peroutka et al., 1988). Acute adverse physiological consequences of MDA and MDMA abuse include convulsions, hyperthermia, cardiac arrhythmias, rhabomyolysis, acute liver and renal failure, hepatatoxicity, and, although rare, death (Dowling and McDonough, 1987; Gordon et al., 1991; Henry et al., 1992). Although the deleterious effects that MDA and MDMA have on the peripheral nervous system and organs outside the CNS warrant investigation, a vast amount of the research and current studies on MDA and MDMA focus on the neurotoxic impact that these amphetamine derivatives have on the CNS. MDA and MDMA abuse is associated with irreversible and long-term damage to the CNS and consequent psychiatric symptoms, including paranoid psychosis (McGuire and Fahy, 1991), anxiety and/or depression (McCann and Ricaurte, 1991; Parrott, 2001; Parrot et al., 2002), impaired memory (Bolla et al., 1998) and disturbed cognitive function (Verkes et al., 2001; Parrott, 2000; McGuire, 2000; McCann et al., 1999; 1998). Substantial evidence, in both humans (Bolla et al., 1998; McCann et al., 1998; Ricaurte et al., 2000) and various animal models (Battaglia et al., 1987; O’Hearn et al., 1988; Ricaurte et al., 1988; Wilson et al., 1989) exists associating the neuropsychiatric and neurotoxic consequences of MDA and MDMA abuse with long-term deficits in brain 5-HT concentrations and severe morphological damage to serotonergic neurons

IV. SEROTONERGIC NEUROTOXICITY OF MDA AND MDMA A. THE SEROTONIN SYSTEM Serotonin (5-hydroxytryptamine, 5-HT; Figure 1.3), perhaps the oldest neurotransmitter in evolution, was first discovered in the late 1930s as researchers were identifying substances that were capable of causing smooth muscle contraction. The substance, which the researchers named enteramine, was identified in rat gastric mucosa. In the late 1940’s another vasoconstricting substance was identified in the serum and

8 named 5-HT. The structure of serotonin (Figure 1.4) was reported in 1949 and it was soon realized that 5-HT and enteramine were the same compound. 5-HT was identified in the vertebrate brain in the early 1950s (Twarog and Page, 1953). The neurochemical anatomy of brain 5-HT neurons, described in 1964 using fluorescence histochemical analysis (Dahlstrom and Fuxe, 1964), consists of a population of morphologically diverse neurons whose cell bodies are present largely within the brainstem raphe nuclei and particular regions of the reticular formation. Raphe clusters of 5-HT neurons are found rostrally from the level of the interpeduncular nucleus in the midbrain to the level of the pyramidal decussation in the medulla. The midline raphe nuclei consist of the caudal linear nucleus, the dorsal raphe nucleus, the median raphe nucleus, raphe magnus nucleus, raphe pallidus nucleus, and the raphe obscurus nucleus. Outside the raphe nuclei there are collections of 5-HT containing cell bodies in a region adjacent to the medial lemniscus called the B9 cell cluster, in the ventrolateral medulla called the B3 cluster, and in the central gray of the medulla oblongata (B4). Although there are only about 20,000 serotonergic neurons in the rat brain (around 300K in humans) the extensive axonal projection system arising from these neurons bears a tremendous number of collateral branches so that the 5-HT system densely innervates nearly all regions of the CNS, including the prefrontal cortex, hypothalamus, hippocampus and caudate putamen (Figure 1.4). Furthermore, 5-HT nerve terminals interact with at least 13 distinct types of G-protein coupled receptors and ion-gated ligand channels. The long evolutionary history of 5-HT may help explain the extreme physical and functional diversity of the serotonergic neurotransmitter system. 5-HT is synthesized within the serotonergic nerve terminal from the amino acid L- tryptophan. The tryptophan hydroxylase (TPH)-mediated formation of 5- hydroxytrytophan (5-HTP) and the subsequent L-amino acid decarboxylase-mediated decarboxylation of 5-HTP combine to produce 5-HT. In the CNS, 5-HT acts as a neurotransmitter on a large variety of receptors, which may be located pre or post synaptically. Once released from 5-HT nerve terminals in an activity-dependent manner,

9 E

B D C

A

NH2

HO

N H Serotonin (5-hydroxytryptamine, 5-HT)

Figure 1.3: Serotonergic projections in the brain. 5-HT cell bodies are located in and around the raphe nucleus (A) and 5-HT axons innervate several regions of the brain, including the hippocampus (B), hypothalamus (C), caudate putamen (D) and cortex (E). See text for details.

10 5-HT regulates a variety of neurological as well as several endocrine related processes, including, the onset of sleep, sensory perception, mood and depression, body temperature, hyperacusis (noise sensitivity) and photophobia (sensitivity to light). Clearance of 5-HT from the brain is accomplished by multiple mechanisms. For instance, 5-HT is predominantly cleared from the synapse via transporter-dependent active reuptake into the presynaptic nerve terminal where it undergoes monoamine oxidase (MAO-A)-mediated metabolism to 5-hydroxyindolacetaldehyde and finally to 5- hydroxyindolacetic acid (5-HIAA). Alternatively, 5-HT may be conjugated with glucuronide or sulfate and targeted for elimination. Thus, 5-HT is a classically defined neurotransmitter, displaying features such as intracellular storage, activity dependent release, the existence of both pre- and postsynaptic receptors, active cellular uptake via the serotonin transporter (SERT) and metabolizing or clearance proteins. Due to the functional diversity of the serotonergic system and variety of neurological processes regulated by 5-HT, alterations in the normal functioning of the serotonergic system can lead to numerous psychiatric and neurological disorders. For instance, elevated 5-HT concentrations are associated with schizophrenia, psychosis, mania, anxiety, autism, and Alzheimer's disease. In contrast, decreased 5-HT concentrations are often linked to depression, suicide, insomnia, alcohol abuse and dependence upon various substances, and obsessive-compulsive behavior. Therefore, therapeutic strategies have been designed to provide drugs that restore the normal balance of 5-HT and help maintain proper serotonergic function. Perhaps the most common of these “serotonergic” drugs are  the selective serotonin reuptake inhibitors (SSRI’s), such as Prozac , frequently prescribed for depression. Interestingly, the long-term psychiatric disorders associated with MDA and MDMA abuse, including paranoid psychosis, persistent anxiety, and depression are similar to the behavioral effects caused by alterations in brain 5-HT activity, therefore implicating the serotonergic system in the acute biochemical changes and the long-term neurotoxicity of MDA and MDMA.

11 B. EFFECTS OF MDA AND MDMA ON THE SEROTONIN SYSTEM MDA and MDMA specifically target the serotonergic neurotransmitter system (Callahan et al., 2001; Ricaurte, 2000a, b [review]; McCann et al., 1998; Scheffel et al., 1998). The biphasic effects of MDA and MDMA are manifest as, i) an immediate, acute release of cellular 5-HT and dopamine (DA) into the synaptic cleft (Gudelsky and Nash, 1996; Brodkin et al., 1993; Berger et al., 1992), which is associated with the “desired” effects of these amphetamines, and ii) by long-term neurotoxicity characterized by significant depletions of brain 5-HT and 5-HIAA concentrations (Harkin et al., 2001; Commins et al., 1987), inhibition of SERT activity and 5-HT reuptake (Steele et al., 1987; Johnson et al., 1992; Fleckenstein et al., 1999), inhibition of TPH and 5-HT synthesis (Kuhn and Arthur, 1998, Kuhn et al., 1999; Kuhn and Geddes, 2000; 1998; Stone et al., 1989a, b; 1986), and morphological damage to the serotonergic nerve terminal exhibited as a loss of SERT proteins (Battaglia et al., 1987) and the degeneration of 5-HT axonal projections and nerve terminals (Callahan et al., 2001; Molliver et al., 1990; O’Hearn et al., 1988), The major adverse consequences of MDA and MDMA- induced neurotoxicity, 5-HT depletion (Harkin et al., 2001; Taffe et al., 2001) and axonal damage (Callahan et al., 2001) represent long-term impairments of the brain’s serotonergic system, which likely contributes to a vast range of psychiatric symptoms (Parrott, 2000; McGuire, 2000 [reviews]; McCann et al., 1997). Although the selectivity of MDA and MDMA for the serotonergic system has been demonstrated, the mechanism(s) involved are not fully understood. Interestingly, MDA and MDMA do not appear to destroy 5-HT cell bodies; rather, the toxic insult is restricted to biochemical amine deficiencies and morphological damage to axonal projections (Ricaurte et al., 1988). MDA and MDMA-induced axonal degeneration has been described as “axonal pruning”, in which the extensive 5-HT axonal projections and branches are literally pruned back as one would do to a tree. That being said, there a number of in-vitro studies suggesting that MDA and MDMA lead to neuronal death and apoptosis in a variety of “serotonergic” cell models. Moreover, several factors, including the involvement of

12 oxidative species and dopamine (DA) metabolism, associated with apoptotic cell death suggest that these amphetamine derivatives may initiate apoptotic cell death (Montiel- Daurte et al., 2002, 2004). Ongoing research into the serotonergic neurotoxicity of these popular amphetamine derivatives encompasses two fundamental questions: i) what is the basis for the selectivity that MDA and MDMA show for 5-HT neurons? ii) What are the mechanism(s) underlying the biochemical changes and axonal damage induced by MDA and MDMA? At this time, MDA and MDMA neurotoxicity appears dependent on, i) the systemic metabolism of the parent drugs, ii) the generation of reactive oxygen species (ROS), iii) the inhibition of SERT and TPH activity, and iv) the metabolism of DA within the serotonergic nerve terminal. The studies described in the following chapters focus on addressing each of these issues by integrating the areas and examining the potential importance of systemic metabolism, ROS generation, SERT activity, and DA metabolism in mediating MDA and MDMA-induced serotonergic neurotoxicity.

V. ROLE OF METABOLISM IN MDA AND MDMA NEUROTOXICITY The degree of serotonergic neurotoxicity induced by MDA and MDMA appears dependent on the route and frequency of administration (O’Shea et al., 1998). For instance, subcutaneous (sc) injection of MDMA is twice as potent as orally administered MDMA and multiple dosing regimes are more toxic than single doses (Ricaurte et al., 1988). Interestingly, consistent with a dependence on the route of drug administration, direct injection of MDA and MDMA into the brain fails to reproduce the acute or long- term neurotoxic effects evident following peripheral administration (Molliver et al., 1986; Schmidt et al., 1987, 1988; Paris et al., 1992; Esteban et al., 2001) suggesting that systemic metabolism of the parent drugs likely plays an essential role in the neurotoxicity of MDA and MDMA. In support of this hypothesis, Gollamudi et al. (1989) demonstrated that inhibition of hepatic cytochrome P450 attenuates MDMA-mediated serotonergic neurotoxicity, whereas stimulating P450 activity potentiates both the

13 metabolism and neurotoxicity of MDMA. Furthermore, differences in MDMA neurotoxic responses between rats (serotonergic) and mice (dopaminergic) have been attributed to potential metabolic differences between the two species (Logan et al., 1988). Finally, MDMA does not inhibit TPH activity in-vitro, supporting a role for bioactivation (Schmidt and Taylor, 1988). However, direct administration of several major metabolites of MDA and MDMA into the brain fails to produce the serotonergic specific neurotoxicity observed following peripheral administration of the parent amphetamines (Walker et al., 1999; Monks and Lau, 1997 [review]; McCann and Ricaurte, 1991). For example, administration of α-methyldopamine (α-MeDA) or 3-O-methyl-α- methyldopamine, major metabolites of MDA and MDMA, into brain fails to reproduce long-term serotonergic deficits (McCann and Ricaurte, 1991). Although direct central injection of 2,4,5-trihydroxyamphetamine or 2,4,5-trihydroxymethamphetamine, putative in-vivo metabolites of MDA and MDMA, are toxic to 5-HT neurons, they also insult the dopaminergic system, and thus, are lacking the 5-HT selectivity characteristic of MDA and MDMA (Johnson et al., 1992; Zhao et al., 1992). In addition, because transport mechanisms to facilitate the uptake of these metabolites into the brain have not been identified, their potential role in MDA and MDMA-induced neurotoxicity remains questionable.

A. METABOLISM OF MDA AND MDMA MDA and MDMA are metabolized to various metabolites through multiple metabolic pathways (see Green et al., 2003 for review). However, due to the inherent reactivity of GSH conjugates (Monks and Lau, 1997; Bolton et al., 2000; Monks and Jones, 2002) we are concerned herein with the formation of α-MeDA thioethers (Figure 1.4). The parent amphetamines, MDA and MDMA, undergo cytochrome P450-mediated (CYP2D6, 2B6, 3A4) demethylenation to α-MeDA and N-methyl-α-methyldopamine (N- Me-α-MeDA), respectively (Midha et al., 1978; Lim et al., 1988; Lin et al., 1992; Kumagai et al., 1991; Kreth et al., 2000). However, MDMA may also be N-demethylated

14 NHCH N H O 3 (A) O 2 CH CH O 3 CYP2D6, 2B6, O 3 (1) (2) CYP2D6, 2B6, CYP2D6, 2B6, 3A4 3A4 NH HO NHCH3 HO 2 (3) (4) CH CH HO 3 HO 3

[O2] [O2]

O NHCH3 O NH2 (5) (6) CH CH O 3 O 3

GSH GSH

NH HO NHCH3 HO 2 (7) (8) CH CH HO 3 HO 3 SG SG

[O2][OGSH 2] GSH

SG SG HO NHCH3 HO NH2 (9) (10) CH CH HO 3 HO 3 SG SG

Figure 1.4: Metabolism of MDA and MDMA. MDMA (1) and MDA (2) are de-methylenated by a variety of hepatic CYP450 isoforms to N-Me-α-MeDA (3) and α-MeDA (4), respectively. MDMA may also be demethylated to MDA (reaction A). N-Me-α- MeDA and α-MeDA are readily oxidized to the corresponding o- quinones (5 and 6). The o-quinones are quickly scavenged by GSH forming 5-(GSyl)-N-Me-α-MeDA (7) and 5-(GSyl)-α-MeDA (8). Subsequent oxidation to the o-quinone thioethers permits the reaction of a second molecule of GSH to form 2,5-bis(GSyl)-N-Me-α-MeDA (9) and 2,5-bis(GSyl)-α-MeDA (10).

15 prior to breaking the methylene ring, although this route accounts for less than 10% of the total metabolic products (de la Torre et al., 2000). Similar to the parent drugs, intracerebral-ventricular (icv) injections of α-MeDA fails to produce either the acute release or long-term depletions in 5-HT (Miller et al., 1997; McCann and Ricaurte, 1991), suggesting that further metabolism may be essential for the development of a toxic response. α-MeDA undergoes rapid oxidation to an ortho-quinone (Patel et al., 1991; Hiramatsu et al., 1990), which possesses redox-cycling potential and electrophilic properties, and are capable of oxidizing cellular sulfhydryls (Bolton et al., 2000; Monks and Lau, 1997 [reviews]). The quinone species derived from α-MeDA and N-Me-α- MeDA are highly reactive and are quickly scavenged in the liver by glutathione (GSH), leading to the formation of 5-(GSyl)-α-MeDA, 2,5-bis(GSyl)-α-MeDA, 5-(GSyl)-N-Me- α-MeDA and 2,5-bis(GSyl)-N-Me-α-MeDA (Hiramatsu et al., 1990; Patel et al., 1991; Miller et al., 1995). The conjugation to GSH usually serves a protective and detoxification role, as GSH-conjugated xenobiotics are targeted for elimination. However, the addition of GSH to α-MeDA and N-Me-α-MeDA decreases the half-wave oxidation potential (E1/2) of the catechol and enhances their biochemical reactivity (Miller et al., 1996), a trait shared by a variety of quinone species (Monks and Jones, 2002; Bolton et al., 2000). Consequently, such reactions provide a rationale for the potential contribution of GSH conjugates of α-MeDA to MDA and MDMA-induced neurotoxicity. Our laboratory has demonstrated that 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA produce serotonergic neurotoxic responses similar to that of the parent compounds (Bai et al., 1999; Miller et al., 1997, 1995). Therefore, the extent of the neurotoxic impact of MDA and MDMA may be a reflection of systemic conjugation of catechol metabolites to GSH and the ability of the thioether conjugates to penetrate the BBB and subsequently induce the generation of ROS.

16 B. METABOLISM OF α-MEDA THIOETHERS The metabolism of GSH-conjugated metabolites via the mercapturic acid metabolic pathway prepares and targets xenobiotics for elimination from the body. The relationship between GSH conjugation and the formation of mercapturic acids was characterized in 1959 (Barnes et al., 1959). The source of mercapturic acids was identified as GSH conjugates. The mercapturic acid pathway has been extensively characterized in the kidney (Meister and Tate, 1976) and recently identified in the brain. Consequently the α- MeDA thioethers, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA, are metabolized by γ- glutamyltranspeptidase (γ-GT) and dipeptidases to the corresponding cysteinyl conjugates, which are readily N-acetylated to form N-acetylcysteine conjugates, or mercapturic acids (Figure 1.5; Miller et al., 1995; Monks and Jones, 2002; Carvalho et al., 2002). The pharmacokinetics of α-MeDA thioether elimination has been reported (Miller et al.1995). Following direct administration into the brain, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA are rapidly converted to the corresponding mercapturic acids, 5- (N-acetylcysteine)-α-MeDA (5-(NAC)-α-MeDA) and 2,5-bis(N-acetylcysteine)-α- MeDA (2,5-bis(NAC)-α-MeDA), which accumulate and persist in the brain and may contribute to the neurotoxicity of MDA and MDMA. To our knowledge, the in-vivo formation of the N-Me-α-MeDA thioethers, 5-(GSyl)-N-Me-α-MeDA and 2,5-bis(GSyl)- N-Me-α-MeDA has yet to be observed, however, conjugation of GSH and N-Me-α- MeDA and the subsequent mercapturic metabolism to 5-(NAC)-N-Me-α-MeDA and 2,5- bis(NAC)-N-Me-α-MeDA has been demonstrated in liver microsomes (Hiramatsu et al., 1990). In addition, N-Me-α-MeDA has been identified in the blood and urine of humans following MDMA administration and appears to be the major metabolite compared to α- MeDA (de la Torre et al., 2000). Therefore, given the common metabolic route as well as the drastic similarities in the behavioral, pharmacological, and toxicological properties of MDA and MDMA, it is likely that formation of N-Me-α-MeDA thioethers and the subsequent metabolism to the corresponding mercapturic acids occur in-vivo. That being

17 A. B. NH HO 2 HO NHCH3

CH (1) (2) CH HO 3 HO 3 GLU-CYS-GLY GLU-CYS-GLY GLU γ-GT GLU γ-GT NH HO 2 HO NHCH3

CH CH HO 3 HO 3 CYS-GLY CYS-GLY GLY dipeptidases GLY dipeptidases

NH HO 2 HO NHCH3

CH (3) (4) CH HO 3 HO 3 Acetyl- CYS Acetyl- CYS CoA N-acetyltransferase CoA N-acetyltransferase COOH COOH NH HO 2 HO NHCH3

CH (5) (6) CH HO 3 HO 3 N-acetyl-CYS N-acetyl-CYS

O O SH H O N GSH = GLU-CYS-GLY HO N OH NH2 H O

Figure 1.5: Mercapturic metabolism of α-MeDA (A) and N-Me- α-MeDA (B) thioethers. 5-(GSyl)-α-MeDA (1) and 5-(GSyl)-N- Me-α-MeDA (2) are metabolized by γ-GT and dipeptidases to the corresponding cysteine conjugates, 5-(CYS)-α-MeDA (3) and 5- (CSY)-N-Me-α-MeDA (4) which are readily N-acetylated to the N- acetylcysteine conjugates, 5-(NAC)-α-MeDA (5) and 5-(NAC)-N- Me-α-MeDA (6).

18 said, future studies should examine the in-vivo brain pharmacokinetics of the GSH conjugation and mercapturic metabolism of N-Me-α-MeDA.

VI. THE NEUROTOXICITY OF QUINONE-THIOETHERS A. QUINONE CHEMISTRY Quinones represent a large class of electron-deficient, biologically reactive metabolic intermediates (Figure 1.6) that are capable of exerting a variety of toxicological effects including neurotoxicity, nephrotoxicity and hematotoxicity (Bolten et al., 2000; Monks and Jones, 2002). The mammalian production of quinones occurs through the oxidative metabolism of a variety of exogenous and endogenous aromatic compounds. Ironically, Although biologically significant as redox reactive co-factors and electron acceptors (for example the mitochondrial protein ubiquinone) quinone-induced toxicity also appears to be associated with their ability to redox cycle and to act as cellular electrophiles (Monks and Lau, 1997). Therefore, the mechanism(s) of quinone toxicity are twofold: quinones can act as oxidants or redox active molecules, undergoing redox cycling and generating reactive oxygen species (ROS). ROS generation results in increased levels of oxidative stress and cellular damage via the oxidation of macromolecules such as lipids, proteins and DNA. On the other hand, quinones may also act as electrophiles, capable of covalently modifying a variety of cellular nucleophiles. Due to the intrinsic nucleophilicity of sulfhydryl (-SH) groups, the major targets of quinone reactivity are often cysteinyl thiol moieties. For instance, the conjugation of GSH, a tri-peptide responsible for maintaining appropriate cellular redox states and thiol groups (Baez et al., 1997) to quinones occurs through the quinone-mediated Michael addition of the quinone onto the sulfhydryls on the cysteine residue of GSH (Monks and Lau, 1997). If not appropriately controlled by antioxidants, quinones are free to interact with cellular macromolecules, causing a toxic response. For instance, N-acetyl-p-benzoquinone mediates the hepatotoxicity of acetaminophen (Esterline et al., 1989; Harman et al., 1991). Quinones are also generated from the oxidation of a large variety of endogenous

19

O O NCOCH3 O

O O

(1) (2) (3)

O

O O O O (4) (5)

O NH2 R O α R = CH3; -MeDA R = H; DA o-quinone (6)

Figure 1.6: Selected examples of quinone structures. Metabolites of benzene, para-benzoquinone (1) and ortho-benzoquinone (2), N- acetyl-para-benzoquinonimine (NAPQI; toxic metabolite of acetaminophen) (3), benzo[a]pyrene-o-quinone (metabolite of PAH) (4), 4-OHEN-o-quinone (estrogen metabolite) (5), and α-MeDA- or DA-o-quinone (6).

20 compounds, including DA (Graham, 1978; Shen and Dryhurst, 1998; Stokes et al., 1999; Montine et al., 2002) and 5-HT (Jiang et al., 1999). DA o-quinones may contribute to neurodegenerative disorders such as Parkinson’s disease, presumably by promoting striatal dopaminergic neurotoxicity (Jones et al., 2000, 2003; Shen and Dryhurst, 1998). Thus, the immense biological reactivity and potential toxic properties of quinones warrants the significant attention of researchers in a variety of fields.

B. REDOX CYCLING AND FORMATION OF ROS Catechols, or hydroquinones, can undergo 1 or 2 electron auto-oxidations, or cytochrome P450-, or prostaglandin H synthase-mediated 1 or 2 electron oxidations producing ortho-quinones (o-quinones; Figure 1.7). In addition, microsomal, cytoplasmic, and mitochondrial reductases, including cytochrome P450 reductase, NAD(P)H:quinone oxidoreductase, microsomal NADH-cytochrome b5 reductase, and mitochondrial NADH-ubiquinone oxidoreductase, may catalyze the one-electron reduction of quinones resulting in the formation of the corresponding semiquinone species (Kappus and Sies, 1981, O’Brian, 1991). Most semiquinones are readily re- oxidized back to their quinone forms and thus, can enter a redox cycle (Figure 1.7) with molecular oxygen leading to the formation of superoxide anion radicals and hydrogen peroxide. Such ROS may inflict oxidative damage on cells and tissue by oxidizing a variety of cellular macromolecules, including lipids, DNA and, proteins. Hydrogen peroxide is formed during the redox cycling of quinones by the spontaneous dismutation of two superoxide anions or, alternatively, from the reduction of two superoxide anion by superoxide dismutase (SOD). Normally removed by SOD and catalase (Figure 1.8; equations 1 and 2), hydrogen peroxide can be toxic to several tissues, including nervous tissue (Colton et al., 1991). However, the toxicity of hydrogen peroxide is not mediated directly, but rather indirectly via the Fenton-based, transition metal-dependent formation of highly reactive free radicals such as hydroxyl radicals, hydroxyl anion, and ferric iron (Figure 1.8; equation 3). Superoxide anions interact with Fe3+ (Figure 1.8; equation 4).

21 NH O 2

O •− O 2

O 2 • O NH2

H O •− O 2 NAD(P)H Quinone Oxireductase NAD(P)H O 2

H O NH2

H O

Figure 1.7: Redox cycling of α-MeDA. The redox cycling in the brain of α-MeDA and N-Me-α-MeDA thioethers produces superoxide anion radicals. See text (page 22) for details.

22 The iron-catalyzed Haber-Weiss reaction (Figure 1.8; equation 5) combines the reactions between hydrogen peroxide, superoxide radicals and transition metals. The brain, especially the hippocampus, is extremely rich in transitions metal such as iron and copper (Danscher et al., 1976; Hartter and Barnea, 1980), hence, the redox cycling of quinone thioethers in brain tissue can lead to the formation of reactive free radicals. Natural defenses (antioxidants), including catalase, ascorbate, vitamin E, and GSH exist to protect against oxidative insults, however, many times the excessive generation of ROS can overwhelm the protective mechanisms and consequently lead to cellular oxidative stress.

C. GLUTATHIONE (GSH) GSH, a tri-peptide consisting of glutamate, cysteine, and glycine (Figure 1.5), is present in virtually all mammalian tissue and serves as a cellular antioxidant that protects the cell from oxidative injury (Meister and Anderson, 1983). GSH is involved in a variety of biochemical reactions, including the maintenance of cellular sulfhydryls and the cell’s thiol-disulfide status, the elimination of hydrogen peroxide and, the detoxification of a variety of foreign compounds. Conjugation of xenobiotics and quinones to GSH and the subsequent elimination as mercapturic acids usually serves a protective and detoxifying process. The nucleophilic thiol group on GSH mediates the conjugation to a variety of quinones and electrophiles in primarily three ways. First, many compounds, including quinones and epoxides, react with GSH in a non-enzymatic nucleophilic addition or substitution reaction. Second, a family of glutathione-S-transferases (GST) catalyzes the enzymatic conjugation of GSH to a variety of hydrophobic substrates. Finally, mediated by GSH peroxidase, GSH may donate a proton (hydrogen atom) to reduce and detoxify hydrogen peroxide, free radicals, and organic peroxides. However, conjugation to GSH does not necessarily inactivate the catechol moiety of quinone species and multiple studies have demonstrated an enhancement of biological reactivity and toxicity (Wefers and Sies, 1983; Monks and Lau, 1997). Because, i) conjugation with GSH preserves the

23 •− + SOD O2 + 2H H2O2 + O2 (1)

catalase 2H2O2 2H20 + O2 (2) GSH peroxidase

GSH GSSG

2+ 3+ - Fe + H2O2 Fe + •OH + OH (3)

2+ •− 2+ Fe +O2 Fe + O2 (4)

•− Fe - O2 + H2O2 •OH + OH + O2 (5)

Figure 1.8: Chemical equations for the production and elimination of reactive oxygen species. See text (page24 ) for details.

24 catechol function and redox reactivity of α-MeDA and ii) α-MeDA thioethers posses neurotoxicological properties (Miller et al., 1997; Bai et al., 1999), studies described in the following chapters were designed to help in understanding the mechanisms underlying the serotonergic neurotoxicity of these compounds.

D. CHEMISTRY AND TOXICITY OF QUINONE-THIOETHERS Quinone-thioether metabolites, which retain the oxidative reactivity of the parent quinones, have been implicated in a number of neurological disorders, including Parkinson’s disease and Alzheimer’s disease (Wong et al., 1993; Shen and Dryhurst, 1996). For instance, the o-quinone species of DA is readily scavenged by L-cysteine, resulting in the formation of 5-S-cysteinyl-DA (Shen and Dryhurst, 1996; Shen et al., 1997). These metabolites possess a lower oxidation potential than the parent catechol permitting redox cycling and the generation of ROS (Zhang et al., 2000), and may represent the final toxic product responsible for the dopaminergic cell death that defines Parkinson’s disease (Zhang and Dryhurst, 1994). Indeed, Parkinson’s disease, which is defined by severe dopaminergic degeneration, has been associated with elevated levels of 5-S-cysteinyl-DA and 5-S-cysteinyl-DOPAC in the striatum and with cellular oxidative stress, and apoptotic cell death (Cheng et al., 1996; Hastings et al., 1997; Montine et al., 2000. 2002). Moreover, intrastriatal injections of DA produce a toxic response associated with formation of protein-bound cysteinyl-catechols (Hastings et al., 1996). In contrast, abnormalities in the serotonergic system of Alzheimer’s patients may be attributed to the formation of tryptamine–4,5-dione and its thioether metabolite, 7-S- glutathionyl-tryptamine-4-5-dione (Wong et al., 1993). The substitution of various functional groups onto the catechol moiety can significantly affect a compounds redox potential, or ease of oxidation. Lower redox potentials are associated with greater instability and reactivity. For instance, the GSH, cysteine, and N-acetylcysteine (NAC) derivatives of DA have lower redox potentials then the parent catechol (Picklo et al., 1999), and the conjugation of GSH and NAC to

25 menadione (Buffinton et al., 1989), decreases the half-wave-oxidation potential (E 1/2). In most cases, the redox potential of the quinone decreases as the degree of substitution increases (Puckett-Vaughn et al., 1993). In agreement with these findings, and relevant to this study, the conjugation of GSH and NAC to α-MeDA lowers the oxidation potential of the parent catechol (Miller et al., 1996). Moreover, the bis-substituted thioether has a lower oxidation potential than the mono-substituted thioether, supporting the hypothesis that increased substitution leads to increased reactivity. Finally, the finding that 2,5- bis(GSyl)-α-MeDA is a more potent serotonergic neurotoxicant than 5-(GSyl)-α-MeDA (Bai et al., 1999; 2001), implies a direct relationship between increased redox reactivity and the degree of toxicity. In addition to cellular damage induced by the excessive generation of ROS produced during the redox cycling of quinone-thioethers, quinone-GSH and NAC conjugates also posses electrophilic properties and readily interact with and covalently modify a variety of cellular macromolecules, including DNA, lipids, and proteins. The formation of covalent bonds, or arylation, may inhibit the normal function of the affected macromolecule, thus producing a toxic response. For instance, acetaminophen-induced hepatotoxicity has been attributed to a reactive metabolite that, following depletion of intracellular GSH, arylates cellular macromolecules and promotes a toxic consequence (Monks and Jones, 2002). More importantly, within the serotonergic neurotransmitter system, DA derived o-quinones form covalent protein adducts with, and inhibits the activity of, tryptophan hydroxylase (TPH), a key enzyme in the synthesis of 5-HT, suggesting a possible relationship between the 5-HT neurotoxicity of MDA and MDMA and DA metabolism (Kuhn et al., 1998; Kuhn and Geddes, 2000). Thus, in summary, the lower redox potentials, increased reactivity, and electrophilic nature of the thioether metabolites of α-MeDA provide a rationale for their potential contribution to MDA and MDMA induced neurotoxicity.

26 VII. FACTORS AFFECTING QUINONE THIOETHER NEUROTOXICITY A number of variables affect the ability of a compound to induce both cellular and tissue toxicity. For instance, to produce toxicity in a specific tissue, the xenobiotic must initially be transported to the target organ. Disposition of the xenobiotic is dependent on the chemical and physical properties of the compound in question, including molecular size and lipid solubility. α-MeDA thioethers, for example, are hydrophilic in nature. To reach the brain, their target organ, these compounds must cross the lipid-rich BBB. Consequently, an active transport mechanism must be involved and individual differences in transporter expression or activity may influence the brain uptake of α- MeDA thioethers. In addition, some xenobiotics are not inherently toxic, and require metabolic transformation, or “bioactivation” to reactive, toxic metabolites. Therefore, the degree of toxicity is ultimately dependent on the ability of the body to metabolize such compounds to toxic metabolites. Individual variability in the expression, distribution and activity of the enzymes involved in catalyzing these reactions may affect the toxicity of a compound. The neurotoxicity induced by quinone thioethers may be dependent on a number of factors, including the expression and distribution of metabolic enzymes such as cytochrome P450’s and glutathione-S-transferase (GST), mercapturic enzymes such as γ-GT, dipeptidases and N-acetyltransferases, GSH and SERT transporters, and protective enzymes such as SOD, catalase, and GSH, all of which may vary dramatically between individuals. Hence, some individuals, dependent on the spectrum of variables noted above, may be inherently more susceptible to the development of MDA and MDMA- induced neurotoxicity.

A. EXPRESSION AND DISTRIBUTION OF CYTOCHROME P450s AND GLUTATHIONE-S-TRANSFERASE

Individual variability of the reactions involved in the formation of α-MeDA thioethers may influence the susceptibility of an individual to MDA and MDMA-induced neurotoxicity. For instance, genetic polymorphisms in the CYP2D family of cytochrome

27 P450’s, the enzymes responsible for the demethylenation of MDA to α-MeDA, and of MDMA to N-Me-α-MeDA, may predispose certain individuals to the neurotoxicity of these amphetamine derivatives and may have deleterious consequences when ingested in combination with other CYP2D6 inhibitors (Oesterheld and Armstrong, 2004). Indeed, inhibition of hepatic CYP2D decreases the demethylenation (Ramamoorthy et al., 2002) and neurotoxicity of MDMA (Gollamudi et al.1989). Moreover, female Dark Agouti rats, which are deficient in CYP2D1 (equivalent to the human CYP2D6) and serve as an animal model for human “poor metabolizers”, display lower levels in both the demethylenation and neurotoxicity of MDMA compared to their male counterparts (Colado et al., 1999; Chu et al., 1996). Interestingly, although no long-term neurotoxicity was observed in female Dark Agouti rats, the lack of CYP2D1 significantly increased the acute lethality of MDMA (Malpass et al., 1999). Although we hypothesize that the formation of α-MeDA thioethers from MDMA occurs systemically and is mediated by hepatic enzymes, the identification of CYP450 isoforms, CYP2D6, CYP2B6, CYP3A4 in the brain (Volk et al., 1991; Kumagai et al., 1994; Kreth et al., 2000) raises the possibility that α-MeDA thioethers may be formed centrally. However, because direct administration of the parent drugs, MDA and MDMA, into the brain fails to reproduce the neurotoxic response, it is unlikely that demethylenation of MDA and MDMA within the brain significantly contributes to the insult on the serotonergic system. Nevertheless, the heterogeneous distribution of cytochrome P450 isoforms in the brain may explain why particular regions of the brain are more susceptible to the toxicity of certain xenobiotics than others. Once α-MeDA and N-Me-α-MeDA are formed via the cytochrome P450-mediated demethylenation of MDA and MDMA, the catechols undergo conjugation to GSH (Miller et al., 1995). Therefore, theoretically, the individual variability in the regional expression and distribution of GST may affect the susceptibility to MDA and MDMA neurotoxicity. GSTs constitute a family of predominantly hepatic, cytosolic and microsomal isoenzymes that are involved in the detoxication of electrophilic xenobiotics.

28 Although there appears to be no documented evidence, individual differences in the expression of GSTs and the presence of genetic polymorphisms may predispose individuals to the neurotoxicity of MDA and MDMA. Again, we suggest that the hepatic conjugation of α-MeDA to GSH contributes to the neurotoxicity of MDA and MDMA; however, an isoenzyme-specific distribution of GSTs has been demonstrated in neurons of the brainstem, forebrain, and cerebellum. For example, the thalamus/hypothalamus had the highest GST activity and greatest concentration of total GST protein and mu-class GST, whereas the brainstem had the greatest concentration of pi-class GST. This regional variation in GST expression may be reflective of regional susceptibility to degeneration after exposure to toxic insults, including α-MeDA (Johnson et al., 1993). However, we must consider again the evidence demonstrating that direct injection of α-MeDA into the brain fails to reproduce the neurotoxicity of MDA and MDMA, suggesting a role for systemic metabolism (Miller et al., 1995; Esteban et al., 2001). Future studies should examine the potential involvement of GST in the neurotoxicity of MDA and MDMA.

B. EXPRESSION AND DISTRIBUTION OF γ-GT γ-Glutamyltranspeptidase (γ-GT), a membrane-bound enzyme with its active site located in the extracellular space, catalyzes the first step in the catabolism of GSH and GSH conjugates. The cleavage of the glutamyl bond, releases glutamate, and leaves a cysteinyl-glycine dipeptide (Orlowski and Meister, 1970). High γ-GT expression and activity are common in tissues with increased rates of amino acid transport, including cerebral microvessels (Okonkwo et al., 1974), the brush border of renal proximal tubular cells (Pfaller et al., 1974, Monks and Lau, 1994), and the choroid plexus (Okonkwo et al., 1974). Interestingly, the nephrotoxicity of quinone-thioethers in renal proximal tubular epithelial cells is dependent on the concentration of γ-GT in the brush border membrane (Monks and Lau, 1994). Because γ-GT activity is enriched in microvessel endothelia cells lining the BBB (Ghersi-Egea et al., 1993), it may play an important role in the neurotoxicity of α-MeDA thioethers by influencing their transport into the brain

29 (Bai et al., 2001). In fact, inhibition of γ-GT increases both the brain uptake of 5-(GSyl)- α-MeDA (Miller et al., 1995) and the neurotoxicity of MDA and MDMA (Chapter 3; Bai et al., 2001). Although little is known about human variability in γ-GT, individual differences in endothelial cell γ-GT expression and activity, as well as temporal changes within an individual, may affect the neurotoxicity of MDA and MDMA. The regional variability of γ-GT expression and activity in the brain may provide a biochemical basis for the different susceptibilities displayed by distinct brain regions to certain neurotoxicants (Philbert et al., 1995). For instance, following direct administration of 5-(GSyl)-α-MeDA into the brain, a significant correlation can be made between the rate of 5-(GSyl)-α-MeDA metabolism and the level of γ-GT activity (Miller et al., 1995). Thus, the formation of the mercapturic metabolites, 5-(CYS)-α-MeDA and 5-(NAC)-α-MeDA, was more pronounced in regions rich in γ-GT activity, including the hypothalamus and hippocampus. Interestingly, high levels of γ-GT activity were also detected in the pons/medulla; however, the rate of α-MeDA thioether metabolism was dramatically lower than the rate observed in the hippocampus and hypothalamus, brain regions susceptible to MDMA-induced neurotoxicity. Therefore, because increased γ-GT activity and rate of mercapturic metabolism are observed in brain regions vulnerable to the neurotoxicity of MDA and MDMA, it appears likely that the extent of the neurotoxic insult may be dependent, in part on the regional variability of γ-GT expression. Future studies should address this issue.

C. EXPRESSION AND DISTRIBUTION OF SOD, CATALASE AND GSH MDA and MDMA induced serotonergic neurotoxicity appears to be mediated through the excessive generation of ROS (Shankaran et al., 2001; Cadet et al., 2001; Zhou et al., 2003a, b). Therefore, the extent of MDA and MDMA-induced oxidative damage may be a reflection of the inter-individual and regional variability in expression and activity of antioxidants, including super oxide dismutase (SOD) (Bergeron et al., 1996; Moreno et al., 1997; Kunikowska and Jenner, 2002; Colombrita et al., 2003),

30 catalase (Brannan et al., 1981; Zimatkin and Lindros, 1996; Moreno et al., 1997), and GSH (Philbert et al., 1991; Calabrese et al., 2002). Genetic polymorphisms and individual differences in antioxidant gene expression may predispose some individuals to the neurotoxic effect of MDA and MDMA. Over-expression of SOD, for example, protects against both MDMA-induced ROS generation and neurotoxicity (Cadet et al., 1995, 2001). GSH protects cells from oxidative damage by serving as both a co-factor in the elimination of hydrogen peroxide and a nucleophile, scavenging cellular electrophiles, including quinones. Regional differences in GSH concentrations in the brain vary depending on species. However, the highest concentrations of GSH are consistently observed in the forebrain, cerebellum and hippocampus (Cooper, 1998; Sasaki and Senda, 1999). In addition to regional variability in GSH, certain regions of the brain, such as the substania nigra and caudate putamen, express high levels of GSH peroxidase (Brannan et al., 1980). Finally, the higher enzyme activities observed in glia relative to neuronal cells (Savolainen, 1978) provide an explanation for the high susceptibility of neurons to oxidative stress.

D. EXPRESSION AND DISTRIBUTION TPH AND SERT PROTEINS Because MDA, MDMA, and their respective α-MeDA thioethers target the serotonergic neurotransmitter system, inherent regional variability in the expression and activity of “serotonergic” proteins may influence the susceptibility of certain brain regions to the neurotoxic effects of these amphetamine derivatives. TPH and the SERT, two major cellular targets of MDA and MDMA, display distinct regional differences in expression and function. Consequently, brain tissue rich in TPH and SERT protein expression are selective targets of MDA and MDMA. Thus, TPH has been identified in hippocampal and cortical nerve terminals (Cohen et al., 1995), regions of the brain vulnerable to MDA- and MDMA-induced damage. Interestingly, TPH immunoreactivity was observed in close proximity to blood microvessels, which may contribute to the ability of these amphetamines, once they have crossed the BBB, to readily inhibit TPH

31 and produce 5-HT neurotoxicity. Interestingly, TPH activity has also been demonstrated in the raphe nuclei, a region rich in 5-HT cell bodies and presumably insensitive to the toxic insults of MDA and MDMA (Ehret et al., 1987; Austin and O'Donnell, 1999). SERT proteins are most often co-localized with TPH expression, thus SERT expression is highest in the hypothalamus, hippocampus, striatum and cortex (Choi et al., 2000; Kretzschmar et al., 2003; Brust et al., 2003), regions with high susceptibility to MDA and MDMA. Perhaps the most convincing evidence that variability in TPH and SERT expression affects vulnerability to the toxicity of these substituted amphetamines suggests that 5-HT cells within a particular region can demonstrate high variability in expression of SERT and TPH mRNA expression (Rattray et al., 1999). Rattray and colleagues observed significant differences and vast heterogeneity in SERT and TPH expression levels between the dorsal and median raphe nucleus with the dorsal displaying significantly higher expression levels. Because the SERT and TPH are molecular targets of MDA and MDMA, this may explain why cells originating in the dorsal raphe nuclei appear to be more sensitive to substituted amphetamine derivatives (Manounas et al., 1991). Finally, inter-individual variability among humans, and polymorphisms in TPH and SERT genes, is associated with a variety of psychiatric disorders, including depression, bi-polar, and attention deficit disorder (Gutierrez et al., 1998; Du et al., 2001; Rotondo et al., 2002; Li et al., 2003). Consequently, it is likely that human variability in SERT and TPH expression may predispose some individuals to the neurotoxicity of MDA and MDMA.

VIII. ROLE OF 5-(GSYL)-α-MEDA AND 2,5-BIS(GSYL)-α-MEDA IN MDA AND MDMA SEROTONERGIC NEUROTOXICITY

Quinones and quinone thioethers, including α-MeDA and N-Me-α-MeDA GSH conjugates, induce toxicity by their inherent behavior as oxidants and electrophiles. In the brain, the behavioral, pharmacological, and toxicological affects α-MeDA thioethers are remarkably similar to that observed for MDA and MDMA. For example, a single

32 intracerebral ventricular (icv) injection of 5-(GSyl)-α-MeDA into the brain rats produces MDA/MDMA-like neurobehavioral changes, including hyperactivity, forepaw treading, Straub tail, low posture, and salivation, and acute increases in extracellular 5-HT and DA concentrations (Miller et al., 1996), biochemical effects characteristic of MDA and MDMA. However, a single dose-dosing regime fails to produce any long-term deficits in brain 5-HT. Therefore, because multiple dose-dosing regimes are common in MDMA neurotoxicity research (Ricaurte et al., 1988) and presumably represents the “real-life” experience of MDMA users, Miller et al. (1997) used a similar method, administering 4 doses of the α-MeDA thioethers over a 36 hr period. Consistent with the single dose regime findings, multiple icv injections of 5-(NAC)-α-MeDA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA induced both the behavioral and acute pharmacological effects characteristic of MDA and MDMA. Furthermore, 2,5-bis(GSyl)-α-MeDA produced selective and long-term depletion of 5-HT in several brain regions rich in serotonergic axonal projections, including the striatum, hippocampus, and hypothalamus (Miller et al., 1997). Interestingly, consistent with the findings that MDMA induced neurotoxicity spares the 5-HT cell bodies, the concentration of 5-HT in the pons/medulla, a region rich in 5-HT cell bodies remained intact. The reasons underlying the lack of neurotoxic insult by the mono-substituted GSH and NAC metabolites is unclear, but perhaps the complex pharmacokinetics associated with intraventricular administration (icv) affected the disposition and toxicity. On the other hand, the bis-substituted thioether adversely affected the serotonergic system, consistent with the theory of increased bioreactivity with increased degree of substitution. In contrast to icv administration, the direct injection of both the mono- and bis- substituted thioethers, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA, as well as 5- (NAC)-α-MeDA into the striatum, cortex and hippocampus produced a prolonged depletion in 5-HT concentrations in brain regions heavily innervated with 5-HT axons, as well as neurobehavioral effects similar to MDA and MDMA (Bai et al., 1999). The neurotoxicity of the thioether conjugates of α-MeDA, which was more potent than the

33 parent amphetamine, MDA, also displayed selectivity for the serotonergic system, in that there was little effect on DA and norpinephrine (NE) levels. Furthermore, similar to icv administration, the neurotoxicity following the direct injection of 5-(GSyl)-α-MeDA, 2,5-bis(GSyl)-α-MeDA, and 5-(NAC)-α-MeDA into the striatum, cortex, and hippocampus was limited to the serotonergic nerve terminals, as there was no effect on 5- HT concentrations in brain regions enriched with 5-HT cell bodies, the midbrain (Bai et al., 1999). Moreover, immunofluorescence analysis with antibodies directed against the SERT, has demonstrated that intrastriatal administration of 5-(GSyl)-α-MeDA, 2,5- bis(GSyl)-α-MeDA, and 5-(NAC)-α-MeDA results in 5-HT axonal degeneration (Bai, 2000). In summary, direct injections of thioether metabolites of α-MeDA into the brain produce a serotonergic toxicity similar to, and more potent than that observed following peripheral administration of MDA or MDMA, supporting the hypothesis that α-MeDA- thioethers contribute to the neurotoxic effects of these amphetamine derivatives. Ultimately, in order for a systemically formed metabolite to induce neurotoxicity, the metabolite must first be transported across the BBB and into the brain. Preliminary studies have identified α-MeDA thioethers in the brain following intravenous administration, and a potential transport mechanism has been suggested (Miller et al., 1995; Bai et al., 2001). Studies described in this dissertation were designed to examine two major facets of α-MeDA thioether-induced neurotoxicity, i) the transport of the α- MeDA thioethers across the BBB following peripheral administration of MDMA, and ii) the mechanisms underlying the neurotoxic insult induced by these compounds and the parent amphetamines.

IX. TRANSPORT OF α-MEDA THIOETHERS INTO THE BRAIN A. THE BLOOD-BRAIN BARRIER (BBB) The cerebral-spinal-fluid (CSF), which is secreted from epithelial cells of the choroids plexus and bathes the brain, and interstitial, or extracellular fluid (ECF), which

34 is found in intercellular spaces between neurons, glial, and capillaries, comprise two of the brain’s three fluid compartments. There are no restrictions to the mutual transportation of compounds between the CSF and ECF; however, anatomical barriers exist restricting the flow of substances from the third fluid compartment of the brain, the blood plasma, into the CSF and ECF. Taken together, the blood-brain (blood-ECF) and blood-CSF barriers are called the blood-brain-barrier (BBB). In contrast to endothelial cells in most tissues, the endothelial cells in brain capillaries which define the BBB, form tight junctions and lack intercellular pores and pinocytotic vesicles, consequently restricting transport from the bloodstream into the brain (Brightman and Reese, 1969; Pardridge, 1999, 2003). Ironically, these lipid-rich barriers, which are especially effective in protecting the brain from large, water soluble molecules and organic acids and bases, may actually facilitate the brain uptake of small hydrophilic compounds and nutrients (Olendorf and Szabo, 1976; Zlokovic et al., 1994). Finally, in addition to its function as a barrier between the blood and brain, the BBB is involved in the metabolism and elimination of a variety of xenobiotics (Jette et al., 1993; Ghersi-Egea and Strazielle, 2001; Miller, 2003). In contrast, metabolism of xenobiotics, including α-MeDA thioethers by γ-GT, at the BBB may play an important role in the bioactivation of chemicals to neurotoxic metabolites (Miller et al., 1995; Bai et al., 2001; Zheng et al., 2003). γ-GT, which serves as a marker for BBB structure (Ghersi-Egea et al., 1993), is enriched in the BBB and mediates the availability of thioether to the brain. Therefore, the BBB serves both to contribute to the transport of necessary nutrients into the brain and to protect the brain from potentially dangerous substances in the bloodstream. Enzymes of the BBB often contribute to the bioactivation and subsequent toxic insult induced by numerous xenobiotics (Ghersi-Egea et al., 1995). In addition, being rich in lipids, the BBB is a prime target for oxidative damage, which can alter membrane permeability, enzyme activity, and ion flux.

35 B. TRANSPORT MECHANISMS ACROSS THE BBB The anatomy and function of the BBB, including the various transporters and its potential involvement in the bioactivation and neurotoxicity of a variety of substances has been extensively examined. Prior to gaining access to the CNS and neurons in the brain, chemicals in the blood must traverse the highly selective and restrictive membrane barriers of the BBB. Although small lipophilic molecules may cross the BBB via lipid- mediated passive diffusion, larger, more hydrophilic molecules, including α-MeDA thioethers, must gain access to the brain using active/selective transport systems. In addition, elimination of xenobiotic metabolites from the brain requires retrograde transport across the BBB; therefore, bi-directional transport mechanisms must exist at the BBB as many molecules may be transported in both directions by carrier- or receptor- mediated active transport. Receptor-mediated transport is responsible for the brain uptake of a variety of circulating peptides and plasma proteins (Banks, 1999; Moos and Morgan, 2000; Kastin and Pan, 2003, 2004). Carrier-mediated active transport, on the other hand, appears more common for the transport of low molecular weight molecules and nutrients such as glucose (Kolber et al., 1979), and water-soluble vitamins as well as a variety of neurotoxicants (Sugiyama et al., 1999; Kusuhara and Sugiyama, 2001; Tsuji and Tamai, 1999; Tamai and Tsuji, 2000; Zheng et al., 2003), including α-MeDA thioethers (Miller et al., 1995). Two major carrier-mediated uptake mechanisms relevant to the neurotoxicity of hydrophilic quinone-thioethers are the L-amino acid and GSH transporters. L-amino acid carriers belong to a family of three transport systems responsible for the uptake of water- soluble neutral amino acids into the brain. Interestingly, the L-amino acid transporter, which transports cysteine into the brain (Wade and Brady, 1981) prior to its conversion to GSH, demonstrates a strong affinity for a variety of xenobiotic-derived neurotoxic cysteine conjugates (Patel et al., 1993; Mokrazan et al., 1995; Nemoto et al., 2003). Therefore, it seems likely that the cysteine conjugates of α-MeDA and N-Me-MeDA may also gain access to the brain via the L-amino acid transport system. However, because the

36 L-amino acid transporter system is often saturated and cysteine conjugates are extremely unstable, it is likely that both the maintenance of sufficient brain GSH concentrations and the uptake of intact GSH conjugates into the brain are mediated by an alternative transport mechanism. Indeed, the low affinity transport of GSH across the BBB has been observed (Kannan et al., 1990) and a sodium-dependent GSH transporter has been identified (Kannan et al., 1999, 2000).

C. TRANSPORT OF α-MEDA THIOETHERS ACROSS THE BBB For the hypothesis, that systemically formed thioether metabolites of α-MeDA are involved in MDA and MDMA-induced serotonergic neurotoxicity to be correct there must be a transport mechanism facilitating access of α-MeDA thioethers to the brain. The BBB associated metabolism of GSH conjugates to their corresponding cysteine conjugates followed by the brain uptake of the cysteine conjugates via the L-amino acid transporter is one possible route. However, transport mechanisms for intact GSH conjugates have been demonstrated in endothelial cells of the BBB. Interestingly, and highly relevant to this discussion, the GSH transporter displays an affinity for a number of GSH conjugates (Kannan et al., 1992; Patel et al., 1993; Homma et al., 1999; Sugiyama et al., 1999). For instance, GSH bimane (GS-B), a GSH conjugate of monochlorobimane, is actively transported into the brain via a high affinity GSH transport system (Homma et al., 1999) with energy being supplied by the high density of mitochondria within endothelial cells. Therefore, it is reasonable to suggest that GSH transporters also serve as the transport mechanism responsible for the brain uptake of systemically formed α-MeDA and N-Me-α-MeDA thioethers (Figure 1.9). Indeed, co- administration of GSH with intravenously administered 5-(GSyl)-α-MeDA decreases the brain uptake of the metabolite (Miller et al., 1995). Moreover, the inhibition of γ-GT on the BBB increases both the brain uptake of (Miller et al., 1995) and the neurotoxicity of systemically administered MDA and MDMA (Bai et al., 2001), presumably by increasing the pool of intact GSH conjugates available for uptake. Polyphenolic-GSH conjugates

37 HO NH2 NH2 (A) HO CH CH HO 3 GSH transporter HO 3 GLU-CYS-GLY GLU-CYS-GLY γ-GT

NH HO 2 HO NH2

CH CH HO 3 HO 3 CYS-GLY CYS-GLY

NH NH HO 2 (B) HO 2 CH CH HO 3 L-AA HO 3 transporter CYS CYS

NH HO 2 HO NH2

CH CH HO 3 HO 3 N-acetyl-CYS N-acetyl-CYS BBB Figure 1.9: Proposed transport mechanisms across the BBB. α- MeDA thioethers may gain access to the brain via the GSH transporter (A). Alternatively, the membrane-associated metabolism to cysteine conjugates and the subsequent brain uptake via the L-amino acid transporter (B) may also represent a potential transport mechanism. Due to their chemical and structural similarities, it is likely that N-Me- α-MeDA thioethers cross into the brain in a similar fashion.

38 inhibit the activity of γ-GT (Hill et al., 1994). Consequently, α-MeDA thioethers may result in the inhibition of γ-GT thereby increasing the brain uptake of intact GSH conjugates. Thus, preliminary studies suggest the involvement of the GSH transporter in the brain uptake of α-MeDA thioethers; however, inherent variables in the experimental model may confound the interpretation of the results (see Chapter 3 for discussion). Therefore studies described in Chapter 3 address the transport of α-MeDA thioethers into the brain following peripheral administration of MDMA in the living and freely moving animal.

VIII. MECHANISMS INVOLVED IN MDA, MDMA 5-(GSYL)-α-MEDA AND 2,5-BIS(GSYL)-α-MEDA-INDUCED NEUROTOXICITY

The α-MeDA and N-Me-α-MeDA thioethers derived from the metabolism of MDA and MDMA, respectively, produce behavioral and acute neurotransmitter changes characteristic of MDA and MDMA. In addition, and perhaps more importantly, the GSH conjugates are significantly more potent to the serotonergic system than either of the parent drugs, suggesting that systemic metabolism to thioether conjugates contributes to the development of MDA and MDMA-induced neurotoxicity. Although i) the in-vivo formation of α-MeDA thioethers has been demonstrated (Miller et al., 1995), ii) a mechanism for transport across the BBB has been suggested (Miller et al., 1995; Bai et al., 2001), and iii) α-MeDA thioethers induce selective serotonergic neurotoxicity, little attention has been paid to the potential mechanisms underlying the neurotoxic response induced by α-MeDA thioethers. Due to the behavioral and pharmacological similarities between the thioether metabolites and the parent amphetamines, it is reasonable to speculate that the compounds most likely share common neurotoxic mechanisms.

39 A. CELLULAR OXIDATIVE STRESS IN MDA AND MDMA-INDUCED SEROTONERGIC TOXICITY.

Oxidative stress is a condition in which cellular macromolecules, including DNA, proteins and lipids, become oxidized by a variety of oxidants, including ROS and quinones; consequently disrupting the normal activity of the affected molecule and damaging the cell. ROS, quinones, and oxidative stress have been implicated in a number of neurological disorders, including Parkinson’s and Alzheimer’s disease (Hensley et al., 1994; Cadet and Brannock, 1998). ROS and oxidative stress are involved in the serotonergic neurotoxicity induced by MDMA. For instance, MDMA increases the production of ROS in a variety of in-vitro and in-vivo models (Guldelsky and Yamamoto, 1994; Shankaran et al., 1999, 2001; Colado et al., 1997, 1998). Moreover, free radical scavengers and antioxidants attenuate MDMA-induced neurotoxicity (Colado and Green, 1995; Gudelsky, 1996; Shanarkan et al., 2001) and over-expression of SOD, an enzyme involved in oxidation/reduction reactions, protects against the effects of MDMA by attenuating the generation of ROS (Jayanthi et al., 1999; Cadet et al., 1995). In addition, MDMA administration leads to oxidation of sulfhydryls and the inactivation of TPH (Stone et al., 1989a, b). Finally, MDMA-induced changes in cellular integrity indicative of oxidative stress, including lipid peroxidation and protein nitration, have been observed (Sprague and Nichols, 1995a, b; Colado et al., 1997a; Stone et al., 1989b). Thus, MDMA causes the transient oxidation-mediated inhibition of cytochrome oxidase, a key enzyme in the mitochondrial electron transport chain (Burrows et al., 2000). Precisely how MDMA generates ROS is unknown. MDA and MDMA do not posses the ability to redox cycle. However, the α-MeDA thioethers, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α- MeDA, are extremely redox reactive, generating both ROS and quinone species. Therefore redox cycling of 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA between their quinone and semiquinone moieties may provide a major source of MDA and MDMA- induced ROS generation. Alternatively, MDMA induces the release of DA into the extracellular space and 5-(GSyl)-α-MeDA also increases the acute turnover of brain DA

40 (Miller et al., 1995). DA oxidation results in the formation of quinones and the generation of ROS, thus it is possible that DA metabolism within the serotonergic nerve terminal contributes to neurotoxicity of MDA and MDMA.

B. THE INVOLVEMENT OF DOPAMINE IN MDA AND MDMA-INDUCED NEUROTOXICITY

Interestingly, the requirement for an intact dopaminergic system in MDMA- mediated serotonergic neurotoxicity (Shankaran et al., 1999; Sprague et al., 1998; Sprague and Nichols, 1995a, b) suggests that DA does indeed contribute to MDMA- induced increases in ROS. An essential relationship between MDMA-induced serotonergic neurotoxicity and the presence of DA has been demonstrated in several studies (Bankson and Cunningham, 2001 [review]; Aguirre et al., 1998; Stone et al., 1988). For instance, MDMA induces the release of DA in a 5-HT dependent manner (Koch and Galloway, 1997; Guldelsky and Nash, 1996; Schmidt et al., 1992a, b, 1994; Nash and Brodkin, 1991) thereby increasing the pool of DA available to the 5-HT nerve terminal. The inhibition of DA synthesis protects against MDMA-mediated neurotoxicity (Schmidt et al., 1991a; Brodkin et al., 1993). In contrast, the administration of L- dihydroxyphenylalanine (L-dopa), which is converted to DA in-vivo, potentiates the serotonergic neurotoxicity of MDMA (Schmidt et al., 1991b; Aguirre et al., 1998). Therefore, it is tempting to suggest that DA may be taken up into the 5-HT cell where it undergoes deamination and is oxidized to ROS and reactive o-quinones by monoamine oxidase-B (MAO-B) (Sprague et al., 1998). In support of this view, Faraj et al., (1994) and Schmidt et al. (1985) demonstrated that functional SERT proteins are capable of transporting DA into 5-HT cells and fluoxetine, a 5-HT reuptake inhibitor blocks the uptake of DA into hippocampal synaptosomes (Sprague and Nichols, 1995a) suggesting that DA is entering the cell via the SERT. Furthermore, inhibition of MAO-B protects against MDMA-induced neurotoxicity (Sprague and Nichols, 1995a, b) suggesting that the metabolism of DA contributes to MDMA-induced serotonergic neurotoxicity. Similar

41 to other polyphenols, DA oxidation generates a variety of cellular oxidants capable of initiating a toxic response, including ROS and o-quinones (Zhang et al., 2000; Naio and Maruyama, 1999; Graham et al., 1978; Stokes et al., 1999; Hastings and Zigmond, 1997; Halliwell, 1992). Generation of DA-derived ROS can result in lipid peroxidation, DNA damage, protein modification, and cellular death (Berman et al., 1996). Indeed, DA derived ROS and quinone species can interact with mitochondrial proteins (Burrows et al., 2000; Montine et al.1997), TPH (Kuhn et al., 1998, 2000) and the DA transporter (DAT) (Metzger et al., 1998; Berman et al., 1996), thereby contributing to the neurotoxicity of MDA and MDMA.

C. THE INVOLVEMENT OF THE SEROTONIN TRANSPORTER IN MDA AND MDMA-INDUCED NEUROTOXICITY

A significant role for the SERT during MDA and MDMA-induced neurotoxicity is firmly recognized since the transporter serves as a prime molecular target of these amphetamine derivatives (Kramer et al., 1997; 1998; Shankaran et al., 1999; Liechti et al., 2000). Thus, citolapram, a selective 5-HT re-uptake inhibitor, attenuated the toxilogical effects of MDMA in humans (Liechti et al., 2000). Furthermore, preliminary investigations in our laboratory demonstrated a significant loss of SERT immunoreactivity in 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA-treated neuronal cell cultures (Bai, 2000). The SERT, an antidepressant-sensitive, Na+/Cl- dependent transport complex of 12 transmembrane domains (Chen and Rudnick, 2000; Amara et al., 1998; Chen et al., 1997a, b), is located on 5-HT axons and nerve terminals and is responsible for the re-uptake of 5-HT from the synaptic cleft. The SERT is a molecular target for several antidepressant compounds, including fluoxetine (FX, Prozac), sibutramine, and sertraline (Mortensen et al., 2001), which do not deplete 5-HT levels (Kramer et al., 1997). In contrast, MDA and MDMA interact with the SERT to deplete 5- HT and cause morphological damage to serotonergic nerve terminals (Sprague et al., 1998; Schmidt et al., 1987). MDMA-induced depletions of 5-HT are absent in SERT-

42 deficient mice (Bengel et al., 1998) and fluoxetine protects against MDMA-induced ROS generation (Shankaran et al., 1999) and neurotoxicity (Aguirre et al., 1998; Malberg et al., 1996) indicating a role for the SERT in mediating MDMA-induced oxidative damage and serotonergic neurotoxicity. Although the importance of the SERT in contributing to MDA and MDMA-induced serotonergic toxicity has been firmly established and the number of SERT proteins measured by anti-SERT immunohistochemistry serves as a reliable marker for neurotoxicity, the precise involvement of the transporter is still unclear. However, inhibition of SERT function has been attributed to MDMA-induced ROS generation (Sprague and Nichols, 1995a, b; Falk et al., 2002) whereas conversely, an increase in ROS generation may be a consequence of MDMA induced SERT inhibition (Shankaran et al., 1999, 2001). Interestingly, MDMA may compromise the function of the SERT, decreasing its affinity for 5-HT while increasing it’s affinity for DA (Saldana and Barker, 2004). Subsequent transport of DA into the 5-HT nerve terminal followed by MAO- mediated oxidation may contribute to the serotonergic neurotoxicity of MDA and MDMA.

D. CYTOTOXICITY AND APOPTOSIS Interestingly, the neurotoxic impact of MDA and MDMA appears to be selective for the 5-HT axonal projections and nerve terminals, whereas the serotonergic cell bodies remain intact (Ricaurte et al., 1985; Commins et al., 1987). The reasons underlying this selectivity are unknown; yet, to our knowledge, MDA and MDMA-induced cell death or apoptosis has yet to be demonstrated in-vivo. That being said, several studies have observed MDMA-induced apoptosis in a variety of in-vitro cell models (Monteil Daurte et al., 2002, 2004; Simontov and Tauber, 1997; Stumm et al., 1999). Furthermore, DNA fragmentation and apoptotic cell death in hepatic cell cultures was potentiated by the addition of GSH and NAC, supporting a role for GSH and NAC conjugates in the neurotoxicity of MDA and MDMA (Monteil Daurte et al., 2004). Apoptosis, or

43 “programmed cell death” involves a well-established cascade of biochemical events resulting in the death of the cell (see Chapter 7 for discussion). By nature, the underlying mechanisms of MDA and MDMA-induced serotonergic neurotoxicity appear to support the initiation of the apoptotic cascade. For instance, ROS and cellular oxidative damage has been associated with the initiation of apoptosis (Richter, 1997; Kruman et al., 1998; Jones et al., 2000, 2003). The high redox potential and the increased generation of ROS by the α-MeDA thioethers may, therefore, have a strong ability to initiate the apoptotic cascade. In addition, DA-induced apoptosis has been associated with Parkinson’s disease and DA oxidation often results in apoptotic cell death (Hastings et al., 1996; Zhang et al., 2000; Jones et al., 2000). Therefore, given the potential “apoptosis-related” mechanisms of MDA and MDMA-induced neurotoxicity, it is appropriate to speculate that MDA and MDMA may induce apoptosis in-vivo. Nevertheless, direct evidence of this is necessary before drawing any absolute conclusions.

IX. SIGNIFICANCE OF RESEARCH AND DISSERTATION RATIONALE The significance of the research described herein is both biological and clinical in nature. Due to the alarming increase in MDA and MDMA recreational use and reported adverse consequences, there is currently a great desire to understand the pharmacology and toxicology of these amphetamine derivatives. Findings obtained in these studies will therefore assist in understanding the biological and clinical manifestations of MDA and MDMA abuse as well as the role of metabolism in xenobiotic-mediated toxicity. The greater our knowledge of the biochemical mechanism(s) involved in MDA and MDMA- induced serotonergic toxicity, the more capable we will become at treating the adverse effects of these drugs. Perhaps, of more importance, however, is the significance that these findings may have on the treatment of serotonergic-mediated disorders such as anxiety and depression, since malfunctions of 5-HT re-uptake has been associated with these and other serotonergic pathologies (Owens and Nemeroff, 1994). By understanding the actions of serotonergic agents on neurons we gain significant insights into the correct

44 and incorrect functioning of the serotonergic system, and increase the potential we have to modulate 5-HT activity. Therefore, because, i) direct administration of MDA and MDMA into the brain fails to reproduce the serotonergic neurotoxicity observed following peripheral administration, ii) direct administration of several major metabolites of MDA and MDMA into the brain also fails to reproduce the serotonergic neurotoxicity observed following peripheral administration of the parent drugs, iii) MDA and MDMA are metabolized to α-MeDA and N-Me-α-MeDA which are readily scavenged by GSH forming α-MeDA thioethers, iv) direct administration of α-MeDA thioethers into the brain mimics the serotonergic neurotoxicity observed following peripheral administration of the parent amphetamines, and v) conjugation to GSH increases the redox potential of α-MeDA resulting in the generation of ROS, we investigated the transport mechanisms involved in the brain uptake of α-MeDA thioethers and the mechanisms involved in the neurotoxicity of both the parent amphetamines and the α-MeDA thioether metabolites. Finally, for the first time, the selective serotonergic neurotoxicity of N-acetylated-N-Me-α-MeDA metabolites was examined.

45 CHAPTER 2 MATERIALS AND METHODS

I. CHEMICALS (±)MDA and (±)MDMA were supplied by the Research Technology Branch, National Institute on Drug Abuse, Rockville, MD. [3H]5-HT and [3H]DA were obtained from Radiolabeled Chemicals Inc (St. Louis, MO). Fluoxetine, nomifensine, 2’7’- dichlorofluorascein diacetate, mushroom tyrosinase, and GSH were purchased from Sigma (St. Louis, MO). 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA were synthesized and purified as previously described (Miller et al., 1995). Briefly, 5-(GSyl)-α-MeDA was prepared by reacting α-MeDA (Merck Research Laboratories, Rahway, NJ) (2mM), GSH (10mM), and mushroom tyrosinase (100U/ml) in 100 ml sodium phosphate buffer (50mM, pH 7.4). The product was purified by HPLC-CEAS (Shimadzu, LC-6A) using a Beckman Ultrasphere ODS-5 reverse-phase semi-preparative column. Fractions were collected at λ=280 nm. Collected fractions were combined, concentrated by rotary evaporation, frozen over dry ice/acetone and lyophilized to dryness. The resulting powder was re-analyzed by HPLC and Coulometric Electrode Array detection. 2,5-bis(GSyl)-α- MeDA was synthesized by dissolving 100 mg of 5-(GSyl)-α-MeDA in 100 ml of 10% formic acid. Sodium periodate (50mg) was added to the solution to promote quinone formation prior to saturating the reaction with GSH. The resulting mixture was concentrated by rotary evaporation, frozen over dry ice/acetone and lyophilized to dryness. The product was purified by HPLC-CEAS and the major UV absorbing product was eluted with water. Major fractions were collected at λ=280, rotary evaporated and lyophilized to dryness. Re-analysis of the product by HPLC-CEAS with UV and Coulometric Electrode Array detection produced a single peak. All other reagents were purchased from commercial sources. The synthesis of 5-(NAC)-N-Me-α-MeDA was

46 accomplished using a similar protocol as the one described for the GSH conjugates substituting N-Me-α-MeDA and N-acetyl cysteine for α-MeDA and GSH, respectively.

II. IN-VIVO EXPERIMENTS A. MICRODIALYSIS 1. Animals. Male Sprague-Dawley rats (Harlen Sprague-Dawley, Houston, TX) weighing ~250g were used in all in-vivo experiments. Animals were group housed and maintained on a 12 hr light/dark cycle and food and water was available ad libitum. Animals undergoing surgical cannula implantation were handled for ~2 weeks prior to surgery. Following surgery, animals were individually housed and allowed a 5-7 day recovery period.

2. Surgical cannula implantation. Guide cannulas were surgically implanted into the striatal tissue as previously described (Duvauchelle et al., 2000). Briefly, animals were anesthetized with sodium pentobarbital (50 mg/kg ip) supplemented with chloral hydrate (80 mg/kg ip). Atropine sulfate (250 µg/kg sc) was administered prophylactically to alleviate potential respiratory congestion. Animals were stereotaxically implanted with a unilateral guide cannula (21 gauge; Plastics One). The cannula was positioned 1mM above the caudate putamen (Figure 2.1; AP: 0.2 mm; ML: ± 3.0 mm; DV: 2.5 mm; Paxinos and Watson, 1997). To control for hemispheric differences, equal numbers of animals were implanted in the left vs. right hemisphere. The cannulas were afixed to the skull with dental acrylic and four stainless steel screws. Dummy cannulas were placed in the guide cannulas and animals were individually housed and allowed a 5-7 day recovery period.

3. In-Vitro Recovery Calibration. Microdialysis probes were of concentric design and were constructed in-house using PE 20 tubing as the inlet and fused silica (75 um ID) within a 26-gauge shaft (Plastics One, Roanoke), with a 4 mm active membrane

47

A

B

C

D

Figure 2.1: Sterotaxic representation of microdialysis probe placement. Probes were implanted in the region corresponding to the striatum, or caudate putamen (AP: + 0.2 Bregma, ML: + 3.0 Midline, DV: - 2.5 from skull (Paxinos and Watson, 1997). (A) guide canula. 2.5 mm; (B) probe extension, 1.0 mm; (C) cellulose membrane, 4.0mm; (D) glue tip, 0.5mm.

48

(mw cutoff = 13,000 Da; Spectrum, Houston). Prior to dialysis probe recovery, all probes were flushed with 70% EtOH and nanopure water. Hamilton syringes were filled with freshly prepared filtered Ringer’s solution (128.3 mM NaCl, 1.35 mM CaCl2, 2.68 mM

KCl, and 2.0 mM MgCl2), and pumped through the probe at 1.63 µl/min, with the probe tips in a beaker containing the Ringer’s solution for 30 min. To obtain probe recovery calibration, the Ringers solution was than replaced with Ringers solution supplemented with MDMA, 5-(GSyl)-α-MeDA, 2,5-bis(GSyl)-α-MeDA, and 5-(NAC)-N-Me-α-MeDA (20 nM) and maintained at 37° C. Two twenty min in-vitro dialysis samples were collected and dialysate was assayed by HPLC with electrochemical detection and the average of the two samples was used as the percent recovery for each probe. Probe recovery was calculated by comparing the peak heights of each dialysate collected and those from prepared standards. The probes used in this experiment had recovery percentages ranging from 15-18%. In-vitro recovery percentages were used to correct for probe-to-probe variability and data were analyzed using the corrected values.

4. Probe implantation and assay of dialysate. Probes were implanted through the guide cannulas into isofurane or halothane anesthetized rats 12-15 hours prior to the experiment and secured with dental acrylic. Probes were connected to a 1.0 ml gastight Hamilton 1000 series syringe mounted on a syringe pump (Razel, Model A), and freshly prepared Ringer’s solution was pumped through the probe (1.63 µL). Animals implanted with the probe remained in a holding chamber (14 x 14 in) overnight with the syringe pump speed set at 0.261 µl/min. Bedding, food, and water were available in the holding chamber. Thirty min prior to the test session, the pump speed was changed to 1.63 µl/min. Dialysate samples were collected the 20 minutes prior to treatment with MDMA (baseline samples) and every 20 min (~32 µL) following MDMA administration (sc) for 2 hrs in 0.4 ml microcentrifuge tubes containing perchloric acid (0.5 N).

49 5. High Performance Liquid Chromatography. Thioether metabolite concentrations and monoamine levels were quantified by HPLC equipped with a four-channel coulometric electrode array system (HPLC-CEAS; ESA Inc., Chelmsford, MA) with electrode potentials set to 50, 150, 300 and 350 mV. Sample aliquots were loaded onto an ESA HR-80 column (80 mm X 4.6 mm i.d., 3 mm particle size) and separated with a mobile phase consisting of 8 mM ammonium acetate, 4 mM citrate, 54 mM EDTA, 230 mM 1-octanesulfonic acid, and 5% methanol (pH 2.5). The flow rate was set at 1 mL/min. Quantitation of the conjugate metabolites, 5-HT, DA, and 5-HIAA, was achieved by comparing the area under the curve (AUC) with standard curves generated from authentic standards.

6. Tandem Mass Spectrometry (LC-MS/MS). High-performance LC-MS/MS was performed on dialysate samples with a Finnigan-MAT LCQ (Thermo Finnigan, San Jose, CA) electrospray ion-trap mass spectrometer coupled with a MAGIC 2002 microbore high-performance liquid chromatograph (Michrom BioResources, Auburn, CA). MS-MS spectra were acquired by data-dependent scanning with Finnigan Excalibur software (Thermo Finnigan). Total run time was 6 min. Samples were passed through a small molecular ion trap to remove undesired salts. Samples were eluted by Magic 2000 HPLC MS C18 microbore column (5 µm, 200 A, 0.5 × 50 mm; Michrom BioResources) with 50% mobile phase A [acetonitrile/water/acetic acid/trifluoroacetic acid (2:98:0.1:0.02, v/v)] and 50% mobile phase B [acetonitrile/water/acetic acid/trifluoroacetic acid (90:10:0.09:0.02, v/v)] at a flow rate of 20µl/min. The eluate from HPLC was analyzed by LCQ with ESI ion source and ion trap mass analyzer (Finnigan, San Jose, CA). Data- dependent scanning with a list of parent ions was performed; the scan events were set with one full scan (50-1500 amu) followed by three MS2. The program was set to run the MS2’s on ions corresponding to the m/z of the thioether metabolites (Table 2.1). Data- dependent scanning was performed with a default charge state of 2, isolation width of 2.0 amu, normalized collision energy of 35%, an activation time of 30.0 ms, and a required

50

MW m/z

5-(GSyl)-N-Me-α-MeDA 486 487

2,5-bis(GSyl)-N-Me-α-MeDA 791 792

5-(NAC)-N-Me-α-MeDA 342 343

2,5-bis(NAC)-N-Me-α-MeDA 503 504

5-(GSyl)-α-MeDA 472 473

2,5-bis(GSyl)-α-MeDA 777 778

5-(NAC)-α-MeDA 328 329

2,5-bis(NAC)-α-MeDA 489 490

Table 2.1: Molecular weights (MW) and mass/charge (m/z) ratio of α–MeDA and N-Me-α- MeDA thioethers used for LC-MS/MS analysis.

51 minimum signal of 50,000 counts. Global dependent data settings were an exclusion mass width of 1.5 amu, a reject mass width of 1.0 amu with dynamic exclusion enabled, a repeat count of 2, a repeat duration of 1.0 min and an exclusion duration of 1.0 min. Data acquisition and analysis were carried out by Finnigan Xcalibur software and analyzed by comparing fragmentation patterns of known standards and experimental samples.

B. IN-VIVO CHANGES IN BRAIN MONOAMINE CONCENTRATIONS

1. Cannula implantation surgery and intrastriatal administration of 5-(GSyl)-N- Me-α-MeDA. Guide cannulas (2 gauge; Plastics One) were implanted into male Sprague-Dawley rats (AP: 0.2 mm; ML: ± 3.0 mm; DV: 4.5 mm; Paxinos and Watson, 1997) as described above for dialysis experiments (Sec I.A.2; Figure 2.1). Following a 5- 7 day recovery period the dummy cannula was replaced with PE20 tubing connected to a 1.0 ml gastight Hamilton 1000 series syringe containing various concentrations of 5- (NAC)-N-Me-α-MeDA in 8 µL Ringers solution and mounted on a syringe pump (Razel, Model A). The drug solution was perfused into the striatum at a rate of 1.63 µL min for 5 minutes. Fresh Ringers solution was added to the syringe and perfusion continued for 10 min to ensure that all of the drug had been delivered to the brain. Following perfusion the dummy cannula was replaced and the animals were left un- disturbed for 7 days before being euthanized.

2. Brain dissection and tissue preparation. Animals were euthanized by decapitation and their brains quickly removed and placed onto an ice-cold plate. Brains were dissected to obtain brain regions enriched in 5-HT nerve terminals as previously described (Bai et al., 1999). Regions corresponding to the striatum, cortex, hippocampus, and hypothalamus were dissected free and frozen by liquid nitrogen. For neurotransmitter analyses, tissue was weighed and sonicated in ice-cold 0.1 N HClO4 containing 134 µM

EDTA and 263 µM Na2S2O5 for 30s. The sonicated tissues were centrifuged at 13500g

52 (4°C) for 10 min. Supernatants were centrifuged again under the same condition and aliquots (20 µL) were used for HPLC-CEAS analysis.

C. DATA ANALYSIS AND STATISTICS Concentrations of the thioether and N-acetyl conjugates are presented as absolute values (pmol/10 µl) and expressed as the mean ± SE (n = 5-6). Where relevant, the Student’s t test was used to compare MDMA to MDMA + ACIVICIN. Because multiple comparisons were made for each set of data, Bonferroni's method for multiple comparisons was used to correct for type 1 error, thus, p < 0.01. Concentrations of monoamine neurotransmitters and their metabolites are presented as the percent decrease and absolute values (pmol/mg tissue) are reported in the figure legends. Student’s t test was used to compare control to treated groups (p < 0.05; 0.01). Relevant comparisons are detailed within the figure legends. All Pearson correlation analysis was preformed using Graphpad Prism (Graphpad Software, San Diego, CA). Student’s t tests (p < 0.05; 0.01) were used to test the significance of the Pearson co-efficient (r).

II. IN-VITRO EXPERIMENTS A. Cell Cultures 1. JAR cells. JAR cells (American Type Culture Collection, ATCC, Manassas, VA) are a SERT-expressing, human placenta serotonergic cell line (Ramamoorthy et al., 1993; Simantov and Tauber, 1997). These cells readily transport 5-HT and serve as a reliable model for serotonergic cells. JAR cells were cultured in Dulbecco's Modified Eagle's Medium buffered with Hanks Buffered Saline Solution (DMEM/HBSS) supplemented with 10% Fetal Bovine Serum (FBS) and 1% penicillin/streptomyocin. Cells were grown 5 to confluency at 37°C at 5% CO2 and seeded (10 cells/ml) in 6, 12, or 24 well plates depending on experiment.

53 2. Isolation of primary striatal and hippocampal cells. Striatal and hippocampal tissue was removed from fetal Sprague-Dawley rats (E18) as previously described (Zeevalk et al, 1995). Briefly, the pregnant rat was anesthetized with pentobarbital (4-6 cc) and the fetuses were removed and placed in Ca++/Mg++ free HBSS (NaCl, 124 mM; KCl, 5.37 mM; NaH2PO4, 1 mM; glucose, 14.5 mM; Hepes, 25 mM; Mg2SO4; BSA, 3 g/L; Phenol red, 20 drops/L). Brain’s were removed and placed under a dissecting microscope. Striatal and hippocampal tissue were dissected out and placed in 15ml centrifuge tubes containing Ca++/Mg++ free HBSS. Cells were dissociated with trypsin (0.2%) and DNAseI (96 U/ml). Cells were resuspended in DMEM supplemented with 10% FBS, 1% penicillin/streptomyocin, and 30 mM KCL prior to being plated in 6 or 24 well culture plates (or on glass coverslips) at 5 X 105 cells/ml. Cells were grown to confluency (10-14 days) before being used for experiments.

3. SK-N-MC cells. SK-N-MC cells (American Type Culture Collection, ATCC, Manassas, VA) are of neuronal origin. Cells were cultured in Eagles Minimal Essential medium (EMEM, ATCC) supplemented with 10% FBS and 1% penicillin/streptomycin.

B. TRANSIENT TRANSFECTION OF SK-N-MC CELLS. The parental cDNA’s, hSERT and hDAT, inserted into pCDNA 3.1 (Invitrogen), were used for transient expression of the transporter proteins. Transfection was accomplished using the lipofectamine reagent (Invitrogen) following the protocol supplied by the manufacturer. Briefly, 24 hrs prior to transfection, cells were seeded in 24 well plates (~5 X 105 cells/well). Opti-MEM I Reduced Serum Medium (Invitrogen) containing the cDNA was combined with the Lipofectaime Reagent and incubated at room temperature for 30 min. Cells were washed with serum free medium (Opti-MEM) and the cDNA-liposomal solution was added to each well. Cells were incubated with the cDNA complexes at 37°C, 5% CO2, for 5 hours. Growth medium supplemented with 10% FBS was added to each well and the cells were incubated for 24 hrs before being

54 A. Anti-hSERT Anti-hDAT

A

A

N

N

D

D

c

c hSERT hSERT hDAT p hDAT

p

18000 16000 B. 14000 12000 10000 CPM 8000 6000 4000 2000 0 ______pcDNADAT ______pcDNA SERT DA uptake 5-HT uptake

Figure 2.2: Confirmation of protein expression by western analysis (A) and cellular uptake assays (B). Western analysis was conducted with antibodies targeted against the DA and SERT. Uptake results for mock (pcDNA), hDAT, and hSERT transfected SK-N-MC cells are expressed as the total counts per minute (CPM).

55 replaced with EMEM. Confirmation of protein expression was demonstrated by western analysis and by quantifying [3H]5-HT and [3H]DA cellular uptake into hDAT and hSERT-transfected cells (Figure 2.2). Transfected cells were used for experiments 48-72 hrs post transfection.

C. CELLULAR UPTAKE EXPERIMENTS Cellular uptake of [3H]-DA and [3H]-5-HT into hDAT and hSERT transfected SK-N- MC cells was measured 48 hours post transfection as previously described (Mortensen et al., 2001). Several uptake experiments were conducted and details for each experiment can be found in the figure legends. Briefly, cells were washed with Krebs-Ringer (KR) buffer (NaCl, 125 mM; KCl, 5 mM; Hepes, 25 mM, glucose, 6 mM; NaHCO3, 5 mM;

MgSO4*7H2O, 1.2 mM; KH2PO4, 1.2 mM; CaCl2*2H2O, 2.4 mM; pH, 7.4). Cells were then incubated in KR buffer containing 20 nM [3H]5-HT, 100 µM pargyline and 100 µM ascorbate plus MDA, MDMA, 5-(GSyl)-α-MeDA, or 2,5-bis-(GSyl)-α-MeDA (100 µM). Cellular uptake was terminated by washing the cells with KR buffer and uptake of [3H]5- HT into mock-, hDAT-, and hSERT-transfected cells was determined by liquid scintillation spectroscopy (Beckman LS 5000TD). The cellular uptake of [3H]DA into the hSERT-transfected cells was examined using a procedure similar to that used for [3H]5- HT. The kinetics of 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA-induced inhibition of 5-HT cellular uptake was examined with saturation and inhibition experiments. Saturation transport analysis was conducted by incubating the cells with increasing concentrations of substrate ([3H]5-HT) for 30 min in KR buffer containing 100 µM pargyline and 100 µM ascorbate. Non-specific uptake was determined in parallel using fluoxetine (10 µM). Assays were terminated by washing the cells with KR buffer and intracellular accumulation of [3H]5-HT was determined by liquid scintillation spectroscopy. For inhibition experiments, cells were incubated in KR buffer containing 20 nM [3H]5-HT and increasing concentrations of MDA, MDMA, 5-(GSyl)-α-MeDA, or

56 2,5-bis-(GSyl)-α-MeDA for 30 min and intracellular [3H]5-HT accumulation was determined. Assuming Michaelis-Menten kinetics, substrate Km and inhibitor Ki values were determined by non-linear least squares curve fit (GraphPad Prism; Graphpad Software Inc., San Diego, CA; Adkins et al., 2001). Experiments were carried out in triplicate and Ki values are presented as the mean (N = 4) ± the standard error of four independent transfections.

D. IMMUNOCYTOCHEMISTRY Immunocytochemistry was conducted on striatal and hippocampal primary neurons fixed in 4% paraformaldehyde for 30 min. and washed twice with PBS. Cells were incubated with 1% triton-X in PBS for 1 hr and washed two times with PBS prior to blocking non-specific binding with 10% evaporated milk for 1 hr. Cells were rinsed with PBS and exposed to primary antibodies against non-specific enolase (NSE, 1:150) and tryptophan hydroxylase (TPH, 1:1000), or tyrosine hydroxylase (TH, 1:500) for 3 hours. Cells were again rinsed with PBS and incubated with peroxidase labeled secondary IgG antibody for 2 hrs. Cells were stained with 3,3-diaminobenzidine (DAB) and visualized at 40X magnification under light microscopy. Cell counts were made in four random fields and data expressed as the percent of the total cells that stained positive for NSE/TPH or TH.

E. EVALUATION OF ROS GENERATION 2’, 7’-Dichlorofluorescein diacetate (DCF-DA) is oxidized to the fluorescent 2’, 7’- dichlorofluorescein by cellular oxidants and is used as a marker for intracellular generation of ROS. Intracellular ROS generation was monitored as previously described (Jones et al., 2000) with modifications. Cells were loaded with DCF-DA (final concentration. of 10 µM) in KR buffer for 15 min in the dark. Cells were washed with KR buffer and exposed to MDA, MDMA, 5-(GSyl)-α-MeDA or 2,5-bis(GSyl)-α-MeDA in the presence or absence of the transporter blockers, nomifensine (DAT) and fluoxetine

57 (SERT). Assays were terminated by washing cells with KR buffer and ROS generation was monitored at 475nm (excitation) and 525nm (emission) using a FL600 microplate fluorescence reader (Bio-tex). Changes in fluorescence are expressed as % control.

F. TUNEL STAINING Deoxynucleotidyl transferase (TdT) dUTP nick-end labeling): The in-situ terminal deoxynucleotidyl transferase (TdT) dUTP nick-end labeling assay recognizes the free 3’ ends of fragmented DNA. A fluorescent or peroxidase tag is attached to the fragmented 3’ ends and used as a marker of nuclear fragmentation and apoptosis. Staining was performed following the procedure included in the ApopTag kit (Intergen, Purchase, NY) as previously described (Jones et al., 2000). Briefly, cells grown on glass coverslips were fixed with 4% paraformaldehyde in phosphate buffered saline (PBS), and incubated in a solution of proteinase K (5 µg/ml) for 20 min, at 37°C. Cells were washed with PBS and 50 µL aliquots of TdT substrate was added and samples were incubated for 60 min at 37°C followed by incubation in stop/wash buffer. Samples were then exposed to anti- digoxigenin peroxidase antibody followed by staining with DAB/H2O2 in PBS. Coverslips were counter-stained with hematoxylin and mounted on glass superfrosted microscope slides for visualization. Cell counts were made in four random fields at 40X magnification and data was expressed as the percent apoptotic cells.

G. STATISTICS Data are expressed as the mean ± SEM. A one-way analysis of variance (ANOVA) was performed on the data and multiple pairwise comparisons were made using Student Newman-Kuels tests with a modified Bonferroni correction for multiple comparisons. Relevant comparisons for each set of results are detailed within the figures legends and represented by symbols within the figures; non-relevant significant differences within each data set are not reported Differences within and between treatment groups were considered significant at p < 0 .05.

58 CHAPTER 3

ACIVICIN INCREASES THE CONCENTRATION OF α-MeDA THIOETHERS IN THE BRAIN AND POTENTIATES SEROTONERGIC NEUROTOXICITY FOLLOWING PERIPHERAL ADMINISTRATION OF MDMA

I. INTRODUCTION AND RATIONALE The systemic metabolism of MDA and MDMA results in mono- and bis-substituted α-MEDA and N-Me-α-MeDA thioether metabolites that may contribute to the serotonergic neurotoxicity of the parent drugs. Consequently, a transport mechanism must exist that allows the hydrophilic metabolites to pass from the blood into the brain. GSH (Kannan et al., 1990) and a variety of thioether conjugates (Monks and Lau, 1997 [review]; Miller et al., 1996; Patel et al., 1993) are taken up into the brain by intact GSH transporters located on brain endothelial cells. For example, inhibiting γ-GT with acivicin potentiates both the brain uptake index (BUI) of [3H]-5-(GSyl)-α-MeDA following intravenous administration (Miller et al., 1996) and the neurotoxicity of MDA and MDMA (Bai et al., 2001; MDMA data is reported in the present chapter) presumably by increasing the available pool of intact GSH conjugates. Animals were treated with acivicin prior to MDMA administration (10 mg/kg; sc) and brain neurotransmitter levels were analyzed seven days later with HPLC-CEAS. Although these findings suggest that α-MeDA and N-Me-α-MeDA thioether metabolites penetrate the BBB, there are several inherent variables including, the route of administration and method of measurement (see Chapter 3 discussion) in the experimental model that may influence the results. Identifying these metabolites in the brain of an awake and free-moving animal following peripheral administration of MDMA is important. Therefore, experiments described in this chapter were designed to test the hypothesis that α-MeDA thioethers formed during the systemic metabolism of MDA and MDMA access the brain via GSH transporters located in BBB endothelial cells. Microdialysis was used to collect extracellular fluid

59 samples from the brains of animals treated with MDMA, or, in order to examine the involvement of γ-GT, acivicin plus MDMA. Samples were then analyzed with LC- MS/MS and HPLC-CEAS for the presence of GSH/NAC α-MeDA and N-Me-α-MeDA conjugates. Microdialysis permits the assessment of drug-induced biochemical changes in the living and free-moving animal, therefore providing a reliable technique for identifying drugs and drug metabolites in the brain while eliminating the inherent problems associated with tritiated compounds and anesthetized animals. Extracellular fluid samples are collected from a specific brain region and analyzed with a variety of analytical methods. Previous studies using microdialysis techniques to monitor MDMA-induced biochemical changes in the brain have focused on the changes in endogenous neurotransmitters, including 5-HT (Nixdorf et al., 2001; Koch and Galloway, 1997), DA (Nixdorf et al., 2001; Nash and Brodkin, 1991; Nash et al., 1990) and acetylcholine (Acquas et al., 2001; Fischer et al., 2000). Interestingly, although several dialysis studies report MDMA-induced changes in neurotransmitter and glucose/glycogen (Darvesh et al., 2002) levels, data reporting the concentration of MDMA and/or MDMA metabolites in the brain following peripheral administration is lacking. One exception, and perhaps one of the most compelling reports to suggest that the systemic metabolism of MDMA is required to elicit a neurotoxic response (Esteban et al., 2001), measured the concentration of MDMA in the extracellular space of the hippocampus following a single peripheral injection. Estimated concentrations of MDMA in the hippocampal extracellular space reached a peak of approximately 20µM one hr following administration (ip; 15 mg/kg) before rapidly decreasing. Perfusion of MDMA (doses sufficient to produce brain MDMA concentrations similar to peripheral administration) directly into the brain failed to reproduce the long-term neurotoxicity characteristic of MDMA. Surprisingly, to our knowledge, no one has reported the use of microdialysis to identify MDMA metabolites in the brain following peripheral administration of the parent drug. Therefore, because we suspect that the serotonergic neurotoxicity of MDA and MDMA may be due to the

60 systemic metabolism of the parent compound, we used microdialysis followed by HPLC- CEAS and LC-MS/MS analysis in an effort to identify the thioether metabolites of MDMA in the extracellular space of the striatum following peripheral administration (20 mg/kg; sc). We suspect that the α-MeDA and the N-Me-α-MeDA thioethers are substrates for γ- GT and gain access to the brain via GSH transporters. To examine the involvement of γ- GT on striatal metabolite concentrations and long-term depletions in 5-HT and 5-HIAA, a group of animals was treated with acivicin (18 mg/kg; ip) prior to MDMA administration. Once in the brain, the thioether metabolites likely contribute to the serotonergic neurotoxicity of the parent amphetamine. Thus, it follows that the concentrations of metabolites should be positively correlated to the degree of neurotoxicity. Consistent with other studies examining long-term neurotoxicity (Esteban et al., 2001; Bai et al., 1999, 2000) animals were euthanized 7 days after treatment, brains were dissected into multiple regions and neurotransmitter levels were analyzed with HPLC-CEAS. Pearson correlation coefficients were then determined to describe the relationship between metabolite concentration and neurotoxicity.

61 I. RESULTS A. ACIVICIN POTENTIATES MDMA-INDUCED LONG-TERM DECREASES IN BRAIN 5-HT AND 5-HIAA CONCENTRATIONS

The extent of MDMA neurotoxicity is dependent upon the particular dosing regimen employed. Single subcutaneous injections of MDA and MDMA result in a steep dose response curve for 5-HT depletion. Thus, in our studies, a single dose of MDMA (20 mg/kg; sc) depletes brain 5-HT concentrations to 47-64% of control values (data not shown), consistent with other reports (Schmidt et al., 1990). By reducing the dose of MDMA (10 mg/kg), 5-HT and 5-HIAA concentrations are minimally affected, creating an experimental model in which potentiation is readily observed. Pretreatment with acivicin (18 mg/kg) potentiated MDMA-induced serotonergic neurotoxicity. Specifically, MDMA (20 mg/kg) caused modest (18-29%) but significant decreases in brain 5-HT concentrations, decreases that were amplified (52-64%) in acivicin pretreated animals (Figure 3.1). MDMA decreased (19-23%) brain 5-HIAA concentrations (Figure 3.2) and again, pretreatment with acivicin potentiated (41-56%) the effect. acivicin alone had no statistically significant effects on either 5-HT or 5-HIAA concentrations. Of the four brain regions examined, the striatum and hippocampus appeared to be the more sensitive to acivicin /MDMA-mediated decreases in 5-HT and 5- HIAA concentrations. However, significant decreases were also observed in the hypothalamus and cortex, suggesting that MDMA neurotoxicity occurs in areas of the brain densely innervated with serotonergic neurons and rich in SERT proteins (Choi et al., 2000). Ambient body temperature may play a critical role in MDMA-mediated neurotoxicity. For example, inducing hypothermia protects against MDMA-mediated neurotoxicity while raising body temperature potentiated the neurotoxicity of MDMA (Malberg et al., 1996). Consequently, this raised the possibility that the potentiation of MDMA neurotoxicity by acivicin was due to the effect of acivicin on ambient body temperature. Perhaps, acivicin was inducing hyperthermia, which, in turn, potentiated the

62

80 *,† 70 *,† *,† 60 *,†

50

40

30 * * * 20

% decrease in 5-HT concentrations 10

0 Hippocampus Hypothalamus Cortex Striatum

Figure 3.1: Effect of acivicin (18 mg/kg, i.p.) pretreatment on 5- HT concentrations, 7 days following MDMA (10 mg/kg, s.c.) administration. Decreases in 5-HT concentrations are expressed as the mean ± SE (N=6). Absolute values for 5-HT concentrations in the striatum, cortex, hippocampus, and hypothalamus in control animals were 2.44 ± 0.09, 1.17 ± 0.07, 1.53 ± 0.10, and 3.96 ± 0.16 pmol/mg tissue, respectively. Controls (error bars only), acivicin treated animals (hatched bars), MDMA-treated animals (open bars), acivicin/MDMA treated animals (black bars). 5-HIAA concentrations in the acivicin/MDMA treated rats were significantly different from those in the control and MDMA treated rats. The F and the P values are [F (3,20) ± 9.026, p = 0.0006], [F (3,20) ± 16.925, p = 0.0001], [F (3,20) ± 14.070, p = 0.0001], and [F (3,-20) ± 13.597, p = 0.0001] for the striatum, cortex, hippocampus, and hypothalamus, respectively. Values significantly different from controls (*) and MDMA (†) at p < 0.05.

63

70 *,† 60 *,† *,† 50 *,†

40

30 * * * * 20

% decrease concentrationsin 5-HIAA 10

0 Hippocampus Hypothalamus Cortex Striatum

Figure 3.2: Effect of acivicin (18 mg/kg, i.p.) pretreatment on 5- HIAA concentrations, 7 days following MDMA (10 mg/kg, s.c.) administration. Decreases in 5-HIAA concentrations are expressed as the mean ± SE (N=6). Absolute values for 5-HIAA concentrations in the striatum, cortex, hippocampus, and hypothalamus in control animals were 2.03 ± 0.10, 1.06 ± 0.08, 1.09 ± 0.07, and 1.64 ± 0.11 pmol/mg tissue, respectively. Controls (error bars only), acivicin treated animals (hatched bars), MDMA- treated animals (open bars), acivicin/MDMA treated animals (black bars). 5-HIAA concentrations in the acivicin/MDMA treated rats were significantly different from those in the control and MDMA treated rats. F and the P values are [F (3,20) ± 9.942, p = 0.0003], [F (3,20) ± 11.746, p = 0.0001], [F (3,20) ± 12.097, p = 0.0001], and [F (3,20) ± 8.444, p = 0.0008] for the striatum, cortex, hippocampus, and hypothalamus, respectively. Values significantly different from controls (*) and MDMA (†) at p < 0.05.

64 neurotoxic effect of MDMA rather than having a direct effect on the concentration of thioether metabolites gaining access to the brain. We therefore measured body temperature (rectally) every 30 minutes following administration of MDA, MDMA, acivicin, or a combination of MDA or MDMA and acivicin. MDA and MDMA significantly increased body temperature over control animals (Figure 3.3). Acivicin protected against MDA and MDMA-induced increases in body temperature and surprisingly, acivicin alone actually decreased body temp during the 2 hours immediately following administration. These results confirm that the potentiating effect of acivicin on MDMA-mediated neurotoxicity is via the inhibition of γ-GT and not due to any changes in body temperature.

B. LC-MS/MS IDENTIFICATION OF THIOETHER METABOLITES IN BRAIN FOLLOWING PERIPHERAL ADMINISTRATION OF MDMA

Tandem LC-MS/MS analysis provided convincing evidence indicating the presence of GSH and N-acetyl metabolites of α-MeDA and N-Me-α-MeDA in the brain dialysate of animals treated peripherally with MDMA. Consistent with the view that the thioethers cross the BBB via the endothelial cell GSH transporter, pretreatment with acivicin increased both the ability to detect the metabolites with the MS/MS and the concentration of the metabolites quantified by HPLC-CEAS, indicating that the thioether metabolites are substrates for γ-GT. Tandem mass spectrometry (LC-MS/MS) has been used successfully to detect MDMA and other amphetamine designer drugs and metabolites in a variety of biological samples, including blood, urine, and sweat (Clauwaert et al., 2000; Maurer et al., 2000; Kintz et al., 1999; Nordgren and Beck, 2003). Therefore, we analyzed dialysate samples for the presence of α-MeDA and N-MeDA-α-MeDA thioethers following peripheral administration of MDMA by LC-MS/MS. MS2 fragmentation patterns consistent with those of the standards were detected in the majority of MS2 scans of animals treated with MDMA, suggesting the presence of MDMA, and both GSH and NAC metabolites.

65

41

40 * 39 ** * 38 *

37

36

35

34

Body temperature (Celsius)Body 33

32

31 Baseline 0.5 1 1.5 2 2.5 3 3.5 4 Time (hr)

Figure 3.3: Effect of MDA, MDMA, and MDMA + acivicin on rectal body temperature. Animals (n = 6) were treated with vehicle (control; !), MDA (20mg/kg; "), MDMA (20 mg/kg; !), acivicin (18 mg/kg; "), MDA + acivicin (#), or MDMA + acivicin; $) and rectal body temperature was measured every 30 min for 4 hours. MDA and MDMA increased ambient body temperature which was attenuated with acivicin. (*) represents significant differences between MDA/MDMA groups and control, acivicin and MDMA + acivicin treated animals; p < 0.5.

66 Moreover, samples taken prior to drug administration (baseline) never provided fragmentation patterns with any similarity to standard samples, indicating that presence of the drug affected the composition of the dialysate and lead to similar fragmentation profiles as standards (Table 3.1). For example, consistent with previous reports (Clauwaert et al., 2000; Wood et al., 2001; Mortier et al., 2002), the MS2 scan fragments MDMA (m/z = 194) into two major fragment ions with m/z ratios of 163 and 58 (Figure 3.4). Similar to the proposed fragmentation pattern of paramethoxyamphetamine (Mortier et al., 2002), the loss of an ion with m/z = 58 likely corresponds to the fragmentation of the C7-C8 bond on the side chain of the parent compound (Figure 3.4), consequently, leaving a peak ion of m/z = 163. These peak ions were never detected in baseline samples taken prior to MDMA administration. On the other hand, following treatment, distinct similarities in fragmentation profiles are observed for the standards and dialysate samples (Figure 3.5), indicating the presence of MDMA in the brain. The most consistent evidence for the presence of the drugs in the brain obtained by LC-MS/MS was for MDMA, however, similar results are seen for each of the α-MeDA and N-Me-α-MeDA metabolites. For instance, the characteristic MS2 fragmentation profile for a variety of glutathione adducts, including 5-(GSyl)-N-Me-α-MeDA (Figure 3.6), depicts the loss of glutamate (m/z = 129) and glycine ions (m/z = 57), products of peptide bond fragmentation, and an ion corresponding to GSH minus the S residue (m/z = 274), a product of the cleavage of the S-C bond of the cysteine residue (Multib et al., 2000; Guan et al., 2003; Grillie and Hua, 2003;). Again, this pattern was never detected in any of the baseline samples. However, following MDMA, major ion peaks similar to those seen in standards were observed, including the loss of both glutamate and glycine ions (Figure 3.7). Mercapturic acids, including 5-(NAC)-N-Me-α-MeDA (Figure 3.8), are fragmented at the S-C bond of the cysteine residue resulting in loss of an ion with m/z = 130 (Lang et al., 2000; Urban et al., 2003). Baseline and control samples never demonstrated ion peaks corresponding to the loss of m/z = 130. In contrast, 60 minutes following MDMA administration, MS2 fragmentation patterns similar to those of standards were observed

67

MDMA + Control MDMA AT-125 (8) (11) (12) 5-(GSyl)-N-Me-α-MeDA 0 5 7 2,5-bis(GSyl)-N-Me-α-MeDA 0 6 8 5-(NAC)-N-Me-α-MeDA 0 8 10 2,5-bis(NAC)-N-Me-α-MeDA 0 7 9

5-(GSyl)-α-MeDA 0 3 5 2,5-bis(GSyl)-α-MeDA 0 2 5 5-(NAC)-α-MeDA 0 4 6 2,5-bis(NAC)-α-MeDA 0 4 7

Table 3.1: Summary of the number of animals demonstrating evidence for α-MeDA and N-Me-α-MeDA thioethers in striatal dialysate samples identified by LC-MS/MS. Fragmentation patterns for dialysates were compared to standard samples and similarities were used to suggest certain observable trends (see text). Acivicin increased the likely hood of detecting metabolites in the brain dialysate samples. Chi square analys was performed on the data to test the significance between MDMA and MDMA + acivicin groups, however, no significant differences were observed; p < 0.05.

68

Figure 3.4: LC-MS/MS scans of MDMA standard in ACSF. (A) Full MS scan. (B) Zoom of Full MS demonstrating the presence of MDMA. (C) MS/MS scan of the dominant peak (m/z = 194) displaying the fragmentation pattern of ions of m/z = 163 and 58 characteristic of MDMA. The proposed fragmentation pattern of MDMA is presented below.

O NHCH3

CH O 3

163

69 1500 203.8 200 1432.4 1400 194.0 194.8 190 1372.7 194.7 1300 1301.5 180 180.0 194.6 1206.7 1200 1196.0 170 165.0 164.0 194.4 1110.5 163.0 160 162.4 1106.0 1100 1073.9 154.8 194.2 150 1012.2 1000 989.7 968.3 140 929.3 194.0 900 863.8 135.1 852.3 130 838.4 800 193.8 788.0 m/z m/z m/z 120 758.3 117.8 175.2 700 702.5 193.6 110 652.4 106.9 600.8 600 582.4 100 576.3 193.4 550.5 510.6 500 90 472.5 193.2 413.9 494.5 80 400 312.4 193.0 70 300 288.0 217.6 242.8 192.8 58.0 200 199.8 192.7 50 120.8 100 192.6 A. B. C. 40 60

0 0 0

80 60 40 20 80 60 40 20 80 60 40 20

100 100 Relative Abundance Abundance Relative Relative

70

Figure 3.5: LC-MS/MS scans of MDMA in brain dialysate samples. (A) Average of 3 separate spectra for MS/MS of the dominant peak, m/z = 194, prior to treatment (baseline) demonstrating no similarities in fragmentation pattern compared to control. (B) MS/MS scan of the dominant peak (m/z = 194) 20 min following MDMA (20 mg/kg; sc) administration displaying a fragmentation pattern of peak ions of m/z = 163 and 58 characteristic of MDMA.

71

280 200 195.3 260 190 193.0 180 240 177.0 175.0 170 164.0 220 163 163.0 160 162.3 150 200 149.0 194.2 140 180 179.4 176.0 175.1 130 131.1 m/z 164.9 160 m/z 121.0 120 151.2 149.2 110.7 135.2 203.2 110 140 136.4 135.0 133.2 100 123.2 120 90 109.0 100 80 99.7 95.0 87.0 70 80 67.0 95.0 105.7 60 58 58.0 60 50 A. B. 40 40 0

30 40 20 10 0 50

80 60 40 20 100 Relative Abundance Abundance Relative Relative

72

Figure 3.6: LC-MS/MS scans of standard 5-(GSyl)-N-Me-α- MeDA in ACSF. (A) Full MS scan demonstrating a dominate peak of m/z = 487. (B) MS/MS scan of the dominant peak (m/z = 487) displaying a fragmentation pattern of ions characteristic of mercapturic catabolism, including loss of a glutamate (GLU; 129), glycine (GLY; 57), and GSH minus S (274). The characteristic fragmentation pattern of GSH conjugates is shown below. Dotted lines represent MS2-induced fragmentations producing ions with m/z’s of 129, 57, and 274.

2 HO 3 NCH3 1

4 6 CH3 HO 5 274 S O O O N HO N OH

NH2 O

57 129

5-(GSyl)-N-Me-α-MeDA

73 500 495.4 488.3 488.3 480 487.2 486.3 469.0 460 463.4 488.2 450.8 440 429.0 420 488.1 415.1 404.3 400 390.0 488.0 389.0 380 387.7 - (GLU) 129 372.2 367.2 359.3 360 358.1 487.9 354.4 340 327.3 487.8 320 315.9 441.3 m/z m/z 300 300.5 487.7 289.1 484.3 280 487.6 280.1 487.6 260 - 57 (GLY) 240 487.5 227.8 220 213.5 200 487.4 183.9 180 165.0 487.3 163.2 238.2 160 140 487.2 A. B. - 274 (GSH w/o S) 120

0 0

90 80 70 60 50 40 30 20 10 90 80 70 60 50 40 30 20 10 100 100

Relative Abundance Abundance Relative Relative

74

Figure 3.7: LC-MS/MS scans of 5-(GSyl)-N-Me-α-MeDA in brain dialysate sample. (A) MS/MS of m/z = 487 prior to treatment with MDMA (20 mg/kg; sc), baseline. (B) MS/MS scan of the dominant peak (m/z = 487) 20 min following MDMA administration displaying a fragmentation pattern of ions characteristic of mercapturic catabolism and similar to the standard fragmentation patterns, including loss of a glutamate (GLU; 129) and glycine (GLY; 57). Boxed numbers represent ions observed in both standard and treated samples.

75

500 488.2 470 486.9 468.7 480 485.9 460 460 450 447.9 440 440 440.8 428.9 420 426.6 430 400 391.2 420 388.3 380 386.5 411.3 410 370.6 411.4 367.9 360 359.5 405.6 400 350.5 395.5 340 390 389.6 320 380 m/z m/z 300 371.9 370 292.4 367.6 280 360 358.8 260 -129 (GLU) -129 350 350.6 254.8 347.3 240 237.4 340 220 330 327.5 200 320 180 310 -57 (GLY) 160 300 298.8 292.9 140 290 290.7 A. B. 120

0 0

90 80 70 60 50 40 30 20 10 90 80 70 60 50 40 30 20 10 100 100

Relative Abundance Abundance Relative Relative

76

Figure 3.8: LC-MS/MS scans of standard 5-(NAC)-N-Me-α- MeDA in ACSF. (A) Full MS scan demonstrating a dominate peak of m/z = 343. (B) MS/MS scan of the dominant peak (m/z = 343) displaying a fragmentation pattern of ions characteristic of mercapturic acids, including the loss of N-acetyl cysteine (NAC) minus the S residue. The characteristic fragmentation pattern of NAC conjugates is shown below. Dotted lines represent MS2- induced fragmentations producing ions with m/z of 130.

2 HO 3 NCH3 1

4 6 CH3 HO 5 -130 S O

OH N H3C

O 5-(NAC)-N-Me-α-MeDA

77

0

5

3

2 2 2

. . .

4 4 4

4 4 4

0 0 0

9

3 3 3

4 4 4

4

3 3 3

3

1 1 1

. . .

) )

9 )

5 5 5

8

.

2 2 2

4

7

3 3 3

O O O

3

4

0 0 0

8 8 8

2 2 2

3

. . .

2 2 2

2

.

2 2 2

3 3 3

H H H

7

1 1 1

8 8 8

. . .

( ( (

4 7

3 3 3

1 1 1

3 4

1 1 1

3

8 8 8

3 3 3

1 1 1

9 9 9

- - -

. . .

0 0 0

4 4 4

6

0 0 0

9 9 9

4

9 9 9

3 3 3

6

. . .

2 2 2

3

.

3 3 3

5

9 9 9

4

2 2 2

7 7 7

3

. . .

5

5 5 5

) ) )

4

8 8 8

3

+ + +

2 2 2

0 0 0

8 8 8

0 0 0

3

a a a

. . .

.

2 2 2

1 1 1

4

N N N

7 7 7

4

9 9 9

4

. . .

( ( (

2 2 2

3

4

9 9 9

3

1 1 1

6 6 6

. . .

2 2 2

2 2 2

9 9 9

2 2 2

6 6 6

0

0 0 0

.

- - -

2 2 2

6 6 6

3

3

2 2 2

4

4

8 8 8

. . .

3

3

2 2 2

5 5 5

1

2 2 2

.

2

2

9 9 9

4

4

. . .

0 0 0

3

3

7 7 7

4 4 4

3 3 3

2 2 2

2 2 2

9 9 9

) ) )

. . .

1

3 3 3

0 0 0

4

2 2 2

3

3 3 3

2 2 2

4

.

1 1 1

0 0 0

. .

0 .

0 0 0

( ( (

5 5 5

4

z z z

2 2 2

z

/ / /

0

1 1 1

3 /

9 9 9 2 2 2

. . .

4

2 2 2

m m m

C C C

m

3 3 3

3

1 1 1

0 0 0

A A A

. . . 2 2 2

7 7 7

0 0 0

N N N

9

2 2 2

0 0 0

- - -

3

. . .

0 0 0

3

6 6 6

0 0 0

9 9 9

2 2 2

1 1 1

9 9 9

. . .

8

3 3 3

3

8 8 8

0 0 0

3

. . .

1 1 1

3 3 3

0 0 0

8 8 8

9 9 9

. . .

8 8 8

1 1 1

0 0 0

1 1 1

7

8

8 8

. 8

3

1 1

6 1

3

3

3

0 0 0

. . .

6

2 2 2

0 0 0

3

6 6 6

6 6 6

3

1 1 1

1 1 1

0 0 0

. . .

5

9 9 9

) ) )

3

4 4 4

3

1 1 1

0 0 0

0 0 0

3 3 3

4 4 4

3 3 3

1 1 1

. . .

1 1 1

4

( ( (

1 1 1

3

0 0 0

3 3 3

. . .

3

1 1 1

0 0 0

C C C

3 3 3

1 1 1

0 0 0

A A A

. . .

9 3

0 0 0

.

1 1 1

3

2 2 2

N N N

2 2

2 2

3

1 1 1

1 1

3 1

3

9 9 9

. . .

5 5 5

2

1 1 1

7

3

.

1 1 1

3

1

9 9 9

. . .

3

0 0 0

3 3 3

3

0 0 0

0 0 0

1 1 1

. . .

1 1 1

.

1

3

B B B

A 3

0

0 0 0

3

8 8 8

3

0 0 0

0 0 0 0 0 0

0 0 0 0 0 0

0

0 0 0

0 0 0

0

0

0 0

4 4 4 2 2 2

8 8 8 6 6 6

0 0

1 1 1

8 6

4 2

1 e e c c a a n n d d n n u u b b

A A e e v v i i t t a a l l e e R R

78

Figure 3.9: LC-MS/MS scans of 5-(NAC)-N-Me-α-MeDA in brain dialysate sample. (A) MS/MS of m/z = 343 prior to treatment with MDMA (20 mg/kg; sc), baseline. (B) MS/MS scan of the dominant peak (m/z = 343) 60 min following MDMA administration displaying a fragmentation pattern of ions characteristic of mercapturic catabolism and similar to the standard fragmentation patterns, including loss of N-acetylcysteine (NAC). Boxed numbers represent ions observed in both standard and treated samples.

79 3

2

.

.

4

4

4

3

4

2

.

. 3

3

2

2

0

4

0

4

4

3

4

3

3

3

5

.

9

5

.

2

3

3

0

2

0

3 2

2

3

3

1

3

.

.

3

3

1

1

2

3

.

3

3

.

7

6

0

7

0

.

0 3

.

3

5

0

0

5

0

0

0

0

3

3

3

3

3

.

3

9

.

9

6

2

8

9

. 2

4

8

0

0

2

8

8

5

2

.

2

5

7

2

0

0

6

6

2

2

2

.

3

5

2

3

.

0

0

0

4

4

4

2

2

2

8

.

2

0

0

2

2

2

2

z

z

/

2 /

7 2

.

m

m

4

1

2

)

0

3

0

1 0

0

( 0

2

2

C

A

N

-

0

0

8

8

1

1

0

0

6

6

1

1

0

0

4

4

1

1

0

0

2

2

1

1

0

0

0 .

.

0

1

1

A

A

0

0

8

8

0

0 0

0 0

0 0

0 0 0

0

0

0

0 0

0 0

0 0 0

0

0

9 8

7 6

5 4

3 2 1

0

0

9

8

7 6

5 4

1 3 2

1

d d a a c c n n e e 1 A A u u

b b n n a a l l v v i i t t e e R R e e

80 observed, including the loss of an ion with m/z = 130 (Figure 3.9). Table 3.1 summarizes the data and reports the number of animals for which evidence of the presence of each of the eight metabolites was observed. Several trends were observed for the presence or absence of individual metabolites, including; i) GSH metabolites were detected in samples immediately following MDMA administration, whereas NAC metabolites were detected at the later time points, ii) NAC metabolites, being more stable, were easier to identify than the GSH conjugates, iii) acivicin increased the likelihood of observing both the GSH and NAC metabolites.

B. HPLC-CEAS QUANTIFICATION OF THIOETHER METABOLITES IN STRIATAL DIALYSATE SAMPLES FOLLOWING PERIPHERAL ADMINISTRATION OF MDMA.

MS/MS analysis of the brain dialysates provided preliminary evidence identifying the both the GSH and NAC metabolites, however, inherent problems such as sensitivity and noise prohibited drawing any concrete conclusions. On the other hand, because of its increased sensitivity (fmol), HPLC-CEAS permitted a more in-depth analysis of the dialysates. By comparing the AUC of known standards against dialysate samples, metabolite concentrations at several time points following MDMA administration were calculated. None of the metabolites were detected in any of the baseline or control samples. However, following peripheral administration of MDMA, the concentrations of 5-(GSyl)-N-Me-α-MeDA and 2,5-bis(GSyl)-N-Me-α-MeDA detected in the dialysate samples increased rapidly over the first 40 min, peaking at ~34 and ~43 pmol/10µl of sample, respectively (Figure 3.10). Pretreatment with acivicin increased the concentration of both metabolites (Figure 3.11), consistent with the view that the metabolites were gaining access to the brain through the GSH transporter. After the initial increase, both 5- (GSyl)-N-Me-α-MeDA and 2,5-bis(GSyl)-N-Me-α-MeDA concentrations fell dramatically, presumably due to their conversion to mercapturate acids. Accordingly,

81

50

45

40

35

l dialysate sample) l dialysate 30 µ 25

20

15

10

Concentration (pmol/10 5

0 Baseline 0-20 20-40 40-60 60-80 80-100 100-120 min

Figure 3.10: HPLC identification of 5-(GSyl)-N-Me-α-MeDA (%), 2,5bis-(GSyl)-N-Me-α-MeDA ("), 5-(NAC)-N-Me-α- MeDA (!), and 2,5bis-(NAC)-N-Me-α-MeDA (") in brain dialysate samples following MDMA (20 mg/kg; sc) + acivicin (18 mg/kg; ip) administration. The AUC for each metabolite were compared to a standard curve generated with standards. Data are expressed as the absolute concentration (pmol) in 10µl) ± SEM.

82

45 40 A. 35 * 30 25 20 15 10

l dialysate sample) 5 µ 0

B. 40 *

30 *

Concentration (pmol/10 20

10

0 Baseline 0-20 20-40 40-60 60-80 80-100 100-120 min

Figure 3.11: Effect of acivicin on the concentration of 5-(GSyl)- N-Me-α-MeDA (A) and 2,5bis-(GSyl)-N-Me-α-MeDA (B) in brain dialysate samples following MDMA (20 mg/kg; sc). Animals were administered MDMA in the absence (!) or presence (!) of acivicin (18 mg/kg; ip). The AUC for each metabolite were compared to a standard curve generated with standards. Data are expressed as the absolute concentration (pmol) in 10µl) ± SEM. (*) represent significant differences from animals treated with MDMA; p< 0.5.

83 concentrations of the mercapturate metabolites of N-Me-α-MeDA, 5-(NAC)-N-Me-α- MeDA and 2,5-bis(NAC)-N-Me-α-MeDA, rose gradually, prior to reaching a plateau of ~37-38 pmol/10 µl sample 90 min following treatment (Figure 3.10). Acivicin increased the concentration of the both the mono and bis substituted NAC metabolites (Figure 3.12). Identification of both the GSH and mercapturic acid metabolites was more reliable in animals pretreated with acivicin. However, although to a lesser extent, thioether metabolites were also identified in animals treated with only MDMA. Therefore it can be concluded that that the thioether conjugates of MDMA are being transported into the brain.

D. PEARSON CORRELATION BETWEEN STRIATAL METABOLITE CONCENTRATION AND DEGREE OF NEUROTOXICITY

There is a strong positive correlation between the concentration of the N- acetylcysteine metabolites and the degree of neurotoxicity. Seven days after dialysis animals were euthanized, brains were removed, dissected into multiple regions, and tissue samples were analyzed by HPLC-CEAS, for changes in monoamine concentrations. Interestingly, although strongest in the striatum (Figure 3.13), significant Pearson correlations can be drawn between the concentration of NAC metabolites in the dialysate and the decrease in both 5-HT and 5-HIAA in the cortex (Figure 3.14), hippocampus (Figure 3.15), and hypothalamus (Figure 3.16). Thus, animals with greater concentrations of 5-(NAC)-N-Me-α-MeDA in the striatum had consequently greater decreases in brain 5-HT (r = 0.859) and 5-HIAA (r = 0.704). Although a trend was observed suggesting a positive correlation between 2,5-bis(NAC)-N-Me-α-MeDA and decreases in brain 5-HT and 5-HIAA concentrations, no significant correlations were demonstrated (data not shown), possibly due to the increased variability in both the 2,5-bis(NAC)-N-Me-α- MeDA concentrations and long-term depletions in 5-HT and 5-HIAA may provide an explanation. Neither the GSH conjugates of α-MeDA nor N-Me-α-MeDA demonstrated

84 a significant correlation between concentration and neurotoxicity (data not shown). The reason underlying the differences between the thioether and mercapturate metabolites is unknown, however, it is likely that both the increased reactivity of the NAC metabolites and their ability to persist for longer time periods in the brain contribute significantly to the neurotoxicity.

85

50 A. 40

30

20

10 l dialysatesample) µ 0 60 B. 50

40

30 Concentration (pmol/10 20

10

0 Baseline 0-20 20-40 40-60 60-80 80-100 100-120 min

Figure 3.12: Effect of acivicin on the concentration of 5- (NAC)-N-Me-α-MeDA (A) and 2,5bis-(NAC)-N-Me-α-MeDA (B) in brain dialysate samples following MDMA (20 mg/kg; sc). Animals were administered MDMA in the absence (!) or presence (!) of acivicin (18 mg/kg; ip). The AUC for each metabolite were compared to a standard curve generated with standards. acivicin increased the concentration of the NAC conjugates identified in the brain dialysate. Data are expressed as the absolute concentration (pmol) in 10µl) ± SEM.

86 65 A. 60 55 50 45 40 35 % Decrease in 5-HT 30 25 25 27 29 31 33 35 37 39 41 43 [5-(NAC)-N-Me-α-MeDA (pmol)] 45 B. 40

35

30

25 % Decrease5HIAA in 20

15 15 20 25 30 35 40 45 [5-(NAC)-N-Me-α-MeDA (pmol)]

Figure 3.13: Significant correlations between concentrations of 5- (NAC)-N-Me-α-MeDA (pmol) and MDMA (!) and MDMA + acivicin (") and the long term depletions in striatal 5-HT (A) and 5- HIAA (B). [5-(NAC)-N-Me-α-MeDA (pmol)] was calculated by taking the average concentration over the second hour post treatment (eg; between 60-120 min). Pearson’s correlation coefficient for 5-HT and 5-HIAA were r = 0.859 and r = 0.704, respectively. Correlations were considered significant at p < 0.05.

87

70 A.

60

50

40 % Decrease in 5-HT

30

25 27 29 31 33 35 37 39 41 43 45 [5-(NAC)-N-Me-α-MeDA (pmol)] 60 B.

50

40

% Decrease5HIAA in 30

20 20 25 30 35 40 45 [5-(NAC)-N-Me-α-MeDA (pmol)]

Figure 3.14: Significant correlations between concentrations of 5- (NAC)-N-Me-α-MeDA (pmol) and MDMA (!) and MDMA + acivicin (") and the long term depletions in cortical 5-HT (A) and 5- HIAA (B). [5-(NAC)-N-Me-α-MeDA (pmol)] was calculated by taking the average concentration over the second hour post treatment (eg; between 60-120 min). Pearson’s correlation coefficient for 5-HT and 5-HIAA were r = 0.767 and r = 0.636, respectively. Correlations were considered significant at p < 0.05.

88

70 A.

60

50

40 % Decrease5-HT in

30

25 27 29 31 33 35 37 39 41 43 45 [5-(NAC)-N-Me-α-MeDA (pmol)] 70 B.

60

50

40

30 % Decrease in 5HIAA

20 20 25 30 35 40 45 [5-(NAC)-N-Me-α-MeDA (pmol)]

Figure 3.15: Significant correlations between concentrations of 5- (NAC)-N-Me-α-MeDA (pmol) and MDMA (!) and MDMA + acivicin (") and the long term depletions in hippocampal 5-HT (A) and 5-HIAA (B). [5-(NAC)-N-Me-α-MeDA (pmol)] was calculated by taking the average concentration over the second hour post treatment (eg; between 60-120 min). Pearson’s correlation coefficient for 5-HT and 5-HIAA were r = 0.642 and r = 0.691, respectively. Correlations were considered significant at p < 0.05.

89

50 A. 45

40

35

30

% Decrease in 5-HT 25 20 25 27 29 31 33 35 37 39 41 43 [5-(NAC)-N-Me-α-MeDA (pmol)] 45 B. 40

35

30

% Decrease5HIAA in 25

20 20 25 30 35 40 45 [5-(NAC)-N-Me-α-MeDA (pmol)]

Figure 3.16: Significant correlations between concentrations of 5- (NAC)-N-Me-α-MeDA (pmol) and MDMA (!) and MDMA + acivicin (") and the long term depletions in hypothalamic 5-HT (A) and 5-HIAA (B). [5-(NAC)-N-Me-α-MeDA (pmol)] was calculated by taking the average concentration over the second hour post treatment (eg; between 60-120 min). Pearson’s correlation coefficient for 5-HT and 5-HIAA were r = 0.749 and r = 0.701, respectively. Correlations were considered significant at p < 0.05.

90 III. DISCUSSION The transport of drugs and drug metabolites from the capillary lumen of the blood stream into the brain requires the compound to pass though the lipid membranes of the blood-brain-barrier (BBB). Consequently, the ability of a drug to gain access to the brain is dependent on certain characteristics, including water solubility and ionization. Consistent with cellular lipid membranes, the BBB allows the bi-directional passive diffusion of small, nonionic, lipid soluble molecules, whereas large, water soluble molecules must employ carrier transport mechanisms (Oldendorf, 1973). Therefore, as water soluble, ionic compounds, the thioether metabolites of α-MeDA and N-Me-α- MeDA likely require a mechanism of carrier-mediated active transport across the BBB. Specifically, we hypothesize that energy-driven GSH transporters on brain microvessel endothelial cell membranes are involved in the transport of these thioethers into the brain. Sodium-dependent GSH transporters have been demonstrated in an apical localization in mouse and human cerebrovascular endothelial cells (Kannan et al., 1999; 2000) and GSH (Kannan et al., 1990) and a variety of thioether conjugates (Monks and Lau, 1997 [review]; Miller et al., 1995; Patel et al., 1993) compete for this low affinity, saturable brain uptake machinery. Consistent with this view, GSH decreases the concentration of 5- (GSyl)-α-[3H]-MeDA in the brain following intravenous administration (Miller et al., 1995). In addition, the fractional uptake of 5-(GSyl)-α-[3H]-MeDA into brain following a single pass (7.4 ± 0.5%) is comparable to the uptake of GSH (Miller et al., 1995; Cornford et al., 1978). Previous work in our laboratory has provided some preliminary evidence indicating; i) the presence of thioether metabolites of α-MeDA in the brain following peripheral intravenous administration (Miller et al. 1995) and ii) the mechanism of transport across the BBB is likely the GSH transporter (Bai et al., 2001). Unfortunately, although promising and intriguing, the data must be carefully interpreted due to several inherent variables that may have affected end results. First, in studies identifying the metabolites in the brain, animals were administered radiolabeled conjugates intravenously while

91 under heavy anesethia. This raises two questions, are there possible side effects due to the anesethia, and did analysis actually measure the concentration of the metabolites, or the level of radiolabeled precursor? Second, although acivicin did potentiate the neurotoxicity of MDA and MDMA (Figures 3.1 and 3.2) suggesting the uptake of the thioethers into the brain occurred via the GSH transporter, this data represents indirect evidence for the presence of the metabolites in the brain. Therefore, the present studies were designed to specifically identify thioether metabolites in the brain of an awake and freely moving animal following peripheral administration of MDMA. Using microdialysis to collect extracellular samples and LC-MS/MS and HPLC-CEAS to analyze dialysates for the presence of the metabolites, we identified of both thioether metabolites of α-MeDA and N-Me-α-MeDA in the extracellular space of the striatum following peripheral, subcutaneous administration of MDMA. Furthermore, pretreatment with acivicin increased the potential for metabolite identification by LC-MS/MS and, more importantly, significantly increased the concentration of the metabolites in the samples analyzed by HPLC-CEAS (Figures 3.11 – 3,14). A major challenge in detecting thioether metabolites in the brain following peripheral administration is overcoming their extreme reactivity and instability. The inherent instability and rapid rate of metabolism of thioethers to the cysteinyl and N-acetylcysteine metabolites may limit their detection. Following icv administration, 5-(GSyl)-α-MeDA within the brain is rapidly metabolized by γ-GT and dipeptidases to 5-(Cys)-α-MeDA, concentrations of which decline rapidly with the concomitant accumulation of 5-(NAC)- α-MeDA (Miller et al., 1995). In contrast to 5-(GSyl)-α-MeDA and 5-(Cys)-α-MeDA, 5- (NAC)-α-MeDA is eliminated relatively slowly from the brain. The accumulation and persistence of mercapturic acids in the brain may therefore contribute to the long-term neurotoxicity of MDA and MDMA and therefore, identification of the N-acetylcysteine metabolites, rather than the GSH conjugates, is more easily accomplished. Indeed, LC- MS/MS analysis identified several mercapturic acid metabolites of both α-MeDA and N- Me-α-MeDA in the majority of the animals treated with MDMA, while the identification

92 of GSH conjugates was more difficult to accomplish. Nevertheless, thioethers were identified in several animals. Unfortunately, the concentration of these metabolites in the brain is quite small, and we were not able to quantify by LC-MS/MS due to the low (nmol) sensitivity of the mass spectrometer. HPLC with electrochemical detection provided a more sensitive (fmol) method for quantifying the concentrations of the thioether metabolites in the brain dialysate sample following peripheral administration of MDMA. Consistent with the hypothesis that thioether metabolites of N-Me-α-MeDA penetrate into the brain and contribute to the neurotoxicity of MDMA, significant findings include; i) the presence of 5-(GSyl)-N-Me- α-MeDA and 2,5-bis(GSyl)-N-Me-α-MeDA, ii) the presence of the, 5-(NAC)-N-Me-α- MeDA and 2,5-(NAC)-N-Me-α-MeDA, iii) acivicin increases the amount of each metabolite found in the striatal dialysate sample, and iv) there is a positive correlation between the concentration of the metabolites and the degree of long-term neurotoxicity in several brain regions. Interestingly, N-Me-α-MeDA thioethers were identified in more animals than the α-MeDA metabolites with both analytical methods (LC-MS/MS and HPLC-CEAS). Reasons for such a difference are largely unknown, however, nearly 90% of MDMA is de-methylenated directly to the N-Me-α-MeDA whereas only 10% undergoes de-methylation to MDA prior to demethylenation (de laTorre et al., 2000). Although the evidence suggests that the increase in uptake of thioether conjugates and the increased neurotoxicity of MDA and MDMA induced by acivicin is due to its effect on inhibiting γ-GT activity, two other scenarios are possible, i) acivicin has low to moderate affinity for L-amino acid transporters, and ii) acivicin produces hyperthermia, thereby, potentiating the neurotoxicity of MDMA. First, transport of the γ-GT generated cysteinyl conjugates across the BBB into the brain via the L-amino acid transporter (Monks and Lau, 1997) may provide one potential route by which systemically formed water-soluble metabolites gain access to the brain. Amino acid transporters have been identified within the BBB (Oldendorf et al., 1976), and a variety of neurotoxicants are transported into the brain via these carriers. For instance, a Na+-independent L-

93 transporter for neutral amino acids transports dichlorovinylcysteine, a reactive cysteinyl metabolite of the neurotoxicant, dichloroacetylene, across the BBB. However, the uptake of the corresponding GSH conjugate is mediated by an as yet unknown carrier system (Patel et al., 1993). In contrast, the extreme instability of cysteinyl conjugates, which are rapidly N-acetylated, suggests that these metabolites gain access to the brain as intact GSH conjugates rather than as the corresponding cysteinyl conjugate. Furthermore, the significant potentiation of MDA and MDMA neurotoxicity by acivicin, and the acivicin- induced increases in thioether metabolites within the brain suggests that the majority of α-MeDA and N-Me-α-MeDA thioethers are transported across the BBB as intact GSH conjugates. Secondly, the effect of acivicin on MDMA-induced neurotoxicity may be indirect and due to its potential effects on ambient body temperature. MDMA-mediated serotonergic neurotoxicity may be exacerbated by drug-induced hyperthermia (Malberg et al., 1996, Malberg and Seiden, 1998; Colado et al., 1997, 1998; Yaun et al., 2002). Inducing hypothermia protects against MDMA-mediated neurotoxicity, whereas raising body temperature potentiates the neurotoxicity of MDMA (Malberg et al., 1996). However, it should be noted that a decrease in body temperature could slow the chemical reactions required for a variety of neurotoxicants (Bowyer et al., 1993). On the other hand, several studies suggest that MDMA-induced hyperthermia is not essential for MDMA-induced neurotoxicity. For example, pretreatment with fluoxetine, and inhibition of MAO-B with antisense nucleotides, provided protection against MDMA-induced serotonergic neurotoxicity yet did not inhibit MDMA-induced increases in ambient body temperature (Falk et al., 2002). In addition, mazindol, a DAT inhibitor, protects against MDMA-induced neurotoxicity without altering MDMA-induced hyperthermia (Shankaran et al., 1999). The present results indicating that acivicin had no effect on MDMA-induced increases in ambient body temperature (Figure 3.3) are consistent with the hypothesis that neurotoxicity is independent of MDMA’s effects on body temperature.

94 The metabolism of GSH conjugated metabolites within the capillary circulation of the brain by γ-GT and dipeptidases, results in the formation of their corresponding cysteinyl conjugates. For example, 5-(GSyl)-N-Me-α-MeDA is converted to 5-(Cys)-N-Me-α- MeDA. γ-GT, a membrane-bound enzyme located on microvessel endothelial cells with its active site in the extracellular space, catalyzes the first step in the mercapturic acid metabolic pathway. Pretreatment of animals with acivicin increases the uptake of 5- (GSyl)-α-[3H]-MeDA into brain (Miller et al., 1995). Therefore, inhibition of BBB endothelial γ-GT may enhance the delivery of GSH conjugated metabolites into the brain by preventing their metabolic clearance, thereby increasing the pool of intact GSH conjugates available for uptake via GSH transporters. Thus, if α-MeDA and N-Me-α- MeDA thioethers contribute to MDA and MDMA neurotoxicity, inhibition of γ-GT should potentiate the neurotoxicity of MDA and MDMA. Indeed, pretreatment with acivicin significantly potentiated MDA and MDMA-induced decreases in brain 5-HT and 5-HIAA (Bai et al., 2001). Interestingly, quinone-thioethers inhibit a variety of enzymes that utilize GSH as a substrate, including γ-GT (Monks and Lau, 1998). Brain uptake of GSH, for instance, is inhibited by several GSH conjugates, including S-alkyl, sulfobromophthalein, monoethyl ester, probenecid, and ophthalmic acid (Kannan et al., 1992). Thus, α-MeDA or N-Me-α-MeDA GSH conjugates may inhibit γ-GT at the BBB, and subsequently increase the brain uptake of neurotoxic metabolites of MDA and MDMA. Although little is known about human variability of γ-GT, it is likely that variations in both, i) regional γ-GT and GSH transporter activity between individuals and ii) temporal changes in γ-GT activity within an individual contributes to the neurotoxic response. Therefore, the extent to which the conjugates are transported into brain may be an important determinant of individual susceptibility to MDA and MDMA serotonergic neurotoxicity. In summary, the present are consistent with the hypothesis that MDA and MDMA- induced serotonergic neurotoxicity is dependent on i) systemic metabolism of the parent drugs to neurotoxic thioether metabolites, and ii) the ability of the thioethers to gain

95 access to the brain. Inhibition of γ-GT with acivicin significantly potentiated the neurotoxicity of peripherally administered MDMA, presumably by increasing the pool of intact GSH conjugates available for uptake via the GSH transporter. In addition, microdialysis studies led to the identification of thioether metabolites of N-Me-α-MeDA in striatal dialysate samples following peripheral administration of MDMA. Pretreatment with acivicin increased the concentration of the thioether metabolites in the brain, and the concentrations of metabolite in the brain were significantly correlated with the extent of long-term neurotoxicity.

96 CHAPTER 4

IN-VIVO SEROTONERGIC NEUROTOXICITY OF 5-(NAC)-N-ME-α-MEDA FOLLOWING INTRASTRIATAL ADMINISTRATION

I. INTRODUCTION AND RATIONALE

Direct injection of MDA and MDMA into the brain fails to reproduce the acute or long-term neurotoxic effects evident following peripheral administration (Paris and Cunningham, 1992; Molliver et al., 1986; Schmidt et al., 1987; 1988), suggesting that systemic metabolism of the parent drugs plays an essential role in the neurotoxicity of MDA and MDMA. However, several metabolites of MDA and MDMA, including α- MeDA, do not reproduce the “serotonin syndrome” when administered directly into the brain (Walker et al., 1999; Miller et al., 1997; McCann and Ricaurte, 1991). Further phase II metabolism of α-MeDA results in conjugation to GSH, and formation of the putative metabolites, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA (Miller et al., 1995). When administered directly into the brain, 5-(GSyl)-α-MeDA, 2,5-bis(GSyl)-α- MeDA and 5-(NAC)-α-MeDA produce long-term serotonergic neurotoxicity similar to that produced by MDA (Jones and Monks, 2001; Bai et al., 1999). Ultimately, the extent of the neurotoxic response is dependent on the route and frequency of drug administration. For instance, a single intracerebroventricular (icv) injection of 5-(GSyl)- α-MeDA failed to produce any long-term deficits in brain 5-HT (Miller et al., 1996). However, once a multiple dosing regime common to MDA and MDMA studies and adapted to mimic “real life” situations was employed, long-term decreases in 5-HT and 5- HIAA were observed following icv administration of 2,5-bis(GSyl)-α-MeDA (Miller et al., 1997). Interestingly, neither of the mono-substituted metabolites, 5-(GSyl)-α-MeDA and 5-(NAC)-α-MeDA, had any significant effects on the serotonin system, suggesting that the inherent reactivity and instability of the thioether metabolites and the complexity

97 of drug pharmacokinetics following icv administration may affect the ability of the metabolites to reach areas rich in 5-HT nerve terminals. On the other hand, if the route of drug administration is altered, by injecting the metabolites, in a multiple dosing regime, directly into several brain regions rather than into the ventricle, then 5-(GSyl)-α-MeDA, 2,5-bis(GSyl)-α-MeDA and 5-(NAC)-α-MeDA all induce significant long-term decreases in 5-HT and 5-HIAA (Bai et al., 1999). MDA and MDMA demonstrate similar behavioral, pharmacological, and toxicological properties (Ricaurte et al., 1985, Ricaurte et al., 1992). In addition, both amphetamine derivatives are metabolized to α-MeDA (Midha et al., 1978; Lim and Foltz, 1988) and subsequently conjugated by GSH to form GSH-α-MeDA thioethers. Therefore, it is likely that the metabolites of MDA and MDMA share a similar redox and electrophilic properties and mechanism of neurotoxicity. That being said, a significant difference between the amphetamines in the nature of their metabolic pathways indicates that other metabolites, the N-Me-α-MeDA thioethers for example, also contribute to the neurotoxicity of MDMA. For instance, MDMA can be demethylated to MDA, which in turn is O-de-methylenated to α-MeDA, however, the de-methylation to MDA appears to be a minor metabolic pathway, representing ~9% of the initial MDMA concentration (de la Torre et al., 2000). Alternatively, the majority of MDMA is O-de-methylenated to N- Me-α-MeDA, and subsequently conjugated to GSH forming 5-(GSyl)-N-Me-α-MeDA and 2,5-bis(GSyl)-N-Me-α-MeDA (de la Torre et al., 2000). Similar to the GSH-α- MeDA conjugates, the GSH-N-Me-α-MeDA conjugates are metabolized to the mercapturic acids, 5-(NAC)-N-Me-α-MeDA and 2,5-bis(NAC)-N-Me-α-MeDA. Furthermore, the N-Me-α-MeDA conjugates also retain their electrophilic and redox properties. Consequently, in theory, the neurotoxicity of the α-MeDA thioethers should be similar to, if not identical to, the neurotoxicity induced by the N-Me-α-MeDA class of thioethers. Therefore, we hypothesis that 5-(NAC)-N-Me-α-MeDA will produce decreases in brain 5-HT and 5-HT metabolite concentrations following intrastriatal

98 injections. Results provide the first evidence that 5-(NAC)-N-Me-α-MeDA is a serotonergic neurotoxicant and significantly decreases brain 5-HT and 5-HIAA levels following direct administration into the brain.

99 II. RESULTS

A. BEHAVIORAL PROFILE FOLLOWING ADMINISTRATION OF MDMA AND 5-(NAC)-N-ME-α-MEDA

Following intrastriatal administration of 5-(NAC)-N-Me-α-MeDA (7, 14, and 21 nmol; 4X at 12 hr intervals) animals displayed behavioral patterns similar to the peripheral administration of MDMA (20 mg/kg) (Table 4.1). Behavioral effects characteristic of MDA and MDMA induced “serotonin syndrome” (Schmidt et al., 1990; Gordon et al., 1991), including enhanced locomotor activity, hyperactivity, forepaw treading, Straub tail, low body posture and salivation, were observed in 5-(NAC)-N-Me- α-MeDA treated animals, which is consistent with a role for mercapturic acid metabolites in the behavioral changes induced by MDA and MDMA. Interestingly, consistent with the behavioral profile observed when animals are administered the α-MeDA conjugates, 5-(GSyl)-α-MeDA, 2,5-bis(GSyl)-α-MeDA and 5-(NAC)-α-MeDA (Miller et al., 1996; 1997), the effects induced by 5-(NAC)-N-Me-α-MeDA were abrupt, occurring 2-3 minutes after administration. In contrast, in animals treated with MDMA (20 mg/kg; sc) behavioral changes were delayed approximately 15-20 minutes, perhaps indicative of the time required for absorption, metabolism, and distribution. In all treatment groups behavioral effects were most intense following the first injection and with each successive dose, behaviors became less apparent.

B. DECREASES IN 5-HT AND 5-HIAA FOLLOWING INTRASTRIATAL ADMINISTRATION OF 5-(NAC)-N-ME-α-MEDA

Intrastriatal administration of 5-(NAC)-N-Me-α-MeDA (7, 14, and 21 nmol; 4X at 12 hr intervals), in addition to producing a behavioral profile similar to peripheral administration of MDMA, significantly decreased (animals were euthanized seven days after the last injection) brain concentrations of both 5-HT (Figure 4.1) and 5-HIAA (Figure 4.2). The most pronounced decreases in 5-HT and 5-HIAA were observed in the

100

Hyper- Forepaw Low Straub activity treading Posture Tail Salivation

MDMA (20 mg/kg, sc) Y Y Y Y Y

5-(NAC)-N-Me-α-MeDA Y Y Y Y N (7 nmol; intrastriatal) 5-(NAC)-N-Me-α-MeDA Y Y Y Y Y (14 nmol; intrastriatal) 5-(NAC)-N-Me-α-MeDA Y Y Y Y Y (21 nmol; interstriatal)

Table 4.1: Behavioral profile for 5-(NAC)-N-Me-α-MeDA. Behaviors characteristic of MDA and MDMA were observed following intrastriatal multiple injections of 5-(NAC)-N-Me-α-MeDA (4X at 12 hr intervals).

101 60 *,† *,† Control MDMA (20 mg/kg) 5-(NAC)-N-Me-α-MeDA (7nmol) 50 5-(NAC)-N-Me α-MeDA (14nmol) 5-(NAC)-N-Me-α-MeDA (21nmol) * * 40 *

30 *

* 20 * % Decrease in [5HT], pmol/mg tissue

10

0 Striatum Cortex Hippocampus Hypothalamus

Figure 4.1: Effect of MDMA (sc) and 5-(NAC)-N-Me-α-MeDA (intrastriatal) on 5-HT concentrations in the striatum, cortex, hippocampus and hypothalamus seven days following multiple injections (4X at 12 hr intervals). Absolute values for 5-HT concentrations in the striatum, cortex, hippocampus, and hypothalamus in control animals were 2.15 ± 0.09, 1.36 ± 0.08, 1.42 ± 0.13, and 4.02 ± 0.19 pmol/mg tissue, respectively. Data are expressed as the % decrease ± SE in [5-HT] compared to control levels. 5-(NAC)-N-Me-α-MeDA was significantly more potent than MDMA, decreasing 5-HT concentrations in a dose-dependent manner. Within each region, values different from control (*) and MDMA (†) were considered significant at p < 0.01.

102 50 *,† *,† 45

40 *,† * Control 35 * MDMA (20 mg/kg) 5-(NAC)-N-Me-α-MeDA (7nmol) 30 5-(NAC)-N-Me α-MeDA (14nmol) 5-(NAC)-N-Me-α-MeDA (21nmol)

25 *

20 *

15 % Decrease [5-HIAA], Decrease % tissue pmol/mg

10

5

0 Striatum Cortex Hippocampus Hypothalamus

Figure 4.2: Effect of MDMA (sc) and 5-(NAC)-N-Me-α-MeDA (intrastriatal) on 5-HIAA concentrations in the striatum, cortex, hippocampus and hypothalamus seven days following multiple injections (4X at 12 hr intervals). Absolute values for 5-HIAA concentrations in the striatum, cortex, hippocampus, and hypothalamus in control animals were 2.15 ± 0.16, 0.99 ± 0.06, 1.18 ± 0.12, and 1.54 ± 0.14 pmol/mg tissue, respectively. Data are expressed as the % decrease ± SE in [5-HIAA] compared to control levels. 5-(NAC)-N-Me-α-MeDA was significantly more potent than MDMA, decreasing 5-HIAA concentrations in a dose- dependent manner. Within each region, values different from control (*) and MDMA (†) were considered significant at p < 0.01.

103 striatum, which is not surprising given the site of injection. However, 5-HT and 5-HIAA concentrations in the cortex, hippocampus, and hypothalamus in treated animals also decreased in relation to control levels, suggesting that the compound diffused readily throughout the brain. Of these brain regions, the cortex displayed the greatest decrease in 5-HT and 5-HIAA, to levels similar to those observed in striatal tissue In contrast, the effects in the hippocampus and hypothalamus were minimal suggesting that the cortex and striatum share properties which predispose these regions to 5-(NAC)-N-Me-α-MeDA neurotoxicity.

C. 5-(NAC)-N-ME-α-MEDA-INDUCES DECREASE IN DA FOLLOWING INTRASTRIATAL ADMINISTRATION.

Interestingly, in addition to its deleterious effects on serotonergic neurons, 5-(NAC)- N-Me-α-MeDA (7, 14, and 21 nmol; 4X at 12 hr intervals) also had a long-term, dose- dependent effect on the brain’s dopaminergic system following intrastriatal administration. At the highest dose given (21nmol) DA concentrations in the striatum and cortex decreased by ~19% and ~17%, respectively (Figure 4.3). The hippocampus and hypothalamus also demonstrated a moderate decrease in DA concentrations, suggesting the effect may be widespread throughout the brain. Animals displayed a significant degree of variability in the extent of the neurotoxic response. Nevertheless, these results, consistent with previous studies demonstrating a moderate loss of brain DA following intrastriatal administration of GSH and NAC-α-MeDA, (Bai, 2000), suggest that in addition to their effects on the 5-HT system, these compounds may affect long-term DA concentrations. However, further studies should be conducted before any conclusions are drawn concerning the possible dopaminergic neurotoxicity of either these mercaputric metabolites, or the parent drugs, MDA and MDMA.

104 25 Control * MDMA (20 mg/kg) * 5-(NAC)-N-Me-α-MeDA (7nmol) 5-(NAC)-N-Me α-MeDA (14nmol) 5-(NAC)-N-Me-α-MeDA (21nmol) 20

15

10 % Decrease [DA], pmol/mg tissue tissue pmol/mg [DA], Decrease %

5

0 Striatum Cortex Hippocampus Hypothalamus

Figure 4.3: Effect of MDMA (sc) and 5-(NAC)-N-Me-α-MeDA (intrastriatal) on DA concentrations in the striatum, cortex, hippocampus and hypothalamus seven days following multiple injections (4X at 12 hr intervals). Absolute values for DA concentrations in the striatum, cortex, hippocampus, and hypothalamus in control animals were 41.95 ± 3.89, 0.49 ± 0.08, 1.96 ± 0.17, and 1.01 ± 0.1 pmol/mg tissue, respectively. Data are expressed as the % decrease ± SE in [DA] compared to control levels. 5-(NAC)-N-Me-α-MeDA was significantly more potent than MDMA, decreasing DA concentrations in a dose-dependent manner. With each region, values different from control (*) were considered significant at p < 0.01.

105 III. DISCUSSION Consistent with the hypothesis that the systemic metabolism of MDA may be necessary to elicit a neurotoxic response, the thioether acid metabolites of α-MeDA induce long-term serotonergic neurotoxicity following intrastriatal injections (Bai et al., 1999). MDMA, the N-methylated form of MDA, produces behavioral, pharmacological, and toxicological effects similar to MDA; hence, it is likely that the requirement for systemic metabolism also plays a part in MDMA mediated neurotoxicity. Although a small percentage of MDMA is de-methylated to MDA, the majority is directly de- methylenated to N-Me-α-MeDA (de la Torre et al., 2000). Consequently, the GSH and NAC conjugates of N-Me-α-MeDA may contribute to the neurotoxicity of MDMA. Indeed, both the thioether acid metabolites of N-Me-α-MeDA were identified in the brain following peripheral administration of MDMA (Chapter 3). To examine the potential behavioral changes and serotonergic neurotoxicity of 5-(NAC)-N-Me-α-MeDA, brain 5- HT and 5-HIAA concentrations were measured 7 days following a multiple dosing regime (4X at 12 hr intervals) of intrastriatal injections. The doses used in this study likely fall within the range of α-MeDA thioethers levels present in the brain following MDA administration (Miller et al., 1996). Administration of 5-(NAC)-N-Me-α-MeDA (7, 14, and 21 nmol) directly into the striatum produces behavioral changes characteristic of the “serotonin syndrome” (Table 4.1) and significant long-term decreases in striatal 5- HT and 5-HIAA concentrations in brain regions rich in 5-HT nerve terminals (Figures 4.2 and 4.3), and which were greater than those observed following peripheral administration of the parent amphetamine. No significant effects on 5-HT concentrations in brain regions rich in 5-HT cell bodies were observed (data not shown), which is consistent with previous findings suggesting the cell bodies are spared during MDMA-induced neurotoxicity (Ricaurte et al., 1985). Therefore, it is likely that the systemic metabolism of MDMA to N-Me-α-MeDA, the subsequent conjugation to GSH, the transport of 5- (GSyl)-N-Me-α-MeDA into the brain, and finally, the formation of 5-(NAC)-N-Me-α- MeDA, likely play significant roles in the neurotoxic response induced by MDMA.

106 Intrastriatal administration of 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA produces decreases in 5-HT and 5-HIAA in-vivo (Bai et al., 1999) and loss of cell viability and apoptosis in-vitro (Chapter 7). However, it is likely that the mercapturic acid metabolites, due to their longer persistence in the brain and their greater capacity to redox cycle and generate ROS (Monks and Lau, 1997), contribute to the neurotoxicity of the thioethers. Doses of the 5-(NAC)-α-MeDA (Bai et al., 1999) and 5-(NAC)-N-Me-α- MeDA (Figures 4.1, 4.2) required to produce serotonergic neurotoxicity are significantly lower than doses of MDA necessary to produce similar degrees of toxicity although differences in the route of administration must be taken into account. Moreover, 5- (NAC)-α-MeDA is more than 2 orders of magnitude more potent than 5-(GSyl)-α- MeDA (Miller et al., 1997, Bai et al., 1999). For instance, intrastriatal administration of 5-(NAC)-α-MeDA, at a dose as low as 7 nmol, produces behavioral effects similar to MDA and a 50% reduction in striatal 5-HT concentrations. In contrast, concentrations of the GSH conjugates required for neurotoxicity range from 400 to 800 nmol. If approximately 1.6% of MDA is converted to 5-(GSyl)-α-MeDA and 7.5 % of the conjugated metabolite is transported into the brain, then an estimated minimum of ~28 nmol will reach the brain (Miller, 1996). Therefore, because, i) the mercapturic acid metabolites persist longer in the brain than the corresponding GSH conjugates (Chapter 3; Miller et al., 1995) and ii) 5-(NAC)-α-MeDA is a more potent neurotoxicant than its corresponding GSH conjugate, it follows that 5-(NAC)-N-Me-α-MeDA is likely a more potent neurotoxicant than its parent thioether, 5-(GSyl)-N-Me-α-MeDA. Interestingly, 5-(NAC)-N-Me-α-MeDA, in addition to decreasing striatal 5-HT and 5-HIAA concentrations, also had long-term effects on the serotonergic system in the cortex, hippocampus, and hypothalamus. Cortical tissue, like striatal tissue, displayed significant losses of 5-HT and 5-HIAA following intrastriatal administration of 5-(NAC)- N-Me-α-MeDA, whereas moderate decreases were observed in the hippocampus and hypothalamus (Figures 4.1, 4.2). It is not surprising that the greatest neurotoxic response is observed in the striatum, given that this was the site of injection, however, the reason

107 why the cortex demonstrated significant losses of 5-HT and 5-HIAA while minimal effects were observed in the hippocampus and hypothalamus is not known. Interestingly, 5-(GSyl)-α-MeDA, 2,5-bis(GSyl)-α-MeDA, and 5-(NAC)-α-MeDA also produce a type of “cross talk” between the cortex and striatum (Bai et al., 1999). For example, intracortical administration of 5-(GSyl)-α-MeDA, 2,5-bis(GSyl)-α-MeDA, and 5- (NAC)-α-MeDA leads to loses of 5-HT and 5-HIAA in the striatum and vice verse. The cortex and striatum are both innervated primarily by thin and extensively branched serotonergic axons with small fusiform varicosities originating in the dorsal raphe (the “D” system) (Mulligan and Tork, 1988; Tork, 1990). 5-HT neurons in these regions are consequently more sensitive to several neurotoxic substituted amphetamines, including MDA and MDMA (Momounas and Molliver, 1988; Momounas et al., 1991). In contrast, The “M” system consists of thick non-varicose axons, originating in the median raphe nucleus, which primarily sends its serotonergic axonal projections to the hippocampus and hypothalamus (Mulligan and Tork, 1988). In contrast to the adverse effects of substituted amphetamines on the fine axon endings of the “D” system, the thick axons of the “M” system appear to be spared the neurotoxic effects of the amphetamines (Momounas et al., 1991; Molliver et al., 1997) Therefore, a common origin of serotonin projections to the cortex and striatum may provide a mechanism that facilitates neuronal injury to sites distal from the site of drug injection. This may explain why a direct injection into the striatum fails to produce significant serotonergic neurotoxicity in regions that receive serotonergic projections from the median raphe. On the other hand, the greater sensitivity of the hippocampal serotonergic system to 2,5-bis(GSyl)-α-MeDA compared to the cortex following icv administration may be a consequence of the proximity of the ventricle system to the hippocampus, rather than the type of serotonergic axons. Thus, the susceptibility of a particular brain region to the adverse effects of these compounds is likely dependent on multiple factors, including, the type of serotonergic axon targeted and, as discussed above, the route of drug administration.

108 Two major moieties of N-Me-α-MeDA thioethers, the catechol and the GSH-derived side chain, may ultimately contribute to the neurotoxic response induced by the parent amphetamine. However, the finding that α-MeDA, when administered directly into the brain, fails to reproduce the neurobehavioral and neurotoxicological effects of MDA (Miller et al., 1997) suggest that the effects of α-MeDA thioethers are not solely dependent upon the catecholamine moiety, and, more likely, are a consequence of the conjugation to GSH. In support of this view, both the systemic conjugation to GSH and metabolism to mercapturic acids within the brain restore the pharmacological and toxicological properties of the parent drug. Furthermore, it is unlikely that α-MeDA or N- Me-α-MeDA produce neurotoxicity following peripheral administration, given the extent to which such systemically administered catechols would be O-methylated and targeted for elimination from the body. Likewise, due to their rapid metabolism, it is also doubtful that peripheral (oral, ip, sc) administration of the α-MeDA or N-Me-α-MeDA thioether metabolites produces any long-term neurotoxic effects. The disposition of systemically administered catechols and thioethers would be substantially different from that of those generated in-situ and will not reflect the complex pharmacokinetics/pharmacodynamics of metabolite formation following sc administration of the parent amphetamines. Not only did 5-(NAC)-N-Me-α-MeDA have adverse effects on the brain’s serotonergic system, interestingly, intrastriatal injections also lead to moderate long-term decreases in DA concentrations in all brain regions examined (Figure 4.3). These results are consistent with studies demonstrating non-significant, but moderate losses of DA following direct administration of the thioether acid metabolites of α-MeDA into the brain (Miller, 1996; Bai, 2000). Although some evidence suggesting the potential for MDMA-mediated dopaminergic neurotoxicity does exist, to our knowledge, there are no studies confirming any significant adverse effects on the dopaminergic system in humans, non-human primates, or rats. In contrast, in mice, MDMA does, in fact, selectively destroy the brain’s dopaminergic system (Colado et al., 2001, 2004). The basis underlying the difference between neurotransmitter systems may be a reflection of the

109 species differences in metabolism, however, additional studies need to be conducted to further examine this hypothesis and potential for long-term dopaminergic neurotoxicity in humans and rats. Nevertheless, data from our laboratory appears to demonstrate a modest degree of long-term dopaminergic neurotoxicity in rats induced by the thioether acid metabolites under investigation. Indeed, a strong relationship between MDMA-mediated serotonergic neurotoxicity and the DA neurotransmitter system is well documented (Bankson and Cunningham, 2001 [review]; Aguirre et al., 1998; Stone et al., 1988). MDMA appears has effects on the dopaminergic system without adversely affecting long-term DA concentrations. For instance, MDMA induces the release of DA from DA neurons (Koch and Galloway, 1997; Guldelsky and Nash, 1996; Gudelsky et al., 1994; Schmidt et al., 1992a, b, 1994; Nash and Brodkin, 1991) and also inhibits DAT function (Metzger et al., 1998), which taken together, cause substantial increases in extracellular DA concentrations. Moreover, the thioether metabolites of α-MeDA also produce acute changes in DA concentrations following direct administration into the brain, increasing the rate of the brain’s turnover of DA (Miller at el, 1997). Again, little is known about the potential long-term effects that MDA or MDMA may have on the dopaminergic system in humans. However, our results demonstrating that MDMA and 5-(NAC)-N-Me-α-MeDA induced modest decreases in long-term DA concentrations in rats coupled with the reports identifying acute, direct effects of both MDMA and the thioether metabolites of α-MeDA on the DA system, raises the possibility that MDMA may indeed, adversely affect the brain’s dopaminergic system. In summary, we have demonstrated that the mercapturic acid of N-Me-α-MeDA, 5- (NAC)-N-Me-α-MeDA, produces, i) behavioral changes consistent with the “serotonin syndrome” and, ii) long-term serotonergic neurotoxicity in multiple brain regions following intrastriatal administration. These results are consistent with the effects of the parent amphetamine, MDMA, and support the hypothesis that the systemic metabolism of MDMA to its thioether metabolites may be necessary for the development of

110 neurotoxicity. Finally, we demonstrated that peripherally administered MDMA and intrastriatally administered 5-(NAC)-N-Me-α-MeDA produced modest decreases in brain DA concentrations, suggesting the possibility of long-term dopaminergic consequences. Further studies should address the effect of these compounds on the DA system.

111 CHAPTER 5

EFFECT OF 5-(GSYL)-α-MEDA, AND 2,5-BIS(GSYL)-α-MEDA ON CELLULAR UPTAKE OF NEUROTRANSMITTERS IN PRIMARY, JAR, AND hSERT- TRANSFECTED SK-N-MC CELLS

I. INTRODUCTION AND RATIONALE.

The thioether metabolites of α-MeDA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α- MeDA, are potent serotonergic neurotoxicants and may contribute to the neurotoxicity of the parent amphetamines, MDA and MDMA. Taken together, studies described in Chapters 3 and 4, and previous work from our laboratory (Miller et al., 1995, 1997; Bai et al., 1999, 2001) have provided substantial evidence that the metabolites are, i) formed in-vivo, ii) are transported into the brain, and iii) produce behavioral and neurotoxic responses similar to MDA and MDMA. Although the neurotoxicity of 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA has been effectively demonstrated, the precise mechanisms involved remain in question. Therefore, the following chapters describe studies designed to examine the underlying mechanisms of 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α- MeDA including, the involvement of the SERT and the role of DA (Chapter 5), the production of ROS (Chapter 6), and cytotoxicity or apoptosis (Chapter 7), in 5-(GSyl)-α- MeDA and 2,5-bis(GSyl)-α-MeDA-induced serotonergic neurotoxicity. The selectivity that the parent amphetamines exhibit for the serotonin system and 5- HT nerve terminals in particular, suggests α-MeDA thioethers may interact directly with 5-HT neurons. The serotonin transporter (SERT), a Na+/Cl- dependent transport complex of 12 transmembrane domains, is located on 5-HT nerve terminals and is responsible for the re-uptake of serotonin. Consequently, the SERT is a prime target for MDA and MDMA, and indeed, both drugs inhibit SERT activity (Sprague et al., 1998; Schmidt et al., 1987). Inhibition of SERT function (Fleckenstein et al., 1999) and the loss of (Battaglia et al., 1987; Aguirre et al., 1995) SERT proteins are associated with long-term

112 impairments in the serotonergic system, including the degeneration of 5-HT axonal projections and nerve terminals (O’Hearn et al., 1988; Molliver et al., 1990; Callahan et al., 2001). Moreover, fluoxetine, a selective 5-HT reuptake inhibitor, attenuates MDMA- induced neurotoxicity (Aguirre et al., 1998; Malberg et al., 1996) and protects against long-term anxiety and depression, both of which, are consequences of MDMA use (Thompson et al., 2004) indicating a role for the SERT in the MDMA-induced damage to 5-HT nerve terminals. Citalopram, another 5-HT reuptake inhibitor, attenuated the effects of MDMA in humans (Liechti et al., 2000). Finally, the behavioral and toxicological effects of MDMA (hyperactivity and depletion of 5-HT) were absent in SERT-deficient mice (Bengel et al., 1998), suggesting that a functional SERT is essential for MDMA- induced toxicity. Although the importance of the SERT in mediating MDA and MDMA- induced serotonergic toxicity has been firmly established, the precise involvement of the transporter remains unclear. MDMA-induced serotonergic neurotoxicity also appears to be coupled to increases in dopamine (DA) release (Bankson and Cunningham, 2001 [review], Falk et al., 2002; Fernandez et al., 2003). MDMA increases DA release (Guldelsky and Nash, 1996; Gudelsky et al., 1994) and inhibits DAT function (Metzger et al., 1998) thereby causing a substantial increase in extracellular DA. Consequently, there is an excess of extracellular DA available that may contribute to MDA and MDMA serotonergic neurotoxicity. Consistent with this view, functional SERT proteins are capable of transporting DA into SERT-expressing cells (Saldana and Barker, 2004; Faraj et al., 1994; Schmidt and Lovenberg, 1985) and fluoxetine blocks the uptake of DA into hippocampal synaptosomes (Sprague and Nichols, 1995) suggesting that DA is entering the cell via the SERT. The subsequent monoamine oxidase (MAO-B)-mediated oxidation of DA within the 5-HT nerve terminal may contribute to MDMA-induced neurotoxicity (Sprague et al., 1998; Sprague and Nichols; 1995a, b). Interestingly, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA induce acute increases in the rate of DA turnover and extracellular concentrations of DA (Miller et al., 1996, 1997). Therefore, a series of studies were

113 conducted to examine the effects of 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA on the cellular uptake of DA into SERT-expressing, or “serotonergic” cell models. The “serotonergic” cells used in all of the mechanistic studies provided three distinct cell types, displaying slightly different characteristics, yet each expressing high levels of SERT proteins. JAR cells are a SERT-expressing, human placental serotonergic cell line that readily transport 5-HT and serve as a reliable model for serotonergic cells (Ramamoorthy et al., 1993; Simantov and Tauber, 1997). A major advantage of the JAR cell model is the natural expression of SERT. However, these cells are not neuronal in origin. Consequently, care must be taken when extrapolating the data from JAR cells to serotonergic neurons. In contrast, the neuroblastoma cell line, SK-N-MC, is of neuronal origin. Although these cells do not naturally express the SERT or DAT proteins, transient transfection of transporter cDNA is easily accomplished and results in high levels of protein expression. Using transfected SK-N-MC cells permits the attribution of any effects to the presence of a specific transporter. Rat primary striatal and hippocampal cells were also used, for a variety of reasons including, i) MDA, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA induce a microglial response in striatal biopsies and astrogliosis in striatal and hippocampal biopsy cultures; ii) MDA, MDMA and the α-MeDA thioethers significantly decrease 5-HT levels in both the stratum and hippocampus; iii) The striatum and hippocampus are rich in 5-HT projections and display serotonergic axonal damage in response to MDA, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA; and finally, iv) the striatum largely contains a high percentage of DA neurons whereas the hippocampus has a high density of 5-HT neurons (Figure 7.6) permitting a comparison between the different types of cell populations. However, although the use of primary cells conveys a potentially more accurate representation of the actual in-vivo effects, the reliability of attributing a response to a specific transporter is questionable. MDA, MDMA, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA had effects on both 5- HT and DA cellular uptake in all three “serotonergic” cell types.

114 II. RESULTS

A. MDA, MDMA, 5-(GSYL)-α-MEDA, AND 2,5-BIS(GSYL)-α-MEDA INHIBIT 5- HT TRANSPORT INTO MULTIPLE CELL MODELS

1. hSERT-TRANSFECTED SK-N-MC CELLS SK-N-MC cells transiently transfected with either pcDNA, hSERT or hDAT cDNA were used to examine the effect of MDA, MDMA, 5-(GSyl)-α-MeDA, and 2,5- bis(GSyl)-α-MeDA on the cellular uptake of 5-HT. Confirmation of protein expression was demonstrated by western analysis and by quantifying [3H]5-HT and [3H]DA cellular uptake into hDAT and hSERT-transfected cells (Figure 2.2). MDA, MDMA, 5-(GSyl)-α- MeDA, and 2,5-bis(GSyl)-α-MeDA maximally inhibited 5-HT uptake between 2-4 hrs following drug treatment (Figure 5.1). The cellular uptake of 5-HT into untreated hSERT-expressing cells was rapid, and continued to accumulate for 8 hours before slowly returning to baseline levels by 48 hrs. Following 4 hours of exposure, MDA and MDMA inhibited 5-HT uptake by approximately 30% (Figure 5.2). However, 5-(GSyl)- α-MeDA, and 2,5-bis(GSyl)-α-MeDA were more potent inhibitors of 5-HT transport than either of the parent compounds, inhibiting uptake by approximately 60 and 70%, respectively, suggesting that metabolism may contribute to the neurotoxicity of the parent drugs. Interestingly, fluoxetine or nomifensine, a DAT inhibitor, had no significant effect on the inhibition of 5-HT uptake (Figure 5.2). However, consistent with the hypothesis that the SERT is a molecular target for MDA, MDMA, and α-MeDA thioethers, none of the compounds tested had any effect on mock (pcDNA) or hDAT transfected cells (data not shown). Kinetic analysis of the inhibition of 5-HT uptake was preformed on SERT 3 transfected cells. Km and Vmax values for specific [ H]5-HT uptake were determined by 3 saturation transport analysis using increasing concentrations of [ H]5-HT (Figure 5.3). Ki values for each of the compounds were determined by measuring uptake of a single concentration of [3H]5-HT (20 nM) and various concentrations of MDA, MDMA, 5- (GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA. Consistent with the single concentration

115

60

50

40

* 30 * H] 5-HTH] uptake * 3 * * * * * * 20 * * * * * * % Specific [ % Specific * *† *† 10 *† *†

0 0 4 8 12 16 20 24 48 Time (hr)

Figure 5.1: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)- α-MeDA inhibit 5-HT uptake into hSERT-transfected cells. Cellular [3H]5-HT uptake was determined in control (!), MDA (100µM, "), MDMA (100µM, #), 5-(GSyl)-α-MeDA (100µM, ") or 2,5-bis(GSyl)-α-MeDA (100µM, #) treated cells at increasing time points (0.5, 1, 2, 4, 8, 16, 24, and 48 hrs) by liquid scintillation spectroscopy. Data are plotted as the % specific 5-HT uptake and expressed as the mean (N = 4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05.

116

90

80 *† 70 *†

60

50

* 40 * 30

20 % of Inhibition Cellular 5-HT Uptake 10

0 Fluoxetine Nomifensine

Figure 5.2. MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α- MeDA-mediated inhibition of 5-HT uptake into hSERT-transfected SK-N-MC cells; effect of fluoxetine. Cellular [3H]5-HT uptake was measured 4 hrs following treatment of hSERT-transfected cells. Control (black bars), MDA (100 µM; dashed bars), MDMA (100 µM; gray bars), 5-(GSyl)-α-MeDA (100 µM; white bars) or 2,5-bis(GSyl)- α-MeDA (100 µM; checkered bars) data are presented as the % inhibition of 5-HT uptake and is expressed as the mean (N = 4) ± SEM. Groups of cells were incubated with fluoxetine (100µM) or nomifensine (50µM) 20 min prior to treatment. Data are expressed as the percent inhibition of 5-HT uptake compared to control values; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05.

117

Figure 5.3: Kinetics of MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA-mediated inhibition of 5-HT uptake in hSERT-transfected SK-N-MC cells. A. Specific [3H]5-HT uptake was determined by saturation analysis, by treating SK-N-MC cells with increasing concentrations of [3H]5-HT either alone or in the presence of fluoxetine (10 µM) for 30 min, measuring the intracellular accumulation of [3H]5-HT, and taking the difference in

uptake between non-specific and total 5-HT uptake. Km and Vmax are mean values of 4 independent transfections. B. Ki values were determined by incubating cells with 20 nM [3H]5-HT and increasing concentrations of MDA ($, 100µM), MDMA (%, 100µM), 5- (GSyl)-α-MeDA (#, 100µM), or 2,5-bis(GSyl)-α-MeDA (", 100µM) prior to measuring intracellular [3H]5-HT accumulation. The data were analyzed by non-linear least squares curve fit and data are plotted as the % specific 5-HT uptake. 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA were more potent at inhibiting [3H]5-HT µ uptake (Ki = 69 and 49 M, respectively) than either MDA or µ MDMA (Ki = 107 and 102 M, respectively). 95% confidence µ intervals for Ki values are as follows; MDA = 90.63 to 127.3 M; MDMA = 83.58 to 126.9 µM; 5-(GSyl)-α-MeDA = 53.52 to 88.94 µM; 2,5-bis(GSyl)-α-MeDA = 36.58 to 65.14 µM.

118

60 Total

) 50 A.

n

i

e

t

otein) o

r 40

p

g 3

m

/ 30 Specific [ H] 5-HT uptake

n

in/mg pr in/mg i

m

/

l 20

o

m

pmol/mp Non-Specific ( 10

v ( 0 0.0 0.5 1.0 1.5 2.0 [(3H)5-HT] (µM)

100

80 B.

60

40 H] 5-HT Cellular uptake Cellular 5-HT H] 3

20 Specific [ Specific

0 -2 -1 0 1 2 3 4 5 Log [inhibitor] (µM)

119 and time course results, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA were more 3 potent at inhibiting [ H]5-HT uptake (Ki = 69 and 49 µM, respectively) than either MDA or MDMA (Ki = 107 and 102 µM, respectively) (Figure 5.3).

B. SEROTONERGIC JAR CELLS The cellular uptake of 5-HT into untreated JAR cells was rapid and continued to accumulate for 24 hrs. Treatment of JAR cells with MDA or MDMA reduced the cellular uptake of [3H]5-HT by 21% to 23% (Figure 5.4). Consistent with the results in hSERT transfected cells, 5-(GSyl)-α-MeDA or 2,5-bis(GSyl)-α-MeDA were more potent inhibitors of 5-HT cellular uptake, inhibiting the uptake of [3H]5-HT by approximately 36% and 42%, respectively. Pretreatment with fluoxetine potentiated MDA, MDMA, 5- (GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA-mediated inhibition of [3H]5-HT uptake (Figure 5.4), perhaps suggesting that fluoxetine acts at a different site on the SERT than the other compounds (Mortensen et al., 2001). Due to the lack of DA transporters, nomifensine had no effect on 5-HT uptake in JAR cells. MDA, MDMA, 5-(GSyl)-α- MeDA, and 2,5-bis(GSyl)-α-MeDA maximally inhibited 5-HT uptake between 4-8 hrs following drug treatment (Figure 5.5).

Kinetic analysis of the inhibition of 5-HT uptake was performed on JAR cells. Km 3 and Vmax values for specific [ H]5-HT uptake were determined using saturation transport analysis using increasing concentrations of [3H]5-HT. 5-(GSyl)-α-MeDA, and 2,5- 3 bis(GSyl)-α-MeDA were more potent at inhibiting [ H]5-HT uptake (Ki = 66 and 47 µM, respectively) than either MDA or MDMA (Ki = 105 and 97 µM, respectively) (Figure 5.6).

C. PRIMARY RAT STRIATAL AND HIPPOCAMPAL CELLS Both the parent amphetamines and the thioether metabolites of α-MeDA inhibited the uptake of [3H]5-HT into “serotonergic”, or SERT-expressing cells, however, these observations were made in established cell lines. Therefore, examined the potential

120

60 *†

50 *† *†

40 * * 30 * * 20

10 % Inhibition of Cellular 5-HT Uptake

0 Fluoxetine Nomifensine

Figure 5.4: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)- α-MeDA inhibit 5-HT uptake into JAR cells. Cellular [3H]5-HT uptake was measured 4 hrs following treatment of JAR cells. Control (black bars), MDA (100 µM; hatched bars), MDMA (100 µM; dotted bars), 5-(GSyl)-α-MeDA (100 µM; white bars) or 2,5-bis(GSyl)-α- MeDA (100 µM; gray bars) data are presented as the % inhibition of 5-HT uptake and is expressed as the mean (N = 4) ± SEM. Groups of cells were incubated with fluoxetine (100µM) or nomifensine (50µM) 20 min prior to treatment. Data are expressed as the percent inhibition of 5-HT uptake compared to control values; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05.

121

50 *† 45 *† *† *† 40 *† *† *† 35 *† * 30 * * * * 25 * * * * * * * 20 * * 15 *

10

% Inhibition of Cellular 5-HT Uptake5-HT Cellular of Inhibition % 5

0 12481624 Hours

Figure 5.5: Time course for MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA inhibition of 5-HT uptake into JAR cells. Cellular [3H]5-HT uptake was determined in control (&), MDA (100µM, $), MDMA (100µM, $), 5-(GSyl)-α-MeDA (100µM, ') or 2,5-bis(GSyl)-α-MeDA (100µM, #) treated cells at increasing time points (1, 2, 4, 8, 16, and 24 hrs) by liquid scintillation spectroscopy. Data are expressed as the percent inhibition of 5-HT uptake compared to control values; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05.

122

Figure 5.6: Kinetics of MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA-mediated inhibition of 5-HT uptake in JAR cells. A. Specific [3H]5-HT uptake was determined by saturation analysis, by treating SK-N-MC cells with increasing concentrations of [3H]5-HT either alone or in the presence of fluoxetine (10 µM) for 30 min, measuring the intracellular accumulation of [3H]5-HT, and taking the difference in uptake between non-specific and total 5-HT

uptake. Km and Vmax are mean values of 4 independent transfections. 3 B. Ki values were determined by incubating cells with 20 nM [ H]5- HT and increasing concentrations of MDA ($, 100µM), MDMA (%, 100µM), 5-(GSyl)-α-MeDA (#, 100µM), or 2,5-bis(GSyl)-α-MeDA (", 100µM) prior to measuring intracellular [3H]5-HT accumulation. The data were analyzed by non-linear least squares curve fit and data are plotted as the % specific 5-HT uptake. 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA were more potent at inhibiting [3H]5-HT µ uptake (Ki = 66 and 47 M, respectively) than either MDA or MDMA µ (Ki = 105 and 97 M, respectively). 95% confidence intervals for Ki values are as follows; MDA = 92.36 to 119.6 µM; MDMA = 82.98 to 112.76 µM; 5-(GSyl)-α-MeDA = 51.52 to 84.12 µM; 2,5-bis(GSyl)- α-MeDA = 36.22 to 61.39 µM.

123

80 70 A. 60 Total 50 40 Specific [3H] 5-HT uptake 30 20 v (pmol/min/mg protein) (pmol/min/mg v 10 Non-Specific 0 0.00 0.25 0.50 0.75 1.00 1.25 1.50 1.75 2.00 2.25 [(3H)5-HT] (µM)

100

B. 80

60

40 H] 5-HT Cellular uptake Cellular 5-HT H] 3

20 Specific [ Specific

-2 -1 0 1 2 3 4 5 Log [inhibitor] (µM)

124 inhibition of 5-HT uptake in primary cells derived from the striatum and hippocampus. Following 4 hr incubations, the parent drugs inhibited 5-HT uptake in primary hippocampal (Figure 5.7) and striatal (Figure 5.8) cells by ~23% and ~12%, respectively. Consistent with work in the SERT-expressing cell lines, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA were more potent that either MDA or MDMA, decreasing 5-HT uptake in hippocampal cells by 38% and 44%, respectively, and 25% and 28% in striatal cells, suggesting that hippocampal cells are more sensitive to MDA and MDMA-induced neurotoxicity. To determine the role of the SERT and DAT on changes in [3H]5-HT, groups of cells were treated with SERT and DAT transport blockers, fluoxetine and nomifensine, and 5-HT uptake was evaluated following 4 hrs of treatment. Nomifensine had no significant effect on MDA or MDMA inhibition of 5-HT uptake, nor on the effects of the metabolites. Fluoxetine pretreatment had no significant effect on the inhibition of [3H]5-HT uptake in both HP and ST cells (Figures 5.7 and 5.8).

B. MDA, MDMA, 5-(GSYL)-α-MEDA, AND 2,5-BIS(GSYL)-α-MEDA STIMULATE DA TRANSPORT INTO MULTIPLE SERT-EXPRESSING CELL MODELS

1. hSERT-TRANSFECTED SK-N-MC CELLS DA may contribute to the serotonergic neurotoxicity of MDMA via its ability to generate ROS and reactive quinones (Sprague et al., 1998; Shankaran et al., 1999). MDA, MDMA, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA stimulated the cellular uptake of [3H]DA into SERT-transfected cells (Figure 5.9). Again, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA were more potent than either MDA or MDMA, increasing DA uptake ~3 fold (~22% and ~28%, respectively) over control levels. The maximum stimulation of DA uptake occurred between 4-8 hrs (Figure 5.9). Fluoxetine inhibited DA uptake, indicating that DA is transported via the SERT (Figure 5.10). To our knowledge, these studies are the first to demonstrate MDA, MDMA, and α-MeDA thioether stimulation of SERT-dependent DA uptake. Subsequent MAO-mediated DA oxidation

125

70 *† 60

50 *† *

40 *† * 30 ** 20

10 % Inhibition of Cellular 5-HT Cellular Uptake of Inhibition %

0 Fluoxetine Nomifensine

Figure 5.7: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)- α-MeDA inhibit 5-HT uptake into hippocampal cells. Cellular [3H]5-HT uptake was measured 4 hrs following treatment of hippocampal cells. Control (black bars), MDA (100 µM; hatched bars), MDMA (100 µM; dotted bars), 5-(GSyl)-α-MeDA (100 µM; white bars) or 2,5-bis(GSyl)-α-MeDA (100 µM; gray bars) data are presented as the % inhibition of 5-HT uptake and is expressed as the mean (N = 4) ± SEM. Groups of cells were incubated with fluoxetine (100µM) or nomifensine (50µM) 20 min prior to treatment. Data are expressed as the percent inhibition of 5-HT uptake compared to control values; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05.

126

45 *† 40

35 *† *† 30 * 25

20 * 15 *

10 % Inhibition of Cellular 5-HT Cellular Uptake of % Inhibition 5

0 Fluoxetine Nomifensine

Figure 5.8: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)- α-MeDA inhibit 5-HT uptake into striatal cells. Cellular [3H]5-HT uptake was measured 4 hrs following treatment of striatal cells. Control (black bars), MDA (100 µM; hatched bars), MDMA (100 µM; dotted bars), 5-(GSyl)-α-MeDA (100 µM; white bars) or 2,5- bis(GSyl)-α-MeDA (100 µM; gray bars) data are presented as the % inhibition of 5-HT uptake and is expressed as the mean (N = 4) ± SEM. Groups of cells were incubated with fluoxetine (100µM) or nomifensine (50µM) 20 min prior to treatment. Data are expressed as the percent inhibition of 5-HT uptake compared to control values; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05.

127

30 * † * †

25 * † * * 20

* 15 * * * * * * * * * 10 * * * 5 % Increase of Cellular DA Uptake Cellular DA of Increase %

0 0 4 8 12 16 20 24 48 Time (hr)

Figure 5.9: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA induce DA uptake into SERT- expressing SK-N-MC cells. Cellular DA uptake was determined in control ($), MDA (100µM, "), MDMA (100µM, #), 5-(GSyl)-α-MeDA (100µM, ") or 2,5- bis(GSyl)-α-MeDA (100µM, #) samples at increasing time points (0.5, 1, 2, 4, 8, 16, 24, and 48 hrs) and using liquid scintillation spectroscopy. Data are expressed as the percent increase in DA uptake compared to controls; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) treatment groups and MDA (†) are considered significant at p < .05.

128 and metabolism may then contribute to the serotonergic neurotoxicity of MDA and MDMA.

2. SEROTONERGIC JAR CELLS Cellular uptake of [3H]DA into JAR cells was not significantly affected by either MDA or MDMA (Figure 5.11). 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA, on the other hand, stimulated [3H]DA uptake into JAR cells by ~16%, which was maximal between 4 and 8 hrs (Figure 5.12). Studies with nomifensine and fluoxetine revealed the potential involvement of the SERT in 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA- mediated cellular DA uptake. Our finding that nomifensine had no effect on the uptake of [3H]DA into JAR cells is consistent with the fact that JAR cells do not express detectable levels of the DAT (Kubota et al., 2001). In contrast, pretreatment with fluoxetine attenuated the increase in DA uptake, supporting a role for the SERT in 5-(GSyl)-α- MeDA and 2,5-bis(GSyl)-α-MeDA-induced cellular DA uptake.

3. RAT PRIMARY STRIATAL AND HIPPOCAMPAL CELLS MDA and MDMA produced a modest, although statistically insignificant, inhibition of [3H]DA uptake into striatal cells (Figure 5.13). 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)- α-MeDA were more potent then either of the parent amphetamines, inhibiting DA uptake into striatal cells by ~12% and ~19%, respectively (Figure 5.13). Pretreatment with nomifensine attenuated the inhibition of DA uptake, whereas fluoxetine had no effect, suggesting a potential interaction between these compounds and the DAT. Indeed, 5- (GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA induce ROS generation in DAT-transfected SK-N-MC cells (Chapter 6). Conversely, neither of the compounds had any effect on DA or 5-HT cellular uptake in DAT-transfected cells (data not shown). Interestingly, in hippocampal cells, both parent drugs and the thioether metabolites stimulated the cellular uptake of DA (Figure 5.14). In contrast to striatal cells, fluoxetine attenuated the effect that these compounds had on [3H]DA uptake into hippocampal cells, whereas

129 nomifensine had no effect, suggesting that DA is entering these cells via the SERT. Reasons underlying the difference in DA uptake between the striatal and hippocampal cells is unknown, however, the ratio of 5-HT verses DA neurons in the hippocampus compared to the striatum (Chapter 7) may influence the results. Perhaps the compounds are indeed selectively targeting 5-HT neurons.

130

35

*† 30 * * †

25 *

20 φ

15 φ

10 % Increase of Cellular DA Uptake of Increase %

5

0 Fluoxetine Nomifensine

Figure 5.10: Fluoxetine, but not nomifensine attenuates MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA stimulation of DA uptake into SERT-expressing SK-N-MC cells. Cellular DA uptake was determined in hSERT-transfected cells by liquid scintillation spectroscopy 4 hrs following treatment. Control (black bars), MDA (100 µM; dashed bars), MDMA (100 µM; gray bars), 5-(GSyl)-α-MeDA (100 µM; white bars) or 2,5-bis(GSyl)-α- MeDA (100 µM; checkered bars) data are presented as the % increase in DA uptake compared to controls and are expressed as the mean (N = 4) ± SEM. Groups of cells were incubated with fluoxetine (100µM) or nomifensine (50µM) 20 min prior to treatment. Differences between; i) control and treatment groups (*), ii) MDA and other drug groups (†), and iii) treatment only and fluoxetine groups (φ) are considered significant at p < 0.05.

131

*† 20 *† 18

16

14

12

10 φ 8 φ

6

% Increase of Cellular DA Uptake of Increase % 4

2

0 Fluoxetine Nomifensine

Figure 5.11: Fluoxetine, but not nomifensine attenuates 5- (GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA stimulation of DA uptake into JAR cells. Cellular DA uptake was determined in hSERT-transfected cells by liquid scintillation spectroscopy 4 hrs following treatment. Control (black bars), MDA (100 µM; dashed bars), MDMA (100 µM; gray bars), 5-(GSyl)-α-MeDA (100 µM; white bars) or 2,5-bis(GSyl)-α-MeDA (100 µM; checkered bars) data are presented as the % increase of DA uptake compared to controls and are expressed as the mean (N = 4) ± SEM. Groups of cells were incubated with fluoxetine (100µM) or nomifensine (50µM) 20 min prior to treatment. Differences between; i) control and treatment groups (*), ii) MDA and other drug groups (†), and iii) treatment only and fluoxetine groups (φ) are considered significant at p < 0.05.

132

25

20 *† *† *† *† *† 15 *† *† * *† * *† 10 * *

5 * % of Cellular Stimulation DA Uptake

0 1 2 4 8 16 24 Hours

Figure 5.12: Time course for MDA, MDMA, 5-(GSyl)-α- MeDA and 2,5-bis(GSyl)-α-MeDA stimulated DA uptake into JAR cells. Cellular [3H]DA uptake was determined in control (!), MDA (100µM, "), MDMA (100µM, #), 5- (GSyl)-α-MeDA (100µM, ") or 2,5-bis(GSyl)-α-MeDA (100µM, #) treated cells at increasing time points (1, 2, 4, 8, 16, and 24 hrs) by liquid scintillation spectroscopy. Data are expressed as the percent DA uptake compared to controls; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*), ii) MDA and other drug groups (†), and iii) treatment only and fluoxetine groups (φ) are considered significant at p < 0.05.

133

30

25 *

20 *

15 φ 10 φ

5 % Inhibition of Cellular DA Uptake of Cellular % Inhibition

0 Fluoxetine Nomifensine

Figure 5.13: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)- α-MeDA inhibit DA uptake into striatal cells. Cellular [3H]DA uptake was measured 4 hrs following treatment of striatal cells. Control (black bars), MDA (100 µM; hatched bars), MDMA (100 µM; dotted bars), 5-(GSyl)-α-MeDA (100 µM; white bars) or 2,5-bis(GSyl)-α- MeDA (100 µM; gray bars) data are presented as the % inhibition of DA uptake compared to controls and are expressed as the mean (N = 4) ± SEM. Groups of cells were incubated with fluoxetine (100µM) or nomifensine (50µM) 20 min prior to treatment. Differences between; i) control and treatment groups (*), and treatment only and nomifensine groups (φ ) are considered significant at p < 0.05.

134

25

20

15 *

10

5 % of Cellular Stimulation DA Uptake

0 Fluoxetine Nomifensine

Figure 5.14: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)- α-MeDA stimulate DA uptake into hippocampal cells. Cellular [3HDA uptake was measured 4 hrs following treatment of hippocampal cells. Control (black bars), MDA (100 µM; hatched bars), MDMA (100 µM; dotted bars), 5-(GSyl)-α-MeDA (100 µM; white bars) or 2,5- bis(GSyl)-α-MeDA (100 µM; gray bars) data are presented as the % inhibition of DA uptake compared to controls and are expressed as the mean (N = 4) ± SEM. Groups of cells were incubated with fluoxetine (100µM) or nomifensine (50µM) 20 min prior to treatment. Differences between; i) control and treatment groups (*) are considered significant at p < 0.05.

135 III. DISCUSSION Studies described in this chapter are the first to demonstrate that MDA, MDMA, 5- (GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA (100 µM) stimulate DA uptake into multiple SERT-expressing, or “serotonergic”, cell types, including JAR cells, hSERT- transfected SK-N-MC cells, and rat primary hippocampal cells. Interestingly, whereas the parent drugs demonstrated a very modest effect, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α- MeDA were significantly more potent than either of the parent amphetamines at stimulating SERT-mediated DA transport, suggesting that systemic conjugation to GSH likely contributes to the neurotoxicity of MDA and MDMA. Fluoxetine, a selective 5-HT reuptake inhibitor, blocked this effect, indicating that the uptake of DA into “serotonergic” cells was dependent on a functional SERT and hence, DA likely entered the cell via this transporter. In support of this interpretation, fluoxetine also inhibits the uptake of DA into hippocampal synaptosomes (Sprague and Nichols, 1995) suggesting that DA is entering the cell via the SERT. The requirement for DA (Sprague and Nichols, 1995 a, b; Aguirre et al., 1998; Sprague et al., 1998; Shankaran et al., 1999) in MDMA- mediated serotonergic neurotoxicity suggests that DA may contribute to MDMA-induced oxidative damage. Indeed, functional SERT proteins are capable of transporting DA into 5-HT cells (Faraj et al., 1994; Schmidt and Lovenberg, 1985). Moreover, increasing body temperature, which is a consequence of MDMA use, disrupts the function of the SERT, increasing its affinity for DA while inhibiting the transport of 5-HT (Saldana and Barker, 2004). In contrast to the present findings, however, MDMA (10 µM) failed to increase the uptake of DA into transfected hSERT-HEK cells following one hr of exposure (Saldana and Barker, 2004). In fact, MDMA appears to partially inhibit the affinity of the SERT for DA (Saldana and Barker, 2004). In addition, MDMA inhibits the synaptosomal uptake of DA (Bogen et al., 2003), although a precise role for the SERT or an identification of affected cells in synaptosomal preparations cannot be determined. The reasons behind the discrepancy in findings between the present results in hSERT- transfected cells and the aforementioned study (Saldana and Barker, 2004) are unknown.

136 However, several differences in experimental model and design, including type of cell, concentration of MDMA used, length of incubation, and stable verses transient protein expression, may differentially affect the SERT-mediated uptake of DA. In addition, perhaps the metabolism of the parent amphetamines to the α-MeDA thioethers differentially affects SERT function, increasing its affinity for DA. Indeed, the electrophilic and oxidative properties of α-MeDA-thioethers are greater than those of the parent amphetamines. Consistent with results in transfected SK-N-MC cells, MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA (100 µM) also induced DA uptake into JAR cells, which constitutively express the SERT and exhibit 5-HT transport. An essential relationship between MDMA-induced serotonergic neurotoxicity and the presence of DA has been demonstrated (Aguirre et al., 1998; Bankson and Cunningham, 2001 [review]; Falk et al., 2002; Fernandez et al., 2003). For instance, MDMA stimulates the release of DA in a 5-HT dependent manner (Nash and Brodkin, 1991; Gudelsky et al., 1994; Guldelsky and Nash, 1996). Interestingly, MDMA also inhibits DAT function (Metzger et al., 1998; Hansen et al., 2002). 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA inhibited DA uptake into striatal cell populations (Figure 5.13), thereby increasing the extracellular pool of DA and implying a role for DA in MDA and MDMA-induced serotonergic neurotoxicity. In contrast, depletion of intracellular DA stores with reserpine did not protect against MDMA-induced serotonergic neurotoxicity (Yaun et al., 2002), which argues against any contribution from DA. However, reserpine also inhibits the MDMA stimulated release of 5-HT (Sabol and Seiden, 1998) and possibly norepinephrine (Falk et al., 2002) from intracellular stores. Therefore, because reserpine has multiple effects, it is difficult to demonstrate that a specific effect on DA was related to the modulation of neurotoxicity. In fact, other studies report that reserpine does indeed protect against MDMA-induced neurotoxicity (Stone et al., 1988; Schmidt et al., 1990). Finally, the administration of DA (Simantov and Tauber, 1997) or the DA precursor, L-dihydroxyphenylalanine (L-dopa) (Schmidt et al., 1991; Simantov and Tauber, 1997; Aguire et al., 1998) potentiates the serotonergic neurotoxicity of MDMA.

137 2,5-bis(GSyl)-α-MeDA not only causes a long-term depletion of 5-HT, it also produces an acute, modest increase in extracellular DA (Miller et al., 1997). Consequently, these actions result in a significant increase in the concentration of extracellular DA available for uptake into serotonergic neurons. Therefore, although it has not been previously demonstrated, several reports have suggested that MDMA stimulates the uptake of DA into 5-HT cells (Aguirre et al.1998; Sprague et al., 1998). Consistent with this view, we hypothesize that the α-MeDA- thioethers stimulate the SERT-dependent cellular uptake of DA into serotonergic nerve terminals, where it undergoes oxidation by MAO-B to generate ROS and reactive quinone species. Indeed, inhibition of MAO-B protects against MDMA-induced serotonergic neurotoxicity (Sprague and Nichols, 1995 a, b; Falk et al., 2002), supporting a contribution from DA. On the other hand, if MDMA or the α-MeDA thioethers are permitted access to intracellular MAO, either via transport by the SERT or passive diffusion, they may interact directly with MAO-B, undergoing oxidation and generating ROS. However, MDMA does not appear to inhibit MAO-B activity (Leonardi and Azmitia, 1994). Moreover, MAO-B deficient mice, in which MDMA-induced serotonergic neurotoxicity is attenuated, actually show an increase in dopaminergic neurotoxicity (Fornai et al., 2001). In addition, inhibition of DAT function in mice decreases dopaminergic neurotoxicity but not MDMA-mediated DA release (Camarero et al., 2002), which suggests that MDMA has access to the nerve terminal DA stores by a DAT-independent mechanism. Furthermore, the finding that MAO-B deficient animals have increased dopaminergic neurotoxicity appears to rule out any potential role for MAO-B in metabolizing MDMA. In any event, it is our opinion that there is insufficient processing of MDMA by MAO-A or -B to produce toxic levels of ROS and reactive metabolites, and that the attenuating effects of MAO-B inhibitors on MDMA-induced neurotoxicity is due to inhibition of DA oxidation within 5-HT nerve terminals. Further studies examining the contribution of DA oxidation to 5-(GSyl)-α-MeDA and 2,5-

138 bis(GSyl)-α-MeDA induced neurotoxicity in these cells should focus on the potential role of MAO-B. Although there is a substantial amount of evidence supporting a requirement for DA in MDMA-induced neurotoxicity, there is an argument against a direct relationship between DA and MDMA-mediated serotonergic neurotoxicity, suggesting that the role of DA is limited to its effect on raising ambient body temperature (Malberg et al., 1996, Malberg and Seiden, 1998; Colado et al., 1997, 1998). For example, induction of hypothermia protects against MDMA-mediated neurotoxicity, whereas raising ambient body temperature, potentiates the neurotoxicity of MDMA (Malberg at el., 1996). However, decreases in body temperature can slow the chemical reactions required for a variety of neurotoxicants and vice verse, increases in temperature can accelerate biochemical reactions (Bowyer et al., 1993) complicating interpretations of the effects of body temperature. Furthermore, fluoxetine provides protection against MDMA-induced serotonergic neurotoxicity, yet did not inhibit MDMA-induced hyperthermia, and inhibiting MAO-B with antisense nucleotides attenuates the neurotoxicity of MDMA with no significant effect on ambient body temperature, indicating a temperature- independent mechanism (Falk et al., 2002). Mazindol protects against MDMA-induced neurotoxicity without altering MDMA-induced hyperthermia, suggesting that the neuroprotection provided by mazindol is related to its ability to increase extracellular DA and subsequent ROS generation (Shankaran et al., 1999). Finally, AT-125 attenuated MDMA-induced increases in ambient body temperature, (Chapter 3) suggesting a hypothermic-independent mechanism. Understandably, the effects of ambient body temperature must be considered when interpreting data on MDMA-mediated neurotoxicity, as many of the pharmacological manipulations that provide protection against MDMA-induced neurotoxicity also attenuate MDMA-induced hyperthermia. It is also an important consideration for the recreational MDMA user because use of the drug often occurs in settings that tend to increase body temperature, such as dance parties, or “raves”. Ultimately, we share the opinion of Falk et al. (2002) and Schmidt et al. (1990),

139 who suggest that increased body temperature can potentiate neurotoxic events, but that the underlying basis of MDMA-induced neurotoxicity is unrelated to the production of hyperthermia. In contrast to its potential role as the transporter in MDA, MDMA, 5-(GSyl)-α- MeDA and 2,5-bis(GSyl)-α-MeDA-stimulated cellular uptake of DA into 5-HT cells, the SERT’s conventional function, the re-uptake of 5-HT, is inhibited by both the parent amphetamines and α-MeDA thioethers. Consistent with their potential role in the neurotoxicity of MDA and MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA were significantly more potent inhibitors of 5-HT reuptake than the parent amphetamines. MDMA interacts with the SERT to, i) inhibit the reuptake of 5-HT into the cell and, ii) induce ROS generation (Liechti et al., 2000; Shankaran et al., 1999; Kramer et al., 1997). For instance, 5-HT reuptake inhibitors, including fluoxetine and citolapram, attenuate both MDMA-induced ROS generation and serotonergic neurotoxicity. Moreover, SERT- deficient mice are insensitive to the neurotoxic effects of MDMA (Bengal et al., 1998). Although the involvement of the SERT in MDMA neurotoxicity is firmly established, the exact nature of the interaction is unknown. To our knowledge, no one has demonstrated the formation of MDMA-adducted SERT proteins. Due to the high electrophilic properties of the thioether metabolites, we suspect that 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA may interact with and covalently modify the SERT with higher affinity than the parent drugs. Alternatively, DA quinones, which are formed during the metabolism of DA, and are also highly electrophilic, may also bind to and compromise SERT function. The human SERT contains conformationally sensitive cysteine residues (Cys 200 and 209) on the extracellular loop between transmembrane domains 3 and 4 (Amara et al., 1995; Chen et al., 1997a, b, 2000), as well as on the cytoplasmic side (Androutsellis-Theotokis et al., 2001). These residues are involved in protein folding and stability, but may also be involved in substrate and inhibitor binding and consequently, may serve as prime targets for quinone-protein adduct formation. In summary, the remarkable ability of the thioether metabolites to form highly reactive electrophiles,

140 coupled with the potential vulnerability of the SERT, make the interaction of the thioethers and the transporter extremely likely. Kinetic analysis of MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA induced inhibition of 5-HT uptake in hSERT-transfected SK-N-MC cells and JAR cells revealed that the α-MeDA thioethers are more potent inhibitors of 5-HT cellular uptake than either of the parent amphetamines. Although it is difficult to compare the kinetics of drugs in different experimental models, a comparison to other known SERT inhibitors suggests that both the parent drugs and the α-MeDA thioethers are significantly less potent at inhibiting 5-HT transport (Barker et al.1998). Thus, in hSERT-transfected HeLa cells, fluoxetine and citolapram have Ki values in the nmol range, whereas MDA,

MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA demonstrate Ki values in the low µmol range. Again, due to the different experimental models, it is difficult to extrapolate the results between cell types; nevertheless, it is safe to assume that the parent amphetamines and the α-MeDA thioethers are comparatively weak inhibitors of SERT function. On the other hand, the kinetic values for MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA (100µM) reported here are consistent with previous analyses (Mortensen et al., 1999). Therefore, although MDA, MDMA, and α-MeDA thioethers are weaker SERT inhibitors then traditional SSRI’s, in contrast to typical 5-HT re-uptake inhibitors, it appears that their interaction with the SERT induces both an uptake of DA, and a neurotoxic response in 5-HT cells. MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA (100µM) inhibits 5- HT transport in striatal and hippocampal cells, inhibition being greater in hippocampal cells. The difference in the response between these brain regions is not surprising, given that hippocampal cell populations are predominantly “serotonergic” whereas more “dopaminergic” cells are observed in striatal cell cultures (Chapter 7) and MDA and MDMA are serotonergic neurotoxicants. Furthermore, differences in axonal characterizes (“M” vs “D” systems; see Chapter 4 discussion) may affect the ability of these compounds to interact with the SERT and inhibit 5-HT transport. Unfortunately, it is

141 difficult to identify exactly which type of cells in the hippocampal and striatal cell populations are affected by MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α- MeDA. Interestingly, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA demonstrated contrasting effects on DA transport into hippocampal and striatal cells. The α-MeDA- thioethers modestly inhibited DA uptake into striatal cells but similar to their effects in JAR and hSERT-transfected cells, significantly stimulated DA uptake into hippocampal cells (Figures 5.13 and 5.14). Again, it is difficult identify the specific type of cell affected by these compounds. However, the stimulation of DA uptake in hippocampal cells is attenuated by fluoxetine, suggesting DA uptake was SERT-dependent, and hence, the cells are serotonergic. It is possible that the stimulation of DA uptake also occurs in SERT-expressing cells in the striatum, but perhaps, these compounds concomitantly inhibit DA transport into DAT-expressing cells masking the increase in DA uptake via the SERT. Indeed, nomifensine attenuated the inhibition of DA transport, suggesting a DAT-dependent component. Moreover, MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA modestly inhibited DA uptake into hDAT-transfected SK-N-MC cells and, consistent with these findings, other laboratories have suggested a potential interaction between MDMA and the DAT (Metzger et al., 1998; Shankaran et al., 1999). Although this action does not appear to produce a neurotoxic response, it may contribute to the serotonergic neurotoxicity of MDMA by increasing the concentration of extracellular DA. In summary, MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA significantly inhibited the transport of 5-HT into SERT-expressing, or “serotonergic” cells with the α-MeDA-thioethers being more potent than the parent amphetamines consistent with a role for metabolism in MDMA-induced neurotoxicity. Interestingly, the inhibition of 5-HT uptake was coupled to a stimulation of SERT-dependent DA transport into “serotonergic” cells, supporting a role for DA in the serotonergic neurotoxicity of these amphetamine analogs and their thioether metabolites. The precise involvement of DA in developing a neurotoxic response is still a matter of controversy. Whether the

142 contribution of DA is indirect, via increased ambient body temperature, or more likely, direct, via its uptake into 5-HT nerve terminals and the subsequent oxidation to ROS and electrophilic quinones remains unclear. Nevertheless, it is apparent that i) DA is required for the neurotoxicity and, ii) the interaction of MDMA, or rather, the α-MeDA- thioethers, and the SERT results in a disruption of SERT function, decreasing its affinity for 5-HT but increasing the cellular uptake of DA.

143 CHAPTER 6

MDA, MDMA, 5-(GSYL)-α-MEDA, AND 2,5-BIS(GSyl)-α-MeDA INDUCE ROS GENERATION IN PRIMARY, JAR AND hSERT-TRANSFECTED SK-N-MC CELLS

I. INTRODUCTION AND RATIONALE The biological and oxidative reactivity of the thioether conjugates of various polyphenolic compounds suggests that neurotoxic mechanisms may include the generation of ROS and quinone species (Monks and Jones, 2002; Monks and Lau, 1997). 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA, polyphenolic GSH conjugates of α- MeDA and the parent amphetamines, MDA and MDMA, share numerous pharmacological and toxicological properties, including long-term decreases in brain 5- HT levels, morphological damage to serotonergic nerve terminals, and activation of neuro-protective systems (i.e. astrocytes, microglia) (Bai, 2000), inhibition of 5-HT uptake, and stimulation of SERT-mediated DA cellular uptake. The precise mechanism underlying the neurotoxic response has yet to be established. However, polyphenolic GSH-conjugates remain capable of redox cycling between their quinone and semiquinone species, consequently generating hydrogen peroxide and hydroxyl radicals, which readily oxidize cellular macromolecules, including proteins, lipids and DNA. In particular, the SERT, a “serotonergic” protein, is a prime target for ROS-induced oxidation (Haughey et al., 1999). Therefore, the present chapter describes studies designed to examine MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA induced ROS generation in multiple SERT-expressing cell types, hSERT transfected SK-N-MC, JAR and primary hippocampal and striatal cells. Cellular oxidative stress and damage appear to contribute significantly to the serotonergic neurotoxicity of MDA and MDMA. In fact, MDMA-induced oxidative stress and free radical damage has been observed in human erythrocytes (Zhou et al., 2003a, b). MDMA increases the ROS generation (Shankaran et al., 1999, 2001; Colado et

144 al., 1998; Guldelsky and Yamamoto, 1994). Microdialysis studies using salicylic acid revealed that administration of MDMA produces significant levels of ROS in the brain (Guldelsky and Yamamoto, 1994; Shankaran et al., 1999; 2001). Furthermore, free radical scavengers and antioxidants attenuate MDMA-induced toxicity in-vitro (Colado and Green, 1995) and in-vivo (Gudelsky, 1996) and over-expression of SOD protects against the effects of MDMA by decreasing the generation of ROS (Jayanthi et al., 1999; Cadet et al., 1995). MDMA-induced morphological and biochemical changes in cellular integrity indicative of oxidative stress, including lipid peroxidation and protein nitration, have been demonstrated (Sprague and Nichols, 1995a, b; Colado et al., 1997a; Stone et al., 1989b). For example, MDMA inhibits cytochrome oxidase, a key enzyme in the mitochondrial electron transport chain, presumably via the generation of excess ROS (Burrows et al., 2000). In addition, MDMA administration leads to the oxidation of sulfhydryl moieties of TPH resulting in the inhibition of TPH activity and 5-HT synthesis (Stone et al, 1989a, b). Interestingly, the parent amphetamines, lacking a catechol moiety, are unable to redox cycle. Consequently, the mechanism by which MDA and MDMA generate ROS is most likely indirect, via the formation and biochemical activity of redox active metabolites or perhaps, the oxidation and metabolism of DA. One potential reactive metabolite, α- MeDA, however, fails to reproduce the serotonergic neurotoxicity of the parent drugs (Miller et al., 1997). On the other hand, the half-wave oxidation potentials of 5-(GSyl)-α- MeDA and 2,5-bis(GSyl)-α-MeDA are lower than α-MeDA and hence, they are more biologically reactive (Miller et al., 1996). Finally, it appears that DA oxidation may contribute to MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA induced ROS generation and neurotoxicity (see Chapter 5 for discussion). We have demonstrated that these compounds stimulate the transport of DA into SERT-expressing cells (Chapter 5), suggesting ROS may be generated as a by-product of MAO-B-mediated oxidation of DA within the 5-HT nerve terminal. In all probability, both redox cycling of the polyphenolic α-MeDA thioethers and the oxidation of DA contribute to the increase in

145 cellular ROS and subsequent oxidative damage. To test the hypothesis that 5-(GSyl)-α- MeDA and 2,5-bis(GSyl)-α-MeDA are more potent generators of ROS than the either MDA or MDMA, hDAT and hSERT transfected SK-N-MC cells, JAR cells, and primary rat hippocampal and striatal cells were loaded with DCF-DA, treated with the compounds and changes in DCF-DA fluorescence were monitored.

146 II. RESULTS A. hSERT AND hDAT- TRANSFECTED SK-N-MC CELLS MDA, MDMA, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA (100 µM) induce rapid ROS generation in hSERT- and hDAT-transfected SK-N-MC cells as indicated by changes in DCF-DA fluorescence (Figure 6.1 and 6.2). The rate of ROS generation declines rapidly after the initial burst, decreases rapidly but persists at low levels for up to 8 hrs. The rapid decrease is presumably due to either metabolism of the drugs and/or exhaustion of reducing equivalents required to support redox cycling. 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA were more efficient inducers of ROS generation, in comparison to MDA and MDMA, in both hSERT and hDAT transfected cells (Figure 6.1 and 6.2). MDA and MDMA induced a concentration-dependent increase in ROS generation in hSERT (Figure 6.3), and hDAT (Figure 6.4), transfected SK-N-MC cells, although 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA displayed a greater ability to generate ROS. Although the absolute concentration of ROS induced by each compound was slightly greater in hSERT transfected cells, the kinetics were similar for, i) both the parent drugs and their metabolites, and ii) both cell types. None of the compounds tested had any effect on ROS generation in mock-transfected cells (data not shown), indicating a requirement for transporters in ROS generation. However, nomifensine had no effect on ROS generation in hDAT-transfected cells (Figure 6.6), indicating that ROS generation, although hDAT-dependent, is insensitive to nomifensine. In contrast, fluoxetine significantly inhibited ROS generation in hSERT-transfected cells (Figure 6.5), indicating that MDA, MDMA, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA induced ROS generation in hSERT-expressing cells requires a functional SERT.

B. SEROTONERGIC JAR CELLS MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA (100 µM) produced significant increases in ROS generation in JAR cells loaded with DCF-DA in a concentration-dependent manner (Figure 6.7). Consistent with time course data obtained

147

70

60 *†

50 *†

40

30 * *

20

10 % Increase in ROS generation (DCF fluorescence) 0 0 0.5 1424 8 16 Hours

Figure 6.1: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA induce ROS generation in hSERT- transfected cells. ROS generation was determined in hSERT transfected SK-N-MC cells by loading the cells with DCF-DA and incubating Control (!), MDA (100 µM; "), MDMA (100 µM; !), 5-(GSyl)-α-MeDA (100 µM; ") or 2,5-bis(GSyl)-α-MeDA (100 µM; #) ) samples for increasing time periods (0.5, 1, 4, 8, 16, and, 24 hrs). Data are expressed the percent increase in ROS generation compared to controls; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05.

148

50

45 *†

40 *† 35

30

25 * 20 *

15

10

% Increase in ROS generation (DCF fluorescence) (DCF generation in ROS Increase% 5

0 0 0.5 1424 8 16 Hours

Figure 6.2: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA induce ROS generation in hDAT- transfected cells. ROS generation was determined in hSERT transfected SK-N-MC cells by loading the cells with DCF-DA and incubating Control (!), MDA (100 µM; "), MDMA (100 µM; !), 5-(GSyl)-α-MeDA (100 µM; ") or 2,5-bis(GSyl)-α-MeDA (100 µM; #) samples for increasing time periods (0.5, 1, 4, 8, 16, and, 24 hrs). Data are expressed the percent increase in ROS generation compared to controls; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05.

149

80 *† 70 *† *† *† 60 *† *

50 *† * * * 40 * * * 30 * * 20 *

10 %Increase in ROS fluorescence)%Increase (DCF generation in ROS 0 10______50 100 200 400 µM

Figure 6.3: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α- MeDA-induced ROS generation in hSERT-transfected SK-N-MC cells is concentration-dependent. ROS generation was determined in hSERT transfected cells using increasing concentrations (10, 50, 100, 200, 400 µM) of the compounds and measuring the change in DCF fluorescence 4 hrs following treatment. Control (!), MDA (100 µM; "), MDMA (100 µM; !), 5-(GSyl)-α-MeDA (100 µM; ") or 2,5- bis(GSyl)-α-MeDA (100 µM; #) data are presented as the % control ROS generation and is expressed as the mean (N = 4) ± SEM. Groups of cells were incubated with fluoxetine (100µM) or nomifensine (50µM) 20 min prior to treatment. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05.

150

80

70 *† 60 *† * 50 * * 40 * * * 30 * * 20 * *

10 *

% Increase in ROS generation (DCF fluorescence) 0 10______50 100 200 400 µM

Figure 6.4: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)- α-MeDA-induced ROS generation in hDAT-transfected SK-N- MC cells is concentration-dependent. ROS generation was determined in hSERT transfected cells using increasing concentrations (10, 50, 100, 200, 400 µM) of the compounds and measuring the change in DCF fluorescence 4 hrs following treatment. Control (!), MDA (100 µM; "), MDMA (100 µM; !), 5-(GSyl)-α- MeDA (100 µM; ") or 2,5-bis(GSyl)-α-MeDA (100 µM; #) data are presented as the % control ROS generation and is expressed as the mean (N = 4) ± SEM. Groups of cells were incubated with fluoxetine (100µM) or nomifensine (50µM) 20 min prior to treatment. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05.

151 70 * †

60 * † 50

40 * * 30 φ φ 20 φ φ

10 % Increase in ROS generation,ROS % fluorescence IncreaseDCF-DA in

0 Fluoxetine Nomifensine Figure 6.5: Fluoxetine attenuates MDA, MDMA, 5-(GSyl)-α- MeDA and 2,5-bis(GSyl)-α-MeDA induced ROS generation in hSERT-transfected SK-N-MC cells. ROS generation was determined in hSERT transfected S-K-N-MC cells after a 30 min incubation with vehicle (black bars), MDA (hatched bars), MDMA (gray bars), 5-(GSyl)-α-MeDA (white bars) or 2,5- bis(GSyl)-α-MeDA (checkered bars) by measuring changes in DCF-DA fluorescence. Fluoxetine (50 µM) or nomifensine (50µM) was added to the cell medium 20 min prior to drug treatment. Data are expressed the percent increase in ROS generation compared to controls; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05. Differences between fluoxetine and treatment only groups are significant at p < 0.05 (φ).

152

70

60 *† 50 *†

40

30 * * 20

10

% Increase in ROS generation, fluorescenceROS DCF-DA % Increase in 0 Fluoxetine Nomifensine

Figure 6.6: Fluoxetine attenuates MDA, MDMA, 5-(GSyl)-α- MeDA and 2,5-bis(GSyl)-α-MeDA induced ROS generation in hDAT-transfected SK-N-MC cells. ROS generation was determined in hDAT transfected S-K-N-MC cells after a 30 min incubation with vehicle (black bars), MDA (hatched bars), MDMA (gray bars), 5-(GSyl)-α-MeDA (white bars) or 2,5-bis(GSyl)-α- MeDA (checkered bars) (100 µM,) by measuring changes in DCF- DA fluorescence. Fluoxetine (50 µM) or nomifensine (50µM) was added to the cell medium 20 min prior to drug treatment. Data are expressed as the mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) treatment groups and MDA (†) are considered significant at p < 0.05. Differences between fluoxetine and treatment only groups are significant at p < 0.05 (φ).

153 obtained from hSERT-transfected cells, the rate of ROS generation increases rapidly followed by a steep decline, quickly returning to baseline levels, presumably due to either metabolism of the drugs and/or exhaustion of reducing equivalents required to support redox cycling (Figure 6.8). Following a 30 min incubation period, MDA and MDMA produced ~28-30% increases in ROS generation. Pre-exposure to the SERT inhibitor, fluoxetine (100 µM), attenuated both MDA and MDMA-induced ROS production. In contrast, nomifensine, a DAT inhibitor, had no effect on the generation of ROS (figure 6.9), confirming the participation of the SERT in MDA and MDMA-mediated ROS generation. 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA were significantly more potent at increasing ROS generation than the parent drugs (~59% and ~63% respectively), an effect attenuated with fluoxetine (Figures 6.9), suggesting that the thioether metabolite-mediated generation of ROS is also SERT dependent.

C. RAT PRIMARY STRIATAL AND HIPPOCAMPAL CELLS MDA and MDMA increased in ROS generation in both primary hippocampal and striatal cells (Figure 6.10 and 6.11, respectively). Fluoxetine attenuated the stimulation of ROS generation whereas nomifensine had no effect, which is indicative of a SERT- dependent mechanism and consistent with the effects observed in transfected SK-N-MC, and JAR cells. 5-(glutathion-S-yl)–αMeDA and 2,5-bis(glutathion-S-yl)-αMeDA were more potent ROS generators than either of the parent amphetamines, and significantly increased the generation of hippocampal ROS by ~53% and ~64% respectively (Figure 6.10). The α-MeDA-thioethers had a similar effect on striatal cells, stimulating a significant increase in ROS generation (Figure 6.11). However, the stimulatory effect of these compounds on the generation of ROS was greater in the hippocampal cell populations as compared to striatal cells.

154

100 90 *† *† *† 80 *† 70 *† 60 *† * 50 * * * * 40 * 30 * * * *

(DCF-DA Flourescence) * 20 * % Increase in ROS generation 10 0 10 uM 50 uM 100 uM 200 uM 400 uM

Figure 6.7: MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA-induced ROS generation in JAR cells is concentration-dependent. ROS generation was determined in JAR cells using increasing concentrations (10, 50, 100, 200, 400 µM) of the compounds and measuring the change in DCF fluorescence 4 hrs following treatment. Control (%), MDA (!), MDMA (&), 5-(GSyl)-α-MeDA (#) or 2,5-bis(GSyl)-α-MeDA (") data are presented as the % control ROS generation and is expressed as the mean (N = 4) ± SEM. Groups of cells were incubated with fluoxetine (100µM) or nomifensine (50µM) 20 min prior to treatment. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05.

155

80

70

*† 60 *† 50

40

† 30 * * (DCF-DA Fluorescence)(DCF-DA

% Increase in ROS generation in ROS Increase % 20 * * 10

0 0 hr 0.5 hr 1 hr 4 hr8 hr 16 hr 24 hr

Figure 6.8: Time course for MDA, MDMA, 5-(GSyl)-α- MeDA and 2,5-bis(GSyl)-α-MeDA-induced ROS generation in JAR cells. ROS generation was determined in JAR cells by loading the cells with DCF-DA and incubating Control (!), MDA (100 µM; "), MDMA (100 µM; !), 5-(GSyl)-α-MeDA (100 µM; ") or 2,5-bis(GSyl)-α-MeDA (100 µM; #) ) samples for increasing time periods (0.5, 1, 4, 8, 16, and, 24 hrs). Data are expressed the percent increase in ROS generation compared to controls; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05.

156

70 *† *† 60

50

40 * * 30 φ φ 20 φ

(DCF-DA fluorescence)(DCF-DA φ % Increase in ROS generation in ROS Increase % 10

0 ______+ fluoxetine + nomifensine

Figure 6.9: Fluoxetine attenuates MDA, MDMA, 5-(GSyl)- α-MeDA and 2,5-bis(GSyl)-α-MeDA induced ROS generation in JAR cells. ROS generation was determined in JAR cells after a 4 hr incubation with vehicle (black bars), MDA (dotted bars), MDMA (hatched bars), 5-(GSyl)-α-MeDA (white bars) or 2,5-bis(GSyl)-α-MeDA (checkered bars) by measuring changes in DCF-DA fluorescence. Fluoxetine (50 µM) or nomifensine (50µM) was added to the cell medium 20 min prior to drug treatment. Data are expressed the percent increase in ROS generation compared to controls; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05. Differences between fluoxetine and treatment only groups are significant at p < 0.05 (φ).

157

70 *†

60 *†

50

40 φ * * 30 φ

φ φ

(DCF-DA fluorescnce) 20 % Increase in ROS generation ROS in % Increase 10

0 ______+ fluoxetine + nomifensine

Figure 6.10: Fluoxetine attenuates MDA, MDMA, 5-(GSyl)-α- MeDA and 2,5-bis(GSyl)-α-MeDA induced ROS generation in hippocampal cells. ROS generation was determined in hippocampal cell cultures after a 4 hr incubation with vehicle (black bars), MDA (dotted bars), MDMA (hatched bars), 5- (GSyl)-α-MeDA (white bars) or 2,5-bis(GSyl)-α-MeDA (checkered bars) by measuring changes in DCF-DA fluorescence. Fluoxetine (50 µM) or nomifensine (50µM) was added to the cell medium 20 min prior to drug treatment. Data are expressed as the mean (N=4) ± SEM. Data are expressed the percent increase in ROS generation compared to controls; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05. Differences between fluoxetine and treatment only groups are significant at p < 0.05 (φ).

158

70

*† 60

50 *†

40

30 ** φ φ 20 φ (DCF-DA fluorescnce)(DCF-DA φ

% increase in ROS generation ROS in increase % 10

0 ______+ fluoxetine + nomifensine

Figure 6.11: Fluoxetine attenuates MDA, MDMA, 5-(GSyl)-α- MeDA and 2,5-bis(GSyl)-α-MeDA induced ROS generation in striatal cells. ROS generation was determined in striatal cell cultures after a 4 hr incubation with vehicle (black bars), MDA (dotted bars), MDMA (hatched bars), 5-(GSyl)-α-MeDA (white bars) or 2,5-bis(GSyl)-α-MeDA (checkered bars) by measuring changes in DCF-DA fluorescence. Fluoxetine (50 µM) or nomifensine (50µM) was added to the cell medium 20 min prior to drug treatment. Data are expressed the percent increase in ROS generation compared to controls; mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) MDA and other drug groups (†) are considered significant at p < 0.05. Differences between fluoxetine and treatment only groups are significant at p < 0.05 (φ).

159 III. DISCUSSION The excess production of ROS and subsequent oxidative damage is a well- established mechanism of cellular toxicity resulting in the oxidation and inactivation of multiple cellular macromolecules. Consistent with previous reports (Guldelsky and Yamamoto, 1994; Shankaran et al., 1999; 2001), MDA and MDMA each induced significant increases in ROS generation. However, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)- α-MeDA were more potent ROS generators than either of the parent drugs in each cell model examined indicating that the α-MeDA thioethers posses lower redox potentials. Interestingly, although enhanced ROS concentrations were observed in “dopaminergic”, or hDAT-transfected SK-N-MC cells, the stimulation of ROS generation was more pronounced in the “serotonergic”, or hSERT-transfected cells, suggesting that the SERT may be more susceptible to MDMA-induced ROS generation. Likewise, ROS generation was greater in primary hippocampal cells, which are predominately “serotonergic” than in striatal cells, which are predominantly “dopaminergic”, again, suggestive of a greater affinity for 5-HT cells. Moreover, the increased susceptibility of the hippocampus compared to the striatum may be a reflection of the differences in serotonergic axonal projections and nerve terminals, the “D” and “M” systems (see chapter 4 discussion). Nevertheless, α-MeDA thioethers did induce transporter-dependent ROS generation in DAT-expressing cells. Therefore, perhaps further examination of the effects of α-MeDA thioethers on the dopaminergic system may contribute to the understanding the mechanisms underlying MDMA-induced ROS generation and neurotoxicity. The source of ROS generation in MDMA-induced neurotoxicity is still debatable, however, two major mechanisms appear likely. First, the oxidation and redox cycling of metabolites of MDMA may contribute to the generation of ROS (Hiramatsu et al., 1990; Miller et al., 1996; Colado et al., 1997; Bai et al., 1999). However, the direct injection of several major metabolites of MDA and MDMA, including α-MeDA, into the brain does not reproduce the serotonergic neurotoxicity of MDA (Miller et al., 1995). In contrast, the α-MeDA thioethers, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA retain their

160 redox cycling capabilities, readily generate ROS, and, when injected directly into the brain, reproduce the “serotonergic syndrome” characteristic of MDMA. Moreover, half- wave oxidation potentials (E1/2) of the thioether metabolites of α-MeDA are lower than that of α-MeDA (Miller et al., 1996), indicative of easier oxidation and increased reactivity. ROS are generated during the cyclic conversion between the quinone and semiquinone species. In addition to their inherent reactivity as oxidants, quinones may also act as electrophiles, covalently modifying several cellular nucleophiles, including TPH (Kuhn et al., 1998), the DAT (Metzger et al, 1998), and the SERT (reference). Therefore 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA-induced oxidative damage may result from their ability to act as both oxidants and electrophiles. Secondly, the oxidation of DA within the 5-HT nerve terminal generates ROS and may contribute to the serotonergic neurotoxicity of MDA, MDMA, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA (Graham et al., 1978; Schmidt and Lovenberg, 1985; Faraj et al., 1994; Gudelsky et al., 1994; Sprague and Nichols, 1998; Aguirre et al., 1998). DA is auto-oxidized (Graham et al., 1978; Naio and Maruyama, 1999; Zhang et al., 2000) or enzymatically oxidized (Stokes et al., 1999) to ortho-quinones, which, are capable of i) redox cycling resulting in the generation of ROS and the peroxidation of lipids (Sprague and Nichols, 1995) and, ii) covalently modifying several cellular nucleophiles, including mitochondrial proteins (Montine et al., 1997), tryptophan and tyrosine hydroxylase (Kuhn and Arthur, 1998, 1999), and the DAT (Berman et al., 1996; Metzger et al., 1998). Alternatively, the monoamine oxidase-B (MAO-B)-mediated enzymatic catabolism of DA to 3,4-dihydroxyphenylacetic acid (DOPAC) leads to the formation of hydrogen peroxide, the precursor for hydroxyl radicals (Halliwell, 1992; Hastings and Zigmond, 1997; Stokes et al., 1999). Our finding that 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α- MeDA stimulated an increase of DA transport into SERT-expressing cells is consistent with the participation of DA oxidation in the neurotoxicity of MDA and MDMA. Interestingly, the major source of ROS generation during MDA and MDMA neurotoxicity may change depending on the amount of time since the original insult. For

161 example, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA stimulated DA uptake into SERT-expressing cells is maximal at 4-8 hrs (Figure 5.) whereas maximal ROS generation occurs within the first 30 min but persists at low levels for up to 8 hrs (Figure 6.). Thus, the redox cycling of 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA may contribute to the initial burst in ROS generation. Subsequently, the electrophilic activity of the metabolites may permit interactions with the SERT, thereby stimulating the cellular uptake and oxidation of DA, resulting in the continued generation of ROS. Two proteins, the SERT and TPH, are targets for both ROS and quinone-mediated oxidation (Stone et al., 1988; Sprague and Nichols, 1995; Kuhn and Arthur, 1998; Burrows et al., 2000; Kuhn and Geddes, 2000; Fornai et al., 2003). For instance, SERT inhibitors, including fluoxetine and citolapram, attenuate MDMA-induced ROS generation and serotonergic neurotoxicity (Kramer et al., 1997; Shankaran et al., 1999; Liechti et al., 2000), suggesting that the increase in ROS generation may be a consequence of the interaction between the transporter and MDMA, or more likely, reactive MDMA metabolites. In the present study, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)- α-MeDA-induced ROS generation was shown to be transporter-dependent, since increased ROS were observed in SERT and DAT transfected cells, but not in mock- transfected cells (Figure 6.). Consistent with previous studies with MDMA (Kramer et al., 1997; Shankaran et al., 1999), fluoxetine attenuated the increase in ROS generation in all SERT-expressing cells, suggesting a SERT-dependent mechanism for ROS-mediated SERT inhibition. Although the involvement of the SERT in MDMA neurotoxicity is firmly established, the exact nature of the interaction is unknown. To our knowledge, the formation of MDMA-adducted SERT proteins has yet to be demonstrated. Cysteinyl residues, which contain sulfhydryls moieties, serve as common targets for ROS-mediated oxidation. Indeed, the human SERT contains conformationally sensitive cysteine residues (cys 200 and 209) on the extracellular loop between transmembrane domains 3 and 4 (Amara et al., 1995; Chen et al., 2000), and which are thought to be involved in protein folding and stability but may also be involved in substrate and inhibitor binding.

162 Therefore, it is possible that these cysteinyl residues are oxidized and alkylated by α- MeDA quinone thioether-mediated ROS generation and nucleophilic attack, respectively. Biochemical markers of MDA and MDMA-induced serotonergic damage is the inhibition of TPH, the rate-limiting enzyme in the synthesis of 5-HT, which is located in the nerve terminals of serotonergic neurons (Stone et al., 1986; 1989a, b). MDMA induced-inhibition of TPH activity has been attributed to ROS and quinone-mediated oxidation of the protein. ROS and catechol quinones, which are generated directly or indirectly by MDMA, inhibit and covalently modify TPH (Kuhn et al., 1998, 2000). Interestingly, Kuhn and colleagues demonstrated that the oxidation of DA leads to the inactivation of TPH, the oxidation of protein sulfhydryls, and the formation of redox active quino-proteins, which are toxic to serotonergic neurons, suggesting that DA oxidation may contribute to MDMA-induced neurotoxicity. 5-(GSyl)-α-MeDA, and 2,5- bis(GSyl)-α-MeDA, readily cycle between their reduced (catechol) and oxidized (o- quinone) forms, and it is therefore possible that the covalent modification of TPH by these quinone species may inhibit TPH activity and lead to the formation of a redox active and electrophilic quino-protein, which may, subsequently, contribute to the degeneration of 5-HT axons. Consequently, future studies should be conducted to examine the potential deleterious effects of α-MeDA thioethers on TPH. In addition to the SERT and TPH, MDMA appears to induce the oxidation of cytochrome oxidase, an electron transport protein involved in cellular energy (ATP) production (Burrows et al., 2000), membrane lipids (Sprague et al., 1995), and DNA (Fornai et al., 2003). Interestingly, decreases in cytochrome oxidase activity were restricted to DA-rich brain regions, the striatum, nucleus accumbens, and substantia nigra, suggesting the potential involvement of DA in MDMA inhibition of cellular respiration. DA may compromise mitochondrial function via the formation of ROS and quinone species. Indeed, ROS and DA-derived quinones are known to directly inhibit mitochondrial enzymes associated with energy production (Ben-Schachar et al., 1995; Montine et al., 2000). Moreover, MDMA-induced peroxidation of membrane lipids has

163 been demonstrated (Sprague et al., 1995). Oxidation of lipids compromises cellular membrane integrity and consequently, may lead to a neurotoxic response. Finally, DNA is oxidized via the MDMA-induced generation of ROS causing ubiquitinated neuronal inclusions and nuclear condensations (Stumm et al., 1999; Fornia et al., 2003). In summary, ROS and quinone species, by-products of α-MeDA-thioether and DA oxidation, appear to contribute to the neurotoxic effects of the parent amphetamines, MDA and MDMA. 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA are significantly more potent ROS generators than either MDA or MDMA, inhibit SERT function, and therefore, should posses an increased capacity to directly or indirectly interact with other nucleophilic or reduced cellular molecules. Several molecular targets, including lipids, DNA, and multiple proteins (SERT, TPH, and cytochrome oxidase) have been identified as undergoing oxidation and quinone mediated covalent modification. Further studies examining the origin and precise involvement of ROS and oxidative stress should focus on the proteins affected and perhaps, the specific amino acids undergoing oxidation and alkylation by ROS and electrophilic quinones, respectively.

164 CHAPTER 7

MDA, MDMA, 5-(GSYL)-α-MEDA, AND 2,5-BIS(GSyl)-α-MeDA INDUCE LOSS OF CELL VIABILITY AND APOPTOSIS IN PRIMARY, JAR, AND hSERT-TRANSFECTED CELLS

I. INTRODUCTION AND RATIONALE MDA and MDMA-induced serotonergic neurotoxicity is manifest as depletions in brain 5-HT, inhibition of SERT and TPH activity, decreases in SERT protein expression and morphological damage and degeneration of 5-HT nerve terminals and axons. Interestingly, although these drugs produce severe, long-term damage to the serotonergic system, it appears that the 5-HT cell bodies of damaged neurons remain intact (Ricaurte et al., 1985; Harvey et al., 1993). Moreover, MDA and MDMA do not affect 5-HT concentrations in brain regions rich in serotonergic neuronal bodies, including the raphe nuclei (Harvey et al., 1993; Miller et al., 1997). Consistent with the lack of MDMA- induced neurotoxicity in the raphe nuclei, the α-MeDA thioethers, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA failed to produce any long-term deficits in 5-HT in the midbrain/diencephalon and pons/medulla (Miller et al., 1997; Bai et al., 1999). In addition, 5-(NAC)-N-Me-α-MeDA also fails to affect 5-HT concentrations in the midbrain regions following intrastriatal administration (see Chapter 4 for discussion). Whether α-MeDA metabolites produces morphological damage at the level of the cell body is yet to be determined. Contrary to findings suggesting a lack of MDMA-induced cell body neurotoxicity, several studies have demonstrated MDMA-induced apoptotic cell death in various cell models (Simantov and Tauber, 1997; Stumm et al., 1999; Montiel-Duarte et al., 2002). In fact, a recent study (Montiel-Duarte et al., 2004) demonstrated MDMA induced apoptosis in hepatic stellate cells. Interestingly, the apoptotic response was potentiated by GSH and N-acetyl cysteine, but diminished when cytochrome P450 was inhibited. Such in-vitro studies raise questions about the mechanism of MDA and MDMA induced neurotoxicity

165 and any effects the drugs may have on 5-HT cell bodies. The above studies also provide a rationale for the further examination of cell viability and MDMA-induced apoptosis. Therefore, using multiple cell models, the present chapter describes studies examining the potential role of cytotoxicity and apoptosis in MDA, MDMA, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA-induced neurotoxicity. We suspect that the α-MeDA thioethers are more potent inducers of apoptosis than either of the parent amphetamines. There are two principal mechanisms of cell death: necrosis and apoptosis. Necrosis is a passive process manifested by cellular swelling, denaturation of cytoplasmic proteins, a complete breakdown of cellular organelles, and cell lysis, eventually resulting in an inflammatory response. In contrast, apoptosis is an active process characterized by an organized, or “programmed”, cascade of biochemical events including, increases in cytoplasmic Ca++, the loss of mitochondrial membrane potential (∆ψm), the release of cytochrome c from the mitochondria, and the activation of cysteine proteases, or caspases (Sastry and Rao, 2000). Morphologically, apoptotic cells display distinct characteristics such as DNA fragmentation, reduction of volume, membrane blebbing, and nuclear condensation (Sastry and Rao, 2000). The machinery for apoptosis is present in virtually all mammalian cells and can be activated by a wide variety of intra- and extra-cellular signals. Interestingly, MDMA-induced neurotoxicity has been linked to several of the above apoptotic criteria, including increases in intracellular Ca++ (Beitia et al., 1999; Gudelsky and Yamamoto, 2003), inhibition of mitochondrial activity (Burrows et al., 2002), and the release of cytochrome c (Jimeniz et al., 2004) and caspase activation (Montiel-Duarte et al., 2002; Jimeniz et al., 2004).

166 II. RESULTS A. APOPTOSIS IN hSERT-AND hDAT-TRANSFECTED CELLS MDA and MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA induced apoptosis, as indicated by positive TUNEL staining, in hSERT-expressing SK-N-MC cells (Figure 7.1). Consistent with the hypothesis that metabolic activation may contribute to MDA and MDMA-induced neuronal toxicity, 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA were significantly more potent at inducing TUNEL-positive cells (~33% and ~43%), respectively) than the parent amphetamine. Fluoxetine attenuated apoptosis, whereas nomifensine had no effect (Figure 7.2), suggesting that the apoptotic response is SERT-dependent. In contrast to the significant level of apoptotic cell death in SERT-expressing cells, cells expressing the DAT were minimally affected by MDA, MDMA, or the α-MeDA thioethers (Figure 7.2), suggesting that MDA and MDMA-induced apoptosis is dependent on the presence of the SERT and selective for 5-HT cells. Thus, although 5- (GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA induced ROS generation in DAT- transfected SK-N-MC cells (Chapter 6), it appears the DAT-expressing cells are spared from cell death. No evidence for MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)- α-MeDA-induced apoptosis was observed in mock-transfected cells, which lack both the DAT and SERT proteins (data not shown).

B. APOPTOSIS IN SEROTONERGIC JAR CELLS. TUNEL staining revealed that MDA and MDMA induced apoptosis in serotonergic JAR cells (Figure 7.3). Consistent with the findings in SERT-expressing cells, 5-(GSyl)- α-MeDA and 2,5-bis(GSyl)-α-MeDA were significantly more potent than either of the parent drugs, inducing apoptosis in 43–53% of cells examined. Apoptosis was attenuated with fluoxetine, whereas nomifensine had no effect (Figure 7.3), suggesting a requirement for the SERT in MDA and MDMA-induced apoptosis.

167

45 *†

40

35 *†

30 φ 25 * 20 φ * 15

10 % (apoptotic) cells TUNEL-positive 5

0 ______+ fluoxetine + nomifensine

Figure 7.1: MDA, MDMA 5-(GSyl)-α-MeDA, and 2,5- (GSyl)-α-MeDA-induced apoptosis in hSERT-transfected SK-N-MC cells. Cells were incubated with vehicle (black bars), MDA (hatched bars), MDMA (dotted bars), 5-(GSyl)-α- MeDA (checkered bars), or 2,5-(GSyl)-α-MeDA (white bars) (100µM) for 24 hrs prior to TUNEL staining. For inhibition studies, cells were treated with fluoxetine (100 µM) or nomifensine (50 µM) for 30 min prior to, and concurrent with, incubation in the test compounds. Data are expressed as the mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) treatment groups and MDA (†) are considered significant at p < 0.05. Differences between fluoxetine and treatment only groups are significant at p < 0.05 (φ).

168

45 40 35 30 25 20 15 10 5 % (apoptotic) cellsTUNEL-positive 0 ______+ fluoxetine + nomifensine

Figure 7.2: Lack of MDA, MDMA 5-(GSyl)-α-MeDA, and 2,5- (GSyl)-α-MeDA-induced apoptosis in hDAT-transfected SK- N-MC cells. Cells were incubated with vehicle (black bars), MDA (hatched bars), MDMA (dotted bars), 5-(GSyl)-α-MeDA (checkered bars), or 2,5-(GSyl)-α-MeDA (white bars) (100µM) for 24 hrs prior to TUNEL staining. For inhibition studies, cells were treated with fluoxetine (100 µM) or nomifensine (50 µM) for 30 min prior to, and concurrent with, incubation in the test compounds. Data are expressed as the mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) treatment groups and MDA (†) are considered significant at p < 0.05. Differences between fluoxetine and treatment only groups are significant at p < 0.05 (φ).

169

60

*†

50

s s l

l *†

l l

e e

c c

) )

c c

i i t

t 40

o o

t t

p p

o o

p p

a a

( (

30 e

e * * φ v

v φ

i i

t t

i i

s s

o o

p p -

- 20 φ L

L φ

E E

N N

U U

T T

10

% %

0 ______fluoxetine nomifensine

Figure 7.3: MDA, MDMA 5-(GSyl)-α-MeDA, and 2,5-(GSyl)- α-MeDA-induced apoptosis in JAR cells. Cells were incubated with vehicle (black bars), MDA (dotted bars), MDMA (hatched bars), 5-(GSyl)-α-MeDA (white bars), or 2,5-(GSyl)-α-MeDA (checkered bars) (100µM) (100µM) for 24 hrs prior to TUNEL staining. For inhibition studies, cells were treated with fluoxetine (100 µM) or nomifensine (50 µM) for 30 min prior to, and concurrent with, incubation in the test compounds. Data are expressed as the mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) treatment groups and MDA (†) are considered significant at p < 0.05. Differences between fluoxetine and treatment only groups are significant at p < 0.05 (φ).

170 C. CELL VIABILITY AND APOPTOSIS IN STRIATAL AND HIPPOCAMPAL CELLS

Serotonergic nerve terminals contain TPH, whereas dopaminergic cells are rich in tyrosine hydroxylase (TH). Primary cell cultures from the striatum displayed large populations of TH positive, or “dopaminergic”, cells, whereas hippocampal primary cultures were significantly TPH positive, or “serotonergic” (Figure 7.4). Therefore, using TPH and TH as markers for cell viability, immunochemical analysis demonstrated significant differences in MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α- MeDA-induced neurotoxicity. MDA and MDMA induced a minimal decrease in TPH positive cells in both hippocampal (Figure 7.5) and striatal (Figure 7.6) cells. 5-(GSyl)-α- MeDA and 2,5-bis(GSyl)-α-MeDA were considerably more cytotoxic than MDA or MDMA (Figures 7.5 and 7.6). None of the compounds tested had any effect on TH positive cells (data not shown) demonstrating the serotonergic selectively of these compounds. The loss of TPH positive cells was significantly greater in hippocampal compared to striatal cells, perhaps a reflection of the different proportions of “serotonergic” verses “dopaminergic” cells. Fluoxetine protected cells against the loss of TPH positive cells, whereas nomifensine had no effect (Figures 7.5 and 7.6), suggesting the involvement of the SERT. However, the protection afforded by fluoxetine was significant in hippocampal cells only. Consistent with findings in JAR and SERT-transfected cells, MDA, MDMA, 5- (GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA induced significant apoptosis in hippocampal cells (Figure 7.7), and failed to produce significant levels of apoptosis in striatal cells (Figure 7.8), indicating a greater sensitivity of the hippocampal cell populations to the apoptotic effects of the parent amphetamines and α-MeDA thioethers. 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA were more potent inducers of apoptotic cell death than MDA or MDMA, supporting the hypothesis that metabolism contributes to the neurotoxicity of MDA and MDMA. Fluoxetine, but not nomifensine, protected both hippocampal and striatal cells from MDA, MDMA 5-(GSyl)-α-MeDA and 2,5-

171 bis(GSyl)-α-MeDA-induced apoptosis (Figures 7.7 and 7.8). Interestingly, for each compound, the percentage of apoptotic cells was greater than the percentage of cells lost as indicated by TPH staining. Although the reason for this is unclear, it is possible that some of the apoptotic cells are not positive for TPH.

172

% TH positive % TPH positive Striatal 46.6 ± 7.21 24.32 ± 4.32 cells

Hippocampal 24.58 ± 5.26 54.18 ± 6.58 cells

Figure 7.4: Tyrosine hydroxylase (TH) and tryptophan hydroxylase (TPH) immunoreactivity in hippocampal and striatal cells. Untreated striatal and hippocampal cells were incubated with antibodies against either TPH or TH, counterstained with DAB and visualized under a light microscope. Data are expressed as the mean ± SE (n=4).

173

70

60 φ φ 50 ** 40 *†

30 *†

% TPH positive cells positive TPH % 20

10

0 Fluoxetine Nomifensine

Figure 7.5: Effect of MDA, MDMA, 5-(GSyl)-α-MeDA and 5- (GSyl)–αMeDA on TPH immunoreactivity in primary hippocampal cells. Cells were incubated with vehicle (black bars), MDA (dotted bars), MDMA (hatched bars), 5-(GSyl)-α- MeDA (white bars), or 2,5-(GSyl)-α-MeDA (checkered bars) (100µM) (100µM) for 24 hrs prior to staining with antibodies against TPH. For inhibition studies, cells were treated with fluoxetine (100 µM) or nomifensine (50 µM) for 30 min prior to, and concurrent with, incubation in the test compounds. Data are expressed as the % TPH-positive cells; mean ± SE (n=4). (*) represents significant difference from control (p < 0.05); (†) represents significant difference from MDA-treated groups; (φ) represents significant difference between relevant treatment only and treatment + fluoxetine groups, (p < 0.05).

174

35

30

25 * 20 *† 15

10 % TPH positive cells positive TPH %

5

0 Fluoxetine Nomifensine

Figure 7.6: Effect of MDA, MDMA, 5-(GSyl)-α-MeDA and 5- (GSyl)–αMeDA on TPH immunoreactivity in primary striatal cells. Cells were incubated with vehicle (black bars), MDA (dotted bars), MDMA (hatched bars), 5-(GSyl)-α-MeDA (white bars), or 2,5- (GSyl)-α-MeDA (checkered bars) (100µM) (100µM) for 24 hrs prior to staining with antibodies against TPH. For inhibition studies, cells were treated with fluoxetine (100 µM) or nomifensine (50 µM) for 30 min prior to, and concurrent with, incubation in the test compounds. Data are expressed as the % TPH positive cells; mean ± SE (n=4). (*) represents significant difference from control (p < 0.05); (†) represents significant difference from MDA-treated groups;, (p < 0.05).

175

60

*† 50

40 *

30 * φ * φ φ φ 20

10 % (apoptotic)% cellsTUNEL-positive

0 fluoxetine nomifensine

Figure 7.7: MDA, MDMA 5-(GSyl)-α-MeDA, and 2,5-(GSyl)- α-MeDA-induced apoptosis in primary hippocampal cells. Cells were incubated with vehicle (black bars), MDA (dotted bars), MDMA (hatched bars), 5-(GSyl)-α-MeDA (white bars), or 2,5-(GSyl)-α-MeDA (checkered bars) (100µM) (100µM) for 24 hrs prior to TUNEL staining. For inhibition studies, cells were treated with fluoxetine (100 µM) or nomifensine (50 µM) for 30 min prior to, and concurrent with, incubation in the test compounds. Data are expressed as the mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) treatment groups and MDA (†) are considered significant at p < 0.05. Differences between fluoxetine and treatment only groups are significant at p < 0.05 (φ).

176

50 45 40 35 30 *† * 25 20 15 10

% TUNEL-positive (apoptotic) cells 5 0 fluoxetine nomifensine

Figure 7.8: MDA, MDMA 5-(GSyl)-α-MeDA, and 2,5- (GSyl)-α-MeDA-induced apoptosis in primary striatal cells. Cells were incubated with vehicle (black bars), MDA (dotted bars), MDMA (hatched bars), 5-(GSyl)-α-MeDA (white bars), or 2,5-(GSyl)-α-MeDA (checkered bars) (100µM) for 24 hrs prior to TUNEL staining. For inhibition studies, cells were treated with fluoxetine (100 µM) or nomifensine (50 µM) for 30 min prior to, and concurrent with, incubation in the test compounds. Data are expressed as the mean (N=4) ± SEM. Differences between; i) control and treatment groups (*) and, ii) treatment groups and MDA (†) are considered significant at p < 0.05.

177 III. DISCUSSION Consistent with previous reports (Simantov and Tauber, 1997; Stumm et al., 1999; Montiel-Duarte et al., 2002, 2004), MDA and MDMA induced apoptosis in various SERT-expressing, or “serotonergic” cell types, JAR cells, hSERT-transfected SK-N-MC cells, and primary hippocampal cells. Cytotoxicity, defined by the loss of TH or TPH positive immunoreactivity in primary cell cultures, was significantly greater in cells treated with α-MeDA thioethers compare to cell treated with the parent drugs. Apoptosis, characterized by TUNEL stain labeling of fragmented DNA, also proved to be more pronounced in cell treated with 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA. Cytotoxicity and apoptotic cell death were attenuated by fluoxetine indicating the involvement of the SERT. The neurotoxic response to MDA and MDMA is presumably limited to the serotonergic nerve terminals, whereas serotonin cell bodies are spared. However, in-vitro evidence exists suggesting that MDMA is capable of inducing apoptosis in multiple cell models, including human serotonergic JAR cells (Figure 7.3; Simantov and Tauber, 1997), rat neocortical neurons, (Stumm et al., 1999), and rat liver cells (Montiel-Duarte et al., 2002, 2004). Interestingly, consistent with findings in non- serotonergic human NMB cells (Simantov and Tauber, 1997), the compounds examined had little, if any, deleterious effects on DAT transfected SK-N-MC cells or primary striatal cells, populations which are considered primarily “dopaminergic”, in agreement with the selective serotonergic neurotoxicity characteristic of MDMA. In addition to MDMA, various amphetamine derivatives, including d-amphetamine (Stumm et al., 1999), d-fenfluramine (Bengel et al., 1998), and methamphetamine (Cadet et al., 1997; Davidson et al., 2001) produce biochemical and neuroanatomical changes associated with apoptosis. For instance, methamphetamine induced DNA fragmentation (“DNA Laddering”) and nuclear condensation, two morphological markers of apoptosis, in neuronal cell models (Cadet et al., 1997; Stumm et al., 1999). Furthermore, several neurochemical changes associated with the apoptotic cascade, including cytochrome c release, mitochondrial impairment, and caspase activation have been observed in-vitro

178 following exposure to MDMA (Montiel-Daurte et al., 2002). Interestingly, genes and proteins of the bcl-2 family regulated the apoptotic response induced by various amphetamines in a variety of cell models (Cadet et al., 1997; Di Migilo et al., 2000; Montiel-Daurte et al., 2002). MDMA neurotoxicity, for example, is correlated with decreases in bcl-xL (Montiel-Daurte et al., 2002), and attenuated by the over expression of bcl-2 (Cadet et al., 1997). Bcl-2 and related genes are proto-oncogenes that promote cell survival (Nunez and Clarke, 1994), block apoptotic damage and protect against cellular oxidative stress (Hockenbery et al., 1993; Kane et al., 1993). Therefore, because i) MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA induced neurotoxicity involves the generation of ROS, ii) 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA posses high redox cycling potential, iii) the anti-oxidative and anti-apoptotic proteins of the bcl-2 family are associated with MDMA exposure, and iv) increased oxidative stress has been linked to the induction of apoptosis, it remains possible that MDMA-induced ROS generation and oxidation of cellular macromolecules may initiate an apoptotic response. If not appropriately controlled, a cascade of complex intracellular events resulting in apoptotic death often accompanies ROS-induced cellular injury. Oxidation-mediated neuronal apoptosis has been associated with changes in intracellular Ca++ homeostasis (Kruman et al., 1998; Richter et al., 1997), mitochondrial impairment (Kim et al., 1999), the release of cytochrome c from the mitochondria (Kantrow et al., 1999), and the activation of caspases (Takai et al., 1998), and such pathways have been extensively examined. For example, cellular oxidants can affect Ca++ homeostasis and mitochondrial function via the reduction of sulfhydryls contained within Na+/Ca++ transport proteins (Kim et al., 1999; Richter et al., 1997). Interestingly, MDMA increases intracellular Ca++ via the generation of ROS and oxidation of membrane proteins and lipids (Beitia et al., 1999; Gudelsky and Yamamoto, 2003). MDMA induced ROS generation also inhibits the electron transport chain and production of ATP (Burrows et al., 2000). Oxidative species can impair mitochondrial function by reducing proteins involved in electron transport and

179 mitochondrial permeability (Kantrow et al., 2000; Fontaine and Bernardi, 1999). Increases in mitochondrial permeability lead to the collapse of the proton motive force and disrupts ionic homeostasis (Fontaine and Bernardi, 1999; LeMasters et al., 1999; Tatton and Olanow, 1999). Oxidation-induced increases in intracellular Ca++, and oxidation of mitochondrial membrane proteins permits the free exchange of ions and other cellular components between the mitochondrial matrix and the cytosol, resulting the rupture of the mitochondrial membrane, and release of cytochrome c (Thress et al., 1999). Cytochrome c, a protein involved in mitochondrial respiration, diffuses freely in the aqueous matrix between the inner and outer mitochondrial membranes, functioning as an electron shuttle between complex III (cytochrome bc1) and complex IV (cytochrome c oxidase) of the electron transport chain (Cortese and Hackenbock, 1993; Gupte and Hackenbrock, 1988). The release of cytochrome c is involved in the apoptotic cascade via the activation of cysteine proteases, or caspases (Qiu et al., 2000; Green and Reed, 1998). Again, similar to the effects of MDMA on Ca++ levels and mitochondrial function, MDMA exposure results in the oxidation of cytochrome oxidase, a key component of the electron transport chain, thereby impairing cellular respiration (Burrows et al., 2000). Interesting, inhibition of cytochrome oxidase was observed in DA-rich regions of the brain, suggesting that the oxidation of DA to ROS and reactive quinones may contribute to MDMA-induced inhibition of electron transport. This is important given the relationship demonstrated between MDMA neurotoxicity and DA metabolism suggesting that DA may be involved in the serotonergic neurotoxicity of MDMA. A potential role for apoptosis in MDMA-mediated neurotoxicity is supported by the requirement for DA. DA-induced oxidation of proteins in the electron transport chain inhibits mitochondrial respiration and initiates apoptotic cell death (Ben Schachar et al., 1995; Zhang et al., 1994). Not only does the oxidation of DA cause apoptotic cell death in-vivo (Hastings et al., 1996; Luo et al., 1999; Zhang et al., 2000), and in-vitro (Stokes et al., 1999, Jones et al., 2000; Pedrosa and Soares-Da-Silva, 2002; Haque et al., 2003), it

180 also potentiates MDMA-induced apoptosis in serotonergic cells (Simantov and Tauber, 1997). Using the HEK293 cell line, primary rat striatal cultures, and intrastriatal DA injections, Luo et al. (1998; 1999) demonstrated that DA-induced apoptosis was mediated by the “oxidation-involved” mechanism. Moreover, inhibition of neuronal DA uptake and ROS generation attenuate DA-induced apoptosis, implying that DA neurotoxicity may be mediated via oxidative mechanisms (Simantov et al., 1996; Cantuti-Castelvetri and Joseph, 1999). Hence, if DA is involved in MDMA-mediated serotonergic neurotoxicity, it is possible that the SERT-dependent cellular uptake of DA into 5-HT neurons may initiate an apoptotic response. Finally, MDMA activates caspase 3, a downstream protease of the apoptotic cascade (Monteil-Daurte et al., 2002). Caspases comprise a large class of proteolytic enzymes that are involved in apoptosis through the initiation of DNA fragmentation, disassembly of the cell, and the de-activation of several anti-apoptotic proteins (Thornberry and Lazebnik, 1998). Caspase-3-mediated cleavage of cellular proteins and DNA is activated by increased levels of cellular oxidants (Takai et al., 1998; Virag et al., 1998), and is dependent on increases in mitochondrial permeability (Susin et al., 1999), and the release of cytochrome c (Uehara et al., 1999). DA oxidation and neurotoxicity is associated with the activation of caspase 3 (Jones et al., 2003) implying that if DA is required for MDMA-induced serotonergic neurotoxicity, the apoptotic response may be due to the oxidative properties of DA. Furthermore, activation of caspases leads to the fragmentation and cleavage of multiple cellular macromolecules, including DNA. Although not yet attributed to the action of caspases, MDMA induces significant DNA fragmentation and nuclear condensation in cortical and striatal cells (Fornai et al., 2003). Further studies should help elucidate the potential roll of caspase activation in MDMA- induced cytotoxicity. In addition to inducing TUNEL-positive apoptosis, MDA, MDMA, 5-(GSyl)-α- MeDA and 2,5-bis(GSyl)-α-MeDA decreased the number of cells positive for TPH immunoreactivity in both hippocampal and striatal cell cultures (Figures 7.7 and 7.8).

181 These results are consistent with previous findings indicating MDA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA-induced the loss of SERT immunoreactivity (Bai, 2000) Serotonergic cytotoxicity, or loss of TPH positive cells, which was significantly greater in hippocampal cells compared to striatal cells, was attenuated with fluoxetine, indicating the participation of the SERT and suggesting that these compounds display selectivity for 5-HT cells. Moreover, none of the compounds examined had any effect on TH-positive cells. This may explain the difference in the extent of TUNEL positive cells between the hippocampal and striatal cell populations, given the relative proportion of cells positive for TPH or TH. Hippocampal cultures, containing a higher percentage of TPH containing cells are, perhaps, more sensitive to the effects of MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α-MeDA Interestingly, in hippocampal populations, the number of apoptotic cells was greater than the number of cells lost, as indicated by TPH immunoreactivity. Although the reason for this is unclear, it is possible that some of the apoptotic cells are not positive for TPH, or “serotonergic” in nature, thus arguing against the pure selectivity that these amphetamines and their metabolites demonstrate for 5-HT cells. Nevertheless, although some “non-serotonergic” cells may be affected, i) the attenuating effect of fluoxetine on both apoptosis and the loss of TPH-positive cells, ii) the lack of toxicity observed in TH-positive cells, and iii) the loss of SERT immunoreactivity suggests that MDA, MDMA, 5-(GSyl)-α-MeDA and 2,5-bis(GSyl)-α- MeDA primarily target the serotonergic system. Although cell culture models posses several advantages, including the ability to examine detailed biochemical and cellular mechanisms, it is often difficult to translate findings observed in-vitro to “real life” in-vivo mechanistic actions. With the ability to manipulate cell cultures, researchers gain control over variables that may affect the outcome. Consequently, drug and toxicant induced cellular changes may be attributed to specific characteristics of the particular cell type being cultured. Ultimately, to accurately define drug-induced mechanisms it is important that observations made in in-vitro models are also demonstrated in-vivo. That being said, one must take great care in

182 interpreting results demonstrated in-vivo due to the numerous inherent factors that may influence the outcome. It appears that ample in-vitro evidence exists suggesting MDA, MDMA and the α-MeDA thioethers induced apoptotic cell death and loss of cell viability to warrant further investigation in animal models. Therefore, to draw unambiguous conclusions on the cytotoxic effect of these compounds, future studies should be conducted examining the potential for MDA and MDMA-induced apoptotic cell death.

183 CHAPTER 8

CONCLUDING REMARKS

I. CONCLUSIONS The studies described in this dissertation were designed to test the hypothesis that systemically formed thioether metabolites of α-MeDA and N-Me-α-MeDA contribute to the serotonergic neurotoxicity of MDA and MDMA. Three specific hypotheses were tested; i) that the GSH conjugates access the brain via the GSH transporter located on the BBB, ii) that the mercapturic acid, 5-(NAC)-N-Me-α-MeDA, decreases long-term brain 5-HT concentrations following direct injection into the brain, and iii) that the mechanisms underlying the serotonergic neurotoxicity of MDA, MDMA, and α-MeDA thioethers involves the inhibition of SERT, the generation of ROS, the DA oxidation. Previous work from our laboratory has demonstrated that the thioether metabolites of α-MeDA, 5-(GSyl)-α-MeDA, 2,5-bis(GSyl)-α-MeDA, and 5-(NAC)-α-MeDA, are formed in-vivo, produce acute behavioral and biochemical changes similar to MDA and MDMA, and are significantly more potent serotonergic neurotoxicants than either of the parent amphetamines (Miller et al., 1995, Miller et al., 1997; Bai et al., 1999; Monks and Jones, 2002). Although the selective serotonergic neurotoxicity of these metabolites has been demonstrated, there is little understanding of the underlying mechanism(s) involved. Therefore, the focus of the present studies were, i) to investigate the transport of α- MeDA and N-Me-α-MeDA thioethers from the blood stream into the brain, and ii) to elucidate the toxicological mechanisms involved in the selective serotonergic neurotoxicity of these metabolites. Finally, because the formation of N-Me-α-MeDA thioethers is a direct result of the demethylenation of MDMA, a set of studies (Chapter 4) examined the effect of intrastriatally administered 5-(NAC)-N-Me-α-MeDA on brain neurotransmitter concentrations. The central administration of MDA or MDMA directly into the brain fails to reproduce the long-term serotonergic neurotoxicity observed following peripheral

184 administration of these substituted amphetamine derivatives, indicating a requirement for systemic metabolism in the serotonergic neurotoxicity of MDA and MDMA. In addition, several major metabolites, including α-MeDA, lack the characteristic “serotonin syndrome” induced by MDA and MDMA. Conjugation to GSH lowers the oxidation potential of α-MeDA and N-Me-α-MeDA, suggesting that the thioethers are more biochemically reactive (Miller et al., 1995), and may therefore play a role in the neurotoxic insult induced by MDA and MDMA. However, prior to affecting the brain, a systemically formed or administered compound must first gain access to the brain via transport across the highly selective BBB. Following intravenous administration in heavily anesthetized animals, 5-(GSyl)-α-MeDA has been identified in the brain (Miller et al., 1995). More over acivicin, an inhibitor of γ-GT, increases both the uptake of 5- (GSyl)-α-MeDA (Miller et al., 1995) and the neurotoxicity of peripherally administered MDA and MDMA (Chapter 3; Bai et al., 2001) implicating GSH transporters as potential uptake mechanisms. Using microdialysis and analytical techniques (LC-MS/MS and HPLC-CEAS) to collect and analyze brain extracellular fluid samples, respectively, thioether metabolites of α-MeDA and N-Me-α-MeDA were observed in the brain following peripheral administration of MDMA (Chapter 3), a finding consistent with the involvement of systemic metabolism in the serotonergic neurotoxicity of these amphetamine derivatives. Acivicin increased the concentration of thioethers in the brain, suggesting the compounds entered the brain via the GSH transporter. Furthermore, and perhaps most indicative of a potential role for thioethers in MDMA neurotoxicity, there was a significant correlation between the concentration of the N-Me-α-MeDA metabolites in the brain and long-term depletions of brain 5-HT and 5-HIAA, reliable markers for MDA and MDMA induced neurotoxicity. The in-vivo formation, brain uptake, and neurotoxicity of α-MeDA thioethers have been demonstrated with a variety of techniques. Thus, logically, the next set of studies was designed to examine the potential molecular and biochemical mechanisms that mediate the neurotoxic insult, including the inhibition of SERT, the SERT-dependent

185 cellular uptake of DA, and the generation of ROS. The SERT, which is a molecular target of a variety of mood-altering drugs, including MDA and MDMA, appears to be functionally compromised by α-MeDA thioethers. Interestingly, numerous medically available drugs, including antidepressants such as Prozac, produce their biochemical effects via the interaction with the SERT and inhibition of 5-HT uptake without causing significant deleterious results. In contrast, the interaction of MDA, MDMA, and the α- MeDA thioethers with the SERT results in a neurotoxic response. For instance, we observed an MDA, MDMA, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA-induced inhibition of cellular 5-HT uptake in a variety of SERT-expressing cell models (Chapter 5). The reason for the discrepancy in the neurotoxicity between the amphetamines and other SSRI’s is unknown, however the reactivity and oxidative properties of the α-MeDA thioethers of MDA and MDMA may provide an explanation. The interaction between α-MeDA thioethers and the SERT not only inhibits the uptake of 5-HT, it also appears to stimulate the uptake of DA into SERT-expressing cells. The requirement for DA in MDA and MDMA-induced serotonergic neurotoxicity has been suggested by a number of studies (Sprague and Nichols, 1995 a, b; Aguirre et al., 1998; Sprague et al., 1998; Shankaran et al., 1999). 5-(GSyl)-α-MeDA and 2,5- bis(GSyl)-α-MeDA induce acute increases in the rate of DA turnover and extracellular concentrations of DA (Miller et al., 1996, 1997). Consequently, the uptake of DA into 5- HT nerve terminals and the subsequent MAO-B-mediated metabolism of DA may produce intracellular oxidants. Indeed, it appears that SERT proteins are capable of transporting DA into 5-HT nerve terminals (Schmidt and Lovenberg, 1985). To our knowledge, this is the first set of studies demonstrating MDA, MDMA, 5-(GSyl)-α- MeDA, and 2,5-bis(GSyl)-α-MeDA stimulation of SERT-dependent cellular DA uptake (Chapter 5). The α-MeDA thioethers, presumably due to their increased reactivity, were more potent than the parent drugs. The precise role of DA in the neurotoxic insult remains debatable. Nevertheless, regardless of whether the role of DA is direct, via its oxidation and metabolism inside 5-HT nerve terminals, or indirect, via its affect on

186 ambient body temperature, further studies examining the contribution of DA in MDA, MDMA, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA induced serotonergic neurotoxicity should be conducted. The excessive generation of ROS observed in MDA and MDMA induced neurotoxicity has been attributed to the interaction between the drugs and the SERT, and vice verse, the inhibition of the SERT has been attributed to the increase in cellular oxidants. Consequently, it is well established that MDA and MDMA-induced neurotoxicity is mediated via the generation of ROS (Guldelsky and Yamamoto, 1994; Shankaran et al., 1999, 2000). Because MDA and MDMA are not redox active, the origin of ROS is likely indirect via the formation of metabolites of the metabolism of DA. α- MeDA thioethers display significant redox reactivity and are highly electrophilic in nature; thus, a potential role for the metabolites in MDA and MDMA-induced ROS generation appears likely. 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA are significantly more potent ROS generators than either of the parent amphetamines in a variety of “serotonergic” cell models (Chapter 6). It is likely that the ROS originates from both the redox cycling of α-MeDA thioethers and the SERT-mediated cellular uptake and subsequent oxidation of DA. Interestingly, although MDA, MDMA, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA-induced ROS generation was observed in DAT-expressing, or “dopaminergic” cells, ROS induced cellular damage and cytotoxicity appears dependent on the SERT. The cytotoxic, or neurotoxic, insult induced by MDA and MDMA appears to be limited to the 5-HT nerve terminals whereas the 5-HT cell bodies remain intact. That being said, a variety of in-vitro studies suggest that MDA and MDMA induce apoptotic cell death (Monteil-Daurte et al., 2002, 2004; Stumm et al., 1999). Thus, because oxidative stress, DA oxidation, and transporter inhibition have been associated with apoptosis, we examined MDA, MDMA, 5-(GSyl)-α-MeDA, and 2,5-bis(GSyl)-α-MeDA induced apoptosis and cytotoxicity in a variety of SERT-expressing cell models (Chapter 7). Consistent with a role for apoptosis in the neurotoxic insult induced by MDA and

187 MDMA, both the parent amphetamines and the α-MeDA thioethers induced apoptotic cell death in each of the SERT-expressing cell models examined. In contrast, treatment of DAT-expressing cells did not result in significant apoptosis, suggesting the compounds are selective for “serotonergic” cells. Interestingly, although the compounds induced ROS generation in both SERT and DAT-expressing SK-N-MC cells, it appears that the ROS-induced cytotoxicity is dependent on the presence of the SERT. Finally, α-MeDA thioethers are one metabolic product of MDA and MDMA; however, the major metabolic route for MDMA appears to be the direct demethylenation to N-Me-α-MeDA (de la Torre et al., 2000). Subsequent conjugation to GSH and mercapturic metabolism leads to the formation of the N-acetyl conjugate, 5-(NAC)-N-α- MeDA. Therefore, we examined the potential serotonergic neurotoxicity of 5-(NAC)-N- α-MeDA following intrastriatal administration (Chapter 4). Following injections of the mercapturic metabolite directly into the brain, significant decreases in brain 5-HT and 5- HIAA concentrations were observed. 5-(NAC)-N-α-MeDA was significantly more potent than peripheral administered MDMA, a finding which is consistent with the hypothesis that systemic metabolism of the parent amphetamine derivatives, MDA and MDMA, is required for the development of neurotoxicity.

II. FUTURE STUDIES Studies described in this dissertation were designed to build on previous findings from our laboratory implicating α-MeDA thioethers as contributors to the serotonergic neurotoxicity of MDA and MDMA and transport mechanisms across the BBB were identified. Thioether metabolites were identified in the brain following peripheral administration of MDMA providing convincing evidence that these systemically formed compounds do indeed reach the brain, most likely via GSH transporters. Future studies should focus on the precise mechanisms of transport, including potential regional variations in the locations of increased brain uptake. Perhaps the target regions for MDA and MDMA are a result of their proximity to the point of entry into the brain.

188 A second major focus of this dissertation was the in-vitro investigation into the detailed mechanisms underlying the neurotoxicity of α-MeDA thioethers, including ROS generation, SERT inhibition, involvement of DA, and apoptosis. Future studies should aim to answer several questions that have been raised by the studies reported here and provide a more in-depth examination of these potential mechanisms. For instance, the precise role of DA remains to be defined, the molecular targets, and specifically, the targeted amino acids, for ROS and quinone-induced oxidative damage need to be identified, and the role, if any, for apoptotic cell death in-vivo should be investigated. In addition, these compounds appear to have a minimal effect on the dopaminergic system; therefore, future studies should investigate the DA system as potential targets of α- MeDA thioethers. Finally, because TPH is also a target for MDA and MDMA-induced oxidative damage, it would be interesting to examine the potential modification of TPH by α-MeDA and N-MeDA-α-MeDA thioethers.

189 REFERENCES

Acquas E, Marrocu P, Pisanu A, Cadoni C, Zernig G, Saria A and Di Chiara G. (2001) Intravenous administration of ecstasy (3,4-methylendioxy-methamphetamine) enhances cortical and striatal acetylcholine release in vivo. Eur J Pharmacol. 418, 207-211.

Adkins EM, Barker EL and Blakely RD. (2001) Interactions of tryptamine derivatives with serotonin transporter species variants implicate transmembrane domain I in substrate recognition. Mol Pharmacol. 59, 514-523.

Aguire N, Barrionuevo M, Lasheras B and Del Rio, J. (1998) The role of dopaminergic systems in the perinatal sensitivity to 3,4-methylenedioxymethamphetamine-induced neurotoxicity in rats. J Pharmacol Exp Ther. 286, 1159-1165.

Amara SG, Sonders MS, Zahniser NR, Povlock SL and Daniels GM. (1998) Molecular physiology and regulation of catecholamine transporters. Adv Pharmacol. 42, 164- 168.

Androutsellis-Theotokis A, Ghassemi F and Rudnick G. (2001) A conformationally sensitive residue on the cytoplasmic surface of serotonin transporter. J Biol Chem. 276, 45933-459338.

Austin MC and O'Donnell SM. (1999) Regional distribution and cellular expression of tryptophan hydroxylase messenger RNA in postmortem human brainstem and pineal gland. J Neurochem. 72, 2065-2073.

Bai F, Lau SS and Monks TJ. (1999) Glutathione and N-acetylcysteine conjugates of α- methyldopamine produce serotonergic neurotoxicity. Possible role in methylenedioxyamphetamine-mediated neurotoxicity. Chem Res Toxicol. 12, 1150- 1157.

Bai F, Jones DC, Lau SS, and Monks TJ. (2001) The serotonergic neurotoxicity of 3,4- (±)-methylenedioxyamphetamine and 3,4-(±)-methylenedioxy-methamphetamine (ecstasy) is potentiated by inhibition of γ-glutamyl transpeptidase. Chem Res Toxicol. 14, 863-870.

Bai F (2000) Serotonergic toxicity of alpha-methyldopamine thioethers: Role in methylenedioxyamphetamine mediated neurotoxicity. Doctoral Dissertation, School of Pharmacy, University of Texas

190 Bankson MG and Cunningham KA (2001) 3,4-methylenedioxymethamphetamine (MDMA) as a unique model of 5-HT receptor function and 5-HT-dopamine interactions. J Pharmacol Exp Ther. 297, 846-852.

Barnes MM, James SP and Wood PB (1959) The formation of mercapturic acids. 1. Formation of mercapturic acid and the levels of glutathione in tissues. Biochem. J. 71, 680-690.

Beal MF, Kowall NW, Ellison DW, Mazurek MF, Swartz KJ, and Martin JB. (1886) Replication of the neurochemical characteristics of Huntington's disease by quinolinic acid. Nature. 321, 168-171.

Baez S, Segura-Aguilar J, Widersten M, Johansson A and Mannervik B. (1997) Glutathione transferases catalyze the detoxification of oxidized metabolites (o- quinones) of catecholamines and may serve as an antioxidant system preventing degenerative cellular processes. J. Biochem. 324, 25-28.

Banks WA. (1999) Physiology and pathology of the blood-brain barrier: implications for microbial pathogenesis, drug delivery and neurodegenerative disorders. J Neurovirol. 5, 538-555.

Barker EL, Perlman MA, Adkins EM, Houlihan WJ, Pristupa ZB, Niznik HB and Blakely RD. (1998) High affinity recognition of serotonin transporter antagonists defined by species-scanning mutagenesis. An aromatic residue in transmembrane domain I dictates species-selective recognition of citolapram and mazindol. J Biol Chem. 273, 19459-19468.

Battaglia G, Yeh SY, O’Hearn E, Molliver ME, Kuhar MJ and DeSouza EB. (1987) 3,4- Methylenedioxymethamphetamine and 3,4-methylenedioxy-amphetamine destroy serotonin terminals in rat brain: quantification of neurodegeneration by measurement of [3H]-paroxetine-labeled serotonin uptake sites. J Pharmacol Exp Ther. 242, 911- 916.

Beitia G, Cobreros A, Sainz L and Cenarruzabeitia E. (1999) 3,4-Methylenedioxy- methamphetamine (ecstasy)-induced hepatotoxicity: effect on cytosolic calcium signals in isolated hepatocytes. Liver. 19, 234-241.

Bengel D, Murphy DL, Andrews AM, Wichems CH, Feltner D, Heils A, Mossner R, Westphal H and Lesch KP. (1998) Altered brain serotonin homeostasis and locomotor insensitivity to 3, 4-methylenedioxymethamphetamine ("Ecstasy") in serotonin transporter-deficient mice. Mol Pharmaco. 53, 649-655.

191 Ben-Sharchar D, Zuk R and Glinka Y. (1995) Dopamine neurotoxicity: inhibition of mitochondrial respiration. J. Neurochem. 64, 718-723.

Berger UV, Gu XF and Azmitia EC. (1992) The substituted amphetamines 3,4- methylenedioxymethamphetamine, methamphetamine, p-chloroamphetamine and fenfluramine induce 5-HT release via a common mechanism blocked by FX and cocaine. European J. Pharm. 215, 153-160.

Bergeron C, Petrunka C and Weyer L. (1996) Copper/zinc superoxide dismutase expression in the human central nervous system. Correlation with selective neuronal vulnerability. Am J Pathol. 148, 273-279.

Berman SB, Zigmond MJ and Hastings TG. (1996) Modification of dopamine transporter function: effect of reactive oxygen species and dopamine. J Neurochemistry. 67, 593- 600.

Bogen IL, Haug KH, Myhre O and Fonnum F. (2003) Short- and long-term effects of MDMA ("ecstasy") on synaptosomal and vesicular uptake of neurotransmitters in vitro and ex vivo. Neurochem Int. 43, 393-400.

Bolla KI, McCann UD and Ricaurte GA. (1998) Memory impairment in abstinent MDMA ("ecstasy") users. Neurology. 51, 1532-1537.

Bolten JL, Trush MA, Penning TM, Dryhurst G and Monks TJ. (2000) Role of quinones in toxicology. Chem. Res. Toxicol. 13, 1-25.

Bowyer JF, Gough B, Slikker W Jr, Lipe GW, Newport GD and Holson RR. (1993) Effects of a cold environment or age on methamphetamine-induced dopamine release in the caudate putamen of female rats. Pharmacol Biochem Behav. 44, 87-98.

Bradham C, Qian T, Streetz K, Trautwein C, Brenner D and Lemasters J. (1998) The mitochondrial permeability transition is required for tumor necrosis factor alpha- mediated apoptosis and cytochrome c release. Mol. Cellular Biol. 18, 6353-6364.

Brannan TS, Maker HS and Raes IP. (1981) Regional distribution of catalase in the adult rat brain.J Neurochem. 36, 307-309.

Brightman MW and Reese TS. (1969) Junctions between intimately apposed cell membranes in the vertebrate brain. J Cell Biol. 40, 648-677.

192 Brodkin J, Malyala A and Nash JF. (1993) Effect of acute monoamine depletion on 3,4- methylenedioxymethamphetamine-induced neurotoxicity. Pharmacol Biochem Behav. 45, 647-653.

Buffinton GD, Ollinger K, Brunmark A and Cadenas E. (1989) DT-diaphorase-catalysed reduction of 1,4-naphthoquinone derivatives and glutathionyl-quinone conjugates. Effect of substituents on autoxidation rates. Biochem J. 257, 561-571.

Brust P, Hinz R, Kuwabara H, Hesse S, Zessin J, Pawelke B, Stephan H, Bergmann R, Steinbach J and Sabri O. (2003) In vivo measurement of the serotonin transporter with (S)-([18F]fluoromethyl)-(+)-McN5652. Neuropsychopharm. 28, 2010-2019.

Burrows KB, Guldelsky G and Yamamoto BK. (2000) Rapid and transient inhibition of mitochondrial function following methamphetamine or 3,4-methylenedioxy- methamphetamine administration. Eur. J. Pharmacol. 398, 11-18.

Cadet JL, Ladenheim B, Hirata H, Rothman RB, Ali S, Carlson E, Epstein C and Moran TH. (1995) Superoxide radicals mediate the biochemical effects of methylenedioxymethamphetamine (MDMA): Evidence from using CuZn-superoxide dismutase transgenic mice. Synapse. 21, 169–176.

Cadet JL and Brannock C. (1998) Free radicals and the pathobiology of brain dopamine systems. Neurochem Int. 32, 117-131.

Cadet JL, Thiriet N and Jayanthi S. (2001) Involvement of free radicals in MDMA- induced neurotoxicity in mice. Ann Med Interne (Paris). 152 Suppl 3:IS, 57-59.

Caldwell J (1980) Amphetamines and related stimulants: chemical, biochemical, clinical, and sociological aspects. CRC Press. Boca Raton, FLA.

Camarero J, Sanchez V, O'Shea E, Green AR and Colado MI. (2002) Studies, using in vivo microdialysis, on the effect of the dopamine uptake inhibitor GBR 12909 on 3,4- methylenedioxymethamphetamine ('ecstasy')-induced dopamine release and free radical formation in the mouse striatum. J Neurochem. 81, 961-972.

Cao Y, Li M, Mager S and Lester HA. (1998) Amino acid residues that control pH modulation of transport-associated current in mammalian serotonin transporters. J Neurosci. 18, 7739-7749.

193 Calabrese V, Scapagnini G, Ravagna A, Fariello RG, Giuffrida Stella AM and Abraham NG. (2002) Regional distribution of heme oxygenase, HSP70, and glutathione in brain: relevance for endogenous oxidant/antioxidant balance and stress tolerance. J Neurosci Res. 68, 65-75.

Callahan BT, Cord BJ and Ricaurte GA. (2001) Long-term impairment of anterograde axonal transport along fiber projections originating in the rostal RN after treatment with fenfluramine or 3,4-Methylenedioxymeth-amphetamine. Synapse. 40, 113-121.

Cantuti-Castelverti I and Joseph JA. (1999) Differential effect of dopamine catabolism and uptake inhibition on dopamine-induced calcium dysregulation and viability loss. Free Radical Biol. & Medicine. 27, 1393-1404.

Carvalho M, Hawksworth G, Milhazes N, Borges F, Monks TJ, Fernandes E, Carvalho F, and Bastos ML. (2002) Role of metabolites in MDMA (ecstasy)-induced nephrotoxicity: an in vitro study using rat and human renal proximal tubular cells. Arch Toxicol. 76, 581-588.

Carvalho M, Milhazes N, Remiao F, Borges F, Fernandes E, Amado F, Monks TJ, Carvalho F and Bastos ML. (2004) Hepatotoxicity of 3,4-methylene- dioxyamphetamine and alpha-methyldopamine in isolated rat hepatocytes: formation of glutathione conjugates. Arch Toxicol. 78, 16-24.

Chen JG, Sachpatzidis A and Rudnick G. (1997a) The third transmembrane domain of the serotonin transporter contains residues associated with substrate and cocaine binding. J Biol Chem. 272, 28321-28327.

Chen JG, Liu-Chen S and Rudnick G. (1997b) External cysteine residues in the serotonin transporter. Biochemistry. 36, 1479-8614.

Chen JG and Rudnick G. (2000) Permeation and gating residues in serotonin transporter. PNAS. 97, 1044-1049.

Chen JG, Liu-Chen S, and Rudnick G. (1998) Determination of external loop topology in the serotonin transporter by site-directed chemical labeling J Biol Chem. 273, 12675- 12681.

Cheng FC, Kuo JS, Huang HM, Yang DY, Wu TF, and Tsai TH. (2000) Determination of catecholamines in pheochromocytoma cell (PC-12) culture medium by microdialysis-microbore liquid chromatography. J. Chromatography. 870, 405-411.

194 Christophersen AS. (2000) Amphetamine designer drugs - an overview and epidemiology. Toxicol Letters. 112-113, 127-131.

Choi SR, Hou C, Oya S, Mu M, Kung MP, Siciliano M, Acton PD and Kung HF. (2000) Selective in vitro and in vivo binding of [(125)I]ADAM to serotonin transporters in rat brain. Synapse. 38, 403-412.

Chu T, Kumagai Y, DiStefano EW and Cho AK. (1996) Disposition of methylenedioxymethamphetamine and three metabolites in the brains of different rat strains and their possible roles in acute serotonin depletion. Biochem Pharmacol. 51, 789-796.

Clauwaert KM, Van Bocxlaer JF and De Leenheer AP. (2001) Stability study of the designer drugs "MDA, MDMA and MDEA" in water, serum, whole blood, and urine under various storage temperatures. Forensic Sci Int. 124, 36-42.

Cohen Z, Ehret M, Maitre M and Hamel E. (1995) Ultrastructural analysis of tryptophan hydroxylase immunoreactive nerve terminals in the rat cerebral cortex and hippocampus: their associations with local blood vessels. Neuroscience. 66, 555-569.

Colado MI, Murray TK and Green AR. (1993) 5-HT loss in rat brain following 3,4- methylenedioxymethamphetamine (MDMA), p-chloroamphetamine and fenfluramine administration and effects of chlormethiazole and dizocilpine. Br J Pharmacol. 108, 583-589.

Colado MI and Green AR. (1995) The spin trap reagent a-phenil-N-tertbutyl nitrone prevents “ecstasy”-induced neurodegeneration of 5-hydroxytryptamine neurons. Eur. J. Pharmacol. 28, 343–346.

Colado MI, O’Shea E, Granados R, Misra A, Murray TK and Green AR. (1997) A study of the neurotoxic effect of MDMA (“ecstasy”) on 5-HT neurons in the brains of mothers and neonates following administration of the drug during pregnancy. Br J Pharmacol 12, 827–833.

Colado MI, O'Shea E, Granados R, Murray TK and Green AR. (1997b) In vivo evidence for free radical involvement in the degeneration of rat brain 5-HT following administration of MDMA ('ecstasy') and p-chloroamphetamine but not the degeneration following fenfluramine. Br J Pharmacol. 121, 889-900.

195 Colado MI, Granados R, O'Shea E, Esteban B. and Green AR. (1998) Role of hyperthermia in the protective action of clomethiazole against MDMA ('ecstasy')- induced neurodegeneration, comparison with the novel NMDA channel blocker AR- R15896AR.Br J Pharmacol. 124, 479-484.

Colado MI, O'Shea E, Granados R, Esteban B, Martin AB and Green AR (1999) Studies on the role of dopamine in the degeneration of 5-HT nerve endings in the brain of Dark Agouti rats following 3,4-methylenedioxymethamphetamine (MDMA or 'ecstasy') administration. Br J Pharmacol. 126, 911-924.

Colado MI, Camarero J, Mechan AO, Sanchez V, Esteban B, Elliott JM AND Green AR. (2001) A study of the mechanisms involved in the neurotoxic action of 3,4- methylenedioxymethamphetamine (MDMA, 'ecstasy') on dopamine neurons in mouse brain. Br J Pharmacol. 134, 1711-1723.

Colado MI, O'Shea E and Green AR. (2004) Acute and long-term effects of MDMA on cerebral dopamine biochemistry and function. Psychopharmacology (Berl). Epub ahead of print

Colombrita C, Calabrese V, Stella AM, Mattei F, Alkon DL and Scapagnini G. (2003) Regional rat brain distribution of heme oxygenase-1 and manganese superoxide dismutase mRNA: relevance of redox homeostasis in the aging processes. Exp Biol Med (Maywood). 228, 517-524.

Colton C, Yao J, Grossman Y and Gilbert D. (1991) The effect of xanthine/xanthine oxidase generated reactive oxygen species on synaptic transmission. Free Radic Res Commun. 14, 385-393.

Commins DL, Vosmer G, Virus RM, Woolverton WL, Schuster CR and Seiden LS. (1987) Biochemical and histological evidence that methylenedioxy- methylamphetamine (MDMA) is toxic to neurons in rat brain. J Pharmacol Exp Ther. 241 338-345.

Cooper AJ. (1998) Role of astrocytes in maintaining cerebral glutathione homeostasis and in protecting the brain against xenobiotics and oxidative stress. In Glutathione in the Nervous System. 91-115; Christopher A Shaw edit, Taylor and Francis, Washington DC

Cornford EM, Braun LD, Crane PD and Oldendorf WH. (1978) Blood-brain barrier restriction of peptides and the low uptake of enkephalins. Endocrinology. 103, 1297- 1303.

196 Cortese J and Hackenbrock C. (1993) Motional dynamics of functional cytochrome c delivered by low pH fusion into the intermembrane space of intact mitochondria., Biochem. Biophys. Acta. 1142, 194-202.

Dahlstrom A, and Fuxe K. (1964). Evidence for the existence of monoamine-containing neurons in the central nervous system. I. Demonstration of monoamines in the cell bodies of brainstem neurons. Acta Physiol Scand. 62: (Suppl. 232), 1-55. de la Torre R, Farre M, Roset PN, Hernandez Lopez C, Mas M, Ortuno J, Menoyo E, Pizarro N, Segura J and Cami J. (2000) Pharmacology of MDMA in humans. Ann N Y Acad Sci. 914, 225-237.

Danscher G, Fjerdingstad EJ, Fjerdingstad E and Fredens K. (1976) Heavy metal content in subdivisions of the rat hippocampus (zinc, lead and copper). Brain Res. 112, 442- 446.

Darvesh AS, Shankaran M and Gudelsky GA. (2002) 3,4-Methylenedioxy- methamphetamine produces glycogenolysis and increases the extracellular concentration of glucose in the rat brain. J Pharmacol Exp Ther. 301, 138-144.

Davidson C, Gow AJ, Lee TH and Ellinwood EH. (2001) Methamphetamine neurotoxicity: necrotic and apoptotic mechanisms and relevance to human abuse and treatment. Brain Res Brain Res Rev. 36, 1-22.

De Miglio MR, Muroni MR, Simile MM, Calvisi DF, Tolu P, Deiana L, Carru A, Bonelli G, Feo F and Pascale RM. (2000) Implication of Bcl-2 family genes in basal and D- amphetamine-induced apoptosis in preneoplastic and neoplastic rat liver lesions. Hepatology. 31, 956-965.

Dowling GP and McDonough ET. (1987) 'Eve' and 'Ecstasy'. A report of five deaths associated with the use of MDEA and MDMA. JAMA. 257, 1615-1627

Du L, Bakish D and Hrdina PD. (2001) Tryptophan hydroxylase gene 218A/C polymorphism is associated with somatic anxiety in major depressive disorder. J Affect Disord. 65, 37-44.

Duvauchelle CL, Ikegami A, Robens J, Asami S, Kressin K and Castaneda E. (2000) Contextual effects on NAcc dopamine and behavioral activity after repeated intravenous cocaine administration. Brain Research. 862, 49-58.

197 Ehret M, Gobaille S, Cash CD, Mandel P and Maitre M. (1887) Regional distribution in rat brain of tryptophan hydroxylase apoenzyme determined by enzyme-linked immunoassay. Neurosci Lett. 73, 71-6.

Esteban B, O'Shea E, Camarero J, Sanchez V, Green AR and Colado MI. (2001) 3,4- Methylenedioxymethamphetamine induces monoamine release, but not toxicity, when administered centrally at a concentration occurring following a peripherally injected neurotoxic dose. Psychopharmacology (Berl). 154, 251-260.

Falk EM, Cook VJ, Nichols DE and Sprague JE. (2002) An antisense oligonucleotide targeted at MAO-B attenuates rat striatal serotonergic neurotoxicity induced by MDMA. Pharmacol Biochem Behav. 72, 617-622.

Faraj BA, Olkowski ZL and Jackson RT. (1994) Active [3H]-dopamine uptake by human lymphocytes: correlates with SERT activity. Pharmacol. 48, 320-327.

Fernandez F, Porras G, Mormede P, Spampinato U and Chaouloff F (2003) Effects of 3,4-methylenedioxymethamphetamine on locomotor activity and extracellular dopamine in the nucleus accumbens of Fischer 344 and Lewis rats. Neurosci Letters. 335, 212-216.

Fischer HS, Zernig G, Schatz DS, Humpel C and Saria A. (2000) MDMA ('ecstasy') enhances basal acetylcholine release in brain slices of the rat striatum. Eur J Neurosci. 12, 1385-90.

Fleckenstein AE, Haughey HM, Metzger RR, Kokoshka JM, Riddle EL, Hanson JE, Gibb JW and Hanson GR. (1999) Differential effects of psychostimulants and related agents on dopaminergic and serotonergic transporter function. Eur J Pharmacol. 382, 45-49.

Fontaine E and Bernardi P. (1999) Progress on the mitochondrial permeability transition pore: Regulation by complex I and Ubiquinone analogs. J. Bioenergetics & Biomembranes. 31, 335-345.

Fornai F, Giorgi FS, Gesi M, Chen K, Alessri MG and Shih JC. (2001) Biochemical effects of the monoamine neurotoxins DSP-4 and MDMA in specific brain regions of MAO-B-deficient mice. Synapse. 39, 213-221.

Ghersi-Egea JF, Perrin R, Leininger-Muller B, Grassiot MC, Jeandel C, Floquet J, Cuny G, Siest G and Minn A. (1993) Subcellular localization of cytochrome P450, and activities of several enzymes responsible for drug metabolism in the human brain. Biochem Pharmacol. 45, 647-658.

198 Ghersi-Egea JF, Leininger-Muller B, Cecchelli R and Fenstermacher JD. (1995) Blood- brain interfaces: relevance to cerebral drug metabolism. Toxicol Lett. 82-83, 645-53.

Ghersi-Egea JF and Strazielle N. (2001) Brain drug delivery, drug metabolism, and multidrug resistance at the choroid plexus. Microsc Res Tech. 52, 83-8.

Gollamudi R, Ali SF, Lipe G, Newport G, Webb P, Lopez M, Leakey JE, Kolta M and Slikker W Jr. (1989) Influence of inducers and inhibitors on the metabolism in vitro and neurochemical effects in vivo of MDMA. Neurotoxicology. 10, 455-466.

Gordon CJ, Watkinson WP, O'Callaghan JP, and Miller DB. (1991) Effects of 3,4- methylenedioxymethamphetamine on autonomic thermoregulatory responses of the rat. Pharmacol. Biochem. Behav. 38, 339-44.

Graham DG, Tiffany SM, Bell WR and Gutknecht WF. (1978) Auto-oxidation verses covalent binding of quinones as the mechanism of toxicity of dopamine, 6- hydroxydopamine, and related compounds toward C1300 neuroblastoma cells in- vitro. Mol Pharmacol 14, 644-653.

Green DR and Reed JC. (1998) Mitochondria and apoptosis. Science. 281, 1309-1312.

Green AR, Mechan AO, Elliott JM, O'Shea E and Colado MI. (2003) The pharmacology and clinical pharmacology of 3,4-methylenedioxymeth-amphetamine (MDMA, "ecstasy"). Pharmacol Rev. 55(3), 463-508.

Grillo MP and Hua F. (2003) Identification of zomepirac-S-acyl-glutathione in vitro in incubations with rat hepatocytes and in vivo in rat bile. Drug Metab Dispos. 31, 1429-1436.

Grupte S and Hackenbrock C. (1988) Multi-dimensional diffusion modes and collision frequencies of cytochrome c with its redox partners. J. Bio. Chem., 263, 5241-5247.

Guan X, Hoffman B, Dwivedi C and Matthees DP. (2003) A simultaneous liquid chromatography/mass spectrometric assay of glutathione, cysteine, homocysteine and their disulfides in biological samples. J Pharm Biomed Anal. 31, 251-61.

Gudelsky GA and Yamamoto BK. (1994) MDMA increases the extracellular concentration of 2,3-dihydroxybenzoic acid in the ST: Evidence for increased hydroxyl radical formation. Soc Neurosci Abst. 24, 419.

199 Gudelsky GA and Nash JF. (1996) Carrier-mediated release of 5-HT by 3,4- methylenedioxymethamphetamine: implications for 5-HT-dopamine interactions. J Neurochemistry. 66, 243-249.

Gudelsky GA. (1996) Effect of ascorbate and cysteine on the 3,4-methyl- enedioxymethamphetamine-induced depletion of brain 5-HT. J Neural Transm 103, 1397–1404.

Gudelsky GA and Yamamoto BK. (2003) Neuropharmacology and neurotoxicity of 3,4- methylenedioxymethamphetamine. Methods Mol Med. 79, 55-73.

Gutierrez B, Pintor L, Gasto C, Rosa A, Bertranpetit J, Vieta E and Fananas L. (1998) Variability in the serotonin transporter gene and increased risk for major depression with melancholia. Hum Genet. 103, 319-322.

Halliwell B. (1992) Reactive oxygen species and the central nervous system. J. Neurochem. 59, 1609-1623.

Hansen JP, Riddle EL, Sandoval V, Brown JM, Gibb JW, Hanson GR and Fleckenstein AE. (2002) Methylenedioxymethamphetamine decreases plasmalemmal and vesicular dopamine transport: mechanisms and implications for neurotoxicity. J Pharmacol Exp Ther. 300, 1093-1100.

Haque ME, Asanuma M, Higashi Y, Miyazaki I, Tanaka K and Ogawa N. (2003) Overexpression of Cu-Zn superoxide dismutase protects neuroblastoma cells against dopamine cytotoxicity accompanied by increase in their glutathione level. Neurosci Res. 47, 31-37.

Harkin A, Connor TJ, Mulrooney J, Kelly JP and Leonard BE. (2001) Prior exposure to methylenedioxyamphetamine (MDA) induces a serotonergic loss and changes in spontaneous exploratory and amphetamine-induced behaviors in rats. Life Sciences. 68, 1367-1382.

Hartter DE and Barnea A. (1980) Brain tissue accumulates by two-ligand dependent saturable process. J Biol Chem. 263, 799-805.

Harvey JA, McMaster SE and Romano AG. (1993) Methylenedioxyamphetamine: neurotoxic effects on serotonergic projections to brainstem nuclei in the rat. Brain Res. 619, 1-14.

Hastings TG, Lewis DA, and Zigmond MJ. (1996) Role of oxidation in the neurotoxic effects of intrastriatal dopamine injections. Proc. Natl. Acad. Sci USA. 93, 1956-1961.

200

Hastings TG and Zigmond MJ. (1997) Loss of dopaminergic neurons in Parkinsonism: possible role of reactive dopamine metabolites. J. Neural Transm. [suppl] 49, 103- 110.

Haughey HM, Brown JM, Fleckenstein AE and Hanson GR. (1999) Oxygen radicals differentially affect Na+/Cl(-)-dependent transporters. Eur J Pharmacol. 379, 107- 110.

Hegadoren KM, Baker GB and Bourin M. (1999) 3,4-Methylenedioxy analogues of amphetamine: defining the risks to humans. Neurosci Biobehav Rev. 4, 539-553.

Henry JA. Jeffreys KJ, and Dawling S (1992) Toxicity and deaths from 3,4- methylenedioxymethamphetamine. Lancet 340, 384-387.

Hensley K, Carney JM, Mattson MP, Aksenova M, Harris M, Wu JF, Floyd RA and Butterfield DA. (1994) A model for beta-amyloid aggregation and neurotoxicity based on free radical generation by the peptide: relevance to Alzheimer disease. Proc Natl Acad Sci U S A. 91, 3270-3274.

Hill BA, Davison KL, Dulik DM, Monks TJ and Lau SS. (1994) Metabolism of 2- (glutathion-S-yl)hydroquinone and 2,3,5- (triglutathion-S-yl)hydroquinone in the in situ perfused rat kidney: relationship to nephrotoxicity. Toxicol Appl Pharmacol. 129, 121-132.

Hiramatsu M, Kumagai Y, Unger SE and Cho, AK (1990) Metabolism of methylenedioxymethamphetamine: formation of dihydroxymethamphetamine and a quinone identified as its glutathione adduct. J Pharmacol Exp Ther 254, 521-527.

Hirsch T, Marzo I and Kroemer G. (1997) Role of the mitochondrial permeability transition pore in apoptosis. Biosci Reports. 17, 67-76.

Hockenbery DM, Oltvai ZN, Yin XM, Milliman CL and Korsmeyer SJ. (1993) Bcl-2 functions in an antioxidant pathway to prevent apoptosis. Cell. 75, 241-251.

Homma M, Suzuki H, Kusuhara H, Naito M, Tsuruo T and Sugiyama Y. (1999) High- affinity efflux transport system for glutathione conjugates on the luminal membrane of a mouse brain capillary endothelial cell line (MBEC4). J Pharmacol Exp Ther. 288, 198-203.

201 Jayanthi S, Ladenheim B, Andrews AM and Cadet JL. (1999) Overexpression of human copper/zinc superoxide dismutase in transgenic mice attenuates oxidative stress caused by methylenedioxymethamphetamine (Ecstasy). Neuroscience. 91, 1379- 1387.

Jette L, Tetu B and Beliveau R. (1993) High levels of P-glycoprotein detected in isolated brain capillaries. Biochim Biophys Acta. 1150, 147-154.

Jiang XR, Wrona MZ, and Dryhurst G. (1999) Tryptamine-4,5-dione, a putative endotoxic metabolite of the superoxide-mediated oxidation of serotonin, is a mitochondrial toxin: possible implications in neurodegenerative brain disorders. Chem Res Toxicol. (5), 429-436.

Jimeniz A, Jorda EG, Verdaguer E, Pubill D, Sureda FX, Canudas AM, Escubedo E, Camarasa J, Camins A and Pallas M. (2004) Neurotoxicity of amphetamine derivatives is mediated by caspase pathway activation in rat cerebellar granule cells. Toxicol Appl Pharmacol. 196, 223-234.

Johnson JA, el Barbary A, Kornguth SE, Brugge JF and Siegel FL. (1993) Glutathione S- transferase isoenzymes in rat brain neurons and glia. J Neurosci. 13, 2013-2023.

Johnson M, Elayan I, Hanson GR, Foltz RL, Gibb JW and Lim HK. (1992) Effects of 3,4-dihydroxymethamphetamine and 2,4,5-trihydroxymethamphetamine, two metabolites of 3,4-methylenedioxyamphetamine, on central serotonergic and dopaminergic systems. J Pharmacol Exp Ther. 261, 447-453.

Johnston L, Bachman J and O’Malley P. (2000) The monitoring of future national results on adolescent drug abuse: overview of key findings, 1999. National institute of Drug Abuse, NIH publication 00-4690.

Jones DJ, Gunasekar PG, Borowitz JB and Isom GE. (2000) Oxidative stress-mediated DA-induced apoptosis is enhanced by cyanide in differentiate PC12 cells. J Neurochem. 74, 2296-2304.

Jones DC, Prabhakaran K, Li L, Gunasekar PG, Shou Y, Borowitz JL and Isom GE. (2003) Cyanide enhancement of dopamine-induced apoptosis in mesencephalic cells involves mitochondrial dysfunction and oxidative stress. Neurotoxicology. 24, 333- 342.

Kane DJ, Sarafian TA, Anton R, Hahn H, Gralla EB, Valentine JS, Ord T and Bredesen DE. (1993) Bcl-2 inhibition of neural death: decreased generation of reactive oxygen species. Science. 262, 1274-1287.

202

Kannan R, Kuhlenkamp JF, Jeandidier E, Trinh H, Ookhtens M. and Kaplowitz N. (1990) Evidence for carrier-mediated transport of glutathione across the blood-brain barrier in the rat. J. Clin. Invest. 85, 2009-2013.

Kannan R, Kuhlenkamp JF, Ookhtens M, Kaplowitz N. (1992) Transport of glutathione at blood-brain barrier of the rat: inhibition by glutathione analogs and age- dependence. J Pharmacol Exp Ther. 263, 964-970.

Kannan R, Mittur A, Bao Y, Tsuruo T and Kaplowitz N. (1999) GSH transport in immortalized mouse brain endothelial cells: evidence for apical localization of a sodium-dependent GSH transporter. J Neurochem. 73, 390-399.

Kannan R, Chakrabarti R, Tang D, Kim KJ and Kaplowitz N. (2000) GSH transport in human cerebrovascular endothelial cells and human astrocytes: evidence for luminal localization of Na+-dependent GSH transport in HCEC. Brain Res. 852, 374-382.

Kantrow SP, Tatro LG and Pantadosi CA. (2000) Oxidative stress and adenine nucleotide control of mitochondrial permeability transition. Free Radical Biol. & Medicine. 28, 251-260.

Kappus H and Sies H. (1981) Toxic drug effects associated with oxygen metabolism: redox cycling and lipid peroxidation. Experientia. 37, 1233-1241.

Kastin AJ and Pan W. (2003) Peptide transport across the blood-brain barrier. Prog Drug Res. 61, 79-100.

Kastin AJ and Pan W. (2004) Polypeptide delivery across the blood-brain barrier. Curr Drug Target CNS Neurol Disord. 3, 131-136.

Kim KJ, Jang YY, Han ES and Lee CS. (1999) Modulation of brain mitochondrial membrane permeability and synaptosomal Ca++ transport by dopamine oxidation. Mol. Cellular Biochem. 201, 89-98.

Kintz P and Samyn N. (1999) Determination of "Ecstasy" components in alternative biological specimens.J Chromatogr B Biomed Sci Appl. 733, 137-143.

Klevin MS and Seiden LS. (1992) Methamphetamine-induced neurotoxicity: structure- activity relationships. Ann NY Acad Sci. 654, 292-301.

203 Kolber AR, Bagnell CR, Krigman MR, Hayward J and Morell P. (1979) Transport of sugars into microvessels isolated from rat brain: a model for the blood-brain barrier. J Neurochem. 33, 419-431.

Koch S and Galloway MP. (1997) MDMA induced dopamine release in-vivo: role of endogenous 5-HT. J Neural Trans 104, 135-146.

Kramer HK, Poblete JC and Azmitia EC. (1997) Activation of protein kinase C (PKC) by 3,4-methylenedioxymethamphetamine (MDMA) occurs through the stimulation of 5- HT receptors and transporter. Neuropsychopharmacoogy. 17, 118-129.

Kramer HK, Poblete JC and Azmitia EC. (1998) Characterization of the translocation of protein kinase C (PKC) by 3,4-methylenedioxymethamphetamine (MDMA/ecstasy) in synaptosomes: evidence for a presynaptic localization involving the of the serotonin transporter (SERT). Neuropsychopharmacol. 19, 265-277.

Kreth K, Kovar K, Schwab M and Zanger UM. (2000) Identification of the human cytochromes P450 involved in the oxidative metabolism of "Ecstasy"-related designer drugs. Biochem Pharmacol. 59, 1563-1571.

Kretzschmar M, Brust P, Zessin J, Cumming P, Bergmann R and Johannsen B. (2003) Autoradiographic imaging of the serotonin transporter in the brain of rats and pigs using S-([18F]fluoromethyl)-(+)-McN5652. Eur Neuropsychopharmacol. 13, 387- 397.

Kruman I, Guo Q, and Mattson MP. (1998) Calcium and reactive oxygen species mediate staurosporine-induced mitochondrial dysfunction and apoptosis in PC12 cells. J Neurosci Res. 51, 293-308.

Kubota N, Kiuchi Y, Nemoto M, Oyamada H, Ohno M, Funahashi H, Shioda S and Oguchi K. (2001) Regulation of serotonin transporter gene expression in human glial cells by growth factors. Eur J Pharmacol. 417, 69-76.

Kuhn DM and Arthur RE. (1999) L-DOPA-quinone inactivates tryptophan hydroxylase and converts the enzyme to a redox-cycling quinoprotein. Brain Res Mol Brain Res 73, 78-84.

Kuhn DM and Arthur RE. (1998) Dopamine inactivates tryptophan hydroxylase and forms a redox cycling quino-protein: possible endogenous toxin to serotonin neurons. J Neurosci. 18, 7111-7117.

204 Kuhn DM and Geddes TJ. (2000) Molecular footprints of neurotoxic amphetamine action. Ann N Y Acad Sci. 914, 92-103.

Kumagai Y, Lin L, Schmitz DA and Cho AK. (1991) Hydroxyl radical mediated demethylenation of (methylenedioxy)phenyl compounds. Chem Res Toxicol 4, 330- 334.

Kumagai Y, Lin LY, Hiratsuka A, Narimatsu S, Suzuki T, Yamada H, Oguri K, Yoshimura H and Cho AK. (1994) Participation of cytochrome P450-2B and -2D isozymes in the demethylenation of methylenedioxymethamphetamine enantiomers by rats. Mol Pharmacol. 45, 359-365.

Kunikowska G and Jenner P. (2002) The distribution of copper, zinc- and manganese- superoxide dismutase, and glutathione peroxidase messenger ribonucleic acid in rat basal ganglia. Biochem Pharmacol. 63, 1159-1164.

Kusuhara H and Sugiyama Y. (2001a) Efflux transport systems for drugs at the blood- brain barrier and blood-cerebrospinal fluid barrier (Part 1). Drug Discov Today. 6, 150-156.

Kusuhara H and Sugiyama Y. (2001b) Efflux transport systems for drugs at the blood- brain barrier and blood-cerebrospinal fluid barrier (Part 2). Drug Discov Today. 6, 206-212.

Lang W, Mao J, Wang Q, Niu C, Doyle TW and Almassian B. (2000) Isolation and identification of metabolites of porfiromycin formed in the presence of a rat liver preparation. J Pharm Sci. 89, 191-198.

LeMasters JJ, Qian T, Bradham CA, Brenner DA, Cascio WE, Trost LC, Nishimura N, Nieminen AL and Herman B. (1999) Mitochondrial dysfunction in the pathogenesis of necrotic and apoptotic cell death. J Bioenergetic & Biomembranes. 31, 305-319.

Leonardi ET and Azmitia EC. (1994) MDMA (ecstasy) inhibition of MAO type A and type B: comparisons with fenfluramine and fluoxetine (Prozac). Neuropsychopharmacology. 10, 231-238.

Liechti ME, Baumann C, Gamma A and Vollenweider FX (2000) Acute psychological effects of 3,4-Methylenedioxymethamphetamin3 (MDMA, “ecstasy”) are attenuated by the 5-HT reuptake inhibitor, Citalopram. Neuropsychopharmacology. 22, 513-521.

205 Li J, Wang YF, Zhou RL, Yang L, Zhang HB and Wang B. (2003) [Association between tryptophan hydroxylase gene polymorphisms and attention deficit hyperactivity disorder with or without learning disorder] Zhonghua Yi Xue Za Zhi. 83, 2114-2118.

Lim HK and Foltz RL. (1988) In vivo and in vitro metabolism of 3,4- methylenedioxymethamphetamine in the rat: identification of metabolites using an ion trap detector. Chem. Res. Toxicol. 1, 370-378.

Lin LY, Kumangi Y and Cho AK. (1992) Enzymatic and chemical demethylenenation of (methylenedioxy) amphetamine and (methylenedioxy)-methamphetamine by rat brain microsomes. Chem. Res. Toxicol. 5, 401-406.

Logan BJ, Laverty R, Sanderson WD, and Yee YB. (1988) Differences between rats and mice in MDMA (methylenedioxymethamphetamine) neurotoxicity. Eur J Pharmacol. 152, 227-234.

Luo Y, Umegaki H, Wang X, Abe R and Roth GS. (1998) Dopamine induces apoptosis through an oxidation-involved SAPK/JNK activation pathway. J Biol Chem. 273, 3756-3764.

Luo Y, Hattori A, Munoz J, Qin ZH and Roth GS. (1999) Intrastriatal dopamine injections induces apoptosis through oxidation-involved activation of transcription factors AP-1 and NF-κB in rats. Mol. Pharmacol. 56, 254-264.

Malberg JE, Sabol KE and Seiden LS. (1996) Co-administration of MDMA with drugs that protect against MDMA neurotoxicity produces different effects on body temperature in the rat. J. Pharm. Exp. Ther. 278, 258-267.

Malberg JE and Seiden LS. (1998) Small changes in ambient temperature cause large changes in 3,4-methylenedioxymethamphetamine (MDMA)-induced serotonin neurotoxicity and core body temperature in the rat. J Neurosci. 18, 5086-5094.

Malpass A, White JM, Irvine RJ, Somogyi AA and Bochner F. (1999) Acute toxicity of 3,4-methylenedioxymethamphetamine (MDMA) in Sprague-Dawley and Dark Agouti rats. Pharmacol Biochem Behav. 64, 29-34.

Mamounas LA and Molliver ME. (1988) Evidence for dual serotonergic projections to neocortex: axons from the dorsal and median raphe nuclei are differentially vulnerable to the neurotoxin p-chloroamphetamine (PCA). Exp Neurol. 102, 23-36.

206 Mamounas LA, Mullen CA, O'Hearn E and Molliver ME. (1991) Dual serotonergic projections to forebrain in the rat: morphologically distinct 5-HT axon terminals exhibit differential vulnerability to neurotoxic amphetamine derivatives. J Comp Neurol. 4, 558-586.

Maurer HH, Bickeboeller-Friedrich J, Kraemer T and Peters FT. (2000) Toxicokinetics and analytical toxicology of amphetamine-derived designer drugs ('Ecstasy'). Toxicol Lett. 112-113, 133-142.

McCann UD and Ricaurte GA. (1991) Lasting neuropsychiatric sequelae of (+/-) methylenedioxymethamphetamine (ecstasy) in recreational users. J. Clin. Psychopharm. 11, 302-305.

McCann UD, Lowe KA and Ricaurte GA. (1997) Long lasting effects of recreational drugs of abuse on the central nervous system. Neuroscientist. 3, 399-411.

McCann UD, Mertle M, Eligulashvili V and Ricaurte GA. (1999) Cognitive performance in (±) 3,4-methylenedioxyamphetamine (MDMA, "ecstasy") users: a control study. Psychopharmacology. 143, 417-425.

McCann UD, Szabo Z, Scheffel U, Dannals RF and Ricaurte GA. (1998) Positron emission tomographic evidence of toxic effects of MDMA ("ecstasy") on brain serotonin neurons in human beings. Lancet. 352, 1433-1437.

McGuire P. (2000) Long-term psychiatric and cognitive effects of MDMA use. Toxicol Letters. 112-113, 153-156.

McGuire P and Faye T. (1992) Flashbacks following MDMA. Br J Psychiatry. 160, 276.

Meister A and Tate SS. (1976) Glutathione and related γ-glutamyl compounds. Biosynthesis and utilization. Annu Rev Biochem. 45, 559-604.

Meister A and Anderson ME. (1983) Glutathione. Annul Rev Biochem. 52, 711-760.

Metzger RR, Hanson GR, Gibb JW and Fleckenstein AE. (1998) 3,4- Methylenedioxymethamphetamine-induced acute changes in dopamine transporter function. Eur J Pharmacol. 349, 205-210.

Midha KK, Hubbard JW, Bailey K and Cooper JK. (1978) Alpha-Methyldopamine, a key intermediate in the metabolic disposition of 3,4-methylenedioxymethamphetamine in vivo in dog and monkey. Drug Metab. Dispos. 6, 623-630.

207 Miller DS. (2003) Confocal imaging of xenobiotic transport across the blood-brain barrier. J Exp Zoolog Part A Comp Exp Biol. 300, 84-90.

Miller RT, Lau SS and Monks TJ. (1995) Metabolism of 5-(GSyl)-α-methyldopamine following intracerebroventricular administration to male Sprague-Dawley rats. Chem Res Toxicol. 8, 634-641.

Miller RT, Lau SS and Monks TJ. (1996) Effect of 5-(GSyl)-α-methyldopamine on dopamine, serotonin and norepinepherine concentrations following intracerebroventricular administration to male Sprague-Dawley rats. Chem Res Toxicol. 9, 457-465.

Miller RT, Lau SS and Monks TJ. (1997) 2,5-bis-(GSyl)-α-methyldopamine, a putative metabolite of (±)-3,4-methylenedioxyamphetamine, decreases brain serotonin concentrations. Eur J Pharmacol. 323, 173-180.

Miller RT. (1996) Metabolism-dependent bioactivation of methylenedioxy-amphetamine to neurotoxic quinol-thioethers. Doctoral Dissertation, School of Pharmacy, University of Texas

Mokrzan EM, Kerper LE, Ballatori N and Clarkson TW. (1995) Methylmercury-thiol uptake into cultured brain capillary endothelial cells on amino acid system L. J Pharmacol Exp Ther. 272, 1277-1284.

Molliver ME, O’Hearn E, Battaglia G and DeSouza EB. (1986) Direct intracerebral administration of MDA and MDMA does not produce serotonin neurotoxicity. Soc Neurosci Abst. 12, 1234.

Molliver ME, Berger UV, Mamounas LA, Molliver DC, O'Hearn E and Wilson MA. (1990) Neurotoxicity of MDMA and related compounds: anatomic studies. Ann N Y Acad Sci. 600, 649-661

Monks TJ and Lau SS. (1994) Glutathione conjugate mediated toxicities. In Handbook of Experimental Pharmacology, Vol. 112: Conjugation-Deconjugation Reactions in Drug Metabolism and Toxicity (Kauffman, F. C., Ed.) pp 459-509, Springer Verlag, New York.

Monks TJ and Lau SS. (1997) Biological reactivity of polyphenolic glutathione conjugates. Chem Res Toxicol. 10, 1296-1313.

Monks TJ and Lau SS. (1998)The pharmacology and toxicology of polyphenolic- glutathione conjugates. Annu Rev Pharmacol Toxicol. 38, 229-355.

208 Monks TJ and Jones DC. (2002) The metabolism and toxicity of quinones, quinonimines, quinone methides, and quinone-thioethers. Curr Drug Metab. 3, 425-438.

Montiel-Duarte C, Varela-Rey M, Oses-Prieto JA, Lopez-Zabalza MJ, Beitia G, Cenarruzabeitia E and Iraburu MJ. (2002) 3,4-Methylenedioxy-methamphetamine ("Ecstasy") induces apoptosis of cultured rat liver cells. Biochim Biophys Acta. 1588, 26-32.

Montiel-Duarte C, Ansorena E, Lopez-Zabalza MJ, Cenarruzabeitia E, and Iraburu MJ. (2004) Role of reactive oxygen species, glutathione and NF-kappaB in apoptosis induced by 3,4-methylenedioxymethamphetamine ("Ecstasy") on hepatic stellate cells. Biochem Pharmacol. 67, 1025-1033.

Montine TJ, Picklo MJ, Amarnath V, Whetsell WO and Graham DG. (1997) Neurotoxicity of endogenous cysteinylcatechols. Exp Neurol. 148, 26-33.

Montine TJ, Amarnath V, Picklo MJ, Sidell KR, Zhang J and Graham DG. (2000) Dopamine mercapturate can augment dopaminergic neurodegeneration. Drug Metab Rev. 32, 363-736.

Moos T and Morgan EH. (2000) Transferrin and transferrin receptor function in brain barrier systems. Cell Mol Neurobiol. 20, 77-95.

Moreno S, Nardacci R and Ceru MP. (1997) Regional and ultrastructural immunolocalization of copper-zinc superoxide dismutase in rat central nervous system. J Histochem Cytochem. 45, 1611-1622.

Mortensen OV, Kristensen AS, Rudnick G and Wiborg O. (1999) Molecular cloning, expression and characterization of a bovine serotonin transporter. Brain Res Mol Brain Res. 71, 120-126.

Mortensen OV, Kristensen AS and Wilborg O. (2001) Species-scanning mutagenesis of the SERT reveals residues essential in selective, high affinity recognition of antidepressants. J Neurochem. 79, 237-247.

Mortier KA, Maudens KE, Lambert WE, Clauwaert KM, Van Bocxlaer JF, Deforce DL, Van Peteghem CH and De Leenheer AP. (2002) Simultaneous, quantitative determination of opiates, amphetamines, cocaine and benzoylecgonine in oral fluid by liquid chromatography quadrupole-time-of-flight mass spectrometry. J Chromatogr B Anal Tech Biomed Life Sci. 779, 321-330.

209 Mulligan KA and Tork I. (1988) Serotonergic innervation of the cat cerebral cortex. J Comp Neurol. 270, 86-110.

Mutlib AE, Shockcor J, Espina R, Graciani N, Du A and Gan LS. (2000) Disposition of glutathione conjugates in rats by a novel glutamic acid pathway: characterization of unique peptide conjugates by liquid chromatography/mass spectrometry and liquid chromatography/NMR. J Pharmacol Exp Ther. 294, 735-745.

Naoi M and Maruyama W (1999) Cell death of dopamine neurons in aging and Parkinson’s disease. Mechanisms of Ageing and Devlop. 111, 175-188.

Nash JF and Brodkin J. (1991) Microdialysis studies on 3,4-Methylenedioxy- methamphetamine-indcued dopamine release: Effect of dopamine reuptake inhibitors. J. Phram. And Exp. Ther. 259, 820-825.

Nash JF and Nichols DE. (1991) Microdialysis studies on 3,4-methylenedioxy- amphetamine and structurally related analogues. Eur J Pharmacol. 200, 53-58.

Nash JF, Meltzer HY and Gudelsky GA. (1990) Effect of 3,4-methylenedioxy- methamphetamine on 3,4-dihydroxyphenylalanine accumulation in the striatum and nucleus accumbens. J Neurochem. 54, 1062-1070.

Nemoto T, Shimma N, Horie S, Saito T, Okuma Y, Nomura Y and Murayama T. (2003) Involvement of the system L amino acid transporter on uptake of S-nitroso-L- cysteine, an endogenous S-nitrosothiol, in PC12 cells. Eur J Pharmacol. 458, 17-24.

Nixdorf WL, Burrows KB, Gudelsky GA and Yamamoto BK. (2001) Enhancement of 3,4-methylenedioxymethamphetamine neurotoxicity by the energy inhibitor malonate. J Neurochem. 77, 647-654.

Nordgren HK and Beck O. (2003) Direct screening for MDMA and MDA by liquid chromatography-tandem mass spectrometry. J Anal Toxicol. 27, 15-19.

Nunez G and Clarke MF. (1994) The Bcl-2 family of proteins: regulators of cell death and survival. Trends Cell Biol. 4, 399-403.

O’Brien PJ. (1991) Molecular mechanisms of quinone cytotoxicity. Chem Biol Interact. 80, 1-41.

Oesterheld JR and Armstrong SC. (2004) Ecstasy: pharmacodynamic and pharmacokinetics interactions. Psychosomatics. 45, 84-87.

210 O'Hearn E, Battaglia G, De Souza EB, Kuhar MJ and Molliver ME. (1988) Methylenedioxyamphetamine (MDA) and methylenedioxymethamphetamine (MDMA) cause selective ablation of serotonergic axon terminals in forebrain: immunocytochemical evidence for neurotoxicity. Neurosci. 8, 2788-2803.

Okonkwo PO, Orlowski M and Green JP. (1974) Enzymes of the y-glutamyl cycle in the choroid plexus and brain. J.Neurochem. 22, 1053-1058.

Oldendorf WH. (1973) Carrier-mediated blood-brain barrier transport of short-chain monocarboxylic organic acids. Am J Physiol. 224, 1450-1463.

Oldendorf WH and Szabo J. (1976) Amino acid assignment to one of three blood-brain barrier amino acid carriers. Am J Physiol. 230, 94-108.

Orlowski M, and Meister A. (1970) The y-glutamyl cycle. A possible transport system for amino acids. Proc. Natl. Acad. Sci. 67, 1248-1255.

O'Shea E, Granados R, Esteban B, Colado MI and Green AR. (1998) The relationship between the degree of neurodegeneration of rat brain 5-HT nerve terminals and the dose and frequency of administration of MDMA ('ecstasy'). Neuropharmacology. 37, 919-26.

O'Shea E, Easton N, Fry JR, Green AR and Marsden CA. (2002) Protection against 3,4- methylenedioxymethamphetamine-induced neurodegeneration produced by glutathione depletion in rats is mediated by attenuation of hyperthermia. J Neurochemisrty. 81, 686-695.

Owens MJ and Nemeroff CB. (1994) Role of serotonin in the pathophysiology of depression: focus on the serotonin transporter. Clin Chem. 40, 288-295.

Pardrige WM. (1999) Blood-brain barrier biology and methodology. J Neurovirol. 5, 556-569.

Pardrige WM. (2003) Molecular biology of the blood-brain barrier. Methods Mol Med. 89, 385-399.

Paris JM and Cunningham KA. (1992) Lack of serotonin neurotoxicity after intraraphe microinjection of (+)-3,4-methylenedioxymethamphetamine (MDMA). Brain Res Bull. 28, 115-119.

211 Parrott AC. (2000) Human research on MDMA (3,4-methylenedioxy-methamphetamine) neurotoxicity: cognitive and behavioral indices of change. Neuropsychobiology. 42, 17-24.

Parrott AC. (2001) Human psychopharmacology of Ecstasy (MDMA): a review of 15 years of empirical research. Hum Psychopharmacol. 16, 557-577.

Parrott AC, Buchanan T, Scholey AB, Heffernan T, Ling J, and Rodgers J. (2002) Ecstasy/MDMA attributed problems reported by novice, moderate and heavy recreational users. Hum Psychopharmacol. 17, 309-132.

Patel NJ, Fullone JS and Anders MW. (1993) Brain uptake of S-(1,2- dichlorovinyl)glutathione and S-(1,2-dichlorovinyl)-L-cysteine, the glutathione and cysteine S-conjugates of the neurotoxin dichloroacetylene. Brain Res Mol Brain Res. 17, 53-58.

Patel NJ, Kumagai Y, Unger SE and Cho AK. (1991) Transformation of dopamine and alpha-methyldopamine. Chem Res Toxicol. 4, 421-426.

Pedrosa R and Soares-da-Silva P. (2002) Oxidative and non-oxidative mechanisms of neuronal cell death and apoptosis by L-3,4-dihydroxyphenylalanine (L-DOPA) and dopamine. Br J Pharmacol. 137, 1305-1313.

Peroutka SJ, Newman H, and Harris H. (1988) Subjective effects of 3,4- methylenedioxymethamphetamine in recreational users. Neuropsychopharm. 1, 273- 277.

Pfaller W, Gstraunthaler G, Kotanko P, Wolf H, and Curthoys NP. (1984) Immunocytochemical localization of y-glutamyltransferase on isolated renal cortical tubular fragments. Histochemistry. 80, 289-293.

Philbert MA, Beiswanger CM, Waters DK, Reuhl KR and Lowndes HE. (1991) Cellular and regional distribution of reduced glutathione in the nervous system of the rat: histochemical localization by mercury orange and o-phthaldialdehyde-induced histofluorescence. Toxicol Appl Pharmacol. 107, 215-27.

Philbert MA, Beiswanger CM, Manson MM, Green JA, Novak RF, Primiano T, Reuhl KR and Lowndes HE. (1995) Glutathione S-transferases and gamma-glutamyl transpeptidase in the rat nervous systems: a basis for differential susceptibility to neurotoxicants. Neurotoxicology. 16, 349-362.

212 Picklo MJ, Amarnath V, Graham DG, and Montine TJ. (1999) Endogenous catechol thioethers may be pro-oxidant or antioxidant. Free Radical Biol Med. 27, 271-277.

Puckett-Vaughn DL, Stenken JA, Scott DO, Lunte SM and Lunte CE. (1993) Enzymatic formation and electrochemical characterization of multiply substituted glutathione conjugates of hydroquinone. Life Sci. 52, 1239-1247.

Qiu JH, Asai A, Chi S, Saito N, Hamada H and Kirino T. (2000) Proteasome inhibitors induce cytochrome c-caspase 3-like protease-mediated apoptosis in cultured cortical neurons. J Neurosci. 20, 259-265.

Ramamoorthy S, Giovanetti E, Qian Y and Blakely RD. (1998) Phosphorylation and regulation of anti-depressant sensitive SERTs. J. Biol. Chem. 273, 2458-2466.

Ramamoorthy Y, Yu AM, Suh N, Haining RL, Tyndale RF and Sellers EM. (2002) Reduced (+/-)-3,4-methylenedioxymethamphetamine ("Ecstasy") metabolism with cytochrome P450 2D6 inhibitors and pharmacogenetic variants in vitro. Biochem Pharmacol. 63, 2111-2119.

Rattray M, Michael GJ, Lee J, Wotherspoon G, Bendotti C, Priestley JV. (1999) Intraregional variation in expression of serotonin transporter messenger RNA by 5- hydroxytryptamine neurons. Neuroscience. 88, 169-183.

Ricaurte G, Bryan G, Strauss L, Seiden L and Schuster C. (1985) Hallucinogenic amphetamine selectively destroys serotonin nerve terminals. Science. 229, 986-988.

Ricuarte GA, Delanney LE, Irwin I, and Langston J W (1988) Toxic effects of MDMA on central serotonergic neurons in primate: importance of route and frequency of drug administration. Brain Research. 446, 165-168.

Ricaurte GA and McCann UD. (1998) Neurotoxic amphetamine analogues: effects in monkeys and implications for humans. Ann N Y Acad Sci. 648, 371-382.

Ricaurte, GA, McCann, UD, Szarbo, Z and Scheffel, U. (2000a) Toxicodynamics and long-term toxicity of the recreational drug, 3,4-methylenedioxymethamphetamine (MDMA, ecstasy). Toxicol. Let. 112-113, 143-146

Ricaurte GA, Yaun J and McCann UD. (2000b) 3,4-Methylenedioxy-methamphetamine (MDMA, ecstasy)-induced 5-HT-neurotoxicity: studies in animals. Neuropsychobiology. 42, 5-10.

213 Richter C. (1997) Reactive oxygen and nitrogen species regulate mitochondrial Ca2+ homeostasis and respiration. Biosci Rep. 17, 53-66.

Richter C, Ghafourifar P, Schweizer M and Laffranchi R. (1997) Nitric oxide and mitochondrial Ca++. Biochemical Soc. Trans. 25, 914 – 918.

Rotondo A, Mazzanti C, Dell'Osso L, Rucci P, Sullivan P, Bouanani S, Gonnelli C, Goldman D and Cassano GB. (2002) Catechol o-methyltransferase, serotonin transporter, and tryptophan hydroxylase gene polymorphisms in bipolar disorder patients with and without comorbid panic disorder. Am J Psychiatry. 59, 23-29.

Sabol KE and Seiden LS. (1998) Reserpine attenuates D-amphetamine and MDMA- induced transmitter release in vivo: a consideration of dose, core temperature and dopamine synthesis. Brain Res. 806, 69-78.

Saldana SN and Barker EL. (2004) Temperature and 3,4-methylenedioxy- methamphetamine alter human serotonin transporter-mediated dopamine uptake. Neurosci Lett. 354, 209-212.

Sanchez V, Camarero J, Esteban B, Peter MJ, Green AR and Colado MI (2001) The mechanisms involved in the long-lasting neuroprotective effect of fluoxetine against MDMA ('ecstasy')-induced degeneration of 5-HT nerve endings in rat brain. Br J Pharmacol. 134, 46-57.

Sasaki T and Senda M. (1999) Evaluation of glutathione localization in brain using 99mTc meso-HMPAO. J Nucl Med. 40, 1056-1060.

Sastry PS and Rao KS. (2000) Apoptosis and the nervous system. J. Neurochem. 74, 1- 20.

Savolainen H. (1978) Superoxide dismutase and glutathione peroxidase activities in rat brain. Res Commun Chem Pathol Pharmacol. 21, 173-176.

Scheffel U, Szabo Z, Mathews WB, Finley PA, Dannals RF, Ravert HT, Szabo K, Yuan J and Ricaurte GA. (1998) In vivo detection of short- and long-term MDMA neurotoxicity--a positron emission tomography study in the living baboon brain. Synapse. 29, 183-192.

Schmidt CJ and Lovenberg W. (1985) In-vitro demonstration of dopamine uptake by neostriatal serotonergic neurons in the rat. Neurosci Letters. 59, 9-14.

214 Schmidt CJ. (1987) Neurotoxicity of the psychedelic amphetamine, methylenedioxy- methamphetamine. J Pharm Exp Ther. 240, 1-7.

Schmidt CJ and Taylor VL. (1988) Direct central effects of acute methylenedioxymethamphetamine on serotonergic neurons. Eur J Pharmacol. 156, 121-131.

Schmidt CJ, Abbate GM, Black CK and Taylor VL. (1990) Selective 5- hydroxytryptamine2 receptor antagonists protect against the neurotoxicity of methylenedioxymethamphetamine in rats. J Pharmacol Exp Ther. 255, 478-483.

Schmidt CJ, Black CK, Abbate GM and Taylor VL. (1990) Methylenedioxy- methamphetamine-induced hyperthermia and neurotoxicity are independently mediated by 5-HT2 receptors. Brain Res. 529, 85-90.

Schmidt CJ, Taylor VL, Abbate GM and Nieduzak TR. (1991a) 5-HT2 antagonists stereoselectively prevent the neurotoxicity of 3,4-methylenedioxymethamphetamine by blocking the acute stimulation of dopamine synthesis: reversal by L-dopa. J Pharmacol Exp Ther. 256, 230-235.

Schmidt CJ, Black CK and Taylor VL. (1991b) L-DOPA potentiation of the serotonergic deficits due to a single administration of 3,4-methylenedioxymethamphetamine, p- chloroamphetamine or methamphetamine to rats. Eur J Pharmacol. 203, 41–49.

Schmidt CJ, Black CK, Taylor VL, Fadayel GM, Humphreys TM, Nieduzak TR and Sorensen SM. (1992a) The 5-HT 2 receptor antagonist, MDL 28,133A, disrupts the serotonergic-dopaminergic interaction mediating the neurochemical effects of 3,4- methylenedioxymethamphetamine. Eur J Pharmacol. 220, 151–159.

Schmidt CJ, Fadayel GM, Sullivan CK and Taylor VL. (1992b) 5-HT 2 receptors exert a state-dependent regulation of dopaminergic function: studies with MDL 100,907 and the amphetamine analogue 3,4-methylenedioxy-methamphetamine. Eur J Pharmacol. 223, 65-74.

Schmidt CJ, Sullivan CK and Fadayel GM. (1994) Blockade of striatal 5-hydroxy- tryptamine 2 receptors reduces the increase in extracellular concentrations of dopamine produced by the amphetamine analogue 3,4-methylenedioxymethamphet- amine. J Neurochem. 62, 1382–1389.

Shankaran M, Yamamoto BK and Gudelsky GA. (1999) Involvement of the SERT in the formation of hydroxy radicals induced by 3,4-methylenedioxymethamphetamine. Eur J Pharmacol. 385, 103-110.

215 Shankaran M, Yamamoto BK and Gudelsky GA. (2001) Ascorbic acid prevents 3,4- methylenedioxymethamphetamine (MDMA)-induced hydroxy radical formation and the behavioral and neurochemical consequences of the depletion of brain 5-HT. Synapse. 40, 55-64.

Shen XM and Dryhurst G. (1996) Further insights into the influence of L-cysteine on the oxidation chemistry of dopamine: reaction pathways of potential relevance to Parkinson's disease. Chem Res Toxicol. 9, 751-763.

Shen XM, Zhang F, and Dryhurst G. (1997) Oxidation of dopamine in the presence of cysteine: characterization of new toxic products. Chem Res Toxicol. 10, 147-155.

Shen XM and Dryhurst G. (1998) Iron- and manganese-catalyzed autoxidation of dopamine in the presence of L-cysteine: possible insights into iron- and manganese- mediated dopaminergic neurotoxicity. Chem Res Toxicol. 11, 824-837.

Shulgin AT. (1990) History of MDMA. Ecstasy: the clinical, pharmacological, and neurotoxicological effects of the drug MDMA. Pp1-20. Kluwer Academic, Boston MA.

Simantov R and Tauber M. (1997) The abused drug MDMA (ecstasy) induces programmed death of human serotonergic cells. FASEB J. 11, 141-146.

Sprague JE and Nichols DE. (1995a) Inhibition of MAO-B protects against MDMA- induced neurotoxicity in the ST. Psychopharmacology. 118, 357-359.

Sprague JE and Nichols DE. (1995b) The monoamine oxidase-B inhibitor L-deprenyl protects against 3,4-Methylenedioxymethamphetamine-induced lipid peroxidation and long-term serotonergic deficits. J Pharmacol Exp Ther. 273, 667-673.

Sprague JE, Everman SL and Nichols DE. (1998) An integrated hypothesis for the serotonergic axonal loss induced by 3,4-methylenedioxymethamphetamine. Neurotoxicology. 19, 427-441.

Steele TD, Nichols DE and Yim GK. (1987) Sterochemical effects of 3,4- methylenedioxymethamphetamine (MDMA) and related amphetamine derivatives on inhibition of uptake of [3H]monoamines into synaptosomes for different regions of rat brain. Biochem Pharm. 36, 2297-2303.

Steele TD, McCann UD, and Ricautre GA. (1994) 3, 4-methylenediuoxy- methamphetamine (MDMA), “Ecstasy): pharmacology and toxicology in animals and humans. Addiction. 89, 539-551.

216 Stokes AH, Hastings TG and Vrana KE. (1999) Cytotoxic and genotoxic potential of dopamine. J Neurosci Res. 55, 659-665.

Stone DM, Hanson GR, and Gibb JW. (1986) The effects of 3,4- methylenedioxymethamphetamine on monoaminergic systems in rat brain. Eur. J. Pharmacol. 128, 41-48.

Stone DM, Johnson M, Hanson GR and Gibb JW. (1988) Role of endogenous dopamine in the central serotonergic deficits induced by 3,4-methylenedioxymethamphetamine. J Pharmacol Exp Ther. 247, 79–87.

Stone DM, Hanson GR, and Gibb JW. (1989a) In vitro reactivation of rat cortical tryptophan hydroxylase following in vivo inactivation by methylene- dioxymethamphetamine. J. Neurochem. 53, 572-581.

Stone DM, Johnson M, Hanson GR, and Gibb JW. (1989b) Acute inactivation of tryptophan hydroxylase by amphetamine analogs involves the oxidation of sulfhydryl sites. Eur J Pharmacol. 172, 93-97.

Stumm G, Schlegel J, Schafer T, Wurz C, Mennel HD, Krieg JC and Vedder H. (1999) Amphetamines induce apoptosis and regulation of bcl-x splice variants in neocortical neurons. FASEB J. 13, 1065-1072.

Sugiyama Y, Kusuhara H and Suzuki H. (1999) Kinetic and biochemical analysis of carrier-mediated efflux of drugs through the blood-brain and blood-cerebrospinal fluid barriers: importance in the drug delivery to the brain. J Control Release. 62, 179-186.

Susin SA, Lorenzo HK, Zamzami N, Marzo I, Brenner C and Larochette N. (1999) Mitochondrial release of caspase-2 and –9 during the apoptotic process. J. Exp. Med. 189, 381-394.

Taffe MA, Weed MR, Davis S, Huitron-Resendiz S, Schroeder R, Parsons LH, Henriksen SJ and Gold LH. (2001) Functional consequences of repeated 3,4- methylenedioxymethamphetamine (MDMA) treatment in Rhesus monkeys. Neuropsychopharm. 24, 230-239.

Takai N, Nakanishi H, Tanabe K, Nishioku T, Sugiyama T, Fujiwara M and Yamamoto K. (1998) Involvement of caspase-like proteinases in apoptosis of neuronal PC12 cells and primary cultured microglia induced by 6-OHDA. J Neurosci Res. 54, 214- 222.

217 Tamai I and Tsuji A. (2000) Transporter-mediated permeation of drugs across the blood- brain barrier. J Pharm Sci. 89, 1371-1388.

Tatton WG and Olanow CW. (1999) Apoptosis in neurodegenerative diseases: the role of the mitochondria. Biochimica et Biophysica Acta. 1410, 195-213.

Thiessen PN and Cook DA. (1973) The properties of 3, 4-methylenediuoxyamphetamine (MDA). A review of literature. Clin Toxicol. 6, 45-52.

Thompson MR, Li KM, Clemens KJ, Gurtman CG, Hunt GE, Cornish JL, and McGregor IS. (2004) Chronic fluoxetine treatment partly attenuates the long-term anxiety and depressive symptoms induced by MDMA ('Ecstasy') in rats. Neuropsychopharmacology. 29, 694-704.

Thornberry N and Lazebnik Y. (1999) Caspases: Enemies within. Science. 281, 1312- 1316.

Thress K, Kornbluth S and Smith JJ. (1999) Mitochondria at the crossroad of apoptotic death. J Bioeng and Biomembranes. 31, 321-326.

Tilson HA. (1996) Evolution and current status of neurotoxicity risk assessment. Drug Metab Rev. 28, 121-139.

Tork I. (1990) Anatomy of the serotonergic system. Ann N Y Acad Sci. 600, 9-34.

Tsuji A and Tamai I. (1999) Carrier-mediated or specialized transport of drugs across the blood-brain barrier. Adv Drug Deliv Rev. 36, 277-290.

Twarog BM and Page IH. (1953) Serotonin content of some mammalian tissues and urine and a method for its determination. Am J Physiol. 175, 157-161.

Uehara T, Kikuchi Y, and Nomura Y. (1999) Caspase activation accompanying cytochrome c release from mitochondria is possibly involved in nitric oxide-induced neuronal apoptosis in SH-SY5Y cells. J of Neurochemistry. 72, 196-205.

Urban M, Gilch G, Schepers G, van Miert E and Scherer G. (2003) Determination of the major mercapturic acids of 1,3-butadiene in human and rat urine using liquid chromatography with tandem mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci. 796, 131-140.

218 Verkes RJ, Gijsman HJ, Pieters MS, Schoemaker RC, de Visser S, Kuijpers M, Pennings EJ, de Bruin D, Van de Wijngaart G, Van Gerven JM, and Cohen AF. (2001) Cognitive performance and serotonergic function in users of ecstasy. Psychopharmacology (Berl). 153, 196-202.

Virág L, Scott GS, Cuzzocrea S, Marmer D, Salzman AL and Szabó C. (1998) Peroxynitrite-induced thymocyte apoptosis: the role of caspases and poly (ADP- ribose) synthase (PARS) activation. Immunology. 94, 345-355.

Volk B, Hettmannsperger U, Papp T, Amelizad Z, Oesch F, and Knoth R. (1991) Mapping of phenytoin-inducible cytochromeP450 immunoreactivity in the mouse central nervous system. Neurosci. 42, 215-135.

Wade LA and Brady HM. (1981) Cysteine and cystine transport at the blood-brain barrier. J Neurochem. 37, 730-740.

Walker TM, Davenport-Jones JE, Fox RM and Atterwill CK. (1999) The neurotoxic effects of methylenedioxymethamphetamine (MDMA) and its metabolites on rat brain spheroids in culture. Cell Bio. Toxicol. 15, 137-142.

Wefers H and Sies H. (1983) Hepatic low-level chemiluminescence during redox cycling of menadione and the menadione-glutathione conjugate: relation to glutathione and NAD(P)H:quinone reductase (DT-diaphorase) activity. Arch Biochem Biophys. 224, 568-578.

Wilson MA, Ricaurte GA, and Molliver ME. (1989) Distinct morphologic classes of serotonergic axons in primates exhibit differential vulnerability to the psychotropic drug 3,4-methylenedioxymethamphetamine. Neuroscience. 28, 121-137.

Wong KS, Goyal RN, Wrona MZ, Blank CL, and Dryhurst G. (1993) 7-S-glutathionyl- tryptamine-4,5-dione: a possible aberrant metabolite of serotonin. Biochem Pharmacol. 46, 1637-1652.

Wood M, De Boeck G, Samyn N, Morris M, Cooper DP, Maes RA and De Bruijn EA. (2003) Development of a rapid and sensitive method for the quantitation of amphetamines in human plasma and oral fluid by LC-MS-MS. J Anal Toxicol. 27, 78- 87.

Yuan J, Cord BJ, McCann UD, Callahan BT and Ricaurte GA. (2002) Effect of depleting vesicular and cytoplasmic dopamine on methylenedioxy-methamphetamine neurotoxicity. J Neurochem. 80, 960-669.

219 Zeevalk GD, Derr-Yellin E and Nicklas WJ. (1995) Relative vulnerability of dopamine and GABA neurons in mesencephalic culture to inhibition of succinate dehydrogenase by malonate and 3-nitropropionic acid and protection by NMDA receptor blockade. J Pharmacol Exp Ther. 275, 1124-1130.

Zhang F and Dryhurst G. (1994) Effects of L-cysteine on the oxidation chemistry of dopamine: new reaction pathways of potential relevance to idiopathic Parkinson's disease. J Med Chem. 37, 1084-1098.

Zhang J, Kravtsov V, Amarnath V, Picklo MJ, Graham DG, and Montine TJ. (2000) Enhancement of dopaminergic neurotoxicity by the mercapturate of dopamine: relevance to Parkinson’s disease. J Neurochem 74, 970-978.

Zheng W, Aschner M and Ghersi-Egea JF. (2003) Brain barrier systems: a new frontier in metal neurotoxicological research. Toxicol Appl Pharmacol. 192, 1-11.

Zhao Z, Castagnoli N Jr, Ricaurte GA, Steele T and Martello M. (1992) Synthesis and neurotoxicological evaluation of putative metabolites of the serotonergic neurotoxin 2-(methylamino)-1-[3,4-methylenedioxy)phenyl]propane-[(methylenedioxy)- methamphetamine]. Chem Res Toxicol 5, 89-94.

Zhou JF, Zhou YH, Zhang L, Chen HH and Cai D. (2003a) 3,4- methylenedioxymethamphetamine (MDMA) abuse markedly inhibits acetylcholinesterase activity and induces severe oxidative damage and liperoxidative damage. Biomed Environ Sci. 16, 53-61.

Zhou JF, Chen P, Zhou YH, Zhang L and Chen HH. (2003b) 3,4- Methylenedioxymethamphetamine (MDMA) abuse may cause oxidative stress and potential free radical damage. Free Radic Res. 37, 491-497.

Zimatkin SM and Lindros KO. (1996) Distribution of catalase in rat brain: aminergic neurons as possible targets for ethanol effects. Alcohol Alcohol. 31, 167-174.

Zlokovic BV, Mackic JB, McComb JG, Weiss MH, Kaplowitz N and Kannan R. (1994) Evidence for transcapillary transport of reduced glutathione in vascular perfused guinea-pig brain. Biochem Biophys Res Commun. 201, 402-408.

220

VITA

Douglas Campbell Jones was born in Richland Center, WI on March 25, 1970, the son of Keith and Mary Ellen Jones. After graduating from Washington High School in 1988, he attended Kansas University, earning a Bachelor of Arts in Psychology in 1992. From 1993-1997, he attended The University of Colorado as a continuing education student and was employed as a Research laboratory technician for the university. In 1997, he entered the graduate program in Medicinal Chemistry and Molecular Pharmacology at Purdue University. After earning a Master of Science degree in 2000, he entered the Pharmacology and Toxicology Department at the University of Texas at Austin and joined the laboratory of Dr. Terrence Monks. During his graduate career, Doug was awarded the Burton Fellowship and served as the graduate student representative for the Gulf Coast Chapter of the Society of Toxicology from 2002-2003.

Permanent address: 10712 Sierra Oaks, Austin Texas, 78759

This dissertation was typed by Douglas C. Jones.

221