The Role of Prostaglandin H Synthase (Phs)

THE ROLE OF PROSTAGLANDIN H SYNTHASE (PHS) BIOACTIVATION AND NUCLEAR FACTOR ERYTHROID 2-RELATED FACTOR 2 (NRF2)-MEDIATED PROTECTION IN ENDOGENOUS AND METHAMPHETAMINE-INITIATED NEUROTOXICITY

Annmarie Ramkissoon

A thesis submitted in the conformity with the requirements for the Degree of Doctor of Philosophy Graduate Department of Pharmaceutical Sciences University of Toronto

THE ROLE OF PROSTAGLANDIN H SYNTHASE (PHS) BIOACTIVATION AND NUCLEAR FACTOR ERYTHROID 2-RELATED FACTOR 2 (NRF2)-MEDIATED PROTECTION IN ENDOGENOUS AND METHAMPHETAMINE-INITIATED NEUROTOXICITY

Doctor of Philosophy, 2011 Annmarie Ramkissoon Graduate Department of Pharmaceutical Sciences University of Toronto

ABSTRACT

Endogenous brain compounds and xenobiotics, including the neurotoxins such as the amphetamine analogs 3,4-methylenedioxymethamphetamine (MDMA, Ecstasy), methamphetamine (METH, Speed) and methylenedioxyamphetamine (MDA, active metabolite of MDMA), may be bioactivated by prostaglandin H synthase (PHS) to free radicals that generate reactive oxygen species (ROS). In the absence of adequate antioxidant or repair mechanisms, ROS oxidize macromolecules such as DNA, protein and lipids, which can lead to toxicity. In vitro, we evaluated bioactivation using both purified ovine PHS-1 and cultured cells stably overexpressing either human PHS-1 or hPHS-2 isozymes. We found the neurotransmitter dopamine, its precursors and some metabolites, as well as METH and MDA, can be bioactivated by ovine and/or human PHS in an isozyme-dependent fashion that generates ROS, which oxidize DNA and protein and increase toxicity. This process is blocked by both the PHS inhibitor acetylsalicylic acid (ASA) and the ROS detoxifying enzyme catalase. Our data are the first to reveal isozyme-dependent bioactivation by PHS as a potential mechanism for enhanced susceptibility to both exogenous and endogenous neurotoxins, the latter of which may be particularly important in aging. METH-initiated ROS can also activate redox-sensitive transcription factors such as nuclear factor erythroid 2-related factor 2 (Nrf2), which is involved in the induction of an array of protective mechanisms in both adult and fetal brain. Using Nrf2

ii knockout mice, we showed Nrf2 has a novel neuroprotective role in METH-initiated oxidative stress, neurotoxicity and functional deficits in both fetal development and adulthood, especially with multiple exposures allowing time for the induction of neuroprotective mechanisms. Our studies are the first to show that Nrf2 afforded protection against both motor coordination deficits and olfactory deficits caused by METH in utero and in adults, suggesting that deficiencies in Nrf2 activation constitute a risk factor for ROS-mediated neurotoxicity in the fetus and adult.

iii

ACKNOWLEDGEMENTS

I would like to thank Dr. Wells for the opportunity to work in his lab and for all the life lessons he has meticulously taught me over the years. I would also like to thank my advisory committee members, Drs. Catherine Bergeron, Stephen Kish, Jeffrey Henderson and Jack

Uetrecht for their constructive criticism and for always challenging me to do more.

Special thanks to Dr. Winnie Jeng for being a great mentor and who provided me a starting point for my thesis on PHS and encouraged me to think creatively and pursue other interests. Thanks to Dr. Luisa Goncalves for taking on the EPR studies while balancing her life in a new country with a new baby and to Dr. Tom Preston for all his advice about cell culture and encouraging me to work diligently. Thanks to Lily Morikawa from the Pathology Core,

Toronto Centre for Phenogenomics, for her help with immunohistochemistry, the Uetrecht lab for use of their RT-PCR machine and Xiaochu Zhang for all the advice. Thanks to former undergraduate students Ada Ho and James Poon for being enthusiastic helpers. Also, thanks to all the members of the Wells lab and Henderson lab and special thanks to Crystal Lee and

Kelvin Hui for being great role models. Their work ethic and integrity were very inspiring to me and helped me through the difficult times.

I would also like to thank my friends, especially CT, HC, MW, TA, YZ, AS and MJ who provided me a variety of resources for entertainment. Last, but not least, I would like to thank my mom, dad and especially my brother whose Xbox and PlayStation kept me mentally sane throughout the years.

TABLE OF CONTENTS

CHAPTER 1: INTRODUCTION ...... 1

1.1 RATIONALE, HYPOTHESIS AND OBJECTIVES ...... 2 1.1.1 RATIONALE ...... 2 1.1.2 HYPOTHESIS AND OBJECTIVES ...... 9

1.2 BRIEF OVERVIEW ...... 12

1.2.1 ROS IN THE BRAIN ...... 12 1.2.1.1 Introduction to Reactive Oxygen Species ...... 12 1.2.1.2 Sources of Reactive Oxygen Species ...... 15 1.2.1.2.1 Mitochondria ...... 15 1.2.1.2.2 Enzymes ...... 17 1.2.1.2.3 Excitotoxicity ...... 22 1.2.1.2.4 Immune Response-Microglia ...... 23 1.2.1.3 Neuroprotective Mechanisms ...... 24 1.2.1.3.1 Blood Brain Barrier ...... 24 1.2.1.3.2 Antioxidant Mechanisms ...... 25

1.3 PROSTAGLANDIN H SYNTHASES (PHSs) ...... 33

1.3.1 ROLE OF PROSTAGLANDIN SYNTHESIS AND THEIR RECEPTORS ...... 33

1.3.2 GENETICS OF PHS ...... 38 1.3.2.1 Genes ...... 38 1.3.2.2 Transcriptional regulation ...... 40 1.3.2.3 Post-transcriptional regulation ...... 45

1.3.3 PRIMARY PROTEIN STRUCTURES OF PHSs ...... 46

1.3.4 PHS ENZYMOLOGY ...... 49

1.3.5 INHIBITION OF PHSs ...... 53

1.3.6 CELLULAR LOCALIZATION AND CENTRAL NERVOUS SYSTEM (CNS) EXPRESSION OF PHSs ...... 57

1.3.7 PHS IN ROS GENERATION ...... 59

1.3.8 PHS IN NEURODEGENERATIVE DISEASES ...... 64

1.4 NUCLEAR FACTOR ERYTHROID 2-RELATED FACTORS (NRFs) ...... 71 v

1.4.1 OVERVIEW OF NRF1, NRF2 AND NRF3 ...... 71

1.4.2 MECHANISM OF ACTION OF NRF2 ...... 72

1.4.3 GENETICS OF NRF2 ...... 75

1.4.4 PROTEIN STRUCTURE OF NRF2 ...... 76

1.4.5 REGULATORS OF NRF2 ...... 80 1.4.5.1 Negative regulation by Kelch-like ECH-associated protein 1 (Keap1) ... 80 1.4.5.2 Negative regulation by proteasome degradation ...... 83 1.4.5.3 Regulation of transcriptional complex in nucleus ...... 84

1.4.6 ANTIOXIDANT RESPONSE ELEMENT (ARE) ...... 87

1.4.7 ACTIVATORS OF NRF2 ...... 91

1.4.8 NRF2 IN NEUROTOXICITY AND CNS DISEASES ...... 95 1.4.8.1 Nrf2 expression ...... 95 1.4.8.2 Nrf2 in neurodegenerative diseases ...... 96 1.4.8.3 Nrf2 in chemical-induced neurotoxicities ...... 100

1.4.9 NRF KNOCKOUT MOUSE MODELS ...... 102

1.4.10 EVIDENCE FOR POLYMORPHISMS IN THE KEAP1- NRF2-ARE PATHWAY ...... 104

1.5 METHAMPHETAMINE ...... 107

1.5.1 HISTORY AND USES ...... 107

1.5.2 PHARMACOKINETICS ...... 110

1.5.3 DISTRIBUTION ...... 112

1.5.4 METABOLISM BY CYTOCHROMES P450 (CYPs) and ELIMINATION ...... 115

1.5.5 RECEPTOR-MEDIATED PHARMACOLOGICAL ACTIONS OF METH ...... 119

1.5.6 EFFECTS OF METH ABUSE ...... 126

1.5.7 EVIDENCE FROM ANIMAL AND HUMAN STUDIES FOR NEUROTOXICITY ...... 130 vi

CHAPTER 2: STUDIES ...... 136

2.1 STUDY 1: PROSTAGLANDIN H SYNTHASE-1-CATALYZED BIOACTIVATION OF NEUROTRANSMITTERS, THEIR PRECURSORS, AND METABOLITES: OXIDATIVE DNA DAMAGE AND ELECTRON SPIN RESONANCE SPECTROSCOPY STUDIES ...... 137

2.1.1 ABSTRACT ...... 138 2.1.2 INTRODUCTION ...... 139 2.1.3 MATERIALS AND METHODS ...... 143 2.1.4 RESULTS ...... 149 2.1.5 DISCUSSION ...... 166

2.2 STUDY 2: HUMAN PROSTAGLANDIN H SYNTHASE (hPHS)-1- AND hPHS- 2-DEPENDENT BIOACTIVATION, OXIDATIVE MACROMOLECULAR DAMAGE AND CYTOTOXICITY OF DOPAMINE, ITS PRECURSOR AND METABOLITES ...... 174

2.2.1 ABSTRACT ...... 175 2.2.2 INTRODUCTION ...... 176 2.2.3 MATERIALS AND METHODS ...... 179 2.2.4 RESULTS ...... 185 2.2.5 DISCUSSION ...... 201

2.3 STUDY 3: HUMAN PROSTAGLANDIN H SYNTHASE (hPHS)-1 AND hPHS-2 IN AMPHETAMINE ANALOG BIOACTIVATION, DNA OXIDATION AND CYTOTOXICITY ...... 207

2.3.1 ABSTRACT ...... 208 2.3.2 INTRODUCTION ...... 209 2.3.3 MATERIALS AND METHODS ...... 212 2.3.4 RESULTS ...... 218 2.3.5 DISCUSSION ...... 227

2.4 STUDY 4: METHAMPHETAMINE OXIDATIVE STRESS, NEUROTOXICITY AND FUNCTIONAL DEFICITS ARE MODULATED BY NRF2 ...... 233

2.4.1 ABSTRACT ...... 234 2.4.2 INTRODUCTION ...... 235 2.4.3 MATERIALS AND METHODS ...... 238 2.4.4 RESULTS ...... 249 2.4.5 DISCUSSION ...... 266

2.5 STUDY 5: DEVELOPMENTAL ROLE OF NUCLEAR FACTOR-E2-RELATED FACTOR 2 (NRF2) IN PROTECTION AGAINST METHAMPHETAMINE vii

FETAL TOXICITY AND POSTNATAL FUNCTIONAL DEFICITS IN NRF2-DEFICIENT MICE ...... 272

2.5.1 ABSTRACT ...... 273 2.5.2 INTRODUCTION ...... 274 2.5.3 MATERIALS AND METHODS ...... 277 2.5.4 RESULTS ...... 284 2.5.5 DISCUSSION ...... 303

CHAPTER 3: SUMMARY, CONCLUSIONS AND FUTURE STUDIES ...... 309

3.1 SUMMARY AND CONCLUSIONS ...... 310

3.2 FINAL THOUGHTS ...... 325

3.3 FUTURE STUDIES ...... 326

CHAPTER 4: REFERENCES ...... 327

CHAPTER 5: APPENDICES ...... 399

I. SUPPLEMENTAL DATA ...... 400

II. COPYRIGHT ACKNOWLEDGEMENTS ...... 406

viii

LIST OF TABLES

SECTION 1: INTRODUCTION TABLE 1. IMPORTANT REACTIVE SPECIES ...... 13 TABLE 2. PROSTANOID RECEPTOR SUBTYPES, EXPRESSION AND FUNCTIONS ...... 37 TABLE 3. TRANSCRIPTIONAL ACTIVATION OF PHS-1 ...... 42 TABLE 4. TRANSCRIPTIONAL ACTIVATION OF PHS-2 ...... 43 TABLE 5. COMPARISON OF DOMAIN REGIONS OF PHS-1 AND 2 ...... 48 TABLE 6. SUMMARY OF PHS-MEDIATED EFFECTS IN NEURODEGENERATIVE DISEASES ...... 67 TABLE 7. NRF DOMAINS AND FUNCTION ...... 79 TABLE 8. REGULATORY MECHANISMS PROVIDED BY KEAP1 ...... 82 TABLE 9. NRF2 INTERACTIONS IN THE NUCLEUS ...... 85 TABLE 10. EXAMPLES OF GENES CONTAINING ARES IN THEIR PROMOTER REGION ...... 89 TABLE 11. SELECTIVE EXAMPLES OF NRF2 ACTIVATORS ...... 94 TABLE 12. EVIDENCE FOR NRF2 ACTIVATION OR DEREGULATION IN NEURODEGENERATIVE DISEASES ...... 98 TABLE 13. AMPHETAMINE ANALOG BINDING AFFINITIES TO UPTAKE TRANSPORTERS ...... 123 TABLE 14. METH RECEPTOR-MEDIATED EFFECTS ...... 125 TABLE 15. EFFECTS OF METH ...... 129

SECTION 2: STUDIES Study 1 TABLE 1. ESR DATA FOR ZN2+-COMPLEXED O-SEMIQUINONES OBTAINED FROM THE AUTOOXIDATION OF NEUROTRANSMITTERS, THEIR PRECURSORS, AND METABOLITES ...... 155 TABLE 2. HYPERFINE SPLITTING CONSTANTS OBTAINED BY EPR-SPIN TRAPPING EXPERIMENTS FOR HRP-CATALYZED OXIDATION OF

NEUROTRANSMITTERS, THEIR PRECURSORS AND METABOLITES ...... 156 TABLE 3. HYPERFINE SPLITTING CONSTANTS OBTAINED BY EPR-SPIN TRAPPING EXPERIMENTS FOR PHS-CATALYZED OXIDATION OF NEUROTRANSMITTERS, THEIR PRECURSORS, AND METABOLITES ...... 157 TABLE 4. UV/VIS SPECTRAL CHARACTERIZATION OF PRODUCTS OBTAINED UPON HRP- AND PHS-1-MEDIATED OXIDATION OF NEUROTRANSMITTERS, THEIR PRECURSORS, AND METABOLITES ...... 158

LIST OF FIGURES SECTION 1: INTRODUCTION Figure 1. Postulated bioactivation of endogenous substrates to a free radical intermediate by PHS which generates ROS to activate Nrf2 ...... 4 Figure 2. Enzymatic pathways involved in reactive intermediate-mediated neurotoxicity .. 14 Figure 3. Sources of reactive oxygen species (ROS) in the brain ...... 16 Figure 4. Cells of the central nervous system (CNS) and their function ...... 26 Figure 5. Biosynthesis of prostaglandins ...... 36 Figure 6. Cell-dependent PHS-2-activation ...... 44 Figure 7. Cyclooxygenase and peroxidase catalysis by PHSs ...... 52 Figure 8. Examples of PHS inhibitors ...... 56 Figure 9. Mechanisms of PHS bioactivation of substrates ...... 61 Figure 10. Oxidation of aminochrome by NADPH-cytochrome P450 reductase and NQO1 ...... 63 Figure 11. Mechanism of Nrf2 in Cytoprotection ...... 74 Figure 12. Amphetamine, its analogs and neurotransmitters, their precursors and metabolites ...... 108 Figure 13. Metabolism of methamphetamine by cytochromes P450 ...... 116 Figure 14. MDMA metabolism by CYPs and P450 reductase ...... 118 Figure 15. METH actions at the dopaminergic nerve terminal ...... 120

SECTION 2: STUDIES Study 1 Figure 1. ESR spectra of Zn2+-complexed o-semiquinones generated during HRP-catalyzed oxidation ...... 159 Figure 2. Relative radical production for Zn2+-complexed semiquinones obtained by autoxidation and by HRP-mediated oxidation of DA, its precursor, and metabolites and of epinephrine and its metabolites ...... 160 Figure 3. X-band ESR spectra of radical species trapped with DMPO ...... 161 Figure 4. X-band ESR spectra of radical species derived from epinephrine trapped with PBN ...... 162 xi

Figure 5. UV/vis spectra of the products obtained by HRP-mediated oxidation of neurotransmitters, their precursors, and metabolites ...... 163 Figure 6. UV/vis spectra of the products obtained by PHS-1-mediated oxidation of neurotransmitters, their precursors, and metabolites ...... 164 Figure 7. Ovine PHS-1-dependent DNA oxidation by neurotransmitters, their precursors, and metabolites ...... 165 Study 2 Figure 1. PGE2 activity in Chinese Hamster Ovary (CHO) cells and cells stably expressing human prostaglandin synthase-1 (hPHS-1) or hPHS-2 ...... 190 Figure 2. Dopamine (DA), its precursor and metabolites in hPHS-1- or hPHS-2-catalyzed bioactivation and cytotoxicity ...... 191 Figure 3. ASA protection in hPHS-catalyzed bioactivation and cytotoxicity of dopamine (DA), its precursor and metabolite ...... 192 Figure 4. Polyethylene-conjugated (PEG)-catalase protects against hPHS-1/2-catalyzed substrate bioactivation and ROS-mediated cytotoxicity...... 193 Figure 5. Protein oxidation caused by hPHS-catalyzed bioactivation of dopamine (DA), its precursor and metabolites ...... 195 Figure 6. DNA oxidation caused by hPHS-catalyzed bioactivation of dopamine (DA), its precursor and metabolites ...... 197 Figure 7. DNA oxidation caused by hPHS-catalyzed bioactivation of dopamine (DA), its precursor and metabolites ...... 199 Study 3 Figure 1. PHS activity in CHO-K1 cells stably expressing human prostaglandin H synthase-1 (hPHS-1) or hPHS-2 ...... 222 Figure 2. METH and MDA in hPHS-1- or hPHS-2-catalyzed bioactivation and cytotoxicity ...... 223 Figure 3. ASA protection in hPHS-catalyzed bioactivation and cytotoxicity of METH ...... 224 Figure 4. ASA protection in hPHS-catalyzed bioactivation and cytotoxicity of 100 uM METH or MDA ...... 225 Figure 5. DNA oxidation caused by hPHS-catalyzed bioactivation of METH and MDA, and protection by ASA ...... 226 xii

Study 4 Figure 1. Effect of methamphetamine (METH) on Nrf2-mediated gene expression in Nrf2 wild-type and knockout mice ...... 256 Figure 2. DNA oxidation in METH-treated Nrf2 wild-type and knockout mice ...... 257 Figure 3. Tyrosine hydroxylase (TH), dopamine transporter (DAT) and glial fibrillary acidic protein (GFAP) immunoblots and densitometric analysis of the striatum in METH-treated Nrf2 wild-type and KO mice ...... 258 Figure 4. Tyrosine hydroxylase (TH) and glial fibrillary acidic protein (GFAP) immunoblots and densitometric analysis of the olfactory bulbs in METH-treated Nrf2 wild-type and KO mice ...... 260 Figure 5. Nissl, DAT and GAD-65 analysis of olfactory bulb in saline and METH-treated Nrf2 wild-type and KO mice ...... 261 Figure 6. METH-initiated functional deficits in motor coordination and activity in Nrf2 wild-type and KO mice ...... 262 Figure 7. Olfactory detection, habituation and discrimination in METH-treated Nrf2 wild-type and KO mice ...... 263 Figure 8. Olfactory discrimination in METH-treated mice 4 wk after dosing ...... 264 Figure 9. Quantification of hyperthermia, methamphetamine (METH) and amphetamine (AMPH) in METH-treated Nrf2 wild-type and KO mice ...... 265 Study 5 Figure 1. Effect of in utero methamphetamine (METH) exposure on live pups and Nrf2 genotype distribution at weaning ...... 291 Figure 2. Impaired development in in utero METH-exposed fetuses from Nrf2 heterozygous dams...... 293 Figure 3. Effect of METH on Nrf2-mediated gene expression in Nrf2 (+/+) and (-/-) fetal brain ...... 295 Figure 4. DNA oxidation is increased in fetal brains after in utero METH exposure ...... 296 Figure 5. In utero exposure to METH on GD 17 results in olfactory deficits in offspring of Nrf2 (+/-) dams ...... 297 Figure 6. 2-day dosing with METH on GD 16 and GD 17 leads to Nrf2 and gender-dependent postnatal olfactory deficits ...... 298 xiii

Figure 7. Enhanced postnatal motor activity deficits in offspring of Nrf2 (+/-) dams exposed to METH ...... 300 Figure 8. Nissl, DAT and GAD-65 analysis of the olfactory bulbs of male offspring exposed in utero to METH ...... 301

xiv

LIST OF ABBREVIATIONS

2-dG 2-deoxyguanosine

3-MT 3-methoxytyramine

3-NP 3-nitropropionic acid

5-HETE 5-hydroxyeicosatetraenoic acid

5-HPETE 5-hydroperoxyeicosatetraenoic acid

5HTR Serotonin receptor

5-LOX 5-Lipoxygenase

6-OHDA 6-hydroxydopamine

8-oxo-dG 8-oxo-2’-deoxyguanosine

15-PGDH 15-hydroxyprostaglandin dehydrogenase

α Alpha adrenergic receptor

β Beta adrenergic receptor

AA Arachidonic acid

Aβ Beta-amyloid

ABC Adenosine-5'-triphosphate-binding cassette

AD Alzheimer's disease

ADHD Attention deficit-hyperactivity disorder

AHR Aryl hydrocarbon receptor

ALDH Aldehyde dehydrogenase

ALS Amyotrophic lateral sclerosis

AMPA α-amino-3-hydroxy-5-methylisoxazole-4-propionic acid

AP-2 Activator protein 2

ARE Antioxidant response element

ASA Acetylsalicylic acid

ATF Activating transcription factor

ATM Ataxia telangiectasia mutated

ATP Adenosine-5'-triphosphate

Bach1 Breakpoint cluster region/Abelson murine leukemia viral oncogene

homolog

BBB Blood-brain-barrier

B-NF Beta-naphthoflavone bp Basepair

BRG1 Brahma-related gene 1

BTB Broad complex, Tramtrack, Bric-a-brac bZip Basic leucine zipper

Ca2+ Calcium cAMP Cyclic adenosine monophosphate

CBP CREB-binding protein

C/EBP CAAT enhancer binding protein cGMP Cyclic guanosine-3′,5′-monophosphate

CHO Chinese hamster ovary

CNC Cap’n’collar

CNS Central nervous system

COMT Catechol-o-methyl-transferase

COX Cyclooxygenase

xvi cPLA2 Cytosolic phospholipase A2

CRE Cyclic AMP response elements

CREB cAMP responsive element binding protein

CRTH2 Chemoattractant receptor-homologous, expressed on T helper-2 cells

CSB Cockayne syndrome B

CYPs Cytochrome P450s

D3T 1,2-dithiole-3-thione

DA Dopamine

DAG Diacylglycerol

DAT Dopamine transporter

DETAPAC Diethyllenetriamine-pentacetic acid

DHMA Dihydroxymandelic acid

DGR Double glycine repeats

DJ-1 or PARK7 Parkinson disease (autosomal recessive, early onset) 7)

DMF Dimethyl fumarate

DMPO 5,5-dimethyl-1-pyrroline 1-oxide

DMSO Dimethyl sulfoxide

DNA Deoxyribonucleic acid

DNPH 2,4-dinitrophenylhydrazine dNTPs Dexoynucleotide triphosphates

DOPAC Dihydroxyphenylacetic acid

DTT Dithiothreitol

DUOX Dual oxidases

xvii

ECH Erythroid cell-derived protein with CNC homolog

EGF Epidermal growth factor eNOS Endothelial NOS

EP PGE receptor

EPL External plexiform layer

EpRE Electrophile response element

ER Estrogen receptor

ERK Extracellular signal-regulated kinases

ESR Electron spin ressonance spectroscopy

ETS E-twenty six

ETYA 5,8,11,14-eicosatetraynoic acid

Fe2+ Ferrous iron

Fe3+ Ferric iron

Fe3+ PPIX Ferric iron protoporphyrin IX (heme)

FP PGF receptor

G6PD Glucose-6-phosphate dehydrogenase

GABA Gamma-aminobutyric acid

GAD Glutamic acid decarboxylase

GCL Granule cell layer

GCLg Glutamate cysteine ligase

GCS Gamma-glutamylcysteine synthetase regulatory subunit

GD Gestational day

GFAP Glial fibrillary acidic protein

xviii

GGCS Gamma-glutamylcysteine synthetase

GnRH Gonadotrophin-releasing hormone

GPCR G protein-coupled receptor

GPx Glutathione peroxidase

GSH Glutathione

GSSG Glutathione disulfide

GSTs Glutathione S-transferases

GT Guanine-thymine

H+ Hydrogen

H2O2 Hydrogen peroxide hfcc hyperfine coupling constant

HHA 3,4-dihydroxyamphetamine

HHMA 3,4-dihydroxymethamphetamine

HMA 4-hydroxy-3-methoxyamphetamine

HMMA 4-hydroxy-3-methoxymethamphetamine

HO Heme oxygenase

HO. Hydroxyl radical hox Homeobox hPHS Human PHS

HPLC High performance liquid chromatography

HRP Horseradish peroxidase

HVA Homovanillic acid

IFN-γ Interferon-gamma

xix

IL Interleukin iNOS inducible NOS i.p. Intraperitoneal

IP PGI receptor

Ip3 Inositol triphosphate iPLA2 Calcium-independent phospholipase A2

IVR Intervening or linker region

JNK c-Jun N-terminal kinases kb Kilobases kDa Kilodalton

Keap1 (Kelch-like erythroid cell-derived protein with CNC homolog (ECH)-

associated protein 1)

LDH Lactate dehydrogenase

L-DOPA L-Dihydroxyphenylalanine

LH luteinizing hormone

LPO Lipoxygenase

Maf Avian musculoaponeurotic fibrosarcoma

MAO Monoamine oxidase

MAPK Mitogen-activated protein kinase

MBD Membrane binding domain

MDA 3,4-methylenedioxyamphetamine

MDMA 3,4-methylenedioxymethamphetamine

MEH Microsomal epoxide hydrolase

MEKK1 Mitogen-activated protein kinase kinase kinase

METH Methamphetamine

Mn Manganese mPGES Microsomal PGE synthase

MPTP 1-methyl-4-phenyl-1,2,3,6-tetrohydropyridine

MRP Multidrug resistance-associated protein

MS Multiple sclerosis mtNOS Mitochondrial nitric oxide synthase

NAD+ Nicotinamide adenine dinucleotide

NADPH Nicotinamide adenine dinucleotide phosphate

NE Norepinephrine

Neh Nrf2-ECH homology

NEPP Electrophilic neurite outgrowth-promoting prostaglandin

NET Norepinephrine transporter

NES Nuclear export signal

NF-E2 p45-nuclear factor erythroid-derived 2

NF-IL6 Nuclear factor for interleukin-6 expression

NF-κB Nuclear factor kappa B

NHB1 N-terminal homology box 1

NLS Nuclear localization sequence

NMDA N-methyl-d-aspartate nNOS Neuronal nitric oxide synthase

NO Nitric oxide

xxi

NO. Nitric oxide radical

NOS Nitric oxide synthase

NOX NADPH oxidase

NQO1 NAD(P)H:quinone oxidoreductase

Nrf2 Nuclear factor erythroid 2-related factor 2

NSAIDs Non-steroidal anti-inflammatory drugs

NTR N-terminal region

O2-. Superoxide anion

OD Olfactory discrimination

OGG1 Oxoguanine glycosylase 1

ONOO Peroxynitrite

P450 Cytochromes 450

PBN N-tert-butyl-alpha-phenyl nitrone

PD Parkinson's disease

PDGF Platelet-derived growth factor

PET Positron emission tomography

PG Prostaglandin

PGG2 Prostaglandin G2

PGH2 Prostaglandin H2

PHS Prostaglandin H synthase

PKA Protein kinase A

PKC Protein kinase C

PLA2 Phospholipase A2

xxii

PlsEtn-PLA2 Plasmalogen-selective phospholipase A2

PMA Phorbol-12-myristate-13-acetate

PNMT Phenylethanolamine-N-methyltransferase

PNW Postnatal week

PPAR Peroxisome proliferator-activated receptor

Prx Peroxiredoxins

PPRE Peroxisome proliferator response element

RAR Retinoic acid receptor

ROH Alcohol

ROOH Alkyl hydroperoxide

ROS Reactive oxygen species

RSA Radical spin adduct

RT-PCR real-time polymerase chain reaction

SE Serotonin

SERT Serotonin transporter

SFhERRbeta Estrogen-related receptor beta (ERR)-beta--short-form

SMRT Silencing mediator for retinoid and thyroid hormone receptors

SOD Superoxide dismutase

SOD1 Cu,Zn- Superoxide dismutase

SOD2 Mn- Superoxide dismutase

SOD3 Extracellular Cu,Zn- Superoxide dismutase

Sp1 Specificity protein 1 sPLA2 Secretory phospholipase A2

xxiii

SRE Sterol response element

TBARS Thiobarbituric acid-reactive substances. tBHQ Tertiary butylhydroquinone

TGF Tumor growth factor

TNF-α Tumor necrosis factor-alpha

TP Thromboxane receptor

TPA 12-O-tetradecanoylphorbol-13-acetate

TRE (12-O-tetradecanoylphorbol-13-acetate (TPA)-responsive element)

Trx Thioredoxins

TXs Thromboxanes

UGT Uridine 5'-diphospho (UDP)-glucuronosyltransferase.

VEGF Vascular endothelial growth factor

VMAT-2 Vesicular monoamine transporter-2 v-src Rous sarcoma virus.

XRE Xenobiotic response elements

xxiv

LIST OF PUBLICATIONS & PRESENTATIONS ARISING FROM THIS THESIS

Published Manuscripts, Submitted

1. Goncalves, L. L., Ramkissoon, A. and Wells, P. G. (2009). Prostaglandin H synthase-1- catalyzed bioactivation of neurotransmitters, their precursors, and metabolites: oxidative DNA damage and electron spin resonance spectroscopy studies. Chem Res Toxicol 22(5): 842-852.

2. Ramkissoon, A. and Wells, P. G. (2011). Human prostaglandin H synthase (hPHS)-1- and hPHS-2-dependent bioactivation, oxidative macromolecular damage, and cytotoxicity of dopamine, its precursor, and its metabolites. Free Radic Biol Med 50(2): 295-304.

3. Ramkissoon, A. and Wells, P. G. (2011). Human prostaglandin H synthase (hPHS)-1 and hPHS-2 in amphetamine analog bioactivation, DNA oxidation and cytotoxicity. Toxicol Sci 120(1): 154-162.

4. Jeng, W., Ramkissoon, A. and Wells, P. G. (2011). Reduced DNA oxidation in aged prostaglandin H synthase-1 knockout mice. Free Radic Biol Med 50(4): 550-556.

5. Ramkissoon, A. and Wells, P.G. (2010). Methamphetamine oxidative stress, neurotoxicity and functional deficits are modulated by Nrf2. (submitted)

6. Ramkissoon, A. and Wells, P.G. (2010). Developmental role of nuclear factor-E2-related factor 2 (Nrf2) in protection against methamphetamine fetal toxicity and postnatal functional deficits in Nrf2-deficient mice. (submitted)

Abstracts 1. Ramkissoon A. and Wells P.G. (2009). Acute methamphetamine (METH)-initiated oxidative stress and neurotoxicity are not modulated by nuclear factor-E2-related factor 2 (Nrf2). Toxicol. Sci. (Supplement: The Toxicologist) 108 (1): 448 (No. 2152).

xxv

2. Ramkissoon A. and Wells P.G. (2009). Methamphetamine-initiated cognitive neurodevelopmental deficits in nuclear factor-E2-related factor 2 (Nrf2)-deficient knockout mice. Birth Defects Research Part A: Clinical and Molecular Teratology 85: 449 (No. P67).

3. Ramkissoon A. and Wells P.G. (2008). Human prostaglandin H synthase (hPHS)-1- and hPHS-2-dependent protein and DNA oxidation caused by dopamine, its precursor and metabolites. Toxicol. Sci. (Supplement: The Toxicologist) 102(1): 373 (No. 1814).

4. Ramkissoon A. and Wells P.G. (2007). Acetylsalicylic acid blocks human prostaglandin H synthase-1 and 2-dependent cytotoxicity caused by dopamine, its precursor and metabolites. Program No. 382.13. Society for Neuroscience meeting, San Diego, California.

5. Ramkissoon A. and Wells P.G. (2007). Prostaglandin H synthase (PHS)-1/2-dependent oxidative DNA damage and cytotoxicity caused by neurotransmitters, their precursors and metabolites. Abstract No. 5947. FASEB Experimental Biology meeting, Washington, DC.

6. Ramkissoon A., Goncalves L.L., Jeng W. and Wells P.G. (2005). Prostaglandin H synthase-1 (PHS-1)-catalyzed bioactivation of dopamine and its precursor L-DOPA to free radical intermediates that initiate oxidative DNA damage. Abstract No. 39. Proceedings of the annual meeting of the Society of Toxicology of Canada.

7. Wells, P. G., Ramkissoon, A. and Jeng, W. (2004). Neuroprotection against endogenous oxidative stress in aging prostaglandin H synthase-1 (PHS-1) knockout mice. Toxicological Sciences (Supplement: The Toxicologist) 78(S-1): 63 (No. 305).

xxvi

CHAPTER 1: INTRODUCTION

1.1 RATIONALE, HYPOTHESIS AND OBJECTIVES

1.1.1 RATIONALE

Reactive oxygen species (ROS), such as superoxide anions, hydrogen peroxide and hydroxyl radicals, can oxidize cellular macromolecules including proteins, lipids, DNA and

RNA (Halliwell and Gutteridge, 2007). Such damage, if not repaired, can accumulate over time and can lead to loss of cellular function, functional deficits and even cell death that may play a role in ROS-mediated neurodevelopmental deficits and neurodegeneration (Wells et al., 2010).

However, ROS also can activate redox-sensitive transcription factors that can lead to the transcription of an array of genes that modulate toxicity. The brain is especially susceptible to oxidative stress due to its high rate of oxygen consumption, low antioxidant levels and a variety of ROS-generating enzymatic reactions (Halliwell and Gutteridge, 2007).

Prostaglandin H synthase (PHS), also known as cyclooxygenase (COX), is a heme- containing enzyme that catalyzes the initial steps in the production of prostaglandins and thromboxanes (Casarett et al., 2008). There are two isoforms of this enzyme, PHS-1 and PHS-

2. PHS-1 is constitutively expressed in most tissues including the brain. PHS-2 is non- constitutive but inducible in most tissues, but is constitutive in the macula densa of the kidney and the vas deferens, and is highly expressed and inducible in brain regions such as the cortex, hippocampus and striatum (Breder and Saper, 1996; Kaufmann et al., 1997; Yamagata et al.,

1993). Also, elevated levels of PHS-2 in neurons have been found in the brains of patients with

Alzheimer's disease (AD) and Parkinson's disease (PD) (Pasinetti, 1998; Teismann and Ferger,

2001). Subcellular locations of PHS include the endoplasmic reticulum and the nuclear envelope

(Morita et al., 1995; Spencer et al., 1998).

PHS is a bifunctional enzyme that has cyclooxygenase and peroxidase activities (Breder 3

et al., 1995; Kaufmann et al., 1997). The cyclooxygenase component converts arachidonic acid to the endoperoxide-hydroperoxide prostaglandin G2 (PGG2) (Figure 1). The peroxidase component reduces the hydroperoxide to prostaglandin H2 (PGH2), in the process oxidizing a co-substrate (Casarett et al., 2008; Marnett, 1990; Wells et al., 2010). It is during this latter step that endogenous compounds or xenobiotics can serve as co-substrates that are bioactivated to free radical intermediates that can generate ROS (Casarett et al., 2008; Marnett, 1990; Wells et al., 2010).

The contribution of PHS to xenobiotic bioactivation, embryonic deoxyribonucleic acid

(DNA) oxidation and teratogenicity has been characterized in our laboratory for several teratogens including phenytoin, thalidomide and benzo[a]pyrene (Parman et al., 1998; Parman and Wells, 2002; Parman et al., 1999; Winn and Wells, 1997). Recently our laboratory has found that the amphetamine analogs 3,4-methylenedioxymethamphetamine (MDMA, Ecstasy), methamphetamine (METH, Speed) and methylenedioxyamphetamine (MDA, active metabolite of MDMA) can be bioactivated by brain PHS-1 to free radical intermediates that generate ROS and oxidatively damage brain DNA, with neurodegenerative consequences (Jeng et al., 2006;

Jeng and Wells, 2010). MDMA, METH and MDA are synthetic substances that are common drugs of abuse as they promote the release of neurotransmitters and induce euphoria and hallucinations (Colado et al., 2001; Kalant, 2001; McGuire, 2000). These acute effects appear to be receptor-mediated, and are due to the rapid release of neurotransmitters from synaptic vesicles (Camarero et al., 2002; Kalant, 2001). Alternatively, various animal and human studies have shown that there are long-term neurodegenerative effects of these amphetamine analogs persisting weeks or months after the drug is eliminated, such as decreases in neurotransmitter 4

Figure 1. Postulated bioactivation of endogenous substrates to a free radical intermediate by PHS which generates ROS to activate Nrf2. Cyclooxygenase and hydroperoxidase are the components of PHS. Arachidonic acid released from membrane phospholipids by phospholipase A2 serves as the co-substrate in the cyclooxygenase-dependent pathway, generating a hydroperoxide (PGG2), which can then be reduced by hydroperoxidase to an alcohol. In the peroxidase pathway, xenobiotics or endogenous compounds may serve as the reducing co-substrate, themselves being oxidized to a reactive free radical intermediate that can initiate the formation of reactive oxygen species (ROS) including superoxide anions (O2-.), hydrogen peroxide (H2O2) or hydroxyl radicals (.OH). If ROS are not detoxified by enzymes such as superoxide dismutase (SOD) or catalase, they can oxidatively damage cellular macromolecules and/or alter signal transduction, thereby causing irreversible damage and neurotoxicity. Abbreviations: Nrf2, nuclear factor erythroid 2- related factor 2; PGG2, prostaglandin G2, PHS, prostaglandin H synthase; ROS, reactive oxygen species (Modified from: Winn and Wells, 1995, Parman et al., 1998). 5

levels and nerve terminal markers, and increases in astrogliosis in brain regions such as the striatum, cortex, hippocampus and olfactory bulb (Cadet et al., 2001; Colado et al., 2001; Deng et al., 2007; Jeng and Wells, 2010; O'Callaghan and Miller, 1994; Wilson et al., 1996; Jeng et al., 2006). Such long-term effects are thought to be at least in part ROS-mediated, as demonstrated by increased levels of protein carbonyls, lipid peroxidation and oxidative DNA damage in degenerated regions in mouse models (Camarero et al., 2002; Gluck et al., 2001;

Jayanthi et al., 1998; Jeng et al., 2006). Also, postmortem brains of chronic METH users exhibit elevated levels of the lipid peroxidation products 4-hydroxynonenal and malondialdehyde in the caudate nucleus and to a lesser extent in the frontal cortex (Fitzmaurice et al., 2006).

The term „”neurotoxicity” in this thesis refers to toxicity as a consequence of exposure to a toxic agent that result in toxicities including, oxidative damage, swollen organelles, shrunken cell membranes, autophagic vacuoles that may lead to cell death or nerve terminal degeneration.

This may result in an increase in the expression of glial fibrillary acidic protein (GFAP), which is a marker of astrogliosis. In METH studies in rodent models, loss of presynaptic nerve terminals has been identified by silver degeneration staining to obtain evidence of METH- induced neuronal damage as identified by morphological signs of axonal degeneration

(O'Callaghan and Miller, 1994; Ricaurte et al., 1982). This has been correlated to a reduction in neurotransmitters or enzymes for their synthesis, or transporters present in the nerve terminal, which are associated with functional deficits (discussed later). These biochemical changes and functional deficits may reflect receptor-mediated mechanisms and be reversible within days as the drug is eliminated. However, damage to critical components may be only partially reversible, and can persist for months after drug exposure (Jeng et al., 2006; Jeng and Wells, 6

2010). Currently, it is not clear from animal models whether the toxic effects of METH and related amphetamine analogs are only long lasting or, at least in some cases, permanent.

Amphetamine analogs can be metabolized by cytochrome P450s (CYPs), especially

CYP2D6, to reactive intermediates that generate ROS; however, the levels of CYPs in the brain are low compared to the liver (Warner et al., 1997), and other metabolic pathways may contribute to amphetamine toxicity. Experiments with MDMA in PHS-1 knockout mice revealed the important in vivo role of this isozyme in the molecular mechanism of ROS- mediated neurodegeneration, where knockouts showed decreased oxidized DNA, nerve terminal degeneration and locomotor functional deficits when compared to wild-type controls (Jeng and

Wells, 2010). The same outcomes caused by METH and MDA in CD-1 mice were reduced by pretreatment with the PHS inhibitor acetylsalicylic acid (ASA, aspirin) (Jeng et al., 2006).

As well, aging studies with untreated PHS-1 knockout mice aged 1.5 to 2 years showed significantly decreased DNA oxidation in brain regions such as the cortex, hippocampus, cerebellum and brainstem when compared to age matched wild-type mice (Jeng et al., 2011).

Conversely, aged mice overexpressing human PHS-2 in neurons were found to have increased levels of apoptosis and astrogliosis and poor performance on the Morris water maze compared to age-matched non-transgenic mice (Andreasson et al., 2001). Among other possibilities, these studies suggest that endogenous compounds in the brain may be bioactivated by PHS to free radical intermediates, which initiate the formation of ROS that oxidize cellular macromolecules, potentially leading to neurotoxicity. Phenols, catechols and amines are good substrates for

PHS, and many neurotransmitters, their precursors and metabolites, as well as the amphetamine analogs contain these functional groups. When incubated in vitro with PHS, the neurotransmitter dopamine can be converted to a reactive quinone that can covalently bind to 7

DNA forming depurinating adducts at the N-7 of guanine and protein sulfhydryl groups

(Hastings, 1995; Mattammal et al., 1995). Dopamine quinones can also undergo one-electron reductions catalyzed by NADPH cytochrome P450 reductase (Segura-Aguilar et al., 1998). This reaction can create a redox cycling process with oxygen, leading to the formation of ROS

(Segura-Aguilar et al., 1998). However, other enzymatic processes may contribute to ROS generation from dopamine and other endogenous sources.

In an in vitro system containing PHS-1 and 2-deoxyguanosine, not only dopamine, but also epinephrine and the dopamine precursor L-DOPA were able to generate ROS that oxidized

2-deoxyguanosine to form 8-oxo-2‟-deoxyguanosine (8-oxo-dG) (Jeng et al., 2011). Such oxidized bases in non-dividing or permanent cells, such as neurons, may contribute to the neurodegeneration associated with aging. The role of PHSs in the bioactivation of endogenous brain compounds and the amphetamine analogs needs further investigation. There are structural similarities between the amphetamine analogs and dopamine, as well as its precursors and metabolites, suggesting these substrates may be bioactivated similarly, and may share a common mechanism of toxicity. Dopamine, its precursors and metabolites may be also be important in amphetamine neurotoxicity as these analogs release neurotransmitters in the cytosol where they can be bioactivated leading to enhanced neurotoxicity.

Most studies to date have investigated bioactivation using purified PHS or tissue microsomes containing ovine and rodent PHS. Studies using purified PHS or microsomes to compare PHS-1/2 bioactivating efficacy are difficult to interpret due to differing isozyme catalytic stability. To avoid this confounding factor, we used ovine PHS-1 as well as Chinese hamster ovary (CHO)-K1 cell lines stably expressing human PHS-1 (hPHS-1) or hPHS-2.

These cell lines provided us with the opportunity to study human PHS as opposed to ovine 8

enzymes in vitro, or rodent enzymes in vivo, either of which may provide different information on substrate specificity and/or efficacy (Wiese et al., 2001). Similarly, we could determine whether drugs can be differentially bioactivated by hPHS-1 versus hPHS-2.

In addition to oxidatively damaging cellular macromolecules, ROS generated through the bioactivation of xenobiotics or endogenous compounds can lead to altered ROS-mediated signal transduction (Figure 1). The redox-sensitive nuclear transcription factor nuclear factor erythroid 2-related factor 2 (Nrf2) is frequently activated during various types of oxidative stress. Under normal conditions, Nrf2 is located in the cytoplasm as an inactive form associated with its repressor protein Keap1 (Itoh et al., 1999b). Oxidation of redox-sensitive cysteine residues in Keap1 leads to dissociation of Nrf2 from Keap1, allowing Nrf2 to translocate to the nucleus where it binds to the antioxidant response element (ARE), an enhancer sequence that regulates the transcription of cytoprotective enzymes (Chan et al., 2001; Itoh et al., 1999b).

Nrf2 has been shown to be protective against benzo[a]pyrene carcinogenicity and the neurotoxicity of 3-nitropropionic acid and kainic acid (Calkins et al., 2005; Kraft et al., 2006;

Ramos-Gomez et al., 2001). Also, aged Nrf2 knockout animals have been shown to develop vacuolar leukoencephalopathy and astrogliosis suggesting a role for Nrf2 in the brain (Hubbs et al., 2007). It is currently uncertain whether exposure to oxidative stress initiated in specific brain regions by amphetamine analogs can activate Nrf2 and modulate neurotoxicity. One study using Nrf2 knockout mice suggests that METH-initiated oxidative stress and neurotoxicity does not involve Nrf2 (Pacchioni et al., 2007); however, only a single endpoint, dopamine depletion in the striatum, was evaluated at a single time point, two weeks post-treatment. Also, multiple day dosing, allowing time for Nrf2-mediated activation of protective effects were not assessed.

Studies in rat striatum indicate that METH can activate Nrf2 and regulate HO-1 expression 9

(Jayanthi et al., 2009), however whether this plays a neuroprotective effect is not known.

Nrf2 also may play a developmental role, as in both humans and rodents, in utero exposure to METH is associated with fetal developmental toxicities including reduced birth weight (Plessinger, 1998; Smith et al., 2006), and postnatal neurobehavioral and functional deficits (Acuff-Smith et al., 1996; Chang et al., 2004; Cho et al., 1991; Jeng et al., 2005;

Slamberova et al., 2006; Wong et al., 2008). These effects are inconsistent among studies, possibly due to differences in dosage, gestational timing and duration of exposure, as well as genetic differences that modulate toxicological susceptibility. The viability of Nrf2-deficient knockout mice has suggested that Nrf2 plays no role in normal mouse development (Chan et al.,

1996), although Nrf2 mRNA and protein are expressed during the period of organogenesis

(Chan et al., 1996). Accordingly, it is uncertain whether ROS-mediated Nrf2 activation in specific brain regions caused by METH can modulate its neurotoxicity in fetal and adult brain.

Dose- and time-ranging studies investigating various brain regions using multiple endpoints are needed to properly determine whether Nrf2 may play a neuroprotective role during toxicity.

1.1.2 HYPOTHESIS AND OBJECTIVES

The focus of this thesis was to elucidate critical mechanisms involved in ROS generation and macromolecular damage, as well as mechanisms in ROS-mediated signaling and cytoprotective responses, in the context of neurodegeneration associated with amphetamines and aging. Specifically, we sought to evaluate the bioactivation of endogenous brain compounds and the amphetamine analogs by PHSs to free radicals that generate neurotoxic ROS, and the protective role of Nrf2-regulated protective responses in the fetus and adult brain.

We hypothesized that the ovine and human PHS-1 and PHS-2 can catalyze the 10

bioactivation of MDA and METH as well as neurotransmitters, their precursors and metabolites to free radical intermediates that initiate ROS formation, oxidative macromolecular damage and ultimately neurotoxicity. In addition, we hypothesized that METH-initiated ROS formation may activate Nrf2-regulated cytoprotective responses, and that this ROS-mediated signaling pathway may play an important neuroprotective role in METH neurotoxicity, and during fetal development, protecting against endogenous and drug-enhanced oxidative stress. We hypothesized that the absence of Nrf2 in knockout mice will result in enhanced neurotoxicity that may be dose- and time-dependent, as well as localized by brain region and/or cellular type, and lead to functional and cognitive deficits.

The objectives were as follows:

1. To determine if ovine PHS can bioactivate neurotransmitters, their precursors and metabolites to free radical intermediates that initiate DNA oxidation. Our evaluation focused upon the dopaminergic pathway of neurotransmitters including not only dopamine (DA), but also its precursor L-dihydroxyphenylalanine (L-DOPA) and its metabolites dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA) and 3-methoxytyramine (3-

MT).

2. To determine the bioactivating efficacy of the human PHS isozymes using CHO-K1 cell lines stably expressing hPHS-1 or hPHS-2 or untransfected CHO-K1 cells to evaluate DA, its precursor and metabolites in oxidative macromolecular damage and cytotoxicity, corroborated by the potential protective effects of PHS inhibition and ROS detoxification.

3. To similarly determine the bioactivating efficacy of the human PHS isozymes using CHO-

K1 cell lines stably expressing hPHS-1 or hPHS-2 or untransfected CHO-K1 cells to evaluate

METH and MDA in oxidative macromolecular damage and cytotoxicity, corroborated by the 11

potential protective effects of PHS inhibition.

4. To determine whether METH can activate Nrf2 and thereby alter susceptibility to METH- initiated oxidative stress, neurotoxicity and functional deficits using Nrf2 knockout mice.

5. To determine in pregnant Nrf2 knockout mice if there is a developmental role for Nrf2 in the context of in utero METH-initiated oxidative stress during organogenesis by determining the activation of Nrf2 in fetal brain, and its protective impact on METH-initiated DNA oxidation, fetal toxicities and postnatal functional deficits.

1.2 BRIEF OVERVIEW

1.2.1 ROS IN THE BRAIN

1.2.1.1 Introduction to Reactive Oxygen Species

ROS, such as superoxide anions, hydrogen peroxide and hydroxyl radicals are produced due to the incomplete reduction of oxygen during aerobic metabolism (Halliwell and Gutteridge,

2007). Superoxide anion, the product of a one-electron reduction of oxygen, is the precursor of most ROS and a mediator in oxidative chain reactions. Dismutation of superoxide anions, either spontaneously or through a reaction catalyzed by superoxide dismutases (SODs) produces hydrogen peroxide. This may be fully reduced to water or partially reduced to hydroxyl radicals, one of the strongest oxidants in nature. The formation of hydroxyl radicals is also catalyzed by reduced transition metals. Important radicals are summarized in Table 1. These

ROS are highly reactive as they contain an unpaired electron and can oxidize molecular targets such as proteins, lipids and DNA in a process known as oxidative stress (Halliwell and

Gutteridge, 2007). Such damage, if not repaired, can accumulate over time and can lead to loss of cellular function and even cell death (Figure 2) (Wells and Winn, 1996; Wells et al., 2009).

The brain is especially susceptible to oxidative stress due to the high rate of oxygen consumption, low antioxidant levels and ROS-generating enzymatic reactions (Halliwell and

Gutteridge, 2007). ROS have been implicated in many neurodegenerative diseases such as AD

(Gabbita et al., 1998), PD (Shimura-Miura et al., 1999) and multiple sclerosis (MS)

(Vladimirova et al., 1998). 13

Table 1. Important Reactive Species (Reiter, 1995; Halliwell and Gutteridge, 2007) Reactive Species Notes 2-. -6 Superoxide anion O Free radical, source of H2O2, t1/2 = 1x10 s . Hydrogen peroxide H2O2 Oxidizing agent, source of OH , t1/2 = min . -9 Hydroxyl radical HO Very reactive, t1/2 = 1x10 s Nitric oxide radical NO. Free radical, reacts with O2-. Peroxynitrite ONOO- Can decompose to OH. 2+ 2-. Ferrous iron Fe Reacts with H2O2, oxidation leads to O Ferric iron Fe3+ Oxidized form of ferrous iron 14

Figure 2. Enzymatic pathways involved in reactive intermediate-mediated neurotoxicity. Susceptibility to the neurotoxic effects involves the balance between drug bioactivation, elimination, detoxification and pathways of cytoprotection and repair. Under normal conditions or during low oxidative stress, cells are able to detoxify endogenous and xenobiotic reactive intermediates and ROS with appropriate enzymes: GSH reductase, GSH peroxidase, G6PD, SOD, catalase, peroxiredoxins. However, when bioactivation exceeds detoxification, high levels of reactive intermediates can lead to cellular damage. Such damage can be repaired by: p53, ATM, OGG1, CSB, Trx. If not repaired, molecular damage and/or alteration of signal transduction can lead to neurotoxicity. Abbreviations: ATM, ataxia telangiectasia mutated; CSB, Cockayne syndrome B; G6PD, glucose-6-phosphate dehydrogenase; GSH, glutathione; LPO, lipoxygenase; OGG1, oxoguanine glycosylase 1; P450, cytochromes 450; PHS, prostaglandin H synthase; ROS, reactive oxygen species; SOD, superoxide dismutase. Modified from Wells and Winn 1996.

1.2.1.2 Sources of Reactive Oxygen Species

There are numerous sources of ROS in the brain that can lead to oxidative damage to macromolecules and neurotoxicity. These are summarized below and in Figure 3.

1.2.1.2.1 Mitochondria

Mitochondria play a central role in the survival and death of neurons. Mitochondria exerts multiple influences on neuronal function including the generation of adenosine-5'- triphosphate (ATP) and sequestering of calcium. The mitochondrial respiratory chain is also the major site for the generation of superoxide anions (Turrens and Boveris, 1980) and hydrogen peroxide (Boveris et al., 1976; Boveris et al., 1972). The electron transport chain, which is embedded in the inner membrane of the mitochondria, consists of five multiprotein complexes.

Although molecular oxygen is reduced to water in complex IV by a sequential four-electron transfer, a proportion can be reduced by a one-electron addition that occurs in complex I

(Turrens and Boveris, 1980) and also in complex III (Boveris et al., 1976; Sugioka et al., 1988), resulting in the formation of ROS. The mitochondrial matrix, however, contains the antioxidant enzyme manganese (Mn)-superoxide dismutase that may combat the high rate of superoxide production in the mitochondrial inner membrane (Halliwell and Gutteridge, 2007). The production of hydrogen peroxide by mitochondria appears to account for 1-2% of the total oxygen consumed in vitro (Chance et al., 1979). As a result, the steady state concentrations of superoxide anions and hydrogen peroxide in the mitochondrial matrix have been estimated to be around 1 x 10−10 M and 5 × 10−9 M, respectively (Cadenas and Davies, 2000). Alterations in mitochondrial function can play an important role in increasing ROS steady state levels and contribute to neurotoxicity.

Figure 3. Sources of reactive oxygen species (ROS) in the brain. When pro-oxidants exceed the antioxidative and repair mechanisms, oxidative damage can occur leading to neurotoxicity. Abbreviations: ATM, ataxia telangiectasia mutated; CSB, Cockayne syndrome B; Fe, iron; G- 6-P, glucose-6-phosphate; GSH, glutathione; GSSG, glutathione disulfide; H2O2, hydrogen peroxide; HO•, hydroxyl radical; LPO, lipoxygenase; NADP+, nicotinamide adenine dinucleotide phosphate; O2-•, superoxide anion, OGG1, oxoguanine glycosylase 1; P450, cytochromes P450; PHS, prostaglandin H synthase; SOD, superoxide dismutase. Modified from Wells et al., 2010. 17

1.2.1.2.2 Enzymes

NADPH oxidase (NOX)

NOX is found in the brain. There are seven different isoforms including NOXs 1 to 5 and dual oxidases (DUOX) DUOX1 and DUOX2. Little is known about the role of NOX5,

DUOX1 and DUOX2 in the brain (Sorce and Krause, 2009). All NOX family members are transmembrane proteins that transport electrons across biological membranes to reduce oxygen to superoxide (Bedard and Krause, 2007). When the complex is active, it generates superoxide by transferring an electron from nicotinamide adenine dinucleotide phosphate (NADPH) in the cytosol to oxygen on the luminal or extracellular space. Superoxide is the primary product of the electron transfer, but hydrogen peroxide is also generated (Babior et al., 1975). The NOX isoforms have been detected in various brain regions (Infanger et al., 2006; Kim et al., 2005;

Serrano et al., 2003). Most studied has been NOX2 which can be found in regions including the cortex, hippocampus (Kim et al., 2005) and striatum (Serrano et al., 2003). Within these regions, NOX enzymes have been investigated with respect to inflammatory processes regulated by NOX2 in microglia (Block et al., 2007; Lavigne et al., 2001). However, NOXs are also present in astrocytes (Abramov et al., 2005) and in mouse and human neurons (Serrano et al.,

2003; Haorah et al., 2008). There is also increasing evidence for a role of NOX2 in neurodegeneration, including in AD and amyotrophic lateral sclerosis (ALS) (Shimohama et al.,

2000; Wu et al., 2006).

Phospholipase A2 (PLA2)

PLA2 enzymes are esterases that cleave the acyl ester bond at the sn-2 position of membrane phospholipids to produce free fatty acids, for example AA (Farooqui et al., 2006). 18

These enzymes are broadly classified into groups including secretory phospholipase A2

(sPLA2), cytosolic phospholipase A2 (cPLA2), plasmalogen-selective phospholipase A2

(PlsEtn-PLA2), and calcium-independent phospholipase A2 (iPLA2) (Farooqui et al., 2006;

Hirashima et al., 1992; Matsuzawa et al., 1996). They are present in various regions of the brain and are expressed in astrocytes (Gattaz et al., 1995; Stephenson et al., 1999) and in neurons (Matsuzawa et al., 1996; Sandhya et al., 1998; Farooqui et al., 2006). Enzymatic activity is regulated by calcium, ROS and neurotransmitters, and activation releases arachidonic acid that is metabolized by cyclooxygenases and lipoxygenases, in the process generating ROS

(discussed in detail later). cPLA2 activities may contribute to neurotoxicity in AD and MS

(Gattaz et al., 1995; Kalyvas and David, 2004) and cPLA2-deficient mice are resistant to 1- methyl-4-phenyl-1,2,3,6-tetrohydropyridine (MPTP)-induced neurotoxicity, an animal model for PD (Klivenyi et al., 1998).

Nitric oxide synthases (NOSs)

There are four members of the nitric oxide synthases (NOS) family. A constitutive isoform of NOS called neuronal NOS (nNOS) is found in neurons (Cork et al., 1998; Bredt et al., 1990). Another isoform is inducible NOS (iNOS) in which inflammatory mediators such as lipopolysaccharide (LPS) and cytokines cause its expression in microglia and astrocytes

(Murphy, 2000; Lee et al., 1993) and possibly in neurons (Heneka and Feinstein, 2001). Other isoforms include endothelial NOS (eNOS) and mitochondrial NOS (mtNOS) (Brown, 2007;

Elfering et al., 2002; Guix et al., 2005). L-arginine is used by NOS to produce NO and citrulline in a process requiring NADPH and O2 (Iyengar et al., 1987). Although all NOS isoforms can potentially produce superoxide anions, iNOS is the most likely to produce 19

superoxide anions in vivo due to L-arginine depletion during inflammation (Xia and Zweier,

1997). Nitric oxide (NO) is mainly used for guanylate cyclase activation with the subsequent production of cyclic guanosine-3′,5′-monophosphate (cGMP) (Ignarro, 1991). NO, however, can react with superoxide anions forming peroxynitrite anion (ONOO-) which is a very reactive anion that can oxidize proteins (Beckman et al., 1990). NO free radicals also triggers apoptosis when it binds to cytochrome c oxidase and induces the formation of superoxide anions in the mitochondria, generating ONOO- (Brown, 1999). These enzymes are involved in various neurodegenerative diseases including AD (Smith et al., 1997). In MS, NOS may be induced and there is evidence of oxidative stress where ONOO- is believed to contribute to the cellular damage (Bagasra et al., 1995; Liu et al., 2001).

Monoamine oxidase (MAO)

During catecholamine metabolism, ROS can also be generated through enzymatic reactions. For example, MAOs, which are associated with mitochondrial membranes, catalyze the oxidation of amines to their corresponding aldehydes and ammonia with the formation of hydrogen peroxide as a by product (Shih, 1991). There are 2 isoforms of this enzyme, MAO-A and MAO-B. MAO-A is in neurons in the catecholinergic cell areas (Westlund et al., 1985).

MAO-B is contained in serotonergic neurones in the median raphe and in astrocytes (Westlund et al., 1985; Levitt et al., 1982). MAO-B is selectively inhibited by L-deprenyl (Magyar and

Knoll, 1977), MAO-A is selectively inhibited by clorgyline (Johnston, 1968). Both forms oxidize dopamine, tyramine, and octopamine (Youdim and Riederer, 1993). MAO-B is also responsible for the oxidation of MPTP to MPP+ which damages dopaminergic neurons

(Heikkila et al., 1984). Alterations in MAO-B activity have been implicated in PD 20

(Hotamisligil et al., 1994) and patients with PD have elevated MAO-B activity in the substantia nigra (Riederer and Jellinger, 1983).

Cytochromes P450 (CYPs)

Cytochromes P450 (CYPs) are a superfamily of heme-containing monooxygenases that metabolize a large number of compounds. CYPs are involved in the biosynthetic pathways of steroid and bile acid production, and most CYPs metabolize xenobiotics. P450s carry out the oxidation of carbon and nitrogen groups usually resulting in the addition of an -OH (Casarett et al., 2008). In catalyzing the metabolism of a drug, CYPs use NADPH to reduce O2 leading to the production of hydrogen peroxide and superoxide anion radicals. CYP2E1 metabolism of a number of substrates is known to lead to increased ROS (Mari and Cederbaum, 2000). CYP2E1 mRNA has been detected in several mammalian brain regions (Dutheil et al., 2009; Farin and

Omiecinski, 1993). It has also been proposed that selective localization of CYP2E1 in dopamine-containing neurons may contribute to nigrostriatal toxicity in chemically induced

Parkinson's disease (Jenner, 1998). While it is uncertain whether other CYPs may contribute to

ROS generation, the expression of CYPs in the brain is 1-2% of that in the liver (Sasame et al.,

1977; Warner et al., 1997). CYP2D6 is an isozyme involved in the metabolism of many drugs active in the central nervous system, such as antidepressants and antipsychotics. This enzyme is coded by a polymorphic gene, with 7% of the Caucasian population showing no enzymatic activity („poor metabolizers‟). Approximately 20–30% of Caucasians carry one active and one inactive allele, and show intermediate enzyme activity (here referred to as „intermediate metabolizers‟). Individuals carrying two active alleles are „extensive metabolizers‟ (Sachse et al., 1997). Whether this enzyme directly forms ROS in the brain is uncertain. 21

Xanthine oxidoreductase

Guanine degrades into xanthine and during ATP catabolism during hypoxia hypoxanthine can be formed (Harrison, 2002). Xanthine oxidoreductase is a widely distributed enzyme that catalyzes the oxidation of hypoxanthine to xanthine and of xanthine to uric acid

(Harrison, 2002). The enzyme occurs in two forms, xanthine dehydrogenase and xanthine oxidase. During these reactions H2O2 and superoxide anions are produced. Only xanthine dehydrogenase is capable of reducing nicotinamide adenine dinucleotide (NAD+). Both forms can reduce molecular oxygen, although xanthine oxidase is more effective (Saito and Nishino,

1989). In the reoxidation of fully reduced xanthine oxidase, the first two steps each involve transfer of two electrons to oxygen, generating hydrogen peroxide. Xanthine oxidase then transfers its remaining electrons in separate steps, with each electron independently reducing O2 to produce superoxide anions (Berry and Hare, 2004). Xanthine dehydrogenase and xanthine oxidase can be interconverted by means of sulfhydryl reagents. When xanthine dehydrogenase is treated with proteases, like trypsin, it is irreversibly transformed into xanthine oxidase (Berry and Hare, 2004). Reversible conversion occurs due to conditions that oxidize thiol groups of

Cys535 and Cys992, exposure to sulphydryl agents and exposure to anaerobic conditions (Berry and Hare, 2004). Xanthine oxidase is localized to the vascular endothelium of brain (Betz,

1985). It is thought to play a neurotoxic role during ischemia-reperfusion injury (Beetsch et al.,

1998; Battelli et al., 1998). During ischemia, ATP is broken down into hypoxanthine and upon reoxygenation, xanthine oxidase converts the excessive hypoxanthine to xanthine thereby generating ROS (Berry and Hare, 2004). 22

1.2.1.2.3 Excitotoxicity

Excitotoxicity is a phenomenon whereby prolonged activation of excitatory amino acid receptors leads to cell death (Lucas and Newhouse, 1957; Olney and Sharpe, 1969). Glutamate has been identified as the principal transmitter mediating fast excitatory synaptic responses in the vertebrate brain. Glutamate distribution within the brain is extensive. Glutamate is present at concentrations of 5–15 μmol/g weight of wet tissue in humans, and regional distribution of glutamate in the brain is similar between species from rat to human (Erecinska and Silver, 1990;

Perry et al., 1987). Glutamate receptors are divided into ionotropic, which are ligand-gated ion channels, and metabotropic receptors that are linked to G-proteins (Dong et al., 2009). These are further subdivided, for example, the ionotropic receptors are characterized by their selective affinity for the specific agonists N-methyl-d-aspartate (NMDA), α-amino-3-hydroxy-5- methylisoxazole-4-propionic acid (AMPA) and kainic acid (Monaghan et al., 1989).

Excitotoxicity can result in excessive intracellular calcium, generation of free radicals, activation of the mitochondrial permeability transition and secondary excitotoxicity.

Excitotoxicity more directly is toxicity that is related to calcium influx subsequent to glutamate receptor activation, and it is greatly attenuated in the absence of calcium (Carriedo et al., 2000).

Excitotoxicity leads to ROS production and the oxidative agents in turn promote excitotoxic mechanisms, usually by disrupting calcium homeostasis and activating calcium-dependent proteases (Avshalumov and Rice, 2002; Carriedo et al., 2000). Termination of the excitatory action of glutamate is mediated by a high-affinity uptake system on both pre- and post-synaptic neuronal cell membranes and the membranes of adjacent glial cells (Danbolt, 2001). Glutamate transporter 1, which is located on astrocytes (Conti and Weinberg, 1999), is responsible for most of the total glutamate transport (Danbolt, 2001). Excitotoxicity not only affects neurons, but 23

also astrocytes and oligodendrocytes as well (Chen et al., 2000; Oka et al., 1993).

Excitotoxicity, through these various mechanisms, has been proposed to explain the pathology characteristic of neurodegenerative diseases such as AD, Huntington‟s disease (HD) and ALS

(Foran and Trotti, 2009; Song et al., 2003; Parameshwaran et al., 2008).

1.2.1.2.4 Immune Response-Microglia

Microglia are considered the local immune cells of the brain and are present in various regions (Mittelbronn et al., 2001). Microglia density varies by brain region in the adult human and in adult mice; these cells are in the grey matter, with the highest concentrations being found in the hippocampus, olfactory telencephalon, basal ganglia and substantia nigra (Mittelbronn et al., 2001; Lawson et al., 1990). Microglia are normally in a resting state and become activated in response to injury (Nimmerjahn et al., 2005; Fetler and Amigorena, 2005). The resting microglial cells are activated by detecting LPS, beta-amyloid (Aβ), thrombin, interferon-gamma

(IFN-γ), and other proinflammatory cytokines (Dheen et al., 2007). A crucial function of microglia is their ability to generate significant immune responses. For example, microglia initiate responses including the production of cytokines, chemokines, ROS and nitric oxide

(Nakamichi et al., 2005; Colton et al., 1994). Activated microglia release interleukin (IL)-1, tumor necrosis factor-alpha (TNF-α) and chemokines for lymphocyte recruitment (Hartlage-

Rubsamen et al., 1999; Floden et al., 2005; Dheen et al., 2007). Phagocytic and cytotoxic functions of microglia are also triggered during CNS injury. During these processes, significant amounts of ROS can be produced. When activated, microglia can contribute to ROS production via NOX enzymes as discussed previously (Block et al., 2007; Lavigne et al., 2001). The brain microglia can generate significant quantities of superoxide anion and NO (Colton et al., 1994). 24

However, species differences in their generation has been observed with mouse microglia generating large amounts of NO when stimulated. In contrast, human and hamster microglia do not produce measurable amounts of NO under the same stimulation conditions, but both human and hamster microglia generate significantly more superoxide anion than rat microglia (Colton et al., 1996). Neurotoxins such as MPTP as well as LPS can also overactivate microglia leading to increased ROS generation and neuronal death (Gao et al., 2002; Gao et al., 2003). Microglia have been shown to play a role in AD, PD and HD (McGeer et al., 1988; McGeer et al., 1987;

Sapp et al., 2001).

1.2.1.3 Neuroprotective Mechanisms

ROS are involved not only in cellular damage, but also serve as a cellular signaling molecule in a large number of reversible regulatory processes. The physiological generation of

ROS in the brain can occur as a byproduct of many biological reactions as discussed above.

However, the brain has structural and both enzymatic and non-enzymatic antioxidants to limit

ROS generation and enhance detoxification (Figures 1 and 2). Neurons are outnumbered by the various supportive cells that in non-pathological conditions can provide protection as well

(Figure 4).

1.2.1.3.1 Blood Brain Barrier

The central nervous system is isolated from the bloodstream by the blood-brain-barrier

(BBB). This unique barrier is supported by the endothelial cells of the brain capillaries, which form complex tight junctions. The capillaries are also surrounded by a sheath of astrocytes end- feet, which provides a high density of cells to restrict passage from the blood to the brain (Reese 25

and Karnovsky, 1967; LeFevre and Peters, 1966; Abbott et al., 2010). While the barrier makes it difficult for polar or large molecules to cross into the brain in high quantity, the BBB is quite dynamic in regulating the exchange of substances between blood and brain (Pardridge et al.,

1990). Transport of nutrients, ions and hormones is required to maintain optimal conditions for neuronal and glial functions (Reese and Karnovsky, 1967; LeFevre and Peters, 1966; Pardridge et al., 1990; Abbott et al., 2010). In addition, enzymes such as MAO, gamma-glutamyl transpeptidase and several CYP enzymes may provide metabolic protection in the brain (Ghersi-

Egea et al., 1994; el-Bacha and Minn, 1999; DeBault and Cancilla, 1980; Dauchy et al., 2008).

This is especially important as a wide range of lipid-soluble molecules can diffuse though the

BBB and enter the brain passively. A number of ATP-binding cassette (ABC) energy- dependent efflux transporters (ATP-binding cassette transporters) also actively pump many of these agents out of the brain (Begley, 2004; Dauchy et al., 2008). There is also evidence that

BBB dysfunction, either through structural changes or the alteration of transporters, may contribute to neurotoxicity associated with MS, PD and AD (Correale and Villa, 2007; Desai et al., 2007; Cirrito et al., 2005; Bartels et al., 2008; Kortekaas et al., 2005).

1.2.1.3.2 Antioxidant Mechanisms

Endogenous antioxidants are needed to protect the brain, as most exogenous antioxidants do not efficiently cross the blood–brain barrier due to their hydrophilic nature (Moosmann and

Behl, 2002). To limit ROS-mediated damage, there are a number of endogenous antioxidant and protective xenobiotic-metabolizing enzymes present in the brain. These can include different enzymes to regenerate antioxidant potential as well as phase II detoxification enzymes.

In the brain, expression of phase II detoxification enzymes occurs in astrocytes and much less in 26

Figure 4. Cells of the central nervous system (CNS) and their function. Neurons are structurally diverse with a cell body, axon and synaptic terminal that are involved in neurotransmitter synthesis and storage. Also, present in the CNS are glial cells including oligodendrocytes, astrocytes and microgila with distinct functions.

neurons (Murphy et al., 2001). Chemical inducers can increase detoxification proteins in both neurons and astrocytes. Furthermore, neurons in close proximity to astrocytes may obtain protective factors from the astrocytes (Johnson et al., 2002).

Glutathione (GSH) is a tripeptide called gamma-glutamyl-cysteinyl-glycine. GSH is synthesized from L-glutamate and cysteine via the enzyme gamma-glutamylcysteine synthetase

(GGCS) (also known as glutamate cysteine ligase, GCLg) forming gamma-glutamylcysteine.

This reaction is the rate-limiting step in glutathione synthesis. Glycine is added to the C- terminal of gamma-glutamylcysteine via the enzyme GSH synthetase to then form GSH

(Dringen, 2000). The GSH system is one of the most important antioxidant systems in the brain. Neuronal cells contain high concentrations of GSH and display high activity of GSH peroxidase (GPx) and GSH reductase (Dringen, 2000). However, astrocytes also have important roles to play in supplying GSH substrates to neurons. This mechanism of substrate supply minimizes the neurotoxic effects of large amounts of extracellular cysteine, which can activate glutamate receptors (Zeevalk et al., 2008). Astrocytes synthesize and export GSH, which can then undergo transpeptidation to cysteinylglycine and gamma-glutamyl amino acid by the ecto-enzyme gamma-glutamyl transpeptidase. The cysteinylglycine generated can then be utilized by neurons to manufacture GSH, after first undergoing dipeptide cleavage to its constituent amino acids.

GSH reduces disulfide bonds formed within cytoplasmic proteins to cysteines by serving as an electron donor and in the process being oxidized to GSH disulfide (GSSG). GSH can also conjugate with electrophilic compounds, mediated by the glutathione S-transferases (GSTs), and these conjugates can be removed from the cell (Dringen, 2000). In the reaction catalyzed by

GPx, the tripeptide GSH serves as an electron donor to reduce H2O2 to water. Besides H2O2, 28

GPx also reduces organic hydroperoxides to their corresponding alcohols (Ursini et al., 1995).

During this reduction of peroxides, GSH is oxidized by GPx to GS•, which combine to form

GSSG. Within cells, GSH is regenerated from GSSG in a reaction catalyzed by GSH reductase.

This enzyme transfers electrons from NADPH to GSSG, thereby regenerating GSH (Dringen et al., 2005). GPx knockout mice appear normal at birth, but when exposed to neurotoxins exhibit enhanced levels of neuropathology (Klivenyi et al., 2000). In astrocytes, when H2O2 clearance by catalase is inhibited, the glutathione system can almost completely compensate for its detoxification (Liddell et al., 2006a; Liddell et al., 2006b).

SOD are metal-containing and include Cu,Zn-SOD (SOD1), Mn-SOD (SOD2) and extracellular Cu,Zn-SOD (SOD3) (Johnson and Giulivi, 2005; Marklund, 1982). They catalyze the dismutation of superoxide anion to molecular oxygen and hydrogen peroxide (Johnson and

Giulivi, 2005). SOD1 dismutates superoxide throughout the cytoplasm and is also found in peroxisomes and the nucleus. SOD1 is mainly expressed in astrocytes and neurons. By immunohistochemistry, motor neurons of the spinal cord appear to have greater SOD1 protein expression (Johnson and Giulivi, 2005). SOD2 is inducible and located in mitochondria and is predominantly localized in neurons throughout the brain and spinal cord (Maier and Chan,

2002). SOD2 KO mice suffer from postnatal neurodegeneration highlighted by cell death in the cortex and brainstem regions (Melov et al., 1998). SOD1 knockout mice do not exhibit these effects and are more sensitive to neurotoxicity caused by MPTP (Zhang et al., 2000). However transgenics overexpressing SOD1 are protected from 3-nitropropionic acid (3-NP) and METH

(Beal et al., 1995; Jayanthi et al., 1998). In the case of SOD3, overexpression has been found to be detrimental to neurons impairing long-term potentiation (Thiels et al., 2000); however, others have found mice overexpressing SOD3 may be protected from neurobehavioural deficits 29

during aging (Hu et al., 2006).

Detoxification of hydrogen peroxide is performed by antioxidant enzymes such as GPx, catalase and peroxiredoxins (Prx). Catalase is an intracellular antioxidant enzyme that catalyzes the conversion of hydrogen peroxide into water and molecular oxygen. It is located in the peroxisomes and to a lesser extent in the cytosol, and is ubiquitously expressed in a wide variety of brain cells; however, catalase expression is relatively low in brain compared to other tissues.

For example, activity of catalase is two orders of magnitude lower in brain than in kidney or liver (Ho et al., 1997).

The family of Prx enzymes consists of six antioxidant proteins that are involved in the degradation of H2O2, organic hydroperoxides and peroxynitrite, and are dependent on a reactive cysteine at the active site (Rhee et al., 2005). Prx proteins reduce peroxides by employing an electron donor and redox-sensitive cysteines that undergo oxidation/reduction cycles (Sarafian et al., 1999). Prx proteins are abundantly expressed in the cytosol; however, several isoforms can also be found in mitochondria, peroxisomes, nuclei and membranes (Hofmann et al., 2002).

Different brain regions and cell types show distinct basal expression profiles of Prx. For example, Prx proteins 1 and 6 are expressed in glial cells, whereas Prx proteins 2–5 are localized in neurons (Sarafian et al., 1999).

Another family of proteins acting in conjunction with Prx are thioredoxins (Trx), with

Trx1 in the cytosol and Trx2 in the mitochrondria. They bind to proteins and form intermediates that reduce protein disulphide bridges, in the process becoming oxidized themselves. Oxidized Trx proteins can be reduced through the action of Trx reductase using

NADPH as a cofactor (Patenaude et al., 2005). Trx1 has been detected in human brain which showed positive Trx1-like staining in white matter astrocytes (Patenaude et al., 2005). The 30

mitochondrial isoform, Trx2, is abundant and widely distributed in rat brain (Rybnikova et al.,

2000). Brain regions showing highest expression at the RNA and protein levels include the olfactory bulb, frontal cortex, thalamus, cerebellum and the brainstem (Rybnikova et al., 2000).

A range of hormones, chemicals, and stress conditions have been shown to induce Trx1 expression in the brain (Patenaude et al., 2005).

Inducible heme oxygenase (HO) HO-1 and constitutive HO-2 belong to the family of heat shock proteins and protect brain cells from oxidative stress by degrading toxic heme into the biliverdin, free iron and carbon monoxide (Wagener et al., 2003). Subsequently, biliverdin is converted by biliverdin reductase into bilirubin which also has antioxidant properties where it becomes oxidized back to biliverdin (Dore et al., 1999). HOs have both antioxidative and anti- inflammatory properties, and their expression is induced by a variety of different stimuli, including their substrate heme and oxidative stress. HOs are localized in membranes including the endoplasmic reticulum, nucleus and plasma membrane (Ryter et al., 2006). Polymorphisms in the lengths of guanine-thymine (GT) repeats (11–40) within the HO-1 promoter appear to be an important determinant of HO-1 expression and function in humans. Long GT sequences code for a protein with lowered basal levels and reduced induction when stimulated. Robust

HO-1 activity is associated with the short-GT polymorphisms, and appears to be protective against atherosclerosis-linked conditions (Exner et al., 2004). HO-1 is expressed in unstressed brain; however, the level is minimal and HO-1 is confined to glial cells and sparse populations of neurons in the cerebellum (Purkinje cells), thalamus, hypothalamus, brain stem, hippocampal dentate gyrus and cerebral cortex (Baranano and Snyder, 2001; Schipper et al., 2009). In contrast, HO-2 mRNA and protein are widely distributed and strongly expressed in neurons of with the highest concentrations in hippocampal pyramidal cells and dentate gyrus, olfactory 31

epithelium and olfactory bulb, and cerebellar granule and Purkinje cell layers (Schipper et al.,

2009). Following exposure to a variety of stimuli, HO-1 induction occurs in neuronal and non- neuronal brain cells, although it has been argued that astrocytes have a greater capacity than neurons for robust HO-1 expression.

NAD(P)H:quinone oxidoreductase (NQO1) is a cytosolic flavoprotein that catalyzes the two-electron reduction of quinones to hydroquinones by using either NADPH or NADH as the donor. This prevents quinone electrophiles from participating in reactions that could lead to either sulfhydryl depletion, or to one-electron reductions that can generate semiquinones and various reactive oxygen intermediates as a result of redox cycling (Dinkova-Kostova and

Talalay, 2010; Siegel and Ross, 2000). In addition, the hydroquinone products of the NQO1 reaction can be further metabolized to glucuronide and sulfate conjugates, which can then be excreted. NQO1 is mainly expressed in astrocytes and brain endothelial cells and prevents the generation of ROS (Siegel and Ross, 2000; van Horssen et al., 2006; van Muiswinkel et al.,

2004). In the human, NQO1 is mainly expressed in astrocytes, vascular endothelium and in a subpopulation of dopaminergic neurons (van Muiswinkel et al., 2004).

Other antioxidants in the brain include vitamins. Vitamin E, also known as α- tocopherol, is a lipid soluble antioxidant concentrated in the cell membranes. It contributes an electron to the peroxyl radical that is formed during the chain reaction of lipid peroxidation, and therefore is classified as a chainbreaking antioxidant (Halliwell and Gutteridge, 2007). The vitamin E radical produced when the parent compound donates an electron is unreactive, and it eventually degrades or is recycled to vitamin E by ascorbate. α-Tocopherol is important for normal brain physiology as patients with a prolonged deficiency of this vitamin suffer from neurological deficits (Muller and Goss-Sampson, 1990). The function of vitamin C (ascorbic 32

acid) in the brain is a double-edged sword with respect to free radical damage (Halliwell and

Gutteridge, 2007). In general, both the gray and white matter of the brain contain ascorbic acid, which can act as a scavenger and functions as an antioxidant in recycling the vitamin E radical back to vitamin E. Vitamin C is an effective antioxidant as it can react with most other biologically relevant radicals and oxidants to generate the ascorbyl radical which has a low reactivity due to resonance stabilization of the unpaired electron and readily dismutates back to ascorbate (Carr and Frei, 1999). In contrast, vitamin C can also be a strong prooxidant through interactions with iron (Reiter, 1995; Carr and Frei, 1999). Ascorbate can maintain metal ions in their reduced state and hence can lead to reaction of the reduced metal ions with hydrogen peroxide generation hydroxyl radicals (Carr and Frei, 1999).

1.3 PROSTAGLANDIN H SYNTHASES (PHSs)

1.3.1 ROLE OF PROSTAGLANDIN SYNTHESIS AND THEIR RECEPTORS

Eicosanoids comprise a class of bioactive lipid mediators derived from the metabolism of polyunsaturated fatty acids by PHSs and lipoxygenases leading to prostanoids and leukotrienes respectively (Funk, 2001; Smith et al., 2000). Prostanoids can be further divided into prostaglandins (PGs) and thromboxanes (TXs). The most typical actions are the relaxation and contraction of various types of smooth muscles. They also modulate neuronal activity by either inhibiting or stimulating neurotransmitter release or inducing central actions such as fever and sleep induction (Narumiya et al., 1999). PGs also regulate secretion and motility in the gastrointestinal tract as well as transport of ions and water in the kidney (Narumiya et al., 1999).

AA serves as the metabolic precursor for eicosanoid synthesis. AA is not available in large quantities in the free acid form, but is stored on the backbone of membrane phospholipids. To be used for biosynthesis, PLA2 (reviewed previously) liberates AA from phospholipids in the membrane (Schaloske and Dennis, 2006). This is the rate-limiting step in eicosanoid synthesis

(Irvine, 1982).

5-Lipoxygenase (5-LOX, reviewed previously) performs the initial enzymatic step in leukotriene synthesis (Murphy and Gijon, 2007), creating 5-hydroperoxyeicosatetraenoic acid

(5-HPETE) by incorporating one molecular oxygen at the C-5 position of AA. Depending on cellular conditions, 5-HPETE has a number of potential metabolic fates. It can be secreted in its peroxide form, reduced to 5-hydroxyeicosatetraenoic acid (5-HETE), or undergo a catalytic rearrangement in the 5-LOX active site to form leukotriene A4.

PHSs, also known as COX, consist of two isozymes, PHS-1 (COX-1) and PHS-2 (COX- 34

2). PHSs contain two distinct active sites, a COX and a hydroperoxidase site, both of which use the same tyrosyl radical and heme-iron for catalysis. The COX site incorporates molecular O2 at the 11- and 15-carbon on AA to form the hydroperoxy endoperoxide PGG2. The hydroperoxidase site reduces the peroxide to the corresponding alcohol, the hydroxy endoperoxide PGH2, which is the substrate for various PG synthases (Hamberg et al., 1974;

Ohki et al., 1979). PGH2 can form a number of different bioactive products through the action of PG synthases (Figure 5). This includes a number of important signaling molecules, including

PGI2 (also known as prostacyclin), PGD2, PGE2, PGF2α and thromboxane A2 (TXA2). PGI2 is formed by prostacyclin synthase, a member of the CYP superfamily (Wu and Liou, 2005). PGI2 binds the G protein-coupled receptor (GPCR) PGI2 receptor (IP) (Boie et al., 1994) as well as the transcription factors peroxisome proliferator-activated receptor (PPAR)α, PPARδ, and

PPARγ (Kojo et al., 2003). PGD2 is major product of rat brain homogenate formed by PGD synthases (Abdel-Halim et al., 1977). PGD2 has two known receptors, PGD2 receptor 1 (DP1) and chemoattractant receptor-homologous, molecule expressed on T helper-2 cells (CRTH2)

(DP2) (Pettipher, 2008). The effects of PGE2 have been implicated in many biological processes (Park et al., 2006). PGE2 synthesis occurs through three unique enzymes, named the cytosolic PGE synthase (cPGES), microsomal PGE synthase-1 (mPGES-1) and mPGES-2 (Park et al., 2006). PGF2α, made by PGF synthase, has been implicated in a number of physiological processes and disease states (Basu, 2007). Only one PGF2α-specific receptor has been cloned

(Basu, 2007), a GPCR termed PGF receptor (FP), which upon binding ligand results in an elevation of intracellular calcium. TXA2 biosynthesis is catalyzed by a member of the CYP superfamily, thromboxane A synthase. It also binds a specific GPCR, termed the thromboxane

A receptor (TP), which leads to increased intracellular calcium (Kinsella et al., 1997). 35

Cyclopentenone PGs are a family of molecules that are formed by dehydration of PGE2 and

PGD2 (Straus and Glass, 2001). Dehydration of PGE2 leads to PGB2. PGD2 dehydration leads to

15-deoxy-12,14-prostaglandin J2 (15d-PGJ2). 15d-PGJ2 has been identified as a high affinity ligand for the transcription factor PPARγ as well as a less potent activator of PPARα and

PPARδ (Powell, 2003).

There are nine types and subtypes of receptor for prostanoids, designated PGD receptor

(DP1) and the CRTH2 or (DP2), EP1, EP2, EP3, and EP4 subtypes of PGE receptor, PGF receptor (FP), PGI receptor (IP), and TXA2 receptor (TP) (Narumiya et al., 1999). All of these prostanoid receptors are GPCRs and their main signal transduction pathways leads to a rise in intracellular cyclic adenosine monophosphate (cAMP) and/or increases in calcium. Their functions and expression are presented in Table 2.

In addition to their synthesis, extracellular levels of PGs also depend on transport processes, which are regulated by PG transporter (an influx transporter) and the multidrug resistance-associated protein 4 (an efflux transporter) (Schuster, 2002). Also, inactivation in the cytoplasm can occur through hydroxyprostaglandin dehydrogenase (also known 15-PGDH).

15-PGDH is highly expressed in normal tissues but is lacking in human colon (Backlund et al.,

2005). Lack of 15-PGDH expression in tumours results in increased endogenous PGE2 levels

(Backlund et al., 2005). 36

Figure 5. Biosynthesis of prostaglandins. Arachidonic acid (AA) is released from phospholipids by phospholipase A2 and is used in both prostaglandin synthesis and the lipoxygenase-eicosanoid pathway. Cyclooxygenase and hydroperoxidase are components of prostaglandin H synthase (PHS). PG, prostaglandin; PGG2, prostaglandin G2; PGH2, prostaglandin H2; HPETE, hydroperoxyeicosatetraenoic acid; HETE, hydroxyeicosatetraenoic acid. 37

Table 2. Prostanoid receptor subtypes, expression and functions. (Andreasson, 2010; Boie et al., 1995; Coleman et al., 1994; Hasumoto et al., 1997; Hata and Breyer, 2004; Hirata et al., 1994; Katsuyama et al., 1997; Namba et al., 1992; Narumiya et al., 1999; Oida et al., 1997; Oida et al., 1995; Sugimoto et al., 1994; Watabe et al., 1993). Receptor Ligand Expression Function EP PGE2 - Cerebral cortex, hippocampus, - Acute signs of inflammation EP1 Purkinje cells - Bone formation - Kidney, lung, stomach - Cerebral blood flow - Gastrointestinal tract

EP2 - Least abundant, inducible - Role in synaptic plasticity - Neurons in forebrain, - Role in cerebral blood flow hypothalamus - Induced in uterus, implantation - Low in Gastrointestinal tract

- Widely distributed - Fever generation EP3 - Kidney, uterus - mRNA in neurons of cortex, hippocampus, midbrain - Monominergic neurons of the substantia nigra - Smooth muscle of gastrointestinal tract

EP4 - Forebrain neurons - Anti-apoptotic effects - Kidney - Inflammatory and anti- - Uterus inflammatory effects DP1 PGD2 - Low levels in human tissue - Immune response - Low in lung, stomach, uterus - Sleep Induction - Leptomeninges of brain

DP2 PGD2 - Th2 cell specific - Allergy (CRTH2) - Lymphocytes - Chemotaxis

FP PGF2α - High in corpus luteum, ovaries - Luteolysis in pregnancy

IP PGI2 - Neurons in dorsal root ganglion - Mediation of pain - Megakaryocytes - Inflammation - Smooth muscle of arteries - Vasodilator - Kidney TP TXA2 - Vasculature of lung, kidney, - Regulation of immunity heart - Hemodynamics - Thymus, Spleen - Vasoconstrictor - Bronchoconstrictor 38

1.3.2 GENETICS OF PHS

1.3.2.1 Genes

Despite their close structural and functional similarities, the PHS isozymes are encoded by different genes that are differentially regulated, leading to distinct expression patterns and biological functions. Sheep PHS-1 was determined by cDNA cloning in 1989 (DeWitt and

Smith, 1988; Merlie et al., 1988; Yokoyama et al., 1988) followed by cloning and sequence analysis of human PHS-1 (Yokoyama and Tanabe, 1989). It is located on chromosome 9

(Yokoyama and Tanabe, 1989). The human and mouse gene for PHS-1 is approximately 22 kilobases (kb) in length with 11 exons and 10 introns and is transcribed as a 2.8 kb mRNA.

PHS-1 is a glycoprotein that in processed form has 576 amino acids with an apparent molecular mass of 70 kilodalton (kDa).

The PHS-1 gene promoter lacks a TATA and CAAT box but is GC-rich, and contains multiple transcription start sites (Wang et al., 1993). These promoter features are usually characteristic of housekeeping genes that are constitutively expressed under basal conditions.

Within the 5′ flanking region of the human PHS-1 promoter there are three functional specificity protein 1 (Sp1) binding sites at −610, −111, and −89 relative to the ATG start site. Reporter gene assays have demonstrated that the Sp1 sites at -610 and -111 are functionally important in maintaining basal constitutive expression of PHS-1 (Xu et al., 1997).

PHS-2 was discovered in 1991, as a primary response gene (Kujubu et al., 1991; Xie et al., 1991). The gene for PHS-2 is approximately 8.3 kb long with 10 exons and it is transcribed as 4.6, 4.0, and 2.8 kb mRNA variants. The human PHS-2 gene is located on chromosome 1.

The cDNA for COX-2 encodes a polypeptide that with the signal peptide region sequence 39

contains 604 amino acids and shares 61% homology with the human COX-1 polypeptide

(Sawaoka et al., 2003). The gene structures of PHS-1 and PHS-2 demonstrate conservation of exon–intron junctions (Kosaka et al., 1994). Unlike PHS-1, sequence analysis of the 5′-flanking region of the human PHS-2 gene has identified several potential transcriptional regulatory elements, including a peroxisome proliferator response element (PPRE), two cyclic AMP response elements (CRE), a sterol response element (SRE), two nuclear factor kappa B (NF-κB) sites, an Sp1 site, a CAAT enhancer binding protein (C/EBP, or nuclear factor for interleukin-6 expression (NF-IL6)) motif, two activator protein 2 (AP-2) sites, an E-box, and a TATA box.

The promoter regions of PHS-2 genes have sequences of typical immediate early genes such as c-fos and c-jun (Smith et al., 2000; Tanabe and Tohnai, 2002).

A comparison of the genes for PHS-1 and PHS-2 showed that the first and last exons differ in size. There is a 42-base deletion in exon 1 of the human PHS-2 gene, which encodes a smaller signal peptide than the PHS-1 gene (Kosaka et al., 1994). PHS-2 exon 1 encodes the signal peptide region and is only 14% identical in amino acid sequence to the corresponding exon of PHS-1 (Appleby et al., 1994). Furthermore, exon 10 of the PHS-2 gene has a larger 3‟- untranslated sequence and a 54-base insert in the protein-coding region, which encodes 18 PHS-

2-specific amino acids. Exon 3 of the human PHS-2 gene contains an additional three- nucleotide insert, which codes for a proline residue absent in exon 4 of human PHS-1 (Kosaka et al., 1994). For PHS-2, the human and mouse genes have similar structure in genomic organization. Sequence comparison showed the first 200 basepair (bp) of the human PHS-2 promoter share 67-65% identity with that of mouse and rat respectively (Tazawa et al., 1994).

There are also some interspecies differences in the sequences of the human and mouse PHS-2 genes. For example, the mouse PHS-2 promoter has one NF-κB motif and two C/EBP sites 40

instead of the two NF-κB sites and one C/EBP motif found in the human PHS-2 promoter.

However, for PHS-1, human and mouse genes share approximately 60% sequence identity in the 230-bp 5‟-flanking region (Wang et al., 1993).

A new member of the PHS family, cyclooxygenase-3 (COX-3) also known as COX-1b, has been identified and characterized in canine tissues (Chandrasekharan et al., 2002). Canine

COX-3 mRNA is identical to the PHS-1 mRNA, except that the intron-1 is retained. In canines,

COX-3 is 90 nucleotides in length and represents an in-frame insertion into the portion of the

PHS-1 open reading frame encoding the N-terminal hydrophobic signal peptide

(Chandrasekharan et al., 2002). Since the normal start codon resides in exon 1 and the 90-bp intron-1 sequence maintains the open reading frame, canine COX-3 mRNA creates an enzymatically active PHS-1-related peptide containing a 30-amino acid insertion near the N terminus (Chandrasekharan et al., 2002). Recently, COX-3 mRNA has been detected in tissues from rat (Kis et al., 2005), mouse (Shaftel et al., 2003) and human (Chandrasekharan et al.,

2002; Dinchuk et al., 2003). It does not appear that a full-length, catalytically active form of

COX-3 exists in humans. Retention of intron-1, however, which is 98 bp in rat and mouse and

94 bp in human should lead to a shift in the reading frame and to the synthesis of a protein very different from PHS-1 and possibly without enzymatic activity (Dinchuk et al., 2003).

1.3.2.2 Transcriptional regulation

In the human PHS-1 promoter there are three functional Sp1 binding sites at -610, -111, and -89 relative to the ATG start site. Reporter gene assays have demonstrated that the Sp1 sites at -610 and -111 are functionally important in maintaining basal expression of PHS-1 (Xu et al.,

1997). Deletion of either site leads to a reduction of 50% in basal transcription, and with 41

deletion of both sites leading to a reduction of about 75%. There is an AP-1 site located in intron 8 of the PHS-1 gene that is highly conserved across species and that interacts with the -

111 Sp1 site of the promoter to regulate induced expression of PHS-1 in MEG-01 cells (DeLong and Smith, 2005). PHS-1 gene expression is controlled and can be upregulated by tumor- promoting phorbol esters or growth factors as seen in some cell lines (Table 3). As discussed previously, the PHS-2 5‟ untranslated region has many cis-acting regulatory elements, which suggests complex regulation of the gene by a number of signaling pathways including the mitogen-activated protein kinase (MAPK). Depending on the cell type and the stimulus, distinct combinations of cis-regulatory elements will be utilized to activate PHS-2 transcription (Figure

6). PHS-2 inducers of transcriptional activation range from growth factors and hormones to shear stress (Table 4).

PPARs interfere with the transcriptional activation of the PHS-2 gene. Repression by

PPARα results from interference with NF-κB signaling pathways. PPARγ activated by ligands can block both AP-1 and NF-κB mediated gene expression of PHS-2 (Inoue et al., 2000). The mechanism of glucocorticoid-mediated repression of PHS-2 gene expression also involves suppression of AP-1 and NF-κB-dependent transcription, but also has post-transcriptional mechanisms of repression, possibly involving the regulation PHS-2 mRNA stability (Ristimaki et al., 1996). 42

Table 3. Transcriptional activation of PHS-1 Activator Cell Type Reference Tumor Epithelial cells (Kitzler et al., 1995) promoter Megakaryoblasts (Ueda et al., 1997) i.e. PMA Cytokines Fibroblasts (Kirtikara et al., 1998) i.e. IL-1β IL-2 TNFα Growth Fibroblasts (Diaz et al., 1998) Factors Vascular endothelial cells (Bryant et al., 1998) i.e. TGFβ VEGF Abbreviations: IL-1β, interleukin-1β; IL-2; interleukin-2; PMA, Phorbol-12-myristate-13- acetate; TGF, tumor growth factor; TNFα, tumor necrosis factor alpha; VEGF, vascular endothelial growth factor . 43

Table 4. Transcriptional activation of PHS-2 Activator Cell Type Reference Fibroblasts (Zhu et al., 2002) Tumor Endothelial cells (Inoue et al., 1995) promoter Epithelial cells (Subbaramaiah et al., 2001) i.e. PMA Macrophages (Inoue et al., 1995) Osteoblasts (Okada et al., 2000) Cytokines (Jones et al., 1993; Wu et al., 2003) Endothelial cells i.e. IL-1β (Deng et al., 2004; Warnock and Fibroblasts TNFα Hunninghake, 1995) Growth

Factors Fibroblasts (Kujubu et al., 1991) i.e. EGF

PDGF Macrophages (Inoue et al., 1994; Reddy and LPS Herschman, 1994) Endothelial cells (Haeffner et al., 1997)

Osteoblasts (Ogasawara et al., 2001) Shear stress Endothelial cells (Okahara et al., 1998)

Hormones i.e. LH Granulosa cells (Morris and Richards, 1996) GnRH

Oncogenes Fibroblasts (Xie et al., 1994) i.e. v-src Abbreviations: EGF, epidermal growth factor; GnRH, gonadotrophin-releasing hormone; IL- 1β, interleukin-1β; LH, luteinizing hormone; LPS, lipopolysaccharide; PDGF, platelet-derived growth factor; PMA, Phorbol-12-myristate-13-acetate; TNFα, tumor necrosis factor alpha; v- src, rous sarcoma virus. 44

Figure 6. Cell-dependent PHS-2-activation. The figure illustrates the numerous functional regulatory elements in the PHS-2 gene promoters that can be involved in the transcriptional regulation of expression when exposed to PMA. Only certain of these pathways are operative in individual cell types. Abbreviations: AP, activator protein; ATF, activating transcription factor; C/EBP, a CAAT enhancer binding protein; CRE, cyclic AMP response elements; NF-KB, nuclear factor kappa B; PMA, Phorbol-12-myristate-13-acetate. 45

1.3.2.3 Post-transcriptional regulation

While the open reading frame is conserved in both PHS-1 and 2, the promoter or 5‟ untranslated region (discussed above) and the 3‟ untranslated region are divergent. The 3' untranslated region of the PHS-2 mRNA is approximately 1.5 kb longer than that of the PHS-1 transcript and contains 23 copies of the Adenine (A)- and Uridine (U)-rich AUUUA motif that has been associated with RNA instability and may participate in post-transcriptional regulation of COX-2 expression (Appleby et al., 1994; Sawaoka et al., 2003). AUUUA motifs may also contribute to the different length of mRNA transcripts. PHS-1 contains only 1 AUUUA motif, contributing to stable mRNA (Hla, 1996).

The N-terminal active-site region of the exon 10-encoded polypeptide is similar (57%) between the two isoenzymes; however, hCox-2 contains a unique 18-amino-acid insertion in the

C-terminal region which contains a potential N-linked glycosylation site (Appleby et al., 1994).

COX-1 can be glycosylated at three sites, whereas COX-2 has four functional N-glycosylation sites. The last glycosylation site of COX-2 (Asn-594) is variably glycosylated (Appleby et al.,

1994). A role for this addition is not well established but it may be a marker for PGHS-2 for rapid proteolysis or provide a signal for subcellular trafficking. Some inducers of PHS transcription can act to stabilize the AU rich regions through activation of the c-Jun N-terminal kinases (JNK) and mitogen-activated protein kinase kinase kinase (MEKK1) pathways (Chen et al., 1998; Ming et al., 1998). IL-1 and TNF-α, for example, can both activate signal pathways and stabilize mRNA of PHS-2 through interaction with AU-rich elements, thereby increasing the half-life of mRNA from 1 to 4 hr (Mahboubi et al., 1998; Ristimaki et al., 1996).

Conversely, glucocorticoids such as dexamethasone can destabilize mRNA, effectively acting as 46

a PHS-2 inhibitor (Ristimaki et al., 1996).

1.3.3 PRIMARY PROTEIN STRUCTURES OF PHSs

The cDNA for PHS-2 encodes a polypeptide that before cleavage of the signal sequence contains 604 amino acids, and is 61% identical to the sequence of the human PHS -1 polypeptide (Sawaoka et al., 2003). Crystallographic structures of PHS-2 show striking similarity with PHS-1 (Kurumbail et al., 1996; Luong et al., 1996). The PHS enzymes are glycosylated, integral membrane proteins with globular catalytic domains. After posttranslational processing in the endoplasmic reticulum , the mature PHS-1 and PHS-2 proteins have apparent molecular masses of 67-72 kDa and exist as homodimers which bind 1 mole of heme per mole monomer.

PHS-1 and PHS-2 have many different domains, starting from the amino terminus with the signal peptide, the epidermal growth factor (EGF)-like region, the membrane binding domain (MBD), and the catalytic domain with distinct peroxidase and cyclooxygenase sites

(Kulmacz et al., 2003). Each region has important functions and may vary between the isozymes as summarized in Table 5. Additional structural features include dimerization domains through which PHS-1 and PHS-2 dimers are held together via hydrophobic interactions, hydrogen bonding, and salt bridges between the dimerization domains of each monomer (Kulmacz et al., 2003; Kurumbail et al., 1996; Luong et al., 1996).

Heterodimerization of PHS-1 and PHS-2 subunits does not occur (Luong et al., 1996).

Asparagine (N)-linked polysaccharides, are dispersed at several points along the polypeptide. Potential sites for N-linked glycosylation are conserved at residues 68, 104, 144, and 410 in PHS-1; PHS-2 lacks the site at 104 but has two additional consensus sites in the C- 47

terminal insert at residues 579 and 591 (murine PHS-2 numbering) (Otto et al., 1993).

Glycosylation of asparagine 410 in PHS-1 is essential for cyclooxygenase and peroxidase activities, probably by promoting proper protein folding (Otto et al., 1993). Blocking glycosylation destroys the activity of both isoforms. At the carboxy terminus of the catalytic domain of PHS -1 and PHS -2 are sequences that act as a signal for retention of proteins in the lumen of the endoplasmic reticulum and nuclear envelope. PHS-2 appears to be relatively more concentrated within the nuclear envelope (Morita et al., 1995; Song and Smith, 1996; Spencer et al., 1998). The major isozyme differences in the primary structure are that PHS-2 has a shorter signal peptide and an 18-amino acid C-terminal insertion. Deletion of the entire insertion site has little on the cyclooxygenase activity of human PHS-2 (Guo and Kulmacz, 2000). The catalytic domain also contains a major structural landmark called the Arg277 loop, which when cleaved destroys PHS-1 peroxidase activity but not in PHS-2. The primary sequence in the

Arg277 loop region is much less conserved between the isoforms than in the overall sequence, with only 25% identity between the human isoforms (Kulmacz et al., 2003). 48

Table 5. Comparison of domain regions of PHS-1 and 2 (Kulmacz et al., 2003) Domain Function Isozyme Differences - Directs polypetides to lumen of ER - Length: 22-26 aa (PHS-1) and nuclear envelope 17 aa (PHS-2) Amino terminal - 57 -65% conserved between human - Larger hydrophobic core in PHS-1 and signal peptide and mouse (Kulmacz et al., 2003) translocates faster to ER compared to PHS-2 (Xie et al., 1991)

- 50 aa at the N terminus - Highly conserved in both PHSs Epidermal - Has 3 disulfides interlocking growth factor - 1 disulfide linking Cys37 to (EGF) domain Cys159 to attach EGF domain to the catalytic domain

Membrane - Anchors enzyme to lipid bilayer - Amino acid sequence sharing only 33% binding domain - Forms the mouth of a hydrophobic identity in this region between isozymes (MBD) channel that leads to the cox site

Catalytic - 80% of the protein contains the - Ile-523 in PHS-1 is a valine in PHS -2. Domain cyclooxygenase and peroxidase sites - Ile-434 in PHS -1 is a valine in PHS-2

Cyclooxygenase - Converts AA to PGG2 via - Increases the volume of the PHS-2 oxygenation reactions cyclooxygenase site by 25% over that in PHS-1 (Luong et al., 1996)

- His 513 in PHS-1 is an Arg in PHS-2 and is required for the time-dependent inhibition of PHS-2 (Kurumbail et al., 1996) - PHS-2 competes more effectively arachidonic acid (Chen et al., 1999) - Arg120 critical residue for arachidonic acid binding in PHS-1, unessential in binding substrate in PHS-2 (Rieke et al., 1999) Peroxidase - Reduces PGG2 to PGH2 - Low conservation of sidechain structure

- Contains heme prosthetic group near the peroxidase site in residues 445– 457 - Structural stability of the peroxidase active site greater in PHS-1 than in PHS-2 (Xiao et al., 1998) - PHS-1 catalyzes a two-electron reduction of hydroperoxidase substrates almost exclusively, whereas PHS-2 catalyzes 60% two-electron and 40% one- electron reductions (Landino et al., 1997) 49

1.3.4 PHS ENZYMOLOGY

PHS-1 and 2 contains both cyclooxygenase and hydroperoxidase activity and is involved in AA metabolism as discussed previously. The peroxidase reaction occurs at a heme- containing active site located near the protein surface, while the cyclooxygenase reaction occurs in a hydrophobic channel in the core of the enzyme. An activated cyclooxygenase component with a crucial tyrosyl radical at Tyr385 is required to initiate hydrogen abstraction from AA

(Karthein et al., 1988). The Tyr385 radical is actually formed through a process dependent on the hydroperoxidase activity of PHS (Karthein et al., 1988) (Figure 7). The first step involves a

2-electron reduction of the hydroperoxide substrate to the alcohol product, which is supported by the two-electron oxidation of the resting ferric heme (FeIII) to a oxyferryl (FeIV) protoporphyrin cation radical (Compound 1) (Figure 7) (Smith and Song, 2002). The process of a 1-electron reduction of the protoporphyrin cation radical, leads to an oxyferryl group (FeIV) plus a neutral protoporphyrin IX (Compound 2) and a Tyr385 tyrosyl radical in the cyclooxygenase active site (Landino et al., 1997; Rouzer and Marnett, 2003; Smith et al., 2000;

Tsai et al., 1995; Tsai and Kulmacz, 2010). The first step of the cyclooxygenase reaction is then the hydrogen abstraction from C13 of AA to form a radical involving C11–C15 (Figure 7). O2 is attacked by the C11 radical, and cyclization forms a 9,11-dioxo bridge and leaves a carbon- centered radical at C8. Cyclization then occurs between C8 and C12 to generate an allyl radical at C15, in which a second O2 can give rise to a hydroperoxy radical at C15. The PGG2 radical is reduced to form PGG2 (Hamberg and Samuelsson, 1967; Landino et al., 1997; Rouzer and

Marnett, 2003; Smith et al., 2000; Tsai et al., 1995; Tsai and Kulmacz, 2010). The hydroperoxidase site reduces PGG2 to the corresponding alcohol, PGH2, and the process can regenerate tyrosyl radicals. However, these radicals can also crosslink and lead to inactivation 50

of PHSs. It is during the process of reduction that 1- or 2-electron oxidation of endogenous and exogenous compounds can occur (discussed later).

The hydroperoxidase activity can function independently of ongoing cyclooxygenase catalysis. In contrast, the cyclooxygenase reaction is peroxide-dependent and requires that the heme group at the hydroperoxidase site undergo a 2-electron oxidation to form the tyrosyl radical (Landino et al., 1997). In vitro studies have shown that the catalytic sites of the cyclooxygenase and hydroperoxidase are active for less than 2 min, which may be due to the various radicals formed and their involvement in enzyme inactivation through internal protein cross-linking (Chen et al., 1999; Rouzer and Marnett, 2003).

In terms of kinetic properties, PHS-1 and PHS-2 have similar Km values for arachidonate (5 μM) and O2 (5 μM) (Kulmacz et al., 1994; Laneuville et al., 1995; Smith et al.,

2000). Cyclooxygenase turnover rates (3500 mol/min of arachidonate per mole of dimer) are similar as well (Kulmacz et al., 1994). There are however some isozyme-dependent substrate specificities. AA with its 20-carbon chain and four cis double bonds (i.e. 20:4) is a fatty acid substrate for oxygenation, and 20:3 is 30−50% as effective as 20:4 (Laneuville et al., 1995).

18:2 and α-18:3 are poor substrates for COX-1 but are better substrates for COX-2 (Laneuville et al., 1995). Both substrates are converted to monohydroperoxide products. Furthermore, the concentration of peroxide needed to activate and sustain cyclooxygenase activity was approximately 2 nM for COX-2 and 20 nM for COX-1, therefore COX-2 may be catalytically active at much lower concentrations of hydroperoxide (Kulmacz and Wang, 1995). The catalytic activities of PHS-1 and -2 isozymes also respond differently to ASA treatment, where the cyclooxygenase activity of PHS-1 was completely inhibited, whereas ASA-treated COX-2 converted AA to 15-HPETE reduced to 15-HETE (Meade et al., 1993). ASA treatment leads to 51

acetylation of the serine residue in the cyclooxygenase active site that block PHS-1 oxygenation, but the larger active site in PHS-2 allows AA to bind after ASA treatment. The presence of the acetyl group alters the conformation of the AA so that the product of oxygenation is 15-HPETE rather than PGG2 (Lecomte et al., 1994; Meade et al., 1993).

Figure 7. Cyclooxygenase and peroxidase catalysis by PHSs. A 2-electron oxidation of the heme group by a hydroperoxide leads to compound I with iron as Fe4+ and protoporphyrin as a cation radical. This radical oxidizes Tyr 385 of cyclooxygenase generating a protein tyrosyl radical. This leads to hydrogen abstraction from C13 of AA to form a radical involving C11– C15. Through cyclization reactions and 2O2 additions, PGG2 is formed and as in step 1 of the peroxidase reaction, hydroperoxidase site reduces PGG2 to the corresponding alcohol, PGH2. Abbreviations: Fe3+ PPIX, ferric iron protoporphyrin IX (heme); ROOH, alkyl hydroperoxide; ROH, alcohol; AA, arachidonic acid; Fe4+ =O PPIX, oxyferryl heme; Compound I, an oxyferryl group (Fe(IV) = O) plus a protoporphyrin IX radical cation; intermediate II, an oxyferryl group plus a neutral protoporphyrin IX plus a Tyr385 tyrosyl radical; compound II, an oxyferryl group plus a neutral protoporphyrin IX; intermediate III, a spectral intermediate of a heme group with a protein radical located on an amino acid sidechain other than Tyr385. Reprinted from Smith, W. L. and Song, I. (2002). The enzymology of prostaglandin endoperoxide H synthases-1 and -2. Prostaglandins Other Lipid Mediat 68-69: 115-128 with permission from Elsevier (Smith and Song, 2002). http://www.sciencedirect.com/science/journal/10988823 53

1.3.5 INHIBITION OF PHSs

The main PHS inhibitors are the non-steroidal anti-inflammatory drugs (NSAIDs)

(Figure 8). In 1971, John Vane used a cell-free homogenate of guinea pig lung to demonstrate that aspirin, indomethacin and salicylate, all popular NSAIDs, were inhibitors of PHS which constituted the mechanism of action of these drugs (Vane, 1971). These classical PHS inhibitors are not selective and inhibit both PHS-1 and PHS-2. They also target the cyclooxygenase component of PHS. NSAIDs are widely prescribed as analgesics and anti- inflammatory agents.

There are different classes of PHS inhibitors (Blobaum and Marnett, 2007). The differences among them are based on their selectivity for the different isozymes, PHS-1 and

PHS-2 (Capone et al., 2007). These classes are: (i) ASA: shown to trigger a covalent acetylation in the enzyme, irreversibly blocking its activity; (ii) competitively acting NSAIDS such as indomethacin, naproxen and ibuprofen; and, (iii) PHS specific inhibitors (coxibs) (Blobaum and

Marnett, 2007; Capone et al., 2007; Simmons et al., 2004).

ASA is a covalent modifier of PHS-1 and PHS-2 as it acetylates serine 530 of PHS-1

(Loll et al., 1995; Picot et al., 1994). As mentioned previously, because the catalytic pocket of the channel in cyclooxygenase is larger in PHS-2 than in PHS-1, access of ASA to the Ser530 of

PHS-2 is reduced due to a lack of stabilization in the binding pocket, and acetylation efficiency in PHS-2 is limited. This accounts for the lowered sensitivity of PHS-2 compared with PHS-1 to inhibition by ASA (Loll et al., 1995; Picot et al., 1994). Therapeutically, this 10- to 100-fold greater selective inhibition of PHS-1 over PHS-2 by low-dose aspirin is employed in the prophylactic treatment of thromboembolic disease and myocardial infarction as ASA can inhibit

PHS-1 in platelets and the vascular endothelium at low doses (Pedersen and FitzGerald, 1984). 54

Another PHS-1 specific inhibitor is SC-560.

Other NSAIDs inhibit PHS-1 and PHS-2 by competing with AA for binding in the COX active site. However, NSAIDs significantly differ from each other in whether they bind the

COX active site in a time-dependent or independent manner. Some NSAIDs like ibuprofen have very rapid reversible binding, and therefore are not time-dependent (Selinsky et al., 2001).

Conversely, NSAIDs such as indomethacin and diclofenac are time-dependent, in that they require typically seconds to minutes to bind the COX active site. Once bound, however, these drugs typically bind with high affinity and may require hours to be washed out of the active site

(Selinsky et al., 2001). Carboxyl-containing NSAIDs form a salt bridge between the carboxylate of the NSAID and the moiety of Arg120, which provides a positive charge that binds the negative charges of carboxylic acid substrates (Loll et al., 1995; Mancini et al., 1995).

These inhibitors block entry of AA to the COX active site.

Selective inhibitors of PHS-2 were introduced in 1999. The first NSAIDs to be introduced as selective PHS-2 inhibitors were celecoxib (Celebrex) and rofecoxib (Vioxx). In place of the carboxyl group of the NSAIDs, the structure of celecoxib contains a sulfonamide group and that of rofecoxib contains a methylsulfone as does DUP-697 (Gierse et al., 1996;

Kurumbail et al., 1996). Each of these compounds is a weak time-independent inhibitor of

PHS-1, but a potent time-dependent inhibitor of PHS-2. Like time-dependent carboxyl- containing NSAIDs, time dependence for celecoxib and rofecoxib requires these compounds to enter and be stabilized in the catalytic pocket of the COX component of PHS (Gierse et al.,

1996; Kurumbail et al., 1996). However, because these drugs lack a carboxyl group, stabilization of binding for both of these drugs does not require Arg120. Instead, the combination of hydrophobic and hydrogen bonding interactions stabilizes binding. Specifically, 55

the selective, time-dependent inhibition of PHS-2 is due to the insertion of the methylsulfonyl or sulfonamide group of the inhibitor past Val-523 in PHS-2 and into the side pocket. This is precluded in PHS-1 by the extra steric bulk of Ile-523 (Copeland et al., 1994; Gierse et al.,

1996). The sulphur-containing phenyl rings of these drugs bind into the side pocket of the cyclooxygenase catalytic channel of PHS-2, but interact weakly with the active site of PHS-1

(Kurumbail et al., 1996).

Previously it was believed that the inhibition of PHS-2 mediated the therapeutic actions of NSAIDs, while the inhibition of PHS-1 caused unwanted side-effects, particularly in the gastrointestinal tract. PHS-1 is the major PHS isoform expressed in platelets and gastric mucosa. NSAID toxicity in the gastrointestinal mucosa, leading to ulceration and bleeding, is the result of inhibition of PHS-1 activity in platelets and a reduction in prostanoids important for protecting the stomach from erosion and ulceration (FitzGerald and Patrono, 2001). Coxibs selective for PHS-2 were developed to reduce the incidence of serious upper gastrointestinal toxicity associated with the administration of non-selective NSAIDs, and hence inhibition of

PHS-1-derived prostanoids. However, the reduced incidence of serious gastrointestinal adverse effects compared to non-selective NSAIDs has been countered by an increased incidence (i.e.

1% in placebo group vs 3.4 % in celecoxib group) of myocardial infarction and stroke detected in placebo controlled trials involving celecoxib and rofecoxib (Bresalier et al., 2005; Solomon et al., 2005). Therefore, chronic administration of the selective PHS-2 inhibitors for prophylactic purposes may also carry certain risks. 56

Figure 8. Examples of PHS inhibitors. Aspirin is a covalent modifier of PHS-1 and PHS-2. Indomethacin is a time-dependent PHS inhibitor. Selective PHS-2 inhibitors are celecoxib, rofecoxib (Vioxx) and DUP-697 while SC-560 is a selective PHS-1 inhibitor. 57

1.3.6 CELLULAR LOCALIZATION AND CENTRAL NERVOUS SYSTEM (CNS)

EXPRESSION OF PHSs

PHS-1 is a constitutive isoform that is widely distributed in various cell types, and is thought to mediate physiological responses. PHS-2 is rapidly induced in several cell types in response to various stimuli, such as neuronal activity, cytokines and pro-inflammatory molecules (Cao et al., 1996; Matsumura et al., 1998; Yamagata et al., 1993). Both PHS-1 and

PHS-2 are expressed under physiological conditions in some organs, such as brain, kidney, heart, liver, spleen, kidney and small intestine (Yamagata et al., 1993; Yasojima et al., 1999a).

PHS-1 can also be induced during T-cell development (Rocca et al., 1999). Several lines of evidence suggest that PHS-1 also has a role in inflammation and, like PHS-2, can be upregulated in certain conditions (see below).

In the human brain, PHS-1 mRNA has been found in regions including the hippocampus, midfrontal cortex, amygdala, substantia nigra, thalamus, occipital cortex, motor cortex, caudate and the cerebellum (Yasojima et al., 1999a). PHS-1 protein is constitutively expressed in both glia and neurons. For example, PHS-1 immunoreactivity was present in microglial cells in gray and white matter in the hippocampus and cortex (Hoozemans et al.,

2001; Yermakova et al., 1999). In rat and ovine brain, PHS-1 immunoreactivity is enriched in midbrain, pons, and medulla (Breder et al., 1995).

Mouse peripheral dorsal root ganglion neurons also appear to constitutively express

PHS-1 exclusively, and lack detectable PHS-2 expression, either under basal conditions or during peripheral inflammatory states (Dou et al., 2004). Recent studies also have indicated a proinflammatory role of PHS-1 in the pathophysiology of acute and chronic LPS-induced neurotoxicity and brain injury (Candelario-Jalil et al., 2007; Choi et al., 2008; Pepicelli et al., 58

2005; Schwab et al., 2002), and have found increased PHS-1 immunopositive microglia in association with amyloid plaques in AD (Hoozemans et al., 2001; Yermakova and O'Banion,

2001; Yermakova et al., 1999).

Several studies have shown the presence of PHS-2 mRNA and protein in different brain regions such as cerebral cortex, substantia nigra, caudate, thalamus, hippocampus and amygdala

(Kaufmann et al., 1996; Yamagata et al., 1993; Yasojima et al., 1999a). PHS-2 immunoreactivity is localized to the perinuclear regions and seems to be primarily neuronal (Ho et al., 2001; Hoozemans et al., 2001; Yamagata et al., 1993; Yasojima et al., 1999a; Tomimoto et al., 2000). It may not be detected in glia under physiologic conditions, except in radial glia of the spinal cord (Ghilardi et al., 2004). However, astrocytes and microglia can express PHS-2 after exposure to pro-inflammatory mediators in vitro or following CNS injury in vivo

(Maslinska et al., 1999; O'Banion et al., 1996; Tomimoto et al., 2000). In hippocampal and cortical glutamatergic neurons, PHS-2 has a central role in synaptic activity and long-term synaptic plasticity (Breder et al., 1995; Kaufmann et al., 1996; Yang and Chen, 2008). Within neurons, PHS-2 immunoreactivity has been localized to postsynaptic sites and dendritic and axonal domains of neurons (Kaufmann et al., 1996). The dendritic spine is a neuronal structure that can modulate large fluctuations in calcium and is believed to function in altering the efficiency of transmission at excitatory synapses. The level of neuronal PHS-2 expression within the CNS appears to be coupled to excitatory neuronal activity as PHS-2 protein expression in the brain and spinal cord is upregulated by seizure activity and peripheral inflammation (Beiche et al., 1996; Samad et al., 2001; Yamagata et al., 1993). 59

1.3.7 PHS IN ROS GENERATION

As discussed previously, co-oxidation of endogenous and exogenous substrates can occur during the catalytic process of PG biosynthesis by PHSs. When the hydroperoxidase site reduces PGG2 to the corresponding alcohol, PGH2, the process can generate tyrosyl radicals and 1- or 2-electron oxidation of endogenous and exogenous compounds (Eling and Curtis,

1992) (Figure 9a). Peroxyl radical-mediated bioactivation can also result when there is a direct transfer of the hydroperoxide oxygen to the co-substrate. This occurs during the bioactivation of compounds such as aflatoxin B1 and benzo[a]pyrene-7,8-dihydrodiol, the latter of which is bioactivated to the reactive 9,10-epoxide intermediate that is the proximate carcinogen (Figure

9b) (Eling and Curtis, 1992; Marnett, 1990; Reed and Marnett, 1982). Compounds can also be oxidized by PHS to C, N or S free radicals that trap O2 forming peroxyl radical; for example, retinoic acid is oxidized to carbon-centered radicals that react with O2 to form peroxyl free radicals (Figure 9c). Peroxyl radicals are stable oxy radicals and are able to diffuse some distance from the site of their generation to form ROS and lead to toxicity (Eling and Curtis,

1992; Marnett, 1990). A phenylbutazone carbon-centered radical can be formed by PHS hydroperoxidase (Hughes et al., 1988), and heterocyclic amines can be bioactivated by PHS to oxide intermediates that covalently bind to DNA, which can initiate cancer (Liu and Levy,

1998).

As shown in Figure 1, 1-electron oxidations of endogenous or exogenous substrates can lead to formation of free radicals that generate ROS and oxidize macromolecules such as protein, lipid, RNA or DNA (Wells et al., 2010). A variety of reducing compounds can serve as peroxidase cosubstrates for PHS-1 and PHS-2, promoting conversion of peroxide activators to the corresponding alcohols (Markey et al., 1987). Addition of aromatic amines, phenols or 60

hydroquinone, epinephrine, melatonin and serotonin (SE) facilitate the conversion of PGG2 to

PGH2 (Kulmacz, 2005; Markey et al., 1987). These can lead to free radical formation only when peroxidase reductants react via a one-electron-transfer. Phenols, catechols and amines are good substrates for PHS, and many neurotransmitters, their precursors and metabolites contain these functional groups. The neurotransmitter dopamine can be converted to a reactive quinone that can covalently bind to DNA and protein sulfhydryl groups (Hastings, 1995; Mattammal et al., 1995). Dopamine quinones can also undergo one-electron reductions catalyzed by NADPH cytochrome P450 reductase (Figure 10) (Segura-Aguilar et al., 1998). This reaction can create a redox cycling process with oxygen, leading to the formation of ROS (Segura-Aguilar et al.,

1998). NQO1 can catalyze 2-electron reductions hence preventing semiquinone radicals. Aside from binding covalently to protein and DNA, these compounds are able to generate ROS that react with DNA to form over 20 types of macromolecular lesions (Cooke et al., 2003), including the oxidation of 2‟-deoxyguanosine in DNA by hydroxyl radicals to form 8-oxodG (Wells et al.,

2010).

The amphetamines METH, MDMA, and its major metabolite MDA can be bioactivated by mouse brain PHS-1 to free radical intermediates that generate ROS and oxidatively damage brain DNA leading to neurodegeneration (Jeng et al., 2006; Jeng and Wells, 2010). Also, PHS- catalyzed bioactivation, ROS formation and embryonic DNA oxidation has been implicated in the teratogenicity of numerous xenobiotics including phenytoin, thalidomide, benzo[a]pyrene and METH (Parman et al., 1998; Parman and Wells, 2002; Parman et al., 1999; Winn and

Wells, 1997; Jeng et al., 2005; Wong et al., 2008).

There are several methodological issues of importance in determining the in vivo relevance of PHS-catalyzed bioactivation. The first is that bioactivation by the COX-2/COX-1 61

Figure 9. Mechanisms of PHS bioactivation of substrates. A) In peroxidase-mediated bioactivation, as ROOH (hydroperoxide) is reduced to ROH, cosubstrate (AH) can be oxidized to a free radical (A.). B) In peroxyl radical-mediated bioactivation, the peroxyl radical generated from cyclooxygenase (ROO.) can transfer oxygen to the substrate (X). C) A cosubstrate-derived oxidant can be formed as the substrate free radical can trap then transfer oxygen to other compounds. Reprinted from Smith, B. J., Curtis, J. F. and Eling, T. E. (1991). Bioactivation of xenobiotics by prostaglandin H synthase. Chem Biol Interact 79(3): 245-264. with permission from Elsevier. http://www.sciencedirect.com/science/journal/00092797 (Smith et al., 1991). 62

ratio for a particular xenobiotic or endobiotic will vary according to whether it is measured in intact cells, cellular homogenates, purified enzymes or recombinant proteins expressed in bacterial, insect or animal cells (Kulmacz et al., 2003; Liu and Levy, 1998; Liu et al., 1995;

Wiese et al., 2001). Secondly, PHS-catalyzed bioactivation also varies when measured in different types of cells derived from various species depending on their potential for antioxidative processes and repair (Jeng et al., 2011; Breder et al., 1995; Parman and Wells,

2002; Pepicelli et al., 2005; Wiese et al., 2001).

Figure 10. Oxidation of aminochrome by NADPH-cytochrome P450 reductase and NQO1. NADPH-cytochrome P450 reductase catalyzes one-electron reduction of aminochrome to o- semiquinone which is very reactive. The continuous NADPH oxidation and oxygen consumption leads to autoxidation. Aminochrome o-semiquinone autoxidize by reducing oxygen to superoxide radicals giving rise a redox cycling. NQO1 catalyzes two-electron reduction of aminochrome to o-hydroquinone. 64

1.3.8 PHS IN NEURODEGENERATIVE DISEASES

A number of studies have investigated the expression of PHSs in neurodegenerative diseases, especially those involving extensive neuroinflammatory effects that are thought to be mediated by PHS-2 (Table 6). ROS generation and down-stream effectors (i.e., PG and/or thromboxane synthases and their respective receptors) that are responsible for the deleterious and/or protective effects of PHS activation in neurodegenerative diseases remain an area of active research. However, research on the expression of PHS in neurodegenerative diseases have been conflicting, for example in some cases showing PHS induction or decreases in lesions of

AD (Hoozemans et al., 2001; Pasinetti and Aisen, 1998; Yermakova and O'Banion, 2001). Of particular interest is the use of NSAIDS and specific PHS-2 inhibitors and their potential neuroprotective effects. The use of NSAIDS to slow pathology of neurodegenerative diseases or aid in prevention of cognitive decline has been inconsistent. COX-2 inhibitor rofecoxib failed to slow cognitive decline in patients with mild-to-moderate AD (Aisen et al., 2003). In contrast, others showed that cyclooxygenase-2 inhibition improves β-amyloid mediated suppression of memory and synaptic plasticity (Kotilinek et al., 2008). The observed inconsistencies may be due to any or all of the following: 1) Treatment of neurodegenerative diseases may require long term prophylactic dosing and PHS-2 inhibitors are associated with increased cardiovascular toxicities as discussed previously. 2) It has to be noted that the protective effects of NSAIDs may be via non-PHS-inhibitory mechanisms, such as activation of

PPARs or through second messenger systems, suggesting that selective inhibition of COX-2 may not be the optimal therapeutic strategy (Jang and Surh, 2005). 3) Furthermore, treatment may need to change as the disease progresses as the stage of the disease may modulate protein expression. 4) The observed variability in PHS expression in the studies may be related to 65

disease stage as well.

PHS has been shown to be involved with many neurodegenerative diseases as investigated in both patients and using animal models (Table 6). Alzheimer‟s disease (AD) is the most common cause of dementia with neurodegeneration in the elderly. It is clinically characterized by a progressive memory loss and other cognitive impairments. The characteristics of AD include deposits of amyloid fibrils in senile plaques, presence of abnormal tau protein filaments in neurofibrillary tangles, and extensive neuronal degeneration and loss in regions such as the frontal cortex and hippocampus. AD brains also exhibit several additional pathological abnormalities, including reactive gliosis, microglial activation, and chronic inflammatory processes (Ritchie and Lovestone, 2002; Yasojima et al., 1999b). Recently β- amyloid-induced neuronal apoptosis has been associated with COX-2 upregulation directly through the activation of NF-κB (Jang and Surh, 2005). Generally, chronic therapy with the

PHS-2 inhibitors rofecoxib or naproxen do not aid in decreasing cognitive decline, however, non-selective PHS inhibitors may prove helpful (Aisen et al., 2003; Thal et al., 2005).

MS is a demyelinating disease of the brain characterized by perivascular infiltration of lymphocytes and macrophages into the brain. Glutamate-mediated excitotoxic death of oligodendrocytes has also been reported to contribute to the pathogenesis of demyelinating diseases (Martino et al., 2002; Matute et al., 2001). Since inflammation is associated with demyelination, oligodendrocyte death, axonal damage and, ultimately, neuronal loss, numerous studies have investigated a potential role for PHS (Table 6).

PD is a neurodegenerative disease that results in the loss of dopaminergic transmission in the substantia nigra and striatum, which leads to rigidity, resting tremors and slowness of movement. Idiopathic PD accounts for the majority (>90%) of the cases. The remaining cases 66

are mostly familial PD forms that are correlated with mutations of genes such as synuclein and parkin (Polymeropoulos et al., 1997). Idiopathic PD, unlike the familial form occurring earlier, usually begins in the fifth decade of life and progresses over long periods of time (10 to 20 years). Biochemical analyses have implicated mitochondrial dysfunction as a mechanism in idiopathic cases of PD (Schapira et al., 1990). Since PD progression has an inflammatory pathology, a potential role for PHS has been investigated (Table 6).

ALS is characterized by the progressive loss of motor neurons, typically resulting in death within 5 years of onset. It has been suggested that inflammatory-related processes may promote motor neuron death. Although the sporadic form of ALS is the most frequent, 5% to

10% of cases are familial, being associated with several genes. Missense mutations in the gene encoding for the Cu/Zn superoxide dismutase (SOD1) account for a familial form of ALS linked to chromosome 21q and present in 20% of the inherited cases (Bendotti and Carri, 2004).

Mutant SOD1 produces motor neuron injury by a toxic gain of function and several hypotheses exist, including aberrant free radical handling, abnormal protein aggregation and increased susceptibility to excitotoxicity although the exact mechanism of action is unclear (Bruijn et al.,

1998). Transgenic mice expressing the human mutant SOD1 with a phenotype that mimics clinical and pathological characteristics of the human disease have been developed and have been used to study PHS-mediated effects (Table 6). 67

Table 6. Summary of PHS-mediated effects in neurodegenerative diseases Disease/Model PHS-mediated evidence References Multiple Sclerosis (MS) - PHS-2-positive cells were present in all chronic (Rose et al., 2004) - MS patients active lesions (Bezzi et al., 1998)

- Associated with cells expressing the macrophage/ microglial marker CD64, associated with activated macrophages

- Possible oligodendroglial excitotoxic death

- Mouse - PHS-2 expression is confined within infiltrating (Deininger and experimental macrophages Schluesener, 1999) autoimmune - PHS-2 induction in astrocytes during relapse phase encephalomyelitis (EAE) - PHS inhibitor indomethacin suppressed active EAE (Reder et al., 1994)

- Rat model of - PHS-2 expression was restricted to major (Minghetti et al., delayed-type infiltrating neutrophils and phagocytes, 1999) hypersensitivity leading to - Macrophages and/or endothelial cells near lesion demyelination - Neuronal PHS-2 not affected

- No obvious PHS-2 staining in astrocytes and microglia

Table 6. Summary of PHS-mediated effects in neurodegenerative diseases continued Disease/Model PHS-mediated evidence References Alzheimer’s disease (AD) - Patients with - PHS-2 mRNA levels reported as either decreased (Chang et al., 1996), AD or increased (Pasinetti, 1998) - Increased PHS-2 in neurons (Hoozemans et al., - PHS-2-positive neurons decreased with the severity 2001; Yermakova et of dementia al., 1999)

- Early AD, an increase in PHS-2 (Ho et al., 1999; - PHS-1 expressed by microglial cells in association Hoozemans et al., with amyloid deposits 2002), - Increased PHS-1 (Yermakova et al., 1999; Kitamura et al., 1999)

- PHS-2 inhibitors (rofecoxib or naproxen) failed to (Aisen et al., 2003; slow cognitive decline Thal et al., 2005), - NSAIDS decrease the severity of cognitive (Etminan et al., symptoms 2003; Rogers et al., - Indomethacin appeared to protect the degree of 1993) cognitive decline

- Mouse models - PHS-2 inhibitors may protect against AD by (Kotilinek et al., i.e. overexpressing blocking the PHS-2-mediated PGE2 response at 2008) amyloid precursor synapses protein (APP) - Overexpressing human PHS-2 show an increase (Xiang et al., 2002) in amyloid plaques - Amyloid plaques are surrounded by a few PHS-2 (Matsuoka et al., immunoreactive astrocytes 2001) - Celebrex (PHS-2 inhibitor) did not decrease (Jantzen et al., 2002) amyloid load in APP transgenic mice - Indomethacin, but not nimesulide, showed a (Sung et al., 2004) significant reduction in the amyloid burden - A subset of NSAIDs, such as ibuprofen, have been (Morihara et al., shown to reduce serum levels of amyloid, a primary 2005) component of senile plaques in AD

Table 6. Summary of PHS-mediated effects in neurodegenerative diseases continued Disease/Model PHS-mediated evidence References Parkinson’s Disease (PD) - Increased expression of PHS-2 in activated (Knott et al., 2000) - PD Patients microglial cells in the substantia nigra - Unchanged neuronal and astroglial PHS-2 expression - Moderate PHS-1 immunoreactivity in neurons and glia

- PD Patients - PHS-2 is specifically induced in substantia nigra (Teismann et al., Postmortem dopaminergic neurons 2003) analysis

- PD mouse - PGE2 levels were increased in both human and (Teismann and models (MPTP) mouse tissues Ferger, 2001; rodent model Teismann et al., 2003) - PHS-2 KO mice exhibited resistance to (Feng et al., 2002) dopaminergic neuron degeneration due to MPTP

- Increased levels of PHS-2 generates toxic (Teismann and dopamine-quinone species leading to dopaminergic Ferger, 2001) neuronal degeneration in substantia nigra neurons

- Selective PHS-2 inhibitor rofecoxib and paracoxib (Reksidler et al., neuroprotective effect on tyrosine hydroxylase 2007) expression, motor and cognitive functions in MPTP- rat model

Table 6. Summary of PHS-mediated effects in neurodegenerative diseases continued Disease/Model PHS-mediated evidence References Amyotrophic lateral sclerosis (ALS) - ALS patients - PHS-2 mRNA and protein were increased in spinal (Yasojima et al., postmortem cords, localized to both neurons and glial cells 2001) analysis (Maihofner et al., - ALS patients - Increased levels of PGE2 in the CSF 2003) (Almer et al., 2002)

- Celecoxib for 12 months to ALS subjects did not (Cudkowicz et al., slow the decline in muscle strength or affect survival 2006)

- ALS mouse - PHS-2 mRNA and protein were increased in spinal (Almer et al., 2001) model (i.e. cords, localized to both neurons and glial cells transgenic expressing human - Celecoxib delayed the onset of disease, reduced (Drachman et al., mutated SOD1) spinal neurodegeneration and glial activation 2002)

- PHS-1 KO does not improve preservation of motor (Almer et al., 2006) neurons and survival of transgenic mutant mice

- Selective PHS-2 inhibitors (e.g. celecoxib and (Klivenyi et al., rofecoxib) improve motor performance, extend 2004) survival and reduce CSF levels of PGE2

- Nimesulide decreased spinal cord PGE2 levels and (Pompl et al., 2003) delayed the onset of ALS type motor impairment in mice - Nimesulide did not affect the onset of end-stage disease

1.4 NUCLEAR FACTOR ERYTHROID 2-RELATED FACTORS (NRFs)

1.4.1 OVERVIEW OF NRF1, NRF2 AND NRF3

Cap‟n‟collar (CNC) proteins form a family of basic leucine zipper (bZip) transcription factors conserved across species from worms to humans. Most CNC factors are transcriptional activators as they contain a conserved 43–amino acid CNC domain usually located N-terminally to a DNA binding domain. CNC transcription factors comprise the Caenorhabditis elegans

SKN-1 (Skinhead family member 1) (Bowerman et al., 1992), the Drosophila CNC (Mohler et al., 1991), and four related vertebrate counterparts. The latter include the p45-nuclear factor erythroid-derived 2 (NF-E2) and the NF-E2-related factors Nrf1, Nrf2, and Nrf3 (Chan et al.,

1995; Chan et al., 1993a; Chan et al., 1993b; Kobayashi et al., 1999; Kobayashi et al., 2002;

Moi et al., 1994; Chan et al., 1998). These factors were originally named from studies assessing activation of β-globin gene expression through characterization of the regulatory locus control region. This regulatory region contained a tandem AP-1-NF-E2 motif, which had strong enhancer activity in erythroid cells. Investigation of the transcription factors that bind to this

AP-1-NF-E2 site led to the discovery of the above CNC and bZip transcription factors. While p45-NF-E2 was localized in erythroid cells, Nrf1 and Nrf2 were expressed in many tissues

(discussed below) (Chan et al., 1995; Itoh et al., 1995; Moi et al., 1994). The related vertebrate transcription factors Breakpoint cluster region/Abelson murine leukemia viral oncogene homolog 1 (Bach1) and Bach2 are characterized by the additional presence of a Broad complex,

Tramtrack, Bric-a-brac (BTB) protein interaction domain along with CNC homology 1; however, Bach 1 and Bach2, function as transcriptional repressors (Oyake et al., 1996;

Dhakshinamoorthy et al., 2005).

1.4.2 MECHANISM OF ACTION OF NRF2

The Nrf transcription factors have both physiological and stress response functions depending on the cellular environment, and are important signaling pathways that can detect oxidative stress and initiate protective mechanisms (Kobayashi et al., 2002; Lee et al., 2005).

Nrf2 is part of a vital network in protective cellular responses in many tissues. In the absence of oxidative stress, Nrf2 binds to Keap1 (Kelch-like erythroid cell-derived protein with CNC homolog (ECH)-associated protein 1) which keeps Nrf2 in the cytosol and hence inactive (Itoh et al., 1999b) (Figure 11). The Nrf2-Keap1 molecule is also bound to an ubiquitin ligase complex which promotes proteasomal degradation of Nrf2. In response to oxidative stress, Nrf2 is released from Keap1. Another protein in the cytosol called DJ-1 (also known as PARK7

(Parkinson disease (autosomal recessive, early onset) 7), is thought to stabilize Nrf2 and prevent its association with its inhibitor Keap1, thereby Nrf2 translocates to the nucleus. In the nucleus,

Nrf2 can dimerize with other bZIP proteins such as Maf (named for the avian musculoaponeurotic fibrosarcoma oncogene) (Alam et al., 1999). Maf proteins possess the characteristic bZip domain but lack a transactivation domain (Motohashi et al., 2000). Nrf2 factors do not homodimerize, but Nrf2-Maf protein heterodimers and Nrf2-Jun protein complexes can to function to modulate transcription when they bind to the antioxidant response element (ARE) (also known as the electrophile response element (EpRE)), an enhancer sequence on DNA that regulates the transcription of cytoprotective components of the cell

(Alam et al., 1999; Itoh et al., 1999b; Kobayashi et al., 2002; Lee et al., 2005; Motohashi and

Yamamoto, 2004; Venugopal and Jaiswal, 1996; Venugopal and Jaiswal, 1998). In addition to

Nrf2, Nrf1 and Nrf3 also regulate transcription of ARE-containing genes, however evidence from binding to the ARE suggests Nrf2 is more efficient than Nrf1 or Nrf3 (Kobayashi et al., 73

2002; Venugopal and Jaiswal, 1998). Ultimately, the binding of Nrfs depends on the presence of various nuclear proteins/cofactors that aid or repress binding to the ARE (discussed below). 74

Figure 11. Mechanism of Nrf2 in Cytoprotection. Nrf2 is frequently activated during various types of oxidative stress. Under normal conditions, Nrf2 is located in the cytoplasm as an inactive form associated with its repressor protein Keap1. The Nrf2-Keap1 molecule is also bound to an ubiquitin ligase complex which promotes proteasomal degradation of Nrf2. Oxidation of redox-sensitive cysteines in Keap1, during oxidative stress, leads to dissociation of Nrf2 from Keap1. DJ-1 is thought to stabilize Nrf2 and prevent its association with its inhibitor Keap1, thereby preventing the ubiquitination of Nrf2. This allows Nrf2 to translocate to the nucleus where it heterodimerizes with members of the small Maf protein family and binds to the antioxidant response element (ARE) that regulates the transcription of cytoprotective enzymes such as heme oxygenase (HO-1), NAD(P)H:quinone oxidoreductase (NQO1), oxoguanine glycosylase 1 (OGG1), glutathione-S-transferase (GST) and multidrug resistance-associated protein 1 (MRP1) among others listed in Table 10. 75

1.4.3 GENETICS OF NRF2

The Nrfs share many structural and functional similarities but are encoded by different genes. Chromosomal localization of the human Nrf2 gene (official symbol NFE2L2) maps to

2q31, Nrf1 (official symbol NFE2L1) maps to 17q21-3, Nrf3 (official symbol NFE2L3) to

7p15.2 and p45-NF-E2 (official symbol NFE2) to 12q13 (Chan et al., 1998; Kobayashi et al.,

1999) (NCBI genome data base). In mice, Nrf2 maps to chromosome 2, Nrf1 maps to chromosome 11, Nrf3 maps to chromosome 6 and p45-NF-E2 chromosome 15. These genes map close to the homeobox (hox) A-D gene cluster suggesting a single ancestral gene for the

CNC family members that may have diverged to give rise to the four closely related CNC factors (Chan et al., 1995; Kobayashi et al., 1999).

The Nrf2 gene consists of five exons with four introns (Chan et al., 1996). The sequence of the first exon of the human Nrf2 gene shows 70% homology with that of the mouse Nrf2 gene, with the first intron being approximately 25 kb (Yamamoto et al., 2004). Alternatively, in humans, the Nrf1 gene spans 15 kb and has nine exons with two polyadenylation sites (Luna et al., 1994). Alternative first exons, differential splicing, and alternate polyadenylation sites give rise to at least four different transcripts of Nrf1 (Luna et al., 1994; Luna et al., 1995). The promoter regions of mouse and human Nrf2 have been analyzed within the first 0.5 to 2 kb pair region of exon 1 (Chan et al., 1996; Kwak et al., 2002; Mahboubi et al., 1998; Miao et al.,

2005; Yamamoto et al., 2004). In the mouse promoter of Nrf2, within the first 500 bp upstream of exon 1, the promoter is GC rich with Sp1 and AP-2 binding sites but contains no TATA box nor a CCAAT box (Chan et al., 1996). Further analysis in showed the Nrf2 promoter contains two ARE sequences (Kwak et al., 2002). The genes encoding Keap1, Nrf2 and MafG (a small

Maf protein) also contain the ARE, and therefore their expression is stimulated by Nrf2 76

(Katsuoka et al., 2005a; Kwak et al., 2002; Lee et al., 2007). Studies have also determined the presence of xenobiotic response elements (XREs) (Miao et al., 2005). The XRE is the ultimate target of a protein complex that includes the ligand-activated aryl hydrocarbon receptor (AHR) that translocates to the nucleus after binding to a chaperone nuclear transporter (Rushmore and

Kong, 2002). Interestingly, the human Nrf2 promoter contains 5 copies of XRE-like elements in the 2-kb region of the promoter compared to only 3 copies in rodents (Miao et al., 2005).

Multiple single nucleotide polymorphisms exist in the promoter of human NRF2 (discussed later), and one of these (−617 C/A) significantly reduces gene expression (Marzec et al., 2007).

1.4.4 PROTEIN STRUCTURE OF NRF2

Human and mouse Nrf2 proteins contain 605 and 597 amino acids respectively. At the amino acid level they have 80% homology. There are six highly conserved domains called

Nrf2-ECH homology (Neh)1 to Neh6 in Nrf2 (Itoh et al., 1999a). Their function and presence in the various CNC-related proteins are summarized in Table 7. The first conserved domain,

Neh1, contains the CNC homology region and bZip domain, which are highly conserved in the

CNC family. Neh1 is located in the C-terminal half of the molecule. The Neh2 domain is located in the proximal N terminus. The Neh2 domain of Nrf2 binds with the Kelch domain of

Keap1 in an inhibitory interaction (Kobayashi et al., 2002), but Nrf1 with a Neh2-like domain is not regulated by Keap1 (Wang and Chan, 2006; Zhang et al., 2006b). Furthermore, the Neh2 domain is not conserved in Nrf3 (Chenais et al., 2005).

Next to Neh2 are the Neh4 and Neh5 regions, which are transcriptional activation domains. The Neh4 and Neh5 domains act cooperatively to recruit cAMP Responsive Element

Binding protein (CREB)-Binding Protein (CBP) to ARE-regulated genes (Katoh et al., 2001). 77

Within the central part of Nrf2 lies the Neh6 domain, which functions in the destabilization of

Nrf2 only under conditions of oxidative stress (McMahon et al., 2004). The C terminus of Nrf2 contains the Neh3 domain, which is highly conserved between species and is important for the transcriptional activity of the protein (Nioi et al., 2005). Deletion of the final 16 amino acids of

Nrf2 gives rise to a molecule that is transcriptionally silent but translocates normally to the nucleus and binds DNA. The function of the Neh3 domain is dependent upon the VFLVPK motif, which is conserved across species and among other members of the CNC-bZIP family, including Nrf1, Nrf3 and NF-45 (Nioi et al., 2005).

Nrf2 is mostly localized to the cytoplasm and upon activation translocates to the nucleus.

Present in the protein domain are both nuclear localization sequence (NLS) and nuclear export signal (NES) motifs (Jain and Jaiswal, 2006). Nrf2 is also partly associated with the mitochondria under basal conditions. Conversely, Nrf1 and Nrf3 are integral membrane proteins targeted to the endoplasmic reticulum through a conserved N-terminal homology box 1

(NHB1) domain (Wang and Chan, 2006; Zhang et al., 2009). Nrf2 is a more potent transcriptional activator than either Nrf1 or Nrf3 (Kobayashi et al., 1999; Venugopal and

Jaiswal, 1998), which may relate to the membrane association of Nrf1 and Nrf3. Also, as seen in Table 7, p45 and Nrf1 contain the transcription activation region Neh5 but not Neh4. Nrf3 does not contain regions homologous to either Neh4 or Neh5 (Katoh et al., 2001). Nrf2 has both motifs, which may increase its transcriptional activity (Katoh et al., 2001).

Posttranslational modifications, including phosphorylation, acetylation and regulated cleavage, can influence Nrf2 activity. The posttranslational phosphorylation of Nrf2 by a tyrosine kinase at Ser-40 can disrupt the association of Nrf2 with Keap1 (Huang et al., 2002), while phosphorylation at Tyr-568 may be required for the nuclear export of the transcription 78

factor (Jain and Jaiswal, 2006). Acetylation of Nrf2 by its transcriptional coactivator CBP promotes Nrf2 DNA binding to ARE promoters (Sun et al., 2009). Also, caspases can remove the N-terminal transactivation domain and convert Nrf2 to an apoptosis-promoting ARE repressor (Ohtsubo et al., 1999). Modulation of Nrf2 activity will be discussed below. 79

Table 7. Nrf domains and function (Cullinan et al., 2004; Furukawa and Xiong, 2005; Itoh et al., 1995; Itoh et al., 1999a; Katoh et al., 2001; McMahon et al., 2004; Nioi et al., 2005) Protein p45- Function Nrf2 Nrf1 Nrf3 Domain NF - bZip region fused to a CNC region - Dimerizes with small MAF proteins Neh1 - Binds DNA as a heterodimer + + + + - Nuclear localization sequence - Nuclear export signal - Negative regulatory domain of Nrf2 function - Interaction with Keap1 - Contains double glycine repeat (DLG) and a peptide sequence (ETGE) motif important for Neh2 Keap1 binding + + - Contains lysines between DLG and ETGE that may be ubiquitinated - Contains a peptide sequence (DIDLID) element (influences degradation) - Interacts with chromodomain helicase DNA- Neh3 binding protein 6 (CHD6) + + + + - May be involved with transcription - Transactivation domain Neh4 - Interacts with CREB-binding protein (CBP) + - Transactivation domain Neh5 - Interacts with CBP + + + - Nuclear export signal - Destabilization of Nrf2 under conditions of Neh6 oxidative stress +

1.4.5 REGULATORS OF NRF2

As seen from its various protein domains, Nrf2 has many different components that can regulate its transcriptional activity. Nrf2 interactions with Keap1 and its degradation by the proteasome are important in the cytosolic compartment where Nrf2 levels are kept low, with a half-life of 10-30 min under physiological conditions (Alam et al., 2003; Furukawa and Xiong,

2005; He et al., 2006). In the nucleus, Nrf2 can interact with a number of cofactors forming a transcriptional complex that binds to the ARE and modulates transcription.

1.4.5.1 Negative regulation by Kelch-like ECH-associated protein 1 (Keap1)

Nrf2 is localized mainly in the cytoplasm bound to a repressor, KIAA0132 (also called iNrf2, the human homolog to Keap1 (Itoh et al., 1999a). Mammalian forms of Keap1 are metalloproteins having 624 amino acids. Mouse Keap1 shares approximately 94% amino acid sequence homology to its iNrf2 human homolog and are highly conserved between species (Itoh et al., 1999a). Keapl is homologous to the Kelch protein that binds actin in Drosophila (Itoh et al., 1999a). Starting at the N-terminus, the Keap1 protein consists of five domains including the

N-terminal region (NTR), the BTB dimerization domain, the intervening or linker region (IVR), the double glycine repeats (DGR) or Kelch repeats region in the C-terminal region.

The BTB region is involved in forming homodimers with other Keap1 proteins, as well as targeting Nrf2 for ubiquitination (Cullinan et al., 2004; Furukawa and Xiong, 2005; Zipper and Mulcahy, 2002). The IVR has a high cysteine content and serves as a redox sensor, reacting with electrophilic reactive intermediates to form adducts, or reacting with ROS to become oxidized (Dinkova-Kostova et al., 2002; Wakabayashi et al., 2004). Either reaction leads to conformational changes that reduce the interaction between Keap1 and Nrf2. Human Keap1 81

contains 27 cysteine residues, 25 of which are highly conserved across species including mice

(Itoh et al., 1999b). The DRG repeats, comprising six double glycine repeats, are necessary for the interaction with the Neh2 domain of Nrf2 (i.e. motifs DLG and ETGE of the Neh2 of Nrf2), and also interacts with actin in the cytosol (Itoh et al., 1999b; Kang et al., 2004; Tong et al.,

2006; Zipper and Mulcahy, 2002). These domains of Keap1 interact through various functions

(Table 8) to keep Nrf2 sequestered in the cytosol and inactive in the absence of chemical or pathophysiological stress. In fact, overexpression of Keap1 reduces Nrf2-mediated activation of

ARE-regulated genes (Itoh et al., 1999b; Wakabayashi et al., 2004). In addition, deletion of the

DRG, IVR or the C-terminal region abolished the ability of Keap1 to repress the activity of Nrf2

(Kang et al., 2004). Other studies have also shown that Keap1 has a nuclear export signal that is required for termination of Nrf2-ARE signaling by escorting the nuclear export of Nrf2 (Sun et al., 2007). However, it is still unknown how Keap1 enters the nucleus, as it has no known nuclear localization sequence. 82

Table 8. Regulatory mechanisms provided by Keap1 Selective Evidence for Keap1 Functions References Directs Nrf2 for Ubiquitination Ubiquitinated Nrf2 has been detected (Cullinan et al., 2004; Kobayashi et al., 2004; Zhang and Hannink, 2003) Critical cysteine residues in Keap1, C273 and C288, C151 are required for Keap1-dependent ubiquitination of Nrf2 (Zhang and Hannink, 2003)

Keap1 (BTB domain) functions as a substrate adaptor for a Cullin (Cullinan et al., 2004; (Cul)-dependent E3 ubiquitin ligase complex i.e. Kobayashi et al., 2004; Zhang immunoprecipitation of Keap1 with Cul3 and Hannink, 2003)

Inhibiton of cul 3 increases basal expression of Nrf2 (Furukawa and Xiong, 2005; Cullinan et al., 2004; Zhang and Hannink, 2003) Mutation of the lysine residues located between DLG and ETGE (Zhang et al., 2004) motifs in the Neh2 domain of Nrf2 inhibits Keap1-directed ubiquitination

Redox Sensor Cys-151 important during oxidative stress conditions, oxidation (Zhang and Hannink, 2003; leads to Nrf2 activation Zhang et al., 2004)

Mutation of other cysteines within the IVR, N-terminal and C- (Wakabayashi et al., 2004; terminal domains has limited effect on Keap1 function in vitro Zhang and Hannink, 2003)

Detection of protein disulfides. The most reactive residues of (Dinkova-Kostova et al., Keap1 C-257, C-273, C-288, and C-297 in the IVR 2002)

Interactions with cellular components Deletion of DGR of Keap1 results in no binding to Nrf2 (Kang et al., 2004)

Mutations of serine 104 (BTB domain) led to the disruption of the (Zipper and Mulcahy, 2002) human keap1 homodimers and lead to the release of Nrf2

Interactions with PGAM5 (a member of the phosphoglycerate (Lo and Hannink, 2008) mutase family) forms ternary complex containing Keap1 and Nrf2 that is localized to mitochondria. Mediates Nrf2 response to changes in mitochondrial functions

Scaffolding of Keap1 to the actin cytoskeleton to maintain Nrf2 in (Kang et al., 2004) cytosol. Disruption of the actin cytoskeleton promotes nuclear entry of Nrf2 83

1.4.5.2 Negative regulation by proteasome degradation

Under normal conditions, Keap1, through its BTB domain, anchors the Nrf2 transcription factor within the cytoplasm, targeting it for ubiquitination and proteasomal degradation to maintain low levels of Nrf2 (Furukawa and Xiong, 2005). The integrity of the proteasomal system therefore also plays an important role in regulation of Nrf2 transcriptional activity. Keap1 protein functions as a link between Nrf2 and the Cullin3-based E3-ubiquitin ligase ubiquitination complex (Cullinan et al., 2004; Kobayashi et al., 2004; Zhang et al., 2004).

The Neh2 domain of Nrf2 represents the region through which the repressor protein Keap1 recognizes and targets lysines in the Neh2 domain of Nrf2 to the Cullin3-based E3-ubiquitin ligase for ubiquitination and subsequent degradation by the 26S proteasome. These lysines are located between the two Kelch-binding sites on Neh2, DLG and ETGE, and a model has been proposed whereby binding of a Keap1 homodimer to these two sites allows for ubiquitination to occur. As shown in Table 8, another motif in Neh2 called the DIDLID element, which is a peptide sequence, may aid in the recruitment of ubiquitin ligase, is required for the rapid turnover of Nrf2 under normal homeostatic conditions (McMahon et al., 2004).

Studies have also shown that, even in the absence of a Keap1 interaction, Nrf2 can be targeted for proteasomal degradation (McMahon et al., 2004). Under conditions of oxidative stress, the Neh2 directs a less rapid, Keap1-independent degradation of the bZip factor.

Degradation of Nrf2 in stressed cells is predominantly mediated by the redox-insensitive Neh6 domain of Nrf2 (McMahon et al., 2004). Another protein in the cytosol called DJ-1, which has been linked to PD, is also important for the stabilization of Nrf2. Proper function of DJ-1 is necessary for the proper induction of NQO1. Loss of DJ-1 leads to deficits in NQO1 expression.

DJ-1 is thought to stabilize Nrf2 and prevent its association with its inhibitor Keap1, thereby 84

preventing the ubiquitination of Nrf2 (Clements et al., 2006). Without intact DJ-1, the Nrf2 protein is unstable and transcriptional responses may be reduced. However, the Nrf2 pathway can still be activated by tertiary butylhydroquinone (tBHQ), a potent Nrf2 activator, in DJ-1- deficient cells (Gan et al., 2010).

1.4.5.3 Regulation of transcriptional complex in nucleus

In response to oxidative stress, Nrf2 accumulates in the nucleus, where it binds nuclear proteins that can also modulate its activity. Nrf2 activates transcription primarily as a dimer with members of the small Maf proteins, which act as coactivators (Itoh et al., 1997). Different cell types may have varying amounts of these coactivators and repressors, which provides another layer of regulation to the system and may result in cell-specific responses to oxidative stress. Gene transcription is regulated by the balance between activation and repression mechanisms in response to stimuli. The Nrf2-Maf heterodimer is the primary transcriptional complex regulating ARE-dependent gene expression; however, with excessive Maf expression, homodimers can form which will reduce Nrf2-Maf complexes and decrease ARE activation

(Katsuoka et al., 2005a; Katsuoka et al., 2005b; Motohashi et al., 2000). Transcription factors, such as Jun, c-Fos, FRA-1, FRA-2 and Nrf1, also can interact with the ARE, thereby competing with Nrf2 (Jeyapaul and Jaiswal, 2000; Katsuoka et al., 2005a; Katsuoka et al., 2005b;

Venugopal and Jaiswal, 1996; Venugopal and Jaiswal, 1998). Other transcription factors, including ATF, estrogen receptor (ER), PPAR and retinoic acid receptor (RAR) may form inhibitory complexes with Nrf2. Transcription factors can inhibit Nrf2 actions by either competing for binding to AREs or by inhibiting Nrf2 through physical complexes. The various transcription factors and their mode of interaction with Nrf2 in the nucleus are described in 85

Table 9. Nrf2 interactions in the nucleus Cofactor Function Reference Small Maf - Nrf2-Maf dimers have been proposed to function (Dhakshinamoorthy proteins: as both positive regulators and Jaiswal, 2000; MafF, MafG - Negative regulators of ARE-dependent gene Katsuoka et al., 2005a; and MafK transcription i.e. MafK Katsuoka et al., 2005b; (also referred - May form homodimers with each other to repress Marini et al., 1997; to as p18) ARE binding of Nrf2 Wild et al., 1999) - Complexed with Nrf2, transcriptional coactivator (Sun et al., 2009) CBP - Acetylation of Nrf2 promotes its DNA binding to ARE - When phosphorylated can enter the nucleus to (Jain and Jaiswal, 2006; Fyn phosphorylate Nrf2 on Tyr-568 to promote nuclear Jain and Jaiswal, 2007; export of Nrf2 Sun et al., 2009) Fos proteins: - Downregulation of human and mouse ARE- (Venugopal and c-Fos, FosB, mediated transcription Jaiswal, 1996; Fra-1, and Yoshioka et al., 1995) Fra-2 - Possible combinations were ineffective in (Tsuji, 2005; repression or upregulation of hARE-mediated gene Venugopal and Jaiswal, expression. 1996; Venugopal and - Jun-Nrf2 complexes have also been implicated as Jaiswal, 1998) Jun proteins: positive effectors of ARE-dependent genes c-Jun, Jun-B, - Role of c-Jun phosphorylation in activation and Jun-D - Heterodimerization and binding of Nrf-Jun proteins require unknown cytosolic factors - Phosphorylation of JunD at Ser-100, an activated form of JunD, is an ARE regulatory protein - Transcription repressor that competes with Nrf2 (Dhakshinamoorthy et - Binds to Mafs to repress gene activation al., 2005; Igarashi et Bach1 - Increased nuclear export during oxidative stress al., 1998; Kaspar and - Phosphorylation of tyrosine 486 leads to rapid Jaiswal, 2010; Suzuki nuclear export of Bach1 et al., 2003) - Oxidative stress induces nuclear accumulation (Hoshino et al., 2000; Bach2 - Suppresses ARE activity and promotes apoptosis Muto et al., 1998; Muto - Interferes with Nrf2-Maf recognition et al., 2002) - Antagonizes NRF2, but in this instance it entails (Liu et al., 2008) NFKB competition for CBP and recruitment of histone deacetylase to MafK - Chromatin remodeling factor (Zhang et al., 2006a) BRG1 - Interacts with Nrf2 - Enhances ARE activity - Transcriptional repressor (Ki et al., 2005) SMRT - Silences ARE activity possibly through binding Nrf2 protein 86

Table 9. Nrf2 interactions in the nucleus continued Cofactor Function Reference ATF1 (Iwasaki et al., 2007) - Transcriptional repressor of the ferritin H ARE - Blocks Nrf2 binding ATF3 - Can repress Nrf2-mediated signaling (Brown et al., 2008) ATF - Direct ATF3-Nrf2 protein-protein interactions that result in displacement of CBP from the ARE ATF4 (He et al., 2001) - Nrf2-interacting protein possibly with other cofactor proteins increase ARE activation In macrophages: (Ikeda et al., 2000) PPAR gamma PPAR-nrf2 interactions may suppress ARE binding - Estrogen-bound ERα, but not ERβ, is able to repress Nrf2-mediated transcription - Physical interaction between ERα and Nrf2 (Zhou et al., 2007) ER - SFhERRbeta repressed Nrf2 activity possibly through physical interaction in a complex with Nrf2, not by competing for the ARE DNA-binding sites

- Reduces the ability of Nrf2 to mediate induction (Wang et al., 2007b) of ARE-driven genes RAR - Evidence of Nrf2 forming a complex with RARalpha therefore preventing binding to ARE Abbreviations. ARE, antioxidant response element; ATF, activating transcription factor; Bach, Breakpoint cluster region/Abelson murine leukemia viral oncogene homolog; BRG1, Brahma-related gene 1; CBP, (CREB (cAMP Responsive Element Binding protein) Binding Protein); ER, estrogen receptor; Maf, musculoaponeurotic fibrosarcoma; NFKB, nuclear factor kappa B; PPAR gamma, peroxisome proliferator-activated receptor gamma; RAR, retinoic acid receptor; SFhERRbeta, estrogen-related receptor beta (ERR)-beta--short-form; SMRT, silencing mediator for retinoid and thyroid hormone receptors. 87

1.4.6 ANTIOXIDANT RESPONSE ELEMENT (ARE)

The ARE is a DNA sequence found within the promoter regions of numerous cytoprotective genes. The ARE core sequence, also known as the EpRE, was first identified in the promoters of Ya subunits of rat and mouse GST as 5′-GTGACnnnGC-3′, where „n‟ was used to denote any nucleotide (Rushmore et al., 1991). The AP-1 recognition site TRE (12-O- tetradecanoylphorbol-13-acetate (TPA)-responsive element) (5'-TGACTCA-3') and the

ATF/CREB binding sequence (5'-TGACGTCA-3') partially overlap with the ARE sequence

(Dalton et al., 1999; Nguyen et al., 2003). Wasserman and Fahl further characterized the ARE sequence to a “core” sequence of 5′-RTGACnnnGCR-3′ using murine GST-Ya ARE, and identified many other genes that contained the sequence in their promoters (Wasserman and

Fahl, 1997). However, subsequent mutagenesis studies identified deviations from the consensus

ARE and identification of ARE sequences may illustrate differences depending on the gene‟s promoter region and the species source (Erickson et al., 2002; Nioi et al., 2003).

Venugopal and Jaiswal (Venugopal and Jaiswal, 1996) identified a cis-element resembling the ARE sequence in 5‟-flanking regulatory region of the human NQO1 that was physically able to bind Nrf1 and Nrf2, which resulted in an increase in transactivation activity and NQO1 gene induction (Venugopal and Jaiswal, 1996). This was evaluated through reporter transgene and electrophoretic mobility shift assay experiments (Nguyen et al., 2003; Venugopal and Jaiswal, 1996). Despite being integral membrane proteins, both Nrf1 and Nrf3 can bind the

ARE as well (Kobayashi et al., 1999; Venugopal and Jaiswal, 1998). Furthermore, Nrf1 has a

65-kDa isoform that is nuclear (Wang and Chan, 2006). However, many studies have shown that Nrf2 is more efficient at transactivation possibly due to its extra activation domain (Biswas 88

and Chan, 2010; Kobayashi et al., 1999; Venugopal and Jaiswal, 1998). Evidence in vivo was subsequently provided by Itoh and coworkers (1997) who observed an impaired constitutive and butylated hydroxyanisole (BHA)-induced expression of the phase II GSTs Ya and Yb in Nrf2- disrupted mice (Itoh et al., 1997). The Nrf2 promoter also contains two ARE-like sequences located at -492 and -754 from the start codon (Kwak et al., 2002). One motif is described as the perfect ARE (AREL1; TGACTccGC) consensus sequence, while the second one is described as imperfect with one more base before the GC box (AREL2; TGACTgtgGC) (Kwak et al., 2002).

Cooperatively, both AREL1 and AREL2 are necessary to fully activate Nrf2 expression by Nrf2 chemical activators (discussed below) (Kwak et al., 2002). Functional XRE and ARE sequences sometimes exist in proximity in promoters of several important detoxifying genes including GST, NQO1 and Nrf2 itself (Favreau and Pickett, 1991; Rushmore et al., 1990). The

XRE motif is the ultimate target of a protein complex that includes a ligand-activated AHR translocating to the nucleus after binding to a chaperone nuclear transporter (Rushmore and

Kong, 2002). This provides added regulation to detoxifying genes along with the ARE sequences. It is necessary to find functional AREs in the promoter regions of genes in order to seek out genes directly regulated by the Nrf2-ARE pathway and determine how other sequences, such as XRE, interact with them. Common genes with the ARE sequences that are regulated by

Nrf2 activation are listed in Table 10. These include many phase 2 detoxification enzymes and especially those regenerating reduced glutathione. However, a diverse set of genes involved in antioxidation, detoxification and repair are also transcriptionally regulated by Nrf2 (Table 10). 89

Table 10. Examples of genes containing AREs in their promoter region Gene Species References NRF2 (Transcription factor) mouse, human (Kwak et al., 2002) KEAP1 (Negative regulator of Nrf2) mouse, human (Lee et al., 2007) HO-1 (Catabolizes heme to yield biliverdin) mouse, rat (Alam et al., 2003; Alam et human al., 1999; Liby et al., 2005) NQO1 (Catalyzes two-electron reduction and (Itoh et al., 1997; Nioi et al., human detoxification of quinones) 2003; Venugopal and mouse Jaiswal, 1996) OGG1 (DNA repair) human (Dhenaut et al., 2001) MafG (Transcription factor) mouse, human (Katsuoka et al., 2005a) SOD1 (Catalyzes the dismutation of (Park and Rho, 2002) human superoxide radicals) AhR (XRE regulation) mouse (Shin et al., 2007) GST Ya (catalyzing GSH conjugation to (Rushmore et al., 1990; electrophiles) mouse, rat Rushmore et al., 1991; Wasserman and Fahl, 1997) GCS (Enzyme involved in GSH synthesis) (Kwak et al., 2003c; mouse Moinova and Mulcahy, human 1998) GCLg (Catalyzes the conjugation of cysteine (Erickson et al., 2002; Yueh human with l-glutamate, involved in GSH synthesis) and Tukey, 2007) UGT1A1 (Catalyzing phase 2 conjugation) mouse (Yueh and Tukey, 2007) human GPx (Catalyze the reduction of H2O2 using (Banning et al., 2005) human GSH) MRP1 (Transporter) (Hayashi et al., 2003; mouse Kauffmann et al., 2002; human Maher et al., 2007) MRP2 (Transporter) (Kauffmann et al., 2002; mouse Maher et al., 2007; Vollrath human et al., 2006) MRP3 (Transporter) mouse (Maher et al., 2007) MRP4 (Transporter) mouse (Maher et al., 2007) Ferritin H mouse (Pietsch et al., 2003; Tsuji, human 2005) ETS 1 (Transcription factor) human (Wilson et al., 2005) Prx1 (Reduces H2O2) human (Kim et al., 2007) PSMB5 (Proteasome factory) mouse (Kwak et al., 2003b) Trx (Catalyzes reduction of disulfides to (Kim et al., 2001) human sulfhydryls) (Sakurai et al., 2005) MEH (catalyzes epoxides to polar diols) mouse (Kwak et al., 2001) Abbreviations. AhR, arylhydrocarbon receptor; ETS, E-twenty six; GCLg, glutamate cysteine ligase; GCS, gamma-glutamylcysteine synthetase regulatory subunit; GPx, glutathione 90

peroxidase; GST, glutathione S-transferase; MEH, microsomal epoxide hydrolase; MRP, multidrug resistance protein; NQO1, NADPH quinone oxidoreductase 1; OGG1, oxoguanine glycosylase 1; Prx1, peroxiredoxin 1; TRx1, thioredoxin reductase; Trx, thioredoxin; UGT, UDP-glucuronosyltransferases. 91

1.4.7 ACTIVATORS OF NRF2

Oxidative stress is induced by a wide range of factors including drugs, heavy metals and ionizing radiation, as well as by endogenous compounds (Table 11). Oxidative stress leads to the generation of ROS and reactive electrophiles, which can be toxic to cells if basal cytoprotective mechanisms are overwhelmed. These compounds may activate Nrf2 and thereby enhance the protective mechanisms discussed above that protect against toxicity initiated by both electrophilic reactive intermediates and ROS. The Nrf2 transcription system is activated by a wide variety of natural and synthetic chemical compounds. As discussed above, under normal cellular conditions, the cytosolic Keap1-Nrf2 complex is constantly degrading Nrf2.

When a cell is exposed to oxidative stress, Nrf2 dissociates from the Keap1 complex, stabilizes and translocates into the nucleus, leading to activation of ARE-mediated gene expression.

Questions remain as to how Nrf2 is transcriptionally activated by such diverse chemical compounds and agents. The diversity may be related to the possibility of interactions with both

Nrf2 and Keap1.

By far the most studied regulatory mechanism of Nrf2 activation is its interaction with

Keap1. Keap1 has a number of reactive cysteines which may play an important role in sensing oxidative stress and then responding by either the release of Nrf2 or preventing Nrf2 degradation by the Keap1-mediated proteasomal degradation pathway (Kobayashi et al., 2006).

Three key cysteine residues (C151, C273, and C288) were identified by both in vitro alkylation and in vivo site-directed mutagenesis assays to be important in redox sensing (Dinkova-Kostova et al., 2002; Hong et al., 2005). Mutation of C151 completely abolished induction of Nrf2 by many Nrf2 activators, such as sulforaphane and tBHQ (Zhang and Hannink, 2003). Those chemical activators that are strong electrophiles can interact with Keap1 cysteines through direct 92

alkylation or arylation, while free radicals and ROS generation can oxidize the cysteine residues resulting in the formation of disulphide bridges (Kobayashi et al., 2006; Zhang and Hannink,

2003). Talalay and coworkers found many inducers of environmental stress response genes that may act through Nrf2 activation, which they classified into 10 chemically distinct classes: (a) oxidizable diphenols, phenylenediamines, and quinones; (b) Michael acceptors (olefins or acetylenes conjugated to electron-withdrawing groups); (c) isothiocyanates; (d) thiocarbamates;

(e) trivalent arsenicals; (f) dithiolethiones; (g) hydroperoxides; (h) vicinal dimercaptans; (i) heavy metals; and (j) polyenes (Dinkova-Kostova et al., 2005). Because of the 25 conserved cysteine residues, it is possible that the sites of adducts or oxidative modification may vary among the different classes of inducers (Dinkova-Kostova et al., 2005). Electrophiles disrupt the interaction between Keap1 and the Neh2 domain of Nrf2 either by inducing phosphorylation of Nrf2 or by directly modifying cysteine residues in Keap1 (Huang et al., 2002; Wakabayashi et al., 2004). Specifically in the brain, electrophilic neurite outgrowth-promoting prostaglandin

(NEPP) compounds are taken up preferentially into neurons and bind to Keap1. NEPPs prevent

Keap1-mediated inactivation of Nrf2 and, thus, enhance Nrf2 translocation into the nucleus of cultured neuronal cells (Satoh et al., 2006).

Another regulatory mechanism of activation of Nrf2 is phosphorylation. The MAPKs are involved in the regulation of the ARE in a Nrf2-dependent manner (Yu et al., 2000a). The extracellular signal-regulated kinases 2 and 5 (ERK2, ERK5) and JNK1 upregulated the ARE

(Keum et al., 2003; Shen et al., 2004; Yu et al., 1999), while the p38 MAPK appears to suppress it (Yu et al., 2000b). However, it is uncertain which cellular components involved in

ARE regulation may be targets of any of these kinases. Huang et al. have reported that protein kinase C (PKC) can directly phosphorylate Nrf2, and Ser-40 appears to be a site of potential 93

phosphorylation (Huang et al., 2002). Activation may involve one or more of these mechanisms converging to work together depending on the chemical, cell or tissues types and the gene of interest. Compounds like oltipraz and dimethyl fumarate (DMF) (Kappos et al., 2008; Zhang and Munday, 2008), as well as natural products like curcumin (Shen et al., 2006) and sulforaphane (Juge et al., 2007), are effective in mouse models of experimental carcinogenesis, and their chemopreventive and neuroprotective actions are reduced or abolished in Nrf2 knockouts, indicating that their activities are mediated by the induction of Nrf2 (Kappos et al.,

2008; Ramos-Gomez et al., 2001).

Table 11. Selective examples of Nrf2 activators Nrf2 Activator Cell type References Sulforaphane mouse small intestine (Shinkai et al., 2006; hepatic cells Thimmulappa et al., 2002; Intracerebral hemorrhage Zhao et al., 2007a; Zhao et brain tissue al., 2007b) blood brain barrier tissue Diethylmaleate fibroblasts (Itoh et al., 1999a) Dimethyl fumarate (DMF) mouse astrocytes (Itoh et al., 1999a; Murphy et al., 2001) tert-Butylhydroquinone human neuroblastoma cells (Lee et al., 2003a; Lee et al., human neural stem cells 2001; Li et al., 2005; Shih et primary cortical astrocytes al., 2005a; Shih et al., 2005b) mouse striatal tissue mouse cortical tissue Oltipraz Hepatic and gastric tissues (Pietsch et al., 2003; Ramos- hepatic cells and fibroblasts Gomez et al., 2001) Electrophilic neurite Neuronal cells (Satoh et al., 2006) outgrowth-promoting prostaglandin (NEPP) Acrolein Murine keratinocyte PE cells (Kwak et al., 2003a) B-NF (beta-naphthoflavone) hepatic cells and fibroblasts (Pietsch et al., 2003) D3T (1,2-dithiole-3-thione) hepatic cells and fibroblasts (Pietsch et al., 2003; Burton et mouse striatal tissue al., 2006) Curcumin renal epithelial cells (Balogun et al., 2003) Cadmium hepatic cells (Casalino et al., 2007; Stewart et al., 2003) Manganese hepatic cells (Casalino et al., 2007) Methylmercury SH-SY5Y cells and with (Toyama et al., 2007) primary mouse hepatocytes, direct binding to Keap1

1.4.8 NRF2 IN NEUROTOXICITY AND CNS DISEASES

1.4.8.1 Nrf2 expression

Nrf2 is expressed in a variety of tissues including liver, kidney, small intestines, stomach, lung and heart (Chan et al., 1996; Itoh et al., 1997; Lee et al., 2005). Studies evaluating mouse fetal RNA and embryonic stem cell RNA revealed expression of Nrf2, as early as from the blastocyst stage and throughout gestation (Chan et al., 1996). In the mouse brain, it is expressed in different regions including the striatum, substantia nigra, cortex and hippocampus (Calkins et al., 2005; Ramsey et al., 2007; Shih et al., 2005a; Shih et al., 2003;

Shih et al., 2005b). The expression of Nrf2 has been studied in human brain tissue. In the cortex, substantia nigra and hippocampus Nrf2 is expressed in neurons and astrocytes, where it is localized in the cytoplasm and the nucleus (Ramsey et al., 2007). In the motor cortex and spinal cord, Nrf2 expression was predominantly in the cytoplasm, and no nuclear labeling could be observed (Sarlette et al., 2008). Ultimately, the transcription of cytoprotective genes is determined by the activation of Nrf2, the presence of the co-activators and the promoter elements of each gene; hence the potential for regional selectivity upon Nrf2 activation.

Antioxidant enzymes regulated by Nrf2 were discussed previously and are reviewed in Table 10 and discussed in section 1.2.1.3.2.

Of importance is the role of the astrocytes in the production of GSH, which is mainly regulated by Nrf2-mediated activation of the genes for GSH synthesis. Nrf2 and ARE activation in glial cells not only prevents oxidative damage in astrocytes, but also protects neighbouring neurons, via its production and secretion of GSH and other potentially protective cofactors, from glutamate- and hydrogen peroxide-induced neuronal cell death (Calkins et al., 96

2005; Jakel et al., 2007; Kraft et al., 2004; Shih et al., 2005a). In addition, studies with the Nrf2 stabilizer DJ-1 have shown that DJ-1 knockdown in astrocytes diminishes astrocyte-mediated neuroprotection (Mullett and Hinkle, 2009). Furthermore, localization of Nrf2 in neurons, together with evidence that NEPPs prevent excitotoxicity by activating the Keap1/Nrf2/HO-1 pathway, suggest that preferential accumulation of protective compounds can occur in neurons

(Satoh et al., 2006). These processes in neurons constitute a complementary mechanism for targeting areas of neurotoxicity as an alternative to mechanisms involving electrophilic compounds such as tBHQ, which activates the antioxidant-responsive element predominantly in astrocytes.

1.4.8.2 Nrf2 in neurodegenerative diseases

Interestingly, differential regulation of Nrf2 has been detected in PD and AD (Ramsey et al., 2007). Studies assessing the localization of Nrf2 in the substantia nigra of PD brains demonstrated that, in addition to the cytoplasm, a strong nuclear immunoreactivity is observed in neurons (Ramsey et al., 2007). In the case of AD, there is a significant reduction in nuclear

Nrf2 in hippocampal neurons. Furthermore, there is less Nrf2 in nuclear fractions from the cortex in AD cases (Ramsey et al., 2007). This outcome in AD is contrary to the expected Nrf2 localization in the nucleus of neurons, which should occur in oxidative stress. Potentially, inactivation of Nrf2 by proteases during the neuropathological process may contribute to its reduced nuclear localization (Ohtsubo et al., 1999). Likewise, evidence for a role of Nrf2 and its regulation and stabilization are currently being investigated for other neurodegenerative diseases such as ALS, MS and HD (Table 12). Although the Nrf2–ARE signaling route may be activated in these diseases, factors may be operating to counteract Nrf2-activated gene 97

transcription depending on the stage of the disease (de Vries et al., 2008). For example, activation of Nrf2 that is not sufficient to inhibit the neuropathology of PD may be related to altered stabilization by DJ-1 (Clements et al., 2006). Mutations in DJ-1 are associated with autosomal recessive forms and some sporadic cases of PD (Wider and Wszolek, 2007).

Evidence is emerging that DJ-1 has antioxidant properties, through its interaction with Nrf2, and protects neurons against oxidative stress-induced cell injury (Canet-Aviles et al., 2004; Lev et al., 2009). DJ-1 is homogenously expressed in all CNS regions in both human and mouse brain

(Bader et al., 2005; van Horssen et al., 2010; van Horssen et al., 2008). It is found in neurons of different neurotransmitter subtypes and in glial cells, such as astrocytes, microglia and oligodendrocytes. It is also distributed throughout the cytoplasm of the soma and the proximal parts of the processes, with limited detection in the nuclei (Bader et al., 2005). Analyses of

Nrf2, Keap1, DJ-1 and Nrf2-regulated proteins in neurodegenerative diseases are providing insights into the potential involvement of ARE-mediated responses and their localization. This is summarized in Table 12. Protective effects of Nrf2 activation have been demonstrated in various animal models for PD (Chen et al., 2009), AD (Kanninen et al., 2009), ALS (Vargas et al., 2008) and in mouse models of HD (Calkins et al., 2005; Shih et al., 2005a). 98

Table 12. Evidence for Nrf2 activation or deregulation in neurodegenerative diseases Disease Evidence References AD - Nrf2 localizes in the cytoplasm of hippocampal neurons (Marcus et - Not a major component of β-amyloid plaques or al., 1998) neurofibrillary tangles - Nuclear Nrf2 expression level significantly decreased in AD cases (Omar et al., - Cu/Zn-SOD activity was significantly decreased in AD frontal 1999) and AD temporal cortex (Takahashi et - catalase activity was significantly decreased in AD temporal al., 2000) cortex

- Increased expression however decreased activity of many (Schipper, Nrf2-regulated antioxidant genes 2004) - Binding of APP inhibits HO activity - HO-1 immunoreactivity is significantly augmented in neurons and astrocytes of the hippocampus and cerebral cortex and colocalizes to senile plaques, neurofibrillary tangles,

- Specific localization of NQO1 staining and increased activity (SantaCruz in astrocytes and neurites surrounding senile plaques in the et al., 2004) frontal cortex of AD brains

- DJ-1 immunoreactivity was seen in a subset of neurofibrillary (Kumaran et tangles, neuropil threads, and neurites in extracellular plaques in al., 2007) AD

PD - NQO1 expressed in astroglial and endothelial cells and less (van frequently in dopaminergic neurons in PD brains Muiswinkel et al., 2004) - HO-1 in Lewy bodies of affected dopaminergic neurons (Schipper, - Highly overexpressed in astrocytes within the substantia nigra 2004)

- Localization of Nrf2 in the substantia nigra of PD brain in (Ramsey et cytoplasm and nuclear immunoreactivity is observed in neurons al., 2007)

- Defects in DJ-1 linked to PD and may destabilize Nrf2 (Clements et - Mutations in DJ-1 are associated with autosomal recessive al., 2006; forms and some sporadic cases of Parkinson disease Wider and Wszolek, 2007) 99

Table 12. Evidence for Nrf2 activation or deregulation in neurodegenerative diseases continued Disease Evidence References MS - HO-1 is upregulated in glial cells within multiple sclerosis (Schipper et plaques in spinal cord al., 2009) - Nrf2 is upregulated in active MS lesions, in both the nucleus (van Horssen and the cytoplasm of infiltrating macrophages et al., 2010) - Nrf2 upregulated to a lesser extent in reactive astrocytes - Cytoplasmic and nuclear staining of Nrf2 in macrophages and astrocytes - Oligodendrocytes at the edge of MS lesions expressed relatively low levels of Nrf2 - DJ-1 protein expression is predominantly increased in (van Horssen astrocytes in both active and chronic inactive MS lesions et al., 2008) - SOD 1 and 2, catalase, and HO-1, are upregulated in active (Hirotani et demyelinating MS lesions especially in hypertrophic astrocytes al., 2008) and myelin-laden macrophages - DJ-1 levels are increased in cerebral spinal fluid of patients with relapsing–remitting MS HD - HO-1 immunopositivity were enhanced in HD brain sections (Browne et - Activities of erythrocyte Cu/Zn-SOD and GPx reduced in 16 al., 1999) HD patients

ALS - Reduced mRNA expression of Nrf2 throughout the cortical (Sarlette et layers in ALS tissues al., 2008) - Reduced cellular Nrf2 protein expression in both the ventral horn and the primary motor cortex in ALS tissue. - Most of the Nrf2 signal was detected in pyramidal cells - Keap1 protein expression in motor cortex and spinal cord was not different from controls

Abbreviations. AD, Alzheimer's disease; ALS, amyotrophic lateral sclerosis; APP, amyloid precursor protein; Cu/Zn-SOD, Cu/Zn-superoxide dismutase; GPx, glutathione peroxidase; HD, Huntington's disease; HO-1, heme oxygenase-1; MS, multiple sclerosis; NQO1, NAD(P)H:quinone oxidoreductase; PD, Parkinson's disease; TBARS, thiobarbituric acid- reactive substances. 100

1.4.8.3 Nrf2 in chemical-induced neurotoxicities

Whether activation is predominantly neuronal or in astrocytes may depend of the origin of the pathology and/or the site of increased ROS generation, kinase activation or electrophilic activation of Nrf2. For example, neurotoxins such as METH and MPTP that localize and generate the majority of ROS in striatal dopaminergic nerve terminals and cell bodies of the substantia nigra would be expected to generate more neurotoxicity, and hence a greater Nrf2- mediated response. Cytoprotective effects of Nrf2-induced protective mechanisms have been seen in many CNS cell types and animal models for neurodegeneration. Nrf2-driven antioxidant enzymes are able to protect primary astrocytes from hydrogen peroxide-induced apoptosis as well as glutamate toxicity, with protection afforded by pretreatment with the Nrf2 activator tBHQ (Desagher et al., 1996; Lee et al., 2003a; Li et al., 2002; Lucius and Sievers, 1996;

Murphy et al., 2001).

Overexpression of Nrf2 and tBHQ-mediated and sulforaphane-mediated activation of the

Nrf2-ARE pathway in astrocytes increased resistance of neurons to non-excitotoxic glutamate toxicity (Kraft et al., 2004). In most cases, the presence of induced Nrf2 was neuroprotective, whereas Nrf2 deficiency conversely increased sensitivity to neurotoxins that cause inhibition of mitochondrial complex II (malonate or 3-NP), kainic acid, MPTP and 6-hydroxydopamine (6-

OHDA), as well as cerebral ischemia (Burton et al., 2006; Calkins et al., 2005; Jakel et al.,

2005; Jakel et al., 2007; Kraft et al., 2006; Zhao et al., 2007a; Zhao et al., 2007b). Protection or enhanced susceptibility was assessed in multiple brain regions such as the striatum, hippocampus, cortex and substantia nigra, while degenerative endpoints including decreased nerve terminal markers, neurotransmitters, caspase activation (apoptosis), neurotoxic glial 101

activation, oxidative macromolecular damage and even functional effects on motor coordination and learning and memory deficits (Burton et al., 2006; Calkins et al., 2005; Dash et al., 2009;

Jakel et al., 2005; Jakel et al., 2007; Kraft et al., 2006; Zhao et al., 2007a; Zhao et al., 2007b).

Further evidence of Nrf2-mediated protection comes from studies which show that treatment with chemical activators of the Nrf2–ARE pathway prior to the toxic insults can reduce cellular damage in Nrf2 wild-type mice but not in Nrf2 KO mice (Burton et al., 2006; Shih et al., 2005a;

Shih et al., 2005b; Zhao et al., 2007a; Zhao et al., 2007b).

It has been suggested that METH-initiated neurotoxicity does not involve Nrf2, as dopamine depletion in the striatum at two weeks post-treatment was not different between Nrf2

KO mice and their wild-type controls (Pacchioni et al., 2007). However, studies in rat striatum indicate that METH can activate Nrf2 and regulate HO-1 expression (Jayanthi et al., 2009).

Accordingly, it is uncertain whether ROS-mediated Nrf2 activation in specific brain regions caused by METH can modulate its neurotoxicity. Of interest is that some of studies show regional differences in Nrf2 activation and the type of protective genes induced as well as differences in the temporal pattern of induction, which highlights the potential for brain regional susceptibilities to neurotoxicity. For example, in a study of the involvement of the Keap1-Nrf2 system in ischemic brain injury, there was a difference in temporal changes of Nrf2-regulated

TRX and HO-1 between the peri-infarct regions and regions destined to infarct (Tanaka et al.,

2010). In rats, administration of the Nrf2 inducer sulforaphane following cortical impact injury improved behavioral performance when the treatment was initiated 1 hour, but not 6 hours, post- injury (Dash et al., 2009). Chemical inducers of Nrf2 like tBHQ have been shown in rats to increase the levels of glutathione, but not GST or NQO1, in the cortex and striatum (Shih et al.,

2005a). However, under the same conditions, all of these antioxidant systems were induced in 102

the liver, indicating a more robust Nrf2 effect in this organ versus the brain, possibly due to the contribution of hepatic metabolism in xenobiotic detoxification (Shih et al., 2005a).

Some neurotoxins themselves have been shown to activate Nrf2, for example, 6-OHDA was found to activate the Nrf2–ARE system and protect against ROS-mediated damage (Jakel et al., 2005; Jakel et al., 2007). The loss of Nrf2 increased vulnerability to 6-OHDA both in vitro and in vivo, whereas upregulation of Nrf2 using tBHQ protected against 6-OHDA-induced cell death in vitro (Jakel et al., 2005). Also selegiline (deprenyl), a drug used in the treatment of PD, stimulated Nrf2 activity as part of its cytoprotective mechanism of action (Nakaso et al., 2006).

Currently, clinical trials with DMF are ongoing (Kappos et al., 2008). DMF is able to activate

Nrf2 and hence antioxidant enzyme production (Murphy et al., 2001), and oral administration of

DMF significantly reduced the formation of new MS lesions in humans (Kappos et al., 2008).

1.4.9 NRF KNOCKOUT MOUSE MODELS

Many types of Nrf knockout mice have been generated, including double knockouts of

Nrf1 and Nrf2, which have revealed that although all Nrf isoforms can regulate the ARE, they are not functionally redundant (Chan et al., 1996; Derjuga et al., 2004; Itoh et al., 1997; Leung et al., 2003; Chan et al., 1998). Nrf3 knockout mice are viable and have no obvious phenotypes

(Derjuga et al., 2004). Nrf1 has been shown to be important for development as Nrf1 -/- mice die in utero due to anemia as a result of abnormal fetal liver erythropoiesis (Chan et al., 1998).

The viability of mice lacking Nrf2 has been taken to indicate that Nrf2 plays no role in normal mouse development (Chan et al., 1996; Itoh et al., 1997). However, in the context of oxidative stress, Nrf2-deficient fetuses may be more susceptible to developmental toxicities. Nrf2 mRNA 103

and protein are expressed during the period of organogenesis (Chan et al., 1996); hence, Nrf2 may become activated and modulate toxicity during in utero exposure to oxidative stress.

Selective upregulation of Nrf2 during fetal development in instances of increased oxidative stress may be protective, as shown in our study herein, and in studies from another laboratory where activation of Nrf2 in embryos protected against ethanol-induced embryotoxicity (Dong et al., 2008). However, constitutive activation of Nrf2 in Keap1 -/- mice is also detrimental by causing hyperkeratotic lesions that block the esophagus and increase postnatal death, possibly due to starvation (Wakabayashi et al., 2003).

Mice lacking Nrf2 are also highly sensitive to toxic stressors, some of which are neurotoxic, as discussed in the previous section. Primary astrocytes and neurons derived from

Nrf2 KO mice are more sensitive than wild-type controls to oxidative damage, calcium disturbance and mitochondrial toxins (Lee et al., 2003b), and Nrf2 -/- cells stable cells are more sensitive to NO•-induced apoptosis (Dhakshinamoorthy and Porter, 2004). Recent data suggest that Nrf2 plays an essential role in oligodendrocyte function and myelination, as Nrf2 KO mice exhibit severe myelin degeneration and oxidative damage to the myelin sheath, and aged Nrf2

KO animals develop vacuolar leukoencephalopathy and astrogliosis (Hubbs et al., 2007).

Furthermore, in aged untreated animals, myelin degeneration in Nrf2-null mice was seen in the striatum, hippocampus, cerebellum and corpus callosum, but importantly the cortex and brainstem were spared, suggesting a regionally specific impact of chronic Nrf2 loss (Hubbs et al., 2007). In addition, aged female Nrf2 –/– mice develop a multiorgan autoimmune disease similar to human systemic lupus erythematosus (Li et al., 2004; Ma et al., 2006; Yoh et al.,

2001). The reason for possible gender specificity remains unclear and may depend on Nrf2- regulated genes and their interactions with female-specific genes. 104

On the basis of experiments with Nrf2 KO mice and cells derived from them, there is interest in the Nrf2 system not only neurotoxicity, but also in cancer chemoprevention. Nrf2 has been shown to be protective against benzo[a]pyrene carcinogenicity (Ramos-Gomez et al.,

2001). Nrf2 is constitutively activated in diverse human cancers, suggesting that Nrf2-mediated regulation of the antioxidant response pathway may play a role in cancer susceptibility (Hayes and McMahon, 2009). Cancer cells take control of the Nrf2 pathway, apparently to protect themselves from the cellular stresses, which may make chemotherapies ineffective. Nrf2 may also play a role in experimental models of pulmonary diseases such as asthma (Rangasamy et al., 2005); inflammatory disorders such as inflammatory bowel disease (Khor et al., 2006); acetaminophen-induced hepatotoxicity (Chan et al., 2001; Enomoto et al., 2001) and atherosclerosis (Collins et al., 2009).

1.4.10 EVIDENCE FOR POLYMORPHISMS IN THE KEAP1-NRF2-ARE PATHWAY

Single nucleotide polymorphisms have been identified within the promoter region of the human Nrf2 gene (Yamamoto et al., 2004), and genetic variation here along with polymorphisms in the ARE region of Nrf2 target genes (Wang et al., 2007a) may cause variability in Nrf2 activity, and hence the response to oxidative stress and neurotoxicity. Three single nucleotide polymorphisms (–650C/A, –684G/A, –686A/G) and one triplet repeat polymorphism (CCG) have been identified in the promoter region of the Nrf2 gene (Yamamoto et al., 2004). As discussed above, cancer cells with mutations to increase Nrf2 activity may promote their survival (Hayes and McMahon, 2009). Conversely, inherited DNA polymorphisms can reduce the abundance of Nrf2. Promoter polymorphisms at positions –617 105

C/A or –651 G/A reduce the basal expression level of Nrf2, thereby resulting in attenuation of

ARE-mediated gene transcription. The allelic frequencies of these inhibitory polymorphisms are as high as 20% in Europeans, 40% in Asians and 55% in Native Americans (Marzec et al.,

2007). Moreover, Marzec et al. reported that the –617 A single nucleotide polymorphism had a significantly higher risk for developing acute lung injury after major trauma (Marzec et al.,

2007).

Vitiligo is an acquired pigmentary disorder and Nrf2 gene polymorphisms with the –650

C/A allele may be a risk factor associated with the development of vitiligo (Guan et al., 2008).

These polymorphisms are also significantly associated with the development of gastric mucosal inflammation and peptic ulcer diseases (Arisawa et al., 2007; Arisawa et al., 2008a; Arisawa et al., 2008b). The recent analysis of Keap1 (INrf2) in human lung cancer patients and cell lines showed that deletion, insertion and missense mutations in functionally important domains of

Keap1 results in reduction of Keap1 affinity for Nrf2 and elevated expression of cytoprotective genes giving a selective advantage to cancerous cells (Padmanabhan et al., 2006; Singh et al.,

2006). The G430C substitution is a somatic mutation in Keap1 discovered in a human lung cancer patient (Padmanabhan et al., 2006).

A number of somatic genetic events are important in cancerous cells, including Keap1 or

Nrf2 mutations, and have been shown to provide advantages for lung and breast cancer growth as well as chemoresistance (Nioi and Nguyen, 2007; Ohta et al., 2008; Padmanabhan et al.,

2006; Shibata et al., 2008; Singh et al., 2006). Studies on the epigenetic hypermethylation of the Keap1 promoter evaluated cytosine methylation in the Keap1 promoter and demonstrated that the P1 region, including 12 CpG sites, was highly methylated in lung cancer cells and tissues, but not in normal cells (Wang et al., 2008a; Wang et al., 2008b). This would reduce 106

levels of Keap1 and hence activate Nrf2 in cancerous cells (Wang et al., 2008a; Wang et al.,

2008b). Aside from the detrimental consequences of Nrf2/Keap1/ARE mutations arising in cancer cells, a deficient response in this pathway can enhance the risk of diseases and xenobiotic toxicities. 107

1.5 METHAMPHETAMINE 1.5.1 HISTORY AND USES The amphetamine analogs 3,4-methylenedioxymethamphetamine (MDMA, Ecstasy), methylenedioxyamphetamine (MDA) (active major metabolite of MDMA) and methamphetamine (METH, Speed) are synthetic substances that are common drugs of abuse, as they promote the release of neurotransmitters and induce euphoria and hallucinations. They are illicit drugs classified under Controlled Drug Acts in both the United States and Canada.

Amphetamine derivatives intended for recreational use have been referred to as 'Designer drugs' because they are designed to circumvent existing legal restrictions (Anglin et al., 2000).

These amphetamine analogs have a chiral center at the alpha carbon, and thus exist as a pair of optical isomers; for example, d-METH (corresponding to the configuration of S-(+)) and l-METH (corresponding to the configuration of R-(-)). Their pharmacological profiles are stereoselectively distinct. The d-enantiomer is the dominant CNS stimulant and is five times more biologically active than the l-enantiomer, which has greater sympathomimetic activity.

(Kuczenski et al., 1995). The l-enantiomer is also formed as a metabolite of selegiline, an anti-

Parkinsonian drug (Cho, 1990). These drugs belong to a class of sympathomimetic drugs called phenylethylamines and are structurally similar to many endogenous neurotransmitters such as dopamine, SE and epinephrine as well as their metabolites (Figure 12).

Amphetamines were drugs originally developed as synthetics used as substitutes for ephedrine. Japanese scientists synthesized methamphetamine in 1919. In 1932, the Smith, Kline and French pharmaceutical company introduced these drugs in over-the-counter inhalers for asthma and congestion. In the 1930s, the American Medical Association approved the use of amphetamines under names like benzedrine (d/l-amphetamine) for treatment of a 108

Figure 12. Amphetamine, its analogs and neurotransmitters, their precursors and metabolites. Abbreviations: ALDH, aldehyde dehydrogenase; COMT, catechol-o-methyltransferase; MAO, monoamine oxidase; PNMT, Phenylethanolamine-N-methyltransferase. 109

range of disorders such as narcolepsy, depression, PD, attention deficit disorder and even as an appetite suppressant. The therapeutic dose for these drugs in tablet form is typically 5–10 milligrams (Anglin et al., 2000; Cho et al., 2001; Kalant, 2001). The first reported misuse of amphetamine was in 1937 when it was used by students in Minnesota to avoid sleep during examination periods. Thereafter both amphetamine and METH were widely used both clinically and illicitly during the Second World War by the Americans, Germans and Japanese, and became a serious problem in post-war Japan. Increasing popularity of METH as a drug of abuse within the United States led to its illicit production in the 1960s and by the 1970s laws were passed to make methamphetamine illegal to possess without a prescription (Anglin et al., 2000)

First synthesized by Merck in 1912, MDMA was patented in 1914 but never marketed

(Gibb et al., 1990). While possibly one of the drugs used as a stimulant in during the World

Wars, it gained recognition in 1965 when Alexander Shulgin manufactured MDMA in his laboratory, but it was not until the 1970s-80s that MDMA first used recreationally and in psychotherapy and was said to increase patient self-esteem and facilitate therapeutic communication (Gibb et al., 1990; Grinspoon and Bakalar, 1986). In 1985, the U.S. Drug

Enforcement Administration classified MDMA as a Schedule 1 drug due to its high abuse potential and evidence that MDA, a related compound and major MDMA metabolite, induced serotonergic nerve terminal degeneration in rat brain (Ricaurte et al., 1985). However, since the mid 1980s MDMA has been a popular recreational drug at “raves”, causing a state of euphoria allowing the user to socialize and dance all night (Green et al., 2003). The absence of commercially produced METH and MDMA/MDA led to the clandestine production of these drugs. Furthermore, the ease of obtaining precursors for METH synthesis, such as ephedrine and pseudoephedrine found in cough medication, resulted in production of the higher quality d- 110

METH and by the early 1980s, METH became more easily synthesized and readily available for abuse. Despite efforts to limit production of the precursors, METH and MDMA use and the associated behavioural problems and addiction have always been and remain a concern, especially among young people (Anglin et al., 2000; Gibb et al., 1990; Green et al., 2003;

Kalant, 2001).

Amphetamine marketed as Adderall and Dexedrine is prescribed for the treatment of narcolepsy and attention deficit-hyperactivity disorder (ADHD), while METH is indicated for the treatment of ADHD and the short-term treatment of obesity. METH prescriptions are very rare (Golub et al., 2005). The recommended starting dose for METH treatment of ADHD is

5 mg/day in individuals who are at least 6 years old. The maximum recommended dose for

ADHD is 25 mg/day. For obesity, the recommended METH dose is 5 mg before a meal (Golub et al., 2005). During illicit use, METH can be taken through a variety of routes, including ingestion, injection, nasal insufflation and inhalation (smoking) (Cook et al., 1993; Cook et al.,

1992; Huestis and Cone, 2007).

1.5.2 PHARMACOKINETICS

Following ingestion, METH is absorbed across the gastrointestinal tract. Controlled studies with therapeutic formulations (5-10 mg) have indicated tmax values ranging from 3 to 6 hours post-ingestion (Cook et al., 1993; Cook et al., 1992; Huestis and Cone, 2007; Schepers et al., 2003). In chronic METH abusers, the average plasma concentration ranges from 150 ug/L

(ng/mL) to 1700 ug/L (ng/mL) (1 uM-10 uM), while the concentration of its amphetamine metabolite ranges from 30 ug/L to 300 ug/L (Wilson et al., 1996). Following intranasal administration of METH, peak plasma concentrations do not occur until approximately 3–4 111

hours post-exposure (Harris et al., 2003).

Inhalation of METH via smoking provides a bioavailability that ranges from 67% to

90% with the differences depending on smoking technique and the temperature of the flame

(Harris et al., 2003). Following oral ingestion, 67% may be absorbed (Harris et al., 2003).

METH is very lipophilic and distributes extensively across the blood–brain barrier (Riviere et al., 2000). METH is also distributed into breast milk, appearing in the milk within minutes of intravenous use (Bartu et al., 2009). Because of its low molecular weight and high lipid solubility, there is also significant transfer of METH from maternal to fetal blood (Stewart and

Meeker, 1997). METH and amphetamine show no differences in their effect on dopamine release in the striatum, elimination rates or other pharmacokinetic properties (Melega et al.,

1995).

The related amphetamine analogs MDMA and MDA are readily absorbed from the intestinal tract and reach their peak concentration in the plasma about 2 hours after oral administration (de la Torre et al., 2000). Doses of 50 mg, 75 mg and 125 mg to healthy human volunteers produced peak blood concentrations of 106 ng/mL, 131 ng/mL and 236 ng/mL of

MDMA respectively. These concentrations are relatively low, in part because the drug passes readily into the tissues, and much of it is bound to tissue constituents (de la Torre et al., 2000).

Most of the cases of serious toxicity or fatality with MDMA have involved blood levels ranging from 0.5 mg/L to 10 mg/L, (500 - 10,000 ng/mL); that is, up to 40 times higher than the usual recreational range (Kalant, 2001).

The mean value for the half-life of MDMA in humans ranges between 9 and 12 hours

(Cook et al., 1993; Cook et al., 1992; Kalant, 2001; Schepers et al., 2003; Shappell et al., 1996), and is not measurably altered by the route of drug administration (Cook et al., 1993; Cook et al., 112

1992; Harris et al., 2003). The half-life in rodents is only 70 minutes to 3 hours, which is significantly lower than that in humans (Cho et al., 2001; de la Torre and Farre, 2004). To approximate human plasma concentrations, mice typically are administered 4 doses of METH

(5- 20 mg/kg i.p.), with a 2-6-hour interval between each dose (Cho et al., 2001; de la Torre and

Farre, 2004). This dosing regimen causes significant neurotoxicity in mice, and achieves a plasma concentration in mice similar to that resulting from a human METH binge pattern of self-administration (Cho et al., 2001; de la Torre and Farre, 2004; Jayanthi et al., 1998; Kita et al., 2000; O'Callaghan and Miller, 1994; Jeng et al., 2006). These concentrations are similar to plasma concentrations of METH in humans after chronic use in the range 0.176-1.743 mg/L

(176-1743 ng/mL) (Melega et al., 2007; Wilson et al., 1996); however, concentrations in rats average 10X higher in the brain versus plasma (Melega et al., 2007).

Human abuse patterns vary from single day usage to regular users where Cho et al.

(2001) reported a dose range of 20–250 mg or more per “hit” in METH abusers with total daily doses of up to several grams, which is substantially greater than the doses used normally in controlled clinical experiments (Cho et al., 2001; McKetin et al., 2006). Further estimates of plasma concentrations come from impaired drivers testing positive for METH, in whom plasma

METH concentrations were typically 300–550 ng/ml, with plasma concentrations up to 1665 ng/ml in non-fatal cases (Logan, 1996; Melega et al., 2007; Melega et al., 1995).

1.5.3 DISTRIBUTION

Postmortem analysis in humans has shown METH distributes to many different tissues including brain, liver and kidney (Kojima et al., 1984). METH is homogeneously distributed 113

within the brain of chronic human users in the globus pallidus, caudate, hippocampus and temporal cortex (Kalasinsky et al., 2001). Human brain levels of METH range from 44-100 nmol/g brain tissue, but can be as high as 200 nmol/g of METH (Kalasinsky et al., 2001; Wilson et al., 1996). Studies with [11C] d-METH tracing in human brain have shown a relatively rapid distribution across brain regions with high and persistent uptake in both subcortical and cortical areas, slow clearance from gray matter and no observable clearance from white matter regions.

METH uptake correlated with DAT availability in the striatum but not the cerebellum over the time course of the study (Fowler et al., 2008). The highest peak uptake of METH was in the putamen and after i.v. administration, with 7–8% of the injected dose accumulating in the brain within 10 min. Estimates from this study suggest that a typical human dose of 30 mg in a

METH user would result in a brain accumulation of about 2.5 mg of METH (14 μM) (Fowler et al., 2008). This is similar to findings in the non-human primate brain (Fowler et al., 2007), and to studies in rat brain after i.v. METH administration (Fowler et al., 2007; O'Neil et al., 2006;

Riviere et al., 2000). Levels of METH are relatively uniformly distributed throughout the brain of experimental animals administered a single dose of the drug (Fowler et al., 2007; Jonsson and

Nwanze, 1982; O'Neil et al., 2006; Shiue et al., 1995) with concentrations in the frontal cortex, striatum and cerebellum of 65±3, 55±5 and 46±2 nmol/g, respectively (O'Neil et al., 2006).

Male Sprague-Dawley rats that received a pharmacologically active METH i.v. bolus dose (1.0 mg/kg) showed distribution of METH into brain and other tissues. This study also revealed that the highest concentrations were observed in the kidney, liver, brain and heart with a delayed peak concentration in the spleen. The METH metabolite amphetamine also distributes extensively into these tissues, and could significantly contribute to the pharmacological effects after administration of METH (Riviere et al., 2000). 114

METH uptake has also been reported to be higher in the striatum versus other brain areas of chronically treated animals (Miyazaki et al., 2006). Theoretically, the extracellular level of drug in the brain is important for determining its psychopharmacological action via receptor/transporters located on neuronal cells. Plasma concentrations of MDMA and METH after recreational doses are usually in the range of 1-10 μM in humans (de la Torre et al., 2000;

Wilson et al., 1996); however, concentrations in brain may be substantially higher. Studies in rats with a dose of 5-10 mg/kg of MDMA or METH achieve extracellular striatal concentrations of 10-100 uM (Esteban et al., 2001; Melega et al., 1995), which are usually on average 10X higher than plasma concentrations. These intracellular concentrations may also be elevated as active transporters may concentrate these drugs inside the neuronal terminal and in the brains of tolerant abusers during high-dose binges. Other studies have shown that the dopamine transporter (DAT) plays an important role in METH neurotoxicity, as DAT knockout mice are less susceptible than wild-type controls to METH-initiated neurotoxicity (Fumagalli et al.,

1998). Postmortem tissue is devoid of these active transport systems and hence METH levels determined in these samples may be significantly different from those in living subjects.

As mentioned previously, METH is also distributed into breast milk (Bartu et al., 2009), and there is significant transfer from maternal to fetal blood (Stewart and Meeker, 1997). The

40 mg/kg doses used in our studies gives a concentration in fetal brain similar to that in METH- exposed infants (Bost et al., 1989; Stewart and Meeker, 1997; Won et al., 2001). Won et al.

(2001) measured maternal and fetal brain levels of METH and amphetamine after s.c. injection of mouse dams with 40 mg/kg d-METH hydrochloride on gestational day (GD) 14. In maternal striatum, METH levels peaked at approximately 510 ng/mg protein 1 hour after injection, while fetal striatum had 99 ng/mg protein in the striatum, approximately 102 ng/mg protein in the 115

brainstem and 57 ng/mg protein in the cortex which peaked at 1 hour after injection, indicating that the drug can accumulate in fetal brain (Won et al., 2001).

1.5.4 METABOLISM BY CYTOCHROMES P450 (CYPs) and ELIMINATION

METH and amphetamine can be excreted unchanged, but the amount and disposition of the metabolites are influenced by urinary pH (Beckett and Rowland, 1965; Caldwell et al.,

1972). With pKa values of approximately 9.9 for the parent drugs, at normal physiological pH, these compounds are primarily in their ionized form. In the urine under acidic conditions, the drugs are ionized and primarily secreted unchanged, with insignificant reabsorption by the kidneys. Conversely, alkaline urine converts more of these drugs to their neutral form, which are readily reabsorbed by the kidneys, thereby increasing the half-lives of the drugs (Beckett and

Rowland, 1965; Caldwell et al., 1972).

About one-half of METH in humans is excreted unchanged, and the remainder undergoes CYP2D6-catalyzed N-demethylation to amphetamine, which can then be hydroxylated (Caldwell et al., 1972; Cook et al., 1993) (Figure 13). METH also is oxidized by

CYP2D6 to 4-hydroxy-METH, which is the predominant metabolite constituting almost 50% of all metabolites excreted in the urine (Caldwell et al., 1972; Cook et al., 1993). These metabolites also accumulate in the striatum after administration of the parent drug (Melega et al., 1995). Other minor metabolites include norephedrine and 4‐hydroxynorephedrine.

MDMA is N-demethylated to MDA by CYP1A2. MDA can be further metabolized by

CYP2D6/CYP3A4 to the catechol intermediate HHA (3,4-dihydroxyamphetamine), and finally

O-methylated by COMT to 4-hydroxy-3-methoxyamphetamine (HMA) (Figure 14) (Kreth et al., 2000; Meyer et al., 2008). MDMA can also be O-demethylenated by CYP2D6 to 116

Figure 13. Metabolism of methamphetamine by cytochromes P450. Methamphetamine (METH) can be demethylated to amphetamine by CYP2D6. 4-Hydroxylation of METH is catalyzed by CYP2D6 to form 4-hydroxy-derivatives while β-hydroxylase generates norephedrine-derivatives. 117

3,4-dihydroxymethamphetamine (HHMA), followed by O-methylation to 4-hydroxy-3- methoxymethamphetamine (HMMA) (a reaction regulated by COMT). These metabolites are believed to be downstream products formed after the opening of the methylendioxyphenyl ring, a process that is mainly catalyzed by CYP2D6, with low-affinity contributions from CYP1A2,

CYP2B6, and CYP3A4 (Kreth et al., 2000; Meyer et al., 2008).

Systemic metabolism of MDMA may play a role in its neurotoxicity. This was concluded from the observation that direct injection of ecstasy into the brain fails to reproduce the neurotoxic effects seen after systemic administration (Esteban et al., 2001). Metabolites such as HHMA and HHA are easily oxidized to their corresponding quinones by CYPs, hydroxyl radicals and possibly PHS; these quinones can form adducts with GSH or conjugation with sulfate or glucuronide and other thiol-containing compounds (Hiramatsu et al., 1990;

Segura et al., 2001), including proteins critical for neural function and survival. The GSH- derived conjugated metabolites are low in urine (Musshoff, 2000).

Catechol metabolites of METH, MDMA and MDA can be oxidized by CYPs to reactive quinones that redox cycle to semiquinone radicals that generate ROS (Hiramatsu et al., 1990).

While it is uncertain whether CYPs contribute to ROS generation, the expression of CYPs in the brain is around 1-2% of that in the liver (Warner et al., 1997). It is also reported that MDMA is a potent competitive inhibitor of CYP2D6 in human liver microsomes (Heydari et al., 2004; Wu et al., 1997). Furthermore, CYP2D6 mRNA transcripts are widely expressed in different brain regions, but constitute only 3% of expression in the liver. At the protein level, CYP2D6 is present in the human frontal lobe, hippocampus and cerebellum (Dutheil et al., 2009). This enzyme is also coded by a polymorphic gene with some subjects expressing no activity (Sachse et al., 1997); however, it is not known whether CYP2D6 in human brain can bioactivate 118

Figure 14. MDMA metabolism by CYPs and P450 reductase. Abbreviations: HHA, 3,4- dihydroxyamphetamine; HHMA, 3,4-dihydroxymethamphetamine; HMA, 4-hydroxy-3- methoxyamphetamine; HMMA, 4-hydroxy-3-methoxymethamphetamine; MDA, 3,4- methylenedioxyamphetamine; MDMA, 3,4-methylenedioxymethamphetamine. 119

amphetamines to neurotoxic intermediates . Evidence from rats suggest that this may not be the case, since rats deficient in CYP2D1, the rat homolog of CYP2D6 in human, remain susceptible to MDMA neurotoxicity (Colado et al., 2001). The activity of NADPH P450 reductase, which converts quinones to semiquinone radicals, has high levels in the putamen-pallidum region where amphetamine analogs and their metabolites accumulate. However, this activity was low in human brain microsomes, representing only 8% of the value reported in rat brain microsomes

(Ghersi-Egea et al., 1994). Therefore, other mechanisms of ROS generation may be present in the brain that can contribute to the metabolism and neurotoxicity of amphetamine analogs.

1.5.5 RECEPTOR-MEDIATED PHARMACOLOGICAL ACTIONS OF METH

METH can cause reversible, receptor-mediated effects in both the peripheral and central nervous systems. In the periphery, clinical manifestations are usually through the alpha- and beta-adrenergic receptor-mediated sympathomimetic effects (Sulzer et al., 2005; Cruickshank and Dyer, 2009). METH leads to the increased release of the key neurotransmitters by several different processes in the brain. METH enters the presynaptic terminals by both passive diffusion across the lipid membrane and through the plasma membrane catecholamine-uptake transporters such as DA, norepinephrine (NE), and SE transporters (Figure 15) (Cruickshank and Dyer, 2009; Sulzer et al., 2005). Within the cytosol, METH enters the presynaptic vesicle via the membrane-bound vesicular monoamine transporter-2 (VMAT-2) and facilitates the redistribution of the monoamines into the cytosol by disrupting the pH gradient that drives the accumulation of the monoamines within the vesicles (Cruickshank and Dyer, 2009; Sulzer et al.,

2005). This contributes to elevated neurotransmitter concentrations within the cytosol, leading to increased movement into the synapse via the plasma membrane transporters which change 120

Figure 15. METH actions at the dopaminergic nerve terminal. METH are substrates of DAT transporters and are taken up into the cell. Once in the cell, METH interferes with the vesicular monoamine transporter (VMAT), depleting synaptic vesicles of their neurotransmitter content. As a consequence, levels of dopamine (or other transmitter amines) in the cytoplasm increase and quickly become sufficient to cause release into the synapse by reversal of the plasma membrane DAT. 121

from an influx to efflux state; for example, the DAT reverses the direction of DA transport causing the transporter to move dopamine from the cytoplasm into the synapse (Fumagalli et al.,

1998; Khoshbouei et al., 2003; Rothman and Baumann, 2003). The detailed mechanisms by which this occurs are unclear. The net effect of these mechanisms is to acutely increase the levels of neurotransmitters in the synaptic cleft, thereby increasing their potential for receptor binding. Hence, METH acts as an indirect agonist at DA, NE and SE receptors.

KI values of METH, amphetamine and MDMA for the DAT, NE transporter (NET) and

SE transporter (SERT) are presented for human, mouse and rat in Table 13. KI values of uptake inhibition reflect the apparent affinities of the drugs to each transporter. Low KI values for amphetamine and METH suggest decreased neurotransmitter exchange at its transporter.

The human and mouse transporters are similar in their efficacies for each of the tested drugs (KI values within a 4-fold range), whereas the chemical modification substantially increases the potency of MDMA for inhibiting SERT while reducing its potency for inhibiting DAT and NET compared to METH (Fleckenstein et al., 1999; Han and Gu, 2006; Rothman and Baumann,

2003). In vitro studies have shown that the amphetamines are better than SE at releasing DA and NE (Fleckenstein et al., 1999; Han and Gu, 2006; Rothman and Baumann, 2003).

These neurotransmitters then bind to a number of receptors that are activated to mediate complex physiological acute, chronic and adverse responses to METH (Table 14). For example, hyperthermia involves the alpha1 adrenoreceptor (Sprague et al., 2004), DA-1

(Mechan et al., 2002) DA-2 (Bowyer et al., 1994) and SE-2A receptors (Herin et al., 2005). In

ADHD, the mechanism of the therapeutic effects of amphetamine and methamphetamine is not known. It is believed that amphetamines increase levels of catecholamine in the synaptic space by blocking reuptake of NE and DA, hence increasing the effect of these neurotransmitters (Kita 122

et al., 2003). Interestingly, DA receptors and adrenoceptors reportedly have no affinity for amphetamines themselves (Kraemer and Maurer, 2002). 123

Table 13. Amphetamine analog binding affinities to uptake transporters Inhibition Drug Transporter constant (KI) Notes (uM) Amphetamine rDAT 0.034 rDAT in rat synaptosomes (Rothman et al., 2001) 2.3 rDAT cultured cells expressing (Giros and Caron, 1993) 0.094 rDAT rat striatal synaptosomes (Fleckenstein et al., 1999) mDAT 0.56 mDAT expressed in intestinal cells (Han and Gu, 2006) hDAT 0.64 hDAT expressed in intestinal cells (Han and Gu, 2006) rNET 0.039 rNET in rat synaptosomes (Rothman et al., 2001) mNET 0.12 mNET expressed in intestinal cells (Han and Gu, 2006) hNET 0.07 hNET expressed in intestinal cells (Han and Gu, 2006) rSERT 3.8 rSERT in rat synaptosomes (Rothman et al., 2001) 8 rSERT rat striatal synaptosomes (Fleckenstein et al., 1999) mSERT 23 mSERT expressed in intestinal cells (Han and Gu, 2006) hSERT 38 hSERT expressed in intestinal cells (Han and Gu, 2006) METH rDAT 0.291 rDAT rat striatal synaptosomes (Fleckenstein et al., 1999) mDAT 0.47 mDAT expressed in intestinal cells (Han and Gu, 2006) hDAT 0.46 hDAT expressed in intestinal cells (Han and Gu, 2006) 0.082 hDAT expressed in human embryonic kidney cells (Eshleman et al., 1999) mNET 0.19 mNET expressed in intestinal cells (Han and Gu, 2006) hNET 0.11 hNET expressed in intestinal cells (Han and Gu, 2006) 0.0013 hNET expressed in human embryonic kidney cells (Eshleman et al., 1999) rSERT 9 uM rSERT rat striatal synaptosomes (Fleckenstein et al., 1999) 124

Table 13. Amphetamine analog binding affinities to uptake transporters continuted Inhibition Drug Transporter constant (KI) Notes (uM) METH mSERT 9.28 mSERT expressed in intestinal cells (Han and Gu, 2006) hSERT 31 hSERT expressed in intestinal cells (Han and Gu, 2006) 20.7 hSERT expressed in human embryonic kidney cells (Eshleman et al., 1999) MDMA rDAT 1.572 rDAT (Rothman and Baumann, 2003) 1.53 rDAT rat striatal synaptosomes (Fleckenstein et al., 1999) mDAT 4.87 mDAT expressed in intestinal cells (Han and Gu, 2006) hDAT 8.29 hDAT expressed in intestinal cells (Han and Gu, 2006) rNET 0.462 rNET (Rothman and Baumann, 2003) mNET 1.75 mNET expressed in intestinal cells (Han and Gu, 2006) hNET 1.19 hNET expressed in intestinal cells (Han and Gu, 2006) rSERT 0.24 rSERT (Rothman and Baumann, 2003) 2.6 rSERT rat striatal synaptosomes (Fleckenstein et al., 1999) mSERT 0.64 mSERT expressed in intestinal cells (Han and Gu, 2006) hSERT 2.41 hSERT expressed in intestinal cells (Han and Gu, 2006) Abbreviations. h, human; m, mouse; r, rat; DAT, dopamine transporter; NET, norepinephrine transporter; SERT, serotonin transporter 125

Table 14. METH receptor-mediated effects (Berridge and Waterhouse, 2003; Girault and Greengard, 2004; Guan et al., 2008; Hornung, 2003; Nichols and Nichols, 2008; Sofuoglu and Sewell, 2009; Wise, 2004; Xu et al., 2005) Dopamine Serotonin Norepinephrine

D1: ↑cAMP→↑PKA 5HTR1,3-7: α1: ↑Ip3, DAG→ ↑ D2: ↓cAMP Modulates cAMP Ca2+ Receptor subtypes: 5HT2: ↑Ip3, DAG→ α2: modulates cAMP 2nd messenger ↑ Ca2+ β1-3: modulates cAMP

Mesolimbic, Raphe nuclei of Locus Ceruleus nuclei Mesocortical circuit brainstem extends to brainstem extends to Brain regional Nigrostriatal pathway spinal cord and medial basal location Nucleus accumbens cortical regions forebrain, hippocampus, cortex Mood Hyperthermia Increases blood Appetite Anxiety pressure, Effects Motor coordination Cognition motor control, Drug dependence Psychotic behaviour learning, memory and Appetite fear Abbreviations. 5HTR, serotonin receptor; α, alpha adrenergic receptor; β, beta adrenergic receptor; Ca2+, calcium; cAMP, cyclic adenosine monophosphate; D, dopamine receptor; DAG, diacylglycerol; Ip3, inositol triphosphate; PKA, protein kinase A. 126

1.5.6 EFFECTS OF METH ABUSE

Effects following METH exposure can vary widely as the release of neurotransmitters like DA, NE and SE can act to control the messaging systems of the brain for reward and pleasure, sleep, appetite and mood (Cruickshank and Dyer, 2009; Sulzer et al., 2005). At doses of between 5 and 50 mg, the first effect is usually a characteristic “rush,” which is believed to be the result of an initial release of high concentrations of dopamine in the central nervous system

(Cruickshank and Dyer, 2009). The effects are rapid when the drug is smoked or injected, appearing within 2-3 minutes after inhalation or 20 minutes after oral ingestion (Golub et al.,

2005). The immediate effects of METH are similar to the fight-or-flight response and involve increased heart and respiratory rates, blood pressure and body temperature due to peripheral effects of adrenergic stimulation (Cruickshank and Dyer, 2009; Golub et al., 2005; Sulzer et al.,

2005) (Table 15). The rush is often associated with euphoria and increased energy, which can last for several hours given the long half-life of METH. Low doses tend to produce a sense of heightened alertness, attentiveness and energy, whereas high doses produce a sense of well- being, euphoria and enhanced self-esteem (Cruickshank and Dyer, 2009; Golub et al., 2005;

Sulzer et al., 2005).

The adverse effects are both short-term (cardiac problems, hyperthermia, depression, confusion) and chronic (Table 15). With METH use, the effect is sustained for hours, placing an extended burden on the nervous, circulatory and respiratory systems. METH causes several adverse cardiac effects (Wijetunga et al., 2003). In a case control study of users, only 64% of

METH users showed normal heart function compared to 88% of age-matched controls. In addition, 28% of METH users showed severe cardiac dysfunction compared to 7% of age- matched controls. When used chronically, METH may cause long-term CNS consequences that 127

result in impaired memory, impaired motor coordination and psychiatric problems long after termination of use (Scott et al., 2007)(Table 15). This is likely due to the acute reversible effects being receptor-mediated and dependent upon the plasma concentration of the amphetamine analog at that time, whereas effects persisting long after the drug is gone are likely due to macromolecular damage. A study of over 1,000 METH users in treatment found a high incidence of psychiatric problems, such as depression, anxiety, suicide and violent or assaultive behaviors. Residual psychiatric symptoms included a prolonged inability to experience pleasure, as well as increased anxiety and psychotic episodes (Cohen et al., 2003; Zweben et al., 2004).

Gross motor abnormalities in METH abusers, including parkinsonism, however, are not prevalent given the vulnerability of the striatum to METH-associated neurotoxicity (Caligiuri and Buitenhuys, 2005; Moszczynska et al., 2004). Meta-analysis revealed moderate effects for basic motor functioning (i.e. fine-motor speed and coordination) and cognitive processing speed

(Scott et al., 2007), however with abstinence these effects may recover over years and such long-term studies are lacking.

Human studies are confounded especially by a lack of information on cognitive and functional tests prior to METH use, polydrug use and variation in the amount and extent of drug exposures. Furthermore, the effects of METH may be modulated by genetic factors of the individual which cannot be extensively assessed. One study, for example, has shown that patients with METH psychosis lasting for 1 month or more after discontinuing METH may be associated with a polymorphism in the hDAT1 gene (SLC6A3) encoding the dopamine transporter (Ujike et al., 2003).

Several factors add complexity to understanding the stimulatory effects of amphetamines upon monoamines and the associated chronic toxicities. These factors include the multiple 128

receptor subtypes that exist for NE, DA and SE, with distinct binding affinities, second- messenger effects and central nervous system distribution, together with the added dimension that neuronal pathways can interact with each other; for example, monoamine neurons stimulate and/or inhibit excitatory glutamate neurons and inhibitory gamma-aminobutyric acid (GABA) neurons to alter toxicity (Burrows and Meshul, 1999; Lehmann and Lehmann, 2007; Bowyer et al., 1991; Pu et al., 1996).

129

Table 15. Effects of METH (Cohen et al., 2003; Cruickshank and Dyer, 2009; Sekine et al., 2001; Sulzer et al., 2005; Ujike et al., 2003; Wijetunga et al., 2003; Zweben et al., 2004) Acute Effects Adverse Effects Chronic Effects increased heart and respiratory hyperthermia insomnia rates, tremors paranoia increased blood pressure anxiety hallucinations increased body temperature cardiac arrhythmias psychotic symptoms acute myocardial infarction (e.g. delusions, obsessive increased energy stroke compulsive thoughts) alertness attentiveness death can occur depression, anxiety, suicide and violent behaviors euphoria enhanced self-esteem slowed motor function decreased appetite central nervous system impairments leading to cognitive and functional decline in learning and memory

130

1.5.7 EVIDENCE FROM ANIMAL AND HUMAN STUDIES FOR NEUROTOXICITY

METH-initiated neurotoxicity is evident in several neurotransmitter systems, but it is most notable in nigrostriatal dopaminergic pathways. Although an acute, moderate dose of

METH is unlikely to reduce DA stores permanently (Chan et al., 1994), high-doses in experimental animals (10-20 mg/kg) cause a significant reduction in levels of DA and SE and their metabolites, as well as tyrosine hydroxylase and tryptophan hydroxylase activity, enzymes involved in the synthesis of DA and other catecholamines, as well as their receptors and transporters (Axt and Molliver, 1991; Bakhit et al., 1981; Deng et al., 2007; Harvey et al.,

2000; Hotchkiss and Gibb, 1980; Jeng and Wells, 2010; O'Callaghan and Miller, 1994; Wagner et al., 1980). In the case of long-term deficits in these nerve terminal markers (weeks to months/years), destruction of nerve terminals is evident, although catecholamine neuron cell bodies themselves do not seem to be destroyed (Harvey et al., 2000; O'Callaghan and Miller,

1994; Wagner et al., 1980). In rodents, evidence of METH neurotoxic effects have been determined through morphological signs of axonal and nerve terminal degeneration as indicated by silver staining (O'Callaghan and Miller, 1994; Ricaurte et al., 1982). In rodents and nonhuman primates, administration of either a large single dose or repeated high doses of

METH or MDMA produces long-lasting deficits in nerve terminal markers, but amphetamines also produce astrogliosis and these markers co-localize with silver-staining, flurojade B staining as well as caspase activation, to indicate toxicity in both neuronal and non-neuronal cells

(O'Callaghan and Miller, 1994; Yu et al., 2004; Deng et al., 2007; Jayanthi et al., 2004).

Neurotoxic amphetamines also damage glutamatergic cortical neurons as well as gabanergic neurons in the striatum and olfactory bulb (Jayanthi et al., 2004; Jayanthi et al., 2009; Deng et al., 2007; Deng et al., 2001; Eisch et al., 1998; Pu et al., 1996). Furthermore, bioactivation and 131

oxidative damage can occur in brain cells other than neurons, for example in microglia, where the expression and activation of PHSs is associated with the neurotoxic properties of METH

(Choi et al., 2008; Eisch et al., 1998; Pu et al., 1996; Thomas and Kuhn, 2005b).

Regionally, METH most severely affects DA terminals in the striatum; however, toxicity to the cerebral cortex, hippocampus and olfactory bulb have been detected through decreased tyrosine hydroxylase and increased GFAP, apoptosis and oxidative macromolecular damage

(Deng et al., 2007; Jayanthi et al., 2004; Jayanthi et al., 2009; O'Callaghan and Miller, 1994).

The magnitude of the toxicity in the different brain regions is variable, and can be attributed to varied densities of DAT in these regions as well as differences in detoxification and repair pathways (Chu et al., 2008; Wells et al., 2010; Wong et al., 2008). For example, DA levels were more severely reduced in the caudate of the striatum than in the putamen (motor area) in postmortem tissue of METH abusers, which may relate to the low prevalence of Parkinsonian motor symptoms observed in chronic METH abusers, which affects the caudate (Moszczynska et al., 2004). A persistent reduction in most DA markers, as found with positron emission tomography (PET) studies of human METH users, revealed decreased D2 receptors that may represent downregulation from exposure to increased synaptic DA concentrations, however reduction in levels of DA, DAT and tyrosine hydroxylase have been found in postmortem striatum of chronic METH abusers (Volkow et al., 2001a; Volkow et al., 2001b; Volkow et al.,

2001d; Wilson et al., 1996). However, the effects of METH on other markers are highly variable. VMAT2 remained unchanged in one study of human chronic abusers (Wilson et al.,

1996), but was elevated in recently abstinent users (Boileau et al., 2008). Other studies demonstrated modest decreases in VMAT-2 binding in PET studies (Johanson et al., 2006). In rodent models, METH decreased striatal VMAT-2 ligand binding assessed 14 days after 132

treatment (Guilarte et al., 2003).

Several studies have shown brain abnormalities in METH abusers and in animals exposed to methamphetamine that are not limited to brain regions containing DA cells and their terminals, implicating non-DA mechanisms of METH toxicity (Ricaurte et al., 1982; Chung et al., 2007; Ernst et al., 2000; Kuczenski et al., 2007; Thompson et al., 2004; Volkow et al.,

2001c). Modest decreases in SERT have been observed in human chronic METH users; however, it is uncertain if this is due to nerve terminal damage or decreased protein expression

(Kish et al., 2009; Sekine et al., 2006). SE nerve terminals in various brain regions including hippocampus, prefrontal cortex, amygdala, and striatum are sensitive to the toxic effects of

METH (Seiden et al., 1988). Given the effects of METH on serotonergic as well as dopaminergic systems, perhaps other GABAergic, glutaminergic systems may be similarly affected, indicating a more global pattern of degeneration and loss of neuronal connectivity in

METH toxicity. Studies have shown deregulation of GABA systems with a decrease in the density of presynaptic immunolabeling for GABA one week post-drug, and an increase after four weeks (Burrows and Meshul, 1999). Furthermore, METH administration causes dopaminergic neuronal death within the olfactory bulb of mice (Deng et al., 2007), a region rich in dopaminergic neurons and regulatory GABA interneurons (Gheusi et al., 2000; Parrish-

Aungst et al., 2007).

METH can generate ROS such as superoxide anions, hydrogen peroxide and hydroxyl radicals, and can oxidize cellular macromolecules, such as proteins, lipids and DNA (Jayanthi et al., 1999; Jeng and Wells, 2010; Jeng et al., 2006). By causing cumulative oxidative macromolecular damage and/or by chronically altering signal transduction in the brain, ROS may contribute to the initiation and/or progression of a number of neurodegenerative diseases 133

and neurodegeneration associated with aging. Also, postmortem brains of chronic METH users had elevated levels of the lipid peroxidation products 4-hydroxynonenal and malondialdehyde in the caudate nucleus and to a lesser extent in the frontal cortex (Fitzmaurice et al., 2006). METH administration causes a marked increase in 2,3- and 2,5-dihydrobenzoic acid (the product of the reaction of salicylate with hydroxyl radicals) in striatal dialysate of rats (Giovanni et al., 1995).

Antioxidants (e.g., ascorbic acid or vitamin E) can attenuate METH-initiated toxicity (De Vito and Wagner, 1989), as can the overexpression of SOD (Hirata et al., 1996). Glutathione levels are also decreased by repeated administration of METH (Moszczynska et al., 2004). METH can cause major disruptions or even induce antioxidant enzymes in the brain (Jayanthi et al., 1998;

Jayanthi et al., 2009). Following a single-day administration of 10 mg/kg x 4 doses of METH to rats, SOD activity was decreased 16 hr later in the cortex, but not in the striatum, while activities of catalase and GPx were decreased in the striatal region (Jayanthi et al., 1998).

However, these studies did not evaluate the time course of activities, and may have missed possible induction effects due to transcriptional responses of Nrf2.

Experiments with PHS-1 knockout mice and PHS inhibitors revealed the important role of this isozyme in the molecular mechanism of ROS generation resulting from MDMA, METH and MDA administration, where knockouts, or mice treated with the PHS-1/2 inhibitor ASA, showed decreased DNA oxidation, nerve terminal degeneration and locomotor functional deficits when compared to PHS-normal wild-type mice or saline controls (Jeng et al., 2006;

Jeng and Wells, 2010). METH-initiated superoxide anions might combine with NO to yield

ONOO-, which can rapidly degrade to form •OH that oxidize proteins, lipids and DNA. This idea is supported by the protective effect of 7-nitroindazole, an inhibitor of neuronal nitric oxide synthase, in reducing METH-initiated depletion of dopamine and its metabolites and the loss of 134

dopamine transporter binding sites (Di Monte et al., 1996; Imam et al., 2001). Similarly, nNOS knockout mice are protected against METH neurotoxicity, exhibiting less depletion of dopaminergic markers (Itzhak et al., 1998).

A number of pathways have been implicated in MA-induced neurotoxicity, including production of reactive oxygen and nitrogen species, hyperthermia, or triggering of an apoptotic cascade dependent upon mitochondria, but the exact processes are still unclear (Cadet and

Krasnova, 2009). Taken together, these studies suggest that a variety of different acute or chronic neuropathological processes may be associated with METH neurotoxicity, including neuronal injury or death, astrocytosis, cellular membrane alterations, and dysregulation of energy metabolism. From these studies, long-term neurotoxic effects are seen in animal models where direct degeneration has been assessed but toxicity markers in human brain have been inconsistent with respect to assessing long-term, irreversible neurotoxicity. The complexity in assessing neurotoxicity is complicated by the fact that the markers may change regionally and over time; for example, DAT density may return to normal slowly during prolonged drug abstinence (Volkow et al., 2001b). The mechanisms of recovery are unclear, but may be partly explained by the sprouting of remaining axons or a compensatory increase in monoamine levels as seen in rats treated with METH (Cass and Manning, 1999). However, even if DAT density normalizes following abstinence, cognitive deficits may still persist (Volkow et al., 2001b).

Also, VMAT2 may redistribute from vesicles to the plasma membrane, changing its regional localization (Fleckenstein et al., 2009). Importantly, direct evidence in humans for the loss of nerve terminals and/or their corresponding cell bodies or the cells surrounding these neurons, such as astrocytes and microglia, have not been provided (Davidson et al., 2001; Harvey et al.,

2000). Thus, multiple markers are needed in human studies, and the human studies must be 135

complemented by more comprehensive molecular and biochemical studies in animal models, to fully elucidate the molecular mechanisms leading to neurotoxicity.

136

CHAPTER 2: STUDIES 137

2.1 STUDY 1: PROSTAGLANDIN H SYNTHASE-1-CATALYZED BIOACTIVATION OF NEUROTRANSMITTERS, THEIR PRECURSORS, AND METABOLITES: OXIDATIVE DNA DAMAGE AND ELECTRON SPIN RESONANCE SPECTROSCOPY STUDIESa

Luisa L. Goncalves1, Annmarie Ramkissoon2 and Peter G. Wells

a. This research was supported by a grant from the Canadian Institutes of Health Research (CIHR). L.L.G. was supported by a Postdoctoral Fellowship from the Portuguese Foundation for Science and Technology (FCT). A.R. was supported by a Doctoral Fellowship from the CIHR/Rx&D Health Research Foundation.

1. All experiments related to ESR and horseradish peroxidase were carried out and written by Luisa Goncalves.

2. Annmarie Ramkissoon performed, analyzed and wrote DNA oxidation sections and edited the manuscript.

This manuscript is reproduced with permission from: Goncalves, L. L., Ramkissoon, A. and Wells, P. G. (2009). Prostaglandin H synthase-1-catalyzed bioactivation of neurotransmitters, their precursors, and metabolites: oxidative DNA damage and electron spin resonance spectroscopy studies. Chem Res Toxicol 22(5): 842-852. Copyright 2009. American Chemical Society. DOI: http://pubs.acs.org/doi/abs/10.1021/tx800423s

138

2.1.1 ABSTRACT

The role of prostaglandin H synthase-1 (PHS-1) and a related model enzyme, horseradish peroxidase (HRP), in catalyzing the bioactivation of dopamine (DA) and epinephrine and their precursors and metabolites to potential neurodegenerative free radical intermediates was examined. To determine the potential contribution of PHS-dependent reactive oxygen species (ROS) formation, the neurotransmitter DA or its precursor and metabolites were incubated in vitro with purified ovine PHS-1 and calf thymus DNA. DA, its l- dihydroxyphenylalanine (l-DOPA), precursor, and its dihydroxyphenylacetic acid (DOPAC) metabolite were excellent PHS-1 substrates, resulting in PHS-1-dependent ROS formation that initiated oxidative DNA damage, selectively quantified as 8-oxo-2′-deoxyguanosine. Most substrates generated isotropic electron spin resonance (ESR) spectra with a resolved hyperfine structure attributable to ortho-semiquinone free radical intermediates upon autoxidation at pH 6, with up to a 18-fold increase via HRP-catalyzed oxidation. Remarkably, HRP-mediated oxidation of DOPAC and dihydroxymandelic acid (DHMA) produced asymmetric ESR spectra characteristic of an immobilized radical, possibly due to free radical intermediates and melanin or melanin-like polymers. These results show that the precursors and metabolites of endogenous neurotransmitters, while inactive in receptor binding assays, may actually play an important role in free radical formation. Additionally, ROS generated by PHS-catalyzed bioactivation produce oxidative DNA damage in the central nervous system, which may initiate neurodegeneration associated with aging. 139

2.1.2 INTRODUCTION

The brain dopaminergic system is implicated in a variety of physiological and pathological processes. An imbalance between dopaminergic neurotransmission and dopamine

(DA) receptors is associated with numerous neuropsychiatric (e.g., schizophrenia and depression) and neuropathological disorders, such as Parkinson‟s, Alzheimer‟s, and

Huntington‟s diseases (Segura Aguilar and Kostrzewa, 2004; Zhu, 2004). Although it is generally accepted that free radicals are involved in the neurodegenerative process, the exact mechanism of neurodegeneration in vivo is not fully understood. It has been postulated that the oxidation of endogenous substrates, particularly the oxidation of catecholamine neurotransmitters (such as DA, epinephrine, and norepinephrine) may contribute to the degeneration of selective brain regions, as observed for example in the nigro-striatal system of

Parkinson‟s patients (Graumann et al., 2002; Herlenius and Lagercrantz, 2004; Mattammal et al., 1995; Paris et al., 2005; Tse et al., 1976; Volicer and Crino, 1990). Approximately 95% of the released DA is quickly taken up into the neurons (or nerve terminals) and then into the storage vesicles for reuse (Zhu, 2004). Although it is not known exactly what percentage of DA molecules inside a neuron is taken up by the storage vesicles, it has been suggested that the fraction of the intracellular DA that is not taken up into the vesicles might be the cause of cytotoxic damage to dopaminergic neurons (Zhu, 2004).

DA can autooxidize via the loss of one electron, generating semiquinone free radical intermediates that rapidly react to produce reactive oxygen species (ROS) such as superoxide and hydroxyl radicals (Ishii and Fridovich, 1990; Klegeris et al., 1995; Tse et al., 1976). The ensuing redox cycling between the catechols/catecholamines and their quinones/semiquinones can continuously produce large amounts of ROS. Both DA quinone and semiquinone 140

intermediates and superoxide and hydroxyl radicals are cytotoxic and potentially genotoxic

(Bolton et al., 2000; Cavalieri et al., 2002a; Cavalieri et al., 2002b; Fornstedt et al., 1990).

Alternatively, several types of oxidizing enzymes (e.g., cytochromes P450) and peroxidases such as myeloperoxidase, lactoperoxidase, horseradish peroxidase (HRP), and prostaglandin H synthase (PHS) can bioactivate DA and other biogenic amines to the same reactive intermediates, which may bind proteins and DNA (Hastings, 1995; Napolitano et al., 1995).

Additionally, these biogenic amines, when metabolized by peroxidases, are known to form prooxidant radicals, which cooxidize cellular antioxidants, such as ascorbate, NADH, or cysteine (Alanko et al., 1999; Siraki and O'Brien, 2002). The involvement of HRP and PHS in the bioactivation of xenobiotics (Boyd and Eling, 1985; Jones et al., 1998; Kubow and Wells,

1989; Parman et al., 1998; Parman and Wells, 2002; Tafazoli and O'Brien, 2005; Jeng et al.,

2006) and endogenous substrates (Adak et al., 1998; Hastings, 1995; Mattammal et al., 1995;

McCormick et al., 1998) is well-documented. PHS is a dual-function enzyme, having a cyclooxygenase and a hydroperoxidase component. It is constitutively expressed in the brain, and increased brain expression and activity of PHS have been reported in aging and neurodegenerative diseases, such as Parkinson‟s and Alzheimer‟s diseases (Hastings, 1995;

Mattammal et al., 1995). It is generally accepted that the oxidation of a substrate, designated by

AH, by the hydroperoxidase component of PHS in the presence of peroxide (AOOH), includes the following initial steps (ref (Koshkin and Dunford, 1998) and references cited therein), enzyme + AOOH → E-I + AOH (1)

E-I + AH →E-II + A• (2)

E-II + AH → enzyme + A• (3) where E-I and E-II are compounds I and II, respectively. In this reaction, a broad spectrum of 141

chemicals can serve as electron donors, such as aromatic amines, polycyclic aromatic hydrocarbons, and phenolic compounds (including catechols and catecholamines) (Vogel,

2000).

Our laboratory has previously shown in vitro and in vivo that xenobiotics such as phenytoin and amphetamines are cosubstrates during the reduction of hydroperoxides in eicosanoid biosynthesis, the reducing cosubstrate being oxidized by PHS and/or lipoxygenases to N- and C-centered reactive free radical intermediates, characterized by electron spin resonance (ESR) spectroscopy (Jeng et al., 2006; Parman et al., 1998; Parman and Wells, 2002).

These intermediates generate ROS that oxidatively damage brain DNA, proteins, and lipids causing dopaminergic nerve terminal degeneration and functional deficits. Additionally, in vitro studies have shown that DA and SE can serve as cofactors for PHS-1 (Hastings, 1995). More recently, a preliminary study in our laboratory found that, in addition to DA, its precursor, l- dihydroxyphenylalanine (l-DOPA), and epinephrine were excellent PHS-1 substrates, resulting in PHS-dependent ROS formation that initiated the oxidation of 2′-deoxyguanosine (2′-dG) to 8- oxo-2′-deoxyguanosine (8-oxo-dG) (Ramkissoon et al., 2005). Accordingly, we postulated that the precursors and metabolites of neurotransmitters, while inactive in binding to neurotransmitter receptors, may nevertheless be bioactivated by PHS to ROS-generating free radical intermediates that cause oxidative macromolecular damage, possibly contributing to neurodegenerative diseases associated with aging.

This hypothesis is consistent with several observations, including the following: (1)

Some neurotransmitter precursors (e.g., l-DOPA), although inactive in receptor binding assays, are bioactivated by PHS-1 (Ramkissoon et al., 2005). (2) Neurotransmitter precursors and metabolites are distributed throughout the brain and are found in higher amounts in the striatum, 142

substantia nigra pars compacta, and hippocampus (Cannazza et al., 2005; Cheng et al., 1993;

Kalant, 2001; Loutelier-Bourhis et al., 2004). (3) Among different brain regions, there is a positive correlation between increasing levels of PHS and increasing levels of oxidative DNA damage (Jeng et al., 2006). To investigate this hypothesis, we determined the efficacy of neurotransmitters, their precursors, and metabolites to serve as substrates for HRP- and PHS-1- catalyzed bioactivation to free radical intermediates using an in vitro system, as successfully employed for phenytoin and amphetamines (Jeng et al., 2006; Parman et al., 1998). The chemical nature of the putative free radical intermediates was evaluated by spin stabilization- and spin trapping-ESR spectroscopy and by ultraviolet/visible (UV/vis) spectroscopy. An ESR- spin stabilization approach allows the scavenging of a semiquinone by divalent metal ions such as Zn2+ (Jeng et al., 2006; Kalyanaraman et al., 1984b; Kalyanaraman et al., 1984a; Parman et al., 1998), which form chelation complexes with the free radicals to enhance their stability. The use of conventional spin traps such as nitroso-compounds and nitrones to detect these intermediary species do not usually produce detectable signals in the ESR time scale (Felix and

Sealy, 1981; Jeng et al., 2006; Kalyanaraman et al., 1984b; Kalyanaraman et al., 1984a; Parman et al., 1998). UV/vis spectroscopy was used to determine whether HRP and PHS-1 catalyze the bioactivation of these endogenous substrates to the same reactive intermediates. PHS-1 and

DNA were coincubated with DA, its precursor, and metabolites in vitro to determine whether the free radical intermediates formed could enhance ROS formation and cause oxidative DNA damage. 143

2.1.3 MATERIALS AND METHODS Materials

All chemicals used in this study, including DA, l-DOPA, 3-methoxytyramine (MTA), dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA), dihydroxymandelic acid

(DHMA), normetanephrine, 2′-dG, hematin, calf thymus DNA, arachidonic acid (AA),

5,8,11,14-eicosatetraynoic acid (ETYA), 5,5-dimethyl-1-pyrroline-1-oxide (DMPO), α-phenyl-

N-tert-butylnitrone (PBN), and nuclease P1 were of analytical or HPLC grade, were purchased from Sigma-Aldrich (Oakville, ON, Canada), and were used as received. Redistilled phenol was from Aldrich Chemical Co. (Oakville, ON, Canada). Calf intestine alkaline phosphatase was obtained from Roche (Laval, Quebec, Canada). Hydrogen peroxide (H2O2), obtained as a 30%

(wt/wt) solution, was freshly prepared, and the final concentration was determined using ε(230 nm) = 72.4 M1 cm−1 (38). HRP (type VI lyophilized powder; EC 1.11.1.7; donor-H2O2 oxidoreductase; 250−330 U/mg of solid), superoxide dismutase (SOD) from horseradish (EC

1.15.1.1; 1218 U/mg of solid), and catalase (E.C.1.11.1.6; 3260 U/mg solid) were also from

Sigma-Aldrich. These enzymes were used as received, and dilutions from the stock solutions in

Na2HPO4 buffer (100 mM, pH 7.4) were performed as needed. Purified ovine PHS-1

(EC.1.14.99.1, 185714 U/mg) and 8-oxo-dG were obtained from Cayman Chemical Co. (Ann

Arbor, MI). A stock solution of PHS-1 (10000 U/mL) in phosphate buffer (80 mM, pH 7.9) was prepared, preserved at -80 °C, and subsequently used in the incubations. All of the substrates and the spin traps were freshly prepared, flushed with nitrogen, and stored at -80 °C until the moment that they were used.

ESR Spectroscopy 144

ESR-spin stabilization measurements were carried out at room temperature in solutions contained in a quartz flat cell, using a Bruker BioSpin GmbH X band spectrometer. Instrument parameters typically used to obtain the spectra were as follows: microwave power, 31.7 mW; modulation amplitude, 0.2 G; time constant, 5.120 ms; conversion time, 5.120 ms; sweep time,

5.243 s; receiver gain, 5.0 × 10(5); center field, 3486 G; frequency 9.765 GHz; and number of scans, 100. Spin trapping experiments were also made at room temperature in solutions contained in capillary tubes, using a Bruker ER-200 X band spectrometer equipped with an ER

4123D_188r resonator. Instrument parameters typically used to obtained the spectra were as follows: microwave power, 31 mW; modulation amplitude, 0.99 G; scan time, 5 or 30 min; time constant, 2.56 s; sweep time, 5.243 s; receiver gain, 5.0 × 10(5); center field, 3473 G; and frequency 9.77 GHz. The spectrometer was initially calibrated using the Strong Pitch standard.

Hyperfine coupling constants (hfcc) were measured (to ±0.1 G) directly from the magnetic field separation and confirmed by computer simulation. ESR spectra were simulated using WINEPR

SimFonia software from Bruker. Estimates of the relative radical production were made through the comparison of the normalized double integrals (DI/N) of the ESR-spin stabilization spectra obtained upon autoxidation and HRP-mediated oxidation of the neurotransmitters, their precursors, and metabolites.

ESR−Spin Stabilization Experiments

Experimental conditions used in these assays were adapted from the literature

(Kalyanaraman et al., 1984a). To evaluate the ESR signal resulting from the autoxidation of the neurotransmitter substrates, a solution containing the neurotransmitters, their precursors, and metabolites (6 mM) and zinc acetate (0.5 M) in acetate buffer (200 mM, pH 6.0) was prepared, 145

and the spectra were immediately recorded. The enzymatic oxidation of the substrates was initiated with the stepwise addition of H2O2 (5 mM) and HRP (100 U/mL), and the spectra were immediately recorded. Because of the formation of a dark brown precipitate, some incubation mixtures were filtrated through a LC13 PVDF membrane (0.2 μm) before acquiring the spectra.

ESR−Spin Trapping Experiments

DMPO was dissolved in Na2HPO4 buffer (100 mM, pH 7.4) in the dark to a final concentration of 2.2 M. Stock solutions were kept under argon and stored at -80 °C. The DMPO solutions and the incubation mixtures were also kept in foil-covered tubes, and room light was kept to a minimum. A different spin trap, PBN, was also used to confirm the structural assignments of the free radicals. In this case, PBN was dissolved in deionized H2O to a final concentration 56 mM. Stock solutions were freshly prepared, flushed with argon, and kept in ice in the dark until the moment that they were used.Typical reaction mixtures for spin trapping experiments with DMPO contained the substrate (1 mM), HRP (100 U/mL), diethylenetriaminepenta-acetic acid (DTPA, 100 μM) to chelate residual redox active metals, spin trap (100 mM), and H2O2 (4 mM) in sodium phosphate buffer (100 mM, pH 7.4). All incubations were performed in the dark at room temperature. The reaction was initiated by addition of H2O2 to a mixture containing all other components. Samples were protected from light with aluminum foil and immediately analyzed by ESR spectroscopy. Spectral assignments were confirmed using 75 mM substrate and 30 mM H2O2. PHS-1 (1000 U/mL) was incubated with hematin (1.0 μM) for 1 min at 37 °C in 80 mM potassium phosphate buffer, pH 7.9. After the addition of the substrate (1 mM) and the free radical spin trap DMPO or PBN (1 mM), AA 146

(140 μM) was added to initiate the reaction. These conditions have been used previously to reliably activate the enzyme while limiting its ability to self-inactivate (Jeng et al., 2006;

Parman et al., 1998). Samples were covered with foil and subsequently analyzed by ESR and by

UV/vis spectroscopy as described in the following section.To detect the presence of paramagnetic impurities in the incubations, control experiments were also performed using the same experimental conditions as described above. In this case, we ran several assays where each of the following reagents was omitted from the incubations at the time: neurotransmitter, their precursor or metabolite, and HRP or PHS-1. Additionally, to evaluate the contribution from autoxidation of each substrate, we ran time-course experiments (5 and 30 min) where only the substrate and DMPO were present in the incubation. To block PHS-1-catalyzed bioactivation of

L-DOPA, the PHS-1 inhibitor ETYA (40 μM) was incubated with the enzyme at 37 °C for 1 min prior the addition of the substrate and AA.

Spectrophotometric Assays

UV/vis spectra were recorded in a SpectraMAX Plus 384 (Molecular Devices Corp.).

Reaction mixtures obtained upon HRP and PHS-1-mediated oxidation were analyzed in a 96 well plate under aerobic conditions at 37 °C. Spectra were scanned (λmax = 220−700 nm) and recorded every 2 min for 30 min against control samples where the substrate and phenol were omitted from the incubations. Experimental conditions used for each of the enzymes were as follows: HRP (100 U/mL); H2O2 (0.3 mM); neurotransmitters, their precursors, and metabolites

(0.2 mM) in phosphate buffer (100 mM, pH 7.4); PHS-1 (1000 U/mL); hematin (1 μm); phenol

(0.5 mM); AA (140 μM); and neurotransmitters, their precursors, and metabolites (0.2 mM) in phosphate buffer (80 mM, pH 7.9). 147

Ovine PHS-1-Dependent DNA Oxidation by Neurotransmitters, Their Precursors, and

Metabolites

Calf thymus DNA was dissolved overnight in 80 mM potassium phosphate buffer, pH

7.9, at 65 °C. PHS-1 (1000 U/mL) was incubated in 80 mM potassium phosphate buffer, pH 7.9, with phenol (0.5 mM) and hematin (1 μM) for 1 min at 37 °C. Neurotransmitters, precursors, or metabolites (500 μM in 80 mM potassium phosphate buffer, pH 7.9) were added to the mixture along with dissolved DNA (2 mg/mL). AA (140 μM) was added to initiate the reaction or omitted to inhibit PHS-1 activity. The reaction mixture was incubated at 37 °C with gentle agitation for 30 min. The controls for the incubations contained all reagents except the drug

(vehicle control) and all reagents except the enzyme PHS-1. The reaction was stopped on ice, and DNA was precipitated with cold ammonium acetate (10 M) and 100% ethanol. Samples were spun at 10000g for 10 min, and the supernatant was discarded by aspiration. The pellet was washed twice with 70% ethanol and spun at 10000g for 5 min, and the supernatant was discarded by aspiration. The DNA pellet was then dissolved overnight in sodium acetate (20 mM, pH 4.8) at 65 °C. DNA was digested with nuclease P1 (8 U) at 37 °C. After 30 min, Tris-

HCl (1 M, pH 8) was added followed by alkaline phosphatase (8 U), and this was incubated at

37 °C for 1 h. Samples were then filtered using Microcon-YM 10 filters (Millipore Canada Ltd.) and were stored at −80 °C until analysis.

Detection of 8-Hydroxy-2′-deoxyguanosine and 2′-dG

8-Hydroxy-2′-deoxyguanosine was quantified using an isocratic Series 200 HPLC system (Perkin-Elmer Instruments LLC, Shelton, CT) with electrochemical detection. It was 148

equipped with a 5 μm Exsil 80A-ODS C-18 column (50 mm × 4.6 mm, Jones Chromatography,

Ltd.), an electrochemical detector (Coulochem II), a guard cell (model 5020), an analytical cell

(model 5010) (Coulochem, ESA Inc., Chelmsford, MA), and an integrator (Perkin-Elmer NCI

900 Interface). The filtered samples were injected into the HPLC system and eluted using a mobile phase containing ammonium acetate buffer (50 mM, pH 5.2) with 5% methanol at a flow rate of 0.8 mL/min, with a detector oxidation potential of +0.4 V. Chromatographs were analyzed using the TotalChrom chromatography software version 6.2.0 (Perkin-Elmer

Instruments LLC). 2′-dG was quantified using HPLC with UV/vis detection at 260 and 280 nm

(Perkin-Elmer Instruments LLC) and a mobile phase containing potassium phosphate buffer (50 mM, pH 5.5) with 5% methanol.

Statistical Analysis

Multiple comparisons among groups were analyzed by one-way ANOVA with a subsequent Tukey‟s test (GraphPad InStat3.05, GraphPad Software, Inc., San Rafael, CA). The level of significance was determined to be at P < 0.05. 149

2.1.4 RESULTS In Vitro Bioactivation of Neurotransmitters, Their Precursors, and Metabolites to a Free

Radical Intermediate by HRP and PHS-1

In the absence of HRP, the incubation of the two neurotransmitters DA and epinephrine and their precursors and metabolites l-DOPA, DOPAC, and DHMA gave rise to an isotropic

ESR spectrum with resolved hyperfine structure, reflecting autoxidation (Figure 1B). Despite considerable overlapping of the spectral lines, computer-simulated spectra helped to identify these radical species as noncyclized Zn2+-complexed primary ortho-semiquinones (Felix and

Sealy, 1981; Felix and Sealy, 1982; Kalyanaraman et al., 1984a; Kalyanaraman et al., 1987;

Kalyanaraman et al., 1984c). Computer simulation of the spectra (Figure 1C) gave the ESR parameters shown in Table 1. These results are in good agreement with published data for zinc- complexed primary semiquinones for epinephrine (Kalyanaraman et al., 1984a), DA (Felix and

Sealy, 1981), and related molecules (Felix and Sealy, 1981; Kalyanaraman et al., 1984a;

Kalyanaraman et al., 1987). The isotropic g value found for all radicals was 2.0039, suggesting their identical structure. HVA and normetanephrine did not generate radical intermediates due to autoxidation.

In the presence of HRP, the addition of H2O2 to the incubation mixture containing the above-mentioned substrates and Zn2+ in acetate buffer produced marked changes in the ESR spectral patterns as compared to those from autoxidation (Figure 1A). These changes included enhanced signal intensity as compared to those obtained upon autoxidation, along with an increased line width and/or unresolved hyperfine splittings. These latter observations may reflect the formation of more complex structures, such as dimers and/or melanin-like polymers

(Felix and Sealy, 1981; Graham and Jeffs, 1977; Kalyanaraman et al., 1982; Kalyanaraman et 150

al., 1987). The most visible change was observed for DOPAC and DHMA, where asymmetric

ESR spectra characteristic of an immobilized radical were observed. Measurements of the relative radical production for Zn2+-complexed semiquinones obtained by autoxidation (Figure

2) revealed that all neurotransmitters, precursors, and metabolites, except HVA and normetanephrine, exhibited limited autoxidation to free radical intermediates (ortho- semiquinones). This limited radical production via autoxidation was in contrast to substantial

HRP-catalyzed oxidation, with a maximal 18-fold increase for HRP-catalyzed radical production from DOPAC, as shown by the normalized double integral ESR signal. HRP- catalyzed free radical formation from DA, its precursor l-DOPA, and its metabolite DOPAC was 3−8 times greater than that from epinephrine and its DHMA metabolite.

To explore the formation of superoxide and/or hydroxyl free radical intermediates, we also ran spin trapping experiments using HRP, PHS-1, DMPO, and PBN. Unlike PBN, DMPO yields distinct and more persistent spin adducts with the primary O2•- and HO• radicals

(DMPO-OOH and DMPO-OH, respectively). DMPO-OH may also arise from the initial trapping of O2•-, giving DMPO-OOH, which is reported to readily decompose to DMPO-OH at physiological pH (Frejaville et al., 1995; Pietri et al., 1998). Spectral confirmation for HRP- catalyzed formation of free radical spin adducts was obtained for DA and its precursor l-DOPA

(Figure 3). Elimination of any component of the spin trapping reaction mixture resulted in the complete loss or diminution of the spectra. The ESR signal revealed the presence of two different radical spin adducts (RSA): a triplet (RSA1) and a large doublet of triplets (RSA2).

The RSA2 was observed for DA, l-DOPA, and MTA when incubated with active HRP (Table

2). Additionally, in the spectra of DA, l-DOPA, epinephrine, and normetanephrine incubated with HRP, a quartet was also observed at aN = aβH = 14.88 G (RSA3) characteristic of DMPO- 151

OH• spin adducts (Table 2). Considering that artifacts with DMPO have been reported under a variety of experimental conditions (Finkelstein et al., 1980; Hanna et al., 1992; Janzen et al.,

1989; Pietri et al., 1989), we assigned the ESR signal corresponding to RSA1 to an aminoxyl radical from DMPO, which is an artifact due to an oxidized derivative of DMPO and/or due to oxidized DMPO impurities, such as hydroxylamine. Similarly, we also consider the large doublet of triplets signal produced by RSA2, noteworthy for further investigation.

The addition of SOD (1 μM) to the incubations did not measurably affect RSA2 or

RSA3 spin adducts but produced an increase in the signal amplitude of RSA1 for DA and l-

DOPA. In contrast, the addition of catalase (62 μM) inhibited RSA2 and RSA3 spin adducts

(results not shown), suggesting that both radical species are formed through a peroxidative mechanism and are H2O2-dependent. These results may also suggest that HO• may be formed by a Fenton-like reaction.

Similar results were obtained when PHS-1 was used instead of HRP (Figure 3). PHS-1- catalyzed oxidation of DA, l-DOPA, epinephrine, and normetanephrine produced spectra with the same hyperfine splitting pattern: a triplet that corresponded to RSA1 of HRP reactions and a large doublet of triplets that corresponded to RSA2 of HRP reactions summarized in Table 3.

Additionally, in the spectra of DA, l-DOPA, epinephrine, and normetanephrine incubated with

PHS, a quartet was also observed at aN = aβH = 14.88 G (RSA3) characteristic of DMPO-OH• spin adducts (Table 3). In contrast, the ESR spectra for the control experiments showed that neither RSA1 nor RSA2 are formed in phosphate buffer containing only DMPO (Figure 3F) or without substrate (Figure 3G). Preincubation of PHS-1 with the PHS/lipoxygenase inhibitor

ETYA (40 μM) abolished the ESR signal corresponding to RSA2 and RSA3 (Figure 3E). The

40 μM concentration of ETYA is above the Ki value for PHS inhibition in isolated cells and 152

purified enzyme preparations (Hammarstrom, 1977).

To evaluate the presence of radical species generated by autoxidation of the substrates, we also ran time-dependent control incubations (5 and 30 min) in which only the substrate,

DMPO, and phosphate buffer were present, without a bioactivating enzyme. The spectra presented for these controls (without either HRP or PHS-1) (Figure 3H,I) did not show the presence of any radical other than a triplet corresponding to RSA1 in the first 5 min (results not shown). These results suggest that the radical species RSA1 is mainly produced via HRP- or

PHS-1-catalyzed oxidation. Weak ESR signals corresponding to RSA2 were initially visible only after 30 min (Figure 3). To investigate the genesis of the postulated C-centered free radical corresponding to RSA2, we repeated the assays with PHS-1, epinephrine, and l-DOPA using

PBN. The ESR signal obtained for epinephrine (Figure 4B) is composed of two superimposed, overlapping spectra, with hfcc values that are in good agreement with published data for C- centered spin adducts (Buettner, 1987). The ESR spectra of l-DOPA also indicated the presence of C-centered radicals with hfcc values similar to those observed for epinephrine. There was no detectable radical adduct of PBN in the control incubations (without substrate) for these compounds (Figure 4A). On the basis of the results of these studies, we believe that the RSA2 detected in reactions with DMPO and HRP or PHS-1 are C-centered free radicals that are derived from the oxidation of the substrates DA, l-DOPA, MTA, epinephrine, and normetanephrine.

Spectrophotometric Assays

The spectrophotometric profile of the incubation mixtures resulting from HRP and PHS-

1-catalyzed oxidation of DA and epinephrine precursors and metabolites was analyzed by 153

UV/vis spectroscopy between 220 and 700 nm, under physiological pH, in a time scale of 30 min. The UV/vis spectra obtained for the 2 min time point are shown in Figures 5 (HRP) and 6

(PHS-1). Upon catalysis, both enzymes produced colored incubation mixtures, from light yellow to brown, depending on the substrate and the enzyme. Although conversion of the substrates to the putative intermediates (ortho-semiquinones/quinones and aminochromes) occurred at higher rates for HRP, time-dependent incubations did not show noticeable differences in the nature of the chromophores detected for each enzyme in a time scale of 30 min (results not shown).

Neutral aqueous solutions of catechols and catecholamines are known to have absorption spectra in the UV region with two maxima at ca. 230 and 280 nm. In fact, upon HRP- and PHS-1- catalyzed oxidation of substrates, the absorbance was shifted to longer wavelengths, suggesting the formation of new products/chromophores.

Time-dependent incubation of l-DOPA and DA with HRP revealed the early (within 2 min) existence of a chromophore with λmax similar to that given in the literature for their aminochromes (a cyclized quinone) at 300 and 480 nm (Table 4 and Figure 5) (Graham et al.,

1978; Segura-Aguilar et al., 1998). Nevertheless, the products of DOPAC and DHMA at λmax

= 400 nm, respectively, are assumed to be ortho-quinones (Graham et al., 1978). DHMA produced an intense absorption band at λmax = 340 nm (Figure 5), which is near to λmax = 335 nm observed for human melanin pigment (Das et al., 1978). HVA and normetanephrine did not produce detectable amounts of either intermediate.

As indicated above, with PHS-1, important differences in the rates of oxidation were observed among the substrates tested (compare, for example, Figures 5 and 6). PHS-1- catalyzed oxidation of l-DOPA produced a band at λmax = 400 nm assumed to be due to its ortho-quinone, with no detectable amounts of aminochrome. In contrast, for DA, new peaks 154

were formed for both the quinone at λmax = 400 nm and the aminochrome at λmax = 480 nm.2

Similarly, the product of DOPAC, an ortho-quinone, was observed at λmax = 400 nm. No absorption bands were detected for HVA and normetanephrine.

Ovine PHS-1-Dependent DNA Oxidation by Neurotransmitters, Their Precursors, and

Metabolites

DA, l-DOPA, and DOPAC (500 μM) all were bioactivated by ovine PHS-1 in vitro to a free radical reactive intermediate that enhanced ROS formation, resulting in the oxidation of

DNA to produce 8-oxo-dG (Figure 7). While there was evidence of DA autoxidation, as illustrated in the reaction without AA, reactions containing active PHS-1 showed a 72% increase in DNA oxidation as compared to reactions containing inhibited enzyme (p < 0.05). Similarly, l-

DOPA with active PHS-1 caused a 70% increase in DNA oxidation as compared to reactions with inhibited enzyme (p < 0.05). DOPAC also showed a 38-62% increase in DNA oxidation as compared to reactions containing no PHS-1 (p < 0.05) and no AA (p < 0.05), respectively.

Under these reaction conditions at 500 μM, HVA and MTA did not lead to a PHS-1-catalyzed increase in DNA oxidation. 155

Table 1. ESR Data for Zn2+-Complexed o-Semiquinones Obtained from the Autooxidation of Neurotransmitters, Their Precursors, and Metabolitesa. a Radicals were generated from 6 mM solutions of substrate in acetate buffer (200 mM, pH 6) containing Zn2+ (0.5 M). b The ring protons are numbered as 3, 5, and 6 (the side chain substituent is at position 4), and protons in the side chain are designated as β (attached to the carbon atom adjacent to the ring) and γ. c Not detected. 156

Table 2. Hyperfine Splitting Constants Obtained by EPR-spin Trapping Experiments for HRP-Catalyzed Oxidation of Neurotransmitters, Their Precursors and Metabolitesa a Radicals were generated from 1 mM solutions of substrate in phosphate buffer, pH 7.4, by HRP-mediated oxidation (100 U/mL). The in vitro system contained H2O2 (4 mM), substrate, and DMPO (100 mM). b Not detected. 157

Table 3. Hyperfine Splitting Constants Obtained by EPR-spin Trapping Experiments for PHS-Catalyzed Oxidation of Neurotransmitters, Their Precursors, and Metabolitesa a Radicals were generated from 1 mM solutions of substrate in phosphate buffer, pH 7.9, by PHS-1-catalyzed oxidation (1000 U/mL). The in vitro system contained PHS-1 and hematin (1.0 μM). After preincubation for 1 min at 37 °C, substrate, DMPO (1mM), and AA (140 μM) were added to initiate the reaction. b Not detected. 158

Table 4. UV/Vis Spectral Characterization of Products Obtained upon HRP- and PHS-1- Mediated Oxidation of Neurotransmitters, Their Precursors, and Metabolitesa. a The wavelengths of maximal absorption obtained at 2 min of incubation time are given for HRP- mediated oxidation at pH 7.4 (100 mM phosphate buffer) and for PHS-1-mediated oxidation at pH 7.9 (80 mM phosphate buffer). Other experimental conditions are described in the Materials and Methods. b Not determined. c No signal. 159

Figure 1. ESR spectra of Zn2+-complexed o-semiquinones generated during HRP- catalyzed oxidation. (A) By autoxidation (B) and by computer simulation (C) of (1) DOPAC, (2) DHMA, (3) DA, (4) L-DOPA, (5) epinephrine, (6) HVA, and (7) normetanephrine. Control without substrate (D). Spectra were acquired according to conditions described in Materials and Methods. 160

Figure 2. Relative radical production for Zn2+-complexed semiquinones obtained by autoxidation and by HRP-mediated oxidation of DA, its precursor, and metabolites and of epinephrine and its metabolites. The relative radical production was estimated using the normalized double integral (DI/N) obtained from the spin stabilization ESR spectra. 161

Figure 3. X-band ESR spectra of radical species trapped with DMPO. (A) DA, (B) L- DOPA, (C) DOPAC, (D) epinephrine, (E) incubation with L-DOPA and ETYA, (F) control with DMPO, (G) control without substrate, (H) control without HRP, and (I) control without PHS-1. Radical species are identified as ([) RSA1, (*) RSA2, and (X) RSA3. Spectra were acquired according to conditions described in Materials and Methods. 162

Figure 4. X-band ESR spectra of radical species derived from epinephrine trapped with PBN. (A) control without substrate, (B) experimental spectrum obtained upon incubation of epinephrine with PHS-1, and (C) computer simulation of the X band ESR spectrum obtained in B. 163

Figure 5. UV/vis spectra of the products obtained by HRP-mediated oxidation of neurotransmitters, their precursors, and metabolites. HRP-mediated oxidation was performed at 37 °C using the following experimental conditions: HRP (100 U/mL), H2O2 (0.3 mM), and neurotransmitters, their precursors, and metabolites (0.2 mM) in phosphate buffer (100 mM, pH 7.4). Spectra presented were scanned from 220 to 700 nm and recorded at 2 min.

164

Figure 6. UV/vis spectra of the products obtained by PHS-1-mediated oxidation of neurotransmitters, their precursors, and metabolites. PHS-1-mediated oxidation was performed at 37 °C using the following experimental conditions: PHS-1 (1000 U/mL), hematin (1 μm), phenol (0.5 mM), AA (140 μM), and neurotransmitters, their precursors, and metabolites (0.2 mM) in phosphate buffer (80 mM, pH 7.9). Spectra presented were scanned from 220 to 700 nm and recorded at 2 min.

165

Figure 7. Ovine PHS-1-dependent DNA oxidation by neurotransmitters, their precursors, and metabolites. Groups include DA, L-DOPA, DOPAC, HVA, and MTA. Each reaction contained 1000 U/mL PHS-1, 1.0 μM hematin, and 0.5 mM phenol. After preincubation for 1 min at 37 °C, 500 μM neurotransmitter and precursor or metabolite, 2 mg/mL DNA, and 140 μM AA were incubated for 30 min at 37 °C. DNA was precipitated and digested, and oxidative DNA damage was quantified by the formation of 8-oxo-2′-deoxyguanosine (8-oxo-dG)/μg 2′- dG. The vehicle control incubation contained all components except the substrate. The No AA group represents inhibition of PHS-1 activity. 166

2.1.5 DISCUSSION

We investigated the hypothesis that the precursors and metabolites of endogenous neurotransmitters, in addition to the neurotransmitters themselves, may serve as substrates for

PHS-1-catalyzed bioactivation to potential neurodegenerative free radical intermediates. The neurodegenerative potential of neurotransmitter precursors and metabolites is particularly important because such endogenous molecules are generally considered biologically inactive based upon neurotransmitter receptor binding assays. The free radical intermediates formed via

PHS-catalyzed bioactivation of such endogenous substrates can initiate the formation of ROS that oxidize cellular macromolecules like DNA, protein, and lipids, which have been implicated in cellular degeneration in the brain (Jeng et al., 2006). Our results accordingly may provide new insights into neurodegenerative effects associated with aging and mechanisms underlying neurodegenerative diseases like Alzheimer‟s and Parkinson‟s diseases. In particular, we examined the role of mammalian PHS-1 and a related model enzyme, HRP, in catalyzing the bioactivation of DA and epinephrine, as well as their precursors and metabolites to potentially neurodegenerative free radical intermediates. The in vitro studies herein using ESR, UV/vis spectroscopy, and high-performance liquid chromatography with electrochemical detection

(HPLC-EC) provide direct evidence that neurotransmitters, precursors, and metabolites can be bioactivated by HRP and PHS-1 to form not only semiquinone radicals but also carbon-centered free radicals and ROS, such as hydroxyl and/or superoxide free radicals. This PHS-dependent bioactivation resulted in ROS-mediated oxidative DNA damage measured by 8-oxo-dG formation, which has been implicated in neurodegenerative processes (Jeng et al., 2006).

Previous studies with epinephrine have demonstrated some of these reactive outcomes during autoxidation or HRP-catalyzed bioactivation (Adak et al., 1998; Kalyanaraman et al., 1984b; 167

Miller et al., 1990; Misra and Fridovich, 1972; Pileblad et al., 1988). Our work confirms those studies and presents novel ESR, UV/vis spectroscopy, and HPLC-EC data characterizing reactive intermediates from PHS-1-catalyzed bioactivation of the l-DOPA precursor and

DOPAC metabolite of the neurotransmitter DA, as well as DA itself, resulting in oxidative DNA damage. Spin stabilization ESR experiments with Zn2+ proved useful for trapping semiquinone radical species formed by one-electron oxidation of the substrates. With the exception of HVA and normetanephrine, all substrates tested generated ortho-semiquinone free radical intermediates upon autoxidation. This was consistent with the known oxidation of catecholamines involving molecular oxygen under physiological conditions (Adak et al., 1998;

Kalyanaraman et al., 1984b; Miller et al., 1990; Misra and Fridovich, 1972; Pileblad et al.,

1988).

HRP-catalyzed free radical formation from DA, its precursor l-DOPA, and its metabolite

DOPAC was 3−8 times greater than that from epinephrine and DHMA. DA was a slightly better substrate for HRP than l-DOPA, but more importantly, DOPAC was a better substrate than DA for HRP-catalyzed oxidation to free radical intermediates. Kinetics studies using the same ESR approach and enzymatic system also found that DA was a better substrate for HRP than l-DOPA

(Kalyanaraman et al., 1984b), at both pH 8 and pH 5. In spin stabilization ESR experiments, the unresolved hyperfine structure observed in the spectrum of DA and epinephrine may be a consequence of the overlapping of ESR signals due to the formation of secondary semiquinone radicals. In the case of DA and epinephrine, this observation may reflect higher rates of cyclization of their ortho-semiquinones, which are the intermediate molecular products in the pathway to aminochromes, and/or perhaps a higher stability of their aminochromes, from which the secondary radicals are derived. UV/vis spectral characterization of the metabolites generated 168

by HRP-catalyzed oxidation of DA and l-DOPA, using similar experimental conditions, revealed the formation of a chromophore with λmax at 300 and 480 nm (Table 4), characteristic of their aminochromes (Graham et al., 1978; Segura-Aguilar et al., 1998). Previous studies

(Adak et al., 1998; Kalyanaraman et al., 1984b; Kalyanaraman et al., 1984a) established that oxidation of epinephrine and norepinephrine by HRP-H2O2 generates ortho-semiquinones as a primary one-electron oxidation product, which by disproportionation may give rise to ortho- quinones. With deprotonation of the amino group in the side chain, ortho-quinones undergo 1,4- intramolecular cyclization to form leukoadrenochrome.

Interestingly, HRP-catalyzed bioactivation of DOPAC and DHMA produced asymmetric

ESR spectra, characteristic of an immobilized radical. Similar ESR spectra have been obtained from enzymatic oxidation of the DOPA derivative 5-S-cysteinyldopa to synthetic pheomelanins

(Sealy et al., 1982). Synthetic melanins are characterized by an ESR spectrum with a single broad line with g = 2.004 and a line width of 4−10 G. The putative DOPAC- and DHMA- derived polymers have g values of 2.004 and 2.003, respectively, and a dark brown precipitate was observed frequently during the course of the ESR−spin stabilization experiments.

Particularly in the case of DOPAC and DHMA, filtration of the incubations mixtures through a

PVDF membrane was necessary prior to spectral acquisition. Although the mechanism of melanin formation is not thought to involve the direct reaction of free radicals per se, the formation of melanin is a well-documented characteristic in the process of catecholamine oxidation (Felix and Sealy, 1981; Finkelstein et al., 1980; Graham, 1978; Graham and Jeffs,

1977; Kalyanaraman et al., 1982). As described in the literature (Graham, 1978), neuromelanin deposited in the substantia nigra and locus ceruleus is generated from polymerization of oxidized products of DA and norepinephrine. For DOPAC and DHMA, the 1,4-intramolecular 169

cyclization of their ortho-quinones is hindered, so their bioactivation pathway is devoid of leukoadrenochrome and aminochrome formation. The direct attack of nucleophiles and/or other radical species to the aromatic ring of primary ortho-semiquinone free radicals may lead to the formation of oligomeric or polymeric structures that may account for the observed ESR signal.

This hypothesis would be consistent with the further detection of UV/vis absorption bands characteristic of ortho-semiquinone free radicals (Table 4) under similar in vitro conditions.

More importantly, the intense absorption peak observed at λmax = 340 nm for DHMA was assigned to melanin or a melanin-like polymer, since human melanin has a characteristic λmax

= 335 nm (Das et al., 1978). These results may provide the first direct evidence for the formation of free radical intermediates and melanin or melanin-like polymers in the peroxidase- dependent bioactivation of neurotransmitter metabolites, such as DOPAC and DHMA, which might contribute to the neurodegenerative process.

As mentioned previously, catecholamine bioactivation by peroxidases is also associated with the generation of ROS. ESR−spin trapping experiments investigated the formation of free radical intermediates such as HO•, O2•−, and other putative radical species, such as C- and N- centered free radicals (Jeng et al., 2006). Although the generation of superoxide radicals during autoxidation of the semiquinones is generally accepted (Graumann et al., 2002; Klegeris et al.,

1995), the superoxide radical spin adduct formed with DMPO is relatively unstable in aqueous solution. The half-life of its ESR signal is approximately 80 s at pH 6 and 35 s at pH 8 (Sun et al., 1981). The HO• radical, and usually C-centered radicals, forms stable spin adducts with

DMPO in aqueous solutions, with the resulting signal often lasting hours or even days, depending on the temperature. In our studies, the production of hydroxyl free radicals via PHS-

1- and HRP-catalyzed oxidation of DA, l-DOPA, and epinephrine was confirmed by spin 170

trapping−ESR spectroscopy (Tables 2 and 3). The quartet, RSA3, having an intensity ratio of

1:2:2:1 and aN = aβH = 14.88 G is a well-known characteristic of DMPO/OH• spin adducts

(Alanko et al., 1999). The addition of SOD did not affect the RSA3 intensity, suggesting that

HO• was not generated by a Haber-Weiss reaction. To the contrary, the inhibition of RSA3 after addition of catalase suggests that HO• could be generated by Fenton chemistry. However, we used lower concentrations of SOD, below its catalytic range (33 μM). Because catalase is a more effective inhibitor than SOD (Adak et al., 1998), lower concentrations of SOD may have not produced a detectable change in the ESR signal. Moreover, Fenton reactions usually require the presence of catalytic concentrations of metal ions, such as Fe2+. The use of DTPA should render the Fenton reaction a minor contributor to the production of HO•. The PHS inhibitor

ETYA abolished the signal of the DMPO-OH spin adduct (Figure 3), negating the possibility of exclusive production of HO• by Fenton chemistry, and alternatively suggesting that the generation of HO• is peroxidase-mediated. One cannot discard completely the possibility that

HO• may also be formed by homolytic cleavage of H2O2, as HO• has been observed in buffered

(pH 7.2) reaction mixtures containing DMPO and H2O2 after UV irradiation (Buettner and

Oberley, 1978; Ozawa and Hanaki, 1978).

PHS-1-catalyzed bioactivation of DA, its precursor l-DOPA, epinephrine, and normetanephrine generated another ESR signal, possibly due to a C-centered or a phenoxyl

DMPO-spin adduct. Inhibition of the ESR signal by the PHS inhibitor ETYA indicated that the bioactivation of l-DOPA to a free radical intermediate was catalyzed by PHS-1. ETYA is a dual inhibitor of PHSs and lipoxygenases (Downing et al., 1972; Towell and Kalyanaraman, 1991) and has been shown in vitro at the same concentration to inhibit the PHS-catalyzed bioactivation of xenobiotics and neurotransmitters (Kubow and Wells, 1989; Parman et al., 1998; 171

Ramkissoon et al., 2005). The absence of the signal in the control spectrum (Figure 3) when

PHS-1 was omitted from the medium also indicated that bioactivation was due to the active enzyme rather than other components of the system. Interestingly, the hyperfine structure and the hfcc values were very similar to those of a DMPO-spin adduct derived from DA previously reported (Das et al., 1978). Computer simulation of the ESR signal obtained in assays with PBN confirmed the existence of C-centered free radicals for l-DOPA (data not shown). A C-centered free radical with similar hfcc values (aN = 15.21 G and aβH = 23.48 G) was also detected by others (Paris et al., 2005) in an ESR−spin trapping study using RCHT cells with DMPO, but this radical spin adduct was not identified. Our results suggest that this radical species is probably due either to the formation of C-centered free radicals on the side chain of the molecule and/or to a radical species delocalized over the aromatic ring. This hypothesis is consistent with published studies reporting the generation of putative N- and C-centered free radicals in the bioactivation of 3,4-methylenedioxyamphetamine by PHS-1 (Jeng et al., 2006), where the oxidation occurred initially at the primary amino group in the side chain of the molecule.

Furthermore, hfcc values found for this DMPO-radical species are in the range of values described in the literature for DMPO C-centered free radicals (Janzen et al., 1989) and for

DMPO-CH(OH)CH3 types of adducts (Guo et al., 1999; Towell and Kalyanaraman, 1991). Our attempts to determine superhyperfine coupling constants by decreasing the modulation amplitude resulted in a decrease of the signal intensity, which prevented the determination of a more accurate structure for these radical species. Our results suggest that different competing pathways could account for the generation of superoxide and hydroxyl free radicals and C- centered radicals during the peroxidase-mediated bioactivation of catecholamines. One pathway involves the generation of semiquinone free radical intermediates that rapidly react to produce 172

superoxide radicals and other damaging ROS. Redox cycling between catechols/catecholamines and their quinones/semiquinones can continuously produce large amounts of ROS, the chemical basis of which is provided elsewhere (Ishii and Fridovich, 1990; Klegeris et al., 1995; Tse et al.,

1976). Although this might be a minor competing pathway in the overall process, peroxidase- catalyzed bioactivation of endogenous substrates, the oxidation of the amino group, and subsequent formation of a C-centered free radical may also play an important role in the formation of damaging ROS. The generation of a C-centered free radical and its reaction with

O2 and the subsequent oxidation by a peroxidase or by a Fenton-like mechanism may generate

ROS capable of oxidizing macromolecules.

DNA oxidation is a sensitive measure of oxidative stress that can occur in neurodegenerative conditions, and this molecular lesion may cause genomic instability and/or aberrant cell signaling leading to cellular dysfunction and cell death (Jeng et al., 2006). In our

DNA oxidation studies, in the absence of PHS-1, DA showed higher overall DNA oxidation as compared to l-DOPA and DOPAC due to autoxidation. However, in the presence of enzyme,

DNA oxidation was further enhanced over that due to autoxidation, and PHS-1 activity was equally effective in increasing DNA oxidation for DA and l-DOPA (i.e., both had increases of

70%). The increase in PHS-1-dependent DNA oxidation for the catechol DOPAC was lower (an increase of 40%) as compared to DA and l-DOPA, which may be due to its lack of ability to cyclize to form an aminochrome. MTA and HVA did not lead to PHS-1-dependent DNA oxidation under these reaction conditions, possibly because these chemicals are not catechols.

While spin trapping−ESR spectroscopy revealed that MTA produced a free radical, it may not be sufficiently reactive or abundant at these concentrations to initiate ROS formation and DNA oxidation. Higher concentrations of the substrates may be necessary to detect oxidized DNA or 173

to trap free radicals directly by ESR. The DA concentration in nerve terminals is about 50 mM, but this is stored in vesicles (Anden et al., 1966; Zhu, 2004). However, total catechol levels of precursors and metabolites present in the cytosol can be in the millimolar range (Mosharov et al., 2003); hence, the neurodegenerative potential of the precursors and metabolites is distinct from the actual neurotransmitters.

In summary, we have shown that DA, its precursor l-DOPA, and its metabolite DOPAC, as well as epinephrine and its metabolite DHMA, were excellent HRP and PHS-1 substrates, resulting in HRP- and PHS-1-catalyzed bioactivation to various free radical intermediates that initiate ROS formation and DNA oxidation. This oxidative damage to DNA and other cellular macromolecules, in addition to ROS-mediated signal transduction, may contribute to the neurodegenerative process (Jeng et al., 2006; Wong et al., 2008). In addition, we provide spectroscopic evidence for the potential bioactivation of neurotransmitter precursors and metabolites by PHS-1 to ortho-semiquinones and aminochromes. The differences in rates of cyclization of the ortho-semiquinones and/or the absence of an amino group in the side chain of the substrate molecules observed in this and other studies (Hastings, 1995; Napolitano et al.,

1995) imply that precursors and metabolites may have the opportunity to react with external nucleophiles and/or radical species rather than undergo internal cyclization, which might contribute to the neurodegenerative process as well. Perhaps most importantly, our results show that the precursors and metabolites of endogenous neurotransmitters, which are generally considered inactive based upon receptor binding assays, may actually play an important role in free radical-initiated neurodegeneration associated with aging. 174

2.2 STUDY 2: HUMAN PROSTAGLANDIN H SYNTHASE (hPHS)-1- AND hPHS-2- DEPENDENT BIOACTIVATION, OXIDATIVE MACROMOLECULAR DAMAGE AND CYTOTOXICITY OF DOPAMINE, ITS PRECURSOR AND METABOLITESa

Annmarie Ramkissoon and Peter G. Wells

a. Preliminary reports of this research were presented at the annual meetings of the Society for Neuroscience (SfN Proceedings, Abstract No. 382.13, San Diego, CA, Nov. 2007) and the Society of Toxicology (Toxicol. Sci. (Supplement: The Toxicologist) 102(1): 373 (No. 1814), 2008). This research was supported by a grant from the Canadian Institutes of Health Research (CIHR). AR was supported by a doctoral scholarship from CIHR and the Rx&D Health Research Foundation.

This manuscript is reproduced with permission from: Ramkissoon, A. and Wells, P. G. (2011). Human prostaglandin H synthase (hPHS)-1- and hPHS-2-dependent bioactivation, oxidative macromolecular damage, and cytotoxicity of dopamine, its precursor, and its metabolites. Free Radic Biol Med 50(2): 295-304. Copyright 2011. Elsevier Limited. http://www.sciencedirect.com/science/journal/08915849 DOI: doi:10.1016/j.freeradbiomed.2010.11.010

175

2.2.1. ABSTRACT

The dopamine (DA) precursor L-dihydroxyphenylalanine (L-DOPA) and metabolites dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA) and 3-methoxytyramine (3-

MT) may serve as substrates for PHS-catalyzed bioactivation to free radical intermediates. We used CHO-K1 cells expressing hPHS-1 or hPHS-2 to investigate hPHS isozyme-dependent oxidative damage and cytotoxicity. hPHS-1 and hPHS-2 cells incubated with DA, L-DOPA,

DOPAC or HVA exhibited increased cytotoxicity compared to untransfected cells and cytotoxicity increased further by exogenous arachidonic acid (AA), which increased hPHS activity. Preincubation with catalase, which detoxifies reactive oxygen species (ROS), or acetylsalicylic acid, an inhibitor of hPHS-1 and -2, reduced the cytotoxicity caused by DA, L-

DOPA, DOPAC and HVA in hPHS-1 and -2 cells both with and without AA. Protein oxidation was increased in hPHS-1 and -2 cells exposed to DA or L-DOPA, and further increased by AA addition. DNA oxidation was enhanced earlier and at lower substrate concentrations than protein oxidation in both hPHS-1 and -2 cells by DA, L-DOPA, DOPAC and HVA, and further enhanced by AA addition. hPHS-2 cells appeared more susceptible than hPHS-1 cells, while untransfected CHO-K1 cells were less susceptible. Thus, isozyme-specific, hPHS-dependent oxidative damage and cytotoxicity caused by neurotransmitters, their precursors and metabolites may contribute to neurodegeneration associated with aging. 176

2.2.2 INTRODUCTION

Reactive oxygen species (ROS), such as superoxide anions, hydrogen peroxide (H2O2) and hydroxyl radicals, can oxidize cellular macromolecules (Halliwell and Gutteridge, 2007).

Such damage, if not repaired, can accumulate over time and can lead to loss of cellular function and cell death (Wells et al., 2009). The brain is susceptible to oxidative stress due to high oxygen consumption, low antioxidants and ROS-generating enzymatic reactions (Halliwell and

Gutteridge, 2007). Prostaglandin H synthase (PHS), also known as cyclooxygenase (COX), catalyzes the initial steps in the production of prostaglandins (Casarett et al., 2008). There are two isoforms, PHS-1 and PHS-2. PHS-1 is constitutively expressed in most tissues including the brain (Yasojima et al., 1999a). PHS-2 is non-constitutive and inducible in most tissues, but is constitutive in the macula densa of the kidney and the vas deferens, and is highly expressed in brain regions such as the cortex, hippocampus and striatum (Kaufmann et al., 1997; Teismann et al., 2003; Yasojima et al., 1999a). PHS is a bifunctional enzyme that has cyclooxygenase and peroxidase activities (Marnett, 1990; Ohki et al., 1979). The cyclooxygenase component converts arachidonic acid (AA) to prostaglandin G2. The peroxidase component reduces this to prostaglandin H2, in the process oxidizing a co-substrate (Markey et al., 1987). Endogenous compounds or xenobiotics can serve as co-substrates in this reaction, forming free radicals that can generate ROS (Hughes et al., 1988; Markey et al., 1987; Parman et al., 1998).

Aging studies with PHS-1 knockout mice showed decreased DNA oxidation accumulation in the brain while aged mice overexpressing human PHS-2 in neurons had increased levels of apoptosis compared to age-matched wild-type mice (Andreasson et al., 2001;

Wells et al., 2004). These results suggest that there are endogenous compounds in the brain that may be bioactivated by PHS. Phenols, catechols and amines are good substrates for PHS, and 177

many neurotransmitters, their precursors and metabolites contain these functional groups.

Studies using neuronal cell lines overexpressing rodent PHS-2 have shown enhanced oxidative stress and neurotoxicity resulting from dopamine (DA) and 6-hydroxy-dopamine oxidation

(Chae et al., 2008; Tyurina et al., 2006). Also, a rat embryonic mesencephalon culture system using selective PHS inhibitors showed that activity of PHS-2, but not PHS-1, was required for

6-hydroxy-dopamine to kill dopaminergic neurons (Carrasco et al., 2005).

There remain important questions concerning the role of PHS isozymes in the bioactivation of endogenous brain compounds. Most studies have investigated bioactivation using purified PHS, tissue microsomes or rodent and ovine PHS. These are difficult to interpret due to differing isozyme catalytic stability (Xiao et al., 1998). We used Chinese hamster ovary-

K1 (CHO-K1) cell lines stably expressing human PHS-1 (hPHS-1) or hPHS-2, in which we can stimulate PHS activity by adding exogenous AA. This provides us with the opportunity to study human PHS as opposed to ovine enzymes in vitro, or rodent enzymes in vivo, either of which may provide different information on substrate specificity and/or potency. The use of non- neuronal cell lines limits the contribution of other brain metabolizing enzymes, transporters and receptor- mediated second messenger effects that would interfere with the evaluation and interpretation of hPHS-catalyzed bioactivation. Specifically, these cell lines do not express the

DA transporter (Pifl et al., 1993), which would interfere with our interpretation of the relative bioactivation of DA versus its metabolite and precursor bioactivation, as these compounds are not substrates for the transporter. Both hPHS isozymes may play a significant role in the bioactivation of brain substrates to ROS-generating free radicals that contribute to the mechanism of neurodegeneration associated with aging. Substrates for hPHS-1/2 in the brain include not only neurotransmitters, such as DA, but also their precursors and metabolites, the 178

latter of which are generally considered inactive based upon receptor binding assays, but may independently initiate oxidative macromolecular damage and cytotoxicity (Kalant et al., 2006). 179

2.2.3 MATERIALS AND METHODS

Chemicals and Reagents

Arachidonic acid (AA), DA, L-DOPA, DOPAC, homovanillic acid (HVA), 3- methoxytyramine (3-MT) proteinase K, nuclease P1, acetylsalicylic acid (ASA), dimethyl sulfoxide were obtained from Sigma-Aldrich (Oakville, ON, Canada). Alkaline phosphatase and Complete, Mini, EDTA-free protease inhibitor cocktail tablets was obtained from Roche

Diagnostics (Laval, QC, Canada). RNAse A/T1 mix was obtained from Fermentas Canada Inc.

(Burlington, ON, Canada). Proteinase K was obtained from BioShop Canada Inc. (Burlington,

ON, Canada). 8-hydroxy-2‟-deoxyguanosine, DUP-697 and SC-560 were obtained from

Cayman Chemical Co. (Ann Arbor, MI). All other reagents were of analytical or HPLC grade.

hPHS-1/2 CHO-K1 Cell Lines

Chinese hamster ovary-K1 (CHO-K1) cell lines overexpressing hPHS-1/2 were obtained from Dr. Fred F. Kadlubar at the National Center for Toxicological Research, Arkansas, who originally obtained them from Dr. Stacia Kargman of the Merck Frosst Center for Therapeautic

Research in Quebec, Canada. During the construction of these cell lines, hPHS-1 and hPHS-2 protein levels were determined from protein samples from each cell line using immunoblot analysis (Kargman et al., 1996). This confirmed that each protein was stably expressed in their respective cell line. Cell lines were grown as a monolayer culture in a humidified environment with 5% CO2 at 37°C as previously described except the hPHS-1 cell line was grown in Ham's

F12 media supplemented with 10% fetal bovine serum (Sigma-Aldrich, Oakville, Ont., Canada),

L-glutamine, non-essential amino acids, 100 U/mL penicillin and 100 ug/mL streptomycin and selected using G418 (Sigma-Aldrich, Oakville, ON, Canada) (Kargman et al., 1996). 180

Untransfected CHO-K1 cells were obtained from Dr. Peter H. Backx, University of Toronto, and were grown in the same media as the transfected cells but without selection agents.

Determination of PHS activity in CHO-K1 cell lines stably expressing human PHS-1 or PHS-2

(hPHS-1/2) and CHO-K1

To determine PHS activity in these cell lines, the cells were activated with exogenous arachidonic acid (10 uM) for 15 minutes and prostaglandin E2 (PGE2), the primary product of

PHS-catalyzed arachidonic acid metabolism, was measured using a commercially available high sensitivity chemiluminescence enzyme immunoassay kit from Cayman Chemical, Ann Arbor,

MI, according to the manufacturer‟s instructions. Briefly, cells were harvested by trypsinization of adherent cultures and resuspended with media. Approximately 200,000 to 300,000 cells were incubated with AA (10 uM) at 37oC for 15 min. The reaction was terminated with 10 uL of 1 M

HCl followed by 20 uL of 0.5 M NaOH, and the samples were spun with the supernatant collected for PGE2 quantification according to kit instructions (Kargman et al., 1996). To inhibit PHS activity, cells were pretreated with the PHS-1 specific inhibitor SC-560 (5 uM) or the PHS-2 specific inhibitor Dup-697 (5 uM) prior to AA addition (Kargman et al., 1996;

Spencer et al., 1998).

Measurement of Cytotoxicity by Lactate Dehydrogenase (LDH) release assay

Lactate dehydrogenase (LDH) release is a reliable, reproducible and sensitive technique to evaluate cytotoxicity. At the early stages of cytotoxicity, LDH is released from cells with a damaged membrane. The CytoTox-ONE TM Homogeneous Membrane Integrity Assay from

Promega was used to assess LDH release. Cells were plated in 96-well plates at a cell 181

concentration of 20,000 to 30,000 cells in 200 uL and allowed to adhere overnight. PHS activity in the cells was activated by preincubation with AA (100 uM) for 2 min, following which DA, its precursor or metabolites were dissolved in dimethyl sulfoxide (DMSO) (total <

0.5% [v/v] in media) as the vehicle, added at various concentrations and the cells were incubated at 37oC for 6 or 24 hr. The concentrations tested were at physiological levels or higher to account for a shorter exposure time. Our early (6 hr) and later (24 hr) time-points ensured that we did not miss critical responses that may require time to accumulate. Plates were removed from the incubator for 25 min and 50 uL of supernatant was collected to quantify LDH release according to kit instructions. Fluorescence with an excitation wavelength of 560 nm and emission wavelength of 590 nm was recorded. Percent LDH release was calculated from the experimental LDH release compared to the maximal LDH release from 20,000 to 30,000 cells.

For some experiments, cells were preincubated for 4 hr with 250 U/ml of polyethylene glycol (PEG)-conjugated catalase, which detoxifies hydrogen peroxide, or heat-inactivated

PEG-catalase. To make consistent comparisons between the different substrates, cell lines and endpoints, we elected to use an irreversible inhibitor of both hPHS-1 and hPHS-2, ASA (5 mM for 1 hr preincubation), which did not interfere with any of the methods used. After preincubation, the media was removed, cells were washed and new warm media was added prior to treatment with AA and test compounds.

Treatment of cells to assess DNA and Protein Oxidation

Cells were grown to 90% confluence on 10 cm tissue culture plates. PHS activity in the cells was activated by AA (100 uM) for 2 min following which DA, its precursor or metabolites were added at various concentrations and incubated at 37oC for 1, 6 or 24 hr. The 182

concentrations were chosen in the range tested in our cytotoxicity studies to be consistent. For

DNA oxidation studies, cells were scraped into media and 3 plates were pooled in 50 mL Falcon tubes. Cellular pellets were obtained by spinning at 1,100 x g for 8 min, the pellets were washed with Hanks solution and stored at -80oC prior to DNA extraction. For protein oxidation studies, similar procedures were followed but cell pellets were suspended in potassium phosphate buffer (100 mM KH2PO4-K2HPO4, pH 7.4, 1 mM EDTA) with freshly added

Complete, Mini, EDTA-free protease inhibitor cocktail tablets obtained from Roche Diagnostics

(Laval, QC, Canada). Samples were sonicated and spun at 10,000 x g for 15 min at 4oC, after which the supernatant was removed and treated with streptomycin sulfate solution to a final concentration of 1 % for 15 min at room temperature to precipitate any extracted DNA.

Samples were then spun at 6,000 x g for 10 min at 4oC and supernatants were stored at -80oC until analysis.

DNA Extraction from cells

The chaotropic NaI method was used (Ravanat et al., 2002) with the following modifications. To the cellular pellet was added 1 ml of lysis buffer A (320 mM sucrose, 5 mM

MgCl2, 10 mM Tris, 0.1 mM desferoxamine pH 7.5, 1% Triton X-100). After hand homogenization, the nuclei were collected following centrifugation at 1,000 x g for 10 min at

4°C and washed with 1 ml of buffer A. To the nuclear pellet, 200 μl of buffer B were added (10 mM Tris, 5 mM EDTA-Na2, 0.15 mM desferoxamine, pH 8.0, 1% w/v SDS) and hand- homogenized to facilitate lysis of the nuclear membrane. Then, 5 μl of RNase A/T1 mix were added and the samples were incubated for 1.5 hr at 50°C, after which 13.3 μl of proteinase K

(10 mg/ml in 10 mM Tris/1 mM EDTA, pH 8.5) were added prior to incubation at 50°C for 1.5 183

hr. Subsequently, 300 uL of the NaI solution (7.6 M NaI, 40 mM Tris, 20 mM EDTA-Na2, 0.3 mM desferrioxamine, pH 8.0) were added and the tubes were inverted for 2 min. Then, 500 ul of isopropanol were added and DNA was precipitated by gently inverting the tube for 2 min. DNA was recovered by centrifugation at 10,000 x g for 10 min at 4°C and washed with 1 ml of 70% ethanol. After centrifugation (10,000 x g for 5 min), DNA was washed again using 1 ml of 70% ethanol four times. DNA was recovered by centrifugation, dissolved in sodium acetate buffer

(20 mM sodium acetate, 0.1 mM desferoxamine, pH 5.2) for 1 hr at 37oC and then briefly sonicated prior to DNA digestion. DNA was digested with nuclease P1 (5 units) at 37oC. After

1 hr, Tris-HCl (1 M, pH 8.5) was added followed by alkaline phosphatase (8 units) and this was incubated at 37oC for 1 hr. Samples were then filtered using Microcon®-YM 10 filters

(Millipore Canada Ltd.) and stored at -80oC until analysis.

Detection of 8-hydroxy-2’-deoxyguanosine (8-oxo-dG) and 2-deoxyguanosine (2-dG)

Oxidation of 2-dG to 8-oxo-dG was quantified using an isocratic Series 200 high- performance liquid chromatographic (HPLC) system (PerkinElmer Instruments LLC, Shelton,

Connecticut). The system was equipped with a SUPELCOSIL LC-18 column (250 mm x 4.6 mm, Sigma-Aldrich, Oakville, ON, Canada), an electrochemical (EC) detector (Coulochem II,

ESA, Chelmsford, MA), a guard cell (ESA, model 5020), an analytical cell (ESA, model 5010) and a recording integrator (PerkinElmer NCI 900 Interface). The filtered samples were injected into the HPLC system and were eluted using a mobile phase which contained 50 mM sodium phosphate buffer (pH 5.5)-methanol (95:5, v/v) at a flow rate of 1.0 ml/min. The nucleoside dG was detected by UV absorption at 280 nm and 8-oxo-dG was monitored by EC detection with channel 1 set at 100 mV and channel 2 set at 400 mV. The guard cell was set at 450 mV. 184

Chromatographs were analyzed using the TotalChrom chromatography software version 6.2.0

(Perkin Elmer Instruments LLC). DNA oxidation was expressed as pmol 8-oxodG per ug dG.

Detection of Protein Carbonyls

Protein oxidation was assessed by protein carbonyl content. Protein carbonyls react with

2,4-dinitrophenylhydrazine (DNPH) to form a hydrazone that can be detected spectrophotometrically. Collected supernatants were analyzed according to instructions in the

Protein Carbonyl Assay kit (Cayman Chemical Co., Ann Arbor, MI), except cells were washed in 20% trichloroacetic acid and the final protein pellet was resuspended in 300 uL of guanidine hydrochloride. Protein carbonyls were measured at absorbance 375 nm and were standardized to total protein recovered as measured at absorbance 280 nm.

Statistical analysis

Multiple comparisons among groups were analyzed by one-way ANOVA with a subsequent

Tukey‟s test (GraphPad InStat®3.05, GraphPad Software, Inc., San Rafael, CA, USA). The level of significance was determined to be at P < 0.05.

185

2.2.4 RESULTS

PHS activity in CHO-K1 cell lines stably expressing human PHS-1 or PHS-2 (hPHS-1/2) and

CHO-K1 cells

PHS-1/2 activities were in the range of published values (Kargman et al., 1996) and activities were reduced using standard inhibitors or by withholding the PHS substrate AA

(Figure 1). Without exogenous AA, PHS activity in untransfected CHO cells averaged 0.3 +/-

0.09 ng of PGE2 per million cells, while PHS activity in hPHS-1-expressing cells had 4.5 +/- 0.4 ng of PGE2 per million cells and hPHS-2-expressing cells averaged 2.5 +/- 0.3 ng of PGE2 per million cells, corresponding to respective increases in hPHS-1 and hPHS-2 activity of about 15- fold and 8-fold with the addition of AA. Activation of hPHS-1-expressing cells with exogenous

AA averaged 58 +/- 14 ng of PGE2 per million cells, or about 13-fold higher than transfected cells without AA, and activities were reduced using the standard PHS-1 inhibitor SC 560

(p<0.01). hPHS-2 cells averaged 24 +/- 3 ng of PGE2 per million cells, or about 10-fold higher than transfected cells without AA, and activities were reduced using the standard PHS-2 inhibitor DUP 697 (p<0.01). Activity in untransformed CHO-K1 cells activated with exogenous AA was only 3% of that in activated hPHS-2-expressing cells. The addition or withholding of AA provided a method of further modulating PHS activity in these cells lines to study co-substrate bioactivation. The untransfected and hPHS-expressing cell lines were used as models to investigate PHS-catalyzed bioactivation of endogenous substrates leading to ROS generation and cytotoxicity.

Cytotoxicity caused by DA, its precursor and metabolites in hPHS1/2-expressing cells

LDH release was measured to evaluate the potential of these substrates to cause 186

cytotoxicity. In all cell lines, AA (100 uM) addition or the vehicle DMSO (0 uM) on its own did not lead to enhanced cytotoxicity (Figure 2). After a 24 hr incubation with DA, both hPHS-

1 and hPHS-2 cells showed enhanced cytotoxicity compared to CHO-K1 cells (Figure 2).

Within the hPHS cell lines, AA addition further increased cytotoxicity (p<0.05). The same was seen for L-DOPA, which was equally cytotoxic as DA over the concentrations tested with AA addition. With AA activation in the hPHS cell lines, cytotoxicity increased by 1-2-fold for both

DA and L-DOPA. hPHS-2 cells appeared to be more susceptible to cytotoxicity even though hPHS-1 cells had 2-fold higher activity than the hPHS-2 cells. With 500 and 1,000 uM DOPAC, hPHS-1 and -2 cells showed enhanced cytotoxicity when activated with AA addition compared to reactions without AA (p<0.05) (Figure 2). HVA caused enhanced cytotoxicity in the hPHS-

1 cell line only at 1,000 uM with AA addition (p<0.05), while in hPHS-2 cells it enhanced cytotoxicity at both 500 uM and 1,000 uM with AA addition (p<0.05). DOPAC and HVA were not as potent as DA or L-DOPA in hPHS-1/2 cells, and were not cytotoxic in untransfected

CHO-K1 cells. LDH release was seen in hPHS-1 and hPHS-2 cells as early as 6 hr after treatment (data not shown) with DA, L-DOPA, DOPAC and HVA compared to the vehicle control (p<0.001). Cytotoxicity was increased further by AA-activation (p<0.05), while cytotoxicity was lower in untransfected CHO-K1 cells (p<0.05) (Figure 2). 3-MT did not show any hPHS1/2-mediated cytotoxicity (Figure 2). A 10 uM concentration of the compounds was not cytotoxic after a 24 hr incubation (data not shown).

ASA and PEG-catalase protect against cytotoxicity caused by DA, its precursor and metabolites in hPHS1/2 expressing cells

By acetylating a serine residue at the catalytic site, ASA irreversibly inhibits both hPHS- 187

1 and hPHS-2 cyclooxygenase activity (Holtzman et al., 1992; Lecomte et al., 1994).

Pretreatment with ASA (5 mM) reduced by up to 50% the cytotoxicity caused by DA in hPHS-1 cells with AA addition (p<0.001) (Figure 3). In hPHS-2 cells, ASA reduced the cytotoxicity associated with and without AA addition (p<0.001). This pattern of cytoprotection by ASA was also seen for L-DOPA, DOPAC and HVA (p<0.001).

To determine whether hPHS-1 and -2-enhanced cytotoxicity of DA, its precursor and metabolites are mediated by ROS, the cells were preincubated with 250 U/ml of PEG-catalase, which detoxifies H2O2, then treated with 1,000 uM of substrate for 6 hr with or without AA, and

LDH release was measured (Figure 4). Under these conditions, catalase is still enzymatically active 10 hours after addition to the media. Catalase protected against the toxicity caused by

DA in hPHS-1 and -2 cells (p<0.001). This protection by catalase was also seen for L-DOPA,

DOPAC and HVA (p<0.001). Protection was due to the catalytic activity of catalase, since heat-inactivated catalase offered no protection.

Protein oxidation caused by hPHS1/2-catalayzed bioactivation of DA, its precursor and metabolites

These cell lines were also used to investigate hPHS isozyme-dependent, ROS-mediated protein oxidation caused by DA, its precursor and metabolites (1,000 uM). Protein oxidation, quantified by protein carbonyl content, was not increased in any cell line or with any treatment within a 6-hr incubation with substrates when compared to controls (data not shown). However, after 24 hr incubation, protein carbonyls were increased over 2-fold in both hPHS-1 and -2 cells exposed to DA compared to controls (p<0.05), and were further increased by AA addition 188

(p<0.05) (Figure 5). Controls exhibited a mean of 2.8 nmol of carbonyls per mg protein, while

DA-treated PHS-1/2 cells with AA addition averaged a 2.3-fold increase to 6.5 nmol of carbonyls per mg protein. After 24 hr, L-DOPA-treated hPHS-1/2 cells did not show enhanced protein carbonyl formation. However, upon AA addition, the level of carbonyls increased 1.6- fold, from 3.1 to 5.0 nmol of carbonyls per mg protein (p<0.05). Enhanced protein oxidation was not detected for DOPAC, HVA or 3-MT. AA (100 uM) addition or the vehicle DMSO (0 uM) on its own did not lead to enhanced protein oxidation.

DNA oxidation caused by hPHS1/2-catalayzed bioactivation of DA, its precursor and metabolites

PHS-dependent, ROS-mediated oxidative damage to DNA was quantified by the formation of 8-oxo-dG, which is formed when a hydroxyl radical reacts with 2dG in DNA.

After 1 hr and 6 hr, AA (100 uM) addition or the vehicle DMSO (0 uM) on its own did not enhance DNA oxidation. After a 1-hr incubation, DNA oxidation was enhanced in both hPHS-1 and -2 cells by DA, L-DOPA, DOPAC and HVA (1,000 uM) compared to the vehicle control

(p<0.05) (Figure 6). AA addition caused a further 10-30% increase in DNA oxidation in hPHS cells. hPHS-2 cells were more susceptible to DNA oxidation than hPHS-1 cells when treated with DA, L-DOPA and HVA (p<0.05). DA-treated hPHS-2 cells with AA addition caused the highest DNA oxidation, followed by L-DOPA and HVA. Interestingly, HVA caused up to 1.8- times more DNA oxidation in hPHS-1/2 cells compared to DOPAC. With a 1000 uM concentration of some compounds, we could detect DNA oxidation within only 1 hr after treatment, whereas a lower concentration (10 uM) required a longer incubation before DNA 189

oxidation could be reliably measured by our technique. At 10 uM of DA, L-DOPA, DOPAC and HVA, DNA oxidation was enhanced by these substrates within 6 hr (Figure 7). 190

Figure 1: PGE2 activity in Chinese Hamster Ovary (CHO) cells and cells stably expressing human prostaglandin synthase-1 (hPHS-1) or hPHS-2. The primary product of PHS-catalyzed arachidonic acid (AA) metabolism, was measured using a prostaglandin E2 (PGE2) enzyme immunoassay. Cells were activated with AA addition (10 uM) and a sample of the media was analyzed for PGE2 levels. SC-560 (5 uM) was used as a specific hPHS-1 inhibitor and DUP-697 (5 uM) was used as a specific hPHS-2 inhibitor. *= significantly different from AA in the same cell line (p<0.01). Each bar represents the mean with standard deviation, n = 4-6. 191

Figure 2: Dopamine (DA), its precursor and metabolites in hPHS-1- or hPHS-2-catalyzed bioactivation and cytotoxicity. CHO cell lines expressing hPHS-1 or 2 or untransfected CHO- K1 cells were activated with AA (100 uM) and then treated with DA, its precursor L- dihydroxyphenylalanine (L-DOPA) or its metabolites dihydroxyphenylacetic acid (DOPAC), homovanillic acid (HVA) or 3-methoxytyramine (3-MT) at concentrations of 250, 500 and 1,000 uM for 24 hr. 0 uM represents vehicle control. Cytotoxicity was measured by the release of lactate dehydrogenase (LDH). a=p<0.001 compared to 0 uM of the same cell line. *=p<0.05 compared to No AA of the same cell line. Each bar represents the mean with standard deviation, n = 6. 192

Figure 3: ASA protection in hPHS-catalyzed bioactivation and cytotoxicity of dopamine (DA), its precursor and metabolite. CHO cell lines expressing hPHS-1 or -2 were pre-treated with the PHS-1/2 inhibitor acetylsalicylic acid (ASA) (5 mM) for 1 hr. Cells were washed, activated with AA (100 uM) and treated with DA, its precursor L-DOPA or its metabolites DOPAC or HVA at a concentration of 1,000 uM for 24 hr. Cytotoxicity was measured by the release of LDH. a=p<0.001 compared to 0 uM of the same cell line. *=p<0.05 compared to No AA of the same cell line. b=p<0.001 compared to 1000 uM with or without AA. Each bar represents the mean with standard deviation, n = 6. 193

194

Figure 4: Polyethylene-conjugated (PEG)-catalase protects against hPHS-1/2-catalyzed substrate bioactivation and ROS-mediated cytotoxicity. CHO cell lines expressing hPHS- 1/2 were preincubated with the stabilized antioxidative enzyme PEG-catalase or heat-inactivated PEG-catalase (250 U/mL) for 4 hr, activated with AA (100 uM) and treated with DA, its precursor L-DOPA or its metabolites DOPAC or HVA at a concentration of 1,000 uM for 6 hr. Cytotoxicity was measured by the release of LDH. a=p<0.001 compared to 0 uM of the same cell line. *=p<0.05 compared to No AA of the same cell line. b=p<0.001 compared to 1000 uM with or without AA and without PEG-Catalase. Each bar represents the mean with standard deviation, n = 6. 195

196

Figure 5: Protein oxidation caused by hPHS-catalyzed bioactivation of dopamine (DA), its precursor and metabolites. CHO cell lines expressing hPHS-1 or -2 or untransfected CHO-K1 cells were activated with AA (100 uM) and treated with DA, its precursor L-DOPA or its metabolites DOPAC, HVA or 3-MT at a concentration of 1,000 uM for 24 hr. Protein oxidation was quantified by protein carbonyl content. The assay involves derivatization of the carbonyl group with DNPH. DNPH reacts with protein carbonyls and leads to protein-hydrozone which can be measured by absorbance at 375 nm. Protein levels were measured by absorbance at 280 nm. a=p<0.001 compared to 0 uM of the same cell line. *=p<0.05 compared to No AA of the same cell line. Each bar represents a mean with standard deviation, n = 5. 197

198

Figure 6: DNA oxidation caused by hPHS-catalyzed bioactivation of dopamine (DA), its precursor and metabolites. CHO cell lines expressing hPHS-1 or -2 or untransfected CHO-K1 cells were activated with AA (100 uM) and treated with DA, its precursor L-DOPA or its metabolites DOPAC, HVA or 3-MT at a concentration of 1,000 uM for 1 hr. Oxidized DNA was quantified by 8-oxo-2'-deoxyguanosine (8-oxo-dG) formation using high-performance liquid chromatography with electrochemical detection (HPLC-EC), and was standardized using 2‟-deoxyguanosine levels quantified by HPLC-UV. a=p<0.001 compared to 0 uM of the same cell line. *=p<0.05 compared to No AA of the same cell line. Each bar represents the mean with standard deviation, n = 5. 199

200

Figure 7: DNA oxidation caused by hPHS-catalyzed bioactivation of dopamine (DA), its precursor and metabolites. CHO cell lines expressing hPHS-1 or -2 or untransfected CHO-K1 cells were activated with AA (100 uM) and treated with DA, its precursor L-DOPA or its metabolites DOPAC or HVA at a concentration of 10 uM for 6 hr. Oxidized DNA was quantified by 8-oxo-2'-deoxyguanosine (8-oxo-dG) formation using high-performance liquid chromatography with electrochemical detection (HPLC-EC), and was standardized using 2‟- deoxyguanosine levels quantified by HPLC-UV. a=p<0.001 compared to 0 uM of the same cell line. *=p<0.05 compared to No AA of the same cell line. Each bar represents the mean with standard deviation, n = 4.

201

2.2.5 DISCUSSION

A number of studies have investigated the role of endogenous neurotransmitters, their precursors and metabolites in oxidative stress (Gomez-Santos et al., 2003; Graham et al., 1978;

Hastings, 1995; Jiang et al., 2004; Mattammal et al., 1995; Segura-Aguilar et al., 1998; Lai and

Yu, 1997; Goncalves et al., 2009). Recently, our laboratory has found that the amphetamines

3,4-methylenedioxymethamphetamine (Ecstasy), methamphetamine (Speed) and 3,4- methylenedioxyamphetamine (Ecstasy metabolite) can be bioactivated by mouse brain PHS-1 to free radical intermediates that generate ROS and oxidatively damage brain DNA leading to neurodegeneration (Jeng et al., 2006; Jeng and Wells, 2010). Also, the contribution of PHS to xenobiotic bioactivation, embryonic DNA oxidation and teratogenicity using phenytoin, thalidomide and benzo[a]pyrene has been characterized in our laboratory (Parman et al., 1998;

Parman and Wells, 2002; Parman et al., 1999). However, the specific role of the human isozymes hPHS-1 and hPHS-2 in bioactivation and ROS-mediated cytotoxicity has not been fully characterized, particularly with regard to the precursor and metabolites of DA.

PHS-1 and PHS-2 have 60% homology at the amino acid level and have similar Km values for AA (10-5 M range) (Kulmacz et al., 2003; Warner and Mitchell, 2004). Within the cyclooxygenase and peroxidase components there are major differences between the isoforms which can alter bioactivation and free radical generation (for review see (Kulmacz et al., 2003)).

The COX active site is larger in PHS-2, resulting in a more flexible substrate channel (Lecomte et al., 1994). PHS-2 can also be activated at lower peroxide levels, where PHS-1 remains dormant (Kulmacz, 1998). A number of residues on the sidechain near the peroxidase site differ between the two isozymes, which may affect the binding of cosubstrates, and the peroxidase activity of PHS-1 in vitro is more stable than that of PHS-2 (Kulmacz et al., 2003; Xiao et al., 202

1998). Substrate/xenobiotic bioactivation by PHS can differ both by PHS isozyme and enzyme source, particularly for ovine or human sources (Liu and Levy, 1998; Liu et al., 1995; Wiese et al., 2001). Furthermore, in inhibitor studies, ibuprofen was found to be more potent as an inhibitor of COX-2 in intact cells than in either broken cells or with purified enzymes (Kargman et al., 1996; Mitchell et al., 1993). These differences have permitted the creation of selective

PHS-1 and PHS-2 inhibitors, and raise the question of whether cosubstrates of these enzymes are similarly selective or more potent for the human enzymes.

We have shown herein that hPHS can catalyze the bioactivation of both precursors (L-

DOPA) and metabolites (DOPAC, HVA) of DA, as well as DA itself, leading to macromolecular oxidative damage and enhanced cytotoxicity that is isozyme-dependent.

Pretreatment with the PHS-1/2 inhibitor ASA protected against the cytotoxicity of DA, L-

DOPA, DOPAC and HVA in both hPHS cell lines, corroborating the dependence upon hPHS-

1/2-catalyzed bioactivation. hPHS-1 and -2 is expressed within the cell, so bioactivation is occurring intracellularly. Also, we show that the constituents of the medium do not measurably contribute to oxidative stress or cytotoxicity, as the untransfected cells do not exhibit high levels of toxicity.

DA concentrations in nerve terminals can reach 50 mM (Anden et al., 1966), but is primarily stored in vesicles away from cytosolic enzymes. Physiological DA in synaptic clefts may approximate 10 uM, but this concentration can substantially increase in the presence of neurotoxic compounds and decreased vesicular regulation (Garris et al., 1994; Mosharov et al.,

2003). These intracellular concentrations may also be elevated as active transporters may concentrate DA inside the neuronal terminal. Striatal cytosolic concentrations of the metabolites and precursor are between 4-8 uM and total catechol levels can reach the millimolar 203

range, accumulating for hours (Mosharov et al., 2003). In our experiments, we tested concentrations of these compounds at physiological levels of 10 uM, but also at higher concentrations to account for our relatively brief exposure of 24 hours. While a 10 uM concentration did not cause hPHS-1/2-catalyzed cytotoxicity (data not shown), it did enhance

PHS-catalyzed DNA oxidation in hPHS1/2 but not untransfected CHO-K1 cells (Figure 7).

This oxidatively damaged DNA in vivo could accumulate with pathogenic consequences not reflected by cell death in culture. While our lower concentrations illustrated susceptibility in the hPHS cell lines, these effects in vivo may be enhanced after longer durations of exposure and accumulate as seen during aging. Also, while CHO-K1 cells may not fully model the cytotoxic response of neural cells to enhanced PHS-catalyzed bioactivation, our DNA oxidation data is consistent with in vivo results in aged PHS knockout mice where regional accumulation of DNA oxidation was reduced when compared to age-matched wild-type mice (Wells et al., 2004).

hPHS-2-treated cells exhibited cytotoxicity equivalent to that in hPHS-1 cells despite hPHS-1 cells having higher PGE2 levels. This may indicate that some compounds are more efficacious cosubstrates for the peroxidase site of hPHS-2. Our results also showed that the enhanced cytotoxicity is mediated by ROS, as pretreatment with the stabilized antioxidative enzyme PEG-catalase significantly reduced cytotoxicity. hPHS-catalyzed bioactivation of these substrates may cause high levels of H2O2 that produce hydroxyl radicals through the Fenton or

Haber-Weiss reactions (Halliwell and Gutteridge, 2007). Others have shown that both AA and

H2O2 can serve as substrates for PHS-catalyzed oxidation of DA (Hastings, 1995; Mattammal et al., 1995). H2O2 can serve as a substrate for the peroxidase component of PHS, which can function independently of the COX site (Kulmacz et al., 2003). Therefore, the hPHS-2 isozyme may be better able to use endogenous peroxides as a cosubstrate in the bioactivation of 204

endogenous compounds to free radical intermediates, thereby enhancing its contribution to the observed oxidative damage and cytotoxicity. Increased COX-2 activity in cortical neurons has been shown to greatly increase the neurotoxic effects of low doses of Fe2+ and H2O2 (Im et al.,

2006).

To assess oxidative macromolecular damage as a potential molecular mechanism of neurodegeneration as well as a biomarker of oxidative stress, we measured protein carbonyl formation and DNA oxidation. The observed oxidative damage to macromolecules such as

DNA and protein confirms the intracellular nature of the effects. Protein oxidation also can lead to protein cross-linking and/or peptide bond cleavage. Oxidation of proteins may decrease protein activity and contribute to decreased energy metabolism in Alzheimer‟s disease (Swaab et al., 1998). Our results showed that hPHS-1/2-enhanced protein oxidation occurred only with

DA and L-DOPA, and only after 24 hr, the time of maximal cytotoxicity. These results are consistent with studies using COX-2-overexpressing SH-SY5Y cells that similarly showed enhanced protein oxidation due to DA (Chae et al., 2008). Interestingly, we were not able to detect increased protein carbonyls with DOPAC treatment in hPHS-1/2 cells despite significant cytotoxicity.

DNA oxidation is a sensitive measure of oxidative stress that can occur in neurodegenerative conditions even in the absence of lipid peroxidation (Jeng et al., 2002), and may cause genomic instability and aberrant cell signaling leading to cellular dysfunction or cell death (Allen and Tresini, 2000; Hailer-Morrison et al., 2003; Hyun et al., 2003; Wong et al.,

2008). Oxidation of dopamine by transition metals present in the brain may lead to the formation of both DNA adducts and oxidative DNA damage in dopaminergic cells (Levay et al.,

1997). Furthermore, hPHSs are found in the inner and outer membranes of the nuclear envelope 205

and hence near DNA (Morita et al., 1995; Spencer et al., 1998). DA, L-DOPA, DOPAC and

HVA all led to hPHS-1/2-catalyzed substrate bioactivation and DNA oxidation that were isozyme-dependent. The higher level of oxidative DNA damage caused by these endogenous substrates in hPHS-2 cells, particularly with AA addition, suggests that cells expressing this isozyme may at higher risk for neurodegeneration. Also, elevated levels of PHS-2 in neurons have been found in the brains of patients with Alzheimer's disease and Parkinson's disease

(Teismann et al., 2003; Pasinetti, 1998). Herein, substrate concentrations as low as 10 uM show early oxidative DNA damage, considerably more rapid than that to protein (1 and 6 hr vs 24 hr) and in the absence of cytotoxicity. These results suggest that oxidative damage to DNA may play both an earlier and a more potent role in the pathogenic mechanism. Increased neurodevelopmental deficits in the offspring of DNA repair-deficient Ogg1 knockout mice exposed in utero to the ROS-initiating teratogen methamphetamine revealed the pathogenic potential of oxidative DNA damage (Wong et al., 2008).

Throughout our studies hPHS-1/2 cells treated with DA and L-DOPA generated the most oxidative stress and cytotoxicity compared to the metabolites. Several can serve as cosubstrates with varying efficacies for the peroxidase component of PHS (Markey et al., 1987), but not all can form reactive free radicals that generate ROS. DA and L-DOPA may form the most free radicals, but DOPAC and HVA can do so as well. Although DA, L-DOPA and DOPAC are catechols, only DA and L-DOPA have a side chain amino group to form N-and C-centered radicals as well as semiquinones, which may enhance ROS generation and cytotoxicity

(Arriagada et al., 2004; Hastings, 1995; Mattammal et al., 1995; Segura-Aguilar et al., 1998;

Goncalves et al., 2009). DOPAC and HVA also cannot cyclize to form an aminochrome, which has been implicated in oxidative damage and toxicity (Hastings, 1995; Segura-Aguilar et al., 206

1998; Arriagada et al., 2004). 3-MT is not a catechol, and its amino group does not generate detectable free radicals or cytotoxicity. Free radical generation from these compounds can affect several different molecular targets. DA causes mitochondrial aggregation (Gomez-Santos et al., 2003), and DA and its metabolites inhibit mitochondrial respiration (Arriagada et al.,

2004; Gluck and Zeevalk, 2004). DA free radicals can inactivate ATPase proton pumps in storage vesicles (Terland et al., 2006). Furthermore, other related endogenous compounds, such as 5-S-cysteinyl DA conjugates, may be bioactivated by hPHS-1/2, causing oxidative DNA base modifications that initiate neuronal damage (Spencer et al., 2002).

In summary, our results show that the precursor and metabolites of DA, although generally considered inactive based upon DA receptor binding assays, are potential substrates for human PHS1/2-catalyzed isozyme-dependant bioactivation to potentially neurodegenerative free radical intermediates. ROS formed by the reactions of these free radical intermediates cause oxidative damage to DNA and protein and enhanced cytotoxicity that are isozyme- dependent. Since the cytosolic concentration of precursors and metabolites is often higher than that of the neurotransmitters, and hPHS-2 is induced during aging and neurotoxicity, this mechanism may contribute substantially to neurodegeneration associated with aging and drugs like amphetamines, which initiate neurotransmitter release. 207

2.3 STUDY 3: HUMAN PROSTAGLANDIN H SYNTHASE (hPHS)-1 AND hPHS-2 IN AMPHETAMINE ANALOG BIOACTIVATION, DNA OXIDATION AND CYTOTOXICITYa

Annmarie Ramkissoon and Peter G. Wells

a. This work was supported by a grant from the Canadian Institutes of Health Research (CIHR). AR was supported by a doctoral scholarship from CIHR and the Rx&D Health Research Foundation.

This manuscript is reproduced with permission from: Ramkissoon, A. and Wells, P. G. (2011). Human prostaglandin H synthase (hPHS)-1 and hPHS-2 in amphetamine analog bioactivation, DNA oxidation and cytotoxicity. Toxicol Sci 120(1): 154-162. (Ramkissoon and Wells, 2011b). Copyright 2011. Oxford University Press-Society of Toxicology. DOI: http://toxsci.oxfordjournals.org/content/120/1/154.long 208

2.3.1 ABSTRACT

Neurotoxicity of the amphetamine analogs methamphetamine (METH) and 3,4- methylenedioxyamphetamine (MDA) (the active metabolite of Ecstasy) may involve their prostaglandin H synthase (PHS)-dependent bioactivation to free radical intermediates that generate reactive oxygen species (ROS) and oxidatively damage cellular macromolecules. We used CHO-K1 cell lines either untransfected or stably expressing human PHS-1 (hPHS-1) or hPHS-2 to investigate hPHS isozyme-dependent oxidative damage and cytotoxicity. Both

METH and MDA (250 uM-1000 uM) caused concentration-independent cytotoxicity in hPHS-1 cells, suggesting maximal bioactivation at the lowest concentration. In hPHS-2 cells, with half the activity of hPHS-1 cells, METH (250 uM-1000 uM) cytotoxicity was less than that for hPHS-1 cells, but was increased by exogenous arachidonic acid (AA), which increased hPHS activity. While 10 uM MDA and METH were not cytotoxic, at 100 uM both analogs caused

AA-dependent and concentration-dependent increases in cytotoxicity and DNA oxidation in both hPHS-1/2 cells. The hPHS-2 isozyme appeared to provide more efficacious bioactivation of these amphetamine analogs at lower concentrations. Acetylsalicylic acid, an irreversible inhibitor of both hPHS-1 and hPHS-2, blocked cytotoxicity and DNA oxidation in both cell lines, and untransfected CHO-K1 cells lacking PHS activity were similarly resistant.

Accordingly, isozyme-dependent hPHS-catalyzed bioactivation of METH and MDA can cause oxidative macromolecular damage and cytotoxicity, which may contribute to their neurotoxicity. 209

2.3.2 INTRODUCTION

Prostaglandin H synthase (PHS), also known as cyclooxygenase (COX), is a heme- containing enzyme that catalyzes the initial steps in the production of prostaglandins and thromboxanes (Kaufmann et al., 1997). There are two isoforms of this enzyme, PHS-1 and

PHS-2. PHS-1 is constitutively expressed in most tissues including the brain. PHS-2 is non- constitutive and inducible in most tissues, but is constitutive in the macula densa of the kidney and the vas deferens, and is highly expressed in brain regions such as the cortex, hippocampus and striatum (Kaufmann et al., 1997; Yamagata et al., 1993). Also, elevated levels of PHS-2 in neurons have been found in the brains of patients with Alzheimer's disease and Parkinson's disease (Pasinetti, 1998; Teismann et al., 2003). PHS subcellular locations include the endoplasmic reticulum and the nuclear envelope (Morita et al., 1995).

PHS is a bifunctional enzyme that has cyclooxygenase and peroxidase activities

(Kaufmann et al., 1997). The cyclooxygenase component converts arachidonic acid (AA) to the endoperoxide-hydroperoxide prostaglandin G2. The peroxidase component reduces the hydroperoxide to prostaglandin H2 and in the process a co-substrate can be oxidized (Wells et al., 2009; Marnett, 1990). It is during this step that endogenous compounds or xenobiotics can serve as co-substrates which form free radicals that can generate ROS which oxidatively damage macromolecules such as protein, lipids and DNA (Wells et al., 2009; Marnett, 1990).

The contribution of PHS to xenobiotic bioactivation, embryonic DNA oxidation and developmental toxicity has been characterized for numerous teratogens including phenytoin and structurally related antiepileptic drugs, thalidomide, benzo[a]pyrene and methamphetamine

(METH, Speed) (Wells et al., 2009). Recently, our laboratory has found that the amphetamine 210

analogs 3,4-methylenedioxymethamphetamine (MDMA, Ecstasy), METH and 3,4- methylenedioxyamphetamine (MDA) (the active metabolite of MDMA) can be bioactivated by brain PHS-1 to free radical intermediates that generate ROS and oxidatively damage brain DNA

(Jeng et al., 2006; Jeng and Wells, 2010). MDMA, METH and MDA are common drugs of abuse that promote the release of neurotransmitters, such as dopamine, and through receptor- mediated mechanisms induce euphoria and hallucinations (Wilson et al., 1996; Kalant, 2001).

Alternatively, various animal studies have shown that there are long-term effects of these amphetamine analogs such as nerve terminal degeneration in brain regions such as the striatum, cortex and hippocampus (Cadet and Krasnova, 2009; Jeng et al., 2006). Such chronic effects are thought to be at least in part ROS-mediated, as demonstrated by increased levels of protein carbonyls, lipid peroxidation and oxidatively damaged DNA in degenerated brain regions which can lead to mitochondrial and endoplasmic reticular dysfunction and proapoptotic signaling

(Cadet and Krasnova, 2009; Jayanthi et al., 1998; Jeng et al., 2006).

The unregulated release of the neurotransmitter dopamine can be converted to a reactive quinone that can covalently bind to DNA and protein sulfhydryl groups (Hastings, 1995;

Mattammal et al., 1995). Dopamine quinones can also undergo one-electron reductions catalyzed by NADPH cytochrome P450 reductase (Segura-Aguilar et al., 1998). This reaction can create a redox cycling process with oxygen, leading to the formation of ROS (Segura-

Aguilar et al., 1998). Amphetamine analogs can also be metabolized by cytochrome P450s, especially CYP2D6, to reactive intermediates that generate ROS; however, the levels of P450 in the brain are low (Warner et al., 1997) and other metabolic pathways may contribute to amphetamine toxicity.

Experiments with PHS-1 knockout mice revealed the important role of this isozyme in 211

the molecular mechanism of ROS-mediated neurodegeneration resulting from MDMA administration, where knockouts showed gene dose-dependent decreased DNA oxidation, nerve terminal degeneration and locomotor functional deficits when compared to wild-type animals

(Jeng and Wells, 2010). The same outcomes caused by METH and MDA in CD-1 mice were reduced by pretreatment with the PHS inhibitor acetylsalicylic acid (ASA, aspirin) (Jeng et al.,

2006).

Most studies to date have investigated bioactivation using purified ovine PHS-1 or rodent PHS-1. However, xenobiotic bioactivation by PHS can differ both by PHS isozyme and enzyme source, mainly for ovine or human sources (Liu and Levy, 1998; Wiese et al., 2001).

We wanted to study the bioactivating potential of both human PHS isozymes under stable conditions, as studies using purified PHS are difficult to interpret due to differing isozyme catalytic stability (Xiao et al., 1998). To avoid this confounding factor, we used CHO-K1 cell lines stably expressing hPHS-1 or hPHS-2 and added exogenous AA to further enhance PHS activity. METH and MDA have been shown to be substrates for both ovine PHS-1 and rodent

PHS-1, resulting in their bioactivation to free radical intermediates, which generate ROS causing neurodegenerative oxidative damage to DNA and enhanced neurodegenerative effects

(Jeng et al., 2006). However, there are species differences in metabolism, where the half-life of the amphetamine analogs in rodents is 2-3 hr versus 10 - 12 hr in humans (Kalant, 2001). We sought to determine whether human PHSs can similarly bioactivate these compounds to DNA- damaging reactive intermediates, and if so, whether there is isozyme specificity.

212

2.3.3 MATERIALS AND METHODS

Chemicals and reagents

AA, proteinase K, nuclease P1, acetylsalicylic acid (ASA) and dimethyl sulfoxide

(DMSO) were obtained from Sigma-Aldrich (Oakville, ON, Canada). Alkaline phosphatase was obtained from Roche Diagnostics (Laval, QC, Canada). RNAse A/T1 mix was obtained from Fermentas Canada Inc. (Burlington, ON, Canada). Proteinase K was obtained from

BioShop Canada Inc. (Burlington, ON, Canada). 8-hydroxy-2‟-deoxyguanosine, DUP-697 and

SC-560 were obtained from Cayman Chemical Co. (Ann Arbor, MI). 5-(and-6)-chloromethyl-

2',7'-dichlorodihydrofluorescein diacetate, acetyl ester (CM-H2DCFDA) was obtained from

Invitrogen (Burlington, ON, Canada). All other reagents were of analytical or HPLC grade.

Drugs

Pure racemic (D/L)-METH was provided by the Healthy Environments and Consumer

Safety Branch of Health Canada (Ottawa, Ontario, Canada). MDA was provided by the

Research Technology Branch of the National Institute on Drug Abuse (Rockville, MD, USA).

The identity and purity of the amphetamine analogs was determined by using a Bio-Rad

REMEDi HS system and confirmed by liquid chromatography-mass spectrometry-mass spectrometry (Clinical Biochemistry, Hospital for Sick Children, Toronto, ON, Canada).

Chinese hamster ovary hPHS-1/2 cell lines

Research in Quebec, Canada. Cell lines were grown as a monolayer culture in a humidified environment with 5% CO2 at 37°C as previously described except the hPHS-1 cell line was grown in Ham's F12 media supplemented with 10% fetal bovine serum (Sigma-Aldrich,

Oakville, Ont., Canada), L-glutamine, non-essential amino acids, 100 U/mL penicillin and 100 ug/mL streptomycin and selected using G418 (Sigma-Aldrich, Oakville, ON, Canada)

(Kargman et al., 1996). Untransfected CHO-K1 cells were obtained from Dr. Peter H. Backx,

University of Toronto, and were grown in the same media as the transfected cells but without selection agents.

Measurement of PHS activity and the efficacy of PHS isozyme-specific inhibitors

Cells were seeded overnight at 100,000 cells in a 96-well plate in 200 uL of media in a humidified environment with 5% CO2 at 37°C. The next day media was replaced with 200 uL of 1X PBS and 21 uM CM-H2DCFDA for 30 min at 37°C. To inhibit PHS activity, cells were pretreated with the PHS-1-specific inhibitor SC-560 (5 uM) or the PHS-2-specific inhibitor

Dup-697 (5 uM) or with the dual PHS-1/2 inhibitor ASA (5 mM) for 30 min, after which cells were treated with AA (10 uM) to activate PHS activity. Fluorescence with an excitation wavelength of 485 nm and emission wavelength of 530 nm was recorded at 0 min, 15 min and

30 min after treatment.

Measurement of cytotoxicity by lactate dehydrogenase (LDH) release

Lactate dehydrogenase (LDH) is a cytoplasmic enzyme that can be detected to evaluate the loss of plasma membrane integrity. The CytoTox-ONE TM Homogeneous Membrane

Integrity Assay from Promega was used to assess LDH release. Cells were plated in 96-well 214

plates at a cell concentration of 20,000 to 30,000 cells in 200 uL and allowed to adhere overnight. PHS activity in the cells was activated by preincubation with AA (100 uM) for 2 min, following which METH or MDA was dissolved in DMSO as the vehicle, (total < 0.5%

[v/v] in media), added at various concentrations and the cells were incubated at 37oC for 24 hr.

The concentrations tested were similar to plasma concentrations in MDMA and METH abusers after recreational doses and in the range of extracellular striatal concentrations in rodents after a neurotoxic dosing regimen (Wilson et al., 1996; de la Torre and Farre, 2004; Esteban et al.,

2001; Melega et al., 1995). The sampling times for rapid processes of ROS generation (15, 30 min) and DNA oxidation (6 hr), and the more delayed outcome of cytotoxicity (24 hr) were based upon our previous in vitro cellular and in vivo studies of ROS-initiating xenobiotics (Jeng et al., 2006; Jeng and Wells, 2010; Preston et al., 2009; Ramkissoon and Wells, 2011a). Plates were removed from the incubator for 25 min and 50 uL of supernatant were collected to quantify LDH release according to kit instructions. Fluorescence with an excitation wavelength of 560 nm and emission wavelength of 590 nm was recorded. Percent LDH release was calculated from the experimental LDH release compared to the maximal LDH release from

20,000 to 30,000 cells.

To make consistent comparisons between the different substrates, cell lines and endpoints, we elected to use an irreversible inhibitor of both hPHS-1 and hPHS-2, ASA (5 mM for 1 hr preincubation), which did not interfere with any of the detection methods used. After preincubation, the medium was removed, cells were washed and new warm medium was added prior to treatment with AA and test compounds.

Treatment of cells to assess DNA oxidation 215

Cells were grown to 90% confluence on 10 cm tissue culture plates. PHS activity in the cells was activated by AA (100 uM) for 2 min following which METH or MDA was added at

100 uM, which was the lowest concentration causing cytotoxicity. The cells were then incubated at 37oC for 6 hr. For DNA oxidation studies, cells were scraped into media and 3 plates were pooled in 50 mL Falcon tubes. Cellular pellets were obtained by centrifugation at

1,100 x g for 8 min, the pellets were washed with Hanks‟ balanced salt solution and stored at -

80oC prior to DNA extraction.

DNA extraction from cells

The chaotropic NaI extraction method was used (Ravanat et al., 2002) with the following modifications. To the cellular pellet was added 1 ml of lysis buffer A (320 mM sucrose, 5 mM MgCl2, 10 mM Tris, 0.1 mM desferoxamine pH 7.5, 1% Triton X-100). After hand homogenization, the nuclei were collected following centrifugation at 1,000 x g for 10 min at 4°C and washed with 1 ml of buffer A. To the nuclear pellet, 200 μl of buffer B were added

(10 mM Tris, 5 mM EDTA-Na2, 0.15 mM desferoxamine, pH 8.0, 1% w/v SDS) and hand- homogenized to facilitate lysis of the nuclear membrane. Then, 5 μl of RNase A/T1 mix were added and the samples were incubated for 1.5 hr at 50°C, after which 13.3 μl of proteinase K

(10 mg/ml in 10 mM Tris/1 mM EDTA, pH 8.5) were added prior to incubation at 50°C for 1.5 hr. Subsequently, 300 uL of the NaI solution (7.6 M NaI, 40 mM Tris, 20 mM EDTA-Na2, 0.3 mM desferoxamine, pH 8.0) were added and the tubes were inverted for 2 min. Then, 500 ul of isopropanol were added and DNA was precipitated by gently inverting the tube for 2 min. DNA was recovered by centrifugation at 10,000 x g for 10 min at 4°C and washed with 1 ml of 70% ethanol. After centrifugation (10,000 x g for 5 min), DNA was washed again using 1 ml of 70% 216

ethanol four times. DNA was recovered by centrifugation, dissolved in sodium acetate buffer

(20 mM sodium acetate, 0.1 mM desferoxamine, pH 5.2) for 1 hr at 37oC and then briefly sonicated prior to DNA digestion. DNA was digested with nuclease P1 (5 units) at 37oC. After

1 hr, Tris-HCl (1 M, pH 8.5) was added followed by alkaline phosphatase (8 units) and this solution was incubated at 37oC for 1 hr. Samples were filtered using Microcon®-YM 10 filters

(Millipore Canada Ltd.) and stored at -80oC until analysis.

Detection of 8-oxo-2’-deoxyguanosine (8-oxo-dG) and 2-deoxyguanosine (2-dG)

Oxidation of 2-dG to 8-oxo-dG was quantified using an isocratic Series 200 high- performance liquid chromatographic (HPLC) system (PerkinElmer Instruments LLC, Shelton,

Connecticut). The system was equipped with a SUPELCOSIL LC-18 column (250 mm x 4.6 mm, Sigma-Aldrich, Oakville, ON, Canada), an electrochemical (EC) detector (Coulochem II,

8-oxodG per ug dG.

Statistical analysis 217

Multiple comparisons among groups were analyzed by one-way ANOVA with a subsequent Tukey‟s test (GraphPad InStat®5.0, GraphPad Software, Inc., San Rafael, CA,

USA). The level of significance was determined to be at P < 0.05. 218

2.3.4 RESULTS

PHS activity in CHO-K1 cell lines stably expressing human PHS-1 or PHS-2 (hPHS-1/2)

PHS-1/2 activities for the CHO-K1 cell lines used herein were measured previously by quantification of prostaglandin E2 formation (Ramkissoon and Wells, 2011a). Exogenous AA substrate was used to increase the PHS activity in the various cell lines. The resulting AA- dependent PHS activities were in the range of published values in that CHO-K1 cells had the lowest activity, and hPHS-1 cells exhibited double the activity found in hPHS-2 cells (Kargman et al., 1996; Ramkissoon and Wells, 2011a). Without AA addition, hPHS activity in hPHS-1 cells and hPHS-2 cells was respectively 15-fold and 8-fold higher than that in untransfected cells (Ramkissoon and Wells, 2011a). In the study herein, we specifically evaluated the activity of the peroxidase component of hPHS using the probe CM-H2DCFDA, which is oxidized by the peroxidase component to a fluorescent product when PHS is activated by AA addition. With

AA, both hPHS-1 and hPHS-2 cells showed increased activity within 15 min of AA addition

(p<0.01), and activity was further increased by 30 min (p<0.01) (Figure 1). hPHS-1 cells exhibited double the activity of the hPHS-2 cell line during a 15 min incubation (Figure 1, upper and lower panels). Activities were reduced using isozyme-specific or dual inhibitors of hPHS-1 and hPHS-2 (Figure 1), or by withholding the PHS substrate AA (data not shown). The untransfected and hPHS-expressing cell lines were used as models to investigate isozyme specificity for hPHS-catalyzed bioactivation of the amphetamine analogs leading to ROS generation and cytotoxicity.

Cytotoxicity caused by METH and MDA in hPHS1/2-expressing cells

LDH release was measured to assess the potential of these amphetamine analogs to 219

cause cytotoxicity in the various cell lines. In all cell lines, AA (100 uM) addition on its own did not lead to enhanced cytotoxicity (Figure 2). The untransfected CHO-K1 cell line lacking

PHS was resistant to cytotoxicity even after concentrations of 1000 uM of METH or MDA

(Figures 2 and 3). After a 24 hr incubation with METH (250 uM-1000 uM), the hPHS-1 cell line showed enhanced cytotoxicity by up to 4-fold compared to untransfected CHO-K1 cells

(p<0.01) (Figure 2). Cytotoxicity was maximal at the lowest analog concentration (250 uM), with no concentration-dependent increase, and the addition of AA, to further increase PHS activity, did not further enhance cytotoxicity (Figure 2). In the hPHS-2 cell line, METH (250 uM-1000 uM), caused an increase in cytotoxicity which was lower than that observed in hPHS-

1 cells (Figure 2). At 1000 uM, METH caused a 3-fold increase in LDH release in hPHS-2 cells compared to vehicle controls (p<0.01), and AA addition enhanced cytotoxicity by an average of 40% in hPHS-2 cells incubated with 500 and 1000 uM METH (p<0.05). Increasing concentrations of METH enhanced cytotoxicity, with over a 2-fold increase in LDH release going from 250 to 1000 uM METH.

MDA (250 uM-1000 uM) effects were similar to those for METH, in that MDA- exposed hPHS-1 cells compared to untransfected CHO-K1 cells had a 4-fold, concentration- independent increase in cytotoxicity (p<0.01) (Figure 2). In the hPHS-2 cell line, as with

METH, MDA (250 uM-1000 uM) caused an increase in cytotoxicity (Figure 2). At 1000 uM,

MDA in hPHS-2 cells caused a 2.5-fold increase in LDH release compared to vehicle controls

(p<0.01), and AA addition enhanced cytotoxicity by an average of 35% for 500 and 1000 uM

MDA (p<0.05). As with METH in this cell line, increasing concentrations of MDA enhanced cytotoxicity, with a 50% increase in LDH release going from 250 to 1000 uM MDA.

To corroborate PHS-dependency, we pretreated both hPHS cell lines with the dual PHS- 220

1/2 inhibitor ASA, and then treated with METH at a concentration of 500 uM, which increased cytotoxicity in both PHS cell lines. PHS inhibition blocked METH-initiated cytotoxicity in both hPHS cell lines with and without AA (p<0.01) (Figure 3). Since maximal toxicity in the PHS-1 cell line appeared to have been achieved with a 250 uM concentration, lower amphetamine analog concentrations were employed to more fully reveal the roles of AA-dependence and isozyme specificity on concentration-response relationships obscured at higher concentrations.

MDA and METH at 100 uM caused a 40-50% increase in cytotoxicity in both the hPHS-1/2 cell lines, but only when PHS activity was increased with AA addition (p<0.05) (Figure 4). The magnitude of cytotoxicity caused by the lower concentration (100 uM) of amphetamine analogs in the PHS-1 cell line (Figure 4) was less than one-third of that observed with the 250 uM concentrations of METH and MDA shown in Figure 2. At the 100 uM concentration of amphetamine analogs, hPHS-2 cells exhibited cytotoxicity similar to that observed in hPHS-1 cells, despite hPHS-2 cells having only half the PHS activity of hPHS-1 cells. ASA pretreatment blocked cytotoxicity in both cell lines (Figure 4). MDA and METH at 10 uM did not increase cytotoxicity in either hPHS-1/2 cell lines (data not shown).

DNA oxidation caused by hPHS1/2-catalayzed bioactivation of METH and MDA

DNA oxidation was quantified by the formation of 8-oxo-dG, which is formed when a hydroxyl radical reacts with 2-deoxyguanosine in DNA. Basal levels of DNA oxidation were not different in the various cell lines (Figure 5). After a 6-hr incubation with 100 uM of METH or MDA, the untransfected CHO-K1 cell line lacking PHS was resistant to DNA oxidation

(Figure 5). DNA oxidation was enhanced in both hPHS-1 and hPHS-2 cells by 100 uM METH and MDA, but only when PHS activity was increased with AA addition (p<0.05) (Figure 5). 221

AA addition caused average 65% and 35% respective increases in METH- and MDA-initiated

DNA oxidation in hPHS-1/2 cells (p<0.05) (Figure 5). Furthermore, ASA pretreatment blocked

DNA oxidation in both cell lines (Figure 5). At a lower concentration of either analog (10 uM),

DNA oxidation was not enhanced within 6 hr (data not shown). 222

Figure 1: PHS activity in CHO-K1 cells stably expressing human prostaglandin H synthase-1 (hPHS-1) or hPHS-2. The activity of hPHS-1 and hPHS-2 was assessed using the probe CM-H2DCFDA which is oxidized to a fluorescent product by the peroxidase component of PHS when PHS is activated by the addition of arachidonic acid (AA) substrate. Cells were pretreated with the probe, PHS was activated by AA addition (10 uM) and fluorescence values were obtained at 0, 15 and 30 min (Upper panel: hPHS-1, Lower panel: hPHS-2). SC-560 (5 uM) was used as a specific hPHS-1 inhibitor and DUP-697 (5 uM) was used as a specific hPHS- 2 inhibitor. ASA (5 mM) was used as an irreversible inhibitor of both hPHS-1 and hPHS-2. a=p< 0.01 compared to AA only at 0 min. b=p<0.01 compared to AA only at 15 min. c=p<0.01 compared to AA only at 30 min. Each bar represents the mean with the standard deviation, n = 6. 223

Figure 2: METH and MDA in hPHS-1- or hPHS-2-catalyzed bioactivation and cytotoxicity. CHO-K1 cell lines either untransfected or stably expressing hPHS-1 or hPHS-2 were activated with AA (100 uM) and incubated with methamphetamine (METH) or 3,4- methylenedioxyamphetamine (MDA) at concentrations of 250, 500 and 1,000 uM for 24 hr. Cytotoxicity was measured by the release of lactate dehydrogenase (LDH). a=p<0.01 compared to 0 uM of the same cell line. *=p<0.05 compared to No AA of the same cell line and drug concentration. Each bar represents the mean with the standard deviation, n = 6. 224

Figure 3: ASA protection in hPHS-catalyzed bioactivation and cytotoxicity of METH. CHO-K1 cell lines expressing hPHS-1 or hPHS-2 were pre-treated with the dual PHS-1/2 inhibitor acetylsalicylic acid (ASA) (5 mM) for 1 hr. Cells were washed, activated with AA (100 uM) and treated with METH (500 uM) for 24 hr. Cytotoxicity was measured by the release of LDH. a=p<0.01 compared to 0 uM of the same cell line. b=p<0.01 compared to 500 uM with or without AA of the same cell line. *=p<0.05 compared to No AA of the same cell line and drug concentration. Each bar represents the mean with the standard deviation, n = 6. 225

Figure 4: ASA protection in hPHS-catalyzed bioactivation and cytotoxicity of 100 uM METH or MDA. CHO-K1 cell lines expressing hPHS-1 or hPHS-2 were pre-treated with the PHS-1/2 inhibitor ASA (5 mM) for 1 hr. Cells were washed, activated with AA (100 uM) and treated with METH or MDA (100 uM) for 24 hr. Cytotoxicity was measured by the release of LDH. a=p<0.05 compared to drug-treated cells with No AA of the same cell line. Each bar represents the mean with the standard deviation, n = 8. 226

Figure 5: DNA oxidation caused by hPHS-catalyzed bioactivation of METH and MDA, and protection by ASA. CHO-K1 cell lines either untransfected or stably expressing hPHS-1 or hPHS-2 were activated with AA (100 uM) and treated with METH or MDA at a concentration of 100 uM for 6 hr. Oxidized DNA was quantified by 8-oxo-2'-deoxyguanosine (8-oxo-dG) formation using high-performance liquid chromatography with electrochemical detection (HPLC-EC), and was standardized using 2‟-deoxyguanosine levels quantified by HPLC-UV. Cell lines were also pre-treated with the dual PHS-1/2 inhibitor ASA and treated as described above. a=p<0.05 compared to 0 uM and 100 uM with no AA of the same cell line. Each bar represents the mean with the standard deviation, n = 5. 227

2.3.5 DISCUSSION

The studies herein using cell lines stably expressing hPHS-1 or hPHS-2 provide the first direct evidence that both human PHS isozymes can readily bioactivate METH and MDA to free radical intermediates that generate ROS, oxidize DNA and enhance cytotoxicity. Within the cyclooxygenase and peroxidase components there are major differences between the isoforms which can alter bioactivation and free radical generation (for review see (Kulmacz et al., 2003).

In the cyclooxygenase component, the COX active site is larger in PHS-2 than PHS-1 and results in a more flexible substrate channel (Lecomte et al., 1994). In the peroxidase component, there are a number of different residues between the enzymes on the sidechain near the peroxidase site, which may affect the binding of cosubstrates, and the peroxidase activity of

PHS-1 in vitro is more stable than that of PHS-2 (Kulmacz et al., 2003; Xiao et al., 1998).

Other studies have shown differential PHS-catalyzed bioactivation depending upon both the enzyme source (ovine or human) and the isozyme (Liu and Levy, 1998; Wiese et al., 2001).

In our studies, at higher concentrations (250 uM-1000 uM) in the hPHS-1 cell line, both

METH and MDA substantially increased cytotoxicity compared to CHO-K1 cells, but in a concentration-independent and AA-independent fashion, indicating maximal cytotoxicity was achieved in the hPHS-1 cells with the 250 uM amphetamine analog concentration. At a lower analog concentration (100 uM), there was a clear AA-dependence for hPHS-1- as well as hPHS-

2-dependent DNA oxidation and cytotoxicity. This indicates that at lower concentrations of amphetamine analogs, higher PHS activity, via exogenous AA addition, is required to generate sufficient free radicals and ROS to oxidize DNA and subsequently lead to cytotoxicity. The concentration dependence for amphetamine cytotoxicity in hPHS-2 cells was indicated by the 228

over 2-fold higher LDH release for METH going from a concentration of 250 uM to 1000 uM, and the similar but lesser 50% increase for MDA over the same concentration range. A similar concentration-dependence for amphetamine analog cytotoxicity in PHS-1 cells was evident across studies for LDH release, which for both METH and MDA was increased over 3-fold going from a concentration of 100 uM to 1000 uM. A similar but lesser amphetamine concentration-dependent increase in cytotoxicity across studies was observed with the hPHS-2 cell line. The AA-dependence of hPHS activity and analog-initiated DNA oxidation and cytotoxicity was confirmed by hPHS inhibition using isozyme-specific (SC-560 for PHS-1,

DUP-697 for PHS-2) and/or dual PHS-1/2 (ASA) inhibitors. ASA inhibited both the basal level of PHS activity in the hPHS cell lines and the increased activity with exogenous AA addition, and accordingly blocked amphetamine-initiated cytotoxicity and DNA oxidation in each hPHS cell line both with and without AA addition.

At the 100 uM analog concentration with AA, the equivalent cytotoxic responses observed in hPHS-1 and hPHS-2 cells, despite only half the PHS activity in the latter cell line, provides evidence that the bioactivating efficacy of hPHS-1 is substantially less than that of hPHS-2 for these amphetamine analogs at lower concentrations on the concentration-response curve. There is no evidence that higher concentrations of these amphetamine analogs are selective hPHS-1 inhibitors, but our results do not preclude this possibility. At the lower concentration of 100 uM, cytotoxicity in the hPHS-1 cell line dropped significantly (15% LDH release compared to 55% with 250-1,000 uM METH and MDA), indicating a concentration- response relationship for hPHS-1 cells consistent with a maximal response and/or saturation or inhibition of hPHS-1 catalyzed bioactivation at the higher concentrations, which was not evident in the hPHS-2 cells. Similar enhanced cytotoxicity in the hPHS-2 cell line has been shown for 229

endogenous substrates; namely, dopamine, its precursor and some metabolites (Ramkissoon and

Wells, 2011a). It has been reported that MDMA is not only a substrate for CYP2D6, but also a potent competitive inhibitor of CYP2D6 in human liver microsomes (Heydari et al., 2004). Our data are consistent with the non-linear pharmacokinetics of MDMA in human exposures where small increases in the dose of MDMA ingested lead to disproportionate increases in MDMA plasma concentrations, and hence increased vulnerability to toxicity (de la Torre et al., 2000).

This may have occurred, in our PHS cell lines, in which a small increase in the concentration of amphetamine analogs caused increased toxicity. Currently, it is unknown if the isozymes can form different types of free radicals that are more toxic and that may contribute to the enhanced toxicity in hPHS-1 cells at the high concentrations.

About one-half of METH in humans is excreted unchanged (Caldwell et al., 1972; Cook et al., 1993), and the remainder undergoes CYP2D6-catalyzed N-demethylation to amphetamine, and 4-hydroxylation. 4-Hydroxy-methamphetamine is the predominant metabolite, constituting almost 50% of all metabolites excreted. These metabolites accumulate in the striatum after administration of the parent drug (Melega et al., 1995). MDMA is N- demethylated to MDA or O-demethylenated to 3,4-dihydroxymethamphetamine (HHMA), which is further metabolized to catechol intermediates. These metabolites are believed to be downstream products formed after the opening of the methylendioxyphenyl ring, a process that is mainly mediated by CYP2D6 (Kreth et al., 2000), with low-affinity contributions from

CYP1A2, CYP2B6, and CYP3A4 (Kreth et al., 2000; Meyer et al., 2008). Unstable catechols can be oxidized to reactive quinones that redox cycle to semiquinone radicals that generates

ROS (Hiramatsu et al., 1990). While it is uncertain whether other cytochromes P450 (CYPs) may contribute to ROS generation, the expression of CYPs in the brain is around 1-2% of that in 230

the liver (Warner et al., 1997). Our studies highlight the potential for the metabolites of METH and MDMA to serve as substrates for hPHSs, which may serve as an alternative to CYPs for neurotoxic free radical generation.

There are structural similarities between the amphetamine analogs and dopamine, as well as its precursors and metabolites, suggesting these substrates may be bioactivated similarly and may share a common mechanism for initiating toxicity. Dopamine, its precursors and metabolites may be also be important in amphetamine neurotoxicity, as amphetamine and its analogs acutely release neurotransmitters, such as dopamine, into the cytosol where they can be bioactivated, leading to enhanced neurotoxicity (Kalant, 2001). Previous studies in CHO cells have shown that dopamine is not present in this cell line (Montine et al., 1994). Our METH and

MDA data reveal that these compounds on their own, in the absence of dopamine, can increase cytotoxicity and oxidative stress in the context of PHS activation. This is especially important with chronic abuse of amphetamine analogs, which has been shown to reduce the concentration of dopamine and its metabolites in vivo (Gluck et al., 2001). Furthermore, PHS-2 is induced by

METH administration and mice lacking the gene for COX-2 were resistant to METH-initiated neurotoxicity (Thomas and Kuhn, 2005a).

Plasma concentrations of MDMA and METH after recreational doses are usually in the range of 4 μM in humans (Wilson et al., 1996; de la Torre et al., 2000); however, concentrations in brain may be substantially higher. Studies in rats with doses of 5-10 mg/kg of MDMA or

METH achieve extracellular striatal concentrations of 10-100 uM (Melega et al., 1995; Esteban et al., 2001), which are usually on average 10 times higher than plasma concentrations. We tested METH and MDA at pharmacologic concentrations between 10-100 uM and assessed cytotoxicity 24 hr later. While our 100 uM amphetamine analog concentrations revealed DNA 231

oxidation and cytotoxicity in the hPHS cell lines, these effects may be enhanced after longer durations of exposure as seen in vivo where 3-7 days following as little as a single-day‟s dosing are required for the development of long-term neurodegeneration, as distinct from acute and reversible neurotoxicity (Jeng et al., 2006; Jeng and Wells, 2010). Also, while other in vitro studies using neuronal cell cultures have found significant toxicity with high mM concentrations

(Huang et al., 2009; Nara et al., 2010), our non-neuronal cell lines exhibited enhanced cytotoxicity with enhanced hPHS activity with concentrations as low as 100 uM after just 24 hours. In contrast, at concentrations up to 1000 uM, our untransfected cell lines with low PHS activity were not sensitive to oxidative stress or cytotoxicity. Neuronal cultures require higher concentrations of METH for toxic effects, suggesting that the in vivo regional environment may provide signals necessary to enhance PHS activity and METH bioactivation, and hence the initiation of nerve terminal degeneration. These signals include cytokines, ROS, calcium, neurotransmitters and xenobiotics such as the tumor promoter 12-O-tetradecanoylphorbol-13- acetate (TPA) which can increase AA release from phospholipids, activating PHS and its bioactivation of METH and MDA in vivo (Schaloske and Dennis, 2006; Wells and Vo, 1989).

Additionally, growth factors and cytokines involved in inflammatory responses can also induce

PHS-2 (Jones et al., 1993; Kujubu et al., 1991). Furthermore, bioactivation can occur in brain cells other than neurons, for example in microglia, where the expression and activation of PHSs is associated with the neurotoxic properties of METH (Choi et al., 2009; Thomas and Kuhn,

2005b).

In summary, our studies showed that human PHS-1 and PHS-2 can bioactivate pharmacologically relevant concentrations of the amphetamine analogs METH and MDA to free radical intermediates that initiate the oxidation of DNA and lead to cytotoxicity. Bioactivation 232

was both AA-dependent and isozyme-dependent, with the hPHS-2 isozyme appearing to provide more efficacious bioactivation of these amphetamine analogs. This hPHS-dependent mechanism may play a significant role in the neurodegenerative effects of amphetamine analogs, where abuse of these drugs and other environmental factors enhance PHS activity and reduce catecholamine levels, thereby enhancing analog bioactivation and free radical generation. 233

2.4 STUDY 4: METHAMPHETAMINE OXIDATIVE STRESS, NEUROTOXICITY AND FUNCTIONAL DEFICITS ARE MODULATED BY NRF2a

Annmarie Ramkissoon and Peter G. Wells

a. A preliminary report of this research was presented at the annual meeting of the Society of Toxicology (Toxicol. Sci. (Supplement: The Toxicologist) 108 (1): 448 (No. 2152), 2009). This research was supported by a grant from the Canadian Institutes of Health Research (CIHR). AR was supported by a doctoral scholarship from CIHR and the Rx&D Health Research Foundation. 234

2.4.1 ABSTRACT Activation of redox-sensitive transcription factors like nuclear factor-E2-related factor 2 can enhance the transcription of cytoprotective genes during oxidative stress. We investigated whether Nrf2 is activated by methamphetamine thereby altering its neurotoxicity. Multiple-day dosing with METH enhanced DNA oxidation and decreased tyrosine hydroxylase and dopamine transporter staining in the striatum, all more severe in Nrf2 knockout mice, as were deficits in motor coordination and olfactory discrimination. Similarly, METH increased striatal glial fibrillary acidic protein, indicating neurotoxic glial activation. Nrf2-mediated protection against

METH was also observed in the glial cells and in the GABAergic system of the olfactory bulbs, whereas dopaminergic parameters were unaffected. With 1-day dosing of METH, there were no differences between Nrf2 wild-type and Nrf2 knockout mice in either basal or METH-enhanced

DNA oxidation and neurotoxicity markers. Nrf2-mediated pathways accordingly protect against the neurodegenerative effects and functional deficits initiated by METH, and perhaps other

ROS-enhancing neurotoxins, when repeated exposures allow time for transcriptional activation and protein induction. In human users of METH, this mechanism may be essential when differences in drug abuse patterns may alter the induction and duration of Nrf2 activation thereby modulating susceptibility to the neurotoxic effects of METH. 235

2.4.2 INTRODUCTION Methamphetamine (METH, Speed) along with other amphetamines are common drugs of abuse that promote the release of neurotransmitters and acutely induce euphoria and hallucinations (Kalant, 2001). These acute receptor-mediated effects are due to the rapid release of neurotransmitters from synaptic vesicles (Kalant, 2001). Alternatively, various animal studies have shown that amphetamines cause long-term effects like nerve terminal degeneration in brain regions such as the striatum, cortex, hippocampus and olfactory bulb (O'Callaghan and

Miller, 1994; Deng et al., 2007; Jeng and Wells, 2010; Jeng et al., 2006). Such chronic effects are thought to be mediated at least in part by reactive oxygen species (ROS) (Gluck et al., 2001;

Jayanthi et al., 1998; Jeng et al., 2006). Bioactivation of METH by enzymes like prostaglandin

H synthases (PHSs) can lead to the formation of reactive free radical intermediates that can enhance the formation of ROS (Jeng et al., 2006; Jeng and Wells, 2010), which can oxidatively damage cellular macromolecules including proteins, lipids and DNA (Halliwell and Gutteridge,

2007). Such damage, if not repaired, can accumulate over time and can lead to loss of cellular function and even cell death (Wells et al., 2009). However, ROS can also activate redox- sensitive transcription factors that can lead to the transcription of genes which modulate toxicity.

The redox-sensitive transcription factor nuclear factor erythroid 2-related factor 2 (Nrf2) is frequently activated during various types of oxidative stress in the brain (Burton et al., 2006;

Calkins et al., 2005; Chan et al., 2001). Under normal conditions, Nrf2 is located in the cytoplasm as an inactive form associated with its repressor protein Kelch-like ECH Associating

Protein 1 (Keap1) (Itoh et al., 1999b). Oxidation of redox-sensitive cysteines in Keap1 during oxidative stress leads to dissociation of Nrf2 from Keap1. This allows Nrf2 to translocate to the nucleus where it heterodimerizes with members of the small avian musculoaponeurotic 236

fibrosarcoma (Maf) protein family and binds to the antioxidant response element (ARE), an enhancer sequence that regulates the transcription of cytoprotective enzymes such as heme oxygenase-1 (HO-1) and NAD(P)H:quinone oxidoreductase (NQO1) and oxoguanine glycosylase 1 (Ogg1) (Dhenaut et al., 2001; Lee et al., 2003b). Generally, its target gene functions include glutathione synthesis and homeostasis, free radical detoxification, antioxidative response and repair. It has been suggested that METH-initiated neurotoxicity does not involve Nrf2, as dopamine depletion in the striatum at two weeks post-treatment was not different between Nrf2 knockout mice and their wild-type controls (Pacchioni et al., 2007).

However, studies in rat striatum indicate that METH can activate Nrf2 and regulate HO-1 expression (Jayanthi et al., 2009). Accordingly, it is uncertain whether ROS-mediated Nrf2 activation in specific brain regions caused by METH can modulate its neurotoxicity.

We hypothesized that METH-initiated ROS generation activates the redox-sensitive transcription factor Nrf2 and plays an important neuroprotective role during neurotoxicity. If so, then METH neurotoxicity should be greater in Nrf2 knockout mice compared to their wild- type controls. Since METH-initiated, ROS-mediated DNA oxidation varies among different brain regions (Jeng et al., 2006), several brain regions needed to be assessed using a wide range of endpoints to reliably determine whether Nrf2 plays a neuroprotective role in METH neurotoxicity. Given the relatively short half-life of METH and its rapid initiation of DNA oxidation, we also anticipated that any protective role for Nrf2 would be revealed only if METH administration was preceded by pretreatment with a Nrf2 activator, employing a time interval sufficient to allow for gene transcription and protein induction. In this study, ROS-initiating

METH was used as both the Nrf2 activator and the neurotoxin, as would occur in drug abusers using METH repeatedly. Our results in Nrf2 knockout mice provide the first evidence that Nrf2 237

is protective against METH-initiated DNA oxidation and dopaminergic and GABAergic neurotoxicity in selective brain regions, but only when METH treatment is preceded by exposure to oxidative stress sufficiently earlier to allow time for increased gene expression and the induction of cytoprotective proteins. Our results also provides evidence for METH-initiated long-term deficits in olfactory discrimination, which appear to constitute a more sensitive functional biomarker of long-term METH neurodegeneration in mice than standard tests currently employed. 238

2.4.3 MATERIALS AND METHODS Chemicals and reagents

Alkaline phosphatase and Complete, Mini, EDTA-free protease inhibitor cocktail tablets were obtained from Roche Diagnostics (Laval, QC, Canada). RNAse A/T1, Taq polymerase and dNTPs were purchased from Fermentas Canada Inc. (Burlington, ON, Canada). Proteinase

K and dithiothreitol (DTT) were obtained from BioShop Canada Inc. 8-hydroxy-2‟- deoxyguanosine was obtained from Cayman Chemical Co. (Ann Arbor, MI). RNAlater® and

DNase were purchased from Qiagen (Mississauga, ON, Canada). 2-dG and pepstatin A were obtained from Sigma-Aldrich (Oakville, ON, Canada). All other reagents were of analytical or

HPLC grade.

Drugs

Pure racemic (D/L)-METH was provided by the Healthy Environments and Consumer

Safety Branch of Health Canada (Ottawa, Ontario, Canada). The identity and purity of (D/L)-

METH was determined by using a Bio-Rad REMEDi HS system and confirmed by high- performance liquid chromatography-mass spectrometry-mass spectrometry (Clinical

Biochemistry, Hospital for Sick Children, Toronto, ON, Canada).

Nrf2 knockout mouse colony

The pathway of Nrf2 activation by METH in the brain was investigated in vivo using young adult Nrf2 knockout mice that were provided by Professor Masayuki Yamamoto

(Department of Medical Biochemistry, Tohoku University Graduate School of Medicine,

Sendai, Japan) through the RIKEN Bioresource Center, Experimental Animal Division. The 239

colony was maintained on a CD1 background and was inbred by mating heterozygous littermates. F9 littermates were used and all Nrf2 genotypes were confirmed as previously described (Itoh et al., 1997). Adult Nrf2 wild-type and knockout (KO) mice 2-4 months old were used in this study, following confirmation of the Nrf2 genotype in breeding females and males. Mice were housed in plastic cages with ground corn cob bedding (Beta Chip;

Northeastern Products, Warrensburg, NY, USA) and maintained in temperature-controlled rooms with a 12 hr light-dark cycle. Food (Laboratory Rodent Chow 5001; Ralston Purina,

Strathroy, ON, Canada) and tap water were provided ad libitum. All animal studies were approved by the University of Toronto Animal Care Committee in accordance with the standards of the Canadian Council on Animal Care.

Dosing regimens

Drugs were dissolved with sterilized 0.9% saline, and the drug or its vehicle were given by intraperitoneal (i.p.) injection in a fixed volume of 0.1 ml/10 g body weight. Nrf2 wild-type and KO mice were treated with a single 10 mg/kg i.p. dose of METH in 0.9% saline (vehicle) to determine temporal induction profiles of Nrf2-regulated antioxidative pathways. In the acute

“binge” dosing regimen, mice were administered 4 doses of METH at 10 mg/kg, i.p. each (total dose of 40 mg/kg), with a 2-hr interval between each dose. To further investigate the role of

Nrf2 in METH-initiated neurotoxicity, we employed a 2-day METH dosing regimen, as distinct from the single-day regimen discussed above. METH (10 mg/kg i.p.) was administered on Day

1 to initiate the expression of Nrf2-regulated genes, followed by a 24 hr period to allow for the

Nrf2-mediated induction of proteins involved in antioxidative pathways, DNA repair and other potentially protective pathways. After the 24 hr period on Day 2, a second dose of METH in 240

varying amounts was administered to determine the protective efficacy of induced Nrf2- regulated pathways on METH-initiated neurotoxicity. On Day 2 we tried four 10 mg/kg i.p. doses of METH, with a 2-hr interval between each dose (total 2-day dose of 50 mg/kg). This dosing regimen increased lethality, so the Day 2 dose was reduced by one-half to four 5 mg/kg doses of METH, with a 2-hr interval between each dose (total 2-day dose of 30 mg/kg). The latter regimen causes significant neurotoxicity in mice, and is required to achieve a plasma concentration in mice similar to that resulting from a human METH binge pattern of self- administration (O'Callaghan and Miller, 1994; Cho et al., 2001). To further investigate dose- response relationships and the role of Nrf2 in modulating METH-initiated functional deficits, we tested two lower dose regimens: (1) a “Low-dose” 2-day dosing regimen using the same single 10 mg/kg i.p. dose of METH dose given on Day 1, but only a single 10 mg/kg i.p. dose of

METH on Day 2; or, (2) a 4-day dosing regimen of 10 mg/kg i.p. each day for 4 days.

Analysis of mRNA levels by quantitative real-time PCR (RT-PCR)

Brains were removed and dissected to isolate the striatum and olfactory bulbs which were then stored in RNAlater solution (Qiagen, Mississauga, ON, Canada). Total RNA from these regions was extracted using the RNeasy® lipid tissue mini kit according to the manufacturer‟s instructions (Qiagen, Mississauga, ON, Canada). To ensure purified RNA, on- column DNase digestion of DNA was performed using the RNase-free DNase set (Qiagen,

Mississauga, ON, Canada). The ratio of readings at 260 nm to 280 nm were determined to be in the range of 1.9-2.1 in 10 mM Tris-Cl, pH7.5 buffer. A total of 2.5 µg of RNA were reverse- transcribed to cDNA using the First Strand cDNA Synthesis Kit (Fermentas, Burlington, ON,

Canada) using random hexamer primers. The PCR reaction was performed in 20 µl using the 241

Lightcycler® FastStart DNA Master SYBR Green 1 reagents (Roche Diagnostics, Laval, QC,

Canada) and 2 uL of cDNA and 0.5 uM of reverse and forward primers. Specific primers were as follows: HO-1-f (CAA-GCC-GAG-AAT-GCT-GAG-TTC-ATG), HO-1-r (GCA-AGG-GAT-

GAT-TTC-CTG-CCA-G), NQO1-f (GCG-AGA-AGA-GCC-CTG-ATT-GTA-CTG), NQO1-r

(TCT-CAA-ACC-AGC-CTT-TCA-GAA-TGG), OGG1-f (ACT-GCA-TCT-GCT-TAA-TGG-

CC), OGG1-r (CGA-AGG-TCA-GCA-CTG-AAC-AG), β-actin-f (AGA-GCA-TAG-CCC-

TCG-TAG-AT), β-actin-r (CCC-AGA-GCA-AGA-GAG-GTA-TC). Quantification was performed on a Lightcycler® 2.0 detection system (Roche Diagnostics, Laval, QC, Canada).

PCR cycles proceeded as follows: initial denaturation for 10 min at 95°C, then 40 cycles of denaturation (10 s, 95°C), annealing (30 s, 55-60°C depending on the specific primer), and extension (20 s, 72°C). The melting-curve analysis showed the specificity of the amplifications.

A series of dilutions were used to generate a relative standard curve to determine the efficiency of amplifications. The relative quantification method was used to quantify mRNA levels using the housekeeping gene β-actin as the reference gene.

DNA extraction

The chaotropic NaI method was used (Ravanat et al., 2002) with the following modifications. To the dissected brain regions was added 1 ml of lysis buffer A (320 mM sucrose, 5 mM MgCl2, 10 mM Tris, 0.1 mM desferoxamine pH 7.5, 1% Triton X-100). After hand homogenization, the nuclei were collected by centrifugation at 1000 x g for 10 min at 4°C and washed with 1 ml of buffer A. To the nuclear pellet, obtained after centrifugation (1000 x g for 10 min at 4°C) 400 μl of buffer B were added (10 mM Tris, 5 mM EDTA-Na2, 0.15 mM desferoxamine, pH 8.0, 1% w/v SDS) and hand-homogenized to allow lysis of the nuclear 242

membrane. Then, 10 μl of RNase A/T1 mix was added and the samples were incubated for 2 hr at 50°C. A 20 μl volume of proteinase K (20 mg/ml in 10 mM Tris/1 mM EDTA, pH 8.5) was added prior to incubation at 50°C for 2 hr. Samples were then spun at 10,000 x g for 15 min to remove particulate material. Subsequently, 600 uL of the NaI solution (7.6 M NaI, 40 mM Tris,

20 mM EDTA-Na2, 0.3 mM desferoxamine, pH 8.0) were added and the tubes were inverted for

2 min. Then, 1 mL of isopropanol was added and DNA was precipitated by gently inverting the tube for 2 min. DNA was recovered by centrifugation at 10,000 x g for 10 min at 4°C and washed with 1 ml of 70% ethanol. After centrifugation (10,000 x g for 5 min), DNA was washed again using 1 ml of 70% ethanol four times. DNA was recovered by centrifugation and dissolved into sodium acetate buffer (20 mM sodium acetate, 0.1 mM desferoxamine, pH 5.2) for 1 hr at 37oC and by briefly sonicating on ice prior to DNA digestion. DNA was digested with nuclease P1 (5 units) at 37oC. After 1 hr, Tris-HCl (1 M, pH 8.5) was added followed by alkaline phosphatase (8 units) and this solution was incubated at 37oC for 1 hr. Samples were filtered using Microcon®-YM 10 filters (Millipore Canada Ltd.) and stored at -80oC until analysis.

Detection of 8-hydroxy-2’-deoxyguanosine (8-oxodG) and 2-deoxyguanosine (2-dG)

Oxidation of 2-dG to 8-oxodG was quantified using an isocratic Series 200 high- performance liquid chromatography (HPLC) system (PerkinElmer Instruments LLC, Shelton,

Connecticut) equipped with a SUPELCOSIL LC-18 column (250mm x 4.6mm, Sigma-Aldrich,

Oakville, ON, Canada), an electrochemical (EC) detector (Coulochem II), a guard cell (model

5020), an analytical cell (model 5010) (Coulochem, ESA, Chelmsford, MA), a UV detector

(series 200, PerkinElmer Instruments LLC, Shelton, Connecticut), and an integrator 243

(PerkinElmer NCI 900 Interface). The filtered samples were injected into the HPLC system and eluted using a mobile phase which contained 50 mM sodium phosphate buffer (pH 5.5)- methanol (95:5, v/v) at a flow rate of 1.0 ml/min. The nucleoside dG was detected by UV absorption at 280 nm and 8-oxodG was monitored by EC detection with channel 1 set at 100 mV and channel 2 set at 400 mV. The guard cell was set at 450 mV. Chromatographs were analyzed using the TotalChrom chromatography software version 6.2.0 (Perkin Elmer

Instruments LLC). DNA oxidation was expressed as pmols 8-oxodG per ug dG.

Immunoblot analysis

Tissue samples were homogenized in RIPA buffer with protease inhibitors on ice and sonication (30 sec; 4°C). Then the homogenates were incubated for 30 min at 4°C with rocking and centrifuged at 14,000 x g for 10 min at 4°C. The supernatant was quantified for total cellular protein using the Bradford Protein Assay kit (Bio-Rad, Hercules, CA). Aliquots of total protein were dissolved in loading buffer (Blue Loading Buffer Pack, New England Biolabs Ltd,

Pickering, ON, Canada). Subsequently, the samples were boiled for 5 min and run on a 10%

SDS polyacrylamide gel under reducing and denaturing conditions with a protein marker (cat. #

P7708S, New England Biolabs Ltd, Pickering, ON, Canada) to assess their approximate molecular weights. The proteins were then transferred onto nitrocellulose membranes (Bio-Rad,

Hercules, CA) and the membranes were blocked for 1 hr. The membranes were incubated with polyclonal anti-glial fibrillary acidic protein (GFAP) (1:10,000 in blocking solution; Dako,

Carpinteria, CA) or anti-tyrosine hydroxylase (TH) (1:2000 in blocking solution; Chemicon

International Inc., Temecula, CA) or anti-dopamine transporter (DAT) (1:2000 in blocking solution; Sigma-Aldrich, Oakville, ON, Canada) overnight at 4°C with rocking. The secondary 244

antibody was goat anti-rabbit IgG-horseradish peroxidase (HRP) (diluted 1:5000 in blocking solution, Santa Cruz Biotechnologies Inc., Santa Cruz, CA). Peroxidase activity was visualized by an enhanced chemiluminescence (ECL) Plus detection system and proteins were identified as a single band corresponding to their molecular weight (Amersham Biosciences, Piscataway,

NJ). Anti-β-actin antibody-HRP (1:5000 in blocking solution) (Sigma, Saint Louis, MO, USA) was used as a housekeeping protein to standardize amounts of protein loaded in each line.

Optical densities were determined using a computerized image analysis system (AlphaEaseFC,

FluorChem SA for Windows) and densities for all proteins were confirmed to be in the linear range with no saturation. Values were calculated as density of a given protein standardized to density of β-actin and expressed as a percent of control (saline +/+).

Nissl staining and immunohistological analysis of the olfactory bulbs

Animals were deeply anesthetized and perfused transcardially with 40 ml of PBS and 40 ml of 4% paraformaldehyde in PBS, pH 7.2. Brains were postfixed in the perfusing solution overnight at 4°C and then cryoprotected for at least 24 hr in 20% sucrose. The brains were then frozen and coronally cut at the olfactory bulbs (Bregma +4.20 mm) at a thickness of 10 um

(Toronto Centre for Phenogenomics, Pathology Core). Sections mounted on glass slides were

Nissl-stained with cresyl violet to visualize cells bodies of the olfactory bulb. Separate sections were incubated with hydrogen peroxide to quench endogenous peroxidase. Tissue sections were blocked (protein block, X0909, Dako Canada Inc., Mississauga, ON, Canada) followed by an overnight incubation at 4°C with anti-DAT (1:10000, Sigma-Aldrich, Oakville, ON, Canada) or mouse anti-glutamic acid decarboxylase (GAD)-6 primary antibody (1:100). The anti-GAD-6 antibody, developed by Dr. David I. Gottlieb (Washington University School of Medicine, St. 245

Louis, MO) was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by The University of Iowa (Department of Biology,

Iowa City, IA 52242), and was used to visualize the glutamic acid decarboxylase-65 (GAD-65) isoform which identifies GABAergic neurons in brain. For DAT and GAD-6, the secondary antibodies were 1:200 biotinylated goat anti-rabbit IgG or 1:200 biotinylated horse anti-mouse

IgG (Vector Laboratories, Burlington, ON, Canada) respectively. Detection was achieved with the Vectastain Elite ABC-Peroxidase kit with Nova red (Vector Laboratories Burlington, ON,

Canada) according to the manufacturer‟s instructions and were counterstained with hematoxylin. For quantification, immuno-positive cells were counted using 2 coronal sections from each mouse and 12 fields within the granule cell layer (GCL) of the olfactory bulbs were analyzed bilaterally and are averaged per field.

Functional deficits: motor coordination studies

Mice that succeeded in 3 trials (2 hr intervals) without falling were tested. In the testing phase, mice were tested for motor coordination at a constant speed of 5 rpm for 90 s, then the rotarod was gradually increased to a constant speed of 20 rpm for a maximum of 5 min. The performance time and speed at which the mice fell from the rod were recorded.

246

Functional deficits: activity monitoring

Mice were given 60 min to acclimatize to the testing room. Activity was measured using the Linton AM1053 Standard (Dual Layer) X, Y, Z IR Activity Monitor and associated

Amonlite software, version 1.4 (Linton Instrumentation, Norfolk, UK). The monitor consisted of two levels of IR light beams set up such that the lower set of beams measured the activity of mice walking in the bottom of the cage, while the higher set measured the activity of mice standing on their hind legs as „rearing‟ activity. Activity measurements were performed during the afternoon and the animals were left alone in the room while measurements were taken in order to minimize external stimuli that might have interfered with their normal activity. The measure of activity was recorded every 5 min for a total of 60 min, such that 12 measurements were taken from each animal in each experiment. The sum of the distance travelled as well as rearing time (counts every time mice were rearing in a given second) from those 12 measurements was then calculated and analyzed statistically.

Functional deficits: olfactory discrimination (OD) test

OD was assessed using the protocol described in Tillerson and coworkers (Tillerson et al., 2006) with the following modifications. Mice are separated into individual cages for 24 hr prior to the test. Mice were given 60 min to acclimatize before the test. The solutions used were

100 ng/mL of paprika or cinnamon. A Petri dish was placed in the animal‟s home cage. Each mouse was presented with two 25 μl solutions: the first odor (paprika) on one side of the dish and water (control) on the other side. Time spent sniffing both the water and scented solution was recorded. Sniffing was defined when the animal‟s nose was located 1 cm or less from the odor. Five successive 3 min trials were performed, separated by 15 min inter-trial intervals. 247

Habituation response was measured on the first five trials by examining investigation time across trials. On the sixth trial, the mouse was presented with the alternative odor, cinnamon, for 3 min along with the water control. As some mice spent equal time investigating the novel odor and water control, the data were expressed as the ratio of time spent detecting odor to water. Discrimination was defined as the difference in ratios between time spent detecting the familiar odor (fifth trial) and time spent detecting the novel odor on the sixth trial.

Quantification of hyperthermia

Both Nrf2 wild-type and Nrf2 KO mice were dosed with METH. METH was dissolved in 0.9% saline and administered in a 2-day dosing regimen using a single 10 mg/kg i.p. dose on

Day 1, followed 24 hr later on Day 2 by four 5 mg/kg i.p. doses of METH, with a 2-hr interval between each dose (total dose of 30 mg/kg). Saline vehicle was used as the control.

Immediately before and at 1, 3, 5 and 7 hr after the second METH injection, rectal core temperature was recorded using a precision digital thermometer (Precision Thermometer 4610,

YSI Inc, Ohio, USA). The probe was inserted into the rectum and maintained until the temperature reading had stabilized. Following injections, mice were maintained in an animal room at 21-22oC.

Quantification of serum levels of METH and amphetamine (AMPH)

Both Nrf2 wild-type and Nrf2 KO mice were dosed with METH in a 2-day dosing regimen using a single 10 mg/kg i.p. dose on Day 1, followed 24 hr later on Day 2 by a single

10 mg/kg i.p. dose. Serum concentrations of METH and its major active metabolite amphetamine (AMPH) were measured 30 min, 1 hr and 3 hr after the Day 2 dose. The 248

procedure involved solvent extraction followed by HPLC/MS/MS with deuterated analytes as internal standards (Clinical Laboratory, Centre for Addiction and Mental Health, Toronto, ON,

Canada). For 40-50 ul of mouse serum, the limit of detection was 1 ng/ml.

Statistical analysis

Multiple comparisons among groups were analyzed by one-way ANOVA followed by a

Tukey‟s test (GraphPad InStat®3.05, GraphPad Software, Inc., San Rafael, CA, USA). The level of significance was determined to be at P < 0.05. 249

2.4.4 RESULTS Analysis of mRNA levels by quantitative RT-PCR

Nrf2 wild-type and knockout mice were administered one dose of METH (10 mg/kg i.p.) or saline to determine if the Nrf2-regulated genes heme oxygenase-1 HO-1, NQO1 and OGG1 are modulated. In the striatum of Nrf2 wild-type mice, HO-1 mRNA levels were increased with

METH treatment by maximal 150% within 6-12 hr after dosing, decreasing thereafter but still significantly elevated at 48 hr (p<0.01) (Figure 1). In METH-treated Nrf2 KO mice, HO-1 mRNA was induced only moderately (p<0.05), returning to basal levels within 24 hr. In the striatum of Nrf2 wild-type mice, METH induced NQO1 maximally at 12 hr with an increase of

40% (p<0.01), whereas NQO1 in Nrf2 KO mice remained at basal levels. In the striatum of

Nrf2 wild-type mice, METH induced OGG1 within 6 hr, whereas OGG1 levels were conversely decreased initially in Nrf2 KO mice with a subsequent return to basal levels, followed by induction 48 hr after dosing (p<0.05).

In the olfactory bulbs, HO-1 mRNA was induced by 70% within 6 hr in the Nrf2 wild- type mice (p<0.01) and remained elevated 24 hr later, whereas HO-1 remained at basal levels in the Nrf2 KO mice (Figure 1). In the olfactory bulbs of Nrf2 wild-type mice, METH induced

NQO1 maximally at 12 hr with an increase of 60% (p<0.01), whereas NQO1 in Nrf2 KO mice remained at basal levels. In the olfactory bulbs, there was no induction of OGG1 mRNA in either of Nrf2 wild-type or KO mice within 12 hr. After 24 hr in the Nrf2 KO mice, there was an 18 % increase in OGG1 levels (p<0.05). Therefore, a single dose of METH can induce Nrf2- regulated antioxidant and cytoprotective proteins in mouse brain in a tissue-specific fashion.

DNA oxidation studies 250

In the acute dosing regimen, after 1 hr, there was no increase in DNA oxidation in either saline- and METH-treated mice of either genotype in the striatum (Figure 2a). After 3 hr,

METH enhanced DNA oxidation by up to 2-fold in the striatum of Nrf2 wild-type and Nrf2 KO mice when compared to saline controls (p<0.01) (Figure 2a). By 6 hr, DNA oxidation levels were decreased in METH-treated mice compared to the 3 hr time-point but still significantly higher than those of saline-treated mice (p<0.05) (Figure 2a). However, there was no difference in DNA oxidation between METH-treated Nrf2 wild-type and Nrf2 KO mice with the acute dosing regimen or in basal levels.

With the 2-day dosing regimen, within 3 hr METH enhanced DNA oxidation in the striatum of both Nrf2 wild-type and Nrf2 KO mice compared to saline controls (Figure 2b).

METH-treated Nrf2 wild-type mice had a 50% increase in DNA oxidation compared to saline- treated Nrf2 wild-type mice (p<0.05). METH-treated Nrf2 KO mice had a 160% increase in

DNA oxidation compared to saline-treated Nrf2 wild-type mice (p<0.01). In contrast to the acute dosing regimen, with the 2-day dosing regimen there was a 60% higher increase in the

METH-treated Nrf2 KO mice compared to Nrf2 wild-type mice at 3 hr (p<0.05), and this difference persisted at the 6 hr time-point (Figure 2b).

Striatum TH, DAT and GFAP immunoblot studies

To investigate the role of Nrf2 in METH-initiated neurotoxicity, Nrf2 wild-type or Nrf2

KO mice were administered an acute dosing regimen of four 10 mg/kg i.p. doses of METH

(total dose of 40 mg/kg), with a 2-hr interval between each dose, and the mice were euthanized

3 days after the last dose. The striatum, a major site of METH-initiated neurotoxicity, was analyzed by immunoblot for TH, the rate- limiting enzyme in catecholamine synthesis which is 251

found in the striatum. Decreases in the levels of TH in the striatum may indicate neurotoxicity

(O'Callaghan and Miller, 1994). METH treatment decreased the relative amount of TH staining in the striatum of Nrf2 wild-type and Nrf2 knockout mice by an average of 50% when compared to saline-treated wild-type controls (p<0.05) (Figure 3a). Immunoblot analysis of GFAP to specifically detect astrogliosis, which is a feature of astrocytes adjacent to neuronal injury

(O'Callaghan and Miller, 1994) revealed the striatum had twice as much GFAP in METH- treated mice compared to saline controls (p<0.01) (Figure 3b). However, there was no difference in the relative amount of TH or GFAP staining between METH-treated Nrf2 wild- type and Nrf2 KO mice at this dose and time (Figure 3a and b).

We next evaluated a 2-day dosing regimen using a single 10 mg/kg, i.p. dose of METH on Day 1 to activate Nrf2-regulated protective pathways, followed 24 hr later on Day 2 by four

5 mg/kg i.p. doses of METH, with a 2-hr interval between each dose (total dose of 30 mg/kg). 1 wk after the last dose, the striatum was analyzed for TH and DAT. METH-treated Nrf2 wild- type and KO mice had respective 45% and 66% decreases in TH levels compared to saline- treated wild-type mice (p<0.05) (Figure 3c). The METH-treated Nrf2 KO mice showed enhanced susceptibility with a 38% decrease in TH compared to METH-treated Nrf2 wild-type mice (p<0.05). DAT immunoblots showed a similar trend (Figure 3e). METH-treated Nrf2 wild-type and KO mice had respective 34% and 56% decreases in DAT levels compared to saline-treated wild-type mice (p<0.05). The METH-treated Nrf2 KO mice showed enhanced susceptibility with a 30% decrease in DAT compared to METH-treated Nrf2 wild-type mice

(p<0.05).

In addition, striatal GFAP levels were increased 35% and 65% respectively in METH- treated Nrf2 wild-type and Nrf2 KO mice compared to saline-treated wild-type mice (p<0.05) 252

(Figure 3d). The METH-treated Nrf2 KO mice showed enhanced susceptibility with a 22% increase in GFAP compared to METH-treated Nrf2 wild-type mice (p<0.05). These results show that Nrf2-mediated pathways provide neuroprotection in the striatum against the cellular neurotoxic effects of multiple exposures to METH with respect to nerve terminal markers and glial response.

Olfactory bulbs TH and GFAP immunoblot studies

METH administration causes dopaminergic neuronal death within the olfactory bulb

(Deng et al., 2007), a region rich in dopamine and regulatory GABA interneurons (Gheusi et al.,

2000; Parrish-Aungst et al., 2007). These parameters were evaluated in Nrf2 wild-type and KO mice with the 2-day dosing regimen using a single 10 mg/kg i.p. dose of METH on Day 1, followed 24 hr later on Day 2 by four 5 mg/kg i.p. doses of METH, with a 2-hr interval between each dose (total dose of 30 mg/kg). Immunoblots of the olfactory bulbs revealed that GFAP was increased 133% in METH-treated wild-type and Nrf2 KO mice compared to saline controls

7 days after dosing (p<0.05) (Figure 4a). Contrary to the striatum, TH levels in the olfactory bulbs remained unchanged in METH-treated mice compared to saline controls (Figure 4b).

To determine more subtle effects, a single 10 mg/kg METH dose was given on Day 1, but only a single 10 mg/kg METH dose was given on Day 2 (total dose of 20 mg/kg).

Immunoblots of the olfactory bulbs 7 days after this Low-dose 2-day regimen revealed an average of 40% increase in GFAP in the METH-treated Nrf2 KO mice compared to both

METH-treated wild-type and saline controls (p<0.05) (Figure 4c). TH levels in the olfactory bulbs were not significantly altered in METH-treated wild-type and KO mice compared to saline controls (Figure 4d). 253

Nissl staining and Immunohistological Analysis of the Olfactory bulbs

Using the Low-dose 2-day dosing regimen, histological sections of the olfactory bulbs of

METH-treated Nrf2 wild-type and KO mice were further characterized to determine other potential neurotoxic effects that may be modulated by Nrf2. Nissl staining of cell bodies in the olfactory bulb showed the characteristic olfactory bulb layers with no microscopic differences between METH-treated Nrf2 wild-type and KO mice or saline-treated mice (Figure 5a). As with TH immunoblots, immunohistological analysis of DAT in the olfactory bulb did not indicate any significant changes in METH-treated wild-type and KO mice compared to saline controls (Figure 5a). Considering the abundance of GABAergic neurons in the olfactory bulbs

(Parrish-Aungst et al., 2007), we quantified GAD-65 positive cells in the GCL of the olfactory bulbs (Figure 5b). METH-treated Nrf2 KO mice had a 37% decrease in GAD-65 immunoreactive cells compared to METH-treated Nrf2 wild-type or saline treated mice (p<0.05)

(Figure 5b), suggesting a loss of GABAergic nerve terminal markers in Nrf2 KO mice.

Functional deficits: motor coordination and activity monitoring

To investigate the role of Nrf2 in modulating METH-initiated functional deficits, we tested a Low-dose 2-day dosing regimen using the same 10 mg/kg METH dose on Day 1, but only a single 10 mg/kg METH dose on Day 2. However, this Low-dose regimen did not lead to significant motor coordination deficits in any of the groups (data not shown). A 4-day dosing regimen of 10 mg/kg each day was then evaluated (Figure 6). A significant trend toward impairment in motor coordination was observed in the rotarod test in METH-treated Nrf2 KO mice by 4 wk of testing (Figures 6a and b). After 8 wk, activity monitoring of the distance travelled in a 60 min trial or rearing time, did not show significant differences between saline 254

and METH-treated mice or between the different genotypes tested (Figures 6c and d). Also, no gender dependent differences were detected so these data were pooled.

Functional deficits: olfactory discrimination (OD) test

OD was investigated with a High-dose 2-day dosing regimen using a single 10 mg/kg i.p. dose of METH on Day 1 to activate Nrf2-regulated protective pathways, followed 24 hr later on Day 2 by four 5 mg/kg doses of METH, with a 2-hr interval between each dose (total dose of

30 mg/kg). Also employed was a Low-dose 2-day dosing regimen using the same single 10 mg/kg i.p. dose of METH on Day 1, but only a single 10 mg/kg dose of METH on Day 2 (total dose of 20 mg/kg). High-dose and low dose data showed the ability of mice to detect and habituate to the first odor (paprika) (Figure 7a). During the discrimination phase, time spent investigating the novel odor was found for the saline group of both wild-type and KO mice

(p<0.01) (Figures 7b and c). In contrast to the saline mice, High-dose METH-treated mice did not significantly increase their investigation time of the novel odor. This OD deficit was observed within 7 days after the last METH treatment (Figure 7b) and continued for up to 4 weeks (Figure 8). Furthermore, both Nrf2 wild-type and KO animals showed this deficit with no significant differences between the groups.

To determine more subtle effects, we further investigated OD deficits with the Low-dose

2-day dosing regimen (Figure 7c). Only the METH-treated Nrf2 KO mice exhibited the olfactory discrimination deficit within 7 days of treatment (p<0.05) (Figure 7c), and this deficit persisted for 1 month thereafter (Figure 8). This effect was not due to an inability to perform the task, as both saline and METH animals explored odors for approximately the same amount of time during the detection and habituation phases of the trials. 255

Hyperthermia in Nrf2 wild-type and Nrf2 KO mice

Nrf2 wild-type and KO mice were dosed on Day 1 with 10 mg/kg i.p. dose of METH to induce Nrf2-regulated antioxidant genes, followed 24 hr later on Day 2 with four 5 mg/kg i.p. doses of METH (total dose of 30 mg/kg) or saline, with a 2-hr interval between each treatment.

METH-treated mice showed significant elevations in core temperature, from 37.5oC to 40oC, within 1 hr of treatment compared to saline controls (p<0.001), and temperatures remained elevated for 5 hr (Figure 9a). However, there were no differences between Nrf2 wild-type and

KO mice, indicating that body temperature did not contribute to the Nrf2-mediated protection from METH-initiated neurodegeneration (Figure 9a).

Quantification of serum levels of METH and amphetamine (AMPH)

METH was administered to Nrf2 wild-type and Nrf2 KO mice with a 2-day dosing regimen using a single 10 mg/kg i.p. dose on Day 1 followed 24 hr later on Day 2 by a single 10 mg/kg i.p. dose. There was no difference in the peak of METH serum levels between wild-type and KO mice 30 min to 1 hr after METH administration (Figure 9b). By 3 hr, METH was reduced by a similar extent in both Nrf2 wild-type and KO mice (Figure 9b). AMPH levels were increased similarly in the Nrf2 wild-type and KO METH-treated mice (Figure 9c). There were no significant differences between Nrf2 wild-type and KO mice. 256

Figure 1. Effect of methamphetamine (METH) on Nrf2-mediated gene expression in Nrf2 wild-type and knockout mice. Mice were administered one dose of METH (10 mg/kg i.p.) or saline to determine if the Nrf2-regulated genes heme oxygenase-1 (HO-1), NAD(P)H:quinone oxidoreductase (NQO1) and oxoguanine glycosylase 1 (OGG1) are modulated in the striatum and olfactory bulbs as assessed by RT-PCR analysis of mRNA levels. Time-points included 6, 12, 24 and 48 hr after dosing. mRNA levels were standardized using B-actin. a=p<0.01 compared to saline-treated wild-type mice. b=p<0.01 compared to saline-treated Nrf2 KO mice. c=p<0.05 compared to saline-treated wild-type mice. * =p<0.01 compared to METH-treated KO mice. For METH and saline-treated groups, 4-6 mice were used per time-point and genotype. 257

Figure 2. DNA oxidation in METH-treated Nrf2 wild-type and knockout mice. Nrf2 wild- type or Nrf2 KO mice were administered either: (a) an acute, single-day dosing regimen of four 10 mg/kg i.p. doses of METH (total dose of 40 mg/kg), with a 2-hr interval between each dose; or, (b) a 2-day dosing regimen with a single 10 mg/kg i.p. dose of METH on Day 1 to activate Nrf2-regulated protective pathways, followed 24 hr later on Day 2 with four 5 mg/kg i.p. doses of METH, with a 2-hr interval between each dose (total dose of 30 mg/kg). The mice were euthanized at 1, 3 or 6 hr after the last dose, and the striatum was isolated. The control vehicle was 0.9% saline. Samples were analyzed for oxidatively damaged DNA reflected by 8-oxo-2- deoxyguanosine (8-oxodG) standardized using dG levels. a = p<0.01 compared to saline-treated Nrf2 wild-type mice, b = p<0.05 compared to saline-treated Nrf2 wild-type mice. *=p<0.05 compared to METH-treated Nrf2 wild-type mice. For METH and saline-treated groups, 4-6 mice were used per time-point and genotype. 258

259

Figure 3. Tyrosine hydroxylase (TH), dopamine transporter (DAT) and glial fibrillary acidic protein (GFAP) immunoblots and densitometric analysis of the striatum in METH- treated Nrf2 wild-type and KO mice. In the acute dosing regimen, densitometric analysis of striatal (a) TH and (b) GFAP in Nrf2 wild-type or Nrf2 KO mice 3 d after the last of 4 doses of saline or METH (10 mg/kg i.p. x 4). For the 2-day dosing regimen, a single 10 mg/kg i.p. dose of METH was given on Day 1 to activate Nrf2-regulated protective pathways, followed 24 hr later on Day 2 by four 5 mg/kg i.p. doses of METH, with a 2-hr interval between each dose (total dose of 30 mg/kg). The mice were euthanized 1 wk after the last dose and the striatum was analyzed for (c) TH, (d) GFAP and (e) DAT. β-actin was used to standardize protein levels. a = p<0.05 compared to saline-treated Nrf2 wild-type mice. b = p<0.01 compared to saline-treated Nrf2 wild-type mice. *=p<0.05 compared to METH-treated Nrf2 wild-type mice. Each bar represents results of densitometric analysis of immunoblots from 4-6 mice. The gels for each dose and endpoint were imaged and exposed at the same time. 260

Figure 4. Tyrosine hydroxylase (TH) and glial fibrillary acidic protein (GFAP) immunoblots and densitometric analysis of the olfactory bulbs in METH-treated Nrf2 wild-type and KO mice. For the High-Dose 2-day dosing regimen, a single 10 mg/kg i.p. dose of METH was given on Day 1 to activate Nrf2-regulated protective pathways, followed 24 hr later on Day 2 by four 5 mg/kg i.p. doses of METH, with a 2-hr interval between each dose (total dose of 30 mg/kg). The mice were euthanized 1 wk after the last dose and the olfactory bulbs were analyzed for (a) GFAP and (b) TH. For the Low-Dose 2-day dosing regimen, the same single 10 mg/kg i.p. dose of METH was given on Day 1, but on Day 2 only a 10 mg/kg dose of METH was administered, (total dose of 20 mg/kg). The mice were euthanized 1 wk after the last dose and the olfactory bulbs were analyzed for (c) GFAP and (d) TH. β-actin was used to standardize protein levels. a = p<0.05 compared to saline-treated Nrf2 wild-type mice. * = p<0.05 compared to METH-treated Nrf2 wild-type mice. Each bar represents results of densitometric analysis of immunoblots from 4-6 mice. The gels for each dose and endpoint were imaged and exposed at the same time. 261

Figure 5. Nissl, DAT and GAD-65 analysis of olfactory bulb in saline and METH-treated Nrf2 wild-type and KO mice. The Low-Dose 2-day dosing regimen used a single 10 mg/kg i.p. dose of METH on Day 1, and a single 10 mg/kg dose of METH on Day 2 (total dose of 20 mg/kg). The mice were euthanized 1 wk after the last dose and the olfactory bulbs were analyzed. (a) Nissl (200X), DAT (200X) and GAD-65 (200X) staining of coronal sections through the olfactory bulb. (b) Quantification of GAD-65 immuno-positive cells in the granule cell layer. a = p<0.05 compared to saline-treated Nrf2 wild-type mice. b = p<0.05 compared to METH-treated Nrf2 wild-type mice. Immunohistological staining is representative of 4 mice for each group. 262

Figure 6. METH-initiated functional deficits in motor coordination and activity in Nrf2 wild-type and KO mice. A 4-day dosing regimen employed a single 10 mg/kg i.p. dose of METH each day for 4 days (total dose of 40 mg/kg). Motor coordination was assessed by the rotarod test (a) 1 wk and (b) 4 wk after the last dose. Latency to fall was measured in seconds. * = p<0.1 compared to METH-treated Nrf2 wild-type mice. The activity of these mice was also assessed 8 wk after the last dose. (c) Total distance travelled and (d) Rearing time were collected by the activity monitor over a 60 min period. 7-10 mice were used for each group. 263

Figure 7. Olfactory detection, habituation and discrimination in METH-treated Nrf2 wild-type and KO mice. For the High-Dose 2-day dosing regimen, a single 10 mg/kg i.p. dose of METH was given on Day 1 to activate Nrf2-regulated protective pathways, followed 24 hr later on Day 2 by four 5 mg/kg i.p. doses of METH, with a 2-hr interval between each dose (total dose of 30 mg/kg). For the Low-Dose 2-day dosing regimen the same single 10 mg/kg i.p. dose of METH was given on Day 1, but a lower single 10 mg/kg i.p. dose of METH was given on Day 2 (total dose of 20 mg/kg). (a) High-dose data shown where mice were able to detect and habituate to the first odor (paprika) after 4 trials. Detection =p<0.01 compared to respective Trial 1 water groups. (b) High-dose data and (c) Low-dose data for the discrimination phase of the test. On the sixth trial, mice were presented with the novel odor (cinnamon) for 3 min compared with the water control. The data are expressed as the ratio of time spent detecting odor to water. Discrimination was defined as the difference in ratios between time spent detecting the familiar odor (fifth trial) and time spent detecting the novel odor on the sixth trial (p<0.01). No discrimination=p<0.05 compared to METH-treated wild-type mice. Each group consisted of 10 mice that were tested 1 wk after dosing. 264

Figure 8. Olfactory discrimination in METH-treated mice 4 wk after dosing. High-dose data and Low-dose data for discrimination was defined as the difference in ratios between time spent detecting the familiar odor (fifth trial) and time spent detecting the novel odor on the sixth trial (p<0.01). No discrimination=p<0.05 compared to METH-treated wild-type mice. Each group consisted of 10 mice that were tested 4 wk after dosing. 265

Figure 9. Quantification of hyperthermia, methamphetamine (METH) and amphetamine (AMPH) in METH-treated Nrf2 wild-type and KO mice. (a) METH was administered in a 2-day dosing regimen, with a single 10 mg/kg i.p. dose given on Day 1, followed 24 hr later on Day 2 by four 5 mg/kg i.p. doses of METH (total of 30 mg/kg), with a 2-hr interval between each dose. Rectal core temperature was recorded immediately before and at 1, 3, 5 and 7 hr after the second METH injection. a= p<0.001 compared to saline-treated wild-type mice. 4 mice were used for each group. (b and c) For determination of serum drug concentrations, METH was administered in a 2-day dosing regimen using a single 10 mg/kg i.p. dose on Day 1, followed 24 hr later on Day 2 by a single 10 mg/kg i.p. dose, after which serum concentrations of METH (b) or its major metabolite amphetamine (AMPH) (c) were measured 30 min, 1 hr and 3 hr after dosing. A minimum of 3 mice were used for each time-point. 266

2.4.5 DISCUSSION

Nrf2 has been shown to be protective against benzo[a]pyrene carcinogenicity and the neurotoxicity of 3-nitropropionic acid and kainic acid (Kraft et al., 2006; Calkins et al., 2005;

Ramos-Gomez et al., 2001). Also, aged Nrf2 knockout animals have been shown to develop vacuolar leukoencephalopathy and astrogliosis (Hubbs et al., 2007). We evaluated the role of

Nrf2 in ROS-mediated neurotoxicity and functional deficits initiated by either acute “binge” or multiple-day dosing regimens with METH. Using RT-PCR analysis, we determined that a single dose of METH (10 mg/kg) can activate Nrf2 as indicated by increased levels of mRNA of

Nrf2-regulated cytoprotective genes in Nrf2 wild-type but not KO mouse brains. The temporal profile and magnitude of enhanced transcription varied with both the gene and brain region analyzed. In Nrf2 wild-type mice, the striatum, for example, had robust changes in HO-1 whereas the olfactory bulb had a large and more prolonged transcription of NQO1. These differences highlight the potential for regional susceptibility to neurotoxins modulated by selective Nrf2 activation. A similar differential activation of Nrf2 regulated-genes in the striatum versus substantia nigra has been reported for MPTP (Chen et al., 2009).

Our study confirms that METH can initiate ROS formation as indicated by enhanced oxidatively damaged DNA in the striatum. With our acute single-day dosing regimen, presumably the synthesis of proteins controlled by Nrf2 activation is not sufficiently rapid, given the relatively short half-life of METH and its rapid initiation of oxidative macromolecular damage, to provide protection against METH-initiated DNA oxidation and/or to enhance its repair. Consistent with this theory, our mRNA analysis of OGG1, involved in DNA repair, showed enhanced levels 6 to 12 hours after a 10 mg/kg dose of METH, and the synthesis of new

OGG1 protein would follow this temporal profile, too late to protect against the DNA oxidation 267

that peaks around 3 hours. In contrast to the single-day treatment regimen, with the 2-day regimen, the second day‟s administration of METH coincided with a maximal induction of antioxidative and cytoprotective proteins that protected the Nrf2 wild-type mice, but not the

Nrf2 KO mice, the latter of which exhibited no Nrf2-dependent enhanced transcription.

METH-initiated decreases in the levels of TH and DAT in the striatum of mice, and converse increases in striatal GFAP, have both been correlated with indices of long-term neurotoxicity as assessed by quantification of silver degeneration staining (O'Callaghan and

Miller, 1994). With our low-dose 2-day METH dosing regimen, the Nrf2 KO mice were more prone to neurotoxicity as illustrated by the changes in these markers. On the other hand, with the acute single-day dosing regimen, neurotoxicity was enhanced with METH exposure, however there were no differences between the METH-treated wild-type and KO mice due to inadequate time for the induction of protective proteins in the Nrf2 wild-type mice. Taken together these results showed in the striatum that prior activation of Nrf2 can have a significant protective effect on METH-initiated neurotoxic outcomes that correlated with neuronal DNA damage. To assess the potential confounding effect of Nrf2-dependent hepatic metabolism of

METH, we measured serum concentrations of METH and its major amphetamine metabolite.

No differences were found between Nrf2 wild-type and KO mice, which ruled out altered

METH metabolism in the enhanced susceptibility of Nrf2 KO mice to METH neurotoxicity.

These concentrations are similar to plasma concentrations of METH in humans after chronic use in the range 0.176- 1.743 mg/L (176-1743 ng/mL) (Wilson et al., 1996), however concentrations in rats are usually on average 10X higher in the brain versus plasma (Melega et al., 1995). Also, the hyperthermic response of the Nrf2 wild-type and KO mice to METH were virtually identical, indicating that this potential modulator of METH-initiated neurotoxicity was not 268

contributing to the vulnerability of the Nrf2 KO mice.

METH-initiated Nrf2 activation and neurotoxicity are not limited to the striatal region.

Previous studies have shown that a single 40 mg/kg dose of METH causes dopaminergic neuronal death within the olfactory bulb of mice (Deng et al., 2007). Our mRNA analysis of the olfactory bulb revealed that Nrf2 can activate cytoprotective genes in this region following a

METH dose of only 10 mg/kg. In fact, in METH-treated mice, GFAP levels increased about 2- fold more in the olfactory bulb (133%) compared to the striatum (66%), indicating that the olfactory bulb is highly susceptible to glial activation by METH. However, in the olfactory bulb, there was no difference between the METH-treated Nrf2 wild-type and KO mice with the high-dose 2-day dosing regimen. However, when the day 2 METH dose was reduced from 20 to 10 mg/kg (low-dose regimen), Nrf2 was protective, indicating that sufficiently high doses of

METH can overwhelm Nrf2-dependent cytoprotective pathways.

The olfactory bulb is the major structure within the brain responsible for olfactory processing and is involved with certain forms of learning and memory (Brennan and Keverne,

1997; Gheusi et al., 2000). Proper function in these underlying pathways is important for odor detection, habituation and discrimination. In humans, olfactory discrimination (OD) declines with age (Kaneda et al., 2000), and olfactory deficits have been shown to be an early marker of various neurodegenerative diseases, for example preceding motor deficits in Parkinson‟s disease by years (Doty et al., 1988; Ross et al., 2008; Talamo et al., 1989; Larsson et al., 2006). In our study, with the high-dose 2-day dosing regimen, animals treated with METH were unable to discriminate between a previously habituated odor and a novel odor as early as 7 days and up to

1 month after treatment. This constitutes the first evidence of METH-initiated OD deficits in mice. In mice, OD deficits are evident in DAT and D2 receptor KO mice (Tillerson et al., 269

2006). However, contrary to this report and to our study of the striatum, we did not observe any

METH-initiated changes in TH levels indicative of dopaminergic neurotoxicity in the olfactory bulb. This may be due to the lower doses used in our studies compared to previous studies where a single high dose decreased TH levels in the olfactory bulb (Deng et al., 2007). We lowered our METH doses even further where the METH-treated Nrf2 KO mice showed enhanced susceptibility to neurotoxic glial activation in the olfactory bulb and OD deficits, while the wild-type controls were normal. Again, there were no changes in TH or DAT levels in METH-treated mice of either Nrf2 genotype. Consistent with studies in DAT KO mice with

OD deficits (Tillerson et al., 2006), analysis of Nissl staining of the olfactory bulb in METH- treated mice with OD deficits demonstrated that the various layers were well organized as they are in saline-treated mice.

In the olfactory bulb, another molecular target of interest is the GABAergic system.

Studies with METH-treated rats have shown that METH can decrease striatal GABAergic markers within a week post-treatment (Jayanthi et al., 2004; Burrows and Meshul, 1999). Our studies, with doses lower than those used in previous studies, demonstrated decreased levels of

GAD-65 in the GCL of the olfactory bulb in METH-treated Nrf2 KO mice consistent with the enhanced OD deficits in these mice. A number of studies have linked OD and olfactory function to the GABAergic systems within the olfactory bulb, where aging and/or decreased neurogenesis led to decreased new GCL GABA neurons, which correlated with OD deficits

(Abraham et al., 2010; Enwere et al., 2004; Gheusi et al., 2000). It is not clear whether METH is directly causing degeneration of GABAergic neurons, or if changes in GAD-65 are due to deregulation of olfactory bulb network. It is also possible that METH-initiated changes in dopamine levels at the synapse, as opposed to altered enzyme levels or transporter levels, may 270

contribute to compensatory changes in the GABAergic system of the olfactory bulb, thereby contributing to the OD deficits observed. Interestingly, with the low-dose 2-day regimen,

METH did not measurably enhance motor coordination functional deficits as assessed by rotarod performance, suggesting that OD constitutes a more sensitive measure of METH- initiated neurodegeneration under these conditions and may be a relevant and sensitive test to assess early human functional deficits. With chronic 4-day dosing, however, METH-treated

Nrf2 KO mice developed a sensitivity to motor coordination deficits that was not reflected in activity monitoring or rearing experiments, which measure spontaneous behaviors as opposed to the rotarod which measures motor functions requiring more dexterity and fine motor skill.

It is evident from our studies that repeated dosing with METH in the context of reduced

Nrf2 activation can enhance susceptibility to functional deficits. This mechanism may be important in human users of METH where differences in drug abuse patterns may alter the induction and duration of Nrf2 activation thereby modulating susceptibility to the neurotoxic effects of METH. Aside from drug abusers, the Nrf2-dependent protective effects observed herein for METH could be relevant for children chronically taking amphetamine and structurally related drugs therapeutically for attention deficit hyperactivity syndrome and related conditions, although neurodegenerative complications from therapeutic use have not been reported. Single nucleotide polymorphisms have been identified within the promoter region of the human Nrf2 gene (Yamamoto et al., 2004) and genetic variation here along with polymorphisms in the ARE region of Nrf2 target gene (Wang et al., 2007a) may cause variability in Nrf2 activity and hence the response to oxidative stress and neurotoxicity. In addition, there are also dietary inducers of Nrf2-regulated genes such as sulforaphane, a compound generated from cruciferous vegetables such as broccoli, as well as tert-butylhydroquinone (tBHQ), a metabolite of the food 271

preservative butylated hydroxyanisole (BHA), that may modulate risk.

In our study, DNA oxidation, dopaminergic and GABAergic neurotoxicity, neurotoxic glial activation and functional OD and motor deficits demonstrated regional brain susceptibilities to METH-initiated oxidative stress that was modulated by Nrf2 loss. This is the first evidence of METH-initiated deficits in OD, which appear to constitute a more sensitive functional biomarker of long-term METH neurodegeneration in mice than standard tests currently employed. In our acute dosing regimen, the response of genes and subsequent protein induction controlled by the Nrf2-induced ARE appears to too slow to provide protection against

METH neurotoxicity. In contrast, Nrf2-mediated protection was observed with our 2-day

METH dosing regimen, in which a day 1 dose was given to initiate the expression of Nrf2- regulated genes, followed 24 hours later by a day 2 dose, to allow for the Nrf2-mediated induction of proteins involved in antioxidative pathways, DNA repair and other potentially protective pathways. These pathways acted in concert to protect against molecular and behavioral toxicities associated with the day 2 dose of METH in wild-type but not Nrf2 KO mice, indicating a neuroprotective role for Nrf2 during chronic METH abuse. 272

2.5 STUDY 5: DEVELOPMENTAL ROLE OF NUCLEAR FACTOR-E2-RELATED FACTOR 2 (NRF2) IN PROTECTION AGAINST METHAMPHETAMINE FETAL TOXICITY AND POSTNATAL FUNCTIONAL DEFICITS IN NRF2-DEFICIENT MICEa

Annmarie Ramkissoon and Peter G. Wells

a. Preliminary reports of this research were presented at the annual meeting of the Teratology Society (U.S.A.) (Birth Defects Research Part A: Clinical and Molecular Teratology 85: 449 (No. P67), 2009). This research was supported by a grant from the Canadian Institutes of Health Research (CIHR). AR was supported by a doctoral scholarship from CIHR and the Rx&D Health Research Foundation. 273

2.5.1 ABSTRACT

Nrf2 mediates protective responses to oxidative stress, but its developmental role is unknown. Herein, we treated pregnant Nrf2-deficient knockout mice with methamphetamine

(METH), which increases fetal reactive oxygen species (ROS) and oxidatively damaged DNA.

METH-exposed -/- Nrf2-deficient fetuses were unable to increase transcription of the ROS- protective genes heme oxygenase-1, NAD(P)H:quinone oxidoreductase or oxoguanine glycosylase 1, unlike wild-type controls, and exhibited enhanced DNA oxidation, fetal resorptions, a red phenotype with edema, and reduced fetal weight, with greater toxicity in female -/- Nrf2 fetuses. Postnatal neurodevelopmental deficits in activity and olfactory function were similarly enhanced, and the olfactory bulb GABAergic marker GAD-65 decreased, in -/-

Nrf2-deficient offspring exposed in utero to METH. This is the first evidence for in utero

METH-initiated olfactory deficits, which may be a sensitive postnatal functional biomarker for long-term neurotoxicity. During development, Nrf2 plays an essential role in protecting the fetus from increased oxidative stress. 274

2.5.2 INTRODUCTION

The fetal effects of methamphetamine (METH, Speed) abuse by pregnant women is of increasing concern (Arria et al., 2006). In both humans and rodents, effects of in utero exposure to METH is associated with fetal developmental toxicities including reduced birth weight and cerebral hemorrhage (Plessinger, 1998; Smith et al., 2006), and postnatal neurobehavioral and functional deficits (Acuff-Smith et al., 1996; Chang et al., 2004; Cho et al., 1991; Jeng et al.,

2005; Slamberova et al., 2006; Wong et al., 2008). These effects are inconsistent among studies, possibly due to differences in dosage, gestational timing and duration of exposure, as well as genetic differences that modulate toxicological susceptibility.

METH causes a rapid release of neurotransmitters from synaptic vesicles, inducing euphoria and hallucinations (Kalant, 2001). Long-term effects following adult exposures in mice can lead to nerve terminal degeneration in brain regions such as the striatum, cortex, hippocampus and olfactory bulb (O'Callaghan and Miller, 1994; Deng et al., 2007; Jeng et al.,

2006; Jeng and Wells, 2010). In the developing fetus, the mechanism of METH toxicity is uncertain, although neurotransmitter release in the developing brain can modulate proliferation, migration and differentiation of cells, and may cause permanent deficits in the fetal central nervous system (Levitt et al., 1997; Nguyen et al., 2001). We have shown that METH-initiated reactive oxygen species (ROS) and oxidative DNA damage in the fetal brain can enhance motor coordination deficits (Jeng et al., 2005; Wong et al., 2008). ROS-mediated damage, if not repaired, can accumulate over time and can lead to loss of cellular function and even cell death

(Wells et al., 2009). However, ROS can also activate redox-sensitive transcription factors such nuclear factor erythroid 2-related factor 2 (Nrf2), as commonly occurs in the adult brain during various types of oxidative stress (Chan et al., 2001; Burton et al., 2006; Calkins et al., 2005). 275

Under normal conditions, Nrf2 is located in the cytoplasm as an inactive form associated with its repressor protein Kelch-like ECH Associating Protein 1 (Keap1) (Itoh et al., 1999b).

Oxidation of redox-sensitive cysteines in Keap1 during oxidative stress allows Nrf2 to dissociate from Keap1 and translocate to the nucleus, where it heterodimerizes with other proteins and binds to the antioxidant response element (ARE), an enhancer sequence that regulates the transcription of cytoprotective enzymes such as glutathione S-transferase (GST), heme oxygenase-1 (HO-1) and NAD(P)H:quinone oxidoreductase (NQO1) (Itoh et al., 1999b;

Chan et al., 2001). Similarly, Nrf2 can also lead to the transcription of a variety of other ROS- related protective genes such as the multidrug resistance-associated protein 1, catalase, superoxide dismutase, glucose-6-phosphate dehydrogenase (G6PD) and oxoguanine glycosylase

1 (OGG1) (Dhenaut et al., 2001; Hayashi et al., 2003; Lee et al., 2003b; Shih et al., 2003).

The viability and absence of a phenotype in Nrf2-deficient knockout mice has suggested that Nrf2 plays no role in normal mouse development (Chan et al., 1996). However, evidence from mice genetically deficient in ROS-protective enzymes like G6PD indicates that even normal developmental oxidative stress can be embryopathic suggesting that fetal Nrf2 may play an important protective role. Nrf2 mRNA and protein are expressed during the period of organogenesis (Chan et al., 1996), so Nrf2 may be activated and modulate fetal toxicity during in utero exposure to ROS-initiating xenobiotics like METH, which in adult exposures increases the expression of Nrf2-regulated genes in mouse brain (Ramkissoon and Wells, 2010). In this study, METH was used as both the Nrf2 activator and the neurotoxin, as would occur in drug abusers using METH repeatedly during pregnancy. Our studies show that in utero exposure to

METH during the fetal period can enhance developmental, gender-dependent toxicities including increased fetal death, edema and decreased weight in Nrf2-deficient fetuses, providing 276

the first evidence that Nrf2 plays an essential protective role during development. Our results also provide the first evidence for in utero METH-initiated olfactory deficits measured by odor detection, habituation and discrimination, which appear to constitute a novel and sensitive postnatal functional biomarker of long-term postnatal neurotoxicity. These postnatal long-term olfactory deficits initiated by in utero exposure to METH along with motor functional deficits are also modulated by Nrf2 activation, indicating a broadly protective role for Nrf2 during development.

277

2.5.3 MATERIALS AND METHODS

Chemicals and Reagents

Alkaline phosphatase and Complete, Mini, EDTA-free protease inhibitor cocktail tablets were obtained from Roche Diagnostics (Laval, QC, Canada). RNAse A/T1, Taq polymerase and dexoynucleotide triphosphates (dNTPs) were purchased from Fermentas Canada Inc.

(Burlington, ON, Canada). Proteinase K and dithiothreitol (DTT) were obtained from BioShop

Canada Inc. 8-Hydroxy-2‟-deoxyguanosine was obtained from Cayman Chemical Co. (Ann

Arbor, MI). RNAlater® and DNase were purchased from Qiagen (Mississauga, ON, Canada). 2-

Deoxyguanosine (2-dG) and pepstatin A were obtained from Sigma-Aldrich (Oakville, ON,

Canada). All other reagents were of analytical or HPLC grade.

Drugs

Pure racemic (D/L)-METH was provided by the Healthy Environments and Consumer

Safety Branch of Health Canada (Ottawa, Ontario, Canada). The identity and purity of (D/L)-

METH was determined by using a Bio-Rad REMEDi HS system and confirmed by liquid chromatography-mass spectrometry-mass spectrometry (Clinical Biochemistry, Hospital for

Sick Children, Toronto, ON, Canada).

Nrf2 knockout mouse colony

Nrf2 knockout (-/-) mice were provided by Dr. Masayuki Yamamoto (Institute of Basic

Medical Sciences and Center for Tsukuba Advanced Research Alliance, University of Tsukuba,

Tennoudai, Japan) through the RIKEN Bioresource Center, Experimental Animal Division. The colony was maintained on a CD1 background and was inbred by mating heterozygous 278

littermates. F10 pregnant dams were used for the developmental studies. All Nrf2 genotypes were confirmed as previously described by Itoh and coworkers (Itoh et al., 1997).

Heterozygous (+/-) females were housed overnight with congenic +/- male breeders. The presence of a vaginal plug the next morning was designated as gestational day (GD) 1. Pregnant females were isolated and housed in plastic cages with ground corn cob bedding (Beta Chip;

Northeastern Products, Warrensburg, NY, USA) and maintained in temperature-controlled rooms with a 12 hr light-dark cycle. Food (Laboratory Rodent Chow 5001; Ralston Purina,

Dosing regimens

METH was dissolved with sterilized 0.9% saline, and the drug or its vehicle were given via intraperitoneal (i.p.) injection in a fixed volume of 0.1 ml/10 g body weight. Mice were administered a single dose of METH (10, 20 or 40 mg/kg) or 0.9% saline (vehicle) on GD 17.

For 2-day dosing, METH (5 or 10 mg/kg, i.p.) was administered on GD 13 or GD 16 to initiate the expression of Nrf2-regulated genes, followed by a 24 hr period to allow for the Nrf2- mediated induction of proteins involved in antioxidative pathways, DNA repair and other potentially protective pathways. After the 24 hr period on GD 14 or GD 17, another dose of

METH (5 or 10 mg/kg, i.p.) was administered to determine the protective efficacy of induced

Nrf2-regulated pathways on METH-initiated developmental toxicity. On GD 19, dams were killed by cervical dislocation. The uterine horns were exteriorized, the number of fetuses and resorptions were noted, and the fetuses were removed and weighed. Samples were genotyped 279

for Nrf2 and gender, the latter via amplification of a part of the Y-chromosome specific Zfy gene

(Sah et al., 1995). For analysis of mRNA levels, pregnant dams were killed 6 or 24 hr after drug treatment and fetal brains were isolated. For analysis of DNA oxidation, pregnant dams were killed 4 hr after drug treatment and fetal brains were isolated. For behavioral studies, dams were allowed to deliver spontaneously, and tail snips were obtained from the offspring for genotyping. The offspring were tested at age 4 or 8 weeks. For Nissl staining and immunohistochemistry, brains from 10-week-old progeny were isolated and stored in 4% paraformaldehyde followed by 20% sucrose.

Analysis of mRNA levels by quantitative real-time PCR (RT-PCR)

Fetal brains were removed and stored in RNAlater solution (Qiagen, Mississauga, ON,

Canada). Total RNA was extracted using the RNeasy® lipid tissue mini kit according to the manufacturer‟s instructions (Qiagen, Mississauga, ON, Canada). To ensure purified RNA, on- column DNase digestion of DNA was performed using the RNase-free DNase set (Qiagen,

Canada) with random hexamer primers. The PCR reaction was performed in 20 µl using the

Lightcycler® FastStart DNA Master SYBR Green 1 reagents (Roche Diagnostics, Laval, QC,

Canada) and 2 uL of cDNA and 0.5 uM of reverse and forward primers. Specific primers were as follows: HO-1-f (CAA-GCC-GAG-AAT-GCT-GAG-TTC-ATG), HO-1-r (GCA-AGG-GAT-

GAT-TTC-CTG-CCA-G), NQO1-f (GCG-AGA-AGA-GCC-CTG-ATT-GTA-CTG), NQO1-r

(TCT-CAA-ACC-AGC-CTT-TCA-GAA-TGG), OGG1-f (ACT-GCA-TCT-GCT-TAA-TGG- 280

CC), OGG1-r (CGA-AGG-TCA-GCA-CTG-AAC-AG), β-actin-f (AGA-GCA-TAG-CCC-

TCG-TAG-AT), β-actin-r (CCC-AGA-GCA-AGA-GAG-GTA-TC). Quantification was performed on a Lightcycler® 2.0 detection system (Roche Diagnostics, Laval, QC, Canada).

PCR cycles proceeded as follows: initial denaturation for 10 min at 95°C, then 40 cycles of denaturation (10 sec, 95°C), annealing (30 sec, 55-60°C depending on the specific primer), and extension (20 sec, 72°C). The melting-curve analysis showed the specificity of the amplifications. A series of dilutions were used to generate a relative standard curve to determine the efficiency of amplifications. The relative quantification method was used to quantify mRNA levels using the housekeeping gene β-actin as the reference gene.

DNA extraction

The chaotropic NaI method (Ravanat et al., 2002) was used with the following modifications. After isolation of the nuclear pellet, 10 μl of RNase A/T1 mix were added and the samples were incubated for 2 hr at 50°C. A 20 μl volume of proteinase K (20 mg/ml in 10 mM Tris/1 mM EDTA, pH 8.5) was added prior to incubation at 50°C for 2 hr. Samples were then spun at 10,000 x g for 15 min to remove particulate material. Subsequently, 600 uL of the

NaI solution (7.6 M NaI, 40 mM Tris, 20 mM EDTA-Na2, 0.3 mM desferoxamine, pH 8.0) were added and the tubes were inverted for 2 min. Then, 1 mL of isopropanol was added and

DNA was precipitated by gently inverting the tube for 2 min. DNA was recovered by centrifugation and washed with 1 ml of 70% ethanol for five times. DNA was recovered by centrifugation and dissolved into sodium acetate buffer (20 mM sodium acetate, 0.1 mM desferoxamine, pH 5.2) for 1 hr at 37oC and by briefly sonicating on ice prior to DNA digestion.

DNA was digested with nuclease P1 (5 units) at 37oC. After 1 hr, Tris-HCl (1 M, pH 8.5) was 281

added followed by alkaline phosphatase (8 units) and this solution was incubated at 37oC for 1 hr. Samples were filtered using Microcon®-YM 10 filters (Millipore Canada Ltd.) and stored at

-80oC until analysis.

Detection of 8-hydroxy-2’-deoxyguanosine (8-oxodG) and 2-deoxyguanosine (2-dG)

Oxidation of 2-dG to 8-oxodG was quantified using an isocratic Series 200 high- performance liquid chromatography (HPLC) system with electrochemical detection

(PerkinElmer Instruments LLC, Shelton, Connecticut) as previously described (Goncalves et al.,

2009).

Functional deficits: Olfactory discrimination (OD) test

OD was assessed using the protocol described in Tillerson and coworkers (Tillerson et al., 2006), with the following modifications. Mice were separated into individual cages 24 hr prior to the test, and were given 60 min to acclimatize immediately prior to the test. The solutions used were 100 ng/mL of paprika or cinnamon. A petri dish was placed in the animal‟s home cage. Each mouse was presented with two 25 μl solutions: the first odor (paprika) on one side of the dish and water (control) on the other side. Time spent sniffing both the water and scented solution was recorded. Sniffing was defined when the animal‟s nose was located 1 cm or less from the odor. Five successive 3-min trials were performed, separated by 15-min inter- trial intervals. Habituation response was measured on the first five trials by examining investigation time across trials. On the sixth trial, the mouse was presented with the alternative odor, cinnamon, for 3 min along with the water control. As some mice spent equal time investigating the novel odor and water control, we expressed the data as the ratio of time spent detecting odor to that for water. Discrimination was defined as the difference in ratios between 282

time spent detecting the familiar odor (fifth trial) and time spent detecting the novel odor on the sixth trial.

Functional deficits: Motor Coordination Studies

Functional deficits were determined by assessment of motor coordination using the rotarod test as described previously (Jeng et al., 2006). Mice were given 60 min to acclimatize to the room followed by trial conditioning and training on a constant speed rotarod (rod diameter=3.6 cm). Mice were required to balance on the stationary rod for 30 sec to accustom themselves before being allowed to run with a constant speed of 5 rpm for 90 sec. Mice that succeeded in 3 trials (2 hr intervals) without falling were tested. In the testing phase, mice were tested for motor coordination at a constant speed of 5 rpm for 90 sec, after which the rotarod was gradually increased to a constant speed of 20 rpm for a maximum of 5 min. The performance time and speed at which the mice fell from the rod were recorded.

Functional Deficits: Activity Monitoring

Mice were given 60 min to acclimatize to the testing room. Activity was measured using the Linton AM1053 Standard (Dual Layer) X, Y, Z IR Activity Monitor and associated

Amonlite software, version 1.4 (Linton Instrumentation, Norfolk, UK). The monitor consisted of two levels of IR light beams set up such that the lower set of beams measured the activity of mice walking at the bottom of the cage, while the higher set measured the activity of mice standing on their hind legs as „rearing‟ activity. Activity measurements were performed during the afternoon and the animals were left alone in the room while measurements were taken in order to minimize external stimuli that might have interfered with their normal activity. The measure of activity was recorded every 5 min for a total of 60 min, such that 12 measurements 283

were taken from each animal in each experiment. The sum of the distance travelled as well as rearing time (counts every time mice were rearing in a given second) from those 12 measurements was then calculated and analyzed statistically.

Nissl staining and immunohistological analysis

The brains were cryoprotected in 20% sucrose then frozen and coronally cut at the olfactory bulbs (Bregma +4.20 mm) at a thickness of 10 um (Toronto Centre for

Phenogenomics, Pathology Core). Sections were Nissl-stained with cresyl violet or with anti- dopamine transporter (DAT) antibody (1:10000, Sigma-Aldrich, Oakville, ON, Canada) or mouse anti-glutamic acid decarboxylase (GAD)-6 primary antibody (1:100). The anti-GAD-6 antibody, developed by Dr. David I. Gottlieb (Washington University School of Medicine, St.

Louis, MO) was obtained from the Developmental Studies Hybridoma Bank developed under the auspices of the NICHD and maintained by The University of Iowa (Department of Biology,

Iowa City, IA 52242). For DAT and GAD-6, the secondary antibodies were 1:200 biotinylated goat anti-rabbit IgG or 1:200 biotinylated horse anti-mouse IgG (Vector Laboratories,

Burlington, ON, Canada) respectively.

Statistical analysis

Multiple comparisons among groups were analyzed by one-way ANOVA with a subsequent Tukey‟s test (GraphPad InStat®3.05, GraphPad Software, Inc., San Rafael, CA,

USA). Pairs of continuous data were analyzed by Student‟s t-test, and binomial data were examined using the Chi-square or Fisher‟s Exact test. The level of significance was determined to be at P < 0.05. 284

2.5.4 RESULTS

Effect of in utero METH exposure on number of live pups from Nrf2 heterozygous dams

METH was administered in a single dose of 10, 20 or 40 mg/kg, i.p. on gestational day

(GD) 17 to pregnant +/- Nrf2 dams bred with +/- Nrf2 males. Of the 25 dams treated with

METH, 1 dam dosed with 40 mg/kg died. METH caused a dose-dependent decrease in the number of live pups born compared to saline-exposed controls, with a 50% reduction at the dose of 20 mg/kg, and few pups surviving at the 40 mg/kg dose (p<0.05) (Figure 1a). While some pups were born alive, they died a few hours later. We also investigated a 2-day METH dosing regimen of 10 mg/kg i.p. on both GDs 16 and 17 to pregnant Nrf2 (+/-) dams, however there were no differences in litter size (Figure 1a insert).

Effect of in utero METH exposure on Nrf2 genotype distribution at weaning

At weaning, the saline and GD 17 METH-exposed progeny were separated by gender and then genotyped for Nrf2 (Figure 1b and c). In the saline-treated progeny, both male and females gave the expected Nrf2 genotype frequencies of wild-type (+/+), heterozygous (+/-) and homozygous null (-/-) mice close to the ideal Mendelian 1:2:1 ratio expected from heterozygous parents (Figure 1b and c). In contrast, for the female progeny, the relative frequency of

METH-exposed -/- Nrf2-deficient mice was less than that for the saline-exposed -/- Nrf2 controls (p<0.01) (Figure 1b). Similarly, there was a non-significant trend for decreased female

METH-exposed +/- Nrf2-deficient mice (43%) compared to saline-exposed +/- Nrf2 controls

(54%). Conversely, there was an increased proportion of METH-exposed +/+ Nrf2-normal female progeny (43%) compared to saline-exposed +/+ Nrf2 controls (21%) (p<0.01) (Figure

1b). In the METH-exposed male progeny, there was a non-significant trend for fewer -/- Nrf2- 285

deficient mice (18%) compared to saline-exposed -/- Nrf2 controls (27%) (Figure 1c).

Impaired development in in utero METH-exposed fetuses from Nrf2 heterozygous dams

To examine effects on the fetus, a METH dose of 20 mg/kg i.p. was given on GD 17 to pregnant +/- Nrf2 dams. On GD 19, the fetuses were examined and were found to exhibit varying fetal toxicities. Among the METH-treated dams, one entire litter died in utero with extensive late resorptions that could not be genotyped for Nrf2 or Zfy. The demise of homozygous Nrf2 fetuses in utero may have interfered with the survival of other fetuses in the same litter. The number of late resorptions in METH-treated dams was 11 out of 43 implantation sites whereas saline-treated dams had no resorptions out of 33 implantation sites.

Some fetuses appeared to be underdeveloped, while -/- and +/- Nrf2-deficient mice manifested a red phenotype and edema (Figure 2a). Of the 35 METH-exposed fetuses, 14 had a red phenotype, including 6 Nrf2 -/- females, 3 Nrf2 -/- males, 2 Nrf2 +/- males, 2 Nrf2 +/- females and 1 Nrf2 +/+ female. This phenotype was not observed in any saline-exposed fetuses.

In METH-exposed female fetuses, Nrf2-deficient +/- and -/- mice exhibited a gene dose- dependent reduction in fetal weight, with Nrf2 -/- fetuses less than Nrf2 +/- fetuses, which in turn were less than Nrf2-normal +/+ fetuses (p<0.01) (Figure 2b) METH-exposed female Nrf2

-/- fetuses weighed 50% less than METH-exposed Nrf2 +/+ fetuses (p<0.01). Similarly, fetal weights for Nrf2 +/- and -/- fetuses were respectively 25% and 55% less than for their genotype- matched saline controls (p<0.01). In the METH-exposed male fetuses, both Nrf2-deficient +/- and -/- mice had reduced fetal weights that were 25% less than their genotype-matched saline controls (p<0.05) (Figure 2c). Among METH-exposed male mice, the respective decreases in fetal body weights in Nrf2-deficient -/- and +/- mice compared to Nrf2-normal +/+ mice were 286

not significant.

Analysis of mRNA levels in Nrf2 +/+ and -/- fetal brains by quantitative real-time PCR (RT-

PCR)

METH (10 mg/kg i.p.) was administered to Nrf2 +/- pregnant mice on GD 16 to initiate the expression of Nrf2-regulated genes, followed by a 24 hr period to allow for the Nrf2- mediated induction of proteins involved in antioxidative pathways, DNA repair and other potentially protective pathways in the fetal brain. After the 24 hr period, a second dose of

METH (10 mg/kg i.p.) was administered on GD 17 to determine if expression of the Nrf2- regulated genes heme oxygenase-1 (HO-1), NAD(P)H:quinone oxidoreductase (NQO1) and 8- oxoguanine glycosylase 1 (OGG1) are modulated in the fetal brain after in utero METH exposure. In the brains of METH-exposed Nrf2-normal +/+ fetuses, HO-1 mRNA levels were increased by 38% within 6 hr after dosing, and remained elevated (61%) at 24 hr compared to saline controls (p<0.05) (Figure 3). In METH-exposed Nrf2-deficient -/- mice, HO-1 mRNA remained at basal levels throughout the 24-hr sampling period. In the brains of Nrf2 +/+ fetuses, in utero METH exposure induced NQO1 at 6 hr, with an increase of 60% (p<0.05) that remained 35% over controls after 24 hr (p<0.05). In METH-exposed Nrf2 -/- mice, NQO1 mRNA remained at basal levels throughout the 24 hr sampling period. Interestingly, basal levels of NQO1 expression in saline-treated Nrf2 -/- mice were consistently lower by an average of 20% compared to Nrf2 +/+ mice, although this difference was not significant (Figure 3).

This trend was also observed for OGG1, where basal levels in the Nrf2 -/- fetal brains were lower (Figure 3). In the fetal brain of Nrf2 +/+ mice, METH induced OGG1 expression within

6 hr by 70% compared to saline controls (p<0.05), and mRNA levels remained elevated by 50% 287

at 24 hr (p<0.05). In METH-exposed Nrf2 -/- mice, OGG1 mRNA remained at basal levels throughout the 24 hr sampling period.

DNA oxidation is increased in METH-exposed fetal brain

There were no differences in DNA oxidation in fetal brains of saline-exposed Nrf2 +/+,

+/- or -/- fetuses. Following in utero METH exposure on GDs 16 and 17, DNA oxidation was increased within 4 hr by 80-100 % respectively in Nrf2 +/+ and +/- compared to their genotype- matched saline controls (p<0.01) (Figure 4). DNA oxidation in the fetal brains from METH- exposed Nrf2-deficient -/- progency was increased 2.7-fold compared to saline-exposed Nrf2 -/- controls (p<0.01), and increased by 45% compared to METH-exposed Nrf2 +/+ and +/- mice

(p<0.05) (Figure 4).

In utero exposure to METH results in olfactory deficits in offspring of Nrf2 (+/-) dams

METH was administered in a single dose of 10 or 20 mg/kg i.p. on GD 17 to pregnant

Nrf2 +/- females. Male offspring were tested for olfactory deficits measured by odor detection, habituation and discrimination. Offspring exposed in utero to METH (20 mg/kg) had an odor detection deficit across all Nrf2 genotypes that was detected at postnatal week (PNW) 4 (Figure

5). This deficit was not observed with a lower dose of METH (10 mg/kg), after which offspring exposed in utero were able to detect and habituate to an odor at PNW 4 (p<0.01)

(Figure 5). However, even at the lower dose, METH-exposed mice were not able to discriminate the novel odor from the habituated odor (Figure 5, lower panel). This deficit in long-term postnatal olfactory discrimination caused by a single in utero exposure to low-dose

METH was independent of the Nrf2 genotype. 288

To determine if an initial activating exposure was necessary to reveal Nrf2 dependency, we subsequently investigated a 2-day dosing regimen administering METH 10 mg/kg i.p. on

GDs 16 and 17 to pregnant Nrf2 +/- dams. At PNW 4, both male and female offspring were tested for olfactory deficits (Figure 6). Both METH-exposed Nrf2 +/+ and -/- female offspring exhibited a detection deficit where they did not explore the odors compared to saline controls

(p<0.01), precluding their subsequent evaluation for olfactory discrimination deficits (Figure 6).

METH-exposed Nrf2 +/+ male offspring were able to detect and habituate to paprika (Figure

6), but during the discrimination phase exhibited a discrimination deficit where they did not investigate the novel odor (p<0.05) (Figure 6). A Nrf2-dependent protective effect was observed for olfactory detection in males exposed to the 2-dose regimen, with which METH- exposed Nrf2 -/- male offspring exhibited a detection deficit where they did not explore either of the odors (Figure 6).

To determine more subtle effects, we further investigated a lower 2-day dosing regimen administering METH 5 mg/kg i.p. on GDs 16 and 17 to pregnant Nrf2 +/- females. At PNW 4, both male and female offspring, Nrf2 +/+ and -/- were able to detect, habituate and discriminate the odors and exhibited no olfactory deficits (Figures S2, S3). Similarly, exposures on GDs 13 and 14 to METH 10 mg/kg i.p. did not cause olfactory deficits in mice of any Nrf2 genotype

(Figures S4, S5).

Enhanced postnatal motor activity deficits in offspring of Nrf2 +/- dams exposed to METH.

A 2-day METH dosing regimen of 10 mg/kg i.p. on GDs 16 and 17 to pregnant Nrf2 +/- females was used to evaluate postnatal motor activity performance. After testing for olfactory deficits, the offspring were also assessed for motor coordination deficits using a rotarod 289

apparatus at PNW 6 and an activity monitor at PNW 8. This level of METH exposure did not cause motor coordination deficits in any group (Figure 7, upper panels).

During activity monitoring, METH-exposed Nrf2 +/+ female offspring had a 27% decrease in distance travelled compared to saline controls, however this deficit was not statistically significant (Figure 7). METH-exposed Nrf2 -/- female offspring had a 46% decrease in distance travelled compared to female Nrf2 -/- saline controls (p<0.05) (Figure 7).

A non-significant Nrf2-dependent trend was observed for distance traveled, which was 26% lower in female METH-exposed Nrf2 -/- offspring than Nrf2 +/+ mice. Both METH-exposed

Nrf2 +/+ and -/- female offspring exhibited a 30-32% decrease in rearing counts compared to their genotype-matched saline controls, however this was not statistically significant (Figure 7).

Nrf2 +/+ male offspring had an 18% decrease in distance travelled compared to saline controls, however this was not statistically significant (Figure 7). METH-exposed Nrf2 -/- male offspring had a significant 45% decrease in distance travelled compared to both Nrf2 -/- saline controls and METH-exposed Nrf2 +/+ offspring (p<0.05), the latter revealing a Nrf2-dependent effect (Figure 7). METH-exposed Nrf2 +/+ males had a non-significant 30% decrease in rearing counts compared to their Nrf2 +/+ saline controls, while Nrf2 -/- male offspring exhibited a 45% decrease in rearing counts compared to their Nrf2 -/- saline controls (p<0.05)

(Figure 7).

Nissl staining and Immunohistological Analysis of the Olfactory bulbs

Male offspring exposed to saline or METH (10 mg/kg, i.p.) in utero on GDs 16 and 17 were sacrificed and histological sections of the olfactory bulbs were further characterized to 290

determine potential cellular and subcellular neurotoxic effects that may be modulated by Nrf2.

Nissl staining of cell bodies in the olfactory bulb showed the characteristic olfactory bulb layers, with no microscopic differences among METH-exposed Nrf2 +/+ and -/- offspring or their saline-exposed controls (Figure 8a). Immunohistological results for the dopamine transporter

(DAT) in the olfactory bulb were not altered in METH-exposed Nrf2 +/+ or -/- offspring or their saline controls (Figure 8a). Considering the abundance of GABAergic neurons in the olfactory bulbs (Parrish-Aungst et al., 2007), we quantified glutamic acid decarboxylase-65 (GAD-65)- positive cells in the granule cell layer (GCL) of the olfactory bulbs (Figure 8b). GAD-65- positive cells were reduced by 20-40% in METH-exposed mice compared to saline-exposed mice, although this apparent reduction was not statistically significant. Microscopic inspection suggested a similar qualitative pattern of decreased staining in the external plexiform layer

(EPL). These changes were not dependent on Nrf2 genotype, nor were they associated with the extent of functional olfactory deficits. 291

292

Figure 1. Effect of in utero methamphetamine (METH) exposure on live pups and Nrf2 genotype distribution at weaning. METH was dissolved in 0.9% saline and administered in 1 dose of 10, 20 or 40 mg/kg, i.p. on gestational day (GD) 17 to pregnant +/- Nrf2 dams. Saline vehicle was used as the control. The mice were allowed to litter out. (a) The number of live pups were counted at birth. a= p<0.05 compared to saline-treated litters. (X,Y) indicates the number of dams (X) and pups (Y) analyzed. Insert: Pregnant Nrf2 (+/-) mice were administered one dose of METH (10 mg/kg i.p.) or saline on GD 16 and GD 17. At weaning, the saline and single day METH-exposed progeny were separated into females (b) and males (c) and then genotyped for Nrf2. For a particular treatment, the number of mice of each genotype was calculated as a percent of total. a=p<0.01 compared to genotype matched saline-treated mice. In brackets are the numbers of mice used for the analysis. 293

294

Figure 2. Impaired development in in utero METH-exposed fetuses from Nrf2 heterozygous dams. METH was dissolved in 0.9% saline and administered in 1 dose of 20 mg/kg, i.p. on GD 17 to pregnant +/- Nrf2 dams. Saline vehicle was used as the control. On GD 19, dams were killed by cervical dislocation. The fetuses were removed and were weighed. Samples were genotyped for Nrf2 and amplification of a part of the Y-chromosome specific Zfy gene to assess gender. (a) Representative litter indicating METH-initiated fetal toxicity separated by genotype and gender. F=female, M=male. (b,c) Fetal weight by treatment, Nrf2 genotype and gender (b) female and (c) male. a=p<0.01 compared to genotype matched saline- treated mice. b=p<0.05 compared to genotype matched saline-treated mice. *=p<0.01 compared to METH-treated (+/+) and (+/-) mice. (X,Y) indicates the number of dams (X) and pups (Y) analyzed. 295

Figure 3. Effect of METH on Nrf2-mediated gene expression in Nrf2 (+/+) and (-/-) fetal brain. Pregnant Nrf2 (+/-) mice were administered one dose of METH (10 mg/kg i.p.) or saline on GD 16 and GD 17 to determine if the Nrf2-regulated genes heme oxygenase-1 (HO-1), NAD(P)H:quinone oxidoreductase (NQO1) and oxoguanine glycosylase 1 (OGG1) are modulated in the fetal brain as assessed by RT-PCR analysis of mRNA levels. Time-points included 0, 6 and 24 hr after dosing. mRNA levels were standardized using B-actin. a=p<0.05 compared to saline-treated wild-type mice. * =p<0.05 compared to METH-treated Nrf2 (-/-) mice. Fetal brains from 4-6 mice were used per group from 4 dams per timepoint. 296

Figure 4. DNA oxidation is increased in fetal brains after in utero METH exposure. Pregnant Nrf2 (+/-) mice were administered one dose of METH (10 mg/kg i.p.) or saline on GD 16 and GD 17. Dams were sacrificed 4 hr after the last injection and fetal brain was isolated to determine oxidative DNA damage quantified by 8-oxodG and standardized to dG. a = p<0.01 compared to genotype matched saline-treated mice. *=p<0.05 compared to METH-treated Nrf2 (+/+) and (+/-) mice. Fetal brains from 4-6 mice were used per group from 4 dams.

297

Figure 5. In utero exposure to METH on GD 17 results in olfactory deficits in offspring of Nrf2 (+/-) dams. METH was dissolved in 0.9% saline and administered in 1 dose of 10 or 20 mg/kg, i.p. to pregnant Nrf2 (+/-) mice on GD 17. The mice were allowed to litter out and at weaning were genotyped for Nrf2. Male offspring were tested for olfactory deficits measured by odor detection, habituation and discrimination on postnatal week (PNW) 4. Detection =p<0.01 compared to genotype-matched Trial 1 saline-treated water group. Habituation=p<0.01 compared to genotype-matched Trial 1 paprika group. Discrimination was defined as the difference in time spent detecting the familiar odor (fifth trial) and time spent detecting the novel odor on the sixth trial. Discrimination=p<0.01 compared to genotype-matched trial 5 paprika group. No discrimination=p<0.05 compared to genotype-matched trial 6 saline-treated cinnamon group. Each group consisted of 7-8 mice. 298

299

Figure 6. 2-day dosing with METH on GD 16 and GD 17 leads to Nrf2 and gender- dependent postnatal olfactory deficits. Pregnant Nrf2 (+/-) mice were administered one dose of METH (10 mg/kg i.p.) or saline on GD 16 and GD 17. The mice were allowed to litter out and at weaning were genotyped for Nrf2. Male and female (+/+) and (-/-) offspring were tested for olfactory deficits measured by odor detection, habituation and discrimination on postnatal week PNW 4. No Detection =p<0.01 compared to genotype-matched Trial 1 saline-treated paprika group. Discrimination was defined as the difference in ratios between time spent detecting the familiar odor (fifth trial) and time spent detecting the novel odor on the sixth trial (p<0.01). No discrimination=p<0.05 compared to saline-treated wild-type mice. Each group consisted of 7-8 mice.

300

Figure 7. Enhanced postnatal motor activity deficits in offspring of Nrf2 (+/-) dams exposed to METH. Pregnant Nrf2 (+/-) mice were administered one dose of METH (10 mg/kg i.p.) or saline on GD 16 and GD 17. The mice were allowed to litter out and at weaning were genotyped for Nrf2. Male and female (+/+) and (-/-) offspring were tested for motor coordination deficits as assessed by the rotarod test at PNW 6 wk. The activity of these mice was also assessed at PNW 8. Total distance travelled and rearing time were collected by the activity monitor over a 60 min period. a=p<0.05 compared to genotype-matched saline-exposed mice. * = p<0.05 compared to METH-exposed Nrf2 (+/+) mice. Each group consisted of 7-8 mice. 301

302

Figure 8. Nissl, DAT and GAD-65 analysis of the olfactory bulbs of male offspring exposed in utero to METH. Pregnant Nrf2 (+/-) mice were administered one dose of METH (10 mg/kg i.p.) or saline on GD 16 and GD 17. The mice were allowed to litter out and at weaning were genotyped for Nrf2. After functional testing, the olfactory bulbs of male offspring were analyzed. (a) Nissl (100X), DAT (200X) and GAD-65 (100X) staining of coronal sections through the olfactory bulb. (b) Quantification of GAD-65 immuno-positive cells in the granule cell layer. Immunohistological staining is representative of 3 mice for each group. The * indicates a difference from saline-exposed Nrf2 -/- fetuses (p<0.05), and the # indicates a marginal difference from saline-exposed Nrf2 +/+ mice (p=0.068).

303

2.5.5 DISCUSSION

Our results show that deficient fetal activation of Nrf2-regulated protective responses to oxidative stress can enhance susceptibility to macromolecular and cellular damage caused by

METH in the fetal brain, and to both METH-initiated fetal developmental toxicities and postnatal neurobehavioural deficits. In the developing embryo and fetus, basal activities of most antioxidant enzymes including SOD and catalase are notably lower than their respective activities in the adult liver (Wells et al., 2009; el-Hage and Singh, 1990; Winn and Wells, 1999).

Other ROS-protective enzymes like the antioxidative enzyme G6PD, and the repair enzyme

OGG1, which repairs oxidatively damaged DNA, can have activities in fetal tissues that are respectively at least equal to or up to 2-fold higher than adult activities (Wong et al., 2008).

This diverse group of enzymes are inducible, at the transcriptional level, by ROS-initiated activation of Nrf2, which heterodimerizes with members of the small avian musculoaponeurotic fibrosarcoma (Maf) protein family in the nucleus and binds the ARE (Itoh et al., 1997).

Polymorphisms have been identified within the promoter region of the human Nrf2 gene

(Yamamoto et al., 2004), and genetic variation here along with polymorphisms in the ARE region of Nrf2 target genes (Wang et al., 2007a) may cause variability in fetal Nrf2 activity, and hence the response to oxidative stress and developmental and functional toxicity. In addition, there are also dietary inducers of Nrf2-regulated genes such as sulforaphane, a compound generated from cruciferous vegetables such as broccoli, which may modulate risk.

The studies herein show that in utero exposure to the ROS-generating teratogen METH can activate Nrf2 in the fetal brain, as mRNA levels for antioxidant and DNA repair enzymes were increased within 6 hours in the brains of Nrf2 +/+ fetuses, but not Nrf2 -/- fetuses, when using the 2-day dosing regimen of 10 mg/kg on both GDs 16 and 17. The absence of Nrf2 304

activation in Nrf2-deficient -/- fetuses resulted in a greater increase in METH-initiated oxidative

DNA damage in Nrf2 -/- fetal brains, indicating that the increase in mRNA levels of Nrf2- regulated genes in Nrf2-normal +/+ fetuses was sufficient to induce the spectrum of ROS- protective proteins, which in concert protected Nrf2 +/+ fetal brains from oxidative macromolecular damage.

Enhanced fetal toxicities such as increased resorptions, the red phenotype and edema, and reduced fetal weight, in METH-exposed Nrf2-deficient -/- fetuses compared to Nrf2 +/+ littermates indicates that fetal Nrf2-regulated antioxidant and repair responses, as opposed to maternal factors, play an important role in determining fetal development. The red phenotype may be indicative of vascular changes or free radical membrane damage. The same fetal importance of Nrf2-dependent protection with in utero METH exposure was evident in postnatal neurodevelopmental deficits, including Nrf2-dependent deficits in olfactory detection and activity in Nrf2 deficient offspring. Nrf2 gene dose-dependency was observed for METH- initiated fetal weight reductions in female fetuses, and similar non-significant trends in postnatal activity and rearing deficits in female offspring. The range of METH doses up to 40 mg/kg used in our studies give concentrations in fetal brain similar to those in METH-exposed infants (Bost et al., 1989; Won et al., 2001). However, with lower doses of METH and depending on prior induction of Nrf2-protective pathways, the fetus was able to combat and reduce the detrimental effects of oxidative stress, whereas Nrf2-deficient fetuses exhibited enhanced toxicities. The highest dose (40 mg/kg) of METH may overwhelm Nrf2-dependent cytoprotective pathways leading to severe fetal toxicities and death. However, such outcomes may also result in part from mechanisms independent of Nrf2 and/or oxidative stress, and some Nrf2-independent effects were observed with METH, including olfactory discrimination deficits in both genders, 305

and olfactory detection deficits in female offspring.

Female Nrf2-deficient -/- fetuses exposed in utero to METH were more susceptible than

-/- males to fetal death and decreased fetal birth weight. For neurodevelopmental outcomes, in utero METH exposure caused Nrf2-dependent postnatal deficits in olfactory detection that appeared to be more severe in female Nrf2 +/+ and -/- mice. The basis for this apparent gender specificity remains unclear, but may depend in part on METH and Nrf2-regulated genes and their interactions with gender-specific genes. Some Nrf2-deficient fetuses did not exhibit toxicities, possibly due to the involvement of alternative mechanisms specific to the fetuses. For example, Nrf1 is related to Nrf2, and these proteins constitute a family of transcription factors that share homology at several protein domains and are capable of binding to the ARE

(Venugopal and Jaiswal, 1998). However, Nrf2 is reported to be more efficient than Nrf1 in upregulation of ARE-mediated gene expression (Venugopal and Jaiswal, 1998), and our mRNA analysis did not show any compensatory increases by Nrf1 in antioxidant responses in Nrf2 -/- fetal brains, at least for the genes investigated in our study. Nevertheless, Nrf1 has been shown to be important for development, as Nrf1 -/- mice die in utero due to anemia resulting from abnormal fetal liver erythropoiesis (Chan et al., 1998). Constitutive activation of Nrf2 in Keap1

-/- mice is also detrimental, causing hyperkeratotic lesions that block the esophagus and increase postnatal death, attributed to starvation (Wakabayashi et al., 2003). However, selective upregulation of Nrf2 during fetal development under conditions of increased oxidative stress may be protective, as shown in our study, and in studies where activation of Nrf2 in embryos can protect against ethanol-induced embryotoxicity (Dong et al., 2008).

Our studies show that in utero METH exposure can cause postnatal neurodevelopmental impairments, including olfactory deficits, that in some but not all cases can be modulated by 306

relatively moderate changes in Nrf2-regulated genes acting in concert. Rodents rely on olfaction for various functions including mating and exploration of their environment.

Olfactory discrimination tests can assess forms of learning and memory, although this depends on adequate detection of and habituation to odors (Brennan and Keverne, 1997; Gheusi et al.,

2000). While a single in utero exposure to METH (20 mg/kg) led to olfactory detection deficits across all genotypes, the use of a 2-day dosing regimen of 10 mg/kg on both GDs 16 and 17 resulted in distinct patterns of deficits in odor detection and discrimination that in some cases were both gender- and Nrf2-dependant. METH-exposed female Nrf2 +/+ and -/- mice and male

Nrf2 -/- were not able to detect the odors presented, and this deficit correlated with reduced spontaneous motor activity. This suggests that reduced sense of olfaction may manifest as a reduction in exploration of the environment as assessed by activity monitoring, although our results do not preclude the possibility of other sensory effects that affect activity and/or the anxiety levels of these mice.

In utero exposure to METH did not cause postnatal motor coordination deficits determined by rotarod performance, suggesting no apparent loss of dexterity or fine motor skill.

Although we have observed motor coordination deficits in the offspring of other mouse strains exposed in utero to METH (Jeng et al., 2005; Wong et al., 2008), the doses used herein were lower, and presumably below the threshold necessary for this neurodevelopmental toxicity. In humans, olfactory deficits have been shown to be an early marker of various neurodegenerative diseases and may precede motor deficits in Parkinson‟s disease by years (Doty et al., 1988; Ross et al., 2008; Talamo et al., 1989; Larsson et al., 2006). Male Nrf2 +/+ offspring exposed in utero to METH did not exhibit decreased motor activity and were able to detect and habituate to odors; however, they were not able to discriminate a novel odor, indicating a cognitive deficit. 307

Olfactory discrimination deficits were also evident in our related studies of adult mice treated with two doses of METH separated by 24 hr, in which discrimination was also modulated by

Nrf2 (Ramkissoon and Wells, 2010). Interestingly, earlier gestational exposures to the 2-day dosing regimen of METH on GDs 13 and 14 did not result in olfactory deficits, possibly because the earliest synapses in the olfactory bulb are not detected until GD 14, with synaptic connections increasing by GD 17 and continuing to develop in postnatal life (Hinds and Hinds,

1976). This indicates that in utero exposures to METH later in gestation, during the period of active synaptogenesis in the olfactory bulb, may have a greater impact on olfactory function.

Molecular changes associated with olfactory deficits indicate enhanced DNA oxidation in the fetal brains and a reduction in GAD-65 levels in the olfactory bulbs of fetuses exposed in utero to METH. METH enhanced the level of oxidatively damage DNA in fetal brains from all

Nrf2 genotypes, but significantly more so in Nrf2-deficient -/- fetuses. Although METH- initiated DNA oxidation in fetal brains was greater for Nrf2 -/- than +/+ fetuses, GAD-65 in adult male brains exposed in utero to METH was decreased to similar degree in Nrf2 +/+ and -/- mice. This suggests that while the observed olfactory deficits caused by METH may involve a loss of GABAergic neurons, those deficits that were Nrf2-dependent may involve additional mechanisms, although we cannot exclude the possibility that there may subtle differences in the patterns of oxidatively damaged DNA within the olfactory bulb or at the level of specific genes.

A number of studies have linked olfactory discrimination deficits and olfactory function to

GABAergic systems within the olfactory bulb, where aging and/or decreased neurogenesis led to decreased new GCL GABA neurons, which correlated with olfactory discrimination deficits

(Gheusi et al., 2000; Enwere et al., 2004). While there were no differences in the levels of DAT in fetal brains of METH-exposed and saline-exposed mice, consistent with previous studies 308

showing no changes in tyrosine hydroxylase levels (Jeng et al., 2005; Wong et al., 2008), the dopaminergic system may also contribute to olfaction (Tillerson et al., 2006), possibly through compensational changes in the GABAergic system of the olfactory bulb.

In conclusion, a deficiency in Nrf2 activation during pregnancy and exposure to METH led to fetal developmental toxicity and postnatal functional deficits, revealing an important protective role for Nrf2 in enhanced developmental oxidative stress. ROS-initiated Nrf2 activation allows for induction of proteins involved in antioxidative pathways, DNA repair and other potentially protective pathways which can act together to protect against molecular and behavioral toxicities associated with subsequent doses of METH. This mechanism may be constitute an important determinant of fetal risk in pregnant abusers of METH, or other ROS- generating neurotoxins, where interindividual variations in drug abuse patterns, dietary exposures and genetic predispositions may alter the induction and duration of Nrf2 activation in the fetus, thereby varying the fetal susceptibility to METH toxicities. The discovery of postnatal olfactory deficits following in utero METH exposure, at lower doses that do not affect postnatal motor coordination, suggests the potential of a novel and more sensitive test for neurodevelopmental deficits. Therapeutic strategies to enhance the Nrf2-dependent ROS- protective responses of the developing fetus may prove worthwhile. 309

CHAPTER 3: SUMMARY, CONCLUSIONS AND FUTURE STUDIES

310

3.1 SUMMARY AND CONCLUSIONS

Due to the continued use of METH and the uncertainty of its long-term neurotoxic consequences in humans, studies investigating mechanisms that enhance potential biomarkers of neurotoxicity and functional deficits are needed to characterize determinants of susceptibility and hence risk. These determinants may also play a role in neurodevelopmental deficits caused by in utero exposure to amphetamines, as well as neuropathologies associated with the aging brain. In this thesis, we identified determinants such as a role for human PHS-1 and 2 in bioactivation and ROS-generation of METH, MDA and endogenous brain compounds, such as

DA, as well as its precursor, L-DOPA and metabolites DOPAC and HVA. Furthermore, we identified a role of the cytoprotective transcriptional responses mediated by Nrf2 during METH- initiated oxidative stress in the adult brain. We determined that Nrf2 is activated by METH, but the resulting expression of cytoprotective proteins was not sufficiently rapid to block the neurotoxic effects of a single drug dose. In contrast, with multiple dosing over 2 days, allowing time for protein expression, Nrf2 had a protective effect against the second METH dose, evidenced by an enhanced susceptibility of Nrf2 -/- mice to METH-initiated oxidative stress, neurotoxicity and functional deficits. Similarly, following pretreatment of pregnant dams one day before with a low Nrf2-activating dose of METH, Nrf2 also played an important protective role in the developing fetus and in fetal brain where it blocked fetal toxicities and postnatal neurodevelopmental deficits caused by in utero exposure to METH.

Study 1: Prostaglandin H synthase-1-catalyzed bioactivation of neurotransmitters, their precursors, and metabolites: oxidative DNA damage and electron spin resonance spectroscopy studies. 311

In study 1, we used ovine PHS-1 to study the components of the dopaminergic pathway including DA, its precursor and metabolites, which structurally are phenols, catechols and amines that are considered potential substrates for PHS. We determined that ovine PHS-1 does indeed generate free radical intermediates not only from DA, but also from its L-DOPA precursor and DOPAC metabolite. The precursor and metabolites traditionally are considered pharmacologically inactive based upon receptor binding assays, but herein were shown to be as potent as DA as substrates for PHS-catalyzed bioactivation, and in generating ROS-mediated oxidative DNA damage. Electron spin resonance (ESR) spectroscopy revealed that the neurotransmitter DA, its precursor L-DOPA as well as epinephrine and normetanephrine, were all bioactivated by PHS-1 to putative carbon-centered free radical intermediates (Goncalves et al., 2009). While our in vitro DNA oxidation study focused on the dopaminergic pathway, we have found that other neurotransmitters such as epinephrine and its metabolites are similarly potent substrates for PHS-catalyzed bioactivation to free radical intermediates that enhance

ROS-mediated DNA oxidation (Jeng et al., 2011; Goncalves et al., 2009). Furthermore, other related endogenous compounds, such as 5-S-cysteinyl conjugates of DA and epinephrine, may be bioactivated by hPHS-1/2, causing oxidative DNA base modifications that initiate CNS damage. Our study reveals a mechanism of toxicity by which other endogenous brain compounds that are substrates for PHS-catalyzed bioactivation may exhibit neurodegenerative effects that extend beyond dopaminergic neurons, warranting further testing. Although treatment of PD with the drug L-DOPA provides initial improvement in motor function, long- term use of L-DOPA is associated with uncontrolled movements (dyskinesia) and increased cognitive impairment (Colosimo and De Michele, 1999; Factor et al., 1995). PHS-catalyzed bioactivation of L-DOPA leading to DNA oxidation provides a novel potential molecular 312

mechanism for the neurotoxicity caused by chronic treatment with L-DOPA.

While our study was in vitro, incubations with purified ovine PHS-1 have been used previously in our laboratory to establish a role for PHS-catalyzed bioactivation in free radical formation, ROS generation and DNA oxidation for phenytoin, benzo[a]pyrene, METH, MDMA and MDA. Subsequently, the relevance of these in vitro results were corroborated by in vivo studies; for example, in MDMA-treated PHS-1 KO mice, which accumulated less oxidative damage and were protected from various toxicities compared to their wild-type controls (Jeng and Wells, 2010). Furthermore, in our recent studies of untreated aging PHS-1 KO mice, brains of 2 year-old heterozygous (+/-) and homozygous (-/-) PHS-1 knockout mice and their wild-type

(+/+) PHS-1-normal controls were analyzed for DNA oxidation. Compared to aging +/+ PHS-

1-normal mice, aging -/- PHS-1-knockout mice had less oxidative DNA damage in the cortex, hippocampus, cerebellum and brainstem, highlighting the in vivo relevance of mouse PHS-1 in the generation of DNA oxidation that accumulates with aging (Jeng et al., 2011). In the oxidatively damaged brain regions, aged PHS-1 +/+ and +/- mice had significantly higher DNA oxidation compared to young adults, indicating that the cumulative damage is an age-dependent characteristic. However, the striatum and brainstem were not fully protected in the aging PHS-1

-/- mice, which suggests a bioactivating role for the PHS-2 isoform, as distinct from the PHS-1 isoform, and/or ROS production via other pathways such as the oxidation of DA by monoamine oxidase, enhanced activity of NADPH oxidases or mitochondrial metabolism.

Study 2: Human prostaglandin H synthase (hPHS)-1- and hPHS-2-dependent bioactivation, oxidative macromolecular damage and cytotoxicity of dopamine, its precursor and metabolites. 313

In study 2, we extended our findings from ovine PHS-1 to evaluate the same substrates from study 1 for bioactivation by human PHS-1 and hPHS-2. In part to avoid the confounding effects of differential PHS isozyme stability in broken cell preparations, we used CHO-K1 cells either untransfected or stably expressing hPHS-1 or hPHS-2 to investigate hPHS isozyme- dependent oxidative damage and cytotoxicity. Our results provide the first evidence that human

PHSs can bioactivate not only DA, but also its precursor and several metabolites, resulting in enhanced oxidative damage to both protein and DNA, and enhanced cytotoxicity. Furthermore, we were able to reduce these cytotoxic effects either with PHS-1/2 inhibition by ASA or by pretreatment with catalase, which detoxifies hydrogen peroxide. This bioactivation was PHS isozyme-specific, with a more potent contribution from hPHS-2 despite the higher PHS activity in PHS-1-expressing cells. Particularly important was the high level of bioactivation and cytotoxicity for the DA precursor and metabolites, since they are generally considered inactive based upon DA receptor binding assays. Oxidative damage to DNA occurred at lower substrate concentrations and earlier than protein oxidation, suggesting that DNA oxidation may play an important initiating role in the molecular mechanism of cytotoxicity.

In our experiments, we tested concentrations of these compounds at physiological levels of 10 uM, but also at higher concentrations to account for our relatively brief exposure of 24 hours. While a 10 uM concentration did not lead hPHS-1/2-catalyzed cytotoxicity, enhanced

DNA oxidation was observed in hPHS1/2-transfected but not untransfected CHO-K1 cells.

Unlike the limited outcome of cell death for in vitro studies, accumulation of oxidatively damaged DNA in vivo may have pathogenic consequences not reflected by cell death in culture

(discussed below). While our lower concentrations of some substrates illustrated susceptibility in the hPHS cell lines, these effects in vivo may be enhanced and accumulate after longer 314

durations of exposure, as seen during aging.

In our study, the use of non-neuronal cell lines limited the contribution of other brain metabolizing enzymes, transporters and receptor-mediated second messenger effects that might have confounded our interpretation of the role of hPHS-catalyzed bioactivation. In particular, these cell lines do not express the DA transporter (Pifl et al., 1993) which could interfere with our interpretation of the relative bioactivation of DA versus its metabolite and precursor bioactivation, as the latter two are not substrates for the transporter.

To assess cytotoxicity, we found LDH release to be a reliable, reproducible and sensitive technique for evaluating the cytotoxicity of the range of compounds tested. At the early stages of cytotoxicity, LDH is released from cells with a damaged membrane. Cell viability is dependent on the integrity of the cell membrane. Together with other endpoints of oxidative macromolecular damage to protein and DNA, our study provided a comprehensive assessment of the different types of toxicity that may occur, which could include apoptosis and/or necrosis.

Our approach assessed the full cytotoxic impact rather than distinguishing between the types of cell death. Perhaps more importantly, we showed that oxidative damage can occur in the absence of cell death, and the cumulative effect of this oxidative damage may have important functional consequences independent of cell death. A neurotoxic role for DNA oxidation is consistent with the enhanced neurodevelopmental deficits observed in the offspring of DNA- repair deficient Ogg1 and Csb knockout mice exposed in utero to the ROS-initiating teratogen

METH (McCallum et al., 2011; Wong et al., 2008).

hPHS-1 and hPHS-2 are expressed within the cell and the protein is not present in the media, so bioactivation is occurring intracellularly. Also, we show that the constituents of the medium do not measurably contribute to oxidative stress or cytotoxicity, as the untransfected 315

cells do not exhibit high levels of toxicity. The observed oxidative damage to macromolecules such as DNA and protein confirms the intracellular nature of the effects. Previous studies in

CHO-K1 cells have shown the DA, L-DOPA and DOPAC can be taken up into the cells, possibly including passive diffusion (Montine et al., 1997). Furthermore, endogenous hPHSs in normal cells are found in the inner and outer membranes of the nuclear envelope (Morita et al.,

1995; Spencer et al., 1998), and hence proximate to DNA, which may enhance the potential for oxidative DNA damage.

Study 3: Human prostaglandin H synthase (hPHS)-1 and hPHS-2 in amphetamine analog bioactivation, DNA oxidation and cytotoxicity.

We have previously shown that the amphetamine analogs METH and MDA (the active metabolite of Ecstasy) may involve their PHS-dependent bioactivation to free radical intermediates that generate ROS and oxidatively damage cellular macromolecules, which can initiate neurotoxicity (Jeng et al., 2006; Jeng and Wells, 2010). However, those studies all have used non-human (rodent, ovine) forms of PHS-1, often as microsomal preparations or in purified enzyme form, which can differentially affect PHS isozyme substrate specificity and enzyme activity. We used CHO-K1 cells either untransfected or stably expressing hPHS-1 or hPHS-2 to investigate hPHS isozyme-dependent oxidative damage and cytotoxicity. Our results provide the first evidence that both human PHSs can bioactivate METH and MDA resulting in enhanced oxidative damage to DNA and cytotoxicity at pharmacologically relevant concentrations. ASA, an irreversible inhibitor of both hPHS-1 and hPHS-2, blocked DNA oxidation and cytotoxicity in both cell lines, corroborating the requirement for PHS activity.

Our METH and MDA data highlight the hypothesis that these compounds on their own, in the 316

absence of dopamine, can increase cytotoxicity and oxidative stress in the context of PHS- catalyzed activation, which may serve as an alternative to cytochromes P450 for neurotoxic free radical generation. This hPHS-dependent mechanism may play a significant role in the neurodegenerative effects of amphetamine analogs, where chronic abuse of these drugs and other environmental factors enhance PHS activity, especially through the induction of PHS-2, and reduce catecholamine levels, thereby enhancing the potential of amphetamine analog bioactivation and free radical generation by PHS. In fact recent studies have shown that DA is not necessary for neurotoxicity of dopaminergic neurons and other regions in the brain, such as the cortex and hippocampus, with low DA and that express PHSs are affected by the amphetamine analogues (O'Callaghan and Miller, 1994; Yuan et al., 2010).

Our results show a concentration-response relationship for hPHS-1 cells consistent with maximal response, and/or saturation or inhibition of hPHS-1 catalyzed bioactivation, at the higher METH concentrations (250 uM-1000 uM). A similar non-linear effect has been reported for MDMA, which may serve as a substrate and potential inhibitor of CYP2D6 in human liver microsomes (Heydari et al., 2004). METH and MDA at 100 uM more fully revealed the roles of AA-dependence and isozyme specificity on concentration-response relationships obscured at higher concentrations. Bioactivation was PHS isozyme-specific, with a greater contribution from hPHS-2 despite the higher PHS activity in hPHS-1-expressing cells. The concentrations tested were similar to plasma concentrations measured in MDMA and METH abusers after recreational doses, and were in the range of extracellular striatal concentrations in rodents after a neurotoxic dosing regimen (10 -100 uM) (de la Torre et al., 2000; Esteban et al., 2001; Melega et al., 1995; Wilson et al., 1996); however, higher concentration were also tested given our short exposure times of 6-24 hr. 317

The cell may attempt to combat METH-initiated oxidative stress and enhance cytoprotective mechanisms (discussed below in study 4), raising the question of whether there are differences in these cellular response mechanisms among the different cell lines. Given the wide range of antioxidative enzymes and repair mechanisms, including DNA repair by oxoguanine glycosylase 1 (OGG1), that are potentially relevant to METH toxicity, it would be difficult to obtain activities for each potentially relevant enzyme. However, our DNA oxidation data (study 3: Fig. 5) show that, in untreated cells (0 uM of METH substrate), the levels of

DNA oxidation across the CHO-K1 and PHS cell lines are similar even with PHS activation by

AA. After METH treatment, even if there were changes in the levels of these detoxifying or repair pathways, they were not sufficient to compensate for the levels of bioactivation in the hPHS cell lines, evidenced by enhanced LDH release and DNA oxidation.

Study 4: Methamphetamine-initiated oxidative stress, neurotoxicity and functional deficits modulated by Nrf2

Our results in Nrf2 knockout mice provide the first evidence that Nrf2 is protective against METH-initiated DNA oxidation, neurotoxicity and functional deficits. We also show that this protective role is evident only if METH treatment is preceded by exposure to oxidative stress sufficiently earlier to allow time for increased Nrf2-dependent gene expression and induction of cytoprotective proteins. This mechanism is important not only in the striatum, but also in the olfactory bulb where neurotoxicity occurs.

The Nrf2-dependent effects observed herein were shown to be independent of changes in drug metabolism or the induction of hyperthermia, two of the most commonly considered confounding issues. Our results provide insights into potential risk factors not only in cases of 318

repeated drug abuse involving amphetamines (e.g. METH, ecstasy), but also for the therapeutic use of amphetamine and related drugs for attention deficit hyperactivity syndrome and related conditions. The doses of METH used in our study give concentrations that are similar to plasma concentrations of METH in humans after chronic use in the range 0.176- 1.743 mg/L (176-1743 ng/mL) (Wilson et al., 1996); however, concentrations in rats are usually on average 10 times higher in the brain versus plasma (Melega et al., 1995). These intracellular concentrations may also be elevated as active transporters may concentrate these drugs inside the neuronal terminal and in the brains of tolerant abusers during high-dose binges.

About one-half of METH in humans is excreted unchanged, and the remainder undergoes CYP2D6-catalyzed N-demethylation to amphetamine, which can then be hydroxylated. The GSH-derived conjugated metabolites are low in urine. Our concern was primarily about potential Nrf2-dependent differences in liver metabolism, which could have confounded our interpretation of the results. Our results excluded this possibility. It remains possible that the lack of some Nrf2-dependent genes in the brain, but not evident in liver, might contribute to the observations.

Our study found that the temporal profile and magnitude of Nrf2-enhanced transcription of cytoprotective genes varied with both the gene and brain region analyzed, suggesting Nrf2 may be an important determinant of regional susceptibility. While our studies only assessed

Nrf2-mediated activation as assessed by mRNA levels, further studies will be needed to determine the activities of the various Nrf2-regulated proteins in each brain region. Additional potential determinants of selectivity for drugs are a localized imbalance in key ROS-relevant pathways of drug bioactivation and ROS formation versus protective levels of antioxidants (e.g. vitamin E) and antioxidative enzymes and, in the case of DNA damage, pathways of DNA 319

repair. Other determinants could include tissue- or cell-selective differences in influx or efflux transporters for METH. There may be cell-specific activation of Nrf2, as with other neurotoxins where astrocytes show increased levels of Nrf2-regulated proteins, and especially GSH and its related synthetic enzymes. Since we only looked regionally for the Nrf2 activation, our studies could not distinguish between different cellular types. However, the susceptibility of the Nrf2 mice may be due in part to a lack of detoxification within astrocytes and/or neurons. So even though METH is concentrated in the nerve terminal by DAT, and this is the site believed to generate the most ROS, our data in the OB with low doses affecting the GABA system highlights the possibility that other cells and other neuronal types may activate Nrf2.

In our study, TH and DAT levels to assess nerve terminal function, GFAP expression, and free radical formation via DNA oxidation were assessed as molecular markers of METH neurotoxicity in the striatum and olfactory bulb. Regional differences in the changes in these markers were evident, in that TH and DAT were decreased in the striatum but not in the olfactory bulb while GFAP was increased in both regions. Whether these changes are irreversible and represent actual degeneration was not assessed in our study (i.e. molecular markers assessed for up to one week after dosing) however, the functional outcomes measured were present for a least one month after dosing with higher deficits in the Nrf2 deficient mice.

This suggests that other markers of neurotoxicity may be necessary to fully evaluate the effects of Nrf2 on METH toxicities, including other neurotransmitter systems (we assessed GABA in the olfactory bulb), effects on myelin, microglia, astrocytes, long-term potentiation and oxidation of proteins and lipids, all of which may lead to malfunctioning brain cells and brain toxicities (Thiels et al., 2000; Bowyer et al., 2008; Melo et al., 2008).

We believe that DNA oxidation is not simply a marker of oxidative stress, but is a 320

molecular lesion that can lead to gene modifications that can disrupt transcription, translation and replication, and can ultimately lead to genomic instability and altered cellular function and/or cell death (Allen and Tresini, 2000; Hailer-Morrison et al., 2003). In related fields, 8- oxodG is a mutagenic lesion involved in the initiation of cancer, and our studies of pregnant

OGG1 knockout mice exposed in utero to METH indicate that this lesion also plays a non- mutational role in the mechanism of neurodevelopmental deficits in repair-deficient offspring

(McCallum et al., 2011; Wong et al., 2008). The OGG1 knockout mice exhibited higher oxidative DNA brain damage as early as 4 hr after in utero exposure to METH but postnatally had no changes in TH or TUNEL staining.

We recently revised our method for DNA oxidation, replacing the traditional phenol

DNA extraction with the chaotropic NaI method (Ravanat et al., 2002), and adding the antioxidant desferoxamine. Using these modifications, our results for endogenous DNA oxidation are in the femtomolar range, consistent with published data for this modified approach

(Ravanat et al., 2002). Extracted DNA was analyzed by HPLC with electrochemical detection for the oxidized base 8-oxo-dG, as described previously (Winn and Wells, 1995). Several

METH studies have shown the generation of reactive radicals that are capable of causing mitochondrial dysfunction, reduced energy metabolism and eventual cell death (Imam et al.,

2001; Pubill et al., 2005). The increase in DA-initiated ROS is thought to occur subsequent to the oxidation of the massive amount of DA released in the brain after exposure to METH (Cadet and Brannock, 1998). Further redox cycling of dopamine quinone formed during DA catabolism would enhance the concentration of ROS within DA terminals. These series of events could cause the degeneration of nerve terminals through membrane destabilization or through changes in calcium homeostasis (Cadet and Brannock, 1998). However, recent studies 321

have shown that neurotoxicity can occur in the absence of DA, in regions where DA levels are low i.e. the cortex and hippocampus, and in other neurotransmitters systems such as SE and

GABA and suggests a potential role for PHS bioactivation and Nrf2-mediated protection that extends beyond dopaminergic effects (Burrows and Meshul, 1999; Deng et al., 2007; Jayanthi et al., 2004; Jayanthi et al., 2009; Kish et al., 2009; O'Callaghan and Miller, 1994; Yuan et al.,

2010).

Olfactory neurotoxicity can occur with low doses of METH and involves changes to

GABAergic markers, yet is independent of changes to dopaminergic nerve terminal markers in other brain regions associated with higher METH doses. Future studies will be needed to determine whether the actual concentrations of DA or other neurotransmitters may be affected and act together to contribute to olfactory deficits. It also would be worthwhile to determine whether this cognitive deficit is due in part to METH effects in the cortex and hippocampus, which play a role in cognitive function. Our studies provide evidence for METH-initiated long- term deficits in olfactory discrimination, which appear to constitute a more sensitive functional biomarker of long-term neurodegeneration in mice than tests assessing only motor function.

Higher doses of METH were necessary to cause motor coordination deficits as measured with the rotarod apparatus, while no changes were observed in spontaneous motor activity tests.

Of importance are the specific effects in the brain that may potentially increase the susceptibility of METH abusers to neurodegenerative diseases and behavioural abnormalities.

Most METH studies have focused on its neurotoxicity related to dopaminergic systems present predominantly in striatum and substantia nigra, which are regions important in PD. This disease is characterized by symptoms of motor dysfunction such as tremor, rigidity and the inability to initiate movement. However, there are a number of non-motor dysfunctions affecting mood, 322

cognition, sleeping patterns and the sense of smell (olfactory deficits) (Berendse et al., 2001;

Doty et al., 1988; Factor et al., 1995). PD has been associated with the death of dopaminergic nerve cells, where an 80% loss of dopaminergic neurons is required for the motor symptoms of the disease to emerge (Berendse et al., 2001; Doty et al., 1988; Factor et al., 1995). However, with high doses of METH abuse, parkinsonian symptoms of gross motor deficits are not observed however there may be deficits in fine motor reactivity (Scott et al., 2007).

The olfactory bulb is the major structure within the brain responsible for olfactory processing, and also is involved with certain forms of learning and memory (Brennan and

Keverne, 1997; Gheusi et al., 2000). Proper function in these underlying pathways is important for odor detection, habituation and discrimination. Olfactory deficits can precede motor deficits by 4 years in men (Berendse et al., 2001; Ross et al., 2008) and, while the mechanism is unknown, the studies of the olfactory bulbs of PD patients are inconsistent with some suggesting no depletion of dopamine or a decrease in TH immunoreactivity, suggesting that hypoosmia in

PD patients may not be due to malfunction of dopaminergic neurons in the olfactory bulb

(Huisman et al., 2008). On the other hand, studies by the same group have shown that TH immunoreactivity is twice the amount in those PD patients with hypoosmia versus age-matched controls, suggesting that high DA levels might lead to olfactory deficits, but DA concentrations were not measured (Huisman et al., 2004). However, therapeutic treatment with L-DOPA, which bypasses TH, does not improve olfactory deficits in PD (Doty et al., 1992), suggesting that hypoosmia in these patients does not involve the DA pathway. In humans, olfactory discrimination declines with age and has been associated with other neurodegenerative disease such as HD and AD (Albers et al., 2006; Kaneda et al., 2000; Larsson et al., 2006; Talamo et al., 1989), which do not necessarily involve dysfunctional DA pathways. 323

Olfactory function is important in animal courtship, mating, and reproduction, so olfactory bulbectomy has been used in animal models of depression because these animals show neurovegetative signs (Song and Leonard, 2005). In mice, olfactory discrimination deficits are evident in DAT and D2 receptor KO mice (Tillerson et al., 2006), and more recently in

VMAT2-deficient mice (Taylor et al., 2009). Interestingly, studies have shown that DAT plays an important role in METH neurotoxicity as DAT knockout mice are less susceptible than wild- type controls to METH-initiated changes in markers of nerve terminal degeneration (Fumagalli et al., 1998); however, olfactory discrimination was not evaluated in these DAT KO mice, which show this deficit even in the absence of any neurotoxin.

Studies with METH-treated rats have shown that METH can decrease GABAergic markers in the striatum within a week post-treatment, and can cause olfactory deficits (Burrows and Meshul, 1999; Jayanthi et al., 2004; O'Dell et al., 2011). Olfactory bulbs of METH-treated mice have shown decreases in DA concentrations and TH-like immunostaining with increases in

TUNEL-labeled OB neurons; however, olfactory discrimination deficits were not assessed

(Deng et al., 2007). A number of studies have linked olfactory function to the GABAergic systems within the olfactory bulb, where aging and/or decreased neurogenesis led to decreased new GCL GABA neurons, which correlated with OD deficits (Abraham et al., 2010; Enwere et al., 2004; Gheusi et al., 2000). Such studies provide potential mechanisms and behavioural studies to be considered in humans as currently studies evaluating olfactory discrimination deficits in METH users are not present. OD studies can be evaluated in METH users that do not intranasally abuse METH by using testing regimens such as the odor identification and discrimination parts of the Sniffin' Sticks battery currently used to assess OD deficits in

Parkinson's disease patients (Boesveldt et al., 2008). 324

Study 5: Developmental role of nuclear factor-E2-related factor 2 (Nrf2) in protection against methamphetamine fetal toxicity and postnatal functional deficits in Nrf2-deficient mice.

The viability and normal phenotype of knockout mice lacking Nrf2 has led to the general perception that Nrf2 plays a limited if not negligible role in normal mouse development.

However, this assumption has not been evaluated under conditions of oxidative stress, where

Nrf2-mediated expression of proteins that mitigate the effects of ROS and their oxidative damage to DNA may protect the developing fetus from adverse structural and functional developmental consequences. In our study of pregnant +/- Nrf2 dams bred with +/- males, the

ROS-initiating teratogen METH was used as both an initial activator of Nrf2, and in a subsequent exposure as a neurodevelopmental toxin, as would occur in drug abusers using

METH repeatedly during pregnancy. Prior exposure to a Nrf2 activator is essential in testing the developmental role of Nrf2, which can protect against oxidative stress by enhancing the expression of a battery of ROS-protective proteins. Our studies are the first to show that fetal

Nrf2 is a critical transcription factor in the expression of an array of proteins that protect the fetus from oxidative DNA damage, fetal weight loss, lethality and postnatal neurodevelopmental deficits resulting from oxidative stress caused by in utero exposure to METH. Several of these

METH-initiated Nrf2-mediated effects were gender-dependent, with greater fetal weight loss and lethality in female -/- Nrf2 fetuses, and greater postnatal olfactory and motor coordination deficits respectively in female and male -/- Nrf2 offspring. Nrf2-deficient -/- knockout fetuses, unlike their +/+ littermates, were unable to enhance the transcription of Nrf2-regulated genes like HO-1, NQO1 and OGG1 that protect against oxidative stress and repair oxidatively damaged DNA. Our results also provide the first evidence that in utero exposure to a relatively 325

low dose of METH causes postnatal olfactory deficits measured by odor detection, habituation and discrimination, which appear to constitute a sensitive functional biomarker of METH- initiated long-term postnatal neurodevelopmental deficits. A loss of staining for GAD-65, but not the dopamine transporter, in the olfactory bulb of METH-exposed offspring suggests that this deficit is at least in part due to a loss of GABAergic function, as was observed in the adult

METH-exposed mice. The observation that fetal toxicity and postnatal olfactory and motor coordination deficits were modulated by fetal Nrf2 activation may be particularly significant in pregnant women exposed to METH or other ROS-generating neurotoxins.

3.2 FINAL THOUGHTS

As new polymorphisms of PHSs and Nrf2 and their functional impact are identified and characterized, along with environmental modulators of these proteins, our data for the roles of

PHS-1, PHS-2 and Nrf2 in modifying oxidative stress cellular responses will contribute to a better understanding of not only regional susceptibility, but also individual risks for neurodegenerative effects and fetal damage and neurodevelopmental deficits caused by exogenous and endogenous initiators of oxidative stress. In such cases, variations in drug abuse patterns, dietary exposures and genetic predisposition may alter the induction and duration of

PHS and Nrf2 activation, thereby varying the susceptibility to ROS-initiating compounds.

Hence, genetic and environmental differences in PHS and Nrf2 may constitute important determinants of neurodegenerative and developmental risk from oxidative stress, which may be reduced by novel therapeutic strategies that limit PHS activity and/or enhance Nrf2-mediated protective responses.

326

3.3 FUTURE STUDIES

The role of PHSs in bioactivation should be evaluated for other neurotransmitter pathways including SE and NE their precursors and metabolites. Furthermore, psychoactive drugs can be investigated in terms of bioactivation by PHSs and also their potential for Nrf2 activation. For example, a potential mechanism of methylphenidate (Ritalin) free radical intermediate formation, ROS generation and activation of Nrf2. Further studies in vivo with mice selectively expressing hPHS-1 and hPHS-2, combined with specific regional/cellular knockdown of PHSs, may provide insights into novel mechanisms of bioactivation; for example, in neurons versus astrocytes or microglia.

The role of Nrf2 could also be evaluated in mice for other substrates like MDMA and

MDA. These studies should be extended to other species such as rats to elucidate whether differences in Nrf2 activation across species may account for differing sensitivities to neurotoxicity. Likewise, endpoints of toxicity need to be expanded to cover other cell types and other brain regional susceptibilities and associated functional deficits.

The role of Nrf2 in development and fetal toxicities warrants further investigation, including the red phenotype and edema produced by in utero METH exposure, the mechanisms for which remain to be established. The red phenotype could reflect vasodilation or alterations in haemoglobin, while the fetal edema could result from free radical-mediated damage to vascular membranes. Likewise, better molecular and biochemical markers for fetal brain toxicities are needed to elucidate the mechanisms of olfactory deficits, hypoactivity and/or anxiety that are manifested postnatally. 327

CHAPTER 4: REFERENCES

Abbott, N. J., Patabendige, A. A., Dolman, D. E., Yusof, S. R. and Begley, D. J. (2010). Structure and function of the blood-brain barrier. Neurobiol Dis 37(1): 13-25. Abdel-Halim, M. S., Hamberg, M., Sjoquist, B. and Anggard, E. (1977). Identification of prostaglandin D2 as a major prostaglandin in homogenates of rat brain. Prostaglandins 14(4): 633-643. Abraham, N. M., Egger, V., Shimshek, D. R., Renden, R., Fukunaga, I., Sprengel, R., Seeburg, P. H., Klugmann, M., Margrie, T. W., Schaefer, A. T. and Kuner, T. (2010). Synaptic inhibition in the olfactory bulb accelerates odor discrimination in mice. Neuron 65(3): 399-411. Abramov, A. Y., Jacobson, J., Wientjes, F., Hothersall, J., Canevari, L. and Duchen, M. R. (2005). Expression and modulation of an NADPH oxidase in mammalian astrocytes. J Neurosci 25(40): 9176-9184. Acuff-Smith, K. D., Schilling, M. A., Fisher, J. E. and Vorhees, C. V. (1996). Stage-specific effects of prenatal d-methamphetamine exposure on behavioral and eye development in rats. Neurotoxicol Teratol 18(2): 199-215. Adak, S., Bandyopadhyay, U., Bandyopadhyay, D. and Banerjee, R. K. (1998). Mechanism of horseradish peroxidase catalyzed epinephrine oxidation: obligatory role of endogenous O2- and H2O2. Biochemistry 37(48): 16922-16933. Aisen, P. S., Schafer, K. A., Grundman, M., Pfeiffer, E., Sano, M., Davis, K. L., Farlow, M. R., Jin, S., Thomas, R. G. and Thal, L. J. (2003). Effects of rofecoxib or naproxen vs placebo on Alzheimer disease progression: a randomized controlled trial. JAMA 289(21): 2819-2826. Alam, J., Killeen, E., Gong, P., Naquin, R., Hu, B., Stewart, D., Ingelfinger, J. R. and Nath, K. A. (2003). Heme activates the heme oxygenase-1 gene in renal epithelial cells by stabilizing Nrf2. Am J Physiol Renal Physiol 284(4): F743-752. Alam, J., Stewart, D., Touchard, C., Boinapally, S., Choi, A. M. and Cook, J. L. (1999). Nrf2, a Cap'n'Collar transcription factor, regulates induction of the heme oxygenase-1 gene. J Biol Chem 274(37): 26071-26078. Alanko, J., Riutta, A., Holm, P., Mucha, I., Vapaatalo, H. and Metsa-Ketela, T. (1999). Modulation of arachidonic acid metabolism by phenols: relation to their structure and antioxidant/prooxidant properties. Free Radic Biol Med 26(1-2): 193-201. Albers, M. W., Tabert, M. H. and Devanand, D. P. (2006). Olfactory dysfunction as a predictor of neurodegenerative disease. Curr Neurol Neurosci Rep 6(5): 379-386. 328

Allen, R. G. and Tresini, M. (2000). Oxidative stress and gene regulation. Free Radic Biol Med 28(3): 463-499. Almer, G., Guegan, C., Teismann, P., Naini, A., Rosoklija, G., Hays, A. P., Chen, C. and Przedborski, S. (2001). Increased expression of the pro-inflammatory enzyme cyclooxygenase-2 in amyotrophic lateral sclerosis. Ann Neurol 49(2): 176-185. Almer, G., Kikuchi, H., Teismann, P. and Przedborski, S. (2006). Is prostaglandin E(2) a pathogenic factor in amyotrophic lateral sclerosis? Ann Neurol 59(6): 980-983. Almer, G., Teismann, P., Stevic, Z., Halaschek-Wiener, J., Deecke, L., Kostic, V. and Przedborski, S. (2002). Increased levels of the pro-inflammatory prostaglandin PGE2 in CSF from ALS patients. Neurology 58(8): 1277-1279. Anden, N. E., Hfuxe, K., Hamberger, B. and Hokfelt, T. (1966). A quantitative study on the nigro-neostriatal dopamine neuron system in the rat. Acta Physiol Scand 67(3): 306-312. Andreasson, K. (2010). Emerging roles of PGE2 receptors in models of neurological disease. Prostaglandins Other Lipid Mediat 91(3-4): 104-112. Andreasson, K. I., Savonenko, A., Vidensky, S., Goellner, J. J., Zhang, Y., Shaffer, A., Kaufmann, W. E., Worley, P. F., Isakson, P. and Markowska, A. L. (2001). Age- dependent cognitive deficits and neuronal apoptosis in cyclooxygenase-2 transgenic mice. J Neurosci 21(20): 8198-8209. Anglin, M. D., Burke, C., Perrochet, B., Stamper, E. and Dawud-Noursi, S. (2000). History of the methamphetamine problem. J Psychoactive Drugs 32(2): 137-141. Appleby, S. B., Ristimaki, A., Neilson, K., Narko, K. and Hla, T. (1994). Structure of the human cyclo-oxygenase-2 gene. Biochem J 302 ( Pt 3): 723-727. Arisawa, T., Tahara, T., Shibata, T., Nagasaka, M., Nakamura, M., Kamiya, Y., Fujita, H., Hasegawa, S., Takagi, T., Wang, F. Y., Hirata, I. and Nakano, H. (2007). The relationship between Helicobacter pylori infection and promoter polymorphism of the Nrf2 gene in chronic gastritis. Int J Mol Med 19(1): 143-148. Arisawa, T., Tahara, T., Shibata, T., Nagasaka, M., Nakamura, M., Kamiya, Y., Fujita, H., Yoshioka, D., Okubo, M., Hirata, I. and Nakano, H. (2008a). Nrf2 gene promoter polymorphism and gastric carcinogenesis. Hepatogastroenterology 55(82-83): 750-754. Arisawa, T., Tahara, T., Shibata, T., Nagasaka, M., Nakamura, M., Kamiya, Y., Fujita, H., Yoshioka, D., Okubo, M., Sakata, M., Wang, F. Y., Hirata, I. and Nakano, H. (2008b). Nrf2 gene promoter polymorphism is associated with ulcerative colitis in a Japanese population. Hepatogastroenterology 55(82-83): 394-397. Arria, A. M., Derauf, C., Lagasse, L. L., Grant, P., Shah, R., Smith, L., Haning, W., Huestis, M., 329

Strauss, A., Della Grotta, S., Liu, J. and Lester, B. (2006). Methamphetamine and other substance use during pregnancy: preliminary estimates from the Infant Development, Environment, and Lifestyle (IDEAL) study. Matern Child Health J 10(3): 293-302. Arriagada, C., Paris, I., Sanchez de las Matas, M. J., Martinez-Alvarado, P., Cardenas, S., Castaneda, P., Graumann, R., Perez-Pastene, C., Olea-Azar, C., Couve, E., Herrero, M. T., Caviedes, P. and Segura-Aguilar, J. (2004). On the neurotoxicity mechanism of leukoaminochrome o-semiquinone radical derived from dopamine oxidation: mitochondria damage, necrosis, and hydroxyl radical formation. Neurobiol Dis 16(2): 468-477. Avshalumov, M. V. and Rice, M. E. (2002). NMDA receptor activation mediates hydrogen peroxide-induced pathophysiology in rat hippocampal slices. J Neurophysiol 87(6): 2896-2903. Axt, K. J. and Molliver, M. E. (1991). Immunocytochemical evidence for methamphetamine- induced serotonergic axon loss in the rat brain. Synapse 9(4): 302-313. Babior, B. M., Curnutte, J. T. and Kipnes, B. S. (1975). Pyridine nucleotide-dependent superoxide production by a cell-free system from human granulocytes. J Clin Invest 56(4): 1035-1042. Backlund, M. G., Mann, J. R., Holla, V. R., Buchanan, F. G., Tai, H. H., Musiek, E. S., Milne, G. L., Katkuri, S. and DuBois, R. N. (2005). 15-Hydroxyprostaglandin dehydrogenase is down-regulated in colorectal cancer. J Biol Chem 280(5): 3217-3223. Bader, V., Ran Zhu, X., Lubbert, H. and Stichel, C. C. (2005). Expression of DJ-1 in the adult mouse CNS. Brain Res 1041(1): 102-111. Bagasra, O., Michaels, F. H., Zheng, Y. M., Bobroski, L. E., Spitsin, S. V., Fu, Z. F., Tawadros, R. and Koprowski, H. (1995). Activation of the inducible form of nitric oxide synthase in the brains of patients with multiple sclerosis. Proc Natl Acad Sci U S A 92(26): 12041-12045. Bakhit, C., Morgan, M. E., Peat, M. A. and Gibb, J. W. (1981). Long-term effects of methamphetamine on the synthesis and metabolism of 5-hydroxytryptamine in various regions of the rat brain. Neuropharmacology 20(12A): 1135-1140. Balogun, E., Hoque, M., Gong, P., Killeen, E., Green, C. J., Foresti, R., Alam, J. and Motterlini, R. (2003). Curcumin activates the haem oxygenase-1 gene via regulation of Nrf2 and the antioxidant-responsive element. Biochem J 371(Pt 3): 887-895. Banning, A., Deubel, S., Kluth, D., Zhou, Z. and Brigelius-Flohe, R. (2005). The GI-GPx gene is a target for Nrf2. Mol Cell Biol 25(12): 4914-4923. 330

Baranano, D. E. and Snyder, S. H. (2001). Neural roles for heme oxygenase: contrasts to nitric oxide synthase. Proc Natl Acad Sci U S A 98(20): 10996-11002. Bartels, A. L., Willemsen, A. T., Kortekaas, R., de Jong, B. M., de Vries, R., de Klerk, O., van Oostrom, J. C., Portman, A. and Leenders, K. L. (2008). Decreased blood-brain barrier P-glycoprotein function in the progression of Parkinson's disease, PSP and MSA. J Neural Transm 115(7): 1001-1009. Bartu, A., Dusci, L. J. and Ilett, K. F. (2009). Transfer of methylamphetamine and amphetamine into breast milk following recreational use of methylamphetamine. Br J Clin Pharmacol 67(4): 455-459. Basu, S. (2007). Novel cyclooxygenase-catalyzed bioactive prostaglandin F2alpha from physiology to new principles in inflammation. Med Res Rev 27(4): 435-468. Battelli, M. G., Buonamici, L., Virgili, M., Abbondanza, A. and Contestabile, A. (1998). Simulated ischaemia-reperfusion conditions increase xanthine dehydrogenase and oxidase activities in rat brain slices. Neurochem Int 32(1): 17-21. Beal, M. F., Ferrante, R. J., Henshaw, R., Matthews, R. T., Chan, P. H., Kowall, N. W., Epstein, C. J. and Schulz, J. B. (1995). 3-Nitropropionic acid neurotoxicity is attenuated in copper/zinc superoxide dismutase transgenic mice. J Neurochem 65(2): 919-922. Beckett, A. H. and Rowland, M. (1965). Urinary excretion of methylamphetamine in man. Nature 206(990): 1260-1261. Beckman, J. S., Beckman, T. W., Chen, J., Marshall, P. A. and Freeman, B. A. (1990). Apparent hydroxyl radical production by peroxynitrite: implications for endothelial injury from nitric oxide and superoxide. Proc Natl Acad Sci U S A 87(4): 1620-1624. Bedard, K. and Krause, K. H. (2007). The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology. Physiol Rev 87(1): 245-313. Beetsch, J. W., Park, T. S., Dugan, L. L., Shah, A. R. and Gidday, J. M. (1998). Xanthine oxidase-derived superoxide causes reoxygenation injury of ischemic cerebral endothelial cells. Brain Res 786(1-2): 89-95. Begley, D. J. (2004). ABC transporters and the blood-brain barrier. Curr Pharm Des 10(12): 1295-1312. Beiche, F., Scheuerer, S., Brune, K., Geisslinger, G. and Goppelt-Struebe, M. (1996). Up- regulation of cyclooxygenase-2 mRNA in the rat spinal cord following peripheral inflammation. FEBS Lett 390(2): 165-169. Bendotti, C. and Carri, M. T. (2004). Lessons from models of SOD1-linked familial ALS. Trends Mol Med 10(8): 393-400. 331

Berendse, H. W., Booij, J., Francot, C. M., Bergmans, P. L., Hijman, R., Stoof, J. C. and Wolters, E. C. (2001). Subclinical dopaminergic dysfunction in asymptomatic Parkinson's disease patients' relatives with a decreased sense of smell. Ann Neurol 50(1): 34-41. Berridge, C. W. and Waterhouse, B. D. (2003). The locus coeruleus-noradrenergic system: modulation of behavioral state and state-dependent cognitive processes. Brain Res Brain Res Rev 42(1): 33-84. Berry, C. E. and Hare, J. M. (2004). Xanthine oxidoreductase and cardiovascular disease: molecular mechanisms and pathophysiological implications. J Physiol 555(Pt 3): 589- 606. Betz, A. L. (1985). Identification of hypoxanthine transport and xanthine oxidase activity in brain capillaries. J Neurochem 44(2): 574-579. Bezzi, P., Carmignoto, G., Pasti, L., Vesce, S., Rossi, D., Rizzini, B. L., Pozzan, T. and Volterra, A. (1998). Prostaglandins stimulate calcium-dependent glutamate release in astrocytes. Nature 391(6664): 281-285. Biswas, M. and Chan, J. Y. (2010). Role of Nrf1 in antioxidant response element-mediated gene expression and beyond. Toxicol Appl Pharmacol 244(1): 16-20. Blobaum, A. L. and Marnett, L. J. (2007). Structural and functional basis of cyclooxygenase inhibition. J Med Chem 50(7): 1425-1441. Block, M. L., Zecca, L. and Hong, J. S. (2007). Microglia-mediated neurotoxicity: uncovering the molecular mechanisms. Nat Rev Neurosci 8(1): 57-69. Boesveldt, S., Verbaan, D., Knol, D. L., Visser, M., van Rooden, S. M., van Hilten, J. J. and Berendse, H. W. (2008). A comparative study of odor identification and odor discrimination deficits in Parkinson's disease. Mov Disord 23(14): 1984-1990. Boie, Y., Rushmore, T. H., Darmon-Goodwin, A., Grygorczyk, R., Slipetz, D. M., Metters, K. M. and Abramovitz, M. (1994). Cloning and expression of a cDNA for the human prostanoid IP receptor. J Biol Chem 269(16): 12173-12178. Boie, Y., Sawyer, N., Slipetz, D. M., Metters, K. M. and Abramovitz, M. (1995). Molecular cloning and characterization of the human prostanoid DP receptor. J Biol Chem 270(32): 18910-18916. Boileau, I., Rusjan, P., Houle, S., Wilkins, D., Tong, J., Selby, P., Guttman, M., Saint-Cyr, J. A., Wilson, A. A. and Kish, S. J. (2008). Increased vesicular monoamine transporter binding during early abstinence in human methamphetamine users: Is VMAT2 a stable dopamine neuron biomarker? J Neurosci 28(39): 9850-9856. 332

Bolton, J. L., Trush, M. A., Penning, T. M., Dryhurst, G. and Monks, T. J. (2000). Role of quinones in toxicology. Chem Res Toxicol 13(3): 135-160. Bost, R. O., Kemp, P. and Hnilica, V. (1989). Tissue distribution of methamphetamine and amphetamine in premature infants. J Anal Toxicol 13(5): 300-302. Boveris, A., Cadenas, E. and Stoppani, A. O. (1976). Role of ubiquinone in the mitochondrial generation of hydrogen peroxide. Biochem J 156(2): 435-444. Boveris, A., Oshino, N. and Chance, B. (1972). The cellular production of hydrogen peroxide. Biochem J 128(3): 617-630. Bowerman, B., Eaton, B. A. and Priess, J. R. (1992). skn-1, a maternally expressed gene required to specify the fate of ventral blastomeres in the early C. elegans embryo. Cell 68(6): 1061-1075. Bowyer, J. F., Davies, D. L., Schmued, L., Broening, H. W., Newport, G. D., Slikker, W., Jr. and Holson, R. R. (1994). Further studies of the role of hyperthermia in methamphetamine neurotoxicity. J Pharmacol Exp Ther 268(3): 1571-1580. Bowyer, J. F., Robinson, B., Ali, S. and Schmued, L. C. (2008). Neurotoxic-related changes in tyrosine hydroxylase, microglia, myelin, and the blood-brain barrier in the caudate- putamen from acute methamphetamine exposure. Synapse 62(3): 193-204. Bowyer, J. F., Scallet, A. C., Holson, R. R., Lipe, G. W., Slikker, W., Jr. and Ali, S. F. (1991). Interactions of MK-801 with glutamate-, glutamine- and methamphetamine-evoked release of [3H]dopamine from striatal slices. J Pharmacol Exp Ther 257(1): 262-270. Boyd, J. A. and Eling, T. E. (1985). Metabolism of aromatic amines by prostaglandin H synthase. Environ Health Perspect 64: 45-51. Breder, C. D., Dewitt, D. and Kraig, R. P. (1995). Characterization of inducible cyclooxygenase in rat brain. J Comp Neurol 355(2): 296-315. Breder, C. D. and Saper, C. B. (1996). Expression of inducible cyclooxygenase mRNA in the mouse brain after systemic administration of bacterial lipopolysaccharide. Brain Res 713(1-2): 64-69. Bredt, D. S., Hwang, P. M. and Snyder, S. H. (1990). Localization of nitric oxide synthase indicating a neural role for nitric oxide. Nature 347(6295): 768-770. Brennan, P. A. and Keverne, E. B. (1997). Neural mechanisms of mammalian olfactory learning. Prog Neurobiol 51(4): 457-481. Bresalier, R. S., Sandler, R. S., Quan, H., Bolognese, J. A., Oxenius, B., Horgan, K., Lines, C., Riddell, R., Morton, D., Lanas, A., Konstam, M. A. and Baron, J. A. (2005). Cardiovascular events associated with rofecoxib in a colorectal adenoma 333

chemoprevention trial. N Engl J Med 352(11): 1092-1102. Brown, G. C. (1999). Nitric oxide and mitochondrial respiration. Biochim Biophys Acta 1411(2-3): 351-369. Brown, G. C. (2007). Mechanisms of inflammatory neurodegeneration: iNOS and NADPH oxidase. Biochem Soc Trans 35(Pt 5): 1119-1121. Brown, S. L., Sekhar, K. R., Rachakonda, G., Sasi, S. and Freeman, M. L. (2008). Activating transcription factor 3 is a novel repressor of the nuclear factor erythroid-derived 2- related factor 2 (Nrf2)-regulated stress pathway. Cancer Res 68(2): 364-368. Browne, S. E., Ferrante, R. J. and Beal, M. F. (1999). Oxidative stress in Huntington's disease. Brain Pathol 9(1): 147-163. Bruijn, L. I., Houseweart, M. K., Kato, S., Anderson, K. L., Anderson, S. D., Ohama, E., Reaume, A. G., Scott, R. W. and Cleveland, D. W. (1998). Aggregation and motor neuron toxicity of an ALS-linked SOD1 mutant independent from wild-type SOD1. Science 281(5384): 1851-1854. Bryant, C. E., Appleton, I. and Mitchell, J. A. (1998). Vascular endothelial growth factor upregulates constitutive cyclooxygenase 1 in primary bovine and human endothelial cells. Life Sci 62(24): 2195-2201. Buettner, G. R. (1987). Spin trapping: ESR parameters of spin adducts. Free Radic Biol Med 3(4): 259-303. Buettner, G. R. and Oberley, L. W. (1978). Considerations in the spin trapping of superoxide and hydroxyl radical in aqueous systems using 5,5-dimethyl-1-pyrroline-1-oxide. Biochem Biophys Res Commun 83(1): 69-74. Burrows, K. B. and Meshul, C. K. (1999). High-dose methamphetamine treatment alters presynaptic GABA and glutamate immunoreactivity. Neuroscience 90(3): 833-850. Burton, N. C., Kensler, T. W. and Guilarte, T. R. (2006). In vivo modulation of the Parkinsonian phenotype by Nrf2. Neurotoxicology 27(6): 1094-1100. Cadenas, E. and Davies, K. J. (2000). Mitochondrial free radical generation, oxidative stress, and aging. Free Radic Biol Med 29(3-4): 222-230. Cadet, J. L. and Brannock, C. (1998). Free radicals and the pathobiology of brain dopamine systems. Neurochem Int 32(2): 117-131. Cadet, J. L. and Krasnova, I. N. (2009). Molecular bases of methamphetamine-induced neurodegeneration. Int Rev Neurobiol 88: 101-119. Cadet, J. L., Thiriet, N. and Jayanthi, S. (2001). Involvement of free radicals in MDMA-induced neurotoxicity in mice. Annales de Medecine Interne 152(Suppl 3): IS57-59. 334

Caldwell, J., Dring, L. G. and Williams, R. T. (1972). Metabolism of ( 14 C)methamphetamine in man, the guinea pig and the rat. Biochem J 129(1): 11-22. Caligiuri, M. P. and Buitenhuys, C. (2005). Do preclinical findings of methamphetamine- induced motor abnormalities translate to an observable clinical phenotype? Neuropsychopharmacology 30(12): 2125-2134. Calkins, M. J., Jakel, R. J., Johnson, D. A., Chan, K., Kan, Y. W. and Johnson, J. A. (2005). Protection from mitochondrial complex II inhibition in vitro and in vivo by Nrf2- mediated transcription. Proc Natl Acad Sci U S A 102(1): 244-249. Camarero, J., Sanchez, V., O'Shea, E., Green, A. R. and Colado, M. I. (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(5): 961-972. Candelario-Jalil, E., de Oliveira, A. C., Graf, S., Bhatia, H. S., Hull, M., Munoz, E. and Fiebich, B. L. (2007). Resveratrol potently reduces prostaglandin E2 production and free radical formation in lipopolysaccharide-activated primary rat microglia. J Neuroinflammation 4: 25. Canet-Aviles, R. M., Wilson, M. A., Miller, D. W., Ahmad, R., McLendon, C., Bandyopadhyay, S., Baptista, M. J., Ringe, D., Petsko, G. A. and Cookson, M. R. (2004). The Parkinson's disease protein DJ-1 is neuroprotective due to cysteine-sulfinic acid-driven mitochondrial localization. Proc Natl Acad Sci U S A 101(24): 9103-9108. Cannazza, G., Di Stefano, A., Mosciatti, B., Braghiroli, D., Baraldi, M., Pinnen, F., Sozio, P., Benatti, C. and Parenti, C. (2005). Detection of levodopa, dopamine and its metabolites in rat striatum dialysates following peripheral administration of L-DOPA prodrugs by mean of HPLC-EC. J Pharm Biomed Anal 36(5): 1079-1084. Cao, C., Matsumura, K., Yamagata, K. and Watanabe, Y. (1996). Endothelial cells of the rat brain vasculature express cyclooxygenase-2 mRNA in response to systemic interleukin-1 beta: a possible site of prostaglandin synthesis responsible for fever. Brain Res 733(2): 263-272. Capone, M. L., Tacconelli, S., Di Francesco, L., Sacchetti, A., Sciulli, M. G. and Patrignani, P. (2007). Pharmacodynamic of cyclooxygenase inhibitors in humans. Prostaglandins Other Lipid Mediat 82(1-4): 85-94. Carr, A. and Frei, B. (1999). Does vitamin C act as a pro-oxidant under physiological conditions? FASEB J 13(9): 1007-1024. Carrasco, E., Casper, D. and Werner, P. (2005). Dopaminergic neurotoxicity by 6-OHDA and 335

MPP+: differential requirement for neuronal cyclooxygenase activity. J Neurosci Res 81(1): 121-131. Carriedo, S. G., Sensi, S. L., Yin, H. Z. and Weiss, J. H. (2000). AMPA exposures induce mitochondrial Ca(2+) overload and ROS generation in spinal motor neurons in vitro. J Neurosci 20(1): 240-250. Casalino, E., Calzaretti, G., Landriscina, M., Sblano, C., Fabiano, A. and Landriscina, C. (2007). The Nrf2 transcription factor contributes to the induction of alpha-class GST isoenzymes in liver of acute cadmium or manganese intoxicated rats: comparison with the toxic effect on NAD(P)H:quinone reductase. Toxicology 237(1-3): 24-34. Casarett, L. J., Doull, J. and Klaassen, C. D. (2008). Casarett and Doull's toxicology : the basic science of poisons. New York, McGraw-Hill. Cass, W. A. and Manning, M. W. (1999). Recovery of presynaptic dopaminergic functioning in rats treated with neurotoxic doses of methamphetamine. J Neurosci 19(17): 7653-7660. Cavalieri, E. L., Li, K. M., Balu, N., Saeed, M., Devanesan, P., Higginbotham, S., Zhao, J., Gross, M. L. and Rogan, E. G. (2002a). Catechol ortho-quinones: the electrophilic compounds that form depurinating DNA adducts and could initiate cancer and other diseases. Carcinogenesis 23(6): 1071-1077. Cavalieri, E. L., Rogan, E. G. and Chakravarti, D. (2002b). Initiation of cancer and other diseases by catechol ortho-quinones: a unifying mechanism. Cell Mol Life Sci 59(4): 665-681. Chae, S. W., Kang, B. Y., Hwang, O. and Choi, H. J. (2008). Cyclooxygenase-2 is involved in oxidative damage and alpha-synuclein accumulation in dopaminergic cells. Neurosci Lett 436(2): 205-209. Chan, J. Y., Cheung, M. C., Moi, P., Chan, K. and Kan, Y. W. (1995). Chromosomal localization of the human NF-E2 family of bZIP transcription factors by fluorescence in situ hybridization. Hum Genet 95(3): 265-269. Chan, J. Y., Han, X. L. and Kan, Y. W. (1993a). Cloning of Nrf1, an NF-E2-related transcription factor, by genetic selection in yeast. Proc Natl Acad Sci U S A 90(23): 11371-11375. Chan, J. Y., Han, X. L. and Kan, Y. W. (1993b). Isolation of cDNA encoding the human NF-E2 protein. Proc Natl Acad Sci U S A 90(23): 11366-11370. Chan, J. Y., Kwong, M., Lu, R., Chang, J., Wang, B., Yen, T. S. and Kan, Y. W. (1998). Targeted disruption of the ubiquitous CNC-bZIP transcription factor, Nrf-1, results in anemia and embryonic lethality in mice. EMBO J 17(6): 1779-1787. 336

Chan, K., Han, X. D. and Kan, Y. W. (2001). An important function of Nrf2 in combating oxidative stress: detoxification of acetaminophen. Proc Natl Acad Sci U S A 98(8): 4611-4616. Chan, K., Lu, R., Chang, J. C. and Kan, Y. W. (1996). NRF2, a member of the NFE2 family of transcription factors, is not essential for murine erythropoiesis, growth, and development. Proc Natl Acad Sci U S A 93(24): 13943-13948. Chan, P., Di Monte, D. A., Luo, J. J., DeLanney, L. E., Irwin, I. and Langston, J. W. (1994). Rapid ATP loss caused by methamphetamine in the mouse striatum: relationship between energy impairment and dopaminergic neurotoxicity. J Neurochem 62(6): 2484- 2487. Chance, B., Sies, H. and Boveris, A. (1979). Hydroperoxide metabolism in mammalian organs. Physiol Rev 59(3): 527-605. Chandrasekharan, N. V., Dai, H., Roos, K. L., Evanson, N. K., Tomsik, J., Elton, T. S. and Simmons, D. L. (2002). COX-3, a cyclooxygenase-1 variant inhibited by acetaminophen and other analgesic/antipyretic drugs: cloning, structure, and expression. Proc Natl Acad Sci U S A 99(21): 13926-13931. Chang, J. W., Coleman, P. D. and O'Banion, M. K. (1996). Prostaglandin G/H synthase-2 (cyclooxygenase-2) mRNA expression is decreased in Alzheimer's disease. Neurobiol Aging 17(5): 801-808. Chang, L., Smith, L. M., LoPresti, C., Yonekura, M. L., Kuo, J., Walot, I. and Ernst, T. (2004). Smaller subcortical volumes and cognitive deficits in children with prenatal methamphetamine exposure. Psychiatry Res 132(2): 95-106. Chen, C. J., Liao, S. L. and Kuo, J. S. (2000). Gliotoxic action of glutamate on cultured astrocytes. J Neurochem 75(4): 1557-1565. Chen, C. Y., Del Gatto-Konczak, F., Wu, Z. and Karin, M. (1998). Stabilization of interleukin-2 mRNA by the c-Jun NH2-terminal kinase pathway. Science 280(5371): 1945-1949. Chen, P. C., Vargas, M. R., Pani, A. K., Smeyne, R. J., Johnson, D. A., Kan, Y. W. and Johnson, J. A. (2009). Nrf2-mediated neuroprotection in the MPTP mouse model of Parkinson's disease: Critical role for the astrocyte. Proc Natl Acad Sci U S A 106(8): 2933-2938. Chen, W., Pawelek, T. R. and Kulmacz, R. J. (1999). Hydroperoxide dependence and cooperative cyclooxygenase kinetics in prostaglandin H synthase-1 and -2. J Biol Chem 274(29): 20301-20306. Chenais, B., Derjuga, A., Massrieh, W., Red-Horse, K., Bellingard, V., Fisher, S. J. and Blank, 337

V. (2005). Functional and placental expression analysis of the human NRF3 transcription factor. Mol Endocrinol 19(1): 125-137. Cheng, F. C., Kuo, J. S., Shih, Y., Lai, J. S., Ni, D. R. and Chia, L. G. (1993). Simultaneous measurement of serotonin, catecholamines and their metabolites in mouse brain homogenates by high-performance liquid chromatography with a microbore column and dual electrochemical detection. J Chromatogr 615(2): 225-236. Cho, A. K. (1990). Ice: a new dosage form of an old drug. Science 249(4969): 631-634. Cho, A. K., Melega, W. P., Kuczenski, R. and Segal, D. S. (2001). Relevance of pharmacokinetic parameters in animal models of methamphetamine abuse. Synapse 39(2): 161-166. Cho, D. H., Lyu, H. M., Lee, H. B., Kim, P. Y. and Chin, K. (1991). Behavioral teratogenicity of methamphetamine. J Toxicol Sci 16 Suppl 1: 37-49. Choi, S. H., Aid, S. and Bosetti, F. (2009). The distinct roles of cyclooxygenase-1 and -2 in neuroinflammation: implications for translational research. Trends Pharmacol Sci 30(4): 174-181. Choi, S. H., Langenbach, R. and Bosetti, F. (2008). Genetic deletion or pharmacological inhibition of cyclooxygenase-1 attenuate lipopolysaccharide-induced inflammatory response and brain injury. FASEB J 22(5): 1491-1501. Chu, P. W., Seferian, K. S., Birdsall, E., Truong, J. G., Riordan, J. A., Metcalf, C. S., Hanson, G. R. and Fleckenstein, A. E. (2008). Differential regional effects of methamphetamine on dopamine transport. Eur J Pharmacol 590(1-3): 105-110. Chung, A., Lyoo, I. K., Kim, S. J., Hwang, J., Bae, S. C., Sung, Y. H., Sim, M. E., Song, I. C., Kim, J., Chang, K. H. and Renshaw, P. F. (2007). Decreased frontal white-matter integrity in abstinent methamphetamine abusers. Int J Neuropsychopharmacol 10(6): 765-775. Cirrito, J. R., Deane, R., Fagan, A. M., Spinner, M. L., Parsadanian, M., Finn, M. B., Jiang, H., Prior, J. L., Sagare, A., Bales, K. R., Paul, S. M., Zlokovic, B. V., Piwnica-Worms, D. and Holtzman, D. M. (2005). P-glycoprotein deficiency at the blood-brain barrier increases amyloid-beta deposition in an Alzheimer disease mouse model. J Clin Invest 115(11): 3285-3290. Clements, C. M., McNally, R. S., Conti, B. J., Mak, T. W. and Ting, J. P. (2006). DJ-1, a cancer- and Parkinson's disease-associated protein, stabilizes the antioxidant transcriptional master regulator Nrf2. Proc Natl Acad Sci U S A 103(41): 15091-15096. Cohen, J. B., Dickow, A., Horner, K., Zweben, J. E., Balabis, J., Vandersloot, D. and Reiber, C. 338

(2003). Abuse and violence history of men and women in treatment for methamphetamine dependence. Am J Addict 12(5): 377-385. Colado, M. I., Camarero, J., Mechan, A. O., Sanchez, V., Esteban, B., Elliott, J. M. and Green, A. R. (2001). A study of the mechanisms involved in the neurotoxic action of 3,4- methylenedioxymethamphetamine (MDMA, 'ecstasy') on dopamine neurones in mouse brain. Br J Pharmacol 134(8): 1711-1723. Coleman, R. A., Smith, W. L. and Narumiya, S. (1994). International Union of Pharmacology classification of prostanoid receptors: properties, distribution, and structure of the receptors and their subtypes. Pharmacol Rev 46(2): 205-229. Collins, A. R., Lyon, C. J., Xia, X., Liu, J. Z., Tangirala, R. K., Yin, F., Boyadjian, R., Bikineyeva, A., Pratico, D., Harrison, D. G. and Hsueh, W. A. (2009). Age-accelerated atherosclerosis correlates with failure to upregulate antioxidant genes. Circ Res 104(6): e42-54. Colosimo, C. and De Michele, M. (1999). Motor fluctuations in Parkinson's disease: pathophysiology and treatment. Eur J Neurol 6(1): 1-21. Colton, C., Wilt, S., Gilbert, D., Chernyshev, O., Snell, J. and Dubois-Dalcq, M. (1996). Species differences in the generation of reactive oxygen species by microglia. Mol Chem Neuropathol 28(1-3): 15-20. Colton, C. A., Snell, J., Chernyshev, O. and Gilbert, D. L. (1994). Induction of superoxide anion and nitric oxide production in cultured microglia. Ann N Y Acad Sci 738: 54-63. Conti, F. and Weinberg, R. J. (1999). Shaping excitation at glutamatergic synapses. Trends Neurosci 22(10): 451-458. Cook, C. E., Jeffcoat, A. R., Hill, J. M., Pugh, D. E., Patetta, P. K., Sadler, B. M., White, W. R. and Perez-Reyes, M. (1993). Pharmacokinetics of methamphetamine self-administered to human subjects by smoking S-(+)-methamphetamine hydrochloride. Drug Metab Dispos 21(4): 717-723. Cook, C. E., Jeffcoat, A. R., Sadler, B. M., Hill, J. M., Voyksner, R. D., Pugh, D. E., White, W. R. and Perez-Reyes, M. (1992). Pharmacokinetics of oral methamphetamine and effects of repeated daily dosing in humans. Drug Metab Dispos 20(6): 856-862. Cooke, M. S., Evans, M. D., Dizdaroglu, M. and Lunec, J. (2003). Oxidative DNA damage: mechanisms, mutation, and disease. FASEB J 17(10): 1195-1214. Copeland, R. A., Williams, J. M., Giannaras, J., Nurnberg, S., Covington, M., Pinto, D., Pick, S. and Trzaskos, J. M. (1994). Mechanism of selective inhibition of the inducible isoform of prostaglandin G/H synthase. Proc Natl Acad Sci U S A 91(23): 11202-11206. 339

Cork, R. J., Perrone, M. L., Bridges, D., Wandell, J., Scheiner, C. A. and Mize, R. R. (1998). A web-accessible digital atlas of the distribution of nitric oxide synthase in the mouse brain. Prog Brain Res 118: 37-50. Correale, J. and Villa, A. (2007). The blood-brain-barrier in multiple sclerosis: functional roles and therapeutic targeting. Autoimmunity 40(2): 148-160. Cruickshank, C. C. and Dyer, K. R. (2009). A review of the clinical pharmacology of methamphetamine. Addiction 104(7): 1085-1099. Cudkowicz, M. E., Shefner, J. M., Schoenfeld, D. A., Zhang, H., Andreasson, K. I., Rothstein, J. D. and Drachman, D. B. (2006). Trial of celecoxib in amyotrophic lateral sclerosis. Ann Neurol 60(1): 22-31. Cullinan, S. B., Gordan, J. D., Jin, J., Harper, J. W. and Diehl, J. A. (2004). The Keap1-BTB protein is an adaptor that bridges Nrf2 to a Cul3-based E3 ligase: oxidative stress sensing by a Cul3-Keap1 ligase. Mol Cell Biol 24(19): 8477-8486. Dalton, T. P., Shertzer, H. G. and Puga, A. (1999). Regulation of gene expression by reactive oxygen. Annu Rev Pharmacol Toxicol 39: 67-101. Danbolt, N. C. (2001). Glutamate uptake. Prog Neurobiol 65(1): 1-105. Das, K. C., Abramson, M. B. and Katzman, R. (1978). Neuronal pigments: spectroscopic characterization of human brain melanin. J Neurochem 30(3): 601-605. Dash, P. K., Zhao, J., Orsi, S. A., Zhang, M. and Moore, A. N. (2009). Sulforaphane improves cognitive function administered following traumatic brain injury. Neurosci Lett 460(2): 103-107. Dauchy, S., Dutheil, F., Weaver, R. J., Chassoux, F., Daumas-Duport, C., Couraud, P. O., Scherrmann, J. M., De Waziers, I. and Decleves, X. (2008). ABC transporters, cytochromes P450 and their main transcription factors: expression at the human blood- brain barrier. J Neurochem 107(6): 1518-1528. Davidson, C., Gow, A. J., Lee, T. H. and Ellinwood, E. H. (2001). Methamphetamine neurotoxicity: necrotic and apoptotic mechanisms and relevance to human abuse and treatment. Brain Res Brain Res Rev 36(1): 1-22. de la Torre, R. and Farre, M. (2004). Neurotoxicity of MDMA (ecstasy): the limitations of scaling from animals to humans. Trends Pharmacol Sci 25(10): 505-508. de la Torre, R., Farre, M., Ortuno, J., Mas, M., Brenneisen, R., Roset, P. N., Segura, J. and Cami, J. (2000). Non-linear pharmacokinetics of MDMA ('ecstasy') in humans. Br J Clin Pharmacol 49(2): 104-109. De Vito, M. J. and Wagner, G. C. (1989). Methamphetamine-induced neuronal damage: a 340

possible role for free radicals. Neuropharmacology 28(10): 1145-1150. de Vries, H. E., Witte, M., Hondius, D., Rozemuller, A. J., Drukarch, B., Hoozemans, J. and van Horssen, J. (2008). Nrf2-induced antioxidant protection: a promising target to counteract ROS-mediated damage in neurodegenerative disease? Free Radic Biol Med 45(10): 1375-1383. DeBault, L. E. and Cancilla, P. A. (1980). gamma-Glutamyl transpeptidase in isolated brain endothelial cells: induction by glial cells in vitro. Science 207(4431): 653-655. Deininger, M. H. and Schluesener, H. J. (1999). Cyclooxygenases-1 and -2 are differentially localized to microglia and endothelium in rat EAE and glioma. J Neuroimmunol 95(1-2): 202-208. DeLong, C. J. and Smith, W. L. (2005). An intronic enhancer regulates cyclooxygenase-1 gene expression. Biochem Biophys Res Commun 338(1): 53-61. Deng, W. G., Zhu, Y. and Wu, K. K. (2004). Role of p300 and PCAF in regulating cyclooxygenase-2 promoter activation by inflammatory mediators. Blood 103(6): 2135- 2142. Deng, X., Ladenheim, B., Jayanthi, S. and Cadet, J. L. (2007). Methamphetamine administration causes death of dopaminergic neurons in the mouse olfactory bulb. Biol Psychiatry 61(11): 1235-1243. Deng, X., Wang, Y., Chou, J. and Cadet, J. L. (2001). Methamphetamine causes widespread apoptosis in the mouse brain: evidence from using an improved TUNEL histochemical method. Brain Res Mol Brain Res 93(1): 64-69. Derjuga, A., Gourley, T. S., Holm, T. M., Heng, H. H., Shivdasani, R. A., Ahmed, R., Andrews, N. C. and Blank, V. (2004). Complexity of CNC transcription factors as revealed by gene targeting of the Nrf3 locus. Mol Cell Biol 24(8): 3286-3294. Desagher, S., Glowinski, J. and Premont, J. (1996). Astrocytes protect neurons from hydrogen peroxide toxicity. J Neurosci 16(8): 2553-2562. Desai, B. S., Monahan, A. J., Carvey, P. M. and Hendey, B. (2007). Blood-brain barrier pathology in Alzheimer's and Parkinson's disease: implications for drug therapy. Cell Transplant 16(3): 285-299. DeWitt, D. L. and Smith, W. L. (1988). Primary structure of prostaglandin G/H synthase from sheep vesicular gland determined from the complementary DNA sequence. Proc Natl Acad Sci U S A 85(5): 1412-1416. Dhakshinamoorthy, S., Jain, A. K., Bloom, D. A. and Jaiswal, A. K. (2005). Bach1 competes with Nrf2 leading to negative regulation of the antioxidant response element (ARE)- 341

mediated NAD(P)H:quinone oxidoreductase 1 gene expression and induction in response to antioxidants. J Biol Chem 280(17): 16891-16900. Dhakshinamoorthy, S. and Jaiswal, A. K. (2000). Small maf (MafG and MafK) proteins negatively regulate antioxidant response element-mediated expression and antioxidant induction of the NAD(P)H:Quinone oxidoreductase1 gene. J Biol Chem 275(51): 40134- 40141. Dhakshinamoorthy, S. and Porter, A. G. (2004). Nitric oxide-induced transcriptional upregulation of protective genes by Nrf2 via the antioxidant response element counteracts apoptosis of neuroblastoma cells. J Biol Chem 279(19): 20096-20107. Dheen, S. T., Kaur, C. and Ling, E. A. (2007). Microglial activation and its implications in the brain diseases. Curr Med Chem 14(11): 1189-1197. Dhenaut, A., Hollenbach, S., Eckert, I., Epe, B., Boiteux, S. and Radicella, J. P. (2001). Characterization of hOGG1 promoter structure, expression during cell cycle and overexpression in mammalian cells. Adv Exp Med Biol 500: 613-616. Di Monte, D. A., Royland, J. E., Jakowec, M. W. and Langston, J. W. (1996). Role of nitric oxide in methamphetamine neurotoxicity: protection by 7-nitroindazole, an inhibitor of neuronal nitric oxide synthase. J Neurochem 67(6): 2443-2450. Diaz, A., Chepenik, K. P., Korn, J. H., Reginato, A. M. and Jimenez, S. A. (1998). Differential regulation of cyclooxygenases 1 and 2 by interleukin-1 beta, tumor necrosis factor-alpha, and transforming growth factor-beta 1 in human lung fibroblasts. Exp Cell Res 241(1): 222-229. Dinchuk, J. E., Liu, R. Q. and Trzaskos, J. M. (2003). COX-3: in the wrong frame in mind. Immunol Lett 86(1): 121. Dinkova-Kostova, A. T., Holtzclaw, W. D., Cole, R. N., Itoh, K., Wakabayashi, N., Katoh, Y., Yamamoto, M. and Talalay, P. (2002). Direct evidence that sulfhydryl groups of Keap1 are the sensors regulating induction of phase 2 enzymes that protect against carcinogens and oxidants. Proc Natl Acad Sci U S A 99(18): 11908-11913. Dinkova-Kostova, A. T., Holtzclaw, W. D. and Kensler, T. W. (2005). The role of Keap1 in cellular protective responses. Chem Res Toxicol 18(12): 1779-1791. Dinkova-Kostova, A. T. and Talalay, P. (2010). NAD(P)H:quinone acceptor oxidoreductase 1 (NQO1), a multifunctional antioxidant enzyme and exceptionally versatile cytoprotector. Arch Biochem Biophys 501(1): 116-123. Dong, J., Sulik, K. K. and Chen, S. Y. (2008). Nrf2-mediated transcriptional induction of antioxidant response in mouse embryos exposed to ethanol in vivo: implications for the 342

prevention of fetal alcohol spectrum disorders. Antioxid Redox Signal 10(12): 2023- 2033. Dong, X. X., Wang, Y. and Qin, Z. H. (2009). Molecular mechanisms of excitotoxicity and their relevance to pathogenesis of neurodegenerative diseases. Acta Pharmacol Sin 30(4): 379-387. Dore, S., Takahashi, M., Ferris, C. D., Zakhary, R., Hester, L. D., Guastella, D. and Snyder, S. H. (1999). Bilirubin, formed by activation of heme oxygenase-2, protects neurons against oxidative stress injury. Proc Natl Acad Sci U S A 96(5): 2445-2450. Doty, R. L., Deems, D. A. and Stellar, S. (1988). Olfactory dysfunction in parkinsonism: a general deficit unrelated to neurologic signs, disease stage, or disease duration. Neurology 38(8): 1237-1244. Doty, R. L., Stern, M. B., Pfeiffer, C., Gollomp, S. M. and Hurtig, H. I. (1992). Bilateral olfactory dysfunction in early stage treated and untreated idiopathic Parkinson's disease. J Neurol Neurosurg Psychiatry 55(2): 138-142. Dou, W., Jiao, Y., Goorha, S., Raghow, R. and Ballou, L. R. (2004). Nociception and the differential expression of cyclooxygenase-1 (COX-1), the COX-1 variant retaining intron-1 (COX-1v), and COX-2 in mouse dorsal root ganglia (DRG). Prostaglandins Other Lipid Mediat 74(1-4): 29-43. Downing, D. T., Barve, J. A., Gunstone, F. D., Jacobsberg, F. R. and Jie, M. L. (1972). Structural requirements of acetylenic fatty acids for inhibition of soybean lipoxygenase and prostaglandin synthetase. Biochim Biophys Acta 280(2): 343-347. Drachman, D. B., Frank, K., Dykes-Hoberg, M., Teismann, P., Almer, G., Przedborski, S. and Rothstein, J. D. (2002). Cyclooxygenase 2 inhibition protects motor neurons and prolongs survival in a transgenic mouse model of ALS. Ann Neurol 52(6): 771-778. Dringen, R. (2000). Metabolism and functions of glutathione in brain. Prog Neurobiol 62(6): 649-671. Dringen, R., Pawlowski, P. G. and Hirrlinger, J. (2005). Peroxide detoxification by brain cells. J Neurosci Res 79(1-2): 157-165. Dutheil, F., Dauchy, S., Diry, M., Sazdovitch, V., Cloarec, O., Mellottee, L., Bieche, I., Ingelman-Sundberg, M., Flinois, J. P., de Waziers, I., Beaune, P., Decleves, X., Duyckaerts, C. and Loriot, M. A. (2009). Xenobiotic-metabolizing enzymes and transporters in the normal human brain: regional and cellular mapping as a basis for putative roles in cerebral function. Drug Metab Dispos 37(7): 1528-1538. Eisch, A. J., Schmued, L. C. and Marshall, J. F. (1998). Characterizing cortical neuron injury 343

with Fluoro-Jade labeling after a neurotoxic regimen of methamphetamine. Synapse 30(3): 329-333. el-Bacha, R. S. and Minn, A. (1999). Drug metabolizing enzymes in cerebrovascular endothelial cells afford a metabolic protection to the brain. Cell Mol Biol (Noisy-le-grand) 45(1): 15-23. el-Hage, S. and Singh, S. M. (1990). Temporal expression of genes encoding free radical- metabolizing enzymes is associated with higher mRNA levels during in utero development in mice. Dev Genet 11(2): 149-159. Elfering, S. L., Sarkela, T. M. and Giulivi, C. (2002). Biochemistry of mitochondrial nitric- oxide synthase. J Biol Chem 277(41): 38079-38086. Eling, T. E. and Curtis, J. F. (1992). Xenobiotic metabolism by prostaglandin H synthase. Pharmacol Ther 53(2): 261-273. Enomoto, A., Itoh, K., Nagayoshi, E., Haruta, J., Kimura, T., O'Connor, T., Harada, T. and Yamamoto, M. (2001). High sensitivity of Nrf2 knockout mice to acetaminophen hepatotoxicity associated with decreased expression of ARE-regulated drug metabolizing enzymes and antioxidant genes. Toxicol Sci 59(1): 169-177. Enwere, E., Shingo, T., Gregg, C., Fujikawa, H., Ohta, S. and Weiss, S. (2004). Aging results in reduced epidermal growth factor receptor signaling, diminished olfactory neurogenesis, and deficits in fine olfactory discrimination. J Neurosci 24(38): 8354-8365. Erecinska, M. and Silver, I. A. (1990). Metabolism and role of glutamate in mammalian brain. Prog Neurobiol 35(4): 245-296. Erickson, A. M., Nevarea, Z., Gipp, J. J. and Mulcahy, R. T. (2002). Identification of a variant antioxidant response element in the promoter of the human glutamate-cysteine ligase modifier subunit gene. Revision of the ARE consensus sequence. J Biol Chem 277(34): 30730-30737. Ernst, T., Chang, L., Leonido-Yee, M. and Speck, O. (2000). Evidence for long-term neurotoxicity associated with methamphetamine abuse: A 1H MRS study. Neurology 54(6): 1344-1349. Eshleman, A. J., Carmolli, M., Cumbay, M., Martens, C. R., Neve, K. A. and Janowsky, A. (1999). Characteristics of drug interactions with recombinant biogenic amine transporters expressed in the same cell type. J Pharmacol Exp Ther 289(2): 877-885. Esteban, B., O'Shea, E., Camarero, J., Sanchez, V., Green, A. R. and Colado, M. I. (2001). 3,4- Methylenedioxymethamphetamine induces monoamine release, but not toxicity, when administered centrally at a concentration occurring following a peripherally injected 344

neurotoxic dose. Psychopharmacology (Berl) 154(3): 251-260. Etminan, M., Gill, S. and Samii, A. (2003). Effect of non-steroidal anti-inflammatory drugs on risk of Alzheimer's disease: systematic review and meta-analysis of observational studies. BMJ 327(7407): 128. Exner, M., Minar, E., Wagner, O. and Schillinger, M. (2004). The role of heme oxygenase-1 promoter polymorphisms in human disease. Free Radic Biol Med 37(8): 1097-1104. Factor, S. A., Molho, E. S., Podskalny, G. D. and Brown, D. (1995). Parkinson's disease: drug- induced psychiatric states. Adv Neurol 65: 115-138. Farin, F. M. and Omiecinski, C. J. (1993). Regiospecific expression of cytochrome P-450s and microsomal epoxide hydrolase in human brain tissue. J Toxicol Environ Health 40(2-3): 317-335. Farooqui, A. A., Ong, W. Y. and Horrocks, L. A. (2006). Inhibitors of brain phospholipase A2 activity: their neuropharmacological effects and therapeutic importance for the treatment of neurologic disorders. Pharmacol Rev 58(3): 591-620. Favreau, L. V. and Pickett, C. B. (1991). Transcriptional regulation of the rat NAD(P)H:quinone reductase gene. Identification of regulatory elements controlling basal level expression and inducible expression by planar aromatic compounds and phenolic antioxidants. J Biol Chem 266(7): 4556-4561. Felix, C. C. and Sealy, R. C. (1981). Photolysis of melanin precursors: Formation of semiquinone radicals and their complexation with diamagnetic metal ions. Photochem Photobiol 34(4): 423-429. Felix, C. C. and Sealy, R. C. (1982). o-Benzosemiquinone and its metal chelates: Electron spin resonance investigation of radicals from the photolysis of catechol in the presence of complexing metal ions. J Am Chem Soc 104(6): 1555-1560. Feng, Z. H., Wang, T. G., Li, D. D., Fung, P., Wilson, B. C., Liu, B., Ali, S. F., Langenbach, R. and Hong, J. S. (2002). Cyclooxygenase-2-deficient mice are resistant to 1-methyl-4- phenyl1, 2, 3, 6-tetrahydropyridine-induced damage of dopaminergic neurons in the substantia nigra. Neurosci Lett 329(3): 354-358. Fetler, L. and Amigorena, S. (2005). Neuroscience. Brain under surveillance: the microglia patrol. Science 309(5733): 392-393. Finkelstein, E., Rosen, G. M. and Rauckman, E. J. (1980). Spin trapping of superoxide and hydroxyl radical: practical aspects. Arch Biochem Biophys 200(1): 1-16. FitzGerald, G. A. and Patrono, C. (2001). The coxibs, selective inhibitors of cyclooxygenase-2. N Engl J Med 345(6): 433-442. 345

Fitzmaurice, P. S., Tong, J., Yazdanpanah, M., Liu, P. P., Kalasinsky, K. S. and Kish, S. J. (2006). Levels of 4-hydroxynonenal and malondialdehyde are increased in brain of human chronic users of methamphetamine. J Pharmacol Exp Ther 319(2): 703-709. Fleckenstein, A. E., Haughey, H. M., Metzger, R. R., Kokoshka, J. M., Riddle, E. L., Hanson, J. E., Gibb, J. W. and Hanson, G. R. (1999). Differential effects of psychostimulants and related agents on dopaminergic and serotonergic transporter function. Eur J Pharmacol 382(1): 45-49. Fleckenstein, A. E., Volz, T. J. and Hanson, G. R. (2009). Psychostimulant-induced alterations in vesicular monoamine transporter-2 function: neurotoxic and therapeutic implications. Neuropharmacology 56 Suppl 1: 133-138. Floden, A. M., Li, S. and Combs, C. K. (2005). Beta-amyloid-stimulated microglia induce neuron death via synergistic stimulation of tumor necrosis factor alpha and NMDA receptors. J Neurosci 25(10): 2566-2575. Foran, E. and Trotti, D. (2009). Glutamate transporters and the excitotoxic path to motor neuron degeneration in amyotrophic lateral sclerosis. Antioxid Redox Signal 11(7): 1587-1602. Fornstedt, B., Bergh, I., Rosengren, E. and Carlsson, A. (1990). An improved HPLC- electrochemical detection method for measuring brain levels of 5-S-cysteinyldopamine, 5-S-cysteinyl-3,4-dihydroxyphenylalanine, and 5-S-cysteinyl-3,4-dihydroxyphenylacetic acid. J Neurochem 54(2): 578-586. Fowler, J. S., Kroll, C., Ferrieri, R., Alexoff, D., Logan, J., Dewey, S. L., Schiffer, W., Schlyer, D., Carter, P., King, P., Shea, C., Xu, Y., Muench, L., Benveniste, H., Vaska, P. and Volkow, N. D. (2007). PET studies of d-methamphetamine pharmacokinetics in primates: comparison with l-methamphetamine and ( --)-cocaine. J Nucl Med 48(10): 1724-1732. Fowler, J. S., Volkow, N. D., Logan, J., Alexoff, D., Telang, F., Wang, G. J., Wong, C., Ma, Y., Kriplani, A., Pradhan, K., Schlyer, D., Jayne, M., Hubbard, B., Carter, P., Warner, D., King, P., Shea, C., Xu, Y., Muench, L. and Apelskog, K. (2008). Fast uptake and long- lasting binding of methamphetamine in the human brain: comparison with cocaine. Neuroimage 43(4): 756-763. Frejaville, C., Karoui, H., Tuccio, B., Le Moigne, F., Culcasi, M., Pietri, S., Lauricella, R. and Tordo, P. (1995). 5-(Diethoxyphosphoryl)-5-methyl-1-pyrroline N-oxide: a new efficient phosphorylated nitrone for the in vitro and in vivo spin trapping of oxygen-centered radicals. J Med Chem 38(2): 258-265. Fumagalli, F., Gainetdinov, R. R., Valenzano, K. J. and Caron, M. G. (1998). Role of dopamine 346

transporter in methamphetamine-induced neurotoxicity: evidence from mice lacking the transporter. J Neurosci 18(13): 4861-4869. Funk, C. D. (2001). Prostaglandins and leukotrienes: advances in eicosanoid biology. Science 294(5548): 1871-1875. Furukawa, M. and Xiong, Y. (2005). BTB protein Keap1 targets antioxidant transcription factor Nrf2 for ubiquitination by the Cullin 3-Roc1 ligase. Mol Cell Biol 25(1): 162-171. Gabbita, S. P., Lovell, M. A. and Markesbery, W. R. (1998). Increased nuclear DNA oxidation in the brain in Alzheimer's disease. J Neurochem 71(5): 2034-2040. Gan, L., Johnson, D. A. and Johnson, J. A. (2010). Keap1-Nrf2 activation in the presence and absence of DJ-1. Eur J Neurosci 31(6): 967-977. Gao, H. M., Jiang, J., Wilson, B., Zhang, W., Hong, J. S. and Liu, B. (2002). Microglial activation-mediated delayed and progressive degeneration of rat nigral dopaminergic neurons: relevance to Parkinson's disease. J Neurochem 81(6): 1285-1297. Gao, H. M., Liu, B., Zhang, W. and Hong, J. S. (2003). Critical role of microglial NADPH oxidase-derived free radicals in the in vitro MPTP model of Parkinson's disease. FASEB J 17(13): 1954-1956. Garris, P. A., Ciolkowski, E. L., Pastore, P. and Wightman, R. M. (1994). Efflux of dopamine from the synaptic cleft in the nucleus accumbens of the rat brain. J Neurosci 14(10): 6084-6093. Gattaz, W. F., Maras, A., Cairns, N. J., Levy, R. and Forstl, H. (1995). Decreased phospholipase A2 activity in Alzheimer brains. Biol Psychiatry 37(1): 13-17. Ghersi-Egea, J. F., Leninger-Muller, B., Suleman, G., Siest, G. and Minn, A. (1994). Localization of drug-metabolizing enzyme activities to blood-brain interfaces and circumventricular organs. J Neurochem 62(3): 1089-1096. Gheusi, G., Cremer, H., McLean, H., Chazal, G., Vincent, J. D. and Lledo, P. M. (2000). Importance of newly generated neurons in the adult olfactory bulb for odor discrimination. Proc Natl Acad Sci USA 97(4): 1823-1828. Ghilardi, J. R., Svensson, C. I., Rogers, S. D., Yaksh, T. L. and Mantyh, P. W. (2004). Constitutive spinal cyclooxygenase-2 participates in the initiation of tissue injury- induced hyperalgesia. J Neurosci 24(11): 2727-2732. Gibb, J. W., Johnson, M., Stone, D. and Hanson, G. R. (1990). MDMA: historical perspectives. Ann N Y Acad Sci 600: 601-611; discussion 611-602. Gierse, J. K., McDonald, J. J., Hauser, S. D., Rangwala, S. H., Koboldt, C. M. and Seibert, K. (1996). A single amino acid difference between cyclooxygenase-1 (COX-1) and -2 347

(COX-2) reverses the selectivity of COX-2 specific inhibitors. J Biol Chem 271(26): 15810-15814. Giovanni, A., Liang, L. P., Hastings, T. G. and Zigmond, M. J. (1995). Estimating hydroxyl radical content in rat brain using systemic and intraventricular salicylate: impact of methamphetamine. J Neurochem 64(4): 1819-1825. Girault, J. A. and Greengard, P. (2004). The neurobiology of dopamine signaling. Arch Neurol 61(5): 641-644. Giros, B. and Caron, M. G. (1993). Molecular characterization of the dopamine transporter. Trends Pharmacol Sci 14(2): 43-49. Gluck, M. R., Moy, L. Y., Jayatilleke, E., Hogan, K. A., Manzino, L. and Sonsalla, P. K. (2001). Parallel increases in lipid and protein oxidative markers in several mouse brain regions after methamphetamine treatment. J Neurochem 79(1): 152-160. Gluck, M. R. and Zeevalk, G. D. (2004). Inhibition of brain mitochondrial respiration by dopamine and its metabolites: implications for Parkinson's disease and catecholamine- associated diseases. J Neurochem 91(4): 788-795. Golub, M., Costa, L., Crofton, K., Frank, D., Fried, P., Gladen, B., Henderson, R., Liebelt, E., Lusskin, S., Marty, S., Rowland, A., Scialli, J. and Vore, M. (2005). NTP-CERHR Expert Panel Report on the reproductive and developmental toxicity of amphetamine and methamphetamine. Birth Defects Res B Dev Reprod Toxicol 74(6): 471-584. Gomez-Santos, C., Ferrer, I., Santidrian, A. F., Barrachina, M., Gil, J. and Ambrosio, S. (2003). Dopamine induces autophagic cell death and alpha-synuclein increase in human neuroblastoma SH-SY5Y cells. J Neurosci Res 73(3): 341-350. Goncalves, L. L., Ramkissoon, A. and Wells, P. G. (2009). Prostaglandin H synthase-1- catalyzed bioactivation of neurotransmitters, their precursors, and metabolites: oxidative DNA damage and electron spin resonance spectroscopy studies. Chem Res Toxicol 22(5): 842-852. Graham, D. G. (1978). Oxidative pathways for catecholamines in the genesis of neuromelanin and cytotoxic quinones. Mol Pharmacol 14(4): 633-643. Graham, D. G. and Jeffs, P. W. (1977). The role of 2,4,5-trihydroxyphenylalanine in melanin biosynthesis. J Biol Chem 252(16): 5729-5734. Graham, D. G., Tiffany, S. M., Bell, W. R., Jr. and Gutknecht, W. F. (1978). Autoxidation versus 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(4): 644-653. 348

Graumann, R., Paris, I., Martinez-Alvarado, P., Rumanque, P., Perez-Pastene, C., Cardenas, S. P., Marin, P., Diaz-Grez, F., Caviedes, R., Caviedes, P. and Segura-Aguilar, J. (2002). Oxidation of dopamine to aminochrome as a mechanism for neurodegeneration of dopaminergic systems in Parkinson's disease. Possible neuroprotective role of DT- diaphorase. Pol J Pharmacol 54(6): 573-579. Green, A. R., Mechan, A. O., Elliott, J. M., O'Shea, E. and Colado, M. I. (2003). The pharmacology and clinical pharmacology of 3,4-methylenedioxymethamphetamine (MDMA, "ecstasy"). Pharmacol Rev 55(3): 463-508. Grinspoon, L. and Bakalar, J. B. (1986). Can drugs be used to enhance the psychotherapeutic process? Am J Psychother 40(3): 393-404. Guan, C. P., Zhou, M. N., Xu, A. E., Kang, K. F., Liu, J. F., Wei, X. D., Li, Y. W., Zhao, D. K. and Hong, W. S. (2008). The susceptibility to vitiligo is associated with NF-E2-related factor2 (Nrf2) gene polymorphisms: a study on Chinese Han population. Exp Dermatol 17(12): 1059-1062. Guilarte, T. R., Nihei, M. K., McGlothan, J. L. and Howard, A. S. (2003). Methamphetamine- induced deficits of brain monoaminergic neuronal markers: distal axotomy or neuronal plasticity. Neuroscience 122(2): 499-513. Guix, F. X., Uribesalgo, I., Coma, M. and Munoz, F. J. (2005). The physiology and pathophysiology of nitric oxide in the brain. Prog Neurobiol 76(2): 126-152. Guo, Q. and Kulmacz, R. J. (2000). Distinct influences of carboxyl terminal segment structure on function in the two isoforms of prostaglandin H synthase. Arch Biochem Biophys 384(2): 269-279. Guo, Q., Zhao, B., Shen, S., Hou, J., Hu, J. and Xin, W. (1999). ESR study on the structure- antioxidant activity relationship of tea catechins and their epimers. Biochim Biophys Acta 1427(1): 13-23. Haeffner, A., Thieblemont, N., Deas, O., Marelli, O., Charpentier, B., Senik, A., Wright, S. D., Haeffner-Cavaillon, N. and Hirsch, F. (1997). Inhibitory effect of growth hormone on TNF-alpha secretion and nuclear factor-kappaB translocation in lipopolysaccharide- stimulated human monocytes. J Immunol 158(3): 1310-1314. Hailer-Morrison, M. K., Kotler, J. M., Martin, B. D. and Sugden, K. D. (2003). Oxidized guanine lesions as modulators of gene transcription. Altered p50 binding affinity and repair shielding by 7,8-dihydro-8-oxo-2'-deoxyguanosine lesions in the NF-kappaB promoter element. Biochemistry (Mosc) 42(32): 9761-9770. Halliwell, B. and Gutteridge, J. M. C. (2007). Free Radicals in Biology and Medicine. New 349

York, Oxford University Press Inc. Hamberg, M. and Samuelsson, B. (1967). Oxygenation of unsaturated fatty acids by the vesicular gland of sheep. J Biol Chem 242(22): 5344-5354. Hamberg, M., Svensson, J., Wakabayashi, T. and Samuelsson, B. (1974). Isolation and structure of two prostaglandin endoperoxides that cause platelet aggregation. Proc Natl Acad Sci U S A 71(2): 345-349. Hammarstrom, S. (1977). Selective inhibition of platelet n-8 lipoxygenase by 5,8,11- eicosatriynoic acid. Biochim Biophys Acta 487(3): 517-519. Han, D. D. and Gu, H. H. (2006). Comparison of the monoamine transporters from human and mouse in their sensitivities to psychostimulant drugs. BMC Pharmacol 6: 6. Hanna, P. M., Chamulitrat, W. and Mason, R. P. (1992). When are metal ion-dependent hydroxyl and alkoxyl radical adducts of 5,5-dimethyl-1-pyrroline N-oxide artifacts? Arch Biochem Biophys 296(2): 640-644. Haorah, J., Ramirez, S. H., Floreani, N., Gorantla, S., Morsey, B. and Persidsky, Y. (2008). Mechanism of alcohol-induced oxidative stress and neuronal injury. Free Radic Biol Med 45(11): 1542-1550. Harris, D. S., Boxenbaum, H., Everhart, E. T., Sequeira, G., Mendelson, J. E. and Jones, R. T. (2003). The bioavailability of intranasal and smoked methamphetamine. Clin Pharmacol Ther 74(5): 475-486. Harrison, R. (2002). Structure and function of xanthine oxidoreductase: where are we now? Free Radic Biol Med 33(6): 774-797. Hartlage-Rubsamen, M., Lemke, R. and Schliebs, R. (1999). Interleukin-1beta, inducible nitric oxide synthase, and nuclear factor-kappaB are induced in morphologically distinct microglia after rat hippocampal lipopolysaccharide/interferon-gamma injection. J Neurosci Res 57(3): 388-398. Harvey, D. C., Lacan, G., Tanious, S. P. and Melega, W. P. (2000). Recovery from methamphetamine induced long-term nigrostriatal dopaminergic deficits without substantia nigra cell loss. Brain Res 871(2): 259-270. Hastings, T. G. (1995). Enzymatic oxidation of dopamine: the role of prostaglandin H synthase. J Neurochem 64(2): 919-924. Hasumoto, K., Sugimoto, Y., Yamasaki, A., Morimoto, K., Kakizuka, A., Negishi, M. and Ichikawa, A. (1997). Association of expression of mRNA encoding the PGF2 alpha receptor with luteal cell apoptosis in ovaries of pseudopregnant mice. J Reprod Fertil 109(1): 45-51. 350

Hata, A. N. and Breyer, R. M. (2004). Pharmacology and signaling of prostaglandin receptors: multiple roles in inflammation and immune modulation. Pharmacol Ther 103(2): 147- 166. Hayashi, A., Suzuki, H., Itoh, K., Yamamoto, M. and Sugiyama, Y. (2003). Transcription factor Nrf2 is required for the constitutive and inducible expression of multidrug resistance- associated protein 1 in mouse embryo fibroblasts. Biochem Biophys Res Commun 310(3): 824-829. Hayes, J. D. and McMahon, M. (2009). NRF2 and KEAP1 mutations: permanent activation of an adaptive response in cancer. Trends Biochem Sci 34(4): 176-188. He, C. H., Gong, P., Hu, B., Stewart, D., Choi, M. E., Choi, A. M. and Alam, J. (2001). Identification of activating transcription factor 4 (ATF4) as an Nrf2-interacting protein. Implication for heme oxygenase-1 gene regulation. J Biol Chem 276(24): 20858-20865. He, X., Chen, M. G., Lin, G. X. and Ma, Q. (2006). Arsenic induces NAD(P)H-quinone oxidoreductase I by disrupting the Nrf2 x Keap1 x Cul3 complex and recruiting Nrf2 x Maf to the antioxidant response element enhancer. J Biol Chem 281(33): 23620-23631. Heikkila, R. E., Manzino, L., Cabbat, F. S. and Duvoisin, R. C. (1984). Protection against the dopaminergic neurotoxicity of 1-methyl-4-phenyl-1,2,5,6-tetrahydropyridine by monoamine oxidase inhibitors. Nature 311(5985): 467-469. Heneka, M. T. and Feinstein, D. L. (2001). Expression and function of inducible nitric oxide synthase in neurons. J Neuroimmunol 114(1-2): 8-18. Herin, D. V., Liu, S., Ullrich, T., Rice, K. C. and Cunningham, K. A. (2005). Role of the serotonin 5-HT2A receptor in the hyperlocomotive and hyperthermic effects of (+)-3,4- methylenedioxymethamphetamine. Psychopharmacology (Berl) 178(4): 505-513. Herlenius, E. and Lagercrantz, H. (2004). Development of neurotransmitter systems during critical periods. Exp Neurol 190 Suppl 1: S8-21. Heydari, A., Yeo, K. R., Lennard, M. S., Ellis, S. W., Tucker, G. T. and Rostami-Hodjegan, A. (2004). Mechanism-based inactivation of CYP2D6 by methylenedioxymethamphetamine. Drug Metab Dispos 32(11): 1213-1217. Hinds, J. W. and Hinds, P. L. (1976). Synapse formation in the mouse olfactory bulb. I. Quantitative studies. J Comp Neurol 169(1): 15-40. Hiramatsu, M., Kumagai, Y., Unger, S. E. and Cho, A. K. (1990). Metabolism of methylenedioxymethamphetamine: formation of dihydroxymethamphetamine and a quinone identified as its glutathione adduct. J Pharmacol Exp Ther 254(2): 521-527. Hirashima, Y., Farooqui, A. A., Mills, J. S. and Horrocks, L. A. (1992). Identification and 351

purification of calcium-independent phospholipase A2 from bovine brain cytosol. J Neurochem 59(2): 708-714. Hirata, H., Ladenheim, B., Carlson, E., Epstein, C. and Cadet, J. L. (1996). Autoradiographic evidence for methamphetamine-induced striatal dopaminergic loss in mouse brain: attenuation in CuZn-superoxide dismutase transgenic mice. Brain Res 714(1-2): 95-103. Hirata, M., Kakizuka, A., Aizawa, M., Ushikubi, F. and Narumiya, S. (1994). Molecular characterization of a mouse prostaglandin D receptor and functional expression of the cloned gene. Proc Natl Acad Sci U S A 91(23): 11192-11196. Hirotani, M., Maita, C., Niino, M., Iguchi-Ariga, S., Hamada, S., Ariga, H. and Sasaki, H. (2008). Correlation between DJ-1 levels in the cerebrospinal fluid and the progression of disabilities in multiple sclerosis patients. Mult Scler 14(8): 1056-1060. Hla, T. (1996). Molecular characterization of the 5.2 KB isoform of the human cyclooxygenase- 1 transcript. Prostaglandins 51(1): 81-85. Ho, L., Pieroni, C., Winger, D., Purohit, D. P., Aisen, P. S. and Pasinetti, G. M. (1999). Regional distribution of cyclooxygenase-2 in the hippocampal formation in Alzheimer's disease. J Neurosci Res 57(3): 295-303. Ho, L., Purohit, D., Haroutunian, V., Luterman, J. D., Willis, F., Naslund, J., Buxbaum, J. D., Mohs, R. C., Aisen, P. S. and Pasinetti, G. M. (2001). Neuronal cyclooxygenase 2 expression in the hippocampal formation as a function of the clinical progression of Alzheimer disease. Arch Neurol 58(3): 487-492. Ho, Y. S., Magnenat, J. L., Bronson, R. T., Cao, J., Gargano, M., Sugawara, M. and Funk, C. D. (1997). Mice deficient in cellular glutathione peroxidase develop normally and show no increased sensitivity to hyperoxia. J Biol Chem 272(26): 16644-16651. Hofmann, B., Hecht, H. J. and Flohe, L. (2002). Peroxiredoxins. Biol Chem 383(3-4): 347-364. Holtzman, M. J., Turk, J. and Shornick, L. P. (1992). Identification of a pharmacologically distinct prostaglandin H synthase in cultured epithelial cells. J Biol Chem 267(30): 21438-21445. Hong, F., Sekhar, K. R., Freeman, M. L. and Liebler, D. C. (2005). Specific patterns of electrophile adduction trigger Keap1 ubiquitination and Nrf2 activation. J Biol Chem 280(36): 31768-31775. Hoozemans, J. J., Rozemuller, A. J., Janssen, I., De Groot, C. J., Veerhuis, R. and Eikelenboom, P. (2001). Cyclooxygenase expression in microglia and neurons in Alzheimer's disease and control brain. Acta Neuropathol 101(1): 2-8. Hoozemans, J. J., Veerhuis, R., Janssen, I., van Elk, E. J., Rozemuller, A. J. and Eikelenboom, 352

P. (2002). The role of cyclo-oxygenase 1 and 2 activity in prostaglandin E(2) secretion by cultured human adult microglia: implications for Alzheimer's disease. Brain Res 951(2): 218-226. Hornung, J. P. (2003). The human raphe nuclei and the serotonergic system. J Chem Neuroanat 26(4): 331-343. Hoshino, H., Kobayashi, A., Yoshida, M., Kudo, N., Oyake, T., Motohashi, H., Hayashi, N., Yamamoto, M. and Igarashi, K. (2000). Oxidative stress abolishes leptomycin B- sensitive nuclear export of transcription repressor Bach2 that counteracts activation of Maf recognition element. J Biol Chem 275(20): 15370-15376. Hotamisligil, G. S., Girmen, A. S., Fink, J. S., Tivol, E., Shalish, C., Trofatter, J., Baenziger, J., Diamond, S., Markham, C., Sullivan, J. and et al. (1994). Hereditary variations in monoamine oxidase as a risk factor for Parkinson's disease. Mov Disord 9(3): 305-310. Hotchkiss, A. J. and Gibb, J. W. (1980). Long-term effects of multiple doses of methamphetamine on tryptophan hydroxylase and tyrosine hydroxylase activity in rat brain. J Pharmacol Exp Ther 214(2): 257-262. Hu, D., Serrano, F., Oury, T. D. and Klann, E. (2006). Aging-dependent alterations in synaptic plasticity and memory in mice that overexpress extracellular superoxide dismutase. J Neurosci 26(15): 3933-3941. Huang, H. C., Nguyen, T. and Pickett, C. B. (2002). Phosphorylation of Nrf2 at Ser-40 by protein kinase C regulates antioxidant response element-mediated transcription. J Biol Chem 277(45): 42769-42774. Huang, Y. N., Wu, C. H., Lin, T. C. and Wang, J. Y. (2009). Methamphetamine induces heme oxygenase-1 expression in cortical neurons and glia to prevent its toxicity. Toxicol Appl Pharmacol 240(3): 315-326. Hubbs, A. F., Benkovic, S. A., Miller, D. B., O'Callaghan, J. P., Battelli, L., Schwegler-Berry, D. and Ma, Q. (2007). Vacuolar leukoencephalopathy with widespread astrogliosis in mice lacking transcription factor Nrf2. Am J Pathol 170(6): 2068-2076. Huestis, M. A. and Cone, E. J. (2007). Methamphetamine disposition in oral fluid, plasma, and urine. Ann N Y Acad Sci 1098: 104-121. Hughes, M. F., Mason, R. P. and Eling, T. E. (1988). Prostaglandin hydroperoxidase-dependent oxidation of phenylbutazone: relationship to inhibition of prostaglandin cyclooxygenase. Mol Pharmacol 34(2): 186-193. Huisman, E., Uylings, H. B. and Hoogland, P. V. (2004). A 100% increase of dopaminergic cells in the olfactory bulb may explain hyposmia in Parkinson's disease. Mov Disord 353

19(6): 687-692. Huisman, E., Uylings, H. B. and Hoogland, P. V. (2008). Gender-related changes in increase of dopaminergic neurons in the olfactory bulb of Parkinson's disease patients. Mov Disord 23(10): 1407-1413. Hyun, J. W., Jung, Y. C., Kim, H. S., Choi, E. Y., Kim, J. E., Yoon, B. H., Yoon, S. H., Lee, Y. S., Choi, J., You, H. J. and Chung, M. H. (2003). 8-hydroxydeoxyguanosine causes death of human leukemia cells deficient in 8-oxoguanine glycosylase 1 activity by inducing apoptosis. Mol Cancer Res 1(4): 290-299. Igarashi, K., Hoshino, H., Muto, A., Suwabe, N., Nishikawa, S., Nakauchi, H. and Yamamoto, M. (1998). Multivalent DNA binding complex generated by small Maf and Bach1 as a possible biochemical basis for beta-globin locus control region complex. J Biol Chem 273(19): 11783-11790. Ignarro, L. J. (1991). Signal transduction mechanisms involving nitric oxide. Biochem Pharmacol 41(4): 485-490. Ikeda, Y., Sugawara, A., Taniyama, Y., Uruno, A., Igarashi, K., Arima, S., Ito, S. and Takeuchi, K. (2000). Suppression of rat thromboxane synthase gene transcription by peroxisome proliferator-activated receptor gamma in macrophages via an interaction with NRF2. J Biol Chem 275(42): 33142-33150. Im, J. Y., Kim, D., Paik, S. G. and Han, P. L. (2006). Cyclooxygenase-2-dependent neuronal death proceeds via superoxide anion generation. Free Radic Biol Med 41(6): 960-972. Imam, S. Z., el-Yazal, J., Newport, G. D., Itzhak, Y., Cadet, J. L., Slikker, W., Jr. and Ali, S. F. (2001). Methamphetamine-induced dopaminergic neurotoxicity: role of peroxynitrite and neuroprotective role of antioxidants and peroxynitrite decomposition catalysts. Ann N Y Acad Sci 939: 366-380. Infanger, D. W., Sharma, R. V. and Davisson, R. L. (2006). NADPH oxidases of the brain: distribution, regulation, and function. Antioxid Redox Signal 8(9-10): 1583-1596. Inoue, H., Nanayama, T., Hara, S., Yokoyama, C. and Tanabe, T. (1994). The cyclic AMP response element plays an essential role in the expression of the human prostaglandin- endoperoxide synthase 2 gene in differentiated U937 monocytic cells. FEBS Lett 350(1): 51-54. Inoue, H., Tanabe, T. and Umesono, K. (2000). Feedback control of cyclooxygenase-2 expression through PPARgamma. J Biol Chem 275(36): 28028-28032. Inoue, H., Yokoyama, C., Hara, S., Tone, Y. and Tanabe, T. (1995). Transcriptional regulation of human prostaglandin-endoperoxide synthase-2 gene by lipopolysaccharide and 354

phorbol ester in vascular endothelial cells. Involvement of both nuclear factor for interleukin-6 expression site and cAMP response element. J Biol Chem 270(42): 24965- 24971. Irvine, R. F. (1982). How is the level of free arachidonic acid controlled in mammalian cells? Biochem J 204(1): 3-16. Ishii, T. and Fridovich, I. (1990). Dual effects of superoxide dismutase on the autoxidation of 1,4-naphthohydroquinone. Free Radic Biol Med 8(1): 21-24. Itoh, K., Chiba, T., Takahashi, S., Ishii, T., Igarashi, K., Katoh, Y., Oyake, T., Hayashi, N., Satoh, K., Hatayama, I., Yamamoto, M. and Nabeshima, Y. (1997). An Nrf2/small Maf heterodimer mediates the induction of phase II detoxifying enzyme genes through antioxidant response elements. Biochem Biophys Res Commun 236(2): 313-322. Itoh, K., Igarashi, K., Hayashi, N., Nishizawa, M. and Yamamoto, M. (1995). Cloning and characterization of a novel erythroid cell-derived CNC family transcription factor heterodimerizing with the small Maf family proteins. Mol Cell Biol 15(8): 4184-4193. Itoh, K., Ishii, T., Wakabayashi, N. and Yamamoto, M. (1999a). Regulatory mechanisms of cellular response to oxidative stress. Free Radic Res 31(4): 319-324. Itoh, K., Wakabayashi, N., Katoh, Y., Ishii, T., Igarashi, K., Engel, J. D. and Yamamoto, M. (1999b). Keap1 represses nuclear activation of antioxidant responsive elements by Nrf2 through binding to the amino-terminal Neh2 domain. Genes Dev 13(1): 76-86. Itzhak, Y., Gandia, C., Huang, P. L. and Ali, S. F. (1998). Resistance of neuronal nitric oxide synthase-deficient mice to methamphetamine-induced dopaminergic neurotoxicity. J Pharmacol Exp Ther 284(3): 1040-1047. Iwasaki, K., Hailemariam, K. and Tsuji, Y. (2007). PIAS3 interacts with ATF1 and regulates the human ferritin H gene through an antioxidant-responsive element. J Biol Chem 282(31): 22335-22343. Iyengar, R., Stuehr, D. J. and Marletta, M. A. (1987). Macrophage synthesis of nitrite, nitrate, and N-nitrosamines: precursors and role of the respiratory burst. Proc Natl Acad Sci U S A 84(18): 6369-6373. Jain, A. K. and Jaiswal, A. K. (2006). Phosphorylation of tyrosine 568 controls nuclear export of Nrf2. J Biol Chem 281(17): 12132-12142. Jain, A. K. and Jaiswal, A. K. (2007). GSK-3beta acts upstream of Fyn kinase in regulation of nuclear export and degradation of NF-E2 related factor 2. J Biol Chem 282(22): 16502- 16510. Jakel, R. J., Kern, J. T., Johnson, D. A. and Johnson, J. A. (2005). Induction of the protective 355

antioxidant response element pathway by 6-hydroxydopamine in vivo and in vitro. Toxicol Sci 87(1): 176-186. Jakel, R. J., Townsend, J. A., Kraft, A. D. and Johnson, J. A. (2007). Nrf2-mediated protection against 6-hydroxydopamine. Brain Res 1144: 192-201. Jang, J. H. and Surh, Y. J. (2005). Beta-amyloid-induced apoptosis is associated with cyclooxygenase-2 up-regulation via the mitogen-activated protein kinase-NF-kappaB signaling pathway. Free Radic Biol Med 38(12): 1604-1613. Jantzen, P. T., Connor, K. E., DiCarlo, G., Wenk, G. L., Wallace, J. L., Rojiani, A. M., Coppola, D., Morgan, D. and Gordon, M. N. (2002). Microglial activation and beta -amyloid deposit reduction caused by a nitric oxide-releasing nonsteroidal anti-inflammatory drug in amyloid precursor protein plus presenilin-1 transgenic mice. J Neurosci 22(6): 2246- 2254. Janzen, E. G., Jandrisits, L. T., Shetty, R. V., Haire, D. L. and Hilborn, J. W. (1989). Synthesis and purification of 5,5-dimethyl-1-pyrroline-N-oxide for biological applications. Chem Biol Interact 70(1-2): 167-172. Jayanthi, S., Deng, X., Noailles, P. A., Ladenheim, B. and Cadet, J. L. (2004). Methamphetamine induces neuronal apoptosis via cross-talks between endoplasmic reticulum and mitochondria-dependent death cascades. FASEB J 18(2): 238-251. Jayanthi, S., Ladenheim, B., Andrews, A. M. and Cadet, J. L. (1999). Overexpression of human copper/zinc superoxide dismutase in transgenic mice attenuates oxidative stress caused by methylenedioxymethamphetamine (Ecstasy). Neuroscience 91(4): 1379-1387. Jayanthi, S., Ladenheim, B. and Cadet, J. L. (1998). Methamphetamine-induced changes in antioxidant enzymes and lipid peroxidation in copper/zinc-superoxide dismutase transgenic mice. Ann N Y Acad Sci 844: 92-102. Jayanthi, S., McCoy, M. T., Beauvais, G., Ladenheim, B., Gilmore, K., Wood, W., 3rd, Becker, K. and Cadet, J. L. (2009). Methamphetamine induces dopamine D1 receptor-dependent endoplasmic reticulum stress-related molecular events in the rat striatum. PLoS One 4(6): e6092. Jeng, W., Chrishti, A., Westaway, D. and Wells, P. G. (2002). Oxidative DNA damage precedes amyloid plaque formation in amyloid precursor protein (APP) trasgenic mice: an animal model of Alzheimer's disease. Toxicol Sci (Supplement: The Toxicologist) 66(1-S): 13 (No. 63). Jeng, W., Ramkissoon, A., Parman, T. and Wells, P. G. (2006). Prostaglandin H synthase- catalyzed bioactivation of amphetamines to free radical intermediates that cause CNS 356

regional DNA oxidation and nerve terminal degeneration. FASEB J 20(6): 638-650. Jeng, W., Ramkissoon, A. and Wells, P. G. (2011). Reduced DNA oxidation in aged prostaglandin H synthase-1 knockout mice. Free Radic Biol Med 50(4): 550-556. Jeng, W. and Wells, P. G. (2010). Reduced 3,4-Methylenedioxymethamphetamine (MDMA, Ecstasy)-Initiated Oxidative DNA Damage and Neurodegeneration in Prostaglandin H Synthase-1 Knockout Mice. ACS Chem Neurosci 1(5): 366-380. Jeng, W., Wong, A. W., Ting, A. K. R. and Wells, P. G. (2005). Methamphetamine-enhanced embryonic oxidative DNA damage and neurodevelopmental deficits. Free Radic Biol Med 39(3): 317-326. Jenner, P. (1998). Oxidative mechanisms in nigral cell death in Parkinson's disease. Mov Disord 13 Suppl 1: 24-34. Jeyapaul, J. and Jaiswal, A. K. (2000). Nrf2 and c-Jun regulation of antioxidant response element (ARE)-mediated expression and induction of gamma-glutamylcysteine synthetase heavy subunit gene. Biochem Pharmacol 59(11): 1433-1439. Jiang, J., Borisenko, G. G., Osipov, A., Martin, I., Chen, R., Shvedova, A. A., Sorokin, A., Tyurina, Y. Y., Potapovich, A., Tyurin, V. A., Graham, S. H. and Kagan, V. E. (2004). Arachidonic acid-induced carbon-centered radicals and phospholipid peroxidation in cyclo-oxygenase-2-transfected PC12 cells. J Neurochem 90(5): 1036-1049. Johanson, C. E., Frey, K. A., Lundahl, L. H., Keenan, P., Lockhart, N., Roll, J., Galloway, G. P., Koeppe, R. A., Kilbourn, M. R., Robbins, T. and Schuster, C. R. (2006). Cognitive function and nigrostriatal markers in abstinent methamphetamine abusers. Psychopharmacology (Berl) 185(3): 327-338. Johnson, D. A., Andrews, G. K., Xu, W. and Johnson, J. A. (2002). Activation of the antioxidant response element in primary cortical neuronal cultures derived from transgenic reporter mice. J Neurochem 81(6): 1233-1241. Johnson, F. and Giulivi, C. (2005). Superoxide dismutases and their impact upon human health. Mol Aspects Med 26(4-5): 340-352. Johnston, J. P. (1968). Some observations upon a new inhibitor of monoamine oxidase in brain tissue. Biochem Pharmacol 17(7): 1285-1297. Jones, D. A., Carlton, D. P., McIntyre, T. M., Zimmerman, G. A. and Prescott, S. M. (1993). Molecular cloning of human prostaglandin endoperoxide synthase type II and demonstration of expression in response to cytokines. J Biol Chem 268(12): 9049-9054. Jones, S. R., Gainetdinov, R. R., Wightman, R. M. and Caron, M. G. (1998). Mechanisms of amphetamine action revealed in mice lacking the dopamine transporter. J Neurosci 357

18(6): 1979-1986. Jonsson, G. and Nwanze, E. (1982). Selective (+)-amphetamine neurotoxicity on striatal dopamine nerve terminals in the mouse. Br J Pharmacol 77(2): 335-345. Juge, N., Mithen, R. F. and Traka, M. (2007). Molecular basis for chemoprevention by sulforaphane: a comprehensive review. Cell Mol Life Sci 64(9): 1105-1127. Kalant, H. (2001). The pharmacology and toxicology of "ecstasy" (MDMA) and related drugs. CMAJ 165(7): 917-928. Kalant, H., Grant, D. and Mitchell, J. (2006). Principles of medical pharmacology. Toronto, Elsevier Health Sciences Kalasinsky, K. S., Bosy, T. Z., Schmunk, G. A., Reiber, G., Anthony, R. M., Furukawa, Y., Guttman, M. and Kish, S. J. (2001). Regional distribution of methamphetamine in autopsied brain of chronic human methamphetamine users. Forensic Sci Int 116(2-3): 163-169. Kalyanaraman, B., Felix, C. C. and Sealy, R. C. (1982). Photoionozation of melanin precursors: An electron spin resonance investigation using the spin trap 5,5-dimethyl-1-pyrroline-1- oxide (DMPO). Photochem Photobiol 26: 5– 12. Kalyanaraman, B., Felix, C. C. and Sealy, R. C. (1984a). Electron spin resonance-spin stabilization of semiquinones produced during oxidation of epinephrine and its analogues. J Biol Chem 259(1): 354-358. Kalyanaraman, B., Felix, C. C. and Sealy, R. C. (1984b). Peroxidatic oxidation of catecholamines. A kinetic electron spin resonance investigation using the spin stabilization approach. J Biol Chem 259(12): 7584-7589. Kalyanaraman, B., Premovic, P. I. and Sealy, R. C. (1987). Semiquinone anion radicals from addition of amino acids, peptides, and proteins to quinones derived from oxidation of catechols and catecholamines. An ESR spin stabilization study. J Biol Chem 262(23): 11080-11087. Kalyanaraman, B., Sealy, R. C. and Sivarajah, K. (1984c). An electron spin resonance study of o-semiquinones formed during the enzymatic and autoxidation of catechol estrogens. J Biol Chem 259(22): 14018-14022. Kalyvas, A. and David, S. (2004). Cytosolic phospholipase A2 plays a key role in the pathogenesis of multiple sclerosis-like disease. Neuron 41(3): 323-335. Kaneda, H., Maeshima, K., Goto, N., Kobayakawa, T., Ayabe-Kanamura, S. and Saito, S. (2000). Decline in taste and odor discrimination abilities with age, and relationship between gustation and olfaction. Chem Senses 25(3): 331-337. 358

Kang, M. I., Kobayashi, A., Wakabayashi, N., Kim, S. G. and Yamamoto, M. (2004). Scaffolding of Keap1 to the actin cytoskeleton controls the function of Nrf2 as key regulator of cytoprotective phase 2 genes. Proc Natl Acad Sci U S A 101(7): 2046-2051. Kanninen, K., Heikkinen, R., Malm, T., Rolova, T., Kuhmonen, S., Leinonen, H., Yla-Herttuala, S., Tanila, H., Levonen, A. L., Koistinaho, M. and Koistinaho, J. (2009). Intrahippocampal injection of a lentiviral vector expressing Nrf2 improves spatial learning in a mouse model of Alzheimer's disease. Proc Natl Acad Sci U S A 106(38): 16505-16510. Kappos, L., Gold, R., Miller, D. H., Macmanus, D. G., Havrdova, E., Limmroth, V., Polman, C. H., Schmierer, K., Yousry, T. A., Yang, M., Eraksoy, M., Meluzinova, E., Rektor, I., Dawson, K. T., Sandrock, A. W. and O'Neill, G. N. (2008). Efficacy and safety of oral fumarate in patients with relapsing-remitting multiple sclerosis: a multicentre, randomised, double-blind, placebo-controlled phase IIb study. Lancet 372(9648): 1463- 1472. Kargman, S., Wong, E., Greig, G. M., Falgueyret, J. P., Cromlish, W., Ethier, D., Yergey, J. A., Riendeau, D., Evans, J. F., Kennedy, B., Tagari, P., Francis, D. A. and O'Neill, G. P. (1996). Mechanism of selective inhibition of human prostaglandin G/H synthase-1 and - 2 in intact cells. Biochem Pharmacol 52(7): 1113-1125. Karthein, R., Dietz, R., Nastainczyk, W. and Ruf, H. H. (1988). Higher oxidation states of prostaglandin H synthase. EPR study of a transient tyrosyl radical in the enzyme during the peroxidase reaction. Eur J Biochem 171(1-2): 313-320. Kaspar, J. W. and Jaiswal, A. K. (2010). Antioxidant-induced phosphorylation of tyrosine 486 leads to rapid nuclear export of Bach1 that allows Nrf2 to bind to the antioxidant response element and activate defensive gene expression. J Biol Chem 285(1): 153-162. Katoh, Y., Itoh, K., Yoshida, E., Miyagishi, M., Fukamizu, A. and Yamamoto, M. (2001). Two domains of Nrf2 cooperatively bind CBP, a CREB binding protein, and synergistically activate transcription. Genes Cells 6(10): 857-868. Katsuoka, F., Motohashi, H., Engel, J. D. and Yamamoto, M. (2005a). Nrf2 transcriptionally activates the mafG gene through an antioxidant response element. J Biol Chem 280(6): 4483-4490. Katsuoka, F., Motohashi, H., Ishii, T., Aburatani, H., Engel, J. D. and Yamamoto, M. (2005b). Genetic evidence that small maf proteins are essential for the activation of antioxidant response element-dependent genes. Mol Cell Biol 25(18): 8044-8051. Katsuyama, M., Sugimoto, Y., Morimoto, K., Hasumoto, K., Fukumoto, M., Negishi, M. and 359

Ichikawa, A. (1997). 'Distinct cellular localization' of the messenger ribonucleic acid for prostaglandin E receptor subtypes in the mouse uterus during pseudopregnancy. Endocrinology 138(1): 344-350. Kauffmann, H. M., Pfannschmidt, S., Zoller, H., Benz, A., Vorderstemann, B., Webster, J. I. and Schrenk, D. (2002). Influence of redox-active compounds and PXR-activators on human MRP1 and MRP2 gene expression. Toxicology 171(2-3): 137-146. Kaufmann, W. E., Andreasson, K. I., Isakson, P. C. and Worley, P. F. (1997). Cyclooxygenases and the central nervous system. Prostaglandins 54(3): 601-624. Kaufmann, W. E., Worley, P. F., Pegg, J., Bremer, M. and Isakson, P. (1996). COX-2, a synaptically induced enzyme, is expressed by excitatory neurons at postsynaptic sites in rat cerebral cortex. Proc Natl Acad Sci U S A 93(6): 2317-2321. Keum, Y. S., Owuor, E. D., Kim, B. R., Hu, R. and Kong, A. N. (2003). Involvement of Nrf2 and JNK1 in the activation of antioxidant responsive element (ARE) by chemopreventive agent phenethyl isothiocyanate (PEITC). Pharm Res 20(9): 1351-1356. Khor, T. O., Huang, M. T., Kwon, K. H., Chan, J. Y., Reddy, B. S. and Kong, A. N. (2006). Nrf2-deficient mice have an increased susceptibility to dextran sulfate sodium-induced colitis. Cancer Res 66(24): 11580-11584. Khoshbouei, H., Wang, H., Lechleiter, J. D., Javitch, J. A. and Galli, A. (2003). Amphetamine- induced dopamine efflux. A voltage-sensitive and intracellular Na+-dependent mechanism. J Biol Chem 278(14): 12070-12077. Ki, S. H., Cho, I. J., Choi, D. W. and Kim, S. G. (2005). Glucocorticoid receptor (GR)- associated SMRT binding to C/EBPbeta TAD and Nrf2 Neh4/5: role of SMRT recruited to GR in GSTA2 gene repression. Mol Cell Biol 25(10): 4150-4165. Kim, M. J., Shin, K. S., Chung, Y. B., Jung, K. W., Cha, C. I. and Shin, D. H. (2005). Immunohistochemical study of p47Phox and gp91Phox distributions in rat brain. Brain Res 1040(1-2): 178-186. Kim, Y. C., Masutani, H., Yamaguchi, Y., Itoh, K., Yamamoto, M. and Yodoi, J. (2001). Hemin-induced activation of the thioredoxin gene by Nrf2. A differential regulation of the antioxidant responsive element by a switch of its binding factors. J Biol Chem 276(21): 18399-18406. Kim, Y. J., Ahn, J. Y., Liang, P., Ip, C., Zhang, Y. and Park, Y. M. (2007). Human prx1 gene is a target of Nrf2 and is up-regulated by hypoxia/reoxygenation: implication to tumor biology. Cancer Res 67(2): 546-554. Kinsella, B. T., O'Mahony, D. J. and Fitzgerald, G. A. (1997). The human thromboxane A2 360

receptor alpha isoform (TP alpha) functionally couples to the G proteins Gq and G11 in vivo and is activated by the isoprostane 8-epi prostaglandin F2 alpha. J Pharmacol Exp Ther 281(2): 957-964. Kirtikara, K., Morham, S. G., Raghow, R., Laulederkind, S. J., Kanekura, T., Goorha, S. and Ballou, L. R. (1998). Compensatory prostaglandin E2 biosynthesis in cyclooxygenase 1 or 2 null cells. J Exp Med 187(4): 517-523. Kis, B., Snipes, J. A. and Busija, D. W. (2005). Acetaminophen and the cyclooxygenase-3 puzzle: sorting out facts, fictions, and uncertainties. J Pharmacol Exp Ther 315(1): 1-7. Kish, S. J., Fitzmaurice, P. S., Boileau, I., Schmunk, G. A., Ang, L. C., Furukawa, Y., Chang, L. J., Wickham, D. J., Sherwin, A. and Tong, J. (2009). Brain serotonin transporter in human methamphetamine users. Psychopharmacology (Berl) 202(4): 649-661. Kita, T., Shimada, K., Mastunari, Y., Wagner, G. C., Kubo, K. and Nakashima, T. (2000). Methamphetamine-induced striatal dopamine neurotoxicity and cyclooxygenase-2 protein expression in BALB/c mice. Neuropharmacology 39(3): 399-406. Kita, T., Wagner, G. C. and Nakashima, T. (2003). Current research on methamphetamine- induced neurotoxicity: animal models of monoamine disruption. J Pharmacol Sci 92(3): 178-195. Kitamura, Y., Shimohama, S., Koike, H., Kakimura, J., Matsuoka, Y., Nomura, Y., Gebicke- Haerter, P. J. and Taniguchi, T. (1999). Increased expression of cyclooxygenases and peroxisome proliferator-activated receptor-gamma in Alzheimer's disease brains. Biochem Biophys Res Commun 254(3): 582-586. Kitzler, J., Hill, E., Hardman, R., Reddy, N., Philpot, R. and Eling, T. E. (1995). Analysis and quantitation of splicing variants of the TPA-inducible PGHS-1 mRNA in rat tracheal epithelial cells. Arch Biochem Biophys 316(2): 856-863. Klegeris, A., Korkina, L. G. and Greenfield, S. A. (1995). Autoxidation of dopamine: a comparison of luminescent and spectrophotometric detection in basic solutions. Free Radic Biol Med 18(2): 215-222. Klivenyi, P., Andreassen, O. A., Ferrante, R. J., Dedeoglu, A., Mueller, G., Lancelot, E., Bogdanov, M., Andersen, J. K., Jiang, D. and Beal, M. F. (2000). Mice deficient in cellular glutathione peroxidase show increased vulnerability to malonate, 3- nitropropionic acid, and 1-methyl-4-phenyl-1,2,5,6-tetrahydropyridine. J Neurosci 20(1): 1-7. Klivenyi, P., Beal, M. F., Ferrante, R. J., Andreassen, O. A., Wermer, M., Chin, M. R. and Bonventre, J. V. (1998). Mice deficient in group IV cytosolic phospholipase A2 are 361

resistant to MPTP neurotoxicity. J Neurochem 71(6): 2634-2637. Klivenyi, P., Kiaei, M., Gardian, G., Calingasan, N. Y. and Beal, M. F. (2004). Additive neuroprotective effects of creatine and cyclooxygenase 2 inhibitors in a transgenic mouse model of amyotrophic lateral sclerosis. J Neurochem 88(3): 576-582. Knott, C., Stern, G. and Wilkin, G. P. (2000). Inflammatory regulators in Parkinson's disease: iNOS, lipocortin-1, and cyclooxygenases-1 and -2. Mol Cell Neurosci 16(6): 724-739. Kobayashi, A., Ito, E., Toki, T., Kogame, K., Takahashi, S., Igarashi, K., Hayashi, N. and Yamamoto, M. (1999). Molecular cloning and functional characterization of a new Cap'n' collar family transcription factor Nrf3. J Biol Chem 274(10): 6443-6452. Kobayashi, A., Kang, M. I., Okawa, H., Ohtsuji, M., Zenke, Y., Chiba, T., Igarashi, K. and Yamamoto, M. (2004). Oxidative stress sensor Keap1 functions as an adaptor for Cul3- based E3 ligase to regulate proteasomal degradation of Nrf2. Mol Cell Biol 24(16): 7130-7139. Kobayashi, A., Kang, M. I., Watai, Y., Tong, K. I., Shibata, T., Uchida, K. and Yamamoto, M. (2006). Oxidative and electrophilic stresses activate Nrf2 through inhibition of ubiquitination activity of Keap1. Mol Cell Biol 26(1): 221-229. Kobayashi, M., Itoh, K., Suzuki, T., Osanai, H., Nishikawa, K., Katoh, Y., Takagi, Y. and Yamamoto, M. (2002). Identification of the interactive interface and phylogenic conservation of the Nrf2-Keap1 system. Genes Cells 7(8): 807-820. Kojima, T., Une, I., Yashiki, M., Noda, J., Sakai, K. and Yamamoto, K. (1984). A fatal methamphetamine poisoning associated with hyperpyrexia. Forensic Sci Int 24(1): 87- 93. Kojo, H., Fukagawa, M., Tajima, K., Suzuki, A., Fujimura, T., Aramori, I., Hayashi, K. and Nishimura, S. (2003). Evaluation of human peroxisome proliferator-activated receptor (PPAR) subtype selectivity of a variety of anti-inflammatory drugs based on a novel assay for PPAR delta(beta). J Pharmacol Sci 93(3): 347-355. Kortekaas, R., Leenders, K. L., van Oostrom, J. C., Vaalburg, W., Bart, J., Willemsen, A. T. and Hendrikse, N. H. (2005). Blood-brain barrier dysfunction in parkinsonian midbrain in vivo. Ann Neurol 57(2): 176-179. Kosaka, T., Miyata, A., Ihara, H., Hara, S., Sugimoto, T., Takeda, O., Takahashi, E. and Tanabe, T. (1994). Characterization of the human gene (PTGS2) encoding prostaglandin- endoperoxide synthase 2. Eur J Biochem 221(3): 889-897. Koshkin, V. and Dunford, H. B. (1998). Reaction of prostaglandin endoperoxide synthase with cis,cis-eicosa-11,14-dienoic acid. J Biol Chem 273(11): 6046-6049. 362

Kotilinek, L. A., Westerman, M. A., Wang, Q., Panizzon, K., Lim, G. P., Simonyi, A., Lesne, S., Falinska, A., Younkin, L. H., Younkin, S. G., Rowan, M., Cleary, J., Wallis, R. A., Sun, G. Y., Cole, G., Frautschy, S., Anwyl, R. and Ashe, K. H. (2008). Cyclooxygenase- 2 inhibition improves amyloid-beta-mediated suppression of memory and synaptic plasticity. Brain 131(Pt 3): 651-664. Kraemer, T. and Maurer, H. H. (2002). Toxicokinetics of amphetamines: metabolism and toxicokinetic data of designer drugs, amphetamine, methamphetamine, and their N-alkyl derivatives. Ther Drug Monit 24(2): 277-289. Kraft, A. D., Johnson, D. A. and Johnson, J. A. (2004). Nuclear factor E2-related factor 2- dependent antioxidant response element activation by tert-butylhydroquinone and sulforaphane occurring preferentially in astrocytes conditions neurons against oxidative insult. J Neurosci 24(5): 1101-1112. Kraft, A. D., Lee, J. M., Johnson, D. A., Kan, Y. W. and Johnson, J. A. (2006). Neuronal sensitivity to kainic acid is dependent on the Nrf2-mediated actions of the antioxidant response element. J Neurochem 98(6): 1852-1865. Kreth, K., Kovar, K., Schwab, M. and Zanger, U. M. (2000). Identification of the human cytochromes P450 involved in the oxidative metabolism of "Ecstasy"-related designer drugs. Biochem Pharmacol 59(12): 1563-1571. Kubow, S. and Wells, P. G. (1989). In vitro bioactivation of phenytoin to a reactive free radical intermediate by prostaglandin synthetase, horseradish peroxidase, and thyroid peroxidase. Mol Pharmacol 35(4): 504-511. Kuczenski, R., Everall, I. P., Crews, L., Adame, A., Grant, I. and Masliah, E. (2007). Escalating dose-multiple binge methamphetamine exposure results in degeneration of the neocortex and limbic system in the rat. Exp Neurol 207(1): 42-51. Kuczenski, R., Segal, D. S., Cho, A. K. and Melega, W. (1995). Hippocampus norepinephrine, caudate dopamine and serotonin, and behavioral responses to the stereoisomers of amphetamine and methamphetamine. J Neurosci 15(2): 1308-1317. Kujubu, D. A., Fletcher, B. S., Varnum, B. C., Lim, R. W. and Herschman, H. R. (1991). TIS10, a phorbol ester tumor promoter-inducible mRNA from Swiss 3T3 cells, encodes a novel prostaglandin synthase/cyclooxygenase homologue. J Biol Chem 266(20): 12866-12872. Kulmacz, R. J. (1998). Cellular regulation of prostaglandin H synthase catalysis. FEBS Lett 430(3): 154-157. Kulmacz, R. J. (2005). Regulation of cyclooxygenase catalysis by hydroperoxides. Biochem Biophys Res Commun 338(1): 25-33. 363

Kulmacz, R. J., Pendleton, R. B. and Lands, W. E. (1994). Interaction between peroxidase and cyclooxygenase activities in prostaglandin-endoperoxide synthase. Interpretation of reaction kinetics. J Biol Chem 269(8): 5527-5536. Kulmacz, R. J., van der Donk, W. A. and Tsai, A. L. (2003). Comparison of the properties of prostaglandin H synthase-1 and -2. Prog Lipid Res 42(5): 377-404. Kulmacz, R. J. and Wang, L. H. (1995). Comparison of hydroperoxide initiator requirements for the cyclooxygenase activities of prostaglandin H synthase-1 and -2. J Biol Chem 270(41): 24019-24023. Kumaran, R., Kingsbury, A., Coulter, I., Lashley, T., Williams, D., de Silva, R., Mann, D., Revesz, T., Lees, A. and Bandopadhyay, R. (2007). DJ-1 (PARK7) is associated with 3R and 4R tau neuronal and glial inclusions in neurodegenerative disorders. Neurobiol Dis 28(1): 122-132. Kurumbail, R. G., Stevens, A. M., Gierse, J. K., McDonald, J. J., Stegeman, R. A., Pak, J. Y., Gildehaus, D., Miyashiro, J. M., Penning, T. D., Seibert, K., Isakson, P. C. and Stallings, W. C. (1996). Structural basis for selective inhibition of cyclooxygenase-2 by anti- inflammatory agents. Nature 384(6610): 644-648. Kwak, M. K., Itoh, K., Yamamoto, M. and Kensler, T. W. (2002). Enhanced expression of the transcription factor Nrf2 by cancer chemopreventive agents: role of antioxidant response element-like sequences in the nrf2 promoter. Mol Cell Biol 22(9): 2883-2892. Kwak, M. K., Itoh, K., Yamamoto, M., Sutter, T. R. and Kensler, T. W. (2001). Role of transcription factor Nrf2 in the induction of hepatic phase 2 and antioxidative enzymes in vivo by the cancer chemoprotective agent, 3H-1, 2-dimethiole-3-thione. Mol Med 7(2): 135-145. Kwak, M. K., Kensler, T. W. and Casero, R. A., Jr. (2003a). Induction of phase 2 enzymes by serum oxidized polyamines through activation of Nrf2: effect of the polyamine metabolite acrolein. Biochem Biophys Res Commun 305(3): 662-670. Kwak, M. K., Wakabayashi, N., Greenlaw, J. L., Yamamoto, M. and Kensler, T. W. (2003b). Antioxidants enhance mammalian proteasome expression through the Keap1-Nrf2 signaling pathway. Mol Cell Biol 23(23): 8786-8794. Kwak, M. K., Wakabayashi, N., Itoh, K., Motohashi, H., Yamamoto, M. and Kensler, T. W. (2003c). Modulation of gene expression by cancer chemopreventive dithiolethiones through the Keap1-Nrf2 pathway. Identification of novel gene clusters for cell survival. J Biol Chem 278(10): 8135-8145. Lai, C. T. and Yu, P. H. (1997). Dopamine- and L-beta-3,4-dihydroxyphenylalanine 364

hydrochloride (L-Dopa)-induced cytotoxicity towards catecholaminergic neuroblastoma SH-SY5Y cells. Effects of oxidative stress and antioxidative factors. Biochem Pharmacol 53(3): 363-372. Landino, L. M., Crews, B. C., Gierse, J. K., Hauser, S. D. and Marnett, L. J. (1997). Mutational analysis of the role of the distal histidine and glutamine residues of prostaglandin- endoperoxide synthase-2 in peroxidase catalysis, hydroperoxide reduction, and cyclooxygenase activation. J Biol Chem 272(34): 21565-21574. Laneuville, O., Breuer, D. K., Xu, N., Huang, Z. H., Gage, D. A., Watson, J. T., Lagarde, M., DeWitt, D. L. and Smith, W. L. (1995). Fatty acid substrate specificities of human prostaglandin-endoperoxide H synthase-1 and -2. Formation of 12-hydroxy-(9Z, 13E/Z, 15Z)- octadecatrienoic acids from alpha-linolenic acid. J Biol Chem 270(33): 19330- 19336. Larsson, M., Lundin, A. and Robins Wahlin, T. B. (2006). Olfactory functions in asymptomatic carriers of the Huntington disease mutation. J Clin Exp Neuropsychol 28(8): 1373-1380. Lavigne, M. C., Malech, H. L., Holland, S. M. and Leto, T. L. (2001). Genetic requirement of p47phox for superoxide production by murine microglia. FASEB J 15(2): 285-287. Lawson, L. J., Perry, V. H., Dri, P. and Gordon, S. (1990). Heterogeneity in the distribution and morphology of microglia in the normal adult mouse brain. Neuroscience 39(1): 151-170. Lecomte, M., Laneuville, O., Ji, C., DeWitt, D. L. and Smith, W. L. (1994). Acetylation of human prostaglandin endoperoxide synthase-2 (cyclooxygenase-2) by aspirin. J Biol Chem 269(18): 13207-13215. Lee, J. M., Calkins, M. J., Chan, K., Kan, Y. W. and Johnson, J. A. (2003a). Identification of the NF-E2-related factor-2-dependent genes conferring protection against oxidative stress in primary cortical astrocytes using oligonucleotide microarray analysis. J Biol Chem 278(14): 12029-12038. Lee, J. M., Li, J., Johnson, D. A., Stein, T. D., Kraft, A. D., Calkins, M. J., Jakel, R. J. and Johnson, J. A. (2005). Nrf2, a multi-organ protector? FASEB J 19(9): 1061-1066. Lee, J. M., Moehlenkamp, J. D., Hanson, J. M. and Johnson, J. A. (2001). Nrf2-dependent activation of the antioxidant responsive element by tert-butylhydroquinone is independent of oxidative stress in IMR-32 human neuroblastoma cells. Biochem Biophys Res Commun 280(1): 286-292. Lee, J. M., Shih, A. Y., Murphy, T. H. and Johnson, J. A. (2003b). NF-E2-related factor-2 mediates neuroprotection against mitochondrial complex I inhibitors and increased concentrations of intracellular calcium in primary cortical neurons. J Biol Chem 278(39): 365

37948-37956. Lee, O. H., Jain, A. K., Papusha, V. and Jaiswal, A. K. (2007). An auto-regulatory loop between stress sensors INrf2 and Nrf2 controls their cellular abundance. J Biol Chem 282(50): 36412-36420. Lee, S. C., Dickson, D. W., Liu, W. and Brosnan, C. F. (1993). Induction of nitric oxide synthase activity in human astrocytes by interleukin-1 beta and interferon-gamma. J Neuroimmunol 46(1-2): 19-24. LeFevre, P. G. and Peters, A. A. (1966). Evidence of mediated transfer of monosaccharides from blood to brain in rodents. J Neurochem 13(1): 35-46. Lehmann, K. and Lehmann, D. (2007). Transmitter balances in the olfactory cortex: adaptations to early methamphetamine trauma and rearing environment. Brain Research 1141: 37- 47. Leung, L., Kwong, M., Hou, S., Lee, C. and Chan, J. Y. (2003). Deficiency of the Nrf1 and Nrf2 transcription factors results in early embryonic lethality and severe oxidative stress. J Biol Chem 278(48): 48021-48029. Lev, N., Ickowicz, D., Barhum, Y., Lev, S., Melamed, E. and Offen, D. (2009). DJ-1 protects against dopamine toxicity. J Neural Transm 116(2): 151-160. Levay, G., Ye, Q. and Bodell, W. J. (1997). Formation of DNA adducts and oxidative base damage by copper mediated oxidation of dopamine and 6-hydroxydopamine. Exp Neurol 146(2): 570-574. Levitt, P., Harvey, J. A., Friedman, E., Simansky, K. and Murphy, E. H. (1997). New evidence for neurotransmitter influences on brain development. Trends Neurosci 20(6): 269-274. Levitt, P., Pintar, J. E. and Breakefield, X. O. (1982). Immunocytochemical demonstration of monoamine oxidase B in brain astrocytes and serotonergic neurons. Proc Natl Acad Sci U S A 79(20): 6385-6389. Li, J., Johnson, D., Calkins, M., Wright, L., Svendsen, C. and Johnson, J. (2005). Stabilization of Nrf2 by tBHQ confers protection against oxidative stress-induced cell death in human neural stem cells. Toxicol Sci 83(2): 313-328. Li, J., Lee, J. M. and Johnson, J. A. (2002). Microarray analysis reveals an antioxidant responsive element-driven gene set involved in conferring protection from an oxidative stress-induced apoptosis in IMR-32 cells. J Biol Chem 277(1): 388-394. Li, J., Stein, T. D. and Johnson, J. A. (2004). Genetic dissection of systemic autoimmune disease in Nrf2-deficient mice. Physiol Genomics 18(3): 261-272. Liby, K., Hock, T., Yore, M. M., Suh, N., Place, A. E., Risingsong, R., Williams, C. R., Royce, 366

D. B., Honda, T., Honda, Y., Gribble, G. W., Hill-Kapturczak, N., Agarwal, A. and Sporn, M. B. (2005). The synthetic triterpenoids, CDDO and CDDO-imidazolide, are potent inducers of heme oxygenase-1 and Nrf2/ARE signaling. Cancer Res 65(11): 4789-4798. Liddell, J. R., Dringen, R., Crack, P. J. and Robinson, S. R. (2006a). Glutathione peroxidase 1 and a high cellular glutathione concentration are essential for effective organic hydroperoxide detoxification in astrocytes. Glia 54(8): 873-879. Liddell, J. R., Hoepken, H. H., Crack, P. J., Robinson, S. R. and Dringen, R. (2006b). Glutathione peroxidase 1 and glutathione are required to protect mouse astrocytes from iron-mediated hydrogen peroxide toxicity. J Neurosci Res 84(3): 578-586. Liu, G. H., Qu, J. and Shen, X. (2008). NF-kappaB/p65 antagonizes Nrf2-ARE pathway by depriving CBP from Nrf2 and facilitating recruitment of HDAC3 to MafK. Biochim Biophys Acta 1783(5): 713-727. Liu, J. S., Zhao, M. L., Brosnan, C. F. and Lee, S. C. (2001). Expression of inducible nitric oxide synthase and nitrotyrosine in multiple sclerosis lesions. Am J Pathol 158(6): 2057- 2066. Liu, Y. and Levy, G. N. (1998). Activation of heterocyclic amines by combinations of prostaglandin H synthase-1 and -2 with N-acetyltransferase 1 and 2. Cancer Lett 133(1): 115-123. Liu, Y., Levy, G. N. and Weber, W. W. (1995). Activation of 2-aminofluorene by prostaglandin endoperoxide H synthase-2. Biochem Biophys Res Commun 215(1): 346-354. Lo, S. C. and Hannink, M. (2008). PGAM5 tethers a ternary complex containing Keap1 and Nrf2 to mitochondria. Exp Cell Res 314(8): 1789-1803. Logan, B. K. (1996). Methamphetamine and driving impairment. J Forensic Sci 41(3): 457-464. Loll, P. J., Picot, D. and Garavito, R. M. (1995). The structural basis of aspirin activity inferred from the crystal structure of inactivated prostaglandin H2 synthase. Nat Struct Biol 2(8): 637-643. Loutelier-Bourhis, C., Legros, H., Bonnet, J. J., Costentin, J. and Lange, C. M. (2004). Gas chromatography/mass spectrometric identification of dopaminergic metabolites in striata of rats treated with L-DOPA. Rapid Commun Mass Spectrom 18(5): 571-576. Lucas, D. R. and Newhouse, J. P. (1957). The toxic effect of sodium L-glutamate on the inner layers of the retina. AMA Arch Ophthalmol 58(2): 193-201. Lucius, R. and Sievers, J. (1996). Postnatal retinal ganglion cells in vitro: protection against reactive oxygen species (ROS)-induced axonal degeneration by cocultured astrocytes. 367

Brain Res 743(1-2): 56-62. Luna, L., Johnsen, O., Skartlien, A. H., Pedeutour, F., Turc-Carel, C., Prydz, H. and Kolsto, A. B. (1994). Molecular cloning of a putative novel human bZIP transcription factor on chromosome 17q22. Genomics 22(3): 553-562. Luna, L., Skammelsrud, N., Johnsen, O., Abel, K. J., Weber, B. L., Prydz, H. and Kolsto, A. B. (1995). Structural organization and mapping of the human TCF11 gene. Genomics 27(2): 237-244. Luong, C., Miller, A., Barnett, J., Chow, J., Ramesha, C. and Browner, M. F. (1996). Flexibility of the NSAID binding site in the structure of human cyclooxygenase-2. Nat Struct Biol 3(11): 927-933. Ma, Q., Battelli, L. and Hubbs, A. F. (2006). Multiorgan autoimmune inflammation, enhanced lymphoproliferation, and impaired homeostasis of reactive oxygen species in mice lacking the antioxidant-activated transcription factor Nrf2. Am J Pathol 168(6): 1960- 1974. Magyar, K. and Knoll, J. (1977). Selective inhibition of the "B form" of monoamine oxidase. Pol J Pharmacol Pharm 29(3): 233-246. Mahboubi, K., Young, W. and Ferreri, N. R. (1998). Tumour necrosis factor-alpha-dependent regulation of prostaglandin endoperoxide synthase-2. Cytokine 10(3): 175-184. Maher, J. M., Dieter, M. Z., Aleksunes, L. M., Slitt, A. L., Guo, G., Tanaka, Y., Scheffer, G. L., Chan, J. Y., Manautou, J. E., Chen, Y., Dalton, T. P., Yamamoto, M. and Klaassen, C. D. (2007). Oxidative and electrophilic stress induces multidrug resistance-associated protein transporters via the nuclear factor-E2-related factor-2 transcriptional pathway. Hepatology 46(5): 1597-1610. Maier, C. M. and Chan, P. H. (2002). Role of superoxide dismutases in oxidative damage and neurodegenerative disorders. Neuroscientist 8(4): 323-334. Maihofner, C., Probst-Cousin, S., Bergmann, M., Neuhuber, W., Neundorfer, B. and Heuss, D. (2003). Expression and localization of cyclooxygenase-1 and -2 in human sporadic amyotrophic lateral sclerosis. Eur J Neurosci 18(6): 1527-1534. Mancini, J. A., Riendeau, D., Falgueyret, J. P., Vickers, P. J. and O'Neill, G. P. (1995). Arginine 120 of prostaglandin G/H synthase-1 is required for the inhibition by nonsteroidal anti- inflammatory drugs containing a carboxylic acid moiety. J Biol Chem 270(49): 29372- 29377. Marcus, D. L., Thomas, C., Rodriguez, C., Simberkoff, K., Tsai, J. S., Strafaci, J. A. and Freedman, M. L. (1998). Increased peroxidation and reduced antioxidant enzyme 368

activity in Alzheimer's disease. Exp Neurol 150(1): 40-44. Mari, M. and Cederbaum, A. I. (2000). CYP2E1 overexpression in HepG2 cells induces glutathione synthesis by transcriptional activation of gamma-glutamylcysteine synthetase. J Biol Chem 275(20): 15563-15571. Marini, M. G., Chan, K., Casula, L., Kan, Y. W., Cao, A. and Moi, P. (1997). hMAF, a small human transcription factor that heterodimerizes specifically with Nrf1 and Nrf2. J Biol Chem 272(26): 16490-16497. Markey, C. M., Alward, A., Weller, P. E. and Marnett, L. J. (1987). Quantitative studies of hydroperoxide reduction by prostaglandin H synthase. Reducing substrate specificity and the relationship of peroxidase to cyclooxygenase activities. J Biol Chem 262(13): 6266- 6279. Marklund, S. L. (1982). Human copper-containing superoxide dismutase of high molecular weight. Proc Natl Acad Sci U S A 79(24): 7634-7638. Marnett, L. J. (1990). Prostaglandin synthase-mediated metabolism of carcinogens and a potential role for peroxyl radicals as reactive intermediates. Environ Health Perspect 88: 5-12. Martino, G., Adorini, L., Rieckmann, P., Hillert, J., Kallmann, B., Comi, G. and Filippi, M. (2002). Inflammation in multiple sclerosis: the good, the bad, and the complex. Lancet Neurol 1(8): 499-509. Marzec, J. M., Christie, J. D., Reddy, S. P., Jedlicka, A. E., Vuong, H., Lanken, P. N., Aplenc, R., Yamamoto, T., Yamamoto, M., Cho, H. Y. and Kleeberger, S. R. (2007). Functional polymorphisms in the transcription factor NRF2 in humans increase the risk of acute lung injury. FASEB J 21(9): 2237-2246. Maslinska, D., Wozniak, R., Kaliszek, A. and Modelska, I. (1999). Expression of cyclooxygenase-2 in astrocytes of human brain after global ischemia. Folia Neuropathol 37(2): 75-79. Matsumura, K., Cao, C., Ozaki, M., Morii, H., Nakadate, K. and Watanabe, Y. (1998). Brain endothelial cells express cyclooxygenase-2 during lipopolysaccharide-induced fever: light and electron microscopic immunocytochemical studies. J Neurosci 18(16): 6279- 6289. Matsuoka, Y., Picciano, M., Malester, B., LaFrancois, J., Zehr, C., Daeschner, J. M., Olschowka, J. A., Fonseca, M. I., O'Banion, M. K., Tenner, A. J., Lemere, C. A. and Duff, K. (2001). Inflammatory responses to amyloidosis in a transgenic mouse model of Alzheimer's disease. Am J Pathol 158(4): 1345-1354. 369

Matsuzawa, A., Murakami, M., Atsumi, G., Imai, K., Prados, P., Inoue, K. and Kudo, I. (1996). Release of secretory phospholipase A2 from rat neuronal cells and its possible function in the regulation of catecholamine secretion. Biochem J 318 ( Pt 2): 701-709. Mattammal, M. B., Strong, R., Lakshmi, V. M., Chung, H. D. and Stephenson, A. H. (1995). Prostaglandin H synthetase-mediated metabolism of dopamine: implication for Parkinson's disease. J Neurochem 64(4): 1645-1654. Matute, C., Alberdi, E., Domercq, M., Perez-Cerda, F., Perez-Samartin, A. and Sanchez-Gomez, M. V. (2001). The link between excitotoxic oligodendroglial death and demyelinating diseases. Trends Neurosci 24(4): 224-230. McCallum, G. P., Wong, A. W. and Wells, P. G. (2011). Cockayne Syndrome B (Csb) Protects against Methamphetamine-Enhanced Oxidative DNA Damage in Murine Fetal Brain and Postnatal Neurodevelopmental Deficits. Antioxid Redox Signal 14(5): 747-756. McCormick, M. L., Gaut, J. P., Lin, T. S., Britigan, B. E., Buettner, G. R. and Heinecke, J. W. (1998). Electron paramagnetic resonance detection of free tyrosyl radical generated by myeloperoxidase, lactoperoxidase, and horseradish peroxidase. J Biol Chem 273(48): 32030-32037. McGeer, P. L., Itagaki, S., Boyes, B. E. and McGeer, E. G. (1988). Reactive microglia are positive for HLA-DR in the substantia nigra of Parkinson's and Alzheimer's disease brains. Neurology 38(8): 1285-1291. McGeer, P. L., Itagaki, S., Tago, H. and McGeer, E. G. (1987). Reactive microglia in patients with senile dementia of the Alzheimer type are positive for the histocompatibility glycoprotein HLA-DR. Neurosci Lett 79(1-2): 195-200. McGuire, P. (2000). Long term psychiatric and cognitive effects of MDMA use. Toxicol Lett 112-113: 153-156. McKetin, R., Kelly, E. and McLaren, J. (2006). The relationship between crystalline methamphetamine use and methamphetamine dependence. Drug Alcohol Depend 85(3): 198-204. McMahon, M., Thomas, N., Itoh, K., Yamamoto, M. and Hayes, J. D. (2004). Redox-regulated turnover of Nrf2 is determined by at least two separate protein domains, the redox- sensitive Neh2 degron and the redox-insensitive Neh6 degron. J Biol Chem 279(30): 31556-31567. Meade, E. A., Smith, W. L. and DeWitt, D. L. (1993). Differential inhibition of prostaglandin endoperoxide synthase (cyclooxygenase) isozymes by aspirin and other non-steroidal anti-inflammatory drugs. J Biol Chem 268(9): 6610-6614. 370

Mechan, A. O., Esteban, B., O'Shea, E., Elliott, J. M., Colado, M. I. and Green, A. R. (2002). The pharmacology of the acute hyperthermic response that follows administration of 3,4- methylenedioxymethamphetamine (MDMA, 'ecstasy') to rats. Br J Pharmacol 135(1): 170-180. Melega, W. P., Cho, A. K., Harvey, D. and Lacan, G. (2007). Methamphetamine blood concentrations in human abusers: application to pharmacokinetic modeling. Synapse 61(4): 216-220. Melega, W. P., Williams, A. E., Schmitz, D. A., DiStefano, E. W. and Cho, A. K. (1995). Pharmacokinetic and pharmacodynamic analysis of the actions of D-amphetamine and D-methamphetamine on the dopamine terminal. J Pharmacol Exp Ther 274(1): 90-96. Melo, P., Pinazo-Duran, M. D., Salgado-Borges, J. and Tavares, M. A. (2008). Correlation of axon size and myelin occupancy in rats prenatally exposed to methamphetamine. Brain Res 1222: 61-68. Melov, S., Schneider, J. A., Day, B. J., Hinerfeld, D., Coskun, P., Mirra, S. S., Crapo, J. D. and Wallace, D. C. (1998). A novel neurological phenotype in mice lacking mitochondrial manganese superoxide dismutase. Nat Genet 18(2): 159-163. Merlie, J. P., Fagan, D., Mudd, J. and Needleman, P. (1988). Isolation and characterization of the complementary DNA for sheep seminal vesicle prostaglandin endoperoxide synthase (cyclooxygenase). J Biol Chem 263(8): 3550-3553. Meyer, M. R., Peters, F. T. and Maurer, H. H. (2008). The role of human hepatic cytochrome P450 isozymes in the metabolism of racemic 3,4-methylenedioxy-methamphetamine and its enantiomers. Drug Metab Dispos 36(11): 2345-2354. Miao, W., Hu, L., Scrivens, P. J. and Batist, G. (2005). Transcriptional regulation of NF-E2 p45-related factor (NRF2) expression by the aryl hydrocarbon receptor-xenobiotic response element signaling pathway: direct cross-talk between phase I and II drug- metabolizing enzymes. J Biol Chem 280(21): 20340-20348. Miller, D. M., Buettner, G. R. and Aust, S. D. (1990). Transition metals as catalysts of "autoxidation" reactions. Free Radic Biol Med 8(1): 95-108. Ming, X. F., Kaiser, M. and Moroni, C. (1998). c-jun N-terminal kinase is involved in AUUUA- mediated interleukin-3 mRNA turnover in mast cells. EMBO J 17(20): 6039-6048. Minghetti, L., Hughes, P. and Perry, V. H. (1999). Restricted cyclooxygenase-2 expression in the central nervous system following acute and delayed-type hypersensitivity responses to bacillus Calmette-Guerin. Neuroscience 92(4): 1405-1415. Misra, H. P. and Fridovich, I. (1972). The role of superoxide anion in the autoxidation of 371

epinephrine and a simple assay for superoxide dismutase. J Biol Chem 247(10): 3170- 3175. Mitchell, J. A., Akarasereenont, P., Thiemermann, C., Flower, R. J. and Vane, J. R. (1993). Selectivity of nonsteroidal antiinflammatory drugs as inhibitors of constitutive and inducible cyclooxygenase. Proc Natl Acad Sci U S A 90(24): 11693-11697. Mittelbronn, M., Dietz, K., Schluesener, H. J. and Meyermann, R. (2001). Local distribution of microglia in the normal adult human central nervous system differs by up to one order of magnitude. Acta Neuropathol 101(3): 249-255. Miyazaki, I., Asanuma, M., Diaz-Corrales, F. J., Fukuda, M., Kitaichi, K., Miyoshi, K. and Ogawa, N. (2006). Methamphetamine-induced dopaminergic neurotoxicity is regulated by quinone-formation-related molecules. FASEB J 20(3): 571-573. Mohler, J., Vani, K., Leung, S. and Epstein, A. (1991). Segmentally restricted, cephalic expression of a leucine zipper gene during Drosophila embryogenesis. Mech Dev 34(1): 3-9. Moi, P., Chan, K., Asunis, I., Cao, A. and Kan, Y. W. (1994). Isolation of NF-E2-related factor 2 (Nrf2), a NF-E2-like basic leucine zipper transcriptional activator that binds to the tandem NF-E2/AP1 repeat of the beta-globin locus control region. Proc Natl Acad Sci U S A 91(21): 9926-9930. Moinova, H. R. and Mulcahy, R. T. (1998). An electrophile responsive element (EpRE) regulates beta-naphthoflavone induction of the human gamma-glutamylcysteine synthetase regulatory subunit gene. Constitutive expression is mediated by an adjacent AP-1 site. J Biol Chem 273(24): 14683-14689. Monaghan, D. T., Bridges, R. J. and Cotman, C. W. (1989). The excitatory amino acid receptors: their classes, pharmacology, and distinct properties in the function of the central nervous system. Annu Rev Pharmacol Toxicol 29: 365-402. Montine, T. J., Picklo, M. J., Amarnath, V., Whetsell, W. O., Jr. and Graham, D. G. (1997). Neurotoxicity of endogenous cysteinylcatechols. Exp Neurol 148(1): 26-33. Montine, T. J., Underhill, T. M., Linney, E. and Graham, D. G. (1994). Fibroblasts that express aromatic amino acid decarboxylase have increased sensitivity to the synergistic cytotoxicity of L-dopa and manganese. Toxicol Appl Pharmacol 128(1): 116-122. Moosmann, B. and Behl, C. (2002). Antioxidants as treatment for neurodegenerative disorders. Expert Opin Investig Drugs 11(10): 1407-1435. Morihara, T., Teter, B., Yang, F., Lim, G. P., Boudinot, S., Boudinot, F. D., Frautschy, S. A. and Cole, G. M. (2005). Ibuprofen suppresses interleukin-1beta induction of pro- 372

amyloidogenic alpha1-antichymotrypsin to ameliorate beta-amyloid (Abeta) pathology in Alzheimer's models. Neuropsychopharmacology 30(6): 1111-1120. Morita, I., Schindler, M., Regier, M. K., Otto, J. C., Hori, T., DeWitt, D. L. and Smith, W. L. (1995). Different intracellular locations for prostaglandin endoperoxide H synthase-1 and -2. J Biol Chem 270(18): 10902-10908. Morris, J. K. and Richards, J. S. (1996). An E-box region within the prostaglandin endoperoxide synthase-2 (PGS-2) promoter is required for transcription in rat ovarian granulosa cells. J Biol Chem 271(28): 16633-16643. Mosharov, E. V., Gong, L. W., Khanna, B., Sulzer, D. and Lindau, M. (2003). Intracellular patch electrochemistry: regulation of cytosolic catecholamines in chromaffin cells. J Neurosci 23(13): 5835-5845. Moszczynska, A., Fitzmaurice, P., Ang, L., Kalasinsky, K. S., Schmunk, G. A., Peretti, F. J., Aiken, S. S., Wickham, D. J. and Kish, S. J. (2004). Why is parkinsonism not a feature of human methamphetamine users? Brain 127(Pt 2): 363-370. Motohashi, H., Katsuoka, F., Shavit, J. A., Engel, J. D. and Yamamoto, M. (2000). Positive or negative MARE-dependent transcriptional regulation is determined by the abundance of small Maf proteins. Cell 103(6): 865-875. Motohashi, H. and Yamamoto, M. (2004). Nrf2-Keap1 defines a physiologically important stress response mechanism. Trends Mol Med 10(11): 549-557. Muller, D. P. and Goss-Sampson, M. A. (1990). Neurochemical, neurophysiological, and neuropathological studies in vitamin E deficiency. Crit Rev Neurobiol 5(3): 239-263. Mullett, S. J. and Hinkle, D. A. (2009). DJ-1 knock-down in astrocytes impairs astrocyte- mediated neuroprotection against rotenone. Neurobiol Dis 33(1): 28-36. Murphy, R. C. and Gijon, M. A. (2007). Biosynthesis and metabolism of leukotrienes. Biochem J 405(3): 379-395. Murphy, S. (2000). Production of nitric oxide by glial cells: regulation and potential roles in the CNS. Glia 29(1): 1-13. Murphy, T. H., Yu, J., Ng, R., Johnson, D. A., Shen, H., Honey, C. R. and Johnson, J. A. (2001). Preferential expression of antioxidant response element mediated gene expression in astrocytes. J Neurochem 76(6): 1670-1678. Musshoff, F. (2000). Illegal or legitimate use? Precursor compounds to amphetamine and methamphetamine. Drug Metab Rev 32(1): 15-44. Muto, A., Hoshino, H., Madisen, L., Yanai, N., Obinata, M., Karasuyama, H., Hayashi, N., Nakauchi, H., Yamamoto, M., Groudine, M. and Igarashi, K. (1998). Identification of 373

Bach2 as a B-cell-specific partner for small maf proteins that negatively regulate the immunoglobulin heavy chain gene 3' enhancer. EMBO J 17(19): 5734-5743. Muto, A., Tashiro, S., Tsuchiya, H., Kume, A., Kanno, M., Ito, E., Yamamoto, M. and Igarashi, K. (2002). Activation of Maf/AP-1 repressor Bach2 by oxidative stress promotes apoptosis and its interaction with promyelocytic leukemia nuclear bodies. J Biol Chem 277(23): 20724-20733. Nakamichi, K., Saiki, M., Sawada, M., Yamamuro, Y., Morimoto, K. and Kurane, I. (2005). Double-stranded RNA stimulates chemokine expression in microglia through vacuolar pH-dependent activation of intracellular signaling pathways. J Neurochem 95(1): 273- 283. Nakaso, K., Nakamura, C., Sato, H., Imamura, K., Takeshima, T. and Nakashima, K. (2006). Novel cytoprotective mechanism of anti-parkinsonian drug deprenyl: PI3K and Nrf2- derived induction of antioxidative proteins. Biochem Biophys Res Commun 339(3): 915-922. Namba, T., Sugimoto, Y., Hirata, M., Hayashi, Y., Honda, A., Watabe, A., Negishi, M., Ichikawa, A. and Narumiya, S. (1992). Mouse thromboxane A2 receptor: cDNA cloning, expression and northern blot analysis. Biochem Biophys Res Commun 184(3): 1197- 1203. Napolitano, A., Crescenzi, O., Pezzella, A. and Prota, G. (1995). Generation of the neurotoxin 6-hydroxydopamine by peroxidase/H2O2 oxidation of dopamine. J Med Chem 38(6): 917-922. Nara, A., Aki, T., Funakoshi, T. and Uemura, K. (2010). Methamphetamine induces macropinocytosis in differentiated SH-SY5Y human neuroblastoma cells. Brain Res 1352: 1-10. Narumiya, S., Sugimoto, Y. and Ushikubi, F. (1999). Prostanoid receptors: structures, properties, and functions. Physiol Rev 79(4): 1193-1226. Nguyen, L., Rigo, J. M., Rocher, V., Belachew, S., Malgrange, B., Rogister, B., Leprince, P. and Moonen, G. (2001). Neurotransmitters as early signals for central nervous system development. Cell Tissue Res 305(2): 187-202. Nguyen, T., Sherratt, P. J. and Pickett, C. B. (2003). Regulatory mechanisms controlling gene expression mediated by the antioxidant response element. Annu Rev Pharmacol Toxicol 43: 233-260. Nichols, D. E. and Nichols, C. D. (2008). Serotonin receptors. Chem Rev 108(5): 1614-1641. Nimmerjahn, A., Kirchhoff, F. and Helmchen, F. (2005). Resting microglial cells are highly 374

dynamic surveillants of brain parenchyma in vivo. Science 308(5726): 1314-1318. Nioi, P., McMahon, M., Itoh, K., Yamamoto, M. and Hayes, J. D. (2003). Identification of a novel Nrf2-regulated antioxidant response element (ARE) in the mouse NAD(P)H:quinone oxidoreductase 1 gene: reassessment of the ARE consensus sequence. Biochem J 374(Pt 2): 337-348. Nioi, P. and Nguyen, T. (2007). A mutation of Keap1 found in breast cancer impairs its ability to repress Nrf2 activity. Biochem Biophys Res Commun 362(4): 816-821. Nioi, P., Nguyen, T., Sherratt, P. J. and Pickett, C. B. (2005). The carboxy-terminal Neh3 domain of Nrf2 is required for transcriptional activation. Mol Cell Biol 25(24): 10895- 10906. O'Banion, M. K., Miller, J. C., Chang, J. W., Kaplan, M. D. and Coleman, P. D. (1996). Interleukin-1 beta induces prostaglandin G/H synthase-2 (cyclooxygenase-2) in primary murine astrocyte cultures. J Neurochem 66(6): 2532-2540. O'Callaghan, J. P. and Miller, D. B. (1994). Neurotoxicity profiles of substituted amphetamines in the C57BL/6J mouse. J Pharmacol Exp Ther 270(2): 741-751. O'Dell, S. J., Feinberg, L. M. and Marshall, J. F. (2011). A neurotoxic regimen of methamphetamine impairs novelty recognition as measured by a social odor-based task. Behav Brain Res 216(1): 396-401. O'Neil, M. L., Kuczenski, R., Segal, D. S., Cho, A. K., Lacan, G. and Melega, W. P. (2006). Escalating dose pretreatment induces pharmacodynamic and not pharmacokinetic tolerance to a subsequent high-dose methamphetamine binge. Synapse 60(6): 465-473. Ogasawara, A., Arakawa, T., Kaneda, T., Takuma, T., Sato, T., Kaneko, H., Kumegawa, M. and Hakeda, Y. (2001). Fluid shear stress-induced cyclooxygenase-2 expression is mediated by C/EBP beta, cAMP-response element-binding protein, and AP-1 in osteoblastic MC3T3-E1 cells. J Biol Chem 276(10): 7048-7054. Ohki, S., Ogino, N., Yamamoto, S. and Hayaishi, O. (1979). Prostaglandin hydroperoxidase, an integral part of prostaglandin endoperoxide synthetase from bovine vesicular gland microsomes. J Biol Chem 254(3): 829-836. Ohta, T., Iijima, K., Miyamoto, M., Nakahara, I., Tanaka, H., Ohtsuji, M., Suzuki, T., Kobayashi, A., Yokota, J., Sakiyama, T., Shibata, T., Yamamoto, M. and Hirohashi, S. (2008). Loss of Keap1 function activates Nrf2 and provides advantages for lung cancer cell growth. Cancer Res 68(5): 1303-1309. Ohtsubo, T., Kamada, S., Mikami, T., Murakami, H. and Tsujimoto, Y. (1999). Identification of NRF2, a member of the NF-E2 family of transcription factors, as a substrate for caspase- 375

3(-like) proteases. Cell Death Differ 6(9): 865-872. Oida, H., Hirata, M., Sugimoto, Y., Ushikubi, F., Ohishi, H., Mizuno, N., Ichikawa, A. and Narumiya, S. (1997). Expression of messenger RNA for the prostaglandin D receptor in the leptomeninges of the mouse brain. FEBS Lett 417(1): 53-56. Oida, H., Namba, T., Sugimoto, Y., Ushikubi, F., Ohishi, H., Ichikawa, A. and Narumiya, S. (1995). In situ hybridization studies of prostacyclin receptor mRNA expression in various mouse organs. Br J Pharmacol 116(7): 2828-2837. Oka, A., Belliveau, M. J., Rosenberg, P. A. and Volpe, J. J. (1993). Vulnerability of oligodendroglia to glutamate: pharmacology, mechanisms, and prevention. J Neurosci 13(4): 1441-1453. Okada, Y., Voznesensky, O., Herschman, H., Harrison, J. and Pilbeam, C. (2000). Identification of multiple cis-acting elements mediating the induction of prostaglandin G/H synthase-2 by phorbol ester in murine osteoblastic cells. J Cell Biochem 78(2): 197-209. Okahara, K., Sun, B. and Kambayashi, J. (1998). Upregulation of prostacyclin synthesis-related gene expression by shear stress in vascular endothelial cells. Arterioscler Thromb Vasc Biol 18(12): 1922-1926. Olney, J. W. and Sharpe, L. G. (1969). Brain lesions in an infant rhesus monkey treated with monsodium glutamate. Science 166(903): 386-388. Omar, R. A., Chyan, Y. J., Andorn, A. C., Poeggeler, B., Robakis, N. K. and Pappolla, M. A. (1999). Increased Expression but Reduced Activity of Antioxidant Enzymes in Alzheimer's Disease. J Alzheimers Dis 1(3): 139-145. Otto, J. C., DeWitt, D. L. and Smith, W. L. (1993). N-glycosylation of prostaglandin endoperoxide synthases-1 and -2 and their orientations in the endoplasmic reticulum. J Biol Chem 268(24): 18234-18242. Oyake, T., Itoh, K., Motohashi, H., Hayashi, N., Hoshino, H., Nishizawa, M., Yamamoto, M. and Igarashi, K. (1996). Bach proteins belong to a novel family of BTB-basic leucine zipper transcription factors that interact with MafK and regulate transcription through the NF-E2 site. Mol Cell Biol 16(11): 6083-6095. Ozawa, T. and Hanaki, A. (1978). Hydroxyl Radical produced by the Reaction of Superoxide Ion with Hydrogen Peroxide : Electron Spin Resonance Detection by Spin Trapping Chemical & pharmaceutical bulletin 26(8): 2572-2575. Pacchioni, A. M., Vallone, J., Melendez, R. I., Shih, A., Murphy, T. H. and Kalivas, P. W. (2007). Nrf2 gene deletion fails to alter psychostimulant-induced behavior or neurotoxicity. Brain Res 1127(1): 26-35. 376

Padmanabhan, B., Tong, K. I., Ohta, T., Nakamura, Y., Scharlock, M., Ohtsuji, M., Kang, M. I., Kobayashi, A., Yokoyama, S. and Yamamoto, M. (2006). Structural basis for defects of Keap1 activity provoked by its point mutations in lung cancer. Mol Cell 21(5): 689-700. Parameshwaran, K., Dhanasekaran, M. and Suppiramaniam, V. (2008). Amyloid beta peptides and glutamatergic synaptic dysregulation. Exp Neurol 210(1): 7-13. Pardridge, W. M., Triguero, D., Yang, J. and Cancilla, P. A. (1990). Comparison of in vitro and in vivo models of drug transcytosis through the blood-brain barrier. J Pharmacol Exp Ther 253(2): 884-891. Paris, I., Martinez-Alvarado, P., Cardenas, S., Perez-Pastene, C., Graumann, R., Fuentes, P., Olea-Azar, C., Caviedes, P. and Segura-Aguilar, J. (2005). Dopamine-dependent iron toxicity in cells derived from rat hypothalamus. Chem Res Toxicol 18(3): 415-419. Park, E. Y. and Rho, H. M. (2002). The transcriptional activation of the human copper/zinc superoxide dismutase gene by 2,3,7,8-tetrachlorodibenzo-p-dioxin through two different regulator sites, the antioxidant responsive element and xenobiotic responsive element. Mol Cell Biochem 240(1-2): 47-55. Park, J. Y., Pillinger, M. H. and Abramson, S. B. (2006). Prostaglandin E2 synthesis and secretion: the role of PGE2 synthases. Clin Immunol 119(3): 229-240. Parman, T., Chen, G. and Wells, P. G. (1998). Free radical intermediates of phenytoin and related teratogens. Prostaglandin H synthase-catalyzed bioactivation, electron paramagnetic resonance spectrometry, and photochemical product analysis. J Biol Chem 273(39): 25079-25088. Parman, T. and Wells, P. G. (2002). Embryonic prostaglandin H synthase-2 (PHS-2) expression and benzo[a]pyrene teratogenicity in PHS-2 knockout mice. FASEB J 16(9): 1001-1009. Parman, T., Wiley, M. J. and Wells, P. G. (1999). Free radical-mediated oxidative DNA damage in the mechanism of thalidomide teratogenicity. Nat Med 5(5): 582-585. Parrish-Aungst, S., Shipley, M. T., Erdelyi, F., Szabo, G. and Puche, A. C. (2007). Quantitative analysis of neuronal diversity in the mouse olfactory bulb. J Comp Neurol 501(6): 825- 836. Pasinetti, G. M. (1998). Cyclooxygenase and inflammation in Alzheimer's disease: experimental approaches and clinical interventions. J Neurosci Res 54(1): 1-6. Pasinetti, G. M. and Aisen, P. S. (1998). Cyclooxygenase-2 expression is increased in frontal cortex of Alzheimer's disease brain. Neuroscience 87(2): 319-324. Patenaude, A., Murthy, M. R. and Mirault, M. E. (2005). Emerging roles of thioredoxin cycle enzymes in the central nervous system. Cell Mol Life Sci 62(10): 1063-1080. 377

Pedersen, A. K. and FitzGerald, G. A. (1984). Dose-related kinetics of aspirin. Presystemic acetylation of platelet cyclooxygenase. N Engl J Med 311(19): 1206-1211. Pepicelli, O., Fedele, E., Berardi, M., Raiteri, M., Levi, G., Greco, A., Ajmone-Cat, M. A. and Minghetti, L. (2005). Cyclo-oxygenase-1 and -2 differently contribute to prostaglandin E2 synthesis and lipid peroxidation after in vivo activation of N-methyl-D-aspartate receptors in rat hippocampus. J Neurochem 93(6): 1561-1567. Perry, T. L., Yong, V. W., Bergeron, C., Hansen, S. and Jones, K. (1987). Amino acids, glutathione, and glutathione transferase activity in the brains of patients with Alzheimer's disease. Ann Neurol 21(4): 331-336. Pettipher, R. (2008). The roles of the prostaglandin D(2) receptors DP(1) and CRTH2 in promoting allergic responses. Br J Pharmacol 153 Suppl 1: S191-199. Picot, D., Loll, P. J. and Garavito, R. M. (1994). The X-ray crystal structure of the membrane protein prostaglandin H2 synthase-1. Nature 367(6460): 243-249. Pietri, S., Culcasi, M. and Cozzone, P. J. (1989). Real-time continuous-flow spin trapping of hydroxyl free radical in the ischemic and post-ischemic myocardium. Eur J Biochem 186(1-2): 163-173. Pietri, S., Liebgott, T., Frejaville, C., Tordo, P. and Culcasi, M. (1998). Nitrone spin traps and their pyrrolidine analogs in myocardial reperfusion injury: hemodynamic and ESR implications--evidence for a cardioprotective phosphonate effect for 5- (diethoxyphosphoryl)-5-methyl-1-pyrroline N-oxide in rat hearts. Eur J Biochem 254(2): 256-265. Pietsch, E. C., Chan, J. Y., Torti, F. M. and Torti, S. V. (2003). Nrf2 mediates the induction of ferritin H in response to xenobiotics and cancer chemopreventive dithiolethiones. J Biol Chem 278(4): 2361-2369. Pifl, C., Giros, B. and Caron, M. G. (1993). Dopamine transporter expression confers cytotoxicity to low doses of the parkinsonism-inducing neurotoxin 1-methyl-4- phenylpyridinium. J Neurosci 13(10): 4246-4253. Pileblad, E., Slivka, A., Bratvold, D. and Cohen, G. (1988). Studies on the autoxidation of dopamine: interaction with ascorbate. Arch Biochem Biophys 263(2): 447-452. Plessinger, M. A. (1998). Prenatal exposure to amphetamines. Risks and adverse outcomes in pregnancy. Obstet Gynecol Clin North Am 25(1): 119-138. Polymeropoulos, M. H., Lavedan, C., Leroy, E., Ide, S. E., Dehejia, A., Dutra, A., Pike, B., Root, H., Rubenstein, J., Boyer, R., Stenroos, E. S., Chandrasekharappa, S., Athanassiadou, A., Papapetropoulos, T., Johnson, W. G., Lazzarini, A. M., Duvoisin, R. 378

C., Di Iorio, G., Golbe, L. I. and Nussbaum, R. L. (1997). Mutation in the alpha- synuclein gene identified in families with Parkinson's disease. Science 276(5321): 2045- 2047. Pompl, P. N., Ho, L., Bianchi, M., McManus, T., Qin, W. and Pasinetti, G. M. (2003). A therapeutic role for cyclooxygenase-2 inhibitors in a transgenic mouse model of amyotrophic lateral sclerosis. FASEB J 17(6): 725-727. Powell, W. S. (2003). 15-Deoxy-delta12,14-PGJ2: endogenous PPARgamma ligand or minor eicosanoid degradation product? J Clin Invest 112(6): 828-830. Preston, T. J., Henderson, J. T., McCallum, G. P. and Wells, P. G. (2009). Base excision repair of reactive oxygen species-initiated 7,8-dihydro-8-oxo-2'-deoxyguanosine inhibits the cytotoxicity of platinum anticancer drugs. Mol Cancer Ther 8(7): 2015-2026. Pu, C., Broening, H. W. and Vorhees, C. V. (1996). Effect of methamphetamine on glutamate- positive neurons in the adult and developing rat somatosensory cortex. Synapse 23(4): 328-334. Pubill, D., Chipana, C., Camins, A., Pallas, M., Camarasa, J. and Escubedo, E. (2005). Free radical production induced by methamphetamine in rat striatal synaptosomes. Toxicol Appl Pharmacol 204(1): 57-68. Ramkissoon, A., Goncalves, L. L., Jeng, W. and Wells, P. G. (2005). Prostaglandin H synthase- 1 (PHS-1)-catalyzed bioactivation of dopamine and its precursor L-DOPA to free radical intermediates that initiate oxidative DNA damage. Proceedings of the Society of Toxicology of Canada, Montreal, Quebec, Abstract No 39 Ramkissoon, A. and Wells, P. G. (2010). Methamphetamine-initiated oxidative stress, neurotoxicity and functional deficits are modulated by nuclear factor-E2-related factor 2 (Nrf2). Submitted. Ramkissoon, A. and Wells, P. G. (2011a). Human prostaglandin H synthase (hPHS)-1- and hPHS-2-dependent bioactivation, oxidative macromolecular damage, and cytotoxicity of dopamine, its precursor, and its metabolites. Free Radic Biol Med 50(2): 295-304. Ramkissoon, A. and Wells, P. G. (2011b). Human prostaglandin H synthase (hPHS)-1 and hPHS-2 in amphetamine analog bioactivation, DNA oxidation and cytotoxicity. Toxicol Sci 120(1): 154-162. Ramos-Gomez, M., Kwak, M. K., Dolan, P. M., Itoh, K., Yamamoto, M., Talalay, P. and Kensler, T. W. (2001). Sensitivity to carcinogenesis is increased and chemoprotective efficacy of enzyme inducers is lost in nrf2 transcription factor-deficient mice. Proc Natl Acad Sci U S A 98(6): 3410-3415. 379

Ramsey, C. P., Glass, C. A., Montgomery, M. B., Lindl, K. A., Ritson, G. P., Chia, L. A., Hamilton, R. L., Chu, C. T. and Jordan-Sciutto, K. L. (2007). Expression of Nrf2 in neurodegenerative diseases. J Neuropathol Exp Neurol 66(1): 75-85. Rangasamy, T., Guo, J., Mitzner, W. A., Roman, J., Singh, A., Fryer, A. D., Yamamoto, M., Kensler, T. W., Tuder, R. M., Georas, S. N. and Biswal, S. (2005). Disruption of Nrf2 enhances susceptibility to severe airway inflammation and asthma in mice. J Exp Med 202(1): 47-59. Ravanat, J. L., Douki, T., Duez, P., Gremaud, E., Herbert, K., Hofer, T., Lasserre, L., Saint- Pierre, C., Favier, A. and Cadet, J. (2002). Cellular background level of 8-oxo-7,8- dihydro-2'-deoxyguanosine: an isotope based method to evaluate artefactual oxidation of DNA during its extraction and subsequent work-up. Carcinogenesis 23(11): 1911-1918. Reddy, S. T. and Herschman, H. R. (1994). Ligand-induced prostaglandin synthesis requires expression of the TIS10/PGS-2 prostaglandin synthase gene in murine fibroblasts and macrophages. J Biol Chem 269(22): 15473-15480. Reder, A. T., Thapar, M., Sapugay, A. M. and Jensen, M. A. (1994). Prostaglandins and inhibitors of arachidonate metabolism suppress experimental allergic encephalomyelitis. J Neuroimmunol 54(1-2): 117-127. Reed, G. A. and Marnett, L. J. (1982). Metabolism and activation of 7,8-dihydrobenzo[a]pyrene during prostaglandin biosynthesis. Intermediacy of a bay-region epoxide. J Biol Chem 257(19): 11368-11376. Reese, T. S. and Karnovsky, M. J. (1967). Fine structural localization of a blood-brain barrier to exogenous peroxidase. J Cell Biol 34(1): 207-217. Reiter, R. J. (1995). Oxidative processes and antioxidative defense mechanisms in the aging brain. FASEB J 9(7): 526-533. Reksidler, A. B., Lima, M. M., Zanata, S. M., Machado, H. B., da Cunha, C., Andreatini, R., Tufik, S. and Vital, M. A. (2007). The COX-2 inhibitor parecoxib produces neuroprotective effects in MPTP-lesioned rats. Eur J Pharmacol 560(2-3): 163-175. Rhee, S. G., Chae, H. Z. and Kim, K. (2005). Peroxiredoxins: a historical overview and speculative preview of novel mechanisms and emerging concepts in cell signaling. Free Radic Biol Med 38(12): 1543-1552. Ricaurte, G., Bryan, G., Strauss, L., Seiden, L. and Schuster, C. (1985). Hallucinogenic amphetamine selectively destroys brain serotonin nerve terminals. Science 229(4717): 986-988. Ricaurte, G. A., Guillery, R. W., Seiden, L. S., Schuster, C. R. and Moore, R. Y. (1982). 380

Dopamine nerve terminal degeneration produced by high doses of methylamphetamine in the rat brain. Brain Res 235(1): 93-103. Riederer, P. and Jellinger, K. (1983). Neurochemical insights into monoamine oxidase inhibitors, with special reference to deprenyl (selegiline). Acta Neurol Scand Suppl 95: 43-55. Rieke, C. J., Mulichak, A. M., Garavito, R. M. and Smith, W. L. (1999). The role of arginine 120 of human prostaglandin endoperoxide H synthase-2 in the interaction with fatty acid substrates and inhibitors. J Biol Chem 274(24): 17109-17114. Ristimaki, A., Narko, K. and Hla, T. (1996). Down-regulation of cytokine-induced cyclooxygenase-2 transcript isoforms by dexamethasone: evidence for post-transcriptional regulation. Biochem J 318 ( Pt 1): 325-331. Ritchie, K. and Lovestone, S. (2002). The dementias. Lancet 360(9347): 1759-1766. Riviere, G. J., Gentry, W. B. and Owens, S. M. (2000). Disposition of methamphetamine and its metabolite amphetamine in brain and other tissues in rats after intravenous administration. J Pharmacol Exp Ther 292(3): 1042-1047. Rocca, B., Spain, L. M., Pure, E., Langenbach, R., Patrono, C. and FitzGerald, G. A. (1999). Distinct roles of prostaglandin H synthases 1 and 2 in T-cell development. J Clin Invest 103(10): 1469-1477. Rogers, J., Kirby, L. C., Hempelman, S. R., Berry, D. L., McGeer, P. L., Kaszniak, A. W., Zalinski, J., Cofield, M., Mansukhani, L., Willson, P. and et al. (1993). Clinical trial of indomethacin in Alzheimer's disease. Neurology 43(8): 1609-1611. Rose, J. W., Hill, K. E., Watt, H. E. and Carlson, N. G. (2004). Inflammatory cell expression of cyclooxygenase-2 in the multiple sclerosis lesion. J Neuroimmunol 149(1-2): 40-49. Ross, G. W., Petrovitch, H., Abbott, R. D., Tanner, C. M., Popper, J., Masaki, K., Launer, L. and White, L. R. (2008). Association of olfactory dysfunction with risk for future Parkinson's disease. Ann Neurol 63(2): 167-173. Rothman, R. B. and Baumann, M. H. (2003). Monoamine transporters and psychostimulant drugs. Eur J Pharmacol 479(1-3): 23-40. Rothman, R. B., Baumann, M. H., Dersch, C. M., Romero, D. V., Rice, K. C., Carroll, F. I. and Partilla, J. S. (2001). Amphetamine-type central nervous system stimulants release norepinephrine more potently than they release dopamine and serotonin. Synapse 39(1): 32-41. Rouzer, C. A. and Marnett, L. J. (2003). Mechanism of free radical oxygenation of polyunsaturated fatty acids by cyclooxygenases. Chem Rev 103(6): 2239-2304. 381

Rushmore, T. H., King, R. G., Paulson, K. E. and Pickett, C. B. (1990). Regulation of glutathione S-transferase Ya subunit gene expression: identification of a unique xenobiotic-responsive element controlling inducible expression by planar aromatic compounds. Proc Natl Acad Sci U S A 87(10): 3826-3830. Rushmore, T. H. and Kong, A. N. (2002). Pharmacogenomics, regulation and signaling pathways of phase I and II drug metabolizing enzymes. Curr Drug Metab 3(5): 481-490. Rushmore, T. H., Morton, M. R. and Pickett, C. B. (1991). The antioxidant responsive element. Activation by oxidative stress and identification of the DNA consensus sequence required for functional activity. J Biol Chem 266(18): 11632-11639. Rybnikova, E., Damdimopoulos, A. E., Gustafsson, J. A., Spyrou, G. and Pelto-Huikko, M. (2000). Expression of novel antioxidant thioredoxin-2 in the rat brain. Eur J Neurosci 12(5): 1669-1678. Ryter, S. W., Alam, J. and Choi, A. M. (2006). Heme oxygenase-1/carbon monoxide: from basic science to therapeutic applications. Physiol Rev 86(2): 583-650. Sachse, C., Brockmoller, J., Bauer, S. and Roots, I. (1997). Cytochrome P450 2D6 variants in a Caucasian population: allele frequencies and phenotypic consequences. Am J Hum Genet 60(2): 284-295. Sah, V. P., Attardi, L. D., Mulligan, G. J., Williams, B. O., Bronson, R. T. and Jacks, T. (1995). A subset of p53-deficient embryos exhibit exencephaly. Nat Genet 10(2): 175-180. Saito, T. and Nishino, T. (1989). Differences in redox and kinetic properties between NAD- dependent and O2-dependent types of rat liver xanthine dehydrogenase. J Biol Chem 264(17): 10015-10022. Sakurai, A., Nishimoto, M., Himeno, S., Imura, N., Tsujimoto, M., Kunimoto, M. and Hara, S. (2005). Transcriptional regulation of thioredoxin reductase 1 expression by cadmium in vascular endothelial cells: role of NF-E2-related factor-2. J Cell Physiol 203(3): 529- 537. Samad, T. A., Moore, K. A., Sapirstein, A., Billet, S., Allchorne, A., Poole, S., Bonventre, J. V. and Woolf, C. J. (2001). Interleukin-1beta-mediated induction of Cox-2 in the CNS contributes to inflammatory pain hypersensitivity. Nature 410(6827): 471-475. Sandhya, T. L., Ong, W. Y., Horrocks, L. A. and Farooqui, A. A. (1998). A light and electron microscopic study of cytoplasmic phospholipase A2 and cyclooxygenase-2 in the hippocampus after kainate lesions. Brain Res 788(1-2): 223-231. SantaCruz, K. S., Yazlovitskaya, E., Collins, J., Johnson, J. and DeCarli, C. (2004). Regional NAD(P)H:quinone oxidoreductase activity in Alzheimer's disease. Neurobiol Aging 382

25(1): 63-69. Sapp, E., Kegel, K. B., Aronin, N., Hashikawa, T., Uchiyama, Y., Tohyama, K., Bhide, P. G., Vonsattel, J. P. and DiFiglia, M. (2001). Early and progressive accumulation of reactive microglia in the Huntington disease brain. J Neuropathol Exp Neurol 60(2): 161-172. Sarafian, T. A., Verity, M. A., Vinters, H. V., Shih, C. C., Shi, L., Ji, X. D., Dong, L. and Shau, H. (1999). Differential expression of peroxiredoxin subtypes in human brain cell types. J Neurosci Res 56(2): 206-212. Sarlette, A., Krampfl, K., Grothe, C., Neuhoff, N., Dengler, R. and Petri, S. (2008). Nuclear erythroid 2-related factor 2-antioxidative response element signaling pathway in motor cortex and spinal cord in amyotrophic lateral sclerosis. J Neuropathol Exp Neurol 67(11): 1055-1062. Sasame, H. A., Ames, M. M. and Nelson, S. D. (1977). Cytochrome P-450 and NADPH cytochrome c reductase in rat brain: formation of catechols and reactive catechol metabolites. Biochem Biophys Res Commun 78(3): 919-926. Satoh, T., Okamoto, S. I., Cui, J., Watanabe, Y., Furuta, K., Suzuki, M., Tohyama, K. and Lipton, S. A. (2006). Activation of the Keap1/Nrf2 pathway for neuroprotection by electrophilic [correction of electrophillic] phase II inducers. Proc Natl Acad Sci U S A 103(3): 768-773. Sawaoka, H., Dixon, D. A., Oates, J. A. and Boutaud, O. (2003). Tristetraprolin binds to the 3'- untranslated region of cyclooxygenase-2 mRNA. A polyadenylation variant in a cancer cell line lacks the binding site. J Biol Chem 278(16): 13928-13935. Schaloske, R. H. and Dennis, E. A. (2006). The phospholipase A2 superfamily and its group numbering system. Biochim Biophys Acta 1761(11): 1246-1259. Schapira, A. H., Cooper, J. M., Dexter, D., Clark, J. B., Jenner, P. and Marsden, C. D. (1990). Mitochondrial complex I deficiency in Parkinson's disease. J Neurochem 54(3): 823- 827. Schepers, R. J., Oyler, J. M., Joseph, R. E., Jr., Cone, E. J., Moolchan, E. T. and Huestis, M. A. (2003). Methamphetamine and amphetamine pharmacokinetics in oral fluid and plasma after controlled oral methamphetamine administration to human volunteers. Clin Chem 49(1): 121-132. Schipper, H. M. (2004). Heme oxygenase expression in human central nervous system disorders. Free Radic Biol Med 37(12): 1995-2011. Schipper, H. M., Song, W., Zukor, H., Hascalovici, J. R. and Zeligman, D. (2009). Heme oxygenase-1 and neurodegeneration: expanding frontiers of engagement. J Neurochem 383

110(2): 469-485. Schuster, V. L. (2002). Prostaglandin transport. Prostaglandins Other Lipid Mediat 68-69: 633- 647. Schwab, J. M., Beschorner, R., Meyermann, R., Gozalan, F. and Schluesener, H. J. (2002). Persistent accumulation of cyclooxygenase-1-expressing microglial cells and macrophages and transient upregulation by endothelium in human brain injury. J Neurosurg 96(5): 892-899. Scott, J. C., Woods, S. P., Matt, G. E., Meyer, R. A., Heaton, R. K., Atkinson, J. H. and Grant, I. (2007). Neurocognitive effects of methamphetamine: a critical review and meta-analysis. Neuropsychol Rev 17(3): 275-297. Sealy, R. C., Hyde, J. S., Felix, C. C., Menon, I. A., Prota, G., Swartz, H. M., Persad, S. and Haberman, H. F. (1982). Novel free radicals in synthetic and natural pheomelanins: distinction between dopa melanins and cysteinyldopa melanins by ESR spectroscopy. Proc Natl Acad Sci U S A 79(9): 2885-2889. Segura-Aguilar, J., Metodiewa, D. and Welch, C. J. (1998). Metabolic activation of dopamine o- quinones to o-semiquinones by NADPH cytochrome P450 reductase may play an important role in oxidative stress and apoptotic effects. Biochim Biophys Acta 1381(1): 1-6. Segura Aguilar, J. and Kostrzewa, R. M. (2004). Neurotoxins and neurotoxic species implicated in neurodegeneration. Neurotox Res 6(7-8): 615-630. Segura, M., Ortuno, J., Farre, M., McLure, J. A., Pujadas, M., Pizarro, N., Llebaria, A., Joglar, J., Roset, P. N., Segura, J. and de La Torre, R. (2001). 3,4-Dihydroxymethamphetamine (HHMA). A major in vivo 3,4-methylenedioxymethamphetamine (MDMA) metabolite in humans. Chem Res Toxicol 14(9): 1203-1208. Seiden, L. S., Commins, D. L., Vosmer, G., Axt, K. and Marek, G. (1988). Neurotoxicity in dopamine and 5-hydroxytryptamine terminal fields: a regional analysis in nigrostriatal and mesolimbic projections. Ann N Y Acad Sci 537: 161-172. Sekine, Y., Iyo, M., Ouchi, Y., Matsunaga, T., Tsukada, H., Okada, H., Yoshikawa, E., Futatsubashi, M., Takei, N. and Mori, N. (2001). Methamphetamine-related psychiatric symptoms and reduced brain dopamine transporters studied with PET. Am J Psychiatry 158(8): 1206-1214. Sekine, Y., Ouchi, Y., Takei, N., Yoshikawa, E., Nakamura, K., Futatsubashi, M., Okada, H., Minabe, Y., Suzuki, K., Iwata, Y., Tsuchiya, K. J., Tsukada, H., Iyo, M. and Mori, N. (2006). Brain serotonin transporter density and aggression in abstinent 384

methamphetamine abusers. Arch Gen Psychiatry 63(1): 90-100. Selinsky, B. S., Gupta, K., Sharkey, C. T. and Loll, P. J. (2001). Structural analysis of NSAID binding by prostaglandin H2 synthase: time-dependent and time-independent inhibitors elicit identical enzyme conformations. Biochemistry 40(17): 5172-5180. Serrano, F., Kolluri, N. S., Wientjes, F. B., Card, J. P. and Klann, E. (2003). NADPH oxidase immunoreactivity in the mouse brain. Brain Res 988(1-2): 193-198. Shaftel, S. S., Olschowka, J. A., Hurley, S. D., Moore, A. H. and O'Banion, M. K. (2003). COX- 3: a splice variant of cyclooxygenase-1 in mouse neural tissue and cells. Brain Res Mol Brain Res 119(2): 213-215. Shappell, S. A., Kearns, G. L., Valentine, J. L., Neri, D. F. and DeJohn, C. A. (1996). Chronopharmacokinetics and chronopharmacodynamics of dextromethamphetamine in man. J Clin Pharmacol 36(11): 1051-1063. Shen, G., Hebbar, V., Nair, S., Xu, C., Li, W., Lin, W., Keum, Y. S., Han, J., Gallo, M. A. and Kong, A. N. (2004). Regulation of Nrf2 transactivation domain activity. The differential effects of mitogen-activated protein kinase cascades and synergistic stimulatory effect of Raf and CREB-binding protein. J Biol Chem 279(22): 23052-23060. Shen, G., Xu, C., Hu, R., Jain, M. R., Gopalkrishnan, A., Nair, S., Huang, M. T., Chan, J. Y. and Kong, A. N. (2006). Modulation of nuclear factor E2-related factor 2-mediated gene expression in mice liver and small intestine by cancer chemopreventive agent curcumin. Mol Cancer Ther 5(1): 39-51. Shibata, T., Kokubu, A., Gotoh, M., Ojima, H., Ohta, T., Yamamoto, M. and Hirohashi, S. (2008). Genetic alteration of Keap1 confers constitutive Nrf2 activation and resistance to chemotherapy in gallbladder cancer. Gastroenterology 135(4): 1358-1368, 1368 e1351- 1354. Shih, A. Y., Imbeault, S., Barakauskas, V., Erb, H., Jiang, L., Li, P. and Murphy, T. H. (2005a). Induction of the Nrf2-driven antioxidant response confers neuroprotection during mitochondrial stress in vivo. J Biol Chem 280(24): 22925-22936. Shih, A. Y., Johnson, D. A., Wong, G., Kraft, A. D., Jiang, L., Erb, H., Johnson, J. A. and Murphy, T. H. (2003). Coordinate regulation of glutathione biosynthesis and release by Nrf2-expressing glia potently protects neurons from oxidative stress. J Neurosci 23(8): 3394-3406. Shih, A. Y., Li, P. and Murphy, T. H. (2005b). A small-molecule-inducible Nrf2-mediated antioxidant response provides effective prophylaxis against cerebral ischemia in vivo. J Neurosci 25(44): 10321-10335. 385

Shih, J. C. (1991). Molecular basis of human MAO A and B. Neuropsychopharmacology 4(1): 1-7. Shimohama, S., Tanino, H., Kawakami, N., Okamura, N., Kodama, H., Yamaguchi, T., Hayakawa, T., Nunomura, A., Chiba, S., Perry, G., Smith, M. A. and Fujimoto, S. (2000). Activation of NADPH oxidase in Alzheimer's disease brains. Biochem Biophys Res Commun 273(1): 5-9. Shimura-Miura, H., Hattori, N., Kang, D., Miyako, K., Nakabeppu, Y. and Mizuno, Y. (1999). Increased 8-oxo-dGTPase in the mitochondria of substantia nigral neurons in Parkinson's disease. Ann Neurol 46(6): 920-924. Shin, S., Wakabayashi, N., Misra, V., Biswal, S., Lee, G. H., Agoston, E. S., Yamamoto, M. and Kensler, T. W. (2007). NRF2 modulates aryl hydrocarbon receptor signaling: influence on adipogenesis. Mol Cell Biol 27(20): 7188-7197. Shinkai, Y., Sumi, D., Fukami, I., Ishii, T. and Kumagai, Y. (2006). Sulforaphane, an activator of Nrf2, suppresses cellular accumulation of arsenic and its cytotoxicity in primary mouse hepatocytes. FEBS Lett 580(7): 1771-1774. Shiue, C. Y., Shiue, G. G., Cornish, K. G. and O'Rourke, M. F. (1995). Comparative PET studies of the distribution of (-)-3,4-methylenedioxy-N-[11C]methamphetamine and (-)- [11C]methamphetamine in a monkey brain. Nucl Med Biol 22(3): 321-324. Siegel, D. and Ross, D. (2000). Immunodetection of NAD(P)H:quinone oxidoreductase 1 (NQO1) in human tissues. Free Radic Biol Med 29(3-4): 246-253. Simmons, D. L., Botting, R. M. and Hla, T. (2004). Cyclooxygenase isozymes: the biology of prostaglandin synthesis and inhibition. Pharmacol Rev 56(3): 387-437. Singh, A., Misra, V., Thimmulappa, R. K., Lee, H., Ames, S., Hoque, M. O., Herman, J. G., Baylin, S. B., Sidransky, D., Gabrielson, E., Brock, M. V. and Biswal, S. (2006). Dysfunctional KEAP1-NRF2 interaction in non-small-cell lung cancer. PLoS Med 3(10): e420. Siraki, A. G. and O'Brien, P. J. (2002). Prooxidant activity of free radicals derived from phenol- containing neurotransmitters. Toxicology 177(1): 81-90. Slamberova, R., Pometlova, M. and Charousova, P. (2006). Postnatal development of rat pups is altered by prenatal methamphetamine exposure. Prog Neuropsychopharmacol B ol Psychiatry 30(1): 82-88. Smith, B. J., Curtis, J. F. and Eling, T. E. (1991). Bioactivation of xenobiotics by prostaglandin H synthase. Chem Biol Interact 79(3): 245-264. Smith, L. M., LaGasse, L. L., Derauf, C., Grant, P., Shah, R., Arria, A., Huestis, M., Haning, 386

W., Strauss, A., Della Grotta, S., Liu, J. and Lester, B. M. (2006). The infant development, environment, and lifestyle study: effects of prenatal methamphetamine exposure, polydrug exposure, and poverty on intrauterine growth. Pediatrics 118(3): 1149-1156. Smith, M. A., Richey Harris, P. L., Sayre, L. M., Beckman, J. S. and Perry, G. (1997). Widespread peroxynitrite-mediated damage in Alzheimer's disease. J Neurosci 17(8): 2653-2657. Smith, W. L., DeWitt, D. L. and Garavito, R. M. (2000). Cyclooxygenases: structural, cellular, and molecular biology. Annu Rev Biochem 69: 145-182. Smith, W. L. and Song, I. (2002). The enzymology of prostaglandin endoperoxide H synthases- 1 and -2. Prostaglandins Other Lipid Mediat 68-69: 115-128. Sofuoglu, M. and Sewell, R. A. (2009). Norepinephrine and stimulant addiction. Addict Biol 14(2): 119-129. Solomon, S. D., McMurray, J. J., Pfeffer, M. A., Wittes, J., Fowler, R., Finn, P., Anderson, W. F., Zauber, A., Hawk, E. and Bertagnolli, M. (2005). Cardiovascular risk associated with celecoxib in a clinical trial for colorectal adenoma prevention. N Engl J Med 352(11): 1071-1080. Song, C. and Leonard, B. E. (2005). The olfactory bulbectomised rat as a model of depression. Neurosci Biobehav Rev 29(4-5): 627-647. Song, C., Zhang, Y., Parsons, C. G. and Liu, Y. F. (2003). Expression of polyglutamine- expanded huntingtin induces tyrosine phosphorylation of N-methyl-D-aspartate receptors. J Biol Chem 278(35): 33364-33369. Song, I. and Smith, W. L. (1996). C-terminal Ser/Pro-Thr-Glu-Leu tetrapeptides of prostaglandin endoperoxide H synthases-1 and -2 target the enzymes to the endoplasmic reticulum. Arch Biochem Biophys 334(1): 67-72. Sorce, S. and Krause, K. H. (2009). NOX enzymes in the central nervous system: from signaling to disease. Antioxid Redox Signal 11(10): 2481-2504. Spencer, A. G., Woods, J. W., Arakawa, T., Singer, II and Smith, W. L. (1998). Subcellular localization of prostaglandin endoperoxide H synthases-1 and -2 by immunoelectron microscopy. J Biol Chem 273(16): 9886-9893. Spencer, J. P., Whiteman, M., Jenner, P. and Halliwell, B. (2002). 5-s-Cysteinyl-conjugates of catecholamines induce cell damage, extensive DNA base modification and increases in caspase-3 activity in neurons. J Neurochem 81(1): 122-129. Sprague, J. E., Brutcher, R. E., Mills, E. M., Caden, D. and Rusyniak, D. E. (2004). Attenuation 387

of 3,4-methylenedioxymethamphetamine (MDMA, Ecstasy)-induced rhabdomyolysis with alpha1- plus beta3-adrenoreceptor antagonists. Br J Pharmacol 142(4): 667-670. Stephenson, D., Rash, K., Smalstig, B., Roberts, E., Johnstone, E., Sharp, J., Panetta, J., Little, S., Kramer, R. and Clemens, J. (1999). Cytosolic phospholipase A2 is induced in reactive glia following different forms of neurodegeneration. Glia 27(2): 110-128. Stewart, D., Killeen, E., Naquin, R., Alam, S. and Alam, J. (2003). Degradation of transcription factor Nrf2 via the ubiquitin-proteasome pathway and stabilization by cadmium. J Biol Chem 278(4): 2396-2402. Stewart, J. L. and Meeker, J. E. (1997). Fetal and infant deaths associated with maternal methamphetamine abuse. J Anal Toxicol 21(6): 515-517. Straus, D. S. and Glass, C. K. (2001). Cyclopentenone prostaglandins: new insights on biological activities and cellular targets. Med Res Rev 21(3): 185-210. Subbaramaiah, K., Lin, D. T., Hart, J. C. and Dannenberg, A. J. (2001). Peroxisome proliferator- activated receptor gamma ligands suppress the transcriptional activation of cyclooxygenase-2. Evidence for involvement of activator protein-1 and CREB-binding protein/p300. J Biol Chem 276(15): 12440-12448. Sugimoto, Y., Shigemoto, R., Namba, T., Negishi, M., Mizuno, N., Narumiya, S. and Ichikawa, A. (1994). Distribution of the messenger RNA for the prostaglandin E receptor subtype EP3 in the mouse nervous system. Neuroscience 62(3): 919-928. Sugioka, K., Nakano, M., Totsune-Nakano, H., Minakami, H., Tero-Kubota, S. and Ikegami, Y. (1988). Mechanism of O2- generation in reduction and oxidation cycle of ubiquinones in a model of mitochondrial electron transport systems. Biochim Biophys Acta 936(3): 377-385. Sulzer, D., Sonders, M. S., Poulsen, N. W. and Galli, A. (2005). Mechanisms of neurotransmitter release by amphetamines: a review. Prog Neurobiol 75(6): 406-433. Sun, F. F., McGuire, J. C., Morton, D. R., Pike, J. E., Sprecher, H. and Kunau, W. H. (1981). Inhibition of platelet arachidonic acid 12-lipoxygenase by acetylenic acid compounds. Prostaglandins 21(2): 333-343. Sun, Z., Chin, Y. E. and Zhang, D. D. (2009). Acetylation of Nrf2 by p300/CBP augments promoter-specific DNA binding of Nrf2 during the antioxidant response. Mol Cell Biol 29(10): 2658-2672. Sun, Z., Zhang, S., Chan, J. Y. and Zhang, D. D. (2007). Keap1 controls postinduction repression of the Nrf2-mediated antioxidant response by escorting nuclear export of Nrf2. Mol Cell Biol 27(18): 6334-6349. 388

Sung, S., Yang, H., Uryu, K., Lee, E. B., Zhao, L., Shineman, D., Trojanowski, J. Q., Lee, V. M. and Pratico, D. (2004). Modulation of nuclear factor-kappa B activity by indomethacin influences A beta levels but not A beta precursor protein metabolism in a model of Alzheimer's disease. Am J Pathol 165(6): 2197-2206. Suzuki, H., Tashiro, S., Sun, J., Doi, H., Satomi, S. and Igarashi, K. (2003). Cadmium induces nuclear export of Bach1, a transcriptional repressor of heme oxygenase-1 gene. J Biol Chem 278(49): 49246-49253. Swaab, D. F., Lucassen, P. J., Salehi, A., Scherder, E. J., van Someren, E. J. and Verwer, R. W. (1998). Reduced neuronal activity and reactivation in Alzheimer's disease. Prog Brain Res 117: 343-377. Tafazoli, S. and O'Brien, P. J. (2005). Peroxidases: a role in the metabolism and side effects of drugs. Drug Discov Today 10(9): 617-625. Takahashi, M., Dore, S., Ferris, C. D., Tomita, T., Sawa, A., Wolosker, H., Borchelt, D. R., Iwatsubo, T., Kim, S. H., Thinakaran, G., Sisodia, S. S. and Snyder, S. H. (2000). Amyloid precursor proteins inhibit heme oxygenase activity and augment neurotoxicity in Alzheimer's disease. Neuron 28(2): 461-473. Talamo, B. R., Rudel, R., Kosik, K. S., Lee, V. M., Neff, S., Adelman, L. and Kauer, J. S. (1989). Pathological changes in olfactory neurons in patients with Alzheimer's disease. Nature 337(6209): 736-739. Tanabe, T. and Tohnai, N. (2002). Cyclooxygenase isozymes and their gene structures and expression. Prostaglandins Other Lipid Mediat 68-69: 95-114. Tanaka, N., Ikeda, Y., Ohta, Y., Deguchi, K., Tian, F., Shang, J., Matsuura, T. and Abe, K. (2010). Expression of Keap1-Nrf2 system and antioxidative proteins in mouse brain after transient middle cerebral artery occlusion. Brain Res (in press). Taylor, T. N., Caudle, W. M., Shepherd, K. R., Noorian, A., Jackson, C. R., Iuvone, P. M., Weinshenker, D., Greene, J. G. and Miller, G. W. (2009). Nonmotor symptoms of Parkinson's disease revealed in an animal model with reduced monoamine storage capacity. J Neurosci 29(25): 8103-8113. Tazawa, R., Xu, X. M., Wu, K. K. and Wang, L. H. (1994). Characterization of the genomic structure, chromosomal location and promoter of human prostaglandin H synthase-2 gene. Biochem Biophys Res Commun 203(1): 190-199. Teismann, P. and Ferger, B. (2001). Inhibition of the cyclooxygenase isoenzymes COX-1 and COX-2 provide neuroprotection in the MPTP-mouse model of Parkinson's disease. Synapse 39(2): 167-174. 389

Teismann, P., Tieu, K., Choi, D. K., Wu, D. C., Naini, A., Hunot, S., Vila, M., Jackson-Lewis, V. and Przedborski, S. (2003). Cyclooxygenase-2 is instrumental in Parkinson's disease neurodegeneration. Proc Natl Acad Sci U S A 100(9): 5473-5478. Terland, O., Almas, B., Flatmark, T., Andersson, K. K. and Sorlie, M. (2006). One-electron oxidation of catecholamines generates free radicals with an in vitro toxicity correlating with their lifetime. Free Radic Biol Med 41(8): 1266-1271. Thal, L. J., Ferris, S. H., Kirby, L., Block, G. A., Lines, C. R., Yuen, E., Assaid, C., Nessly, M. L., Norman, B. A., Baranak, C. C. and Reines, S. A. (2005). A randomized, double- blind, study of rofecoxib in patients with mild cognitive impairment. Neuropsychopharmacology 30(6): 1204-1215. Thiels, E., Urban, N. N., Gonzalez-Burgos, G. R., Kanterewicz, B. I., Barrionuevo, G., Chu, C. T., Oury, T. D. and Klann, E. (2000). Impairment of long-term potentiation and associative memory in mice that overexpress extracellular superoxide dismutase. J Neurosci 20(20): 7631-7639. Thimmulappa, R. K., Mai, K. H., Srisuma, S., Kensler, T. W., Yamamoto, M. and Biswal, S. (2002). Identification of Nrf2-regulated genes induced by the chemopreventive agent sulforaphane by oligonucleotide microarray. Cancer Res 62(18): 5196-5203. Thomas, D. M. and Kuhn, D. M. (2005a). Cyclooxygenase-2 is an obligatory factor in methamphetamine-induced neurotoxicity. J Pharmacol Exp Ther 313(2): 870-876. Thomas, D. M. and Kuhn, D. M. (2005b). MK-801 and dextromethorphan block microglial activation and protect against methamphetamine-induced neurotoxicity. Brain Res 1050(1-2): 190-198. Thompson, P. M., Hayashi, K. M., Simon, S. L., Geaga, J. A., Hong, M. S., Sui, Y., Lee, J. Y., Toga, A. W., Ling, W. and London, E. D. (2004). Structural abnormalities in the brains of human subjects who use methamphetamine. J Neurosci 24(26): 6028-6036. Tillerson, J. L., Caudle, W. M., Parent, J. M., Gong, C., Schallert, T. and Miller, G. W. (2006). Olfactory discrimination deficits in mice lacking the dopamine transporter or the D2 dopamine receptor. Behav Brain Res 172(1): 97-105. Tomimoto, H., Akiguchi, I., Wakita, H., Lin, J. X. and Budka, H. (2000). Cyclooxygenase-2 is induced in microglia during chronic cerebral ischemia in humans. Acta Neuropathol 99(1): 26-30. Tong, K. I., Katoh, Y., Kusunoki, H., Itoh, K., Tanaka, T. and Yamamoto, M. (2006). Keap1 recruits Neh2 through binding to ETGE and DLG motifs: characterization of the two-site molecular recognition model. Mol Cell Biol 26(8): 2887-2900. 390

Towell, J. and Kalyanaraman, B. (1991). Detection of radical adducts of 5,5-dimethyl-1- pyrroline n-oxide by the combined use of high-performance liquid chromatography with electrochemical detection and electron spin resonance. Anal Biochem 196(1): 111-119. Toyama, T., Sumi, D., Shinkai, Y., Yasutake, A., Taguchi, K., Tong, K. I., Yamamoto, M. and Kumagai, Y. (2007). Cytoprotective role of Nrf2/Keap1 system in methylmercury toxicity. Biochem Biophys Res Commun 363(3): 645-650. Tsai, A., Kulmacz, R. J. and Palmer, G. (1995). Spectroscopic evidence for reaction of prostaglandin H synthase-1 tyrosyl radical with arachidonic acid. J Biol Chem 270(18): 10503-10508. Tsai, A. L. and Kulmacz, R. J. (2010). Prostaglandin H synthase: resolved and unresolved mechanistic issues. Arch Biochem Biophys 493(1): 103-124. Tse, D. C., McCreery, R. L. and Adams, R. N. (1976). Potential oxidative pathways of brain catecholamines. J Med Chem 19(1): 37-40. Tsuji, Y. (2005). JunD activates transcription of the human ferritin H gene through an antioxidant response element during oxidative stress. Oncogene 24(51): 7567-7578. Turrens, J. F. and Boveris, A. (1980). Generation of superoxide anion by the NADH dehydrogenase of bovine heart mitochondria. Biochem J 191(2): 421-427. Tyurina, Y. Y., Kapralov, A. A., Jiang, J., Borisenko, G. G., Potapovich, A. I., Sorokin, A., Kochanek, P. M., Graham, S. H., Schor, N. F. and Kagan, V. E. (2006). Oxidation and cytotoxicity of 6-OHDA are mediated by reactive intermediates of COX-2 overexpressed in PC12 cells. Brain Res 1093(1): 71-82. Ueda, N., Yamashita, R., Yamamoto, S. and Ishimura, K. (1997). Induction of cyclooxygenase- 1 in a human megakaryoblastic cell line (CMK) differentiated by phorbol ester. Biochim Biophys Acta 1344(1): 103-110. Ujike, H., Harano, M., Inada, T., Yamada, M., Komiyama, T., Sekine, Y., Sora, I., Iyo, M., Katsu, T., Nomura, A., Nakata, K. and Ozaki, N. (2003). Nine- or fewer repeat alleles in VNTR polymorphism of the dopamine transporter gene is a strong risk factor for prolonged methamphetamine psychosis. Pharmacogenomics J 3(4): 242-247. Ursini, F., Maiorino, M., Brigelius-Flohe, R., Aumann, K. D., Roveri, A., Schomburg, D. and Flohe, L. (1995). Diversity of glutathione peroxidases. Methods Enzymol 252: 38-53. van Horssen, J., Drexhage, J. A., Flor, T., Gerritsen, W., van der Valk, P. and de Vries, H. E. (2010). Nrf2 and DJ1 are consistently upregulated in inflammatory multiple sclerosis lesions. Free Radic Biol Med 49(8): 1283-1289. van Horssen, J., Schreibelt, G., Bo, L., Montagne, L., Drukarch, B., van Muiswinkel, F. L. and 391

de Vries, H. E. (2006). NAD(P)H:quinone oxidoreductase 1 expression in multiple sclerosis lesions. Free Radic Biol Med 41(2): 311-317. van Horssen, J., Schreibelt, G., Drexhage, J., Hazes, T., Dijkstra, C. D., van der Valk, P. and de Vries, H. E. (2008). Severe oxidative damage in multiple sclerosis lesions coincides with enhanced antioxidant enzyme expression. Free Radic Biol Med 45(12): 1729-1737. van Muiswinkel, F. L., de Vos, R. A., Bol, J. G., Andringa, G., Jansen Steur, E. N., Ross, D., Siegel, D. and Drukarch, B. (2004). Expression of NAD(P)H:quinone oxidoreductase in the normal and Parkinsonian substantia nigra. Neurobiol Aging 25(9): 1253-1262. Vane, J. R. (1971). Inhibition of prostaglandin synthesis as a mechanism of action for aspirin- like drugs. Nat New Biol 231(25): 232-235. Vargas, M. R., Johnson, D. A., Sirkis, D. W., Messing, A. and Johnson, J. A. (2008). Nrf2 activation in astrocytes protects against neurodegeneration in mouse models of familial amyotrophic lateral sclerosis. J Neurosci 28(50): 13574-13581. Venugopal, R. and Jaiswal, A. K. (1996). Nrf1 and Nrf2 positively and c-Fos and Fra1 negatively regulate the human antioxidant response element-mediated expression of NAD(P)H:quinone oxidoreductase1 gene. Proc Natl Acad Sci U S A 93(25): 14960- 14965. Venugopal, R. and Jaiswal, A. K. (1998). Nrf2 and Nrf1 in association with Jun proteins regulate antioxidant response element-mediated expression and coordinated induction of genes encoding detoxifying enzymes. Oncogene 17(24): 3145-3156. Vladimirova, O., O'Connor, J., Cahill, A., Alder, H., Butunoi, C. and Kalman, B. (1998). Oxidative damage to DNA in plaques of MS brains. Mult Scler 4(5): 413-418. Vogel, C. (2000). Prostaglandin H synthases and their importance in chemical toxicity. Curr Drug Metab 1(4): 391-404. Volicer, L. and Crino, P. B. (1990). Involvement of free radicals in dementia of the Alzheimer type: a hypothesis. Neurobiol Aging 11(5): 567-571. Volkow, N. D., Chang, L., Wang, G. J., Fowler, J. S., Ding, Y. S., Sedler, M., Logan, J., Franceschi, D., Gatley, J., Hitzemann, R., Gifford, A., Wong, C. and Pappas, N. (2001a). Low level of brain dopamine D2 receptors in methamphetamine abusers: association with metabolism in the orbitofrontal cortex. Am J Psychiatry 158(12): 2015-2021. Volkow, N. D., Chang, L., Wang, G. J., Fowler, J. S., Franceschi, D., Sedler, M., Gatley, S. J., Miller, E., Hitzemann, R., Ding, Y. S. and Logan, J. (2001b). Loss of dopamine transporters in methamphetamine abusers recovers with protracted abstinence. J Neurosci 21(23): 9414-9418. 392

Volkow, N. D., Chang, L., Wang, G. J., Fowler, J. S., Franceschi, D., Sedler, M. J., Gatley, S. J., Hitzemann, R., Ding, Y. S., Wong, C. and Logan, J. (2001c). Higher cortical and lower subcortical metabolism in detoxified methamphetamine abusers. Am J Psychiatry 158(3): 383-389. Volkow, N. D., Chang, L., Wang, G. J., Fowler, J. S., Leonido-Yee, M., Franceschi, D., Sedler, M. J., Gatley, S. J., Hitzemann, R., Ding, Y. S., Logan, J., Wong, C. and Miller, E. N. (2001d). Association of dopamine transporter reduction with psychomotor impairment in methamphetamine abusers. Am J Psychiatry 158(3): 377-382. Vollrath, V., Wielandt, A. M., Iruretagoyena, M. and Chianale, J. (2006). Role of Nrf2 in the regulation of the Mrp2 (ABCC2) gene. Biochem J 395(3): 599-609. Wagener, F. A., Volk, H. D., Willis, D., Abraham, N. G., Soares, M. P., Adema, G. J. and Figdor, C. G. (2003). Different faces of the heme-heme oxygenase system in inflammation. Pharmacol Rev 55(3): 551-571. Wagner, G. C., Ricaurte, G. A., Seiden, L. S., Schuster, C. R., Miller, R. J. and Westley, J. (1980). Long-lasting depletions of striatal dopamine and loss of dopamine uptake sites following repeated administration of methamphetamine. Brain Res 181(1): 151-160. Wakabayashi, N., Dinkova-Kostova, A. T., Holtzclaw, W. D., Kang, M. I., Kobayashi, A., Yamamoto, M., Kensler, T. W. and Talalay, P. (2004). Protection against electrophile and oxidant stress by induction of the phase 2 response: fate of cysteines of the Keap1 sensor modified by inducers. Proc Natl Acad Sci U S A 101(7): 2040-2045. Wakabayashi, N., Itoh, K., Wakabayashi, J., Motohashi, H., Noda, S., Takahashi, S., Imakado, S., Kotsuji, T., Otsuka, F., Roop, D. R., Harada, T., Engel, J. D. and Yamamoto, M. (2003). Keap1-null mutation leads to postnatal lethality due to constitutive Nrf2 activation. Nat Genet 35(3): 238-245. Wang, L. H., Hajibeigi, A., Xu, X. M., Loose-Mitchell, D. and Wu, K. K. (1993). Characterization of the promoter of human prostaglandin H synthase-1 gene. Biochem Biophys Res Commun 190(2): 406-411. Wang, R., An, J., Ji, F., Jiao, H., Sun, H. and Zhou, D. (2008a). Hypermethylation of the Keap1 gene in human lung cancer cell lines and lung cancer tissues. Biochem Biophys Res Commun 373(1): 151-154. Wang, W. and Chan, J. Y. (2006). Nrf1 is targeted to the endoplasmic reticulum membrane by an N-terminal transmembrane domain. Inhibition of nuclear translocation and transacting function. J Biol Chem 281(28): 19676-19687. Wang, X., Tomso, D. J., Chorley, B. N., Cho, H. Y., Cheung, V. G., Kleeberger, S. R. and Bell, 393

D. A. (2007a). Identification of polymorphic antioxidant response elements in the human genome. Hum Mol Genet 16(10): 1188-1200. Wang, X. J., Hayes, J. D., Henderson, C. J. and Wolf, C. R. (2007b). Identification of retinoic acid as an inhibitor of transcription factor Nrf2 through activation of retinoic acid receptor alpha. Proc Natl Acad Sci U S A 104(49): 19589-19594. Wang, X. J., Sun, Z., Villeneuve, N. F., Zhang, S., Zhao, F., Li, Y., Chen, W., Yi, X., Zheng, W., Wondrak, G. T., Wong, P. K. and Zhang, D. D. (2008b). Nrf2 enhances resistance of cancer cells to chemotherapeutic drugs, the dark side of Nrf2. Carcinogenesis 29(6): 1235-1243. Warner, M., Hellmold, H., Yoshida, S., Liao, D., Hedlund, E. and Gustafsson, J. A. (1997). Cytochrome P450 in the breast and brain: role in tissue-specific activation of xenobiotics. Mutat Res 376(1-2): 79-85. Warner, T. D. and Mitchell, J. A. (2004). Cyclooxygenases: new forms, new inhibitors, and lessons from the clinic. FASEB J 18(7): 790-804. Warnock, L. J. and Hunninghake, G. W. (1995). Multiple second messenger pathways regulate IL-1 beta-induced expression of PGHS-2 mRNA in normal human skin fibroblasts. J Cell Physiol 163(1): 172-178. Wasserman, W. W. and Fahl, W. E. (1997). Functional antioxidant responsive elements. Proc Natl Acad Sci U S A 94(10): 5361-5366. Watabe, A., Sugimoto, Y., Honda, A., Irie, A., Namba, T., Negishi, M., Ito, S., Narumiya, S. and Ichikawa, A. (1993). Cloning and expression of cDNA for a mouse EP1 subtype of prostaglandin E receptor. J Biol Chem 268(27): 20175-20178. Wells, P. G., McCallum, G. P., Chen, C. S., Henderson, J. T., Lee, C. J., Perstin, J., Preston, T. J., Wiley, M. J. and Wong, A. W. (2009). Oxidative stress in developmental origins of disease: teratogenesis, neurodevelopmental deficits, and cancer. Toxicol Sci 108(1): 4- 18. Wells, P. G., McCallum, G. P., Lam, K. C., Henderson, J. T. and Ondovcik, S. L. (2010). Oxidative DNA damage and repair in teratogenesis and neurodevelopmental deficits. Birth Defects Res C Embryo Today 90(2): 103-109. Wells, P. G., Ramkissoon, A. and Jeng, W. (2004). Neuroprotection against endogenous oxidative stress in aging prostaglandin H synthase-1 (PHS-1) knockout mice. Toxicol Sci (Supplement: The Toxicologist) 78(S-1): 63 (No. 305). Wells, P. G. and Vo, H. P. (1989). Effects of the tumor promoter 12-O-tetradecanoylphorbol-13- acetate on phenytoin-induced embryopathy in mice. Toxicol Appl Pharmacol 97(3): 398- 394

405. Wells, P. G. and Winn, L. M. (1996). Biochemical toxicology of chemical teratogenesis. Crit Rev Biochem Mol Biol 31(1): 1-40. Westlund, K. N., Denney, R. M., Kochersperger, L. M., Rose, R. M. and Abell, C. W. (1985). Distinct monoamine oxidase A and B populations in primate brain. Science 230(4722): 181-183. Wider, C. and Wszolek, Z. K. (2007). Clinical genetics of Parkinson's disease and related disorders. Parkinsonism Relat Disord 13 Suppl 3: S229-232. Wiese, F. W., Thompson, P. A. and Kadlubar, F. F. (2001). Carcinogen substrate specificity of human COX-1 and COX-2. Carcinogenesis 22(1): 5-10. Wijetunga, M., Seto, T., Lindsay, J. and Schatz, I. (2003). Crystal methamphetamine-associated cardiomyopathy: tip of the iceberg? J Toxicol Clin Toxicol 41(7): 981-986. Wild, A. C., Moinova, H. R. and Mulcahy, R. T. (1999). Regulation of gamma-glutamylcysteine synthetase subunit gene expression by the transcription factor Nrf2. J Biol Chem 274(47): 33627-33636. Wilson, J. M., Kalasinsky, K. S., Levey, A. I., Bergeron, C., Reiber, G., Anthony, R. M., Schmunk, G. A., Shannak, K., Haycock, J. W. and Kish, S. J. (1996). Striatal dopamine nerve terminal markers in human, chronic methamphetamine users. Nat Med 2(6): 699- 703. Wilson, L. A., Gemin, A., Espiritu, R. and Singh, G. (2005). ets-1 is transcriptionally upregulated by H2O2 via an antioxidant response element. FASEB J 19(14): 2085-2087. Winn, L. M. and Wells, P. G. (1995). Phenytoin-initiated DNA oxidation in murine embryo culture, and embryo protection by the antioxidative enzymes superoxide dismutase and catalase: evidence for reactive oxygen species-mediated DNA oxidation in the molecular mechanism of phenytoin teratogenicity. Mol Pharmacol 48(1): 112-120. Winn, L. M. and Wells, P. G. (1997). Evidence for embryonic prostaglandin H synthase- catalyzed bioactivation and reactive oxygen species-mediated oxidation of cellular macromolecules in phenytoin and benzo[a]pyrene teratogenesis. Free Radic Biol Med 22(4): 607-621. Winn, L. M. and Wells, P. G. (1999). Maternal administration of superoxide dismutase and catalase in phenytoin teratogenicity. Free Radic Biol Med 26(3-4): 266-274. Wise, R. A. (2004). Dopamine, learning and motivation. Nat Rev Neurosci 5(6): 483-494. Won, L., Bubula, N., McCoy, H. and Heller, A. (2001). Methamphetamine concentrations in fetal and maternal brain following prenatal exposure. Neurotoxicol Teratol 23(4): 349- 395

354. Wong, A. W., McCallum, G. P., Jeng, W. and Wells, P. G. (2008). Oxoguanine glycosylase 1 protects against methamphetamine-enhanced fetal brain oxidative DNA damage and neurodevelopmental deficits. J Neurosci 28(36): 9047-9054. Wu, D., Otton, S. V., Inaba, T., Kalow, W. and Sellers, E. M. (1997). Interactions of amphetamine analogs with human liver CYP2D6. Biochem Pharmacol 53(11): 1605- 1612. Wu, D. C., Re, D. B., Nagai, M., Ischiropoulos, H. and Przedborski, S. (2006). The inflammatory NADPH oxidase enzyme modulates motor neuron degeneration in amyotrophic lateral sclerosis mice. Proc Natl Acad Sci U S A 103(32): 12132-12137. Wu, G., Mannam, A. P., Wu, J., Kirbis, S., Shie, J. L., Chen, C., Laham, R. J., Sellke, F. W. and Li, J. (2003). Hypoxia induces myocyte-dependent COX-2 regulation in endothelial cells: role of VEGF. Am J Physiol Heart Circ Physiol 285(6): H2420-2429. Wu, K. K. and Liou, J. Y. (2005). Cellular and molecular biology of prostacyclin synthase. Biochem Biophys Res Commun 338(1): 45-52. Xia, Y. and Zweier, J. L. (1997). Superoxide and peroxynitrite generation from inducible nitric oxide synthase in macrophages. Proc Natl Acad Sci U S A 94(13): 6954-6958. Xiang, Z., Ho, L., Valdellon, J., Borchelt, D., Kelley, K., Spielman, L., Aisen, P. S. and Pasinetti, G. M. (2002). Cyclooxygenase (COX)-2 and cell cycle activity in a transgenic mouse model of Alzheimer's disease neuropathology. Neurobiol Aging 23(3): 327-334. Xiao, G., Chen, W. and Kulmacz, R. J. (1998). Comparison of structural stabilities of prostaglandin H synthase-1 and -2. J Biol Chem 273(12): 6801-6811. Xie, W., Fletcher, B. S., Andersen, R. D. and Herschman, H. R. (1994). v-src induction of the TIS10/PGS2 prostaglandin synthase gene is mediated by an ATF/CRE transcription response element. Mol Cell Biol 14(10): 6531-6539. Xie, W. L., Chipman, J. G., Robertson, D. L., Erikson, R. L. and Simmons, D. L. (1991). Expression of a mitogen-responsive gene encoding prostaglandin synthase is regulated by mRNA splicing. Proc Natl Acad Sci U S A 88(7): 2692-2696. Xu, W., Zhu, J. P. and Angulo, J. A. (2005). Induction of striatal pre- and postsynaptic damage by methamphetamine requires the dopamine receptors. Synapse 58(2): 110-121. Xu, X. M., Tang, J. L., Chen, X., Wang, L. H. and Wu, K. K. (1997). Involvement of two Sp1 elements in basal endothelial prostaglandin H synthase-1 promoter activity. J Biol Chem 272(11): 6943-6950. Yamagata, K., Andreasson, K. I., Kaufmann, W. E., Barnes, C. A. and Worley, P. F. (1993). 396

Expression of a mitogen-inducible cyclooxygenase in brain neurons: regulation by synaptic activity and glucocorticoids. Neuron 11(2): 371-386. Yamamoto, T., Yoh, K., Kobayashi, A., Ishii, Y., Kure, S., Koyama, A., Sakamoto, T., Sekizawa, K., Motohashi, H. and Yamamoto, M. (2004). Identification of polymorphisms in the promoter region of the human NRF2 gene. Biochem Biophys Res Commun 321(1): 72-79. Yang, H. and Chen, C. (2008). Cyclooxygenase-2 in synaptic signaling. Curr Pharm Des 14(14): 1443-1451. Yasojima, K., Schwab, C., McGeer, E. G. and McGeer, P. L. (1999a). Distribution of cyclooxygenase-1 and cyclooxygenase-2 mRNAs and proteins in human brain and peripheral organs. Brain Res 830(2): 226-236. Yasojima, K., Schwab, C., McGeer, E. G. and McGeer, P. L. (1999b). Up-regulated production and activation of the complement system in Alzheimer's disease brain. Am J Pathol 154(3): 927-936. Yasojima, K., Tourtellotte, W. W., McGeer, E. G. and McGeer, P. L. (2001). Marked increase in cyclooxygenase-2 in ALS spinal cord: implications for therapy. Neurology 57(6): 952-956. Yermakova, A. V. and O'Banion, M. K. (2001). Downregulation of neuronal cyclooxygenase-2 expression in end stage Alzheimer's disease. Neurobiol Aging 22(6): 823-836. Yermakova, A. V., Rollins, J., Callahan, L. M., Rogers, J. and O'Banion, M. K. (1999). Cyclooxygenase-1 in human Alzheimer and control brain: quantitative analysis of expression by microglia and CA3 hippocampal neurons. J Neuropathol Exp Neurol 58(11): 1135-1146. Yoh, K., Itoh, K., Enomoto, A., Hirayama, A., Yamaguchi, N., Kobayashi, M., Morito, N., Koyama, A., Yamamoto, M. and Takahashi, S. (2001). Nrf2-deficient female mice develop lupus-like autoimmune nephritis. Kidney Int 60(4): 1343-1353. Yokoyama, C., Takai, T. and Tanabe, T. (1988). Primary structure of sheep prostaglandin endoperoxide synthase deduced from cDNA sequence. FEBS Lett 231(2): 347-351. Yokoyama, C. and Tanabe, T. (1989). Cloning of human gene encoding prostaglandin endoperoxide synthase and primary structure of the enzyme. Biochem Biophys Res Commun 165(2): 888-894. Yoshioka, K., Deng, T., Cavigelli, M. and Karin, M. (1995). Antitumor promotion by phenolic antioxidants: inhibition of AP-1 activity through induction of Fra expression. Proc Natl Acad Sci U S A 92(11): 4972-4976. 397

Youdim, M. B. and Riederer, P. (1993). Dopamine metabolism and neurotransmission in primate brain in relationship to monoamine oxidase A and B inhibition. J Neural Transm Gen Sect 91(2-3): 181-195. Yu, J., Wang, J., Cadet, J. L. and Angulo, J. A. (2004). Histological evidence supporting a role for the striatal neurokinin-1 receptor in methamphetamine-induced neurotoxicity in the mouse brain. Brain Res 1007(1-2): 124-131. Yu, R., Chen, C., Mo, Y. Y., Hebbar, V., Owuor, E. D., Tan, T. H. and Kong, A. N. (2000a). Activation of mitogen-activated protein kinase pathways induces antioxidant response element-mediated gene expression via a Nrf2-dependent mechanism. J Biol Chem 275(51): 39907-39913. Yu, R., Lei, W., Mandlekar, S., Weber, M. J., Der, C. J., Wu, J. and Kong, A. N. (1999). Role of a mitogen-activated protein kinase pathway in the induction of phase II detoxifying enzymes by chemicals. J Biol Chem 274(39): 27545-27552. Yu, R., Mandlekar, S., Lei, W., Fahl, W. E., Tan, T. H. and Kong, A. N. (2000b). p38 mitogen- activated protein kinase negatively regulates the induction of phase II drug-metabolizing enzymes that detoxify carcinogens. J Biol Chem 275(4): 2322-2327. Yuan, J., Darvas, M., Sotak, B., Hatzidimitriou, G., McCann, U. D., Palmiter, R. D. and Ricaurte, G. A. (2010). Dopamine is not essential for the development of methamphetamine-induced neurotoxicity. J Neurochem 114(4): 1135-1142. Yueh, M. F. and Tukey, R. H. (2007). Nrf2-Keap1 signaling pathway regulates human UGT1A1 expression in vitro and in transgenic UGT1 mice. J Biol Chem 282(12): 8749-8758. Zeevalk, G. D., Razmpour, R. and Bernard, L. P. (2008). Glutathione and Parkinson's disease: is this the elephant in the room? Biomed Pharmacother 62(4): 236-249. Zhang, D. D. and Hannink, M. (2003). Distinct cysteine residues in Keap1 are required for Keap1-dependent ubiquitination of Nrf2 and for stabilization of Nrf2 by chemopreventive agents and oxidative stress. Mol Cell Biol 23(22): 8137-8151. Zhang, D. D., Lo, S. C., Cross, J. V., Templeton, D. J. and Hannink, M. (2004). Keap1 is a redox-regulated substrate adaptor protein for a Cul3-dependent ubiquitin ligase complex. Mol Cell Biol 24(24): 10941-10953. Zhang, J., Graham, D. G., Montine, T. J. and Ho, Y. S. (2000). Enhanced N-methyl-4-phenyl- 1,2,3,6-tetrahydropyridine toxicity in mice deficient in CuZn-superoxide dismutase or glutathione peroxidase. J Neuropathol Exp Neurol 59(1): 53-61. Zhang, J., Ohta, T., Maruyama, A., Hosoya, T., Nishikawa, K., Maher, J. M., Shibahara, S., Itoh, K. and Yamamoto, M. (2006a). BRG1 interacts with Nrf2 to selectively mediate 398

HO-1 induction in response to oxidative stress. Mol Cell Biol 26(21): 7942-7952. Zhang, Y., Crouch, D. H., Yamamoto, M. and Hayes, J. D. (2006b). Negative regulation of the Nrf1 transcription factor by its N-terminal domain is independent of Keap1: Nrf1, but not Nrf2, is targeted to the endoplasmic reticulum. Biochem J 399(3): 373-385. Zhang, Y., Kobayashi, A., Yamamoto, M. and Hayes, J. D. (2009). The Nrf3 transcription factor is a membrane-bound glycoprotein targeted to the endoplasmic reticulum through its N- terminal homology box 1 sequence. J Biol Chem 284(5): 3195-3210. Zhang, Y. and Munday, R. (2008). Dithiolethiones for cancer chemoprevention: where do we stand? Mol Cancer Ther 7(11): 3470-3479. Zhao, J., Moore, A. N., Redell, J. B. and Dash, P. K. (2007a). Enhancing expression of Nrf2- driven genes protects the blood brain barrier after brain injury. J Neurosci 27(38): 10240-10248. Zhao, X., Sun, G., Zhang, J., Strong, R., Dash, P. K., Kan, Y. W., Grotta, J. C. and Aronowski, J. (2007b). Transcription factor Nrf2 protects the brain from damage produced by intracerebral hemorrhage. Stroke 38(12): 3280-3286. Zhou, W., Lo, S. C., Liu, J. H., Hannink, M. and Lubahn, D. B. (2007). ERRbeta: a potent inhibitor of Nrf2 transcriptional activity. Mol Cell Endocrinol 278(1-2): 52-62. Zhu, B. T. (2004). CNS dopamine oxidation and catechol-O-methyltransferase: importance in the etiology, pharmacotherapy, and dietary prevention of Parkinson's disease. Int J Mol Med 13(3): 343-353. Zhu, Y., Saunders, M. A., Yeh, H., Deng, W. G. and Wu, K. K. (2002). Dynamic regulation of cyclooxygenase-2 promoter activity by isoforms of CCAAT/enhancer-binding proteins. J Biol Chem 277(9): 6923-6928. Zipper, L. M. and Mulcahy, R. T. (2002). The Keap1 BTB/POZ dimerization function is required to sequester Nrf2 in cytoplasm. J Biol Chem 277(39): 36544-36552. Zweben, J. E., Cohen, J. B., Christian, D., Galloway, G. P., Salinardi, M., Parent, D. and Iguchi, M. (2004). Psychiatric symptoms in methamphetamine users. Am J Addict 13(2): 181- 190.

399

CHAPTER 5: APPENDICES 400

I. SUPPLEMENTAL DATA 401

Figure S1. Tyrosine hydroxylase (TH) immunoblots and densitometric analysis of the striatum in METH-treated Nrf2 wild-type and KO mice. Mice were dosed with saline or the acute dosing regimen of 4 x 5 mg/kg i.p. doses of METH, with a 2-hr interval between each dose, and densitometric analysis of striatal TH in Nrf2 wild-type or Nrf2 KO mice 7 d after the last dose was assessed. β-actin was used to standardize protein levels. a = p<0.05 compared to saline-treated Nrf2 wild-type mice. Each bar represents results of densitometric analysis of immunoblots from 4-6 mice. 402

Figure S2. In utero exposure to METH on GD 16 and 17: Detection and habituation. METH was dissolved in 0.9% saline and administered in 1 dose of 5 mg/kg, i.p. to pregnant Nrf2 (+/-) mice on GD 16 and 17. The mice were allowed to litter out and at weaning offspring were genotyped for Nrf2. Male and female offspring were tested for olfactory deficits measured by odor detection and habituation on postnatal week (PNW) 4. No statistically different detection or habituation deficits were observed between treatment groups or between genotypes. Each bar consisted of 7-8 mice. 403

Figure S3. In utero exposure to METH on GD 16 and 17: Discrimination. Pregnant Nrf2 (+/-) mice were administered one dose of METH (5 mg/kg i.p.) or saline on GD 16 and GD 17. The mice were allowed to litter out and at weaning offspring were genotyped for Nrf2. Male and female (+/+) and (-/-) offspring were tested for olfactory deficits measured by odor detection, habituation and discrimination on postnatal week PNW 4. Discrimination was defined as the difference in ratios between time spent detecting the familiar odor (fifth trial) and time spent detecting the novel odor on the sixth trial (p<0.01). Each group consisted of 7-8 mice.

404

Figure S4. In utero exposure to METH on GD 13 and 14: Detection and habituation. METH was dissolved in 0.9% saline and administered in 1 dose of 10 mg/kg, i.p. to pregnant Nrf2 (+/-) mice on GD 13 and 14. The mice were allowed to litter out and at weaning offspring were genotyped for Nrf2. Male and female offspring were tested for olfactory deficits measured by odor detection and habituation on postnatal week (PNW) 4. No statistically different detection or habituation deficits were observed between treatment groups or between genotypes. Each bar consisted of 7-11 mice. 405

Figure S5. In utero exposure to METH on GD 13 and 14: Discrimination. Pregnant Nrf2 (+/-) mice were administered one dose of METH (10 mg/kg i.p.) or saline on GD 13 and GD 14. The mice were allowed to litter out and at weaning were genotyped for Nrf2. Male and female (+/+) and (-/-) offspring were tested for olfactory deficits measured by odor detection, habituation and discrimination on postnatal week PNW 4. Discrimination was defined as the difference in ratios between time spent detecting the familiar odor (fifth trial) and time spent detecting the novel odor on the sixth trial (p<0.01). Each group consisted of 7-11 mice. 406

II. COPYRIGHT ACKNOWLEDGEMENTS

Permissions were received from:

Elsevier Limited

American Chemical Society

Oxford University Press and the Society of Toxicology