Toxicity of the aquatic herbicide, reward®, on the fathead minnow with pulsed-exposure proteomic profile

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

Contents lists available at ScienceDirect

Comparative Biochemistry and Physiology - Part D journal homepage: www.elsevier.com/locate/cbpd

Toxicity of the aquatic herbicide, reward®, on the fathead minnow with pulsed-exposure proteomic profile

T

⁎

Michael L. Moretona, , Bonnie P. Lob, Denina B.D. Simmonsc, Vicki L. Marlatta a

Department of Biological Sciences, Simon Fraser University, Burnaby, BC, Canada Nautilus Environmental, Burnaby, BC, Canada c Ontario Tech University, Oshawa, ON, Canada b

A R T I C LE I N FO

A B S T R A C T

Keywords: Toxicology Aquatic herbicide Diquat dibromide Fathead minnow Proteomics

The objectives of this study were to assess the lethal and sub-lethal effects of the aquatic herbicide commercial formulation, Reward® (373 g/L DB), using application scenarios prescribed by the manufacturer. Specifically, a 14 d period between applications of Reward® in a water body undergoing treatment is required, yet the effects of these ‘pulse’ exposure scenarios on aquatic wildlife such as fish are unknown. In the first experiment early life stage FHM were exposed to a continuous DB concentrations from 0.105–12.6 mg/L which yielded a larval 7 d LC50 of 2.04 mg/L as well as a significant decrease in body mass (25.0 ± 11.6%) at the 1.18 mg/L Reward® concentration. In a second experiment, FHM larvae were exposed for 24 h and then reared in clean water for 14 d followed by a second 24 h exposure to Reward®. The 16 d LC50 value was 4.19 mg/L. In a third experiment, adult FHM were exposed in a pulse/discontinuous manner to Reward® with a calculated 21 d LC50 value of 6.71 mg/L. No significant changes in gonadosomatic index or fecundity of the F1 generation's hatch success were found when eggs from exposed adults were then reared in clean water. Proteome analyses of whole FHM larvae from the discontinuous/pulse exposure showed the primary gene ontology molecular functions of the proteins in fish exposed to 3.78 mg/L DB that resulted in ~30% mortality with positive or negative differential abundance (pvalue < .2) were: structural molecule activity; identical protein binding; structural constituent of cytoskeleton; ion binding; calcium ion binding; cytoskeletal protein binding; actin binding; and, ATP binding. These findings suggest that concentrations causing adverse effects occur above the maximum concentration predicted by the manufacturer when applied according to the label (i.e. > 0.37 mg/L).

1. Introduction Diquat dibromide (DB) is a non-selective, systemic herbicide widely used for industrial and recreational control of terrestrial and aquatic weeds in Canada, the United States, and Europe ((European Commission, 2000; Health Canada, 2010; USEPA, 2002). DB is also the active ingredient in a commonly used aquatic herbicide commercial formulation Reward® (Syngenta, 2003), which is used in both aquatic and terrestrial applications. Although we could find no published reports of environmental concentrations of DB in Canadian waters, DB is highly water soluble (> 700 g/L; Shiu et al., 1990). It also exhibits high soil adsorption coefficients for silt and clay, thus binds tightly to soil and sediment causing its rapid evacuation from the water column (Ketterings et al., 2007). To what degree soil or sediment can become saturated with DB, therefore preventing clearance from the water column by sedimentation in subsequent applications, is not currently

⁎

known (Birmingham and Colman, 1983). The use of DB as an aquatic pesticide draws attention to the sensitivity of aquatic organisms. There is some indication that earlier life stages show lower survival tolerance to DB than adults as seen in the walleye (Stizostedion vitreum) which indicates a larval 24 h LC50 of 2.9 mg/L versus adult 24 h LC50 of 7.8 mg/L (Paul et al., 1994). Gilderhus (1965) reported the 96 h LC50 for rainbow trout to be 11.2 mg/L. Hardier 96 h LC50 values were seen in grass carp (Ctenopharyngodon idella; 53.0 mg/L; Salah El-Deen and Rogers, 1993) and mosquitofish (Gambubsia affins; 289 mg/L; Leung et al., 1983). According to application methods and rates published by the manufacturer, the maximum concentration of DB that could accumulate in a water body is 0.37 mg/L (Syngenta, 2003). As mandated by Canada's Pest Management Regulatory Agency (PMRA), there is a mandatory two week waiting period between multiple applications of Reward® (Health Canada, 2010). This recovery period exists to protect non-target aquatic organisms from the peripheral effects of pesticide

Corresponding author at: 8888 University Drive, Burnaby, BC V5A 1S6, Canada. E-mail address: [email protected] (M.L. Moreton).

https://doi.org/10.1016/j.cbd.2019.100635 Received 14 August 2019; Received in revised form 5 October 2019; Accepted 5 October 2019 Available online 01 November 2019 1744-117X/ © 2019 Published by Elsevier Inc.

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

M.L. Moreton, et al.

containers. Previous work has suggested that DB does not volatilize or adhere to glass surfaces when subject to aeration for 24 h (McCuaig, 2018).

use. In plants, its primary mode of action is to disrupt photosystem I by cycling electrons between NADPH and reactive oxygen species (Jones and Vale, 2000). In vertebrates, recent research has demonstrated that adverse effects in non-target animals are due to increasing intracellular quantities of reactive oxygen species (Sus scrofa; Zheng et al., 2017). There are currently no studies published that examine chronic effects such as DB's potential to impact reproduction, or growth and development, which is a more realistic possibility given predicted aquatic concentrations and the likelihood of multiple applications in aquatic environments. Fathead Minnows (FHM; Pimephales promelas) are small (0.81–3.37 g) teleosts that school in lakes, rivers and ponds across North America (Crane and Ferrari, 2015). The FHM has also been adopted in North America as a hardy and convenient model species that has been extensively characterized for toxicological research throughout each stage of its lifecycle (Ankley et al., 2001). Routine apical endpoints using the FHM model in toxicological studies include body morphometrics, development, deformity rates, behaviour, and survival (Kolok and L'Etoile-Lopes, 2005). FHM are also an established model organism for evaluating the effects of endocrine-disrupting toxicants, particularly through a reproductive assay using sexually mature adults (Jensen et al., 2001; Parrott and Blunt, 2005; OECD, 2012). Phenotypic feminization of males and changes in vitellogenin levels are indicators that a toxicant interferes with the reproductive endocrine axis (Yonkos et al., 2010; Fort et al., 2015; Zhang et al., 2015). Utilizing the FHM model species to evaluate the adverse effects of DB will aid in more accurately assessing the chronic risks of this herbicide on nontarget teleosts native to this geographic region. Advances in omics has opened up a broad world of understanding how systems respond to exogenous influences, yielding insight of organismal-level changes from the molecular drivers of physiology (Forné et al., 2010). Unlike conventional apical endpoints (e.g. bodyweight, length, survival, etc.), employing omics techniques is useful for determining molecular mechanisms of action that can help define the route of toxicity of a chemical on an organism, something that even extends to treating human disease (Ankley and Villeneuve, 2006; Amatruda and Patton, 2008). Therefore, the objectives of this study were to explore the adverse effects of Reward® using the FHM model by examining established whole-organism level effects (survival, body length, and weight) while also examining molecular level endpoints (e.g. gene expression and proteomics). In addition to measuring these overt whole body and molecular level effects of Reward® in larval and adult FHMs, the effects of Reward® on reproductive capacity, sexual dimorphism, offspring survival and hatching were assessed in adult FHM exposures to determine the potential for endocrine and developmental disruptions. The novelty of these experiments include: using a commercial formulation containing DB, Reward®, instead of DB alone; mimicking the real-world exposure scenario after aquatic applications of this herbicide by discontinuous exposure periods (as mandated by the PMRA); and performing exploratory proteomics analyses after Reward® exposures on developing FHMs.

2.2. Animals Adult FHM were imported from Aquatic Biosystems (Fort Collins, Colorado, USA) under approval from the Canadian Food Inspected Agency (import number: Q-2016-00463-4). Upon arrival at the Alcan Research Centre, Simon Fraser University (Burnaby, BC, Canada) the fish, which were separated by sex and acclimated at 25 ± 2 °C under aeration (dissolved oxygen levels > 9.0 mg/L) for 14 d in 14 L dechlorinated municipal water in 20 L glass aquaria containing one PVC breeding tile as described in OECD (2012) Test 229. Daily egg production was monitored and recorded by removing each PVC breeding tile and collecting the eggs. Eggs for the larval study were reared in aerated moderately hard water [96 mg/L NaHCO3, 60 mg/L CaSO4, 60 mg/L MgSO4, 4 mg/L KCl dissolved in deionized water] in 1.5 L glass aquaria (loading density 100–300 eggs/L). Daily water renewals (80%) were performed during embryo and larval rearing. Once hatched, larvae were transferred to 400 mL glass beakers containing 250 mL of moderately hard water (5 larvae/vessel) and reared to 24 h post-hatch, at which point the exposure experiment began. Larvae were fed live brine shrimp ad libitum (Artemia salina; Canadian Aqua Farm, Maple Ridge, BC, Canada) twice per day by administering 50–150 μL aliquots of hatched and fresh water rinsed brine shrimp such that larvae had brine shrimp available throughout the day light hours. Each beaker was fed the same amount of brine shrimp per feeding based on a consistent volume dispensed to each test vessel per feeding. Visual confirmation of feeding was monitored by observing orange coloration of the larval gastrointestinal tract post-feeding and recorded daily. 2.3. 7 d continuous larval FHM exposure to Reward® The 7 d continuous exposure experiment was conducted in quadruplicate test vessels and initiated when larvae were ≤ 24 h post-hatch with the following treatment groups: moderately hard water control; and, nominal diquat ion concentrations of 0.12, 0.37, 1.18, 3.79 and 12.12 mg/L after serial dilutions of Reward®. Unless otherwise stated, experimental methods adhered to the Test of Larval Growth and Survival Using Fathead Minnows (Environment Canada, 2011). Daily water quality was measured in one replicate test vessel in each control and treatment group in all experiments prior to water changes, and included: dissolved oxygen, conductivity, temperature, and pH. Water changes during this larval exposure were performed every 24 h, renewing 80% of the volume with freshly prepared solutions. During the daily water changes and feeding, survival was monitored and excess food and feces were removed. After the 7 d exposure period, feeding was suspended 12 h prior to euthanization. Larvae were collected by gently pouring the test solution for a single test vessel into a net; larvae were immediately euthanized in tricaine methanesulfonate (MS222; 0.4 g/L) buffered to pH = 7 with sodium bicarbonate. Larvae were removed and carefully placed onto pre-weighed aluminum weigh boats, placed into a 60 °C incubator for 24 h, then re-weighed at random. The final measurement of dry biomass was calculated for each replicate as an average per individual larva.

2. Methods 2.1. Chemical The commercial formulation of Reward® was obtained from Syngenta Canada and contains 373 g/L diquat dibromide [6,7-dihydrodipyriod (1,2-a:2′,1′-c) pyrazinediium dibromide]. Diquat dibromide is manufactured as a bromide salt but can also be expressed as diquat ion; Reward® contains a concentration of 240 g/L diquat ion. The commercial formulation, Reward®, was used to prepare all test concentrations based on a diquat ion concentration of 240 g/L. All nominal and measured concentrations represent the diquat ion concentration in the test water after serial dilutions of Reward® with dechlorinated municipal water prepared in glass or food grade plastic

2.4. 16 d pulse larval FHM exposure to Reward® FHM larvae were exposed to varying concentrations of Reward® for two 24 h period pulses that were 14 d apart, modified from the Test of Larval Growth and Survival Using Fathead Minnows (Environment Canada, 2011). All larvae used in this study were ≤ 24 h post-hatch and were collected from adult FHM (Aquatic Biosystems, Fort Collins, CO) reared and bred in the Alcan Research Centre, Simon Fraser University (Burnaby, BC) as described above. In this pulse exposure experiment, 2

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

M.L. Moreton, et al.

were distributed per vessel with a male: female ratio of 2:4 and acclimated for 2 weeks in municipal dechlorinated tap water in aquaria containing 7.5 cm diameter PVC pipe sectioned longitudinally and cut into ~9 cm long segments as an egg-laying substrate. The fish were maintained at 25 ± 2 °C and under a 16 h light: 8 h dark photoperiod. Fish were fed 2.76% bodyweight of Finfish starter #1 fish food (Ziegler Bros, Inc., PA) twice a day at least 3 h apart. Aquarium water was renewed every 48 h removing uneaten food and feces, replacing 80% of the tank volume. For the first 24 h Reward® pulse, 80% of water in each tank was removed and refilled with 120% the desired nominal concentrations. Water samples were collected as a 1 L composite from all 4 replicate tanks (250 mL/tank) at each test concentration and the control immediately following the addition of first of Reward® pulse on d 1; samples were analyzed through LC/MS-MS by ALS Environmental (Burnaby, BC) and yielded measured concentrations of 0, 0.10, 0.33, 1.17, 3.57, and 12.6 mg/L (based on the active ingredient diquat ion concentration). Following the first 24 h pulse, an 80% clean water renewal was performed and then every 48 h with clean water. On d 15, a second 24 h Reward® pulse exposure occurred as described above. Daily measures of water quality (dissolved oxygen, conductivity, temperature, and pH) were performed, and ammonia water concentrations were measured 3 times per week. Daily egg production was recorded throughout both the acclimation and exposure period. To measure hatch success in the F1 generation after Reward® pulse exposures, eggs from the control, 0.10, 0.33 and 0.17 mg/L DB test concentrations (from all 4 replicate tanks) were collected on d 20 of the experiment and distributed into 250 ml of aerated moderately hard water in 400 mL glass beakers (10 eggs/beaker; 3 beakers per treatment). Eggs were incubated at 25 ± 2 °C for 5 d with daily hatch assessments, water renewals and water quality monitoring (dissolved oxygen, conductivity, temperature, and pH). Termination of the adult FHM exposure occurred during d 20/21, with two of the four replicate tanks for each test concentration terminated on d 20 and the other two on day 21. After euthanasia in 0.4 g/L MS222 buffered to pH = 7 with sodium bicarbonate, body weight and snout-fork length were recorded, and the liver rapidly dissected and snap frozen on dry ice and transferred to long-term storage at −80 °C. The gonads were then removed, weighed and along with dissected fish heads (decapitated slightly anterior to the operculum). Fish heads were preserved in Davidson's for 24 h, followed by rinsing in tap water and immersion in 10% Neutral Buffered Formalin at room temperature for long term storage. Subsequent quantification and classification of nuptial tubercles on the snout of the preserved heads for each fish were performed and adhered to the OECD Fish Short Term Reproduction Assay test guideline (OECD, 2012).

the following treatments were included: moderately hard water control; and, Reward® concentrations of 0.12, 0.37, 1.18, 3.79 and 12.12 mg/L based on the active ingredient diquat ion. Ten larvae were placed into 250 mL of the control or DB concentrations in 400 mL glass beakers. Feeding began immediately with freshly hatched, live brine shrimp as described above. After 24 h of exposure, all test beakers underwent an 80% water renewal with clean moderately hard water. This renewal with fresh moderately hard water was repeated every 24 h for 14 d, at which point the water was removed to 20% volume and refilled with a 120% concentration of test solution to yield the desired nominal concentration for a second 24 h exposure. Feeding was suspended at least 12 h prior to termination. Larvae from each vessel were removed and euthanized with 0.4 g/L MS222 buffered to pH = 7 with sodium bicarbonate. Once euthanized, excess water was carefully blotted and the wet body weight and length were measured before snap freezing the whole body on dry ice and storing at −80 °C for future proteomics analysis. 2.5. 16 d pulse larval FHM proteomics Our exploratory proteomics approach was employed on FHM larvae from this 16 d pulse exposure. Proteome analysis was performed on individual whole body larval control fish and on fish in the 0.37 and 3.78 mg/L DB pulse separated by 14 days in clean water. For each concentration and control treatment a total of 15 individuals were tested, with 3–4 individual fish randomly selected from each of the four replicates. The previously frozen larvae were homogenized for 3–5 min at 20 Hz in 50 uL 1.0 M tris-HCl (pH 7.46) with a stainless steel bead (Retsch MM300). After centrifuging samples at 14,000 ×g for 10 mins, the supernatant was removed and maintained at −80 °C until analysis. The frozen homogenates were thawed on ice and then reduced and alkylated prior to digestion by heating in formic acid using previously described methods (Simmons et al., 2012). The sample digests were then diluted so that total protein concentration was 2 ( ± 0.2) mg/mL. Unlabeled peptide digests were separated and identified using the Waters ionKey/MS microflow LC-MS System tandem to the Xevo G2-XS QTof Quadrupole Time-of-Flight mass spectrometer. Separation was achieved by direct injection onto an iKey Peptide BEH C18 reverse phase separation device (300 Å, 1.7 μm, 150 μm × 50 mm) with a 75 min gradient (0–5 min 5% Solvent B, 5–35 min 5–40% solvent B, 35–45 min 40–60% solvent B, 45–50 min 85% solvent B, 50–75 min 5% Solvent B) using 0.1% formic acid in water (solvent A) and 0.1% formic acid in acetonitrile (Solvent B). We acquired the data in positive and sensitivity mode, with the following source settings: capillary voltage = 3.00 kV, sampling cone = 20 kV, and source offset = 80 kV, source temperature = 150 °C, desolvation gas temperature = 350 °C, desolvation gas flow = 135 L/h and cone gas flow = 140 L/h. Peptide spectra were acquired using DDA mode with a survey mass scan of 300–3000 Da. The threshold ion count to trigger tandem mass spectrometry acquisition was 2000. The gain time was 0.5 s, in continuum format, with 20 precursor ions per scan, and a scan rate of 0.5 s for each precursor. Scanning stopped after the accumulated TIC =100,000 counts or after 5 s. Peak selection priority was based upon charge state: 2 > 3 > 4 > 5 > 6+ with no ions in the include and exclude list. Collision energy was ramped for each charge state, 25–40 V for m/z 400–1000.

2.7. Data analysis Statistical analysis was performed using SPSS v. 24 (IBM Corporation, Armonk, New York, USA). Survival and body morphometric data were analyzed using one-way analysis of variance (ANOVA) followed by a Tukey's post hoc (P < .05). These data passed the criteria for normality and homogeneity of variance based on evaluating these data via the Shapiro-Wilk's test for normality and Levene's homogeneity of variance test. LC50s were calculated using the binomial method if survival dropped from 100% to 0% between two test concentrations (i.e. 2 data points) (Environment Canada, 2007). In cases where survival displayed a more gradual dose response (i.e. 3 or greater data points), the probit method or the trimmed Spearman-Karber method was employed (Environment Canada, 2007). For proteomics, spectral data were extracted, filtered, and searched against a concatenated Uniprot teleost fish database (downloaded March 15, 2018) using PEAKs Studio 8.5 (Bioinformatics Solutions Inc., Waterloo, ON). Equivalent human ortholog gene symbols for each protein ID were found using a combination of the Uniprot retrieve/ID mapping and

2.6. 21 d pulse adult FHM exposure to Reward® The adult FHM exposure adhered to the standardized Organisation for Economic Co-operation and Development Test: 229 Fish Short Term Reproduction Assay (OECD, 2009), but was modified to reflect a pulse exposure design. Specifically, adult FHMs were exposed in quadruplicate glass tanks to two 24 h exposures of waterborne Reward® that were administered 14 d apart, with rearing in clean water in between pulses and for 5 to 6 days after the second pesticide pulse. Adult FHM 3

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

M.L. Moreton, et al.

conductivity 308–405 μs/cm; the temperature 20.5–28.8 °C; and pH 6.93–8.75.

Table 1 Nominal values of diquat ion present in treatment solutions for 21 d adult exposure, based on serial dilution of concentrated Reward® herbicide compared with values measured by ALS Environmental (Burnaby, BC). Nominal concentration (mg/L)

Measured concentration (mg/L)

0 0.1156 0.37 1.184 3.788 12.124

< 0.001 0.102 0.33 1.17 3.57 12.6

3.4. 21 d pulse adult FHM exposure to Reward® The survival of the adult FHM exposed to two 24 h exposures to Reward® 14 days apart with rearing in clean water in between pulses and 5 to 6 days after the second pulse was 100% in diquat concentrations of 0.10, 0.33, 1.17 and 3.57 mg/L, but was decreased to 0% survival by 12.6 mg/L diquat ion (Fig. 3). Body weight, length, and gonadosomatic index (GSI = 100 ∗ gonad weight / bodyweight) in adult FHMs from the 0.10, 0.33, 1.17 and 3.57 mg/L diquat ion concentrations showed no significant differences compared to control, whether analyzed together or by sex (p > .05). The 21 d LC50 value for this FHM pulse exposure experiment was 6.71 mg/L. The daily range of water quality metrics were: dissolved oxygen 6.0–8.1 mg/L; the conductivity 6.2–319 μs/cm; the temperature 21.2–27.5 °C; and pH 6.91–7.89. There were no differences in average or cumulative egg production in any of the quadruplicate adult Reward® pulse exposures over the course of the 21 d (Fig. 4A and B). Fig. 4C illustrates the male tubercle score averages in the surviving fish in the Reward® treatments and the controls and no significant differences were observed. There was no evidence of tubercles developing on females (tubercle score = 0 for each female FHM from all treatments; data not shown). To measure hatch success and time to hatch in the F1 generation after exposure of adults to two 24 h pulses of Reward® administered 14 days apart, eggs from the control, 0.10, 0.33 and 0.17 mg/L diquat ion test concentrations (from all 4 replicate tanks) were collected after the second pulse and reared until hatching was complete (Fig. 5). There was a significant decrease in the average % of hatched eggs in the 0.33 mg/L diquat ion group compared to the 0.10 mg/L (86.7 ± 3.3% and 100.0%, respectively (P = .046)), but there was no difference between any of the treatment groups and control. The number of days for embryos in each treatment to reach 50% (H50) or 90% (H90) hatched was not significantly different between any treatments (mean time-tohatch values ( ± SE): H50 = 2.16–2.5 ± 0.08 d and H90 = 3.81–4.5 ± 0.15 d; P = .808 and P = .0553 respectively; oneway analysis of variance; data not shown).

BLAST tools. Gene symbols and intensity counts for each sample were then exported from PEAKs and uploaded to Metaboanalyst 3.0 (Xia and Wishart, 2016) which we used to process and filter (using default settings and inter-quantile range), scale and normalize (using median normalization and Pareto scaling). Using the principal component analysis tool, we identified and removed 3 outliers from the dataset. We also used the volcano plot tool to determine fold change compared to the control for each treatment. Gene ontologies were mapped using the Uniprot retrieve/ID mapping and inclusion of those GO columns in the mapping table. 3. Results 3.1. Water chemistry The concentration of DB in composite samples collected from quadruplicate water control and Reward® test vessels was performed during a single water collection immediately after placing the fish during the pulse adult exposure only, and these measured concentrations were similar to nominal values (mean difference from nominal concentration = 5.6 ± 2.0%; Table 1). All larval FHM data presented in this research are based on the nominal concentrations of diquat ion during Reward® exposure experiments. 3.2. 7 d continuous larval FHM exposure to Reward® Survival was unaffected at 0.12, 0.37 and 1.18 mg/L diquat ion during Reward® FHM larvae exposures, but decreased to 0% in the two highest concentrations tested (3.79 and 12.1 mg/L diquat ion; Fig. 1). After 96 h, the larval LC50 was 3.82 mg/L and 7 d larval LC50 was 2.11 mg/L. After 7 days, body weight decreased significantly for larvae in the 1.18 mg/L concentration (P = .027) but was not statistically different in the 0.12 or 0.37 mg/L treatments. Thus, the lowest observed effect concentration (LOEC) on body weight was 1.18 mg/L and this was a 25% decrease (188.5 ± 14.9 μg) compared to the average control individual larval body weights (total biomass of all individuals in a replicate divided by number of individuals) of 251.5 ± 21.8 μg (P = .027; Fig. 1). The daily dissolved range of water quality metrics were: oxygen 7.45–8.25 mg/L; conductivity 317–396 μs/cm; temperature 20.7–26.5 °C; and pH 7.52–8.68.

3.5. Proteomics In total, 1303 proteins were detected in all of FHM whole larvae sampled from the 0.37 and 3.78 mg/L diquat ion pulse exposure experiment, whereby two 24 h pulses of Reward® were administered 14 d apart. Among those, only 180 proteins were detected in enough individual FHM larval samples to reliably calculate fold change with an associated test statistic (p-value) (132 proteins for 0.37 compared to control and 121 for 3.78 compared to control). In the fish from the 0.37 mg/L DB exposure, three proteins; titin (Ttn), helicase SRCAP (Srcap), and myosin heavy chain 4 (Myh4), were increased in abundance with a P < .1. In the fish from the 3.78 mg/L DB exposure, two proteins; AHNAK nucleoprotein (Ahnak), and actin gamma 1 (Actg1) were increased in abundance with a P < .1. In addition, there were thirty-one more proteins with differential abundance levels where 0.1 < P < .2. Table 2 lists all such proteins that changed compared to the control (P < .2) in at least one of either the 0.37 or 3.78 mg/L exposures. Of those, there were 11 proteins with decreased abundance and 14 with increased abundance in the 0.37 mg/L treatment compared to the control. Additionally, there were 12 proteins that decreased in abundance and 8 that increased in abundance in the 3.78 mg/L treatment compared to the control. We also conducted a partial-least squares discriminant analysis (PLS-DA), a supervised multivariate technique, to see if the three treatments (0, 0.37, and 3.78 mg/L) could be reasonably distinguished or classified (Fig. 6a). Component 1 was the most predictive and had a significant Q2 value after 10-fold cross validation.

3.3. 16 d pulse larval FHM exposure to Reward® FHM larvae survival was unaffected at test concentrations of 0.12, 0.37 and 1.8 mg/L diquat ion during the two 24 h pulse exposures of Reward® that were administered 14 d apart, but survival significantly decreased at 3.79 and 12.1 mg/L diquat ion (Fig. 2). The 3.79 mg/L diquat ion concentration decreased survival to 72.2 ± 6.0% (P = .00038) and at 12.1 mg/L diquat ion, 100% mortality occurred after the 2nd 24 h pulse. No significant differences were observed in body length or weight compared to the controls. The 16 d LC50 value for this exposure experiment was 4.19 mg/L. The daily dissolved range of water quality metrics were: dissolved oxygen 6.57–8.41 mg/L; the 4

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

M.L. Moreton, et al.

A 100

Survival (%)

80 60 40 20 0 0

0.12 0.37 1.18 3.79 12.1 Diquat ion (mg/L)

Mean Individual Weight (µg)

B 300

a

ab

250

a b

200 150 100 50 0 0

0.12 0.37 1.18 3.79 12.1 Diquat ion (mg/L)

Fig. 1. The effects of continuous 7 d Reward® exposures on FHM larval A) survival and B) mean individual larval weight (total dry weight of all individuals in a replicate divided by number of individuals). Values presented are means of 4 replicates ± standard error (5 larvae/replicate and 4 replicates per test concentration). Different superscripts indicate significant differences between treatments (one-way analysis of variance followed by a Tukey's post-hoc, P < .05). Nominal exposure test concentrations were prepared using diluted Reward® Landscape and Aquatic Herbicide concentrate that contained 240 g/L diquat ion.

LC50 of 3.82 mg/L and a 7 d LC50 of 2.11 mg/L DB on 8 d post-hatch FHM larvae, which is lower than previous reports for the pure active ingredient in rainbow trout of unknown ages (96 h LC50 11.2 mg/L DB; Gilderhus, 1965). In addition, sub-lethal adverse effects on growth were observed at lower concentrations with a decrease in larval body weight at 1.18 mg/L in the present 7 d larval continuous exposure to the commercial formulation (Reward®). Although no effects on growth of FHM larvae or adults were evident after two 24 h pulse exposures ranging from 0.11 to 12.12 mg/L DB, lethality was observed and the LC50 values of 4.19 mg/L and 6.71 mg/L were obtained in these two life stages, respectively. These data show that larval fathead minnow are more sensitive than adult life stages to this commercial formulation and, as expected, sub-chronic continuous exposure elicits greater mortality and sub-lethal effects (i.e. decreased growth) compared to a pulse exposure scenarios. Proteome analyses of whole FHM larvae from the

Eighty percent of the top 20 proteins based upon variable importance in projection (VIP) were also proteins with abundance fold change p-values < .2 from the 3.78 mg/L treatment (Fig. 6b), which suggests that these 16 proteins (marked in bold in Table 2) are likely predictive of effects or differences between the treatments, even with the relaxed alpha-level of 20%. The primary gene ontology molecular functions of the proteins in fish exposed to 3.78 mg/L and with positive and negative differential abundance (P < .2) are displayed in Fig. 7. 4. Discussion The objective of this study was to examine the acute and subchronic toxicity of the commercial formulation (Reward®) of the aquatic herbicide DB, on multiple life developmental stages of the fathead minnow. The continuous exposure experiments indicated a 96 h 5

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

M.L. Moreton, et al.

A a

a

a

a

100 b

Survival (%)

80 60 40 20 0 0

0.12 0.37 1.18 3.79 12.1 Diquat ion (mg/L)

0

0.12 0.37 1.18 3.79 12.1

B 10 Length (mm)

8 6 4 2 0 Diquat ion (mg/L) C

6

Weight (mg)

5 4 3 2 1 0 0

0.12 0.37 1.18 3.79 12.1 Diquat ion (mg/L)

Fig. 2. The effects of two 24 h pulses of Reward® on FHM larvae after 16 d on A) survival, B) fin-fork to snout length and C) wet weight of individuals. Values presented are the means ± standard error (10 larvae/replicate, 4 replicates per treatment). Superscript letters indicate significant differences between treatments (one-way analysis of variance followed by Tukey's post-hoc, P < .05). Nominal exposure test concentrations were prepared using diluted Reward® Landscape and Aquatic Herbicide concentrate that contained 240 g/L diquat ion. 6

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

M.L. Moreton, et al.

A

Fig. 3. Effects of two 24 h Reward® exposures over 21 d separated by 14 d of recovery in clean water on adult FHM A) survival, B) body length by sex, C) body weight by sex, and D) gonadosomatic index (GSI = 100 ∗ average gonad weight ÷ body weight). Clear bars represent females and shaded bars represent males. Values represent means ± standard error (2 males, 4 females per replicate and 4 replicates per treatment). No significant differences between treatments were observed (one-way analysis of variance followed by Tukey's post-hoc, (P < .05). Measured exposure test concentrations were prepared using diluted Reward® Landscape and Aquatic Herbicide concentrate that contained 240 g/L diquat ion.

100 Survival (%)

80 60 40 20

discontinuous/pulse exposure showed the primary gene ontology molecular functions of the proteins in fish exposed to 3.78 mg/L DB that resulted in ~30% mortality with positive or negative differential abundance (p-value < .2) were: structural molecule activity; identical protein binding; structural constituent of cytoskeleton; ion binding; calcium ion binding; cytoskeletal protein binding; actin binding; and, ATP binding. Although environmental levels of DB have not been reported in Canada, the manufacturer claims that the highest water concentration of DB is 0.37 mg/L after aquatic applications complying with the label of Reward® to treat aquatic pest plants. The manufacturer of Reward® also purports that concentrations are expected to dissipate to 0.01 mg/L within 24 h, though the conditions (e.g. water temperature, pH, turbidity, etc.) for this dissipation rate are unpublished. Based on these predicted concentrations of Reward® in the field after direct applications to water bodies according to the manufacturer label instructions the most sensitive endpoint (LOEC of 1.18 mg/L for larval growth in FHM during a 7 d continuous exposure) is 3.2 times higher than the maximum expected concentration after the aforementioned aquatic applications. It is noted that these expected concentrations are based on appropriately calculated application after the herbicide dissolves into solution represented by the entire volume of water. Thus, prudence in aquatic application methodology, including accurate calculations of water body volumes to ensure application rates do not exceed the expected maximum concentration of 0.37 mg/L is critical. Furthermore, no monitoring studies reporting environmental concentrations of DB have been performed in Canada. A California-based monitoring study reported levels of DB in surface waters present at 0.18–0.40 mg/L 4.3–4.5 d after application, though it should be noted there was also DB present in their control samples as well (Siemering et al., 2008). Therefore, whether this maximum expected environmental concentration of 0.37 mg/L is exceeded during aquatic pest plant treatments, run-off events from terrestrial applications, or if this herbicide persists in natural aquatic systems in concentrations that translate into acute or sub-chronic toxicity exposure scenarios is unknown and should be the focus of future studies. Pulse/discontinuous exposure scenarios of FHM larvae to DB in a commercial formulation (Reward®) are less toxic than the cumulative effects associated with continuous exposures at the same concentrations. While this phenomenon was expected, it has not been previously tested. And, in the case of Reward® as an aquatic herbicide with directions for use indicating applications are to be administered directly to a water body 14 days apart, this was an important line of investigation to capture a more realistic environmental exposure scenario. Indeed, whether organisms recover, become resilient or more sensitive to these discontinuous exposures to DB and other pesticides that are likely input into water in pulses is poorly studied. Nonetheless, this phenomenon of continuous exposure translating into lower concentrations being more toxic was evident based on the LC50 value for pulse exposures separated by 14 d of non-exposure being twice that observed for the 7 d continuous exposure scenario (4.19 mg/L vs. 2.11 mg/L, respectively). One other longer study of diquat ion exposure on FHM larvae conducted during the egg to fry stage during a 34 d subchronic toxicity study, reported a NOEC of 0.12 mg/L and a LOEC of 0.32 mg/L based on survival (European Commission, 2000). However, compared to the LOEC for growth of 1.18 mg/L and LC50 of 2.11 mg/L

0 0

0.102 0.33 1.17 3.57 12.6 Diquat ion (mg/L)

B

Body Length (mm)

60 50

Female

40

Male

30 20 10 0 0

0.102 0.33 1.17 3.57 12.6 Diquat ion (mg/L)

C

Body Weight (g)

3 Female

2.5

Male

2 1.5 1 0.5 0

0 0.102 0.33 1.17 3.57 12.6 Diquat ion (mg/L) D 20

GSI (%)

16

Female Male

12 8 4 0 0

0.102 0.33 1.17 3.57 12.6 Diquat ion (mg/L) 7

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

M.L. Moreton, et al.

A

Fig. 4. Effects of two 24 h Reward® exposures separated by 14 d of recovery in clean water on adult FHM A) average egg production per d, B) total egg production sum of all replicates over the course of the exposure, and C) average male tubercle score. Values presented are means of 4 replicates ± standard error (2 males, 4 females per replicate and 4 replicates per treatment). No significant differences between treatments were observed (one-way analysis of variance followed by Tukey's post-hoc, P > .05). Measured exposure test concentrations were prepared using diluted Reward® Landscape and Aquatic Herbicide concentrate that contained 240 g/L diquat ion.

Egg Production per Day

70 60 50 40 30 20 10 0 0 0.102 0.33 1.17 3.57 12.6 Diquat ion (mg/L) B

Total Egg Production (x1000)

4 3.5 3 2.5 2 1.5 1 0.5 0 0

0.102 0.33 1.17 3.57 12.6 Diquat ion (mg/L)

C 35

Tubercle Score

30 25 20 15 10 5 0 0

0.102 0.33 1.17 Diquat ion (mg/L) 8

3.57

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

M.L. Moreton, et al.

Fig. 5. Percent of eggs hatched of eggs collected from adult FHM exposed to two 24 h periods in Reward® for treatments that laid eggs on termination day. Eggs were collected, incubated in clean moderately hard water and monitored daily for hatching. The results are displayed for A) overall hatch success and B) progress of hatch success from d 1–6. Relative length of dashes represents increasing concentration from the shortest dash (control) to the solid line (highest), where concentrations represent measured diquat ion (mg/L). Values represent averages and standard error (10 eggs per treatment and 3 replicates per treatment). Superscripts indicate significant differences between treatments (one-way analysis of variance followed by Tukey's post-hoc P < .05). Measured exposure test concentrations were prepared using diluted Reward® Landscape and Aquatic Herbicide concentrate that contained 240 g/L diquat ion.

stages. The LC50 values for the FHM continuous and pulse exposures to DB in a commercial formulation in this study are similar, and in some instances lower, than the values reported in other teleosts during exposure to pure DB. For example, the acute toxicity of the continuous 96 h exposure for larval FHM (LC50 of 3.82 mg/L) in the present study is lower than presented previously for exposures conducted using the pure active ingredient, DB, on rainbow trout of unknown ages (96 h LC50 of 11.2 mg/L DB; Gilderhus, 1965). A sub-chronic toxicity study conducted on juvenile rainbow trout for 21 d (48.6 mm fingerling) reported an LC50 of 2.9 mg/L DB (European Commission, 2000), which is similar to the 7 d continuous LC50 of 2.11 mg/L for larval FHM (initiated in larvae aged < 24 h post-hatch) in the present study, which suggests the older rainbow trout are able to survive in higher concentrations of DB compared to FHM larvae. The adult FHM LC50 of 6.71 mg/L in the pulse Reward® exposures (i.e. two 24 h pulses separated by 2 weeks in clean water) in the present study is slightly lower than the range of LC50s reported for continuous acute exposures in other teleosts. For example, several studies included in the United States Environmental Protection

DB in the present 7 d study for larval FHM, the longer duration of exposure and observation of a more pronounced toxicity by Surprenant (European Commission, 2000), suggests that there was no plateau of adverse effects at these low concentrations. Additionally, in the present study no effects on larval growth were observed during the pulse exposure scenario up to and including 3.79 mg/L, while the LOEC for larval growth in the 7 d continuous exposure was 1.18 mg/L. This suggests that the prescribed 14 d period between applications of this commercial aquatic herbicide is unlikely to impact conventional whole organism (or apical) endpoints in the FHM up to 3.79 mg/L. This is further supported by the lack of effects observed over 21 days on adult FHM with respect to fecundity and growth during the pulse exposure experiments up to 3.57 mg/L diquat ion in the present study, and 100% mortality in the 12.6 mg/L diquat ion treatments. Therefore, with the highest predicted water diquat ion concentration of 0.37 mg/L DB during pulse aquatic applications of Reward® according to the manufacturer's instructions for direct application to water bodies, it appears that no lethality or effects on growth would be observed during or within days after this application scenario in FHMs of multiple life 9

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

M.L. Moreton, et al.

Table 2 Change in protein abundance in homogenates of FHM larvae exposed to DB in two 24-h periods separated by 14 days. Negative logarithm fold change values (−log2(FC)) represent the means of the individuals in the listed nominal concentration expsoure group (n = 15) compared with the means of control group (n = 15). Entries with bold font represent 80% of the top 20 variable importance in projection proteins from the Partial Least Squares Discriminant Analysis (PLS-DA). Gene symbol (human ortholog)

HSPD1 DCHS1 SYNE2 UBB MYL2 HMCN1 TNKS MYL1 FREM2 FHOD3 CAPN5 MYH3 MYH13 ACTB LRP1 TTN AHNAK ZP3 RNF213 RAG1 SPTBN1 HELZ2 LAMA3 ACTG1 RGL1 DOP1B UNC80 ZSWIM6 MYH9 TNRC6B FAT2 MYH4 PCLO ANKRD11 SRCAP

Protein name

0.37 dB

60 kDa heat shock protein, mitochondrial Protocadherin-16 Nesprin-2 Polyubiquitin-B Myosin regulatory light chain 2, ventricular/cardiac muscle isoform Hemicentin-1 Poly [ADP-ribose] polymerase tankyrase-1 Myosin light chain 1/3, skeletal muscle isoform FRAS1-related extracellular matrix protein 2 FH1/FH2 domain-containing protein 3 Calpain-5 Myosin-3 Myosin-13 Actin, cytoplasmic 1 Prolow-density lipoprotein receptor-related protein 1 Titin Neuroblast differentiation-associated protein AHNAK Zona pellucida sperm-binding protein 3 E3 ubiquitin-protein ligase RNF213 V(D)J recombination-activating protein 1 Spectrin beta chain, non-erythrocytic 1 Helicase with zinc finger domain 2 Laminin subunit alpha-3 Actin, cytoplasmic 2 Ral guanine nucleotide dissociation stimulator-like 1 Protein dopey-2 Protein unc-80 homolog Zinc finger SWIM domain-containing protein 6 Myosin-9 Trinucleotide repeat-containing gene 6B protein Protocadherin Fat 2 Myosin-4 Protein piccolo Ankyrin repeat domain-containing protein 11 Helicase SRCAP

3.78 dB

log2(FC)

P-Value

log2(FC)

P-Value

−10.98 −11.10 −2.04 −5.76 −2.39 −1.55 −5.33 2.76 −6.59 −3.52 −2.95 4.43 −8.08 −1.23 −2.97 1.92 6.31 – 7.35 – – – 2.54 – 2.88 2.94 5.09 5.64 6.67 6.82 6.99 7.26 7.42 8.69 9.44

0.15 0.18 0.25 0.16 0.18 0.39 0.20 0.31 0.14 0.17 0.16 0.34 0.19 0.18 0.16 0.01 0.14 – 0.32 – – – 0.16 – 0.16 0.16 0.15 0.17 0.16 0.18 0.19 0.07 0.17 0.16 0.06

−11.63 −9.26 −8.30 −6.40 −6.38 −6.27 −5.97 −5.14 −4.75 −4.13 −3.55 −3.55 −1.82 −1.72 −1.08 0.42 2.71 3.90 4.60 5.37 5.49 7.65 7.96 9.11 – – – – – – – – – – –

0.17 0.20 0.13 0.17 0.11 0.18 0.22 0.14 0.17 0.19 0.18 0.18 0.39 0.14 0.50 0.46 0.06 0.16 0.11 0.20 0.20 0.19 0.20 0.09 – – – – – – – – – – –

are more sensitive to DB, or if this difference in sensitivity compared to other teleosts is due to varying experimental conditions (e.g. water hardness, temperature, etc.) or the use of the pure active ingredient instead of a commercial formulation of DB. In the current study, there were no effects of DB ranging from 0.11 to 12.12 mg/L on the reproductive endocrine axis of the adult FHM after Reward® exposures administered as two 24 h pulses separated by two weeks in clean water. This interpretation is based on the lack of effects on gonadosomatic index, egg production and male nuptial tubercle numbers between control and Reward®-treated adult FHMs. Although the estrogenic biomarker vitellogenin was not measured in the present study, previous work investigating DB's potential for estrogenic activity suggests no estrogenic effects of this herbicide. Specifically, Xie et al. (2005) reported no changes in vitellogenin protein in rainbow trout exposed to 2.07 mg/L DB and Kojima et al. (2004) reported no transactivation of estrogen receptors-α or -β in Chinese hamster ovary cells tested in vitro (Kojima et al., 2004). During this FHM adult exposure experiment, eggs were collected and reared in clean water to examine effects on the F1 offspring, and although there was significantly lowered success in eggs between the 1.17 and 0.37 mg/L DB treatment groups, there was no significant effects on hatch success compared with the control group in the lowest (0.102 mg/L) or highest concentrations (1.17 mg/L). No previous studies have examined the effects of DB on hatching success in fish nor on transgenerational effects after parental exposures. These data indicate that transient exposure of adult FHM during typical application of DB in the commercial formulation, Reward®, is not likely to affect reproduction or F1 early development.

Agency (USEPA) Registration Eligibility Decision for Diquat Dibromide (USEPA, 1995) assessment to evaluate DB toxicity to fish reported adult rainbow trout 96 h LC50 values between 14.8 mg/L and < 18.7 mg/L. The adult FHM also appears to be more sensitive than adult Chinook salmon that exhibited a 48 h LC50 of 28.5 mg/L and the northern pike a 96 h LC50 of 16 mg/L after continuous exposure to pure DB (USEPA, 1995). Because of a lack of published adult FHM tests, it is difficult to conclude whether FHM are more sensitive to DB than other species, or if the commercial formulation induces greater toxicity. Further testing comparing the commercial formulation toxicity to existing DB data in different teleosts is warranted since pesticide products are what organisms will be exposed to in the environment. The sensitivity of the larval life stage of the FHM to Reward® exposures in the present study appears to be higher compared to other early life stage teleosts during pure DB exposures, though it is unclear whether the commercial formulation causes greater toxic effects or if FHM are more susceptible to DB. In the present study the 7 d larval FHM LOEC based on body weight was 1.18 mg/L, which reflects the LOEC on weight and length reported by the USEPA for early life stage FHM of 1.5 mg/L to DB alone (USEPA, 1995). These similar values may suggest similarity in toxicity between DB and Reward®. The FHM are slightly more sensitive than the embryo largemouth bass (Micropterus dolomieu), with a 96 h LC50 of 4.8 mg/L to DB. The FHM LC50 value reflects the embryo smallmouth bass (Micropterus salmoides) 96 h LC50 of 3.9 mg/L (Paul et al., 1994). In contrast, walleye embryos exposed to DB yielded a 96 h LC50 of 0.75 mg/L, being over 5 times more sensitive than the FHM larvae exposed to Reward® (de Peyster and Long, 1993). Further testing is required to ascertain if the larval walleye and FHM 10

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

M.L. Moreton, et al.

Fig. 6. Partial-least squares discriminant analysis (a) and comparison with proteins that had significantly different abundance (P-value < .20) (b).

proteins, and additional proteomic analyses at more concentrations are needed to verify this phenomenon. In addition, three proteins were significantly increased in abundance in the 0.37 mg/L DB treatment (pvalue < .1), but not in the higher DB treatment. This also indicates some unique protein changes between these two treatments and may account to some degree for the differences in survival observed between these two treatments. In terms of the toxic effects of Reward®, the primary gene ontology molecular functions of the proteins in fish exposed to 3.78 mg/L DB that resulted in ~30% mortality with positive or negative differential abundance (p-value < .2) were: structural molecule activity; identical protein binding; structural constituent of cytoskeleton; ion binding; calcium ion binding; cytoskeletal protein binding; actin binding; and, ATP binding. Although examining larval fish undergoing complicated developmental processes likely presents high background fluctuations in proteins involved in intra- and extracellular communication and cell structure and ion homeostasis, it appears as though Reward® was able to disrupt protein abundance involved in these development processes. Although, mortality was evident in the high concentration tested in this study, whether the protein changes in the larvae surviving the higher treatment and/or if larvae in the lower

4.1. Proteomics Although very few proteins (132 proteins for 0.37 mg/L DB compared to control and 121 for 3.78 mg/L DB compared to control) changed in abundance relative to the total number of proteins detected (1303) in the FHM larval pulse/discontinuous exposure experiment, those that did change still provide important insights into the effects of waterborne Reward® on larval teleosts (Valcu and Kempenaers, 2015). Indeed, partial-least squares discriminant analysis revealed that the proteome changes in larvae in the three treatments analyzed in this study (i.e. control, 0.37 and 3.78 mg/L DB) were distinguishable, and 16 of these proteins from the 3.78 mg/L DB treatment in which ~30% mortality was observed are likely predictive of effects or differences between the treatments. Although no mortality or effects on body morphometrics were observed in the lower test concentration (0.37 mg/L DB) analyzed in this experiment, 9 of these aforementioned 16 proteins predictive of adverse effects were also changed in abundance at 0.37 mg/L DB (p-value < .2). This suggests a concentration response at the proteome level in whole FHM larvae with increasing concentrations of DB resulting in more dramatic changes in these 16 11

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

M.L. Moreton, et al.

Fig. 7. Gene Ontology Molecular Functions among the proteins that had significantly different abundance (P < .20) in the 3.78 mg/L DB exposure group.

observed (Shankar et al., 2010). In the present study, the simultaneous increase in actin, cytoplasmic 2 and AHNAK protein abundance after 3.78 mg/L DB exposure provides further evidence of a potential link between these two proteins, as was observed in the AHNAK knockdown experiments in the mammary carcinoma cell study (Shankar et al., 2010). Actin, cytoplasmic 2 (a gamma actin), is proposed to be ubiquitously expressed in all eukaryotic cells and is utilized in components of the cytoskeleton and as a mediator of internal cell motility, particularly surrounding cytoskeletal rearrangement during major cellular events like mitosis (Po'uha and Kavallaris, 2015). The functions of these two proteins, AHNAK and actin, cytoplasmic 2, also support results of many of the gene ontology molecular function findings in this study (i.e. structural molecule activity; identical protein binding; structural constituent of cytoskeleton; ion binding; calcium ion binding; cytoskeletal protein binding; actin binding; and, ATP binding). We hypothesize that 3.78 mg/L DB exposure was toxic to larval FHM, at least in part, due to increased AHNAK and actin, cytoplasmic 2 protein abundance based on several major cellular processes that would be affected due to abnormal levels of these proteins, such as: cellular calcium homeostasis; intra- and/or extra-cellular vesicle formation and associated protein transport; and cytoskeletal formation. Future studies using teleost models exploring the function of these particular proteins, abnormal expression levels that are lethal, and under what toxicant exposure scenarios are necessary to test this hypothesis. Worthy of note in terms of interpreting proteomics data was that two 24 h pulses of 0.37 mg/L DB separated by 14 days did not cause mortality or changes in body morphometrics, but several proteins changed in abundance with a P-value of < 0.2 and three proteins with a P-value < .1 (titin, myosin-4 and helicase SRCAP). Titin and myosin-4 are involved in striated muscle structure and function while helicase SRCAP is involved in remodelling chromatin, regulation of

treatment with some evidence of less dramatic protein changes would result in abnormal growth and development longer term warrants further study. The overexpression of AHNAK and actin, cytoplasmic 2 in the FHM larvae exposed to 3.78 mg/L DB ion has not been previously reported, but it is likely that this contributed to some extent to the lethality observed in the fish in this test concentration. Although some studies in mammals have discovered functions involved in intra- and extra-cellular vesicle formation for AHNAK and cytoskeletal support and internal cell motility for actin, cytoplasmic 2, teleost specific studies examining these proteins are scarce. In humans AHNAK appears to be ubiquitously expressed in most tissues and organs and several biological functions for AHNAK have been reported, but it is proposed that much remains to be discovered in terms of its function(s) in vertebrates (Lee et al., 2018). For example, some of the earliest experiments using mammalian cell lines revealed that AHNAK is involved in calcium flux regulation and has been proposed to interact with s100 proteins to regulate cellular Ca2+ homeostasis (Gentil et al., 2001). In addition, reports of AHNAK participating in the formation of enlargeosomes, which are cytoplasmic vesicles, was originally observed in rat neuronal cell lines (Borgonovo et al., 2002). More recent studies supporting AHNAKs role in vesicle formation using electron microscopy indicate that AHNAK plays a functional role in the formation of cell protrusions and extracellular vesicles (Silva et al., 2016). Specifically, in human mammary carcinoma cells (MDA-MB-231) AHNAK enables cells to produce extracellular vesicles that increase neighboring fibroblast cell motility (Silva et al., 2016). Interestingly, in the mammary carcinoma cells, AHNAK knockdown experiments demonstrated that cell viability was unaffected but pseudopod retraction, reduced actin cytoskeleton dynamics, reversion of the transformation process from epithelial to mesenchymal and inhibition of cell migration and invasion were 12

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

M.L. Moreton, et al.

transcription activation, DNA repair and recombination, and increased in ~2, 7 and 9 fold in abundance, respectively, with no concomitant decreased survival or body weight. Interestingly, the abundance of these 3 proteins was not significantly different between the control and 3.78 mg/L DB treatment (not detected or P > .45) where significant mortality was observed. This finding of some unique proteins and different whole organism outcomes between these two treatments further supports that the changes in these proteins observed in the 0.37 mg/L treatment, specifically, titin, myosin-4 and helicase SRCAP, were not adversely affecting FHM larvae in this discontinuous/pulse exposure despite the statistical significance criteria of P < .1. Ongoing study and interpretation of hormetic response pathways and statistically versus biologically significant changes in proteins, particularly in teleost models, is of paramount importance in order to unravel the function(s) of poorly studied proteins in fish as well as the toxic effects of environmental contaminants at the molecular level. In addition, examining the temporal profile for proteome changes and whole organism adverse effects was beyond the scope of this study, but this is also required to better understand how to extrapolate proteome changes to whole organism adverse outcomes in the short and long term.

present and future. Aquat. Toxicol. 78, 91–102. https://doi.org/10.1016/J. AQUATOX.2006.01.018. Ankley, G.T., Jensen, K.M., Kahl, M.D., Korte, J.J., Makynen, E.A., 2001. Description and evaluation of a short-term reproduction test with the fathead minnow (Pimephales promelas). Environ. Toxicol. Chem. 20, 1276–1290. Birmingham, B.C., Colman, B., 1983. Potential phytotoxicity of diquat accumulated by aquatic plants and sediments. Water Air Soil Pollut. 19, 123–131. https://doi.org/10. 1007/BF00211798. Borgonovo, B., Cocucci, E., Racchetti, G., Podini, P., Bachi, A., Meldolesi, J., 2002. Regulated exocytosis: a novel, widely expressed system. Nat. Cell Biol. https://doi. org/10.1038/ncb888. Crane, A.L., Ferrari, M.C.O., 2015. Minnows trust conspecifics more than themselves when faced with conflicting information about predation risk. Anim. Behav. 100, 184–190. https://doi.org/10.1016/J.ANBEHAV.2014.12.002. Environment Canada, 2007. Guidance Document on Statistical Methods for Environmental Toxicity Tests. Environ. Prot. Ser, Ottawa, Canada. Environment Canada, 2011. Biological Test Method: Test of Larval Growth and Survival Using Fathead Minnows. Environ, Canada. European Commission, 2000. Directorate E - Public, animal and plant health Unit E1 Legislation relating to crop products and animal nutrition. Diquat: 1688/VI/97-final. EU Pesticide Database. Forné, I., Abián, J., Cerdà, J., 2010. Fish proteome analysis: model organisms and nonsequenced species. Proteomics 10, 858–872. https://doi.org/10.1002/pmic. 200900609. Fort, D.J., Mathis, M., Fort, C.E., Fort, H.M., Bacon, J.P., 2015. Application of endocrine disruptor screening program fish short-term reproduction assay: reproduction and endocrine function in fathead minnow (Pimephales promelas) and killifish (Fundulus heteroclitus) exposed to Bermuda pond sediment. Environ. Toxicol. Chem. 34, 1283–1295. https://doi.org/10.1002/etc.2880. Gentil, B.J., Delphin, C., Mbele, G.O., Deloulme, J.C., Ferro, M., Garin, J., Baudier, J., 2001. The giant protein AHNAK is a specific target for the calcium-and zinc-binding S100B protein: potential implications for Ca2+ homeostasis regulation by S100B. J. Biol. Chem. https://doi.org/10.1074/jbc.M010655200. Gilderhus, P.A., 1965. Effects of Diquat on Bluegills and Their Food Organisms. (doi:10.1577/1548-8640(1967)29[67:EODOBA]2.0.CO;2). Health Canada, 2010. Re-evaluation Decision Diquat Dibromide. Jensen, K.M., Korte, J.J., Kahl, M.D., Pasha, M.S., Ankley, G.T., 2001. Aspects of basic reproductive biology and endocrinology in the fathead minnow (Pimephales promelas). Comp. Biochem. Physiol. C. Toxicol. Pharmacol. 128, 127–141. Jones, G.M., Vale, J.A., 2000. Mechanisms of toxicity, clinical features, and management of diquat poisoning: a review. J. Toxicol. Clin. Toxicol. 38, 123–128. Ketterings, Q., Reid, S., Rao, R., 2007. Cation Exchange Capacity (CEC) Agronomy Fact Sheet Series. Kojima, H., Katsura, E., Takeuchi, S., Niiyama, K., Kobayashi, K., 2004. Screening for estrogen and androgen receptor activities in 200 pesticides by in vitro reporter gene assays using Chinese hamster ovary cells. Environ. Health Perspect. 112, 524–531. https://doi.org/10.1289/ehp.6649. Kolok, A.S., L’Etoile-Lopes, D., 2005. Do copper tolerant fathead minnows produce copper tolerant adult offspring? Aquat. Toxicol. 72, 231–238. https://doi.org/10.1016/j. aquatox.2004.12.010. Lee, H., Kim, K., Woo, J., Park, J., Kim, H., Lee, K.E., Kim, H., Kim, Y., Moon, K.C., Kim, J.Y., Park, I.A., Shim, B.B., Moon, J.H., Han, D., Ryu, H.S., 2018. Quantitative proteomic analysis identifies AHNAK (neuroblast differentiation-associated protein AHNAK) as a novel candidate biomarker for bladder urothelial carcinoma diagnosis by liquid-based cytology. Mol. Cell. Proteomics. https://doi.org/10.1074/mcp.ra118. 000562. Leung, T.S., Naqvi, S.M., Leblanc, C., 1983. Toxicities of two herbicides (Basagran, Diquat) and an algicide (Cutrine-plus) to mosquitofish Gambusia affinis. Environmental Pollution. Series A, Ecological and Biological 30 (2), 153–160. https://doi.org/10.1016/0143-1471(83)90012-0. In press. McCuaig, L., 2018. Hepatic Proteome and Toxic Response of Early-life Stage Rainbow Trout (Oncorhynchus mykiss) to the Aquatic Herbicide, Reward®. Simon Fraser University. OECD, 2009. Test No. 230: 21-day Fish Assay, OECD Guidelines for the Testing of Chemicals, Section 2. OECD Publishinghttps://doi.org/10.1787/9789264076228-en. OECD, 2012. Test No. 229: Fish Short Term Reproduction Assay, OECD Guidelines for the Testing of Chemicals, Section 2. OECD. doi:https://doi.org/10.1787/ 9789264185265-en. Parrott, J.L., Blunt, B.R., 2005. Life-cycle exposure of fathead minnows (Pimephales promelas) to an ethinylestradiol concentration below 1 ng/L reduces egg fertilization success and demasculinizes males. Environ. Toxicol. 20, 131–141. https://doi.org/ 10.1002/tox.20087. Paul, E.A., Simonin, H.A., Symula, J., Bauer, R.W., 1994. The toxicity of diquat, endothall, and fluridone to the early life stages of fish. J. Freshw. Ecol. 9, 229–239. https://doi.org/10.1080/02705060.1994.9664890. de Peyster, A., Long, W.F., 1993. Fathead minnow optomotor response as a behavioral endpoint in aquatic toxicity testing. Bull. Environ. Contam. Toxicol. 51, 88–95. https://doi.org/10.1007/BF00201005. Po’uha, S.T., Kavallaris, M., 2015. Gamma-actin is involved in regulating centrosome function and mitotic progression in cancer cells. Cell Cycle 14, 3908–3919. https:// doi.org/10.1080/15384101.2015.1120920. Salah El-Deen, M.A., Rogers, W.A., 1993. Changes in total protein and transaminase activities of grass carp exposed to diquat. J. Aquat. Anim. Health 5, 280–286. https:// doi.org/10.1577/1548-8667(1993)005<0280:CITPAT>2.3.CO;2. Shankar, J., Messenberg, A., Chan, J., Underhill, T.M., Foster, L.J., Nabi, I.R., 2010. Pseudopodial actin dynamics control epithelial-mesenchymal transition in metastatic

5. Conclusions One of the novel aspects of this study was testing an environmentally-realistic exposure scenario (i.e. two pulses) of a commercial formulation of the herbicide Reward®. The importance of adhering to Reward®'s label instructions when applying this herbicide directly to bodies of water is evident based on the FHM toxicity studies presented here. According to the manufacturer's instructions, the expected concentrations of DB in the water column is 0.37 mg/L immediately after single applications of Reward® separated by two weeks, thus, this herbicide does not appear to be toxic to adult or larval FHMs when used as directed. However, a LOEC of 1.18 mg/L for larval growth in FHM during a 7 d continuous exposure experiment was observed in the present study, which is 3.2 times higher than the maximum expected concentration after the aforementioned aquatic applications. It is noted that these expected concentrations are based on appropriately calculated application after the herbicide dissolves into solution represented by the entire volume of water. Thus, prudence in aquatic application methodology, including accurate calculations of water body volumes to ensure application rates do not exceed the expected maximum concentration of 0.37 mg/L is critical. Furthermore, no monitoring studies reporting environmental concentrations of DB have been performed in Canada. Therefore, whether this maximum expected environmental concentration of 0.37 mg/L is exceeded during aquatic pest plant treatments, run-off events from terrestrial applications, or if this herbicide persists in natural aquatic systems in concentrations that translate into acute or sub-chronic toxicity exposure scenarios is unknown and should be the focus of future studies. Declaration of competing interest The authors declare that they have no conflict of interest. Acknowledgments We are grateful to Dr. Chris Kennedy and Dr. David Huebert for their input on this manuscript. This work was funded by the National Contaminants Advisory Group, Fisheries and Oceans Canada, Canada (#19327). References Amatruda, J.F., Patton, E.E., 2008. Chapter 1 genetic models of cancer in. Zebrafish. 1–34. https://doi.org/10.1016/S1937-6448(08)01201-X. Ankley, G.T., Villeneuve, D.L., 2006. The fathead minnow in aquatic toxicology: past,

13

Comparative Biochemistry and Physiology - Part D 33 (2020) 100635

M.L. Moreton, et al.

Valcu, C.-M., Kempenaers, B., 2015. Proteomics in behavioral ecology. Behav. Ecol. 26, 1–15. https://doi.org/10.1093/beheco/aru096. Xia, J., Wishart, D.S., 2016. Using MetaboAnalyst 3.0 for comprehensive metabolomics data analysis. Curr. Protoc. Bioinforma. 55, 14.10.1–14.10.91. https://doi.org/10. 1002/cpbi.11. Xie, L., Thrippleton, K., Irwin, M.A., Siemering, G.S., Mekebri, A., Crane, D., Berry, K., Schlenk, D., 2005. Evaluation of estrogenic activities of aquatic herbicides and surfactants using an rainbow trout vitellogenin assay. Toxicol. Sci. 87, 391–398. https:// doi.org/10.1093/toxsci/kfi249. Yonkos, L.T., Fisher, D.J., Van Veld, P.A., Kane, A.S., McGee, B.L., Staver, K.W., 2010. Poultry litter-induced endocrine disruption in fathead minnow, sheepshead minnow, and mummichog laboratory exposures. Environ. Toxicol. Chem. 29, 2328–2340. https://doi.org/10.1002/etc.282. Zhang, Y., Krysl, R.G., Ali, J.M., Snow, D.D., Bartelt-Hunt, S.L., Kolok, A.S., 2015. Impact of sediment on agrichemical fate and bioavailability to adult female fathead minnows: a field study. Environ. Sci. Technol. 49, 9037–9047. https://doi.org/10.1021/ acs.est.5b01464. Zheng, P., Yu, B., He, J., Yu, J., Mao, X., Luo, Y., Luo, J., Huang, Z., Tian, G., Zeng, Q., Che, L., Chen, D., 2017. Arginine metabolism and its protective effects on intestinal health and functions in weaned piglets under oxidative stress induced by diquat. Br. J. Nutr. 117, 1495–1502. https://doi.org/10.1017/S0007114517001519.

cancer cells. Cancer Res. https://doi.org/10.1158/0008-5472.CAN-09-4439. Shiu, W.Y., Ma, K.C., Mackay, D., Seiber, J.N., Wauchope, R.D., 1990. Solubilities of Pesticide Chemicals in Water Part II: Data Compilation. Springer, New York, NY, pp. 15–187. https://doi.org/10.1007/978-1-4612-3434-0_2. Siemering, G.S., Hayworth, J.D., Greenfield, B.K., 2008. Assessment of potential aquatic herbicide impacts to California aquatic ecosystems. Arch. Environ. Contam. Toxicol. 55, 415–431. https://doi.org/10.1007/s00244-008-9137-2. Silva, T.A., Smuczek, B., Valadão, I.C., Dzik, L.M., Iglesia, R.P., Cruz, M.C., Zelanis, A., de Siqueira, A.S., Serrano, S.M.T., Goldberg, G.S., Jaeger, R.G., Freitas, V.M., 2016. AHNAK enables mammary carcinoma cells to produce extracellular vesicles that increase neighboring fibroblast cell motility. Oncotarget. https://doi.org/10.18632/ oncotarget.10307. Simmons, D.B.D., Bols, N.C., Duncker, B.P., McMaster, M., Miller, J., Sherry, J.P., 2012. Proteomic profiles of white sucker (Catostomus commersonii) sampled from within the Thunder Bay Area of concern reveal up-regulation of proteins associated with tumor formation and exposure to environmental estrogens. Environ. Sci. Technol. 46, 1886–1894. https://doi.org/10.1021/es204131r. Syngenta, 2003. Chemical nomenclature:physical properties:chemical stability. In: Regulation. USEPA, 1995. Reregistration Eligibility Decision (RED) Diquat Dibromide. USEPA, 2002. Report of the Food Quality Protection Act (FQPA) Tolerance Reassessment Progress and Risk Management Decision (TRED).

14