Developmental Expression of Heme Degradation Genes in Zebrafish

DEVELOPMENTAL EXPRESSION OF HEME DEGRADATION GENES IN ZEBRAFISH

(DANIO RERIO) SUGGEST NOVEL ROLES IN HEMATOPOIESIS AND EYE

DEVELOPMENT

ANDREW HOLOWIECKI

MATTHEW J. JENNY, COMMITTEE CHAIR JOHN YODER CAROL DUFFY STEVE MARCUS PATRICK FRANTOM

A DISSERTATION

Submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in the Department of Biological Sciences in the Graduate School of The University of Alabama

TUSCALOOSA, ALABAMA

2015

Heme is an important cofactor for numerous proteins, including hemoglobin used for oxygen transport. However, in its unbound form, heme is highly toxic and can perpetuate the formation of harmful reactive oxygen species (ROS). Thus, the heme degradation pathway, an evolutionarily conserved enzymatic pathway, facilitates the metabolism of heme into excretable by-products that may also have significant biological functions. Heme oxygenase (HO-1, HO-2) degrades heme into biliverdin (BV). Subsequently, BV is converted into the potent antioxidant bilirubin (BR) by biliverdin reductase (BVRa or BVRb). The ability of BR to efficiently quench

ROS through a recycling reaction mediated by BVR is of great interest with strong medicinal implications. In support of this argument, individuals with slightly elevated levels of BR, a condition known as Gilbert’s syndrome, have lower incidences of diseases related to oxidative stress, such as cardiovascular disease. The goal of this research is to characterize the transcriptional regulation of genes involved in the heme degradation pathway. Chapter 2 provides a thorough characterization of expression of both sets of the HO-1 and HO-2 paralogs, as well as BVRa and BVRb, during normal development and in response to oxidative stress. It also provides new data on the sex-specific differences in expression of these genes under control and stressful conditions. The observed quantitative changes in divergent gene expression provide strong initial support for subfunction partitioning of the HO paralogs. Chapter 3 presents qualitative data on the expression patterns of HO-1a, BVRa and BVRb during development, conclusively demonstrating spatiotemporal patterns that are consistent with hematopoiesis.

Additionally, in vivo promoter analysis demonstrates specific expression of HO-1a in eye lens epithelial cells during development. Finally, Chapter 4 presents functional data on the roles of

GATA-1, the master regulator of erythroid development, and NRF2a, the master regulator of oxidative stress, on the spatiotemporal regulation of HO-1a, BVRa and BVRb during normal development and in response to oxidative stress. Furthermore, all three genes are expressed in developmental tissues that are highly sensitive to oxidative stress. In Chapter 5, we describe alternative experimental approaches, suggest future studies, and discuss these results in context with the current knowledge base.

iii

DEDICATION

I dedicate this study to:

My wife Kristy who has supported me through this, completion of this goal would not have been possible without your patience, love, and understanding. My daughter Lilly, who has been with me throughout this entire process, from birth she has looked over my data sets, critiqued my figures, and provided spot on advice on how to handle stress. My daughter Georgia, who entered the family half way through this project, your enthusiasm for life is contagious and inspirational.

Your comedic timing is spot on. I love how you always listen to me prepare for seminars and repeat back “big science words”. My new son, Henry, I hope you will be proud of me.

My mother, whom has set an example of what it means to work hard: first shift, second shift, or third shift for nearly 40 years at a steel plant. You continued to do this to help me during my time in graduate school and to help support me and my family. Clearly I could not have done this without you. My father, who passed away during my first year here, you instilled a strong sense of responsibility in me. I understand that it is a privilege to get to pursue an education. Thank you. My mother-in-law and father-in-law who have provided support and guidance: You have put in many miles to visit us down here. Thank you.

LIST OF ABBREVIATIONS AND SYMBOLS

AHR aryl hydrocarbon receptor

ALM anterior lateral mesoderm

AP-1 activator protein-1

ARNT aryl hydrocarbon receptor nuclear translocator proteins bHLH PAS basic-helix-loop-helix Per-ARNT-SIM protein family

BVR biliverdin reductase

BVRa biliverdin reductase alpha

BVRb biliverdin reductase beta

BV biliverdin

BV-IXα alpha isomer of biliverdin

BV-IX-β beta isomer of biliverdin

BV-IXδ delta isomer of biliverdin

BV-IXγ gamma isomer of biliverdin

BR bilirubin

BR-IXα alpha isomer of bilirubin

BR-IX-β beta isomer of bilirubin bZip basic leucine zipper DNA binding domain

Cd cadmium

v cDNA complementary deoxyribonucleic acid

CHT caudal hematopoietic tissue

CMP’s common myelo-erythroid progenitors

CRISPRs clustered regularly interspaced short palindromic repeats

CUL3 cullin3

CYP1A cytochrome-p4501A

DA dorsal aorta

DDC duplication, degeneration, and complementation model

DIG digoxigenin

EDCs endocrine-disrupting chemicals eGFP enhanced green fluorescent protein

EHT endothelial-to-hematopoietic transition

EMPs erythomyeloid progenitors

ETS E26 transformation-specific family

FIH-1 factor inhibiting hypoxia inducible factor 1α

GRPX glutathione peroxidase

GRX glutathione reductase

GSH reduced glutathione

GSSG oxidized glutathione

Hif1α hypoxia inducible factor-1α

HO heme oxygenase

HO-1 heme oxygenase 1

HO-2 heme oxygenase 2

HOOH hydrogen peroxide

HRE hypoxia response element

HSCs hematopoietic stem cells

HSF heat shock factor

HSP heat shock protein

ICM intermediate cell mass

IDHc cytosolic isocitrate dehydrogenase

IDHm mitochondrial isocitrate dehydrogenase

INRF2 inhibitor of nuclear factor erythroid 2-related factor 2

ISVs intersegmental veins

Keap1 kelch-like ECH-associated protein 1

KOH potassium hydroxide

Le lens

LTZ lens transition zone mCherry red fluorescent protein (“m” monomer conformation)

MAPK mitogen-activated protein kinase

MEPS Mass Embryo Production System

MO morpholino

MTF-1 metal-regulatory transcription factor 1

NAD nicotinamide adenine dinucleotide

NADH nicotinamide adenine dinucleotide (reduced form)

NADP nicotinamide adenine dinucleotide phosphate

NADPH nicotinamide adenine dinucleotide phosphate (reduced form)

vii

NES nuclear export signal

NLS nuclear localization signal

NFκB nuclear factor–κB

NNT nicatinamide nucleotide transhydrogenase

NRF2 nuclear factor erythroid 2-related factor 2

OD o-dianisidine

OP olfactory placode

PBI posterior blood island

PBS phosphate buffered saline

PBST phosphate buffered saline with 0.1% tween

PCBs polychlorinated biphenols

PFA paraformaldehyde

PL posterior lens

PLM posterior lateral mesoderm

PMBC primordial midbrain channel

PPP pentose phosphate pathway

PQQ pyrroloquinoline quinone

PRDX peroxiredoxin qPCR quantitative polymerase chain reaction

RBC red blood cell

ROS reactive oxygen species

RPE retinal pigment epithelium

SOD superoxide dismutase

viii

S/T/K serine, threonine, tyrosine kinase

TALENs transcription activator-like effector nucleases tBHQ tert-butylhydroquinone

TCDD dioxin

VEGF vascular endothelial growth factor

UROD uroporphyrinogen decarboxylase

WBC’s white blood cell

XRE xenobiotic response element

ACKNOWLEDGEMENTS

First I would like to thank Dr. Matthew Jenny for taking a chance on me, providing guidance, and continually challenging me. Additionally, I would like to thank my committee members:

Your dedication to teaching in the classroom played a big part in my success. Many of my ideas originated from classroom discussions. Finally, I would like to acknowledge members of the

Jenny lab.

CONTENTS ABSTRACT ...... ii

DEDICATION ...... iv

LIST OF ABREVIATIONS AND SYMBOLS ...... v

ACKNOWLEDGEMENTS ...... x

LIST OF TABLES ...... xiv

LIST OF FIGURES ...... xv

1 UTILIZING ZEBRAFISH (DANIO RERIO) AS A DEVELOPMENTAL AND TOXICOLOGICAL MODEL ...... 1

1.1 Introduction ...... 1

1.2 Zebrafish as a Developmental, Toxicological, and Disease model ...... 2

1.3 The Zebrafish Genome and Subfunction Partitioning ...... 4

1.4 Oxidative Stress and Redox Signaling ...... 5

1.5 Antioxidant Defense Mechanisms ...... 6

1.6 Transcriptional Regulators of Stress Response and their Roles in Development ...... 8

1.7 NRF2 is the Master Regulator of Oxidative Stress ...... 8

1.8 MTF-1 is the Master Regulator of Metal Homeostasis...... 9

1.9 bHLH-PAS Proteins Act as Environmental Sensors ...... 10

1.10 Heat Shock Factors ...... 11

1.11 The Heme Degradation Pathway ...... 12

1.12 Heme Degradation: Production of Second Messengers and the Potent Antioxidant BR ...... 14

1.13 Arguments against BR Recycling ...... 16

1.14 Primary Research Focus on BVRa and the Neglect of BVRb ...... 17

1.15 Regulation of the Heme Degradation Pathway ...... 18

1.16 Roles for HO and BVR in Hematopoiesis...... 19

1.17 Transcriptional Regulators of Hematopoiesis ...... 20

1.18 Zebrafish Hematopoietic Stem Cell Differentiation ...... 22

1.19 Conclusion ...... 24

2 ONTOGENIC EXPRESSION OF HO AND BVR IN RESPONSE TO OXIDATIVE STRESS ...... 25

2.1 Abstract ...... 25

2.2 Introduction ...... 26

2.3 Materials and Methods ...... 29

2.4 Results ...... 34

2.5 Discussion ...... 51

3 TEMPORAL AND SPATIAL EXPRESSION OF HO AND BVR DURING NORMAL DEVELOPMENT AND IN RESPONSE TO OXIDATIVE STRESS ...... 57

3.1 Abstract ...... 57

3.2 Introduction ...... 58

3.3 Materials and Methods ...... 59

3.4 Results ...... 68

3.5 Discussion ...... 81

xii

4 FUNCTIONAL CHARACTERIZATION OF HO AND BVR GENE REGULATION DURING NORMAL DEVELOPMENT AND IN RESPONSE TO STRESS ...... 87

4.1 Abstract ...... 87

4.2 Introduction ...... 88

4.3 Materials and Methods ...... 90

4.4 Results ...... 93

4.5 Discussion ...... 108

5 CONCLUSIONS, EXPERIMENTAL CHALLENGES, AND FUTURE DIRECTIONS ...... 114

5.1 Summary ...... 114

5.2 Further Characterization of Roles for HO-1a, BVRa, and BVRb in Hematopoiesis ...... 116

5.3 Further Characterization of Roles for NRF2 as a HSC Regulator ...... 118

5.4 Heme Oxygenase Paralogs May Have Underwent Subfunction Partitioning ...... 119

5.5 Developmental Roles for HO-1a in Zebrafish Eye Development ...... 121

5.6 Novel Roles for HO-1, BVRa and BVRb during Early Development ...... 121

5.7 Concluding Remarks ...... 124

REFERENCES ...... 125

xiii

LIST OF TABLES

2.1 qPCR and 5’ RACE PCR Primers ...... 34

3.1 qPCR, Cloning, and Gateway Cloning Primers ...... 68

xiv

LIST OF FIGURES

1.1 Glutathione Recycling ...... 7

1.2 Antioxidant Recycling Pathways ...... 16

1.3 Differences in the Structure and Enzymatic Activity of BVRa and BVRb ...... 18

1.4 Three Main Transcription Factors Regulate Hematopoietic Development ...... 23

2.1 Zebrafish HO-1a and HO-1b Contain Conserved Heme Oxygenase Domains and Heme Signature Motifs ...... 35

2.2 Phylogenetic Analysis of Heme Oxygenase Proteins ...... 36

2.3 HO Expression in Adult Zebrafish Tissues ...... 37

2.4 BVR Expression in Adult Zebrafish Tissues ...... 38

2.5 HO-1 Induction in Adult Zebrafish in Response to Cd Exposure ...... 39

2.6 HO-2 Induction in Adult Zebrafish in Response to Cd Exposure ...... 40

2.7 Adult BVR Expression in Response to Cd Exposure ...... 41

2.8 Changes in Expression of HO and BVR Between 24-124 hpf ...... 42

2.9 Zebrafish are More Sensitive to Cd and tBHQ During Peaks in Expression of Genes Responsible for Producing NADPH...... 44

2.10 Effects of Acute Cd or tBHQ Exposure on Gene Expression in 72 hpf Zebrafish ...... 45

2.11 Effects of Continually Increasing Cd Exposure on Genes Involved Heme Degradation ...... 46

2.12 Effects of Short Term and Continuous Cd Exposures on Genes Involved in Iron Homeostasis...... 47

2.13 Effects of Acute Hemin Exposure on HO and BVR Expression ...... 49

2.14 MTF-1 and NRF2 Binding Motifs within Zebrafish HO...... 50 and BVR Promoters

3.1 HO-1a in Vivo DNA Expression Construct for Tissue Specific Expression ...... 63

3.2 Changes in BVR Expression during Primitive Hematopoiesis...... 69

3.3 Changes in BVR Expression during the Onset of Definitive Hematopoiesis ...... 70

3.4 Changes in BVRa and BVRb Expression at 48 and 72 hpf ...... 71

3.5 Changes in BVRa and BVRb Expression at 96 and 120 hpf ...... 71

3.6 Effects of Cd on BVRa Expression during Definitive Hematopoiesis ...... 72

3.7 Effects of Cd on BVRb Expression during Definitive Hematopoiesis ...... 73

3.8 Effects of Acute Cd Exposure on BVR Expression...... 74

3.9 HO-1a Expression within the ICM and Embryonic RBC’s during Primitive Hematopoiesis ...... 75

3.10 Comparison of HO-1a mRNA and mCherry Expression Driven by the HO-1a Promoter in Eye Tissues ...... 76

3.11 HO-1a Expression in Liver, Kidney, and Eye Tissue Between 4-5 dpf ...... 77

3.12 WISH, in Vivo Promoter Analysis, and qPCR Show Differences in HO-1a Induction by Cd at 4, 5 and 6 dpf ...... 79

3.13 HO-1a Lens Expression is Specific to Lens Epithelial Cells ...... 80

3.14 GATA-1 Response Elements within Promoter Regions of HO-1a, BVRa, and BVRb ...... 83

4.1 GATA-1 Knockdown Results in Decreases in RBC’s ...... 94

4.2 GATA-1 Knockdown Attenuates BVRa and BVRb Expression during Primitive Hematopoiesis ...... 95

4.3 GATA-1 Knockdown Decreases BVRb Expression in Primitive RBC’s at 48 hpf ...... 96

xvi

4.4 Effects of GATA-1 Knockdown on BVR Expression during Definitive Hematopoiesis ...... 97

4.5 GATA-1 Knockdown Does not Decrease HO-1a Promoter Driven mCherry Expression within the ICM ...... 99

4.6 GATA-1 Knockdown Alters Spatial Distribution of RBC...... 100

4.7 NRF2a Knockdown Increases RBC Count at 48 hpf...... 101

4.8 NRF2a Knockdown Results in an Increase in BVRa Expression within the ICM at 24 hpf ...... 102

4.9 NRF2a Morphants Express BVRb Transcripts within the Dorsal Aorta ...... 103

4.10 NRF2a Knockdown Does not Alter BVRa and BVRb Expression at 48 hpf...... 104

4.11 NRF2a Knockdown Results in a Decrease in BVRb Expression in Cd Challenged Embryos at 96 hpf...... 106

4.12 Effects of NRF2a Knockdown on HO-1a Expression in Control and Cd Challenged Fish...... 108

xvii

CHAPTER 1

UTILIZING ZEBRAFISH (DANIO RERIO) AS A DEVELOPMENTAL AND TOXICOLOGICAL MODEL

1.1 Introduction

Efficient cellular responses to oxidative stress are crucial for proper development, survival, and proliferation. Oxidative stress results from an over-abundance of reactive oxygen species (ROS) which results in the disruption of cellular homeostasis. Unbound cellular heme, a normal metabolic by-product, is highly reactive, and can generate damaging ROS. Accordingly, multiple mechanisms exist which tightly regulate cellular stress responses. Heme oxygenase

(HO) catalyzes the breakdown of heme into carbon monoxide (CO), free iron, and the bile pigment biliverdin (BV). A novel antioxidant cycling pathway catalyzed by biliverdin reductase

(BVR) utilizes cytosolic NADPH to reduce BV to the potent antioxidant bilirubin (BR) (Dore

1999, Baranano and Snyder 2001, Baranano 2002). This pathway is thought to confer protection against ROS to lipophilic compartments of the cell through the protective benefits of bilirubin

(BR).

While there are two isoforms of HO (HO-1 and HO-2) and BVR (BVRa and BVRb), studies evaluating the antioxidant benefits of this pathway often neglect HO-2 and BVRb. Furthermore, very little is known regarding the role of these enzymes during embryonic development. Thus, the main goal of this research was to characterize the transcriptional regulation of the HO and

BVR isoforms at various ontogenic stages under normal conditions, as well as in response to

oxidative stress using the zebrafish (Danio rerio) as a model organism. As a result of a teleost- specific genome duplication event, zebrafish have paralogous copies of both HO-1 and HO-2.

Characterization of any potential divergence in the regulation of the paralogs could provide insight into the partitioning of function or evolution of novel function. For the first aim, real- time RT-PCR was used to characterize HO and BVR expression during development and in adult tissues under normal conditions and in response to pro-oxidant exposures. To gain further insights regarding developmental roles of HO and BVR, aim 2 used in situ hybridization and in vivo promoter analysis techniques to evaluate changes in the spatial and temporal patterns of gene expression during early development under control and stressed conditions. Finally, the third aim investigated the roles of GATA-1, the master regulator of erythropoiesis, and NRF2a, the master regulator of oxidative stress, in the regulation of HO-1a, BVRa and BVRb via targeted knockdown of the transcription factors using antisense morpholino technology coupled with in situ hydridization and in vivo promoter assays.

1.2 Zebrafish as a Developmental, Toxicological, and Disease Model

The zebrafish (Danio rerio) has become a popular model organism for developmental and toxicological studies for various reasons which include a high rate of fecundity, rapid transparent development, small size, and an annotated genome (Hill 2005, Spitsbergen and Kent 2003, Zon

2005, Amsterdam et al. 2004, Murphey et al. 2006, Traver et al. 2003). Development from zygote to early larva occurs within 72 hours and during this time milestones including gastrulation, segmentation, axis formation, circulation, and organogenesis are completed, all of which are readily observable using basic microscopy techniques (Kimmel 1995). These benefits have facilitated the use of many molecular techniques such as in vivo promoter analysis, in situ

hybridization, mRNA overexpression, and multiple techniques for targeted gene knockdown including antisense morpholinos (MOs), transcription activator-like effector nucleases

(TALENS), and the widely accepted Clustered Regularly Interspaced Short Palindromic Repeats

(CRISPR) type II system, which collectively allow for mechanistic studies of specific genes in embryo toxicity (Timme-Laragy 2012, Ablain et al. 2015, Malicki et al. 2002).

Remarkably, ~80% of human genes implicated in disease have zebrafish counterparts (Howe et al. 2013). This high degree of developmental conservation amongst vertebrates has facilitated the development of numerous disease models. Zebrafish disease models with direct human relevance have been established for Duchenne muscular dystrophy (Kawahara et al. 2014), kidney disease (Sun et al. 2004), heart disease (Xu et al. 2002), anemia (Donovan et al. 2000, Bai et al. 2010), cancer (Amatruda et al. 2002, Bai et al. 2010), retinal degeneration (Li and Dowling

1997), cataractogenesis (Goishi et al. 2006), anxiety (Peitsaro et al. 2003), and autism (Turner et al. 2015). Additionally, zebrafish have proven to be an effective model for the study of tissue regeneration (Poss, Keating, and Nechiporuk 2003, Jopling et al. 2010, Dohn and Waxman

2012). Collectively, the transparent embryonic development, sequenced genome, and amiability to both forward and reverse genetics techniques makes the zebrafish an ideal model to further characterize HO and BVR isoforms. Furthermore, zebrafish have proven to be an excellent model for developmental and toxicological studies with results being highly applicable towards human development (Patton 2001, Hill 2005). Thus, the research presented in this dissertation may reveal novel insights which are directly relevant to human development and cellular defense mechanisms.

1.3 The Zebrafish Genome and Subfunction Partitioning

The zebrafish genome contains over 26,000 protein coding genes (Collins et al. 2012), a significant number which result from a teleost specific genome duplication that occurred 100 million years ago (Amores et al. 1998, Postlethwait 2007, Meyer 2005). Thus, teleosts may possess two copies of a gene in comparison to their mammalian counterpart. Accordingly, the duplication, degeneration, and complementation model (DDC) suggests that duplicated genes can gain a new function, partition their function, or become pseudogenes which can be eventually lost through selection if the redundancy is not needed (Force 1999). The partitioning of gene function can be experimentally manipulated to gain novel insights into their functions, as mutations which are lethal in mammalian models can often be performed in zebrafish (Hill

2005). For example, zebrafish contain two copies of transferrin receptor 1 (tfr1a and tfr1b), however, only tfr1a is necessary for the production of hemoglobin (Wingert et al. 2004).

Similarly, zebrafish possess paralogs of multiple stress response transcription factors which have undergone subfunction partitioning, such as nuclear factor erythroid 2-related factor 2 (NRF2a and NRF2b) (Timme-Laragy 2012), the aryl hydrocarbon receptor (ahr1a and ahr1b) (Karchner,

Franks, and Hahn 2005), and the aryl hydrocarbon receptor repressor (ahrra and ahrrb) (Jenny et al. 2009). While paralogs of HO-1 (HO-1a and HO-1b) and HO-2 (HO-2a and HO-2b) have been identified within the zebrafish genome (Nakajima and Mukaigasa 2011), functional characterization of the various isoforms is still lacking. Investigations into the differences in the transcriptional regulation of these duplicated (paralogous) genes is the first step into determining if diversification and subfunction partitioning of these enzymes has occurred.

1.4 Oxidative Stress and Redox Signaling

Oxygen-derived free radicals, termed reactive oxygen species (ROS), may be produced as a byproduct of endogenous aerobic respiration but also result from the continuous onslaught of environmental insults. Oxidative stress occurs when there is an imbalance between ROS and protective antioxidants. Excessive oxidative stress can drain cells of their reducing power and

ATP. Generators of ROS include heavy metals, various drugs and toxic xenobiotics, hypoxic reperfusion and temperature stress. The perpetuation of oxidative stress can lead to cellular and nuclear damage and is associated with several diseases which affect the kidney, skin, eye, heart, lung, joint, brain, and digestive system (Niture, Khatri, and Jaiswal 2014).

Three common mechanisms that generate cellular oxidative stress include redox cycling,

Fenton reactions and disruption of mitochondrial electron transport. Redox-cycling occurs when a xenobiotic undergoes a single electron reduction to form a radical species which may then react

- with molecular oxygen to form the superoxide anion radical [O2 ], resulting in the conversion of the xenobiotic back to its original oxidative state. This continuous redox cycling between the xenobiotic and oxygen molecules can continue causing significant damage to the cell. Peroxides in the presence of transition metals such as iron (Fe2+ and Fe3+) undergo Fenton chemistry to produce the highly unstable hydroxyl radical [HO]. Importantly, the highly reactive hydroxyl radical cannot be detoxified through enzymatic mechanisms and can cause significant damage to cellular macromolecules. Finally, mitochondrial damage or exposure to chemicals that interfere with the major electron transport complexes can significantly increase the amount of endogenous

ROS associated with oxidative phosphorylation.

While numerous pathways have evolved to confer protection, these pathways all require high levels of cellular reductants (NADPH and NADH) and ATP. To this end, toxicants which

disrupt cellular processes, deplete cellular resources, or alter energy production are all considered major mechanisms of oxidative stress. Presented next is an overview of the primary defense mechanisms used to combat oxidative stress, the transcriptional regulators which regulate these mechanisms, and the roles that these transcription factors have in developmental processes.

1.5 Antioxidant Defense Mechanisms

Cellular health is exemplified by high ratios of reduced to oxidized nicotinamide adenine dinucleotide phosphate (NADPH: NADP+), nicotinamide adenine dinucleotide (NADH: NAD+), and glutathione (GSH: GSSG), as well as a high ratio of ATP to ADP. As oxidative stress depletes these cellular resources ultimately leading to cell death, protection against ROS is of high importance. To this end multiple signaling pathways and proteins work together to sense, respond and eliminate oxidative stress. The frontline defense against ROS is the generation of

GSH which acts as a potent antioxidant that effectively quenches ROS. GSH is maintained by the classic glutathione recycling pathway via the action of NADPH-dependent glutathione reductase (GRX) (Figure 1.1). Continuous generation of reduced GSH is dependent on the availability of reducing power in the form of NADPH which is generated via the pentose phosphate pathway (PPP) (Wu 2011), NADPH producing cytosolic and mitochondrial isocitrate dehydrogenases (Jennings and Stevenson 1991, Kirsch 2001, Jo et al. 2001), as well as the membrane bound nicotinamide nucleotide transhydrogenase (NNT) (Yin, Sancheti, and Cadenas

2012).

Figure 1.1. Glutathione Recycling. Reduced glutathione (GSH) is the most abundant cellular antioxidant and acts as a scavenger of ROS. GSH quenches ROS by donating an electron from its reactive sulfhydryl group and becoming oxidized to GSSG in the process. Healthy cells typically contain GSH:GSSG ratios of 100:1. Continued maintenance of this ratio is essential, and requires a favorable redox environment. NADPH serves as the preferred electron donor in cytosolic and mitochondrial redox reactions, and is needed for continuous generation of GSH. Sources of NADPH include the Pentose Phosphate Pathway (PPP), soluble and mitochondrial isocitrate dehydrogenases (IDH), as well as the mitochondrial nicotinamide nuclear transhydrogenase (NNT).

Other defense mechanisms include antioxidant enzymes such as superoxide dismutase

(SOD), catalase, peroxiredoxin (PRDX), and glutathione peroxidase (GRPX). SOD eliminates

- - reactive superoxide anions (O2 ) via the dismutation of (O2 ) to hydrogen peroxide (HOOH).

Conversely, catalase, PRDX, and GRPX prevent the formation of highly reactive hydroxyl radicals (HO) by reducing hydrogen peroxide to water. This is of utmost importance as (HO) cannot be detoxified through cell-mediated processes (Holmstrom and Finkel 2014). Finally, the

HO/BVR enzymatic pathway is a more recently proposed antioxidant mechanism that protects the lipophilic regions of the cell in a BV:BR recycling reaction that is analogous to the GSH recycling pathway. Further details of the HO/BVR pathway are provided in later sections of this chapter.

1.6 Transcriptional Regulators of Stress Response and Their Roles in Development

Cells respond to environmental challenges such as temperature extremes, oxygen fluctuations and toxic exposures by a variety of adaptive mechanisms. In response to oxidative challenges, several different transcription factors are activated resulting in translocation to the nucleus and binding to consensus motifs within the promoter of their target genes. These transcription factors, often referred to the master regulators of the stress response, are responsible for activating the expression of different gene batteries which in turn ultimately result in the production of proteins involved in the synthesis of antioxidants, expression of antioxidant enzymes or expression of other proteins that play a role in detoxification or cellular repair. In addition to serving as protectors of cellular health, many of these transcription factors have essential developmental roles. Major mechanisms of developmental toxicity are often associated with the excessive activation of these transcription factors leading to disruption of normal cellular processes. The proceeding sections discuss some of the main transcriptional regulators of the antioxidant and cellular stress response and their roles in development.

1.7 NRF2 is the Master Regulator of Oxidative Stress

Nuclear factor erythroid 2-related factor 2 (NRF2) serves as the master regulator of oxidative stress. Upon oxidative insult, NRF2 promotes a strong antioxidant response by activating a battery of cytoprotective genes (Niture et al. 2010, Niture, Khatri, and Jaiswal 2014). Under normal physiological conditions, NRF2 is maintained in the cytoplasm by kelch-like ECH- associated protein 1 (Keap1), also known as inhibitor of NRF2 (INRF2), which facilitates the ubiquitination of NRF2 by cullin3 (CUL3) (Sporn and Liby 2012) leading to degradation by the

26S proteasome (Niture, Khatri, and Jaiswal 2014). Two highly reactive thiol groups within

Keap1 are directly acted upon by ROS, resulting in release of NRF2 preventing it from being ubiquitinated. This results in the increase of NRF2 protein levels and subsequently increased nuclear translocation. Upon nuclear entrance, NRF2 recognizes and binds to consensus antioxidant response element (ARE) binding sequences (TGAG/CNNNGC) in the promoter regions of antioxidant genes resulting in their transcriptional upregulation (Motohashi et al.

2004, Venugopal 1998, Uruno and Motohashi 2011). NRF2 is responsible for regulating the expression of the enzymes involved in the biosynthesis of GSH, as well as the enzymes that are responsible for maintaining high cellular levels of reduced GSH. NRF2 is also a known regulator of HO-1 expression under conditions of oxidative stress (Alam et al. 1999). In addition to regulating antioxidant genes, NRF2 maintains cellular levels of NADPH by upregulating genes in the Pentose Phosphate Pathway (Wu 2011). Finally, NRF2 appears to have roles in hematopoiesis and stem cell survival (Merchant et al. 2011, Tsai et al. 2013), which will be highly relevant to the research presented in this dissertation.

1.8 MTF-1 is the Master Regulator of Metal Homeostasis

Metal responsive transcription factor 1 (MTF-1), the master regulator of metals homeostasis, is activated upon exposure to heavy metals such as cadmium, mercury, and lead resulting in the upregulation of metallothioneins, metal binding proteins involved in sequestration of toxic metals and chaperoning of essential metals. MTF-1 translocates to the nucleus in a Zn- dependent manner where it binds to the core consensus sequence 5’-TGCRCNC-3’ (R=A or G,

N=any nucleotide) within the promoter regions of its target genes (Stuart 1985, Günther, Lindert, and Schaffner 2012). In addition to its well-known roles in maintaining metals homeostasis,

MTF-1 is necessary for liver development (Gunes 1998) and has been implicated in the

regulation of genes involved in oxidative stress signaling, muscle development, and eye development (O'Shields et al. 2014, Hogstrand C 2008). MTF-1 also plays a role in regulating - glutamylcysteine synthetase, the rate-limiting enzyme involved in the biosynthesis of GSH

(Andrews 2001). Other roles for MTF-1 include roles in maintaining white blood cells (WBC’s)

(Wang 2004), regulating cellular of iron and heme levels (O'Shields et al. 2014, Balesaria et al.

2010, Troadec et al. 2010), and responding to hypoxic stress (Green et al. 2001).

1.9 bHLH-PAS Proteins Act as Environmental Sensors

The aryl hydrocarbon receptor (AHR) is a ligand-activated transcription factor belonging to the basic-helix-loop-helix (bHLH) Per-ARNT-SIM (PAS) protein family of transcription factors that act as environmental stress sensors and participate in developmental cell signaling processes

(Hahn, Allan, and Sherr 2009). The AHR is bound by chaperone proteins (e.g. HSP90) and is localized in the cytoplasm under normal conditions. Exposure to AHR agonists results in the dissociation of AHR from the chaperones followed by nuclear translocation and dimer formation with AHR nuclear translocator (ARNT) proteins. The complex (ligand-AHR-ARNT) binds to the consensus xenobiotic response element (XRE) 5’-T/GNGCGTGA/CG/CA-3’ (Lusska, Shen, and Whitlock 1993, Yao and Denison 1992) located in the promoter region of AHR regulated genes. These receptors are activated by xenobiotics, such as dioxin (TCDD) and polychlorinated biphenols (PCBs), and upon activation induce the expression of detoxifying enzymes such as cytochrome-P450 1A (CYP1A) (Whitlock 1999, Tijet et al. 2006). CYP1A proteins are monooxygenases that utilize molecular oxygen as a substrate to hydroxylate xenobiotics. When

CYP1A proteins metabolize bulky xenobiotics or become overwhelmed by excessive xenobiotic exposure, they become proficient generators of ROS. Thus, there is significant crosstalk

between the AHR and NRF2 pathways to regulate the expression of antioxidant proteins. In addition to its well-known role as an environmental sensor, numerous studies have identified potential regulatory roles for AHR in hematopoiesis (Smith et al. 2013, Singh et al. 2011,

Boitano et al. 2010, Aluru, Jenny, and Hahn 2014) in addition to roles in photoreceptor development and hemostasis (Aluru, Jenny, and Hahn 2014).

Under hypoxic conditions hypoxia inducible factor-1α (HIF-1α), another member of the bHLH-PAS protein family, also utilizes ARNT (also known as HIF-1β) as a dimerization partner to form a transcriptional complex which binds to the consensus hypoxia response element (HRE)

5’-RCGTG-3’ to regulate a battery of genes which defend against the low oxygen levels

(Semenza et al. 1996). In response to decreasing oxygen levels, HIF-1α increases O2 transport by activating genes necessary for blood vessel development and erythrocyte production

(Semenza 2000). The reperfusion of O2 associated with the increased expression of these target genes can result in the production of oxidative stress. Thus, several studies have supported roles for crosstalk between HIF-1α and NRF2, as well as MTF-1 (Murphy et al. 2008, Murphy et al.

2005, Ryter, Alam, and Choi 2006, Malec et al. 2010, Simmons, Fan, and Ramabhadran 2009).

Finally, HIF-1α has also been shown to regulate the expression of HO-1 under cellular stress conditions (Lee et al. 1997).

1.10 Heat Shock Factors

Heat shock factors (HSFs) increase the production of heat shock proteins (HSPs) in response to a variety of physiological and environmental stressors including heat, oxidative stress and heavy metals (Akerfelt, Morimoto, and Sistonen 2010). While different stressors result in the upregulation of different HSPs (Vihervaara and Sistonen 2014), the mechanism for upregulation

is the same regardless of the initiator. HSFs bind to heat shock elements (HSEs) located in the upstream promoter of HSPs. HSEs consist of a minimum of three inverted repeats of the pentameric sequence nGAAn or nTTCnnGAAnnTTCn (Akerfelt, Morimoto, and Sistonen 2010).

Many HSPs act as molecular chaperones and assist in protein-protein interactions, protein folding, and protein transport. In addition to well characterized roles in response to oxidative stress, roles for HSPs in the regulation of the erythroid specific transcription factor GATA-1 (de

Thonel et al. 2010) and the myeloid specific transcription factor PU.1 (Jego et al. 2014) have been noted.

1.11 The Heme Degradation Pathway

Heme oxygenase, the rate limiting enzyme of heme degradation (Choi and Alam 1996), was first described nearly 50 years ago using rat liver and spleen as the source of enzyme (Tenhunen,

Marver, and Schmid 1969). The degradation of heme to biliverdin (BV) by heme oxygenase 1

(EC 1.14.99.3) or heme oxygenase 2 proceeds as follows:

1/2 1/2 + 2+ 1/2 + Heme b + 3O2 + 3 NADPH + 3 H  BV+ Fe + CO + 3 NADP + 3H2O.

Subsequently, using rat and guinea pig tissues as the source of enzyme it was determined that biliverdin reductase (BVR) (EC 1.3.1.24) further reduced BV into bilirubin (BR) (Tenhunen et al. 1970, Singleton and Laster 1965) as follows:

BR + NAD(P)+   BV + NAD(P)H + H+

Reduction of the easily excretable BV to the insoluble and toxic BR was initially puzzling

(Colleran and Heirwegh 1979), however, it was later found that BV could not cross the placenta and conversion to BR was necessary for clearance of BV from the fetus of placental mammals

(McDonagh, Palma, and Schmid 1981). Interestingly, studies have demonstrated that BR can act

as a potent antioxidant (Stocker 1987, Stocker, Glazer, and Ames 1987, Stocker and Ames 1987) in a recycling reaction analogous to the GSH recycling pathway (Baranano 2002). Two isoforms of BVR (BVRa and BVRb) which produce different BR isomers in placental animals have been described (Yamaguchi 1994). Bilirubin-IXα (BR- IXα) is predominantly found in adult human bile and cannot cross the placenta, while bilirubin-IXβ (BR-IXβ) is found predominantly in fetal bile and can cross the placenta (Yamaguchi 1994, Yamaguchi 1995). Collectively, the reported antioxidant properties of BR (Stocker 1987, Stocker, Glazer, and Ames 1987, Stocker and Ames

1987), as well as the distinct isozymes of BVR and subsequent localization of their BR end product led to the hypothesis that BVRb conferred fetal protection to oxidative stress by generating BR-IXβ, while BVRa conferred protection against oxidative stress in adult tissues by generating BR- IXα (Yamaguchi 1994). This hypothesis served as a plausible explanation regarding the initial observation of the energetically wasteful conversion of BV to BR.

However, even though this argument is plausible and is often presented as fact, a degree of controversy as to the efficiency in which BR can cross the placenta remains.

The initial report by McDonagh that BV is unable to cross the placenta and therefore must be reduced to BR (McDonagh, Palma, and Schmid 1981) for proper clearance has come under question, most notably by McDonagh himself who suggests that the greater levels of BR-IXβ detected in fetal bile are an artifact which exists due to the properties of BV-IXα, BV-IXβ, BR-

IXα, and BR-IXβ (McDonagh 2001). McDonagh points to the different placental permeabilities of BR-IXβ and BR-IXα as the reason for the greater detection of BR-IXβ in comparison to BR-

IXα within fetal tissues. Specifically, the BR-IXβ isomer is polar, thus no further conjugation is necessary for excretion and as such, BR-IXβ remains within fetal bile because it can be metabolized. In contrast, the BR-IXa isomer requires further conjugation to be eliminated from

placental or adult tissues. This presents a problem as fetal tissues contain low levels of glucuronosyl transferase. Thus, it is McDonagh’s argument that unconjugated BR-IXα crosses the placenta due to its requirement to become further conjugated for excretion and that this is the reason that low levels of BR-IXα are detected within fetal tissues. Additionally, it should be noted that numerous studies have indicated the difficult nature of quantifying the contents of bile acids (Colleran and Heirwegh 1979, McDonagh 1979, McDonagh 2010, van den Hurk 2006).

Regardless of which BR isomers can or cannot cross the placenta, the hypothesis that the two separate isoforms of BVR have different ontogenic roles, one to solely confer fetal protection against ROS (BVRb) while the other (BVRa) confers adult protection against ROS, does not account for the widespread distribution of these bile pigments in non-placental animals, such as avians, amphibians and fish, as well as plants and cyanobacteria (McDonagh and Palma 1982,

Cornelius 1991, Lee Shing 1987, McDonagh 2001). To this end, roles for BV and BR other than as antioxidants have been proposed including forming camouflage in egg shells (Keneddy and

Vevers 1973), serving as a biosynthetic precursor in plants (McDonagh 2001), acting as a morphogen in Xenopus (Falchuk et al. 2002, Montorzi, Dziedzic, and Falchuk 2002) and mediating digestion in animals (Qin 2007). Interestingly, a lizard species has been shown to have green blood as a result of extremely high levels of BV (Austin and Jessing 1994) although the physiological benefits or consequences of this are unknown.

1.12 Heme Degradation: Production of Second Messengers and the Potent Antioxidant BR

The discovery of a second HO isoform (HO-2) that is highly expressed in brain tissues was initially unexpected as heme turnover was not typically associated with brain tissues (Barinaga ,

Sun 1990). However, the subsequent finding that nitric oxide (NO) acts as a key chemical

messenger (Culotta and Koshland 1992) prompted intense research looking for other gaseous transmitters. In this regard, it was later found that HO-2 brain expression is localized to regions which also contain guanylyl cyclase (Verma et al. 1993, Maines 1993). As CO stimulates cyclic guanosine monophosphate (cGMP) production, these studies suggested that HO-2 functions to modulate cGMP levels in neural tissues through the production of CO. The messaging ability of

CO, coupled with the previously mentioned finding that BR acts as a strong antioxidant in vitro

(Stocker 1987) and as a peroxyl scavenger in human plasma (Stocker, Glazer, and Ames 1987), resulted in a renewed interest in the heme degradation pathway and planted the foundations for a second explanation regarding the apparently wasteful generation of BR. Levels of BR in the nanomolar range act as a potent neuroprotectant, with 10 nM providing protection to nearly

10,000 fold levels of peroxide in cultured cells (Dore 1999). Barañano et al. proposed a model explaining how such low concentrations of BR can confer such a degree of cellular protection

(Barañano 2002). In this model, BVR utilizes cytosolic NADPH to reduce BV to the potent antioxidant BR. When acting as an antioxidant, BR is oxidized back to BV in a manner analogous to the classic glutathione (GSH) recycling pathway (Figure 1.1), thus providing protection to lipophilic cellular compartments. A comparison of the GSH and BR recycling pathways is shown in Figure 1.2.

Figure 1.2. Antioxidant Recycling Pathways. Reduced glutathione (GSH) confers protection against oxidative stress within cytosolic cellular compartments while bilirubin (BR) is thought to confer protection against oxidative stress in membrane compartments of the cell. When acting as an antioxidant, GSH is oxidized to GSSG which through the action of glutathione reductase (GR) is recycled back to GSH. Similarly, when BR acts as an antioxidant it is oxidized back to biliverdin (BV) which can be recycled back to BR via the action of biliverdin reductase (BVR). BV is generated via endoplasmic reticulum (ER) bound heme oxygenase (HO-1). Electrons from NADPH are transported to HO-1 by cytochrome P450 Reductase (P450) (Schacter et al. 1972). HO-2 and BVRb are omitted from this figure for the sake of clarity. Figure is adapted from (Barañano 2002) and (Huber et al. 2009).

1.13 Arguments against BR Recycling

While micromolar concentrations of BR are effective antioxidants in vitro (Stocker 1987,

Stocker, Glazer, and Ames 1987, Stocker and Ames 1987), the model suggesting that BVR acts to continuously generate BR for the purpose of cellular protection against oxidative stress is the result of only two studies. Dore et al., showed that BR concentrations as low as 10 nM conferred protection to cells against 10,000 fold higher levels of hydrogen peroxide (Dore 1999). In a study by Barañano et al., knockdown of BVR resulted in increased levels of ROS and cell death in HeLa cells providing further support of a BVR recycling antioxidant process (Barañano 2001).

However, the validity of this model has been called into question, notably by the researchers who first described BR as an antioxidant of possible physiological importance (Maghzal et al. 2009,

McDonagh 2010, McDonagh 2004).

Central to the argument against BR recycling are two tenants: 1) BR exists predominantly as an unconjugated pigment bound to albumin rendering it unable to become oxidized and produce

BV, and 2) BR oxidation yields stoichiometric amounts of BV (McDonagh and Dolphin 1979,

Maghzal et al. 2009). When mimicking physiological conditions, Maghzal et al., could not generate BV upon oxidation of BR (Maghzal et al. 2009). Furthermore, they could not observe a difference in response to stress when depleting or overexpressing BVR in Hela cells or when co- expressing human HO-1 and BVR in HO-1 null yeast cells. In response to Maghzal et al.

(2009), Snyder and colleagues cite differences in H2O2 timing and dosing as an explanation for the observed differences in protection conferred by BR (Sedlak 2009a).

1.14 Primary Research Focus on BVRa and the Neglect of BVRb

The scientific literature is dramatically skewed in favor of BVRa in attempts to identify novel therapeutic roles. This may be the result of the unique biochemical and molecular properties of

BVRa which displays a dual cofactor dual pH specificity, utilizing either NADH (pH 7.0) or

NADPH (pH 8.7) (Kutty and Maines 1981). Furthermore, BVRa is a rare serine, threonine, tyrosine kinase (S/T/Y) and contains numerous regulatory motifs including the basic leucine zipper DNA binding domain (bZip), nuclear export and localization signals (NES, NLS), and

SH2 recognition sites, which collectively allow BVRa to interact with other proteins and participate in insulin signaling and mitogen-activated protein kinase (MAPK) pathways

(Kapitulnik 2008, Lerner-Marmarosh 2005, Lerner-Marmarosh 2008). Whereas BVRa only

reduces the BV-IXα isomer, BVRb is a more promiscuous enzyme and reduces numerous BV isomers (BV-IXγ, BV-IX-β, BV-IXδ (McDonagh 2001) as well as multiple flavins, pyrroloquinoline quinone (PQQ), and ferric ion (Cunningham, Gore, and Mantle 2000, Xu et al.

1993, Shalloe 1996). Further differences in the structure and catalytic activity of BVRa and

BVRb are diagrammed in Figure 1.3.

Figure 1.3. Differences in the Structure and Enzymatic Activity of BVRa and BVRb. BVRa contains numerous regulatory motifs including bZip (Ahmad, Salim, and Maines 2002), YMKM, NES, NLS, SH2 recognition, C-box (high affinity ERK binding site), and D-box (binding site for MAPK kinases and substrates)(Salim 2001, Gibbs, Tudor, and Maines 2012). BVRb is also known as flavin reductase, green heme-binding protein, methemoglobin reductase, NADPH dehydrogenase and diaphorase (Shalloe 1996, Xu 1992, Chikuba 1994).

1.15 Regulation of the Heme Degradation Pathway

The first enzyme in the heme degradation pathway, HO-1, is highly inducible by oxidative stress and is regulated by the redox sensitive transcription factors NRF2 and HIF-1α (Lee 1997, Alam

1994), as well as nuclear factor–κB (NFκB) and activator protein-1 (AP-1) (Alam 2007).

Conversely, the HO-2 isoform is typically considered to be non-inducible. However, the promoter region of the HO-2 gene contains a functional glucocorticoid response element (Raju

1997) and has been confirmed to be regulated by glucocorticoid levels in rat testes (Liu 2000).

Regarding the transcriptional regulation of BVRa and BVRb, it has been suggested that BVRb

may be regulated by NRF2 (Wu 2011, Moon 2012) and GATA-1 (Galloway 2005). However, these studies are not definitive and failed to evaluate whether BVRa is also regulated by NRF2 or

GATA-1. Both rat and human BVRa (hBVRa) promoters contain regulatory elements for NFκB and HIF-1 (Gibbs, Miralem, and Maines 2010) which is in agreement with hBVRa being activated by hypoxia (Salim 2001) and inhibited by NFκB (Gibbs, Miralem, and Maines 2010).

Additionally, computer analysis of the hBVRa promoter suggested numerous candidate regulatory elements including AHR and HSF (Gibbs, Miralem, and Maines 2010). To my knowledge, so far to date there are no definitive studies which have characterized the BVRb promoter.

1.16 Roles for HO and BVR in Hematopoiesis

HO-1 seems to be extremely important for embryonic survival as the majority of HO-1-/- mice die in utero (Poss and Tonegawa 1997a) . Furthermore, patients heterozygous for HO-1 or patients with polymorphisms within the HO-1 promoter are at increased risks for spontaneous miscarriage (Yachie et al. 1999, Denschlag et al. 2004). HO-1 has been shown to be expressed in macrophages in rat fetal liver (Watanabe et al. 2004) (the yolk syncytial layer and blood cells in zebrafish (Thisse et al. 2004), and developing mouse erythroid cells (Garcia-Santos et al.

2014) . Furthermore, HO-1-/- mice have an increased number of circulating RBCs with an abnormally long lifespan due to decreases in macrophages which engulf aged RBCs (Fraser et al.

2015). Collectively, HO-1 expression in differentiating erythroid cells (Garcia-Santos et al.

2014) and the increased lifespan of RBCs in HO-1-/- mice (Fraser et al. 2015) suggests a role for

HO-1 in maintaining RBC turnover.

Bioinformatic analysis has shown some GATA binding motifs to be located within the hBVRa promoter (Gibbs, Miralem, and Maines 2010). As GATA-1 is the master regulator of erythroid development, this suggests that hBVRa may be regulated in some capacity by members of the GATA transcription factor family. However, there is no direct evidence that hBVRa is involved in hematopoiesis. Furthermore, this study failed to define which of the six known transcription factors within the GATA family are thought to actively bind to these response elements (Gibbs, Miralem, and Maines 2010). However, it has been shown that GATA-1 knockdown in zebrafish results in a loss of BVRb expression, suggesting a role for BVRb in hematopoiesis (Galloway 2005).

1.17 Transcriptional Regulators of Hematopoiesis

Numerous transcription factors regulate the hematopoietic program and are expressed in multiple waves in a tissue specific fashion (Akashi et al. 2000). However, two main transcription factors are traditionally associated with determining whether a hematopoietic stem cell (HSC) differentiates into a myeloid cell or an erythroid cell. GATA-1 is the main transcription factor associated with erythrocyte development (Ferreira et al. 2005). The GATA family of transcription factors are divided into two subfamilies: 1) GATA-1, GATA-2, and GATA-3 which are mainly expressed within the hematopoietic system (Weiss and Orkin 1995) and 2) GATA-4,

GATA-5, and GATA-6 which are expressed in various tissues (Molkentin 2000). All six of these transcription factors contain two zinc finger motifs (Ko and Engel 1993, Martin and Orkin

1990, Merika and Orkin 1993, Whyatt, deBoer, and Grosveld 1993, Yamamoto et al. 1990), an

N-terminal finger which is involved in stabilizing DNA binding (Ghirlando and Trainor 2003,

Martin and Orkin 1990, Trainor, Ghirlando, and Simpson 2000, Whyatt et al. 1997) and a C-

terminal zinc finger which recognizes and binds the consensus sequence WGATAR (where W =

A/T and R=A/G) (Martin and Orkin 1990, Yang and Evans 1992). Using mouse model systems and knockout technology it has been shown that GATA-1 is necessary for normal erythroid development (Fujiwara et al. 1996, McDevitt et al. 1997, Pevny et al. 1995, Pevny et al. 1991,

Weiss, Keller, and Orkin 1994, Weiss and Orkin 1995) and this role is conserved in zebrafish

(Detrich et al. 1995, Thompson et al. 1998).

Conversely, PU.1 (also known as SPI-1) serves as the main transcriptional regulator of myeloid and lymphoid lineages (Hromas et al. 1993, Klemsz et al. 1990, Scott et al. 1994). It is a member of the large E26 transformation-specific (ETS) transcription factor family that is involved in the development of numerous tissues. PU.1 interacts physically with GATA-1 and it is through this interaction that PU.1 and GATA-1 antagonize each other (Cantor and Orkin

2002). Like GATA-1, and in a broader sense vertebrate hematopoiesis, studies of PU.1 were performed in mouse models. However, it was studies utilizing zebrafish (Danio rerio) which demonstrated conservation of the vertebrate hematopoietic program (Detrich et al. 1995,

Thompson et al. 1998). These studies confirmed the mechanisms of GATA-1 and PU.1 cross antagonism, and substantially enhanced our understanding of the complex regulatory pathways involved in blood cell differentiation (Galloway 2005, Monteiro, Pouget, and Patient 2011,

Ransom 2004, Rhodes et al. 2005).

A third player in the differentiation process is transcriptional intermediary factor 1γ (TIF1γ)

(also known as TRIM33). The zebrafish moonshine mutant, which has a defective copy of tif1γ, is severely anemic and only forms erythroid progenitor cells (Ransom 2004). Using the moonshine mutant, Monteiro et al., drastically changed the accepted model of how erythroid/myeloid fate decisions are made and established that TIF1γ is a crucial player in these

decisions (Monteiro, Pouget, and Patient 2011). Previously, the accepted model involved the antagonism between GATA-1 and PU.1, and stated that these proteins positively auto-regulate themselves (Rhodes et al. 2005). However, Monteiro et al., showed that the resulting fate outcomes are not only dependent on the interactions between GATA-1 and PU.1, but also on the spatial and temporal contexts of their interactions which are modulated by TIF1γ (Monteiro,

Pouget, and Patient 2011). The current model of hematopoietic development in zebrafish is summarized in Figure 1.4.

1.18 Zebrafish Hematopoietic Stem Cell Differentiation

Blood cell development occurs in three waves, a primitive wave, a pro-definitive wave and a definitive wave (Ciau-Uitz et al. 2014). Early in development (~14 hpf) transcription factors necessary for blood and endothelial cell formation are expressed within the blood islands of the anterior lateral mesoderm (ALM) and the posterior lateral mesoderm (PLM), which produce myeloid and erythroid cells, respectively (Bennett et al. 2001, Davidson et al. 2003, Amigo et al.

2009). These primitive blood cells arise within a region termed the intermediate cell mass (ICM)

(Detrich et al. 1995, Thompson et al. 1998) and begin to actively circulate at ~24 hpf, marking the onset of the primitive wave. The second wave, the pro-definitive wave begins with the production of multipotent erythomyeloid progenitors (EMPs) from the posterior blood island

(PBI) (Bertrand et al. 2007). These EMPs subsequently migrate to the caudal hematopoietic tissue (CHT), a transient location in which they differentiate into erytho-myeloid and lymphoid lineages (Murayama et al. 2006, Jin 2007). Although controversial, a subset of these EMPs from the PBI are thought to directly seed the kidney marrow, effectively by-passing the transient stage within the CHT (Bertrand et al. 2008).

Figure 1 4. Three Main Transcription Factors Regulate Hematopoietic Development. Transcription factors needed for blood and endothelial cell formation are expressed within the blood islands of the ALM (PU.1-myeloid lineage) and the PLM (GATA-1-erythroid lineage) early in development (~14 hpf). During the primitive wave of hematopoiesis (~24 hpf) embryonic RBC’s arise from the ICM and begin to circulate, primarily within the yolk sac circulation valley. Within the ICM, GATA-1 negatively regulates PU.1, both GATA-1 and PU.1 positively autoregulate themselves, and TIF1γ is required for the expression of both transcription factors. During the pro-definitive wave (~24-30 hpf) EMPs emerge from PBI and “seed” the CHT. HSC’s also arise from the dorsal aorta (DA) and seed the CHT. Finally, during the definitive wave some HSC’s within the CHT differentiate into erythroid and myeloid cells while others migrate to the kidney marrow which is the location of adult hematopoiesis. During this wave, TIF1γ and GATA-1 inhibit PU.1 expression which increases RBC levels. Furthermore, during this time period GATA-1 acts as a repressor, inhibiting TIFγ as well as negatively repressing itself. Abbreviations used are as follows: ALM, anterior lateral mesoderm; PLM, posterior lateral mesoderm; hpf, hours past fertilization; ICM, inner cell mass; EMPs, erythromyeloid progenitors; PBI, posterior blood island, HSC’s, hematopoietic stem cells, CHT, caudal hematopoietic tissue, RBC's, red blood cells. Figure is adapted from (Monteiro, Pouget, and Patient 2011, Ciau-Uitz et al. 2014).

In the definitive wave, true HSCs arise from the hemogenic endothelium (localized in the ventral wall of the dorsal aorta (DA) of the aorta/gonad/mesonephros region (AGM) in a process termed the “endothelial-to-hematopoietic transition” (EHT) (Bertrand et al. 2010, Kissa and

Herbomel 2010, Lam et al. 2010). While HSCs involved in the EHT arise from the DA, they do not begin to further differentiate until they enter into circulation and seed the CHT, where they will differentiate into erythroid and myeloid lineages (Medvinsky, Rybtsov, and Taoudi 2011,

Chen and Turpen 1995, Murayama et al. 2006, Jin et al. 2009, Monteiro, Pouget, and Patient

2011). Finally, HSCs from the CHT will migrate and seed the kidney marrow (Kissa et al. 2008,

Jin 2007), the adult hematopoietic organ.

1.19 Conclusion

While primarily known for their roles in heme catabolism, the widespread distribution of HO and

BVR in nature suggests roles for these isoforms other than facilitating the elimination of bile acids in placental mammals. However, there are no studies which have thoroughly investigated the evolutionary and developmental aspects of HO and BVR. The central hypothesis of this dissertation is that HO and BVR serve as regulators of normal embryonic development in addition to their well characterized roles in cellular response to toxicant induced oxidative stress.

This study aims to provide the most comprehensive developmental overview of these genes which are primarily known for their roles in heme catabolism. In Chapter 2 we quantitatively characterize the ontogenic expression of HO and BVR under normal cellular conditions and in response to oxidative stress. To gain further insights into novel functions for these genes involved in heme metabolism, in Chapter 3 we characterize the spatial and temporal expression of these genes during development. Finally, in Chapter 4 we functionally evaluate two potential transcriptional regulators of HO and BVR during normal development and in response to stress.

CHAPTER 2

ONTOGENIC EXPRESSION OF HO AND BVR IN RESPONSE TO OXIDATIVE STRESS

2.1 Abstract

To further characterize the developmental and protective roles of zebrafish heme degradation genes we evaluated their expression patterns during development and within adult tissues under normal conditions and in response to differing inducers of oxidative stress. Using quantitative

RT-PCR (qRT-PCR), we identified significant sex-specific differences in basal and induced gene expression of heme oxygenase (HO) and biliverdin reductase (BVR) isoforms in adult tissues of male and female zebrafish under control and cadmium (Cd) induced conditions. We also determined developmental timepoints of tolerance and sensitivity towards xenobiotic induced oxidative stress from Cd and tert-butylhydroquinone (tBHQ) exposure. This was followed up by documentation of the comparative developmental expression of these same genes under control conditions, as well as in response to three different pro-oxidants (Cd, tBHQ and hemin). Finally, we evaluated if genes involved in maintaining iron homeostasis (ferroportin 1, hepcidin, and hemopexin) responded to Cd induced stress in the same manner as genes involved in heme degradation. Collectively, these data suggest that genes involved in heme degradation may be regulated in some capacity by transcription factors involved in maintaining iron homeostasis or in the response to oxidative stress. In this chapter we provide a thorough characterization of ontogenic expression of heme degradation genes in zebrafish. We also provide the first

evaluation of developmental and tissue specific expression of zebrafish heme oxygenase paralogs, providing the first important step towards evaluating the potential for subfunction partitioning.

2.2 Introduction

Heme molecules are essential for life, having roles in cellular differentiation, apoptosis and serving as a cofactor for numerous proteins (Larsen et al. 2012). However, unbound cellular heme is highly reactive and can generate damaging reactive oxygen species (ROS) via Fenton chemistry. Two isoforms of the heme-degrading enzyme, heme oxygenase 1 and 2 (HO-1 and

HO-2), catalyze the breakdown of heme into carbon monoxide (CO), free iron, and the bile pigment biliverdin (BV) (Tenhunen, Marver, and Schmid 1969, Abraham and Kappas 2008).

Biliverdin reductase a (BVRa) and biliverdin reductase b (BVRb) further reduce BV to the proposed antioxidant bilirubin (BR) (Stocker 1987, Yamaguchi 1994, Stocker, Glazer, and Ames

1987). It has been suggested that BVRa functions to protect cellular lipophilic components from oxidative stress through the production of BR via an antioxidant recycling pathway similar to the classic glutathione recycling pathway (Dore 1999, Barañano 2002, Sedlak 2004, Sedlak 2009b).

The action of BR as an effective antioxidant is notable for its potential benefits to cellular health.

However, an excess of BR can be problematic resulting in jaundice and toxic insult (Greenberg

2002). Although heme and BR have important cellular roles, the potential toxicity associated with these two molecules illustrates the necessity for tight regulation of this enzymatic pathway.

While the heme degradation pathway has been widely studied, much of the research has focused primarily on HO-1 and BVRa in mammalian systems with little focus on developmental roles. To this end, HO-1 has been shown to be highly inducible by a variety of stressors in

mammalian systems, and is detected at high levels in tissues involved in heme metabolism including spleen and liver (Tenhunen, Marver, and Schmid 1969), bone marrow (Abraham et al.

1989, Brown et al. 1988) and erythroid cells (Garcia-Santos et al. 2014). The other heme degrading enzyme, HO-2, is expressed at high levels in the testes (Trakshel, Kutty, and Maines

1986), brain (Trakshel, Kutty, and Maines 1988, Maines 1986, Ewing and Maines 1992), and vascular tissues (Zakhary et al. 1996). Although HO-2 is generally regarded as being constitutively expressed and non-inducible (Maines 1986), this isoform has been shown to be activated by corticosterone, opiates and menadione (Liu 2000, Maines 2005, Vukomanovic et al.

2011).

The BVR isoforms are distinct in their enzymatic actions and their evolutionary origins

(Yamaguchi et al. 1993). BVRa has been shown to be highly expressed in rat kidney, liver, spleen and brain tissues (McCoubrey, Cooklis, and Maines 1995), as well as human liver

(Maines and Trakshel 1993) and the plasma membrane of macrophages (Wegiel et al. 2009).

BVRb, also known as flavin reductase, green heme binding protein (Shalloe 1996), methemoglobin reductase, erythrocyte reductase, NADPH dehydrogenase, and diaphorase (Xu

1992) has been demonstrated to be abundant in adult bovine, rat, rainbow trout, and human erythroid cells, as well as rat and bovine liver (Shalloe 1996, Xu 1992, Saleh and McConkey

2012). Although both BVRa and BVRb generate BR, the published research is highly skewed toward BVRa while the BVRb isoform is typically not included in studies related to the antioxidant properties of this pathway. One explanation for this omission may be the result of the unique properties of BVRa. Although both BVR enzymes contain the reductase domain,

BVRa’s domain has dual cofactor (NADH and NADPH) dual pH specificity. BVRa also has several regulatory domains not found in BVRb, such as a serine/threonine/tyrosine kinase

(S/T/Y) domain (Kutty and Maines 1981, Salim 2001, Lerner-Marmarosh 2005), a leucine zipper

DNA-binding domain (Ahmad, Salim, and Maines 2002), as well as nuclear export and localization signals (Maines et al. 2001, Lerner-Marmarosh 2008). However, while there appears to be a high degree of sequence similarity between BVRa proteins throughout nature

(Maines 2007), the extent to which these unique functional attributes of BVRa are conserved has not been rigorously examined. These discrepancies highlight the need for further studies in other model systems.

Due to their rapid transparent development and sequenced genome, zebrafish (Danio rerio) have become an excellent model for developmental and toxicological studies (Hill 2005,

Blechinger 2007, 2002, Howe et al. 2013). Furthermore, zebrafish often have multiple homologues of human genes (Meyer 2005) resulting from a teleost-specific whole genome duplication event (Amores et al. 1998), some of which have undergone subfunction partitioning

(Postlethwait et al. 2004). This is advantageous for ascertaining the myriad functions of a gene in which its human counterpart has multiple roles (Amores et al. 1998). In this regard, purported duplicates of HO-1 and HO-2 have been identified within the zebrafish genome (Nakajima and

Mukaigasa 2011). However, a full characterization of the expression patterns and response to cellular stress of these various isoforms is lacking.

As an initial step towards understanding novel developmental and protective roles of zebrafish heme degradation genes, we set out to quantitatively evaluate differences in their expression during early development and in adult tissues under normal and stressful conditions.

We also wanted to determine how these genes responded to different inducers of oxidative stress during development (Cd, hemin, and tBHQ), and if any changes in HO and BVR gene expression correlated with temporal sensitivity to Cd or tBHQ induced oxidative stress. As

previous studies from this lab supported novel roles for MTF-1 as a regulator of iron homeostasis

(O'Shields et al. 2014), we also wanted to determine if genes involved in maintaining cellular iron levels (ferroportin 1, FP1; hepcidin, HaMP; and hemopexin, HPX) responded to metal induced stress in a manner similar to HO and BVR. Interesting differences in developmental expression and adult tissue distribution, as well as responses to oxidative stress were noted. The results of these experiments and their implications are presented here.

2.3 Materials and Methods

2.3.1 Chemicals

Cadmium (Cd) chloride (CAS # 654054-66-7) and tert-butylhydoquinone (tBHQ) (CAS #1948-

33-0) were purchased from Sigma-Aldrich (St. Louis, MO). Hemin chloride (CAS # 16009-13-5) was purchased from EMD Millipore.

2.3.2 Fish Husbandry

The TL (Tupfel/Long fin mutations) wild-type strain of zebrafish was used for all experiments.

Fertilized eggs were obtained from multiple group breedings from a Mass Embryo Production

System (MEPS; Aquatic Habitats, Apopka, FL) with ~200 fish at a ratio of 2 female per 1 male fish. Procedures used in these experiments were approved by the Animal Care and Use

Committee of the University of Alabama, Tuscaloosa, Alabama, USA.

2.3.3 Phylogenetic Analysis

HO homologs were identified using mammalian HO protein sequences to search the zebrafish genome using BLAST. Proteins were aligned using ClustalW and a phylogenetic tree was

constructed using the maximum likelihood method and the WAG model of amino acid substitution followed by likelihood calculation using the GAMMA model. To assess confidence in individual nodes, a bootstrap analysis with 1000 re-samplings was performed. The consensus bootstrap tree was rooted to the Drosophila melanogaster HO protein. All phylogenetic analysis were performed in MEGA6 (Hall 2013).

2.2.4 Cadmium Exposures in Adult Zebrafish

Adult male and female zebrafish approximately one year in age were subjected to 20 μM Cd for

96 hours (Ali, Mil, and Richardson 2011). Adult fish were grouped in a 4L beaker (3 male and 3 female) with 3L of fish water (360 mg Instant Ocean Sea Salt and 11 mg of NaHCO3 per liter of deionized water; 7.4 pH, 750 µ-siemens conductivity, 28.5 C) or fish water with 20 µM Cd.

Water was continuously aerated with a small air stone, and both control water and the water with

Cd were renewed after 48 hours. Adult zebrafish were anesthetized with 150 mM MS-222 in buffered fish water and euthanized by cervical transection. Brain, liver, and gill tissues were removed from male and female fish and flash frozen in liquid nitrogen. Tissues were stored at -

80 C until RNA isolation. Three biological replicates were collected for each sex and treatment group and consisted of equivalent amounts of total RNA pooled from three individual male or female fish.

2.3.5 Developmental Time Series

A large batch (>3000) of embryos were generated by active breeding of the fish in the MEPS system for 60 minutes. After 60 minutes embryos were collected and transferred to large Petri dishes (150 mm diameter) at a density of 100 embryos per 100 ml of 0.3x Danieau’s solution at

28.5 °C. Embryos were screened for developmental abnormalities within 24 hours of fertilization and subsequently separated into 15 small Petri dishes (100 mm in diameter) with 20 embryos per dish in 25 ml of 0.3x Danieau’s solution at 28.5 °C under a 14 hour light/10 hour dark cycle with water changes every 24 hours. Three replicates of 20 pooled embryos were collected at several developmental timepoints (24, 48, 72, 96, and 120 hpf). Embryos were flash frozen in liquid nitrogen and stored at -80°C until RNA isolation.

2.3.6 Toxicant Exposures during Zebrafish Development

Acute 4 hour challenges with Cd (150 µM), hemin (100 µM or 150 µM), or tBHQ (10.0 μM) were performed starting at 72 hpf (3 replicates of 20 pooled embryos). For continuous Cd exposures (4 replicates of 25 pooled embryos), zebrafish embryo-larvae were exposed to different concentrations of Cd (50, 100, 150 µM) for a 24 hour period beginning at 24 hpf and continuing through 96 hpf. Both control water and the water with Cd were renewed every 24 hours. Following all exposures, embryos were flash frozen in liquid nitrogen and stored at -80°C until RNA isolation.

2.3.7 Rapid Amplification of cDNA Ends for HO-1b Transcript Analysis

Total RNA was collected from embryos from several different developmental timepoints and pooled together for cDNA synthesis. A Marathon® cDNA Amplification Kit (BD Biosciences,

Palo Alto, CA, U.S.A.) was used to synthesize double stranded zebrafish cDNA according to manufacturer’s protocols. Adaptor primers were used with an HO-1b gene specific primer

(Table 2.1) with the following PCR parameters: 95 °C for 3 min, 95 °C for 1:00 min/58 °C for

45 s/68 °C for 1:00 min (30 cycles). The secondary PCR product was gel purified using a

QIAGEN® Qiaex II Gel Extraction Kit and cloned into the pGEM®-T Easy Vector (Promega,

Madison, WI) for sequencing.

2.3.8 Gene Expression Assessment by Real-time RT-PCR

Total RNA was prepared from embryos and adult tissues using the TRIzol® reagent (Life

Technologies, Invitrogen) protocol. The qScript™ cDNA Synthesis Kit (Quanta Biosciences,

Gaithersburg, MD) was used to generate cDNA from 1 µg total RNA per manufacturer’s instructions. PerfeCTa® SYBR® Green Supermix for iQ™ (Quanta Biosciences, Gaithersburg,

MD) was used for real-time RT-PCR experiments in a MyiQ2 Two-Color Real-Time PCR

Detection system (Bio-Rad, Hercules, CA) under the following conditions: 95°C for 3 min, 95°C for 15s/60°C for 30 sec (40 cycles). To ensure that only a single product was amplified, all real- time RT-PCR experiments were followed with a melt curve analysis. Gene expression for the developmental time series, as well as the adult tissue analysis, was quantified by generating a standard curve using serially diluted plasmids containing full length copies of the transcripts.

For the acute Cd, hemin and tBHQ exposures, changes in transcript expression were determined using the comparative Ct method (Schmittgen 2008).

2.3.9 Statistical Analysis of Gene Expression Profiling

Statistical analysis was performed using Prism 5 software (GraphPad Software Inc., San Diego,

CA). For the developmental time series, statistical significance in transcript expression compared to 24 hpf embryos was determined using one-way analysis of variance (ANOVA) followed by a Dunnett’s post hoc test. For the adult tissue experiments, statistical significance in transcript expression between controls and treatment groups was determined using a two-way

ANOVA followed by a Bonferroni post hoc test (*p-value < 0.05). For Cd, tBHQ, and hemin zebrafish embryo exposures, changes in gene expression were calculated using the comparative

Ct method, and statistical significance in comparison to control embryos was determined using one-way ANOVA followed by a Dunnett’s post hoc test.

2.3.10 in Silico Promoter Analysis

We searched 4 kb upstream of the transcription start site and within the first intron of zebrafish

HO-1a, HO-2a, HO-2b, BVRa, and BVRb genes for consensus metal response element (MRE)

MTF-1 binding motifs (TGCRCNC) and consensus antioxidant response element (ARE) NRF2 binding motifs (TTGAYNNNGC) using ApE software by M. Wayne Davis. The promoter region of zebrafish HO-1b is not available at this time.

Table 2.1 qPCR and 5’ RACE PCR Primers

Gene qPCR Sequence HO-1a Forward 5-GCTTCTGCTGTGCTCTCTATACG-3 Reverse 5-CAATCTCTCTCAGTCTCTGTGC -3 HO-1b Forward 5’-GCAGTGATCTGTCTGAACAG-3 Reverse 5-GCTTGTACTGTGTTTGTGTG-3 HO-2a Forward 5-ATGGCGGTCAGTGGAAACACAACC-3 Reverse 5-GGCAACAGCAGCAACCAATGTGGC-3 HO-2b Forward 5’TTTAGGAGGTTGAGTTGGAGTCAG-3 Reverse 5-TTCTGCCTTCTGGTGCACTTCT-3 BVRa Forward 5-CAGGCAGTTTCTGGAGGCAGG-3 Reverse 5-CCAGACCCTTCTGTTGAGC-3 BVRb Forward 5’GCATGTCAGCATTCCTCTTGTGG-3 Reverse 5-CACCAGCAATATGTGGAGG-3 HPX Forward 5-GATGGCCATTTCTACATGATCAAGGACA-3 Reverse 5-GCCCTCAATTCCCAGCACATCC-3 HaMP Forward 5’CACAGCCGTTCCCTTCATACAGCA-3 Reverse 5-GGTCTGCTAGTCTGTGTTCAGCTTC-3 FP1 Forward 5-GTCCTACATTCATTCCTACAACTGAACC-3 Reverse 5-GTCAAGTCGAAGGACCAAAGACCAACT-3 NNT Forward 5-GCGCTGAAGGCTTCCTGCTGAAT-3 Reverse 5-AGGGATCGGTTCATGGCCACACA-3 IDHc Forward 5-ATGGGAGCTCATAAAAGAGAAACTCAT-3 Reverse 5-CCTCAACTGTTACCTTATCATCGGTG-3 IDHm Forward 5-ATCTGGGAGTTCATTAAAGAAAAGCTT-3 Reverse 5-GTAATAGTGGCGCATTTGACAGCAACG-3 Gene 5’ RACE PCR Sequence HO-1b Reverse 5-CGCCTCGTAGATCTTGTAGAGC-3

2.4 Results

2.4.1 5' RACE PCR and Phylogenetic Analysis of Heme Oxygenase

Initial protein alignments of zebrafish HO-1a and HO-1b showed a high degree of conservation at the C-terminal region of the protein. However, the HO-1a isoform is 272 amino acid residues in length while the initial transcript for HO-1b only encoded for 247 amino acids and was missing part of the heme oxygenase domain found in the N-terminal region of the protein. Using

5' RACE PCR we were able to deduce another 22 amino acids. Alignment with the HO-1a sequence suggests that the original HO-1b sequence was deduced from a transcript that was missing the 5' untranslated region (UTR) and initial coding region that is found in exon 2 (Figure

2.1).

Figure 2.1. Zebrafish HO-1a and HO-1b Contain Conserved Heme Oxygenase Domains and Heme Signature Motifs. Zebrafish HO-1a is composed of 7 exons. The heme oxygenase domain (HO Domain) spans exons 3-5, while the highly conserved heme oxygenase signature motif which is necessary for heme binding is located within exon 5. Shown above is a multiple sequence alignment of HO-1 proteins from fish (goldfish and zebrafish) and human. Sequence alignment was generated using Clustal W. Shading highlights regions with differing levels of sequence identity. The HO Domain (red line) and the HO Signature (blue box) are denoted by lines above their respective sequences. Zebrafish HO-1a: NP_001120988.1; Zebrafish HO-1b: NP_991234.1; Human HO-1: NP_002124.1; Goldfish HO-1 GenBank: AHI15729.1

Using this new sequence for zebrafish HO-1b, a phylogenetic analysis was performed to better understand the relationship between zebrafish and other vertebrate HO proteins.

Phylogenetic analysis suggests conservation amongst HO-1 proteins and HO-2 proteins in humans, rats, and chickens (Figure 2.2). HO-1a groups more closely to other fish HO-1 proteins in comparison to vertebrate homologues. Likewise, zebrafish HO-2a and HO-2b show a closer relation in comparison to other vertebrate HO-2 homologues. HO-2 has not been identified in goldfish (Carassius auratus).

Figure 2.2. Phylogenetic Analysis of Heme Oxygenase Proteins. Protein alignments and tree construction were performed in MEGA6. Evolutionary history was inferred by using the Maximum Likelihood method. The Bootstrap method was used (1000 replicates) for the Test of Phylogeny. Drosophila HO (NP_524321.1) was used as an out-group. HO-1 proteins: Zebrafish HO-1a: NP_001120988.1; Zebrafish HO-1b: NP_991234.1; Pufferfish HO-1a: UniProtKB - O73688; Pufferfish HO-1b: GenBank: CAF95107.1; Human HO-1: NP_002124.1; Rat HO-1: NP_036712.1; Chicken HO-1:NP_990675.1; Goldfish HO-1 GenBank: AHI15729.1. HO-2 proteins: Zebrafish HO-2a: NP_001096609.1; Zebrafish HO-2b: XP_002661145.1; Pufferfish HO-2a: GenBank: CAG00172.1; Pufferfish HO-2b: Ensemble: ENSTNIP00000012632.1m; Human HO-2: NP_002125; Rat HO-2: NP_077363.1; Chicken HO- 2: XP_414960.1.

2.4.2 Comparative Expression of HO and BVR in Adult Tissues

Real-time RT-PCR was used to characterize the differential expression of the HO and BVR isoforms in liver, gill and brain tissues of adult male and female zebrafish. HO-1a and HO-1b were differentially expressed in all three tissues in male zebrafish, while HO-1a and HO-1b were only differentially expressed the brain and gills of females (Figure 2.3). Furthermore, liver expression of HO-1a was the only HO-1 isoform to be differentially expressed between males and females. However, HO-2a and HO-2b were differentially expressed between males and females in liver tissue. There was a statistically significant difference in HO-2a and HO-2b expression in the brains of both male and female zebrafish. Finally, HO-2a and HO-2b expression levels were significantly different in the gill tissue of male zebrafish. Although a similar trend was observed in females, it was not a statistically significant difference.

Figure 2.3. HO Expression in Adult Zebrafish Tissues. Relative abundance of HO-1a, HO-1b, HO-2a and HO-2b transcripts were determined in male and female liver, gill, and brain tissues. Statistical significance in HO transcript expression between males and females was determined using two-way ANOVA followed by a Bonferroni post hoc test (p-value < 0.05). The asterisk designates statistical significance between males and females. Different letters represent statistical significance between paralogous genes. n = 3 biological replicates.

There was no significant difference in either BVRa or BVRb expression levels in gill tissues of either male or female zebrafish (Figure 2.4). However, BVRa expression in brain tissue was significantly higher in both males and females compared to BVRb expression. Although there was no significant difference between BVRa and BVRb expression in female liver tissue, BVRb was expressed at a significantly higher level compared to BVRa in male liver tissues. Finally, there was also a significant difference in BVRb liver expression between males and females

(Figure 2.4).

Figure 2.4. BVR Expression in Adult Zebrafish Tissues. Relative abundance of BVRa and BVRb transcripts were determined in male and female liver, gill and brain tissues. Statistical significance in BVR transcript expression between males and females was determined using two- way ANOVA followed by a Bonferroni post hoc test (p-value < 0.05). The asterisk designates statistical significance between males and females. Different letters designate statistical significance between genes. n = 3 biological replicates.

2.4.3 Inducibility of HO and BVR Expression in Adult Tissues in Response to Cd Exposure

A trend toward induction of HO-1 gene expression was noticeable in all three tissues of male zebrafish (Figure 2.5). This same trend was also present in the brain and liver tissue of female zebrafish. However, due to high variation within the female Cd-treatment the observed

differences were not statistically significant. Interestingly, neither HO-1 isoform appeared to be responsive in female gill tissue (Figure 2.5).

Figure 2.5. HO-1 Induction in Adult Zebrafish in Response to Cd Exposure. Relative fold induction of HO-1a and HO-1b transcripts in response to Cd exposure was determined in zebrafish liver, gill, and brain tissues (20 µM Cd for 96 hours). Statistical significance was determined using a two-way ANOVA followed by a Bonferroni post hoc test (p-value < 0.05). n = 3 biological replicates.

Similar trends in Cd-induced gene expression were also observable for the HO-2 isoforms in adult zebrafish tissues. Although not statistically significant, expression levels of HO-2a were generally higher in brain and liver tissues of male and female zebrafish after Cd exposure (Figure

2.6). In contrast, expression levels of HO-2a were actually reduced in the gill tissues of male and female zebrafish after Cd exposure, although the reduction was not statistically significant. HO-

2b expression was statistically induced in the brain and liver tissue of male zebrafish in response to Cd exposure (Figure 2.6). Once again, there was a nonsignificant trend toward induction of

HO-2b in the gill tissues of male zebrafish. Although the expression levels of HO-2b were

generally higher in all three tissues of female zebrafish in response to Cd, the expression was neither statistically significant nor as prominent as the changes observed in male fish (Figure

2.6).

Figure 2.6. HO-2 Induction in Adult Zebrafish in Response to Cd Exposure. Relative fold induction of HO-2a and HO-2b transcripts in response to Cd exposure was determined in zebrafish liver, gill, and brain tissues (20 µM Cd for 96 hours). Statistical significance between control and Cd-treated fish was determined using a two-way ANOVA followed by a Bonferroni post hoc test (*p-value < 0.05). n = 3 biological replicates.

BVRa and BVRb expression was also assessed in adult tissues under control and Cd-exposed conditions (Figure 2.7). Although there was a general trend toward BVRa and BVRb induction in liver tissue of both male and female zebrafish in response to Cd exposure, the differences were not statistically significant. Expression of BVRa and BVRb in control and Cd-treated fish was not significantly different in gill tissue or either males or females. BVRb expressions levels were not significantly induced by Cd exposure in the brain tissues of male or female zebrafish (Figure

2.7). However, BVRa expression in brain tissue was higher in response to Cd treatment in both sexes, although it was only statistically significant in female brain tissue (Figure 2.7).

Figure 2.7. Adult BVR Expression in Response to Cd Exposure. Relative fold induction of BVRa and BVRb transcripts were determined in zebrafish liver, gill, and brain tissues in control and Cd-exposed fish (20 µM Cd for 96 hours). Statistical significance was determined using two- way ANOVA followed by a Bonferroni post hoc test (*p-value < 0.05). n = 3 biological replicates.

2.4.4 Expression of Heme Oxygenase and Biliverdin Reductase during Zebrafish Development

Quantitative expression profiling of HO and BVR isoforms was performed in zebrafish embryos starting at 24 hpf and continuing in 24 hour intervals until 120 hpf (Figure 2.8). HO-2a levels steadily increased during this developmental time period, with low levels being detected from 24 to 72 hpf followed by a statistically significant increase in expression at 96 and 120 hpf compared to 24 hpf (Figure 2.8). Conversely, HO-2b transcripts were generally higher than HO-

2a and relatively constant, showing no statistically significant changes in expression between 24-

120 hpf. Additionally, we noticed that HO-2b was the most highly expressed HO isoform between 48 and 120 hpf, while HO-1a was the most highly expressed isoform at 24 hpf. HO-1b transcript expression was significantly lower compared to HO-1a and just at the limit of

detection throughout the observed developmental period. BVRa and BVRb displayed similar expression patterns between 24-120 hpf, with both isoforms having peaks in expression at 24 hpf which were statistically significant compared to the other timepoints. However, although the expression patterns shared the same trends, BVRb transcript levels were an order of magnitude higher in comparison to BVRa at all of the assessed timepoints (Figure 2.8).

Figure 2.8. Changes in Expression of HO and BVR between 24-124 hpf. Relative transcript abundance each of transcript was determined via qPCR. Statistical significance in transcript expression in comparison to 24 hpf embryos was determined using one- way ANOVA followed by a Dunnet post hoc test (*p-value < 0.05). n = 3 biological replicates of 20 pooled embryos. hpf = hours past fertilization.

2.4.5 Developmental Sensitivity to Toxicity and NADPH Homeostasis

An initial experiment was performed to determine if there are differences in sensitivity during development to pro-oxidants which differ in their mechanisms of toxicity. Embryos at 24, 48,

72, 96, and 120 hpf were dosed with increasing concentrations of either Cd (25, 50,100, and 150

μM) or tBHQ (2.5, 5, and 10 μM). Cd generates ROS through inhibition of electron transport

(Wang et al. 2004), while tBHQ is a lipophilic molecule which generates ROS through redox cycling (Liu 1998, van Ommen 1992, Dinkova-Kostova 2010). Embryos between ~24-96 hpf were mainly unaffected by Cd exposures (Figure 2.9). However, starting at 96 hpf exposure to

Cd resulted in a dose-dependent increase in mortality, with 100 % mortality occurring at the 150

μM dose. In contrast, embryos displayed greater sensitivity to tBHQ starting at 72 hpf and by 96 hpf 100% mortality occurred in response to all tBHQ concentrations.

A favorable redox environment is necessary for the continued maintenance of cellular antioxidants. NADPH maintains cellular homeostasis by serving as the preferred electron donor in cytosolic redox reactions associated with the amelioration of oxidative stress, and is needed for continuous generation of GSH and BR. To this end, we wondered whether there was a correlation between the timepoints in which zebrafish became increasingly sensitive to Cd or tBHQ and the expression of genes which are responsible for generating cellular NADPH.

Accordingly, we evaluated changes in expression of three major producers of NADPH between

24-120 hpf, i.e., mitochondrial nicotinamide nucleotide transhydrogenase (NNT) (Rydstrom

2006), and cytosolic and mitochondrial isocitrate dehydrogenases (IDHc, IDHm)(Jo et al. 2001,

Jennings and Stevenson 1991, Kirsch 2001). Interestingly, all three genes displayed peaks in expression at 96 hpf, the same time point at which embryos were most sensitive to Cd and tBHQ exposures (Figure 2.9).

Figure 2.9. Zebrafish are More Sensitive to Cd and tBHQ During Peaks in Expression of Genes Responsible for Producing NADPH. Presented is the relative fold change in expression compared to 24 hpf embryos. Statistical significance in expression in comparison to 24 hpf embryos was determined using one-way ANOVA (*p-value < 0.05) followed by a Dunnett’s post hoc test. All values are normalized to 18S ribosomal RNA. n = 3 biological replicates of 20 pooled embryos. Cd = cadmium, DMSO = dimethyl sulfoxide, tBHQ = tert-butylhydoquinone. IDHc = cytosolic isocitrate dehydrogenase, IDHm = mitochondrial isocitrate dehydrogenase, NNT = nicatinamide nucleotide transhydrogenase.

2.4.6 Effects of Acute and Chronic Cd Exposures on Heme Degradation and Iron Homeostasis Genes

Zebrafish embryos become sensitive to Cd and tBHQ starting around 72 hpf (Figure 2.9), the timepoint at which HO and BVR isoforms expression is at their lowest (Figure 2.8). To determine if HO-1a, BVRa and BVRb were inducible during this potential period of heightened

sensitivity, we challenged zebrafish eleutheroembryos starting at 72 hpf with Cd and tBHQ

(Figure 2.10). HO-1a and BVRb were both up-regulated in response to 4 hour exposures to higher concentrations of either Cd or tBHQ, while no significant induction of BVRa was observed for either treatment.

Figure 2.10. Effects of Acute Cd or tBHQ Exposure on Gene Expression in 72 hpf Zebrafish. Embryos were exposed to either Cd or tBHQ for 4 hours starting at 72 hpf. Statistical significance in comparison to control embryos was determined using a one-way ANOVA followed by a Dunnett’s post hoc test (*p-value < 0.05). All values are normalized to 18S ribosomal RNA. n = 3 biological replicates of 20 pooled embryos.

Similar results were obtained when embryos were continuously challenged with increasing concentrations of Cd from 24-96 hpf (Figure 2.11). Continuous Cd exposure resulted in significant increases in HO-1a and BVRb transcripts, while BVRa transcripts remained relatively unchanged. Interestingly, HO-2a was mildly yet significantly upregulated at both the intermediate (100 µM) and the highest (150 µM) concentrations of Cd exposures.

Figure 2.11. Effects of Continually Increasing Cd Exposure on Genes Involved Heme Degradation. Zebrafish at 24 hpf were continuously exposed to increasing concentrations of Cd through 4 dpf. Statistical in comparison to control embryos was determined using a one-way ANOVA followed by a Dunnett’s post hoc test. All values are normalized to 18S ribosomal RNA. n = 3 biological replicates of 25 pooled embryos.

2.4.7 Effects of Cd Exposures on Genes Involved in Regulating Cellular Iron Levels

We wondered if genes involved in maintaining iron homeostasis responded to oxidative stress in the same manner as genes involved in heme degradation (Figure 2.10 and 2.11). Neither 4 hour exposures starting at 72 hpf nor continuous exposure to Cd starting at 24 hpf significantly

altered the expression of ferroportin 1 (FP1), a major iron efflux transporter (Figure 2.12).

However, short term (4 hr) exposures did show a trend towards a dose dependent increase in expression. Similarly, hepcidin (HaMP), the master regulator of iron homeostasis, was not induced by continuous Cd exposure (Figure 2.12). However, HaMP was significantly upregulated following the 4 hour exposure at the highest Cd concentration. Interestingly, the heme binding protein, hemopexin (HPX), responded differently to the Cd exposures, i.e., 4 hour exposures starting at 72 hpf resulted in significant up-regulation while long term exposures resulted in significant down-regulation of HPX regardless of the Cd concentration (Figure 2.12).

Figure 2.12. Effects of Short Term and Continuous Cd Exposures on Genes Involved in Iron Homeostasis. Embryos were exposed to either 50 μM, 100, or 150 μM Cd for 4 hours starting at 72 hpf (top) or continuously exposed to increasing concentrations of Cd starting at 24 hpf and continuing through 120 hpf (bottom). Statistical significance in comparison to control larvae was determined using a one-way ANOVA followed by a Dunnett’s post hoc test. All values are normalized to 18S ribosomal RNA. n = 3 biological replicates of 20 pooled embryos for 4 hour Cd exposures. n = 4 biological replicates of 25 pooled embryos for continuous Cd exposures.

2.4.8 Comparative Expression of HO and BVR in Response to Hemin During Early Development

Hemin is a porphyrin which can participate in Fenton reactions by releasing redox-active iron

(Robinson et al. 2009). To this end, we wanted to compare changes in expression of heme degradation genes between Cd and hemin challenged embryos. Zebrafish larvae (72 hpf) were challenged with hemin (100 and 150 M) for four hours to quantitatively assess changes in expression of HO and BVR isoforms in response to cellular stress. Hemin exposure resulted in statistically signifigant increases in HO-1a, HO-1b and BVRb expression (Figure 2.13). In contrast, neither of the HO-2 paralogs nor the BVRa isoform responded to hemin challenges during early development (Figure 2.13).

Figure 2.13. Effects of Acute Hemin Exposure on HO and BVR Expression. Real-time RT-PCR was used to quantify changes in HO and BVR expression following exposure to hemin (4 hour exposure starting at 72 hpf). Statistical significance in comparison to control larvae was determined using a one-way ANOVA followed by a Dunnett’s post hoc test (*p-value < 0.05). All values are normalized to 18S ribosomal RNA. n = 3 biological replicates of 20 pooled embryos.

2.4.9 MTF-1 and NRF2 Promoter Motif Diversity in HO and BVR Genes

Given the novel observations of BVRa, BVRb and HO-2a and HO-2b induction in response to metals or other pro-oxidant exposures, the 5 kb region upstream of exon 1, as well as the entire region of intron 1, for each gene was searched for the binding motifs for MTF-1 and NRF2, the master regulators of metal and oxidative stress response, respectively (Figure 2.14). The HO-1a gene regulatory region contained several NRF2-binding (ARE) sites. In contrast, the HO-2a gene only contained MRE sites in the regulatory region. HO-2b and BVRb contained both ARE and MRE motifs in the promoter region upstream of the first exon, while BVRa contained a single MRE motif. The first introns for both HO-2b and BVRa contained multiple MRE and

ARE motifs.

Figure 2.14. MTF-1 and NRF2 Binding Motifs within Zebrafish HO and BVR Promoters. ARE and MRE motifs were identified in the promoter regions upstream of exon 1 or within intron 1 of the HO and BVR genes. The promoter region for HO-1b is not available at this time. ARE: antioxidant response element (Y= C or T, N=any nucleotide); MRE: metal response element (R= A or G, N= any nucleotide).

2.5 Discussion

HO-1 and HO-2 catabolize unbound heme into BV, CO and free iron. While primarily recognized for their enzymatic activity, potential roles related to the relief of oxidative stress, erythropoiesis (Garcia-Santos et al. 2014) and cell signaling (HO-2, BVRa) (Maines 1997,

Barinaga , Verma et al. 1993) have also been suggested. However, very little is known regarding the developmental expression of these heme degradation genes. This study is the first to not only evaluate developmental expression levels of both HO and BVR isoforms, but also the first to evaluate transcriptional responses to oxidative stress of zebrafish HO paralogs and BVR isoforms.

2.5.1 Adult Tissue Distribution and Response to Cd

Females are known to be more susceptible to Cd toxicity than males, an observation believed to be primarily mediated through the estrogen receptor (Johnson et al. 2003). While exposure to Cd resulted in a general increase in HO-1a and HO-1b liver expression in both males and females

(Figure 2.5), it is of interest that basal expression of HO-1a is nearly 20 times greater in male liver tissue than females, and that HO-1b expression is more than double that of females of the same tissue (Figure 2.3). A similar sexual dimorphic expression pattern was noted for BVRb in liver tissue (Figure 2.4). The liver serves as the primary detoxification organ, and as such the strong dimorphic expression may serve as an additional mechanism that contributes to the gender-specific sensitivity to Cd. Interestingly, BVRa was observed to be the predominant isoform in adult brain tissue and was significantly up-regulated by Cd in females (Figure 2.5 &

2.6). Previous vertebrate studies have identified BVRa as being highly expressed in brain tissues

(McCoubrey, Cooklis, and Maines 1995). While further studies are needed, the inducibility of

BVRa may be indicative of a novel protective role within the nervous system.

Similar to the HO-1 paralogs, differences in adult tissue distribution were noted for HO-2a and HO-2b (Figure 2.3). As HO-2 is thought to act as a neuroprotectant through the liberation of

CO (Verma et al. 1993), we were interested in determining which of the four HO genes in zebrafish was most highly expressed in brain tissues. Consistent with known expression of HO-

2 in other vertebrates (Ewing and Maines 1997, 1992, Maines and Trakshel 1993), both HO-2 paralogs were more highly expressed compared to the HO-1 paralogs in brain tissue, with HO-2b showing the greatest levels of expression (Figure 2.3). In contrast to the notion that HO-2 is the non-inducible HO isoform (Ryter, Alam, and Choi 2006), HO-2b was upregulated in response to

Cd in brain and liver tissues of males (Figure 2.6). Furthermore, HO-2b was shown to be the predominant HO-2 isoform in brain tissues, while HO-2a was the predominant isoform in gill tissues (Figure 2.3). This difference in tissue expression may suggest a partitioning of function with differing roles for HO-2 paralogs in adult tissues.

2.5.2 Developmental Expression of HO and BVR

Although greater research focus has been placed on the BVRa isoform, BVRb was shown to be expressed at nearly an order of magnitude higher than BVRa throughout early zebrafish development 24-120 hpf (Figure 2.8). These developmental differences in expression between

BVR isoforms may indicate an important developmental role for BVRb. Interestingly, all four

HO isoforms differed in their expression patterns during early development. The most significant difference in the expression of the HO-1 paralogs was the extremely high levels of

HO-1a at 24 hpf compared to the barely detectable levels of HO-1b expression (Figure 2.8). The

HO-2 paralogs had a similar contrast in expression patterns. HO-2a expression was quite low at

24 hpf but significantly increased around 96 hpf. In contrast, HO-2b expression was fairly consistent throughout development.

The duplication, degeneration, and complementation model (DDC) suggests that duplicated genes can gain a new function, delegate their original function (subfunction partitioning) or become pseudogenes which may be lost through selection if the redundancy is unnecessary

(Amores et al. 1998). The contrasting expression of the HO paralogs during development and in adult tissues is indicative of a change in transcriptional regulation that may represent the first step to the partitioning of function. Specifically, during early development HO-1a alone may be sufficient to deal with oxidative stress. Alternatively, the results may suggest that HO-1a, and not HO-1b, is required for an early developmental function other than its known role in the oxidative stress response. For another example, HO-2 is known to be highly expressed in mammalian brain tissues where it is thought to participate in cell signaling through the generation of CO (Maines 1997, Barinaga , Verma et al. 1993). Neural tissue is the most prominent tissue during early zebrafish development and the embryonic brain is well established prior to the initiation of the rest of organogenesis. The constitutive expression of HO-2b during development and its prominent expression in the brains of adult zebrafish may be indicative of a partitioned signaling role of HO-2b in neural tissues. In contrast, the significant increase in HO-

2a expression during later larval development and its higher expression in gill tissues of adults, may suggest a totally different role for this paralog. While this is purely speculation at this time, these possibilities highlight the importance of considering the different physiological roles of various tissues when investigating the potential for novel or partitioned function.

2.5.3 Changes in Tolerance and Sensitivity to Oxidative Stress

During zebrafish development there appears to be a transition from anoxia tolerance to declining survivability due to anoxia sensitivity (~24-48 hpf) (Padilla 2001). In agreement with Padilla and Roth, studies by Mendelsohn et al determined that anoxia becomes lethal at approximately

52 hpf (Mendelsohn 2008b). One explanation for this transition from anoxia tolerance to anoxia sensitivity is that a highly coordinated change in metabolism occurs, which results in a change from a more primitive anaerobic form of energy generation to an aerobic form of energy generation, which is conducive to generating ROS (Padilla 2001, Ton 2003). Interestingly, challenges with either Cd or tBHQ between 24-120 hpf revealed an increase in sensitivity towards both pro-oxidants beginning at 72 hpf (Figure 2.9), which is approximately 20 hours after the reported switch to an aerobic form of respiration. Furthermore, zebrafish gills do not begin any significant transport until 72 hpf (Kimmel 1995, Rombough 2002) and at this point in time primarily facilitate the uptake of ions rather than oxygen. Thus, it is not too surprising to see an increase in sensitivity to Cd beginning after 72 hpf (Figure 2.9). Conversely, we expected lipophillic tBHQ to show high levels of toxicity at all timepoints due to its ability to diffuse across cell membranes. While zebrafish were quite tolerant to tBHQ challenges between 24-72 hpf in comparison to Cd, they became increasingly sensitive to tBHQ earlier than Cd, showing some mortality between 48-72 hpf at the highest concentration and displaying sensitivity to all concentrations between 72-96 hpf. It is possible that the observed sensitivity to tBHQ may be the result of an increasing availability of NADPH generated by increasing levels of IDHc,

IDHm, and NNT starting between ~72-96 hpf (Figure 2.8). This increase in NADPH-related gene expression during early development coincides with a change from a more oxidative cellular environment to a more reducing cellular environment (Timme-Laragy et al. 2013a).

Ironically, this increase in NADPH would actually facilitate the unwanted redox cycling that is capable of generating significant amounts of oxidative stress.

2.5.4 Potential Regulators of HO and BVR Expression in Response to Oxidative Stress

While the induction of HO-1a was expected, we documented novel observations regarding HO-

1b, HO-2a and BVRb induction in response to Cd, tBHQ or hemin during development (Figures

2.10, 2.11 and 2.13) and provided further evidence of induction of HO-1, HO-2 and BVRa in adults tissues in response to Cd exposure (Figures 2.5-2.7). NRF2, the master regulator of the oxidative stress response, is known to upregulate HO-1a in response to Cd (Alam et al. 1999,

Alam et al. 2000). While NRF2 has also been implicated in the regulation of BVRb (Wu 2011,

Moon 2012), this remains to be confirmed. MTF-1, the master regulator of metals homeostasis, is also activated in response to Cd (Stuart 1985, Günther, Lindert, and Schaffner 2012) and has been implicated in the regulation of several genes involved in iron and heme homeostasis (Figure

2.12) (Troadec et al. 2010, Balesaria et al. 2010, O'Shields et al. 2014). We searched 5 kb upstream of the transcription start site and within the first intron of zebrafish HO-1a, HO-2a,

HO-2b, BVRa, and BVRb and identified numerous MTF-1 and NRF2 binding sites (Figure

2.13). Interestingly, 4 consensus MRE and 1 consensus ARE binding motifs were identified near exon 1 of BVRb. This would seem to be in agreement with our data showing BVRb to be the more inducible BVR isoform (Figure 2.9, 2.10, and 2.12). It is also of interest that we identified

MRE motifs within the HO-2a and HO-2b promoter regions (Figure 2.14), as HO-2 is typically thought of as being non-inducible. Thus, there is strong evidence that the induction of the HO and BVR genes is likely through NRF2 and/or MTF-1. Additional information on the role of

NRF2 in regulating the HO-1a, BVRa and BVRb genes will be provided in Chapter 4.

2.5.5 Conclusions

These differences in HO-1 and HO-2 developmental expression, adult tissue distribution, and responses to oxidative stress are consistent with potential subfunction partitioning of the paralogous isoforms. For example, the transient high levels of HO-1a expression at 24 hpf are in direct contrast to the almost non-detectable expression of HO-1b at this same timepoint. This could be indicative of a specific function assigned to HO-1a that is no longer performed by HO-

1b. Future studies employing the powerful reverse genetics techniques, such as targeted mutagenesis using the CRISPR-Cas system may be used to characterize these functional differences. Furthermore, with respect to BVRa and BVRb, it is interesting that these two genes of distinct evolutionary origin have evolved to have similar reductase activity and display nearly identical expression patterns during early development. However, the fact that the BVRb transcripts are expressed at an order of magnitude higher suggests a difference in function. In addition, the presence of a diversity of other functional domains within the BVRa isoform offers future opportunities to evaluate the potential differences in function during development.

CHAPTER 3

TEMPORAL AND SPATIAL EXPRESSION OF HO AND BVR DURING NORMAL DEVELOPMENT AND IN RESPONSE TO OXIDATIVE STRESS

3.1 Abstract

In Chapter 2 quantitative analysis of expression of heme degradation genes revealed differences in their expression patterns during early development, within adult tissues and in response to various pro-oxidants. In this chapter, we characterize the spatial and temporal expression patterns of HO-1a, BVRa and BVRb using in situ hybridization and in vivo promoter analysis to gain further insights into the processes by which these genes are regulated. We demonstrate that the spatial and temporal expression patterns of both BVR isoforms are consistent with known hematopoietic regulators during early zebrafish development. Additionally, using in vivo promoter analysis we documented HO-1a expression at ~24 hpf within the intermediate cell mass (ICM), the site of embryonic erythrocyte formation, and within circulating embryonic erythrocytes during later stages of development. These observations support a role for HO-1a as a regulator of erythropoiesis. Furthermore, we show zebrafish HO-1a to be expressed within the eye during early development. Further histological assessments revealed the origin of expression to be from the epithelial cells of the lens. Thus, HO-1a, BVRa and BVRb developmental expression occurs in two cell types that are highly susceptible to oxidative stress. This study presents the most thorough characterization of the spatial and temporal changes in expression of

HO-1a and BVR to date.

3.2 Introduction

In Chapter 2 we characterized quantitative differences in expression of the genes involved in heme degradation (HO-1/HO-2, BVRa/BVRb) during development and in adult tissues under control and stressed conditions. While some information regarding the spatial and temporal expression patterns HO and BVR isoforms has been reported, these studies do not evaluate multiple developmental timepoints and do not assess HO and BVR expression simultaneously.

To further characterize the expression of HO-1a, BVRa and BVRb during early development we utilize two techniques which take advantage of the rapid and transparent development of zebrafish, i.e., in situ hybridization and in vivo promoter analysis.

Galloway et al. demonstrated BVRb expression within the intermediate cell mass (ICM), the embryonic site of RBC formation at 20 hpf (Galloway 2005). Additionally, a study by Craven et al. showed HO-1a to be expressed within the posterior blood island (PBI), an intermediate location of blood cell differentiation at ~36 hpf (Craven et al. 2005). While no other data regarding the spatial and temporal expression patterns of heme degradation genes is available in zebrafish, there are numerous studies which have characterized expression patterns of genes necessary for heme biosynthesis (Nilsson et al. 2009, Hanaoka, Dawid, and Kawahara 2007,

Hanaoka et al. 2006), as well as genes necessary for iron transport and iron homeostasis

(Fraenkel et al. 2009). Additionally, mutagenesis studies targeting some of these genes have produced numerous mutant lines with defects in blood cell formation (Weinstein 1996). All of the above mentioned studies have identified ontogenic expression patterns for heme synthesis and heme transport genes to be within or near the ICM. The known temporal expression patterns of zebrafish BVRb and HO-1a, as well as similar expression patterns noted for heme synthesis and transport, led us to hypothesize that HO-1a and the BVR isoforms may show expression

patterns consistent with those of other erythroid markers and genes which regulate heme production.

It has long been established that HO-1 is inducible by numerous pro-oxidants. However, despite the suggestion that BVR may function to provide cellular protection through the generation of the potent antioxidant bilirubin (BR), very little information regarding the response of either BVR isoform to pro-oxidant challenges have been reported. In chapter 2 we showed

BVRb to be the more inducible BVR isoform during early development. To follow up on this, in this chapter we evaluate if pro-oxidant exposures alter the spatial and temporal expression of

BVRa and BVRb during early development.

3.3 Materials and Methods

3.3.1 Chemicals

Cadmium (Cd) chloride (CAS # 654054-66-7) was purchased from Sigma-Aldrich (St. Louis,

MO). Toluidine Blue (CAS # 92-31-9) was purchased from VWR (Radnor, PA). BM Purple,

Proteinase K, and Anti-Digoxigenin-AP, Fab fragments were obtained from Roche Diagnostics

(Indianapolis, IN). Technovit 7100® (Heraeus-Kulzer, Wehrheim/TS, Germany) was purchased from Electron Microscopy Sciences (Hatfield, PA).

3.3.2 Fish Husbandry

The TL (Tupfel/Long fin mutations) wild-type strain of zebrafish was used for all experiments.

Fertilized eggs were obtained from multiple group breedings from a Mass Embryo Production

System (MEPS; Aquatic Habitats, Apopka, FL) with ~200 fish at a ratio of 2 female per 1 male

fish. Procedures used in these experiments were approved by the Animal Care and Use

Committee of the University of Alabama, Tuscaloosa, Alabama, USA.

3.3.3 Generation of HO-1a and BVR Whole Mount In Situ Hybridization (WISH) Probes

BVRa, BVRb, and HO-1a sense and antisense digoxigenin (DIG) labeled RNA probes were synthesized using the Roche Dig RNA Labeling Kit (SP6/T7) from cDNA subcloned into pGEM®-T Easy Vector (Promega, Madison, WI). The BVRa probe is 413 bp in length and includes 83 bp of the 3’ UTR. The BVRb probe is located within the coding sequence and is 460 bp in length. Finally, the HO-1a probe is 398 bp in length, of which 314 bp are located within the 3’ UTR. Primers used for probe generation are listed in the primers table.

3.3.4 Collection of Zebrafish Embryos for a Developmental Time Series

A large batch (>3000) of embryos were generated by active breeding of the fish in the MEPS system for 60 minutes. After 60 minutes embryos were collected and transferred to large petri dishes (150 mm diameter) at a density of 100 embryos per 100 ml of 0.3x Danieau’s solution at

3.3.5 Whole Mount In Situ Hybridization

Developing zebrafish embryos were collected at 20, 24, 28, 32, 36, 48, 72, 96, and 120 hpf.

Embryos between 20-36 hpf were dechorionated using forceps. All embryos were fixed in 4%

paraformaldehyde (PFA), stored at 4 °C overnight and rehydrated the following day. A 3% hydrogen peroxide (H2O2)/ 0.5% potassium hydroxide (KOH) solution was used to remove pigmentation at later timepoints (≥ 48 hpf) as described by Thisse and Thisse (Thisse 2007).

Procedures for WISH were performed essentially as described by (Thisse 2007). Fixed embryos were rehydrated and permeabilized in a 10 µg/ml solution of proteinase K (Roche Diagnostics) to allow for probe entry. Prior to the addition of probes, embryos were subjected to a pre- hybridization solution which contained yeast tRNA (2 µg/ml), heparin (50 µg/ml), 50% deionized formamide, 5x SSC, 0.1% Tween 20, and 1 M citric acid between 2-4 hours at 70°C.

Sense or antisense probes (~75 ng) suspended in pre-hybridization solution were added and allowed to incubate overnight at 70 °C. The next day, embryos were washed and incubated with blocking buffer for 4 hours. After 4 hours the blocking buffer was removed and replaced with

Anti-Digoxigenin-AP, Fab fragments (Roche Diagnostics) made up in blocking buffer (1:10,000) and incubated overnight at 4 °C. The following day the antibody was removed and the embryos were washed 6 times in phosphate buffered saline with 0.1% tween (PBST). To eliminate background staining at later timepoints, a group of embryos were processed in parallel until the pre-hybridization step, at which point a 1:1000 dilution of the antibody was added to the embryos which were subsequently rocked at room temperature for 2 hours and stored overnight at 4°C (Chitramuthu and Bennett 2013). This pre-adsorbed antibody was then further diluted to a final ratio of 1:10,000 in blocking buffer the next day and added to the embryos which were incubated overnight at 4 °C. Following antibody removal and PBST washes, embryos were further washed in alkaline Tris buffer and stained with BM-Purple AP Substrate precipitating

(Roche Diagnostics). Zebrafish embryos were visualized with a Nikon AZ100M microscope and photographed using a Nikon Digital Sight DS-F-1 camera.

3.3.6 Cd Dosing for WISH and qPCR

For further characterization of BVR expression by both WISH and qPCR assessment, embryos were challenged with 50 µM Cd for 8 hours starting at 96 and 120 hpf, respectively. For HO-1a characterization by WISH and real-time RT-PCR, eleutheroembryos at 5 dpf were challenged with 50 µM Cd for 4 hours.

3.3.7 Gene Expression Assessment by Real-time RT-PCR

Total RNA was prepared from embryos using the TRIzol® reagent (Life Technologies,

Invitrogen) protocol. The qScript™ cDNA Synthesis Kit (Quanta Biosciences, Gaithersburg,

MD) was used to generate cDNA from 1 µg total RNA as per manufacturer’s instructions.

PerfeCTa® SYBR® Green Supermix for iQ™ (Quanta Biosciences, Gaithersburg, MD) was used for qPCR experiments using a MyiQ2 Two-Color Real-Time PCR Detection system (Bio-

Rad, Hercules, CA) under the following conditions: 95°C for 3 min, 95°C for 15s/60°C for 30 sec (40 cycles). To ensure that only a single product was amplified, all qPCR experiments were followed with a melt curve analysis. Changes in gene expression were calculated using the comparative Ct method (Schmittgen 2008) and statistical significance in comparison to control embryos was determined using two-way ANOVA followed by a Bonferroni post hoc test.

3.3.8 HO-1a In Vivo Promoter Construct

To create the HO-1a expression construct, a 5.4 kb DNA region upstream of HO-1a exon 2 was amplified from zebrafish bacterial artificial chromosome (BAC) DKEY-69E1 using Kapa Hifi

DNA Polymerase (Kapa Biosystems, Inc., Boston, Ma) and gene specific primers flanked by attB4 and attB1 sites. Entry clones and destination vectors for Tol2 transgenesis were kindly

provided by the Chien lab (Kwan 2007). To create a 5’ entry vector, the PCR product was recombined with donor vector #219 pDONRP4-P1R using Gateway® BP Clonase™. The transgenic construct was created by performing a 3-way recombination reaction using Gateway®

LR Clonase II ™ Enzyme to combine the HO-1a promoter 5’ entry vector with middle vector

#386 (containing mCherry) and 3’ entry vector #302 (containing the SV40 late polyA signal) into destination vector #395 (containing Tol2 inverted repeats and the cardiac myosin light-chain

(cmlc2:egfp) “heart marker” gene) to create pDestTol2CG2;HO-1a:pME-mCHerry-p3EpolyA.

One Shot® Top 10 cells (Invitrogen) were used to amplify the expression construct and a

QIAGEN® Maxiprep was used for purification and removal of any potential endotoxins. All

Gateway® reactions were performed according to manufacturer’s instructions.

Figure 3.1. HO-1a In Vivo DNA Expression Construct for Tissue Specific Expression. Schematic of the HO-1a in vivo promoter construct. The construct also contains a cardiac myosin light chain (cmcl2) promoter ligated to a GFP reporter gene, which serves as a convenient marker to identify successful transformants. Successful genomic incorporation is indicated by the expression of GFP in the heart at approximately 16-17 hpf. The entire construct is flanked by inverted terminal Tol2 repeats which serve as recognition sequences for the transposase enzyme. Thus, co-injection of this construct with transposase mRNA facilitates genomic incorporation.

3.3.9 Generation of HO-1a Stable Transgenic Lines and In Vivo Promoter Analysis

To generate the stable transgenic line Tg(ho-1a:mCHerry;cmlc2-eGFP)mjj1, embryos at the 1-2 cell stage were co-injected with 50 pg of the HO-1a expression construct pDestTol2CG2;HO-

1a:pME-mCHerry-p3EpolyA and 5 pg of in vitro transcribed transposase mRNA using a

Narishige IM-300 microinjector. Embryos were screened for expression of the cmlc2:eGFP heart marker by 24 hpf and subsequently screened for mCherry expression by 5 days past fertilization (dpf). Embryos strongly expressing the mCherry transgene were raised to adulthood and crossed with wild-type TL fish to create F1 heterozygous progeny as described by

(Kawakami, Shima, and Kawakami 2000, Kawakami 2007). F1 progeny were subsequently raised to adulthood and crossed with wild-type TL fish to create F2 Tg(ho-1a:mCHerry;cmlc2- eGFP)mjj1. The same procedures were used to generate F3 and F4 Tg(ho-1a:mCHerry;cmlc2- eGFP)mjj1.

We initially documented HO-1a expression in F1, F2, and F3, and F4 Tg(ho-

1a:mCHerry;cmlc2-eGFP)mjj1 embryos starting at ~20-24 hpf, and continued to do so every 24 hours for up to 14 days. To more specifically identify timepoints of expression during early development, embryos were collected from a mating between 1 F2 male HO-1a transgenic fish and 2 wild type TL females. Changes in HO-1a expression were documented in embryos every

3 hours starting at ~22 hpf and continuing through 40 hpf. All fluorescent imaging was performed using a Photometrics CoolSNAP ES2 camera attached to a Nikon AZ100M microscope. To ensure that the observed fluorescence was real and not auto-fluorescence, wild- type embryos were imaged at comparable timepoints.

3.3.10 Cd Dosing for In Vivo Promoter Analysis

Tg(ho-1a:mCHerry;cmlc2-eGFP)mjj1 eleutheroembryos at 5 dpf were challenged for 2-4 hours with 25 μM Cd. Immediately following exposures, zebrafish were transferred to 0.3X Danieau’s

solution and allowed to recover at 28.5 °C for at least 24 hours before being imaged for changes in mCherry expression. Wild-type TL zebrafish were utilized as controls.

3.3.11 Histology

Wild type TL and HO-1a transgenic embryos were obtained as described above and were fixed overnight in fresh 4% PFA at 4°C. To prevent issues with fixative based auto-fluorescence,

4% PFA was removed the next morning and embryos were rinsed with 1x PBS (3 five minute washes). Following the rinses, transgenic fish were quickly examined under the fluorescent microscope to ensure that fluorescence was not affected by fixation.

Dehydration, infiltration, and embedding of wild type samples were performed essentially as described by Schonthaler et al except where noted (Schonthaler et al. 2010). Dehydration of wild-type samples was performed at room temperature in a series of 15 minute incubations with increasing concentrations of ethanol (30%, 50%, 70%, 90%, 96%, and 2 x 100%) (Schonthaler et al. 2010). Dehydration of transgenic fish was performed as described by Sullivan-Brown et al.

(Sullivan-Brown, Bisher, and Burdine 2011). Briefly, embryos were subjected to the following solutions after which the solutions were immediately removed (95% 1x PBS/5% ETOH, 75% 1x

PBS /25% ETOH, 50% 1x PBS/50% ETOH, 25% 1x PBS/ 75% ETOH, and 100% ETOH).

Infiltration of wild type embryos was performed at room temperature while gently rocking using Technovit® 7100 (Heraeus-Kulzer, Wehrheim/TS, Germany) with the addition of Hardener

I as follows: (2:1) 100% ETOH/Technovit for 30 minutes; (1:1) Technovit/100% ETOH for 45 minutes; 100% Technovit for 1 hour. Conversely, transgenic fish were incubated with 1 mL of infiltration solution at room temperature, at which point infiltration solution was removed and replaced with 1 mL infiltration solution and kept overnight at 4°C. Following infiltration, wild

type and transgenic samples were transferred to 15 x 15 mm mold for embedding. The same embedding procedure for wild type and transgenic fish was used; however, care was taken to avoid unnecessary light exposure when embedding transgenic samples. Embedding solution was prepared fresh (15ml infiltration solution/1ml of hardener II) in a 50 ml capped conical tube and stirred with a magnetic stir bar for 1.5 minutes. To soften the polymerized blocks, 0.6 mL of polyethylene 400 (peg400) was added to the embedding solution as described by Yeung and

Chan (Yeung and Chan 2014). Embedding mixture was added to the infiltrated samples which were subsequently oriented within 5 minutes. Samples were left overnight to polymerize at

28°C. To limit the amount of light and oxygen exposure, molds were placed in sealable

Tupperware containers lined with aluminum foil. Following embedding, embryos were sectioned at 4-6 μm using a Leica RM 2125 microtome with Edge-Rite High Profile Blades

(4275H, disposable steel blades). During the sectioning process, additional peg400 was added as needed to soften the block and enhance cutting. Sections were hydrated in a water bath at 45-50

⁰C and transferred to a microscope slide and allowed to air dry.

3.3.12 Dehydration and Preparation of ISH Samples for Histology

Zebrafish ISH samples are stored in glycerol after imaging. Thus, to prepare them for histology we remove glycerol by performing 3 washes in 1x PBS for 5 minutes. Five minute dehydration washes were performed as follows: 95% 1x PBS/5% ETOH, 75% 1x PBS /25% ETOH, 50% 1x

PBS/50% ETOH, 25% 1x PBS/ 75% ETOH, and a 1 minute wash in 100% ETOH. Following three rinses, two additional 30 minute washes were performed in 100% ETOH. Embedding and sectioning of ISH samples was performed as described above.

3.3.13 Toluidine Blue Staining

Staining with toluidine blue was performed as described by Sullivan-Brown et al (Sullivan-

Brown, Bisher, and Burdine 2011). Essentially, slides were incubated in a 1% toluidine blue solution for ~ 1 minute and subsequently rinsed for 5 minutes with water and allowed to dry overnight. Permount was used to affix coverslips for all samples except samples used for fluorescent imaging in which case a wet mount was prepared. Imaging was performed using a

Photometrics CoolSNAP ES2 camera attached to a Nikon AZ100M microscope.

3.3.14 In Silico Promoter Analysis

We searched 4 kb upstream of the transcription start site and within the first intron of zebrafish

HO-1a, HO-2a, HO-2b, BVRa, and BVRb for consensus GATA-1 (WGATAR) binding motifs using ApE software by M. Wayne Davis. The promoter region of zebrafish HO-1b is not available at this time.

Table 3.1 qPCR, Cloning, and Gateway Cloning Primers

Gene qPCR Primers Sequence HO-1a Forward 5’-GCTTCTGCTGTGCTCTCTATACG-3’ Reverse 5’-CCAATCTCTCTCAGTCTCTGTGC-3’ HO-1b Forward 5’-GCAGTGATCTGTCTGAACAG-3’ Reverse 5’-GCTTGTACTGTGTTTGTGTG-3’ HO-2a Forward 5’-ATGGCGGTCAGTGGAAACACAACC-3’ Reverse 5’-GGCAACAGCAGCAACCAATGTGGC-3’ HO-2b Forward 5’-TTTAGGAGGTTGAGTTGGAGTCAG-3’ Reverse 5’-TTCTGCCTTCTGGTGCACTTCT-3’ BVRa Forward 5’-CAGGCAGTTTCTGGAGGCAGG-3’ Reverse 5’-CCAGACCCTTCTGTTGAGC-3’ BVRb Forward 5’-GCATGTCAGCATTCCTCTTGTGG-3’ Reverse 5’-CACCAGCAATATGTGGAGG-3’ Gene Cloning Primers Sequence HO-1a Forward 5’-ATTGATATCCTACAGCACAAAGATGGAC-3’ Reverse 5’-TATGGATCCAGTAAAAAAAACTAATCTG-3’ BVRa Forward 5’-TAAGAAAACCACATCAAAGGCTG-3’ Reverse 5’-GCACTTCAGTTATCTGTCTCTATA-3’ BVRb Forward 5’-TGTGGCGATATTTGGCTGCACGG-3’ Reverse 5’-ACCATCAGTACCAACCTGTTATTTTCGC-3’ Gene Gateway Cloning Sequence HO-1a Forward 5’-GGGGACAACTTTGTATAGAAAAGTTGTGCACT TATCAGGCATAGTAAATGGGCA-3’ Reverse 5’- GGGGACTGCTTTTTTGTACAAACTTGCTCTAAAA AAACAGATTGCCCTATTTG-3’

3.4 Results

3.4.1 Spatial and Temporal Expression of BVRa and BVRb

To determine the spatial and temporal expression patterns of BVRa and BVRb, WISH was performed at timepoints throughout development from early embryo to later larval stages.

Between 20-24 hpf, both BVRa and BVRb were strongly expressed within the intermediate cell mass (ICM) (Figure 3.1). However, BVRb expression remained strongly detected throughout the ICM at 28 hpf, while BVRa expression was restricted within the posterior blood island (PBI).

Furthermore, both transcripts were detected within circulating blood cells in the Ducts of Cuvier at 28 hpf (Figure 3.1).

Figure 3.2. Changes in BVR Expression During Primitive Hematopoiesis. WISH was performed on embryos at 20, 24, and 28 hpf using BVRa and BVRb sense and antisense RNA probes. To determine if any observed staining was the result of endogenous phosphatase activity, a set of embryos at each time point were processed under the same conditions without the addition of a probe (data not shown). Abbreviations: hpf = hours past fertilization; ICM = intermediate cell mass; PBI= posterior blood island; RBCs = red blood cells.

Interestingly, by 32 hpf BVRa transcripts were no longer detected within the PBI and remained confined to circulating RBC’s (Figure 3.3). In contrast, BVRb transcripts were expressed in both the PBI and in circulating blood cells in the yolk sac. By 36 hpf, neither isoform was detected within the PBI. However, both BVRa and BVRb expression persisted within circulating RBC’s

(Figure 3.3).

Figure 3.3. Changes in BVR Expression During the Onset of Definitive Hematopoiesis. WISH was performed on embryos at 32 and 36 hpf using BVRa and BVRb sense and antisense RNA probes. To determine if any observed staining was the result of endogenous phosphatase activity, a set of embryos at each time point were processed under the same conditions without the addition of a probe (data not shown). hpf = hours past fertilization; PBI = posterior blood island; RBC’s = red blood cells.

At 52 hpf, BVRa expression was no longer detectable while BVRb expression remained strong within circulating blood cells within the yolk sac (Figure 3.4). At the later timepoints (76,

100 and 124 hpf), BVRb progressed from being faintly detected in the PBI to strongly expressed within the caudal hematopoietic tissue (CHT), a transient location of further blood cell differentiation (Figure 3.4 & 3.5) Furthermore, at the later timepoints (100 and 124 hpf), BVRb was also detectable in the kidney and liver. In contrast, BVRa expression was not detectable at any of these later timepoints (Figure 3.5).

Figure 3.4. Changes in BVRa and BVRb Expression at 48 and 72 hpf. WISH was performed at 52 and 76 hpf using BVRa and BVRb sense and antisense RNA probes. To determine if any observed staining was the result of endogenous phosphatase activity, a set of embryos at each time point were processed under the same conditions without the addition of a probe (data not shown). hpf = hours past fertilization; CHT = caudal hematopoietic tissue; RBC’s = red blood cells.

Figure 3.5. Changes in BVRa and BVRb Expression at 96 and 120 hpf. WISH was performed at 96 and 120 hpf using BVRa and BVRb sense and antisense RNA probes. To determine if any observed staining was the result of endogenous phosphatase activity, a set of embryos at each time point were processed under the same conditions without the addition of a probe (data not shown). hpf = hours past fertilization; CHT = caudal hematopoietic tissue; H = heart; K = kidney.

To further examine the expression patterns of BVRa and BVRb, and determine if Cd- exposure produced any detectable changes in spatial expression at later timepoints, WISH was performed on control and Cd-dosed fish at 104 and 128 hpf (8 hour, 50 µM Cd). BVRa was not detectable at 104 or 128 hpf in control fish. However, slight expression was detected at 104 hpf within the CHT in Cd exposed fish (Figure 3.6).

Figure 3.6. Effects of Cd on BVRa Expression During Definitive Hematopoiesis. WISH was performed on developing zebrafish at 104 and 128 hpf using BVRa sense and antisense RNA probes. Embryos were challenged for 8 hours with 50 µM Cd starting at 96 and 120 hpf. To determine if any observed staining was the result of endogenous phosphatase activity, a set of embryos at each time point were processed under the same conditions without the addition of a probe (data not shown). CHT = caudal hematopoietic tissue; hpf = hours past fertilization.

In contrast, BVRb transcripts were detected within the CHT, as well as the heart and liver, at

104 and 128 hpf in control fish (Figure 3.7). Cd exposure appeared to result in slightly stronger staining of BVRb at both 104 and 128 hpf; however, staining seemed to be strongest at 104 hpf

(Figure 3.7). Similar to these results obtained using in situ hybridization, quantitative

comparisons using real-time RT-PCR demonstrated the induction of BVRb in response to Cd at these timepoints (Figure 3.8).

Figure 3.7. Effects of Cd on BVRb Expression during Definitive Hematopoiesis. In situ hybridization was performed on developing zebrafish at 104 and 128 hpf using BVRb sense and antisense RNA probes. Embryos were challenged for 8 hours with 50 µM Cd starting at 96 and 120 hpf. To determine if any observed staining was the result of endogenous phosphatase activity, a set of embryos at each time point were processed under the same conditions without the addition of a probe (data not shown). CHT= caudal hematopoietic tissue; H= heart; hpf= hours past fertilization; K= kidney; L= liver.

Figure 3.8. Effects of acute Cd exposure on BVR expression. Real-time RT-PCR was used to quantify changes in BVR expression following exposure to Cd. Embryos were challenged for 8 hours with 50 µM Cd starting at 96 and 120 hpf. Statistical significance in comparison to control embryos was determined using a Two-way ANOVA followed by a Bonferroni post hoc test (*p-value < 0.05). All values are normalized to 18S ribosomal RNA. n = 3 biological replicates of 20 pooled embryos.

3.4.2 HO-1a In Vivo Expression

To determine if HO-1a spatial and temporal expression patterns are similar to BVRa and BVRb during early development, we performed in situ hybridization at 20 and 24 hpf (Figure 3.9).

While we did see some evidence of HO-1a expression near the ICM region, the intensity of this signal was much lower in comparison to that observed for BVRa and BVRb (Figure 3.2 & 3.3).

Thus, to further characterize HO-1a expression, a stable transgenic line was created in which the promoter region of HO-1a drives expression of the mCherry fluorescent protein (Figure 3.1). We observed mCherry fluorescence within the ICM between 24-32 hpf (Figure 3.9), the same timepoints in which we detected BVRa and BVRb transcripts within the same region. After 52 hpf we could not detect either BVR isoform within circulating RBC’s using in situ hybridization

(Figure 3.4 & 3.5). However, by ~ 36 hpf we observed mCherry fluorescence within circulating

RBCs which persisted through at least 7 dpf (Movie 3.1 and 3.2). In agreement with the fluorescence that we detected in circulating RBCs, we were able to detect HO-1a transcripts within primitive RBCs of the yolk sac using in situ hybridization after histological processing

(Figure 3.9).

Figure 3.9. HO-1a Expression within the ICM and Embryonic Red Blood Cells during Primitive Hematopoiesis. Strong mCherry expression driven by a zebrafish HO-1a promoter construct is visible within the ICM at 24 and 32 hpf. HO-1a expression could also be detected in red blood cells using in situ hybridization at 36 hpf following histological analysis. ICM = intermediate cell mass, RBCs= red blood cells.

During this same developmental period (between 1-2 dpf), we also observed mCherry fluorescence within the eye (Figure 3.10). To more precisely determine the time at which mCherry began to be translated within presumptive lens tissues, we documented expression every three hours in three separate fish beginning at 22 hpf and continuing through 40 hpf. Faint fluorescence was seen within the lens at 35 hpf (Figure 3.10). The intensity of mCherry

remained strong throughout the remainder of the timepoints. Similarly, we detected HO-1a transcripts within the lens using in situ hybridization at 36 hpf.

Figure 3.10. Comparison of HO-1a mRNA and mCherry Expression Driven by the HO-1a Promoter in Eye Tissues. At 35 hpf mCherry showed some expression in developing eye tissues. Similar expression was seen at 36h hpf using in situ hybridization. hpf= hours past fertilization; Le = lens.

By 4 dpf mCherry fluorescence remained strong within the lens (Figure 3.11). Additionally, strong fluorescence was also observed in the liver and kidney. Fluorescence in all three tissues remained strong at 5 dpf; however, at this time we also began to note mCherry expression within the primordial hindbrain channel (PMHC) (Figure 3.11).

Figure 3.11. HO-1a Expression in Liver, Kidney, and Eye Tissue Between 4-5 dpf. mCherry expression within the eye, liver, kidney and circulating RBC’s at 4 and 5dpf in transgenic fish . Le = lens; L = liver; K = kidney; dpf = days past fertilization, PMBC = primordial midbrain channel.

3.4.3 Effects of Cd Exposure on HO-1a Expression

HO-1 is highly inducible by oxidative stress and has specifically been shown to be upregulated in response to Cd exposures (Alam 1994, Alam, Shibahara, and Smith 1989, Takeda et al. 1994).

To further demonstrate that our transgenic fish is indeed expressing mCherry under the control of the HO-1a promoter, we compared changes in HO-1a expression in both transgenic fish and wild-type fish challenged with Cd (Figure 3.12). While Cd challenges seemed to result in some

increases in fluorescent intensity in our transgenic fish, documenting increases in fluorescent intensity in comparison to control fish proved to be quite challenging. However, we were able to observe an increase in fluorescence within the olfactory placodes (Figure 3.12), similar to that reported by Blechinger et al (Blechinger 2007). A more obvious change in HO-1a expression in both transgenic and wild-type Cd challenged fish was observed in liver tissue using in situ hybridization. A final method, real-time RT-PCR was used to confirm that Cd was indeed resulting in HO-1 induction at both 5 and 6 dpf (Figure 3.12).

Figure 3.12. WISH, In Vivo Promoter Analysis, and qPCR Show Differences in HO-1a Induction by Cd at 4, 5 and 6 dpf. Transgenic fish challenged with 100 µM Cd for 8 hours starting at 4 dpf displayed increases in mCHerry expression within the olfactory placodes (dorsal view). Wild type fish and a second set of transgenic fish were challenged with 50 µM Cd for 4 hours starting at 5 dpf. and immediately fixed in 4% PFA for in situ hybridization following exposures. Real-time RT-PCR was used to quantify changes in HO-1a and HO-1b expression following exposure to Cd (eight hour exposure to 50 µM Cd starting at 96 and 120 hpf). Statistical significance in comparison to control embryos was determined using a Two-way ANOVA followed by a Bonferroni post hoc test (*p-value<0.05). All values are normalized to 18S ribosomal RNA. n = 3 biological replicates of 20 pooled embryos. OP = olfactory placodes; dpf = days past fertilization.

3.4.4 HO-1a Eye Expression is Localized to Lens Epithelial Tissue mCherry expression driven by the HO-1a promoter was detectable within lens tissue of our transgenic fish at ~35 hpf (Figure 3.10) and expression appeared to increase in intensity as development proceeded (Figure 3.11). Similarly, using WISH, we were able to detect HO-1a transcripts in lens tissues between ~36 hpf (Figure 3.10). To further refine the origin of HO-1a promoter driven mCherry fluorescence, we performed histological assessments on our transgenic fish (Figure 3.13). Comparison of our HO-1a transgenic fish to wild-type fish stained with

Toluidine blue revealed the origin of fluorescence to be from within the lens epithelial cells

(Figure 3.13).

Figure 3.13. HO-1a Lens Expression is Specific to Lens Epithelial Cells. mCherry expression at 7 dpf in the lens (left). Sagittal section at 6 dpf (right) shows the origin of mCherry expression driven by the HO-1a promoter to be from the epithelial cells of the lens which is divided into the lens epithelial cells (LE), posterior lens (PL) and the lens transition zone. Toluidine blue staining (center) of 5 dpf eleutheroembryos showing the layering of the retina, the lens (Le), cornea (Cor), and retinal pigment epithelial cells (RPE).

3.5 Discussion

The primary objective of this chapter was to characterize the spatial and temporal expression patterns of HO-1a, BVRa, and BVRb. Using two different techniques, i.e., in situ hybridization and in vivo promoter analysis interesting observations regarding the spatial and temporal expression of these genes were noted. These observations are further discussed below.

3.5.1 Developmental BVR Expression Coincides with Changes in Location of Erythropoiesis

Zebrafish blood cells develop through a conserved program which is broadly divided into primitive and definitive waves. The primitive wave produces primitive macrophages and embryonic RBC's, while the definitive wave produces self-renewing hematopoietic stem cells

(HSC's) which can produce all adult blood cell lineages (Clements and Traver 2013, Detrich et al. 1995). Both waves contribute to the regulation of HSC’s, and display distinctive spatial and temporal components characterized by the rapidly changing expression of specific transcription factors within the anterior lateral mesoderm (ALM) and the posterior lateral mesoderm (PLM)

(Hsia and Zon 2005). At approximately 24 hpf, transcription factors specific to the PLM begin to rapidly decrease in expression in conjunction with the onset of active circulation (Galloway

2003). As shown in Chapter 2 (Figure 2.5), expression of BVRa and BVRb during early development was highest at 24 hpf, a time point which coincides with the onset of circulation

(Kimmel 1995). WISH was used to determine if there are differences in spatial and temporal expression of BVR isoforms during the transition from the primitive to definitive wave of hematopoiesis (Figure 3.2, 3.3, 3.4, & 3.5). At 20 and 24 hpf, both isoforms were expressed within the ICM, which is the first site of hematopoiesis (Willett et al. 1999, Detrich et al. 1995,

Thompson et al. 1998). At 28 hpf BVRa expression appears to be reduced while becoming

localized within the aorta-gonadmesonephros (AGM), the site where the definitive wave originates (Zhang and Rodaway 2007), as well as within the PBI, a transient location for blood development (Jin 2007). Conversely, BVRb expression remained strong within the AGM at 28 hpf, a timepoint coinciding with the transition from the primitive to definitive wave characterized by a decrease in PLM transcription factors and the onset of circulation (Hsia and Zon 2005).

Expression patterns of both BVRa and BVRb (Figure 3.2 & 3.3) are consistent with the expression of GATA-1, a PLM transcription factor which specifies erythroid cells and represses myeloid cells (Galloway 2005, Lam et al. 2009). The expression is also consistent with PU.1

(aka SPI-1), a transcriptional regulator of the myeloid lineage (Hsu et al. 2004, Rhodes et al.

2005). Indeed, elimination of GATA-1 in mutant fish or via morpholino knockdown results in a decrease in BVRb expression at 20 hpf in zebrafish (Galloway 2005), suggesting that BVRb is regulated, at least in part, by GATA-1. The similar expression pattern of BVRa during early development (Figure 3.2) suggests that it may also be regulated by GATA-1. Both BVR isoforms show strong expression within the ICM between 20-24 hpf prior to being expressed in circulating RBC’s. An initial search of the promoter region of BVRa reveals the presence of 16

GATA-1 motifs (WGATAR) (Martin and Orkin 1990, Yang and Evans 1992) within the first intron, and an additional 19 motifs within 6 kb upstream of the transcriptional start site in exon 1

(Figure 3.14).

Figure 3.14. GATA-1 Response Elements within Promoter Regions of HO-1a, BVRa, and BVRb. Multiple GATA-1 motifs were identified within the promoter regions of zebrafish HO- 1a, BVRa, and BVRb. W= A or T, R= A or G

Interestingly, at 28 hpf, BVRa expression is not detected in the anterior end of the ICM. The early expression pattern of BVRa within the ICM and the restricted expression of BVRa at 28 hpf is consistent with the expression pattern of the transcription factor PU.1 which regulates the development of myeloid cells (Hsu et al. 2004, Rhodes et al. 2005). However, at 32 hpf BVRb expression persists within circulating RBC’s and within the PBI, the location from which erythomyeloid progenitor cells (EMPs) emerge to support definitive hematopoiesis (Bertrand et al. 2007). Further expression profiling of BVRa and BVRb was performed during timepoints encompassing the definitive wave of hematopoiesis. BVRa expression was not detected between

52-124 hpf in situ (Figure 3.4 & 3.5). Conversely, BVRb transcripts were detected within circulating RBC’s at 52 hpf, as well as within the CHT at 100 and 124 hpf. Additionally, BVRb was detected in the heart and the liver at 100 and 124 hpf (Figure 3.5). As primitive RBCs

circulate until ~ 4 dpf we cannot be certain that the observed expression of BVRb within the

CHT is not the result of remaining embryonic RBC’s which express BVRb. However, the loss of BVRa expression at the same time point supports the notion that the BVRb signal we are seeing is indeed part of the CHT. The CHT serves as a temporary location in which EMP’s further differentiate into erythomyeloid and lymphoid lineages (Jin 2007, Murayama et al. 2006).

Therefore, further studies determining whether the observed expression patterns of BVRb are indeed a part of this tissue are of interest.

3.5.2 In Vivo Promoter Analysis Supports a Role for HO-1a as a Co-regulator of Erythroid Differentiation

In Chapter 2 we showed HO-1a to be highly expressed at 24 hpf (Figure 2.5) using real-time RT-

PCR. In the current chapter, we were able to observe strong fluorescence within the ICM, the location of primitive RBC development, at 24 hpf (Figure 3.9). Furthermore, by 2 dpf HO-1a appeared to be expressed within circulating RBCs, and by 5 dpf mCherry expression was visible within the kidney, the adult hematopoietic organ (Figure 3.11). Only two papers have identified

HO-1a expression in zebrafish RBC’s (Craven et al. 2005, Thisse et al. 2004). However, recent studies have shown that HO-1 is present in mouse erythroid cells and is upregulated during erythroid differentiation (Garcia-Santos et al. 2014). Furthermore, deficiency of HO-1 in humans results in hemolytic anemia, which is characterized by the presence of fragmented RBCs and intravascular hemolysis (Yachie et al. 1999). Furthermore, HO-1-/- mice have progressive anemia due to the absence of macrophages (Poss and Tonegawa 1997b). Interestingly, bone marrow transplantations in HO-1-/- mice successfully restore normal macrophage levels and the ability to catabolize heme (Kovtunovych et al. 2014). The appearance of circulating RBCs expressing mCherry in our transgenic fish support the observation that HO-1 is transcribed

during erythroid differentiation in mammalian systems (Garcia-Santos et al. 2014) lending additional support to the notion that HO-1a may play a role in blood cell differentiation.

Additionally, the HO-1a transgenic line fish may provide an alternative model to study hematopoietic lineage reprogramming in vivo.

3.5.3 In Vivo Promoter Analysis Suggests Novel Roles for HO-1a in Eye Development

At ~35 hpf we observed HO-1a expression in eye tissue (Figure 3.10) and we were able to determine that the expression was specific to the epithelial cells of the lens (Figure 3.13). While

HO-1 has been shown to have a role in Drosophila eye development (Cui et al. 2008), studies reporting HO-1 expression in vertebrate eye tissue are scarce. However, one study has documented expression within rabbit lens epithelial cells consistent with the hypothesis that HO-

1 plays a protective role during lens formation (Rzymkiewicz, Reddan, and Andley 2000).

Regarding zebrafish HO-1a eye expression, a study focusing on the factor inhibiting hypoxia inducible factor 1α (FIH-1) reported HO-1a expression in zebrafish eye tissue at 26 hpf (So et al.

2014). Of interest, a recent study has suggested that decreasing levels of NRF2 are associated with age related cataract formation, (Gao, Yan, and Huang 2015) the leading cause of human blindness, lending further support to the hypothesis that oxidative stress contributes to cataract formation (Gao et al. 2013, Kalariya et al. 2010, Wu et al. 2014). Thus, our finding that HO-1a is expressed within developing lens epithelial cells suggests a novel role for HO-1 in conferring protection against ROS within developing eye tissues.

3.5.4 Conclusion

The observed co-expression of BVRa and BVRb in tissues that are spatially and temporally associated with hematopoiesis inspires the formation of interesting questions regarding the potential for overlapping functions. However, the presence of additional functional domains within the BVRa isoform also raises the possibility that the two BVR isoforms are playing separate functional roles within the same tissues, e.g. amelioration of oxidative stress (BVRb) versus cellular signaling (BVRa). Finally, the expression of HO-1a in HSC’s and lens epithelial cells, two cell types that are high susceptible to oxidative stress, provides new opportunities for future in vivo studies to investigate novel developmental roles of this enzyme.

Movie 3-1. Circulating RBC's in 2 dpf HO-1a transgenic zebrafish.

Movie 3-2. Fluorescent RBC's in 2 dpf HO-1a transgenic zebrafish.

Movie 3-3. Circulating RBC's in the CHT of 7 dpf HO-1a transgenic zebrafish.

Movie 3-4. Fluorescent RBC's with the CHT of 7 dpf HO-1a transgenic zebrafish.

CHAPTER 4

FUNCTIONAL CHARACTERIZATION OF HO AND BVR GENE REGULATION DURING NORMAL DEVELOPMENT AND IN RESPONSE TO STRESS

4.1 Abstract

In vivo promoter analysis (HO-1a) and in situ hybridization (BVRa and BVRb) has shown the spatial and temporal expression patterns of these genes to be consistent with other genes regulated by the erythroid specific transcription factor GATA-1 (Chapter 3). Additionally, using in situ hybridization and RT-PCR we have shown BVRb to be inducible by Cd exposure

(Chapter 2 and 3). In this chapter we evaluate changes in HO-1, BVRa, and BVRb expression following transient knockdown of GATA-1 and NRF2a, the master regulator of oxidative stress.

The data obtained supports the conclusion that BVRa and BVRb are regulated by GATA-1 in hematopoietic stem cells (HSC’s) and embryonic erythrocytes. In contrast, HO-1a does not appear to be regulated by GATA-1 in HSC’s. Furthermore, although NRF2a regulates the induction of HO-1a in response to Cd treatment, transient knockdown of NRF2a also had no effect on HO-1a transcriptional activation within the intermediate cell mass (ICM). In fact,

NRF2a morphants displayed evidence of enhanced BVRa and BVRb expression within the ICM and the dorsal aorta respectively. These data not only provide new insights into the differential regulation of zebrafish HO and BVR, but also support recent studies which implicate a role for

NRF2 as a regulator of HSCs.

4.2 Introduction

In Chapter 3 we characterized the spatial and temporal expression patterns of zebrafish HO-1a,

BVRa and BVRb during early development and in later larval stages under control and Cd- induced conditions. All three genes consistently displayed spatiotemporal expression patterns consistent with HSC’s that eventually differentiate into erythrocytes (red blood cells, RBC’s).

Recent findings have demonstrated that HO-1 transcripts increase in murine RBC’s during erythropoiesis (Garcia-Santos et al. 2014). Additionally, zebrafish BVRb has been shown to be expressed within the intermediate cell mass (ICM) during early development and this expression appears to be dependent on GATA-1, the master regulator of erythropoiesis, based on studies using mutant fish and GATA-1 antisense morpholinos (Galloway 2005). Our HO-1a in vivo data

(Chapter 3) support the findings of Garcia-Santos et al (Garcia-Santos et al. 2014), which can be explained by the conserved nature of the erythroid differentiation program (Detrich 1995). What is of particular interest is the overlapping expression patterns of both BVR isoforms within the

ICM and the posterior blood island (PBI), locations which are consistent with the developmental progression of erythroid cells.

It has long been understood that both BVR isoforms are distinct in their enzymatic mechanism and evolutionary origins (Yamaguchi et al. 1993). BVRb has a single active reductase domain. In contrast, BVRa has a reductase domain, as well as a DNA-binding domain,

SH2 homology domain and a Ser/Thr/Tyr kinase domain. That both of these enzymes are expressed within the same developmental tissues raises some interesting questions. Namely, what are the mechanisms by which these genes are regulated? To this end the spatial and temporal expression patterns of HO-1a, BVRa, and BVRb are consistent with regulation by

GATA-1, although direct regulation of HO-1 or BVRa by this transcription remains to be confirmed.

Additionally, we have shown BVRb to be induced by multiple pro-oxidants (Chapters 2 and

3), and it has long been known that HO-1 is a target gene of NRF2, the master regulator of the oxidative stress response (Alam et al. 1999). Although induction of stress genes in response to

Cd exposure is also consistent with regulation by MTF-1, a master regulator of metals homeostasis, recent studies have suggested a novel developmental role for NRF2 as a regulator of hematopoietic stem cell homeostasis (Merchant et al. 2011, Tsai et al. 2013). Thus, NRF2 is another excellent candidate transcription factor that may be involved in the regulation of HO-1a and BVRb during hematopoiesis.

It should be noted that the transcription factors often considered to be master regulators of the cellular stress response, which include NRF2 and MTF-1, have also been shown to play important roles in many basic developmental processes (Hogstrand C 2008, Hahn et al.). Thus, many mechanisms of developmental toxicity are often explained by disruption of the constitutive, developmental role of these transcription factors by unexpected cellular stress. This disruption is often the result of competition in the functional outcome of these transcription factors as they attempt to mediate the developmental processes, while simultaneously regulating a different battery of genes involved in the stress response. Given the important role for NRF2 in mediating the cellular response to oxidative stress, and the sensitivity of RBC’s to oxidative stress, confirmation of the role of NRF2 in transcriptional regulation of these two enzymes during hematopoiesis will aid future studies investigating potential mechanisms of developmental toxicity.

In this chapter, I investigate the role of GATA-1 in the spatiotemporal regulation of HO-1a,

BVRa, and BVRb expression during hematopoiesis. Furthermore, I also investigate the potential role of NRF2a, the NRF2 paralog responsible for mediating the oxidative stress response in zebrafish, in the regulation of HO-1a, BVRa, and BVRb during hematopoiesis and in response to cellular stress. Spatial and temporal expression patterns of HO-1a, BVRa and BVRb in

GATA-1 and NRF2a morphants during early development in control and Cd challenged embryos were evaluated using in situ hybridization, in vivo promoter analysis and real-time RT-PCR. The results of these experiments provide novel insights regarding the differential regulation of these zebrafish heme degradation genes during development.

4.3 Materials and Methods

4.3.1 Fish Husbandry

The TL (Tupfel/Long fin mutations) wild-type strain of zebrafish was used for all experiments.

Fertilized eggs were obtained from multiple group breedings from a Mass Embryo Production

System (MEPS; Aquatic Habitats, Apopka, FL) with ~200 fish at a ratio of 2 female per 1 male fish. Pairwise breeding of HO-1a:mCherry transgenics was performed with wild-type TL zebrafish. Procedures used in these experiments were approved by the Animal Care and Use

Committee of the University of Alabama, Tuscaloosa, Alabama, USA.

4.3.2 O-dianisidine Staining For Detection of Hemoglobin o-dianisidine (CAS #119-90-4) was purchase from Alfa Aesar (MA, USA). o-dianisidine in the presence of hemoglobin produces a distinct orange stain via the peroxidase activity of hemoglobin (O'brien 1961) and thus serves as a useful marker for detecting the presence of

RBC’s. For o-dianisidine staining we followed the methods of (Detrich 1995). Briefly, 500 μl of o-dianisidine solution consisting of 70% ETOH, 0.01 M sodium acetate, and 3% H2O2 was added to live, dechorionated embryos for ~5 minutes in the dark. Following incubation, embryos were fixed with 4% paraformaldehyde (PFA) followed by extensive washing in phosphate buffered saline with 0.1% tween (PBST). Fixed embryos were stored in glycerol until imaging.

4.3.3 Whole Mount In Situ Hybridization

Developing zebrafish embryos were collected at 24, 48, and 96 hpf. 24 hpf embryos were dechorionated using forceps. All embryos were fixed in 4% PFA, stored at 4 °C overnight and rehydrated the following day. A 3% hydrogen peroxide (H2O2)/ 0.5% potassium hydroxide

(KOH) solution was used to remove pigmentation at 48 and 96 hpf as previously described by

Thisse and Thisse (Thisse 2007). Procedures for whole mount in situ hybridization (WISH) were performed essentially as described by (Thisse 2007). Fixed embryos were rehydrated and permeabilized in a 10 µg/ml solution of proteinase K (Roche Diagnostics) to allow for probe entry. Prior to the addition of probes, embryos were subjected to a pre-hybridization solution which contained yeast tRNA (2 µg/ml), heparin (50 µg/ml), 50% deionized formamide, 5x SSC,

0.1% Tween 20, and 1 M citric acid between 2-4 hours at 70°C. Sense or antisense probes (~75 ng) suspended in pre-hybridization solution were added and allowed to incubate overnight at 70

°C. The next day, embryos were washed and incubated with blocking buffer for 4 hours. After 4 hours the blocking buffer was removed and replaced with Anti-Digoxigenin-AP, Fab fragments

(Roche Diagnostics) made up in blocking buffer (1:10,000) and incubated overnight at 4 °C.

The following day the antibody was removed and the embryos were washed 6 times in (PBST).

To eliminate background staining at later time points, a group of embryos were processed in

parallel with the experimental group up till the pre-hybridization step, at which point a 1:1000 dilution of the antibody was added to the embryos which were subsequently rocked at room temperature for 2 hours and stored overnight at 4°C to saturate non-specific binding sites

(Chitramuthu and Bennett 2013). This pre-adsorbed antibody was then further diluted to a final ratio of 1:10,000 in blocking buffer the next day and added to the embryos which were incubated overnight at 4 °C. Following antibody removal and PBST washes, embryos were further washed in alkaline Tris buffer and stained with BM-Purple AP Substrate precipitating (Roche

Diagnostics). Zebrafish embryos were visualized with a Nikon AZ100M microscope and photographed using a Nikon Digital Sight DS-F-1 camera.

4.3.4 GATA-1 and NRF2a Transient Knockdown via Morpholino

Morpholinos (MOs) designed to block initiation of translation of zebrafish GATA-1 and NRF2a were synthesized by Gene Tools, LLC (Philomath, OR). The GATA-1 MO (5'-

CTGCAAGTGTAGTATTGAAGATGTC-3') was a generous gift from Dr. Leonard Zon. The previously described NRF2a MO (5'- CATTTCAATCTCCATCATGTCTCAG -3') was obtained from Gene Tools, LLC (Timme-Laragy 2012, Kobayashi and Katoh 2002). The standard control morpholino (5'-CCTCTTACCTCAGTTACAATTTATA-3') from Gene Tools was used to account for any nonspecific effects associated with microinjection. The control and NRF2a MOs are fluorescein tagged for screening purposes to guarantee that only successfully injected embryos were used for the subsequent experiments. The NRF2a MO was diluted to 0.18mM and the GATA-1 MO was diluted to 0.5 mM in nuclease-free water. The control MO was diluted to

0.5 mM and 0.18 mM to serve as injection controls for both transcript-specific MOs. A

Narishige IM-300 microinjector was used to inject 2.1 nl of morpholino into the yolk of two- to

four-cell stage embryos. Injection volumes were calibrated by injecting solutions into mineral oil and measuring the diameter of the sphere with a stage micrometer (volume = 4/3r3; 160 lm diameter is equivalent to 2.1 nl). At 3 hours post-fertilization (hpf), embryos were sorted to remove unfertilized eggs or abnormally developed embryos. MO injection studies utilized wild- type TL embryos for the WISH studies. The HO-1a:mCherry transgenic line was also used for both in vivo promoter analysis and WISH after MO injection.

4.3.5 Cd Dosing of NRF2a Morphants

NRF2a morphants, as well as non-injected embryos and embryos injected with the standard control morpholino, were challenged with 50 µM Cd for 6 hours starting at 96 hpf. Following challenges, embryos were immediately fixed in 4% PFA as described above.

4.4. Results

4.4.1 Effects of GATA-1 Knockdown on BVRa and BVRb Expression

To determine if GATA-1 has a regulatory role for BVRa, and to further evaluate the regulatory role that GATA-1 has on BVRb, we injected 2-4 cell embryos with a GATA-1 MO as described by Galloway et al. (Galloway 2005). To confirm the effectiveness of the knockdown, we assessed RBC production using the hemoglobin specific stain o-dianisidine at 48, 72, and 96 hpf.

At 48 hpf a large number of circulating primitive RBCs are easy to visualize within the yolksac circulation valley of non-injected embryos, as well as within control morphants (Herbomel

1999). Conversely, GATA-1 morphants had a nearly complete loss of RBCs at 48 hpf (Figure

4.1), as well as at 72 and 96 hpf (data not shown).

Figure 4.1. GATA-1 Knockdown Results in Decreases in RBC’s. RBCs were stained using o-dianisidine in non-injected, control MO injected, and GATA-1 morphants at 48, 72, and 96 hpf. Shown above are results at the 48 hpf time point. These primitive RBCs flow in a depression of the yolk surface termed the yolksac circulation valley (Herbomel 1999). The direction of circulation is indicated with blue arrows.

Next we performed in situ hybridization on GATA-1 morphants at 24, 48, and 96 hpf, timepoints which span the “primitive” and “definitive” waves of hematopoiesis (Figure 4.2, 4.3,

& 4.4). Knockdown of GATA-1 resulted in a complete loss of BVRb expression within the ICM at 24 hpf (Figure 4.2). However, while BVRa expression was significantly diminished in

GATA-1 morphants some expression persisted within the posterior region of the ICM at 24 hpf.

Figure 4.2. GATA-1 Knockdown Attenuates BVRa and BVRb Expression During Primitive Hematopoiesis. In situ hybridization at 24 hpf in non-injected embryos (left), control morphants (middle), and GATA-1 morphants (right). Embryos were injected with a GATA-1 MO and the standard control MO at a concentration of 1.0 mM. In addition to sense probe controls, a group of embryos with no probe were processed in parallel as a second control (data not shown). No probe controls showed no background staining. Abbreviations used are as follows: ICM = intermediate cell mass, MO = morpholino.

At 48 hpf, BVRb transcripts were detected in circulating RBCs in non-injected and control morphants, while BVRb expression was no longer detectable in RBC's of GATA-1 morphants

(Figure 4.3). Conversely, BVRa transcripts were not detected at this time point in non-injected embryos nor the control or GATA-1 morphants (Figure 4.3).

Figure 4.3. GATA-1 Knockdown Decreases BVRb Expression in Primitive RBC’s at 48 hpf. In situ hybridization at 48 hpf in non-injected embryos (left), control morphants (middle), and GATA-1 morphants (right). Both the GATA-1 MO and the standard control MO were injected at a concentration of 1 mM. Abbreviations used are as follows: RBC’s= red blood cells; MO= morpholino.

At 96 hpf, GATA-1 morphants had stronger staining for BVRb in the liver, heart, kidney, and

CHT compared to non-injected and control MO-injected embryos (Figure 4.4). Although BVRa expression was not detected in non-injected fish at 96 hpf, GATA-1 morphants displayed a modest increase in staining for BVRa within the CHT (Figure 4.4, inset).

Figure 4.4. Effects of GATA-1 Knockdown on BVR Expression During Definitive Hematopoiesis. In situ hybridization at 96 hpf in non-injected embryos (left), control morphants (middle), and GATA-1 morphants (right). BVRa was not detected at 96 hpf in non-injected embryos, however, BVRa expression was faintly detected within the CHT of GATA-1 morphants (inset far right). Conversely, while BVRb expression was detected in the CHT of non-injected embryos, GATA-1 morphants showed stronger staining within the CHT as well as the heart, kidney, and liver. Embryos were injected with a GATA-1 MO and the standard control MO at a concentration of 1.0 mM. In addition to sense probe controls, a group of embryos with no probe were processed in parallel as a second control (data not shown). CHT= caudal hematopoietic tissue, H= heart, L= liver, K= kidney, MO= morpholino.

4.4.2 Effects of GATA-1 Knockdown in HO-1a:mCherry Transgenic Fish

To investigate the potential novel role for GATA-1 in the regulation of HO-1a, we injected HO-

1a:mCherry transgenic embryos with the GATA-1 MO. mCherry expression within the ICM did not seem to be altered in GATA-1 morphants (Figure 4.5). However, at 48 and 72 hpf rather than having circulating fluorescent RBC’s, GATA-1 morphants displayed strong, stationary fluorescence within the CHT (movie 4.1-4.5). Although there are no reports in the literature of off-target effects resulting from the GATA-1 MO, the zebrafish yquem mutant displays auto- fluorescence within the ICM (Nasevicius and Ekker 2000) similar to the mCherry expression observed in the HO-1a transgenic line. Auto-fluoresence in the yquem mutant is the result of an inherited homozygous mutation in uroporphyrinogen decarboxylase (UROD) which causes an accumulation of photosensitive porphyrins (Wang et al. 1998). To determine if GATA-1 knockdown is causing an accumulation of auto-fluorescent porphyrins we injected wild-type

(WT) control embryos with the GATA-1 MO and monitored them for fluorescence at 24 and 48 hpf (Figure 4.5). We did not observe any auto-fluorescence in the WT GATA-1 morphants at either time point. This suggests that the accumulating fluorescent signal that we see in our HO-

1a transgenic GATA-1 morphants is the result of mCherry expression driven by the HO-1a promoter and that GATA-1 knockdown does not decrease HO-1a promoter driven mCherry expression within the ICM (Figure 4.5).

Figure 4.5. GATA-1 Knockdown Does Not Decrease HO-1a Promoter Driven mCherry Expression within the ICM. Fluorescent images of Tg(ho-1a:mCHerry;cmlc2-eGFP)mjj1 at 24 (top left) and 72 hpf (top right) in comparison to Tg(ho-1a:mCHerry;cmlc2-eGFP)mjj1 GATA-1 morphants (middle).Wild type GATA-1 morphants at 24 and 48 hpf (bottom) were used to confirm the lack of auto-fluorescence. Merged images (mCherry and brightfield) are displayed in the bottom right corners of fluorescent images to help illustrate anatomical location of focus. CA = caudal artery, CV = caudal vein, hpf = hours past fertilization, ICM = intermediate cell mass, ISVs = intersegmental veins, MO = morpholino, WT = wild-type.

HO-1a:mCherry expression in GATA-1 morphants was documented through the 120 hpf timepoint, at which point we noticed an increase in circulating RBC’s (movie 4.8), some of which were fluorescent (movie 4.9). However, the number of circulating RBCs in GATA-1 morphants was drastically lower in comparison to non-injected fish (movies 4.6 & 4.7). To determine whether the strong fluorescent signal from within the CHT was a result of impaired blood flow and that subsequent appearance of circulating fluorescent RBC’s was likely the result of the weakening effects of the GATA-1 MO, OD staining was performed on non-injected transgenic fish as well as the GATA-1 transgenic morphants at 5 dpf (Figure 4.6). In agreement with the movies showing many circulating RBCs in the non-injected transgenic fish, RBCs appeared to be evenly dispersed within the caudal artery (CA), caudal vein (CV), and intersegmental veins (ISVs) in non-injected fish. Conversely, hemoglobin staining in GATA-1 morphants showed an accumulation of RBCs within the CV and a reduction in RBC’s within the

CA and ISVs (Figure 4.6)

Figure 4.6. GATA-1 Knockdown Alters Spatial Distribution of RBC’s. Hemoglobin staining in non-injected fish at 5 dpf revealed evenly dispersed RBCs in the CA, CV, and ISVs. Conversely, hemoglobin staining in GATA-1 MO injected HO-1a transgenic fish showed an accumulation of RBCs within the CV at this same time. CA= caudal artery, CV= caudal vein, hpf= hours past fertilization, ISVs- intersegmental veins, MO= morpholino, WT= wild-type.

100

4.4.3 Effects of NRF2a Knockdown on BVR Expression during Primitive Hematopoiesis

NRF2 is the master regulator of oxidative stress and upregulates a battery of protective genes in response to oxidative stress including HO-1 (Alam et al. 1999). Recently however, a different role as a regulator of HSC’s has been suggested (Merchant et al. 2011, Tsai et al. 2013). The purported role of BVR in the recycling of the potent antioxidant BR coupled with the overlapping expression patterns of BVRa and BVRb raises the question of whether targeted knockdown of NRF2a would alter zebrafish RBC production and BVR gene expression. To determine if transient knockdown of zebrafish NRF2a alters RBC production we stained for hemoglobin in control and NRF2a morphants. While o-dianisidine staining is not quantitative,

NRF2a morphants appeared to display greater numbers of primitive RBC’s at 48 hpf in comparison to non-injected and control morphants (Figure 4.7).

Figure 4.7. NRF2a Knockdown Increases RBC Count at 48 hpf. o-dianisidine staining for hemoglobin was performed in non-injected embryos (left), control morphants (middle), and NRF2a morphants (right) at 48 hpf. NRF2a morphants appeared to have more RBCs within the yolksac circulation valley in comparison to non-injected and control morphants. hpf = hours past fertilization, MO = morpholino, RBCs = red blood cells.

101

Since NRF2a knockdown seemed to enhance the amount of RBC’s, we wanted to determine if NRF2a morphants had altered expression patterns of BVRa and BVRb within the ICM or other hematopoietic tissues (Figure 4.8). Transient knockdown of NRF2a seemed to enhance BVRa expression within the ICM at 24 hpf in comparison to non-injected embryos and embryos injected with the standard control MO (Figure 4.8). Due to the strong BVRb staining that was detected within the ICM in non-injected and control morphants at 24 hpf, it is hard to accurately determine whether or not NRF2a knockdown enhanced BVRb expression within the ICM.

Figure 4.8. NRF2a Knockdown Results in an Increase in BVRa Expression within the ICM at 24 hpf. BVRa and BVRb expression at 24 hpf in non-injected embryos (left), control morphants (middle), and NRF2a morphants (right). In addition to sense probe controls (data not shown), a group of embryos with no probe were processed in parallel as a second control (data not shown). No probe controls and sense probes displayed no background staining. ICM=intermediate cell mass, hpf= hours past fertilization, MO= morpholino.

However, we did notice enhanced BVRb expression within the dorsal aorta (DA) (Figure 4.9), the location in which definitive HSC’s are produced (Thompson et al. 1998, Kalev-Zylinska et al. 2002, Burns et al. 2002).

102

Figure 4.9. NRF2aMmorphants Express BVRb Transcripts within the Dorsal Aorta. Dorsal view showing BVRb expression within the DA of NRF2a morphants at 24 hpf (right) in comparison to control morphants (middle), and non-injected embryos (left). In addition to sense probe controls (data not shown), a group of embryos with no probe were processed in parallel as a second control (data not shown). No probe controls and sense probes displayed no background staining. DA = dorsal aorta, hpf = hours past fertilization, MO = morpholino.

4.4.4 Effects of NRF2a Knockdown on BVR Expression during Definitive Hematopoiesis

Since NRF2a knockdown appeared to increase BVRa and BVRb expression during primitive hematopoiesis, we next wanted to determine how knockdown of this transcription factor effected their expression during definitive hematopoiesis. Neither BVRa nor BVRb transcripts seemed to be effected by NRF2a knockdown at 48 hpf (Figure 4.10).

103

4.4.5 Effects of NRF2a Knockdown on BVR Expression in Cd Challenged Embryos

Normal expression patterns of BVRa and BVRb do not appear to be altered in NRF2a morphants at 48 hpf (Figure 4.10).

Figure 4.10. NRF2a Knockdown Does Not Alter BVRa and BVRb Expression at 48 hpf. In situ hybridization at 48 hpf in non-injected embryos (left), control morphants (middle), and NRF2a morphants (right). In addition to sense probe controls (data not shown), a group of embryos with no probe were processed in parallel as a second control (data not shown). No probe controls showed no background staining. MO = morpholino, hpf = hours past fertilization, RBCs = red blood cells.

However, Chapters 2 and 3 demonstrated that BVRb was inducible by multiple pro-oxidants at later developmental periods. To further evaluate the possibility that NRF2a is involved in the

BVR stress response, we evaluated changes in BVR expression patterns by WISH in Cd challenged NRF2a morphants. At 96 hpf, BVRa transcripts remained undetected. In contrast,

BVRb staining seemingly showed equal amounts of expression within the liver, heart, kidney, and CHT of NRF2a morphants compared to non-injected embryos and those injected with the standard control MO (Figure 4.11A). BVRa staining did not change in NRF2a morphants as a result of Cd exposures (Figure 4.11B). However, exposure to Cd resulted in an increase in

104

BVRb staining intensity within the visceral tissues (Figure 4.11B). Interestingly, Cd-exposed

NRF2a morphants did not appear to have as strong of a BVRb signal within the heart, liver, kidney and intestines in comparison to non-injected and control morphants. Additionally, and perhaps most interestingly, Cd challenged NRF2a morphants did not have BVRb staining within the CHT (Figure 4.11B).

105

Figure 4.11. NRF2a Knockdown Results in a Decrease in BVRb Expression in Cd Challenged Embryos at 96 hpf. BVRa and BVRb expression at 96 hpf in non-injected embryos (left), control morphants (middle), and NRF2a morphants (right) in control embryos (A) or embryos challenged for 6 hours with 50 µM Cd starting at 96 hpf (B). In addition to sense probe controls (data not shown), a group of embryos with no probe were processed in parallel as a second control (data not shown). No probe controls showed no background staining. CHT= caudal hematopoietic tissue, H= heart, I= intestines, L= liver, K= kidney, MO= morpholino.

106

4.4.6 NRF2a Knockdown Attenuates HO-1a Liver Expression in Cd Challenged Embryos

GATA-1 knockdown did not result in a decrease in HO-1a promoter driven mCherry expression within hematopoietic tissues (Figure 4.5). Since NRF2 is a known transcriptional regulator of

HO-1 (Alam et al. 1999) and has been implicated in the regulation of HSC’s (Tsai et al. 2013,

Merchant et al. 2011), NRF2a knockdown experiments were performed to determine if HO-

1a:mCherry expression within the ICM would be attenuated. NRF2a morphants did not show a decrease in mCherry expression in comparison to non-injected embryos or control morphants within the ICM (Figure 4.12). To ensure that the NRF2a MO is working properly, we performed in situ hybridization to detect HO-1a transcripts in Cd-challenged NRF2a morphants at 96 hpf

(Figure 4.12). As expected, NRF2a knockdown resulted in decreased WISH HO-1a staining in

Cd-challenged NRF2a morphants.

107

Figure 4.12. Effects of NRF2a Knockdown on HO-1a Expression in Control and Cd Challenged Fish. Fluorescent images of Tg(ho-1a:mCHerry;cmcl2-eGFP)mjj1embryos at 24 hpf (top row) demonstrate the NRF2a knockdown has no effect on HO-1a:mCherry expression within the ICM. The control and NRF2a MOs are fluorescein tagged for screening purposes (images provided bottom right corner of top row). HO-1a induction at 96 hpf in Cd-challenged larvae was attenuated by the NRF2a MO. Larvae were exposed to 50 µM Cd for 6 hours (bottom row). ICM = intermediate cell mass, L = liver, MO = morpholino.

4.5 Discussion

Using antisense MO’s we determined the effects of GATA-1 and NRF2a knockdown on HO-1a,

BVRa, and BVRb expression in control and Cd challenged embryos. Transient knockdown of these master transcriptional regulators provided new insights in the spatiotemporal regulation of

BVRa and BVRb expression, and the induction of BVRb by pro-oxidants. However, unlike the

BVR isoforms, HO-1a expression in hematopoietic tissues does not appear to be regulated by

108

either transcription factor. These observations and their relevance to the current literature are discussed below.

4.5.1 GATA-1 Morphants Have Reduced BVRa Expression During Primitive Hematopoiesis

Loss of GATA-1, either via MO knockdown or in the vlade tepes bloodless mutant, results in loss of BVRb expression (Galloway 2005) during primitive hematopoiesis. As I have shown that

BVRA, BVRb and HO-1a are expressed within the ICM region during this same time period, I hypothesized that GATA-1 may play a regulatory role in all three heme degradation genes (HO-

1a, BVRa and BVRb). While GATA-1 knockdown resulted in a complete loss of BVRb expression within the ICM, BVRa expression was only slightly diminished (Figure 4.2). We found this surprising as BVRb is expressed at an order of magnitude higher than BVRa during development (Figure 2.5). However, these results are consistent with models that describe an erythroid to myeloid lineage switch (Graf 2002). The ICM contains common myelo-erythroid progenitors (CMPs) which will differentiate into either erythroid or myeloid cells. Loss of

GATA-1may result in cell death of erythroid progenitors and subsequent expansion of myeloid progenitors (Model 1), or CMPs differentiate based on the expression levels of GATA-1 and

PU.1 (Model 2) (Graf 2002). GATA-1 and PU.1 antagonize one another, and loss of GATA-1 results in an increase in PU.1 expression and subsequent coercion of ICM progenitors towards a myeloid lineage (Rhodes et al. 2005, Galloway 2005). The continued expression of BVRa in

GATA-1 morphants at 24 hpf leads to a new hypothesis that BVRa may be regulated by both

GATA-1 and PU.1. Alternatively, GATA-1 null mice and GATA-1 deficient cell lines show numerous genes with only reduced transcript levels, similar to our observation regarding BVRa

(Weiss and Orkin 1995, Pevny et al. 1995). Interestingly, GATA-2 has been shown to be able to

109

regulate some erythroid genes in GATA-1 mutant mice during embryonic hematopoiesis

(Takahashi et al. 2000), suggesting the BVRa may be co-regulated by GATA-1 and -2. Future studies may address whether PU.1 or GATA-2 are capable of regulating BVRa expression patterns.

4.5.2 GATA-1 Knockdown Increases BVR Expression During Definitive Hematopoiesis

GATA-1 knockdown did not alter the expression of either BVR isoform at 48 hpf during the definitive wave of hematopoiesis (Figure 4.3). To further investigate the potential relationship between GATA-1 and BVR expression, we evaluated the effects of GATA-1 knockdown at a timepoint in which definitive hematopoiesis is well under way, i.e., 96 hpf (Figure 4.4). While

BVRa was not detected by in situ hybridization at 96 hpf, GATA-1 morphants displayed an increase in BVRa expression within the CHT at this timepoint. Conversely, while BVRb transcripts were detected in the CHT of non-injected and control MO-injected embryos, BVRb staining appeared to be significantly enhanced within the CHT, heart, kidney and liver of GATA-

1 morphants at this timepoint (Figure 4.4). One explanation for this data would be that as the effects of GATA-1 knockdown diminish, we are seeing the re-initiation of the hematopoietic program, i.e., we are seeing the expression levels that we would normally see 1-2 days earlier.

However, it has been demonstrated that a third transcription factor (TIF-1λ) is involved in determining erythroid and myeloid fate decisions, and that these decision are influenced by the location of blood cell populations (Monteiro, Pouget, and Patient 2011). Thus, determining if

TIF-1λ may be involved in the transcriptional regulation of BVRa should be addressed in future studies.

110

4.5.3 GATA-1 and NRF2a Morphants Do Not Have Altered HO-1a:mCherry Expression within the ICM

Transient knockdown of GATA-1 resulted in a decrease in the expression of both BVRa and

BVRb within the ICM (Figure 4.3). Therefore we hypothesized that GATA-1 knockdown would result in a similar loss of mCherry expression within the ICM driven by the HO-1a promoter

(Figure 4.5). However, injection of GATA-1 MO did not appear to attenuate HO-1a:mCherry fluorescence. In fact, GATA-1 morphants appeared to display stronger levels of mCHerry expression within the ICM. We confirmed the fluorescence was indeed mCherry and not the result of auto-fluorescence associated with disruption of heme synthesis and the resulting accumulation of porphyrins (Nasevicius and Ekker 2000, Wang et al. 1998, Ablain et al. 2015).

Visualization of mCherry expression in circulating RBCs (Movies 4.1-4.9) demonstrates that

GATA-1 morphants have restricted blood flow in comparison to non-injected transgenic fish.

Furthermore, around 5 dpf the rate of blood flow begins to increase, presumably as the effects of the GATA-1 MO subside. Using a hemoglobin specific stain (o-dianisidine) we detected an accumulation of RBC’s within the same regions of the tail in which there is an accumulation of fluorescent cells. Collectively, these experiments suggest that GATA-1 knockdown is not decreasing mCherry expression but may result in vascular phenotypes that inhibit circulation.

Although NRF2 is a known HO-1 regulator (Alam et al. 1999), surprisingly, transient knockdown of this transcription factor also did not weaken mCherry expression within the ICM of our transgenic line (Figure 4.12). However, we were able successfully inhibit HO-1a expression in liver tissue in Cd challenged embryos by NRF2a knockdown. Collectively, these results suggest that a different transcription factor is involved in regulating HO-1a expression within hematopoietic tissues.

111

4.5.4 NRF2a Knockdown Enhances BVR Expression during Primitive Hematopoiesis

The effects of NRF2a knockdown on BVR expression (Figure 4.10 & 4.11) and HO-1a expression (Figure 4.12) raise several interesting questions that warrant further attention. NRF2a morphants demonstrated an increase in BVRa staining in the ICM at 24 hpf (Figure 4.7), as well as an increase in RBCs within the yolksac (Figure 4.6). Furthermore, NRF2a knockdown resulted in an increase in BVRb expression within the DA (Figure 4.8). These observations are consistent with the report that NRF2 null mice have an expansion of HSC and progenitor cell compartments (Tsai et al. 2013). Alternatively, as zebrafish possess two copies of NRF2

(NRF2a and NRF2b) which have undergone subfunction partitioning (Timme-Laragy 2012), the ability of NRF2b to regulate either BVR isoform should be considered. While NRF2a is thought to act primarily as an activator and the NRF2b paralog is thought to act primarily as a repressor,

Timme-Laragy et al reported that 198 of 508 total probes were upregulated in NRF2a morphants

(Timme-Laragy 2012). This would be in agreement with our data showing an increase in BVRa and BVRb expression in NRF2a morphants at 24 hpf (Figure 4.8 & 4.9). Interestingly, at later timepoints (96 hpf) NRF2a morphants did not show a difference in BVR expression under normal conditions (Figure 4.11A). However, under Cd challenged conditions BVRb expression was not induced to the same extent as non-injected and control morphants (Figure 4.11B).

Future studies should address the effects of NRF2b knockdown on BVRa and BVRb expression.

4.5.5 Conclusions and Remaining Questions

This chapter presents novel observations on the transcriptional regulation of zebrafish heme degradation genes. GATA-1 has been confirmed to be required for the expression of BVRa and

BVRb during primitive hematopoiesis. In contrast, while NRF2a is not required for expression

112

of the BVR isoforms during primitive hematopoiesis, loss of NRF2a appears to significantly enhance BVRa and BVRb expression. In contrast to observations related to BVR expression, the transcription factor required for the regulation of HO-1a in hematopoietic tissues remains a mystery. In Chapter 5, I will discuss future studies that may be useful in addressing some of the new questions and hypotheses related to the regulation of these heme degradation genes during hematopoiesis and in response to oxidative stress.

Movie 4-1. Circulating RBC's in 2 dpf HO-1a transgenic zebrafish

Movie 4-2. Fluorescent RBC's in 2 dpf HO-1a transgenic zebrafish

Movie 4-3. Circulating RBC's in 2 dpf HO-1a transgenic zebrafish

Movie 4-4. Fluorescent RBC's in 2 dpf HO-1a GATA-1 morphants

Movie 4-5. Circulating RBC's in 3 dpf HO-1a transgenic zebrafish

Movie 4-6. Fluorescent RBC's in 3 dpf HO-1a GATA-1 morphants

Movie 4-7. Circulating RBC's in 5 dpf HO-1a transgenic zebrafish

Movie 4-8. Fluorescent RBC's in 5 dpf HO-1a GATA-1 morphants

113

CHAPTER 5

CONCLUSION

A classic toxicological approach was initially taken with the goal of characterizing the expression of HO and BVR genes in response to pro-oxidant exposure. However, several interesting observations altered the direction of the study leading to a more significant focus on the characterization of potential developmental roles for these genes. Although the results certainly support a functional role for HO and BVR in response to chemically-induced oxidative stress, the more interesting results suggest other roles in mediating the effects of endogenous oxidative stress during development. In this chapter I summarize the data and provide a historical context for the results, as well as further discuss remaining questions and describe future directions for this project.

5.1 Summary

Chapter 2 focused on the quantitative assessment of zebrafish heme degradation gene expression

(HO-1a, HO-1b, HO-2a, HO-2b, BVRa, and BVRb) during development and within adult tissues under normal conditions and in response to exposure to different pro-oxidants. A comparative evaluation of the effects of Cd exposure on expression of genes involved in maintaining iron and heme homeostasis (ferroportin 1, hepcidin, and hemopexin) was also performed. While interesting differences in expression were noted, several findings stood out as being important.

114

Our data show BVRb, the lesser studied BVR isoform, to be the predominantly expressed isoform during early development (24-120 hpf) and to be inducible by proxidants with different mechanisms of toxicity (Cd, hemin, and tBHQ). Additionally, BVRb, as well as HO-1a, were expressed at much higher levels in male liver tissues, which could potentially provide a novel mechanism of the sex-specific differences associated with Cd toxicity. Finally, these experiments represent the first evaluation of developmental and tissue specific expression patterns of the zebrafish heme oxygenase paralogs, thereby providing strong initial support for subfunction partitioning of these genes.

Chapter 3 characterized the spatial and temporal expression patterns of HO-1a, BVRa and

BVRb during normal development and in response to oxidative stress. In situ hybridization

(BVRa and BVRb) was used to demonstrate that the spatial and temporal expression patterns of both BVR isoforms are consistent with known hematopoietic regulators during primitive hematopoiesis. However, during definitive hematopoiesis the data suggest that only the BVRb isoform maintains a role in the continued differentiation of RBCs. Furthermore, in vivo promoter analysis demonstrates that the first gene of the heme degradation pathway (HO-1a) is expressed in a manner similar to BVRa and BVRb, i.e., within the ICM and at later time points within circulating primitive RBC’s. Additionally, using in situ hybridization, in vivo promoter analysis and histological examinations we show HO-1a to be expressed within lens epithelial cells during early development.

Chapter 4 evaluated the role of two potential transcriptional regulators of HO and BVR

(GATA-1 and NRF2a) during normal development and in response to Cd induced stress. A combination of antisense morpholinos and in situ hybridization was used to demonstrate that

BVRb expression is dependent on GATA-1 knockdown at 24 hpf, while BVRa expression is still

115

detectable although significantly reduced in intensity. At an intermediate timepoint (48 hpf),

BVRa is not detectable but the strong expression of BVRb in circulating RBC’s is attenuated in

GATA-1 morphants. In contrast to the 48 hpf observations, at 96 hpf the GATA-1 morphants display evidence of enhanced expression of both BVRa and BVRb. While BVRa and BVRb expression patterns were altered in GATA-1 morphants, knockdown of this transcription factor did not appear to alter mCherry expression driven by the HO-1a promoter within hematopoietic tissues (ICM and primitive RBC’s). While this data certainly suggests that another transcription factor may be regulating HO-1a expression in hematopoietic tissues, this data is also consistent with that reported by Fraser et al (Fraser et al. 2015). Specifically, Fraser et al show that HO-1 deficient mice have an increased RBC lifespan as a result of decreasing amounts of macrophages within bone marrow, spleen, and liver (Fraser et al. 2015). Experiments to determine if HO-1a

MO knockdown alters macrophage numbers would be easy to perform in zebrafish using Sudan

Black, a common granulocyte stain, and tail wound assays (Cvejic et al. 2008, Walters et al.

2010).

Similar experiments were performed to determine if NRF2a knockdown altered the Cd- induced induction of HO-1a, BVRa, and BVRb expression during development. As expected,

NRF2a morphants clearly had attenuated HO-1a expression. Of particular interest, BVRa and

BVRb expression increased during primitive hematopoiesis (24 hpf) in NRF2a morphants, supporting recent studies which suggest a novel role for NRF2 as a regulator of HSC’s.

5.2 Further Characterization of Roles for HO-1a, BVRa, and BVRb in Hematopoiesis

GATA-1 morphants had reduced levels of both BVRa and BVRb within the ICM at 24 hpf

(Chapter 4). While this loss of expression at 24 hpf is in agreement with GATA-1 regulating

116

these genes during primitive hematopoiesis, the following questions remain. First, the decrease in BVRa expression is not complete; specifically, its expression becomes restricted towards the posterior end of the ICM in GATA-1 morphants. This suggests that other transcription factor(s) may be involved in its regulation at this time point. Obvious candidate transcription factors to test would be PU.1 (the master regulator of myeloid development) and TIF1λ (transcriptional modulator which influences erythroid fate decisions). Experiments evaluating regulatory roles of these transcription factors would be highly feasible as there are available GATA-1 and TIF1λ mutant lines (moonshine-TIF1λ mutant and the vlade tepes-GATA-1 mutant, respectively)

(Ransom 2004, Lyons et al. 2002). While a PU.1 mutant line is not available, there is a previously characterized PU.1 MO (Rhodes et al. 2005), as well as a transgenic line in which the

PU.1 promoter drives GFP expression (Hsu et al. 2004). However, we believe that further studies of BVRa and BVRb in the context of hematopoietic development should also incorporate the recent finding that HSC commitment occurs earlier than previously thought (Butko et al.

2015). Runt-related transcription factor 1 (RUNX1) has traditionally served as the earliest marker for HSC commitment from within the DA (Clements and Traver 2013). However, Butko et al. have shown that GATA-2b is expressed genetically upstream of RUNX1 (Butko et al.

2015). Therefore, further characterization of roles for HO-1a, BVRa, and BVRb in hematopoiesis should include evaluations during earlier development (~14-18 hpf), timepoints in which cells of the posterior lateral mesoderm begin to express GATA-2b thus indicating HSC commitment.

117

5.3 Further Characterization of Roles for NRF2 as a HSC Regulator

Loss of NRF2 results in the expansion of HSC and progenitor cell compartments, suggesting that

NRF2 acts as a negative regulator of HSC entry into the cell cycle (Tsai et al. 2013). In agreement with this newly proposed role for NRF2, our data show that knockdown of NRF2a results in an increased expression of both BVRa and BVRb within hematopoietic tissues, as well as an increase in RBC number. However, this finding is somewhat surprising as NRF2 has undergone subfunction partitioning in zebrafish, with NRF2a thought to act as an activator and

NRF2b thought to act as a repressor (Timme-Laragy 2012). While it would certainly be interesting to determine the effects of NRF2b MO knockdown on BVR expression and RBC expansion, numerous questions regarding the proposed developmental role(s) of NRF2 exist.

While Tsai et al suggests that HSC expansion in NRF2 mutants is the result of disruption of normal NRF2 regulation of chemokine receptor signaling (Cxcr4) (Tsai et al. 2013), there are certainly other possibilities. The results presented in this dissertation and our examination of the literature has led us to believe that NRF2 may influence HSC specification by regulating vascular endothelial growth factors (VEGF’s). As discussed in chapter 1, GATA-1 and PU.1 antagonize one another (Galloway 2005, Rhodes et al. 2005). Interestingly, PU.1 MO knockdown results in an increase in VEGF expression within the heart, and conversely GATA-1 knockdown results in a decrease in VEGF heart expression (Helker et al. 2013). In agreement with this, we observed an apparent accumulation of RBC’s within the caudal vein in HO-1a transgenic GATA-1 morphants, presumably as the result of impaired vascular development.

While the increase in BVRb expression that we noted within the DA would be in agreement with the proposed role of NRF2 acting as a regulator of HSC expansion (Tsai et al. 2013), the NRF2a morphants may secrete greater levels of VEGF due to an increase in RBC’s. This is plausible as

118

zebrafish contain nucleated RBC’s which are transcriptionally active (Helker et al. 2013). In this context, roles for RBC’s other than oxygen transport have been reported. Namely, HSC development has been shown to be dependent on blood flow and to be intimately linked with nitric oxide (NO) production (North et al. 2009). This leads to the question of whether NRF2a morphants have increased NO levels or increased levels of nitric oxide synthase. To this end, a zebrafish NRF2a mutant has already been described (Mukaigasa 2012), presenting an opportunity to uncover novel roles for this transcription factor as a HSC regulator.

5.4 Heme Oxygenase Paralogs May Have Underwent Subfunction Partitioning

In chapter 2 were provide the first characterization of developmental and tissue specific expression patterns of the zebrafish heme oxygenase paralogs. Interesting differences in expression during development, within adult tissues, and in response to oxidative stress were noted. These differences suggest that these genes may be differentially regulated and may be performing differing functions. While the existence of a “free” or “uncommitted” heme pool has long been postulated (Ponka 1997), the existence of such a reserve would require tight regulation owing to the toxicity of heme (Khan and Quigley 2011). In this regard using mammalian cell lines, a role for HO-1 as a regulator of heme turnover during erythroid differentiation has recently been suggested (Garcia-Santos et al. 2014). However, further characterization of a role for HO-1 in this context is difficult as the majority of HO-1 null mice die in utero (Poss and

Tonegawa 1997b, Poss and Tonegawa 1997a). This finding extends to humans, as females with

HO-1 microsatellite polymorphisms have an increased chance of having recurrent miscarriages

(Denschlag et al. 2004). To this end, further characterization of developmental roles of these paralogs may provide information regarding developmental roles for HO-1 with direct relevance

119

to human health. The ability to transiently disrupt gene expression using antisense MOs provides one method to further determine differences in the functions of these paralogs. An alternative strategy would be to utilize targeted genome editing strategies such as the CRISPR-

Cas system which has gained widespread acceptance and use in recent years. Concurrent with the popularity of the CRISPR-Cas system, concerns regarding the use of MO’s to study gene knockdown have come under fire as several studies have reported different phenotypes when using MOs in comparison to studies utilizing genome editing strategies (Stainier, Kontarakis, and

Rossi 2015). Our data show HO-1b expression to be negligible during early development

(Chapter 2). However, it should be considered that mRNA levels do not necessarily correlate well with protein levels (Maier, Guell, and Serrano 2009). Nonetheless, these low HO-1b transcript levels do suggest a certain degree of developmental irrelevance making it likely that targeted HO-1a knockouts in zebrafish would be embryonic lethal, as seen in mice. While it would be interesting to determine if HO-1b overexpression could rescue this lethality (or other observed phenotypes), recently it has been shown that gene knockout using CRISPR-Cas can result in genetic compensation (Rossi et al. 2015). This certainly raises the interesting question of whether or not HO-1b would compensate for HO-1a? While we can only speculate on the outcome of HO-1a knockout, the best strategy would be to use both techniques to complement one another.

We also described differences in expression between HO-2 paralogs. HO-2 is known to be highly expressed in brain tissues, thus it was not completely unexpected that we detected high levels of HO-2a and HO-2b within brain tissues. However, we did find it interesting that HO-2b was constitutively expressed during early development while HO-2a transcript levels gradually increased during these early developmental periods. This suggests that these paralogs are

120

performing different functions during early development. Similar experiments to the above proposed experiments regarding discernment of partitioning of function between HO-1a and HO-

1b may yield novel insights regarding the hypothesized role of HO-2 serving as a neuroprotectant and will provide information regarding possible delegation of function between these paralogs.

5.5 Developmental Roles for HO-1a in Zebrafish Eye Development

In chapter 3 we show HO-1a to be expressed within zebrafish lens epithelial cells during early development. This could be another avenue of research with potential relevance to human health. Recently, a known transcriptional regulator of HO-1, NRF2, has been associated with age-related cataract formation, (Gao, Yan, and Huang 2015) the leading cause of human blindness. This lends further support to the hypothesis that oxidative stress contributes to cataract formation (Gao et al. 2013, Kalariya et al. 2010, Wu et al. 2014). The known correlation between oxidative stress and cataract development has led us to speculate that HO-1a may play a role in zebrafish eye development by conferring protection against oxidative stress. Using the previously characterized HO-1a morpholino (Kawahara et al. 2013), new studies could be initiated to determine the effects of transient gene knockdown on eye morphology and the ability to cope with xenobiotic induced oxidative stress.

5.6 Novel Roles for HO-1, BVRa and BVRb during Early Development

The potent antioxidant BR is generated as the final product of heme degradation. Therefore, it is surprising that all the genes involved in this pathway are not equally represented in studies assessing antioxidant roles of BR. The developmental characterization of BVRa and BVRb demonstrates expression of BVRb at nearly an order of magnitude higher than BVRa during

121

early zebrafish development. However, we also noted that both isoforms displayed similar overlapping expression patterns within hematopoietic tissues until approximately 48 hpf. The first gene in the heme degradation pathway, HO-1a was also expressed within hematopoietic tissues during this time frame. These observations raise a number of questions. Namely, what are the physiological reasons necessitating the need for two BV reducing enzymes within hematopoietic tissues during the first 2 days of life? During zebrafish development there is a transition from anoxia tolerance (<24 hours post fertilization, hpf) to declining survivability due to anoxia sensitivity (~24-48 hpf), with anoxia becoming lethal at 52 hpf (Padilla 2001). One explanation for this transition is the occurrence of a highly coordinated change in metabolism from an anaerobic to an aerobic form of energy production which is more conducive to generating reactive oxygen species (ROS) (Mendelsohn 2008a, Mendelsohn 2008b).

Interestingly, the overlapping expression patterns of BVRa and BVRb occur during this same period of metabolic change. While GSH recycling represents the frontline defense against oxidative stress, oscillating levels of GSH and fluctuations in cytosolic redox environment have been reported between 0-48 hpf in developing zebrafish embryos (Timme-Laragy et al. 2013b).

Specifically, during this time period (~0-48 hpf) the cytoplasmic redox environment changes from being highly reduced (0 hpf) to being highly oxidized (~ 3-24 hpf) then transitioning back to a reduced environment by ~48 hpf. This means that the time frame in which cells are most rapidly dividing (between 0-3 hpf) coincides with a period which favors the formation of ROS and in which there are low levels of GSH. In fact, it is not until after 48 hpf that embryos have a favorable redox environment conducive to conferring protection against oxidative stress, i.e., having high levels of GSH. As a highly oxidized redox environment would favor the formation of methemoglobin, it seems reasonable that BVRb (also known as methemoglobin reductase)

122

may function to maintain adequate hemoglobin levels during this time period. Thus, we believe that the high levels of BVRb during early development represent a cellular need to maintain hemoglobin levels. Lending support to this idea, the observed fluctuations in BVRb expression within primitive RBC’s and hematopoietic tissues is consistent with the embryonic to larval globin switch which occurs in zebrafish between 24-36 hpf (Ganis 2012, Brownlie et al. 2003).

Globin switching refers to changes in the production of alpha and beta globins which make up hemoglobin molecules and is thought to allow tolerance towards fluctuating oxygen levels.

Furthermore, globin switching is thought to be regulated by the different waves of hematopoiesis

(primitive and definitive) which are defined by the spatial and temporal expression patterns of erythrocyte precursors. Around 24 hpf, developing zebrafish express a specific set of embryonic globins within the ICM (Brownlie et al. 2003), the same location in which BVRa and BVRb show overlapping expression patterns. However, by 72 hpf most embryonic globin genes cannot be detected via in situ hybridization. Interestingly, between 72-120 hpf expression levels of embryonic globin genes are reinitiated. This renewal of embryonic globin expression is thought to arise from the second hematopoietic wave (the definitive wave of hematopoiesis) (Ganis

2012). These similarities in expression between BVRb and embryonic globins are interesting and further suggest that BVRb is acting to maintain embryonic globin levels by reducing methemoglobin.

Different than the proposed roles for BVRb, in the absence of GSH during early development

BVRa may be functioning to provide protection against ROS by generating BR. Alternatively,

BVRa has been shown to activate HO-1 transcription by transporting cytoplasmic heme into the nucleus (Tudor 2008). Thus, these two genes may be creating a feedback loop to provide adequate levels of heme for incorporation into hemoproteins and to generate BR. Additionally,

123

BV has been shown to act as a dorsal axis determinant in Xenopus (Falchuk et al. 2002,

Montorzi, Dziedzic, and Falchuk 2002). Thus another function for this feedback loop may be to regulate the formation of this morphogen. As previously discussed regarding proposed studies involving the HO parlogs, the best approach to understanding potential developmental roles for

BVRa and BVRb would be to take advantage of the CRISPR-Cas editing strategy (in combination with MO knockdown) to determine the effects of gene knockout on normal development.

5.7 Concluding Remarks

While this initial characterization of all of the isoforms involved in the heme degradation pathway represents the most complete study of its kind, it is far from complete and raises more questions than answers. However, a substantial amount of new information regarding the lesser studied isoforms involved in heme degradation have provided novel insights regarding their transcriptional regulation.

124

REFERENCES

Ablain, J., E. M. Durand, S. Yang, Y. Zhou, and L. I. Zon. 2015. "A CRISPR/Cas9 Vector System for Tissue-Specific Gene Disruption in Zebrafish."Developmental Cell 32 (6):756-764.

Abraham, N. G., J. C. Nelson, T. Ahmed, G. Konwalinka, and R. D. Levere. 1989. "Erythropoietin controls heme metabolic enzymes in normal human bone marrow culture." Experimental Hematology 17 (8):908-13.

Abraham, N.G., and A. Kappas. 2008. "Pharmacological and Clinical Aspects of Heme Oxygenase."Pharmacological Reviews 60 (1):79-127.

Ahmad, Z., M. Salim, and M. Maines. 2002. "Human biliverdin reductase is a leucine zipper-like DNA-binding protein and functions in transcriptional activation of heme oxygenase-1 by oxidative stress."Journal of Biological Chemistry 277 (11):9226-9232.

Akashi, K., D. Traver, T. Miyamoto, and I. L. Weissman. 2000. "A clonogenic common myeloid progenitor that gives rise to all myeloid lineages."Nature 404 (6774):193-7.

Akerfelt, M., R. I. Morimoto, and L. Sistonen. 2010. "Heat shock factors: integrators of cell stress, development and lifespan."Nature Reviews: Molecular Cell Biology 11 (8):545-55.

Alam, J. 1994. "Multiple elements within the 5' distal enhancer of the mouse heme oxygenase-1 gene mediate induction by heavy metals."Journal of Biological Chemistry 269 (40):25049-56.

Alam, J. and Cook, J. 2007. "How many transcription factors does it take to turn on the heme oxygenase-1 gene?"American Journal of Respiratory Cell and Molecular Biology 36 (2):166-74.

Alam, J., S. Shibahara, and A. Smith. 1989. "Transcriptional activation of the heme oxygenase gene by heme and cadmium in mouse hepatoma cells."Journal of Biological Chemistry 264 (11):6371-5.

Alam, J., D., Stewart, C., Touchard, S., Boinapally, M. K. Choi, and J. L. Cook. 1999. "Nrf2, a Cap’n’Collar Transcription Factor, Regulates Induction of the Heme Oxygenase-1 Gene." Journal of Biological Chemistry 274 (37):26071-26078.

Alam, J., C., Wicks, D. Stewart, P. Gong, C. Touchard, S. Otterbein, A. M. K. Choi, M. E. Burow, and J. Tou. 2000. "Mechanism of Heme Oxygenase-1 Gene Activation by Cadmium in MCF-7 Mammary Epithelial Cells: Role of p38 Kinase and Nrf2 Transcription Factor."Journal of Biological Chemistry 275 (36):27694-27702.

Ali, S., H. G. J. van Mil, and M. K. Richardson. 2011. "Large-Scale Assessment of the Zebrafish Embryo as a Possible Predictive Model in Toxicity Testing."PLoS One 6 (6):e21076.

125

Aluru, N., M. J. Jenny, and M.E. Hahn. 2014. "Knockdown of a Zebrafish Aryl Hydrocarbon Receptor Repressor (AHRRa) Affects Expression of Genes Related to Photoreceptor Development and Hematopoiesis."Toxicological Sciences 139 (2):381-395.

Amatruda, J. F., J. L. Shepard, H. M. Stern, and L. I. Zon. 2002. "Zebrafish as a cancer model system."Cancer Cell 1 (3):229-31.

Amigo, J.D., G. E. Ackermann, J. J. Cope, M. Y., J. D. Cooney, D.M., N. B. Langer, E. Shafizadeh, G. C. Shaw, W. Horsely, N. S. Trede, A. J. Davidson, B. A. Barut, Y. Zhou, S. A. Wojiski, D. Traver, T. B. Moran, G. Kourkoulis, K. Hsu, J.P. Kanki, D. I. Shah, H. F. Lin, R. I. Handin, . B. Cantor, and B. H. Paw. 2009. "The role and regulation of friend of GATA-1 (FOG- 1) during blood development in the zebrafish."Blood 114 (21):4654-4663.

Amores, A., A. Force, Y. L. Yan, L. Joly, C. Amemiya, A. Fritz, R. K. Ho, J. Langeland, V. Prince, Y. L. Wang, M. Westerfield, M. Ekker, and J. H. Postlethwait. 1998. "Zebrafish hox clusters and vertebrate genome evolution."Science 282 (5394):1711-1714.

Amsterdam, A, R. M. Nissen, Z. Sun, E. C. Swindell, S. Farrington, and N. Hopkins. 2004. "Identification of 315 genes essential for early zebrafish development."Proceedings of the National Academy of Sciences 101 (35):12792-12797.

Andrews, G. K. 2001. "Cellular zinc sensors: MTF-1 regulation of gene expression."Biometals 14 (3-4):223-37.

Austin, C. C., and K. W. Jessing. 1994. "Green-blood pigmentation in lizards."Comparative Biochemistry and Physiology Part A: Physiology 109 (3):619-626.

Bai, X., J. Kim, Z. Yang, M. J. Jurynec, T. E. Akie, J. Lee, J. LeBlanc, A. Sessa, H. Jiang, A. DiBiase, Y. Zhou, D. J. Grunwald, S. Lin, A. B. Cantor, S. H. Orkin, and L. I. Zon. 2010. "TIF1gamma controls erythroid cell fate by regulating transcription elongation."Cell 142 (1):133-43.

Balesaria, S., B. Ramesh, H. McArdle, H. K. Bayele, and S. K. Srai. 2010. "Divalent metal- dependent regulation of hepcidin expression by MTF-1."FEBS Letters 584 (4):719-25.

Barañano, D., Ferris, C., and Snyder, S. 2001. "Atypical neural messengers."Trends in Neurosciences 24 (2):99-106.

Barañano, D., Rao, M., Ferris, C., and Snyder, S. 2002. "Biliverdin reductase: a major physiologic cytoprotectant."Proceedings of the National Academy of Sciences 99 (25):16093- 16098.

Baranano, D. E., and S. H. Snyder. 2001. "Colloquium Paper: Neural roles for heme oxygenase: Contrasts to nitric oxide synthase."Proceedings of the National Academy of Sciences 98 (20):10996-11002.

Barinaga, M. "Carbon monoxide: killer to brain messenger in one step."Science.

126

Bennett, C. M., J. P. Kanki, J. Rhodes, T. X. Liu, B. H. Paw, M. W. Kieran, D. M. Langenau, A. Delahaye-Brown, L. I. Zon, M. D. Fleming, and A. T. Look. 2001. "Myelopoiesis in the zebrafish, Danio rerio."Blood 98 (3):643-651.

Bertrand, J. Y., N. C. Chi, B. Santoso, S. Teng, D. Y. Stainier, and D. Traver. 2010. "Haematopoietic stem cells derive directly from aortic endothelium during development." Nature 464 (7285):108-11.

Bertrand, J. Y., A. D. Kim, S. Teng, and D. Traver. 2008. "CD41+ cmyb+ precursors colonize the zebrafish pronephros by a novel migration route to initiate adult hematopoiesis." Development 135 (10):1853-62.

Bertrand, J. Y., A. D. Kim, E. P. Violette, D. L. Stachura, J. L. Cisson, and D, Traver. 2007. "Definitive hematopoiesis initiates through a committed erythromyeloid progenitor in the zebrafish embryo."Development 134 (23):4147-4156.

Blechinger, S., Kusch, R., Haugo, K., Matz, C., Chivers, D., and Krone, P. 2007. "Brief embryonic cadmium exposure induces a stress response and cell death in the developing olfactory system followed by long-term olfactory deficits in juvenile zebrafish."Toxicology and Applied Pharmacology 224 (1):72-80.

Blechinger, S., Warren Jr, J., Kuwada, J. and Krone, P. 2002. "Developmental toxicology of cadmium in living embryos of a stable transgenic zebrafish line."Environmental Health Perspectives 110 (10):1041.

Boitano, A. E., J. Wang, R. Romeo, L. C. Bouchez, A. E. Parker, S.E. Sutton, J. R. Walker, C. A. Flaveny, G. H. Perdew, M.S. Denison, P. G. Schultz, and M. P. Cooke. 2010. "Aryl Hydrocarbon Receptor Antagonists Promote the Expansion of Human Hematopoietic Stem Cells."Science 329 (5997):1345-1348.

Brown, A. C., J. D. Lutton, H. A. Pearson, J. C. Nelson, R. D. Levere, and N. G. Abraham. 1988. "Heme metabolism and in vitro erythropoiesis in anemia associated with hypochromic microcytosis."American Journal of Hematology 27 (1):1-6.

Brownlie, A., C. Hersey, A. C. Oates, B. H. Paw, A. M. Falick, H. E. Witkowska, J. Flint, D. Higgs, J. Jessen, N. Bahary, H. Zhu, S. Lin, and L. Zon. 2003. "Characterization of embryonic globin genes of the zebrafish."Developmental Biology 255 (1):48-61.

Burns, C. E., T. DeBlasio, Y. Zhou, J. Zhang, L. Zon, and S. D. Nimer. 2002. "Isolation and characterization of runxa and runxb, zebrafish members of the runt family of transcriptional regulators."Experimental Hematology 30 (12):1381-9.

Butko, E., M. Distel, C. Pouget, B. Weijts, I. Kobayashi, K. Ng, C. Mosimann, F. E. Poulain, A. McPherson, C. W. Ni, D. L. Stachura, N. Del Cid, R. Espin-Palazon, N. D. Lawson, R. Dorsky, W. K. Clements, and D. Traver. 2015. "Gata2b is a restricted early regulator of hemogenic endothelium in the zebrafish embryo."Development 142 (6):1050-61.

127

Cantor, A. B., and S. H. Orkin. 2002. "Transcriptional regulation of erythropoiesis: an affair involving multiple partners."Oncogene 21 (21):3368-76.

Chen, X. D., and J. B. Turpen. 1995. "Intraembryonic origin of hepatic hematopoiesis in Xenopus laevis."Journal of Immunology 154 (6):2557-67.

Chikuba, K., Yubisui, T., Shirabe, K. and Takeshita, M. 1994. "Cloning and Nucleotide Sequence of a cDNA of the Human Erythrocyte NADPH-Flavin Reductase."Biochemical and Biophysical Research Communications 198 (3):1170-1176.

Chitramuthu, B. P., and H. P. Bennett. 2013. "High Resolution Whole Mount In Situ Hybridization within Zebrafish Embryos to Study Gene Expression and Function."Journal of Visualized Experiments: JoVE (80):e50644.

Choi, A. M., and J. Alam. 1996. "Heme oxygenase-1: function, regulation, and implication of a novel stress-inducible protein in oxidant-induced lung injury."American Journal of Respiratory Cell and Molecular Biology 15 (1):9-19.

Ciau-Uitz, A., R. Monteiro, A. Kirmizitas, and R. Patient. 2014. "Developmental hematopoiesis: ontogeny, genetic programming and conservation."Experimental Hematology 42 (8):669-83.

Clements, W. K., and D. Traver. 2013. "Signalling pathways that control vertebrate haematopoietic stem cell specification."Nature Reviews Immunology 13 (5):336-348.

Colleran, E., and K. P. M. Heirwegh. 1979. "Comparative aspects of bile pigment formation and excretion."Comparative Biochemistry and Physiology Part B: Comparative Biochemistry 64 (2):133-139.

Collins, J. E., S. White, S. M. Searle, and D. L. Stemple. 2012. "Incorporating RNA-seq data into the zebrafish Ensembl genebuild."Genome Research 22 (10):2067-78.

Cornelius, C. E. 1991. "Bile Pigments in Fishes: A Review."Vetinary Clinical Pathology 20 (4):106-115.

Craven, S. E., D. French, W. Ye, F. de Sauvage, and A. Rosenthal. 2005. "Loss of Hspa9b in zebrafish recapitulates the ineffective hematopoiesis of the myelodysplastic syndrome."Blood 105 (9):3528-3534.

Cui, L., Y. Yoshioka, O. Suyari, Y. Kohno, X. Zhang, Y. Adachi, S. Ikehara, T. Yoshida, M. Yamaguchi, and S. Taketani. 2008. "Relevant expression of Drosophila heme oxygenase is necessary for the normal development of insect tissues."Biochemical and Biophysical Research Communications 377 (4):1156-1161.

Culotta, E., and D. E. Koshland, Jr. 1992. "NO news is good news."Science 258 (5090):1862-5.

Cunningham, O., M. G. Gore, and T. J. Mantle. 2000. "Initial-rate kinetics of the flavin reductase reaction catalysed by human biliverdin-IXbeta reductase (BVR-B)."Biochemical Journal 345 Pt 2:393-9.

128

Cvejic, A., C. Hall, M. Bak-Maier, M. V. Flores, P. Crosier, M. J. Redd, and P. Martin. 2008. "Analysis of WASp function during the wound inflammatory response – live-imaging studies in zebrafish larvae."Journal of Cell Science 121 (19):3196-3206.

Davidson, A. J., P. Ernst, Y. Wang, M. P. Dekens, P. D. Kingsley, J. Palis, S. J. Korsmeyer, G. Q. Daley, and L. I. Zon. 2003. "cdx4 mutants fail to specify blood progenitors and can be rescued by multiple hox genes."Nature 425 (6955):300-6. de Thonel, A., J. Vandekerckhove, D. Lanneau, S. Selvakumar, G. Courtois, A. Hazoume, M. Brunet, S. Maurel, A. Hammann, J. A. Ribeil, Y. Zermati, A. S. Gabet, J. Boyes, E. Solary, O. Hermine, and C. Garrido. 2010. "HSP27 controls GATA-1 protein level during erythroid cell differentiation."Blood 116 (1):85-96.

Denschlag, D., R. Marculescu, G. Unfried, L. A. Hefler, M. Exner, A. Hashemi, E. K. Riener, C. Keck, C. B. Tempfer, and O. Wagner. 2004. "The size of a microsatellite polymorphism of the haem oxygenase 1 gene is associated with idiopathic recurrent miscarriage."Molecular Human Reproduction 10 (3):211-4.

Detrich, H W, M W Kieran, F Y Chan, L M Barone, K Yee, J A Rundstadler, S Pratt, D Ransom, and L I Zon. 1995. "Intraembryonic hematopoietic cell migration during vertebrate development."Proceedings of the National Academy of Sciences 92 (23):10713-10717.

Detrich, H., Kieran, M., Chan, F., Barone, L., Yee, K., Rundstadler, J., Pratt, S., Ransom, D. and Zon, L. 1995. "Intraembryonic hematopoietic cell migration during vertebrate development." Proceedings of the National Academy of Sciences 92 (23):10713-10717.

Dinkova-Kostova, A. and Talalay, P. 2010. "NAD(P)H:quinone acceptor oxidoreductase 1 (NQO1), a multifunctional antioxidant enzyme and exceptionally versatile cytoprotector." Archives of Biochemistry and Biophysics 501 (1):116-23.

Dohn, T. E., and J. S. Waxman. 2012. "Distinct phases of Wnt/beta-catenin signaling direct cardiomyocyte formation in zebrafish."Developmental Biology 361 (2):364-76.

Donovan, A., A. Brownlie, Y. Zhou, J. Shepard, S. J. Pratt, J. Moynihan, B. H. Paw, A. Drejer, B. Barut, A. Zapata, T. C. Law, C. Brugnara, S. E. Lux, G. S. Pinkus, J. L. Pinkus, P. D. Kingsley, J. Palis, M. D. Fleming, N. C. Andrews, and L. I. Zon. 2000. "Positional cloning of zebrafish ferroportin1 identifies a conserved vertebrate iron exporter."Nature 403 (6771):776-81.

Dore, S., Takahashi, M.,Ferris, C., Zakhary, R., Hester, L., Guastella, D. and Snyder, S. . 1999. "Bilirubin, formed by activation of heme oxygenase-2, protects neurons against oxidative stress injury."Proceedings of the National Academy of Sciences 96 (5):2445-2450.

Ewing, J. F., and M.D. Maines. 1992. "In situ hybridization and immunohistochemical localization of heme oxygenase-2 mRNA and protein in normal rat brain: Differential distribution of isozyme 1 and 2."Molecular and Cellular Neuroscience 3 (6):559-570.

Ewing, J. F., and M. D. Maines. 1997. "Histochemical localization of heme oxygenase-2 protein and mRNA expression in rat brain."Brain Research Protocols 1 (2):165-174.

129

Falchuk, K. H., J. M. Contin, T. S. Dziedzic, Z. Feng, T. C. French, G. J. Heffron, and M. Montorzi. 2002. "A role for biliverdin IXα in dorsal axis development of Xenopus laevis embryos."Proceedings of the National Academy of Sciences 99 (1):251.

Ferreira, R., K. Ohneda, M.Yamamoto, and S. Philipsen. 2005. "GATA1 Function, a Paradigm for Transcription Factors in Hematopoiesis."Molecular and Cellular Biology 25 (4):1215-1227.

Force, A., Lynch, M., Pickett, F., Amores, A., Yan, Y. and Postlethwait, J. 1999. "Preservation of duplicate genes by complementary, degenerative mutations."Genetics 151 (4):1531-1545.

Fraenkel, P. G., Y. Gibert, J. L. Holzheimer, V. J. Lattanzi, S. F. Burnett, K. A. Dooley, R. A. Wingert, and L. I. Zon. 2009. "Transferrin-a modulates hepcidin expression in zebrafish embryos."Blood 113 (12):2843-50.

Fraser, S. T., R. G. Midwinter, L. A. Coupland, S. Kong, B. S. Berger, J. H. Yeo, O. C. Andrade, D. Cromer, C. Suarna, M. Lam, G. J. Maghzal, B. H. Chong, C. R. Parish, and R. Stocker. 2015. "Heme oxygenase-1 deficiency alters erythroblastic island formation, steady-state erythropoiesis and red blood cell lifespan in mice."Haematologica 100 (5):601-10.

Fujiwara, Y., C. P. Browne, K. Cunniff, S. C. Goff, and S. H. Orkin. 1996. "Arrested development of embryonic red cell precursors in mouse embryos lacking transcription factor GATA-1."Proceedings of the National Academy of Sciences 93 (22):12355-8.

Galloway, J., and Zon, L. 2003. "3 Ontogeny of hematopoiesis: Examining the emergence of hematopoietic cells in the vertebrate embryo."Current Topics in Developmental Biology Volume 53:139-158.

Galloway, J., Wingert, R., Thisse, C. Thisse, B. and Zon, L. 2005. "Loss of gata1 but not gata2 converts erythropoiesis to myelopoiesis in zebrafish embryos."Developmental Cell 8 (1):109- 116.

Ganis, J., Hsia, N., Trompouki, E., de Jong, J., DiBiase, A., Lambert, J., Jia, Z., Sabo, P., Weaver, M., Sandstrom, R., Stamatoyannopoulos, J., Zhou, Y., and Zon, L. 2012. "Zebrafish globin switching occurs in two developmental stages and is controlled by the LCR." Developmental Biology 366 (2):185-194.

Gao, Dian, Zhen’e Xu, Xiaoyan Zhang, Chunchao Zhu, Yannan Wang, and Weiping Min. 2013. "Cadmium triggers kidney cell apoptosis of purse red common carp (Cyprinus carpio) without caspase-8 activation."Developmental & Comparative Immunology 41 (4):728-737.

Gao, Y., Y. Yan, and T. Huang. 2015. "Human agerelated cataracts: epigenetic suppression of the nuclear factor erythroid 2 related factor 2 mediated antioxidant system."Molecular Medecine Reports 11 (2):1442-7.

Garcia-Santos, D., M. Schranzhofer, M. Horvathova, M. M. Jaberi, J. A. Bogo Chies, A. Sheftel, and P. Ponka. 2014. "Heme oxygenase 1 is expressed in murine erythroid cells where it controls the level of regulatory heme."Blood.

130

Ghirlando, R., and C. D. Trainor. 2003. "Determinants of GATA-1 binding to DNA: the role of non-finger residues."Journal of Biological Chemistry 278 (46):45620-8.

Gibbs, P. E., T. Miralem, and M. D. Maines. 2010. "Characterization of the human biliverdin reductase gene structure and regulatory elements: promoter activity is enhanced by hypoxia and suppressed by TNF-alpha-activated NF-kappaB."The FASEB Journal 24 (9):3239-3254.

Gibbs, P., C. Tudor, and, and M. Maines. 2012. "Biliverdin reductase: more than a namesake - the reductase, its Peptide fragments, and biliverdin regulate activity of the three classes of protein kinase C."Frontiers in Pharmacology 3:31.

Goishi, K., A. Shimizu, G. Najarro, S. Watanabe, R. Rogers, L. I. Zon, and M. Klagsbrun. 2006. "αA-crystallin expression prevents γ-crystallin insolubility and cataract formation in the zebrafish cloche mutant lens."Development 133 (13):2585-2593.

Graf, T. 2002. "Differentiation plasticity of hematopoietic cells."Blood 99 (9):3089-101.

Green, C. J., P. Lichtlen, N. T. Huynh, M. Yanovsky, K. R. Laderoute, W. Schaffner, and B. J. Murphy. 2001. "Placenta growth factor gene expression is induced by hypoxia in fibroblasts: a central role for metal transcription factor-1."Cancer Research 61 (6):2696-703.

Greenberg, David A. 2002. "The jaundice of the cell."Proceedings of the National Academy of Sciences 99 (25):15837-15839.

Gunes, C., Heuchel, R., Georgiev, O., Muller, K., Lichtlen, P., Bluthmann, H., Marino, S., Aguzzi, A., and Schaffner, W. 1998. "Embryonic lethality and liver degeneration in mice lacking the metal-responsive transcriptional activator MTF-1."The EMBO Journal 17 (10):2846-2854.

Günther, V., U. Lindert, and W. Schaffner. 2012. "The taste of heavy metals: Gene regulation by MTF-1." Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 1823 (9):1416-1425.

Hahn, M. E., L. L. Allan, and D. H. Sherr. 2009. "Regulation of constitutive and inducible AHR signaling: complex interactions involving the AHR repressor."Biochemical Pharmacology 77 (4):485-97.

Hahn, M. E., A. R. Timme-Laragy, S. I. Karchner, and J. J. Stegeman. "Nrf2 and Nrf2-related proteins in development and developmental toxicity:Insights from studies in zebrafish (danio rerio)."Free Radical Biology and Medicine.

Hall, Barry G. 2013. "Building Phylogenetic Trees from Molecular Data with MEGA." Molecular Biology and Evolution 30 (5):1229-1235.

Hanaoka, R., I. B. Dawid, and A. Kawahara. 2007. "Cloning and expression of zebrafish genes encoding the heme synthesis enzymes uroporphyrinogen III synthase (UROS) and protoporphyrinogen oxidase (PPO)."DNA Sequence 18 (1):54-60.

131

Hanaoka, R., S. Katayama, I. B. Dawid, and A. Kawahara. 2006. "Characterization of the heme synthesis enzyme coproporphyrinogen oxidase (CPO) in zebrafish erythrogenesis."Genes to Cells 11 (3):293-303.

Helker, C. S., A. Schuermann, T. Karpanen, D. Zeuschner, H. G. Belting, M. Affolter, S. Schulte-Merker, and W. Herzog. 2013. "The zebrafish common cardinal veins develop by a novel mechanism: lumen ensheathment."Development 140 (13):2776-86.

Herbomel, P., Thisse, B., and Thisse, C. 1999. "Ontogeny and behaviour of early macrophages in the zebrafish embryo."Development 126 (17):3735-3745.

Hill, A., Teraoka, H., Heideman, W., and Peterson, R. 2005. "Zebrafish as a model vertebrate for investigating chemical toxicity."Toxicological Sciences 86 (1):6.

Hogstrand C, Zheng D, Feeney G, Cunningham P, and Kille P. 2008. "Zinc-controlled gene expression by metal-regulatory transcription factor 1 (MTF1) in a model vertebrate, the zebrafish."Biochemical Society Transactions 36:1252-1257.

Holmstrom, K. M., and T. Finkel. 2014. "Cellular mechanisms and physiological consequences of redox-dependent signalling."Nature Reviews. Molecular Cell Biology 15 (6):411-21.

Howe, K., Matthew D. , Clark, C. F. , Torroja, J., Torrance, C., Berthelot, M., Muffato, John E. Collins, S., Humphray, K., McLaren, L., Matthews, S., McLaren, I., Sealy, M., Caccamo, C., Churcher, C., Scott, J., Barrett, R., Koch, G., Rauch, S., White, W., Chow, B., Kilian, L., Quintais, J., Guerra-Assuncao, Y., Zhou, Y., Gu, J., Yen, Jan-Hinnerk., Vogel, T., Eyre, S., Redmond, R., Banerjee, J., Chi, B., Fu, E., Langley, S., Maguire, G., Laird, D., Lloyd, E., Kenyon, S., Donaldson, H., Sehra, J., Almeida-King, J., Loveland, S., Trevanion, M., Jones, M., Quail, D., Willey, A., Hunt, J., Burton, S., Sims, K., McLay, B., Plumb, J., Davis, C., Clee, K., Oliver, R., Clark, C., Riddle, D., Eliott, G., Threadgold, G., Harden, D., Ware, B., Mortimer, G., Kerry, P., Heath, B., Phillimore, A., Tracey, N., Corby, M., Dunn, C., Johnson, J., Wood, S., Clark, S., Pelan, G., Griffiths, M., Smith, R., Glithero, P., Howden, N., Barker, C., Stevens, J., Harley, K., Holt, G., Panagiotidis, J., Lovell, H., Beasley, C., Henderson, D., Gordon, K., Auger, D., Wright, J., Collins, C., Raisen, L., Dyer, K., Leung, L., Robertson, K., Ambridge, D., Leongamornlert, S., McGuire, R., Gilderthorp, C., Griffiths, D., Manthravadi, S., Nichol, G., Barker, S., Whitehead, M., Kay, J., Brown, C., Murnane, E., Gray, M., Humphries, N., Sycamore, D., Barker, D., Saunders, J., Wallis, A., Babbage, S., Hammond, Maryam ., Mashreghi-Mohammadi, L., Barr, S., Martin, P., Wray, A., Ellington, N., Matthews, M., Ellwood, R., Woodmansey, G., Clark, J., Cooper, A., Tromans, D., Grafham, C., Skuce, R., Pandian, R., Andrews, E., Harrison, A., Kimberley, J., Garnett, N., Fosker, R., Hall, P., Garner, D., Kelly, C., Bird, S., Palmer, I., Gehring, A., Berger, C., Dooley, Zubeyde., Ersan, Urun, C., Eser, H., Geiger, M., Geisler, L., Karotki, A., Kirn, J., Konantz, M., Konantz, M., Oberlander, S., Rudolph-Geiger, M., Teucke, K., Osoegawa, B., Zhu, A., Rapp, Sara, Widaa, C., Langford, Fengtang, Yang, N., Carter, J., Harrow, Z., Ning, J., Herrero, Steve M., Searle, A., Enright, R., Geisler, R., Plasterk, C., Lee, M. Westerfield, Pieter J. de Jong, L. I. Zon, J. H. Postlethwait, C. Nusslein-Volhard, T. Hubbard, H. Crollius, J. Rogers, and D. L. Stemple. 2013. " The zebrafish reference genome sequence and its relationship to the human genome."Nature 496 (25):498-503.

132

Hromas, R., A. Orazi, R. S. Neiman, R. Maki, C. Van Beveran, J. Moore, and M. Klemsz. 1993. "Hematopoietic lineage- and stage-restricted expression of the ETS oncogene family member PU.1." Blood 82 (10):2998-3004.

Hsia, N., and L. Zon. 2005. "Transcriptional regulation of hematopoietic stem cell development in zebrafish."Experimental Hematology 33 (9):1007-1014.

Hsu, K., D. Traver, J. L. Kutok, A. Hagen, T. Liu, B. Paw, J. Rhodes, J. Berman, L. Zon, J. Kanki, and A. Look. 2004. "The pu.1 promoter drives myeloid gene expression in zebrafish." Blood 104 (5):1291-1297.

Huber, W. J., 3rd, C. C. Marohnic, M. Peters, J. Alam, J. R. Reed, B. S. Masters, and W. L. Backes. 2009. "Measurement of membrane-bound human heme oxygenase-1 activity using a chemically defined assay system."Drug Metabolism and Disposition:The Biological Fate of Chemicals 37 (4):857-64.

Jego, G., D. Lanneau, A. De Thonel, K. Berthenet, A. Hazoume, N. Droin, A. Hamman, F. Girodon, P. S. Bellaye, G. Wettstein, A. Jacquel, L. Duplomb, A. Le Mouel, C. Papanayotou, E. Christians, P. Bonniaud, V. Lallemand-Mezger, E. Solary, and C. Garrido. 2014. "Dual regulation of SPI1/PU.1 transcription factor by heat shock factor 1 (HSF1) during macrophage differentiation of monocytes."Leukemia 28 (8):1676-86.

Jennings, G. T., and P. M. Stevenson. 1991. "A study of the control of NADP -dependent isocitrate dehydrogenase activity during gonadotropin-induced development of the rat ovary." European Journal of Biochemistry 198 (3):621-625.

Jenny, M. J., S. I. Karchner, D. G. Franks, B. R. Woodin, J. J. Stegeman, and M. E. Hahn. 2009. "Distinct Roles of Two Zebrafish AHR Repressors (AHRRa and AHRRb) in Embryonic Development and Regulating the Response to 2,3,7,8-Tetrachlorodibenzo-p-dioxin." Toxicological Sciences 110 (2):426-441.

Jin, H, R. Sood, J. Xu, F. Zhen, M. A. English, P. Liu, and Z.Wen. 2009. "Definitive hematopoietic stem/progenitor cells manifest distinct differentiation output in the zebrafish VDA and PBI."Development 136 (4):647-654.

Jin, H, Xu, J., and Z. Wen. 2007. "Migratory path of definitive hematopoietic stem/progenitor cells during zebrafish development."Blood 109 (12):5208-5214.

Jo, S., Mi-Kyung S., H. Koh, S. Lee, I. Song, Y. Kim, Y. Lee, K. Jeong, W. Kim, J. Park, B. Song, and T. Huhe. 2001. "Control of Mitochondrial Redox Balance and Cellular Defense against Oxidative Damage by Mitochondrial NADP+-dependent Isocitrate Dehydrogenase." Journal of Biological Chemistry 276 (19):16168-16176.

Johnson, M. D., N. Kenney, A. Stoica, L. Hilakivi-Clarke, B. Singh, G. Chepko, R. Clarke, P. F. Sholler, A. A. Lirio, C. Foss, R. Reiter, B. Trock, S. Paik, and M. B. Martin. 2003. "Cadmium mimics the in vivo effects of estrogen in the uterus and mammary gland." Nature Medecine 9 (8):1081-4.

133

Jopling, C., E. Sleep, M. Raya, M. Marti, A. Raya, and J. C. Izpisua Belmonte. 2010. "Zebrafish heart regeneration occurs by cardiomyocyte dedifferentiation and proliferation." Nature 464 (7288):606-9.

Kalariya, N. M., B. Nair, D. K. Kalariya, N. K. Wills, and F. J. van Kuijk. 2010. "Cadmium- induced induction of cell death in human lens epithelial cells: implications to smoking associated cataractogenesis." Toxicology Letters 198 (1):56-62.

Kalev-Zylinska, M.., J. Horsfield, M. Flores, J.H. Postlethwait, M. Vitas, A.Baas, P.. Crosier, and K. Crosier. 2002. "Runx1 is required for zebrafish blood and vessel development and expression of a human RUNX1-CBF2T1 transgene advances a model for studies of leukemogenesis." Development 129 (8):2015-2030.

Kapitulnik, J. and Maines, M. 2008. "Pleiotropic functions of biliverdin reductase: cellular signaling and generation of cytoprotective and cytotoxic bilirubin."Trends in Pharmacological Sciences 30 (3):129-137.

Karchner, S. I., D. G. Franks, and M. E. Hahn. 2005. "AHR1B, a new functional aryl hydrocarbon receptor in zebrafish: tandem arrangement of ahr1b and ahr2 genes."Biochemical Journal 392 (Pt 1):153-61.

Kawahara, G., M. Gasperini, J. Myers, J. Widrick, A. Eran, P. Serafini, M. Alexander, M. Pletcher, C. Morris, and L. Kunkel. 2013. "Dystrophic muscle improvement in zebrafish via increased heme oxygenase signaling."Human Molecular Genetics.

Kawakami, K. 2007. Tol2: a versatile gene transfer vector in vertebrates. Genome Biol. 8(Suppl 1), S7.

Kawakami, K., A., Shima, and N.,Kawakami. 2000. "Identification of a functional transposase of the Tol2 element, an Ac-like element from the Japanese medaka fish, and its transposition in the zebrafish germ lineage."Proceedings of the National Academy of Sciences 97 (21):11403-11408.

Keneddy, G. Y., and H. G. Vevers. 1973. "Eggshell pigments of the araucano fowl." Comparative Biochemistry and Physiology Part B: Comparative Biochemistry 44 (1):11-25.

Khan, A., and J.G. Quigley. 2011. "Control of intracellular heme levels: Heme transporters and heme oxygenases."Biochimica et Biophysica Acta (BBA) - Molecular Cell Research 1813 (5):668-682.

Kimmel, C., Ballard, W., Kimmel, S., Ullmann, B., and Schilling, T. 1995. "Stages of embryonic development of the zebrafish."American Journal of Anatomy 203 (3):253-310.

Kirsch, M. and De Groot, H. 2001. "NAD(P)H, a directly operating antioxidant?"The FASEB Journal 15 (9):1569-1574.

Kissa, K., and P. Herbomel. 2010. "Blood stem cells emerge from aortic endothelium by a novel type of cell transition."Nature 464 (7285):112-5.

134

Kissa, K., E. Murayama, A. Zapata, A. Cortes, E. Perret, C. Machu, and P. Herbomel. 2008. "Live imaging of emerging hematopoietic stem cells and early thymus colonization."Blood 111 (3):1147-56.

Klemsz, M. J., S. R. McKercher, A. Celada, C. Van Beveren, and R. A. Maki. 1990. "The macrophage and B cell-specific transcription factor PU.1 is related to the ets oncogene."Cell 61 (1):113-24.

Ko, L. J., and J. D. Engel. 1993. "DNA-binding specificities of the GATA transcription factor family."Molecular and Cellular Biology 13 (7):4011-22.

Kobayashi, M., Itoh, K., Suzuki,T.,Osanai, H.,Nishikawa, K.,, and Y. Katoh, Takagi, Y., and Yamamoto, M. 2002. "Identification of the interactive interface and phylogenic conservation of the Nrf2-Keap1 system."Genes to Cells 7:807-820.

Kovtunovych, G., M. C. Ghosh, W. Ollivierre, R. P. Weitzel, M. A. Eckhaus, J. F. Tisdale, A. Yachie, and T. A. Rouault. 2014. "Wild-type macrophages reverse disease in heme oxygenase 1- deficient mice."Blood 124 (9):1522-30.

Kutty, R., and, and M. Maines. 1981. "Purification and characterization of biliverdin reductase from rat liver."Journal of Biological Chemistry 256 (8):3956-3962.

Kwan, K. Fujimoto, E., Grabher, C., Mangum, B., Hardy, M., Campbell, D., Parant, J., Yost, H., Kanki, J. and Chien, C. . 2007. "The Tol2kit: a multisite gateway-based construction kit for Tol2 transposon transgenesis constructs."Developmental Dynamics 236 (11):3088-99.

Lam, E. Y., J. Y. Chau, M. L. Kalev-Zylinska, T. M. Fountaine, R. S. Mead, C. J. Hall, P. S. Crosier, K. E. Crosier, and M. V. Flores. 2009. "Zebrafish runx1 promoter-EGFP transgenics mark discrete sites of definitive blood progenitors."Blood 113 (6):1241-9.

Lam, E., C. Hall, P. Crosier, K. Crosier, and M. Flores. 2010. "Live imaging of Runx1 expression in the dorsal aorta tracks the emergence of blood progenitors from endothelial cells." Blood 116 (6):909-914.

Larsen, R., Z. Gouveia, M. P. Soares, and R. Gozzelino. 2012. "Heme cytotoxicity and the pathogenesis of immune-mediated inflammatory diseases."Frontiers in Pharmacology 3:77.

Lee, P. J., B. H. Jiang, B. Y. Chin, N. V. Iyer, J. Alam, G. L. Semenza, and A. M. Choi. 1997. "Hypoxia-inducible factor-1 mediates transcriptional activation of the heme oxygenase-1 gene in response to hypoxia."Journal of Biological Chemistry 272 (9):5375-81.

Lee, P., Jiang, B., Chin, B., Iyer, N., Alam, J., Semenza, G., and Choi, A. 1997. "Hypoxia- inducible Factor-1 Mediates Transcriptional Activation of the Heme Oxygenase-1 Gene in Response to Hypoxia."Journal of Biological Chemistry 272 (9): 5375–5381.

L., Fang. 1987. "Study of the heme catabolism of fish."Comparative Biochemistry and Physiology Part B: Comparative Biochemistry 88 (2):667-673.

135

Lerner-Marmarosh, N. Miralem, T. Gibbs, P. and Maines, M. 2008. "Human biliverdin reductase is an ERK activator; hBVR is an ERK nuclear transporter and is required for MAPK signaling." Proceedings of the National Academy of Sciences 105 (19):6870-5.

Lerner-Marmarosh, N., Shen, J., Torno, M., Kravets, A., Hu, Z. and Maines, M. 2005. "Human biliverdin reductase: a member of the insulin receptor substrate family with serine/threonine/tyrosine kinase activity."Proceedings of the National Academy of Sciences 102 (20):7109.

Li, L., and J.E. Dowling. 1997. "A dominant form of inherited retinal degeneration caused by a non-photoreceptor cell-specific mutation."Proceedings of the National Academy of Sciences 94 (21):11645-11650.

Liu, N., Wang, X., McCoubrey, W. and Maines, M. 2000. "Developmentally regulated expression of two transcripts for heme oxygenase-2 with a first exon unique to rat testis: control by corticosterone of the oxygenase protein expression."Gene 241 (1):175-183.

Liu, R., Shi, M., Giulivi,C. and Forman, H. 1998. "Quinones increase g-glutamyl transpeptidase expression bymultiple mechanisms in rat lung epithelial cells."American Journal of Physiology 274:L330-L336.

Lusska, A, E Shen, and J P Whitlock. 1993. "Protein-DNA interactions at a dioxin-responsive enhancer. Analysis of six bona fide DNA-binding sites for the liganded Ah receptor."Journal of Biological Chemistry 268 (9):6575-6580.

Lyons, S.., N. Lawson, L. Lei, P. Bennett, B.Weinstein, and P. Liu. 2002. "A nonsense mutation in zebrafish gata1 causes the bloodless phenotype in vlad tepes."Proceedings of the National Academy of Sciences 99 (8):5454-5459.

Maghzal, G. J., M. C. Leck, E. Collinson, C. Li, and R. Stocker. 2009. "Limited role for the bilirubin-biliverdin redox amplification cycle in the cellular antioxidant protection by biliverdin reductase."Journal of Biological Chemistry 284 (43):29251-9.

Maier, T., M. Guell, and L. Serrano. 2009. "Correlation of mRNA and protein in complex biological samples."FEBS Lett 583 (24):3966-73.

Maines, M. and Gibbs, P. 2005. "30 some Years of Heme Oxygenase: from a "Molecular Wrecking Ball" to a "Mesmerizing" Trigger of Cellular Events."Biochemical and Biophysical Research Communications 338 (1):568-577.

Maines, M. D. 1993. "Carbon Monoxide: An Emerging Regulator of cGMP in the Brain." Molecular and Cellular Neurosciences 4 (5):389-97.

Maines, M. D. 1997. "The heme oxygenase system: a regulator of second messenger gases." Annual Review of Pharmacology and Toxicology 37:517-54.

Maines, M. D. 2007. "Biliverdin reductase: PKC interaction at the cross-talk of MAPK and PI3K signaling pathways."Antioxidants & Redox Signaling 9 (12):2187-2195.

136

Maines, M. D., J. F. Ewing, T. J. Huang, and N. Panahian. 2001. "Nuclear localization of biliverdin reductase in the rat kidney: response to nephrotoxins that induce heme oxygenase-1." Journal of Pharmacology and Experimental Therapeutics 296 (3):1091.

Maines, M. D., and G. M. Trakshel. 1993. "Purification and characterization of human biliverdin reductase."Archives of Biochemistry and Biophysics 300 (1):320-6.

Maines, M., Trakshel, G., and Kutty, R. 1986. "Characterization of two constitutive forms of rat liver microsomal heme oxygenase. Only one molecular species of the enzyme is inducible." Journal of Biological Chemistry 261 (1):411-419.

Malec, V., O. R. Gottschald, S. Li, F. Rose, W. Seeger, and J. Hanze. 2010. "HIF-1 alpha signaling is augmented during intermittent hypoxia by induction of the Nrf2 pathway in NOX1- expressing adenocarcinoma A549 cells."Free Radical Biology and Medicine 48 (12):1626-35.

Malicki, J., Z. Pujic, C. Thisse, B.Thisse, and X. Wei. 2002. "Forward and reverse genetic approaches to the analysis of eye development in zebrafish."Vision Research 42 (4):527-533.

Martin, D. I., and S. H. Orkin. 1990. "Transcriptional activation and DNA binding by the erythroid factor GF-1/NF-E1/Eryf 1."Genes & Development 4 (11):1886-98.

McCoubrey, W. K., Jr., M. A. Cooklis, and M. D. Maines. 1995. "The structure, organization and differential expression of the rat gene encoding biliverdin reductase."Gene 160 (2):235-240.

McDevitt, M. A., R. A. Shivdasani, Y. Fujiwara, H. Yang, and S. H. Orkin. 1997. "A "knockdown" mutation created by cis-element gene targeting reveals the dependence of erythroid cell maturation on the level of transcription factor GATA-1."Proceedings of the National Academy of Sciences 94 (13):6781-5.

McDonagh, A. 1979. Bile pigments:biladienes and 5,15-bilatrienes. Edited by ed Dolphin D, The Porphyrins. New York, NY: Academic Press.

McDonagh, A. 2001. "Turning Green to Gold "Nature Structural Biology 8 (3):198-200.

McDonagh, A. 2004. "Bilirubin the beneficent."Pediatrics 114 (6):1741-2; author reply 1742-3.

McDonagh, A. F. 2010. "Controversies in bilirubin biochemistry and their clinical relevance." Seminars in Fetal & Neonatal Medicine 15 (3):141-7.

McDonagh, A. F., L. A. Palma, and R. Schmid. 1981. "Reduction of biliverdin and placental transfer of bilirubin and biliverdin in the pregnant guinea pig."Biochemical Journal 194 (1):273- 282.

McDonagh, AF, and D Dolphin. 1979. The Porphyrins, Vol. VIAcademic Press, New York.

McDonagh, A., and L. Palma. 1982. "Heme catabolism in fish. Bile pigments in gallbladder bile of the electric torpedo, Torpedo californicus."Comparative Biochemistry and Physiology Part B: Comparative Biochemistry 73 (3):501-507.

137

Medvinsky, A., S. Rybtsov, and S.Taoudi. 2011. "Embryonic origin of the adult hematopoietic system: advances and questions."Development 138 (6):1017-1031.

Mendelsohn, B., and Gitlin, J. 2008a. "Coordination of development and metabolism in the pre- midblastula transition zebrafish embryo."Developmental Dynamics 237 (7):1789-1798.

Mendelsohn, B. Kassebaum, B. and Gitlin, J. . 2008b. "The zebrafish embryo as a dynamic model of anoxia tolerance."Developmental Dynamics 237 (7):1780-1788.

Merchant, A., A. Singh, W. Matsui, and S. Biswal. 2011. "The redox-sensitive transcription factor Nrf2 regulates murine hematopoietic stem cell survival independently of ROS levels." Blood 118 (25):6572-6579.

Merika, M., and S. H. Orkin. 1993. "DNA-binding specificity of GATA family transcription factors." Molecular and Cellular Biology 13 (7):3999-4010.

Meyer, A. and Van de Peer, Y. 2005. "From 2R to 3R: evidence for a fish-specific genome duplication (FSGD)."Bioessays 27 (9):937-45.

Molkentin, J. D. 2000. "The zinc finger-containing transcription factors GATA-4, -5, and -6. Ubiquitously expressed regulators of tissue-specific gene expression."Journal of Biological Chemistry 275 (50):38949-52.

Monteiro, R., C. Pouget, and R. Patient. 2011. "The gata1/pu.1 lineage fate paradigm varies between blood populations and is modulated by tif1γ."The EMBO Journal 30 (6):1093-1103.

Montorzi, M., T. S. Dziedzic, and K. H. Falchuk. 2002. "Biliverdin during Xenopus laevis Oogenesis and Early Embryogenesis."Biochemistry 41 (31):10115-10122.

Moon, M., McDevitt, E., Zhu, J., Stanley, B., Krzeminski, J,. Amin, S., Aliaga, C., Miller, T., and Isom, H. 2012. "Elevated Hepatic Iron Activates NF-E2-Related Factor 2-Regulated Pathway in a Dietary Iron Overload Mouse Model."Toxicological Sciences 129 (1):74-85.

Motohashi, H., F. Katsuoka, J. Engel, and M.Yamamoto. 2004. "Small Maf proteins serve as transcriptional cofactors for keratinocyte differentiation in the Keap1–Nrf2 regulatory pathway." Proceedings of the National Academy of Sciences 101 (17):6379-6384.

Mukaigasa, K., Nguyen, L., Li, L., Nakajima, H. Yamamoto, M. and Kobayashi, M. 2012. "Genetic Evidence of an Evolutionarily Conserved Role for Nrf2 in the Protection against Oxidative Stress."Molecular and Cellular Biology.

Murayama, E., K. Kissa, A. Zapata, E. Mordelet, V. Briolat, H. F. Lin, R. I. Handin, and P. Herbomel. 2006. "Tracing hematopoietic precursor migration to successive hematopoietic organs during zebrafish development."Immunity 25 (6):963-75.

Murphey, R. D., H. M. Stern, C. T. Straub, and L. I. Zon. 2006. "A chemical genetic screen for cell cycle inhibitors in zebrafish embryos."Chemical Biology & Drug Design 68 (4):213-9.

138

Murphy, B.., T. Kimura, B.G. Sato, Y. Shi, and G. Andrews. 2008. "Metallothionein Induction by Hypoxia Involves Cooperative Interactions between Metal-Responsive Transcription Factor-1 and Hypoxia-Inducible Transcription Factor-1α."Molecular Cancer Research 6 (3):483-490.

Murphy, B.., B. Sato, T. Dalton, and K. Laderoute. 2005. "The metal-responsive transcription factor-1 contributes to HIF-1 activation during hypoxic stress."Biochemical and Biophysical Research Communications 337 (3):860-867.

Nakajima, H., Nakajima-Takagi, Y., Tsujita, T., Akiyama, S., Wakasa, T.,, and K. Mukaigasa, Kaneko, H.,Tamaru, Y.,Yamamoto,M., and Kobayashi, M. 2011. "Tissue-Restricted Expression of Nrf2 and Its Target Genes in Zebrafish with Gene-Specific Variations in the Induction Profiles."PLoS ONE 6 (10):1-8.

Nasevicius, A., and S. C. Ekker. 2000. "Effective targeted gene ‘knockdown’in zebrafish." Nature Genetics 26 (2):216-220.

Nilsson, R., I. Schultz, E. Pierce, K. Soltis, A. Naranuntarat, D. Ward, J. Baughman, P. Paradkar, P. Kingsley, V.Culotta, J. Kaplan, J. Palis, B. Paw, and V. Mootha. 2009. "Discovery of Genes Essential for Heme Biosynthesis through Large-Scale Gene Expression Analysis."Cell Metabolism 10 (2):119-130.

Niture, S. K., R. Khatri, and A. K. Jaiswal. 2014. "Regulation of Nrf2-an update."Free Radical Biology and Medicine 66:36-44.

Niture, S., J. Kaspar, J. Shen, and A. Jaiswal. 2010. "Nrf2 signaling and cell survival." Toxicology and Applied Pharmacology 244 (1):37-42.

North, T.., W. Goessling, M. Peeters, P. Li, C. Ceol, A. Lord, G. Weber, J. Harris, C. Cutting, P. Huang, E. Dzierzak, and L.I. Zon. 2009. "Hematopoietic Stem Cell Development Is Dependent on Blood Flow." Cell 137 (4):736-748.

O'brien, B. R. A. 1961. "Identification of Haemoglobin by its Catalase Reaction with Peroxide and o-Dianisidine." Biotechnic & Histochemistry 36 (2):57-61.

O'Shields, B., A. G. McArthur, A. Holowiecki, M. Kamper, J. Tapley, and M. J. Jenny. 2014. "Inhibition of endogenous MTF-1 signaling in zebrafish embryos identifies novel roles for MTF- 1 in development."Biochimica et Biophysica Acta (BBA)- Molecular Cell Research.

Padilla, P., and Roth, M. . 2001. "Oxygen deprivation causes suspended animation in the zebrafish embryo."Proceedings of the National Academy of Sciences 98 (13):7331-7335.

Patton, E., and Zon, L. 2001. "The Art and Design of Genetic Screens: Zebrafish."Nature Reviews Genetics 2:956-966.

Peitsaro, N., J. Kaslin, O. Anichtchik, and P. Panula. 2003. "Modulation of the histaminergic system and behaviour by α-fluoromethylhistidine in zebrafish."Journal of Neurochemistry 86 (2):432-441.

139

Pevny, L., C. S. Lin, V. D'Agati, M. C. Simon, S. H. Orkin, and F. Costantini. 1995. "Development of hematopoietic cells lacking transcription factor GATA-1."Development 121 (1):163-72.

Pevny, L., M. C. Simon, E. Robertson, W. H. Klein, S. F. Tsai, V. D'Agati, S. H. Orkin, and F. Costantini. 1991. "Erythroid differentiation in chimaeric mice blocked by a targeted mutation in the gene for transcription factor GATA-1."Nature 349 (6306):257-60.

Ponka, P. 1997. "Tissue-Specific Regulation of Iron Metabolism and Heme Synthesis: Distinct Control Mechanisms in Erythroid Cells."Blood 89 (1):1-25.

Poss, K. D., and S. Tonegawa. 1997a. "Reduced stress defense in heme oxygenase 1-deficient cells."Proceedings of the National Academy of Sciences 94 (20):10925-30.

Poss, K., M. Keating, and A. Nechiporuk. 2003. "Tales of regeneration in zebrafish." Developmental Dynamics 226 (2):202-210.

Poss, K.., and S. Tonegawa. 1997b. "Heme oxygenase 1 is required for mammalian iron reutilization."Proceedings of the National Academy of Sciences 94 (20):10919- 10924.

Postlethwait, J. H. 2007. "The zebrafish genome in context: ohnologs gone missing."Journal of Experimental Zoology Part B: Molecular and Developmental Evolution 308 (5):563-77.

Postlethwait, J., A. Amores, W. Cresko, A. Singer, and Y. Yan. 2004. "Subfunction partitioning, the teleost radiation and the annotation of the human genome."Trends in Genetics 20 (10):481- 490.

Qin, X. 2007. "Inactivation of digestive proteases by deconjugated bilirubin: the possible evolutionary driving force for bilirubin or biliverdin predominance in animals."Gut 56 (11):1641-2.

Raju, V., McCoubrey, W. and Maines, M. 1997. "Regulation of heme oxygenase-2 by glucocorticoids in neonatal rat brain: characterization of a functional glucocorticoid response element."Biochimica et Biophysica Acta (BBA) - Gene Structure and Expression 1351 (1–2):89- 104.

Ransom, D., Bahary, N., Niss, K., Traver, D., Burns, C., Trede, N., Paffett-Lugassy, N., Saganic, W., Lim, C., Hersey, C., Zhou, Yi., Barut, B., Lin, S., Kingsley, P., Palis, J., Orkin, S., and L. Zon. 2004. "The Zebrafish moonshine Gene Encodes Transcriptional Intermediary Factor 1γ, an Essential Regulator of Hematopoiesis."PLoS Biology 2 (8):e237.

Rhodes, J., A. Hagen, K. Hsu, M. Deng, T. Liu, A. Look, and J. Kanki. 2005. "Interplay of Pu.1 and Gata1 Determines Myelo-Erythroid Progenitor Cell Fate in Zebrafish."Developmental Cell 8 (1):97-108.

140

Robinson, S. R., T. N. Dang, R. Dringen, and G. M. Bishop. 2009. "Hemin toxicity: a preventable source of brain damage following hemorrhagic stroke." Redox Report:Communications in Free Radical Research 14 (6):228-35.

Rombough, P. 2002. "Gills are needed for ionoregulation before they are needed for O(2) uptake in developing zebrafish, Danio rerio."The Journal of Experimental Biology 205 (Pt 12):1787- 1794.

Rossi, A., Z. Kontarakis, C. Gerri, H. Nolte, S. Holper, M. Kruger, and D. Y. Stainier. 2015. "Genetic compensation induced by deleterious mutations but not gene knockdowns."Nature 524 (7564):230-3.

Rydstrom, J. 2006. "Mitochondrial NADPH, transhydrogenase and disease."Biochimica et Biophysica Acta (BBA) -Bioenergetics 1757 (5-6):721-726.

Ryter, S., J. Alam, and A. Choi. 2006. "Heme Oxygenase-1/Carbon Monoxide: From Basic Science to Therapeutic Applications." Physiological Reviews 86 (2):583-650.

Rzymkiewicz, D. M., J. R. Reddan, and U. P. Andley. 2000. "Induction of heme oxygenase-1 modulates cis-aconitase activity in lens epithelial cells."Biochemical and Biophysical Research Communications 270 (1):324-8.

Saleh, M. C., and S. McConkey. 2012. "NADH-dependent cytochrome b5 reductase and NADPH methemoglobin reductase activity in the erythrocytes of Oncorhynchus mykiss."Fish Physiology and Biochemistry 38 (6):1807-1813.

Salim, M., Brown-Kipphut, B., and Maines, M. 2001. "Human Biliverdin Reductase Is Autophosphorylated, and Phosphorylation Is Required for Bilirubin Formation."Journal of Biological Chemistry 276 (14):10929-10934.

Schacter, B. A., E. B. Nelson, H. S. Marver, and B. S. Masters. 1972. "Immunochemical evidence for an association of heme oxygenase with the microsomal electron transport system." Journal of Biological Chemistry 247 (11):3601-7.

Schmittgen, T. and Livak, K. 2008. "Analyzing real-time PCR data by the comparative CT method." Nature Protocols 3 (1101-1108).

Schonthaler, H. B., T. A. Franz-Odendaal, C. Hodel, I. Gehring, R. Geisler, H. Schwarz, S. C. Neuhauss, and R. Dahm. 2010. "The zebrafish mutant bumper shows a hyperproliferation of lens epithelial cells and fibre cell degeneration leading to functional blindness."Mechanisms of Development 127 (3-4):203-19.

Scott, E. W., M. C. Simon, J. Anastasi, and H. Singh. 1994. "Requirement of transcription factor PU.1 in the development of multiple hematopoietic lineages."Science 265 (5178):1573-7.

Sedlak, T. and Snyder, S. 2004. "Bilirubin benefits: cellular protection by a biliverdin reductase antioxidant cycle."Pediatrics 113 (6):1776-1782.

141

Sedlak, T. and Snyder, S. 2009a. "Cycling the wagons for biliverdin reductase."Journal of Biological Chemistry 284 (46):le11; author reply le12.

Sedlak, T., Saleh, M., Higginson, D., Paul, B., Juluri, K. and Snyder, S. 2009b. "Bilirubin and glutathione have complementary antioxidant and cytoprotective roles."Proceedings of the National Academy 106 (13):5171-5176.

Semenza, G. 2000. "HIF-1: mediator of physiological and pathophysiological responses to hypoxia."Journal of Applied Physiology 88 (4):1474-1480.

Semenza, G., B. Jiang, S. Leung, R. Passantino, J. Concordet, P. Maire, and A. Giallongo. 1996. "Hypoxia Response Elements in the Aldolase A, Enolase 1, and Lactate Dehydrogenase A Gene Promoters Contain Essential Binding Sites for Hypoxia-inducible Factor 1."Journal of Biological Chemistry 271 (51):32529-32537.

Shalloe, F., Gordon, E, Ennis, O. and Mantle, T. 1996. "Evidence that biliverdin-IXb reductase and flavin reductase are identical."Biochemical Journal 316:38-387.

Simmons, S. O., C. Y. Fan, and R. Ramabhadran. 2009. "Cellular stress response pathway system as a sentinel ensemble in toxicological screening."Toxicological Sciences 111 (2):202-25.

Singh, K. P., R. W. Garrett, F. L. Casado, and T. A. Gasiewicz. 2011. "Aryl hydrocarbon receptor-null allele mice have hematopoietic stem/progenitor cells with abnormal characteristics and functions."Stem Cells and Development 20 (5):769-84.

Singleton, J., and L. Laster. 1965. "Biliverdin Reductase of Guinea Pig Liver."Journal of Biological Chemistry 240 (12):4780-4789.

Smith, B. W., S. S. Rozelle, A. Leung, J. Ubellacker, A. Parks, S. K. Nah, D. French, P. Gadue, S. Monti, D. H. Chui, M. H. Steinberg, A. L. Frelinger, A. D. Michelson, R. Theberge, M. E. McComb, C. E. Costello, D. N. Kotton, G. Mostoslavsky, D. H. Sherr, and G. J. Murphy. 2013. "The aryl hydrocarbon receptor directs hematopoietic progenitor cell expansion and differentiation."Blood 122 (3):376-85.

So, J. H., J. D. Kim, K. W. Yoo, H. T. Kim, S. H. Jung, J. H. Choi, M. S. Lee, S. W. Jin, and C. H. Kim. 2014. "FIH-1, a novel interactor of mindbomb, functions as an essential anti-angiogenic factor during zebrafish vascular development."PLoS One 9 (10):e109517.

Spitsbergen, J. M., and M. L. Kent. 2003. "The state of the art of the zebrafish model for toxicology and toxicologic pathology research--advantages and current limitations."Toxicologic Pathology 31:62-87.

Sporn, M. B., and K. T. Liby. 2012. "NRF2 and cancer: the good, the bad and the importance of context."Nature Reviews: Cancer 12 (8):564-571.

Stainier, D. Y., Z. Kontarakis, and A. Rossi. 2015. "Making sense of anti-sense data." Developmental Cell 32 (1):7-8.

142

Stocker, R, and B N Ames. 1987. "Potential role of conjugated bilirubin and copper in the metabolism of lipid peroxides in bile."Proceedings of the National Academy of Sciences 84 (22):8130-8134.

Stocker, R., A. N. Glazer, and B. N. Ames. 1987. "Antioxidant activity of albumin-bound bilirubin."Proceedings of the National Academy of Sciences 84 (16):5918-22.

Stocker, R., Yamamoto, Y., McDonagh, A., Glazer, A., and Ames, B. 1987. "Bilirubin is an antioxidant of possible physiological importance."Science 235 (4792):1043-1046.

Stuart, G., Searle, P., Palmiter, R. 1985. "Identification of multiple metal regulatory elements in mouse metallothionein-I promoter by assaying synthetic sequences."Nature 317 ( 6040):828- 831.

Sullivan-Brown, J., M. E. Bisher, and R. D. Burdine. 2011. "Embedding, serial sectioning and staining of zebrafish embryos using JB-4 resin."Nature Protocols 6 (1):46-55.

Sun, Y., Rotenberg, M. and Maines, M. 1990. "Developmental expression of heme oxygenase isozymes in rat brain. Two HO-2 mRNAs are detected."Journal of Biological Chemistry 265 (14):8212-7.

Sun, Z., A. Amsterdam, G. J. Pazour, D. G. Cole, M. S. Miller, and N. Hopkins. 2004. "A genetic screen in zebrafish identifies cilia genes as a principal cause of cystic kidney." Development 131 (16):4085-93.

Takahashi, S., R. Shimizu, N. Suwabe, T. Kuroha, K. Yoh, J. Ohta, S. Nishimura, K. C. Lim, J. D. Engel, and M. Yamamoto. 2000. "GATA factor transgenes under GATA-1 locus control rescue germline GATA-1 mutant deficiencies."Blood 96 (3):910-6.

Takeda, K., S. Ishizawa, M. Sato, T. Yoshida, and S. Shibahara. 1994. "Identification of a cis- acting element that is responsible for cadmium-mediated induction of the human heme oxygenase gene."Journal of Biological Chemistry 269 (36):22858-67.

Tenhunen, R., H.. Marver, and R. Schmid. 1969. "Microsomal Heme Oxygenase: Characterization of the Enzyme."Journal of Biological Chemistry 244 (23):6388-6394.

Tenhunen, R., M. Ross, H. Marver, and R. Schmid. 1970. "Reduced nicotinamide adenine dinucleotide phosphate dependent biliverdin reductase. Partial purification and characterization." Biochemistry 9 (2):298-303.

Thisse, B., V. Heyer, A. Lux, V. Alunni, A. Degrave, I. Seiliez, J. Kirchner, J. P. Parkhill, and C. Thisse. 2004. "Spatial and temporal expression of the zebrafish genome by large-scale in situ hybridization screening."Methods in Cell Biology 77:505-19.

Thisse, C., and Thisse, B. 2007. "High-resolution in situ hybridization to whole-mount zebrafish embryos."Nature Protocols 3 (1):59-69.

143

Thompson, M., D. Ransom, S. Pratt, H. MacLennan, M. Kieran, W. Detrich, B. Vail, T. Huber, B. Paw, A. Brownlie, A. Oates, A. Fritz, M. Gates, A. Amores, N. Bahary, W. Talbot, H. Her, D. Beier, J. Postlethwait, and L. Zon. 1998. "The cloche and spadetail Genes Differentially Affect Hematopoiesis and Vasculogenesis."Developmental Biology 197 (2):248-269.

Tijet, N., P. Boutros, I. Moffat, A. Okey, J. Tuomisto, and R. Pohjanvirta. 2006. "Aryl Hydrocarbon Receptor Regulates Distinct Dioxin-Dependent and Dioxin-Independent Gene Batteries."Molecular Pharmacology 69 (1):140-153.

Timme-Laragy, A., Karchner, S., Franks, D., Jenny, M.,Harbeitner, R., Goldstone, J., McArthur, A. and Hahn, M. 2012. "Nrf2b, novel zebrafish paralog of oxidant-responsive transcription factor NF-E2-related factor 2 (NRF2)."Journal of Biological Chemistry 287 (7):4609-27.

Timme-Laragy, A. R., J. V. Goldstone, B. R. Imhoff, J. J. Stegeman, M. E. Hahn, and J. M. Hansen. 2013a. "Glutathione redox dynamics and expression of glutathione-related genes in the developing embryo."Free Radical Biology and Medicine 65:89-101.

Ton, C. Stamatiou, D. and Liew, C. 2003. Gene expression profile of zebrafish exposed to hypoxia during development. Physiological Genomics 13:97-106.

Trainor, C. D., R. Ghirlando, and M. A. Simpson. 2000. "GATA zinc finger interactions modulate DNA binding and transactivation."Journal of Biological Chemistry 275 (36):28157-66.

Trakshel, G M, R K Kutty, and M D Maines. 1986. "Purification and characterization of the major constitutive form of testicular heme oxygenase. The noninducible isoform."Journal of Biological Chemistry 261 (24):11131-11137.

Trakshel, G. M., R. K. Kutty, and M. D. Maines. 1988. "Resolution of the rat brain heme oxygenase activity: absence of a detectable amount of the inducible form (HO-1)."Archives of Biochemistry and Biophysics 260 (2):732-9.

Traver, D., P. Herbomel, E. Patton, R. Murphey, J. Yoder, G. Litman, A. Catic, C. Amemiya, L. Zon, and N. Trede. 2003. "The Zebrafish as a Model Organism to Study Development of the Immune System."In Advances in Immunology, 254-330.

Troadec, M. B., D. M. Ward, E. Lo, J. Kaplan, and I. De Domenico. 2010. "Induction of FPN1 transcription by MTF-1 reveals a role for ferroportin in transition metal efflux."Blood 116 (22):4657-64.

Tsai, J. J., J. A. Dudakov, K. Takahashi, J. H. Shieh, E. Velardi, A. M. Holland, N. V. Singer, M. L. West, O. M. Smith, L. F. Young, Y. Shono, A. Ghosh, A. M. Hanash, H. T. Tran, M. A. Moore, and M. R. van den Brink. 2013. "Nrf2 regulates haematopoietic stem cell function." Nature Cell Biology 15 (3):309-16.

Tudor, C., Lerner-Marmarosh, N., Engelborghs, Y., Gibbs, P. and Maines, M. 2008. "Biliverdin reductase is a transporter of haem into the nucleus and is essential for regulation of HO-1 gene expression by haematin."Biochemical Journal 413 (3):405-416.

144

Turner, T. N., K. Sharma, E. C. Oh, Y. P. Liu, R. L. Collins, M. X. Sosa, D. R. Auer, H. Brand, S. J. Sanders, D. Moreno-De-Luca, V. Pihur, T. Plona, K. Pike, D. R. Soppet, M. W. Smith, S. W. Cheung, C. L. Martin, M. W. State, M. E. Talkowski, E. Cook, R. Huganir, N. Katsanis, and A. Chakravarti. 2015. "Loss of delta-catenin function in severe autism."Nature.

Uruno, A., and H. Motohashi. 2011. "The Keap1–Nrf2 system as an in vivo sensor for electrophiles."Nitric Oxide 25 (2):153-160. van den Hurk, Peter. 2006. "Bile fluorescence, heme oxygenase induction, and increased biliverdin excretion by mixtures of environmental toxicants."Aquatic Toxicology 77 (2):202-209. van Ommen, B., Koster, A., Verhagen, H. and van Bladeren, P. 1992. "The glutathione conjugates of tert-butyl hydroquinone as potent redox cycling agents and possible reactive agents underlying the toxicity of butylated hydroxyanisole."Biochemical and Biophysical Research Communications 189 (1):309-314.

Venugopal, R and Jaiswal, A. 1998. "Nrf2 and Nrf1 in association with Jun proteins regulate antioxidant response element-mediated expression and coordinated induction of genes encoding detoxifying enzymes."Oncogene 17 (24):3145-3156.

Verma, A., D. J. Hirsch, C. E. Glatt, G. V. Ronnett, and S. H. Snyder. 1993. "Carbon monoxide: a putative neural messenger."Science 259 (5093):381-4.

Vihervaara, Anniina, and Lea Sistonen. 2014. "HSF1 at a glance."Journal of Cell Science 127 (2):261-266. doi: 10.1242/jcs.132605.

Vukomanovic, D., B. McLaughlin, M. Rahman, W. Szarek, J. Brien, Z. Jia, and K. Nakatsu. 2011. "Selective activation of heme oxygenase-2 by menadione."Canadian Journal of Physiology and Pharmacology 89 (11):861-864.

Walters, K.., J. Green, J. Surfus, S. Yoo, and A. Huttenlocher. 2010. "Live imaging of neutrophil motility in a zebrafish model of WHIM syndrome."Blood 116 (15):2803-2811.

Wang, H., Q. Long, S. Marty, S. Sassa, and S. Lin. 1998. "A zebrafish model for hepatoerythropoietic porphyria."Nature Genetics 20 (3):239-243.

Wang, Y., J. Fang, S. S. Leonard, and K. M. Krishna Rao. 2004. "Cadmium inhibits the electron transfer chain and induces reactive oxygen species."Free Radical Biology and Medicine 36 (11):1434-1443.

Wang, Y., Wimmer, U., Lichtlen, P., Inderbitzin, D., Stieger, B., Meier, P., Hunziker, L., Stallmach,T., Forrer,R.,Rulicke,T.,Georgiev, O. and Schaffner, W. 2004. "Metal-responsive transcription factor-1 (MTF-1) is essential for embryonic liver development and heavy metal detoxification in the adult liver."The FASEB journal 18 (10):1071-1079.

Watanabe, S., R. Akagi, M. Mori, T. Tsuchiya, and S. Sassa. 2004. "Marked Developmental Changes in Heme Oxygenase-1 (HO-1) Expression in the Mouse Placenta: Correlation Between HO-1 Expression and Placental Development."Placenta 25 (5):387-395.

145

Wegiel, B., C. Baty, D. Gallo, E. Csizmadia, J. Scott, A. Akhavan, B. Chin, E. Kaczmarek, J. Alam, F. Bach, B. Zuckerbraun, and L. Otterbein. 2009. "Cell Surface Biliverdin Reductase Mediates Biliverdin-induced Anti-inflammatory Effects via Phosphatidylinositol 3-Kinase and Akt." Journal of Biological Chemistry 284 (32):21369-21378.

Weinstein, B.. Schier, A., Abdelilah, S., Malicki, J., Solnica-Krezel, L., Stemple, D., Stainier, D., Zwartkruis, F., Driever, W. and Fishman, M. 1996. "Hematopoietic mutations in the zebrafish." Development 123 (1):303-309.

Weiss, M. J., G. Keller, and S. H. Orkin. 1994. "Novel insights into erythroid development revealed through in vitro differentiation of GATA-1 embryonic stem cells."Genes & Development 8 (10):1184-97.

Weiss, M. J., and S. H. Orkin. 1995. "Transcription factor GATA-1 permits survival and maturation of erythroid precursors by preventing apoptosis."Proceedings of the National Academy of Sciences 92 (21):9623-7.

Whitlock, J. P., Jr. 1999. "Induction of cytochrome P4501A1."Annual Review of Pharmacology and Toxicology 39:103-25. doi: 10.1146/annurev.pharmtox.39.1.103.

Whyatt, D. J., E. deBoer, and F. Grosveld. 1993. "The two zinc finger-like domains of GATA-1 have different DNA binding specificities."The EMBO Journal 12 (13):4993-5005.

Whyatt, D. J., A. Karis, I. C. Harkes, A. Verkerk, N. Gillemans, A. G. Elefanty, G. Vairo, R. Ploemacher, F. Grosveld, and S. Philipsen. 1997. "The level of the tissue-specific factor GATA- 1 affects the cell-cycle machinery."Genes and Function 1 (1):11-24.

Willett, C., A. Cortes, A. Zuasti, and A. Zapata. 1999. "Early hematopoiesis and developing lymphoid organs in the zebrafish."Developmental Dynamics 214 (4):323-336.

Wingert, R., A. Brownlie, J. Galloway, K. Dooley, P. Fraenkel, J. Axe, A. Davidson, B. Barut, L. Noriega, X. Sheng, Y. Zhou, and L. Zon. 2004. "The chianti zebrafish mutant provides a model for erythroid-specific disruption of transferrin receptor 1."Development 131 (24):6225-6235.

Wu, H., Y. Yu, L. David, Y. S. Ho, and M. F. Lou. 2014. "Glutaredoxin 2 (Grx2) gene deletion induces early onset of age-dependent cataracts in mice."Journal of Biological Chemistry 289 (52):36125-39.

Wu, K., Cui, J., and Klaassen, C. 2011. "Beneficial role of Nrf2 in regulating NADPH generation and consumption."Toxicological Sciences 123 (2):590-600.

Xu, F., C. P. Mack, K. S. Quandt, M. Shlafer, V. Massey, and D. E. Hultquist. 1993. "Pyrroloquinoline Quinone Acts with Flavin Reductase to Reduce Ferryl Myoglobin in Vitro and Protects Isolated Heart from Reoxygenation Injury."Biochemical and Biophysical Research Communications 193 (1):434-439.

146

Xu, F., Quandt, K. and Hultquist, D. 1992. "Characterization of NADPH-dependent methemoglobin reductase as a heme-binding protein present in erythrocytes and liver." Proceedings of the National Academy of Sciences 89 (6):2130-2134.

Xu, X., S. E. Meiler, T. P. Zhong, M. Mohideen, D. A. Crossley, W. W. Burggren, and M. C. Fishman. 2002. "Cardiomyopathy in zebrafish due to mutation in an alternatively spliced exon of titin."Nature Genetics 30 (2):205-9.

Yachie, A., Y. Niida, T. Wada, N. Igarashi, H. Kaneda, T. Toma, K. Ohta, Y. Kasahara, and S. Koizumi. 1999. "Oxidative stress causes enhanced endothelial cell injury in human heme oxygenase-1 deficiency."Journal of Clinical Investigation 103 (1):129-35.

Yamaguchi, T. and Nakajima, H. 1995. "Changes in the Composition of Bilirubin IX Isomers During Human Prenatal Development."European Journal of Biochemistry 233 (2):467-472.

Yamaguchi, T., Komoda, Y. and Nakajima, H. 1994. "Biliverdin-IX alpha reductase and biliverdin-IX beta reductase from human liver. Purification and characterization."Journal of Biological Chemistry 269 (39):24343.

Yamaguchi, T., A. Komuro, Y. Nakano, M. Tomita, and H. Nakajima. 1993. "Complete Amino Acid Sequence of Biliverdin-IXβ Reductase from Human Liver."Biochemical and Biophysical Research Communications 197 (3):1518-1523.

Yamamoto, M., L. J. Ko, M. W. Leonard, H. Beug, S. H. Orkin, and J. D. Engel. 1990. "Activity and tissue-specific expression of the transcription factor NF-E1 multigene family."Genes & Development 4 (10):1650-62.

Yang, H. Y., and T. Evans. 1992. "Distinct roles for the two cGATA-1 finger domains." Molecular and Cellular Biology 12 (10):4562-70.

Yao, E., and M. Denison. 1992. "DNA sequence determinants for binding of transformed Ah receptor to a dioxin-responsive enhancer."Biochemistry 31 (21):5060-5067.

Yeung, E., and C. Chan. 2014. "Glycol methacrylate: the art of embedding and serial sectioning."Botany 93 (1):1-8.

Yin, F., H. Sancheti, and E. Cadenas. 2012. "Mitochondrial Thiols in the Regulation of Cell Death Pathways."Antioxidants & Redox Signaling.

Zakhary, R, S P Gaine, J L Dinerman, M Ruat, N A Flavahan, and S H Snyder. 1996. "Heme oxygenase 2: endothelial and neuronal localization and role in endothelium-dependent relaxation."Proceedings of the National Academy of Sciences 93 (2):795-798.

Zhang, X. Y., and A. R. Rodaway. 2007. "SCL-GFP transgenic zebrafish: in vivo imaging of blood and endothelial development and identification of the initial site of definitive hematopoiesis."Developmental Biology 307 (2):179-94.

147

Zon, L. and Peterson, R. 2005. "In vivo drug discovery in the zebrafish."Nature Reviews Drug Discovery 4 (1):35-44.

148