Assessing the effects of environmentally relevant concentrations of antidepressant mixtures to fathead minnows exposed over a full life cycle

Science of the Total Environment 648 (2019) 1227–1236

Contents lists available at ScienceDirect

Science of the Total Environment journal homepage: www.elsevier.com/locate/scitotenv

Assessing the effects of environmentally relevant concentrations of antidepressant mixtures to fathead minnows exposed over a full life cycle Joanne L. Parrott a,⁎, Chris D. Metcalfe b a b

Water Science and Technology Directorate, Environment and Climate Change Canada, Burlington, ON L7S 1A1, Canada Water Quality Centre, Trent University, Peterborough, ON K9J 7B8, Canada

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• Lifecycle exposure of fathead minnows to mixture of 5 antidepressants. • Exposure concentrations were 1× and 10× those in municipal wastewater effluents. • Antidepressant mixtures did not affect fish survival, growth, or maturation. • There were no significant changes in reproduction or F1 hatching success.

a r t i c l e

i n f o

Article history: Received 5 June 2018 Received in revised form 16 August 2018 Accepted 18 August 2018 Available online 19 August 2018 Editor: Kevin V. Thomas Keywords: Mixture Fathead minnow Lifecycle Antidepressants Municipal wastewater effluent

a b s t r a c t Antidepressant drugs have been detected in municipal wastewater effluents (MWWEs) at ng/L to low μg/L concentrations. We exposed fathead minnow (Pimephales promelas) over a full lifecycle to a mixture of five antidepressants at concentrations similar to a MWWE (1× AntiD Mix); venlafaxine at 2400 ng/L, citalopram at 240 ng/L, fluoxetine at 90 ng/L, sertraline at 20 ng/L, and bupropion at 90 ng/L, and 10× these concentrations (i.e. 10× AntiD Mix). Mean measured concentrations of venlafaxine, citalopram, fluoxetine, sertraline, and bupropion were 2300, 160, 110, 7 ng/L, and below detection limits, respectively, for the 1× AntiD Mix, and 33,000, 2900, 1000, 210, and 100 ng/L, respectively for the 10× AntiD Mix. During the life-cycle exposure, no significant changes were observed in survival of fathead minnows. When male fish from the exposed treatments reached maturity, their weights were increased compared to control males. There were no significant differences in condition factor, gonadosomatic index, or liver-somatic index in the exposed fish. Exposed fathead minnows produced similar numbers of eggs as control fish, and there were no changes in nest-defense behaviours of male minnows. Egg quality, % fertilization, and % hatching in F1 fry were unaffected by exposure to the antidepressants. Eggs hatched 0.5 d earlier, deformities in fry were 50% lower, and there were transient decreases in length of F1 larvae at 8 days post-hatch in offspring from the treatment with the 10× AntiD Mix. Overall, exposure to the antidepressant mixture at environmentally relevant concentrations (i.e. 1× AntiD Mix) caused no adverse effects in fathead minnows. Exposure to the 10× AntiD Mix increased the weight of adult male minnows and caused subtle effects in F1 offspring. This study is the first to assess sublethal effects in fish exposed to mixtures of antidepressants over a full lifecycle.

⁎ Corresponding author at: Environment and Climate Change Canada, 867 Lakeshore Road, Burlington, ON, L7S 1A1, Canada. E-mail address: [email protected] (J.L. Parrott).

https://doi.org/10.1016/j.scitotenv.2018.08.237 0048-9697/Crown Copyright © 2018 Published by Elsevier B.V. All rights reserved.

1228

J.L. Parrott, C.D. Metcalfe / Science of the Total Environment 648 (2019) 1227–1236

Capsule: No effects were observed in fathead minnow exposed for a lifecycle to antidepressant mixtures at environmentally-relevant concentrations. Crown Copyright © 2018 Published by Elsevier B.V. All rights reserved.

1. Introduction Antidepressants are widely prescribed to manage depression and anxiety (Vaswani et al., 2003). These drugs act at the synaptic junction to prevent the re-uptake of serotonin, noradrenaline, and other neurotransmitters, such as dopamine (Asnis et al., 2004; Caccia, 1998; Preskorn et al., 1994). In humans, some antidepressants are excreted via urine as metabolites (Caccia, 1998) and so both the parent compound and metabolites have the potential to be transported in sewage into municipal wastewater treatment plants (WWTPs). Many antidepressants and their metabolites are not removed efficiently during wastewater treatment as concentrations in effluents are similar to concentrations in effluents (Schultz and Furlong, 2008). Several antidepressant drugs have been detected in WWTP effluents and in receiving waters in Canada and the USA (Lajeunesse et al., 2008; Metcalfe et al., 2010; Schultz and Furlong, 2008; Schultz et al., 2010). Various antidepressants have also been detected in WWTP effluents and river waters in many other countries, including the United Kingdom (Kasprzyk-Hordern and Baker, 2012), Spain (Collado et al., 2014; Gros et al., 2007; Gros et al., 2010; López-Serna et al., 2012), Sweden (Gago-Ferrero et al., 2017), and Japan (Tanoue et al., 2015). A survey of this literature shows that concentrations are typically highest for venlafaxine (175–2660 ng/L), followed by citalopram (50–350 ng/L) and bupropion (150–221 ng/L), and lower for fluoxetine (5–170 ng/L) and sertraline (10–93 ng/L). From these studies, it appears that concentrations of these compounds detected in rivers immediately downstream of WWTP discharges are reduced by about 5 to 10 fold or more. Fish accumulate some antidepressants when exposed in bench-scale experiments or when exposed to municipal wastewater effluents (MWWEs). Schultz et al. (2011) exposed fathead minnows to antidepressants for 21 d and there was evidence of bioaccumulation in the brain tissue of male fish for fluoxetine, sertraline, and bupropion, but minimal bioaccumulation of venlafaxine relative to the concentrations in exposure water (Schultz et al., 2011). Round goby (Neogobius melanostomus) placed in 100% municipal wastewater for 28 d accumulated bupropion, citalopram, in their brain tissue or blood at concentrations in the low ng/g or ng/ml range, respectively (McCallum et al., 2017b). Crucian carp (Carassius carassius) and common carp (Cyprinus carpio) residing in an effluent-dominated stream in Japan accumulated sertraline and norsertraline in their plasma, liver, and brain (Tanoue et al., 2014). Metcalfe et al. (2010) reported bioaccumulation of venlafaxine, citalopram, sertraline and some of their metabolic transformation products in the muscle tissue of fathead minnows caged for 2 weeks downstream of a WWTP, with the highest mean concentration of 3.83 μg/kg of sertraline. Brooks et al. (2005) and Chu and Metcalfe (2007) detected selected antidepressants in the tissues of wild fish collected in areas impacted by discharges of MWWE. Biological effects have been observed in fish exposed in the laboratory to antidepressants in relatively short-term experiments. Larval common carp survival was reduced by exposure to 500 μg/L amitriptyline, nortriptyline, and clomipramine (Sehonova et al., 2017). Reduced survival was observed in adult male fathead minnow exposed for 21 d to nominal concentrations of 0.3 and 1.1 μg/L venlafaxine, or 5.2 ng/L sertraline, but no mortality was seen when fish were exposed to a mixture of fluoxetine at 28 ng/L, sertraline at 22 ng/L, venlafaxine at 798 ng/L and bupropion at 466 ng/L (Schultz et al., 2011). Egg production was reduced in adult zebrafish (Danio rerio) exposed for

6–7 weeks to venlafaxine at a nominal concentration of 10 μg/L (5–6 μg/L measured), and a mixture of acetaminophen, carbamazepine, gemfibrozil, and venlafaxine, all at nominal concentrations of 0.5 μg/L (Galus et al., 2013a; Galus et al., 2013b). Fish behaviour can also be affected by exposure to antidepressants. In experiments assessing seven behavioural endpoints, MargiottaCasaluci et al. (2014) showed exploratory behaviours of fathead minnow increased when fluoxetine exposure concentrations were high enough (72 μg/L for 28 d) to result in fish blood plasma concentration above the human therapeutic plasma concentration. In a series of five behavioural tests, Japanese medaka (Oryzias latipes) exposed to 100 μg/L fluoxetine for 10 d increased exploratory behaviours and reduced mirror-biting behaviours (Ansai et al., 2016). Reduced movements of larval life stages in the dark was observed in Japanese medaka exposed to a mixture of 6 psychotropic drugs, including citalopram at 1510 ng/L, fluoxetine at 300 ng/L and sertraline at 150 ng/L (Chiffre et al., 2016). Escape velocity was lower in larval fathead minnows exposed to a mixture of four antidepressants, including venlafaxine at 5000 ng/L, fluoxetine at 250 ng/L, sertraline at 250 ng/L and bupropion at 2000 ng/L (Painter et al., 2009). Aggression towards a mirror was reduced in round goby exposed for 28 d to municipal wastewater effluent containing bupropion, citalopram, and venlafaxine, as well as many other pharmaceuticals and industrial chemicals (McCallum et al., 2017b). Time to capture prey was increased in hybrid striped bass (Morone saxatilis x Morone chrysops) exposed for 3–6 d to a mixture of fluoxetine and venlafaxine at nominal concentrations between 0.36 and 4.65 μg/L (Bisesi et al., 2014). Disrupted circadian rhythm with decreased locomotion during the day was seen in adult mosquitofish (Gambusia holbrooki) exposed for 7 d to 100 μg/L of an antidepressant mixture comprised of fluoxetine, sertraline, and venlafaxine (Melvin, 2017). There are currently no data reported in the literature on the effects of an antidepressant mixture in fish exposed over an entire lifecycle, although these types of long-term studies are important (Silva et al., 2015). Lifecycle experiments are important for assessing lethal and sublethal responses to contaminants of emerging concern, as they follow fish from the egg stage through all critical stages of development. Previously, we used the fathead minnow lifecycle test to study the biological effects from exposure to a mixture of 6 pharmaceuticals and triclosan (Parrott and Bennie, 2009), the antidepressant venlafaxine (Parrott and Metcalfe, 2017), and other pharmaceuticals, such as propranolol (Parrott and Balakrishnan, 2017) and ethinylestradiol (Parrott and Blunt, 2005). In the present study, we monitored fathead minnows exposed continuously to environmentally relevant concentrations of an antidepressant mixture (i.e. 1× AntiD Mix) and to ten times the environmentallyrelevant concentrations (i.e. 10× AntiD Mix). Exposures began at the embryo stage and continued through hatching (after day 5), growth of larvae (days 7–20) and juveniles (days 30–60), maturation of fish into adults (days 70–90), and mating and reproduction (days 90–165). The endpoints monitored in the lifecycle test included survival, % egg fertilization, % deformities in hatched fry, the overall success of hatch of the F1 generation, as well as the body condition, secondary sex characteristics and gonadal sex of the adults. Nest-defense behaviour of adult males was also assessed. The concentrations in the 1× AntiD Mix were chosen to be consistent with the concentrations of five antidepressants detected in the MWWE from a WWTP serving a city in Ontario, Canada.

J.L. Parrott, C.D. Metcalfe / Science of the Total Environment 648 (2019) 1227–1236

2. Materials and methods 2.1. Flow-through exposures Flow-through exposures of fathead minnows to the antidepressant mixture were accomplished with a modified Mount & Brungs-type diluter, as described previously (Parrott and Bennie, 2009). Water flowed through the diluter and into the mixing cells that received AntiD Mix stock solutions from 2 peristaltic pumps attached to 2 separate Marriott bottles containing the antidepressant mixture solutions. The antidepressant mixture was prepared from venlafaxine-HCl (CAS # 9930078-4, 99.88% purity, lot # A-1237-145), citalopram-HBr (CAS # 5972932-7, 99.59% purity, lot # A-1191-050), fluoxetine-HCl (CAS # 5933367-4, 99.92% purity), sertraline-HCl (CAS # 79559-97-0, 99.94% purity, lot # A-1092-005), bupropion-HCl (CAS # 31677-93-7, 99.96% purity lot # A-1110-037). All antidepressants were purchased from SynFine Research (Richmond Hill, ON, Canada). Target nominal concentrations in aquaria were 23,864 ng/L of venlafaxine, 2401 ng/L of citalopram, 895 ng/L of fluoxetine, 179 ng/L of sertraline, and 868 ng/L of bupropion. An antidepressant mixture super-stock solution was prepared by adding 135 mg of venlafaxine-HCl, 15 mg of citalopram-HBr, 5 mg of fluoxetine-HCl, 1 mg of sertraline-HCl and 5 mg of bupropion-HCl to 25 mL of deionized water. No solvent controls were required for the exposures as the antidepressant super-stock and stock solutions were prepared in water. For the 10× AntiD Mix, 1 mL of super-stock was used to prepare 1 L of stock solution in deionized water. Nominal concentrations in the stock were 4.773 mg/L of venlafaxine, 0.480 mg/L of citalopram, 0.179 mg/L of fluoxetine, 0.036 mg/L of sertraline, and 0.174 mg/L of bupropion. During each 3.5 min cycle, 2 mL of this stock flowed via peristaltic pump to a 400 mL mixing chamber in the diluter, and 398 mL of lab control water was added to the 2 mL stock. This 400 mL then flowed to the tanks receiving the 10× AntiD Mix. For the 1× AntiD Mix treatment, the stock solution was diluted 10-fold and a separate 1 L Marriott bottle and peristaltic pump was used to deliver 2 mL per cycle to 400 mL of water in the mixing chamber. The 400 mL aliquots supplied 4 replicate aquaria used for each treatment with the 1× and 10× antidepressant mixtures, so the flow was 100 mL every 3.5 min into each 12 L aquarium, for a solution replacement time of about 3.4 tank volumes per day.

2.2. Analysis of antidepressants The extraction method used to extract water and wastewater to measure the concentrations of the antidepressants and some primary metabolites in the treatments were essentially as described previously by Metcalfe et al. (2010). Briefly, water samples were extracted using Oasis MCX solid phase extraction cartridges purchased from Waters (Milford, MA, USA), followed by analysis by liquid chromatography with tandem mass spectrometry (LC-MS/MS). Stable isotope internal standards, venlafaxine-d10 hydrochloride, O-desmethyl venlafaxined10 hydrochloride, citalopram-d4 hydrobromide, sertraline-d3 hydrochloride purchased from SynFine Research and fluoxetine-d5 hydrochloride purchased from Sigma-Aldrich (Oakville, ON, Canada) were added to all samples before extraction to correct for variations in the recoveries of target compounds and any matrix effects. The antidepressants and their respective stable isotope surrogates were analyzed by LC-MS/MS with an AB Sciex (Mississauga, ON, Canada) Q-Trap 5500 instrument with an electrospray ionization source, and this instrument was operated in positive ion mode. This system was equipped with an Agilent 1100 series HPLC system. The analytes were separated chromatographically using a Genesis C18 column that was 150 mm long, 2.1 mm ID and 4 μm particle size (Chromatographic Specialties, Brockville, ON, Canada) with a guard column with the same stationary phase (Genesis C18, 10 × 2.1 mm and 4 μm).

1229

MS detection was performed using multiple reaction monitoring with the ion transitions previously described by Metcalfe et al. (2010). For quantification, an internal standard method was used with a nine-point calibration graph covering the range of anticipated analyte concentrations. Internal standards (i.e. stable isotope labelled compounds) were used to correct for analyte recovery and matrix effects. The Limits of Detection (LODs) and Limits of Quantification (LOQs) were calculated as 3× and 10×, respectively, of the standard deviation of the baseline response at the retention time of the target analytes in extracts prepared from either a wastewater or a surface water matrix. Except for bupropion, the LODs and LOQs for all target compounds in wastewater were below 4 ng/L and 13 ng/L, respectively, and the LODs and LOQs for surface water were all below 3 ng/L and 9 ng/L, respectively. For bupropion, the LODs and LOQs in wastewater were 12 ng/L and 40 ng/L respectively, and the LODs and LOQs in surface water were 9 ng/L and 30 ng/L, respectively. The concentrations of the five antidepressants measured in exposure waters from the eight replicate control tanks were always bLOD. 2.3. Fathead minnow exposures All methods for fathead minnow exposures to the antidepressant mixtures and sampling and euthanasia methods were approved under animal use protocols #0710 and #0810 reviewed by the Department of Fisheries and Oceans/Environment Canada Joint Animal Care Committee for the Canada Centre for Inland Waters (Burlington, ON, Canada), as operated under the approval of the Canadian Council of Animal Care. Fathead minnow eggs from a breeding colony maintained at the Canada Centre for Inland Waters were rolled from tiles, assessed for fertility and counted into mesh-bottom 500-mL egg cups. One cup (each containing 30 eggs) was placed in each aquarium. Aquaria were housed in a 25 °C water bath, aerated with air stones, and covered with Plexiglas lids. Effective tank volumes were approximately 12 L. Exposure to the antidepressant mixtures began with eggs that were b12 h postfertilization. The photoperiod was set at 16 h light:8 h dark. Dilution water for the antidepressant mixtures was municipal water for the City of Burlington, ON, Canada that was dechlorinated by charcoal filtration and UV sterilized. For the municipal water, the average hardness was 124 mg/L, alkalinity was 81 mg/L, and conductivity was 353 μS/ cm. Other water quality parameters in the dilution water are summarized in Supplemental Data (Table S1a). Water temperature, dissolved oxygen, pH, and conductivity were measured once a week in all fish exposure aquaria, and these data are summarized in Supplemental Data (Table S1b). Eggs were checked daily for mortality, and dead eggs were removed to discourage fungal growth. At 7 d post-hatch, larvae were counted and transferred from egg cups to the 12 L aquaria. Larval fish were fed newly hatched brine shrimp, with additional frozen and thawed brine shrimp added as fish grew to juveniles, and then switched completely to frozen brine shrimp as adults, as described by Parrott and Bennie (2009). Random culling of fish over time, using a random-number generator which gives all fish in the tank and equal chance of being culled, maintained appropriate densities for remaining fish to grow. At 31 d post-hatch (i.e. 31 dph), aquaria were randomly culled to 20 fish, and at 63 dph, the fish were randomly culled to 15. At 64 dph, three breeding tiles constructed from PVC plastic pipes cut in half were added to each tank to promote maturation of males, and breeding/spawning behaviours. At 77 dph, fish were further culled to 12 fish per tank, and at 92 dph fish, were culled to 10 fish per tank, with culling purposefully aimed at leaving at least 3 males and 5 females in each tank. A final culling was done at 106 dph to 8 fish, with purposeful culling to the most sexually mature fish, and with three mature males and five mature females selected for each tank. All culled fish were sacrificed by an overdose of clove-oil solution (10 drops/L), and weighed and measured.

1230

J.L. Parrott, C.D. Metcalfe / Science of the Total Environment 648 (2019) 1227–1236

Breeding trials started in tanks with sexually mature individuals at 77 dph to 99 dph. Any eggs deposited on tiles were removed every morning, rolled from tiles, counted and assessed for fertilization. Breeding was monitored until 125 dph. Up to 100 eyed eggs from each batch were placed in clean lab water in screen-bottomed egg cups in aquaria (aerated) and observed until hatch. For each antidepressant mixture treatment, we observed a total of 6400 eggs from the 1× AntiD Mix treatments and 5050 eggs from the 10× AntiD Mix treatments, which were distributed among the 4 replicates. For the control treatment, we observed 14,900 eggs, as there were 8 replicate control breeding aquaria. Egg-hatching aquaria were temperature controlled in a 25 °C water bath. Egg production and hatch were monitored daily, and data were collected on the day to first breeding event with a clutch size over 20 eggs, daily egg production, % fertilization, % dead eggs, % hatch, % deformed fry, and % dead fry. Percentages of deformed fry were also assessed by summing the number of all deformed fry in a tank and dividing by the total number of eggs that hatched in that tank. This value gave a second overall assessment of the deformities in each tank and treatment that was not biased by differences in the sizes of egg batches that were laid. Dead eggs were defined as eggs that were fertilized and had begun cell division, but then subsequently died. Deformed fry had hatched but had severe deformities of the spine, or edema of the heart or yolk sac. Between 126 dph and 162–163 dph, repeated behavioural trials were conducted to assess whether the exposures to AntiD Mix affected fish nest defense, using protocols previously described by Parrott and Metcalfe (2018). Briefly, a dummy intruder “fish-on-a-stick” (Supplemental Data, Fig. S1) was held near each male's nest for 1 min, and the time to first contact in seconds and the total number of contacts in 1 min were counted to assess the aggressiveness of the nest-defense behaviours of the males.

and hatching success) using the Shapiro-Wilk Normality Test. In about 84% of the cases, (i.e. 63% of the survival data, 85% of the growth data, 92% of the growth endpoint data, and 97% of the data on egg production and egg/fry health) the data were distributed normally, so we proceeded with analysis of variance (ANOVA) using untransformed data. Growth parameters of length (mm), weight (g), CF, LSI, GSI, ovipositor area of females (mm2), genital papillae area of males (mm2), male index, tubercle index, and hematocrit were assessed for each sex for differences among treatments using ANOVA. This statistical analysis approach was also used to assess F1 parameters (i.e. % hatch, % fry mortality, % larval deformities) for differences among treatments. Nest-defense behaviour parameters (% males contacting a dummy intruder, % males making fast contact (≤10 s), and mean time to first contact with the dummy intruder) were tested using ANOVA. Significant differences from control treatments were assessed using Two-Sample t-Tests (separate variances, Bonferroni's adjusted probabilities) to determine levels of significance. All p-values for these tests are shown in the text, Figures, and Tables. For the assessment of eggs and fry produced by the parent fish exposed to antidepressant mixtures, the data were summarized in two ways. Percentages of unfertilized eggs and dead eggs as well as % eggs hatched, fry mortality and % deformed fry were assessed in each batch of eggs, and averaged per tank. These data were also assessed by summing the incidence of these in all eggs laid in a given tank and dividing by the total number of eggs laid or eggs assessed for that tank. These values provided another way for the reproductive parameters to be analyzed per replicate tank and gave an overall assessment of the success of reproduction and hatch for that treatment.

2.4. Fish dissection

Water quality parameters were very stable during the flow-through exposures. Mean temperatures ranged from 24.1 to 24.4 °C, mean dissolved oxygen ranged from 8.32 to 8.65 mg/L, mean pH ranged from 7.79 to 7.83, and mean conductivity ranged from 368 to 370 μS/cm. None of the water quality values differed between the controls and either of the two treatments with antidepressant mixtures. Dilution water hardness was 124 ± 2 mg/L, and alkalinity (measured as CaCO3) was 81.4 ± 0.35. All water quality data are summarized in Supplemental Data, Table S1a and S1b. Fig. 1 illustrates the concentrations of the target compounds in a sample of MWWE collected from a WWTP on the Grand River in southern Ontario (Canada) and the concentrations of these compounds

At 162–163 dph, all of the remaining adult fish were sacrificed. Fish were anesthetized in clove oil solution (10 drops/L). Blood samples were taken from the dorsal aorta using a heparinized capillary tube. Fish were then killed by spinal severance, weighed, measured, and dissected as described previously by (Parrott and Metcalfe, 2018). Secondary sex characteristics were assessed and the phenotypic sex of the fish was determined from these characteristics. The measure of female secondary sex characteristics included ovipositor length and width measured under a dissecting microscope and the area of the approximately triangular ovipositor was calculated in mm2. For males, the area of the genital papillae was measured. Ovipositor area ranged from 0.53 to 3.4 mm2 for individual females, and genital papillae area ranged from 0.00 to 0.55 mm2 for individual males. Other male secondary sex characteristics that were assessed included the presence of a dorsal fin dot (graded as present/absent), and the appearance of the dorsal head pad, which was rated on a scale of 0 (no pad) to 4 (large, well developed pad and nuptial tubercles). Nuptial tubercles were graded as small or large and counted under a dissecting microscope. Male banding pattern was scored from 0 (no banding) to 3 (dark banding with black head). Tubercle index was calculated as the number of tubercles +2× the number of large tubercles, and for individual males in this experiment it ranged from 4 to 52. The “male index” was calculated as the sum of: 1 point for fin dot + band points (out of a possible 3) + pad points (out of a possible 4) + tubercle index/10. The male index of individual males in this experiment ranged from 4.7 to 11.8.

3. Results 3.1. Exposures

2.5. Statistical analyses Data were analyzed using Systat 11.0 (Systat Software Inc., San José, CA, USA). Data were checked for normality by sex and by treatment (control, 1× AntiD Mix, 10× AntiD Mix) for growth parameters, and by treatment for egg production and reproductive parameters (i.e. egg viability

Fig. 1. Concentrations (ng/L) of five antidepressants (Venlafaxine, Citalopram, Fluoxetine, Sertraline, and Bupropion) in municipal wastewater effluent (MWWE) and measured exposure concentrations for the five compounds in the 1× AntiD Mix and 10× AntiD Mix treatments that fathead minnow were exposed to for a lifecycle.

J.L. Parrott, C.D. Metcalfe / Science of the Total Environment 648 (2019) 1227–1236

1231

Table 1 Nominal and measured concentrations (ng/L) of the five antidepressants (Venlafaxine, Citalopram, Fluoxetine, Sertraline, and Bupropion) in the 1× AntiD Mix and 10× AntiD Mix antidepressant mixtures used in the fathead minnow flow-through lifecycle exposure. Samples were taken and each compound was measured 4 times in the 1× AntiD Mix, and 7–9 times in the 10× AntiD Mix. Mean and standard deviation are also shown. Concentrations of the five antidepressants in MWWE are also shown, as these were the target nominal concentrations for the 1× AntiD Mix. Bupropion was in most cases not detected in aquaria water, so a value of one-half of the Limit of Detection was assigned in order to calculate the mean measured concentration of this compound. All raw data are shown in Supplemental Data, Table S2. Venlafaxine

Citalopram

Fluoxetine

Sertraline

Bupropion

2579

315

145

22

120

Nominal concentrations (ng/L) 1× AntiD Mix 10× AntiD Mix

2386 23,864

240 2401

89 895

18 179

87 868

Mean measured concentrations (ng/L) 1× AntiD Mix Mean (ng/L) Standard deviation 10× AntiD Mix Mean (ng/L) Standard deviation

2302 1024 32,899 11,330

159 78 2866 1250

108 25 1005 333

7 3 207 135

4.5 0 102 193

Concentrations in MWWE (ng/L)

detected in replicate samples (n = 4 to 9) of the water to which fathead minnows were exposed. Target nominal concentrations for the five antidepressants were the sum of the antidepressant plus it major metabolites in the municipal wastewater effluent. Measured concentrations of the antidepressants in the 1× AntiD Mix exposure water were close to the measured concentrations in MWWE, except for lower concentrations of sertraline and concentrations below detection limits for bupropion (Fig. 1). Measured concentrations of venlafaxine, citalopram, and fluoxetine were 66–121% of the expected nominal concentrations in the 1× AntiD Mix, and 112–138% of nominal concentrations for the 10× AntiD Mix (Table 1). Sertraline was 37% of nominal concentration for the 1× AntiD Mix, and 116% of nominal concentrations for the 10× AntiD Mix. The elevated LODs for bupropion meant that this compound was detected only twice in the 10× AntiD Mix at 420 and 465 ng/L, respectively, which was about half of the nominal concentration. Note that none of the metabolites of the antidepressants (i.e. O- and N-desmethyl-venlafaxine, norfluoxetine, desmethyl-sertraline, desmethyl-citalopram) were detected in the AntiD mixes, which was expected, since these transformation products are produced as a result of human metabolism. All data are shown in Supplemental Data, Table S2. 3.2. Survival and growth Lifecycle exposures to the antidepressant mixtures did not affect survival of fish. There were no effects on egg mortality or fry hatching at the beginning of the lifecycle exposures. Fathead minnow eggs exposed to the 1× AntiD Mix or to the 10× AntiD Mix successfully hatched, and fry survival was similar to the control treatment, as illustrated in Fig. 2 (see Supplemental Data, Table S3). Overall, control fish survival was 81% from the egg stage until 163 dph, while fish exposed to 1× AntiD Mix had an overall survival of 72%, and fish exposed to 10× AntiD Mix had an overall survival of 74%. Slight decreases in survival in Fig. 2 were not significant compared to control data (p N 0.114) for all time points in both AntiD Mix treatments. There were few effects of the antidepressant mixtures on growth of the fish over time. Mean length (mm), and weight (g), of AntiD Mixexposed fish at 31, 63 and 77 dph did not differ from those of controls, as shown in Table 2 (see Supplemental Data Table S4 for individual replicate weight and length details). At 63 dph, fish exposed to the 10× AntiD Mix had significantly higher condition factors (CF) compared to control fish (p = 0.010), but increased CFs were not seen in 31 dph or 77 dph fish. At the end of the exposure period, when fish were sampled at 162–163 dph, the mean weight of adult male fish exposed to the 10× AntiD Mix (3.31 g) was greater than the mean weight of control fish (2.90 g), at a p = 0.046 level of significance. The mean weight of male fish exposed to the 1× AntiD Mix (3.05 g) was not significantly different than the mean weight of control males (2.90 g). Lengths, weights, and

CFs of female fish showed no significant differences among controls and AntiD Mix-exposed fish, as shown in Table 3 (see Supplemental Data Table S4 for all data by replicate tank). Liver-somatic indices and gonadosomatic indices of AntiD Mix-exposed fish were also similar to the controls. Mean liver-somatic indices for males ranged from 2.1 to 2.2, and for females 3.3 to 3.7, and mean gonadosomatic indices for male fish ranged from 1.0 to 1.4, and for females from 12.6 to 14.1 (Table 3). Mean hematocrits of AntiD Mix-exposed female fish were higher than controls, although this was significant only for fish from the treatments with the 1× AntiD Mix (p = 0.044), and not for fish from the treatments with the 10× AntiD Mix (p = 0.090). Male fish hematocrits did not differ significantly between control fish and exposed fish (Table 3). 3.3. Maturation and sex characteristics There were no effects of exposure to the antidepressant mixtures on maturation of the fish during the lifecycle test (Supplemental Data, Table S5). From their external characteristics, only a few fish were mature by about 63 dph, with 8% of fish identified as either males or females at this time. At 64 dph, breeding tiles were added to the exposure tanks, which promoted maturation, so that by 77 dph, 32% of the fish were mature. At 92 dph and after the cull to 10 fish per tank, 77% of fish were classified as mature males or females. By 106 dph, 97% of the fish were mature. Breeding began in the control treatments at 87 dph and in the antidepressant mixture treatments,

Fig. 2. Plot of mean survival of fathead minnow eggs, larvae, and adults in each antidepressant mixture concentration (1× AntiD Mix and 10× AntiD Mix) and in water controls. Eggs hatched on day 0. Error bars show ± standard deviation for water controls and 10× AntiD Mix-exposed fish. Survival was not significantly different from controls for either 1× or 10× AntiD Mix treatments (p ≥ 0.114).

1232

J.L. Parrott, C.D. Metcalfe / Science of the Total Environment 648 (2019) 1227–1236

Table 2 Growth measurements over time in culled fathead minnows from the AntiD Mix lifecycle exposure. Mean (±standard deviation, s.d.) weight (mg), length (mm), and condition factor of culled fish from AntiD Mix exposures over time. Fish were randomly sampled at 31, 63, and 77 days post-hatch (dph). Statistical tests were done on tank means, so n in table is the number of tanks (n tanks). The value in bold with an asterisk indicated that the mean was significantly different from the control means, p = 0.010 (from Two sample t-test, with separate variances for Bonferroni's adjusted probability). All data for each replicate are shown in Supplemental Data, Table S4. Day post-hatch

Treatment

n

Weight (mg)

s.d.

Length (mm)

s.d.

Condition factor

s.d.

31

Controls 1× AntiD Mix 10× AntiD Mix Controls 1× AntiD Mix 10× AntiD Mix Controls 1× AntiD Mix 10× AntiD Mix

8 4 4 8 4 4 8 4 4

29.4 15.9 23.3 319 397 438 559 519 585

14.7 13.9 9.5 105 127 142 145 85 204

14.4 11.5 13.7 31.1 33.5 34.5 37.2 35.4 36.8

2.5 3.0 1.7 3.8 3.3 3.5 3.1 2.6 4.1

0.800 0.776 0.746 0.930 0.984 1.026⁎ 1.032 1.091 1.055

0.052 0.059 0.060 0.072 0.062 0.032 0.052 0.046 0.050

63

77

breeding began at 94 and 93 dph for fish from the 1× AntiD Mix and 10× AntiD Mix treatments, respectively. These slight delays in mean time to first breeding were not significant (p N 0.103). Based on other fathead minnow lifecycle tests run previously in our laboratory (Parrott and Balakrishnan, 2017; Parrott and Bennie, 2009), the time-to-maturation and time-to-initiation of breeding were within the normal range for fish from the control and the exposed treatments. Secondary sex characteristics of male fathead minnows were enhanced by exposure to the antidepressant mixtures, as shown in Table 3 (see Supplemental data Table S6 for all data by replicate tank). Male fathead minnows exposed to the 1× AntiD Mix had a significantly greater Tubercle Index compared to control fish (s = 0.038). However, the Tubercle Index for fish from the treatments with the 10× mixture were not significantly different from the control males (p = 0.090). The Male Index was also significantly greater in male fish exposed to 1× AntiD Mix (p = 0.001) compared to controls. However, once again, the Male Index in fish exposed to the 10× AntiD Mix was not significantly different from control males (p = 0.161). Female fathead minnows showed no significant dose-related changes in ovipositor size between control fish and exposed fish. Mean ovipositor areas of female control fish and AntiD Mix-exposed female fish ranged from 1.4 to 1.6 mm2 (Table 3). There were no significant differences in the area of genital papillae for male fish exposed to antidepressant mixtures compared to control fish. The mean area of the genital papillae for males ranged from 0.09 to 0.16 mm2.

3.4. Nest-defense behaviour of males There were no significant differences in the nest-defense behaviour of adult males that had been exposed to antidepressant mixtures over a full lifecycle. Male fathead minnows exposed to antidepressant mixtures and in the control treatments protected their nests more vigorously if they had eggs, relative to males with empty nests. Male fathead minnows made contact with the dummy fish 65% of the time in trials with empty nests (Supplemental Data, Fig. S2 and Table S8), and 93% of the time in trials where the nest had eggs. The number of contacts per minute was the similar for controls and for males from the 1× and 10× AntiD Mix treatments, and averaged 7 contacts per minute for males with empty nests, and 13 contacts per minute for males with eggs in their nests. All nest-defense data are shown in Supplemental Data, Table S7, and data summaries for the behavioural trials are shown in Table S8. 3.5. Egg production and effects in offspring Fish from both antidepressant mixture treatments produced similar numbers of eggs in relation to the controls (Fig. 3). Breeding began on average at 87 dph for the controls and 93–94 dph for the exposed fish. Mean days to the first breeding event, in which there was a clutch size over 20 eggs, did not differ among the antidepressant mixture treatments and the controls, as shown in Table 4 (see Supplemental Data

Table 3 Growth and health measurements of adult fathead minnows exposed to water (controls, 0 μg/L) or to 1× and 10× AntiD Mix treatments for a lifecycle (from egg to 163 days post-hatch). Values are treatment means (±standard deviation, s.d.) for Length (mm), Weight (g), condition factor (CF), Tubercle Index, Male Index, Genital Papillae Area for males (mm2) or Ovipositor Area for females (mm2), Liver-Somatic Index (LSI), Gonadosomatic Index (GSI), and Hematocrit (Hemat %). Statistical tests were done on tank means, so n in table is the number of tanks. The column ‘n Hemat’ shows the number of tanks for the hematocrit data. Values in bold with asterisk indicate means that are significantly different from control means, p values b0.05 (determined using Two sample t-tests, separate variances, Bonferroni's adjusted probabilities). All data for each replicate are shown in Supplemental Data, Table S6. Males Treatment Water 1× AntiD Mix 10× AntiD Mix Treatment Water 1× AntiD Mix 10× AntiD Mix

n (tanks) 8 4 4

Length (mm) 58.6 59.0 60.3 2

s.d. 3.2 1.4 1.2

Weight (g) 2.90 3.05 3.31⁎

s.d. 0.35 0.37 0.25

CF 1.43 1.47 1.50

s.d. 0.09 0.09 0.08

Tubercle index 24.9 35.7⁎

Male index 8.1 9.7⁎

32.3

s.d. 6.0 6.7 6.0

9.0

n (tanks) 8 4 4

Ovi area (mm ) 0.11 0.09 0.16

s.d. 0.09 0.10 0.06

LSI 2.16 2.12 2.06

s.d. 0.38 0.23 0.23

GSI 1.05 1.40 1.15

s.d. 0.15 0.47 0.14

n Hemat 7 4 4

Hemat (%) 40.8 42.7 41.7

s.d. 4.8 1.4 1.2

Treatment Water 1× AntiD Mix 10× AntiD Mix

n (tanks) 8 4 4

Length (mm) 49.1 49.3 50.0

s.d. 1.1 2.1 1.2

Weight (g) 1.32 1.34 1.44

s.d. 0.11 0.21 0.15

CF 1.11 1.11 1.14

s.d. 0.03 0.04 0.04

Tubercle index .

s.d. .

Male index .

Treatment Water 1× AntiD Mix 10× AntiD Mix

n (tanks) 8 4 4

Ovi Area (mm2) 1.51 1.56 1.42

s.d. 0.28 0.34 0.45

LSI 3.25 3.66 3.70

s.d. 0.56 0.45 0.64

GSI 13.2 14.1 12.6

s.d. 1.9 1.9 2.2

n Hemat 7 4 3

Hemat (%) 32.3 36.0⁎

s.d. 3.5 1.9 2.9

s.d. 0.7 0.4 0.9

Females

36.9

s.d. .

J.L. Parrott, C.D. Metcalfe / Science of the Total Environment 648 (2019) 1227–1236

Fig. 3. Cumulative total egg production per female for fathead minnows exposed to antidepressant mixtures (1× AntiD Mix and 10× AntiD Mix) or lab water (controls) for a lifecycle. Egg production was not significantly different from controls for either 1× or 10× AntiD Mix treatments (p ≥ 0.109).

Table S9 for all breeding data per replicate tank). Although there was a trend of lower mean total egg production per tank and lower mean eggs per female in the 10× AntiD Mix treatment (Supplemental Data, Fig. S3), these differences were not significantly different from fish in the control (i.e. p = 0.109, and p = 0.191, respectively for total eggs and eggs per female for Bonferroni's adjusted p using separate variances). Total egg production per 4 replicates was 19,400 for controls and 16,800 and 12,200 for fish exposed to the 1× AntiD Mix and 10× AntiD Mix, respectively. There were no statistically significant differences in mean egg production per female across all treatments, with 1196 for control fish, 1116 for fish exposed to the 1× AntiD Mix, and 863 for fish exposed to the 10× AntiD Mix. The mean number of clutches per tank was significantly lower (p = 0.014) in fish exposed to the 1× AntiD Mix, but was not different than controls in fish exposed to the 10× AntiD Mix (p = 0.063). As shown in Table 4, the fish from the 1× AntiD Mix treatment had on average 18.8 clutches and the fish from the 10× AntiD Mix treatment had 19.0 clutches, compared to controls with 24.8 clutches per replicate tank (see individual tank replicate data are shown in Supplemental Data, Table S6). The mean size of each clutch was not altered significantly by exposure to the antidepressant mixtures, with control fish producing 190 ± 64 (mean ± standard deviation) eggs per clutch, and fish from 1× and 10× AntiD Mix exposures producing 222 ± 32 and 162 ± 71 eggs per clutch, respectively (Table 4).

1233

There were no differences in the survival of embryos or fry produced by parent fish exposed to antidepressant mixture treatments when data were compared to controls (Table 4). The % fertilization, % hatch, and % fry mortality all showed no differences across treatments (see Supplemental Data, Table S6 for details by replicate tank of all parameters assessed in eggs and hatched fry). For fish from the controls and treatments with both antidepressant mixtures there were no differences in fertilization rates. Assessing all of the eggs laid in each tank across all treatments, the mean percentage of unfertilized eggs was about 2.4%, while the percentage of dead eggs was about 8% (Table 4). All fertilized and healthy eggs (or a subset of 100 eggs) from the treatments were placed in egg cups to assess hatching success and time to hatch, for a total monitoring group of nearly 14,900 eggs from the 8 replicate control aquaria, and 6400 and 5050 eggs from the 1× AntiD Mix and 10× AntiD Mix treatments, respectively (having 4 replicate aquaria in each). For eggs produced by parental fish exposed to the 10× AntiD Mix for a lifecycle, the time to hatch (TTH) of F1 eggs was significantly earlier (i.e. 5.0 ± 0.2 d, p = 0.011) compared to TTH for eggs from control fish (i.e. 5.5 ± 0.2 d), as shown in Table 4. In these egg monitoring trials, there were no significant differences in hatchability (i.e. % embryos that hatched), hatching success (i.e. % of embryos that hatched into viable fry), or fry mortalities across the treatments (Table 4). The percentage of deformed fry was significantly lower in eggs produced by parental fish that had been exposed to the antidepressant mixes. Among eggs produced by fish exposed to the 1× AntiD Mix, there were about 4% deformed offspring, compared to 3.3% deformed offspring from fish exposed to the 10× AntiD Mix, and about 7–8% deformed offspring in control fish, as illustrated in Fig. 4 (see Supplemental Data Table S9 for data from all replicate tanks). The values of deformities were calculated using two methods for each tank, to ensure that biases from clutch size did not affect results. Both ways of analyzing the percent deformities data yielded similar results (Fig. 4). The differences in % deformities were significantly different from controls only for the 10× AntiD Mix treatment, using data on the % deformities overall (p = 0.008) for % deformities overall, as well as using data on the total # deformities / total # eggs hatched per tank (p = 0.015). There were transient decreases in growth of F1 larvae produced by parent fish exposed to the 10× Anti D Mix. As shown in Table 6 (see Supplemental Data Fig. S4), at 8 dph, F1 larvae produced by parents exposed to 10× AntiD Mix were significantly shorter at a mean length of 5.6 mm than control fish at a mean length of 6.0 mm (p = 0.003; Bonferroni's adjusted t-test with separate variances). However, the shorter lengths of larval fish observed at 8 dph did not persist, since at 16 dph the lengths of larval fish from parents exposed to 10× AntiD Mix were not statistically different than controls (p = 0.057), as

Table 4 Summary of egg and fry quality parameters from fathead minnow exposed to AntiD Mix for a lifecycle. Values are means (with standard deviation, s.d.) for each treatment. Breeding began at about 86 dph, and was followed until 125 dph. Fish were sampled at 162–163 dph. Parameters include Total Eggs, Eggs per Female, # Females, Date Breeding Started (dph), Clutch Size (# eggs per clutch), # Clutches, Unfertilized Eggs (%), Dead Eggs (%), Time to Hatch (d), Hatchability (%), Hatch Success (%), Fry Mortality (%), Deformed Fry (mean % per tank) and # Fry deformed of # hatched per tank. Statistical tests were done on tank means, so n in table is the number of tanks. Values in bold with asterisk indicate means that are significantly different from control means, p values b0.05 (determined using Two sample t-tests, separate variances, Bonferroni's adjusted probabilities). All data for each replicate are shown in Supplemental Data, Table S9. Treatment

n

Total # eggs per tank

s.d.

Eggs per female

s.d.

# females

s.d.

Breeding started (dph)

s.d.

Clutch size (# eggs per clutch)

s.d.

Water 1× AntiD Mix 10× AntiD Mix

8 4 4

4850 4211 3040

2233 1101 1304

1196 1116 863

532 283 290

4.0 3.8 3.5

0.4 0.5 0.6

86.5 93.8 92.5

6.6 6.1 6.6

190 222 162

64 32 71

Treatment Water 1× AntiD Mix 10× AntiD Mix

8 4 4

Treatment Water 1× AntiD Mix 10× AntiD Mix

8 4 4

# Clutches per tank

s.d.

Unfertilized eggs (%)

s.d.

Dead Eggs (%)

s.d.

Time to hatch (d)

s.d.

Hatchability (%)

s.d.

24.8 18.8⁎ 19.0

4.4 2.5 4.1

2.7 2.2 2.3

1.5 0.6 0.7

8.5 7.6 8.1

3.2 1.7 1.6

5.5 5.4 5.0⁎

0.2 0.1 0.2

82.5 83.8 84.2

4.2 3.0 4.5

Hatch success (%)

s.d.

Fry mortality (%)

s.d.

Deformed fry (mean % per tank)

s.d.

# fry deformed of # hatched per tank

s.d.

75.8 79.6 79.8

5.7 4.0 3.9

1.7 1.2 1.9

1.2 0.7 1.1

7.7 4.0 3.4⁎

3.3 2.6 0.9

7.0 3.9 3.2⁎

3.3 2.8 1.0

1234

J.L. Parrott, C.D. Metcalfe / Science of the Total Environment 648 (2019) 1227–1236

Fig. 4. Percentages of deformed fry in F1 offspring from parent fathead minnows exposed to water (controls, 0) or to antidepressant mixtures (1× AntiD Mix and 10× AntiD Mix) for a lifecycle. Percentages of deformed fry were calculated two ways: As the average percentage of deformed fry for every batch of eggs assessed in a given tank (purple bars), and as the total number of deformed fry produced by a tank divided by the total number of fry produced by that tank (green bars). Asterisks indicate that F1 fry from parents exposed to the 10× AntiD Mix treatment had significantly fewer deformities compared to controls (p = 0.008 for Deformed fry % and p = 0.015 for # Fry deformed of hatched). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)

shown in Table 6 (see all data for replicate aquaria shown in Supplemental Data Table S7). 4. Discussion There were no significant changes in survival of fish exposed over a full life cycle to the antidepressant mixture at environmentally-relevant concentrations (i.e. 1× AntiD Mix) or at higher concentrations (i.e. 10× AntiD Mix). Other researchers have seen no impact of exposure to mixed antidepressants on the survival of fish, although exposures were much shorter in these other studies. Schultz et al. (2011) reported that there were no impacts on survival of adult male fathead minnow exposed for 21 d to a mix of venlafaxine, fluoxetine, sertraline and bupropion, although exposure to venlafaxine and sertraline individually did decrease male survival. Maturing fathead minnows exposed to 10× AntiD mixture had an increased condition factor at 63 dph, and adult male fish were heavier than control males at the end of the experiment. There are few other published reports of increased growth in fish exposed to antidepressants, as more commonly, decreased growth has been observed in fish exposed to these drugs. Wild European perch (Perca fluviatilis) exposed to the anti-anxiety drug oxazepam (1.8 and 910 μg/L for 7 d) did have increased feeding rates (Brodin et al., 2013). However, larval guppies (Poecilia reticulate) exposed to 0.03 μg/L fluoxetine (nominal concentration) for 35 d were smaller (i.e. decreased weight and lengths) compared to control fish (Pelli and Connaughton, 2015). Decreased feeding rate and decreased growth were observed in larval fathead minnow exposed to 51 μg/L fluoxetine for 7 d (Stanley et al., 2007). Exposure to the antidepressant mixtures did not affect the maturation of the fathead minnows over time. Juvenile fish exposed to both antidepressant mixtures developed sex characteristics at the same rate as control fish. Exposure to 1× AntiD mixture enhanced the secondary sex characteristics (i.e. tubercle index and male index) of adult male minnows sampled at the end of the lifecycle test (Table 4). However, these are thought to be chance findings, as male secondary sex characteristics were unaffected by exposure to the 10× AntiD mixture. Exposure to the AntiD mixtures did not affect female secondary sex characteristics (ovipositor area). Other researchers have reported that exposure to some mixtures of antidepressants did not affect male sex

characteristics in fathead minnow, although exposures of adult fish occurred only over 21 d (Schultz et al., 2011). Lifecycle exposure to antidepressant mixtures caused no significant changes in egg production compared to control fish. Mean egg production is highly variable in fathead minnow breeding groups, and so large and consistent differences are needed to show statistical significance. Some studies with fish exposed to antidepressants have shown decreased reproduction, although in these studies the exposure was at the adult stage only. For instance, cumulative egg production by zebrafish was decreased by exposure to a mixture of four pharmaceuticals (i.e. acetaminophen, carbamazepine, gemfibrozil and venlafaxine) all at nominal concentrations of 0.5 and 10 μg/L (Galus et al., 2013b). Other antidepressants such as fluoxetine have also been shown to decrease reproduction in fishes exposed as adults. Zebrafish exposed to 32 μg/L fluoxetine for 7 d produced significantly fewer eggs than controls, and had decreased production of estradiol in ovaries of females (Lister et al., 2009). Other studies have shown no effects of antidepressants on reproduction. Foran et al. (2004) found no effects of fluoxetine on reproductive success among adult Japanese medaka exposed for 28 d (Foran et al., 2004). The differences in effects on reproduction could be due to the nature of the exposure. In lifecycle exposures where exposure is from the egg stage to the reproductive adult, it is possible that fish may acclimate or adjust to the presence of low concentrations of antidepressants (Sehonova et al., 2018). In experiments with shorter exposures, such as those reported by Galus et al. (2013b) and Lister et al. (2009), where adult reproductively-active fish were exposed to antidepressants for 1–3 weeks, the exposures may decrease reproduction as the fish have not adjusted to the presence of antidepressants. There is a current controversy over the repeatability of some of the published research on the apparent effects of antidepressants in fish. While scientists agree that high concentrations of antidepressants can affect endpoints such as fish behaviour, there is considerable debate whether very low concentrations can cause similar effects or whether the differences seen in some fish studies with antidepressants may be chance findings or artifacts (Harris et al., 2014; Margiotta-Casaluci et al., 2014; Sumpter and Margiotta-Casaluci, 2014). Several authors recommend that if results are unusual there should be some attempt to repeat the study to verify the findings (Hanson et al., 2017; Harris and Sumpter, 2015; Sumpter et al., 2014). Fish behavioural studies are subject to high variability, and can be affected by species, sex (Piyapong et al., 2010), and individual organism traits (dominant vs subordinate fish) (Dzieweczynski et al., 2016; Margiotta-Casaluci et al., 2014). Experimental factors such as acclimation and observation time (Melvin and Wilson, 2013) can affect the results, as can fish learning or habituation if behavioural trials are carried out over time (Margiotta-Casaluci et al., 2014). In the current experiments, the antidepressant mixture exposures caused slight changes percentages of egg hatching and fry deformities. Eggs of fish exposed to 10× AntiD Mix hatched slightly earlier than eggs from control fish, and fry produced by fish exposed to the 10× AntiD Mix had fewer deformities than control fry. However, these changes were very small, and we do not consider them to be biologically-relevant. Fathead minnow lay thousands of eggs per breeding season, and success and size of young-of-year fish is inversely proportional to the numbers of minnows present (Vandenbos et al., 2006). Therefore, we do not expect that a 0.5 d shortening of hatch time or a 4% decrease in fry abnormalities would affect overall recruitment success of fathead minnows in the wild. The decrease in hatch time may have been related to the decreased length of fish at 8 dph, as discussed below. Other lab experiments have shown that exposure to some antidepressant mixtures can affect fish embryonic development and hatching times. In an experiment with zebrafish, there were 4–5-fold increases in embryo mortalities and abnormalities when parent fish were exposed for 6 weeks to a mixture of acetaminophen, carbamazepine, gemfibrozil

J.L. Parrott, C.D. Metcalfe / Science of the Total Environment 648 (2019) 1227–1236

and venlafaxine (Galus et al., 2013a). Yang et al. (2014) reported trends of earlier hatching when zebrafish embryos were exposed to the antidepressant amitriptyline at 100 ng/L (i.e. a concentration about 5 × surface water concentrations). Similar early-hatching was observed among the eggs of common carp that had been exposed to amitriptyline, nortriptyline, and clomipramine, and their mixtures, all at 10 μg/L (Sehonova et al., 2017). In the current study, it is not immediately obvious what the mechanism is for the early hatching of F1 eggs from parent fish exposed to the 10× AntiD Mix. There were no differences in the total egg production in the fish exposed to antidepressant mixtures. Previously we observed that lifecycle exposure of fathead minnows to venlafaxine (88 μg/L nominal, 75 μg/L measured) significantly increased the number of eggs produced per female (Parrott and Metcalfe, 2017). We attributed this response to similar responses in fish to the anti-anxiety effects of this drug in mammalian tests species (Borsini et al., 2002; Hascoët et al., 2000; Lapmanee et al., 2012; Rupniak et al., 2000), and we hypothesized that the venlafaxine-exposed fish may have been less anxious and so may have spent more time breeding. In the present experiment, the venlafaxine concentration in the 10× AntiD mix was 33 μg/L, and we did not see the increased egg production that we previously observed with fish exposed to 75 μg/L venlafaxine alone. It is possible that the four other antidepressants in the mixture attenuated the effects of venlafaxine in egg production in fathead minnows. There we no changes in nest-defense of adult male minnows exposed to antidepressant mixtures. All male fish protected nests with eggs more vigorously than they protected empty nests, and this was similar for control males and for 1× AntiD Mix- and 10× AntiD Mixexposed males. Previously, we found that exposure to high concentrations of venlafaxine alone (75 μg/L measured concentration) caused increased nest-defense behaviour in fathead minnow males, with male fish protecting their empty nests as vigorously as males protecting nests with eggs (Parrott and Metcalfe, 2018). In those studies, the highest exposure concentration of venlafaxine caused a significant increase in the number of egg laid per female as well as an increase in the nest-defense behaviours of empty-nest males (Parrott and Metcalfe, 2017, 2018). We hypothesized that the two were linked, with perhaps one causing the other: Female fathead minnows prefer to lay their eggs in nests of males with good nest-guarding abilities, or males with increased numbers of eggs in their nests develop more nest-protective behaviours (that occur even when their nests are empty). Exposure to certain antidepressants can increase boldness in some species of fish. Fluoxetine exposure (100 μg/L for 10 d) increased several tank-exploratory behaviours of Japanese medaka and decreased the time-to-first-contact and frequency of mirror biting in social interaction tests (Ansai et al., 2016). Fluoxetine (76 μg/L for 28 d) exposure also increased several exploratory behaviours in fathead minnows but only at high exposure concentrations (72 μg/L) that raised fish plasma fluoxetine concentrations above the human therapeutic plasma concentration for this drug (Margiotta-Casaluci et al., 2014). Male fathead minnows exposed for 4 weeks to 3–30 μg/L sertraline spent less time in the shelter (during the light period) compared to control males (Valenti Jr et al., 2012). Exposures of parent fish to the 10× AntiD Mix caused a transient decreases in F1 larval growth. The decreased larval lengths could be related to the early hatching of the F1 from 10× AntiD Mix exposed parents. Early hatching of embryos can result in smaller larval size. For example, Yang et al. (2014) reported a trend of earlier hatching and smaller fry when zebrafish embryos were exposed to the antidepressant amitriptyline. Amitriptyline, nortriptyline, and clomipramine (and their mixtures, all at 10 μg/L) caused early hatching and decreased length in the larvae of common carp, as reported by (Sehonova et al., 2017). Lake whitefish (Coregonus clupeaformis) exposed to morpholine (a heterocyclic amine used to prevent bio-fouling in industrial cooling waters) hatched 30% faster, and were 10% smaller in body length compared to

1235

controls (Thome et al., 2017). Fathead minnow embryos exposed to copper hatched faster, and larval fish were shorter in length at 28 dph (Scudder et al., 1988). Overall, our lifecycle exposure of fathead minnows to environmentally-relevant concentrations of a five-antidepressantmixture caused no significant effects on survival, growth, maturation, reproduction, or F1 success. This study adds to the growing body of literature suggesting environmentally-relevant concentrations of antidepressants cause few adverse effects in fish (Margiotta-Casaluci et al., 2014; McCallum et al., 2017a; Parrott and Metcalfe, 2017). 5. Conclusions To our knowledge, this is the first report of the effects of lifecycle exposure of fish to an antidepressant drug mixture. Fathead minnows exposed to environmentally relevant concentrations of an antidepressant mixture and to 10× these concentrations showed no significant changes in survival, growth, maturation, reproduction, and F1 hatching success. There were some slight increases in adult male fish weight from the 10× AntiD mixture. F1 assessment showed that offspring from parents exposed to 10× AntiD Mix fish had a slightly shorter time to hatch and decreased deformities in hatched F1 fry, and transient decreases in length of F1 larvae at 8 dph, that recovered by 16 dph. Overall, the study results show that exposures to an antidepressant mixture at environmentally relevant concentrations did not affect survival or reproductive capacity of this model fish test species. Acknowledgements The study could not have been competed without the dedication and excellence of the fish and chemistry lab teams: Bev Blunt, Christine Lavalle, and Maria Ramil. Funding We are grateful for generous funding from the Chemicals Management Plan of Health Canada. The authors declare that they have no conflicts of interest related to the reporting of these results. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2018.08.237. References Ansai, S., Hosokawa, H., Maegawa, S., Kinoshita, M., 2016. Chronic fluoxetine treatment induces anxiolytic responses and altered social behaviors in medaka, Oryzias latipes. Behav. Brain Res. 303, 126–136. Asnis, G.M., Kohn, S.R., Henderson, M., Brown, N.L., 2004. SSRIs versus non-SSRIs in posttraumatic stress disorder: an update with recommendations. Drugs 64 (4), 383–404. Bisesi, J.H., Bridges, W., Klaine, S.J., 2014. Effects of the antidepressant venlafaxine on fish brain serotonin and predation behavior. Aquat. Toxicol. 148, 130–138. Borsini, F., Podhorna, J., Marazziti, D., 2002. Do animal models of anxiety predict anxiolytic-like effects of antidepressants? Psychopharmacology 163 (2), 121–141. Brodin, T., Fick, J., Jonsson, M., Klaminder, J., 2013. Dilute concentrations of a psychiatric drug alter behavior of fish from natural populations. Science 339 (6121), 814–815. Brooks, B.W., Chambliss, C.K., Stanley, J.K., Ramirez, A., Banks, K.E., Johnson, R.D., Lewis, R.J., 2005. Determination of select antidepressants in fish from an effluentdominated stream. Environ. Toxicol. Chem. 24 (2), 464–469. Caccia, S., 1998. Metabolism of the newer antidepressants: an overview of the pharmacological and pharmacokinetic implications. Clin. Pharmacokinet. 34 (4), 281–302. Chiffre, A., Clérandeau, C., Dwoinikoff, C., Le Bihanic, F., Budzinski, H., Geret, F., Cachot, J., 2016. Psychotropic drugs in mixture alter swimming behaviour of Japanese medaka (Oryzias latipes) larvae above environmental concentrations. Environ. Sci. Pollut. Res. 23 (6), 4964–4977. Chu, S., Metcalfe, C.D., 2007. Analysis of paroxetine, fluoxetine and norfluoxetine in fish tissues using pressurized liquid extraction, mixed mode solid phase extraction cleanup and liquid chromatography-tandem mass spectrometry. J. Chromatogr. A 1163 (1–2), 112–118. Collado, N., Rodriguez-Mozaz, S., Gros, M., Rubirola, A., Barceló, D., Comas, J., RodriguezRoda, I., Buttiglieri, G., 2014. Pharmaceuticals occurrence in a wwtp with significant

1236

J.L. Parrott, C.D. Metcalfe / Science of the Total Environment 648 (2019) 1227–1236

industrial contribution and its input into the river system. Environ. Pollut. 185, 202–212. Dzieweczynski, T.L., Kane, J.L., Campbell, B.A., Lavin, L.E., 2016. Fluoxetine exposure impacts boldness in female Siamese fighting fish, Betta splendens. Ecotoxicology (London, England) 25 (1), 69–79. Foran, C.M., Weston, J., Slattery, M., Brooks, B.W., Huggett, D.B., 2004. Reproductive assessment of Japanese medaka (Oryzias latipes) following a four-week fluoxetine (SSRI) exposure. Arch. Environ. Contam. Toxicol. 46 (4), 511–517. Gago-Ferrero, P., Gros, M., Ahrens, L., Wiberg, K., 2017. Impact of on-site, small and large scale wastewater treatment facilities on levels and fate of pharmaceuticals, personal care products, artificial sweeteners, pesticides, and perfluoroalkyl substances in recipient waters. Sci. Total Environ. 601-602, 1289–1297. Galus, M., Jeyaranjaan, J., Smith, E., Li, H., Metcalfe, C., Wilson, J.Y., 2013a. Chronic effects of exposure to a pharmaceutical mixture and municipal wastewater in zebrafish. Aquat. Toxicol. 132-133, 212–222. Galus, M., Kirischian, N., Higgins, S., Purdy, J., Chow, J., Rangaranjan, S., Li, H., Metcalfe, C., Wilson, J.Y., 2013b. Chronic, low concentration exposure to pharmaceuticals impacts multiple organ systems in zebrafish. Aquat. Toxicol. 132–133, 200–211. Gros, M., Petrović, M., Barceló, D., 2007. Wastewater treatment plants as a pathway for aquatic contamination by pharmaceuticals in the Ebro river basin (Northeast Spain). Environ. Toxicol. Chem. 26 (8), 1553–1562. Gros, M., Petrović, M., Ginebreda, A., Barceló, D., 2010. Removal of pharmaceuticals during wastewater treatment and environmental risk assessment using hazard indexes. Environ. Int. 36 (1), 15–26. Hanson, M.L., Wolff, B.A., Green, J.W., Kivi, M., Panter, G.H., Warne, M.S.J., Ågerstrand, M., Sumpter, J.P., 2017. How we can make ecotoxicology more valuable to environmental protection. Sci. Total Environ. 578, 228–235. Harris, C.A., Sumpter, J.P., 2015. Could the quality of published ecotoxicological research be better? Environ. Sci. Technol. 49 (16), 9495–9496. Harris, C.A., Scott, A.P., Johnson, A.C., Panter, G.H., Sheahan, D., Roberts, M., Sumpter, J.P., 2014. Principles of sound ecotoxicology. Environ. Sci. Technol. 48 (6), 3100–3111. Hascoët, M., Bourin, M., Colombel, M.C., Fiocco, A.J., Baker, G.B., 2000. Anxiolytic-like effects of antidepressants after acute administration in a four-plate test in mice. Pharmacol. Biochem. Behav. 65 (2), 339–344. Kasprzyk-Hordern, B., Baker, D.R., 2012. Enantiomeric profiling of chiral drugs in wastewater and receiving waters. Environ. Sci. Technol. 46 (3), 1681–1691. Lajeunesse, A., Gagnon, C., Sauvé, S., 2008. Determination of basic antidepressants and their N-desmethyl metabolites in raw sewage and wastewater using solid-phase extraction and liquid chromatography-tandem mass spectrometry. Anal. Chem. 80 (14), 5325–5333. Lapmanee, S., Charoenphandhu, N., Krishnamra, N., Charoenphandhu, J., 2012. Anxiolyticlike actions of reboxetine, venlafaxine and endurance swimming in stressed male rats. Behav. Brain Res. 231 (1), 20–28. Lister, A., Regan, C., Van Zwol, J., Van Der Kraak, G., 2009. Inhibition of egg production in zebrafish by fluoxetine and municipal effluents: a mechanistic evaluation. Aquat. Toxicol. 95 (4), 320–329. López-Serna, R., Petrović, M., Barceló, D., 2012. Direct analysis of pharmaceuticals, their metabolites and transformation products in environmental waters using on-line turboflow™ chromatography-liquid chromatography-tandem mass spectrometry. J. Chromatogr. A 1252, 115–129. Margiotta-Casaluci, L., Owen, S.F., Cumming, R.I., De Polo, A., Winter, M.J., Panter, G.H., Rand-Weaver, M., Sumpter, J.P., 2014. Quantitative cross-species extrapolation between humans and fish: the case of the anti-depressant fluoxetine. PLoS One 9 (10). McCallum, E.S., Bose, A.P.H., Warriner, T.R., Balshine, S., 2017a. An evaluation of behavioural endpoints: the pharmaceutical pollutant fluoxetine decreases aggression across multiple contexts in round goby (Neogobius melanostomus). Chemosphere 175, 401–410. McCallum, E.S., Krutzelmann, E., Brodin, T., Fick, J., Sundelin, A., Balshine, S., 2017b. Exposure to wastewater effluent affects fish behaviour and tissue-specific uptake of pharmaceuticals. Sci. Total Environ. 605-606, 578–588. Melvin, S.D., 2017. Effect of antidepressants on circadian rhythms in fish: insights and implications regarding the design of behavioural toxicity tests. Aquat. Toxicol. 182, 20–30. Melvin, S.D., Wilson, S.P., 2013. The utility of behavioral studies for aquatic toxicology testing: a meta-analysis. Chemosphere 93 (10), 2217–2223. Metcalfe, C.D., Chu, S., Judt, C., Li, H., Oakes, K.D., Servos, M.R., Andrews, D.M., 2010. Antidepressants and their metabolites in municipal wastewater, and downstream exposure in an urban watershed. Environ. Toxicol. Chem. 29 (1), 79–89. Painter, M.M., Buerkley, M.A., Julius, M.L., Vajda, A.M., Norris, D.O., Barber, L.B., Furlong, E.T., Schultz, M.M., Schoenfuss, H.L., 2009. Antidepressants at environmentally relevant concentrations affect predator avoidance behavior of larval fathead minnows (Pimephales promelas). Environ. Toxicol. Chem. 28 (12), 2677–2684. Parrott, J.L., Balakrishnan, V.K., 2017. Life-cycle exposure of fathead minnows to environmentally relevant concentrations of the β-blocker drug propranolol. Environ. Toxicol. Chem. 36, 1644–1651. Parrott, J.L., Bennie, D.T., 2009. Life-cycle exposure of fathead minnows to a mixture of six common pharmaceuticals and triclosan. J. Toxicol. Environ. Health, Part A Cur. Issues 72 (10), 633–641.

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 (2), 131–141. Parrott, J.L., Metcalfe, C.D., 2017. Assessing the effects of the antidepressant venlafaxine to fathead minnows exposed to environmentally relevant concentrations over a full life cycle. Environ. Pollut. 229, 403–411. Parrott, J.L., Metcalfe, C.D., 2018. Nest-defense behaviors in fathead minnows after lifecycle exposure to the antidepressant venlafaxine. Environ. Pollut. 234, 223–230. Pelli, M., Connaughton, V.P., 2015. Chronic exposure to environmentally-relevant concentrations of fluoxetine (Prozac) decreases survival, increases abnormal behaviors, and delays predator escape responses in guppies. Chemosphere 139, 202–209. Piyapong, C., Krause, J., Chapman, B.B., Ramnarine, I.W., Louca, V., Croft, D.P., 2010. Sex matters: a social context to boldness in guppies (Poecilia reticulata). Behav. Ecol. 21 (1), 3–8. Preskorn, S.H., Richelson, E., Feighner, J.P., Cunningham, L.A., Roose, S.P., Keller, M.B., 1994. Antidepressant drug selection: criteria and options. J. Clin. Psychiatry 55 (9 SUPPL. A), 6–24. Rupniak, N.M.J., Carlson, E.C., Harrison, T., Oates, B., Seward, E., Owen, S., De Felipe, C., Hunt, S., Wheeldon, A., 2000. Pharmacological blockade or genetic deletion of substance P (NK1) receptors attenuates neonatal vocalisation in Guinea-pigs and mice. Neuropharmacology 39 (8), 1413–1421. Schultz, M.M., Furlong, E.T., 2008. Trace analysis of antidepressant pharmaceuticals and their select degradates in aquatic matrixes by LC/ESI/MS/MS. Anal. Chem. 80 (5), 1756–1762. Schultz, M.M., Furlong, E.T., Kolpin, D.W., Werner, S.L., Schoenfuss, H.L., Barber, L.B., Blazer, V.S., Norris, D.O., Vajda, A.M., 2010. Antidepressant pharmaceuticals in two U.S. effluent-impacted streams: occurrence and fate in water and sediment and selective uptake in fish neural tissue. Environ. Sci. Technol. 44 (6), 1918–1925. Schultz, M.M., Painter, M.M., Bartell, S.E., Logue, A., Furlong, E.T., Werner, S.L., Schoenfuss, H.L., 2011. Selective uptake and biological consequences of environmentally relevant antidepressant pharmaceutical exposures on male fathead minnows. Aquat. Toxicol. 104 (1–2), 38–47. Scudder, B.C., Carter, J.L., Leland, H.V., 1988. Effects of copper on development of the fathead minnow, Pimephales promelas rafinesque. Aquat. Toxicol. 12 (2), 107–124. Sehonova, P., Plhalova, L., Blahova, J., Doubkova, V., Marsalek, P., Prokes, M., Tichy, F., Skladana, M., Fiorino, E., Mikula, P., Vecerek, V., Faggio, C., Svobodova, Z., 2017. Effects of selected tricyclic antidepressants on early-life stages of common carp (Cyprinus carpio). Chemosphere 185 (Supplement C), 1072–1080. Sehonova, P., Svobodova, Z., Dolezelova, P., Vosmerova, P., Faggio, C., 2018. Effects of waterborne antidepressants on non-target animals living in the aquatic environment: a review. Sci. Total Environ. 631–632, 789–794. Silva, L.J.G., Pereira, A.M.P.T., Meisel, L.M., Lino, C.M., Pena, A., 2015. Reviewing the serotonin reuptake inhibitors (SSRIs) footprint in the aquatic biota: uptake, bioaccumulation and ecotoxicology. Environ. Pollut. 197, 127–143. Stanley, J.K., Ramirez, A.J., Chambliss, C.K., Brooks, B.W., 2007. Enantiospecific sublethal effects of the antidepressant fluoxetine to a model aquatic vertebrate and invertebrate. Chemosphere 69 (1), 9–16. Sumpter, J.P., Margiotta-Casaluci, L., 2014. Are some invertebrates exquisitely sensitive to the human pharmaceutical fluoxetine? Aquat. Toxicol. 146, 259–260. Sumpter, J.P., Donnachie, R.L., Johnson, A.C., 2014. The apparently very variable potency of the anti-depressant fluoxetine. Aquat. Toxicol. 151, 57–60. Tanoue, R., Nomiyama, K., Nakamura, H., Hayashi, T., Kim, J.W., Isobe, T., Shinohara, R., Tanabe, S., 2014. Simultaneous determination of polar pharmaceuticals and personal care products in biological organs and tissues. J. Chromatogr. A 1355, 193–205. Tanoue, R., Nomiyama, K., Nakamura, H., Kim, J.W., Isobe, T., Shinohara, R., Kunisue, T., Tanabe, S., 2015. Uptake and tissue distribution of pharmaceuticals and personal care products in wild fish from treated-wastewater-impacted streams. Environ. Sci. Technol. 49 (19), 11649–11658. Thome, C., Mitz, C., Sreetharan, S., Mitz, C., Somers, C.M., Manzon, R.G., Boreham, D.R., Wilson, J.Y., 2017. Developmental effects of the industrial cooling water additives morpholine and sodium hypochlorite on lake whitefish (Coregonus clupeaformis). Environ. Toxicol. Chem. 36 (7), 1955–1965. Valenti Jr., T.W., Gould, G.G., Berninger, J.P., Connors, K.A., Keele, N.B., Prosser, K.N., Brooks, B.W., 2012. Human therapeutic plasma levels of the selective serotonin reuptake inhibitor (SSRI) sertraline decrease serotonin reuptake transporter binding and shelterseeking behavior in adult male fathead minnows. Environ. Sci. Technol. 46 (4), 2427–2435. Vandenbos, R.E., Tonn, W.M., Boss, S.M., 2006. Cascading life-history interactions: alternative density-dependent pathways drive recruitment dynamics in a freshwater fish. Oecologia 148 (4), 573–582. Vaswani, M., Linda, F.K., Ramesh, S., 2003. Role of selective serotonin reuptake inhibitors in psychiatric disorders: a comprehensive review. Prog. Neuro-Psychopharmacol. Biol. Psychiatry 27 (1), 85–102. Yang, M., Qiu, W., Chen, J., Zhan, J., Pan, C., Lei, X., Wu, M., 2014. Growth inhibition and coordinated physiological regulation of zebrafish (Danio rerio) embryos upon sublethal exposure to antidepressant amitriptyline. Aquat. Toxicol. 151, 68–76.