The Effects of the Herbicide Atrazine on Hawaiian Corals

THE EFFECTS OF THE HERBICIDE ATRAZINE ON HAWAIIAN CORALS

A THESIS SUBMITTED TO THE GRADUATE DIVISION OF THE UNIVERSITY OF HAWAI‘I AT MĀNOA IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF

MASTER OF SCIENCE IN MARINE BIOLOGY

MAY 2020

By Laura Anne Damiani

Thesis Committee: Robert H. Richmond, Chairperson Zac Forsman Kirsten Oleson

Keywords: Coral reefs, anthropogenic impacts, Hawaii, biomarkers, atrazine, herbicide, ecotoxicology

ACKNOWLEDGEMENTS

My advisor: Dr. Robert H. Richmond

My committee: Dr. Zac Forsman, Dr. Kirsten Olesen

Colleagues in the Richmond Lab: Alex Barkman, Dr. Kaho Tisthammer, Narrissa Brown, Dr. James Murphy, Maikani Mereng Andres, Aja Reyes, Lauren Wetzell, Ohialehua Bullock, Lauryn Hansen, Danny Zhen

Funding: Marine Biology Graduate Program at the University of Hawai‘i at Mānoa, Edmondson Research Fund, University of Hawai‘i at Mānoa Department of Biology, NOAA, NFWF, Maui Nui Marine Resource Council

Special thanks to: Richard Galindo, John Lam, Kory Misaki, Kathy Souza, Marissa Stone, Dr. Brian Nedved, Claire Lager and the Hagedorn Lab

My family: My sister Isabel, and parents Sandra and Joel Damiani

iii

ABSTRACT

The herbicide atrazine is one of the most commonly applied herbicides used worldwide, despite being shown to disrupt reproductive development in some animals. Atrazine is used for agriculture in Hawai‘i and potentially threatens marine organisms after it enters the ocean through runoff. Various pollutants and pesticides affect the health of corals, but atrazine’s effect on corals is unknown. This study investigates the effects of atrazine on the early life history stages of coral, and the molecular response of corals exposed to atrazine. Fertilization assays were performed with Montipora capitata gametes exposed to select concentrations of atrazine, and adult colonies of M. capitata and Leptastrea purpurea were exposed to atrazine in the laboratory. Along with monitoring for physiological effects, Western blot analyses were performed using the expression of select proteins, including xenobiotic metabolizing enzymes, to detect stress at the cellular level. There was no significant difference in percent fertilization or cell division rates between treatments in M. capitata embryos. Overall, eight biomarkers were identified to be used for Western blots in studying sublethal stress in L. purpurea, and three in M. capitata. This study indicates that atrazine does not greatly impact the coral animal itself within the coral holobiont for the species studied. However, it has the potential to impact its protein expression, including the expression of hydroxysteroid 17-beta dehydrogenase 8 (HSD17B8), which plays a role in regulating reproductive activity. Longer term exposures of several months are suggested for addressing the effects of atrazine on coral metabolism and gamete formation.

Studies such as this that develop and apply molecular biomarkers to corals will help resource managers in learning about the causation of coral stress for coral reef restoration.

iv TABLE OF CONTENTS Acknowledgements……………………………………………………………………………....iii Abstract…..……………………………………………………………………………………….iv List of Figures…………………………………………………………….……………………...vii List of Symbols and Abbreviations…..…………………………………………...…………….viii Introduction………………………………………………………………………………………..1 Materials and Methodology………………………………………………………………….…....5 Study species of coral……………………………………………………………………..5 Atrazine preparation……………………………………………………………………….5 Spawning and gamete collection………………………………………………………….5 Fertilization assays with M. capitata…………………………………………………..….6 M. capitata larval exposures to atrazine………………………………………..…………6 P. damicornis and M. capitata exposure to atrazine………………………………………7 L. purpurea exposure to atrazine…………..…………………………………………...…8 SDS-Page and Western blots with coral proteins…………………………………………9 Results…………………………………………………………………………………………....12 Statistical analyses………………………………………………………….……………12 Fertilization assays with M. capitata…………………………………………………….12 M. capitata larval exposures to atrazine………………………………………………....13 P. damicornis and M. capitata exposure to atrazine and protein response……………....14 L. purpurea exposure to atrazine…………………………………………....…………...15 Larval output……………………………………………………………………..15 L. purpurea settlement…………………………………………………………...16 Protein response in L. purpurea….……………………………………………....16 Discussion…………………………………………………………………………………….….20 Impact of Atrazine on Coral Fertilization…………………………………………...…...20 Impact of Atrazine on Coral Larvae……………………………………………………..21 Coral Biomarkers………………………………………………………………………...22 Limitations……………………………………………………………………………….23 Future Steps……………………………………………………………………………...24 Management Implications………………………………………………………………..25

v Conclusion……………………………………………………………………………….25 Appendix A: Supporting Information….…...……………………………………………………27 List of Tables………………………………………………………………………….…27 List of Figures……………………………………………………………………………27 Appendix B: Background on Biomarkers………………………………………………………..33 References….…………………………………………………………………………………….35

vi LIST OF FIGURES

Figure 1. Gamete trap……………………………………………………………..……………....6

Figure 2. Example of a test chamber for the atrazine exposure…………………………………..7

Figure 3. The experimental set-up for L. purpurea…………………………...……………………..9

Figure 4. Average percent fertilization of M. capitata embryos…………………..……...….….12

Figure 5. Average cell division of M. capitata embryos…………………………………...…...13

Figure 6. M. capitata larval exposure to atrazine…………………………………….…………13

Figure 7. Example of corals at the end of exposure……………………………………………..14

Figure 8. Biomarker protein expressions in M. capitata……………………………..…………….15

Figure 9. Biomarker protein expressions in L. purpurea ………………………….………...... 18

vii LIST OF SYMBOLS AND ABBREVIATIONS

ANOVA: analysis of variance

ATZ: atrazine

BCA: bicinchoninic acid

°C: degrees Centigrade

CaM: calmodulin

CCA: crustose coralline algae cDNA: complementary deoxyribonucleic acid

CYP: cytochrome P450 monooxygenase

CYP1A1: cytochrome P450, family 1, subfamily A, polypeptide 1

CYP17A: cytochrome P450 17A1

DI: deionized

DMSO: dimethyl sulfoxide

ECL: enhanced chemiluminescence

FSW: filtered seawater

GPx-1: glutathione peroxidase 1

GSR: glutathione reductase h: hour

HSD17B8: hydroxysteroid 17-beta dehydrogenase 8

HSP60: heat shock protein 60 kDa: kilodalton

L: liter

LC-MS/MS: liquid chromatography tandem mass spectrometry

viii μgl-1: microgram per liter mg: milligram

MIF: macrophage migration inhibitory factor mL: milliliter

NAP: non animal protein nMDS: non-metric multidimensional scaling

PAH: polycyclic aromatic hydrocarbon

PAM: pulse-amplitude modulation

PCB: polychlorinated biphenyl

PMSF: phenylmethylsulphonyl fluoride ppb: parts per billion ppt: parts per thousand

PSMC: proteasome 26S subunit

PVDF: polyvinylidene difluoride rcf: relative centrifugal force

SE: standard error

SD: standard deviation

SDS: sodium dodecyl sulfate

SELENBP1: selenium binding protein 1

SOD1: superoxide dismutase 1

START: stAR-related lipid transfer

ix INTRODUCTION

Harboring an incredible amount of biodiversity, coral reefs are of great ecological, economic, and cultural importance. Despite only covering 0.1-0.5% of the ocean floor, coral reefs are home to nearly one-third of all marine fish species (McAllister, 1991; Spalding & Grenfell, 1997). They have been compared to rainforests for their rich biodiversity and high local diversity of species (Reaka-Kudla, 1997). Coral reefs provide renewable resources of economic value such as seafood products and even natural marine products used in the pharmaceutical industry (Moberg & Folke, 1999). Grafeld et al. (2017) found that nearshore Hawaiian coral reef fisheries provide more than 7 million meals annually. Reefs physically protect coastlines and dissipate wave energy, preventing coastal erosion (Kench & Brander, 2006). A 2019 report by the U.S. Geological Survey estimated that U.S. coral reefs prevent $825 million in direct flood damages to buildings (Storlazzi et al., 2019). Reefs also provide nature-based tourism activities such as scuba diving, snorkeling, and underwater photography. Coral reefs also have an important cultural aspect to island and nearshore communities. The Kumulipo, or Hawaiian Creation Chant, states that the coral polyp was the first creature to be born from the sea, from which other life evolved from through ancestral deities (McGregor, 2007). Many reef activities, including traditional harvesting, form the fabric of Pacific Island cultures. However, reefs are declining due to global stressors such as rising sea temperatures and acidifying oceans as well as local stressors such as pollution, overfishing, invasive species, and coastal development. Rising water temperatures can cause corals to expel their algal symbionts in a process known as bleaching, which can lead to coral mortality when bleaching events are prolonged. Local stressors have been found to decrease coral resilience to bleaching by increasing the recovery time following these events (Carilli et al., 2009). Focusing coral conservation efforts on reducing local stressors may “buy time” to address larger issues like climate change. One factor that affects reef health is water quality, which may be impacted by anthropogenic inputs such as heavy metals, pesticides, oil spills, and personal care products. Corals are particularly susceptible to endogenous chemicals due to their thin, lipid rich tissue layer that may attract the uptake of lipophilic chemicals (Peters et al., 1997). One chemical of particular concern is atrazine, a widely used herbicide that has been found in the coastal waters of Hawai‘i that interferes with the reproductive systems of a variety of

1 animals. This study was designed to understand whether this herbicide affects coral reproduction, recruitment and development. Atrazine (2-chloro-4-ethytlamino-6-isopropylamine-1,3,5-triazine) is the second most commonly used herbicide in the U.S., with over 70 million pounds applied yearly (Thelin et al. 2013). It is primarily used to control the growth of grass weeds in the production of crops such as corn, sorghum, and sugarcane (Solomon et al., 1996). Organisms that live in water bodies that receive input from streams and rivers in proximity to agricultural fields where atrazine is applied are particularly at risk. Since as early as the 1970’s, Hawai‘i sugarcane, pineapple, and macadamia nut farmers have used atrazine to control weed growth (Shigeura & Ooka, 1984). Atrazine is classified as a restricted-use pesticide, but the Hawai‘i Department of Agriculture does not manage the levels at which atrazine is applied. In the 2013- 2014 statewide pesticide sampling water project, the Hawai‘i State Departments of Health and Agriculture detected atrazine in the majority of their samples from surface water. The highest concentration of atrazine detected was 2.05 μgl-1 from a site on west Kauai, which exceeds the EPA’s aquatic life benchmark for freshwater algae at 1 μgl-1 (Grange, 2014).

Atrazine is a triazine herbicide that works through the mechanism of binding to the QB site on the D1 protein of Photosystem II, thus preventing the binding of plastoquinone (Lavergne, 1982). This behavior leads to chlorophyll-mediated photodamage causing mortality in the impacted plant (Jones et al., 2003). Atrazine resists microbial degradation and can persist in the environment for up to several years (Howard, 1991). Khan (1978) found the environmental half- life of atrazine to be 742 days at a pH of 7.0 and to decrease with lower pH. On the other hand, Obien & Green (1969) found atrazine to degrade quicker in Hawaiian soils, as only 15-30% of the initial quantity applied was detected after 34 days. Atrazine is moderately water soluble at 33 mg/liter at 27°C (Best & Weber, 1974). Atrazine can also decrease phytoplankton biomass, affecting primary production in aquatic ecosystems (Solomon et al., 1996; Starr et al., 2016). Atrazine can disrupt endocrine functions, raising worldwide concerns about its safety and effect on human and animal health (Hayes et al., 2002; Solomon et al., 1996). Atrazine has been reported to cause demasculization and hermaphroditism in amphibians and may promote the conversion of testosterone to estrogen (Hayes et al., 2002). Other studies have documented the effect of atrazine on the demasculinization of other animals such as teleost fish, crustaceans, reptiles, and mammals (Álvarez et al., 2015; Papoulias et al., 2014; Rey et al., 2009; Spanò et al., 2004; Tillitt et al., 2010; Victor-Costa et al., 2010). Throughout these studies, demasculinization

2 consisted of decreases in testicular size and weight, tissue damage to testis, loss of Sertoli cells, and a loss of germ cells (Spanò et al., 2004, Rey et al., 2009, Victor-Costa et al., 2010). These effects occur due to a reduction of androgens, by which various mechanisms have been identified in vertebrates (Hayes et al., 2011). The majority of the research has studied atrazine’s effects on freshwater aquatic organisms from the Midwest. In a rare study focused on corals, Jones et al. (2003) found that the symbiotic zooxanthellae of corals exposed to atrazine for 10 h had decreased photosynthetic efficiency. However, no study has investigated the potential effects of atrazine exposure on the coral animal itself, or on coral reproduction. Because stony corals have estrogenic and androgenic compounds, including estrone, 17β-estradiol, progesterone, and testosterone, there is reason to believe that atrazine may interfere with normal reproductive functions in corals (Atkinson & Atkinson, 1992; Gassman, 1993; Tarrant et al., 1999). This study sought to answer whether atrazine affects different stages of coral reproduction, including fertilization, larval metamorphosis and settlement, and adult gametogenesis. In addition, I investigated sublethal stress in adult corals using a variety of protein biomarkers. In order for reef building corals to sexually reproduce, gametes must be fertilized, develop into free-swimming planulae, settle on suitable substrata, and undergo metamorphosis (Richmond, 1997). They must additionally acquire zooxanthellae either from their parental colony or from their environment, depending on the species. Environmental conditions or anthropogenic stressors may impact the coral at any of these stages, preventing recruitment and survival. For instance, chemicals in the water may interfere with the proper chemical signaling necessary for egg and sperm recognition or prevent planulae from identifying appropriate settlement substrates (Richmond, 1997; Richmond et al., 2018). Previous studies have assessed fertilization success and cell developmental rate for coral gametes as a measurement of toxicity for contaminants such as copper, tributyltin, and lubricating oils (Victor & Richmond, 2005, Hédouin & Gates, 2013, Negri & Heyward, 2001, Mercurio et al., 2004). Mercurio et al. (2004) exposed Acropora microphthlama gametes to vegetable-derived lubricants and mineral-derived oil and found that higher concentrations tested in both reduced fertilization and led to abnormal cell division. In addition, Montipora capitata gametes exhibited lower fertilization and settlement success when exposed to the commercially available herbicide Roundup® (Diu et al., 2015). Because atrazine is a Photosystem II inhibitor and endocrine-disrupting chemical, there is reason to believe atrazine could impair early life stages of coral.

3 This study tested the effects of atrazine on the (1) fertilization, (2) cell division rate, and (3) larval survival of Montipora capitata. I hypothesized that atrazine would negatively impact fertilization success, gamete cell division, and larval survival in M. capitata, with higher concentrations of atrazine having a larger impact. These endpoints were chosen to test whether atrazine is a toxicant to early life stages of coral. Traditional metrics such as coral cover or species loss to measure coral reef health do not provide information on the identity of environmental stressors or how to prevent future coral loss. It is becoming more common to use biomarkers as a tool to assess coral health and to learn about the physiological state of the animal. They have been applied in previous studies to detect coral stress responses resulting from exposure to compounds including fuel oil and anthracene (Montilla et al., 2016; Rougée et al., 2006). For example, Rougée et al. (2006) found an increase in cnidarian cytochrome P450 1-class and 2-class and glutathione S-transferase-pi expression in Pocillopora damicornis exposed to various concentrations of marine fuel oil. The biomarkers tested in this study included cytochrome P450 1A1 (CYP1A1), selenium binding protein 1 (SELENBP1), heat shock protein 60 (HSP60), calmodulin (CaM), superoxide dismutase 1 (SOD1), hydroxysteroid 17-β dehydrogenase 8 (HSD17B8), glutathione peroxidase-1 (GPx-1), and glutathione reductase (GSR). I hypothesized that I would be able to use these biomarkers as a means to detect stress at the molecular level in corals exposed to atrazine.

4 MATERIALS AND METHODOLOGY

Study species of coral Montipora capitata is a major reef-building coral in the Hawaiian Archipelago, growing in branching and plating morphologies (Forsman et al., 2010). It is a hermaphroditic broadcast spawner which releases buoyant egg-sperm bundles 0-3 nights after a new moon during summer months (Padilla-Gamiño & Gates, 2012). Because large quantities of gametes may be collected during a spawning night, this allows for M. capitata planulae to be harvested and used for experiments with a high number of biological replicates. Leptastrea purpurea, or crust coral, is a generalist coral species found throughout the Indo-Pacific to the Red Sea (Veron, 2000). In Hawai‘i, L. purpurea broods planulae year-round, with the highest larval output being between the months of August to October (Nietzer et al., 2018; Narrissa Brown, personal communication). Nietzer et al. (2018) found that L. purpurea colonies released 3.7 larvae on average over a period of 65 days from August to October in 2014. The authors suggest that L. purpurea will be a dependable source of larvae for future research on corals. In addition, L. purpurea is believed to be relatively resilient to higher temperatures and toxicant exposure, and able to thrive in poor water conditions (Narrissa Brown, personal communication).

Atrazine preparation Atrazine solutions were prepared by first making a 1000 μgl-1 stock solution by dissolving analytical grade atrazine (Sigma-Aldrich, St. Louis, MO) in dimethyl sulfoxide (DMSO) and diluting this solution with 0.2-micron filtered seawater (FSW). DMSO was used as a solvent but has been shown to be produced naturally in corals (Gardner et al., 2017). The different atrazine concentrations tested were made by diluting the stock solution with FSW. Atrazine concentrations were chosen by taking into consideration environmental concentrations previously measured in Hawai‘i, as well as the EPA aquatic ecosystem’s Levels of Concern being 10 μgl-1.

Spawning and gamete collection Gametes were collected from Montipora capitata colonies in Kāne‘ohe Bay, O‘ahu, HI, on June 14, 2018. Five gamete traps were placed over large colonies with engorged polyps in the

5 bay and collected after spawning, which occurred at approximately 9 pm (Fig. 1). The egg-sperm bundles were mixed from three or more colonies in order to fertilize the gametes.

Figure 1. Gamete trap placed above M. capitata colony in Kāne‘ohe Bay, O‘ahu.

Fertilization assays with M. capitata Atrazine concentrations of 2, 10, 50, 100, 200, and 1000 μgl-1 were tested. In the lab, 48 mL solutions of atrazine as well as an FSW and a 0.0167% DMSO solvent control were prepared in clear glass jars with Teflon®-lined lids, with six replicates per treatment. Immediately after collecting the gametes from M. capitata, approximately 20 egg-sperm bundles from at least three different colonies were placed in each jar. Embryos (roughly 20 per replicate) were placed in 1 mL of 10% zinc formalin fixative (Z-Fix) after 2, 3, 4, and 5 hours to stop development and preserve them for later study. These time points were chosen because the cells do not begin to divide until after the 2nd hour. The developmental stage of each embryo (2, 4, 8, 16, past 16-cell stage) was assessed using a dissecting microscope. Cells that were beginning to segment into 2 cells were scored as the 2-cell stage.

M. capitata larval exposures to atrazine Eight-day old M. capitata larvae from the June 2018 spawn were exposed to either FSW,

-1 -1 DMSO, 2 μgl atrazine, or 10 μgl atrazine for a period of 3 days. One-hundred larvae were added to 50 mL of each treatment in clear glass jars with Teflon®-lined lids, with three replicates

6 per treatment. After 3 days, larvae were frozen at -80°C and their protein extracted and quantified using the bicinchoninic acid (BCA) assay, as described in Murphy and Richmond (2016).

P. damicornis and M. capitata exposure to atrazine P. damicornis and M. capitata colonies originally collected from Kāne‘ohe Bay were fragmented into nubbins and placed on small ceramic tiles using non-toxic epoxy. Corals were left in a continuous flow seawater table for 3 weeks to recover. Between December 8-13, 2017, P. damicornis and M. capitata nubbins were exposed to either FSW, DMSO, 2 μgl-1 atrazine, or 10 μgl-1 atrazine for a period of 5 days. Test chambers consisted of 2-L glass beakers, with four replicates each, and one P. damicornis and M. capitata nubbin per beaker (Fig. 2). Every other day, a one-third water change was completed. Throughout the exposure period, any visual signs of stress (bleaching, tissue loss) were recorded for each of the corals. Coral nubbins were flash frozen at the end of the experiment, crushed, and proteins were extracted.

Figure 2. Example of a test chamber for the atrazine exposure, with one M. capitata and one P. damicornis nubbin per beaker.

7 L. purpurea exposure to atrazine On September 20, 2018, colonies of Leptastrea purpurea (<3 inches in diameter) were collected off of Ke‘ehi Small Boat Harbor, Honolulu, HI, and transferred to an outdoor flow through tank at Kewalo Marine Laboratory (University of Hawai‘i at Mānoa). Corals were collected under the Department of Land and Natural Resources-Division of Aquatic Resources coral collection permit SAP 2019-25 (O‘ahu, HI, USA). The following day, each colony was transferred to a small tank with a glass Pasteur pipette bubbler and placed in the flow-through tank to regulate temperature. For the next 12 days, the number of larvae produced per colony per day was recorded. The six colonies with the highest total larval output were selected for the exposure in order to have 3 replicates per treatment. Solutions of 1 μgl-1 atrazine and a 1.667*10-5 % DMSO control were freshly prepared and added to 2-L glass beakers to be used as test chambers. The beakers were placed in the same flow-through tank and equipped with glass Pasteur pipettes that provided constant gentle aeration. Colonies were transferred to the test chambers so that there was one individual coral per chamber (Fig. 3). The water table was equipped with a HOBO® data logger (Onset Computer, Bourne, MA) to monitor water temperature and light intensity. Salinity measurements were taken daily using a portable refractometer. In addition, coral polyps were noted as feeding (extended) or retracted every day. A shade cover was placed over the beakers in the water table, and full water changes were completed once every three days of the experiment. Every day, larvae from each colony were counted and removed from the beakers. Larvae were pooled per treatment per day and transferred to a glass dish with 80 mL of fresh coral water and half of a crustose-coralline algae (CCA) chip. Coral water, which has been shown to induce settlement in L. purpurea (Narrissa Brown, personal communication), was made by placing live coral into FSW and removing the coral after 2 h. The dishes with larvae were kept in a 29°C incubator. Fifty percent water changes were done every 3 days. All surviving larvae were scored under a dissecting microscope after 7-10 days as either swimming, metamorphosed and settled, metamorphosed but not settled, or settled but not metamorphosed. Metamorphosed larvae had undergone tissue modification, having secreted a calcareous exoskeleton and developed obvious septal mesenteries, as described in Richmond (1985). Percent metamorphosis was calculated (the total number of larvae either metamorphosed and settled or only metamorphosed but not settled divided by the total number of surviving larvae).

8 At the end of the experiment, corals were flash frozen in liquid nitrogen and stored at -80°C. Corals were exposed for a total of 21 days from October 2 to October 23, 2018. The number of polyps per coral colony was calculated using photos of each coral and ImageJ software.

Figure 3. The experimental set-up with one L. purpurea colony per beaker in a water table at KML.

SDS-Page and Western blots with coral proteins Frozen coral tissue was crushed down to a powder using a Graseby Specac press, and placed in 1.5 mL Eppendorf tubes stored in -80°C. To extract protein, 500 μL homogenization buffer (49.5 mL 0.01 M Tris-Cl buffer pH 8, 500 μl DMSO, and 8.7 mg PMSF) was added to each sample. A handheld homogenizer was used on each sample for 1 minute, and the sample was centrifuged at 13,000 rcf for 20 minutes at 4°C. The supernatant (also known as the whole cell lysate), was collected and aliquoted in separate 1.5 mL tubes. The BCA assay was then used to determine the concentrations of protein as described in Murphy and Richmond (2016). To

9 separate the protein, 50 μg of each protein sample was loaded onto a ten-lane Mini-PROTEAN® TGXTM precast gel (Bio-Rad, Hercules, CA). First, protein lysates were prepared by adding a 4X SDS loading buffer to the samples and heating at 95°C for 5 minutes. The lysates were loaded into appropriate wells on the gel, and the gel run first at 80V for 30 minutes, then increased to 100V for approximately 75 minutes. In addition, 8 μg of HeLa whole cell lysate (Santa Cruz Biotechnology, Dallas, TX) was also loaded prior to running the gel to be used as a positive control. Protein was transferred from gels to PVDF membranes (EMD Millipore, Burlington, MA) overnight at 4°C at a current of 40V using the wet transfer method described in Mahmood and Yang (2012). To check that proteins transferred, membranes were stained with Ponceau S Solution (Biotium, Fremont, CA) for 5 minutes and washed twice with 5% acetic acid for 5 minutes. After visualizing the protein staining, membranes were transferred to deionized (DI) water for two washes of 5 minutes each. Membranes were blocked in either 5% nonfat dry milk (Carnation, Los Angeles, CA) or 1X NAP™-BLOCKER (G-Biosciences, St. Louis, MO) for one hour on a rotary table and incubated overnight at 4°C with the following primary antibodies from ThermoFisher Scientific: anti-CYP1A1 (IgG rabbit clone, PA5-15213, 1:2000 dilution or IgG rabbit clone, 13241-1-AP, ProteinTech, 1:2000 dilution), anti-SELENBP1 (IgG rabbit clone, PA5-37332, 1:1000 dilution), anti-HSP60 (IgG rabbit clone, PA1-41662, 1:1000 dilution), anti- HSD17B8 (IgG rabbit clone, PA5-50423, 1:1000 dilution), anti-calmodulin (IgG rabbit clone, PA5-11662, 1:1000 dilution), anti-SOD1 (IgG rabbit clone, PA1-30195, 1:2000 dilution), anti- GPx-1 (IgG rabbit clone, PA5-30593, 1:1000 dilution), anti-GSR (IgG rabbit clone, PA5-70004, 1:1000 dilution), anti-PSMC4 (IgG rabbit clone, PA5-51676, 1:1000 dilution), anti-MIF (IgG rabbit clone, PA5-27343, 1:1000 dilution), anti-annexin A7 (IgG rabbit clone, PA5-35358, 1:1000 dilution), and anti-cathepsin L (IgG goat clone, PA5-47971, 1:1000 dilution). Dilutions were performed in phosphate buffered saline with Tween® 20 (PBST). After the primary incubation, the membranes were washed in PBST four times. The blots were then incubated with goat anti-rabbit secondary antibody (sc-2004, Santa Cruz Biotechnology, 1:2000 dilution) for two hours at room temperature. Again, blots were washed four times in PBST. The blots were imaged using Pierce® Enhanced Chemiluminescence (ECL) Western Blotting Substrate (Thermo Scientific, Waltham, MA) and a C-DiGit® Blot Scanner (LI-COR Biosciences, Lincoln, NE). Image Studio™ software (LI-COR Biosciences, Lincoln, NE) was used to quantify the net intensity of the band signal using a rectangle and the

10 background subtracted by the Median method. For instances in which band signals on separate blots needed to be compared, such as with the M. capitata samples, the band intensity was normalized to the intensity of the HeLa whole cell lysate.

11 RESULTS

Statistical analyses Data was analyzed using the programs JMP® Pro versions 14 and 15 (SAS Institute Inc.) and RStudio v. 1.1.419. Normality of the distribution of data was assessed by performing a Shapiro-Wilk W Test, and homogeneity of variance tested by performing a Bartlett’s Test. An alpha level of 0.05 was used for all hypothesis test statistics.

Fertilization assays with M. capitata Embryos were first observed 2 hours after eggs were fertilized. Generally, average fertilization rates increased until after the 5th hour, which was the last hour tested (Table S1). Average percent fertilization after the 5th hour ranged from 21% to 57% (Table S1). The mean cell division among all treatments after the final 5th hour was 6.87 ± 6.26 cells. Wilcoxon Kruskal-Wallis by ranks tests were used to compare the medians of the percent fertilization and mean cell division, as the datasets did not fit normal distributions in order to perform ANOVA. Overall, there was no significant difference in average percent fertilization among each treatment after the 5th hour (Wilcoxon, χ = 5.224, P-value = 0.633, Fig. 4). In addition, mean cell division after the 2nd, 3rd, 4th, and 5th hours did not significantly differ among all treatments (Wilcoxon, χ = 2.430, P-value = 0.932, Fig. 5).

Figure 4. Average percent fertilization of M. capitata embryos 5 hours after being fertilized, with standard error bars (Wilcoxon, c = 5.224, P-value = 0.633).

Figure 5. Average cell division of M. capitata embryos 5 hours after being fertilized, with standard error bars (Wilcoxon, c = 2.430, P-value = 0.932).

M. capitata larval exposures to atrazine Percent survival ranged from 87.6% to 95.3% after three days of exposure. Average survival was 92.1 ± 5.5% among all treatments. There was no significant difference in percent survival among the four treatments tested (one-way ANOVA, F = 1.022, P-value = 0.433, Fig. 6). The BCA assay showed that not enough protein was recovered from each replicate to perform Western blots, most likely because coral larvae are made primarily of lipids.

Figure 6. Results of 3-day long M. capitata larval exposure to FSW, DMSO, 2 μgl-1, or 10 μgl-1 atrazine with standard error bars (one-way ANOVA, F = 1.022, P-value = 0.433).

13 P. damicornis and M. capitata exposure to atrazine and protein response Throughout the exposure period, no visual signs of stress such as bleaching or tissue loss were observed for any of the corals (Fig. 7). The M. capitata samples were selected for protein analysis in order to compare the fertilization and larval assays with the adult sublethal stress response to atrazine. Of all the antibodies tested, the following recognized antigens on the M. capitata proteins: anti-HSD17B8, anti-SELENBP1, and anti-SOD1. Of the other antibodies tested, anti-CYP1A1, anti-calmodulin, anti-GSR, anti-HSP60 had high levels of non-specific binding, while anti-annexin A7, anti-MIF, anti-cathepsin, and anti-PSMC4 did not bind to the target protein. A Shapiro-Wilk W test and Bartlett’s test were used to assess normality and homogeneity of variance, respectfully. If found to meet the assumptions, a one-way ANOVA was used to compare the Western blot net intensity between the samples. Anti-SELENBP1 recognized proteins of two different sizes, approximately 44 and 53 kDa, so statistical analyses were performed for both protein sizes. HSD17B8 recognized proteins at approximately 38 and 79 kDa. SOD1 recognized protein at approximately 14 kDa, slightly smaller than the 17 kDa control. Two of the three proteins that showed good banding did not reflect significant differences in protein expression (Fig. 8). However, there was a significant upregulation of HSD17B8 in the DMSO controls compared to the atrazine 10 μgl-1 exposed corals (T-Test, P-value = 0.032) (Fig. 8a). Net intensity refers to the unitless value the Image Studio™ software calculated based on the darkness of the band on the Western blot (Fig. 8).

Figure 7. Example of P. damicornis (left) and M. capitata (right) corals at the end of the atrazine exposure.

14 a. b.

c. d.

Figure 8. Biomarker protein expressions in M. capitata after exposure to atrazine, (a) HSD17B8 (ANOVA, F = 1.09, P-value = 0.391), (b) SELENBP1 44kDa (ANOVA, F = 2.07, P-value = 0.158) (c) SELENBP1 53 kDa (ANOVA, F = 1.65, P-value = 0.231), (d) SOD1 (ANOVA, F = 0.63, P-value = 0.612).

L. purpurea exposure to atrazine Larval output Overall, 1,262 planulae were collected from fourteen colonies measured, averaging 9.0 larvae per coral per day before the start of the experiment. The six colonies with the highest larval output over 10 days were selected for the exposure experiments (Table S2). During the experiment, total larval output ranged from over 1,000 larvae in the third control to 60 larvae in the first atrazine treatment (Table S3). In total, approximately 1,820 coral planulae were collected from the six L. purpurea colonies over the period of the 20-day experiment (Table S3). During the exposure, the third atrazine-treated colony exhibited extreme tissue loss, so the coral was flash frozen on the 13th

15 day of the exposure and not included in the protein analyses. Therefore, only two atrazine treated corals were exposed for the entire 20-day duration. The DMSO replicate #3 also exhibited tissue loss but was not frozen until the end of the experiment. Each coral continued releasing planula larvae up to the point in which the coral was frozen. Corals released between 3 and 51 planulae on average per day during the exposure (Table S3). There was no significant relationship between the number of planulae released and the number of polyps per colony, and no apparent trend seen with larval production related to the lunar cycle or atrazine-exposed corals. A repeated measures MANOVA multivariate approach showed no significant difference in the rate in planulae production during the exposure between the controls and atrazine (MANOVA, F=0.032, P-value = 0.78) when comparing the number of planulae released per polyp every 4 days of the experiment (Figures S1, S2 and S3). The salinity ranged from 34 to 36 ppt, and averaged approximately 35.3 ppt among all test chambers, with no notable fluctuations during the experiment. Temperature ranged from a minimum of 26.5°C to a maximum of 29.8°C, fluctuating depending on the time of day (Fig. S4). The maximum light intensity was 5,511.1 Lux and also fluctuated daily (Fig. S5). L. purpurea settlement Metamorphosed and swimming larvae from the L. purpurea exposed to atrazine appeared normal and did not have apparent deformities. Percent settlement ranged from 0 to 100% among both the treatment and controls. Although there was a large spread in the data, the average percent settlement in the DMSO and atrazine treatments was 44% and 30%, respectively (Fig. S6). After performing a Welch’s T-Test, there was no significant difference found between the settlement means of the atrazine and control treatments (P-value = 0.086). Protein response in L. purpurea The following antibodies recognized antigens on the L. purpurea proteins: anti-CYP1A1, anti-HSD17B8, anti-HSP60, anti-SELENBP1, anti-CaM, anti-SOD1, anti-GPX-1, and anti-GSR. If found to meet the assumptions, a Welch’s T-Test was performed to compare the Western blot net intensity of the controls versus the atrazine-treated corals. If assumptions could not be met for a T-Test, the nonparametric Wilcoxon 2-Sample Test (Mann-Whitney U Test) was performed. GSR exhibited two bands at approximately 36 and 43 kDa. At the 36 kDa, protein expression only appeared in the two atrazine samples and the coral DMSO #3, the coral which had exhibited prior tissue loss. The polyclonal antibody anti-HSP60 recognized two bands at

16 approximately 30 and 60 kDa, so both were analyzed. No significant difference in protein expression was observed among the eight proteins tested (Fig. 8). However, of note was higher expression in the protein HSD17B8 compared to the controls (Fig. 8A, P-value = 0.055). In addition, two controls had higher expression of HSP60 at 60 kDa than the atrazine samples and the control which had exhibited tissue loss.

17 a. b. c.

d. e. f.

g. h. i.

18 Figure 9. Biomarker protein expressions in L. purpurea after exposure to atrazine. (a) HSD17B8 (T-Test, t = 3.24, P-value = .055), (b) HSP60 30 kDa (T-Test, t = -0.37, P-value = 0.74), (c) HSP60 60kDa (T-Test, t = -1.28, P-value = 0.33), (d) Selenium binding protein 1 53kDa (T-Test, t = 0.42, P-value = 0.71), (e) Selenium binding protein 1 44kDa (T-Test, t = 2.78, P-value = 0.11), (f) Calmodulin (T-Test, t = 0.39, P-value = 0.73), (g) SOD1 (T-Test, t = 0.70, P-value = 0.54), (h) GPx-1 (T-Test, t = -0.023, P-value = 0.98), (i) CYP1A1 (T-Test, t = 0.76, P-value = 0.58), (j) GSR (T-Test, t = 0.07, P-value = 0.95).

19 DISCUSSION

Impact of Atrazine on Coral Fertilization In urbanized areas of Hawai‘i, nonpoint source pollution contaminates streams and coastal waters after periods of rainfall. This pollution may include sediment, nutrients, pathogens, or toxic contaminants and chemicals. In reefs neighboring farmland, particularly on Maui and Moloka‘i where atrazine is being applied to grow crops, adult corals and gametes are potentially exposed to this herbicide through runoff. This study sought to address whether the chemical atrazine impacts various life stages of Hawaiian reef-building corals. Once released from the parent colony, coral gametes are likely to come into contact with various toxicants in the water column which may impede further development. Fertilization and cell division in corals has been shown to be disrupted by various land-based contaminants, including heavy metals (Goh, 1991; Hédouin & Gates, 2013; Leigh-Smith et al., 2018), insecticides (Markey et al., 2007), and petroleum products (Negri & Heyward, 2000). In this study, M. capitata egg- sperm bundles were exposed to either 2, 10, 50, 100, 200, of 1000 μgl-1 of atrazine. Results suggest that exposure to ecologically relevant doses of atrazine does not reduce or impact fertilization success or embryonic development in M. capitata embryos. This is perhaps not surprising, because atrazine acts as an endocrine disruptor, and endocrine-like signaling has only been detected in adult corals (Tarrant et al., 1999; Tarrant 2005). Although no previous report has tested the impact of atrazine on the fertilization of corals, previous studies have tested this with other marine animals, including fish, amphibians, and tunicates. Bringolf et al. (2004) found a slight but not statistically significant decrease in percent fertilization of fathead minnow eggs exposed to 5 and 50 μgl-1 atrazine. Additionally, embryos of the ascidian Ciona intestinalis failed to develop past the 2-4 cell stage when exposed to 0.1, 1, and 10 μM atrazine (Cangialosi et al., 2013). Unlike the findings reported in Cangialosi et al. (2013), in which 0% of ascidian embryos developed past the 2-4 cell stage, the M. capitata embryos developed past the 16-cell stage in all atrazine treatments tested. During the M. capitata spawning in June 2018, the fertilization rates among replicates had high variability. However, this experiment had substantial statistical power from a large sample size of 6 replicates per treatment, and approximately 20 embryos scored per treatment per hour. Over 770 coral embryos that were exposed to atrazine were scored for fertilization, which is a comparable sample size to similar coral fertilization studies (eg., Victor & Richmond, 2005;

20 Hédouin & Gates, 2013). Fertilization failures during coral spawning may be due to low sperm motility from ocean pollution, warming and acidification. For instance, Morita et al., 2009 found that relatively small decreases in pH greatly reduced sperm flagellar motility in the broadcast spawner Acropora digitifera. Reproductive failures were also observed with the mushroom coral Fungia scutaria during spawning events at the Hawai‘i Institute of Marine Biology in 2018 and 2019. Conversely, a fertilization study using the same protocols as this study found 100% fertilization in M. capitata in the summer of 2015, which shows that the methods used here are robust (Diu et al., 2015).

Impact of Atrazine on Coral Larvae Because atrazine is lipophilic, I hypothesized that atrazine would readily diffuse across the lipid bilayer of coral gametes, making them particularly susceptible to atrazine, as eggs of hermatypic corals are primarily made of lipids (Arai et al., 1992). Because M. capitata larvae already contain symbiotic zooxanthellae passed on from their parents through vertical transmission, I predicted that atrazine could interfere with photosynthesis of the symbionts, leading to larval mortality from lack of energy. However, M. capitata larvae exhibited high levels of survival when exposed to 2 μgl-1 or 10 μgl-1 atrazine for the period of 3 days. Because M. capitata larvae have been observed to survive over 200 days and may rely on their lipid content as a long-term energy source, 3 days probably was not a sufficient amount of time for the larvae to be impacted by atrazine through the mechanism of disrupting photosynthesis (Benson et al., 2004). In this study, I found that the number of larvae released by L. purpurea during the exposure largely varied depending on the individual coral colony. Although collected from the same site, each colony could have differed in their energy reserves, affecting the number of planulae produced. This study provides evidence that exposure to an ecologically relevant dose of atrazine (1 μgl-1) does not prevent coral from releasing planulae, or from said planulae settling and metamorphosizing. Of the colonies tracked, they continued releasing larvae almost every day for a period of over 30 days. If atrazine had reduced coral fecundity or impacted the larvae, we would have expected to see a more severe decrease in larval output, or noticeable deformities such as the planulae exhibited in Downs et al. (2016), suggesting that atrazine is not toxic to early life stages of coral.

21 Coral Biomarkers Although we cannot talk to corals, we can assess them as ‘patients’ by using biomarkers. The report A Decision Framework for Interventions to Increase the Persistence and Resilience of Coral stresses the need for developing biomarkers for coral health and alleviating local stressors (National Academies of Sciences, Engineering, and Medicine, 2019). At the present, we remain at the discovery phase of coral biomarkers. Once discovered, biomarkers must be tested for scalability and transferability before being implemented in the field for conservation value (Parkinson et al., 2019). The protein biomarkers identified in this study may be tested with other coral proteins to study coral stress responses from exposure to other toxicants or stressors of interest. This study found a down-regulation in the protein HSD17B8 in atrazine-exposed M. capitata, and a slight up-regulation of HSD17B8 in L. purpurea exposed to atrazine. HSD17B8 is a short-chain dehydrogenase (SDR) involved in steroid and lipid metabolism and catalyzes the conversion of estradiol to estrogen (Luu-The, 2001). SDR’s are highly conserved, as Tarrant et al. (2009) found a homolog of the HSD17B8 gene in the cnidarian Nematostella vectensis. The different expression of HSD17B8 in the atrazine-exposed corals indicates that atrazine could be disrupting normal steroidogenesis in corals. Similarly, Yu et al. (2009) found a significant upregulation in gene expression of HSD17B8 from Sertoli cells exposed to developmentally toxic phthalate esters. M. capitata and L. purpurea could have responded differently to atrazine due to genotypic differences, as coral genotypes have been shown to respond uniquely to the same stressors (Rodrigues et al., 2008; Cunning et al., 2016). Furthermore, unique reproductive modes as well as the point in the species’ gametogenesis cycle could have been factors affecting the expression of HSD17B8. The M. capitata corals were just beginning to form gametes in December, while the L. purpurea corals were exposed during their reproductive peak in October, so already had fully developed planulae. In the future, the coral’s mode of reproduction and stage in its reproductive cycle should be taken into consideration when analyzing the effects of potential toxicants. Previous studies (K. Tisthammer, personal communication and in prep.) used liquid chromatography tandem mass spectrometry (LC-MS/MS) to investigate the sublethal effects of various pollutants, including atrazine, PCBs, and Roundup® (the commercial brand of glyphosate) on Porites lobata. Tisthammer found a significant number of differentially expressed proteins in the corals exposed to these toxicants, including proteins related to detoxification,

22 antioxidation, cell proliferation regulation, and proteolysis. For her atrazine experiment, she found that corals exposed to 1 and 10 ppb of atrazine both elicited a proteomic response, in which 202 proteins were up-regulated in both treatments, and 147 proteins were down-regulated compared to the controls. An nMDS plot showed that the corals exposed to atrazine had a unique protein signature compared to the corals exposed to PCBs and Roundup®. These data support the findings in this study that corals respond to atrazine at their metabolic level by shifting protein expression for various functions.

Limitations Although Western blots have previously been used to detect proteins in coral (eg., Rougée et al., 2006; Downs & Downs, 2006), its application to corals is relatively new, with various challenges. Because most antibodies are designed to recognize protein sequences from model organisms like mice and rabbits, not all antibodies will recognize and bind to coral proteins. In order to find antibodies that will work, the sequences of the antibody and coral must be compared, and the antibody tested by running an SDS-PAGE and Western blot with the desired coral protein. Often, an antibody that would work for one coral species would not work with another, perhaps due to interspecies differences in coral protein structure. In addition, Western blotting is a semi-quantitative method, so in order to compare protein signals between different blots, they must be processed under the same conditions (the amount of protein loaded, number of washes, etc.) and the values normalized to a control or house-keeping gene. This study used polyclonal antibodies, which are generally more inexpensive and have higher affinity than monoclonal antibodies. However, polyclonal antibodies have a higher chance of cross-reactivity because they react with different epitopes of an antigen, which may explain the cross-reactivity observed in some M. capitata blots. In addition, each batch of polyclonal antibodies produced by the same manufacturer could vary in their antigen targets, making it difficult to repeat results with a new batch of the same antibody. In the future, polyclonal antibodies may be replaced with monoclonal antibodies to reduce non-specific binding. Another option would be to design polyclonal antibodies using cDNA sequences from coral or other invertebrates to ensure that the antibodies work, as was done in Downs & Downs (2006). Despite their limitations, Western blots remain a relatively cheap and easy way to examine the presence or absence of proteins. If the correct protocols are followed, the bands

23 from imaging the blot can be quantified in terms of protein expression. Extra steps were taken in this study to ensure that the blots were optimized in order to reduce the likelihood of false positives or negatives. Of the two commonly used methods for protein quantification in Western blotting, the BCA assay was selected for being more sensitive than its counterpart, the Bradford assay. The use of precast polyacrylamide gels produced more consistent results. Each blot was stained with Ponceau S Solution, and blots that appeared to have unequal protein transfer were discarded. Non-specific binding was reduced as much as possible by optimizing the antibody concentrations, and by increasing the number of washing steps. Chemiluminescence was used for detection, which is a method sensitive enough to detect very small concentrations of protein (Alegria-Schaffer et al., 2009). Although there are certain limitations inherent to Western blotting, the protocols were standardized and optimized for successful quantification of the proteins.

Future Steps The next research studies should investigate whether atrazine disrupts gonad structure and the formation of gametes in corals, since the chemical has been shown to disrupt development of sex organs in other animals (Hayes et al., 2002, Spanò et al., 2004, Rey et al., 2009, Victor-Costa et al., 2010). To test this, corals would need to be exposed to atrazine prior to and during their gametogenesis cycle, which would require a much longer exposure time, since corals can take months to develop eggs (Padilla-Gamiño et al., 2014). We would also expect to see more of an impact from atrazine on the zooxanthellae, since atrazine is a Photosystem II inhibitor. Techniques such as pulse-amplitude modulation (PAM) fluorometry can be utilized to study the extent to which atrazine impacts the photosynthetic efficiency in the symbionts of Hawaiian corals. In addition, the expression of other proteins related to steroidogenesis such as stAR-related lipid transfer (START) protein and cytochrome P450 17A1 (CYP17A) could be investigated to better characterize atrazine’s effect on steroidogenesis in corals. The relevance of the other biomarkers tested in this study are explored in Appendix B. Although the corals exposed to atrazine did not demonstrate acute toxicity responses in this study, atrazine remains a chemical of concern for coral reef ecosystems. For instance, there is reason to believe that atrazine may be disrupting reproduction in reef fish, as atrazine has been shown to reduce spawning events and increase gonad abnormalities in other fish species (eg.,

24 Tillitt et al., 2010; Papoulias et al., 2014) Atrazine may also alter phytoplankton and algal communities, leading to indirect trophic effects on coral ecosystems (eg., Fairchild et al., 1998; Starr et al., 2016). Since little is known about how atrazine may affect reef organisms, future ecotoxicology studies should investigate its effects on reef fish, marine phytoplankton, and marine algae, by conducting controlled exposures and mesocosm experiments.

Management Implications Overall, managers must take into consideration the impact of local stressors on coral health. Studies have shown that managing local stressors improves reef health, and that corals exposed to fewer local stressors recover faster from bleaching events (Williams et al., 2016; Carilli et al., 2009). Improving water quality will reduce the stress load that corals endure. For instance, managers should focus on regulating policies that improve waste retention on land and limit wastewater discharge. Reducing watershed pollution would consequently help reduce toxicant and sediment loading into the ocean. With fewer stressors and toxicants to excrete, it is believed that corals will have more cellular energy to allocate into growth and reproduction. With the state’s current plans to increase local agriculture to improve food security in Hawai‘i, pesticide use will likely increase in the following decades. It will be even more important to conduct water quality monitoring near agricultural areas, particularly in watersheds near coral reefs, in order to understand the extent to which pesticides and other chemicals are contaminating reefs. Coral reef and watershed managers can also discuss with local farmers the possibility of using alternatives to chemical pesticides, such as biological controls or organic weed killers that are safer for the environment. Because coral spawning and larval dispersal events are predictable in Hawai‘i and elsewhere, management strategies could also be enacted to reduce threats during these sensitive time periods.

Conclusion Reef-building corals now face a severe set of challenges during our current climate crisis. Although this study found that the impact of atrazine was relatively low, there is great need to understand how land-based sources of pollution affect coral reefs. This study documents methods and molecular tools for future studies to investigate the effects of land-based pollutants on coral health. It is essential to investigate the ways in which potential toxicants impact coral at

25 various life history stages for more effective coral reef management. In addition, this study depicts the need to develop reliable biomarkers to test how stressors are impacting coral at a sublethal, molecular level. Increasing reef resilience by reducing local stressors will promote healthy coral populations in the future, which will benefit both the inhabitants of the reef and the people who depend on them.

APPENDIX A: Supporting Information

List of Tables Table S1. Percent fertilization for M. capitata embryos………………………………………...28 Table S2. Larval output of L. purpurea colonies before the start of the experiment….………...29 Table S3. Larval output of six L. purpurea colonies during the experiment……………………29

List of Figures Figure S1. A comparison of average larval output in planula larvae per day before and after the experiment………………………………………………………………………………………..30 Figure S2. The number of planula larvae released during the experiment…..………………….30 Figure S3. A comparison of average larval output in planula larvae per day before and after the experiment……………………………………………..…………………………………………31 Figure S4. Temperature during the L. purpurea atrazine exposure……………………………..31 Figure S5. Light intensity during the L. purpurea atrazine exposure...... 32 Figure S6. Settlement in Leptastrea larvae……………………………………………………...32

27 Treatment Average Percent Fertilization 2h 3h 4h 5h FSW 9.5 ± 11.5 32.9 ± 40. 6 41.07 ± 26. 85 47.15 ± 37.32

DMSO 0 3.9 ± 4.9 15.4 ± 14.8 21.4 ± 8.6

2 μgl-1 ATZ 11.4 ± 27.9 30.0 ± 36.4 53.2 ± 31.4 56.3 ± 27.1

10 μgl-1 ATZ 3.3 ± 4.1 16.3 ± 6.5 38.8 ± 31.6 21.3 ± 8.6

50 μgl-1 ATZ 10.6 ± 15.9 39.4 ± 31.5 47.4 ± 34.7 56.3 ± 27.1

100 μgl-1 ATZ 4.0 ± 6.3 25.2 ± 22.1 40.3 ± 32.1 45.4 ± 21.5

200 μgl-1 ATZ 7.7 ± 8.8 25.0 ± 33.0 28.8 ± 23.7 50.1 ± 34.7

1000 μgl-1 ATZ 2.5 ± 4.3 25.4 ± 34.0 28.2 ± 35.1 50.1 ± 28.4

Table S1. Average percent fertilization with standard deviations for M. capitata embryos fixed 2, 3, 4, and 5 hours after fertilization in controls (FSW and DMSO) and atrazine treated groups (2, 10, 50, 100, 200, 1000 μgl-1).

28 Treatment + Replicate Total Larval Output Average Larval Average Larval over 11 days pre- Output per day (# Output per polyp per exposure (# larvae) larvae/day) day (# larvae/polyp/day) DMSO 1 74 6.72 0.061 DMSO 2 342 31.09 0.451 DMSO 3 131 11.09 0.097 Atrazine 1 146 13.27 0.170 Atrazine 2 441 40.09 0.371 Atrazine 3 76 6.91 0.031

Table S2. Larval output of six L. purpurea colonies before the start of the experiment.

Treatment + Total Larval Average Larval Exposure time Average Larval Replicate Output Output per day (days) Output per polyp (# larvae) (# larvae/day) per day (# larvae/polyp/day) DMSO 1 71 3.55 20 0.032 DMSO 2 266 13.30 20 0.193 DMSO 3 1,031* 51.55* 20 0.452* Atrazine 1 60 3 20 0.038 Atrazine 2 301 15.05 20 0.139 Atrazine 3 91 7 13 0.032

Table S3. Larval output of six L. purpurea colonies during the experiment. *Minimum number; some days larvae were too numerous to accurately count.

Figure S1. A comparison of average larval output in planula larvae per day before and after the experiment for each coral (DMSO and atrazine treatments) with standard error bars.

Figure S2. The number of planula larvae released during the experiment over a period of 20 days for each individual colony exposed to either atrazine or a DMSO solvent control.

Figure S3. A comparison of average larval output in planula larvae per day before and after the experiment for each coral (DMSO and atrazine treatments).

Figure S4. Temperature in degrees Celsius during the L. purpurea atrazine exposure.

Figure S5. Light intensity in Lux during the L. purpurea atrazine exposure.

Figure S6. Settlement in Leptastrea larvae whose parental colony was exposed to atrazine or a DMSO control (T-Test, P-value = 0.086). On the x-axis, Days Exposed represents the number of days the parent colony was exposed when the larvae were collected.

32 APPENDIX B: Background on Biomarkers

SELENBP1 participates in intra-Golgi protein transport and has been hypothesized to play a role in detoxification pathways (Porat et al., 2000). Oakley et al. 2016 found that symbiotic Aiptasia showed upregulation of SELENBP1 compared to their aposymbiotic counterparts (Oakley et al., 2016). Heat shock proteins are a type of molecular chaperone protein that have been shown to increase expression after exposure to thermal stress and other stressful conditions (Sørensen et al., 2003). Various studies have used heat shock proteins as biomarkers to detect a stress response in corals (Downs et al., 2012; Rougée et al., 2006; Wecker et al., 2018). Calmodulin, a calcium modulated protein, is an essential calcium binding protein in eukaryotes necessary for proper physiological functioning (Means & Dedman, 1980). Morita et al. (2008) validated the presence of calmodulin in the sperm of the coral Acropora digitata and found that calmodulin is involved in the regulation of sperm flagellar motility. Herbicides such as atrazine that interfere with photosynthetic electron transport have the potential to increase formation of reactive oxygen species (Bowler et al., 1992). Thus, corals may show differential expression in proteins related to oxidative stress after exposure to atrazine. Superoxide dismutase is an antioxidant enzyme which reacts with superoxide radicals to produce hydrogen peroxide, defending cells from protein damage and DNA mutation (Bowler et al., 1992). An increase in SOD activity has been reported in zebrafish after exposure to atrazine (Blahová et al., 2013). A separate study found that SOD was inhibited in earthworms after living in atrazine-spiked artificial soil (Song et al., 2009). Previous studies have found an increase in SOD activity in coral after exposure to other toxicants including microplastics and the insecticide chlordecone (Tang et al., 2018; Wecker et al., 2018). The enzymes glutathione peroxidase-1 (GPx-1) and glutathione reductase (GSR) also protect cells from oxidative damage. GPx-1 reduces hydrogen peroxide to water, preventing the hydroxyl radical from forming and causing damage to lipids, DNA, and proteins (Lubos et al., 2011). GSR catalyzes glutathione disulfide to the antioxidant glutathione, which reduces oxidative stress. Downs et al. (2006) found that the coral Madracis mirabilis exhibited a decrease in GPx-1 protein levels after exposure to Irgarol 1051 (Downs & Downs, 2006). Cytochrome P450 enzymes (CYP) are heme-containing monooxygenases involved in metabolizing xenobiotics including pesticides. When toxic compounds enter an animal’s body, the monooxygenase system is responsible for producing water-soluble metabolites and

33 conjugates for the organism to excrete (Walker et al., 1996). CYP1A1, a component of Phase 1 xenobiotic metabolism induced by polycyclic aromatic hydrocarbons (PAHs), has been used as a biomarker in previous coral ecotoxicology studies (e.g., Downs et al., 2012; Rougée et al., 2006; Wecker et al., 2018). For instance, Rougée et al. 2006 found upregulation of CYP1A1 in Pocillopora damicornis after exposure to fuel oil (Rougée et al., 2006). The expression of the proteins described above were all tested as biomarkers in this study to explore the sublethal effects of atrazine on corals.

34 REFERENCES

Alegria-Schaffer, A., Lodge, A., & Vattem, K. (2009). Performing and optimizing Western blots with an emphasis on chemiluminescent detection. In Methods in enzymology (Vol. 463, pp. 573-599). Academic Press.

Arai, T., Kato, M., Heyward, A., Ikeda, Y., Iizuka, T., & Maruyama, T. (1992). Lipid composition of positively buoyant eggs of reef building corals. Coral Reefs, 12, 71–75.

Atkinson, S., & Atkinson, M. J. (1992). Detection of estradiol-17β during a mass coral spawn. Coral Reefs, 11(1), 33–35.

Álvarez, N. B., Avigliano, L., Loughlin, C. M., & Rodríguez, E. M. (2015). The adverse effect of the herbicide atrazine on the reproduction in the intertidal varunid crab Neohelice granulata (Dana, 1851). Regional Studies in Marine Science, 1, 1–6.

Benson, A. A., Patton, J. S., Bulletin, S. A. A. R., 1978. (1978). Energy exchange in coral reef ecosystems. Repository.Si.Edu.

Best, J. A., & Weber, J. B. (1974). Disappearance of s-Triazines as Affected by Soil pH Using a Balance-Sheet Approach. Weed Science Society of America, 364–373.

Blahová, J., Plhalová, L., Hostovský, M., Divišová, L., Dobšíková, R., Mikulíková, I., et al. (2013). Oxidative stress responses in zebrafish Danio rerio after subchronic exposure to atrazine. Food and Chemical Toxicology, 61, 82–85.

Bringolf, R. B., Belden, J. B., & Summerfelt, R. C. (2004). Effects of atrazine on fathead minnow in a short-term reproduction assay. Environmental Toxicology and Chemistry, 23(4), 1019–1025.

Bowler, C., Montagu, M. V., & Inzé, D. (1992). Superoxide Dismutase and Stress Tolerance. Dx.Doi.org, 43(1), 83–116.

Cangialosi, M. V., Mansueto, V., & Faqi, A. S. (2013). Bisphenol A (BPA) and atrazine inhibit the embryonic development of Ciona intestinalis (Ascidiacea, Urochordata). Caryologia, 66(2), 97–102.

Carilli, J. E., Norris, R. D., Black, B. A., Walsh, S. M., & McField, M. (2009). Local Stressors Reduce Coral Resilience to Bleaching. Plos One, 4(7), e6324.

Cunning, Ross, Raphael Ritson-Williams, and Ruth D. Gates. "Patterns of bleaching and recovery of Montipora capitata in Kāne‘ohe Bay, Hawai ‘i, USA." Marine Ecology Progress Series 551 (2016): 131-139.

Diu, S., Murphy, J., Brown, N., Tisthammer, K., Richmond, R.H. (2015). The effects of the pesticides glyphosphate and Roundup on fertilization and larval settlement in the coral Montipora capitata. Unpublished manuscript.

35 Downs, C., & Downs, A. (2006). Preliminary Examination of Short-Term Cellular Toxicological Responses of the Coral Madracis mirabilis to Acute Irgarol 1051 Exposure. Archives of Environmental Contamination and Toxicology, 52(1), 47–57.

Downs, C. A., Ostrander, G. K., Rougée, L., Rongo, T., Knutson, S., Williams, D. E., et al. (2012). The use of cellular diagnostics for identifying sub-lethal stress in reef corals. Ecotoxicology, 21(3), 768–782.

Downs, C. A., Kramarsky-Winter, E., Segal, R., Fauth, J., Knutson, S., Bronstein, O., ... & Pennington, P. (2016). Toxicopathological effects of the sunscreen UV filter, oxybenzone (benzophenone-3), on coral planulae and cultured primary cells and its environmental contamination in Hawaii and the US Virgin Islands. Archives of environmental contamination and toxicology, 70(2), 265-288.

Fairchild, J. F., Ruessler, D. S., & Carlson, A. R. (1998). Comparative sensitivity of five species of macrophytes and six species of algae to atrazine, metribuzin, alachlor, and metolachlor. Environmental Toxicology and Chemistry: An International Journal, 17(9), 1830-1834.

Forsman, Z. H., Concepcion, G. T., Haverkort, R. D., Shaw, R. W., Maragos, J. E., & Toonen, R. J. (2010). Ecomorph or Endangered Coral? DNA and Microstructure Reveal Hawaiian Species Complexes: Montipora dilatata/flabellata/turgescens & M. patula/verrilli. Plos One, 5(12), e15021–10.

Gardner, S. G., Raina, J. B., Ralph, P. J., & Petrou, K. (2017). Reactive oxygen species (ROS) and dimethylated sulphur compounds in coral explants under acute thermal stress. Journal of Experimental Biology, 220(10), 1787-1791.

Gassman, N. J. (1993). Aspects of the steroid biochemistry of the scleractinian coral, Favia fragum.

Goh, B. (1991). Mortality and settlement success of Pocillopora damicornis planula larvae during recovery from low levels of nickel.

Grafeld, S., Oleson, K. L. L., Teneva, L., & Kittinger, J. N. (2017). Follow that fish: Uncovering the hidden blue economy in coral reef fisheries. Plos One, 12(8), e0182104– 25.

Grange, F., 2014, 2013-1014 Statewide Pesticide Sampling Pilot Project Water Quality Findings. Prepared by the State of Hawai‘i Department of Health Hazard Evaluation and Emergency Response.

Hayes, T. B., Anderson, L. L., Beasley, V. R., de Solla, S. R., Iguchi, T., Ingraham, H., et al. (2011). Demasculinization and feminization of male gonads by atrazine: Consistent effects across vertebrate classes. The Journal of Steroid Biochemistry and Molecular Biology, 127(1-2), 64–73.

36 Hayes, T. B., Collins, A., Lee, M., Mendoza, M., Noriega, N., Stuart, A. A., & Vonk, A. (2002). Hermaphroditic, demasculinized frogs after exposure to the herbicide atrazine at low ecologically relevant doses. Proceedings of the National Academy of Sciences, 99(8), 5476– 5480.

Hédouin, L., & Gates, R. D. (2013). Assessing fertilization success of the coral Montipora capitata under copper exposure: Does the night of spawning matter? Marine Pollution Bulletin, 66(1-2), 221–224.

Howard, P. H. (1991). Handbook of Environmental Fate and Exposure Data For Organic Chemicals (Vol. III). Lewis Publishers.

Jones, R. J., Muller, J., Haynes, D., & Schreiber, U. (2003). Effects of herbicides diuron and atrazine on corals of the Great Barrier Reef, Australia. Marine Ecology Progress Series, 251, 153–167.

Kench, P. S., & Brander, R. W. (2006). Wave Processes on Coral Reef Flats: Implications for Reef Geomorphology Using Australian Case Studies. Journal of Coastal Research, 22(1), 209–223.

Khan, S. U. (1978). Kinetics of hydrolysis of atrazine in aqueous fulvic acid solution. Pest Management Science, 9(1), 39–43.

Kolinski, S. P. (2004). Sexual reproduction and the early life history of Montipora capitata in Kāne'ohe Bay, O'ahu, Hawai'i.

Lavergne, J. (1982). Mode of action of 3-(3,4-dichlorophenyl)-1,1-dimethylurea. Evidence that the inhibitor competes with plastoquinone for binding to a common site on the acceptor side of Photosystem II. Biochimica Et Biophysica Acta (BBA) - Bioenergetics, 682(3), 345–353.

Leigh-Smith, J., Reichelt-Brushett, A., & Rose, A. L. (2018). The characterization of iron (III) in seawater and related toxicity to early life stages of scleractinian corals. Environmental Toxicology and Chemistry, 37(4), 1104–1114.

Lubos, E., Loscalzo, J., & Handy, D. E. (2011). Glutathione Peroxidase-1 in Health and Disease: From Molecular Mechanisms to Therapeutic Opportunities. Antioxidants & Redox Signaling, 15(7), 1957–1997.

Luu-The, V. (2001). Analysis and characteristics of multiple types of human 17β-hydroxysteroid dehydrogenase. The Journal of Steroid Biochemistry and Molecular Biology, 76(1-5), 143– 151.

Macek, K. J., K. S. Buxton, S. Sauter, S. Gnilka, and J. Dean. 1976. Chronic toxicity of atrazine to selected aquatic invertebrates and fishes. Ecological Research Series 600/ 3-76-047, Environmental Protection Agency, Washington, DC, USA.

37 Mahmood T, Yang P-C (2012). Western blot: Technique, theory, and trouble shooting. North American Journal of Medical Sciences, 4, 429–434.

Markey, K. L., Baird, A. H., Humphrey, C., & Negri, A. P. (2007). Insecticides and a fungicide affect multiple coral life stages. Marine Ecology Progress Series, 127–137.

McGregor, D. P. (2007). Nā Kua'āina. University of Hawai'i Press.

Means, A. R., & Dedman, J. R. (1980). Calmodulin—an intracellular calcium receptor. Nature, 285(5760), 73–77.

Mercurio, P., Burns, K. A., & Negri, A. (2004). Testing the ecotoxicology of vegetable versus mineral based lubricating oils: 1. Degradation rates using tropical marine microbes. Environmental Pollution, 129(2), 165–173.

Moberg, F., & Folke, C. (1999). Ecological goods and services of coral reef ecosystems. Ecological Economics, 29, 215–233.

Montilla, L. M., Ramos, R., García, E., & Cróquer, A. (2016). Caribbean yellow band disease compromises the activity of catalase and glutathione S-transferase in the reef-building coral Orbicella faveolata exposed to anthracene. Diseases of Aquatic Organisms, 119(2), 153–161.

Morita, M., Iguchi, A., & Takemura, A. (2008). Roles of Calmodulin and Calcium/Calmodulin- Dependent Protein Kinase in Flagellar Motility Regulation in the Coral Acropora digitifera. Marine Biotechnology, 11(1), 118–123.

Murphy, J. W. A., & Richmond, R. H. (2016). Changes to coral health and metabolic activity under oxygen deprivation. PeerJ, 4(2), e1956–16.

National Academies of Sciences, Engineering, and Medicine. 2019. A Decision Framework for Interventions to Increase the Persistence and Resilience of Coral Reefs. Washington, DC: The National Academies Press.

Negri, A. P., & Heyward, A. J. (2000). Inhibition of Fertilization and Larval Metamorphosis of the Coral Acropora millepora (Ehrenberg, 1834) by Petroleum Products. Marine Pollution Bulletin, 41(7-12), 420–427.

Negri, A. P., & Heyward, A. J. (2001). Inhibition of Coral Fertilization and Larval Metamorphosis by Tributyltin and Copper. Marine environmental research, 51(1), 17-27.

Nietzer, S., Moeller, M., Kitamura, M., & Schupp, P. J. (2018). Coral Larvae Every Day: Leptastrea purpurea, a Brooding Species That Could Accelerate Coral Research. Frontiers in Marine Science, 5(466), 1–12.

38 Oakley, C. A., Ameismeier, M. F., Peng, L., Weis, V. M., Grossman, A. R., & Davy, S. K. (2016). Symbiosis induces widespread changes in the proteome of the model cnidarian Aiptasia. Cellular Microbiology, 18(7), 1009–1023.

Obien, S. R., & Green, R. E. (1969). Degradation of Atrazine in Four Hawaiian Soils. Weed Science, 17(4), 509–514.

Padilla-Gamiño, J. L., & Gates, R. D. (2012). Spawning dynamics in the Hawaiian reef-building coral Montipora capitata. Marine Ecology Progress Series, 449, 145–160.

Padilla-Gamino, J. L., Hedouin, L., Waller, R. G., Smith, D., Truong, W., & Gates, R. D. (2014). Sedimentation and the reproductive biology of the Hawaiian reef-building coral Montipora capitata. The Biological Bulletin, 226(1), 8-18.

Papoulias, D. M., Tillitt, D. E., Talykina, M. G., Whyte, J. J., & Richter, C. A. (2014). Atrazine reduces reproduction in Japanese medaka (Oryzias latipes). Aquatic Toxicology, 154, 230– 239.

Parkinson, J. E., Baker, A. C., Baums, I. B., Davies, S. W., Grottoli, A. G., Kitchen, S. A., Matz, M. V., Miller, M. W., Shantz, A. A., & Kenkel, C. D. (2019). Molecular tools for coral reef restoration: beyond biomarker discovery. Conservation Letters, e12687.

Peters, E. C., Gassman, N. J., Firman, J. C., Richmond, R. H., & Power, E. A. (1997). Ecotoxicology of tropical marine ecosystems. Environmental Toxicology and Chemistry, 16(1), 12–40.

Porat, A., Sagiv, Y., & Elazar, Z. (2000). A 56-kDa selenium-binding protein participates in intra-Golgi protein transport. Journal of Biological Chemistry, 275(19), 14457–14465.

Reaka-Kudla, M. L. (1997). Biodiversity II: Understanding and Protecting Our Biological Resources (Vol. 2, pp. 83–108).

Rey, F., González, M., Zayas, M. A., Stoker, C., Durando, M., Luque, E. H., & Muñoz-de-Toro, M. (2009). Prenatal exposure to pesticides disrupts testicular histoarchitecture and alters testosterone levels in male Caiman latirostris. General and Comparative Endocrinology, 162(3), 286–292.

Richmond, R. H. (1985). Reversible metamorphosis in coral planula larvae. Mar Ecol Prog Ser, 22(1), 181-5.

Richmond, R. H. (1997). Reproduction and recruitment in corals: critical links in the persistence of reefs. In Life and Death of Coral Reefs (pp. 175–197).

Richmond, R. H., & Jokiel, P. L. (1984). Lunar Periodicity in Larva Release in the Reef Coral Pocillopora Damicornis at Enewetak and Hawai‘i. Bulletin of Marine Science, 34, 280– 287(8).

39 Richmond, R. H., Tisthammer, K. H., & Spies, N. P. (2018). The effects of anthropogenic stressors on reproduction and recruitment of corals and reef organisms. Frontiers in Marine Science, 5, 226.

Rodrigues, Lisa J., Andréa G. Grottoli, and Tamara K. Pease. "Lipid class composition of bleached and recovering Porites compressa Dana, 1846 and Montipora capitata Dana, 1846 corals from Hawaii." Journal of Experimental Marine Biology and Ecology 358.2 (2008): 136-143.

Rougée, L., Downs, C. A., Richmond, R. H., & Ostrander, G. K. (2006). Alteration of normal cellular profiles in the scleractinian coral (Pocillopora damicornis) following laboratory exposure to fuel oil. Environmental Toxicology and Chemistry, 25(12), 3181–3187.

Shafir, S., Rijn, J. V., & Rinkevich, B. (2007). Short and Long Term Toxicity of Crude Oil and Oil Dispersants to Two Representative Coral Species. Environmental Science & Technology, 41(15), 5571–5574.

Sharp, V. A., Brown, B. E., & Miller, D. (1997). Heat shock protein (Hsp 70) expression in the tropical reef coral Goniopora djiboutiensis. Journal of Thermal Biology, 22(1), 11–19.

Shigeura, G. T., & Ooka, H. (1984). Macadamia nuts in Hawai‘i: history and production.

Solomon, K. R., Baker, D. B., Richards, R. P., Dixon, K. R., Klaine, S. J., La Point, T. W., et al. (1996). Ecological risk assessment of atrazine in North American surface waters. Environmental Toxicology and Chemistry, 15(1), 31–76.

Song, Y., Zhu, L. S., Wang, J., Wang, J. H., Liu, W., & Xie, H. (2009). DNA damage and effects on antioxidative enzymes in earthworm (Eisenia foetida) induced by atrazine. Soil Biology and Biochemistry, 41(5), 905–909.

Spalding, M. D., & Grenfell, A. M. (1997). New estimates of global and regional coral reef areas. Coral Reefs, 16, 225–230.

Spanò, L., Tyler, C. R., Aerle, R. V., Devos, P., Mandiki, S. N. M., Silvestre, F., et al. (2004). Effects of atrazine on sex steroid dynamics, plasma vitellogenin concentration and gonad development in adult goldfish (Carassius auratus). Aquatic Toxicology, 66(4), 369–379.

Starr, A. V., Bargu, S., Maiti, K., & DeLaune, R. D. (2016). The Effect of Atrazine on Louisiana Gulf Coast Estuarine Phytoplankton. Archives of Environmental Contamination and Toxicology, 72(2), 178–188.

Storlazzi, C.D., Reguero, B.G., Cole, A.D., Lowe, E., Shope, J.B., Gibbs, A.E., Nickel, B.A., McCall, R.T., van Dongeren, A.R., and Beck, M.W., 2019, Rigorously valuing the role of U.S. coral reefs in coastal hazard risk reduction: U.S. Geological Survey Open-File Report 2019–1027, 42 p.

40 Sørensen, J. G., Kristensen, T. N., & Loeschcke, V. (2003). The evolutionary and ecological role of heat shock proteins. Ecology Letters, 6(11), 1025–1037.

Tang, J., Ni, X., Zhou, Z., Wang, L., & Lin, S. (2018). Acute microplastic exposure raises stress response and suppresses detoxification and immune capacities in the scleractinian coral Pocillopora damicornis. Environmental Pollution, 243(Part A), 66–74.

Tarrant, A. M., Atkinson, S., & Atkinson, M. J. (1999). Estrone and estradiol-17β concentration in tissue of the scleractinian coral, Montipora verrucosa. Comparative Biochemistry and Physiology Part a: Molecular & Integrative Physiology, 122(1), 85–92.

Tarrant, A. M. (2005). Endocrine-like signaling in cnidarians: current understanding and implications for ecophysiology. Integrative and Comparative Biology, 45(1), 201-214.

Tarrant, A. M., Reitzel, A. M., Blomquist, C. H., Haller, F., Tokarz, J., & Adamski, J. (2009). Steroid metabolism in cnidarians: Insights from Nematostella vectensis. Molecular and Cellular Endocrinology, 301(1-2), 27–36.

Tillitt, D. E., Papoulias, D. M., Whyte, J. J., & Richter, C. A. (2010). Atrazine reduces reproduction in fathead minnow (Pimephales promelas). Aquatic Toxicology, 99(2), 149– 159.

Tisthammer, K. H. (2017). Understanding the adaptive ability of corals to changing environments (Order No. 10656652). Available from Dissertations & Theses @ University of Hawai`i at Manoa; Dissertations & Theses @ University of Hawaii; ProQuest Dissertations & Theses Global. (1955177248).

Tisthammer, K. H., Forsman, Z. H., Toonen, R. J., & Richmond, R. H. (2020). Genetic structure is stronger across human-impacted habitats than among islands in the coral Porites lobata. PeerJ, 8, e8550.

Thelin, G.P., and Stone, W.W., 2013, Estimation of annual agricultural pesticide use for counties of the conterminous United States, 1992–2009: U.S. Geological Survey Scientific Investigations Report 2013-5009, 54 p.

Veron, J. E. N. (2000). Corals of the World. Australia Institute of Marine Science. Townsville: Australian Institute of Marine Science.

Victor, S., & Richmond, R. H. (2005). Effect of copper on fertilization success in the reef coral Acropora surculosa. Marine pollution bulletin, 50(11), 1448.

Victor-Costa, A. B., Bandeira, S. M. C., Oliveira, A. G., Mahecha, G. A. B., & Oliveira, C. A. (2010). Changes in testicular morphology and steroidogenesis in adult rats exposed to Atrazine. Reproductive Toxicology, 29(3), 323–331.

41 Walker, C. H., Hopkin, S. P., Sibly, R. M., & Peakall, D. B. (1996). Principles of Ecotoxicology. Taylor & Francis.

Wecker, P., Lecellier, G., Guibert, I., Zhou, Y., Bonnard, I., & Berteaux-Lecellier, V. (2018). Exposure to the environmentally-persistent insecticide chlordecone induces detoxification genes and causes polyp bail-out in the coral P. damicornis. Chemosphere, 195, 190–200.

Williams, I. D., White, D. J., Sparks, R. T., Lino, K. C., Zamzow, J. P., Kelly, E. L., & Ramey, H. L. (2016). Responses of herbivorous fishes and benthos to 6 years of protection at the Kahekili Herbivore Fisheries Management Area, Maui. PLoS One, 11(7).

Yu, X., Hong, S., Moreira, E. G., & Faustman, E. M. (2009). Improving in vitro Sertoli cell/gonocyte co-culture model for assessing male reproductive toxicity: Lessons learned from comparisons of cytotoxicity versus genomic responses to phthalates. Toxicology and applied pharmacology, 239(3), 325–336.