Chemosphere 189 (2017) 498e506

Contents lists available at ScienceDirect

Chemosphere

journal homepage: www.elsevier.com/locate/chemosphere

Tributyltin induces premature hatching and reduces locomotor activity in zebrafish (Danio rerio) embryos/larvae at environmentally relevant levels

* Xuefang Liang a, b, Christopher L. Souders II b, Jiliang Zhang c, Christopher J. Martyniuk b, a School of Ecology and Environment, Inner Mongolia University, Hohhot, 010021, China b Center for Environmental and Human Toxicology, Department of Physiological Sciences, College of Veterinary Medicine, UF Genetics Institute, Interdisciplinary Program in Biomedical Sciences Neuroscience, University of Florida, Gainesville, FL, 32611, USA c Henan Open Laboratory of Key Subjects of Environmental and Animal Products Safety, College of Animal Science and Technology, Henan University of Science and Technology, Henan, China highlights

Exposure to 1 nM TBT induced premature hatching in zebrafish. TBT at 1 nM inhibited locomotor behaviors of zebrafish larvae. Mitochondrial bioenergetics were not affected by TBT. Genes related to muscle function and dopamine signaling were not altered by TBT. article info abstract

Article history: Tributyltin (TBT) is an organotin compound that is the active ingredient of many biocides and antifouling Received 27 July 2017 agents. In addition to its well established role as an endocrine disruptor, TBT is also associated with Received in revised form adverse effects on the nervous system and behavior. In this study, zebrafish (Danio rerio) embryos were 18 September 2017 exposed to environmentally relevant concentrations of TBT (0.01, 0.1, 1 nM) to determine how low levels Accepted 19 September 2017 affected development and behavior. Fish exposed to 1 nM TBT hatched earlier when compared to con- Available online 20 September 2017 trols. Following a 96-h exposure, total swimming distance, velocity, and activity of zebrafish larvae were Handling Editor: David Volz reduced compared to controls. To identify putative mechanisms for these altered endpoints, we assessed embryo bioenergetics and gene expression. We reasoned that the accelerated hatch time could be related Keywords: to ATP production and energy, thus embryos were exposed to TBT for 24 and 48-h exposure prior to Tributyltin hatch. There were no differences among groups for endpoints related to bioenergetics (i.e. basal, ATP- Mitochondrial bioenergetics dependent, and maximal respiration). To address mechanisms related to changes in behavioral activ- Locomotion ity, we measured transcripts associated with muscle function (myf6, myoD, and myoG) and dopamine Dopaminergic signaling signaling (th, dat, dopamine receptors) as dopamine regulates behavior. No transcript was altered in Premature hatching expression by TBT in larvae, suggesting that other mechanisms exist that may explain changes in higher level endpoints. These results suggest that endpoints related to the whole animal (i.e. timing of hatch and locomotor behavior) are more sensitive to environmentally-relevant concentrations of TBT compared to the molecular and metabolic endpoints examined here. © 2017 Elsevier Ltd. All rights reserved.

1. Introduction

The organotin compound tributyltin (TBT) has been widely applied as a biocide in antifouling coatings for the shipping in- * Corresponding author. Center for Environmental and Human Toxicology & dustry since the 1970s (Novelli et al., 2002). TBT is a documented Department of Physiological Sciences, College of Veterinary Medicine, University of endocrine disruptor that induces imposex in mollusks and in- Florida, 2187, Mowry Rd. Bldg 471, Gainesville, FL, 32611, USA. terferes with reproductive processes in bivalves and fish (Ruiz et al., E-mail address: cmartyn@ufl.edu (C.J. Martyniuk). https://doi.org/10.1016/j.chemosphere.2017.09.093 0045-6535/© 2017 Elsevier Ltd. All rights reserved. X. Liang et al. / Chemosphere 189 (2017) 498e506 499

1995; McAllister and Kime, 2003; Shimasaki et al., 2003, 2006; 2015). However, the full scope of TBT-induced nervous and Zhang et al., 2007, 2009; Matthiessen, 2008). Studies also show behavior dysfunction, as well as its role in regulating mitochondrial that TBT can exert adverse effects on non-target aquatic organisms bioenergetics in fish, remains unclear. at very low concentrations in the water. For example, exposure to In this study, zebrafish (Danio rerio) embryos were exposed to 0.1 ng/L TBT for 70 days induced masculinization and sperm ab- environmentally relevant concentrations of TBT (0.01, 0.1, 1 nM). normality in zebrafish (Danio rerio)(McAllister and Kime, 2003). As We first assessed endpoints that were ecologically relevant, and a result, TBT was banned globally by the international community this included survival, presence of deformity, hatch rates, and in 2008 (Gipperth, 2009). As of 2009, the reduction of TBT levels in locomotion. To address potential mechanisms of toxicity, we also seawater and biota on a global scale has been more than 50% since measured oxygen consumption rates (bioenergetics), and the the ban (Kim et al., 2014). Tributyltin in seawater in Jinhae Bay, expression of genes related to muscle function and dopamine South Korea, for example decreased more than 50-fold from ~1995 signaling, the rationale being that these systems are involved in to 97 (mean 11.3 ± 8.2 ng/L) to 2008e09 (mean 0.2 ± 0.3 ng/L) after regulating locomotor activity and there is evidence that neuro- the total ban (Kim et al., 2014). However, due to its persistence and transmitter systems are perturbed in fish with TBT exposure. We widespread use, TBT residues in aquatic environments continue to hypothesized that TBT impairs locomotion in zebrafish larvae, and remain a concern in many regions. For example, in China, TBT levels that this is related to lower ATP production and suppressed in water varied from non-detectable to 93.8 ng/L (measured as Sn) expression of genes associated with muscle and the dopamine between the years of 2000e03 (Cao et al., 2009). In water, TBT also system, which is a dominant neurotransmitter involved in regu- ranged between 0.1 and 103 ng Sn/L in different ports of India from lating behavior. 2007 to 2008 (Garg et al., 2010, 2011) and levels in the seawater along the Croatian Adriatic Coast in 2009 and 2010 ranged 2. Materials and methods 10.1e72.4 ng/L as Sn (Furdek et al., 2012). In addition, TBT in sedi- ment can reach concentrations >1000 ng Sn/g, and this can be one 2.1. Chemicals of the leading sources of TBT contamination for aquatic ecosystems (Kim et al., 2015). Thus, despite a significant reduction in global TBT Tributyltin (TBT, CAS no. 7486-35-3, purity >97%) was pur- levels over the past several years, the biological effects of TBT at low chased from Sigma-Aldrich Co. LLC (USA). Stock solutions of 1, 10 concentrations, and its environmental persistence, ensures that and 100 mM TBT were prepared by dilution in ethanol (EtOH, CAS TBT remains a concern for the environment. no. 64-17-5, purity >99.5þ%, Acros Organics). The exposure solu- In addition to reproductive effects as an anti-androgen, TBT af- tions were prepared by adding 1 ml of each stock solution into fects the nervous system of fish by altering neurotransmitter levels. 100 mL of embryo rearing medium (ERM; 8 g NaCl, 0.4 g KCl, TBT alters monoamine concentrations in the brain and studies 0.035 g Na2HPO4, 0.6 g KH2PO4, 0.14 g CaCl2, 0.12 g MgSO4, 0.35 g report that the concentration of dopamine (DA) and 5- NaHCO3, into 1 L, pH 7.2) yielding the final concentrations of 0.01, hydroxytryptamine (5-HT) of male and female medaka (Oryzias 0.1 and 1 nM TBT (equal 3.17, 31.7, and 317 ng/L). A vehicle treat- latipes) are significantly decreased with exposure at 125 mg TBT/g ment of EtOH served as the control. The ratio of vehicle to ERM was diet/d for 21 days (Nakayama et al., 2007). Conversely, in another 1:100,000 (vol/vol). study, DA was increased in the brain of rockfish Sebastiscus mar- moratus following exposure to 10e1000 ng/L TBT while levels of 5- 2.2. Experimental design HT were decreased with treatment (Yu et al., 2013). In addition, the levels of the neurotransmitter glutamate were increased following Adult zebrafish (ZF, Wild type AB) were raised at the University exposure to 500 mg/kg TBT (Zuo et al., 2009) and the N-methyl-d- of Florida (Gainesville) in the Cancer Genetics Research Center aspartate receptor (NMDAR) signaling pathway was inhibited by (CGRC). The mean water pH was 7.2 ± 1.0 and mean temperature TBT in the brains of rockfish (Zuo et al., 2009; Yu et al., 2013). Thus, was 28.1 ± 1.0 C, monitored daily. Dissolved oxygen was measured ® there is good evidence for neurotransmitter alterations by TBT at 6.6 ppm using a LeMotte Freshwater Fish Farm test kit (Pentair). which underlie the scientific premise that this compound may Fish were exposed to 14 h light and 10 h dark per day. ZF were affect behavior. This is an important component to investigate as randomly selected from a breeding stock and placed in a shallow behavioral endpoints can be used to detect neuronal effects of water breeding tank the night before embryo collection. The ratio environmental stressors (Robinson, 2009). Previous studies was two males to two females per tank and there were two tanks demonstrate that predatory behavior (Yu et al., 2013), antipredator used to generate the fertilized embryos. A divider separated the behavior (Wibe et al., 2001), shoaling and anxiety behavior (Zhang males and females overnight. These dividers were removed at 8:00 et al., 2016b), swimming distances, speed and orientation am when the facility lights turned on and spawning occurred. Both (Triebskorn et al., 1994), as well as feeding behavior of fish, are tanks yielded viable fertilized embryos. These embryos were disrupted by TBT (Zhang et al., 2016a). Thus, changes in behavior pooled in 2 petri dishes, totaling ~450 embryos. Using a light mi- may be related to disruptions in neurotransmitter systems. croscope, unfertilized eggs were identified and removed. The In addition to impairments in the central nervous system (CNS), fertilized eggs (~300e400) determined to be at the same stage in vivo studies have also indicated that TBT is a xenobiotic mito- (shield stage, 6e8 h post fertilization, hpf) were selected and used chondrial toxin that can inhibit mitochondrial ATP production and in subsequent experiments. Embryos were staged using criteria trigger cell death (Corsini et al., 1997; Stridh et al., 1999). Mito- outlined by Kimmel et al. (1995). For mitochondrial respiration chondria are major producers of energy in the cell and deficits in measurements and real-time PCR (qPCR), 10 embryos were equally energy production can result in motor deficits (DiMauro and Schon, distributed into one glass beaker containing 10 mL of each exposure 2008). In addition to production of ATP, mitochondria also control solution. Five replicate beakers were performed for each condition calcium homeostasis, generate reactive oxygen species (ROS), and (control, 0.01, 0.1, and 1 nM TBT) and this experiment was repeated regulate apoptosis, in addition to many other cellular signaling twice to increase sample size for bioassays. Embryos were main- cascades. Evidence suggests that TBT can inhibit energy meta- tained in an incubator at 26 ± 1 C and exposed to the same light to bolism in Chinese rare minnow (Gobiocypris rarus) larvae (Li and Li, dark schedule as above. For the duration of the experiment, em- 2015), and induce oxidative stress and cause central nervous sys- bryos were not disturbed, and this constituted a static waterborne tem damage in juvenile common carp (Cyprinus carpio)(Li et al., exposure over 96 h with 90% water renewal with the chemical each 500 X. Liang et al. / Chemosphere 189 (2017) 498e506 day. For studies on development (survival, deformities, and hatch well with 100 mL ERM (four wells per treatment). Thus, the total success) and locomotion, embryos were exposed in a 96-well cul- volume of ERM in each well was 525 mL. Blanks contained 525 mL ture plate. Embryos at ~8 hpf were transferred to each individual ERM with no embryos (n ¼ 4). The instrument was programmed to well which was filled with 200 mL exposure solution. add a volume of 75 mL each of challenge solutions of oligomycin (75.2 mM), carbonyl cyanide-p- trifluoromethoxyphenylhydrazone 2.3. Embryo development (FCCP, 54 mM), and sodium azide (200 mM) to give final concen- trations in the wells of 9.4 mM, 6 mM, and 20 mM, respectively. The Twenty-four hours prior to this assay, a sterile 96-well plate was protocol consisted of the following time cycles: 2 min for mixing, pseudorandomized into groups of control and TBT treatments per 1 min paused, and then 2 min to measure oxygen levels and pH. Ten column and each well was pre-coated with 200 mL of its corre- cycles of data were collected for basal respiration. Eighteen cycles sponding treatment. Pre-incubation was done to saturate the well were used for oligomycin to inhibit ATP production of embryos. plate and prevent the plastic from binding any chemicals from the Eight cycles were set for the FCCP incubations which maximizes solution. Prior to the assay, the well plate was washed 3 times with respiration. Sodium azide was introduced at the final injection for ERM and allowed to air dry. Embryos were transferred to wells of a 24 cycles to completely inhibit mitochondrial respiration of 96-well culture plate and were treated with the control or one of zebrafish embryos. the three concentrations of TBT. Each exposure group included n ¼ 16e20 embryos. The 96-well plate was incubated at 26 Cinan 2.6. RNA extraction and cDNA synthesis EVOS FL Cell Imaging System (Thermo Scientific, USA) which uses an automated routine to acquire images of each well every hour for After 6 d exposure, larvae in each group were collected and 5 96 h. Thus, a total of 96 pictures were recorded for each fish and larval fish were pooled into one centrifuge tube from each beaker manually scored for several developmental parameters, including (n ¼ 7/condition). Once pooled, samples were flash-frozen using hatching time, mortality, and malformation using the method of liquid nitrogen and stored at 80 C until subsequent RNA Reimers et al. (2004). extraction. Extraction of RNA from ZF larvae was performed using ® 1 mL TRIzol Reagent (Life Technologies, Carlsbad, CA, USA) as per 2.4. Zebrafish locomotion manufacturer's protocol. Immediately after extraction, RNA pellets were dissolved in 20 mL of RNase-DNase free water and purified After exposure for 96 h, the culture plate was transferred from through the RNeasy Mini Kit column, as per manufacturer's pro- the EVOS FL Auto Cell Imaging System (Thermo Scientific, USA) to a tocol (Qiagen, Valencia, CA, USA). Purified RNA samples were DanioVision instrument (Noldus Information Technology). After assessed for quality using the 2100 Bioanalyzer (Agilent Technol- being acclimated at 26 C in the instrument for 14 h, the activities of ogies, Santa Clara, CA, USA). The RNA integrity number (RIN) for fish were recorded by an infrared analog camera within the Dan- samples used in this study ranged from 9.0 to 9.7. The mean RIN ioVision observation chamber, and the resulting video was simul- value was 9.4 (SD ± 0.25). The concentration of RNA was deter- taneously and individually tracked using the EthoVision software mined using the Qubit Fluorometric Quantitation (Thermo Scien- version 12.0 (Noldus Information Technology, Leesburg, VA). Larvae tific, USA) and the 260/280 and 260/230 ratios were examined to were tracked following a standard 50 min “white light routine” of confirm sample purity. alternating 10 min periods of light and dark beginning with a dark The cDNA synthesis was performed using 1.2 mg of column pu- period. Analysis profiles were generated in EthoVision for distance rified RNA using iScript (BioRad), following the manufacturer's moved, velocity, activity, and turn angle and an excel file was protocol. After addition of nuclease free water, sample volume was exported for further analysis. In this visual motor response test, a 15 mL. No reverse transcriptase (NRT) (n ¼ 3) controls were pre- key feature is the robust but transient behavioral activity that oc- pared in the same fashion, except that the NRT contained no reverse curs in response to sudden transitions from light to dark. This test, transcriptase. Once prepared, samples were placed into a T100™ in principle, relies on the integrity of brain function, nervous sys- Thermal Cycler (BioRad, USA). The cDNA was generated using the tem development, locomotor system development and visual following steps: 25 C for 5 min, 42 C for 30 min, 85 C for 5 min, pathways and has been used in various assays (Ali et al., 2012; and a final cycle of 4 C for 5 min. Padilla et al., 2011; Zhang et al., 2017). 2.7. Real-time PCR 2.5. Mitochondrial respiration measurement Primer sets for target genes were collected from literature or Two separate experiments were conducted for mitochondrial designed using Primer 3 (Table S1). The transcripts measured in this respiration. After exposure for either 24 or 48 h, one embryo (n ¼ 4 study included tyrosine hydroxylase (th), dat (also known as solute total) from separate biological replicates from each treatment was carrier family 6 (neurotransmitter transporter), member 3; slc6a3), selected for mitochondrial respiration assessment using a XFe24 dopamine receptors (drd1, drd2a, drd2b, drd3, drd4a, drd4b, and Extracellular Flux Analyzer (Seahorse Bioscience, Massachusetts, drd4c), myogenic factor 6 (myf6), myogenic differentiation 1 USA). This instrument measures oxygen consumption and pH (myoD), and myogenin (myoG). Real-time PCR was performed using change using solid state sensor probes. These probes quantify the the CFX96™ Real-Time PCR Detection System (BioRad) with SSo- ® changing concentrations of dissolved oxygen and free protons in Fast™ EvaGreen Supermix (BioRad, Hercules, CA, USA), 100 nM of the micro-environment surrounding individual intact embryos. The each forward and reverse primer, and 5 mL of cDNA (diluted 20-fold rate of O2 consumption (OCR) can be assigned to oxidative phos- prior to real-time analysis). The two-step thermal cycling param- phorylation and the rate of extracellular acidification (ECAR) to eters were as follows: initial 1-cycle Taq activation at 95 C for 30 s, glycolysis. For this experiment, we used the 24-well Islet Plate to followed by 95 C for 5 s, and primer annealing for 5 s. After 39 accommodate live embryos. Each well of an Islet Plate was filled cycles, a dissociation curve was generated, starting at 65.0 and with 1 mL of XF Calibrant fluid, and the Islet Plate was incubated ending at 95.0 C, with increments of 0.5 C every 5 s. with the sensor plate overnight at 28 C. Each well of an islet Four reference genes (beta actin, b-actin; elongation factor 1, capture microplate contained an initial volume of 425 mL ERM. After ef1a; ribosomal subunit 18, rps18; and ribosomal protein L13a, washing twice with ERM, a single embryo was added to each islet rpl13a) were assessed for stability and appropriateness for X. Liang et al. / Chemosphere 189 (2017) 498e506 501 normalization. Expression values of reference genes were evaluated embryos exposed to 1 nM TBT (Chi square ¼ 9.45, p ¼ 0.002). At 70 statistically using a Kruskal-Wallis test, determining whether hpf, the hatching rate of 1 nM TBT group (26.5%) was higher than expression levels significantly varied across experimental groups. that of the control (7.6%) (Fig. 1A). No differences in hatching rates This was based upon total RNA input for the cDNA synthesis. The were observed for the lower doses of TBT (p ¼ 0.6 and p ¼ 0.2 for mean expression levels of rps18 and rpl13a were determined to be 0.01 and 0.1 nM TBT, respectively) (Fig. 1A). In addition, hatching the most stable combination of reference genes to normalize all time in the 1 nM TBT group (mean ¼ 66.2 hpf) was accelerated by target genes. The target stability function in the CFX96 software 13.8% compared to the control (mean ¼ 76.8 hpf) (df ¼ 75, F ¼ 6.34, determined that the combined M-value for rps18 and rpl13a was p ¼ 0.0007), while no differences were detected in 0.01 and 0.1 nM 0.55 (CV ¼ 0.19). Each primer set was tested for linearity (>0.97) TBT (p ¼ 1 and 0.43, respectively) (Fig. 1B). and efficiency (90e120%) using a 4 or 5 point standard curve generated by a dilution series from a cDNA pool of all samples. The 3.2. TBT exposure inhibited zebrafish larvae locomotor behavior qPCR analysis included 3 NRT samples and 1 NTC sample. Negative controls indicated that column purification sufficiently removed Following a 96 h exposure to TBT, locomotor behavior of genomic DNA. Normalized gene expression was extracted using zebrafish larvae was significantly inhibited (Fig. 2). There was a CFX Manager™ software with the relative DDCq method (baseline decrease in the total distance moved by 84.1% in the 1 nM TBT subtracted) based on the method of Pfaffl (2001). Sample sizes were exposure group (df ¼ 70, F ¼ 22.9, p < 0.0001), whereas, no marked n ¼ 7 for all 4 groups (control, 0.01, 0.1 and 1 nM TBT). Individual effects were detected following 0.01 and 0.1 nM TBT exposure samples that were verified as having expression levels too low to be (Fig. 2A). In addition, a significant decrease in velocity of 84.1% accurately determined were assigned the lowest measureable (df ¼ 70, F ¼ 22.9, p < 0.0001) was observed in 1 nM TBT while no expression value of the gene-specific dataset (limit of detection). All differences were detected for fish exposed to the lower doses primers used in the qPCR analysis amplified one product, indicated (Fig. 2B). Moreover, 1 nM TBT decreased the activity of larvae by by a single melt curve. 85.9% (df ¼ 70, F ¼ 23.5, p < 0.0001) while 0.01 and 0.1 nM TBT did not significantly affect the activity of larvae (Fig. 2C). No alterations 2.8. Prediction of Activity Spectra of Substances (PASS) of turn angle were observed in larvae across all concentrations of TBT compared to control, although an increase of turn angle was To learn more about putative mechanisms of action of TBT, we detected in 1 nM TBT in comparison of 0.01 and 0.1 nM TBT obtained the simplified molecular-input line-entry system (p ¼ 0.004 and 0.017, respectively) (df ¼ 66, F ¼ 4.79, p ¼ 0.0045) (SMILES) format and uploaded this to Prediction of Activity Spectra (Fig. 2D). of Substances (PASS) (Lagunin et al., 2000). Pa (probability “to be active”) estimates the chance that the studied compound belongs 3.3. TBT exposure did not alter mitochondrial bioenergetics to the sub-class of active compounds while Pi (probability “to be inactive”) estimates the chance that the studied compound belongs To assess the effects of TBT on mitochondrial bioenergetics in to the sub-class of inactive compounds. zebrafish embryos, oxygen consumption rate was measured in intact zebrafish embryos after 24 and 48-h exposure, respectively 2.9. Statistical analysis (Figs. 3A and 4A). After 24-h exposure, basal respiration, maximal respiration, mitochondrial respiration, and non-mitochondrial Statistical analysis of variance was performed using SPSS respiration were not changed after TBT exposure (Fig. 3BeE). For (version 17.0). The normality of data was assessed using the 48-h exposed embryos, basal respiration was different among Kolmogorov-Smirnov test and logarithmic transformation was groups after a 48-h exposure to TBT (df ¼ 15, F ¼ 4.01, p ¼ 0.03), performed prior to ANOVA to ensure data conformed to assump- however this difference was detected between 0.1 nM TBT and tions. The Levene's test of homogeneity of variance was used to test 1 nM TBT (p ¼ 0.046) (Fig. 4B). This was also the case for maximal for adherence to one-way analysis of variance (ANOVA) assump- respiration (df ¼ 15, F ¼ 4.36, p ¼ 0.027), and the difference was tions. Hatching time, locomotion analysis, and mitochondrial observed between 0.1 nM and 1 nM TBT (p ¼ 0.035) (Fig. 4C). A respiration were analyzed using ANOVA followed by Tukey's mul- significant difference in mitochondrial respiration was observed tiple comparison tests to compare the differences between groups. between 0.1 nM TBT and 1 nM TBT as well as 0.01 nM and 1 nM TBT All quantitative data above are expressed as mean ± standard error treatments (p ¼ 0.047 and 0.043, respectively) (Fig. 4D). No dif- (SEM). For hatching rates, the Log-rank (Mantel-Cox) test was used ferences in non-mitochondrial respiration were detected (df ¼ 15, to compare the differences between treatments and the control. F ¼ 2.01, p ¼ 0.17) (Fig. 4E). Patterns in energy utilization in The qPCR results were analyzed by Kruskal-Wallis followed by zebrafish embryos after 24 and 48 h TBT exposure was also assessed Dunn's multiple comparison tests. A probability of p < 0.05 was (Figs. 3F and 4F) and OCAR vs ECAR were plotted. Overall, there considered to be statistically significant. All figures were prepared were no dramatic shifts in the metabolic capacity of embryos using Graph-Pad Prism version 6.0 (GraphPad Software Inc., La following TBT exposure and the analysis suggests that the accel- Jolla, CA, USA). erated hatching time may not be related to changes in energy production within the embryo. Additional experiments were con- 3. Results ducted with higher levels of TBT (10 and 100 nM) and these too showed that there was little change in bioenergetics at the doses 3.1. Developmental toxicity of TBT in zebrafish embryos examined (Fig. S2).

Mortality and malformation were not different between treat- 3.4. TBT exposure did not affect the expression of genes related to ments and the control after 96 h TBT exposure (percentage < 5%, myogenesis nor dopamine signaling data not shown). Deformities with TBT were low in occurrence and some fish exhibited spinal curve, pericardial edema and axial In order to determine the effects of TBT on myogenesis in malformation (Fig. S1) but again, this did not differ from those in developing zebrafish, the mRNA levels of myf6, myoD, and myoG the control group. Fish in the control group hatched between 70 were measured (Fig. 5). However, after 6d exposure of TBT, no and 80 hpf (Fig. 1A). An increased rate of hatching was detected for significant differences in myf6, myoD and myoG mRNA levels were 502 X. Liang et al. / Chemosphere 189 (2017) 498e506

Fig. 1. Effects of waterborne TBT exposure on (A) hatching rate and (B) hatching time of zebrafish embryos during 8e96 hpf. Hatching time was analyzed using ANOVA followed by Tukey's multiple comparison. Hatching rate was analyzed using the Log-rank (Mantel-Cox) test; significant change of the curves was detected between control and 1 nM TBT. Data are expressed as mean ± standard error (SEM). Different letters denote significant differences (p < 0.05) between treatments and controls.

Fig. 2. Locomotor behavior of zebrafish embryos exposed to either control or one dose of 0.01, 0.1 or 1 nM TBT over 96 h. The parameters analyzed included (A) the total distance moved, (B) the velocity, (C) the activity, and (D) the turn angle for individual larvae. Data were collected during 50 min White light routine. Results are expressed as mean ± SEM. Different letters denote significant differences (p < 0.05) between treatments and controls. detected at any concentration of TBT (Fig. 5). showed no difference in expression following TBT treatments Since the locomotor behavior was affected by TBT, we hypoth- (Fig. 6A and D-G). Although the level of dat was decreased in 1 nM esized that DA signaling may also be disturbed as this neuro- TBT group compared to 0.1 nM TBT group (2.6-fold, p ¼ 0.03), no transmitter system plays a significant role in locomotion. We significant changes were observed in any TBT treatment compared examined transcriptional levels of genes involved in DA synthesis to the control, (Fig. 6B). Similarly, levels of drd1 were significantly (th), transport (dat), and receptor (drd1, drd2a, drd3, drd4b, and decreased in 1 nM TBT group compared to 0.1 nM TBT group (1.8- drd4c)(Fig. 6). Levels of drd2b and drd4a were also assessed but the fold, p ¼ 0.04), but there were no changes observed in any TBT expression levels were too low to be accurately measured (data not group compared to the control (Fig. 6C). shown). Among the tested genes, th, drd2a, drd3, drd4b, and drd4c X. Liang et al. / Chemosphere 189 (2017) 498e506 503

Fig. 3. Mitochondrial respiration of zebrafish embryos exposed to either control or one dose of 0.01, 0.1 or 1 nM TBT for 24 h. (A) Oxygen consumption rate (OCR) of each condition after the injection of oligomycin, FCCP and sodium azide; (B) Basal respiration, (C) maximal respiration, (D) mitochondrial respiration, and (E) non-mitochondrial respiration following exposure to TBT; (F) shift in mode of energy utilization plotting using indicators of oxidative phosphorylation (measured by OCR) and glycolysis (measured by extracellular acidification rate, ECAR). Data are presented as mean ± S.E.

Fig. 4. Mitochondrial respiration of zebrafish embryos exposed to either control or one dose of 0.01, 0.1 or 1 nM TBT for 48 h. (A) Oxygen consumption rate (OCR) of each condition after the injection of oligomycin, FCCP and sodium azide; (B) Basal respiration, (C) maximal respiration, (D) mitochondrial respiration, and (E) non-mitochondrial respiration following exposure to TBT; (F) shift in mode of energy utilization plotting using indicators of oxidative phosphorylation (measured by OCR) and glycolysis (measured by extracellular acidification rate, ECAR). Data are presented as mean ± S.E. Different letters denote significant differences (p < 0.05) between treatments and controls.

4. Discussion mortality and abnormal embryonic development in zebrafish. It has been reported that doses as low as 0.1, or 1 ng L 1 TBT reduces In the present study, mortality and malformation were not hatch rates and cause morphological abnormalities including dor- significantly different among groups after 96 h exposure to TBT. sal curvature, twisted tails, pericardial edema and craniofacial Hano et al. (2007) observed that 160 ng TBT/g egg induced higher skeletal deformities in rockfish (Sebastiscus marmoratus) embryos 504 X. Liang et al. / Chemosphere 189 (2017) 498e506

Fig. 5. The mRNA levels of (A) myf6, (B) myoD, (C) myoG in zebrafish larvae following treatment with TBT for 6 days. Each point represents normalized expression in each biological replicate (n ¼ 7), and the line represents the median value of the data. Detection limit of each gene was defined as the lowest values within each gene dataset. Different letters denote significant differences (p < 0.05) between treatments and controls.

Fig. 6. The mRNA levels of (A) th, (B) dat, (C) drd1, (D) drd2a, (E) drd3, (F) drd4b, (G) drd4c, which are related to dopaminergic signaling in zebrafish, were examined after exposure to TBT for 6 days. Each point represents normalized expression in each biological replicate (n ¼ 7), and the line represents the median value of the data. Detection limit of each gene was defined as the lowest values within each gene dataset. Different letters denote significant differences (p < 0.05) between treatments and controls.

(Zhang et al., 2011, 2012). A lack of response for some morpho- oxygen consumption rates with TBT exposure, suggesting that metric deformities may be attributed to the route of exposure (i.e. there may be other reasons underlying the accelerated hatch rate. waterborne or injection), or variability due to the experimental Measuring ATP levels within the embryo may be a better indicator model (Gardner et al., 1979; Worek et al., 2002; Baird and Van den of impaired energy stores, and may be a valid endpoint to assess in Brink, 2007). However, hatching rate was increased after exposure future studies. Noteworthy here is that a change in the time to to the highest dose (1 nM) of TBT, and was elevated at early periods hatch is an important stress response for fish larvae (Barton, 2002). of development (60-70hpf). Thus, time to hatch may be a more Prematurely hatched embryos can exhibit developmental de- sensitive indicator for adverse effects in the early developing stage formities such as growth reduction, spinal curvature (body and tail of zebrafish embryos exposed to TBT compared to morphological shape), and yolk sac edema (Barron et al., 2004; Corrales et al., defects. Hatch time can be influenced by the activity of chorionic 2014; Samaee et al., 2015). Thus, changes in hatch rates observed hatching enzymes and embryo movements (Winnicki et al., 1970; with TBT may be more related to the stress axis than altered Cheng et al., 2007). As a result, an increase in respiration rates or oxidative phosphorylation and energy production within the em- blockage of oxygen exchange can lead to early hatching (Leung and bryo. These alternative mechanisms must also be rigorously tested. Bulkley, 1979; Samaee et al., 2015). In a previous study, exposure of Behavioral responses also act as sensitive indicators of toxicant the nano-sized TiO2 particles caused premature hatching in exposure in fish early life stages (Sloman and McNeil, 2012). zebrafish embryos because of the adsorption of TiO2 to the surface Ambient light conditions in the test apparatus are of vital impor- of the chorion, thus affecting oxygen exchange and waste elimi- tance to zebrafish larvae (Ahmad et al., 2012). Based on light-dark nation (Samaee et al., 2015). Thus, we tested the hypothesis that the challenge, behavioral response has been shown to be highly sen- change in hatch rates was associated with changes in bioenergetics sitive to neuroactive chemical compounds (Ali et al., 2012). Studies of the embryo. However, we did not detect any impairment on report that TBT induces behavior impairments in different aquatic X. Liang et al. / Chemosphere 189 (2017) 498e506 505 species (Wibe et al., 2001; Yu et al., 2013; Li and Li, 2015; Zhang cell and deficits in energy production result in neurodegeneration et al., 2016b). In accordance with these studies, we observed that and motor deficits (DiMauro and Schon, 2008). Li and Li (2015) different endpoints of locomotor behavior, i.e., total swimming demonstrated that energy metabolism was inhibited in Chinese distance, velocity, and activity of zebrafish larvae were significantly rare minnow (Gobiocypris rarus) larvae exposed to 800 ng/L TBT, reduced following exposure to 1 nM TBT. In contrast to our data inducing abnormal behaviors in the fish. Despite this evidence, we showing reduced activity, Triebskorn et al. (1994) (Triebskorn et al., did not detect any significant effects of TBT exposure on oxygen 1994) reported that bis(tri-n-butyltin)oxide (TBTO)-treated consumption rates. Based on our data, we propose that rainbow trout (Oncorhynchus mykiss) hatchlings swam longer dis- environmentally-relevant concentrations of tributyltin can affect tances over prolonged times with higher velocity compared to hatching time and behavior in larval zebrafish (Danio rerio) without controls. These conflicting data (hyper- and hypo-activity) may be affecting oxygen consumption rates and bioenergetics of due to the different methodologies used in each study as locomo- development. tion of TBT-treated larval zebrafish was recorded in response to the As embryo bioenergetics (i.e. ATP production) and the expres- stimulation of lightedark conversion. sion of the dopamine system appear not to play a prominent role in We proposed two mechanisms that could explain altered TBT-induced changes in hatch rate and locomotion, we highlight behavior following TBT exposure, those being effects of TBT on the here other potential theoretical bioactivities of TBT for future muscle itself or neurotransmitter systems or both (Triebskorn et al., studies. Based upon Prediction of Activity Spectra of Substances 1994; Dong et al., 2006; Zuo et al., 2009; Yu et al., 2013). The (PASS), the results suggested that TBT acts as a testosterone 17beta- myogenic regulatory transcription factor (MRF) family includes dehydrogenase (NADPþ) inhibitor which is consistent with its anti- myf6 (mrf4), myoD and myoG which are essential for skeletal androgenic effects in aquatic species and terrestrial invertebrates as myogenesis. In fish, myoD is the major myogenic gene in the lateral well as an ubiquinol-cytochrome-c reductase inhibitor (Table S3). somite. MyoD activity drives myoG expression, which is essential However, some of the most likely potential mechanisms for TBT for fast muscle differentiation (Hinits et al., 2009). As fish mature, included those related to glucose and starch metabolism, for myf6 expression is pronounced in the region of slow muscle fibers. example sugar-phosphatase inhibitor, chymosin inhibitor, However, it does not contribute to early myogenesis in zebrafish carboxypeptidase Taq inhibitor, and glucan 1,4-alpha-malto- (Hinits et al., 2007, 2009). Our data showed that mRNA levels of triohydrolase inhibitor. Thus, experiments investigating these myf6, myoD and myoG were not affected by TBT, suggesting that mechanisms may offer new insight into TBT toxicity and may locomotion dysfunction in larvae may not involve changes in these explain, in part, changes observed in higher level endpoints. transcripts. Impairment of muscle was only assessed at the level of In conclusion, environmentally-relevant concentrations of TBT transcripts, and it is important to point out that other endpoints decreased the hatching time of zebrafish embryos and inhibited such as protein levels or skeletal histology may yield additional locomotor activity in larvae. Oxidative respiration was not signifi- physiological clues into the TBT-induced changes observed with cantly affected in developing embryos. Additionally, the expression locomotor behavior. levels of genes related to muscle function (myf6, myoD, and myoG) Dopaminergic signaling is known to regulate cellular processes and dopamine signaling (th, dat, dopamine receptors) were not (e.g. gene transcription and cell proliferation) that are crucial for affected by TBT in the larvae. Thus, changes in higher level end- neurodevelopment, and this has a direct impact on locomotor points appear not to be related to impaired mitochondrial bio- behavior in larval zebrafish (Souza and Tropepe, 2011). Dopamine is energetics nor to the expression levels of transcripts involved in synthesized from tyrosine by tyrosine hydroxylase (TH) and DOPA muscle function and dopamine signaling. Our data suggest that decarboxylase (DDC). After synthesis and release into the synaptic higher level endpoints such as hatching time and locomotor cleft, DA binds to specific post-synaptic dopaminergic receptors behavior may be more sensitive to environmentally-relevant con- (D1, D2, D3 and D4) and is subsequently recycled via the dopamine centrations of TBT compared to the molecular or physiological transporter (DAT) (Souza and Tropepe, 2011). In the present study, endpoints assessed in this study. the expressions of dat and drd1 were decreased after 1 nM TBT treatment compared to 0.1 nM TBT but the mean expression of Acknowledgements these transcripts were not different than the control group. In addition, turn angle was different between 1 nM TBT and the lower The authors have no conflict of interest to declare. This work was doses and may be related to the expression patterns observed here supported by the National Natural Science Foundation of China fi but overall, there was no signi cant impact of TBT on the expres- (21507064), the Natural Science Foundation of Inner Mongolia sion of the dopamine system. In contrast to this study, Yu et al. Autonomous Region of China (2015MS0202), and the University of (2013) detected reduced expression of drd1 in the brains of Sebas- Florida, College of Veterinary Medicine. The support provided by tiscus marmoratus exposed to 10 ng/L TBT for 50 days. Differences China Scholarship Council (CSC) during a visit of Xuefang Liang (No. between the two studies may be related to the exposure period and 201608155003) to University of Florida is acknowledged. the sensitivity of different species. Moreover, gene expression was conducted in whole homogenates of larvae, and this is likely not to Appendix A. Supplementary data reflect transcriptional changes in specific tissue such as those in the CNS. This may also explain the lack of expression changes with Supplementary data related to this article can be found at some transcripts (e.g. drd1). However, our data does not rule out an https://doi.org/10.1016/j.chemosphere.2017.09.093. association between dopamine and locomotion; dopamine con- centrations in the brain or whole body of larvae may also yield important information about TBT and behavioral changes in References fi zebra sh. fi fi Ahmad, F., Noldus, L.P., Tegelenbosch, R.A., Richardson, M.K., 2012. Zebra sh em- Due to the fact that the availability of suf cient ATP is necessary bryos and larvae in behavioural assays. Behav 149, 1241e1281. to maintain normal muscle function and locomotion (Allen et al., Ali, S., Champagne, D.L., Richardson, M.K., 2012. Behavioral profiling of zebrafish 2008), we investigated whether impaired oxidative respiration embryos exposed to a panel of 60 water-soluble compounds. Behav. Brain Res. 228, 272e283. was a potential mechanism related to the effects on hatching time Allen, D.G., Lamb, G.D., Westerblad, H., 2008. Skeletal muscle fatigue: cellular and locomotion. Mitochondria are major producers of energy in the mechanisms. Physiol. Rev. 88, 287e332. 506 X. Liang et al. / Chemosphere 189 (2017) 498e506

Baird, D.J., Van den Brink, P.J., 2007. Using biological traits to predict species (Echinodermata: echinoidea). Environ. Toxicol. Chem. 21, 859e864. sensitivity to toxic substances. Ecotoxicol. Environ. Saf. 67, 296e301. Padilla, S., Hunter, D., Padnos, B., Frady, S., MacPhail, R., 2011. Assessing locomotor Barron, M.G., Carls, M.G., Heintz, R., Rice, S.D., 2004. Evaluation of fish early life- activity in larval zebrafish: influence of extrinsic and intrinsic variables. Neu- stage toxicity models of chronic embryonic exposures to complex polycyclic rotoxicol. Teratol. 33, 624e630. aromatic hydrocarbon mixtures. Toxicol. Sci. 78, 60e67. Pfaffl, M.W., 2001. A new mathematical model for relative quantification in real- Barton, B.A., 2002. Stress in fishes: a diversity of responses with particular reference time RTePCR. Nucleic Acids Res. 29 e45ee45. to changes in circulating corticosteroids. Integr. Comp. Biol. 42, 517e525. Reimers, M.J., Flockton, A.R., Tanguay, R.L., 2004. Ethanol- and acetaldehyde- Cao, D., Jiang, G., Zhou, Q., Yang, R., 2009. Organotin pollution in China: an overview mediated developmental toxicity in zebrafish. Neurotoxicol. Teratol. 26, of the current state and potential health risk. J. Environ. Manage 90, S16eS24. 769e781. Cheng, J., Flahaut, E., Cheng, S.H., 2007. Effect of carbon nanotubes on developing Robinson, P.D., 2009. Behavioural toxicity of organic chemical contaminants in fish: zebrafish (Danio rerio) embryos. Environ. Toxicol. Chem. 26, 708e716. application to ecological risk assessments (ERAs). Can. J. Fish. Aquat. Sci. 66, Corrales, J., Thornton, C., White, M., Willett, K.L., 2014. Multigenerational effects of 1179e1188. benzo[a]pyrene exposure on survival and developmental deformities in Ruiz, J.M., Bryan, G.W., Wigham, G.D., Gibbs, P.E., 1995. Effects of tributyltin (TBT) zebrafish larvae. Aquat. Toxicol. 148, 16e26. exposure on the reproduction and embryonic development of the bivalve Corsini, E., Viviani, B., Marinovich, M., Galli, C.L., 1997. Role of mitochondria and Scrobicularia plana. Mar. Environ. Res. 40, 363e379. calcium ions in tributyltin-induced gene regulatory pathways. Toxicol. Appl. Samaee, S.M., Rabbani, S., Jovanovic, B., Mohajeri-Tehrani, M.R., Haghpanah, V., Pharmacol. 145, 74e81. 2015. Efficacy of the hatching event in assessing the embryo toxicity of the DiMauro, S., Schon, E.A., 2008. Mitochondrial disorders in the nervous system. nano-sized TiO2 particles in zebrafish: a comparison between two different Annu. Rev. Neurosci. 31, 91e123. classes of hatching-derived variables. Ecotoxicol. Environ. Saf. 116, 121e128. Dong, W., Muramoto, W., Nagai, Y., Takehana, K., Stegeman, J.J., Teraoka, H., Shimasaki, Y., Kitano, T., Oshima, Y., Inoue, S., Imada, N., Honjo, T., 2003. Tributyltin Hiraga, T., 2006. Retinal neuronal cell is a toxicological target of tributyltin in causes masculinization in fish. Environ. Toxicol. Chem. 22, 141e144. developing zebrafish. J. Vet. Med. Sci. 68, 573e579. Shimasaki, Y., Oshima, Y., Inoue, S., Inoue, Y., Kang, I.J., Nakayama, K., Imoto, H., Furdek, M., Vahcic, M., Scancar, J., Milacic, R., Kniewald, G., Mikac, N., 2012. Orga- Honjo, T., 2006. Effect of tributyltin on reproduction in Japanese whiting, Sillago notin compounds in seawater and Mytilus galloprovincialis mussels along the japonica. Mar. Environ. Res. 62 (Suppl. 1), S245eS248. croatian adriatic coast. Mar. Pollut. Bull. 64, 189e199. Sloman, K.A., McNeil, P.L., 2012. Using physiology and behaviour to understand the Gardner, D.E., Miller, F.J., Blommer, E.J., Coffin, D.L., 1979. Influence of exposure responses of fish early life stages to toxicants. J. Fish. Biol. 81, 2175e2198. mode on the toxicity of NO2. Environ. Health Perspect. 30, 23e29. Souza, B.R., Tropepe, V., 2011. The role of dopaminergic signalling during larval Garg, A., Meena, R.M., Bhosle, N.B., 2010. Distribution of butyltins in waters and zebrafish brain development: a tool for investigating the developmental basis sediments of the Mandovi and Zuari estuaries, west coast of India. Environ. of neuropsychiatric disorders. Rev. Neurosci. 22, 107e119. Monit. Assess. 165, 643e651. Stridh, H., Fava, E., Single, B., Nicotera, P., Orrenius, S., Leist, M., 1999. Tributyltin- Garg, A., Meena, R.M., Jadhav, S., Bhosle, N.B., 2011. Distribution of butyltins in the induced apoptosis requires glycolytic adenosine trisphosphate production. waters and sediments along the coast of India. Mar. Pollut. Bull. 62, 423e431. Chem. Res. Toxicol. 12, 874e882. Gipperth, L., 2009. The legal design of the international and European Union ban on Triebskorn, R., Kohler,€ H.R., Flemming, J., Braunbeck, T., Negele, R.D., Rahmann, H., tributyltin antifouling paint: direct and indirect effects. J. Environ. Manage 90 1994. Evaluation of bis(tri-n-butyltin)oxide (TBTO) neurotoxicity in rainbow (Suppl. 1), S86eS95. trout (Oncorhynchus mykiss). I. Behaviour, weight increase, and tin content. Hano, T., Oshima, Y., Kim, S.G., Satone, H., Oba, Y., Kitano, T., Inoue, S., Shimasaki, Y., Aquat. Toxicol. 30, 189e197. Honjo, T., 2007. Tributyltin causes abnormal development in embryos of Wibe, Å.E., Nordtug, T., Jenssen, B.M., 2001. Effects of bis(tributyltin)oxide on medaka, Oryzias latipes. Chemosphere 69, 927e933. antipredator behavior in three spine stickleback Gasterosteus aculeatus L. Che- Hinits, Y., Osborn, D.P.S., Carvajal, J.J., Rigby, P.W.J., Hughes, S.M., 2007. Mrf4 (myf6) is mosphere 44, 475e481. dynamically expressed in differentiated zebrafish skeletal muscle. Gene Expr. Winnicki, A., Stankowska-Radziun, M., Radziun, K., 1970. Structural and mechanical Patterns 7, 738e745. changes in the egg membranes of Salmo gairdneri Rich. during the period of Hinits, Y., Osborn, D.P.S., Hughes, S.M., 2009. Differential requirements for myogenic hatching of the larvas. Acta Ichthyol. Piscat. 1, 7e20. regulatory factors distinguish medial and lateral somitic, cranial and fin muscle Worek, F., Reiter, G., Eyer, P., Szinicz, L., 2002. Reactivation kinetics of acetylcho- fibre populations. Development 136, 403e414. linesterase from different species inhibited by highly toxic organophosphates. Kim, N.S., Hong, S.H., An, J.G., Shin, K.-H., Shim, W.J., 2015. Distribution of butyltins Arch. Toxicol. 76, 523e529. and alternative antifouling biocides in sediments from shipping and ship- Yu, A., Wang, X., Zuo, Z., Cai, J., Wang, C., 2013. Tributyltin exposure influences building areas in South Korea. Mar. Pollut. Bull. 95, 484e490. predatory behavior, neurotransmitter content and receptor expression in Kim, N.S., Hong, S.H., Yim, U.H., Shin, K.H., Shim, W.J., 2014. Temporal changes in Sebastiscus marmoratus. Aquat. Toxicol. 128, 158e162. TBT pollution in water, sediment, and oyster from Jinhae Bay after the total ban Zhang, J., Sun, P., Yang, F., Kong, T., Zhang, R., 2016a. Tributyltin disrupts feeding and in South Korea. Mar. Pollut. Bull. 86, 547e554. energy metabolism in the goldfish (Carassius auratus). Chemosphere 152, Kimmel, C.B., Ballard, W.W., Kimmel, S.R., Ullmann, B., Schilling, T.F., 1995. Stages of 221e228. embryonic development of the zebrafish. Dev. Dyn. 203, 253e310. Zhang, J., Souders, C.L., Denslow, N.D., Martyniuk, C.J., 2017. Quercetin, a natural Lagunin, A., Stepanchikova, A., Filimonov, D., Poroikov, V., 2000. PASS: prediction of product supplement, impairs mitochondrial bioenergetics and locomotor activity spectra for biologically active substances. Bioinformatics 16, 747e748. behavior in larval zebrafish (Danio rerio). Toxicol. Appl. Pharmacol. 327, 30e38. Leung, T.S., Bulkley, R.V., 1979. Effects of petroleum hydrocarbons on length of in- Zhang, J., Zhang, C., Sun, P., Shao, X., 2016b. Tributyltin affects shoaling and anxiety cubation and hatching success in the Japanese medaka. Bull. Environ. Contam. behavior in female rare minnow (Gobiocypris rarus). Aquat. Toxicol. 178, 80e87. Toxicol. 23, 236e243. Zhang, J., Zuo, Z., Chen, Y., Zhao, Y., Hu, S., Wang, C., 2007. Effect of tributyltin on the Li, Z., Li, P., 2015. Evaluation of tributyltin toxicity in Chinese rare minnow larvae by development of ovary in female cuvier (Sebastiscus marmoratus). Aquat. Toxicol. abnormal behavior, energy metabolism and endoplasmic reticulum stress. 83, 174e179. Chem. Biol. Interact. 227, 32e36. Zhang, J., Zuo, Z., He, C., Cai, J., Wang, Y., Chen, Y., Wang, C., 2009. Effect of tributyltin Li, Z., Li, P., Shi, Z., 2015. Chronic exposure to tributyltin induces brain functional on testicular development in Sebastiscus marmoratus and the mechanism damage in juvenile common carp (Cyprinus carpio). PLoS One 10, e0123091. involved. Environ. Toxicol. Chem. 28, 1528e1535. Matthiessen, P., 2008. An assessment of endocrine disruption in mollusks and the Zhang, J., Zuo, Z., Sun, P., Wang, H., Yu, A., Wang, C., 2012. Tributyltin exposure potential for developing internationally standardized mollusk life cycle test results in craniofacial cartilage defects in rockfish (Sebastiscus marmoratus) guidelines. Integr. Environ. Assess. Manag. 4, 274e284. embryos. Mar. Environ. Res. 77, 6e11. McAllister, B.G., Kime, D.E., 2003. Early life exposure to environmental levels of the Zhang, J., Zuo, Z., Wang, Y., Yu, A., Chen, Y., Wang, C., 2011. Tributyltin chloride aromatase inhibitor tributyltin causes masculinisation and irreversible sperm results in dorsal curvature in embryo development of Sebastiscus marmoratus damage in zebrafish (Danio rerio). Aquat. Toxicol. 65, 309e316. via apoptosis pathway. Chemosphere 82, 437e442. Nakayama, K., Oshima, Y., Tachibana, T., Furuse, M., Honjo, T., 2007. Alteration of Zuo, Z., Cai, J., Wang, X., Li, B., Wang, C., Chen, Y., 2009. Acute administration of monoamine concentrations in the brain of medaka, Oryzias latipes, exposed to tributyltin and trimethyltin modulate glutamate and N-methyl-d-aspartate tributyltin. Environ. Toxicol. 22, 53e57. receptor signaling pathway in Sebastiscus marmoratus. Aquat. Toxicol. 92, Novelli, A.A., Argese, E., Tagliapietra, D., Bettiol, C., Ghirardini, A.V., 2002. Toxicity of 44e49. tributyltin and triphenyltin to early life-stages of Paracentrotus lividus