Oxidative Stress Mechanisms in Alcohol Developmental Toxicity

Lutfiya Miller

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy Department of Pharmacology and Toxicology University of Toronto

Lutfiya Miller

Doctor of Philosophy

Department of Pharmacology and Toxicology University of Toronto

2014 ABSTRACT

Consumption of alcohol (ethanol, EtOH) during pregnancy can result in a spectrum of anomalies termed the Fetal Alcohol Spectrum Disorders (FASD), characterized by structural and behavioural deficits, while in utero exposure to methanol (MeOH), a structurally similar alcohol, enhances structural malformations in rodents. The mechanisms underlying these toxicities are unclear, however reactive oxygen species (ROS) have been implicated.

I hypothesize that prenatal exposure to EtOH or MeOH enhance the formation of ROS, which oxidatively damage fetal cellular macromolecules (DNA, protein, lipids) and/or alter fetal redox-sensitive signal transduction, causing structural birth defects and postnatal neurodevelopmental deficits. If so, then altering the balance of pathways of ROS formation and detoxification, and repair of oxidatively damaged DNA, should determine teratological risk.

This hypothesis was tested using the free radical spin trapping agent phenylbutylnitrone (PBN), and pharmacological and genetic modulations of the ROS-forming enzyme NADPH oxidase

(NOX), the antioxidative enzyme catalase and the DNA repair enzyme oxoguanine glycosylase 1

(OGG1).

Embryonic EtOH exposure oxidatively damaged embryonic DNA and enhanced structural malformations, which were: (1) blocked by the free radical spin trapping agent PBN;

(2) decreased in transgenic mice expressing human catalase (hCat); (3) exacerbated in ii acatalasemic (aCat) mice; and, (4) exacerbated in OGG1 knockout mice in a genotype- dependent fashion. Fetal EtOH exposure enhanced fetal brain DNA oxidation and postnatal behavioural deficits in CD-1 mice, and in OGG1 knockout mice in a gene dose-dependent fashion; both effects were blocked by PBN. Interestingly, untreated OGG1 knockout mice exhibited a learning deficit compared to wild-type littermates, constituting the first report of a phenotype for these mice.

Embryonic MeOH exposure enhanced embryopathies, embryonic protein oxidation and upregulated mRNA and protein expression of embryonic ROS-initiating NOXs in CD-1 embryos, providing the first evidence of NOX involvement in MeOH embryotoxicity.

Embryopathies were respectively decreased and increased in hCat and aCat embryos.

These results suggest that ROS are involved in the teratological mechanism of EtOH and

MeOH, and that embryonic catalase, NOX, and DNA damage and repair are determinants of risk.

iii

Acknowledgments

My thesis was a collaborative work, which wouldn’t have been possible for some very important people who were there with me from day one who made the entire experience as wonderful as it all was.

I am grateful to my advisor Peter Wells for his ongoing support, insightful feedback, continuous encouragement and intellectual challenges. I have now developed a great skill of summarizing large bodies of literature into a single table! My experience in his lab has taught me to be a diligent scientist while working both independently and collaboratively.

I would like to thank my committee members Jack Uetrecht and Michael Wiley for their ongoing assessment and input throughout the course of my Ph.D. Whether it was confirmation of a birth defect or a chemical mechanism of electron movement, or feedback during my committee meetings, their insight was always helpful.

I am grateful to my lab mate Aaron Shapiro for his ongoing assistance and support not only with scientific collaborations and technical computer issues, but also for his friendship. Late hours in the lab troubleshooting protocols or marking exams were always more fun with someone to take a sushi or shawarma break with. Thanks to his initiative and desire to save the lab a buck, we always had an interesting time at the conferences we attended over the years.

Amy Sharma, a fellow graduate student in Dr. Uetrecht’s lab, assisted me greatly with my

NADPH oxidase studies, without her I wouldn’t have completed that study in such a timely fashion or had such a fun time doing it. The suspense of waiting for our results always kept us on our toes and in the lab late hours of the evening to find out if our data agreed! Our brief yet

iv ongoing ‘inter-experiment’ chats in the hallway and at the waterbath while thawing our samples always kept the days interesting and gave me a good laugh whenever I needed it.

I had tremendous help from Jun Cheng and Daniel Pinto, two undergraduate students I had the opportunity to supervise over the course of the final years in my Ph.D who not only assisted greatly in the timely tasks of behavioural testing, but also taught me to be a better leader, and train students with an interest in science. Their continued feedback, open minds and subsequent success taught me how fulfilling teaching can really be.

I would like to thank Crystal Lee for teaching me the laborious technique of culturing mouse embryos when I first started in the lab, a skill that allowed me to publish several papers and attend many conferences, as well as teaching me rabbit embryo culture.

None of this would have been possible without the ongoing love, support, encouragement

(and patience!) of Joshua Pinsler, who listened to several practice presentations, gave thoughtful and insightful feedback and taught me practical ways of presenting my research to easily communicate with broad audiences. I’m sure by now he knows more about oxidative stress than he ever thought he would know!

Last but certainly not least, I would like to thank my entire family for always being there when I needed it, Shameen, Shaheema, Reeiyad, Riedwaan, Faiza and Frank. I wouldn’t have come this far without their love and encouragement.

Table of Contents

Acknowledgments ...... iv Table of Contents ...... vi List of Tables ...... xi List of Figures ...... xiii List of Abbreviations ...... xxi List of publications and presentations arising from this thesis ...... xxvii List of Appendices ...... xxx Chapter 1 Introduction ...... 1 1.1 Rationale and Research Objectives ...... 1 1.2 Overview of Development ...... 5 1.2.1 Introduction ...... 5 1.2.2 Week 1: The start of human development ...... 9 1.2.3 Week 2: Formation of the embryonic bilaminar disc ...... 11 1.2.4 Week 3: Germ layer formation and differentiation of early tissues and organs ...... 11 1.2.5 Weeks 4-8: Organogenesis ...... 18 1.2.6 Week 9 – Birth: Fetal Period ...... 19 1.3 Mechanisms of Teratogenesis ...... 21 1.3.1 Receptor-mediated mechanisms ...... 21 1.3.2 Reactive intermediate-mediated mechanisms ...... 28 1.3.2.1 Electrophilic reactive intermediates ...... 28 1.3.2.1 Free Radical reactive intermediates ...... 30 1.4 Reactive Oxygen Species (ROS) and antioxidative defense mechanisms ...... 30 •- 1.4.1 Superoxide anion (O2 ) ...... 33

1.4.2 Hydrogen peroxide (H2O2) ...... 33 1.4.3 Hydroxyl radical (•OH) ...... 33 1.4.4 Endogenous sources of ROS ...... 34 1.4.5 Exogenous sources of ROS ...... 37 1.4.6 Antioxidative defenses ...... 37 1.5 Deleterious effects of ROS ...... 42 1.5.1 Altered signal transduction ...... 42 1.5.2 Oxidative macromolecular damage ...... 46 1.5.2.1 DNA oxidation ...... 46

1.5.2.2 Protein oxidation ...... 49 1.5.2.3 Lipid peroxidation ...... 51 1.6 Studying mechanisms of teratogenesis ...... 53 1.6.1 Whole embryo culture as a model for teratogenesis ...... 54 1.6.2 Passive avoidance task as a model for behavioural deficits ...... 66 1.6.3 Chemical probes ...... 67 1.6.3.1 Free radical spin trapping agent phenylbutylnitrone (PBN) ...... 69 1.6.3.2 NADPH Oxidase (NOX) inhibitors ...... 73 1.6.3.3 Polyethylene glycol (PEG)-conjugated catalase ...... 77 1.7 NADPH oxidases (NOX) ...... 81 1.7.1 NOX2 ...... 84 1.7.2 NOX1 ...... 85 1.7.3 NOX3 ...... 85 1.7.4 NOX4 ...... 85 1.7.5 NOX5 ...... 86 1.7.6 Dual Oxidases (DUOX) 1/2 ...... 86 1.7.7 Developmental NOX ontogeny ...... 87 1.7.8 Role of NOX in disease ...... 89 1.7.9 NOX knockout mouse models ...... 92 1.7.10 Role of NOX in teratogenesis ...... 92 1.8 Catalase ...... 97 1.8.1 Catalase gene ...... 97 1.8.2 Enzyme structure and catalytic mechanism ...... 97 1.8.3 Catalase in disease ...... 103 1.8.4 Embryonic catalase expression ...... 105 1.8.5 Transgenic human catalase-expressing mouse ...... 109 1.8.6 Acatalasemic mouse ...... 119 1.9 8-Oxoguanine Glycosylase 1 (OGG1) ...... 126 1.9.1 8-oxoguanine ...... 126 1.9.2 Consequences of 8-oxoG formation ...... 126 1.9.3 Base excision repair (BER) pathway ...... 128 1.9.3.1 Short-patch BER pathway ...... 128 1.9.3.2 Long-patch BER pathway ...... 128 1.9.4 OGG1 gene ...... 131 1.9.5 OGG1 protein structure ...... 134

vii

1.9.6 OGG1 catalytic mechanism ...... 134 1.9.7 Developmental OGG1 expression ...... 136 1.9.8 OGG1 knockout mouse ...... 136 1.9.9 OGG1 in teratogenesis ...... 139 1.10 Ethanol (EtOH) ...... 144 1.10.1 Pharmacokinetics ...... 144 1.10.2 Teratogenesis ...... 154 1.10.2.1 Mechanisms of EtOH teratogenesis ...... 155 1.10.2.1 Evidence for the involvement of ROS in EtOH teratogenesis ...... 158 1.11 Methanol (MeOH) ...... 166 1.11.1 Pharmacokinetics ...... 166 1.11.2 Teratogenesis ...... 169 1.11.2.1 Evidence for the involvement of ROS in MeOH teratogenesis ...... 169 Chapter 2 Study 1: Altered methanol embryopathies in embryo culture with mutant catalase-deficient mice and transgenic mice expressing human catalase ...... 174 2.1 Abstract ...... 175 2.2 Introduction ...... 176 2.3 Methods ...... 178 2.4 Results ...... 185 2.5 Discussion ...... 189 Chapter 3 Enhanced NADPH oxidases and reactive oxygen species in the mechanism of methanol-initiated protein oxidation and embryopathies in mouse embryo culture ...... 195 3.1 Abstract ...... 196 3.2 Introduction ...... 197 3.3 Methods ...... 200 3.4 Results ...... 207 3.5 Discussion ...... 227 Chapter 4 Enhanced embryonic catalase protects against ethanol-initiated embryopathies in acatalasemic and transgenic human catalase-expressing mice in embryo culture ...... 233 4.1 Abstract ...... 234 4.2 Introduction ...... 235 4.3 Methods ...... 239 4.4 Results ...... 242 4.5 Discussion ...... 252

viii

Chapter 5 Embryonic catalase protects against ethanol-initiated DNA oxidation and teratogenesis in acatalasemic and transgenic human catalase-expressing mice ...... 257 5.1 Abstract ...... 258 5.2 Introduction ...... 259 5.3 Methods ...... 262 5.4 Results ...... 269 5.5 Discussion ...... 287 5.6 Supplemental Figures ...... 293 Chapter 6 Embryonic DNA repair and gender are risk factors in ethanol embryopathies in oxoguanine glycosylase 1 (OGG1) knockout mice: a role for oxidatively damaged DNA and protection by a free radical spin trapping agent ...... 305 6.1 Abstract ...... 306 6.2 Introduction ...... 307 6.3 Methods ...... 310 6.4 Results ...... 316 6.5 Discussion ...... 329 6.6 Supplemental Figures ...... 333 Chapter 7 Embryonic DNA repair and ethanol-initiated behavioural deficits in oxoguanine glycosylase 1 (OGG1) knockout mice: a pathogenic role for oxidatively damaged DNA and protection by a free radical spin trapping agent ...... 335 7.1 Abstract ...... 336 7.2 Introduction ...... 337 7.3 Methods ...... 342 7.4 Results ...... 349 7.5 Discussion ...... 360 Chapter 8 The free radical spin trapping agent phenylbutylnitrone reduces fetal brain DNA oxidation and postnatal cognitive deficits by in utero exposure to a non- structurally teratogenic dose of ethanol: a role for oxidative stress ...... 364 8.1 Abstract ...... 365 8.2 Introduction ...... 366 8.3 Methods ...... 370 8.4 Results ...... 376 8.5 Discussion ...... 395 8.6 Supplemental Figures ...... 401 Chapter 9 Summary, Conclusions and Future Studies ...... 405 9.1 Summary and Conclusions ...... 406 9.1.1 Ethanol studies ...... 410

9.1.2 Methanol studies ...... 414 9.1.3 Embryopathic potential of physiological levels of embryonic ROS ...... 419 9.1.4 Strain differences in EtOH teratogenicity ...... 420 9.2 Future Studies ...... 428 Chapter 10 REFERENCES ...... 431 Chapter 11 Appendices ...... 474 11.1 Species differences in methanol and formic acid pharmacokinetics in mice, rabbits and primates ...... 475 11.2 Oxidative stress and species differences in the metabolism, developmental toxicity and carcinogenic potential of methanol and ethanol1 ...... 512 11.3 In vivo teratogenesis pictures ...... 602

List of Tables

Table 1.1: Neural germ layer derivatives ...... 14

Table 1.2: Key events occurring during the embryonic period ...... 20

Table 1.3: Characteristics of xenobiotic toxicity initiated by reactive intermediates compared with reversible, receptor-mediated interactions ...... 23

Table 1.4: Comparison of EtOH-initiated embryopathies in whole embryo culture ...... 58

Table 1.5: Comparison of MeOH-initiated embryopathies in whole embryo culture ...... 59

Table 1.6: Effect of in utero ethanol exposure on postnatal performance on a passive avoidance task ...... 68

Table 1.7: Pharmacological effects of PBN ...... 72

Table 1.8: NADPH oxidase inhibitors ...... 74

Table 1.9: Safety evaluation of PEG-catalase in mice and rats ...... 78

Table 1.10: PEG-Catalase in models of developmental toxicity ...... 80

Table 1.11: Tissue distribution and subcellular localization of NADPH oxidase (NOX) enzymes ...... 82

Table 1.12: Genetically modified NADPH oxidase (NOX) mouse models ...... 93

Table 1.13: Mouse models of Base excision repair (BER) pathway enzymes resulting in lethality ...... 138

Table 1.14: Phenotype of OGG1 deficient mice and their response to oxidative stressors ...... 143

Table 1.15: Blood alcohol concentration (BAC) and its effects on human behaviour ...... 146

Table 1.16: Approximate mean lethal dose of EtOH for humans and some common laboratory animals...... 147

Table 1.17: Estimated volume of distribution in men and women ...... 148

Table 1.18: Ethanol elimination rates measured in humans and rodents ...... 153

Table 1.19: Structural birth defects initiated by in utero EtOH exposure in rodents ...... 156

Table 1.20: Brain regions affected by in utero EtOH exposure ...... 157

Table 1.21: Mechanisms of EtOH-initiated developmental toxicity ...... 158

Table 1.22: Evidence of ROS in EtOH-initiated embryopathies and teratogenesis ...... 160

Table 1.23: Measurement of oxidatively damaged macromolecules in fetal brain tissues after in utero EtOH exposure ...... 164

Table 1.24: Indirect evidence for ROS in EtOH-initiated behavioural deficits ...... 165

Table 1.25: Developmental toxicity of methanol (MeOH) in rodents ...... 170

Table 1.26: Evidence for MeOH-initiated oxidative stress ...... 171

xii

List of Figures

Chapter 1

Figure 1.1: Thesis outline ...... 3

Figure 1.2: Human development and critical periods for drug exposure ...... 7

Figure 1.3: Comparison of key developmental events in mice and humans...... 8

Figure 1.4: Cleavage of the zygote to form a blastocyst ...... 10

Figure 1.5: Gastrulation and the tissues derived from the three germ layers ...... 12

Figure 1.6: Beginning of neurulation ...... 15

Figure 1.7: Transverse section through an embryo illustrating neurulation ...... 16

Figure 1.8: Receptor- versus reactive intermediate-mediated mechanisms of teratogenesis ...... 22

Figure 1.9: Renin-angiotensin signaling pathway and AT1 receptor inhibition by sartan drugs. 25

Figure 1.10: AHR pathway ...... 26

Figure 1.11: Reactive intermediate-mediated mechanisms of developmental toxicity ...... 29

Figure 1.12: Generation of ROS in biological systems ...... 32

Figure 1.13: ROS generation by assembly of the NOX regulatory subunits in phagocytes...... 36

Figure 1.14: Reactions of antioxidative enzymes and antioxidants...... 39

Figure 1.15: Redox control of various proteins and transcription factors by reversible and irreversible modifications of cysteine ...... 44

Figure 1.16: Major products of oxidative DNA lesions ...... 47

Figure 1.17: Reaction of hydroxyl radicals (•OH) with guanine residues of DNA to form the molecular lesion 7-8,dihydro-8-oxoguanine (8-oxoG)...... 48

xiii

Figure 1.18: Endpoints measured in whole embryo culture ...... 56

Figure 1.19: EtOH embryopathies in mouse and rat whole embryo culture (48 hours) ...... 61

Figure 1.20: MeOH embryopathies in mouse and rat whole embryo culture (24 hours) ...... 62

Figure 1.21: MeOH embryopathies in mouse and rat whole embryo culture (48 hours) ...... 63

Figure 1.22: Comparison of MeOH vs. EtOH embryopathies in mice ...... 64

Figure 1.23: Comparison of MeOH vs. EtOH embryopathies in rats ...... 65

Figure 1.24: PBN trapping hydroxyl radical ...... 70

Figure 1.25: Mechanisms of action of DPI inhibiting NOXs ...... 76

Figure 1.26: Blood-circulating life in mice of PEG-catalase...... 79

Figure 1.27: Subunit composition of the seven mammalian NADPH oxidase (NOX) isoforms 83

Figure 1.28: Developmental NOX mRNA ontogeny ...... 88

Figure 1.29: Developmental NOX protein expression...... 90

Figure 1.30: Primary sequence and secondary structure of human catalase...... 98

Figure 1.31: Tertiary structure of human catalase ...... 99

Figure 1.32: Catalytic mechanism of catalase...... 100

Figure 1.33: Quantitative analysis of the 12.2 and 2.4 kb catalase-related transcripts during development in mice...... 106

Figure 1.34: Developmental profile of the catalase-specific RNA and enzyme activity in the carcass and liver during development in mice...... 108

Figure 1.35: Time course of embryonic catalase activity after maternal injection of PEG- catalase...... 110

Figure 1.36: Map of the P1 bacteriophage clones containing the entire human CAT gene...... 111 xiv

Figure 1.37: Catalase expression in heterozygous Tg(CAT)+/o, homozygous Tg(CAT)+/+ and wild type mice ...... 112

Figure 1.38: Distribution of catalase in the mitochondria and peroxisomes of Tg(CAT)+/+ and wild-type mice...... 114

Figure 1.39: Representative gel for genotyping showing DNA bands for mice expressing human catalase (hCat) and their proximate C567BL/6 wild-type (WT) controls...... 116

Figure 1.40: Embryopathies in wild-type (WT) mouse embryos and embryos expressing hCat exposed to phenytoin or its vehicle...... 117

Figure 1.41: Maternal tissue activity of catalase in aCat mice...... 120

Figure 1.42: Maternal tissue activities of catalase and GPx in aCat mice...... 121

Figure 1.43: Prevention of phenytoin embryopathies in aCat mice with catalase protein therapy...... 123

Figure 1.44: Representative gel for genotyping showing DNA bands for catalase-deficient (acatalasemic, aCat) mice and their proximate C3HeB/FeJ WT controls...... 125

Figure 1.45: Base pairing properties of 8-oxoG residues in DNA...... 127

Figure 1.46: BER pathway overview ...... 129

Figure 1.47: Alignment of amino acid sequences of α- and β-human (h), mouse (m) and yeast (y) Ogg1 proteins...... 132

Figure 1.48: Splicing variants of the hOGG1 primary transcript...... 133

Figure 1.49: Structure of the α- and β-hOgg1 isoforms...... 135

Figure 1.50: Enzymatic mechanism of hOGG1...... 137

Figure 1.51: Targeted disruption of the murine OGG1 locus...... 140

Figure 1.52: Cleavage activity of OGG1 on 8-oxoG in various tissues from ogg1 null mice... 141

Figure 1.53: Pathways of ethanol metabolism ...... 150

Figure 1.54: ROS generation by CYP2E1-mediated metabolism of EtOH...... 152

Figure 1.55: Metabolic pathways of methanol in rodents and humans ...... 167

Figure 1.56: Role of folate in the metabolism of formic acid...... 168

Chapter 2: Study 1

Figure 2.1: Representative gel for genotyping showing DNA bands for mice expressing human catalase (hCat) and their proximate C567BL/6 wild-type (WT) controls...... 180

Figure 2.2: Representative gel for genotyping showing DNA bands for catalase-deficient (acatalasemic, aCat) mice and their proximate C3HeB/FeJ WT controls...... 182

Figure 2.3: Effect of hCat expression on methanol (MeOH) embryopathies...... 186

Figure 2.4: Effect of catalase deficiency on MeOH embryopathies in aCat embryos...... 188

Chapter 3: Study 2

Figure 3.1: Upregulation of embryonic p22phox mRNA by in utero methanol (MeOH) exposure and the effect of pretreatment with the free radical spin trap phenylbutylnitrone (PBN) and the NOX inhibitor diphenyleneiodonium (DPI)...... 209

Figure 3.2: Upregulation of embryonic p22phox protein by in utero MeOH exposure and the effect of pretreatment with PBN and DPI...... 211

Figure 3.3: Enhanced oxidatively damaged embryonic protein following in utero MeOH exposure and the effect of pretreatment with PBN and DPI...... 214

Figure 3.4: Developmental endpoints assessed in mouse whole embryos in culture...... 216

xvi

Figure 3.5: Effect of pretreatment with PBN on concentration-dependent MeOH embryopathies in embryo culture...... 218

Figure 3.6: Effect of pretreatment with DPI on MeOH embryopathies in embryo culture...... 221

Figure 3.7: Effect of the prostaglandin H synthase (PHS) inhibitor eicosatetraynoic acid (ETYA) on MeOH embryopathies in embryo culture...... 224

Figure 3.8: Effect of the PHS inhibitor ETYA on phenytoin embryopathies in embryo culture...... 226

Chapter 4: Study 3

Figure 4.1: Protection against ethanol (EtOH) embryopathies in transgenic mice expressing human catalase (hCat)...... 244

Figure 4.2: Exacerbation of EtOH embryopathies in catalase-deficient acatalasemic (aCat) mice...... 248

Figure 4.3: Strain differences in EtOH embryopathies in C57BL/6 WT embryos and C3H WT embryos...... 251

Chapter 5: Study 4

Figure 5.1: Number of pups per litter, gender distribution of offspring and maternal lethality in transgenic mice expressing human catalase (hCat) compared to C57BL/6 wild-type (WT) and acatalasemic (aCat) mice compared to C3H WT controls...... 271

Figure 5.2: Enhanced in vivo EtOH developmental toxicity in acatalasemic (aCat) mice, and converse protection in transgenic mice expressing human catalase (hCat)...... 273

Figure 5.3: Effect of polyethylene glycol-conjugated catalase (PEG-Cat) pretreatment on EtOH developmental toxicity in acatalasemic (aCat) mice and C57BL/6 mice...... 278

xvii

Figure 5.4: Enhanced EtOH-initiated embryonic DNA oxidation in gestational day 12 embryos of acatalasemic (aCat) mice, and converse protection in embryos of transgenic mice expressing human catalase (hCat)...... 282

Figure 5.5: Effect of catalase deficiency and strain on plasma EtOH pharmacokinetics in acatalasemic (aCat) mice and transgenic mice expressing human catalase (hCat), and the effect of strain on in vivo EtOH developmental toxicity in C3H WT mice and C57BL/6 WT mice. .. 286

Supplemental Figure S5.6: Spectrum of ethanol (EtOH)-initiated malformations in acatalasemic (aCat) mice...... 294

Supplemental Figure S5.7: Spectrum of EtOH-initiated malformations in transgenic mice expressing human catalase (hCat)...... 296

Supplemental Figure S5.8: Effect of pretreatment with polyethylene glycol-conjugated catalase (PEG-Cat) on the spectrum of EtOH-initiated malformations in acatalasemic (aCat) mice...... 298

Supplemental Figure S5.9: The effect of PEG-Cat pretreatment on the spectrum of EtOH- initiated malformations in C57BL/6 mice...... 300

Supplemental Figure S5.10: The effect of gender on EtOH-initiated malformations in acatalasemic (aCat) mice and transgenic mice expressing human catalase (hCat)...... 302

Supplemental Figure S5.11: The effect of mouse strain on EtOH-initiated malformations in acatalasemic (aCat) mice and transgenic mice expressing human catalase (hCat)...... 304

Chapter 6: Study 5

Figure 6.1: Effect of ethanol (EtOH) and phenylbutylnitrone (PBN) pretreatment on oxidatively damaged DNA in oxoguanine glycosylase 1 (OGG1) wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout embryos...... 318

Figure 6.2: Effect of oxoguanine glycosylase 1 (ogg1) genotype on ethanol (EtOH) embryopathies in embryo culture...... 320

xviii

Figure 6.3: Genotype-dependent correlation between head length and anterior neuropore closure...... 323

Figure 6.4: Effect of PBN pretreatment on EtOH embryopathies in oxoguanine glycosylase 1 (ogg1) knockout embryos in culture...... 325

Figure 6.5: Determination of embryonic gender by genotype, and the effect of gender and oxoguanine glycosylase 1 (ogg1) genotype on EtOH embryopathies in embryo culture...... 328

Supplemental Figure S6.6: Effect of oxoguanine glycosylase 1 (ogg1) genotype on ethanol (EtOH) embryopathies in embryo culture...... 334

Chapter 7: Study 6

Figure 7.1: Postulated involvement of reactive oxygen species (ROS) and oxidatively damaged cellular macromolecules, and particularly DNA, which is repaired by oxoguanine glycosylase 1 (OGG1), in the mechanism of postnatal behavioural deficits initiated by in utero exposure to ethanol (EtOH). The free radical spin trapping agent phenylbutylnitrone (PBN) can detoxify drug and ROS free radical intermediates...... 341

Figure 7.2: Representative gel for genotyping showing DNA bands for oxoguanine glycosylase 1 (OGG1) wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout mice...... 345

Figure 7.3: Oxidatively damaged DNA in fetal brains from oxoguanine glycosylase 1 (OGG1) wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout mice...... 351

Figure 7.4: Time course for postnatal learning development in oxoguanine glycosylase 1 (OGG1) wild-type (+/+), heterozygous (+/-) and knockout (-/-) progeny exposed in utero to EtOH with or without PBN pretreatment...... 353

Figure 7.5: Effect of PBN pretreatment on EtOH-initiated learning deficits in oxoguanine glycosylase 1 (OGG1) wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout progeny...... 356

xix

Figure 7.6: Fetal weight in oxoguanine glycosylase 1 (OGG1) wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout mice...... 359

Chapter 8: Study 7

Figure 8.1: Effect of ethanol (EtOH) exposure during the fetal period on passive avoidance learning in CD-1 mice, and protection by phenylbutylnitrone (PBN)...... 378

Figure 8.2: Passive avoidance task retention at Trial #1 between time points in CD-1 mice exposed to EtOH during the fetal period...... 381

Figure 8.3: Effect of in utero EtOH exposure during the fetal period on postnatal weight in CD-1 mice...... 383

Figure 8.4: Effect of PBN on EtOH-enhanced oxidatively damaged DNA in fetal brains of CD-1 mice exposed to EtOH during the fetal period...... 386

Figure 8.5: Teratogenicity and fetal toxicity in CD-1 mice at birth following EtOH treatment during the embryonic period...... 388

Figure 8.6: Comparison of in vivo ethanol teratogenesis in C57BL/6 and CD-1 mice following treatment during the embryonic period...... 391

Figure 8.7: Comparison of EtOH pharmacokinetics in C57BL/6 and CD-1 mice following a single dose of ethanol...... 394

Supplementary Figure S8.8: Long-term passive avoidance test retention at trial #1 between time points in CD-1 mice exposed to ethanol (EtOH) during the fetal period...... 402

Supplementary Figure S8.9: Comparison of passive avoidance behaviour data analyzing the fetus as the unit versus the litter as the unit...... 404

List of Abbreviations

4-HNE hydroxynonenal

8-oxodG 8-oxo-2’-deoxyguanosine aCat acatalasemic

ACE angiotensin converting enzyme

AHR aryl hydrocarbon receptor

AHRE aryl hydrocarbon responsive element

ALDH2 aldehyde dehydrogenase

AP site apurinic or apyrimidinic site

AP-1 activator protein 1

APEX1 AP endonuclease 1

ARBD alcohol-related birth defects

ARND alcohol-related neurological defects

ARNT aryl hydrocarbon receptor nuclear translocator

AT angiotensinogen

AT1 angiotensin II type 1 receptor

AT2 angiotensin II type 2 receptor

AT-I angiotensin

AT-II angiotensin II

ATP adenosine triphosphate

B[a]P benzo[a]pyrene

BAC blood alcohol concentration

BER base excision repair xxi

BMP bone morphogenetic protein

CGD chronic granulomatous disease

CN cranial nerve

CNS central nervous system

COX2 cyclooxygenase 2

Cul3 cullin 3

CuZnSOD copper-zinc superoxide dismutase

CYP cytochrome P450 cys cysteine cyt b558 flavocytochrome b558

D3T 3H-1,2 dithiole-3-thione

DHA docosahexanoic acid dpc days post coitum

DPI diphenyleneiodonium chloride dpo days post ovulation

DUOX dual oxidase

EGFR epidermal growth factor receptor

EPA eicosapentanoic acid

ER estrogen receptor

ETC electron transport chain

EtOH ethanol

FAD flavin adenine dinucleotide

FAS fetal alcohol syndrome

xxii

FASD fetal alcohol spectrum disorders

FEN1 flap endonuclease 1

FGF fibroblast growth factor

G6PD glucose-6-phosphate dehydrogenase

GD gestational day

GI gastrointestinal

GPx glutathione peroxidase

Grx glutaredoxin

GSH glutathione

GSH Rd glutathione reductase

GST glutathione S-transferase

H2O2 hydrogen peroxide hCat transgenic human catalase expressing

HhH helix-hairpin-helix

HOCl hypochlorous acid

IFN-γ interferon gamma

IL-1β interleukin 1 beta

IL-6 interleukin 6 iNOS inducible nitric oxide synthase

IP isoprostane

JNK jun terminal kinase

Keap1 kelch-like ECH-associated protein 1

LIG1 DNA ligase 1

xxiii

LIG3 DNA ligase 3

LPO lipoxygenase

LPS lipopolysaccharide mA milliampere

MDA malondialdehyde

MeOH methanol

MMR mismatch repair

MnSOD manganese superoxide dismutase

MTS mitochondrial targeting signal

NADH nicotinamide adenine dinucleotide

NADPH nicotinamide adenine dinucleotide phosphate

Neo neomycin resistance gene polyadenylation signal

NER nucleotide excision repair

NF-κB nuclear factor kappa-light-chain-enhancer of activated B cells

NLS nuclear localization sequence

NO• nitric oxide

NOS nitric oxide synthase

NOX NADPH oxidase

NOXA1 NOX activator 1

NOXO1 NOX organizer 1

O2 oxygen

O2•- superoxide anion

OGG1 oxoguanine glycosylase 1

xxiv

OH hydroxyl radical

ONOO peroxynitrite

ORF open reading frame

PAH polycyclic aromatic hydrocarbons

PBN alpha-phenyl-N-tert-butylnitrone

PDI protein disulfide isomerase

PEG polyethylene glycol

PHS prostaglandin H synthase

PMA phorbol myristate acetate

PNS peripheral nervous system

POLβ DNA polymerase beta

PRR proline rich region

Prx peroxiredoxin

PTK protein tyrosine kinase

PTP protein tyrosine phosphatase

PUFA polyunsaturated fatty acid

ROS reactive oxygen species

SH3 Src homology 3 siRNA small interfering RNA

SNP single nucleotide polymorphism

SOD superoxide dismutase

TCDD 2,3,7,8-tetrachlorodibenzo-p-dioxin

TGF-β transforming growth factor beta

xxv

TNF-α tumor necrosis factor alpha

Trx thioredoxin wpo weeks post ovulation

XRCC1 X-ray repair complementing defective repair in Chinese hamster cells 1

xxvi

List of publications and presentations arising from this thesis

Publications and submitted manuscripts

1. Sweeting, J.N., Siu, M., McCallum, G., Miller, L., Wells, P.G. (2010) Species differences in methanol and formate disposition in mice, rabbits and primates. Toxicology and Applied Pharmacology 247(1): 28-35.

2. Miller, L., Wells, P.G. (2011) Altered methanol embryopathies in embryo culture with mutant catalase-deficient mice and transgenic mice expressing human catalase. Toxicology and Applied Pharmacology 252(1): 55-61.

3. Wells, P.G., McCallum, G.P., Miller, L., Siu, M.T., Sweeting, J.N. Oxidative stress and species differences in the metabolism, developmental toxicity and carcinogenic potential of methanol and ethanol. In: The Toxicology of Methanol, JJ Clary (ed.), pp. 169-253, Wiley, Hoboken, U.S.A. 2013.

4. Miller, L., Shapiro, A., Cheng, J., Wells, P.G. (2013) The free radical spin trapping agent phenylbutylnitrone reduces fetal brain DNA oxidation and postnatal cognitive deficits caused by in utero exposure to a non-structurally teratogenic dose of ethanol: a role for oxidative stress. Free Radical Biology & Medicine 60: 223-232.

5. Miller, L., Shapiro, A., Wells, P.G. (2013) Embryonic catalase protects against ethanol- initiated teratogenesis and DNA oxidation in acatalasemic and transgenic human catalase- expressing mice. Toxicological Sciences (in press).

6. Miller, L., Sharma, A., Wells, P.G. (2013) Enhanced NADPH oxidases and reactive oxygen species in the mechanism of methanol-initiated protein oxidation and embryopathies in mouse embryo culture. Toxicological Sciences (submitted).

7. Miller, L., Wells, P.G. (2013) Embryonic catalase protection against ethanol embryopathies in acatalasemic and human catalase-expressing mice in embryo culture. Toxicology and Applied Pharmacology (in preparation).

8. Miller, L., Shapiro, A., Wells, P.G. (2013) Embryonic DNA repair and gender are risk factors in Oxoguanine Glycosylase 1 (OGG1) knockout mice: A role for oxidatively damaged DNA and protection by a free radical spin trapping agent. Nature Medicine (in preparation).

9. Miller, L., Pinto, D., Shapiro, A., Wells, P.G. Embryonic DNA repair and ethanol-initiated behavioural deficits in Oxoguanine glycosylase 1 (OGG1) knockout mice: a role of oxidatively damaged DNA and protection by a free radical spin trapping agent. Antioxidants & Redox Signaling (in preparation).

xxvii

Published abstracts 1. Miller, L., Wells, P.G. (2009) Ethanol embryopathies, oxidative stress, and protection by a free radical spin trapping agent. Birth Defects Research Part A: Clinical and Molecular Teratology 85(5): 399 (No. 4).

2. Scaysbrook, J.N., Siu, M., McCallum, G.P., Miller, L., Wells, P.G. (2009) Species differences in methanol and formic acid disposition and catalase activity in mice, rabbits and primates. Birth Defects Research Part A: Clinical and Molecular Teratology 85: 453 (No. P74).

3. Miller, L., Wells, P.G. (2010) Ethanol teratogenesis, oxidative stress and protection by a free radical spin trapping agent. Birth Defects Research Part A: Clinical and Molecular Teratology 88(5): 354 (No. 13).

4. Miller, L., Wells P.G. (2010) Reduced ethanol embryopathies in embryo culture in transgenic mice expressing human catalase. Toxicological Sciences (Supplement: The Toxicologist 114(1): 184 [No. 863]).

5. Shapiro, A.M, Poon, J.C.H., Miller, L., Kanawaty, A., Henderson, J.T., Ramkissoon, A., Wells, P.G. (2010) In utero bisphenol A exposure may cause structural and functional changes in the developing murine nervous system. Toxicological Sciences (Supplement: The Toxicologist 114(1): 183 [No. 859]).

6. Miller, L., Wells P.G. (2010) Altered methanol embryopathies in embryo culture with mutant catalase-deficient mice and transgenic mice expressing human catalase. Proceedings of the 42nd Annual Symposium of the Society of Toxicology of Canada, Abstract No. 39.

7. Miller, L., Cheng, J., Wells, P.G. (2011) The free radical spin trapping agent phenylbutylnitrone reduces postnatal cognitive deficits caused by in utero ethanol exposure in CD-1 mice. Proceedings of the 43rd Annual Symposium of the Society of Toxicology of Canada.

8. Miller, L., Wells, P.G. (2012) Methanol embryopathies in mouse embryo culture following pretreatment with a free radical spin trapping agent and inhibitors of prostaglandin H synthase and NADPH oxidases. Toxicol. Sci. (Supplement: The Toxicologist) 126(1): 417-418 (No. 1940).

9. Miller, L., Shapiro, A., Wells, P.G. (2012) Endogenous and exogenously enhanced embryonic catalase protects against ethanol-initiated teratogenesis and DNA oxidation in human catalase-expressing mice. Proceedings of the Annual Great Lakes Mammalian Development meeting, Abstract No. 19, Toronto, Ontario.

10. Wells, P.G., Abramov, J.P., Lam, K.C.H., Miller, L., McPherson, J.P., Ondovcik, S.L., Ramkissoon, A., Shapiro, A.M., Siu, M., Sweeting, J.N., Wiley, M.J. (2012) Embryonic and fetal biochemical determinants of reactive oxygen species-mediated chemical teratogenesis and neurodevelopmental deficits. Proceedings of the 7th Meeting of the Canadian Oxidative Stress Consortium, p. 43, Lakehead University, Thunder Bay. xxviii

11. Miller, L., Shapiro, A. and Wells, P.G. (2012) Embryonic catalase protects against ethanol embryopathies in acatalasemic and human catalase-expressing mice in embryo culture. Proceedings of the Annual Meeting of the Canadian Society of Pharmacology and Therapeutics, 19 (2): e262 (Abstract No. 6).

12. Miller, L., Shapiro, A., Wells, P.G. (2012) Embryonic DNA repair and gender are risk factors in ethanol embryopathies in Oxoguanine Glycosylase 1 (OGG1) knockout mice: A role for oxidatively damaged DNA and protection by a free radical spin trapping agent Birth Defects Research Part A: Clinical and Molecular Teratology 94(5): 317 (Abstract No. 13).

13. Wells, P.G., Miller, L., Ramkissoon, A., Shapiro, A.M. (2012) Fetal oxidative stress, DNA damage and repair in methamphetamine neurodevelopmental deficits in mice. Birth Defects Research Part A: Clinical and Molecular Teratology 94(5): 304 (Abstract No. S16)

14. Shapiro, A., Miller, L., Wells, P.G. (2012) Developmental consequences of embryonic oxidative stress from methamphetamine and ethanol in BRCA1-deficient mice in vivo or in embryo culture. Birth Defects Research Part A: Clinical and Molecular Teratology 94(5): 317 (Abstract No. 14).

15. Miller, L., Pinto, J., Shapiro, A., Wells, P.G. (2013) Embryonic DNA repair and ethanol- initiated behavioural deficits in oxoguanine glycosylase 1 (OGG1) knockout mice: a role for oxidatively damaged DNA and protection by a free radical spin trapping agent. Toxicological Sciences (Supplement: The Toxicologist) Vol 132(1): 218 (Abstract No. 1022).

16. Shapiro, A., Miller, L., Wells, P.G. (2013) Breast Cancer 1 (BRCA1)-deficient mice develop normally but are more susceptible to ethanol- and methamphetamine-initiated embryopathies. Toxicological Sciences (Supplement: The Toxicologist) Vol 132(1): 213 (Abstract No. 999).

xxix

List of Appendices

2. Wells, P.G., McCallum, G.P., Miller, L., Siu, M.T. and Sweeting, J.N. Oxidative stress and species differences in the metabolism, developmental toxicity and carcinogenic potential of methanol and ethanol. In: The Toxicology of Methanol, JJ Clary (ed.), pp. 169-253, Wiley, Hoboken, U.S.A. 2013.

xxx 1

Chapter 1 Introduction

Note: Text in italics indicates content taken from the following publication for which I am a co- author: Wells, P.G., McCallum, G.P., Miller, L., Siu, M.T. and Sweeting, J.N. Oxidative stress and species differences in the metabolism, developmental toxicity and carcinogenic potential of methanol and ethanol. In: The Toxicology of Methanol, JJ Clary (ed.), pp. 169-253, Wiley, Hoboken, U.S.A. 2013. 1.1 Rationale and Research Objectives

Compounds that can disrupt normal embryo or fetal development are collectively termed

‘teratogens’. These can include drugs or environmental chemicals, as well as other agents such as ionizing radiation. Embryonic development is an exquisitely regulated process, involving both

spatial and temporal control of signal transduction pathways to control cellular division,

differentiation, migration and apoptosis, which make the developing embryo particularly

sensitive to the effects of teratogens. Teratogens can exert their effects either through receptor-

mediated pathways or reactive intermediate-mediated pathways, which can result in the

- formation of reactive oxygen species (ROS) including superoxide (O2• ), hydrogen peroxide

(H2O2) and the hydroxyl radical (•OH). In utero ethanol (EtOH) exposure can result in both

structural and behavioural deficits, collectively termed ‘Fetal Alcohol Spectrum Disorders

(FASD)’, which is considered to be the leading preventable cause of mental retardation in the

Western world (Mattson et al., 2011). The mechanisms underlying these toxicities remain unclear, and although ROS have been implicated, their source and determinants of risk remain unclear. Similarly, methanol (MeOH), a structurally related alcohol, has been shown to be developmentally toxic in mice and rats, although developmental toxicity to humans remains unclear. Given their structural similarity, MeOH may initiate birth defects through similar ROS- mediated pathways. The focus of my thesis was to identify potential determinants of risk responsible for the teratogenic effects of EtOH and MeOH, using a combination of in vivo and

2 whole embryo culture approaches, assessing both structural and behavioural outcomes in the offspring (Fig. 1.1).

The objectives of my research were:

1. To determine the contribution of ROS to EtOH- and MeOH-initiated structural

embryopathies upon exposure during the embryonic period of development with the use of

the free radical spin trapping agent phenylbutylnitrone (PBN) in a whole embryo culture

model, thereby examining embryonic mechanisms independent of maternal factors.

2. To determine the contribution of ROS and embryonic DNA oxidation in EtOH-initiated

structural teratogenesis in vivo upon exposure during the embryonic period of development

with the use of a free radical spin trapping agent PBN, to compare to the findings obtained

in whole embryo culture.

3. To determine the contribution of ROS and fetal brain DNA oxidation in EtOH-initiated

behavioural deficits in vivo upon exposure during the fetal period of development with the

use of the free radical spin trapping agent PBN.

4. To examine and compare the relative sensitivities of the organogenesis stage embryo and

the developing fetal brain to EtOH exposure during the embryonic and fetal periods,

respectively.

5. To determine whether in utero MeOH exposure can upregulate embryonic NADPH

oxidase (NOX) mRNA and protein expression, thereby enhancing ROS formation that can

oxidatively damage embryonic cellular macromolecules and increase structural

Free radical NOX, PHS Spin trapping agent Inhibitors • Phenylbutylnitrone (PBN) • NADPH oxidase (NOX) Antioxidative • Prostaglandin H Enzyme synthase (PHS) • Enhanced catalase ALCOHOL • Deficient catalase ROS •Embryo • Ethanol (EtOH) • Fetus • Methanol (MeOH)

Oxidatively Damaged DNA Repair Macromolecules • Oxoguanine glycosylase 1 • 8-oxo-2’-deoxyguanosine (8-oxodG) (OGG1) • Protein carbonyls

Structural Deficits Functional Deficits

• Embryopathies in culture • Learning and memory • Teratogenesis in vivo

Figure 1.1: Thesis outline

embryopathies, and whether inhibition of NOX and ROS with the chemical inhibitors

diphenyleneiodonium (DPI) and PBN, respectively, can mitigate these outcomes.

6. To determine whether MeOH can initiate ROS formation and structural embryopathies

through the prostaglandin H synthase (PHS) pathway.

7. To determine if endogenous or exogenously enhanced embryonic catalase can protect

against EtOH and MeOH structural embryopathies in whole embryo culture.

8. To determine if endogenous or exogenously enhanced embryonic catalase can protect

against EtOH teratogenicity and embryonic DNA oxidation in vivo.

9. To determine if deficient embryonic DNA repair, specifically oxoguanine glycosylase 1

(OGG1), involved in base excision repair (BER), is a risk factor for EtOH-initiated

embryonic DNA damage and structural embryopathies upon exposure during the

embryonic period.

10. To determine if deficient fetal DNA repair, specifically OGG1, is a risk factor for EtOH-

initiated behavioural deficits and fetal brain DNA oxidation upon exposure during the fetal

period.

I have demonstrated that NOXs are a notable source of MeOH-initiated ROS, and NOX and ROS inhibition with DPI and PBN, respectively, protect against NOX mRNA and protein upregulation, protein oxidation and structural embryopathies. The antioxidative enzyme catalase, which in the embryo is expressed at only about 5% of maternal levels, is nevertheless important in protecting the embryo from both MeOH and EtOH embryopathies and teratogenesis, as catalase-deficient embryos exhibited increased DNA oxidation (8-oxo-2-deoxyguanosine, 8-

5 oxodG), structural embryopathies and teratogenesis, while exogenous and enhanced endogenous forms of catalase were protective, implicating ROS in the mechanism of developmental toxicity, and demonstrating the important embryoprotective role of the endogenous enzyme. Finally, embryonic DNA repair through the BER pathway initiated by OGG1, which repairs the 8-oxodG lesion, proved to be an important risk factor for both structural and behavioural deficits.

Unrelated to alcohol developmental toxicities, untreated ogg1 knockout mice exhibited learning deficits, constituting the first reported phenotype for these mice, and revealing the potential developmental consequences of endogenously generated DNA damage. EtOH increased 8- oxodG formation in the developing embryo and fetal brain in ogg1 knockout mice in a gene dose-dependent fashion, which correlated with increased structural embryopathies and behavioural deficits, respectively. These results suggest that in utero MeOH and EtOH enhance

NOX-initiated ROS formation, which if not detoxified by embryonic catalase, can oxidatively damage DNA resulting in structural and functional birth defects in the developing embryo or fetus.

1.2 Overview of Development

1.2.1 Introduction

‘Teratology’ is the study of the causes, mechanisms and patterns of abnormal development

(Moore and Persaud, 2007; Wells et al., 2009a). ‘Teratogenesis’ refers to adverse development

in the developing fetus which can result in either in utero death, structural birth defects, or

functional (behavioural) deficits that persist after birth, and is the leading cause of infant

mortality. A ‘teratogen’ is any agent capable of altering embryonic or fetal development

including xenobiotics (drugs and environmental chemicals), infectious agents and irradiation. An

6 important concept in the study of teratogenesis is that certain periods of embryonic or fetal development are more susceptible to disruption by teratogens than others, called ‘windows of susceptibility’ (Wells et al., 2009a) (Fig 1.2). Generally, exposure to xenobiotics during the first

2 weeks of human gestation result in either in utero death or no effect, as the cells have not yet begun to differentiate, and there is substantial cellular redundancy acting as backup protection.

Exposure during the embryonic period when organs are being formed (organogenesis), which in humans span from weeks 4-8 (Moore and Persaud, 2007), results in structural birth defects, the specific type being dependent upon which organ was undergoing critical developmental events at the time of exposure. Lastly, exposure during the fetal period, which in humans spans from the

9th week to birth (Moore and Persaud, 2007), after the organs have been formed and are now

developing function, results in growth retardation, or functional birth defects such as cognitive or

motor deficits, altered biochemical enzymatic activities, cancer, or an array of metabolic diseases

including diabetes, obesity and cardiovascular diseases (Wells et al., 2009a; Wells et al.,

2009b). It was once generally believed that human embryos were completely protected from

environmental chemicals by the mother and the extraembryonic membranes, until 1941 when

cases of rubella virus initiating severe birth defects were reported (Lenz, 1988). Another well-

documented case of teratogenesis began in 1956 with thalidomide, a sedative drug given to

pregnant women, which resulted in thousands of children being born with stunted limbs and

other anomalies (Wells et al., 2009a). Mouse models are frequently used to study teratogenesis,

and although quite different from humans, mice share many similar developmental events (Fig.

1.3), facilitating their use in studies of developmental toxicity. This section will give an overview

of the stages of human and mouse development in order to better understand how birth defects

may occur, and why exposure to teratogens during certain times of development can result in

different teratological outcomes.

Figure 1.2: Human development and critical periods for drug exposure From: (Wells et al., 2009a)

Figure 1.3: Comparison of key developmental events in mice and humans. From: (Ko, 2001) Abbreviations used: dpc, days post conception; dpo, days post ovulation ; wpo, weeks post ovulation

1.2.2 Week 1: The start of human development

Development begins when a sperm fertilizes a egg to form a single totipotent cell, termed a

zygote, a process that takes approximately 24 hours to complete. The zygote undergoes

cleavage, a series of mitotic divisions that produce a greater number of cells called blastomeres,

which get smaller with each division as they are encased in a zona pellucida that restricts space

(Fig. 1.4) (Moore and Persaud, 2007). The spherical collection of these blastomeres is termed a

morula. After the 9-cell stage, the morula undergoes compaction, whereby the blastomeres align

tightly with one another and slightly change their shape forming a compact ball of cells

permitting greater intercellular interactions. Approximately 3 days after fertilization as a result of

compaction, the morula is divided into an inner cell mass and a surrounding outer cell layer

(trophoblast). The trophoblast forms the embryonic part of the placenta, while the inner cell mass

gives rise to the embryo proper. At this stage the morula enters the uterus. Approximately 24

hours after the morula enters the uterus, blastogenesis occurs, which is marked by the formation

of a fluid-filled cavity within the morula (Moore and Persaud, 2007). At this stage, the conceptus

is referred to as a blastocyst (Dudek, 2011; Moore and Persaud, 2007). Six days after

fertilization following dissolution of the zona pellucida, the free-floating blastocyst attaches to

the uterine endometrial epithelium at the embryonic pole, followed by rapid proliferation of the

trophoblast into two layers: an inner cytotrophoblast and an outer highly invasive syncytiotrophoblast. The syncytiotrophoblast invades the uterine wall, permitting implantation

of the blastocyst into the endometrium, which is complete by the end of the 2nd week. (Moore

and Persaud, 2007).

Figure 1.4: Cleavage of the zygote to form a blastocyst From: (Moore and Persaud, 2007)

1.2.3 Week 2: Formation of the embryonic bilaminar disc

While implantation is occurring, the inner cell mass develops into a bilaminar

embryonic disc composed of epiblast and hypoblast which will eventually give rise to the 3 germ layers that form all of the embryonic tissues and organs (Fig. 1.5) (Moore and Persaud,

2007). At this stage, nutrition to the embryo is derived from maternal blood via diffusion through lacunae, which form within the syncytiotrophoblast, the primordia of placental intervillous spaces.

By day 12, lacunar networks have formed which permit greater nutrient and gas exchange between maternal blood and the conceptus (Dudek, 2011). By day 14, the prechordal plate, a thickening of the hypoblast, forms within the bilaminar disc, marking the cranial region of the embryo. The prechordal plate is important in organizing development of the head region. The extraembryonic structures that form during this period include the amniotic cavity, amnion, umbilical vesicle (yolk sac), connecting stalk and the chorionic sac. The embryo is completely implanted in the maternal endometrium by day 10 (Moore and Persaud, 2007). Exposure to teratogens during the first two weeks is most likely to lead to in utero death resulting in spontaneous abortion (Wells et al., 2009b).

1.2.4 Week 3: Germ layer formation and differentiation of early tissues and

organs

The key events that occur in the third week are gastrulation, neurulation, development of

the primitive gut, the somites, body cavities and the cardiovascular system (Moore and Persaud,

2007).

ECTODERM

Bilaminar • Epidermis EPIBLAST embryonic disc • Nervous system (CNS/PNS) • Eye • Inner ear HYPOBLAST • Neural crest cells which form connective tissues of the head BILAMINAR DISC MESODERM

• Skeletal muscles • Blood cells • Lining of blood vessels • Visceral smooth muscular GASTRULATION coats • Serosal linings • Ducts and organs of the reproductive and excretory systems • Most of the cardiovascular system • Connective tissues in the ECTODERM trunk

ENDODERM MESODERM

Trilaminar • Epithelial lining of embryonic ENDODERM disc respiratory, digestive, reproductive and urinary systems • Glandular cells of associated organs (liver, TRILAMINAR DISC pancreas, thymus, thyroid and parathyroid glands)

Figure 1.5: Gastrulation and the tissues derived from the three germ layers Modified from: (Moore and Persaud, 2007) Abbreviations: CNS, central nervous system; PNS, peripheral nervous system

Gastrulation is the conversion of the bilaminar embryonic disc (hypoblast, epiblast) to a trilaminar embryonic disc to form the 3 germ layers – the endoderm, the mesoderm, and the ectoderm, which occurs by day 21 of development (Dudek, 2011). The beginning of gastrulation is marked by appearance of the primitive streak on the caudal end surface of the epiblast. The primitive streak is a thickening of the epiblast due to proliferation and migration of epiblast cells into the caudal midline. At the site of the primitive streak, cells detach from the epiblast layer and ingress, replacing the cells in the hypoblast with endoderm and also creating the mesoderm layer by migrating into the interval between the piblast and the hypoblast. At the end of gastrulation in week 5, the remaining cells in the epiblast form the ectoderm. Each of the germ layers formed during gastrulation will give rise to specific tissues that contribute to the development of the organs (Moore and Persaud, 2007).

Neurulation is the process of neural tube formation and closure (Dudek, 2011). The central nervous system, namely the brain and spinal cord, are formed from neuroectoderm of the neural plate, while the neural crest cells and the neural tube both contribute to the development of the peripheral nervous system (Dudek, 2011) (Table 1.1, Fig 1.6).

The notochord serves as the primary axis around which the axial skeleton will form, and induces the thickening of the overlying embryonic ectoderm to form the neural plate, the central nervous system primordium (Fig. 1.7) (Moore and Persaud, 2007). Some mesenchymal cells migrate cranially on either side of the notochord process, and meet to form the cardiogenic area where the heart will begin to develop at the end of the third week. On approximately day 18, a longitudinal median neural groove is formed by invagination of the neural plate along its central axis, with neural folds on either side (Moore and Persaud, 2007). By the end of the third week the neural folds have started to come together and fuse, converting the neural plate into a

Table 1.1: Neural germ layer derivatives

Neuroectoderm All neurons within brain and spinal cord Retina, iris epithelium, ciliary body epithelium, optic nerve (CN II), optic chiasm, optic tract, dilator and sphincter pupillae muscles Astrocytes, oligodendrocytes, ependymocytes, tanycytes, choroid plexus cells Neurohypophysis Pineal gland Neural Crest Cranial neural crest cells: Pharyngeal arch skeletal and connective tissue components Bones of neurocranium Pia and arachnoid Parafollicular cells of the thyroid Aoricopulmonary septum Odontoblasts (dentin of teeth) Sensory ganglia of CN V, CN VII, CN IX, CN X Ciliary (CN III), pterygopalatin (CN VII), submandibular (CN VII), and otic (CN IX) parasympathetic ganglia Trunk neural crest cells: Melanotyes Schwann cells Chromaffin cells of adrenal medulla Dorsal root ganglia Sympathetic chain ganglia Prevertebral sympathetic ganglia Enteric parasympathetic ganglia of the gut (Meisssner and Auerbach; CN X) Abdominal/pelvic cavity, parasympathetic ganglia From: (Dudek, 2011) Abbreviations used: CN, cranial nerve

Figure 1.6: Beginning of neurulation From: (Moore and Persaud, 2007)

Figure 1.7: Transverse section through an embryo illustrating neurulation From: (Moore and Persaud, 2007)

17 neural tube (Moore and Persaud, 2007).

As the neural folds meet, the neural tube separates from the surface ectoderm, and cells from the neural crest (NCC) undergo an epithelial to mesenchymal transition, losing their attachments with neighbouring cells, and migrate throughout both the trunk and cranial regions of the embryo, eventually differentiating into a wide variety of structures in the embryo (Table

1.1) (Dudek, 2011). Exposure to teratogens during this time period can result in abnormalities in neural tube formation leading to neural tube defects such as exencephaly, anencephaly and spina bifida (Moore and Persaud, 2007). NCCs are highly sensitive to EtOH toxicity as demonstrated in NCC culture studies in vitro (Chen et al., 2000; Chen and Sulik, 1996). Given the fate of these cells in craniofacial development (Cordero et al., 2011), it can be understood why children exposed in utero to EtOH during the time of NCC migration develop the distinct pattern of craniofacial abnormalities commonly observed in FASD.

Somites are paired cuboidal structures that develop from differentiated and condensed paraxial mesoderm, located on either side of the neural tube, the first appearing at the end of the

3rd week, developing in a craniocaudal fashion (Fig. 1.7) (Moore and Persaud, 2007). Somites

further differentiate into: (i) sclerotome, which gives rise to the cartilage and bone of the

vertebral column; (ii) myotome, which gives rise to epimeric and hypomeric muscles; and, (iii) dermatome, which gives rise to dermis and the hypodermis (Dudek, 2011).

Embryonic body cavities start out as dispersed coelomic spaces in the lateral mesoderm

(Moore and Persaud, 2007). These spaces coalesce to form a horseshoe-shaped cavity called the intraembryonic coelem, which divides the lateral mesoderm into two distinct layers: a somatic or parietal layer which will contribute to the embryonic body wall, and a splanchnic or visceral

18 layer which will contribute to the embryonic gut. The intraembryonic coelem is later divided into pericardial, pleural and peritoneal cavities (Moore and Persaud, 2007).

Development of the cardiovascular system begins at the start of the third week, with angiogenesis and vasculogenesis beginning in the extraembryonic mesoderm, a process required to meet the nutrient and oxygen requirements of the developing embryo (Moore and Persaud,

2007). The heart starts out as a primordial tube, which joins with blood vessels to form a primordial cardiovascular system. Chorionic villi branch out and develop arteriocapillary networks, which eventually connect with the embryonic heart (Dudek, 2011). Oxygen and nutrients in the maternal blood in the intervillous spaces diffuse through the walls of the villi and enter the capillary networks. By the end of the 3rd week, the heart begins to beat and blood is

circulating, making the cardiovascular system the first functioning organ system in the

developing embryo (Moore and Persaud, 2007).

1.2.5 Weeks 4-8: Organogenesis

This period is considered the most sensitive with respect to structural teratogenesis, as

most physical internal and external structures in the embryo form during this time (Moore and

Persaud, 2007; Wells et al., 2009b). By the end of organogenesis, the primordia of all structures

in the embryos have developed but have minimal function except for the cardiovascular system.

Due to enhanced growth at the cranial and caudal ends of the embryo, the sides of the bilaminar

disc cannot keep up, resulting in folding of the embryo, an event that turns the flat trilaminar

embryonic disc into a cylindrical C-shaped embryo (Moore and Persaud, 2007). Due to the high

number of structures being formed during this period, exposure to xenobiotics during

19 organogenesis can result in major structural malformations depending upon which organ system was undergoing critical developmental events when teratogen exposure occurred, and the specific mechanism of action of the particular xenobiotic to which the embryo was exposed

(Wells et al., 2009b). Table 1.2 highlights the key events occurring from weeks 3 – 8 of embryonic development.

1.2.6 Week 9 – Birth: Fetal Period

At this stage, the embryo is now called a fetus, denoted by the completed formation of

primordial organs, which now begin to develop functionality, and the rapid growth that takes

place during this period resulting in a marked increase in fetal size and weight (Moore and

Persaud, 2007). Since all of the organs including the central nervous system have formed during

organogenesis and are now developing functionality, teratogen exposure during the fetal period

is less likely to result in structural defects, and more likely to result in growth retardation, or

functional or behavioural deficits that may persist into adulthood (Wells et al., 2009b).

Table 1.2: Key events occurring during the embryonic period Week Key Events Appearance Actual Size Primitive streak appears Notochord appears Angiogenesis in yolk sac 3 Gastrulation occurs

Cranial neural folds elevated Day 21 Early somites form Neural tube forms Rostral/caudal neuropores close Otic/lens placode present Head and tail folding begins 4 Lateral folding begins Upper/lower limb buds present Pharyngeal arches appear Day 28 Heart is distinct Upper limb is paddle shaped

Optic cup is present 5 Hand plate forms

Lower limb is paddle shaped Day 35 Foot plate forms Auricular hillocks appear

6 Digital rays appear in hand plate

Cerebral vesicles prominent Day 42 Digital rays appear in foot plate Notches in hand plate form Both limbs extend ventrally 7 Elbow region is apparent

Notches in foot plate form Day 48 Fingers are formed Head is rounded Umbilical herniation distinct 8 Toes are formed Tail disappears Day 56 Distinct human appearance

Modified from: (Dudek and Fix, 1998)

1.3 Mechanisms of Teratogenesis

Xenobiotic exposure to the embryo or fetus during pregnancy can typically alter development through one of three ways: (i) reversible binding of the parent compound or a stable metabolite to a receptor; (ii) bioactivation to an electrophilic reactive intermediate that can covalently bind to cellular macromolecules such as DNA and proteins; or, (iii) bioactivation to a free radical reactive intermediate that can initiate the formation ROS, which can alter signal transduction pathways, or oxidatively damage cellular macromolecules (Fig. 1.8, Table 1.3)

(Wells et al., 2009a; Wells et al., 2009b). The outcome of xenobiotic exposure is determined by the balance of toxification pathways and detoxification pathways, and the ability to repair the resulting damage. If pathways are kept in a balance, no resultant damage occurs, but if toxification is greater than detoxification, damage may occur, possibly resulting in teratogenesis if the damage is not repaired (Wells et al., 2009b; Wells et al., 2010).

1.3.1 Receptor-mediated mechanisms

Receptor-mediated toxicity occurs when a parent compound or its stable active metabolite(s) bind to either embryonic cell-surface or nuclear receptors in place of the endogenous ligand, resulting in aberrant gene expression and protein expression changes, or bind to embryonic enzymes resulting in either activation/inactivation of enzymatic pathways (Wells et al., 2009a). Binding of a xenobiotic to a cell-surface receptor may directly initiate downstream signal transduction pathways if it is an agonist for that receptor, or may conversely inhibit downstream signal transduction pathways if the xenobiotic is an antagonist (Wells et al., 2009a;

Wells et al., 2009b).

Xenobiotic

Bioactivation

Rece ptor Reactive Intermediate

•Covalent binding (electrophi le) •Oxidative damage (free radica l)

Cellular TERATOGENESIS Macromolecules

Figure 1.8: Receptor- versus reactive intermediate-mediated mechanisms of teratogenesis. The mechanism involving receptor-mediated interactions are reversible, and the other mechanism results in drugs being converted to reactive intermediates which react irreversibly with cellular macromolecules.

From: (Wells et al., 2009a)

Table 1.3: Characteristics of xenobiotic toxicity initiated by reactive intermediates compared with reversible, receptor-mediated interactions Mechanism of tissue interaction Characteristic Reactive Intermediate Receptor-mediated  Reactive intermediary  Parent compound and/or a metabolite (highly unstable) stable, major metabolite  Electrophile Initiating species  Free radical

 Often a minor metabolite

amounting only to 1-10% of total xenobiotic/metabolites  Multiple sites within  Specific receptor on one type of different cellular macromolecule (usually a Molecular target macromolecules (DNA, protein) protein, lipid and carbohydrate)  Irreversible  Reversible binding  Covalent binding Target interaction (arylation/alkylation)  Oxidation Duration of target  Cumulative  Transient interaction  Unrelated to therapeutic  Generally an extension of the Toxic effectsa effectb therapeutic effect  Toxicity can occur at  Toxicity occurs when therapeutic therapeutic plasma drug or safe plasma concentration are Toxic dose/ concentrations or “safe” exceeded concentration concentrations of environmental chemicals  Toxicity occurs well after the time of the peak plasma xenobiotic concentration, and usually after the  Toxicity usually increases with xenobiotic is no longer rising plasma xenobiotic Onset of toxicity detectable in plasma or concentration, and decreases urine with or shortly after declining  Depending upon both the concentrations xenobiotic and the toxicity, this delay can be hours, days, months or years a – Effect in this case refers to the effect of therapeutic drugs, and is not relevant to environmental chemicals b – There are some exceptions, such as alkylating anticancer drugs, where drug toxicity does result from the same mechanisms by which tumor cells are killed

From: (Wells and Winn, 1996)

An example of a xenobiotic binding to a cell-surface receptor leading to teratogenesis is the interaction of a class of drugs called ‘sartans’ with angiotensin type 2 receptors thereby interfering with renin-angiotensin signaling in the developing fetus (Alwan et al., 2005).

Angiotensin (AT-I) is formed from angiotensinogen (AT) by renin, which is released from the kidney. AT-I is rapidly converted to angiotensin II (A-II) by angiotensin converting enzyme

(ACE), which also degrades bradykinin, a vasodilator. A-II, through binding to A-II type 1

(AT1) and type 2 (AT2) receptors promotes the release of aldosterone, induces vasoconstriction, and increases sodium and fluid retention to increase blood pressure (Fig. 1.9). Sartans are AT1 receptor inhibitors used in the treatment of hypertension and heart failure. AT1 and AT2 receptors are both expressed throughout mature fetal tissues including the kidney (Norwood et al., 1997; Sequeira Lopez and Gomez, 2004; Tufro-McReddie et al., 1993), and exposure to sartans during pregnancy have resulted in fetal renal damage, growth retardation, aberrant heart development and in utero death when administered during the 2nd or 3rd trimester of pregnancy.

Indeed, newborn AT-I knockout mice exhibit delayed glomerular maturation and lesions on the

renal cortex (Kim et al., 1995; Niimura et al., 1995).

Xenobiotic binding to a nuclear receptor is exemplified by the environmental toxin

2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD), which binds to the aryl hydrocarbon receptor

(AHR), a ligand-activated transcription factor that regulates gene transcription by binding to enhancer DNA sequences, resulting in altered gene expression and teratogenesis, depending upon which genes are altered, and when the disturbance occurred (Wells et al., 2009a). The AHR is an intracellular protein with broad substrate specificity (Denison and Nagy, 2003). A wide range of compounds can activate AHR, the most potent being TCDD. Under normal conditions,

AHR is found in the cytosol bound to various chaperone proteins (Fig. 1.10). Once a ligand binds AHR, it dissociates from its chaperone proteins and translocates to the nucleus. Once in the

Figure 1.9: Mechanism involving a cell-surface receptor interaction. Renin-angiotensin signaling pathway and AT1 receptor inhibition by sartan drugs. Modified from: (Alwan et al., 2005)

Figure 1.10: Mechanism involving a nuclear receptor. Aryl Hydrocarbon Receptor (AHR) pathway Modified from: (Wells et al., 2009a)

27 nucleus, the ligand-bound AHR dimerizes with the aryl hydrocarbon receptor nuclear translocator (ARNT). This complex binds to specific DNA elements termed AH responsive elements (AHREs), which are located typically in the 5’-flanking region of the target genes.

Binding to AHREs recruits co-activator proteins that either enhance or inhibit gene transcription, resulting in a modified biological response (Wells et al., 2009a). Exposure to AHR ligands during development aberrantly activates AHR, resulting in an abnormal pattern of AHR- regulated gene transcription and subsequent teratogenesis (Wells et al., 2009a). The role of AHR in developmental toxicity is evidenced by Ahr null mice, which exhibit a significantly lower incidence of developmental abnormalities than their wild-type littermates (Peters et al., 1999).

AHR regulates CYP1A1 (Lin et al., 2003), which can convert environmental polycyclic aromatic hydrocarbons (PAHs) to highly mutagenic, carcinogenic or teratogenic metabolites (Uppstad et al., 2010). These metabolites can further induce CYP1A1 thereby stimulating their own metabolism resulting in metabolites that can lead to teratogenicity, independent of AHR activation. TCDD does not form a reactive metabolite, so induction of metabolizing enzyme to stimulate its own metabolism cannot account for the toxicity observed (Wells et al., 2009a).

Indeed, it is the parent compound that mediates toxicity through AHR activation. AHR signaling regulates the expression of hundreds of genes both positively and negatively, and is also involved in cross-talk with several pathways involving Wnt (Chesire et al., 2004; Mathew et al., 2009), estrogen receptor (ER)(Khan et al., 2006; Matthews and Gustafsson, 2006; Ruegg et al., 2008), epidermal growth factor receptor (EGFR) and transforming growth factor beta (TGFβ) (Chang et al., 2007; Wolff et al., 2001), among others. Upon TCDD-initiated AHR activation and subsequent CYP1A1 upregulation, CYP1A1 is able to metabolize endogenous substrates or xenobiotics resulting in enhanced covalent binding and/or oxidative stress (Shertzer et al., 1998;

Wells et al., 2009b). In developmental models, in utero TCDD exposure results in teratogenesis

28 in rodents (Couture et al., 1990; Thackaberry et al., 2005; Vezina et al., 2009) and humans

(Alaluusua and Lukinmaa, 2006). Vitamin E, an antioxidant, has demonstrated partial protection upon in utero exposure to TCDD (Hassoun et al., 1997).

1.3.2 Reactive intermediate-mediated mechanisms

Reactive intermediate-mediated mechanisms involve the conversion or bioactivation of the parent compound to either an electrophilic reactive intermediate that can covalently bind to cellular macromolecules, or to a free radical intermediate which can initiate oxidative stress and the formation of ROS (Wells et al., 2009a; Wells et al., 2009b). Toxicity depends largely upon the balance of bioactivating pathways compared to detoxifying pathways, and the extent of macromolecular repair (Fig 1.11).

1.3.2.1 Electrophilic Reactive Intermediates

Electrophilic reactive intermediates such as epoxides contain electron-deficient

(electrophilic) centers, which are slightly positively charged moieties that can react with electron-rich (nucleophilic) groups contained within proteins or DNA, thereby irreversibly covalently binding, forming an adduct (Wells et al., 2009b). Bioactivation of a drug to an electrophilic reactive intermediate is usually catalyzed by cytochrome P450 (CYP) enzymes, resulting in teratogenicity (Juchau, 1989; Juchau et al., 1998). If the adduct is formed with a developmentally important protein or gene, this may result in altered development or potential death of the developing fetus. The embryo is particularly susceptible to the formation of reactive intermediates given the low levels of detoxifying enzymes such as epoxide hydrolase, catalase

(Abramov and Wells, 2011a; Winn and Wells, 1999), and glutathione (GSH) S-transferases

(GSTs). Benzo[a]pyrene (B[a]P) is a good example of a xenobiotic that forms an electrophilic

. Figure 1.11: Reactive intermediate-mediated mechanisms of developmental toxicity. Xenobiotics can be bioactivated to electrophiles and/or free radical reactive-intermediates. Free radicals can initiate the formation of reactive oxygen species (ROS) which can alter signal transduction and/or oxidatively damage cellular macromolecules. Toxicity depends on the balance of the formation of these reactive intermediates versus metabolism/elimination of the parent compound, detoxification, cytoprotection, and macromolecular repair. Abbreviations used: BRCA1, breast cancer 1; CSB, Cockayne syndrome B; CYP, cytochromes P450; EH, epoxide hydrolase; FMO, flavin containing monooxygenases; G6PD, glucose-6-phosphate dehydrogenase; GPx, glutathione peroxidase; GSH, glutathione; GST, glutathione-S-transferase; LPO, lipoxygenase; OGG1, oxoguanine glycosylase 1; PHS, prostaglandin H synthase; SOD, superoxide dismutase; SULT, sulfotransferase; Trx, thioredoxin; UGT, UDP-glucuronosyl transferase. Modified from: (Wells et al., 2009b; Wells et al., 2010)

reactive epoxide intermediate upon CYP1A1-catalyzed bioactivation of the parent compound

(Jernstrom et al., 1984).

1.3.2.1 Free Radical Reactive Intermediates

The mechanism of teratogenicity for several teratogens including phenytoin,

methamphetamine, cyclophosphamide and thalidomide has implicated a role for the generation

of a free radical intermediate (Fantel, 1996; Wells and Winn, 1996). Given the low CYP

expression in the embryo, particularly in mice, it is unlikely that these enzymes are responsible

for bioactivation of these xenobiotics to free radical intermediates (Wells et al., 2009b). The

embryo does however express high levels of PHSs and lipoxygenases (LPOs) that can bioactivate xenobiotics to free radical intermediates (Winn and Wells, 1997; Yu and Wells,

1995). These free radicals can go on to directly or indirectly react with O2 to form ROS,

- including O2• , •OH or H2O2, which can oxidatively damage cellular macromolecules or alter

embryonic signal transduction (Wells et al., 2009b).

1.4 Reactive Oxygen Species (ROS) and antioxidative defense

mechanisms

Molecular oxygen (O2) is essential for the survival of aerobic organisms which derive

energy from oxidative phosphorylation, a process whereby the redox energy of the mitochondrial

electron transport chain is converted to the high energy phosphate bond of adenosine

triphosphate (ATP), using O2 as the final electron acceptor (Murphy, 2009). This final step

involves a series of electron transfer reactions that ultimately results in a 4-electron reduction of

O2. During this process however, partially reduced metabolites of O2 can be formed, which are

31 highly reactive, and are collectively termed ROS (Fig. 1.12). The three most common ROS

- produced in biological organisms include O2• , •OH, and H2O2, which will be discussed in

further detail below. The 2 former species are considered free radicals, as they contain an

unpaired electron, while the latter does not, and so it is considered to be a non-radical species.

Free radicals contain an unpaired electron in their valence shell, and contain at least one oxygen

atom that can accept, donate or transfer electrons (Halliwell and Gutteridge, 2007). Their high

reactivity is due to the presence of one or more unpaired electrons, which can react with cellular

macromolecules such as DNA, proteins and lipids to increase their stability, and in doing so,

oxidatively damage these cellular macromolecules. The need for O2 to survive despite its ability

to create potentially harmful byproducts underlies the ‘oxygen paradox’ (Novo and Parola,

2008). The apparent paradox of oxygen, and the resulting ROS and reactive nitrogen species

(RNS), being required for cellular function while also being deleterious is a result of the

concentration of these chemical species. A good example of this is nitric oxide (NO•), which

when produced at low concentrations by constitutive nitric oxide synthase (NOS) in vascular

endothelial cells, acts as a signaling molecule that mediates vasodilation (Wilcox et al., 1992).

However, NO• acts as a highly toxic oxidant in the immune system when produced in high

concentrations by the inducible NOS (iNOS) in macrophages to kill invading microorganisms

(MacMicking et al., 1995). As a consequence, organisms have evolved antioxidative defense

pathways to carefully control the amount of ROS present at any given time, as well as to use

these ROS and RNS under physiological conditions for signal transduction, to alter gene

transcription, and affect cellular growth and differentiation (Finkel, 1998; Rhee, 1999; Sauer et

al., 2001). The major reactions of ROS that will be described are outlined in Fig. 1.12.

OXIDATIVE DAMAGE DNA, Protein, Lipids

 HOCl Cl OH  HYDROXYL HYPOCHLOROUS RADICAL ACID

FENTON 2 2 REACTION Fe ,Cu

e  e O  O •SOD (109 M-1s-1) H O 2 2 • Spontaneous (105 M-1s-1) 2 2 MOLECULAR SUPEROXIDE HYDROGEN OXYGEN ANION PEROXIDE

• CATALASE NO • • GPx

 H O  O ONOO 2 2 PEROXYNITRITE

Figure 1.12: Generation of ROS in biological systems Modified from: (Halliwell and Gutteridge, 2007; Novo and Parola, 2008) Abbreviations used: SOD, superoxide dismutase; GPx, glutathione peroxidase; e-, electron

•- 1.4.1 Superoxide anion (O2 )

- O2• is formed through a one-electron reduction of O2 (Fig. 1.12) (Halliwell and

Gutteridge, 2007). It is relatively unreactive, although it can react with nitric oxide (NO•) to

- - yield the highly reactive peroxynitrite (ONOO ). O2• can be converted into H2O2 enzymatically

by superoxide dismutase (SOD), or nonenzymatically (Cadenas and Davies, 2000; Halliwell and

Gutteridge, 2007).

1.4.2 Hydrogen peroxide (H2O2)

H2O2 is a non-radical reactive oxygen species, although still a pro-oxidant, and is formed

- from the two-electron reduction of O2 via the dismutation of O2• , which can occur either

enzymatically catalyzed by SOD, nonenzymatically, or from the direct reduction of O2 (Fig.

1.12). H2O2 can easily cross biological membranes, and can be removed either by catalase or

- glutathione peroxidase (GPx). H2O2 can react with O2• in the presence of divalent metal ions

(Fe2+, Cu2+) to produce the highly reactive hydroxyl radical, a reaction which is termed the

‘Fenton Reaction’ (Halliwell and Gutteridge, 2007). This reaction is particularly useful in phagocytic cells which use it to form the highly reactive hypochlorous acid (HOCl) during the oxidative burst to kill invading microorganisms (Halliwell and Gutteridge, 2007; Lambeth,

2004).

1.4.3 Hydroxyl radical (•OH)

•OH results from a three-electron reduction of O2, formed by the Fenton reaction (Fig.

1.12) (Thannickal and Fanburg, 2000; Winterbourn, 1995) or by the decomposition of ONOO-.

It has a very short half-life (10-9 s) and very high reactivity (Halliwell and Gutteridge, 2007). As

34 a consequence, it cannot diffuse from the site of generation, and can very quickly react with and damage surrounding cellular macromolecules including DNA, proteins and lipids. Oxidation of

DNA can result in the 8-oxodG lesion, which has been shown to be not only a marker of oxidative damage, but also a developmentally pathogenic lesion (McCallum et al., 2011c; Wells et al., 2009b; Wong et al., 2008). Oxidative damage to proteins can result in a loss of activity or complete denaturation, while oxidative damage to lipid can initiate lipid peroxidation

(Gutteridge, 1984).

1.4.4 Endogenous sources of ROS

A major source of endogenous ROS production is the mitochondrial electron transport

chain (ETC), which undergoes oxidative phosphorylation to create high-energy ATP, using O2 as the final electron acceptor for the cytochrome c oxidase, the enzyme that catalyzes the four- electron reduction of O2 to H2O. ROS can be formed during these electron transfer reactions,

resulting in the formation of highly reactive O2 metabolites (Thannickal and Fanburg, 2000).

Approximately 5% of electrons that go through complex I and complex III of the ETC are

- - diverted into O2• , however, the mitochondria contains high concentrations of SOD to keep O2• concentrations very low (Tyler, 1975).

The endoplasmic reticulum contains CYP enzymes that use O2 to oxidize a variety of

endogenous substrates including unsaturated fatty acids, thereby reducing molecular O2 to produce O2•- or H2O2, along with a substrate free radical if the oxidation reaction is not complete

Additionally, CYP enzymes are a source of iron that can participate in the Fenton reaction.

(Halliwell and Gutteridge, 2007).

Peroxisomes contain peroxisomal oxidases such as urate oxidases, glycolate oxidases, D- amino oxidases, L-α-hydroxyacid oxidases and fatty acyl-CoA oxidases, all of which can generate H2O2 when oxidizing a variety of substrates in peroxidative reactions (Boveris et al.,

1972; Rojkind et al., 2002).

Soluble enzymes such as xanthine oxidase and aldehydes oxidase can generate ROS

during catalytic cycling (Freeman and Crapo, 1982).

Auto-oxidation reactions including small molecules such as dopamine have been

implicated in the pathogenesis of neurodegenerative conditions such Parkinson’s disease

(Freeman and Crapo, 1982; Offen et al., 1997; Offen et al., 1996; Yoshikawa et al., 1994).

ROS-forming NADPH oxidases (NOXs) are found in phagocytic cells, as well as in non-

phagocytic cells. In phagocytic cells, NOX is comprised of 2 membrane bound subunits

gp91phox/NOX2 and p22phox, which together make up the flavocytochrome b558 (cyt b558),

so called because of the heme absorbance peak at 558 nm in the reduced state, along with four

cytosolic regulatory subunits p40phox, p47phox, p67phox, and the GTPase Rac1/2 (Fig 1.13).

NOX is dormant in non-activated phagocytes, however upon phagocytic cell stimulation, the

regulatory subunits are recruited to the plasma membrane where they associate with the cyt b558

complex, thereby activating NOX activity and stimulating the production of O2•- during

phagocytosis according to the following reaction (Lambeth, 2004; Vignais, 2002):

   NADPH  2O2  2O2  NADP  H

•- The O2 is rapidly converted to H2O2 by dismutation inside the cell, and in the presence

- of iron, O2• and H2O2 can react to form •OH via the Fenton Reaction. In non-phagocytic cells, the structure of NOXs is similar although the main gp91phox/NOX2 is replaced with a NOX

Figure 1.13: Reactive Oxygen Species generation by assembly of the NADPH Oxidase (NOX) regulatory subunits in phagocytes. The + NOX complex takes an electron from NADPH and passes it onto O2 to create O2•- and NADP . Modified from: (Lambeth, 2004)

isoform (NOX1, NOX3, NOX4, NOX5, dual oxidases (DUOX) 1/2), which are differentially regulated. In these cells, NOXs are constitutively active, and are a major source of ROS-initiated signal transduction capable of responding to a wide variety of stimuli and cellular conditions

(Brown and Griendling, 2009; Lambeth, 2004). NOXs will be discussed in further detail in section 1.7

1.4.5 Exogenous sources of ROS

ROS can be formed exogenously by exposure to ionizing radiation, xenobiotics or tissue injury (Wells et al., 2009b). Ionizing radiation increases cellular energy, which is absorbed by the cell water, resulting in breakage of the covalent oxygen-hydrogen bonds, thereby forming

•OH (Halliwell and Gutteridge, 2007). Xenobiotics can initiate ROS in several ways, including:

(1) uncoupling of oxidative phosphorylation via disruption of the mitochondrial electron transport chain; (2) bioactivation by CYPs, LPOs and PHSs to form free radical intermediates which can react with O2 to generate ROS; (3) redox cycling of hydroxylated metabolites, whereby a single molecule of xenobiotic can result in amplified ROS production, as seen with paraquat and menadione (Halliwell and Gutteridge, 2007); and, (4) induction or activation of

ROS-producing NOXs (Dong et al., 2010).

1.4.6 Antioxidative defenses

Although O2 can be poisonous, aerobic organisms are able to survive in the presence of

O2 because of evolved enzymatic and non-enzymatic antioxidative defense mechanisms to detoxify ROS (Halliwell and Gutteridge, 2007). The term ‘antioxidant’ is quite broad, including any one of the following: (1) compounds that catalytically remove ROS such as SOD, catalase or

GPx; (2) compounds that reduce the formation of ROS such as mitochondrial uncoupling

38 proteins, or proteins that decrease the availability of pro-oxidants such as metal ions (iron, copper) which include hemoglobin, transferrins or albumin; (3) proteins that protect cellular macromolecules from oxidation, such as chaperone proteins; and, (4) ‘sacrificial’ compounds such as GSH, ascorbate or α-tocopherol, which are oxidized in lieu of other more important cellular macromolecules (Halliwell and Gutteridge, 2007; Novo and Parola, 2008). The developing embryo is particularly susceptible to the deleterious effects of ROS since it has very low levels of most antioxidative enzymes compared to adult tissues (Ozolins et al., 1996; Wells et al., 2009b; Wells and Winn, 1996). Antioxidants are not completely effective, as oxidative damage to cellular macromolecules still occurs, presumably because ROS are also used as second messengers in signal transduction pathways, so the challenge is to evolve a system that will minimize oxidative damage but still allow for physiologically necessary signal transduction to occur (Halliwell and Gutteridge, 2007).

SOD (EC 1.15.1.1) converts superoxide to hydrogen peroxide and oxygen via a dismutation reaction (Fig.1.14). Copper-zinc SOD (CuZnSOD) is located in the cytosol, lysosomes, the nucleus, the space between inner and outer mitochondrial membranes, and in peroxisomes (Fridovich, 1995; Okado-Matsumoto and Fridovich, 2001). The Zn2+ does not contribute to the catalytic cycle, but instead functions to stabilize the enzyme (Halliwell and

Gutteridge, 2007). Manganese SOD (MnSOD), also referred to as SOD2, is located primarily in

- the mitochondria (Fridovich, 1995). Dismutation of O2• can occur non-enzymatically as well, although the rate constant for this reaction is about 5 x 105 M-1s-1 at pH 7.0, and is pH-dependent.

The dismutation catalyzed by SOD occurs at a rate of at least 1.5 x 109 M-1s-1, and is independent

- of pH between pH 5.3 to 9.5. Furthermore, non-enzymatic dismutation of O2• would require the

- collision of O2• and H2O2, which does not occur often given their low intracellular

-9 -10 - concentrations (10 -10 M). It is much more likely that O2• would collide with SOD given its

1) Superoxide dismutase (SOD) isoforms

2   Enzyme  Cu  O2  Enzyme  Cu  O2     Enzyme  Cu  O2  2H  Enzyme  Cu  H 2O2

  Net 2O2  2H  H 2O2  O2 2) Catalase

Catalase  Fe(III)  H 2O2  CompoundI  H 2O

CompoundI  H 2O2  Catalase  Fe(III)  H 2O  O2

Net 2H 2O2  2H 2O  O2 3) Glutathione peroxidase (GPx) 2H O  2GSH  2H O  2O  GSSG 2 2 2 2 2NADPH 2NADP  2GSH 4) Thioredoxin (Trx)  TrxRed  Trx  S2  NADPH  H  Trx  SH 2  NADPH spon taneous Trx  SH 2  Pr otein  S2  Trx  S2  Pr otein  SH 2 5) Glutaredoxin (Grx)

Grx  S2  2GSH  Grx  SH 2  GSSG

Grx  SH 2  Pr otein  S  SG  Grx  S  SG  Pr otein  SH 2 SH SH

Grx  S  SG  GSH  Grx  SH 2  GSSG SH

Grx  SH 2  Pr otein  S2  Grx  S2  Pr otein  SH 2 GSSG  NADPH  H  GSSGRed 2GSH  NADP

Figure 1.14: Reactions of antioxidative enzymes and antioxidants. Modified from: (Halliwell and Gutteridge, 2007; Novo and Parola, 2008)

40 intracellular concentration of 10-5 M (Halliwell and Gutteridge, 2007). The dismutation of superoxide catalyzed by SOD produces H2O2, which must be detoxified by either catalase or peroxidase enzymes; otherwise in the presence of iron, H2O2 can undergo the Fenton reaction to produce •OH radicals. Catalase (EC 1.11.1.6), which is located primarily in peroxisomes, catalyses the direct decomposition of H2O2 to O2 while peroxidase enzymes such as glutathione peroxidase (GPx) (EC 1.11.1.9), remove H2O2 by using it to oxidize another substrate (Halliwell and Gutteridge, 2007) (Fig. 1.14).

Catalase, like SOD, catalyzes a dismutation reaction where one H2O2 is reduced to H2O and the other is oxidized to O2. Catalase is also capable of catalyzing a peroxidase reaction, which is fairly slow. An example of this is the oxidation of ethanol to acetaldehyde (Kirkman and Gaetani, 2007). Catalase will be discussed further in section 1.8.

As mentioned above, GPx removes H2O2 by using it to oxidize another substrate.

Specifically, GPx couples the reduction of H2O2 to H2O with the oxidation of reduced GSH, a thiol-containing tripeptide. In doing so, GSH is oxidized to GSH disulfide (GSSG). Glutathione reductase (GSH Rd) is responsible for the regeneration of GSH by reducing GSSG, coupling it with the oxidation of NADPH as a reducing cofactor creating NADP+. NADP+ is converted back to NADPH by glucose-6-phosphate dehydrogenase (G6PD) (Halliwell and Gutteridge, 2007).

Thioredoxins (Trx) (EC 1.8.1.9) and glutaredoxins (Grx) (EC 1.20.4.1) are proteins that contain reduced thiol (R-SH) groups that are oxidized and converted to a disulfide (R-S-S-R) bond when reducing a target compound (Fig. 1.14). Trx has been shown to catalyze the reduction of H2O2 thereby minimizing oxidative stress (Kang et al., 1998). Oxidized Trx (R-S-S-R) is converted back to reduced Trx (R-SH) by thioredoxin reductases (Arner and Holmgren, 2000).

Grx and Trx have overlapping functions in the repair of oxidatively damaged amino acid side

41 chains in proteins by donating an electron thereby reducing itself. The Trx ‘system’ is composed of Trx, Trx reductase, and NADPH, while the Grx system is comprised of glutaredoxin, GSH,

GSH reductase and NADPH (Holmgren, 1989).

The peroxiredoxins (Prx) are a class of H2O2 scavengers that are located primarily in the cytosol, but have also been identified in mitochondria, peroxisomes and nucleus. They are homodimers comprised of 3 classes: typical 2-Cys Prx, atypical 2-CysPrx, and 1-Cys Prx, all of which share the same catalytic antioxidative mechanism whereby an active site N-terminal cysteine on one homodimer is oxidized by H2O2 to produce water and form an intermolecular disulfide bond with the C-terminal cys of the other homodimer component to form an intermolecular disulfide. The fully reduced enzyme is then reconstituted by Trx reduction

(Hofmann et al., 2002; Wood et al., 2003).

Small molecules such as GSH, and vitamins C and E can also act as antioxidants.

GSH is present in the cell in the range of about 3 – 10 mM (Frei, 1994). In its reduced state, it is capable of donating a reducing equivalent to other molecules or compounds, such as ROS, thereby stabilizing them.

Vitamin C, or ascorbic acid, is a water soluble antioxidant found in the cytosol at concentrations of 2 – 5 mM (Frei, 1994; Padayatty et al., 2003). It is an electron donor, which accounts for all of its antioxidative actions.

Vitamin E, or α-tocopherol is a lipid-soluble antioxidant found in the plasma membrane capable of donating electrons to peroxyl radicals thereby preventing the propagation of lipid peroxidation (Reiter, 1995).

It is important to keep in mind that the actions of these antioxidants can be complex, as is the case with vitamin E, whereby it has numerous alternative activities that are unrelated to its antioxidant capability. While protective at lower doses (Chen et al., 2009a), vitamin E at higher doses has been shown to oxidatively damage DNA (Chen and Wells, 2006) and enhance embryopathies initiated by in utero exposure to phenytoin (Wells et al., 2005).

The sequestration of metal ions is also a mechanism of antioxidative defense (Novo and

Parola, 2008). Transition metals can exacerbate ROS formation by reacting with H2O2 resulting in •OH formation via the Fenton reaction, therefore proteins that sequester metal ions such as transferrin, ferritin, ceruloplasmin, metallothionein and lactoferrin can be considered antioxidative as they prevent this from occurring (Novo and Parola, 2008).

1.5 Deleterious effects of ROS

ROS are important second messengers in signal transduction pathways, and their production is kept in balance detoxification to maintain cellular homeostasis. If this balance is disrupted either by enhanced ROS production, or deficient ROS detoxification, oxidative stress occurs, which can result in altered signal transduction or oxidative damage to cellular macromolecules, either of which can result in altered cellular function or cell death in the developing embryo leading to teratogenesis (Wells et al., 2009b).

1.5.1 Altered signal transduction

ROS are widely implicated in highly regulated cellular signal transduction pathways, which are selective to different cell types and their subcellular organelles. ROS signaling has

43 been linked to numerous pathways involved in cellular proliferation, differentiation, migration and apoptosis (Allen and Tresini, 2000; Thannickal and Fanburg, 2000). ROS-mediated signal transduction has been attributed to H2O2, which is less reactive and has greater diffusibility than

•- • O2 or OH, allowing it to selectively oxidize sulfhydryl groups of specific cysteine (cys) residues on proteins, acting as a ‘sulfur switch’. At physiologically relevant concentrations of H2O2, these oxidative modifications to cys are reversible and constitute a control mechanism of protein function, whereas higher exposures could lead to excessive and irreversible S-oxidation resulting in loss of protein function and teratogenic consequences (Fig. 1.15). Nearly all proteins contain cys resides that can be oxidized. The reactivity of these residues is due to the sulfur-containing thiol group (R-SH), which can stably exist in multiple oxidation states. Cys thiol groups are oxidized by ROS to a sulfenic acid (RSOH) intermediate, which can either react with neighbouring thiol groups to forming intra/intermolecular protein-protein disulfides, or with

GSH to produce a GSH-protein disulfide (Gs-Pr, mixed disulfides) (Janssen-Heininger et al.,

2008; Leonard and Carroll, 2011; Thannickal and Fanburg, 2000). Through these oxidative modifications, ROS can modulate the activity of several targets including protein tyrosine phosphatases (Caselli et al., 1998), protein tyrosine kinases (Matsuzawa and Ichijo, 2008), small

G proteins (Heo and Campbell, 2005; Lander et al., 1997), protein disulfide isomerases (Chen et al., 2008), transcription factors such as nuclear factor erythroid 2-related factor 2 (Nrf2), apoptosis-signaling kinase-1 (ASK-1), and nuclear factor transcription factor kappa B (NF-κB)

(Chen et al., 2009b; Halliwell and Gutteridge, 2007). ROS-initiated transcription factor activation can alter cellular gene expression leading to the downstream cellular response

(Brown and Griendling, 2009). At higher ROS concentrations, further oxidation of sulfenic acids

(RSOH) can lead to the formation of sulfinic (RSO2H) and sulfonic (RSO3H) acid products, which are generally considered to be irreversible (Salmeen et al., 2003).

Keap1 SH  Keap1 SOH Nrf2 Cul3 TRANSCRIPTION FACTORS Ubiquitination Nrf2 ARE NO GENE GENE EXPRESSION EXPRESSION

PDI SH  PDI SOH PROTEIN DISULFIDE ISOMERASES

INACTIVE ACTIVE

GTPase SH  GTPase SOH SMALL G PROTEINS

INACTIVE ACTIVE

PTK SH  PTK SOH PROTEIN TYROSINE KINASES

INACTIVE ACTIVE

PTP SH  PTP SOH PROTEIN TYROSINE PHOSPHATASES

ACTIVE INACTIVE

Pr SH  Pr SOH  Pr SO2H Pr SO3H

SULFENIC SULFINIC SULFONIC ACID ACID ACID

REVERSIBLE IRREVERSIBLE

ROS CONCENTRATION

Figure 1.15: Redox control of various proteins and transcription factors by reversible and irreversible modifications of cysteine Abbreviations used: Keap1, kelch-like ECH-associated protein 1; Nrf2, nuclear factor erythroid 2-related factor 2; Cul3, cullin 3; ARE, antioxidant response element; ROS, reactive oxygen species. Based upon data from: (Caselli et al., 1998; Chen et al., 2008; Heo and Campbell, 2005; Lander et al., 1997; Matsuzawa and Ichijo, 2008)

A specific example of xenobiotic initiated ROS-mediated signal transduction altering embryonic development is phenytoin, an antiepileptic drug given during pregnancy to mitigate seizure occurrence. Our lab has shown the involvement of the NF-κB signaling pathway in phenytoin- initiated embryopathies in culture (Kennedy et al., 2004). Using antisense oligonucleotides, inhibition of the downstream NF-κB signaling cascade blocked embryopathies, suggesting the involvement of ROS-initiated NF-κB signaling in phenytoin-initiated embryopathies. The NF-κB family of transcription factors regulates the expression of many genes involved in development, as well as immunity and the inflammatory response (Baeuerle and Baltimore, 1996). Another example is thalidomide, a drug used for its sedative-hypnotic effects in pregnant women, and is still in clinical use for the treatment of leprosy and multiple myeloma due to its strong immunomodulatory, anti-inflammatory and anti-angiogenic properties (Matthews and McCoy,

2003). When taken during the 3rd-8th week of pregnancy, thalidomide can initiate birth defects predominantly involving the limbs (phocomelia is the most well known), but can also affect the ears, eyes, heart, kidneys and other internal organs (Knobloch and Ruther, 2008). The mechanisms by which thalidomide initiates birth defects are unclear, however oxidative stress and alterations in signal transduction pathways have been implicated (Wells et al., 2009b).

Thalidomide has been demonstrated to suppress numerous survival signaling pathways including the canonical Wnt/β-catenin pathway (Knobloch et al., 2007) and Akt signaling, while upregulating PTEN and stimulating caspase-dependent cell death (Knobloch et al., 2008).

Additionally, thalidomide-initiated limb defects have been postulated to result from its binding to cereblon, a protein that forms an E3 ubiquitin ligase complex with damaged DNA binding protine 1 and Cul4a that is important for limb development and FGF8 expression (Ito et al.,

2010).

1.5.2 Oxidative macromolecular damage

1.5.2.1 DNA oxidation

Exposure of DNA to highly reactive hydroxyl radicals can yield several products, depending where the oxidation occurs, which can accumulate and may contribute to teratological outcomes

(Wells et al., 2009b). Approximately 20 forms of oxidatively damaged DNA have been identified, the most commonly measured being 8-oxodG (Figs. 1.16, 1.17) (Dizdaroglu, 2005; Halliwell and Gutteridge, 2007). Consequences of 8-oxodG accumulation in dividing cells may lead to transversion mutations which can affect the expression and activity of proteins required for normal development and function, and if not repaired, may affect gene transcription, DNA replication and cell division, which may lead to cancer and/or embryopathies (Evans et al.,

2004; Neeley and Essigmann, 2006; Wells et al., 2009b). Oxidatively damaged DNA may also directly initiate embryopathies via non-mutagenic mechanisms, possibly including altered gene transcription, which can occur in several ways (Wells et al., 2010). One is the ability of the 8- oxodG –initiated GC:TA transversion to disrupt RNA polymerase II function thereby stalling basal transcriptional machinery (Viswanathan and Doetsch, 1998). 8-oxodG may also affect the binding efficiency of transcription factors such as NF-κB to specific promoter elements, thereby altering gene expression (Hailer-Morrison et al., 2003). Hydroxyl radicals alternatively can attack nuclear proteins, which results in the formation of protein radicals which can then bind to

DNA to form DNA-protein cross links that can interfere with gene transcription, replication and repair (Halliwell and Gutteridge, 2007).

Figure 1.16: Major products of oxidative DNA lesions From: (Dizdaroglu, 2005)

Figure 1.17: Reaction of hydroxyl radicals (•OH) with guanine residues of DNA to form the molecular lesion 7-8,dihydro-8-oxoguanine (8-oxoG). From: (Wells et al., 2009b)

To ensure cellular viability in the presence of high concentrations of ROS, DNA repair mechanisms exist to ensure replicative fidelity and normal gene expression, including base excision repair (BER), nucleotide excision repair (NER) and mismatch repair (MMR)

(Christmann et al., 2003). In addition to its mutagenic activity, 8-oxodG is a developmentally pathogenic lesion. This was demonstrated in knockout mice lacking OGG1, a component of the

BER pathway that repairs 8-oxodG, whereby Ogg1 knockout mice exposed in utero to methamphetamine and tested postnatally for motor coordination deficits performed significantly worse than wild-type controls, suggesting 8-oxodG can contribute to postnatal neurodevelopmental deficits (Wong et al., 2008). This was further supported in mice lacking

Cockayne Syndrome B (CSB), a protein involved in DNA repair, whereby in utero methamphetamine exposure enhanced fetal brain DNA oxidation and postnatal motor coordination deficits in the CSB-deficient littermates, but not the wild-type littermates

(McCallum et al., 2011c).

1.5.2.2 Protein oxidation

Protein oxidation can impair the function of signal transduction proteins, receptors and enzymes, and subsequently cause secondary damage to other cellular macromolecules depending upon whether or not the protein was developmentally important (Wells et al., 2009b).

If damage occurs to developmentally important proteins including enzymes or receptors, aberrant development and teratogenesis can occur (Wells et al., 2009b). Oxidation of a protein is initiated by the •OH-dependent abstraction of the α-hydrogen atom of an amino acid residue to form a carbon-centered radical, which can rapidly react with oxygen to form subsequent radical intermediates that can react with other amino acid residues in the same or a different protein, forming a new carbon-centered radical (Berlett and Stadtman, 1997). All amino acid side chains

50 are susceptible to attack by •OH radicals. A well-studied measure of protein oxidation is protein carbonyl formation. Carbonyl groups can be introduced into proteins either by direct metal- catalyzed oxidation of lysine, arginine, proline and threonine residues, or by reaction with aldehydes produced during lipid peroxidation, such as 4-hydroxy-2-nonenal (HNE), and malondialdehyde (MDA), or with reactive carbonyl derivatives (Berlett and Stadtman, 1997;

Nystrom, 2005). The presence of elevated levels of carbonylated proteins has been used as a marker of ROS-mediated protein oxidation and several methods of detection have been developed (Levine et al., 1994). Methods to measure carbonyl groups on proteins include: (1) spectrophotometrically upon derivatization with dinitrophenylhydrazine (DNPH) to form a protein-hydrozone product for which absorbance can be measured at 375 nm (Zusterzeel et al.,

2000); (2) using a fluorophore which binds in a 1:1 ratio with protein carbonyl groups in a sample, and measure excitation and emission at 480-490 nm and 525-535 nm, respectively, which can then be standardized to a fluorescence standard curve (Mohanty et al., 2010); (3) immunoblotting, whereby samples are transferred to an sodium dodecyl sulfate (SDS)-PAGE gel, subjected to electrophoresis, derivatized with DNPH and then probed with an antibody against DNPH (Satapati et al., 2012). Oxidized protein products are removed from the cell by recognition and degradation by cellular proteases, and loss of developmentally important proteins in the embryo could lead to subsequent embryopathies, which has been observed with phenytoin-initiated protein oxidation (Nystrom, 2005; Winn and Wells, 1999). Interestingly, ubiquitination has been shown not to be involved in the removal of oxidized proteins, as cells lacking ubiquitination activity were able to degrade oxidized proteins, but this activity was inhibited by proteosome inhibitors (Shringarpure et al., 2003).

1.5.2.3 Lipid peroxidation

(1) Initiation: lipid peroxidation begins either by addition of a •OH across a double bond, forcing the electrons to move onto the adjacent carbon forming a carbanion, or more commonly, by hydrogen abstraction creating a lipid radical (Gutteridge, 1984; Halliwell and Gutteridge,

2007). Transition metals such as iron and copper can participate in electron exchange with O2 to form •OH.

(2) Propagation: lipid radicals can stabilize by rearrangement to a conjugated diene, or can react with molecular oxygen to give rise to a peroxyl radical (ROO•) (Halliwell and Chirico,

1993). This peroxyl radical can abstract a hydrogen atom from an adjacent fatty acid side chain, forming new carbon centered radicals that can react with oxygen to form new peroxyl radicals.

The peroxyl radical can combine with the abstracted hydrogen atom to form a lipid hydroperoxide (ROOH). Cyclic peroxides can form when a peroxyl radical attacks a double bond within the same fatty acid residue (Halliwell and Chirico, 1993; Halliwell and Gutteridge,

2007).

(3) Termination: the chain reaction terminates when two lipid peroxyl radicals combine to produce a non-radical species, or when a radical is halted by binding to antioxidants such as α- tocopherol (Halliwell and Chirico, 1993; Halliwell and Gutteridge, 2007).

Lipid peroxidation can produce DNA-damaging aldehydes such as MDA, 4-HNE, and F- isoprostanes. The decomposition of lipid peroxides by heating or reaction with metal ions

52 creates a wide variety of cytotoxic products, which can produce more radicals that can initiate further lipid peroxidation (Gutteridge and Quinlan, 1983; Halliwell and Chirico, 1993).

MDA is produced either from the peroxidation of polyunsaturated fatty acids (PUFAs), with more than two double bonds, or enzymatically during the metabolism of eicosanoids.

PUFAs are fatty acids that contain a double bond at the carbon-6 position, such as linoleic acid and arachidonic acid (Spiteller, 1998). At physiological pH, most MDA exists as the enolate ion which has low reactivity towards amino groups in proteins (Halliwell and Gutteridge, 2007). At a lower pH, MDA exists as the undissociated enol form in equilibrium with its keto form, and exhibits a higher reactivity towards proteins, able to attack residues resulting in intra- and intermolecular cross-links (Esterbauer et al., 1991). MDA can react with DNA, more specifically guanine bases, to create G to T transversions, A to G transitions, C to T transitions, frameshifts and deletions, with potentially mutagenic consequences (Marnett, 2000). MDA is metabolized to malonic semialdehyde by aldehyde dehydrogenase, and this product is decarboxylated to acetaldehyde, and is finally metabolized to acetic acid again by aldehyde dehydrogenase

(Halliwell and Gutteridge, 2007).

4-HNE is formed during the oxidation of n–6 PUFAs. Basal cellular levels of HNE in healthy tissues are approximately 1 μM or lower; however, under conditions of oxidative stress,

HNE concentrations can rise to between 2-20 μM, which is cytotoxic, leading to inhibition of

DNA and protein synthesis, cellular proliferation and NER (Feng et al., 2004; Parola et al.,

1999). HNE can react rapidly with thiol and amino groups on proteins (i.e. histidine, lysine) and amino groups on DNA bases, with guanine being the preferred target (Halliwell and Gutteridge

2007). HNE reacts with DNA to form an etheno-adduct by adding an NH2 group to the double

53 bond of the aldehyde to yield 1,N2-propano-21-deoxyguanosine adducts (Choudhury et al., 2004;

Schaur, 2003).

Isoprostanes (IPs) are formed from PUFAs with at least three double bonds, which include linolenic acid, arachidonic acid (F2-isoprostanes), eicosapentanoic acid (EPA) (F3 isoprostanes) and docosahexanoic acid (DHA) (F4 isoprostanes) (Fam and Morrow, 2003;

Roberts and Fessel, 2004). IPs can form isoketals that quickly react with amino groups to form adducts with lysine residues on membrane proteins, resulting in protein cross-linking, ultimately damaging the cellular membrane (Poliakov et al., 2004).

1.6 Studying mechanisms of teratogenesis

Mechanisms of teratogenesis can be explored using various model systems including whole embryo culture to examine the developmental fate of the embryo during organogenesis, in vivo teratology studies to examine the outcome of drug exposure on the developing fetus, and behavioural studies to investigate the postnatal consequence of in utero drug exposure on the developing brain. Any of these models can be applied to genetically modified animals that either lack, or have enhanced levels of key enzymes proposed to be involved in the teratological mechanism. Additionally, in any of these models, pharmacological probes that either enhance or inhibit specific molecular or macromolecular targets can be used to further investigate the contribution of specific pathways to teratogenesis.

1.6.1 Whole embryo culture as a model for teratogenesis

Whole embryo culture is the technique of explanting the organogenesis stage embryo, and culturing it in vitro in an environment that supports life in the absence of the maternal system (New, 1978). Culture of mouse and rat embryos was developed over 30 years ago as a method of screening for potential teratogens as well elucidating their mechanisms (New, 1978;

Sadler et al., 1982; Webster et al., 1997). There are several advantages and limitations to the use of this technique compared to studies in vivo, which must be weighed prior to using it in a study of developmental toxicity. The major advantages are: (1) the lack of a confounding maternal system that can metabolize drugs or exhibit hormonal effects; (2) drug exposure is precisely controlled, as both the drug concentration and duration of exposure can be precisely modulated;

(3) embryos at a similar developmental stage can be selected for culture; (4) embryopathies can be monitored and assessed throughout the entire study in real-time; (5) genetically modified animals and chemical and biochemical probes can be used to investigate molecular mechanisms;

(6) embryopathies between different species can be examined (mice, rats and rabbits); (7) developmental events occur in a manner similar to that in vivo; (8) short time frame to obtain results, as studies typically range from 24 – 48 hr; and (9) only small amounts of drug or probe are required, which is useful if the quantity of the drug or probe are limited, or if they are expensive. Some of the advantages can also be limiting, for example the time period during which embryos can be grown in culture is limited to 48 hr in the organogenesis period, and although developmental events occur in a similar fashion as those in vivo, results from culture cannot be directly extrapolated to teratological outcomes in vivo. From a practical standpoint, embryo culture requires technical training and skill for dissecting under a microscope, which may take several months to learn.

Endpoints measured in culture include (1) anterior neuropore closure, (2) somite development, (3) turning, (4) crown-rump length, (5) head length, (6) yolk sac diameter and (7) heart rate (Fig. 1.18).

Closure of the anterior and posterior neuropores creates a separate blood vascular circulation for the neural tube, and circulation is no longer continuous with the amnionic fluid.

When neuropores fail to close, the separation of vascular circulation is not established, which can lead to neural tube defects in the fetus (Moore and Persaud, 2007). Somite development directly relates to embryonic growth and can thus be correlated to specific developmental events. Any xenobiotic-initiated decrease in the number of somites developed over a specific period of time is indicative of a developmental delay (Moore and Persaud, 2007). The development of somites begins in the 3rd week of human development, and approximately day 7 in the mouse embryo

(Becker et al., 1996; Moore and Persaud, 2007). When the culture period begins, embryos containing anywhere from 6-9 somite pairs, depending on the strain of mouse, are selected for culture. Over the 24 hr culture period, somites develop at a rate of approximately 1 pair per hour, reaching up to 24 additional somite pairs. Embryonic turning is an important developmental event marked by folding of the embryo that establishes the body form necessary for continued development, while crown-rump length, head length and yolk sac diameter are general measures of overall growth (Moore and Persaud, 2007). In rodents, the embryo develops from a belly-flop position to the classic “C” or “fetal” position, whereas human embryos develop from the beginning in the fetal position.

Some drugs that have been shown to decrease anterior neuropore closure, somite development embryonic turning and parameters of growth include phenytoin (Abramov and

Wells, 2011b; Winn and Wells, 1995), thalidomide (Lee et al., 2011), EtOH and MeOH.

Figure 1.18: Endpoints measured in mouse whole embryo culture. ANP, anterior neuropore: In some cases, the ANP can fail to close, leaving an open ANP. YSD, yolk sac diameter: the diameter of yolk sac that encircles embryo can be decreased in some toxicities. CRL, crown-rump length: the length of the embryo measured from the top of the head to its lowest point, or “rump”, and is used as a general measure of overall growth. Turning: embryos normally turn from a belly-flop position to a classic “C” or “fetal” position. HL, head length: the length of the embryo’s head measured from the back of its head to the front of its head. Somites: paired cuboidal structures indicated by the curved line that develops into various tissues. See page 55 for details.

Numerous studies have examined the consequences of EtOH and MeOH exposure to the developing mouse and rat embryo in whole embryo culture (Tables 1.4, 1.5). From comparisons derived from embryo culture studies in the literature, it can be observed that mice are more sensitive to the teratogenic effects of both MeOH and EtOH, when compared to rats (Figs. 1.19,

1.20, 1.21). Additionally, when comparing embryopathies in the same species treated with EtOH or MeOH, both drugs exhibit a similar spectrum of embryopathies, although EtOH appears to be a more potent embryotoxin in mice than MeOH, as demonstrated by decreased anterior neuropore closure, turning and somite development (Fig. 1.22). EtOH appears to be similarly more potent than MeOH in rats, as demonstrated by the decreased crown-rump length, head length, somite development and protein content (Fig. 1.23). However, results from embryo culture may vary among different laboratories performing the technique, as Fig. 1.23 demonstrates; for example, even the same strain of rat exposed to the same drug (EtOH) can vary in somite development at a similar developmental age. Indeed, variations in these parameters occur within the litter, although with a sufficiently large sample size, interindividual differences should be representative of the mean, which would ideally be similar in cases of identical exposures.

Table 1.4: Comparison of ethanol (EtOH)-initiated embryopathies in whole embryo culture Developmental Parameters Reference Culture [Drug] Species, Plug Somite Strain (GD) duration stage (n) mg/ CRL ANC Turning YSD Somite mM HL (mm) Protein (μg) (hr) ml (mm) (%) (%) (mm) Development 0 0 4.6 5.6 2.2 30.5 Mouse, 1 21.7 4.1 4.9 2.0 28.4 (Xu et al., NS 3 to 5 48 C57BL/6J 2 43.5 3.4 4.4 1.7 27.6 2005) 4 86.9 3.0 4.2 1.3 21.6 0 0 100 96 23.8 160 Rat, 2 43.4 100 97 22.4 141 ON, (Wynter et al., Sprague- NS 48 0.5 4 86.8 100 94 19.9 120 1983) Dawley 6 130.2 100 89 17.4 82 8 173.6 100 88 12.8 41 0 0 25.7 212 Rat, (not 0.9 20 23.4 171 (Snyder et al., NS NS 48 stated) 2.3 50 21.8 141 1992) 4.6 100 21.1 113 0 0 4.5 5.1 2.3 29.2 333.3 Rat, (not NS, (Brown et al., NS 48 stated) 0.5 1.5 32.6 4.3 5.1 2.2 28.6 295.4 1979) 3 65.2 3.8 4.8 1.8 26.6 223.4 0 0 3.27 4.0 2.3 Rat, (not (Lee et al., ON, 0 NS 48 0.8 17.14 3.14 3.9 1.6 stated) 3.9 85.69 2.54 3.1 1.2 2005) 7.9 171.4 2.23 2.72 0.69 0 0 4.4 32 Rat, 0.8 17.4 4.3 31.4 (Priscott, Sprague- ON, 0 NS 48 2 43.4 4.2 31.7 1982) Dawley 3.9 84.7 3.8 30.3 5.9 128.1 3.2 28.1 Abbreviations used: ANC, anterior neuropore closure; CRL, crown-rump length; HL, head length; NS, not stated; ON, overnight; YSD, yolk sac diameter. See text on page 57 for details.

Table 1.5: Comparison of MeOH-initiated embryopathies in whole embryo culture Culture Developmental Parameters Reference Somite [Drug] Species, Plug Duration stage (n) CRL ANC YSD Protein (ug/ Strain (GD) (hr) mg/ml mM (mm) (%) Turn (%) (mm) HL (mm) Somite Development embryo) 0 0.0 2.84 3.59 1.44 21.5 89.1 2 62.4 2.63 3.45 1.31 20.8 103.9 Mouse, CD- ON, 0 3 to 5 48 1 4 124.8 2.6 3.38 1.25 20.5 127.6 6 187.3 2.63 3.32 1.26 19.3 56.3 8 249.7 2.23 3.1 1.07 21 ND 0 0.0 3.2 4.3 1.73 24.2 227.3 (Andrews et al., 1993) 2 62.4 3.2 4 1.6 23.3 213.1 Rat, Sprague- ON, 0 NS 48 4 124.8 3.23 4.14 1.6 23.1 154.7 Dawley 6 187.3 4 21.6 8 249.7 3.1 4.1 1.59 22 209.5 12 374.5 2.9 3.3 1.53 19.3 ND Mouse, CD- 0 0.0 2.68 3.48 1.35 21.37 (Abbott et al., ON, 0 3 to 5 24 1 4 124.8 2.7 3.54 1.23 20.77 1995) 0 0.0 100 100 20.8 132.1 1 31.2 100 84.6 19.4 110.9 Mouse, CD- 2 62.4 100 84.6 18.3 117.9 2h, 0 5 to 7 24 1 3 93.6 93.3 86.7 19.6 107.5 4 124.8 83.3 75 19.6 102.8 6 187.3 38.5 30.8 17.8 87.7 (Degitz et al., 0 0.0 87.5 87.5 18.3 70.3 2004) 1 31.2 100 100 18.91 72.3 2 62.4 100 91.7 18.1 68.4 Mouse, C57 2h, 0 NS 24 3 93.6 81.8 62.7 18.4 66.1 4 124.8 60 53.3 17.5 56.1 6 187.3 9.1 36.4 16.5 54.0 0 0.0 3 92 93 251.4 4 124.8 2.81 67 50 251.9 Mouse, CD- 2h, 0 8 to 10 24 1 8 249.7 2.7 67 87 216.1 10 312.1 NA 0 0 95.8 12 374.5 NA NA NA NA (Hansen et al., 0 3.21 100 92 288.9 2005) Rat, 8 249.7 3.05 100 40 200.9 Sprague- 2h, 0 NS 24 12 374.5 NA 50 0 84.3 Dawley 16 499.4 NA 40 0 55.0 20 624.2 NA NA NA NA

Table 1.5 (cont’d): Comparison of MeOH-initiated embryopathies in whole embryo culture Developmental Parameters Culture Reference Species, Plug Somite [Drug] Duration Strain (GD) stage (n) CRL ANC Turn YSD Protein (uμg/ (hr) HL (mm) Somite Development (mm) (%) (%) (mm) embryo) mg/ml mM

Rat, 0 3.4 93 97 25.5 Sprague- ON, 0 8 to 10 24 12 374.5 3.1 77 96 22.1 Dawley (Harris et al., 24 749.1 2.7 50 100 19 2004) Mouse, 0 0 1.955 88 90 2.139 22.3 ON, 1 7 to 9 24 C57 4 124.8 2.181 28.5 21.4 2.161 18.6 Mouse, 0 0 2.119 69 69 2.495 0.945 23 ON, 1 7 to 9 24 C3H (Miller and 4 124.8 1.995 54 46 2.501 0.879 20 Wells, 2011)

Abbreviations used: ANC, anterior neuropore closure; CRL, crown-rump length; Dev. Score, HL, head length; NS, not stated; ON, overnight; YSD, yolk sac diameter. See text on page 57 for details.

MOUSE VS. RAT EMBRYO CULTURE: 48 HOURS EtOH

6 5 Yolk Sac Diameter Crown-rump Length 5 4 4 mm mm 3 3

2 2

2.5 400 Head Length Protein Content 2.0 300 g)

1.5  200 ( mm 1.0 100

0.5 0 0 50 100 150 200 250

40 Somite Development DRUG CONCENTRATION (mM) 30 C57BL/6 Mouse: Xu et al., 2005 SD Rat: Wynter et al., 1983 n 20 Rat: Snyder et al 1992 Rat: Brown et al., 1979 10 SD Rat: Priscott, 1982

0 0 50 100 150 200 250

DRUG CONCENTRATION (mM)

Figure 1.19: EtOH embryopathies in mouse and rat whole embryo culture (48 hours). See text on page 57 for details. (data adapted from Table 1.4)

MOUSE VS. RAT EMBRYO CULTURE: 24 HOURS MeOH

150 Anterior Neuropore Closure C57 mouse: Degitz et al 2004 CD-1 mouse: Degitz et al 2004 100 CD-1 mouse: Hansen et al 2005 CD-1 mouse: Miller and Wells, unpublished C57 mouse: Miller and Wells, 2011 50 C3H mouse: Miller and Wells, 2011 Sprague-Dawley Rat: Hansen et al., 2005 RATS Sprague-Dawley Rat: Harris et al., 2004 PERCENTAGE (%) PERCENTAGE 0 0 200 400 600 800 1000

DRUG CONCENTRATION (mM)

30 Somite Development C57 mouse: Degitz et al 2004 CD-1 mouse: Degitz et al 2004 25 CD-1 mouse: Miller and Wells, unpublished C57 mouse: Miller and Wells, 2011 C3H mouse: Miller and Wells, 2011 20 CD-1 mouse: Abbott et al 1995 Sprague-Dawley Rat: Harris et al., 2004 RAT

Somites developed (n) developed Somites 15 0 200 400 600 800 1000

DRUG CONCENTRATION (mM)

Figure 1.20: MeOH embryopathies in mouse and rat whole embryo culture (24 hours). See text on page 57 for details. (data adapted from Table 1.5) From: (Wells et al., 2013)

MOUSE VS. RAT EMBRYO CULTURE: 48 HOURS MeOH

Somite Development 26 3.5 Crown-rump Length CD-1 mouse 24 Sprague-Dawley Rat 3.0

n 22 mm 2.5 20

18 2.0 0 100 200 300 400 500 0 100 200 300 400 500

DRUG CONCENTRATION (mM) DRUG CONCENTRATION (mM)

2.0 Head Length 250 Protein Content 1.8 200 1.6 150 g mm  1.4 100 1.2 50 1.0 0 0 100 200 300 400 500 0 100 200 300

DRUG CONCENTRATION (mM) DRUG CONCENTRATION (mM)

45 Developmental Score 5.0 Yolk-sac Diameter

40 4.5 4.0

x 35 mm 3.5 30 3.0 25 2.5 0 100 200 300 400 500 0 100 200 300 400 500

DRUG CONCENTRATION (mM) DRUG CONCENTRATION (mM)

Figure 1.21: MeOH embryopathies in mouse and rat whole embryo culture (48 hours). See text on page 57 for details. (data adapted from Table 1.5)

MOUSE EMBRYO CULTURE: 24 HOURS MeOH or EtOH

150 Anterior Neuropore C57BL/6 mouse: Degitz et al 2004 (MeOH) Closure CD-1 mouse: Degitz et al 2004 (MeOH) 100 CD-1 mouse: Hansen et al 2005 (MeOH) CD-1 mouse: Miller and Wells, unpublished (MeOH) C57BL/6 mouse: Miller and Wells, 2011 (MeOH) 50 C3H mouse: Miller and Wells, 2011 (MeOH) CD-1 mouse: Miller and Wells, unpublished (EtOH) EtOH C57BL/6 mouse: Miller and Wells, unpublished PERCENTAGE (%) PERCENTAGE (EtOH) 0 0 100 200 300 400 DRUG CONCENTRATION (mM)

150 C57BL/6 mouse: Degitz et al 2004 (MeOH) Turning CD-1 mouse: Degitz et al 2004 (MeOH) CD-1 mouse: Hansen et al 2005 (MeOH) 100 CD-1 mouse: Miller and Wells, unpublished (MeOH) C57BL/6 mouse: Miller and Wells, 2011 (MeOH) C3H mouse: Miller and Wells, 2011 (MeOH) 50 CD-1 mouse: Miller and Wells, unpublished (EtOH) EtOH C57BL/6 mouse: Miller and Wells, unpublished (EtOH)

PERCENTAGE (%) PERCENTAGE 0 0 100 200 300 400

DRUG CONCENTRATION (mM)

26 C57BL/6 mouse: Degitz et al 2004 (MeOH) Somite Development 24 CD-1 mouse: Degitz et al 2004 (MeOH) CD-1 mouse: Miller and Wells, unpublished (MeOH) 22 C57BL/6 mouse: Miller and Wells, 2011 (MeOH) 20 C3H mouse: Miller and Wells, 2011 (MeOH) CD-1 mouse: Abbott et al 1995 (MeOH) 18 CD-1 mouse: Miller and Wells, unpublished (EtOH) 16 C57BL/6 mouse: Miller and Wells, unpublished EtOH (EtOH) 14 Somites Developed (n) 0 50 100 150 200 250

DRUG CONCENTRATION (mM)

Figure 1.22: Comparison of MeOH vs. EtOH embryopathies in mice. See text on page 57 for details.

Data adapted from Tables 1.4 and 1.5 From: (Wells et al., 2013)

RAT EMBRYO CULTURE: 48 HOURS MeOH or EtOH

5 Sprague-Dawley Rat: Andrews et al, 1993 (MeOH) Crown-rump Length Outbred rats: Brown et al 1979 (EtOH) 4 Rat: Lee et al 2005 (EtOH) 3 Sprague-Dawley Rat: Priscott 1982 ( EtOH) Rat: Snyder et al. 1992 (EtOH) 2 1 MeOH 0

Crown-rump length (mm) length Crown-rump 0 100 200 300 400 500 DRUG CONCENTRATION (mM)

2.5 Head Length 2.0 Sprague-Dawley Rat: Andrews et al, 1993 (MeOH) Outbred rats: Brown et al 1979 (EtOH) 1.5 Rat: Lee et al 2005 (EtOH)

1.0 MeOH 0.5 Head length (mm) length Head 0.0 0 100 200 300 400 500 DRUG CONCENTRATION (mM)

40 Somite Development Sprague-Dawley Rat: Andrews et al, 1993 (MeOH) 30 Outbred rats: Brown et al 1979 (EtOH) Rat: Snyder et al. 1992 (EtOH) 20 Sprague-Dawley Rat: Priscott 1982 ( EtOH) Sprague-Dawley Rat: Wynter et al 1983 (EtOH) 10 MeOH

Somites Developed (n) Somites 0 0 100 200 300 400 500

400 Protein Content Sprague-Dawley Rat: Andrews et al, 1993 (MeOH) Outbred rats: Brown et al 1979 (EtOH) 300 MeOH Rat: Snyder et al. 1992 (EtOH) Sprague-Dawley Rat: Wynter et al 1983 (EtOH) 200

100 Sprague-Dawley

0 0 100 200 300 Protein content (ug/embryo) content Protein DRUG CONCENTRATION (mM)

Figure 1.23: Comparison of MeOH vs. EtOH embryopathies in rats. See text on page 57 for details. Data adapted from Tables 1.4 and 1.5 From: (Wells et al., 2013)

1.6.2 Passive avoidance task as a model for behavioural deficits

The passive avoidance task is a test of learning and memory. In order to successfully complete the task, new information must be stored in the hippocampus (Lorenzini et al., 1996), which is a brain region shown to be affected by prenatal EtOH exposure (Gil-Mohapel et al.,

2010). Disruptions of the hippocampus observed in animals exposed to EtOH in utero, such as hippocampal neuronal loss (Walker et al., 1980; Uban et al., 2010) and structural changes in hippocampal neurons (Gil-Mohapel et al., 2002), provide evidence for EtOH-initiated neurotoxicity related to the passive avoidance task that would alter the ability to learn the task, resulting in decreased performance. Additionally, damage to the hippocampus and amygdala has been shown to impair passive avoidance performance (Phillips and LeDoux, 1992; Russo et al.,

1976; Stubley-Weatherly et al., 1996). The test apparatus consists of a plastic box with two equal sized chambers, one with clear plastic walls designated as the ‘light chamber’, and another with red walls designated as the ‘dark chamber’, separated by a guillotine door. The floors of both chambers consist of stainless steel rods. The rods of the dark chamber are connected to a power supply that provides a 1.0 milliampere (mA) shock. To ensure the current runs through the mouse and provides a shock without producing a short circuit, alternating rods are connected to the anode and cathode such that the mouse has to grasp onto the rods in order complete the circuit and receive the shock. In trial #1, the mouse is placed in the light chamber and is allowed to explore for 20 seconds, after which time the guillotine door is lifted. Once the mouse enters the dark chamber, the guillotine door is closed, and a shock of 1.0 mA is administered for 4 seconds. The mouse is then returned back to its home cage. The apparatus is sprayed with 70%

EtOH and wiped in between each mouse tested to eliminate any olfactory cues. The same procedure is repeated 24 and 48 hours later for trials #2 and #3, respectively. The time it takes

67 the mouse to enter the dark chamber in both trials after opening the guillotine door is recorded as

‘latency to enter the dark chamber’. The longer the latency, the better the learning and memory of the animal, and particularly the better the animal is at remembering not to enter the dark chamber. Several studies have demonstrated a learning deficit upon in utero exposure to EtOH

(Table 1.6).

1.6.3 Chemical probes

Pharmacological probes have proven useful in investigating mechanisms of teratogenesis.

Such probes include small interfering RNA (siRNA), chemical enzyme inhibitors, protein therapy and free radical spin trapping agents. This section will focus on the latter 3, which have been used in several of my studies. Specifically, the free radical spin trapping agent PBN, the

NOX enzyme inhibitor DPI, and protein therapy with polyethylene glycol (PEG)-conjugated catalase were used in mechanistic studies outlined in this thesis, and will be discussed further in the following sections. There are several advantages and limitations to the use of these probes to elucidate mechanisms of teratogenicity, provided the appropriate concentration of probe is used, understanding probe selectivity, mechanism of action, and alternative pharmacological effects of the probes being used, and how these may confound interpretation of results. Results obtained from studies employing such chemical probes should be corroborated with other, in some cases more definitive, studies using alternative and in some cases more rigorous approaches, such as genetically modified animals or siRNA to selectively target genes or proteins of interest.

Table 1.6: Effect of in utero EtOH exposure on postnatal performance on a passive avoidance task

Strain/ Duration Age Dose BAC (mg/100mL) Shock Outcome Reference Species (GD) conducted

Liquid diet: # of trials to meet criteria: EtOH C57BL/6 0.2 mA, PN25 (male (Becker and Randall, 25% EDC (3.7% w/v) (GD6-18) 6 to birth ND (2.2) vs. control (1.5) mice 2 secs only) 1989) 17% EDC (GD18-birth) (p < 0.05)

280 (maternal) 0.13 mA, # of trials to meet criteria: EtOH 2 x 2.9 g/kg i.g (20% w/v) LS/SS mice 7 to 18 256 mg/100ml until mouse PND 19 (4) vs. control (2.3) (Gilliam et al., 1987) 6 hr apart (amnionic fluid) exits chamber (p < 0.001)

# of trials to meet criteria: Sprague- 0.5 mA, 6 g/kg 8 to 18 ND PND 20 EtOH (4.3) vs. control (3.8) (Mattson et al., 1993) Dawley rats 0.5 secs P < 0.05

Dose dependent ↑in number of 0.8 mA PND 16 trials to meet criterion (p < 0.001) Liquid diet: Long Evans Group 1: 35% EDC (6.6% v/v) 10.9 (35%) 5 to 20 0% (4.2) vs. 17%: (Abel, 1982) hooded rats Group 2: 17% EDC (3.3% v/v) 1.8 (17%) p < 0.02 1.0 mA PND 40-42 Group 3: 0% EDC 0% vs. 35% (5.8): p < 0.002 1.0 mA PND 114 No differences Liquid diet # of trials to meet criteria: Group 1: 32% EDC PND 18 0% (1.8) vs. 8% (3.4) p < 0.05 Group 2: 19% EDC 10.4 (32%) (females) 0% vs. 19% (2.8) p < 0.05 Long Evans Group 3: 8% EDC 6 to 16 2.9 (19%) 0.5 mA 0% vs. 32% (5) p < 0.05 (Riley et al., 1979) rats Group 4: 0% EDC 0.4 (8%) Liquid diet PND 41-53 0% (1.5) vs. 32% (2.3) Group 1: 32% EDC (males) p < 0.05 Group 2: 0% EDC

C57BL/6 Deficit in acquisition and Liquid diet 25%, 0% EDC 5 to 18 ND NS PND 70 (Randall et al., 1985) mice performance in shuttle avoidance Abbreviations used: BAC, blood alcohol concentration; EDC, ethanol-derived calories; GD, gestation day; LS, long sleep; mA, milliampere; ND, not determined; PND, postnatal day; SS, short sleep. See text on page 66 for details.

1.6.3.1 Free radical spin trapping agent phenylbutylnitrone (PBN)

PBN is a free radical spin trapping agent that contains a nitroso group (R=N–O), and is used to scavenge ROS through its ability to stabilize a nitroxyl free radical by delocalizing electrons into the double bond between oxygen and nitrogen (Halliwell and Gutteridge, 2007). By reacting with the free radical, a more stable and longer-lived radical is produced, which is detectable using electron paramagnetic resonance (EPR) spectrometry. PBN is a commonly used spin trap to investigate xenobiotic-initiated free radical-mediated mechanisms of teratogenicity (Floyd et al., 2002; Halliwell and Gutteridge, 2007). PBN traps both EtOH-derived carbon-centred lipid free radicals (Reinke et al 1987) and hydroxyethyl radicals comprising more than 80% of EtOH- derived radicals following reactions with hydroxyl radical, which can lead to lipid peroxidation both in vitro and in vivo (Reinke, 2002) (Fig. 1.24). PBN has been shown to protect against birth defects caused by other ROS-initiating teratogens, and to reduce the associated oxidative damage to cellular macromolecules including DNA, protein and lipids (Wells et al., 2009b). For example, PBN blocked oxidation of embryonic cellular macromolecules and teratogenesis in pregnant rabbits treated with the sedative/antileprotic/anticancer drug thalidomide (Lee et al.,

2011; Parman et al., 1999), and in pregnant mice treated with the antiepileptic drug phenytoin

(Liu and Wells, 1994; Wells et al., 1989), as well as trapping free radicals of structurally similar drugs and reducing macromolecular oxidative damage in vitro (Parman et al., 1998).

The protective effect of PBN against ROS may involve not only directly trapping free radicals, but also indirectly by blocking inflammatory pathways that produce ROS. In particular,

PBN can inhibit: (1) the induction iNOS, thereby decreasing NO• production; (2) activation of

NFκB, which is involved in ROS signaling; (3) inducible cyclooxygenase (COX) 2 mRNA expression and COX catalytic activity (Kotake et al., 1998), involved in xenobiotic bioactivation

Hydroxylamine adduct Reduction

Radical Adduct

PBN Oxidation

Nitrone Adduct

Figure 1.24: Phenylbutylnitrone (PBN) trapping hydroxyl radical. Hydroxyl radicals are trapped and stabilized by PBN, thereby making them amenable to quantification by ESR and decreasing their toxicity in embryo culture and in vivo. Modified from: (Ramirez et al., 2007)

to free radical intermediates; (4) transcription of proinflammatory cytokines such as tumor necrosis factor-α (TNF- α) and interferon-gamma (IFN-γ), which is correlated with decreased activation of the NF-κB and activator protein-1 (AP-1) transcription factors (Sang et al., 1999);

(5) cleave of caspase-3, which is involved in apoptosis (McLaughlin et al., 2003); and, (6) activation of NOXs (Chang et al., 2009) which are activated in the embryo by in utero EtOH exposure, increasing ROS production, embryonic DNA oxidation and structural teratogenesis

(Dong et al., 2010). PBN has neuroprotective and antiaging properties (Carney and Floyd, 1991) and improves learning in lipopolysaccharide (LPS)-treated neonatal rats (Fan et al., 2008).

It is important to note that the above alternative pathways overlap with those implicated in the teratogenic mechanism of several ROS-initiating teratogens including phenytoin, thalidomide and benzo[a]pyrene (Kasapinovic et al., 2004; Kennedy et al., 2004; Knobloch and

Ruther, 2008; Knobloch et al., 2008; Parman and Wells, 2002). PBN exhibits alternative effects that are independent of ROS scavenging, so its ability to offer protection should be weighed against careful assessment of these alternative effects, and not be singularly interpreted as direct evidence for the involvement of ROS in a teratological mechanism (Table 1.7). To interpret protection by PBN as proof of oxidative stress involvement, the full spectrum of its biological effects should be considered when used in vivo or in embryo culture studies. Outcomes should be confirmed by the measurement of oxidatively damaged macromolecules such as proteins, DNA or lipids, among other approaches; otherwise, the potential confounding contribution of alternative effects of PBN cannot be ruled out.

Table 1.7: Pharmacological effects of PBN

Effect Reference (Miyajima and Kotake, 1997; Prevent iNOS gene upregulation and nitric oxide production Miyajima and Kotake, 1995; Tabatabaie et al., 2000)

Inhibit IL-1β, H2O2 and sorbitol-mediated p38 activation (phosphorylation) and significantly increase phosphatase (Robinson et al., 1999) activity

Inhibit NADH-linked H2O2 biosynthesis in respiring mitochondria; (ROS responsible for the inactivation of (Hensley et al., 1998) phosphatases and acting as signalling molecules) Inhibit NADPH oxidase activation, and mRNA expression of (Chang et al., 2009) hyperoxia-induced TNF-α, IL-6 and TGF-β mRNA

Suppress mRNA expression of proapoptotic genes (fas-A, (Stewart et al., 1999) bcl-x, fas-L, ICE, YAMA, ICH L/S, bax, bcl2) Inhibit NF-κB, iNOS mRNA, COX2 mRNA and COX catalytic (Kotake et al., 1998) activity Inhibit transcription of proinflammatory cytokines (TNF-α, (Pogrebniak et al., 1992; Sang IFN-γ, IL-6, IL-1, c-fos) which is correlated with decreased et al., 1999) activation of the NF-κB and AP-1 transcription factors Inhibit upregulation of IL-1 β, TNF- α and iNOS mRNA and (Lin et al., 2006) suppressed hypoxia ischemia induced NF-κb activation Inhibit expression of proapoptotic immediate early gene (Marterre et al., 1991) product c-Fos

Inhibit COX (weak) (Nakae et al., 1998) Inhibit breakdown of nitric oxide (enhancing vasodilation) (Inanami and Kuwabara, 1995) Enhance IL-10 levels (protective) (Kotake et al., 1999) Reduce c-jun mRNA and c-Jun expression; enhanced p- (Hirano et al., 2005) JNK1 and JNK-1 Abolish caspase-3 cleavage (McLaughlin et al., 2003) Inhibit expression of Fas, decrease cleaved caspase-8, (Inoue et al., 2007) cytochrome c and cleaved caspase-9 Some of the above effects may contribute to the mechanism by which PBN decreases developmental toxicity of drugs. Abbreviations used: COX2, cyclooxygenase 2; H2O2, hydrogen peroxide; IL-1β, interleukin 1 beta; IL-6, interleukin 6; IFN-γ, interferon gamma; IL-1, interleukin-1; IL-10, interleukin-10; iNOS, inducible nitric oxide synthase; JNK, jun terminal kinase; NADH, nicotinamide adenine dinucleotide; NADPH, nicotinamide adenine dinucleotide phosphate; NF-κB, nuclear factor kappa-light-chain-enhancer of activated B cells; TNF-α, tumor necrosis factor alpha; TGF-β, transforming growth factor beta.

1.6.3.2 NADPH Oxidase (NOX) inhibitors

Several studies have used inhibitors of NOX enzymes to implicate the involvement of NOX and their production of ROS in the particular mechanism being studied, although each inhibitor has its advantages and disadvantages, as outlined in Table 1.8. Given the lack of specificity, or off-target effects inherent to these inhibitors, their ability to modulate an outcome should not be taken as definitive evidence of the involvement of NOXs. Indeed, although an effect may support the implication of NOX enzymes, as with other chemical probes used to elucidate mechanisms of teratogenesis, alternative and in some cases more rigorous approaches should be employed to confirm the mechanism underlying the observed outcome, such as the use of NOX knockout mice or siRNA. To that end, the use of genetically modified mouse models has been employed in numerous studies to implicate NOX involvement in a variety of mechanisms and conditions, as outlined in detail in section 1.7.9, and in Table 1.13.

DPI was used as an inhibitor in my studies to implicate the role of NOXs in MeOH- initiated embryopathies. The mechanism of action of DPI is outlined in Fig. 1.25. Similar to

PBN, while DPI is a broad spectrum NOX inhibitor, it also has alternative pharmacological effects including but not limited to: (1) cholinesterase inhibition (Tazzeo et al., 2009); (2) smooth muscle calcium pump inhibition (Tazzeo et al., 2009); (3) inhibition of flavin-containing enzymes such as NOS (Stuehr et al., 1991; Wind et al., 2010), XO (Sanders et al., 1997; Wind et al., 2010) and P450 reductase (Tew, 1993); (4) mitochondrial respiratory chain complex I inhibition (Majander et al., 1994); (5) inhibition of the pentose phosphate pathway (Herrick et al., 2006); (6) antiproliferative effects via G2 cell cycle arrest and down-regulation of cyclin B

(Scaife, 2004; Scaife, 2005). These alternative pharmacological effects must be taken into account when interpreting results from studies using this probe.

Table 1.8: NADPH oxidase inhibitors

Chemical Nox Off-target Name Mechanism of Action References Structure inhibited effects

Inhibits assembly of oxidase by Nonselective (Diatchuk et AEBSF suppressing NOX2 serine protease al., 1997) association between inhibitor NOX2 and p47phox

Inhibits assembly of oxidase by (Simons et suppressing al., 1989; Apocynin NOX2 H O scavenger association between 2 2 Stolk et al., p47phox and 1994; t Hart NOX2/p22phox et al., 1990)

Inhibits CYP450 oxidoreductase, Inhibits flavoprotein by NADH (Cross et al., abstracting electrons dehydrogenase, 1984; Gatley All NOX DPI from FAD preventing NADH- and Sherratt, isoforms electron flow through ubiquinone 1976; Stuehr cyt b558 oxidoreductase, et al., 1991) NOS, xanthine oxidase

NOX2, (Bhandarkar Fulvene Undefined None reported NOX4 et al., 2009)

Undefined. Structural (Laleu et al., GK- similarity to NADPH NOX1, 2010; None reported 136901 suggests competitive NOX4 Sedeek et substrate inhibitor al., 2010)

Undefined, targets NOX1 oxidase but not (Gianni et al., ML171 NOX1 None reported its subunits (NOXO1, 2010) NOXA1, Rac1)

Inhibits assembly of [H]- oxidase by (DeLeo et al., GP91 RKKRRQRRRCS suppressing NOX2 None reported 1995; Rey et DSTAT TRIRRQL-NH2 association between al., 2001) NOX2 and p47phox

Table 1.8 (cont’d): NADPH oxidase inhibitors

Chemical Nox Off-target Name Mechanism of Action References Structure inhibited effects

(Fan et al., Potential ROS 2005; Plumbagin Undefined NOX4 scavenger Rossary et al., 2007) RRRPRPPYLP Binds to SH3 domains Inhibits SH3- on p47phox and domain (Shi et al., PR-39 RPRPPPFFPPR NOX2 LPPRIPPGFPP prevents its binding to containing 1996) RFPPRFP p22phox proteins

NOX2, (Cayatte et S17834 Undefined None reported NOX4 al., 2001)

(ten NOX2, VAS2870 Undefined None reported Freyhaus et NOX4 al., 2006)

Non- (Wind et al., VAS3947 Undefined None reported selective 2010) We cannot exclude a protective mechanism through nitric oxide synthase (NOS), which is known to contribute in some cases to embryopathies in mouse whole embryo culture (Kasapinovic et al., 2004). The IC50 (half maximal inhibitory concentration of an enzyme) value for AEBSF is 1.2 mM, for apocynin is 1.9 mM, and for VAS 3947 is 13 μM (Wind et al., 2010).

DPI + e- Phenyl radical

FAD e- FAD

FAD - e NADP+ + H +

NADPH

DPI ADDUCT

NADPH donates its DPI abstracts electron Radical adds back directly electron to FAD of from reduced redox centre Onto prosthetic group or NOX enzyme to form a phenyl radical Adjacent proteins  LOSS OF RECOVERABLE FAD

Figure 1.25: Mechanisms of action of DPI inhibiting NOXs Modified from: (O'Donnell et al., 1993) Abbreviations used: FAD, flavin adenine dinucleotide; NADPH, nicotinamide adenine dinucleotide phosphate; NOX, NADPH oxidase; DPI, diphenyleneiodonium

1.6.3.3 Polyethylene glycol (PEG)-conjugated catalase

PEG is an inert linear polymer with a molecular formula of H-O-(CH2-CH2-O)n-CH3 and a molecular weight of 5000 Da when n=150 (Beckman et al., 1988). Conjugation of catalase to

PEG is performed to overcome 2 key problems with the administration of exogenous antioxidative catalase: (1) catalase has a very short half-life, in the order of 6-10 minutes after intravenous (i.v.) injection, and (2) catalase cannot traverse cell membranes (Beckman et al.,

1988). Since ROS travel only very short distances prior to reacting with cellular macromolecules, catalase must be able to diffuse to intracellular sites of ROS generation with a reasonable half- life to effectively act as an antioxidant. By conjugating catalase with PEG, its half-life is increased from 6-10 minutes, to 40 hours in the rat, slows down its renal clearance by increasing its molecular weight by up to 300% (Abuchowski et al., 1977b), and is better able to cross cell membranes (Boni et al., 1984). Additionally, when PEG was conjugated to bovine liver catalase, approximately 95% of catalase activity was maintained (Abuchowski et al., 1977a). Safety evaluation of PEG-catalase toxicity in mice and rats in acute, subacute and subchronic studies revealed that rodents can tolerate large doses without any changes in survival, appearance, behaviour, food intake, blood chemistry, hematology, urinalysis or body weight, and exhibit minimal signs of toxicity (Viau et al., 1986) (Table 1.9). Additionally, the route of administration did not significantly change its disposition or half-life (Fig. 1.26). Studies in whole embryo culture and in vivo have demonstrated the ability of PEG-catalase to protect against ROS-initiated oxidative damage and decrease the associated birth defects, as well as increase embryonic activity of catalase when administered directly to the culture medium in whole embryo culture, or by maternal administration (Table 1.10).

Table 1.9: Safety evaluation of PEG-catalase in mice and rats Species, Dose Study type N Dose (kU/kg) ROA Adverse effects Strain schedule Splenic hypertrophy and generalized 10,000 1X i.p splenic stimulation, transient 4-6% body Swiss 8 males, 8 weight loss; regained weight by day 7 Acute Webster females Splenic hypertrophy and generalized mice 20,000 1X i.p splenic stimulation, transient 4-6% body weight loss; regained weight by day 7 25 i.p none Sprague- 8 males, 8 5X/week for Subacute Dawley rats females 50 4 weeks i.p none 100 i.p none 5 i.p none Sprague- 8 males, 8 2X/week for Subchronic Dawley rats females 12.5 3 months i.p none 25 i.p none Adapted from: (Viau et al., 1986)

Figure 1.26: Blood plasma elimination-time curve for PEG-catalase in mouse. Enzyme was administered as a single i.p, i.v. (tail vein), or i.m. (thigh) injection. Groups of 12 mice were used for each route of enzyme administration. Blood samples were collected from the supraorbital plexus for the assay of catalase activity. Each point is the average of 4 mice. (+) i.v.; (square) i.p.; (diamond) i.m. The half-life for PEG-catalase was approximately75 hours. Red lines indicate points taken for half life extrapolation (3.6 – 1.8 log catalase units/ml plasma); blue line is the extrapolated elimination half-life; half-life calculated by subtracting 0.25 days from 3.4 days, which were the intersection points for the above-mentioned values of catalase. From: (Viau et al., 1986)

Table 1.10: PEG-Catalase in models of developmental toxicity

WHOLE EMBRYO CULTURE Drug PEG-Cat Strain Drug Outcome Reference Concentration concentration

80 μM; aCat embryopathies ↓ by (Abramov and aCat, hCat PHT 1680 U/ml 40 ug/ml PEG-cat co-treatment Wells, 2011a)

80 μM; DNA oxidation ↓ by PEG- (Winn and CD-1 PHT 1680 U/ml 40 ug/ml cat co-treatment Wells, 1995)

50 kU/kg maternal aCat embryopathies ↓ by Miller and aCat, hCat EtOH 4 mg/ml administration 8 maternal PEG-cat Wells, hr prior to embryo pretreatment (submitted) culture IN VIVO Strain Drug Drug Dose PEG-Cat dose Outcome Reference

Teratogenesis and DNA 50 kU/kg 6 hr (Abramov and aCat, hCat PHT 65 mg/kg oxidation ↓ by PEG-cat pretreatment Wells, 2011a) pretreatment aCat and C57BL/6 WT teratogenesis and DNA (Miller et al., aCat, hCat EtOH 4 g/kg 50 kU/kg oxidation ↓ by PEG-cat 2013c) pretreatment 10 - 50 kU/kg 8 hr (Winn and CD-1 PHT 65 mg/kg ↓ teratogenesis pretreatment Wells, 1999) For the first 2 studies in whole embryo culture, PEG-Cat was administered in the culture medium along with the drug. For the third study in embryo culture, PEG-Cat was administered to the pregnant dam (i.p) 8 hours prior to the initiation of culture. For all 3 in vivo studies, PEG-Cat was administered to the dam (i.p.) 8 hours prior to maternal drug administration. Abbreviations used: aCat, acatalasemic; EtOH, ethanol; hCat, transgenic human catalase- expressing; i.p, intraperitoneal; PEG-cat, PEG-catalase; PHT, phenytoin; WT, wild-type

1.7 NADPH oxidases (NOX)

NADPH oxidases are evolutionarily highly conserved multi-subunit enzymes whose

•- primary function is the production of ROS, specifically O2 and H2O2 produced from the oxidation of NADPH (Lalucque and Silar, 2003). The catalytic subunit is comprised of six hydrophobic transmembrane helices that contain four conserved heme-binding histidine residues believed to be required for electron transfer across the plasma membrane, and a long cytosolic C- terminal tail that contains binding sites for NADPH and flavin adenine dinucleotide (FAD)

(Groemping and Rittinger, 2005). The C-terminus of the cytosolic tail contains a putative

NADPH binding site where electrons are transferred from NADPH to FAD bound to the N- terminus of the tail (Biberstine-Kinkade et al., 2001; Han et al., 2001; Li et al., 2005b). The electrons are then transferred through the heme groups to reduce molecular oxygen, thereby

- producing O2• or H2O2, depending on the NOX isoform (Table 1.11). The NOX family is comprised of 7 oxidases, including NOX1, NOX2, NOX3, NOX4, NOX5, Dual oxidase

(DUOX) 1, and DUOX2 (Drummond et al., 2011; Selemidis et al., 2008). In addition to the catalytic NOX subunit, some NOXs require cytosolic regulatory subunits for enzymatic localization, stability and activation including p22phox, p47phox, p67phox, p40phox and the small GTPase Rac (Lambeth, 2004; Vignais, 2002). NOX1, NOX3 and NOX4 are almost identical to NOX2, NOX5 contains a calcium-binding domain in addition to the NOX2 catalytic subunit, while DUOX1 and DUOX2 contain an additional extracellular peroxidase domain capable of oxidizing cellular substrates along with the reduction of H2O2 (Fig 1.27)

Table 1.11: Tissue distribution and subcellular localization of NADPH oxidase (NOX) enzymes

Sequence Necessary Tissue Subcellular ROS homology Subunits Distribution Localization Produced to NOX2 Colon, Plasma endothelium, NOXO1, membrane,  placenta, O NOX1 60% NOXA1, Caveolae, 2 prostate, p22phox, Rac endosomes, smooth muscle, perinuclear uterus Plasma p22phox, Wide membrane, p47phox,  NOX2 100% distribution; perinuclear, O2 p67phox, phagocytes nuclear pore, p40phox, Rac endosomes Inner ear, fetal NOXO1, kidney and  NOX3 56% NOXA1, Not determined O2 spleen, brain, p22phox, Rac skull bone Kidney, focal adhesions, endothelium, endoplasmic smooth muscle, reticulum, NOX4 39% p22phox hematopoietic nucleus, H 2O2 stem cells, mitochondria, fibroblasts, perinuclear, neurons cytoplasm Testis, spleen, lymph nodes, vascular smooth musccle, bone Plasma NOX5* 27% none H O marrow, membrane 2 2 pancrease, placenta, ovary, uterus, stomach Apical thyroid, lung DUOXA1, membrane, H O DUOX1/2 50% epithelium, 2 2 DUOXA2 endoplasmic prostate reticulum

Modified from: (Brown and Griendling, 2009; Chen et al., 2009b; Lassegue et al., 2012)

*Present in humans, not present in rodents

NOX1 p22phox NOX2 p22phox NOX3 p22phox

RAC1 RAC RAC NOXO1 p47phox NOXO1 NOXA1 p67phox NOXA1 p40phox

Peroxidase-like domain DUOXA1/ DUOX1/ DUOXA2 NOX4 p22phox NOX5 DUOX2

Ca2+ EF hand Ca2+ Ca2+ motifs Ca2+ Ca2+ EF hand Ca2+ motifs

Figure 1.27: Subunit composition of the seven mammalian NADPH oxidase (NOX) isoforms Modified from: (Drummond et al., 2011) supplemented with data from (Brown and Griendling, 2009; Chen et al., 2009b; Lassegue et al., 2012)

1.7.1 NOX2

The original NOX2, first termed gp91phox, was first discovered in 1973 (Babior et al.,

1973). NOX expression and activation are dependent upon p22phox co-expression (Kawahara et al., 2005) with NOX2 being rapidly degraded in the case of p22phox deficiency. Together,

NOX2 and p22phox comprise the cyt b558 complex. p22phox is bound to the NOX catalytic subunit, and contains 3 transmembrane domains, and a proline-rich region (PRR) at its C terminus capable of interacting with an Src homology 3 (SH3) domain of the organizer protein p47phox, which is essential for NOX activation (Fig. 1.13) (Groemping and Rittinger, 2005;

Leto et al., 1994). p47phox is a cytosolic protein that translocates to the cyt b558 complex upon stimulation. The p47phox has 2 SH3 domains, one of which is bound to an upstream proline-rich region of the protein under non-stimulated conditions.

Upon stimulation, a number of serine residues on p47phox become phosphorylated, exposing the bound SH3 domain allowing it to interact with the PRR of p22phox (Fontayne et al., 2002). p47phox is essential for the recruitment of another regulatory protein, p67phox, which interacts via its second SH3 domain (de Mendez et al., 1997). Once p67 has translocated to the membrane, it can interact directly with NOX2, and recruit the binding of a small GTPase Rac.

Finally, p40phox interacts with NOX2 via binding domains on p47phox and p67phox. NOX2 is expressed ubiquitously in phagocytic cells, as well as endothelial cells (Brown and Griendling,

2009).

1.7.2 NOX1

The first homologue of NOX2 was discovered in 1999, and was originally termed Mox1

(mitogen oxidase 1) (Suh et al., 1999). NOX1 shares 60% homology with NOX2, and is very similar in structure. The main difference from NOX2 is that the cytosolic regulators, p47phox and p67phox, are replaced by their respective homologs, Nox organizer 1 (NoxO1) and Nox activator 1 (NoxA1) (Ago et al., 2003; Geiszt et al., 2003a). NOX1 is expressed in vascular smooth muscle cells, colon and airway epithelium, and uterus (Banfi et al., 2003; Brown and

Griendling, 2009; Cheng et al., 2001; Suh et al., 1999).

1.7.3 NOX3

NOX3 was discovered in 2000 based on its sequence homology to NOX2 and 1, and shares 56% homology with NOX2 (Kikuchi et al., 2000). Structurally, NOX3 is similar to NOX2 in its requirement of p22phox for stabilization, the NADPH- and FAD-binding domains and the heme-containing histidines, although it uses NoxO1 and NoxA1 for activation, like NOX1.

NOX3 is found in the inner ear, brain, skull bone, and fetal kidney and spleen. (Banfi et al.,

2004a; Cheng et al., 2001; Ueno et al., 2005).

1.7.4 NOX4

NOX4 was first identified in 2000, and was initially termed ‘Renox’ as it was found in the kidney (Geiszt et al., 2000; Shiose et al., 2001). NOX4 shares 39% structural homology to

NOX2, and requires p22phox for activity and stability. However, NOX4 does not require the

PRR of p22phox for activation, indicating that any interactions between the PRR and p47phox or

NoxO1 are not essential to oxidase activity (Kawahara et al., 2005). In contrast to NOX2, NOX1

•- and NOX3, the main ROS produced by NOX4 is H2O2 rather than O2 . This is possibly due to a delayed dissociation of superoxide from the catalytic subunit due to hindrance by a third

•- cytoplasmic loop that is present in NOX4, thereby allowing a second O2 molecule to form and dismutate prior to being released from the enzyme (Takac et al., 2011). NOX4 is expressed in the kidney, endothelium, smooth muscle, hematopoietic stem cells, heart, pancreas, fibroblasts, osteoclasts and neurons, but not in phagocytic cells (Hilenski et al., 2004; Piccoli et al., 2005;

Vallet et al., 2005).

1.7.5 NOX5

NOX5 shares 27% homology to NOX2, and possesses the same NADPH- and FAD- binding sites as NOX2. However, NOX5 contains a unique cytosolic N-terminal Ca2+ binding domain with calmodulin-linked EF-hand motifs that render NOX5 highly sensitive to Ca2+

(Banfi et al., 2001), and does not require p22phox for activation (Kawahara et al., 2005), nor any of the cytosolic regulatory proteins for any of the other NOXs. To date, five splice variants of

NOX5 have been identified (Nox5α, Nox5β, Nox5γ, Nox5δ and Nox5s), which differ in their tissue distribution, and the sequence of their Ca2+ binding regions. For example, Nox5s does not contain a Ca2+ binding domain, and is constitutively active (BelAiba et al., 2007). NOX5 is expressed in testis, spleen, lymph nodes, vascular smooth muscle, bone marrow, pancreas, placenta, ovary, uterus, stomach, and endothelial cells (Banfi et al., 2001; Banfi et al., 2004b;

BelAiba et al., 2007; Cheng et al., 2001).

1.7.6 Dual Oxidases (DUOX) 1/2

DUOX 1/2 share 50% sequence homology with NOX2, and associate with DUOXA1 and

DUOXA2. They differ from NOX2 in that they contain an N-terminus peroxidase region capable

of producing H2O2. DUOX1/2 are expressed in the thyroid gland and lung epithelium involved in hormone biosynthesis and host defense (De Deken et al., 2000; Geiszt et al., 2003b).

1.7.7 Developmental NOX ontogeny

H2O2 production has been blocked in the preimplantation mouse embryo at the 2-cell stage using the NOX inhibitor DPI (Nasr-Esfahani and Johnson, 1991), suggesting that NOX was the source of H2O2 and is expressed as early as the 2-cell stage. Furthermore, NOX has also been detected in rabbit blastocysts (Manes and Lai, 1995), as well as in mouse embryonic stem cells (Sauer et al., 2008). Direct measurement of mRNA transcripts of the components of the cyt b558 complex have been detected in mouse embryos as early as E 5.5 for p22phox mRNA, which is followed shortly thereafter by NOX2 mRNA, expressed at E 9.0, which both continue to increase as development proceeds, as seen by p22phox mRNA expression in fetal liver, spleen, and bone marrow, and NOX2 mRNA expression in liver (Baehner et al., 1999) (Fig.

1.28). The lack of gp91phox/NOX2 detection in whole embryo earlier than E 9.0, may suggest either that: (1) the embryo does not have a functional NOX enzyme until E 9.0, and that studies implying that ROS formation in early embryos is NOX-initiated should be interpreted carefully; or, (2) the NOX enzymes in the preimplantation embryo are extraembryonic only.

Extraembryonic NOX has been detected in mouse trophoblast giant cells, the cells responsible for implantation at the embryo-maternal interface (Bevilacqua et al., 2012), as well as in ectoplacental cones taken from E 7.5 CD-1 embryos (Gomes et al., 2012). The NOX in these extraembryonic cells seems to be functional at this early stage, as they are inducible by phorbol myristate acetate (PMA), a compound used to induce NOX activity (Gomes et al., 2012). These data suggest that NOX is important in the preimplantation embryo, possibly for ROS production

Figure 1.28: Developmental NOX mRNA ontogeny. Expression of various components increase from embryonic day 5.5 to 11 and remain constant from day 14-19. Modified from: (Baehner et al., 1999)

to assist implantation into the maternal uterus. NOX expression in the embryo at the 2-cell stage and at implantation is consistent with the requisite generation of ROS during these developmental periods in order for normal development to proceed (Nasr-Esfahani et al., 1990;

Nasr-Esfahani and Johnson, 1991). In whole embryo, we have observed p22phox mRNA and protein expression at E 12.0, both of which are inducible by MeOH (Miller et al., submitted), and the only evidence to suggest that the embryo proper has a functional NOX enzyme at E 9.0 is that NOX activity is inducible by in utero EtOH exposure as measured by O2•- production using a lucigenin-based activity assay (Dong et al., 2010).

Expression of mRNA transcripts of regulatory proteins p67phox and p47phox have been detected as early as E 7.0 and E 7.5, respectively, and both continue to be expressed throughout development (Baehner et al., 1999). Indeed, mRNA expression of p22phox, p68phox, Noxo1,

Noxa1 and Rac1 are all induced by in utero EtOH exposure as early as E 9.0 (Dong et al., 2010), supporting the notion that NOX is active by E 9.0. NOX proteins have been detected in fetal liver and spleen (Fig. 1.29)

1.7.8 Role of NOX in disease

The importance of the phagocytic NOX is exemplified by chronic granulomatous disease

(CGD), a rare autoimmune disease resulting from mutations in any one of the genes encoding gp91phox/NOX2, p22phox, p67phox or p47phox, resulting in low or absent activity in phagocytes (Heyworth et al., 2003; Vignais, 2002). With CGD, the host cannot mount an adequate oxidative burst to kill microorganisms, resulting in severe and recurrent infections, demonstrating the important protective role of NOX-derived free radicals. The incidence of CGD

Figure 1.29: Developmental NOX protein expression. Expression of various components increase from embryonic day 5.5 to 11 and remain constant from day 14-19. Modified from: (Baehner et al., 1999)

is approximately 1/250,000 people, with a mutation in NOX2 being the most prevalent, accounting for 1/3 of cases, followed by mutations in p47phox that account for 1/4 of cases.

Mutations in p67phox and p22phox are rare, while there have been no reports of a p40phox mutant. CGD is inherited in an X-linked recessive fashion, with approximately 90% of affected subjects being male with a heterozygous mutation on the NOX2 gene located on their X chromosome (p21.1 region) (Hauck et al., 2008). Subtypes of CGD include: (i) the classical X- linked CGD resulting in the absence of the NOX2 subunit, and a complete lack of respiratory burst in patient neutrophils; (ii) autosomal recessive CGD results in the absence of one of the other NOX subunits, and abnormal respiratory burst activity; and, (iii) variant or atypical CGD is diagnosed when the patient has a sufficient amount of the required subunits, but respiratory burst activity is not sufficient to combat infection (Assari, 2006). NOX-derived ROS have also been implicated in a number of disease states including alcohol-induced liver disease (Kono et al.,

2000), central nervous system disorders such as Alzheimer’s disease, multiple sclerosis and

Parkinson’s disease (Gao et al., 2012; Shimohama et al., 2000; Sorce et al., 2012), diabetes

(Sedeek et al., 2012; Vaquer et al., 2012), and cardiovascular diseases such as hypertension, atherosclerosis, cardiac hypertrophy and heart failure (Cave et al., 2006; Zhang et al., 2012).

The wide distribution of tissues affected by NOX-derived ROS mirrors the ubiquitous distribution of NOX isoforms throughout the body, and as such may constitute novel therapeutic targets. To that end, overactive NOX has been linked to enhanced neuroinflammation, a feature common to several neurodegenerative diseases (Gao and Hong, 2008; Glass et al., 2010; Philips and Robberecht, 2011), while suppression of NOX activity both in vitro (Liu et al., 2011; Qian et al., 2008; Qian et al., 2007; Yang et al., 2006; Zhang et al., 2008; Zhang et al., 2010b) and in animal models (Choi et al., 2005; Hu et al., 2010; Wu et al., 2002; Zhang et al., 2010a;

Zhang et al., 2004), has been shown to attenuate neuronal impairment in models of Parkinson’s disease, representing a potentially valuable therapeutic option.

1.7.9 NOX knockout mouse models

Several knockout mouse models specifically targeting deletion of different NOX isoforms have been generated (Table 1.12). The phenotypes presented in the mice are a function of the physiological role of the NOX that has been deleted. For example, deletion of p47phox, a regulatory cytosolic subunit required for the activation of NOX2 in phagocytes results in a mouse model of chronic granulomatous disease (Jackson et al., 1995). Similarly, a knockout of

DUOXA1, a regulatory factor for DUOX2 important for thyroid hormone biosynthesis, results in severe hypothyroidism (Grasberger et al., 2012). NOX3 knockout mice exhibit balance and gait disorders, as NOX3 is important in the development of the inner ear (Paffenholz et al., 2004).

NOX5 is absent in rodents, precluding the generation of a knockout model for this particular

NOX (Fulton, 2009).

1.7.10 Role of NOX in teratogenesis

In utero exposure to EtOH has been shown to increase NOX activity and mRNA expression, along with increased DNA oxidation, apoptosis and embryopathies (Dong et al.,

2010). Inhibition of NOX using the inhibitor DPI completely blocked the toxic effects produced by EtOH, suggesting that EtOH mediates some of its embryopathic effects through NOX activation. To date, no other reports of NOX involvement in birth defects have been published. I have shown in my studies that embryonic exposure to methanol results in increased p22phox mRNA and protein expression, along with an increase in protein oxidation, all of which are decreased by pretreatment with the NOX inhibitor DPI (Miller et al., submitted).

Table 1.12: Genetically modified NADPH oxidase (NOX) mouse models

Background Tissue Subunit Modulation Method Condition Phenotype Reference strain analyzed No change in body weight or vascular development; upregulated catalase (2.6 fold) and MnSOD (1.5 fold); ~1.6 fold ↑NOX activity (NOX-dependent O2 production); 22% impaired Basal acetylcholine-induced endothelium- pBSCX1-LEL dependent vasodilation, prevented with Aortic plasmid with CX1 apocynin (50 μmol/L) or catalase (Dikalova et al., Overexpression vascular promoter containing C57BL/6 treatment; 20% ↓ NO levels; 25% ↓ 2005; Dikalova (4-5 fold) smooth human NOX1 dimerized eNOS; et al., 2010) muscle cDNA ~2.1 fold ↑NOX activity (NOX- dependent O production); ~1.3 fold ↑ Angiotensin 2 H O generation; 129% ↑ blood II-stimulated 2 2 pressure; exacerbated impairment of (0.7 acetylcholine-induced endothelium- mg/kg/day) dependent vasodilation; 126% ↓ NO NOX1 levels; 57% ↓ dimerized eNOS; Vascular Basal 7% ↓ blood pressure vs. wild-type Replace exons 3-6 system (Gavazzi et al., Knockout C57BL/6 Angiotensin ↓ 8-hydroxyguanosine with neocasette (Thoracic II-stimulated immunolabelling; 21% ↓ blood 2006) aorta) (3 mg/kg/day) pressure Normal growth; no difference in body weight, blood pressure or heart rate; Thoracic no change in superoxide production Basal aorta (DHE staining); no difference in mRNA NOX1, 2, 4; no difference in cGMP Replace exons 3-6 (Matsuno et al., Knockout C57BL/6 levels with neocasette 2005) 38% ↓ superoxide production (DHE Angiotensin staining); 13% ↓ blood pressure; no II-stimulated difference in heart rate; no difference in (0.7 NOX2 or 4 mRNA; 1.6 fold ↑ in cGMP mg/kg/day) levels

Table 1.12 (cont’d): Genetically modified NADPH oxidase (NOX) mouse models

Background Tissue Modulation Method Condition Phenotype Reference Tableubunit strain analyzed

C57BL/6 x Impaired otoconial morphogenesis, impaired (Paffenholz NOX3 Mutant Inner ear Basal C3HeB/FeJ balance and gait, normal hearing et al., 2004)

1.7 fold ↑ in catalase; slightly ↓ left ventricular and body weight; decreased cardiac function Overexpression Adenovirus (Ago et al., FVB Cardiomyocytes Basal when aged; NOX4 gene-dependent ↑ in (4 fold) transduction 2010) apoptotic cell death; 1.7 fold ↑ superoxide production (DHE staining)

Nox4 cDNA was cloned No basal dysfunction; protection against downstream of contractile heart dysfunction, hypertrophy and Overexpression C57BL/6 Basal the mouse α- cardiac dilatation during exposure to chronic myosin heavy overload (suprarenal aortic constriction) chain promoter Cardiomyocytes, (Zhang et al., kidney 2010c) No obvious phenotype; exaggerated targeted deletion contractile heart dysfunction, hypertrophy, and of the translation Knockout C57BL/6 Basal cardiac dilatation during exposure to chronic initiation site and overload (suprarenal aortic constriction); NOX4 exons 1 and 2 55% ↓ H2O2 generation

No change in systemic and pulmonary blood pressure, kidney function, cerebral blood flow, Exons 14 and 15 cerebral vasculature and brain structure; flanked by loxP Aorta, lung, protected against transient and permanent (Kleinschnitz Knockout sites followed by C57BL/6 Basal kidney ischemic stroke induced MCAO: et al., 2010) a floxed 67.5% ↓ infarct volumes, nemocyin casette 67% ↑ neurological function and 60% ↑ basal motor function and coordination

LoxP-mediated Normal cardiac phenotype; 50% ↓ p22phox deletion of exon 9 protein but not mRNA; 35% ↓ superoxide (Kuroda et Knockout C57BL/6 Cardiomyocytes Basal and part of exon production (DHE); 55% mortality upon al., 2010) 10 transverse aortic constriction

Table 1.12 (cont’d): Genetically modified NADPH oxidase (NOX) mouse models Background Tissue Referenc Subunit Modulation Method Condition Phenotype strain analyzed e p22hpox was No change in blood pressure or Basal cloned aortic hypertrophy; 2-fold ↑ H2O2; downstream of Angiotensin II- Exacerbation of blood pressure smooth muscle stimulated (0.7 increase; 1.3-fold ↑ aortic Overexpressi α-actin (Weber et C57BL/6 Aortic tissue mg/kg/day) hypertrophy 2.5-fold ↑ H2O2; on (2 fold) promoter Angiotensin II- al., 2005) fragment SMP- stimulated (0.7 Inhibited exacerbation of blood 8 and upstream p22phox mg/kg/day) + pressure increase; 12.5% ↓ aortic of a SV40 poly ebselen (GPx hypertrophy; no change in H2O2; A fragment mimetic) CGD-like immune defect (highly Tyr121His susceptible to necrotizing (Nakano substitution Lung, spleen, pneumonia caused by 1x106 CFU Mutant nmf333 Basal et al., (missense inner ear inhaled Burkholderia cepacia); 2008) mutation) impaired bacterial clearance in respiratory tracts; balance disorder Spontaneous point mutation Neutrophils, (Huang et Mutant C57BL/6 Basal Complete absence of O production at -2 position of bone marrow 2 al., 2000) exon 8 Fertile, no change in weight, normal peripheral blood total and differential leukocyte counts prior to development of severe infections; 75% ↑ infection rate; 2-fold p47phox Basal Neocasette leukocyte count in peritoneal cavity (Jackson Neutrophils, Knockout insertion into C57BL/6 upon thioglycolate injection; no et al., phagocytes exon 7 superoxide production (NBT 1995) reduction, rhodamine) upon PMA stimulation Whole blood killed 10-fold fewer S. Phagocytes in vitro aureus; 2-fold gram stain-identifiable organisms per phagocyte

Table 1.12 (cont’d): Genetically modified NADPH oxidase (NOX) mouse models Background Tissue Subunit Modulation Method Condition Phenotype Reference strain analyzed Embryonic death prior to R58A mutation in exon (Ellson et p40phox Mutant C57BL/6 Neutrophils Basal GD 10 (homozygous 3 al., 2006) mutant); Mouse Noxo1 cDNA Arrest of otoconia was subcloned into genesis (lack of calcium pStec-1 vector [S1] Vestibular carbonate mineralization containing an artiﬁcial C57BL/6 X and and accumulation of an (Kiss et al., NOXO1 Knockout intron and a CMV Basal SJL.F2 cochlear otoconial protein, 2006) promoter upstream epithelia otoconin-90/95); severe and an SV40 polyA imbalance; rescued by signal site downstream NOXO1 transgene of Noxo1 LoxP-mediated Viable, fertile, no overt Conditional (Flaherty et NOXA1 deletion of exon 3-6 of C57BL/6 Inner ear Basal phenotypic knockout al., 2010) NOXA1 abnormalities Growth retardation, lack golgi processing of N- glycans, loss of H O Targeting construct to 2 2 release from thyroid delete a region tissue; severe goiteous (Grasberger DUOXA1/2 Knockout extending from exon 5 C57BL/6 Thyroid Basal congenital et al., 2012) of DUOXA2 to coding hypothyroidism with exon 5 of DUOXA1 undetectable serum T4 and maximally disinhibited TSH levels Abbreviations used: CGD, chronic granulomatous disease; DHE, dihydroethidium; DUOX, dual oxidase; eNOS, endothelial nitric oxide synthase; GD, gestational day; MCAO, mouse model of middle cerebral artery occlusion; MnSOD, manganese superoxide dismutase; NBT, nitroblue tetrazolium; NO, nitric oxide; PMA, phorbol 12-myristate 13-acetate

1.8 Catalase

1.8.1 Catalase gene

Catalase is coded for by a single locus that has been mapped to 11p13. The catalase gene is 34 kilobases (kb) in length and contains 12 introns, 13 exons, and encodes the catalase protein of 526 amino acids (Bell et al., 1986; Quan et al., 1986) (Fig 1.30).

1.8.2 Enzyme structure and catalytic mechanism

Catalase (EC 1.11.1.6) is a tetramer composed of four identical heme-containing subunits, each with an approximate molecular weight of 60,000 kDa. Each subunit has the following four domains: (1) a long N-terminal threading arm; (2) a wrapping loop around the subunit exterior which includes the proximal heme ligand; (3) an anti-parallel eight-stranded β- barrel providing the residues that bind to the distal side of the heme, and (4) an α-helical domain

(Figs. 1.30, 1.31) (Goth et al., 2004; Putnam et al., 2000). Each subunit remains bound to the next by hooking its N-terminal threading arm into the long wrapping loop found in the next subunit (Putnam et al., 2000). Catalase is found primarily in peroxisomes, and catalyzes the decomposition of H2O2 to H2O and O2 by the following 2-step catalytic mechanism (Fig 1.32):

(1) Free catalase reacts with H2O2 to form compound I:

III  IV Enz(Por  Fe )  H 2O2  CpdI(Por  Fe  O)  H 2O

(2) Two electrons are transferred from an electron donor to form H2O and an oxidized product:

 IV CpdI(Por  Fe  O)  H 2O2  Enz  H 2O  O2

Figure 1.30: Primary sequence and secondary structure of human catalase. A 3-D version is shown in Figure 1.31. The components involved in mutations causing catalase deficiency are discussed on page 119.

From: (Putnam et al., 2000)

Figure 1.31: Tertiary structure of human catalase. See text on pages 97 and 119 for discussion.

From: (Putnam et al., 2000)

100

Figure 1.32: Catalytic mechanism of catalase. (a) Hydrogen peroxide is selected by and concentrated at the active site through use of a narrow hydrophobic channel with two fixed water sites next to the active site (His75 and Asn148) and at the other end of the hydrophobic channel (Asp128 and Gln168). The spacing of the water sites is too long to allow a hydrogen bonding network containing only water molecules, and is too short to allow the insertion of an additional water molecule. After coordination to the heme, hydrogen bonding in the hydrophobic channel again becomes disfavoured and should prepare the channel for the next water molecule (continued on next 2 pages) From: (Putnam et al., 2000)

101

Figure 1.32: Catalytic mechanism of catalase (continued) (b) Heterolytic cleavage of the peroxide bond is driven through interactions with the electron- rich active site metal and the electron-withdrawing His75 and Asn148. During compound I formation, a charge relay system acts to increase electron density at the active site and remove charge-charge repulsion between Arg354 and the porphyrin p-cation radical (continued on next page) From: (Putnam et al., 2000)

102

Figure 1.32: Catalytic mechanism of catalase (continued) (c) Oxidation of the second peroxide molecule likely occurs near the active site to a peroxide molecule bound by His75 and Asn148. A similar coordination of the EtOH hydroxyl group would bind the ethyl group within the hydrophobic channel. Reversal of the charge relay system after oxidation brings the molecule back to the resting state. From: (Putnam et al., 2000)

103

In step 2, the electron donor can be either a second molecule of H2O2 representing the antioxidative mode of the enzyme, or another substrate such as EtOH or MeOH, representing the peroxidative mode of catalase (Kirkman and Gaetani, 2007; Putnam et al., 2000). The ability of catalase to metabolize EtOH is minimal, as acatalasemic mice administered 14C-EtOH demonstrated no inhibition of EtOH metabolism, whereas MeOH metabolism was significantly inhibited (Karinje and Ogata, 1990). The reaction with H2O2 is first order, and is entirely dependent upon the concentration of H2O2. At high H2O2 concentrations, the rate of reaction is rapid and catalase preferentially oxidizes 2 molecules of H2O2, while at low H2O2 concentrations

(<10-6M) catalase will use another substrate such as EtOH or MeOH (Hashida et al., 2002;

Masuoka et al., 2003; Nagababu et al., 2003; Percy, 1984). Catalase is expressed in the liver, where the activity is the highest, and is also found in brain, kidney, blood and bone marrow

(Halliwell and Gutteridge, 2007; Ogata, 1991).

1.8.3 Catalase in disease

Mutations in the catalase gene resulting in decreased catalase activity have been associated predominantly with Takahara’s disease, while minor associations have been made for increased risk for diabetes mellitus, hypertension and vitiligo. Decreased catalase results in higher H2O2 concentrations, which can react with metal ions such as iron or copper in the Fenton reaction to form •OH, which can oxidatively damage cellular macromolecules, injure the cell membrane, alter mitochondrial electron transport as well as alter homocysteine metabolism

(Goth and Eaton, 2000; Goth and Vitai, 2003; Heales, 2001; Peiro et al., 2001; Wells et al.,

2009b)

An association of inherited catalase deficiency and diabetes mellitus was first reported by Goth and Eaton (2000), whereby Hungarian families with acatalasemia had a 12% increased

104

risk of diabetes. The suggested mechanism underlying this association is that enhanced H2O2 production can damage the redox sensitive pancreatic β-cells thereby leading to decreased insulin production (Goth and Eaton, 2000; Goth et al., 2001).

The first study to demonstrate a potential association between genetic variations in catalase and hypertension was performed in China in 2001, whereby a single nucleotide polymorphism (SNP) C to T at -773 found in hypertensive individuals demonstrated strong evidence of an association (p < 0.002) with hypertension, defined as a systolic blood pressure over 160 mmHg. This SNP is located in the promoter region of the catalase gene and is involved in transcription factor binding; however, catalase levels were not measured in this study, and it remains unclear how this SNP relates to blood pressure (Jiang et al., 2001).

Vitiligo is a disease involving depigmentation of the skin due to altered melanin biosynthesis. Skin taken from patients with vitiligo have demonstrated consistently reduced levels of catalase compared to normal controls (Schallreuter et al., 1991), which supported the authors’ hypothesis that enhanced ROS are involved in the disorder. However, other studies showed no association between SNP T to C at position 20 of the 5’ flanking region, or T to C at exon 60 of exon 10 and vitiligo (Casp et al., 2002), while another has demonstrated a significant correlation between the polymorphism T/C at codon 389 in exon 9 and vitiligo (Lv et al., 2011)

With regard to neurodegenerative disease, catalase activities were measured in various brain regions from patients with Parkinson’s disease and were found to be reduced in the substantia nigra, caudate and putamen (Ambani et al., 1975). Similarly, oxidative stress has been implicated in Alzheimer’s disease, specifically an accumulation of H2O2 in affected patients.

Therefore, decreased H2O2 detoxification may result in increased neurotoxicity (Goulas et al.,

2002).

105

Acatalasemia is inherited as an incomplete autosomal recessive trait characterized by deficient catalase activity in erythrocytes (Ogata, 1991). Clinical features in patients with acatalasemia include progressive gangrene of the mouth, which is termed ‘Takahara’s disease’

(Takahara, 1971; Takahara, 1952). The low levels of catalase in the blood and in tissues of the mouth permits bacterial H2O2 accumulation which would oxidize hemoglobin, thereby depriving oxygen from the infected area, resulting in necrosis (Takahara, 1971). Cases of acatalasemia have been identified in 11 different countries (Austria, Germany, Hungary, Iran, Israel, Japan,

Korea, Mexico, Peru, Switzerland, US), totaling 113 cases from 59 families (Goth et al., 2004).

Acatalasemia in patients from Japan, Switzerland and Hungary are well characterized, although it remains poorly characterized in other countries. The frequency of acatalasemia is 0.8:1,000 in

Japan, 0.04:1,000 in Switzerland, and 0.05:1,000 in Hungary (Goth, 2001; Goth et al., 2001;

Takahara, 1952).

1.8.4 Embryonic catalase expression

In mice, embryonic catalase is detectable as early as gestational day 8 as 2 distinct bands:

(1) a 12.2 kb RNA band that is observed only during in utero development; and, (2) a 2.4 kb band, which is seen during development, at birth and after birth, and is the mature message of the catalase gene. The 12.2 kb RNA can be spliced and processed to produce the mature 2.4 kb

RNA, although this processing machinery is not fully functional during development, so as the embryo develops and acquires the ability to process this 12.2 kb fragment, more mature catalase can be produced as required, although this 12.2 kb RNA is not detected at all at and beyond birth, while the 2.4 kb RNA is present at birth and in adults (Fig. 1.33) (el-Hage and Singh,

1989). This suggests either that the 12.2 kb RNA is no longer produced, or that the splicing of this fragment occurs rapidly enough that it is no longer detectable. Interestingly, as

106

Figure 1.33: Quantitative analysis of the 12.2 and 2.4 kb catalase-related transcripts during development in mice. Relative changes of RNA levels were quantitated by densitometry of autoradiograms (northerns) taking the value of the adult liver as 1. The vertical lines indicate the standard errors (n=3). A: Quantitative changes of the 12.2 kb RNA in liver (1) and carcass (2). B: Quantitative analysis of the 2.4kb RNA in liver (1) and carcass (2). Note that the time scales for the two tissues studied are independent and different.

From: (el-Hage and Singh, 1989)

107 development proceeds, relative levels of mRNA do not correspond to activity of the enzyme

(Fig. 1.34) (el-Hage and Singh, 1990). Indeed, embryonic catalase is active on day 8, and increases as development proceeds with lowest expression in the heart (Harris et al., 2003), although total activity is less than 10% of maternal activity (Abramov and Wells, 2011a; Winn and Wells, 1999). In humans, catalase activity was not detected in fetal liver at 6 weeks, 34% of fetal liver peroxisomes showed catalase immunoreactivity, which increased to 100% by 8 weeks

(Espeel et al., 1993; Espeel et al., 1997). Enhanced endogenous catalase has been associated with decreased malformations in Sprague-Dawley rat embryos (Sivan et al., 1997), while decreased catalase activity has been associated with increased malformations in diabetic rats

(Cederberg and Eriksson, 1997; Cederberg et al., 2000) and increased malformations associated with methanol exposure (Miller and Wells, 2011). Additionally, exogenous forms of catalase have demonstrated protection in several developmental models; for example, EUK-134, a synthetic SOD/catalase mimetic that scavenges hydrogen peroxide, protected against EtOH- induced limb malformations in vivo (Chen et al., 2004a). Administration of PEG-cat protected against phenytoin teratogenicity and blocked the associated oxidative damage to DNA (Abramov and Wells, 2011a) and protein (Winn and Wells, 1999) in vivo. Similarly in whole embryo culture, PEG-cat treatment alone reduced baseline DNA oxidation in untreated CD-1 mouse embryos, suggesting that PEG-cat reduces endogenous ROS, while pretreatment prior to phenytoin treatment blocked phenytoin-initiated embryonic DNA oxidation and structural embryopathies in whole embryo culture (Winn and Wells, 1995). PEG-cat pretreatment also blocked structural embryopathies in catalase-deficient mice treated in culture with phenytoin

(Abramov and Wells, 2011b). In vivo, maternal treatment with PEG-cat not only enhances catalase activity in the dam, but also increases embryonic catalase activity approximately 2-fold,

108

Figure 1.34: Developmental profile of the catalase-specific RNA and enzyme activity in the carcass and liver during development in mice.

Both plots represent values relative to the values of the adult liver.

From: (el-Hage and Singh, 1990)

109 remaining elevated for about 24 hours (Fig. 1.35) (Abramov and Wells, 2011a; Winn and Wells,

1999), while in embryo culture, PEG-cat addition directly into the culture medium increases embryonic catalase activity (Abramov and Wells, 2011a).

1.8.5 Transgenic human catalase-expressing mouse

A transgenic mouse expressing human catalase in addition to murine catalase was generated on a C57BL/6 background to investigate the contribution of catalase in the aging process (Chen et al., 2003). This was done by inserting an 80 kb fragment of the human catalase gene from a P1 clone, which contains (i) the 33 kb human catalase gene, (ii) 41 kb of the 5’ flanking region, and (iii) 6 kb of the 3’ flanking region (Fig. 1.36). By inserting the entire gene along with its flanking regions, enhancer or promoter elements necessary to express the catalase gene in a tissue-specific manner that mimics the expression of endogenous catalase are included.

The resulting tissue expression and subcellular localization was similar to that of endogenous mouse catalase, with an approximate 4-fold increase in tissue catalase activity (Figs. 1.37, 1.38).

In order to identify these mice, a polymerase chain reaction (PCR) using primers that target exon

13 of the human catalase gene can be used, which will amplify a 450 bp region of the gene.

When run on an agarose gel, the presence of a band at 450 bp identifies a transgenic mouse, while the absence of a band identifies a wild-type (Fig. 1.39). The mice used in my studies are represented as homozygous Tg(CAT)+/+ mice in figures 1.37 and 1.38, herein denoted as ‘hCat’.

Despite a significant increase in catalase activity in hCat embryos compared to their

C57BL/6 wild-type controls, levels of embryonic catalase activity in both strains are less than

10% of maternal hepatic levels (Fig 1.40). Additionally, embryonic catalase activity has been shown to be a determinant of risk for phenytoin-initiated embryopathies in embryo culture.

110

Figure 1.35: Time course of embryonic catalase activity after maternal injection of PEG- catalase. PEG-catalase (10 kU/kg) was administered 4, 8, 18 and 24 h prior to the scheduled time of phenytoin injection as in the teratological studies on either days 11 and 12, or day 12 only. Dams were sacrificed at the scheduled time of phenytoin treatment on gestational day 12, and catalase activity was determined. The number of embryos is given in parentheses. *Indicate a difference from the PBS-treated controls (* p < 0.05). From: (Winn and Wells, 1999)

111

Figure 1.36: Map of the P1 bacteriophage clones containing the entire human CAT gene. The transcriptional start site (+1) and 13 exons (black boxes) of the human CAT gene are shown. The 33kb human catalase gene is boxed in yellow. See text on page 109 for discussion. From: (Chen et al., 2003)

112

Figure 1.37: Catalase expression in heterozygous Tg(CAT)+/o, homozygous Tg(CAT)+/+ and wild-type mice

(Legend on next page)

113

(A) Catalase activity in various tissues of the Tg(CAT)+/o, Tg(CAT)+/+, and wild-type mice. Data are expressed as means + SEM from 10 mice. The data were analyzed by ANOVA followed by a Student–Newman–Keuls multiple comparison test. Catalase activity levels in tissues of the Tg(CAT)+/+ mice were significantly different from both wild-type and Tg(CAT)+/o mice at the P < 0.01 level (**). The differences between the catalase activity levels in tissues of wild-type and Tg(CAT)+/o mice were significant at the P < 0.01 level (*). (B) Levels of catalase protein in various tissues of the Tg(CAT)+/+ and wild-type mice were determined by western blot analysis using a polyclonal antibody against bovine catalase. A representative western blot is shown in which the amount of protein loaded was 3 μg for liver, 4 μg for kidney, and 20 μg for the rest of the tissues. β-Actin was used as a loading control. The migration of molecular standards is indicated on the left. Lanes with odd numbers are for wild- type mice; lanes with even numbers are for Tg(CAT)+/+ mice. The graph shows the comparison of catalase activity and protein levels in various tissues of the Tg(CAT)+/+ mice expressed as percentage of the wild-type control. The data are expressed as means + SEM from five mice. (C) Catalase activity in homogenate and subcellular fractions of liver in wild-type and Tg(CAT)+/+ mice. The data are expressed as means + SEM from five mice From: (Chen et al., 2004b)

114

Figure 1.38: Distribution of catalase in the mitochondria and peroxisomes of Tg(CAT)+/+ and wild-type mice. (Legend on next page)

115

(A) The levels of catalase protein in the liver homogenates and the peroxisomal and mitochondrial fractions of Tg(CAT)+/+ (Tg) and wild-type (WT) mice were determined using Western blots. Three micrograms of protein from the homogenate or peroxisomes and 10 μg of protein from the mitochondria were subjected to Western blot analysis, and the membranes were probed with antibodies to catalase, β-actin, and ATPase. (B) The catalase protein levels were determined from the Western blots corrected for loading controls. Data are expressed as means + SEM from three mice and were analyzed by Student’s t test. Catalase protein levels were significantly higher in the homogenate and peroxisomal fraction from the livers of Tg(CAT)+/+ mice compared to wild type mice (*P < 0.01). There was no significant difference in the catalase protein levels in the mitochondrial fraction. From: (Chen et al., 2004b)

116

Figure 1.39: Representative gel for genotyping showing DNA bands for mice expressing human catalase (hCat) and their proximate C567BL/6 wild-type (WT) controls.

Primers targeting exons 12/13 of the human catalase gene amplify in the hCat mice a 450 bp fragment, which is absent in the WT animals. The presence of a 450 bp band identifies the hCat genotype, while no band signifies WT.

From: (Miller and Wells, 2011)

117

Figure 1.40: Embryopathies in wild-type (WT) mouse embryos and embryos expressing hCat exposed to phenytoin or its vehicle. (Legend on the next page)

118

(A) WT embryos with normal catalase activity and hCat embryos expressing hCat (high activity) were cultured and assessed for embryopathies.

(B) Embryos represented in panel A were used to measure endogenous catalase activity in each individual embryo using the Ferrous Oxidation of Xylenol orange (FOX) assay.

(C) Catalase activity was measured in maternal liver samples using the FOX assay and compared with activities in WT and hCat embryos. VEH, vehicle; PHT, phenytoin. The beta symbol indicates a difference from VEH-exposed embryos of the same genotype (p < 0.05); the asterisk indicates a difference from WT embryos of the same treatment group (VEH or PHT) (p < 0.05).

(n, N) = (litters, embryos)

From: (Abramov and Wells, 2011b)

119

1.8.6 Acatalasemic mouse

An acatalasemic mouse with deficient catalase activity was generated in 1964 by screening 12,000 mice that were irradiated to generate a mutation of the catalase gene (Feinstein et al., 1964). The effective mutation created a single nucleotide transversion mutation (G  T) located in the third position of codon 11, changing a glutamine to a histidine, resulting in loss of protein activity. This amino acid is located within a region that forms the first major α-helix in the N-terminal arm of the catalase subunit, thereby rendering the catalase protein unstable. Since each subunit remains bound to the next by hooking its N-terminal threading arm into the long wrapping loop found in the next subunit (Putnam et al., 2000), it is possible that the instability is created by an inability to remain stably bound to the next subunit. Upon analysis of transcriptional and translational products between aCat and WT mice, it was determined that mRNA levels were similar in all tissues, so the acatalasemic phenotype is not a result of changes at the transcription level, but is rather at the level of mRNA translation and/or catalase protein turnover (Shaffer et al., 1987). Catalase expression is quite variable among different tissues in aCat mice, as they demonstrate approximately 1%, 20% and 50% catalase activity in red blood cells, kidney, and liver tissue, respectively, compared to wild-type mice (Aebi et al., 1968). Our laboratory has demonstrated that maternal aCat activity is substantially lower than their

C3HeB/FeJ wild-type (C3H WT) controls (Fig. 1.41). Glutathione peroxidase activity is upregulated to a variable extent in some tissues, demonstrating a potential compensatory mechanism to regulate H2O2 detoxification (Fig. 1.42), but not sufficient to protect against the

ROS-initiating teratogen phenytoin (Abramov and Wells, 2011a). Catalase activity in the aCat embryos is substantially lower than their C3H WT controls, while maternal pretreatment with

PEG-cat resulted in a sustained elevation of embryonic catalase activity, and was a determinant of risk for

120

Figure 1.41: Maternal tissue activity of catalase in aCat mice. The activity of catalase in the various maternal tissues of WT and aCat mice was determined using a ferrous oxidation of xylenol orange (FOX) assay. Each group consisted of samples from five animals. Asterisks indicate a difference from the respective WT group for the same organ (* p < 0.05). From: (Abramov and Wells, 2011b)

121

Figure 1.42: Maternal tissue activities of catalase and glutathione peroxidase (GPx) in aCat mice. The activity of GPx in the various maternal tissues of WT and aCat mice was determined using the a glutathione peroxidase activity assay. Each group consisted of samples from five animals. Asterisks indicate a difference from the respective WT group for the same organ (* p < 0.05). While catalase-deficient mice exhibited a small but significant compensatory increase in GPx, this was not sufficient to protect them from ROS. Modified from: (Abramov and Wells, 2011b)

122 phenytoin embryopathies in (Fig. 1.43). To maintain a colony of these mice, they are compared to a proximate C3H WT mouse, as opposed to a congenic strain, due to the inability to appropriately distinguish a homozygous mutant from a heterozygous mutant. To genotype these mice, primers are targeted towards the region that encodes for the mutation, thereby amplifying this region creating a 493 bp PCR band (Fig. 1.44). These PCR products are subjected to enzymatic digestion by NdeI, which recognizes the G to T point mutation and cuts the 493 bp fragment into 2 fragments of 255 and 237 bp. However, cutting of the wild-type 493 bp PCR fragment is incomplete, resulting in both a wild-type band at 493 bp, and a mutant band between

237-255 bp, characteristic of an aCat heterozygote.

123

Figure 1.43: Prevention of phenytoin embryopathies in aCat mice with catalase protein therapy. (Legend on next page)

124

(A) aCat and WT embryos were cultured with phenytoin and assessed. (B) Embryos represented in panel A were used to measure endogenous catalase activity in each individual embryo using the FOX assay. (C) Catalase activity was measured in maternal liver samples using the FOX assay and compared with activities in WT and aCat embryos. The alpha symbol indicates a difference from the respective WT embryonic or maternal genotype (p < 0.05); the asterisk indicates a difference from aCat embryos exposed only to phenytoin (p < 0.05); the star symbol indicates a difference from activity in maternal livers with the same genotype (p < 0.05). (n, N) = (litters, embryos). From: (Abramov and Wells, 2011b)

125

Figure 1.44: Representative gel for genotyping showing DNA bands for catalase-deficient (acatalasemic, aCat) mice and their proximate C3HeB/FeJ WT controls. Primers target a 493 bp segment of the mouse catalase gene, and PCR products are then subject to NdeI enzymatic digestion resulting in the presence of a 237/255 bp fragment. The presence of two bands identifies the aCat genotype, while one band signifies WT. From: (Miller and Wells, 2011)

126

1.9 8-Oxoguanine Glycosylase 1 (OGG1)

1.9.1 8-oxoguanine

Physiological levels of ROS are important in the developing embryo for signal transduction to occur, however conditions of oxidative stress can result in oxidative damage to

DNA, and structural or functional birth defects. The reaction of •OH with DNA can result in several types of lesions (reviewed in Section 1.5.2.1), the most prevalent being the 7,8-dihydro-

8-oxoguanine (8-oxoG) lesion (Wells et al., 2009b), often measured as the nucleoside 8-oxo-2’- deoxyguanosine (8-oxodG). The low redox potential of guanine makes it particularly vulnerable to oxidative attack (Neeley and Essigmann, 2006). 8-oxodG is not only a biomarker for oxidative stress, but is also a pathogenic lesion, which will be described further in section 1.9.

1.9.2 Consequences of 8-oxoG formation

During DNA replication, the presence of 8-oxoG can result in the formation of G:C to

T:A transversions, as the oxidized guanine can functionally mimic thymine in its syn conformation, stably base pairing with adenine with 2 hydrogen bonds (Shibutani et al., 1991)

(Fig 1.45, top). When paired with cytosine, 8-oxoG is in the normal anti conformation, and forms a stable base pair with 3 hydrogen bonds (Fig 1.45, bottom). While 8-oxoG can lead to mutations, it may also directly initiate embryopathies via non-mutagenic mechanisms, possibly including altered gene transcription, which can occur in several ways as discussed in Section

15.2.1.

127

Figure 1.45: Base pairing properties of 8-oxoG residues in DNA. The 8oxoG exists predominantly in its keto form at physiological pH, which forms a stable Watson-Crick base pair with cognate cytosine along with the normal anti conformation about the N-glycosylic bond. The keto form of 8oxoG also pairs with non-cognate adenine to form a stable Hoogsteen mispair, adopting the syn conformation about the N-glycosylic bond. From: (Klungland and Bjelland, 2007)

128

1.9.3 Base excision repair (BER) pathway

BER is the primary and essential repair system involved in the removal of damaged DNA bases. Two pathways exist for BER: short-patch, and long-patch (Fig. 1.46). Short-patch repairs

1 nucleotide (Dianov et al., 1992), while long-patch repairs 2-13 nucleotides (Frosina et al.,

1996; Klungland and Lindahl, 1997).

The initiating steps in these two pathways are similar (Robertson et al., 2009), beginning with recognition and removal of a damaged base by a DNA glycosylase, creating an apurinic or apyrimidinic site (AP site). The AP site in the DNA backbone is processed by AP endonuclease

1 (APEX1), which creates a single-stranded DNA nick 5’ to the AP site by hydrolyzing the phosphodiester bond next to the AP site, resulting in a 3’-OH and a 5’-deoxyribose-5-phosphate

(5’-dRP) (Memisoglu and Samson, 2000). The remaining steps are different for the short- and long-patch pathways.

1.9.3.1 Short-patch BER pathway

In the short-patch BER pathway, DNA polymerase β (DNA pol β) removes the 5’-dRP residue and fills the gap with the correct nucleotide. DNA ligase III then seals the nick with the help of X-ray repair complementing defective repair in Chinese hamster cells 1 (XRCC1), thereby restoring DNA integrity (Coppede, 2011).

1.9.3.2 Long-patch BER pathway

DNA pol β adds several nucleotides from the 3’ end of the nick, displacing the strand for several nucleotides creating a ‘flap’ structure and consists of the 5’-sugar phosphate end. DNA

Pol δ or ε can also carry out this step (Stucki et al., 1998). Flap endonuclease 1 (FEN1) removes

129

Figure 1.46: BER pathway overview (Legend on next page)

130

Diagram illustrating the DNA base excision repair (BER) pathway. In this pathway, a DNA glycosylase catalyzes the removal of a damaged base, creating an abasic (AP) site. The APEX1 endonuclease catalyzes the incision of the DNA backbone leaving behind a 5’-deoxyribose phosphate (5’dRP, indicated with a black circle). In short-patch BER, polymerase β (Pol β) displaces the AP site and polymerizes DNA to fill in the gap. Pol β then catalyzes the removal of the displaced AP site, and the ligase III/XRCC1 complex seals the ends. If the 5’-dRP is refractory to the action of Pol β, then an additional synthesis of DNA (long-patch BER) is required to displace the modified 5’-sugar phosphate as part of a flap (in grey), which is then removed by flap endonuclease (FEN1). Pol β adds the first nucleotide into the gap and is substituted by polymerase δ/ε which continues long-patch BER. DNA ligase I (LIG1) completes the long-patch pathway. Abbreviations used: AP, abasic (apurinic, apyrimidinic); APEX, AP endonuclease 1; BER, base excision repair; FEN, flap endonuclease 1; LIG1, DNA ligase I; dRP, deoxyribose phosphate; Pol β, DNA polymerase β; XRCC1, X-ray repair complementing defective repair in Chinese hamster cells 1 From: (Coppede, 2011)

131 the resulting ‘flap’ structure of displaced oligonucleotides (Klungland and Lindahl, 1997), producing a nick that can be ligated by DNA ligase I (LIG1), completing the repair pathway

(Coppede, 2011).

1.9.4 OGG1 gene

The human OGG1 (hOGG1) gene maps to chromosome 3p26, and was originally cloned in yeast as a functional homolog of the bacterial MutM gene which encodes for the Fpg protein

(Boiteux et al., 1987). Although these two genes shared no significant structural homology, the identification of ogg1 was based on its ability to minimize the mutator phenotype of a strain of E.

Coli with inactive Fpg and mutY (Nash et al., 1996; van der Kemp et al., 1996). Subsequently, several groups identified the human ogg1 gene as both a structural and functional homolog to the yeast ogg1 (yOGG1) gene (Aburatani et al., 1997; Arai et al., 1997; Bjoras et al., 1997; Kuo and Sklar, 1997; Lu et al., 1997; Radicella et al., 1997; Roldan-Arjona et al., 1997), sharing

38% homology (Boiteux and Radicella, 2000) (Fig. 1.47).

Two forms of mRNA have been identified that encode for two hOgg1 peptides, which arise from alternative splicing. A 1.7 kb mRNA encodes for a 345 amino acid peptide termed α- hOgg1, while the 2.1 kb mRNA encodes for a 434 amino acid peptide termed β-Ogg1 (Fig.

1.48). Both forms are present in most tissues, however the α form is the most abundant and is found in the nucleus, while the β form is found in the mitochondria (Aburatani et al., 1997;

Kohno et al., 1998; Monden et al., 1999; Nishioka et al., 1999).

132

Figure 1.47: Alignment of amino acid sequences of α- and β-human (h), mouse (m) and yeast (y) Ogg1 proteins. The amino acids identical in the four sequences are shown (*). The consensus sequence for the Helix-hairpin-Helix motif, mitochondrial targeting (MTS), and nuclear localization sequence (NLS). Highly conserved regions are boxed. Regions III and IV contain the active site, whereas I, II, and V have unknown function. The catalytically important amino acids, Lys241 (yOgg1) or Lys249 (hOgg1) and Asp230 (yOggg1) or Asp268 (hOgg1) are also shown using a black triangle and a black star, respectively. The site of divergence between α- and β-hOgg1 generated by alternative splicing is shown by an arrow. Cleavage site for the mitochondrial β-hOgg1 is also indicated by dark triangles. (Boiteux and Radicella, 2000)

133

Figure 1.48: Splice variants of the hOGG1 primary transcript. Arrows indicate the direction of transcription for the OGG1 gene. MTS, putative mitochondrial targeting signal; NLS, nuclear localization sequence. See text on page 131 for details. From: (Boiteux and Radicella, 2000)

134

1.9.5 OGG1 protein structure

Humans express two splice-variant forms of the OGG1 protein, which share identical first

316 amino acids, but differ at their carboxy terminus. The open reading frame (ORF) of α- hOgg1 encodes the 345 amino acid 39 kDa α-OGG1 protein, while the ORF of β-Ogg1 encodes the 424 amino acid 47 kDa β-OGG1 protein. Although a mitochondrial targeting signal (MTS) spans the first 30 amino acids of the N-terminal end of both isoforms, only the β-OGG1 protein is localized to the mitochondria. This is because the α-OGG1 contains a nuclear localization sequence (NLS) at the C-terminal end from amino acids 335-342 that suppresses the MTS, thereby localizing it to the nucleus (Fig. 1.49). This NLS is absent in β-OGG1, resulting in its localization to the mitochondria (Nash et al., 1996; Nishioka et al., 1999).

1.9.6 OGG1 catalytic mechanism

8-OxoG enhances bending in DNA that distorts the double-helix, which is recognized by

OGG1 (Miller et al., 2003). Once recognized, OGG1 swivels the 8-oxoG residue out of the helix into its recognition pocket. Lys249, the catalytic residue for OGG1, attacks the glycosidic bond of 8-oxoG thereby expelling the oxidized residue, forming a covalently linked enzyme-substrate intermediate (Fig 1.50). This intermediate undergoes rearrangement to a Schiff base, which undergoes a lyase reaction to expel the 8-oxoG base via β-elimination resulting in strand nicking of the DNA backbone (Bruner et al., 2000).

135

Figure 1.49: Structure of the α- and β-hOgg1 isoforms. See text on pages 131 and 133 for discussion. The α- and β-hOgg1 isoforms are composed of 345 and 424 amino acids, respectively. The catalytic lysine 249 is indicated. Alanine 316 marks the last common residue between forms α and β. Abbreviations used: NLS, nuclear localization sequence; MTS, mitochondrial targeting sequence.

From: (Boiteux and Radicella, 2000)

136

1.9.7 Developmental OGG1 expression

Both α-OGG1 and β-OGG1 mRNAs have been detected in human fetal brain and liver

(Nishioka et al., 1999). In rats, gestational day (GD) 16 whole fetuses demonstrated 3-15 fold higher mRNA compared to adult liver, with 2-3 fold higher activity (Riis et al., 2002). Primers used to detect rat OGG1 mRNA targeted exons 2 and 3 of the rat OGG1 gene, suggesting that both α- and β-OGG1 mRNA were detected. When examined from GD 17 until 30 months postnatally, OGG1 protein and activity measured in GD 17 cerebral cortex were found to be at their highest and declined in an age-dependent manner (Chen et al., 2002). OGG1 mRNA was measured in rat frontal cortex, and there was no change in mRNA levels on postnatal days 1, 7 or in the adult (Verjat et al., 2000). These differential tissue expression profiles suggest varying tissue-specific requirements for BER during development and in the brain. The importance of

BER during earlier stages of development is highlighted by knockout mouse models of specific

BER enzymes, which result in lethality (Table 1.13).

1.9.8 OGG1 knockout mouse

A mouse lacking ogg1 was generated by 2 different groups (Klungland et al., 1999;

Minowa et al., 2000). To disrupt the ogg1 gene, a neomycin resistance gene polyadenylation signal (Neo) was inserted to replace a portion of DNA containing a highly conserved helix-

137

Figure 1.50: Enzymatic mechanism of hOGG1. The enzyme uses an active-site lysine residue (Lys 249) to attack C-1’ of the deoxyribose sugar and expel the 8oxoG base. From: (Bruner et al., 2000)

138

Table 1.13: Mouse models of Base Excision Repair (BER) pathway enzymes resulting in lethality Knockout Mouse Phenotype Reference Model Pyknotic nuclei at E 6.5; (Ludwig et al., 1998; APE1 Embryonic lethal at E 7.5 Meira et al., 2001) Embryonic lethal at E 18.5; growth (Gu et al., 1994; DNA Pol β retardation, respiratory failure and Sugo et al., 2000) defective neurogenesis Embryonic lethal at E 4.5; FEN +/- (Kucherlapati et al., FEN1 are normal; FEN -/- blastocysts fail 2002; Larsen et al., to form inner cell mass 2003) Embryonic lethal at E 8.5. XRCC1 Increased DNA breakage, (Sato et al., 2003) apoptosis prior to death Embryonic lethal at E 15.5 - 16.5; Lig I defective fetal erythropoiesis (Bentley et al., 1996) resulting in severe anemia Abbreviations used: APE1, AP Endonuclease 1; DNA Pol β, DNA Polymerase β; FEN1, Flap endonuclease 1; XRCC1, X-ray repair complementing defective repair in Chinese hamster cells 1; Lig I: DNA Ligase I.

139 hairpin- helix (HhH) motif required for DNA glycosylase/AP lyase activity (Bjoras et al., 1997;

Lu et al., 1997) (Fig. 1.51), resulting in a loss of OGG1 activity in a variety of tissues (Fig.

1.52). Several studies demonstrate that OGG1-deficient mice accumulate 8-oxoG following exposure to oxidative stress; however, no phenotype has been published for these mice (Table

1.14). My results described later in this thesis provide the first evidence of a phenotype relating to neurodevelopmental deficits.

1.9.9 OGG1 in teratogenesis

As previously mentioned, 8-oxoG has been demonstrated to be not only a biomarker for oxidative stress, but also a pathogenic lesion, as demonstrated in pregnant DNA repair knockout mice treated with the ROS-initiating drug methamphetamine, whereby the null progeny deficient in OGG1 exhibit enhanced motor coordination deficits compared to their DNA repair-normal wild-type littermates (Wong et al., 2008). Similarly, CSB-deficient mice exhibited enhanced metamphetamine-initiated fetal brain DNA oxidation measured as 8-oxoG and postnatal motor coordination deficits, while their wild-type littermates did not (McCallum et al., 2011c).

Although there is a large body of literature that describes DNA repair, there is still a limited understanding about the explicit involvement of specific DNA repair pathways in protecting the embryo from DNA-damaging agents, and particularly from ROS-initiating teratogens.

140

Figure 1.51: Targeted disruption of the murine OGG1 locus. (a) Physical map of the genomic DNA containing the OGG1 gene. The location of the helix– hairpin–helix motif is indicated; amino acids identical in yeast, human, and murine OGG1 are boxed; the murine amino acid sequence shown is identical to the human sequence except for the two residues underlined; the Lys and Asp residues associated with DNA glycosylase/AP lyase activity are shown with an arrow and a circle, respectively. Genomic fragments were subcloned on either side of the Neo gene to generate a construct that deleted ~4.6 kb including this motif at the targeted locus. Genomic DNA is shown by boxes; vector sequences are shown by a line. Restriction digest with EcoRI gives rise to a ~12-kb fragment at the wild-type locus and a ~5-kb fragment at the targeted locus that are detected by hybridization with a 5’ flanking probe (striped box).

(b) Representative wild-type (+/+), heterozygous (+/-), and ogg1 null (-/-) live-born F2 progeny genotyped by EcoRI digestion of tailsnip DNA and hybridization with the 5’ probe.

From: (Klungland et al., 1999)

141

Figure 1.52: Cleavage activity of OGG1 on 8-oxoG in various tissues from ogg1 null mice. (Legend on next page)

142

(a) Nuclear extracts were prepared of the organs indicated from wild-type (+/+), heterozygous (+/-), and ogg1 null (-/-) mice and incubated with a double-stranded oligonucleotide substrate (b) with a 8-oxoG:C mismatch and 32P-labeled at the 5’ end of the 8-oxoG-containing strand. Reaction products were analyzed by PhosphorImager after denaturing PAGE. OGG1 activity excises the 8-oxoG and cleaves the 32P-labeled, 8-oxoG containing strand (49 nt) 3’ of the lesion; subsequent cleavage of the terminal sugar phosphate by HAP1 endonuclease in the extract produces a 21-nt, 32P-labeled species. Lower molecular weight bands result from exonuclease degradation at the exposed 3’ residues of the 21-nt cleavage product. Levels of two control enzyme activities, uracil-DNA glycosylase and hNth1, were normal in all extracts. (c) comparative activity on 8-oxoG DNA lesions in extracts from ogg1 null mice. Nuclear extracts prepared from testes of wild-type (WT), heterozygous (Het.) and ogg1 knockout (KO) mice were assayed for cleavage at 8-oxoG:C by using appropriate DNA substrates. Adapted from: (Klungland et al., 1999)

143

Table 1.14: Phenotype of OGG1-deficient mice and their response to oxidative stressors Phenotype Treatment Effect Reference 8-oxoG 1.7 fold ↑ in 8-oxoG accumulation vs. (Klungland et None accumulation WT in liver nuclear extracts al., 1999) 3-fold 8-oxoG accumulation at 6 (Minowa et al., None weeks, 7-fold accumulation at 14 2000) weeks compared to WT (de Souza- None 9-fold ↑ in liver Pinto et al., 2001) (Arai et al., KBrO Increased 8-oxoG accumulation in liver 3 2002) Increased 8-oxoG in GD 12 embryo, Miller et al., EtOH fetal brain unpublished Mutation No change at 10 weeks in testes, 2-3 (Klungland et None frequency fold increase in liver at 10 weeks al., 1999) (Minowa et al., None 2.3 fold ↑ in liver at 16-20 weeks 2000) No activity in nuclear extracts from (Klungland et Repair activity None liver, testes, lung, spleen, liver, al., 1999) mitochondrial liver extracts; (Arai et al., None No activity in liver 2002; Minowa et al., 2000) (de Souza- No activity in mitochondrial liver None Pinto et al., extracts 2001) (Klungland et Tumor al., 1999; None None Formation Minowa et al., 2000) (Kunisada et UVB light Enhanced skin tumor formation al., 2005) Structural Miller & Wells, None None Embryopathies unpublished Increased embryopathies in KO > HET Miller & Wells, EtOH > WT unpublished Behavioural Miller & Wells, None Learning and memory deficits Deficits unpublished Miller & Wells, EtOH Learning and memory deficits unpublished

Abbreviations used: 8-oxoG, 8-oxo-2-deoxyguanosine; GD, gestational day; HET, heterozygous; KO, knockout; WT, wild-type

144

1.10 Ethanol (EtOH)

EtOH is generally consumed for its psychoactive properties, acting as a central nervous system depressant at higher doses (above 100 mg/dl or 23.77 mM), and a stimulant at lower levels (50 mg/dl or 11.89 mM) (Pohorecky and Brick, 1988). This section will provide an overview of EtOH pharmacokinetics, factors that affect the pharmacokinetic profile, consequences of EtOH intoxication, the effects of in utero EtOH exposure and the role of oxidative stress in the mechanism.

1.10.1 Pharmacokinetics

EtOH is a small molecule with high water solubility that can easily traverse biological membranes by simple diffusion according to Fick’s first law of diffusion, and through the same transmembrane protein channels that permit the passage of water (Kalant, 1971; Wallgren and

Barry, 1970). Given these properties, EtOH can cross any bodily surface including skin

(Anderson et al., 1991), lung (Goldstein and Pal, 1971) and gastrointestinal (GI) tract.

Absorption across the GI tract is dependent upon the dose ingested, the rate of gastric emptying, the presence of food and the type of alcohol consumed (Kalant, 1971; Wallgren and Barry,

1970). EtOH-metabolizing enzymes including CYP2E1, catalase and various isozymes of ADH, have been characterized in GI mucosal cells, and variations in ADH contribute to modulation of

EtOH metabolism (Seitz and Poschl, 1997).

EtOH distributes from the blood into total body water, with concentrations in tissues quickly reaching equilibrium with plasma EtOH (Kalant, 1971). Tissues with greater blood flow such as the kidneys and brain reach equilibrium quicker than those tissues with slower blood flow such as resting muscle (Wilkinson and Rheingold, 1981). During pregnancy in humans,

145

EtOH easily crosses the placenta, with the blood alcohol concentration (BAC) in the fetus achieving similar levels, and following a similar time course as maternal BAC, albeit with a slower rise and fall in amniotic fluid compared to maternal blood, demonstrating a high degree of

EtOH exchange between these compartments (Brien et al., 1983). Similar results have been demonstrated in pregnant rats (Hayashi et al., 1991; Kesaniemi and Sippel, 1975) and humans

(Brien et al., 1983; Idanpaan-Heikkila et al., 1972). The role of the acetaldehyde metabolite in the developmental toxicity of EtOH is not clear; in vivo studies have detected low levels of acetaldehyde, in the range of 0.2-4 μM, in fetal blood and amniotic fluid, and 11-35 μM in maternal blood (Hayashi et al., 1991), while studies in whole embryo culture found that acetaldehyde is embryotoxic in concentrations as low as 25 μM (Campbell and Fantel, 1983).

Table 1.15 outlines the behavioural effects of increasing amounts of alcohol in adults

(Pohorecky and Brick, 1988). Comparing plasma EtOH concentrations between mice and humans, the peak plasma EtOH concentration achieved in CD-1 mice following administration of 4 g/kg i.p is approximately 400 mg/dL (100 mM) (Blakley and Scott 1984b), which is a nearly lethal concentration in humans. Although the concentration-response curve for mice and humans may be shifted, such that a higher dose is required in a mouse to achieve the same effect observed in humans. Indeed, the lethal dose of EtOH in mice is twice as high as that in humans

(Table 1.16). Interestingly, the lethal plasma concentrations in humans differ in the studies published in Tables 1.15 (86 mM) and Table 1.16 (118 mM). The wide variation in published estimations necessitates caution in interpreting these reports. Peak plasma EtOH levels ranged from 21 – 47 mM in alcoholic women who bore children with FASD, by consuming anywhere from 3 – 6 drinks per day (May et al., 2008).

146

Table 1.15: Blood alcohol concentration (BAC) and its effects on human behaviour

*140lb/64kg woman, ŧ1 standard drink (13.5g EtOH=355ml 5% beer=146ml 10% wine=44ml 40% spirit Modified from: (Pohorecky and Brick, 1988)

147

Table 1.16: Approximate mean lethal dose of EtOH for humans and some common laboratory animals.

Species BAC (%) (g/100ml) mM Oral dose (g/kg) Rat 0.9 214 9 Mouse 0.8 190 8 Dog 0.55 118 5 Chicken 0.55 130 5 Human 0.5 118 4 Guinea pig 0.5 118 4 Cat 0.5 118 4

Modified from: (Wallgren and Barry, 1970)

148

The same dose of EtOH given to different individuals produces widely variable BAC

(Bloom et al., 1982). Some factors contributing to this variability include the type and concentration of the alcoholic beverage, the presence of food in the stomach, differences in the expression of EtOH metabolizing enzymes, volume of distribution (Vd), and varied amounts of water and fat in the body (Pohorecky and Brick, 1988). Since EtOH equilibrates in total body water, the Vd is equal to total body water. Therefore, in order to accurately determine BAC, it must be corrected for by the total body water of that individual. Indeed, nearly identical BACs were obtained when corrected for by total body water in mouse strains exhibiting different proportions of body water (Faulkner et al., 1990).

Numerous studies investigating gender differences in EtOH metabolism in humans have reported contradictory results. Despite a higher elimination rate in females, they have been reported to achieve higher peak BACs compared to males (Ammon et al., 1996; Arthur et al.,

1984; Cole-Harding and Wilson, 1987; Frezza et al., 1990; Jones and Jones, 1976; Martin et al., 1985; Mishra et al., 1989; Niaura et al., 1987; Radlow and Hurst, 1985; Sutker et al.,

1983). The higher BAC in females despite their faster elimination rate is thought to be due to a lower Vd in females (Table 1.17) (Goist and Sutker, 1985; Kalant, 1996; Marshall et al., 1983;

Martin et al., 1985; Mishra et al., 1989).

Table 1.17: Estimated volume of distribution in men and women

V (L/kg) d Reference Women Men 0.56 0.68 (Arthur et al., 1984) 0.43 0.58 (Goist and Sutker, 1985) 0.59 0.73 (Marshall et al., 1983)

On the other hand, no difference in BAC was observed when adjusted for by body fat/water content, body weight or liver size (Goist and Sutker, 1985; Kalant, 1996; Kwo et al.,

149

1998; Mills and Bisgrove, 1983; Mulvihill et al., 1997; Mumenthaler et al., 1999; Smith et al.,

1993; Sutker et al., 1983; Thomasson, 2002). Similar findings of a higher peak BAC in female mice compared to male mice has been demonstrated when administered intragastrically, while the reverse has been observed when administered intraperitoneally (Desroches et al., 1995). This could imply that females have lower gastric ADH activity, an observation which has been demonstrated in both humans (Baraona et al., 2001) and rats (Lee et al., 1992). Another study demonstrated a higher peak BAC in male mice regardless of the route of administration

(Middaugh et al., 1992). The explanation for this difference is unclear. Although the exact mechanism is unknown, it has been suggested that female Msh2-/-p53-/- mice deficient in mismatch repair may be more susceptible to developmental deficits due to X chromosome inactivation resulting in Barr bodies, which can lead to developmental arrest and apoptosis, an observation not seen in males of the same strain (Cranston et al., 1997). These results are consistent with the gender differences observed herein in our ogg1-deficient mice in whole embryo culture.

Studies in rodents examining the effect of pregnancy on ethanol metabolism have been similarly inconclusive. Some have reported a higher BAC in pregnant animals compared to virgin females (Badger et al., 2005; Traves and Lopez-Tejero, 1994), while others have found no difference (Gordon et al., 1985; Traves et al., 1995).

Differences in age can also contribute to differing BACs. Older rats (12 months) had lower total body water content than younger rats (2 to 3 months), which completely accounted for the difference in BAC observed between the two groups (Wiberg et al., 1971).

Once absorbed, EtOH undergoes hepatic and gastric first-pass metabolism (Fig. 1.53).

The first step in the breakdown of EtOH is catalyzed by alcohol dehydrogenase (Km=1-2 mM),

150

Figure 1.53: Pathways of ethanol metabolism. Abbreviations used: ADH1, alcohol dehydrogenase; ALDH2, aldehyde dehydrogenase; CYP2E1, cytochromes P450 2E1; NADPH, nucleotide adenine dinucleotide phosphate.

151

which occurs predominantly in the liver (Larsen, 1959; Winkler et al., 1969) and gastric mucosa

(Lim et al., 1993). Studies in both humans (DiPadova et al., 1987) and rats (Julkunen et al.,

1985a; Julkunen et al., 1985b) have shown that BACs are lower after oral or intragastric administration than after intravenous administration when administered the same dose. The hydrogen is removed with the coenzyme NAD+ as an electron acceptor, which then transfers the hydrogen to other acceptors in the reoxidation of NADH (Wallgren and Barry, 1970). Oxidation of EtOH accounts for over 90% of its removal (Holford, 1987).

Catalase and cytochrome P450 (CYP) 2E1 can also catalyze this reaction, although the contribution of catalase is minimal due to the low peroxide levels in tissues (Hay et al., 1956;

Nelson et al., 1957), and the contribution of CYP2E1 only occurs when the concentration of

EtOH is 10 mM or higher (Km = 8-10 mM) (Grunnet, 1973; Thieden, 1971). CYP2E1-mediated metabolism of EtOH has been shown to generate ROS as a byproduct (Fig. 1.54). Acetaldehyde is rapidly oxidized to acetic acid predominantly by aldehyde dehydrogenase (ALDH2) in the liver (Wallgren and Barry, 1970), while other enzymes including aldehyde oxidase, xanthine oxidases and CYP2E1 have been shown to oxidize acetaldehyde to acetate (Terelius et al., 1991).

The mitochondria contains low Km (0.01 mM) and high Km (1 mM) ALDH2 isoforms, the former of which is responsible for the majority of acetaldehyde oxidation arising from EtOH oxidation in rodents (Corrall et al., 1976; Grunnet, 1973; Parrilla et al., 1974; Siew et al.,

1976) and humans (Forte-McRobbie and Pietruszko, 1985; Henehan et al., 1985).

The majority of EtOH (90-95%) is metabolized via acetaldehyde to CO2 and water while a small percentage (10%) is excreted unchanged, all of which are eliminated in the urine, breath and sweat. (Wiberg et al., 1971), The rates of EtOH elimination vary by species, with rodents eliminating EtOH much faster than humans (Table 1.18). When blood levels of EtOH have

152

Figure 1.54: ROS generation by CYP2E1-mediated metabolism of EtOH. CYP2E1 is represented by the heme iron in the blue ovals. When CYP2E1 uses oxygen to metabolize alcohol, reactive oxygen species (ROS) can be generated by the following chain of events: ethanol binds to the enzyme (step 1). As the first electron is passed to the heme of CYP2E1 and oxygen is bound (step 2), the electron can move and exist on the oxygen, essentially generating superoxide bound to the heme of CYP2E1 (step 3). Occasionally, the superoxide will break down, releasing free superoxide and generating the starting enzyme. If the second electron is added to the enzyme (step 4), then a second form of reduced oxygen is produced that is identical to a heme-bound form of the two-electron–reduced oxygen (i.e., peroxide) (step 5). When this product breaks down, it picks up two hydrogens to generate hydrogen peroxide. The production of these ROS by CYP2E1 is referred to as an “uncoupled reaction” because the oxygen does not end up in the substrate. If the reduced oxygen species remains bound, then the enzyme will transfer one oxygen atom to the substrate and the other atom becomes water, producing an unstable intermediate (i.e., a gem-diol) product that decomposes to acetaldehyde (step 6). From: (Koop, 2006)

153

Table 1.18: Ethanol elimination rates measured in humans and rodents

Elimination Species (mg/kg/h) Human 92 132 88 92 84 97 Mouse 650 550 Rat 628 397 270 283-352 271-512 359 322

Adapted from: (Wallgren and Barry, 1970)

154 returned to zero, changes associated with a ‘hangover’ become apparent. Physiological factors contributing to a ‘hangover’ have been attributed to many causes, including but not limited to :

(1) dehydration and electrolyte imbalance by inhibiting the release of antidiuretic hormone or vasopressin from the pituitary gland (Kleeman et al., 1955); (2) gastrointestinal disturbances including gastritis caused by irritation inflammation and irritation of the stomach lining, and delayed gastric emptying (Anylian et al., 1978); (3) hypoglycemia by inhibiting gluconeogenesis

(Krebs et al., 1969); (4) disruption of sleep by interfering with growth hormone secretion at night and inducing adrenocorticotropic hormone from the pituitary gland which then stimulates the release of cortisol, a hormone involved in stress response and carbohydrate metabolism (Swift and Davidson, 1998). Physical symptoms include fatigue, headache, thirst and malaise, and psychological symptoms include anxiety, depression and irritability (Pohorecky and Brick,

1988).

1.10.2 Teratogenesis

In utero embryonic and fetal exposure to EtOH can result in a spectrum of morphological and behavioral or neurodevelopmental anomalies, collectively termed Fetal Alcohol Spectrum

Disorders. This spectrum can include Alcohol Related Birth Defects (ARBD), Alcohol Related

Neurological Disorders (ARND) and Fetal Alcohol Syndrome (FAS). FAS, representing the complete phenotype is characterized by characteristic craniofacial dysmorphology (short palpebral fissures, flat mid-face, smooth or flat philtrum, thin upper lip), growth retardation, and neurodevelopmental deficits (Jones, 2011). The estimated incidence of FASD is 1 out of 100 live births in Canada (Stade et al., 2009).

155

In rodent models, prenatal EtOH exposure causes morphological birth defects (structural teratogenesis) in the developing fetus in both rats (Lee et al., 2005; Wentzel and Eriksson, 2008) and mice (Dong et al., 2010; Wentzel and Eriksson, 2006) in vivo (Table 1.19) and in embryo culture (Table 1.4, Fig. 1.19), and initiates postnatal behavioral dysfunction (functional teratogenesis) in mice (Becker and Randall, 1989; Gilliam et al., 1987) and rats (Brocardo et al.,

2012; Mattson et al., 1993; Mooney and Varlinskaya, 2011) (Table 1.6).

FASD is thought to be one of the leading preventable causes of neurodevelopmental disorders in the Western world (Mattson et al., 2011). Children exposed to EtOH prenatally exhibit a broad spectrum of behavioural anomalies including motor function deficits, hyperactivity, decreased IQ, and deficits in executive function, verbal language, memory and attention (Mattson et al., 2011). Studies using magnetic resonance imaging (MRI) of the brains of children with FASD have shown volume reductions in brain regions such as the hippocampus

(Coles et al., 2011), basal ganglia (Mattson et al., 1996), cerebellum (Sowell et al., 1996) and corpus callosum (Riley et al., 1995), and animal models exhibit similar reductions in the hippocampus (Wigal and Amsel, 1990) and cerebellum (Heaton et al., 2002) (Table 1.20). The estimated annual cost of FASD in Canada has been reported to be in the range of $148.4 million

– $6.7 billion, which takes into account both direct costs such as health care, education and social services, as well as indirect costs including loss of productivity (Popova et al., 2011).

1.10.2.1 Mechanisms of EtOH teratogenesis

Several mechanisms have been postulated to account for the pleiotropic effects EtOH exerts on the developing embryo including but not limited to acetaldehyde formation, cell-cell interaction and cell-adhesion disassembly, deregulation of neurotransmission systems, growth factor and trophic support alteration, aberrant glycosylation and glucose uptake, interference

156

Table 1.19: Structural birth defects initiated by in utero EtOH exposure in rodents

Strain/ Dosing Exposure Analysis ROA Outcome Reference Species regimen period date 2x2.9 g/kg ↓crown‐rump length, C57BL/6 (Parnell et al., (23.7%), 4 GD 8 GD 17 brain and whole body mouse 2009) h apart, i.p volume (MRI) ↑resorptions and C57BL/6, 4.8 g/kg (Weston et al., GD 8 GD 17 ↓fetal weight ICR mice (25%), i.p 1994) (C57BL/6 > ICR) 2x2.9 g/kg ↑ facial C57BL/6 (23.7%), GD 8, (Kotch and Sulik, GD 8 malformations, mouse 4h apart, 10, 14 1992a) resorptions i.p i.p ↑fetal mortality, microphthalmia and C57BL/6, 2x3.97g/kg GD 7, 8, 9, digital malformations (Fukui et al., DBA 4h apart 10, 11, 12, GD 18 (C57BL/6), 1990) mice (25%), i.p or 13 ↑ophthalmic defects (DBA), ↓fetal brain weight 2*2.9 4h ↑ facial C57BL/6 apart or GD 7, 8, 9 malformations (day 7, (Webster et al., GD 18 mouse 5.8 g/kg or 10 8), limb defects (day 9, 1983) (25%) i.p 10) CBA, ↑ prenatal death, C3H, 20% liquid Throughout malformations, ↓ GD 18 (Chernoff, 1980) C57BL/6 diet gestation fetal weight (CBA > mice C3H > C57BL/6) Liquid diet ↑ resorption rate, malformations C57BL/6 25% liquid (Randall et al., GD 5 ‐ 10 GD 19 (skeleton, eye, head, mouse diet 1977) heart, abdominal/urogenital)

Abbreviations used: GD, gestational day; i.p., intraperitoneal

157

Table 1.20: Brain regions affected by in utero EtOH exposure

Constituent Brain Region Function Effect of Exposure References Parts

Cognitive function, motor abilities; Caudate (Archibald et al., Caudate Nucleus: 2001; Mattson nucleus, MRI studies: reduction spatial and Riley, 1999; Basal Ganglia Putamen, in size, notably the memory, Mattson et al., Globus caudate nucleus shifting from 1994; Mattson et Pallidus one task to al., 1996) another, inhibition of bad behaviour MRI studies: thinning or complete absence of Connects the region; displacement left and right relative to other brains (Roebuck et al., Large bundle cerebral Corpus structures. Deficits 1998; Sowell et of nerve hemispheres, Callosum linked to: attention, al., 2001; fibers allowing them intellectual function, Witelson, 1985) to learning, reading, communicate psychosocial and executive functioning. Disproportionate Located at (Archibald et al., Cognitive and reduction in size, Cerebellum the base of 2001; Sowell et motor skills especially in anterior the brain al., 1996) vermis (Bunsey and Volume asymmetry, left Eichenbaum, Located lobe smaller than the 1996; Riikonen within Declarative Hippocampus right; Deficits in spatial et al., 1999; temporal lobe Memory and other types of Squire, 1992; of brain memory Uecker and Nadel, 1996)

Abbreviations used: MRI, magnetic resonance imaging

158 with signal transduction pathways, hypoxia, apoptosis and oxidative stress (see next section)

(Table 1.21). No single mechanism can account for all of the structural and behavioural phenotypes observed in children exposed prenatally to EtOH, and it would not be surprising if more than one mechanism can contribute to the same adverse outcome. Furthermore, if two pregnant women were to consume the same amount of alcohol, the outcome to the developing offspring will not always be the same as there are numerous risk factors associated with FASD including but not limited to: (i) dose and pattern of alcohol consumption (binge versus chronic, and the frequency); (ii) timing of consumption (embryonic vs. fetal period); (iii) maternal and embryonic genetics and (iv) environmental factors (polypharmacy including drugs of abuse, maternal nutrition, socioeconomic status) (Martinez and Egea, 2007). Interestingly, acetaldehyde has been shown to activate the Fanconi anemia-breast cancer susceptibility (FA-BRCA) DNA damage response network along with a 4-fold increase in the acetaldehyde DNA-adduct N2- ethylidene-dGuo (Abraham et al., 2011). Additionally, EtOH has been shown to inhibit fetal

Shh production, a signaling molecule important for normal embryonic development (Aoto et al.,

2008).

1.10.2.1 Evidence for the involvement of ROS in EtOH teratogenesis

EtOH-initiated ROS formation has been measured both directly and indirectly in cell culture, xenopus culture, rodent embryo culture and in vivo in several species (Table 1.22). I used EtOH as a ROS-initiating teratogen to reveal novel molecular and biochemical determinants of both structural and functional deficits.

Evidence of ROS in the mechanism of structural teratogenesis includes: decreased forelimb malformations in EtOH-treated embryos pretreated with the SOD/catalase mimetic EUK-134

159

Table 1.21: Mechanisms of EtOH-initiated developmental toxicity

Mechanism References (Campbell and Fantel, 1983; Giavini et al., 1992; Holownia et al., 1996; O'Shea and Kaufman, Acetaldehyde formation 1981; Sreenathan et al., 1982; Sreenathan et al., 1984; Webster et al., 1983); (Bearer, 2001; Bearer et al., 1999; Cell-cell interaction and cell-adhesion Charness et al., 1994; assembly Ramanathan et al., 1996; Wilkemeyer and Charness, 1998) (Ikonomidou et al., 1999; Sari and Deregulation of neurotransmission systems Zhou, 2004; Zhou et al., 2002) (Cui et al., 1997; Ge et al., 2004; Heaton et al., 2000; Li et al., 2004; Growth factor and trophic support alteration Luo and Miller, 1996; Miller et al., 2002) (Fattoretti et al., 2003; Hu et al., 1995; Singh et al., 1992; Snyder Aberrant glycosylation and glucose uptake et al., 1992; Snyder and Singh, 1989; Tomas et al., 2002; Tomas et al., 2003) (Allansson et al., 2001; Bonthius et al., 2004; Bonthius et al., 2003; Interference with signal transduction Deltour et al., 1996; Duester, pathways 1991; Kumada et al., 2006; Zhang et al., 1998) Hypoxia (Savoy-Moore et al., 1989) (Bhave and Hoffman, 1997; Cartwright et al., 1998; De et al., 1994; Holownia et al., 1997; Apoptosis Ikonomidou et al., 2000; Kotch and Sulik, 1992a; Kotch and Sulik, 1992b; Light et al., 2002) (Chen et al., 2004a; Dong et al., 2008; Dong et al., 2010; Nayanatara Oxidative stress et al., 2009; Wentzel and Eriksson, 2008)

See text on pages 156 and 159 for discussion

160

Table 1.22: Evidence of ROS in EtOH-initiated embryopathies and teratogenesis

Exposure Evidence for oxidative Species [EtOH] Probe Reference duration damage CELL CULTURE Fetal rat ↑ ROS (DCF), Lipid peroxidation L-cysteine, L-cystine, anti- (Montoliu et al., astrocytes (GD 25, 50 mM 7-10 days (MDA and 4-HNE), ↓ GSH; CYP2E1 antibody 1995) 21) protection with GSH addition Sprague-Dawley ↑ MDA (measured with TBA (Devi et al., fetal rat (GD 20) 2 mg/ml 24 hr None test) by 14% at 6 hr and by 18 1996) hepatocytes hr at 24 hr vs. control Primary C57BL/6 SOD (300 U/mL), CAT (500 NBT probe for superoxide anion mouse embryo 50, 100, 150, U/mL) α-tocopherol (300 radicals, concentration- (Chen and Sulik, 48 hr (6-16 somite pair) 200 mM μM) Xanthine/Xanthine- dependent ↓ cytotoxicity with 1996) cranial NCCs Oxidase antioxidant treatment Co-treatment with either: 1 Primary C57BL/6 or 10 uM DFX (iron mouse embryo chelator), 10, 50, 250 μM Cell viability (trypan blue (Chen et al., (6-16 somite pair) 100 mM 16 hr PHE (iron chelator), 0.01, exclusion) 2000) cranial neural 0.1, 1 mM NAC crest cells (antioxidant) XENOPUS CULTURE Ocular defects (39%) ↓ with 0.1-0.5% in 10 Overexpression of CAT and enhanced catalase (21%), 2-cell stage- (Peng et al., Xenopus laevis mL medium both human cpPRDX5 and cpPRDX5 (19%), mtPRDX5 stage 25 2004) (171-856 μM) mtPRDX5 (13%) expression, H2O2 measurement, NO production

161

Table 1.22 (cont’d): Evidence of ROS in EtOH-initiated embryopathies and teratogenesis

Exposure Evidence for oxidative Species [EtOH] Probe Reference duration damage WHOLE EMBRYO CULTURE

Capsaicin (trans-8-methyl- Capsaicin co-treatment ↓ EtOH N-vanillyl-6-nonenamide) embryopathies, ↑ mRNA (Kim et al., ICR mouse 1.0 μl/ml 48 hr (1 × 10-8 μg/ml or 1 × 10-7 expression of cGPx and 2008) μg/ml) PHGPx, ↑ SOD activity

Black ginseng co-treatment ↓ Black Ginseng (1, 10, 100 ethanol embryopathies, restored (Lee et al., ICR mouse 1.0 μl/ml 48 hr μg/ml) mRNA levels of cGPx, PHGPx 2009) and SePP

EtOH ↑ superoxide anion 6 hr drug, 5 mg/ml in SOD (300 U/mL) co- production (NBT) lipid (Kotch et al., C57BL/6 Mouse 30 hr ECM treatment peroxidation (MDA), ↓ by SOD 1995) medium co-exposure WEC: 17-171 mM EtOH, 86.5-432 M AcOH; WEC: 48 hr, WEC: ↑ 8-OHdG (ELISA), ROS Rat (strain not CAT (60, 100 μg/ml), (Lee et al., Midbrain cell Cell culture: (DCF); Cell culture: ↓ specified) vitamin E (no protection) 2005) culture: 6-856 96 hr cytotoxicity with probes mM EtOH; 32 μM-4 mM AcOH

162

Table 1.22 (cont’d): Evidence of ROS in EtOH-initiated embryopathies and teratogenesis

Strain/ EtOH Dose Exposure Evidence for oxidative Probe Reference Species (ROA) duration damage IN VIVO 2 x 2.9 g/kg (4h DPI ↓ EtOH-initiated 8-oxodG, (Dong et al., C57BL/6 mouse 6 hr DPI (NOX inhibitor) apart) i.p. apoptosis, NOX activity & mRNA 2010) GD 9 (plug = EUK-134 ↓ limb malformations (Chen et al., C57BL/6 mouse 2.9 g/kg i.p. EUK-134: SOD/CAT GD 1) (67.3% to 35.9%) 2004a) 16 days 4.8% ethanol in before NAC in drinking water on GD (Parnell et al., C57BL/6 mouse drinking water mating, GD NAC ↓ ocular abnormalities 7, 8 (eq. 300 and 600 mg/kg) 2010) (p.o.) 14 (sacrificed) 2wk prior to 5-20% in gestation + (Wentzel and C57BL/6 mouse drinking water throughout Transgenic SOD mice Fetal hepatic 8-isoprostanes Eriksson, 2006) (p.o.) gestation to GD 18

2 x 2.9 g/kg 4h GD 8 (plug = 18 hr D3T pretreatment (5 (Dong et al., C57BL/6 mouse ROS (DCF) in embryos apart (i.p.) GD 0), 6 hr mg/kg) to upregulate Nrf2 2008)

15 days prior to Lipid peroxidation (TBARS) in mating+ (Nayanatara et Wistar rat 2 g/kg (i.p.) none fetal liver, brain, kidneys, testes entire al., 2009) (nmol MDA/g wet tissue) gestation period Abbreviations used: CAT, catalase; cpPRDX5, cytosolic/peroxisomal peroxiredoxin 5; cGPx, cytoplasmic glutathione peroxidase; DCF, dichlorofluorescein; DFX, desferoxamine; GD, gestational day; MDA, malondialdehyde; 4-HNE, 4-hydroxynonenal; GSH, glutathione; mtPRDX5, mitochondrial peroxiredoxin 5; NAC; N-acetyl cysteine; NBT, nitro blue tetrazolium; NCC, neural crest cell; NO, nitric oxide; Nrf2, nuclear factor erythroid 2-related factor 2; PHE, phenanthroline; PHGPx, phospholipid glutathione peroxidase; ROS, reactive oxygen species; SePP, selenoprotein P; SOD, superoxide dismutase; TBA, thiobarbituric acid, TBARS, thiobarbituric acid reactive substances;

163

(Chen et al., 2004a); elevation in fetal hepatic isoprostane formation in superoxide SOD knockout mice exposed to EtOH in utero (Wentzel and Eriksson, 2006); a protective effect of

3H-1,2 dithiole-3-thoine (D3T), a nuclear factor erythroid 2-related factor 2 (Nrf2) inducer, against EtOH-initiated ROS formation in mouse embryos exposed in utero to EtOH (Dong et al.,

2008); increased lipid peroxidation in fetal liver, brain, kidneys and testes (Nayanatara et al.,

2009); and a protective effect of the NOX inhibitor DPI, against EtOH-initiated ROS formation and oxidative DNA damage in mouse embryos (Dong et al., 2010).

Evidence of ROS in the mechanism of behavioural teratogenesis is derived from studies measuring oxidatively damaged macromolecules in fetal brain tissues (Table 1.23), as well as several observational studies using chronic EtOH administration reporting protection by antioxidative compounds (Table 1.24). These include: (1) Dolivin, an antioxidant containing vitamin E and hypoxen, enhanced passive avoidance test performance in adult rat progeny exposed to EtOH in utero; (2) vitamins C and E administered with EtOH to pregnant guinea pigs between GDs 2-67 improved task-retention memory (Nash et al 2007); and (3) the antioxidant salymarin improve social recognition tasks (males) and spatial working memory (females) in rat progeny exposed to EtOH throughout gestation via a maternal liquid diet containing 35% EtOH

(Busby et al., 2002; Neese et al., 2004; Reid et al., 1999). These antioxidant studies are consistent with a role for ROS in the neurodevelopmental deficits caused by in utero EtOH exposure; however, ROS and/or their molecular effects (e.g. macromolecular oxidation) were not measured in fetal brain, so protective mechanisms unrelated an antioxidative effect in these studies cannot be excluded.

164

Table 1.23: Measurement of oxidatively damaged cellular macromolecules in fetal brain tissues after in utero EtOH exposure Dosing Exposure Species ROA Tissue analyzed Endpoint measured Reference regimen Duration 4 g/kg EtOH, 12 hr GD 17, 18, GD19 Fetal and maternal ↑ MDA and conjugated (Henderson et al., Rat i.g. intervals 19 brain and liver dienes 1995) (5 doses total) ↓ mRNA levels of mitochondrial genes 0, 2, 4.5, p.o. encoding complexes II-A, IV Long 6.5, GD 6- (liquid PN 1 Cerebellum and V, ↑ expression of p53 (Chu et al., 2007) evans rat 9.25% delivery diet) and NOX 1 and 3; ↑ v/v EtOH immunoreactivity of 4-HNE and 8-OHdG

Throughout Frontal cortex, striatum, 9 pregnancy Rat NS hypothalamus, hippocampus ↑ conjugated dienes (Petkov et al., 1992) g/kg/day and and cerebellum lactation

Sprague- 4 g/kg, GD 17, 18, Mitochondrial from fetal (Ramachandran et al., Dawley i.g. ↑ 4-HNE 25% v/v 19 brains 2001) Rat

Abbreviations used: 4-HNE, 4-hydroxynonenal; 8-OHdG, 8-oxo-2’-deoxyguanosine; GD, gestational day; i.g., intragastric; MDA, malondialdehyde; NOX, NADPH oxidase; NS, not stated; p.o., per os; PN, postnatal; ROA, route of administration;

165

Table 1.24: Indirect evidence for ROS in EtOH-initiated behavioural deficits

Exposure Age Model Dose Probe Probe dose ROA Test Result Reference Time tested

107.7 mM Range: C57BL/6 ADNF-9 ADNF-9; Morris water- EtOH ↓ learning 3X ADNF-9, (Vink et al., 5.9 g/kg GD 8 i.p. PND 35- Mouse + NAP 121.22 mM maze (male) NAP ↑ learning 2005) 50 NAP)

Radial arm PND 75 EtOH ↓ learning 2.5X Sprague- Throughout 400 Liquid maze (Busby et al., 35% EDC SI Dawley Rat pregnancy mg/kg/day diet Social 2002) PND 90 EtOH ↓ learning 4X recognition

s.c. Sprague- Throughout 400 Social EtOH-treated pups had ↓ (Reid et al., 35% EDC SY PND 90 Dawley Rat pregnancy mg/kg/day recognition social recognition 1999) p.o. EtOH: required 3.88 trials to GD 1-7 meet criterion; protected by SY Fischer/344 400 Liquid Radial arm EtOH: required 4.13 trials to (Neese et 35% EDC GD 8-14 SY PND 60 Rat mg/kg/day diet maze meet criterion; protected by SY al., 2004) EtOH: required 6.75 trials to GD 15-21 meet criterion; protected by SY

Abbreviations used: ADNF-9, activity-dependent neurotrophic factor 9; EDC, ethanol-derived calories; EtOH, ethanol; GD, gestational day; i.p, intraperitoneal; NAP, NAPVSIPQ peptide; PND, postnatal day; ROA, route of administration; SI, silybin/phospholipid; SY, silymarin; s.c., subcutaneous

166

1.11 Methanol (MeOH)

1.11.1 Pharmacokinetics

In humans, MeOH exposure results either from environmental or accidental exposure

(Harris et al., 2004). Once in the body, methanol (MeOH) is metabolized by alcohol dehydrogenase to formaldehyde (Cederbaum and Qureshi, 1982), which is then oxidized by glutathione-dependent formaldehyde dehydrogenase to formic acid, which is metabolized to CO2 and H2O by a folate-dependent pathway (Fig. 1.55) (Sweeting et al., 2010). In rodents however,

MeOH is also metabolized by catalase to formaldehyde, which as in humans is oxidized by glutathione-dependent formaldehyde dehydrogenase to formic acid, which is metabolized to CO2 and H2O by a folate-dependent pathway (Harris et al., 2004).

Species susceptibility to MeOH toxicity has been shown to be related to the ability to metabolize formic acid. As mentioned above, formic acid is metabolized to carbon dioxide and water by a folate-dependent dehydrogenase. Using 10-formyl-tetrahydrofolate (THF) synthetase, formic acid is sd converted into 10-formyl-THF, which in turn is converted into carbon dioxide and water by 10-formyl-THF dehydrogenase (Black et al., 1985) (Fig. 1.56). This metabolic pathway is maintained in all species; however, varying levels of the cofactor THF limit the efficiency of formic acid oxidation (McMartin et al., 1981). Specifically, the mouse has over fourfold higher basal THF levels than humans, effectively decreasing the accumulation of formic acid and its associated toxicities including blindness and ultimately death (Sokoro et al., 2008).

167

Figure 1.55: Metabolic pathways of methanol in rodents and humans

168

Figure 1.56: Role of folate in the metabolism of formic acid. In the presence of tetrahydrofolate (THF), formate is metabolized into 10-formyl-THF by 10-formyl-THF-synthetase. 10-Formyl-THF is subsequently converted into CO2 and H2O by 10-formyl-THF dehydrogenase. Modified from: (Wells et al., 2013)

169

1.11.2 Teratogenesis

Human MeOH exposure results mainly from environmental exposure, and although MeOH is a developmental toxicant in mice and rats, the underlying mechanism is unclear, and it is not yet clear whether or not MeOH is developmentally toxic in humans (Andrews et al., 1993; Degitz et al., 2004; Rogers and Mole, 1997; Rogers et al., 1993). In rodents, studies demonstrate that

MeOH is both embryopathic in whole embryo culture (Table 1.5, Fig. 1.19, 1.20) and teratogenic in vivo (Bolon et al., 1993; Bolon et al., 1994; De-Carvalho et al., 1994; Rogers et al., 2004; Rogers and Mole, 1997; Rogers et al., 1993; Youssef et al., 1997) (Table 1.25).

Although the underlying mechanism of developmental toxicity is unclear, studies have implicated a role for enhanced oxidative stress, or increased ROS (Wells et al., 2013).

1.11.2.1 Evidence for the involvement of ROS in MeOH teratogenesis

Studies in adult animals have demonstrated enhanced lipid peroxidation in brain, liver, bile, erythrocytes, urine as well as in lymphoid organs of rats intoxicated both acutely and chronically with MeOH (Table 1.26).

(Skrzydlewska et al., 2000); (2) MeOH-derived adducts to the free radical spin trapping agent

PBN detected in bile and urine of PBN-pretreated, MeOH-exposed rats (Kadiiska and Mason,

2000); and, (3) MeOH embryopathies in rat whole embryo culture are enhanced by the depletion

170

Table 1.25: Developmental toxicity of methanol (MeOH) in rodents

Exposure Species Strain Dosing Regimen ROA Effect of MeOH Reference Duration Exencephaly, cleft palate, 1,000 - 15,000 ppm, in utero death, decreased CD-1 Inhalation GD 6-15 (Rogers et al., 1993) 7hr/day fetal weight, cervical rib ossification Resorptions, NTDs, cleft GD 6-15 palate, CD-1 10,000 ppm, 6 hr/day Inhalation (Bolon et al., 1993) GD 7-9 Resorptions, cleft palates GD 9-11 Cleft palates Mouse CD-1 15,000 ppm, 6 hr/day Inhalation GD 7-9 NTDs (Bolon et al., 1994)

10,000 ppm, 7 hr/day (Rogers and Mole, CD-1 Inhalation GD 6-13 Cleft palate, NTDs 10,000 ppm, 7 hr/day 1997) x 2 (24 hr apart) 3.4 g/kg x 2 (4 hr apart) C57BL/6 i.p. GD 7 Ocular, craniofacial (Rogers et al., 2004) 4.9 g/kg x 2 (4 hr (dose-dependent apart) increase) Dose-dependent skeletal Sprague- 5,000 - 20,000 ppm, 7 Inhalation GD 1-19 and cardiovascular (Nelson et al., 1985) Dawley hr/day defects (De-Carvalho et al., Rat Wistar 2.5 g/kg/day Gavage GD 6-15 Skeletal malformations 1994) Dose dependent ocular Long-Evans 1.02 - 4.09 g/kg/day Gavage GD 10 defects and incidence of (Youssef et al., 1997) undescended testes Abbreviations used: GD, gestational day; NTD, neural tube defect; ppm, parts per million.

171

Table 1.26: Evidence for MeOH-initiated oxidative stress

Dose Species ROA Duration Probe Effect of MeOH Effect of Probe Tissues analyzed Reference (g/kg)

↑ Lipid peroxidation (TBARS), ↓ Wistar (Skrzydlewska and i.g. 1.5, 3 Acute N/A GSH-Px, SOD, GSSG-R N/A Liver, Serum Rat Farbiszewski, 1997e) activities, ↓ [ascorbate]

↑ Lipid peroxidation (TBARS), ↓ Wistar GSH-Px, SOD, GSSG-R Liver, Erythrocytes, (Skrzydlewska and i.g. 6 Acute N/A N/A Rat activities, ↓ sulfhydryl Serum Farbiszewski, 1997d) compounds, [ascorbate]

↓[GSH], GSH-Px and GSSG-R Liver, Erythrocytes, (Skrzydlewska and Wistar rat i.g. 3, 6 Acute N/A N/A activities Serum Farbiszewski, 1997c)

↓ lipid peroxidation; U- ↑ Lipid peroxidation (MDA); ↓ Liver, Erythrocytes, (Skrzydlewska and Wistar rat i.g. 3 Acute ↑GSH, 83836E GSH-Px, GSSG-R, TAS Serum Farbiszewski, 1997b) antioxidative enzymes

↑ Lipid peroxidation (TBARS) 1.5, 3, CuZnSOD, CAT; ↓ GSH-Px, (Skrzydlewska and Wistar rat i.g. Acute N/A N/A Liver 6 GSSG-R, [GSH], [Ascorbate], Farbiszewski, 1997a) TAS

↑ Lipid peroxidation (TBARS), ↓ [GSH]/[ascorbate] Brain: ↓SOD, (Skrzydlewska et al., Wistar rat i.g. 6 7 d N/A GSH-Px, GSSG-R activity Liver: N/A Brain, Liver 1998) ↑SOD, CAT activity, ↓GSH-Px, GSSG-R activity

172

Table 1.26 (cont’d): Evidence for MeOH-initiated oxidative stress

Dose Species ROA Duration Probe Effect of MeOH Effect of Probe Tissues analyzed Reference (g/kg)

Direct detection of MeOH (Skrzydlewska et al., Wistar rat i.g. 6 7 d none radical, lipid peroxidation, N/A Liver 2000) protein carbonyl formation

↓ TBARS, ↑ Lipid peroxidation (TBARS), ↑ U-83836E, membrane (Dobrzynska et al., Wistar rat i.g. 3 Acute membrane surface charge Erythrocytes NAC surface charge 1999) density density

Sprague- i.p. (Kadiiska and Mason, Dawley 4.5, 7 Acute POBN Direct detection of a MeOH-derived PBN adduct Bile and urine i.g. 2000) Rat

↑ Lipid peroxidation (TBARS), Wistar 1, 15, 30 Enzymatic/non-enzymatic (Parthasarathy et al., i.p. 2.37 none N/A Lymphoid organs albino rat d antioxidants ↑ at 1d; ↓ at 15, 30 2006a) d

↑ Lipid peroxidation (TBARS), Wistar 1, 15, 30 enzymatic/non enzymatic Hypothalamus, (Parthasarathy et al., i.p. 2.37 none N/A albino rat d antioxidants ↑ at 1d; ↓ at 15, 30 adrenal gland 2006c) d; DNA fragmentation ↑ at 30 d

Abbreviations used: ROA, route of administration; i.g, intragastric; TBARS, thiobarbituric acid reactive substances; N/A, not applicable; MeOH, methanol; NAC, N-acetyl cysteine; POBN, 4-pyridyl N-oxide;

173

of GSH (Harris et al., 2004). Additionally, MeOH-initiated oxidative stress, as evidenced by the production of the lipid peroxidation product MDA, along with increases in antioxidative enzyme activities, was observed in the lymphoid organs of adult rats (Parthasarathy et al., 2006a).

Evidence implicating ROS in the mechanism of MeOH developmental toxicity is limited, however studies in whole embryo culture have provided evidence that ROS may be involved: (1) aCat embryos exhibited enhanced embryopathies, while hCat embryos were protected (see later in this thesis (chapter 4); and, (2) Sprague-Dawley rat embryos exposed to MeOH in culture exhibited enhanced embryopathies when GSH was depleted (Harris et al., 2004). Although the enzyme(s) catalyzing the formation of embryonic ROS remain to be determined, as shown later in this thesis (chapter 3), PHSs do not appear to contribute, as MeOH embryopathies are not blocked by pretreatment with the PHS inhibitor ETYA, while NOXs appear to play a role (Miller et al., unpublished data). CYP2E1 expression is negligible during the embryonic period, and low compared to adult activity during the fetal period, particularly in rodents (Vieira et al 1996;

Juchau et al 1998; Hines, 2008), so CYP2E1-mediated superoxide formation would appear to be an unlikely mechanism for embryonic ROS formation, at least in rodents. Preliminary studies described later in this thesis suggest that MeOH and/or its metabolites can activate and/or induce the expression of embryonic NOXs that produce embryopathic ROS and protein oxidation, as

MeOH embryopathies and protein oxidation are reduced by pretreatment with the NOX inhibitor

DPI and the free radical spin trapping agent PBN (Miller et al., unpublished data). This would be consistent with the NOX-dependent ROS mechanism previously reported for the related alcohol

EtOH (Dong et al., 2010). Additional potential mechanisms, such as interference with the mitochondrial electron transport chain or epigenetics have yet to be investigated.

174

Chapter 2 Study 1: Altered methanol embryopathies in embryo culture with mutant catalase-deficient mice and transgenic mice expressing human catalase

Running title: Catalase in methanol embryopathies

Lutfiya Miller† and Peter G. Wells*†

*Faculty of Pharmacy and †Department of Pharmacology and Toxicology University of Toronto Toronto, Ontario, Canada

a. Preliminary reports of this research were presented at the 2010 annual meeting of the Society of Toxicology of Canada [Proceedings of the 42nd Annual Symposium of the Society ofToxicology of Canada, Abstract No. 39] These studies were supported by a grant from the Methanol Foundation and Canadian Institutes of Health Research. b. Full report of this research has been published: Miller, L., Wells, P.G. (2011) Altered methanol embryopathies in embryo culture with mutant catalase-deficient mice and transgenic mice expressing human catalase. Toxicology and Applied Pharmacology 252(1): 55-61. c. Individual contributions: Lutfiya Miller- embryo culture; Peter G. Wells- supervisor.

175

2.1 Abstract

The mechanisms underlying the teratogenicity of methanol (MeOH) in rodents, unlike its acute toxicity in humans, is unclear, but may involve reactive oxygen species (ROS). Embryonic catalase, although expressed at about 5% of maternal activity, may protect the embryo by detoxifying ROS. This hypothesis was investigated in whole embryo culture to remove confounding maternal factors, including metabolism of MeOH by maternal catalase. C57BL/6

(C57) mouse embryos expressing human catalase (hCat) or their wild-type (C57 WT) controls, and C3Ga.Cg-Catb/J acatalasemic (aCat) mouse embryos or their wild-type C3HeB/FeJ (C3H

WT) controls, were explanted on gestational day (GD) 9 (plug = GD 1), exposed for 24 hr to 4 mg/mL MeOH or vehicle, and evaluated for functional and morphological changes. hCat and

C57 WT vehicle-exposed embryos developed normally. MeOH was embryopathic in C57 WT embryos, evidenced by decreases in anterior neuropore closure, somites developed and turning, whereas. hCat embryos were protected. Vehicle-exposed aCat mouse embryos had lower yolk sac diameters compared to C3H WT controls, suggesting endogenous ROS are embryopathic.

MeOH was more embryopathic in aCat embryos than WT controls, with reduced anterior neuropore closure and head length only in catalase-deficient embryos. These data suggest that

ROS may be involved in the embryopathic mechanism of methanol, and that embryonic catalase activity may be a determinant of teratological risk.

Key Words: Methanol; catalase; embryopathies; reactive oxygen species; transgenic mice; mutant mice; teratogenesis

176

2.2 Introduction

Methanol (MeOH) exposure results from its use as an industrial solvent and accidental ingestion (Harris et al., 2004), and is a developmental toxicant in both rats (Nelson et al., 1985) and mice in vivo and in embryo culture (Abbott et al., 1995; Andrews et al., 1993; Degitz et al.,

2004; Rogers and Mole, 1997; Rogers et al., 1993)

As discussed below, reactive oxygen species (ROS) have been implicated in the embryopathic mechanism for MeOH, and catalase is an antioxidative enzyme that converts potentially embryotoxic hydrogen peroxide to oxygen and water. Although expressed in the embryo at levels about 5% of those in maternal liver, there is some evidence that catalase may protect the embryo from ROS-initiated embryopathies (Wells et al., 2009b). On the other hand, catalase in the rodent can also metabolize methanol to its more reactive metabolite formaldehyde, as well as contributing, at least in adult animals, to the conversion of the downstream formic acid metabolite to carbon dioxide and water (Tephly et al., 1964).

Accordingly, the roles of ROS and embryonic catalase in the developmental toxicity of MeOH are unclear.

A number of studies have implicated free radical-initiated, ROS-mediated involvement in the mechanism of MeOH developmental toxicity, including: (1) direct detection of a MeOH radical by electron spin resonance spectrometry, and oxidative protein damage in MeOH- intoxicated rats (Skrzydlewska et al., 2000); (2) MeOH-derived adducts to the free radical spin trapping agent phenylbutylnitrone (PBN) detected in bile and urine of PBN-pretreated, MeOH- exposed rats (Kadiiska and Mason, 2000); and, (3) MeOH embryopathies in rat whole embryo culture are enhanced by the depletion of glutathione (GSH) (Harris et al., 2004). Additionally,

MeOH-initiated oxidative stress, as evidenced by the production of the lipid peroxidation product

177 malondialdehyde (MDA), along with increases in antioxidative enzyme activities, were observed in the lymphoid organs of adult rats (Parthasarathy et al., 2006a). Free radical intermediates and

ROS have been implicated in the teratological mechanism of other xenobiotics like phenytoin and structurally related antiepileptic drugs, benzo[a]pyrene, methamphetamine and thalidomide .

Exogenous catalase has been demonstrated to provide protection against other ROS- initiating teratogens such as phenytoin, ethanol and diabetes in vivo and/or in embryo culture in mice, rats or Xenopus (Cederberg and Eriksson, 1997; Chen et al., 2004a; Peng et al., 2004;

Winn and Wells, 1995). However, the protective efficacy of the endogenous enzyme, given its low embryonic activity, is unclear.

To determine the role of ROS and catalase in the mechanism of MeOH embryonic toxicity, we examined MeOH embryopathies in cultured transgenic mouse embryos endogenously expressing human catalase (hCat), as well as catalase-deficient (acatalasemic, aCat) mouse embryos. The use of whole embryo culture avoided the potential confounding role of maternal catalase in the metabolism of MeOH and its formic acid metabolite. The protection observed in embryos with increased endogenous catalase (hCat), together with the enhanced embryopathies observed in catalase-deficient embryos, suggest that ROS may be involved in the embryopathic mechanism of MeOH, and provides the first evidence that endogenous embryonic catalase activity may be a determinant of teratological risk for MeOH.

178

2.3 Methods

Chemicals. Fetal bovine serum, Hanks’ balanced salt solution (HBSS), Waymouth’s MB 752/1 medium, HEPES, and penicillin-streptomycin were purchased from Gibco Laboratories

(Toronto, Ontario, Canada). Isofluorane, USP was obtained from Baxter (Mississauga, Ontario,

Canada). HPLC grade MeOH was purchased from EMD Sereno Canada, Inc. (Mississauga, ON).

Saline (0.9 %, sterile) was purchased from Baxter Corporation (Mississauga, ON). Compressed

CO2 (5% in air) was purchased from Linde Canada (Mississauga, ON).

Animals and diet. All animal protocols used were approved by the institutional animal care committee in conformance with the guidelines established by the Canadian Council on Animal

Care. Homozygous C57BL/6J human-catalase (hCat) expressing mice (Tg(CAT)23Jv) were generously provided by Dr. Arlan Richardson at the University of Texas Health Science Center at San Antonio (San Antonio, TX), and were generated as described previously (Chen et al.,

2003). C57BL/6J (+/+) wild-type (C57 WT) male and female mice (Charles River Canada) were mated to generate a proximate wild-type colony to the hCat strain. Homozygous C3Ga.Cg-Catb/J acatalasemic (aCat) mice were purchased from Jackson Laboratory. C3HeB/FeJ (+/+) wild-type

(C3H WT) male and female mice were mated in order to generate a proximate wild-type colony to the aCat strain. All four mouse lines were bred and maintained separately. Mice were housed in plastic vented cages (Allentown, NJ) with ground corncob bedding (Bed-O’ Cobs Laboratory

Animal Bedding, The Andersons Industrial Products Group, Maumee, OH).

The cages were maintained in a light- and temperature-controlled room (14-hr light/10-hr dark cycle, 20°C, 50 % humidity). Food (Harlan Labs: 2018, Harlan Teklad, Montreal, QC) and

179 tap water were provided ad libitum. Mice were acclimatized for 1 week before use. One male was housed with three females from 5:00 P.M. to 9:00 A.M. the next day. The presence of a vaginal plug was designated gestational day (GD) 1, and these females were isolated and housed together in groups of four or less per cage.

Genotyping hCat and C57 WT. DNA from adult tail snips or ear notches was extracted by boiling at 95ºC in a 300 µl solution containing 10 mM NaOH/0.1 mM EDTA and genotyped by a

PCR-based assay. Primers were used to target a 450 bp fragment of exon 12/13 of the human- catalase gene: forward (5’-GAGGTCCACCCTGACTACGGG-3’) and reverse (5’-

GCCTTCTCCCTTGCCGCCAAG-3’) (The Centre for Applied Genomics, Hospital for Sick

Children, Toronto, Canada) (Fig. 2.1). The reaction conditions were: 1 µl per reaction genomic

DNA, 5 µl per sample of 10X Taq Buffer with KCl (100mM Tris-HCl (pH 8.8 at 25C), 500 mM

KCl, 0.8% (v/v) Nonidet P40) (Fermentas Life Sciences, Burlington, Toronto), 2.5 µl per sample of 25 mM MgCl2 (Fermentas Life Sciences), 0.5 µl per sample of 5 U/µl Taq polymerase

(Fermentas Life Sciences), 1.25 µl per sample of 10 mM deoxyribonucleotides (dNTP)

(Fermentas Life Sciences), 1 µl per sample of 50 mM of each primer, 6 µl per sample of 6X

DNA loading dye (0.03% bromophenol blue/0.03% xylene cyanol FF, Fermentas) and 37.75 µl per sample of ddH20 for a final volume of 50 µl. Cycling conditions were: 95ºC for 5 min; 35 cycles of: 95ºC, 30 s; 60ºC, 30 s; 72ºC, 1 min, and completed with a final extension at 72ºC for

10 min and then placed on hold at 4ºC. The PCR products were separated on a gel consisting of

1.5% (w/v) agarose, 40 mM Tris, 19.4mM glacial acetic acid and 2.5 mM EDTA.

180

Figure 2.1: Representative gel for genotyping showing DNA bands for mice expressing human catalase (hCat) and their proximate C567BL/6 wild-type (WT) controls.

Primers targeting exon12/13 of the human catalase gene amplify in the hCat mice a 450 bp fragment, which is absent in the WT animals. The presence of a 450 bp band identifies the hCat genotype, while no band signifies WT.

181

Genotyping aCat and C3H WT. DNA from adult tail snips or ear notches was extracted by boiling at 95ºC in a 300 µl solution containing 10 mM NaOH/0.1 mM EDTA and genotyped by a

PCR-based assay. Primers were used to target a 493 bp fragment of the mouse catalase gene: forward (5’- TCCTTCCAATCCCGTCCTTTCT-3’) and reverse (5’-

AAATGCCAAACTCGGAGCCATC-3’) (The Centre for Applied Genomics, Hospital for Sick

Children, Toronto, Canada) (Fig. 2.2). The reaction conditions were: 2 μl per reaction genomic

DNA, 2 μl per sample of 10X Hot Start PCR Buffer, 1.2 μl per sample of 25 mM MgCl2, 0.1 μl per sample of 5U/μl Maxima Hot Start Taq DNA polymerase, 0.2 μl per sample of 10 mM deoxyribonucleotides (dNTP) (Fermentas Life Sciences, Burlington, Toronto), 0.5 μl per sample of 20 mM of each primer, and 13.5 μl of ddH2O for a final volume of 20 μl. Cycling conditions were: 95°C for 5 min; 10 touchdown cycles of: 94°C for 20 s, 65°C for 20 s (-1° each cycle),

72°C for 40 s; 30 cycles of: 94°C for 20 s, 55°C for 20 s, 72°C for 40 s, and completed with a final extension at 72°C for 10 min and then held at 4°C. The PCR products were subjected to enzymatic digestion with restriction endonuclease NdeI (CA↓TATG) (Fermentas Life Sciences).

The reaction conditions were: 10 μl PCR reaction mixture, 18 μl nuclease-free water, 2 μl 10X

Buffer O and NdeI (10u/μl) (Fermentas Life Sciences). The reaction was incubated at 37°C overnight, and terminated by incubating at 65°C for 20 min, and then held at 4°C. Enzymatic digestion is necessary because a 115GT point mutation in the mouse catalase gene is responsible for the deficient catalase activity, as well as creating the recognition site for NdeI, which will cut the 493 bp WT band into 2 fragments of 255 bp and 237 bp. Cutting of the WT band is incomplete, which results in both a wild-type band, and a band between 237-255 bp, characteristic of an aCat heterozygote, which renders an aCat heterozygote indistinguishable from an aCat homozygote, thus precluding the merging of these two strains. The digestion products were separated on a gel consisting of 1.5% (w/v) agarose, 40 mM Tris, 19.4 mM glacial

182

Figure 2.2: Representative gel for genotyping showing DNA bands for catalase-deficient (acatalasemic, aCat) mice and their proximate C3HeB/FeJ WT controls.

Primers target a 493 bp segment of the mouse catalase gene, and PCR products are then subject to NdeI enzymatic digestion resulting in the presence of a 237/255 bp fragment. The presence of two bands identifies the aCat genotype, while one band signifies WT.

183 acetic acid, and 2.5 mM EDTA.

Embryo Culture. On GD 9, pregnant hCat dams or their proximate C57BL/6J WT controls, and aCat dams or their proximate C3HeB/FeJ WT controls, were euthanized by cervical dislocation.

The uterus containing the embryos was immediately removed, and rinsed with HBSS maintained at 37°C. Using a dissecting microscope (Stemi SV11, Carl Zeiss, Oberkochen, Germany), the individual implantation sites were exposed using number five watchmaker’s forceps (Fine

Science Tools, North Vancouver, BC, Canada). The decidua, trophoblast, parietal endoderm and

Reichert’s membrane were carefully removed, leaving the amnion, parietal endoderm, visceral yolk sac and ectoplacental cone intact. Because male rat serum (MRS) contains undefined nutrients and factors required by murine embryos for survival and growth, it was used as a component of the medium in which the embryos were cultured. Blood was obtained from retired

Sprague-Dawley male rat breeders (Charles River Canada), which were exsanguinated under isofluorane anaesthesia. The blood was centrifuged for 5 min at 1000 x g at 4ºC (Model TJ-6 centrifuge; Beckman Instruments, Toronto, Ontario, Canada) and kept on ice until blood was obtained from all animals. All blood samples were then centrifuged for 30 min at 1900 x g at 4ºC

(Model J2-21M centrifuge; Beckman Instruments). Pooled serum was heat-inactivated for 1 hr at

58ºC and gassed (5% CO2 in air; Linde, Mississauga, Ontario, Canada) for 30 min to evaporate residual protein-bound isofluorane. The heat-inactivated MRS was divided into aliquots and stored at -80ºC. All embryos dissected between the 7- to 8-somite stage were washed in pregassed (5% CO2 in air) holding medium (HM) (17 ml male rat serum, 2.5 mM HEPES, 50 units/ml of penicillin, 50 mg/ml of streptomycin, and 10X Waymouth’s MB 752/l medium) and subsequently transferred to a 24-well plate (BD Biosciences, Franklin Lakes New Jersey, US) containing 2 mL embryo culture medium (ECM; HM supplemented with 15 mL fetal bovine serum) (1 embryo per well) containing either methanol (4 mg/ml) or saline vehicle for a total of

184

24 hr. Each well was sealed airtight with PVC sealing tape (Immunochemistry technologies,

Bloomington, MN). The 24-well plate was incubated at 37ºC in a Sanyo model MCO-17A CO2 incubator (Sanyo Electric Co., Ltd., Japan) on a platform rocker (Bellco Biotechnology,

Vineland, NJ). After 24 hr, embryonic morphological and developmental parameters were observed using a dissecting microscope (Stemi SV11, Carl Zeiss, Oberkochen, Germany).

Developmental parameters included dorsal-ventral flexure (turning), anterior neuropore closure, and somite development. The embryo is measured as “turned” if it curls from a ‘belly-flop’ position into the fetal position during the period of culture (Moore and Persaud, 2007). The anterior neuropore is listed as ‘closed’ if fusion of the anterior neural folds has occurred. If fusion has not occurred, it is listed as ‘open’. Morphological assessments included yolk sac diameter (in mm), crown-rump length (in mm) and head length (in mm). Heart rate was a functional assessment. Anterior neuropore closure and turning were quantified as percentages of the total number of embryos assessed, while crown-rump length and somite development were scored as a percentage of only the number of embryos that turned.

Data analysis. Statistical significance between treatment groups was determined using GraphPad

Prism®, Version 5 (GraphPad Software, Inc., San Diego, CA). Anterior neuropore closure and turning were compared with a chi-square (Fisher’s exact) test. Somite development, crown-rump length, yolk-sac diameter and heart were compared using a one-way ANOVA with a post-hoc

Bonferroni test. A t-test was used to compare saline-exposed vs. MeOH-exposed aCat embryos for crown-rump length and head length. The minimal level of significance used throughout was p

< 0.05.

185

2.4 Results

Embryos expressing human catalase (hCat)

Comparison of growth of hCat and C57BL/6 WT saline-exposed embryos. No differences in any parameters were observed for baseline embryonic growth and development between saline- exposed hCat and C57BL/6 WT embryos (Fig. 2.3).

MeOH embryopathies in C57BL/6 WT embryos. Exposure to 4 mg/ml MeOH for 24 hr resulted in dysmorphogenesis evidenced by significant decreases in anterior neuropore closure

(60%), turning (69%) and somite development (13%), along with a significant increase in heart rate (31%), compared to saline-exposed WT controls (Fig. 2.3).

MeOH embryopathies in hCat embryos. MeOH was embryopathic in hCat embryos, evidenced by significant increases in crown-rump length (19%) and heart rate (51%) compared to saline-exposed hCat controls (Fig. 3).

Comparison of MeOH embryopathies in hCat versus C57BL/6 WT embryos. Compared to

MeOH-exposed WT controls, hCat embryos were almost completely protected from MeOH embryopathies, as evidenced by increases back to saline control levels for anterior neuropore closure (p < 0.05), somite development (p < 0.05) and turning (p = 0.1) (Fig. 2.3).

Catalase-deficient embryos (acatalasemic, aCat)

Comparison of growth of aCat and C3H WT saline-exposed embryos. There was a significant decrease in yolk-sac diameter (15%) in aCat embryos compared to WT embryos

186

40 Somites Developed Anterior Neuropore Closure 100 (10) (10) (13) 30 (9)* (9) ‡ (10) (3) *‡ 20 50 (14) 10 (n, mean + SD) + mean (n, PERCENTAGE (%) 0 0 S M S M S M S M

4 150 Turning Crown-rump Length 3 ‡ (10) (3) 100 (10) a (9) (9) (13) 2 (10) 50 ‡ (6) (14) 1 (mm, mean + SD) + mean (mm, PERCENTAGE (%) PERCENTAGE 0 0 S M S M S M S M

Heart Rate Yolk Sac Diameter 150 3 ‡‡ ‡‡‡ (10) (14) (13) (14) (10) (13) 100 2 (10) (10)

1 50 (mm, mean + SD) + mean (mm, 0 0 S M S M (beats/min, mean SD) + S M S M WT hCat WT hCat

Figure 2.3: Effect of hCat expression on methanol (MeOH) embryopathies.

On gestational day (GD) 9, WT (C57BL/6) embryos or hCat embryos containing 7 or 8 somite- pairs were explanted and incubated for 24 hr with MeOH (4 mg/ml) or its saline vehicle. Saline and MeOH treatments are indicated by ‘S’ and ‘M’ respectively. The number of embryos is given in parentheses. Double-daggers indicate a difference from saline control of the same genotype (‡ = p < 0.05, ‡ ‡ = p < 0.01, ‡ ‡ ‡ = p < 0.001), asterisks indicate a difference from the WT MeOH-exposed group (* = p < 0.05), the letter ‘a’ indicates p = 0.1 compared to the WT MeOH-exposed group.

187 exposed to saline vehicle (Fig. 2.4). Non-significant trends were apparent for decreased turning

(33%), and possibly anterior neuropore closure (20%).

MeOH embryopathies in C3H WT embryos. Exposure to 4 mg/ml MeOH for 24 hr resulted in dysmorphogenesis evidenced by a significant decrease in somite development (13%), with non-significant decreases in anterior neuropore closure (15%) and turning (23%), compared to saline-exposed WT controls (Fig. 2.4). This C3H WT strain was more resistant to MeOH embryopathies than the C57 WT strain, the latter of which exhibited a greater extent and severity of embryopathies.

MeOH embryopathies in aCat embryos. MeOH was highly embryopathic in aCat embryos, evidenced by significant decreases in anterior neuropore closure (100%), somite development

(21%) and head length (14%), along with a non-significant decrease in turning (27%), compared to saline-exposed aCat controls (Fig. 2.4).

Comparison of MeOH embryopathies in aCat and C3H WT embryos. aCat embryos were substantially more susceptible than WT controls to MeOH embryopathies, evidenced by decreased anterior neuropore closure (100%) (p < 0.05), yolk-sac diameter (13%) (p < 0.05) and crown-rump length (37%) (p = 0.05) in hCat embryos compared to MeOH-exposed WT controls

(Fig. 2.4).

188

100 Anterior Neuropore Closure 40 Somites Developed 80 (16) 30 ‡‡ (11) 60 (13) (11) (6) (5) ‡ 20 (2) 40 20 *‡ 10 (n, mean SD) +

PERCENTAGE (%) (11) 0 0 S M S M S M S M 4 100 Turning Crown-rump Length 80 (16) 3 (11) (6) b 60 (5) (13) (11) 2 (2) 40 (11) 20 1 (mm, mean + SD) + mean (mm, PERCENTAGE (%) PERCENTAGE 0 0 S M S M S M S M

5 Yolk Sac Diameter 200 4 Heart Rate 150 (16) (13) †† (11) 3 (13) (11) (11) (11)* (16) 100 2 1 50 (mm, mean + SD) mean + (mm, 0 0 S M S M SD) mean + (beats/min, S M S M

2.0 WT aCat Head Length 1.5 (7) (13) * (7) ‡‡‡ 1.0 (11)

0.5 (mm, mean + SD) + mean (mm, 0.0 S M S M WT aCat

Figure 2.4: Effect of catalase deficiency on MeOH embryopathies in aCat embryos.

On GD 9, WT (C3HeB/FeJ) embryos or aCat embryos containing 7 or 8 somite-pairs were explanted and incubated for 24 hr with MeOH (4 mg/ml) or its saline vehicle. Saline and MeOH treatments are indicated by ‘S’ and ‘M’ respectively. The number of embryos is given in parentheses. Double-daggers indicate a difference from saline control of the same genotype (‡ = p < 0.05, ‡ ‡ = p < 0.01, ‡ ‡ ‡ = p < 0.001), asterisks indicate a difference from wild-type MeOH exposed group (* = p < 0.05), letter ‘b’ indicates p = 0.05 compared to the WT MeOH-exposed group.

189

2.5 Discussion

The protection against MeOH embryopathies observed with enhanced endogenous catalase (hCat embryos), and the conversely enhanced susceptibility with a catalase deficiency

(aCat embryos) implicates ROS in the mechanism of MeOH developmental toxicity. This is the first evidence that an endogenous embryonic antioxidative enzyme, in this case catalase, plays an important role in protecting the developing embryo from MeOH embryopathies, as well as from the embryopathic effects of endogenous ROS. Although not all MeOH embryopathies were enhanced in aCat embryos, the almost complete protection in hCat embryos suggest that ROS play a major role in the molecular mechanism, and that the protective contribution of embryonic catalase is similarly substantial. This protective role of embryonic catalase is remarkable given that embryonic levels are only about 5% of maternal hepatic activity (Winn and Wells, 1997). A similar protective effect of endogenous embryonic catalase against endogenous and drug- enhanced oxidative stress in embryo culture has been observed in the hCat and aCat strains used herein when exposed to the antiepileptic drug phenytoin or its vehicle (Abramov and Wells,

2011).

The use of whole embryo culture was employed to discriminate the role of endogenous embryonic catalase by removing confounding maternal factors, including the metabolism of

MeOH and its formic acid metabolite by maternal catalase (Dorman et al., 1995). Maternal factors may have confounded the interpretation of related in vivo studies of MeOH teratogenicity using these same mouse strains, in which the protective role in hCat mice was equivocal, while both the aCat mice and their WT strain appeared resistant to MeOH teratogenicity (Siu and

Wells, 2010). In our studies herein, the C3H WT controls for the aCat mice were substantially more resistant to MeOH embryopathies than the C57BL/6 WT controls for the hCat strain. The

190 mechanism of resistance is not known, but this resistance, possibly along with maternal factors, may explain why the aCat mice in vivo exhibited no teratological effects of MeOH. This in vivo resistance is in contrast to our embryo culture studies herein, where aCat embryos were more susceptible than WT controls to the embryopathic effects of even endogenous oxidative stress, and dramatically more so to MeOH embryopathies.

Catalase is an antioxidative enzyme which detoxifies ROS by converting potentially toxic hydrogen peroxide to water and oxygen (Halliwell and Gutteridge, 2007). Despite the low level of embryonic catalase, this study and others demonstrate that catalase is a developmentally important enzyme protecting the embryo from both endogenous and xenobiotic-initiated ROS, which may adversely affect development by altering signal transduction pathways and/or oxidatively damaging cellular macromolecules (Sauer et al., 2001). Maintenance of a balance between ROS formation and detoxification is important for the developing embryo in which the control of numerous pathways involved in cell migration and differentiation are redox sensitive

(Moore and Persaud, 2007). Enhanced endogenous ROS has been shown to be embryopathic both in vivo and in vitro in whole embryo culture, wherein low levels of embryonic antioxidative enzymes such as glucose-6-phosphate dehydrogenase (G6PD) resulted in enhanced embryonic and fetal toxicity in the absence of xenobiotic exposure (Nicol et al., 2000), while protection against embryonic DNA oxidation and embryopathies was afforded by treatment with exogenous antioxidative enzymes including catalase and superoxide dismutase (SOD) (Winn and Wells,

1997; Winn and Wells, 1999; Winn and Wells, 1995).

MeOH itself is embryopathic both in vivo and in whole embryo culture (Abbott et al.,

1995; Andrews et al., 1993; Degitz et al., 2004; Rogers and Mole, 1997; Rogers et al., 1993), although the underlying mechanisms are unclear. Studies have shown that MeOH can form free

191 radicals (Kadiiska and Mason, 2000; Skrzydlewska et al., 2000) that can enhance the production of ROS, which in turn have been implicated in the mechanism of developmental toxicity (Harris et al., 2004).

In this study, enhanced levels of endogenous catalase protected hCat embryos from

MeOH embryopathies as evidenced by maintenance of somite development, and percentages of embryos that turned and closed their anterior neuropore compared to MeOH-exposed WT embryos, which exhibited significant decreases in each of these parameters. Additionally, catalase deficiency exacerbated MeOH embryopathies as evidenced by significant reductions in anterior neuropore closure, yolk sac diameter and head length in MeOH-exposed aCat embryos compared to saline-exposed embryos of the same genotype. Somite development is an important developmental parameter that relates directly to the growth of the embryo, and can be correlated to specific events throughout development. Decreased somite development is indicative of developmental delay and, given that embryonic systems have a finite period during which they may develop, MeOH-initiated decreases in somite development may lead to either the delay or complete failure of a system to develop to completion (Moore and Persaud, 2007). Enhanced endogenous catalase protected MeOH-exposed hCat embryos from decreases in somite development compared to WT MeOH-exposed embryos, consistent with the involvement of

ROS in the mechanism of MeOH developmental toxicity, and a protective role for embryonic catalase. While MeOH-exposed aCat and WT embryos exhibited significant decreases in somite development compared to saline-exposed embryos of the same genotype, there was only a small and statistically non-significant decrease MeOH-exposed aCat embryos compared to MeOH- exposed WT controls. This may be due in part to the apparent relative resistance of this strain to

MeOH embryopathies, and/or to an insufficient decrease in embryonic catalase in the aCat mice.

192

Similar to the protection observed for somite development by enhanced endogenous catalase levels, a 48% higher percentage of MeOH-exposed hCat embryos compared to their WT controls completed turning (69% vs. 21%, p < 0.05). Conversely, saline-exposed aCat embryos exhibited a non-significant trend for a 31% decrease in turning compared WT controls, and an even greater but still non-significant 60% decrease was observed in MeOH-exposed aCat embryos compared to WT controls. These consistent trends in aCat mice for both saline and

MeOH-exposed embryos suggest the decreases may become significant with the analysis of a larger number of embryos. Embryonic turning involves the folding of the embryo thus establishing body form necessary for development to continue (Moore and Persaud, 2007). These findings suggest that redox balance may be an important factor in embryonic turning, and that embryonic catalase is an important regulator.

Enhanced endogenous catalase completely protected hCat embryos from MeOH-initiated decreases in the closure of their anterior neuropore, while catalase deficiency caused 100% failure of anterior neuropore closure in MeOH-exposed aCat embryos. This suggests that ROS are an important risk factor for neural tube defects in the developing embryo, and that the embryonic level of catalase in particular, and possibly other antioxidative enzymes, play a major protective role. Failure of the anterior neuropore to close can manifest clinically as neural tube defects such as exencephaly and anencephaly, which ultimately result in death of the embryo

(Moore and Persaud, 2007).

MeOH-initiated reductions in embryonic head length were similarly enhanced respectively in aCat embryos compared to both saline-exposed aCat embryos, and MeOH- exposed WT embryos. This is not surprising given that ethanol, a structurally similar alcohol, specifically targets craniofacial cells leading to the characteristic facial features of the Fetal

193

Alcohol Spectrum Disorder (Sulik, 2005). MeOH may act through a similar mechanism.

Although hCat embryos exposed to MeOH exhibited an increase in crown-rump length over their saline controls, the difference was slight, and not different from the WT MeOH-exposed embryos; moreover, crown-rump length for the saline-exposed hCat embryos was less than that for saline-exposed WT embryos.

Yolk sac diameter is an indicator of the amount of visceral yolk-sac present, which is important for nutrient transport as well as protecting the embryo by creating a selective barrier for ions and substrates such as xenobiotics (Hansen et al., 2005). It was interesting that MeOH- exposed aCat embryos exhibited larger yolk-sac diameter than saline-exposed aCat embryos. The significance of this increase is unclear, but might constitute a compensatory mechanism, whereby the deficiency in the ability to detoxify MeOH-enhanced ROS might be partially mitigated by a larger surface area of yolk sac that provides enhanced protection via glutathione- dependent pathways and/or efflux transporters.

MeOH increased heart rate in hCat and their C57BL/6 WT controls, but not in aCat or their C3H WT controls, indicating a differential strain susceptibility, although this affect was not modulated by altered catalase activity. Although the effect of MeOH on heart rate in the developing embryo has not been investigated, a similar alcohol, ethanol, induces tachycardia in adult rats. This has been attributed to direct stimulation of sympathetic nerves innervating the heart (Sparrow et al., 1987), but the mechanism in embryos remains to be determined.

It is not surprising that some parameters were not affected by alterations in catalase expression, demonstrating selectivity in the developmental effects of MeOH-initiated ROS formation and the protective role catalase, both of which can vary by tissue, cell type and intracellular localization. Selectivity with a different pattern has been observed elsewhere in

194 embryo culture studies with the ROS-initiating teratogen phenytoin (Abramov and Wells,

2011b). Herein, had all parameters been equally affected, this might have suggested an excessively toxic concentration of MeOH. The selectivity also may have been due in part to differential redox signalling in embryonic tissues over the course of the culture period, rendering only some parameters sensitive to MeOH-initiated ROS (Ufer et al., 2010).

In summary, our results suggest that ROS production may contribute to the mechanism of

MeOH developmental toxicity in mice, and that the level of embryonic catalase, although low, is an important determinant of risk.

195

Chapter 3 Enhanced NADPH oxidases and reactive oxygen species in the mechanism of methanol-initiated protein oxidation and embryopathies in mouse embryo culture

Running title: Methanol embryopathies and NADPH oxidases

Lutfiya Miller†, Amy Sharma* and Peter G. Wells*†

†Department of Pharmacology and Toxicology, and

*Faculty of Pharmacy

University of Toronto

Toronto, Ontario, Canada

a. Preliminary reports of this research were presented at the 2012 annual meeting of the Society of Toxicology (U.S.A.) [The Toxicologist,126(1):417-418 (Abstract No. 1940)]. These studies were supported by a grant from the Canadian Institutes of Health Research. b. Individual contributions: Lutfiya Miller- dosing and sample collection, embryo culture, mRNA analysis, protein oxidation; Amy Sharma- western blotting; Peter G. Wells- supervisor.

196

3.1 Abstract

Methanol (MeOH) teratogenicity in rodents may be mediated in part by reactive oxygen species (ROS), the source of which is unknown. To determine if MeOH enhances embryonic

ROS-producing NADPH oxidases (NOXs), embryonic p22phox mRNA and protein and oxidatively damaged protein were measured in (GD) 12 MeOH-exposed CD-1 mouse embryos with or without pretreatment with the free radical spin trap phenylbutylnitrone (PBN) or the

NADPH oxidase inhibitor diphenyleneiodonium chloride (DPI). MeOH exposure upregulated p22phox mRNA and protein expression, and enhanced protein oxidation, within 3-6 hr.

Compared to embryos exposed to MeOH alone, PBN and DPI pretreatment decreased MeOH- enhanced p22phox mRNA expression, DPI but not PBN blocked p22phox protein expression, and both blocked protein oxidation. To assess developmental relevance, mouse embryos were exposed in culture for 24 hr to MeOH or vehicle with or without pretreatment with PBN, DPI or the prostaglandin H synthase (PHS) inhibitor eicosatetraynoic acid (ETYA), and evaluated for abnormalities. ETYA did not prevent MeOH embryopathies, despite blocking phenytoin embryopathies (ROS-initiating positive control), precluding bioactivation of MeOH or its metabolites by PHS. Concentration-dependent MeOH embryopathies were blocked by both DPI and PBN pretreatment, suggesting that enhanced embryonic NOX-catalyzed ROS formation and oxidative stress may contribute to the mechanism of methanol embryopathies.

Key Words: Methanol; phenylbutylnitrone; free radical spin trap; diphenyleneiodonium;

NADPH oxidases; prostaglandin H synthase; phenytoin; embryopathies; oxidative stress; reactive oxygen species; embryo culture

197

3.2 Introduction

The use of methanol (MeOH) as an alternative fuel source would result in increased environmental MeOH exposure to humans (Harris et al., 2004), with a potential risk for the developing fetus during pregnancy. In utero exposure to the structurally similar alcohol, ethanol

(EtOH) can initiate a spectrum of structural and functional deficits in the developing child termed Fetal Alcohol Spectrum Disorder (FASD), however the developmental toxicity of MeOH in humans is less well understood. MeOH like EtOH is a developmental toxicant in rats (Nelson et al., 1985) and mice both in vivo and in whole embryo culture (Abbott et al., 1995; Andrews et al., 1993; Degitz et al., 2004; Rogers and Mole, 1997; Rogers et al., 1993), however EtOH and

MeOH have very different developmental toxicities and toxicities in adults. In particular, while the evidence for oxidative stress in the teratogenicity of EtOH is much more consistent than it is for MeOH, alimited evidence implicates reactive oxygen species (ROS) in the mechanism of

MeOH teratogenicity (Wells et al., 2013), including exacerbation of MeOH embryopathies in rat embryo culture by depletion of the antioxidative peptide glutathione (GSH) (Harris et al., 2004) and respective enhancement or prevention of MeOH embryopathies in genetically altered cultured mouse embryos with deficient or enhanced catalase activity (Miller and Wells, 2011).

Accordingly, one would not necessarily expect the same biochemical effects from these two seemingly similar alcohols.

ROS mediate a number of fundamental cellular processes during embryonic development including proliferation, differentiation and migration (Allen and Tresini, 2000; Thannickal and

Fanburg, 2000; Ufer et al., 2010). Oxidative stress occurs when ROS production exceeds antioxidative capacity, resulting in excessive ROS, which can be harmful either by altering signal transduction pathways, or by oxidatively damaging cellular macromolecules that are important

198 for normal embryonic development (Halliwell and Gutteridge, 2007; Wells et al., 2009b). Free radical intermediates and ROS have been implicated in the teratological mechanism of other xenobiotics like phenytoin (Abramov and Wells, 2011a; Abramov and Wells, 2011b; Parman et al., 1998), benzo[a]pyrene (Winn and Wells, 1997), thalidomide (Arlen and Wells, 1996; Lee et al., 2011; Wells et al., 2009b), methamphetamine (McCallum et al., 2011c; Wong et al., 2008) and EtOH (Dong et al., 2008; Dong et al., 2010).

In addition to the bioactivation of xenobiotics to ROS-initiating free radical intermediates

(Wells et al., 2009b), potential sources of ROS in the developing embryo include the mitochondrial electron transport chain, xanthine oxidases and NADPH oxidases (NOXs) (Brown and Griendling, 2009; Coso et al., 2012; Ufer et al., 2010; Wu and Cederbaum, 2003). NOX enzymes are involved in signal transduction pathways that regulate growth, differentiation, migration and apoptosis (Brown and Griendling, 2009), and their primary function is the

•- production of ROS, specifically O2 and H2O2 produced via the oxidation of NADPH (Lalucque and Silar, 2003; Lambeth, 2004). Recently, embryonic NOXs have been shown to be a notable source of ROS in the developmental toxicity of EtOH, contributing to DNA oxidation, apoptosis and structural teratogenesis, all of which were inhibited by the NOX inhibitor diphenyleneiodonium (DPI) (Dong et al., 2010). Given the similarity of EtOH and MeOH in their ability to initiate ROS formation, and exhibit similar structural embryopathies in whole embryo culture and in vivo, we speculated that embryonic NOXs may be similarly involved in the mechanism of developmental toxicity for MeOH, specifically in mediating MeOH-initiated

ROS production.

To determine the contribution of NOX-initiated ROS and oxidative stress to the mechanism of MeOH developmental toxicity, we examined the pattern of embryonic NOX

199 mRNA and protein expression, and oxidatively damaged protein, after in utero MeOH exposure.

We determined whether these changes were related to MeOH-initiated structural embryopathies in embryo culture by pretreating pregnant dams with the NOX inhibitor diphenyleneiodonium

(DPI) and the free radical spin trapping agent phenylbutylnitrone (PBN) in vivo, and similarly exposing embryos in culture to MeOH with or without DPI and PBN pretreatment. We also evaluated the potential involvement of MeOH bioactivation by prostaglandin H synthases

(PHSs), which can bioactivate other teratogens to ROS-initiating reactive intermediates, by pretreating MeOH-exposed embryos in culture with the PHS inhibitor eicosatetraynoic acid

(ETYA). MeOH rapidly induced embryonic NOX, as determined by p22phox mRNA and protein, and oxidatively damaged embryonic proteins, and these effects and embryopathies in culture were blocked by PBN and DPI, but not by ETYA, the latter suggesting that PHS does not bioactive MeOH or its metabolites. Our results indicate that embryonic ROS formation and oxidative stress may be involved in the mechanism of MeOH embryopathies, and that embryonic

NOXs may be an important source for ROS production.

200

3.3 Methods

Animal Breeding. Virgin female CD-1 mice and male mice (Charles River Canada,Saint-

Constant, Quebec) were housed in plastic cages with ground corncob bedding (Beta Chip;

Northeastern Products, Warrensburg, NY). The cages were maintained in a temperature- controlled room with a 14-hr light/10-hr dark cycle. Food (Laboratory Rodent Chow 5001; PMI

Feeds, St. Louis, MO) and tap water were provided ad libitum. Mice were acclimatized for 1 week before use. One male was housed with three females for 2 hr. The presence of a vaginal plug after 2 hr was designated gestational day (GD) 0. Pregnant females were housed together in groups of four or less per cage. One group of pregnant females was sacrificed on GD 8 for embryo culture studies, while a second group of pregnant females was sacrificed on GD 12 for analysis of NOX mRNA and protein expression, and protein oxidation. 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.

Chemicals. Fetal bovine serum, Hanks balanced salt solution, Waymouth’s MB 752/1 medium,

HEPES, and penicillin- streptomycin were from Gibco Laboratories (Toronto, Ontario). α-

Phenyl-N-tert-butylnitrone (PBN), diphenyleneiodonium chloride (DPI), 5,8,11,14-

Eicosatetraynoic acid (ETYA), phenytoin (diphenylhydantoin sodium salt) and N,N- dimethylformamide (DMF) were from Sigma Aldrich (St. Louis, MO, US). HPLC grade MeOH was purchased from EMD Sereno Canada, Inc. (Mississauga, ON). Saline (0.9%, sterile) was purchased from Baxter Corporation (Mississauga, ON). Compressed CO2 (5% in air) was purchased from Linde Canada (Mississauga, ON).

201

RNA Extraction and semiquantitative reverse-transcription PCR for NOX mRNA. On GD

12, pregnant CD-1 dams were dosed with MeOH (4 g/kg i.p) with or without 30 min pretreatment with PBN (40 mg/kg, i.p) or DPI (4 mg/kg, i.p). GD 12 embryos were used to permit an assessment of NOX mRNA expression in single embryos while still in the embryonic period. The dose of MeOH was based on previous studies from our laboratory (Siu et al., 2013b), with a single administration to facilitate mechanistic investigations uncomplicated by secondary effects from multiple dosing. The dose of PBN was based on previous studies demonstrating protective efficacy against phenytoin-initiated protein oxidation in embryonic microsomes (Liu and Wells, 1994), thalidomide-initiated DNA oxidation and birth defects in rabbit embryos

(Parman et al., 1999), and ethanol-initiated fetal brain DNA oxidation and postnatal behavioural deficits (Miller et al., 2013b). The dose range for DPI encompassed that previously shown to inhibit NOX activity and mRNA expression in vivo (Dong et al., 2010). Dams were sacrificed 3,

6 and 12 hr post-injection, and embryos were removed and analyzed for p22phox mRNA expression using a reverse-transcription PCR assay. RNA was extracted using the Qiagen

RNeasy mini kit. cDNA was synthesized using OligodT primer (Invitrogen, Burlington, ON).

P22phox and GADPH cDNA were amplified using a PCR based assay with the following primers: p22phox forward: 5'-CGTGGCTACTGCTGGACGTT-3'; p22phox reverse: 5'-

GCACACCTGCAGCGATAGAG-3'; GAPDH forward 5'-AACGACCCCTTCATTGAC-3' and

GAPDH reverse 5'-TCCACGACATACTCAGCAC-3'. Cycling conditions were: 95°C for 5 min;

35 cycles of: 94°C for 1 min, 55°C for 1.5 min, 72°C for 2 min and completed with a final extension at 72°C for 10 min and then placed on hold at 4°C. 6X DNA loading dye (0.03% bromophenol blue/0.03% xylene cyanol FF) was added to each sample. The PCR products were separated on a gel consisting of 1.5% (w/v) agarose, 40mM Tris, 19.4mM glacial acetic acid,

2.5mM EDTA and 8 μl ethidium bromide. The agarose gel was run at 100 V for 60 min, and

202 then visualized and photographed under an ultraviolet light. P22phox band density was measured and standardized to GAPDH band density to quantify p22phox mRNA levels.

SDS-PAGE and Immunoblotting of embryonic tissue for p22phox protein expression. On

GD 12, pregnant CD-1 dams were dosed with MeOH (4 g/kg i.p) with or without 30 min pretreatment with PBN (40 mg/kg, i.p) or DPI (4 mg/kg, i.p), dams were sacrificed 3, 6 and 12 hr post-injection, and embryos were removed and analyzed for p22phox protein expression by western blot. Embryos were homogenized in working cell lysis RIPA buffer (Cell Signaling

Technologies, Pickering, ON) containing 1X HALT Protease Inhibitor Cocktail (Pierce,

Rockford, IL) with a Polytron 2100 homogenizer and centrifuged at 10000 × g for 10 min and supernatant was collected and again centrifuged at 10 000 × g for 30 min. The protein concentrations of the supernatants were determined with the Bicinchoninic acid (BCA) assay

(Pierce, Rockford, IL). The supernatant was mixed with Pierce reducing sample loading buffer in a 4:1 protein to buffer ratio and boiled for 5 min. Sodium dodecyl sulfate polyacrylamide gel electrophoresis was performed using the Protean-3 minigel system (BioRad, Mississauga, ON).

Gels were pre-cast from Bio-Rad Canada (4-20% gradient), and were run at 110 V.

Electrophoresis running buffer (Bio-Rad) consisted of 25 mM Tris base, 192 mM glycine, and

0.1% sodium dodecyl sulfate, pH 8.3. Transfer to nitrocellulose membrane (0.2 µM, BioRad) occurred at 0.10 mA for 90 min at 4ºC using the same Protean-3 minigel system (BioRad,

Mississauga, ON). Tris-glycine transfer buffer (Bio-Rad) consisted of 25 mM Tris, 192 mM glycine, and 20% methanol at pH 8.5. Membranes were washed twice in tris-buffered saline tween-20 (TBST) wash solution for 5 min. Membranes were then blocked in 5% non-fat milk blocking solution in TBST. Blocking was done for 90 min at room temperature. Membranes

203 were then rinsed with three changes of TBST for 5 min each and incubated with 1.5 µg/ml of primary anti-cytochrome b245 Light Chain/p22phox polyclonal antibody (Abcam ab75941) and

10% normal goat serum in TBST overnight at 4 ºC. This antibody detects two distinct bands (19 and 21 kDa), but only the 21kDa band was used for analysis (Palm et al., 2010; Sakellariou et al., 2013). The 19kDa band is a possible cleavage fragment. A 60 min wash (ten minute intervals) in TBST after overnight blocking was followed by a 60 min incubation in secondary antisera (1:10 000 dilution) in TBST containing 10% goat serum. The secondary antisera was goat anti-rabbit horseradish peroxidase antisera (Sigma). Membranes were washed 4 times for 20 min with TBST. All blots were incubated with enhanced chemiluminescence stain for 5 min and analyzed with a FluorChem8800 imager. To probe for the glyceraldehyde 3-phosphate dehydrogenase (GAPDH) loading control, membranes were stripped of primary anti-cytochrome b245 Light Chain/p22phox using Pierce Restore Plus buffer (Pierce, Rockford, IL) for 15 to 20 min at room temperature followed by a one hour blocking step. Membranes were then incubated in mouse monoclonal anti-GAPDH antisera (1:40,000) and processed as above except the secondary antisera was goat anti-mouse horseradish peroxidase antisera diluted 1:10,000

(Jackson ImmunoResearch, Baltimore Pike, West Grove, PA.).

Protein Oxidation. On GD 12, pregnant CD-1 dams were dosed with MeOH (4 g/kg i.p) with or without 30 min pretreatment with PBN (40 mg/kg, i.p) or DPI (4 mg/kg, i.p), dams were sacrificed 3, 6 and 12 hr post-injection, and embryos were removed and analyzed for protein oxidation using a fluorometric protein carbonyl kit (Cayman Labs) and carried out according to manufacturer’s instructions. Briefly, samples were homogenized in 100 mM Tris pH 7.4, and protein concentrations were determined by the Bradford Assay using bovine serum albumin as a

204 standard (BioRad, Hercules, CA). Samples were adjusted to 5-10 mg/ml protein, and were then incubated with 50 μL of 200 μM fluorophore overnight at room temperature protected from light.

20% TCA solution was added to each sample, and washed 3 times with acetone. Samples were derivitized with Guanidine HCl, and analyzed for fluorescence at excitation wavelength 485 nm and emission wavelength 535 nm. Protein oxidation was calculated by fluorescence fit within a standard curve of fluorescence.

Embryo Culture. On GD 8.5, pregnant dams were sacrificed by cervical dislocation and cultured as previously described (Miller and Wells, 2011). Although protein oxidation and NOX expression were assessed in GD 12 embryos, p22phox mRNA and protein expression are evident on both GDs 8 and 12 (Baehner et al., 1999; Dong et al., 2010). Embryos dissected at the 6- somite stage were washed in pregassed (5% CO2 in air) holding medium (HM) (17 ml male rat serum, 2.5mM HEPES, 50 units/ml of penicillin, 50 mg/ml of streptomycin, and Waymouth’s

MB 752/l medium) and transferred to a 48–well plate (BD Biosciences, Franklin Lakes, NJ) containing either PBN or its 80:20 saline:water vehicle, or DPI (0.5 – 6 M) or its DMF vehicle in 1 mL embryo culture medium (ECM; HM supplemented with 15 mL fetal bovine serum). The concentration range for DPI was chosen in part to encompass efficacy in preventing MeOH embryopathies, and was within the range shown to inhibit NOX activity and superoxide production in vitro (Cross and Jones, 1986). A range of PBN concentrations (0.125, 0.187, 0.22,

0.44, 1.0 and 2.2 mM) was tested initially as no previously published studies have used PBN in whole embryo culture, with 1.0 mM producing embryotoxicity, and 2.2 mM causing embryolethality (data not shown). PBN concentrations varying from 0.125-0.22 mM produced equivalent protective efficacy, so the data from these concentration groups were pooled. Each

205 plate well was sealed air tight with optically clear sealing tape (Sarstedt, Montreal, QC, Canada).

After 30 min, embryos were washed briefly in HM, and transferred to a 24-well plate (BD

Biosciences, Franklin Lakes, NJ) containing 2 mL ECM (1 embryo per well) containing either

MeOH (4 or 6 mg/ml), saline vehicle, phenytoin (20 μg/mL) or its vehicle (0.002N NaOH) for a total of 24 hr. Concentrations of MeOH were based on a midrange concentration from previous studies and encompassed the LC50 (Abbott et al., 1995; Andrews et al., 1993; Degitz et al.,

2004; Hansen et al., 2005; Miller and Wells, 2011). The concentration of phenytoin was chosen based on published data (Abramov and Wells, 2011b; Miranda et al., 1994; Winn and Wells,

1995). For PHS inhibition, embryos were co-treated with ETYA (40 μM) in 2 mL ECM in a 24- well plate. The concentration of ETYA was based on studies demonstrating protective efficacy

(Miranda et al., 1994). The 24-well plate was incubated at 37ºC in a Sanyo model MCO-17A

CO2 incubator (Sanyo Electric Co., Ltd, Japan) on a platform rocker (Bellco Biotechnology,

Vineland, NJ). After 24 hr, embryonic morphological and developmental parameters were observed using a dissecting microscope (Stemi SV11, Carl Zeiss, Oberkochen, Germany).

Developmental parameters included dorsal-ventral flexure (turning), anterior neuropore closure and somite development. Functional and morphological assessments included heart rate, yolk sac diameter (in mm), crown-rump length (in mm) and head length (in mm). Heart rate was measured in culture by counting the beats per minute in real time. Overall lethality was measured in all groups as a percentage. An embryo was considered dead if it had no heart beat.

Statistical analysis. Statistical significance between treatment groups was determined using

GraphPad Prism®, Version 5 (GraphPad Software, Inc.). Continuous data (p22phox mRNA and protein expression, protein carbonyl content, somites developed, crown-rump length, yolk sac

206 diameter, heart rate and head length) were analyzed by one-way ANOVA with a post-hoc

Bonferroni test. Binomial data were analyzed by chi-square test (anterior neuropore closure, turning, lethality).

207

3.4 Results

MeOH-initiated upregulation of p22phox mRNA

MeOH exposure increased p22phox NOX mRNA expression by 10% at 3 hr (p < 0.001), which increased to a maximum of 20% at 6 hr (p < 0.001), and remained elevated at 12 hr with a 10% increase compared to vehicle-treated embryos (p < 0.001) (Fig. 3.1, upper panel).

Effect of PBN and DPI on MeOH-initiated upregulation of p22phox mRNA upregulation

Compared to MeOH-exposed embryos, PBN- and DPI pretreatment decreased MeOH-initiated p22phox mRNA upregulation by 40% (p < 0.01) and 83% (p < 0.001), respectively (Fig. 3.1, lower panel).

MeOH-initiated upregulation of p22phox protein expression

MeOH-exposure increased p22phox NOX protein expression by 60% at 3 hr (p < 0.001), by

1.56-fold at 6 hr (p < 0.001) and by 1.4-fold at 12 hrs (p < 0.01) compared to vehicle-treated embryos (Fig. 3.2, upper panel).

Effect of PBN and DPI on MeOH-initiated upregulation of p22phox protein expression

Compared to MeOH-exposed embryos, PBN-pretreatment resulted in a nonsignificant 34% decrease in p22phox protein expression, while pretreatment with DPI resulted in a 67% decrease

(p < 0.05) (Fig. 3.2, lower panel).

208

GAPDH (40.2 kDa)

p22phox (21 kDa)

††† 1.4 ††† (9) ††† 1.2 (11) (7) (9) 1.0 0.8 (%of control) 0.6 normalized to GAPDH to normalized

p22phox mRNA expression p22phox mRNA Veh 3hr 6hr 12hr MeOH

GAPDH (40.2 kDa) p22phox (21 kDa)

1.4 ††† (11) ** 1.2 (3) ***(3) (9) 1.0 0.8 (%of control) 0.6 normalized to GAPDH to normalized

p22phox mRNA expression Veh 6hr +PBN +DPI MeOH

209

P22 phox co-localizes with NOX2 and is required for NOX function. Upper panel: On gestational day (GD) 12, pregnant CD-1 dams were dosed with MeOH (4 g/kg i.p.), dams were sacrificed at 3, 6 and 12 hr post-injection, embryos were removed and snap-frozen for p22phox mRNA analysis. Lower panel: On GD 12, pregnant CD-1 dams were dosed with MeOH (4 g/kg i.p.), with or without 30 min pretreatment with either the free radical spin trap PBN (40 mg/kg, i.p) or the NOX inhibitor DPI (4 mg/kg, i.p). The number of embryos is given in parentheses.

Embryos were taken from 3-6 different litters. Data were analyzed by a one-way ANOVA with a post-hoc Bonferroni test. Single dagger indicates a difference from vehicle-treated embryos (††† p < 0.001). Asterisks indicate a difference from MeOH-exposed embryos (** p < 0.01, *** p <

0.001).

210

GAPDH (40.2 kDa)

p22phox (21 kDa) p22phox (19 kDa) ††† ††† (9) 2.0 (9) †† (7) 1.5 (8) 1.0

(%of control) (%of 0.5

normalized to GAPDH 0.0

p22phox protein expression p22phox protein Veh 3hr 6hr 12hr MeOH

GAPDH (40.2 kDa)

p22phox (21 kDa) p22phox (19 kDa)

††† 2.0 (9) (9) (9)* 1.5 (8) 1.0

(%of control) (%of 0.5

normalized to GAPDH 0.0

p22phox protein expression p22phox protein Veh 6hr +PBN +DPI MeOH

211

Figure 3.2: Upregulation of embryonic p22phox protein by in utero MeOH exposure and the effect of pretreatment with PBN and DPI.

Upper panel: On GD 12, pregnant CD-1 dams were dosed with MeOH (4 g/kg i.p.), dams were sacrificed at 3, 6 and 12 hr post-injection, embryos were removed and snap-frozen for p22phox protein analysis. The p22phox antibody detects 2 bands (19 and 21kDa), but only the 21kDa band was quantified for analysis. The 19kDa band is a possible cleavage fragment. Lower panel:

On GD 12, pregnant CD-1 dams were dosed with MeOH (4 g/kg i.p), with or without 30 min pretreatment with either PBN (40 mg/kg, i.p) or DPI (4 mg/kg, i.p). The number of embryos is given in parentheses. Embryos were taken from 3-6 different litters. Data were analyzed by a one-way ANOVA with a post-hoc Bonferroni test. Single daggers indicate a difference from vehicle-treated embryos (†† p < 0.01, ††† p < 0.001). Asterisk indicates a difference from

MeOH-exposed embryos (* p < 0.05).

212

MeOH-initiated protein oxidation

MeOH increased embryonic protein oxidation as measured by protein carbonyls by 74% at 3 hr (p < 0.001) reaching a maximum at 6 hr with a 93% increase, returning back to baseline levels by 12 hr (Fig. 3.3, upper panel)

Effect of PBN and DPI on MeOH-initiated protein oxidation

Compared to MeOH-exposed embryos, pretreatment with PBN and DPI decreased protein oxidation by 42% (p < 0.001) and 49% (p < 0.001), respectively (Fig. 3.3, lower panel).

Concentration-dependent MeOH embryopathies

Embryos exposed to MeOH at 4 and 6 mg/ml for 24 hr exhibited concentration- dependent dysmorphogenesis as evidenced by significant decreases in anterior neuropore closure

(ANC) (45% and 83%, respectively), somite development (17% and 20%, respectively), turning

(35% and 73%, respectively) and head length (20% and 13%, respectively), along with significant increases in heart rate (46% and 50%, respectively) and yolk sac diameter (15% at 4 mg/ml) (Figs. 3.4, 3.5). MeOH increased embryonic lethality in a concentration-dependent fashion, with a nonsignificant 6% increase at 4 mg/ml (p > 0.05), and a 27% increase at 6 mg/ml

(p < 0.05) compared to control embryos exposed to saline vehicle (Fig. 3.5).

213

††† (13) 0.6 ††† (13) 0.4 (13) (13)

0.2

0.0

Protein carbonyl content carbonyl Protein Veh 3hr 6hr 12hr (nmol/mg protein, mean + SD) + mean protein, (nmol/mg MeOH

††† 0.6 (13)

0.4 ***(6) (13) ***(6)

0.2

0.0

Protein carbonyl content carbonyl Protein Veh 6hr +PBN +DPI (nmol/mg protein, mean + SD) + mean protein, (nmol/mg MeOH

214

Figure 3.3: Enhanced oxidatively damaged embryonic protein following in utero MeOH exposure and the effect of pretreatment with PBN and DPI.

Upper panel: On GD 12, pregnant CD-1 dams were dosed with MeOH (4 g/kg i.p), dams were sacrificed at 3, 6 and 12 hr post-injection, embryos were removed and snap-frozen for analysis of protein oxidation measured as protein carbonyls. Lower panel: On GD 12, pregnant CD-1 dams were dosed with MeOH (4 g/kg i.p), with or without 30 min pretreatment with either PBN

(40 mg/kg, i.p) or DPI (4 mg/kg, i.p). The number of embryos is given in parentheses. Embryos were taken from 3-6 different litters. Data were analyzed by a one-way ANOVA with a post-hoc

Bonferroni test. Single daggers indicate a difference from vehicle-treated embryos (††† p <

0.001). Asterisks indicate a difference from MeOH-exposed embryos (*** p < 0.001).

215

CLOSED ANP OPEN ANP

YOLK SAC DIAMETER

CROWN-RUMP HEAD LENGTH LENGTH NOT TURNED SOMITES

TURNED

216

Figure 3.4: Developmental endpoints assessed in mouse whole embryos in culture.

Embryos were explanted on gestational day (GD) 8.5 (plug = GD 0) and cultured for 24 hr.

During this period, the embryo would normally “turn” from an elongated form to the “fetal position” (turning), and the anterior neuropore (ANP) at the top of the head would close.

Somites are paired segmental structures along the back of the spine that develop into the bones, muscle and dermis of the axial skeleton. Embryos with 6 somite pairs are selected, and the number of somite pairs developing during the culture period, normally at a rate of about 1 pair per hour, constitutes an accurate and highly reproducible “clock” indicating the stage of embryonic development.

217

150 35 Anterior Neuropore Closure Somites Developed 30 100 * MeOH PBN + MeOH †† 25 ** ** 50 20 †† †† ††† (n, mean SD) + PERCENTAGE (%) PERCENTAGE 0 15

3.0 150 Turning Crown-rump Length

2.5 100 * ††† 2.0 50 ††† (n, mean + SD) mean+ (n, PERCENTAGE (%) PERCENTAGE 0 1.5

4.0 200 Yolk Sac Diameter Heart Rate 3.5 150 ††† 3.0 †† ** 100 2.5 †† 2.0 † 50 (mm, mean + SD) + mean (mm, 1.5 0 SD) mean + (beats/min, 2.0 60 Head Length Lethality 1.5 40 * † 1.0 † 20 0.5 †† (mm, mean + SD) + mean (mm, PERCENTAGE (%) PERCENTAGE 0.0 0 0 4 6 mg/ml 0 4 6 mg/ml ______0 124 187 mM ______0 124 187 mM METHANOL CONCENTRATION METHANOL CONCENTRATION

218

Figure 3.5: Effect of pretreatment with PBN on concentration-dependent MeOH embryopathies in embryo culture.

On GD 8.5, CD-1 embryos were explanted and incubated for 24 hr with increasing concentrations of MeOH (4 or 6 mg/ml) or its saline vehicle, with or without a 30-min preincubation with 0.125-0.22 mM PBN. The protective efficacy of PBN was equivalent at concentrations of 0.125, 0.187 and 0.22 mM, and these data were combined. Embryos with 6 somite pairs were used for culture. Each group consisted of 7-40 embryos that came from 7-16 different litters. Continuous data (somite development, crown-rump length, yolk sac diameter, heart rate and head length) were analyzed by one-way ANOVA with a post-hoc Bonferroni test.

Binomial data (anterior neuropore closure, turning, lethality) were analyzed by chi-square test.

Daggers indicate a difference from saline-exposed embryos († p < 0.05, †† p < 0.01, ††† p <

0.001). Asterisks indicate a difference from MeOH-exposed embryos at the same concentration

(* = p < 0.05, ** = p < 0.01).

219

Effect of PBN and DPI on MeOH embryopathies

At the lower concentration of MeOH (4 mg/ml), PBN pretreatment protected embryos against MeOH embryopathies as evidenced by a 52% increase in turning (p < 0.05), a 79% increase in anterior neuropore closure (p < 0.05), a 25% increase in somite development (p <

0.01) and a 21% increase in head length (p < 0.05) (Fig. 3.5), while DPI pretreatment (0.5 M) increased anterior neuropore closure by 36% (p < 0.05), somite development by 42% (p < 0.001) and head length by 25% (p < 0.001) compared to MeOH-exposed embryos (Fig. 3.6). DPI pretreatment also increased turning by 35% but this increase was not significant.

The protective efficacy of DPI declined with higher concentrations (1 μM-6 μM), and conversely increased MeOH embryopathies in a DPI concentration-dependent manner, as evidenced by 100% failure of embryos to close their anterior neuropore at 6 μM (p < 0.01), with

3.4-fold and 7.1-fold increases in lethality at 3 (p < 0.05) and 6 μM (p < 0.001), respectively, while somite development increased by 35% (p < 0.001), 20% (p < 0.01) and 23% (p < 0.01) at

1, 3 and 6 uM, respectively compared to MeOH-exposed embryos (Fig. 3.6).

At the higher MeOH concentration (6 mg/ml), PBN pretreatment was protective, enhancing somite development by 29% (p < 0.01), and turning by 2.4-fold, although the latter increase was not significant (p = 1.095) (Fig. 3.5). PBN also increased heart rate by 49% compared to saline-exposed embryos (p < 0.05). MeOH-initiated lethality at the higher concentration was reduced by PBN pretreatment by 39%, although this apparent protection was only marginally significant (0.05 < p < 0.1).

220

150 Anterior Neuropore Closure 60 Somite Development

100 * *** (14) (13) (14) 40 ††† † ††† (11) ***(11) (16) (12) **(4) **(3) 50 (10) (8) 20 * + SD) mean (n, PERCENTAGE (%) PERCENTAGE (5) 0 0 150 5 Turning Crown-Rump Length 4 100 (14) (13) (14) 3 (12) (10) (11) (11) (16) (5) (4) (3) (8) 2 50 1 PERCENTAGE (%) (mm, mean + SD) mean (mm, 0 0 250 Heart Rate 6 Yolk Sac Diameter 200 †† 150 4 (16) (13) (16) (13) (9) (9) (6) (14) ***(8) (6) 100 (14) **(8) 2 50 (mm, mean + SD) 0 0 (beats/min, mean + SD) mean (beats/min, 2.5 Head Length 60 Lethality 2.0 ***(12) 1.5 †† 40 (14) ***(13) (14) (16) (8) 1.0 (6) (10)* 20 (11) 0.5 (15) (17) (14) (mm, mean +SD) 0.0 (%) PERCENTAGE 0 VH ______0 0.5 1 3 6 VH ______0 0.5 1 3 6 DPI Concentration (M) DPI Concentration (M) ______VH METHANOL VH METHANOL

221

Figure 3.6: Effect of pretreatment with DPI on MeOH embryopathies in embryo culture.

On GD 8.5, CD-1 embryos were removed and incubated for 24 hr with MeOH (4 mg/ml) or its saline vehicle (VH), with or without 30 min pretreatment with DPI (1-6 μM). The number of embryos is given in parentheses. Embryos with 6 somite pairs were used for culture. Embryos were taken from 3-14 different litters. Continuous data (somite development, crown-rump length, yolk sac diameter, heart rate and head length) were analyzed by one-way ANOVA with a post- hoc Bonferroni test. Binomial data (anterior neuropore closure, turning, lethality) were analyzed by chi-square test. Daggers indicate a difference from vehicle-exposed embryos († p < 0.05, †† p < 0.01, ††† p < 0.001). Asterisks indicate a difference from embryos exposed MeOH alone (* p < 0.05, ** p < 0.01, *** p < 0.001). Embryos with 6 somite-pairs were used for culture.

Crown-rump length and somite development were only scored if the embryo had turned.

222

Effect of eicosatetraynoic acid (ETYA) on MeOH and phenytoin embryopathies

Compared to embryos exposed to MeOH alone, co-treatment with ETYA had no effect on any MeOH embryopathies (Fig. 3.7). To confirm that the absence of an effect of ETYA on

MeOH embryopathies was not due to methodological error, we also evaluated the effect of

ETYA on phenytoin (PHT) embryopathies as a positive control, since PHT is bioactivated by

PHS to an embryopathic free radical intermediate (Wells et al., 2009b). ETYA co-treatment (40

μM) protected against PHT embryopathies, completely blocking its effects on ANC (p < 0.05), somite development (p < 0.01), crown-rump length (p < 0.001) and head length (p < 0.01), with a non-significant trend for increased turning (Fig. 3.8).

223

80 150 Anterior Neuropore Closure Somite Development 60 100 (14) (10) † † ††† ††† ‡‡ 40 ‡‡‡ ‡‡‡ (15) (10) (16) (13) (7) (6) 50 20 (n, mean + SD) mean (n,

PERCENTAGE (%) PERCENTAGE 0 0 150 5 Turning Crown-Rump Length 4 100 (16) (11) + SD) 3 (15) (11) (7) (6) † † 2 50 (16) (14) 1 (mm, mean PERCENTAGE (%) PERCENTAGE 0 0

6 Yolk Sac Diameter 300 Heart Rate ††† †† 4 200 ‡ †† (16) (11) (16) (14) (11) (16) (14) 2 100 (16) (mm, mean + SD) 0 0 VH ETYA VH ETYA + SD) mean (beats/min, VH METHANOL 3 Head Length 2 ††† †† (16) (11) ‡‡‡ ‡‡ (14) 1 (16) mm, mean SD + 0 VH ETYAVH ETYA VH METHANOL

224

Figure 3.7: Effect of the prostaglandin H synthase (PHS) inhibitor eicosatetraynoic acid (ETYA) on MeOH embryopathies in embryo culture.

On GD 8.5, CD-1 embryos were removed and incubated for 24 hr with MeOH (4 mg/) or its saline vehicle (VH), with or without co-treatment with ETYA (40 μM). The number of embryos is given in parentheses. Embryos with 6 somite pairs were used for culture. Embryos were taken from 7-14 different litters. Crown-rump length and somite development were scored only if the embryo had turned. Continuous data (somite development, crown-rump length, yolk sac diameter, heart rate and head length) were analyzed by one-way ANOVA with a post-hoc

Bonferroni test. Binomial data (anterior neuropore closure, turning, lethality) were analyzed by chi-square test. Daggers indicate a difference from vehicle-treated embryos († p < 0.05, †† p <

0.01, ††† p < 0.001). Double daggers indicate a difference from embryos exposed to ETYA alone (‡ p < 0.05, ‡‡ p < 0.01, ‡‡‡ p < 0.001).

225

200 40 Anterior Neuropore Closure Somite Development 150 30 (10) (12) ††† **(8) (12) (10) (8)* 100 20 (6) † (10) 10 50 SD) + mean (n, PERCENTAGE (%) 0 0

150 4 Turning Crown-Rump Length (8) (12) (10) 3 (10) ***(8) 100 † (12) † (6) (10) 2 50 1 (mm, mean + SD) + mean (mm, PERCENTAGE (%) PERCENTAGE 0 0

5 200 Yolk Sac diameter Heart Rate 4 150 3 (12) ††† **(8) (12) (8) (10) (10) (10) (10) 100 2 1 50 (mm, mean + SD) + mean (mm, 0 0 VH ETYA VH ETYA SD) + mean (beats/min, VH PHENYTOIN 3 Head Length

(12) (10) ††† **(8) 1 (10) (mm, mean + SD) + mean (mm, 0 VH ETYA VH ETYA VH PHENYTOIN

226

Figure 3.8: Effect of the PHS inhibitor ETYA on phenytoin embryopathies in embryo culture.

On GD 8.5, CD-1 embryos were removed and incubated for 24 hr with phenytoin (80 μM) or its saline vehicle (VH), with or without co-treatment with ETYA (40 μM). Phenytoin is bioactivated by PHS to a ROS-initiating embryopathic free radical intermediate, and accordingly serves as a positive control for the study evaluating the effect of ETYA pretreatment on MeOH embryopathies. The number of embryos is given in parentheses. Embryos with 6 somite pairs were used for culture. Embryos were taken from 7-14 different litters. Crown-rump length and somite development were scored only if the embryo had turned. Continuous data (somite development, crown-rump length, yolk sac diameter, heart rate and head length) were analyzed by one-way ANOVA with a post-hoc Bonferroni test. Binomial data (anterior neuropore closure, turning, lethality) were analyzed by chi-square test. Daggers indicate a difference from vehicle- treated embryos († p < 0.05, ††† p < 0.001). Asterisks indicate a difference from phenytoin- exposed embryos (* p < 0.05, ** p < 0.01, *** p < 0.001)

227

3.5 Discussion

In utero exposure to MeOH via maternal administration upregulated embryonic p22phox mRNA and protein expression, which was associated with increased oxidative stress as measured by oxidatively damaged embryonic protein. In embryo culture, MeOH caused a concentration-dependent increase in embryopathies similar to those observed for EtOH, a structurally similar alcohol and ROS-initiating teratogen (Hunter et al., 1994; Kotch et al., 1995;

Xu et al., 2005). Maternal pretreatment with the NOX inhibitor DPI decreased MeOH-initiated embryonic p22phox mRNA and protein expression, and the level of oxidatively damaged embryonic protein, and pretreatment with the free radical spin trap PBN decreased embryonic p22phox mRNA expression and the level of oxidatively damaged protein. This in vivo evidence for MeOH induction of embryonic NOX and ROS-mediated oxidative damage to embryonic protein, together with the protection against MeOH embryopathies in culture provided by PBN and DPI pretreatment, suggest that the developmental toxicity of MeOH is due at least in part to embryonic NOX induction and NOX-catalyzed ROS formation. This putative mechanism is similar to that postulated for EtOH embryopathies (Dong et al., 2010). On the other hand, the failure of pretreatment with the PHS inhibitor ETYA to block MeOH embryopathies, despite protecting against the embryopathic effects of the ROS-initiating teratogen phenytoin, a known substrate for PHS-catalyzed bioactivation (Wells et al., 2009b), suggests that neither MeOH nor its metabolites are bioactivated by embryonic PHSs to pathogenic free radical intermediates.

In vivo evidence for the metabolism of MeOH to a free radical intermediate, and the formation of ROS and oxidative stress, includes: (1) Direct detection of MeOH-initiated ROS formation in the liver of MeOH-intoxicated adult rats after MeOH intoxication, and an increase in both lipid and protein oxidation (Skrzydlewska et al., 2000); (2) increased lipid peroxidation

228 measured as thiobarbituric acid-reactive substances (TBARS) in the livers of adult rats exposed to 6 g/kg MeOH, as well as a variably altered antioxidant defense profile in the brain and liver

(Skrzydlewska and Farbiszewski, 1998); (3) direct detection of a MeOH-derived α-(4-pyridyl 1- oxide)-N-tert-butylnitrone (POBN) adduct in the bile and urine of MeOH intoxicated rats 2 hr after acute methanol intoxication (Kadiiska and Mason, 2000); (4) Co-administration of N- acetylcysteine or the antioxidant U-83836E decreased MeOH-initiated lipid peroxidation in erythrocytes of MeOH-intoxicated rats (Dobrzynska et al., 1999); (5) MeOH-exposure increased formation of lipid peroxidation product malondialdehyde (MDA) and increased antioxidative enzyme activities in the lymphoid organs of adult rats (Parthasarathy et al., 2006b).

Similarly, developmental studies in rat and mouse embryo culture have provided evidence for ROS involvement in the mechanism of MeOH teratogenesis in rodents, including enhancement of MeOH embryopathies by depletion of glutathione in rat embryo culture (Harris et al., 2004), and respectively increased and reduced embryopathies in genetically altered mice with deficient or enhanced activities of embryonic catalase (Miller and Wells, 2011).

However, in contrast to embryo culture studies, in vivo studies from our laboratory have found no evidence for ROS involvement in the mechanism of MeOH teratogenesis. These include an in vivo study in mice genetically modified to express either reduced or enhanced levels of embryonic catalase, in which enhanced activity in C57BL/6J mice did not alter MeOH teratogenicity (reduced activity could not be evaluated as the parent C3H strain was resistant)

(Siu et al., 2013b). In another in vivo study of MeOH teratogenicity in CD-1 mice, PBN pretreatment was not protective, while embryonic GSH depletion by maternal pretreatment with buthionine sulfoximine, an inhibitor of GSH synthesis, expanded the spectrum of fetal malformations and the number of malformations per fetus, suggesting that formaldehyde-

229 macromolecular adducts rather than ROS were initiating teratogenesis, with GSH protecting via its role as a cofactor for formaldehyde dehydrogenase rather than as an antioxidant (Siu et al.,

2013a). Other malformations were not affected, suggested alternative ROS-independent mechanisms for some target tissues. Similarly in adult male mice treated acutely or chronically with MeOH, oxidatively damaged DNA and lipid peroxidation were not enhanced in most tissues, even in DNA repair-deficient oxoguanine glycosylase 1 (ogg1) knockout mice

(McCallum et al., 2011a; McCallum et al., 2011b). The basis for the discrepancy in ROS involvement observed herein and elsewhere in rodent embryo culture and some adult rodent studies, but not in other in vivo studies, is unknown, but may include the influence of diffusible modulatory factors from maternal or placental sources with in vivo studies. In the in vivo component of the studies herein, maternal administration enhanced embryonic levels of oxidatively damaged protein, which was blocked by pretreatment with DPI and PBN, indicating that MeOH can enhance embryonic NOX-catalyzed ROS formation in vivo. The embryo culture-in vivo discrepancy among some MeOH studies has not been observed with other ROS- initiating teratogens like phenytoin, benzo[a]pyrene, thalidomide and methamphetamine (Wells et al., 2009b). For these and other reasons, it is unclear which model (embryo culture vs. in vivo), species (rabbits are resistant) and rodent strain (at least one mouse and rat strain are resistant) would best predict the human developmental risk from MeOH exposure (Wells et al., 2013).

Herein, MeOH exposure increased embryonic protein oxidation from 3 to 12 hr, occurring maximally by 2-fold at 6 hr, demonstrating that MeOH is capable of initiating oxidative stress in vivo in the developing embryo. The time course of MeOH-initiated protein oxidation observed from 3 hr to 12 hr is similar to that in previous studies occurring from 6 hr to

7 days following chronic MeOH administration (Skrzydlewska et al., 2000; Skrzydlewska and

Farbiszewski, 1998). MeOH exposure similarly upregulated embryonic p22phox mRNA and

230 protein from 3 to 12 hr. This time course is remarkably similar to that for protein oxidation, consistent with our hypothesis that MeOH-initiated p22phox mRNA upregulation results in increased translation of NOX enzyme that produces increased ROS, resulting in enhanced oxidative stress and protein oxidation, which may contribute to the observed structural embryopathies initiated by MeOH exposure.

Diphenyleneiodonium (DPI) is a chemical inhibitor that inhibits NOX enzymes by abstracting the NADPH-donated electron in the FAD domain of the gp91phox subunit, converting DPI to a phenyl radical in the process. The DPI-derived phenyl radical then covalently binds to the FAD domain of gp91phox, rendering it inactive, and unable to transfer the electron for ROS production (O'Donnell et al., 1993). As all NOX isoforms contain a FAD domain, DPI is a broad spectrum NOX inhibitor. DPI both inhibits NOX activity and blocks the upregulation of DUOX1 and p22phox mRNA in embryos exposed to EtOH in utero (Dong et al.,

2010). These actions of DPI are consistent with its inhibition of MeOH-initiated p22phox mRNA expression herein. This consistency suggests that both EtOH and MeOH upregulate similar NOX components, and perhaps share a similar source of ROS production. Furthermore, DPI inhibited

EtOH-initiated embryonic DNA oxidation, as well as cellular apoptosis (Dong et al., 2010), demonstrating that DPI is capable of inhibiting oxidative stress, and that embryonic NOXs may be a candidate source of ROS. These effects are consistent with the protection by DPI against

MeOH-initiated protein oxidation observed herein. However, since p22phox is necessary for the function of NOX isoforms 1-4 (Brown and Griendling, 2009), it remains unclear which embryonic isoform(s) is or are responsible for enhanced ROS generation and resulting embryopathies. Although primarily known for its efficacy in trapping of free radicals, phenylbutylnitrone (PBN) pretreatment also decreased the MeOH-initiated upregulation of p22phox mRNA, but not protein. A previous study reported PBN inhibition of p67phox protein

231 levels (Chang et al., 2009), although the mechanism underlying this inhibition remains unclear.

By inhibiting NOX, PBN may be inhibiting the formation of ROS, which along with its efficacy as a free radical spin trapping agent (see below) would be consistent with a hypothesis involving oxidative stress, given its protection against MeOH embryopathies in culture. PBN also inhibited protein oxidation at the same time points for which NOX mRNA was inhibited, consistent with embryonic NOX-initiated ROS-formation and oxidative stress. PBN also inhibits mRNA expression for tumor necrosis factor alpha (TNF-α) (Chang et al., 2009; Lin et al., 2006; Sang et al., 1999), which stimulates NOX activity (Li et al., 2005a), suggesting an upstream mechanism by which PBN may decrease NOX expression.

PBN is a free radical spin trapping agent that traps and stabilizes superoxide and hydroxyl radicals, characterized by electron spin resonance spectrometry (ESR) (Finkelstein et al., 1980). PBN has been used to identify free radical intermediates for a number of teratogenic

ROS-initiating xenobiotics such as phenytoin and structurally related antiepileptic drugs (Liu and

Wells, 1994; Parman et al., 1998; Wells et al., 1989), MeOH (Kadiiska and Mason, 2000) and cocaine (Zimmerman et al., 1994). PBN also reduces both embryonic DNA oxidation and developmental toxicity caused by a number of ROS-initiating teratogens (Wells et al., 2009b).

Herein, the protection afforded by PBN against MeOH embryopathies support a role for ROS in the mechanism of toxicity.

In conclusion, MeOH upregulated embryonic NOX mRNA and protein expression in vivo, and caused concentration-dependent embryopathies in mouse whole embryo culture, the latter of which has been shown previously for rats. The MeOH-initiated upregulation of NOX, together with the protection by PBN and DPI against MeOH-initiated oxidatively damaged protein and most of the embryopathic effects of MeOH, suggest that NOX-catalyzed ROS

232 formation is involved in the mechanism of MeOH developmental toxicity. In contrast to embryo culture, although there is corroborating evidence of ROS involvement in MeOH toxicity in some in vivo studies, other in vivo studies including several from our laboratory, found little or no evidence of a role for ROS in MeOH teratogenicity in mice, or MeOH toxicity in adult males from several species. The reasons for this discrepancy are largely unknown, although in vivo toxicity may involve formaldehyde adducts to embryonic cellular macromolecules. The observations herein in vivo and in embryo culture show that MeOH embryopathies are not dependent upon maternal metabolism, and suggest that ROS-mediated embryopathies caused by

MeOH will likely be determined by the balance among embryonic NOX-catalyzed pathways for

MeOH-initiated ROS formation and embryonic pathways for detoxification.

233

Chapter 4 Enhanced embryonic catalase protects against ethanol-initiated embryopathies in acatalasemic and transgenic human catalase-expressing mice in embryo culture

Running title: Embryonic catalase & ethanol embryopathies

Lutfiya Miller† and Peter G. Wells*†

* Division of Biomolecular Sciences, Faculty of Pharmacy, and

†Department of Pharmacology and Toxicology, Faculty of Medicine

University of Toronto

Toronto, Ontario, Canada

a. Preliminary reports of this research were presented at the 2010 annual meeting of the Society of Toxicology [The Toxicologist 114(1): 184 (Abstract No. 863)] and at the 2012 annual meeting of the Canadian Society of Pharmacology and Therapeutics. [Proceedings of the Annual Meeting of the Canadian Society of Pharmacology and Therapeutics 19(2): e262 (Abstract No. 6) These studies were supported by a grant from the Methanol Foundation and Canadian Institutes of Health Research. b. Individual contributions: Lutfiya Miller- embryo culture; Peter G. Wells- supervisor.

234

4.1 Abstract

Reactive oxygen species (ROS) have been implicated in the mechanism of ethanol

(EtOH) teratogenicity, but the protective role of the embryonic antioxidative enzyme catalase is unclear, as embryonic activity is less than 5% of maternal levels. We addressed this question in a whole embryo culture model. C57BL/6 mouse embryos expressing human catalase (hCat) or their wild-type (C57BL/6 WT) controls, and C3Ga.Cg-Catb/J catalase-deficient, acatalasemic

(aCat) mouse embryos or their wild-type C3HeB/FeJ (C3H WT) controls, were explanted on gestational day (GD) 9 (plug = GD 1), exposed for 24 hr to 2 or 4 mg/mL EtOH or vehicle, and evaluated for functional and morphological changes. hCat and C57BL/6 WT vehicle-exposed embryos developed normally, while EtOH was embryopathic in C57BL/6 WT embryos, evidenced by decreases in anterior neuropore closure, somites developed, turning and head length, whereas hCat embryos were protected (p < 0.001). Maternal pretreatment of C57BL/6

WT dams with 50 kU/kg PEG-catalase (PEG-cat) 8 hr prior to embryo culture, which increases embryonic catalase activity, blocked all EtOH embryopathies (p < 0.001). Vehicle-exposed aCat mouse embryos had lower yolk sac diameters compared to WT controls, suggesting endogenous

ROS are embryopathic. EtOH was more embryopathic in aCat embryos than WT controls, evidenced by reduced head length and somite development (p < 0.01), and trends for reduced anterior neuropore closure, turning and crown-rump length. Maternal pretreatment of aCat dams with PEG-Cat blocked all EtOH embryopathies (p < 0.05). These data suggest that embryonic catalase is a determinant of risk for EtOH teratogenicity.

Key Words: Ethanol; reactive oxygen species; C57BL/6 mice; aCat mice; PEG-Catalase

235

4.2 Introduction

Ethanol (EtOH) is a well-established human teratogen, with in utero exposure resulting in a spectrum of anomalies collectively termed the Fetal Alcohol Spectrum Disorder (FASD)

(Jones, 2011; Jones and Smith, 1973). The precise mechanisms underlying EtOH developmental toxicity remain unclear, although reactive oxygen species (ROS) have been implicated.

In vivo, EtOH has been shown to be a developmental toxicant in rats (Lee et al., 2005;

Wentzel and Eriksson, 2008) and mice (Dong et al., 2010; Miller et al., 2013b; Miller et al.,

2013c; Wentzel and Eriksson, 2006), resulting in a spectrum of structural birth defects similar to those observed in humans. EtOH can also cause embryopathies in rats and mice in whole embryo culture (Brown et al., 1979; Hunter et al., 1994; Lee et al., 2005; Priscott, 1982; Snyder et al.,

1992; Wynter et al., 1983; Xu et al., 2005), a model that removes the confounding effects of maternal factors, allowing a more definitive examination of embryonic mechanisms of toxicity

(New, 1978).

ROS are important for the normal development of the embryo, being involved in signal transduction pathways linked to cellular proliferation, migration and differentiation (Dennery,

2007; Hansen and Harris, 2013). ROS are maintained in a balance with antioxidants in order to control the amount of ROS present at any given time, as embryonic development requires exquisite spatiotemporal control (Moore and Persaud, 2007). Excessive ROS production or deficient removal of ROS can result in oxidative stress which can alter signal transduction or oxidatively damaged cellular macromolecules including proteins, lipids and DNA (Hansen and

Harris, 2013; Wells et al., 2009b).

236

Oxidative stress and the protective role of embryonic antioxidative enzymes like catalase have also been implicated in the mechanism of teratogenesis for ROS-initiating teratogens including thalidomide, phenytoin and structurally related antiepileptic drugs (AEDs), methamphetamine and benzo[a]pyrene (Abramov and Wells, 2011a; Abramov and Wells, 2011b; Lee et al., 2011;

Parman et al., 1999; Winn and Wells, 1997; Winn and Wells, 1999; Winn and Wells, 1995;

Wong et al., 2008).

Several lines of evidence support the involvement of oxidative stress in the mechanism of

EtOH developmental toxicity. Superoxide dismutase (SOD) knockout mice exposed to EtOH in utero demonstrated enhanced levels of fetal hepatic isoprostane formation compared to their wild-type littermates (Wentzel and Eriksson, 2006); pretreatment with the free radical spin trapping agent α-phenyl-N-tert-butylnitrone (PBN) protected against EtOH-initiated fetal brain

DNA oxidation and postnatal behavioural deficits in mice exposed to EtOH during the fetal period (Miller et al., 2013b); pretreatment with 3H-1,2 dithiole-3-thione (D3T), a nuclear factor erythroid 2-related factor 2 (Nrf2) inducer, protected against EtOH-initiated ROS formation with in mouse embryos exposed in utero to EtOH (Dong et al., 2008); and, treatment with diphenyleneiodonium (DPI), an inhibitor of ROS-producing NADPH oxidase (NOX) enzymes, protected against EtOH teratogenicity, DNA oxidation and apoptosis in mouse embryos exposed in utero to EtOH (Dong et al., 2010).

Embryonic catalase is an antioxidative enzyme responsible for the detoxification of ROS, specifically hydrogen peroxide (H2O2), while also capable of metabolizing the alcohols EtOH and methanol (MeOH) to their corresponding aldehydes (Thurman and Handler, 1989). If not detoxified by catalase, oxidative stress can occur as a result of enhanced ROS levels, resulting in altered signal transduction and oxidative damage to cellular macromolecules, which have been

237 implicated in structural and functional teratogenesis (Wells et al., 2009b). . Despite the low level of embryonic catalase expression, approximating 5% of maternal levels, this quantitatively minor enzyme has been shown to protect against phenytoin-initiated DNA oxidation and embryopathies in whole embryo culture and teratogenesis in vivo, as well as protecting against MeOH embryopathies in embryo culture (Abramov and Wells, 2011a; Abramov and Wells, 2011b;

Miller and Wells, 2011; Winn and Wells, 1997; Winn and Wells, 1999; Winn and Wells,

1995).

Similarly, an embryoprotective effect has been observed with exogenous protein therapy in several models, in which pretreatment using exogenous forms of catalase or SOD reduced the developmental toxicity of ROS-initiating teratogens and conditions such as phenytoin, ethanol and diabetes in vivo and/or in embryo culture in mice, rats or xenopus (Cederberg and Eriksson,

1997; Chen et al., 2004a; Peng et al., 2004; Winn and Wells, 1995). Although administration of exogenous forms of catalase is protective, the protective efficacy of the low embryonic level of the endogenous enzyme against EtOH embryopathies is unclear.

Herein, we sought to determine the role of endogenous embryonic catalase in protecting against ROS-mediated structural embryopathies caused by EtOH. We used a whole embryo culture model including both transgenic mice endogenously expressing human catalase (hCat), and mutant catalase-deficient (acatalasemic, aCat) mice. These hCat and aCat mice have been previously shown to exhibit respectively enhanced and deficient levels of embryonic catalase

(Abramov and Wells, 2011a). The specific role of catalase, as distinct from other unappreciated genetic determinants, was confirmed by the use of PEG-cat, an exogenous form of catalase protein therapy. We demonstrate herein that enhanced activity of embryonic catalase, whether endogenous or exogenous, protects against EtOH embryopathies, while endogenous catalase

238 deficiency exacerbates embryopathies. These results show that the quantitatively minor level of endogenous embryonic catalase is an important determinant of risk for EtOH embryopathies, and corroborate other studies implicating ROS in the teratological mechanism.

239

4.3 Methods

Chemicals. Ethanol was purchased from Commercial Alcohol Inc. (Brampton, ON). Bouin’s fixative and PEG-catalase were purchased from Sigma Chemical (St. Louis, MO). Taq polymerase, 10X Hot start buffer, MgCl2, deoxyribonucleotides (dNTPs), 6X loading dye, 10X

Buffer O and restriction endonuclease NdeI were purchased from Fermentas Life Sciences

(Burlington, ON).

Animals and diet. All animal protocols were approved by the institutional animal care committee in conformance with the guidelines established by the Canadian Council on Animal

Care. Transgenic C57BL/6J mice expressing human catalase (Tg(CAT)23Jv) (hCat) were generously provided by Dr. Arlan Richardson (University of Texas Health Science Center at San

Antonio, San Antonio, TX), and were generated as described previously (Chen et al., 2003).

C57BL/6J wild-type (C57BL/6 WT) male and female mice (Charles River Canada, St. Constant,

QC) were mated to generate a proximate wild-type colony to the hCat strain. Breeding pairs of

C3Ga.Cg-Catb/J acatalasemic (aCat) and catalase-normal wild-type C3HeB/FeJ (+/+) wild-type mice (C3H WT) were purchased from The Jackson Laboratory (Bar Harbor, ME). C3H WT male and female mice were mated in order to generate a proximate wild-type colony to the aCat strain, which were mated homozygously. All four mouse lines were bred and maintained separately. Mice were housed in vented plastic cages from Allentown, Inc. (Allentown, NJ) with ground corncob bedding (Bed-O’Cobs Laboratory Animal Bedding, The Andersons Industrial

Products Group, Maumee, OH). Mouse cages were maintained in a room with controlled light

(14 hr light-10 hr dark cycle) and climate (20ºC, 50% humidity), and provided with rodent chow

240

(Harlan Labs: 2018, Harlan Teklad, Montreal, QC) and tap water ad libitum. Mice were acclimatized for 1 week prior to use. Virgin female mice were mated with males overnight from

5:00 PM to 9:00 AM by housing one male in a cage containing 1-3 females. The presence of a vaginal plug the next morning was designated as gestational day (GD) 1 and plugged females were separated, weighed and housed in groups of up to four per cage until use.

Embryo Culture. On GD 9, pregnant hCat dams or their proximate C57BL/6J WT controls, and aCat dams or their proximate C3H WT controls, were euthanized by cervical dislocation, and embryos were cultured as previously described (Miller and Wells, 2011). Embryos dissected at the 7-9-somite stage were washed in pregassed (5% CO2 in air) holding medium (HM) (17 ml male rat serum, 2.5mM HEPES, 50 units/ml of penicillin, 50 mg/ml of streptomycin, and 10X

Waymouth’s MB 752/l medium) and transferred to a 24-well plate (BD Biosciences, Franklin

Lakes, NJ) containing 2 mL embryo culture medium (ECM; HM supplemented with 15 mL fetal bovine serum) (1 embryo per well) containing either EtOH (2 or 4 mg/ml) or saline vehicle for a total of 24 hr. Each plate well was sealed air tight with optically clear sealing tape (Sarstedt,

Montreal, QC). The 24-well plate was incubated at 37ºC in a Sanyo model MCO-17A CO2 incubator (Sanyo Electric Co., Ltd, Japan) on a platform rocker (Bellco Biotechnology,

Vineland, NJ). After 24 hr, embryonic morphological and developmental parameters were observed using a dissecting microscope (Stemi SV11, Carl Zeiss, Oberkochen, Germany).

Developmental parameters included dorsal-ventral flexure (turning), anterior neuropore closure

(ANC) and somite development. Functional and morphological assessments included heart rate, yolk sac diameter (in mm), crown-rump length (in mm) and head length (in mm). Overall lethality was measured in all groups as a percentage. Somite development and crown-rump

241 length were determined only in those embryos that had turned, as the continued development of these two parameters is dependent upon turning.

Genotyping hCat and C57BL/6 WT mice. DNA was isolated from ear notches by heating the sample in 75 µL alkaline lysis reagent (25 mM NaOH, 0.2 mM disodium EDTA, pH 12) for 1 hour. Samples were then neutralized by addition of 75 µL of neutralizing buffer (40 mM Tris-

HCl, pH 5), and genotyped using a PCR-based assay as previously described (Miller and Wells,

2011). The PCR products were separated on a gel consisting of 1.5% (w/v) agarose, 40 mM Tris,

19.4mM glacial acetic acid and 2.5 mM EDTA.

Genotyping aCat and C3H WT mice. DNA from ear notches was extracted as described above and genotyped by a PCR-based assay as previously described (Miller and Wells, 2011). The digestion products were separated on a gel consisting of 1.5% (w/v) agarose, 40 mM Tris, 19.4 mM glacial acetic acid and 2.5 mM EDTA.

Statistical analysis. Statistical analysis was performed using GraphPad Prism, Version 5

(GraphPad Software, Inc., San Diego, CA). Continuous data including somite development, crown-rump length, yolk sac diameter, head length and heart rate were analyzed using a 1-way analysis of variance (ANOVA) with a post-hoc Bonferroni test. Binomial data including anterior neuropore closure and turning were analyzed using a chi-square test, or a chi-square test for trend where a trend was observed. The minimal significance level used throughout was p < 0.05.

242

4.4 Results

Embryos expressing human catalase (hCat)

Comparison of growth of hCat and C57BL/6 WT saline-exposed embryos

There were no differences in any parameters for baseline embryonic growth and development between saline-exposed hCat and C57BL/6 WT embryos (Fig. 4.1)

EtOH embryopathies in C57BL/6 WT embryos

Exposure to 4 mg/ml EtOH for 24 h was embryopathic as evidenced by decreased ANC (p <

0.01), somite development (p < 0.01), turning (p < 0.01), and head length (p < 0.01), and increased heart rate (p < 0.01) (Fig. 4.1)

EtOH embryopathies in hCat embryos

There were no differences in any growth parameters in hCat embryos, while heart rate was increased (p < 0.05) (Fig. 4.1)

Comparison of EtOH embryopathies in hCat versus C57BL/6 embryos

Compared to EtOH-exposed C57BL/6 WT embryos, hCat embryos exhibited increased ANC (p

< 0.01), somite development (p < 0.001), turning (p < 0.01), and head length (p < .001) (Fig.

4.1).

243

Anterior neuropore closure Somites developed 150 40 WT hCat WT hCat 30 (20) * *** (21) *** 100 (7) (21) (13)** (19) †† (6) (11) †† 20 (7) 50 (19) 10 (n, mean SD) +

PERCENTAGE (%) 0 0

Turning Crown-rump length 150 4 WT hCat WT hCat (20) * (21) 3 100 (7) (13)** (19) (7) (6) (21) (11) †† 2 50 (19) 1 (mm, mean + SD) + mean (mm, PERCENTAGE (%) 0 0

Yolk sac diameter Heart rate 200 3 WT hCat WT hCat 150 †† (20) (21) (13) †† †† (19) (7) (13) 2 (19) (7) 100 (20) (21)

1 50 (mm, mean + SD) (mm, mean + 0 0 (beats/min, mean + SD) + mean (beats/min, Veh EtOH PEG-Cat Veh EtOH + EtOH

Head length 2.0 WT hCat 1.5 (12) ††† **(7) (9) ***(8) (14) 1.0

0.5 (mm, mean + SD) mean(mm, + 0.0 Veh EtOH PEG-Cat Veh EtOH + EtOH

244

Figure 4.1: Protection against ethanol (EtOH) embryopathies in transgenic mice expressing human catalase (hCat).

On gestational day (GD) 9, wild-type (WT) C57BL/6 or hCat embryos containing 7 or 8 somite- pairs were explanted and incubated for 24 h with EtOH (4 mg/ml) or its saline vehicle. A separate group of pregnant C57BL/6 dams were dosed with polyethylene glycol-conjugated catalase (PEG-cat) (50 kU/kg, i.p) 8 hours prior to the initiation of culture to assess the effect of enhanced exogenous catalase on EtOH embryopathies. Crown-rump length and turning were assessed only if the embryo had turned. Continuous data were analyzed by one-way ANOVA with a post-hoc Bonferroni test, and binomial data were analyzed by a chi-square test. The number of embryos is given in parentheses. Single daggers indicate a difference from saline control of the same genotype (†† p < 0.01, ††† p < 0.001). Asterisks indicate a difference from

EtOH-exposed C57BL/6 embryos (* p < 0.05, ** p < 0.01, *** p < 0.001).

245

Effect of PEG-Cat in C57BL/6 embryos

Compared to EtOH-exposed C57BL/6 WT embryos, maternal pretreatment with PEG-Cat increased ANC (p < 0.05), somite development (p < 0.001), turning (p < 0.05), and head length

(p < 0.01) (Fig. 4.1).

Catalase-deficient embryos (acatalasemic, aCat)

Comparison of growth of aCat and C3H WT saline-exposed embryos

Saline-treated aCat embryos had a lower yolk-sac diameter than C3H WT saline-treated embryos

(p < 0.05), and there was a non-significant trend for a lower percentage of aCat embryos that turned compared to C3H WT embryos (Fig. 4.2).

Concentration-dependent EtOH embryopathies in C3H WT embryos

C3H WT embryos exposed to EtOH at 2 and 4 mg/ml exhibited concentration-dependent dysmorphogenesis as evidenced by decreased somite development (p < 0.05) at the low concentration and high concentration, and decreased anterior neuropore closure (p < 0.05), turning (p < 0.05), yolk-sac diameter (p < 0.01) and head length (p < 0.05) at the high concentration, with nonsignificant trends for decreased ANC and turning (Fig. 4.2).

246

Concentration-dependent EtOH embryopathies in aCat embryos aCat embryos exposed to EtOH at 2 and 4 mg/ml exhibited concentration-dependent dysmorphogenesis as evidenced by decreased somite development (p < 0.001) at the low concentration, and head length at the low (p < 0.01) and high (p < 0.001) EtOH concentrations.

None of the aCat embryos exposed to 4 mg/ml EtOH closed their neuropore or turned, the latter precluding evaluation of somite development and crown-rump length at this concentration (Fig.

4.2).

Comparison of EtOH embryopathies in aCat and C3H WT embryos

Compared to C3H WT EtOH-exposed embryos, aCat embryos exhibited decreases in somite development (p < 0.01) at the low concentration, and head length at the low (p < 0.01) and high

(p < 0.05) concentrations of EtOH (Fig. 4.2). aCat embryos consistently exhibited lower percentages of anterior neuropore closure and turning at both the low and high concentration of

EtOH compared to C3H WT embryos, although this observation was not significant.

Effect of PEG-cat in aCat embryos

Compared to EtOH-exposed aCat embryos, maternal pretreatment with PEG-cat increased ANC

(p < 0.05), turning (p < 0.05), and head length ( p < 0.05). Somite development and crown-rump length were restored to levels not different from aCat saline-exposed embryos in PEG-cat- pretreated embryos; however, these parameters could not be statistically compared to embryos exposed to 4 mg/ml as they failed to turn (Fig. 4.2).

247

Anterior neuropore closure Somites developed

100 40 WT aCat  WT aCat 80 (7) (16) 30 † (17) (11) ††† (5) (7) 60 (11) (10) †††** 20 (5) (2) 40 (9) (7) 20 10 (n, mean + SD) mean + (n,

PERCENTAGE (%) PERCENTAGE (4) 0 0

Turning Crown-rump length

100 4 WT aCat  WT aCat 80 (16) (7) 3 (17) (11) 60 (10) (5) (5) (7) (11) 2 (2) 40 (9) (7) 20 1 (mm, mean + SD) + mean (mm, PERCENTAGE (%) PERCENTAGE (4) 0 0

Yolk sac diameter Heart rate 5 200 aCat WT aCat WT (4) 4 150 (16) (7) (7) (11) (16) †† ‡ (17) 3 (17) (7) (16) (16) (11) (7) (4) 100 2 50 1 (mm, mean + SD) mean(mm, + 0 0 (beats/min, mean SD) + 0 2 4 0 2 4 PEG-Cat + EtOH (mg/ml) EtOH (mg/ml) EtOH (4 mg/ml)

Head length 2.0 WT aCat 1.5 a (16) (17) † (11) (7) (16) **† * 1.0 (7) †† (4) 0.5 mm, mean + SD mean + mm, 0.0 0 2 4 0 2 4 PEG-Cat + EtOH (mg/ml) EtOH (mg/ml) EtOH (4 mg/ml)

248

Figure 4.2: Exacerbation of EtOH embryopathies in catalase-deficient acatalasemic (aCat) mice.

On GD 9, C3HeB/FeJ (C3H) WT or aCat embryos containing 7 or 8 somite-pairs were explanted and incubated for 24 h with EtOH (2 or 4 mg/ml) or its saline vehicle. A separate group of pregnant aCat dams were dosed with PEG-cat (50 kU/kg, i.p) 8 hours prior to the initiation of culture to assess the effect of enhanced exogenous catalase on EtOH embryopathies. Crown- rump length and turning were assessed only if the embryo had turned. Continuous data were analyzed by one-way ANOVA with a post-hoc Bonferroni test, and binomial data were analyzed by a chi-square test. The number of embryos is given in parentheses. Single daggers indicate a difference from saline control of the same genotype (†† p < 0.01, ††† p < 0.001). Asterisks indicate a difference from EtOH-exposed C57BL/6 embryos (* p < 0.05, ** p < 0.01, *** p <

0.001).

249

Strain differences in C3H WT and C57BL/6 WT embryos

Compared to C57BL/6 saline-treated embryos, saline-treated C3H embryos had 27% lower ANC

(p < 0.05), 27%, lower turning (p < 0.05), 16% increased crown-rump length (p < 0.05), and 25% larger yolk-sac diameter (p < 0.001) (Fig. 4.3). There were no differences observed for somite development, heart rate or head length.

Embryonic DNA oxidation

Previous in vivo studies using these genetically altered animal models showed that maternal

EtOH treatment caused an increase in oxidatively damaged DNA in the WT animals for all strains, and EtOH-initiated DNA oxidation was respectively decreased and increased in hCat and aCat embryos compared to their EtOH-exposed WT controls (Miller et al., 2013c).

250

Anterior neuropore closure Somites developed 150 40

30 (20) (19) (11) 100 * (16) 20 50 10 (n, meanSD) + PERCENTAGE (%) PERCENTAGE 0 0

Turning Crown-rump length 150 4

3 100 (20) (11)* * (19) (16) 2 50 1 (n, mean SD) + PERCENTAGE (%) PERCENTAGE 0 0

Yolk sac diameter Heart Rate 5 200

4 150 ***(16) 3 (16) (20) 100 (20) 2 1 50 (mm, mean + SD) + mean (mm, 0 0 (beats/min, mean SD) + C57BL/6 C3H

Head length 4

2 (12) (16) 1 (n, meanSD) + 0 C57BL/6 C3H

251

Figure 4.3: Strain differences in EtOH embryopathies in C57BL/6 WT embryos and C3H WT embryos.

On GD 9, C3HeB/FeJ (C3H) WT or C57BL/6 WT embryos containing 7 or 8 somite-pairs were explanted and incubated for 24 h with saline. Crown-rump length and turning were assessed only if the embryo had turned. Continuous data were analyzed by one-way ANOVA with a post-hoc

Bonferroni test, and binomial data were analyzed by a chi-square test. The number of embryos is given in parentheses. Asterisks indicate a difference from EtOH-exposed C57BL/6 embryos (* p

< 0.05, *** p < 0.001).

252

4.5 Discussion

Prenatal exposure to EtOH can cause a spectrum of structural and behavioural anomalies collectively termed FASD, and while the underlying mechanisms are unclear, oxidative stress has been implicated (Brocardo et al., 2011). Our studies herein using mutant catalase-deficient mice and transgenic mice expressing human catalase demonstrate that endogenous embryonic catalase activity is an important risk factor in EtOH teratogenesis, which is consistent with published studies implicating oxidative stress in the pathogenic mechanism.

Our laboratory has previously shown that the mutant aCat mice used herein have decreased levels of catalase activity (Abramov and Wells, 2011a; Abramov and Wells, 2011b), and in in vivo studies exhibit enhanced embryonic DNA oxidation and birth defects when treated with EtOH (Miller et al., 2013c) or phenytoin (Abramov and Wells, 2011a). Conversely, hCat mice have enhanced levels of embryonic catalase activity (Abramov and Wells, 2011a;

Abramov and Wells, 2011b) and are protected against embryonic DNA oxidation and birth defects caused by both EtOH in vivo (Miller et al., 2013c) and phenytoin in vivo and in embryo culture (Abramov and Wells, 2011a; Abramov and Wells, 2011b). Endogenous embryonic catalase has similarly been shown protect against embryopathies caused by the related alcohol

MeOH in embryo culture (Miller and Wells, 2011).

Physiological levels of ROS are necessary for normal embryonic development, as they are involved in signal transduction pathways that mediate appropriate cellular growth, differentiation and proliferation (Dennery, 2007; Hansen and Harris, 2013). However, excessive

ROS can result in altered signal transduction or oxidatively damaged cellular macromolecules, either of which can result in aberrant embryonic development (Hansen and Harris, 2013; Wells

253

et al., 2009b). Catalase is an antioxidative enzyme responsible for the detoxification of H2O2, a molecule involved in signal transduction pathways and also capable of forming the highly toxic hydroxyl radical by reacting with iron via the Fenton reaction (Lloyd et al., 1997). Since previous studies have demonstrated the potential protective effect of exogenous catalase

(Cederberg and Eriksson, 1997; Chen et al., 2004a; Peng et al., 2004; Winn and Wells, 1995), we sought to determine the importance of the endogenous quantitatively minor embryonic catalase enzyme in EtOH embryopathies.

In the studies herein using mice with partial genetic reductions or enhancements in catalase activity, the protection offered EtOH embryopathies by both genetically enhanced endogenous embryonic catalase and exogenously administered polyethylene glycol-conjugated catalase (PEG-cat), and the converse exacerbation of EtOH embryopathies in acatalasemic mice, show that embryonic catalase, despite its low level, is a determinant of risk for EtOH teratogenicity, and corroborate studies implicating ROS in the pathogenic mechanism. Not all embryopathies were exacerbated in the aCat mice, however the almost complete protection observed in the hCat embryos exposed to EtOH demonstrates a comprehensive protective effect of endogenous catalase. The protection afforded by endogenous catalase is remarkable, given that it is expressed at only about 5% of maternal hepatic activity (Abramov and Wells, 2011a;

Winn and Wells, 1999).

While somite development and crown-rump length could not be assessed in the aCat embryos at the 4mg/ml concentration of EtOH, precluding a comparison to their WT embryos, it is clear that this concentration is highly embryopathic to aCat embryos as none of the aCat embryos turned. For this reason, a lower concentration of EtOH (2 mg/ml) was used in the aCat and C3H embryos, which allowed for comparison of somite development and crown-rump

254 length, both of which were significantly decreased in the aCat embryos compared to C3H WT embryos. Conversely, 4 mg/ml EtOH was embryopathic in the C57BL/6 embryos, while hCat embryos were completely protected, further demonstrating the important protective role of embryonic catalase.

aCat mouse embryos exposed only to vehicle had smaller yolk sac diameters compared to

C3H WT controls, suggesting endogenous ROS are embryopathic. The embryopathic potential of endogenous ROS has been similarly observed in mutant progeny deficient in glucose-6- phosphate dehydrogenase (G6PD), which provides NADPH necessary for ROS detoxification; the G6PD-deficient fetuses exhibit increased in utero death compared to their wild-type littermates (Nicol et al., 2000). Similarly, DNA repair-deficient knockout embryos lacking ataxia telangiectasia mutated (ATM), which directs the repair of oxidative DNA damage, exhibited more embryopathies than wild-type littermates, with an atm gene dose-dependent increase in susceptibility from wild-type to heterozygous to homozygous knockout (Bhuller and Wells,

2006).

Strain differences in baseline growth were observed in the saline-treated C57BL/6 and

C3H WT strains, with the C3H embryos exhibiting lower anterior neuropore closure and turning, but a higher yolk-sac diameter and crown-rump length. Because of these strain differences,

PEG-cat was used to ensure that the observed differences in embryopathies were due to the modifications in catalase expression, and not due to other unappreciated genetic differences in these strains. While not measured herein, our laboratory has demonstrated that PEG-cat administration in pregnant dams in vivo or directly to the culture medium in embryo culture results in sustained elevated levels of embryonic catalase activity over 24 hours (Abramov and

Wells, 2011a; Winn and Wells, 1999). Herein, consistent with a catalase-dependent mechanism,

255 when pretreated with PEG-cat, all embryopathic parameters exacerbated by EtOH in the catalase-deficient embryos were protected, confirming the pivotal role of embryonic catalase in the modulation of structural embryopathies.

PEG-cat pretreatment is an effective way to offer antioxidative protection, as evidenced by protection against: (1) phenytoin-initiated embryonic DNA oxidation and embryopathies in embryo culture (Abramov and Wells, 2011b; Winn and Wells, 1995); (2) phenytoin-initiated embryonic DNA oxidation and teratogenicity in vivo (Abramov and Wells, 2011a; Winn and

Wells, 1999); (3) EtOH-initiated embryonic DNA oxidation and teratogenicity in vivo (Miller et al., 2013c); ( 4) benzene-initiated oxidative stress, as measured by the ratio of reduced to oxidized glutathione, and teratogenesis in vivo (Wan and Winn, 2008); and, (5) valproic acid- initiated intracellular ROS production in cell culture (Defoort et al., 2006). Similarly, EUK-134, a synthetic SOD/catalase mimetic that scavenges hydrogen peroxide, has been shown to protect against EtOH-induced limb malformations in vivo (Chen et al., 2004a).

Herein, an embryo culture method was used to remove any potentially confounding maternal factors, including maternal metabolism of EtOH, to facilitate an explicit examination of embryonic mechanisms. The embryo culture model for EtOH in hCat and aCat mice reproduces the outcomes observed from in vivo studies of EtOH (Miller et al., 2013c) and several other

ROS-initiating xenobiotics and conditions (Wells et al., 2009b), but this does not appear to be true for the related alcohol MeOH. In the case of MeOH, although hCat and aCat embryos in culture were respectively resistant and more susceptible to MeOH than their WT controls, suggesting a ROS-dependent mechanism of teratogenesis, this modulation was not observed with in vivo studies (Miller and Wells, 2011; Siu et al., 2013b). The mechanisms underlying this in

256 vitro/in vivo discrepancy for MeOH in the same mouse strains is unknown, but illustrates the potential for maternal modulation.

The range of EtOH concentrations used herein were from 2-4 mg/ml (43 – 87 mM), which is comparable to concentrations used in other studies investigating the mechanism of

EtOH developmental toxicity (Brown et al., 1979; Hunter et al., 1994; Lee et al., 2005;

Priscott, 1982; Snyder et al., 1992; Wynter et al., 1983; Xu et al., 2005). This plasma concentration is easily achievable in humans after consumption of anywhere between 3 – 10 drinks (May et al., 2008).

In summary, our results demonstrate that endogenous embryonic catalase, while quantitatively minor, is an important determinant of risk for EtOH teratogenicity, and corroborate studies implicating ROS in the pathogenic mechanism.

257

Chapter 5 Embryonic catalase protects against ethanol- initiated DNA oxidation and teratogenesis in acatalasemic and transgenic human catalase-expressing mice

Running title: Embryonic catalase in ethanol teratogenesis

Lutfiya Miller*, Aaron M. Shapiro† and Peter G. Wells*†

*Department of Pharmacology and Toxicology, Faculty of Medicine; and,

†Division of Biomolecular Sciences, Faculty of Pharmacy

University of Toronto

Toronto, Ontario, Canada

a. Preliminary reports of this research were presented at the 2012 annual Great Lakes Mammalian Development meeting [Proceedings of Annual Great Lakes Mammalian Development meeting, Abstract No. 19] These studies were supported by grants from the Methanol Foundation and Canadian Institutes of Health Research. b. Full report of this research has been published: Miller, L., Shapiro, A.M., Wells, P.G. (2013) Embryonic catalase protects against ethanol-initiated teratogenesis and DNA oxidation in acatalasemic and transgenic human catalase-expressing mice. Toxicological Sciences 134(2):400-11 c. Individual contributions: Lutfiya Miller- mating, genotyping, dosing and sample collection, fetal assessments; Aaron M. Shapiro- DNA oxidation measurements; Peter G. Wells- supervisor.

258

5.1 Abstract

Reactive oxygen species (ROS) are implicated in Fetal Alcohol Spectrum Disorders

(FASD) caused by alcohol (ethanol, EtOH). Although catalase detoxifies hydrogen peroxide

(H2O2), embryonic catalase activity is only about 5% of maternal levels. To determine the roles of ROS and embryonic catalase in FASD, pregnant mice with enhanced (expressing human catalase, hCat) or deficient (acatalasemic, aCat) catalase activity, or their respective wild-type

(WT) controls, were treated intraperitoneally on gestational day 9 with 4 or 6 g/kg EtOH or its saline vehicle, and embryos and fetuses were respectively evaluated for oxidatively damaged

DNA and structural anomalies. Untreated hCat and aCat dams had respectively more and less offspring than their WT controls. hCat progeny were protected from all EtOH fetal anomalies at the low dose (p < 0.01), and from reduced head diameter and resorptions at the high dose (p <

0.001). Conversely, aCat progeny were more sensitive to dose-dependent EtOH fetal anomalies

(p < 0.001), and exhibited enhanced a 50% increase in maternal lethality (p < 0.05) at the high dose. Maternal pretreatment of aCat mice with PEG-Catalase (PEG-Cat) reduced EtOH fetal anomalies (p < 0.001). EtOH-initiated embryonic DNA oxidation was reduced in hCat mice and

WT mice pretreated with PEG-Cat, and enhanced in aCat mice. Plasma concentrations of EtOH in catalase-altered mice were similar to controls, precluding a pharmacokinetic basis for altered

EtOH teratogenesis. Endogenous embryonic catalase, despite its low level, is an important embryoprotective enzyme for EtOH teratogenesis, and a likely determinant of individual risk.

Key Words: Catalase; DNA oxidation; ethanol; mutant mice; reactive oxygen species; teratogenesis

259

5.2 Introduction

Ethanol (EtOH, alcohol) consumption during pregnancy can result in a spectrum of anomalies in the developing child including structural malformations and behavioural deficits, collectively termed Fetal Alcohol Spectrum Disorders (FASD) (Jones, 2011; Jones and Smith,

1973). Reactive oxygen species (ROS) have been implicated in the embryopathic mechanism for

EtOH, and catalase is both an antioxidative enzyme that converts potentially embryotoxic hydrogen peroxide (H2O2) to oxygen and water (Kirkman and Gaetani, 2007), although it also metabolizes alcohols such as ethanol and methanol (Thurman and Handler, 1989). If not detoxified, H2O2 can potentially alter embryonic signal transduction pathways and/or oxidatively damage cellular macromolecules (DNA, proteins, lipids) which may result in a loss of their function (Halliwell and Gutteridge, 2007; Wells et al., 2009b). Although expressed in the embryo at levels only about 5% of those in adult liver (el-Hage and Singh, 1990; Winn and

Wells, 1999), embryonic catalase nevertheless appears to protect the embryo from ROS-initiated embryopathies and teratogenesis (Abramov and Wells, 2011a; Abramov and Wells, 2011b;

Miller and Wells, 2011; Wells et al., 2009b; Winn and Wells, 1995). However, the roles of

ROS and particularly embryonic catalase in the developmental toxicity of EtOH are unclear.

Evidence supporting the involvement of ROS-initiated oxidative stress in the mechanism of EtOH developmental toxicity includes elevation in fetal hepatic isoprostane formation in superoxide dismutase (SOD) knockout mice exposed to EtOH in utero (Wentzel and Eriksson,

2006); increased lipid peroxidation in fetal liver, brain, kidneys and testes (Nayanatara et al.,

2009); and a protective effect of diphenyleneiodonium (DPI), a NADPH oxidase (NOX) inhibitor, against EtOH-initiated ROS formation and oxidative DNA damage in mouse embryos

(Dong et al., 2010). Further evidence to support the involvement of oxidative stress in EtOH

260 teratogenicity is reviewed in detail elsewhere (Brocardo et al., 2011). Oxidative stress has been implicated in the teratogenic mechanism of other ROS-initiating teratogens including phenytoin and structurally related antiepileptic drugs, benzo[a]pyrene, methamphetamine and thalidomide

(Wells et al., 2009b).

Supplementation with exogenous catalase has been demonstrated to provide protection against other ROS-initiating teratogens such as phenytoin, ethanol and diabetes in vivo and/or in embryo culture in mice, rats or xenopus (Cederberg and Eriksson, 1997; Chen et al., 2004a;

Peng et al., 2004; Winn and Wells, 1995), as has exogenous SOD for benzo[a]pyrene in embryo culture (Winn and Wells, 1997). EUK-134, a synthetic SOD/catalase mimetic that scavenges

H2O2, protected against EtOH-induced limb malformations in vivo (Chen et al., 2004a).

More broadly, administration of polyethylene glycol-conjugated catalase (PEG-Cat) has been demonstrated to elevate embryonic catalase activity in embryo culture and in vivo, and protect against the effects of ROS-initiating teratogens, including oxidative damage to embryonic cellular macromolecules caused by phenytoin and benzo[a]pyrene, and phenytoin developmental toxicity (Abramov and Wells, 2011a; Abramov and Wells, 2011b; Wells et al., 2009b; Winn and Wells, 1997; Winn and Wells, 1999). However, although administration of exogenous forms of catalase can be protective, the protective efficacy of the low embryonic level of the endogenous enzyme against EtOH teratogenesis is unclear.

To determine the role of ROS and embryonic catalase in the mechanism of EtOH developmental toxicity, we examined EtOH teratogenesis in transgenic mice endogenously expressing human catalase (hCat), exhibiting enhanced catalase activity, as well as in mutant catalase-deficient (acatalasemic, aCat) mice. The involvement of catalase in EtOH teratogenesis was confirmed in aCat mice by treatment with PEG-Cat, an exogenous form of catalase protein

261 therapy. Oxidatively damaged DNA was measured in gestational day (GD) 12 embryos 6 hours after EtOH exposure to determine both the involvement of ROS and the potentially pathogenic role of 8-oxo-2’-deoxyguanosine (8-oxodGuo), a DNA lesion that has been implicated in neurodevelopmental deficits caused by the ROS-initiating drug methamphetamine (McCallum et al., 2011c; Wong et al., 2008). Lastly, to ensure that any differences in teratogenicity between catalase-altered strains and their wild-type controls were not due to differences in EtOH metabolism, plasma concentrations of EtOH were measured in male mice of all strains. We demonstrate herein for the first time that enhanced activity of embryonic catalase, whether endogenous or exogenous, protects against EtOH-initiated embryonic DNA oxidation and teratogenesis, whereas a deficiency in catalase conversely enhances DNA oxidation and teratogenesis. Plasma concentrations of EtOH in catalase-altered mice were not different from those in their respective wild-type controls, suggesting that the differences in DNA oxidation and teratogenicity were not due to altered EtOH metabolism. These results suggest that ROS, and particularly 8-oxodGuo, contribute to the teratological mechanism of EtOH, and show that endogenous embryonic catalase activity, while low, may be a determinant of risk.

262

5.3 Methods

Chemicals. Ethanol was purchased from Commercial Alcohol Inc. (Brampton, ON). Bouin’s fixative and PEG-catalase were purchased from Sigma Chemical (St. Louis, MO).

Taq polymerase, 10X Hot start buffer, MgCl2, deoxyribonucleotides (dNTPs), 6X loading dye,

10X Buffer O and restriction endonuclease NdeI were purchased from Fermentas Life Sciences

(Burlington, ON)

Animals and diet. All animal protocols used were approved by the institutional animal care committee in conformance with the guidelines established by the Canadian Council on Animal

Care. Transgenic C57BL/6J mice expressing human catalase (Tg(CAT)23Jv) (hCat) were generously provided by Dr. Arlan Richardson (University of Texas Health Science Center at San

Antonio, San Antonio, TX), and were generated as described previously (Chen et al., 2003).

C57BL/6J wild-type (C57BL/6 WT) male and female mice (Charles River Canada, St. Constant,

Quebec) were mated to generate a proximate wild-type colony to the hCat strain. Breeding pairs of C3Ga.Cg-Catb/J acatalasemic (aCat) and catalase-normal wild-type C3HeB/FeJ (+/+) wild- type mice (C3H WT) were purchased from The Jackson Laboratory (Bar Harbor, Maine, USA).

C3H WT male and female mice were mated in order to generate a proximate wild-type colony for the aCat strain, which were mated homozygously. All four mouse lines were bred and maintained separately. Mice were housed in plastic vented cages (Allentown, NJ) with ground corncob bedding (Bed-O’ Cobs Laboratory Animal Bedding, The Andersons Industrial Products

Group, Maumee, OH).

263

The cages were maintained in a light- and temperature-controlled room (14-hr light/10-hr dark cycle, 20°C, 50 % humidity). Food (Harlan Labs: 2018, Harlan Teklad, Montreal, QC) and tap water were provided ad libitum. Mice were acclimatized for 1 week before use. One male was housed with three females from 5:00 P.M. to 9:00 A.M. the next day. The presence of a vaginal plug was designated as gestational day (GD) 1 because the breeders were mixed at 5:00 p.m. and allowed to mate overnight, so insemination may have occurred as much as 15 hours prior to the detection of the plug the next morning. Plugged females were isolated and housed together in groups of four or less per cage. The enzyme catalase was chosen as the focus of our studies in part due to its proximate role in the detoxification of hydrogen peroxide, and hence one of the most direct approaches for a proof of principle examination of ROS dependence in EtOH teratogenesis, and the protective importance of quantitatively minor embryonic enzymes. Also, previous studies using exogenous approaches for varying catalase activity had suggested that exogenous approaches could alter teratological outcomes following exposure to ROS-initiating teratogens (Winn and Wells, 1997; Winn and Wells, 1999; Winn and Wells, 1995). In contrast, some antioxidative enzymes require pairing with complementary downstream pathways, as in the case of superoxide dismutase (SOD), which forms hydrogen peroxide that requires catalase- catalyzed detoxification, and exogenous SOD administration can be embryotoxic (Winn and

Wells, 1999). Having found an important protective role for the endogenous embryonic enzyme catalase, it would not be surprising if a similarly protective role would be found for other quantitatively minor embryonic antioxidative enzymes, although efficacies may vary with the type of ROS-initiating teratogen and its intracellular location of ROS formation.

264

HCl, pH 5), and genotyped using a PCR-based assay as previously described (Miller and Wells,

2011). The PCR products were separated on a gel consisting of 1.5% (w/v) agarose, 40 mM Tris,

19.4mM glacial acetic acid and 2.5 mM EDTA.

Genotyping aCat and C3H WT mice. DNA from ear notches was extracted as described above and genotyped by a PCR-based assay as previously described (Miller and Wells, 2011). The PCR products were subjected to enzymatic digestion with restriction endonuclease NdeI (CA↓TATG).

The reaction conditions were: 10 μl PCR reaction mixture, 18 μl nuclease-free water, 2 μl 10X

Buffer O and NdeI (10 u/μl). The reaction was incubated at 37°C overnight, and terminated by incubating at 65°C for 20 min, and then held at 4°C. The digestion products were separated on a gel consisting of 1.5% (w/v) agarose, 40 mM Tris, 19.4 mM glacial acetic acid and 2.5 mM

EDTA.

Teratology. On GD 9, aCat and hCat females and their respective C3H and C57BL/6 WT controls were treated intraperitoneally (i.p) with vehicle or EtOH (4 or 6 g/kg), and sacrificed on

GD 19 by cervical dislocation. The doses of EtOH given in this study are comparable to those used in other studies in vivo in C57BL/6 mice (Dong et al., 2010), C3H mice (Chernoff, 1977) and CD-1 mice (Blakley and Scott, 1984b) achieving blood alcohol concentrations (BAC)

265 ranging from 79-398 mg/100mL blood (13-86 mM), which are achievable in humans after consumption of 3-10 drinks (May et al., 2008).

PEG-Catalase (PEG-Cat) (50 kU/kg) dissolved in 0.9% saline was administered i.p. to a separate group of pregnant aCat and C57BL/6 dams 8 hr prior to EtOH administration. On GD

19, the number of implantations, resorptions and pups was recorded. The pups were assessed for growth parameters including fetal weight, head length, head diameter, and crown-rump length.

Pups were placed in Bouin’s fixative for 3 days for soft tissue examination, and subsequently transferred to 70% EtOH. Fetuses were examined for gender by identifying the presence or absence of testes, identifying animals as males or females respectively, and for internal soft tissue anomalies using the Wilson’s freehand razorblade-sectioning technique (Barrow and

Taylor, 1969). We chose a single-dose model as opposed to 2 divided doses to measure primary molecular and biochemical events in the initiation of teratogenesis, and minimize the confounding effects of secondary and later processes arising from divided and multiple-dose models. The single-dose model results in embryopathic outcomes similar to those reported for divided-dose models, and the underlying mechanisms are likely relevant to FASD, so this approach is commonly used in mechanistic studies of EtOH teratogenesis (Blakley and Scott,

1984b; Boehm et al., 1997; Downing et al., 2009).

DNA extraction and digestion. DNA was extracted from GD 12 embryos explanted from dams

6 hours after EtOH (4 g/kg) treatment (i.p), with or without PEG-Cat (50 kU/kg) pretreatment

(i.p) 8 hours prior to EtOH administration, using a chaotropic sodium iodide (NaI) method previously described by Ravanat et al. (2002). GD 12 embryos were used in part as they contained sufficient DNA to allow the measurement of oxidatively damaged DNA in single embryos. Briefly, samples were homogenized in 1 mL of cold lysis buffer (320 mM sucrose, 5

266

mM MgCl2, 10 mM Tris, 0.1 mM desferoxamine, 1% Triton X-100, pH 7.5) and centrifuged twice at 10,000 X g for 10 min at 4°C to isolate the nuclear pellet containing DNA. The pellet was then incubated with 200 uL enzyme reaction solution (1% w/v SDS, 5 mM EDTA-NA2,

0.15 mM desferoxamine 10 mM Tris-HCl pH 8.0) for 1 hr at 50ºC with RNase A/T1 mix (624 and 312 U/ml final activities, respectively). Proteinase K (1.8 mg/ml final concentration) was added for an additional 1 hr. The DNA pellet obtained after extraction was then washed 5 times with 70% EtOH. Samples were resuspended in 200 μl of sodium acetate buffer (20 mM, pH 4.8) and sonicated into solution. DNA purity was determined by measuring the absorbance ratio at

260/280 nm of 2 μL of sample in Na-acetate buffer (total volume 200 μL) with an acceptable purity cutoff of 1.5. Samples were digested with nuclease P1 (5 U/sample, 1 hr, 37°C) and calf intestinal alkaline phosphatase (6 U/sample, 1 hr, 37°C) and filtered through Amicon UltraTM filter units (YM-10, 10,000 MW cutoff; Millipore, Billerica, MA, USA) to remove DNA digestion enzymes and large particulates. Quantification of 8-oxo-7,8-dihydro-2’- deoxyguanosine (8-oxodGuo) was standardized to endogenous deoxyguanosine (dGuo), which were analyzed using high-performance liquid chromatography (HPLC) coupled respectively to tandem mass spectrometric (MS/MS) and ultraviolet (UV) detectors.

Quantification of oxidatively damaged DNA using HPLC-MS/MS. To determine the levels of

8-oxodGuo and dGuo in the same sample without dilution, dGuo was first analyzed using a UV detector set to wavelength λ=280 nm on a Perkin Elmer Series 200 HPLC system (Woodbridge,

ON) and a mobile phase consisting of phosphate buffer pH 5.2 and methanol (95:5) at an isocratic rate of 0.8 mL/min for 50 min. The same sample was then reanalyzed using an Agilent

1100 series HPLC (Mississauga, ON) with an AB Sciex API4000 QTRAP triple quadrapole

267 mass spectrometer (Concord, ON) on a water:methanol gradient with a flow rate of 0.6 mL/min for 9.5 min under the control of AB Sciex Analyst® Software. The multiple reaction monitoring

(MRM) transition used to quantify 8-oxodGuo was m/z 284.0168.0 in the positive ion selection mode to capture the parent compound (Q1), 8-oxodGuo, and the product ion (Q3), 8- oxoguanine. Data from both runs were combined and expressed as fmol 8-oxodGuo per μg dGuo.

Ethanol metabolism. To confirm whether genetic modulation of catalase altered EtOH pharmacokinetics in any of the mouse strains, aCat and hCat males and their respective C3H and

C57BL/6 WT controls were treated i.p with EtOH (2.8 g/kg, 25% v/v solution in 0.9% saline) using a 26 gauge (G) 3/8 needle or saline control. Males were chosen to avoid potentially confounding effects of hormonal modulation. Prior to blood sampling, mice were anaesthetized by placing a 15 mL conical tube containing isoﬂurane-soaked gauze over the snout of the animal until it was unresponsive. Blood samples were collected in heparinised vacutainers (lithium heparin 68 USP units per tube, Becton, Dickinson and Company, Oakville, ON) by puncturing the portal vein. Blood samples were collected at 0.5, 1, 2, 4 and 8 hr post injection. Plasma was isolated from the samples by centrifugation at 1000 X g for 15 min at 4 °C. Plasma samples were frozen at −20 °C until time of analysis. At each time point, plasma samples were collected from

3 to 6 mice. Plasma samples from saline controls and EtOH-exposed mice were analyzed for

EtOH concentrations by headspace gas chromatography (GC) based on previously published methods (Porter and Moyer, 1994).

Data analysis. Statistical significance between treatment groups was determined using GraphPad

Prism®, Version 5 (GraphPad Software, Inc., San Diego, CA). Continuous data (fetal weight,

268 head length, head diameter and crown-rump length) were analyzed by a Kruskal-Wallis test with a post-hoc Dunn’s test. Binomial data (resorptions, total anomalies, malformations by strain and gender) were analyzed by a chi-square test, or a chi-square test for trend where a dose-dependent relationship was observed. Metabolism data were analyzed by two-way ANOVA. The level of significance used throughout was p < 0.05. The fetus rather than the litter is used as the unit of analysis because the studies herein involve mechanisms for a drug that produces reactive intermediates, which are too reactive to travel from the mother to the fetus. Susceptibility is primarily determined at the level of the fetus, based upon the balance of fetal pathways for ROS formation vs. detoxification and DNA repair. This fetal dependence has been demonstrated in heterozygous animals bred to produce +/+, +/- and -/- littermates, wherein their susceptibility to

ROS varies in a fetal gene dose-dependent fashion (Wells et al., 2009b). This level of discrimination is lost when the litter rather than the fetus is used as the unit of analysis.

269

5.4 Results

Spontaneous embryopathies and maternal outcomes in untreated mice aCat mice

The untreated C3H progeny developed normally, as parameters measured herein were comparable to other control strains used in our laboratory (Abramov and Wells, 2011a; Nicol et al., 2000). Compared to untreated C3H wild-type (WT) controls, untreated aCat mice had fewer pups per litter (p < 0.01) (Fig. 5.1A), and a 3-fold greater incidence of resorptions (p < 0.01)

(Fig. 5.2), but no difference in the incidence of spontaneous malformations (Figs. 5.2, S5.6a).

Untreated C3H and related aCat mice both exhibited a 3% incidence of spontaneous umbilical defects, which has not been previously reported. There was an equal distribution of males and females among the litters (Fig. 5.1B), and no maternal lethality (Fig. 5.1C).

hCat mice

The untreated C57BL/6 progeny developed normally, as parameters measured herein were comparable to other control strains used in our laboratory (Abramov and Wells, 2011a).

Compared to untreated C57BL/6 WT controls, untreated hCat mice had larger and more pups per litter at birth (p < 0.001) (Fig. 5.1A), but no difference in baseline resorptions (Fig. 5.2). Both the hCat and the C57BL/6 WT mice exhibited a nonsignificant spontaneous 2% incidence of ocular malformations (Figs.5. 2, S5.7a), which is consistent with, albeit lower than, previous reports of 4.4% (Chase, 1942) to 10% (Kalter, 1968). Ocular malformations included

270

A. PUPS PER LITTER C. MATERNAL LETHALITY

NUMBER OF PUPS NUMBER OF PUPS ACATALASEMIC (aCat) MICE PER LITTER PER LITTER 100 (x, y) = dams, pups (x, y) = dams, pups 100 C3H WT aCat (32, 201) (13, 100) (24, 205) 80 (31, 148) † 60 (13) 50 40 (8) n, mean + SD meann, +

percentage (%) percentage 20 (7) (7) (11) (9)

MATERNAL LETHALITY 0 0 VH 4 6 VH 4 6 C3H WT aCat C57BL/6 hCat GENOTYPE GENOTYPE EtOH EtOH B. GENDER DISTRIBUTION HUMAN CATALASE (hCat)- EXPRESSING MICE GENDER DISTRIBUTION 100 (x, y) = dams, total pups Female (n) = number of pups WT hCat Male 80 100 60 (20,124) (38, 208) (28, 232) (21, 176) (81) 40 (60) (64) (99) (109) (116)(116) 50 (95) (10)

percentage (%) percentage 20

PERCENTAGE (%) PERCENTAGE (9) (7) (6) (10) (5)

MATERNAL LETHALITY 0 0 VH 4 6 VH 4 6 C3H WT aCat C57BL/6 WT hCat EtOH GENOTYPE EtOH

271

The number of pups per litter (Panel A) and gender distribution of offspring (Panel B) was assessed in untreated pregnant aCat and hCat dams or their respective C3H and C57BL/6 WT controls. The number of litters and pups is shown in parentheses. (Panel C) On gestational day

(GD) 9, pregnant aCat and hCat dams or their respective C3H and C57BL/6 WT controls were treated intraperitoneally (i.p.) with either ethanol (EtOH) (4 or 6 g/kg) or its saline vehicle (VH).

The number of dams is shown in parentheses. Continous data were analyzed by one-way

ANOVA with a post-hoc Bonferroni test (pups per litter). Binomial data (gender distribution and maternal lethality) were analyzed by chi-square test. Asterisks indicate a difference from the respective WT group (** = p < 0.01, *** = p < 0.001). Daggers indicate a difference from VH- exposed dams of the same genotype († = p < 0.05).

272

ACATALASEMIC (aCat) mice HUMAN CATALASE (hCat)-EXPRESSING MICE

13 1.6 13 FETAL WEIGHT HEAD LENGTH FETAL WEIGHT HEAD LENGTH 1.2 12 1.4 * *** 12 * *** **† *** *** 1.2 ‡‡‡ ††† 11 11 1.0 ††† 1.0 ††† 10 ††† 10 ††† g,mean + SD WT g, mean + SD C57BL/6 mm, mean + SD 0.8 mm, mean + SD 0.8 aCat hCat ††† ††† 9 † 9 ††† ‡‡‡ 0.6

9 3.0 9 3.0 HEAD DIAMETER CROWN-RUMP LENGTH HEAD DIAMETER CROWN-RUMP LENGTH ††† 8 *** *** 8 2.5 ‡‡‡ ‡‡‡ 2.5 ** ††† ††† †††

7 ††† 7 2.0 2.0 † mm, mean + mean SDmm, mm, mean + mean SDmm, †† + mean SDmm, + mean SDmm, ††† ††† 6 ††† ‡‡‡ ‡‡‡ ‡‡ 6 1.5 1.5

100 100 RESORPTIONS RESORPTIONS TOTAL MALFORMATIONS TOTAL MALFORMATIONS 100 ***††† ‡‡‡ 80 80 100 † ‡‡‡ ††† 60 60 *** * ††† ††† 50 50 40 40 ** ***††† ‡‡‡ ‡ † ***††† (mean +SD) (mean (mean +(mean SD) (mean + SD)(mean † +(mean SD) ††† PERCENTAGE PERCENTAGE PERCENTAGE PERCENTAGE 20 20 ‡‡ ‡‡‡ †† 0 0 0 0 VH 4 6 VH 4 6 VH 4 6 VH 4 6

ETHANOL (g/kg) ETHANOL (g/kg) ETHANOL (g/kg) ETHANOL (g/kg)

273

Figure 5.2: Enhanced in vivo EtOH developmental toxicity in acatalasemic (aCat) mice, and converse protection in transgenic mice expressing human catalase (hCat).

On GD 9, pregnant aCat and hCat dams or their respective C3H and C57BL/6 WT controls were treated i.p. with either EtOH (4 or 6 g/kg) or its saline vehicle (VH), and sacrificed on GD 19.

Total malformations assessed include cleft palate, ventricular dilation, exencephaly, hepatic and limb defects, microcephaly, ocular and otic defects, renal defects, spina bifida, tail and umbilical defects. Percentage was calculated by dividing the total number of fetuses affected with >1 anomaly by the number of total live fetuses. Continuous data (fetal weight, head length, head diameter, crown-rump length) were analyzed by Kruskal-Wallis test with a post-hoc Dunn’s test.

Binomial data (resorptions and total malformations) were analyzed by chi-square test for trend.

Number of litters was 2-11, with an average of 7 litters per group overall, and 8 litters per group for only the 4 g/kg dose and saline controls. Single daggers indicate a difference from saline controls of the same genotype († = p < 0.05, †† = p < 0.01, ††† = p < 0.001). Double daggers indicate a difference from the 4 g/kg EtOH-exposed group of the same genotype (‡ = p < 0.05,

‡‡ = p < 0.01, ‡‡‡ = p < 0.001). Asterisks indicate a difference from WT controls for the same treatment group (* p < 0.05, ** = p < 0.01, *** = p< 0.001).

274 microphthalmia, anophthalmia and open eye. There was an equal distribution of males and females among the litters (Fig. 5.1B), and no maternal lethality (Fig. 5.1C).

Embryopathies in ethanol-exposed mice aCat mice

EtOH-exposed aCat mice exhibited a dose-dependent reduction in fetal weight (p < 0.001) and head length (p < 0.001), and an increase in total malformations (p < 0.001), with a reduction in crown-rump length only at the high dose (p < 0.01) (Figs. 5.2, S5.6b-c). Although baseline glutathione peroxidase activity is slightly but significantly elevated in the aCat embryos, this increase was not sufficient to protect against ROS-initiated birth defects caused by phenytoin

(Abramov and Wells, 2011a). C3H WT mice were not affected at the low dose, but at the high dose exhibited reductions in head diameter (p < 0.01) and crown-rump length (p < 0.001), increased resorptions (p < 0.01) and a non-significant increase in total malformations. Compared to EtOH-exposed C3H WT controls, aCat mice at the low EtOH dose had lower fetal weight (p <

0.01) and head length (p < 0.001), and increased total malformations (p < 0.01), and at the high dose exhibited lower fetal weight (p < 0.001), head length (p < 0.001) and crown-rump length (p

< 0.001) (Fig. 5.2). When the low and high EtOH dose groups were combined, aCat mice had a greater incidence of umbilical defects than their C3H WT controls (p < 0.05) (Fig. S5.6d). The umbilical defects observed included gastroschisis and bleeding from the umbilicus. At the high dose of EtOH, maternal lethality was increased to 50% in the aCat mice, about 2-fold higher than that in C3H WT controls (p < 0.05) (Fig. 5.1C). No maternal lethality was observed at the low dose of EtOH.

275 hCat mice

EtOH-exposed C57BL/6 WT mice exhibited a dose-dependent decrease in fetal weight (p <

0.001), head length (p < 0.001), head diameter (p < 0.001) and crown-rump length (p < 0.001), and an increase in resorptions (p < 0.001) and total malformations (p < 0.001) compared to saline controls (Fig. 5.2). hCat mice were not affected at the low EtOH dose, but at the high dose exhibited reductions in all growth parameters, as well as increased resorptions and total malformations (p < 0.001). Compared to EtOH-exposed C57BL/6 WT mice at the low dose, hCat mice were protected against EtOH-initiated decreases in fetal weight (p < 0.01), head length

(p < 0.001), head diameter (p < 0.001) and crown-rump length (p < 0.001), and sustained fewer resorptions (p < 0.001) and total malformations (p < 0.05). At the high dose, hCat mice were protected against the reductions in head diameter (p < 0.05) and resorptions (p < 0.001) observed in C57BL/6 WT mice. At the low dose of EtOH, C57BL/6 WT controls exhibited increased incidences of cleft palate (p < 0.01), dilated ventricles (p < 0.05), exencephaly (0.05 < p < 0.1), hepatic defects (0.05 < p < 0.1), ocular malformations (p < 0.05), renal malformations (0.05 < p

< 0.1), and umbilical defects (0.05 < p < 0.1) compared to hCat mice (Fig. S5.7b).

Exencephalies, ventricular dilations and cleft palates resulting from EtOH exposure have previously been reported in both humans and rodents (Sulik, 2005). At the high dose of EtOH, there were no significant differences in the incidence of individual malformations between

C57BL/6 WT and hCat mice (Fig. S5.7c). When EtOH treatment groups were combined,

C57BL/6 mice exhibited higher incidences of cleft palate (0.05 < p < 0.1), dilated ventricles (p <

0.05) and ocular malformations (0.05 < p < 0.1) (Fig. S5.7d). At the low dose of EtOH, no maternal lethality was observed. At the high dose of EtOH, maternal lethality was increased non- significantly to 20% in the C57BL/6 WT mice, whereas hCat mice exhibited no lethality (Fig.

5.1C).

276

Embryopathies in mice pretreated with polyethylene glycol-conjugated catalase (PEG-Cat) aCat mice

In aCat mice receiving no EtOH, PEG-Cat treatment decreased resorptions by 63% compared to saline-treated aCat mice (p < 0.05), but no other parameters were affected (Figs. 5.3, S5.8a).

Compared to aCat mice treated with the low dose of EtOH alone, PEG-Cat pretreatment was broadly protective against EtOH teratogenesis, increasing fetal weight (p < 0.01), head length (p

< 0.001) and crown-rump length (p < 0.01), and decreasing total malformations (p < 0.05) (Fig.

5.3). There were no differences in malformations when examined by individual malformation

(Fig. S5.8b). The high dose of EtOH (6 g/kg) was maternally toxic in the aCat mice, precluding the use of this dose in investigating mechanisms exclusively within the embryo.

C57BL/6 WT mice

In C57BL/6 WT mice receiving no EtOH, PEG-Cat treatment increased fetal weight compared to the saline group (p < 0.001), and had no effect on the low baseline incidence of malformations

(Figs. 5.3, S5.9a). Compared to the WT group treated with the low dose of EtOH alone, PEG-

Cat pretreatment was broadly protective against EtOH teratogenesis, increasing fetal weight (p <

0.001), head length (p < 0.001), head diameter (p < 0.001) and crown- rump length (p < 0.001).

There was also a protective effect for PEG-Cat pretreatment in reductions of 81% in resorptions

(p < 0.01) and 53% in malformations caused by EtOH (p < 0.01). PEG-Cat pretreatment decreased the incidence of ocular malformations (p < 0.05), and possibly dilated ventricles (0.05

< p < 0.1), compared to C57BL/6 mice exposed to EtOH alone (Fig. S5.9b). The high dose of

277

PEG-Catalase (PEG-Cat) IN PEG-Catalase (PEG-Cat) IN ACATALASEMIC (aCat) MICE C57BL/6 WILD-TYPE MICE

13 1.4 13 FETAL WEIGHT HEAD LENGTH FETAL WEIGHT HEAD LENGTH 1.2 * 12 *** 12 * *** 1.2 *** ††† 11 ††† 11 1.0 1.0 ††† 10 ††† 10 ††† g, mean + SD mean g, g, mean + SD mean g, aCat C57BL/6 WT ‡‡‡ mm, mean + SD mm, mean + SD 0.8 PEG-Cat PEG-CAT 9 † 0.8 9 †††

9 3.0 9 3.0 HEAD DIAMETER CROWN-RUMP LENGTH HEAD DIAMETER CROWN-RUMP LENGTH

8 2.5 * 8 *** *** 2.5 ††† ‡‡ ††† 7 2.0 7 ††† ††† mm, mean +mm, SD mm, mean +mm, SD mean +mm, SD mean +mm, SD †† † 2.0 ‡‡‡ 6 1.5 6

100 100 100 100 RESORPTIONS TOTAL MALFORMATIONS RESORPTIONS TOTAL MALFORMATIONS 80 † 80 80 80 ††† ††† ‡‡‡ 60 60 60 60 ** ††† ‡‡‡ *** ** ††† 40 40 40 40 †††* *† (mean + SD)(mean (mean + SD)(mean (mean + SD)(mean + SD)(mean PERCENTAGE PERCENTAGE PERCENTAGE 20 PERCENTAGE 20 20 20

0 0 0 0 VH 4 6 VH 4 6 VH 4 6 VH 4 6

ETHANOL (g/kg) ETHANOL (g/kg) ETHANOL (g/kg) ETHANOL (g/kg)

278

Figure 5.3: Effect of polyethylene glycol-conjugated catalase (PEG-Cat) pretreatment on EtOH developmental toxicity in acatalasemic (aCat) mice and C57BL/6 mice.

On GD 9, pregnant aCat or C57BL/6 dams were treated i.p. with either EtOH (4 g/kg) or its saline vehicle, with or without an 8 hr pretreatment with polyethylene glycol-conjugated catalase

(PEG-Cat) (50 kU/kg) and sacrificed on GD 19. Total malformations assessed include cleft palate, ventricular dilation, exencephaly, hepatic and limb defects, microcephaly, ocular and otic defects, renal defects, spina bifida, tail and umbilical defects. Percentage was calculated by dividing the total number of fetuses affected with >1 anomaly by the number of total live fetuses.

Continuous data (fetal weight, head length, head diameter, crown-rump length) were analyzed by

Kruskal-Wallis test with a post-hoc Dunn’s test. Binomial data (resorptions and total malformations) were analyzed by chi-square test for trend. Number of litters was 2-9, with a mean of 7 litters per group overall, and 9 litters per group for only the 4 g/kg dose and saline controls. Data for aCat mice and C57BL/6 WT mice are reproduced from Fig. 2. Single daggers indicate a difference from saline controls of the same treatment group († = p < 0.05, †† = p <

0.01,††† = p < 0.001). Double daggers indicate a difference from the 4 g/kg EtOH-exposed group of the same genotype (‡‡‡ = p < 0.001). Asterisks indicate a difference from the 4 g/kg

EtOH-exposed group of the same genotype without PEG-Cat pretreatment (* = p < 0.05, ** = p

< 0.01, *** = p< 0.001).

279

EtOH (6 g/kg) was maternally toxic in the C57BL/6 mice, precluding the use of this dose in investigating mechanisms exclusively within the embryo.

Gender Effects aCat mice and hCat mice

There were no significant gender differences in structural malformations when examined by individual anomaly, separated by treatment and strain, for either aCat mice (Figs. S5.6, S5.8) or hCat mice (Figs. S5.7, S5.9).

Combined treatments and strains

When treatment groups and strains were combined, females exhibited a 6-fold greater incidence of total structural malformations than males, independent of treatment or strain (p < 0.001) (Fig.

S5.10a), with specifically greater incidences of cleft palate (p < 0.001), dilated ventricles (p <

0.001), exencephaly (p < 0.05), ocular defects (p < 0.001), tail defects (p < 0.05) and umbilical defects (p < 0.05) (Fig. S5.10b).

Strain Effects

When treatment groups and genders were combined, C57BL/6 WT mice and hCat mice together exhibited a 2-fold increase in overall structural malformations compared to C3H WT and aCat

280 mice grouped together (p < 0.001) (Fig. S5.11a), Specifically, C57BL/6 WT and hCat mice combined exhibited greater incidences of cleft palate (p < 0.05), dilated ventricles (p < 0.05) and ocular defects (p < 0.001) compared to C3H WT and aCat mice combined (Fig. S5.11b).

Oxidatively damaged embryonic DNA in ethanol-exposed progeny aCat mice

EtOH-exposed aCat embryos exhibited a 79% increase in DNA oxidation compared to C3H WT

EtOH-exposed embryos (p < 0.001), and a 6.7-fold increase compared to saline-treated aCat embryos (p < 0.001) (Fig. 5.4, upper left panel). EtOH-exposed C3H WT embryos exhibited a

3.7-fold increase in DNA oxidation compared to saline-treated C3H WT embryos (p < 0.001).

hCat mice

EtOH-exposed hCat embryos did not exhibit any increase in DNA oxidation compared to saline- treated hCat embryos, whereas EtOH-exposed C57BL/6 WT embryos exhibited a 4.3-fold increase in DNA oxidation compared to saline-treated C57BL/6 WT embryos (p < 0.001) (Fig.

5.4, upper right panel). EtOH-exposed C57BL/6 WT had greater DNA oxidation than EtOH- exposed hCat embryos (p < 0.001).

DNA oxidation in PEG-Catalase treated mice aCat mice

281

DNA OXIDATION Ethanol (EtOH)

1000 1000 ***††† hCat aCat (4) 800 g dGuo) 800  ††† 600 ††† 600 (3) (3) 400 400 ***(3) % of control % of 200 (3) (3) 200 (4) (3)

DNA OXIDATION DNA (1) 0 0 VH EtOH VH EtOH (8oxodGuo, fmol/ (8oxodGuo, VH EtOH VH EtOH WT aCat WT hCat

PEG-Catalase + EtOH

1000 1000 ††† aCat C57BL/6 WT (4) 800 g dGuo) 800  ††† 600 600 (3) 400 400 ***(3) *** % of control % of (3) (3) 200 (3) 200 (4) (3) DNA OXIDATION DNA 0 0 VH PEG-Cat PEG-Cat VH PEG-Cat PEG-Cat (8oxodGuo, fmol/ (8oxodGuo, 4 g/kg EtOH 4 g/kg EtOH

282

On GD 12, pregnant aCat and hCat dams or their respective C3H and C57BL/6 WT controls were treated i.p. with EtOH (4 g/kg) or saline vehicle (VH) and sacrificed 6 hr later. Single embryos from 3-4 different litters were analyzed for 8-oxo-2’-deoxyguanosine (8-oxodGuo) formation using high-performance liquid chromatography with tandem mass spectrometry. The number of embryos is given in parentheses. Data were analyzed by one-way ANOVA with a post-hoc Bonferroni test. Daggers indicate a difference from saline controls of the same genotype

(††† = p < 0.001). Asterisks indicate a difference from WT embryos exposed to EtOH (*** = p <

0.001).

283

In aCat mice, PEG-Cat treatment alone did not change DNA oxidation compared to saline- exposed aCat embryos (Fig. 5.4, lower left panel). EtOH exposure caused a 6.7-fold increase in embryonic DNA oxidation compared to saline-exposed controls (p < 0.001), and this increase was completely blocked by PEG-Cat pretreatment, which decreased EtOH-initiated DNA oxidation by 79% (p < 0.001), to basal levels.

C57BL/6 mice

In C57BL/6 WT mice receiving no EtOH, PEG-Cat treatment alone did not change DNA oxidation compared to saline-exposed C57BL/6 embryos (Fig. 5.4, lower right panel). EtOH exposure caused a 4.4-fold increase in embryonic DNA oxidation compared to saline-exposed controls (p < 0.001), and PEG-Cat pretreatment reduced this increase by 67% (p < 0.001), to levels that were not different from saline controls.

Ethanol metabolism aCat and C3H WT mice

In male aCat mice administered 2.8 g/kg EtOH i.p., there were no differences in EtOH plasma concentrations over time compared to C3H WT controls (Fig. 5.5, left upper panel).

284 hCat and C57BL/6 WT mice

In male hCat mice administered 2.8 g/kg EtOH i.p., there were no differences in EtOH plasma concentrations over time compared to C57BL/6 WT controls (Fig. 5.5, left middle panel).

Comparison of metabolism between C3H and C57BL/6 wild-type strains

In the first 2 hr following EtOH administration, C3H WT mice exhibited higher plasma EtOH concentrations than C57BL/6 WT mice at 0.5 (p < 0.05) and 2 hr (p < 0.01) post-injection, demonstrating a strain-dependent difference in EtOH metabolism (Fig. 5.5, left lower panel).

Comparison of developmental toxicity between C3H and C57BL/6 background strains

Untreated C57BL/6 WT mice exhibited increased fetal weight (p < 0.001) compared to untreated

C3H WT mice (Fig. 5.5). Compared to EtOH-exposed C3H WT mice, C57BL/6 mice exhibited decreased head length (p < 0.001) and crown-rump length (p < 0.001) at the low EtOH dose, and increased resorptions (p < 0.001) and total malformations (p < 0.01) at the high EtOH dose.

285

PLASMA EtOH DEVELOPMENTAL TOXICITY CONCENTRATION C3H Wild-type (WT) vs. C57BL/6 WT

12.0 *** FETAL WEIGHT HEAD LENGTH 150 1.2 aCat 11.5 C3H - Vehicle ** C3H - EtOH 100 11.0 aCat - Vehicle 1.0 aCat - EtOH 10.5

50 g, mean + SD C3H WT 0.8 C57BL/6 WT mm, mean + SD 10.0

(mmol/L, mean + SD) + mean (mmol/L, 0 9.5 Plasma EtOH concentration EtOH Plasma 9 HEAD DIAMETER CROW N-RUMP LENGTH 2.6 100 hCat C57 - Vehicle 8 ** 80 C57 - EtOH 2.4 hCat - Vehicle 60 hCat - EtOH 7 40 2.2 mm, mean + SD mm, mean + SD 20 6 2.0 (mmol/L, mean SD) + 0 Plasma EtOH concentration EtOH Plasma

100 RESORPTIONS TOTAL MALFORMATIONS 150 ** C3H WT vs. C57BL/6 WT *** 80  100  C57 - EtOH 100 60 C3H - EtOH 40 50 (mean +(mean SD) 50 +(mean SD) PERCENTAGE PERCENTAGE 20

(mmol/L, mean + SD) + mean (mmol/L, 0 0 0 0 2 4 6 8 10 VH 4 6 VH 4 6 Plasma EtOH concentration EtOH Plasma Time (h) ETHANOL (g/kg) ETHANOL (g/kg)

286

(Left panel) Male acatalasemic (aCat) mice and their C3H WT controls, and transgenic mice expressing human catalase (hCat) and their C57BL/6 WT controls, were treated with a single dose of EtOH (2.8 g/kg i.p), and sacrificed at various time points to measure plasma concentrations of EtOH. Each time point indicates the mean of 3-6 samples. Data were analyzed by two-way ANOVA with a post-hoc Bonferroni test. Alpha symbols indicate a difference from

C57BL/6 mice at the same time point ( = p < 0.05,  = p < 0.01). (Right panel) On GD 9, pregnant C3H and C57BL/6 WT dams were treated i.p. with either EtOH (4 or 6 g/kg) or its saline vehicle (VH), and sacrificed on GD 19. Total malformations assessed include cleft palate, ventricular dilation, exencephaly, hepatic and limb defects, microcephaly, ocular and otic defects, renal defects, spina bifida, tail and umbilical defects. Percentage was calculated by dividing the total number of fetuses affected with >1 anomaly by the number of total live fetuses.

Continuous data (fetal weight, head length, head diameter, crown-rump length) were analyzed by

Kruskal-Wallis test with a post-hoc Dunns test. Binomial data (resorptions and total malformations) were analyzed by chi-square test for trend. Within-group treatment differences not shown here are shown in Fig. 2. Number of litters was 7 and 9 for C3H and C57BL/6 mice, respectively. Data for C3H mice and C57BL/6 WT mice are reproduced from Fig. 2.Asterisks indicate a difference between strains within the same treatment group (** = p < 0.01, *** = p <

0.001).

287

5.5 Discussion

EtOH-enhanced ROS may adversely affect development via changes in signal transduction (Hansen and Harris, 2013) and/or by oxidatively damaging cellular macromolecules

(Wells et al., 2009b). The antioxidative enzyme catalase detoxifies H2O2 to water and oxygen

(Halliwell and Gutteridge, 2007), and studies employing exogenous forms of catalase against

ROS-initiating teratogens suggest a protective developmental role for this enzyme (Wells et al.,

2009b). However, endogenous embryonic catalase activity is low, and its importance in protecting against EtOH teratogenesis is unclear. Studies using antioxidants like EUK-134, which mimics the activity of superoxide dismutase and catalase (Chen et al., 2004a), may be complicated by unappreciated additional effects that confound interpretation of the results, and do not permit an assessment of the importance of embryonic catalase in EtOH teratogenesis.

While catalase knockout mice are viable (Ho et al., 2004), they have not been used in teratological studies, nor are null progeny likely to be common in humans. The protection shown herein by both genetically enhanced endogenous embryonic catalase, and by exogenously administered PEG-Cat, against EtOH-initiated embryonic DNA oxidation and teratogenesis, and their converse exacerbation in aCat mice, suggest that ROS and oxidatively damaged DNA are involved in the mechanism of EtOH developmental toxicity, and that embryonic catalase, while low, is an important determinant of risk.

While not measured herein, we have shown that aCat embryos have substantially lower catalase activity than their C3H WT controls, while the hCat embryos have higher activity compared to their C57BL/6 WT controls, all of which are only about 5% of maternal hepatic activity (Abramov and Wells, 2011a). Both maternal PEG-Cat administration (Abramov and

288

Wells, 2011a; Winn and Wells, 1999), and its direct addition to embryo culture medium

(Abramov and Wells, 2011b; Winn and Wells, 1995), significantly increase embryonic catalase activity, which is sustained over 24 hours, consistent with its protection against embryopathies observed herein.

Increased high-dose EtOH-initiated maternal lethality in the C57BL/6 WT dams but not the hCat dams suggests a ROS-mediated mechanism of toxicity in the mother, and a protective role for maternal catalase, corroborated by the conversely enhanced lethality in the aCat dams but not their C3H WT controls. At the embryonic level, the increased number of pups per litter in untreated hCat mice may involve preimplantation conceptal antioxidative protection, while the similar protection in EtOH-exposed hCat mice involves a postimplantation mechanism. Given the reduced maternal lethality in EtOH-exposed hCat mice, a distal protective contribution from enhanced maternal health cannot be excluded. In either case, enhanced catalase expression seems to improve reproduction, further corroborated by the converse outcomes in untreated aCat mice, which have fewer pups per litter than C3H WT controls, and when treated with EtOH exhibit greater maternal lethality.

EtOH-treated aCat mice exhibited a dose-dependent increase in teratogenesis compared to their C3H WT controls, which was blocked by PEG-Cat pretreatment, while EtOH-treated hCat mice were almost completely protected compared to their susceptible C57BL/6 WT controls, suggesting a ROS-dependent mechanism of toxicity, and a protective role for catalase.

Previous studies using PEG-Cat have demonstrated decreased teratological outcomes following exposure to other ROS-initiating teratogens (Winn and Wells, 1997; Winn and Wells,

1999; Winn and Wells, 1995). This protection is remarkable given that embryonic levels are only ~5-10% of maternal hepatic activity (Abramov and Wells, 2011b; Wells et al., 2009b;

289

Winn and Wells, 1997). The higher dose of EtOH increased toxicity for most of the developmental outcomes in the hCat mice, comparable to their C57BL/6 WT controls, suggesting that there is a threshold of oxidative stress beyond which catalase no longer offers protection. EtOH-treated C57BL/6 mice exhibited a significant increase in ocular malformations, which was not observed in the hCat mice, suggesting a ROS-mediated mechanism. However, this increase may involve an exacerbation of strain predisposition. The C3H WT controls for the aCat mice were substantially more resistant to EtOH teratogenesis than the C57BL/6 WT controls for the hCat mice, exhibiting only minor toxicity at the high dose, indicating a marked strain difference. While the mechanism of this strain-dependent resistance is not known, we have observed it with other ROS-initiating teratogens including phenytoin and methanol (Abramov and Wells, 2011a; Sweeting et al., 2011).

Females exhibited a 6-fold higher incidence of total structural malformations than males in both strains, although previous studies examining gender differences have yielded inconclusive results (Blanchard et al., 1987; Gianoulakis, 1990; Gilliam et al., 2011; McGivern et al., 1984; Zimmerberg et al., 1989). Dosing regimen, timing and mouse strain are all variables within studies that could contribute to these observed differences. Numerous studies investigating gender differences in EtOH metabolism in humans have reported contradictory results (Ammon et al., 1996; Arthur et al., 1984; Cole-Harding and Wilson, 1987; Desroches et al., 1995; Frezza et al., 1990; Goist and Sutker, 1985; Jones and Jones, 1976; Kalant, 1996;

Kwo et al., 1998; Marshall et al., 1983; Martin et al., 1985; Middaugh et al., 1992; Mills and

Bisgrove, 1983; Mishra et al., 1989; Mulvihill et al., 1997; Mumenthaler et al., 1999; Niaura et al., 1987; Radlow and Hurst, 1985; Smith et al., 1993; Sutker et al., 1983; Thomasson,

2002), although the explanation for these differences is unclear. In light of these discrepancies, the gender difference observed herein would need to be replicated in order to ensure that this

290 difference is real. If so, the molecular basis is unknown, but could reflect gender differences in one or more embryonic pathways of NOX-dependent ROS formation, numerous ROS detoxification pathways and DNA repair. Studies in rodents examining the effect of pregnancy on ethanol metabolism have been similarly inconclusive (Badger et al., 2005; Gordon et al.,

1985; Traves et al., 1995; Traves and Lopez-Tejero, 1994).

8-oxodGuo is a developmentally pathogenic lesion (McCallum et al., 2011c; Wong et al., 2008), suggesting that the increased embryonic DNA oxidation and teratogenesis in the aCat mice, and the converse decreases in the hCat mice and mice pretreated with PEG-Cat, are causally related. This protection by catalase is likely conceptal as opposed to maternal because the half-life of maternally derived hydroxyl radicals (Halliwell and Gutteridge, 2007) is too short to permit them to travel from the mother to the embryo and initiate DNA oxidation within embryonic target tissues. Alterations in maternal catalase would be less likely to alter embryonic

DNA oxidation. This supposition is consistent with definitive evidence of the protective importance of embryonic catalase demonstrated in embryo culture studies (Abramov and Wells,

2011b; Miller and Wells, 2011).

The 50% increase in baseline fetal resorptions in the untreated aCat mice compared to their C3H WT controls is consistent with published data using this mouse strain (Abramov and

Wells, 2011a), and may contribute to their reduced apparent fertility, suggesting that embryonic catalase is important for protecting against even normal physiological oxidative stress, which can be embryopathic. A pathogenic potential for endogenous ROS is consistent with the converse absence of increased resorptions in untreated hCat mice compared to their C57BL/6 WT controls. A similar embryopathic potential for endogenous embryonic ROS has been observed in vivo wherein low levels of embryonic enzymes important for antioxidative protection such as

291 glucose-6-phosphate dehydrogenase (G6PD) resulted in enhanced embryonic and fetal toxicity in the absence of xenobiotic exposure (Nicol et al., 2000). In embryo culture, protection against embryonic DNA oxidation and embryopathies in untreated progeny was afforded by treatment with exogenous antioxidative enzymes including catalase and superoxide dismutase (SOD)

(Winn and Wells, 1995), while untreated aCat mouse embryos exhibited enhanced embryopathies compared to C3H WT embryos (Miller and Wells, 2011).

In addition to its antioxidative activity, catalase via its peroxidative activity can also metabolize some alcohols like methanol and EtOH (Thurman and Handler, 1989). However, there were no differences in the plasma concentrations of EtOH between aCat mice and their

C3H WT controls, nor between hCat mice and their C57BL/6 WT controls, demonstrating that the protection afforded by catalase was not due to altered pharmacokinetics in these strains.

These observations are consistent with previous studies demonstrating no differences in EtOH disposition in acatalasemic mice either compared to wild-type mice despite differences in locomotor activity (Aragon et al., 1992), or using 14-C labeled EtOH (Karinje and Ogata, 1990).

Plasma concentrations of EtOH were lower in the C57BL/6 WT mice than in C3H WT mice, suggesting that the apparently greater susceptibility of the C57BL/6 strain to EtOH developmental toxicity compared to C3H mice is due to factors other than strain-dependent differences in plasma concentrations of EtOH. Despite these strain differences, the protection provided by PEG-Cat in both the aCat mice and the C57BL/6 WT mice confirmed that the catalase genotype-dependent differences in EtOH teratogenicity observed in aCat and hCat mice were due to alterations in catalase itself, and not due to strain differences in metabolism or susceptibility to EtOH.

292

Our results in mice with genetically altered catalase expression provide proximate evidence that ROS and oxidatively damaged DNA contribute to EtOH teratogenesis, and constitute the first direct evidence that embryonic catalase, although low compared to adult activity, is an important determinant of risk.

293

5.6 Supplemental Figures

aCat & C3H WILD-TYPE

a. BASELINE INCIDENCE OF MALFORMATIONS BY GENDER

Cleft palate N=40 N=55 FEMALE Dilated ventricles Exencephaly MALE Hepatic WT aCat Limb Microcephaly Ocular Otic Renal Spina bifida Tail defects Umbilical -40 -20 0 20 40

b. INCIDENCE OF EtOH-INITIATED MALFORMATIONS BY GENDER: 4 g/kg EtOH

Cleft palate FEMALE Dilated ventricles N=43 N=41 Exencephaly MALE Hepatic WT aCat Limb Microcephaly Ocular Otic Renal Spina bifida Tail defects Umbilical -40 -20 0 20 40

c. INCIDENCE OF EtOH-INITIATED MALFORMATIONS BY GENDER: 6 g/kg EtOH

Cleft palate FEMALE Dilated ventricles Exencephaly MALE Hepatic WT aCat Limb Microcephaly Ocular Otic Renal Spina bifida N=41 N=15 Tail defects Umbilical -40 -20 0 20 40

d. COMBINED INCIDENCE OF MALFORMATIONS IN aCat AND C3H WT MICE: 4 and 6 g/kg EtOH

Cleft palate FEMALE Dilated ventricles Exencephaly MALE Hepatic Limb WT aCat Microcephaly Ocular Otic Renal Spina bifida N=84 N=56 Tail defects Umbilical * -40 -20 0 20 40 PERCENTAGE (%)

294

Supplemental Figure S5.6: Spectrum of ethanol (EtOH)-initiated malformations in acatalasemic (aCat) mice.

On gestational day (GD) 9, pregnant aCat mice or their C3H wild-type (WT) controls were treated intraperitoneally (i.p.) with either EtOH (4 or 6 g/kg) or its saline vehicle (VH), and sacrificed on GD 19. Fetuses were stored in Bouin's fixative for 3 days, and transferred to 70%

EtOH for 1 day. Soft tissue examination was done using Wilson's freehand razorblade technique, and gender was determined identifying the presence (male) or absence (female) of testes. Data were analyzed by chi-square test. Number of litters analyzed in each group was 5-11. Asterisk indicates a difference between genotypes (* = p < 0.05).

295

hCat & C57BL/6 WILD-TYPE

a. BASELINE INCIDENCE OF MALFORMATIONS BY GENDER

Cleft palate N=70 N=53 FEMALE Dilated ventricles Exencephaly MALE Hepatic WT hCat Limb Microcephaly Ocular Otic Renal Spina bifida Tail defects Umbilical -40 -20 0 20 40 b. INCIDENCE OF EtOH-INITIATED MALFORMATIONS BY GENDER: 4 g/kg EtOH

Cleft palate FEMALE Dilated ventricles N=58 ** N=86 Exencephaly * a MALE Hepatic WT a hCat Limb Microcephaly Ocular Otic * Renal a Spina bifida Tail defects Umbilical a -40 -20 0 20 40 c. INCIDENCE OF EtOH-INITIATED MALFORMATIONS BY GENDER: 6 g/kg EtOH

Cleft palate FEMALE Dilated ventricles Exencephaly MALE Hepatic WT hCat Limb Microcephaly Ocular Otic Renal Spina bifida N=19 N=37 Tail defects Umbilical -40 -20 0 20 40 d. COMBINED INCIDENCE OF MALFORMATIONS IN hCat AND C57BL/6 WT MICE: 4 and 6 g/kg EtOH

Cleft palate a FEMALE Dilated ventricles Exencephaly * MALE Hepatic WT hCat Limb Microcephaly Ocular a Otic Renal Spina bifida N=77 N=123 Tail defects Umbilical -40 -20 0 20 40 PERCENTAGE (%)

296

Supplemental Figure S5.7: Spectrum of EtOH-initiated malformations in transgenic mice expressing human catalase (hCat).

On GD 9, pregnant hCat dams or their C57BL/6 wild-type (WT) controls were treated intraperitoneally (i.p.) with either EtOH (4 or 6 g/kg) or its saline vehicle (VH), and sacrificed on

GD 19. Fetuses were stored in Bouin's fixative for 3 days, and transferred to 70% EtOH for 1 day. Soft tissue examination was done using Wilson's freehand razorblade technique, and gender was determined identifying the presence (male) or absence (female) of testes. Data were analyzed by chi-square test. Number of litters was 2-10, with an average of 6 litters per group overall, and 8 litters per group for only the 4 g/kg dose and saline controls. Asterisks indicate a difference between genotypes (* = p < 0.05, ** = p < 0.01). Letter 'a' indicates a difference between genotypes (0.05 < p < 0.1).

297

MALFORMATIONS

a. PEG-Catalase IN SALINE-TREATED aCat MICE

Cleft palate N=55 N=38 FEMALE Dilated ventricles Exencephaly MALE Hepatic PEG-Cat aCat Limb Microcephaly Ocular Otic Renal Spina bifida Tail defects Umbilical

b. PEG-Catalase IN EtOH-TREATED aCat MICE

Cleft palate FEMALE Dilated ventricles Exencephaly MALE Hepatic PEG-Cat aCat Limb Microcephaly Ocular Otic Renal Spina bifida N=55 N=59 Tail defects Umbilical -20 0 20 PERCENTAGE (%)

298

Supplemental Figure S5.8: Effect of pretreatment with polyethylene glycol-conjugated catalase

(PEG-Cat) on the spectrum of EtOH-initiated malformations in acatalasemic (aCat) mice.

On GD 9, pregnant aCat dams were treated i.p. with either EtOH (4 or 6 g/kg) or its saline vehicle (VH), with or without an 8 hr pretreatment with PEG-Cat, and sacrificed on GD 19.

Fetuses were stored in Bouin's fixative for 3 days, and transferred to 70% EtOH for 1 day. Soft tissue examination was done using Wilson's freehand razorblade technique, and gender was determined identifying the presence (male) or absence (female) of testes. Data were analyzed by chi-square test. Number of litters analyzed in each group was 6-9.

299

MALFORMATIONS

a. PEG-Catalase IN SALINE-TREATED C57BL/6 WT MICE

Cleft palate N=70 N=44 FEMALE Dilated ventricles Exencephaly MALE Hepatic WT PEG-Cat Limb Microcephaly Ocular Otic Renal Spina bifida Tail defects Umbilical

b. PEG-Catalase IN EtOH-TREATED C57BL/6 WT MICE

Cleft palate FEMALE Dilated ventricles a Exencephaly MALE Hepatic WT PEG-Cat Limb Microcephaly Ocular Otic * Renal Spina bifida N=58 N=41 Tail defects Umbilical -20 0 20 PERCENTAGE (%)

300

Supplemental Figure S5.9: The effect of PEG-Cat pretreatment on the spectrum of EtOH- initiated malformations in C57BL/6 mice.

On GD 9, aCat mice or their C3H WT controls were treated i.p. on GD 9 with either EtOH (4 or

6 g/kg) or its saline vehicle (VH), with or without an 8 hr pretreatment with PEG-Cat, and sacrificed on GD 19. Fetuses were stored in Bouin's fixative for 3 days, and transferred to 70%

PEG-Cat-pretreated mice (* = p < 0.05). Letter 'a' indicates a difference from PEG-Cat- pretreated mice (0.05 < p < 0.1).

301

GENDER DIFFERENCES

a. TOTAL MALFORMATIONS BETWEEN GENDERS 50 40 *** 30 (372) 20

10 (368) PERCENTAGE (%) 0 FEMALE MALE GENDER

b. COMBINED INCIDENCE OF MALFORMATIONS: GENDER DIFFERENCES

Cleft palate FEMALE Dilated ventricles *** Exencephaly *** MALE Hepatic * Limb FEMALE MALE Microcephaly Ocular Otic *** Renal Spina bifida N=372 N=368 Tail defects * Umbilical * -10 0 10 PERCENTAGE (%)

302

Supplemental Figure S5.10: The effect of gender on EtOH-initiated malformations in acatalasemic (aCat) mice and transgenic mice expressing human catalase (hCat).

On GD 9, pregnant hCat dams or C57BL/6 wild-type (C57 WT) controls, or aCat dams or C3H wild-type (C3H WT) controls were treated i.p. with either ethanol (4 or 6 g/kg) or its saline vehicle (VH), and sacrificed on GD 19. (A) Offspring from all four strains were grouped irrespective of treatment or genotype and stratified by gender to assess the effect of gender on total malformations. Data were analyzed by one-way ANOVA with a post-hoc Bonferroni test.

Asterisks indicate a difference from males (*** p < 0.001). (B) Total malformations assessed include cleft palate, ventricular dilation, exencephaly, hepatic and limb defects, microcephaly, ocular and otic defects, renal defects, spina bifida, tail and umbilical defects. Percentage was calculated by dividing the total number of fetuses affected with >1 anomaly by the number of total live fetuses. Data were analyzed by chi-square test. Total number of litters analyzed was

111. Asterisks indicate a difference from males (* = p < 0.05, *** = p < 0.001).

303

STRAIN DIFFERENCES a. TOTAL MALFORMATIONS BETWEEN STRAINS 50 40 *** 30 (408) 20 (332) 10

PERCENTAGE (%) 0 C3H WT C57BL/6 WT + + aCat hCat STRAIN

b. COMBINED INCIDENCE OF MALFORMATIONS: STRAIN DIFFERENCES

Cleft palate C3H Dilated ventricles * Exencephaly * aCat Hepatic C57BL/6 C57BL/6 Limb C3H WT + hCat hCat Microcephaly + aCat Ocular Otic *** Renal Spina bifida N=332 N=408 Tail defects Umbilical -20 -10 0 10 20 PERCENTAGE (%)

304

Supplemental Figure S5.11: The effect of mouse strain on EtOH-initiated malformations in acatalasemic (aCat) mice and transgenic mice expressing human catalase (hCat).

On GD 9, pregnant hCat dams or C57BL/6 wild-type (C57 WT) controls, or aCat dams or C3H wild-type (C3H WT) controls, were treated intraperitoneally (i.p.) with either EtOH (4 or 6 g/kg) or its saline vehicle (VH), and sacrificed on GD 19. (A) Offspring from all four strains were grouped irrespective of treatment or gender and stratified by genotype to assess the effect of strain differences on total malformations. Data were analyzed by one-way ANOVA with a post- hoc Bonferroni test. Asterisks indicate a difference from grouped C3H WT + aCat mice (*** p <

0.001). (B) Total malformations assessed include cleft palate, ventricular dilation, exencephaly, hepatic and limb defects, microcephaly, ocular and otic defects, renal defects, spina bifida, tail and umbilical defects. Percentage was calculated by dividing the total number of fetuses affected with >1 anomaly by the number of total live fetuses. Data were analyzed by chi-square test. Total number of litters analyzed was 111. Asterisks indicate a difference from grouped C3H WT + aCat mice (* = p < 0.05, *** = p < 0.001).

305

Chapter 6 Embryonic DNA repair and gender are risk factors in ethanol embryopathies in oxoguanine glycosylase 1 (OGG1) knockout mice: a role for oxidatively damaged DNA and protection by a free radical spin trapping agent

Running title: Embryonic DNA repair & ethanol embryopathies

Lutfiya Miller† and Peter G. Wells*†

* Division of Biomolecular Sciences, Faculty of Pharmacy, and

†Department of Pharmacology and Toxicology, Faculty of Medicine

University of Toronto

Toronto, Ontario, Canada

a. Preliminary reports of this research were presented at the 2010 annual meeting of the Teratology Society (U.S.A.) [Birth Defects Research Part A: Clinical and Molecular Teratology 94(5): 317 (Abstract No. 13)]. These studies were supported by a grant from Canadian Institutes of Health Research. b. Individual contributions: Lutfiya Miller- mating, genotyping, dosing and sample collection, embryo culture; Peter G. Wells- supervisor.

306

6.1 Abstract

Reactive oxygen species (ROS) have been implicated in the teratogenicity of alcohol

(ethanol, EtOH). To determine the involvement of embryonic oxidative DNA damage, DNA repair-deficient oxoguanine glycosylase 1 (OGG1) knockout embryos were exposed in culture to

EtOH (2 or 4 mg/ml), with or without pretreatment with the free radical spin trap phenylbutylnitrone (PBN) (0.125 mM). Visceral yolk sacs were used to genotype embryos for

DNA repair status and gender. EtOH caused a concentration-dependent decrease in anterior neuropore closure (ANPC), somite development, turning, crown-rump length (CRL), yolk-sac diameter (YSD) and head length (HL) (p < 0.001) in all 3 ogg1 genotypes. There was a further ogg1 gene dose-dependent decrease from +/+ to -/- embryos in ANPC, somite development, turning, CRL and HL (p < 0.05), and a gene-dependent correlation between HL and ANPC (p <

0.01). Female embryos exhibited lesser ANPC and turning than males (p < 0.05), suggesting underlying gender-dependent target-specific determinants. PBN pretreatment increased ANPC, somite development, turning, CRL, YSD and HL (p < 0.001), although this protection against

EtOH was slightly less effective in -/- embryos. Oxidatively damaged DNA determined as 8- oxo-2’-deoxyguanosine (8-oxodGuo), which is repaired by OGG1, was measured in single embryos in vivo after maternal EtOH treatment (4 g/kg i.p). EtOH increased embryonic 8- oxodGuo in an ogg1 gene-dependent fashion, with the highest levels in -/- embryos. These results suggest that embryonic DNA repair status and gender are determinants of risk, and ROS- initiated embryonic DNA oxidation is involved in EtOH embryopathies.

307

6.2 Introduction

Ethanol (EtOH) consumption during pregnancy can result in both structural and/or functional deficits in the developing child collectively termed Fetal Alcohol Spectrum Disorders

(FASD), with the complete phenotype including characteristic craniofacial dysmorphology, growth retardation, and behavioural deficits (Jones, 2011). Although the mechanisms underlying these toxicities are unclear, reactive oxygen species (ROS) have been implicated (Brocardo et al., 2011). EtOH initiates oxidative damage to DNA in the developing embryo both in vitro (Lee et al., 2005) and in vivo (Dong et al., 2010; Miller et al., 2013c), as well as in developing fetal brain (Miller et al., 2013b). In the latter case, CD-1 progeny exposed in utero to a single maternal administration of EtOH during the fetal period exhibited increased fetal brain DNA oxidation and postnatal learning and memory deficits, both of which were protected by pretreatment with the free radical spin trapping agent phenylbutylnitrone (PBN) (Miller et al.,

2013b).

While ROS are necessary for normal embryonic development, any imbalance in the formation or detoxification of ROS resulting in enhanced ROS production and oxidative stress can adversely affect embryonic development, either through aberrant signal transduction or by oxidatively damaging cellular macromolecules including DNA, protein and lipid (Hansen and

Harris, 2013; Wells et al., 2009b). Oxidatively damaged DNA, and specifically the 8-oxo-2- deoxyguanosine (8-oxodGuo) lesion, is initiated by the reaction of hydroxyl radicals with guanine, and is one of the most prevalent forms of oxidative damage (Wells et al., 2009b).

Several other ROS-initiating teratogens, including phenytoin and structurally related antiepileptic drugs, thalidomide, benzo[a]pyrene and methamphetamine, increase 8-oxodGuo formation and teratogenesis in the developing embryo and fetus (Abramov and Wells, 2011a; Parman et al.,

308

1999; Winn and Wells, 1997; Winn and Wells, 1995; Wong et al., 2008). The pathogenic role of oxidatively damaged DNA and particularly the 8-oxodGuo lesion is revealed by the increased embryopathies in DNA repair-deficient knockout mice lacking key repair proteins or enzymes such as p53 (Nicol et al., 1995), ataxia telangiectasia mutated (ATM) (Bhuller and Wells, 2006;

Laposa et al., 2004), oxoguanine glycosylase 1 (OGG1) (Wong et al., 2008) and Cockayne

Syndrome B (CSB) (McCallum et al., 2011c) following exposure to ROS-initiating agents including benzo[a]pyrene, ionizing radiation, phenytoin and methamphetamine. Several potential non-mutagenic mechanisms exist whereby the 8-oxodGuo lesion may initiate teratogenesis, including altering the expression and activity of proteins required for normal embryonic development, alteration of gene transcription or expression via its ability to regulate binding affinity of various transcription factors including nuclear factor kappa B (NF-κB) to specific promoter elements, or apoptosis resulting from 8-oxodGuo accumulation (Wells et al.,

2009b; Wells et al., 2010).

OGG1 is the major DNA repair enzyme catalyzing the rate-limiting step to excise and repair the 8-oxodGuo lesion in the base excision repair (BER) pathway (Boiteux and Radicella,

2000). This DNA lesion has been implicated in the structural birth defects caused by the ROS- initiating teratogens thalidomide (Lee et al., 2011; Parman et al., 1999) and phenytoin

(Abramov and Wells, 2011a; Winn and Wells, 1995). It is the sole enzyme that contributes to

DNA repair in the developing fetal brain, where its activity is double that in maternal tissues

(Wong et al., 2008). Ogg1 knockout mice exposed in utero to methamphetamine exhibit a gender-dependent increase in fetal brain 8-oxodGuo and postnatal motor coordination deficits compared to their wild-type littermates, highlighting roles for 8-oxodGuo in the embryopathic mechanism, and gender as a risk determinant (Wong et al., 2008), While a gender-dependent

309 effect was observed in the female ogg1 knockout pups, it is unclear whether gender-dependency extends to other ROS-initiating teratogens such as EtOH for structural embryopathies.

Herein, to determine the role of ROS, oxidatively damaged DNA, DNA repair and gender in the mechanism of EtOH-initiated morphological deficits, ogg1 wild-type (+/+), heterozygous

(+/-) and homozygous (-/-) knockout progeny of +/- ogg1 dams and male breeders were exposed to EtOH in whole embryo culture on gestational day (GD) 9, with or without PBN pretreatment, and their morphological development and gender were assessed 24 hours later, the latter by PCR.

Oxidatively damaged DNA was measured in single GD 12 embryos 6 hours after in utero EtOH exposure to determine both the involvement of ROS and the potentially pathogenic role of 8- oxodGuo. Our results provide the most definitive evidence to date that ROS-initiated formation of 8-oxodGuo is involved in the mechanism of EtOH structural embryopathies, with both the macromolecular lesion in embryos and structural defects being blocked by pretreatment with the free radical spin trapping agent phenylbutylnitrone (PBN). Genotyping for gender revealed that female embryos were more susceptible to neural tube defects and dorsal-ventral flexion than males, demonstrating a gender-dependency in EtOH developmental toxicity. These results provide novel insights into the mechanism and determinants of risk for ethanol embryopathies.

310

6.3 Methods

Chemicals. Alpha-phenyl-N-t-butylnitrone (phenylbutylnitrone, PBN) minimum 98% purity by gas chromatography (GC), Hanks’ Balanced Salt Solution (HBSS), Waymouth’s MB 752/1 medium, fetal bovine serum, and penicillin-streptomycin were obtained from Invitrogen Canada

(Burlington, ON). 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), nuclease P1 and all other reagents unless otherwise specified were purchased from Sigma Aldrich (St. Louis,

MO). Proteinase K was purchased from Bioshop (Burlington, ON), saline (0.9 %, sterile) from

Baxter Corporation (Mississauga, ON) and ethanol from Commercial Alcohol Inc. (Brampton,

ON).

Animals and diet. All animal protocols were approved by the institutional animal care committee in conformance with the guidelines established by the Canadian Council on Animal

Care. Ogg1 knockout mice on a 129SV/C57BL/6J background strain were originally generated by Klungland and coworkers (1999), and generously provided by Dr. Tomas Lindahl (Imperial

Cancer Research Fund, UK) through Dr. Christi A. Walter at the University of Texas Health

Science Center at San Antonio. Mice were housed in vented plastic cages from Allentown, Inc.

(Allentown, NJ) with ground corncob bedding (Bed-O’Cobs Laboratory Animal Bedding, The

Andersons Industrial Products Group, Maumee, OH). Mouse cages were maintained in a room with controlled light (14 hr light-10 hr dark cycle) and climate (20ºC, 50% humidity), and provided with rodent chow (Harlan Labs: 2018, Harlan Teklad, Montreal, QC) and tap water ad libitum. Mice were acclimatized for 1week prior to use. Virgin heterozygous females were mated with a heterozygous male overnight from 5:00 p.m. to 9:00 a.m. by housing one male in a cage

311 containing 1-3 females. The presence of a vaginal plug the next morning was designated as gestational day (GD) 1 and plugged females were separated, weighed and housed in groups of up to four per cage until use.

DNA extraction and digestion. DNA was extracted from single GD 12 embryos using a chaotropic sodium iodide (NaI) method previously described (Ravanat et al., 2002). Briefly, samples were homogenized in 1 mL of cold lysis buffer (320 mM sucrose, 5 mM MgCl2, 10mM

Tris, 0.1 mM desferoxamine, 1% Triton X-100, pH 7.5) and centrifuged twice at 10,000 X g for

10 min at 4°C to isolate the nuclear pellet containing DNA. The pellet was then incubated with

200 uL enzyme reaction solution (1% w/v SDS, 5 mM EDTA-NA2, 0.15 mM desferoxamine 10 mM Tris-HCl pH 8.0) for 1 hr at 50ºC with RNase A/T1 mix (624 and 312 U/ml final activities, respectively). Proteinase K (1.8 mg/ml final concentration) was added for an additional 1 hr. The

DNA pellet obtained after extraction was then washed 5 times with 70% EtOH. Samples were resuspended in 200 μl of sodium acetate buffer (20 mM, pH 4.8) and sonicated into solution.

DNA purity was determined by measuring the absorbance ratio at 260/280 nm of 2 μL of sample in Na-acetate buffer (total volume 200 μL). Samples were digested with nuclease P1 (5

U/sample, 1 hr, 37°C) and calf intestinal alkaline phosphatase (6 U/sample, 1 hr, 37°C) and filtered through Amicon UltraTM filter units (YM-10, 10,000 MW cutoff; Millipore, Billerica,

MA) to remove DNA digestion enzymes and large particulates.

312

Quantification of oxidative damage to DNA using 8-OHdG ELISA. Oxidative

DNA damage was quantified using the 8-OHdG ELISA kit (JaICA, Fukuroi, Japan) according to the manufacturer’s instructions.

Embryo Culture. On GD 8.5, pregnant heterozygous ogg1 dams were sacrificed by cervical dislocation. The uterus containing the embryos was immediately removed, and rinsed with HBSS maintained at 37°C. Using a dissecting microscope (Stemi SV11, Carl Zeiss, Oberkochen,

Germany), the individual implantation sites were exposed using number five watchmaker’s forceps (Fine Science Tools, North Vancouver, BC). The decidua, trophoblast, parietal endoderm and Reichert’s membrane were carefully removed, leaving the amnion, parietal endoderm, visceral yolk sac and ectoplacental cone intact. Embryos dissected at the 7-9-somite stage were washed in pregassed (5% CO2 in air) holding medium (HM) (17 ml male rat serum, 2.5mM

HEPES, 50 units/ml of penicillin, 50 mg/ml of streptomycin, and Waymouth’s MB 752/l medium) and transferred to a 48–well plate (BD Biosciences, Franklin Lakes, NJ) containing either PBN (0.125 mM) or its 80:20 saline:water vehicle in 1 mL embryo culture medium

(ECM; HM supplemented with 15 mL fetal bovine serum). Each plate well was sealed air tight with optically clear sealing tape (Sarstedt, Montreal, QC). After 30 min, embryos were washed briefly in HM, and transferred to a 24-well plate (BD Biosciences, Franklin Lakes, NJ) containing 2 mL ECM (1 embryo per well) containing either EtOH (2 or 4 mg/ml) or saline vehicle for a total of 24 hr. The 24-well plate was incubated at 37ºC in a Sanyo model MCO-17A

CO2 incubator (Sanyo Electric Co., Ltd., Japan) on a platform rocker (Bellco Biotechnology,

Vineland, NJ). After 24 hr, embryonic morphological and developmental parameters were observed using a dissecting microscope (Stemi SV11, Carl Zeiss, Oberkochen, Germany).

313

Developmental parameters included dorsal-ventral flexure (turning), anterior neuropore closure and somite development. Functional and morphological assessments included heart rate, yolk sac diameter (in mm), crown-rump length (in mm) and head length (in mm). Overall lethality was measured in all groups as a percentage. Visceral yolk sacs were snap frozen for genotyping ogg1 genotype and gender.

Genotyping for ogg1 genotype. DNA was isolated from visceral yolk-sacs by heating the sample in 75 µL alkaline lysis reagent (25 mM NaOH, 0.2 mM disodium EDTA, pH 12) for 1 hour. Samples were then neutralized by addition of 75 µL of neutralizing buffer (40 mM Tris-

HCl, pH 5), and genotyped using a PCR-based assay. Primers (synthesized by The Center for

Applied Genomics, The Hospital for Sick Children, Toronto, ON) used to amplify the 500 bp band for the ogg1 gene were: ogg1-sense (5’-ACTGCATCTGCTTAATGGCC-3’) (forward primer) and ogg1-antisense (5’-CGAAGGTCAGCACTGAACAG-3’) (reverse primer). Primers used to amplify the 300 bp band for the ogg1 knockout gene, the primers used were neo-sense

(5’-CTGAATGAACTGCAGGACGA-3’) (forward primer) and neo-antisense (5’-

CTCTTCGTCCAGATCATCCT-3’) (reverse primer). PCR reaction conditions were: 8 μl genomic DNA, 5 μl per sample of 10X Hotstart Buffer (Fermentas Life Sciences, Burlington,

ON), 3 μl per sample of MgCl2 (Fermentas Life Sciences), 0.25 μl Maxima Hotstart Taq polymerase (Fermentas Life Sciences), 1 μl per sample of 10 mM deoxyribonucleotides (dNTP)

(Fermentas Life Sciences), 0.75 μL per sample of each 20 μM primer and 29.75 μl per sample of ddH2O for a final volume of 50 μl. Cycling conditions were: 95°C for 5 min; 35 cycles of: 94°C for 1 min, 55°C for 1.5 min, 72°C for 2 min and completed with a final extension at 72°C for 10 min and then placed on hold at 4°C. 6X DNA loading dye (0.03% bromophenol blue/0.03%

314 xylene cyanol FF,) was added to each sample. The PCR products were separated on a gel consisting of 1.5% (w/v) agarose, 40mM Tris, 19.4mM glacial acetic acid, 2.5mM EDTA and 8

μl ethidium bromide. The agarose gel was run at 100 V for 60 min, and then visualized and photographed under an ultraviolet light.

Genotyping for gender. DNA was isolated from visceral yolk sacs as outlined above. Two sets of primers were used, one specific for the male-specific Zfy gene (Sah et al., 1995), and the other specific for the autosomal Nf1 gene. Primers (synthesized by The Center for Applied Genomics,

The Hospital for Sick Children) used to amplify Zfy were: forward 5’-

AAGATAAGCTTACATAATCACATGGA-3’ and reverse 5’-

CCTATGAAATCCTTTGCTGCACATGT-3’. Primers used to amplify Nf1 were: forward 5’-

GGTATTGAATTGAAGCAC-3’ and reverse 5’- TTCAATACCTGCCCAAGG-3’. PCR reaction conditions were: 8 μl genomic DNA, 5 μl per sample of 10X Hotstart Buffer (Fermentas

Life Sciences), 3 μl per sample of MgCl2 (Fermentas Life Sciences), 0.25 μl Maxima Hotstart

Taq polymerase (Fermentas Life Sciences), 1 μl per sample of 10 mM deoxyribonucleotides

(dNTP) (Fermentas Life Sciences), 0.75 μL per sample of each 20 μM primer and 29.75 μl per sample of ddH2O for a final volume of 50 μl. Cycling conditions were: 95°C for 5 min; 35 cycles of: 94°C for 1 min, 55°C for 1.5 min, 72°C for 2 min and completed with a final extension at 72°C for 10 min and then placed on hold at 4°C. 6X DNA loading dye (0.03% bromophenol blue/0.03% xylene cyanol FF,) was added to each sample. The PCR products were separated on a gel consisting of 1.5% (w/v) agarose, 40mM Tris, 19.4mM glacial acetic acid, 2.5mM EDTA and 8 μl ethidium bromide. The agarose gel was run at 100 V for 60 min, and then visualized and photographed under an ultraviolet light.

315

Statistical analysis. Statistical analysis was performed using GraphPad Prism, Version 5

(GraphPad Software, Inc., San Diego, CA). Multivariate analysis was performed using a 1- or 2- way analysis of variance (ANOVA) with a post-hoc Bonferroni test. Binomial data were analyzed using a chi-square test, or a chi-square test for trend where a trend was observed. The minimal significance level used throughout was p < 0.05.

316

6.4 Results

DNA oxidation

Saline-treated embryos of all 3 genotypes exhibited similar baseline DNA oxidation (Fig. 6.1, whereas. EtOH-treated embryos exhibited an ogg1 gene dose-dependent increase in DNA oxidation. Compared to saline-treated controls of the same ogg1 genotype, EtOH enhanced

DNA oxidation by 2.3-fold in +/+ embryos (p < 0.001), 3.8-fold in +/- embryos (p < 0.01) and

4.6-fold in -/- embryos (p < 0.001). EtOH-initiated DNA oxidation was increased in an ogg1 gene dose-dependent fashion, as evidenced by EtOH-treated -/- embryos exhibiting 1.4-fold and

2.7-fold higher DNA oxidation than +/- (p < 0.05) and +/+ (p < 0.001) EtOH-treated embryos, respectively. PBN-pretreatment decreased DNA oxidation in +/+, +/- and -/- embryos in a gene dose-dependent fashion by 36% (p < 0.001), 62% (p < 0.05) and 70% (p < 0.001), respectively.

Effect of OGG1 on ethanol embryopathies

Saline-treated embryos from all 3 genotypes developed normally (Figs. 6.2, S6.6). Compared to saline-treated embryos, the low concentration of EtOH was embryopathic in +/+ embryos as evidenced by a decrease in somite development (p < 0.01). At the low concentration of EtOH,

+/- DNA repair-deficient embryos exhibited decreases in anterior neuropore closure (p < 0.05), somites developed (p < 0.001), turning (p < 0.05), crown-rump length (p < 0.01) and head length

(p < 0.01), all of which were further decreased at the high concentration of EtOH. At the low concentration of EtOH, -/- DNA repair-deficient embryos exhibited decreases in anterior neuropore closure (p < 0.001), somite development (p < 0.001), turning (p < 0.01), crown-rump

317

100 ***# 80 †††

g DNA) SALINE  PBN † 60 EtOH + SD PBN + EtOH 40 mean  20

8-oxodGuo (fmol/ 8-oxodGuo 0 +/+ +/- -/- 100 ***# SALINE ††† PBN 80 ††  (3) EtOH g DNA)

 PBN + EtOH (3) 60  + SD †††

40 †††

mean  (3) (5) 20 (3) (5) (3) (3) (3) (3) (3) (3) 8-oxodGuo (fmol/ 8-oxodGuo 0 +/+ +/- -/- EMBRYONIC OGG1 GENOTYPE

318

Pregnant ogg1 +/- dams mated with +/- male breeders were treated on gestational day (GD) 12 with EtOH (4 g/kg i.p.) or saline vehicle, with or without pretreatment with the free radical spin trapping agent PBN (40 mg/kg i.p.) and sacrificed 6 hr later. Single embryos from at least 3 litters containing embryos from all 3 ogg1 genotypes were analyzed for 8-oxo-2’- deoxyguanosine (8-oxodGuo) formation using an ELISA kit with an 8-oxodGuo-specific antibody. The top and bottom panels show different representations of the same data. Data were analyzed by one-way ANOVA with a post-hoc Bonferroni test. Lines indicate analysis by

Student’s t-test. Daggers indicate a difference from vehicle control group for the same genotype

(† = p < 0.05, †† = p < 0.01, ††† = p < 0.001). Asterisks indicate a difference from +/+ embryos for the same treatment group (*** = p < 0.001). Pound symbols indicate a difference from +/- embryos for the same treatment group (# = p < 0.05). Psi symbols indicate a difference from

EtOH-treated embryos for the same genotype (ΨΨΨ = p < 0.001). The number of embryos is given in parentheses.

319

150 40 Anterior Neuropore Closure Somites Developed

30 (11) †† ††† (18)††† (12) (11) (18) (12) ‡‡ *** 100 (13)(5) (10)  ††† ††† (17) † 20 (5) (5) ‡ (19) 50 †† ††† †††** (14) SD) + mean (n, 10 (24) PERCENTAGE (%) (21)††† (18) N/A 0 0 4 150 Crown-rump Length Turning

3 (11) † (18) (11) (18) (12) †† (12) 100 (13)(5) (10)†† (5) * (17) 2 ††† † (5) †† (19) 50 (14) ††* ††† 1 (24) (21) SD) + mean (mm, PERCENTAGE (%) ††† (18) N/A 0 0 2.0 4 Yolk Sac Diameter Head Length (11) † 3 (17) (12) 1.5 ‡‡ ‡‡ (14) (18) (11) † *** (19)(24) (21) † ††† (18) ††† (12) (18) (17) (19) †††** (14) (24) 1.0 (21)††† 2 (18)

1 0.5 (mm, mean + SD) + mean (mm, (mm, mean + SD) + mean (mm,

0 0.0 +/+ +/- -/- +/+ +/- -/- EMBRYONIC OGG1 GENOTYPE EMBRYONIC OGG1 GENOTYPE

250 Heart Rate 200 ††† † Saline (14) (24) (21)(18) 2 mg/ml EtOH 150 (19) (17) (11) (18) (12) 100 4 mg/ml EtOH

50 (beats/min, mean SD) (beats/min, + 0 +/+ +/- -/- EMBRYONIC OGG1 GENOTYPE

320

Figure 6.2: Effect of oxoguanine glycosylase 1 (ogg1) genotype on ethanol (EtOH) embryopathies in embryo culture.

Heterozygous ogg1 dams and males were mated overnight, plug = gestational day (GD) 1. On

GD 9, ogg1 wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout embryos from the same litter containing 7 or 8 somite-pairs were explanted and cultured in the presence of

EtOH (2 or 4 mg/ml) or saline for 24 hr. The number of embryos is given in parentheses.

Continuous data (somites developed, crown-rump length, yolk sac diameter, head length, heart rate) were analyzed by one-way ANOVA with a post-hoc Bonferroni test. Binomial data

(anterior neuropore closure, turning) were analyzed by chi-square test for trend. Daggers indicate a difference from saline of the same genotype († = p < 0.05, †† = p < 0.01, ††† = p < 0.001).

Double-daggers indicate a difference from the 2 mg/ml EtOH group of the same genotype (‡ = p

< 0.05, ‡‡ = p < 0.01). Asterisks indicate a difference from +/+ embryos for the same treatment group (* = p < 0.05, ** = p < 0.01, *** = p < 0.001). Psi symbols indicate a difference from +/- embryos for the same treatment group (ψ = p < 0.05, ψψ = p < 0.01).

321 length (p < 0.001) and head length (p < 0.001). The high concentration of EtOH was exceptionally embryopathic in -/- embryos, as evidenced by 100% failure to close their anterior neuropore (p < 0.001) or complete turning (p < 0.001), precluding an evaluation of somites developed or crown-rump length. Head length was also decreased (p < 0.001). There was an ogg1 gene dose-dependent EtOH effect exhibited for decreased anterior neuropore closure (p <

0.01), somites developed (p < 0.001), turning (p < 0.05), crown-rump length (p < 0.05) and head length (p < 0.001), whereby -/- embryos were more affected than +/- and +/+ embryos, and +/- embryos were more affected than +/+ embryos (Fig. 6.2). The head length of embryos that failed to close their anterior neuropore was decreased in a gene dose-dependent fashion, whereby -/- embryos with an open neuropore had smaller head length than +/+ embryos with an open neuropore (p < 0.05) (Fig. 6.3).

Effect of PBN in embryo culture

PBN-pretreatment was broadly protective, reducing EtOH embryopathies in all 3 genotypes, as evidenced by increases in anterior neuropore closure (p < 0.01), somites developed (p < 0.001), turning (p < 0.01), crown-rump length (p < 0.05), yolk sac diameter (p < 0.05) and head length (p

< 0.001) (Fig. 6.4). There was a gene dose-dependent protection by PBN as evidenced by PBN- pretreated +/+ embryos having increased somites developed (p < 0.05) and head length (p <

0.05) than PBN-pretreated -/- embryos.

322

8 HEAD LENGTH VS. ANTERIOR NEUROPORE 6 CLOSURE C = Closed O = Open †† ††† 4 (9) (17) (28)†††* (22) (26) (15) 2 HEAD LENGTH (mm, mean + SD) + mean (mm, 0 C O C O C O +/+ +/- -/- OGG1 GENOTYPE

323

Figure 6.3: Genotype-dependent correlation between head length and anterior neuropore closure.

Embryos from culture were grouped according to open or closed neuropores, irrespective of treatment. The number of embryos is given in parentheses. Data were analyzed by one-way

ANOVA with a post-hoc Bonferroni test. Daggers indicate a difference from the closed neuropore (C) bar for embryos of the same genotype (†† = p < 0.01, ††† = p < 0.001). Asterisk indicates a difference from the open neuropore (O) bar of WT embryos (* = p < 0.05).

324

35 Anterior Neuropore Closure Somites Developed 100 * ** 30 * ** *** † 50 25

20 mm, mean + SD + mean mm, 0 SD + mean mm, 15

3.0 Crown-rump Length 100 Turning * ** * * 2.5 * 50 2.0 mm, mean + SD + mean mm, 0 SD + mean mm, 1.5

3.0 Yolk Sac Diameter 1.4 Head Length † 1.2 2.5 * *** ** ** 1.0

2.0 0.8 mm, mean + SD + mean mm, mm, mean + SD + mean mm, 0.6 +/+ +/- -/- +/+ +/- -/- EMBRYONIC OGG1 GENOTYPE EMBRYONIC OGG1 GENOTYPE

160 Heart Rate + SD 140 4 mg/ml EtOH 120 0.125 mM PBN + 4 mg/ml EtOH 100 

beats/min, mean beats/min, 80 +/+ +/- -/- EMBRYONIC OGG1 GENOTYPE

325

Figure 6.4: Effect of PBN pretreatment on EtOH embryopathies in oxoguanine glycosylase 1

(ogg1) knockout embryos in culture.

As discussed in fig. 6.2, ogg1 wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout embryos from the same litter containing 7 or 8 somite-pairs were explanted and cultured in the presence of EtOH (4 mg/ml) or saline for 24 hr, with or without 30 min PBN pretreatment (0.125 mM). The number of embryos assessed range from 5-21 per group. Data for

+/+, +/- and -/- EtOH-treated embryos is reproduced from fig. 2. Continuous data (somites developed, crown-rump length, yolk sac diameter, head length, heart rate) were analyzed by one- way ANOVA with a post-hoc Bonferroni test. Binomial data (anterior neuropore closure, turning) were analyzed by chi-square test for trend. Asterisks indicate a difference from EtOH- treated embryos for the same genotype (* = p < 0.05, ** = p < 0.01, *** = p < 0.001). Dagger indicates a difference from +/+ embryos for the same treatment group († = p < 0.05).

326

Gender differences in EtOH embryopathies

No gender differences were evident in control +/+ embryos treated with saline. At the low concentration of EtOH, compared to male +/- littermates, female +/- embryos exhibited decreased turning (p < 0.05) and anterior neuropore closure (p < 0.05). At the high concentration of EtOH, female +/+ embryos exhibited decreased turning compared to +/+ male littermates (p <

0.05) (Fig 6.5).

327

TURNING p < 0.05 p < 0.05 150 150 Turning Turning

(6) (9) (3) (5) (9) (9) 100 100 (9) (9) (8) (9) 50 †† 50 * ‡‡ (10) ††* (9) (7) ††† (14) PERCENTAGE (%) PERCENTAGE ††† (15) ††† (%) PERCENTAGE ** (5) (11) (7) 0 0

ANTERIOR NEUROPORE CLOSURE p < 0.05 150 150 Anterior Neuropore Closure Anterior Neuropore Closure

(6) (9) (3) (5) (9) (9) 100 100 (8) (9) (9) † †† †† 50 †† 50 (9) **†† (9) (7) ‡ (10)†††

PERCENTAGE (%) PERCENTAGE (14)*** (%) PERCENTAGE * ††† (15) ††† ††† (5) (11) (7) 0 0 +/+ +/- -/- +/+ +/- -/- FEMALES MALES

Saline 2 mg/ml EtOH 4 mg/ml EtOH

328

Figure 6.5: Determination of embryonic gender by genotype, and the effect of gender and oxoguanine glycosylase 1 (ogg1) genotype on EtOH embryopathies in embryo culture.

As discussed in fig. 6.2, ogg1 wild-type (+/+), heterozygous (+/-) or knockout (-/-) embryos from the same litter containing 7 or 8 somite-pairs were explanted and cultured in the presence of

EtOH (2 or 4 mg/ml) or saline for 24 hr. After the 24 hr culture period, the visceral yolk sac from each embryo was snap frozen and analyzed for gender. Primers targeting the ZFY gene of the male sex-determining region Y amplify a 610 bp product in male embryos that is not present in females. Upper panel: The presence of a 610 bp fragment signifies a male genotype, while its absence indicates a female. Primers targeting the autosomal Nf1 gene is used as a loading control. Lower panels: Data for turning and anterior neuropore closure were separated by gender and analyzed for gender differences using a chi-square test for trend. Daggers indicate a difference from saline of the same genotype († = p < 0.05, †† = p < 0.01, ††† = p < 0.001).

Double-daggers indicate a difference from the 2 mg/ml EtOH group of same genotype (‡ = p <

0.05, ‡‡ = p < 0.01). Psi symbols indicate a difference from heterozygous embryos of the same treatment group (ψψ = p < 0.01). Asterisks indicates a difference from +/+ embryos for the same

EtOH-treated group (* = p < 0.05, ** = p < 0.01, *** = p < 0.001). Gender differences are shown graphically by the lines in the figure.

329

6.5 Discussion

Herein, using oxoguanine glycosylase 1 (OGG1) knockout mice that are unable to repair the 8-oxodGuo lesion, we provide the most direct evidence to date that oxidatively damaged

DNA, and particularly the 8-oxodGuo lesion, plays a pathogenic role in the mechanism of EtOH teratogenicity, and that embryonic DNA repair, specifically OGG1, is a determinant of risk for

EtOH embryopathies. The pathogenic role of 8-oxodGuo was evidenced by ogg1 gene dose- dependent increases in DNA oxidation and embryopathies, both of which were protected by pretreatment with the free radical spin trapping agent PBN, corroborating the initiating role of

While adult ogg1 knockout mice accumulate 8-oxodGuo (Klungland et al., 1999), saline- treated +/- and -/- OGG1-deficient embryos exhibited no increase in baseline DNA oxidation, and they all developed normally in culture. This suggests that oxidative stress may be lower during the embryonic period than in adults, and/or that considerable time is necessary for the accumulation of measurably increased 8-oxodGuo lesions in OGG1-deficient animals. The latter possibility would be consistent with the increased levels of 8-oxodGuo in aging but not young wild-type mice expressing prostaglandin H synthase 1 (PHS1), which bioactivates endogenous substrates like neurotransmitters to free radical intermediates, compared to aging PHS-knockout controls (Jeng et al., 2011).

Previous studies have demonstrated that EtOH can initiate 8-oxodGuo in the developing embryo and fetus. EtOH-exposed organogenesis-stage embryos exhibited enhanced DNA

330 oxidation, apoptosis and NADPH oxidase (NOX) induction, an enzyme dedicated to ROS production (Lambeth, 2004), all of which were blocked by pretreatment with the NOX inhibitor diphenylene iodonium chloride (DPI). During the later fetal period, in utero EtOH exposure enhanced the level of oxidatively damaged DNA in fetal brain, and caused postnatal learning and memory deficits in the progeny of CD-1 mice (Miller et al., 2013b).

Phenylbutylnitrone (PBN) is a free radical spin trapping agent that minimizes the damaging effects of excessive ROS by reacting with and thereby stabilizing free radicals (Janzen et al., 1985). PBN has been demonstrated to protect against oxidative damage to cellular macromolecules initiated by ROS-initiating teratogens such as EtOH (Miller et al., 2013b), phenytoin (Liu and Wells, 1994; Wells et al., 1989) and thalidomide (Lee et al., 2011; Parman et al., 1999), as well as mitigating the associated developmental toxicity.

In embryos herein exposed only to saline vehicle, the decreased embryonic DNA oxidation with PBN pretreatment suggests that PBN may be scavenging endogenous ROS, which can be developmentally pathogenic. The pathogenic potential of endogenously initiated ROS has been previously observed in the progeny of untreated dams deficient in glucose-6-phosphate dehydrogenase (G6PD) (Nicol et al., 2000) and catalase (Abramov and Wells, 2011a), which are important in ROS detoxification, and mice deficient in ataxia telangiectasia mutated (ATM)

(Bhuller and Wells, 2006), which is important in DNA repair, in each case compared to their wild-type normal littermates.

In EtOH-exposed progeny, PBN pretreatment reduced DNA oxidation to control levels in all ogg1 genotypes, consistent with a protective mechanism via free radical scavenging and ROS reduction, although other ROS-independent effects of PBN cannot be excluded. The decrease in

EtOH-initiated DNA oxidation afforded by PBN was paralleled by protection against EtOH-

331 initiated structural embryopathies in an ogg1 gene dose-dependent fashion. PBN-pretreated

EtOH-treated -/- embryos had significantly lesser somite development and head length compared to PBN-pretreated EtOH-treated +/+. This somewhat reduced ability of PBN to protect -/- knockout embryos from some embryopathic effects of EtOH compared to similarly treated +/+ and +/- littermates reveals the predominant protective importance of embryonic DNA repair.

PBN has demonstrated broad protective ability in models of developmental toxicity with other ROS-initiating teratogens by reducing oxidative damage to cellular macromolecules (DNA, protein, lipids) as well as reducing the associated birth defects in several models. In pregnant

New Zealand White rabbits, PBN blocked embryonic DNA oxidation in rabbit embryos exposed in utero to the sedative/antileprotic/anticancer drug thalidomide, as well as decreased the incidence of birth defects including phocomelia, the hallmark of thalidomide teratogenicity (Lee et al., 2011; Parman et al., 1999). In pregnant CD-1 mice, PBN blocked protein oxidation in mouse embryos exposed in utero to the antiepileptic drug phenytoin as well as decreased birth defects (Liu and Wells, 1994; Wells et al., 1989). Furthermore in vitro, PBN has been demonstrated to trap free radical intermediates of phenytoin and structurally similar antiepileptic drugs, thereby reducing macromolecular oxidation (Parman et al., 1998).

Independent of treatment, there was an ogg1 gene dose-dependent correlation between head length and anterior neuropore closure, whereby -/- embryos with open anterior neuropores had significantly smaller head length than +/+ embryos with open anterior neuropores. Closure of the anterior neuropore and development of the head may be particularly sensitive to oxidative stress, and hence similarly susceptible to alterations in embryonic DNA repair capacity.

Female +/+ and +/- embryos exhibited a lower incidence of turning at the high and low concentrations of EtOH, respectively, and female +/- embryos exhibited a lower incidence of

332 anterior neuropore closure at the low concentration of EtOH. This concentration-dependent susceptibility by genotype suggests that differences in EtOH developmental toxicity at the high concentration can be sustained and observed in the +/+ embryos as they have full capacity to repair their DNA, while gender-differences at the low concentration of EtOH were observed in the +/- embryos possibly due to their decreased ability to repair DNA. There were no gender differences among the -/- embryos as the low and high concentrations of EtOH were equally toxic in both genders, highlighting the important role of embryonic DNA repair as a determinant of risk. Failure to close the anterior neuropore can result in neural tube defects such as exencephaly and anencephaly (Moore and Persaud, 2007). This is consistent with previous observations that incidences of human anencephaly, and exencephaly in mouse models, have been observed to occur more frequently in females than males in humans and animal models

(Copp et al., 1990). Additionally, mice lacking p53 demonstrated gender-dependent incidence of exencephaly, whereby 100% of exencephalic embryos were female (Nicol et al., 1995; Sah et al., 1995). The mechanisms underlying this gender bias in neural tube defects is unclear.

In summary, EtOH-initiated embryonic DNA oxidation, and particularly 8-oxodGuo formation, in vivo, and embryopathies in culture, were both enhanced in ogg1 knockout mice in a gene dose-dependent fashion, indicating a pathogenic role for 8-oxodGuo, and an embryoprotective role for OGG1. The reduction in both EtOH-initiated embryonic DNA oxidation and structural embryopathies by pretreatment with the free radical spin trapping agent

PBN suggests a ROS-dependent mechanism. Among -/- ogg1 progeny, the greater increase in some EtOH embryopathies in females compared to male littermates reveals a gender-dependent risk, the mechanism of which remains to be determined. These results suggest that embryonic

DNA repair may be a possible genetic determinant of teratological risk for children exposed in utero to EtOH.

333

6.6 Supplemental Figures

150 Anterior Neuropore Closure 30 Somites Developed

100 SALINE 25

50 20

EtOH 2mg/ml SD + mean mm, 15

PERCENTAGE (%) 0 EtOH 4mg/ml

150 3.0 Turning Crown-rump Length

100 2.5

50 2.0 mm, mean + SD + mean mm, mm, mean + SD mean + mm, 0 1.5

3.0 1.4 Yolk Sac Diameter Head Length 1.2 2.5 1.0

2.0 0.8 mm, mean + SD + mean mm, mm, mean + SD + mean mm, 0.6 1.5 +/+ +/- -/- EMBRYONIC OGG1 GENOTYPE EMBRYONIC OGG1 GENOTYPE

Saline2 mg/ml EtOH 4 mg/ml EtOH

334

Supplemental Figure S6.6: Effect of oxoguanine glycosylase 1 (ogg1) genotype on ethanol

(EtOH) embryopathies in embryo culture.

Heterozygous ogg1 dams and males were mated overnight, plug = gestational day (GD) 1. On

GD 9, ogg1 wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout embryos from the same litter containing 7 or 8 somite-pairs were explanted and cultured in the presence of

EtOH (2 or 4 mg/ml) or saline for 24 hr. The number of embryos is given in parentheses. All data except heart rate are taken from fig. 2 and represented in linear format without statistical symbols to facilitate group comparisons. Refer to fig. 2 for numbers per group and statistical analyses.

335

Chapter 7 Embryonic DNA repair and ethanol-initiated behavioural deficits in oxoguanine glycosylase 1 (OGG1) knockout mice: a pathogenic role for oxidatively damaged DNA and protection by a free radical spin trapping agent

Running title: Embryonic DNA repair & ethanol behavioural deficits

Lutfiya Miller†, Daniel J. Pinto† and Peter G. Wells*†

*Division of Biomolecular Sciences, Faculty of Pharmacy; and

†Department of Pharmacology and Toxicology, Faculty of Medicine

University of Toronto

Toronto, Ontario, Canada

a. A preliminary report of this research was presented at the 2010 annual meeting of the Society of Toxicology [The Toxicologist 132(1): 218 (Abstract No. 1022)]. These studies were supported by a grant from Canadian Institutes of Health Research. b. Individual contributions: Lutfiya Miller- mating, genotyping, dosing and sample collection; Daniel J. Pinto- genotyping and behavioural testing, trained and supervised by LM; Peter G. Wells- supervisor.

336

7.1 Abstract

Consumption of alcohol (ethanol, EtOH) during pregnancy can cause a spectrum of structural, cognitive and behavioural problems in the developing child termed Fetal Alcohol Spectrum

Disorders (FASD). Reactive oxygen species (ROS) have been implicated in the mechanism of behavioural teratogenicity, although the role of embryonic DNA repair and oxidatively damaged

DNA are unclear. To determine this, a passive avoidance learning test was assessed postnatally in the progeny of oxoguanine glycosylase 1 (OGG1)-deficient mice treated once on gestational day 17 with 2 g/kg i.p. of EtOH or its saline vehicle, with or without pretreatment with the free radical spin trapping agent alpha-phenyl-N-tert-butylnitrone (PBN) (40 mg/kg i.p.). Saline- exposed progeny exhibited an ogg1 gene dose-dependent learning deficit, demonstrating for the first time a phenotype of behavioural deficits in these knockout mice. EtOH-exposed progeny exhibited learning deficits at 6, 9 and 12 weeks of age, as well as enhanced fetal brain DNA oxidation measured as 8-oxo-2’-deoxyguanosine (8-oxodGuo), in a gene dose-dependent fashion. PBN pretreatment protected +/+, +/- and -/- ogg1 progeny from EtOH-initiated learning deficits and fetal brain DNA oxidation, although slightly less so in +/- and -/- progeny. These results provide the most direct evidence to date that ROS-mediated embryonic 8-oxodGuo is a pathogenic lesion in EtOH-initiated behavioural deficits, and embryonic DNA repair status is a determinant of risk.

337

7.2 Introduction

The consequences of in utero EtOH exposure are well characterized, and include a spectrum of anomalies termed the Fetal Alcohol Spectrum Disorder (FASD), which includes both structural and functional birth defects, with the complete phenotype including characteristic craniofacial dysmorphology, growth retardation and behavioural deficits (Jones, 2011). The estimated incidence is 1 out of 100 live births in Canada (Stade et al., 2009), and is postulated to be one of the leading preventable causes of neurodevelopmental disorders in the Western world

(Mattson et al., 2011). In humans, prenatal EtOH exposure has been shown to initiate a broad spectrum of behavioural anomalies in children including decreased IQ, motor coordination deficits, hyperactivity and deficits in executive function, verbal language, memory and attention

(Mattson et al., 2011). The mechanisms underlying these deficits are unclear, however reactive oxygen species (ROS) have been implicated (Brocardo et al., 2012; Miller et al., 2013b). ROS formation produced by environmental chemicals or xenobiotics, such as EtOH, can result in oxidative stress, which can potentially dysregulate otherwise normal development by altering signal transduction pathways or oxidatively damaging cellular macromolecules including proteins, lipids, and DNA (Brocardo et al., 2011; Dong et al., 2010; Wells et al., 2009b). The mechanisms by which EtOH enhance ROS formation are not fully known, but likely involves at least in part its activation of ROS-generating NADPH oxidase (NOX) enzymes (Dong et al.,

2010), and possibly via free radical formation during its metabolism by cytochromes P450

(Koop, 2006).

Oxoguanine glycosylase 1 (OGG1) is a critical DNA repair enzyme involved in the excision and repair of the 8-oxo-2-deoxyguanosine (8-oxodGuo) lesion (Boiteux and Radicella,

338

2000), a biomarker of oxidative stress as well as a developmentally pathogenic lesion

(McCallum et al., 2011c; Wells et al., 2009b; Wong et al., 2008). OGG1 may be particularly important in the developing brain, where its activity is double that in maternal tissues (Wong et al., 2008). The progeny of CD-1 mice exposed in utero to EtOH exhibited an increase in oxidatively damaged DNA (8-oxodGuo) in fetal brain DNA as well as postnatal learning and memory deficits (Miller et al., 2013b). Similarly, in utero exposure of CD-1 mice to the ROS- initiating drug methamphetamine exhibit increased fetal brain levels of 8-oxodGuo as well as postnatal motor coordination deficits (Jeng et al., 2005). The pathogenic role of 8-oxodGuo in methamphetamine-initiated neurodevelopmental deficits was revealed in studies using knockout mice lacking ogg1 (Wong et al., 2008) or csb (McCallum et al., 2011c) which also repairs the 8- oxodGuo lesion. In these studies, knockout progeny exposed in utero to methamphetamine exhibited an ogg1 or csb gene dose-dependent increase in 8-oxodGuo in fetal brain of +/- and -/- pups compared to wild-type littermates, and increased motor coordination deficits measured by rotarod testing. ROS-initiating teratogens in addition to methamphetamine, including phenytoin and thalidomide, similarly increase 8-oxodGuo formation in the developing embryo and fetus

(Abramov and Wells, 2011a; Lee et al., 2011; Parman et al., 1999; Winn and Wells,

1995).Taken together, these data suggest that 8-oxodGuo is a pathogenic molecular lesion in the embryo, and ogg1 variability is a determinant of risk for behavioural deficits caused by oxidative stress. However, the roles of oxidatively damaged DNA and ogg1 in EtOH-initiated behavioural deficits are unknown.

Phenylbutylnitrone (PBN), a free radical spin trapping agent that traps free radicals thereby minimizing the damaging effect of excess ROS (Janzen et al., 1985), has been shown to protect against ROS-initiated oxidative damage to cellular macromolecules as well as the

339 associated birth defects caused by EtOH (Miller et al., 2013b), phenytoin (Liu and Wells, 1994;

Wells et al., 1989) and thalidomide (Lee et al., 2011; Parman et al., 1999).

Herein, to determine the pathogenic contribution of ROS-mediated formation of the 8- oxodGuo lesion, and the protective role of fetal DNA repair (Fig. 7.1), in the mechanism of

EtOH-initiated behavioural deficits, ogg1 wild-type (+/+) and heterozygous (+/-) and homozygous (-/-) knockout progeny produced as littermates from the breeding of +/- ogg1 dams and male breeders were exposed in utero on gestational day (GD) 17 to a single maternal administration of EtOH with or without PBN pretreatment. Oxidatively damaged DNA was measured in fetal brains 6 hours after EtOH exposure for some litters, and for other litters cognition was assessed at postnatal weeks 6, 9 and 12 using a passive avoidance learning test.

Our results in untreated mice provide the first demonstration of a behavioural phenotype for the ogg1 knockout mouse. In EtOH-exposed mice, deficient embryonic DNA repair and enhanced

8-oxodGuo contributed to the mechanism of EtOH-initiated postnatal behavioural deficits, which were blocked by pretreatment with PBN, corroborating a ROS-dependent mechanism of DNA oxidation. These results provide the most direct evidence to date of a pathogenic role for 8- oxodGuo in the mechanism of EtOH-initiated behavioural deficits, and the activity of OGG1 activity in fetal brain as a determinant of risk.

340

341

Figure 7.1: Postulated involvement of reactive oxygen species (ROS) and oxidatively damaged cellular macromolecules, and particularly DNA, which is repaired by oxoguanine glycosylase 1

(OGG1), in the mechanism of postnatal behavioural deficits initiated by in utero exposure to ethanol (EtOH). The free radical spin trapping agent phenylbutylnitrone (PBN) can detoxify drug and ROS free radical intermediates.

342

7.3 Methods

Chemicals. Alpha-phenyl-N-tert-butylnitrone (phenylbutylnitrone, PBN), minimum 98% purity by gas chromatography (GC), and all other reagents unless otherwise specified were purchased from Sigma Aldrich (St. Louis, MO). Saline (0.9 %, sterile) was purchased from Baxter

Corporation (Mississauga, ON) and ethanol from Commercial Alcohol Inc. (Brampton, ON).

Animals and diet. All animal protocols were approved by the institutional animal care committee in conformance with the guidelines established by the Canadian Council on Animal

Care. Ogg1 knockout mice on a 129SV/C57BL/6J background strain were originally generated by Klungland and coworkers (Klungland et al., 1999), and generously provided by Dr. Tomas

Lindahl (Imperial Cancer Research Fund, UK) through Dr. Christi A. Walter at the University of

Texas Health Science Center at San Antonio. Mice were housed in vented plastic cages from

Allentown, Inc. (Allentown, NJ) with ground corncob bedding (Bed-O’Cobs Laboratory Animal

Bedding, The Andersons Industrial Products Group, Maumee, OH). Mouse cages were maintained in a room with controlled light (14 hr light-10 hr dark cycle) and climate (20ºC, 50% humidity), and provided with rodent chow (Harlan Labs: 2018, Harlan Teklad, Montreal, QC) and tap water ad libitum. Mice were acclimatized for 1 week prior to use. Virgin +/- ogg1 females were mated with a +/- male overnight from 5:00 P.M. to 9:00 A.M. by housing one male in a cage containing 1-3 females. Heterozygous matings were employed to generate progeny of all 3 genotypes (+/+, +/-. -/-) within the same litter. The presence of a vaginal plug the next morning was designated as gestational day (GD) 1, and plugged females were separated, weighed and housed in groups of up to four dams per cage until use.

343

Genotyping. DNA was isolated from tail snips of fetal progeny by heating the sample at 95°C in

75 µL alkaline lysis reagent (25 mM NaOH, 0.2 mM disodium EDTA, pH 12) for 1 hr. Samples were then neutralized by addition of 75 µL of neutralizing buffer (40 mM Tris-HCl, pH 5), and genotyped using a PCR-based assay. Primers (synthesized by The Center for Applied Genomics,

The Hospital for Sick Children, Toronto, ON) used to amplify the 500 base pair (bp) band for the ogg1 gene were: ogg1-sense (5’-ACTGCATCTGCTTAATGGCC-3’) (forward primer) and ogg1-antisense (5’-CGAAGGTCAGCACTGAACAG-3’) (reverse primer). Primers used to amplify the 300 bp band for the neo-cassette responsible for disruption of the ogg1 gene in the ogg1 knockout mice were neo-sense (5’-CTGAATGAACTGCAGGACGA-3’) (forward primer) and neo-antisense (5’-CTCTTCGTCCAGATCATCCT-3’) (reverse primer). PCR reaction conditions were: 8 μl genomic DNA, 5 μl per sample of 10X Hotstart Buffer (Fermentas Life

Sciences, Burlington, ON), 3 μl per sample of MgCl2 (Fermentas Life Sciences), 0.25 μl Maxima

Hotstart Taq polymerase (Fermentas Life Sciences), 1 μl per sample of 10 mM deoxyribonucleotides (dNTP) (Fermentas Life Sciences), 0.75 μL per sample of each 20 μM primer and 29.75 μl per sample of ddH2O for a final volume of 50 μl. Cycling conditions were:

95°C for 5 min; 35 cycles of: 94°C for 1 min, 55°C for 1.5 min, 72°C for 2 min and completed with a final extension at 72°C for 10 min and then placed on hold at 4°C. 6X DNA loading dye

(0.03% bromophenol blue/0.03% xylene cyanol FF) was added to each sample. The PCR products were separated on a gel consisting of 1.5% (w/v) agarose, 40 mM Tris, 19.4 mM glacial acetic acid, 2.5 mM EDTA and 8 μl ethidium bromide. The agarose gel was run at 100 V for 60 min, and then visualized and photographed under an ultraviolet light (Fig. 7.2).

344

345

Figure 7.2: Representative gel for genotyping showing DNA bands for oxoguanine glycosylase

1 (OGG1) wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout mice.

Primers targeting the OGG1 gene amplify a 500 bp band present and intact in the +/+ and +/- mice, which is disrupted by the insertion of a neo cassette in the -/- mice. Primers targeting the neo cassette amplify a 300 bp band present in the +/- and -/- mice that is not present in the +/+ mice.

346

Dosing for measurement of DNA oxidation and postnatal behavioural testing. On GD 17

(within the fetal period), pregnant +/- ogg1 dams were pretreated with PBN (40 mg/kg i.p.) or its

0.79% saline vehicle control at 9:00 AM using a 26 gauge (G) 3/8 needle. Thirty min later, dams were treated with a 25% v/v solution of EtOH (2 g/kg i.p.) or its 0.9% saline vehicle. Food was removed from the cages for 6 hr post injection to avoid potential effects on EtOH pharmacokinetics. One group of pregnant dams was sacrificed 6 hr after EtOH exposure, fetal brains were removed, snap frozen and stored at -80C until analysis for DNA oxidation. A second group was allowed to deliver spontaneously for postnatal behavioural testing of their

EtOH-exposed offspring.

DNA extraction and digestion. DNA was extracted from GD 17 fetal brains using a modified version of a chaotropic sodium iodide (NaI) method previously described (Ravanat et al., 2002).

Briefly, samples were homogenized in 1 mL of cold lysis buffer (320 mM sucrose, 5 mM MgCl2,

10mM Tris, 0.1 mM desferoxamine, 1% Triton X-100, pH 7.5) and centrifuged twice at 10,000

X g for 10 min at 4°C to isolate the nuclear pellet containing DNA. The pellet was then incubated with 200 uL enzyme reaction solution (1% w/v SDS, 5 mM EDTA-NA2, 0.15 mM desferoxamine 10 mM Tris-HCl pH 8.0) for 1 hr at 50ºC with RNase A/T1 mix (624 and 312

U/ml final activities, respectively). Proteinase K (1.8 mg/ml final concentration) was added for an additional 1 hr. The DNA pellet obtained after extraction was then washed 5 times with 70%

EtOH. Samples were resuspended in 200 μl of sodium acetate buffer (20 mM, pH 4.8) and sonicated into solution. DNA purity was determined by measuring the absorbance ratio at

260/280 nm of 2 μL of sample in Na-acetate buffer (total volume 200 μL). Samples were digested with nuclease P1 (5 U/sample, 1 hr, 37°C) and calf intestinal alkaline phosphatase (6

347

U/sample, 1 hr, 37°C) and filtered through Amicon UltraTM filter units (YM-10, 10,000 MW cutoff; Millipore, Billerica, MA, USA) to remove DNA digestion enzymes and large particulates.

Quantification of oxidative damage to DNA using 8-OHdG ELISA. Oxidatively damaged

DNA was quantified using the 8-OHdG ELISA kit (JaICA, Fukuroi, Japan) according to the manufacturer’s instructions.

Passive avoidance test. At 6, 9, 12 and 16 weeks of age, ogg1 wild-type, and +/- and -/- knockout offspring were tested using a passive avoidance test as previously described (Miller et al., 2013b). Briefly, the test apparatus consisted of a plastic box with two equal sized chambers, one with clear plastic walls designated as the ‘light chamber’, and another with red walls designated as the ‘dark chamber’, separated by a guillotine door. The floors of both chambers consisted of stainless steel rods, wherein the rods of only the dark chamber were connected to a power supply that provided a mild 1.0 milliampere (mA) shock.

In trial #1, the mouse was placed in the light chamber and was allowed to explore for 20 sec. The guillotine door was lifted, and once the mouse entered the dark chamber, the guillotine door was closed, and a shock of 1.0 mA was administered for 4 sec. The mouse was then returned back to its home cage, the apparatus was wiped with 70% EtOH to remove any olfactory cues, and the same procedure was repeated 24 and 48 hr later for trials #2 and #3, respectively. The time it took the mouse to enter the dark chamber in both trials after opening the guillotine door was recorded as “latency to enter the dark chamber”. To examine the effect of treatment and ogg1 genotype on learning, the latency to enter the dark chamber was plotted for

348 trial 3 only. After trial #3, the pups were weighed at 6, 9 and 12 weeks to assess whether or not in utero EtOH exposure during the fetal period had any effect on growth.

Statistical analysis. Statistical analysis was performed using GraphPad Prism, Version 5

(GraphPad Software, Inc., San Diego, CA). Data were analyzed using a 1- or 2-way analysis of variance (ANOVA) with a post-hoc Bonferroni test. The minimal significance level used throughout was p < 0.05.

349

7.4 Results

DNA oxidation in fetal brains

EtOH-exposed +/+, +/- and -/- ogg1 progeny exhibited 2.4-fold (p < 0.001), 2.5 fold (p < 0.001) and 2.2-fold (p < 0.001) increases in oxidatively damaged DNA, measured as the 8-oxodGuo lesion, in fetal brain compared to their respective, genotypically matched saline-exposed littermates (Fig. 7.3). PBN pretreatment of saline-exposed fetuses did not change baseline DNA oxidation. Pretreatment with PBN decreased EtOH-enhanced DNA oxidation in fetal brains by

46% in +/+ (p < 0.01) and 23% in +/- (marginal, p 0.05 < p < 0.1) progeny, and similarly but not significantly in -/- littermates. EtOH-exposed -/- OGG1-deficient progeny had 25% higher DNA oxidation than EtOH-treated +/+ ogg1 progeny (p < 0.05). A similar pattern for increasing DNA oxidation in PBN-pretreated, EtOH-exposed fetal brains was observed in +/+, +/- and -/- ogg1 mice, with 36% higher DNA oxidation in +/- vs. +/+ mice (p < 0.05).

Passive avoidance test

Ogg1 genotype effect: Saline-exposed progeny

Saline-exposed +/+ ogg1 DNA repair-normal progeny had a greater latency time compared to +/- and -/- DNA repair-deficient progeny at 6 and 9 weeks (p < 0.001), while there were no differences observed at 12 weeks (Fig. 7.4).

350

100 * ††† 80  EtOH g DNA) †††  ††† 60  PBN+EtOH + SD

40 SALINE mean PBN 20

8-oxodGuo (fmol/ 8-oxodGuo 0 +/+ +/- -/- FETAL OGG1 GENOTYPE 150 SALINE †††* PBN EtOH g DNA) (6)  PBN + EtOH 100 ††† †††

+ SD (7) (5)  (7)  (6) (6) (6)

mean 50 (6)(5) (6)(6) (6)

8-oxodGuo (fmol/ 8-oxodGuo 0 +/+ +/- -/- FETAL OGG1 GENOTYPE

351

Figure 7.3: Oxidatively damaged DNA in fetal brains from oxoguanine glycosylase 1 (OGG1) wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout mice.

Pregnant ogg1 +/- mice bred with +/- males were treated on gestational day (GD) 17 with ethanol (2 g/kg i.p.) or saline vehicle (VH), with or without pretreatment with the free radical spin trapping agent phenylbutylnitrone (PBN) (40 mg/kg i.p.) and sacrificed 6 hr later. Litters containing embryos from all 3 genotypes were analyzed for 8-oxo-2'-deoxyguanosine (8- oxodGuo) formation using an ELISA kit with an 8-oxodGuo-specific antibody. Data were analyzed by one-way ANOVA with a post-hoc Bonferroni test. Single daggers indicate a difference from the vehicle control group for the same genotype (††† = p < 0.001). Asterisk indicates a difference from +/+ fetuses for the same treatment group (* = p < 0.05). Psi symbols indicate a difference from EtOH-treated fetuses for the same genotype (ψ = p < 0.05, ψψ = p <

0.01). Alpha symbol indicates a difference from EtOH-treated fetuses for the same genotype (α

= 0.05 < p < 0.1) The number of fetal brains analyzed is given in parentheses.

352

400 SALINE PBN EtOH PBN+EtOH 300 ** 200 *** # ** (seconds) +/+ ** 100 *** *** ** ** * DARK CHAMBER DARK +/- *** LATENCY TO ENTER -/- *** ** 0 ## *** 6 9 12 6 9 12 6 9 12 6 9 12 AGE (WEEKS) AGE (WEEKS) AGE (WEEKS) AGE (WEEKS)

PBN PBN+EtOH SALINE 400 (6) (17) (18)  EtOH (13) (13) (8) (10) (15) (13) **      (13) (24) (15) * 300 ** **  ***   *** ***  200 ** # ** *** *** *** ***

(seconds) ## 100 DARK CHAMBER

LATENCY TO ENTER TO LATENCY 0 6 9 12 6 9 12 6 9 12 AGE (WEEKS) 6 9 12 6 9 12 6 9 12 6 9 12 6 9 12 6 9 12 6 9 12 6 9 12 6 9 12 FETAL OGG1 GENOTYPE +/+ +/- -/- +/+ +/- -/- +/+ +/- -/- +/+ +/- -/-

353

Figure 7.4: Time course for postnatal learning development in oxoguanine glycosylase 1

(OGG1) wild-type (+/+), heterozygous (+/-) and knockout (-/-) progeny exposed in utero to

EtOH with or without PBN pretreatment.

(A graphical presentation of genotypical differences is provided in Fig. 7.5.) Data were analyzed by one-way ANOVA with a post-hoc Bonferroni test. Alpha symbols indicate a difference from week 6 of the same genotype (α = p < 0.05, αα = p < 0.01, ααα = p < 0.001). Beta symbols indicate a difference from week 9 of the same genotype (β = p < 0.05, βββ = p < 0.001).

Asterisks indicate a difference from +/+ progeny for the same time point (* = p < 0.05, ** = p <

0.01, *** = p < 0.001). Pound symbols indicate a difference from +/- progeny for the same time point (# = p < 0.05, ## = p < 0.01). The number of pups tested is given in parentheses.

354

Ogg1 genotype effect: PBN-exposed progeny

At 6 weeks, PBN-treated +/+ ogg1 progeny had a greater latency time compared to +/- (p < 0.01) and -/- (p < 0.001) progeny (Fig. 7.4). At 9 weeks, +/+ (p < 0.01) and +/- (p < 0.05) progeny had higher latency times compared to -/- littermates.

Ogg1 genotype effect: EtOH-exposed progeny

At 6 weeks, there were no differences in latency times among +/+, +/- and -/- progeny exposed in utero to EtOH (Fig. 7.4). At 9 weeks, EtOH-exposed +/+ progeny had a greater latency time compared to similarly exposed +/- (p < 0.01) and -/- (p < 0.001) littermates, and +/- progeny had a greater latency time than -/- littermates (p < 0.01). At 12 weeks, +/+ progeny had a greater latency time than -/- littermates (p < 0.01) exposed to EtOH.

Ogg1 genotype effect: PBN-pretreated EtOH-exposed progeny

At 6 weeks, there were no differences in latency times among +/+, +/- and -/- progeny exposed in utero to PBN followed by EtOH (Fig. 7.4). At 9 weeks, +/- progeny had greater latency time compared to similarly exposed +/- (p < 0.01) and -/- littermates (p < 0.001). At 12 weeks, +/+ had greater latency than -/- progeny (p < 0.05) exposed to PBN followed by EtOH.

Treatment effect

Saline-exposed +/+, +/- and -/- progeny had greater latency times compared to their EtOH- treated +/+ (p < 0.001), +/- (p < 0.001) and -/- (p < 0.001) littermates, respectively (Fig 7.5).

Similarly, PBN-pretreated EtOH-exposed +/+, +/- and -/- progeny had greater latency times compared to their EtOH-treated +/+ (p < 0.05), +/- (p < 0.001) and -/- (p < 0.01) littermates, respectively. There were no differences in latency time among saline-exposed and PBN-exposed

355

400

300 PBN SALINE 200 * *** PBN+EtOH (seconds) ††† ††† ** DARK CHAMBER DARK 100 EtOH

LATENCY ENTER TO †††

0 +/+ +/- -/- AGE (WEEKS)

400 SALINE (17) (13)* (15)(17) ***(13) PBN (10) (6) 300 (13) EtOH ††† **(8) PBN+EtOH (13) ††† (24) ††† 200 (15) (seconds) 100 DARK CHAMBER LATENCY TO ENTER TO LATENCY 0 +/+ +/- -/- FETAL OGG1 GENOTYPE

356

Figure 7.5: Effect of PBN pretreatment on EtOH-initiated learning deficits in oxoguanine glycosylase 1 (OGG1) wild-type (+/+), and heterozygous (+/-) and homozygous (-/-) knockout progeny.

Pregnant mice were treated as described in fig. 7.3, and the progeny were tested postnatally for a passive avoidance test at weeks 6, 9 and 12. Data are shown for trial #3 at 12 weeks. Data in the bottom panel are a different representation of the same results shown in the top panels. Data were analyzed by one-way ANOVA with a post-hoc Bonferroni test. Daggers indicate a difference from saline for the same genotype (††† = p < 0.001). Asterisks indicate a difference from the

PBN+EtOH group for the same genotype (* = p < 0.05, ** = p < 0.01, *** = p < 0.001).

357 progeny for any of the genotypes.

Fetal Weight

There were no treatment- or genotype-dependent differences in fetal weight on GD 17, nor in the postnatal weight of progeny at any time point, suggesting that the dose of EtOH used was minimally toxic (Fig 7.6).

358

3 WEEKS

(18) (15) 15 (19) (6) (13) (22) (16) (5) (13) (13) (21) (17) (22) (30) (15) (24) (13)(16) (10) (12) 10 WEIGHT

(g, n + SD) (g, n + 5 FETAL BODY FETAL

0 6 WEEKS

30 *** *** *** ***

WEIGHT 10 (g, n + SD) (g, n + FETAL BODY FETAL

0 9 WEEKS 40

** ** 30 ** **

20 WEIGHT

(g,SD) n + 10 FETAL BODY FETAL

0 12 WEEKS

40 ** ** ** 30 **

20 WEIGHT

(g, n + SD) (g, n + 10 FETAL BODY FETAL

0 Veh PBN Veh PBN Veh PBN Veh PBN Veh PBN Veh PBN Veh PBN Veh PBN Veh PBN Veh PBN TREATMENT ______EtOH ______EtOH ______EtOH ______EtOH ______EtOH WT HET KO GENDER FEMALE MALE

359

Figure 7.6: Fetal weight in oxoguanine glycosylase 1 (OGG1) wild-type (+/+), and heterozygous

(+/-) and homozygous (-/-) knockout mice.

Pregnant ogg1 +/- mice bred with +/- males were treated on gestational day (GD) 17 with ethanol (2 g/kg i.p.) or saline vehicle (VH), with or without pretreatment with the free radical spin trapping agent phenylbutylnitrone (PBN) (40 mg/kg i.p.), the progeny were weighed postnatally at weeks 3, 6, 9 and 12. Data were analyzed by one-way ANOVA with a post-hoc

Bonferroni test. Asterisks indicate a difference from females for the same genotype and treatment group. The number of fetuses is given in parentheses.

360

7.5 Discussion

EtOH exposure during pregnancy causes a spectrum of behavioural deficits affecting both learning and memory in humans and in animal models (Brocardo et al., 2012; Jones, 2011). Our study in ogg1 DNA repair-deficient knockout mice demonstrates for the first time: (1) a behavioural phenotype for these mice in the absence of drug exposure; (2) that a single prenatal

EtOH exposure on GD 17 during fetal development increases DNA oxidation measured as the 8- oxodGuo lesion in fetal brain; and, (3) this single EtOH exposure decreases postnatal performance in the passive avoidance test at a minimally toxic dose that does not affect postnatal body weight. These results, reflecting a deficit in learning and memory, reveal the exquisite susceptibility of developing brain function in the fetal period to impairment by in utero exposure to what we previously have shown is a relatively non-teratogenic dose of EtOH with respect to structural birth defects (Miller et al., 2013b), consistent with the absence of any effect of EtOH in progeny body weight observed herein. Perhaps most noteworthy, our results provide the most direct evidence to date of a causal role for oxidatively damaged DNA in the fetal brain, and protective role for DNA repair, in the mechanism of EtOH-initiated neurodevelopmental deficits.

While the 8-oxodGuo lesion is a biomarker for oxidative stress, several studies suggest that it is also a pathogenic lesion, likely via altered transcription as opposed to increased mutagenesis

(Wells et al., 2010). As observed herein for EtOH, whereby fetal brain DNA oxidation and learning deficits were increased in an ogg1gene dose-dependent fashion, CD-1 progeny exposed in utero to EtOH exhibited enhanced fetal brain DNA oxidation and learning deficits (Miller et al., 2013b), ogg1 knockout progeny exposed in utero to methamphetamine exhibited enhanced motor coordination deficits (Wong et al., 2008), and knockout mice deficient in Cockayne

361

Syndrome B (CSB), another DNA repair protein also exhibited enhanced motor coordination deficits compared to their DNA repair-normal wild-type littermates (McCallum et al., 2011c).

Several potential mechanisms exist whereby in utero formation of the 8-oxodGuo lesion in fetal brain may initiate postnatal behavioural deficits, including altering the expression and activity of proteins required for normal embryonic development, alteration of gene transcription or expression via its ability to regulate binding affinity of various transcription factors including nuclear factor kappa B (NF-κB) to specific promoter elements, and/or apoptosis resulting from

8-oxodGuo accumulation, none of which are mutually exclusive (Wells et al., 2009b). Likely downstream consequences include altered cellular division, differentiation, migration, function and intercellular communication in the brain.

Fetal weights were not affected by EtOH exposure, indicating that the behavioural deficit observed was not secondary to a structural growth deficit, and that the dose used was minimally fetotoxic. Saline-exposed +/+ progeny exhibited a consistently better learning and memory performance than +/- and -/- ogg1 knockout littermates in an ogg1 gene dose-dependent fashion.

This neurodevelopmental deficit in untreated animals constitutes the first demonstration of a phenotype for ogg1 DNA repair-deficient mice, which somewhat surprisingly do not exhibit an increase in cancer (Klungland et al., 1999), suggesting that the fundamental importance of this gene lies in development rather than providing lifelong protection against cancer. The OGG1- dependent enhanced susceptibility to neurodevelopmental deficits was similarly observed in

EtOH-exposed progeny, which exhibited a shift in the ogg1 gene dose-response curve, corroborating the hypothesis that EtOH-initiated behavioural deficits are initiated at least in part by ROS-mediated oxidatively damaged DNA in the developing fetal brain which, if not repaired by OGG1, can result in abnormal postnatal brain function.

362

As observed in our single-exposure threshold model herein, previous studies have observed similar postnatal neurodevelopmental deficits with chronic EtOH exposure throughout pregnancy in both rats (Abel, 1982; Mattson et al., 1993; Riley et al., 1979) and mice (Becker and Randall,

1989; Fiore et al., 2009; Gilliam et al., 1987), whereby EtOH-exposed progeny required more trials to learn the passive avoidance task than control progeny.

The protective effect of the free radical spin trapping agent PBN in blocking EtOH-initiated

8-oxodGuo formation in +/+ ogg1 fetal brains and postnatal behavioural deficits in +/+, +/- and -

/- ogg1 progeny suggests that ROS are involved in the mechanism of EtOH-initiated DNA oxidation and neurodevelopmental deficits, and that PBN is protecting via a mechanism proximal to DNA repair. PBN similarly protects against birth defects initiated by other ROS-initiating teratogens including thalidomide (Lee et al., 2011; Parman et al., 1999) and phenytoin (Liu and

Wells, 1994; Wells et al., 1989), and reduce the associated oxidative damage to cellular macromolecules including DNA, protein and lipids (Wells et al., 2009b).

The increased susceptibility of ogg1 DNA repair-deficient knockout mice to fetal DNA oxidation and postnatal behavioral deficits caused by in utero EtOH exposure indicates both the pathogenic importance of the 8-oxodGuo lesion, and the protective importance of fetal DNA repair enzymes in preventing behavioral abnormalities. If alcohol is consumed by a pregnant woman whose child has a DNA repair variant, inherited from either the mother or father, that renders OGG1 inactive and/or deficient, her child could be at increased risk of developing

FASD. Recent studies have associated the Ser326Cys polymorphism in the human OGG1 gene with delayed enzyme function (Kershaw and Hodges, 2012), lung cancer in Chinese population

(Kirkali et al., 2009) and non-small cell lung cancer (Janik et al., 2011).

363

In summary, OGG1-deficient fetuses exposed once in utero to only saline vehicle exhibit enhanced DNA oxidation and postnatal cognitive deficits, revealing the first evidence of a phenotype for these knockout mice. DNA oxidation and postnatal neurodevelopmental deficits are further increased by in utero exposure of ogg1 knockout fetuses to a single dose of EtOH that does not affect body weight. These results reveal the exquisite sensitivity of the developing brain to both endogenous and EtOH-enhanced oxidative stress. The protection afforded by pretreatment with the free radical spin trapping agent PBN suggests that both the macromolecular damage and consequential cognitive deficits are ROS-mediated. The ogg1 gene dose-dependent exacerbation of EtOH-initiated 8-oxodGuo formation in fetal brain and postnatal cognitive deficits among littermates with deficient DNA repair provides the most direct evidence to date that: (1) the 8-oxodGuo lesion plays a pathogenic role in the mechanism of postnatal neurodevelopmental deficits initiated by in utero fetal exposure to endogenous oxidative stress or

EtOH; and, (2) OGG1 activity in the fetal brain constitutes a determinant of risk.

364

Chapter 8 The free radical spin trapping agent phenylbutylnitrone reduces fetal brain DNA oxidation and postnatal cognitive deficits by in utero exposure to a non- structurally teratogenic dose of ethanol: a role for oxidative stress

Running title: Fetal DNA oxidation & ethanol behavioural deficits

Lutfiya Miller†, Aaron M. Shapiro*, Jun Cheng† and Peter G. Wells*†

* Division of Biomolecular Sciences, Faculty of Pharmacy, and

†Department of Pharmacology and Toxicology, Faculty of Medicine

University of Toronto

Toronto, Ontario, Canada

a. Preliminary reports of this research were presented at the 2009 and 2010 annual meetings of the Teratology Society [Birth Defects Research Part A: Clinical and Molecular Teratology 85(5): 399 (Abstract No. 4); Birth Defects Research Part A: Clinical and Molecular Teratology 88(5): 354 (Abstract No.134] and at the 2011 annual meeting of the Society of Toxicology of Canada [Proceedings of the 43rd Annual Meeting of the Society of Toxicology]. These studies were supported by a grant from the Canadian Institutes of Health Research. b. Full report of this research has been published: Miller, L., Shapiro, A. M., Cheng, J. and Wells, P. G. (2013). The free radical spin trapping agent phenylbutylnitrone reduces fetal brain DNA oxidation and postnatal cognitive deficits caused by in utero exposure to a non-structurally teratogenic dose of ethanol: A role for oxidative stress. Free Radical Biology & Medicine 60, 223-32. c. Individual contributions: Lutfiya Miller- mating, dosing and sample collection, behavioural testing; Aaron M. Shapiro-DNA oxidation measurements; Jun Cheng- behavioural testing, trained and supervised by LM; Peter G. Wells- supervisor.

365

8.1 Abstract

Reactive oxygen species (ROS), while implicated in morphological birth defects caused by ethanol (EtOH) during pregnancy, have not been directly linked to its behavioural deficits.

To determine the latter, a pathogenic oxidative DNA lesion was measured in fetal brain, and a passive avoidance learning test was assessed postnatally in the progeny of CD-1 mice treated once on gestational day 17 with 4 g/kg of EtOH or its saline vehicle, with or without pretreatment with the free radical spin trapping agent alpha-phenyl-N-tert-butylnitrone (PBN)

(40 mg/kg). EtOH-exposed CD-1 progeny, unlike C57BL/6 progeny, had no morphological birth defects, but exhibited a learning deficit at 12 weeks of age (p < 0.001), which continued to

16 weeks in males (p < 0.01). Peak blood EtOH concentrations were 2.5-fold higher in C57BL/6 mice compared to CD-1 mice given the same dose. PBN pretreatment of CD-1 dams blocked both EtOH-initiated DNA oxidation in fetal brain (p < 0.05) and postnatal learning deficits (p <

0.01), providing the first direct evidence for ROS in the mechanism of EtOH-initiated neurodevelopmental deficits.

366

8.2 Introduction

In utero embryonic and fetal exposure to ethanol (EtOH) can result in a spectrum of morphological and behavioral or neurodevelopmental anomalies, collectively termed Fetal

Alcohol Spectrum Disorders (FASD). This spectrum can include Alcohol Related Birth Defects

(ARBD), Alcohol Related Neurological Disorders (ARND) and Fetal Alcohol Syndrome (FAS).

FAS, representing the complete phenotype is characterized by characteristic craniofacial dysmorphology (short palpebral fissures, flat midface, smooth or flat philtrum, thin upper lip), growth retardation, and neurodevelopmental deficits (Jones, 2011). The incidence of FASD is reported as high as 0.2-2 per 1000 live births (Riley and McGee, 2005), and is thought to be one of the leading preventable causes of neurodevelopmental disorders in the Western world

(Mattson et al., 2011) Children exposed to EtOH prenatally exhibit a broad spectrum of behavioural anomalies including motor function deficits, hyperactivity, decreased IQ, and deficits in executive function, verbal language, memory and attention (Mattson et al., 2011).

Studies using magnetic resonance imaging (MRI) of the brains of children with FASD have shown volume reductions in brain regions such as the hippocampus (Coles et al., 2011), basal ganglia (Mattson et al., 1996), cerebellum (Sowell et al., 1996) and corpus callosum (Riley et al., 1995), and animal models exhibit similar reductions in the hippocampus (Wigal and

Amsel, 1990) and cerebellum (Heaton et al., 2002).

Prenatal EtOH exposure causes morphological birth defects (structural teratogenesis) in the developing fetus in both rats (Lee et al., 2005; Wentzel and Eriksson, 2008) and mice (Dong et al., 2010; Wentzel and Eriksson, 2006) in vivo and in embryo culture, as well as initiating postnatal behavioral dysfunction (functional teratogenesis) in mice (Becker and Randall, 1989;

367

Gilliam et al., 1987) and rats(Brocardo et al., 2012; Mattson et al., 1993; Mooney and

Varlinskaya, 2011).

Some of the mechanisms implicated in EtOH teratogenicity include fetal hypoxia, disrupted cell proliferation and apoptosis, altered membrane fluidity and oxidative stress

(Brocardo et al., 2011; Henderson et al., 1999). Oxidative stress results from an imbalance between reactive oxygen species (ROS) formation and removal by antioxidants leading to elevated ROS levels, causing altered signal transduction and oxidative damage to cellular macromolecules (Halliwell and Gutteridge, 2007), which have been implicated in structural and functional teratogenesis (Wells et al., 2009b).

ROS-initiated oxidative stress caused by EtOH has been implicated in the mechanism of its structural teratogenicity, although ROS involvement in its functional, behavioral teratogenicity has not been directly determined. Evidence in structural teratogenesis includes: decreased forelimb malformations in EtOH-treated embryos pretreated with the SOD/catalase mimetic EUK-134 (Chen et al., 2004a); elevation in fetal hepatic isoprostane formation in superoxide dismutase (SOD) knockout mice exposed to EtOH in utero (Wentzel and Eriksson,

2006); a protective effect of 3H-1,2 dithiole-3-thoine (D3T), a nuclear factor erythroid 2-related factor 2 (Nrf2) inducer, against EtOH-initiated ROS formation in mouse embryos exposed in utero to EtOH (Dong et al., 2008); increased lipid peroxidation in fetal liver, brain, kidneys and testes (Nayanatara et al., 2009); and a protective effect of diphenyleneiodonium (DPI), a

NADPH oxidase (NOX) inhibitor, against EtOH-initiated ROS formation and oxidative DNA damage in mouse embryos (Dong et al., 2010). Further evidence supporting the involvement of oxidative stress in structural components of the FASD is reviewed in detail elsewhere (Brocardo et al., 2011).

368

The involvement of ROS in EtOH-initiated behavioural deficits is less clear, although the low antioxidative potential of the brain along with its high consumption of oxygen make it a susceptible site for oxidative stress (Floyd, 1999; Halliwell and Gutteridge, 2007). Various conditions that enhance oxidative stress in the fetal brain (Henderson et al., 1999) are correlated with EtOH-initiated behavioural deficits (Vink et al., 2005), which can be ameliorated by antioxidant pretreatments (Busby et al., 2002; Neese et al., 2004; Reid et al., 1999; Vink et al.,

2005), but direct measurements of ROS and/or oxidation of lipid, protein or DNA in fetal brain are lacking in these observational studies.

The free radical spin trapping agent alpha-phenyl-N-tertiary-butylnitrone (PBN) traps both EtOH-derived carbon-centred lipid free radicals (Reinke et al., 1987) and hydroxyethyl radicals comprising more than 80% of ethanol-derived radicals following reactions with hydroxyl radical, which can lead to lipid peroxidation both in vitro and in vivo (Reinke, 2002).

PBN pretreatment also prevents structural embryopathies and teratogenesis, and oxidative macromolecular damage, in embryos and fetuses exposed in culture and/or in utero to other

ROS-initiating teratogens such as thalidomide (Lee et al., 2011; Parman et al., 1999) and phenytoin (Liu and Wells, 1994; Wells et al., 1989). PBN has neuroprotective and antiaging properties (Carney and Floyd, 1991; Halliwell and Gutteridge, 2007) and improves learning in lipopolysaccharide (LPS)-treated neonatal rats (Fan et al., 2008). However, the effect of PBN on

EtOH-initiated ROS-mediated neurodevelopmental deficits has not been investigated.

Herein, to determine the role of ROS in the mechanism of EtOH-initiated neurodevelopmental behavioural deficits, the progeny of CD-1 mice were exposed in utero on gestational day (GD) 17 to a single maternal administration of a minimally teratogenic dose of

EtOH for the CD-1 strain, with or without PBN pretreatment, and their cognition was assessed

369 from postnatal weeks 6 to 16 weeks using a passive avoidance learning test. Oxidatively damaged DNA was measured in fetal brains 6 hours after EtOH exposure to determine both the involvement of ROS and the potentially pathogenic role of 8-oxo-2’-deoxyguanosine (8- oxodGuo), a DNA lesion that has been implicated in neurodevelopmental deficits caused by the

ROS-initiating drug methamphetamine (McCallum et al., 2011c; Wong et al., 2008). Our results provide the most direct evidence to date of a role for ROS and 8-oxodGuo in the mechanism of

EtOH neurodevelopmental deficits, with both the macromolecular lesion in fetal brain and the postnatal functional deficits being blocked by pretreatment with PBN. To compare the relative risks of functional (behavioural) and structural teratogenesis, other pregnant CD-1 mice were treated earlier in development, on GD 9, during organogenesis when the developing embryo is highly susceptible to structural teratogenesis (Chen et al., 2004a; Wells et al., 2009b), and our results show that ROS-mediated neurodevelopmental deficits result from a single exposure to a minimally teratogenic dose of EtOH that does not cause structural malformations in this strain.

This absence of structural malformations in CD-1 mice was contrasted with the C57BL/6 strain, which is known to be susceptible to EtOH-initiated malformations (Boehm et al., 1997), and we also discovered a substantial reduction in EtOH plasma concentrations in CD-1 mice compared to the C57BL/6 strain, revealing a novel mechanism for the strain-dependent shift in the EtOH dose-response relationship for structural teratogenesis.

370

8.3 Methods

Chemicals. Alpha-phenyl-N-tert-butylnitrone (phenylbutylnitrone, PBN) minimum 98% purity by gas chromatography (GC), Bouin’s fixative, nuclease P1, deoxyguanosine (dG) standard and all other reagents unless otherwise specified were purchased from Sigma Aldrich (St. Louis,

MO). Proteinase K was purchased from Bioshop (Burlington, Ontario [ON]), saline (0.9 %, sterile) from Baxter Corporation (Mississauga, ON) and ethanol from Commercial Alcohol Inc.

(Brampton, ON).

Animals and diet. All animal protocols were approved by the institutional animal care committee in conformance with the guidelines established by the Canadian Council on Animal

Care. CD-1 and C57BL/6 mice were obtained from Charles River Canada Ltd. (Saint-Constant,

Quebec [QC]). Mice were housed in vented plastic cages from Allentown, Inc. (Allentown, NJ) with ground corncob bedding (Bed-O’Cobs Laboratory Animal Bedding, The Andersons

Industrial Products Group, Maumee, OH). Mouse cages were maintained in a room with controlled light (14 hr light-10 hr dark cycle) and climate (20ºC, 50% humidity), and provided with rodent chow (Harlan Labs: 2018, Harlan Teklad, Montreal, QC) and tap water ad libitum.

Mice were acclimatized for 1 week prior to use. Virgin female mice were mated with males overnight from 5:00 PM to 9:00 AM by housing one male in a cage containing 1-3 females. The presence of a vaginal plug the next morning was designated as GD 1 and plugged females were separated, weighed and housed in groups of up to four per cage until use.

371

Dosing for behavioural testing. On GD 17 within the fetal period, CD-1 dams were pretreated with PBN (40 mg/kg i.p.) or its 0.79% saline vehicle control at 9:00 AM using a 26 gauge (G)

3/8 needle. Thirty minutes later, dams were treated with a 25% v/v solution of EtOH (4 g/kg i.p.) or its 0.9% saline vehicle. Food was removed from the cages for 6 hours post injection to avoid potential effects on EtOH pharmacokinetics.

Passive avoidance test. At 6, 9, 12 and 16 weeks of age, offspring were tested using a passive avoidance test. The test apparatus consisted of a plastic box with two equal sized chambers, one with clear plastic walls designated as the ‘light chamber’, and another with red walls designated as the ‘dark chamber’, separated by a guillotine door. The floors of both chambers consisted of stainless steel rods. The rods of the dark chamber were connected to a power supply which provided a 1.0 milliampere (mA) shock. In order to ensure the current ran through the mouse and provided a shock without producing a short circuit, alternating rods were connected to the anode and cathode such that the mouse had to grasp onto the rods in order complete the circuit and receive the shock.

In trial #1, the mouse was placed in the light chamber and was allowed to explore for 20 seconds, after which time the guillotine door was lifted. Once the mouse entered the dark chamber, the guillotine door was closed, and a shock of 1.0 mA was administered for 4 seconds. The mouse was then returned back to its home cage. The apparatus was sprayed with 70% EtOH and wiped in between each mouse tested to eliminate any olfactory cues. The same procedure was repeated

24 hours later for trial #2. The time it took the mouse to enter the dark chamber in both trials after opening the guillotine door was recorded as ‘latency to enter the dark chamber’. To examine whether or not the mouse had learned the test to avoid the dark chamber, the time it

372 took the mouse to enter the dark chamber in trial #2 was compared to trial #1. If the animal did not enter the dark chamber within 5 minutes in trial #2, it was considered have learned the passive avoidance test. After the last trial, the pups were weighed at 6, 9, 12 and 16 weeks to assess whether or not in utero EtOH exposure during the fetal period had any effect on growth.

Dosing for DNA oxidation. On GD 17, a separate group of CD-1 dams were treated in the same manner as those for behavioural testing. Six hours after EtOH exposure, the dam was sacrificed, the uterus was exteriorized, fetal brains were removed from each of the pups, snap frozen in liquid nitrogen and stored at -80ºC until use. Samples were randomly selected from at least 3 separate litters for analysis from each treatment group.

DNA extraction and digestion. DNA was extracted from GD 17 fetal brains using a chaotropic sodium iodide (NaI) method previously described by Ravanat et al.(2002). Briefly, samples were homogenized in 1 mL of cold lysis buffer (320 mM sucrose, 5 mM MgCl2, 10mM Tris, 0.1 mM desferoxamine, 1% Triton X-100, pH 7.5) and centrifuged twice at 10,000 X g for 10 min at 4°C to isolate the nuclear pellet containing DNA. The pellet was then incubated with 200 uL enzyme reaction solution (1% w/v SDS, 5 mM EDTA-NA2, 0.15 mM desferoxamine 10 mM Tris-HCl pH 8.0) for 1 hr at 50ºC with RNase A/T1 mix (624 and 312 U/ml final activities, respectively).

Proteinase K (1.8 mg/ml final concentration) was added for an additional 1 hr. The DNA pellet obtained after extraction was then washed 5 times with 70% EtOH. Samples were resuspended in

200 μl of sodium acetate buffer (20 mM, pH 4.8) and sonicated into solution. DNA purity was determined by measuring the absorbance ratio at 260/280 nm of 2 μL of sample in Na-acetate buffer (total volume 200 μL). Samples were digested with nuclease P1 (5 U/sample, 1 hr, 37°C)

373 and calf intestinal alkaline phosphatase (6 U/sample, 1 hr, 37°C) and filtered through Amicon

UltraTM filter units (YM-10, 10,000 MW cutoff; Millipore, Billerica, MA, USA) to remove DNA digestion enzymes and large particulates. Quantification of 8-oxo-7,8-dihydro-2’- deoxyguanosine (8-oxodGuo) was standardized to endogenous deoxyguanosine (dGuo), which were analyzed using high-performance liquid chromatography (HPLC) coupled respectively to tandem mass spectrometric (MS/MS) and ultraviolet (UV) detectors.

Quantification of oxidative damage to DNA using HPLC-MS/MS. To determine the levels of

8-oxodGuo and dGuo in the same sample without dilution, dGuo was first analyzed using a UV detector set to wavelength λ=280 nm on a Perkin Elmer Series 200 HPLC system (Woodbridge,

ON) and a mobile phase consisting of phosphate buffer pH 5.2 and methanol (95:5) at an isocratic rate of 0.8 mL/min for 50 min. The same sample was then reanalyzed using an Agilent

1100 series HPLC (Mississauga, ON) with an AB Sciex API4000 QTRAP triple quadrapole mass spectrometer (Concord, ON) on a water:methanol gradient with a flow rate of 0.6 mL/min for 9.5 min under the control of AB Sciex Analyst® Software. The multiple reaction monitoring

Teratology. On GD 9, during the period of organogenesis, CD-1 dams were pretreated with a single injection of PBN (40 mg/kg i.p.) or its 0.79% saline vehicle control at 9:00 AM using a 26

374 gauge (G) 3/8 needle. Thirty minutes later, dams were treated with a 25% v/v solution of EtOH

(4 or 6 g/kg i.p.) or its 0.9% saline vehicle. To determine strain-dependent differences in the dose-response relationship for EtOH structural teratogenesis, C57BL/6 dams were treated with a

25% v/v solution of EtOH (4 or 6 g/kg i.p.). On GD 19, dams were sacrificed and fetuses were examined for growth parameters and external anomalies. Fetal carcasses were then placed in

Bouin’s fixative for 3 days, and subsequently transferred to 70% EtOH for at least one day. The gender of fetuses was determined by the presence or absence of testicles, and internal soft tissue anomalies were assessed using the Wilson’s freehand razorblade-sectioning technique (Barrow and Taylor, 1969).

Metabolism. Male CD-1 and C57BL/6 mice were treated i.p with EtOH (2.8 g/kg, 25% v/v solution in 0.9% saline) using a 26 gauge (G) 3/8 needle or saline control. Prior to blood sampling, mice were anaesthetized by placing a 15 mL conical tube containing isoﬂurane-soaked gauze over the snout of the animal until it was unresponsive. Blood samples were collected in heparinised vacutainers (lithium heparin 68 USP units per tube, Becton, Dickinson and

Company, Oakville, ON) by puncturing the portal vein. Blood samples were collected at 0.5, 1,

2, 4 and 8 hr post injection. Plasma was isolated from the samples by centrifugation at 1000×g for 15 min at 4 °C. Plasma samples were frozen at −20 °C until time of analysis. At each time point, plasma samples were collected from 3 to 6 mice. Plasma samples from saline controls and

EtOH-treated mice were analyzed for EtOH concentrations by headspace gas chromatography

(GC) based on previously published methods (Porter and Moyer, 1994). The Agilent GC system consisted of a 6890N GC, a G1888 headspace sampler, and a G1540N flame ionization detector

(FID) coupled with a Restek Rtx-200 capillary column (30 m, 0.53 mm ID, 3 μm).

375

Statistical analysis. Statistical analysis was performed using GraphPad Prism, Version 5

(GraphPad Software, Inc., San Diego, CA). Multivariate analysis was performed using a 1- or 2- way analysis of variance (ANOVA) with a post-hoc Bonferroni test. Binomial data were analyzed using a chi-square test. The minimal significance level used throughout was p < 0.05.

376

8.4 Results

Passive avoidance test. At 6 and 9 weeks, there were no differences in latency to enter the dark chamber for CD-1 male or female progeny for either trial #1 or trial #2.

At 12 weeks, female progeny exposed in utero to saline treatment learned the task, as evidenced by a higher latency to enter the dark chamber for trial #2 compared to trial #1 (p < 0.05), while there was no difference between trial 1 and 2 for the saline-exposed male progeny (Fig. 8.1).

Female progeny exposed in utero to PBN alone learned the task as evidenced by a higher latency in trial #2 compared to trial #1 (p < 0.01), while there was no difference for males. EtOH- exposed female progeny exhibited a reduced latency to enter the dark chamber compared to saline-exposed females at trial #2 (p < 0.01), and also compared to females exposed to PBN alone (p < 0.05). EtOH-exposed male progeny exhibited a reduced latency to enter the dark chamber compared to saline-exposed males (p < 0.05), and also compared to males exposed to

PBN alone (p < 0.05). PBN pretreatment blocked the EtOH-initiated deficits, evidenced by an increased latency, similar to saline controls, for both male and female progeny at 12 weeks in both trials #1 and #2, compared to EtOH-exposed progeny (p < 0.001).

At 16 weeks, there were no differences between trials #1 or #2 for the female or male progeny exposed in utero to saline, or PBN alone. There also was no difference between EtOH- and saline-exposed female progeny for either trials #1 or #2. In contrast, EtOH-exposed males exhibited a lower latency to enter the chamber at trial #1 compared to males exposed to either saline (p < 0.05) or PBN alone (p < 0.01), and this EtOH-initiated deficit was blocked by pretreatment with PBN.

377

6 WEEKS 400

300 FEMALES MALES

(18) 200 (27) (20) (20) (21) (19) (15) 100 (17)

0 9 WEEKS 400

300

200

100

0 12 WEEKS 400   ****** ** 300 **

200 † ‡‡ ††

(seconds, mean + SD) mean + (seconds, ‡ † 100 ‡

0 16 WEEKS 400 LATENCY TO ENTER CHAMBER DARK ** 300 † ‡‡ 200

100

0 TRIAL NO. ______1 2______1 2______1 2 ______1 2 ______1 2______1 2______1 2 ______1 2 TREATMENT SALINE PBN ______SALINE PBN SALINE PBN ______SALINE PBN 4g/kg EtOH 4g/kg EtOH

378

Figure 8.1: Effect of ethanol (EtOH) exposure during the fetal period on passive avoidance learning in CD-1 mice, and protection by phenylbutylnitrone (PBN).

Pregnant CD-1 mice were treated intraperitoneally (i.p.) on gestational day (GD) 17 with ethanol

(4 g/kg) or saline vehicle, with or without pretreatment with the free radical spin trapping agent

PBN (40 mg/kg i.p.), and the progeny were tested postnatally for a passive avoidance test at weeks 6, 9, 12 and 16. Data were analyzed by one-way ANOVA with a post-hoc Bonferroni test, and shown as the mean + standard deviation. Daggers indicate a difference from saline for the same trial († = p < 0.05, †† = p < 0.01, ††† = p < 0.001). Double daggers indicate a difference from PBN for the same trial (‡ = p < 0.05, ‡‡ = p < 0.01, ‡‡‡ = p < 0.001). Alpha symbols indicate a difference from trial 1 within the same treatment group (α = p < 0.05, αα = p < 0.01,

ααα = p < 0.001). Asterisks indicate a difference from EtOH for the same trial (* = p < 0.05, ** = p < 0.01, *** = p < 0.001). N = 3-4 litters per group, the number of pups tested is given in parentheses.

379

Long-term passive avoidance test retention. Long-term passive avoidance test retention between time points compares the latency time at trial #1 from 6 to 16 weeks, reflecting the ability of a mouse to remember the shock that had occurred at the last time point (Figs. 8.2 and

S8.8). Male and female progeny exposed to saline exhibited a gender-dependent gradual increase in the latency to enter the dark chamber in trial #1 from 6 to 16 weeks, indicative of learning.

Trial 1 time for saline-exposed males at 12 weeks of age was significantly greater than the trial 1 time at 6 weeks, whereas in females this difference was not apparent until 16 weeks of age. At 16 weeks of age, trial #1 time for saline-exposed males was significantly greater than trial #1 time at

6, 9 and 12 weeks of age. Trial #1 time at 16 weeks for saline-exposed females was significantly greater than trial #1 time at 6 and 12 weeks. PBN-treated females and males both had a significantly greater trial #1 time at 12 weeks compared to 6 weeks, and trial #1 time at 16 weeks was significantly greater than trial #1 time at 6, 9 and 12 weeks.

EtOH-exposed male and female progeny had no increase in learning from weeks 6-12, although at 16 weeks their trial #1 time was significantly greater than that at 6, 9 or 12 weeks. PBN pretreatment blocked the EtOH-initiated learning deficits in both female and male progeny, evidenced by a gradual increase in trial #1 time from 6 to 16 weeks in EtOH-exposed females and males, similar to the saline-exposed progeny, and trial #1 times at both 12 and 16 weeks were significantly greater than at 6 or 9 weeks.

Fetal Weight. At 6, 9, 12 and 16 weeks, there were no differences in fetal weight among any of the groups for either gender, suggesting that the dose of EtOH used was minimally toxic for the

CD-1 strain (Fig. 8.3). Males were heavier than females at all time points throughout the study

380

300 FEMALES *** PBN + EtOH *** ( ) 200 PBN Saline( ) * ‡ 100 EtOH (seconds, mean) DARK CHAMBER LATENCY TO ENTER

0 300 MALES *** PBN + EtOH *** PBN ( ) 200 Saline( ) † EtOH 100 (seconds mean) DARK CHAMBER † LATENCY TO ENTER

0 6 9 12 16 TIME (weeks)

381

Figure 8.2: Passive avoidance task retention at Trial #1 between time points in CD-1 mice exposed to EtOH during the fetal period.

Pregnant CD-1 mice were treated on gestational day (GD) 17 with ethanol (4 g/kg i.p.) or saline vehicle, with or without pretreatment with the free radical spin trapping agent PBN (40 mg/kg i.p.), and the progeny were tested postnatally for a passive avoidance task at weeks 6, 9, 12 and

16. These data show trial 1 time across the time points tested to determine longer-term task retention over weeks, as opposed to short-term task retention between trials 1 and 2. Data were analyzed by repeated measures two-way ANOVA with a post-hoc Bonferroni test, and shown as the mean. Daggers indicate a difference between saline- and EtOH-treated animals for the same time point († p < 0.05). Double daggers indicate a difference between saline and PBN-pretreated

EtOH-treated animals. Asterisks indicate a difference from EtOH alone (*** p < 0.001). N = 3-4 litters per group, 15-27 pups per group.

382

FEMALES MALES

100 80 80 SALINE Saline FEMALES MALES 60 PBN 60 EtOH PBN PBN + EtOH 80 40 EtOH 40 PBN + EtOH g, mean + SD 20 SD + mean g, 20 (20) (19) (13) 60 0 0 (20) (15) 6 9 12 16 6 9 12 16 (19) TIME (weeks) (13)(27) TIME (weeks) (13)(15) (18) (20) (16) (19) (13)(15) (13)(27)(18) (19)(20) 40 (16) (13)(15) (13)(27)(18)(16) (13)(27)(18)(16) BODY WEIGHT BODY SD) + mean (g, 20

0 6 9 12 16 6 9 12 16 TIME (weeks) TIME (weeks)

383

Figure 8.3: Effect of in utero EtOH exposure during the fetal period on postnatal weight in CD-1 mice.

Pregnant CD-1 mice were treated on gestation day (GD) 17 with ethanol (4 g/kg i.p.) or saline vehicle, with or without pretreatment with the free radical spin trapping agent phenylbutylnitrone

(PBN) (40 mg/kg i.p.), and the progeny were weighed on postnatal weeks 6, 9, 12 and 16. The data are shown as the mean + standard deviation. N = 3-4 litters per group, 15-27 pups per group.

384 regardless of treatment (p < 0.001). From 6 to 16 weeks, females gained an average of 14.51 g, and males gained an average of 18.9 g. There were no differences in weight gain among treatment groups.

DNA oxidation. EtOH-exposed progeny exhibited a 2.7-fold increase in oxidatively damaged

DNA, measured as the pathogenic 8-oxodGuo lesion, in fetal brain compared to saline-exposed controls (p < 0.01) (Fig. 8.4). PBN treatment of saline-exposed mice decreased DNA oxidation by 55% compared to saline-exposed controls, although this reduction was only marginally significant (p = 0.06). Pretreatment with PBN blocked EtOH-enhanced DNA oxidation (p <

0.05), resulting in a value that was not different from the saline-exposed group.

Teratology and fetal toxicity at 4 g/kg EtOH

CD-1 mice. At a maternal EtOH dose of 4 g/kg i.p., there were no differences in fetal weight, head length, head diameter or total fetal structural malformations in progeny exposed to EtOH compared to saline controls, nor was there evidence of maternal toxicity (Fig. 8.5). EtOH did cause a small increase in fetal resorptions compared to saline-exposed progeny (p < 0.05). PBN pretreatment had no significant effect on any of the measured parameters. Although PBN pretreatment decreased EtOH-initiated resorptions by about 50%, resulting in a level similar to saline controls, this apparent reduction was not statistically different from the group exposed to

EtOH alone.

385

1000 †† 800 (4) 600

g dGuo, mean +SD) (4)*  400 (4) a 200 (3) 0

DNA OXIDATION IN FETAL BRAIN FETAL IN OXIDATION DNA Veh PBN Veh PBN

(fmol 8-oxo-dGuo/(fmol Veh EtOH

386

Figure 8.4: Effect of PBN on EtOH-enhanced oxidatively damaged DNA in fetal brains of CD-1 mice exposed to EtOH during the fetal period.

Pregnant CD-1 mice were treated intraperitoneally on gestational day (GD) 17 with ethanol (4 g/kg) or saline vehicle, with or without pretreatment with the free radical spin trapping agent

PBN (40 mg/kg) and sacrificed 6 hr later. Fetal brains were analyzed for 8-oxo-2’- deoxyguanosine (8-oxodGuo) formation using high-performance liquid chromatography with tandem mass spectrometry. Data were analyzed by one-way ANOVA with a post-hoc Bonferroni test, and are shown as the mean + standard deviation. Daggers indicate a difference from the saline control group (†† = p < 0.01). Asterisk indicates a difference from the EtOH-treated group

(* = p < 0.05). Letter ‘a’ indicates difference from the saline control group (a = 0.5 < p < 0.1)

The number of fetal brains analyzed is given in parentheses. N=3-4 litters per group.

387

1.9 15 FETAL WEIGHT HEAD LENGTH (x,y=dams, fetuses) (x,y=dams, fetuses) 1.7 14

(7,85) 13 1.5 (7,85) (10,133) (10,122) (7,86) (10,133) (7,86) 12 (10,122)

11 g, mean+ SD

mm, mean + SD 10 1.0

10 3.5 HEAD DIAMETER CROWN-RUMP LENGTH (x,y=dams, fetuses) (x,y=dams, fetuses) (7,85) 9 3.0 (7,85) (10,133) (10,122) (7,86) (10,133) (10,122) (7,86) 8 2.5 mm, mean + mm,SD mean mm,mean + SD 7 2.0

100 100 TOTAL ANOMALIES FETAL RESORPTIONS (x,y=dams, implantations) (x,y=dams, implantations) 80 80

60 60

40 40 mean + SD mean + mean + SD + mean † 20 20 PERCENTAGE (%) PERCENTAGE (10,135) (7,86) (%) PERCENTAGE (10,138) (7,88) (10,138) (7,85) (10,135) (7,86) 0 0 VH PBN VH PBN VH PBN VH PBN EtOH EtOH

388

Figure 8.5: Teratogenicity and fetal toxicity in CD-1 mice at birth following EtOH treatment during the embryonic period.

Pregnant CD-1 mice were treated intraperitoneally on gestational day (GD) 9 (plug = GD 1) with

EtOH (4 g/kg) or saline vehicle, with or without pretreatment with the free radical spin trapping agent PBN (40 mg/kg), and sacrificed on GD 19 for fetal assessments. Pups were assessed for growth deficits and total anomalies including cleft palate, neural tube defects, microphthalmia, vestigial tail, kinky tail, and gastroschisis. Percentage was calculated by dividing the total number of fetuses affected with >1 anomaly by the number of total live fetuses. Continuous data were analyzed by one-way ANOVA with a post-hoc Bonferroni test, and binomial data were analyzed using a chi-square test. The data are shown as the mean + standard deviation. Daggers indicate a difference from the saline group († = p < 0.05). The number of litters and pups is given in parentheses.

389

Strain-dependent teratology and fetal toxicity

The dose-response relationship for EtOH-initiated fetal toxicity and structural malformations was compared at doses of 4 and 6 g/kg i.p. for the CD-1 mice used throughout this study, and the

C57BL/6 strain known to be susceptible to EtOH teratogenicity (Fig. 8.6).

C57BL/6 mice. At the low dose of EtOH (4 g/kg), EtOH reduced fetal weight, head length, head diameter and crown-rump length, and increased fetal resorptions and total structural malformations (p < 0.001) (Fig. 8.6). At the high dose (6 g/kg), EtOH caused 9% maternal lethality, as well as complete litter resorptions in 8 out of the 11 animals dosed, resulting in a small sample size for this dose. Comparing the low- and high-dose treatment groups, pups in the high-dose group had lower fetal body weights than pups in the low-dose group, as well as more resorptions (p < 0.001).

Comparison of CD-1 and C57BL/6 mouse strains. Compared to the C57BL/6 strain, saline- exposed CD-1 pups had significantly greater fetal weight, head length, head diameter and crown- rump length, but there were no differences in spontaneous anomalies or resorptions (Fig. 8.6).

At 4 g/kg EtOH, C57BL/6 progeny were substantially more susceptible to EtOH toxicity than

CD-1 pups, as evidenced by reduced fetal weight, head length, head diameter and crown-rump length, and increased resorptions and structural malformations (p < 0.001). In contrast, the

EtOH-exposed CD-1 strain at this dose were similar to saline controls. At 6 g/kg EtOH, CD-1 pups had significantly reduced fetal weight, head length, head diameter and crown-rump length, and increased total structural malformations and resorptions compared to both saline-exposed and low-dose EtOH-exposed CD-1 progeny (p < 0.001). However, in comparison to the

C57BL/6 strain, the CD-1 were substantially less affected by the 6 g/kg EtOH treatment (p <

390

2.5 14 C57BL/6 FETAL WEIGHT HEAD LENGTH 2.0 CD-1 13  + SD *** *** + SD  12 †††*** 1.5 *** *** †††*** 11 1.0 g, meang, ††† 10 ††† mean mm, ††† ††† 0.5 9 10 3.5 HEAD DIAMETER CROWN-RUMP LENGTH  9  3.0 *** + SD + SD *** † *** *** *** ***†† ††† 8 2.5  7 2.0 ††† ††† mm, mean mm, ††† mm, mean  6 1.5

200 RESORPTIONS 150 TOTAL ANOMALIES *** ††† †††** 100 100 * † ††† ** 50 † ††† 0 PERCENTAGE (%) PERCENTAGE PERCENTAGE (%) PERCENTAGE 0 0 4 6 0 4 6 EtOH (g/kg) EtOH (g/kg)

391

Figure 8.6: Comparison of in vivo ethanol teratogenesis in C57BL/6 and CD-1 mice following treatment during the embryonic period.

Pregnant CD-1 and C57BL/6 dams were treated on GD 9 with either EtOH (4 g/kg) or its saline vehicle, and sacrificed on GD 19. Offspring were assessed for growth deficits and total anomalies including cleft palate, neural tube defects, microphthalmia, vestigial tail, kinky tail, and gastroschisis. Percentage was calculated by dividing the total number of fetuses affected with >1 anomaly by the number of total live fetuses. Continuous data were analyzed by one-way

ANOVA with a post-hoc Bonferroni test, and binomial data were analyzed using a chi-square test. The data are shown as the mean + standard deviation. Daggers indicate a difference from the saline group of the same strain († p < 0.05,†† < 0.01). Asterisks indicate a difference between strains for the same treatment (* p < 0.05, ** p < 0.01, *** p < 0.001). N=7-10 litters per group

392

0.001), evidenced by greater fetal weight, head length, head diameter and crown-rump length, and less anomalies and resorptions than the 6 g/kg EtOH-exposed C57BL/6 progeny (p < 0.001).

Metabolism. Saline-treated CD-1 and C57BL/6 mice had no detectable plasma EtOH concentration (data not shown). C57BL/6 mice treated with 2.8 g/kg EtOH achieved a peak

EtOH plasma concentration of 81 mM by 30 minutes, which was 2.5 fold higher than that achieved in the CD-1 mice, 32 mM (p < 0.001) (Fig. 8.7). Plasma concentrations at both 1 and 2 hr were at least 3-fold higher in the C57BL/6 mice compared to the CD-1 mice (p < 0.001). By 4 hr, there were no differences between CD-1 and C57BL/6 mice, and levels were undetectable by

8 hr.

393

150

100 *** + SD) *** C57BL/6 *** CD-1 50

0 2 4 6 8 10 (mmol/L, mean (mmol/L, Time (h)

Plasma EtOH concentration EtOH Plasma -50

394

Figure 8.7: Comparison of EtOH pharmacokinetics in C57BL/6 and CD-1 mice following a single dose of ethanol.

Male CD-1 and C57BL/6 mice were treated with EtOH (2.8 g/kg i.p), and sacrificed at various time points to measure blood ethanol concentrations. Each time point contains 3-6 samples. Data were analyzed using a one-way ANOVA with a post-hoc Bonferroni test, and shown as the mean

+/- standard deviation (SD). SDs for CD-1 mice are smaller than the symbols. Asterisks indicate a difference from CD-1 mice at the same time point (*** p < 0.001).

395

8.5 Discussion

EtOH exposure during pregnancy causes a spectrum of behavioural deficits affecting both learning and memory in animal models (Brocardo et al., 2012; Museridze et al., 2008). Our study of CD-1 mice found that a single prenatal EtOH exposure on GD 17 during fetal development increases oxidatively damaged DNA in fetal brain, and decreases postnatal performance in the passive avoidance test, all at a lower dose that does not cause structural malformations or reduced fetal body weight in this strain. Our use of a single exposure and a relatively low dose for the CD-1 strain reveals a sensitive threshold for a minimal effect of EtOH on the developing fetal brain. These results, reflecting a deficit in learning and memory, reveal the exquisite susceptibility of brain function in the fetal period to impairment by in utero exposure to a relatively non-teratogenic dose of EtOH for this strain. This is the first report of passive avoidance learning deficits after a single in utero exposure to EtOH, although previous studies have observed such deficits with chronic EtOH exposure throughout pregnancy in both rats (Abel, 1982; Mattson et al., 1993; Riley et al., 1979) and mice (Becker and Randall, 1989;

Fiore et al., 2009; Gilliam et al., 1987).

We use the fetus rather than the litter as the experimental unit for statistical analysis in studies of mechanisms involving drug reactive intermediates and reactive oxygen species (ROS), which generally are too unstable to leave the cell of their formation. As a result, susceptibility is primarily determined at the level of the individual fetus, particularly in regard to fetal pathways for the formation and detoxification of ROS, and repair of oxidatively damaged DNA. This fetal level of control for ROS susceptibility has been demonstrated in the breeding of heterozygous knockout mice deficient in ROS-protective pathways resulting in all 3 +/+, +/- and -/- fetal

396 genotypes as littermates, which differ in their susceptibility to ROS in a gene dose-dependent fashion (Wells et al., 2009b). This genetic variability among littermates is lost when the litter rather than the fetus is used as the experimental unit, although in this study the key differences remained significant when the data were analyzed by litter (Fig. S8.9). Although the mice used herein are not genetically modified, outbred strains such as the CD-1 mouse demonstrate great genetic variability, being polymorphic at a significant number of loci (Aldinger et al., 2009).

This genetic variability may account for the variability in responses seen in individual pups tested in the passive avoidance task, despite exposure to the same treatment as their littermates in the same dam.

Compared with CD-1 mice treated in the earlier period of embryonic development, the substantially enhanced, dose-dependent susceptibility of the C57BL/6 strain to EtOH-initiated fetal toxicity and structural malformations evident at birth suggests the possibility for a similar strain-dependent susceptibility in functional brain development, which may be reflected in similar genetic predispositions in humans. A similar strain-dependent shift in sensitivity has been observed between C57BL/6 mice and another outbred strain, ICR mice, both in embryo culture and in neural crest cell culture, whereby ICR embryos require higher EtOH concentrations than

C57BL/6 embryos to initiate major structural malformations (Hunter et al., 1994; Kotch et al.,

1995); and neural crest cells from ICR mice are relatively less sensitive to EtOH-initiated toxicity compared to those from C57BL/6 mice in a time- and concentration-dependent fashion

(Chen et al., 2000). The doses of EtOH given in this study are comparable to those used in other studies in vivo in C57BL/6 mice (Dong et al., 2010) and in CD-1 mice (Blakley and Scott,

1984b). The peak plasma EtOH concentration achieved in CD-1 mice upon administration of 4 g/kg has been reported as approximately 400 mg/dL (100 mM) (Blakley and Scott, 1984a). This concentration is higher than that achieved in humans (20-200 mg/dL, 5-47 mM) (Khaole et al.,

397

2004; May et al., 2008), and suggests that the concentration-response curve for at least one mouse strain is shifted to the right compared to humans, such that a higher EtOH dose and concentration are required in a mouse to achieve the same effect observed in humans. Indeed, the dose of 4 g/kg used herein was relatively low for the CD-1 strain, demonstrated by the lack of structural anomalies, which was also observed by (1984b). In regard to strain differences in susceptibility to EtOH, herein we found that C57BL/6 mice achieve a 2.5-fold higher peak EtOH plasma concentration than CD-1 mice when administered the same dose, which explains at least in part why we observed structural anomalies C57BL/6 mice, but not in the CD-1 strain.

Although males were used in the pharmacokinetic study in part to minimize hormonal confounders, the relative strain-dependent genetic differences in metabolism would be expected to be gender independent. Although pregnancy may increase EtOH metabolism by alcohol and aldehyde dehydrogenases in some rodent species and strains, resulting in lower plasma concentrations (Badger et al., 2005), other rodent studies have found metabolism was decreased or unchanged during pregnancy (Petersen et al., 1977; Traves et al., 1995; Traves and Lopez-

Tejero, 1994), possibly varying with the species and/or strain, as well as the route of administration and whether animals were exposed chronically or acutely, so the absolute plasma concentrations in each mouse strain may or may not be different in pregnant females compared to the males studied herein. The substantially lower plasma concentrations achieved in CD-1 mice following i.p. administration also show that the widely assumed resistance of CD-1 mice to

EtOH teratogenicity is actually due at least in large part to enhanced metabolism rather than a shift in the plasma concentration-response curve between “resistant” CD-1 mice and the susceptible C57BL/6 strain. A similar mechanism likely contributes in some cases to differences in susceptibility to FASD among children, including different children from the same mother.

398

With fetal exposure to EtOH in CD-1 mice, the protective effect of the free radical spin trapping agent PBN in blocking both EtOH-initiated DNA oxidation in fetal brain and postnatal behavioural deficits suggests that ROS, and possibly DNA oxidation, are involved in the mechanism of neurodevelopmental deficits. PBN is known to block embryonic and/or fetal

DNA oxidation caused by a number of ROS-initiating teratogens, and similarly to block the structural malformations caused by these teratogens (Wells et al., 2009b). For example, PBN blocks oxidation of embryonic cellular macromolecules (DNA, protein, lipids) and teratogenesis in pregnant rabbits treated with the sedative/antileprotic/anticancer drug thalidomide (Lee et al.,

2011; Parman et al., 1999), and in pregnant mice treated with the antiepileptic drug phenytoin

(Liu and Wells, 1994; Wells et al., 1989), as well as trapping free radicals of structurally similar drugs and reducing macromolecular oxidation in vitro (Parman et al., 1998). The importance of oxidatively damaged DNA in the developing embryo and fetus is corroborated by preliminary results in oxoguanine glycosylase 1 (OGG1) mice that lack DNA repair, which when exposed to

EtOH during either the embryonic or fetal period, are more susceptible to both structural and behavioural deficits compared to wild-type littermates, effects which are blocked by pretreatment with PBN (Miller et al., 2013a; Miller and Wells, 2013). The protective effect of PBN against

ROS may involve not only directly trapping free radicals, but also indirectly by blocking inflammatory pathways that produce ROS. In particular, PBN can inhibit: (1) the induction of the inducible nitric oxide synthase (iNOS), thereby decreasing nitric oxide production (NO); (2) nuclear factor κB (NFκB) activation; (3) inducible cyclooxygenase (COX2) mRNA expression and COX catalytic activity (Kotake et al., 1998); (4) transcription of proinflammatory cytokines such as tumor necrosis factor-α (TNF- α) and interferon-gamma (IFN-γ), which is correlated with decreased activation of the NFκB and activator protein-1 (AP-1) transcription factors (Sang et al., 1999); (5) caspase-3 cleavage (McLaughlin et al., 2003); and, (6) activation of NADPH

399 oxidases (NOXs) (Chang et al., 2009), which are activated by EtOH in the embryo in vivo, increasing ROS, embryonic DNA oxidation and structural teratogenesis (Dong et al, 2010).

Particularly with EtOH, the inhibition of NOXs by PBN in the case of EtOH may complement

PBN trapping free radicals, although contributions from the other mechanisms cannot be excluded.

In addition to serving as a biomarker of oxidative stress, DNA oxidation appears to be a developmentally pathogenic lesion, likely via altered gene transcription rather than enhanced mutagenesis (Wells et al., 2010). In pregnant DNA repair knockout mice treated with the ROS- initiating drug methamphetamine, the null progeny deficient in either oxoguanine glycosylase 1

(OGG1) (Wong et al., 2008) or Cockayne Syndrome B (CSB) (McCallum et al., 2011c) exhibit enhanced motor coordination deficits compared to their DNA repair-normal wild-type littermates. In the saline-exposed fetuses herein, the 55% decrease in brain DNA oxidation with

PBN pretreatment suggests that PBN may be scavenging endogenous ROS, which can be developmentally pathogenic, as revealed in untreated progeny that are genetically deficient in glucose-6-phosphate dehydrogenase (G6PD) (Nicol et al., 1995; Nicol et al., 2000), catalase

(Abramov and Wells, 2011a) or ataxia telangiectasia mutated (ATM) , in each case compared to their wild-type normal littermates. In EtOH-exposed progeny, PBN pretreatment reduced DNA oxidation to control levels, consistent with a protective mechanism via free radical scavenging and ROS reduction, although other ROS-dependent mechanisms may contribute.

In addition to the single exposure used herein, several studies using chronic EtOH administration have reported protection by antioxidative compounds: (1) Dolivin, an antioxidant containing vitamin E and hypoxen, enhanced passive avoidance test performance in adult rat progeny exposed to EtOH in utero (Museridze et al., 2008); (2) vitamins C and E administered

400 with EtOH to pregnant guinea pigs between GDs 2-67 improved task-retention memory (Nash et al., 2007); and (3) the antioxidant salymarin improved social recognition tasks (males) and spatial working memory (females) in rat progeny exposed to EtOH throughout gestation via a maternal liquid diet containing 35% EtOH (Busby et al., 2002; Neese et al., 2004; Reid et al.,

1999). These antioxidant studies are consistent with a role for ROS in the neurodevelopmental deficits caused by in utero EtOH exposure; however, ROS and/or their molecular effects (e.g. macromolecular oxidation) were not measured in fetal brain, so protective mechanisms unrelated an antioxidative effect in these studies cannot be excluded.

In summary, a single in utero exposure in the fetal period to a minimally toxic dose of EtOH for this CD-1 mouse strain causes postnatal cognitive deficits in the progeny, despite the absence of structural teratogenesis or fetal toxicity at birth. The reduction in both EtOH-initiated DNA oxidation in fetal brain and postnatal cognitive deficits by pretreatment with the free radical spin trapping agent PBN suggests a ROS-dependent mechanism, possibly involving oxidatively damaged DNA. The susceptibility of CD-1 mouse progeny to EtOH-initiated neurodevelopmental deficits is remarkable in light of the relative resistance of this strain, compared to C57BL/6 progeny, to the structural teratogenesis and fetal toxicity at birth caused by EtOH administration during the embryonic period, which suggests the possibility of genetic determinants of risk for children exposed to EtOH during pregnancy.

401

8.6 Supplemental Figures

FEMALES SALINE MALES

400 aaa aaa c bbb N=15 c 300 N=17 aa 200

100

0 PBN

400 aaa aaa bbb bbb cc cc N=27 N=20 300 aa a 200

100

0 EtOH

400 aaa bbb 300 a ccc bb N=20 N=19 c 200 (seconds, mean + SD) 100

0 PBN + EtOH LATENCY TO ENTER DARK CHAMBER

aaa aaa 400 aaa bb bbb b aaa N=18 bb 300 N=21

200

100

0 6 9 12 16 6 9 12 16

TASK RETENTION BETWEEN TIME POINTS (weeks)

402

Supplementary Figure S8.8: Long-term passive avoidance test retention at trial #1 between time points in CD-1 mice exposed to ethanol (EtOH) during the fetal period.

Pregnant CD-1 mice were treated intraperitoneally (i.p) on gestational day (GD) 17 with ethanol

(4 g/kg) or saline vehicle, with or without pretreatment with the free radical spin trapping agent phenylbutylnitrone (PBN) (40 mg/kg), and the progeny were tested postnatally for a passive avoidance task at weeks 6, 9, 12 and 16. These data plot trial 1 time across the time points tested to determine longer-term task retention over weeks, as opposed to short-term task retention between trials 1 and 2. Data were analyzed by one-way ANOVA with a post-hoc Bonferroni test, and shown as the mean + standard deviation. Letter ‘a’ indicates difference between from 6 weeks (a p < 0.05, aa p < 0.01, aaa p < 0.001). Letter ‘b’ indicates difference between from 9 weeks (b p < 0.05, bb p < 0.01, bbb p < 0.001). Letter ‘c’ indicates difference between from 12 weeks (c p < 0.05, cc p < 0.01, ccc p < 0.001).

403

FETUS AS THE UNIT 400   *** (13) ***(16) (27) (16) 300

(13) (27) 200 †† ‡ (18)

DARK CHAMBER 100 LATENCY TO ENTER

(seconds, mean + SD) (18)

400 LITTER AS THE UNIT ** **(3) (3) 300 (3) (4)

200 (3) (4) † (3)

DARK CHAMBER 100 LATENCY TO ENTER

(seconds, mean + SD) (3)

0 TRIAL NO. ______1 2______1 2______1 2 ______1 2 TREATMENT SALINE PBN ______SALINE PBN 4g/kg EtOH

404

Supplementary Figure S8.9: Comparison of passive avoidance behaviour data analyzing the fetus as the unit versus the litter as the unit.

The data for passive avoidance behaviour were analyzed using the fetus as the unit (top panel) and the litter as the unit (bottom panel) using one-way ANOVA with a post-hoc Bonferroni test, and shown as the mean + standard deviation. Daggers indicate a difference from saline for the same trial († = p < 0.05, †† = p < 0.01). Double daggers indicate a difference from PBN for the same trial (‡ = p < 0.05). Alpha symbols indicate a difference from trial 1 within the same treatment group (α = p < 0.05, αα = p < 0.01). Asterisks indicate a difference from EtOH for the same trial (* = p < 0.05, ** = p < 0.01). The number of fetuses and litters tested is given in parentheses.

405

Chapter 9 Summary, Conclusions and Future Studies

406

9.1 Summary and Conclusions

In utero EtOH exposure can result in a spectrum of structural and behavioural anomalies in the developing child that vary by timing, duration and dose, collectively termed FASD (Jones and Jones, 1976; Jones, 2011). The complete phenotype is termed Fetal Alcohol Syndrome

(FAS), characterized by craniofacial malformations, growth deficits and behavioural deficits.

The mechanisms underlying this toxicity are unclear, however evidence suggests that ROS and oxidative stress are involved in the mechanism (Brocardo et al., 2011).

Reactive oxygen species (ROS) are necessary for embryonic development, as they are involved in signal transduction pathways that mediate the spatiotemporally controlled pathways underlying normal embryonic development (Hansen and Harris, 2013). Enhanced oxidative stress can oxidatively damage cellular macromolecules (including DNA, proteins and lipids) or alter signal transduction pathways, mechanisms which have been implicated in a spectrum of adverse structural and behavioural developmental consequences (Wells et al., 2009b).

MeOH, a structurally similar alcohol, causes a different spectrum of developmental and adult toxicities than EtOH, and limited evidence suggests the involvement of ROS ROS (Wells et al., 2013). Herein, we sought to determine potential sources of MeOH-initiated ROS, as well as to determine embryonic pathways that may be a determinant of risk. PBN, a free radical spin trapping agent, has offered protection against other ROS-initiating teratogens including phenytoin and thalidomide. We sought to determine whether PBN would offer similar protection against MeOH or EtOH embryopathies and teratogenesis, thereby implicating ROS in the mechanism of toxicity.

407

Additionally, previous studies have demonstrated the contribution of embryonic catalase as a determinant of risk for phenytoin teratogenesis, as well as the potential for exogenous forms of catalase to protect against EtOH embryopathies. Using genetically modified mouse models with partial genetic enhancements and deficiencies in catalase, both in vivo and in whole embryo culture, we investigated the contribution of endogenous embryonic catalase to MeOH and EtOH developmental toxicity.

EtOH has been shown to oxidatively damage embryonic DNA as well as fetal brain DNA lending to structural and behavioural deficits. We sought to determine whether embryonic DNA repair, specifically by OGG1, the enzyme that repairs 8-oxodGuo, is a determinant of risk, thereby implicating 8-oxodGuo as an embryopathic macromolecular lesion.

The results are summarized as follows (Fig. 9.1):

1. Endogenous embryonic catalase is protective in embryos exposed in culture to MeOH

and EtOH, as evidenced by decreased embryopathies in hCat embryos with

genetically enhanced levels of catalase, and their converse exacerbation in catalase-

deficient aCat embryos.

2. Endogenous embryonic catalase protects against EtOH-initiated structural deficits and

DNA oxidation in vivo, as demonstrated by decreased teratogenicity and embryonic

DNA oxidation measured as 8-oxodGuo in hCat mice, and converse increases in aCat

mice.

408

Figure 9.1. Summary of results of thesis studies using EtOH and MeOH.

409

1. Embryonic NADPH oxidases are potential sources of embryonic ROS upon MeOH

exposure, as demonstrated by enhanced embryonic p22phox mRNA and protein

expression, and embryonic protein oxidation upon in utero MeOH exposure, all of

which are blocked by pretreatment with the NOX inhibitor DPI, or with the free

radical spin trapping agent PBN.

2. PBN, a free radical spin trapping agent, can mitigate both the structural and

behavioural deficits initiated by in utero EtOH exposure, as evidenced by decreased

structural birth defects and postnatal behavioural deficits when pretreated with PBN,

implicating ROS in the pathogenic mechanism.

3. Embryonic DNA repair, specifically by OGG1, the enzyme that repairs 8-oxodGuo, is

a determinant of risk for both the structural and behavioural deficits caused by EtOH.

Embryonic EtOH exposure during organogenesis enhanced both embryopathies and

DNA oxidation in an ogg1 gene dose-dependent fashion, while EtOH exposure in the

later fetal period enhanced both behavioural deficits and fetal brain DNA oxidation in

an ogg1 gene dose-dependent fashion.

Taken together, these results suggest that in utero MeOH and/or EtOH enhance NOX- initiated ROS formation, which if not detoxified by embryonic catalase, can oxidatively damage

DNA, resulting in structural and functional birth defects in the developing embryo or fetus.

These results suggest that targeting embryonic pathways of redox control and DNA repair may constitute a therapeutic approach in children prenatally exposed to MeOH or EtOH, while deficient DNA repair may constitute a risk factor. Indeed, mothers could eat cruciferous vegetables such as broccoli which contains sulforaphane, an isothiocyanate compound that

410 activates Nrf2 and stimulates Ogg1 expression, in addition to numerous other enzymes for ROS detoxification and xenobiotic elimination (Boddupalli et al., 2012; Singh et al., 2013).

9.1.1 Ethanol studies

Catalase is an antioxidative enzyme responsible for the detoxification of H2O2, a molecule involved in signal transduction pathways and also capable of forming the highly toxic hydroxyl radical by reacting with iron via the Fenton reaction (Lloyd et al., 1997).

In our studies using mutant catalase-deficient mice and transgenic mice expressing human catalase, the protection offered by catalase against EtOH embryopathies in whole embryo culture and EtOH-initiated DNA oxidation and teratogenesis in vivo by both genetically modified mice expressing enhanced levels of endogenous embryonic catalase (hCat mice), and exogenously administered polyethylene glycol-conjugated catalase (PEG-cat), together with the converse exacerbation of both EtOH embryopathies in culture, and EtOH-initiated DNA oxidation and teratogenesis in vivo in acatalasemic (aCat) mice, show that embryonic catalase, despite its low level, is a determinant of risk for EtOH teratogenicity. These results also corroborate studies implicating ROS in the pathogenic mechanism.

While not measured herein, our laboratory has shown that aCat embryos have substantially lower catalase activity than their C3H WT controls, while hCat embryos conversely have higher activity compared to their C57BL/6 WT controls, all of which are only about 5% of their respective maternal hepatic activities (Abramov and Wells, 2011a). Both maternal PEG-Cat administration in in vivo studies (Abramov and Wells, 2011a; Winn and Wells, 1999), and its direct addition to embryo culture medium (Abramov and Wells, 2011b; Winn and Wells, 1995), significantly increase embryonic catalase activity, which is sustained over 24 hours, consistent with its protection against embryopathies and teratogenesis observed herein.

411

The protection afforded by endogenous catalase is remarkable, given that it is expressed at only about 5% of maternal hepatic activity (Abramov and Wells, 2011a; Winn and Wells,

1997; Winn and Wells, 1999). A similar protective effect of endogenous embryonic catalase against endogenous and drug-enhanced oxidative stress in embryo culture has been observed in the hCat and aCat strains used herein when exposed to the ROS-initiating antiepileptic drug phenytoin or its vehicle, the latter revealing the embryopathic potential of physiological levels of embryonic ROS (Abramov and Wells, 2011).

Herein, using oxoguanine glycosylase 1 (OGG1) knockout mice that are unable to repair the 8-oxodGuo lesion, we provide the most direct evidence to date that oxidatively damaged

DNA, and particularly the 8-oxodGuo lesion, plays a pathogenic role in the mechanism of EtOH teratogenicity, and that embryonic DNA repair, specifically OGG1, is a determinant of risk for

ROS in the oxidative macromolecular damage. Furthermore, we discovered a gender-dependent risk for embryopathies, whereby females exhibited more embryopathies than males of the same ogg1 genotype treated with the same concentration of EtOH. The basis for this gender effect is unknown, but could involve differences in EtOH-metabolizing enzymes, hormonal influences on

EtOH absorption/metabolism, differences in antioxidative enzymes that detoxify EtOH-initiated

ROS, or differences in downstream DNA repair enzymes.

In addition, our study in ogg1 DNA repair-deficient knockout mice treated with EtOH during the fetal period and assessed postnatally for cognitive deficits demonstrates for the first time: (1) a behavioural phenotype for these mice in the absence of drug exposure; (2) that a

412 single prenatal EtOH exposure on GD 17 during fetal development increases DNA oxidation measured as the 8-oxodGuo lesion in fetal brain; and, (3) this single EtOH exposure decreases postnatal performance in the passive avoidance test at a minimally toxic dose that does not affect postnatal body weight of the progeny. These results, reflecting a deficit in learning and memory, reveal the exquisite susceptibility of developing brain function in the fetal period to impairment by in utero exposure to what we previously have shown is a relatively non-teratogenic dose of

EtOH in this strain with respect to structural birth defects (Miller et al., 2013b), consistent with the absence of any effect of EtOH in progeny body weight observed herein. Perhaps most noteworthy, our results provide the most direct evidence to date of a causal role for oxidatively damaged DNA in the fetal brain, and a protective role for DNA repair, in the mechanism of

EtOH-initiated neurodevelopmental deficits.

Interestingly, the dose employed to measure oxidatively damaged DNA in ogg1-deficient embryos during the embryonic period was 50% higher than that used in the later fetal period to study effects on functional brain development. It is interesting to note that while CD-1 mice were exposed in utero to a dose of 4 g/kg EtOH on GD 17, this dose was too high for ogg1- deficient mice and their congenic wild-type controls. Indeed, the dose was decreased by half in order to exhibit the same spectrum of behavioural deficits in the absence of maternal toxicity.

Remarkably, the dose of 4 g/kg employed during the embryonic period which resulted in

20 fmol 8-oxodG/μg DNA in ogg1 +/+ in whole embryos (2-fold increase over vehicle-treated whole embryos), while only half that dose (2 g/kg) during the fetal period resulted in a 3-fold higher level of 60 fmol 8-oxodG/μg DNA in ogg1 +/+ fetal brain (2.2-fold increase over vehicle- treated fetal brains). These results highlight the unique sensitivity of the developing fetal brain,

413 whereby half the dose of EtOH is sufficient to produce the same amount of DNA oxidation in an

EtOH-exposed embryo.

In contrast to results in vivo, while adult ogg1 knockout mice accumulate 8-oxodGuo

(Klungland et al., 1999), saline-treated +/- and -/- OGG1-deficient embryos exhibited no increase in baseline DNA oxidation, and they all developed normally in culture. This suggests that oxidative stress may be lower during the embryonic period than in adults, and/or that considerable time is necessary for the accumulation of measurably increased 8-oxodGuo lesions in OGG1-deficient animals. The latter possibility would be consistent with the increased levels of 8-oxodGuo in aging but not young wild-type mice expressing prostaglandin H synthase 1

(PHS1), which bioactivates endogenous substrates like neurotransmitters to free radical intermediates, compared to aging PHS-knockout controls that exhibited no increase in brain levels of 8-oxodGuo (Jeng et al., 2011).

While the 8-oxodGuo lesion is a biomarker for oxidative stress, several studies suggest that it is also a pathogenic lesion, likely via altered transcription as opposed to increased mutagenesis

(Wells et al., 2010). As observed herein for EtOH, whereby fetal brain DNA oxidation and learning deficits were increased in an ogg1 gene dose-dependent fashion, CD-1 progeny exposed in utero to EtOH exhibited enhanced fetal brain DNA oxidation and learning deficits (Miller et al., 2013b), ogg1 knockout progeny exposed in utero to methamphetamine exhibited enhanced motor coordination deficits (Wong et al., 2008), and knockout mice deficient in Cockayne

Syndrome B (CSB), another DNA repair protein, also exhibited enhanced motor coordination deficits compared to their DNA repair-normal wild-type littermates (McCallum et al., 2011c).

Several potential mechanisms exist whereby in utero formation of the 8-oxodGuo lesion in fetal brain may initiate postnatal behavioural deficits, including altering the expression and activity of

414 proteins required for normal embryonic development, alteration of gene transcription or expression via its ability to regulate binding affinity of various transcription factors including nuclear factor kappa B (NF-κB) to specific promoter elements, and/or apoptosis resulting from

9.1.2 Methanol studies

As observed for EtOH, in the studies herein, aCat and hCat mice exposed to MeOH in whole embryo culture exhibited respectively, increased and decreased embryopathies, suggesting a role of ROS in the embryopathic mechanism.

MeOH is known to be embryopathic in rodents both in vivo and in whole embryo culture

(Abbott et al., 1995; Andrews et al., 1993; Degitz et al., 2004; Rogers and Mole, 1997; Rogers et al., 1993), although the underlying mechanisms are unclear. Studies have shown that MeOH can form free radicals (Kadiiska and Mason, 2000; Skrzydlewska et al., 2000) that can enhance the production of ROS, which in turn have been implicated in the mechanism of developmental toxicity (Harris et al., 2004). In vivo evidence for the metabolism of MeOH to a free radical intermediate, and enhanced formation of ROS and oxidative stress, includes: (1) Direct detection of MeOH-initiated ROS formation in the liver of MeOH-intoxicated adult male rats after MeOH intoxication, and an increase in both lipid and protein oxidation (Skrzydlewska et al., 2000); (2) increased lipid peroxidation measured as thiobarbituric acid-reactive substances (TBARS) in the livers of adult rats exposed to 6 g/kg MeOH, as well as a variably altered antioxidant defense profile in the brain and liver (Skrzydlewska and Farbiszewski, 1998); (3) direct detection of a

MeOH-derived adduct with the free radical trapping agent α-(4-pyridyl 1-oxide)-N-tert-

415 butylnitrone (POBN) in the bile and urine of MeOH-intoxicated rats 2 hr after treatment with a toxic dose of MeOH (Kadiiska and Mason, 2000); (4) Co-administration of the glutathione precursor N-acetylcysteine or the antioxidant U-83836E decreased MeOH-initiated lipid peroxidation in erythrocytes of MeOH-intoxicated rats (Dobrzynska et al., 1999); (5) MeOH- exposure increased formation of the lipid peroxidation product malondialdehyde (MDA) and increased antioxidative enzyme activities in the lymphoid organs of adult rats (Parthasarathy et al., 2006b).

PBN pretreatment was not protective, while embryonic GSH depletion by maternal pretreatment with buthionine sulfoximine, an inhibitor of GSH synthesis, expanded the spectrum of fetal malformations and the number of malformations per fetus, albeit without altering the incidence of the most common MeOH-initiated malformations, suggesting that formaldehyde- macromolecular adducts rather than ROS were causing the new malformations in the expanded

416 teratological spectrum, with GSH protecting via its role as a cofactor for formaldehyde dehydrogenase rather than as an antioxidant (Siu et al., 2013a). The absence of any alteration in the most common MeOH-initiated malformations suggested alternative ROS-independent mechanisms for some target tissues. Similarly in adult male mice treated acutely or chronically with MeOH, oxidatively damaged DNA and lipid peroxidation were not enhanced in most tissues, even in DNA repair-deficient oxoguanine glycosylase 1 (ogg1) knockout mice

(McCallum et al., 2011a; McCallum et al., 2011b). The basis for the discrepancy in ROS involvement observed herein and elsewhere in rodent embryo culture and some adult rodent studies, but not in the in vivo studies from our laboratory, is unknown, but may include the influence of diffusible modulatory factors from maternal or placental sources with in vivo studies. In contrast, in the in vivo component of the studies herein, whereby CD-1 dams were dosed with 4 g/kg MeOH i.p. on GD 9, maternal administration enhanced embryonic levels of oxidatively damaged protein, which was blocked by pretreatment with DPI and PBN, indicating that MeOH can enhance embryonic NOX-catalyzed ROS formation in vivo, as has been reported for EtOH (Dong et al., 2010). The discrepancy in evidence for ROS involvement with embryo culture vs. some but not all in vivo MeOH studies has not been observed with other ROS- initiating teratogens like phenytoin, benzo[a]pyrene, thalidomide and methamphetamine (Wells et al., 2009b), nor for EtOH, as discussed above. For these and other reasons, it is unclear which model (embryo culture vs. in vivo), species (rabbits are resistant) and rodent strain (at least one mouse and rat strain are resistant) would best predict the human developmental risk from MeOH exposure (Wells et al., 2013).

The use of whole embryo culture was employed to discriminate the role of endogenous embryonic catalase by removing confounding maternal factors, including the metabolism of

MeOH and its formic acid metabolite by maternal catalase (Dorman et al., 1995). Maternal

417 factors may have confounded the interpretation of related in vivo studies of MeOH teratogenicity using these same mouse strains, in which the protective role in hCat mice was equivocal, while both the aCat mice and their WT strain appeared resistant to MeOH teratogenicity (Siu et al.,

2013b). In our studies herein, the C3H WT controls for the aCat mice were substantially more resistant to MeOH embryopathies than the C57BL/6 WT controls for the hCat strain. The mechanism of resistance is not known, but this resistance, possibly along with maternal factors, may explain why the aCat mice in vivo exhibited no teratological effects of MeOH. This in vivo resistance is in contrast to our embryo culture studies herein, where aCat embryos were more susceptible than WT controls to the embryopathic effects of even endogenous oxidative stress, and dramatically more so to MeOH embryopathies.

Herein, we demonstrated for the first time that in utero exposure to MeOH via maternal administration upregulated embryonic p22phox mRNA and protein expression, which was associated with increased oxidative stress as measured by oxidatively damaged embryonic protein. In embryo culture, MeOH caused a concentration-dependent increase in embryopathies similar to that observed for EtOH, a structurally similar alcohol and ROS-initiating teratogen

(Hunter et al., 1994; Kotch et al., 1995; Xu et al., 2005). Maternal pretreatment with the NOX inhibitor DPI decreased MeOH-initiated embryonic p22phox mRNA and protein expression, and the level of oxidatively damaged embryonic protein, while pretreatment with the free radical spin trap PBN decreased embryonic p22phox mRNA expression and the level of oxidatively damaged protein. This in vivo evidence for MeOH induction of embryonic NOX and ROS-mediated oxidative damage to embryonic protein, together with the protection against MeOH embryopathies in culture provided by PBN and DPI pretreatment, suggest that the developmental toxicity of MeOH is due at least in part to embryonic NOX induction and NOX-catalyzed ROS formation. This putative mechanism is similar to that postulated for EtOH embryopathies (Dong

418 et al., 2010). On the other hand, the failure of pretreatment with the PHS inhibitor ETYA to block MeOH embryopathies, despite protecting against the embryopathic effects of the ROS- initiating teratogen phenytoin, a known substrate for PHS-catalyzed bioactivation (Wells et al.,

2009b), suggests that neither MeOH nor its metabolites are bioactivated by embryonic PHSs to pathogenic free radical intermediates.

419

Although primarily known for its efficacy in trapping of free radicals, phenylbutylnitrone

(PBN) pretreatment also decreased the MeOH-initiated upregulation of p22phox mRNA, but not protein. A previous study reported PBN inhibition of p67phox protein levels (Chang et al.,

2009), although the mechanism underlying this inhibition remains unclear. By inhibiting NOX,

PBN may be inhibiting the formation of ROS, which along with its efficacy as a free radical spin trapping agent (see below) would be consistent with a hypothesis involving oxidative stress, given its protection against MeOH embryopathies in culture. PBN also inhibited protein oxidation at the same time points for which NOX mRNA was inhibited, consistent with embryonic NOX-initiated ROS-formation and oxidative stress. PBN also inhibits mRNA expression for tumor necrosis factor alpha (TNF-α) (Chang et al., 2009; Lin et al., 2006; Sang et al., 1999), which stimulates NOX activity (Li et al., 2005a), suggesting an upstream mechanism by which PBN may decrease NOX expression.

9.1.3 Embryopathic potential of physiological levels of embryonic ROS

The embryopathic potential of physiological levels of endogenous embryonic ROS was exemplified in several studies herein: (1) aCat mouse embryos exposed only to vehicle had smaller yolk sac diameters compared to C3H WT controls in whole embryo culture; (2) the 50% increase in baseline fetal resorptions in the untreated aCat mice compared to their C3H WT controls, and the converse absence of increased resorptions in untreated hCat mice compared to their C57BL/6 WT controls, the former of which is consistent with published data using this mouse strain (Abramov and Wells, 2011a), and may contribute to their reduced apparent fertility, suggesting that embryonic catalase is important for protecting against even normal physiological oxidative stress, which can be embryopathic; (3) ogg1-deficient fetuses exposed once in utero to only saline vehicle exhibit enhanced DNA oxidation and postnatal cognitive deficits in an ogg1

420 gene dose-dependent fashion. This neurodevelopmental deficit in untreated animals constitutes the first demonstration of a phenotype for ogg1 DNA repair-deficient mice, which somewhat surprisingly do not exhibit an increase in cancer (Klungland et al., 1999). (4) PBN treatment alone decreased baseline fetal brain DNA oxidation in CD-1 mice exposed once to only saline vehicle in utero on GD 17; and, (5) in ogg1 -/- embryos exposed only to saline vehicle, the decreased embryonic DNA oxidation with PBN pretreatment suggests that PBN may be scavenging endogenous ROS.

The embryopathic potential of endogenous ROS has been similarly observed in mutant progeny deficient in glucose-6-phosphate dehydrogenase (G6PD), which provides NADPH necessary for ROS detoxification: G6PD-deficient fetuses exhibit increased in utero death compared to their wild-type littermates (Nicol et al., 2000). Similarly, DNA repair-deficient knockout embryos lacking ataxia telangiectasia mutated (ATM), which directs the repair of oxidative DNA damage, exhibited more embryopathies than wild-type littermates, with an atm gene dose-dependent increase in susceptibility from wild-type to heterozygous to homozygous knockout (Bhuller and Wells, 2006).

9.1.4 Strain differences in EtOH teratogenicity

When comparing across all studies performed in this thesis, strain differences in susceptibility to

EtOH developmental toxicity become apparent (Fig. 9.2, upper panel; Fig. 9.3): (1) C57BL/6 mice were more susceptible to EtOH structural teratogenesis when exposed in utero on GD 9 to a dose of 4 g/kg, whereas CD-1 mice required a dose of 6 g/kg to exhibit the same spectrum of teratogenesis, with C57BL/6 mice exhibiting a higher EtOH plasma concentration than CD-1 mice suggesting a pharmacokinetic basis for increased susceptibility; (2) C3H mice were completely resistant to EtOH teratogenesis at a dose of 4 g/kg, while C57BL/6 mice were

421 susceptible, despite exhibiting similar plasma concentrations of EtOH, precluding a pharmacokinetic basis for this difference in susceptibility. Interestingly, previous studies comparing hepatic ADH in various strains of mice (Swiss, AKR, CBA, DBA, C57Brcd,

C57BL/6 and C3H) demonstrated that C57BL/6 mice have the highest hepatic ADH activity (~7

μmol NADH/min/g liver), while C3H mice exhibited the lowest activity (~3.8 μmol

NADH/min/g liver). The reported lower ADH activity in C57BL/6 mice compared to C3H mice is not consistent with our studies herein, in which the C57BL/6 and C3H strains exhibited similar plasma EtOH concentrations. The reasons for this discrepancy are unclear, but the equivalent plasma EtOH concentrations in C57BL/6 and C3H strains in our study preclude a pharmacokinetic basis for the strain differences in susceptibility to teratogenicity observed herein, whereby C57BL/6 mice were more susceptible to EtOH teratogenesis than the resistant

C3H strain (Rao et al., 1997). Aside from mice genetically modified to have more or less antioxidative enzymes or DNA repair enzymes, these shifts in the dose-response curve in wild- type mice may reflect strain differences in inductive ability of ROS-producing NOX enzymes, antioxidative enzymes capable of detoxifying ROS such as catalase, superoxide dismutase, glutathione peroxidase and thioredoxins, or differences in DNA repair enzymes, all of which would influence the amount of resulting oxidative damage. Similarly for functional teratogenesis, strain differences in susceptibility were evident (Fig. 9.2, lower panel; Fig. 9.4), and the doses necessary to cause postnatal behavioural deficits were lower than those required to cause structural teratogenesis (Fig. 9.2, lower panel; Fig. 9.5).

From comparisons derived from embryo culture studies herein together with those in the literature, mice are more sensitive than rats to the teratogenic effects of both MeOH and EtOH

(Figs. 1.19, 1.20, 1.21). Additionally, when comparing embryopathies in the same species treated with EtOH or MeOH, both drugs exhibit a similar spectrum of embryopathies, although EtOH

422

Figure 9.2. Summary of structural and behavioural teratogenesis and the dose of ethanol (EtOH) employed across several strains of mice. Upper panel: comparison of structural birth defects across CD-1, C3H and C57BL/6 wild-type (WT) mice, transgenic mice expressing human catalase (hCat) on a C57BL/6 background, and catalase deficient mice (aCat) on a C3H background, treated with EtOH during the embryonic period. Lower panel: comparison of postnatal behavioural deficits between CD-1 mice and Ogg1 +/+ mice on a 129SV/C57BL/6 background treated with EtOH during the fetal period.

423

Figure 9.3. Summary of susceptibility to structural teratogenesis across various strains of mice exposed to ethanol (EtOH) in utero during the embryonic period. On gestational day (GD) 9 (plug = GD 1), dams of various strains were dosed with EtOH (4-6 g/kg i.p) and assessed for structural teratogenesis on GD 19. Mouse strains: CD-1, C3H and C57BL/6 wild-type (WT) mice, transgenic mice expressing human catalase (hCat) on a C57BL/6 background, and catalase deficient mice (aCat) on a C3H background, and Ogg1 +/+ mice on a 129SV/C57BL/6 background .

424

Figure 9.4. Summary of susceptibility to postnatal behavioural teratogenesis across various strains of mice exposed to EtOH in utero during the fetal period. On GD 17 (plug = GD 1), pregnant CD-1 or Ogg1 +/- mice, the latter bred to ogg1 +/- males, were dosed with EtOH (2-4g/kg i.p), and offspring were assessed postnatally for behavioural deficits. Mouse strains: CD-1, and Ogg1 +/+ WT and -/- DNA repair-deficient on a 129SV/C57BL/6 background.

425

Figure 9.5. Comparison of susceptibility of CD-1 embryos to structural birth defects and postnatal neurodevelopmental deficits initiated by in utero exposure to EtOH. Structural defects: CD-1 dams were dosed on GD 9 with EtOH (4 and 6 g/kg i.p) and assessed for structural teratogenesis. Postnatal behavioural deficits: CD-1 dams were dosed on GD 17 with EtOH (4 g/kg i.p) and the offspring were assessed postnatally for behavioural deficits.

426 appears to be more embryotoxic than MeOH, as demonstrated in mice by a greater decrease with

EtOH in anterior neuropore closure, turning and somite development (Fig. 1.22), and to a lesser extent in rats, by a greater decrease with EtOH in crown-rump length, head length, somite development and protein content (Fig. 1.23). However, results from embryo culture may vary somewhat among those employing the technique, as Fig. 1.22 demonstrates, where even the same strain of rat evaluated in different studies from the same laboratory can differ in protein content at a similar developmental age. Although variations in such parameters occur routinely within the litter, this factor should be mitigated by the use of a sufficiently large number of litters, so comparisons of different studies can be difficult even in cases of similar exposures.

The doses of EtOH given to assess teratogenesis (4 g/kg) are comparable to those used in other studies in vivo for C57BL/6 mice (Dong et al., 2010) and for CD-1 mice (Blakley and Scott,

FAS (20-200 mg/dL, 5-47 mM) (Khaole et al., 2004; May et al., 2008), and suggests that the concentration-response curve for at least one mouse strain is shifted to the right compared to humans, such that a higher EtOH dose and concentration are required in a mouse to achieve the same effect observed in humans. Indeed, the dose of 4 g/kg used herein was relatively low for the CD-1 strain, demonstrated by the lack of structural anomalies, which was also observed by

Blakley and coworkers (Blakley and Scott, 1984b).

It is interesting to note that there is a 10-fold difference in the effective concentration of

PBN used in mouse vs. rabbit whole embryo culture. While 0.22 mM PBN is an efficacious concentration in protecting against EtOH and MeOH embryopathies in mouse embryo culture,

427 rabbit embryos require 2.2 mM PBN to protect against thalidomide embryopathies, a concentration that is lethal for mouse embryos. The mechanism of this remarkable species difference is unknown, and may include differences in PBN actions unrelated to free radical spin trapping (Table 1.7). In vivo, a dose of 40 mg/kg PBN is effective in protecting against EtOH teratogenicity in both mice and rabbits (Miller et al., 2013b; Parman et al., 1999), while doses of

75 mg/kg and 100 mg/kg are maternally toxic, as evidenced by maternal sedation and complete litter resorptions in mice. Further, DPI was efficacious against MeOH embryopathies in mouse embryo culture at a concentration of 0.5 μM, while increasing concentrations resulted in increased concentration-dependent toxicity. Given that DPI exhibits a broad array of alternative effects involving the pentose phosphate pathway and glucose metabolism, it is conceivable that the toxicity at higher concentrations arose from these alternative effects.

These observations taken together highlight the importance of redox balance and DNA repair within the embryo, and show that embryotoxicity can result from: (1) physiological levels of embryonic ROS in the absence of drug exposure in animals with deficient DNA repair; (2) excessive ROS caused by drugs like EtOH and MeOH; and, (3) excessive levels of normally protective antioxidative agents.

428

9.2 Future Studies

NOX enzymes have been implicated in the developmental toxicity of EtOH and MeOH.

Herein, using DPI, a chemical inhibitor of NOX, and measuring p22phox mRNA and protein expression, we have implicated NOX in MeOH structural embryopathies, while work from Dong et al., (2010) implicated NOX in EtOH structural embryopathies using DPI and measuring mRNA of several NOX isoforms (1-4, DUOXs 1, 2) as well as their associated regulatory subunits (p22phox, p67phox, Noxo1, Noxa1, Rac1). While these studies are insightful, they do not directly implicate NOX involvement. These studies demonstrate associations rather than causal mechanisms. DPI may have alternative effects that are not related to the intended pharmacologic use of the probe. Furthermore, my studies implicate NOX in MeOH embryopathies through enhanced p22phox mRNA expression, but p22phox is required for the activation of NOXs 1-4 but not 5. It would be useful to employ NOX knockout mice to investigate mechanisms of EtOH or MeOH teratogenesis. The breadth of knockout mice available would provide a higher level of rigor as to which NOX is specifically in the mechanism. Once identified, modulating or restoring levels of the particular NOX that is absent would demonstrate a causal relationship.

In utero EtOH exposure results in characteristic craniofacial malformations that are consistent in humans and rodent models of FASD. It is understood that neural crest cells are responsible for development of the face, and have been shown to be sensitive to oxidative stress

(Chen et al., 2000). Understanding the NOX isoform ontogeny within neural crest cell subtypes in affected vs. non-affected offspring, as well as understanding the mechanism of activation,

429 would provide a valuable insight into whether specific NOX isoforms are responsible for susceptibility or resistance, and may provide a basis for predicting toxicity.

Studies herein have demonstrated that PBN can offer protection in a variety of different models of developmental toxicity including: (1) postnatal behavioural deficits and fetal brain

DNA oxidation resulting from EtOH exposure during the fetal period in CD-1 mice, as well as in ogg1 knockout mice; (2) EtOH- and MeOH-initiated structural embryopathies in whole embryo culture; (3) MeOH-initiated embryonic protein oxidation; and, (4) EtOH-initiated structural teratogenesis resulting from EtOH exposure during the embryonic period in CD-1 mice. Other studies have also demonstrated that PBN can offer protection in models of aging, and neurodevelopmental deficits such as Parkinson’s disease and Alzheimer’s disease.

Understanding how PBN can provide protection in such diverse models of toxicity is an interesting question. It would be useful to employ a microarray analysis of embryonic, fetal and postnatal brain tissue pretreated with PBN to determine exactly which genes PBN is modulating in these different models to offer protection. If a common gene, or set of genes, is involved, they can be targeted therapeutically to mitigate the effects underlying toxicity.

Herein I investigated the contribution of embryonic catalase to EtOH and MeOH developmental toxicity using acatalasemic mice and transgenic human catalase-expressing mice with partial reductions and enhancements in catalase activity, respectively. While a catalase knockout mouse exists (Ho et al., 2004), no developmental studies have been performed to date.

Clinically, the human equivalents of catalase knockouts are unlikely, so results from teratological studies performed in catalase null mice would be to provide mechanistic insight, as opposed to mimicking the human condition. While herein we demonstrated aCat mice have 50% baseline resorptions, catalase null mice develop normally, and are fertile. It is possible that this is due to

430 background strain differences, as the catalase null mice were generated on a C57BL/6 background, and this strain does not exhibit a high rate of spontaneous resorptions. Similarly, it is possible that the null mice exhibit upregulation of other antioxidative genes and related defenses for survival. Previous studies using transgenic SOD mice have demonstrated that enhanced SOD was protective against EtOH teratogenesis (Wentzel and Eriksson, 2006) however, too much SOD may be embryopathic, as previously demonstrated with phenytoin

(Winn and Wells, 1999). I propose to treat transgenic human catalase-expressing mice with increasing doses of PEG-SOD prior to teratogen exposure to confirm the pairing of these 2 enzymes in protecting against oxidative stress. I hypothesize that the hCat mice can tolerate higher doses of SOD than their C57BL/6 WT controls, and will therefore be protected against teratogen exposure because their enhanced level of catalase will be able to detoxify the excess

H2O2 formed by the enhanced SOD.

Herein, we demonstrate that ogg1 deficiency is a determinant of risk for EtOH embryopathies in whole embryo culture. It would be useful to follow up these findings with in vivo teratological studies in ogg1 knockout mice to determine whether the observations in culture are reflective of those in vivo. Conversely, it would be useful to employ transgenic ogg1 mice that overexpress OGG1 to determine whether enhanced ability to repair the oxidative 8-oxodGuo lesion confers protection against EtOH- or MeOH-initiated teratogenesis. The phenotype demonstrated herein in untreated ogg1 knockout mice is interesting, as this is the first report of a learning and memory deficit in these mice. It would be informative to perform immunohistochemistry in the brains of ogg1 progeny exposed in utero to saline or EtOH to determine what neuronal changes may be implicated in the phenotype of learning and memory in these mice, as well as EtOH-initiated changes.

431

Chapter 10 REFERENCES

Abbott, B. D., Ebron-McCoy, M. and Andrews, J. E. (1995). Cell death in rat and mouse embryos exposed to methanol in whole embryo culture. Toxicology 97, 159-71.

Abel, E. L. (1982). In utero alcohol exposure and developmental delay of response inhibition. Alcohol Clin Exp Res 6, 369-76.

Abraham, J., Balbo, S., Crabb, D. and Brooks, P. J. (2011). Alcohol metabolism in human cells causes DNA damage and activates the Fanconi anemia-breast cancer susceptibility (FA-BRCA) DNA damage response network. Alcohol Clin Exp Res 35, 2113-20.

Abramov, J. P. and Wells, P. G. (2011a). Embryonic catalase protects against endogenous and phenytoin-enhanced DNA oxidation and embryopathies in acatalasemic and human catalase-expressing mice. FASEB J 25, 2188-200.

Abramov, J. P. and Wells, P. G. (2011b). Embryoprotective role of endogenous catalase in acatalasemic and human catalase-expressing mouse embryos exposed in culture to developmental and phenytoin-enhanced oxidative stress. Toxicol Sci 120, 428-38.

Abuchowski, A., McCoy, J. R., Palczuk, N. C., van Es, T. and Davis, F. F. (1977a). Effect of covalent attachment of polyethylene glycol on immunogenicity and circulating life of bovine liver catalase. J Biol Chem 252, 3582-6.

Abuchowski, A., van Es, T., Palczuk, N. C. and Davis, F. F. (1977b). Alteration of immunological properties of bovine serum albumin by covalent attachment of polyethylene glycol. J Biol Chem 252, 3578-81.

Aburatani, H., Hippo, Y., Ishida, T., Takashima, R., Matsuba, C., Kodama, T., Takao, M., Yasui, A., Yamamoto, K. and Asano, M. (1997). Cloning and characterization of mammalian 8-hydroxyguanine-specific DNA glycosylase/apurinic, apyrimidinic lyase, a functional mutM homologue. Cancer Res 57, 2151-6.

Aebi, H., Suter, H. and Feinstein, R. N. (1968). Activity and stability of catalase in blood and tissues of normal and acatalasemic mice. Biochem Genet 2, 245-51.

Ago, T., Kuribayashi, F., Hiroaki, H., Takeya, R., Ito, T., Kohda, D. and Sumimoto, H. (2003). Phosphorylation of p47phox directs phox homology domain from SH3 domain toward phosphoinositides, leading to phagocyte NADPH oxidase activation. Proc Natl Acad Sci U S A 100, 4474-9.

Ago, T., Kuroda, J., Pain, J., Fu, C., Li, H. and Sadoshima, J. (2010). Upregulation of Nox4 by hypertrophic stimuli promotes apoptosis and mitochondrial dysfunction in cardiac myocytes. Circ Res 106, 1253-64.

AICR (2007). Food, Nutrition, Physical Activity, and the Prevention of Cancer: a Global Perspective., Ed.^ Eds.). World Cancer Research Fund / American Institute for Cancer Research, Washington DC.

Alaluusua, S. and Lukinmaa, P. L. (2006). Developmental dental toxicity of dioxin and related compounds--a review. Int Dent J 56, 323-31.

Aldinger, K. A., Sokoloff, G., Rosenberg, D. M., Palmer, A. A. and Millen, K. J. (2009). Genetic variation and population substructure in outbred CD-1 mice: implications for genome-wide association studies. PLoS One 4, e4729.

Algar, E. M., Seeley, T. L. and Holmes, R. S. (1983). Purification and molecular properties of mouse alcohol dehydrogenase isozymes. Eur J Biochem 137, 139-47.

Allansson, L., Khatibi, S., Olsson, T. and Hansson, E. (2001). Acute ethanol exposure induces [Ca2+]i transients, cell swelling and transformation of actin cytoskeleton in astroglial primary cultures. J Neurochem 76, 472-9.

432

Allen, R. G. and Tresini, M. (2000). Oxidative stress and gene regulation. Free Radic Biol Med 28, 463-99.

Alwan, S., Polifka, J. E. and Friedman, J. M. (2005). Angiotensin II receptor antagonist treatment during pregnancy. Birth Defects Res A Clin Mol Teratol 73, 123-30.

Ambani, L. M., Van Woert, M. H. and Murphy, S. (1975). Brain peroxidase and catalase in Parkinson disease. Arch Neurol 32, 114-8.

Ames, B. N. (1989). Mutagenesis and carcinogenesis: endogenous and exogenous factors. Environ Mol Mutagen 14 Suppl 16, 66-77.

Ammon, E., Schafer, C., Hofmann, U. and Klotz, U. (1996). Disposition and first-pass metabolism of ethanol in humans: is it gastric or hepatic and does it depend on gender? Clin Pharmacol Ther 59, 503-13.

Anderson, C., Andersson, T. and Molander, M. (1991). Ethanol absorption across human skin measured by in vivo microdialysis technique. Acta Derm Venereol 71, 389-93.

Andrews, J. E., Ebron-McCoy, M., Logsdon, T. R., Mole, L. M., Kavlock, R. J. and Rogers, J. M. (1993). Developmental toxicity of methanol in whole embryo culture: a comparative study with mouse and rat embryos. Toxicology 81, 205-15.

Andrews, J. E., Ebron-McCoy, M., Kavlock, R. J. and Rogers, J. M. (1995). Developmental toxicity of formate and formic acid in whole embryo culture: a comparative study with mouse and rat embryos. Teratology. 51, 243-51.

Anylian, G. H., Dorn, J. and Swerdlow, J. (1978). The manifestations, aetiology and assessment of ethanol-induced hangover. S Afr Med J 54, 193-8.

Aoto, K., Shikata, Y., Higashiyama, D., Shiota, K. and Motoyama, J. (2008). Fetal ethanol exposure activates protein kinase A and impairs Shh expression in prechordal mesendoderm cells in the pathogenesis of holoprosencephaly. Birth Defects Res A Clin Mol Teratol 82, 224-31.

Apaja, M. (1980). Evaluation of toxicity and carcinogenicity of malonaldehyde. Acta Universitatis Ouluensis 55, 28-59.

Aragon, C. M., Pesold, C. N. and Amit, Z. (1992). Ethanol-induced motor activity in normal and acatalasemic mice. Alcohol 9, 207-11.

Arai, K., Morishita, K., Shinmura, K., Kohno, T., Kim, S. R., Nohmi, T., Taniwaki, M., Ohwada, S. and Yokota, J. (1997). Cloning of a human homolog of the yeast OGG1 gene that is involved in the repair of oxidative DNA damage. Oncogene 14, 2857-61.

Arai, T., Kelly, V. P., Minowa, O., Noda, T. and Nishimura, S. (2002). High accumulation of oxidative DNA damage, 8-hydroxyguanine, in Mmh/Ogg1 deficient mice by chronic oxidative stress. Carcinogenesis 23, 2005-10.

Archibald, S. L., Fennema-Notestine, C., Gamst, A., Riley, E. P., Mattson, S. N. and Jernigan, T. L. (2001). Brain dysmorphology in individuals with severe prenatal alcohol exposure. Dev Med Child Neurol 43, 148-54.

Arlen, R. R. and Wells, P. G. (1996). Inhibition of thalidomide teratogenicity by acetylsalicylic acid: evidence for prostaglandin H synthase-catalyzed bioactivation of thalidomide to a teratogenic reactive intermediate. J Pharmacol Exp Ther 277, 1649-58.

Arner, E. S. and Holmgren, A. (2000). Physiological functions of thioredoxin and thioredoxin reductase. Eur J Biochem 267, 6102-9.

Arthur, M. J., Lee, A. and Wright, R. (1984). Sex differences in the metabolism of ethanol and acetaldehyde in normal subjects. Clin Sci (Lond) 67, 397-401.

433

Assari, T. (2006). Chronic Granulomatous Disease; fundamental stages in our understanding of CGD. Med Immunol 5, 4.

Baan, R., Straif, K., Grosse, Y., Secretan, B., El Ghissassi, F., Bouvard, V., Altieri, A. and Cogliano, V. (2007). Carcinogenicity of alcoholic beverages. Lancet Oncol 8, 292-3.

Babior, B. M., Kipnes, R. S. and Curnutte, J. T. (1973). Biological defense mechanisms. The production by leukocytes of superoxide, a potential bactericidal agent. J Clin Invest 52, 741-4.

Badger, T. M., Hidestrand, M., Shankar, K., McGuinn, W. D. and Ronis, M. J. (2005). The effects of pregnancy on ethanol clearance. Life Sci 77, 2111-26.

Baehner, R. L., Millar-Groff, S. and Bringas, P. (1999). Developmental expression of NADPH phagocytic oxidase components in mouse embryos. Pediatr Res 46, 152-7.

Baeuerle, P. A. and Baltimore, D. (1996). NF-kappa B: ten years after. Cell 87, 13-20.

Banfi, B., Molnar, G., Maturana, A., Steger, K., Hegedus, B., Demaurex, N. and Krause, K. H. (2001). A Ca(2+)- activated NADPH oxidase in testis, spleen, and lymph nodes. J Biol Chem 276, 37594-601.

Banfi, B., Clark, R. A., Steger, K. and Krause, K. H. (2003). Two novel proteins activate superoxide generation by the NADPH oxidase NOX1. J Biol Chem 278, 3510-3.

Banfi, B., Malgrange, B., Knisz, J., Steger, K., Dubois-Dauphin, M. and Krause, K. H. (2004a). NOX3, a superoxide-generating NADPH oxidase of the inner ear. J Biol Chem 279, 46065-72.

Banfi, B., Tirone, F., Durussel, I., Knisz, J., Moskwa, P., Molnar, G. Z., Krause, K. H. and Cox, J. A. (2004b). Mechanism of Ca2+ activation of the NADPH oxidase 5 (NOX5). J Biol Chem 279, 18583-91.

Baraona, E., Abittan, C. S., Dohmen, K., Moretti, M., Pozzato, G., Chayes, Z. W., Schaefer, C. and Lieber, C. S. (2001). Gender differences in pharmacokinetics of alcohol. Alcohol Clin Exp Res 25, 502-7.

Barrow, M. V. and Taylor, W. J. (1969). A rapid method for detecting malformations in rat fetuses. J Morphol 127, 291-305.

Baumbach, G. L., Cancilla, P. A., Martin-Amat, G., Tephly, T. R., McMartin, K. E., Makar, A. B., Hayreh, M. S. and Hayreh, S. S. (1977). Methyl alcohol poisoning. IV. Alterations of the morphological findings of the retina and optic nerve. Archives of ophthalmology 95, 1859-65.

Bearer, C. F., Swick, A. R., O'Riordan, M. A. and Cheng, G. (1999). Ethanol inhibits L1-mediated neurite outgrowth in postnatal rat cerebellar granule cells. J Biol Chem 274, 13264-70.

Bearer, C. F. (2001). L1 cell adhesion molecule signal cascades: targets for ethanol developmental neurotoxicity. Neurotoxicology 22, 625-33.

Becker, H. C. and Randall, C. L. (1989). Effects of prenatal ethanol exposure in C57BL mice on locomotor activity and passive avoidance behavior. Psychopharmacology (Berl) 97, 40-4.

Becker, H. C., Diaz-Granados, J. L. and Randall, C. L. (1996). Teratogenic actions of ethanol in the mouse: a minireview. Pharmacol Biochem Behav 55, 501-13.

Beckman, J. S., Minor, R. L., Jr., White, C. W., Repine, J. E., Rosen, G. M. and Freeman, B. A. (1988). Superoxide dismutase and catalase conjugated to polyethylene glycol increases endothelial enzyme activity and oxidant resistance. J Biol Chem 263, 6884-92.

434

BelAiba, R. S., Djordjevic, T., Petry, A., Diemer, K., Bonello, S., Banfi, B., Hess, J., Pogrebniak, A., Bickel, C. and Gorlach, A. (2007). NOX5 variants are functionally active in endothelial cells. Free Radic Biol Med 42, 446-59.

Bell, G. I., Najarian, R. C., Mullenbach, G. T. and Hallewell, R. A. (1986). cDNA sequence coding for human kidney catalase. Nucleic Acids Res 14, 5561-2.

Bentley, D., Selfridge, J., Millar, J. K., Samuel, K., Hole, N., Ansell, J. D. and Melton, D. W. (1996). DNA ligase I is required for fetal liver erythropoiesis but is not essential for mammalian cell viability. Nat Genet 13, 489-91.

Berlett, B. S. and Stadtman, E. R. (1997). Protein oxidation in aging, disease, and oxidative stress. J Biol Chem 272, 20313-6.

Bevilacqua, E., Gomes, S. Z., Lorenzon, A. R., Hoshida, M. S. and Amarante-Paffaro, A. M. (2012). NADPH oxidase as an important source of reactive oxygen species at the mouse maternal-fetal interface: putative biological roles. Reprod Biomed Online 25, 31-43.

Bhandarkar, S. S., Jaconi, M., Fried, L. E., Bonner, M. Y., Lefkove, B., Govindarajan, B., Perry, B. N., Parhar, R., Mackelfresh, J., Sohn, A., Stouffs, M., Knaus, U., Yancopoulos, G., Reiss, Y., Benest, A. V., Augustin, H. G. and Arbiser, J. L. (2009). Fulvene-5 potently inhibits NADPH oxidase 4 and blocks the growth of endothelial tumors in mice. J Clin Invest 119, 2359-65.

Bhave, S. V. and Hoffman, P. L. (1997). Ethanol promotes apoptosis in cerebellar granule cells by inhibiting the trophic effect of NMDA. J Neurochem 68, 578-86.

Bhuller, Y. and Wells, P. G. (2006). A developmental role for ataxia-telangiectasia mutated in protecting the embryo from spontaneous and phenytoin-enhanced embryopathies in culture. Toxicol Sci 93, 156-63.

Biberstine-Kinkade, K. J., DeLeo, F. R., Epstein, R. I., LeRoy, B. A., Nauseef, W. M. and Dinauer, M. C. (2001). Heme-ligating histidines in flavocytochrome b(558): identification of specific histidines in gp91(phox). J Biol Chem 276, 31105-12.

Bjoras, M., Luna, L., Johnsen, B., Hoff, E., Haug, T., Rognes, T. and Seeberg, E. (1997). Opposite base-dependent reactions of a human base excision repair enzyme on DNA containing 7,8-dihydro-8-oxoguanine and abasic sites. EMBO J 16, 6314-22.

Black, K. A., Eells, J. T., Noker, P. E., Hawtrey, C. A. and Tephly, T. R. (1985). Role of hepatic tetrahydrofolate in the species difference in methanol toxicity. Proc Natl Acad Sci U S A 82, 3854-8.

Blakley, P. M. and Scott, W. J., Jr. (1984a). Determination of the proximate teratogen of the mouse fetal alcohol syndrome. 2. Pharmacokinetics of the placental transfer of ethanol and acetaldehyde. Toxicol Appl Pharmacol 72, 364-71.

Blakley, P. M. and Scott, W. J., Jr. (1984b). Determination of the proximate teratogen of the mouse fetal alcohol syndrome. 1. Teratogenicity of ethanol and acetaldehyde. Toxicol Appl Pharmacol 72, 355-63.

Blanchard, B. A., Riley, E. P. and Hannigan, J. H. (1987). Deficits on a spatial navigation task following prenatal exposure to ethanol. Neurotoxicol Teratol 9, 253-8.

Blomstrand, R. and Ingemansson, S. O. (1984). Studies on the effect of 4-methylpyrazole on methanol poisoning using the monkey as an animal model: with particular reference to the ocular toxicity. Drug Alcohol Depend 13, 343-55.

Bloom, F., Lad, P., Pittman, Q. and Rogers, J. (1982). Blood alcohol levels in rats: non-uniform yields from intraperitoneal doses based on body weight. Br J Pharmacol 75, 251-4.

Boddupalli, S., Mein, J. R., Lakkanna, S. and James, D. R. (2012). Induction of phase 2 antioxidant enzymes by broccoli sulforaphane: perspectives in maintaining the antioxidant activity of vitamins a, C, and e. Front Genet 3, 7.

435

Boehm, S. L., 2nd, Lundahl, K. R., Caldwell, J. and Gilliam, D. M. (1997). Ethanol teratogenesis in the C57BL/6J, DBA/2J, and A/J inbred mouse strains. Alcohol 14, 389-95.

Boiteux, S., O'Connor, T. R. and Laval, J. (1987). Formamidopyrimidine-DNA glycosylase of Escherichia coli: cloning and sequencing of the fpg structural gene and overproduction of the protein. EMBO J 6, 3177-83.

Boiteux, S. and Radicella, J. P. (2000). The human OGG1 gene: structure, functions, and its implication in the process of carcinogenesis. Arch Biochem Biophys 377, 1-8.

Bolon, B., Dorman, D. C., Janszen, D., Morgan, K. T. and Welsch, F. (1993). Phase-specific developmental toxicity in mice following maternal methanol inhalation. Fundam Appl Toxicol 21, 508-16.

Bolon, B., Welsch, F. and Morgan, K. T. (1994). Methanol-induced neural tube defects in mice: pathogenesis during neurulation. Teratology 49, 497-517.

Boni, L. T., Hah, J. S., Hui, S. W., Mukherjee, P., Ho, J. T. and Jung, C. Y. (1984). Aggregation and fusion of unilamellar vesicles by poly(ethylene glycol). Biochim Biophys Acta 775, 409-18.

Bonthius, D. J., Karacay, B., Dai, D. and Pantazis, N. J. (2003). FGF-2, NGF and IGF-1, but not BDNF, utilize a nitric oxide pathway to signal neurotrophic and neuroprotective effects against alcohol toxicity in cerebellar granule cell cultures. Brain Res Dev Brain Res 140, 15-28.

Bonthius, D. J., Karacay, B., Dai, D., Hutton, A. and Pantazis, N. J. (2004). The NO-cGMP-PKG pathway plays an essential role in the acquisition of ethanol resistance by cerebellar granule neurons. Neurotoxicol Teratol 26, 47-57.

Boveris, A., Oshino, N. and Chance, B. (1972). The cellular production of hydrogen peroxide. Biochem J 128, 617- 30.

Bradford, B. U., Seed, C. B., Handler, J. A., Forman, D. T. and Thurman, R. G. (1993). Evidence that catalase is a major pathway of ethanol oxidation in vivo: dose-response studies in deer mice using methanol as a selective substrate. Arch Biochem Biophys 303, 172-6.

Brien, J. F., Loomis, C. W., Tranmer, J. and McGrath, M. (1983). Disposition of ethanol in human maternal venous blood and amniotic fluid. Am J Obstet Gynecol 146, 181-6.

Brocardo, P. S., Gil-Mohapel, J. and Christie, B. R. (2011). The role of oxidative stress in fetal alcohol spectrum disorders. Brain Res Rev 67, 209-25.

Brocardo, P. S., Boehme, F., Patten, A., Cox, A., Gil-Mohapel, J. and Christie, B. R. (2012). Anxiety- and depression-like behaviors are accompanied by an increase in oxidative stress in a rat model of fetal alcohol spectrum disorders: Protective effects of voluntary physical exercise. Neuropharmacology 62, 1607-18.

Brooks, P. J., Enoch, M. A., Goldman, D., Li, T. K. and Yokoyama, A. (2009). The alcohol flushing response: an unrecognized risk factor for esophageal cancer from alcohol consumption. PLoS Med 6, e50.

Brown, D. I. and Griendling, K. K. (2009). Nox proteins in signal transduction. Free Radic Biol Med 47, 1239-53.

Brown, N. A., Goulding, E. H. and Fabro, S. (1979). Ethanol embryotoxicity: direct effects on mammalian embryos in vitro. Science 206, 573-5.

Bruner, S. D., Norman, D. P. and Verdine, G. L. (2000). Structural basis for recognition and repair of the endogenous mutagen 8-oxoguanine in DNA. Nature 403, 859-66.

Buhler, R., Hempel, J., Kaiser, R., von Wartburg, J. P., Vallee, B. L. and Jornvall, H. (1984). Human alcohol dehydrogenase: structural differences between the beta and gamma subunits suggest parallel duplications in

436 isoenzyme evolution and predominant expression of separate gene descendants in livers of different mammals. Proc Natl Acad Sci U S A 81, 6320-4.

Bunsey, M. and Eichenbaum, H. (1996). Conservation of hippocampal memory function in rats and humans. Nature 379, 255-7.

Burbacher, T., Shen, D., Grant, K., Sheppard, L., Damian, D., Ellis, S. and Liberato, N. (1999). Reproductive and offspring developmental effects following maternal inhalation exposure to methanol in nonhuman primates. Research report (Health Effects Institute), i-ii, 1-117; discussion 119-33.

Busby, A., La Grange, L., Edwards, J. and King, J. (2002). The use of a silymarin/phospholipid compound as a fetoprotectant from ethanol-induced behavioral deficits. J Herb Pharmacother 2, 39-47.

Cadenas, E. and Davies, K. J. (2000). Mitochondrial free radical generation, oxidative stress, and aging. Free Radic Biol Med 29, 222-30.

Campbell, M. A. and Fantel, A. G. (1983). Teratogenicity of acetaldehyde in vitro: relevance to the fetal alcohol syndrome. Life Sci 32, 2641-7.

Carney, E. W., Tornesi, B., Markham, D. A., Rasoulpour, R. J. and Moore, N. (2008). Species-specificity of ethylene glycol-induced developmental toxicity: toxicokinetic and whole embryo culture studies in the rabbit. Birth defects research 83, 573-81.

Carney, J. M. and Floyd, R. A. (1991). Protection against oxidative damage to CNS by alpha-phenyl-tert-butyl nitrone (PBN) and other spin-trapping agents: a novel series of nonlipid free radical scavengers. J Mol Neurosci 3, 47-57.

Cartwright, M. M., Tessmer, L. L. and Smith, S. M. (1998). Ethanol-induced neural crest apoptosis is coincident with their endogenous death, but is mechanistically distinct. Alcohol Clin Exp Res 22, 142-9.

Caselli, A., Marzocchini, R., Camici, G., Manao, G., Moneti, G., Pieraccini, G. and Ramponi, G. (1998). The inactivation mechanism of low molecular weight phosphotyrosine-protein phosphatase by H2O2. J Biol Chem 273, 32554-60.

Casp, C. B., She, J. X. and McCormack, W. T. (2002). Genetic association of the catalase gene (CAT) with vitiligo susceptibility. Pigment Cell Res 15, 62-6.

Castro, G. D., Costantini, M. H., Delgado de Layno, A. M. and Castro, J. A. (2002). Rat liver microsomal and nuclear activation of methanol to hydroxymethyl free radicals. Toxicol Lett 129, 227-36.

Cave, A. C., Brewer, A. C., Narayanapanicker, A., Ray, R., Grieve, D. J., Walker, S. and Shah, A. M. (2006). NADPH oxidases in cardiovascular health and disease. Antioxid Redox Signal 8, 691-728.

Cayatte, A. J., Rupin, A., Oliver-Krasinski, J., Maitland, K., Sansilvestri-Morel, P., Boussard, M. F., Wierzbicki, M., Verbeuren, T. J. and Cohen, R. A. (2001). S17834, a new inhibitor of cell adhesion and atherosclerosis that targets nadph oxidase. Arterioscler Thromb Vasc Biol 21, 1577-84.

Cederbaum, A. I. and Qureshi, A. (1982). Role of catalase and hydroxyl radicals in the oxidation of methanol by rat liver microsomes. Biochem Pharmacol 31, 329-35.

Cederberg, J. and Eriksson, U. J. (1997). Decreased catalase activity in malformation-prone embryos of diabetic rats. Teratology 56, 350-7.

Cederberg, J., Galli, J., Luthman, H. and Eriksson, U. J. (2000). Increased mRNA levels of Mn-SOD and catalase in embryos of diabetic rats from a malformation-resistant strain. Diabetes 49, 101-7.

437

Chang, X., Fan, Y., Karyala, S., Schwemberger, S., Tomlinson, C. R., Sartor, M. A. and Puga, A. (2007). Ligand- independent regulation of transforming growth factor beta1 expression and cell cycle progression by the aryl hydrocarbon receptor. Mol Cell Biol 27, 6127-39.

Chang, Y. S., Kim, Y. J., Yoo, H. S., Sung, D. K., Kim, S. Y., Kang, S. and Park, W. S. (2009). Alpha-phenyl-N- tert-butylnitrone attenuates hyperoxia-induced lung injury by down-modulating inflammation in neonatal rats. Exp Lung Res 35, 234-49.

Charness, M. E., Safran, R. M. and Perides, G. (1994). Ethanol inhibits neural cell-cell adhesion. J Biol Chem 269, 9304-9.

Chase, H. B. (1942). Studies on an Anophthalmic Strain of Mice. III. Results of Crosses with Other Strains. Genetics 27, 339-48.

Chen, C. S. and Wells, P. G. (2006). Enhanced tumorigenesis in p53 knockout mice exposed in utero to high-dose vitamin E. Carcinogenesis 27, 1358-68.

Chen, C. S., Squire, J. A. and Wells, P. G. (2009a). Reduced tumorigenesis in p53 knockout mice exposed in utero to low-dose vitamin E. Cancer 115, 1563-75.

Chen, D., Cao, G., Hastings, T., Feng, Y., Pei, W., O'Horo, C. and Chen, J. (2002). Age-dependent decline of DNA repair activity for oxidative lesions in rat brain mitochondria. J Neurochem 81, 1273-84.

Chen, K., Craige, S. E. and Keaney, J. F., Jr. (2009b). Downstream targets and intracellular compartmentalization in Nox signaling. Antioxid Redox Signal 11, 2467-80.

Chen, S. Y. and Sulik, K. K. (1996). Free radicals and ethanol-induced cytotoxicity in neural crest cells. Alcohol Clin Exp Res 20, 1071-6.

Chen, S. Y., Periasamy, A., Yang, B., Herman, B., Jacobson, K. and Sulik, K. K. (2000). Differential sensitivity of mouse neural crest cells to ethanol-induced toxicity. Alcohol 20, 75-81.

Chen, S. Y., Dehart, D. B. and Sulik, K. K. (2004a). Protection from ethanol-induced limb malformations by the superoxide dismutase/catalase mimetic, EUK-134. Faseb J 18, 1234-6.

Chen, W., Shang, W. H., Adachi, Y., Hirose, K., Ferrari, D. M. and Kamata, T. (2008). A possible biochemical link between NADPH oxidase (Nox) 1 redox-signalling and ERp72. Biochem J 416, 55-63.

Chen, X., Mele, J., Giese, H., Van Remmen, H., Dolle, M. E., Steinhelper, M., Richardson, A. and Vijg, J. (2003). A strategy for the ubiquitous overexpression of human catalase and CuZn superoxide dismutase genes in transgenic mice. Mech Ageing Dev 124, 219-27.

Chen, X., Liang, H., Van Remmen, H., Vijg, J. and Richardson, A. (2004b). Catalase transgenic mice: characterization and sensitivity to oxidative stress. Arch Biochem Biophys 422, 197-210.

Cheng, G., Cao, Z., Xu, X., van Meir, E. G. and Lambeth, J. D. (2001). Homologs of gp91phox: cloning and tissue expression of Nox3, Nox4, and Nox5. Gene 269, 131-40.

Chernoff, G. F. (1977). The fetal alcohol syndrome in mice: an animal model. Teratology 15, 223-9.

Chernoff, G. F. (1980). The fetal alcohol syndrome in mice: maternal variables. Teratology 22, 71-5.

Chesire, D. R., Dunn, T. A., Ewing, C. M., Luo, J. and Isaacs, W. B. (2004). Identification of aryl hydrocarbon receptor as a putative Wnt/beta-catenin pathway target gene in prostate cancer cells. Cancer Res 64, 2523-33.

438

Choi, S. H., Lee, D. Y., Chung, E. S., Hong, Y. B., Kim, S. U. and Jin, B. K. (2005). Inhibition of thrombin-induced microglial activation and NADPH oxidase by minocycline protects dopaminergic neurons in the substantia nigra in vivo. J Neurochem 95, 1755-65.

Choudhury, S., Pan, J., Amin, S., Chung, F. L. and Roy, R. (2004). Repair kinetics of trans-4-hydroxynonenal- induced cyclic 1,N2-propanodeoxyguanine DNA adducts by human cell nuclear extracts. Biochemistry 43, 7514-21.

Christmann, M., Tomicic, M. T., Roos, W. P. and Kaina, B. (2003). Mechanisms of human DNA repair: an update. Toxicology 193, 3-34.

Chu, J., Tong, M. and de la Monte, S. M. (2007). Chronic ethanol exposure causes mitochondrial dysfunction and oxidative stress in immature central nervous system neurons. Acta Neuropathol 113, 659-73.

Clary, J. J. (2003). Methanol, is it a developmental risk to humans? Regul Toxicol Pharmacol 37, 83-91.

Cole-Harding, S. and Wilson, J. R. (1987). Ethanol metabolism in men and women. J Stud Alcohol 48, 380-7.

Coles, C. D., Goldstein, F. C., Lynch, M. E., Chen, X., Kable, J. A., Johnson, K. C. and Hu, X. (2011). Memory and brain volume in adults prenatally exposed to alcohol. Brain Cogn 75, 67-77.

Cook, R. J., Champion, K. M. and Giometti, C. S. (2001). Methanol toxicity and formate oxidation in NEUT2 mice. Arch Biochem Biophys 393, 192-8.

Coon, M. J. and Koop, D. R. (1987). Alcohol-inducible cytochrome P-450 (P-450ALC). Archives of toxicology 60, 16-21.

Copp, A. J., Brook, F. A., Estibeiro, J. P., Shum, A. S. and Cockroft, D. L. (1990). The embryonic development of mammalian neural tube defects. Prog Neurobiol 35, 363-403.

Coppede, F. (2011). An overview of DNA repair in amyotrophic lateral sclerosis. ScientificWorldJournal 11, 1679- 91.

Cordero, D. R., Brugmann, S., Chu, Y., Bajpai, R., Jame, M. and Helms, J. A. (2011). Cranial neural crest cells on the move: their roles in craniofacial development. Am J Med Genet A 155A, 270-9.

Corrall, R. J., Havre, P., Margolis, J., Kong, M. and Landau, B. R. (1976). Subcellular site of acetaldehyde oxidation in rat liver. Biochem Pharmacol 25, 17-20.

Coso, S., Harrison, I., Harrison, C. B., Vinh, A., Sobey, C. G., Drummond, G. R., Williams, E. D. and Selemidis, S. (2012). NADPH oxidases as regulators of tumor angiogenesis: current and emerging concepts. Antioxid Redox Signal 16, 1229-47.

Couture, L. A., Abbott, B. D. and Birnbaum, L. S. (1990). A critical review of the developmental toxicity and teratogenicity of 2,3,7,8-tetrachlorodibenzo-p-dioxin: recent advances toward understanding the mechanism. Teratology 42, 619-27.

Cranston, A., Bocker, T., Reitmair, A., Palazzo, J., Wilson, T., Mak, T. and Fishel, R. (1997). Female embryonic lethality in mice nullizygous for both Msh2 and p53. Nat Genet 17, 114-8.

Cross, A. R., Parkinson, J. F. and Jones, O. T. (1984). The superoxide-generating oxidase of leucocytes. NADPH- dependent reduction of flavin and cytochrome b in solubilized preparations. Biochem J 223, 337-44.

Cross, A. R. and Jones, O. T. (1986). The effect of the inhibitor diphenylene iodonium on the superoxide-generating system of neutrophils. Specific labelling of a component polypeptide of the oxidase. Biochem J 237, 111-6.

Cruzan, G. (2009). Assessment of the cancer potential of methanol. Critical reviews in toxicology 39, 347-63.

439

Cui, S. J., Tewari, M., Schneider, T. and Rubin, R. (1997). Ethanol promotes cell death by inhibition of the insulin- like growth factor I receptor. Alcohol Clin Exp Res 21, 1121-7.

Cummings, A. M. (1993). Evaluation of the effects of methanol during early pregnancy in the rat. Toxicology 79, 205-14.

De-Carvalho, R. R., Delgado, I. F., Souza, C. A., Chahoud, I. and Paumgartten, F. J. (1994). Embryotoxicity of methanol in well-nourished and malnourished rats. Braz J Med Biol Res 27, 2915-23.

De, A., Boyadjieva, N. I., Pastorcic, M., Reddy, B. V. and Sarkar, D. K. (1994). Cyclic AMP and ethanol interact to control apoptosis and differentiation in hypothalamic beta-endorphin neurons. J Biol Chem 269, 26697-705.

De Deken, X., Wang, D., Many, M. C., Costagliola, S., Libert, F., Vassart, G., Dumont, J. E. and Miot, F. (2000). Cloning of two human thyroid cDNAs encoding new members of the NADPH oxidase family. J Biol Chem 275, 23227-33. de Mendez, I., Homayounpour, N. and Leto, T. L. (1997). Specificity of p47phox SH3 domain interactions in NADPH oxidase assembly and activation. Mol Cell Biol 17, 2177-85. de Souza-Pinto, N. C., Eide, L., Hogue, B. A., Thybo, T., Stevnsner, T., Seeberg, E., Klungland, A. and Bohr, V. A. (2001). Repair of 8-oxodeoxyguanosine lesions in mitochondrial dna depends on the oxoguanine dna glycosylase (OGG1) gene and 8-oxoguanine accumulates in the mitochondrial dna of OGG1-defective mice. Cancer Res 61, 5378-81.

Defoort, E. N., Kim, P. M. and Winn, L. M. (2006). Valproic acid increases conservative homologous recombination frequency and reactive oxygen species formation: a potential mechanism for valproic acid-induced neural tube defects. Mol Pharmacol 69, 1304-10.

Degitz, S. J., Rogers, J. M., Zucker, R. M. and Hunter, E. S., 3rd (2004). Developmental toxicity of methanol: Pathogenesis in CD-1 and C57BL/6J mice exposed in whole embryo culture. Birth Defects Res A Clin Mol Teratol 70, 179-84.

DeLeo, F. R., Nauseef, W. M., Jesaitis, A. J., Burritt, J. B., Clark, R. A. and Quinn, M. T. (1995). A domain of p47phox that interacts with human neutrophil flavocytochrome b558. J Biol Chem 270, 26246-51.

Deltour, L., Ang, H. L. and Duester, G. (1996). Ethanol inhibition of retinoic acid synthesis as a potential mechanism for fetal alcohol syndrome. Faseb J 10, 1050-7.

Denison, M. S. and Nagy, S. R. (2003). Activation of the aryl hydrocarbon receptor by structurally diverse exogenous and endogenous chemicals. Annu Rev Pharmacol Toxicol 43, 309-34.

Dennery, P. A. (2007). Effects of oxidative stress on embryonic development. Birth Defects Res C Embryo Today 81, 155-62.

Desroches, D., Orevillo, C. and Verina, D. (1995). Sex- and strain-related differences in first-pass alcohol metabolism in mice. Alcohol 12, 221-6.

Devi, B. G., Schenker, S., Mazloum, B. and Henderson, G. I. (1996). Ethanol-induced oxidative stress and enzymatic defenses in cultured fetal rat hepatocytes. Alcohol 13, 327-32.

Dianov, G., Price, A. and Lindahl, T. (1992). Generation of single-nucleotide repair patches following excision of uracil residues from DNA. Mol Cell Biol 12, 1605-12.

Diatchuk, V., Lotan, O., Koshkin, V., Wikstroem, P. and Pick, E. (1997). Inhibition of NADPH oxidase activation by 4-(2-aminoethyl)-benzenesulfonyl fluoride and related compounds. J Biol Chem 272, 13292-301.

440

Dikalova, A., Clempus, R., Lassegue, B., Cheng, G., McCoy, J., Dikalov, S., San Martin, A., Lyle, A., Weber, D. S., Weiss, D., Taylor, W. R., Schmidt, H. H., Owens, G. K., Lambeth, J. D. and Griendling, K. K. (2005). Nox1 overexpression potentiates angiotensin II-induced hypertension and vascular smooth muscle hypertrophy in transgenic mice. Circulation 112, 2668-76.

Dikalova, A. E., Gongora, M. C., Harrison, D. G., Lambeth, J. D., Dikalov, S. and Griendling, K. K. (2010). Upregulation of Nox1 in vascular smooth muscle leads to impaired endothelium-dependent relaxation via eNOS uncoupling. Am J Physiol Heart Circ Physiol 299, H673-9.

DiPadova, C., Worner, T. M., Julkunen, R. J. and Lieber, C. S. (1987). Effects of fasting and chronic alcohol consumption on the first-pass metabolism of ethanol. Gastroenterology 92, 1169-73.

Dizdaroglu, M. (2005). Base-excision repair of oxidative DNA damage by DNA glycosylases. Mutat Res 591, 45- 59.

Dobrzynska, I., Skrzydlewska, E. and Figaszewski, Z. (1999). Influence of trolox derivative and N-acetylcysteine on surface charge density of erythrocytes in methanol intoxicated rats. Environ Toxicol Pharmacol 8, 15-21.

Dobrzynska, I., Skrzydlewska, E., Kasacka, I. and Figaszewski, Z. (2000). Protective effect of N-acetylcysteine on rat liver cell membrane during methanol intoxication. J Pharm Pharmacol 52, 547-52.

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 prevention of fetal alcohol spectrum disorders. Antioxid Redox Signal 10, 2023-33.

Dong, J., Sulik, K. K. and Chen, S. Y. (2010). The role of NOX enzymes in ethanol-induced oxidative stress and apoptosis in mouse embryos. Toxicol Lett 193, 94-100.

Dorman, D. C., Bolon, B., Struve, M. F., LaPerle, K. M. D., Wong, B. A., Elswick, B. and Welsch, F. (1995). Role of formate in methanol-induced exencephaly in CD-1 mice. Teratology 52, 30-40.

Downing, C., Balderrama-Durbin, C., Broncucia, H., Gilliam, D. and Johnson, T. E. (2009). Ethanol teratogenesis in five inbred strains of mice. Alcohol Clin Exp Res 33, 1238-45.

Drummond, G. R., Selemidis, S., Griendling, K. K. and Sobey, C. G. (2011). Combating oxidative stress in vascular disease: NADPH oxidases as therapeutic targets. Nat Rev Drug Discov 10, 453-71.

Dudek, R. W. and Fix, J. D. (1998). Embryology. 2nd ed. Williams & Wilkins, Pennsylvania.

Dudek, R. W. (2011). Embryology. 5th ed. Lippincott Williams & Wilkins, Philadelphia.

Duester, G. (1991). A hypothetical mechanism for fetal alcohol syndrome involving ethanol inhibition of retinoic acid synthesis at the alcohol dehydrogenase step. Alcohol Clin Exp Res 15, 568-72.

Ekins, B. R., Rollins, D. E., Duffy, D. P. and Gregory, M. C. (1985). Standardized treatment of severe methanol poisoning with ethanol and hemodialysis. The Western journal of medicine 142, 337-40. el-Hage, S. and Singh, S. M. (1989). Regulation of catalase-specific mRNA and its processing during development in mice. Dev Genet 10, 339-44. 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, 149-59.

Ellson, C., Davidson, K., Anderson, K., Stephens, L. R. and Hawkins, P. T. (2006). PtdIns3P binding to the PX domain of p40phox is a physiological signal in NADPH oxidase activation. EMBO J 25, 4468-78.

441

Espeel, M., Briere, N., De Craemer, D., Jauniaux, E. and Roels, F. (1993). Catalase-negative peroxisomes in human embryonic liver. Cell Tissue Res 272, 89-92.

Espeel, M., Depreter, M., Nardacci, R., D'Herde, K., Kerckaert, I., Stefanini, S. and Roels, F. (1997). Biogenesis of peroxisomes in fetal liver. Microsc Res Tech 39, 453-66.

Esterbauer, H., Schaur, R. J. and Zollner, H. (1991). Chemistry and biochemistry of 4-hydroxynonenal, malonaldehyde and related aldehydes. Free Radic Biol Med 11, 81-128.

Evans, M. D., Dizdaroglu, M. and Cooke, M. S. (2004). Oxidative DNA damage and disease: induction, repair and significance. Mutat Res 567, 1-61.

Fam, S. S. and Morrow, J. D. (2003). The isoprostanes: unique products of arachidonic acid oxidation-a review. Curr Med Chem 10, 1723-40.

Fan, C., Katsuyama, M., Nishinaka, T. and Yabe-Nishimura, C. (2005). Transactivation of the EGF receptor and a PI3 kinase-ATF-1 pathway is involved in the upregulation of NOX1, a catalytic subunit of NADPH oxidase. FEBS Lett 579, 1301-5.

Fan, L. W., Tien, L. T., Mitchell, H. J., Rhodes, P. G. and Cai, Z. (2008). Alpha-phenyl-n-tert-butyl-nitrone ameliorates hippocampal injury and improves learning and memory in juvenile rats following neonatal exposure to lipopolysaccharide. Eur J Neurosci 27, 1475-84.

Fantel, A. G. (1996). Reactive oxygen species in developmental toxicity: review and hypothesis. Teratology 53, 196-217.

Farbiszewski, R., Witek, A. and Skrzydlewska, E. (2000). N-acetylcysteine or trolox derivative mitigate the toxic effects of methanol on the antioxidant system of rat brain. Toxicology 156, 47-55.

Fattoretti, P., Bertoni-Freddari, C., Casoli, T., Di Stefano, G., Giorgetti, G. and Solazzi, M. (2003). Ethanol-induced decrease of the expression of glucose transport protein (Glut3) in the central nervous system as a predisposing condition to apoptosis: the effect of age. Ann N Y Acad Sci 1010, 500-3.

Faulkner, T. P., Cantleberry, S. B., Watts, V. J. and Hussain, A. S. (1990). Comparative pharmacokinetics of ethanol in inbred strains of mice using doses based on total body water. Alcohol Clin Exp Res 14, 82-6.

Feinstein, R. N., Seaholm, J. E., Howard, J. B. and Russell, W. L. (1964). Acatalasemic Mice. Proc Natl Acad Sci U S A 52, 661-2.

Feng, Z., Hu, W. and Tang, M. S. (2004). Trans-4-hydroxy-2-nonenal inhibits nucleotide excision repair in human cells: a possible mechanism for lipid peroxidation-induced carcinogenesis. Proc Natl Acad Sci U S A 101, 8598-602.

Finkel, T. (1998). Oxygen radicals and signaling. Curr Opin Cell Biol 10, 248-53.

Finkelstein, E., Rosen, G. M. and Rauckman, E. J. (1980). Spin trapping of superoxide and hydroxyl radical: practical aspects. Arch Biochem Biophys 200, 1-16.

Fiore, M., Laviola, G., Aloe, L., di Fausto, V., Mancinelli, R. and Ceccanti, M. (2009). Early exposure to ethanol but not red wine at the same alcohol concentration induces behavioral and brain neurotrophin alterations in young and adult mice. Neurotoxicology 30, 59-71.

Flaherty, J. P., Spruce, C. A., Fairfield, H. E. and Bergstrom, D. E. (2010). Generation of a conditional null allele of NADPH oxidase activator 1 (NOXA1). Genesis 48, 568-75.

Floyd, R. A. (1999). Antioxidants, oxidative stress, and degenerative neurological disorders. Proc Soc Exp Biol Med 222, 236-45.

442

Floyd, R. A., Hensley, K., Forster, M. J., Kelleher-Andersson, J. A. and Wood, P. L. (2002). Nitrones, their value as therapeutics and probes to understand aging. Mech Ageing Dev 123, 1021-31.

Fontayne, A., Dang, P. M., Gougerot-Pocidalo, M. A. and El-Benna, J. (2002). Phosphorylation of p47phox sites by PKC alpha, beta II, delta, and zeta: effect on binding to p22phox and on NADPH oxidase activation. Biochemistry 41, 7743-50.

Forte-McRobbie, C. M. and Pietruszko, R. (1985). Aldehyde dehydrogenase content and composition of human liver. Alcohol 2, 375-81.

Fraser, A. D. and MacNeil, W. (1989). Gas chromatographic analysis of methyl formate and application in methanol poisoning cases. Journal of analytical toxicology 13, 73-6.

Fratta, I. D., Sigg, E. B. and Maiorana, K. (1965). Teratogenic effects of thalidomide in rabbits, rats, hamsters, and mice. Toxicology and applied pharmacology 7, 268-86.

Freeman, B. A. and Crapo, J. D. (1982). Biology of disease: free radicals and tissue injury. Lab Invest 47, 412-26.

Frei, B. (1994). Reactive oxygen species and antioxidant vitamins: mechanisms of action. Am J Med 97, 5S-13S; discussion 22S-28S.

Frezza, M., di Padova, C., Pozzato, G., Terpin, M., Baraona, E. and Lieber, C. S. (1990). High blood alcohol levels in women. The role of decreased gastric alcohol dehydrogenase activity and first-pass metabolism. N Engl J Med 322, 95-9.

Fridovich, I. (1995). Superoxide radical and superoxide dismutases. Annu Rev Biochem 64, 97-112.

Frosina, G., Fortini, P., Rossi, O., Carrozzino, F., Raspaglio, G., Cox, L. S., Lane, D. P., Abbondandolo, A. and Dogliotti, E. (1996). Two pathways for base excision repair in mammalian cells. J Biol Chem 271, 9573-8.

Fukui, Y., Naruse, I., Hoshino, K., Kameyama, Y. and Takeuchi, Y. (1990). Strain Difference in Effects of Ethanol Treatment on Fetal Development in C57BL and DBA Mice. Congenital Anomalies 30, 177-186.

Fulton, D. J. (2009). Nox5 and the regulation of cellular function. Antioxid Redox Signal 11, 2443-52.

Gao, H. M. and Hong, J. S. (2008). Why neurodegenerative diseases are progressive: uncontrolled inflammation drives disease progression. Trends Immunol 29, 357-65.

Gao, H. M., Zhou, H. and Hong, J. S. (2012). NADPH oxidases: novel therapeutic targets for neurodegenerative diseases. Trends Pharmacol Sci 33, 295-303.

Gatley, S. J. and Sherratt, H. S. (1976). The effects of diphenyleneiodonium and of 2,4- dichlorodiphenyleneiodonium on mitochondrial reactions. Mechanism of the inhibition of oxygen uptake as a consequence of the catalysis of the chloride/hydroxyl-ion exchange. Biochem J 158, 317-26.

Gavazzi, G., Banfi, B., Deffert, C., Fiette, L., Schappi, M., Herrmann, F. and Krause, K. H. (2006). Decreased blood pressure in NOX1-deficient mice. FEBS Lett 580, 497-504.

Ge, Y., Belcher, S. M. and Light, K. E. (2004). Alterations of cerebellar mRNA specific for BDNF, p75NTR, and TrkB receptor isoforms occur within hours of ethanol administration to 4-day-old rat pups. Brain Res Dev Brain Res 151, 99-109.

Geiszt, M., Kopp, J. B., Varnai, P. and Leto, T. L. (2000). Identification of renox, an NAD(P)H oxidase in kidney. Proc Natl Acad Sci U S A 97, 8010-4.

Geiszt, M., Lekstrom, K., Witta, J. and Leto, T. L. (2003a). Proteins homologous to p47phox and p67phox support superoxide production by NAD(P)H oxidase 1 in colon epithelial cells. J Biol Chem 278, 20006-12.

443

Geiszt, M., Witta, J., Baffi, J., Lekstrom, K. and Leto, T. L. (2003b). Dual oxidases represent novel hydrogen peroxide sources supporting mucosal surface host defense. Faseb J 17, 1502-4.

Gianni, D., Taulet, N., Zhang, H., DerMardirossian, C., Kister, J., Martinez, L., Roush, W. R., Brown, S. J., Bokoch, G. M. and Rosen, H. (2010). A novel and specific NADPH oxidase-1 (Nox1) small-molecule inhibitor blocks the formation of functional invadopodia in human colon cancer cells. ACS Chem Biol 5, 981-93.

Gianoulakis, C. (1990). Rats exposed prenatally to alcohol exhibit impairment in spatial navigation test. Behav Brain Res 36, 217-28.

Giavini, E., Broccia, M. L., Prati, M., Bellomo, D. and Menegola, E. (1992). Effects of ethanol and acetaldehyde on rat embryos developing in vitro. In Vitro Cell Dev Biol 28A, 205-10.

Gil-Mohapel, J., Boehme, F., Kainer, L. and Christie, B. R. (2010). Hippocampal cell loss and neurogenesis after fetal alcohol exposure: insights from different rodent models. Brain Res Rev 64, 283-303.

Gilliam, D., Valdez, N., Branson, S., Dixon, A. and Downing, C. (2011). Maternal effects on ethanol teratogenesis in a cross between A/J and C57BL/6J mice. Alcohol 45, 441-9.

Gilliam, D. M., Stilman, A., Dudek, B. C. and Riley, E. P. (1987). Fetal alcohol effects in long- and short-sleep mice: activity, passive avoidance, and in utero ethanol levels. Neurotoxicol Teratol 9, 349-57.

Glass, C. K., Saijo, K., Winner, B., Marchetto, M. C. and Gage, F. H. (2010). Mechanisms underlying inflammation in neurodegeneration. Cell 140, 918-34.

Goist, K. C., Jr. and Sutker, P. B. (1985). Acute alcohol intoxication and body composition in women and men. Pharmacol Biochem Behav 22, 811-4.

Goldstein, D. B. and Pal, N. (1971). Alcohol dependence produced in mice by inhalation of ethanol: grading the withdrawal reaction. Science 172, 288-90.

Gomes, S. Z., Lorenzon, A. R., Vieira, J. S., Rocha, C. R., Bandeira, C., Hoshida, M. S., Lopes, L. R. and Bevilacqua, E. (2012). Expression of NADPH oxidase by trophoblast cells: potential implications for the postimplanting mouse embryo. Biol Reprod 86, 56.

Gordon, B. H., Baraona, E., Miyakawa, H., Finkelman, F. and Lieber, C. S. (1985). Exaggerated acetaldehyde response after ethanol administration during pregnancy and lactation in rats. Alcohol Clin Exp Res 9, 17-22.

Goth, L. and Eaton, J. W. (2000). Hereditary catalase deficiencies and increased risk of diabetes. Lancet 356, 1820- 1.

Goth, L. (2001). A new type of inherited catalase deficiencies: its characterization and comparison to the Japanese and Swiss type of acatalasemia. Blood Cells Mol Dis 27, 512-7.

Goth, L., Lenkey, A. and Bigler, W. N. (2001). Blood catalase deficiency and diabetes in Hungary. Diabetes Care 24, 1839-40.

Goth, L. and Vitai, M. (2003). The effects of hydrogen peroxide promoted by homocysteine and inherited catalase deficiency on human hypocatalasemic patients. Free Radic Biol Med 35, 882-8.

Goth, L., Rass, P. and Pay, A. (2004). Catalase enzyme mutations and their association with diseases. Mol Diagn 8, 141-9.

Goulas, A., Fidani, L., Kotsis, A., Mirtsou, V., Petersen, R. C., Tangalos, E. and Hardy, J. (2002). An association study of a functional catalase gene polymorphism, -262C-->T, and patients with Alzheimer's disease. Neurosci Lett 330, 210-3.

444

Goyal, M. M. and Basak, A. (2010). Human catalase: looking for complete identity. Protein & cell 1, 888-97.

Grady, S. and Osterloh, J. (1986). Improved enzymic assay for serum formate with colorimetric endpoint. Journal of analytical toxicology 10, 1-5.

Grasberger, H., De Deken, X., Mayo, O. B., Raad, H., Weiss, M., Liao, X. H. and Refetoff, S. (2012). Mice deficient in dual oxidase maturation factors are severely hypothyroid. Mol Endocrinol 26, 481-92.

Groemping, Y. and Rittinger, K. (2005). Activation and assembly of the NADPH oxidase: a structural perspective. Biochem J 386, 401-16.

Grunnet, N. (1973). Oxidation of acetaldehyde by rat-liver mitochondria in relation to ethanol oxidation and the transport of reducing equivalents across the mitochondrial membrane. Eur J Biochem 35, 236-43.

Gu, H., Marth, J. D., Orban, P. C., Mossmann, H. and Rajewsky, K. (1994). Deletion of a DNA polymerase beta gene segment in T cells using cell type-specific gene targeting. Science 265, 103-6.

Gutteridge, J. M. and Quinlan, G. J. (1983). Malondialdehyde formation from lipid peroxides in the thiobarbituric acid test: the role of lipid radicals, iron salts, and metal chelators. J Appl Biochem 5, 293-9.

Gutteridge, J. M. (1984). Lipid peroxidation initiated by superoxide-dependent hydroxyl radicals using complexed iron and hydrogen peroxide. FEBS Lett 172, 245-9.

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 42, 9761-70.

Halliwell, B. and Chirico, S. (1993). Lipid peroxidation: its mechanism, measurement, and significance. Am J Clin Nutr 57, 715S-724S; discussion 724S-725S.

Halliwell, B. and Gutteridge, J. (2007). Free Radicals in Biology and Medicine. 4th ed. Oxford, Oxford University Press.

Han, C. H., Nisimoto, Y., Lee, S. H., Kim, E. T. and Lambeth, J. D. (2001). Characterization of the flavoprotein domain of gp91phox which has NADPH diaphorase activity. J Biochem 129, 513-20.

Hansen, J., Contreras, K. M. and Harris, C. (2005). Methanol, formaldehyde, and sodium formate exposure in rat and mouse conceptuses: a potential role of the visceral yolk sac in embryotoxicity. Birth Defects Res A Clin Mol Teratol 73, 72-82.

Hansen, J. M., Harris, K. K., Philbert, M. A. and Harris, C. (2002). Thalidomide modulates nuclear redox status and preferentially depletes glutathione in rabbit limb versus rat limb. The Journal of pharmacology and experimental therapeutics 300, 768-76.

Hansen, J. M. and Harris, C. (2013). Redox control of teratogenesis. Reprod Toxicol 35, 165-79.

Hantson, P., Haufroid, V. and Wallemacq, P. (2005). Formate kinetics in methanol poisoning. Human & experimental toxicology 24, 55-9.

Harris, C., Wang, S. W., Lauchu, J. J. and Hansen, J. M. (2003). Methanol metabolism and embryotoxicity in rat and mouse conceptuses: comparisons of alcohol dehydrogenase (ADH1), formaldehyde dehydrogenase (ADH3), and catalase. Reprod Toxicol 17, 349-57.

Harris, C., Dixon, M. and Hansen, J. M. (2004). Glutathione depletion modulates methanol, formaldehyde and formate toxicity in cultured rat conceptuses. Cell Biol Toxicol 20, 133-45.

445

Hashida, K., Sakakura, Y. and Makino, N. (2002). Kinetic studies on the hydrogen peroxide elimination by cultured PC12 cells: rate limitation by glucose-6-phosphate dehydrogenase. Biochim Biophys Acta 1572, 85-90.

Hassoun, E. A., Walter, A. C., Alsharif, N. Z. and Stohs, S. J. (1997). Modulation of TCDD-induced fetotoxicity and oxidative stress in embryonic and placental tissues of C57BL/6J mice by vitamin E succinate and ellagic acid. Toxicology 124, 27-37.

Hauck, F., Heine, S., Beier, R., Wieczorek, K., Muller, D., Hahn, G., Gahr, M., Rosen-Wolff, A. and Roesler, J. (2008). Chronic granulomatous disease (CGD) mimicking neoplasms: a suspected mediastinal teratoma unmasking as thymic granulomas due to X-linked CGD, and 2 related cases. J Pediatr Hematol Oncol 30, 877-80.

Hay, M. G., Kinard, F. W. and Nelson, G. H. (1956). Catalase activity and ethanol metabolism in the rat. Proc Soc Exp Biol Med 92, 772-3.

Hayashi, M., Shimazaki, Y., Kamata, S., Kakiichi, N. and Ikeda, M. (1991). Disposition of ethanol and acetaldehyde in maternal blood, fetal blood, and amniotic fluid of near-term pregnant rats. Bull Environ Contam Toxicol 47, 184- 9.

Hayreh, M. S., Hayreh, S. S., Baumbach, G. L., Cancilla, P., Martin-Amat, G., Tephly, T. R., McMartin, K. E. and Makar, A. B. (1977). Methyl alcohol poisoning III. Ocular toxicity. Archives of ophthalmology 95, 1851-8.

Heales, S. J. (2001). Catalase deficiency, diabetes, and mitochondrial function. Lancet 357, 314.

Heaton, M. B., Mitchell, J. J., Paiva, M. and Walker, D. W. (2000). Ethanol-induced alterations in the expression of neurotrophic factors in the developing rat central nervous system. Brain Res Dev Brain Res 121, 97-107.

Heaton, M. B., Paiva, M., Mayer, J. and Miller, R. (2002). Ethanol-mediated generation of reactive oxygen species in developing rat cerebellum. Neurosci Lett 334, 83-6.

Henderson, G. I., Devi, B. G., Perez, A. and Schenker, S. (1995). In utero ethanol exposure elicits oxidative stress in the rat fetus. Alcohol Clin Exp Res 19, 714-20.

Henderson, G. I., Chen, J. J. and Schenker, S. (1999). Ethanol, oxidative stress, reactive aldehydes, and the fetus. Front Biosci 4, D541-50.

Henehan, G. T., Ward, K., Kennedy, N. P., Weir, D. G. and Tipton, K. F. (1985). Subcellular distribution of aldehyde dehydrogenase activities in human liver. Alcohol 2, 107-10.

Hensley, K., Pye, Q. N., Maidt, M. L., Stewart, C. A., Robinson, K. A., Jaffrey, F. and Floyd, R. A. (1998). Interaction of alpha-phenyl-N-tert-butyl nitrone and alternative electron acceptors with complex I indicates a substrate reduction site upstream from the rotenone binding site. J Neurochem 71, 2549-57.

Heo, J. and Campbell, S. L. (2005). Superoxide anion radical modulates the activity of Ras and Ras-related GTPases by a radical-based mechanism similar to that of nitric oxide. J Biol Chem 280, 12438-45.

Herrick, J. R., Brad, A. M. and Krisher, R. L. (2006). Chemical manipulation of glucose metabolism in porcine oocytes: effects on nuclear and cytoplasmic maturation in vitro. Reproduction 131, 289-98.

Heyworth, P. G., Cross, A. R. and Curnutte, J. T. (2003). Chronic granulomatous disease. Curr Opin Immunol 15, 578-84.

Hilenski, L. L., Clempus, R. E., Quinn, M. T., Lambeth, J. D. and Griendling, K. K. (2004). Distinct subcellular localizations of Nox1 and Nox4 in vascular smooth muscle cells. Arterioscler Thromb Vasc Biol 24, 677-83.

Hines, R. N. (2008). The ontogeny of drug metabolism enzymes and implications for adverse drug events. Pharmacol Ther 118, 250-67.

446

Hirano, H., Tabuchi, Y., Kondo, T., Zhao, Q. L., Ogawa, R., Cui, Z. G., Feril, L. B., Jr. and Kanayama, S. (2005). Analysis of gene expression in apoptosis of human lymphoma U937 cells induced by heat shock and the effects of alpha-phenyl N-tert-butylnitrone (PBN) and its derivatives. Apoptosis 10, 331-40.

Ho, Y. S., Xiong, Y., Ma, W., Spector, A. and Ho, D. S. (2004). Mice lacking catalase develop normally but show differential sensitivity to oxidant tissue injury. J Biol Chem 279, 32804-12.

Hofmann, B., Hecht, H. J. and Flohe, L. (2002). Peroxiredoxins. Biol Chem 383, 347-64.

Holford, N. H. (1987). Clinical pharmacokinetics of ethanol. Clin Pharmacokinet 13, 273-92.

Holmgren, A. (1989). Thioredoxin and glutaredoxin systems. J Biol Chem 264, 13963-6.

Holownia, A., Ledig, M., Mapoles, J. and Menez, J. F. (1996). Acetaldehyde-induced growth inhibition in cultured rat astroglial cells. Alcohol 13, 93-7.

Holownia, A., Ledig, M. and Menez, J. F. (1997). Ethanol-induced cell death in cultured rat astroglia. Neurotoxicol Teratol 19, 141-6.

Horton, V. L., Higuchi, M. A. and Rickert, D. E. (1992). Physiologically based pharmacokinetic model for methanol in rats, monkeys, and humans. Toxicol Appl Pharmacol 117, 26-36.

Hovda, K. E., Mundal, H., Urdal, P., McMartin, K. and Jacobsen, D. (2007). Extremely slow formate elimination in severe methanol poisoning: a fatal case report. Clinical toxicology (Philadelphia, Pa 45, 516-21.

Hsu, G. W., Ober, M., Carell, T. and Beese, L. S. (2004). Error-prone replication of oxidatively damaged DNA by a high-fidelity DNA polymerase. Nature 431, 217-21.

Hu, I. C., Singh, S. P. and Snyder, A. K. (1995). Effects of ethanol on glucose transporter expression in cultured hippocampal neurons. Alcohol Clin Exp Res 19, 1398-402.

Hu, L. F., Lu, M., Tiong, C. X., Dawe, G. S., Hu, G. and Bian, J. S. (2010). Neuroprotective effects of hydrogen sulfide on Parkinson's disease rat models. Aging Cell 9, 135-46.

Huang, C. K., Zhan, L., Hannigan, M. O., Ai, Y. and Leto, T. L. (2000). P47(phox)-deficient NADPH oxidase defect in neutrophils of diabetic mouse strains, C57BL/6J-m db/db and db/+. J Leukoc Biol 67, 210-5.

Hunter, E. S., 3rd, Tugman, J. A., Sulik, K. K. and Sadler, T. W. (1994). Effects of short-term exposure to ethanol on mouse embryos in vitro. Toxicol In Vitro 8, 413-21.

IARC (2006). IARC monographs on the evalution of carcinogenic risks to humans volume 88: formaldehyde, 2- butoxyethanol and 1-tert-butoxypropan-2-ol, Ed.^ Eds.). World Health Organization, Geneva.

Idanpaan-Heikkila, J., Jouppila, P., Akerblom, H. K., Isoaho, R., Kauppila, E. and Koivisto, M. (1972). Elimination and metabolic effects of ethanol in mother, fetus, and newborn infant. Am J Obstet Gynecol 112, 387-93.

Ikonomidou, C., Bosch, F., Miksa, M., Bittigau, P., Vockler, J., Dikranian, K., Tenkova, T. I., Stefovska, V., Turski, L. and Olney, J. W. (1999). Blockade of NMDA receptors and apoptotic neurodegeneration in the developing brain. Science 283, 70-4.

Ikonomidou, C., Bittigau, P., Ishimaru, M. J., Wozniak, D. F., Koch, C., Genz, K., Price, M. T., Stefovska, V., Horster, F., Tenkova, T., Dikranian, K. and Olney, J. W. (2000). Ethanol-induced apoptotic neurodegeneration and fetal alcohol syndrome. Science 287, 1056-60.

Inanami, O. and Kuwabara, M. (1995). alpha-Phenyl N-tert-butyl nitrone (PBN) increases the cortical cerebral blood flow by inhibiting the breakdown of nitric oxide in anesthetized rats. Free Radic Res 23, 33-9.

447

Infurna, R. and Weiss, B. (1986). Neonatal behavioral toxicity in rats following prenatal exposure to methanol. Teratology 33, 259-65.

Inoue, Y., Asanuma, T., Smith, N., Saunders, D., Oblander, J., Kotake, Y., Floyd, R. A. and Towner, R. A. (2007). Modulation of Fas-FasL related apoptosis by PBN in the early phases of choline deficient diet-mediated hepatocarcinogenesis in rats. Free Radic Res 41, 972-80.

IPCS (1997). Environmental Health Criteria 196 - Methanol, Ed.^ Eds.). World Health Organization, Geneva.

Ito, T., Ando, H., Suzuki, T., Ogura, T., Hotta, K., Imamura, Y., Yamaguchi, Y. and Handa, H. (2010). Identification of a primary target of thalidomide teratogenicity. Science 327, 1345-50.

Jackson, S. H., Gallin, J. I. and Holland, S. M. (1995). The p47phox mouse knock-out model of chronic granulomatous disease. J Exp Med 182, 751-8.

Janik, J., Swoboda, M., Janowska, B., Ciesla, J. M., Gackowski, D., Kowalewski, J., Olinski, R., Tudek, B. and Speina, E. (2011). 8-Oxoguanine incision activity is impaired in lung tissues of NSCLC patients with the polymorphism of OGG1 and XRCC1 genes. Mutat Res 709-710, 21-31.

Janssen-Heininger, Y. M., Mossman, B. T., Heintz, N. H., Forman, H. J., Kalyanaraman, B., Finkel, T., Stamler, J. S., Rhee, S. G. and van der Vliet, A. (2008). Redox-based regulation of signal transduction: principles, pitfalls, and promises. Free Radic Biol Med 45, 1-17.

Janzen, E. G., Stronks, H. J., Dubose, C. M., Poyer, J. L. and McCay, P. B. (1985). Chemistry and biology of spin- trapping radicals associated with halocarbon metabolism in vitro and in vivo. Environ Health Perspect 64, 151-70.

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, 317-26.

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, 550-6.

Jernstrom, B., Dock, L. and Martinez, M. (1984). Metabolic activation of benzo[a]pyrene-7,8-dihydrodiol and benzo[a]pyrene-7,8-dihydrodiol-9,10-epoxide to protein-binding products and the inhibitory effect of glutathione and cysteine. Carcinogenesis 5, 199-204.

Jiang, F., Zhang, Y. and Dusting, G. J. (2011). NADPH oxidase-mediated redox signaling: roles in cellular stress response, stress tolerance, and tissue repair. Pharmacological reviews 63, 218-42.

Jiang, Z., Akey, J. M., Shi, J., Xiong, M., Wang, Y., Shen, Y., Xu, X., Chen, H., Wu, H., Xiao, J., Lu, D., Huang, W. and Jin, L. (2001). A polymorphism in the promoter region of catalase is associated with blood pressure levels. Hum Genet 109, 95-8.

Johlin, F. C., Fortman, C. S., Nghiem, D. D. and Tephly, T. R. (1987). Studies on the role of folic acid and folate- dependent enzymes in human methanol poisoning. Mol Pharmacol 31, 557-61.

Jones, B. M. and Jones, M. K. (1976). Alcohol effects in women during the menstrual cycle. Ann N Y Acad Sci 273, 576-87.

Jones, K. L. and Smith, D. W. (1973). Recognition of the fetal alcohol syndrome in early infancy. Lancet 302, 999- 1001.

Jones, K. L. (2011). The effects of alcohol on fetal development. Birth Defects Res C Embryo Today 93, 3-11.

Juchau, M. R. (1989). Bioactivation in chemical teratogenesis. Annu Rev Pharmacol Toxicol 29, 165-87.

448

Juchau, M. R., Boutelet-Bochan, H. and Huang, Y. (1998). Cytochrome-P450-dependent biotransformation of xenobiotics in human and rodent embryonic tissues. Drug Metab Rev 30, 541-68.

Julkunen, R. J., Di Padova, C. and Lieber, C. S. (1985a). First pass metabolism of ethanol--a gastrointestinal barrier against the systemic toxicity of ethanol. Life Sci 37, 567-73.

Julkunen, R. J., Tannenbaum, L., Baraona, E. and Lieber, C. S. (1985b). First pass metabolism of ethanol: an important determinant of blood levels after alcohol consumption. Alcohol 2, 437-41.

Kadiiska, M. B. and Mason, R. P. (2000). Acute methanol intoxication generates free radicals in rats: an ESR spin trapping investigation. Free Radic Biol Med 28, 1106-14.

Kalant, H. (1971). Absorption, diffusion, distribution, and elimination of ethanol: effects on biological membranes (B. Kissin and H. Begleiter, Ed.^ Eds.), pp. 1-62. Plenum Press, New York.

Kalant, H. (1996). Pharmacokinetics of Ethanol: Absorption, Distribution, and Elimination (H. Begleiter and B. Kissin, Ed.^ Eds.), pp. 15-58. Oxford University Press, New York.

Kalter, H. (1968). Sporadic congenital malformations of newborn inbred mice. Teratology 1, 193-9.

Kang, S. W., Chae, H. Z., Seo, M. S., Kim, K., Baines, I. C. and Rhee, S. G. (1998). Mammalian peroxiredoxin isoforms can reduce hydrogen peroxide generated in response to growth factors and tumor necrosis factor-alpha. J Biol Chem 273, 6297-302.

Karinje, K. U. and Ogata, M. (1990). Methanol metabolism in acatalasemic mice. Physiol Chem Phys Med NMR 22, 193-8.

Kasapinovic, S., McCallum, G. P., Wiley, M. J. and Wells, P. G. (2004). The peroxynitrite pathway in development: phenytoin and benzo[a]pyrene embryopathies in inducible nitric oxide synthase knockout mice. Free Radic Biol Med 37, 1703-11.

Kavet, R. and Nauss, K. M. (1990). The toxicity of inhaled methanol vapors. Crit Rev Toxicol 21, 21-50.

Kawahara, T., Ritsick, D., Cheng, G. and Lambeth, J. D. (2005). Point mutations in the proline-rich region of p22phox are dominant inhibitors of Nox1- and Nox2-dependent reactive oxygen generation. J Biol Chem 280, 31859-69.

Kedishvili, N. Y., Bosron, W. F., Stone, C. L., Hurley, T. D., Peggs, C. F., Thomasson, H. R., Popov, K. M., Carr, L. G., Edenberg, H. J. and Li, T. K. (1995). Expression and kinetic characterization of recombinant human stomach alcohol dehydrogenase. Active-site amino acid sequence explains substrate specificity compared with liver isozymes. J Biol Chem 270, 3625-30.

Kennedy, J. C., Memet, S. and Wells, P. G. (2004). Antisense evidence for nuclear factor-kappaB-dependent embryopathies initiated by phenytoin-enhanced oxidative stress. Mol Pharmacol 66, 404-12.

Kerns, W. D., Pavkov, K. L., Donofrio, D. J., Gralla, E. J. and Swenberg, J. A. (1983). Carcinogenicity of formaldehyde in rats and mice after long-term inhalation exposure. Cancer Res 43, 4382-4392.

Kershaw, R. M. and Hodges, N. J. (2012). Repair of oxidative DNA damage is delayed in the Ser326Cys polymorphic variant of the base excision repair protein OGG1. Mutagenesis 27, 501-10.

Kesaniemi, Y. A. and Sippel, H. W. (1975). Placental and foetal metabolism of acetaldehyde in rat. I. Contents of ethanol and acetaldehyde in placenta and foets of the pregnant rat during ethanol oxidation. Acta Pharmacol Toxicol (Copenh) 37, 43-8.

Khan, S., Barhoumi, R., Burghardt, R., Liu, S., Kim, K. and Safe, S. (2006). Molecular mechanism of inhibitory aryl hydrocarbon receptor-estrogen receptor/Sp1 cross talk in breast cancer cells. Mol Endocrinol 20, 2199-214.

449

Khaole, N. C., Ramchandani, V. A., Viljoen, D. L. and Li, T. K. (2004). A pilot study of alcohol exposure and pharmacokinetics in women with or without children with fetal alcohol syndrome. Alcohol Alcohol 39, 503-8.

Kikuchi, H., Hikage, M., Miyashita, H. and Fukumoto, M. (2000). NADPH oxidase subunit, gp91(phox) homologue, preferentially expressed in human colon epithelial cells. Gene 254, 237-43.

Kim, H. S., Krege, J. H., Kluckman, K. D., Hagaman, J. R., Hodgin, J. B., Best, C. F., Jennette, J. C., Coffman, T. M., Maeda, N. and Smithies, O. (1995). Genetic control of blood pressure and the angiotensinogen locus. Proc Natl Acad Sci U S A 92, 2735-9.

Kim, M. R., Lee, K. N., Yon, J. M., Lee, S. R., Jin, Y., Baek, I. J., Lee, B. J., Yun, Y. W. and Nam, S. Y. (2008). Capsaicin prevents ethanol-induced teratogenicity in cultured mouse whole embryos. Reprod Toxicol 26, 292-7.

Kinoshita, A., Wanibuchi, H., Morimura, K., Wei, M., Nakae, D., Arai, T., Minowa, O., Noda, T., Nishimura, S. and Fukushima, S. (2007). Carcinogenicity of dimethylarsinic acid in Ogg1-deficient mice. Cancer science 98, 803-14.

Kirkali, G., de Souza-Pinto, N. C., Jaruga, P., Bohr, V. A. and Dizdaroglu, M. (2009). Accumulation of (5'S)-8,5'- cyclo-2'-deoxyadenosine in organs of Cockayne syndrome complementation group B gene knockout mice. DNA Repair (Amst) 8, 274-8.

Kirkman, H. N. and Gaetani, G. F. (1984). Catalase: a tetrameric enzyme with four tightly bound molecules of NADPH. Proc Natl Acad Sci U S A 81, 4343-7.

Kirkman, H. N. and Gaetani, G. F. (2007). Mammalian catalase: a venerable enzyme with new mysteries. Trends Biochem Sci 32, 44-50.

Kiss, P. J., Knisz, J., Zhang, Y., Baltrusaitis, J., Sigmund, C. D., Thalmann, R., Smith, R. J., Verpy, E. and Banfi, B. (2006). Inactivation of NADPH oxidase organizer 1 results in severe imbalance. Curr Biol 16, 208-13.

Klaassen, C. D. (2008). Cassarett and Doull's Toxicology: The Basic Science of Poisons. 7th ed. ed. McGraw-Hill Companies, Inc., New York.

Klaunig, J. E. and Kamendulis, L. M. (2004). The role of oxidative stress in carcinogenesis. Annu Rev Pharmacol Toxicol 44, 239-67.

Kleeman, C. R., Rubini, M. E., Lamdin, E. and Epstein, F. H. (1955). Studies on alcohol diuresis. II. The evaluation of ethyl alcohol as an inhibitor of the neurohypophysis. J Clin Invest 34, 448-55.

Kleinschnitz, C., Grund, H., Wingler, K., Armitage, M. E., Jones, E., Mittal, M., Barit, D., Schwarz, T., Geis, C., Kraft, P., Barthel, K., Schuhmann, M. K., Herrmann, A. M., Meuth, S. G., Stoll, G., Meurer, S., Schrewe, A., Becker, L., Gailus-Durner, V., Fuchs, H., Klopstock, T., de Angelis, M. H., Jandeleit-Dahm, K., Shah, A. M., Weissmann, N. and Schmidt, H. H. (2010). Post-stroke inhibition of induced NADPH oxidase type 4 prevents oxidative stress and neurodegeneration. PLoS Biol 8.

Klungland, A. and Lindahl, T. (1997). Second pathway for completion of human DNA base excision-repair: reconstitution with purified proteins and requirement for DNase IV (FEN1). EMBO J 16, 3341-8.

Klungland, A., Rosewell, I., Hollenbach, S., Larsen, E., Daly, G., Epe, B., Seeberg, E., Lindahl, T. and Barnes, D. E. (1999). Accumulation of premutagenic DNA lesions in mice defective in removal of oxidative base damage. Proc Natl Acad Sci U S A 96, 13300-5.

Klungland, A. and Bjelland, S. (2007). Oxidative damage to purines in DNA: role of mammalian Ogg1. DNA Repair (Amst) 6, 481-8.

Knobloch, J., Shaughnessy, J. D., Jr. and Ruther, U. (2007). Thalidomide induces limb deformities by perturbing the Bmp/Dkk1/Wnt signaling pathway. FASEB J 21, 1410-21.

450

Knobloch, J. and Ruther, U. (2008). Shedding light on an old mystery: thalidomide suppresses survival pathways to induce limb defects. Cell Cycle 7, 1121-7.

Knobloch, J., Schmitz, I., Gotz, K., Schulze-Osthoff, K. and Ruther, U. (2008). Thalidomide induces limb anomalies by PTEN stabilization, Akt suppression, and stimulation of caspase-dependent cell death. Mol Cell Biol 28, 529-38.

Ko, M. S. (2001). Embryogenomics: developmental biology meets genomics. Trends Biotechnol 19, 511-8.

Kohno, T., Shinmura, K., Tosaka, M., Tani, M., Kim, S. R., Sugimura, H., Nohmi, T., Kasai, H. and Yokota, J. (1998). Genetic polymorphisms and alternative splicing of the hOGG1 gene, that is involved in the repair of 8- hydroxyguanine in damaged DNA. Oncogene 16, 3219-25.

Kono, H., Rusyn, I., Yin, M., Gabele, E., Yamashina, S., Dikalova, A., Kadiiska, M. B., Connor, H. D., Mason, R. P., Segal, B. H., Bradford, B. U., Holland, S. M. and Thurman, R. G. (2000). NADPH oxidase-derived free radicals are key oxidants in alcohol-induced liver disease. J Clin Invest 106, 867-72.

Koop, D. R. (2006). Alcohol metabolism's damaging effects on the cell: a focus on reactive oxygen generation by the enzyme cytochrome P450 2E1. Alcohol Res Health 29, 274-80.

Kotake, Y., Sang, H., Miyajima, T. and Wallis, G. L. (1998). Inhibition of NF-kappaB, iNOS mRNA, COX2 mRNA, and COX catalytic activity by phenyl-N-tert-butylnitrone (PBN). Biochim Biophys Acta 1448, 77-84.

Kotake, Y., Sang, H., Wallis, G. L. and Stewart, C. A. (1999). Phenyl N-tert-butylnitrone provides protection from endotoxin shock through amplified production of the anti-inflammatory cytokine interleukin-10. Arch Biochem Biophys 371, 129-31.

Kotch, L. E. and Sulik, K. K. (1992a). Experimental fetal alcohol syndrome: proposed pathogenic basis for a variety of associated facial and brain anomalies. Am J Med Genet 44, 168-76.

Kotch, L. E. and Sulik, K. K. (1992b). Patterns of ethanol-induced cell death in the developing nervous system of mice; neural fold states through the time of anterior neural tube closure. Int J Dev Neurosci 10, 273-9.

Kotch, L. E., Chen, S. Y. and Sulik, K. K. (1995). Ethanol-induced teratogenesis: free radical damage as a possible mechanism. Teratology 52, 128-36.

Krebs, H. A., Freedland, R. A., Hems, R. and Stubbs, M. (1969). Inhibition of hepatic gluconeogenesis by ethanol. Biochem J 112, 117-24.

Kucherlapati, M., Yang, K., Kuraguchi, M., Zhao, J., Lia, M., Heyer, J., Kane, M. F., Fan, K., Russell, R., Brown, A. M., Kneitz, B., Edelmann, W., Kolodner, R. D., Lipkin, M. and Kucherlapati, R. (2002). Haploinsufficiency of Flap endonuclease (Fen1) leads to rapid tumor progression. Proc Natl Acad Sci U S A 99, 9924-9.

Kumada, T., Lakshmana, M. K. and Komuro, H. (2006). Reversal of neuronal migration in a mouse model of fetal alcohol syndrome by controlling second-messenger signalings. J Neurosci 26, 742-56.

Kunisada, M., Sakumi, K., Tominaga, Y., Budiyanto, A., Ueda, M., Ichihashi, M., Nakabeppu, Y. and Nishigori, C. (2005). 8-Oxoguanine formation induced by chronic UVB exposure makes Ogg1 knockout mice susceptible to skin carcinogenesis. Cancer Res 65, 6006-10.

Kunitoh, S., Tanaka, T., Imaoka, S., Funae, Y. and Monna, Y. (1993). Contribution of cytochrome P450s to MEOS (microsomal ethanol-oxidizing system): a specific and sensitive assay of MEOS activity by HPLC with fluorescence labeling. Alcohol and alcoholism (Oxford, Oxfordshire) 1B, 63-8.

Kuo, F. C. and Sklar, J. (1997). Augmented expression of a human gene for 8-oxoguanine DNA glycosylase (MutM) in B lymphocytes of the dark zone in lymph node germinal centers. J Exp Med 186, 1547-56.

451

Kuroda, J., Ago, T., Matsushima, S., Zhai, P., Schneider, M. D. and Sadoshima, J. (2010). NADPH oxidase 4 (Nox4) is a major source of oxidative stress in the failing heart. Proc Natl Acad Sci U S A 107, 15565-70.

Kwo, P. Y., Ramchandani, V. A., O'Connor, S., Amann, D., Carr, L. G., Sandrasegaran, K., Kopecky, K. K. and Li, T. K. (1998). Gender differences in alcohol metabolism: relationship to liver volume and effect of adjusting for body mass. Gastroenterology 115, 1552-7.

Laleu, B., Gaggini, F., Orchard, M., Fioraso-Cartier, L., Cagnon, L., Houngninou-Molango, S., Gradia, A., Duboux, G., Merlot, C., Heitz, F., Szyndralewiez, C. and Page, P. (2010). First in class, potent, and orally bioavailable NADPH oxidase isoform 4 (Nox4) inhibitors for the treatment of idiopathic pulmonary fibrosis. J Med Chem 53, 7715-30.

Lalucque, H. and Silar, P. (2003). NADPH oxidase: an enzyme for multicellularity? Trends Microbiol 11, 9-12.

Lambeth, J. D. (2004). NOX enzymes and the biology of reactive oxygen. Nat Rev Immunol 4, 181-9.

Lander, H. M. (1997). An essential role for free radicals and derived species in signal transduction. FASEB J 11, 118-24.

Lander, H. M., Hajjar, D. P., Hempstead, B. L., Mirza, U. A., Chait, B. T., Campbell, S. and Quilliam, L. A. (1997). A molecular redox switch on p21(ras). Structural basis for the nitric oxide-p21(ras) interaction. J Biol Chem 272, 4323-6.

Laposa, R. R., Henderson, J. T., Xu, E. and Wells, P. G. (2004). Atm-null mice exhibit enhanced radiation-induced birth defects and a hybrid form of embryonic programmed cell death indicating a teratological suppressor function for ATM. Faseb J 18, 896-8.

Larsen, E., Gran, C., Saether, B. E., Seeberg, E. and Klungland, A. (2003). Proliferation failure and gamma radiation sensitivity of Fen1 null mutant mice at the blastocyst stage. Mol Cell Biol 23, 5346-53.

Larsen, J. A. (1959). Extrahepatic metabolism of ethanol in man. Nature 184(Suppl 16), 1236.

Lassegue, B., San Martin, A. and Griendling, K. K. (2012). Biochemistry, physiology, and pathophysiology of NADPH oxidases in the cardiovascular system. Circ Res 110, 1364-90.

Lee, C. J., Goncalves, L. L. and Wells, P. G. (2011). Embryopathic effects of thalidomide and its hydrolysis products in rabbit embryo culture: evidence for a prostaglandin H synthase (PHS)-dependent, reactive oxygen species (ROS)-mediated mechanism. Faseb J 25, 2468-83.

Lee, L., Schmidt, K. L., Tornwall, M. S., Henagan, J. M. and Miller, T. A. (1992). Gender differences in ethanol oxidation and injury in the rat stomach. Alcohol 9, 421-5.

Lee, R. D., An, S. M., Kim, S. S., Rhee, G. S., Kwack, S. J., Seok, J. H., Chae, S. Y., Park, C. H., Choi, Y. W., Kim, H. S., Cho, H. Y., Lee, B. M. and Park, K. L. (2005). Neurotoxic effects of alcohol and acetaldehyde during embryonic development. J Toxicol Environ Health A 68, 2147-62.

Lee, S. R., Kim, M. R., Yon, J. M., Baek, I. J., Park, C. G., Lee, B. J., Yun, Y. W. and Nam, S. Y. (2009). Black ginseng inhibits ethanol-induced teratogenesis in cultured mouse embryos through its effects on antioxidant activity. Toxicol In Vitro 23, 47-52.

Lenz, W. (1988). A short history of thalidomide embryopathy. Teratology 38, 203-15.

Leonard, S. E. and Carroll, K. S. (2011). Chemical 'omics' approaches for understanding protein cysteine oxidation in biology. Curr Opin Chem Biol 15, 88-102.

Leto, T. L., Adams, A. G. and de Mendez, I. (1994). Assembly of the phagocyte NADPH oxidase: binding of Src homology 3 domains to proline-rich targets. Proc Natl Acad Sci U S A 91, 10650-4.

452

Levine, R. L., Williams, J. A., Stadtman, E. R. and Shacter, E. (1994). Carbonyl assays for determination of oxidatively modified proteins. Methods Enzymol 233, 346-57.

Li, J. M., Fan, L. M., Christie, M. R. and Shah, A. M. (2005a). Acute tumor necrosis factor alpha signaling via NADPH oxidase in microvascular endothelial cells: role of p47phox phosphorylation and binding to TRAF4. Mol Cell Biol 25, 2320-30.

Li, X. J., Grunwald, D., Mathieu, J., Morel, F. and Stasia, M. J. (2005b). Crucial role of two potential cytosolic regions of Nox2, 191TSSTKTIRRS200 and 484DESQANHFAVHHDEEKD500, on NADPH oxidase activation. J Biol Chem 280, 14962-73.

Li, Z., Ding, M., Thiele, C. J. and Luo, J. (2004). Ethanol inhibits brain-derived neurotrophic factor-mediated intracellular signaling and activator protein-1 activation in cerebellar granule neurons. Neuroscience 126, 149-62.

Lieber, C. S. (2004). The discovery of the microsomal ethanol oxidizing system and its physiologic and pathologic role. Drug Metab Rev 36, 511-29.

Light, K. E., Belcher, S. M. and Pierce, D. R. (2002). Time course and manner of Purkinje neuron death following a single ethanol exposure on postnatal day 4 in the developing rat. Neuroscience 114, 327-37.

Lim, R. T., Jr., Gentry, R. T., Ito, D., Yokoyama, H., Baraona, E. and Lieber, C. S. (1993). First-pass metabolism of ethanol is predominantly gastric. Alcohol Clin Exp Res 17, 1337-44.

Lin, P., Hu, S. W. and Chang, T. H. (2003). Correlation between gene expression of aryl hydrocarbon receptor (AhR), hydrocarbon receptor nuclear translocator (Arnt), cytochromes P4501A1 (CYP1A1) and 1B1 (CYP1B1), and inducibility of CYP1A1 and CYP1B1 in human lymphocytes. Toxicol Sci 71, 20-6.

Lin, S., Cox, H. J., Rhodes, P. G. and Cai, Z. (2006). Neuroprotection of alpha-phenyl-n-tert-butyl-nitrone on the neonatal white matter is associated with anti-inflammation. Neurosci Lett 405, 52-6.

Liu, L. and Wells, P. G. (1994). In vivo phenytoin-initiated oxidative damage to proteins and lipids in murine maternal hepatic and embryonic tissue organelles: potential molecular targets of chemical teratogenesis. Toxicol Appl Pharmacol 125, 247-55.

Liu, Y., Lo, Y. C., Qian, L., Crews, F. T., Wilson, B., Chen, H. L., Wu, H. M., Chen, S. H., Wei, K., Lu, R. B., Ali, S. and Hong, J. S. (2011). Verapamil protects dopaminergic neuron damage through a novel anti-inflammatory mechanism by inhibition of microglial activation. Neuropharmacology 60, 373-80.

Lloyd, R. V., Hanna, P. M. and Mason, R. P. (1997). The origin of the hydroxyl radical oxygen in the Fenton reaction. Free Radic Biol Med 22, 885-8.

Lorenzini, C. A., Baldi, E., Bucherelli, C., Sacchetti, B. and Tassoni, G. (1996). Role of dorsal hippocampus in acquisition, consolidation and retrieval of rat's passive avoidance response: a tetrodotoxin functional inactivation study. Brain Res 730, 32-9.

Lu, K., Collins, L. B., Ru, H., Bermudez, E. and Swenberg, J. A. (2010). Distribution of DNA adducts caused by inhaled formaldehyde is consistent with induction of nasal carcinoma but not leukemia. Toxicol Sci 116, 441-51.

Lu, R., Nash, H. M. and Verdine, G. L. (1997). A mammalian DNA repair enzyme that excises oxidatively damaged guanines maps to a locus frequently lost in lung cancer. Curr Biol 7, 397-407.

Ludwig, D. L., MacInnes, M. A., Takiguchi, Y., Purtymun, P. E., Henrie, M., Flannery, M., Meneses, J., Pedersen, R. A. and Chen, D. J. (1998). A murine AP-endonuclease gene-targeted deficiency with post-implantation embryonic progression and ionizing radiation sensitivity. Mutat Res 409, 17-29.

453

Luo, J. and Miller, M. W. (1996). Ethanol inhibits basic fibroblast growth factor-mediated proliferation of C6 astrocytoma cells. J Neurochem 67, 1448-56.

Lutwak-Mann, C. (1938). Alcohol dehydrogenase of animal tissues. Biochem J 32, 1364-74.

Lv, Y. J., Liao, W. J., Luan, Q., Wang, H., Wang, L. and Li, Q. (2011). The polymorphism of catalase T/C codon 389 in exon 9 and vitiligo susceptibility: a meta-analysis. J Eur Acad Dermatol Venereol 25, 955-8.

MacMicking, J. D., Nathan, C., Hom, G., Chartrain, N., Fletcher, D. S., Trumbauer, M., Stevens, K., Xie, Q. W., Sokol, K., Hutchinson, N. and et al. (1995). Altered responses to bacterial infection and endotoxic shock in mice lacking inducible nitric oxide synthase. Cell 81, 641-50.

Majander, A., Finel, M. and Wikstrom, M. (1994). Diphenyleneiodonium inhibits reduction of iron-sulfur clusters in the mitochondrial NADH-ubiquinone oxidoreductase (Complex I). J Biol Chem 269, 21037-42.

Makar, A. B., Tephly, T. R. and Mannering, G. J. (1968). Methanol metabolism in the monkey. Mol Pharmacol 4, 471-83.

Makar, A. B. and Tephly, T. R. (1977). Methanol poisoning VI: role of folic acid in the production of methanol poisoning in the rat. Journal of toxicology and environmental health 2, 1201-9.

Makar, A. B. and Tephly, T. R. (1982). Improved estimation of formate in body fluids and tissues. Clinical chemistry 28, 385.

Mambo, E., Chatterjee, A., de Souza-Pinto, N. C., Mayard, S., Hogue, B. A., Hoque, M. O., Dizdaroglu, M., Bohr, V. A. and Sidransky, D. (2005). Oxidized guanine lesions and hOgg1 activity in lung cancer. Oncogene 24, 4496- 508.

Manes, C. and Lai, N. C. (1995). Nonmitochondrial oxygen utilization by rabbit blastocysts and surface production of superoxide radicals. J Reprod Fertil 104, 69-75.

Mani, J. C., Pietruszko, R. and Theorell, H. (1970). Methanol activity o alcohol dehydrogenases from human liver, horse liver, and yeast. Arch Biochem Biophys 140, 52-9.

Mannering, G. J., Van Harken, D. R., Makar, A. B., Tephly, T. R., Watkins, W. D. and Goodman, J. I. (1969). Role of the intracellular distribution of hepatic catalase in the peroxidative oxidation of methanol. Ann N Y Acad Sci 168, 265-80.

Maritim, A. C., Sanders, R. A. and Watkins, J. B., 3rd (2003). Diabetes, oxidative stress, and antioxidants: a review. J Biochem Mol Toxicol 17, 24-38.

Marnett, L. J. (2000). Oxyradicals and DNA damage. Carcinogenesis 21, 361-70.

Marshall, A. W., Kingstone, D., Boss, M. and Morgan, M. Y. (1983). Ethanol elimination in males and females: relationship to menstrual cycle and body composition. Hepatology 3, 701-6.

Marterre, W. F., Jr., Kindy, M. S., Carney, J. M., Landrum, R. W. and Strodel, W. E. (1991). Induction of the protooncogene c-fos and recovery of cytosolic adenosine triphosphate in reperfused liver after transient warm ischemia: effect of nitrone free-radical spin-trap agents. Surgery 110, 184-91.

Martin, N. G., Perl, J., Oakeshott, J. G., Gibson, J. B., Starmer, G. A. and Wilks, A. V. (1985). A twin study of ethanol metabolism. Behav Genet 15, 93-109.

Martinez, S. E. and Egea, G. (2007). Novel molecular targets for the prevention of fetal alcohol syndrome. Recent Pat CNS Drug Discov 2, 23-35.

454

Masuoka, N., Kodama, H., Abe, T., Wang, D. H. and Nakano, T. (2003). Characterization of hydrogen peroxide removal reaction by hemoglobin in the presence of reduced pyridine nucleotides. Biochim Biophys Acta 1637, 46- 54.

Mathew, L. K., Simonich, M. T. and Tanguay, R. L. (2009). AHR-dependent misregulation of Wnt signaling disrupts tissue regeneration. Biochem Pharmacol 77, 498-507.

Matsuno, K., Yamada, H., Iwata, K., Jin, D., Katsuyama, M., Matsuki, M., Takai, S., Yamanishi, K., Miyazaki, M., Matsubara, H. and Yabe-Nishimura, C. (2005). Nox1 is involved in angiotensin II-mediated hypertension: a study in Nox1-deficient mice. Circulation 112, 2677-85.

Matsuzawa, A. and Ichijo, H. (2008). Redox control of cell fate by MAP kinase: physiological roles of ASK1-MAP kinase pathway in stress signaling. Biochim Biophys Acta 1780, 1325-36.

Matthews, J. and Gustafsson, J. A. (2006). Estrogen receptor and aryl hydrocarbon receptor signaling pathways. Nucl Recept Signal 4, e016.

Matthews, S. J. and McCoy, C. (2003). Thalidomide: a review of approved and investigational uses. Clin Ther 25, 342-95.

Mattson, S. N., Carlos, R. and Riley, E. P. (1993). The behavioral teratogenicity of alcohol is not affected by pretreatment with aspirin. Alcohol 10, 51-7.

Mattson, S. N., Riley, E. P., Jernigan, T. L., Garcia, A., Kaneko, W. M., Ehlers, C. L. and Jones, K. L. (1994). A decrease in the size of the basal ganglia following prenatal alcohol exposure: a preliminary report. Neurotoxicol Teratol 16, 283-9.

Mattson, S. N., Riley, E. P., Sowell, E. R., Jernigan, T. L., Sobel, D. F. and Jones, K. L. (1996). A decrease in the size of the basal ganglia in children with fetal alcohol syndrome. Alcohol Clin Exp Res 20, 1088-93.

Mattson, S. N. and Riley, E. P. (1999). Implicit and explicit memory functioning in children with heavy prenatal alcohol exposure. J Int Neuropsychol Soc 5, 462-71.

Mattson, S. N., Schoenfeld, A. M. and Riley, E. P. (2001). Teratogenic effects of alcohol on brain and behavior. Alcohol Res Health 25, 185-91.

Mattson, S. N., Crocker, N. and Nguyen, T. T. (2011). Fetal alcohol spectrum disorders: neuropsychological and behavioral features. Neuropsychol Rev 21, 81-101.

May, P. A., Gossage, J. P., Marais, A. S., Hendricks, L. S., Snell, C. L., Tabachnick, B. G., Stellavato, C., Buckley, D. G., Brooke, L. E. and Viljoen, D. L. (2008). Maternal risk factors for fetal alcohol syndrome and partial fetal alcohol syndrome in South Africa: a third study. Alcohol Clin Exp Res 32, 738-53.

McCallum, G. P., Siu, M., Ondovcik, S. L., Sweeting, J. N. and Wells, P. G. (2011a). Methanol exposure does not lead to accumulation of oxidative DNA damage in bone marrow and spleen of mice, rabbits or primates. Mol Carcinog 50, 163-72.

McCallum, G. P., Siu, M., Sweeting, J. N. and Wells, P. G. (2011b). Methanol exposure does not produce oxidatively damaged DNA in lung, liver or kidney of adult mice, rabbits or primates. Toxicol Appl Pharmacol 250, 147-53.

McCallum, G. P., Wong, A. W. and Wells, P. G. (2011c). Cockayne syndrome B protects against methamphetamine-enhanced oxidative DNA damage in murine fetal brain and postnatal neurodevelopmental deficits. Antioxid Redox Signal 14, 747-56.

McGivern, R. F., Clancy, A. N., Hill, M. A. and Noble, E. P. (1984). Prenatal alcohol exposure alters adult expression of sexually dimorphic behavior in the rat. Science 224, 896-8.

455

McLaughlin, B., Hartnett, K. A., Erhardt, J. A., Legos, J. J., White, R. F., Barone, F. C. and Aizenman, E. (2003). Caspase 3 activation is essential for neuroprotection in preconditioning. Proc Natl Acad Sci U S A 100, 715-20.

McMartin, K. E., Martin-Amat, G., Makar, A. B. and Tephly, T. R. (1977). Methanol poisoning. V. Role of formate metabolism in the monkey. J Pharmacol Exp Ther 201, 564-72.

McMartin, K. E., Martin-Amat, G., Noker, P. E. and Tephly, T. R. (1979). Lack of a role for formaldehyde in methanol poisoning in the monkey. Biochem Pharmacol 28, 645-9.

McMartin, K. E., Virayotha, V. and Tephly, T. R. (1981). High-pressure liquid chromatography separation and determination of rat liver folates. Arch Biochem Biophys 209, 127-36.

Meira, L. B., Devaraj, S., Kisby, G. E., Burns, D. K., Daniel, R. L., Hammer, R. E., Grundy, S., Jialal, I. and Friedberg, E. C. (2001). Heterozygosity for the mouse Apex gene results in phenotypes associated with oxidative stress. Cancer Res 61, 5552-7.

Memisoglu, A. and Samson, L. (2000). Base excision repair in yeast and mammals. Mutat Res 451, 39-51.

Middaugh, L. D., Frackelton, W. F., Boggan, W. O., Onofrio, A. and Shepherd, C. L. (1992). Gender differences in the effects of ethanol on C57BL/6 mice. Alcohol 9, 257-60.

Miller, J. H., Fan-Chiang, C. C., Straatsma, T. P. and Kennedy, M. A. (2003). 8-Oxoguanine enhances bending of DNA that favors binding to glycosylases. J Am Chem Soc 125, 6331-6.

Miller, L. and Wells, P. G. (2011). Altered methanol embryopathies in embryo culture with mutant catalase- deficient mice and transgenic mice expressing human catalase. Toxicol Appl Pharmacol 252, 55-61.

Miller, L., Pinto, D. and Wells, P. G. (2013a). Embryonic DNA repair and ethanol-initiated behavioural deficits in oxoguanine glycosylase 1 (OGG1) knockout mice: a role of oxidatively damaged DNA and protection by a free radical spin trapping agent. in press.

Miller, L., Shapiro, A. M., Cheng, J. and Wells, P. G. (2013b). The free radical spin trapping agent phenylbutylnitrone reduces fetal brain DNA oxidation and postnatal cognitive deficits caused by in utero exposure to a non-structurally teratogenic dose of ethanol: A role for oxidative stress. Free Radic Biol Med 60, 223-32.

Miller, L., Shapiro, A. M. and Wells, P. G. (2013c). Embryonic catalase protects against ethanol-initiated DNA oxidation and teratogenesis in acatalasemic and transgenic human catalase-expressing micea. Toxicol Sci.

Miller, L. and Wells, P. G. (2013). Embryonic DNA repair and gender are risk factors in ethanol embryopathies in Oxoguanine glycosylase 1 (OGG1) knockout mice: a role for oxidatively damaged DNA and protection by a free radical spin trapping agent. in press.

Miller, R., King, M. A., Heaton, M. B. and Walker, D. W. (2002). The effects of chronic ethanol consumption on neurotrophins and their receptors in the rat hippocampus and basal forebrain. Brain Res 950, 137-47.

Mills, K. C. and Bisgrove, E. Z. (1983). Body sway and divided attention performance under the influence of alcohol: dose-response differences between males and females. Alcohol Clin Exp Res 7, 393-7.

Minowa, O., Arai, T., Hirano, M., Monden, Y., Nakai, S., Fukuda, M., Itoh, M., Takano, H., Hippou, Y., Aburatani, H., Masumura, K., Nohmi, T., Nishimura, S. and Noda, T. (2000). Mmh/Ogg1 gene inactivation results in accumulation of 8-hydroxyguanine in mice. Proc Natl Acad Sci U S A 97, 4156-61.

Miranda, A. F., Wiley, M. J. and Wells, P. G. (1994). Evidence for embryonic peroxidase-catalyzed bioactivation and glutathione-dependent cytoprotection in phenytoin teratogenicity: modulation by eicosatetraynoic acid and buthionine sulfoximine in murine embryo culture. Toxicol Appl Pharmacol 124, 230-41.

456

Mishra, L., Sharma, S., Potter, J. J. and Mezey, E. (1989). More rapid elimination of alcohol in women as compared to their male siblings. Alcohol Clin Exp Res 13, 752-4.

Miyajima, T. and Kotake, Y. (1995). Spin trapping agent, phenyl N-tert-butyl nitrone, inhibits induction of nitric oxide synthase in endotoxin-induced shock in mice. Biochem Biophys Res Commun 215, 114-21.

Miyajima, T. and Kotake, Y. (1997). Optimal time and dosage of phenyl N-tert-butyl nitrone (PBN) for the inhibition of nitric oxide synthase induction in mice. Free Radic Biol Med 22, 463-70.

Moeller, B. C., Lu, K., Doyle-Eisele, M., McDonald, J., Gigliotti, A. and Swenberg, J. A. (2011). Determination of N2-hydroxymethyl-dG adducts in the nasal epithelium and bone marrow of nonhuman primates following 13CD2- formaldehyde inhalation exposure. Chem Res Toxicol 24, 162-4.

Mohanty, J. G., Bhamidipaty, S., Evans, M. K. and Rifkind, J. M. (2010). A fluorimetric semi-microplate format assay of protein carbonyls in blood plasma. Anal Biochem 400, 289-94.

Monden, Y., Arai, T., Asano, M., Ohtsuka, E., Aburatani, H. and Nishimura, S. (1999). Human MMH (OGG1) type 1a protein is a major enzyme for repair of 8-hydroxyguanine lesions in human cells. Biochem Biophys Res Commun 258, 605-10.

Montoliu, C., Sancho-Tello, M., Azorin, I., Burgal, M., Valles, S., Renau-Piqueras, J. and Guerri, C. (1995). Ethanol increases cytochrome P4502E1 and induces oxidative stress in astrocytes. J Neurochem 65, 2561-70.

Mooney, S. M. and Varlinskaya, E. I. (2011). Acute prenatal exposure to ethanol and social behavior: effects of age, sex, and timing of exposure. Behav Brain Res 216, 358-64.

Moore, K. and Persaud, T. V. N. (2007). THE DEVELOPING HUMAN: Clinically Oriented Embryology. 8th ed. W.B. Saunders, Philadelphia, USA.

Mukhopadhyay, P., Rajesh, M., Batkai, S., Patel, V., Kashiwaya, Y., Liaudet, L., Evgenov, O. V., Mackie, K., Hasko, G. and Pacher, P. (2010). CB1 cannabinoid receptors promote oxidative stress and cell death in murine models of doxorubicin-induced cardiomyopathy and in human cardiomyocytes. Cardiovasc Res 85, 773-84.

Mulvihill, L. E., Skilling, T. A. and Vogel-Sprott, M. (1997). Alcohol and the ability to inhibit behavior in men and women. J Stud Alcohol 58, 600-5.

Mumenthaler, M. S., Taylor, J. L., O'Hara, R. and Yesavage, J. A. (1999). Gender differences in moderate drinking effects. Alcohol Res Health 23, 55-64.

Murphy, M. P. (2009). How mitochondria produce reactive oxygen species. Biochem J 417, 1-13.

Museridze, D., Tsaishvili, T., Svanidze, I. and Khanayeva, Z. (2008). Disorders in memory and learning in offspring of alcoholized female rats, and a possibility for correction of these changes. Neurophysiology 40, 115-120.

Nagababu, E., Chrest, F. J. and Rifkind, J. M. (2003). Hydrogen-peroxide-induced heme degradation in red blood cells: the protective roles of catalase and glutathione peroxidase. Biochim Biophys Acta 1620, 211-7.

Nakae, D., Kotake, Y., Kishida, H., Hensley, K. L., Denda, A., Kobayashi, Y., Kitayama, W., Tsujiuchi, T., Sang, H., Stewart, C. A., Tabatabaie, T., Floyd, R. A. and Konishi, Y. (1998). Inhibition by phenyl N-tert-butyl nitrone of early phase carcinogenesis in the livers of rats fed a choline-deficient, L-amino acid-defined diet. Cancer Res 58, 4548-51.

Nakano, Y., Longo-Guess, C. M., Bergstrom, D. E., Nauseef, W. M., Jones, S. M. and Banfi, B. (2008). Mutation of the Cyba gene encoding p22phox causes vestibular and immune defects in mice. J Clin Invest 118, 1176-85.

Nash, C. M., Ibram, F., Dringenberg, H. C., Reynolds, J. N. and Brien, J. F. (2007). Effects of maternal administration of vitamins C and E on ethanol neurobehavioral teratogenicity in the guinea pig. Alcohol 41, 577-86.

457

Nash, H. M., Bruner, S. D., Scharer, O. D., Kawate, T., Addona, T. A., Spooner, E., Lane, W. S. and Verdine, G. L. (1996). Cloning of a yeast 8-oxoguanine DNA glycosylase reveals the existence of a base-excision DNA-repair protein superfamily. Curr Biol 6, 968-80.

Nasr-Esfahani, M. H., Aitken, J. R. and Johnson, M. H. (1990). Hydrogen peroxide levels in mouse oocytes and early cleavage stage embryos developed in vitro or in vivo. Development 109, 501-7.

Nasr-Esfahani, M. M. and Johnson, M. H. (1991). The origin of reactive oxygen species in mouse embryos cultured in vitro. Development 113, 551-60.

Nayanatara, A. K., Nagaraja, H. S., Ramaswamy, C., Bhagyalakshmi, K., Ramesh Bhat, M. and Harini, N. (2009). Estimation of Tissue Lipid Peroxidation Level and Organ Weight in Litters of Wistar Rats Exposed to Prenatal Alcohol Ingestion. J Physiol Biomed Sci 22, 44-47.

NEDO (1985). Toxicological research of methanol as a fuel for power stations, Ed.^ Eds.). New Energy Development Organization, Tokyo, Japan.

NEDO (1986). Toxicological research on methanol as a fuel for power stations.

NEDO (1987). Toxicological research on methanol as a fuel for power stations: A summary report on test with monkeys, rats and mice.

Neeley, W. L. and Essigmann, J. M. (2006). Mechanisms of formation, genotoxicity, and mutation of guanine oxidation products. Chem Res Toxicol 19, 491-505.

Neese, S., La Grange, L., Trujillo, E. and Romero, D. (2004). The effects of ethanol and silymarin treatment during gestation on spatial working memory. BMC Complement Altern Med 4, 4.

Nelson, B. K., Brightwell, W. S., MacKenzie, D. R., Khan, A., Burg, J. R., Weigel, W. W. and Goad, P. T. (1985). Teratological assessment of methanol and ethanol at high inhalation levels in rats. Fundam Appl Toxicol 5, 727-36.

Nelson, G. H., Kinard, F. W. and Hay, M. G. (1957). Rate of metabolism of ethanol in the mouse. Am J Physiol 190, 169-71.

New, D. A. (1978). Whole-embryo culture and the study of mammalian embryos during organogenesis. Biol Rev Camb Philos Soc 53, 81-122.

Niaura, R. S., Nathan, P. E., Frankenstein, W., Shapiro, A. P. and Brick, J. (1987). Gender differences in acute psychomotor, cognitive, and pharmacokinetic response to alcohol. Addict Behav 12, 345-56.

Nicholls, P. (1975). Formate as an inhibitor of cytochrome c oxidase. Biochem Biophys Res Commun 67, 610-6.

Nicholls, P. (1976). The effect of formate on cytochrome aa3 and on electron transport in the intact respiratory chain. Biochim Biophys Acta 430, 13-29.

Nicol, C. J., Harrison, M. L., Laposa, R. R., Gimelshtein, I. L. and Wells, P. G. (1995). A teratologic suppressor role for p53 in benzo[a]pyrene-treated transgenic p53-deficient mice. Nat Genet 10, 181-7.

Nicol, C. J., Zielenski, J., Tsui, L. C. and Wells, P. G. (2000). An embryoprotective role for glucose-6-phosphate dehydrogenase in developmental oxidative stress and chemical teratogenesis. Faseb J 14, 111-27.

Niimura, F., Labosky, P. A., Kakuchi, J., Okubo, S., Yoshida, H., Oikawa, T., Ichiki, T., Naftilan, A. J., Fogo, A., Inagami, T. and et al. (1995). Gene targeting in mice reveals a requirement for angiotensin in the development and maintenance of kidney morphology and growth factor regulation. J Clin Invest 96, 2947-54.

458

Nishioka, K., Ohtsubo, T., Oda, H., Fujiwara, T., Kang, D., Sugimachi, K. and Nakabeppu, Y. (1999). Expression and differential intracellular localization of two major forms of human 8-oxoguanine DNA glycosylase encoded by alternatively spliced OGG1 mRNAs. Mol Biol Cell 10, 1637-52.

Norwood, V. F., Craig, M. R., Harris, J. M. and Gomez, R. A. (1997). Differential expression of angiotensin II receptors during early renal morphogenesis. Am J Physiol 272, R662-8.

Novo, E. and Parola, M. (2008). Redox mechanisms in hepatic chronic wound healing and fibrogenesis. Fibrogenesis Tissue Repair 1, 5.

NTP (2002). NTP-CERHR expert panel report on the reproductive and developmental toxicity of methanol, Ed.^ Eds.). U.S. Department of Health and Human Services.

NTP (2004). NTP-CERHR expert panel report on the reproductive and developmental toxicity of methanol, Ed.^ Eds.). U.S. Department of Health and Human Services.

Nystrom, T. (2005). Role of oxidative carbonylation in protein quality control and senescence. EMBO J 24, 1311-7.

O'Donnell, B. V., Tew, D. G., Jones, O. T. and England, P. J. (1993). Studies on the inhibitory mechanism of iodonium compounds with special reference to neutrophil NADPH oxidase. Biochem J 290 ( Pt 1), 41-9.

O'Shea, K. S. and Kaufman, M. H. (1981). Effect of acetaldehyde on the neuroepithelium of early mouse embryos. J Anat 132, 107-18.

Offen, D., Ziv, I., Sternin, H., Melamed, E. and Hochman, A. (1996). Prevention of dopamine-induced cell death by thiol antioxidants: possible implications for treatment of Parkinson's disease. Exp Neurol 141, 32-9.

Offen, D., Ziv, I., Panet, H., Wasserman, L., Stein, R., Melamed, E. and Barzilai, A. (1997). Dopamine-induced apoptosis is inhibited in PC12 cells expressing Bcl-2. Cell Mol Neurobiol 17, 289-304.

Ogata, M. (1991). Acatalasemia. Hum Genet 86, 331-40.

Okado-Matsumoto, A. and Fridovich, I. (2001). Subcellular distribution of superoxide dismutases (SOD) in rat liver: Cu,Zn-SOD in mitochondria. J Biol Chem 276, 38388-93.

Oshino, N., Jamieson, D., Sugano, T. and Chance, B. (1975). Optical measurement of the catalase-hydrogen peroxide intermediate (Compound I) in the liver of anaesthetized rats and its implication to hydrogen peroxide production in situ. Biochem J 146, 67-77.

Otani, M. (1978). Alcohol dehydrogenase activity in animal tissues. Nichidai Igaku Zasshi. 37, 341-350.

Ozolins, T. R., Siksay, D. L. and Wells, P. G. (1996). Modulation of embryonic glutathione peroxidase activity and phenytoin teratogenicity by dietary deprivation of selenium in CD-1 mice. J Pharmacol Exp Ther 277, 945-53.

Padayatty, S. J., Katz, A., Wang, Y., Eck, P., Kwon, O., Lee, J. H., Chen, S., Corpe, C., Dutta, A., Dutta, S. K. and Levine, M. (2003). Vitamin C as an antioxidant: evaluation of its role in disease prevention. J Am Coll Nutr 22, 18- 35.

Paffenholz, R., Bergstrom, R. A., Pasutto, F., Wabnitz, P., Munroe, R. J., Jagla, W., Heinzmann, U., Marquardt, A., Bareiss, A., Laufs, J., Russ, A., Stumm, G., Schimenti, J. C. and Bergstrom, D. E. (2004). Vestibular defects in head-tilt mice result from mutations in Nox3, encoding an NADPH oxidase. Genes Dev 18, 486-91.

Palm, F., Nangaku, M., Fasching, A., Tanaka, T., Nordquist, L., Hansell, P., Kawakami, T., Nishijima, F. and Fujita, T. (2010). Uremia induces abnormal oxygen consumption in tubules and aggravates chronic hypoxia of the kidney via oxidative stress. Am J Physiol Renal Physiol 299, F380-6.

459

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, 25079-88.

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, 582-5.

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, 1001-9.

Parnell, S. E., O'Leary-Moore, S. K., Godin, E. A., Dehart, D. B., Johnson, B. W., Allan Johnson, G., Styner, M. A. and Sulik, K. K. (2009). Magnetic resonance microscopy defines ethanol-induced brain abnormalities in prenatal mice: effects of acute insult on gestational day 8. Alcohol Clin Exp Res 33, 1001-11.

Parnell, S. E., Sulik, K. K., Dehart, D. B. and Chen, S. Y. (2010). Reduction of ethanol-induced ocular abnormalities in mice through dietary administration of N-acetylcysteine. Alcohol 44, 699-705.

Parola, M., Bellomo, G., Robino, G., Barrera, G. and Dianzani, M. U. (1999). 4-Hydroxynonenal as a biological signal: molecular basis and pathophysiological implications. Antioxid Redox Signal 1, 255-84.

Parrilla, R., Okawa, K., Lindros, K. O., Zimmerman, U. J., Kobayashi, K. and Williamson, J. R. (1974). Functional compartmentation of acetaldehyde oxidation in rat liver. J Biol Chem 249, 4926-33.

Parthasarathy, N., Kumar, R. S., Manikandan, S. and Devi, R. S. (2006a). Methanol-induced oxidative stress in rat lymphoid organs. J Occup Health 48, 20-27.

Parthasarathy, N. J., Kumar, R. S., Manikandan, S. and Devi, R. S. (2006b). Methanol-induced oxidative stress in rat lymphoid organs. J Occup Health 48, 20-7.

Parthasarathy, N. J., Kumar, R. S., Manikandan, S., Narayanan, G. S., Kumar, R. V. and Devi, R. S. (2006c). Effect of methanol-induced oxidative stress on the neuroimmune system of experimental rats. Chem Biol Interact 161, 14- 25.

Paula, E. M., Mathangi, D. C. and Namasivayam, A. (2003). Free radical changes in methanol toxicity. Indian journal of physiology and pharmacology 47, 207-11.

Paz-Elizur, T., Krupsky, M., Blumenstein, S., Elinger, D., Schechtman, E. and Livneh, Z. (2003). DNA repair activity for oxidative damage and risk of lung cancer. J Natl Cancer Inst 95, 1312-9.

Peiro, C., Lafuente, N., Matesanz, N., Cercas, E., Llergo, J. L., Vallejo, S., Rodriguez-Manas, L. and Sanchez- Ferrer, C. F. (2001). High glucose induces cell death of cultured human aortic smooth muscle cells through the formation of hydrogen peroxide. Br J Pharmacol 133, 967-74.

Peng, S. L., Fatenejad, S. and Craft, J. (1997). Scleroderma: a disease related to damaged proteins? Nat Med 3, 276- 8.

Peng, Y., Yang, P. H., Guo, Y., Ng, S. S., Liu, J., Fung, P. C., Tay, D., Ge, J., He, M. L., Kung, H. F. and Lin, M. C. (2004). Catalase and peroxiredoxin 5 protect Xenopus embryos against alcohol-induced ocular anomalies. Invest Ophthalmol Vis Sci 45, 23-9.

Percy, M. E. (1984). Catalase: an old enzyme with a new role? Can J Biochem Cell Biol 62, 1006-14.

Perkins, R. A., Ward, K. W. and Pollack, G. M. (1995). A pharmacokinetic model of inhaled methanol in humans and comparison to methanol disposition in mice and rats. Environ Health Perspect 103, 726-33.

Peters, J. M., Narotsky, M. G., Elizondo, G., Fernandez-Salguero, P. M., Gonzalez, F. J. and Abbott, B. D. (1999). Amelioration of TCDD-induced teratogenesis in aryl hydrocarbon receptor (AhR)-null mice. Toxicol Sci 47, 86-92.

460

Petersen, D. R., Panter, S. S. and Collins, A. C. (1977). Ethanol and acetaldehyde metabolism in the pregnant mouse. Drug Alcohol Depend 2, 409-20.

Petkov, V. V., Stoianovski, D., Petkov, V. D. and Vyglenova, I. (1992). [Lipid peroxidation changes in the brain in fetal alcohol syndrome]. Biull Eksp Biol Med 113, 500-2. First published on Izmeneniia perekisnogo okisleniia lipidov v mozge pri fetal'nom alkogol'nom sindrome.

Philips, T. and Robberecht, W. (2011). Neuroinflammation in amyotrophic lateral sclerosis: role of glial activation in motor neuron disease. Lancet Neurol 10, 253-63.

Phillips, R. G. and LeDoux, J. E. (1992). Differential contribution of amygdala and hippocampus to cued and contextual fear conditioning. Behav Neurosci 106, 274-85.

Piccoli, C., Ria, R., Scrima, R., Cela, O., D'Aprile, A., Boffoli, D., Falzetti, F., Tabilio, A. and Capitanio, N. (2005). Characterization of mitochondrial and extra-mitochondrial oxygen consuming reactions in human hematopoietic stem cells. Novel evidence of the occurrence of NAD(P)H oxidase activity. J Biol Chem 280, 26467-76.

Pietruszko, R. (1975a). Human liver alcohol dehydrogenase--inhibition of methanol activity by pyrazole, 4- methylpyrazole, 4-hydroxymethylpyrazole and 4-carboxypyrazole. Biochem Pharmacol 24, 1603-7.

Pietruszko, R. (1975b). Mammalian liver alcohol dehydrogenases. Adv Exp Med Biol 56, 1-31.

Pikkarainen, P. H. and Raiha, N. C. (1967). Development of alcohol dehydrogenase activity in the human liver. Pediatr Res 1, 165-8.

Pogrebniak, H. W., Merino, M. J., Hahn, S. M., Mitchell, J. B. and Pass, H. I. (1992). Spin trap salvage from endotoxemia: the role of cytokine down-regulation. Surgery 112, 130-9; discussion 138-9.

Pohorecky, L. A. and Brick, J. (1988). Pharmacology of ethanol. Pharmacol Ther 36, 335-427.

Poliakov, E., Meer, S. G., Roy, S. C., Mesaros, C. and Salomon, R. G. (2004). Iso[7]LGD2-protein adducts are abundant in vivo and free radical-induced oxidation of an arachidonyl phospholipid generates this D series isolevuglandin in vitro. Chem Res Toxicol 17, 613-22.

Popova, S., Stade, B., Bekmuradov, D., Lange, S. and Rehm, J. (2011). What do we know about the economic impact of fetal alcohol spectrum disorder? A systematic literature review. Alcohol Alcohol 46, 490-7.

Porter, W. H. and Moyer, T. P. (1994). Determination of alcohols by headspace analysis. In: Burtis, C.A., Ashwood, E.R. (Eds.), Tietz Textbook of Clinical Chemistry. W.B. Saunders Company, Philadelphia, pp. 1173-1174.

Priscott, P. K. (1982). The effects of ethanol on rat embryos developing in vitro. Biochem Pharmacol 31, 3641-3.

Putnam, C. D., Arvai, A. S., Bourne, Y. and Tainer, J. A. (2000). Active and inhibited human catalase structures: ligand and NADPH binding and catalytic mechanism. J Mol Biol 296, 295-309.

Qian, L., Xu, Z., Zhang, W., Wilson, B., Hong, J. S. and Flood, P. M. (2007). Sinomenine, a natural dextrorotatory morphinan analog, is anti-inflammatory and neuroprotective through inhibition of microglial NADPH oxidase. J Neuroinflammation 4, 23.

Qian, L., Wei, S. J., Zhang, D., Hu, X., Xu, Z., Wilson, B., El-Benna, J., Hong, J. S. and Flood, P. M. (2008). Potent anti-inflammatory and neuroprotective effects of TGF-beta1 are mediated through the inhibition of ERK and p47phox-Ser345 phosphorylation and translocation in microglia. J Immunol 181, 660-8.

Quan, F., Korneluk, R. G., Tropak, M. B. and Gravel, R. A. (1986). Isolation and characterization of the human catalase gene. Nucleic Acids Res 14, 5321-35.

461

Radicella, J. P., Dherin, C., Desmaze, C., Fox, M. S. and Boiteux, S. (1997). Cloning and characterization of hOGG1, a human homolog of the OGG1 gene of Saccharomyces cerevisiae. Proc Natl Acad Sci U S A 94, 8010-5.

Radlow, R. and Hurst, P. M. (1985). Temporal relations between blood alcohol concentration and alcohol effect: an experiment with human subjects. Psychopharmacology (Berl) 85, 260-6.

Ramachandran, V., Perez, A., Chen, J., Senthil, D., Schenker, S. and Henderson, G. I. (2001). In utero ethanol exposure causes mitochondrial dysfunction, which can result in apoptotic cell death in fetal brain: a potential role for 4-hydroxynonenal. Alcohol Clin Exp Res 25, 862-71.

Ramanathan, R., Wilkemeyer, M. F., Mittal, B., Perides, G. and Charness, M. E. (1996). Alcohol inhibits cell-cell adhesion mediated by human L1. J Cell Biol 133, 381-90.

Ramirez, D. C., Gomez-Mejiba, S. E. and Mason, R. P. (2007). Immuno-spin trapping analyses of DNA radicals. Nat Protoc 2, 512-22.

Randall, C. L., Taylor, J. and Walker, D. W. (1977). Ethanol-induced malformations in mice. Alcohol Clin Exp Res 1, 219-24.

Randall, C. L., Becker, H. C. and Middaugh, L. D. (1985). Effect of prenatal ethanol exposure on activity and shuttle avoidance behavior in adult C57 mice. Alcohol Drug Res 6, 351-60.

Rang, H., Dale, M., Ritter, J. and Moore, P. (2003). Pharmacology, 5th Edition. 5 ed. Churchill Livingstone, London.

Rao, U. N., Aravindakshan, M., Satyanarayan, V. and Chauhan, P. S. (1997). Genotype- and gender-dependent hepatic alcohol dehydrogenase (ADH) activity in developing mice. Alcohol 14, 527-31.

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, 1911-8.

Reddy, J. K., Lalwani, N. D., Qureshi, S. A., Reddy, M. K. and Moehle, C. M. (1984). Induction of hepatic peroxisome proliferation in nonrodent species, including primates. The American journal of pathology 114, 171-83.

Reid, C., Edwards, J., Wang, M., Manybeads, Y., Mike, L., Martinez, N., La Grange, L. and Reyes, E. (1999). Prevention by a silymarin/phospholipid compound of ethanol-induced social learning deficits in rats. Planta Med 65, 421-4.

Reinke, L. A., Lai, E. K., DuBose, C. M. and McCay, P. B. (1987). Reactive free radical generation in vivo in heart and liver of ethanol-fed rats: correlation with radical formation in vitro. Proc Natl Acad Sci U S A 84, 9223-7.

Reinke, L. A. (2002). Spin trapping evidence for alcohol-associated oxidative stress. Free Radic Biol Med 32, 953-7.

Reiter, R. J. (1995). Oxidative processes and antioxidative defense mechanisms in the aging brain. Faseb J 9, 526- 33.

Rey, F. E., Cifuentes, M. E., Kiarash, A., Quinn, M. T. and Pagano, P. J. (2001). Novel competitive inhibitor of NAD(P)H oxidase assembly attenuates vascular O(2)(-) and systolic blood pressure in mice. Circ Res 89, 408-14.

Rhee, S. G. (1999). Redox signaling: hydrogen peroxide as intracellular messenger. Exp Mol Med 31, 53-9.

Riikonen, R., Salonen, I., Partanen, K. and Verho, S. (1999). Brain perfusion SPECT and MRI in foetal alcohol syndrome. Dev Med Child Neurol 41, 652-9.

462

Riis, B., Risom, L., Loft, S. and Poulsen, H. E. (2002). OGG1 mRNA expression and incision activity in rats are higher in foetal tissue than in adult liver tissue while 8-oxo-2'-deoxyguanosine levels are unchanged. DNA Repair (Amst) 1, 709-17.

Riley, E. P., Lochry, E. A. and Shapiro, N. R. (1979). Lack of response inhibition in rats prenatally exposed to alcohol. Psychopharmacology (Berl) 62, 47-52.

Riley, E. P., Mattson, S. N., Sowell, E. R., Jernigan, T. L., Sobel, D. F. and Jones, K. L. (1995). Abnormalities of the corpus callosum in children prenatally exposed to alcohol. Alcohol Clin Exp Res 19, 1198-202.

Riley, E. P. and McGee, C. L. (2005). Fetal alcohol spectrum disorders: an overview with emphasis on changes in brain and behavior. Exp Biol Med (Maywood) 230, 357-65.

Roberts, L. J., 2nd and Fessel, J. P. (2004). The biochemistry of the isoprostane, neuroprostane, and isofuran pathways of lipid peroxidation. Chem Phys Lipids 128, 173-86.

Robertson, A. B., Klungland, A., Rognes, T. and Leiros, I. (2009). DNA repair in mammalian cells: Base excision repair: the long and short of it. Cell Mol Life Sci 66, 981-93.

Robinson, K. A., Stewart, C. A., Pye, Q. N., Nguyen, X., Kenney, L., Salzman, S., Floyd, R. A. and Hensley, K. (1999). Redox-sensitive protein phosphatase activity regulates the phosphorylation state of p38 protein kinase in primary astrocyte culture. J Neurosci Res 55, 724-32.

Roe, O. (1955). The metabolism and toxicity of methanol. Pharmacological reviews 7, 399-412.

Roebuck, T. M., Mattson, S. N. and Riley, E. P. (1998). A review of the neuroanatomical findings in children with fetal alcohol syndrome or prenatal exposure to alcohol. Alcohol Clin Exp Res 22, 339-44.

Rogers, J. M., Mole, M. L., Chernoff, N., Barbee, B. D., Turner, C. I., Logsdon, T. R. and Kavlock, R. J. (1993). The developmental toxicity of inhaled methanol in the CD-1 mouse, with quantitative dose-response modeling for estimation of benchmark doses. Teratology 47, 175-88.

Rogers, J. M. and Mole, M. L. (1997). Critical periods of sensitivity to the developmental toxicity of inhaled methanol in the CD-1 mouse. Teratology 55, 364-72.

Rogers, J. M., Brannen, K. C., Barbee, B. D., Zucker, R. M. and Degitz, S. J. (2004). Methanol exposure during gastrulation causes holoprosencephaly, facial dysgenesis, and cervical vertebral malformations in C57BL/6J mice. Birth Defects Res B Dev Reprod Toxicol 71, 80-8.

Rojkind, M., Dominguez-Rosales, J. A., Nieto, N. and Greenwel, P. (2002). Role of hydrogen peroxide and oxidative stress in healing responses. Cell Mol Life Sci 59, 1872-91.

Roldan-Arjona, T., Wei, Y. F., Carter, K. C., Klungland, A., Anselmino, C., Wang, R. P., Augustus, M. and Lindahl, T. (1997). Molecular cloning and functional expression of a human cDNA encoding the antimutator enzyme 8- hydroxyguanine-DNA glycosylase. Proc Natl Acad Sci U S A 94, 8016-20.

Rossary, A., Arab, K. and Steghens, J. P. (2007). Polyunsaturated fatty acids modulate NOX 4 anion superoxide production in human fibroblasts. Biochem J 406, 77-83.

Ruegg, J., Swedenborg, E., Wahlstrom, D., Escande, A., Balaguer, P., Pettersson, K. and Pongratz, I. (2008). The transcription factor aryl hydrocarbon receptor nuclear translocator functions as an estrogen receptor beta-selective coactivator, and its recruitment to alternative pathways mediates antiestrogenic effects of dioxin. Mol Endocrinol 22, 304-16.

Russo, M. T., De Luca, G., Degan, P., Parlanti, E., Dogliotti, E., Barnes, D. E., Lindahl, T., Yang, H., Miller, J. H. and Bignami, M. (2004). Accumulation of the oxidative base lesion 8-hydroxyguanine in DNA of tumor-prone mice defective in both the Myh and Ogg1 DNA glycosylases. Cancer Res 64, 4411-4.

463

Russo, N. J., 2nd, Kapp, B. S., Holmquist, B. K. and Musty, R. E. (1976). Passive avoidance and amygdala lesions: relationship with pituitary-adrenal system. Physiol Behav 16, 191-9.

Sadler, T. W., Horton, W. E. and Warner, C. W. (1982). Whole embryo culture: a screening technique for teratogens? Teratog Carcinog Mutagen 2, 243-53.

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, 175-80.

Sakellariou, G. K., Vasilaki, A., Palomero, J., Kayani, A., Zibrik, L., McArdle, A. and Jackson, M. J. (2013). Studies of mitochondrial and nonmitochondrial sources implicate nicotinamide adenine dinucleotide phosphate oxidase(s) in the increased skeletal muscle superoxide generation that occurs during contractile activity. Antioxid Redox Signal 18, 603-21.

Salmeen, A., Andersen, J. N., Myers, M. P., Meng, T. C., Hinks, J. A., Tonks, N. K. and Barford, D. (2003). Redox regulation of protein tyrosine phosphatase 1B involves a sulphenyl-amide intermediate. Nature 423, 769-73.

Sanders, S. A., Eisenthal, R. and Harrison, R. (1997). NADH oxidase activity of human xanthine oxidoreductase-- generation of superoxide anion. Eur J Biochem 245, 541-8.

Sang, H., Wallis, G. L., Stewart, C. A. and Kotake, Y. (1999). Expression of cytokines and activation of transcription factors in lipopolysaccharide-administered rats and their inhibition by phenyl N-tert-butylnitrone (PBN). Arch Biochem Biophys 363, 341-8.

Sari, Y. and Zhou, F. C. (2004). Prenatal alcohol exposure causes long-term serotonin neuron deficit in mice. Alcohol Clin Exp Res 28, 941-8.

Satapati, S., Sunny, N. E., Kucejova, B., Fu, X., He, T. T., Mendez-Lucas, A., Shelton, J. M., Perales, J. C., Browning, J. D. and Burgess, S. C. (2012). Elevated TCA cycle function in the pathology of diet-induced hepatic insulin resistance and fatty liver. J Lipid Res 53, 1080-92.

Sato, M., Girard, L., Sekine, I., Sunaga, N., Ramirez, R. D., Kamibayashi, C. and Minna, J. D. (2003). Increased expression and no mutation of the Flap endonuclease (FEN1) gene in human lung cancer. Oncogene 22, 7243-6.

Sauer, H., Wartenberg, M. and Hescheler, J. (2001). Reactive oxygen species as intracellular messengers during cell growth and differentiation. Cell Physiol Biochem 11, 173-86.

Sauer, H., Ruhe, C., Muller, J. P., Schmelter, M., D'Souza, R. and Wartenberg, M. (2008). Reactive oxygen species and upregulation of NADPH oxidases in mechanotransduction of embryonic stem cells. Methods Mol Biol 477, 397- 418.

Savoy-Moore, R. T., Dombrowski, M. P., Cheng, A., Abel, E. A. and Sokol, R. J. (1989). Low dose alcohol contracts the human umbilical artery in vitro. Alcohol Clin Exp Res 13, 40-2.

Scaife, R. M. (2004). G2 cell cycle arrest, down-regulation of cyclin B, and induction of mitotic catastrophe by the flavoprotein inhibitor diphenyleneiodonium. Mol Cancer Ther 3, 1229-37.

Scaife, R. M. (2005). Selective and irreversible cell cycle inhibition by diphenyleneiodonium. Mol Cancer Ther 4, 876-84.

Schallreuter, K. U., Wood, J. M. and Berger, J. (1991). Low catalase levels in the epidermis of patients with vitiligo. J Invest Dermatol 97, 1081-5.

Schaur, R. J. (2003). Basic aspects of the biochemical reactivity of 4-hydroxynonenal. Mol Aspects Med 24, 149-59.

464

Schoeb, T. R., McConnell, E. E., Juliana, M. M., Davis, J. K., Davidson, M. K. and Lindsey, J. R. (2009). Mycoplasma pulmonis and lymphoma in bioassays in rats. Veterinary pathology 46, 952-9.

Sedeek, M., Callera, G., Montezano, A., Gutsol, A., Heitz, F., Szyndralewiez, C., Page, P., Kennedy, C. R., Burns, K. D., Touyz, R. M. and Hebert, R. L. (2010). Critical role of Nox4-based NADPH oxidase in glucose-induced oxidative stress in the kidney: implications in type 2 diabetic nephropathy. Am J Physiol Renal Physiol 299, F1348- 58.

Sedeek, M., Montezano, A. C., Hebert, R. L., Gray, S. P., Di Marco, E., Jha, J. C., Cooper, M. E., Jandeleit-Dahm, K., Schiffrin, E. L., Wilkinson-Berka, J. L. and Touyz, R. M. (2012). Oxidative stress, nox isoforms and complications of diabetes-potential targets for novel therapies. J Cardiovasc Transl Res 5, 509-18.

Seitz, H. K. and Poschl, G. (1997). The role of gastrointestinal factors in alcohol metabolism. Alcohol Alcohol 32, 543-9.

Selemidis, S., Sobey, C. G., Wingler, K., Schmidt, H. H. and Drummond, G. R. (2008). NADPH oxidases in the vasculature: molecular features, roles in disease and pharmacological inhibition. Pharmacol Ther 120, 254-91.

Sequeira Lopez, M. L. and Gomez, R. A. (2004). The role of angiotensin II in kidney embryogenesis and kidney abnormalities. Curr Opin Nephrol Hypertens 13, 117-22.

Shaffer, J. B., Sutton, R. B. and Bewley, G. C. (1987). Isolation of a cDNA clone for murine catalase and analysis of an acatalasemic mutant. J Biol Chem 262, 12908-11.

Shaham, J., Bomstein, Y., Meltzer, A., Kaufman, Z., Palma, E. and Ribak, J. (1996). DNA--protein crosslinks, a biomarker of exposure to formaldehyde--in vitro and in vivo studies. Carcinogenesis 17, 121-5.

Shertzer, H. G., Nebert, D. W., Puga, A., Ary, M., Sonntag, D., Dixon, K., Robinson, L. J., Cianciolo, E. and Dalton, T. P. (1998). Dioxin causes a sustained oxidative stress response in the mouse. Biochem Biophys Res Commun 253, 44-8.

Shi, J., Ross, C. R., Leto, T. L. and Blecha, F. (1996). PR-39, a proline-rich antibacterial peptide that inhibits phagocyte NADPH oxidase activity by binding to Src homology 3 domains of p47 phox. Proc Natl Acad Sci U S A 93, 6014-8.

Shibutani, S., Takeshita, M. and Grollman, A. P. (1991). Insertion of specific bases during DNA synthesis past the oxidation-damaged base 8-oxodG. Nature 349, 431-4.

Shimizu, T., Kondo, K. and Hayaishi, O. (1981). Role of prostaglandin endoperoxides in the serum thiobarbituric acid reaction. Arch Biochem Biophys 206, 271-6.

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, 5-9.

Shiose, A., Kuroda, J., Tsuruya, K., Hirai, M., Hirakata, H., Naito, S., Hattori, M., Sakaki, Y. and Sumimoto, H. (2001). A novel superoxide-producing NAD(P)H oxidase in kidney. J Biol Chem 276, 1417-23.

Shringarpure, R., Grune, T., Mehlhase, J. and Davies, K. J. (2003). Ubiquitin conjugation is not required for the degradation of oxidized proteins by proteasome. J Biol Chem 278, 311-8.

Siew, C., Deitrich, R. A. and Erwin, V. G. (1976). Localization and characteristics of rat liver mitochondrial aldehyde dehydrogenases. Arch Biochem Biophys 176, 638-49.

Simons, J. M., Hart, L. A., van Dijk, H., Fischer, F. C., de Silva, K. T. and Labadie, R. P. (1989). Immunodulatory compounds from Picrorhiza kurroa: isolation and characterization of two anti-complementary polymeric fractions from an aqueous root extract. J Ethnopharmacol 26, 169-82.

465

Singh, B., Chatterjee, A., Ronghe, A. M., Bhat, N. K. and Bhat, H. K. (2013). Antioxidant-mediated up-regulation of OGG1 via NRF2 induction is associated with inhibition of oxidative DNA damage in estrogen-induced breast cancer. BMC Cancer 13, 253.

Singh, S. P., Pullen, G. L., Srivenugopal, K. S., Yuan, X. H. and Snyder, A. K. (1992). Decreased glucose transporter 1 gene expression and glucose uptake in fetal brain exposed to ethanol. Life Sci 51, 527-36.

Siu, M. and Wells, P. G. (2010). Catalase in the mechanism of methanol developmental toxicity. Toxicol Sci (Supplement: The Toxicologist) 114, 183 (Abstract No. 858).

Siu, M., Shapiro, A. M., Wiley, M. J. and Wells, P. G. (2013a). A role for glutathione, independent of oxidative stress, in the developmental toxicity of methanol (submitted).

Siu, M. T., Wiley, M. J. and Wells, P. G. (2013b). Methanol teratogenicity in mutant mice with deficient catalase activity and transgenic mice expressing human catalase. Reprod Toxicol 36, 33-9.

Sivan, E., Lee, Y. C., Wu, Y. K. and Reece, E. A. (1997). Free radical scavenging enzymes in fetal dysmorphogenesis among offspring of diabetic rats. Teratology 56, 343-9.

Skrzydlewska, E. and Farbiszewski, R. (1997a). Decreased antioxidant defense mechanisms in rat liver after methanol intoxication. Free Radic Res 27, 369-75.

Skrzydlewska, E. and Farbiszewski, R. (1997b). Trolox-derivative antioxidant protects against methanol-induced damage. Fundam Clin Pharmacol 11, 460-5.

Skrzydlewska, E. and Farbiszewski, R. (1997c). Glutathione consumption and inactivation of glutathione-related enzymes in liver, erythrocytes and serum of rats after methanol intoxication. Arch Toxicol 71, 741-5.

Skrzydlewska, E. and Farbiszewski, R. (1997d). Antioxidant status of liver, erythrocytes, and blood serum of rats in acute methanol intoxication. Alcohol 14, 431-7.

Skrzydlewska, E. and Farbiszewski, R. (1997e). Liver and serum antioxidant status after methanol intoxication in rats. Acta Biochim Pol 44, 139-45.

Skrzydlewska, E. and Farbiszewski, R. (1998). Lipid peroxidation and antioxidant status in the liver, erythrocytes, and serum of rats after methanol intoxication. J Toxicol Environ Health A 53, 637-49.

Skrzydlewska, E., Witek, A. and Farbiszewski, R. (1998). The comparison of the antioxidant defense potential of brain to liver of rats after methanol ingestion. Comp Biochem Physiol C Pharmacol Toxicol Endocrinol 120, 289-94.

Skrzydlewska, E., Elas, M., Farbiszewski, R. and Roszkowska, A. (2000). Effect of methanol intoxication on free- radical induced protein oxidation. J Appl Toxicol 20, 239-43.

Smith, E. N. and Taylor, R. T. (1982). Acute toxicity of methanol in the folate-deficient acatalasemic mouse. Toxicology 25, 271-87.

Smith, G. D., Shaw, L. J., Maini, P. K., Ward, R. J., Peters, T. J. and Murray, J. D. (1993). Mathematical modelling of ethanol metabolism in normal subjects and chronic alcohol misusers. Alcohol Alcohol 28, 25-32.

Snyder, A. K. and Singh, S. P. (1989). Effects of ethanol on glucose turnover in pregnant rats. Metabolism 38, 149- 52.

Snyder, A. K., Jiang, F. and Singh, S. P. (1992). Effects of ethanol on glucose utilization by cultured mammalian embryos. Alcohol Clin Exp Res 16, 466-70.

466

Soffritti, M., Belpoggi, F., Tibaldi, E., Esposti, D. D. and Lauriola, M. (2007). Life-span exposure to low doses of aspartame beginning during prenatal life increases cancer effects in rats. Environmental health perspectives 115, 1293-7.

Sokoro, A. A., Zhang, Z., Eichhorst, J. C., Zello, G. A., House, J. D., Alcorn, J. and Lehotay, D. C. (2008). Formate pharmacokinetics during formate administration in folate-deficient young swine. Metabolism 57, 920-6.

Sorce, S., Krause, K. H. and Jaquet, V. (2012). Targeting NOX enzymes in the central nervous system: therapeutic opportunities. Cell Mol Life Sci 69, 2387-407.

Sowell, E. R., Jernigan, T. L., Mattson, S. N., Riley, E. P., Sobel, D. F. and Jones, K. L. (1996). Abnormal development of the cerebellar vermis in children prenatally exposed to alcohol: size reduction in lobules I-V. Alcohol Clin Exp Res 20, 31-4.

Sowell, E. R., Mattson, S. N., Thompson, P. M., Jernigan, T. L., Riley, E. P. and Toga, A. W. (2001). Mapping callosal morphology and cognitive correlates: effects of heavy prenatal alcohol exposure. Neurology 57, 235-44.

Sparrow, M. G., Roggendorf, H. and Vogel, W. H. (1987). Effect of ethanol on heart rate and blood pressure in nonstressed and stressed rats. Life Sci 40, 2551-9.

Spiteller, G. (1998). Linoleic acid peroxidation--the dominant lipid peroxidation process in low density lipoprotein-- and its relationship to chronic diseases. Chem Phys Lipids 95, 105-62.

Squire, L. R. (1992). Memory and the hippocampus: a synthesis from findings with rats, monkeys, and humans. Psychol Rev 99, 195-231.

Sreenathan, R. N., Padmanabhan, R. and Singh, S. (1982). Teratogenic effects of acetaldehyde in the rat. Drug Alcohol Depend 9, 339-50.

Sreenathan, R. N., Singh, S. and Padmanabhan, R. (1984). Effect of acetaldehyde on skeletogenesis in rats. Drug Alcohol Depend 14, 165-74.

Stade, B., Ali, A., Bennett, D., Campbell, D., Johnston, M., Lens, C., Tran, S. and Koren, G. (2009). The burden of prenatal exposure to alcohol: revised measurement of cost. Can J Clin Pharmacol 16, e91-102.

Stanton, M. E., Crofton, K. M., Gray, L. E., Gordon, C. J., Boyes, W. K., Mole, M. L., Peele, D. B. and Bushnell, P. J. (1995). Assessment of offspring development and behavior following gestational exposure to inhaled methanol in the rat. Fundam Appl Toxicol 28, 100-10.

Stewart, C. A., Hyam, K., Wallis, G., Sang, H., Robinson, K. A., Floyd, R. A., Kotake, Y. and Hensley, K. (1999). Phenyl-N-tert-butylnitrone demonstrates broad-spectrum inhibition of apoptosis-associated gene expression in endotoxin-treated rats. Arch Biochem Biophys 365, 71-4.

Stolk, J., Hiltermann, T. J., Dijkman, J. H. and Verhoeven, A. J. (1994). Characteristics of the inhibition of NADPH oxidase activation in neutrophils by apocynin, a methoxy-substituted catechol. Am J Respir Cell Mol Biol 11, 95- 102.

Stubley-Weatherly, L., Harding, J. W. and Wright, J. W. (1996). Effects of discrete kainic acid-induced hippocampal lesions on spatial and contextual learning and memory in rats. Brain Res 716, 29-38.

Stucki, M., Pascucci, B., Parlanti, E., Fortini, P., Wilson, S. H., Hubscher, U. and Dogliotti, E. (1998). Mammalian base excision repair by DNA polymerases delta and epsilon. Oncogene 17, 835-43.

Stuehr, D. J., Fasehun, O. A., Kwon, N. S., Gross, S. S., Gonzalez, J. A., Levi, R. and Nathan, C. F. (1991). Inhibition of macrophage and endothelial cell nitric oxide synthase by diphenyleneiodonium and its analogs. Faseb J 5, 98-103.

467

Sugo, N., Aratani, Y., Nagashima, Y., Kubota, Y. and Koyama, H. (2000). Neonatal lethality with abnormal neurogenesis in mice deficient in DNA polymerase beta. EMBO J 19, 1397-404.

Suh, Y. A., Arnold, R. S., Lassegue, B., Shi, J., Xu, X., Sorescu, D., Chung, A. B., Griendling, K. K. and Lambeth, J. D. (1999). Cell transformation by the superoxide-generating oxidase Mox1. Nature 401, 79-82.

Suit, P. F. and Estes, M. L. (1990). Methanol intoxication: clinical features and differential diagnosis. Cleveland Clinic journal of medicine 57, 464-71.

Sulik, K. K. (2005). Genesis of alcohol-induced craniofacial dysmorphism. Exp Biol Med (Maywood) 230, 366-75.

Sutker, P. B., Tabakoff, B., Goist, K. C., Jr. and Randall, C. L. (1983). Acute alcohol intoxication, mood states and alcohol metabolism in women and men. Pharmacol Biochem Behav 18 Suppl 1, 349-54.

Sweeting, J. N., Siu, M., McCallum, G. P., Miller, L. and Wells, P. G. (2010). Species differences in methanol and formic acid pharmacokinetics in mice, rabbits and primates. Toxicol Appl Pharmacol 247, 28-35.

Sweeting, J. N., Siu, M., Wiley, M. J. and Wells, P. G. (2011). Species- and strain-dependent teratogenicity of methanol in rabbits and mice. Reprod Toxicol 31, 50-8.

Swift, R. and Davidson, D. (1998). Alcohol hangover: mechanisms and mediators. Alcohol Health Res World 22, 54-60. t Hart, B. A., Simons, J. M., Knaan-Shanzer, S., Bakker, N. P. and Labadie, R. P. (1990). Antiarthritic activity of the newly developed neutrophil oxidative burst antagonist apocynin. Free Radic Biol Med 9, 127-31.

Tabatabaie, T., Graham, K. L., Vasquez, A. M., Floyd, R. A. and Kotake, Y. (2000). Inhibition of the cytokine- mediated inducible nitric oxide synthase expression in rat insulinoma cells by phenyl N-tert-butylnitrone. Nitric Oxide 4, 157-67.

Takac, I., Schroder, K., Zhang, L., Lardy, B., Anilkumar, N., Lambeth, J. D., Shah, A. M., Morel, F. and Brandes, R. P. (2011). The E-loop is involved in hydrogen peroxide formation by the NADPH oxidase Nox4. J Biol Chem 286, 13304-13.

Takahara, S. (1952). Progressive oral gangrene probably due to lack of catalase in the blood (acatalasaemia); report of nine cases. Lancet 2, 1101-4.

Takahara, S. (1971). Actalasemia and hypocatalasemia in the Orient. Semin Hematol 8, 397-416.

Tazzeo, T., Worek, F. and Janssen, L. (2009). The NADPH oxidase inhibitor diphenyleneiodonium is also a potent inhibitor of cholinesterases and the internal Ca(2+) pump. Br J Pharmacol 158, 790-6. ten Freyhaus, H., Huntgeburth, M., Wingler, K., Schnitker, J., Baumer, A. T., Vantler, M., Bekhite, M. M., Wartenberg, M., Sauer, H. and Rosenkranz, S. (2006). Novel Nox inhibitor VAS2870 attenuates PDGF-dependent smooth muscle cell chemotaxis, but not proliferation. Cardiovasc Res 71, 331-41.

Teng, S., Beard, K., Pourahmad, J., Moridani, M., Easson, E., Poon, R. and O'Brien, P. J. (2001). The formaldehyde metabolic detoxification enzyme systems and molecular cytotoxic mechanism in isolated rat hepatocytes. Chem Biol Interact 130-132, 285-96.

Tephly, T. R., Parks, R. E. and Mannering, G. J. (1964). Methanol metabolism in the rat. Journal of Pharmacology and Experimental Therapeutics 143, 292-300.

Tephly, T. R. (1991). The toxicity of methanol. Life Sci 48, 1031-41.

Terelius, Y., Norsten-Hoog, C., Cronholm, T. and Ingelman-Sundberg, M. (1991). Acetaldehyde as a substrate for ethanol-inducible cytochrome P450 (CYP2E1). Biochem Biophys Res Commun 179, 689-94.

468

Teschke, R., Hasumura, Y. and Lieber, C. S. (1974). NADPH-dependent oxidation of methanol, ethanol, propanol and butanol by hepatic microsomes. Biochem Biophys Res Commun 60, 851-7.

Tew, D. G. (1993). Inhibition of cytochrome P450 reductase by the diphenyliodonium cation. Kinetic analysis and covalent modifications. Biochemistry 32, 10209-15.

Thackaberry, E. A., Nunez, B. A., Ivnitski-Steele, I. D., Friggins, M. and Walker, M. K. (2005). Effect of 2,3,7,8- tetrachlorodibenzo-p-dioxin on murine heart development: alteration in fetal and postnatal cardiac growth, and postnatal cardiac chronotropy. Toxicol Sci 88, 242-9.

Thannickal, V. J. and Fanburg, B. L. (2000). Reactive oxygen species in cell signaling. Am J Physiol Lung Cell Mol Physiol 279, L1005-28.

Thieden, H. I. (1971). The effect of ethanol concentration on ethanol oxidation rate in rat liver slices. Acta Chem Scand 25, 3421-7.

Thomasson, H. (2002). Gender Differences in Alcohol Metabolism (M. Galanter, H. Begleiter, R. Deitrich, D. Gallant, D. Goodwin, E. Gottheil, A. Paredes, M. Rothschild, D. Thiel and H. Edwards, Ed.^ Eds.), pp. 163-179. Springer US.

Thurman, R. G. and Handler, J. A. (1989). New perspectives in catalase-dependent ethanol metabolism. Drug Metab Rev 20, 679-88.

Tomas, M., Fornas, E., Megias, L., Duran, J. M., Portoles, M., Guerri, C., Egea, G. and Renau-Piqueras, J. (2002). Ethanol impairs monosaccharide uptake and glycosylation in cultured rat astrocytes. J Neurochem 83, 601-12.

Tomas, M., Lazaro-Dieguez, F., Duran, J. M., Marin, P., Renau-Piqueras, J. and Egea, G. (2003). Protective effects of lysophosphatidic acid (LPA) on chronic ethanol-induced injuries to the cytoskeleton and on glucose uptake in rat astrocytes. J Neurochem 87, 220-9.

Tran, M. N., Wu, A. H. and Hill, D. W. (2007). Alcohol dehydrogenase and catalase content in perinatal infant and adult livers: potential influence on neonatal alcohol metabolism. Toxicology letters 169, 245-52.

Traves, C. and Lopez-Tejero, D. (1994). Ethanol elimination in alcohol-treated pregnant rats. Alcohol Alcohol 29, 385-95.

Traves, C., Camps, L. and Lopez-Tejero, D. (1995). Liver alcohol dehydrogenase activity and ethanol levels during chronic ethanol intake in pregnant rats and their offspring. Pharmacol Biochem Behav 52, 93-9.

Tufro-McReddie, A., Harrison, J. K., Everett, A. D. and Gomez, R. A. (1993). Ontogeny of type 1 angiotensin II receptor gene expression in the rat. J Clin Invest 91, 530-7.

Tyl, R. W., Price, C. J., Marr, M. C., Myers, C. B., Seely, J. C., Heindel, J. J. and Schwetz, B. A. (1993). Developmental toxicity evaluation of ethylene glycol by gavage in New Zealand white rabbits. Fundam. Appl. Toxicol. 20, 402-12.

Tyler, D. D. (1975). Polarographic assay and intracellular distribution of superoxide dismutase in rat liver. Biochem J 147, 493-504.

Uchida, K. and Stadtman, E. R. (1992). Modification of histidine residues in proteins by reaction with 4- hydroxynonenal. Proc Natl Acad Sci U S A 89, 4544-8.

Uecker, A. and Nadel, L. (1996). Spatial locations gone awry: object and spatial memory deficits in children with fetal alcohol syndrome. Neuropsychologia 34, 209-23.

469

Ueno, N., Takeya, R., Miyano, K., Kikuchi, H. and Sumimoto, H. (2005). The NADPH oxidase Nox3 constitutively produces superoxide in a p22phox-dependent manner: its regulation by oxidase organizers and activators. J Biol Chem 280, 23328-39.

Ufer, C., Wang, C. C., Borchert, A., Heydeck, D. and Kuhn, H. (2010). Redox control in mammalian embryo development. Antioxid Redox Signal 13, 833-75.

Uotila, L. and Koivusalo, M. (1974). Formaldehyde dehydrogenase from human liver. Purification, properties, and evidence for the formation of glutathione thiol esters by the enzyme. J Biol Chem 249, 7653-63.

Uppstad, H., Ovrebo, S., Haugen, A. and Mollerup, S. (2010). Importance of CYP1A1 and CYP1B1 in bioactivation of benzo[a]pyrene in human lung cell lines. Toxicol Lett 192, 221-8.

Vallet, P., Charnay, Y., Steger, K., Ogier-Denis, E., Kovari, E., Herrmann, F., Michel, J. P. and Szanto, I. (2005). Neuronal expression of the NADPH oxidase NOX4, and its regulation in mouse experimental brain ischemia. Neuroscience 132, 233-8. van der Kemp, P. A., Thomas, D., Barbey, R., de Oliveira, R. and Boiteux, S. (1996). Cloning and expression in Escherichia coli of the OGG1 gene of Saccharomyces cerevisiae, which codes for a DNA glycosylase that excises 7,8-dihydro-8-oxoguanine and 2,6-diamino-4-hydroxy-5-N-methylformamidopyrimidine. Proc Natl Acad Sci U S A 93, 5197-202.

Vaquer, G., Magous, R., Cros, G., Wojtusciszyn, A., Renard, E., Chevassus, H., Petit, P., Lajoix, A. D. and Oiry, C. (2012). Short-term intravenous insulin infusion is associated with reduced expression of NADPH oxidase p47(phox) subunit in monocytes from type 2 diabetes patients. Fundam Clin Pharmacol.

Verjat, T., Dhenaut, A., Radicella, J. P. and Araneda, S. (2000). Detection of 8-oxoG DNA glycosylase activity and OGG1 transcripts in the rat CNS. Mutat Res 460, 127-38.

Vetrano, A. M., Heck, D. E., Mariano, T. M., Mishin, V., Laskin, D. L. and Laskin, J. D. (2005). Characterization of the oxidase activity in mammalian catalase. J Biol Chem 280, 35372-81.

Vezina, C. M., Lin, T. M. and Peterson, R. E. (2009). AHR signaling in prostate growth, morphogenesis, and disease. Biochem Pharmacol 77, 566-76.

Viau, A. T., Abuchowski, A., Greenspan, S. and Davis, F. F. (1986). Safety evaluation of free radical scavengers PEG-catalase and PEG-superoxide dismutase. J Free Radic Biol Med 2, 283-8.

Vieira, I., Sonnier, M. and Cresteil, T. (1996). Developmental expression of CYP2E1 in the human liver. Hypermethylation control of gene expression during the neonatal period. Eur J Biochem 238, 476-83.

Vignais, P. V. (2002). The superoxide-generating NADPH oxidase: structural aspects and activation mechanism. Cell Mol Life Sci 59, 1428-59.

Vinet, B. (1987). An enzymic assay for the specific determination of methanol in serum. Clinical chemistry 33, 2204-8.

Vink, J., Auth, J., Abebe, D. T., Brenneman, D. E. and Spong, C. Y. (2005). Novel peptides prevent alcohol-induced spatial learning deficits and proinflammatory cytokine release in a mouse model of fetal alcohol syndrome. Am J Obstet Gynecol 193, 825-9.

Viswanathan, A. and Doetsch, P. W. (1998). Effects of nonbulky DNA base damages on Escherichia coli RNA polymerase-mediated elongation and promoter clearance. J Biol Chem 273, 21276-81.

Von Burg, R. (1994). Methanol. J Appl Toxicol 14, 309-13.

470

Wagner, F. W., Burger, A. R. and Vallee, B. L. (1983). Kinetic properties of human liver alcohol dehydrogenase: oxidation of alcohols by class I isoenzymes. Biochemistry 22, 1857-63.

Wallage, H. R. and Watterson, J. H. (2008). Formic acid and methanol concentrations in death investigations. Journal of analytical toxicology 32, 241-7.

Wallgren, H. and Barry, H. (1970). Actions of Alcohol: Biochemical, Physiological and Psychological Aspects. Elsevier Publishing Company, Amsterdam.

Wan, J. and Winn, L. M. (2008). In utero exposure to benzene increases embryonic c-Myb and Pim-1 protein levels in CD-1 mice. Toxicol Appl Pharmacol 228, 326-33.

Wang, M., Cheng, G., Villalta, P., Kassie, F. and Hecht, S. (2008). Analysis of formaldehyde-DNA adducts in leukocyte DNA by liquid chromatography-electrospray ionization-tandem mass spectrometry: Application to rats treated with aspartame. AARC, San Diego, CA. Abstract 2250.

Wang, Z., Chandrasena, E. R., Yuan, Y., Peng, K. W., van Breemen, R. B., Thatcher, G. R. and Bolton, J. L. (2010). Redox cycling of catechol estrogens generating apurinic/apyrimidinic sites and 8-oxo-deoxyguanosine via reactive oxygen species differentiates equine and human estrogens. Chem Res Toxicol 23, 1365-73.

Weber, D. S., Rocic, P., Mellis, A. M., Laude, K., Lyle, A. N., Harrison, D. G. and Griendling, K. K. (2005). Angiotensin II-induced hypertrophy is potentiated in mice overexpressing p22phox in vascular smooth muscle. Am J Physiol Heart Circ Physiol 288, H37-42.

Webster, W. S., Walsh, D. A., McEwen, S. E. and Lipson, A. H. (1983). Some teratogenic properties of ethanol and acetaldehyde in C57BL/6J mice: implications for the study of the fetal alcohol syndrome. Teratology 27, 231-43.

Webster, W. S., Brown-Woodman, P. D. and Ritchie, H. E. (1997). A review of the contribution of whole embryo culture to the determination of hazard and risk in teratogenicity testing. Int J Dev Biol 41, 329-35.

Wells, P. G., Zubovits, J. T., Wong, S. T., Molinari, L. M. and Ali, S. (1989). Modulation of phenytoin teratogenicity and embryonic covalent binding by acetylsalicylic acid, caffeic acid, and alpha-phenyl-N-t- butylnitrone: implications for bioactivation by prostaglandin synthetase. Toxicol Appl Pharmacol 97, 192-202.

Wells, P. G. and Winn, L. M. (1996). Biochemical toxicology of chemical teratogenesis. Crit Rev Biochem Mol Biol 31, 1-40.

Wells, P. G., Kim, P. M., Laposa, R. R., Nicol, C. J., Parman, T. and Winn, L. M. (1997). Oxidative damage in chemical teratogenesis. Mutat Res 396, 65-78.

Wells, P. G., Bhuller, Y., Chen, C. S., Jeng, W., Kasapinovic, S., Kennedy, J. C., Kim, P. M., Laposa, R. R., McCallum, G. P., Nicol, C. J., Parman, T., Wiley, M. J. and Wong, A. W. (2005). Molecular and biochemical mechanisms in teratogenesis involving reactive oxygen species. Toxicol Appl Pharmacol 207, 354-66.

Wells, P. G., Lee, C. J. J., McCallum, G. P., Perstin, J. and Harper, P. A. (2009a). Chapter 6. Receptor- and reactive intermediate-mediated mechanisms of teratogenesis. Vol 196, pp 131-162. (J. Uetrecht, Ed.^ Eds.). Springer, Heidelberg.

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. (2009b). Oxidative stress in developmental origins of disease: teratogenesis, neurodevelopmental deficits, and cancer. Toxicol Sci 108, 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, 103-9.

471

Wells, P. G., McCallum, G. P., Miller, L., Siu, M. T. and Sweeting, J. N. (2013). Oxidative stress and species differences in the metabolism, differences in the metabolism, developmental toxicity and carcinogenic potential of methanol and ethanol (J. J. Clary, Ed.^ Eds.), pp. 169-253. Wiley, Hoboken, U.S.A.

Wentzel, P. and Eriksson, U. J. (2006). Ethanol-induced fetal dysmorphogenesis in the mouse is diminished by high antioxidative capacity of the mother. Toxicol Sci 92, 416-22.

Wentzel, P. and Eriksson, U. J. (2008). Genetic influence on dysmorphogenesis in embryos from different rat strains exposed to ethanol in vivo and in vitro. Alcohol Clin Exp Res 32, 874-87.

Weston, W. M., Greene, R. M., Uberti, M. and Pisano, M. M. (1994). Ethanol effects on embryonic craniofacial growth and development: implications for study of the fetal alcohol syndrome. Alcohol Clin Exp Res 18, 177-82.

Wiberg, G. S., Samson, J. M., Maxwell, W. B., Coldwell, B. B. and Trenholm, H. L. (1971). Further studies on the acute toxicity of ethanol in young and old rats: relative importance of pulmonary excretion and total body water. Toxicol Appl Pharmacol 20, 22-9.

Wigal, T. and Amsel, A. (1990). Behavioral and neuroanatomical effects of prenatal, postnatal, or combined exposure to ethanol in weanling rats. Behav Neurosci 104, 116-26.

Wilcox, C. S., Welch, W. J., Murad, F., Gross, S. S., Taylor, G., Levi, R. and Schmidt, H. H. (1992). Nitric oxide synthase in macula densa regulates glomerular capillary pressure. Proc Natl Acad Sci U S A 89, 11993-7.

Wilkemeyer, M. F. and Charness, M. E. (1998). Characterization of ethanol-sensitive and insensitive fibroblast cell lines expressing human L1. J Neurochem 71, 2382-91.

Wilkinson, P. K. and Rheingold, J. L. (1981). Arterial-venous blood alcohol concentration gradients. J Pharmacokinet Biopharm 9, 279-307.

Wind, S., Beuerlein, K., Eucker, T., Muller, H., Scheurer, P., Armitage, M. E., Ho, H., Schmidt, H. H. and Wingler, K. (2010). Comparative pharmacology of chemically distinct NADPH oxidase inhibitors. Br J Pharmacol 161, 885- 98.

Winkler, K., Lundquist, F. and Tygstrup, N. (1969). The hepatic metabolism of ethanol in patients with cirrhosis of the liver. Scand J Clin Lab Invest 23, 59-69.

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, 112-20.

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, 607-21.

Winn, L. M. and Wells, P. G. (1999). Maternal administration of superoxide dismutase and catalase in phenytoin teratogenicity. Free Radic Biol Med 26, 266-74.

Winterbourn, C. C. (1995). Toxicity of iron and hydrogen peroxide: the Fenton reaction. Toxicol Lett 82-83, 969-74.

Witelson, S. F. (1985). The brain connection: the corpus callosum is larger in left-handers. Science 229, 665-8.

Wolff, S., Harper, P. A., Wong, J. M., Mostert, V., Wang, Y. and Abel, J. (2001). Cell-specific regulation of human aryl hydrocarbon receptor expression by transforming growth factor-beta(1). Mol Pharmacol 59, 716-24.

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, 9047-54.

472

Wood, Z. A., Schroder, E., Robin Harris, J. and Poole, L. B. (2003). Structure, mechanism and regulation of peroxiredoxins. Trends Biochem Sci 28, 32-40.

Wu, D. and Cederbaum, A. I. (2003). Alcohol, oxidative stress, and free radical damage. Alcohol Res Health 27, 277-84.

Wu, D. C., Jackson-Lewis, V., Vila, M., Tieu, K., Teismann, P., Vadseth, C., Choi, D. K., Ischiropoulos, H. and Przedborski, S. (2002). Blockade of microglial activation is neuroprotective in the 1-methyl-4-phenyl-1,2,3,6- tetrahydropyridine mouse model of Parkinson disease. J Neurosci 22, 1763-71.

Wynter, J. M., Walsh, D. A., Webster, W. S., McEwen, S. E. and Lipson, A. H. (1983). Teratogenesis after acute alcohol exposure in cultured rat embryos. Teratog Carcinog Mutagen 3, 421-8.

Xie, Y., Yang, H., Cunanan, C., Okamoto, K., Shibata, D., Pan, J., Barnes, D. E., Lindahl, T., McIlhatton, M., Fishel, R. and Miller, J. H. (2004). Deficiencies in mouse Myh and Ogg1 result in tumor predisposition and G to T mutations in codon 12 of the K-ras oncogene in lung tumors. Cancer Res 64, 3096-102.

Xu, Y., Xiao, R. and Li, Y. (2005). Effect of ethanol on the development of visceral yolk sac. Hum Reprod 20, 2509-16.

Yang, S., Yang, J., Yang, Z., Chen, P., Fraser, A., Zhang, W., Pang, H., Gao, X., Wilson, B., Hong, J. S. and Block, M. L. (2006). Pituitary adenylate cyclase-activating polypeptide (PACAP) 38 and PACAP4-6 are neuroprotective through inhibition of NADPH oxidase: potent regulators of microglia-mediated oxidative stress. J Pharmacol Exp Ther 319, 595-603.

Yoshikawa, T., Minamiyama, Y., Naito, Y. and Kondo, M. (1994). Antioxidant properties of bromocriptine, a dopamine agonist. J Neurochem 62, 1034-8.

Youssef, A. F., Baggs, R. B., Weiss, B. and Miller, R. K. (1997). Teratogenicity of methanol following a single oral dose in Long-Evans rats. Reprod Toxicol 11, 503-10.

Yu, W. K. and Wells, P. G. (1995). Evidence for lipoxygenase-catalyzed bioactivation of phenytoin to a teratogenic reactive intermediate: in vitro studies using linoleic acid-dependent soybean lipoxygenase, and in vivo studies using pregnant CD-1 mice. Toxicol Appl Pharmacol 131, 1-12.

Yuan, J. (1993). Estimation of variance for AUC in animal studies. Journal of pharmaceutical sciences 82, 761-3.

Zhang, D., Hu, X., Wei, S. J., Liu, J., Gao, H., Qian, L., Wilson, B., Liu, G. and Hong, J. S. (2008). Squamosamide derivative FLZ protects dopaminergic neurons against inflammation-mediated neurodegeneration through the inhibition of NADPH oxidase activity. J Neuroinflammation 5, 21.

Zhang, F., Qian, L., Flood, P. M., Shi, J. S., Hong, J. S. and Gao, H. M. (2010a). Inhibition of IkappaB kinase-beta protects dopamine neurons against lipopolysaccharide-induced neurotoxicity. J Pharmacol Exp Ther 333, 822-33.

Zhang, F., Shi, J. S., Zhou, H., Wilson, B., Hong, J. S. and Gao, H. M. (2010b). Resveratrol protects dopamine neurons against lipopolysaccharide-induced neurotoxicity through its anti-inflammatory actions. Mol Pharmacol 78, 466-77.

Zhang, F. X., Rubin, R. and Rooney, T. A. (1998). Ethanol induces apoptosis in cerebellar granule neurons by inhibiting insulin-like growth factor 1 signaling. J Neurochem 71, 196-204.

Zhang, M., Brewer, A. C., Schroder, K., Santos, C. X., Grieve, D. J., Wang, M., Anilkumar, N., Yu, B., Dong, X., Walker, S. J., Brandes, R. P. and Shah, A. M. (2010c). NADPH oxidase-4 mediates protection against chronic load- induced stress in mouse hearts by enhancing angiogenesis. Proc Natl Acad Sci U S A 107, 18121-6.

473

Zhang, M., Perino, A., Ghigo, A., Hirsch, E. and Shah, A. (2012). NADPH oxidases in heart failure: Poachers or gamekeepers? Antioxid Redox Signal.

Zhang, W., Wang, T., Qin, L., Gao, H. M., Wilson, B., Ali, S. F., Hong, J. S. and Liu, B. (2004). Neuroprotective effect of dextromethorphan in the MPTP Parkinson's disease model: role of NADPH oxidase. Faseb J 18, 589-91.

Zhou, F. C., Sari, Y., Li, T. K., Goodlett, C. and Azmitia, E. C. (2002). Deviations in brain early serotonergic development as a result of fetal alcohol exposure. Neurotox Res 4, 337-42.

Zimmerberg, B., Mattson, S. and Riley, E. P. (1989). Impaired alternation test performance in adult rats following prenatal alcohol exposure. Pharmacol Biochem Behav 32, 293-9.

Zimmerman, E. F., Potturi, R. B., Resnick, E. and Fisher, J. E. (1994). Role of oxygen free radicals in cocaine- induced vascular disruption in mice. Teratology 49, 192-201.

Zusterzeel, P. L., Mulder, T. P., Peters, W. H., Wiseman, S. A. and Steegers, E. A. (2000). Plasma protein carbonyls in nonpregnant, healthy pregnant and preeclamptic women. Free Radic Res 33, 471-6.



474

Chapter 11 Appendices

11.1 Species differences in methanol and formic acid pharmacokinetics in mice, rabbits and primates…………………………………………………………………………………..p. 475

11.2 Oxidative stress and species differences in the metabolism, developmental toxicity and carcinogenic potential of methanol and ethanol…………………………………………p. 512

11.3 In vivo teratogenesis pictures………………………………………………………p. 602

475

11.1 Species differences in methanol and formic acid pharmacokinetics in mice, rabbits and primates

Running title: Species-dependent methanol metabolism

J. Nicole Sweetingd*, Michelle Siud*, Gordon P. McCallumd*, Lutfiya Miller† and Peter G. Wells*†

*Faculty of Pharmacy and

†Department of Pharmacology and Toxicology

University of Toronto

Toronto, Ontario, Canada a. Preliminary reports of this research were presented at the 2007, 2008 and 2009 annual meetings of the Teratology Society (Birth Defects Res. Part A: Clinical and Molecular Teratology 79(5): 418 (no. P25) and 419 (no. P26), 2007; Birth Defects Res. Part A: 82(5): 373 (no. P48) and 377 (no. P56) and, Birth Defects Res. Part A: 85: 453 (no. P74), 2009). This work was supported by grants from the Methanol Foundation and the Canadian Institutes of Health Research (CIHR). b. Full report of this research has been published: Sweeting, J. N., Siu, M., McCallum, G. P., Miller, L., Wells, P. G. (2010) Species differences in methanol and formic acid pharmacokinetics in mice, rabbits and primates. Toxicology and Applied Pharmacology. 247(1): 28-35. c. Individual contributions: J. Nicole Sweeting- rabbit dosing and sample collection; Michelle Siu- mouse dosing and sample collection; Gordon P. McCallum- monkey dosing and sample collection; Lutfiya Miller- assisted with dosing and collection of a portion of mouse samples; Peter G. Wells- supervisor. d. These authors contributed equally.

476

ABSTRACT

Methanol (MeOH) is metabolized primarily by alcohol dehydrogenase in humans, but by

catalase in rodents, with species variations in the pharmacokinetics of its formic acid (FA)

metabolite. The teratogenic potential of MeOH in humans is unknown, and its teratogenicity in

rodents may not accurately reflect human developmental risk due to differential species

metabolism, as for some other teratogens. To determine if human MeOH metabolism might be

better reflected in rabbits than rodents, the plasma pharmacokinetics of MeOH and FA were

compared in male CD-1 mice, New Zealand white rabbits and cynomolgus monkeys over time

(24, 48 and 6 hr, respectively) following a single intraperitoneal injection of 0.5 or 2 g/kg MeOH

or its saline vehicle. Following the high dose, MeOH exhibited saturated elimination kinetics in

all 3 species, with similar peak concentrations and a 2.5-fold higher clearance in mice than

rabbits. FA accumulation within 6 hr in primates was 5-fold and 43-fold higher than in rabbits

and mice respectively, with accumulation being 10-fold higher in rabbits than mice. Over 48 hr,

FA accumulation was nearly 5-fold higher in rabbits than mice. Low-dose MeOH in mice and

rabbits resulted in similarly saturated MeOH elimination in both species, but with approximately

2-fold higher clearance rates in mice. FA accumulation was 3.8-fold higher in rabbits than mice.

Rabbits more closely than mice reflected primates for in vivo MeOH metabolism, and particularly FA accumulation, suggesting developmental studies in rabbits may be useful for assessing potential human teratological risk.

Key Words: Methanol, formic acid, pharmacokinetics, species differences, rabbits, mice and primates

477

INTRODUCTION

Human exposure to methanol (MeOH) can result from its use as an industrial solvent, through accidental ingestion and as a potential alternate fuel source (Harris et al., 2004). Most toxicological studies use rodents as models of human MeOH toxicity, but there are key species differences in MeOH metabolism. Primates, including humans, metabolize MeOH to formaldehyde using the enzyme alcohol dehydrogenase (ADH), whereas rodents use catalase

(Cederbaum and Qureshi, 1982) (fig. 1). Humans and rodents metabolize formaldehyde to formic acid (FA) by formaldehyde dehydrogenase (ADH III). FA is subsequently converted to the non-toxic metabolites carbon dioxide and water by a folate-dependent dehydrogenase.

Humans have limited folate, resulting in FA accumulation following higher MeOH exposures(Perkins et al., 1995). Conversely, folate is not limited in rodents, which metabolize

FA via both catalase- and folate-dependent pathways, thereby preventing FA accumulation

(Clary, 2003; Harris et al., 2004). Aside from its role in rodent MeOH metabolism, catalase in all species provides cytoprotection against reactive oxygen species (ROS) by detoxifying hydrogen peroxide (Halliwell and Gutteridge, 2007; Wells et al., 2009b), which complicates the interpretation of rodent data for MeOH toxicity.

Human MeOH overdose causes acute ocular toxicity, CNS depression and death, apparently due to FA accumulation and subsequent metabolic acidosis. Rodents are resistant to this acute toxicity, presumably due to the absence of FA accumulation. Conversely, rodents can be susceptible to delayed adverse consequences of MeOH exposure, such as fetal neural tube defects and cleft palates following in utero exposure (Bolon et al., 1994; Rogers and Mole,

1997), and possibly cancer in adult rats (Soffritti et al., 2007), although the latter remains controversial (Cruzan, 2009; Schoeb et al., 2009).

478

Mechanisms of acute MeOH toxicity likely differ from those underlying its delayed toxicities. Although the mechanisms underlying delayed MeOH toxicities are unclear, a role for

ROS has been implicated in a rat embryo culture study, where depletion of the antioxidative peptide glutathione (GSH) increased the embryopathic effect of MeOH and its metabolites

(Harris et al., 2004). The proximal toxicant of MeOH developmental toxicity is unknown, although in rat embryo culture exogenous administration of formaldehyde is substantially more embryopathic than MeOH and FA (Hansen et al., 2005; Harris et al., 2004). Direct in vitro exposure to formaldehyde, which is highly reactive and transient, may not accurately reflect in utero exposure via its production from MeOH metabolism, however the role of formaldehyde in

MeOH developmental toxicity cannot be overlooked. ROS also have been implicated in adult

MeOH toxicity, including in vivo rat studies reporting that MeOH increased lipid peroxidation in lymphoid organs and the brain, as well as decreasing GSH and activity of the antioxidative enzyme superoxide dismutase (SOD) (Farbiszewski et al., 2000; Parthasarathy et al., 2006b).

The species differences between rodents and humans in MeOH metabolism and susceptibility to at least acute toxicity suggest that rodents may not constitute the most predictive model of human risk for delayed adverse effects of MeOH. Conversely, the rabbit is an attractive non-primate alternative for several reasons. Rabbits, but not mice, are susceptible to enhanced embryonic oxidative DNA damage and teratogenic effects of thalidomide, a known

ROS-initiating drug, which may prove relevant to the delayed effects of MeOH (Fratta et al.,

1965; Parman et al., 1999). Additionally, rabbits but not rats are insensitive to ethylene glycol teratogenicity, where the toxicokinetic profile in rabbits more closely reflects that in humans

(Carney et al., 2008). Finally, there is in vitro evidence in liver homogenates to suggest that

ADH activity in alcohol metabolism is more similar in rabbits than mice to that in humans

(Otani, 1978).

479

To determine if primate metabolism of MeOH, and particularly FA accumulation, was reflected more accurately in rabbits than rodents, we examined the in vivo plasma pharmacokinetics of MeOH and FA in cynomolgus monkeys, rabbits and mice following single doses of MeOH either below or above the reported saturation level for catalase (600 mg/kg)

(NEDO, 1986; NEDO, 1987). Males were used to avoid potential confounding effects of hormonal differences on MeOH metabolism between animals and species. Our results provide the first in vivo evidence that rabbits more closely than mice reflect primate MeOH metabolism, particularly in regard to FA accumulation. Developmental studies are needed to determine if rabbits similarly predict more accurately than mice the human risk for delayed adverse MeOH effects.

480

METHODS

Chemicals. HPLC grade MeOH was purchased from EMD Sereno Canada, Inc. (Mississauga,

ON). Saline (0.9 %, sterile) was purchased from Baxter Corporation (Mississauga, ON).

Isoflurane was purchased from Abbott Laboratories Ltd. (Saint-Laurant, QC). Compressed oxygen (99 %) was purchased from BOC Gases (Mississauga, ON). Alcohol oxidase (AO;

A6941) and formate dehydrogenase (FDH; F8649) from Candida boidinii, diaphorase from

Clostridium kluyveri (D5540), sodium formate, and β-nicotinamide adenine dinucleotide (β-

NAD+; N1511) were purchased from Sigma-Aldrich (St. Louis, MO). Formaldehyde dehydrogenase from Pseudomonas putida was obtained from MP Biomedicals (Solon, OH). p- iodonitrotetrazolium violet (INT) was purchased from Alfa Aesar (Ward Hill, MA).

Animals and diet. All animal protocols used were approved by the institutional animal care committee in conformance with the guidelines established by the Canadian Council on Animal

Care.

Mice. Male CD-1 mice were purchased from Charles River Laboratories (Saint-Constant, QC) and were 2 – 5 months old and approximately 33 – 52 g at the time of study. Mice were housed in vented cages from Allentown, Inc. (Allentown, NJ) with ground corn cob bedding (Bed-O’

Cobs Laboratory Animal Bedding, The Andersons Industrial Products Group, Maumee, OH).

Animal rooms were climate- and light-controlled (20°C, 50 % humidity, 14 hr light-10 hr dark cycle). Mice were fed rodent chow (Harlan Labs: 2018, Harlan Teklad, Montreal, QC) and

481 provided with water ad libitum. All mice were transported from the animal facility to the main laboratory for dosing and sacrificing, but were allowed 1 hr for acclimatization prior to the commencement of each study.

Rabbits. Male New Zealand white (NZW) rabbits were purchased from Charles River

Laboratories. At the time of experiments, rabbits were between the ages of 5-12 months, with a weight range of 3.25 – 4.75 kg. Rabbits were housed in plastic cages (Allentown, Inc.) in rooms maintained at 20°C and 60 % humidity, with an automated 12 hr light-dark cycle. Rabbits were fed a diet of standard high-fibre rabbit chow (Lab Diet: 5236 Hi-Fibre, PMI Nutrition

International LLC, Brentwood, MO), and provided with water ad libitum. Three days prior to the commencement of each study, rabbits did not receive any vegetable supplementation to their diet to avoid exposure to exogenous sources of antioxidants.

Primates. Studies were conducted with male cynomolgus monkeys (Macaca fascicularis) at

Charles River Laboratories (Sparks, NV). At the time of experiments, monkeys were between the ages of 3.4-5.7 years, with a weight range of 2.8-4.8 kg. Monkeys were acclimatized to individual stainless-steel cages two weeks prior to commencement of the study in rooms maintained between 18-29°C, with an automated 12 hr light-dark cycle. Monkeys were fed a certified primate chow diet (# 5048) from Purina Mills (St. Louis, MO) supplemented with fruit or vegetables 2-3 times weekly, and provided with water ad libitum.

482

Dosing and blood collection. Mice were administered either a single low dose (0.5 g/kg bw) or high dose (2 g/kg bw) of MeOH (20 % [w/v] in sterile saline) or a saline vehicle control. MeOH was administered via intraperitoneal (ip) injection using a 26 gauge (G) 3/8 needle. Prior to blood collection, mice were anaesthetized by placing a 15 mL conical tube containing isoflurane- soaked gauze over the snout of the animal until it was unresponsive. Blood samples

(approximately 1 mL) were collected in heparinised vacutainers (lithium heparin 68 USP units per tube, Becton, Dickinson and Company, Oakville, ON) either directly from the atrium or by puncturing the portal vein or inferior vena cava, using a 20 G 1½ needle. For mice administered a high dose of MeOH, blood samples were collected at 1, 4, 8, 12, 16, 20 and 24 hr post injection, while samples for the low dose mice were collected at 1, 2, 4, 6, 8, 10 and 12 hr post injection. Plasma was isolated from the samples by centrifugation at 1,000 x g for 15 min at

4°C. Plasma samples were frozen at -20°C until time of analysis. At each time point, plasma samples were collected from 3-6 mice.

Due to the size of the animals, rabbits were anaesthetized with 3 % isoflurane in 2 L of oxygen for approximately 5 min. Rabbits were then administered a single high or low dose (0.5 or 2 g/kg bw) of MeOH (20% [w/v] in sterile saline) or its saline vehicle (control) by ip injection using a 23 G needle. Following injection, rabbits were exposed to 100 % oxygen for approximately 2 min to quicken recovery from the anaesthesia. Blood samples (1-3 mL) were collected in heparinised vacutainers (lithium heparin 68 USP units per tube, Becton, Dickinson and Company) through the ear vein prior to dosing and at 15 and 30 min, and 1, 4, 6, 12, 18, 24,

30, 36 and 48 hr post injection. Blood samples were centrifuged at 1,000 x g for 15 min at 4°C, and the plasma supernatant was collected and frozen at –80°C. A total of 3 rabbits were used in each treatment group.

483

Primates were lightly sedated to effect with ketamine (~5-10 mg/kg im) and then administered MeOH (2 g/kg bw; 20% [w/v] in sterile saline) or a saline vehicle control by ip injection using a 22 G needle. Blood samples (6 mL) were drawn from the femoral vein and collected in heparinised tubes at 15 and 30 min, and 1, 4 and 6 hr post injection. Whole blood was processed to plasma by centrifugation at 1,500 x g for 15 min at 4°C, flash-frozen in liquid

N2 and stored at –70°C. Three primates were used in both the 2 g/kg MeOH and saline control groups.

Analytical Methods. Plasma samples from saline controls and MeOH-treated rabbits and mice were analyzed for MeOH and FA concentrations by headspace gas chromatography (GC) based on previously published methods (Fraser and MacNeil, 1989; Porter and Moyer, 1994). The

Agilent GC system consisted of a 6890N GC, a G1888 headspace sampler, and a G1540N flame ionization detector (FID) coupled with a Restek Rtx-200 capillary column (30 m, 0.53 mm ID, 3

µm). MeOH concentrations were quantified from linear standard curves using the peak area ratios of MeOH to the internal standard n-propanol using Chemstation software (version

B.02.01; Agilent). Similarly, FA concentrations were determined based on the peak area ratios of methyl formate to the internal standard methyl proprionate.

Due to local biosafety restrictions, MeOH and FA concentrations were analyzed in primate serum samples with validated enzymatic methods. MeOH concentrations were determined by the method of Vinet (Vinet, 1987) adapted to accommodate analysis in a microplate format. Briefly, reaction mixtures contained 200 µL of enzymatic reagent (2.5 mM

β-NAD+, and 500 U/L FDH in 100 mM phosphate buffer, pH 7.6), and 3.5 µL of sample or calibration standards. Reactions were incubated for 5 min at 30 oC and then started by addition

484 of 16.5 µL of AO solution (5U/mL in 100 mM phosphate buffer, pH 7.6). Calibration standards of 12.5, 6.25, 3.125 and 0 mmol/L were prepared by diluting a 1.25 mol/L aqueous solution of

MeOH with unexposed primate serum. Serum samples from MeOH-exposed primates were diluted in unexposed primate serum in order to fit within the calibration curve. Absorbance was monitored continuously at 340 nm in a Spectramax M2 Plate Reader (Molecular Devices

Corporation, Sunnyvale, CA) and the initial linear reaction velocities were used to calculate a standard curve to determine unknown concentrations, which were determined from the average of duplicate determinations. Serum FA concentrations were determined by a modification of the colorimetric endpoint assay (Grady and Osterloh, 1986) which was developed to improve the dynamic range of the fluorometric method of Makar and Tephly (Makar and Tephly, 1982) using acetonitrile serum extraction and INT, a chromophore with higher molar absorptivity. Briefly, reaction mixtures contained 600 µL enzymatic reagent (2.5 mM β-NAD+, 800 U/L diaphorase,

2.0 mmol/L INT in 100 mM phosphate buffer, pH 6.0), and 20 µL of supernatant from an acetonitrile (1:1) serum precipitation. Reactions were initiated by the addition of 12 µL of FDH

(5400 U/L in 100mM phosphate buffer, pH 6.0 containing 2.5 mM β-NAD+, and 800 U/L diaphorase), left for 15 min at room temperature and then the absorbance was read at 500 nm in a

Spectramax M2 Plate Reader (Molecular Devices Corporation). Serum FA concentrations were calculated from the average of duplicate determinations using a six point standard curve prepared from 2-fold serial dilutions of unexposed primate serum spiked with 320 mg/ml of sodium formate.

Data Analysis. MeOH and FA data for mice, rabbits and primates were graphed using

GraphPad Prism®, Version 5 (GraphPad Software, Inc., San Diego, CA). The area under the

485 plasma concentration-time curve (AUC) in rabbits and primates was calculated using this software from time zero to the time of the last measurable concentration of MeOH or FA in each graph to avoid overestimates of the AUC values. For mouse data, the AUC was calculated using the mean plasma concentration of MeOH or FA at each time point; however, as different animals were sacrificed at each time point, the standard deviation of the AUC was determined by the method of Yuan (Yuan, 1993). For both rabbit and primate data, an AUC was determined for individual animals, as each animal was sampled repetitively until 48 or 6 hr post-injection, respectively. The clearance (Cl) of MeOH and FA was determined using the equation Cl = dose/AUC. Cl values for mice represent the mean for all animals studied, whereas the Cl data for rabbits and primates represents the elimination of MeOH and FA for each animal.

Statistical Analysis. Statistical analysis was performed using GraphPad Prism®, Version 5

(GraphPad Software, Inc.). Differences in MeOH Cl and FA AUC values between mice and rabbits over the complete elimination period were analyzed for significance using the Student`s t-test. For AUC values in which all 3 species were compared, a one-way analysis of variance

(ANOVA) with Tukey’s post-hoc test was used. The level of significance was determined to be at p < 0.05.

486

RESULTS

COMPLETE TIME STUDIES:

In the time course studies discussed below, plasma concentrations of methanol and FA following saline administration were not detectable, therefore these data were not included in the figures.

1. MeOH and FA plasma pharmacokinetics following low-dose (0.5 g/kg) MeOH treatment

Mice. Male CD-1 mice exhibited zero-order (saturation) elimination kinetics of MeOH, with a peak plasma MeOH concentration of approximately 18 mmol/L being achieved between 1-4 hr post-injection (fig. 2, supplementary fig. S1). MeOH concentrations were below the level of detection by 8-10 hr, leading to a mean clearance (Cl) rate of 3.67 ± 0.46 ml/(min x kg bw) (fig.

3, table 1). Additionally, there was an absence of sustained plasma FA accumulation at this dose. A peak in plasma FA occurred at 6 hr (0.14 mmol/L) but was undetectable 2 hr later, corresponding to an area under the plasma concentration-time curve (AUC (0-24 h)) of 0.53 ±

0.08 (mmol/L) x hr.

Rabbits. Male NZW rabbits similarly eliminated MeOH by zero-order kinetics (figs. 2, S1).

The mean peak plasma MeOH concentration achieved was 24.67 ± 6.74 mmol/L, occurring within 15 min in rabbits 1 and 2, and within 1 hr in rabbit 3 (fig. S2). MeOH was completely eliminated from the plasma by 18 hr, corresponding to a mean Cl rate of 1.50 ± 0.26 ml/(min x

487 kg bw) (fig. 3, table 1). In each of these rabbits, FA began to accumulate in the plasma within

15 min, with a mean peak concentration of 0.17 mmol/L being achieved between 4-6 hr.

Metabolite accumulation was maintained for 12-30 hr, resulting in a mean AUC (0-48 h) of 3.02

± 1.28 (mmol/L) x hr.

Species comparison. The MeOH Cl rate in rabbits was less than one-half that in mice at the low dose, although the peak plasma concentrations were not significantly different (fig. 3). There was a somewhat greater species difference in FA accumulation, with about a 5.7-fold higher FA AUC in rabbits compared to mice. See Table 1 for values and comparisons.

2. MeOH and FA plasma pharmacokinetics following high-dose (2 g/kg) MeOH treatment

Mice. In male CD-1 mice, the 2 g/kg dose of MeOH resulted in saturable elimination kinetics

(figs. 4, S3). A mean peak plasma MeOH concentration of 79 mmol/L was achieved within 1 hr, and was 4.4 times greater than that achieved following exposure to the low dose of MeOH (fig.

2). This increase is consistent with the 4-fold difference between the low and high doses. The high dose of MeOH resulted in a plasma Cl rate of 1.53 ± 0.11 ml/(min x kg bw) (fig. 3, table 1).

Even at the higher dose, mice exhibited little FA accumulation, with a mean peak plasma level of

0.3 mmol/L at 8 hr, returning to near basal levels within 16-20 hr (fig. 4). Since there was a slight accumulation of FA at 20 hr following saline exposure, the final FA AUC (0-24 h) for the

2 g/kg treatment group represents the difference between the MeOH-treated mice and saline

488 controls, and was 3.06 ± 0.54 (mmol/L) x hr (fig. 3, table 1). This FA AUC (0-24 h) resulting from high-dose MeOH was approximately 5.8-fold higher than that (0.53 (mmol/L) x hr) from the low MeOH dose, which in light of the 4-fold difference in doses may suggest saturable FA elimination at the higher dose.

Rabbits. Male NZW rabbits exhibited saturable elimination of MeOH following exposure to the high dose of MeOH (figs. 4, S3). A mean peak plasma MeOH concentration of 114 mmol/L was achieved within 15 min. This peak concentration was 5.0 times greater than that achieved following the 0.5 g/kg dose of MeOH, consistent with the 4-fold difference between the high and low doses. The resulting mean plasma MeOH Cl rate following the high dose of MeOH was

0.57 ± 0.12 ml/(min x kg bw) (fig. 3, table 1). FA accumulation commenced within 15 min and was maintained for almost 48 hr. FA accumulation reached a level 2.8 times greater than that achieved following exposure to the saline control, corresponding to a mean AUC (0-48 h) of

10.63 ± 1.35 (mmol/L) x hr (fig. 3, table 1). This high-dose AUC reflects a 3.5-fold greater value than that achieved with the low dose, consistent with the 4-fold difference between the two doses.

Species comparison. With high-dose MeOH, the MeOH Cl rate in rabbits was over 60% slower than that achieved in mice (fig. 3, table 1). This trend is consistent with the aforementioned species differences in Cl rates observed following the low dose of MeOH exposure. The species difference in FA accumulation observed with low-dose MeOH was substantially enhanced with the high dose, after which the FA AUC in rabbits was nearly 4-fold higher than that in mice.

See Table 1 and fig. 3 for values and comparisons.

489

3. Comparison of low- and high-dose studies

Mice. MeOH Cl in mice treated with high-dose MeOH was about 40% of that following the low dose, while the FA AUC (0-24 h) was 5.8-fold higher (fig. 3, table 1).

Rabbits. MeOH Cl in rabbits treated with high-dose MeOH was about 40% of that following the low dose, while the FA AUC (0-48 h) was about 3.5-fold higher (fig. 3, table 1).

SIX-HOUR STUDIES:

1. Primates

Cynomolgus male monkeys treated with 2 g/kg of MeOH or its saline vehicle control were sampled for 6 hr, at which time they were sacrificed for future molecular studies. All pharmacokinetic values for MeOH and FA accordingly reflected this 6-hr window rather than a complete elimination profile. Within 6 hr, MeOH elimination was clearly saturable (figs. 5, S5,

S6). The mean peak plasma MeOH concentration of 94.49 ± 14.22 mmol/L was achieved within

30 min. FA began to accumulate within 15 min, reaching a relatively substantial concentration of 2.15 ± 0.77 mmol/L at 6 hr, at which time it was still increasing.

490

2. Species comparison including primates over 6 hr

Plasma concentrations of MeOH over 6 hr were similar in primates, rabbits and mice (fig.

5). The 6-hr plasma MeOH AUC for mice was approximately 1.8 and 1.5-fold lower than the respective AUC values for rabbits and primates (figs. 6, S7; table 2). The mean plasma concentration of FA achieved in primates over the 6-hr time course was approximately 11-fold and 7-fold greater than that achieved in mice and rabbits over the same period, respectively (fig.

5). This difference in primate FA formation is an underestimate, since the concentrations in mice and rabbits, unlike primates, were not still increasing at 6 hr. The FA AUC levels achieved within 6 hr in primates was about 52-fold and 5-fold higher than the levels in mice and rabbits respectively (fig. 6). The 6-hr FA AUC in rabbits was 10-fold higher than that in mice.

The 6-hr species difference between mice and rabbits treated with high-dose MeOH (figs.

5, 6, S7, S8; table 2) exceeded that observed in the complete time study for these two species treated with high-dose MeOH, where the complete FA AUC (0-48 h) in rabbits was about 4-fold higher than that in mice (AUC (0-24 h)), and the rabbit MeOH Cl rate was about 37% of that in mice (figs. 3, 4; table 2).

491

DISCUSSION

Humans primarily use ADH and CYP2E1 to metabolize MeOH to formaldehyde, and subsequently use a capacity-limited, folate-dependent pathway to metabolize FA, which in

MeOH overdose accumulates and is believed to cause the acute MeOH effects of blindness, metabolic acidosis and death (Coon and Koop, 1987; Wallage and Watterson, 2008). In contrast, rodents use catalase to metabolize MeOH, and subsequently use catalase together with a folate- dependent pathway that is not capacity-limited to metabolize FA, which does not accumulate following MeOH overdose (Clary, 2003; Harris et al., 2004). Accordingly, MeOH metabolism in rodents does not reflect that in humans and, as would be expected given the absence of rodent

FA accumulation, rodents may not be the most accurate reflection of human risk for the acute toxic effects of MeOH.

Less well understood are the adverse effects of MeOH on the developing embryo and fetus, the mechanisms of which may be quite different from the FA-initiated acute toxicities seen in humans. Although developmental toxicities have not been reported in humans, the potential is assumed largely on the basis of studies in pregnant rodents (Bolon et al., 1994; Degitz et al.,

2004; Harris et al., 2003). One potential mechanism for developmental abnormalities involves oxidative stress. Even low levels of embryonic and fetal oxidative stress and the formation of

ROS have been implicated in the mechanism of developmental toxicities observed in both untreated and xenobiotic-exposed pregnant animal models, and limited evidence suggests a potential role for ROS in the developmental toxicity of MeOH (Farbiszewski et al., 2000; Harris et al., 2004; Parthasarathy et al., 2006b). Importantly, ROS can be detoxified by the same catalase enzyme that rodents, but not humans, use to metabolize MeOH. However, it is must be noted that there may be developmental differences in enzyme activity and MeOH metabolism

492 between the embryo and the adult. A recent study suggests that catalase may actually be playing a more influential role than ADH in perinatal human MeOH metabolism (Tran et al., 2007). In light of the profound differences between rodents and humans in metabolic pathways used to metabolize MeOH, and the dual roles of catalase in ROS detoxification and rodent MeOH metabolism, rodent models, while useful in evaluating molecular mechanisms, may not accurately reflect the human risk for MeOH developmental toxicity. We therefore evaluated rabbits as an animal model potentially more similar to humans in their metabolism of MeOH, and if so, possibly more similar to humans in their risk for MeOH developmental toxicity.

We selected the rabbit as an animal model that may potentially more accurately reflect human developmental risk from MeOH for several reasons: (1) one in vitro study in hepatic homogenates suggested that the rabbit might be more similar than mice to humans in their use of

ADH rather than catalase in MeOH metabolism (Otani, 1978); (2) rabbits and humans, but not rodents, are susceptible to the teratogenic effects of thalidomide, which may be caused in part by

ROS formation (Hansen et al., 2002; Parman et al., 1999); and, (3) rats, but not rabbits are susceptible to glycolic acid-initiated birth defects following exposure to ethylene glycol (Carney et al., 2008; Tyl et al., 1993). Male rabbits and mice were chosen for the metabolism studies herein to avoid potential confounding effects of hormonal variability in females within and between species. Two species comparisons were made, the first over the initial 6 hours following high-dose MeOH, which in addition to mice and rabbits included cynomolgus monkeys that were sacrificed at this time for molecular studies. The second species comparison for only mice and rabbits was over the full elimination periods for mice (24 hr) and rabbits (48 hr) following both low- and high-dose MeOH.

493

The doses used in our studies (2 and 0.5 g/kg) were chosen based upon published reports to deliver tissue concentrations respectively above and below the saturation level for catalase in mice. However, MeOH elimination following the lower dose still exhibited saturation kinetics in our CD-1 mice, albeit less so than following high-dose MeOH. Since MeOH metabolism is catalyzed by catalase in mice, the low-dose results indicate that a dose lower than that reported elsewhere would be necessary to avoid saturable metabolism in this strain. The saturation kinetic profile for MeOH in rabbits, which appear to use ADH to metabolize MeOH (Otani,

1978), our studies herein provide the first in vivo evidence that rabbits more closely than mice reflect the human metabolism of MeOH, particularly with respect to FA accumulation. Perhaps most importantly, in the comparison of all 3 species over the first 6 hr, the accumulation of FA in primates was 52-fold higher than that in mice. Although rabbits exhibited less FA accumulation than primates, by 6 hr rabbits nevertheless exhibited 10-fold higher FA accumulation than mice, more closely reflecting the human profile (Hantson et al., 2005). MeOH AUC over this initial 6- hr period did not differ substantially among species.

Over the complete elimination period, rabbits presented a more predictable dose- dependent pattern of FA accumulation than mice, with a 3.5-fold increase resulting from the 4- fold dose increase going from the low to high dose of MeOH. Mice in contrast accumulated FA to a greater extent, with a 5.8-fold increase following the same increase in MeOH dose. Despite these dose-dependent differences within species, the substantially greater FA accumulation in rabbits compared with mice was about 4-fold higher with high-dose MeOH, and 5.7-fold higher at the low dose. Although both species achieved similar peak concentration levels of FA, the rabbit showed consistently sustained levels over a greater period of time over the course of the study. Conversely, the clearance of MeOH in rabbits was about 40% of that in mice at both the high and low doses of MeOH. Taken together, the lower clearance of MeOH and perhaps most

494 importantly the substantially greater sustained accumulation of FA in rabbits compared to mice indicate that rabbits are substantially more similar than mice to nonhuman primates (our 6-hr study) and humans (Clary, 2003; Perkins et al., 1995) in their metabolism of MeOH. The accumulation of FA may be particularly important in light of the presumed role of this metabolite in the acute toxicities of MeOH, although it is not clear what role FA plays in delayed effects like developmental toxicities (Andrews et al., 1995; Hansen et al., 2005) or cancer, the latter association being controversial due to potentially confounding variables and an absence of corroborating studies (Cruzan, 2009). Additionally, while in vitro whole embryo culture studies in mice and rats suggest that formaldehyde may be a potent proximate teratogen mediating

MeOH-induced birth defects (Hansen et al., 2005; Harris et al., 2004), the presence of elevated formaldehyde concentrations in tissues and body fluids has not been observed in any species following MeOH exposure(Tephly, 1991), likely due to its transient nature. Formaldehyde has plasma half-lives of approximately 1 and 1.5 minutes in rats and primates respectively, due to rapid metabolism to FA and/or the formation of DNA and/or protein adducts (McMartin et al.,

1979; Shaham et al., 1996). During pregnancy, little if any formaldehyde produced in the maternal liver would persist long enough to cross the placenta and enter the embryo. It is not known if toxic amounts of formaldehyde can be produced proximately via the limited activity of embryonic catalase, which at least in mice is less than 5% of maternal activity (Wells et al.,

2009).

The values for MeOH clearance and FA accumulation in humans are difficult to determine accurately due to questionable information for such factors as dose or exposure level and time of exposure, as well as the confounding use of hemodialysis therapy as early as possible. In cases of human MeOH poisoning, a wide range of plasma concentrations have been reported, encompassing the values observed herein in all 3 species, with MeOH levels varying

495 from 8 to 154 mM and plasma FA concentrations varying from 0.3 to over 20 mM (Hantson et al., 2005).

In conclusion, rabbits reflected more closely than mice the pharmacokinetic profile of

MeOH and FA in primates, the latter of which were similar to the limited published values in humans. This included both the lower clearance of MeOH and greater accumulation of FA, the latter of which is presumed to cause at least the acute toxic effects of MeOH. Whether or not this FA metabolite contributes to delayed MeOH effects like developmental abnormalities remains to be determined, although exogenously added FA is embryotoxic in rodent embryo culture (Hansen et al., 2005; Harris et al., 2004). It is worth noting that the interpretive problem is with the use of mice in the prediction of human risk, since mice, despite the species differences in enzymes and enzymatic activities, can nevertheless be useful in determining molecular mechanisms of teratogens, including alcohols like MeOH and ethanol. It is not known whether the absence of published evidence for delayed outcomes in humans reflects species resistance, the absence of sufficiently high exposures or failure to detect outcomes manifested perhaps weeks, months or years following MeOH exposure. Further studies comparing the developmental effects of MeOH in pregnant rabbits and mice are warranted to determine if rabbits may serve as a more accurate model than mice in predicting human risk with respect to toxicity in the embryo and fetus.

496

TABLES

TABLE 1. COMPARISON OF RABBITS AND MICE FOR COMPLETE ELIMINATION. Studies were carried out over the complete elimination period for rabbits (48 hr) and mice (24 hr) (See Figures 2-4, S1-S4). Data represent the mean + SD.

497

TABLE 2. COMPARISON OF PRIMATES, RABBITS AND MICE OVER 6 HOURS. Studies were carried out for 6 hr, at which time primates were sacrificed for molecular analyses (See Figures 5, 6, S5-S8). Data represent the mean + SD.

498

FIGURES

Fig. 1. Species differences in the enzymes catalyzing the metabolism of methanol (MeOH) to formaldehyde and formic acid (FA) in mice and primates, including humans. In addition to its role in MeOH metabolism in mice, catalase is used by all species in the detoxification of reactive oxygen species (ROS).

499

Fig. 2. Mean MeOH and FA pharmacokinetics in male mice and rabbits following a single dose of 0.5 g/kg MeOH, plotted on a semi-logarithmic scale. MeOH was administered through an ip injection as a 20 % solution in sterile saline, and plasma samples were analyzed by GC for MeOH or FA concentrations. Saline groups not shown (all not detectable). Each time point represents the mean of 3-6 mice or 3 rabbits for treated groups. Data points with open symbols represent samples with a mean MeOH or FA concentration below the level of detection, and were arbitrarily assigned a value of 0.1 mmol/L.

500

Fig. 3. MeOH clearance (Cl) and FA areas under the plasma concentration-time curve (AUCs) in male mice and rabbits following a single dose of 0.5 or 2 g/kg MeOH. Cl and AUC values were calculated from the MeOH and FA pharmacokinetics curves of animals dosed with 0.5 or 2 g/kg MeOH (20 % solution in sterile saline) through an ip injection. Saline groups not shown (all not detectable). Each time point represents the mean of 3-6 mice or 3 rabbits for treated groups. * indicates measures for rabbit MeOH Cl and FA AUC that were different from the respective mouse values (p < 0.05).

501

Fig. 4. Mean MeOH and FA pharmacokinetics in male mice and rabbits following a single dose of 2 g/kg MeOH, plotted on a semi-logarithmic scale. MeOH was administered through an ip injection as a 20 % solution in sterile saline, and plasma samples were analyzed by GC for MeOH or FA concentrations. Saline groups not shown (all not detectable). Each time point represents the mean of 3-6 mice or 3 rabbits for treated groups. Data points with open symbols represent samples with a mean MeOH or FA concentration below the level of detection, and were arbitrarily assigned a value of 0.1 mmol/L.

502

Fig. 5. Mean MeOH and FA pharmacokinetics in male mice, rabbits and primates over a 6 hr time period, following a single dose of 2 g/kg MeOH, plotted on a semi-logarithmic scale. MeOH wasadministered through an ip injection as a 20 % solution in sterile saline, and plasma samples were analyzed by GC (mice and rabbits) or by spectrometry (primates) for MeOH or FA concentrations. A total of 3-6 mice, 3 rabbits or 3 primates were sampled at each time point for treated groups. Saline groups not shown (all not detectable). Data points with open symbols represent samples with a MeOH or FA concentration below the level of detection and were arbitrarily assigned a value of 0.1 mmol/L.

503

Fig. 6. Six-hour MeOH and FA AUCs in male mice, rabbits and primates following a single dose of 2 g/kg MeOH. AUC values were calculated from the MeOH and FA pharmacokinetics curves of animals dosed with 2 g/kg MeOH (20 % solution in sterile saline) through an ip injection. A total of 3-6 mice, 3 rabbits and 3 primates were sampled at each time point for treated groups. Saline groups not shown (all not detectable). * indicates that 6-hr rabbit and primate MeOH Cl and FA AUC were different from the respective mouse values. † indicates that the 6-hr FA AUC in primates was different that in rabbits (p < 0.05).

504

SUPPLEMENTARY FIGURES

Fig. S1. Mean methanol (MeOH) and formic acid (FA) pharmacokinetics in male mice and rabbits following a single dose of 0.5 g/kg MeOH, plotted on a linear scale. MeOH was administered through an ip injection as a 20 % solution in sterile saline, and plasma samples were analyzed by GC for MeOH or FA concentrations. Each time point represents the mean of 3- 6 mice or 3 rabbits in the treated groups. Saline groups not shown (all not detectable). Data points with open symbols represent samples with a mean MeOH or FA concentration below the level of detection, and were arbitrarily assigned a value of 0.1 mmol/L.

505

Fig. S2. Pharmacokinetics of MeOH and FA in individual male rabbits following a dose of 0.5 or 2 g/kg MeOH, plotted on a semi-logarithmic scale. MeOH was administered through an ip injection as a 20 % solution in sterile saline, and plasma samples were analyzed by GC for MeOH or FA concentrations. Saline groups not shown (all not detectable). Data points with open symbols represent samples with a MeOH or FA concentration below the level of detection and were arbitrarily assigned a value of 0.1 mmol/L.

506

Fig. S3. Mean MeOH and FA pharmacokinetics in male mice and rabbits following a single dose of 2 g/kg MeOH, plotted on a linear scale. MeOH was administered through an ip injection as a 20 % solution in sterile saline, and plasma samples were analyzed by GC for MeOH or FA concentrations. Each time point represents the mean of 3-6 mice or 3 rabbits in the treated groups. Saline groups not shown (all not detectable). Data points with open symbols represent samples with a mean MeOH or FA concentration below the level of detection, and were arbitrarily assigned a value of 0.1 mmol/L.

507

Fig. S4. MeOH clearance (Cl) rates and FA areas under the plasma concentration-time curve (AUCs) of individual male rabbits following a dose of 0.5 or 2 g/kg MeOH. Cl and AUC values were calculated from the MeOH and FA pharmacokinetics curves of rabbits dosed with 0.5 or 2 g/kg MeOH (20 % solution in sterile saline) through an ip injection.

508

Fig. S5. Mean MeOH and FA pharmacokinetics in male mice, rabbits and primates over a 6 hr time period, following a single dose of 2 g/kg MeOH, plotted on a linear scale. MeOH was administered through an ip injection as a 20 % solution in sterile saline, and plasma samples were analyzed by GC (mice and rabbits) or by spectrometry (primates) for MeOH or FA concentrations. A total of 3-6 mice, 3 rabbits or 3 primates were sampled at each time point for treated groups. Saline groups not shown (all not detectable). Data points with open symbols represent samples with a MeOH or FA concentration below the level of detection and were arbitrarily assigned a value of 0.1 mmol/L.

509

Fig. S6. Pharmacokinetics of MeOH and FA in individual male primates following a single dose of 2 g/kg MeOH, plotted on semi-logarithmic scales. MeOH was administered through an ip injection as a 20 % solution in sterile saline, where plasma samples collected over 6 hr and analyzed by spectrometry for MeOH or FA concentrations.

510

Fig. S7. Six hr MeOH and FA AUCs in individual male primates following a single dose of 2 g/kg MeOH. AUC values were calculated from the MeOH and FA pharmacokinetics curves of animals dosed with 2 g/kg MeOH (20 % solution in sterile saline) through an ip injection.

511

Fig. S8. Mean MeOH and FA pharmacokinetics in male mice, rabbits and primates over a 48 hr time period, following a single dose of 2 g/kg MeOH, plotted on semi-logarithmic scales. MeOH was administered through an ip injection as a 20 % solution in sterile saline, and plasma samples were analyzed by GC (mice and rabbits) or by spectrometry (primates) for MeOH or FA concentrations. A total of 3-6 mice, 3 rabbits or 3 primates were sampled at each time point for treated groups. Saline groups not shown (all not detectable). Data points with open symbols represent samples with a MeOH or FA concentration below the level of detection and were arbitrarily assigned a value of 0.1 mmol/L.

512

11.2 Oxidative stress and species differences in the metabolism, developmental toxicity and carcinogenic potential of methanol and ethanol1

Running title: Oxidative stress in methanol teratogenicity

Peter G. Wellsa,b, Gordon P. McCallum2,a, Lutfiya Miller2,b, Michelle Siu2,a and J. Nicole Sweeting2,a

a Division of Biomolecular Sciences, Faculty of Pharmacy b Department of Pharmacology and Toxicology, Faculty of Medicine University of Toronto Toronto, Ontario, Canada

1. These studies were supported by grants from the Methanol Foundation (U.S.A.) and the Canadian Institutes of Health Research.

2. These authors contributed equally. b. Full report of this research has been published: Wells, P.G., McCallum, G.P., Miller, L., Siu, M.T., Sweeting, J.N. Oxidative stress and species differences in the metabolism, developmental toxicity and carcinogenic potential of methanol and ethanol. In:The Toxicology of Methanol, JJ Clary (ed.), pp. 169-253, Wiley, Hoboken, U.S.A. 2013.

513

CHAPTER OUTLINE

ABSTRACT

1. INTRODUCTION 1.1. Preamble 1.1.1. The problem 1.1.2. Fundamental question 1.1.3. Research objectives 1.1.4. Approach 1.2. Methanol (MeOH) developmental toxicity 1.3. Carcinogenic potential 1.4. Oxidative stress and other potential mechanisms of toxicity 1.5. Factors affecting the human relevance of animal models 1.5.1. Species differences in metabolism 1.5.2. Dose of methanol and route of exposure

2. SPECIES DIFFERENCES IN METHANOL METABOLISM 2.1. Enzymes and pathways 2.1.1. Alcohol dehydrogenase 2.1.2. Catalase 2.1.2.1. Peroxidative role 2.1.2.2. Antioxidative role 2.1.3. Cytochrome P450 (CYP) 2E1 2.1.4. Formaldehyde dehydrogenase 2.1.5. Folate 2.2. Pharmacokinetics of methanol and formic acid

3. SPECIES AND STRAIN DIFFERENCES IN METHANOL TOXICITY 3.1. Acute metabolic acidosis, ocular toxicity and death 3.2. Teratogenesis 3.3. Neurodevelopmental effects 3.4. Carcinogenic potential

4. OXIDATIVE STRESS 4.1. Oxidative stress mechanisms 4.1.1. Embryonic drug exposure and reactive oxygen species (ROS) formation 4.1.2. Signal transduction 4.1.3. Macromolecular damage 4.2. Oxidative stress from methanol exposure 4.2.1. Evidence for MeOH-initiated ROS formation 4.2.2. Mechanism of MeOH-initiated ROS formation 4.3. Teratogenicity of methanol and comparisons to ethanol 4.3.1. Genetic modulation of catalase 4.3.2. Free radical spin trapping 4.4. Carcinogenic potential 4.4.1. Oxidatively damaged DNA 4.4.2. Hydroxynonenal-histidine protein adducts

5. CONCLUSIONS

514

ABSTRACT

Methanol (MeOH) is teratogenic in rodents, which unlike humans use catalase to convert both

MeOH to formaldehyde, and its formic acid (FA) metabolite to carbon dioxide and water. It is not known if MeOH is developmentally toxic in humans, and it is unclear if rodents can predict human risk in light of their different routes of MeOH metabolism. A more accurate model for predicting human risk might be found in non-human primates, or at least in a species that is more like humans than mice and rats in their metabolism of MeOH.

A better understanding of the molecular mechanism of MeOH teratogenesis would also help in understanding the potential human risk. One possible mechanism, involving enhanced oxidative stress and the formation of toxic reactive oxygen species (ROS), also would be relevant to the potential for MeOH to cause cancer, which has been suggested by one controversial study in rats.

We have found that the metabolism of MeOH in cynomolgus monkeys, which closely reflect humans, is more similar to that in rabbits than in mice. In complementary in vitro studies, preliminary data suggest that the peroxidative activity of catalase in rabbits is only 52% of that in mice (p < 0.05), and similar to the activity in humans, consistent with a primary role for catalase in mice in metabolizing MeOH and FA, while humans and possibly rabbits rely upon alcohol dehydrogenase (ADH1). These results suggest that the rabbit might be a more accurate model than the mouse for predicting the human risk for MeOH developmental toxicity.

In developmental studies, rabbits were shown for the first time to be resistant to MeOH teratogenesis, in contrast to C57BL/6 mice, which were susceptible, as had previously been shown by other laboratories. However, we also found that a different (C3H) mouse strain was

515 resistant, providing the first evidence that not all mouse strains are susceptible. These strain- and species-dependent differences in teratological susceptibility could not be explained by differences in the pharmacokinetics of MeOH or FA. The dose (4 g/kg i.p.) of MeOH administered to both species was high, and substantially above the lethal human dose, so it is unlikely that resistance was due to an insufficient dose of MeOH. To facilitate a mechanistic investigation, our studies focused upon the teratogenic effects of a single high total dose of

MeOH given during the embryonic period of organogenesis, so it remains to be determined if a similar pattern of outcomes would occur following extended exposure to MeOH during the later fetal period of development. Given that rabbits, which more closely reflect human MeOH metabolism, and at least one strain of mice, are resistant to MeOH teratogenesis, it is questionable whether the human risk for MeOH developmental toxicity can be accurately assessed in sensitive rodent models.

In mouse strains that are sensitive to the developmental toxicity of MeOH, we examined the role of catalase and ROS in the embryopathic mechanism. In addition to the metabolism of MeOH by catalase in rodents, this antioxidative enzyme also detoxifies ROS (fig. 1). We used the following approaches in vivo and/or in embryo culture: (1) genetically modified mutant mice deficient in catalase (acatalasemic, aCat), and transgenic mice expressing human catalase activity (high catalase activity, hCat); and, (2) pretreatment with the free radical spin trapping agent α-phenyl-N-t-butylnitrone (PBN). The in vivo studies were inconclusive due in part to the resistance of both aCat mice and their wild-type C3H strain, and no apparent effect of enhanced catalase on developmental outcome. However, in embryo culture, MeOH embryopathies were enhanced in aCat mice and reduced in hCat mice, and were blocked by pretreatment with PBN.

These results suggest that ROS play a role in the mechanism of MeOH developmental toxicity in

516 sensitive strains of mice, and that embryonic catalase plays an important protective role via its antioxidative activity, as distinct from its peroxidative, MeOH-metabolizing role in maternal liver or the embryo.

In addition to its potential role in teratogenesis, ROS-initiated oxidatively damaged DNA, and particularly the 8-oxo-2'-deoxyguanine (8-oxodG) lesion, is mutagenic and carcinogenic. In a novel comparative evaluation of adult male mice, rabbits and monkeys, we found that an acute single high dose (2 g/kg i.p.) of MeOH did not enhance 8-oxodG formation in any tissue including bone marrow in any species, nor did the same dose of MeOH injected daily in mice for

15 days. Levels of hydroxynonenal (HNE)-histidine protein adducts, reflecting free radical- mediated production of the potentially carcinogenic lipid peroxidation product HNE, were not enhanced by MeOH in primate bone marrow or spleen, or in rabbit bone marrow or mouse spleen, although modest increases were observed in rabbit spleen and mouse bone marrow.

These results suggest that it is unlikely that human environmental exposure to MeOH would cause cancer via a mechanism involving oxidatively damaged DNA.

The above results from our laboratory are discussed within the context of regulatory issues and published research on MeOH from other laboratories, and in some respects compared to ethanol

(EtOH).

1. INTRODUCTION

1.1. Preamble

517

1.1.1. The regulatory problem

MeOH appears to be metabolized quite differently in rodents such as mice and rats than it is in humans. Nevertheless, most studies to estimate the human safety of MeOH in the developing embryo and fetus (hereafter collectively referred to as the embryo) have been carried out in rodents, raising the question of the accuracy, if not the relevance, of regulations based solely upon such studies.

1.1.2. Fundamental question

From the regulatory viewpoint, the fundamental question is whether the mechanism by which

MeOH harms the embryo is similar in rodents and humans. Since the mechanism of MeOH toxicity in the embryo remains to be established, it is not clear whether or to what extent the reported differences between rodents and humans in the metabolism of MeOH determine its relative toxicity in the embryo of these two species. It is similarly unclear whether the molecular effects of MeOH in rodent embryos reflect those in the developing human.

1.1.3. Research objectives

Our objectives were to determine:

(a) If reactive oxygen species (ROS) play a role in the embryonic toxicity of MeOH in

mice.

518

(b) If the mechanism of this embryotoxicity is the same in the mouse as it is in a non-

rodent species that, unlike the mouse, metabolizes MeOH like humans.

mice and adults from non-rodent species that metabolize MeOH more like humans.

1.1.4. Approach

Our approach was based upon methods and models that we have previously used to determine the role of ROS in adverse embryonic effects that either arise spontaneously or are caused by drugs (Wells et al., 2009b). The mouse was used as the basic animal model, and compared to a second non-rodent species that, unlike the mouse, appears to metabolize MeOH like humans, primarily via alcohol dehydrogenase (ADH1) and cytochrome P450 2E1 (CYP2E1), rather than via catalase as in rodents (Cederbaum and Qureshi, 1982; Makar et al., 1968; Tephly et al.,

1964). The toxicological relevance of catalase in rodents is potentially complex, since in addition to its role in MeOH metabolism, catalase also plays an antioxidative role in the detoxification of

ROS (Wells et al., 2009b) (fig. 1). Mice were chosen as a species for which: (1) MeOH teratogenicity has been well characterized (Degitz et al., 2004; Harris et al., 2003; Rogers and

Mole, 1997); and, (2) genetically modified strains are available for testing the contribution of specific biochemical pathways to the mechanism of teratogenesis. Rabbits were the first non- rodent species investigated as potentially reflective of humans, based in part upon their remarkable species difference from mice in susceptibility to birth defects caused by the ROS- initiating sedative drug thalidomide (Parman et al., 1999).

519

The intraperitoneal (i.p.) route, as opposed to inhalation or oral intake, was used in our studies to ensure that the dose delivered was as consistent as possible, in part for a precise determination of pharmacokinetic parameters, particularly across species like mice, rabbits and primates. Equally important, the resulting consistency in embryonic exposure allows a more reliable interpretation of results from mechanistic studies relating molecular changes to teratological outcomes. The high dose of MeOH, well above the lethal dose in humans, was based upon published studies of

MeOH teratogenicity in mice (Rogers et al., 2004), and retained in rabbits and primates to ensure that any teratogenic potential was not missed due to an insufficiently high dose, as well as to facilitate cross-species comparisons. MeOH was administered acutely rather than chronically to avoid the complicating secondary biochemical changes that accumulate with chronic exposures, which can confound the interpretation of data relevant to the mechanisms of teratogenic initiation. The high MeOH dose employed assured that any potential for initiating oxidative stress would not be missed by the use of an acute dose. This regimen produced teratogenic effects similar to those reported in the literature with other dosing regimens in susceptible mouse strains, so any teratologically relevant ROS involvement would be evident under these conditions.

1.2. Methanol developmental toxicity

MeOH developmental toxicity has been extensively studied in the mouse and rat, both in vivo and in vitro, but no human cases have been reported. The underlying mechanism of this developmental toxicity in rodents is not well understood. Studies in animal models have shown different species- and strain-dependent outcomes from MeOH in utero exposure. The CD-1 and

520

C57BL/6 strains of mice, Sprague-Dawley and Long-Evans rats have all been shown to be susceptible to MeOH teratogenicity in a dose-dependent manner, but to varying degrees and with a spectrum of outcomes depending on the strain or species. However, one strain of rat, the

Holtzman rat, was reported to be resistant to MeOH teratogenicity at doses of up to 3.2 g/kg bw/day with the specific dosing schedule employed in that study (Cummings, 1993)(table 7).

Additionally, rats appear to be less sensitive to the developmental effects of MeOH than mice, where concentrations at least four-fold higher were required in embryo culture to produce the same anomalies in rats as seen with mice (Andrews et al., 1993; Hansen et al., 2005; Harris et al., 2003). Given that the metabolism of MeOH differs between rodents and humans, it would be beneficial to determine if there is an alternative animal model better suited to predicting the risk of developmental outcomes of MeOH exposure in humans.

Teratogenicity of the related alcohol, ethanol (EtOH), is much better understood. In humans, exposure of the embryo to EtOH during gestation is known to lead to a variable spectrum of birth defects termed Fetal Alcohol Spectrum Disorders (FASD), which includes Fetal Alcohol

Syndrome (FAS). There are specific craniofacial abnormalities that are commonly associated with FAS, including a smooth philtrum (groove between nose and upper lip), thin upper lip, and small palpebral fissures, where the eye width is shortened (Jones and Smith, 1973).

Neurological and functional impairments can also result from fetal EtOH exposure (Jones, 2011).

It is difficult to determine the exact dose of EtOH in humans that leads to FASD, precluding a direct comparison of drug potency with MeOH, for which no human developmental effects have been reported. EtOH, like MeOH, is teratogenic in mice, and similarly exhibits strain differences in teratogenicity (Abel, 1982; Becker and Randall, 1989; Gilliam et al., 1987; Mattson et al.,

2001). Our studies further investigating species- and strain-specific differences in susceptibility

521 to both EtOH and MeOH are discussed later in Section 3 on Species and Strain Differences in

MeOH Teratogenicity.

1.3. Carcinogenic potential

EtOH consumed in alcoholic beverages is regarded as carcinogenic in humans based on epidemiological evidence (AICR, 2007; IARC, 2006; NTP, 2002). Mechanistic evidence supporting a causal role of acetaldehyde in alcohol-related esophageal cancer comes from studies in aldehyde dehydrogenase-2 (ALDH2)-deficient individuals who consume large quantities of alcoholic beverages (Baan et al., 2007; Brooks et al., 2009). However, there is a fundamental difference between EtOH and MeOH in that the majority of primary exposures to MeOH of toxicological interest other than accidental poisonings from oral ingestion will occur via exposure to vapours or dermal contact of the skin. It is important to keep this fundamental difference in mind as the exposure to EtOH in subsets of the human population will be orders of magnitude higher than that following environmental and/or industrial exposure to MeOH, and this factor is central to any assessment of risk even if these agents share a common potential mechanism of initiation due to structural or biochemical similarities. Among other unappreciated possibilities, MeOH could theoretically cause cancer either by the covalent binding of its reactive formaldehyde metabolite to DNA, or via the formation of highly reactive and potentially toxic forms of oxygen termed reactive oxygen species (ROS) and ROS-initiated changes in signal transduction or oxidative damage to DNA. Although several studies in mice and rats have found no evidence for MeOH carcinogenesis (NEDO, 1985; NEDO, 1987), one study reported an increase in lymphohematoreticular neoplasms in Sprague-Dawley rats exposed to MeOH in the

522 drinking water (Soffritti et al., 2007). The latter study was confounded by unusual experimental conditions whereby the rat colony had a high potential for Mycoplasma pulmonis lung infection, was not specifically pathogen-free, and had a high background level of lymphoma in the control animals (Cruzan, 2009; Soffritti et al., 2007). Our studies on carcinogenic potential are discussed later in Section 4 on Oxidative Stress.

1.4. Oxidative stress and other potential mechanisms of toxicity

The mechanism by which MeOH causes toxicity in the developing embryo is not clearly understood, but one contributing factor may be enhanced “oxidative stress”, or increased formation of ROS. These ROS include highly toxic “free radical” intermediates such as hydroxyl radicals that have been implicated in a number of human diseases and drug toxicities including cancer, neurodegenerative diseases and birth defects (Halliwell and Gutteridge, 2007; Klaassen,

2008) (See later Section 4 on Oxidative Stress).

Using a number of models involving genetically altered animals and drugs that enhance oxidative stress, results from our laboratory and others have shown that ROS can adversely affect embryonic development in at least two ways (Wells et al., 1997; Wells et al., 2009a;

Wells et al., 2009b). The first is by altering the level of signals, termed “signal transduction”, within embryonic cells that are involved in the activation of genes leading to the production or suppression of proteins ultimately necessary for normal development. The second effect of ROS is to damage cellular macromolecules such as RNA, DNA, proteins and lipid membranes, resulting in their inability to perform their normal developmental role.

523

Other potential mechanisms in the developmental toxicity of MeOH would include covalent binding to cellular macromolecules (RNA, DNA, proteins) by the reactive formaldehyde metabolite, which is substantially more embryotoxic in embryo culture than MeOH itself or its stable formate metabolite (Hansen et al., 2005). Additionally, as yet undetermined effects of

MeOH itself have been postulated to initiate teratogenesis (NTP, 2002). Although formaldehyde remains a candidate for the proximate toxic chemical species due to its high embryotoxic potency in embryo culture (Hansen et al., 2005), the physiological relevance of treating embryos with this reactive metabolite, as distinct from having it produced metabolically within the embryo, remains unclear. However, the lower levels of embryonic formaldehyde dehydrogenase levels in mice compared to rats (Harris et al., 2003), may explain the greater susceptibility of mice to MeOH teratogenesis, consistent with a role for formaldehyde as the proximate chemical teratogen. The acute ocular toxicity and lethality observed in humans poisoned with MeOH appears to be due to metabolic acidosis initiated by high concentrations of the FA metabolite, which appears not to play a role in the developmental toxicity of MeOH (See later Section 3.1).

Our studies on the role of ROS in MeOH developmental toxicity are discussed later in Section 4 on Oxidative Stress.

1.5. Factors affecting the human relevance of animal models

1.5.1. Species differences in metabolism

524

Primates, including humans, use alcohol dehydrogenase (ADH1) to oxidize MeOH to formaldehyde, whereas rodents use the peroxidative activity of catalase (Cederbaum and

Qureshi, 1982; Makar et al., 1968; Tephly et al., 1964) (fig. 1). Both humans and rodents convert formaldehyde into formic acid (FA) by formaldehyde dehydrogenase (ADH3) (Harris et al., 2004; Teng et al., 2001). FA is subsequently metabolized into carbon dioxide and water by a folate-dependent dehydrogenase (Johlin et al., 1987). Folate, however, is limited in humans, leading to an accumulation of the toxic FA metabolite following exposure to high levels of

MeOH (Perkins et al., 1995). Conversely, folate is not limited in rodents (Black et al., 1985), and a combined use of the peroxidative activity of catalase and the aforementioned folate-dependent pathway prevents the accumulation of the acutely toxic FA (Clary, 2003; Cook et al., 2001;

Harris et al., 2004). Accordingly, the metabolism of MeOH in rodents does not reflect that in humans.

Despite these differences in metabolic pathways, the rate of MeOH metabolism appears to be similar between primates and rats following a saturating dose of MeOH (Kavet and Nauss, 1990;

Makar et al., 1968; Tephly et al., 1964). This is in contrast to mice, in which the reported rate of

MeOH metabolism is approximately twice as fast as that in both primates and rats (Mannering et al., 1969) (table 1).

1.5.2. Dose of methanol and route of exposure

Following an 8 h inhalation exposure of 5,000 ppm MeOH, the blood MeOH concentration and daily dose has been determined for mice and rodents and projected for humans, where mice have upwards of a calculated 17-fold higher plasma concentration than that in humans (Perkins et al.,

1995). Different routes of MeOH exposure in various animal models and humans via i.p.

525 injection, oral ingestion and inhalation may produce different peak plasma concentrations of

MeOH, not to mention its metabolites, in both the mother and developing embryo, which may alter developmental risks.

2. SPECIES DIFFERENCES IN METHANOL METABOLISM

2.1. Enzymes and pathways

Additional comments on species differences particularly in comparison to humans are provided above in section 1.5.1.

2.1.1. Alcohol dehydrogenase

Alcohol dehydrogenase (ADH1) (EC 1.1.1.1) is an oxidoreductase enzyme responsible for the metabolism of alcohols into their corresponding aldehydes, using nicotinamide adenine dinucleotide (NAD+) as a cofactor (Lutwak-Mann, 1938) (fig. 2). The enzyme is constructed of two subunits, each subunit being one of 6 possible subunits (α, β, γ, π, χ or σ), where the class 1 isoenzymes of ADH1 are of the αα, αβ or αγ variety (Kedishvili et al., 1995). ADH1 is highly expressed in the liver, and each isoenzyme has a varying capacity to metabolize alcohols

(Wagner et al., 1983).

The physical properties of ADH1 differ considerably among species, where the affinity of

MeOH for ADH1 varies greatly (Mani et al., 1970; Pietruszko, 1975a; Pietruszko, 1975b)

(table 2). In humans and non-human primates, MeOH is preferentially oxidized to formaldehyde using the ADH1 β1β1 isoenzyme (Blomstrand and Ingemansson, 1984; Mannering

526 et al., 1969; Wagner et al., 1983). Saturable elimination results from a limited source of the

ADH1 cofactor NAD+, and occurs following doses greater than 0.1 g/kg MeOH in humans and non-human primates (Horton et al., 1992; Rang et al., 2003). In contrast, rodent ADH1 makes a minimal contribution to MeOH metabolism, as demonstrated in ADH-deficient mice (Bradford et al., 1993). The underlying mechanism responsible for the inability of rodents to efficiently metabolize MeOH by ADH remains unknown; however, species differences in the ADH gene or protein expression (Buhler et al., 1984) and differing isoenzymes (Algar et al., 1983) have been implicated.

ADH activity using EtOH as a substrate has been well characterized in humans and rodents; however, the use of MeOH as a substrate is less understood, and therefore assessment of species differences in ADH activity generally refer to the former substrate (Algar et al., 1983; Harris et al., 2003; Pikkarainen and Raiha, 1967). An interspecies comparison of ADH activity with

EtOH demonstrates parallels between humans and rodents in the increasing ADH activity throughout gestation and postnatally, with highest levels being reached in adulthood. However, human ADH activity is consistently higher than that found in both mice and rats (table 3).

2.1.2. Catalase

Catalase (EC 1.11.1.6) is a tetramer containing 4 strongly bound molecules of NADPH (Kirkman and Gaetani, 1984). Each tetrameric unit contains 4 heme groups, which determine the function of the enzyme based on their redox state. Consequently, catalase can exist in 3 states: (1) the resting state (containing ferric heme groups), which is responsible for the antioxidative activity of catalase; (2) Compound I (containing Fe4+ heme groups), which is responsible for the peroxidative activity of catalase; and (3) Compound II, which is the inactive form of the enzyme

527

(Goyal and Basak, 2010). The tightly bound NADPH molecules prevent the formation of

Compound II (Kirkman and Gaetani, 1984).

2.1.3.1. Peroxidative role

In the presence of H2O2 and a hydrogen donor such as an alcohol, the Compound I state of catalase oxidizes the alcohol to its aldehyde, in the process reducing catalase back to its resting state (fig. 3). In rodents, MeOH is metabolized by the peroxidative activity of catalase as opposed to ADH1, forming the toxic formaldehyde metabolite (Bradford et al., 1993; Karinje and Ogata, 1990). Conversely, humans and non-human primates do not metabolize MeOH using catalase (Mannering et al., 1969), likely due to low levels of endogenous H2O2 (Oshino et al.,

1975). In rodents, catalase is also responsible for the peroxidative oxidation of formic acid to carbon dioxide and water (Smith and Taylor, 1982; Von Burg, 1994) (fig. 3)—a process that does not occur in humans.

In rodent embryos, peroxidative catalase activity increases with gestational age, and comparatively, the mouse embryo has consistently higher specific catalase activity than the rat embryo (Harris et al., 2003) (table 4). The peroxidative activity of catalase follows Michaelis-

Menten kinetics, where rat catalase demonstrates an affinity for MeOH of approximately 1.5 mM

(Perkins et al., 1995).

2.1.3.2. Antioxidative role

In all species, catalase acts as an antioxidant by scavenging H2O2 (fig. 4) (Halliwell and

Gutteridge, 2007; Wells et al., 2009b). Like its peroxidative activity, antioxidative catalase

528 activity increases throughout gestation and into adulthood in both humans and rodents, where mouse embryonic catalase activity is approximately 4 % of adult activity (table 5) (Abramov and

Wells, 2011a; Winn and Wells, 1999). Furthermore, substantial species differences in antioxidative catalase activity have been documented (Reddy et al., 1984; Sweeting et al.,

2010). Unlike the peroxidative activity of catalase, antioxidative catalase activity follows

Michaelis-Menten kinetics only at low concentrations of H2O2, due to the high reaction rate of this process and the ability of high H2O2 levels to inactivate the enzyme. Accordingly, only theoretical kinetic parameters for antioxidative catalase may be calculated, where an apparent

Km of 25 mM has been reported for H2O2 (Vetrano et al., 2005). As such, the apparent affinity of rodent catalase for H2O2 is approximately 20-fold lower than that for MeOH (see section

2.1.3.1), indicating that MeOH will be preferentially metabolized over H2O2. With less H2O2 being scavenged, the potential for ROS formation is enhanced, leading to an increased possibility of oxidative damage.

2.1.3. Cytochrome P450 (CYP) 2E1

Alcohols including EtOH and MeOH can be metabolized by a microsomal EtOH oxidizing system (MEOS) to their respective aldehydes (Teschke et al., 1974). The MEOS system principally relies on the cytochromes P450 (CYP) isozymes CYP2E1 and CYP1A2 for alcohol metabolism, and is dependent on the presence of molecular oxygen and NADPH as the cofactor

(Kunitoh et al., 1993; Lieber, 2004) (fig. 5).

2.1.4. Formaldehyde dehydrogenase (ADH3)

529

The half-life of formaldehyde is approximately 1.5 minutes (McMartin et al., 1979) due to its rapid oxidation into FA and/or the formation of macromolecular adducts (Shaham et al., 1996).

Using ALDH2, formaldehyde can be metabolized into FA in the mitochondria (Teng et al.,

2001). Additionally, formaldehyde dehydrogenase (ADH3) can oxidize formaldehyde into FA in both humans and rodents through a GSH-dependent mechanism (Harris et al., 2004; Uotila and

Koivusalo, 1974)(fig. 6). ADH3 is approximately 100-fold more efficient at oxidizing formaldehyde than ALDH2, suggesting that it is the predominant pathway for formaldehyde metabolism in vivo (Teng et al., 2001).

During development, rat and mouse embryos have similar ADH3 specific activity (Harris et al.,

2003).

2.1.5. Folate

Folic acid (FA) is metabolized to carbon dioxide and water by a folate-dependent dehydrogenase. Using 10-formyl-tetrahydofolate (THF) synthetase, FA is converted to 10- formyl THF, which in turn is converted to carbon dioxide and water by 10-formyl-THF dehydrogenase (Black et al., 1985) (fig. 7). This metabolic pathway is maintained in all species, however varying levels of the cofactor THF limit the efficiency of FA oxidation (McMartin et al., 1981). Specifically, the mouse has over 4-fold higher basal THF levels than humans, effectively decreasing the accumulation of FA and its associated toxicities (Sokoro et al.,

2008)(table 6). In primates, the THF-dependent metabolic pathway of FA oxidation becomes saturated with doses of MeOH exceeding 300 mg/kg (Kavet and Nauss, 1990).

530

In addition to the enzyme-catalyzed metabolism of FA discussed above, there are also

H+/formate co-transporters in the renal tubes, which could potentially be present in the yolk sac and placenta and subsequently effect species differences in the developmental susceptibility to

FA exposure (Hovda et al., 2007).

2.2. Pharmacokinetics of methanol and formic acid

In light of the substantial species differences in MeOH metabolism between humans and rodents

(see section 2.1 for more details), we sought to establish if human MeOH metabolism might be better reflected in rabbits than rodents. The plasma pharmacokinetics of MeOH and FA were compared in male CD-1 mice, New Zealand white (NZW) rabbits and cynomolgus monkeys over time (24, 48 and 6 h, respectively) following a single i.p. injection of 0.5 or 2g/kg MeOH or its saline vehicle. Following the 2 g/kg dose, MeOH exhibited zero-order elimination kinetics in all 3 species, with comparable peak concentrations and a 2.5-fold higher clearance in mice than rabbits (Sweeting et al., 2010) (figs. 8, 9). FA accumulation within 6 h in primates was 5-fold and 43-fold higher than in rabbits and mice respectively, with accumulation being 10-fold higher in rabbits than mice (fig. 10). Over 48 h, FA accumulation was nearly 5-fold higher in rabbits than mice (fig. 9). Low-dose MeOH in mice and rabbits resulted in similarly saturated MeOH elimination in both species, but with approximately 2-fold higher clearance rates in mice (fig. 9).

Furthermore, FA accumulation was 3.8-fold higher in rabbits than mice (fig. 9), suggesting that rabbits more closely than mice reflected primates for in vivo MeOH metabolism, and particularly

FA accumulation (Sweeting et al., 2010).

531

MeOH and FA pharmacokinetics were also studied in two additional mouse strains: C57BL/6J and C3H mice. In comparison to the previously mentioned CD-1 mouse study, employing the same acute dosing regimen, C57BL/6J and C3H mice both exhibited pharmacokinetic profiles similar to that in the CD-1 strain (fig. 11) (Sweeting et al., 2011). Peak plasma concentrations and AUC levels were identical between mouse strains over the initial 12 h period of sample collection.

Our results can be compared with other pharmacokinetic studies performed in other species and through different routes of administration (table 1). MeOH and FA peak concentrations in our mice administered (i.p.) 20 times the dose given to rats (i.v.) were approximately 23-fold and 4- fold higher, respectively, in the mice (Horton et al., 1992). However, the MeOH clearance rates and FA AUC values found in our mouse studies (2 g/kg MeOH i.p.) were comparable to those found in the rat (100 mg/kg MeOH i.v.). MeOH plasma concentration levels in humans at which toxicity is observed have not been reported above 70 mM, which is below the peak concentration seen in mice in which no toxicity is observed, highlighting the species difference in susceptibility to acute toxicity. Non-human primate MeOH and FA peak plasma concentrations, as described earlier, have been reported to be higher than those observed in humans, with levels up to 90 mM and 2 mM respectively (Horton et al., 1992; Sweeting et al., 2010) (Table 1).

3. SPECIES AND STRAIN DIFFERENCES IN METHANOL TOXICITY

3.1. Acute metabolic acidosis, ocular toxicity and death

In humans, acute rather than chronic or developmental toxicities of MeOH are predominantly reported in the literature, often due to accidental ingestion or as a substitute for EtOH. Patients

532 often present with central nervous system (CNS) depression, metabolic acidosis, ocular toxicity, regional brain damage (primarily in the putamen) and death (Suit and Estes, 1990). Plasma concentrations of MeOH and its metabolite FA are often measured and monitored upon arrival at the hospital and throughout treatment. Metabolic acidosis due to FA accumulation, as indicated by a lower blood pH, is believed to be the cause of ocular toxicities and death in humans. This deviation from normal physiological pH disrupts the anion gap and inhibits cytochrome oxidase on the mitochondrial membrane, localized in the optic nerve and retina (Baumbach et al., 1977;

Hayreh et al., 1977). FA inhibits cytochrome oxidase when at concentrations between 5 and 30 mM, which overlaps with concentrations observed in patients demonstrating acute MeOH intoxication (Nicholls, 1976; Nicholls, 1975). If cells are sufficiently damaged, swelling and edema result, which can compress the optic nerves and potentially lead to cell death. As metabolic acidosis worsens, further symptoms may include shallow respiration, cyanosis, coma and electrolyte disturbances, which eventually lead to death (Suit and Estes, 1990). The blood

MeOH concentration required to cause these acute effects vary, but has been reported to fall within the range of 6 – 16 mM for CNS effects, 16 – 47 mM for ocular toxicity, and between 42

– 62 mM for fatalities (IPCS, 1997) (fig. 12). These concentrations, however, are an inaccurate representation of the actual blood levels required to cause these acute effects of MeOH toxicity, as several factors must be taken into consideration: (a) The amount and composition of the ingested MeOH solution is unknown; (b) the time since initial ingestion is unknown, and therefore it is unknown how much of the MeOH has already been metabolised; and, (c) patients will receive some form of intervention to alleviate the symptoms of MeOH poisoning. This includes treatment with EtOH, which acts as a competitive inhibitor of ADH1 due to its higher affinity than MeOH for the metabolizing enzyme, ADH1 (Ekins et al., 1985; Mani et al., 1970).

By saturating ADH1, further metabolism of MeOH and the subsequent acute toxicities due to FA

533 accumulation are inhibited. Hemodialysis, treatment with bicarbonate for acidotic patients, or with drugs such as fomepizole, which is a competitive inhibitor of ADH, are also common interventions {

MeOH acute toxicity in animal models is completely different. Although primates, as expected, exhibit similar symptoms (ocular toxicity, metabolic acidosis) {Baumbach, 1977 #807}, rodents and rabbits do not exhibit either of these manifestations (Roe, 1955). Our laboratory has performed single dose (2 g/kg MeOH), 2-day and 15-day (daily 2 g/kg doses of MeOH on consecutive days) pharmacokinetic studies which found no accumulation of MeOH or FA in mice, whereas a sustained accumulation of both MeOH and FA was observed in rabbits following a single dose of MeOH (McCallum et al., 2011a; McCallum et al., 2011b; Sweeting et al., 2010). However, as previously reported, no evidence of ocular toxicity or metabolic acidosis was observed in either mice or rabbits, despite greater FA accumulation in the latter, and both species appeared normal throughout the studies.

Although acute MeOH toxicity in humans is thought to be caused by FA accumulation and the resulting metabolic acidosis, the mechanism of developmental toxicity is not well understood, but FA accumulation is unlikely to be involved. Pharmacokinetic studies comparing blood plasma concentrations of both MeOH and FA in different strains of mice and NZW rabbits, together with complementary in utero MeOH exposure studies in the same animals, indicate that pharmacokinetics alone are insufficient to explain the marked differences in susceptibility between strains and species to MeOH developmental toxicity. Similarly among mouse strains

(C57BL/6J, C3H and CD-1), pharmacokinetic profiles are similar (see Section 2.2), yet each strain has a different susceptibility to MeOH in utero exposure; C57BL/6J mice and CD-1 mice are susceptible to morphological anomalies, and even these strains exhibit strain-specific

534 abnormalities, whereas C3H mice are relatively resistant (Sweeting et al., 2011) (see Section

3.2). Only morphological abnormalities were measurably affected by MeOH administration, whereas litter parameters such as litter size or the incidence of resorptions and stillbirths were not altered. Furthermore, NZW rabbits exhibited a pharmacokinetic profile more similar to that of humans, in that FA levels accumulated and were sustained for more than twice the time observed in mice (Sweeting et al., 2011) (fig. 11). Rabbits also proved to be resistant to MeOH developmental toxicity, despite their substantially different metabolism of MeOH, including FA accumulation (Sweeting et al., 2011).

3.2. Teratogenesis

Species differences in teratogenicity are clearly exemplified in the case of thalidomide, whereby rodents are resistant while rabbits, non-human primates and humans are susceptible.

Additionally, in utero thalidomide exposure significantly increased DNA oxidation in rabbit embryos but not in mouse embryos, suggesting oxidatively damaged DNA may be involved in the observed species difference (Parman et al., 1999).

There have been no well-documented cases of human MeOH developmental toxicity. Animal models, mostly rodents, have shown species- and strain-specific sensitivity to MeOH during pregnancy, with a variable spectrum of morphological abnormalities. Rodent studies have demonstrated that the mouse is more sensitive than the rat to MeOH developmental toxicity in embryo culture, where concentrations at least four-fold higher were required to obtain the same outcomes in rats as in mice (Andrews et al., 1993; Hansen et al., 2005; Harris et al., 2003). This difference in susceptibility between the rat and mouse may potentially be explained at least in part by differences in MeOH metabolizing enzymes. Although mouse embryos express lower

535 levels of ADH1, responsible for converting the parent drug MeOH to formaldehyde, the embryonic level of ADH3, responsible for converting formaldehyde to FA, is lower in mice than in rats, potentially resulting in higher formaldehyde levels in the mouse embryo (Harris et al.,

2003). Formaldehyde when added directly to the culture medium is 1000-fold more embryotoxic than either MeOH or its FA metabolite in rat embryo culture (Hansen et al., 2005). Addition of all three chemical species to whole embryo culture decreased embryonic glutathione (GSH) levels, but only MeOH and formaldehyde caused dysmorphogenesis (Harris et al., 2004).

Furthermore, depletion of embryonic GSH by pretreatment with L-buthionine-S,R-sulfoximine

(BSO), an inhibitor of GSH synthesis, exacerbated these embryotoxic effects, suggesting a protective role for GSH-dependent detoxification of formaldehyde (Harris et al., 2004).

Combined with the enhanced susceptibility of mouse embryos compared to rats, these studies implicate oxidative stress in the mechanism of MeOH teratogenesis (see later Section 4 on

Oxidative Stress).

A single primate study found no evidence for a teratogenic effect with inhaled MeOH at the doses employed (NEDO, 1987), but this study has yet to replicated. Moreover, the NEDO results were originally judged to be inadequately reported (NTP, 2002), although more information has since been made available through the Methanol Institute

(http://www.methanol.org). Rodent models, however, have been extensively explored. Both rats and mice, as previously mentioned, are susceptible to MeOH teratogenicity, albeit to different degrees. Rats and mice exhibit different morphological outcomes, which are dependent on the rodent strain employed, the gestational day (GD) of exposure and the route of administration.

The rabbit is another potential candidate as a model for human MeOH developmental toxicity.

Parallel studies in C57BL/6J, C3H mice and NZW rabbits reveal that, like C3H mice, rabbits are

536 resistant to MeOH teratogenicity, despite differences in pharmacokinetics and MeOH metabolism between these species. There was no measurable effect of MeOH on any developmental parameters (litter or morphological), similar to C3H mice, with both species being exposed to similar dosing schedules during gestation (Sweeting et al., 2011) (figs. 13, 14).

In addition to species differences in MeOH teratogenicity, there are also strain-specific malformations that result from MeOH exposure during gestation (table 7). Studies primarily in rodents have outlined these differences, emphasizing the difficulty in identifying the most representative animal model for MeOH developmental toxicity in humans. Sprague-Dawley rats exposed to varying concentrations of inhaled MeOH (5,000 to 20,000 ppm for 7 h/day) exhibited only skeletal and cardiovascular effects (NEDO, 1987; Nelson et al., 1985). In contrast,

Holtzman rats exposed to a range MeOH doses by gavage (1.6 – 3.2 g/kg body weight) exhibited no significant embryonic or fetal effects with treatment (Cummings, 1993). Long-Evans rats, unlike the aforementioned strains, exhibited ophthalmic abnormalities and abnormal development of sexual organs when exposed to a range of MeOH doses by gavage (1.3 – 5.2 ml/kg or 1023 - 4090 mg/kg bw according to CERHR calculations (NTP, 2002)) on GD 10

(Youssef et al., 1997). Among mouse strains, CD-1 mice exhibit cephalic neural tube defects

(NTDs) with the highest incidence, along with facial and palatal clefts and skeletal defects

(Bolon et al., 1994). C57BL/6 mice, on the other hand, primarily exhibit ophthalmic abnormalities, including microphthalmia (smaller eyes) and anophthalmia (absence of the eye), along with facial and palatal clefts and skeletal defects (Rogers et al., 2004). Overall, C57BL/6 mice are more susceptible to the effects of in utero MeOH exposure, displaying a much higher incidence of affected pups per litter, more than twice that observed in CD-1 mice. The periods of

537 susceptibility among strains are similar, however, with MeOH administration on GD 7 or 8 resulting in the highest incidence of affected pups.

Comparing the potency of MeOH to the related alcohol, EtOH, in whole embryos of various strains of mouse exposed in culture for 24 hours, for some developmental parameters there is no distinct difference in embryopathies caused by these two alcohols at a similar molar concentration. However, for other parameters, it appears that EtOH reduces anterior neuropore closure, turning and somite development by 50 mM while MeOH reduces these parameters at or above 100 mM, suggesting EtOH is more potent (fig. 15). In rat embryos, EtOH appears to be more potent than MeOH, as EtOH reduces crown-rump length, head length, somite development and protein content by 50-100 mM, while MeOH reduces these parameters at or above 100 mM

(fig. 16). Some outcomes assessed in the same strain by the same lab exhibited differences, as observed with protein content measured in Sprague-Dawley rat embryos (fig. 16). There are no remarkable species differences between mouse and rat embryos exposed to MeOH for 24 hr in culture except for embryolethality, whereby mouse embryos die above 250 mM MeOH, while rat embryos can survive above 350 mM (fig. 17).

Comparisons are simpler in a whole embryo culture model where the xenobiotic concentration and duration of exposure can be controlled, whereas comparing the potency of MeOH and EtOH in vivo is more difficult. Prenatal exposure of rodents to both alcohols causes birth defects, the incidence of which increases with higher doses, with the anomalies varying depending on the time of exposure as well as the peak blood alcohol concentration (BAC) achieved (Becker et al.,

1996) (table 11). Human prenatal EtOH exposure results in a characteristic spectrum of anomalies, collectively termed FASD, which includes decreased fetal weight, distinct

538 craniofacial malformations, and neurobehavioural deficits (Jones and Smith, 1973). On the other hand, while prenatal MeOH exposure is developmentally toxic in mice and rats, the mechanism is unclear, and it is not known whether MeOH is developmentally toxic in humans. There are several confounding factors in attempting to compare the embryopathic potencies of EtOH and

MeOH in vivo. In some studies, the maternal peak BAC achieved in the model may not have been measured, let alone concentrations in embryonic or fetal tissues, so it is difficult to compare one alcohol and dose to another. Additionally, mice of different strains exhibit varied susceptibility to the same dose of drug such that their dose-response curves are shifted, with one strain of mouse being more or less susceptible than another strain under the same experimental conditions (Chernoff, 1980; Weston et al., 1994). Furthermore, dosing regimens, routes of administration, as well as outcomes measured differ across studies, precluding direct comparisons. In vivo studies comparing the same molar equivalent dose of EtOH and MeOH in the same strains and species under the identical conditions measuring the same developmental endpoints and at least the peak maternal BAC, if not embryonic or fetal tissue concentrations, would be useful in determining more definitively the relative teratological potencies of these two alcohols.

3.3. Neurodevelopmental effects

FASD in humans resulting from EtOH exposure during pregnancy may include neurodevelopmental effects, such as cognitive and other behavioural deficits (Jones, 2011).

Behavioural deficits as a result of in utero MeOH exposure remain to be definitively determined due to the lack of reports in humans and investigations in animal models. Only Long-Evans rats given either MeOH in drinking water (GDs 15-17 or 17-19) or inhaled MeOH (GD 7-19) during

539 pregnancy have been studied for postnatal behavioural and cognitive deficits (Infurna and Weiss,

1986; Stanton et al., 1995) (table 8). The first study monitored litter parameters such as litter size, fetal weight and infant mortality, as well as eye opening, suckling and nest-seeking

(homing) behaviour. Pups exposed to MeOH required more time to suckle on postnatal day

(PND) 1 and also required more time to locate nesting material from their home cages on PND

10, suggesting a delay in normal behavioural development. The second study, where rats were given a chronic high dose (15,000 ppm 7 h/day, GDs 7-19) of inhaled MeOH, also looked at litter parameters, and in addition employed a variety of behavioural and cognitive tests for motor activity, olfactory learning, behavioural thermoregulation, T-maze learning, acoustic startle response, reflex modification, pubertal landmarks, passive avoidance and visual evoked- potentials. The only observed effect of MeOH was a decrease in fetal body weight, while no effects were observed with any of the behavioural tests employed. A single non-human primate study was conducted in the primate model Macaca fascicularis, where pregnant monkeys were exposed to a range of concentrations of inhaled MeOH throughout the pre-mating and entire gestational period (Burbacher et al., 1999). Postnatal behaviour was followed for 9 months after birth, employing neurodevelopmental behavioural tests for early reflex response, motor development, spatial memory and social behaviour. In utero MeOH exposure caused a decrease in sensorimotor development only in male progeny in all treated groups. MeOH exposure also demonstrated a failure for fetuses, in all treatment groups and regardless of sex, to show preference for novel stimuli as tested by the Fagan Test of Infant Intelligence. No other tests were affected by in utero MeOH exposure. Results of this study must be interpreted with caution due to the small sample size used to assess the neurobehavioural outcomes, as random fluctuations could potentially account for the statistical differences observed. On the other hand, sensorimotor development is a complex developmental process and the presence or absence of

540 an observed effect at this early stage does not preclude a delayed onset neurotoxic effect later in life.

3.4. Carcinogenic potential

Although the acute toxicity of MeOH in humans is well described, its carcinogenic potential, much like its teratogenic potential, is unknown. It is similarly unknown which chemical species, whether MeOH and/or any of its metabolites, might be a proximal toxicant. Formaldehyde itself is a known carcinogen that causes nasopharyngeal cancer in humans and squamous cell carcinomas of the respiratory epithelium of rats and mice (IARC, 2006; Kerns et al., 1983), but this species is rapidly metabolized by ADH3 to FA, and increases in formaldehyde levels were not observed in body fluids or tissues following acute high-dose MeOH exposures in early studies with formaldehyde detection limits of 25 µM (Makar and Tephly, 1977; McMartin et al.,

1977; McMartin et al., 1979). Formaldehyde is a metabolic intermediate endogenously produced from amino acid metabolism and exists at concentrations of about 0.1 mM in human blood (IARC, 2006). In addition to rapid metabolism to FA, formaldehyde can be sequestered as macromolecular adducts to endogenous nucleophiles such as DNA, lipid and protein. Recent formaldehyde inhalation studies using radiolabeled formaldehyde reported that exogenous formaldehyde DNA adducts were observed only in nasal DNA, and that the exogenous adduct levels did not exceed the levels of endogenous formaldehyde DNA adducts, suggesting that additional mechanism(s) in addition to DNA adduct formation are involved in formaldehyde- dependent nasopharyngeal cancer (Lu et al., 2010; Moeller et al., 2011). Exposure of Sprague-

Dawley rats to MeOH (20,000 ppm) or formaldehyde (1,500 ppm) in drinking water for 7 days failed to increase the basal levels of the major formaldehyde-DNA adduct N6-

541 hydroxymethyldeoxyadenosine (N6-HOMe-dAdo) in leukocyte or hepatocyte DNA (Wang et al.,

2008). Another potential indirect mechanism of MeOH-initiated carcinogenesis is via ROS- mediated oxidative damage to DNA. MeOH could promote ROS formation directly via a free radical intermediate, or indirectly via mechanisms including the activation and/or enhancement of ROS-producing NADPH oxidases, as has been reported for EtOH (Dong et al., 2010). Several studies have reported free radical production during MeOH biotransformation (Castro et al.,

2002; Paula et al., 2003; Skrzydlewska et al., 2000), with the most direct evidence being the direct detection of alpha-(4-pyridyl-1-oxide)-N-tert-butylnitrone (POBN)-hydroxymethyl radical adducts in bile and urine from rats acutely intoxicated with MeOH (Kadiiska and Mason, 2000).

Oxidatively damaged DNA has been implicated for decades as a causal factor in carcinogenesis

(Ames, 1989), the most commonly measured lesion being 8-hydroxyguanine, or its physiologically prevalent keto form, 7,8-dihydro-8-oxoguanine, commonly termed 8-oxo-2'- deoxyguanine (8-oxodG). 8-OxodG is the most abundant promutagenic oxidation product of guanine, yielding G-to-T transversion mutations that could activate oncogenes or inactivate tumor suppressor genes linked to the development of cancers (Hsu et al., 2004; Klaunig and

Kamendulis, 2004). Genetically altered mice with deficiencies in DNA glycosylases that protect against G-to-T transversions provide strong evidence of a causal role for oxidatively damaged

DNA in tumorigenesis (Kinoshita et al., 2007; Russo et al., 2004; Xie et al., 2004).

Dimethylarsinic acid strongly increases 8-oxodG levels and carcinogenicity in lungs of Ogg1 KO mice (Kinoshita et al., 2007). In double mutant mice deficient for Ogg1 and Myh, 8-oxodG accumulates in lung and small intestine, and these organs have multifold increases in cancer incidence with a high frequency of G-to-T transversion mutations that activate the K-ras oncogene in lung cancers (Xie et al., 2004). Deficiencies in the repair of 8-oxoG have also been

542 suggested to be risk factors for the development of human lung cancer (Mambo et al., 2005;

Paz-Elizur et al., 2003).

There are no human data regarding the carcinogenic potential of MeOH, and only four rodent studies, which have reported conflicting results (table 9). Inhalation studies in mice and rats performed by the New Energy Development Organization (NEDO, 1987) in Japan concluded that carcinogenic effects were not evident following chronic exposure to MeOH, and that there were no differences in the incidence of leukemias or lymphomas. The lack of developmental effects in non-human primates is not entirely consistent with the published evidence of developmental toxicity discussed earlier in section 3.2. The NEDO studies were initially difficult to assess as they were available only as summary reports, but more complete English translations are now available from the Methanol Institute (www.methanol.org). An extensive chronic carcinogenic study performed in Sprague-Dawley rats by the Ramazzini Foundation

(Soffritti et al., 2007) exposed rats to a range of MeOH concentrations in drinking water and reported dose-dependent increases in lympho-immunoblastic lymphomas and ear duct carcinomas. However, the Ramazzini Foundation studies have been questioned and its results are not widely accepted (Cruzan, 2009; NTP, 2002). Firstly, its methodology does not allow for an accurate estimate of the actual dose of MeOH consumed by the rats in drinking water. Total water consumption was measured only weekly, rather than daily, per cage, where multiple rats were housed, which may inappropriately average individual variability. The classification of the results made their interpretation difficult, as the diagnostic terms used did not match standard pathological guidelines. Of the classifications made by the pathologists at the Ramazzini

Foundation, nearly half did not match those reviewed by the National Toxicology Program

(NTP, 2004). There were also discrepancies with the study’s control data in comparison to

543 historical control data, where the incidences of total cancers derived from blood-forming cells were consistently approximately four times higher. Furthermore, their report neglected to acknowledge the high incidence of early mortality (>80% of rats in study) and lung pathologies present in nearly all dose groups, which likely contributed to the formation of the reported lympho-immunoblastic lymphomas. Additionally, the report lacked pertinent information that could help the interpretation of results, such as the aforementioned limited data on water consumption and limited information concerning lung pathology. These limitations diminished the validity of the study, and its results would require replication to corroborate the conclusions made in the report. One additional report in the literature is a graduate thesis examining the carcinogenic potential of malonaldehyde in Swiss Webster Mice, in which MeOH was utilized as a vehicle control, although unfortunately there were no concurrent untreated controls in this study (Apaja, 1980). Mice were exposed to MeOH in drinking water at doses of approximately

550, 1,000, or 2,000 mg/kg/day six times per week until their spontaneous death. The incidence of malignant lymphomas in MeOH-treated mice was higher than the overall incidence in historical controls for high-dose females and mid-dose males, but the author concluded that these incidences were "within the normal range of occurrence of malignant lymphomas in Eppley

Swiss mice." The mouse colony was not maintained under specific pathogen-free (SPF) conditions, and there was a high incidence of pneumonia reported (8-28%), which may be a confounding variable.

4. OXIDATIVE STRESS

4.1. Oxidative stress mechanisms

4.1.1. Embryonic drug exposure and reactive oxygen species (ROS) formation

544

Maternal elimination of a teratogen can be an important regulator of the amount reaching the embryo, so maternal pathways such as CYP-catalyzed hydroxylation reactions and UDP- glucuronosyltransferase-catalyzed drug conjugation are important determinants of embryonic teratogen exposure (Wells et al., 2005). However, since ROS, and particularly hydroxyl radicals, are highly reactive and unlikely to escape the cell in which they are formed, let alone the tissue or organ, maternal pathways of ROS formation are unlikely to contribute to embryonic ROS levels, which are determined by proximate, embryonic pathways. The risk of embryopathies will likely be determined by a balance among: (1) the maternal pathways of teratogen elimination and

(2) embryonic pathways of ROS formation and detoxification, and repair of ROS-mediated oxidative macromolecular damage to cellular macromolecules such as DNA, protein and lipids.

When an imbalance in the above pathways occurs, teratogenesis can result even at therapeutic drug doses or maternal plasma concentrations, or at exposures to levels of environmental chemicals generally considered to be safe.

Drugs and environmental chemicals can enhance ROS formation via a number of mechanisms that are not necessarily mutually exclusive for a given xenobiotic. These mechanisms include:

(1) enzymatic bioactivation to a free radical intermediate, catalyzed by cytochromes P450s

(CYPs), prostaglandin H synthases (PHSs) and lipoxygenases (LPOs), among others (Wells et al., 2009b; Wells et al., 2010) (fig. 18); (2) superoxide formation during the metabolism of substrates like EtOH by CYP2E1 (Koop, 2006); (3) redox cycling of catechol metabolites (Wang et al., 2010); (4) interference with the mitochondrial electron transport chain, producing superoxide (Maritim et al., 2003); and, (5) activation and/or induction of enzymes like the

NADPH oxidases (NOXs) that form superoxide and/or hydrogen peroxide (Brown and

Griendling, 2009; Jiang et al., 2011; Lambeth, 2004) (fig. 19).

545

4.1.2. Signal transduction

ROS are widely implicated in highly regulated cellular signal transduction pathways, which are selective to different cell types and their subcellular organelles (Wells et al., 2009b) (fig. 20).

ROS signalling has been linked to numerous pathways involved in cellular proliferation, differentiation, migration and apoptosis (Thannickal and Fanburg, 2000). ROS-mediated signal transduction has been attributed to hydrogen peroxide, which is less reactive and has greater diffusibility than superoxide or hydroxyl radicals, allowing it to selectively oxidize sulfhydryl groups of specific cysteine residues on proteins resulting in several reversible modifications including the formation of protein-protein (Pr-Pr) and glutathione (GSH)-protein disulfides (GS-

Pr, mixed disulfides) (Janssen-Heininger et al., 2008; Thannickal and Fanburg, 2000). At physiologically relevant concentrations of hydrogen peroxide, these modifications are reversible and constitute a control mechanism of protein function, whereas higher exposures could lead to excessive and irreversible S-oxidation resulting in loss of protein function and pathological consequences. Major cellular sources of ROS include the mitochondrial electron transport chain, lipoxygenases, as well as NOXs (Lander, 1997). NOX enzymes are commonly known for their role in phagocytes for initiating the ‘respiratory burst’ by producing high concentrations of superoxide and hydrogen peroxide. Homologs were later found to be expressed ubiquitously in non-phagocytic cells, although they produce more physiological levels of ROS involved in signal transduction (Brown and Griendling, 2009; Lambeth, 2004). ROS can modulate signalling pathways involving protein kinases and transcription factors such as nuclear factor transcription factor kappa B (NF-κB), p53, mitogen-activated protein kinases (MAPK), and protein kinase C

546

(Halliwell and Gutteridge, 2007). ROS-initiated transcription factor activation can alter cellular gene expression leading to the downstream cellular response (Brown and Griendling, 2009). A specific example of xenobiotic initiated ROS-mediated signal transduction altering embryonic development is phenytoin, an antiepileptic drug given during pregnancy to mitigate seizure occurrence. Our lab has shown the involvement of the NF-κB signalling pathway in phenytoin- initiated embryopathies in culture (Kennedy et al., 2004). Using antisense oligonucleotides, inhibition of the downstream NF-κB signalling cascade blocked embryopathies, suggesting the involvement of ROS-initiated NF-κB signalling in phenytoin-initiated embryopathies. The NF-

κB family of transcription factors regulate the expression of many genes involved in development, as well as immunity and the inflammatory response (Baeuerle and Baltimore,

1996). Another example is thalidomide, a drug used for its sedative-hypnotic effects in pregnant women, and is still in clinical use for the treatment of leprosy and multiple myeloma due to its strong immunomodulatory, anti-inflammatory and anti-angiogenic properties. When taken during the 3rd-8th week of pregnancy, thalidomide can initiate birth defects predominantly involving the limbs (phocomelia is the most well known) but can also affect the ears, eyes, heart, kidneys and other internal organs (Knobloch and Ruther, 2008). The mechanisms by which thalidomide initiates birth defects are not clear, however several mechanisms have been suggested including oxidative stress and alterations in signal transduction pathways.

4.1.3. Macromolecular damage

547

Elevated ROS concentrations increase the likelihood that they will react with molecular oxygen to form superoxide via a one-electron reduction, hydrogen peroxide via a second electron reduction or enzymatically catalyzed by superoxide dismutase, or the highly reactive hydroxyl radical via the Fenton reaction with hydrogen peroxide. Hydroxyl radical can oxidatively damage cellular macromolecules including proteins, DNA and lipids (Halliwell and Gutteridge,

2007).

Protein oxidation

Protein oxidation can impair the function of signal transduction proteins, receptors and enzymes, and subsequently cause secondary damage to other cellular macromolecules. If damage occurs to

DNA repair enzymes, levels of DNA damage can increase, while damage to DNA polymerases may decrease replication fidelity (Halliwell and Gutteridge, 2007). Oxidized protein may be recognized as a foreign antigen by the innate immune system and trigger antibody production

(Peng et al., 1997). Oxidation of a protein is initiated by the hydroxyl radical-dependent abstraction of the alpha-hydrogen atom of an amino acid residue to form a carbon-centered radical, which can rapidly react with oxygen to form subsequent radical intermediates that can react with other amino acid residues in the same or a different protein, forming a new carbon centered radical (Berlett and Stadtman, 1997). All amino acid side chains are susceptible to attack by hydroxyl radicals. A well-studied measure of protein oxidation is protein carbonyl formation. Carbonyl groups can be introduced into proteins either by direct metal-catalyzed oxidation of lysine, arginine, proline and threonine residues, or by reaction with aldehydes

548 produced during lipid peroxidation, such as 4-hydroxy-2-nonenal (HNE), and malondialdehyde

(MDA), or with reactive carbonyl derivatives (Berlett and Stadtman, 1997; Nystrom, 2005).

Histidine attacked by the reactive lipid peroxidation aldehyde product HNE can generate an

HNE-histidine adduct that can be measured as a marker of lipid peroxidation (Uchida and

Stadtman, 1992) (fig. 21). Oxidized protein products are removed from the cell by increased recognition and degradation by cellular proteases, and loss of developmentally important proteins in the embryo could lead to subsequent embryopathies, which has been observed with phenytoin-initiated protein oxidation (Nystrom, 2005; Winn and Wells, 1999). The presence of elevated levels of carbonylated proteins has been used as a marker of ROS-mediated protein oxidation and several methods of detection have been developed (Levine et al., 1994).

DNA oxidation

Exposure of DNA to highly reactive hydroxyl radicals can yield several products, depending where the oxidation occurs, which can accumulate and may contribute to teratological outcomes.

Approximately 20 forms of oxidatively damaged DNA have been identified, the most commonly measured being 8-oxodG (Halliwell and Gutteridge, 2007; Wells et al., 2009b). Consequences of 8-oxodG accumulation in dividing cells may lead to transversion mutations which can affect the expression and activity of proteins required for normal development and function, and if not repaired, may affect gene transcription, DNA replication and cell division, which may lead to cancer and/or embryopathies (Evans et al., 2004; Neeley and Essigmann, 2006; Wells et al.,

2009b). Oxidatively damaged DNA may also directly initiate embryopathies via non-mutagenic mechanisms, possibly including altered gene transcription (Wells et al., 2010). Hydroxyl radicals alternatively can attack nuclear proteins, which results in the formation of protein radicals which can then bind to DNA to form DNA-protein cross links that can interfere with gene transcription,

549 replication and repair (Halliwell and Gutteridge, 2007). To ensure cellular viability in the presence of high concentrations of ROS, DNA repair mechanisms exist to ensure replicative fidelity and normal gene expression, including base excision repair (BER), nucleotide excision repair (NER) and mismatch repair (Christmann et al., 2003). In addition to its mutagenic activity, 8-oxodG is a developmentally pathogenic lesion. This was demonstrated in knockout mice lacking oxoguanine glycosylase 1 (OGG1), a component of the BER pathway that repairs

8-oxodG. Ogg1 knockout mice exposed in utero to methamphetamine and tested postnatally for motor coordination deficits performed significantly worse than wild-type controls, suggesting 8- oxodG can contribute to postnatal neurodevelopmental deficits (Wong et al., 2008).

Lipid Peroxidation

Polyunsaturated fats within the cellular membrane are common targets of oxidative damage due to the presence of carbon-carbon double bonds (Halliwell and Gutteridge, 2007). Three steps are involved in lipid peroxidation: initiation, propagation and termination. (1) Initiation: lipid peroxidation begins either by addition of a hydroxyl group across a double bond, forcing the electrons to move onto the adjacent carbon forming a carbanion, or more commonly, by hydrogen abstraction creating a lipid radical (Gutteridge, 1984; Halliwell and Gutteridge, 2007).

Transition metals such as iron and copper can participate in electron exchange with oxygen to form the hydroxyl radical. (2) Propagation: lipid radicals can stabilize by rearrangement to a conjugated diene, or can react with molecular oxygen to give rise to a peroxyl radical (ROO•).

This peroxyl radical can abstract a hydrogen atom from an adjacent fatty acid side chain, forming new carbon centered radicals that can react with oxygen to form new peroxyl radicals. The peroxyl radical can combine with the abstracted hydrogen atom to form a lipid hydroperoxide

(ROOH). Cyclic peroxides can form when a peroxyl radical attacks a double bond within the

550 same fatty acid residue. (3) Termination: the chain reaction terminates when two lipid peroxyl radicals combine to produce a non-radical species, or when a radical is halted by binding to antioxidants such as α-tocopherol.

Lipid peroxidation can produce DNA-damaging aldehydes such as malondialdehyde (MDA), 4- hydroxynonenal (4-HNE), and F-isoprostanes. The decomposition of lipid peroxides by heating or reaction with metal ions creates a wide variety of cytotoxic products, which can produce more radicals that can initiate further lipid peroxidation.

MDA is produced either from the peroxidation of polyunsaturated fatty acids (PUFAs) with more than two double bonds, or enzymatically during the metabolism of eicosanoids. At physiological pH, most MDA exists as the enolate ion which has low reactivity towards amino groups in proteins (Halliwell and Gutteridge, 2007). At a lower pH, MDA exists as the undissociated enol form in equilibrium with its keto form, and exhibits a higher reactivity towards proteins, able to attack residues resulting in intra- and inter-molecular cross-links

(Esterbauer et al., 1991). MDA can react with DNA, more specifically guanine bases, to create G to T transversions, A to G transitions, C to T transitions, frameshifts and deletions, with potentially mutagenic consequences (Marnett, 2000). MDA is metabolized to malonic semialdehyde by aldehyde dehydrogenase, and this product is decarboxylated to acetaldehyde, and is finally metabolized to acetic acid again by aldehyde dehydrogenase (Halliwell and

Gutteridge, 2007). The thiobarbituric acid (TBA) assay is a colorimetric assay commonly used to detect MDA, whereby a sample is heated with TBA in acid and a pink colour develops, reflecting the formation of thiobarbituric acid-reactive substances (TBARS). Although technically easy to perform, the specificity of this assay has been questioned (Gutteridge and

551

Quinlan, 1983). It has been suggested that much of the MDA measured may actually generated during the assay workup, since the amount of free MDA in most lipid peroxidizing systems is too low to be detected by the assay. As much as 98% of the MDA detected was generated during the acid heating step of the assay by the decomposition of lipids, which can result in a wide variety of radical-generating toxic products. This can be circumvented by measuring MDA directly via high-performance liquid chromatography (HPLC) or gas chromatography-mass spectrometry (GC-MS). Another source of uncertainty in the TBARs assay is that several compounds can react with TBA to form chromogens that absorb at 532 nm. Lastly, assaying human body fluids can detect artifactual MDA produced enzymatically during eicosanoid synthesis (Shimizu et al., 1981).

HNE is formed during the oxidation of n-6 polyunsaturated fatty acids (PUFAs), which are fatty acids that contain a double bond at the carbon-6 position, such as linoleic acid and arachidonic acid (Esterbauer et al., 1991; Spiteller, 1998). Basal cellular levels of HNE in healthy tissues are approximately 1 μM or lower; however, under conditions of oxidative stress, HNE concentrations can rise to between 2-20 μM which is cytotoxic, leading to inhibition of DNA and protein synthesis, cellular proliferation and nucleotide excision repair (Feng et al., 2004; Parola et al., 1999). HNE can react rapidly with thiol and amino groups on proteins (i.e. histidine, lysine) and amino groups on DNA bases, with guanine being the preferred target (Halliwell and

Gutteridge, 2007). HNE reacts with DNA to form an etheno-adduct by adding an NH2 group to the double bond of the aldehyde to yield 1,N2-propano-21-deoxyguanosine adducts (Choudhury et al., 2004; Schaur, 2003) (fig. 22). As mentioned previously, HNE and MDA, the end products of lipid peroxidation, are DNA-damaging aldehydes. As such, they may facilitate cancer development in several ways such as: (1) mutagenic adduct formation with DNA bases, (2)

552 formation of subsequent ROS during the peroxidation process which may directly oxidize DNA leading to mutagenesis, or (3) oxidation of DNA repair proteins resulting in decreased replication fidelity with a subsequent increase in the incidence of mutations (Halliwell and Gutteridge,

2007).

2004). IPs can form isoketals that quickly react with amino groups to form adducts with lysine residues on membrane proteins, resulting in protein cross-linking, ultimately damaging the cellular membrane (Poliakov et al., 2004).

4.2. Oxidative stress from methanol exposure

4.2.1. Evidence for MeOH-initiated ROS formation

Studies have demonstrated enhanced lipid peroxidation in brain, liver, bile, erythrocytes, urine as well as lymphoid organs of rats intoxicated both acutely and chronically with MeOH

(Dobrzynska et al., 2000; Parthasarathy et al., 2006b; Parthasarathy et al., 2006c;

Skrzydlewska and Farbiszewski, 1997a; Skrzydlewska and Farbiszewski, 1998) (table 10).

Following an acute exposure of rats to MeOH, one study reported an increase in enzymatic and non-enzymatic antioxidants such as catalase, superoxide dismutase, glutathione and vitamin C, which may be a cellular response to counteract the enhanced ROS formation. In contrast, following a longer duration of MeOH exposure, there was a decrease in enzymatic and non-

553 enzymatic antioxidants, which could reflect cellular membrane and macromolecular damage due to excessive oxidative stress.

4.2.2. Mechanism of MeOH-initiated ROS formation

The mechanisms by which MeOH and/or its metabolites enhance ROS formation have yet to be determined. Numerous studies have implicated free radical-initiated, ROS-mediated involvement in the mechanism of toxicity including: (1) direct detection of a MeOH radical by electron spin resonance spectrometry, and oxidative protein damage in MeOH intoxicated rats (Skrzydlewska et al., 2000); (2) MeOH-derived adducts to the free radical spin trapping agent phenylbutylnitrone (PBN) detected in bile and urine of PBN-pretreated, MeOH-exposed rats

(Kadiiska and Mason, 2000); and, (3) MeOH embryopathies in rat whole embryo culture are enhanced by the depletion of glutathione (GSH) (Harris et al., 2004). Additionally, MeOH- initiated oxidative stress, as evidenced by the production of the lipid peroxidation product MDA, along with increases in antioxidative enzyme activities, was observed in the lymphoid organs of adult rats (Parthasarathy et al., 2006b). Although the enzyme(s) catalyzing this reaction remain to be determined, PHSs do not appear to contribute, as MeOH embryopathies are not blocked by pretreatment with the PHS inhibitor acetylsalicylic acid (ASA) (Miller and Wells, unpublished).

CYP2E1 expression is negligible during the embryonic period, and low compared to adult activity during the fetal period, particularly in rodents (Hines, 2008; Juchau et al., 1998; Vieira et al., 1996), so CYP2E1-mediated superoxide formation would appear to be an unlikely mechanism for embryonic ROS formation, at least in rodents. Preliminary studies suggest that

MeOH and/or its metabolites may activate and/or induce the expression of embryonic NOXs that produce embryopathic ROS, as MeOH embryopathies are reduced by pretreatment with the

554

NOX inhibitor diphenyleneiodonium (DPI) (Miller and Wells, unpublished). This would be consistent with the NOX-dependent ROS mechanism previously reported for the related alcohol

EtOH (Dong et al., 2010). Additional potential mechanisms, such as interference with the mitochondrial electron transport chain, have yet to be investigated.

4.3. Teratogenicity of methanol and comparisons to ethanol

4.3.1. Genetic modulation of catalase

Genetically modified mice with altered catalase activity were used to evaluate the role of catalase in embryo culture, which removes potentially confounding maternal factors, and in vivo. Mutant acatalasemic mice (aCat) with reduced catalase activity were compared to their C3H wild-type

(WT) controls, and transgenic mice expressing human catalase (hCat) with enhanced catalase activity were compared to their C57BL/6J WT controls.

In embryo culture, MeOH was embryopathic in both the WT C57BL/6J and C3H strains, although the range and severity of embryopathies were greater in the C57BL/6J strain (Miller and Wells, 2011). hCat embryos with enhanced catalase activity were protected from MeOH- initiated decreases in anterior neuropore closure, turning and somite development. In contrast,

MeOH-exposed catalase-deficient aCat embryos exhibited reduced anterior neuropore closure and head length, which were not observed in MeOH-exposed WT controls (figs. 23, 24). The respectively contrasting protection and increased embryopathies in hCat and aCat embryos compared to their WT controls suggest that ROS contribute to the mechanism of MeOH teratogenicity, and that embryonic catalase, despite constituting only about 2-5% of maternal activity, plays a developmentally protective role. This interpretation is consistent with similar

555 embryo culture and in vivo studies of the ROS-initiating teratogen phenytoin (Abramov and

Wells, 2011a). The results also suggest that the alternative peroxidative activity of catalase in converting MeOH to formaldehyde, at least in embryonic tissues, does not play a measurable role in modulating embryopathic risk.

In the same genetically modified mice, the effects of altered catalase activity in the more complicated in vivo system were more difficult to interpret. The C57BL/6J WT controls for the hCat dams treated with MeOH (total dose of 4 g/kg ip on GD 8) had offspring with an increased incidence of ophthalmic abnormalities and cleft palates; however, these incidences of anomalies were not significantly different from those seen in hCat mice, although there was a trend for protection (Siu et al., 2013b) (fig. 25). On the other hand, both the aCat mice and their WT C3H controls were relatively resistant to MeOH teratogenicity, the first strain of mouse to be reported as such, precluding an evaluation of the effect of catalase deficiency upon MeOH teratogenesis.

The resistance of the aCat and C3H WT controls could not be explained by pharmacokinetic determinants, since the pharmacokinetic profiles for MeOH and FA in the aCat, hCat and WT mice were similar, with only the hCat mice showing a small increase in MeOH peak plasma concentration and AUC (Sweeting et al., 2010). The resistance of the C3H mice in vivo is consistent with the lower susceptibility of this strain to MeOH embryopathies in embryo culture compared to C57BL/6J embryos. In the hCat mice, it is possible that the 1.5-fold increase in embryonic catalase activity was insufficient to protect against MeOH-initiated ROS formation, although the same hCat embryos were protected in embryo culture. In light of the protective role for embryonic catalase evident in embryo culture using these mice, the in vivo resistance of the aCat strain to MeOH teratogenesis and the non-significant protection in hCat mice may be due to confounding maternal factors other than hepatic metabolism, such as placental transporters for

MeOH and/or FA, or protective maternal hormones.

556

In the catalase-modified mice, embryos were exposed in culture to 4 mg/ml of EtOH (87 mM) or

4 mg/ml of MeOH (125 mM). Although these concentrations are not equimolar, it can be observed that a smaller percentage of MeOH-exposed WT embryos closed their neuropore or turned, despite the relatively higher molar concentration of MeOH, while both EtOH- and

MeOH-exposed WT embryos developed significantly fewer somites than saline-exposed controls

(Miller and Wells, 2011). These results suggest that the embryopathic molar potency of MeOH is less than that for EtOH, and substantially so for some developmental outcomes. Also interesting was that regardless of the molar concentration, enhanced catalase significantly protected embryos from several parameters that were affected by EtOH or MeOH, including anterior neuropore closure, somite development and turning (figs. 23, 24). This suggests that embryonic catalase activity may be a potential risk factor for both EtOH and MeOH, and enhanced embryonic catalase is protective.

4.3.2. Free radical spin trapping agent

In embryo culture, MeOH-exposed CD-1 mouse embryos pretreated with the free radical spin trapping agent phenylbutylnitrone (PBN) exhibited reduced MeOH embryopathies, evidenced by increased anterior neuropore closure, turning and somite development, suggesting ROS involvement in the mechanism of MeOH embryopathies (Miller and Wells, unpublished). This interpretation is consistent with the protective effect of PBN in reducing embryonic/fetal DNA oxidation and the embryopathic effects of other ROS-initiating teratogens in embryo culture and/or in vivo (Lee et al., 2011; Wells et al., 2009b).

4.4. Carcinogenic potential

4.4.1. Oxidatively damaged DNA

557

In vitro and in vivo genotoxicity tests indicate MeOH is not mutagenic (IPCS, 1997), but carcinogenic potential has been claimed in one controversial long-term rodent cancer bioassay that has not been replicated (Soffritti et al., 2007). To determine if MeOH could indirectly damage DNA via ROS-mediated mechanisms, we treated male CD-1 mice, NZW rabbits and cynomolgous monkeys with MeOH (2.0 g/kg i.p.) and 6 hr later assessed oxidative damage to

DNA, measured as 8-oxo-2’-deoxyguanine (8-oxodG) formation by HPLC with electrochemical detection. We found no MeOH-dependent increases in 8-oxodG in bone marrow (fig. 26), spleen, lung, liver or kidney of any species (McCallum et al., 2011a; McCallum et al., 2011b).

Chronic treatment of CD-1 mice with MeOH (2.0 g/kg i.p.) daily for 15 days also did not increase 8-oxodG levels in these organs. To rule out the possibility that the lack of effect of

MeOH exposure on accumulation of oxidatively damaged DNA was due to masking by rapid repair of induced lesions, we performed further studies in DNA repair-deficient oxoguanine glycosylase 1 (Ogg1) knockout (KO) mice. The results were corroborated in untreated Ogg1

KO mice, which accumulated 8-oxodG in bone marrow (fig. 27), spleen, lung, kidney and liver with age, but exhibited no increase following MeOH treatment, despite a 2-fold increase in renal

8-oxodG in Ogg1 KO mice following treatment with a ROS-initiating positive control, the renal carcinogen potassium bromate (KBrO3; 100 mg/kg i.p.) (McCallum et al., 2011a; McCallum et al., 2011b). These observations suggest that MeOH exposure does not promote the accumulation of oxidatively damaged DNA in lung, kidney or liver, and that environmental exposure to MeOH is unlikely to initiate carcinogenesis in these organs by ROS-initiated DNA oxidation.

4.4.2. Hydroxynonenal-histidine protein adducts

558

Based on our negative findings for MeOH-dependent ROS-mediated DNA damage and conflicting reports in the literature of MeOH-induced lipid peroxidation (Parthasarathy et al.,

2006b; Parthasarathy et al., 2006c; Skrzydlewska et al., 2000) determined by the controversial

TBARS method (see section 4.1.3. above), we conducted an additional measure of free radical- initiated macromolecular damage by measuring the HNE-histidine (HNE-His) protein adducts under conditions minimizing the influence of adventitious iron in cellular homogenates (see previous section 4.1.3 for carcinogenic relevance). We detected modest increases in the levels of the HNE-His adduct only in mouse bone marrow (1.4-fold) and in rabbit spleen (1.5-fold) with no increases observed in primate bone marrow or spleen or mouse spleen (McCallum et al.,

2011a) (fig. 28). For comparison, acute exposure to 20 mg/kg i.p. of the redox cycling agent doxorubicin increased the levels of HNE-His protein adducts by 5.3-fold in mouse heart homogenates (Mukhopadhyay et al., 2010). In contrast, the absence of MeOH-initiated HNE-His protein adducts in mouse spleen, and more importantly in either primate bone marrow or spleen, or in rabbit bone marrow, indicates this oxidative effect of methanol is limited and unlikely to occur to a biologically significant degree in adult humans. In general, these observations suggest that a limit dose of MeOH (2.0 g/kg i.p.) can produce modest oxidative stress, but the potential for macromolecular damage is highly specific with respect to macromolecular target, species and tissue. Moreover, perhaps due to a combination of subcellular localization, packaging in nucleosomes, and repair mechanisms, DNA is more resistant than lipids to MeOH-initiated oxidative damage in adults.

5. CONCLUSIONS

559

Our results suggest that:

• As the metabolism of MeOH by rabbits is more similar than mice to that in humans, the rabbit might be a more accurate model than the mouse, and perhaps the rat, for predicting the human risk for MeOH developmental toxicity.

• Given that rabbits, which more closely reflect human MeOH metabolism, and at least one strain of mice and one strain of rat, are resistant to MeOH teratogenesis, it is questionable whether the human risk for MeOH developmental toxicity can be accurately assessed in sensitive rodent models.

• The respectively enhanced and reduced susceptibility of aCat and hCat mice to the embryopathic effects of MeOH in embryo culture, albeit not in vivo, suggest that ROS may contribute to the underlying mechanism of MeOH teratogenicity in rodents. These results also suggest that embryonic catalase plays an important protective role via its antioxidative activity, as distinct from its peroxidative, MeOH-metabolizing role in maternal liver or the embryo. The absence of modulation by altered catalase expression in vivo merits further investigation.

• The reduced embryopathies observed in MeOH-exposed mouse embryos pretreated with a free radical spin trapping agent in whole embryo culture further support a role of oxidative stress in the mechanism of MeOH developmental toxicity.

• MeOH appears to be less embryopathic than EtOH on a molar basis, based upon mouse embryo culture results, although it is difficult to extrapolate this comparison to in vivo studies in susceptible rodent strains where conditions are highly variable.

560

• In regard to the mechanism by which MeOH enhances embryopathic ROS formation, our preliminary results showing a reduction in MeOH embryopathies in culture by pretreatment with a NOX inhibitor suggest that MeOH may activate/induce embryonic NOXs that form ROS intracellularly. Conversely, prostaglandin H synthase-mediated bioactivation of MeOH to a free radical intermediate is unlikely to play a role, as preliminary studies using PHS inhibition did not alter MeOH-initiated embryopathies in culture.

• MeOH did not increase the level of oxidatively damaged DNA in any tissue in any species, despite enhanced baseline DNA oxidation observed in DNA repair-deficient ogg1 knockout mice, and enhanced DNA oxidation observed with a ROS-initiating positive control. Similarly, levels of HNE-His protein adducts, reflecting free radical-mediated production of the potentially carcinogenic lipid peroxidation product HNE, were not enhanced by MeOH in primate bone marrow or spleen, or in rabbit bone marrow or mouse spleen, although modest increases were observed in rabbit spleen and mouse bone marrow. These results suggest that it is unlikely that human environmental exposure to MeOH would cause cancer via a mechanism involving oxidatively damaged DNA.

561

Table 1: Pharmacokinetic data for methanol (MeOH) metabolism in rodents, humans, and non-human primates.

Abbreviations: ROA, route of administration; AUC, area under the plasma concentration-time curve; i.p., intraperitoneal; i.v., intravenous; MeOH, methanol; FA, formic acid; N/D, not determined.

562

Table 2: Comparison of enzyme kinetic parameters for alcohol dehydrogenase (ADH1) with MeOH as a substrate in mice, rats and humans.

Abbreviations: Vmax, maximum rate of enzyme reaction; Km, substrate concentration at ½ Vmax.

563

Table 3: Changes in ADH1 activity throughout development in mice, rats and humans.

564

Table 4: Changes in peroxidative catalase activity, using MeOH as a substrate, throughout development in mice and rats.

565

Table 5: Changes in antioxidative catalase activity throughout development in rodents, humans and primates.

Abbreviations: GD, gestational day.

566

Table 6: Species differences in hepatic levels of tetrahydrofolate (THF).

Abbreviations: THF, tetrahydrofolate.

567

Table 7. Species- and strain-specific malformations following in utero MeOH exposure in mice, rats and non-human primates.

568

569

Abbreviations: ROA, route of administration; GD, gestational day; If no other indication, plug = GD 1; NOAEL, no observable adverse effect level; NTD, neural tube defect; i.p., intraperitoneal; EtOH, ethanol; NS, non-significant; WN, well-nourished; MN, malnourished.

570

Table 8. Species- and strain-specific neurodevelopmental effects of in utero MeOH exposure in rats and non-human primates.

Abbreviations: ROA, route of administration; GD, gestational day (if no other indication, plug = GD 1); PND, postnatal day

571

Table 9: Summary of methanol carcinogenicity studies in animals.

Abbreviations: EPA, Environmental Protection Agency; F, female; M, male; N, number of animals, NEDO, New Energy Development Organization (Japan); NTP, U.S. National Toxicological Program a US EPA. 2009. "Toxicological Review of Methanol in Support of Summary Information on the Integrated Risk Information System (IRIS) (External Peer Review Draft)." EPA/635/R-09/013.

572

Table 10: Evidence for MeOH-initiated oxidative stress

Abbreviations: TBARS, thiobarbituric reactive acid substances; SOD, superoxide dismutase; GSH-Px, glutathione peroxidase; GSSG-R, glutathione reductase; MDA, malondialdehyde; TAS, total antioxidant score; CAT, catalase; U-83836E, antioxidant trolox derivative, N/A, not applicable, i.p., intraperitoneal; i.g., intragastric. a US EPA. 2009. "Toxicological Review of Methanol in Support of Summary Information on the Integrated Risk Information System (IRIS) (External Peer Review Draft)." EPA/635/R-09/013.

573

Table 11: Strain-specific malformations following in utero EtOH exposure in mice.

Abbreviations: GD, gestational day; BAC, blood alcohol concentration; i.p. intraperitoneal; N/D, not determined, MRI, magnetic resonance imaging.

574

Figure 1. Species differences in the enzymes catalyzing the metabolism of methanol (MeOH) to formaldehyde and formic acid (FA) in mice and primates, including humans. In addition to its role in MeOH and FA metabolism in mice, catalase is used by all species in the detoxification of reactive oxygen species (ROS). (From Sweeting et al. (2010). Toxicol. Appl. Pharmacol. 247(1): 28-35)

575

Figure 2. Role of alcohol dehydrogenase (ADH1) in the metabolism of MeOH. MeOH (CH3OH) is oxidized in humans and primates + by ADH1 into formaldehyde (CH2O) using NAD as a cofactor.

576

Figure 3. Role of catalase in the metabolism of MeOH and FA: Peroxidative activity. Using H2O2 as a cofactor, rodent catalase peroxidatively oxidizes (a) MeOH (CH3OH) into formaldehyde (CH2O), and (b) FA (HCOOH) into carbon dioxide and water.

577

Figure 4. Role of catalase in the metabolism of reactive oxygen species (ROS): Antioxidative activity. H2O2 is scavenged by the antioxidative function of catalase in all species, forming the non-toxic metabolites of water and oxygen.

578

Figure 5. Role of cytochrome P450 2E1 (CYP2E1) in the metabolism of MeOH. In the presence of NADPH and molecular oxygen,

CYP2E1 oxidizes MeOH into formaldehyde (CH2O).

579

Figure 6. Role of formaldehyde dehydrogenase (ADH3) in the metabolism of formaldehyde. ADH3 conjugates glutathione (GSH) to formaldehyde forming S- formylglutathione, which is subsequently metabolized into FA through the removal of GSH by thiolase.

580

Figure 7. Role of folate in the metabolism of FA. In the presence of tetrahydrofolate (THF), formate is metabolized into 10-formyl-

THF by 10-formyl-THF-synthetase. 10-Formyl-THF is subsequently converted into CO2 and H2O by 10-formyl-THF dehydrogenase.

581

Figure 8. MeOH and FA disposition in male mice, rabbits and primates over a 6 hr time period, following a single dose of 2 g/kg MeOH, plotted on a semi-log scale. MeOH was administered through an ip injection as a 20 % solution in sterile saline, and plasma samples were analyzed by GC (mice and rabbits) or by spectrometry (primates) for MeOH or FA concentrations. Each time point represents the mean of 3-6 mice, 3 rabbits or 3 primates for treated groups. Data points with open symbols represent samples with a MeOH or FA concentration below the level of detection and were arbitrarily assigned a value of 0.1 mmol/L. (From Sweeting et al. (2010). Toxicol. Appl. Pharmacol. 247: 28-35).

582

Figure 9. MeOH clearance (Cl) and FA areas under the plasma concentration–time curve (AUC) in male mice and rabbits following a single dose of 0.5 or 2 g/kg MeOH. Cl and AUC values were calculated from the MeOH and FA pharmacokinetics curves of animals dosed with 0.5 or 2 g/kg MeOH (20% solution in sterile saline) through an ip injection. Saline groups not shown (all not detectable). Each time point represents the mean of 3–6 mice or 3 rabbits for treated groups. *Indicates measures for rabbit MeOH Cl and FA AUC that were different from the respective mouse values (p < 0.05). (From Sweeting et al. (2010). Toxicol. Appl. Pharmacol. 247(1): 28-35).

583

Figure 10. Six-hour MeOH and FA AUC in male mice, rabbits and primates following a single dose of 2 g/kg MeOH. A total of 3-6 mice, 3 rabbits and 3 primates were sampled at each time point for treated groups. * Indicates that the 6-hour MeOH AUC and FA AUC were different from the respective mouse values. † Indicates that the 6-hour FA AUC in primates was different than that in rabbits (p < 0.05). (From Sweeting et al. (2010). Toxicol. Appl. Pharmacol. 247: 28-35).

584

Figure 11. MeOH and FA pharmacokinetics in male C57BL/6J, C3H and CD-1 mice, and New Zealand white (NZW) rabbits. C57BL/6J and C3H mice were sampled over a 12 h time period, CD-1 mice over 24 h and NZW rabbits over 48 h, following a single intraperitoneal (i.p.) dose of 2 g/kg MeOH. MeOH was administered as a 20% solution in sterile saline, and plasma samples were analyzed by GC for MeOH or FA concentrations. A total of 3–6 mice and 3 rabbits were sampled at each time point. Data points with open symbols represent samples with a MeOH or FA concentration below the level of detection and were arbitrarily assigned a value of 0.1mM. (From Sweeting et al. (2011). Reprod Toxicol. 31: 50-58).

585

Figure 12. Blood MeOH levels measured from human cases of acute toxicity. Blood MeOH levels (mmol/L and equivalent mg/L) with corresponding symptoms of toxicity. (Modified from WHO. (1997). Environmental Health Criteria: http://www.inchem.org/documents/ehc/ehc/ehc196.htm#SectionNumber:1.3)

586

Figure 13. MeOH-initiated birth defects in NZW rabbits. Pregnant does were treated ip with 2 doses of 2 g/kg MeOH (20% solution in saline) or its saline vehicle, with an 8 h interval, on GD 7 or 8. Does were euthanized on GD 29 for assessment of fetuses. Fetal outcomes from GD 7 and 8 MeOH exposure were not different (p > 0.05), and these data were combined. Fetal outcomes from GD 7 and 8 MeOH exposure were not different (p > 0.05), and these data were combined (x,y = number of litters, number of fetuses). (From Sweeting et al. (2011). Reprod. Toxicol. 31: 50-58).

587

Figure 14. MeOH-initiated birth defects in C57BL/6J and C3H mice. Pregnant dams were treated ip with 2 doses of 2 g/kg MeOH (20% solution in saline) or its saline vehicle, with a 4 h interval on 8. Dams were euthanized on GD19 for assessment of fetuses. Each data point represents one litter, with the mean designated by a horizontal bar. The number in parentheses represents the number of litters assessed.  indicates a difference from the respective saline control (p < 0.01) for ophthalmic defects, p < 0.05 for cleft palates). (From Sweeting et al. (2011). Reprod. Toxicol. 31: 50-58).

588

Figure 15. Comparison of outcomes in mouse embryos exposed to MeOH or ethanol (EtOH) for 24 hours in culture. EtOH studies are represented by dashed lines and MeOH studies are represented by solid lines. For unpublished studies by Miller and Wells, each group consisted of 11-40 embryos

589

RAT EMBRYO CULTURE

Crown-rump length (mm) length Crown-rump 0 100 200 300 400 500 DRUG CONCENTRATION (mM)

2.5 Head Length 2.0 Sprague-Dawley Rat: Andrews et al, 1993 (MeOH) Outbred rats: Brown et al 1979 (EtOH) 1.5 Rat: Lee et al 2005 (EtOH)

1.0 MeOH 0.5 Head length (mm) 0.0 0 100 200 300 400 500 DRUG CONCENTRATION (mM)

Somites Developed (n) Developed Somites 0 0 100 200 300 400 500

400 Protein Content Sprague-Dawley Rat: Andrews et al, 1993 (MeOH) Outbred rats: Brown et al 1979 (EtOH) 300 MeOH Rat: Snyder et al. 1992 (EtOH) Sprague-Dawley Rat: Wynter et al 1983 (EtOH) 200

100 Sprague-Dawley

0 0 100 200 300 Protein content (ug/embryo) content Protein DRUG CONCENTRATION (mM)

Figure 16. Comparison of outcomes in rat embryos exposed to ethanol (EtOH) or MeOH for 48 hours in culture. MeOH studies are represented by dashed lines and EtOH studies are represented by solid lines.

590

MOUSE VS. RAT EMBRYO CULTURE

DRUG CONCENTRATION (mM)

Somites developed (n) Somites 15 0 200 400 600 800 1000

DRUG CONCENTRATION (mM)

Figure 17. Comparison of outcomes in mouse and rat embryos exposed to methanol (MeOH) for 24 hours in culture. The rat study is indicated by a dashed line, and the remaining studies employed mice. For unpublished studies by Miller and Wells, each group consisted of 11-40 embryos

591

Figure 18. Biochemical pathways for endogenous and xenobiotic-enhanced formation and detoxification of reactive oxygen species (ROS), and repair of oxidatively damaged cellular ●- macromolecules. ROS include superoxide (O2 ), hydrogen peroxide (H2O2) and hydroxyl radicals (HO●). Teratogenesis is postulated to result from embryonic macromolecular damage and/or ROS-mediated alterations in embryonic signal transduction. If embryonic ROS formation exceeds the proximal capacity for ROS detoxification and/or repair of cellular macromolecules, this imbalance can result in enhanced teratogenesis, even at a therapeutic drug concentration or generally “safe” exposure level for an environmental chemical. Abbreviations: ATM, ataxia telangiectasia mutated protein; CSB, Cockayne syndrome B protein; CYPs, cytochromes P450; Fe, iron; G-6-P, glucose-6-phosphate; GSH, glutathione; GSSG, glutathione disulfide; LPOs, lipoxygenases; NADP+, nicotinamide adenine dinucleotide phosphate; OGG1, oxoguanine glycosylase 1; PHSs, prostaglandin H synthases, SOD, superoxide dismutase. (Modified from Wells et al. (2009). Toxicol. Sci. 108: 4-18).

592

Figure 19. NADPH oxidase (NOX)-mediated ROS formation. NOX enzymes catalyze the formation of superoxide and subsequently hydrogen peroxide (H2O2) through reduction of oxygen using NADPH as a cofactor. The first group consisting of gp91phox (NOX2), NOX3 and NOX4 share similar domain structures and size. Each contains a transmembrane domain composed of 6 alpha helices. A flavin domain attached to the carboxy-terminal can bind co- enzymes flavin adenine dinucleotide (FAD) and NADPH. H2O2 produced by this enzyme can be reduced by an independent peroxidase, such as myeloperoxidase, while oxidizing an extracellular substrate. NOX5 shares the same structural features as gp91phox along with a calcium binding domain which allows for calcium-mediated enzyme regulation. DUOX1 and 2 share the same structural features as NOX5 along with an additional peroxidase domain anchored to the enzyme by an additional transmembrane alpha-helix, capable of reducing hydrogen peroxide while oxidizing extracellular substrates. (Based upon Lambeth et al. (2004). Nat. Rev. Imunol. 4: 181-189).

593

Figure 20. ROS-mediated signal transduction. (Reproduced from Jiang et al. (2011). Pharmacol. Rev. 63: 218-242).

594

Figure 21. Formation of a 1,N-2-propane adduct of histidine following reaction of 4- hydroxynonenol (HNE) with histidine. HNE reacts with histidine via Michael addition to form an enol adduct, which subsequently tautomerizes and undergoes a nucleophilic addition to form a stable adduct.

595

Figure 22. Formation of a 1,N-2-propane adduct of guanine following reaction of 4- hydroxynonenol (HNE) with guanine. HNE reacts with guanine via Michael addition to form an enol adduct, which subsequently tautomerizes and undergoes a nucleophilic addition to form a stable adduct.

596

CATALASE DEFICIENCY

100 40 Anterior Neuropore Closure Somites Developed 80 (16) 30 (11) ‡‡ 60 (13) (11) (6) (5) ‡ 20 (2) 40 *‡ 10

20 SD) mean + (n,

PERCENTAGE (%) (11) 0 0 S M S M S M S M 4 100 Turning Crown-rump Length 80 (16) 3 (11) (6) b 60 (5) (13) (11) 2 (2) 40 (11) 20 1 (mm, mean + SD) + mean (mm, PERCENTAGE (%) 0 0 S M S M S M S M

5 200 Yolk Sac Diameter Heart Rate 4 150 (11) †† (13) (11) 3 (16) (13) (16) (11) (11)* 100 2 50 1 (mm, mean + SD) mean + (mm, 0 0 S M S M (beats/min, mean SD) + S M S M 2.0 WT aCat Head Length 1.5 (7) (13) * (7) ‡‡‡ 1.0 (11)

0.5 (mm, mean + SD) + mean (mm, 0.0 S M S M WT aCat

Figure 23. Effect of MeOH on embryopathies in mice with a catalase deficiency in embryo culture. Double daggers indicate difference from saline treated embryos of same genotype (‡ p < 0.05, ‡‡ p < 0.01, ‡‡‡ p < 0.0001). Asterisk indicates difference between genotypes for the same treatment group (* p < 0.05). Letter b indicates p=0.05 using a t-test

597

ENHANCED HUMAN CATALASE

EXPRESSION

40 Somites Developed Anterior Neuropore Closure 100 (10) (10) (13) 30 (9)* (9) ‡ (10) (3) *‡ 20 50 (14) 10 (n, mean(n, SD) + PERCENTAGE (%) 0 0 S M S M S M S M

4 150 Turning Crown-rump Length 3 ‡ (10) (3) 100 (10) a (9) (9) (13) 2 (10) 50 ‡ (6) (14) 1 (mm, mean + SD) + mean (mm, PERCENTAGE (%) PERCENTAGE 0 0 S M S M S M S M

Heart Rate Yolk Sac Diameter 150 3 ‡‡ ‡‡‡ (10) (14) (13) (14) (10) (13) 100 2 (10) (10)

1 50 (mm, mean + SD) + mean (mm, 0 0 S M S M SD) + mean (beats/min, S M S M WT hCat WT hCat

Figure 24. Effect of MeOH on embryopathies in mice overexpressing human catalase. Double daggers indicate difference from saline treated embryos of the same genotype (‡ p < 0.05, ‡‡ p < 0.01, ‡‡‡ p < 0.0001). Asterisk and letter indicates difference between genotypes for the same treatment group (* p< 0.05; a, p = 0.05). (From Miller and Wells. (2011). Toxicol. Appl. Pharmacol. 252: 55-61).

598

Figure 25. Effect of MeOH on embryopathies in mice with a genetic modulation of catalase in vivo. Pregnant dams were treated on gestational day (GD) 8 with 2 doses of 2 g/kg MeOH i.p. or its saline vehicle at 4 hr intervals, and assessed on GD 19. Each symbol represents one litter, with the number of litters shown in parentheses, and the mean by a horizontal bar. * indicates a difference from the respective saline treatment (p < 0.01 for ophthalmic anomalies and p < 0.05 for cleft palate anomalies). (From Siu and Wells, unpublished)

599

Figure 26. MeOH-initiated 8-oxo-2’deoxyguanine (8-oxodG) formation in the lung, liver, kidney, bone marrow and spleen of male CD-1 mice, NZW rabbits and cynomolgous monkeys at 6 h. Animals were treated intraperitoneally with a single dose of 2.0 g/kg bw MeOH (20% [w/v] in sterile saline) or saline vehicle (controls) and sacrificed at 6 or 24 h post- injection. Genomic DNA was isolated and analyzed for oxidatively damaged DNA damage reflected by the formation of 8-oxodG in the a) lung, b) liver, c) kidney, d) bone marrow and e) spleen of each species. Values are mean + SE; N = 4 for mice and N = 3 for rabbit and monkeys, respectively. (Modified from McCallum et al. (2011). Mol. Carcinogen. 50: 163-172 and McCallum et al. (2011). Toxicol. Appl. Pharmacol. 250: 147-153).

600

Figure 27. MeOH exposure does not increase 8-oxodG levels in bone marrow and spleen of Ogg1 (+/+) or Ogg1 (−/−) mice. Mice were given a single i.p. dose of 2.0 g/kg bw MeOH (20% [w/v] in sterile saline) or saline vehicle control, and sacrificed at 6 and 24 h post-injection. Genomic DNA was isolated and analyzed for oxidative DNA damage reflected by the formation of 8-oxodG. Values are means + SE; N = 4. Symbol * denotes a difference in Ogg1 (−/−) sample compared to the respective group in Ogg1 (+/+) mice (p < 0.05). (Modified from McCallum et al. (2011). Mol. Carcinogen. 50: 163-172 and McCallum et al. (2011). Toxicol. Appl. Pharmacol. 250: 147-153).

601

Figure 28. HNE-His protein adducts in bone marrow and spleen from male CD-1 mice, New Zealand white rabbits or cynomolgus monkeys following acute exposure to MeOH (2.0 g/kg ip). Animals were given a single i.p. dose of 2.0 g/kg bw MeOH (20% [w/v] in sterile saline) or saline vehicle (control) and sacrificed at 6 or 24 h post-injection. Values are means + SE; N = 4 for mice and N = 3 for rabbits and monkeys. Symbol * denotes a difference from saline control (p < 0.05). (Modified from McCallum et al. (2011). Mol. Carcinogen. 50: 163- 172).

602

11.3 In vivo teratogenesis pictures

603

Figure 1. Growth and reproductive parameters in mice exposed in utero to ethanol (EtOH). CD-1 mice were mated overnight, plug was designated gestation day (GD) 1. On GD 9, dams were dosed with EtOH (4 g/kg i.p) and sacrificed on GD19 to assess fetal structural birth defects. Fetuses were stored in Bouin’s solution for 3 days to fix the tissues and transferred to 70% EtOH for storage.

604

Figure 2. Brain and neural tube defects in mice exposed in utero to EtOH. CD-1 mice were mated overnight, plug was designated gestation day (GD) 1. On GD 9, dams were dosed with EtOH (4 g/kg i.p) and sacrificed on GD19 to assess fetal structural birth defects. Fetuses were stored in Bouin’s solution for 3 days to fix the tissues and transferred to 70% EtOH for storage.

605

Figure 3. Ocular defects in mice exposed in utero to EtOH. CD-1 mice were mated overnight, plug was designated gestation day (GD) 1. On GD 9, dams were dosed with EtOH (4 g/kg i.p) and sacrificed on GD19 to assess fetal structural birth defects. Fetuses were stored in Bouin’s solution for 3 days to fix the tissues and transferred to 70% EtOH for storage.

606

Figure 4. Umbilical and renal defects in mice exposed in utero to EtOH. CD-1 mice were mated overnight, plug was designated gestation day (GD) 1. On GD 9, dams were dosed with EtOH (4 g/kg i.p) and sacrificed on GD19 to assess fetal structural birth defects. Fetuses were stored in Bouin’s solution for 3 days to fix the tissues and transferred to 70% EtOH for storage.

607

Figure 5. Tail defects in mice exposed in utero to EtOH. CD-1 mice were mated overnight, plug was designated gestation day (GD) 1. On GD 9, dams were dosed with EtOH (4 g/kg i.p) and sacrificed on GD19 to assess fetal structural birth defects. Fetuses were stored in Bouin’s solution for 3 days to fix the tissues and transferred to 70% EtOH for storage.