University of California Riverside

UNIVERSITY OF CALIFORNIA RIVERSIDE

Mechanisms of Selenomethionine and Hypersaline Developmental Toxicity in Japanese Medaka (Oryzias latipes)

A Dissertation submitted in partial satisfaction of the requirements for the degree of

Doctor of Philosophy

Environmental Toxicology

Allison Justine Kupsco

August 2016

Dissertation Committee: Dr. Daniel Schlenk, Chairperson Dr. Morris Maduro Dr. Nicole zur Nieden

The Dissertation of Allison Justine Kupsco is approved:

Committee Chairperson

University of California, Riverside

Acknowledgements

This dissertation would not have been possible without the support of many people. First, my advisor, Dr. Dan Schlenk, was invaluable in not only his guidance and resources for this project, but also in providing me with the opportunities to grow and develop as a scientist. Without the time and effort he put into guiding me, this would not have been possible. In addition to advice about this dissertation, he also provided me with numerous networking and presentation opportunities through attendance of many conferences. I would also like to thank my committee members, Dr. Nicole zur Nieden and Dr. Morris Maduro for their support and advice. There have been many past and present labmates that have also aided in the dissertation through training, advice and emotional support. Dr. Aileen Maldonado, Dr. Lindley Maryoung, and Dr. Jordan Crago all participated in training me in lab techniques. Rafid Sikder was always eager to help and assisted in fish care and in data collected in Chapter 2. Luisa Becker Bertotto, Scott

Coffin, Marissa Giroux and Sara Vliet provided important emotional support and entertainment during long lab hours. And especially, I want to thank Graciel Diamante for her invaluable advice, commiseration and emotional support throughout our 4 years working together. Without her this would truly have not been possible.

I would like to acknowledge several collaborators and funding sources in conjunction with this work. Firstly, Rodney Johnson and Frank Whiteman at the USEPA

Mid-Continent Ecology Division Laboratory, Duluth, MN, provided the medaka cultures used in this research. David Lyons provided much technical assistance in chemical analysis. I would like to thank Greggory Britten, at University of California-Irvine for his

iv statistical guidance. I would also like to thank Roger Phillips and Eric Kingsley at

Monterey Bay Aquarium, Monterey, CA, for providing water samples and analysis for

Chapter 2.

This research was supported by the National Water Research Institute and

Southern California Salinity Coalition Fellowship, a National Research Service Award

Institutional Training Grant [2T32ES018827-06], the University of California-Riverside/

Agricultural Experiment Station Resource Allocation Program, the University of

Washington Superfund Basic Research Program [NIEHS P42ES04696], and the UCR

Dissertation Year Program Fellowship. Funding for attendance of numerous conferences was provided by the UCR Graduate Student Association Travel grants, Pollutant

Responses in Marine Organisms Student Travel Award, Society of Toxicology Student

Travel Award, Society of Environmental Toxicology and Chemistry North America

Student Travel Award, and T. Roy Fukuto Environmental Toxicology travel awards.

The text and figures in Chapter 1: Part C in part or in full, are a reprint of the material as it appears in “Oxidative Stress, Unfolded Protein Response, and Apoptosis in

Developmental Toxicity” published in International Review of Cellular and Molecular

Biology, Vol. 317, pages 1-66, 2015. The co-author Dr. Daniel Schlenk provided guidance and editing for this work.

The text and figures in Chapter 3 in part or in full, are a reprint of the material as it appears in “Stage Susceptibility of Japanese Medaka (Oryzias latipes) to

v Selenomethionine and Hypersaline Developmental Toxicity” published in Environmental

Toxicology and Chemistry, Vol. 35, pages 1247-1256, 2016. The co-author Dr. Daniel

Schlenk directed and supervised this research.

The text and figures in Chapter 4 in part or in full, are a reprint of the material as it appears in “Mechanisms of Selenomethionine Developmental Toxicity and the Impacts of Combined Hypersaline Conditions on Japanese Medaka (Oryzias latipes)” published in Environmental Science and Technology, Vol. 48, pages 7062-7068, 2014. The co- author Dr. Daniel Schlenk directed and supervised this research.

vi Dedication

This dissertation is dedicated to my fiancé Mike and my parents. Without their support this would not have been possible.

vii ABSTRACT OF THE DISSERTATION

Mechanisms of Selenomethionine and Hypersaline Developmental Toxicity in Japanese Medaka (Oryzias latipes)

Allison Justine Kupsco

Doctor of Philosophy, Graduate Program in Environmental Toxicology University of California, Riverside, August 2016 Dr. Daniel Schlenk, Chairperson

Selenium toxicity to oviparous vertebrates is often attributed to selenomethionine

(SeMet), which can maternally transfer to developing embryos. The mechanism of SeMet toxicity is unclear. Furthermore, salinity of fresh waterways is increasing due to climate change and anthropogenic disturbance. Hypersalinity can potentiate SeMet toxicity to

Japanese medaka (Oryzias latipes). The current study aimed to characterize the molecular mechanisms of SeMet and hypersalinity at sensitive developmental stages.

Developmental toxicity of seawater was compared to desalination brine (DSB) and artificial water based on the San Joaquin River (SJR), CA. DSB toxicity was equal to seawater, while SJR water was the most toxic to embryos and larvae. Flavin-containing monooxygenases (FMOs) initiate SeMet toxicity and are induced by hypersalinity.

However, developmental expression and regulation of FMOs in fish are unknown. Five putative medaka FMOs were identified with differential developmental mRNA expression patterns: two FMOs increased during mid-organogenesis; two FMOs decreased beginning at early organogenesis; and one FMO remained constant. Promoter

viii analysis indicated regulation by developmental factors and the UPR. Treatments with

UPR-inducer tunicamycin increased expression of two FMOs. In contrast, dithiothreitol inhibited the UPR and three FMOs, suggesting that FMOs are differentially regulated by the UPR. The developmental stage sensitivity of medaka embryos to SeMet was investigated in freshwater and SJR water. Stages 9-25 were most sensitive to SeMet; and hypersalinity potentiated SeMet toxicity during the onset of liver organogenesis, osmoregulation, and chondrogenesis. The mechanisms behind the potentiation of SeMet toxicity by hypersalinity indicated no involvement of oxidative stress or apoptosis; however, results suggested a role for the unfolded protein response (UPR) when animals were treated with 50µM SeMet for 12hrs. Mechanisms of SeMet-induced spinal deformities (5µM and 2.5µM for 24hrs) were further elucidated using imaging methods and showed increased oxidative stress and apoptosis in tails of embryos with spinal malformations. Gene expression analysis demonstrated a UPR activation pattern unique from UPR positive controls. Furthermore, these effects prematurely repressed chondrogenesis and induced osteogenesis. Overall, results will be useful for the risk assessment of hypersalinity and Se under hypersaline conditions; and inform studies on developmental mechanisms of toxicity.

ix Table of Contents

Chapter 1: Introduction 1 Selenium 1 Selenium Overview 1 Selenium in the Environment 4 Toxicity of Selenium to Fish 9 Selenium in the SJRV 18 Hypersalinity 21 Salinity Toxicity 21 Hypersalinity in Multiple Stressor Situations 25 Mechanisms of Developmental Toxicity 34 Oxidative Stress 35 The Unfolded Protein Response 52 Apoptosis 65 Integration of Oxidative Stress, the UPR and Apoptosis 72 Proposed Mechanism of Selenium and Hypersaline Toxicity 83 References 85 Chapter 2: Comparative Developmental Toxicity of Desalination Brine and 115 Sulfate Dominated Saltwater in a Euryhaline Fish

Abstract 115 Introduction 116 Methods 117 Results 121 Discussion 127 References 134 Supplemental Information 136

x Chapter 3: Flavin-containing Monooxygenase Developmental Expression 138 and Regulation by the Unfolded Protein Response in Japanese Medaka (Oryzias latipes)

Abstract 138 Introduction 139 Methods 141 Results 145 Discussion 152 References 158 Supplemental Information 162 Chapter 4: Stage Susceptibility of Japanese Medaka (Oryzias latipes) to 172 Selenomethionine and Hypersaline Developmental Toxicity

Abstract 172 Introduction 173 Methods 175 Results 181 Discussion 195 References 203 Chapter 5: Mechanisms of Selenomethionine Developmental Toxicity and 208 the Impacts of Combined Hypersaline Conditions on Japanese Medaka (Oryzias latipes)

Abstract 208 Introduction 209 Methods 212 Results 216 Discussion 220 References 227

xi Chapter 6: Molecular Mechanisms of Selenium-Induced Spinal 232 Deformities in Fish

Abstract 232 Introduction 233 Methods 235 Results 240 Discussion 248 References 255 Chapter 7: Conclusions 259 References 266

xii List of Figures

Figure 1.1. Selenium Metabolism. 3

Figure 1.2. Mosquito fish from Belews lake, NC. 10

Figure 1.3. Proposed mechanism of SeMet induced oxidative stress in the 15 presence of methioninase.

Figure 1.4. Oxidation of SeMet by FMO and reduction by GSH. 15

Figure 1.5. Mean salinity in May in A. Fort Point, CA, and B. Pittsburg, CA. 23

Figure 1.6. Endogenous formation of reactive oxygen species/reactive nitrogen 37 species.

Figure 1.7. Damage and protection from reactive oxygen species. 42

Figure 1.8. Redox-regulated transcription factors. 45

Figure 1.9. The three branches of the unfolded protein response. 53

Figure 1.10. Extrinsic and intrinsic apoptosis. 67

Figure 1.11. Apoptosis in physiological development. 68

Figure 1.12. Interplay between the unfolded protein response and redox 73 potential.

Figure 1.13. Oxidative stress-induced apoptosis. 76

Figure 1.14. Oxidative stress induced by the unfolded protein response. 79

Figure 1.15. Interplay between the unfolded protein response, oxidative stress, 83 and apoptosis in developmental toxicity.

Figure 2.1. Percent total deformities of Japanese medaka embryos following 123 treatment with San Joaquin River saltwater at full strength (13ppt), concentrated (17ppt), and diluted (9ppt), and with San Joaquin River water full strength (100%, 13ppt), mixed with seawater, to 75% (19ppt), 50% (24ppt) and 25% (30ppt).

xiii Figure 2.2. Percent swim bladder inflation in Japanese medaka embryos at 124 hatch following treatment with artificial seawater, RO reject mixed with artificial seawater, San Joaquin River water, and San Joaquin River water mixed with seawater.

Figure 2.3. Percent survival 3 days post hatch of Japanese medaka embryos 125 following treatment with San Joaquin River saltwater at full strength (13ppt), concentrated (17ppt), and diluted (9ppt) and with San Joaquin River water full strength (100%, 13ppt), mixed with seawater, to 75% (19ppt), 50% (24ppt) and 25% (30ppt).

Figure 2.4. Median day to hatch of Japanese medaka embryos treated with 126 artificial seawater (17, 35, 42, 56, and 70ppt corresponding to 50%, 100%, 130% 160% and 200% concentrated); and RO reject mixed with artificial seawater to 100% (50ppt), 75% (46ppt), 50% (42ppt) and 25% (39ppt).

Figure 2.S1. Percent hatch of medaka embryos following treatment with 136 artificial seawater, RO reject mixed with artificial seawater, San Joaquin River water, and San Joaquin River water mixed with seawater.

Figure 2.S2. Percent deformities of Japanese medaka embryos treated with 136 artificial seawater (17, 35, 42, 56, and 70ppth corresponding to 50%, 100%, 130% 160% and 200% concentrated); and RO reject mixed with artificial seawater to 100% (50ppth), 75% (46ppth), 50% (42ppth) and 25% (39ppth).

Figure 2.S3. Survival 3 days post hatch of Japanese medaka embryos treated 137 with artificial seawater (17, 35, 42, 56, and 70ppth corresponding to 50%, 100%, 130% 160% and 200% concentrated); and RO reject mixed with artificial seawater to 100% (50ppth), 75% (46ppth), 50% (42ppth) and 25% (39ppth).

Figure 2.S4. Median day to hatch in days post fertilization of Japanese medaka 137 embryos following treatment with San Joaquin River saltwater at full strength (13ppth), concentrated (17ppth), and diluted (9ppth) and with San Joaquin River water full strength (100%, 13ppth), mixed with seawater, to 75% (19ppth), 50% (24ppth) and 25% (30ppth).

Figure 3.1. Developmental expression of putative medaka FMOs on 1dpf, 147 3dpf, 6dpf and 9dpf expressed as fold change against 1dpf ± standard error (SE). A. FMO4, B. FMO5A, C. FMO5B, D. FMO5C, E. FMO5D.

Figure 3.2. Developmental expression of A. CHOP and B. BiP against 1dpf 149 embryos as fold change ± SE.

xiv Figure 3.3. Expression of UPR targets BiP and CHOP in 6dpf medaka embryos 150 following treatment with 0.02% DMSO, 1µg/ml Tm, 2µg/ml Tm, and 4µg/ml Tm for 24hrs. Expression of medaka FMOs in 6dpf embryos following treatment with 0.02% DMSO, 1µg/ml Tm, 2µg/ml Tm, and 4µg/ml Tm for 24hrs.

Figure 3.4. Expression of UPR targets BiP and CHOP in 6dpf medaka embryos 151 following treatment with Freshwater controls, 2mM DTT and 4mM DTT for 24hrs. Expression of medaka FMOs in 6dpf embryos following treatment with freshwater, 2mM DTT, and 4mM DTT for 24hrs.

Figure 3.S1. Protein Alignments of medaka FMOs by Clustal Omega. Validity 162 of alignments was assessed by TCS. FMO motifs are identified by black boxes.

Figure 4.1. Percent survival of embryos 24 h following 24 h of treatment with 182 0.5µM, 5µM, and 50µM SeMet in either freshwater or saltwater. Each graph represents one stage of treatment: 9, 17, 25, 29, 34 and 38.

Figure 4.2. Percent hatch following 24 h of treatment with 0.5µM, 5µM, and 184 50µM SeMet in either freshwater or saltwater. Each graph represents one stage of treatment: 9, 17, 25, 29, 34 and 38.

Figure 4.3. Percent deformities following 24 h of treatment with 0.5µM, 5µM, 185 and 50µM SeMet in either freshwater or saltwater. Each graph represents one stage of treatment: 9, 17, 25, 29, 34 and 38.

Figure 4.4. Percent of each type of deformity inflation following 24 h of 187 treatment with all concentrations of SeMet and both types of water.

Figure 4.5. Percent of each type of spinal deformity inflation following 24 h of 188 treatment with all concentrations of SeMet and both types of water.

Figure 4.6. Percent average severity of spinal deformities following 24 h of 190 treatment with 0.5µM, 5µM, and 50µM SeMet in either freshwateror saltwater. Each graph represents one stage of treatment: 9, 17, 25, 29, 34 and 38.

Figure 4.7. Percent failed swim bladder inflation following 24 h of treatment 191 with 0.5µM, 5µM, and 50µM SeMet in either freshwater or saltwater. Each graph represents one stage of treatment: 9, 17, 25, 29, 34 and 38.

Figure 4.8. Median day to hatch following 24 h of treatment with 0.5µM, 5µM, 193 and 50µM SeMet in either freshwater or saltwater. Each graph represents one stage of treatment: 9, 17, 25, 29, 34 and 38.

xv Figure 4.9. Whole egg Se content (dry weight (DW) µg/g) in embryos 194 surviving following 0.5µM, 5µM, and 50µM SeMet treatment in either freshwater or saltwater.

Figure 5.1. Effects of combined exposure of SeMet (50µM) and hypersaline 217 conditions on the development and hatchability of Japanese medaka embryos after exposure at 12 hpf.

Figure 5.2. Effects of combined exposure of SeMet (50µM) and hypersaline 218 conditions on lipid peroxidation in Japanese medaka embryos after 12 and 24 hpf.

Figure 5.3. Effects of combined exposure of SeMet (50µM) and hypersaline 219 conditions on BAX and CASP3A transcripts in Japanese medaka embryos after 12 and 24 hpf.

Figure 5.4. Effects of combined exposure of SeMet (50µM) and hypersaline 220 conditions on BiP, ATF6, and ATF4 transcripts in Japanese medaka embryos after 12 and 24 hpf.

Figure 6.1. Effects of 24 hours of 2.5µM and 5µM SeMet on survival and 241 deformities of Japanese medaka embryos at stage 25 against controls

Figure 6.2. Representative images of staining of SeMet treated embryos with 243 10µg/ml DCFDA and imaged on a stereoscope. Proportion of embryos stained with DCFDA in the tail and yolk sac.

Figure 6.3. Effects of 24 hours of 2.5 and 5µM SeMet on mRNA expression of 244 CHOP, PDIA4 and Dnajb9

Figure 6.4. Effects of 24 hours of 1, 2, and 4µg/ml tunicamycin (Tm) on 245 mRNA expression of CHOP, PDIA4 and Dnajb9; and Sox9, Col2a1, Runx2 and Twist as fold change.

Figure 6.5. Effects of 24 hours of 2.5 and 5µM SeMet on mRNA expression of 246 Sox9, Col2a1, Runx2, and Twist

Figure 6.6. Representative images of tails of Japanese medaka embryos stained 248 with 2µg/ml acridine orange following SeMet treatments and the average number of cells stained from tail tip to yolk sac.

xvi List of Tables

Table 1.1. Index for evaluating the impact of Se on fish populations. 11

Table 1.2. Role of unfolded protein response genes in physiological 57 development

Table 2.1. Nominal and measured concentrations of ions by ICP-OES and ion 120 chromatography, and water chemistry of freshwater, Instant Ocean, artificial San Joaquin River Water, and Monterey Bay Aquarium Brine (n = 3).

Table 2.2. No Observable Effect Concentrations (NOEC) and EC50 values for 122 the four different treatments.

Table 3.1. Primers, Accession Numbers, and Concentrations used for qRT- 144 PCR.

Table 3.2. Percent Identity Matrix of medaka FMOs as calculated by Clustal 146 Omega.

Table 3.3. Number of transcription factor binding sites of ATF (Alibaba2.1), 149 ATF4 (manual search), Chop-EBP (Consite), and C/EBPβ (Alibaba2.1) per medaka FMO promoter region.

Table 3.S1. The number of transcription factor binding sites identified in the 163 proximal promoter regions of medaka FMOs by Consite and Alibaba2.1. Full names and functions are summarized from UniprotKB.

Table 4.1. Developmental stages of medaka development chosen for 24 h of 178 SeMet treatment.

Table 4.2. Relationship between egg Se concentration and mean percent 195 deformities for embryos treated with SeMet in freshwater and saltwater.

Table 5.1. Primers, Accession Numbers, and Concentrations used for qRT- 215 PCR.

Table 6.1. Primer sequences and accession numbers for genes evaluated in this 239 study, along with the concentrations used for each primer pair.

xvii Chapter 1: Introduction

Selenium

Selenium (Se) is a widely distributed non-metal, with both metalloid and non- metal properties. Like a metal, Se has several oxidation states and can be present in its elemental form (Se0), or as selenate (4+), selenite (6+) or organoSe (such as selenocysteine or selenomethionine (2-)). Present naturally in sediments, Se release into surface waters and uptake into biota can negatively impact fish and birds. Although Se is an essential micronutrient, it can be toxic to aquatic organisms at 3-10 times concentrations necessary for biological function (Lemly, 1997). Se poisoning events have been documented throughout the USA, with severe long-term disturbances occurring in the San Joaquin River Valley and San Francisco Bay Delta, California. Although a large body of knowledge is available concerning the beneficial and negative impacts of Se, very little work has been performed on its interactions with multiple stressors and its mechanisms of developmental toxicity.

Se Overview

Physiological Roles

Selenium is an essential micronutrient for all kingdoms of life. Fish require approximately 0.15-0.5mg/kg in the diet for survival, and over supplementation can lead to adverse health effects at concentrations greater than 3mg/kg (Watanabe et al., 1997;

Hamilton, 2004). In humans, there are 25 known selenoproteins, which provide a variety

1 of functions including: thyroid hormone metabolism, inflammatory response, anti- oxidative capacity and Se homeostasis (Fairweather-Tait et al., 2010). Many selenoproteins are involved in maintenance of cellular redox potential. Glutathione peroxidases (GPx) can reduce H2O2 with two antioxidant tripeptide glutathione (GSH) molecules, generating water and glutathione disulfide. In humans, there are four members of the GPx family, GPx1, GPx2, GPx3 and GPx4 (Imai & Nakagawa, 2003). GPx1 and

GPx2 are located in the cytoplasm, whereas GPx3 is found primarily in the plasma of the kidney (Imai & Nakagawa, 2003). GPx4 is distinct, as it is the only GPx that can reduce phospholipid hydrogen peroxides (Imai & Nakagawa, 2003). GPx proteins are also conserved in teleosts. Many species, including zebrafish, carp, trout, salmon and tuna have GPx4, and GPx1 and 2 orthologs, whereas a GPx3 ortholog has only been discovered in zebrafish (Pacitti et al., 2013).

Another important class of anti-oxidative selenoproteins are thioredoxin reductases (TrxR). Thioredoxins are small proteins, which act as key regulators of redox signaling through the reduction of disulfides and H2O2. During reduction of disulfide bonds, electrons from NADPH are transferred to TrxR, which reduces thioredoxin to reduce disulfide bonds (Hanschmann et al., 2013). There are three isoforms of TrxR, which have a wide substrate specificity and are regulated by nuclear factor E2-related factor 2 (Nrf-2) (Kim et al., 2001). TrxR2 and TrxR3 orthologs have been discovered in a variety of teleosts, such as zebrafish, trout and salmon, however, none of the fish TrxR genes studied have phylogenetic identity to human TrxR1 (Pacitti et al., 2014).

2 Metabolism

Se incorporation into proteins as selenocysteine is tightly controlled by a complex

Se metabolic pathway (Figure 1.1). The metabolic transformation of Se depends on the form assimilated (Figure 1.1; Rayman et al., 2008). Se in the diet is often in the form of

Selenomethionine (SeMet). SeMet can be converted into selenocysteine via the trans- sulfuration pathway. SeCys is then converted into hydrogen selenide via SeCys beta- lyase (Rayman et al., 2008). From there, hydrogen selenide can be phosphorylated by selenophosphate synthetase to begin SeCys synthesis or the excess Se is excreted as a selenosugar in the urine. Selenide may further undergo oxidation to form Se dioxide, which causes oxidative stress. Inorganic selenate or selenite can be reduced by GSH to selenide. Excess SeMet can also be incorporated nonspecifically into proteins in place of methionine. SELENIUM BIOAVAILABILITY 1487S responsive biomarkers, and these could conceivably include aspects of speciation. Plasma selenium concentration reﬂects dietary exposure to most forms of selenium, but in the absence of a well-described homeostatic regulation there is no absolute plateau, although the concentration will reach a steady state at any constant level of intake after ’10–12 wk (33, 91, 92, 94–97). In addition to dose, the plasma response to dietary selenium is species dependent, so consumption of 2 different forms may result in different plasma selenium concentrations (33, 92, 95, 96, 98, 99).

EFFECT OF GENOTYPE

Figure FIGURE1.1. Selenium 1. Metabolic metabolism. pathway Fairweather of dietary-Tait, sele 20nium10. in humans. Se, selenium; The response by individuals to 6 wk of selenium supple- SeMet, selenomethionine; SeCys, selenocysteine; GSSeSG, selenodiglutathione; mentation with 100 lg sodium selenite/d has been shown to be Downloaded from c-glutamyl-CH3SeCys, c-glutamyl-Se-methylselenocysteine; H2Se, hydrogen 2- inﬂuenced by genetic polymorphisms in the selenoprotein P selenide; HSePO3 , selenophosphate; CH3SeCys, Se-methylselenocysteine; gene (SEPP) (100) and GPX4 gene (101). Biomarkers that are CH3SeH, methylselenol; (CH3)2Se,3 dimethyl selenide; SeO2, selenium + commonly used to assess selenium bioavailability (plasma se- dioxide; (CH3)3Se , trimethyl selenonium ion. Reproduced with permission from reference 5. lenium, selenoprotein P, and GPx3) were associated with 2 common single nucleotide polymorphisms in SEPP in both

baseline and postsupplementation samples (100). The GPX4 ajcn.nutrition.org a paucity of data for meaningful subgroup or dose-response polymorphism was shown to influence lymphocyte GPx4 con- analysis. In the included studies plasma selenium was the most centration and other selenoproteins in vivo (101). A single nu- commonly measured biomarker, and it responded positively to cleotide polymorphism in GPx1 (Pro198Leu) was associated intervention, as did whole-blood and erythrocyte selenium, with selenium deficiency and impaired GPx1 activity (102) and plasma selenoprotein P, and platelet, plasma, erythrocyte and also may be associated with a different response of GPx1 ac- whole-blood GPx activity, albeit with significant heterogeneity in tivity to selenium (103). This observation raises the issue of at UNIV OF CALIFORNIA RIVERSIDE on June 19, 2014 each case. The review concluded that further large-scale in- whether common polymorphisms in selenoprotein genes, such terventions are required to assess the usefulness of selenium- as SEPP, GPX1, GPX4, and selenoprotein S (SELS) (92), will

TABLE 2 Bioavailability of selenium from various foods1 Food (reference) Technique used Results

Selenium (Se)-yeast, 300 lg/d for 10 wk, Absorption from stable isotopically labeled material 89% then single dose of 77Se-yeast (34) (327 lg selenium) 74% Retention (absorption minus urinary excretion) Brazil nuts, 53 lg/d for 3 mo (35) Plasma selenium increase 64.2% Plasma GPx increase 8.2% Whole-blood GPx increase 13.2% Selenomethionine, 100 lg/d for 3 mo (35) Plasma selenium increase 61% Plasma GPx increase 3.4% Whole-blood GPx increase 5.3% Biofortified wheat-flour biscuits, mean Plasma selenium increase after 6-mo feeding trial 72-lg/L increase intake 172 lg/d for 6 mo (11) Fortified wheat-flour biscuits, mean intake Plasma selenium increase after 6-mo feeding trial 16-lg/L increase 208 lg/d for 6 mo (11) Basal diet, 52 lg selenium + cow milk, Fractional absorption in ileostomists 65.5% 15 lg selenium (36) 73.3% Shrimp, 88 lg/d for 6 wk (37) Plasma selenium increase 6.3-lg/L increase Apparent absorption 83% Beef, rice, and powdered milk , 14 lg/d (low) Plasma selenium change 240 lg/L (low); 97 lg/L (high) compared with 297 lg /d (high) for 14 d (38) Muscle selenium 20.37 lg/g protein (low); 0.57 lg/g Platelet GPx protein (high) Red blood cell selenium 2120 nkat/g protein (low); 100 nkat/g Red blood cell GPx protein (high) 242 lg/L (low); 106 lg/L (high) 215 nkat/g protein (low); 13 nkat/g protein (high) Pork, 106 lg/d for 3 wk; 7 d metabolic Apparent absorption 94% balance in final week (39) Retention 58% 1 GPx, glutathione peroxidase; nkat, nanokatal. Selenium in the Environment

Sources and Speciation

Selenium is present naturally in the Earth’s crust, particularly in coals, phosphate, shale, and metallic ore deposits (Luttrell, 1959). Se can be mobilized by a variety of natural and anthropogenic activities. Natural occurrences, such as volcanic eruptions, weathering of seleniferous rocks and soils, and Se volatilization are secondary in Se release in comparison to anthropogenic activities. Mining, oil refinement, and agricultural irrigation are the primary causes of Se release into the environment. Se released from agricultural drainage, treated oil refinery effluent, copper mining discharge and mountaintop mining discharge is often predominately selenate, while selenite predominates oil refinery effluent, fly ash disposal effluent and phosphate mining leachate (Presser & Ohlendorf, 1987; Zhang & Moore 1996; Cutter & San Diego-

McGlone, 1990).

Atmospheric Se plays a significant role in Se dispersion and transport. Natural sources of atmospheric Se are weathering of seleniferous rocks and soils, volcanic activity and sea salt, while anthropogenic contributions include combustion, nonferrous metal melting and use of agricultural products (Wen & Carignan, 2007). Several methylated Se compounds are volatile, though dimethyl selenide is most common due to its high stability (Wen & Carignan, 2007). Inorganic Se species (H2Se and SeO2) result primarily from coal combustion and can also be volatile but are unstable and usually form particulates (Andren & Klein, 1975). Se has a short atmospheric residence time and is

4 thus not transported long distances, with wet deposition predominating Se removal from the atmosphere.

Environmental Fate and Transport of Selenium

The fate and transport of Se into aquatic systems depends greatly on the Se source and the receiving waters. Elemental Se, selenite, selenate and hydrogen selenite are present in water at pH 6 to 8 in either a dissolved or particulate form (Milne, 1998).

Dissolved Se is only present in selenate, selenite and organic selenides, while particulate

Se can be found in any oxidation state, either adsorbed to particulates or within microalgae and periphyton (Doblin et al., 2006). The soluble Se fraction can be removed from the water column following reduction to elemental Se or binding to organic matter.

In the North San Francisco Bay, the largest proportion of particulate Se was found to be organic selenides, followed by elemental Se (Doblin et al., 2006).

In the dissolved fraction, dissolved oxygen typically does not oxidize selenite

2- (SeO3 ) to selenate due to slow reaction kinetics. The process can be enhanced by UV radiation, redox reactive metals and oxidizing bacteria in the water column. The presence of reducing bacteria can contribute to the persistence of selenite in oxic systems. In general, selenite and selenate are found in the oxic zone, while elemental Se is more common in anoxic areas (Stolz & Oremland, 1999).

In sediment, Se can follow a variety of transformation pathways. Elemental Se can only be formed under anoxic conditions (Cutter, 1992). Iron oxides in sediments can reduce selenite to selenate and elemental Se (Myneni et al., 1997). Particulates in the

5 North SF Bay water column contained approximately 30% elemental Se, suggesting a large contribution from resuspended sediments (Doblin et al., 2006). In Kesterson reservoir, CA, sediments under reducing conditions resulted in primarily iron selenides and elemental Se, while oxidizing environments produced more dissolved selenite

(Masscheleyn et al., 1990).

Bioaccumulation and Trophic Transfer

A critical step in Se toxicity is uptake by microorganisms and phytoplankton in the water column. It has been estimated that dissolved Se exposure in the water column makes little contribution to Se bioaccumulation at higher trophic levels (Lemly, 1985).

The greatest accumulation of Se occurs at the base of the food chain (Orr et al., 2012).

Freshwater algae assimilate selenomethionine most readily over other forms; however, they can also take up selenite and selenate (Stewart et al., 2010). In green algae, uptake of selenomethionine and selenate appeared to be transporter mediated, while selenite uptake was constant at all concentrations tested suggesting passive mechanisms (Fournier et al.,

2006). It has been shown that selenate competes with sulfate for uptake in algae, aquatic plants, crustaceans, fungi and Hela cells suggesting a similar mechanism of transport

(Milne, 1998; Shrift, 1954). Furthermore, marine phytoplankton have been shown to concentrate selenate and organic selenides by a factor of up to 106 (Baines & Fisher,

2001). However, there is large variability of Se uptake between algal species, for instance, diatoms and prymnesiophytes enrich Se at much greater degrees than chlorophytes (Baines & Fisher, 2001).

6 Trophic level two organisms accumulate Se primarily in its organic form from the diet. In the bivalve Macoma balthica, dissolved selenite was found to have a negligible contribution to Se assimilation, and elemental Se in ingested sediment was assimilated with an efficiency of 22% (Luoma et al., 1992). However, organic Se in diatoms fed to the mussels had the greatest contribution to Se accumulation (Luoma et al., 1992).

Overall, the amount of bioaccumulation of Se depends greatly on diet choice, the ingestion rate, assimilation efficiency and elimination rate of each individual species

(Stewart et al., 2010).

Trophic transfer factors (TTFs), which characterize the potential of an organism to bioaccumulate a compound, can be calculated for each species using the following formula, where AE is assimilation efficiency, IR is ingestion rate, Ke is the elimination rate constant (Presser & Luoma, 2010):

TTF= (AE*IR)/(Ke)

TTFs in invertebrates have been shown to be primarily greater than 1, demonstrating that Se biomagnifies through the food chain (Presser & Luoma, 2010).

Furthermore, bivalves, such as clams or mussels, can have TTFs as high as 23, which suggests that bivalves accumulate greater levels of Se than other invertebrates, such as crustaceans. In areas such as the San Francisco Bay Delta, bivalves accumulate 4-5 times more Se than mysids (Stewart et al., 2004). This difference can be explained by variability in elimination rate constants between classes (Stewart et al., 2004).

Assimilation efficiencies from the invertebrate diet are generally high and vary between

60% and 100% (Stewart et al., 2010). However, mysids eliminated Se at a rate of up to

7 10 times faster than P. amurensis, resulting in lower Se bioaccumulation (Stewart et al.,

2004). Ingestion rate may also impact species differences in biomagnification.

Considering the high variability of invertebrate Se accumulation, Se availability to fish also greatly depends on diet. In particular, selenomethionine was found to be the primary form of Se in the fish diet (Phibbs et al., 2011). TTFs for Se whole body concentrations range between 0.85 and 1.6 (Presser & Luoma, 2010), suggesting some potential for biomagnification; however, Se content is highly dependent on tissue type.

For instance, rainbow trout (Oncorhynchus mykiss) fed a SeMet spiked diet of 40µg/g dry mass, accumulated Se significantly in all tissues with particularly high accumulation in the gastrointestinal tract, gonads, liver and brain (Misra et al., 2012a). This total Se was composed of SeMet, Se-Cysteine and Se-Cystine (Misra et al., 2012a). In Japanese medaka females fed Se-dosed mayflies, 24% of the total Se was partitioned to the ovaries, while 18% was found in the digestive tract, 14% in the liver and 42% in the remaining carcass (Conley et al., 2014). In contrast, males fed the same diet only partitioned 0.8% to the testes, while 15% went to the liver, 17% to the digestive tract and

64% to the carcass (Conley et al., 2014). Finally, in juvenile white sturgeon exposed to dietary SeMet, Se accumulated preferentially in the kidney followed by the liver, though the gonads, brain and digestive tract were not measured (Tashjian et al., 2006). Se elimination in fish tends to be on the order of 3 to 4 weeks for dietary exposures (Presser and Luoma, 2010). For instance, juvenile striped bass lost between 8-10% Se per day

(Baines et al., 2002). Furthermore, reproductively active Japanese medaka females had an efflux rate 2.5 fold greater than reproductively inactive females (Conley et al., 2014).

8 This difference is attributed to SeMet’s high rate of maternal transfer into embryos, which contributes to Se embryo toxicity. It has been determined that Se is transferred via vitellogenin (VTG) from mother to offspring in the yolk (Janz et al., 2010). VTG is a yolk precursor protein, which is synthesized in the liver following induction by estradiol.

It is hypothesized that mobilization of SeMet from yolk proteins is responsible for the observed malformations and mortality (Janz et al., 2010).

Selenium Toxicity to Fish

Teratogenesis and Embryotoxicity

Oviparous vertebrates, such as fish and birds, are the most sensitive taxa to developmental Se toxicity. And most notably, Se has resulted in loss of entire species from aquatic ecosystems in lakes in North Carolina, Texas, California, and Colorado

(Hamilton, 2004). Se is passed from female fish into developing offspring via maternal transfer- a fact first discovered following the Belews Lake, NC, contamination event in the 1970s. Gillespie and Baumann (1986) collected Bluegill sunfish (Lepomis macrochiru) from the another contaminated site, Hyco Reservoir, NC, and from a control site and artificially crossed them in clean water in all combinations. Although fertilization success and hatch remained constant, larvae resulting from crosses only with contaminated females exhibited gross morphological abnormalities and reduced survival

(Gillespie & Baumann 1986). Furthermore, Se concentrations in the ovaries were consistently higher than the carcass. This work was the first indication that the population

9 declines caused by Se toxicity resulted from maternal transfer rather than aquatic

exposure. Furthermore, although high levels of Se can cause larval mortality, hatchability

of fish is generally not affected by Se (Janz et al., 2010). Thus, deformities are a better

indicator of Se toxicity than mortality or hatchability (Janz et al., 2010).

Se causes a variety of teratogenic effects including: spinal (lordosis, scoliosis and

kyphosis), fin (missing or deformed), gill (missing or deformed), or cranio-facial effects

(deformities of the eyes, mouth or head) (Figure 1.2) (Lemly, 1997). Some effects from

Se are reversible and may not be considered as true deformities, including edema and

cataracts (Lemly, 1997). Using lab and field data for Se available at the time, Lemly

(1997) related whole-body Se conceentrations to the percent of deformed fish in a 260 A. DENNIS LEMLY rensen, 1986). This paper examines the most prominent out- to focus on the earliest life stages, i.e., newly emerging larvae ward manifestationpopulation of selenium on toxicosis—teratogenic an exponential defor- and scale. young fry. However, no such relationship was established for mities—and presents an index for using this pathological symptom as a diagnostic tool to assess impacts on fish popu- SYMPTOMS OF TERATOGENESIS lations in contaminated aquatic habitats. Teratogenic deformities can occur in most, if not all, hard or mortality, as fish with a certainsoft tissuesprevalence of the body. However, of deformities some of the most conspicu- died regardless of Se body OCCURRENCE AND PERSISTENCE OF ous (consequently, the most diagnostic) are found in the skel- TERATOGENIC EFFECTS eton, fins, head, and mouth. These typically involve (1) lordosis—concave curvature of the lumbar region of the spine; (2) Teratogenicburden. deformities Based in fish are a permanenton this pathological informscoliosis—lateralation, Lemly curvature ( of19 the97) spine; developed (3) kyphosis—convex a teratogenic deformity marker of selenium poisoning. They are congenital malforma- curvature of the thoracic region of the spine resulting in a tions that occur due to excessive selenium in eggs. The process ‘‘humpback’’ condition; (4) missing or deformed fins; (5) begins withindex the diet of to parent co fish.nnect Excess dietary observed selenium (>3 numbersmissing or deformed of deformities gills or gill covers (opercle);in the (6) field abnor- with population mortality mg/g) causes elevated concentrations of selenium to be depos- mally shaped head; (7) missing or deformed eyes; and (8) ited in developing eggs, particularly the yolk. When eggs deformed mouth. Several of these symptoms are presented in hatch, larval fish rapidly utilize the selenium-contaminated Figs. 1–3. yolk, both( asTable an energy 1. supply1). and This as a source index of protein estimates for In general, the a carefulimpact fish-in-hand of Se inspection on a is sufficientpopulation to based on the observed building new body tissues. Hard and soft tissues may be de- diagnose any of the major teratogenic deformities. However, formed if the molecular structure of the protein building blocks careful examination with the aid of a dissection microscope is has been distorted due to substitution of selenium for sulfur. needed to make the diagnosis for larvae and fry, small species Some tissuesdeformities may not be generated and at all, estimates resulting in missing that(e.g., 20 small-80% cyprinids, of poeciliids, juveniles etc.), or inor situations adults when itwith deformities would body parts. is necessary to tabulate all of the subtle, less overt symptoms The prevalence of teratogenic deformities increases rapidly (e.g., slightly deformed fins, opercles, etc.). This is particularly once seleniumresult concentrations in 5- in20 eggs% exceeds population 10 mg/g. Hatch- mortalitytrue for larval fish.which Some of would their undeveloped have features a slight could to moderate impact ability of eggs is not affected by elevated selenium even though erroneously be considered a defect when, in fact, they are a there may be a high incidence of deformities in resultant larvae consequence of a premature life stage, not selenium teratogen- and fry, and many may fail to survive (Gillespie and Baumann, esis. However, this is not a serious concern because larval fish 1986; Coyle(Tableet al., 1993). 1. The1) time. for induction of teratogen- have distinctive patterns of development that quickly become esis is when larval fish are relying on their attached yolk sac for apparent to the investigator looking for teratogenesis. With a nourishment and development. Once external feeding begins, the potential for teratogenic effects declines and is soon lost. Feeding excessive selenium (up to lethal levels) to fry or juvenile fish as they are growing will not cause teratogenic malformations to occur (Hamilton et al., 1990; Cleveland et al., 1993). Moreover, dietary selenium levels sufficient to load eggs beyond teratogenic thresholds (diet: 5–20 mg/g) do not cause teratogenesis in, or otherwise generally affect the health or survival of, parent fish (Coyle et al., 1993). Thus, the teratogenic process is strictly an egg–larvae phenomenon. Because of these relationships, teratogenesis can be a very subtle, but important, cause of reproductive failure in fish. Entire populations may disappear with little evidence of ‘‘toxicity’’ since major impacts to early life stages can be taking place at the same time that adult fish appear healthy (Cumbie and Van Horn, 1978;Figure Lemly, 1985). 1.2. Mosquito fish from Belews lake, NC. A. normal, B. Lordosis. Lemly, 1997. Mortality of larval fish can be high if the teratogenic defects are severe enough to impair critical body functions (Woock et al., 1987). However, in some cases the abnormalities may not be life threatening and the malformations can persist into juvenile and adult life stages (Lemly, 1993). This is likely re- 10 stricted to locations where there is little threat from predators since all but the most subtle deformities would probably com- FIG. 1. Normal (A) and teratogenic (B, C) adult mosquitofish (Gambusia promise a fish’s ability to feed and avoid predators. Thus, in affinis) exhibiting dorsoventral deformation (kyphosis and lordosis) of the assessing the prevalence of teratogenic defects, it is important spine. 264 A. DENNIS LEMLY ic-based assessment of impacts to fish populations (Table 1). The index can be applied to virtually any aquatic habitat This index is composed of three ratings that signify increasing because it consists of impact-based assessment. Impacts levels of terata-induced population mortality: 1, negligible im- (terata) are a function of selenium concentrations in fish eggs. pact (<5% population mortality); 2, slight to moderate impact Conditions responsible for getting selenium into fish eggs— (5–20% population mortality); 3, major impact (>20% popu- bioaccumulation in aquatic food chains and consumption of lation mortality). Each rating is based on the anticipated popu- contaminated diets by parent fish—can be highly variable from lation-level impact of the corresponding degree of mortality, location to location and are influenced by such things as hy- i.e., little effect is expected with <5% mortality but substantial drology and landform (amount and timing of precipitation; effects may occur with >20% mortality. Population mortality is stream, lake, or wetland), chemical form of selenium (selenate, calculated in four simple steps: (1) determine the percentage of selenite, organoselenium), and timing and amount of selenium teratogenic fish and the percentage of normal fish in the total inputs relative to spawning periods (Lemly and Smith, 1987). sample, (2) multiply the percentage of teratogenic fish by the Consequently, the potential hazard (likelihood of toxic im- expected mortality rate (80% for larvae, 25% for juveniles and pacts) of selenium to fish and wildlife is also highly variable adults) to estimate the percentage of fish that will survive, (3) (Lemly, 1995, 1996). However, the index is based on a mea- add the percentage of normal fish and the percentage of sur- sure of existing impact (terata), not potential hazard. As such, viving teratogenic fish, (4) subtract this sum from 100%—the terata are an expression of the sum total of parental exposure, result is population mortality, which will be less than terato- regardless of the temporal, spatial, or chemical variations that genic mortality. For example, 20% teratogenic larvae with 80% may exist from site to site. Thus, the applicability of the index mortality translates to 16% population mortality; 20% terato- is not influenced by local environmental conditions that affect genic juveniles/adults with 25% mortality translates to 5% selenium dynamics and biological uptake. It makes no differ- population mortality. Because of the differences in teratogenic ence whether the system is a fast-flowing stream in which mortality between larval and juvenile/adult fish, age-specific selenate predominates and bioaccumulation is low or a termi- indices were developed. As discussed previously, persistence nal wetland experiencing high bioaccumulation from sele- of terata in older life stages (and thus accurate evaluation) can nite—population-level impacts are indicated only if a sufficient be heavily influenced by predation; thus, the index for juve- amount of terata exist. niles and adults may have limited application. The terata–mortality relationships are based on data for two EXAMPLE ASSESSMENTS fish families—Centrarchidae (bass, sunfish) and Cyprinidae (minnows). The resultant index for impacts may or may not be (1) Collect and examine 500 larval fish using ichthyoplank- directly applicable to cold- or cool-water families such as Sal- ton sampling techniques; assess the prevalence of teratogenic monidae (e.g., trout, salmon) or Esocidae (e.g., pike, muskel- deformities and measure selenium concentrations in six com- lunge). However, extrapolation to other families of fish is prob- posite samples of teratogenic individuals. The investigation ably not necessary since centrarchids and cyprinids have char- reveals that 15% have terata and the associated selenium con- acteristics that make them a good indicator or sentinel for other centrations are 10–15 mg/g. The expected population-level species, i.e., they are sensitive to selenium, include nationally mortality is 12% (85% normal + 3% surviving teratogenic 4 important sport fish species, and occupy most of the aquatic 88% total survival), resulting in an index rating of 2. Conclu- habitats in the continental United States (Lee et al., 1980; sion—slight to moderate impact on the population due to ter- Lemly, 1993). atogenic effects of selenium. Table 1.1. Index for evaluating the impact of Se on fish populations. Lemly, 1997. (2) Collect and examine 300 juvenile and 200 adult fish. Assess the prevalence of teratogenic deformities and measure TABLE 1 selenium concentrations in individuals with terata (whole- Index for Evaluating the Impact of Selenium-Induced body; 6 juveniles and 6 adults). The investigation reveals that Teratogenic Mortality on Fish Populations 8% have terata and the associated selenium concentrations are 20–30 mg/g. The expected population-level mortality is 2% Fish life % With % Population Index Anticipated stage terata mortalitya rating impact (92% normal + 6% surviving teratogenic 4 98% total sur- Larvae or fry <6 <5 1 Negligible vival), resulting in an index rating of 1. Conclusion— 6–25 5–20 2 Slight to moderate negligible impact on the population due to teratogenic effects >25 >20 3 Major of selenium. Juveniles or adults <20 <5 1 Negligible 20–80 5–20 2 Slight to moderate (3) Sample 1000 larval, 200 juvenile, and 100 adult fish. >80 >20 3 Major Determine the prevalence of teratogenic deformities and measure selenium concentrations in teratogenic individuals (6 com- a Mortality is expressed as a percentage of the total fish population, not teratogenic mortality. For example, 20% larvae with terata translates to 16% posite samples for larvae, 6 individuals for juveniles and population mortality because up to 20% of those with terata would be expected adults). The investigation reveals that 35% of larvae have to survive to adulthood (e.g., only about 80% of teratogenic larvae die). terata and 3% of juveniles and adults have terata; selenium

Since Lemly’s studies, the literature available on the effects of Se to wild populations has greatly increased. In 2005, Holm et al., performed a comprehensive assessment on the effects of Se on offspring of rainbow trout (Oncorhynchus mykiss) and brook trout (Salvelinus fontinalis) collected downstream of an active coal mine in

Manitoba, Canada. Although no effects were observed in parents with high Se body burdens, significant increases in deformities were found in hatched larvae. No increase in mortality, fertilization success or time to hatch was observed in Se exposed embryos, which is consistent with earlier studies (Holm et al., 2003). A significant correlation was observed between Se content, cranio-facial deformities, and spinal deformities in rainbow trout (Oncorhynchus mykiss) but not for brook trout (Salvelinus fontinalis). Effect concentration (EC)15 values were calculated at 8.8-10.5µg/g wet weight. This work highlights the importance of deformities assessment in Se toxicity. Similar effects have been reported in eggs spawned from northern pike (Esox lucius) downstream of a

11 uranium mine; however, Northern pike may be less sensitive to Se toxicity, as EC20 values were calculated to be 33.55µg/g dry weight (Muscatello et al., 2006).

Mechanisms of Toxicity

Early research on Se toxicity hypothesized that observed effects were due to disruptions in protein folding due to Se substitution for sulfur in methionine (Met) and cysteine (Cys). This hypothesis was supported by research suggesting that SeMet substitution in enzymatic active sites would result in reduced catalytic activity

(Schrauzer, 2000). For instance, the Km of phosphomannose isomerase substituted with

SeMet was four-fold higher than the wild-type enzyme (Bernard et al., 1995). Huber &

Criddle (1967) found that E. coli treated with selenate incorporated SeMet into 55% of the Met active site residues of beta-galactosidase. While Km and Vmax were unaffected, the SeMet-containing enzyme was less stable. Similarly, Boles et al. (1991) found that selenomethionyl thymidylate synthase had a reduced half-life in comparison to wild type with no alterations in Vmax or Km.

However, the definitive role of protein folding in Se toxicity has yet to be determined. For instance, non-specific SeCys incorporation into proteins is of little concern. This is because SeCys synthesis and incorporation is highly regulated by a series of cis- and trans-elements (Low et al., 2000). Thus, only SeMet is incorporated nonspecifically into non-selenoproteins (Schrauzer, 2000). Treatment of E. coli with selenate resulted exclusively in SeMet substitutions (Huber & Criddle, 1967). A number of studies have since demonstrated that proteins containing SeMet in place of sulfur are able to fold

12 appropriately and that SeMet-containing enzymes retain activity (Mechaly et al., 2000;

Yuan et al., 1998). This is supported by the frequent use of SeMet in protein crystallization studies for labelling purposes (Egerer-Sieber, 2006; Madduri et al., 2009).

Thus, the role of protein integrity in SeMet toxicity remains unclear and recently the hypothesis that oxidative stress is responsible for observed effects has been under examination.

Studies on oxidative stress in SeMet toxicity began in birds. Hoffman et al. (1989) reported a dose-dependent decrease in reduced thiols and increase in the ratio of oxidized to reduced GSH (GSSG:GSH) in mallard ducklings treated with seleno-DL-methionine.

This was accompanied by an increase in lipid peroxidation. Similar results were reported in adults (Hoffman, 2002).

Concurrently, studies on Se’s role as an anti-carcinogen were being performed, which found chemical speciation of Se to be particularly crucial to its redox activity.

These studies reported that various metabolites of Se were able to induce redox cycling in vitro (Spallhoz et al., 2004). Spallhoz et al. (2004) attributed these effects to the generation of organic selenide species that were able to oxidize GSH. Compounds such as selenite, methylselenic acid and Se dioxide, were able to generate superoxide in vitro in the presence of GSH, while selenomethionine and selenate were not (Spallholz et al.,

2004). However, if the enzyme methioninase (also called methionine gamma-lyase) was added to SeMet, in both the presence and absence of GSH, superoxide was produced.

Methioninase catalyzes the formation of methylselenol from SeMet (Figure 1.1), suggesting the methylselenol may be responsible for toxicity. Involvement of

13 methioninase is further supported by a study suggesting that recombinant cell lines overexpressing methioninase are more susceptible to SeMet induced oxidative stress and cell death (Miki et al., 2001).

Several studies in fish support the hypothesis of oxidative stress as a mechanism of Se toxicity. Early studies found alterations in GSSG:GSH ratios, previously demonstrated in birds. In juvenile rainbow trout fed 180mg/kg dietary SeMet for 7 days, the GSSG:GSH ratio increased significantly (Schlenk et al., 2003). Oxidative stress was further found to occur in isolated rainbow trout hepatocytes at high concentrations of

500-1000µM SeMet for 24 hours in the form of increased superoxide dismutase activity

(SOD), increased catalase activity (CAT) and increased lipid peroxidation (Misra et al.,

2012b). Similar effects were observed in trout hepatocytes at much lower doses of selenite (Misra & Niyogi, 2009).

In trout embryo homogenates, 1mg/ml SeMet produced superoxide in trout embryo extracts with 1mg/ml additional GSH (Palace et al., 2004). Furthermore, a dose- dependent increase in superoxide production was observed when methylselenol was added to the embryo extracts. This evidence combined with previous studies on the formation of superoxide from SeMet and methionininase led the authors to postulate the mechanisms shown in Figure 1.3 (Palace et al., 2004). However, these studies were performed in vitro and with high concentrations of SeMet, suggesting further research is necessary in order to confirm this mechanism.

Figure 1.3. Proposed mechanism of SeMet induced oxidative stress in the presence of methioninase presented by Palace et al., 2004.

Another proposed mechanism of SeMet toxicity is activation to the SeMet oxide.

SeMet is easily oxidized by peroxynitrite and the corresponding SeMet oxide can be reduced by two equivalents of GSH (Assman et al., 1998). Other thiols such as N-acetyl cysteine and ascorbic acid are also able to reduce SeMet oxide (Krause & Elfarra, 2009).

In addition to oxidants, flavin-containing monooxygenases (FMOs) are can oxidize

SeMet (Chen & Ziegler, 1994; Krause et al., 2006). The ability of FMO to oxidize

SeMet, which can then be reduced and re-oxidized, suggests the potential for redox cycling, which can deplete GSH and lead to oxidative stress (Figure 1.4).

Figure 1.4. Oxidation of SeMet by FMO and reduction by GSH.

15 Although a plethora of evidence is available on the ability of Se to induce oxidative stress, until recently no study had demonstrated rescue of Se-induced toxicity with antioxidants. In studies on the effects of Se in zebrafish embryos, Arnold et al.,

(2016) found that embryonic waterborne treatments with 400µg/L SeMet during embryonic development resulted in approximately 50% deformities and 30% mortality.

These alterations were accompanied by significant decreases in total GSH and

GSH:GSSG ratios. Co-treatments with the GSH precursor, N-acetyl cysteine, significantly reduced the number of deformed fish treated with SeMet and restored GSH concentrations. These results suggest that oxidative stress is necessary for SeMet induced toxicity in zebrafish embryos at this high concentration.

Apoptosis has also been proposed as the cytotoxic mechanism following oxidative stress caused by Se. Caspase 3/7 activity was increased in trout hepatocytes treated with

1000µM SeMet and 100µM selenite for 24 hours (Misra et al., 2012b; Misra & Niyogi

2009). Furthermore, 10µM sodium selenite induced apoptosis and caspase activity in the fish hepatoma cell line, PLHC-1, which was accompanied by ROS production (Selvaraj et al., 2013). Still many of the studies here are performed with high concentrations of

SeMet and in vitro, leaving questions about the surrounding mechanisms of the generation of oxidative stress and the connections between measured oxidative stress and observed deformities unanswered.

16 Selenium Criteria and Regulation

Considering the complex chemistry and bioaccumulation of Se in the environment, water quality regulatory standards are hardly sufficient for Se regulation.

The first standards published by the USEPA in 1980 for Se were calculated on water-only exposures. These criteria were based solely on inorganic Se species, selenate and selenite, and were set at a 24-hour average of 760µg/l and 35µg/l, respectively (USEPA, 1980).

However, with the Belews Lake, NC, poisoning event in the early 1980s, the standard was revised in 1987 to address both waterborne exposures and dietary exposures. The 4- day average water concentration was set at 5µg/L and the 1-hour water concentration at

20µg/L (USEPA, 1987). Although these values were an improvement over the 1980 standards, they still did not include a tissue quality measurement and only considered selenate and selenite. Since that time, the USEPA has been working to find a more appropriate set of standards for Se regulation, and must contend with the complex chemistry and effects caused by Se.

Many have recommended that tissue-level criteria may be necessary for Se regulation (Sappington et al., 2011). However, a number of technical issues make implementation of tissue quality standards difficult. For instance, technical issues affecting outcomes include the monitored species’ life history, age, diet, and tissue, as well as Se speciation, survivor bias (Sappington, 2002; Lemly & Skorupa, 2007). In addition to tissue level criteria others have suggested that site-specific Se monitoring may be necessary (Ohlendorf et al., 2011). A 2004 draft of new aquatic life criteria was the first to propose tissue quality measurements for Se. It recommended a chronic whole

17 body dry weight criterion at 7.91µg/g dw and corrected water column concentration monitoring to sulfate values (USEPA 2004). These values were never finalized and the

USEPA is currently drafting a new set of criteria based on a greater body of current Se literature. The next set of criteria will account for four elements of Se exposure. They will include two based on tissue values (whole body and egg/ovary) and two based on water column values (30-day chronic and intermittent concentrations for lentic and lotic systems). At this time, the USEPA draft has undergone external peer review and is open for public comment.

Selenium in the San Francisco Bay Delta and San Joaquin Valley, CA

The Kesterson Reservoir

Since the late 1800s, massive agricultural development of the San Joaquin River

Valley (SJRV) area has drastically altered the natural landscape. In the early 1970s, construction began on the San Luis Drain to discharge subsurface irrigation to the San

Francisco Bay. As part of the project, the Kesterson Reservoir was added to the

Kesterson National Wildlife Refuge as a series of 12 shallow evaporation ponds totaling

518 hectares. Although the San Luis Drain project was halted in 1975, the reservoir remained and by 1982 flow into the reservoir consisted almost entirely of agricultural drainage (Presser & Ohlendorf, 1987). By 1983, collection of mosquito fish from the area revealed tissue concentrations of Se of approximately, 30ppm, which was 100 times greater than the adjacent reference site (Ohlendorf, 1984). Further analysis of waterfowl,

18 plants, and insects indicated that Se had bioaccumulated from low to high trophic levels

(Ohlendorf et al., 1986).

The elevated Se levels generated severe reproductive impacts on the fish and aquatic birds of Kesterson (Ohlendorf et al., 1986). In 1983, a kill in the San Luis Drain eliminated 8 species of fish from the reservoir. Of 347 bird nests surveyed in 1983, 40% contained at least one dead embryo, while 20% of nests contained at least one embryo exhibiting abnormalities of the eyes, legs, beak, and wings (Ohlendorf et al., 1986). It is estimated that Se exposure killed or deformed 1000 migratory birds between 1983-1985

(Presser & Ohlendorf, 1987). Agricultural drainage to the site was halted in 1986 (Presser

& Ohlendorf, 1987) and in 1988 the area was drained and filled in with soil.

Sources and Chemistry of Se in the SJRV and SFBD

The arid soils and marine sedimentary rocks of the San Joaquin River Valley are naturally seleniferous due to volcanic activity in the late cretaceous period, as well as marine bioaccumulation and subsequent deposition (Presser, 1994; Seiler et al., 1999).

Soil erosion from agricultural irrigation represents the primary source of Se to the waters of the Bay Delta, however, other Se inputs occur from oil refineries and power plants

(Presser & Luoma, 2006).

2- Se released into the San Joaquin River is primarily selenate (SeO4 ), which can be slowly taken up by microorganisms and plants at the base of the food web and converted into organoSe compounds (Fan et al., 2002; Luoma & Presser, 2009). For example, in the San Joaquin River, selenate was approximately 74% of total Se, while

19 2- elemental Se and selenide were about 20% (Cutter, 1989). In contrast, selenite (SeO3 ), is taken up much more readily. Still, organoSe release from dying organisms is the form most readily assimilated (Fan et al., 2002). Proteinaceous organoSe, particularly SeMet, was the dominant form of Se in aquatic organisms from the SJRV (Fan et al. 2002) and dietary Se is assimilated more efficiently than dissolved Se (Stewart et al., 2004).

In the SFBD, filter feeders, in particular the exotic bivalve Potamocorbula amurensis, accumulate Se with an assimilation efficiency of 45-80% (Stewart et al.,

2004). Predatory fish that preferentially feed on these bivalves, such as white sturgeon and Sacramento splittail, accumulate approximately 5 fold higher levels of Se than predators feeding on crustaceans (Stewart et al., 2004). While fish tend to have trophic transfer factors (TTFs) for Se at around 1, clams and mussels can have TTFs from 10 -20, further indicating the important role of diet in Se exposure (Luoma & Presser, 2009).

Current Se status in the Bay Delta

With fresh awareness for Se toxicity in the SFBD and SJRV, increased Se monitoring in the area indicated high potential for Se toxicity. In 2005 and 2006, it was estimated that the Se load from the San Joaquin River into the San Francisco Bay was

11,100 kg and 21,400 kg, respectively, with a flow-weighted mean on 0.31-0.79µg/L

(David et al., 2015). Water quality measurements taken by the Regional Monitoring

Program (RMP), suggested a bay-wide average of 0.2µg/L in 2013, with highest concentrations measured in the lower south bay at 0.25µg/L (SFEI, 2015). While the total load values may be of concern, the concentrations in surface waters are below current

20 regulations of 5µg/L and the US Fish and Wildlife Service guidelines of 1µg/L (David et al., 2015).

Nonetheless, several endemic fish species are currently listed as being at risk from

Se exposure in the SFBD, including Chinook salmon, steelhead, green sturgeon, white sturgeon, delta smelt and Sacramento splittail (Presser & Luoma, 2010). The Regional

Monitoring Program run by the San Francisco Estuarine Institute lists Se as a moderate concern (SFEI, 2015) and a Total Maximum Daily Load (TMDL) for Se in the North Bay is projected to be completed this year. A TMDL for the South Bay will be considered once the TMDL for North Bay is in place (Trowbridge et al., 2016).

Hypersalinity

Hypersaline toxicity

The salinity of many fresh waterways has been increasing over the past century generating hypersaline environments. For the purposes of this work, hypersalinity can be defined as an increase in salinity over historical values. Across the globe, salinity has been increasing to the extent that the Millennium Ecosystem Assessment has rated salinization as one of the most important stressors on freshwater ecosystems (2005).

Anthropogenic activities resulting in freshwater salinization include: agricultural irrigation, mining activity, application of roadway de-icer, effluent from industrial and urban areas (Canedo-Arguelles et al., 2013). Additionally, climate change is another major contributor to increasing salinity of freshwater and estuarine ecosystems. Reduced

21 rainfall can result in decreased runoff into streams that feed estuaries, while rising sea levels provide greater saltwater input.

The SFBD is one such area affected by increased salinity. It is estimated that rising temperatures due to climate change will result in up to 60% reduced snowpack over the next 100 years (Knowles & Cayan, 2002). This would significantly reduce stream outflow and increase salinity in summer months (Knowles & Cayan, 2002).

Although large variations in seasonal salinity are common, an increase over the last 50 years in spring salinity in Fort Point (Stahle et al., 2001) and in Suisun Bay have been reported (Figure 1.5; Peterson et al., 1996). Furthermore, with increasing demand on

California’s drinking water supply, there is a greater demand for desalination technology.

These facilities will discharge concentrated brine to the open ocean or water-restrained estuaries, which may increase ambient salinity (Jenkins et al., 2013).

80 D. Schlenk, R. Lavado / Aquatic Toxicology 105S (2011) 78–82

Table 2

Examples of FMO substrates (Hajjar and Hodgson, 1980; Schlenk, 1995; Cashman, 2008).

Pharmaceuticals Pesticides Endogenous molecules

Imipramine Aldicarb Trimethylamine

Ranidine Croneton Cysteine (and conjugates)

Methimazole Disulfoton Selenomethionine

Phenothiazines Phorate Selenocysteine

Cimetidine Fenthion

Methamphetamine Thiobencarb

Verapamil Thiourea

the anticholinesterase insecticides and reduced detoxiﬁcation in

salmonids (Lavado et al., 2009b) as well as other species of ﬁsh

acclimated to hypersaline conditions (El-Alfy and Schlenk, 1998;

El-Alfy et al., 2001).

In each case where enhanced toxicity was observed, the

pesticides were activated to more toxic oxon and/or sulfoxide inter-

mediates following saltwater acclimation. Aldicarb sulfoxide was

more than 100-fold more toxic than aldicarb in rainbow trout and

Japanese medaka (El-Alfy and Schlenk, 1998; Perkins and Schlenk,

2000). Stereoselective formation of the most potent oxon-sulfoxide

metabolite of fenthion was enhanced by acclimation to hypersaline

conditions as was its precursor, the more potent fenoxon; and the

less potent sulfoxide (Lavado et al., 2009b). In animals unaffected

by combined exposures to hypersaline conditions and pesticides,

Fig. 2. Mean-monthly (May) salinity at (A) Fort Point, modiﬁed from Stahle et al.

(2001) and (B) Pittsburg (CA) and water exported from the delta as percentage of activation to more potent intermediates was not enhanced. For

annual inﬂow (Peterson et al., 1996). example, the toxicity of aldicarb was not enhanced by salinity in

Figure 1.5. Mean salinity in May in A. Fort Point, CA, and B. Pittsburg, CA.striped From bass, and activation of aldicarb to aldicarb sulfoxide was

Schlenk & Lavado,ﬁsh 20 populations11. (Moyle, 2002). The reduced freshwater inﬂow and not increased (Wang et al., 2001).

resulting increase in salinity causes more salt-tolerant species to Oxidation of OPs and carbamates to more potent cholinesterase

move upstream while freshwater species retreat. In addition, lower inhibitors occurs through two monooxygenase enzymes:

Alterations freshwaterin salinity ﬂows canand greatlyhigher waterimpacttemperatures freshwaterdisrupt andspawning estuarine ecosystems.cytochrome P450 (CYP450) and ﬂavin-containing monooxy-

and larval transport (Bennett, 2005). Yet another relatively unex- genases (FMO). Each of the orthologs has been shown to be

plored issue is the relationship between hypersaline conditions induced by hypersaline conditions in euryhaline ﬁsh (Larsen

Maintaining osmotic pressure, ionic composition, and body fluid and cell volumes are

and the toxicological impacts of agricultural activities in the main and Schlenk, 2001; Lavado et al., 2009b). Acclimation to hyper-

drainage basins from the Central Valley. saline conditions requires the production of intracellular organic

critical for fish living in environments that can be hyper- or hypo-osmotic toosmolytes, their bodies. such as trimethylamine N-oxide (TMAO), to counter

increases in osmotic pressure. In addition, TMAO counters the

3. Impacts of hypersaline conditions on xenobiotic toxicity

intracellular toxicity of urea, which also serves as an organic

(i.e. pesticides)

Osmoregulation is controlled by both passive and active mechanisms in theosmolyte gills, (Larsen and Schlenk, 2001). Due to the enhanced pro-

duction of FMO during the osmoregulatory process, chemicals,

Salmonids and other anadromous and catadromous species of

which are biotransformed to more toxic intermediates by FMO

gastrointestinal tractﬁsh and routinely kidneysmove (Boeufbetween & freshPayan,and 20saltwater,01). Smallrelying changesupon in salinity can

(e.g. thioether OPs and carbamates) are made more toxic in the

olfaction to locate critical habitat for spawning. While hypersaline

animals under hypersaline conditions.

conditions can provide a certain degree of protection to aquatic

greatly impact stenohalineorganisms teleosts, from metal whichintoxication can only(Dwyer tolerateet al., 1992a narrow), interac- range of salinities.

tions with agricultural contaminants such as pesticides which may

4. FMO characteristics

co-occur in runoff are less clear (Hall and Anderson, 1995).

In contrast, those thatIn can the toleratecase of aorganophosphate wide range are(OP) calledand thioethereuryhaline.carba- Optimal salinities

While much has been done to characterize the CYP450 system

mate insecticides, signiﬁcantly greater acute toxicity to pesticides

in aquatic organisms, limited work has focused on FMO (Schlenk

are often species specifichas been; however,observed in alterationsseveral euryhaline in salinityﬁsh species, can haveincluding metabolic costs and

et al., 2008). FMOs are a family of monooxygenases that catalyze

salmonids acclimated to hypersaline conditions (Table 1). Accli-

oxidation of soft nucleophilic substrates that contain nitrogen, sul-

mation to hypersaline conditions increased the bioactivation of

influence growth rate (Canedo-Arguelles et al., 2013; Boeuf & Payan, 2001).fur, Thephosphorus or other heteroatoms that form oxide metabolites

(Cashman, 2008) (Table 2). Only 7 genes have been identiﬁed in

Table 1

ﬁsh and only one transcript (FMOh) in rainbow trout liver has been

Chemicals that have higher toxicities in hypersaline conditions.

energetic cost of osmoregulation is estimated to be between 10%-50% of theshown total to energycorrelate with a characteristic catalytic activity (ben-

Chemical Species Reference

zydamine N-oxidase) (Lavado et al., unpublished). Immunoblot

Aldicarb Oryzias latipes El-Alfy et al. (2001) studies with mammalian antibodies to mouse, porcine, human and

Oncorhynchus mykiss Wang et al. (2001)

monkey FMOs have indicated multiple protein bands, but their spe-

Fenthion Morone saxatilis23 Bawardi et al. (2007) ciﬁc catalytic activities have not been well characterized (Schlenk,

Oncorhynchus mykiss

Bawardi et al. (2007) 1998). Several biochemical and physiological studies have indi-

Oreochromis mossambicus Bawardi et al. (2007)

rectly characterized trimethylamine oxidase as an FMO catalytic

l-Selenomethionine Oryzias latipes Lavado et al. (2011) activity, but this enzyme has yet to be successfully puriﬁed or

Phorate Oncorhynchus kisutch Lavado et al. (2011)

recombinantly expressed from an aquatic organism. budget (Boeuf & Payan, 2001). Large alterations in salinity can be lethal to adult fish, although embryonic development appears to be the more sensitive life stage.

Furthermore, effects of salinity can vary based on ionic content, leading scientists to propose a need for ion-specific standards (Canedo-Arguelles et al., 2016). For instance, in a study on coho salmon (Oncorhynchus kisutch), embryos were exposed to 2500mg/L

TDS mimicking ionic content of mining effluent in Alaska (Stekoll et al., 2009). The solution was sulfate dominated, primarily composed of CaSO4, Na2SO4, and MgSO4.

Researchers further calculated EC50 values for the impacts of individual ions on coho

2+ + 2+ 2- + salmon fertilization and found the order of toxicity to be; Ca > K > Mg > SO4 > Na with a range of 102mg/L to 4744mg/L (Stekoll et al., 2009). In stenohaline freshwater organisms, combinations of various salts were tested for toxicity to Daphnia magna,

Ceriodaphnia dubia and Pimephales promelas (Mount et al., 1997). Ion toxicity

+ - 2+ - 2- modeling suggested that in order of toxicity, K > HCO3 = Mg > Cl > SO4 , while

Ca2+ and Na+ were not significant (Mount et al., 1997). With the exception of Ca2+, agreement exists between this model and the one determined for coho salmon (Stekoll et al., 2009). Furthermore, other studies suggest that in addition to ion-specific toxicity, ion imbalance may also play a role in salinity toxicity. For instance, fathead minnows are

2- particularly sensitive to SO4 during the embryo to larval period (Wang et al., 2016). The

+ 2- addition of K to test waters with equal SO4 concentrations significantly reduced toxicity while the addition of Cl- did not (Wang et al., 2016). These studies suggest that hypersalinity can be toxic to euryhaline and stenohaline organisms, but is also specific to ion ratios.

24 Hypersalinity in Multiple Stressor Situations

Although hypersalinity alone can exert toxicity, it can also modify the toxicity of other chemical and physical stressors that fish encounter in their environment. These interactions can be antagonistic, additive or synergistic, and complicate regulation of toxicants. There are multiple studies outlining the impacts of hypersalinity on xenobiotic toxicity (Schlenk & Lavado, 2011) and examples are given below.

Pesticides

In agricultural and urban areas across the US, irrigation and stormwater runoff containing pesticides can result in acute and chronic aquatic exposures. Fish are non- target organisms for pesticides and may have increased susceptibility over mammals.

Effects of pesticides are further complicated by changes in salinity. A common class of pesticides that was frequently used in CA are organophosphates, which act via inhibition of acetylcholinesterase. Phorate is an organophosphate pesticide with restricted use that has been shown to have a synergistic relationship with salinity. Juvenile Coho salmon were treated with 2-80µg/L phorate in freshwater or low (8g/l), medium (16g/l) and high

(32g/l) salinity (Lavado et al., 2011). At high salinity, there was an decrease in the time to mortality at all concentrations of phorate tested, with LC50 values of up to 32-fold lower than in freshwater. Phorate’s toxicity stems from its biotransformation by monoxygenase enzymes to oxon and oxon-sulfoxide metabolites. Metabolism by salmon liver, gill and olfactory tissue microsomes under freshwater and saltwater conditions suggested that

25 increased metabolism to more toxic metabolites may be contributing to the greater observed toxicity (Lavado et al., 2011).

Fenthion is another organophosphate with variable toxicity under hypersaline conditions. Similar to phorate, the acute toxicity of fenthion depends on its biotransformation products produced by FMOs and CYP450s. Sulfoxidation creates a chiral center which diminishes fenthion toxicity (Gadepalli et al., 2007). However, oxidative desulfuration of R-oxide to generate R-fenoxon sulfoxide, can lead to increased acetylcholinesterase inhibition of over 150-fold (Gadepalli et al., 2007). The impact of hypersalinity on fenthion biotransformation and toxicity to rainbow trout (Oncorhynchus mykiss), striped bass (Morone saxatilis X Morone chrysops) and tilapia (Oreochromis mossambicus) was assessed (Bawardi et al., 2007). The LC50 of fenthion was significantly decreased in rainbow trout and striped bass, but not in tilapia (Bawardi et al., 2007). However, these effects were not mirrored in the biotransformation products, suggesting a more complex mechanism of increased toxicity (Bawardi et al., 2007).

Follow-up studies in rainbow trout, found that the oxon and the oxon-sulfoxide products were increased in trout liver microsomes treated with hypersalinity, but not in gill or olfactory tissue microsomes (Lavado et al., 2009).

Further studies on organophosphates and salinity were conducted with chlorpyrifos. Chlorpyrifos, an organophosphate pesticide detected in CA waterways, has been shown to interact with salinity to cause toxicity to salmonids. Juvenile rainbow trout acclimated to 16ppth seawater had a greater median lethal time to death than in freshwater when exposed to acute levels of chlorpyrifos (Maryoung et al., 2014). These

26 differences were not related to alterations in chlorpyrifos metabolism, but possibly to alterations in signal transduction (Maryoung et al., 2014). Additionally, sublethal chlorpyrifos exposures in saltwater decreased olfactory function and behavioral responses to predator avoidance odorants in rainbow trout in comparison to freshwater, suggesting that an interaction between chlorpyrifos and salinity may impact olfactory signal transduction (Maryoung et al., 2015).

In addition to organophosphates, salinity impacts the toxicity of other classes of pesticides. Aldicarb is a carbamate pesticide that also exerts its effects through inhibition of acetylcholinesterase. In one study, mature female Japanese medaka were exposed to sublethal concentrations of aldicarb at 1.5ppth, 12ppth, or 20ppth salinity (El-Alfy &

Schlenk, 1998). Survival decreased from 100% of aldicarb-treated fish in freshwater to

0% of aldicarb-treated fish in 20ppth saltwater, while no difference was observed in aldicarb uptake (Schlenk & El-Alfy, 1998; El-Alfy & Schlenk, 1998). The ability of aldicarb to inhibit acetylcholinesterase is enhanced by S-oxidation to aldicarb sulfoxide, but reduced by the second S-oxidation to aldicarb sulfone (Baron, 1994). FMOs are able to oxidize aldicarb to the sulfoxide, which could increase its toxicity. FMO activity increased significantly in gills of fish at higher salinities, which correlated to significantly greater concentrations of aldicarb sulfoxide in the gills (Schlenk & El-Alfy, 1998).

Similar effects were observed in juvenile rainbow trout, but not in striped bass (Morone saxatilis x chrysops), which had no increase in aldicarb toxicity or FMO activity (Wang et al., 2001).

27 Finally, hypersalinity has also been shown to impact toxicity of pyrethroids, such as bifenthrin. In contrast to other pesticides, hypersalinity decreased the acute toxicity of bifenthrin in rainbow trout; however, species/strain differences exist as the same effects were not observed in steelhead trout (Riar et al., 2013). The plethora of data surrounding the impacts of salinity on pesticide toxicity to fish demonstrates that hypersalinity deserves consideration as a co-stressor in xenobiotic toxicity to fish.

Flavin Containing Monooxygenases

Flavin Containing Monooxygenases (FMOs) were first discovered in the 1960s by

Ziegler & Mitchell, who identified an enzyme that would oxygenate a tertiary amine that was preferentially dealkylated by cytochrome P450s (CYP450s) (Ziegler & Mitchell,

1972; Ziegler and Poulson, 1978). There are 5 functional FMOs and at least 7 pseudogenes in humans (Hernandez et al., 2004). Although FMOs have received less attention than CYP450s, they play a large role in metabolism of xenobiotic and endogenous compounds. FMOs preferentially oxidize soft nucleophiles, such as heteroatoms N, S, P and Se, to form more polar metabolites (Cashman & Zhang 2006).

The catalytic mechanism is distinct from CYP450s, in that FMOs form a stable hydroperoxyflavin in the absence of substrate and can accept reducing equivalents directly from NADPH (Cashman, 2005). First, the flavin adenine dinucleotide (FAD) cofactor is reduced by NADPH. The addition of molecular oxygen then forms the hydroperoxyflavin, which transfers an oxygen atom to the substrate. Water and NADP+ are then released from the FAD to regenerate the active enzyme (Ziegler, 2002).

28 FMOs are highly evolutionarily conserved, and exhibit species, sex, tissue and developmental stage specific patterns of expression. For instance, FMO1 is expressed in the fetal human kidney and liver, yet adult expression is restricted to the kidney (Dolphin et al., 1996). FMO1 expression in the liver is highest during the first trimetster and steadily declines until it is undetectable at one year old (Hines, 2006). FMO3 expression in the liver then begins during the neonate stage, then increases up to 11 years of age

(Hines, 2006). In contrast, other mammals such as rabbits, rats and mice, express FMO1 in the adult liver (Janmohamed et al., 2004). In male and female mice, FMO1 is highly expressed in the liver, lung and kidney throughout development, while expression in the brain is highest at 0 weeks then steadily declines (Janmohamed et al., 2004). FMO3 expression begins at 3 weeks in the liver and lung, but expression is higher in the female liver (Janmohamed et al., 2004).

FMOs have a variety of endogenous substrates, yet a distinct physiological role remains unclear. The use of knockout (KO) lines has established that FMO null animals are viable (Hernandez et al., 2009), suggesting that FMOs are not essential for embryonic development. FMOs have been shown to possibly be involved in cysteine and methionine metabolism (Ripp et al., 1997; Elfarra & Krause, 2005). Mutations in FMO3 cause the disorder trimethylaminuria, which results in a fishy body odor, due to the inability of

FMO3 to metabolize dietary byproduct trimethylamine to trimethylamine N-oxide

(Cashman et al., 2003). In a hyperlipidemic mouse model, FMO3 KO resulted in decreased lipid, glucose, and insulin content, while FMO3 overexpression had the opposite effect (Shih et al., 2015). Furthermore, recent studies have shown that FMO3

29 knockdown results in re-organization of cholesterol balance in mice (Warrier et al.,

2015). These results indicate a role for FMO3 in glucose and lipid homeostasis, suggesting that FMOs may have a greater endogenous role than previously hypothesized.

Mice lacking FMO1, FMO2, and FMO4 have reduced body mass and smaller fat deposits than wildtype (Veeravalli et al., 2014). This difference may be due to enhanced resting energy expenditure, fueled by increased fatty acid beta-oxidation (Veervalli et al., 2014).

Still, FMO2 likely has a limited role in humans, as it is encoded as a nonfunctional truncated protein in Caucasians and Asians due to a single nucleotide polymorphism, however, the functional protein is present in approximately 50% of sub-Saharan Africans and 26% of African-Americans (Dolphin et al., 1998; Veeramah et al., 2008).

Although much information is available on FMOs in mammals, less is known about FMOs in fish. Fish FMOs play a crucial role in osmoregulation via oxidation of trimethylamine to the corresponding oxide, and analysis of this reaction has led to the observation of FMOs in a variety of aquatic vertebrates and invertebrates (reviewed in

Schlenk, 1998). In addition to TMA, fish FMOs have been shown to metabolize a number of xenobiotics, including N,N-dimethylanaline, thioether pesticides, thiocarbamates, thiocarbamines and thioamides (Schlenk, 1998). It is hypothesized that fish FMOs evolved from the same phylogenetic lineage as mammalian FMOs, while many invertebrate FMOs evolved from a separate second lineage (Hao et al., 2009).

However, fish FMOs exhibit a monophyletic origin and cluster together, rather than with mammalian FMOs of the same nomenclature (Hao et al., 2009). Nonetheless, some mammalian FMO antibodies can recognize fish FMOs, including trout FMOs recognized

30 by a porcine FMO1 antibody (Schlenk & Buhler, 1993) and Japanese medaka FMOs recognized by FMO1 and 3 antibodies (El-Alfy & Schlenk, 2002).

To date, only one FMO has been well-characterized in fish. In 2008, Rodriguez-

Fuentes et al., cloned a hepatic trout FMO, which had an amino acid sequence with 74% identity to a putative puffer fish (Takifugu rubipes) FMO, 55% identity to putative zebrafish (Danio rerio) FMO and about 55% identity to human FMO1, 3 and 5. The protein contained all FMO identifying motifs found in other species and a promoter analysis suggested regulation of FMO expression via the glucocorticoid response element and osmoregulatory response element. Gene expression analysis on trout hepatocytes treated with NaCl and cortisol demonstrated an increase in the expression of this specific

FMO (Rodriguez-Fuentes et al., 2008). Juvenile rainbow trout were also acclimated to 8,

12, and 18g/l salinity, which induced expression of FMOA in liver, gills, kidney and olfactory tissue, but decreased expression of FMOB (Lavado et al., 2013). Furthermore,

FMO activity in the liver and kidney was increased at higher salinities (Lavado et al.,

2013). These alterations in FMO expression at differing salinities suggests that toxicity of xenobiotics metabolized by FMO may change depending on the salinity.

This finding corresponds with a body of work showing how hypersalinity regulates FMO activity in fish. Flounder (Platichthys flesus) acclimated to seawater then exposed to low salinity regimens of 25 and 15ppth had significantly decreased FMO activity in the gills and liver after two weeks (Schlenk et al., 1996). Juvenile rainbow trout injected with cortisol or the organic osmolyte, urea, for 48-72 hours had significantly greater hepatic FMO content and activity (El-Alfy et al., 2002). Modulation

31 of FMO activity has also been shown to contribute to toxicity of aldicarb and fenthion in fish as previously discussed (Lavado et al., 2009; Schlenk & El-Alfy, 1998; El-Alfy &

Schlenk, 1998; Wang et al., 2001).

In addition to osmotic stress, FMOs have also been shown to be regulated by other stress responses- oxidative stress and endoplasmic reticulum (ER) stress. In yeast

(Saccharomyces cerevisiae), there is a single FMO (yFMO), which is unable to oxidize amines and primarily participates in thiol oxidation (Suh et al., 1999; Suh et al., 1996).

This includes GSH, cystine, and cystamine, which suggests a role for yFMO in maintenance of redox potential (Suh et al., 1999). Other research indicates that yFMO activity may be redox regulated by oxidation of Cys353 (Suh et al., 2000). Furthermore,

ER stress and the subsequent unfolded protein response (UPR) regulate yFMO (Suh &

Robertus, 2000). Yeast treated with the reductant dithiothreitol (DTT) to induce disruption of protein folding, had increased yFMO activity; however, this was reduced to control levels upon deletion of HAC1 (homolog to mammalian IRE1 (inositol requiring enzyme1)), which regulate the UPR in yeast (Suh & Robertus, 2000). Additionally, yFMO knockout yeast suffered increased duration of the UPR under DTT stress and a greater accumulation of unfolded proteins (Suh & Robertus, 2000). This suggests that yFMO is regulated by and crucial for oxidative and ER stress responses in yeast, a trait which may be conserved in higher eukaryotes.

To date there have been a number of studies examining regulation of FMOs in mammals. Transcription factors HNF1a (hepatic nuclear factor -1a) and HNF4a have been found to regulate transcription of rabbit FMO1 (Luo & Hines, 2001). Human FMO3

32 in HepG2 cells was regulated by transcription factors NFY, Oct-1 and Pbx2/Hox (Klick

& Hines, 2007). Dioxins (TCDD) were shown to induce FMO1, FMO2 and FMO3 in an aryl hydrocarbon receptor (AhR)- dependent manner in both male and female mice (Tijet et al., 2006; Celius et al., 2008). Further analysis demonstrated AHR/ARNT binding to the FMO3 enhancer region (Celius et al., 2008), suggesting the ability of xenobiotics to modulate FMO expression in vertebrates. However, additional studies showed that mammalian FMO regulation is highly complex. A well-known inducer of the AhR, 3- methylcholanthrene (3MC), was able to induce FMO2 and 3 in mouse livers in vivo; however, induction was independent of the AhR (Celius et al., 2010). In Hepa-1 cells,

3MC and benzo(a)pyrene (BaP) induced FMO3, but not FMO1, 4 and 5, and this response required both AHR and ARNT, although no binding to the FMO3 promoter was detected (Celius et al., 2010). The Nrf2/antioxidant pathway was not involved in this response; however, p53 binding to the FMO3 promoter was detected (Celius et al., 2010).

This was confirmed by a recent study suggesting that FMO3 in HepG2 cells is not regulated by Nrf2 or oxidative stress (Rudraiah et al., 2016). Taken together, these results indicate that FMO in vertebrates may not be regulated by oxidative stress as in yeast.

Hypersalinity and Selenium Toxicity

Another compound shown to be oxidized by FMOs is SeMet. SeMet oxidation by

FMOs generates the SeMet oxide (Chen & Ziegler, 1994; Krause et al., 2006). This metabolite may undergo redox cycling to deplete GSH (Assmann et al., 1998).

Considering increases in FMO activity at greater salinities, it is possible that Se toxicity

33 could increase under hypersaline conditions. However, few studies have examined the effect of hypersalinity on Se toxicity to fish. One study found that juvenile rainbow trout acclimated to saltwater based on the SJRV, CA, were resistant to the acute toxicity of

180mg/kg dietary selenomethionine after 7 days in comparison to freshwater controls

(Schlenk et al., 2003). This difference was not due to differences in SeMet absorption

(Schlenk et al., 2003). In contrast, later studies on the combined effects of SeMet and salinity in Japanese medaka (Oryzias latipes) embryos indicated a synergistic effect

(Lavado et al., 2012). Medaka embryos were treated in freshwater or a salinity regimen of SJRV saltwater with 50µM SeMet for 24 hours during early development. It was found that increased salinity, significantly decreased hatchability of embryos exposed to

SeMet, while there was no difference in uptake. Furthermore, SeMet significantly decreased concentrations of reduced GSH in a salinity dependent manner (Lavado et al.,

2012). These changes were accompanied by an increase in SeMet oxide formation at higher salinities, which corresponded to increased FMO activity (Lavado et al., 2012).

This study indicated that embryo toxicity of Se may be increased with hypersalinity via induction of FMOs and that differences in susceptibility of various life stages may be present.

Mechanisms of Developmental Toxicity

Normal development is a tightly regulated, complex, spatiotemporal process, which when disrupted, can lead to developmental toxicity resulting in birth defects or embryonic mortality. The cellular and molecular mechanisms behind many teratogens

34 remain unknown and increasing effort is being applied towards their elucidation.

Recently, significant research has focused around the roles of oxidative stress, the unfolded protein response (UPR) and apoptosis in the pathogenesis of human disease; however, few studies have investigated their intersection in developmental toxicity.

Oxidative stress, the UPR and apoptosis all play key physiological roles in vertebrate development- in diverse processes from early cell proliferation to late organogenesis and morphogenesis. While many classical teratogens or developmental toxicants have very specific molecular targets, stresses caused by alterations in redox state, protein folding or apoptosis tend to have more general effects, resulting in a wide array of malformations.

Both exogenous and endogenous toxicants can disrupt these key processes and while these pathways have all been implicated in teratogenesis individually, recent work and future research will examine the intersection between the key roles these events play in cell signaling and cellular fate. This section will examine the current knowledge of the role of oxidative stress, the UPR and apoptosis in physiological development as well as in developmental toxicity, focusing on studies and advances in vertebrate model systems.

Oxidative Stress

Oxidative stress was defined originally in 1985 by Sies as “disturbances in the pro-oxidant/antioxidant systems in favor of the former”, however, in 2006, Jones suggested a new definition as “a disruption of redox signaling and control”. Considering that the current literature recognizes the integral role of redox in all forms of aerobic life, this new definition represents a more nuanced view. Exogenous toxicants can alter the

35 redox environment to disrupt development and cause teratogenesis. Although researchers often consider exogenous sources of reactive oxygen species (ROS) in developmental toxicity, ROS are produced endogenously and redox regulation plays a major role in normal vertebrate development. In order to understand the mechanisms behind oxidative teratogens, it is necessary to first grasp the physiological role of ROS in vertebrate development.

Generation of Endogenous ROS

Redox plays a critical role in cell signaling and homeostasis. This is particularly true during development, when processes are precisely timed and executed with little room for error. Reactive oxygen species (ROS) and reactive nitrogen species (RNS) are the primary oxidants produced endogenously. Endogenous ROS include super oxide

• - • anion ( O2 ), hydroxyl radical (OH ), and hydrogen peroxide (H2O2), while RNS are mainly nitric oxide (NO-) and peroxynitrite (ONOO-) (Figure 1.6).

36 Figure'1.'Endogenous'forma2on'of'ROS/RNS'

NO2'+'OH''

ONOOH' CO H+' 2' 9 NOS' NO' ONOO9' NO2'+'CO3 ' O2' Fe' H O OH9'+'OH' 2 2' NOX' SOD' 9' XO' O2

Figure 1.6. Endogenous formation of reactive oxygen species/reactive nitrogen species. Abbreviations: COX, cyclooxygenase; CYP450, cytochrome P 450; NOX, NADPH oxidase; NOS, nitric oxide synthase; SOD, superoxide dismutase; XO, xanthine oxidase.

It has been estimated that 90% of ROS are generated by the mitochondria

(Balaban et al., 2005). Oxidative phosphorylation occurs in the mitochondria and is the primary source of ATP is aerobic organisms, providing the majority of cellular energy. In the electron transport chain, electrons are passed between complexes, resulting in the ejection of hydrogen atoms from the inner mitochondrial space. This generates a membrane potential used to drive the formation of ATP. Leakage of electrons from complexes I and III generates superoxide (Balaban, 2005). Superoxide can then dismutate to form hydrogen peroxide, a reaction that occurs spontaneously or is catalyzed by superoxide dismutase (SOD) (Maier & Chan, 2002).

37 Another non-enzymatic reaction generating ROS is the Fenton reaction. In the

Fenton reaction, hydrogen peroxide reacts with mutivalence transition metals (iron, copper, manganese) to form the hydroxyl radical and the hydroxyl anion. The hydroxyl radical is highly reactive with a shorter half-life than other free radicals at 10-9 seconds, and often reacts with molecules in the vicinity of its creation (Pryor, 1986). Furthermore, in the Haber- Weiss reaction, superoxide can increase hydroxyl radical production by oxidizing [4Fe-4S] cluster containing enzymes, making more Fe(II) available for Fenton chemistry (Leonard et al., 2004).

ROS/RNS can be generated enzymatically via enzymes involved in cell signaling, defense or biosynthesis such as nitric oxide synthase (NOS), NADPH oxidase (NOX), xanthine oxidase (XO), lipoxygenases (LOX) and cyclooxygenases (COX). XO is a cytosolic enzyme catalyzing the conversion of hypoxanthine to xanthine and xanthine to uric acid, which uses oxygen as a cofactor and can generate superoxide and the hydroxyl radical (McCord & Fridovich, 1968; Kuppusamy & Zweier, 1989). LOX enzymes are involved in eicosanoid metabolism and catalyze the formation of peroxides on fatty acid substrates (Brash, 1999). 5-Lipoxygenase catalyzes the conversion of arachiodonic acid to leukotrienes can create ROS in lymphocytes, contributing to activation of NF-kb

(Soberman & Christmas, 2003). While it was initially believed that NOX was expressed exclusively in macrophages, it has since been discovered in tissues throughout the body and is another well-known generator of superoxide (DeCoursey & Ligeti, 2005). In response to foreign stimuli, these enzymes catalyze the conversion of large amounts of O2 to superoxide, a process that has been called respiratory or oxidative burst (DeCoursey

38 &Ligeti, 2005). This superoxide contributes to inflammation and is dismutated to H2O2

(DeCoursey & Ligeti, 2005). Cytochrome P 450 monooxygenase (CYP450) enzymes also contribute to ROS generation (Gillette et al., 1957; Nordblom & Coon, 1977).

CYP450s are a large class of microsomal, drug metabolizing enzymes that have several endogenous substrates. Uncoupling of the oxygen bound to the heme active site can generate superoxide, which rapid forms H2O2 (Narasimhulu, 1971; Grinkova et al.,

2013). NO is an important signaling molecule generated by one of three tissue specific isoforms of nitric oxide synthase (iNOS (immune cells), eNOS (endothelial cells) and nNOS (neurons)). NO will react readily with superoxide to form the peroxynitrite

(ONOO-) anion (Bergendi et al., 1999). In acidic media, ONOO- can react with hydrogen and fragment forming the hydroxyl radical, which is extremely reactive (Bergendi et al.,

1999).

Oxidative Damage from ROS/RNS

When ROS/RNS exceed detoxification capacities, oxidative damage to lipids, nucleic acids and proteins may occur. If this damage is irreparable, the cell will undergo cell death, in the form of either apoptosis (programmed cell death) or necrosis, depending on the extent of the damage. OH• is the most reactive ROS to DNA, and can damage

DNA via H abstraction or alkene addition. Hydrogen abstraction occurs from the methyl group of thymine and from the C-H bonds on 2-deoxyribose, leading to multiple oxidized products (Cooke et al., 2003). One major product of oxidative DNA damage is 8- hydroxyguanine, which has been shown to cause G to T and A to C substitutions (Cheng

39 et al., 1992). Oxidatively modified bases can also cause double strand breaks when reacting with C-H bonds on 2-deoxyribose. If DNA repair mechanisms are impaired or overwhelmed, these products can lead to adverse cellular effects, including mutations and cytotoxicity.

Lipids are also susceptible to oxidation by free radicals, particularly polyunsaturated fatty acids with two cis-double bonds separated by a methylene group

(Niki, 2014). Lipid peroxidation may occur as a chain reaction separated into three basic steps: initiation, propagation and termination (Porter et al., 1995). Initiation begins with hydrogen abstraction from a bis-allylic carbon, which rearranges to form relatively a stable cis, trans-pentadienyl radical. The addition of oxygen then forms a lipid peroxylradical, which begins propagation by abstracting a second H from adjacent lipids.

The process continues until termination, when two peroxyl radicals react together to form nonradical products or the products are scavenged by cellular antioxidants. Lipid peroxides can fragment to form reactive aldehydes and can alter membrane stability and permeability of key organelles, such as the mitochondria (Sultana et al., 2013). These cellular level effects translate to organ level pathologies and are implicated in a variety of diseases, including heart disease (Anderson et al., 2012), ocular degeneration (Njie-

Mbye, 2013) and Alzheimer’s disease (Sultana et al., 2013).

Oxidative damage to proteins can take several forms. Oxidation of the protein backbone following H abstraction by OH• can result in protein fragmentation, as can oxidation of glutamyl, aspartyl and prolyl side chains (Berlett & Stadtman, 1997). Sulfur containing amino acids, cysteine and methionine, are particularly sensitive to oxidation

40 (Berlett & Stadtman, 1997). These modifications can result in the loss of enzyme or protein function and the accumulation of unfolded proteins, which can impair cell function, leading to the unfolded protein response (Radak, 2001).

Protection Against ROS/RNS

Considering the many endogenous sources of ROS, aerobic organisms have developed enzymatic and non-enzymatic responses to mediate ROS levels (Figure 1.7).

Important enzymatic mechanisms include super oxide dismutase (SOD) and catalase

(CAT). There have been three isozymes of SOD identified in mammals, which are localized into different cellular compartments. Cu-ZnSOD is located in the cytosol,

MnSOD is in the mitochondria and ECSOD is in the extracellular space (thus far,

ECSOD has only been identified in mice) (Maier & Chan, 2002). In general, SODs stoichiometrically catalyze the dismutation of two superoxide anions with two hydrogen atoms to form H2O2 and O2 (Maier & Chan, 2002). CAT and other peroxidases then reduce two molecules of H2O2 to two water molecules and molecular oxygen (Kirkman

& Gaetani, 2007).

41 Figure'2.'Damage'and'Protec2on'from'ROS'

H2O' O O2' 2' Lipid' SOD' Prx' Lipid' Lipid'' Peroxida2on' CAT' H2O' OH' 2H2O'+'O2' H2O2' R9SH' DNA'Damage' GPx'

R9SS9R' GR' H2O'

Trx,'TrxR' Grx'

R9SH' R9SH' Figure 1.7. Damage and protection from reactive oxygen species. Abbreviations: CAT, catalase; GSH, glutathione; GR, glutathione reductase; Grx, glutaredoxin; GSSG, glutathione disulfide; Prx, peroxiredoxin; ReSH, protein thiol group; ReSSeR, protein with disulfide bond; SOD, superoxide dismutase; Trx, thioredoxin; TrxR, thioredoxin reductase.

Non-enzymatic mechanisms of detoxification include antioxidants, which

scavenge free radicals in the cell and/or reduce oxidized thiols. GSH is a tripeptide of

glycine, cysteine and glutamate, in which the glutamate is connected to the cysteine via a

gamma-carboxyl linkage (Lu, 1999). It exists in 2 forms, a reduced form (GSH) and an

oxidized disulfide form (GSSG). GSH is primarily located in the cytosol, with about 10%

in the mitochondria. GSH is present in the cell at milli-molar concentrations and is the

most abundant non-protein intracellular thiol (DeLeve & Kaplowitz, 1991). GSH can

reduce oxidants (ROS or strong electrophiles) either enzymatically or spontaneously

through its thiol group. Glutathione S-transferase (GST) can enzymatically add GSH to

electrophiles and disulfides. Glutathione reductase then regenerates oxidized glutathione

42 (GSSG) with NADPH as a cofactor. To combat ROS, selenoprotein glutathione peroxidases (GPx) can reduce H2O2 or lipid hydroperoxides with 2GSH, generating water and GSSG. There are four members of the GPx family, GPx1, GPx2, GPx3 and GPx4

(Imai & Nakagawa, 2003). GPx1 and GPx2 are localized to the cytoplasm, whereas GPx3 is primarily in the plasma of the kidney (Imai & Nakagawa, 2003). GPx4 is distinct from the other GPx isoforms. It is the only GPx that can reduce phospholipid hydrogen peroxides (Imai & Nakagawa, 2003).

Other important antioxidants include thioredoxins, glutaredoxins, peroxiredoxins and nucleoredoxins. These small proteins are key regulators of redox signaling through the reduction of disulfides and H2O2. Thioredoxins (Trx) contain the active site motif

Cys-Gly-Pro-Cys (Holmgren, 1968). Electrons from NADPH are transferred to thioredoxin reductase (TrxR), which reduces Trx to reduce disulfide bonds (Hanschmann et al., 2013). TrxR is another selenoprotein with a wide substrate specificity and is regulated by nuclear factor E2-related factor 2 (Nrf-2) (Kim et al., 2001). There are four glutaredoxins (Grx) with different cellular localization in mammals (Hanschmann et al.,

2013). Each antioxidant is ultimately part of the GSH pathway and the resulting oxidants are eventually reduced by GSH (Hanschmann et al., 2013). The six mammalian peroxiredoxins (Prx) are localized to a wide variety of cellular compartments (cytosol, nucleus, mitochondria, peroxisomes, lysosomes, and vesicles) (Hanschmann et al., 2013).

In contrast to Grxs and Trxs, Prxs reduce H2O2 to water and are regenerated by Trx or

GSH (Hanschmann et al., 2013). Finally, nucleoredoxins were first discovered in yeast in

1997 (Kurooka et al., 1997); however, homologs have since been identified across

43 vertebrates (Funato & Miki, 2007). They are localized to the cytoplasm and nucleus, and are involved in the regulation of several redox regulated transcription factors (Funato &

Miki, 2007).

Physiological role of ROS in Signaling and Development

Redox plays a key physiological role in vertebrate development; however, little progress has been made concerning the detailed function and nature of redox signaling in developmental processes. Alterations in the redox environment of developing embryos can lead to teratogenesis through dysregulation of cell signaling at lower levels and cell damage at higher levels. Here, we will examine the current knowledge base concerning the physiological role of ROS/RNS in development and teratogens that have been associated with oxidative stress.

Redox Regulated Transcription Factors

Redox regulation of key transcription factors is critical to development (Figure

1.8). The NF-κΒ family of transcription factors control expression of inflammatory cytokines, growth factors, redox regulated enzymes and apoptotic regulators. Knockout mice of various NF-κΒ family members are born with several defects, including postnatal immune defects (p50-/-, c-Rel-/-), multi-organ inflammation (RelB-/-), disrupted splenic architecture (p52-/-) and prenatal embryonic lethality (p65) (Ufer et al., 2010).

44 Figure'3.'Redox'Regulated'Transcrip2on'Factors'

ROS' ROS' NF9kB' IκB' Hypoxia'

Keap1'

Keap1' PHD' Ref1' NF9kB'

Nrf2' Nrf2' P'

HIF1α' ATF' AP91'

An2oxidants'and'' ARNT' Glucose'&'Iron'Met.,'' Cytokines,'Growth'Factors,' Detox.'Enzymes' Angiogenesis,'Prolif.' NF9kB' An29apopto2c'regulators' Nrf2' P' HIF1α'

Nucleus' Figure 1.8. Redox-regulated transcription factors. Abbreviations: AP-1, activator protein 1; ATF, activating transcription factor; HIF-1a, hypoxia-inducible factor 1a; Keap1, Kelch-like ECH-associated protein 1; NF-kB, nuclear factor kappa-light-chain-enhancer of activated B cells; Nrf2, nuclear factor (erythroid-derived 2)-like 2; PHD, prolyl hydroxylase; Ref-1, redox effector factor 1.

Nrf2 (nuclear factor E2-related factor 2) is a redox activated transcription factor

controlling gene expression through the antioxidant response element (ARE). Under

normal unstressed conditions, Nrf2 is targeted for degradation by Keap-1 ubiquitylation.

However, Keap-1 has several reactive cysteine residues, that impair its capacity for

ubiquitylation when oxidized, allowing Nrf2 to perform its function as a transcription

factor (Nguyen et al., 2009). Nrf2 regulates genes related to detoxification and

antioxidants, as well as lipid carbohydrate and heme metabolism (Hayes & Dinkova-

Kostova, 2014). While not embryonically lethal, Nrf2 knockout mice display auto-

45 immune mediated lesions in multiple tissues, and premature postnatal death (Ma et al.,

2006).

Hypoxia inducible factors (HIF) are important regulators of gene expression under hypoxia, a condition known to generate oxidative stress and discussed further below. HIFs consist of two subunits. The β subunit (HIFβ, also called aryl hydrocarbon receptor nuclear translocator (ARNT), a critical component of the aryl hydrocarbon receptor pathway) is stable, while the α subunit is mediated by O2 levels. Under homeostatic conditions, the α subunit is hydroxylated by a HIF specific prolyl hydroxylase (PHD), targeting it for ubiquitylation and degradation. PHD has low O2 affinity and is thus its activity is highly dependent on O2 levels. Under hypoxic conditions, HIFα is not hydroxylated and forms a heterodimer with HIFβ to modulate expression of target genes (Majmundar et al., 2010). Although the precise mechanism remains unclear, ROS also negatively regulate PHD, implicating HIF activity under oxidative stress (Majmundar et al., 2010; Eom et al., 2013). HIF regulates genes involved in glucose and iron metabolism, erythropoiesis, angiogenesis, and cell proliferation and differentiation (Maxwell et al., 2001; Yoon et al., 2006). In embryonic development, HIF is essential for vascular development during hypoxic conditions in neurogenesis and organogenesis (Trollmann & Gassmann, 2009). Furthermore, all HIF subunits have been shown to be essential for development as knockouts die between E10.5 and E12.5 (Ufer et al., 2010). HIF1α null embryos exhibited abnormal neural tube formation, increased apoptosis and a reduction in the number of somites, as well as impaired erythropoiesis

(Ryan et al., 1998; Yoon et al., 2006).

46 Redox effector factor-1 (Ref-1), also called apurinic/apryimidinic endonuclease-1

(APE-1) contributes to the base excision repair pathway of DNA lesions; however, it is also involved in modulation of redox sensitive transcription factors (Tell et al., 2009).

Oxidation of key cysteine residues impairs DNA binding capacity of Ref-1 but induces its ability to oxidize several transcription factors, including AP-1, NF-κΒ, and the

ATF/CREB family (Tell et al., 2009). Trx can then restore reduced Ref-1 (Tell et al.,

2009). Homozygous Ref-1 knockout embryos die by E6, while heterozygous embryos display increased markers of oxidative stress, which can be rescued by the addition of antioxidants (Ufer et al., 2010).

Antioxidants in Development

Considering the variable oxygen availability throughout development as well as potential onslaught from exogenous sources, it is necessary for vertebrates to have redundant antioxidant responses with high capacity to combat ROS. The redox potential of the developing embryo is closely linked to the GSH/GSSG and thioredoxin potentials.

GSH is essential for vertebrate development; mice lacking gamma-glutamylcysteine synthetase, an essential enzyme for GSH synthesis, die before E8.5 from apoptotic cell death (Shi et al., 2000). GSH levels are high in the maturing oocyte but steadily decrease following fertilization (Gardiner & Reed, 1994). In developing zebrafish (Danio rerio) embryos the redox potential as measured through the GSH/GSSG ratio fluctuated greatly

(Timme-Laragy et al., 2013). The cellular environment was reducing at fertilization (-

225mV), became increasingly oxidized until 12hpf (-170mV), after which point it

47 fluctuated between 24 and 48hpf, increasing to a stable reduced state at hatch at 72hpf (-

225mV) (Timme-laragy et al., 2013).

The Trx family has been shown to play a key role in development. During mouse gastrulation (E7), Trx1 mRNA expression was 11 fold higher than adults, but steadily decreased to adult levels by late gestation (Jurado et al., 2003). Trx1 and Trx2 knockout mice undergo embryonic lethality. Trx1 mice die around implantation (E3.5) following reduced proliferation of the inner cell mass (Matsui, 1996). In contrast, Trx2 mice develop normally until E8.5, but display an open neural tube at E10.5 and are reabsorbed at E12.5 (Nonn et al., 2003). This corresponds with the onset of mitochondrial maturation and oxidative phosphorylation, suggesting a key role for Trx in regulation by endogenous

ROS (Nonn et al., 2003). Trx reductase null mutations are also embryonically lethal, with

TrxR2 being essential for proper cardiac development, and TrxR1 not having a role

(Jakupoglu et al., 2005).

Prxs are also essential for H2O2 metabolism and while null mice are viable, Prx1 plays a key role in motor neuron differentiation through reduction of a disulfide bond in

GDE2 (Yan et al., 2009). Another member of the thioredoxin family, Grx also plays a complex role in vertebrate development. A study in zebrafish found Grx2 regulation of

CRMP2 thiols controlled axonal outgrowth, neuron survival and brain development

(Brautigam et al., 2011).

Many enzymes that diminish oxidative stress, such as CAT, SOD, glutathione reductase and GPx, are present at lower levels in developing embryos than in adult organisms, further complicating the embryo’s ability to cope with oxidative insult (Ufer

48 et al., 2010). CAT null mice are viable with no visible defects, yet reduced CAT levels may indicate reduced ability of the embryo to cope with ROS generating toxicants (Ho et al., 2004).

GPx enzymes are important components of the GSH system. While GPx1, GPx2 and GPx3 knockout embryos are viable with no visible defects, indicating that they are dispensable for embryonic development, GPx4 plays a more complex role (Ufer & Wang,

2011). GPx4 null embryos are embryonically lethal, undergoing resorption by day E8.5

(Imai et al., 2003). GPx4 mRNA expression has been detected beginning at E6.5, and at

E8.0 in the neural tube, extending in to the developing brain (Borchert et al., 2006). It is later detected in the developing organs and limbs (Schneider et al., 2006). GPx4’s role in these tissues appears to be the prevention of mitochondrial membrane oxidation and subsequent apoptosis inducing factor (AIF) dependent apoptosis (Ufer & Wang, 2011).

However, it has also been shown to modulate Nrf2 and NF- κΒ activity, indicating possible contributions to cell signaling (Ufer & Wang, 2011).

ROS/RNS Generating Enzymes in Development

Knock out of many oxidative enzymes, such as NADPH oxidase, NOS, XO and some isoforms of LOX and COX produce viable offspring, indicating that they are not essential for embryonic development (Ufer et al., 2010). However, 12-LOX knock out mice displayed skin defects that led to neonatal death, implicating 12-LOX in development of skin structure (Epp et al., 2007). Furthermore, COX-2 knockout embryos displayed defective kidney development and postnatal myocardial fibrosis, accompanied

49 by an increase of neonatal mortality (Dinchuk et al., 1995). NOS null embryos were viable even though NO is an important signaling molecule, and has been shown to be involved in physiological vascular development (Nath et al., 2004). eNOS expression increased steadily throughout development from approximately 3,000 to 14,000 copies of mRNA/106 copies of GAPDH mRNA, while iNOS and nNOS were expressed between

50 and 400 copies of mRNA/106 copies of GAPDH mRNA (Ufer et al., 2010).

Oxidative Stress and Teratogenesis

The complex nature of redox signaling and control throughout development indicates the potential that even relatively minor alterations to redox potential may generate significant developmental effects. While mild imbalances in the redox potential impact redox regulated signaling, higher levels of oxidative stress can induce teratogenesis through cytotoxicity. Oxidative stress is implicated as the mechanism of action of many developmental toxicants (Hansen, 2006). Environmental exposures to chemicals can cause oxidative damage leading to developmental toxicity. Many pesticides, such as organophosphates, cause acute toxicity via inhibition of acetylcholinesterase, and are also developmental toxicants. For instance, neonatal rats are more sensitive to the organophosphate, chlorpyrifos, than adults (Chakraborti et al.,

1993). Chlorpyrifos has been shown to cause neurobehavioural defects following early exposure in mammals (Chakraborti et al., 1993; Chanda & Pope, 1996). It has also been linked to aberrant behavior, spinal curvature, pericardial edema and mortality in zebrafish

50 embryos (Kienle et al., 2009), whereas exposed frogs displayed tail flexure and decreased neuromuscular activity (Bonfanti et al., 2004).

The mechanisms of chlorpyrifos developmental toxicity are far more complex than its acute toxicity. In addition to acetylcholinesterase inhibition, organophosphates have also been shown to interfere with neuronal cell differentiation and replication, synapse formation, neurite cell growth and other neurodevelopmental processes (Flaskos

& Sachana, 2011). Additionally, oxidative stress has been implicated in chlorpyrifos toxicity. In PC12 cells, a model for neurodifferentiation, the addition of 2-50µM chlorpyrifos significantly induced ROS (Crumpton et al., 2000). This effect was ameliorated with the addition of vitamin E (Slotkin et al., 2007). Chlorpyrifos treatment during gestation and on postnatal days 1-4 did not increase lipid peroxidation in rats; however, oxidative stress was generated following treatment during the peak of neuronal cell differentiation and synaptogenesis (second postnatal week) (Slotkin et al., 2005). It was further discovered that 2mg/mL chlorpyrifos induced expression of oxidative stress related genes following treatments to 7-day-old rats (Ray et al., 2010). Understanding the mechanism of action of chlorpyrifos and phenytoin teratogenicity informs future drug and pesticide design to mitigate potential negative effects. However, many toxicants act via a range of mechanisms and can disrupt other normal cellular processes both directly and via oxidative stress, including protein folding.

51 The Unfolded Protein Response

The endoplasmic reticulum (ER) is the major site of post-translational processing of proteins, which amounts to about 30% of the proteome, and performs several other critical functions including calcium storage and metabolic processes (Hetz, 2012). The fidelity of the ER must be able to handle large secretory loads. Secretory cells produce high volumes of proteins for processing and the same can be said of cells during embryonic development at certain stages. Disruption of proper ER function can lead to the accumulation of unfolded proteins, a state called ER stress. ER stress results in the activation of a complex signaling pathway called the unfolded protein response (UPR).

Through alterations of gene expression and protein translation, the UPR optimizes the ER to correctly manage protein folding and initiate apoptosis or autophagy in cells that are irreversibly damaged. UPR mediators and signaling proteins respond not only to exogenous toxicants that disturb ER function, but also play key roles in embryonic development and organ physiology (Cornejo et al., 2013). In this section I address the physiological role of the UPR in vertebrate development and how induction and perturbation of the UPR can result in teratogenesis.

Signal Transduction of the Three Branches of the UPR

There are three branches of the UPR, each with a distinct signal transduction pathway culminating in transcription of UPR target genes (Figure 1.9). The branches are controlled by inositol-required enzyme 1 (IRE1, α and β), protein kinase RNA-like ER kinase (PERK), and activating transcription factor 6 (ATF6, α and β).

52 Figure'4.'The'3'branches'of'the'UPR'

BiP'

BiP' ATF6' BiP' PERK' PERK' BiP' PERK' IRE1' IRE1' IRE1' ER'Lumen' P' P' P' P' P' Cytoplasm'

eIF2α' eIF2α'P'

Golgi' XBP1'mRNA' RIDD' Transla2onal' AZenua2on' XBP1' ATF4'

ATF6'

ERAD,'Chaperones,'Redox,' CHOP,'AA'transport,'Redox,' ERAD,'Chaperones,'' XBP1' RNA/DNA'processing'' ATF4' Mitochondria' ATF6' XBP1'

Nucleus' Figure 1.9. The three branches of the unfolded protein response. Abbreviations: ATF4, activating transcription factor 4; ATF6, activating transcription factor 6; BiP, immunoglobulin heavy-chain-binding protein; eIF2a, eukaryotic initiation factor 2a; IRE1, inositol-required enzyme 1; PERK, protein kinase RNA-like ER kinase; RIDD, regulated IRE1-dependent decay; SP1/2, serine proteases 1 and 2; XBP-1, X-box protein 1.

The most extensively studied branch of the UPR, IRE1α is also the most evolutionarily conserved and is the only branch present in yeast (Mori, 2009). IRE1α is both a kinase and an endonuclease (Tirasophon et al., 1998). When activated by the UPR,

IRE1α dimerizes and undergoes autotransphosphorylation, leading to activation of the cytosolic RNase domain (Hetz, 2012). It then splices a 26 base pair region from the center of X-box binding protein-1 (XBP-1) mRNA, allowing the translated protein to fulfill its role as a transcription factor (Calfon et al., 2002; Yoshida et al., 2001). XBP-1 targets a wide variety of genes, 29% of which are involved in protein trafficking,

53 processing and secretion and 14.8% of which are involved in protein biosynthesis and post-translational modifications (Acosta-Alvear et al., 2007). Thus, through XBP-1 the

IRE1α branch of the pathway is primarily responsible for the secretion and degradation of unfolded protein through ER-associated degradation (ERAD). However, in addition to genes impacting ER function, XBP-1 also controls expression of genes for cell growth and differentiation, RNA processing, signal transduction, ion channels, DNA replication,

DNA repair, and redox potential (Acosta-Alvear et al., 2007). IRE1 also contributes to the UPR through Regulated IRE1-Dependent Decay (RIDD), a process by which IRE1 mediates decay of certain mRNA within the ER (Hollien & Weissman, 2006). It is through the RIDD pathway that IRE1 regulates cell fate during ER stress (Chen &

Brandizzi, 2013).

ATF6α and β are part of the CREB/ATF family of proteins and are constitutively expressed transmembrane ER proteins (Yoshida et al., 1998a). When unfolded proteins accumulate in the ER, ATF6α translocates to the golgi apparatus, where it is cleaved by site-1 and site-2 proteases into its active form (Nadanaka et al., 2004; Ye et al., 2000). It then can enter the nucleus to act as a transcription factor for ERSE regulated genes

(Okada et al. 2003; Yoshida et al., 1998a). These include ERAD proteins, various ER chaperones, and XBP-1.

Finally, the PERK branch of the UPR is primarily responsible for translational attenuation in order to reduce the load of unfolded proteins. Upon activation, PERK oligomerizes and phosphorylates the translation initiation factor eIF2α (eukaryotic initiation factor 2), inhibiting its activity and attenuating protein translation (Harding et

54 al., 2000). Additionally, PERK regulates other transcription factors and UPR mediators.

Without available eIF2α some mRNAs with specific sequences are preferentially translated, including activating transcription factor 4 (ATF4), a member of the same family of transcription factors as ATF6 (Walter & Ron, 2011).

ATF4 has a wide variety of gene targets and can function as both a transcriptional activator and repressor. Its repressor activities are less well studied and include negative regulation of the cAMP response element (Karpinski et al., 1992) and memory storage

(Chen et al., 2003). ATF4 can function as a transcriptional activator both alone and as a heterodimer with other transcription factors. ATF4’s transcriptional targets are wide- ranging and include genes involved in amino acid transport and metabolism, transcription, mitochondrial function and redox/detoxification (Ameri & Harris. 2008).

Induction and Regulation of the UPR

ER stress can be generated from disruption of calcium homeostasis, ER-localized oxidative stress, or impaired vesicular trafficking, or protein degradation (Hetz, 2012). In studies of ER stress, pharmacological agents are generally used to induce the UPR. These include thapsigargin (Tg), which inhibits Ca2+-dependent ATPase, tunicamycin (Tm), a protein glycosylation inhibitor, and dithiothreitol (DTT), a disruptor of disulfide bonds

(DuRose et al., 2006).

The dynamic nature of the response to ER stress is in part due to varied forms of regulation at several levels. It was originally proposed that the UPR was governed by the inducible ER resident chaperone BiP (Immunoglobulin-heavy-chain-binding protein, also

55 known as GRP78 or HSPA5). An early and pioneering study by Bertolotti et al. (2000) demonstrated BiP was able to complex with the luminal domains of IRE1 and PERK under ER homeostasis and dissociated from the complex under ER stress. Furthermore, the activities of IRE1 and PERK were attenuated when BiP was overexpressed (Bertolotti et al., 2000). In the case of ATF6, BiP binding masks golgi-localization signals, preventing constitutive transactivation (Shen et al., 2002). Once the UPR is activated, BiP dissociates from each branch to perform its duties as a chaperone for unfolded proteins.

However, further research suggests that BiP is not the only regulator of UPR signaling. For example, it has been suggested that decreased glycosylation of ATF6 may control its own activation. This mechanism of ATF6 regulation was proposed following observations that depletion of Ca2+ in the ER induced newly synthesized and partially- glycosylated ATF6 (Hong et al., 2004). Glycosylated ATF6 was no longer able to associate with calreticulin, resulting in faster translocation of ATF6 to the golgi and stronger UPR activation (Hong et al., 2004). In addition, reduction of disulfide bonds within ATF6 also facilitated its transport to the golgi, but was not sufficient for its activation (Nadanaka et al., 2007).

Several studies have been performed characterizing the kinetics of UPR activation. In one study, DTT caused 100% activation of PERK, ATF6 and IRE1 within 1 hour following treatment. PERK and IRE1 were equally as sensitive to Tg, while ATF6 was not (DuRose et al., 2006). Furthermore, all branches took longer (2 to 5 hours) to reach full activation following treatment with Tm (DuRose et al., 2006). PERK signaling

56 was also maintained under prolonged ER stress caused by Tm and Tg, while IRE1α and

ATF6 signaling peaked at 4 hours post treatment then declined (Lin et al., 2007).

The UPR in Physiological Development

Although once thought to merely function as a stress response, evidence supporting a role for the UPR in developmental physiology is increasing. UPR mediators have been implicated in diverse developmental pathways and many are essential for embryonic development, demonstrated by non-viable embryonic knockouts. While the primary role of the UPR is mediation of protein folding and load in secretory cells, various UPR transcription factors and signal transduction pathways are involved in several other processes (Table 1.2). UPR mediators have been implicated in hematopoiesis, osteogenesis, chondrogenesis, angiogeneisis, neurogenesis, hepatogenesis, as well as pancreatic and lens cell development. In this section we will examine each branch of the UPR and its role in physiological development.

Table 1.2. Role of unfolded protein response genes in physiological development

Gene Name Embryonic Lethal Developmental Process Knockout? IREα Yes Immune cell differentiation, hepatogenesis, chondrogenesis, adipogenesis IRE1β No XBP1 Yes Immune cell differentiation, hepatogenesis, zymogen cell differentiation, adipogenesis ATF6α/β Individual KO of α or Neurogenesis, hepatogenesis β- No Double α/β KO- Yes PERK No Pancreatic β-cell differentiation, osteogenesis ATF4 Yes Osteogenesis, lens formation, hematopoiesis BiP Yes Early development, Neurogenesis

57 IRE1/XBP-1

The function of the IRE1 and XBP-1 axis in development is the best studied of the three pathways. IRE1α knockout mice are embryonic lethal and full IRE1α knockout embryos die between E12.5 and E13 (Zhang et al., 2005). Embryos at E13 demonstrated fetal liver hypoplasia and decreased proliferation of hematopoietic stem cells (HSCs)

(Zhang et al., 2005). Subsequent studies on IRE1α null mice have shown severe defects to the labyrinth in the placenta mediated by XBP-1 independent decreased expression of

VEGF-A (Iwawaki et al., 2009). Interestingly, following selective IRE1α expression in the placenta, embryos displayed no liver hypoplasia and were viable (Iwawaki et al.,

2009). In contrast to IRE1α, IRE1β knockout mice are viable (Bertolotti et al., 2001).

However, a study in Xenopus laevis frog embryos found IRE1β to be required for mesoderm development, suggesting evolutionary alterations in function (Yuan et al.,

2008).

XBP-1 mRNA can be detected in the nucleus and cytoplasm as early as the one cell stage (Zhang et al., 2012). Similar to IRE1α null mice, XBP-1-/- mice are not viable, with liver hypoplasia and lethality beginning at E12.5 (Reimold et al., 2000). Livers from knockout mice had decreased proliferation and increased apoptosis of hepatocytes

(Reimold et al., 2000). This corresponded to decreased levels in several acute phase proteins (Reimold et al., 2000).

The UPR has been recognized as a key regulator in secretory cells, and IRE1 and

XBP-1 have been implicated in immune cell development and function. While rag-/- mice transplanted with IRE1α-/- HSCs were able to produce pro–B cells, as well as

58 erythroid, myeloid, and thrombocyte lineages, B-cell receptors were not detected and a reduction in B-cell Ig VDJ recombination was observed (Zhang et al., 2005).

Furthermore, though spliced XBP-1 was not necessary for early B lymphocyte differentiation, it was necessary and sufficient for terminal plasma cell differentiation

(Reimold et al 2001; Zhang et al., 2005). Ectopic XBP-1 expression induced B cell differentiation into plasma cells (Reimold et al., 2001) and induction of XBP-1 spliced mRNA correlated with Ig heavy chain secretion during plasma cell differentiation (Zhang et al., 2005) and induced interleukin-6 (IL-6) (Iwakoshi et al., 2003). The PERK branch of the response was not necessary for this response (Zhang et al., 2005).

The XBP-1 branch of the UPR has also been implicated in several other developmental pathways. For example, XBP-1 is responsible for the development of zymogen cells in the gastric epithelium (Huh et al., 2010). IRE1 regulated chondrocyte differentiation, via inhibition of UPR induced apoptosis following induction by BMP2

(bone morphogenic protein 2) (Han et al., 2013). Finally, XBP-1 transcription is induced by C/EBPβ, an early adipogenic factor, and plays a key role in adipogenesis, as cells lacking XBP-1 and IRE1α display defects in adipogenesis (Sha et al., 2009). It is clear that the IRE1/XBP-1 signal transduction pathway plays a key role in physiological development through both modulation of secretory cell differentiation and function, and through transcriptional regulation.

Considering the diverse network of genes regulated by XBP-1 and the large number of mRNA’s subject to RIDD, the possible roles for IRE1 and XBP-1 in vertebrate development are limitless. XBP-1 targets include genes involved in cell growth

59 and differentiation, including several cyclin-dependent kinases (Acosta-Alvear et al.,

2007). The role of RIDD in physiological development is still unclear. However, several genes essential for embryonic development are RIDD targets. For example, homeo box

B4 (HOXB4) is a RIDD target (Hollien et al., 2009) and is essential for skeletal development in mice (Ramirez-Solis et al., 1993), and HSC differentiation from embryonic stem cells (Fan et al., 2012).

ATF6

ATF6α and ATF6β single knockouts develop normally, while embryos with

ATF6α/β double knockouts are not viable, suggesting an overlap in function (Yamamoto et al., 2007). Mice lacking ATF6α, but not ATF6β, develop liver steatosis when challenged with ER stress and have a depressed UPR (Yamamoto et al., 2010). ATF6α/β double knockouts have been studied in the model fish, Japanese medaka (Ishikawa et al.,

2013). Double knockouts were found to undergo more severe physiological ER stress in the brain, otic vesicles and notochord, accompanied by significant decreases in expression of BiP and other chaperones (Ishikawa et al., 2013). Double knockout embryos had severely degenerated notochords and displayed a similar phenotype to BiP knockout embryos (Ishikawa et al., 2013). Thus, the role for ATF6 in physiological development appears to be tied to the reduced ability of secretory cells to induce chaperone proteins, thus leading to degeneration.

Other studies concerning the role of ATF6 in physiological development are limited. In myoblasts ATF6, was exclusively responsible for the induction of apoptosis

60 during muscle development (Nakanishi et al., 2005). Other members of the ATF6 family sharing structural similarities and localized to the ER have also been shown to play a role in development. OASIS (old astrocyte specifically induced substance) knockout mice display defects in bone formation and bone weakness caused by a decrease in a type I collagen, Col1a1 (Murakami et al., 2009). OASIS was found to bind to a UPR response element (UPRE) sequence in the osteoblast Col1a1 promoter (Murakami et al., 2009).

Furthermore, OASIS plays a key role in astrocyte differentiation (Kondo et al., 2005).

Thus the ATF6 family may play a greater role in development than previously realized and further research is required to fully understand its role in physiological development.

PERK/eIF2α/ATF4

The PERK pathway is essential for many facets of embryonic development.

PERK knockout mice develop severe hyperglycemia and lose pancreatic insulin-secreting

β cells postnatally (Zhang et al., 2006). PERK is required for β cell proliferation and differentiation during development, but not for maintenance in the adult stage (Zhang et al., 2006). Knockouts also display skeletal dysplasia, reduced growth and impaired locomotor activity, accompanied by decreased bone mineralization and abnormal expression of collagen 1 (Zhang et al., 2002). These skeletal defects may be related to the important role of ATF4 in skeletal development.

ATF4-/- mice have reduced and delayed bone mineralization, and reduced adult bone mass (Yang et al., 2004). ATF4 deficiency also reduced Type 1 collagen, as well as osteocalcin, which is not only a late stage marker for osteoblast differentiation, but also

61 possesses an ATF4 binding site in its promoter (Yang et al., 2004). Similarly, BMP2 also induced mild ER stress through the PERK/ATF4 pathway in osteogenesis (Saito et al.,

2011). Alternatively, ATF4 may modulate osteogenesis through RSK2 (Yang et al.,

2004) or regulate growth plate chondrocyte proliferation and differentiation as a transcription factor of Indian hedgehog (Ihh), a factor required for skeletal development

(Wang et al., 2012).

In addition to skeletal deformities, mice lacking ATF4 exhibited severe microphthalmia caused by a complete absence of the lens (Tanaka et al., 1998; Hettmann et al., 2000). They displayed normal lens development until E14.5, at which point the lens degenerated due to severe apoptosis, suggesting a role for ATF4 in later stages of lens fiber cell differentiation (Tanaka et al., 1998). Recent studies have confirmed that

UPR is activated during physiological lens development (Firtina & Duncan, 2011). Other abnormalities associated with ATF4 deficient mice include defects in postnatal hair growth, reduced body size and impaired hematopoiesis resulting in severe anemia

(Masuoka & Townes, 2002).

BiP

As an ER chaperone, BiP plays diverse physiological roles apart from its important role in the UPR. BiP expression in mouse embryonic development begins at the 2-cell stage and its expression is barely detectable until the blastocyst stage when BiP protein levels increase significantly (Kim et al., 1990). BiP null mutations were lethal to embryos at the peri-implantation stage (E7.5) and exhibited decreased proliferation and

62 increased apoptosis of the inner cell mass in comparison to wild type embryos (Luo et al.,

2006). This suggests an important role for BiP in early development that is potentially linked to the increased secretory protein load during this crucial period of proliferation.

Little research has been performed on the role of BiP in physiological development past the peri-implantation stage. Mimura et al. (2007) generated knock-in

BiP mice with a mutant retrieval sequence that disrupted BiP return to the ER from the secretory pathway. Embryos containing this mutant BiP were unable to secrete pulmonary surfactants from alveolar cells and died shortly after birth from respiratory failure. Furthermore, these mice displayed significantly smaller brains, and defective neocortical stratification possibly resulting from decreased reelin secretion, a glycoprotein that regulates neuronal migration (Mimura et al., 2008). A genetic screen for defects in thalamocortical development found that mutated BiP delayed axon extension, caused over fasciculation and impaired corticostriatal boundary crossing (Favero et al.,

2013). These studies indicate an important role for BiP in neural development, yet more research is needed to fully understand its precise functions.

The UPR in Developmental Toxicity

Considering the diverse role of the UPR in all stages of development, toxicants that interfere with the physiological UPR or agents that induce the UPR in developing embryos have the potential to generate developmental toxicity and birth defects.

Induction of ER stress is not limited to exogenous toxicants, the UPR also plays a key role in chondrodysplasia caused by mutations in key extracellular matrix components

63 (Patterson & Dealy, 2014). Mutations in ECM components such as type II collagen, collagen X and cartilage-matrix oligomeric protein (COMP), often result in impaired protein folding, processing or export. As healthy chondrocytes typically secrete large volumes of these ECM proteins, the mutated unfolded proteins are retained in the ER and generate ER stress (Arnold & Fertala, 2013). Activation of the UPR typically results in reduced chondrocyte proliferation, increased apoptosis or altered differentiation, culminating in the observed short stature and deformities of the face and joints in many types of chondrodysplasia (Patterson & Dealy, 2014). For instance, a key glycine to serine mutation in collagen 2A1 results in shortening of the femurs and humerii, suggesting disruptions in endochondral bone ossification (Liang et al., 2014).

Researchers observed significant increases in the gene expression of UPR mediators, decreased proliferation and increased apoptosis resulting in elimination of the hypertrophic zone of developing endochondral bones (Liang et al., 2014).

Although there are few studies currently investigating the effects of exogenous toxicants on the UPR in chondrodysplasia, they may also have the potential to induce similar effects. ATF4 has been shown to play a role in physiological chondrocyte differentiation (Wang et al., 2012). Alteration of that pathway could result in toxicity to chondrocytes. For example, the UPR was induced in zebrafish displaying spinal curvature following treatment with silver nanoparticles (Christen et al., 2013).

Furthermore, induction of ER stress following targeted mutation of collagen 10a1 in terminally differentiated hypertrophic chondrocytes causes reversion into pre-

64 differentiated cells, which have delayed ossification, generating a chondrodysplasia phenotype (Tsang et al., 2007).

Apoptosis

Programmed cell death, also known as apoptosis, is a tightly regulated process by which multicellular organisms dispose of unwanted cells. Apoptosis can occur naturally during embryogenesis or as a result of cell damage or stress. Apoptotic cells are characterized by overall cell shrinkage and membrane blebbing into small apoptotic bodies. This is accompanied by degradation of intracellular components, including condensation and fragmentation of DNA, fragmentation of organelles and proteolysis of many proteins (Taylor et al., 2008).

The primary effectors of the apoptotic response are caspases, cysteine proteases that cleave their substrates following an aspartic acid residue (Kumar, 2007). All caspases are present in cells as inactive zymogens and exist as Initiator or Effector caspases.

Initiator caspases 9, 2, 8 and 10 are the first to become activated in the apoptotic pathway

(Fuentes-Prior & Salvesen, 2004). The N-terminal sequences of initiator caspases contain caspase recruitment domains (CARD) or death effector domains (DED), which allow them to form oligomeric complexes with adaptor proteins Fas associated death domain

(FADD) or Apoptotic protease activating factor 1 (Apaf-1) (Fuentes-Prior & Salvesen,

2004). It has been proposed that dimerization of initiator procaspase monomers allows for activation and cleavage into the caspase form (Boatright et al., 2003). The active form will then go on to process the effector caspases.

65 Effector caspases 3, 7 and 6 lack the recruitment domains of the initiators and are responsible for the proteolytic processing of proteins during the apoptotic response. There have been nearly 1,000 natural substrates of effector caspases identified (Crawford &

Wells, 2011). Proteolysis by effector caspases can lead to protein loss of function, such as with structural proteins and proteins involved in transcription and translation, gain of function, such as upstream apoptotic modulators, change of function, and change of localization (Crawford & Wells, 2011). Furthermore, many targets are thought to be merely bystander proteins, whose cleavage has no immediate effect on the apoptotic response (Crawford & Wells, 2011). Caspase substrates have been found to be involved in cell adhesion, cell structure, nuclear structure, cell cycle, DNA synthesis, cleavage and repair, DNA transcription, RNA synthesis and splicing, protein translation, modification and degradation, lipid metabolism, and neurodegeneration (Fischer et al., 2003).

Extrinsic and Intrinsic Apoptosis

Apoptosis can be initiated by both intracellular and extracellular signals. The extrinsic pathway is initiated by natural killer cells, which release proapoptotic ligands that bind to death receptors on the surface of infected or damaged target cells (Ashkenazi,

2008). This event initiates a signal cascade that recruits procaspase 8, procaspase 10

(humans only) and FADD to form the death inducing signaling complex (DISC)

(Askenazi, 2008). The aggregation of several molecules of caspase 8 and 10 within DISC induces their auto-processing. These initiator caspases are then able to cleave the effector caspases 3 and 7 and propagate the apoptotic response (Figure 1.10).

66 Figure'5.'The'extrinsic'and'intrinsic'pathways'of'apoptosis' Bcl92'

Fas9L' BAX' BAK' FAS' FAS' FADD' FADD'

C9FLIP' CytC' DISC' PC10' PC8' PC10' PC8' CytC'

CytC' CytC' PC9' Apaf91' PC9' Apoptosome' PC9' CytC' Casp8' Casp10' Casp9'

PC3' Casp3' Apoptosis'

Figure 1.10. Extrinsic and intrinsic apoptosis. Abbreviations: Apaf-1, apoptotic protease activating factor 1; BAK, Bcl-2 homologous antagonist/killer; BAX, BCL2-associated X; Bcl-2, B-cell lymphoma 2; c-FLIP, cellular FLICE (FADD-like IL-1b-converting enzyme)- inhibitory protein; Casp, caspase; CytC, cytochrome C; FADD, Fas-associated protein with death domain; FAS-L, FAS ligand; PC, procaspase.

In contrast, the intrinsic pathway of apoptosis, also called the mitochondrial

pathway, is initiated by cell stress or damage, such as oxidative stress or DNA damage

(Figure 1.10). The anti-apoptotic Bcl-2 (B-cell lymphocyte-2) family of proteins tightly

regulates this pathway. This family inhibits the proapoptotic family of proteins, BH3-

only (Bcl-2 homology 3 domain only). These proapoptotic molecules regulate another

family of BH domain proteins, which include, BAX and BAK. As BH3-only protein

signaling overcomes Bcl-2 inhibition, BAX and BAK assemble into oligomers in the

mitochondrial membrane, promoting the release of cytochrome C into the cytoplasm

67 (Taylor et al., 2008). Cytochrome C release initiates assembly of the apoptosome,

containing Apaf-1 and oligomers of procaspase-9, which then goes on to activate effector

caspases and propagate the response.

Apoptosis in Physiological Development

Apoptosis regulates many developmental processes including morphogenesis,

deletion of vestigial structures, cell number regulation and elimination of dangerous cells

as the broad functions of cell death in development (Fuchs and Stellar 2011) (Figure

1.11). These categories characterize the variety of important roles that apoptosis can play

Figure'6.'Apoptosis'in'Physiological'Development''in development.

Apoptosis'

Morphogenesis' Regula2on' Removal' Dele2on'

• Limb' • Plasma'Cell' • Infected' Forma2on' Selec2on' • Ves2gial' Cells' • Endocardial' • Neuron' Structures' • Damaged' Cushion' Selec2on' Cells'

Figure 1.11. Apoptosis in physiological development.

Apoptosis in limb development is mediated by BMP and signaling forms the

joints and the separation of the digits (Mori et al., 1995; Zou & Niswander, 1996; Macias

68 et al., 1997). Apoptosis also shapes structures by removing cells to form cavities and the four-chambered heart from the endocardial cushion (Abdelwahid et al., 2002).

Furthermore, in endochondral bone formation, chondrocytes undergo apoptosis to stimulate calcification and blood vessel perfusion (Gibson, 1998).

Overproduction of cells during development is common and apoptosis is necessary to regulate appropriate cell number. During immune system development lymphocytes are subjected to multiple checkpoints. Random genetic recombination generates unique antigen receptors, which must be properly aligned and evaluated for auto-reactivity (Rathmell & Thompson 2002). If the junctions are not joined in frame or the receptor is auto-reactive the developing T or B cell will undergo apoptosis (Rathmell

& Thompson, 2002).

Apoptosis also plays an important role in the development of the vertebrate nervous system. From 30% to 80% of neural and glial cells undergo apoptosis throughout the course of development (Buss et al., 2006). It has been suggested that cell death in the developing nervous system functions to ensure sufficient innervation, regulate cell number, and correct errors in cell location or axonal pathfinding (Buss et al., 2006).

On the cellular level, knockouts of apoptotic mediators and effectors can result in embryonic mortality or deformities, demonstrating the key role for apoptosis in development. For instance, knockout of caspase 3, caspase 9 or Apaf-1 results in mortality and forebrain malformations, suggesting an important caspase-dependent role for intrinsic apoptosis on brain development (Yoshida et al., 1998b; Kuida et al., 1998;

Kuida et al., 1996). In the extrinsic pathway, knockout of caspase 8 or the FAS adaptor

69 protein, FADD, generates cardiac defects that result in embryonic lethality (Yeh et al.,

1998; Varfolomeev et al., 1998). Knockout of other apoptotic proteins results in disrupted immune cell function autoimmune diseases.

Apoptosis in Developmental Toxicity

Although apoptosis is a key event in development, cell death can severely impact formation and growth during development by acting on either actively proliferating cells or by expanding the area of cells already programmed for apoptosis. In early development, different cell lineages may confer differing susceptibility to apoptosis, depending on their metabolic requirements and microenvironments (Pampfer, 2000). It is generally accepted that if cell damage is too severe and ATP stores entirely depleted, the cell will undergo necrosis. However, apoptosis under less severe exogenous stress in development is still common. Thus, although apoptosis resulting from a developmental toxicant may be the final outcome for a cell or tissue, apoptosis is not the initiating event of toxicity.

Environmental toxicants pose a risk for developmental toxicity caused by apoptosis. TCDD (2,3,7,8-tetrachlorodibenzo-para-dioxin) is a member of the dioxin family and enters the environment following use in herbicide manufacturing and the combustion of organic materials. In mammals, TCDD causes fetal mortality, growth retardation, edema, lymphoid organ hypoplasia, and cleft palate at concentrations below the threshold for maternal toxicity (Birnbaum, 1995). In fish, it causes yolk sac edema, craniofacial abnormalities, hemorrhage, and mortality (King-Heiden et al., 2012). TCDD

70 is an ideal ligand for the aryl hydrocarbon receptor (AhR), a transcription factor regulating xenobiotic metabolism, cell-cell interactions, cell cycle control and the endocrine system (King-Heiden et al., 2012). It is believed that TCDD exerts its effects through the AhR, but oxidative stress has also been suggested as an additional mechanism of toxicity.

Several studies have implicated apoptosis in TCDD developmental toxicity. In

Japanese medaka (Oryzias latipes), apoptosis was detected in the brain and heart of late stage embryos treated with 4pg of TCDD at fertilization (Cantrell et al., 1996). These effects were eliminated following treatment with a CYP450 inhibitor and decreased with antioxidants (Cantrell et al., 1996). Treatment of Fundulus heteroclitus embryos with

TCDD generated apoptosis in the brain, eye, gill, kidney, tail, heat, intestine and blood vessels (Toomey et al., 2001). Apoptosis was detected in the dorsal midbrain of zebrafish embryos following treatment with 0.3ppb TCDD, and was inhibited by co-exposure with antioxidants and CYP450 inhibitors (Dong et al., 2002).

Exact mechanisms of TCDD-induced apoptosis are unclear. Several studies suggest a role for the extrinsic pathway in TCDD induced immunotoxicity. Apoptosis in thymocytes of TCDD treated mice was inhibited in Fas and Fas-L knockouts (Rhile et al.,

1996; Kamath et al., 1999). Furthermore, increased levels of Fas, TRAIL (TNF-related apoptosis-inducing ligand), and DR5 (death receptor 5) mRNA were observed in postnatal mice treated with TCDD during gestation, while no changes in Bax or Bcl-2 expression was observed (Camacho et al., 2004). Prenatal exposure to TCDD also caused

71 a significant decrease in microRNAs with sequences specific for Fas, Fas-L and CYP1A1 and AhR in thymocytes (Singh et al., 2012).

Integration of Oxidative Stress, the UPR and Apoptosis

Positive Feedback in Oxidative Stress/UPR signaling

As ER fidelity and redox homeostasis are essential for nearly all cellular processes, it stands to reason that a complex interplay between the UPR and oxidative stress would exist (Figure 1.12). Oxidative stress can interfere with protein folding via disruption of disulfide bonds or through inhibition of Ca2+ ATPase, resulting in inactive enzymes or important signaling molecules. If oxidative stress in the vicinity of the ER results in the accumulation of mis-folded proteins, the UPR can be activated to combat the response. Furthermore, the UPR itself has been implicated in the generation of ROS

(Harding et al., 2003), resulting in a positive feedback loop between the UPR and oxidative stress. In contrast to the reducing environment of the cytosol, the ER lumen is oxidizing, which promotes formation of disulfide bonds (Van der Vlies et al., 2003).

Protein disulfide isomerase (PDI) is a member of the Prx superfamily and catalyzes disulfide bond formation, isomerization and reduction of proteins within the ER (Ferrari

& Soling, 1999). Following oxidation of protein thiols, the reduced form of PDI is oxidized by ER oxidoreductins (ERO), ERO1α and ERO1β (Pagani et al., 2000). ERO1 uses a flavin-dependent mechanism to transfer electrons to molecular oxygen, which may

72 generate ROS (as superoxide and H2O2) as a result of electron uncoupling (Tu &

Weissman, 2004). Figure'7.'Interplay'between'the'UPR'and'redox'poten2al' ER'Lumen'

NF9kB' ER'Stress''

P' eIF2α'

ROS' ROS'

ERO1' Ca2+' Nrf2' P'

2+' Ca PDI9Ox' PDI9Red'

BiP' ROS'

GSH' GSSG' P' Nrf2' ATF4'

CHOP' ATF4' Nrf2' P'

Nucleus' Figure 1.12. Interplay between the unfolded protein response and redox potential. Abbreviations: ATF4, activating transcription factor 4; BiP, immunoglobulin heavy- chain- binding protein; CHOP, C/EBP homologous protein; eIF2a, eukaryotic initiation factor 2a; ERO1, ER oxidoreductin 1; NF-kB, nuclear factor kappa-light-chain-enhancer of activated B cells; Nrf2, nuclear factor (erythroid-derived 2)-like two; PDI, protein disulfide isomerase; PERK, protein kinase RNA-like ER kinase.

In addition to this causative association between oxidative damage and the UPR,

other more complex interactions between redox and the UPR function in networks of

gene expression and signal transduction throughout embryonic development. Many UPR

transcription factors have gene targets that are involved in the response to oxidative stress

or are redox regulated. The association between oxidative stress and the UPR has recently

been established in several human diseases (Van der Vlies et al., 2003; Cao & Kaufman,

73 2014). However, their relationship with physiological embryonic development and developmental toxicity have yet to be thoroughly explored.

One connection between the UPR and oxidative stress lies within the PERK pathway. Although eIF2α is the primary target of PERK, PERK has the capacity to phosphorylate other substrates, one of which is Nrf2 (Cullinan et al., 2003). PERK phosphorylation promotes release of Nrf2 from Keap1 and its translocation into the nucleus, independently of eIF2α (Cullinan et al., 2003). Furthermore, ER stressed cell survival was reduced in Nrf2 knockouts (Cullinan et al., 2003). Nrf2-/- mice challenged with ER stress by tunicamycin had changes in gene expression of genes involved in apoptosis, cell cycle, glucose biosynthesis, calcium homeostasis, ER/golgi transport and biosynthesis, and drug metabolism and transport (Nair et al., 2007). Another component of the PERK pathway, ATF4 is a Nrf2 interacting protein, specifically in the regulation of heme-oxygenase- 1 (HO-1). This is consistent with data indicating that fibroblasts increase ATF4 DNA-binding in response to anoxia (Estes et al., 1995) and the fact that

ATF4-/- cells are more susceptible to oxidative stress (Harding et al., 2003).

Another important redox-regulated transcription factor, NF-κΒ is also regulated by mediators of the UPR (Pahl & Baeuerle, 1995). However, this signal transduction pathway is thought to be separate from the UPR and has been designated the ER overload response (EOR) (Jiang et al., 2003). In PERK signaling, phosphorylation of eIF2α is required for activation of NF-κΒ under ER stress and stress caused by amino acid starvation (Jiang et al., 2003) possibly via translational inhibition of NF-κΒ inhibitor

74 IκΒα (Deng et al., 2004). The IRE1 and ATF6 branches of the response are also thought to be involved in NF-κΒ signaling (Hu et al., 2006; Yamazaki et al., 2009).

Oxidative Stress Induced Apoptosis

The delicate balance between endogenous ROS and ROS defense systems can be tipped by exogenous toxicants to generate the cellular damage discussed above.

Oxidative damage can initiate apoptosis through a complex web of pathways and mechanisms. Here we will give a brief overview of apoptosis resulting from ROS induced mitochondrial damage, disruptions in calcium homeostasis, and DNA damage

(Figure 1.13).

75 Figure'8.'Oxida2ve'Stress'Induced'Apoptosis'

NADPH' ROS'

HIPK2'

ATP' P' P' P53' P53' MPT'

Ca2+' Ca2+' Ca2+' Ca2+' BAX,'PUMA,' Ca2+' Apaf91,'DR5,' Ca2+' ONOO9' TRAIL' H2O2' CytC' XO'

CytC' Ca2+' Ca2+' Ca2+' ROS' Ca2+' Apoptosis'

Figure 1.13. Oxidative stress-induced apoptosis. Abbreviations: Apaf-1, apoptotic protease activating factor 1; BAX, BCL2-associated X; CytC, cytochrome C; DR5, death receptor 5; HIPK2, homeodomain interacting protein kinase 2; MPT, mitochondrial permeability transition; PUMA, p53-upregulated modulator of apoptosis; TRAIL, TNF- related apoptosis-inducing ligand; XO, xanthine oxidase.

Mitochondrial damage can result in mitochondrial permeability transition (MPT;

pores through the inner and outer mitochondrial membrane), which causes release of

cytochrome C and activation of the intrinsic pathway. MPT is induced by ROS (Kim et

al., 2003) and oxidation of mitochondrial proteins exposes hydrophilic moieties which

leads to protein aggregation and pore formation in the membrane (Kowaltowski et al.,

2001). MPT can also be induced by NADPH oxidation, which can occur as ROS deplete

NADPH through the GSH pathway (Kowaltowski et al., 2001). Furthermore, lipid

peroxidation can destroy the mitochondrial membrane, causing a release of intracellular

76 Ca2+ loss of membrane potential and attenuation of ATP production (Kowaltowski &

Vercesi, 1999), leading to MPT.

ROS can also induce apoptosis via disruption of calcium homeostasis. Calcium is sequestered in the extracellular space, ER and mitochondria in order to keep cytoplasmic

2+ Ca levels low and maintain membrane potential. Peroxynitrite and H2O2 can inactivate

Ca2+ ATPase causing accumulation of cytoplasmic Ca2+ (Viner et al., 1996; Zaidi et al.,

2003). Furthermore, ROS can induce ATP depletion by consumption of NADPH and inactivation of the electron transport chain depriving Ca2+ ATPase of substrate, leading to inactivation. Elevation of intracellular Ca2+ can harm the cell in several ways. Ca2+ uptake by the mitochondria can cause ATP depletion via disruption of membrane potential eventually leading to MPT (Kim et al., 2003). Elevated Ca2+ can also induce proteolytic cleavage of hydrolases that activate XO, leading to increased ROS, providing a positive feedback loop between hypercalcemia and ROS production (Harrison, 2002).

Alternatively, ROS can cause oxidative damage to DNA. When this damage is irreparable the apoptotic response is mediated by the transcription factor p53. P53 phosphorylation by HIPK2 results in its activation and transcription of its pro-apoptotic target genes, Bax, Apaf-1, PUMA (p53-upregulated modulator of apoptosis), and p53AIP1 (Surova & Zhivotovsky, 2013). In addition to these intrinsic proteins, p53 can also transcribe various factors of the extrinsic pathway, including TRAIL and DR5. P53 can also function outside its role as a nuclear transcription factor and has been shown to cause caspase activation, BCL-2 inhibition and MPT in the cytoplasm (Surova &

Zhivotovsky, 2013).

77 ER Stress Induced Apoptosis

Apoptosis induced by the UPR has been shown to proceed via both the intrinsic pathway and extrinsic pathway (Timmins et al., 2009; Hu et al., 2006). Exact mechanisms resulting in apoptosis from the UPR are unknown; however, several UPR mediators have shown to regulate cell death (Figure 1.14). IRE1α has both endoribonuclease and kinase functions, and several mechanisms have been proposed for

IRE1α modulation of apoptosis. One study found that RIDD was a significant contributor to apoptosis (Han et al., 2009). Furthermore, IRE1α was found to cleave select microRNAs that repress translation of proapoptotic caspase-2, an initiator caspase (Upton et al., 2012). IRE1α cleavage of 5 microRNAs resulted in significant increases in caspase

2 protein levels to regulate apoptosis (Upton et al., 2012).

78 Figure'9.'Apoptosis'induced'by'the'UPR'

ER'Stress'' PERK' PERK' IRE1' IRE1' IRE1' IRE1' ER'Lumen' IRE1' IRE1'

P' P' TRAF2' Cytoplasm' P' P' P' P' P' BAX' ASK1' P' Apoptosis' eIF2α'

NF9kB'

Ca2+' JNK' RIDD' ???' ERO1' Ca2+' Ca2+'

ATF4' ROS' Casp2/12'

CHOP' Bcl92' Apoptosis'

Nucleus' Figure 1.14. Oxidative stress induced by the unfolded protein response. Abbreviations: ASK1, apoptosis signal-regulating kinase 1; ATF4, activating transcription factor 4; BAX, BCL2-associated X; Bcl-2, B-cell lymphoma 2; Casp, caspase; CHOP, C/EBP homologous protein; eIF2a, eukaryotic initiation factor 2a; ERO1, ER oxidoreductin 1; IRE1, inositol-required enzyme 1; JNK, c-Jun N-terminal kinase; NF-kB, nuclear factor kappa-light-chain-enhancer of activated B cells; PERK, protein kinase RNA-like ER kinase; RIDD, regulated IRE1-dependent decay; TRAF2, TNF receptor-associated factor 2; XBP-1, X-box protein 1.

Conversely, several important apoptosis related proteins have been shown to be

IRE1α interactors (Chen & Brandizzi, 2013), indicating an integral and complex role for

IRE1 in UPR mediated cell death. Specifically, IREα was shown to be able to bind proapoptotic factors, BAX and BAK (Hetz et al., 2006). Mice lacking BAX-BAK that are challenged with Tm (tunicamycin) have fewer apoptotic cells and display normal PERK and BiP signaling, but have defects in the IRE1 pathway, indicating an important role for

IRE1 and BAX-BAK in UPR induced cell death (Hetz et al., 2006). IRE1α was also

79 shown to interact with BAX-BAK and binding increased under ER stress and may modulate IRE1α activation through their proapoptotic BH3 and BH1 domains (Hetz et al., 2006). Furthermore, expression of several BH3-only proteins has been shown to increase under ER stress (Sovolyova et al., 2014).

Another potential mechanism of IRE1α mediated apoptosis occurs through the c-

Jun NH2-terminal kinase (JNK) pathway, a well-known initiator of apoptosis. ER stress has been shown to induce JNK in an IRE1 dependent manner through IRE1 interaction with TRAF2 (tumor necrosis factor receptor (TNFR)- associated factor-2), which is mediated through its kinase domain (Urano et al., 2000). Another study demonstrated that

ASK-1 is also another essential component of this complex for the induction of apoptosis; primary neurons lacking ASK-1 were resistant to ER stress and apoptosis

(Nishitoh et al., 2002).

The PERK pathway has also been shown to be involved in apoptosis, through the

ATF4 target gene, CHOP (C/EBP homologous protein), a leucine zipper transcription factor activated during cellular stress. As with IRE1, multiple associations between

CHOP and ER stress-induced apoptosis have been drawn. In vivo, CHOP deficient mice had decreased cardiomyocyte apoptosis following transverse aortic constriction in comparison to controls (Fu et al., 2010). Much of the CHOP regulation of apoptosis has been associated with oxidative stress. Elevated CHOP expression has been linked to

BCL-2 down regulation, which is accompanied by a depletion of GSH and increased

ROS production in mouse embryonic fibroblasts (McCullough et al., 2001).

80 Furthermore, ERO1α is a transcriptional target of CHOP (Marciniak et al., 2004).

ERO1α has been implicated in Ca2+ regulation (Anelli et al., 2012) and has been shown to be involved in the release of ER Ca2+ stores during ER stress. Increases in cytoplasmic calcium have been linked to Fas death receptor induction and the extrinsic pathway of apoptosis through calcium/calmodulin-dependent protein kinase IIγ (CaMKIIγ) (Timmins et al., 2009). CaMKIIγ was also linked to intrinsic apoptosis through accumulation of mitochondrial Ca2+ leading to cytochrome c release during ER stress (Timmins et al.,

2009). This was further connected to an increase in ROS via activation of NADPH oxidase (NOX), which was able to induce CHOP through a positive feedback loop (Li et al., 2010). Thus, CHOP induced oxidative stress and Ca2+ signaling may play a role in

ER-stress induced apoptosis.

Recent work has suggested an integral role for caspase 8 and death receptor 5

(DR5) in ER stress induced apoptosis (Lu et al., 2014). ER stress was shown to upregulate DR5 transcripts via CHOP and depletion of caspase 8 and DR5 inhibited ER stress induced apoptosis (Lu et al., 2014). The role of DR5 was shown to be independent of ligand binding (Lu et al., 2014).

Conclusions

The complex spatiotemporal orchestration of development requires great precision, leaving the developing fetus vulnerable to slight perturbations that can result in teratogenesis or lethality. To date, the mechanisms of many teratogens and toxicants remain unknown. Understanding these mechanisms impacts both development of safer

81 pharmaceuticals and of therapeutics to combat teratogens, such as those designed to prevent developmental toxicity resulting from hypoxia. While the mode of action of developmental toxicants was once thought to require specificity, it is becoming increasingly clear that many toxicants may impact several key cellular processes, leading to varied outcomes depending on dose, duration and developmental stage.

Processes such as the unfolded protein response and redox regulated signaling play complex and diverse roles in physiological development. Xenobiotic-induced perturbations in these events can generate oxidative stress or ER stress, which can lead to damage and apoptosis through a variety of mechanisms. As demonstrated, these events can exert their effects both independently and in concert, resulting in the potentiation of cellular damage (Figure 1.15). Although oxidative stress and apoptosis have long been regarded as mechanisms of developmental toxicity, the role of the UPR is only just coming under scrutiny. Much of our current understanding of the UPR in developmental toxicity comes from its newly recognized role in disease, such as in neurodegenerative diseases. As research elucidating its role in developmental toxicity becomes available the

UPR, like oxidative stress, will become a target for new therapeutics.

82 Figure'13.' Toxicant' Interplay' between' the'UPR,' Oxida2ve' Unfolded' ROS' Stress'and' proteins' Apoptosis' in' Redox' Developme Imbalance' ntal'toxicity'

Oxida2ve' Protein' ER'stress' stress'' Damage' ''

DNA/Lipid' Disrupted' Protein/RNA' Damage' Ca2+'signaling'' degrada2on''

Apoptosis'

Figure 1.15. Interplay between the unfolded protein response, oxidative stress, and apoptosis in developmental toxicity.

Proposed Mechanism of Selenium and Hypersaline toxicity

The project described herein was designed to further understand mechanisms of

embryo toxicity of selenomethionine and hypersalinity. Se contamination remains a

concern throughout the US and characterization of multiple stressor interactions and their

mechanisms is crucial for proper Se regulation. Although previous research has indicated

that hypersalinity may increase SeMet toxicity to developing embryos through activation

by FMOs and induction of oxidative stress (Lavado et al., 2012), the stages of embryonic

development sensitive to this interaction and the precise mechanism are unknown. Based

83 on the body of knowledge on mechanisms of oxidative stress, the UPR and apoptosis in teratogenesis, as well as the role of FMO activity SeMet/hypersaline toxicity the following hypotheses were tested:

i. Increasing salinity of artificial seawater, desalination brine and artificial

SJRV saltwater will increase deformities and decrease survival in

Japanese medaka embryos and larvae.

ii. Saltwater of differing ionic contents will induce variable embryo and

larval toxicity to Japanese medaka.

iii. FMOs are expressed throughout Japanese medaka development and are

regulated by the UPR during liver functionalization.

iv. Hypersalinity will increase deformities and mortality of SeMet to Japanese

medaka embryos in a developmental stage-specific and dose-specific

manner.

v. High concentrations of SeMet will induce oxidative stress, the UPR and

apoptosis in Japanese medaka embryos, and hypersalinity will increase

these effects.

vi. SeMet treatments during chondrogenesis and early organogenesis will

induce malformations of the tails in Japanese medaka embryos through

oxidative stress, the UPR and apoptosis.

84 References

Abdelwahid, E., Pelliniemi, L.J., Jokinen, E., 2002. Cell death and differentiation in the development of the endocardial cushion of the embryonic heart. Microsc. Res. Tech. 58, 395-403.

Acosta-Alvear, D., Zhou, Y., Blais, A., Tsikitis, M., Lents, N.H., Arias, C., Lennon, C.J., Kluger, Y., Dynlacht, B.D., 2007. XBP1 controls diverse cell type- and condition- specific transcriptional regulatory networks. Mol. Cell 27, 53-66.

Ameri, K., Harris, A.L., 2008. Activating transcription factor 4. Int. J. Biochem. Cell Biol. 40, 14e21.

Anderson, E. J., Katunga, L. A., Willis, M. S., 2012. Mitochondria as a Source and Target of Lipid Peroxidation Products in Healthy and Diseased Heart. Clin. Exp. Pharmacol. Physiol. 39, 179-193

Andren, A.W., Klein, D.H., 1975. Selenium in coal-fed steam plant emissions. Environ. Sci. Technol. 9, 856–858.

Anelli, T., Bergamelli, L., Margittai, E., Rimessi, A., Fagioli, C., Malgaroli, A., Pinton, P., Ripamonti, M., Rizzuto, R., Sitia, R., 2012. Ero1a regulates Ca(2+) fluxes at the endoplasmic reticulum-mitochondria interface (MAM). Antioxid. Redox Signal. 16, 1077- 1087.

Arnold, W.V., Fertala, A., 2013. Skeletal diseases caused by mutations that affect collagen structure and function. Int. J. Biochem. Cell Biol. 45, 1556-1567.

Arnold, M., Forte, J.E., Osterberg, J.S., Di Giulio, R.T. 2016. Antioxidant Rescue of Selenomethionine-Induced Teratogenesis in Zebrafish Embryos. Arch. Environ. Contam. Toxicol. 70, 311-320.

Ashkenazi, A., 2008. Directing cancer cells to self-destruct with pro-apoptotic receptor agonists. Nat. Rev. Drug Discov. 7, 1001-1012.

Assmann, A., Briviba, K., Sies, H., 1998. Reduction of methionine selenoxide to selenomethionine by glutathione. Arch. Biochem. Biophys. 349, 201–203.

Baines, S.B., Fisher, N.S., 2001. Interspecific differences in the bioconcentration of selenite by phytoplankton and their ecological implications. Mar. Ecol. Prog. Ser. 213, 1- 12.

Baines, S.B., Fisher, N.S., Stewart, R., 2002. Assimilation and retention of selenium and other trace elements from crustacean food by juvenile striped bass (Morone saxitilis). Limnol. Oceanogr. 47, 646- 355.

85 Balaban, R. S., Nemoto, S. Finkel, T., 2005. Mitochondria, Oxidants, and Aging. Cell 120, 483-495.

Baron, R.L., 1994. A carbamate insecticide: a case study of aldicarb. Enviorn. Health Perspect. 102, 23-27

Bawardi, O., Rimoldi, J., Schlenk, D., 2007. Impacts of hypersaline water on the biotransformation and toxicity of fenthion rainbow trout (Oncorhynchus mykiss), striped bass (Morone saxatilis × Morone chrysops) and tilapia (Oreochromis mossambicus). Pest. Biochem. Physiol. 88, 321–327.

Bergendi, L., Benes, L., Durackova, Z., Ferencik, M., 1999. Chemistry, physiology, and pathology of free radicals. Life Sci. 65, 1865-1874.

Berlett, B.S., Stadtman, E.R., 1997. Protein Oxidation in Aging, Disease, and Oxidative Stress. J. Biol. Chem. 272, 20313-20316.

Bernard, A.R., Wells, T.N., Cleasby, A., Borlat, F., Payton, M.A., Proudfoot, A.E., 1995. Selenomethionine labelling of phosphomannose isomerase changes its kinetic properties. Eur. J. Biochem. 230, 111–118.

Bertolotti, A., Zhang, Y., Hendershot, L.M., Harding, H.P., Ron, D., 2000. Dynamic interaction of BiP and ER stress transducers in the unfolded-protein response. Nat. Cell. Biol. 2, 326-332.

Bertolotti, A., Wang, X., Novoa, I., Jungreis, R., Schlessinger, K., Cho, J.H., West, A.B., Ron, D., 2001. Increased sensitivity to dextran sodium sulfate colitis in IRE1b-deficient mice. J. Clin. Invest. 107, 585-593.

Birnbaum, L. S., 1995. Developmental Effects of Dioxins. Environ. Health Persp. 103, 89-94

Boatright, K.M., Renatus, M., Scott, F.L., Sperandio, S., Shin, H., Pedersen, I.M., Ricci, J.E., Edris, W.A., Sutherlin, D.P., Green, D.R., Salvesen, G.S., 2003. A unified model for apical caspase activation. Mol. Cell 11, 529-541.

Boeuf, G., Payan, P., 2001. How should salinity influence fish growth? Comp. Biochem. Physiol. C Pharmacol. Toxicol. 130, 411 -423.

Boles, J.O., Cisneros, R.J., Weir, M.S., Odom, J.D., Villafranca, J.E., Dunlap, R.B., 1991. Purification and characterization of selenomethionyl thymidylate synthase from Escherichia coli; comparison with the wild-type enzyme. Biochemistry 30: 11073–11080.

86 Bonfanti, P., Colombo, A., Orsi, F., Nizzetto, I., Andrioletti, M., Bacchetta, R., Mantecca, P., Fascio, U., Vailati, G., Vismara, C., 2004. Comparative teratogenicity of Chlorpyrifos and Malathion on Xenopus laevis development. Aquat. Toxicol. 70, 189- 200.

Borchert, A., Wang, C. C., Ufer, C., Schiebel, H., Savaskan, N. E., Kuhn, H., 2006. The role of phospholipid hydroperoxide glutathione peroxidase isoforms in murine embryogenesis. J. Biol. Chem. 281, 19655–19664.

Brash, A. R., 1999. Lipoxygenases: Occurrence, Functions, Catalysis, and Acquisition of Substrate. J. Biol. Chem. 274, 23679-23682.

Brautigam, L., Schutte, L. D., Godoy, J. R., Prozorovski, T., Gellert, M., Hauptmann, G., Holmgren, A., Lillig, C. H., Berndt, C., 2011. Vertebrate-specific glutaredoxin is essential for brain development. Proc. Natl. Acad. Sci. USA 108, 20532-20537.

Buss, R.R., Sun, W., Oppenheim, R.W., 2006. Adaptive roles of programmed cell death during nervous system development. Annu. Rev. Neurosci. 29, 1-35.

Calfon, M., Zeng, H., Urano, F., Till, J.H., Hubbard, S.R., Harding, H.P., Clark, S.G., Ron, D., 2002. IRE1 couples endoplasmic reticulum load to secretory capacity by processing the XBP-1 mRNA. Nature 415, 92-96.

Camacho, I.A., Nagarkatti, M., Nagarkatti, P.S., 2004. Evidence for induction of apoptosis in T Cells from murine fetal thymus following perinatal exposure to 2,3,7,8- tetrachlorodibenzo-p-dioxin (TCDD). Toxicol. Sci. 78, 96-106.

Cañedo-Argüelles, M., Kefford, B.J., Piscart, C., Prat, N., Schäfer, R.B., Schulz, C., 2013. Salinisation of rivers: An urgent ecological issue. Environ. Pol. 173, 157-167.

Cañedo-Argüelles, M., Hawkins, C.P., Kefford, B.J, Schäfer, R.B., Dyack, B.J., Brucet, S., Buchwalter D., Dunlop, J., Frör, O., Lazorchak, J., Coring, E., Fernandez, H.R., Goodfellow, W., González Achem, A.L., Hatfield-Dodds, S., Karimov, B.K., Mensah, P., Olson, J.R., Piscart, C., Prat, N., Ponsá, S., Schulz, C.-J., Timpano, A.J. 2016. Ion- specific standards are needed to protect biodiversity. Science 351, 914-916.

Cantrell, S.M., Lutz, L.H., Tillitt, D.E., Hannink, M., 1996. Embryotoxicity of 2,3,7,8- tetrachlorodibenzo-p-dioxin (TCDD): the embryonic vasculature is a physiological target for TCDD-induced DNA damage and apoptotic cell death in Medaka (Oryzias latipes). Toxicol. Appl. Pharmacol. 141, 23-34.

Cao, S.S., Kaufman, R.J., 2014. Endoplasmic reticulum stress and oxidative stress in cell fate decision and human disease. Antioxid. Redox Signal. 21, 396-413.

87 Cashman, J.R., 2005. Some distinctions between flavin-containing and cytochrome P450 monooxygenases. Biochem. Biophys. Res. Commun. 338, 599-604.

Cashman, J.R., Camp, K., Fakharzadeh, S.S., Fennessey, P.V., Hines, R.N., Mamer, S.C., Nguyen, G.P, Schlenk, D., Smith, R.L., Tjoa, S.S., Williams, D.E., Yannicelli, S., 2003. Biochemical and clinical aspects of the human flavin-containing monooxygenase form 3 (FMO3) related to trimethylaminuria. Curr. Drug Metabol. 4, 151–70.

Cashman, J.R., Zhang, J., 2006. Human Flavin-Containing Monooxygenases. Annu. Rev. Pharamol. Toxicol. 46, 65-100.

Celius, T., Roblin, S., Harper, P.A., Matthews, J., Boutros, P.C., Pohjanvirta, R., Okey, A.B., 2008. Aryl hydrocarbon receptor-dependent induction of flavin-containing monooxygenase mRNAs in mouse liver. Drug Metab. Dispos. 36, 2499–2505.

Celius, T., Pansoy, A., Matthews, J., Okey, A.B., Henderson, M.C., Krueger, S.K., Williams, D.E., 2010. Flavin-containing monooxygenase-3: induction by 3- methylcholanthrene and complex regulation by xenobiotic chemicals in hepatoma cells and mouse liver. Toxicol. Appl. Pharmacol. 247, 60–69.

Chakraborti, T.K., Farrar, J.D., Pope, C.N., 1993. Comparative neurochemical and neurobehavioral effects of repeated chlorpyrifos exposures in young and adult rats. Pharmacol. Biochem. Behav. 46, 219-224.

Chanda, S.M., Pope, C.N., 1996. Neurochemical and neurobehavioral effects of repeated gestational exposure to chlorpyrifos in maternal and developing rats. Pharmacol. Biochem. Behav. 53, 771-776.

Chen, A., Muzzio, I.A., Malleret, G., Bartsch, D., Verbitsky, M., Pavlidis, P., Yonan, A.L., Vronskaya, S., Grody, M.B., Cepeda, I., Gilliam, T.C., Kandel, E.R., 2003. Inducible enhancement of memory storage and synaptic plasticity in transgenic mice expressing an inhibitor of ATF4 (CREB-2) and C/EBP proteins. Neuron 39, 655-669.

Chen, Y., Brandizzi, F., 2013. IRE1: ER stress sensor and cell fate executor. Trends Cell. Biol. 23, 547-555.

Chen, G.-P., Ziegler, D.M., 1994. Liver microsome and flavin-containing monooxygenase catalyzed oxidation of organic selenium compounds. Arch. Biochem. Biophys. 312, 566–572.

Cheng, K.C., Cahill, D. S., Kasais, H., Nishimura, S., Loeb, L.A., 1992. 8- hydroxyguanine, an Abundant Form of Oxidative DNA Damage, Causes G-T and A-C Substitutions. J. Biol. Chem. 267, 166-172.

88 Christen, V., Capelle, M., Fent, K., 2013. Silver nanoparticles induce endoplasmatic reticulum stress response in zebrafish. Toxicol. Appl. Pharmacol. 272, 519-528.

Conley, J.M., Watson, A.T.D., Xie, L., Buchwalter, D.B., 2014. Dynamic selenium assimilation, distribution, efflux, and maternal transfer in Japanese medaka fed a diet of Se-enriched mayflies. Environ. Sci. Technol. 48, 2971–2978.

Cooke, M.S., Evans, M D., Dizdaroglu, M., Lunec, J., 2003. Oxidative DNA damage: mechanisms, mutation, and disease. FASEB J. 17, 1195-1214.

Cornejo, V.H., Pihan, P., Vidal, R.L., Hetz, C., 2013. Role of the unfolded protein response in organ physiology: lessons from mouse models. IUBMB Life 65, 962-975.

Crawford, E.D., Wells, J.A., 2011. Caspase substrates and cellular remodeling. Annu. Rev. Biochem. 80, 1055-1087.

Crumpton, T.L., Seidler, F.J., Slotkin, T.A., 2000. Is oxidative stress involved in the developmental neurotoxicity of chlorpyrifos? Dev. Brain Res. 121, 189-195.

Cullinan, S.B., Zhang, D., Hannink, M., Arvisais, E., Kaufman, R.J., Diehl, J.A., 2003. Nrf2 is a direct PERK substrate and effector of PERK-dependent cell survival. Mol. Cell. Biol. 23, 7198-7209.

Cutter, G.A. 1989. The estuarine behavior of selenium in San Francisco Bay. Estuar. Coast. Shelf Sci. 28, 13-34.

Cutter, G.A. 1992. Kinetic Controls on Metalloid Speciation in Seawater. Marine. Chem. 40, 65-80.

Cutter, G.A., San Diego-McGlone, M.L.C., 1990. Temporal variability of selenium fluxes in the San Francisco Bay. Sci. Total Environ. 97, 235-250.

David, N., Gluchowski, D.C., Leatherbarrow, J.E., D. Yee, McKees, L.J., 2015. Estimation of Contaminant Loads from the Sacramento-San Joaquin River Delta to San Francisco Bay. Water Environ. Res. 87, 334-346.

DeCoursey, T. E., Ligeti, E., 2005. Regulation and termination of NADPH oxidase activity. Cell. Mol. Life Sci. 62, 2173-2193.

DeLeve, L. D., Kaplowitz, N., 1991. Glutathione metabolism and its role in hepatotoxicity. Pharmac.Ther. 52, 287-307

Deng, J., Lu, P.D., Zhang, Y., Scheuner, D., Kaufman, R.J., Sonenberg, N., Harding, H.P., Ron, D., 2004. Translational repression mediates activation of nuclear factor kappa B by phosphorylated translation initiation factor 2. Mol. Cell. Biol. 24, 10161-10168.

89 Dinchuk, J. E., Car, B. D., Focht, R. J., Johnston, J. J., Jaffee, B. D., Covington, M. B., Contel, N. R., Eng, V. M., Collins, R. J., Czerniak, P. M., et al., 1995. Renal abnormalities and an altered inflammatory response in mice lacking cyclooxygenase II. Nature 378, 406–409.

Doblin, M.A., Baines S.B., Cutter L.S., Cutter, G.A., 2006. Sources and biogeochemical cycling of particulate selenium in the San Francisco Bay estuary. Estuar. Coast. Mar. Sci. 67, 681-694.

Dolphin, C., Cullingford, T., Shepard, E., Smith, R., Phillips, I., 1996. Differential developmental and tissue-specific regulation of expression of the genes encoding three members of the flavin-containing monooxygenase family of man, FMO1, FMO3 and FMO4. Eur. J. Biochem. 235, 683-689.

Dolphin, C.T., Beckett, D.J., Janmohamed, A., Cullingford, T.E., Smith, R.L., Shephard, E.A., Phillips, I.R., 1998. The flavin-containing monooxygenase 2 gene (FMO2) of humans, but not of other primates, encodes a truncated, nonfunctional protein. J. Biol. Chem. 273, 30599–30607.

Dong, W., Teraoka, H., Yamazaki, K., Tsukiyama, S., Imani, S., Imagawa, T., Stegeman, J.J., Peterson, R.E., Hiraga, T., 2002. 2,3,7,8-Tetrachlorodibenzo-p-dioxin toxicity in the zebrafish embryo: local circulation failure in the dorsal midbrain is associated with increased apoptosis. Toxicol. Sci. 69, 191-201.

DuRose, J.B., Tam, A.B., Niwa, M., 2006. Intrinsic capacities of molecular sensors of the unfolded protein response to sense alternate forms of endoplasmic reticulum stress. Mol. Biol. Cell 17, 3095-3107.

Egerer-Sieber, C., Herl, V., Müller-Uri, F., Kreis, W., Muller, Y.A., 2006. Crystallization and preliminary crystallographic analysis of selenomethionine-labeled progesterone 5beta-reductase from Digitalis lanata Ehrh. Acta Crystallogr. F62, 186-8

El-Alfy, A., Larsen, B.K., Schlenk, D., 2002. Effect of cortisol and urea on flavin monooxygenase activity and expression in rainbow trout, Oncorhynchus mykiss, Mar. Environ. Res., 54, 275.

El-Alfy, A., Schlenk, D., 1998. Potential mechanisms of the enhancement of aldicarb toxicity to Japanese medaka, Oryzias latipes, at high salinity. Toxicol. Appl. Pharmacol., 152, 175.

El-Alfy, A., Schlenk, D., 2002. Effect of 17β-estradiol and testosterone on the toxicity of aldicarb to Japanese medaka, Oryzias latipes. Toxicol. Sci. 68, 381, 2002.

90 Elfarra, A.A., Krause, R.J., 2005. Potential roles of flavin-containing monooxygenases in sulfoxidation reactions of l-methionine, N-acetyl-l-methionine and peptides containing l- methionine. Biochimica et Biophysica Acta 1703, 183-189.

Eom, H., Ahn, J., Kim, Y., Choi, J., 2013. Hypoxia inducible factor-1 (HIF-1)–flavin containing monooxygenase-2 (FMO-2) signaling acts in silver nanoparticles and silver ion toxicity in the nematode, Caenorhabditis elegans. Toxicol. Appl. Pharm. 270, 106- 113.

Epp, N., Furstenberger, G., Muller, K., de Juanes, S., Leitges, M., Hausser, I., Thieme, F., Liebisch, G., Schmitz, G., Krieg, P., 2007. 12R-lipoxygenase deficiency disrupts epidermal barrier function. J. Cell Biol. 177, 173–182.

Estes, S. D., Stoler, D. L., Anderson, G. R., 1995. Normal fibroblasts induce the C/EBP beta and ATF-4 bZIP transcription factors in response to anoxia. Exp. Cell. Res. 220, 47- 54.

Fairweather-Tait, S.J., Collings, R., Hurst, R., 2010. Selenium bioavailability: Current knowledge and future research requirements. Am. J. Clin. Nutr. 91, 1484S–1491S.

Fan, T.W.M., Teh, S.J., Hinton D.E., Higashi, R.M., 2002. Selenium biotransformations into proteinaceous forms by foodweb organisms of selenium-laden drainage waters in California. Aquatic Toxicol. 57, 65-84.

Fan, R., Bonde, S., Gao, P., Sotomayor, B., Chen, C., Mouw, T., Zavazava, N., Tan, K., 2012. Dynamic HoxB4-regulatory network during embryonic stem cell differentiation to hematopoietic cells. Blood 119, e139-e147.

Favero, C.B., Henshaw, R.N., Grimsley-Myers, C.M., Shrestha, A., Beier, D.R., Dwyer, N.D., 2013. Mutation of the BiP/GRP78 gene causes axon outgrowth and fasciculation defects in the thalamocortical connections of the mammalian forebrain. J. Comp. Neurol. 521, 677-696.

Ferrari, D.M., Soling, H.D., 1999. The protein disulphide-isomerase family: unravelling a string of folds. Biochem. J. 339, 1-10.

Firtina, Z., Duncan, M.K., 2011. Unfolded Protein Response (UPR) is activated during normal lens development. Gene Expr. Patterns 11, 135-143.

Fischer, U., Janicke, R.U., Schulze-Osthoff, K., 2003. Many cuts to ruin: a comprehensive update of caspase substrates. Cell Death Differ. 10, 76-100.

Flaskos, J., Sachana, M., 2010. Developmental neurotoxicity of anticholinesterase pesticides. In: Satoh, T., Guptah, R.C. (Eds.), Anticholinesterase Pesticides: Metabolism, Neurotoxicity, and Epidemiology. John Wiley & Sons, Hoboken, pp. 203-223.

91 Fournier, E., Adam, C., Massabuau, J.-C., Garnier-Laplace, J., 2006. Selenium bioaccumulation in Chlamydomonas reinhardtii and subsequent transfer to Corbicula fluminea: role of selenium speciation and bivalve ventilation. Environ. Toxicol. Chem. 25, 2692-2699.

Fu, H.Y., Okada, K., Liao, Y., Tsukamoto, O., Isomura, T., Asai, M., Sawada, T., Okuda, K., Asano, Y., Sanada, S., Asanuma, H., Asakura, M., Takashima, S., Komuro, I., Kitakaze, M., Minamino, T., 2010. Ablation of C/EBP homologous protein attenuates endoplasmic reticulum-mediated apoptosis and cardiac dysfunction induced by pressure overload. Circulation 122, 361-369.

Fuchs, Y., Steller, H., 2011. Programmed cell death in animal development and disease. Cell 147, 742-758.

Fuentes-Prior, P., Salvesen, G.S., 2004. The protein structures that shape caspase activity, specificity, activation and inhibition. Biochem. J. 384, 201-232.

Funato, Y., Miki, H., 2007. Nucleoredoxin, a Novel Thioredoxin Family Member Involved in Cell Growth and Differentiation. Antioxid. Redox Sign. 9, 1035-1057.

Gadepalli, R., Rimoldi, J.M., Fronczek, F.R., Nillos, M., Gan, J., Deng, X., Rodriguez- Fuentes, G., Schlenk, D., 2007. Synthesis of fenthion sulfoxide and fenoxon sulfoxide enantiomers: effect of sulfur chirality on acetylcholinesterase activity. Chem. Res. Toxicol. 20, 257–262.

Gardiner, C. S., Reed, D. J., 1994. Status of glutathione during oxidant-induced oxidative stress in the preimplantation mouse embryo. Biol. Reprod. 51, 1307–1314.

Gibson, G., 1998. Active role of chondrocyte apoptosis in endochondral ossification. Microsc. Res. Tech. 43, 191-204.

Gillespie, R.B., Baumann, P.C., 1986. Effects of high tissue concentrations of selenium on reproduction by bluegills. Trans. Am. Fish. Soc. 115, 208-213.

Gillette, J. R., Brodie, B. B., La Du, B. N., 1957. The oxidation of drugs by liver microsomes: On the role of TPNH and oxygen. J. Pharmacol. Exp. Ther. 119, 532-540.

Grinkova, Y. V., Denisov, I. G., McLean, M. A., Sligar, S. G., 2013. Oxidase uncoupling in heme monooxygenases: Human cytochrome P450 CYP3A4. Nanodiscs. Biochem. Bioph. Res. Co. 430, 1223–1227.

Hamilton, S.J., 2004. Review of selenium toxicity in the aquatic food chain. Sci. Total Environ. 326, 1–31.

92 Han, D., Lerner, A.G., Walle, L.V., Upton, J.-P., Xu, W., Hagen, A., Backes, B.J., Oakes, S.A., Papa, F.R., 2009. IRE1a kinase activation modes control alternate endoribonuclease outputs to determine divergent cell fates. Cell 138, 562-575.

Han, X., Zhou, J., Zhang, P., Song, F., Jiang, R., Li, M., Xia, F., Guo, F.J., 2013. IRE1a dissociates with BiP and inhibits ER stress-mediated apoptosis in cartilage development. Cell. Signal. 25, 2136-2146.

Hanschmann, E., Godoy, J. R., Berndt, C., Hudemann, C., Lillig, C. H., 2013. Thioredoxins, Glutaredoxins, and Peroxiredoxins— Molecular Mechanisms and Health Significance: from Cofactors to Antioxidants to Redox Signaling. Antioxid. Redox Signal. 19, 1539-1605.

Hansen, J.M., 2006. Oxidative stress as a mechanism of teratogenesis. Birth Defects Res. C Embryo Today 78, 293-307.

Hao, D.C., Chen, S.L., Mu, J., Xiao, P.G. 2009. Molecular phylogeny, long-term evolution, and functional divergence of flavin-containing monooxygenases. Genetica, 137, 173-187.

Harding, H.P., Novoa, I., Zhang, Y., Zeng, H., Wek, R., Schapiro, M., Ron, D., 2000. Regulated translation initiation controls stress-induced gene expression in mammalian cells. Mol. Cell 6, 1099-1108.

Harding, H.P., Zhang, Y., Zeng, H., Novoa, I., Lu, P.D., Calfon, M., Sadri, N., Yun, C., Popko, B., Paules, R., Stojdl, D.F., Bell, J.C., Hettmann, T., Leiden, J.M., Ron, D., 2003. An integrated stress response regulates amino acid metabolism and resistance to oxidative stress. Mol. Cell 11, 619-633.

Harrison, R., 2002. Structure and function of xanthine oxidoreductase: where are we now? Free Radic. Biol. Med. 33, 774-797.

Hayes, J. D., Dinkova-Kostova, A. T., 2014. The Nrf2 regulatory network provides an interface between redox and intermediary metabolism. Trends Biochem. Sci. 39, 199-218.

Hernandez, D., Janmohamed, A., Chandan, P., Phillips, I. R., & Shephard, E. A., 2004. Organization and evolution of the flavin-containing monooxygenase genes of human and mouse: identification of novel gene and pseudogene clusters. Pharmacogenetics 14, 117– 130.

Hernandez, D., Janmohamed, A., Chandan, P., Omar, B., Phillips, I., Shepard, E. 2009. Deletion of the mouse Fmo1 gene results in enhanced pharmacological behavioural responses to imipramine. Pharmacogenet. Genom. 19, 289-299.

93 Hettmann, T., Barton, K., Leiden, J.M., 2000. Microphthalmia due to p53-mediated apoptosis of anterior lens epithelial cells in mice lacking the CREB-2 transcription factor. Dev. Biol. 222, 110-123.

Hetz, C., 2012. The unfolded protein response: controlling cell fate decisions under ER stress and beyond. Mol. Cell. Biol. 13, 89-102.

Hetz, C., Bernasconi, P., Fisher, J., Lee, A.H., Bassik, M.C., Antonsson, B., Brandt, G.S., Iwakoshi, N.N., Schinzel, A., Glimcher, L.H., Korsmeyer, S.J., 2006. Proapoptotic BAX and BAK modulate the unfolded protein response by a direct interaction with IRE1alpha. Science 312, 572-576.

Hines, R.N., 2006. Developmental and tissue-specific expression of human flavin- containing monooxygenases 1 and 3. Expert Opin. Drug Metab. Toxicol. 2, 41-49. Ho, Y., Xiong, Y., Ma, W., Spector, A., Ho, D. S., 2004. Mice Lacking Catalase Develop Normally but Show Differential Sensitivity to Oxidant Tissue Injury. J. Biol. Chem. 279, 32804-32812.

Hoffman, D.J., Heinz, G.H., Krynitsky, A.J., 1989. Hepatic glutathione metabolism and lipid peroxidation in response to excess dietary selenomethionine and selenite in mallard ducklings. J. Toxicol. Environ. Health. 27, 263-271.

Hoffman, D.J., Marn, C.M., Marois, K.C., Sproul, E., Dunne, M., Skorupa, J.P., 2002. Sublethal effects in avocet and stilt hatchlings from selenium-contaminated sites. Environ. Toxicol. Chem. 21, 561-566.

Hollien, J., Weissman, J.S., 2006. Decay of endoplasmic reticulum-localized mRNAs during the unfolded protein response. Science 313, 104-107.

Hollien, J., Lin, J.H., Li, H., Stevens, N., Walter, P., Weissman, J.S., 2009. Regulated Ire1- dependent decay of messenger RNAs in mammalian cells. J. Cell Biol. 186, 323- 331.

Holm, J., Palace, V.P., Wautier, K., Evans, R.E, Baron, C.L., Podemski, C., Siwik, P., Sterling, G., 2003. An assessment of the development and survival of rainbow trout (Oncorhynchus mykiss) and brook trout (Salvelinus fontinalis) exposed to elevated selenium in an area of active coal mining. Proceedings of the 26th Annual Larval Fish Conference 2003, Bergen, Norway.

Holm, J., Palace, V.P., Siwik, P., Sterling, G., Evans, R.E., Baron, C.L., Werner, J., Wautier, K., 2005. Development effects of bioaccumulated selenium in eggs and larvae of two salmonid species. Environ. Toxicol. Chem. 24, 2373−81.

Holmgren, A., 1968. Thioredoxin. 6. The Amino Acid Sequence of the Protein from Escherichia coli. B. European J. Biochem. 6, 475-484.

94 Hong, M., Luo, S., Baumeister, P., Huang, J., Gogia, R.K., Li, M., Lee, A.S., 2004. Under-glycosylation of ATF6 as a novel sensing mechanism for activation of the unfolded protein response. J. Biol. Chem. 279, 11354-11363.

Hu, P., Han, Z., Couvillon, A.D., Kaufman, R.J., Exton, J.H., 2006. Autocrine tumor necrosis factor alpha links endoplasmic reticulum stress to the membrane death receptor pathway through IRE1alpha-mediated NF-kappaB activation and down-regulation of TRAF2 expression. Mol. Cell. Biol. 26, 3071-3084.

Huber, R.E., Criddle, R.S., 1967. The isolation and properties of beta-galactosidase from Escherichia coli grown in sodium selenate. Biochim. Biophys. Acta 141, 587–599.

Huh, W.J., Esen, E., Geahlen, J.H., Bredemeyer, A.J., Lee, A., Shi, G., Konieczny, S.F., Glimcher, L.H., Mills, J.C., 2010. XBP1 controls maturation of gastric zymogenic cells by induction of MIST1 and expansion of the rough endoplasmic reticulum. Gastroenterology 139, 2038-2049.

Imai, H., Hirao, F., Sakamoto, T., Sekine, K., Mizukura, Y., Saito, M., Kitamoto, T., Hayasaka, M., Hanaoka, K., Nakagawa, Y., 2003. Early embryonic lethality caused by targeted disruption of the mouse PHGPx gene. Biochem. Bioph. Res. Co. 305, 278-286.

Imai, H., Nakagawa, Y., 2003. Biological significance of phospholipid hydroperoxide glutathione peroxidase (phgpx, gpx4) in mammalian cells. Free Radical Biol. Med. 34, 145-169.

Ishikawa, T., Okada, T., Ishikawa-Fujiwara, T., Todo, T., Kamei, Y., Shigenobu, S., Tanaka, M., Saito, T.L., Yoshimura, J., Morishita, S., Toyoda, A., Sakaki, Y., Taniguchi, Y., Takeda, S., Mori, K., 2013. ATF6alpha/beta-mediated adjustment of ER chaperone levels is essential for development of the notochord in medaka fish. Mol. Biol. Cell 24, 1387-1395.

Iwakoshi, N.N., Lee, A.-H., Vallabhajosyula, P., Otipoby, K.L., Rajewsky, K., Glimcher, L.H. 2003. Plasma cell differentiation and the unfolded protein response intersect at the transcription factor XBP-1. Nat. Immunol. 4, 321-329.

Iwawaki, T., Akai, R., Yamanaka, S., Kohno, K., 2009. Function of IRE1a in the placenta is essential for placental development and embryonic viability. Proc. Natl. Acad. Sci. USA 106, 16657-16662.

Jakupoglu, C., Przemeck, G. K., Schneider, M., Moreno, S. G., Mayr, N., Hatzopoulos, A. K., de Angelis, M. H., Wurst, W., Bornkamm, G. W., Brielmeier, M., Conrad, M., 2005. Cytoplasmic thioredoxin reductase is essential for embryogenesis but dispensable for cardiac development. Mol. Cell Biol. 25, 1980–1988

95 Janmohamed, A., Hernandez, D., Phillips, I.R., Shephard, E.A., 2004. Cell-, tissue-, sex- and developmental stage-specific expression of mouse flavin-containing monooxygenase (FMO’s). Biochem. Pharmacol. 68, 73–83

Janz, D.M., DeForest, D.K., Brooks, M.L., Chapman, P.M., Gilron, G., Hoff, D., Hopkin, W.A., McIntyre, D.O., Mebane, C.A., Palace, V.P., Skorupa, J.P., Wayland, M., 2010. Selenium toxicity to aquatic organisms. In Ecological Assessment of Selenium in the Aquatic Environment. (P.M Chapman, W.J. Adams, M.L. Brooks, C.G Delos, S.N. Luoma, W.A. Maher, H.M. Ohlendorf, T.S. Presser, D.P. Shaw, Eds), pp 141–231, CRC Press, New York.

Jenkins, S., Paduan, J., Roberts, P., Schlenk, D., Weiss, J., 2013. Management of Brine Discharges to Coastal Waters: Recommendations of a Science Advisory Panel. Southern California Coastal Water Research Project. Costa Mesa, CA.

Jiang, H.-Y., Wek, S.A., McGrath, B.C., Scheuner, D., Kaufman, R.J., Cavener, D.R., Wek, R.C., 2003. Phosphorylation of the alpha subunit of eukaryotic initiation factor 2 is required for activation of NF-kB in response to diverse cellular stresses. Mol. Cell. Biol. 23, 5651-5663.

Jones, D. P., 2006. Redefining Oxidative Stress. Antioxid. Redox Signal. 8, 1865-1879.

Jurado, J., Prieto-Alamo, M., Madrid-Risquez, J., Pueyo, C., 2003. Absolute Gene Expression Patterns of Thioredoxin and Glutaredoxin Redox Systems in Mouse. J. Biol. Chem. 278, 45546-45554.

Kamath, A.B., Camacho, I., Nagarkatti, P.S., Nagarkatti, M., 1999. Role of Fas-Fas ligand interactions in 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)-induced immunotoxicity: increased resistance of thymocytes from Fas-deficient (lpr) and fas ligand-defective (gld) mice to TCDD-induced toxicity. Toxicol. Appl. Pharmacol. 160, 141-155.

Karpinski, B. A., Morle, G. D., Huggenvik, J., Uhler, M. D., Lenden, J. M. 1992. Molecular cloning of human CREB-2: an ATF/CREB transcription factor that can negatively regulate transcription from the cAMP response element. Proc. Natl. Acad. Sci. USA 89, 4820–4824.

Kienle, C., Kohler, H., Gerhardt, A., 2009. Behavioural and developmental toxicity of chlorpyrifos and nickel chloride to zebrafish (Danio rerio) embryos and larvae. Ecotox. Environ. Safe. 72, 1740-1747.

Kim, K.S., Kim, Y.K., Lee, A.S., 1990. Expression of the glucose-regulated proteins (GRP94 and GRP78) in differentiated and undifferentiated mouse embryonic cells and the use of the GRP78 promoter as an expression system in embryonic cells. Differentiation 42, 153-159.

96 Kim, Y.C., Masutani, H., Yamaguchi, Y., Itoh, K., Yamamoto, M., Yodoi, J., 2001. Hemin-induced activation of the thioredoxin gene by Nrf2. A differential regulation of the antioxidant responsive element by a switch of its binding factors. J. Biol. Chem. 276, 18399–18406

Kim, J.S., He, L., Lemasters, J.J., 2003. Mitochondrial permeability transition: a common pathway to necrosis and apoptosis. Biochem. Biophys. Res. Commun. 304, 463-470.

King-Heiden, T.C., Mehta, V., Xiong, K.M., Lanham, K.A., Antkiewicz, D.S., Ganser, A., Heideman, W., Peterson, R.E., 2012. Reproductive and developmental toxicity of dioxin in fish. Mol. Cell. Endocrinol. 354, 121-138.

Kirkman, H. N., Gaetani, G. F., 2007. Mammalian catalase: a venerable enzyme with new mysteries. Trends Biochem. Sci. 32, 44-50.

Klick, D.E., Hines, R.N., 2007. Mechanisms regulating human FMO3 transcription. Drug Metab. Rev. 39, 419–442.

Knowles, N., Cayan, D.R., 2002. Potential effects of global warming on the Sacromento/San Joaquin watershed and the San Francisco Estuary. Geophys. Res. Lett. 29, 32-42.

Kondo, S., Murakami, T., Tatsumi, K., Ogata, M., Kanemoto, S., Otori, K., Iseki, K., Wanaka, A., Imaizumi, K., 2005. OASIS, a CREB/ATF-family member, modulates UPR signalling in astrocytes. Nat. Cell. Biol. 7, 186-194.

Kowaltowski, A.J., Castilho, R.F., Vercesi, A.E., 2001. Mitochondrial permeability transition and oxidative stress. FEBS Lett. 495, 12-15.

Kowaltowski, A.J., Vercesi, A.E., 1999. Mitochondrial damage induced by conditions of oxidative stress. Free Radic. Biol. Med. 26, 463-471.

Krause, R.J., Elfarra, A.A., 2009. Reduction of L-methionine selenoxide to seleno-L- methionine by endogenous thiols, ascorbic acid, or methimazole. Biochem. Pharmacol. 77, 134–140.

Krause, R.J., Glocke, S.C., Sicuri, A.R., Ripp, S.L., Elfarra, A.A., 2006. Oxidative metabolism of seleno-L-methionine to L-methionine selenoxide by flavin-containing monooxygenases. Chem. Res. Toxicol. 19, 1643-9.

Kuida, K., Zheng, T.S., Na, S., Kuan, C., Yang, D., Karasuyama, H., Su, M.S., Rakic, P., Flavell, R.A., 1996. Decreased apoptosis in the brain and premature lethality in CPP32- deficient mice. Nature 384, 368-372

97 Kuida, K., Haydar, T.F., Kuan, C.Y., Gu, Y., Taya, C., Karasuyama, H., Su, M.S., Rakic, P., Flavell, R.A., 1998. Reduced apoptosis and cytochrome c-mediated caspase activation in mice lacking caspase 9. Cell 94, 325-337.

Kumar, S., 2007. Caspase function in programmed cell death. Cell Death Differ. 14, 32- 43.

Kuppusamy, P., Zweier, J. L., 1989. Characterization of Free Radical Generation by Xanthine Oxidase. J. Biol. Chem. 264, 9880-9884.

Kurooka, H., Kato, K., Minoguchi, S., Takahashi, Y., Ikeda, J., Habu, S., Osawa, N., Buchberg, A. M., Moriwaki, K., Shisa, H., Honjo, T., 1997. Cloning and Characterization of the Nucleoredoxin Gene That Encodes a Novel Nuclear Protein Related to Thioredoxin. Genomics 39, 331-339.

Lavado, R., Rimoldi, J.R., Schlenk, D., 2009. Mechanisms of fenthion activation in rainbow trout (Oncorhynchus mykiss) acclimated to hypersaline environments. Toxicol. App. Pharmacol. 235, 143-152.

Lavado, R., Maryoung, L.A., Schlenk, D., 2011. Hypersalinity acclimation increases the toxicity of the insectide phorate in coho salmon (Oncorhynchus kisutch). Environ. Sci. Technol. 45, 4623-4629.

Lavado, R., Shi, D., Schlenk, D., 2012. Effects of salinity on the toxicity and biotransformation of L-selenomethionine in Japanese medaka (Oryzias latipes) embryos: mechanisms of oxidative stress. Aquat. Toxicol., 108, 18-22.

Lavado, R., Aparicia-Fabre, R., Schlenk, D., 2013. Effects of salinity acclimation on the pesticide-metabolizing enzyme flavin-containing monooxygenase (FMO) in rainbow trout (Oncorhynchus mykiss). Comp. Biochem. Physiol. Part C, 157, 9-15.

Lemly, A.D. 1985. Toxicology of selenium in a freshwater reservoir: Implications for environmental hazard evalutaion and safety. Ecotoxicol. Environ. Safety 10, 314-338.

Lemly, A.D., 1997. A teratogenic deformity index for evaluating impacts of selenium on fish populations. Ecotox. Environ. Safe. 37, 259-66.

Lemly, D.A., Skorupa, J.P. 2007. Technical issues affecting the implementation of U.S. Environmental Protection Agency’s proposed fish tissue-based aquatic criterion for selenium. Integr. Environ. Assess. Manag. 3, 552–558.

Leonard, S. S., Harris, G. K., Shi, X., 2004. Metal-Induced Oxidative Stress and Signal Transduction. Free Radical Bio. Med. 37, 1921-1942.

98 Li, G., Scull, C., Ozcan, L., Tabas, I., 2010. NADPH oxidase links endoplasmic reticulum stress, oxidative stress, and PKR activation to induce apoptosis. J. Cell Biol. 191, 1113-1125.

Liang, G., Lian, C., Huang, D., Gao, W., Liang, A., Peng, Y., Ye, W., Wu, Z., Su, S., Huang, D., 2014. Endoplasmic reticulum stress-unfolding protein response-apoptosis cascade causes chondrodysplasia in a col2a1 p.Gly1170Ser mutated mouse model. PLoS One 9, e86894.

Lin, J.H., Li, H., Yasamura, D., Cohen, H.R., Zhang, C., Panning, B., Shokat, K.M., LaVail, M.M., Walter, P., 2007. IRE1 signaling affects cell fate during the unfolded protein response. Science 318, 944-949.

Low, S.C., Grundner-Culemann, E., Harney, J., Berry, M., 2000. SECIS-SBP2 interactions dictate selenocysteine incorporation effciency and selenoprotein hierarchy. EMBO J. 19, 6882-6890.

Lu, S., 1999. Regulation of hepatic glutathione synthesis: current concepts and controversies. FASEB J. 13, 1169-1183.

Lu, M., Lawrence, D.A., Marsters, S., Acosta-Alvear, D., Kimmig, P., Mendez, A.S., Paton, A.W., Paton, J.C., Walter, P., Ashkenazi, A., 2014. Opposing unfolded-protein- response signals converge on death receptor 5 to control apoptosis. Science 345, 98-101.

Luo, Z., Hines, R.N., 2001. Regulation of flavin-containing monooxygenase 1 expression by ying yang 1 and hepatic nuclear factors 1 and 4. Mol. Pharmacol. 60, 1421–1430.

Luo, S., Mao, C., Lee, B., Lee, A.S., 2006. GRP78/BiP is required for cell proliferation and protecting the inner cell mass from apoptosis during early mouse embryonic development. Mol. Cell. Biol. 26, 5688-5697.

Luoma, S.N., Presser, T.S., 2009. Emerging opportunities in management of selenium contamination: Environ. Sci. Technol. 43, 8483-8487.

Luoma, S.N., Johns, C., Fisher, N.S., Steinberg, N.A., Oremland, R.S., Reinfelder, J.R. 1992. Determination of selenium bioavailability to a benthic bivalve from particulate and solute pathways. Environ. Sci. Technol. 26, 485-491.

Luttrell, G. 1959. Annotated Bibliography on the geology of Selenium. USGS. Bulletin 1019-M, US Department of the Interior, Washington DC

Ma, Q., Battelli, L., Hubbs, A. F., 2006. Multiorgan Autoimmune Inflammation, Enhanced Lymphoproliferation, and Impaired Homeostasis of Reactive Oxygen Species in Mice Lacking the Antioxidant-Activated Transcription Factor Nrf2. Am. J. Pathol. 168, 1960-1974.

99 Macias, D., Ganan, Y., Sampath, T.K., Piedra, M.E., Ros, M.A., Hurle, J.M., 1997. Role of BMP-2 and OP-1 (BMP-7) in programmed cell death and skeletogenesis during chick limb development. Development 124, 1109-1117.

Madduri, K., Badger, M., Li, Z.-S., Xu, X., Thornburgh, S., Evens, S., Dhadialla, T.S., 2009. Development of stable isotope and selenomethionine labeling methods for proteins expressed in Pseudomonas fluorescens. Protein Exp. Purif. 65, 57-65.

Maier, C.M., Chan, P.K., 2002. Role of Superoxide Dismutases in Oxidative Damage and Neurodegenerative Disorders. The Neuroscientist 8, 323-334.

Majmundar, A.J., Wong, W.J., Simon, M.C., 2010. Hypoxia-Inducible Factors and the Response to Hypoxic Stress. Mol. Cell 40, 294-309.

Marciniak, S.J., Yun, C.Y., Oyadomari, S., Novoa, I., Zhang, Y., Jungreis, R., Nagata, K., Harding, H.P., Ron, D., 2004. CHOP induces death by promoting protein synthesis and oxidation in the stressed endoplasmic reticulum. Genes Dev. 18, 3066-3077.

Maryoung, L., Lavado, R., Schlenk, D., 2014. Impacts of hypersaline acclimation on the acute toxicity of the organophosphate chlorpyrifos to salmonids. Aquat Toxicol. 152, 284- 290.

Maryoung, L.A., Blunt, B., Tierney, K.B., Schlenk, D., 2015. Sublethal toxicity of chlorpyrifos to salmonid olfaction after hypersaline acclimation. Aquat. Toxicol. 161, 94– 101

Masscheleyn, P.H., Deluane, R.D., Patrick, W.H.Jr., 1990. Transformations of Selenium as Affected by Sediment Oxidation-Reduction Potential and pH. Environ. Sci. Technol. 24, 91-96.

Masuoka, H.C., Townes, T.M., 2002. Targeted disruption of the activating transcription factor 4 gene results in severe fetal anemia in mice. Blood 99, 736e745.

Matsui, M., Oshima, M., Oshima, H., Takaku, K., Maruyama, T., Yodoi, J., Taketo, M. M.. 1996. Early embryonic lethality caused by targeted disruption of the mouse thioredoxin gene. Dev. Biol. 178, 179–185.

Maxwell, P. H., Pugh, C. W., Ratcliff, P.J., 2001. Activation of the HIF pathway in cancer. Curr. Opin. Genet. Dev. 11, 293-299.

McCord, J. M., Fridovich, I., 1968. The Reduction of Cytochrome c by Milk Xanthine Oxidase. J. Biol. Chem. 243, 5753-5760.

100 McCullough, K.D., Martindale, J.L., Klotz, L.O., Aw, T.Y., Holbrook, N.J., 2001. Gadd153 sensitizes cells to endoplasmic reticulum stress by down-regulating Bcl2 and perturbing the cellular redox state. Mol. Cell. Biol. 21, 1249-1259.

Mechaly, A., Teplitsky, A., Belakhov, V., Baasov, T., Shoham, G., Shoham, Y., 2000. Overproduction and characterization of seleno-methionine xylanase T-6. J. Biotech. 78, 83-86.

Miki, K., Mingxu, X., Gupta, A., Ba, Y., Tan, Y., Al-Refaie, W., Bouvet, M., Makuuchi, M., Moosa, A.R., Hoffman, R.M., 2001. Methioninase cancer gene therapy with selenomethionine as suicide prodrug substrate. Cancer Res. 61, 6805-6810.

Millennium Ecosystem Assessment, 2005. Ecosystems and Human Well-being: Synthesis. Island Press, Washington, DC.

Milne, J.B., 1998. The Uptake and Metabolism of Inorganic Selenium Species. In: W.T. Frankenberger, Jr. and R.A. Engberg (eds.), Environmental Chemistry of Selenium. Marcel Dekker, New York. pp. 459- 478.

Mimura, N., Hamada, H., Kashio, M., Jin, H., Toyama, Y., Kimura, K., Iida, M., Goto, S., Saisho, H., Toshimori, K., Koseki, H., Aoe, T., 2007. Quality control in the endoplasmic reticulum impairs the biosynthesis of pulmonary surfactant in mice expressing mutant BiP. Cell Death Differ. 14, 1475-1485.

Mimura, N., Yuasa, S., Soma, M., Jin, H., Kimura, K., Goto, S., Koseki, H., Aoe, T., 2008. Altered quality control in the endoplasmic reticulum causes cortical dysplasia in knock-in mice expressing a mutant BiP. Mol. Cell. Biol. 28, 293-301.

Misra, S., Niyogi, S., 2009. Selenite causes cytotoxicity in rainbow trout (Oncorhynchus mykiss) hepatocytes by inducing oxidative stress. Toxicol. in Vitro 23, 1249–1258.

Misra, S., Peak, D., Chen, N., Hamilton, C., Niyogi, S., 2012a. Tissue-specific accumulation and speciation of selenium in rainbow trout (Oncorhynchus mykiss) exposed to elevated dietary selenomethionine. Comp. Biochem. Physiol. C Toxicol. Pharmacol. 155, 560–565.

Misra, S., Hamilton, C., Niyogi, S., 2012b. Induction of oxidative stress by selenomethionine in isolated hepatocytes of rainbow trout (Oncorhynchus mykiss). Toxicol. In Vitro 26, 621-629.

Mori, K., 2009. Signaling pathways in the unfolded protein response: development from yeast to mammals. J. Biochem. 146, 743-750.

101 Mori, C., Nakamura, N., Kimura, S., Irie, H., Takigawa, T., Shiota, K., 1995. Programmed cell death in the interdigital tissue of the fetal mouse limb is apoptosis with DNA fragmentation. Anat. Rec. 242, 103-110.

Mount, D.R., Gulley, D.D., Hockett, J.R., Garrison, T.D., Evans, J.M., 1997. Statistical models to predict the toxicity of major ions to Ceriodaphnia dubia, Daphnia magna and Pimephales promelas (fathead minnows). Environ. Toxicol. Chem. 16, 2009-2019.

Murakami, T., Saito, A., Hino, S., Kondo, S., Kanemoto, S., Chihara, K., Sekiya, H., Tsumagari, K., Ochiai, K., Yoshinaga, K., Saitoh, M., Nishimura, R., Yoneda, T., Kou, I., Furuichi, T., Ikegawa, S., Ikawa, M., Okabe, M., Wanaka, A., Imaizumi, K., 2009. Signalling mediated by the endoplasmic reticulum stress transducer OASIS is involved in bone formation. Nat. Cell. Biol. 11, 1205-1211.

Muscatello, J.R., P.M. Bennett, K.T. Himbeault, A.M. Belknap, Janz, D.M., 2006. Larval deformities associated with selenium accumulation in northern pike (Esox lucius) exposed to metal mining effluent. Environ. Sci. Technol. 40, 6506-6512.

Myneni, S.C.B., T.K. Tokunaga, Brown, G.E. Jr. 1997. Abiotic selenium redox transformations in the presence of Fe(II, III) oxides. Science. 278, 1106–1109.

Nadanaka, S., Yoshida, H., Kano, F., Murata, M., Mori, K., 2004. Activation of mammalian unfolded protein response is compatible with the quality control system operating in the endoplasmic reticulum. Mol. Biol. Cell 15, 2537-2548.

Nadanaka, S., Okada, T., Yoshida, H., Mori, K., 2007. Role of disulfide bridges formed in the luminal domain of ATF6 in sensing endoplasmic reticulum stress. Mol. Cell. Biol. 27, 1027-1043.

Nair, S., Xu, C., Shen, G., Hebbar, V., Gopalakrishnan, A., Hu, R., Jain, M.R., Liew, C., Chan, J.Y., Kong, A.-N., 2007. Toxicogenomics of endoplasmic reticulum stress inducer tunicamycin in the small intestine and liver of Nrf2 knockout and C57BL/6J mice. Toxicol. Lett. 168, 21-39.

Nakanishi, K., Sudo, T., Morishima, N., 2005. Endoplasmic reticulum stress signaling transmitted by ATF6 mediates apoptosis during muscle development. J. Cell Biol. 169, 555-560.

Narasimhulu, S., 1971. Uncoupling of Oxygen Activation from Hydroxylation in the Steroid C-21 Hydroxylare of Bovine Adrenocortical Microsomes. Arch. Biochem. Biophys. 147, 384-390.

Nath, A. K., Enciso, J., Kuniyasu, M., Hao, X.-Y., Madri, J. A., Pinter, E., 2004. Nitric oxide modulates murine yolk sac vasculogenesis and rescues glucose induced vasculopathy. Development, 131, 2485-2496.

102 Nguyen T., Nioi, P., Pickett, C. B., 2009. The Nrf2-Antioxidant Response Element Signaling Pathway and Its Activation by Oxidative Stress. J. Biol. Chem. 284, 13291- 13295.

Niki, E., 2014. Biomarkers of lipid peroxidation in clinical material. Biochim. Biophys. Acta. 1840, 809-817.

Nishitoh, H., Matsuzawa, A., Tobiume, K., Saegusa, K., Takeda, K., Inoue, K., Hori, S., Kakizuka, A., Ichijo, H., 2002. ASK1 is essential for endoplasmic reticulum stress- induced neuronal cell death triggered by expanded polyglutamine repeats. Genes Dev. 16, 1345-1355.

Njie-Mbye, Y. F., Kulkarni-Chitnis, M., Opere, C.A., Barrett, A., Ohia, S. E., 2013. Lipid peroxidation: pathophysiological and pharmacological implications in the eye. Front. Physiol. 4, 1-10.

Nonn, L., Williams, R. R., Erickson, R. P., Powis, G., 2003. The absence of mitochondrial thioredoxin 2 causes massive apoptosis, exencephaly, and early embryonic lethality in homozygous mice. Mol. Cell. Biol. 23, 916–922.

Nordblom, G. D., Coon, M. J., 1977. Hydrogen Peroxide Formation and Stoichiometry of Hydroxylation Reactions Catalyzed by Highly Purified Liver Microsomal Cytochrome P- 450. Arch. Biochem. Biophys. 180, 343-347.

Ohlendorf, H.M. 1984. The biologic system. Pages 8-15/n U.S. Bureau of Reclamation and Ecological Analysts, Proceedings of a research meeting on toxicity problems at Kesterson Reservoir, CA. US Bureau of Reclamation, MidPacific Region, Sacramento, California.

Ohlendorf, H.M., D.J. Hoffman, M.K. Saiki, Aldrich, T.W., 1986. Embryonic mortality and abnormalities of aquatic birds: Apparent impacts of selenium from irrigation drainwater. Sci. Total Environ. 52, 49-63.

Ohlendorf, H.M., Covington, S.M., Byron, E.R., Arenal, C.A., 2011. Conducting site- specific assessments of selenium bioaccumulation in aquatic systems. Integr. Environ. Assess. Manag. 7, 314-24.

Okada, T., Haze, K., Nadanaka, S., Yoshida, H., Seidah, N.G., Hirano, Y., Sato, R., Negishi, M., Mori, K., 2003. A Serine Protease Inhibitor Prevents Endoplasmic Reticulum Stress-induced Cleavage but Not Transport of the Membrane-bound Transcription Factor ATF6. J. Biol. Chem. 278, 31024-31032.

103 Orr, P.L., Wiramanden, C.I., Paine, M.D., Franklin, W., Fraser, C., 2012. Food chain model based on field data to predict westslope cutthroat trout (Oncorhynchus clarkii lewisi) ovary selenium concentrations from water selenium concentrations in the Elk Valley, British Columbia. Environ. Toxicol. Chem. 31, 672-680.

Pacitti, D., Wang, T., Page, M.M., Martin, S.A.M. Sweetman, J., Feldmann, J., Secombes, C.J., 2013. Characterization of cytosolic glutathione peroxidase and phospholipid-hydroperoxide glutathione peroxidase genes in rainbow trout (Oncorhynchus mykiss) and their modulation by in vitro selenium exposure. Aquat. Toxicol. 130-131, 97-11.a

Pacitti, D., Wang, T., Martin, S.A.M. Sweetman, J., Secombes, C.J., 2014. Insights into the fish thioredoxin system: Expression profile of thioredoxin and thioredoxin reductase in rainbow trout (Oncorhynchus mykiss) during infection and in vitro stimulation. Dev. Comp. Immunol. 42, 261-277.

Pagani, M., Fabbri, M., Benedetti, C., Fassio, A., Pilati, S., Bulleid, N.J., Cabibbo, A., Sitia, R., 2000. Endoplasmic reticulum oxidoreductin 1-lbeta (ERO1-Lbeta), a human gene induced in the course of the unfolded protein response. J. Biol. Chem. 275, 23685- 23692.

Pahl, H.L., Baeuerle, P.A., 1995. A novel signal transduction pathway from the endoplasmic reticulum to the nucleus is mediated by transcription factor NF-kB. EMBO J. 14, 2580-2588.

Palace, V.P., Spallholz, J.E., Holm, J., Wautier, K., Evans, R.E., Baron, C.L., 2004. Metabolism of selenomethionine by rainbow trout (Oncorhynchus mykiss) embryos can generate oxidative stress. Ecotoxicol. Environ. Saf. 58,17-21.

Pampfer, S., 2000. Apoptosis in rodent peri-implantation embryos: differential susceptibility of inner cell mass and trophectoderm cell lineages-a review. Placenta 21, S3-S10.

Patterson, S.E., Dealy, C.N., 2014. Mechanisms and models of endoplasmic reticulum stress in chondrodysplasia. Dev. Dyn. 243, 875-893.

Peterson, D.H., Cayan, D.R., Dettinger, M.D., Noble, M.A., Riddle, L.G., Schemel, L.E., Smith, R.E., Uncles, R.J., Walters, R.A., 1996. San Francisco Bay salinity: observations, numerical simulations, and statistical models. In: San Francisco Bay: The ecosystem (Hollinbaugh, J.T., Ed), pp.9-34. AAAS Monograph.

Phibbs, J., Franz, E., Hauck, D., Gallego, M., Tse, J.J., Pickering, I.J., Liber, K., Janz, D.M., 2011. Evaluating the trophic transfer of selenium in aquatic ecosystems using caged fish, X-ray absorption spectroscopy and stable isotope analysis. Ecotox. Environ. Safe. 74, 1855-1863.

104 Porter, N.A., Caldwell, S. E., Mills, K. A., 1995. Mechanisms of Free Radical Oxidation of Unsaturated Lipids. Lipids 30, 277-290.

Presser, T.S., 1994. The Kesterson effect. Environ. Manage. 18, 437 –454.

Presser, T.S., Luoma, S.N., 2006. Forecasting selenium discharges to the San Francisco Bay-Delta estuary: ecological effects of a proposed San Luis drain extension. Professional Paper 1646, U.S. Department of Interior, U.S. Geological Survey, Reston, Virginia.

Presser, T.S., Luoma, S.N., 2010. Ecosystem-Scale Selenium Modeling in Support of Fish and Wildlife Criteria Development for the San Francisco Bay-Delta Estuary, U.S. Geological Survey Administrative Report, California.

Presser, T.S., Ohlendorf, H.M. 1987. Biogeochemical cycling of selenium in the San Joaquin Valley, California, USA. Environ. Manage. 11, 805-821.

Pryor, W., 1986. Oxy-Radicals And Related Species: Their Formation, Lifetimes, and Reactions. Ann. Rev. Physiol. 48, 657-667.

Radak, Z., Kaneko, T., Tahara, S., Nakamoto, H., Pucsok, J., Sasvari, M., Nyakas, C., Goto, S., 2001. Regular exercise improves cognitive function and decreases oxidative damage in rat brain. Neurochem. Int. 38, 17-23.

Ramirez-Solis, R., Zheng, H., Whiting, J., Krumlauf, R., Bradley, A., 1993. Hoxb-4 (Hox- 2.6) mutant mice show homeotic transformation of a cervical vertebra and defects in the closure of the sternal rudiments. Cell 73, 279-294.

Rathmell, J.C., Thompson, C.B., 2002. Pathways of apoptosis in lymphocyte development, homeostasis, and disease. Cell 109, S97-S107.

Ray, A., Liu, J., Ayoubi, P., Pope, C., 2010. Dose-related gene expression changes in forebrain following acute, low-level chlorpyrifos exposure in neonatal rats. Toxicol. Appl. Pharmacol. 248, 144-155.

Rayman, M.P., Infante, H.G., Sargent, M., 2008. Food-chain selenium and human health: spotlight on speciation. Br. J. Nutr. 100, 238–53.

Reimold, A.M., Etkin, A., Clauss, I., Perkins, A., Friend, D.S., Zhang, J., Horton, H.F., Scott, A., Orkin, S.H., Byrne, M.C., Grusby, M.J., Glimcher, L.H., 2000. An essential role in liver development for transcription factor XBP-1. Genes Dev. 14, 152-157.

Reimold, A.M., Iwakoshi, N.N., Manis, J., Vallabhajosyula, P., Szomolanyi-Tsuda, E., Gravallese, E.M., Friend, D., Grusby, M.J., Alt, F., Glimcher, L.H., 2001. Plasma cell differentiation requires the transcription factor XBP-1. Nature 412, 300-307.

105 Rhile, M.J., Nagarkatti, M., Nagarkatti, P.S., 1996. Role of Fas apoptosis and MHC genes in 2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD)-induced immunotoxicity of T cells. Toxicology 110, 153-167.

Riar, N., Crago, J., Jiang, W., Maryoung, L., Gan, J., Schlenk, D., 2013. Effects of Salinity Acclimation on the Endocrine Disruption and Acute Toxicity of Bifenthrin in Freshwater and Euryhaline Strains of Oncorhynchus mykiss. Environ. Toxicol. Chem. 32, 2779-2785.

Ripp, S.L., Overby, L.H., Philpot, R.M., Elfarra, A.A., 1997. Oxidation of cysteine S- conjugates by rabbit liver microsomes and cDNA-expressed flavin-containing monooxygenases: studies with S-(1,2-dichlorovinyl)-l-cysteine, S-(1,2,2-trichlorovinyl)- l-cysteine, S-allyl-l-cysteine, and S-benzyl-l-cysteine. Mol. Pharmacol. 52, 507-515.

Rodriguez-Fuentes, G., Aparicio-Fabre, R., Li, Q., Schlenk, D., 2008. Osmotic regulation of a novel flavin-containing monooxygenase in primary cultured cells from rainbow trout (Oncorhynchus mykiss). Drug Metab. Dispos. 36, 1212–1217.

Rudraiah, S., Gu, X., Hines, R., Manautou, J., 2016. Oxidative stress-responsive transcription factor NRF2 is not indispensable for the human hepatic Flavin-containing monooxygenase-3 (FMO3) gene expression in HepG2 cells. Toxicol. In Vitro, 31, 54-59.

Ryan, H. E., Lo, J., Johnson, R. S., 1998. HIF-1α is required for solid tumor formation and embryonic vascularization. EMBO J. 17, 3005-3015.

Saito, A., Ochiai, K., Kondo, S., Tsumagari, K., Murakami, T., Cavener, D.R., Imaizumi, K., 2011. Endoplasmic reticulum stress response mediated by the PERK-eIF2a-ATF4 pathway is involved in osteoblast differentiation induced by BMP2. J. Biol. Chem. 286, 4809-4818.

Sappington, K.G., 2002. Development of aquatic life criteria for selenium: a regulatory perspective on critical issues and research needs. Aquat. Toxicol. 57, 101-133.

Sappington, K.G., Bridges, T.S., Bradbury, S.P., Erickson, R.J., Hendriks, A.J., Lanno, R.P., Meador, J.P., Mount, D.R., Salazar, M.H., Spry, D.J., 2011. Application of the tissue residue approach in ecological risk assessment. Integr. Environ. Assess. Manag. 7, 116–140.

Schlenk, D., 1998. Occurrence of flavin-containing monooxygenases in non-mammalian eukaryotic organisms, Comp. Biochem. Physiol., 121C, 185.

Schlenk, D., Buhler, D.R., 1993. Immunological characterization of flavin-containing monooxygenases from the liver of rainbow trout (Oncorhynchus mykiss): sexual- and age-dependent differences and the effect of trimethylamine on enzyme regulation, Biochim. Biophys. Acta, 1156, 103.

106 Schlenk, D., El-Alfy, A., 1998. Expression of branchial flavin-containing monooxygenase is directly correlated with salinity-induced aldicarb toxicity in the euryhaline fish (Oryzias latipes), Mar. Environ. Res., 46, 103.

Schlenk, D., Lavado, R., 2011. Impacts of climate change on hypersaline conditions of estuaries and xenobiotic toxicity. Aquat. Toxicol. 105, 78-82.

Schlenk, D., Peters, L.D., Livingstone, D.R., 1996. Correlation of salinity with flavin- containing monooxygenase activity but not cytochrome P450 activity in the euryhaline fish (Platichthys flesus), Biochem. Pharmacol., 52, 815.

Schlenk, D., Zubcov, N., Zubcov, E., 2003. Effects of salinity on the uptake, biotransformation, and toxicity of dietary seleno-L-methionine to rainbow trout. Toxicol. Sci. 75, 309-313.

Schneider, M., Vogt Weisenhorn, D. M., Seiler, A., Bornkamm, G. W., Brielmeier, M., Conrad, M., 2006. Embryonic expression profile of phospholipid hydroperoxide glutathione peroxidase. Gene Expr. Patterns 6, 489-494.

Schrauzer, G.N. 2000. Selenomethionine: a review of its nutritional significance, metabolism and toxicity. J Nutr. 130:1653-1656.

Seiler, R.L., Skorupa, J.P., Peltz, L.A., 1999. Areas Susceptible to Irrigation-Induced Selenium Contamination of Water and Biota in the Western United States. U.S. Geol. Surv. Circ., 1180, 36 pp

Selvaraj, V., Tomblin, J., Armistead, M.Y., Murray, E. 2013. Selenium (sodium selenite) causes cytotoxicity and apoptotic mediated cell death in PLHC-1 fish cell line through DNA and mitochondrial membrane potential damage. Ecotoxicol. Environ. Safe. 87, 80- 88.

SFEI, 2015. The Pulse of the Bay: The State of Bay Water Quality, 2015 and 2065. SFEI Contribution #759. San Francisco Estuary Institute, Richmond, CA.

Sha, H., He, Y., Chen, H., Wang, C., Zenno, A., Shi, H., Yang, X., Zhang, X., Qi, L., 2009. The IRE1a-XBP1 pathway of the unfolded protein response is required for adipogenesis. Cell. Metab. 9, 556-564.

Shen, J., Chen, X., Hendershot, L., Prywes, R., 2002. ER stress regulation of ATF6 localization by dissociation of BiP/GRP78 binding and unmasking of Golgi localization signals. Dev. Cell 3, 99-111.

107 Shi, Z. Z., Osei–Frimpong, J., Kala, G., Kala, S. V., Barrios, R. J., Habib G. M., Lukin D., Danney, C. M., Matzuk, M. M., Lieberman, M. W., 2000. Glutathione synthesis is essential for mouse development but not for cell growth in culture. Proc. Natl. Acad. Sci. USA 97, 5101–5106.

Shih, D.M., Wang, Z., Lee, R., Meng, Y., Che, N., Charugundla, S., Qi, H., Wu, J., Pan, C., Brown, M., Vallin, T., Bennett, B., Graham, M., Hazen, S., Lusis, A. 2015. Flavin containing monooxygenase 3 exerts broad effects on glucose and lipid metabolism and atherosclerosis. J. Lipid Res. 56, 22–37.

Shrift, A. 1954. Sulfur-selenium antagonism. I. Antimetabolic action of selenate on the growth of Chlorella vulgaris. Am. J. Bot. 41, 223-230.

Sies, H., 1985. Oxidative Stress: introductory remarks. In Oxidative stress (H. Sies, Ed.), pp 1- 8, Academic Press, London.

Singh, N.P., Singh, U.P., Guan, H., Nagarkatti, P.S., Nagarkatti, M., 2012. Prenatal exposure to TCDD triggers significant modulation of microRNA expression profile in the thymus that affects consequent gene expression. PLoS One 7, e45054.

Slotkin, T.A., Oliver, C.A., Seidler, F.J., 2005. Critical periods for the role of oxidative stress in the developmental neurotoxicity of chlorpyrifos and terbutaline, alone or in combination. Dev. Brain Res. 157, 172-180.

Slotkin, T.A., MacKillop, E.A., Ryde, I.T., Seidler, F.J., 2007. Ameliorating the developmental neurotoxicity of chlorpyrifos: a mechanisms-based approach in PC12 cells. Environ. Health Perspect. 115, 1306-1313.

Soberman, R. J., Christmas, P., 2003. The organization and consequences of eicosanoid signaling. J. Clin. Invest. 111, 1107–1113.

Sovolyova, N., Healy, S., Samali, A., Logue, S.E., 2014. Stressed to death e mechanisms of ER stress-induced cell death. Biol. Chem. 395, 1-13.

Spallholz, J.E., Palace, V.P., Reid, T.W., 2004. Methioninase and selenomethionine but not Se- methylselenocysteine generate methyselenol and superoxide in an in vitro chemiluminescence assay: experiments and review. Biochem. Pharamacol. 67, 547-554.

Stahle, D.W., Therrel, M.D., Cleaveland, M.K., Cayan, D.R., Dettinget, M.D., and Knowles, N., 2001. Ancient blue oaks reveal human impact on San Francisco Bay salinity. Eos, Trans. Amer. Geophys. Union 82, 144-145.

Stekoll, M. S., Smoker, W.W., Failor-Rounds, B.J., Wang, I.A., Joyce, V.J., 2009. Response of the early developmental stages of hatchery reared salmonids to major ions in a simulated mine effluent. Aquaculture 298, 172-181.

108 Stewart R.A., Luoma, S.N., Schlekat, C.E, Doblin, M.A., Hieb, K.A., 2004. Food web pathway determines how selenium affects aquatic ecosystems: a San Francisco Bay case study. Environ. Sci. Technol. 38, 4519-4526.

Stewart, R., Grosell, M., Buchwalter, D., Fisher, N., Luoma, S., Mathews, T., Orr, P., Wang, W., 2010. Bioaccumulation and Trophic Transfer of Selenium. In Ecological Assessment of Selenium in the Aquatic Environment, Chapman, P. M., Adams, W. J., Brooks, M. L., Delos, C. G., Luoma, S. N., Maher, W. A., Ohlendorf, H. M., Presser, T. S., Shaw, D. P., Eds., SETAC Press: Boca Raton, FL, pp 93– 139.

Stolz, J.F., Oremland, R.S., 1999. Bacterial respiration of arsenic and selenium. FEMS Microbiol. Rev. 23, 615–27.

Suh, J.K, Poulsen, L.L., Ziegler, D.M., Robertus, J.D., 1996. Molecular Cloning and Kinetic Characterization of a Flavin-containing Monooxygenase from Saccharomyces cerevisiae. Arch. Biochem. Biophys. 336, 268-274.

Suh, J.K., Poulsen, L.L., Ziegler, D.M., Robertus, J.D., 1999. Yeast Flavin-containing Monooxygenase Generates Oxidizing Equivalents Controlling Protein Folding in the Endoplasmic Reticulum. Proc. Natl. Acad. Sci. USA 96, 2687-2691.

Suh, J.K., Poulsen, L.L., Ziegler, D.M., Robertus, J.D. 2000. Redox Regulation of Yeast Flavin-Containing Monooxygenase. Arch. Biochem. Biophys. 381, 317-322.

Suh, J.K., Robertus. J.D., 2000. Yeast FMO is induced by the unfolded protein response. Proc. Natl. Acad. Sci. USA. 97, 121-126.

Sultana, R., Perluigi, M., Butterfield, D. A., 2013. Lipid peroxidation triggers neurodegeneration: A redox proteomics view into the Alzheimer disease brain. Free Radical Bio. Med. 62, 157-169.

Surova, O., Zhivotovsky, B., 2013. Various modes of cell death induced by DNA damage. Oncogene 32, 3789-3797.

Tanaka, T., Tsujimura, T., Takeda, K., Sugihara, A., Maekawa, A., Terada, N., Yoshida, N., Akira, S., 1998. Targeted disruption of ATF4 discloses its essential role in the formation of eye lens fibres. Genes Cells 3, 801-810.

Tashjian, D.H., S.J. Teh, A. Sogomoyan, Hung, S.S.O., 2006. Bioaccumulation and chronic toxicity of dietary L-selenomethionine in juvenile white sturgeon (Acipenser transmontanus). Aquatic Toxicol. 79, 401-409.

Taylor, R.C., Cullen, S.P., Martin, S.J., 2008. Apoptosis: controlled demolition at the cellular level. Mol. Cell. Biol. 9, 231-241.

109 Tell, G., Quadrifoglio, F., Tiribelli, C., Kelley, M.R., 2009. The Many Functions of APE1/Ref-1: Not Only a DNA Repair Enzyme. Antioxid. Redox Signal. 11, 601-691.

Tijet, N., Boutros, P.C., Moffat, I.D., Okey, A.B., Tuomisto, J., Pohjanvirta, R., 2006. The aryl hydrocarbon receptor regulates distinct dioxin-dependent and dioxin- independent gene batteries. Mol. Pharmacol. 69, 140–153.

Timme-Laragy, A. R., Goldstone, J. V., Imhoff, B. R., Stegeman, J. J., Hahn, M. E., Hansen, J. M., 2013. Glutathione redox dynamics and expression of glutathione-related genes in the developing embryo. Free Radical Biol. Med. 65, 89-101.

Timmins, J.M., Ozcan, L., Seimon, T.A., Li, G., Malagelada, C., Backs, J., Backs, T., Bassel- Duby, R., Olson, E.N., Anderson, M.E., Tabas, I., 2009. Calcium/calmodulin- dependent protein kinase II links ER stress with Fas and mitochondrial apoptosis pathways. J. Clin. Invest. 119, 2925-2941.

Tirasophon, W., Welihinda, A.A., Kaufman, R.J., 1998. A stress response pathway from the endoplasmic reticulum to the nucleus requires a novel bifunctional protein kinase/ endoribonuclease (Ire1p) in mammalian cells. Genes Dev. 12, 1812-1824.

Toomey, B.H., Bello, S., Hahn, M.E., Cantrell, S., Wright, P., Tillitt, D.E., Di Giulio, R.T., 2001. 2,3,7,8-Tetrachlorodibenzo-p-dioxin induces apoptotic cell death and cytochrome P4501A expression in developing Fundulus heteroclitus embryos. Aquat. Toxicol. 53, 127-138.

Trollmann, R., Gassmann, M., 2009. The role of hypoxia-inducible transcription factors in the hypoxic neonatal brain. Brain Dev.-Jpn. 31, 503-509.

Trowbridge, P.R., Davis, J.A., Mumley, T., Taberski, K., Feger, N., Valiela, L., Ervin, J., Arsem, N., Olivieri, A., Carroll, P., Coleman, J., Salop, P., Sutton, R., Yee, D., McKee, L.J., Sedlak, M., Grosso, C., Kelly, J. 2016. The Regional Monitoring Program for Water Quality in San Francisco Bay, California, USA: Science in support of managing water quality. Regional Studies in Marine Science. 4, 21-33

Tsang, K.Y., Chan, D., Cheslett, D., Chan, W.C., So, C.L., Melhado, I.G., Chan, T.W., Kwan, K.M., Hunziker, E.B., Yamada, Y., Bateman, J.F., Cheung, K.M., Cheah, K.S., 2007. Surviving endoplasmic reticulum stress is coupled to altered chondrocyte differentiation and function. PLoS Biol. 5, e44.

Tu, B.P., Weissman, J.S., 2004. Oxidative protein folding in eukaryotes: mechanisms and consequences. J. Cell Biol. 164, 341-346.

Ufer, C., Wang, C. C., 2011. The roles of glutathione peroxidases during embryo development. Front. Mol. Neurosci. 4, 1-14.

110 Ufer, C., Wang, C. C., Borchert, A., Heydeck, D., Kuhn, H., 2010. Redox Control in Mammalian Embryo Development. Antioxid. Redox Sign. 13, 833-874.

Upton, J.-P., Wang, L., Han, D., Wang, E.S., Huskey, N.E., Lim, L., Tuitt, M., McManus, Ruggero, D., Goga, A., Papa, F.R., Oakes, S.A., 2012. IRE1a cleaves select micro- RNAs during ER stress to derepress translation of proapoptotic caspase-2. Science 338, 818-822.

Urano, F., Wang, X., Bertolotti, A., Zhang, Y., Chung, P., Harding, H.P., Ron, D., 2000. Coupling of stress in the ER to activation of JNK protein kinases by transmembrane protein kinase IRE1. Science 287, 664-666.

USEPA, 1980. Ambient water quality criteria for selenium. EPA-440/5-80-070. National Technical Information Service, Springfield, VA.

USEPA, 1987. Ambient water quality criteria for selenium. EPA-440/5-87-006. National Technical Information Service, Springfield, VA.

USEPA, 2004. Draft aquatic life water quality criteria for selenium. Office of Water and Office of Science and Technology, Washington (DC).

Van Der Vlies, D., Makkinje, M., Jansens, A., Braakman, I., Verklejj, A.J., Wirtz, K.W., Andries, J., 2003. Oxidation of ER resident proteins upon oxidative stress: effects of altering cellular redox/antioxidant status and implications for protein maturation. Anti- oxid. Redox Signal. 5, 381-387.

Varfolomeev, E.E., Schuchmann, M., Luria, V., Chiannilkulchai, N., Beckmannm, J.S., Mett, I.L., Rebrikov, D., Brodianski, V.M., Kemper, O.C., Kollet, O., Lapidot, T., Soffer, D., Sobe, T., Avraham, K.B., Goncharov, T., Holtmann, H., Lonai, P., Wallach, D., 1998. Targeted disruption of the mouse caspase 8 gene ablates cell death induction by the TNF receptors, Fas/Apo1, and DR3 and is lethal prenatally. Immunity 9, 267-276.

Veeramah, K., Thomas, M.G., Weale, M.E., Zeitlyn, D., Tarekegn, A., Bekele, E., Mendell, N.R., Shepard, E.A., Bradman, N., Phillips, I.R., 2008. The potentially deleterious functional variant flavin-containing monooxygenase 2*1 is at high frequency throughout sub-Saharan Africa. Pharmacogenet. Genom. 18, 877–86.

Veeravalli, S., Omar, B., Houseman, L., Hancock, M., Gonzalez Malagon, S., Scott, F., Janmohamed, A., Phillips, I., 2014. The phenotype of a flavin-containing monooyxgenase knockout mouse implicates the drug-metabolizing enzyme FMO1 as a novel regulator of energy balance. Biochem. Pharmacol. 90, 88-95.

Viner, R.I., Huhmer, A.F., Bigelow, D.J., Schoneich, C., 1996. The oxidative inactivation of sarcoplasmic reticulum Ca(2+)-ATPase by peroxynitrite. Free Radic. Res. 24, 243- 259.

111 Walter, P., Ron, D., 2011. The unfolded protein response: from stress pathway to homeostatic regulation. Science 334, 1081-1086.

Wang, J., Grisle, S., Schlenk, D., 2001. Effects of salinity on aldicarb toxicity to juvenile rainbow trout (Oncorhynchus mykiss) and striped bass (Morone saxatilis × chrysops), Toxicol. Sci., 64, 200-207.

Wang, W., Lian, N., Ma, Y., Li, L., Gallant, R.C., Elefteriou, F., Yang, X., 2012. Chondrocytic Atf4 regulates osteoblast differentiation and function via Ihh. Development 139, 601-611.

Wang, N., Dorman, R.A., Ingersoll, C. G., Hardesty, D. K., Brumbaugh, W.G., Hammer, E. J., Bauer, C. R., Mount, D. R. 2016. Acute and Chronic Toxicity of Sodium Sulfate to Four Freshwater Organisms in Water-Only Exposures. Environ. Toxicol. Chem. 35, 115- 127.

Warrier M, Shih DM, Burrows AC, Ferguson, D., Gromovsky, A.D., Brown, A.L., Marshall, S., McDaniel, A., Schugar, R.C., Wang, Z., Sacks, J., Rong, X., Vallim, T.A., Chou J., Ivanova, P.T., Myers, D.S., Brown, H.A., Lee, R.G., Crooke, R.M., Graham, M.J., Liu, X., Parini, P., Tontonoz, P., Lusis, A.J., Hazen, S.L., Temel, R.E., Brown, J.M. 2015. The TMAO-generating enzyme flavin monooxygenase 3 is a central regulator of cholesterol balance. Cell Rep. 10, 326–338.

Watanabe, T., Kiron, V., Satoh, S., 1997. Trace minerals in fish nutrition. Aquaculture 151, 185-207.

Wen, H., Carignan, J. 2007. Reviews on atmospheric selenium: Emissions, speciation and fate. Atmos. Environ. 41, 7151-7165.

Yamamoto, K., Sato, T., Matsui, T., Sato, M., Okada, T., Yoshida, H., Harada, A., Mori, K., 2007. Transcriptional induction of mammalian ER quality control proteins is mediated by single or combined action of ATF6a and XBP1. Dev. Cell 13, 365-376.

Yamamoto, K., Takahara, K., Oyadomari, S., Okada, T., Sato, T., Harada, A., Mori, K., 2010. Induction of liver steatosis and lipid droplet formation in ATF6a-knockout mice burdened with pharmacological endoplasmic reticulum stress. Mol. Biol. Cell 21, 2975- 2986

Yamazaki, H., Hiramatsu, N., Hayakawa, K., Tagawa, Y., Okamura, M., Ogata, R., Huang, T., Nakajima, S., Paton, A.W., Paton, J.C., Kitamura, M., 2009. Activation of the Akt-NF-kB pathway by subtilase cytotoxin through the ATF6 branch of the unfolded protein response. J. Immunol. 183, 1480-1487.

112 Yan, Y., Sabharwal, P., Rao, M., Sockanathan, S., 2009. The Antioxidant Enzyme Prdx1 Controls Neuronal Differentiation by Thiol- Redox-Dependent Activation of GDE2. Cell 138, 1209-1221.

Yang, X., Matsuda, K., Bialek, P., Jacquot, S., Masuoka, H.C., Schinke, T., Li, L., Brancorsini, S., Sassone-Corsi, P., Townes, T.M., Hanauer, A., Karsenty, G., 2004. ATF4 is a substrate of RSK2 and an essential regulator of osteoblast biology; implication for Coffine-Lowry Syndrome. Cell 117, 387-398.

Ye, J., Rawson, R.B., Komuro, R., Chen, X., Dave, U.P., Prywes, R., Brown, M.S., Goldstein, J.L., 2000. ER stress induces cleavage of membrane-bound ATF6 by the same proteases that process SREBPs. Mol. Cell 6, 1355-1364.

Yeh, W.C., de la Pompa, J.L., McCurrach, M.E., Shu, H.B., Elia, A.J., Shahinian, A., Ng, M., Wakeham, A., Khoo, W., Mitchell, K., El-Deiry, W.S., Lowe, S.W., Goeddel, D.V., Mak, T.W., 1998. FADD: essential for embryo development and signaling from some, but not all, inducers of apoptosis. Science 279, 1954-1958.

Yoon, D., Pastore, Y. D., Divolky, V., Liu, E., Mlodnicka, A. E., Rainey, K., Ponka, P., Semenza, G. L., Schumacher, A., Prchal, J. T., 2006. Hypoxia-inducible Factor-1 Deficiency Results in Dysregulated Erythropoiesis Signaling and Iron Homeostasis in Mouse Development. J. Biol. Chem. 281, 25703-25711.

Yoshida, H., Haze, K., Yanagi, H., Yura, T., Mori, K., 1998a. Identification of the cis- acting endoplasmic reticulum stress response element responsible for transcriptional induction of mammalian glucose-regulated proteins, involvement of basic-leucine zipper transcription factors. J. Biol. Chem. 273, 33741-33749.

Yoshida, H., Kong, Y.Y., Yoshida, R., Elia, A.J., Hakem, A., Hakem, R., Penninger, J.M., Mak, T.W., 1998b. Apaf1 is required for mitochondrial pathways of apoptosis and brain development. Cell 94, 739-750.

Yoshida, H., Matsui, T., Yamamoto, A., Okada, T., Mori, K., 2001. XBP1 mRNA is induced by ATF6 and spliced by IRE1 in response to ER stress to produce a highly active transcription factor. Cell 107, 881-891.

Yuan, T., Weljie, A.M., Vogel, H.J., 1998. Tryptophan fluorescence quenching by methionine and selenomethionine residues of calmodulin: orientation of peptide and protein binding. Biochem. 37, 3187-3195.

Yuan, L., Cao, Y., Oswald, F., Knochel, W., 2008. IRE1b is required for mesoderm formation in Xenopus embryos. Mech. Dev. 125, 207-222.

113 Zaidi, A., Barron, L., Sharov, V.S., Schoneich, C., Michaelis, E.K., Michaelis, M.L., 2003. Oxidative inactivation of purified plasma membrane Ca2+-ATPase by hydrogen peroxide and protection by calmodulin. Biochemistry 42, 12001-12010.

Zhang Y., Moore, J.N., 1996. Selenium Fractionation and Speciation in a Wetland System. Environ. Sci. Technol. 30, 2613-2619.

Zhang, P., McGrath, B., Li, S., Frank, A., Zambito, F., Reinert, J., Gannon, M., Ma, K., McNaughton, K., Cavener, D.R., 2002. The PERK eukaryotic initiation factor 2 alpha kinase is required for the development of the skeletal system, postnatal growth, and the function and viability of the pancreas. Mol. Cell. Biol. 22, 3864-3874.

Zhang, W., Feng, D., Li, Y., Lida, K., McGrath, B., Cavener, D.R., 2006. PERK EIF2AK3 control of pancreatic beta cell differentiation and proliferation is required for postnatal glucose homeostasis. Cell. Metab. 4, 491-497.

Zhang, K., Wong, H.N., Song, B., Miller, C.N., Scheuner, D., Kaufman, R.J., 2005. The unfolded protein response sensor IRE1a is required at 2 distinct steps in B cell lymphopoiesis. J. Clin. Invest. 115, 268-281.

Ziegler, D.M. 2002. An overview of the mechanism, substrate specificities and structure of FMOs. Drug. Metab. Rev. 34, 503–511.

Ziegler, D. M., Mitchell, C. H., 1972. Microsomal oxidase IV: properties of mixed- function amine oxidase isolated from pig liver microsomes. Arch. Biochem. Biophys. 150, 116–125.

Ziegler, D. M., Poulsen, L.L., 1978. Hepatic microsomal mixed-function amine oxidase. Methods Enzymol. 52C, 142–151.

Zou, H., Niswander, L., 1996. Requirement for BMP signaling in interdigital apoptosis and scale formation. Science 272, 738-741.

114 Chapter 2: Comparative Developmental Toxicity of Desalination Brine and Sulfate

Dominated Saltwater in a Euryhaline Fish

Abstract

Desalination of seawater and brackish groundwater is a promising sustainable solution to meet growing water needs of cities across the US with water restrictions.

However, the environmental impacts of the resulting filtrate (often referred to as brine) discharged to surface water need to be evaluated before large-scale desalination can be successful in the US. Developing fish embryos and larvae are especially sensitive to changes in salinity as well as varying ionic compositions of water. Limited research has been performed on the impacts of hypersalinity on chronic vertebrate embryonic development, particularly on sublethal adverse effects. To investigate these impacts, embryos of the euryhaline fish model, Japanese medaka (Oryzias latipes) were treated with: 1. Graphite filtered freshwater; 2. Artificial seawater (17, 35, 42, 56, and 70 parts per thousand (ppt)); 3. Effluent from a desalination facility at Monterey Bay Aquarium,

CA, diluted to 75%, 50%, and 25% with 35ppt artificial seawater to simulate mixing (39,

42, 46 and 50ppt); 4. Artificial San Joaquin River water (CA, USA) (9, 13, and 17 ppt); and 5. Artificial San Joaquin River water diluted to 75%, 50%, and 25% with artificial seawater to simulate estuarine mixing in the San Francisco Bay (13, 19, 24, and 30 ppt).

Percent hatch, survival post hatch, deformities, swim bladder inflation, and median day to hatch were recorded to calculate EC50 and NOEC values. No significant difference was observed between artificial seawater and Monterey Bay aquarium effluent; EC50 values varied between 45-55 ppt. However, San Joaquin river water decreased survival post

115 hatch and increased deformities (EC50 > 17 ppt) in comparison to artificial seawater and

San Joaquin River water mixed with seawater, suggesting that unique ion compositions such as sulfate may play a role in embryo and larval toxicity.

Introduction

Population growth, drought, and climate change have placed strain on water resources around the world, leading to great interest in alternative methods of potable water production, such as desalination of seawater and brackish groundwater. Many countries currently utilize desalination, and several desalination facilities are already active along the California, USA, coast. However, 20 additional facilities are being considered for production within the next 15 years (Cooley et al., 2006). Although desalination has the potential to service millions of CA residences with potable water, possible environmental impacts need to be evaluated to ensure proper regulation.

Concerns associated with desalination include high energy costs, and toxicity of reverse osmosis (RO) reject or effluent (aka brine) to receiving surface waters (Lattemann &

Hopner, 2008). The majority of proposed desalination facilities are intending to discharge effluent directly into the open ocean. However, other facilities have indicated that brine may also be discharged into rivers and estuaries of the San Francisco Bay Delta (Cooley et al., 2006). Embayment locations have decreased dissolution and dispersal potential, which can possibly result in an increase in ambient salinity (Jenkins et al., 2013).

Furthermore, desalination of brackish groundwater will produce reject with a different ionic content from seawater, which may have different toxicity to aquatic organisms.

116 Embryonic and larval developmental periods in teleosts are often the most sensitive to environmental stressors (Westernhagen, 1998). Although studies have been conducted on the toxicity of desalination brine to marine invertebrate development

(Voorhees et al., 2013) and the acute toxicity of ion imbalance to freshwater organisms

(Goodfellow et al., 2000), there is a lack on information of brine toxicity to euryhaline vertebrate development. Furthermore, few studies have addressed sublethal impacts of desalination effluent on developing organisms.

To better characterize the risks that may arise from various “brines” in CA waterways, we examined the impacts of artificial seawater, reverse osmosis (RO) brine from a facility at Monterey Bay Aquarium (CA, USA) and an artificial brackish water in the San Joaquin River, CA, on the embryonic development of the euryhaline fish model

Japanese medaka (Oryzias latipes). The flexibility of the Japanese medaka to a wide- range of salinities allows for direct comparisons between freshwater, ambient seawater and greater salinities. We assessed both lethal (percentage hatch and survival post hatch) and sublethal (percentage deformities, swim bladder inflation, median day to hatch) endpoints in order to provide CA regulators with information regarding potential ecological risks associated with brine discharge from desalination.

Methods

Embryo Collection and Exposure

Japanese medaka (Oryzias latipes) were cultured at the University of California-

Riverside at 27°C. Tanks were aerated under 14hour light and 10hour dark cycles. Adults

117 were fed twice a day a diet of live brine shrimp. We specifically used Japanese medaka because they are a euryhaline species that are widely accepted as a model for testing toxicity.

Embryos were pooled from all adults and separated at random for treatment at the

2-4-cell stage in 60 x 15 mm petri dishes with 15-20 eggs per dish. The viable eggs were separated according to Kirchen and West (1976). Four treatment waters were used in this study: freshwater, artificial seawater, artificial San Joaquin River, CA, water, and RO effluent brine discharged from a facility at Monterey Bay Aquarium, CA. Medium-hard, graphite filtered freshwater treatments were used as a quality control with each replicate.

If embryo survival in freshwater was below 80%, the replicate was discarded. Instant

Ocean Aquarium Salts were used to prepare the artificial seawater (Table 2.1). All water dilutions used in the study were prepared daily from stock waters prior to treatment and aerated to 70% saturation before treatment. Water and dishes were replaced daily until the end of the experiment. Full strength seawater (100%) was assumed to be at 35ppt.

Accordingly, artificial seawater treatments were prepared at: 17ppt (50%), 35ppt (100%),

42ppt (130%), 56ppt (160%) and 70ppt (200%) (n = 3-5).

Artificial San Joaquin River (SJR) saltwater was prepared based on water from the Westlands Water District (Table 2.1). This district resides around 10 km south of

Mendota within San Joaquin River Drainage Basin, CA. Use of this water is relevant since desalination projects are proposed for this area. Full strength SJR water is 13 ppt

(Table 2.1) as measured with a refractometer. Accordingly, the artificial SJR water was concentrated from 13 ppt to 17 ppt via evaporation by heating or diluted to 9 ppt in order

118 to ascertain the toxicity of SJR brine alone (n = 5). The low solubility of CaSO4 present in the water prevented concentration above 17 ppt. Subsequently, to simulate discharge of

SJR brine into marine environments, full strength SJR water (13 ppt) was diluted 25%

(19 ppt), 50% (24 ppt) and 75% (30 ppt) with full strength artificial seawater (35 ppt) (n

= 4).

The Monterey Bay Aquarium, CA, maintains a small seawater desalination facility to produce industrial grade water, which is not treated with any chemicals prior to

RO concentration. The aquarium provided RO desalination brine at 50 ppt to compare with artificial seawater, which was stored at 4°C throughout the experiment. Full strength

Monterey Bay Aquarium (MB) brine was sequentially diluted with full strength seawater in order to simulate mixing within an ocean environment. Effluent was diluted 25% (46 ppt), 50% (42 ppt), and 75% (39 ppt) in 35 ppt artificial seawater (n = 5).

Embryos were observed daily under a compound microscope for hatch, mortality, and deformities. Un-anesthetized larvae were examined for spinal (curvatures such as lordosis, kyphosis and scoliosis), cranio-facial (ocular abnormalities and abnormalities in head shape), cardiac (edema), fin (fin shape) and yolk sac (edema) malformations. The presence of an inflated swim bladder was also noted, as the swim bladder is important for buoyancy and balance control, however, it was scored separately from total deformities, as this occurs naturally in control fish and is often reversible. In addition, the survival of the hatched larvae was monitored for three days post hatch.

119 Ion and Water Quality Measurements

All tested waters were analyzed for major ionic constituents. Cations (Ca2+, Na+,

Mg2+, Sr2+ and K+) were analyzed with inductively coupled plasma optical emission spectroscopy (ICP-OES) (Perkin Elmer Optima 3000 DV, Waltham, MA), while anions

- 2- (Cl and SO4 ) were analyzed with an IonPac ASIV high capacity anion-exchange column on a Dionex 500 ion chromatograph (Sunnyvale, CA). Nitrogen species were measured with aquarium test strips (Tetra) for prepared waters, and according to

Kingsley et al. (2014) for brine from Monterey Bay Aquarium. Results in comparison to nominal concentrations are summarized in Table 2.1.

Table 2.1. Nominal and measured concentrations of ions by ICP-OES and ion chromatography, and water chemistry of freshwater, Instant Ocean, artificial San Joaquin River Water, and Monterey Bay Aquarium Brine (n = 3).

Instant Ocean SJR MB Brine Freshwater Measured Nominala Measured Nominalb Measured Measured Conc. Ion Conc. (mg/L) Conc. (mg/L) Conc. (mg/L) Conc. (mg/L) Conc. (mg/L) (mg/L) Na+ 10780 9875 ± 17.6 4252 4245 ± 36.9 15094 ± 0.48 51 ± 2.04 K+ 420 489 ± 2 22 30 ± 0.36 723 ± 1.91 3.2 ± 0.12 Mg2+ 1320 1089 ± 2.77 131 116.3 ± 0.75 1710 ± 0.79 7.7 ± 0.39 Ca2+ 400 375 ± 0.57 392 382 ± 2.91 588 ± 0.48 66 ± 2.91 Sr2+ 8.8 14 ± 0.03 0 0 17 ± 0.04 0.42 ± 0.02 Cl- 19290 17120 ± 400 942 978 ± 7.61 26619 ± 244 32.5 ± 0.4 2- SO4 2660 2294 ±43.5 8974 8628 ± 122 3355 ± 36.9 76.1 ± 0.97 - c NO3 0 ND 0 ND 1.96 ± 0.1 10 ± 10 - NO2 0 ND 0 ND 0.01 ± 0 0.25 ± 0.35 + NH3 0 ND 0 ND ND 0.25 ± 0.35 pH - 8.3 - 7.7 8.2 8.4 a Havonec, 2015 b Schlenk, et al., 2003 c Not Detected

120 Statistics

Data was tested for normality and homogeneity of variances, which was not met.

The Kruskal-Wallis test was performed with a Dunn’s Test post hoc and a Holm-Sidak correction. Statistical tests were performed in the statistical program Stata 13.1

(StataCorp LP) with the dunntest package. Statistical significance was determined at p ≤

0.05. EC50 values were calculated with the probit method and 95% confidence intervals were calculated with the Litchfield-wilcoxon method (Litchfield & Wilcoxon, 1949).

Calculated EC50 values that were greater than the highest test concentration are listed as > the highest concentration.

Results

In the present study, vertebrate developmental and larval toxicity of desalination brine was compared to artificial seawater and saltwater of differing ionic contents.

Desalination reject brine and San Joaquin River water were diluted with full strength seawater to simulate mixing in marine waters. SJR water and SJR mixed with seawater had no significant effect on percent hatch at any salinity (Figure 2.S1). In contrast, percent hatch was significantly decreased to 37% in 56 ppt seawater (160% full strength) and 22% in 70ppt (200%) (p = 0.03) (Figure 2.S1). Similarly, significant decreases in percent hatch were observed with 50 ppt RO brine (100% effluent; p = 0.001; Figure

2.S1). EC50 values for the RO brine and seawater, were 55.9 ppt (95% CI 48.1-65 ppt) and > 50 ppt, respectively (Table 2.2).

121 Table 2.2. No Observable Effect Concentrations (NOEC) and EC50 values for the four different treatments.

End point Saltwater Range Tested NOEC EC50 95% CI Hatch Artificial Seawater 17, 35, 42, 56, 70 42 55.9 48.1-65 Brine + Seawater 39, 42, 46, 50 42 >50 San Joaquin River 9, 13, 17 >17 San Joaquin River + Seawater 13, 19, 24, 30 >30 Survival Post Artificial Seawater 17, 35, 42, 56, 70 35 49 46.5- Hatch 51.6 Brine + Seawater 39, 42, 46, 50 46 San Joaquin River 9, 13, 17 NA San Joaquin River +Seawater 13, 19, 24, 30 >30 Deformities Artificial Seawater 17, 35, 42, 56 35 >56 Brine + Seawater 39, 42, 46, 50 46 >50 San Joaquin River 9, 13, 17 13 >18 San Joaquin River + Seawater 13, 19, 24, 30 >30 Swim Bladder Artificial Seawater 17, 35, 42, 56 17 >56 Inflation Brine + Seawater 39, 42, 46, 50 42 45.1 38-53.6 San Joaquin River 9, 13, 17 13 >17 San Joaquin River + Seawater 13, 19, 24, 30 >30

Total deformities were measured following hatch in live larvae. All embryos

treated with 70 ppt seawater were dead immediately post hatch and were not assessed for

deformities. The percent of seawater deformities increased to 18% following exposure to

42 ppt seawater (p = 0.01) and to 50% in 56 ppt seawater (p = 0.001) (Figure 2.S2). The

percent of deformities increased to 22% when medaka were treated with 50 ppt of RO

brine (p = 0.001) (Figure 2.S2). SJR water mixed with seawater resulted in no increase in

deformities; however, 17 ppt SJR water alone caused 53% deformities (p = 0.0004)

(Figure 2.1). This is in contrast to 17ppt seawater exposures (50% concentrated), which

failed to cause deformities. The EC50 values reflect this difference: > 56 ppt for seawater,

> 50 ppt for RO brine, and > 17 ppt for SJR (Table 2.2).

122

Figure 2.1. Percent total deformities ± standard error (SE) of Japanese medaka embryos following treatment with (A) San Joaquin River saltwater at full strength (13ppt), concentrated (17ppt), and diluted (9ppt) (n = 5) and (B) with San Joaquin River water full strength (100%, 13ppt), mixed with seawater, to 75% (19ppt), 50% (24ppt) and 25% (30ppt) (n = 4). Differing letters indicate significant differences between treatments (p ≤ 0.05) following Dunn’s test.

Swim bladder inflation is important for maintenance of buoyancy and balance in the water column for pelagic fish species. A concentration response of swim bladder inflation to salinity was observed in all saltwater/brine treatments except SRJ water mixed with seawater. Swim bladder inflation decreased significantly to 46% in embryos treated with 46 ppt RO brine (25% dilution) and to 52% in 35 ppt artificial seawater

(Figure 2.2; p = 0.0008 and 0.002 respectively). Furthermore, SJR water significantly

123 decreased swim bladder inflation at 13 and 17 ppt to 73% and 53%, respectively (Figure

2.2; p = 0.01 and 0.0004). Calculated EC50 values were > 56 ppt for seawater, 45.1 ppt

(95% CI 38-53.6) for desalination brine, and >17 ppt for SJR water. This further suggests that SJR water is more toxic to developing fish than seawater, based solely on salinity.

Figure 2.2. Percent swim bladder inflation ± SE in Japanese medaka embryos at hatch following treatment with (A) artificial seawater (n = 3-5), (B) RO reject mixed with artificial seawater (n = 5), (C) San Joaquin River water (n = 4), and (D) San Joaquin River water mixed with seawater (n = 5). Differing letters indicate significant differences between treatments (p ≤ 0.05) following Dunn’s test.

Embryos were monitored for 3 days post hatch (until feeding) for survival.

Survival decreased to 78% in 42 ppt seawater (p = 0.04), to 20% in in 56 ppt (p = 0.0014) and to 0% in 70 ppt (p = 0.00014; Figure 2.S3). RO brine treated larvae survival

124 decreased to 90% in 50 ppt (full strength effluent; p = 0.0014; Figure 2.S3). Although no decrease in survival was observed in SJR water mixed with seawater embryos, SJR water alone resulted in 17% survival at 9 ppt (p = 0.0065), 50% survival at 13 ppt (p = 0.04) and 1% survival at 17 ppt (p < 0.0001; Figure 2.3). Due to the U-shaped nature of the dose response curve for SJR water, an EC50 for survival could only be calculated for seawater (49 ppt; 95% CI 46.5-51.6 ppt).

Figure 2.3. Percent survival 3 days post hatch ± SE of Japanese medaka embryos following treatment with (A) San Joaquin River saltwater at full strength (13ppt), concentrated (17ppt), and diluted (9ppt) (n = 5) and (B) with San Joaquin River water full strength (100%, 13ppt), mixed with seawater, to 75% (19ppt), 50% (24ppt) and 25% (30ppt) (n = 4). Differing letters indicate significant differences between treatments (p ≤ 0.05) following Dunn’s test.

Finally, the median day to hatch was calculated for each treatment group. No significant alteration in time to hatch was observed in SJR water alone or when mixed with seawater (Figure 2.S4). However, RO brine significantly increased time to hatch from 9.1 ± 0.14 days post fertilization (dpf) in freshwater to 10.4 ± 0.25 to 10.8 ± 0.2 dpf at all concentrations tested (p < 0.005 for all; Figure 2.4). Artificial seawater also significantly increased day to hatch beginning at 17 ppt to 10.25 ± 0.31 dpf, at 42 ppt to

125 10.4 ± 0.25 dpf, to 11 ± 0.37 dpf in 56ppt exposures and to 14.5 ± 0.71 dpf in 70 ppt exposures (p = 0.047, 0.03, 0.007, and 0.007, respectively).

Figure 2.4. Median day to hatch ± SE of Japanese medaka embryos treated with (A) artificial seawater (17, 35, 42, 56, and 70ppt corresponding to 50%, 100%, 130% 160% and 200% concentrated) (n = 3-5); and (B) RO reject mixed with artificial seawater to 100% (50ppt), 75% (46ppt), 50% (42ppt) and 25% (39ppt) (n = 5). Differing letters indicate significant differences between treatments (p ≤ 0.05) following Dunn’s test.

126 Discussion

Although data is available concerning acute saltwater impacts on invertebrate development, no true chronic vertebrate embryonic development tests have been conducted. Furthermore, most tests on euryhaline or marine organisms have been performed only with seawater effluent (Voorhees et al., 2013; Inoue & Takei, 2002; Iso et al., 1994), while tests with effluent from brackish water have been conducted only with freshwater organisms (Mount et al., 1997; Wang et al., 2016; Soucek & Kennedy, 2005).

In the present research, we characterize EC50 and NOEC values for toxicity of three different saltwaters, including seawater desalination brine, on vertebrate embryonic and larval development and examine sublethal endpoints of toxicity. We accomplish this with a common, euryhaline model species, the Japanese medaka. Use of a euryhaline model is pertinent because it is relevant for estuarine environments and allows for a wide range of salinities to be tested.

EC50 values measured in this study ranged between 45-60 ppt for artificial seawater and RO reject from Monterey Bay. Overlapping 95% confidence intervals indicated no difference between the RO reject and seawater toxicity to embryos and larvae. Although reverse osmosis may concentrate trace toxicants present at low levels, this appears to not have impacted toxicity and the results were expected considering no chemical treatment was added to the effluent. This study is one of the few to examine sublethal indicators of brine toxicity. No difference in EC50 or NOEC was observed between percent hatch, survival and deformities. However, NOEC values were lower for failure of swim bladder inflation, suggesting effects following alterations in salinity

127 above 35 ppt seawater. Uninflated swim bladders may negatively impact fish; for instance, Japanese medaka larvae with uninflated swim bladders consumed more oxygen than those with inflated swim bladders (Marty et al., 1995). This metric may be a good indicator of sublethal effects from brine discharge. Additionally, the median day to hatch was significantly increased by exposures of greater than 35 ppt. Increased time to hatch may decrease larval competitiveness and survival in the wild (Kestemont et al., 2003).

Current regulations on brine discharge in California are characterized as increases in salinity over ambient or by absolute salinity. The US EPA has set a limit of an increase

< 4 ppt, while the San Diego Regional Water Quality Control Board and Santa Ana

Regional Water Quality Control Board set an absolute salinity of < 40 ppt in 2006 and

2012, respectively, for oceanic discharge (Jenkins et al., 2013). Other countries utilizing desalination as a water source, such as Australia, Oman, or Japan suggest increments of

<1 or 2 ppt for oceanic discharge (Jenkins et al., 2013). Embryos used in the current study were spawned in freshwater and no significant toxicity was observed in 35 ppt seawater. Hence, according our results and assuming a seawater salinity of 35 ppt, the high flexibility of Japanese medaka embryos and larvae indicates that they would be protected to seawater effluent under these guidelines.

Previous studies have been performed investigating the lethal effects of seawater on marine/euryhaline embryos and larvae. Inoue & Takei (2002) reported that Japanese medaka embryos spawned in 50% seawater hatched at a rate of 90%, and a rate of 60% in full seawater (Inoue & Takei, 2002). Conversely, in the current study, percentage hatch was unaffected by 35 ppt and 17 ppt seawater. Time to hatch was affected in both our

128 study and the Inoue and Takei (2002) study approximately 1-2 days. Studies on artificial seawater have also been conducted in flounder (Pleuronectes yokohumae) eggs and larvae (Iso et al., 1994). Saltwater acclimated flounder larvae experienced a mortality rate of 12% in 50 ppt seawater and 100% in 60 ppt13. Furthermore, embryo percent hatch remained at 100% at 60 ppt, but decreased to 0% at 70 ppt (Iso et al., 1994).

While informative, the above experiments were performed on marine organisms with only artificial seawater and give no indication of differences between RO brine and seawater toxicity. Short-term development tests on purple sea urchin, red abalone, sand dollar, and bay mussels have been conducted with desalination reject brine from

Monterey Bay and indicated EC50 values between 36 and 43 ppt (Voorhees et al., 2013).

The invertebrate embryos used in these tests were spawned in seawater, while the medaka embryos used in this study were spawned in freshwater. Thus, the medaka embryos were subjected to greater overall changes in salinity than the invertebrates at fertilization, yet greater EC50 values were measured. This suggests that marine invertebrates are more sensitive to alterations in salinity than euryhaline Japanese medaka. However, larval assays performed with saltwater acclimated topsmelt, suggested an EC50 of 62ppt

(Voorhees et al., 2013). This value is greater than the EC50 values calculated for larval survival here (49ppt), suggesting that medaka larvae are more sensitive to high salinities.

This difference could be due differences in methods, as topsmelt were transferred into the test solution as larvae, while the Japanese medaka were reared in the corresponding salt and spawned in freshwater.

129 Although no differences in desalination brine and artificial seawater were noted, significant differences between SJR alone, seawater alone, and seawater mixed with SJR were observed. For instance, 17 ppt SJR water resulted in 50% deformities and 1% survival post hatch, while treatment with 17 ppt seawater generated no deformities and

84% survival. Furthermore, a hormetic dose response curve was observed for survival of larvae treated with SJR water. This may be due to a lack of necessary ions in the 9 ppt treatments, which were supplemented at 13 ppt, but toxic at 17 ppt. SJR water mixed with seawater produced results similar to seawater alone. These differences suggest that the ion ratios and imbalance are larger drivers of embryo and larval toxicity, rather than the total dissolved solids. Furthermore, sublethal indicators of toxicity (deformities and swim bladder inflation) were a more sensitive metric than percent hatch in the current study, suggesting that they may be of use in assessing desalination impacts of brackish groundwater.

In a study on coho salmon (Oncorhynchus kisutch Walbaum), embryos were exposed to 2500mg/L (2.5 ppt) TDS mimicking ionic content of mining effluent in

Alaska (Stekoll et al., 2009). The solution was sulfate dominated, primarily composed of

CaSO4, Na2SO4, and MgSO4, similar to the water used in this study. Approximately 50% mortality was observed in embryos treated from fertilization to swim up (Stekoll et al.,

2009). Researchers further calculated EC50 values for the impacts of individual ions on

2+ + 2+ 2- coho salmon fertilization and found the order of toxicity to be; Ca > K > Mg > SO4

> Na+ with a range of 102mg/L (0.102 ppt) to 4744mg/L (4.7 ppt) (Stekoll et al., 2009).

130 Similar to the results observed here, embryos exposed through swim-up experienced high post-hatch mortality (Stekoll et al., 2009).

In stenohaline freshwater organisms, combinations of various salts were tested for toxicity to Daphnia magna, Ceriodaphnia dubia and Pimephales promelas (Mount et al.,

1997). LD50 values for the combination of NaCl and Na2SO4, the primary salts present in the artificial San Joaquin water, were approximately 6000mg/L (6 ppt) for the 96 hr fathead minnow test and 5700mg/L (5.7 ppt) for the 48 hr Daphnia magna test (Mount et

+ - 2+ al., 1997). Ion toxicity modeling suggested that in order of toxicity, K > HCO3 = Mg >

- 2- 2+ + Cl > SO4 , while Ca and Na were not significant (Mount et al., 1997). With the exception of Ca2+, agreement exists between this model and the one determined for coho salmon (Stekoll et al., 2009).

However, other research suggests that ion ratios may be more important than concentration in determining toxicity. In manufacturing effluent, high ratios of Ca2+ to

Na+ were found to be responsible for fathead minnow toxicity (Dorn & Rogers, 1989).

Ca2+ to Na+ ratios of 15:1 caused high mortality, while 1:20 did not (Dorn & Rogers,

2+ 1989). EC50 values for Ca were calculated for Daphnia pulex, Mysidopsis bahia, and

Pimephales promelas and ranged between 266 (0.27ppt) and 927mg/L (0.926ppt)

(Goodfellow et al., 2000). However, these values were all based on adult 96hr exposures to larval and juvenile freshwater organisms, and likely vary for euryhaline or saltwater organisms during embryonic development. Very low levels of K+ and Mg2+ are found in the SJR water, indicating low potential for toxicity; however, 392mg/L (0.392 ppt) Ca2+

131 present in the SJR is greater than the calculated EC50 values for coho salmon and fathead minnows (Goodfellow et al., 2000; Stekoll et al., 2009).

2+ 2- In addition to Ca , a major component of the SJR water is SO4 . A comprehensive study on sulfate toxicity to freshwater species (mussels, midges, fathead minnows, and a cladoceran) has been performed (Wang et al., 2016). Chronic development tests with fathead minnows suggested growth/weight EC20’s of 305-

477mg/L (0.305-0.477 ppt) for survival, and 106-185mg/L (0.106-0.185 ppt) (Wang et al., 2016). Short term (7-14 day) embryonic LC50 values ranged from 478mg/L to

644mg/L (0.478-0.644 ppt). These values are substantially lower than those used in this study, most likely because the species tested were stenohaline freshwater organisms in comparison to the euryhaline organism tested here. However, Wang et al. (2016) report that increasing the concentration of K+ from 1mg/L to 3mg/L (0.003 ppt) decreased sulfate toxicity, while increasing Cl- had no effect. The SJR water used here contained

2- + very high ratios of SO4 to K , which may have contributed to toxicity.

In contrast, other studies on sulfate toxicity to a freshwater invertebrate, Hyalella

2- - azteca, found that SO4 toxicity decreased with increasing levels of Cl by 5.5 fold

(Soucek & Kennedy, 2005). Furthermore, sulfate toxicity was also reduced with increasing water hardness as CaCO3, which was not present in the prepared SJR water.

Additional studies on the impacts of these factors in H. azteca and D. magna, indicated

2- that chloride concentrations from 5-25mg/L (0.005-0.025ppt) decreased SO4 toxicity, while greater concentrations caused it to increase, suggesting an additive effect (Soucek,

2- 2001). Although SO4 levels predominate in SJR water, based on these studies, its

132 possible that the Cl- levels may also have an effect. We can conclude that further chronic testing of ion ratios on vertebrate development of a variety of endemic species is necessary to fully understand the impacts of brackish water desalination.

Overall our results indicate that developing euryhaline organisms (specifically

Japanese medaka) will be protected from seawater brine toxicity under the current regulations. However, a general TDS measurement for National Pollutant Discharge

Elimination System (NPDES) permits is insufficient for brine of different ionic contents, and alternative regulations need to be in place for brackish groundwater desalination.

Recently, researchers have advocated for the use of site-specific and ion-specific standards for adequate ecosystem protection (Canedo-Arguelles, 2016), with which we agree in the case of desalination. Further research on ion imbalance and endemic species is necessary to determine site-specific impacts of brackish water desalination brine.

Additionally, the implications of ion imbalance need to be considered for estuarine disposal of groundwater desalination reject.

133 References

Cañedo-Argüelles, M., Hawkins, C.P., Kefford, B.J, Schäfer, R.B., Dyack, B.J., Brucet, S., Buchwalter D., Dunlop, J., Frör, O., Lazorchak, J., Coring, E., Fernandez, H.R., Goodfellow, W., González Achem, A.L., Hatfield-Dodds, S., Karimov, B.K., Mensah, P., Olson, J.R., Piscart, C., Prat, N., Ponsá, S., Schulz, C.-J., Timpano, A.J., 2016. Ion- specific standards are needed to protect biodiversity. Science 351, 914-916.

Cooley, H., Gleick, P.H., Wolff, G., 2006. Desalination, with a Grain of Salt- A California Perspective. Pacific Institute for Studies in Development, Environment, and Security: Oakland, CA.

Dorn, P.B., Rodgers, J.H.Jr., 1989. Variability associated with identification of toxics in NPDES effluent toxicity tests. Environ. Toxicol. Chem. 8, 893–902.

Goodfellow, W.L., Ausley, L.W., Burton, D.T., Denton, D.L., Dorn, P.B., Grothe, D.R., Heber, M.A., Norberg-King T.J., Rodgers, J.H. Jr., 2000. Major ion toxicity in effluents: A review with permitting recommendations. Environ. Toxicol. Chem. 19, 175–182.

Havonec, T.A., 2015. Synthetic Sea Salts: Are They All Equal? A Primer on Instant Ocean®: The Premier Sea Salt for Your Marine Aquarium. Marineland, Spectrum Brands, Blacksburg, VA.

Inoue, K., Takei, Y., 2002. Diverse adaptability in oryzias species to high environmental salinity. Zool. Sci. 19, 727−734.

Iso, S., Suizu S., Maejima, A., 1994. The lethal effect of hypertonic solution and avoidance of marine organisms in relation to the discharged brine from desalination plant. Desalination 97, 389- 399.

Kestemont, P., Jourdan, S., Houbart, M., Melard, C., Paspatis, M., Fontaine, P., Cuvier, A., Kentouri, M., Baras, E., 2003. Size heterogeneity, cannibalism and competition in cultured predatory fish larvae: biotic and abiotic influences. Aquaculture 227, 333–356

Kingsley, E., Modisette, N., Phillips, R., 2014. Development and Use of Nitrate and Total Ammonia Testing Procedures at the Monterey Bay Aquarium. 2nd Aquality Symposium: Water Quality Treatment in Zoos and Aquariums.

Kirchen, R.V., West, W.R., Carolina Biological Supply, C., 1976. The Japanese medaka: its care and development. Carolina Biological Supply Company.

134 Lattemann, S., Hopner, T., 2008. Environmental impact and impact assessment of seawater desalination. Desalination 220, 1-15

Litchfield, J. T. Jr., Wilcoxon, F., 1949. A Simplified Method for Evaluating Dose-Effect Experiments. J. Pharmacol. Exp. Ther. 96, 99-113.

Marty, G.D., Hinton, D.E., Cech, J.J., 1995. Oxygen Consumption by Larval Japanese Medaka with Inflated or Uninflated Swim Bladders. T. Am. Fish. Soc. 124623-627.

Schlenk, D., Zubcov, N., Zubcov, E., 2003. Effects of salinity on the uptake, biotransformation, and toxicity of dietary seleno-L-methionine to rainbow trout. Toxicol. Sci. 75, 309−313.

Soucek, D.J., 2007. Comparison of hardness- and chloride-regulated acute effects of sodium sulfate on two freshwater crustaceans. Environ. Toxicol. Chem. 26, 773-9

Soucek, D.J., Kennedy, A.J., 2005. Effects of hardness, chloride, and acclimation on the acute toxicity of sulfate to freshwater invertebrates. Environ. Toxicol. Chem. 24, 1204- 1210.

Stekoll, M.S., Smoker, W.W., Failor-Rounds, B.J., Wang, I.A., Joyce, V.J., 2009. Response of the early developmental stages of hatchery reared salmonids to major ions in a simulated mine effluent. Aquaculture 298, 172-181.

Voorhees, J.P., Phillips, B.M., Anderson B.S., Siegler K., Katz S., Jennings L., Tjeerdema R.S., Jensen J., Carpio-Obeso M.D., 2013. Hypersalinity toxicity thresholds for nine California ocean plan toxicity test protocols. Arch. Environ. Contam. Toxicol. 65, 665–670.

Wang, N., Dorman, R.A., Ingersoll, C.G., Hardesty, D.K., Brumbaugh, W.G., Hammer, E.J., Bauer, C.R., Mount, D.R., 2016. Acute and Chronic Toxicity of Sodium Sulfate to Four Freshwater Organisms in Water-Only Exposures. Environ. Toxicol. Chem. 35, 115- 127.

Westernhagen, H. von., 1998. Sublethal Effects of Pollutants on Fish Eggs and Larvae. In Fish Physiology, (D.S. Hoar, D.J. Randall, Eds.), Vol. Xia, pp. 253. Academic Press, San Diego, CA, pp 253.

135 Supplemental Information

Figure 2.S1. Percent hatch ± SE of medaka embryos following treatment with (A) artificial seawater, (B) RO reject mixed with artificial seawater, (C) San Joaquin River water, and (D) San Joaquin River water mixed with seawater. Differing letters indicate significant differences between treatments (p ≤ 0.05) following Dunn’s test.

Figure 2.S2. Percent deformities ± SE of Japanese medaka embryos treated with (A) artificial seawater (17, 35, 42, 56, and 70ppth corresponding to 50%, 100%, 130% 160% and 200% concentrated); and (B) RO reject mixed with artificial seawater to 100% (50ppth), 75% (46ppth), 50% (42ppth) and 25% (39ppth). Differing letters indicate significant differences between treatments (p ≤ 0.05) following Dunn’s test.

136

Figure 2.S3. Survival 3 days post hatch ± SE of Japanese medaka embryos treated with (A) artificial seawater (17, 35, 42, 56, and 70ppth corresponding to 50%, 100%, 130% 160% and 200% concentrated); and (B) RO reject mixed with artificial seawater to 100% (50ppth), 75% (46ppth), 50% (42ppth) and 25% (39ppth). Differing letters indicate significant differences between treatments (p ≤ 0.05) following Dunn’s test.

Figure 2.S4. Median day to hatch in days post fertilization ± SE of Japanese medaka embryos following treatment with (A) San Joaquin River saltwater at full strength (13ppth), concentrated (17ppth), and diluted (9ppth) and (B) with San Joaquin River water full strength (100%, 13ppth), mixed with seawater, to 75% (19ppth), 50% (24ppth) and 25% (30ppth). Differing letters indicate significant differences between treatments (p ≤ 0.05) following Dunn’s test.

137 Chapter 3: Flavin-containing Monooxygenase Developmental Expression and

Regulation by the Unfolded Protein Response in Japanese Medaka (Oryzias latipes)

Abstract

In fish, flavin-containing monooxygenases (FMOs) play a key role in xenobiotic metabolism, are regulated by environmental conditions such as osmotic pressure, and are differentially regulated during development in mammals. Japanese medaka (Oryzias latipes) are a common model euryhaline organism for toxicological studies. The goal of the current research was to characterize the developmental expression and regulation of

FMOs in Japanese medaka embryos in order to better understand the role of FMOs in this model species. Five putative medaka FMOs were characterized from the medaka genome through the National Center for Biotechnology Information (NCBI) database by protein motifs and alignments, then identified as FMO4, FMO5A, FMO5B, FMO5C and

FMO5D for the current study. FMO gene expression was analyzed at stages 19 (1dpf), 29

(3hpf), 36 (6dpf) and 39 (9dpf) and distinct developmental patterns of expression were observed. FMO4 and FMO5D were found to increase 3-fold during organogenesis (6dpf), while FMO5B and FMO5C decreased significantly beginning in early organogenesis

(3dpf). No alteration in expression was observed for FMO5A. Promoter analysis was performed for transcription factor binding sites using Consite and Alibaba and indicated regulation during embryonic development and a role for the unfolded protein response in

FMO modulation. FMO regulation by the UPR was assessed with treatments of 1µg/ml,

2µg/ml, and 4µg/ml Tunicamycin (Tm), and 2mM and 4mM dithiothreitol (DTT), well- known inducers of endoplasmic reticulum stress, for 24hrs from 5- 6dpf. High

138 concentrations to Tm were found to induce FMO4 and FMO5A up to two-fold, while

DTT significantly decreased expression of FMO5A, FMO5B, and FMO5C. Results suggest that FMO regulation during fish development is complex, and that medaka FMOs are variably regulated by the UPR during organogenesis and stage-dependent suggesting potential stage-dependent activation or detoxification of xenobiotics.

Introduction

Flavin-Containing Monooxygenases (FMOs) are multigene family of enzymes found in the endoplasmic reticulum of vertebrates, catalyzing the oxygenation of nucleophilic-heteroatom containing compounds. FMOs can play a significant role in xenobiotic metabolism and have a variety of endogenous substrates. There are primarily

5 functional vertebrate FMOs (Hernandez et al., 2004) and these genes exhibit unique developmental, species-specific expression patterns. For instance, FMO1 is expressed in the fetal human kidney and liver, yet adult expression is restricted to the kidney (Dolphin et al., 1996). Meanwhile, FMO3 expression in the liver begins at birth and dominates in the liver into adulthood (Hines, 2006). FMO knockout mice are viable, with alterations in lipid metabolism (Hernandez et al., 2009). This unique expression pattern has implications for xenobiotic and endobiotic metabolism during development and suggests some capacity for a developmental role.

Although research is available on mammalian FMOs, less is known about FMO expression and regulation in fish. FMOs are highly evolutionarily conserved in vertebrates and have been discovered in a range of organisms, including marine

139 invertebrates and teleosts (reviewed in Schlenk, 1998). It is hypothesized that fish FMOs evolved from the same phylogenetic lineage as mammalian FMOs, while many invertebrate FMOs evolved from a second lineage (Hao et al., 2009). However, fish

FMOs exhibit a monophyletic origin and cluster together, rather than with mammalian

FMOs of the same nomenclature (Hao et al., 2009). Fish FMOs have been shown to metabolize a number of xenobiotics, including N,N-dimethylanaline, thioether pesticides, thiocarbamates, thiocarbamines and thioamides (Schlenk, 1998). Furthermore, some mammalian FMO antibodies can recognize fish FMOs, including Japanese medaka

FMOs recognized by FMO1 and 3 antibodies (El-Alfy & Schlenk, 2002).

Regulation of FMOs by exogenous factors is under-studied. It was long believed that mammalian FMOs were not inducible. However, recent research found that dioxins induced FMO1, FMO2 and FMO3 in mice (Tijet et al., 2006). Still the mechanism behind these effects are complex and may not involve classic aryl-hydrocarbon receptor function as a transcription factor (Celius et al., 2010). In fish, FMOs have been shown to be regulated by alterations in salinity (Schlenk et al., 1996; Rodriguez-Fuentes et al., 2008;

Lavado et al., 2013), and an osmotic response element was discovered in the proximal promoter region of a trout FMO which was induced after NaCl treatment (Rodriguez-

Fuentes et al., 2008).

Considering the range of xenobiotics metabolized by fish FMOs in the environment, as well as the growing use of fish model species in toxicological research, the goal of the current study was to investigate the developmental expression patterns and regulation of FMO in the Japanese medaka (Oryzias latipes). Medaka are a common

140 model organism for both environmental toxicological research as well as developmental toxicology, making them an ideal organism for this study. Putative medaka FMOs were identified by National Center for Biotechnology Information (NCBI) and their expression characterized throughout development. Promoter analysis indicated the presence of transcription factor binding sites for the unfolded protein response (UPR), a coordinated stress response to the accumulation of unfolded proteins in the endoplasmic reticulum.

Thus, developmental expression levels of the UPR mediator, BiP (GRP78), and apoptotic mediator, CHOP (C/EBP-homologous protein) were determined. Medaka embryos were treated with UPR-inducers Tunicamycin (Tm) and Dithiothreitol (DTT) during liver organogenesis to investigate the ability of the UPR to modulate medaka FMO expression during development. Results indicated stage-specific expression of FMOs and an association with UPR during medaka development.

Methods

FMO Sequence Analysis

Putative FMOs from Japanese Medaka (Oryzias latipes) were identified by NCBI predictive inference software, and the predicted amino acid sequences analyzed for FMO- identifying motifs (including NADPH binding, FAD-binding, and EGLEP (Cashman &

Zhang, 2006)). Five potential FMOs were identified: one FMO4-like and four FMO5-like proteins. For simplicity and to remain in accordance with NCBI, the FMO5-like proteins are identified as FMO5A-D in the present study. Protein alignments were performed in

Clustal Omega 1.2.1 (Sievers et al., 2011) and verified with TCS (transitive consistency

141 score; http://tcoffee.crg.cat/apps/tcoffee/do:core; Chang et al., 2014;). Proximal promoter regions that were 1000 base pairs upstream of the predicted transcriptional start site were analyzed for conserved motifs and transcription factor binding sites using Consite

(Lenhard et al., 2003; http://consite.genereg.net/) and Alibaba2.1 (http://www.gene- regulation.com/pub/programs.html) using high match conservations of 85% and 80%, respectively. These two resources were chosen because they draw information from different databases; Consite uses the JASPAR database (Sandelin et al., 2004), while

Alibaba2.1 uses the TRANSFAC database (Matys et al., 2003). UniprotKB was used to identify the function of the transcription factors (http://www.uniprot.org/).

Embryo Collection and Exposures

Japanese Medaka were cultured at University of California-Riverside in accordance with animal use protocols (AUP #20140002). Fish were maintained on a

14:10 hour light:dark cycle at a ratio of 2:3 male:female at 27oC. Embryos were collected daily into de-chlorinated tap water and rinsed thoroughly. For gene expression analysis by stage, randomized embryos were collected at various stages throughout development: stage 19 (1 day post fertilization (dpf), 25 embryos/replicate), stage 29 (3dpf, 15 embryos/replicate), stage 36 (6dpf, 10 embryos/replicate), stage 39 (9dpf, 10 embryos/replicate) (n = 3-5).

Positive controls were used to induce the Unfolded protein response (UPR) in embryos. As FMO expression by stage indicated that the greatest changes in FMO expression occurred at 6dpf, embryos were treated with either Tunicamycin (Tm) or dithiothreitol (DTT) beginning at 5dpf for 24hrs. Tm is a canonical UPR-inducer that acts

142 by inhibiting protein glycosylation, while DTT is a strong reductant and has been used in several studies to reduce disulfide bonds. Tm (Sigma Aldrich) is a mixture of four homologs and was dissolved in DMSO to a stock concentration of 2mg/ml. At 5dpf, embryos were treated with 0.2% DMSO, 4µg/ml Tm in 0.2% DMSO, 2µg/ml Tm in 0.2%

DMSO, or 1µg/ml Tm in 0.1% DMSO for 24 hours (5-10 embryos/ replicate). DMSO alone, 1µg/ml Tm, and 2µg/ml Tm did not cause overt toxicity, however, 4µg/ml Tm caused a reduction in blood circulation in all embryos. Following treatment, embryos were rinsed and frozen for gene expression analysis (n = 7-10). For DTT studies, a 0.1M

DTT (Sigma Aldrich) stock solution was prepared fresh immediately prior to treatments to insure minimal oxidation of thiol moieties. At 5dpf, embryos were treated with freshwater, 2mM DTT, or 4mM DTT (5-10 embryos/replicate) for 24 hours. Post treatment, embryos were assessed for survival. Embryos treated with 2mM DTT had equal survival to controls. However, 43 ± 8% mortality was observed in embryos treated with 4mM DTT. Live embryos were frozen for analysis (n = 7-9).

FMO and UPR Gene Expression analysis

mRNA was isolated from pools of embryos using the Lipid Tissue RNeasy kit

(Qiagen, Valencia, CA) according to the manufacturers instructions. mRNA quantity and quality was determined with the ND-1000 (Nanodrop, Wilmington, DE). The Reverse

Transcription System (Promega Corporation, Madison, WI) was used to prepare cDNA from 1µg of mRNA according to the manufacturers instructions.

Predicted mRNA sequences for the putative FMO genes were used to design primers with the Integrated DNA Technologies PrimerQuest software and optimized

143 using PCR Miner (Zhao et al., 2005) (Table 3.1). Because many housekeeping genes vary by developmental stage, the average of EF1a and 18S was used to normalize stage expression data (Vandesompele et al., 2002). For instance, standard deviation for 18S and

EF1 individually was 1.6-1.3. However, when averaged the standard deviation was decreased to 0.3. For FMO studies at a single stage, 18S expression had less deviation within and between treatments than EF1a and was used as a housekeeping gene for UPR treatments. The iScript One-step RT-PCR kit with SYBR Green (Bio-Rad, Hercules, CA), with the reverse transcriptase omitted, was used on a MyiQ Thermocycler (Biorad) for qPCR expression studies. Samples were activated at 95°C for 5 min, then 40 cycles of

10s at 95°C and 30s of 55°C. Melting curve analysis was performed from 54 to 95°C in

0.5°C increments with continuous fluorescence measurement. Fold change was calculated against the controls according to Schmittgen & Livak (2008).

Table 3.1. Primers, Accession Numbers, and Concentrations used for qRT-PCR. Gene Primers (5’-3’) Accession # Conc. 18Sa CCTGCGGCTTAATTTGACTC AB105163.1 0.5µM GACAAATCGCTCCACCAACT EF1a CTACATCAAGAAGATCGGCTACAA NM_001104662.1 2.5µM CGACAGGGACAGTTCCAATAC CHOP CTCCTCTTCAGCATCCCAATC NM_001278882.1 2.5µM GCCTTTCGATCTCTGCCTTTA BiP GGAGGATTCTGACCTGAAGAAG NM_001278801.1 0.5µM GGTGACAGTAGGCTGGTTATC FMO4 GTCACCTTACCATCTCGTACAC XM_011483884.1 2.5µM CCTGTAGTTGAGTTTCCGGTAG FMO5A GGTGGACTGTGGAGGTTTAAG XM_004068272.2 2.5µM GTCGGCATACATTCGGTAGTAG FMO5B CGAGAAAGGTGGAGGTGAATAA XM_004068171.2 2.5µM AACTTGAGGACGGAACTTGG FMO5C TGCTTTACCTGCGTCTCTATG XM_011474346.1 5µM GGCTGAGTGTAATGACCAGAA FMO5D GCTCACAGGAGGAGAACATTAG XM_004068170.2 5µM ACTGGGATGATTGGTCCTTTAG aFrom Dong et al., 2010

144

Statistical Analysis

Statistical significance of qPCR data was determined in the statistical package R

(R Core Team, 2012; Vienna, Austria). Data was checked for normality and homogeneity of variance. Overall significance was assessed with an ANOVA and a Tukey’s test post hoc (for stage expression data) or a Dunnett’s test (for UPR treatments).

Results

Characterization of Putative medaka FMOs

NCBI predicted 5 potential medaka FMO genes: one FMO4-like gene present on chromosome 14 and 4 FMO5-like genes present on chromosome 4. For simplicity and to distinguish the cluster of FMO5-like genes from the FMO4-like gene, the FMO5-like proteins are named FMO5A, FMO5B, FMO5C, and FMO5D in the present study. The

FMO4-like gene is called FMO4 hence forth. FMO4 contained the least percent identity to the FMO5 enzymes, with approximately 27-29% identity to all (Table 3.2). In contrast, the FMO5-like enzymes retained approximately 50-55% identity to each other. Protein alignments suggested high conservation between the FMOs, with good alignment scores of 88-92 as determined by TCS; however, proteins were less conserved near the C- terminus (Figure 3.S1). All FMOs contained the FMO identifying sequence

(FXGXXXHXXXY), the FAD –binding motif, and the NADPH-binding motif (Cashman

& Zhang, 2006) (Figure 3.S1).

145 Table 3.2. Percent Identity Matrix of medaka FMOs as calculated by Clustal Omega.

FMO4 FMO5A FMO5B FMO5C FMO5D FMO4 100 28.39 26.6 28.9 29.67 FMO5A 100 54.6 54.9 50.92 FMO5B 100 52.95 49.81 FMO5C 100 52.91 FMO5D 100

Developmental Expression of FMOs

qPCR confirmed that mRNAs of all medaka FMOs were expressed. Transcripts of putative FMOs were measured on days 1, 3 6, and 9 post fertilization and data is presented relative to 1dpf. Each FMO exhibited a distinct expression pattern. FMO4 expression significantly increased 3-fold at 6dpf in relation to 1dpf, 3dpf and 9dpf (p =

0.005) (Figure 3.1A). No change in FMO5A was observed (Figure 3.1B). Expression of

FMO5B significantly decreased to 0.18 of 1dpf, remaining constant at 3dpf, 6dpf, and

9dpf (p < 0.001) (Figure 3.1C). FMO5C transcripts also decreased to approximately 0.37 of 1dpf at 3dpf (p < 0.001), then increased to 0.60 at 6dpf (p = 0.01), then returned to

0.43 at 9dpf (p < 0.001) (Figure 3.1D). Finally, FMO5D followed a similar pattern to

FMO4, increasing 3.7-fold over controls at 6dpf (p = 0.01), then decreasing to 2 fold greater than 1dpf at 9dpf (p = 0.35) (Figure 3.1E).

146

Figure 3.1. Developmental expression of putative medaka FMOs on 1dpf, 3dpf, 6dpf and 9dpf expressed as fold change against 1dpf ± standard error (SE). A. FMO4, B. FMO5A, C. FMO5B, D. FMO5C, E. FMO5D. Differing letters represent significant differences between stages following a Tukey’s HSD test at p ≤ 0.05 (n = 3-5).

Promoter Analysis of Medaka FMOs

Considering the variable expression patterns of medaka FMOs during development, we were further interested in examining their regulation by developmental programming and exogenous factors. Proximal promoter regions were analyzed for transcription factor binding sites (TFBS) using Consite and Alibaba2.1. A total of 74

TFBS were identified by the two programs (Table 3.S1). UniprotKB was used to determine the function of the TFBS identified. Of the 74 TFBS, 63 were involved in embryonic development (8 in liver development), 25 in apoptosis or cell death, 22 in immunity and inflammation, 12 in cellular responses to hypoxia, and 7 in oxidative stress or redox cycling (Table 3.S1).

147 Interestingly, FMO5B and FMO5D were found to contain TFBS for CHOP-EBP

(C/EBP homologous protein) (Table 3.3), which is an important apoptotic mediator for the Unfolded Protein Response (UPR). Furthermore, both promoter regions contained

TFBS for the ATF (Activating transcription factor) family of proteins, of which ATF4 is the primary effector of the PERK (PKR-like endoplasmic reticulum kinase) branch of the

UPR and activator of CHOP. All FMOs contained regions for C/EBPβ, which is upregulated during the UPR and may be involved in UPR induced apoptosis (Chen et al.,

2004; Meir et al., 2010). The UPR is a known perpetrator of many pathological diseases, can be activated by xenobiotics, and plays a key role in many developmental pathways

(Kupsco & Schlenk, 2015). Other TFBS found in the FMO promoters have shown to be related to the UPR, including oxidative stress, hypoxia and calcium signaling (Kupsco &

Schlenk, 2015). These results prompted us to perform a manual search for the UPR response element (TGACGTG(G/A) or ER Stress response element (CCAAT-N9-

CCACG) in the FMO promoter regions; however, none were identified. Still, manual searches for ATF4 further identified an ATF4 binding sites off by one base pair

((A/G/C)TTT(T/G/A)C(G/A)TCA; Bouman et al., 2011) in the promoter regions of

FMO4 and FMO5C (Table 3.3), suggesting that medaka FMOs may be regulated by the

UPR.

148 Table 3.3. Number of transcription factor binding sites of ATF (Alibaba2.1), ATF4 (manual search), Chop-EBP (Consite), and C/EBPβ (Alibaba2.1) per medaka FMO promoter region. ATF ATF4a CHOP-EBP C/EBPβ FMO4 - 1 - 2 FMO5A - - - 1 FMO5B 1 - 2 1 FMO5C - 2 - 1 FMO5D 1 1 1 2 aFrom Bouman et al., 2011

Developmental Expression of the UPR

These findings led us to investigate the developmental expression patterns of

CHOP and a major UPR regulator, BiP (GRP78). Although the UPR is conserved in medaka (Ishikawa et al., 2011), to date no information on expression patterns is available.

CHOP mRNA levels remained constant between 1dpf and 3dpf, then increased significantly 6-fold over 1dpf at 6dpf and 62-fold at 9dpf (p < 0.001) (Figure 3.2A). BiP expression increased significantly to 24-fold at 9dpf (p < 0.001) (Figure 3.2B).

Figure 3.2. Developmental expression of A. CHOP and B. BiP against 1dpf embryos as fold change ± SE. Differing letters represent significant differences between stages following a Tukey’s HSD test at p ≤ 0.05 (n = 4).

149 Regulation of FMOs by the UPR

We hypothesized that liver functionalization would occur during late organogenesis occurring at around 6dpf, and thus chose to further investigate FMO regulation during that time period. Tm is a well-studied inducer of the UPR that acts via inhibition of N-glycosylation (Takatsuki et al., 1971). BiP and CHOP expression were analyzed as a positive control to demonstrate induction of the UPR. Both are highly inducible and thus a good marker for induction of the UPR. Tm robustly induced CHOP and BiP to 204 and 71-fold, respectively, over controls after treatment with 4µg/ml Tm (p

< 0.001) (Figure 3.3A). FMO4 was significantly induced to 1.6-fold over controls by

4µg/ml Tm (p = 0.001) (Figure 3.3B). FMO5A was also significantly induced to over

2.5-fold following 4µg/ml Tm treatments (p = 0.003). Expression levels of FMO5B, 5C and 5D remained constant (Figure 3.3B).

Figure 3.3. A. Expression of UPR targets BiP (black bars) and CHOP (white bars) in 6dpf medaka embryos following treatment with 0.02% DMSO, 1µg/ml Tm, 2µg/ml Tm, and 4µg/ml Tm for 24hrs. B. Expression of medaka FMOs in 6dpf embryos following treatment with 0.02% DMSO (black bars), 1µg/ml Tm (white bars), 2µg/ml Tm (gray bars), and 4µg/ml Tm (striped bars) for 24hrs. Values expressed as fold change ± SE. Asterisks indicate significant difference from controls following a Dunnett’s test (p ≤ 0.05, n = 7-14).

150

As a variety of compounds are able to induce the UPR via different disruptions in protein folding and processing, DTT was also used in this study. DTT reduces disulfide bonds which subsequently induces the UPR. DTT significantly reduced CHOP expression to approximately 0.6 of controls following 2mM and 4mM treatments (p =

0.02) (Figure 3.4A). Although a similar trend was observed with BiP expression, no significant difference was present (p = 0.2). In accordance with the trend in BiP and

CHOP expression, DTT significantly decreased expression of FMO5A, FMO5B and

FMO5C (Figure 3.4B). FMO5A expression decreased to 0.56 fold after 2mM DTT (p =

0.04) and 0.33 fold after 4mM DTT (p = 0.01). FMO5B expression decreased to 0.58 fold after 2mM (p = 0.04) and 0.38 fold after 4mM DTT (p = 0.003). Finally, FMO5C decreased to 0.38 fold following 2mM DTT (p = 0.02) and 0.25 fold following 4mM

DTT (p = 0.01). No change was detected in FMO4 or FMO5D expression.

Figure 3.4. A. Expression of UPR targets BiP (black bars) and CHOP (white bars) in 6dpf medaka embryos following treatment with Freshwater controls, 2mM DTT and 4mM DTT for 24hrs. B. Expression of medaka FMOs in 6dpf embryos following treatment with freshwater (black bars), 2mM DTT (white bars), and 4mM DTT (striped bars) for 24hrs. Values expressed as fold change ± SE. Asterisks indicate significant difference from controls following a Dunnett’s test (p ≤ 0.05, n = 7-14).

151 Discussion

In the present study, we demonstrated that Japanese medaka have at least 5 putative FMO genes, which we designated as FMO4, FMO5A, FMO5B, FMO5C, and

FMO5D. Among mammals, different FMO isoforms have between 50% and 58% identity, while orthologs have around 80% identity (Lawton et al., 1994). The medaka

FMO5 enzymes exhibit approximately 50% identity, which would be considered distinct isoforms under the current nomenclature. In contrast, the FMO4 isoform has only 30% identity to other medaka FMOs, suggesting that it may have evolved differently.

This work is the first to examine developmental expression of fish FMOs, important information for studies on drug metabolism in these model organisms. Medaka

FMOs exhibited distinct developmental expression patterns. FMO4 and FMO5D increased up to 6dpf, then decreased, while FMO5B and FMO5C decreased. These expression patterns are similar to those found in humans. For example, FMO1 expression in the human liver is highest during the first trimester and steadily declines until it is undetectable at 1year old (Hines, 2006). FMO3 expression in the liver then begins during the neonate stage, then increases up to 11 years of age (Hines, 2006). In contrast, other mammals such as rabbits, rats and mice, express FMO1 in the adult liver (Janmohamed et al., 2004).

Although no tissue-specific expression could be measured here, the stages analyzed were chosen to represent different periods throughout development: at 1 dpf embryos have completed neurulation and are beginning somitogenesis; at 3dpf the embryo is in the early stages of organogenesis; at 6dpf the embryo is in organogenesis

152 and the heart develops; and at 9dpf the embryo is in the late stage of organogenesis as the spleen develops. In particular, these stages were chosen based upon liver development, as the liver is the primary organ responsible for xenobiotic metabolism, and FMO levels are highest in the liver in vertebrates (Zhang & Cashman, 2006). In medaka, liver development begins at stage 25 (2dpf) when the hepatic bud forms from the endoderm

(Watanabe et al., 2004). Genes essential for liver development, foxa3 (Also called HNF-3 gamma) and Gata6 are expressed in the liver anlagae beginning at this stage and expression increases until stage 36 (6dpf) (Watanabe et al., 2004). Although foxa3 expression continues to increase, Gata6 is no longer found in the liver at this stage

(Watanabe et al., 2004), suggesting that proliferation and growth of hepatocytes may begin between stages 34 (5dpf) and 36 (6dpf).

Many of the developmental transcription factors identified during promoter analysis were associated with liver development and organogenesis. For instance, Sox17 is an endoderm-specific transcription factor, the germ layer that will eventually form the liver (Zorn et al., 2007). The Sox17 TFBS was present in all medaka FMOs, with 3 present in each of the FMO5B and FMO5D promoter regions. Additionally, response elements for hepatic nuclear factors HNF-1 and HNF-3 were also found in promoter regions of all medaka FMOs, and FMO5C and FMO5D were enriched for 7 different binding sites. HNF1 and HNF3 have high expression during liver development, and expression is restricted to hepatocytes in adult human livers (Nagy et al., 1994). C/EBPα and C/EBPβ have also been implicated in liver development and in expression of important enzymes in the adult liver, such as albumin and HNF4 (Westmacott et al.,

153 2006; Bossard et al., 1997). Since we hypothesized that functionalization of the liver occurred during 5dpf-6dpf and several UPR-associated TFBS were identified in several

FMO promoters, we chose to investigate these stages for UPR and FMO gene expression.

The UPR is conserved in medaka (Ishikawa et al., 2011); however, little is known about its developmental patterns of expression. BiP is the main chaperone mediating activity of the UPR (Hetz, 2012). Under physiological conditions BiP is bound to the three branches of the UPR (IRE1 (isositol requiring enzyme 1), PERK (PKR-like endoplasmic reticulum kinase) and ATF6 (activating transcription factor 6)), inhibiting their activation. However, if unfolded proteins are sensed in the ER, BiP will dissociate from the three branches and allow them to become active (Hetz, 2012). CHOP is a downstream target of the PERK branch of the UPR and has been shown to promote apoptosis following chronic ER stress (Hetz, 2012). CHOP and BiP expression were found to increase significantly throughout development, beginning at 6dpf. This is unsurprising as the UPR plays an important physiological role in many developmental pathways (Kupsco & Schlenk, 2015). For instance, both the IRE1 and ATF6 branches of the UPR are involved in hepatogenesis. Mice lacking key components of the IRE1 pathway exhibit liver hypoplasia, decreased proliferation of hepatocytes and increased apoptosis of hepatocytes (Zhang et al., 2005; Reimold et al., 2000). ATF6 knockout mice suffer from liver steatosis (Yamamoto et al., 2007). Interestingly, FMO knockouts also suffer from impaired lipid metabolism (Hernandez et al., 2009). The crucial role of the

UPR in liver development coupled with developmental regulation of FMOs suggests that

FMOs may be regulated by the UPR during development.

154 In the current study, Tm and DTT both modified medaka FMO expression. Tm is a potent inducer of ER stress, which has been shown to induce the UPR in medaka

(Ishikawa et al., 2011). Tm significantly induced CHOP and BiP expression in 6dpf medaka embryos. High concentrations of Tm also induced expression of FMO4 and

FMO5A. FMO4 was found to contain a possible ATF4 element, which is an upstream regulator of CHOP and an integral part of the UPR. However, FMO5A did not have a

TFBS for CHOP or ATF4. It is possible that a nearby enhancer region may be acting during ER stress to activate FMO5A; however, this requires further investigation.

C/EBPβ is involved in ER stress induced apoptosis through the ATF4/PERK branch

(Meir et al., 2010). C/EBPβ was present in all medaka FMOs, but it is unclear why it would only act as an activator for FMO5A and FMO4. Similarly, although FMO5B,

FMO5C and FMO5D contained TFBS for either CHOP or ATF4, none responded to Tm.

It has been suggested that many motifs present in proximal promoters may not actually be bound in vivo (Hardison & Taylor, 2012). It is for these reasons that we tested the functionality of these regions was tested in vivo in the current study. Treatments of medaka embryos with Tm suggest that only two FMOs may be regulated by the UPR, although promoter analysis would suggest otherwise.

In contrast to Tm, DTT significantly repressed the UPR in medaka embryos after

24hrs. DTT is a classical-inducer of the UPR and has been shown to induce BiP and

CHOP in other organisms (DuRose et al., 2006). In medaka, DTT strongly inhibited translation in a medaka OLCAB-e3 cell line (Ishikawa et al., 2011), but when used in vivo, 2mM DTT was lethal to medaka embryos treated at 2dpf and was deemed not

155 suitable to induce ER stress (Ishikawa et al., 2013). Although DTT is able to induce the

UPR, different types of ER stress produce different kinetics in UPR activation (DuRose et al., 2006). For example, in a model of cerebellar granule neuron (CGN) ER stress,

DTT was used to induce a “fast motion” response, while Tm was used to induce a “slow motion” response (Li et al., 2011). DTT increased CHOP expression and IRE1 phosphorylation within the first 3 hours of treatment, but levels creased significantly after

6 and 10hrs back to controls (Li et al., 2011). In contrast, Tm induction of CHOP and

IRE1 phosphorylation remained elevated throughout (Li et al., 2011). Thus, it is possible that the 24hr treatments performed in this study were too long to capture the height of the

UPR from DTT, and that the response was repressed at analysis.

Additionally, DTT is a powerful reductant and is thought to induce ER stress via inhibition of disulfide bond formation. As a monooxygenase of many sulfur-containing compounds, FMOs have been found oxidize endogenous substrates such as methionine and cysteine conjugates (Elfarra, 1995; Ripp et al., 1997), suggesting they play a role in metabolism of endogenous sulfur containing compounds. Thus, it’s possible that the addition of a reducing agent may impact FMO expression independently of the UPR. In the present study, both concentrations of DTT significantly reduced expression of

FMO5A, FMO5B and FMO5C. FMO4 remained unchanged by DTT, even though it was significantly induced by Tm. Furthermore, FMO5B and FMO5C were unaffected by Tm, but reduced by DTT. These results suggest that regulation of FMOs under reducing stress may be either modulated by a different branch of the UPR or or have an alternative regulatory pathway.

156 In conclusion, medaka FMOs exhibit developmentally specific patterns of expression and are variably induced by the UPR or inhibited by UPR inhibition or reductive stress. Further research is needed to fully understand developmental regulation of UPRs in fish. However, a growing role for the UPR has been established in development, disease, and xenobiotic toxicity highlighting the importance of these findings. Developmental expression patterns of FMOs and modulation by stress response pathways are significant for toxicological research using fish as a model organism for either embryonic development or environmental factors.

157 References

Bossard, P., McPherson, C.E., Zaret, K.S., 1997. In Vivo Footprinting with Limiting Amounts of Embryo Tissues: A Role for C/EBPβ in Early Hepatic Development. Methods

Bouman L, Schlierf A, Lutz AK, Shan J, Deinlein A, Kast J, Galehdar Z, Palmisano V,Pa tenge N, Berg D, Gasser T, Augustin R, Trumbach D, Irrcher I, Park DS, Wurst W,Kilber g MS, Tatzelt J, Winklhofer KF (2011) Parkin is transcriptionally regulated by ATF4: evidence for an interconnection between mitochondrial stress and ER stress. Cell Death Differ. 18: 769–782

Cashman, J.R., Zhang, J., 2006. Human Flavin-Containing Monooxygenases. Annu. Rev. Pharamol. Toxicol. 46, 65-100.

Chang, J.-M., Di Tommaso, P., Notredame, C., 2014. TCS: A New Multiple Sequence Alignment Reliability Measure to Estimate Alignment Accuracy and Improve Phylogenetic Tree Reconstruction. Mol. Biol. Evol. 31, 1625-1637.

Chen, C., Dudenhausen, E.E., Pan, Y.-X., Zhong, C., Kilberg, M.S., 2004. Human CCAAT/Enhancer-binding Protein Gene Expression Is Activated by Endoplasmic Reticulum Stress through an Unfolded Protein Response Element Downstream of the Protein Coding Sequence. J. Biol.Chem. 27, 27948–27956.

El-Alfy, A., Schlenk, D., 2002. Effect of 17β-estradiol and testosterone on the toxicity of aldicarb to Japanese medaka, Oryzias latipes. Toxicol. Sci. 68, 381, 2002.

Elfarra, A., 1995. Potential role of the flavin-containing monooxygenases in the metabolism of endogenous compounds. Chemico-Biological Interactions 96, 47-55.

158 Hao, D.C., Chen, S.L., Mu, J., Xiao, P.G. 2009. Molecular phylogeny, long-term evolution, and functional divergence of flavin-containing monooxygenases. Genetica, 137, 173-187.

Hardison, R.C., Taylor, J., 2012. Genomic approaches towards finding cis-regulatory modules in animals. Nature Rev. Genetics 13, 469-483.

Hetz, C., 2012. The unfolded protein response: controlling cell fate decisions under ER stress and beyond. Mol. Cell. Biol. 13, 89-102.

Hines, R.N., 2006. Developmental and tissue-specific expression of human flavin- containing monooxygenases 1 and 3. Expert Opin. Drug Metab. Toxicol. 2, 41-49.

Ishikawa, T.; Taniguchi, Y.; Okada, T.; Takeda, S.; Mori, K. Vertebrate unfolded protein response: Mammalian signaling pathways are conserved in Medaka fish. Cell. Struct. Funct. 2011, 36 (2), 247−59.

Janmohamed, A., Hernandez, D., Phillips, I.R., Shephard, E.A., 2004. Cell-, tissue-, sex- and developmental stage-specific expression of mouse flavin-containing monooxygenase (FMO’s). Biochem. Pharmacol. 68, 73–83

Lawton M, Cashman J, Cresteil T, Dolphin C, Elfarra A, et al. 1994. A nomenclature for the mammalian flavin-containing monooxygenase gene family based on amino acid sequence identities. Arch. Biochem. Biophys. 308:254–57

Lenhard, B. et al. Identification of conserved regulatory elements by comparative genome analysis. J. Biol. 2, 13 (2003).

Matys, V. et al. TRANSFAC: transcriptional regulation, from patterns to profiles. Nucleic Acids Res. 31, 374–378 (2003).

159 Meir, O.M., Dvash, E., Werman, A., Rubinstein, M. 2010. C/EBP-b Regulates Endoplasmic Reticulum Stress–Triggered Cell Death in Mouse and Human Models. PLoS One 5, e9516.

Nagy, P., Bisgaard, H.C., Thorgeirsson, S.S., 1994. Expression of Hepatic Transcription Factors during Liver Development and Oval Cell Differentiation. J. Cell Biol. 126, 223- 233.

R Core Team., 2012. R: A Language and Environment for Statistical Computing. R Foundation for Statistical Computing, Vienna, Austria.

Ripp, S.L., Overby, L.H., Philpot, R.M., Elfarra, A., 1997. Oxidation of Cysteine S- Conjugates by Rabbit Liver Microsomes and cDNA-Expressed Flavin-Containing Mono- oxygenases: Studies with S-(1,2-Dichlorovinyl)-L-cysteine, S- (1,2,2-Trichlorovinyl)-L- cysteine, S-Allyl-L-cysteine, and S- Benzyl-L-cysteine. Mol. Pharmacol. 51, 507-515.

Sandelin, A., Alkema, W., Engstrom, P., Wasserman, W. W. & Lenhard, B. JASPAR: an open-access database for eukaryotic transcription factor binding profiles. Nucleic Acids Res. 32, D91–D94 (2004).

Schlenk, D., 1998. Occurrence of flavin-containing monooxygenases in non-mammalian eukaryotic organisms, Comp. Biochem. Physiol., 121C, 185.

Schmittgen, T.D., Livak, K.J., 2008. Analyzing real-time PCR data by the comparative C(T) method. Nat. Protoc. 3, 1101−8.

Sievers, F. et al. Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega. Mol. Syst. Biol. 7, 539 (2011)

Takatsuki, A., Arima, K., Tamura, G., 1971. Tunicamycin, a New Antibiotic. I. Isolation and characterization of Tunicamycin. J. Antibiotics 24, 215-223.

160 Tijet, N., Boutros, P.C., Moffat, I.D., Okey, A.B., Tuomisto, J., Pohjanvirta, R., 2006. The aryl hydrocarbon receptor regulates distinct dioxin-dependent and dioxin- independent gene batteries. Mol. Pharmacol. 69, 140–153.

Vandesompele, J., De Preter, K., Pattyn, F., Poppe, B., Van Roy, N., De Paepe A., Speleman, F., 2002. Accurate normalization of real-time quantitative RT-PCR data by geometric averaging of multiple internal control genes. Genome BIol. 3, research0034.1– 0034.11.

Watanabe T, Asaka S, Kitagawa D, Saito K, Kurashige R, Sasado T, et al. Mutations affecting liver development and function in Medaka, Oryzias latipes, screened by multiple criteria. Mech Dev 2004; 121: 791–802.

Westmacott, A., Burke, Z.D., Oliver, G., Slack, J.M., Tosh, D., 2006. C/EBPα and C/EBPβ are markers of early liver development. Int. J. Dev. Biol. 50, 653-657.

Zhang, K., Cashman, J.R., 2006. Quantitative Analysis of FMO Gene mRNA Levels in Human Tissues. Drug Metab. Dispos. 34, 19-26.

Zhao, S., Fernald, R.D., 2005. Comprehensive algorithm for quantitative real-time polymerase chain reaction. J. Comput. Biol. 12, 1047−64.

Zorn A.M, Wells J.M., 2007. Molecular basis of vertebrate endoderm development. Int. Rev. Cytol. 259, 49–111

161 Supplemental Information

Figure 3.S1. Protein Alignments of medaka FMOs by Clustal Omega. Validity of alignments was assessed by TCS. FMO motifs are identified by black boxes.

162 Table 3.S1. The number of transcription factor binding sites identified in the proximal promoter regions of medaka FMOs by Consite and Alibaba2.1. Full names and functions are summarized from UniprotKB.

163

164

165

166

167

168

169

170

171 Chapter 4: Stage susceptibility of Japanese medaka (Oryzias latipes) to selenomethionine and hypersaline developmental toxicity

Abstract

Anthropogenic disturbance of seleniferous soils can lead to selenium contamination of waterways. Although selenium is an essential micronutrient, bioaccumulation and maternal transfer of proteinaceous selenomethionine (SeMet) can result in embryotoxicity. Furthermore, as the climate changes, the salinity of spawning grounds in water-restrained estuaries is increasing. While a small increase in salinity may not directly impact adult fish, it may alter the detoxification strategies of developing organisms. Previous research indicates that hypersalinity may potentiate SeMet embryo toxicity at an early developmental stage. However, embryonic development is a complex, spatiotemporal process with a constantly shifting cellular microenvironment. In order to generate thresholds and an adverse outcome pathway for the interactions between selenium and salinity, we sought to identify windows of susceptibility for lethality and deformities in the Japanese medaka (Oryzias latipes). Embryos were treated in freshwater or saltwater for 24hours with 0.5µM, 5µM, and 50µM SeMet at 6 different developmental stages (9, 17, 25, 29, 34, and 38). Survival, hatch, deformities (total, type and severity), and day to hatch were quantified. Se embryo tissue measurements were performed. SeMet exposures of 5µM and 50µM significantly decreased survival and hatch at all stages. However, SeMet uptake was stage dependent and increased with stage. Stage 17 (early neurulation) was identified as the most susceptible stage for lethality and deformities. SeMet in saltwater caused significantly greater toxicty than

172 freshwater at stage 25 (early organogenesis), suggesting a role for liver and osmoregulatory organogenesis in toxicity.

Introduction

Selenium (Se) is an essential micronutrient that can be toxic to oviparous animals in excess (Lemly, 1997). Present naturally in soils, Se is released into waterways following anthropogenic activity, such as irrigation of arid soils; coal, phosphate and uranium mining; and coal burning power plants (Outridge et al., 1999; Muscatello et al.,

2006; Presser et al., 2004; Ramirez & Rogers, 2002). Free Se is often present in water in its inorganic forms of selenate and selenite, and can be taken up by microorganisms and converted into various organic forms, including selenomethionine (SeMet) (Fan et al.,

2002). Dietary uptake has been shown to be the primary route of exposure in fish (Phibbs et al., 2011), and the high bioaccumulation factor of Se results in increased exposure to upper trophic levels. Subsequent maternal offloading to eggs can cause embryonic lethality and teratogenesis (Lemly, 2002). The teratogenic effects of Se have been well characterized in fish, and can include spinal, fin and cranio-facial deformities; as well as alterations in organ physiology (Lemly, 1997).

Selenium’s high bioaccumulation factor confounds water quality monitoring efforts and many have criticized free Se water measurements in favor of ovary/egg measurements (Lemly, 1993a). While these tissue quality measurements are an improvement over water measurements in the estimation of Se risk, site-specific testing for Se may be necessary, particularly in multiple stressor situations (Ohlendorf et al.,

173 2011). One common stressor is hypersalinity. As climate change increases salinization of many traditionally fresh waterways (Wong et al., 2014), effects of the interaction between SeMet and hypersalinity need to be further elucidated. This issue is particularly relevant in areas such as the San Francisco Bay Delta, where both salinity and Se have been cited as major contributors to declining fish populations (Delta Science Council,

2013). In the Delta, rising sea levels leading to increased saltwater intrusion combine with decreased freshwater input to increase salinity (Delta Science Council, 2013).

Furthermore, desalination is under consideration to meet California’s growing drinking water needs, and disposal of desalination brine back into the estuary will further increase salinization. Recent research indicates that hypersaline conditions may potentiate SeMet developmental toxicity; embryos treated with SeMet in saltwater had significantly decreased hatch and increased deformities compared to those treated in freshwater

(Lavado et al., 2012; Kupsco & Schlenk, 2014].

Although toxicity and teratogenicity of Se have been well characterized in the field (Lemly, 1997; Lemly, 2002; Phibbs et al., 2011), there have been no studies examining the stage specific susceptibility of fish to Se. Embryonic development is a complex, spatiotemporal process with constant alterations in the molecular and cellular microenvironment. These changes can modify the mechanism and overall toxicity of a toxicant to the embryo. The identification of critical exposure windows to Se can produce lethal and sublethal thresholds for toxicity, as well as provide additional insight into the mechanism of toxicity for adverse outcome pathway (AOP) development. AOPs connect a molecular initiating event with a population level outcome via different levels of

174 biological organization (Ankley et al., 2010). Identification of the critical stage of Se toxicity allows for further elucidation of an AOP for compounds similar to Se, and provides a starting point for future studies on mechanisms of SeMet toxicity.

Furthermore, this approach is ideal for studying effects of multiple stressors in development because it allows for isolation of the height of the interaction between Se and hypersalinity.

In the present study, 3 concentrations of waterborne SeMet in freshwater or saltwater were used to model of SeMet exposure to Japanese medaka (Oryzias latipes) embryos. Embryos were exposed to SeMet at various stages throughout development.

The aim was to determine sensitive developmental stages to SeMet and hypersaline toxicity, and to identify thresholds for lethal and sublethal toxicity of SeMet under both freshwater and saltwater conditions using tissue Se values. We hypothesized that earlier developmental stages would be more susceptible to SeMet toxicity than later stages, and that hypersalinity would increase SeMet toxicity in a stage specific manner. These findings allow for further elucidation of the mechanism of SeMet embryo toxicity alone and under hypersaline conditions.

Methods

Animals

Japanese medaka (Oryzias latipes) were used for this study because they are a good euryhaline model organism, frequently used in toxicity testing, and have well documented embryonic development. Medaka develop over 10-14 days and hatch as

175 feeding larvae, allowing for isolation of a specific developmental period, and little structural and functional development occurs post hatch. Furthermore, medaka have been previously shown to be sensitive to SeMet and hypersaline toxicity (Lavado et al., 2012;

Kupsco & Schlenk, 2014).

Japanese medaka were cultured at the University of California- Riverside in medium-hard water at 27°C and were housed in a 2:3 ratio of males to females. They were maintained in a 14 hour light, 10 hour dark cycle and fed twice daily a diet of live brine shrimp. Embryos were collected 0−1 h following fertilization and nonviable embryos were discarded. Oil droplet migration to the vegetal pole was used to determine viable embryos (Kirchen & West, 1976).

Overall experimental design

The effects of 3 factors on embryonic development were examined: water type, developmental stage of treatment, and SeMet concentration. In order to assess these effects, fertilized embryos were placed into dishes containing either freshwater or saltwater. They were monitored throughout development for one of the following stages:

9, 17, 25, 29, 34, or 38. Once the embryos had reached the appropriate stage, they were treated with 0µM, 0.5µM, 5µM, or 50µM SeMet for 24 h. Following exposure, embryos were removed from the SeMet treatment, rinsed, then replaced into freshwater or saltwater and allowed to develop until hatch. Survival post treatment, hatch, deformities and day to hatch were recorded.

Water treatment

To assess the effects of water type, embryos were placed into 60x15mm petri

176 dishes containing either medium-hard fresh water or artificial saltwater from the San

Joaquin River that had been prepared in the lab (approximately 10-20 embryos per replicate, 4-6 replicates per treatment). Salts were purchased from Fisher Scientific and deionized water was obtained via a Milli-Q water purification system (Millipore).

Saltwater was prepared based on values collected from Westlands Water District, which is located approximately 10 km south of Mendota in the San Joaquin River Drainage

Basin, CA (Schlenk et al., 2003). Salinity was 13‰ with 15.21g/L suspended solids.

Although the saltwater used was not embryotoxic, it caused specific tissue death and larval mortality immediately post hatch in 45% percent of replicates, and 60 ± 7.5% of larvae were deformed. The replicates exhibiting these features were discarded and not considered in the analysis.

Developmental stage of treatment

Embryos were treated at one of six stages, 9, 17, 25, 29, 34 or 38, based on

Iwamatsu (2004). Stages were spaced regularly throughout development and were chosen to represent key developmental milestones that may be susceptible to selenium and hypersaline toxicity (Table 4.1). Stage 9 (5 hours post fertilization (hpf)), the late morula stage, represents early exposure window. Stage 17 (25hpf) is the start of neurulation, and was considered based on the well-documented impacts of Se on spinal development

(Lemly, 1997). The liver is the primary site of xenobiotic metabolism. Thus, to examine a role for the liver in SeMet and hypersaline toxicity, stage 25 (50hpf) was examined. Stage

29 (74hpf) is when the craniofacial cartilage begins to develop (Langille & Hall, 1987) and SeMet has been implicated in craniofacial abnormalities (Lemly, 1997). Bone

177 mineralization in Medaka occurs at stage 34 (121hpf) (Yasutake et al., 2004). Finally, stage 38 (192hpf) occurs just before hatching and is late organogenesis, when full SeMet activation is most likely to occur (Iwamatsu, 2004).

Table 4.1. Developmental stages of medaka development chosen for 24 h of SeMet treatment. Stage Hpf Developmental Milestone 9 5 Late morula 17 25 Early neurulation 25 50 Liver anlagae 29 74 Cranio-facial cartilage 34 121 Bone mineralization 38 192 Late organogenesis

Selenomethionine treatment

Once embryos reached the appropriate stage, they were treated with SeMet in freshwater or saltwater (Seleno-L-methionine (purity 98%) from Sigma-Aldrich).

Although the primary mechanism of environmental embryo exposure is via maternal transfer (Fan et al., 2002) and thus waterborne exposures do not entirely reflect environmental conditions, they are a simplified method of determining SeMet’s effects.

Furthermore, free SeMet is the main cause of Se toxicity (Fan et al., 2002), and precise exposures can be determined with tissue level analytical chemistry. Although the chorion can function as a barrier against waterborne toxicants, it is water permeable and exhibits varying permeability to organic osmolytes (Alderice, 1988). Hence we expect SeMet treatments with the chorion to be valid. Embryos were treated with 0.5µM, 5µM, and

50µM SeMet, which correspond to approximately 0.01mg/L, 0.1mg/L, and 1mg/L, respectively. Based on previous work, it is estimated that 50µM SeMet treatment for 24 h would be equivalent to the maximum Se tissue concentrations observed in the field

178 following Se contamination (Lavado et al., 2012). The 5µM and 0.5µM SeMet treatments would be indicative of a less impacted environment. Following 24 h of treatment, embryos were replaced into either freshwater or saltwater and allowed to develop until hatch. Water was exchanged every other day. At 24 h post treatment, survival was assessed and dead embryos were discarded. Embryos were checked daily for hatch and mortality. Dead embryos were removed. Deformities at hatch were recorded for type

(cardiac, spinal, craniofacial, fin or swim bladder). Spinal deformities were categorized as lordosis, scoliosis and kyphosis. Severity of each deformity was recorded on a scale from 1-3. Embryos that had not hatched after 1 month were considered dead.

Tissue analysis of selenium

Exposures outlined in the Overall Experimental Design section were repeated.

Instead of allowing embryonic development to continue post treatment, live embryos were frozen immediately following the 24 h treatment for analysis with the chorion intact. As dead embryos may lack the capacity to regulate SeMet uptake, and would thus provide inaccurate measurements, they were not included in the analysis. Not all treatment groups yielded enough live embryos for analysis, so only samples of the lowest concentration demonstrating significance in the survival endpoint were collected.

Samples of 40-50 embryos were lyophilized and digested in 50% nitric acid at 90◦C for 5 h. Following digestion, samples were diluted to a final nitric acid content of 10% and Se content was analyzed on an Agilent 7000 inductively coupled plasma-mass spectrometer

(ICPMS) with a detection limit of 0.025µg/g. Oyster tissue was used for quality control

(National Institute of Standards and Technology) with a percent recovery of 98 ± 2.7%.

179 Statistical analysis

In order to determine the statistical significance of each factor (water type, SeMet concentration, and developmental stage) as well as differences between the levels within the factors, 3 different statistical methods were used. Binomial data (nominal data with 2 outcomes; survival, hatch, total deformities, and swim bladder inflation) were analyzed using logistic regression and pairwise significance was assessed with multiple comparisons. Multinomial data (data with more than 2 nominal outcomes) were assessed with multinomial logistic regressions and multiple comparisons were performed between the levels of each factor. Multinomial logistic regressions were performed on data reporting the type of deformity and type of spinal deformity; spinal deformities and lordosis were used as reference deformities, respectively. Continuous data (day to hatch, average severity of spinal deformity, and SeMet tissue concentration) were checked for normality and homoscedasticity, and log transformed if they violated these assumptions.

3-way ANOVA and the Tukey HSD test were used to assess significance. The relationship between Se tissue content and percent deformities was evaluated using linear regression analysis and ANCOVA. P values of ≤ 0.05 were considered statistically significant. All calculations were performed in the statistical program R64 (R Core Team,

2012).

180 Results

Survival

Logistic regression analysis indicated that SeMet exposures of 5µM and 50µM significantly decreased embryo survival 24 h post treatment at stages 17-38, while no differences between controls and 0.5µM SeMet were observed (Figure 4.1). When embryos were exposed at stages 17, 25 and 29 survival decreased significantly between

5µM SeMet and 50µM SeMet (p = 0.02, 0.023 and 0.026, respectively; Figure 4.1B, C and D). At stage 9, only 50µM SeMet significantly decreased survival (p = 0.005; Figure

4.1A). Significant differences were also observed between stages. At 5µM and 50µM

SeMet, embryos treated at stage 9 had significantly greater survival than all other stages

(p < 0.001). Saltwater also significantly impacted survival following exposure with 50µM

SeMet (p = 0.011), and the largest difference was observed at stage 25, where survival differed between freshwater (42%) and saltwater (2%) (Figure 4.1C).

181

Figure 4.1. Percent survival ± standard error (SE) of embryos 24 h following 24 h of treatment with 0.5µM, 5µM, and 50µM SeMet in either freshwater (black bars) or saltwater (white bars) (n = 3-6). Each graph represents one stage of treatment: A. stage 9; B. stage 17; C. stage 25; D. stage 29; E. stage 34; and F. stage 38. Statistical significance was assessed using a logistic regression followed by multiple comparisons and p ≤ 0.05 was considered statistically significant. Differing letters represent significant differences between concentrations at a specific stage and forks denote differences between fresh and saltwater at a specific dose. Symbols represent significant differences between stages (different graphs) at specific doses; e.g. the percent survival at 5µM at stage 9 with the hash tag (A) is significantly different than the percent survival at 5µM with the at-sign in stages 17-38 (B-F).

Hatch

Percent hatch of embryos decreased with increasing concentrations of SeMet

(Figure 4.2). When embryos were treated at stage 17 for 24 h with 0.5µM SeMet, hatch decreased significantly from controls (logistic regression and multiple comparisons: p =

0.039) (Figure 4.2B). Treatments with 5µM SeMet at all other stages decreased hatch

182 significantly (p < 0.02 for all). Significant differences between freshwater and saltwater exposures were observed after 5µM and 50µM SeMet (p = 0.019 and 0.001, respectively). Consistent with percent survival, the largest differences in embryo hatch between fresh and salt water were recorded at stage 25 (40-50% difference in hatch;

Figure 4.2C). Stage specific differences in hatch were also observed. After 5µM SeMet treatment at stage 17, hatch was significantly less than all other stages (p < 0.01 for all).

After 50µM treatment, embryos treated at stage 17 had significantly less hatch than all stages except stage 29 (p < 0.001). Furthermore, embryos treated with 50µM SeMet at stage 29 had significantly less percent hatch than those treated at stages 9, 34 and 38 (p =

0.001, 0.011 and 0.023, respectively).

183

Figure 4.2. Percent hatch ± SE following 24 h of treatment with 0.5µM, 5µM, and 50µM SeMet in either freshwater (black bars) or saltwater (white bars) (n = 3-6). Each graph represents one stage of treatment: A. stage 9; B. stage 17; C. stage 25; D. stage 29; E. stage 34; and F. stage 38. Statistical significance was assessed using a logistic regression followed by multiple comparisons and p ≤ 0.05 was considered statistically significant. Differing letters represent significant differences between concentrations at a specific stage and forks denote differences between fresh and saltwater at a specific concentration. Symbols represent significant differences between stages (different graphs) at specific doses.

Deformities

The percent of total deformities (excluding failed swim bladder inflation) increased significantly with 5µM SeMet treatment during stages 9, 17, 25 and 34 (logistic regression; p < 0.025; Figure 4.3). Total deformities also increased following 50µM

SeMet during stages 9 and 25 (p = 0.003 and 0.001, respectively; Figure 4.3A and C).

After exposure to 5µM SeMet, saltwater and freshwater induced significantly different numbers in total deformities (p = 0.028), and embryos treated at stage 25 had greater

184 deformities than embryos treated at stages 29 and 34 (p = 0.035 and 0.045, respectively).

Embryos treated at stages 17 and 25 with 50µM SeMet also had significantly greater percent deformities than those treated at stage 34 (p = 0.031 and 0.013).

Figure 4.3. Percent deformities ± SE following 24 h of treatment with 0.5µM, 5µM, and 50µM SeMet in either freshwater (black bars) or saltwater (white bars) (n = 3-6). Each graph represents one stage of treatment: A. stage 9; B. stage 17; C. stage 25; D. stage 29; E. stage 34; and F. stage 38. Statistical significance was assessed using a logistic regression followed by multiple comparisons and p ≤ 0.05 was considered statistically significant. Differing letters represent significant differences between concentrations at a specific stage and forks denote differences between fresh and saltwater at a specific concentration. Symbols represent significant differences between stages (different graphs) at specific concentration.

Types of deformities

In addition to total deformities, the type of each deformity was also measured as spinal, heart, fin and cranio-facial. In the multinomial regression model, dose and water type were found to have no significant effect on the type of deformity, however, there

185 were significant differences in types of deformities within and between stages (p = 0.001;

Figure 4.4). With a frequency ranging from 60% to 89% of total deformities, embryos exhibited significantly greater spinal than other deformities when treated with SeMet at all stages except for stage 9 (p < 0.01). Embryos treated at stage 9 had significantly more spinal than fin deformities (p < 0.001). Significant differences were also detected between treatments at different stages. Embryos treated at stage 25 had significantly less cardiac deformities in relation to spinal deformities than embryos treated at stages 9 (p <

0.001), 29 (p = 0.03), and 34 (p = 0.021). In contrast, embryos treated with SeMet at stage 9 exhibited significantly greater cardiac deformities than those treated at stage 17 (p

= 0.002) and 38 (p = 0.017); and significantly greater fin deformities than 17 and 34 (p =

0.033 and 0.019). Embryos exposed at stages 9 and 38 also had greater cranio-facial deformities than stage 25 (p = 0.003 and 0.03 respectively).

186 Figure 4.4. Percent of each type of deformity inflation ± SE following 24 h of treatment with all concentrations of SeMet and both types of water (n =14-24 replicates with 10-20 embryos per replicate). Types of deformities measured include: spinal (lined bars), cardiac (black bars), cranio-facial (white bars) and fin (checkered bars). Statistical significance was assessed with a multinomial logistic regression using spinal deformities as the reference. P ≤ 0.05 was considered statistically significant. Differing letters represent significant differences within a stage, while symbols represent significant differences amongst one type of deformity between stages.

Spinal deformities

Type of spinal deformity was also scored as lordosis (inward curvature of the lower spine), kyphosis (rounded outward spinal curvature) or scoliosis (sideways spinal curvature) and assessed via multinomial logistic regression using lordosis as the reference deformity (Supplemental Data, Figure 4.5). As with type of deformity, stage was the only factor to have a significant effect on the type of spinal deformity in SeMet treated embryos. Embryos treated at stage 25 had significantly more lordosis than scoliosis or

187 kyphosis (65% in comparison to 12% and 23% for scoliosis and kyphosis respectively (p

= 0.001 and 0.002)). In contrast, embryos treated later, at stage 38, had significantly higher incidence of kyphosis than lordosis (p = 0.002). Types of spinal deformities also differed significantly depending on the treatment stage. The embryos treated at stage 25 had significantly less scoliosis and kyphosis than stages 9 (p = 0.042 and 0.022), 17 (p =

0.025 and 0.001), 29 (p = 0.005 and 0.004) and 38 (p < 0.001). Exposure at stage 38 generated significantly greater kyphosis in hatched embryos than embryos treated at 9,

17, 25 and 34 (p = 0.003, 0.017, 0.001 and 0.04, respectively).

Figure 4.5. Percent of each type of spinal deformity inflation ± SE following 24 h of treatment with all concentrations of SeMet and both types of water (n = 8-10 replicates). Types of spinal deformities measured include: lordosis (lined bars), scoliosis (black bars) and kyphosis (white bars). Statistical significance was assessed with a multinomial logistic regression using lordosis as the reference. P ≤ 0.05 was considered statistically significant. Differing letters represent significant differences within a stage, while symbols represent significant differences amongst one type of deformity between stages.

188 Severity of deformities was also considered using the graduated severity index and deformities were scored from 1- 3 (Holm et al., 2003). A score of 1 indicated a mild defect in spinal curvature, cranio-facial structure, or mild pericardial edema, whereas scores of 2 and 3 indicated a moderate and severe defects, respectively (Holm et al.,

2003). The sample size for cardiac, cranio-facial and fin deformities was too small to accurately calculate severity, however, average spinal severities were calculated for each treatment and significance was determined using a 3-way ANOVA and Tukey’s HSD test post hoc (Supplemental Data, Figure 4.6). No significant differences were observed between fresh and saltwater, however, a significant interaction between stage, water type and SeMet dose was detected (p = 0.001). Spinal deformities were more severe in embryos treated with 5µM and 50µM SeMet than 0.5µM SeMet (p = 0.011 and 0.0004, respectively). Spinal deformities in embryos treated at stage 9 were significantly more severe than those treated at stages 34 and 25 (p < 0.001). Furthermore, those treated at stage 34 had less severe spinal deformities than embryos treated at stages 9, 17, 29 and 38

(p = 0.001, 0.013, 0.021, and 0.018 respectively).

189

Figure 4.6. Percent average severity of spinal deformities ± SE following 24 h of treatment with 0.5µM, 5µM, and 50µM SeMet in either freshwater (black bars) or saltwater (white bars) (n = 1-5). Each graph represents one stage of treatment: A. stage 9; B. stage 17; C. stage 25; D. stage 29; E. stage 34; and F. stage 38. Statistical significance was assessed using a 3-factor ANOVA followed by a Tukey’s HSD test post-hoc. Differing letters indicate significant differences between treatments and are applicable between graphs. P ≤ 0.05 was considered statistically significant.

Swim bladder inflation

Although swim bladder inflation is not traditionally considered a deformity, and was thus kept separate from the other deformities measured, the failure of the swim bladder to inflate can also impact embryo survival. The percent of embryos with failed swim bladders following hatch significantly increased following treatment with 5µM

SeMet at stages 9, 17 and 25 (logistic regression; p = 0.006, 0.001 and 0.002, respectively), however, none of these differences were found at the 50µM treatment

(Figure 4.7A, B and C). At stage 17, embryos treated with 0.5µM SeMet also had

190 significantly more failed swim bladders (p = 0.007). Embryos exposed during stages 9,

17 and 25 had significantly greater incidence of swim bladder failure than those exposed at stage 34 (p = 0.02, 0.007 and 0.006). Finally, following exposure to 0.5µM SeMet at stage 17 and 38, embryos had increased swim bladder failure than those treated with

0.5µM SeMet at stage 25 (p = 0.022 and 0.039).

Figure 4.7. Percent failed swim bladder inflation ± SE following 24 h of treatment with 0.5µM, 5µM, and 50µM SeMet in either freshwater (black bars) or saltwater (white bars) (n = 3-6). Each graph represents one stage of treatment: A. stage 9; B. stage 17; C. stage 25; D. stage 29; E. stage 34; and F. stage 38. Statistical significance was assessed using a logistic regression followed by multiple comparisons and p ≤ 0.05 was considered statistically significant. Differing letters represent significant differences between concentrations at a specific stage and forks denote differences between fresh and saltwater at a specific concentration. Symbols represent significant differences between stages (different graphs) at specific concentrations.

191 Day to hatch

SeMet dose had little significant effect on the median day to embryo hatch

(Supplemental Data, Figure 4.8). Only embryos treated at stage 34 with 5µM SeMet took a greater time to hatch than those treated with 50µM (approximately 13 days in comparison to 10 days). However, saltwater did significantly decrease the median day to hatch in controls and after 50µM of treatment (p = 0.003 and 0.027, respectively).

Furthermore, significant differences in day to hatch were observed between stages.

Embryos treated with SeMet at stages 9 and 17, hatched significantly sooner than those treated at stages 25, 29 and 34 (p < 0.05 for all).

192

Figure 4.8. Median day to hatch ± SE following 24 h of treatment with 0.5µM, 5µM, and 50µM SeMet in either freshwater (black bars) or saltwater (white bars) (n = 3-6). Each graph represents one stage of treatment: A. stage 9; B. stage 17; C. stage 25; D. stage 29; E. stage 34; and F. stage 38. Statistical significance was assessed using a 3-factor ANOVA followed by a Tukey’s HSD test post-hoc and p ≤ 0.05 was considered statistically significant. Differing letters represent significant differences between concentrations at a specific stage and forks denote differences between fresh and saltwater at a specific concentration.

Tissue Se content

In order to determine tissue level Se content, ICP-MS was used to quantify total

Se in embryos following the 24 h treatments (Figure 4.9). Mean control embryo Se content was approximately 3.29 ± 0.24 µg/g. No significant differences were observed between freshwater and saltwater controls at any stage (ANOVA and Tukey HSD, p =

0.812). However, significant differences were observed between treatment concentrations

(all p < 0.001) and treatment stages (all p < 0.001). A dose dependent increase in Se

193 content was observed for stages 25, 29, 34 and 38 (Figure 4.9). Embryos assimilated significantly less Se when treated with 5µM SeMet at stage 9 than stages 25, 29, 34, and

38 (Average 5.59µg/g in comparison to 24.81µg/g, 33.89µg/g, 51.53µg/g, and 49.52µg/g, respectively; p < 0.001 for all; Supplemental Data, Figure 4.9). Embryos treated with

0.5µM SeMet at stage 17 also assimilated less Se than those treated at stages 29, 34 and

38 (Average 5.0µg/g in comparison to 16.25µg/g, 16.51µg/g and 15.86µg/g, respectively; p ≤ 0.001 for all).

Figure 4.9. Whole egg Se content (dry weight (DW) µg/g) in embryos surviving following 0.5µM, 5µM, and 50µM SeMet treatment in either freshwater (black bars) or saltwater (white bars) (n = 3). Samples were only collected for SeMet concentrations where the lowest level of significance was observed in the apical endpoints. Statistical significance was assessed with a 3-way ANOVA followed by a Tukey’s HSD test post- hoc and p ≤ 0.05 was considered statistically significant. Differing letters represent significant differences between concentrations at a specific stage and differing symbols denote differences stages for a specific concentration

194 The relationship between mean Se content and average percent deformities was assessed using linear regression (Supplemental Data, Table 4.2). A significant correlation was detected for stages 25, 29, 34 and 38. An ANCOVA was performed to ascertain differences between stages and a significant interaction between Se content and stage was detected. Embryos treated at stage 25 had significantly increased deformities by Se content than stages 29, 34 and 38 (i.e. greater slope of regression).

Table 4.2. Relationship between egg Se concentration and mean percent deformities for embryos treated with SeMet in freshwater and saltwater. Linear regressions were performed for each stage of development and significance was determined at p ≤ 0.05.

Stage r2 Value Regression P Value 9 0.04 0.645 17 0.67 0.182 25 0.77 0.022* 29 0.69 0.042* 34 0.71 0.034* 38 0.96 0.001*

Discussion

Stage sensitivity of Japanese medaka embryos to SeMet and osmotic stress was evaluated using lethal and sublethal indicators. Pairing these endpoints with Se tissue level measurements allows for the determination of stage specific thresholds to SeMet under hypersaline conditions, such as those found in the San Francisco Bay Delta. As climate change and agricultural development increases the salinity of the bay delta, these values become increasingly important for environmental protection. Furthermore, multiple desalination projects are planned for the area, and disposal of desalination brine back into the estuary may further increase salinity of waterways and concentrate Se within discharges.

195 Japanese medaka exhibited salinity and stage sensitivity to SeMet at different concentrations. In general, a concentration response to SeMet was observed. However,

SeMet concentration was not correlated with embryo Se dose at all stages. For example, embryo Se content was not equal for 5µM SeMet between stages, although significant differences were observed for all endpoints measured. Significantly less SeMet was assimilated by embryos treated at stages 9 and 17 than later stages. These differences may be mediated by the chorion. Japanese medaka have a thick (10µm) chorion, which begins to harden immediately following fertilization, reaching maximum hardness at

6hpf, then gradually softening again beginning at 6 days post fertilization (Suga, 19663;

Yamagami et al., 1992). The chorion may impede aqueous toxicant uptake (Blaxter,

1988), and in one study dechorionated Japanese medaka embryos at stage 13 were more sensitive to thiobencarb toxicity than chorionated embryos (Villalobos et al., 2000).

Further research would be needed in order to distinguish between the role of the chorion and other mechanisms in aqueous SeMet uptake.

Nevertheless, the waterborne exposures performed in the present study were only used as a model of SeMet exposure and thus the chorion would not play a role in environmental Se assimilation. In oviparous organisms, exposures to Se in embryos typically occur via maternal transfer (Fan et al., 2002). Maternal SeMet is incorporated into vitellogenin, a yolk precursor protein, during synthesis, which is then utilized by the embryo throughout development (Janz et al., 2010). The efficiency of this process varies greatly by species and reproductive strategy (Janz et al., 2010), however, yolk utilization and thus mobilization of SeMet from vitellogenin would be the major determinant of

196 toxicity. Early studies in the medaka have examined transfer of maternal S35-dl- methionine into the yolk and subsequent transfer from the yolk into the embryo (Monroy et al., 1961). Although no transfer was observed through gastrulation, a rapid increase in radioactivity was detected beginning at stage 19 (closure of the blastopore) and increased steadily until hatch (Monroy et al., 1961), at which point the embryo becomes a feeding larva. Considering free SeMet is responsible for Se embryo toxicity (Fan et al., 2002), this evidence suggests that SeMet mobilization from the yolk would occur beginning during the stage 17 treatments performed, but only at the very end of the stage 9 treatments.

Even with differences in Se assimilation due to chorion interference, stage specific differences in SeMet toxicity could be observed. Stage 17 appears to be the most sensitive stage for both lethal and sublethal endpoints; significant effects on hatch and deformities were observed at tissue concentrations of 3-6µg/g Se, which was not significantly different from controls. Additionally, although stage 9 appeared less sensitive to higher concentrations of SeMet (significantly greater survival than all other stages treated with 5µM and 50µM SeMet), significant deformities and decreased hatch were measured for Se tissue doses of 5-6µg/g. These values are lower than those reported in the literature. Studies of Northern pike (Esox lucius) exposed to mining effluent in

Saskatchewan, Canada, found 33.55µg/g DW egg Se to be the EC20 for deformities

(Muscatello et al., 2006). Another study on rainbow trout (Oncorhynchus mykiss) exposed to Se in Alberta, Canada, found an egg EC15 of 21.1µg/g DW Se for skeletal deformities (Holm et al., 2005). These reported values are more similar to the values

197 following 5µM SeMet treatment at stages 25 and 29 presented here, which ranged from

23 to 37 µg/g Se DW. However, percent of population affected at these stages and this treatment level was greater than 15-20%, and ranged from 40-100% for hatch and deformities. These discrepancies may be due to differences in species sensitivity or different exposure methods.

Nonetheless, it is expected that earlier developmental stages would be more susceptible to developmental toxicants than later ones. It is generally accepted that susceptibility to teratogenesis follows a specific pattern throughout development. During cleavage and pre-differentiation, embryos are considered less susceptible to a toxicant, which is expected to either cause mortality or induce sufficient repair (Timbrell, 2009).

Susceptibility to teratogens is thought to then gradually increase during gastrulation and peaks during early organogenesis, to steady decline until hatch (Timbrell, 2009). At stage

9, the embryo is in the early blastula stage, synchronous cleavage is still occurring, and cells are in a single mass. During the 24 h treatment duration, the embryo undergoes gastrulation and neurulation, and somite formation begins (Iwamatsu, 2004). At stage 17, early neurulation begins with the formation of an embryonic body and head. Embryos treated at this stage begin optic, brain, and heart formation during the treatment period

(Iwamatsu, 2004). Thus, the increased sensitivity of stage 17 over stage 9 follows the accepted dogma. However, early organogenesis continues through stages 25 and 29, where sensitivity is decreased in relation to Se content. However, embryos treated at stage 29 were more susceptible to SeMet toxicity to hatch than other stages.

198 In addition to stage specific effects, effects of salinity on SeMet toxicity were observed and were stage and SeMet concentration specific. Significant differences between fresh and saltwater were only measured at higher concentrations of SeMet- at the 50µM SeMet treatment level for percent survival, hatch, and day to hatch, and at the

5µM (approximately 24µg/g) level for percent hatch, and deformities. These differences were not a result of differences in SeMet uptake between fresh and salt water. This indicates that salinity has an increased toxicity at higher levels of SeMet.

Specifically, stage 25 can be identified as particularly sensitive to salinity. Less than 3% of embryos survived or hatched when treated with SeMet in saltwater, in contrast to 40%-60% in freshwater embryos. Embryos treated with 5µM SeMet in saltwater also had greater deformities and failed swim bladders than those treated in freshwater (100% in comparison to 66%). This indicates that saltwater is able to potentiate SeMet toxicity when embryos are treated at stage 25 for 24 h.

These differences may be explained by the developmental changes occurring during this period. Osmoregulation in medaka begins at stage 25 (Thermes et al., 2010), which may influence SeMet toxicity under osmotic stress at this stage. Furthermore, stage

25 is when the liver anlage appears, although full functionality will not be reached until late organogenesis (Hinton et al., 2004). The liver is the primary site of xenobiotic metabolism. Flavin-containing monooxygenases (FMOs) are located predominantly in the liver during development in mice (Janmohamed et al., 2004) and are able to oxidize

SeMet to the corresponding oxide in mice and humans (Krause et al., 2006). Although little work has investigated the role of FMOs during teleost development, FMOs have

199 been shown to be upregulated during hypersaline conditions in euryhaline fish (Lavado et al., 2012; Schlenk et al., 1996; El-Alfy et al., 2002) and previous research suggests the

FMO activation of SeMet can generate oxidative stress and toxicity in Japanese medaka

(Lavado et al., 2012). Developmental expression of FMOs has not been investigated in medaka, however, FMO may play a role in increased SeMet toxicity under hypersaline conditions at stage 25.

The extensive deformities assessment performed here is also highly relevant to

SeMet toxicity during osmotic stress. Lemly (1997) found a significant exponential correlation between selenium tissue concentrations and percent terata in Se impacted populations, and estimated that 80% of deformed larvae die regardless of their Se concentration. While this value may vary with the threat of predation, it suggests the importance of the consideration of teratogenesis when evaluating Se toxicity in natural populations. Based on a number of Se toxicity studies, Lemly (1997) developed an index for the impact of teratogenesis at the population level. According to the index, 6-25% deformities in a larval population would result in 5-20% population mortality, while greater than 25% deformities would result in greater than 20% population mortality, which would constitute a major population loss. Our results indicate that 5-20µg/g SeMet exposures at stages 9, 17 and 25 generated significant increases in total deformities of 20-

80%, which would lead to moderate to severe population mortality.

Type of deformity may also influence probability of larval survival. For instance, spinal deformities can impact swimming performance and thus an animals’ ability of obtain food or escape predators. Consistent with previous studies, spinal deformities were

200 the most common type of deformity observed in the SeMet treated embryos [Lemly,

1997; Muscatello et al., 2006], while fin deformities were the least common. The specificity of the deformity varied by stage, with embryos treated at stages 17, 25, 29 and

38 with significantly more spinal deformities than any other type. Spinal deformities were further divided into lordosis, kyphosis and scoliosis. Lordosis is the most prevalent type of spinal deformity caused by Se in the environment, and our results indicate that lordosis was the most common deformity from embryos treated at stage 25. This would suggest that SeMet’s mode of action is most specific at this stage.

Pericardial edema was the most prevalent deformity measured in the cardiac category and edema is generally considered to be a temporary and reversible deformity and is thus not always an accurate measure of teratogenesis past the embryo-larval stage

(Lemly, 1993b). Conflicting reports on the effects of edema on larvae survival have been reported. Pyron and Beitinger (1989) found that fathead minnow larvae from Se treated mothers with edema had low survivability, whereas Hermanutz (1992) found no effect of edema on survival in fathead minnow exposed to Se in outdoor streams.

Failed swim bladder is common in aquaculture and has been observed in wild populations at levels of 0.1-8% (Egloff, 1996). In aquaculture, failed swim bladder inflation can result in significant losses and has been found to be associated with lordosis

(Chatain, 1994), which may suggest that not all spinal deformities measured in the present study were caused directly by SeMet. While one study found that fish without swim bladders are able to survive to adulthood, these fish expended more energy to maintain their position in the water column and to catch prey (Czesny et al., 2005).

201 Furthermore, Japanese medaka larvae without swim bladder consumed significantly more oxygen (Marty et al., 1995). This information indicates that fish without a swim bladder may have reduced fitness in the wild. Interestingly, salinity also has an impact on swim bladder inflation. In Australian bass (Mucquaria nouemczculeut) a decrease in salinity from 25 to 10 ppth resulted in decreased swim bladder inflation (Battaglene & Talbot,

1990), whereas in gilthead sea bream the opposite effect was observed (Wooley & Qin,

2010). However, there was no significant effect of salinity on swim bladder inflation in the present study.

The present study is the first comprehensive work on the stage specific effects of

SeMet in fish embryos. Our results suggest that earlier stages of SeMet are more susceptible to lethality and deformities. We are further able to identify the peak of hypersaline potentiation of SeMet toxicity at stage 25. This information is valuable for determination of site-specific Se thresholds and for characterization of an adverse outcome pathway for selenium in both freshwater and saltwater.

202 References

Alderice, D.F., 1988. Osmotic and ionic regulation in teleost eggs and larvae. In Fish Physiology Vol 11, part A- The Physiology of Developing Fish- Eggs and Larvae (W.S. Hoar, D.J. Randall, Eds.), Vol 11, pp. 163-251. Academic Press, Inc., San Diego.

Ankley, G.T., Bennett, R.S., Erickson, R.J., Hoff, D.J., Hornung, M.W., Johnson, R.D., Mount, D.R., Nichols, J.W., Russom, C.L., Schmieder, P.K., Serrrano, J.A., Tietge, J.E., Villeneuve, D.L., 2010. Adverse outcome pathways: a conceptual framework to support ecotoxicology research and risk assessment. Environ. Toxicol. Chem. 29, 730-41.

Battaglene, S.C., Talbot, R.B., 1990. Initial swim bladder inflation in intensively reared Australian bass larvae. Macquaria novemacctleata (Steindachner) (Perciformes: Percichthyidae). Aquacult. 86, 431-442.

Blaxter, J.H.S., 1988. Pattern and variety in development. In Fish Physiology (W.S. Hoar, D.J. Randall, Eds.), Vol. XIA, pp. 1–58. Academic Press, San Diego.

Chatain, B., 1994. Abnormal swimbladder development and lordosis in sea bass (Dicentrarchus labrax) and sea bream (Sparus auratus). Aquacult. 119, 371-379.

Czesny, S.J., Graeb, B.D.S., Dettmers, J.M., 2005. Ecological consequences of swimbladder noninflation for larval yellow perch. Trans. Am. Fish. Soc 134, 1011–1020.

Delta Science Council., 2013. Final Draft Delta Plan. Delta Stewardship Council, Sacramento, CA.

Egloff, M., 1996. Failure of Swim Bladder of Perch, Perca fluviatilis L. Found in Natural Populations. Aquat. Sci. 58.

El-Alfy, A., Larsen, B., Schlenk, D., 2002. Effect of cortisol and urea on flavin monooxygenase activity and expression in rainbow trout, Oncorhynchus mykiss. Mar. Environ. Res. 54, 275−8.

Fan, T.W., Teh, S.J., Hinton, D.E., Higashi, R.M., 2002. Selenium biotransformations into proteinaceous forms by foodweb organisms of selenium-laden drainage waters in California. Aquat. Toxicol. 57, 65-84.

Hermanutz, R.O., 1992. Malformation of the fathead minnow (Pimephales promelas) in an ecosystem with elevated selenium concentrations. Bull. Environ. Contam. Toxicol. 49, 290–294.

Hinton, D.E., Wakamatsu, Y., Ozato, K., Kashiwada, S., 2004. Imaging Liver Development Remodeling in the See-Through Medaka Fish. Comp. Hepatol. 3, 30–34.

203 Holm, J., Palace, V.P., Siwik, P., Sterling, G., Evans, R.E., Baron, C.L., Werner, J., Wautier, K., 2005. Development effects of bioaccumulated selenium in eggs and larvae of two salmonid species. Environ. Toxicol. Chem. 24, 2373−81.

Holm, J., Palace, V.P., Wautier, K., Evans, R.E., Baron, C.L., Podemski, C., Siwik, P., Sterling, G., 2003. An assessment of the development and survival of wild rainbow trout (Oncorhynchus mykiss) and brook trout (Salvelinus fontinalis) exposed to elevated selenium in an area of active coal mining. The Big Fish Bang. Proceedings of the 26th Annual Larval Fish Conference. Institute of Marine Research, Bergen, Norway, 257-73.

Iwamatsu, T., 2004. Stages of Normal Development in the Medaka Oryzias latipes. Mechan. Dev. 121, 605-618.

Janmohamed, A., Hernandez, D., Phillips, I.R., Shepard, E.A., 2004. Cell-, tissue-, sex- and developmental stage-specific expression of mouse flavin-containing monooxygenases (Fmos). Biochem. Pharmacol. 68, 73-83.

Janz, D.M., DeForest, D.K., Brooks, M.L., Chapman, P.M., Gilron, G., Hoff, D., Hopkins, W.A., McIntyre, D.O., Mebane, C.A., Palace, V.P., Skorupa, J.P., Wayland, M., 2010. Selenium Toxicity to Aquatic Organisms. In Ecological Assessment of Selenium in the Aquatic Environment (P.M. Chapman, W.J. Adams, M.L. Brooks, C.G Delos, S.N Luoma, W.A. Maher, H.M. Ohlendorf, T.S. Presser, D.P. Shaw, Eds.), pp 139-240. Society of Environmental Toxicology and Chemistry (SETAC), Pensacola.

Kirchen, R.V., West, W.R., Carolina Biological Supply C., 1976. The Japanese medaka: its care and development. Carolina Biological Supply Company.

Krause, R.J., Glocke, S.C., Sicuri, A.R., Ripp, S.L., Elfarra, A.A., 2006. Oxidative Metabolism of Seleno-L-Methionine to L-Methionine Selenoxide by Flavin- Containing Monooxygenases. Chem. Res. Toxicol. 19, 1643-9.

Kupsco, A., Schlenk, D., 2014. Mechanisms of selenomethionine developmental toxicity and the impacts of combined hypersaline conditions on Japanese medaka (Oryzias latipes). Env. Sci. Technol. 48, 7062-8.

Langille, R.M., Hall, B.K., 1987. Development of the Head Skeleton of the Japanese Medaka, Oryzias latipes (Teleostei). J. Morphol. 193, 135-158.

Lemly, A.D., 1993a. Guidelines for evaluating selenium data from aquatic monitoring and assessment studies. Environ. Mon. Assess. 28, 83–100.

204 Lemly, A.D., 1993b. Teratogenic effects of selenium in natural populations of freshwater fish. Ecotoxicol. Environ. Saf. 26, 181–204.

Lemly, A.D., 1997. A teratogenic deformity index for evaluating impacts of selenium on fish populations. Ecotox. Environ. Safe. 37, 259-66.

Lemly, A.D., 2002. Symptoms and implications of selenium toxicity in fish: the Belews Lake case example. Aquat. Toxicol. 57, 39-49.

Marty, G.D., Hinton, D.E., Cech, J.J., 1995. Oxygen consumption by larval Japanese Medaka with inflated or uninflated swim bladders. Trans. Am. Fish. Soc. 124, 623-627.

Monroy, A., Ishida, M., Nakano, E., 1961. The Pattern of Transfer of the Yolk Material to the Embryo during the Development of the Teleostean Fish, Oryzias latipes. Embryologia. 6, 151-8.

Muscatello, J.R., Bennett, P.M., Himbeault, K.T., Belknap, A.M., Janz, D.M., 2006. Larval Deformities Associated with Selenium Accumulation in Northern Pike (Esox lucius) Exposed to Metal Mining Effluent. Environ. Sci. Technol. 40, 6506-6512.

Ohlendorf, H.M., Covington, S.M., Byron, E.R., Arenal, C.A., 2011. Conducting site- specific assessments of selenium bioaccumulation in aquatic systems. Integr. Environ. Assess. Manag. 7, 314-24.

Outridge, P.M., Scheuhammer, A.M., Fox, G.A., Braune, B.M., White, L.M., Gregorich, L.J., Keddy, C., 1999. An assessment of the potential hazards of environmental selenium for Canadian water birds. Environ. Rev. 7, 81-96.

Presser, T.S., Piper, D.Z., Bird, K.J., Skorupa, J.P., Hamilton, S.J., Detwiler, S.J., Huebner, M.A., 2004. Chapter 11 The phosphoria formation: A model for forecasting global selenium sources to the environment. In Handbook of Exploration and Environmental Geochemistry (R.H. James, Ed.), Vol. 8., pp. 299-319. Elsevier Science B.V.

Pyron, M., Beitinger, T.L., 1989. Effect of selenium on reproductive behavior and fry of fathead minnows. Bull. Environ. Contam. Toxicol. 42, 609–613.

205 R Core Team., 2012. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. ISBN 3-900051-07-0, Available from: http://www.R-project.org/.

Ramirez, P., Rogers, B.P., 2002. Selenium in a Wyoming grassland community receiving wastewater from an in situ uranium mine. Arch. Environ. Con. Tox. 42, 431-6.

Schlenk, D., Zubcov, N., Zubcov, E., 2003. Effects of salinity on the uptake, biotransformation, and toxicity of dietary seleno-L-methionine to rainbow trout. Toxicol. Sci. 75, 309-313.

Suga, N., 1963. Changes of the toughness of the chorion of fish eggs. Embryologia. 8, 63-74.

Thermes, V., Lin, C.C., Hwang, P.P., 2010. Expression of Ol-foxi3 and Na+/K+-ATPase in ionocytes during the development of euryhaline medaka (Oryzias latipes) embryos. Gene Expr. Patterns.10, 185-192.

Timbrell, J.A., 2009. Chapter 6: Toxic Responses to Foreign Compounds. In Principles of Biochemical Toxicology, 4th ed., pp 193-287. Informa Healthcare USA.

Villalobos, S.A., Hamm, J.T., Teh, S.J., Hinton, D.E., 2000. Thiobencarb-induced embryotoxicity in medaka (Oryzias latipes): stage-specific toxicity and the protective role of chorion. Aquat. Toxicol. 48, 309-26.

Wong, P.P., Losada, I.J., Gattuso, J.P., Hinkel, J., Khattabi, A., McInnes, K.L., Saito, Y., Sallenger, A., 2014. Coastal systems and low-lying areas. Climate Change 2014: Impacts, Adaptation, and Vulnerability. Part A: Global and Sectoral Aspects. Contribution of Working Group II to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change (Field, C.B., V.R. Barros, D.J. Dokken, K.J. Mach, M.D. Mastrandrea, T.E. Bilir, M. Chatterjee, K.L. Ebi, Y.O. Estrada, R.C. Genova, B. Girma, E.S. Kissel, A.N. Levy, S. MacCracken, P.R. Mastrandrea, and L.L.White, Eds.), pp. 361-409. Cambridge University Press, Cambridge and New York.

Woolley, L.D., Qin, J.G., 2010. Swim Bladder Inflation and its Implication to the Culture of Marine Fish Larvae. Rev. Aquacult. 2, 181-190.

Yamagami, K., Hamazaki, T.S., Yasumasu, S., Masuda, K., Iuchi, I., 1992. Molecular and cellular basis of formation, hardening, and breakdown of the egg envelope in fish. Int. Rev. Cytol. 136, 51–92.

206 Yasutake, J., Inohaya, K., Kudo, A., 2004. Twist functions in vertebral column formation in medaka, Oryzias latipes. Mech. Dev. 121, 883-94.

207 Chapter 5: Mechanisms of selenomethionine developmental toxicity and the impacts of combined hypersaline conditions on Japanese medaka (Oryzias latipes)

Abstract

Selenium (Se) is an essential micronutrient that can cause embryotoxicty at levels

7-30 times above essential concentrations. Exposure to hypersaline conditions and 50µM selenomethionine (SeMet) decreased embryo hatch and depleted glutathione in Japanese medaka embryos without affecting Se accumulation. To better understand the impacts of non-chemical stressors on developmental toxicity of Se in fish, several adverse outcome pathways were evaluated in the Japanese medaka (Oryzias latipes). We treated medaka embryos at 12 hours post fertilization with 50µM SeMet for 12hours in freshwater or in

13ppth hypersalinity and evaluated the contributions of oxidative stress, the unfolded protein response and apoptosis to reduced hatch. Exposure to SeMet and hypersalinity decreased embryo hatch to 3.7% ± 1.95, and induced teratogenesis in 100% ± 0 of hatched embryos. In contrast, treatments of freshwater, saltwater and SeMet in freshwater resulted in 89.8% ± 3.91-86.7% ± 3.87 hatch, and no significant increase in deformities.

We found no significant differences in lipid peroxidation, indicating that oxidative stress may not be responsible for the observed toxicity in embryos at this time point (24hr).

Although significant changes in apoptosis were not observed, we witnessed up to 100 fold increases in transcripts of the endoplasmic reticulum (ER) chaperone, immunoglobulin binding protein (BiP) and trends towards increasing downstream signals, activating transcription factor 4 (ATF4) and ATF6 indicating potential contributions of the unfolded protein response to the effects of SeMet and hypersaline

208 conditions. These data indicate that multiple adverse outcome pathways may be responsible for the developmental toxicity of Se and salinity, and these pathways may be time dependent.

Introduction

Selenium (Se) is an essential micronutrient; levels only 7-30 times greater than required can be toxic (Lemly, 1997). This is a concern in aquatic environments, where anthropogenic activities can release large quantities of Se, and include agricultural runoff of irrigation waters in arid regions (Outridge et al., 1999); waste rock from coal, phosphate and uranium mining (Muscatello et al., 2006; Presser et al., 2004; Ramirez &

Rogers, 2002); and combustion waste from coal burning power plants (Wen & Carignan,

2007). Se usually enters the waterways in its inorganic forms of selenate (Se+4) or selenite (Se+6), which can be taken up by microbes and primary producers and converted into various organic forms, including the amino acid, selenomethionine (SeMet) (Fan et al., 2002). Consumers, such as fish and birds, are exposed to Se primarily in the diet and

SeMet has been shown to be the major form of Se in the fish diet (Phibbs et al., 2011).

One concern for SeMet toxicity is its bioaccumulation potential. SeMet has been demonstrated to move through the food chain by trophic transfer to higher-level organisms (Luoma & Rainbow, 2005; Lemly, 2002). Following a Se poisoning event at

Belews lake, NC, Lemly found Se to have bioaccumulated from 519 times in periphyton to 3975 times in the visceral tissues of fish (Lemly, 2002). This is of particular concern

209 for oviparous carnivores, for which maternal offloading may impair development or reproductive success through respective teratogenesis or embryo lethality (Lemly, 2002).

Because of the potential for biomagnification, traditional water quality measurements of Se concentrations may be ineffective. Recently, the USEPA has begun advocating tissue concentration measurements for Se monitoring (USEPA, 2004).

However, even these measurements may not provide an accurate picture of Se effects on an ecosystem, because fish encounter multiple stressors in their environments, which can alter Se toxicity. Recent evidence suggests that hypersalinity may compound Se toxicity

(Lavado et al., 2012). This is of particular importance in areas such as the San Joaquin

River Valley, CA, and the San Francisco Bay Delta area, where many historically freshwater waterways are in danger of salinization (Enright & Culberson, 2010). These areas are often spawning grounds for protected species such as the endangered delta smelt (Hypomesus transpacificus) and threatened steelhead trout (Oncorhynchus mykiss).

The mechanisms behind Se induced teratogenesis and mortality remain unclear.

Several studies point to oxidative stress as one mode of action for Se toxicity (Palace et al., 2004; Misra et al., 2012; Miller et al., 2007; Hoffman et al., 1996). However, oxidative stress is most likely only one factor influencing SeMet toxicity. The unfolded protein response (UPR) is a cellular and molecular response to perturbations in endoplasmic reticulum (ER) homeostasis (See Hetz (2012) for review). Oxidative stress, calcium disruption, and glycosylation inhibition, can all disrupt protein folding, leading to the accumulation of unfolded proteins in the ER. Protein folding chaperones, such as

BiP (immunoglobulin- binding protein; Grp78), initiate the UPR through dissociation

210 from the mediators of the three branches, PERK (protein kinase RNA (PKR)-like ER kinase), IRE1a (inositol- requiring protein-1) and ATF6 (activating transcription factor

6). While the three branches of the response are highly interconnected, they can be generally divided into three categories. The PERK branch is responsible for translational attenuation through Activating Transcription Factor 4 (ATF4). The IRE1a branch is responsible for transcription of ER-Associated Degradation (ERAD) genes through X box protein 1 (XBP1). And finally, ATF6 is responsible for transcription of protein folding enzymes and chaperones. If the response is unable to attenuate the stress, the

UPR will initiate cell death, often in the form of programmed cell death (apoptosis).

We have previously demonstrated that 50µM of SeMet and hypersalinity treatment for 24 hours significantly decreased embryo hatch, decreased total reduced glutathione and increased flavin containing-monooxygenase (FMO) activity in medaka embryos (Lavado et al., 2012). Exposure of Japanese medaka embryos to SeMet under varied salinities did not impact overall Se accumulation but significant differences in toxicity were observed (Lavado et al., 2012). The purpose of the current study was to further elucidate the mechanisms behind SeMet and hypersaline induced embryo mortality after 12 hours of SeMet treatment at developmental stages not previously examined. We hypothesized that SeMet would induce oxidative stress, the UPR and apoptosis in Japanese medaka embryos and that hypersaline conditions would potentiate these effects. This research will aid in the development for site specific monitoring for Se in CA.

211 Methods

Chemicals and Reagents

Seleno-L-methionine (Purity 98%), 1-butanol, phosphoric acid, thiobarbituric acid and all other reagents were purchased from Sigma Aldrich (St. Louis, MO). A Milli-Q water purification system (Millipore, Billerica, MA) was used to obtain deionized water.

Ethanol (Fisher, Pittsburg, PA) was of molecular biology grade.

Embryo Collection and Exposure

Japanese Medaka were cultured at the University of California- Riverside and housed in a 2:3 ratio of males to females in medium-hard water at 27°C and a photoperiod of 14 hours light and 10 hours dark. Adults were fed twice daily a diet of live brine shrimp. Embryos were collected 0-1 hour following fertilization. Viable embryos were determined based on oil droplet migration to the vegetal pole as outlined by Kirchen and West (1976). Nonviable embryos were discarded and viable embryos were placed into 60x15mm petri dishes containing either freshwater or a make-up of saltwater from the San Joaquin River Valley (20-30 embryos per replicate, and 5-10 replicate per group).

Although, water-borne SeMet exposures do not represent the most likely environmental exposure (the primary exposures for SeMet are dietary or via maternal transfer), they are sufficient to study the mechanistic effects of SeMet on medaka embryos. San Joaquin

River Valley saltwater was prepared in the lab according to a recipe from Westlands

Water District located about 10 km south of Mendota in the San Joaquin River Drainage

Basin, CA (Schlenk et al., 2003). Salinity was measured with a conductivity meter, and corresponds to about 13‰ and 15.3g/L suspended solids.

212 Following 12 hours of equilibration in fresh or salt water, the replicates were divided into three groups. The first subset of embryos were frozen -80ºC to represent a time zero control (12hpf). Other embryos were treated with a 50µM solution of Seleno-L-

Methionine (Cat. No. S3132, Sigma-Aldrich) in freshwater or saltwater and exposed for

12 hours, then collected and frozen at -80°C for analysis (24hpf). SeMet concentrations were chosen based on previous research and were intended to represent the upper levels of bioaccumulation measured in embryos (Lavado et al., 2012). Previous studies have also demonstrated uptake of SeMet into the embryo, indicating this system was an effective exposure method (Lavado et al., 2012). The final subset of embryos was left in freshwater or saltwater for 24 hours to compare to the SeMet treated 12hpf samples.

Modified Embryo Larval Toxicity Assay

The medaka embryo-larval toxicity assay was adapted from Farwell et al. (2006).

Embryos were collected and treated as above with one replicate equal to 15-20 embryos per dish. Following 12 hours of SeMet treatment, embryos were rinsed and transferred to new dishes containing freshwater or saltwater. Water was changed every other day and dishes were monitored for mortality with removal of dead embryos. Embryo hatch was monitored for 21 days post fertilization. At hatch, embryos were assessed for deformities and terminated immediately. Percent hatch and percent of hatched embryos with deformities were recorded.

Analysis of Gene Expression

Total mRNA was isolated from embryos using the Lipid Tissue RNeasy kit

(Qiagen, Valencia, CA) following the manufacturers instructions. mRNA quantity and

213 quality was measured using the ND-1000 (NanoDrop, Wilmington, DE). mRNA (1µg) was converted to cDNA using the Reverse Transcription System (Promega Corporation,

Madison, WI), according to the manufacturers instructions.

Primers were designed using IDTDNA PrimerQuest software and optimized using

PCR Miner (Zhao & Fernald, 2005) (Table 5.1). As no BAX gene for Japanese medaka has been annotated in the NCBI database, BLAST was used on the medaka genome

(http://compbio.dfci.harvard.edu/cgi-bin/tgi/Blast/index.cgi) against BAX from zebrafish

(Danio rerio) to develop primers. Similarity between the genes had an E value of 1*10^-

48. EF1α was run as a housekeeping gene. qPCR was performed with the iScript™ One- step RT-PCR kit with SYBR® Green from Bio-Rad (Hercules, CA, USA), omitting the reverse transcriptase, on a MyiQ5 Thermo cycler (Bio-rad). The samples were denatured and the polymerase activated at 95°C for 5 minutes, then 40 cycles of 10s at 95°C and

30s of 55°C. Samples were subject to melting curve analysis from 65-85°C in 0.5°C increments with continuous fluorescence measurement. qPCR was analyzed according to

Schmittgen and Livak (2008) and fold change was calculated against the 12hpf freshwater controls. All data was compared against the 12hpf controls in order represent how the gene expression changed over the 12hr time period, and how the treatments affected this change. Rather than observing a discrete point in development, we feel it is necessary to understand how these treatments altered normal development.

214 Table 5.1. Primers, Accession Numbers, and Concentrations used for qRT-PCR.

Gene Forward and Reverse Primers (5’-3’) Accession Number Concentration EF1-a CTACATCAAGAAGATCGGCTACAA NM_001104662.1 2.5µM CGACAGGGACAGTTCCAATAC CASP3A CCAAATCCCAGGTCTACTGATG NM_001104670.1 5µM AGGCAAAGGAGGCAAACTTA BAX GCTGGTCATAAAGGCTCTCATC NM_131562.2 2.5µM CCAGATTGCTCGAACCGTAAA BiP GGAGGATTCTGACCTGAAGAAG NM_001278801.1 0.5µM GGTGACAGTAGGCTGGTTATC ATF6 CAAGCCAACTCCAGTCAGTATC NM_001278901.1 0.5µM GCCGACTCTCGGTTCTTTATC ATF4 CTTAGAGGTGAAGGTGCCTATG XM_004066069.1 2.5µM TGAGGAAGGAGACCTGTTAGA

Analysis of Oxidative Stress

Thiobarbituric Reactive Substances (TBARS) were measured to estimate malondialdehyde (MDA) formation in medaka embryos (Jentzsch, 1996). Embryos (15-

20) were weighed and homogenized in 1.15% KCl then centrifuged at 3000rpm for 5 minutes at 4°C. The supernatant was then used in the assay. Samples were run in duplicate on a Wallac Victor2 multilabel plate reader (PerkinElmer, Waltham, MA) with excitation at 535nm and emission at 585nm.

Statistical Analysis

Statistical significance was assessed using a student’s T-Test or 2-way ANOVA in the statistical program R. Statistical significance was determined at p ≤ 0.05, unless otherwise noted. If overall significance was determined following two-way ANOVA,

Tukey’s HSD test was performed post-hoc. Data was checked for normality and homogeneity of variances. Any non-normal data was log transformed. For data that remained non-normal following log transformation, Kruskal-Wallis tests were performed with Dunn’s test post-hoc.

215 Results

Embryo-Larval Toxicity Assay

While there was no difference in hatch between saltwater and freshwater control embryos, 50µM SeMet significantly decreased embryo hatch in saltwater treated embryos to 3.7% (Figure 5.1A). However, SeMet treatment in freshwater did not significantly decrease hatch. The median day to hatch was not significantly affected by SeMet treatment in freshwater or saltwater (Figure 5.1B). SeMet and hypersaline treatment also significantly increased the deformities in treated embryos (Figure 5.1C). All SeMet and hypersaline treated embryos had deformities upon hatch. The most common deformities observed were kyphosis, lordosis, craniofacial and yolk sac edema (Figure 5.1D).

216

Figure 5.1. Effects of combined exposure of SeMet (50µM) and hypersaline conditions on the development and hatchability of Japanese medaka embryos after exposure at 12 hpf. White bars represent freshwater and black bars represent saltwater. A) Percent hatch, B) median day to hatch, C) percent deformities in hatched embryos, D) examples of deformities; control is top image, bottom two images demonstrate lordosis, kyphosis, cranio-facial abnormalities and yolk-sac edema. Each value represents the mean ± standard error (SE) of 5-10 replicates. Statistical significance is indicated by differing letters (Two-way ANOVA, Tukey HSD test or Kruskal Wallis, Dunn’s test p ≤ 0.05).

Oxidative Stress and Apoptosis

There was no significant difference in amount of lipid peroxidation between any of the treatments and no difference between embryos at 12hpf and 24hpf (Figure 5.2).

BAX transcript levels decreased significantly from 12hpf to 24hpf (0.2 fold, p = 0.011),

217 while Caspase 3A levels remained constant (Figure 5.3A). There was no significant difference in BAX or CASP3A gene expression following treatment with hypersalinity or

SeMet (Figure 5.3B and C).

Figure 5.2. Effects of combined exposure of SeMet (50µM) and hypersaline conditions on lipid peroxidation in Japanese medaka embryos after 12 and 24 hpf. A) TBARS was measured as nmol/g wet weight tissue in freshwater controls at 12 hpf and 24 hpf. B) TBARS measured in embryos in freshwater, saltwater, SeMet in freshwater and SeMet in saltwater at 24hpf. Hydrogen peroxide (3% for 3 hours) was run as a positive control. Each value represents the mean ± SE of 5-10 replicates Statistical significance is denoted by differing letters at p ≤ 0.05 (One-way ANOVA, Tukey HSD test).

218

Figure 5.3. Effects of combined exposure of SeMet (50µM) and hypersaline conditions on BAX and CASP3A transcripts in Japanese medaka embryos after 12 and 24 hpf. A) Change in BAX and CASP3A expression between 12hpf and 24hpf. Expression of (B) BAX and (C) CASP3A in freshwater, saltwater, SeMet in freshwater and SeMet in saltwater. Each value represents the mean ± SE of 5-10 replicates. EF1-α was run as a housekeeping gene. Statistical significance is indicated by differing letters (Two-way ANOVA, Tukey HSD Test p ≤ 0.05).

Gene Expression of UPR Mediators

UPR gene expression changed from 12hpf to 24 hpf in medaka embryos (Figure

5.4). ATF6 gene expression decreased significantly from 12hpf to 24 hpf (0.3 fold, p=0.019), while BiP and ATF4 expression increased significantly (2 fold, p =0.048, and 4 fold, p = 0.008, respectively). BiP mRNA was increased in both saltwater and freshwater

SeMet treatments up to 39 fold over 12 hr controls and 107 fold over 12 hours controls in the SeMet freshwater and SeMet saltwater treatments, respectively. Though there was a trend towards an increase in ATF6 expresssion with SeMet treatment (p = 0.119), SeMet and hypersalinity did not significantly alter ATF6 expression. Similarly, trends in ATF4 expression indicated a potential difference between freshwater and saltwater treatment (p

= 0.07), with SeMet decreasing ATF4 in freshwater, yet increasing it in saltwater.

219

Figure 5.4. Effects of combined exposure of SeMet (50µM) and hypersaline conditions on BiP, ATF6, and ATF4 transcripts in Japanese medaka embryos after 12 and 24 hpf. A) Expression of BiP, ATF6 and ATF4 in whole embryos at 12hpf and 24hpf in freshwater. Expression of (B) BiP, (C) ATF6 and (D) ATF4 in 24hpf control embryos and SeMet treated embryos in freshwater and saltwater. Each value represents the mean ± SE of 5-10 replicates. EF1-α was run as a housekeeping gene. Statistical significance is indicated by differing letters (Two-way ANOVA, Tukey HSD Test p ≤ 0.05).

Discussion

The mechanism of action of SeMet toxicity is not well understood, particularly in the presence of multiple stressors, which may confound regulatory monitoring. We observed a decrease in hatch following treatment with hypersalinity and SeMet, while surviving embryos had deformities. Previous work at this developmental stage and SeMet concentration reported a significant decrease in embryo hatch following 24 hours of

50µM SeMet treatment in freshwater (Lavado et al., 2012). While, we did not observe this trend, our results were expected considering the duration of SeMet treatment was half

220 as that previously studied. Overall, our results are consistent with a plethora of other data reporting SeMet’s lethal and teratogenic effects in the field (e.g. Lemly, 2002).

Japanese medaka are a euryhaline species; adults are able to spawn and embryos hatch in full seawater (Inoue & Takei, 2002). We observed no significant difference in toxicity between freshwater and hypersaline controls, indicating that our results are not due to osmotic stress alone. Others have found salinity to potentiate SeMet toxicity in embryos

(Lavado et al., 2012). The mechanism behind this remains to be elucidated, however,

FMO may play a role. FMOs have been shown to oxygenate SeMet, which may contribute to its toxicity (Chen & Ziegler, 1994; Lavado et al., 2012). Several studies have found FMO activity is osmo-regulated (Schlenk et al., 1996; El-Alfy et al., 2002) and can increase under hypersaline conditions (Lavado et al., 2012). This increased FMO activity may increase SeMet oxygenation, which in turn may increase its embryo toxicity.

In contrast, studies have shown that SeMet activation can occur following methioninase generation of methylselenol (Spallholz et al., 2004), which can subsequently generate oxidative stress in rainbow trout embryos (Palace et al., 2004).

Methylselenol has also been implicated in induction of caspase-mediated apoptosis in cancer cell lines (Wang et al., 2002). However, considering that neither oxidative stress nor apoptosis was induced in SeMet and hypersaline treated embryos and that FMO has been found to be osmo-regulated, we conclude that this pathway is not a major contributor to the observed toxicity.

Of interest is that SeMet and hypersalinity did not generate oxidative stress as measured by lipid peroxidation after 12 hours of treatment. However, several groups,

221 including ours, have identified oxidative stress as one of the main modes of action of

SeMet toxicity (Lavado et al., 2012; Enright & Culberson, 2010; Palace et al., 2004;

Misra et al., 2015). As lipid peroxidation is an endpoint for severe oxidative stress,

TBARS is not sensitive to small changes in cellular redox (Hackett et al., 1988).

Furthermore, TBARS is a whole embryo measurement of oxidative stress, and does not consider localized effects. Thus, oxidative stress may still be occurring in SeMet and hypersaline treatments, yet it may not be detected by our assays or was not as high as that observed after 24 hr (Lavado et al., 2012). While our results do not eliminate oxidative stress as a mechanism of SeMet induced embryotoxicty, they indicate that other processes may play an important role particularly at the 12hour time point of exposure.

While we observed no difference in whole-embryo lipid peroxidation between 12hpf and

24hpf, many studies demonstrate that the redox status of embryos also undergoes great changes throughout development. Two contradicting reports demonstrate the changes in redox status of medaka embryos throughout development. Wu et al. (2011) measured changes in oxidative stress in the whole embryo each day post fertilization using

Dichloro-dihydro-fluorescein diacetate (DCHFDA; a dye that fluoresces following oxidation) and found overall reactive oxygen species (ROS) increased gradually until hatch. In contrast, another group studying medaka development and silver nanoparticle toxicity, found total ROS decreased throughout development (Wu & Zhou, 2012).

However, in addition to total ROS, both of these studies examined multiple biomarkers for oxidative stress and found no common patterns between them. Hence, the current

222 studies on redox status throughout medaka development confirm that further studies are necessary in order to understand this complex process.

Another adverse outcome pathway that may contribute to SeMet toxicity in medaka at this early life stage is the UPR. Methylselenic acid (MSA) induced the UPR in

PC-3 cells, a human prostate cancer cell line (Wu et al., 2005) and we observed increases in BiP expression following SeMet treatment, suggesting that SeMet may be disrupting

ER homeostasis. Although not significant, BiP mRNA expression was higher in SeMet and hypersaline treated embryos than in embryos in SeMet and freshwater. This would suggest that a greater UPR is being induced in SeMet and hypersaline conditions, a mechanism that requires further exploration. The trend towards alterations in ATF4 and

ATF6 gene expression also indicated a possible role for the PERK and ATF6 branches of the UPR in SeMet toxicity. PERK also activates Nrf2 (Cullinan & Diehl, 2006; Nair et al., 2007), a major transcription factor for antioxidant genes during oxidative stress. This provides a link between postulated mechanisms of oxidative stress and the UPR. It must be noted that each branch of the UPR has a different activation step (eg. ATF6 splicing,

PERK phosphorylation) (Cribb et al., 2005) according to varying time scales and specific types of ER stress (DuRose et al., 2006). DuRose et al (2006) compared UPR responses generated by dithriothreitol (DTT; disrupts disulfide bond formation), thapsigargin (Tg; inhibits ER Calcium-dependent ATPase) and tunicamycin (Tm; inhibits protein glycosylation). The ATF6 response to DTT was the most rapid of the three branches, yet it was significantly less sensitive to Tg and Tm (DuRose et al., 2006). In contrast, PERK responded most rapidly to calcium disruption caused by Tg. These responses occurred on

223 different time scales and at different magnitudes; overall, activation of the branches in response to DTT was faster and stronger than to Tm and Tg (DuRose et al., 2006). The varied time scales of each branch of the response to different types of ER stress indicate the necessity to document the UPR throughout development following SeMet and hypersaline treatment. Furthermore, as mentioned above concerning TBARS experiments, these gene expression studies cannot discern between localized effects.

Changes in transcription in a few key cells may not result in a distinct difference in fold change in qPCR.

The UPR plays a key physiological role in development (Cornejo et al., 2013).

The IRE1a-XBP1 pathway has been shown to regulate a variety of developmental processes, including plasma cell differentiation (Reimold et al., 2001), liver development

(Reimold et al., 2000), chondrogenesis (Han et al., 2013) and adipogenesis (Sha et al.,

2009). The PERK/ATF4 branch of the UPR has been found to be involved in osteoblast formation (Saito et al., 2011). BiP also plays a major role in development as BiP knockout mice are not viable after the peri-implantation stage (Weng et al., 2011). BiP has a much higher expression during neural development than during adulthood (Weng et al., 2011). The importance of these responses has been found to be conserved in the medaka (Ishikawa et al., 2011). Recent work established that ATF6a/b results in embryonic lethality in medaka, as in mice (Ishikawa et al., 2013). They found the physiological response to be strong at 2 days post fertilization, where it was localized to the brain, otic vesicle and notochord (Ishikawa et al., 2013). Perturbations in the physiological ER stress response during embryogenesis may result in the inability of the

224 embryo to manage the high demand for protein folding. If the perturbation is not stopped it could result in teratogenesis and embryo lethality. Lordosis, kyphosis and craniofacial abnormalities were the primary forms of teratogenesis documented in this study.

Considering the studies above illustrating the role of the UPR in the development of the notochord and cartilage, it is possible that UPR disruption may be generating the deformities witnessed here.

Apoptosis is one common outcome generated by the UPR (Hetz, 2012). We had hypothesized that apoptosis played a significant role leading to embryo lethality generated by SeMet. However, we observed no evidence of apoptosis in our embryos.

Neither BAX nor CASP3A gene expression were unchanged by SeMet treatment. Wu et al. (2005) demonstrated that 5µM MSA was able to cause apoptosis following UPR induction in PC-3 cells. However, MSA is highly redox reactive and not a natural form of

Se in the fish diet. BiP has been shown to be a negative regulator of apoptosis (Hetz,

2012), so the high levels of BiP mRNA induction observed may indicate a repression of apoptosis. Autophagy is another well-documented outcome from UPR (Benbrook &

Long, 2012), and may also contribute to the deformities and reduction of hatch observed.

The comparison of apoptosis taken at 12hpf and 24hpf clearly shows that these important processes fluctuate greatly during development. Indeed, apoptosis has been shown to fluctuate in tissue-specific patterns throughout Japanese medaka development, occurring mostly in the head, spinal column and tailbud (Iijima & Yokoyama, 2007).

Furthermore, apoptosis plays a key role in neural development (Nijhawan et al., , 2000).

225 The overwhelming evidence for changes in apoptosis, oxidative stress, and the UPR throughout development, indicates that developing Japanese medaka embryos have important windows of susceptibility to SeMet and hypersaline stress. Thus, while the roles of oxidative stress and apoptosis in the developmental toxicity of SeMet may be limited from 12-24 hours, they may be increased as the oxidation state of the embryos increases, or as apoptosis is more active at later stages of development. It is important to map these processes throughout embryogenesis, so that we may better understand the developmental toxicity of SeMet and hypersaline conditions.

In summary, hypersaline conditions derived from the San Joaquin River Valley,

CA, enhanced the toxicity of SeMet in the developing medaka embryo. While the UPR may have played a role, oxidative stress and apoptosis measured in the whole embryo was not associated with SeMet induced mortality and teratogenesis at this early stage.

Additional studies will further consider the role of oxidative stress and the UPR throughout medaka development and investigate developmental periods most susceptible to SeMet and hypersaline toxicity.

226 References

Benbrook, D.M., Long, A., 2012. Integration of autophagy, proteasomal degradation, unfolded protein response and apoptosis. Exp. Oncol. 34, 286-97.

Chen, G.P., Ziegler, D.M., 1994. Liver microsome and flavin-containing monooxygenase catalyzed oxidation of organic selenium compounds. Arch. Biochem. Biophys. 312, 566- 72.

Cornejo, V.H., Pihan, P., Vidal, R.L., Hetz, C., 2013. Role of the unfolded protein response in organ physiology: lessons from mouse models. IUBMB Life 65, 962-75.

Cribb, A.E., Peyrou, M., Muruganandan, S., Schneider, L., 2005. The endoplasmic reticulum in xenobiotic toxicity. Drug. Metab. Rev. 37, 405-442.

Cullinan, S.B., Diehl, J.A., 2006. Coordination of ER and oxidative stress signaling: The PERK/Nrf2 signaling pathway. Int. J. of Biochem. Cell B. 36, 317-332.

El-Alfy, A., Larsen, B., Schlenk, D., 2002. Effect of cortisol and urea on flavin monooxygenase activity and expression in rainbow trout, Oncorhynchus mykiss. Mar. Environ. Res. 54, 275-8.

Enright, C., Culberson, S.D., 2010. Salinity trends, variability, and control in the northern reach of the San Francisco Estuary. San Francisco Estuary and Watershed Science 7.

Farwell, A., Nero, V., Croft, M., Bal, P., Dixon, D.G., 2006. Modified Japanese medaka embryo-larval bioassay for rapid determination of developmental abnormalities. Arch. Environ. Contam. Toxicol. 51, 600-7.

Hackett, C., Linley-Adams, M., Lloyd, B., Walker, V., 1988. Plasma malondialdehyde: a poor measure of in vivo lipid peroxidation. Clin. Chem. 34, 208.

Han, X., Zhou, J., Zhang, P., Song, F., Jiang, R., Li, M., Xia, F., Guo, F. J., 2013. IRE1alpha dissociates with BiP and inhibits ER stress-mediated apoptosis in cartilage development. Cell Signal. 25, 2136-46.

Hetz, C., 2012. The unfolded protein response: controlling cell fate decisions under ER stress and beyond. Nat. Rev. Mol. Cell. Biol. 13, 89-102.

227 Hoffman, D. J., Heinz, G.H., LeCaptain, L.J., Eisemann, J.D., Pendleton, G.W., 1996. Toxicity and oxidative stress of different forms of organic selenium and dietary. Arch. Environ. Contam. Toxicol. 31, 120-7.

Iijima, N., Yokoyama, T., 2007. Apoptosis in the Medaka Embryo in the Early Developmental Stage. Acta Histochem. Cytoc. 40, 1-7.

Inoue, K., Takei, Y., 2002. Diverse Adaptability in Oryzias Species to High Environmental Salinity. Zool. Sci. 19, 727-734.

Ishikawa, T., Okada, T., Ishikawa-Fujiwara, T., Todo, T., Kamei, Y., Shigenobu, S., Tanaka, M., Saito, T. L., Yoshimura, J., Morishita, S., Toyoda, A., Sakaki, Y., Taniguchi, Y., Takeda, S., Mori, K., 2013. ATF6alpha/beta-mediated adjustment of ER chaperone levels is essential for development of the notochord in medaka fish. Mol. Biol. Cell. 24, 1387-95.

Ishikawa, T., Taniguchi, Y., Okada, T., Takeda, S., Mori, K., 2011. Vertebrate unfolded protein response: mammalian signaling pathways are conserved in Medaka fish. Cell. Struct. Funct. 36, 247-59.

Jentzsch, A.M., Bachmann, H., Furst, P., Biesalski, H.K., 1996. Improved analysis of malondialdehyde in human body fluids. Free Radical Bio. Med. 20, 251-6.

Kirchen, R.V., West, W.R., Carolina Biological Supply, C., 1976. The Japanese medaka: its care and development. Carolina Biological Supply Company.

Lavado, R., Shi, D., Schlenk, D., 2012. Effects of salinity on the toxicity and biotransformation of L-selenomethionine in Japanese medaka (Oryzias latipes) embryos: Mechanisms of oxidative stress. Aquat. Toxicol. 108, 18-22.

Lemly, A.D., 1997. A teratogenic deformity index for evaluating impacts of selenium on fish populations. Ecotox. Environ. Safe. 37, 259-66.

Lemly, A.D., 2002. Symptoms and implications of selenium toxicity in fish: the Belews Lake case example. Aquat. Toxicol. 57, 39-49.

Luoma, S.N., Rainbow, P.S., 2005. Why is metal bioaccumulation so variable? Biodynamics as a unifying concept. Environ. Sci. Technol. 39, 1921-31.

Miller, L.L., Wang, F., Palace, V.P., Hontela, A., 2007. Effects of acute and subchronic exposures to waterborne selenite on the physiological stress response and oxidative stress indicators in juvenile rainbow trout. Aquat. Toxicol. 83, 263-271.

228 Misra, S., Hamilton, C., Niyogi, S., 2012. Induction of oxidative stress by selenomethionine in isolated hepatocytes of rainbow trout (Oncorhynchus mykiss). Toxicol. In Vitro 26, 621-629.

Nair, S., Xu, C., Shen, G., Hebbar, V., Gopalakrishnan, A., Hu, R., Jain, M. R., Liew, C., Chan, J.Y., Kong, A.N., 2007. Toxicogenomics of endoplasmic reticulum stress inducer tunicamycin in the small intestine and liver of Nrf2 knockout and C57BL/6J mice. Toxicol. Lett. 168, 21-39.

Nijhawan, D., Honarpour, N., Wang, X., 2000. Apoptosis in neural development and disease. Annu. Rev. Neurosci. 23, 73-87.

Ramirez, P., Jr., Rogers, B. P., 2002. Selenium in a Wyoming grassland community receiving wastewater from an in situ uranium mine. Arch. Environ. Con. Tox. 42, 431-6.

229 Saito, A., Ochiai, K., Kondo, S., Tsumagari, K., Murakami, T., Cavener, D.R., Imaizumi, K., 2011. Endoplasmic reticulum stress response mediated by the PERK-eIF2(alpha)- ATF4 pathway is involved in osteoblast differentiation induced by BMP2. J. Biol. Chem. 286, 4809-18.

Schlenk, D., Peters, L.D., Livingstone, D.R., 1996. Correlation [corrected] of salinity with flavin-containing monooxygenase activity but not cytochrome P450 activity in the euryhaline fish (Platichthys flesus). Biochem. Pharmacol. 52, 815-8.

Schlenk, D., Zubcov, N., Zubcov, E., 2003. Effects of salinity on the uptake, biotransformation, and toxicity of dietary seleno-L-methionine to rainbow trout. Toxicol. Sci. 75, 309-313.

Schmittgen, T.D., Livak, K.J., 2008. Analyzing real-time PCR data by the comparative C(T) method. Nat. Protoc. 3, 1101-8.

Sha, H., He, Y., Chen, H., Wang, C., Zenno, A., Shi, H., Yang, X., Zhang, X., Qi, L., 2009. The IRE1alpha-XBP1 pathway of the unfolded protein response is required for adipogenesis. Cell. Metab. 9, 556-64.

Spallholz, J.E., Palace, V.P., Reid, T.W., 2004. Methioninase and selenomethionine but not Se-methylselenocysteine generate methylselenol and superoxide in an in vitro chemiluminescent assay: implications for the nutritional carcinostatic activity of selenoamino acids. Biochem. Pharmacol. 67, 547-54.

USEPA, 2004. Draft aquatic life water quality criteria for selenium. Office of Water and Office of Science and Technology, Washington D.C. http://water.epa.gov/scitech/swguidance/standards/criteria/aqlife/selenium/upload/comple te-2.pdf

Wang, Z., Jiang, C., Lü, J., 2002. Induction of caspase-mediated apoptosis and cell-cycle G1 arrest by selenium metabolite methylselenol. Mol. Carcinogen. 34, 113-120.

Wen, H., Carignan, J., 2007. Reviews on atmospheric selenium: Emissions, speciation and fate. Atmos. Environ. 41, 7151-7165.

Weng, W.C., Lee, W.T., Hsu, W.M., Chang, B.E., Lee,H., 2011. Role of glucose- regulated Protein 78 in embryonic development and neurological disorders. J. Formos. Med. Assoc. 110, 428-37.

Wu, M., Shariat-Madar, B., Haron, M.H., Khan, I.A., Dasmahapatra, A.K., 2011. Ethanol-induced attenuation of oxidative stress is unable to alter mRNA expression pattern of catalase, glutathione reductase, glutathione-S-transferase (GST1A), and superoxide dismutase (SOD3) enzymes in Japanese rice fish (Oryzias latipes) embryogenesis. Comp. Biochem. Physiol. C. Toxicol. Pharmacol. 153, 159-67.

230 Wu, Y., Zhang, H., Dong, Y., Park, Y. M., Ip, C., 2005. Endoplasmic reticulum stress signal mediators are targets of selenium action. Cancer. Res. 65, 9073-9.

Wu, Y., Zhou, Q., 2012. Dose- and time-related changes in aerobic metabolism, chorionic disruption, and oxidative stress in embryonic medaka (Oryzias latipes): underlying mechanisms for silver nanoparticle developmental toxicity. Aquat. Toxicol. 124-125, 238-46.

Zhao, S., Fernald, R.D., 2005. Comprehensive algorithm for quantitative real-time polymerase chain reaction. J. Comput. Biol. 12, 1047-64.

231 Chapter 6: Molecular Mechanisms of Selenium-Induced Spinal Deformities in Fish

Abstract

Selenium toxicity to oviparous vertebrates is often attributed to selenomethionine

(SeMet), which can biomagnify through maternal transfer. Although oxidative stress is implicated in SeMet toxicity, knowledge gaps remain in how SeMet causes characteristic spinal deformities. In the present study, we use the Japanese medaka (Oryzias latipes) model to investigate the role of oxidative stress, cell death, and the unfolded protein response (UPR) on skeletal gene expression and SeMet toxicity, linking localization of cellular effects to observed abnormalities. Medaka embryos were treated with 2.5µM or

5µM SeMet for 24hr at stage 25 (48 hours post fertilization). Post treatment, embryos were separated into normal, deformed (mild, moderate or severe), or dead categories.

Dichlorofluorescein staining demonstrated oxidative stress in tails of embryos with observable spinal malformations. Furthermore, acridine orange staining for apoptosis identified significantly more dead cells in tails of treated embryos. Gene expression studies for the UPR suggest a potential role for CHOP (c/ebp homologous protein) induced apoptosis deformed embryos after 5µM SeMet, accompanied by a significant decrease in PDIA4 (protein disulfide isomerase A4) and no change in Dnajb9 (ER DNA J

Domain-Containing Protein 4). This expression was distinct from the UPR induced by well-studied ER stress inducer, tunicamycin, which robustly activated CHOP, PDIA4 and

Dnajb9. Finally, SeMet treatment significantly decreased transcripts of cartilage development, Sox9 (SRY box 9) and Collagen 2a1 mRNA, while increasing Runx2 in deformed embryos, without altering Twist. Results suggest that oxidative stress, the UPR

232 and cell death play key roles in SeMet induced deformities and altered skeletal development factors.

Introduction

Selenium (Se) is an essential micronutrient with a narrow margin between essentiality and toxicity to oviparous vertebrates. Although Se is present naturally in soils, anthropogenic disturbance can release Se into waterways. Free waterborne Se is bioaccumulated at low trophic levels, and integrated into selenomethionine (SeMet), which is then incorporated non-specifically into proteins (Fan et al., 2002). Thus, vertebrate consumption of Se in the diet is often in the form of SeMet. SeMet can be maternally transferred in vitellogenin to developing embryos, where it exerts developmental toxicity causing teratogenesis and mortality. Common abnormalities from

Se include spinal deformities, such as lordosis, cranio-facial abnormalities, and fin deformities (Lemly, 1997). In particular, skeletal deformities are characteristic of Se toxicity.

Speculations about the mechanism of toxicity have suggested a role for oxidative stress in Se embryo toxicity (Lavado et al., 2011; Palace et al., 2004; Misra et al., 2012;

Arnold et al., 2016). However, most studies were performed with high concentrations of

SeMet, only a few link oxidative stress to an adverse outcome, and none have showed it occurring at the sites of malformations. Furthermore, other research suggests that oxidative stress is not the only molecular and cellular disturbance caused by SeMet

(Kupsco & Schlenk, 2014).

233 Oxidative stress has been linked to the unfolded protein response (UPR) in several studies (Cao & Kaufman, 2014). The UPR is an integrated stress response activated by an increase in unfolded proteins in the endoplasmic reticulum (ER), which causes ER stress.

The response induces increases in protein folding capacity, translational attenuation, mRNA degradation, and proteolysis (Hetz, 2012). Activation of the UPR occurs when the master regulatory chaperone BiP dissociates from the three branches, ATF6 (Activating transcription factor 6), PERK (PKR-like endoplasmic reticulum kinase) and IRE1

(Inositol requiring enzyme 1), each with some independent function. IRE1 is primarily responsible for degradation of mRNA and proteins. PERK is responsible for translational attenuation and ATF6 for an increase in folding capacity (Hetz, 2012). If the stress remains uncorrected, the UPR will initiate apoptosis via C/EBP homologous protein

(CHOP). Apoptosis is programmed cell death that is also a common outcome from oxidative stress, which may affect developing chondrocytes. The UPR is further connected to skeletal development through the PERK pathway, and PERK and ATF4

(Activating transcription factor 4) knockout models result in osteopenia and reduced bone mineralization and collagens (Zhang et al., 2002; Yang et al., 2004).

In the present study, we further characterized mechanisms of SeMet-induced developmental toxicity with the model organism, the Japanese medaka (Oryzias latipes).

We investigated the localization of oxidative stress and apoptosis in malformations following SeMet exposure and examined gene expression of UPR and skeletogenesis genes. We chose downstream UPR target genes PDIA4 (protein disulfide isomerase associated 4), which is involved in disulfide bond formation and activated via ATF6;

234 CHOP, which is involved in UPR induced apoptosis through PERK and ATF4; and

Dnajb9 (also ERdj4), a chaperone involved in ER associated degradation. Skeletal development is regulated by various chondrogenic and osteogenic factors. In the present study, we focused on SRY-box protein 9 (Sox9), collagen 2a1 (Col2a1), a collagen secreted in large amounts during chondrogenesis; Runx2 (Runt-related transcription factor 2), a transcription factor critical to osteogenesis; and Twist (Twist-related protein

1), a BHLH transcription factor responsible for repression Runx2-induced osteogenesis until the end of chondrogenesis. We hypothesized that acute SeMet exposure at stage 25 would result in an increase in oxidative stress, and the UPR leading to apoptosis and mis- regulation of cartilage and bone formation.

Methods

Embryo Collection and Exposures

Japanese medaka (Oryzias latipes) were maintained at a ratio of 4:6 males:females, on a 14 hour light to 10 hour dark cycle. Fish were fed daily a diet of live brine shrimp at University of California-Riverside in accordance with animal use protocols (AUP # 20140002). Embryos were collected into de-chlorinated freshwater and rinsed thoroughly. At stage 25 (48hours post fertilization; Iwamatsu, 2004) embryos (15-

25/replicate) were separated into controls or treated with 5µM or 2.5µM seleno-L- methionine (Sigma Aldrich, 98% purity) for 24 hours (until stage 29). Previous studies on

Japanese medaka treated with 5µM SeMet for 24hrs at stage 25 demonstrated an embryo

Se content of 24µg/g dry weight (Kupsco & Schlenk, 2016). Stage 25 was chosen as the

235 treatment stage because it was specific for Se induced lordosis, and there was a significant correlation between Se content and deformities (Kupsco & Schlenk, 2016).

Post treatment, embryos were rinsed and examined under a microscope for deformities.

Embryos were characterized as either normal (indistinguishable from controls), mildly deformed (some abnormality in the tip of the tail), moderately deformed (abnormality in the entire tail), severely deformed (abnormalities to tail, body and head), or dead

(embryos lacking a heart beat). They were then separated for analysis of oxidative stress, apoptosis, the UPR and skeletal gene expression.

UPR positive controls were also treated for gene expression comparison.

Canonical UPR-inducer, Tunicamycin (Tm; a glycosylation inhibitor) (Sigmal Aldrich), was dissolved in a stock concentration of 2mg/ml in DMSO. Embryos were treated at stage 25 with a 0.2% DMSO control, 4µg/ml Tm in 0.2% DMSO, 2µg/ml Tm in 0.2%

DMSO and 1µg/ml Tm in 0.1% DMSO for 24 hours (10-15 embryos/ replicate). No toxicity was observed in DMSO controls, 1µg/ml Tm, or 2µg/ml Tm treatments. Embryos treated with 4µg/ml Tm displayed small lesions in the end of the tail, similar to the observed deformities following SeMet exposure. Embryos were frozen and pooled for analysis of UPR and skeletal gene expression (n = 6).

Oxidative Stress

Control embryos, and normal, mildly deformed, moderately deformed, and severely deformed embryos treated with 2.5µM and 5µM SeMet were incubated in

10µg/ml 2’,7’-Dichlorofluorescin diacetate (DCFDA) (97%, sigma Aldrich) for 1hr, on a shaker in the dark at room temperature. Embryos were rinsed 2x10mins in freshwater and

236 10mins in 300µg/ml tricaine. Embryos were imaged with a SPOT Pursuit camera under a green fluorescent filter on a Leica MZIII Pursuit stereoscope. Embryos were scored for the presence or absence of staining of the tail or yolk sac (n = 7-20).

Acridine orange staining for apoptosis

As acridine orange was unable to penetrate the chorion, deformed embryos from

5µM and 2.5µM SeMet treatments and controls were dechorionated according to

Porazinski et al. 2010. In brief, embryos were rolled on fine grit sandpaper to remove hairs on chorion exterior, then incubated with 1mg/mL protease for 1 hour. Embryos were rinsed and incubated with 20% hatching enzyme for 10-30 mins, under observation.

Dechorionated embryos were rinsed and incubated in 2µg/ml acridine orange (0.2%

DMSO) for 30mins at room temperature on shaker in the dark. After incubation, embryos were rinsed 2x10mins in freshwater and 10mins in 300µg/ml tricaine. The dechorionation process is harsh due to the use of proteases and thus only controls, and normal and mildly deformed embryos survived for imaging. Embryos were mounted in 4% methylcellulose for visualization with a 488nm filter on a fluorescent Nikon Eclipse Ti microscope. The number of cells in the tail up to the yolk sac attachment was counted on live embryos.

Images were analyzed with NIS Elements AR 3.0 software (n = 20 controls and 8 treated).

Gene expression analysis

Following deformities analysis, dead embryos were discarded and embryos were pooled into control, normal, or deformed, and frozen for gene expression analysis (n =10-

17). The Lipid Tissue RNeasy kit (Qiagen, Valencia, CA) was used to isolate total

237 mRNA from pooled embryo samples according to the manufacturers instructions. mRNA quality and quantity was assessed with the ND-1000 (Nanodrop, Wilmington, DE). cDNA was prepared with 1µg mRNA using the Reverse Transcription System (Promega

Corporation, Madison, WI), according to the manufacturers instructions.

Primers were designed using Integrated DNA Technologies PrimerQuest software and optimized using PCR Miner (Zhao et al., 2005) (Table 6.1). EF1α was run as a housekeeping gene. qPCR was performed with the iScript One-step RT-PCR kit with

SYBR Green from Bio-Rad (Hercules, CA), omitting the reverse transcriptase, on a

MyiQ Thermo cycler (Biorad). The samples were denatured and the polymerase activated at 95 °C for 5 min, then 40 cycles of 10s at 95 °C and 30s of 55 °C. Samples were subjected to melting curve analysis from 65 to 85 °C in 0.5 °C increments with continuous fluorescence measurement. qPCR was analyzed according to Schmittgen &

Livak (2008) and fold change was calculated against the controls.

238 Table 6.1. Primer sequences and accession numbers for genes evaluated in this study, along with the concentrations used for each primer pair.

Gene Primers Accession # Concentration

Sox9b GGATCATCCCGACTATAAA AY870393.1 0.5 µM TGTGACTGACCTGAATG Col2a1 TGTTGGCAGAGTTGGGAATAG BJ518303.1 0.5 µM AAGAGTGGTGACTTCTGGATTG Twist GCCAGATACATCGACTTCCTTT NM_001104707.1 0.5 µM CGTACCATGTTGGGTAGGATTT RunX2 CTCTGGGAAGATGAACGATGTG NM_001104850.1 2.5 µM CAGGACCGAGCAAAGAAAGT EF1a CTACATCAAGAAGATCGGCTACAA NM_001104662.1 2.5 µM CGACAGGGACAGTTCCAATAC CHOP CTCCTCTTCAGCATCCCAATC NM_001278882.1 2.5 µM GCCTTTCGATCTCTGCCTTTA Dnajb9 CCTGCTCATATCCGAGTTCATC XM_004082887.2 2.5 µM CGTACTCTCTTCTTCTCTTGTCATC PDIA4 AAAGACAACGATCCTCCCATAC XM_004081002.2 2.5 µM CTCCTTTGCTGCCTTCTCATA

Statistical Analysis

Statistical significance was assessed using the statistical package R and analysis varied depending on the type of data. Deformities analysis scored embryos as binary data of one of 5 outcomes (normal, mild, moderate, severe, or dead) and was analyzed with multiple chi-square tests for desired comparisons. P-values were corrected with the

Holm-Sidak method. Similarly, the presence or absence of DCFDA staining is also binomial data. Because only comparisons to controls were deemed necessary and the number of observations was less than 5, a chi-square test was not appropriate and significance was assessed with a fishers-exact test and corrected with a Holm-Sidak correction for the number of P-values. The number of cells stained with acridine orange is ordinal data and significance was assessed with a Kruskall-Wallis test followed by a

Dunn’s test post hoc. Finally, gene expression results were analyzed as delta Ct and were

239 normally-distributed, continuous data. These results were analyzed with a two-way

ANOVA for SeMet treatment and deformity status as the factors. A Tukey’s test was performed post hoc. Statistical significance was determined for p ≤ 0.05 for all tests.

Results

Observed effects

SeMet caused a range in the severity of deformities (Figure 6.1). No difference in outcome was observed between 2.5µM and 5µM SeMet for any category, while outcomes from both 2.5µM and 5µM were significantly different from controls in all categories (p

< 0.0001). Treatments with 2.5µM and 5µM SeMet resulted in approximately, 44% normal embryos, 27% mildly deformed, 7% moderately deformed, 7% severely deformed and 10% dead embryos. With the exception of numbers of moderately and severely deformed embryos, which were not significantly different from each other, there were significantly different proportions of embryos in each category.

240

Figure 6.1. Effects of 24 hours of 2.5µM (striped bars) and 5µM SeMet (black bars bars) on survival and deformities of Japanese medaka embryos at stage 25 against controls (white bars). Embryos were scored as dead (lacking heartbeat), severely deformed, moderately deformed, mildly deformed or normal. Differing letters represent significant differences between categories at a single concentration, while differing symbols represent difference between concentrations in a single category following multiple chi- square tests with corrected p values (p ≤ 0.05).

Oxidative stress

DCFDA is fluorescent upon oxidation by reactive oxygen species (ROS) and allows measurements of localized oxidative stress. All embryos displayed characteristic staining of developing melanocytes in the head and trunk; however, no other aberrant staining was detected in controls. In SeMet treated embryos, fluorescence was detected in abnormalities in the tail and in visible yolk sac lesions (Figure 6.2A, B, C). DCFDA staining was not detected in controls or embryos appearing normal post treatment. For a quantitative measurement, staining was scored as present or absent on the tail and yolk sac of each embryo. Staining of the yolk sac was present in 60% of embryos with moderate deformities following treatment with 2.5µM SeMet (p = 0.0004 against

241 controls) in contrast to only 25% of embryos treated with 5µM SeMet (not significant from control). Oxidative stress-induced lesions on the yolk sac increased to 82% in severely malformed embryos with 2.5µM SeMet (p = 0) and 88% in severely deformed embryos treated with 5µM SeMet (p = 0). Oxidative stress in the tail was significantly greater than controls in mildly (35%, p = 0.0042), moderately (92%, p = 0) and severely

(62%; p = 0.0006) deformed embryos treated with 5µM SeMet. However, only severely deformed embryos treated with 2.5µM SeMet had significantly more oxidative stress in the tails in comparison to controls (45%, p = 0.003).

242

Figure 6.2. Representative images of staining of SeMet treated embryos with 10µg/ml DCFDA and imaged on a stereoscope. A. Control (n = 20); B. 2.5µM normal (n =17); C. 2.5µM mildly deformed (n = 21); D. 2.5µM moderately deformed (n = 10); E. 2.5µM severely deformed (n = 11); F. 5µM normal (n = 18); G. 5µM mildly deformed (n = 20); H. 5µM moderately deformed (n = 12); I. 5µM severely deformed (n = 8). J. Proportion of embryos stained with DCFDA in the tail (white bars) and yolk sac (black bars). Asterisks represent significant difference from control at p ≤ 0.05 following fisher’s exact test and a holm-sidak correction.

The Unfolded Protein Response

Following SeMet treatments, embryos were separated under a microscope into

Normal or Deformed. These categories were used to reduce variability from combining normal and deformed embryos. UPR target genes CHOP (also called DDIT3), PDIA4

(also Erp72), and Dnajb9 (also Erdj4) were chosen for analysis as they are downstream

243 targets of different branches of the UPR: PERK, ATF6 and IRE1, respectively (Shoulders et al., 2013). A trend in an increase in CHOP was observed in 5µM treated deformed embryos (p = 0.057) (Figure 6.3A). In contrast, both SeMet treatments significantly reduced PDIA4 expression in normal and deformed embryos. (Figure 6.3B). No alternations in Dnajb9 expression were observed (Figure 6.3C). Treatments with the positive control Tm significantly upregulated expression of all genes with 2µg/ml and

4µg/ml treatments to levels greater than SeMet treatments (Figure 4A).

Figure 6.3. Effects of 24 hours of 2.5 and 5µM SeMet on mRNA expression of A. CHOP, B. PDIA4 and C. Dnajb9 as fold change over controls ± standard error (SE). Embryos were separated into Normal Controls (n = 15) and Normal treated (n = 5-7; white bars) and Deformed (n = 10-11; Black bars). Differing letters represent significant differences between groups following ANOVA and Tukeys’s HSD test post hoc (p ≤ 0.05).

244

Figure 6.4. Effects of 24 hours of 1, 2, and 4µg/ml tunicamycin (Tm) on mRNA expression of (A) CHOP (white bars), PDIA4 (black bars) and Dnajb9 (stripped bars); and (B) Sox9 (black bars), Col2a1 (stripped bars), Runx2 (white bars) and Twist (gray bars) as fold change over controls ± standard error (SE) (n = 5-8). Differing letters represent significant differences between groups following ANOVA and Tukeys’s HSD test post hoc (p ≤ 0.05).

Skeletal Development

Selenomethionine causes characteristic lordosis in fish and aquatic birds. This suggests that factors involved in skeletal development may be altered in response to cell death or the UPR. In the present study we chose to investigate expression of Sox9b, the primary factor involved in chondrogenic differentiation, Col2a1, the major collagen in cartilage, Runx2, a transcription factor important in initiation of bone development, and

Twist, a Runx2 inhibitor that controls the transition from cartilage to bone.

Sox9 expression was significantly decreased in deformed embryos (p = 0.03) and embryos treated with 5µM SeMet had significantly less Sox9 expression than controls (p

= 0.05) (Figure 6.5A). This change correlated with a significant increase in Runx2 expression in deformed embryos exposed to 5µM SeMet (p = 0.03) (Figure 6.5C).

245 Finally, no alterations in Col2a1 or Twist were observed in comparison to controls, however, a trend towards a decrease in Col2a1 was observed in 2.5µM treated embryos

(p = 0.054) and a significant effect of SeMet was detected by 2-way ANOVA for Twist expression (Figure 6.5B, 5D). The Tm positive controls were also tested for changes in skeletal gene expression. No significant differences from DMSO controls were observed for Col2a1, Runx2 or Twist (Figure 6.4B). Sox9 decreased significantly from all other treatments following exposure to 4µg/ml Tm (p = 0.002).

Figure 6.5. Effects of 24 hours of 2.5 and 5µM SeMet on mRNA expression of A. Sox9, B. Col2a1, C. Runx2, and D. Twist as fold change against controls ± standard error (SE). Embryos were separated into Normal Controls (n = 15) and Normal treated (n = 5-7; white bars) and Deformed (n = 10-11; Black bars). Differing letters represent significant differences between groups following ANOVA and Tukeys’s HSD test post hoc (p ≤ 0.05).

246 Apoptosis

Acridine orange has been shown to stain apoptotic cells. Apoptosis is a normal process during embryonic development, and thus a number of apoptotic cells were observed in control embryos. In controls, stained cells were spread evenly throughout the body and head with a concentration in the tip of the tail. The number of stained cells were quantified from tail tip to the fusion with the yolk sac (Figure 6.6). A dose-dependent increase in the number of apoptotic cells in the tails of SeMet treated embryos was observed. Control embryos had an average of 88 ± 6.7 cells stained, while deformed embryos treated with 2.5µM SeMet had 114 ± 10.3 (p = 0.04) and embryos treated with

5µM had an average of 196 ± 20.4 apoptotic cells (p = 0.0005).

247

Figure 6.6. Representative images of tails of Japanese medaka embryos stained with 2µg/ml acridine orange following SeMet treatments. A. Control (n = 19), B. 2.5µM deformed (n = 8), C. 5µM deformed (n = 7), D. The average number of cells ± SE stained from tail tip to yolk sac. Differing letters represent significant differences between treatments following ANOVA and Tukeys’s HSD test post hoc (p ≤ 0.05).

Discussion

This study is the first to investigate the localization of cellular and molecular effects of SeMet and to examine these effects in relation to differentiation and formation of skeletal structures. We demonstrate that oxidative stress and cell death occur at the sites of SeMet induced malformations in Japanese medaka embryos. We further establish the potential involvement of the UPR in these processes, and connect these changes to alterations in expression of key skeletogenesis genes.

248 Oxidative stress has long been a postulated mechanism for SeMet toxicity to fish embryos. SeMet induced superoxide in the presence of methioninase in vitro and in trout embryos (Spallholz et al., 2004; Palace et al., 2004) and significantly reduced total glutathione concentrations in Japanese medaka embryos, zebrafish embryos, and rainbow trout hepatocytes (Lavado, 2012; Arnold et al., 2016, Misra et al., 2012). However, previously these alterations were not quantitatively linked to deformities or mortality.

Recently, the addition of antioxidant N-acetyl cysteine to 400µg/ml SeMet treatments in zebrafish embryos was found to significantly reduce SeMet induced deformities to control levels (Arnold et al., 2016). This corresponded with increased total glutathione and an increased ratio of reduced to oxidized glutathione back to control levels with

SeMet co-exposures (Arnold et al., 2016). This demonstrates that oxidative stress is responsible for SeMet induced deformities in zebrafish embryos.

While research on oxidative stress is crucial to advance our understanding of

SeMet-induced toxicity, oxidative stress is likely only one event responsible for observed malformations. In the current study, we assess the contributions of the UPR and oxidative stress as mechanisms of SeMet toxicity. Previous research has implicated a role of the

UPR in SeMet embryo toxicity to Japanese medaka through induction of the main regulatory chaperone BiP (Kupsco & Schlenk, 2014). Additionally, methyl-selenic acid

(MSA) was found to induce all three branches of the UPR in PC-3 cells (Zu et al., 2006).

Here we sought to further define the pathway through examination of the expression of downstream UPR target genes: CHOP, PDIA4 and dnajb9. We also used a global UPR inducer, tunicamycin, as a positive control. We found that 5µM SeMet treatment

249 produced a trend towards an increase in CHOP in deformed embryos. In contrast. SeMet significantly decreased PDIA4 expression in normal and deformed embryos, while expression in dead embryos was equal to controls. Treatment with Tm however, resulted in a dose dependent increase in the expression of all UPR target genes.

A large body of evidence links the UPR to oxidative stress in a positive feedback loop (Cao & Kaufman, 2014). For instance, oxidative stress can activate the UPR via disruption of disulfide bonds and inhibition of Calcium ATPase (Van der Vlies et al.,

2003). The UPR can subsequently further oxidative stress via electron leakage from the protein disulfide isomerase reductants, ER oxidoreductins, such as ERO1 (Tu &

Weissman, 2004). Additionally, antioxidant protein Nrf2 (Nuclear factor (erythroid- derived 2)-like 2) is a PERK substrate (Cullinan et al., 2003) and has been shown to interact with ATF4 in vitro to induce heme oxygenase 1 (He et al., 2001). Furthermore, many studies link oxidative stress and the UPR in a variety of pathogenic diseases (Cao

& Kaufman, 2014).

In the present study, possible CHOP mRNA in deformed embryos treated with

5µM SeMet suggests a role for the PERK/ATF4 pathway. CHOP is a transcription factor activated downstream of ATF4, which has been associated with apoptosis through a number of pathways. Furthermore, CHOP and ATF4 overexpression were found to induce oxidative stress prior to apoptosis, and knockdown of ERO1 reduced these effects

(Han et al., 2013). Another metalloid, inorganic arsenic, caused oxidative stress, and induced the UPR to increase CHOP expression as well as apoptosis via the mitochondrial pathway (Lu et al., 2014).

250 In contrast to CHOP, PDIA4 expression decreased in SeMet treated embryos.

PDIA4 is a protein disulfide isomerase under control of the ATF6 branch of the UPR

(Shoulders et al., 2013). PDIA4 protein levels have been shown to increase over time with Tm and thapsigargin, another potent UPR inducer (Mintz et al., 2008). However, the redox activity of PDIA4 suggests that it may be differentially regulated during oxidative stress induced UPR. Model studies on glucotoxicity and diabetes found that ribose treated islet cells had increased CHOP expression and decreased PDIA4 (Wali et al., 2014). In another study, cell death increased in adenocarcinoma cells with PDIA4 knockdown, which was correlated with increased caspase 3/7 activity and caspase 9 cleavage (Tufo et al., 2014). In contrast, PDIA4 expression was shown to increase following ablation of

PERK in 832/13 beta cells (Feng et al., 2009). These results support the hypothesis of

PDIA4 inhibition during activation of the PERK pathway, which is consistent with our results.

Dnajb9 (ERdj4) is a heat shock protein chaperone regulated by the UPR. Dnajb9 mRNA expression was significantly induced by activation of XBP1 of the IRE1 branch, but not by the ATF6 branch (Shoulders et al., 2013). However, other studies on ATF4 knockout mouse embryonic fibroblasts reported that ATF4 may also upregulate Dnajb9

(Han et al., 2013). However, no alterations in Dnajb9 were observed following SeMet treatment.

In comparison to Tm treatments, UPR activation by SeMet did not follow a global

UPR response. This suggests that SeMet may not be a robust inducer of the UPR, but instead may activate the pathway at low levels. Complex timing of the UPR can

251 complicate observed effects. For instance, treatment of mouse embryonic fibroblasts with

Tm increased eIF2a phosphorylation (translational inhibition) after 1-2 hours, followed by sequential activation of ATF4 and CHOP, while caspase 3 was not activated until 24 hours (Han et al., 2013). Furthermore, different types of stressors can activate the three branches of the UPR independently, as dithiothreitol (disulfide bond reductant) activated

ATF6 much faster than Tm or thapsigargin (Ca2+ ATPase inhibitor), and thapsigargin induced IRE1 more strongly than PERK or ATF6 (DuRose et al., 2006). Thus, time course studies may be necessary to elucidate the role of the UPR in SeMet toxicity.

Finally, skeletogenesis was examined via expression of key factors regulating bone and cartilage development. Bone development in fish is similar to mammals, however, fish bones are formed in a process called perichondral formation, during which the chondrocytes do not commit apoptosis and are instead incorporated into perichondral ossification. Despite this difference, many key factors for differentiation of bone and cartilage are conserved in Japanese medaka (Renn et al., 2006), including those examined here, Col2a1, Twist and Runx2. Collagen 2a1 is the major component of the collagen matrix and is induced by transcription factor SRY box 9 (sox9) (Bell et al., 1997). Sox9 and Col2a1 are co-expressed in developing zebrafish embryos (Chiang et al., 2001). In

Japanese medaka, TCDD has been shown to disrupt hypural skeletogenesis, accompanied by decreased Col2a1 expression (Dong et al., 2012). In the current study, Sox9 expression decreased significantly in deformed embryos following 5µM SeMet treatment, this suggests that cartilage formation may be disrupted by SeMet treatment.

252 In addition to Sox9 and Col2a1, we also examined Runx2 expression as an indicator of bone development. At stage 25, Runx2 is expressed anterior to the first myotome and by stage 30 it is expressed throughout the embryonic head (Inohaya &

Kudo, 2007). An increase in Runx2 expression coupled with a decrease in Sox9 could indicate premature bone development following SeMet exposures.

Lastly, Twist expression was analyzed in this study. Twist is a Runx2 inhibitor, and suppresses Runx2 by interacting with its DNA binding domain (Bialek et al., 2004).

Twist is expressed during early vertebrae formation. By stage 25, Twist is expressed in medaka sclerotomal cells of the trunk at the ventromedial part of the somites, and medaka

Twist knockouts do not develop neural arches (Yasutake et al., 2004). SeMet failed to alter Twist expression, suggesting that changes in Runx2 expression are not due to decreased Twist.

The UPR is involved in bone development (Saito et al., 2011). PERK and ATF4 knockout mice display skeletal dysplasia, reduced bone mineralization and reduced growth (Zhang et al., 2002; Yang et al., 2004). Furthermore, chondrodysplasias caused by mutations in extracellular matrix proteins are often mediated through the UPR leading to reduced chondrocyte proliferation and increased apoptosis (Arnold & Fertala, 2013;

Patterson and Dealy, 2014). Collagens are decreased during ER stress (Mintz et al.,

2008), which is consistent with our SeMet treatments. Tm significantly decreased Sox9, suggesting that the UPR may also impact chondrogenesis. There was also a trend towards an increase in Runx2 after 4µg/ml Tm. The significant decrease in Sox9 and potential increase in RunX2 induced by Tm is similar to the gene expression pattern observed in

253 SeMet treated embryos. However, the lack of a robust UPR response suggests that the

UPR may not be exclusively responsible for the observed changes in gene expression.

Apoptosis of proliferating chondrocytes may be contributing; however, further studies are needed to confirm this hypothesis.

In conclusion, we have demonstrated that oxidative stress and cell death occur in the tails of malformed embryos treated with 2.5µM and 5µM SeMet. Furthermore, a unique activation pattern of the UPR occurred potentially through the PERK pathway.

Finally, these changes were accompanied by alterations in skeletogenesis. Future research will determine more precise UPR activation patterns and further elucidate the molecular mechanisms of SeMet’s effects on chondro- and osteogenesis. This research is environmentally relevant considering wide-spread Se contamination across the US and may provide insights into studies on similar toxicants causing skeletal abnormalities.

254 References

Arnold, M., Forte, J.E., Osterberg, J.S., Di Giulio, R.T. 2016. Antioxidant Rescue of Selenomethionine-Induced Teratogenesis in Zebrafish Embryos. Arch. Environ. Contam. Toxicol. 70, 311-320.

Arnold, W.V., Fertala, A., 2013. Skeletal diseases caused by mutations that affect collagen structure and function. Int. J. Biochem. Cell Biol. 45, 1556-1567.

Bell D.M., Leung K.K., Wheatley S.C., Ng L.J., Zhou S., Ling K.W., Sham M.H., Koopman P., Tam P.P., Cheah K.S.,1997. SOX9 directly regulates the type-II collagen gene. Nat. Genet. 16, 174–178.

Bialek, P., Kern, B., Yang, X., Schrock, M., Sosic, D., Hong, N., Wu, H., Yu, K., Ornitz, D.M., Olson, E.N., Justice, M.J., Karsenty, G., 2004. A twist code determines the onset of osteoblast differentiation. Dev. Cell 6, 423–435.

Cao S.S., Kaufman R.J., 2014. Endoplasmic reticulum stress and oxidative stress in cell fate decision and human disease. Antioxid. Redox Signal. 21, 396–413

Chiang, E.F., Pai, C.I., Wyatt, M., Yan, Y.L., Postlethwait, J., Chung, B., 2001. Two Sox9 genes on duplicated zebrafish chromosomes: expression of similar transcription activators in distinct sites. Dev. Biol. 231, 149–163.

Cullinan, S.B., Zhang, D., Hannink, M., Arvisais, E., Kaufman, R.J., Diehl, J.A., 2003. Nrf2 is a direct PERK substrate and effector of PERK-dependent cell survival. Mol. Cell. Biol. 23, 7198-7209

Dong, W., Hinton, D.E., Kullman, S.W., 2012. TCDD disrupts hypural skeletogenesis during medaka embryonic development. Toxicol. Sci. 125, 91-104.

Feng, D., Wei, J., Gupta, S., McGrath, B.C., Cavener, D.R., 2009. Acute ablation of PERK results in ER dysfunctions followed by reduced insulin secretion and cell proliferation. BMC Cell. Biol. 10, 61

255 Han, J. Back, S.H., Hur, J., Lin, Y.H., Gildersleeve, R., Shan, J., Yuan, C.L., Krokowski, D., Wang, S., Hatzoglou, M., Kilberg, M.S., Sartor, M.A., Kaufman, R.J., 2013. ER- stress-induced transcriptional regulation increases protein synthesis leading to cell death. Nature Cell Biol. 15, 481–490.

He, C.H., Gong, P., Hu, B., Stewart, D., Choi, M.E., Choi, A.M., Alam, J., 2001. Identification of activating transcription factor 4 (ATF4) as an Nrf2-interacting protein. Implication for heme oxygenase-1 gene regulation. J. Biol. Chem. 276, 20858-20865.

Hetz, C., 2012. The unfolded protein response: Controlling cell fate decisions under ER stress and beyond. Nat. Rev. Mol. Cell. Biol. 13 (2), 89−102.

Inohaya, K., Kudo, A., 2000. Temporal and spatial patterns of cbfal expression during embryonic development in the teleost, Oryzias latipes. Dev. Genes Evol. 210, 570–574.

Iwamatsu, T., 2004. Stages of normal development in the medaka Oryzias latipes. Mech. Develop. 12, 605–618.

Kupsco, A., Schlenk, D., 2014. Mechanisms of Selenomethionine Developmental Toxicity and the Impacts of Combined Hypersaline Conditions on Japanese Medaka (Oryzias latipes). Environ. Sci. Technol. 48, 7062-8

Kupsco, A., Schlenk, D., 2016. Stage Susceptibility of Japanese Medaka (Oryzias Latipes) to Selenomethionine and Hypersaline Developmental Toxicity. Environ. Toxicol. Chem. 35, 1247–1256.

Lavado, R., Shi, D., Schlenk, D., 2011. Effects of salinity on the toxicity and biotransformation of L-selenomethionine in Japanese medaka (Oryzias latipes) embryos: Mechanisms of oxidative stress. Aquat. Toxicol. 108, 18−22.

Lemly, A.D., 1997. A teratogenic deformity index for evaluating impacts of selenium on fish populations. Ecotoxicol. Environ. Saf. 37 (3), 259−66.

Lu, T.H., Tseng, T.J., Su, C.C., Tang, F.C., Yen, C.C., Liu, Y.Y., Yang, C.Y., Wu, C.C., Chen, K.L., Hung, D.Z., Chen, Y.W., 2014. Arsenic induces reactive oxygen species- caused neuronal cell apoptosis through JNK/ERK-mediated mitochondria-dependent and GRP 78/CHOP-regulated pathways. Toxicol. Lett. 224, 130–140

Miller, L. L.; Wang, F.; Palace, V. P.; Hontela, A. 2007. Effects of acute and subchronic exposures to waterborne selenite on the physiological stress response and oxidative stress indicators in juvenile rainbow trout. Aquat. Toxicol. 83, 263−271.

Mintz, M., Vanderver, A., Brown, K.J., Lin, J., Wang, Z., Kaneski, C., Schiffmann, R., Nagaraju, K., Hoffman, E.P., Hathout, Y., 2008. Time series proteome profiling to study endoplasmic reticulum stress response. J. Proteome. Res. 7, 2435–2444.

256 Misra, S., Hamilton, C., Niyogi, S., 2012. Induction of oxidative stress by selenomethionine in isolated hepatocytes of rainbow trout (Oncorhynchus mykiss). Toxicol. In Vitro 26, 621−629.

Patterson, S.E., Dealy, C.N., 2014. Mechanisms and models of endoplasmic reticulum stress in chondrodysplasia. Dev. Dyn. 243, 875-893.

Porazinski, S.R., Wang, H., Furutani-Seiki, M., 2010. Dechorionation of medaka embryos and cell transplantation for the generation of chimeras. J. Vis. Exp. 22, 2055

Renn, J., Winkler, C., Schartl, M., Fischer, R., Goerlich, R., 2006. Zebrafish and medaka as models for bone research including implications regarding space-related issues. Protoplasma 229, 209–214.

Saito, A., Ochiai, K., Kondo, S., Tsumagari, K., Murakami, T., Cavener, D.R., Imaizumi, K., 2011. Endoplasmic reticulum stress response mediated by the PERK-eIF2(alpha)- ATF4 pathway is involved in osteoblast differentiation induced by BMP2. J. Biol. Chem. 286, 4809−18.

Schmittgen, T.D., Livak, K.J., 2008. Analyzing real-time PCR data by the comparative C(T) method. Nat. Protoc. 3, 1101−8.

Shoulders, M.D., Ryno, L.M., Genereux, J.C., Moresco, J.J., Tu, P.G., Wu, C., Yates, J.R., Su, A.I., Kelly, J.W., Wiseman, R.L., 2013. Stress-independent activation of XBP1s and/or ATF6 reveals three functionally diverse ER proteostasis environments. Cell Rep. 3, 1279-1292

Spallholz, J.E., Palace, V.P., Reid, T.W., 2004. Methioninase and selenomethionine but not Se-methylselenocysteine generate methylselenol and superoxide in an in vitro chemiluminescent assay: Implications for the nutritional carcinostatic activity of selenoamino acids. Biochem. Pharmacol. 67, 547−54.

Tu, B.P., Weissman, J.S., 2004. Oxidative protein folding in eukaryotes: mechanisms and consequences. J. Cell Biol. 164, 341-346.

Tufo, G., Jones, A.W., Wang, Z., Hamelin, J., Tajeddine, N., Esposti, D.D., Martel, C., Boursier, C., Gallerne, C. and Migdal, C., 2014. The protein disulfide isomerases PDIA4 and PDIA6 mediate resistance to cisplatin-induced cell death in lung adenocarcinoma. Cell Death Differ. 21, 685–695.

257 Van Der Vlies, D., Makkinje, M., Jansens, A., Braakman, I., Verklejj, A.J., Wirtz, K.W., Andries, J., 2003. Oxidation of ER resident proteins upon oxidative stress: effects of altering cellular redox/antioxidant status and implications for protein maturation. Antioxid. Redox Signal. 5, 381-387.

Wali, J.A., Rondas, D., McKenzie, M.D., Zhao, Y., Elkerbout, L., Fynch, S., Gurzov, E.N., Akira, S., Mathieu, C., Kay, T. W., 2014. The proapoptotic BH3-only proteins Bim and Puma are downstream of endoplasmic reticulum and mitochondrial oxidative stress in pancreatic islets in response to glucotoxicity. Cell. Death. Dis. 5, e1124.

Yasutake, J., Inohaya, K., Kudo, A., 2004. Twist functions in vertebral column formation in medaka, Oryzias latipes. Mech. Dev. 121, 883–894.

Zhao, S., Fernald, R.D., 2005. Comprehensive algorithm for quantitative real-time polymerase chain reaction. J. Comput. Biol. 12, 1047−64.

Zu, K., Bihani, T., Lin, A., Park, Y.M., Mori, K., Ip, C., 2006. Enhanced selenium effect on growth arrest by BiP/GRP78 knockdown in p53-null human prostate cancer cells. Oncogene 25, 546–554.

258 Chapter 7: Conclusions

Selenium (Se) is an essential micronutrient that causes developmental toxicity to oviparous vertebrates. Thresholds between essentiality and toxicity are narrow, and the ability of Se to bioaccumulate and transfer maternally continues to lead to Se contamination across the US. In particular, the San Francisco Bay Delta (SFBD) has experienced historical Se contamination and continues to struggle with Se levels in certain areas. Se toxicity is presented as a variety of deformities, including spinal, cranio- facial, fin and organ-level deformities, as well as lethality (Lemly, 1997). Se can be converted to selenomethionine (SeMet) by micro-organisms in aquatic systems. SeMet is ingested in the diet of oviparous organisms, incorporated into yolk proteins, and passed to developing embryos, where it is responsible for observed toxicity (Janz et al., 2010).

Studies on Se have thus far focused on sulfur substitution or oxidative stress as mechanisms of toxicity. FMOs can activate SeMet to the corresponding SeMet oxide, which can be reduced by GSH and re-oxidized to undergo redox cycling (Chen &

Ziegler, 1994; Krause et al., 2006). However, little research is available directly linking oxidative stress with observed adverse outcomes. Furthermore, there has been less focus on other potential mechanisms, such as the unfolded protein response and apoptosis, which have been shown to work in a positive feedback loop with oxidative stress.

In addition to Se, fish encounter other stressors in their environments, such as hypersalinity. Salinity of many fresh waterways and estuaries has been increasing due to climate change and anthropogenic disturbance, such as input of desalination brine and agricultural activities. In CA, salinity in the SFBD has increased over the past 100 years

259 (Stahle et al., 2001), and researchers estimate that it will continue to rise due to climate change (Knowles & Cayan, 2002). Hypersalinity can cause toxicity to fish by altering osmotic pressure, cell/body volume, oxygen consumption and metabolism (Canedo-

Arguelles et al., 2013; Boeuf & Payan, 2001). Saltwater toxicity appears to be dependent both on ionic-content and ion ratios, leading to some scientists to call for ion-specific standards (Canedo-Arguelles et al., 2016). Although hypersalinity can be toxic alone, sublethal concentrations of saltwater can significantly modify the impacts of toxicants, such as SeMet. In fact, previous research demonstrated that SeMet developmental toxicity increased under hypersaline conditions and that this alteration was due to increases in FMO activity at higher salinities (Lavado et al., 2012).

The primary goals of this dissertation were to characterize developmental toxicity of SeMet and hypersalinity, and to elucidate the molecular mechanisms behind observed mortality and deformities. Specifically, the objectives were five-fold: 1. To evaluate toxicity of saltwater of varying ionic compositions to Japanese medaka embryos; 2. To characterize developmental expression patterns and potential regulation of FMOs during development; 3. To investigate the stage-specificity of SeMet and hypersaline toxicity; 4.

To examine oxidative stress, the UPR, and apoptosis as mechanisms in SeMet and hypersaline toxicity; and 5. To investigate the localization of oxidative stress, the UPR and apoptosis in SeMet induced spinal-deformities. These studies were all performed with the model organism, the Japanese medaka (Oryzias latipes). Medaka were an appropriate model organism for this work because they are euryhaline (able to tolerate

260 both fresh and salt water); and have a fully sequenced genome, rapid development, and a clear chorion, making molecular investigations and developmental assessments possible.

In order to assess growing concerns about environmental impacts of climate change and desalination, the embryo-larval toxicity of sulfate-based brackish water was compared to desalination brine and artificial seawater. A salinity-response of each water type was performed using dilutions with seawater to mimic oceanic mixing effects. No difference in toxicity was observed between artificial seawater and desalination brine, and EC50 and NOEC values for seawater and brine varied between 45-50ppt suggesting that medaka would be protected under current regulations. In contrast, artificial water based on the San Joaquin River (SJR) demonstrated greater larval toxicity and increased deformities compared to either seawater based on salinity alone. These studies are the first to examine sublethal toxicity of desalination brine in medaka, and the first to compare it to different ionic compositions. Results suggest that ion-specific effects may influence hypersaline toxicity and need to be incorporated into environmental regulations.

It was previously demonstrated that FMOs are induced by hypersaline conditions and may potentiate SeMet toxicity in euryhaline species of fish. Yet, FMOs have been uncharacterized during development in medaka. Thus, the goal of this chapter was to characterize the expression and regulation of medaka FMOs during development. Five putative medaka FMOs were identified by NBCI and found to contain FMO-identifying motifs. FMO5A-D were approximately 50% identical to each other, whereas FMO4 was approximately 30% identical to the FMO5 isoforms. All FMOs were expressed

261 throughout development, displaying stage-specific expression patterns. FMO4 and

FMO5D increased during organogenesis then decreased, whereas FMO5B and 5C expression decreased beginning in early organogenesis and remained low. FMO5A expression remained constant throughout development.

Promoter analysis indicated a 74 transcription factor binding sites (TFBS) in the

FMO promoters, which were enriched for embryonic development (particularly liver development), apoptosis, oxidative stress, immunity and inflammation. Interestingly,

TFBS for the UPR were also discovered, particularly CHOP, ATF4 and C/EBPβ.

Embryos were treated from 5-6dpf (organogenesis) with tunicamycin or dithiothreitol to assess potential regulation by the UPR. High concentrations of tunicamycin increased

FMO4 and FMO5A expression, whereas dithiothreitol significantly decreased FMO5A,

5B and 5C. These results demonstrate that medaka FMOs are expressed during embryonic development and are under variable regulation by different types of UPR stressors. It is thus possible that FMOs may be regulated by the UPR induced by SeMet toxicity, which would result in greater SeMet activation.

To confirm hypersaline potentiation of SeMet (an FMO substrate) developmental toxicity, the combined effects of hypersaline SJR water and SeMet were investigated.

Previous research suggested that SJR water could potentiate SeMet toxicity (Lavado et al., 2012); however, these studies were performed at a single early developmental stage.

Embryonic development is a complex spatiotemporal process. Often xenobiotic toxicity varies by developmental stage because of the constantly changing cellular composition and microenvironment. It was hypothesized that SeMet would exhibit stage-specific

262 toxicity that would be modified under hypersaline conditions. Six developmental stages were treated with a concentration response of SeMet in freshwater or SJR saltwater for 24 hours. Stages 9 and 17 (0-2dpf) were more susceptible to SeMet lethal and sublethal toxicity than later stages, even though SeMet uptake was stage-dependent and increased with stage. Embryos treated at very early stages (0-2dpf) and very late stages (6-9dpf) exhibited a range of deformities, while those treated in between were more specific for characteristic spinal deformities. Hypersalinity significantly increased SeMet toxicity at stage 25 (48hpf) in Japanese medaka, when early liver development, osmoregulation, and chondrocyte differentiation begin.

Once susceptible stages were established, the molecular mechanism behind SeMet and hypersaline induced mortality was investigated. Since 50µM SeMet caused high mortality after 24 hours, embryos were treated for 12 hours so that mechanisms could be investigated prior to death. Hypersalinity significantly increased mortality and deformities in SeMet-treated medaka embryos in comparison to those treated in freshwater. No increase in lipid peroxidation (a measure of oxidative stress), or mRNA expression of apoptotic mediators BAX and CASP3A were observed. However, a significant increase in BiP, the major regulator of the unfolded protein response (UPR) was observed suggesting a role for UPR in SeMet and hypersaline toxicity. This is in accordance with Chapter 2, where the UPR was shown to modulate medaka FMO expression.

These studies were the first to implicate the UPR in SeMet toxicity and to suggest that oxidative stress may not be solely response for observed effects. However,

263 measurements were based on whole embryo homogenates and the high concentration of

SeMet suggested that overt toxicity may be of concern. Subsequent research sought to further elucidate molecular mechanisms of SeMet toxicity, focusing on spinal deformities induced by lower concentrations of SeMet. Although spinal deformities (particularly lordosis) have been established as characteristic of Se toxicity, specific mechanisms behind those deformities were unknown. For this study medaka embryos were treated at stage 25 (48hpf) for 24hrs with 5µM and 2.5µM SeMet. This stage was chosen because previous studies indicated a significant correlation between Se tissue content and percentage of deformed embryos, and deformities were specific to lordosis. Hypersaline treatments were eliminated from these experiments to avoid over-complication of results.

SeMet-treated medaka embryos exhibited a range of deformities from mild to severe post treatment, however, deformities in the developing tail were the most prevalent. Dichlorofluorescein was used to image oxidative stress, and demonstrated staining in tails of embryos with observable spinal malformations. Furthermore, acridine orange staining for apoptosis was increased in the tails of deformed embryos. Expression of UPR downstream target genes was analyzed and significant decrease in PDIA4 was observed, correlating with a trend towards an increase in CHOP in embryos treated with

5µM SeMet. Tunicamycin was used as a positive control and resulted in significant increases in all downstream targets, suggesting that SeMet did not activate a classical, robust UPR. However, observed effects were linked to significant alterations in genes involved in chondrogenesis (Sox9, Col2a1) and osteogenesis (Runx2), indicating that

SeMet significantly repressed chondrogenesis, while promoting osteogenesis. Overall

264 results suggest that SeMet-induced spinal deformities in Japanese medaka embryos may be due to oxidative stress, the UPR and apoptosis in developing chondrocytes.

In summary, hypersalinity and SeMet induce molecular changes in Japanese medaka embryos leading to toxicity in a stage- and ion-specific manner. These results will aid in the development of regulatory standards for ions and Se in CA and across the

US. The studies also highlight the importance of integrating developmental biology into environmental toxicology, as developmental processes can significantly impact xenobiotic toxicity. If studies are only conducted at a single developmental stage, important toxic effects may be overlooked. Furthermore, our results indicate the importance of examining both sublethal and lethal effects of these compounds, as mechanisms may be different between them. Finally, the current body of literature on the combined role of oxidative stress, the UPR and apoptosis is growing. Thus, the understanding of molecular mechanisms of toxicity gained by this research and will inform future work in environmental and developmental toxicology.

265 References

Boeuf, G., Payan, P., 2001. How should salinity influence fish growth? Comp. Biochem. Physiol. C Pharmacol. Toxicol. 130, 411 -423.

Cañedo-Argüelles, M., Kefford, B.J., Piscart, C., Prat, N., Schäfer, R.B., Schulz, C., 2013. Salinisation of rivers: An urgent ecological issue. Environ. Pol. 173, 157-167.

Chen, G.-P., Ziegler, D.M., 1994. Liver microsome and flavin-containing monooxygenase catalyzed oxidation of organic selenium compounds. Arch. Biochem. Biophys. 312, 566–572.

Knowles, N., Cayan, D.R., 2002. Potential effects of global warming on the Sacromento/San Joaquin watershed and the San Francisco Estuary. Geophys. Res. Lett. 29, 32-42.

Lemly, A.D., 1997. A teratogenic deformity index for evaluating impacts of selenium on fish populations. Ecotox. Environ. Safe. 37, 259-66.

266