<Abstract Centered> an ABSTRACT of the THESIS OF

AN ABSTRACT OF THE DISSERTATION OF

Brian A. Head for the degree of Doctor of Philosophy in Molecular and Cellular Biology presented on February 26, 2021.

Title: Vitamin E Protects Developmental Gene Expression and Metabolic Networks in Zebrafish Embryos

Abstract approved:

______

Maret G. Traber

Vitamin E (VitE) is necessary for vertebrate embryonic development. VitE prevents lipid peroxidation (LPO), which requires detoxification by cellular antioxidant systems subsequently involving reducing power derived from energy metabolism. Thus, VitE protects metabolic networks in the developing embryo and the integrated gene expression networks compensating for and impaired by LPO-induced errors. In this thesis, I present four experimental chapters regarding VitE’s vital role in the time- dependent protection of embryo developmental gene expression and metabolic networks using the zebrafish model. I begin by evaluating gastrulation, neural plate formation and neural crest cell migration defined by expression of key transcription factors goosecoid, sox10, and pax2a. VitE-deficient (E–) zebrafish embryos appear normal during neuroectoderm formation at 6 hours post-fertilization (hpf), however, by 12 hpf E– embryos are developmentally abnormal with mis-localized expression of pax2a and sox10 in the early midbrain-hindbrain boundary and neural crest cells. At 24 hpf E– further have reduced expression of these latter transcription factors distributed throughout the spinal neurons and in trunk neural crest cells. Patterning defects precede the obvious morphological impairments in E– embryos as well as histologic evidence that all brain regions develop erroneously by 24 hpf. In addition, the α-tocopherol transfer protein (TTPA) gene is expressed at the leading edges of the brain ventricle border; the brains of E– embryos were often over- or under inflated by 24 hpf. I then evaluated the whole transcriptome profile of VitE-sufficient (E+) and E– embryos at 12-, 18- and 24 hpf using RNAseq. By gene annotation, gene set enrichment, and integrated metabolomic analyses I found that E– embryos experience significant disruption to expression of genes associated with energy metabolism. Specifically, E– relative to E+ embryos have increased expression of genes involved in glycolysis and the pentose phosphate pathway, while they have decreased expression of genes involved in anabolic pathways and gene transcription. More importantly, when both gene expression and the metabolome in embryos at 24 hpf were analyzed together, the mechanistic Target of Rapamycin (mTOR) signaling pathway was found diminished in E– embryos. VitE protected gene expression associated with energy metabolism, as integrated by the cell cycle survival and growth protein complex mTOR. Gene expression disruption may precede the errors in embryonic patterning visualized with sox10 and pax2a. To ascertain the metabolic impacts of a VitE deficiency, we then developed sensitive UPLC-MS/MS methods to identify and quantify amino acids and thiol-related antioxidants in the E+ and E– embryos. Betaine, the oxidation product of choline was increased in E– embryos at 12-, 24- and 48 hpf. At 24 hpf, E– embryos contained less choline and by 48 hpf glutathione (GSH) was also decreased. It is likely that choline, an integral component of the membrane lipid, phosphatidyl choline, is used to generate betaine, which is a substrate for the methionine cycle and needed to produce methyl donors. These methyl donors, including S-adenosyl methionine (SAM) were also decreased in E– embryos. SAM is used to generate homocysteine for synthesis of cysteine, the rate-limiting amino acid in glutathione (GSH) synthesis. Ultimately, our analyses indicate that thiols become depleted in E– embryos because LPO generates products that requires compensation using limited amino acids and methyl donors that are also developmentally relevant. The first three experimental chapters demonstrate the interconnected reality of VitE-deficiency

induced LPO that disrupts antioxidant balance; thiol and energy metabolic status; and gene expression patterns leading to increased morbidity and mortality outcomes. Finally, we aimed to generate a transgenic model to track VitE-dependent cells in vivo with a fluorescently tagged TTPA protein. Using CRISPR-Cas technology, pigment-less Casper zebrafish embryos were injected with cas9 complex proteins and a plasmid containing the mScarlet coding sequence targeted 5’ of the first exon of ttpa. Homozygote mutant fish, named the RedEfish, were found to express the mScarlet protein and produce red fluorescence in the olfactory pits, digestive tract including the liver, and posterior tail fin by 7 days post-fertilization (dpf). At 14 dpf the RedEfish also expressed mScarlet in the caudal vertebrae edges identified as likely dorsal root ganglia and the caudal vein plexus. The TTPA protein function was not impaired by the addition of the mScarlet genomic sequence, however, the VitE status of the RedEfish was 30% lower relative to Casper zebrafish of the same age, suggesting some TTPA dysfunction. This model provides a valuable new means of testing and visualizing VitE necessity during vertebrate development. This thesis provides ample evidence supporting VitE as a necessary dietary factor for embryonic survival. It is thus critical to continue evaluating the physiologically relevant nature of a simple dietary deficiency in a developmental window defined by neurogenesis and specific gene expression and metabolic needs.

Copyright by Brian A. Head

February 26, 2021

Vitamin E Protects Developmental Gene Expression and Metabolic Networks in Zebrafish Embryos

Brian A. Head

A DISSERTATION

submitted to

Oregon State University

in partial fulfillment of the requirements for the degree of

Doctor of Philosophy

Presented February 26, 2021

Commencement June 2021

Doctor of Philosophy dissertation of Brian A. Head presented on February 26, 2021.

APPROVED:

______

Major Professor, representing Molecular and Cellular Biology

______

Director of the Molecular and Cellular Biology Program

______

Dean of the Graduate School

I understand that my dissertation will become part of the permanent collection of Oregon State University libraries. My signature below authorizes release of my dissertation to any reader upon request.

______

Brian A. Head, Author

ACKNOWLEDGEMENTS

I could easily write another hundred pages expressing my gratitude for each person that contributed to my projects and helped me along this journey. First and foremost, my major advisor, Dr. Maret Traber, thank you. You are a leader, a model of perseverance and a champion for the scientist you saw in me. I am forever grateful for the hand that pushed me to think, question and grow. If it were easy, everybody would be doing it and I’m thankful to have had to opportunity to add to the great Vitamin E puzzle. Thank you for being my mentor.

To Dr. Robyn Tanguay, thank you for welcoming me into your lab, providing me space and resources to grow and for showing us all how not to sacrifice authenticity in the pursuit of our goals. To Dr. Chrissa Kioussi, thank you for your invaluable insight, your encouragement of my learning and your amazing ability to find joy and laughter through it all. And to the remaining members of my doctoral committee, Dr. Stephen

Ramsey and Dr. Dave Dallas, your experience and enthusiasm to be a part of these projects has been so appreciated.

To Scott Leonard, thank you for being my bench-side therapist, the best coffee date, and of course, greatest asset in our lab. I’m not sure how I could have survived without your guidance. To Dr. Jaewoo Choi, your expertise is unmatched, and Sandra

Uesugi, your office was always a safe space to gripe. Thank you to my crew at LPI and everyone who inevitably donated wisdom, reagents, or a smile to keep me going. To

Carrie Barton, thank you for answering all my text messages, being my library of

zebrafish knowledge and being the star that you are. To Jane La Du, your experience

and wit makes the dimmest fluorescent signal shine. Thank you to everyone at SARL, you are the best host for this parasite, and a special thank you to my thesis buddy, my

constant competitor and my friend, Dr. Prarthana Shankar.

Thank you to my Mom, Dad, Grandma, and Grandpa for supporting me every

step of the way, and to my friends, the family I’ve found here in Oregon. Randi Wilson,

you are the sister I never I asked for, and Pat Di Bello, thanks. Thank you to my

colleagues and friends Dr. Serhan Mermer, Dr. Sarah Erhlicher, and Faye Andrews for

talking life and science with me. To the friends that don’t quite understand what I do,

thank you for reminding me that there is life outside the lab. To Abdul, my best friend,

thank you for your unwavering love and support. Thank you to my four-legged

officemates Bohdie, Sophie and Samson for keeping me company as I write this thesis

beside you. And finally, thank you Corvallis, Oregon, you will always feel like home.

CONTRIBUTION OF AUTHORS

The authors contributions were as follows:

Chapter One: BH wrote.

Chapter Two: BH, JLD, CK, RLT and MGT contributed to conception or design of the work. All authors contributed to acquisition, analysis, or interpretation of data and have drafted or substantively revised the work.

Chapter Three: BH, SAR, CK, RLT and MGT contributed to conception or design of the work. All authors contributed to acquisition, analysis or interpretation of data and have drafted or substantively revised the work.

Chapter Four: BH, JZ, SWL, JC, RLT, and MGT contributed to conception or design of the work. All authors contributed to acquisition, analysis or interpretation of data and have drafted or substantively revised the work.

Chapter Five: BH wrote. RLT and MGT contributed to conception or design of the work. BH, JLD, CB, JZ, RLT and MGT contributed to acquisition, analysis, or interpretation of data.

Chapter Six: BH wrote

TABLE OF CONTENTS

Page

CHAPTER ONE: Introduction ...... 1

Vitamin E: forms and antioxidant function ...... 1

Regulation of Vitamin E Trafficking by TTPA ...... 3

Vitamin E deficiency ...... 5

Embryonic VitE Requirement ...... 7

Zebrafish as a developmental model ...... 8

Zebrafish Embryonic VitE Requirement ...... 9

Neurogenesis ...... 10

Hypothesis & Objectives ...... 12

CHAPTER TWO: ...... 13

Vitamin E is necessary for zebrafish nervous system development ...... 13

ABSTRACT ...... 14

INTRODUCTION ...... 14

MATERIALS AND METHODS ...... 17

Zebrafish husbandry ...... 17

Morphological and histological assessment ...... 18

Whole mount in situ hybridization...... 19

Relative reverse transcription PCR ...... 20

Statistical analyses ...... 20

RESULTS ...... 21

Morphological abnormalities associated with VitE deficiency ...... 21

Ttp localization ...... 21

Nervous system development markers ...... 22 TABLE OF CONTENTS (Continued)

Page

Quantitation of gene expression...... 24

Histological analysis ...... 24

DISCUSSION ...... 25

CHAPTER THREE: ...... 37

Vitamin E deficiency disrupts gene expression networks during zebrafish development ...... 37

ABSTRACT ...... 38

INTRODUCTION ...... 38

MATERIALS AND METHODS ...... 40

Zebrafish husbandry ...... 40

RNA extraction and sequencing ...... 41

Data deposition ...... 42

Data processing and statistical analysis ...... 42

Gene ontology and network enrichment ...... 43

Western blotting ...... 43

RESULTS ...... 43

Hierarchical clustering and gene annotation ...... 44

Top differentially expressed genes in E– embryos ...... 45

DISCUSSION ...... 48

CHAPTER FOUR: ...... 60

Vitamin E deficiency dysregulates thiols, amino acids and related molecules during zebrafish embryogenesis ...... 60

ABSTRACT ...... 61 TABLE OF CONTENTS (Continued)

Page

INTRODUCTION ...... 61

MATERIALS AND METHODS ...... 63

Materials and Reagents ...... 63

Zebrafish husbandry ...... 64

Thiol extraction and determination by UPLC-MS/MS ...... 65

Assessment of thiol extraction and quantitation method development ...... 67

Analysis of amino acid and related metabolites ...... 68

Repeatability and reproducibility of measurements: thiols, amino acids, and related metabolites ...... 69

Statistical analyses, principal component analyses and hierarchical clustering ...... 69

RESULTS ...... 70

Embryo morphology and egg quality assessment ...... 70

Assessment of thiol extraction and quantitation method development ...... 71

Repeatability and reproducibility of measurements: thiols, amino acids and related metabolites ...... 72

Impact of VitE deficiency on GSH and GSSG ...... 73

Integrated analysis of thiols and amino acids ...... 73

Hierarchical clustering of E+ and E– outcomes ...... 74

Pathways relating AA and Thiol Status of E+ and E– embryos ...... 75

DISCUSSION ...... 76

CHAPTER FIVE: ...... 86

RedEfish: generation of the polycistronic mScarlet:T2A:Ttpa zebrafish line ...... 86

ABSTRACT ...... 87 TABLE OF CONTENTS (Continued)

Page

INTRODUCTION ...... 88

METHODS ...... 89

Zebrafish husbandry & strains...... 89

Construction of donor plasmid and guide RNA vectors for knock-in ...... 89

Microinjection ...... 89

Mutation detection and insertion mapping ...... 90

In vivo fluorescence detection ...... 91

Egg quality and embryo morphological assessment ...... 91

Whole mount immunolocalization and Western blot analysis ...... 91

RESULTS ...... 93

Characterization of RedEfish ...... 93

CHAPTER SIX: ...... 104

CONCLUSIONS AND FUTURE DIRECTIONS ...... 104

Future Directions ...... 108

Pharmacologic modulation mTOR and energy-related pathways ...... 108

Characterization of VitE-trafficking dependent cell types ...... 109

Final thoughts ...... 111

BIBLIOGRAPHY ...... 112

SUPPLEMENTARY...... 130 LIST OF FIGURES

Figure Page

1. Morphological abnormalities associated with VitE deficiency at 12-, 24- and 48 hours post-fertilization ...... 29

2. Ttpa signal localized throughout early embryo and brain ventricle borders regardless of VitE status ...... 30

3. Gastrulation marker goosecoid (gsc) is not affected by VitE deficiency ...... 31

4. Midbrain-hindbrain boundary formation shown by pax2a expression is dysregulated by VitE deficiency ...... 32

5. Neural crest cell migration impaired during development by VitE deficiency ...... 33

6. Notochord collagen markers col2a1a and col9a2 affected by dietary treatment ...... 34

7. Relative gene expression of neurogenesis markers ...... 35

8. Histological analysis of 24 hpf zebrafish embryos with morphological defects associated with VitE status ...... 36

9. Hierarchical clustering and gene annotation of differentially expressed genes in E– and E+ embryos at 12, 18 and 24 hpf ...... 53

10. Venn diagrams and KEGG pathways enriched by over representation analysis of differentially expressed genes changed consistently over time ...... 54

11. Gene ontology terms enriched by over representation analysis and gene set enrichment analysis of all differentially expressed genes ...... 56

12. Gene ontology terms enriched by gene set enrichment analysis of all expressed genes ...... 57

13. Mechanistic target of rapamycin (mTOR) signaling pathway disrupted in 24 hpf E– embryos ...... 58

14. Embryos are similar sized at each stage with E– developmentally delayed ...... 81

15. Schematic representation of 2D PLS-DA scores plot and important features ...... 82

16. Heatmap visualization of metabolite features and relative concentrations over time ...... 83

17. Quantitative analysis of thiols, amino acids and related substances ...... 84 LIST OF FIGURES (Continued)

Figure Page

18. Knock-in strategy and designed protein sequence ...... 100

19. In vivo fluorescence detected in 7- and 14 days post-fertilization RedEfish ...... 102

20. Immunolocalization and Western blot of TTPA protein ...... 103

S1. Gene set enrichment analysis plots generated using WebGestalt ...... 130

S2. RNA integrity numbers generated by the Center for Genome and Research Biocomputing ...... 131

S3. Multi-dimensional scaling plots of E+ and E– log10CPM clustered at each time point prior to data filtration and normalization ...... 132

S4. Quantitative analysis of other amino acids ...... 133

S5. Genotyping confirmation indicates successful genomic integration in heterozygote RedEfish ...... 142

S6. RFP expression confirmation in 3 days post-fertilization c269, Casper and RedEfish ...... 143

S7. Malondialdehyde increased in E– embryos at 24 hours post-fertilization ...... 144 LIST OF TABLES

Table Page

1. Top differentially expressed genes in E– embryos relative to E+ at 12, 18 and 24 hpf ...... 55

2. Top transcription factors differentially expressed in E– embryos across the developmental window measured ...... 59

3. GSH and GSSG concentrations (pmol/embryo) over time by embryo group ...... 85

4. Primers designed for insertion mapping ...... 101

S1. Quantitative analysis of other amino acids ...... 134

S2. MRM-parameters for the analysis of derivatized thiols and thiol disulfides ...... 135

S3. MRM-parameters and labeled-internal standards used for the analysis of amino acids and related substances ...... 136

S4. Linearity for the quantitative analysis of amino acids ...... 137

S5. Within-day and between-day coefficients of variation (%) of thiol analytes ...... 138

S6. Within-day and between-day coefficients of variation (%) of amino acid analytes 139 LIST OF ABBREVIATIONS

VitE Vitamin E

αToc α-tocopherol

TTPA α-Tocopherol transfer protein hpf Hours post-fertilization dpf Days post-fertilization

GSH Glutathione

GSSG Oxidized glutathione

LPO Lipid peroxidation

Cys Cysteine

Hcys Homocysteine 1 CHAPTER ONE: INTRODUCTION

I will first address how Vitamin E (VitE) is regulated by vertebrates and what are the causes and consequences of its deficiency, especially with regards to the nervous system. I will then focus on how VitE can be studied in the vertebrate model organism, zebrafish (Danio rerio) and provide a primer on zebrafish neurodevelopment. Subsequently, I will discuss the effects of a VitE deficiency on zebrafish neurodevelopment and on embryo-specific networks of gene expression patterns. To probe the metabolic consequences arising from VitE deficiency, I will discuss methyl donor and thiol antioxidant status during zebrafish development. Finally, to assess the specific cells that are affected by Vit E deficiency during neurogenesis, I will discuss the generation of a polycistronic zebrafish transgenic line using

CRISPR-Cas9 technology. I will conclude with a summary of the major outcomes, the relationship this work has with regards to human health, along with a brief discussion of future directions.

Vitamin E: forms and antioxidant function

Discovered as Factor X, VitE was first described as essential to restoring fertility to rats [1]. Maternal VitE deficiency caused abnormal rat fetal placentation leading to remarkable underdevelopment of early vascularization [2] followed by embryonic death on embryonic day (E) 12 or 13 and ultimately fetal resorption [1]. Thus, the embryonic VitE requirement is time-dependent [3-5]. The molecular basis for the VitE requirement, however, remains under investigation nearly 100 years later.

VitE is an umbrella term describing a family of eight plant-derived, fat-soluble compounds: four tocopherols (α-, β-, γ-, δ-tocopherol) and four tocotrienols (α-, β-, γ-, δ- tocotrienols). All natural VitE forms are synthesized only by photosynthesizing organisms [6], thus human dietary VitE sources are predominantly plant-based. Tocopherols and 2 tocotrienols can be distinguished by the saturation of their aliphatic side-chain at the 4’, 8’ and 12’ carbon positions; tocotrienols are unsaturated. Tocopherols contain 3 chiral centers positioned along the hydrophobic tail at the 2, 4’ and 8’ position. Thus, when α-tocopherol

(αT) is chemically synthesized, there are eight stereoisomers: RRR, RSR, RSS, RRS, SRR,

SSR, SRS and SSS. All natural tocopherols, however, are entirely in the RRR form. Finally,

α-, β-, γ-, and δ- forms can be distinguished by the methylation pattern of the chromanol ring.

Each chromanol ring possesses a hydroxyl group at the 6 position that can donate its hydrogen atom giving VitE its key function as an antioxidant. VitE forms, however, are not equally distributed throughout the body and do not contribute equally as antioxidants. In vivo,

αT is the main lipid soluble antioxidant; the mechanism for this selectivity is discussed further below.

The lipid peroxidation (LPO) chain reaction occurs when a peroxyl radical, which is formed by carbon centered radical (R•) reacting with molecular oxygen, reacts with other lipids [7]. αT reacts with peroxyl radicals 1000 times faster than they can react with other unsaturated lipids in the membrane [8]. VitE acts as a lipid-soluble, “chain-breaking” antioxidant by rapidly donating its hydrogen to peroxyl radicals to form lipid hydroperoxides

(LOOH) and the VitE radical (tocopheroxyl radical), the latter of which can be reduced by ascorbate (Vitamin C [VitC]), generating the ascorbate free radical (Asc•–) [9, 10]. Lipid hydroperoxides can then be reduced by the enzyme, glutathione peroxidase 4 (GPX4), to form a more stable lipid alcohol (LOH). Carbon centered radicals are also reducible by glutathione (GSH) forming a lipid (LH) prior propagation reaction with molecular oxygen. αT exerts its peroxyl radical scavenging function primarily in circulating lipoproteins, lipids and biological membranes, where it colocalizes with polyunsaturated fatty acids (PUFA), especially docosahexaenoic acid (DHA) [11] in cell membrane PUFA-rich lipid domains [12].

3 αT influences membrane fluidity [12] and promotes membrane repair in vitro and in vivo [13]. αT contributes to membrane fluidity via protection of oxidizable lipid and stabilizes membrane domains to assist cell signaling cascades dependent on membrane protein- protein interactions and ion permeability [14]. αT has been implicated in the regulation of gene expression related to lipid metabolism, cell signaling and cell cycle progression [14-16], however, this role is generally attributed to secondary effects due to αT inhibiting LPO propagation [17]. In summary, αT likely regulates membrane-associated gene expression patterns by protection of membrane structure.

Regulation of Vitamin E Trafficking by TTPA

αT is a required dietary nutrient and must be consumed by animals to prevent its deficiency. In the intestine, VitE is absorbed into enterocytes along with dietary lipids by micellar uptake in the intestinal epithelium, incorporated into chylomicrons synthesized by the enterocyte and secreted into circulation. During the first pass through the circulation, chylomicrons distribute some VitE to peripheral tissues during endothelial triglyceride hydrolysis by lipoprotein lipase (LPL). The remaining lipids and VitE travel through circulation to be taken up by the liver as chylomicron remnants. Upon remnant endocytosis, mediated by apolipoprotein E association with remnants and low-density lipoprotein receptor (LDL-R) and LDL-R related protein (LRP) on hepatocyte membranes, the early endosome containing chylomicron particles merge with lysosomes forming the recycling endosome [18, 19].

Cytosolic αT transfer protein (TTPA) associates with the endosomal membrane. The association of TTPA with endosomal recycling [20] has been shown using NPC1 (Niemann-

Pick disease, type C1) knockout mice. The neurons from the cerebellum of these NPC1-/- mice, despite expressing TTPA, have impaired VitE trafficking and accumulate αT instead of secreting it supporting TTPA association with the late endosome [20]. The function of TTPA

4 in αT trafficking is further reinforced by evidence that TTPA localizes to perinuclear vesicles harboring recycling endosome markers including CD71, transferrin and Rab8 [18]. In in vitro studies of VitE trafficking, upon VitE treatment of cultured hepatocytes expressing TTPA, the

αT-TTPA complex moves to the plasma membrane, where αT is released and its secretion is facilitated by ABC-family transporters (e.g. ATP binding cassette subfamily A member 1

[ABCA1]) to extrahepatic lipoprotein acceptors [18].

TTPA preferentially binds RRR and 2R forms of αT [21], transports them to the membrane and secretes αT via ABCA1 to lipoproteins to be delivered to peripheral tissue

[22, 23]. αT is also exchanged spontaneously between membranes and lipoproteins, as well as between circulating lipoproteins [24]. The phospholipid transfer protein (PLTP) also facilitates this exchange mechanism for αT between lipoproteins and the peripheral tissues

[25]. Indeed, all mechanisms for delivery of lipids and lipoproteins to cells serve to non- specifically deliver VitE to tissues [26]. αT is the most biologically active and distributed form of VitE because of this TTPA preference for αT. In addition to TTPA function, the catabolism of non-αT VitE forms serves to potentiate tissue αT enrichment. Excess αT, and especially tocotrienols, β-, γ, and δ-tocopherols, in the liver are catabolized by the cytochrome P450 isoform CYP4F2 into water-soluble catabolic products that are excreted in the urine [27, 28].

TTPA, initially characterized as a tocopherol binding protein by Catignani et al. (1975), remains the chief regulator of αT status [29, 30]. TTPA, expressed primarily in the liver, is also found in very low concentrations in many other tissues including the lung, spleen, kidney, brain, adrenals, rodent uterus and placenta [31-37]. Human placental TTPA expression is second in abundance to that in the liver [38]. Its expression is necessary for proper placental development in mice [31] and is found localized to the site of implantation [32]. TTPA is expressed in both the term and first trimester human placenta [39]. Importantly, prior to

5 implantation TTPA is expressed in human embryo yolk sac [40]. Ttpa translation is also essential for zebrafish embryogenesis [41]. Zebrafish ttpa expression is localized to the most dorsal and anterior regions of the head and tailbud between 12- and 24 hours post-fertilization

(hpf) [41]. This observation was a key piece of information for my studies described in

Chapters 2 and 5.

Vitamin E deficiency

In the United States, recommended VitE intakes are 15 mg αT per day for men and women between 14 and 65 years of age [42]. Adequate circulating αT concentrations in humans are above 12 µmol/L. Less than 12 µmol/L causes fragile red blood cells in human subjects undergoing VitE-depletion studies [43, 44]. Less than 8 µmol/L is associated with neurological abnormalities.

VitE deficiency in humans, as a result of inadequate dietary αT intake is quite rare with deficiency most often attributed to genetic defects of lipoprotein metabolism (e.g. abetalipoproteinemia [45]) or TTPA [37, 46, 47]. TTPA is responsible for circulating αT levels because mutations in the TTPA gene resulting in defective TTPA cause extremely low serum

αT levels (< 1 µmol/L) due to rapid decreases in circulating αT. Deleterious mutations in

TTPA result in ataxia with vitamin E deficiency (AVED), a neurological disorder characterized by peripheral neuropathy and spinocerebellar ataxia. AVED, an autosomal recessive disorder, is a condition described by extraordinarily low plasma VitE concentrations [48].

There are at least 25 known mutations to TTPA that result in varying degrees of severity of

AVED with more severe mutations truncating or preventing protein function [46]. AVED symptoms include low circulating and tissue αT, progressive ataxia, loss of motor control and lesions in muscle and peripheral nerves [48-51]. This phenotype is shared by other VitE deficient animal models, including chickens that develop ataxia and encephalomalacia [52,

6 53], rats and mice that experience localized degeneration of cerebellar Purkinje neurons,

Bergmann glial cells [54] and spinal cord neurons [55], and Rhesus monkeys that experience neuroaxonal degeneration [56]. VitE supplementation in humans can prevent or halt progression of the disorder [46, 57].

Inadequate VitE intake in humans is also linked to increased Alzheimer’s disease

(AD) incidence and progression [58, 59], and a recent meta-analysis showed that low serum

VitE is positively associated with AD prevalence [60]. Similarly, risk of sporadic amyotrophic lateral sclerosis (ALS), a progressive nervous system disease characterized by oxidative stress, was decreased with increasing duration of VitE supplement usage [61, 62]. Cortex and hippocampal amyloid content was reduced in Tg2576 mice provided VitE enriched diets following repeated brain injury, suggesting a protective effect against oxidative damage, especially non-enzymatic, free-radical induced LPO [63]. VitE supplementation also improved cellular redox systems, including superoxide dismutase (SOD) and sirtuin 2 (Sir2), following brain injury protecting against cognitive deficits in rats [64]. TTPA-/- mice had increased cortical neuroprostanes causing brain mitochondrial respiratory dysfunction [65].

Defective TTPA also causes whole-body VitE deficiency in mice, but especially reduces VitE status in the central nervous system, which has lower expression of both TTPA and receptors for HDL [66] relative to other body tissues that would otherwise scavenge VitE from circulating lipoproteins [67]. HDL receptor scavenger receptor class B type 1 (SRB1) is also necessary for normal embryonic development [68] because it delivers VitE to prevent neural tube defects in mice [69]. Thus, there is a strong connection between nervous system health and

VitE sufficiency in the embryo and developed brain. Nonetheless it is not clear how VitE may be critical during embryonic neurodevelopment.

7 Embryonic VitE Requirement

Low maternal VitE (serum and cord) in humans is linked to prematurity, low birth weight [70], increased risk of miscarriage and intrauterine growth restriction [71]. Preterm babies were found with lower VitE relative to term babies of mothers with similar cord blood

VitE levels [72]. Early VitE supplementation reduced the risk of preeclampsia, improved endothelial function, and reduced placental dysfunction [73, 74]. Circulating VitE in women at early and late term was positively correlated to fetal growth and decreased risk for low birth weight [75]. TTPA expression was found in both embryonic trophoblast and fetal capillaries

[38], and importantly the yolk sac [40], indicating a VitE requirement prior to implantation.

Thus, VitE is critical to proper human fetal development.

TTPA-/- mice are unable to carry offspring to term due to fetal resorption and placental failure [31, 76]. Mouse TTPA is expressed in uterine and placental tissues [31], similar to human expression, but is localized specifically to the implantation site [32] and may be required for placentation only [3]. Defective uterine TTPA in the mouse caused neural tube defects in a majority of embryos at E 10.5 [31], the same Carnegie Stage as rats at E 12.

TTPA-/- mice require consistent dietary VitE doses to produce litters, although weanlings from this group were consistently 2- to 4-fold lower enriched for brain αT regardless of maternal

VitE supplementation relative to TTPA expressing mothers [77]. TTPA-/- mice also experience many neurologic consequences, albeit studied primarily in adulthood: 12- to 16-week old mice experienced prefrontal cortex-related stress and learning deficits [78], 6-mo old mice raised on VitE deficient diets experienced foot misplacement neurologic defects with glial cell apoptosis identified by 1 year of age [79], 1-year old mice showed ataxia, degenerating neurons and retinal degeneration [55], and 17-month old mice experienced reduced Purkinje neuron branching in the cerebella resulting in impaired motor coordination and cognitive

8 function [54]. Early age neurologic outcomes because of defective TTPA are not widely reported [79]. Modification of VitE status by targeted mutation of apolipoprotein B (ApoB) [80] and SRB1 [69], however, causes neural tube defects (NTD) in mice. Methods to produce VitE deficient mouse models often result in low reproductive success, reduced offspring viability and increased fetal mortality [77].

Zebrafish as a developmental model

Zebrafish embryos are a frequently used model for developmental biology given they develop externally, are transparent, easy to maintain and are produced in large quantities to allow sensitive analyses [81]. More importantly, early development gene expression networks are highly conserved between the zebrafish and human [82]. For this reason, the zebrafish is a powerful tool to study numerous human-health related conditions, including birth defects

[83-88], oxidative stress, and inflammation [89-92].

Rapid and external development means that the fertilized egg must contain all the necessary nutrients for proper growth. VitE deficiency has up to this point been described as an impediment to placental formation, however, the zebrafish develops independent of their mother, are dependent on nutritive sources previously deposited in the egg by the mother, and yet VitE deficient embryos experience severe morphologic abnormalities. In addition, lecithotrophic animals, those that rely on maternal deposited yolk, are an important model for lipid and lipid-soluble vitamin metabolism. The zebrafish embryo develops separately from its yolk [93, 94], which is bound by the yolk syncytial layer (YSL) that secretes nutrients from the yolk into the developing organism. The YSL expresses many of the genes needed for lipid and lipid-soluble vitamin transport [95] likely facilitating the transfer of VitE from the yolk to the embryo. The YSL even contributes to embryonic patterning and helps regulate early neurogenesis pathways in the zebrafish [96, 97]. Thus, the zebrafish embryo is an excellent

9 model with which to study the molecular basis of VitE deficiency separate from placental outcomes.

Zebrafish Embryonic VitE Requirement

The embryonic VitE requirement is time-dependent and is needed between embryonic days (E) 9.5 and 11.5 in rats [5], a developmentally similar period to that of zebrafish between 12 and 24 hours post-fertilization (hpf) [3, 4]. During this window the zebrafish embryo undergoes significant growth, nearly tripling in length [98]. This period coincides with segmentation of the mesoderm into somites, notochord vacuolation, and primary and secondary neurulation resulting in brain regionalization and neural tube expansion. The Traber lab showed both by (1) using morpholinos to block the embryonic translation of the mRNA for the Ttpa [41] and by (2) evaluating neuronal structures in VitE- deficient embryos (E–) [99, 100] that VitE is essential during the early stages of zebrafish neurogenesis. Specifically, blocking ttpa mRNA translation with an oligonucleotide in zebrafish was 100% lethal by 24 hpf with noticeable impairment of brain and eye development beginning at 12 hpf [41]. A model system of VitE deficiency was also developed using parental dietary VitE restrictions. VitE deficient embryos (E–) experienced increased morphologic abnormalities with nearly 2-fold increase and 4-fold increase in mortality outcomes by 24 and 120 hpf, respectively, relative to VitE-sufficient embryos [99]. By 36 hpf

E– embryos evaluated for gene expression networks by microarray show transcriptional differences relating to energy metabolism, cell cycle progression, and developmental pathways [101]. Between 24- and 120 hpf, E– embryos display initial reduction of TCA cycle and glycolytic intermediates that increase greatly over time relative to E+ embryos [100]. This metabolic shift may be responsible for increasing reducing equivalents needed to detoxify spent cellular antioxidant networks or simply may indicate cellular dysfunction and decline.

Altogether, VitE is required in early zebrafish development to protect animal health likely

10 through regulation of gene expression networks and critical metabolic regenerative pathways. This VitE requirement is separate from placental TTPA expression because the zebrafish embryo survives without need for a placenta; thus, VitE is required in a currently unknown but conserved mechanism by the embryo that requires further investigation.

Prior to liver formation, the yolk and YSL act as the embryonic metabolic hub producing lipoproteins that are trafficked to the developing structures through early circulatory ducts [102]. Ttpa is expressed in the yolk syncytium, indicating a need to export VitE from the yolk [41]. VitE is strongly retained in the teleost yolk during development with 97% of αT still in the yolk of newly hatched Atlantic halibut larvae, and another 30% transferred into the larval body by first feeding [103]. The liver of the Atlantic halibut is not functional until 50% of the initial yolk is depleted [104]. In zebrafish, anterior endodermal progenitor cells that precede the liver are identifiable as early as 22 hpf with liver primordial lobes distinguishable by 48 hpf [105]. Ceruplasmin, a marker of hepatocyte function, is expressed around 32 hpf in zebrafish embryos [106], which suggests that the yolk remains the primary driver of lipid and lipid-related metabolism until 32 hpf when the brain has already regionalized into the fore-, mid-, and hindbrain [107]. Thus, it is critical to investigate the requirement for both VitE and TTPA in the embryo prior to liver formation especially because of the advanced formation of the early brain and nervous system.

Neurogenesis

Vertebrate neurogenesis is a complex and coordinated process that begins at gastrulation and is regulated by a complex set of cell-signaling pathways that stimulate proliferation, differentiation and specification of neuronal cells and eventually tissue formation

[86]. Neurulation differs between vertebrates; primary neurulation whereby the neural plate folds forming a tube in mammals contrasts with zebrafish neural keel formation which

11 consists of a thickening of the neuroectoderm into a rod that subsequently cavitates forming the neural tube [108]. Secondary neurulation is similar between vertebrates and occurs in the caudal embryo by inflation of neural rod, epithelial transition to form a lumen and an expanded neural tube. Regardless, cell types and tissue layers are similarly derived with conserved cellular and molecular developmental mechanisms [109].

Specifically, in zebrafish, the neural progenitor cell populations are induced during gastrulation, as the embryonic shield forms and the neuroectoderm involutes [110, 111]. Early neuroectoderm in zebrafish embryos is marked by the expression of goosecoid (gsc) [112].

Following the formation of the neural plate, neural crest (NC) cells migrate and give rise to neurons of the peripheral and enteric nervous system, craniofacial cartilage and bone, and even pigment cells [113-115]. Neurodevelopmental progression is well-studied with numerous pathways and genes identified as critical to development of the neural keel, neural tube, brain regionalization and more. For example, paired box transcription factor 2a (pax2a) is expressed in the midbrain-hindbrain boundary (mhb), neural keel, optic vesicle and pronephric mesoderm when the animal has developed 1-4 somites [116, 117]. These pax2a expressing structures continue to form the isthmic organizer, neural tube, otic placode and pronephric ducts up to 24 hpf [118-120]. SRY-box transcription factor 10 (sox10) expression begins during gastrulation and is primarily localized to cranial and migratory NC between 12 and 24 hpf [113, 121]. Sox10 is a member of the SoxE family of genes, a group of transcription factors required for NC survival and migration [122]. In addition to their function in NC, SoxE genes are responsible for regulation of collagen involved in cartilage formation, including the critical notochord-specific gene, col2a1a [123].

NC migration and differentiation is highly dependent on nutrient status [124, 125].

Both pax2a and sox10 expression patterns are irregular in oxidative stress models [126-128]

12 and errors in neurogenesis are marked by transcriptional changes of the Pax [129-131] and the SoxE gene family [131-134].

Neural tube defects are often initiated by unifying environmental factors such as oxidative stress, nutrient insufficiencies, and energy dysregulation [69, 131, 135-137]. Neural tube closure and defects are primarily initiated early during organism axis development, gastrulation, and neural crest cell migration.

Hypothesis & Objectives

The goal of this work is to investigate the mechanisms by which VitE prevents vertebrate embryo morbidity and mortality. Specifically, is VitE necessary to preserve neurodevelopmental pathways in the zebrafish embryo, a conserved model of vertebrate neurogenesis? My central hypothesis is that VitE is central to protection of cellular redox status and maintenance of metabolic signal integration, which coordinates gene expression networks, especially in the developing nervous system. To investigate this hypothesis, I developed three aims: 0) evaluate morphology of VitE deficient embryos, 1) Evaluate VitE deficiency impacts on gene expression networks coordinating zebrafish development. 2)

Evaluate VitE deficiency induced thiol and methyl donor status. 3) Develop a model system to study the role of TTPA in early zebrafish embryogenesis.

13 CHAPTER TWO:

Vitamin E is necessary for zebrafish nervous system development

Brian Head, Jane La Du, Robyn L. Tanguay, Chrissa Kioussi, Maret G. Traber

Scientific Reports 2020 September 21: 15028 doi: 10.1038/s41598-020-71760-x

14 ABSTRACT

Vitamin E (VitE) deficiency results in embryonic lethality. Knockdown of the gene ttpa encoding for the VitE regulatory protein [α-tocopherol transfer protein (α-TTP)] in zebrafish embryos causes death within 24 hours post-fertilization (hpf). To test the hypothesis that VitE, not just α-TTP, is necessary for nervous system development, adult 5D strain zebrafish, fed either VitE sufficient (E+) or deficient (E–) diets, were spawned to obtain E+ and E– embryos, which were subjected to RNA in situ hybridization and RT-qPCR. Ttpa was expressed ubiquitously in embryos up to 12 hpf. Early gastrulation (6 hpf) assessed by goosecoid expression was unaffected by VitE status. By 24 hpf, embryos expressed ttpa in brain ventricle borders, which showed abnormal closure in E– embryos. They also displayed disrupted patterns of paired box 2a (pax2a) and SRY-box transcription factor 10 (sox10) expression in the midbrain-hindbrain boundary, spinal cord and dorsal root ganglia. In E– embryos, the collagen sheath notochord markers (col2a1a and col9a2) appeared bent.

Severe developmental errors in E– embryos were characterized by improper nervous system patterning of the usually carefully programmed transcriptional signals. Histological analysis also showed developmental defects in the formation of the fore-, mid- and hindbrain and somites of E– embryos at 24 hpf. Ttpa expression profile was not altered by the VitE status demonstrating that VitE itself, and not ttpa, is required for development of the brain and peripheral nervous system in this vertebrate embryo model.

INTRODUCTION

Vitamin E (VitE) is necessary during embryo development and prevents fetal resorption in VitE-deficient rats [1], We have previously shown that VitE deficiency dysregulates whole animal phospholipid metabolism, energy status and antioxidant systems using VitE deficient zebrafish (Danio rerio) embryos [100, 138]. VitE deficient vertebrates

15 exhibit neurodevelopmental defects [139] including exencephaly [80], increased dorsal root ganglia (DRG) turnover [140] and defective blood brain barriers [141]. By contrast, VitE supplementation confers protective effects against oxidative stress-induced neural tube defects (NTD) in mice [142] and in human diabetic embryopathy [143 ].

NTDs are a broad class of embryologic defects that occur during development; human embryos experience NTD onset between 22-30 days post-fertilization (dpf) [144], rats between 9 and 12 dpf [4], and at the embryological period similar to zebrafish embryos less than 24 hours post-fertilization (hpf) [98]. By this developmental stage, the gene ttpa encoding for the α-tocopherol transfer protein (α-TTP) is expressed in the zebrafish embryo brain, eye and tailbud [41], which suggests that these developing tissues require VitE and/or α-TTP.

We have previously shown that a morpholino ttpa knockout at the one cell stage impairs brain and eye formation in zebrafish embryos [41]. Thus, either α-TTP, its ligand VitE or both are required proper for nervous system development.

The accepted α-TTP function is that it is involved in hepatic VitE trafficking, where it facilitates α-tocopherol secretion into the plasma [145]. In humans, Ttpa genetic defects cause ataxia with VitE deficiency (AVED). This disorder results in VitE-deficiency in early childhood and a sensory neuropathy caused by dying degenerative peripheral nerves [37,

146]. Studies in mice have shown that VitE deficiency causes a localized degeneration of cerebellar Purkinje neurons and Bergmann glial cells [54] and spinal cord neurons [55]. VitE supplementation in humans can prevent or halt progression of the disorder [46]. Thus, there is a strong connection between nervous system health and VitE sufficiency, nonetheless it is not clear how VitE may be critical during embryonic neurodevelopment.

We posit that there are several steps where VitE may be essential during neurodevelopment. Early brain development is a coordinated process that begins as early as

16 gastrulation and is regulated by a complex set of cell signaling pathways that stimulate proliferation, differentiation and specification of neuronal cells and eventually tissue formation. Of note, neurulation differs between vertebrates; primary neurulation whereby the neural plate folds forming a tube in mammals contrasts with zebrafish neural keel formation which consists of a thickening of the neuroectoderm into a rod that subsequently cavitates forming the neural tube [108]. Secondary neurulation is similar between vertebrates and occurs in the caudal embryo by inflation of neural rod, epithelial transition to form a lumen and an expanded neural tube. Regardless, cell types and tissue layers are similarly derived with conserved cellular and molecular developmental mechanisms [109].

Following the formation of the neural plate, neural crest (NC) cells migrate and give rise to neurons of the peripheral and enteric nervous system, craniofacial cartilage, and bone, and even pigment cells [113-115]. Neurodevelopmental progression is well-studied with numerous pathways and genes identified as critical to development of the neural keel, neural tube, brain regionalization and more. For example, paired box transcription factor 2a (pax2a) is expressed in the midbrain-hindbrain boundary (mhb), neural keel, optic vesicle and pronephric mesoderm when the animal has developed 1-4 somites [116, 117]. These pax2a expressing structures continue to form the isthmic organizer, neural tube, otic placode and pronephric ducts up to 24 hpf [118-120]. SRY-box transcription factor 10 (sox10) expression begins during gastrulation and is primarily localized to cranial and migratory NC between 12 and 24 hpf [113, 121]. Sox10 is a member of the SoxE family of genes, a group of transcription factors required for NC survival and migration [122]. In addition to NC, SoxE genes are responsible for regulation of collagen involved in cartilage formation, including the

17 critical notochord-specific gene, col2a1a [123]. NC migration and differentiation is highly dependent on nutrient status [124, 125]. Both pax2a and sox10 expression patterns are irregular in oxidative stress models [126-128] and errors in neurogenesis are marked by transcriptional changes of the Pax [129-131] and SoxE family of genes [131-134].

We hypothesized that VitE deficient zebrafish would experience nervous tissue disruption during embryogenesis. Prior studies from our lab have shown that impairing ttpa translation causes neural and eye tissue deformations, and by 24 hpf results in 100% zebrafish embryo death [41]. These data suggest that either α-TTP itself is necessary, or that its functional role to deliver VitE is essential for neurogenesis. Thus, we have sought to identify VitE-dependent processes by localizing gene expression of key neurogenesis markers in VitE deficient (E–) and sufficient (E+) embryos. To establish the role of VitE in support of neurodevelopment at two specific developmental time points, the relative gene expressions of specific markers in E+ and E– embryos were quantified at 12- and 24 hpf.

Additionally, a comparative histological analysis was performed to assess in E+ and E– embryo developmental defects at 24 hpf.

MATERIALS AND METHODS

Zebrafish husbandry

All experimental protocols and methods were carried out in accordance with the animal use and care protocol (# 5068) approved by the Institutional Animal Care and Use

Committee at Oregon State University. Tropical 5D strain zebrafish were reared in Sinnhuber

Aquatic Research Laboratory at Oregon State University under standard laboratory conditions of 28°C on a 14 h light/10h dark photoperiod according to standard zebrafish breeding protocols12. At 55 days post-fertilization (dpf), adult zebrafish were randomly

18 allocated to two experimental diets, vitamin E deficient (E–) or sufficient (E+), as described

[99, 147, 148].

Vitamin E was extracted from the diets prior to feeding and assayed by HPLC-UV prior to feeding, as described 70. In the E+ diet, α- and γ-tocopherols were 229.8 ± 8.1 and

16.5 ± 0.5 mg/kg ± SEM (n=5 measurements), respectively. In the E– diet, α- and γ- tocopherols were 6.3 ± 0.2 and 2.4 ± 0 mg/kg, respectively. E– and E+ embryos were obtained by group spawning of adult fish fed either E– or E+ diets for a minimum of 80 days up to 9 months. Embryos were collected, staged, and incubated until use in standard embryo media (15 mM NaCl, 0.5 mM KCl, 1 mM MgSO4, 0.15 mM KH2PO4, 0.05 mM Na2HPO4, 1 mM CaCl2, NaHCO3 in fish system water).

Morphological and histological assessment

At each time point (12-, 24-, 48 hpf), embryos are assessed for morbidity and mortality outcomes, as described [149, 150]; at 12 hpf, embryos are assessed for viability and developmental progression if development exceeds 3 hours the expected stage. 24- and 48 hpf embryos are assessed similarly with more attention to embryonic morphological outcomes, including brain, eye, somite, and notochord formation as well as early pericardial and yolk sac edema onset. 24 hpf embryos from both diet groups were euthanized with tricaine (MS222, ethyl 3-aminobenzoate methanesufonate salt, Sigma Aldrich) in accordance with animal care and use guidelines, fixed overnight in 4% paraformaldehyde (PFA) solution, embedded in paraffin blocks, serially sectioned, and stained with Hematoxylin and Eosin for histological assessment (Oregon Veterinary Diagnostic Lab, Corvallis, OR). Live imaging of embryos was conducted using a Keyence BZ-x700 All-in-one microscope with a 2x or 10x objective. Embryos were mounted in 3% methylcellulose at room temperature in 35mm glass bottom MatTek dishes and covered in embryo media and tricaine to anesthetize embryos.

19 Image adjustments, including cropping, brightness and contrast were performed using Adobe

Photoshop.

Whole mount in situ hybridization

The whole mount in situ hybridization protocol with minor modifications was used as described50,73. Briefly, RNA was extracted from pooled E+ or E– embryo lysates. cDNA was synthesized using the High-Capacity SuperScript kit (Applied Biosystems). Primers were designed for a 1 kb portion of the 3’ UTR region of gene targets of interest and contained an

RNA polymerase promoter sequence, either T3 or T7, located at the 5’ end of the reverse primer to make an antisense RNA probe. DNA templates were prepared using PCR amplification followed by synthesis of antisense RNA probes utilizing an in vitro RNA transcription reaction kit with DIG-UTP labeled nucleotides. PCR product and RNA probe quantity and quality were checked at each step by electrophoresis on 1.2% agarose gels and spectrophotometrically with a BioTek3 plate reader. A hybridization buffer was prepared with

50% formamide, 5X sodium saline citrate (SSC), 50 µg/mL heparin, 500 µg/mL tRNA, 0.1%

Tween, and 9 mM citric acid in RNAse–free water.

Embryos used for in situ hybridization techniques were euthanized by tricaine prior to fixation, in accordance with animal care and use guidelines. Embryos at the desired time points were dechorionated by hand using sharp forceps, then fixed overnight in 4% PFA.

Fixed embryos were washed 3 times with sterile phosphate buffered saline (PBS) and stored in methanol at -20°C. For analysis, embryos were rehydrated in PBS, permeabilized with proteinase K (10 µg/mL), fixed with 4% PFA and hybridized with antisense probe in the hybridization buffer with 10% dextran sulfate. The following day, embryos were washed PBS with 0.1% Tween and incubated with 1:5000 anti-DIG-Alkaline Phosphatase Label (Roche) in blocking buffer [2% sheep serum (vol/vol) and bovine serum albumin (2 mg/mL) in PBS

20 with 0.1% Tween] on a gentle rocker at 4°C overnight. Embryos were then washed with Fast

Red Buffer [filter (0.2 µm) sterilized, Tris HCl (36 mM) and NaCl (2 M) in sterile water at pH

8.2]. Finally, color development was observed with Fast Red (Sigma Aldrich) to detect alkaline phosphatase-digoxigenin tagged probes and generate an insoluble red precipitate and rhodamine fluorophore. Fixed embryo imaging was conducted using a Keyence BZ-x700

All-in-one microscope with a 10x or 20x objective. Embryos were mounted in 3% methylcellulose at room temperature in 35mm glass bottom MatTek dishes. All images from the same time point and probe were taken with the same exposure in the same optical plane and processed with the BZ-x Analyzer Software (Keyence). Additional image adjustments, including cropping, brightness and contrast applied were performed using Adobe Photoshop.

Relative reverse transcription PCR

RNA was extracted from embryos staged developmentally at 12 and 24 hpf, pooled

(n=10 embryos/sample) and homogenized; 4 sample pools per VitE status. RNA integrity

(9.02 ± 0.15, n=16) was assessed using the 2100 BioAnalyzer Instrument (Agilent) at the

Center for Genome Research and Biocomputing at Oregon State University. Reverse transcription to synthesize cDNA was performed with the High-Capacity Superscript Kit

(Applied Biosystems). Primers were designed for 100-200 bp amplicons in the 3’ UTR region of the target genes (Table 2). PCR was performed with final concentrations 1X Sso Advanced

Universal SYBR Green Supermix (Bio-Rad), 500 nM forward and reverse primers and ~20 ng template.

Statistical analyses

All statistical analyses were evaluated using Prism (GraphPad Software Inc., CA).

Viability outcomes assessed linear regression with comparison of slopes. Gene expression data were normalized using 18S ribosomal subunit expression and analyzed with the 2-∆∆Ct

21 method, as described.74 Differences between groups, evaluated by T-test were determined to be statistically significant if P < 0.05.

RESULTS

Morphological abnormalities associated with VitE deficiency

The E– embryos experienced severe developmental morbidity and mortality outcomes, including deformation of the brain and eye and ill-defined somites by 12 hpf

(Figure 1A-B). By 24 hpf, surviving E– embryos displayed ill-defined somites and stunted fin formation (Figure 1C-D). By 48 hpf, surviving E– embryos experienced severe pericardial and yolk sac edema, as well as disruption to tail development (Figure 1E-F). Overall, mortality was greater (P=0.031) in the E– relative to E+ embryos over the first 48 hpf (Figure 1G).

Similar to our previous report [100], E– compared with E+ embryos experienced a greater incidence of severe brain and somite malformations and increased developmental delay

(Figure 1H). Malformations of the brain and eye were unique to E– embryos at 24 hpf.

Notably, the E+ mortality was not significantly different from the mortality rate of embryos from parents fed a standard lab diet [151].

Ttp localization

We reported previously ttpa is necessary for zebrafish embryogenesis and its expression is localized to the most dorsal and anterior regions of the head and tailbud [41].

Because ttpa knockouts were lethal by 24 hpf, we hypothesized that ttpa expression is critical in regions requiring VitE delivery. To test this hypothesis, we compared ttpa expression in E+ and E– embryos. Surprisingly, at 6 hpf ttpa was ubiquitously expressed throughout the gastrulating animal regardless of VitE status (Figure 2A-B). By 12 hpf, both E+ and E– embryos show similar ttpa expression patterns in the both the animal and yolk syncytial layer

(Figure 2C-D). By 24 hpf, ttpa expression was detected in the developing ventricles of the brain (Figure 2E-F). VitE status did not appear to impact ttpa expression or its localization.

22 However, VitE deficiency was associated with impaired brain development indicated by deformed mhb (arrow) and an uninflated third ventricle (*). Thus, VitE status did not regulate ttpa signal but was required for normal brain regionalization and structure by pharyngula stage prim-5 at 24 hpf when pigment formation also begins and the first heartbeat can be detected in zebrafish embryo [98].

Nervous system development markers

Early neurogenesis is defined by a specific set of transcriptional regulators that when present up- or down-regulate gene expression to cause proliferation and differentiation of neuronal tissue. Goosecoid (gsc), for example, is a nervous system patterning marker that indicates proper gastrulation and is expressed in the involuting neuroectoderm of zebrafish embryos at 6 hpf [112, 152]. Gsc mRNA expressions at 6 hpf in E+ or E– embryos were not apparently different (Figure 3) suggesting that VitE might not be critical for gastrulation at 6 hpf, a time point equivalent to mouse embryonic day 6 and human embryonic day 12 [153].

Because early embryonic axis patterning was not impacted by inadequate VitE at 6 hpf, subsequent times (12 and 24 hpf) were investigated. Pax2a was used to evaluate the midbrain-hindbrain boundary (mhb) formation. At 12 hpf, E+ embryos have clearly developed mhb and otic placodes (op) (Figure 4A,C). By contrast, at 12 hpf in E– embryos, the pax2a signal was ill-defined and diffused, suggesting impaired formation of the mhb (Figure 4C-D).

Specifically, the width of the mhb in E– is 1.5 times greater than the mhb in E+ embryos.

Moreover, the op margins in the E– were further apart than in the E+ embryos. These abnormalities may signal developmental delays, defects, or both. By 24 hpf, pax2a is normally found in the anterior mhb, hind brain and spinal cord neurons.[154] In both E+ and

E– embryos pax2a expression was prominent in the optic stalk (os) region of the forebrain

(Figure 4E–F). E– embryos had irregular pax2a expression characterized by a shortened distance between os and mhb. The pax2a expression pattern in the otic vesicle did not differ

23 between E+ and E– embryos. By contrast, pax2a expression in spinal cord neurons (sn) was more abundant with structures more clearly defined in the E+ compared with the E– embryos

(Figure 4E–F). Overall, pax2a expression was most severely impacted in the mhb of E– 12 hpf embryos and in the spinal cord neurons of the E– 24 hpf embryos.

Sox10, another key transcription factor that plays a role in cell fate specification [121], was evaluated at 12- and 24 hpf in E+ and E– embryos. The E+ embryos exhibited sox10 expression pattern similar to the established patterns defined during zebrafish development

[154]. E– relative to E+ embryos, however, demonstrate a restricted distribution of sox10 cell lineages in NC (Figure 5A-B). By 24 hpf, sox10 was abundantly expressed in E+ animals, in migratory NC, ov and down the spinal cord (sc) (Figure 5C). Evidence of NC migration was observed by the increased sox10 expression in the trunk of the animal into the dorsal root ganglia and early enteric neurons (not shown). E– embryos have reduced developmental progress relative to E+ based on the visibly different expression of sox10-expressing cells along the growing trunk with diffuse ov margins and reduced expression in NC. These findings were similar to the abnormal sn expression of pax2a in 24 hpf E– embryos, discussed above.

To determine the extent to which VitE status was associated with downstream neurogenesis targets, we used markers to localize notochord formation. The notochord is a conserved structural element and signal source point for neural tube and eventual spinal cord development [155, 156], where collagen synthesis, driven by col2a1a and col9a2 expression, is critical. In the 12 hpf embryos, notochord formation did not vary by VitE status (Figure 6A-

B). By 24 hpf, however, both E+ and E– embryos showed notochord abnormalities with severe notochord bending more prevalent in E– embryos (Figure 6C-J). The wavy notochord phenotype was present in some embryos from parents fed a standard lab diet (not shown)

24 and may be an artifact of the fixative process. Similarly, we note that the notochord was more severely curved along the yolk axis in E– embryos and is not likely an artifact of fixation.

Quantitation of gene expression

To determine the extent to which VitE status altered the relative abundance of the same gene targets which had been localized, qPCR was performed. We hypothesized that transcriptional regulation would be altered and would be reflected in the patterning disruptions. Using relative fold change ratios (2-∆∆Ct), gene expressions at 12- or 24 hpf were evaluated in E– compared with E+ embryos (Figure 7). Gene expressions of sox10, pax2a, col2a1a and col9a2 were not significantly different at 12 hpf. At 24 hpf, only pax2a gene expression was lower (P<0.05) in E– compared with E+ embryos. Thus, Pax2a expression was the only gene changed of those tested, which suggests that the mhb requires VitE for normal development.

Histological analysis

Sectioned E+ and E– embryos at 24 hpf were hematoxylin and eosin stained to assess the structural effects downstream of the dysregulated gene express patterns (Figure

8). E+ embryo (Figure 8A) forebrains opened into a teardrop shape with eyes to each side, while E– embryos (Figure 8B) exhibited an elongated ventricle with an improperly inflated lumen. In E+ embryos (Figure 8C), the diencephalic ventricle at the midbrain opened into a diamond shape with distinct hinge points, while in E– embryos (Figure 8D) the ventricles had somewhat distorted diamond shaped openings. The extent of hindbrain inflation was similar in E+ and E– embryos; however, roughly 72% were developed in the E+ (Figure 8E) while only 41% were developed in the E– embryos (Figure 8F). Somitic cells in E+ embryos (Figure

8G) formed distinct, V-shaped epithelial boundaries, while in E– embryos somitic cells were

25 loosely packed and formed U-shapes (Figure 8H). The vacuolated notochord cells appeared normal regardless of VitE status (Figure 8I, 8J).

DISCUSSION

Regardless of VitE status, ttpa is expressed throughout the developing zebrafish embryo at gastrulation as the neuroectoderm involutes forming the embryonic shield.

Subsequently, ttpa was localized in the rostral head region of the zebrafish embryo at 24 hpf

(Figure 2), as we previously reported [41]. Specifically, ttpa was expressed at 24 hpf in the midbrain-hindbrain region and ventricle borders suggesting the critical need for the α-TTP in these regions of the early central nervous system (CNS). Because the known hepatic α-TTP role is to deliver VitE from recycling endosomes to the cellular membrane for local cellular redistribution [18], our findings suggest that α-TTP in these brain regions is necessary to deliver VitE to newly formed neurons and/or differentiating surrounding neuronal cells. In adults, most neurons do not express ttpa, but Purkinje cells and associated Bergmann glia cells do [54]. Purkinje neurons and Bergmann glia cells are found in the cerebellum, a brain region critical for motor control and derived from the primordial hindbrain [157]. Shown here, ttpa was expressed at 24 hpf in the zebrafish embryo mhb. Our data suggest ttpa expression in the developing 24 hpf zebrafish embryo may be found in anatomically similar regions to where Purkinje progenitor neurons are found in rodent brain development [158]. We also showed that the lack of VitE did not change the abundance of ttpa expression as measured by qPCR, but its lack disrupted brain ventricle inflation and overall brain structure.

Little is known about VitE trafficking in the brain. Our findings support the hypothesis that VitE is necessary in the hindbrain-midbrain boundary to support formation of the ventricular shape, size, and constriction, which may be a result of two different VitE functions.

First, VitE is a potent lipophilic antioxidant that is a necessary antioxidant to protect the

26 polyunsaturated fatty acid-rich membranes of nervous tissue [17]. Second, VitE enhances membrane fluidity and repair [13]. The developing brain undergoes significant membrane expansion and growth requiring a degree of neuronal plasticity facilitated by membrane composition [159], and our work supports the idea that VitE serves critical roles during this process. The histological analysis further supports that VitE is necessary during somitic cell formation, which is highly dependent on convergence and extension of the lateral mesoderm

[160]. E– embryos display somitic defects consistent with impaired convergence and extension, which may further impact neural tube formation and closure by 24 hpf. Collagen sheath markers col2a1a and col9a2 (Figure 6 D, 6H) also demonstrate that E– embryos had severely bent axes. Similarly, the ttpa morpholino-knockout embryos displayed a shrunken body axis [99]. also indicating dysregulated convergent extension. Patterns of genes associated with convergent extension will be investigated in the future based on the mounting evidence regarding NTDs, body axis development and the VitE-deficient phenotype shown throughout this study.

Histological analysis also validates morphological deformations observed by whole mount in situ hybridization. Both techniques demonstrated that the forebrains in E– embryos had ill-defined boundaries and were overly inflated (Figs. 2F, 8B). Similarly, the E– midbrain ventricles had abnormal morphology with neuroepithelial bending at the hingepoints (Figs.

2F, 8D). Somitogenesis plays a critical role in determining NC cell migration [161]. Errors in somite formation observed in E– embryos (Figure 8H) may explain the reduced signal of sox10 in NC cells migrating ventrally along the zebrafish midline and primordial spinal cord

(Figure 5D).

Because ttpa was expressed from 6- to 24 hpf (Figure 2), VitE must be essential in the same developmental period, especially in the structures that express ttpa. We aimed to determine at what period embryonic development goes awry. To investigate gastrulation,

27 which occurs at nearly 50% epiboly with the formation of the embryonic shield [112, 162], we evaluated gsc expression [162], but found that gsc was not affected by VitE status.

Interestingly, Niemann-Pick C Disease-1 depleted embryos have impaired VitE trafficking and also have normal gsc expression at the same early timepoint [163]. Although modulation of lipid metabolism did not affect embryonic gsc expression [164]; a ventralizing effect at the shield stage was observed with gpx4b knockdown, which increases oxidative damage by impairing lipid hydroperoxide detoxification [165]. E– embryos undergo normal gastrulation, which suggests that the oxidative damage due to VitE deficiency is insufficient to cause very early-stage abnormalities. Thus, VitE appears to be needed at a subsequent time point, such as during the formation of the neural keel, rod folding and neural tube cavitation, which occur between 12- and 24 hpf.

Genetic markers associated with NC proliferation, migration, and the formation of the mhb were also studied. Sox10 expression is found in zebrafish embryo NC and their progenitor cells preceding formation of NC-derived sensory neurons and dorsal root ganglia

[121]. Mislocated sox10 expression patterns were observed at 12- and 24 hpf in E– embryos

(Figure 5). These patterns have been seen in zebrafish embryos with folate deficiency- induced neuropathy, which was associated with irregular sox10 expression at similar developmental periods [131]. Pax2a defines the mhb, or isthmic organizer, a constricted portion of the still open neural tube that coordinates patterning in both the midbrain and cerebellum [116]. Localization of the pax2a gene in the E– embryo at two developmental timepoints shows that the mhb is ill-formed at 12 hpf (Figure 4D) and subsequently structures critical for the peripheral nervous system, including spinal cord neurons, are less abundant

(Figure 4F). These data suggest that oxidative damage may be impairing neurodevelopment.

VitE deficiency causes lipid peroxidation resulting in whole embryo metabolic disruption including increased phospholipid turnover and choline utilization with subsequent methyl

28 donor depletion [100, 166]. Folic acid deficiency similarly depletes methyl donors producing similar pax2a expression pattern errors to those described herein in the E– embryos [131].

We suggest that VitE deficiency may act in a similar manner by depleting methyl donors and thus disrupt brain development because we found pax2a gene expression was also mislocated and quantitatively reduced (Fig 7B). Further study is needed to explore the mechanisms involved.

This study shows that VitE deficiency causes severe developmental impairment at early embryonic stages. VitE does not regulate ttpa expression patterns but does alter the tissue structures where it is found (Figure 2D, 2F). Major pattern disruptions at 12 and 24 hpf, as indicated by pax2a and sox10, may indicate VitE-dependent regions of the developing nervous system. In addition to disrupted brain regionalization and NC migration, E– embryos experience severe morphologic abnormalities indicated by bent axes as indicated by col2a1a and col9a2. Additionally, histological evidence agrees with the dysregulated gene expression patterns and brain morphology in E– embryos, indicating impaired brain ventricle inflation and somite formation. These experiments represent a major step in determining the molecular basis of VitE in the developing vertebrate nervous system.

Figure 1. Morphological abnormalities associated with VitE deficiency at 12-, 24- and 48 hours post-fertilization , Representative bright field images of E+ and E– embryos showed normal development in E+ embryos and abnormalities in the E– embryos. At 12 hpf, E+ embryos (A) had defined somites and eyes, while E– embryos (B) showed abnormal somite formation and a reduced dorsal region where the eye is usually located. At 24 hpf, E+ embryos (C) had clearly defined eyes, heads, and somites, while the E– embryos (D) had pericardial edema, less well-defined somites and notochord. At 48 hpf, E+ embryos (E) had extended fins and pigmentation throughout the body, while the E– embryos (F) experienced severe pericardial edema, stunted fin formation; some E– embryos experience errors in tail formation. (G) Early mortality, defined as nonviable beyond that time point, was increased in E– embryos at 12 hpf. By 48 hpf only about 35% of the original E– clutch survived with about 75% of E+ embryos surviving (P<0.031) At 24 hpf (H), E– embryos that were alive experienced greater incidences of developmental delay, as well as morphological malformations that include brain, eye, and somites deformities. * indicate delay or defects in E– embryos relative to E+. Scale bar represents 500 m. Representative embryos are shown. Figure panels A-F were generated with the BZ-x700 microscope, processed with BZ-X Analyzer Software with image adjustments made 𝜇𝜇in Adobe Photoshop. This figure was created with Adobe Photoshop, v21.2.1.

Figure 2. Ttpa signal localized throughout early embryo and brain ventricle borders regardless of VitE status , Ttpa expression in E+ and E– embryos is indicated with red fluorescence; dorsal direction is indicated by arrow (A-D). At 6 hpf (dorsal shield stage, A-B), Ttpa expression was present throughout the animal poles [E+ embryos, n=5/5 (n= number of animals with the observed defect/total number of animals observed)], E– embryos, n=6/6). At 12 hpf (90% epiboly, C-D), ttpa expression was present both in the embryo and in the yolk syncytial layer (E+ embryos, n=6/6; E– embryos, n=7/8 ), arrow indicates anterior region of the embryo. At 24 hpf (E-F), Ttpa expression was localized in the brain ventricle borders and within cells of the fore (f), mid- (m), and hindbrain (h). Arrows indicate the midbrain-hindbrain boundary where diencephalic ventricle expansion was altered; * represent inflation in E+ embryos (E, n=6/6) or lack thereof in E– embryos (F, n=3/7). Scale bar represents 500 m (A-D) and 50 m (E-F); representative embryos are shown. Figure panels were generated with the BZ-x700 microscope with, processed with BZ-X Analyzer Software with ima𝜇𝜇ge adjustments equally𝜇𝜇 applied across time points in Adobe Photoshop. This figure was created with Adobe Photoshop, v21.2.1.

Figure 3. Gastrulation marker goosecoid (gsc) is not affected by VitE deficiency , Gsc expression at 6 hpf in the neuroectoderm of the dorsal embryonic shield, as indicated by red fluorescence in (A) E+ embryos (n=9/9) and (B) E– embryos (n=8/8). Arrow indicates dorsal region of the embryo. Scale bar represents 500 µm; representative embryos are shown. Figure panels were generated with the BZ-x700 microscope, processed with BZ-X Analyzer Software with image adjustments equally applied across time points in Adobe Photoshop. This figure was created with Adobe Photoshop, v21.2.1.

Figure 4. Midbrain-hindbrain boundary formation shown by pax2a expression is dysregulated by VitE deficiency , Pax2a expression in early optic stalk (os), midbrain-hindbrain boundary (mhb) and otic placode (op) in 12 hpf embryos, lateral views (A-B); arrow indicates dorsal region. At 12 hpf, E+ embryos (C) had defined mhb and op borders (n=8/8), while E– embryos (D) had diffuse mhb and op borders (n=8/12 assessed). Shown are representative E+ embryo with mhb 25 µm wide and op 49 µm apart, while representative E– embryo measurements were mhb 42 µm wide and op 63 µm apart. At 24 hpf, in E+ (E) and E– embryos (F) pax2a was expressed in the os, mhb, otic vesicles (ov) and spinal cord neurons (sn). Distance between os and mhb, a measure of first brain ventricle inflation, were greater in a representative E+ (91 µm) relative to an E– (80 µm) embryo. E+ embryo (E) spinal cord neurons at the same fluorescence exposure had significantly increased pax2a signal (n=9/9), as compared with E– embryos (n=6/9). Scale bar represents 100 µm; representative embryos are shown. Figure panels were generated with the BZ-x700 microscope, processed with BZ- X Analyzer Software with image adjustments equally applied across time points in Adobe Photoshop. This figure was created with Adobe Photoshop, v21.2.1.

Figure 5. Neural crest cell migration impaired during development by VitE deficiency , Sox10 was expressed in the cells of the neural border at 12 hpf (A-B); arrows indicate dorsal region of the embryo. The extent of the distribution of neural crest cells (nc) expressing sox10 was greater (n=5/6) in E+ embryos (A) relative to E– embryos (n=4/8) (B). Measurements in representative embryos were E+ 174 µm relative to E– 145 µm. NC migrate ventrally away from the spinal cord (sc) to differentiate into cranial, cardiac, enteric, and sensory neurons and glia. At 24 hpf, sox10 expression around the otic vesicle (ov) was clearly defined in E+ embryos (C) and less well-defined in E– embryos (D). In E+ embryos (C), the migratory NC that flank the developing sc show increased sox10 expression (n=10/12), while E– embryos (D) had fewer migratory NC with sox10 expression (n=6/12). Nonetheless, similar spinal cord (s) sox10 expression was observed in E+ and E– embryos. Scale bar represents 100 m; representative embryos are shown. Figure panels were generated with the BZ-x700 microscope, processed with BZ-X Analyzer Software with image adjustments equally applied𝜇𝜇 across time points in Adobe Photoshop. This figure was created with Adobe Photoshop, v21.2.1.

Figure 6. Notochord collagen markers col2a1a and col9a2 affected by dietary treatment , At 12 hpf, col2a1a expression was localized in the developing notochord of E+ (A, n=5/5) and E– (B, n=6/6) embryos. Arrow indicates dorsal region of the embryo. At 24 hpf, the notochord sheath in E+ and E– embryos showed col2a1a (C, D) and col9a2 (G, H) expression with a wavy notochord phenotype observed in both groups. The wavy notochord was more apparent when viewing dorsally (E-F, I-J) relative to lateral view (C-D, G-H). At 24 hpf, ~25% of E+ embryos (E, n=4/12) showed a wavy col2a1a expression, while ~66% of E– embryos (F, n=6/9) were obviously wavy; similarly, a wavy col9a2 expression was observed in 33% of E+ embryos (I, n=3/10), while ~60% of E– embryos had a wavy expression (J, n=5/8). Scale bar represents 500 m (C-J) and 100 m (A-B); representative embryos are shown. Figure panels were generated with the BZ-x700 microscope, processed with BZ-X Analyzer Software with image𝜇𝜇 adjustments equally𝜇𝜇 applied across time points in Adobe Photoshop. This figure was created with Adobe Photoshop, v21.2.1.

Figure 7. Relative gene expression of neurogenesis markers , Log2 fold change of genes measured in E– relative to E+ by RT-qPCR. Ttpa expression was not measured at 12 hpf. Data are as the median fold change, with the box from the 25th to the 75th percentile, the whiskers show the minimum and maximum value. * indicates a significant difference in gene expression between E– and E+ Ct values. This figured was created with Prism software.

Figure 8. Histological analysis of 24 hpf zebrafish embryos with morphological defects associated with VitE status , Hematoxylin and eosin staining of E+ and E– embryos at 24 hpf was used to evaluate morphological defects, including fore- (F) and mid- (M), and hind- (H) brain ventricle inflation, somite formation and notochord vacuolation. Transverse section of E+ embryos had tear-drop shaped-F with eyes to each side (A), whereas E– embryos had a neural lumen, but improperly inflated-F (B). Serial transverse sections of E+ embryos showed normal M ventricle inflation with distinct hinge-points, indicated by * (C). E– embryos (D) had reduced M ventricle inflation with similar hinge points. E+ and E– embryos had similar H inflation leading to the neural tube in the trunk (E-F). Sagittal sections of the trunk indicate defined somite epithelial boundaries in E+ embryos (G) and indistinct epithelial boundaries in E– somitic cells (H). Notochord vacuolation appeared normal in both E+ (I) and E– (J) embryos. Scale bar represents 50 m; representative embryos are shown. Figure panels were generated with the BZ-x700 microscope, processed with BZ-X Analyzer Software with image adjustments equally applied across𝜇𝜇 time points in Adobe Photoshop. This figure was created with Adobe Photoshop, v21.2.1.

37 CHAPTER THREE:

Vitamin E deficiency disrupts gene expression networks during zebrafish development

Brian Head, Stephen A. Ramsey, Chrissa Kioussi, Robyn L. Tanguay, Maret G. Traber

Nutrients 2021 January 30: 13(2), 468 doi: 10.3390/nu13020468

38 ABSTRACT

Vitamin E (VitE) is essential for vertebrate embryogenesis, but the mechanisms involved remain unknown. To study embryonic development, we fed zebrafish adults (>55 days) either VitE sufficient (E+) or deficient (E–) diets for >80 days, then the fish were spawned to generate E+ and E– embryos. To evaluate the transcriptional basis of the metabolic and phenotypic outcomes, E+ and E– embryos at 12, 18 and 24 h post-fertilization

(hpf) were subjected to gene expression profiling by RNASeq. Hierarchical clustering, over- representation analyses and gene set enrichment analyses were performed with differentially expressed genes. E– embryos experienced overall disruption to gene expression associated with gene transcription, carbohydrate, and energy metabolism, intracellular signaling and the formation of embryonic structures. mTOR was apparently a major controller of these changes. Thus, embryonic VitE deficiency results in genetic and transcriptional dysregulation as early as 12 hpf, leading to metabolic dysfunction and ultimately lethal outcomes.

INTRODUCTION

Vitamin E (VitE) is a potent lipophilic antioxidant and is localized in membranes to protect against lipid peroxidative damage [17]. VitE contributes to membrane fluidity via protection of oxidizable lipid and stabilizes membrane domains to assist cell signaling cascades dependent on membrane protein–protein interactions and ion permeability [14].

VitE must be provided in the diet because it is produced only by photosynthetic organisms, where it accumulates in plant seeds and early embryonic structures [6]. Although VitE was discovered as a dietary component necessary to prevent fetal resorption in rats [1], the molecular basis for this requirement still remains under investigation.

The vertebrate embryonic VitE requirement is time-dependent and is needed between embryonic days (E) 9.5 and 11.5 in rats [5], a developmentally similar period to that of zebrafish between 12 and 24 h post-fertilization (hpf) [3, 4]. During this window, the zebrafish

39 embryo undergoes significant growth, nearly tripling in length [98]. This period coincides with segmentation of the mesoderm into somites, notochord vacuolation, and primary and secondary neurulation resulting in brain regionalization and neural tube expansion. We showed both by (1) using morpholinos to block the embryonic translation of the mRNA for the α-tocopherol transfer protein (α-TTP) [41] and by (2) evaluating neuronal structures in

VitE-deficient embryos (E–) [99, 100, 167] that VitE is essential during the early stages of zebrafish neurogenesis. Specifically, blocking TTP mRNA translation with an oligonucleotide in zebrafish was 100% lethal by 24 hpf with noticeable impairment of brain and eye development beginning at 12 hpf [41]. VitE deficiency caused by defective lipoprotein metabolism also produces neurodevelopmental defects in mice, including neural tube defects

[69] and exencephaly [80]. Additional neurologic impairments have been reported in Ttpa-/- mice, which show degeneration of cerebellar Purkinje [54] and spinal cord neurons [55].

In addition to morphologic derangements, E– zebrafish embryos also experience lethal dysregulation of glycolytic metabolism in the first 120 hpf [100], which could be remedied by glucose supplementation at 24 hpf. Notably, E– embryos experience a hypermetabolic state at 24 hpf, which switches to a hypometabolic state by 48 hpf.

Importantly, carbohydrate metabolism is necessary to provide precursor molecules for the rapid growth of zebrafish embryos especially between 0 and 48 hpf [168]. In addition, E– embryos between 24 and 48 hpf experience errors in amino acid metabolism and generation of precursor ribonucleotides, tricarboxylic acid (TCA) cycle intermediates and demonstrate over-production of free saturated fatty acids [100, 138]. E– embryos by 12 hpf experience increased betaine concentrations, suggesting that this methyl donor is needed for the methionine cycle to maintain S-adenosyl methionine production [169]. These data suggest that epigenetic regulation may be impacted by VitE deficiency. Lee et al. showed that blocking

40 folic acid metabolism and thus preventing methyl donor recycling causes also neurologic defects in zebrafish embryos as early as 10 hpf [88].

Zebrafish development is coordinated by temporal and spatial gene expression networks. Transcriptional regulation is a key component of embryonic fate [170]. The mechanistic target of the rapamycin (mTOR) complex senses energy status, amino acid abundance and redox imbalance to regulate cell survival and proliferation [171, 172].

Additionally, in zebrafish, mTOR signaling is necessary in myelination pathways throughout development of the central and peripheral nervous system [173]. VitE deficiency may disrupt metabolic pathways regulated by and restored by mTOR signal integration. Metabolic dysfunction is likely the cause and consequence of transcriptional network alterations that regulate metabolic needs for rapid growth and development [174]. Thus, the VitE deficient state may have marked effects on temporal gene expression patterns in the zebrafish embryo specifically during neural tube formation and brain development. The objective of this study was to identify transcriptional targets of VitE deficiency in embryo development longitudinally.

Due to early disruption of the metabolic state in E– embryos [100, 138], we examined changes prior to 24 hpf. We hypothesize that VitE is necessary to protect zebrafish transcriptional networks associated with metabolism, cell signaling and embryo energy status. In addition, we sought to identify transcription factors necessary for zebrafish embryo nervous system development that are VitE-dependent and produce the morphological defects previously observed [167, 169].

MATERIALS AND METHODS

Zebrafish husbandry

All experimental protocols and methods were carried out in accordance with the animal use and care protocol (# 5068) approved by the Institutional Animal Care and Use

Committee at Oregon State University. Tropical 5D strain zebrafish were reared in the

41 Sinnhuber Aquatic Research Laboratory at Oregon State University under standard laboratory conditions of 28 °C on a 14 h light/10 h dark photoperiod according to standard zebrafish breeding protocols [98]. At 55 days post-fertilization (dpf), adult zebrafish were randomly allocated to two experimental diets, vitamin E deficient (E–) or sufficient (E+), as described [99].

VitE was extracted from the diets prior to feeding and assayed by HPLC-UV prior to feeding, as described [175]. In the E+ diet, α- and γ-tocopherols were 361 ± 10 and 2.6 ± 0.0 mg/kg ± SEM (n = 3 measurements), respectively. In the E– diet, α- and γ-tocopherols were

1.1 ± 0.0 and 0.5 ± 0.0 mg/kg (n = 3 measurements), respectively. E– and E+ embryos were obtained by group spawning of adult fish fed either E– or E+ diets for a minimum of 80 days up to 9 months. Embryos were collected, staged and incubated until use in standard embryo media (15 mM NaCl, 0.5 mM KCl, 1 mM MgSO4, 0.15 mM KH2PO4, 0.05 mM Na2HPO4, 1 mM CaCl2, NaHCO3 in fish system water, (EM)). Embryos at each time point (12, 18, 24 hpf) were staged according to the appropriate developmental landmarks [98]; embryos at 12 hpf with the presence of 6 somites, 18 hpf with the presence of 18 somites and defined otic vesicle, and at 24 hpf (prim-5 stage) upon first pigment formation in the retinal epithelium and presence of 30 somites. To evaluate transcriptional differences based solely on VitE status, only embryos identified as morphologically normal and at the appropriate developmental landmarks were used for analysis. Embryos were euthanized with tricaine (MS222, ethyl 3- aminobenzoate methanesulfonate salt, Sigma Aldrich) in accordance to animal care and use guidelines.

RNA extraction and sequencing

RNA was extracted from embryos staged developmentally at 12, 18 and 24 hpf, pooled (n = 10 embryos/sample) and homogenized, with 4 sample pools per VitE status. RNA

42 integrity (9.02 ± 0.15, n = 24) was assessed using the 2100 BioAnalyzer Instrument (Agilent) at the Center for Genome Research and Biocomputing at Oregon State University

(Appendix). RNA libraries were prepared with the Lexogen QuantSeq 3′mRNA Seq Library

Prep Kit-FWD kits (Lexogen, Vienna, Austria). After library preparation, samples were pooled and sequenced using single-end sequencing with 100 bp reads on an Illumina HiSeq3000 instrument (Illumina, San Diego, CA, USA).

Data deposition

RNASeq data for this study has been deposited in Gene Expression Omnibus with accession number GSE164848 and can be viewed at: https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE164848.

Data processing and statistical analysis

Data processing was performed using the default parameters provided by the program manuals unless stated otherwise. Sequence read quality was first assessed using

FastQC v0.11.5. Adapters and poly(A) tails were trimmed using bbduk v35.92. Trimmed reads were then aligned to the current reference genome (GRCz11, D. rerio) with current annotations (Ensembl 100) using STAR aligner, v2.7.3a. Raw gene counts were summarized with edgeR, Bioconductor v3.24.3 in R, v4.0.2. Data were normalized by the TMMwsp

(Trimmed Mean of M-values with singleton pairing) method.

Multidimensional scaling plots (MDS) were computed to assess the similarity of group effects, identified as the interaction of diet and time, and the sequencing lane effects between samples. edgeR generalized linear model (glm) was used to fit the data and estimate the common, trended and tagwise dispersions given the experimental design. A quasi-likelihood

F-test (QLF) was used to compute differential gene expression between E+ and E– embryo gene expression at each time point measured (n = 3 time points; 12, 18, 24 hpf). Heatmaps were generated using z-score transformed counts per million (CPM) of differentially

43 expressed (DE) genes identified by adjusted p-value, or Benjamini–Hochberg False

Discovery Rate (FDR) < 0.1.

Gene ontology and network enrichment

DE genes were subject to Gene Ontology (GO) enrichment analysis using the

Bioconductor R package GOseq, v.1.40.0. Over representation analysis (ORA) was performed with DE genes and DE genes with absolute value log2FC ≥ 1 found at all time points using WebGestalt [176]. Gene set enrichment (GSEA) was performed with pre-ranked genes computed with log2FC and p-value using WebGestalt. Differentially expressed transcription factors (TFs) were identified and described by predicted anatomical regions using ZFIN [177]. All network enrichment was considered significant if FDR < 0.25.

Metabolomic data from 24 hpf embryos previously acquired [100] were integrated with 24 hpf transcriptomic profiles using MetaboAnalyst [178].

Western blotting

Protein was extracted from pooled (n = 30) E+ and E– embryos at 24 hpf in RIPA buffer supplemented with protease and protein phosphatase inhibitors (Calbiochem, La Jolla,

CA). Rat liver extract was used as a positive control for the assay. Extracted protein (25 μg) was subjected to SDS-Page and processed by immunoblotting. All gels and blots were run simultaneously. The proteins were visualized by SuperSignalTM West Pico Chemiluminescent

Substrate (ThermoFisher, Carlsbad, CA) and quantified using the Bio-Rad Image System

(Hercules, CA, USA). Phospho-protein (p-) to unphosphorylated protein ratio is calculated for the representative lanes shown.

RESULTS

Morphological deformities including yolk sac and pericardial edemas, bent axes, and apparent developmental delays generally occur after 24 hpf in E– embryos as previously reported [167]. To evaluate gene expression differences based on VitE status alone and not

44 phenotypic changes, only embryos identified as morphologically normal were used. All appropriate developmental landmarks were used for the following analyses. Thus, the transcriptional profiles reported herein are powerful determinants of the underlying effects of

VitE deficiency because they are not changes induced by physical deformities that induce further stress on the animal.

Hierarchical clustering and gene annotation

To determine the VitE-dependent effects on global transcriptomic changes during embryo development, RNA was isolated from E+ and E– whole embryo lysates (n = 4 pools,

10 embryos/pool) at the three time points, 12, 18 and 24 hpf. Multidimensional scaling analysis indicated strong separation between the gene expression profiles by time point and further by embryo VitE status (Supplementary Figure S1). Sequencing lane effects were similarly analyzed but with little to no effect on gene expression profiles, thus are disregarded from further analysis. Of the 22,796 sequenced genes, 2355 were identified as different (p <

0.05) across all time points and by VitE status. From this selected group of genes, 12% (n =

285) were considered differentially expressed (DE) under a false discovery rate threshold

(FDR < 0.1). A heatmap of DE genes was clustered with z-score averages according to VitE status, age and gene (Figure 9). Gene clusters from the heatmap were annotated for gene ontology (GO) terms associated with biological processes (BP), molecular function (MF) and cellular component (CC). Cluster 1, identifiable by reduced gene expression in E– embryos at 12 hpf, was significantly (FDR < 0.25) associated with organic acid binding (GO:0043177), iron ion binding (GO:0005506), and carbohydrate binding (GO:0030246). Vitamin binding

(FDR = 0.06, GO:0019842) refers to genes associated with L-ascorbic acid binding capacity, including egln2, p4ha1b, and plod1a. Cluster 2 was significantly associated with peptidase activity (GO:0008233) and unfolded protein binding (GO:0051082). Cluster 3, identifiable by increased gene expression in E+ embryos at 12 and 18 hpf, was significantly associated with

45 protein-containing complex binding (GO:0044877), cell adhesion molecule binding

(GO:0050839) and cytoskeletal protein binding (GO:0008092). Cluster 4 genes were defined by GO terms vacuole (GO:0005773) and extracellular region part (GO:0044421). The term

“vacuole” was associated with endolysosomal trafficking-associated genes abcc6a, galca, and zgc:110239.

Overall, in E– relative to E+ embryos, there were 110 DE genes with increased expression levels (Figure 10A) and 80 DE genes with decreased expression levels (Figure

10B). DE genes that were observed consistently with increased expression in E– embryos were associated significantly with the following KEGG (Kyoto Encyclopedia of Genes and

Genomes) pathways—glycolysis/gluconeogenesis, carbon metabolism, biosynthesis of amino acids, pentose phosphate pathway and fructose and mannose pathways (FDR < 0.25,

Figure 10C). DE genes that were observed consistently with decreased expression in E– embryos were associated significantly with pyrimidine metabolism and proteasome pathways

(Figure 10D).

Top differentially expressed genes in E– embryos

The top 10 DE genes (e.g., with the largest log-fold changes) with increased or decreased expression levels in E– embryos relative to E+ embryos were identified at each of the three time points (12, 18 or 24 hpf; Table 1). DE genes that increased at all three time points in E– embryos include olfce2 (olfactory receptor C family, e2), which is predicted to be involved in G protein-coupled receptor signaling, and cngb3.2 (cyclic nucleotide gated channel subunit beta 3, tandem duplicate 2), which is predicted to have cGMP binding activity and is involved in cation transmembrane transport. DE genes that decreased at all three time points in E– embryos include serpina7 (serpin peptidase inhibitor, clade A, member 7), a serine-type endopeptidase inhibitor implicated in mTOR-related neuropathology in zebrafish

[179] and numerous unnamed genes si:dkey-73p2.2, si:dkey-15j16.6, and si:dkey-23k10.3.

46 nitr3r.1l (novel immune-type receptor 3, related 1-like) was found initially (12 and 18 hpf) to be increased highly in E– embryos, but was decreased highly at 24 hpf.

To evaluate the gene product properties of highly DE genes identified, over- representation analysis was performed using DE genes with absolute value log2FC ≥ 1, or 2- fold differences between E– and E+ gene expression values. Top GO terms (FDR < 0.05) are shown in ascending p-value order (Figure 12). DE genes in E– embryos annotated for

Biological Process (BP) indicated highly related metabolic processes, including the generation of precursor metabolites and energy and carbohydrate metabolic processes

(Figure 12A). The Molecular Function (MF) domain annotations included oxidoreductase activity, glutamate and peptide receptor activity and binding related to cell cycle progression

(Figure 12B). Annotations associated with Cellular Component (CC) included further description of energy generation by the respiratory chain complex and other membrane protein complexes (Figure 12C).

Genes (n = 22,796) were ranked by their fold-change in expression level between E– and E+ sample groups. Using the ranked genes, Gene Set Enrichment Analysis (GSEA) was carried out using WebGestalt (Figure 13). All GSEA enrichments shown are significant (FDR

< 0.25). At 12 hpf, E– vs E+ embryos have increased expression of genes associated with vitamin C binding, demethylation and demethylase activity, metabolic processes and decreased expression of genes associated with collagen synthesis and protein import (Figure

13A). At 18 hpf, E– vs E+ embryos have increased expression of genes associated with vitamin C binding and decreased expression of genes associated with oxidoreductase activity, gene expression regulation, endothelium development and membrane protein transport activity (Figure 13B). At 24 hpf, E– vs E+ embryos have increased expression genes associated with neuron ensheathment activity and decreased expression of genes associated with gene transcription regulation (Figure 13C).

47 Metabolomics data, previously acquired [100], were integrated with the gene expression profiles reported herein (both data sets were from 24 hpf E+ and E– embryos) and submitted to MetaboAnalyst. VitE status significantly (FDR = 2.39 × 10−28) altered the mTOR signaling pathway (Figure 14A), as shown with metabolites (circles) and genes

(squares), which were either increased (red) or decreased (green) in E– relative to E+ embryos. This finding was also validated by Western blot analysis of protein extracts obtained from 24 hpf E– and E+ embryos using β-actin to quantify protein levels and the phosphorylated-to-unphosphorylated protein ratio calculated (Figure 14B). Raptor, a key component of mTOR complex 1 (mTORC1) was decreased in E– embryos. The phosphorylated protein (p-) abundance indicates active or inactive state relative to the unmodified protein, depending on the protein. In E– relative to E+ embryos, at 24 hpf p-Rps6 kinase (p-Rps6k) was decreased more than two-fold, p-eukaryotic translation initiation factor

2A (p-Eif2a) was decreased two-fold, and p-eukaryotic translation initiation factor 4E-binding protein 1 (p-Eif4ebp1) was increased 1.5-fold. These outcomes all support the finding that mTOR signaling was dysregulated at 24 hpf in E– embryos in comparison to E+ embryos.

Over-representation analysis was used for transcription factor (TF) enrichment to determine the extent to which gene expression trends are associated with similar TF regulatory units or promoter regions. DE genes shared the consensus site for TF binding

YGTCCTTGR motif (n = 12 genes, FDR = 0.029). Top DE TFs were identified from the dataset (Table 2). The most DE TF, nr2f1b, is expressed in the midbrain hindbrain boundary region and somites of the zebrafish embryo during segmentation [180]. Other highly DE TFs include mafaa, expressed in the neural tube and myotomes during segmentation [181], and zgc:101562, orthologous to human ZCAN10, ZNF398 and ZNF777. Due to VitE’s role in prevention of propagation of lipid peroxidation, other genes of interest including ttpa, ttpal, gpx4a, gpx4b and others were evaluated but were not found to be differentially expressed.

48 DISCUSSION

The outcomes of this study show that VitE deficiency disrupts numerous gene expression networks, including energy metabolism, oxidoreductase activity, intra- and intercellular signaling, and developmental transcriptional regulation, during critical developmental windows in zebrafish embryos. We previously identified that VitE deficiency prevents proper neural crest cell migration and impairs midbrain–hindbrain development

[167].

E– zebrafish embryos at 24 hpf experience metabolic dysfunction with reduced glycolytic intermediates and increased basal oxygen consumption rates [100]. These metabolic changes are echoed by the gene expression profiles of E– embryos between 12 to 24 hpf reported herein. Similarly, rat embryos between E10.5 and E12.5, developmentally comparable to zebrafish embryos at 24 hpf [4, 98], exhibit increased expression levels of genes involved in glycolytic pathways, presumably to increase production of reducing equivalents and precursor molecules via the pentose phosphate pathway. [182]. We found that the greatest changes in gene expression in E– embryos were metabolic processes that generate NADPH, ATP and other precursor compounds through glycolysis, the TCA cycle, and oxidative phosphorylation gene expression networks. Tixier et al. showed that glycolytic pathway induction is required to fuel the rapid growth of zebrafish embryos between 0 and

48 hpf [168]. Glycolysis-related gene expression in E– embryos at 12, 18, and 24 hpf is further supported by increased expression of transcription factor tead1b (Table 2, FDR = 0.024).

Glycolytic genes are necessary for anabolic growth and key morphogenesis pathways in embryonic development, which we previously reported are significantly altered in E– embryos

[167, 169]. E– relative to E+ embryos at 36 hpf had similar misexpression of carbohydrate metabolism related genes in embryos in our previous microarray experiments [101]. We reported that these same TCA cycle and glycolytic gene products were altered using a

49 proteomic profile in adult zebrafish fed VitE-deficient diets [183]. Vitamin E-deficient rat livers were analyzed by NMR and were found also to have decreased glucose levels [184]. Thus,

VitE deficiency appears to impair production of energy.

Metabolic pathways are also critical for cellular compensation in response to lipid peroxidation (LPO). We reported previously the depletion of long chain polyunsaturated fatty acids, specifically docosahexaenoic acid (22:6n-3), in the E– embryos [138] and E– adult zebrafish [147]. Additionally, an LPO biomarker, malondialdehyde (MDA) was increased in

E– adult zebrafish [185]. LPO requires detoxification by oxidation of glutathione (GSH). To replenish GSH requires synthesis using the limiting amino acid, cysteine. We have found in

E– embryos that choline is depleted, betaine is increased [169] and the methyl donor status

[169], along with the thiol status, is disrupted. In addition, oxidized glutathione (GSSG), which requires enzymatic reduction using NAD(P)H, is increased [169]. Thus, the E– embryo, over time, is unable to compensate for the increasing LPO.

Zgc:101562 and mafaa, two differentially expressed transcription factors in E– embryos, are associated with gene expression related to glucose regulation and are expressed in mouse neural tissue at E12.5 [177, 186, 187] and the zebrafish neural tube during the segmentation period (between 10 and 24 hpf) [28,38,39], respectively. Together, key TFs dysregulated by VitE deficiency, associated with metabolism and localized to early neurologic structures of the vertebrate embryo, highlight VitE’s necessary role in maintaining growth pathways and preventing neurodevelopmental defects we reported previously [167].

Importantly, we showed that glucose injections at 24 hpf rescued ~70% of E– embryos [100], providing a clue that earlier intervention may alter transcriptional responses and fully rescue

E– animals through remediation of glycolysis. Importantly, vertebrate neurogenesis formation requires mTOR signaling [188, 189] and mTOR is expressed in the zebrafish trunk and brain at 24 hpf [190].

50 Overall, genes associated with metabolism and mTOR signal integration, used to regenerate the metabolic intermediates described, appear to be the mechanism for the compensatory responses to VitE deficiency. mTOR senses energy and amino acid status signals to stimulate cell survival, growth and proliferation [171]. mTOR is reported also to sense redox status through modulation of glutathione status in the mouse cerebellum [172].

The alteration in mTORC1 at 24 hpf reported herein show that VitE deficiency has a major impact on this key regulator. Additional studies are needed to evaluate the longitudinal changes, along with quantitation of the key mediators, which are likely to accompany the depleted glucose status, oxidized lipids and other cellular stressors.

mTOR contributes to ensheathment in the zebrafish embryo [173] and requires cholesterol for myelination, cytoskeletal support and membrane-specific signaling pathways

[191]. We report herein that E– embryo transcripts are enriched for cytoskeleton filament, anchored membrane components and experience metabolic disruption sensed by mTOR.

Our data indicate that mTOR signaling is indeed disturbed in E– embryos by 24 hpf with a 3- fold reduction of phosphorylated Rps6k and 2-fold reduction of phosphorylated Eif2a relative to E+ embryos. In addition, expression of genes associated with ensheathment of neurons was positively enriched in E– embryos at 24 hpf. Previously, we showed that a VitE deficiency impacts neural crest cell gene expression patterns and somite integrity as early as 12 hpf.

Myelination and ensheathment pathways immediately follow the migration of sensory axons away from the central nervous system (CNS) towards the periphery [192], a process dependent on neural crest cells [193, 194]. Hypothetically, the errors at 12 hpf may precede the effects observed in 24 hpf E– embryos. Sensory neurons are derived from Sox10- expressing neural crest cells [195], which we previously reported to be reduced in number and mis-localized in E– embryos at 12 and 24 hpf [167]. Thus, we suggest that VitE deficiency

51 may disrupt mTOR signaling cascades and subsequent ensheathment pathways that begin during axonal migration from the CNS at 24 hpf.

Many other pathways were impacted by VitE deficiency, of note is collagen remodeling. egln3, expressed in the neural plate in 12 hpf zebrafish embryos [181], is a hypoxia inducible factor (HIF) target and regulates the expression of collagen remodeling genes p4ha1b, plod1a and col1a1b [196, 197]. VitE radicals produced as a result of lipid peroxidation reactions are chemically reduced by ascorbic acid (Vitamin C (VitC)), thereby depleting this critical antioxidant micronutrient [198]. Zebrafish, like humans, likely do not synthesize VitC and require it in their diet [199, 200]. VitC was increasingly depleted in E– relative to E+ embryos up to 120 hpf [100]. We observed herein that vitamin binding, annotated primarily with VitC-associated gene expression, was positively enriched in E– embryos at 12 and 18 hpf, indicating VitC’s important early role in development [201-203].

VitC is required for collagen prolyl hydroxylation, catalyzed by the p4ha family of genes, including p4ha1b found in this study [204]. Together, a VitE deficiency and induced VitC deficiency can dramatically alter collagen remodeling pathways in zebrafish embryo vascularization, skin growth and even axonal projections of the brain [205]. Negative enrichment of genes associated with collagen trimer at 12 hpf and endothelium development at 18 hpf in E– embryos further highlights VitE’s role in maintaining structural support during embryonic development of the notochord and somites [206].

In addition, there were numerous genes differentially expressed at 12, 18 and 24 hpf in E– embryos that might suggest multiple developmental pathways disturbed, including— nr2f1b, a transcription factor decreased in E– embryos at 12 hpf (log2FC = −2.21, FDR =

0.007) that regulates vascularization and hindbrain regionalization in zebrafish [207]; nitr3r.1l, an immune receptor significantly increased in E– embryos at 12 and 18 hpf (log2FC = 6.14 and 2.06, respectively) but decreased at 24 hpf (log2FC = −2.31) [208]; and cngb3.2, a cation

52 transmembrane transporter increased in E– embryos at 18 and 24 hpf (log2FC = 2.94 and

1.59, respectively) implicated in the pathology of retinitis pigmentosa [209], a symptom of

VitE deficiency in humans [210]. We also compared our data to that of other VitE deficiency models. Unfortunately, comparisons between species including the mouse, rat, and horse, between tissue types, and by age of development are limited. Thus, we found little to no overlap between our dataset and that of others published [78, 79, 211, 212]. It is most likely that the zebrafish embryo prior to 24 hpf simply does not share similar gene expression profiles of highly differentiated tissues such as the cerebellum, spinal cord or liver.

Myelination pathways, for example, are not highly expressed in early development and thus are not differentially expressed in the E– embryo.

In summary, we show that VitE deficiency as early as 12 hpf disrupts pathways underlying growth and development. This study also points to major metabolic dysfunctions occurring as early as 12 hpf. Simultaneously, E– embryos experience LPO disruption to membrane structure and signaling pathways, both of which are known secondary effects due to VitE deficiency. Errors occurring as early as 12 hpf signal catastrophic decline and impairment to neurodevelopment that relies on early signaling capacity for cellular migration, proliferation and overall tissue and organ development. There remains a need to further investigate the origins of VitE deficiency-induced embryonic death as a means of preventing developmental defects and improving prenatal health span.

Figure 9. Hierarchical clustering and gene annotation of differentially expressed genes in E– and E+ embryos at 12, 18 and 24 hpf , Heatmap of all differentially expressed (DE) genes (n = 286, FDR < 0.1) clustered with z-score averages according to VitE status, age and gene with red color indicating greater expression relative to all-condition average and blue color indicating lower expression relative to all-condition average amongst all expression values (rows correspond to genes, columns correspond to conditions). Clusters within heatmap annotated by Gene Ontology (GO) terms organized by Benjamini–Hochberg False Discovery Rate (FDR).

Figure 10. Venn diagrams and KEGG pathways enriched by over representation analysis of differentially expressed genes changed consistently over time , (A–B) Venn diagrams separating DE genes consistently (B) increased or (C) decreased in E– embryos relative to E+ embryos at each age. (C–D) KEGG pathways associated with DE genes (FDR < 0.1) consistently (C) increased or decreased (D) in E– embryos vs. E+ embryos at each time point (12, 18, and 24 hpf). * indicates significant FDR < 0.25.

55 Table 1. Top differentially expressed genes in E– embryos relative to E+ at 12, 18 and 24 hpf , Top 10 DE genes by absolute value log2FC at each time point with log2FC in E– embryos relative to E+. ↑ and ↓ indicate a DE gene found in the top 10 consistently increased or decreased in E– embryos at each age (12, 18, 24 hpf).

12hpf 18hpf 24hpf Log2FC in Log2FC in Log2FC in Gene Symbol Gene Symbol Gene Symbol E– E– E– nitr3r.1l ↑↑↓ 6.129 cngb3.2 ↑↑ 2.944 olfce2 ↑↑↑ 5.822 si:dkey-90m5.4 3.135 olfce2 ↑↑↑ 2.472 si:dkey-23k10.3 ↓↑ 4.389 si:dkey-238d18.15 tnnc1a 3.048 ighd ↑↑ 2.326 3.404 ↑↑ olfce2 ↑↑↑ 3.023 anpepa ↑↑ 2.129 polr3c 2.768 cd40lg 2.924 nitr3r.1l ↑↑↓ 2.068 topaz1 2.242 si:dkey-238d18.15 slc25a55b 2.496 2.003 anpepa ↑↑ 1.815 ↑↑ ribc1 2.386 rnps1 1.801 si:dkey-238d18.5 1.813 ighd ↑↑ 2.327 pvalb4 1.714 cngb3.2 ↑↑ 1.722 si:dkey-159f12.2 2.225 snx21 1.662 dntt 1.516 scdb 2.145 si:dkey-90m5.4 1.635 f2rl1.2 1.505

b3gat2 −4.178 serpina7 ↓↓↓ −4.396 nitr3r.1l ↑↑↓ −2.692 serpina7 ↓↓↓ −2.993 si:dkey-15j16.6 ↓↓ −2.998 si:ch211-191i18.2 −2.657 loxhd1b ↓↓ −2.787 si:dkey-73p2.2 ↓↓↓ −2.901 si:dkey-15j16.6 ↓↓ −2.648 si:dkey-73p2.2 −2.673 zgc:158445 −2.839 serpina7 ↓↓↓ −2.629 ↓↓↓ asb13a.1 −2.666 si:ch1073-13h15.3 −2.680 si:dkey-73p2.2 ↓↓↓ −2.231 rfesd −2.532 si:dkey-23k10.3 ↓↑ −2.460 zgc:173585 −1.617 aqp8b −2.518 prkra −2.174 loxhd1b ↓↓ −1.575 nr2f1b −2.205 grin1b −1.573 zgc:101562 −1.453 proca −2.069 kiaa1549lb −1.570 insl5a −1.206 si:ch1073-268j14.1 −2.041 ormdl3 −1.417 dock11 −1.125

Figure 11. Gene ontology terms enriched by over representation analysis and gene set enrichment analysis of all differentially expressed genes , GO annotations associated with most highly DE genes (n = 286, FDR < 0.1, |log2(FC)| ≥ 1) enriched across all ages (12, 18 and 24 hpf) and Vitamin E (VitE) status by over-representation analysis. Terms are separated by (A) Biological Process, (B) Molecular Function and (C) Cellular Component. * indicates significant FDR < 0.25 with all enrichments shown considered significant.

Figure 12. Gene ontology terms enriched by gene set enrichment analysis of all expressed genes , GO annotations associated with all genes (n = 22,796) pre-ranked, analyzed by time (12, 18, 24 hpf) and organized by normalized enrichment score. Top 10 significant (FDR < 0.25) GO terms are separated by expression at (A) 12 hpf, (B) 18 hpf and (C) 24 hpf. Approximate number of genes found enriched in each term indicated by dot size. Positive enrichment score indicates increased expression in E– relative to E+ embryos, negative enrichment score indicates decreased expression in E– relative to E+ embryos. Normalized enrichment score automatically computed with enrichment and adjusted p-value by WebGestalt.

Figure 13. Mechanistic target of rapamycin (mTOR) signaling pathway disrupted in 24 hpf E– embryos , (A) E+ and E– embryos at 24 hpf metabolomic data acquired previously were integrated with gene expression profiles using MetaboAnalyst to generate network profiles. The mTOR signaling pathway was significantly enriched with both metabolites and genes expressed in 24 hpf E– embryos. Red boxes or circles represent increased, while green boxes represent reduced metabolite or gene expression, respectively, in E– relative to E+ embryos. (B) mTOR complex 1 (mTORC1)-associated proteins including Raptor, Rps6k, Eif2a and Eif4ebp1 were evaluated by activity determined by phosphorylation status in pooled protein extracts of rat liver (positive control), E+ and E– embryos (n = 30 embryos/pool) at 24 hpf. Phospho-protein (p-) to unphosphorylated protein ratio is calculated for the representative lanes shown.

59 Table 2. Top transcription factors differentially expressed in E– embryos across the developmental window measured , Transcription factors (TFs) identified within DEG list annotated for TF family, anatomical region of expression at the measured time point and log2FC in E– embryos relative to E+.

Log2FC in E– vs. E+ Symbol Name FDR 12 hpf 18 hpf 24 hpf

Nuclear receptor subfamily 2, group F, nr2f1b −2.21 0.17 0.12 0.007 member 1b

v-maf avian musculoaponeurotic mafaa −1.83 −0.22 0.13 0.008 fibrosarcoma oncogene homolog Aa

hsf1 Heat shock transcription factor 1 0.68 0.55 −0.08 0.013 zgc:101 zgc:101562 −0.94 −1.11 −1.45 0.020 562

tead1b TEA domain family member 1b 0.46 0.26 0.15 0.024

mbd3b Methyl-CpG binding domain protein 3b −0.32 −0.35 −0.23 0.024

TOX high mobility group box family tox2 −0.48 −0.41 −0.08 0.027 member 2 hmgb2b High mobility group box 2b −0.14 0.49 0.04 0.047 bhlhe40 Basic helix-loop-helix family, member e40 0.86 0.17 0.12 0.061

tbx18 T-box transcription factor −0.81 −0.33 −0.42 0.066

CHAPTER FOUR:

Vitamin E deficiency dysregulates thiols, amino acids and related molecules during zebrafish embryogenesis

Jie Zhang1, Brian Head1, Scott W. Leonard, Jaewoo Choi, Robyn L. Tanguay, Maret G.

Traber.

Redox Biology 2021 January: (38) 101784 doi: 10.1016/j.redox.2020.101784 1Equal contribution.

61 ABSTRACT

Vitamin E (α-tocopherol, VitE) was discovered as a nutrient essential to protect fetuses, but its molecular role in embryogenesis remains undefined. We hypothesize that the increased lipid peroxidation due to VitE deficiency drives a complex mechanism of overlapping biochemical pathways needed to maintain glutathione (GSH) homeostasis that is dependent on betaine and methyl group donation. We assess amino acids and thiol changes that occur during embryogenesis [12, 24 and 48 hours post fertilization (hpf)] in VitE- sufficient (E+) and deficient (E–) embryos using two separate, novel protocols to quantitate changes using UPLC-MS/MS. Using partial least squares discriminant analysis, we found that betaine is a critical feature separating embryos by VitE status and is higher in E– embryos at all time points. Other important features include; glutamic acid, increased in E– embryos at 12 hpf; choline, decreased in E– embryos at 24 hpf; GSH, decreased in E– embryos at 48 hpf. By 48 hpf, GSH was significantly lower in E– embryos (P < 0.01), as were both S- adenosylmethionine (SAM, P < 0.05) and S-adenosylhomocysteine (SAH, P < 0.05), while glutamic acid was increased (P < 0.01). Since GSH synthesis requires cysteine (which was unchanged), these data suggest that both the conversion of homocysteine and the uptake of

– cystine via the Xc exchanger are dysregulated. Our data clearly demonstrates the highly inter-related dependence of methyl donors (choline, betaine, SAM) and the methionine cycle for maintenance of thiol homeostasis in E– embryos. Additional quantitative flux studies are needed to clarify the quantitative importance of these routes.

INTRODUCTION

Vitamin E (α-tocopherol, VitE) was discovered because it is a necessary nutrient to maintain pregnancy in rats [1]. Importantly, VitE sufficiency is closely associated with the developing nervous system in poultry [213], rodents [214], and zebrafish [41]. In particular, zebrafish embryos are a frequently used model for developmental biology given they develop

62 externally, are transparent, easy to maintain and are produced in large quantities to allow sensitive analyses [81]. The fertilized egg contains all of the necessary nutrients and is a self- contained unit that develops into a swimming fish in 5 days. Therefore, we developed a unique model of VitE deficiency using zebrafish embryos obtained by spawning adult fish fed

VitE deficient (E–) diets.

The use of E– embryos has allowed study of the interactions of interdependent, overlapping metabolic systems during embryonic development [99, 100, 138, 215]. Critically, by 5 days 80% of E– embryos have morphologic abnormalities or have died [100]. Not surprisingly, since VitE is a lipid soluble, peroxyl radical scavenger, E– embryos experience lipid peroxidation, depletion of phosphatidyl choline with docosahexaenoic acid (DHA-PC) and dysregulated phospholipid metabolism [100, 166]. Additionally, we reported using targeted metabolomics analyses over a time course of 5 days that E– embryos had dysregulated energy metabolism [100], as well as mitochondrial dysfunction, measured by extracellular oxygen consumption [100]. Specifically, glucose was depleted and its supplementation could partially rescue the E– embryos [100, 138]. We also found in E– embryos that choline was depleted by 24 hours post-fertilization (hpf) and that over time betaine became dysregulated, along with glutathione (GSH) depletion [100, 138]. Moazzami et al [216], who utilized non-targeted 1H-NMR-metabolomics to investigate long-term effects of VitE deficiency in rats, also found that betaine concentrations were higher and expression of genes related to energy metabolism were lower in deficient rats. Taken together, these data show that VitE is necessary to maintain energy homeostasis, and point to betaine as a critical nutrient that becomes dysregulated as deficiency proceeds.

Dysregulation of betaine concentrations in E– embryos is consequential because betaine is a methyl donor that plays a significant role in zebrafish development. Betaine homocysteine S-methyltransferase (BHMT, EC 2.1.1.5) catalyzes the transfer of methyl

63 groups from betaine to homocysteine (Hcy) to form methionine (Met). Yang et al [217] showed that the bhmt gene is highly expressed at 12, 24 and 72 hpf in zebrafish embryos.

Additionally, cystathionine ß-synthase (CBS, EC 4.2.1.22) catalyzes the first step of the transsulfuration pathway, converting Hcy to cystathionine. Cbs knockdown in zebrafish causes a bent embryonic axis that can be rescued with betaine [218]. Thus, understanding betaine regulation may provide a critical linkage between metabolic pathways and thiol redox status to understand consequences of VitE deficiency.

This linkage between VitE deficiency, increased lipid peroxidation and thiol status has been studied in E– embryos using lipidomics [166] and metabolomics [100]. These latter studies suggested that choline, betaine, and the methionine cycle were dysregulated by inadequate VitE protection during embryogenesis. The dysregulation of these pathways in

E– embryos causes abnormal nervous system formation, especially the midbrain-hindbrain boundary, spinal cord and dorsal root ganglia [167]. Thus, we raise the question “why is a water-soluble methyl donor dysregulated during VitE deficiency?” We hypothesize that an increased requirement for reduced glutathione (GSH) during the increased lipid peroxidation observed in E– embryos drives a complex mechanism of overlapping biochemical pathways needed to maintain thiol homeostasis that is dependent on betaine and methyl group donation. To test this hypothesis, we have developed highly sensitive, analytical methodologies to measure both critical amino acids and thiols longitudinally in E– and E+ zebrafish embryos.

MATERIALS AND METHODS

Materials and Reagents

Supplies were obtained as follows: ammonium formate (Optima, Fisher); formic acid

(Optima LC-MS grade, Fisher Chemical); tris (2-carboxyethyl) phosphine hydrochloride

(TCEP, Sigma-Aldrich); N-ethylmaleimide (NEM; Sigma-Aldrich); ethylenediaminetetraacetic

64 acid (EDTA; EMD Millipore); 5-sulfosalicylic acid (SSA; Sigma-Aldrich); zirconium oxide beads (Next Advance); amino acid standard mix (Sigma-Aldrich); stable isotope-labeled amino acid standard mix (Cambridge Isotope); choline chloride (Fluka); betaine (Sigma-

Aldrich); S-adenosylmethionine (SAM) (Cayman Chemical); S-adenosylhomocysteine (SAH)

13 (Cayman Chemical). The stable isotope-labeled internal standard (IS) GSH-(glycine- C2,

15N) trifluoroacetate salt (Sigma-Aldrich, St Louis, MO) was used for recovery analysis and quantitation of free thiols. All other reagents and solvents were of analytical grade.

Zebrafish husbandry

The Institutional Animal Care and Use Committee (IACUC) of Oregon State University approved the protocol (ACUP Number #5068). Tropical 5D strain zebrafish were reared in the Sinnhuber Aquatic Research Laboratory at Oregon State University under standard laboratory conditions of 28 on a 14-h light/10-h dark photoperiod according to standard zebrafish breeding protocols℃ [98]. At 55 days post-fertilization, adult zebrafish were randomly allocated to one of two experimental diets, VitE deficient (E–) or sufficient (500 mg RRR-α- tocopheryl acetate/kg diet, E+), as previously described [99, 100, 147]. We obtained E– and

E+ embryos by spawning adult 5D zebrafish that had been fed either E– or E+ diets, respectively, for a minimum of 80 days. Control embryos were obtained from adults fed standard zebrafish food (Gemma micro, https://zebrafish.skrettingusa.com/). Diet and embryo α- and γ-tocopherol concentrations (Supplement Table 1) were determined using high-performance liquid chromatography with electrochemical detection [175].

Embryos were collected, staged and incubated in standard embryo media (EM; reverse osmosis water with 15 mM NaCl, 0.5 mM KCl, 1 mM CaCl2, 1 mM MgSO4, 0.15 mM

KH2PO4, 0.05 mM Na2HPO4 and 0.7 mM NaHCO3) [219]. Reproductive output was determined by pair spawning females followed by total egg count per female. For extraction

65 protocols, only living embryos as evaluated by spontaneous embryo movement are used for analysis and these embryos were randomly sampled from those viable at the given sampling time. To reduce variability due to differences in hatching, all embryos used for analysis were unhatched with chorion on to investigate changes unrelated to hatching, which may significantly alter the embryos’ redox potential [220]. For sample preparation, embryos at 12,

24 and 48 hpf were transferred to 1.7 ml microcentrifuge tubes. The tubes were kept on ice for 30 min to euthanize the animals, the liquid removed, and prepared as described below, then the samples were snap frozen in liquid nitrogen and stored at -80°C. Live embryos were used for morphological assessment. At 12, 24 and 48 hpf, embryos were mounted in 3% methylcellulose on a glass slide dish, anesthetized with tricaine and imaged using a Keyence

BZ-x700 All-in-One Microscope. Images were taken with the 2x and 10x objective using brightfield imaging. All image enhancements (contrast, saturation) were made uniformly in

Adobe Photoshop. Egg traits were assessed by egg yolk diameter followed by egg volume estimated based on the volume of an ellipsoid using the equation [221, 222]:

4 = π ( ) ( ) 3 2 𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉 ∗ 𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 ∗ 𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙𝑙 𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 Thiol extraction and determination by UPLC-MS/MS

Free thiols in embryos were extracted, derivatized and measured by UPLC-MS/MS, as adapted from [6-8]. The extracted and quantitated thiols included GSH, glutathione disulfide (GSSG), cysteine, homocysteine, and cystine (cysteine disulfide, CysS). Briefly, embryos (n=10) were homogenized in the derivatizing protein and precipitating solution (20 mM NEM, 50 mM EDTA and 2% SSA in 15% methanol, final concentrations) using a bullet blender (Next Advance, Inc. Troy, NY) with 0.5 mm zirconium oxide beads for 2 minutes at speed 10. Prior to homogenization, an aliquot of the internal standard (IS) [GSH–(glycine-

13 15 C2, N)] was added to each sample for quantitation and to correct for non-specific losses.

66 NEM incubation times between 1 min to 1 hour have been previously reported for a variety of biological samples, including red blood cells and human plasma [223], but we found that this length of time was insufficient for the more complex whole zebrafish embryo homogenate. To determine the optimal length of time necessary for complete thiol derivation with NEM, a time course was performed. The minimum time necessary for maximum derivation was 9 h (see below). The reaction can continue at room temperature protected from light for up to 24 hours after homogenization with the derivation solution without any subsequent increase in oxidation of GSH to GSSG (data not shown). For the data in this report, the embryo homogenate and derivatizing solution mixture were incubated at room temperature for a minimum of 9 h. Following derivatization, samples were centrifuged at room temperature for 5 minutes at 14,000 rpm. Supernatants were transferred to injection vials, then stored at 10° C until UPLC-MS/MS analysis.

Derivatized-thiols and thiol disulfides were separated using a C18 column

[temperature 40° C, Acquity UPLC BEH C18 (2.1 x 100mm x 1.7µm; Waters, Milford, MA)] and a UPLC (Waters Acquity H system) coupled with a Xevo TQD mass spectrometer

(Waters, Milford, MA) equipped with an electrospray ionization source operated in positive mode (ESI+). MS tune conditions were as follows: desolvation gas 800 L/h; capillary 2.00 kV; cone 37 V; ESI+ probe temperature 500° C. MassLynx v.4.1 software (Waters, Milford, MA) was used for instrument control and data acquisition. Analytes were detected and quantified by multiple reaction monitoring (MRM). Chromatographic separation was achieved at a flowrate of 0.375 mL/min using a gradient with mobile phase A (MP A, water with 0.1% formic acid) and mobile phase B (MP B, acetonitrile with 0.1% formic acid), as follows: Initial conditions were 0% MP B increased by slight convex curve to 60% for 1.10 min. After 1.10 min a gradient was initiated for 0.2 min (1.10-1.30 min), and MP B was increased linearly to

67 95%. From 1.30 to 1.67 min MP B was held at 95%. At 2.14 min MP B was increased to

100% and held at 100% until 4.0 min, then solvents were returned to initial conditions. Total run time was 5 min including column equilibration. The MRM-transitions and retention times of the individual derivatized thiols and thiol disulfides are shown in Supplement Table 2.

Standard curves were linear (r2 >0.98) between 10 fmol and 100 pmol injected for all thiols tested.

To determine extraction efficiency, derivatization and recovery of NEM-conjugated free thiols, zebrafish embryo pools (n=5 pools, 10 embryos per pool) were spiked with known

IS concentrations at two points in the high (50-100 pmol injected) and two points in the low

(0.5-1 pmol injected) end of the standard curve. The embryo pools were then homogenized in derivatizing solution and measured, as described. Recovery was calculated:

= 100 [( + ) – ] /

Assessment𝑅𝑅 𝑅𝑅of𝑅𝑅𝑅𝑅 thiol𝑅𝑅𝑅𝑅𝑅𝑅𝑅𝑅 extraction∗ 𝑠𝑠𝑠𝑠𝑠𝑠 and𝑠𝑠𝑠𝑠𝑠𝑠 𝑎𝑎quanti𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎tation𝐼𝐼𝐼𝐼 method𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠𝑠 𝑎𝑎 development𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎𝑎 𝐼𝐼𝐼𝐼

Embryo pools were homogenized in derivatizing solution, then at periodic intervals (1,

2, 3, 6, 9, 12 and 24 h) samples were centrifuged, supernatants were transferred to injection vials and subjected to analysis. Labeled internal standard (IS) added to the standard curve was used to estimate maximum derivatization in the absence of the embryo homogenate, which presumably was caused by matrix effects in those homogenates. We determined that the half-life to the maximum derivatization of homogenates was 3.3 h and was similar for all thiols tested. By 9 h nearly 90% of free thiols in the sample were derivatized with NEM. Using the 9 h incubation protocol, the recovery of added standard to sample homogenates were consistently >75% for all analyses at two points in the high [50-100 pmol injected] and two points in the low [0.5-1 pmol injected] end of the standard curve. Thus, a solution containing

5-sulfosalicylic acid (SSA), ethylenediaminetetraacetic acid (EDTA) and NEM effectively

68 protects 90% of free thiols in zebrafish embryo homogenates without any increase in glutathione disulfide (GSSG).

Analysis of amino acid and related metabolites

To extract amino acids and related compounds from embryos, 50 µL TCEP (100 mg/ml) and 350 µl acetonitrile were added to 1.7 ml microcentrifuge tubes each containing

10 embryos. The samples were homogenized with 0.5 mm zirconium oxide beads in bullet blender for 3 min, at speed 10, then centrifuged at 4 at 15,000 × g for 5 min. An appropriate aliquot of the supernatant was diluted with Mobile Phase℃ B (MP B, 10 mM ammonium formate in 85% acetonitrile) and IS containing multiple-labeled amino acids added. An aliquot was transferred to low volume injection vials and stored at –80 until LC-MS/MS analysis.

Amino acids and related compounds were analyzed,℃ as described [224] with modifications, using the UPLC-MS/MS equipment described above. The MS/MS was operated in ESI+ mode with equipment parameter settings as follows: desolvation gas 1000

L/h, capillary voltage 1.00 kV, desolvation temperature 550 and cone gas flow 10 L/h.

Collision energy (CE) and cone voltage were optimized for each℃ compound of interest and analyzed using MRM. The column used was a BEH Amide (Waters® Acquity UPLC 1.7 µM,

2.1 × 100 mm) with a VanGuard guard column (Waters® Acquity UPLC 1.7 µM, 2.1 × 5 mm).

The column was maintained at a temperature of 40 . Optimal chromatographic separation was achieved at a flowrate of 0.3 mL/min using a gradient℃ with mobile phase C (MP C, 10 mM ammonium formate) and MP D as follows: initial conditions were 100% MP D, hold 6 min. After 6 min a gradient was initiated for 0.1 min (6.0-6.1 min), then MP D was decreased to 94.1%. From 6.1 to 10 min MP D was linearly decreased to 82.4% and from 10 to 14 min,

MP D was decreased to 70.6%. After 14 min solvents were returned to the initial conditions.

Total run time was 20 min including column equilibration. MassLynx v.4.1 software was used

69 for instrument control and data acquisition. Retention times, MRM transitions and CE are shown for the unlabeled and labeled analytes (Supplement Table 3). The range of linearity

(pmol/injected) and correlation coefficients for the standard curves are shown in Supplement

Table 4.

Repeatability and reproducibility of measurements: thiols, amino acids, and related metabolites

Measurements of triplicates varied <10%; this variation is within acceptable limits set by the ICH calling for 3 replicates each at 3 test concentrations with variation of ±20% [225].

Inter-day assessments were made with thiols extracted and derivatized from three zebrafish homogenate pool of 24 hpf embryos analyzed five times. Derivatized analytes (GSH, cysteine and homocysteine) had between day % coefficient of variation (CV) in the range of 2.3 to

7.2% (Supplement Table 5). Within-day precision was also calculated with one zebrafish homogenate pool of 24 hpf embryos analyzed 5-times resulting in %CV values ranging from

2.7 to 11.7% (Supplement Table 5). The %CV for amino acid (AA) quantitation of five replicate injections of standards on one day ranged from 1.1 to 12.3%. The %CV ranged from

0.8 to 16.1% when the standards were injected on three different days. Higher %CV were observed for those compounds [s-adenosylhomocysteine (SAH), S-adenosylmethionine

(SAM), choline and betaine] for which no labeled IS was available. The highest within day

%CV of this latter group (20.8%) was observed for SAH, which is present at low concentrations in embryos (Supplement Table 6).

Statistical analyses, principal component analyses and hierarchical clustering

We validated our outcomes by assessing 1) sensitivity [limits of detection (LOD) and of quantification (LOQ)], 2) accuracy and precision, and 3) recovery of authentic standards added to embryos prior to extraction. LOD and LOQ were established with signal-to-noise ratio (S/N) of 3/1 and 10/1, respectively (consistent with the International Council for

70 Harmonization of Technical Requirements for Pharmaceuticals for Human Use (ICH) standards [225]). Precision (CV%) of inter- and intra-day assays was measured and calculated by replicate analysis (n=5 or 3, as indicated). For thiol analyses, labeled GSH-

13 15 ( C2- N) was used as an IS to determine derivatization efficiency and recovery throughout method development and sample analysis. For amino acid analysis, labeled IS additions for most analytes were used. Results for both assays were corrected for IS recovery with quantitation relative to the respective standard curves.

Data from both analyses were combined into one dataset. All data (metabolite concentrations per embryo) were normalized by log transformation and auto scaling prior to statistical analysis. Two-dimensional (2D) principal component analysis (PCA) score and loading plots were generated by MetaboAnalyst 3.0 (https://www.metaboanalyst.ca/) [178,

226]. Comparisons were made at each independent time point between diet groups using a supervised partial least squares-discriminant analysis (PLSDA) clustering method. PLSDA variable of importance (VIP) features were generated to identify metabolites that differentiate the diet groups. Heatmaps were generated between E+ and E– diet groups. Hierarchical clustering was performed using the Ward algorithm and separated by Euclidean distance with factors visually arranged by time point. Statistical differences between groups, over time were assessed using 2-way ANOVA with Tukey’s multiple comparison tests (Prism 6.0, GraphPad,

La Jolla, CA). Statistical significance between differences was set at P < 0.05. Differences between reproductive output and egg yolk volume in 3 hpf embryos was assessed with t- tests between E+ and E– groups. Data are reported as group mean ± standard error of the mean (SEM).

RESULTS

Embryo morphology and egg quality assessment

71 E+ and E– embryos were first assessed for proper development by morphologic screening at 12, 24 and 48 hpf with representative embryos shown (Figure 13). E+ embryos at 12 hpf were developmentally normal with 5 to 6 somites located dorsally in addition to anterior head and eye (Figure 13A). E– embryos, however, were noticeably delayed indicated by reduced somite progression (Figure 13B, *). By 24 hpf, E+ embryos had tightly packed V- shaped myotomes distributed throughout the trunk, defined eyes and early otic vesicle, and extended tails (Figure 13C). E– embryos at the same time had loosely packed myotomes and reduced tail extension (Figure 13D, *). At 48 hpf, E+ embryos are developmentally normal, indicated by pigment cell migration dorso-laterally, circulation present in the pericardium and straight notochord extending from head to tail (Figure 13E). E– embryos experience greater incidence of both yolk sac and pericardial edema (*) and bent axis (FIgure

13F). To determine if initial maternal contribution and initial egg quality predicts embryonic morphological deformities, reproductive parameters were assessed. E+ and E– females at approximately 1 year old and 9 months on diet produced similar numbers of eggs with 67.4

±7.7 eggs per E+ female and 75.7 ±17.8 eggs per E– female (p=0.674, n=7 and 13, respectively). Egg quality was assessed at 3 hpf or 1 thousand-cell stage prior to onset of any visible morphological abnormalities in either group. Egg yolk volume was found to be similar between embryos with 0.155mm3 ±0.003mm3 in E+ embryos and 0.153mm3

±0.006mm3 in E– embryos (p=0.758, n=7 per group).

Assessment of thiol extraction and quantitation method development

To evaluate the extent to which GSH depletion could explain altered metabolism, we undertook development of methods to accurately assess GSH status in zebrafish embryos.

N-ethylmaleimide (NEM) in acidic solution is the current accepted approach for GSH quantitation by derivation and protection of free thiol groups. We first investigated the amount

72 of time required to derivatize free thiols contained in the zebrafish embryo homogenate.

Embryo pools were homogenized in derivatizing solution, then at periodic intervals (1, 2, 3,

6, 9, 12 and 24 h) samples were centrifuged, supernatants were transferred to injection vials and subjected to analysis. Labeled internal standard (IS) added to the standard curve was used to estimate maximum derivatization in the absence of the embryo homogenate, which presumably was caused by matrix effects in those homogenates. We determined that the half-life to the maximum derivatization of homogenates was 3.3 h and was similar for all thiols tested. By 9 h nearly 90% of free thiols in the sample were derivatized with NEM. Using the

9 h incubation protocol, the recovery of added standard to sample homogenates were consistently >75% for all analyses at two points in the high [50-100 pmol inj] and two points in the low [0.5-1 pmol inj] end of the standard curve. Thus, a solution containing 5- sulfosalicylic acid (SSA), ethylenediaminetetraacetic acid (EDTA) and NEM effectively protects 90% of free thiols in zebrafish embryo homogenates without any increase in glutathione disulfide (GSSG).

Repeatability and reproducibility of measurements: thiols, amino acids and related metabolites

Measurements of triplicates varied <10%; this variation is within acceptable limits set by the ICH calling for 3 replicates each at 3 test concentrations with variation of ±20% [225].

7.2% (Supplement Table 5). Within-day precision was also calculated with one zebrafish homogenate pool of 24 hpf embryos analyzed 5-times resulting in %CV values ranging from

2.7 to 11.7% (Supplement Table 5). The %CV for amino acid (AA) quantitation of five replicate injections of standards on one day ranged from 1.1 to 12.3%. The %CV ranged from

73 0.8 to 16.1% when the standards were injected on three different days. Higher %CV were observed for those compounds [s-adenosylhomocysteine (SAH), s-adenosylmethionine

(SAM), choline and betaine] for which no labeled IS was available. The highest within day

%CV of this latter group (20.8%) was observed for SAH, which is present at low concentrations in embryos (Supplement Table 6).

Impact of VitE deficiency on GSH and GSSG

Embryo GSH concentrations increased in all groups from 12 to 24 hpf, then between

24 and 48 hpf in control and E+ embryos GSH concentrations plateaued (Table 3).

Previously, we have observed that the E– embryos undergo a metabolic switch between 24 and 48 hpf, with an apparent energy metabolism crisis with only some of the embryos surviving [100, 138]. In the present study, the E– embryos at 48 hpf were unable to maintain the same GSH content relative to the 24 hpf time point but were not significantly different than at 12 hpf. The GSH/GSSG ratio provides a measure of redox status. Up to 24 hpf, this ratio is very high (2500 to 4500 mol/mol) in all groups, but at 48 hpf the ratio has decreased in all groups to <725 mol/mol.

Integrated analysis of thiols and amino acids

To evaluate how VitE deficiency and the changes in thiol redox status altered free amino acid composition, quantitative assays were used to measure longitudinal (12, 24, 48 hpf) changes in E+ and E– zebrafish embryos. The thiol and the amino acid data from the same embryo cohorts were combined and evaluated using PLSDA. This analysis maximizes covariance between the metabolite features and VitE status, such that sample clusters are expected to separate with 95% confidence regions (CR) at each time point (12, 24 and 48 hpf). Component 1 describes metabolite features and component 2 describes treatment groups. At 12 hpf, component 1 for 54.2% of the variation and component 2 accounts for

16.6% of the variation (Figure 14A) . The Q2 index was greater than 0.90 with 4 components

74 indicating excellent model fit. Top VIP scores indicating most important features that were high in abundance in E– embryos at 12 hpf were betaine, glutamic acid, serine and histidine; by contrast, SAM was low in E– embryos (Figure 14B). At 24 hpf, E+ and E– embryos again clustered separately with non-overlapping CR (Figure 14C). Components 1 and 2 account for 43.1% and 24.3% of the variation, respectively. The Q2 index was greater than 0.90 with

4 components. Betaine, which was lower in E– embryos at 24 hpf, remained the most important feature separating embryo groups. Importantly, choline, threonine and tyrosine were all greater in E+ relative to E– embryos (Figure 14D). At 48 hpf, components 1 and 2 account for 15.7% and 57.7% of the variation, respectively (Figure 14E). The Q2 index was greatest with 3 components indicating excellent fit. Top VIP scores metabolites in E+ embryos include GSH and SAH with higher abundant metabolites in E– embryos including betaine and glutamic acid.

Hierarchical clustering of E+ and E– outcomes

Hierarchical clustering was used to classify the metabolite features into related groups

(Figure 15). A dendrogram was used to sort the metabolites (rows) according to similarities in their patterns using the Ward clustering algorithm, while the color scheme from blue

(lowest) to peach (highest) show the relative concentrations. Each of the time intervals (12,

24, and 48 hpf) are indicated for the E+ and E– embryos. In general, the compounds increased with embryo developmental age except for choline, cysteine, SAH, SAM, which were lower at 48 than at 24 hpf. These latter compounds clustered at the top of the heatmap.

Glutamate and Hcy clustered together and with the previous grouping. Betaine and GSSG clustered together and increased with time as shown by the dark peach color at 48 hpf.

(Figure 15). Cystine showed a similar pattern. Of note, compounds (GSH, serine, proline, and isoleucine) clustered together near the middle of the heat map showed similar trends.

The compounds that clustered in the bottom of the figure were methionine, leucine,

75 phenylalanine, threonine, tyrosine, valine, lysine, arginine, alanine, and histidine, which were correlated with each other.

Pathways relating AA and Thiol Status of E+ and E– embryos

Betaine and GSSG are oxidation products of choline and GSH, respectively; therefore, we have integrated our data with the metabolic pathways leading from choline and betaine to GSH and GSSG. The concentrations of the thiols and AA were measured in a single batch of control, E+ and E– embryos at 12, 24 and 48 hpf (Figure 4). We particularly focused on the two pathways to maintain cysteine concentrations because cysteine is the rate limiting amino acid for GSH synthesis. Notably, cysteine concentrations were unchanged between E+ and E– embryos, while by 48 hpf GSH concentrations were significantly (P<0.01) lower in E– (529 ± 21 pmol/embryo) compared with E+ embryos (655 ± 30 pmol/embryo,

Table 3). AA levels varied by time, with most reaching their highest concentrations at 24 hpf.

Choline showed no apparent differences among control, E+ and E– embryos. By contrast, betaine in E– embryos significantly increased at each of the time points: 12, 24 and 48 hpf, and were 6.7-, 2.0- and 1.6-fold higher than in E+ embryos, respectively. By contrast, SAM and SAH significantly decreased in E– embryos at 48 hpf, with 0.58- and 0.57-fold changes compared to E− embryos. Although methionine increased significantly with time, no statistical differences were found between the groups. Early (12 hpf) significant changes in E– embryos were observed for serine (1.3-fold increase compared to E+) and glutamate (1.4-fold increase compared to E+). Additionally, glutamate concentrations were higher in 48 hpf E– embryos

(1.4-fold compared to E+). Although threonine, alanine, tyrosine, isoleucine, leucine, phenylalanine, lysine, arginine, and histidine significantly increased with time, no differences were found in the concentrations in control, E+ and E– embryos (Figure S1). E+ embryos at

12 hpf showed significantly higher proline than in control samples (1.30-fold increase). At 24

76 hpf, valine in E– embryos decreased significantly compared to the control (0.81-fold change)

(Figure S1).

DISCUSSION

We hypothesized that the requirement for GSH during the increased lipid peroxidation observed in E– embryos drives a complex mechanism of overlapping biochemical pathways needed to maintain thiol homeostasis that is dependent on betaine and methyl group donation. Adapting sensitive extraction and quantitation techniques allows us to show that this is the case in the developing zebrafish embryo.

Our approach to evaluate the global changes in thiol and amino acids in response to

VitE deficiency was to analyze the quantitative changes in these compounds over time.

Previously, we showed that thiol status was dysregulated in E– embryos [100, 138] and in

E– embryos that were subsequently fed VitE adequate diets from day 5 to 12 d [100].

However, those studies used metabolomics approaches with relative quantitation between

E+ and E– animals. Further, no specific precautions were taken to prevent thiol depletion and thus, did not accurately quantitate thiols, especially GSH. In order to measure changes over time, a precise and effective method for determination of embryo thiol status was developed using two LC-MS/MS methods to evaluate the thiol and AA status of zebrafish embryos. Two methods were necessary because thiols are highly redox sensitive, and thus, require both a reducing environment and masking of the sulfhydryl group during isolation and analysis.

Sulfhydryl group masking is often achieved with 2-vinylpyridine, iodoacetic acid or monobromobimane [227]. Nonetheless, NEM was chosen as the primary derivatizing agent because NEM reacts more quickly with free-sulfhydryl groups and more efficiently permeates cell membranes during the homogenization process to protect the reactive sulfhydryl groups from oxidation [227]. We found that oxidation sensitive thiols in embryos required 9 h for complete derivatization. The method described herein provides simultaneous GSH and

77 GSSG detection by UPLC-MS/MS with a 4-minute run time with MRM detection of GSSG as low as 50 femtomoles GSSG injected, and thus is an improvement on previous methods

[228, 229]. The AA method described herein simultaneously quantitates 23 AA and related metabolites in zebrafish embryos with a ~20-minute run time with most concentrations detected as low as at 0.1 pmol/inj. High sensitivity and specificity provide accurate quantitation of embryo AA at critical developmental time points. Metabolite identification and quantification is further improved with the use of stable isotope-labeled AA, as IS.

Zebrafish embryos are a closed system up to 5 days post-fertilization and do not obtain external nutrients. Prior to 48 hpf, the neural cord and brain are developed and segmented, and circulation initiates through closed circuits just before the first heart beat

[219]. They, thus, are a valuable model of metabolic flux during early embryo development and lipid peroxidation arising during VitE deficiency before the presence of a fully formed liver. During this period, E– embryos experience greater incidence of developmental defects and delays indicated by reduced somites, bent axes and incomplete tail extension, and impaired brain and eye development. We therefore sought to determine the metabolic derangements associated with VitE deficiency during this critical developmental window.

Previously, we observed choline depletion, methyl donor alterations and disturbed cellular energy metabolism in E– embryos [100, 138]. These outcomes raise the question, “how is choline related to lipid peroxidation?” Choline is an integral part of phosphatidyl choline (PC), a membrane phospholipid, and we have shown that PC with docosahexaenoic acid (DHA-

PC) is depleted in E– embryos [166]. To replace oxidized DHA-PC, PC can be synthesized from choline via the cytidine diphosphate-choline (CDP-choline) pathway or via the phosphatidylethanolamine-N-methyl transferase (PEMT) pathway [230]. The PEMT pathway uses methyl groups (SAM) to methylate phosphatidylethanolamine (PE) to form PC [230].

Overall, choline could be used directly in the CDP-choline pathway or could be converted to

78 betaine for use in the methionine-Hcy cycle (Figure 16). Our data shows that E– embryos contained increased betaine concentrations at all times investigated (12, 24 and 48 hpf;

Figure 16). Potentially, VitE deficiency up-regulates betaine production to increase available methyl groups, suggesting that those E– embryos that cannot maintain the up-regulation of betaine production are unable to survive. Notably, betaine was a driver of differences between E+ and E– embryos as documented by the PLSDA analysis (Figure 14B, D, F)

During VitE inadequacy, lipid peroxidation generates lipid hydroperoxides, which are reduced by phospholipid glutathione peroxidase (GPX4) using GSH [231]. To replace GSH, either energy (as NADPH) is required, or GSH could be synthesized de novo. γ-glutamyl- cysteinyl synthase (γ-GSC) is the rate-limiting enzyme and cysteine the rate limiting amino acid for GSH synthesis. There are two cysteine sources [232]: 1) uptake of cystine in exchange for glutamate via the Xc– exchanger, then cystine is reduced to cysteine, and 2)

Hcy is redirected from the methionine cycle. In the present study, neither cystine nor cysteine concentrations were different between E+ and E– embryos at any time point, however, glutamate was significantly different (P<0.01) at 12 and 48 hpf. Since we cannot determine if the glutamate is within the embryo proper or the yolk sac, these data suggest that the animal is generating a glutamate surplus to promote cysteine availability via the Xc- exchanger.

Nonetheless, GSH by 48 hpf is significantly depleted (Figure 16J, Table 3). Livers from rats fed a VitE deficient diet for 290 days were similarly deplete of total GSH with no further increase of γ-GCS transcription suggesting a lack of VitE does not increase de novo synthesis of GSH [233]. As shown by Timme-Laragy [220], total GSH increases during zebrafish embryo development, while the GSH/GSSG ratio fluctuates as a response to cellular differentiation and organogenesis. Thus, the inability of the E– embryos to maintain GSH at

48 hpf indicates that by this time they are undergoing redox stress.

79 Future studies will include quantitative flux analysis of methionine and choline metabolism throughout development as it relates to overall antioxidant status to determine the flow of nutrients and metabolic regulation. Untargeted metabolomics assays show that these metabolites are dysregulated in the developing zebrafish embryo [100].

In the heat map (Figure 15), GSH clustered with serine, proline and isoleucine, while

Hcy, glutamate and cysteine clustered with methionine cycle related metabolites (choline,

SAM and SAH). These findings suggest that Hcy may be diverted from the methionine cycle in the response to VitE deficiency. Nonetheless, the importance of cystine cannot be ruled out since it was a major variable in the VIP analyses (Figure 14B, F). The hierarchical clustering analyses used provide valuable insight into metabolic relationships especially in a model with increased redox stress initiated by VitE deficiency. Unfortunately, the whole embryo homogenates used for analysis may not provide the most accurate relationships as the yolk contains abundant maternally derived resources. In addition, actively metabolic yolk syncytial cells that regulate early cardiac and neural tube morphogenesis [234] may experience varying degrees of VitE-deficiency induced stress and are not separate from other cell types in this analysis.

This study demonstrates that VitE deficiency alters overall thiol status in the developing embryo. Simultaneously, concentrations of the methyl donor, betaine, are greatly increased in the E– embryos. Betaine is thought to protect the pre-implantation human embryo by protecting epigenetic regulation via its methyl donation pathways [235]. However, betaine addition to culture medium of alcohol-treated mouse embryos suggested that betaine plays a role in protection from reactive oxygen species. We suggest that betaine’s critical role in the embryo is also to maintain thiol status. The quantitative analyses used here have provided some clues, but further flux analysis is necessary to define how specific metabolites change during embryogenesis. Ultimately, this study delivers a more accurate depiction of

80 redox-sensitive thiols and the biochemical pathway disruption as a result of VitE deficiency in the developing zebrafish embryo.

Figure 14. Embryos are similar sized at each stage with E– developmentally delayed. Live imaging of embryos was performed at each developmental stage used for assays and shows that embryos are roughly the same size measured by egg diameter, egg yolk diameter and estimated egg yolk volume. At 12 hpf, E+ embryos (A) develop normally indicated by 5 to 6 somites. At the same stage, E– embryos (B) are developmentally delayed with reduced somites (*) and smaller anterior head region. By 24 hpf, the otic vesicle is apparent, myotomes are neatly packed in V-shapes and the tail begins to extend in E+ embryos (C). In a representative E– embryo, however, the otic vesicle is less visible, posterior myotomes are loosely packed and the tail extension is reduced (D, *). E+ embryos at 48 hpf (E) show pigment migration dorso-laterally and have experienced the first heartbeat with visible circulation in the pericardium. E– embryos at the same stage (F) have greater incidence of both yolk sac and pericardial edema (*). (A-B) are taken with a 10x objective, (C-F) are taken with a 2x objective. Image was generated in Adobe Photoshop and any manipulations (contrast, saturation) were performed uniformly across images at the same developmental stage.

Figure 15. Schematic representation of 2D PLS-DA scores plot and important features. Metabolite features and concentrations from embryos (E+ and E–; n=5 per group per time point per analysis) were normalized and assessed by partial least squares – discriminant analysis (PLSDA). 2D score plots were produced to separate feature differences between E+ and E– embryos at (A) 12-, (C) 24-, and (E) 48 hpf. E+ is shown in blue, E– is shown in red with corresponding 95% confidence regions (CR) drawn around each sample group. Variable importance of projection (VIP) scores from Component 1 of each PLS model indicate the most important features separating E+ and E– embryos at (B) 12-, (D) 24-, and (F) 48 hpf. Black and white boxes to the right of the VIP score plot indicate relative concentrations of each metabolite between E+ and E– embryos at the given time point. .

Figure 16. Heatmap visualization of metabolite features and relative concentrations over time. Metabolite features and concentrations from embryos (E+ and E–; n=5 per group per time point per analysis) were normalized by log transformation and auto scaling. Heatmap was generated with Ward clustering algorithm and separated by Euclidean distance. Map is arranged by VitE status and time point with all measured metabolite features shown. Colored box to the right of the map indicates relative concentration of metabolite between E+ and E– embryos across all time points. Heatmap dendrogram separated primarily into two clusters by relative metabolite concentration: cluster one contains choline, cysteine, SAH, SAM, glutamic acid, and homocysteine with all other metabolites measured in cluster two. Cluster two is further separated by three family groups; one group contains betaine and GSSG, one group contains GSH, serine, proline, and isoleucine, and one group contains the remaining 11 metabolites, which are primarily amino acids.

Figure 17. Quantitative analysis of thiols, amino acids and related substances. Y-axes vary according to the concentration of each compound (pmol/embryo), X-axes shown times as 12, 24 and 48 hpf. Embryos (E+, E– or control; n=5 per group per time point per analysis) were analyzed both by the AA protocol (choline, betaine, SAM, methionine, serine, and glutamate) and the thiol protocol (homocysteine, cysteine, cystine, GSH and GSSG). Data shown are mean ± SEM. Symbols denote significant differences, as determined by Tukey post-hoc comparisons, at the time interval shown: E+ vs E– ‡ P ≤ 0.05, ‡‡ P≤ 0.01, ‡‡‡ P ≤ 0.001, ‡‡‡‡ P ≤ 0.0001; control vs E– * P ≤ 0.05, ** P ≤ 0.01, **** P ≤ 0.0001; control vs E+ # P ≤ 0.05, ## P ≤ 0.01, ### P ≤ 0.001. (A) Choline (Interaction, P = 0.0670, Time P < 0.0001, Group P = 0.7725), (B) Betaine (Interaction, P < 0.0001, Time P < 0.0001, Group P < 0.0001), (C) SAM (Interaction, P = 0.2115, Time P < 0.0001, Group P = 0.0666), (D) Homocysteine (Interaction, P = 0.3539, Time P = 0.0802, Group P = 0.4504), (E) Methionine (Interaction, P = 0.3661, Time P < 0.0001, Group P = 0.7944), (F) SAH (Interaction, P = 0.0405, Time P < 0.0001, Group P = 0.0271), (G) Serine (Interaction, P = 0.0357, Time P < 0.0001, Group P = 0.4039), (H) Cystine (Interaction, P = 0.7759, Time P < 0.0001, Group P < 0.0001), (I) Cysteine (Interaction, P = 0.1655, Time P < 0.0001, Group P = 0.2320), (J) GSH (Interaction, P < 0.0001, Time P < 0.0001, Group P < 0.0001), (K) Glutamate (Interaction, P = 0.2791, Time P < 0.0001, Group P = 0.0001), (L) GSSG (Interaction, P = 0.0977, Time P < 0.0001, Group P = 0.0622).

Table 3. GSH and GSSG concentrations (pmol/embryo) over time by embryo group.

Diet group GSH1 GSSG2 GSH:GSSG3

Control

12 hpf 569 ± 18 0.13 ± 0.01 a 4547 ± 432a

24 hpf 747 ± 37 a 0.30 ± 0.05 a 2678 ± 388 b

48 hpf 809 ± 37 a 1.14 ± 0.07 717 ± 52 c

E+

12 hpf 454 ± 13 0.13 ± 0.01 a 3514 ± 317 a

24 hpf 709 ± 27 a 0.18 ± 0.01 a 4113 ± 325 a,b

48 hpf 656 ± 30 a 0.99 ± 0.09 690 ± 82 c

E–

12 hpf 489 ± 9 a 0.12 ± 0.00 a 4185 ± 244a

24 hpf 695 ± 11b 0.28 ± 0.04 a 2771 ± 475 b

48 hpf 529 ± 21 a 1.11 ± 0.11 489 ± 41c

Concentrations are shown for GSH and GSSG (pmol/embryo, mean ± SEM, n=5 group per time). 1GSH: Interaction, P<0.0001; For the E+ vs E– comparisons: at 12 and 24 hpf, NS; 48 hpf, P<0.01. Within each group, values that do not share the same letter are different (P<0.001) at the indicated times. 2GSSG, time main effect P<0.0001. Within each group, values that do not share the same letter are different (P<0.001) at the indicated times3GSH:GSSG (mol:mol ± SEM, n=5 per time point): interaction P=0.0045; values were significantly different (P<0.05) 12 vs 24, except NS for the E+ 12 and 24 hpf comparison; all other times within each group, P<0.001.

86 CHAPTER FIVE:

RedEfish: generation of the polycistronic mScarlet:T2A:Ttpa zebrafish line

Brian Head, Jane La Du, Carrie Barton, Jie Zhang, Robyn L. Tanguay, Maret G. Traber

Unpublished manuscript

87 ABSTRACT

The alpha-tocopherol transfer protein (TTPA) is necessary for zebrafish embryo development prior to liver formation. Characterized initially in rodent and human hepatocytes,

TTPA transports VitE from recycling endosomes to the membrane where it can be delivered to peripheral tissues by lipoproteins. To evaluate embryo TTPA function, we generated a fluorescent-tagged zebrafish transgenic line with CRISPR-Cas9 technology. Casper

(colorless) zebrafish adults were spawned, their one-cell stage embryos collected and injected with a plasmid containing the mScarlet coding sequence in combination with cas9 protein complexed to single guide RNA molecules targeting 5’ of the ttpa genomic region.

Embryos were genotyped for proper insertion of the mScarlet coding sequence, raised to adulthood and successively in-crossed to produce the homozygote RedEfish

(mScarlet:GSG-T2A:TTPA). The RedEfish were characterized by in vivo fluorescence detection at 24 hours (hpf), 5- and 14 days post-fertilization (dpf). At 24 hpf, mScarlet:GSG-

T2A:TTPA is detectable by a faint mScarlet fluorescence distributed throughout the developing brain, posterior tailbud and yolk sac. At 5 dpf, the RedEfish is identifiable by red fluorescence in olfactory pits, gill arches, pectoral fins, posterior tail region and remaining yolk sac. Subsequently (14 dpf), the mScarlet protein is found in the olfactory pits, distributed throughout the digestive tract, along the lateral line and in the tail. Validation red fluorescent protein and TTPA expression was performed by immunolocalization and Western blot of 5 dpf embryos. The RedEfish does not experience different morphological outcomes or developmental delays relative to the Casper transgenic line. RedEfish does contain less VitE than the Casper line indicating an unknown effect on TTPA protein function. Potentially, the

RedEfish will be a powerful model to study TTPA function during embryo development.

88 INTRODUCTION

Vitamin E (VitE) is necessary for embryonic development [1]. Humans develop VitE deficiency if they have genetic defects in the α-tocopherol transfer protein (TTPA [Ttpa in zebrafish]). In hepatocytes, Ttpa functions primarily in VitE trafficking from recycling endosomes to the membrane in exchange for inositol-triphosphate (Ins3P) [18]. Limited information exists, however, about the role of Ttpa in the developing zebrafish embryo. ttpa, the gene encoding for Ttpa, is expressed in the zebrafish brain ventricle borders, eye and tailbud at 24 hours post-fertilization (hpf) [41, 167]. Translation- and splice-blocking morpholinos preventing Ttpa synthesis caused 100% lethality in zebrafish embryos by 24 hpf

[41]. In humans, in addition to the liver, TTPA is also expressed in placental uterine lining at the site of implantation [32] and in human yolk sac [40]. These findings suggest that TTPA has additional functions beyond those seen in the liver. Potentially, during embryogenesis

TTPA must be critical for 1) VitE transfer from the yolk to the developing embryo, 2) VitE transport from the placenta to the embryo, and 3) VitE trafficking in cells of the developing nervous system.

We hypothesize that the Ttpa localization in the embryo will provide critical information concerning the essential function of Ttpa to deliver VitE to target sites. To test this hypothesis, we will develop a transgenic zebrafish model containing a red fluorophore tagged Ttpa.

CRISPR-cas9 technology can be used to efficiently knock-in a genomic mScarlet sequence

5’ upstream of the ttpa genomic sequence separated by a self-cleaving viral peptide sequence (GSG-T2A). We propose mScarlet because it is a powerful red fluorophore with more intensity (0.54 quantum yield units) and longer lifetime (3.1 ns) than other known red fluorescent proteins [236]. The protein is not cytotoxic and has been shown previously to be effective for creation of transgenic zebrafish lines [237]. Ultimately, our goal is to generate a

89 fluorescent tagged Ttpa model to trace VitE-dependent and Ttpa-dependent cells at the onset of development in the zebrafish embryo. We report herein the creation of the RedEfish transgenic line.

METHODS

Zebrafish husbandry & strains

All experimental protocols and methods were carried out in accordance with the animal use and care protocol (# 5068) approved by the Institutional Animal Care and Use

Committee at Oregon State University. Tg[Casper; mitfaw2/w2, mpv17a9/a9] strain zebrafish

(hereafter called Casper) [238] were reared in Sinnhuber Aquatic Research Laboratory at

Oregon State University under standard laboratory conditions of 28°C on a 14 h light/10h dark photoperiod according to standard zebrafish breeding protocols [98]. 5D tropical strain and the c269Tg reporter line zebrafish were used as additional negative and positive controls for experiments, noted throughout.

Construction of donor plasmid and guide RNA vectors for knock-in

The donor plasmid containing homology arms to the ttpa genomic sequence, mScarlet sequence, Gly-Ser-Gly linker (GSG) sequence and 2A self-cleaving peptide (T2A) was constructed and supplied by In Vivo Biosystems, Inc (Eugene, OR). Constructs were assembled by Gibson ligation and confirmed by digest at restriction sites Pvul and Pvull, followed by sequencing. A single guide RNA (sgRNA) sequence was designed with the custom designed short CRISPR RNA (crRNA) sequence fused to the scaffold trans-activating

CRISPR RNA (tracRNA) sequence (Figure 18).

Microinjection

sgRNA and cas9 protein were complexed and co-injected into one-cell stage zebrafish embryos. Each embryo was injected directly into the embryo with approximately 1

90 nL of injection mix (10.5 µM sgRNA, 500 ng/µL cas9, 50 ng/µL donor plasmid, and 50 µM

NU7441 in 1% DMSO). Embryos were then placed on ice-cold 1.5% agarose for no more than 15 minutes. NU7441 and cold exposure after injection were used to improve knock-in efficiency of mScarlet sequence by homology-directed repair mechanisms (HDR) [239].

NU7441 in 1% DMSO (50 µM) injected at the levels described did not have deleterious effects on normal zebrafish embryos, as assessed by morbidity, mortality, and behavioral outcomes at 5 days post-fertilization (dpf) [150]. Injected embryos were collected, staged, and incubated in standard embryo media [(EM),15 mM NaCl, 0.5 mM KCl, 1 mM MgSO4, 0.15 mM KH2PO4,

0.05 mM Na2HPO4, 1 mM CaCl2, NaHCO3 in fish system water].

Mutation detection and insertion mapping

Genomic DNA (gDNA) was extracted from injected embryos (n=32) at 5 dpf after each trial, as described [240]. Embryos were euthanized with tricaine [MS222, ethyl 3- aminobenzoate methanesulfonate salt, (Sigma Aldrich, St. Louis, MO, USA)] in accordance with animal care and use guidelines, placed in 50 µL 50 mM NaOH and incubated at 95°C for 15 min. DNA extraction was stopped with 10 µL 1M TRIS pH 7.4, and stored in 4°C until use.

PCR reactions were performed with primers designed to capture the complete insertion region (Table 4). Target regions were amplified by PCR and sequenced by Sanger

Sequencing at the Center for Genome Research and Biocomputing at Oregon State

University. Adult injected zebrafish were fin clipped at 3 months, gDNA extracted and PCR performed as described above. Identified carriers were out-crossed with Casper zebrafish to produce F1 heterozygotes Tg[mScarlet:ttpa; mScarlet+/-, mitfaw2/w2, mpv17a9/a9], which were then in-crossed to produce F2 homozygotes (hereafter called RedEfish).

91 In vivo fluorescence detection

Live embryo imaging was conducted using a Keyence BZ-x700 All-in-one microscope with a 2x or 10x objective and 644 nm filter. Live embryos were mounted in 3% methylcellulose at room temperature in 35 mm glass bottom MatTek dishes and covered in embryo media and tricaine to anesthetize live embryos. Image adjustments, including cropping, brightness and contrast were performed uniformly using Adobe Photoshop.

Egg quality and embryo morphological assessment

RedEfish embryos were collected, staged, and incubated in standard EM. Egg traits were assessed by egg yolk diameter followed by egg volume estimated based on the volume of an ellipsoid using the equation [221, 222]:

= (4/3)π ( ) ( ) 2 At 24 hpf𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉𝑉 and 5 dpf, RedEfish∗ 𝑠𝑠𝑠𝑠 and𝑠𝑠𝑠𝑠𝑠𝑠 𝑎𝑎Casper𝑎𝑎𝑎𝑎𝑎𝑎 𝑟𝑟𝑟𝑟𝑟𝑟 embryos𝑟𝑟𝑟𝑟𝑟𝑟 ∗ 𝑙𝑙𝑙𝑙𝑙𝑙 were𝑙𝑙𝑙𝑙 𝑎𝑎 𝑎𝑎𝑎𝑎assessed𝑎𝑎 𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟𝑟 for morbidity and mortality outcomes, as described [149, 150]. RedEfish and Casper 5 dpf larval α- and γ- tocopherol concentrations were determined using high-performance liquid chromatography with electrochemical detection, as described with comparison to authentic standards [175].

Parametric t-tests with Welch’s correction were used to determine differences between groups (p<0.05).

Whole mount immunolocalization and Western blot analysis

Distribution of mScarlet protein was assessed in embryos (3 dpf) using the polyclonal anti-RFP antibody [anti-rabbit RFP, 1:3000; abcam, ab62341 (Cambridge, MA, USA)].

Monoclonal antibodies generated against acetylated tubulin [mouse anti-goat, 1:4000; Sigma

Aldrich (St. Louis, MO USA)] label axons in the developing embryo as a positive control.

Embryos (n=15 per group) at the described time points were fixed overnight in 4% paraformaldehyde (PFA) in phosphate-buffered saline (PBS), washed in PBS the following

92 morning and stored at 4°C in PBS and 0.1% sodium azide until use. Fixed embryos were washed in PBS + 0.1% Tween20 (PBST), permeabilized with 0.005% trypsin in PBS on ice for 5 min, rinsed with PBST and postfixed in 4% PFA. Permeabilized embryos were blocked in 10% normal goat serum in PBS + 0.5% Triton X-100 for an hour followed by incubation with the primary antibody overnight at 4°C in 1% normal goat serum in PBS and 0.5% Triton

X-100. Embryos were washed 4 times for 30 min in PBST, incubated with secondary antibody

[Alexa-488 goat anti-mouse, 1:1000; ThermoFisher (Carlsbad, CA, USA)] for 2 hrs and washed again 4 times for 30 min in PBST.

Distribution of Ttpa protein was assessed in zebrafish embryos (5 dpf) using the polyclonal anti-TTPA antibody [anti-rabbit TTPA, 1:300; abcam, ab155323 (Cambridge, MA,

USA)]. Embryos previously fixed in 4% PFA were washed in PBST and stored in 100% methanol. Samples were rehydrated and treated with 10mg/mL proteinase K in PBST.

Samples were then blocked in 5% normal goat serum and 3mg/mL bovine serum albumin in

PBST followed by incubation with the primary antibody. Embryos were washed, incubated with secondary antibody [Alexa-594 goat anti-rabbit, 1:1000; ThermoFisher (Carlsbad, CA,

USA)], washed and visualized.

All fixed embryo imaging was conducted using Keyence BZ-x700 All-in-one microscope with a 2x, 10x, or 20x objective and 470 nm or 694 nm filter, as described above.

Image adjustments, including cropping, brightness and contrast were performed uniformly using Adobe Photoshop.

Protein was extracted from pooled (n = 25) embryos (24 hpf) in RIPA buffer supplemented with protease and protein phosphatase inhibitors (Calbiochem, La Jolla, CA).

Rat liver extract was used as a positive control for the assay. Extracted protein (20 μg) was subjected to SDS-Page and processed by immunoblotting. Antibodies used were anti-ATP

93 synthase F1 subunit α (ab110273), anti-TTPA (ab155323), anti-red fluorescent protein

(ab62341), goat anti-rabbit and goat anti-mouse secondary antibodies. All gels and blots were run simultaneously using 2% bovine serum albumin or 3% nonfat dry milk in PBST blocking buffer. Proteins were visualized by SuperSignalTM West Femto Chemiluminescent

Substrate (ThermoFisher, Carlsbad, CA) and quantified using the Bio-Rad Image System

(Hercules, CA, USA).

RESULTS

Characterization of RedEfish

Mosaicism in the founder generation was not immediately apparent by fluorescent imaging. PCR genotyping was utilized to identify sequence expressing fish. About 7.5%

(n=19 of 254) of 3 month old (mo) adults screened were identified as transgenic; one screened adult (male) retained the mScarlet:T2A construct. Success was defined as integration of the mScarlet sequence 5’ of ttpa without erroneous insertions or deletions

(INDELs), as determined by Sanger Sequencing. The mosaic founder male was outcrossed with female Casper zebrafish (mitfaw2/w2, mpv17a9/a9) to produce F1 embryos. F1 embryos were raised to 3 mo for further in-crossing with the F0 mosaic founder. This in-cross produced

F2 heterozygotes, identified by PCR and sequencing. In vivo fluorescence of heterozygotes was apparent by 5 dpf, however, the yolk sac remnants were a significant source of autofluorescence. F3 in-crosses were performed to generate homozygotes. Sequencing of heterozygote embryos indicating two gene products, one containing the mScarlet:ttpa construct and the unmodified ttpa exon 1 (Supplementary Figure S5). Sequencing about the targeted region show no INDELs either in the 5’ region of mScarlet or in the 5’ region of ttpa exon1 (Supplementary Figure S5).

94 In vivo fluorescence was detected during the early larval stage at 7 dpf. RedEfish express mScarlet in the olfactory pits (Figure 19A), throughout the liver and digestive tract

(Figure 19B-C), and express punctate signal throughout the tail fin and fin edges. By 14 dpf,

RedEfish express mScarlet throughout the digestive tract. Most striking, mScarlet expression was found in the caudal vein plexus, caudal vertebrae, and tail fin. The red fluorescence in the vertebrae is localized to dorsal root ganglia (Figure 19E).

Whole mount immunolocalization was performed to validate the presence of mScarlet and TTPA. GSG-T2A linker sites are not 100% efficient peptide cleavage sites, which can cause either (1) mScarlet fused to exon 1 of Ttpa, which may potentially produce fluorescent but nonfunctional protein, or (2) skipped and reduced Ttpa, as was described [241]. RedEfish embryos were fixed in 4% PFA and subject to antibody staining with an anti-RFP antibody.

Un-injected Casper zebrafish embryos were used as a negative control, with the c269 allelic variant zebrafish embryo used as a positive control. Anti-acetylated tubulin is shown as a control antibody. All embryos are 3 dpf. Red fluorescent protein (Rfp) was expressed in

RedEfish brain, trunk, yolk syncytium and pectoral fins. Rfp was not detected in Casper embryos and was detected in the positive control c269 embryo (Supplementary Figure S6).

Immunolocalization with anti-TTPA validated that Ttpa is expressed in the zebrafish olfactory pits, gill arches and digestive tract lining in 5 dpf embryos (Figure 20A). Western blot analysis was also used to evaluate the success of protein expression. We anticipated a protein band at 59.6 kDa (sum of mScarlet (26.4 kDa) and Ttpa (32.7 kDa) if the construct was produced, but not hydrolyzed. Ttpa was detected in Casper, RedEfish and the c269 lines

(24 hpf) (Figure 20B). Two bands were present for Ttpa; however, the two bands were present in all samples including rat liver used as a positive control. No band for a larger protein at approximately 60 kDa was observed. An antibody for Rfp was also used to determine if mScarlet protein is expressed in the RedEfish line, although this antibody has

95 not previously been used with mScarlet. The anti-RFP antibody used for immunlocalization was not effective in Western blot analysis (data not shown).

Embryo morphology and egg quality assessment

Egg quality was assessed in third generation RedEfish and Casper embryos at 6 hpf

(gastrulation), when ttpa is first expressed in zebrafish [41, 167]. Egg yolk volume was found similar between RedEfish (0.220 ± 0.024 mm3) and Casper (0.205 ± 0.023 mm3) embryos

(p=0.18, n=10 per group). Morbidity and mortality outcomes at 5 dpf were also similar with less than 5% of a given clutch experiencing a developmental deformity (n=100 embryos each in 3 independent spawning events). Finally, because of potential modification to ttpa by insertion of the mScarlet coding sequence, VitE status was assessed in RedEfish and Casper embryos at 5 dpf. γ-tocopherol concentrations were lower in RedEfish (2.21 ± 0.09 pmol/embryo) relative to Casper strain (2.80 ± 0.24 pmol/embryo; p=0.001, n=5 homogenate pools per group, n=15 embryos per sample, mean ± std. deviation). α-tocopherol concentrations were also lower in RedEfish (70.07 ± 5.85 pmol/embryo) relative to Casper strain (104.61 ± 16.18 pmol/embryo) (p=0.002, n=5 homogenate pools per group, n=15 embryos per sample, mean ± std. deviation).

DISCUSSION

We present the RedEfish, a new model for VitE trafficking. This model will be useful to study early embryo development and as the zebrafish grows to assess the requirement for

Ttpa. Based on the concept that mScarlett accumulates in cells that synthesize Ttpa, we provide further support that Ttpa must be associated with numerous developing embryonic structures including throughout the nervous system, circulatory system, and digestive tract by mere Ttpa presence via expression of mScarlet. These data also suggest that VitE is required by these cells as TTPA’s known function is to traffick α-Tocopherol.

96 Lecithotrophic animals, those that rely on maternal deposited yolk, are an important model for lipid and lipid-soluble vitamin metabolism during yolk utilization. The zebrafish embryo develops separately from its yolk [93, 94], which is bound by the yolk syncytial layer

(YSL) that secretes nutrients from the yolk into the developing organism. The YSL expresses many of the genes needed for lipid and lipid-soluble vitamin transport [95, 242] likely facilitating the transfer of VitE from the yolk to the embryo. Zebrafish express immediately after the gastrulation the microsomal triglyceride transfer protein (mttp) in the YSL, which facilitates apolipoprotein B (ApoB) secretion to increase neutral lipid secretion and accumulation throughout the embryo until the yolk depletes [243, 244]. Mice similarly require mttp to export triglyceride in ApoB-containing lipoproteins and mttp-null mice die mid- gestation [245, 246]. Defects in mttp causing abetalipoproteinemia in humans also cause

VitE deficiency [247]. Thus, VitE transport from the yolk likely depends on the lipid metabolic machinery of the YSL. The gene ttpa was found in zebrafish embryo YSL, as well as early neural structures including the eye, brain and posterior tailbud [41, 167]. We report mScarlet fluorescence is detected in the YSL in the zebrafish embryo through yolk depletion and continues to produce punctate expression throughout the larval digestive tract until at least

14 dpf. Ttpa may be involved in retrieving VitE from yolk stores as well as from the enterocytes upon first feeding at 5 dpf, although this is not yet supported in other embryonic models. We previously report that VitE-deficiency impacts neural crest cells expressing sox10 [167], which may be important to enteric neuron maturation.

Ttpa, a cytosolic protein, localizes to recycling endosome, which is also marked by expression of Npc1 and 2. Both Npc1 and Npc2 are expressed specifically in the YSL in the zebrafish embryo [248], a metabolically active organ that exists up to 9 dpf [234]. Npc1-/- zebrafish mutants have increased unesterified cholesterol in trunk neuromasts indicating localization of impaired recycling endosomal metabolism and possible co-localization of Ttpa

97 [249], supported by early embryonic mScarlet expression in our line. Indeed, NPC1-/- mice experience negatively impacted hepatic and cortical VitE levels [250], with accumulation of

VitE in lysosomal compartments [20]. Our model supports previous findings that Ttpa and

Npc may localize to the same cell or tissue types involving endosomal recycling, including neuromasts and the lateral line especially in 5 dpf embryos.

mScarlet signal was not immediately apparent in the mosaic founder zebrafish nor in the early embryonic stages, up to 5 dpf, in the heterozygote or homozygote RedEfish.

Although some fluorescence was detected during this developmental window, successful transgenesis required molecular validation because of the significant autofluorescence from the yolk. The bright yolk sac may be indicative of Ttpa and simultaneous mScarlet expression, however, that outcome is currently unclear. We hypothesized that the presence of VitE separate from yolk-derived transport may stimulate embryonic tissue expression of Ttpa and mScarlet. VitE supplementation increases Ttpa mRNA expression in isolated rat liver after food deprivation [251]. In addition, VitE supplementation stabilizes and prevents degradation of TTPA in HepG2 cells [252]. Punctate protein expression may also be achieved by supplementation because TTPA colocalizes to lysosomes, indicated by LAMP1, and fluorescent tocopherol in isolated mouse hepatocytes [22]. Although feeding induced red fluorescence detected at 7, 14 and 21 dpf, signal was primarily localized to the gut and digestive tract lining rather than the liver, as expected. Larval zebrafish guts remain highly auto-fluorescent [253], likely a result of their microbiome [254]. Further investigation of the

RedEfish model should exclude examination of the auto-fluorescent yolk and digestive tract.

Importantly, VitE is critically associated with nervous system development.

Chronically VitE deficient mice [255] and zebrafish [148] experience cognitive dysfunction resulting from brain specific lipid peroxidation. In addition, blocking translation of Ttpa prevented formation of the eye and brain of 24 hpf zebrafish embryos [41]. TTPA is found in

98 the cerebellum of adult rats [33] and in mice with VitE deficiency caused by deletion of TTPA exhibit lipofuscin accumulation as a result of oxidative stress specifically in the dorsal root ganglia (DRG) [55]. DRG from TTPA-/- mice have increased apoptosis [79], causing reduced mechano-sensitivity and excitability prevented only by very high levels of VitE supplementation [211]. Knowing this, we anticipated mScarlet expression as a proxy for Ttpa expression in the DRG of adult zebrafish. Although fluorescence was detected in the

RedEfish vertebrae at 21 dpf, some red fluorescence was also present in another Casper zebrafish strain without red fluorescent protein expression (data not shown). The mScarlet expression detected in the RedEfish may be partially caused by autofluorescence, although it is unknown what produces this effect in 21 dpf zebrafish embryos.

mScarlet can be visualized using existing microscope resources with an absorbance and emission spectra of 569 nm and 594 nm, respectively. Further, the T2A site was chosen due to high “cleavage” efficiency for a 2A self-cleaving peptide sequence in zebrafish [241].

However, 2A sites present a technical challenge as some bi-cistronic proteins remain un- cleaved because of ribosomal read-through or even falling-off [256]. Because VitE levels were found lower in RedEfish relative to the Casper line, it appears that ribosomal skipping is occurring to produce reduced Ttpa and thus reduced VitE levels in the embryo. Western blot analysis did not confirm reduced Ttpa expression, however, this analysis also did not indicate a larger protein with conjoined mScarlet and Ttpa.

Guide RNA (gRNA) sequences were selected for a region 5’ of the first ttpa exon to create a polycistronic 2A construct encoding mScarlet and ttpa in tandem. The T2A self- cleaving peptide has the highest cleavage efficiency of the 2A peptides and has been used to produce independent fluorophore and target proteins in immortalized cell lines, as well as zebrafish embryos [241]. We were able to confirm, using the commercially available TTPA- antibody that Ttpa is expressed in the RedEfish at 24 hpf, as well as in other transgenic fish

99 lines (c269 and Casper). Because of antibody limitations, however, we unable to confirm the presence of Rfp by Western blot analysis. It is possible that the anti-RFP antibody is not suitable for detecting mScarlet and other protein expression methods will be necessary.

In summary, the RedEfish is a new model to study VitE trafficking in early embryonic development in a model organism suitable for further genetic modification, as well as easy observation by external development. We provide ample evidence supporting the proper germ-line transmission of the mScarlet coding sequence 5’ of the first exon of ttpa. This line will provide valuable understanding of VitE trafficking given VitE sufficient and deficient diets because for the first time we will be able to observe in real-time the effects of VitE status on

VitE-dependent cells, tissues, and organs. Further, interrogations to prevent morbidity and mortality caused by VitE deficiency will be visualized in vivo. Finally, we plan to isolate mScarlet expressing cells from these embryos and repeat analyses found in Chapters 3 and

4 to evaluate VitE-dependent cell metabolomic and transcriptional profiles.

100

Figure 18. Knock-in strategy and designed protein sequence. LHA refers to the left homology arm and RHA refers to the right homology arm of the donor plasmid containing the mScarlet coding sequence. Amino acid sequences are denoted by color to indicate protein: mScarlet (red), GSG-T2A linker site (Green), exon 1 of ttpa (blue).

101 Table 4. Primers designed for insertion mapping. Primers designed for left and right homology arms were first utilized to identify germ-line transmission of the mScarlet sequence. Successful insertion and validation of mScarlet expression determined with mScarlet cassette primers. All primers designed with NCBI primer Blast.

Primer purpose Sequence (5’ – 3’) Product size (bp)

Left homology arm F AGGCGTGTGTCTCTGTAAGG 418 Left homology arm R GCCGTACATGAACTGAGGGG

Right homology arm F CGCTACCTGGCGGACTTC 309 Right homology arm R AATCCGGGAGGAATCAACAGG mScarlet cassette F CGCTCTGAGAACAACATGACAC 121 no insertion, mScarlet cassette R GACAAATATGGTGCAATCCGGG 879 insertion

102

Figure 19. In vivo fluorescence detected in 7- and 14 days post-fertilization RedEfish. Autofluorescence was detected in the pigment cells surrounding the eyes (e) in all embryos (A), however, mScarlet expression was unique to the olfactory pits (op) found in the RedEfish (A). mScarlet expression was identified in the digestive tract (B), specifically the liver (*) and with more punctate expression viewed in the anterior gut region (C). Further punctate expression was observed in RedEfish tail fin (D, *). By 14 dpf, mScarlet expression was identified in the caudal vertebrae and dorsal root ganglia (drg) (E). Fluorescence was also detected in the caudal vein plexus (cpv). Autofluorescence of the tail fin bones was noted in RedEfish and Casper larvae.

103

Figure 20. Immunolocalization and Western blot of TTPA protein. (A) 5D strain fish expressed TTPA in olfactory pits, gill arches and gut lining. Anti-TTPA was used to detect and validate TTPA protein expression in 5D embryos initially to determine regionalization followed by in vivo observation. The same antibody confirmed that TTPA expression is not affected by the insertion of the mScarlet coding sequence (B) in RedEfish relative to Casper and the c269 transgenic lines.

104 CHAPTER SIX:

CONCLUSIONS AND FUTURE DIRECTIONS

Vitamin E was discovered in 1922 as a vital nutrient required to prevent fetal resorption in rats [1]. Nearly a century later, the mechanism underlying this protective effect remains under investigation. The primary objective of this thesis was to investigate the time- dependent requirement for VitE in the vertebrate embryo through transcriptional and metabolic profiling of developing zebrafish embryos.

I hypothesize that VitE is necessary to protecting oxidation sensitive polyunsaturated fatty acids in the developing zebrafish embryo. Lipid peroxyl radicals abstract hydrogens from nearby lipids to form lipid radicals and lipid hydroperoxides; the latter requires detoxification by intrinsic antioxidant systems including the glutathione peroxidase enzyme family that utilizes glutathione as a substrate to reduce lipid hydroperoxides to more stable lipid alcohols.

Oxidized glutathione (GSSG), the by-product of the reactions requires reduction by thr enzyme glutathione reductase and NADPH, a key cofactor in numerous anabolic reactions, which is generated primarily by glycolysis and the pentose phosphate pathway. Together,

VitE and the lack thereof intimately connects redox balance and energy status. Thus, the VitE deficient zebrafish embryo is unable to prevent lipid peroxidation reactions, which incurs depletion of antioxidant status and of energy stores all to protect developing cells, tissues, and organs. Indeed, work by Melissa McDougall, et al. showed that a VitE deficiency in zebrafish embryos alters phospholipid remodeling [166], likely as a result of oxidizing lipid species contained within membranes, and dysregulation of energy status between 24- and

120 hours post-fertilization [100, 138]. This thesis continues that work by providing further evidence that gene expression profiles relating to energy metabolism are disrupted as early as 12 hpf.

105 I have shown that E– embryos have increased expression of genes at 12-, 18- and

24 hpf that strongly associate with glycolysis, carbon metabolism and the pentose phosphate pathway. Across the same developmental windows, E– embryos had reduced gene expression associated with anabolic reactions and gene transcriptional pathways.

Longitudinally, 12 hpf E– embryos had gene expression profiles positively enriched with genes associated with methylation and energy metabolic status. By 18 hpf, gene expression profiles were negatively associated with gene transcription pathways.

I also closely examined the morphology using specific mRNA markers of neurogenesis. At 6 hpf, E– embryos experience an otherwise normal appearing gastrulation marked by expression of the gene goosecoid in the early involuting ectoderm. However, by

12 hpf, pax2a and sox10, markers of the midbrain-hindbrain boundary and neural plate, are aberrant in the E– embryo. By 24 hpf, these patterns are worsened in E– embryos, as observed by reduced expression of pax2a in the trunk spinal cord neurons and reduced sox10 in the migrating neural crest cells distributed through the embryo trunk. These errors in gene expression patterns translate into morphological deformities in E– embryos at 24 hpf including over-inflated forebrain, altered midbrain hinge points, and somite packing throughout the trunk. Similar morphological impairments plague E– embryos with bent axes identified by in vivo imaging as well as deformed collagen, specifically col2a1a and col9a2, distributed throughout the E– 24 hpf notochord.

Thus, Chapters 2 and 3 show that VitE deficiency impacts gene expression abundance and patterning as early as 12 hpf and causes morphological errors that ultimately lead to increased morbidity and mortality outcomes. Because gene expression is inherently tied to cellular metabolism, I sought to characterize both the cause and consequence of transcriptional changes. Indicated above, 12 hpf E– embryos have altered expression of methylation status and carbon metabolism genes. To this end, I showed that metabolic

106 profiles of E– embryos at 12 hpf were driven primarily by increased betaine and glutamic acid status. Betaine is critical in the regeneration of methyl donors via betaine mitochondrial oxidation to choline, which re-methylates homocysteine to form methionine. Glutamic acid is necessary for both the synthesis of glutathione as well as the exchange for extracellular cystine, the oxidized disulfide cysteine and the rate-limiting amino acid for glutathione synthesis. Increased betaine remains a key feature of E– embryo metabolic profiles at 24- and 48 hpf with the emergence of reduced choline at 24 hpf and reduced glutathione at 48 hpf. Targeted metabolomic assays were developed to accurately extract and quantitatively detect amino acids and thiol antioxidants in zebrafish embryos. Our precise LC/MS measurements indicated E– embryo dysregulation of betaine at 12-, 24- and 48 hpf used to regenerate S-adenosyl methionine and cysteine, which is used synthesize glutathione.

Nonetheless, glutathione status is not maintained in 48 hpf E– embryos. Further unpublished evidence points to increased LPO in the E– embryo at 24 hpf with increased malondialdehyde

(MDA), a by-product of linoleic acid oxidation (Supplementary Figure S7), supporting the claim of increased LPO driving the loss of glutathione.

Not only did VitE deficiency severely impact metabolic status as early as 12 hpf, but it continued to disrupt GSH regenerative pathways up to 48 hpf. It is most likely that early alterations in E– betaine and methyl donor status signal embryonic epigenetic regulation that alters the transcriptional landscape [257, 258] with energy and metabolic status disturbing necessary cellular growth, proliferation and even cellular migration [124, 125, 259].

Chapters 2, 3, and 4 characterize the E– embryo as having disrupted gene expression associated with 1-carbon metabolism and depletion of key nutrients that likely impact the formation of critical neural structures like the midbrain-hindbrain boundary between 12- and

24 hpf. Zebrafish embryogenesis between this window is defined by extensive elongation and rudimentary organ development. More importantly, just prior 12 hpf the brain primordium

107 thickens into the neural keel, which will inevitably form the neural rod and tube. At 16 hpf the embryo brain begins to regionalize into 3 distinct brain structures (fore-, mid- and hindbrain) and neural crest cells begin to migrate throughout the head and trunk [98]. The developmental window we focused on is evidently associated with important steps in neurogenesis that are dependent on pre-determined gene expression networks [260, 261] and adequate energy metabolic status [262-264]. Ultimately, VitE deficient embryos experience significant disruption to both, however, the initiating steps preceding these events remain under investigation. My studies in Chapter 3 support that mTOR, the master regulator of cellular growth and survival, is the key regulator that signals these metabolic status cues, determines E– embryo health outcomes.

Finally, the objective of Chapter 5 was to generate a transgenic, fluorescently tagged

Ttpa zebrafish model system to study expression, localization, and function of the embryonic

Ttpa protein. Using CRISPR-Cas9 technology, the mScarlet fluorescent protein and T2A self- cleaving peptide coding sequences were inserted 5’ of the genomic ttpa sequence in Casper zebrafish embryos. The embryos were grown up and repeatedly in-crossed to generate the homozygote RedEfish Tg(mScarlet:GSG-T2A:TTPA) line. Although not immediately apparent in early development, red fluorescence indicating expression of Ttpa was observed throughout the 5 dpf RedEfish digestive tract, gill arches, hindbrain, and olfactory pits. mScarlet was also detected in the lateral line, caudal vein plexus and at the edges of the posterior tail region. As the larval zebrafish begins to feed after 5 dpf, mScarlet fluorescence was observed with more punctate signal in the gut lining, gills, and olfactory pits. Of note, by

15 dpf the RedEfish is characterized by fluorescence in the vertebral column and other bone structures of the tail. Evidence of mScarlet expression indicates that VitE is transported between these structures to deliver VitE to nearby tissues, although dependency on VitE may

108 not be immediately obvious. The RedEfish provides a new model for VitE research, discussed in the future directions.

Future Directions

Pharmacologic modulation mTOR and energy-related pathways

I have shown in this thesis and in agreement with previous work that VitE is necessary to protecting cellular energy metabolism. Previously, McDougall, et al., found that E– embryos experienced disruption to energy metabolism through increased glycolytic intermediates, TCA cycle intermediates and even short chain fatty acid abundance between

24- and 120 hpf [100, 138]. Miller, et al. found that 36 hpf E– embryos experienced alterations to gene expression related to survival, growth, and especially carbon-derived energy metabolism [101]. A primary objective of this work was to further explore those pathways with thorough evaluation of gene and metabolic pathway derangement across embryonic development beginning at 12 hpf. E– embryos reported herein display alterations gene expression associated with carbon energy metabolism, anabolic reactions, and gene transcription between 12- and 24 hpf. During the same period, E– embryos have increased betaine status, likely to regenerate critical thiol pro-antioxidant cysteine and antioxidant glutathione. In addition, these alterations are sensed by mTOR complex 1 (mTORC1), a complex of proteins that regulates cell survival, growth, and proliferation pathways dependent on nutrient and redox status. I showed that mTORC1 formation is less abundant in E– embryos at 24 hpf. Its downstream targets are subsequently less active evidenced by reduced phosphorylation status of Rps6k and Eif2α. Likely, mTOR is a critical regulator of cellular energy status, which also plays a major role in nervous system development [265-

267]. To interrogate this pathway, mTOR can be targeted by inhibitors including rapamycin to selectively reduce mTORC1 formation in E+ embryos to replicate the phenotype of E–

109 embryos. Alternatively, glucose injections can be used to stimulate mTORC1 formation to increase Rps6k phosphorylation and signaling for increased cell growth in E– embryos.

Future objectives should replicate the protein analysis performed in Chapter 3 as a means of evaluating mTOR’s causal role in E– embryo health decline. mTOR complex 2 (mTORC2) formation also plays a critical role in cell growth via regulation of lipid and glucose metabolism. Rictor, the regulatory protein in mTORC2 formation, should be measured in E+ and E– embryos, as well as those treated with mTOR mimetics including rapamycin. Since immediate downstream targets of mTOR are associated with numerous signals, future directions should include analysis of upstream regulators that induce mTORC1 or mTORC2 formation. TSC complex regulates mTORC1 formation and is regulated by energy status via

AMPK, oxygen levels via REDD1, growth factors via ERK signaling and energy status via

Akt. All possible routes of mTORC1 formation, or lack thereof, should be noted by both gene and protein expression analyses. Pharmacologic modes of interrogation are the preferred mechanism because germline knock-down models are primarily lethal [266, 268]. mTOR’s causal role in E– embryo morbidity and mortality will be evaluated by addressing mTOR status and its upstream regulators.

Characterization of VitE-trafficking dependent cell types

The RedEfish is a new model for in vivo evaluation of VitE status and the effect on its direct trafficking partner TTPA. Early zebrafish embryos, prior to 72 hpf, rely on lipid metabolism in the YSL. TTPA, characterized primarily for its hepatic VitE trafficking function, likely differs within the more basal lipid metabolic machinery of the YSL. Miller, et al. found that the gene ttpa is expressed in the 24 hpf embryo brain, eyes, and tail bud [41].

Additionally, preventing protein synthesis by targeting RNA caused embryonic lethality by 18 hpf [41]. Thus, we hypothesized that the protein Ttpa is required prior to 18 hpf and likely in the developing neural structures. We then found that a VitE deficient zebrafish expressed

110 ttpa in the brain ventricle borders, and although not different in abundance relative to an E+ embryo, the brain was severely deformed with over- and under inflated brain ventricles [167].

The RedEfish provides an opportunity to visualize the structures requiring VitE in vivo. Cells expressing Ttpa likely require VitE or deliver VitE to neighboring cells due to the limitations of embryonic lipoprotein metabolism and not well-established circulatory system prior to 24 hpf. The RedEfish should be thoroughly characterized by confocal microscopy to determine precisely what structures express Ttpa. Because pax2a and sox10, both markers of neurodevelopment, were also impaired by VitE status at 12- and 24 hpf, it is likely that cells that express those markers are VitE-dependent. Transgenic reporter lines can be used in- crossed with the RedEfish to demonstrate co-localization of two target proteins, Ttpa and neural crest cell-specific Sox10. When crossed, the RedE:Sox10 fish subject to a VitE sufficient and deficient diet will demonstrate that Ttpa localization is not affected by VitE status, however, the neighboring and subsequently VitE-dependent cells will be reduced or deformed. Further, the experiments performed throughout this thesis can be utilized to characterize both transcriptional and metabolic responses to those Ttpa-expressing cell types isolated from the embryo. I posit, however, that it is neighboring cells that require VitE rather than those expressing Ttpa proving another challenge. To overcome this, I hypothesize that instead cells that express HDL, LDL and lipoprotein receptors like scavenger receptor class B, member 1 (scarb1, also known as SR-B1) that binds early circulating lipoproteins in the developing nervous system may instead be important targets of VitE requirement. For example, the gene encoding for LDL receptor adaptor protein 1 (LDLRAP1) is expressed in a 24 hpf zebrafish in the midbrain-hindbrain boundary, dorsal hindbrain, and somites [269].

Earlier 10 hpf embryos express the ldlrap1 gene in the axial fin fold, also known as tailbud.

These regions are critically damaged given a VitE deficiency at the same developmental windows. Further, scarb1 and apoa1a are expressed at the onset of gastrulation at 5.25 hpf

111 in YSL and in the caudal neural tube by 24 hpf, indicating an early and prolonged need for

HDL metabolism throughout developing CNS [269-271]. Thus, I propose a more thorough examination of lipid metabolism in the early embryo to fully characterize the cause and effects of a VitE deficiency-induced death.

Final thoughts

Vitamin E is necessary for embryonic survival. This thesis elucidates that VitE acts within a network of metabolic measures to protect the developing embryonic nervous system from LPO. Up to 24 hpf, the embryo is composed of sensitive ectodermal tissue and its neural derivatives. Lethal knockdowns of TTPA and removal of VitE from the diet further support that VitE is necessary to those specific cells, tissue types, and early organs. The 24 hpf zebrafish is comparable to the mouse at E8 and human embryo within the first 3 weeks post- fertilization. Thus, it is critical to understand the effects of even a mild nutritional deficiency during a period prior implantation and possibly beyond notice.

112 BIBLIOGRAPHY

[1] H.M. Evans, K.S. Bishop, On the Existence of a Hitherto Unrecognized Dietary Factor Essential for Reproduction, Science 56(1458) (1922) 650-1. doi:10.1126/science.56.1458.650 [2] H.M. Evans, G.O. Burr, The anti-sterility vitamine fat soluble E, P Natl Acad Sci USA 11(6) (1925) 334. [3] K. Jishage, T. Tachibe, T. Ito, N. Shibata, S. Suzuki, T. Mori, T. Hani, H. Arai, H. Suzuki, Vitamin E is essential for mouse placentation but not for embryonic development itself, Biol Reprod 73(5) (2005) 983-7. doi:10.1095/biolreprod.105.043018 [4] P.L. Altman, D.S. Dittmer, Growth, including reproduction and morphological development, in: Committee on Biological Handbooks (Ed.) Biological Handbooks, Federation of American Societies for Experimental Biology, Washington, D.C., 1962, p. 608. [5] C.E. Steele, E.H. Jeffery, A.T. Diplock, The effect of vitamin E and synthetic antioxidants on the growth in vitro of explanted rat embryos, J Reprod Fertil 38(1) (1974) 115-23. doi:10.1530/jrf.0.0380115 [6] L. Mene-Saffrane, D. DellaPenna, Biosynthesis, regulation and functions of tocochromanols in plants, Plant Physiol Biochem 48(5) (2010) 301-9. doi:10.1016/j.plaphy.2009.11.004 [7] B. Li, D.A. Pratt, Methods for determining the efficacy of radical-trapping antioxidants, Free Radic Biol Med 82 (2015) 187-202. doi:10.1016/j.freeradbiomed.2015.01.020 [8] X. Leng, J.J. Kinnun, D. Marquardt, M. Ghefli, N. Kucerka, J. Katsaras, J. Atkinson, T.A. Harroun, S.E. Feller, S.R. Wassall, alpha-Tocopherol Is Well Designed to Protect Polyunsaturated Phospholipids: MD Simulations, Biophys J 109(8) (2015) 1608-18. doi:10.1016/j.bpj.2015.08.032 [9] M.G. Traber, G.R. Buettner, R.S. Bruno, The relationship between vitamin C status, the gut-liver axis, and metabolic syndrome, Redox Biol 21 (2019) 101091. doi:10.1016/j.redox.2018.101091 [10] M.G. Traber, J.F. Stevens, Vitamins C and E: beneficial effects from a mechanistic perspective, Free Radic Biol Med 51(5) (2011) 1000-13. doi:10.1016/j.freeradbiomed.2011.05.017 [11] S. Lemaire-Ewing, C. Desrumaux, D. Neel, L. Lagrost, Vitamin E transport, membrane incorporation and cell metabolism: Is alpha-tocopherol in lipid rafts an oar in the lifeboat?, Mol Nutr Food Res 54(5) (2010) 631-40. doi:10.1002/mnfr.200900445 [12] J. Atkinson, T. Harroun, S.R. Wassall, W. Stillwell, J. Katsaras, The location and behavior of alpha-tocopherol in membranes, Mol Nutr Food Res 54(5) (2010) 641-51. doi:10.1002/mnfr.200900439 [13] A.C. Howard, A.K. McNeil, P.L. McNeil, Promotion of plasma membrane repair by vitamin E, Nat Commun 2 (2011) 597. doi:10.1038/ncomms1594 [14] J.M. Zingg, Vitamin E: Regulatory Role on Signal Transduction, IUBMB Life 71(4) (2019) 456-478. doi:10.1002/iub.1986 [15] A. AZZI, R. GYSIN, P. KEMPNÁ, A. MUNTEANU, Y. NEGIS, L. VILLACORTA, T. VISARIUS, J.M. ZINGG, Vitamin E mediates cell signaling and regulation of gene expression, Annals of the New York Academy of Sciences 1031(1) (2004) 86-95. [16] J.-M. Zingg, A. Azzi, Non-antioxidant activities of vitamin E, Current medicinal chemistry 11(9) (2004) 1113-1133. [17] M.G. Traber, J. Atkinson, Vitamin E, antioxidant and nothing more, Free Radic Biol Med 43(1) (2007) 4-15. doi:10.1016/j.freeradbiomed.2007.03.024

113 [18] S. Chung, M. Ghelfi, J. Atkinson, R. Parker, J. Qian, C. Carlin, D. Manor, Vitamin E and Phosphoinositides Regulate the Intracellular Localization of the Hepatic alpha-Tocopherol Transfer Protein, J Biol Chem 291(33) (2016) 17028-39. doi:10.1074/jbc.M116.734210 [19] A.D. Cooper, Hepatic uptake of chylomicron remnants, J Lipid Res 38(11) (1997) 2173- 92. [20] L.F. Yevenes, A. Klein, J.F. Castro, T. Marin, N. Leal, F. Leighton, A.R. Alvarez, S. Zanlungo, Lysosomal vitamin E accumulation in Niemann-Pick type C disease, Biochimica et biophysica acta 1822(2) (2012) 150-60. doi:10.1016/j.bbadis.2011.11.009 [21] M.G. Traber, G.W. Burton, K.U. Ingold, H.J. Kayden, RRR- and SRR-alpha-tocopherols are secreted without discrimination in human chylomicrons, but RRR-alpha-tocopherol is preferentially secreted in very low density lipoproteins, J Lipid Res 31(4) (1990) 675-85. [22] J. Qian, S. Morley, K. Wilson, P. Nava, J. Atkinson, D. Manor, Intracellular trafficking of vitamin E in hepatocytes: the role of tocopherol transfer protein, J Lipid Res 46(10) (2005) 2072-82. doi:10.1194/jlr.M500143-JLR200 [23] M.G. Traber, R.J. Sokol, G.W. Burton, K.U. Ingold, A.M. Papas, J.E. Huffaker, H.J. Kayden, Impaired ability of patients with familial isolated vitamin E deficiency to incorporate alpha-tocopherol into lipoproteins secreted by the liver, J Clin Invest 85(2) (1990) 397-407. doi:10.1172/JCI114452 [24] M.G. Traber, J.C. Lane, N.R. Lagmay, H.J. Kayden, Studies on the transfer of tocopherol between lipoproteins, Lipids 27(9) (1992) 657-63. doi:10.1007/BF02536020 [25] G.M. Kostner, K. Oettl, M. Jauhiainen, C. Ehnholm, H. Esterbauer, H. Dieplinger, Human plasma phospholipid transfer protein accelerates exchange/transfer of alpha- tocopherol between lipoproteins and cells, Biochem J 305 ( Pt 2) (1995) 659-67. doi:10.1042/bj3050659 [26] M.G. Traber, Mechanisms for the prevention of vitamin E excess, J Lipid Res 54(9) (2013) 2295-306. doi:10.1194/jlr.R032946 [27] T.J. Sontag, R.S. Parker, Cytochrome P450 omega-hydroxylase pathway of tocopherol catabolism. Novel mechanism of regulation of vitamin E status, J Biol Chem 277(28) (2002) 25290-6. doi:10.1074/jbc.M201466200 [28] T.J. Sontag, R.S. Parker, Influence of major structural features of tocopherols and tocotrienols on their omega-oxidation by tocopherol-omega-hydroxylase, J Lipid Res 48(5) (2007) 1090-8. doi:10.1194/jlr.M600514-JLR200 [29] G.L. Catignani, An α-tocopherol binding protein in rat liver cytoplasm, Biochemical and biophysical research communications 67(1) (1975) 66-72. [30] G.L. Catignani, J.G. Bieri, Rat liver alpha-tocopherol binding protein, Biochimica et biophysica acta 497(2) (1977) 349-57. doi:10.1016/0304-4165(77)90192-1 [31] K. Jishage, M. Arita, K. Igarashi, T. Iwata, M. Watanabe, M. Ogawa, O. Ueda, N. Kamada, K. Inoue, H. Arai, H. Suzuki, Alpha-tocopherol transfer protein is important for the normal development of placental labyrinthine trophoblasts in mice, J Biol Chem 276(3) (2001) 1669-72. doi:10.1074/jbc.C000676200 [32] D.E. Kaempf-Rotzoll, K. Igarashi, J. Aoki, K. Jishage, H. Suzuki, H. Tamai, O. Linderkamp, H. Arai, Alpha-tocopherol transfer protein is specifically localized at the implantation site of pregnant mouse uterus, Biol Reprod 67(2) (2002) 599-604. doi:10.1095/biolreprod67.2.599 [33] A. Hosomi, K. Goto, H. Kondo, T. Iwatsubo, T. Yokota, M. Ogawa, M. Arita, J. Aoki, H. Arai, K. Inoue, Localization of alpha-tocopherol transfer protein in rat brain, Neurosci Lett 256(3) (1998) 159-62. doi:10.1016/s0304-3940(98)00785-x [34] K. Gohil, R. Godzdanker, E. O'Roark, B.C. Schock, R.R. Kaini, L. Packer, C.E. Cross, M.G. Traber, Alpha-tocopherol transfer protein deficiency in mice causes multi-organ

114 deregulation of gene networks and behavioral deficits with age, Ann N Y Acad Sci 1031 (2004) 109-26. doi:10.1196/annals.1331.012 [35] Y. Sato, H. Arai, A. Miyata, S. Tokita, K. Yamamoto, T. Tanabe, K. Inoue, Primary structure of alpha-tocopherol transfer protein from rat liver. Homology with cellular retinaldehyde-binding protein, J Biol Chem 268(24) (1993) 17705-10. [36] M. Arita, Y. Sato, A. Miyata, T. Tanabe, E. Takahashi, H.J. Kayden, H. Arai, K. Inoue, Human alpha-tocopherol transfer protein: cDNA cloning, expression and chromosomal localization, Biochem J 306 ( Pt 2) (1995) 437-43. doi:10.1042/bj3060437 [37] K. Ouahchi, M. Arita, H. Kayden, F. Hentati, M. Ben Hamida, R. Sokol, H. Arai, K. Inoue, J.L. Mandel, M. Koenig, Ataxia with isolated vitamin E deficiency is caused by mutations in the alpha-tocopherol transfer protein, Nat Genet 9(2) (1995) 141-5. doi:10.1038/ng0295-141 [38] K. Muller-Schmehl, J. Beninde, B. Finckh, S. Florian, J.W. Dudenhausen, R. Brigelius- Flohe, M. Schuelke, Localization of alpha-tocopherol transfer protein in trophoblast, fetal capillaries' endothelium and amnion epithelium of human term placenta, Free Radic Res 38(4) (2004) 413-20. doi:10.1080/10715760310001659611 [39] D.E. Rotzoll, R. Scherling, R. Etzl, H. Stepan, L.C. Horn, J.M. Poschl, Immunohistochemical localization of alpha-tocopherol transfer protein and lipoperoxidation products in human first-trimester and term placenta, Eur J Obstet Gynecol Reprod Biol 140(2) (2008) 183-91. doi:10.1016/j.ejogrb.2008.03.013 [40] E. Jauniaux, T. Cindrova-Davies, J. Johns, C. Dunster, J. Hempstock, F.J. Kelly, G.J. Burton, Distribution and transfer pathways of antioxidant molecules inside the first trimester human gestational sac, J Clin Endocrinol Metab 89(3) (2004) 1452-8. doi:10.1210/jc.2003- 031332 [41] G.W. Miller, L. Ulatowski, E.M. Labut, K.M. Lebold, D. Manor, J. Atkinson, C.L. Barton, R.L. Tanguay, M.G. Traber, The alpha-tocopherol transfer protein is essential for vertebrate embryogenesis, PLoS One 7(10) (2012) e47402. doi:10.1371/journal.pone.0047402 [42] N.I. Krinsky, G. Beecher, R. Burk, A. Chan, j.J. Erdman, R. Jacob, I. Jialal, L. Kolonel, J. Marshall, P.R. Taylor Mayne, Dietary reference intakes for vitamin C, vitamin E, selenium, and carotenoids, Institute of Medicine (2000). [43] M.G. Traber, A. Kamal-Eldin, Oxidative stress and vitamin E in anemia, Nutritional Anemia 155 (2007). [44] E.R. Monsen, Dietary reference intakes for the antioxidant nutrients: vitamin C, vitamin E, selenium, and carotenoids, J Am Diet Assoc 100(6) (2000) 637-40. doi:10.1016/S0002- 8223(00)00189-9 [45] R. Zamel, R. Khan, R.L. Pollex, R.A. Hegele, Abetalipoproteinemia: two case reports and literature review, Orphanet Journal of Rare Diseases 3(1) (2008) 19. doi:10.1186/1750- 1172-3-19 [46] I. Di Donato, S. Bianchi, A. Federico, Ataxia with vitamin E deficiency: update of molecular diagnosis, Neurological sciences : official journal of the Italian Neurological Society and of the Italian Society of Clinical Neurophysiology 31(4) (2010) 511-5. doi:10.1007/s10072-010-0261-1 [47] L. Cavalier, K. Ouahchi, H.J. Kayden, S. Di Donato, L. Reutenauer, J.-L. Mandel, M. Koenig, Ataxia with Isolated Vitamin E Deficiency: Heterogeneity of Mutations and Phenotypic Variability in a Large Number of Families, The American Journal of Human Genetics 62(2) (1998) 301-310. doi:10.1086/301699 [48] R.J. Sokol, Vitamin E deficiency and neurologic disease, Annu Rev Nutr 8 (1988) 351- 73. doi:10.1146/annurev.nu.08.070188.002031

115 [49] M. Schuelke, B. Finckh, E.A. Sistermans, M.G. Ausems, C. Hubner, A. von Moers, Ataxia with vitamin E deficiency: biochemical effects of malcompliance with vitamin E therapy, Neurology 55(10) (2000) 1584-6. doi:10.1212/wnl.55.10.1584 [50] M.G. Traber, R.J. Sokol, S.P. Ringel, H.E. Neville, C.A. Thellman, H.J. Kayden, Lack of tocopherol in peripheral nerves of vitamin E-deficient patients with peripheral neuropathy, N Engl J Med 317(5) (1987) 262-5. doi:10.1056/NEJM198707303170502 [51] R.J. Sokol, H.J. Kayden, D.B. Bettis, M.G. Traber, H. Neville, S. Ringel, W.B. Wilson, D.A. Stumpf, Isolated vitamin E deficiency in the absence of fat malabsorption--familial and sporadic cases: characterization and investigation of causes, J Lab Clin Med 111(5) (1988) 548-59. [52] R.I. Hislop, T.E. Whittle, Nutritional encephalomalacia in the chicken, Worlds Poult Sci J 23(2) (1967) 133-50. doi:10.1079/wps19670013 [53] B.S. Bains, A.R. Watson, Nutritional myopathy--a cause of ataxia in broiler chickens, N Z Vet J 26(1-2) (1978) 31-2. doi:10.1080/00480169.1978.34481 [54] L. Ulatowski, R. Parker, G. Warrier, R. Sultana, D.A. Butterfield, D. Manor, Vitamin E is essential for Purkinje neuron integrity, Neuroscience 260 (2014) 120-9. doi:10.1016/j.neuroscience.2013.12.001 [55] T. Yokota, K. Igarashi, T. Uchihara, K. Jishage, H. Tomita, A. Inaba, Y. Li, M. Arita, H. Suzuki, H. Mizusawa, H. Arai, Delayed-onset ataxia in mice lacking alpha -tocopherol transfer protein: model for neuronal degeneration caused by chronic oxidative stress, Proc Natl Acad Sci U S A 98(26) (2001) 15185-90. doi:10.1073/pnas.261456098 [56] J.S. Nelson, C.D. Fitch, V.W. Fischer, G.O. Broun, Jr., A.C. Chou, Progressive Neuropathologic Lesions in Vitamin E-Deficient Rhesus Monkeys1, Journal of Neuropathology & Experimental Neurology 40(2) (1981) 166-186. doi:10.1097/00005072- 198103000-00008 [57] A. Kohlschutter, B. Finckh, M. Nickel, A. Bley, C. Hubner, First Recognized Patient with Genetic Vitamin E Deficiency Stable after 36 Years of Controlled Supplement Therapy, Neurodegener Dis 20(1) (2020) 35-38. doi:10.1159/000508080 [58] P.P. Zandi, J.C. Anthony, A.S. Khachaturian, S.V. Stone, D. Gustafson, J.T. Tschanz, M.C. Norton, K.A. Welsh-Bohmer, J.C. Breitner, Reduced risk of Alzheimer disease in users of antioxidant vitamin supplements: the Cache County Study, Archives of neurology 61(1) (2004) 82-88. [59] L.A. Boothby, P.L. Doering, Vitamin C and vitamin E for Alzheimer's disease, Annals of Pharmacotherapy 39(12) (2005) 2073-2080. [60] A. Lloret, D. Esteve, P. Monllor, A. Cervera-Ferri, A. Lloret, The effectiveness of vitamin E treatment in Alzheimer’s disease, Int J Mol Sci 20(4) (2019) 879. [61] H. Wang, É.J. O'Reilly, M.G. Weisskopf, G. Logroscino, M.L. McCullough, A. Schatzkin, L.N. Kolonel, A. Ascherio, Vitamin E intake and risk of amyotrophic lateral sclerosis: a pooled analysis of data from 5 prospective cohort studies, American journal of epidemiology 173(6) (2011) 595-602. [62] A. Ascherio, M.G. Weisskopf, E.J. O'Reilly, E.J. Jacobs, M.L. McCullough, E.E. Calle, M. Cudkowicz, M.J. Thun, Vitamin E intake and risk of amyotrophic lateral sclerosis, Annals of Neurology: Official Journal of the American Neurological Association and the Child Neurology Society 57(1) (2005) 104-110. [63] V. Conte, K. Uryu, S. Fujimoto, Y. Yao, J. Rokach, L. Longhi, J.Q. Trojanowski, V.M. Lee, T.K. McIntosh, D. Pratico, Vitamin E reduces amyloidosis and improves cognitive function in Tg2576 mice following repetitive concussive brain injury, J Neurochem 90(3) (2004) 758-64. doi:10.1111/j.1471-4159.2004.02560.x

116 [64] W. Aiguo, Y. Zhe, F. Gomez-Pinilla, Vitamin E protects against oxidative damage and learning disability after mild traumatic brain injury in rats, Neurorehabil Neural Repair 24(3) (2010) 290-8. doi:10.1177/1545968309348318 [65] S.L. Cuddihy, S.S. Ali, E.S. Musiek, J. Lucero, S.J. Kopp, J.D. Morrow, L.L. Dugan, Prolonged alpha-tocopherol deficiency decreases oxidative stress and unmasks alpha- tocopherol-dependent regulation of mitochondrial function in the brain, J Biol Chem 283(11) (2008) 6915-24. doi:10.1074/jbc.M702572200 [66] A. Tran-Dinh, A. Levoye, D. Couret, L. Galle-Treger, M. Moreau, S. Delbosc, C. Hoteit, P. Montravers, P. Amarenco, T. Huby, O. Meilhac, High-Density Lipoprotein Therapy in Stroke: Evaluation of Endothelial SR-BI-Dependent Neuroprotective Effects, Int J Mol Sci 22(1) (2020). doi:10.3390/ijms22010106 [67] K. Gohil, S. Oommen, H.T. Quach, V.T. Vasu, H.H. Aung, B. Schock, C.E. Cross, G.T. Vatassery, Mice lacking alpha-tocopherol transfer protein gene have severe alpha- tocopherol deficiency in multiple regions of the central nervous system, Brain Res 1201 (2008) 167-76. doi:10.1016/j.brainres.2008.01.044 [68] N. Santander, C. Lizama, L. Murgas, S. Contreras, A.J.M. Martin, P. Molina, A. Quiroz, K. Rivera, F. Salas-Perez, A. Godoy, A. Rigotti, D. Busso, Transcriptional profiling of embryos lacking the lipoprotein receptor SR-B1 reveals a regulatory circuit governing a neurodevelopmental or metabolic decision during neural tube closure, BMC Genomics 19(1) (2018) 731. doi:10.1186/s12864-018-5110-2 [69] N. Santander, C. Lizama, M.J. Parga, A. Quiroz, D. Perez, G. Echeverria, L. Ulloa, V. Palma, A. Rigotti, D. Busso, Deficient Vitamin E Uptake During Development Impairs Neural Tube Closure in Mice Lacking Lipoprotein Receptor SR-BI, Sci Rep-Uk 7 (2017). doi:10.1038/s41598-017-05422-w [70] R.S. Shah, R. Rajalakshmi, R.V. Bhatt, M.N. Hazra, B.C. Patel, N.B. Swamy, T.V. Patel, Vitamin E status of the newborn in relation to gestational age, birth weight and maternal vitamin E status, Br J Nutr 58(2) (1987) 191-8. doi:10.1079/bjn19870086 [71] A. Gagné, S.Q. Wei, W.D. Fraser, P. Julien, Absorption, transport, and bioavailability of Vitamin E and its role in pregnant women, Journal of Obstetrics and Gynaecology Canada 31(3) (2009) 210-217. [72] G. Baydas, F. Karatas, M.F. Gursu, H.A. Bozkurt, N. Ilhan, A. Yasar, H. Canatan, Antioxidant Vitamin Levels in Term and Preterm Infants and Their Relation to Maternal Vitamin Status, Archives of Medical Research 33(3) (2002) 276-280. doi:10.1016/S0188- 4409(02)00356-9 [73] L.C. Chappell, P.T. Seed, A.L. Briley, F.J. Kelly, R. Lee, B.J. Hunt, K. Parmar, S.J. Bewley, A.H. Shennan, P.J. Steer, L. Poston, Effect of antioxidants on the occurrence of pre-eclampsia in women at increased risk: a randomised trial, Lancet 354(9181) (1999) 810-6. doi:10.1016/s0140-6736(99)80010-5 [74] R. Brigelius-Flohé, F.J. Kelly, J.T. Salonen, J. Neuzil, J.-M. Zingg, A. Azzi, The European perspective on vitamin E: current knowledge and future research, The American Journal of Clinical Nutrition 76(4) (2002) 703-716. doi:10.1093/ajcn/76.4.703 [75] T.O. Scholl, X. Chen, M. Sims, T.P. Stein, Vitamin E: maternal concentrations are associated with fetal growth, Am J Clin Nutr 84(6) (2006) 1442-8. doi:10.1093/ajcn/84.6.1442 [76] Y. Terasawa, Z. Ladha, S.W. Leonard, J.D. Morrow, D. Newland, D. Sanan, L. Packer, M.G. Traber, R.V. Farese, Jr., Increased atherosclerosis in hyperlipidemic mice deficient in alpha -tocopherol transfer protein and vitamin E, Proc Natl Acad Sci U S A 97(25) (2000) 13830-4. doi:10.1073/pnas.240462697

117 [77] K.M. Ranard, M.J. Kuchan, J.W. Erdman, Jr., Breeder Diet Strategies for Generating Ttpa-Null and Wild-Type Mice with Low Vitamin E Status to Assess Neurological Outcomes, Curr Dev Nutr 4(11) (2020) nzaa155. doi:10.1093/cdn/nzaa155 [78] K. Gohil, B.C. Schock, A.A. Chakraborty, Y. Terasawa, J. Raber, R.V. Farese, Jr., L. Packer, C.E. Cross, M.G. Traber, Gene expression profile of oxidant stress and neurodegeneration in transgenic mice deficient in alpha-tocopherol transfer protein, Free Radic Biol Med 35(11) (2003) 1343-54. doi:10.1016/s0891-5849(03)00509-4 [79] C.J. Finno, M.H. Bordbari, G. Gianino, B. Ming-Whitfield, E. Burns, J. Merkel, M. Britton, B. Durbin-Johnson, E.A. Sloma, M. McMackin, G. Cortopassi, V. Rivas, M. Barro, C.K. Tran, I. Gennity, H. Habib, L. Xu, B. Puschner, A.D. Miller, An innate immune response and altered nuclear receptor activation defines the spinal cord transcriptome during alpha-tocopherol deficiency in Ttpa-null mice, Free Radic Biol Med 120 (2018) 289- 302. doi:10.1016/j.freeradbiomed.2018.02.037 [80] G.E. Homanics, N. Maeda, M.G. Traber, H.J. Kayden, D.B. Dehart, K.K. Sulik, Exencephaly and hydrocephaly in mice with targeted modification of the apolipoprotein B (Apob) gene, Teratology 51(1) (1995) 1-10. doi:10.1002/tera.1420510102 [81] G.R. Garcia, P.D. Noyes, R.L. Tanguay, Advancements in zebrafish applications for 21st century toxicology, Pharmacol Ther 161 (2016) 11-21. doi:10.1016/j.pharmthera.2016.03.009 [82] M.C. Fishman, Zebrafish--the canonical vertebrate, Science 294(5545) (2001) 1290- 1291. [83] W. Driever, L. Solnica-Krezel, A. Schier, S. Neuhauss, J. Malicki, D. Stemple, D. Stainier, F. Zwartkruis, S. Abdelilah, Z. Rangini, A genetic screen for mutations affecting embryogenesis in zebrafish, Development 123(1) (1996) 37-46. [84] A. Amsterdam, R.M. Nissen, Z. Sun, E.C. Swindell, S. Farrington, N. Hopkins, Identification of 315 genes essential for early zebrafish development, Proceedings of the National Academy of Sciences 101(35) (2004) 12792-12797. [85] N. Gonsar, A. Coughlin, J.A. Clay-Wright, B.R. Borg, L.M. Kindt, J.O. Liang, Temporal and spatial requirements for Nodal-induced anterior mesendoderm and mesoderm in anterior neurulation, Genesis 54(1) (2016) 3-18. doi:10.1002/dvg.22908 [86] L.M. Kindt, A.R. Coughlin, T.R. Perosino, H.N. Ersfeld, M. Hampton, J.O. Liang, Identification of transcripts potentially involved in neural tube closure using RNA sequencing, Genesis 56(3) (2018). doi:ARTN e23096 10.1002/dvg.23096 [87] P. Ma, M.R. Swartz, L.M. Kindt, A.M. Kangas, J.O. Liang, Temperature Sensitivity of Neural Tube Defects in Zoep Mutants, Zebrafish 12(6) (2015) 448-56. doi:10.1089/zeb.2015.1113 [88] M.S. Lee, J.R. Bonner, D.J. Bernard, E.L. Sanchez, E.T. Sause, R.R. Prentice, S.M. Burgess, L.C. Brody, Disruption of the folate pathway in zebrafish causes developmental defects, BMC Dev Biol 12 (2012) 12. doi:10.1186/1471-213X-12-12 [89] L. Fang, Y.I. Miller, Emerging applications for zebrafish as a model organism to study oxidative mechanisms and their roles in inflammation and vascular accumulation of oxidized lipids, Free Radic Biol Med 53(7) (2012) 1411-20. doi:10.1016/j.freeradbiomed.2012.08.004 [90] M. van der Vaart, O. Svoboda, B.G. Weijts, R. Espin-Palazon, V. Sapp, T. Pietri, M. Bagnat, A.R. Muotri, D. Traver, Mecp2 regulates tnfa during zebrafish embryonic development and acute inflammation, Dis Model Mech 10(12) (2017) 1439-1451. doi:10.1242/dmm.026922

118 [91] C.Y. Usenko, S.L. Harper, R.L. Tanguay, Fullerene C60 exposure elicits an oxidative stress response in embryonic zebrafish, Toxicol Appl Pharmacol 229(1) (2008) 44-55. doi:10.1016/j.taap.2007.12.030 [92] C.Y. Usenko, S.L. Harper, R.L. Tanguay, In vivo evaluation of carbon fullerene toxicity using embryonic zebrafish, Carbon N Y 45(9) (2007) 1891-1898. doi:10.1016/j.carbon.2007.04.021 [93] C.B. Kimmel, R.D. Law, Cell lineage of zebrafish blastomeres. II. Formation of the yolk syncytial layer, Dev Biol 108(1) (1985) 86-93. doi:10.1016/0012-1606(85)90011-9 [94] C.B. Kimmel, R.D. Law, Cell lineage of zebrafish blastomeres. I. Cleavage pattern and cytoplasmic bridges between cells, Dev Biol 108(1) (1985) 78-85. doi:10.1016/0012- 1606(85)90010-7 [95] R.L. Miyares, V.B. de Rezende, S.A. Farber, Zebrafish yolk lipid processing: a tractable tool for the study of vertebrate lipid transport and metabolism, Dis Model Mech 7(7) (2014) 915-27. doi:10.1242/dmm.015800 [96] E.A. Ober, S. Schulte-Merker, Signals from the yolk cell induce mesoderm, neuroectoderm, the trunk organizer, and the notochord in zebrafish, Developmental biology 215(2) (1999) 167-181. [97] H.I. Sirotkin, S.T. Dougan, A.F. Schier, W.S. Talbot, Bozozok and squint act in parallel to specify dorsal mesoderm and anterior neuroectoderm in zebrafish, Development 127(12) (2000) 2583-2592. [98] C.B. Kimmel, W.W. Ballard, S.R. Kimmel, B. Ullmann, T.F. Schilling, Stages of embryonic development of the zebrafish, Dev Dyn 203(3) (1995) 253-310. doi:10.1002/aja.1002030302 [99] G.W. Miller, E.M. Labut, K.M. Lebold, A. Floeter, R.L. Tanguay, M.G. Traber, Zebrafish (Danio rerio) fed vitamin E-deficient diets produce embryos with increased morphologic abnormalities and mortality, J Nutr Biochem 23(5) (2012) 478-86. doi:10.1016/j.jnutbio.2011.02.002 [100] M. McDougall, J. Choi, H.K. Kim, G. Bobe, J.F. Stevens, E. Cadenas, R. Tanguay, M.G. Traber, Lethal dysregulation of energy metabolism during embryonic vitamin E deficiency, Free Radic Biol Med 104 (2017) 324-332. doi:10.1016/j.freeradbiomed.2017.01.020 [101] G.W. Miller, L. Truong, C.L. Barton, E.M. Labut, K.M. Lebold, M.G. Traber, R.L. Tanguay, The influences of parental diet and vitamin E intake on the embryonic zebrafish transcriptome, Comp Biochem Physiol Part D Genomics Proteomics 10 (2014) 22-9. doi:10.1016/j.cbd.2014.02.001 [102] L. Fang, C. Liu, Y.I. Miller, Zebrafish models of dyslipidemia: relevance to atherosclerosis and angiogenesis, Transl Res 163(2) (2014) 99-108. doi:10.1016/j.trsl.2013.09.004 [103] I. Rønnestad, K. Hamre, Ø. Lie, R. Waagbø, Ascorbic acid and α‐tocopherol levels in larvae of Atlantic halibut before and after exogenous feeding, Journal of fish biology 55(4) (1999) 720-731. [104] E. Kjorsvik, A. Reiersen, Histomorphology of the early yolk‐sac larvae of the Atlantic halibut (Hippoglossus hippoglossus L.)—an indication of the timing of functionality, Journal of Fish Biology 41(1) (1992) 1-19. [105] J. Chu, K.C. Sadler, New school in liver development: lessons from zebrafish, Hepatology 50(5) (2009) 1656-1663. [106] S. Korzh, A. Emelyanov, V. Korzh, Developmental analysis of ceruloplasmin gene and liver formation in zebrafish, Mech Develop 103(1) (2001) 137-139. doi:10.1016/S0925- 4773(01)00330-6

119 [107] L.S. Ross, T. Parrett, S. Easter, Axonogenesis and morphogenesis in the embryonic zebrafish brain, Journal of Neuroscience 12(2) (1992) 467-482. [108] R. Schmidt, U. Strahle, S. Scholpp, Neurogenesis in zebrafish - from embryo to adult, Neural Dev 8 (2013) 3. doi:10.1186/1749-8104-8-3 [109] E. Nikolopoulou, G.L. Galea, A. Rolo, N.D. Greene, A.J. Copp, Neural tube closure: cellular, molecular and biomechanical mechanisms, Development 144(4) (2017) 552-566. doi:10.1242/dev.145904 [110] Y.K. Bae, T. Shimizu, M. Hibi, Patterning of proneuronal and inter-proneuronal domains by hairy- and enhancer of split-related genes in zebrafish neuroectoderm, Development 132(6) (2005) 1375-85. doi:10.1242/dev.01710 [111] S. Jia, D. Wu, C. Xing, A. Meng, Smad2/3 activities are required for induction and patterning of the neuroectoderm in zebrafish, Dev Biol 333(2) (2009) 273-84. doi:10.1016/j.ydbio.2009.06.037 [112] I. Seiliez, B. Thisse, C. Thisse, FoxA3 and goosecoid promote anterior neural fate through inhibition of Wnt8a activity before the onset of gastrulation, Dev Biol 290(1) (2006) 152-63. doi:10.1016/j.ydbio.2005.11.021 [113] P. Betancur, M. Bronner-Fraser, T. Sauka-Spengler, Genomic code for Sox10 activation reveals a key regulatory enhancer for cranial neural crest, Proc Natl Acad Sci U S A 107(8) (2010) 3570-5. doi:10.1073/pnas.0906596107 [114] W. Tang, M.L. Martik, Y. Li, M.E. Bronner, Cardiac neural crest contributes to cardiomyocytes in amniotes and heart regeneration in zebrafish, Elife 8 (2019). doi:10.7554/eLife.47929 [115] J.M. Newbern, Molecular control of the neural crest and peripheral nervous system development, Curr Top Dev Biol 111 (2015) 201-31. doi:10.1016/bs.ctdb.2014.11.007 [116] H.C. Gibbs, A. Chang-Gonzalez, W. Hwang, A.T. Yeh, A.C. Lekven, Midbrain- Hindbrain Boundary Morphogenesis: At the Intersection of Wnt and Fgf Signaling, Front Neuroanat 11 (2017) 64. doi:10.3389/fnana.2017.00064 [117] A.H. Monsoro-Burq, PAX transcription factors in neural crest development, Semin Cell Dev Biol 44 (2015) 87-96. doi:10.1016/j.semcdb.2015.09.015 [118] M.N. McCarroll, Z.R. Lewis, M.D. Culbertson, B.L. Martin, D. Kimelman, A.V. Nechiporuk, Graded levels of Pax2a and Pax8 regulate cell differentiation during sensory placode formation, Development 139(15) (2012) 2740-50. doi:10.1242/dev.076075 [119] T. Shima, W. Znosko, M. Tsang, The characterization of a zebrafish mid-hindbrain mutant, mid-hindbrain gone (mgo), Dev Dyn 238(4) (2009) 899-907. doi:10.1002/dvdy.21916 [120] S. Krauss, T. Johansen, V. Korzh, U. Moens, J.U. Ericson, A. Fjose, Zebrafish pax[zf- a]: a paired box-containing gene expressed in the neural tube, EMBO J 10(12) (1991) 3609-19. [121] T.J. Carney, K.A. Dutton, E. Greenhill, M. Delfino-Machin, P. Dufourcq, P. Blader, R.N. Kelsh, A direct role for Sox10 in specification of neural crest-derived sensory neurons, Development 133(23) (2006) 4619-30. doi:10.1242/dev.02668 [122] C.E. Haldin, C. LaBonne, SoxE factors as multifunctional neural crest regulatory factors, Int J Biochem Cell Biol 42(3) (2010) 441-4. doi:10.1016/j.biocel.2009.11.014 [123] H. Akiyama, M.C. Chaboissier, J.F. Martin, A. Schedl, B. de Crombrugghe, The transcription factor Sox9 has essential roles in successive steps of the chondrocyte differentiation pathway and is required for expression of Sox5 and Sox6, Genes Dev 16(21) (2002) 2813-28. doi:10.1101/gad.1017802 [124] P.R. Brauer, T.H. Rosenquist, Effect of elevated homocysteine on cardiac neural crest migration in vitro, Dev Dyn 224(2) (2002) 222-30. doi:10.1002/dvdy.10105

120 [125] M.J. Boot, R.P. Steegers-Theunissen, R.E. Poelmann, L. Van Iperen, J. Lindemans, A.C. Gittenberger-de Groot, Folic acid and homocysteine affect neural crest and neuroepithelial cell outgrowth and differentiation in vitro, Dev Dyn 227(2) (2003) 301-8. doi:10.1002/dvdy.10303 [126] J. Eason, A.L. Williams, B. Chawla, C. Apsey, B.L. Bohnsack, Differences in neural crest sensitivity to ethanol account for the infrequency of anterior segment defects in the eye compared with craniofacial anomalies in a zebrafish model of fetal alcohol syndrome, Birth Defects Res 109(15) (2017) 1212-1227. doi:10.1002/bdr2.1069 [127] H.M. French, M. Reid, P. Mamontov, R.A. Simmons, J.B. Grinspan, Oxidative stress disrupts oligodendrocyte maturation, J Neurosci Res 87(14) (2009) 3076-87. doi:10.1002/jnr.22139 [128] R. Bajo-Graneras, D. Sanchez, G. Gutierrez, C. Gonzalez, S. Do Carmo, E. Rassart, M.D. Ganfornina, Apolipoprotein D alters the early transcriptional response to oxidative stress in the adult cerebellum, J Neurochem 117(6) (2011) 949-60. doi:10.1111/j.1471- 4159.2011.07266.x [129] P. Blader, U. Strahle, Ethanol impairs migration of the prechordal plate in the zebrafish embryo, Dev Biol 201(2) (1998) 185-201. doi:10.1006/dbio.1998.8995 [130] C.P. Heisenberg, C. Brennan, S.W. Wilson, Zebrafish aussicht mutant embryos exhibit widespread overexpression of ace (fgf8) and coincident defects in CNS development, Development 126(10) (1999) 2129-40. [131] T.T. Kao, C.Y. Chu, G.H. Lee, T.H. Hsiao, N.W. Cheng, N.S. Chang, B.H. Chen, T.F. Fu, Folate deficiency-induced oxidative stress contributes to neuropathy in young and aged zebrafish--implication in neural tube defects and Alzheimer's diseases, Neurobiol Dis 71 (2014) 234-44. doi:10.1016/j.nbd.2014.08.004 [132] J.Y. Chen, X. Tan, Z.H. Wang, Y.Z. Liu, J.F. Zhou, X.Z. Rong, L. Lu, Y. Li, The ribosome biogenesis protein Esf1 is essential for pharyngeal cartilage formation in zebrafish, FEBS J 285(18) (2018) 3464-3484. doi:10.1111/febs.14622 [133] C. Ciarlo, C.K. Kaufman, B. Kinikoglu, J. Michael, S. Yang, D.A. C, S. Blokzijl-Franke, J. den Hertog, T.M. Schlaeger, Y. Zhou, E. Liao, L.I. Zon, A chemical screen in zebrafish embryonic cells establishes that Akt activation is required for neural crest development, Elife 6 (2017). doi:10.7554/eLife.29145 [134] L. Barske, P. Rataud, K. Behizad, L. Del Rio, S.G. Cox, J.G. Crump, Essential Role of Nr2f Nuclear Receptors in Patterning the Vertebrate Upper Jaw, Dev Cell 44(3) (2018) 337- 347 e5. doi:10.1016/j.devcel.2017.12.022 [135] K.Y. Leung, Y.J. Pai, Q. Chen, C. Santos, E. Calvani, S. Sudiwala, D. Savery, M. Ralser, S.S. Gross, A.J. Copp, N.D.E. Greene, Partitioning of One-Carbon Units in Folate and Methionine Metabolism Is Essential for Neural Tube Closure, Cell Rep 21(7) (2017) 1795-1808. doi:10.1016/j.celrep.2017.10.072 [136] T.I. Chang, M. Horal, S.K. Jain, F. Wang, R. Patel, M.R. Loeken, Oxidant regulation of gene expression and neural tube development: Insights gained from diabetic pregnancy on molecular causes of neural tube defects, Diabetologia 46(4) (2003) 538-45. doi:10.1007/s00125-003-1063-2 [137] M.C. Fisher, S.H. Zeisel, M.H. Mar, T.W. Sadler, Perturbations in choline metabolism cause neural tube defects in mouse embryos in vitro, FASEB J 16(6) (2002) 619-21. doi:10.1096/fj.01-0564fje [138] M. McDougall, J. Choi, H.K. Kim, G. Bobe, J.F. Stevens, E. Cadenas, R. Tanguay, M.G. Traber, Lipid quantitation and metabolomics data from vitamin E-deficient and - sufficient zebrafish embryos from 0 to 120 hours-post-fertilization, Data Brief 11 (2017) 432- 441. doi:10.1016/j.dib.2017.02.046

121 [139] N. Santander, C. Lizama, M.J. Parga, A. Quiroz, D. Perez, G. Echeverria, L. Ulloa, V. Palma, A. Rigotti, D. Busso, Deficient vitamin E uptake during development impairs neural tube closure in mice lacking lipoprotein receptor SR-BI, Sci Rep 7(1) (2017) 5182. doi:ARTN 5182 10.1038/s41598-017-05422-w [140] T. Cecchini, P. Ferri, S. Ciaroni, R. Cuppini, P. Ambrogini, S. Papa, P. Del Grande, Postnatal proliferation of DRG non-neuronal cells in vitamin E-deficient rats, Anat Rec 256(2) (1999) 109-15. doi:10.1002/(SICI)1097-0185(19991001)256:2<109::AID- AR1>3.0.CO;2-T [141] D. Goti, Z. Balazs, U. Panzenboeck, A. Hrzenjak, H. Reicher, E. Wagner, R. Zechner, E. Malle, W. Sattler, Effects of lipoprotein lipase on uptake and transcytosis of low density lipoprotein (LDL) and LDL-associated alpha-tocopherol in a porcine in vitro blood-brain barrier model, J Biol Chem 277(32) (2002) 28537-44. doi:10.1074/jbc.M203989200 [142] S. Al Deeb, K. Al Moutaery, M. Arshaduddin, M. Tariq, Vitamin E decreases valproic acid induced neural tube defects in mice, Neurosci Lett 292(3) (2000) 179-82. doi:10.1016/s0304-3940(00)01457-9 [143] E. Sivan, E.A. Reece, Y.K. Wu, C.J. Homko, M. Polansky, M. Borenstein, Dietary vitamin E prophylaxis and diabetic embryopathy: morphologic and biochemical analysis, Am J Obstet Gynecol 175(4 Pt 1) (1996) 793-9. doi:10.1016/s0002-9378(96)80001-9 [144] N.D. Greene, A.J. Copp, Neural tube defects, Annu Rev Neurosci 37 (2014) 221-42. doi:10.1146/annurev-neuro-062012-170354 [145] N. Kono, H. Arai, α-Tocopherol Transfer Protein, in: E. Niki (Ed.), Vitamin E; Chemistry and Nutritional Benefits, Royal Society of Chemistry, Cambridge, UK, 2019, pp. 64-74. [146] D.P. Muller, J.K. Lloyd, O.H. Wolff, Vitamin E and neurological function, Lancet 1(8318) (1983) 225-8. doi:10.1016/s0140-6736(83)92598-9 [147] K.M. Lebold, D.B. Jump, G.W. Miller, C.L. Wright, E.M. Labut, C.L. Barton, R.L. Tanguay, M.G. Traber, Vitamin E deficiency decreases long-chain PUFA in zebrafish (Danio rerio), J Nutr 141(12) (2011) 2113-8. doi:10.3945/jn.111.144279 [148] M. McDougall, J. Choi, L. Truong, R. Tanguay, M.G. Traber, Vitamin E deficiency during embryogenesis in zebrafish causes lasting metabolic and cognitive impairments despite refeeding adequate diets, Free Radic Biol Med 110 (2017) 250-260. doi:10.1016/j.freeradbiomed.2017.06.012 [149] L. Truong, S.L. Harper, R.L. Tanguay, Evaluation of embryotoxicity using the zebrafish model, Methods Mol Biol 691 (2011) 271-9. doi:10.1007/978-1-60761-849-2_16 [150] L. Truong, D.M. Reif, L. St Mary, M.C. Geier, H.D. Truong, R.L. Tanguay, Multidimensional In Vivo Hazard Assessment Using Zebrafish, Toxicol Sci 137(1) (2014) 212-233. doi:10.1093/toxsci/kft235 [151] M.L. Markovich, N.V. Rizzuto, P.B. Brown, Diet affects spawning in zebrafish, Zebrafish 4(1) (2007) 69-74. doi:10.1089/zeb.2006.9993 [152] C. Thisse, B. Thisse, M.E. Halpern, J.H. Postlethwait, Goosecoid expression in neurectoderm and mesendoderm is disrupted in zebrafish cyclops gastrulas, Dev Biol 164(2) (1994) 420-9. doi:10.1006/dbio.1994.1212 [153] E.M. Otis, R. Brent, Equivalent ages in mouse and human embryos, Anat Rec 120(1) (1954) 33-63. doi:10.1002/ar.1091200104 [154] B. Thisse, S. Pflumio, M. Fürthauer, B. Loppin, V. Heyer, A. Degrave, R. Woehl, A. Lux, T. Steffan, X.Q. Charbonnier, C. Thisse, Expression of the zebrafish genome during embryogenesis, ZFIN Direct Data Submission, ZFIN Direct Data Submission, 2001.

122 [155] R.M. Dale, J. Topczewski, Identification of an evolutionarily conserved regulatory element of the zebrafish col2a1a gene, Dev Biol 357(2) (2011) 518-31. doi:10.1016/j.ydbio.2011.06.020 [156] S. Wopat, J. Bagwell, K.D. Sumigray, A.L. Dickson, L.F.A. Huitema, K.D. Poss, S. Schulte-Merker, M. Bagnat, Spine Patterning Is Guided by Segmentation of the Notochord Sheath, Cell Rep 22(8) (2018) 2026-2038. doi:10.1016/j.celrep.2018.01.084 [157] M. Hoshino, Neuronal subtype specification in the cerebellum and dorsal hindbrain, Dev Growth Differ 54(3) (2012) 317-26. doi:10.1111/j.1440-169X.2012.01330.x [158] K. Muguruma, A. Nishiyama, Y. Ono, H. Miyawaki, E. Mizuhara, S. Hori, A. Kakizuka, K. Obata, Y. Yanagawa, T. Hirano, Y. Sasai, Ontogeny-recapitulating generation and tissue integration of ES cell-derived Purkinje cells, Nat Neurosci 13(10) (2010) 1171-80. doi:10.1038/nn.2638 [159] J. Egawa, M.L. Pearn, B.P. Lemkuil, P.M. Patel, B.P. Head, Membrane lipid rafts and neurobiology: age-related changes in membrane lipids and loss of neuronal function, J Physiol 594(16) (2016) 4565-79. doi:10.1113/JP270590 [160] C. Yin, L. Solnica-Krezel, Convergence and extension movements mediate the specification and fate maintenance of zebrafish slow muscle precursors, Dev Biol 304(1) (2007) 141-55. doi:10.1016/j.ydbio.2006.12.030 [161] D.W. Raible, J.S. Eisen, Restriction of neural crest cell fate in the trunk of the embryonic zebrafish, Development 120(3) (1994) 495-503. [162] K. Gritsman, W.S. Talbot, A.F. Schier, Nodal signaling patterns the organizer, Development 127(5) (2000) 921-32. [163] T. Schwend, E.J. Loucks, D. Snyder, S.C. Ahlgren, Requirement of Npc1 and availability of cholesterol for early embryonic cell movements in zebrafish, J Lipid Res 52(7) (2011) 1328-44. doi:10.1194/jlr.M012377 [164] J.C.H. Ho, C.D. Hsiao, K. Kawakami, W.K.F. Tse, Triclosan (TCS) exposure impairs lipid metabolism in zebrafish embryos, Aquat Toxicol 173 (2016) 29-35. doi:10.1016/j.aquatox.2016.01.001 [165] X. Rong, Y. Zhou, Y. Liu, B. Zhao, B. Wang, C. Wang, X. Gong, P. Tang, L. Lu, Y. Li, C. Zhao, J. Zhou, Glutathione peroxidase 4 inhibits Wnt/beta-catenin signaling and regulates dorsal organizer formation in zebrafish embryos, Development 144(9) (2017) 1687-1697. doi:10.1242/dev.144261 [166] M.Q. McDougall, J. Choi, J.F. Stevens, L. Truong, R.L. Tanguay, M.G. Traber, Lipidomics and H2(18)O labeling techniques reveal increased remodeling of DHA- containing membrane phospholipids associated with abnormal locomotor responses in alpha-tocopherol deficient zebrafish (danio rerio) embryos, Redox Biol 8 (2016) 165-74. doi:10.1016/j.redox.2016.01.004 [167] B. Head, J. La Du, R.L. Tanguay, C. Kioussi, M.G. Traber, Vitamin E is necessary for zebrafish nervous system development, Sci Rep 10(1) (2020) 15028. doi:10.1038/s41598- 020-71760-x [168] V. Tixier, L. Bataille, C. Etard, T. Jagla, M. Weger, J.P. Daponte, U. Strahle, T. Dickmeis, K. Jagla, Glycolysis supports embryonic muscle growth by promoting myoblast fusion, Proc Natl Acad Sci U S A 110(47) (2013) 18982-7. doi:10.1073/pnas.1301262110 [169] J. Zhang, B. Head, S.W. Leonard, J. Choi, R.L. Tanguay, M.G. Traber, Vitamin E deficiency dysregulates thiols, amino acids and related molecules during zebrafish embryogenesis, Redox Biol 38 (2020) 101784. doi:10.1016/j.redox.2020.101784 [170] R.J. White, J.E. Collins, I.M. Sealy, N. Wali, C.M. Dooley, Z. Digby, D.L. Stemple, D.N. Murphy, K. Billis, T. Hourlier, A. Fullgrabe, M.P. Davis, A.J. Enright, E.M. Busch-

123 Nentwich, A high-resolution mRNA expression time course of embryonic development in zebrafish, Elife 6 (2017). doi:10.7554/eLife.30860 [171] M. Shimobayashi, M.N. Hall, Multiple amino acid sensing inputs to mTORC1, Cell Res 26(1) (2016) 7-20. doi:10.1038/cr.2015.146 [172] C. Garza-Lombo, P. Petrosyan, M. Tapia-Rodriguez, C. Valdovinos-Flores, M.E. Gonsebatt, Systemic L-buthionine-S-R-sulfoximine administration modulates glutathione homeostasis via NGF/TrkA and mTOR signaling in the cerebellum, Neurochem Int 121 (2018) 8-18. doi:10.1016/j.neuint.2018.10.007 [173] C.A. Kearns, A.M. Ravanelli, K. Cooper, B. Appel, Fbxw7 Limits Myelination by Inhibiting mTOR Signaling, J Neurosci 35(44) (2015) 14861-71. doi:10.1523/JNEUROSCI.4968-14.2015 [174] C. Milanese, C.R. Bombardieri, S. Sepe, S. Barnhoorn, C. Payan-Gomez, D. Caruso, M. Audano, S. Pedretti, W.P. Vermeij, R.M.C. Brandt, A. Gyenis, M.M. Wamelink, A.S. de Wit, R.C. Janssens, R. Leen, A.B.P. van Kuilenburg, N. Mitro, J.H.J. Hoeijmakers, P.G. Mastroberardino, DNA damage and transcription stress cause ATP-mediated redesign of metabolism and potentiation of anti-oxidant buffering, Nat Commun 10(1) (2019) 4887. doi:10.1038/s41467-019-12640-5 [175] M. Podda, C. Weber, M.G. Traber, L. Packer, Simultaneous determination of tissue tocopherols, tocotrienols, ubiquinols, and ubiquinones, J Lipid Res 37(4) (1996) 893-901. [176] Y. Liao, J. Wang, E.J. Jaehnig, Z. Shi, B. Zhang, WebGestalt 2019: gene set analysis toolkit with revamped UIs and APIs, Nucleic Acids Res 47(W1) (2019) W199-W205. doi:10.1093/nar/gkz401 [177] L. Ruzicka, D.G. Howe, S. Ramachandran, S. Toro, C.E. Van Slyke, Y.M. Bradford, A. Eagle, D. Fashena, K. Frazer, P. Kalita, P. Mani, R. Martin, S.T. Moxon, H. Paddock, C. Pich, K. Schaper, X. Shao, A. Singer, M. Westerfield, The Zebrafish Information Network: new support for non-coding genes, richer Gene Ontology annotations and the Alliance of Genome Resources, Nucleic Acids Res 47(D1) (2019) D867-D873. doi:10.1093/nar/gky1090 [178] Z. Pang, J. Chong, S. Li, J. Xia, MetaboAnalystR 3.0: Toward an Optimized Workflow for Global Metabolomics, Metabolites 10(5) (2020). doi:10.3390/metabo10050186 [179] C. Scheldeman, J.D. Mills, A. Siekierska, I. Serra, D. Copmans, A.M. Iyer, B.J. Whalley, J. Maes, A.C. Jansen, L. Lagae, E. Aronica, P.A.M. de Witte, mTOR-related neuropathology in mutant tsc2 zebrafish: Phenotypic, transcriptomic and pharmacological analysis, Neurobiol Dis 108 (2017) 225-237. doi:10.1016/j.nbd.2017.09.004 [180] C.E. Love, V.E. Prince, Expression and retinoic acid regulation of the zebrafish nr2f orphan nuclear receptor genes, Dev Dyn 241(10) (2012) 1603-15. doi:10.1002/dvdy.23838 [181] C. Thisse, B. Thisse, High-resolution in situ hybridization to whole-mount zebrafish embryos, Nat Protoc 3(1) (2008) 59-69. doi:10.1038/nprot.2007.514 [182] T. Tanimura, T.H. Shepard, Glucose metabolism by rat embryos in vitro, Proc Soc Exp Biol Med 135(1) (1970) 51-4. doi:10.3181/00379727-135-34985 [183] I. Motorykin, M.G. Traber, R.L. Tanguay, C.S. Maier, Proteome-driven elucidation of adaptive responses to combined vitamin E and C deficiency in zebrafish, J Proteome Res 13(3) (2014) 1647-56. doi:10.1021/pr401108d [184] A.A. Moazzami, R. Andersson, A. Kamal-Eldin, Changes in the metabolic profile of rat liver after alpha-tocopherol deficiency as revealed by metabolomics analysis, NMR Biomed 24(5) (2011) 499-505. doi:10.1002/nbm.1615 [185] K.M. Lebold, C.V. Lohr, C.L. Barton, G.W. Miller, E.M. Labut, R.L. Tanguay, M.G. Traber, Chronic vitamin E deficiency promotes vitamin C deficiency in zebrafish leading to

124 degenerative myopathy and impaired swimming behavior, Comp Biochem Physiol C Toxicol Pharmacol 157(4) (2013) 382-9. doi:10.1016/j.cbpc.2013.03.007 [186] C. Zhang, M. Skamagki, Z. Liu, A. Ananthanarayanan, R. Zhao, H. Li, K. Kim, Biological Significance of the Suppression of Oxidative Phosphorylation in Induced Pluripotent Stem Cells, Cell Rep 21(8) (2017) 2058-2065. doi:10.1016/j.celrep.2017.10.098 [187] P. Kraus, S. V, H.B. Yu, X. Xing, S.L. Lim, T. Adler, J.A. Pimentel, L. Becker, A. Bohla, L. Garrett, W. Hans, S.M. Holter, E. Janas, K. Moreth, C. Prehn, O. Puk, B. Rathkolb, J. Rozman, J. Adamski, R. Bekeredjian, D.H. Busch, J. Graw, M. Klingenspor, T. Klopstock, F. Neff, M. Ollert, T. Stoeger, A.O. Yildrim, O. Eickelberg, E. Wolf, W. Wurst, H. Fuchs, V. Gailus-Durner, M.H. de Angelis, T. Lufkin, L.W. Stanton, Pleiotropic functions for transcription factor zscan10, PLoS One 9(8) (2014) e104568. doi:10.1371/journal.pone.0104568 [188] K.J. Fishwick, R.A. Li, P. Halley, P. Deng, K.G. Storey, Initiation of neuronal differentiation requires PI3-kinase/TOR signalling in the vertebrate neural tube, Dev Biol 338(2) (2010) 215-25. doi:10.1016/j.ydbio.2009.12.001 [189] M. Kadoya, N. Sasai, Negative Regulation of mTOR Signaling Restricts Cell Proliferation in the Floor Plate, Front Neurosci 13 (2019) 1022. doi:10.3389/fnins.2019.01022 [190] K. Makky, J. Tekiela, A.N. Mayer, Target of rapamycin (TOR) signaling controls epithelial morphogenesis in the vertebrate intestine, Dev Biol 303(2) (2007) 501-13. doi:10.1016/j.ydbio.2006.11.030 [191] E.S. Mathews, B. Appel, Cholesterol Biosynthesis Supports Myelin Gene Expression and Axon Ensheathment through Modulation of P13K/Akt/mTor Signaling, J Neurosci 36(29) (2016) 7628-39. doi:10.1523/JNEUROSCI.0726-16.2016 [192] K. Blauth, S. Banerjee, M.A. Bhat, Axonal ensheathment and intercellular barrier formation in Drosophila, Int Rev Cell Mol Biol 283 (2010) 93-128. doi:10.1016/S1937- 6448(10)83003-5 [193] A. Woodhoo, L. Sommer, Development of the Schwann cell lineage: from the neural crest to the myelinated nerve, Glia 56(14) (2008) 1481-90. doi:10.1002/glia.20723 [194] P. Barraud, A.A. Seferiadis, L.D. Tyson, M.F. Zwart, H.L. Szabo-Rogers, C. Ruhrberg, K.J. Liu, C.V. Baker, Neural crest origin of olfactory ensheathing glia, Proc Natl Acad Sci U S A 107(49) (2010) 21040-5. doi:10.1073/pnas.1012248107 [195] H.F. McGraw, C.D. Snelson, A. Prendergast, A. Suli, D.W. Raible, Postembryonic neuronal addition in zebrafish dorsal root ganglia is regulated by Notch signaling, Neural Dev 7 (2012) 23. doi:10.1186/1749-8104-7-23 [196] D. Marchi, K. Santhakumar, E. Markham, N. Li, K.H. Storbeck, N. Krone, V.T. Cunliffe, F.J.M. van Eeden, Bidirectional crosstalk between Hypoxia-Inducible Factor and glucocorticoid signalling in zebrafish larvae, PLoS Genet 16(5) (2020) e1008757. doi:10.1371/journal.pgen.1008757 [197] N. Pescador, Y. Cuevas, S. Naranjo, M. Alcaide, D. Villar, M.O. Landazuri, L. Del Peso, Identification of a functional hypoxia-responsive element that regulates the expression of the egl nine homologue 3 (egln3/phd3) gene, Biochem J 390(Pt 1) (2005) 189-97. doi:10.1042/BJ20042121 [198] J.E. Packer, T.F. Slater, R.L. Willson, Direct observation of a free radical interaction between vitamin E and vitamin C, Nature 278(5706) (1979) 737-8. doi:10.1038/278737a0 [199] G. Drouin, J.-R. Godin, B. Pagé, The genetics of vitamin C loss in vertebrates, Current genomics 12(5) (2011) 371-378.

125 [200] M. Nishikimi, K. Yagi, Molecular basis for the deficiency in humans of gulonolactone oxidase, a key enzyme for ascorbic acid biosynthesis, The American journal of clinical nutrition 54(6) (1991) 1203S-1208S. [201] S. Hasselholt, P. Tveden-Nyborg, J. Lykkesfeldt, Distribution of vitamin C is tissue specific with early saturation of the brain and adrenal glands following differential oral dose regimens in guinea pigs, Br J Nutr 113(10) (2015) 1539-49. doi:10.1017/S0007114515000690 [202] P. Tveden-Nyborg, L. Vogt, J.G. Schjoldager, N. Jeannet, S. Hasselholt, M.D. Paidi, S. Christen, J. Lykkesfeldt, Maternal vitamin C deficiency during pregnancy persistently impairs hippocampal neurogenesis in offspring of guinea pigs, PLoS One 7(10) (2012) e48488. doi:10.1371/journal.pone.0048488 [203] M.E. Meredith, F.E. Harrison, J.M. May, Differential regulation of the ascorbic acid transporter SVCT2 during development and in response to ascorbic acid depletion, Biochem Biophys Res Commun 414(4) (2011) 737-42. doi:10.1016/j.bbrc.2011.09.146 [204] C. D'Aniello, F. Cermola, A. Palamidessi, L.G. Wanderlingh, M. Gagliardi, A. Migliaccio, F. Varrone, L. Casalino, M.R. Matarazzo, D. De Cesare, G. Scita, E.J. Patriarca, G. Minchiotti, Collagen Prolyl Hydroxylation-Dependent Metabolic Perturbation Governs Epigenetic Remodeling and Mesenchymal Transition in Pluripotent and Cancer Cells, Cancer Res 79(13) (2019) 3235-3250. doi:10.1158/0008-5472.CAN-18-2070 [205] S. Bretaud, P. Nauroy, M. Malbouyres, F. Ruggiero, Fishing for collagen function: About development, regeneration and disease, Seminars in cell & developmental biology, Elsevier, 2019, pp. 100-108. [206] C.A. Henry, B.D. Crawford, Y.L. Yan, J. Postlethwait, M.S. Cooper, M.B. Hille, Roles for zebrafish focal adhesion kinase in notochord and somite morphogenesis, Dev Biol 240(2) (2001) 474-87. doi:10.1006/dbio.2001.0467 [207] R.F. Li, T.Y. Wu, Y.Z. Mou, Y.S. Wang, C.L. Chen, C.Y. Wu, Nr2f1b control venous specification and angiogenic patterning during zebrafish vascular development, J Biomed Sci 22 (2015) 104. doi:10.1186/s12929-015-0209-0 [208] D.N. Meyer, B.B. Baker, T.R. Baker, Ancestral TCDD Exposure Induces Multigenerational Histologic and Transcriptomic Alterations in Gonads of Male Zebrafish, Toxicol Sci 164(2) (2018) 603-612. doi:10.1093/toxsci/kfy115 [209] M. Iribarne, I. Masai, Neurotoxicity of cGMP in the vertebrate retina: from the initial research on rd mutant mice to zebrafish genetic approaches, J Neurogenet 31(3) (2017) 88-101. doi:10.1080/01677063.2017.1358268 [210] K. Iwasa, K. Shima, K. Komai, Y. Nishida, T. Yokota, M. Yamada, Retinitis pigmentosa and macular degeneration in a patient with ataxia with isolated vitamin E deficiency with a novel c.717 del C mutation in the TTPA gene, J Neurol Sci 345(1-2) (2014) 228-30. doi:10.1016/j.jns.2014.07.001 [211] C.J. Finno, J. Peterson, M. Kang, S. Park, M.H. Bordbari, B. Durbin-Johnson, M. Settles, M.C. Perez-Flores, J.H. Lee, E.N. Yamoah, Single-Cell RNA-seq Reveals Profound Alterations in Mechanosensitive Dorsal Root Ganglion Neurons with Vitamin E Deficiency, iScience 21 (2019) 720-735. doi:10.1016/j.isci.2019.10.064 [212] A. Fischer, J. Pallauf, K. Gohil, S.U. Weber, L. Packer, G. Rimbach, Effect of selenium and vitamin E deficiency on differential gene expression in rat liver, Biochem Biophys Res Commun 285(2) (2001) 470-5. doi:10.1006/bbrc.2001.5171 [213] P.F. Surai, R.C. Noble, B.K. Speake, Tissue-specific differences in antioxidant distribution and susceptibility to lipid peroxidation during development of the chick embryo, Biochimica et biophysica acta 1304(1) (1996) 1-10. doi:10.1016/s0005-2760(96)00099-9

126 [214] G.T. Vatassery, C.K. Angerhofer, F.J. Peterson, Vitamin E concentrations in the brains and some selected peripheral tissues of selenium-deficient and vitamin E-deficient mice, J Neurochem 42(2) (1984) 554-8. doi:10.1111/j.1471-4159.1984.tb02713.x [215] K.M. Lebold, J.S. Kirkwood, A.W. Taylor, J. Choi, C.L. Barton, G.W. Miller, J. La Du, D.B. Jump, J.F. Stevens, R.L. Tanguay, M.G. Traber, Novel liquid chromatography-mass spectrometry method shows that vitamin E deficiency depletes arachidonic and docosahexaenoic acids in zebrafish (Danio rerio) embryos, Redox Biol 2 (2013) 105-13. doi:10.1016/j.redox.2013.12.007 [216] A.A. Moazzami, S. Frank, A. Gombert, N. Sus, B. Bayram, G. Rimbach, J. Frank, Non-targeted 1 H-NMR-metabolomics suggest the induction of master regulators of energy metabolism in the liver of vitamin E-deficient rats, Food & function 6(4) (2015) 1090-1097. [217] S.L. Yang, S.S. Aw, C. Chang, S. Korzh, V. Korzh, J. Peng, Depletion of Bhmt elevates sonic hedgehog transcript level and increases beta-cell number in zebrafish, Endocrinology 152(12) (2011) 4706-17. doi:10.1210/en.2011-1306 [218] S. Prabhudesai, C. Koceja, A. Dey, S. Eisa-Beygi, N.R. Leigh, R. Bhattacharya, P. Mukherjee, R. Ramchandran, Cystathionine beta-synthase is necessary for axis development in vivo, Front Cell Dev Biol 6 (2018) 14. doi:10.3389/fcell.2018.00014 [219] M. Westerfield, The zebrafish book : a guide for the laboratory use of zebrafish (Brachydanio rerio), University of Oregon Press, Eugene, OR, 2007. [220] A.R. Timme-Laragy, J.V. Goldstone, B.R. Imhoff, J.J. Stegeman, M.E. Hahn, J.M. Hansen, Glutathione redox dynamics and expression of glutathione-related genes in the developing embryo, Free Radic Biol Med 65 (2013) 89-101. doi:10.1016/j.freeradbiomed.2013.06.011 [221] D. Jardine, M.K. Litvak, Direct yolk sac volume manipulation of zebrafish embryos and the relationship between offspring size and yolk sac volume, Journal of Fish Biology 63(2) (2003) 388-397. doi:10.1046/j.1095-8649.2003.00161.x [222] S. Uusi-Heikkilä, C. Wolter, T. Meinelt, R. Arlinghaus, Size-dependent reproductive success of wild zebrafish Danio rerio in the laboratory, Journal of Fish Biology 77(3) (2010) 552-569. doi:10.1111/j.1095-8649.2010.02698.x [223] T. Moore, A. Le, A.K. Niemi, T. Kwan, K. Cusmano-Ozog, G.M. Enns, T.M. Cowan, A new LC-MS/MS method for the clinical determination of reduced and oxidized glutathione from whole blood, J Chromatogr B Analyt Technol Biomed Life Sci 929 (2013) 51-5. doi:10.1016/j.jchromb.2013.04.004 [224] H. Prinsen, B.G.M. Schiebergen-Bronkhorst, M.W. Roeleveld, J.J.M. Jans, M.G.M. de Sain-van der Velden, G. Visser, P.M. van Hasselt, N.M. Verhoeven-Duif, Rapid quantification of underivatized amino acids in plasma by hydrophilic interaction liquid chromatography (HILIC) coupled with tandem mass-spectrometry, J Inherit Metab Dis 39(5) (2016) 651-660. doi:10.1007/s10545-016-9935-z [225] ICH Expert Working Group, ICH Harmonised Tripartite Guideline, Validation of analytical procedures: Text and methodology Q2(R1), International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use, 2005, p. 17. [226] I.T. Jolliffe, J. Cadima, Principal component analysis: a review and recent developments, Philos Trans A Math Phys Eng Sci 374(2065) (2016) 20150202. doi:10.1098/rsta.2015.0202 [227] D. Giustarini, I. Dalle-Donne, A. Milzani, P. Fanti, R. Rossi, Analysis of GSH and GSSG after derivatization with N-ethylmaleimide, Nat Protoc 8(9) (2013) 1660-9. doi:10.1038/nprot.2013.095

127 [228] L.S. New, E.C. Chan, Evaluation of BEH C18, BEH HILIC, and HSS T3 (C18) column chemistries for the UPLC-MS-MS analysis of glutathione, glutathione disulfide, and ophthalmic acid in mouse liver and human plasma, J Chromatogr Sci 46(3) (2008) 209-14. doi:10.1093/chromsci/46.3.209 [229] A. Vallverdu-Queralt, A. Verbaere, E. Meudec, V. Cheynier, N. Sommerer, Straightforward method to quantify GSH, GSSG, GRP, and hydroxycinnamic acids in wines by UPLC-MRM-MS, J Agric Food Chem 63(1) (2015) 142-9. doi:10.1021/jf504383g [230] C.J. DeLong, Y.J. Shen, M.J. Thomas, Z. Cui, Molecular distinction of phosphatidylcholine synthesis between the CDP-choline pathway and phosphatidylethanolamine methylation pathway, J Biol Chem 274(42) (1999) 29683-8. doi:10.1074/jbc.274.42.29683 [231] F. Ursini, M. Maiorino, Lipid peroxidation and ferroptosis: The role of GSH and GPx4, Free Radic Biol Med 152 (2020) 175-185. doi:10.1016/j.freeradbiomed.2020.02.027 [232] G.J. McBean, The transsulfuration pathway: a source of cysteine for glutathione in astrocytes, Amino Acids 42(1) (2012) 199-205. doi:10.1007/s00726-011-0864-8 [233] L. Barella, P.Y. Muller, M. Schlachter, W. Hunziker, E. Stocklin, V. Spitzer, N. Meier, S. de Pascual-Teresa, A.M. Minihane, G. Rimbach, Identification of hepatic molecular mechanisms of action of alpha-tocopherol using global gene expression profile analysis in rats, Biochimica et biophysica acta 1689(1) (2004) 66-74. doi:10.1016/j.bbadis.2004.02.002 [234] T. Sakaguchi, Y. Kikuchi, A. Kuroiwa, H. Takeda, D.Y. Stainier, The yolk syncytial layer regulates myocardial migration by influencing extracellular matrix assembly in zebrafish, Development 133(20) (2006) 4063-72. doi:10.1242/dev.02581 [235] S.A. Craig, Betaine in human nutrition, Am J Clin Nutr 80(3) (2004) 539-49. doi:10.1093/ajcn/80.3.539 [236] D.S. Bindels, L. Haarbosch, L. van Weeren, M. Postma, K.E. Wiese, M. Mastop, S. Aumonier, G. Gotthard, A. Royant, M.A. Hink, T.W.J. Gadella, mScarlet: a bright monomeric red fluorescent protein for cellular imaging, Nature Methods 14(1) (2017) 53-56. doi:10.1038/nmeth.4074 [237] M.R. Velez, Generation of new transgenic zebrafish lines for studying neuronal circuits underlying behavior in zebrafish, 2018. [238] P. Antinucci, R. Hindges, A crystal-clear zebrafish for in vivo imaging, Sci Rep 6 (2016) 29490. doi:10.1038/srep29490 [239] Y.A. Aksoy, D.T. Nguyen, S. Chow, R.S. Chung, G.J. Guillemin, N.J. Cole, D. Hesselson, Chemical reprogramming enhances homology-directed genome editing in zebrafish embryos, Commun Biol 2 (2019) 198. doi:10.1038/s42003-019-0444-0 [240] G.E. Truett, P. Heeger, R.L. Mynatt, A.A. Truett, J.A. Walker, M.L. Warman, Preparation of PCR-quality mouse genomic DNA with hot sodium hydroxide and tris (HotSHOT), Biotechniques 29(1) (2000) 52, 54. doi:10.2144/00291bm09 [241] J.H. Kim, S.R. Lee, L.H. Li, H.J. Park, J.H. Park, K.Y. Lee, M.K. Kim, B.A. Shin, S.Y. Choi, High cleavage efficiency of a 2A peptide derived from porcine teschovirus-1 in human cell lines, zebrafish and mice, PLoS One 6(4) (2011) e18556. doi:10.1371/journal.pone.0018556 [242] V.H. Quinlivan, S.A. Farber, Lipid Uptake, Metabolism, and Transport in the Larval Zebrafish, Front Endocrinol (Lausanne) 8 (2017) 319. doi:10.3389/fendo.2017.00319 [243] A. Schlegel, D.Y. Stainier, Microsomal triglyceride transfer protein is required for yolk lipid utilization and absorption of dietary lipids in zebrafish larvae, Biochemistry 45(51) (2006) 15179-87. doi:10.1021/bi0619268 [244] E. Marza, C. Barthe, M. Andre, L. Villeneuve, C. Helou, P.J. Babin, Developmental expression and nutritional regulation of a zebrafish gene homologous to mammalian

128 microsomal triglyceride transfer protein large subunit, Dev Dyn 232(2) (2005) 506-18. doi:10.1002/dvdy.20251 [245] M. Raabe, E. Kim, M. Veniant, L.B. Nielsen, S.G. Young, Using genetically engineered mice to understand apolipoprotein-B deficiency syndromes in humans, Proc Assoc Am Physicians 110(6) (1998) 521-30. [246] M. Raabe, L.M. Flynn, C.H. Zlot, J.S. Wong, M.M. Veniant, R.L. Hamilton, S.G. Young, Knockout of the abetalipoproteinemia gene in mice: reduced lipoprotein secretion in heterozygotes and embryonic lethality in homozygotes, Proc Natl Acad Sci U S A 95(15) (1998) 8686-91. doi:10.1073/pnas.95.15.8686 [247] G.W. Burton, M.G. Traber, Vitamin E: antioxidant activity, biokinetics, and bioavailability, Annu Rev Nutr 10 (1990) 357-82. doi:10.1146/annurev.nu.10.070190.002041 [248] S. Louwette, L. Regal, C. Wittevrongel, C. Thys, G. Vandeweeghde, E. Decuyper, P. Leemans, R. De Vos, C. Van Geet, J. Jaeken, K. Freson, NPC1 defect results in abnormal platelet formation and function: studies in Niemann-Pick disease type C1 patients and zebrafish, Hum Mol Genet 22(1) (2013) 61-73. doi:10.1093/hmg/dds401 [249] W.C. Tseng, H.E. Loeb, W. Pei, C.H. Tsai-Morris, L. Xu, C.V. Cluzeau, C.A. Wassif, B. Feldman, S.M. Burgess, W.J. Pavan, F.D. Porter, Modeling Niemann-Pick disease type C1 in zebrafish: a robust platform for in vivo screening of candidate therapeutic compounds, Dis Model Mech 11(9) (2018). doi:10.1242/dmm.034165 [250] L. Ulatowski, R. Parker, C. Davidson, N. Yanjanin, T.J. Kelley, D. Corey, J. Atkinson, F. Porter, H. Arai, S.U. Walkley, D. Manor, Altered vitamin E status in Niemann-Pick type C disease, J Lipid Res 52(7) (2011) 1400-10. doi:10.1194/jlr.M015560 [251] H. Fechner, M. Schlame, F. Guthmann, P.A. Stevens, B. Rustow, α-and δ-tocopherol induce expression of hepatic α-tocopherol-transfer-protein mRNA, Biochemical Journal 331(2) (1998) 577-581. [252] V. Thakur, S. Morley, D. Manor, Hepatic alpha-tocopherol transfer protein: ligand- induced protection from proteasomal degradation, Biochemistry 49(43) (2010) 9339-44. doi:10.1021/bi100960b [253] M.J. Taormina, E.A. Hay, R. Parthasarathy, Passive and Active Microrheology of the Intestinal Fluid of the Larval Zebrafish, Biophys J 113(4) (2017) 957-965. doi:10.1016/j.bpj.2017.06.069 [254] C.M. Whipps, L.G. Moss, D.M. Sisk, K.N. Murray, D.M. Tobin, J.B. Moss, Detection of autofluorescent Mycobacterium chelonae in living zebrafish, Zebrafish 11(1) (2014) 76-82. doi:10.1089/zeb.2012.0863 [255] K. Fukui, K. Nakamura, M. Shirai, A. Hirano, H. Takatsu, S. Urano, Long-term vitamin E-deficient mice exhibit cognitive dysfunction via elevation of brain oxidation, Journal of nutritional science and vitaminology 61(5) (2015) 362-368. [256] Z. Liu, O. Chen, J.B.J. Wall, M. Zheng, Y. Zhou, L. Wang, H. Ruth Vaseghi, L. Qian, J. Liu, Systematic comparison of 2A peptides for cloning multi-genes in a polycistronic vector, Sci Rep-Uk 7(1) (2017) 2193. doi:10.1038/s41598-017-02460-2 [257] A. Chango, I.P. Pogribny, Considering maternal dietary modulators for epigenetic regulation and programming of the fetal epigenome, Nutrients 7(4) (2015) 2748-70. doi:10.3390/nu7042748 [258] X. Fan, T. Hou, S. Zhang, Y. Guan, J. Jia, Z. Wang, The cellular responses of autophagy, apoptosis, and 5-methylcytosine level in zebrafish cells upon nutrient deprivation stress, Chemosphere 241 (2020) 124989. doi:10.1016/j.chemosphere.2019.124989

129 [259] D. Bhattacharya, A.P. Azambuja, M. Simoes-Costa, Metabolic Reprogramming Promotes Neural Crest Migration via Yap/Tead Signaling, Dev Cell 53(2) (2020) 199-211 e6. doi:10.1016/j.devcel.2020.03.005 [260] J.A. Farrell, Y. Wang, S.J. Riesenfeld, K. Shekhar, A. Regev, A.F. Schier, Single-cell reconstruction of developmental trajectories during zebrafish embryogenesis, Science 360(6392) (2018). doi:10.1126/science.aar3131 [261] C.B. Kimmel, R.M. Warga, T.F. Schilling, Origin and organization of the zebrafish fate map, Development 108(4) (1990) 581-94. [262] M. Erecinska, S. Cherian, I.A. Silver, Energy metabolism in mammalian brain during development, Progress in neurobiology 73(6) (2004) 397-445. [263] Y. Song, M.A. Selak, C.T. Watson, C. Coutts, P.C. Scherer, J.A. Panzer, S. Gibbs, M.O. Scott, G. Willer, R.G. Gregg, D.W. Ali, M.J. Bennett, R.J. Balice-Gordon, Mechanisms underlying metabolic and neural defects in zebrafish and human multiple acyl-CoA dehydrogenase deficiency (MADD), PLoS One 4(12) (2009) e8329. doi:10.1371/journal.pone.0008329 [264] J.L. Anderson, J.D. Carten, S.A. Farber, Zebrafish lipid metabolism: from mediating early patterning to the metabolism of dietary fat and cholesterol, Methods Cell Biol 101 (2011) 111-41. doi:10.1016/B978-0-12-387036-0.00005-0 [265] K. Switon, K. Kotulska, A. Janusz-Kaminska, J. Zmorzynska, J. Jaworski, Molecular neurobiology of mTOR, Neuroscience 341 (2017) 112-153. doi:10.1016/j.neuroscience.2016.11.017 [266] M. Murakami, T. Ichisaka, M. Maeda, N. Oshiro, K. Hara, F. Edenhofer, H. Kiyama, K. Yonezawa, S. Yamanaka, mTOR is essential for growth and proliferation in early mouse embryos and embryonic stem cells, Mol Cell Biol 24(15) (2004) 6710-8. doi:10.1128/MCB.24.15.6710-6718.2004 [267] K.E. Hentges, B. Sirry, A.C. Gingeras, D. Sarbassov, N. Sonenberg, D. Sabatini, A.S. Peterson, FRAP/mTOR is required for proliferation and patterning during embryonic development in the mouse, Proc Natl Acad Sci U S A 98(24) (2001) 13796-801. doi:10.1073/pnas.241184198 [268] Y.G. Gangloff, M. Mueller, S.G. Dann, P. Svoboda, M. Sticker, J.F. Spetz, S.H. Um, E.J. Brown, S. Cereghini, G. Thomas, S.C. Kozma, Disruption of the mouse mTOR gene leads to early postimplantation lethality and prohibits embryonic stem cell development, Mol Cell Biol 24(21) (2004) 9508-16. doi:10.1128/MCB.24.21.9508-9516.2004 [269] B. Thisse, S. Pflumio, M. Fürthauer, B. Loppin, V. Heyer, A. Degrave, R. Woehl, A. Lux, T. Steffan, X.Q. Charbonnier, C. Thisse, Expression of the zebrafish genome during embryogenesis, ZFIN on-line publication, 2001. [270] P.J. Babin, C. Thisse, M. Durliat, M. Andre, M.A. Akimenko, B. Thisse, Both apolipoprotein E and A-I genes are present in a nonmammalian vertebrate and are highly expressed during embryonic development, Proc Natl Acad Sci U S A 94(16) (1997) 8622-7. doi:10.1073/pnas.94.16.8622 [271] J.P. Otis, E.M. Zeituni, J.H. Thierer, J.L. Anderson, A.C. Brown, E.D. Boehm, D.M. Cerchione, A.M. Ceasrine, I. Avraham-Davidi, H. Tempelhof, K. Yaniv, S.A. Farber, Zebrafish as a model for apolipoprotein biology: comprehensive expression analysis and a role for ApoA-IV in regulating food intake, Dis Model Mech 8(3) (2015) 295-309. doi:10.1242/dmm.018754 [272] Y.-L. Hong, S.-L. Yeh, C.-Y. Chang, M.-L. Hu, Total Plasma Malondialdehyde Levels in 16 Taiwanese College Students Determined by Various Thiobarbituric Acid Tests and an Improved High-Performance Liquid Chromatography-based Method, Clinical Biochemistry 33(8) (2000) 619-625. doi:10.1016/S0009-9120(00)00177-6

130 SUPPLEMENTARY

Supplementary Figure S1. Gene set enrichment analysis plots generated using WebGestalt

131

Supplementary Figure S2. RNA integrity numbers generated by the Center for Genome and Research Biocomputing

132

Supplementary Figure S3. Multi-dimensional scaling plots of E+ and E– log10CPM clustered at each time point prior to data filtration and normalization

133

Supplementary Figure S4. Quantitative analysis of other amino acids. Y-axes vary according to the concentration of each compound (pmol/embryo), X-axes shown times as 12, 24 and 48 hpf. Embryos (E+, E– or control; n=5 per group per time point) were analyzed using the AA protocol. Data shown are mean ± SEM; Y-axes vary according to the concentration of each compound (pmol/embryo). Symbols denote significant differences, as determined by Tukey post-hoc comparisons, at the time interval shown: E+ vs E–, no differences shown; control vs E– * P ≤ 0.05; control vs E+ # P ≤ 0.05. (A) Threonine (Interaction, P = 0.0974, Time P < 0.0001, Group P = 0.9076), (B) Alanine (Interaction, P = 0.2206, Time P < 0.0001, Group P = 0.7531), (C) Proline (Interaction, P = 0.2412, Time P < 0.0001, Group P = 0.0361), (D) Tyrosine (Interaction, P = 0.1992, Time P < 0.0001, Group P = 0.4684), (E) Valine (Interaction, P = 0.0592, Time P < 0.0001, Group P = 0.9817), (F) Isoleucine (Interaction, P = 0.4799, Time P < 0.0001, Group P = 0.4272), (G) Leucine (Interaction, P = 0.0765, Time P < 0.0001, Group P = 0.7684), (H) Phenylalanine (Interaction, P = 0.0935, Time P < 0.0001, Group P = 0.9031), (I) Lysine (Interaction, P = 0.1766, Time P < 0.0001, Group P = 0.4825), (J) Arginine (Interaction, P = 0.2019, Time P < 0.0001, Group P = 0.6090), (K) Histidine (Interaction, P = 0.2068, Time P < 0.0001, Group P = 0.8806).

134 Supplementary Table S1. Diet and embryo α- and γ-tocopherols measured by HPLC with electrochemical detection

α-tocopherol γ-tocopherol

Diet (mg/kg)

E+ 230 ± 8 16.4 ± 0.5

E– 6.3 ± 0.2 2.4 ± 0.1

Embryos (pmol/embryo)

E+ 68.7 ± 5.2 0.57 ± 0.05

E– 6.2 ± 0.2 0.11 ± 0.00

Shown are mean ± SEM.

135 Supplementary Table S2. MRM-parameters for the analysis of derivatized thiols and thiol disulfides1

Compound RT MRM-transition Collision

(min) (m/z) Energy (V)

Glutathione-NEM 1.98 433.1 > 304.1 16

Glutathione-13C2-15N-NEM 1.98 436.1 > 307.1 14

Cysteine-NEM 2.05 247.1 > 158.1 20

Homocysteine-NEM 2.01 261.3 > 126.0 25

Glutathione disulfide, GSSG 1.90 613.4 > 355.2 25

Cysteine disulfide, Cystine 0.88 241.0 > 74.0 20

1MRM-multiple reaction monitoring

136 Supplementary Table S3. MRM-parameters and labeled-internal standards used for the analysis of amino acids and related substances

Compound RT1 MRM-transition CE

Unlabeled (m/z) Labeled (m/z)

13 15 Methionine; C5-, N-Methionine 3.96 150 > 104 156 > 109 10

13 15 Tyrosine; C9-, N-Tyrosine 4.83 182 > 136 192 > 145 15

13 15 Phenylalanine; C9-, N- 3.07 166 > 120 176 > 129 15 Phenylalanine

13 15 Leucine; C6-, N-Leucine 3.18 132 > 86 139 > 92 10

13 15 Isoleucine ; C6-, N-Isoleucine 3.50 132 > 86 139 > 92 10

13 15 Threonine; C4-, N-Threonine 8.88 120 > 74 125 > 78 10

13 15 Valine; C5-, N-Valine 4.51 118 > 72 124 > 77 10

13 15 Proline; C5-, N-Proline 4.75 116 > 70 122 > 75 10

13 15 Alanine; C3-, N-Alanine 7.42 90 > 44 94 > 47 10

13 15 Lysine; C6-, N2-Lysine 12.4 147 > 84 155 > 90 15

13 15 Glutamate acid; C5-, N- 10.4 148 > 84 154 > 89 16 Glutamate

13 15 Arginine; C6-, N4-Arginine 12.0 175 > 70 185 > 75 20

13 15 Serine; C3-, N-Serine 9.74 10 6 > 60 110 > 63 10

Choline 2.26 104 > 60 17

Betaine 3.75 118 > 58 25

S‐adenosylmethionine 13.3 399 > 250 14

S‐adenosylhomocysteine 10.6 385 > 134 21

1RT: retention time (min), 2 CE: Collision Energy (V)

137 Supplementary Table S4. Linearity for the quantitative analysis of amino acids

Compound Linearity r2 (pmol/injected)

Threonine 0.10-1.56 0.999

Alanine 0.20-1.56 0.995

Tyrosine 0.39-1.56 0.999

Proline 0.10-1.56 0.999

Valine 0.10-1.56 0.999

Methionine 0.10-1.56 0.999

Isoleucine 0.10-1.56 0.999

Leucine 0.10-1.56 0.999

Phenylalanine 0.10-1.56 0.999

Lysine 0.10-1.56 0.999

Arginine 0.10-1.56 0.999

Histidine 0.10-1.56 0.999

Glutamate 0.10-1.56 0.997

Serine 0.20-1.56 0.993

SAH 0.041-5.19 0.996

SAM 0.13-4.19 0.992

Choline 0.015-1.96 0.998

Betaine 0.048-1.54 0.999

138 Supplementary Table S5. Within-day and between-day coefficients of variation (%) of thiol analytes1

Concentration Compound Range (pmol Within-day Between-day injected)

Glutathione-NEM 0.50 8.4% 6.9%

100 2.4% 2.5%

Cysteine-NEM 0.50 2.7% 7.2%

100 3.4% 3.5%

Homocysteine-NEM 0.50 8.7% 9.1%

100 5.3% 15.2%

Glutathione disulfide, GSSG 0.50 6.0% 6.7%

100 3.8% 8.2%

Cysteine disulfide, Cystine 0.5 8.4% 1.4%

100 0.6% 9.1%

1Three homogenate pools of zebrafish embryos (n= 10 embryos per analysis, 24 hpf) were used to calculate the within day (n=5 injections) % coefficient of variation (%CV = SD/mean x 100). A separate three pools were used to calculate %CV between days (n=3 analysis).

139 Supplementary Table S6. Within-day and between-day coefficients of variation (%) of amino acid analytes

Concentration Within-day Between- Compound range (pmol (n = 5a) day (n = 3b) injected)

Threonine 0.39 3.35%

0.78 11.5%

1.56 5.90% 9.40%

Alanine 0.20 26.3%

0.78 7.57%

1.56 2.80% 8.70%

Tyrosine 0.39 26.9%

0.78 8.10%

1.56 3.20% 0.80%

Proline 0.1 6.20%

0.39 10.3%

1.57 1.20% 8.90%

Valine 0.1 2.98%

0.39 9.54%

1.56 1.10% 5.50%

Methionine 0.1 2.83%

0.39 9.00%

1.56 1.80% 5.40%

Isoleucine 0.1 15.7%

0.39 8.44%

1.56 1.80% 6.09%

140 Concentration Within-day Between- Compound range (pmol (n = 5a) day (n = 3b) injected)

Leucine 0.1 12.8%

0.39 7.27%

1.56 1.70% 9.60%

Phenylalanine 0.1 5.50%

0.39 6.20%

1.56 3.10% 5.40%

Lysine 0.1 17.4%

0.39 8.40%

1.56 3.30% 11.8%

Arginine 0.1 7.20%

0.39 8.10%

1.56 1.90% 4.90%

Histidine 0.1 10.7%

0.39 17.6%

1.56 4.20% 10.0%

Glutamate 0.195 14.2%

0.39 19.3%

1.56 9.00% 9.00%

Serine 0.2 23.2%

0.39 9.12%

1.56 12.30% 16.1%

SAH 0.041 22.0%

141 Concentration Within-day Between- Compound range (pmol (n = 5a) day (n = 3b) injected)

1.30 21.6%

5.19 20.80% 16.1%

SAM 0.13 10.4%

1.05 1.54%

4.2 13.50% 2.48%

Choline 0.015 15.0%

0.21 29.6%

1.96 9.30% 30.7%

Betaine 0.096 26.6%

0.20 13.2%

1.54 12.60% 19.6% a Within-day assays were measured by injecting the highest concentration of the standard pool 5 times for analysis. b Between-day assays were measured by injecting different concentrations of the standard pool on three different days for analysis.

142

Supplementary Figure S5. Genotyping confirmation indicates successful genomic integration in heterozygote RedEfish. (A) PCR products generated from heterozygote (+/-) F3 RedEfish and Casper (WT) fin clips. Plasmid sequence used for positive control of mScarlet cassette. (B) Sanger sequencing confirmed no insertions or deletions 5’ of the mScarlet coding sequence (Kozak seq.) and in the first exon of ttpa.

143

Supplementary Figure S6. RFP expression confirmation in 3 days post-fertilization (dpf) c269, Casper and RedEfish. Anti-acetylated tubulin was used as a control for the assay. c269 embryos express red fluorescent protein (RFP) in their mid- and hindbrain indicated with *. Casper embryos do not express RFP, however, autofluorescence and non-specific binding of the antibody is found just posterior of the yolk sac at 3 dpf. RedEfish (noted here as mScarlet F3) express RFP in the pectoral fins, mid- and hindbrain, down the trunk and partially in the yolk syncytium (*) at 3 dpf.

144

Supplementary Figure S7. Malondialdehyde increased in E– embryos at 24 hours post- fertilization. Malondialdehyde (MDA) is presented as pmol per mg protein measured in E+ and E– samples (5 replicates per group, n=25 embryos per sample) ± SEM. MDA concentrations were measured in an aliquot of embryo homogenate according to the method by Hong et al. [272]. Quantitation was done using an external standard of 1,1,3,3- tetraethoxypropane (Sigma, St Louis, MO) prepared using the same method. Protein concentrations were measured by DC Protein Assay (Bio-Rad, Hercules, CA), per kit instructions. Absorbance was read at 750 nm on the BioTek Synergy HT (BioTek Instruments, Inc., Winooski, VT). A parametric t-test with Welch’s correction was used to compare data from the two groups.