The presence of MWCNTs reduces developmental toxicity of PFOS in early life stage of zebrafish

Environmental Pollution xxx (2016) 1e9

Contents lists available at ScienceDirect

Environmental Pollution journal homepage: www.elsevier.com/locate/envpol

The presence of MWCNTs reduces developmental toxicity of PFOS in early life stage of zebrafish* Shutao Wang a, *, Changlu Zhuang a, b, Jia Du a, Chuan Wu a, Hong You a a b

State Key Laboratory of Urban Water Resource and Environment, Harbin Institute of Technology, Harbin 150028, China Life Science and Environmental Science Research Center, Harbin Institute of Commerce, Harbin 150028, China

a r t i c l e i n f o

a b s t r a c t

Article history: Received 1 September 2016 Received in revised form 13 December 2016 Accepted 18 December 2016 Available online xxx

Both carbon nanotubes (CNTs) and perfluorooctane sulfonate (PFOS) are used widely. There is considerable concern regarding their ecotoxicity. CNTs might interact with PFOS in water and result in different impacts compared with those after single exposures. To our knowledge, the developmental toxicity of PFOS in the presence of multi-walled carbon nanotubes (MWCNTs) in the early life stage of zebrafish (from 3 h post fertilization (hpf) to 96 hpf) was investigated for the first time in this study. The embryos and larvae were exposed to PFOS (0.2, 0.4, 0.8, and 1.6 mg/L), MWCNTs (50 mg/L), and a mixture of both. Compared with PFOS exposure, the adverse effects induced by PFOS on the hatching rate of zebrafish embryos and the heart rate and body length of zebrafish larvae were reduced in the presence of MWCNTs, and mortality and malformation were also alleviated. In addition, zebrafish larvae exposed to PFOS showed decreased activities of superoxide dismutase, catalase, and glutathione peroxidase, as well as decreased levels of reactive oxygen species and malondialdehyde, in the presence of MWCNTs, indicating that oxidative stress and lipid peroxidation was relieved. Thus, the presence of MWCNTs reduces the developmental toxicity of PFOS in the early life stage of zebrafish. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Multi-walled carbon nanotubes (MWCNTs) Perfluorooctane sulfonate (PFOS) Developmental toxicity Zebrafish embryo and larvae Oxidative stress Lipid peroxidation (LPO)

1. Introduction Carbon nanotubes (CNTs), a typical carbon-based nanomaterial, have achieved considerable attention. CNTs are used for many applications, including catalysis, absorbent materials, and battery electrodes, owing to their unique chemical and physical characteristics. In general, there are two kinds of CNTs: single-walled carbon nanotubes (SWCNTs) and multi-walled carbon nanotubes (MWCNTs). During the production and use of CNTs, some of them might enter the environment, especially the aquatic system. Assessment of their ecotoxicity is necessary for understanding the potential adverse effects of these materials on the environment. Considerable attention is paid to the ecological safety of MWCNTs because of their high chemical activity (Boncel et al., 2015; Ema

* This paper has been recommended for acceptance by Dr. Harmon Sarah Michele. * Corresponding author. School of Municipal and Environmental Engineering, Harbin Institute of Technology, 73, Huanghe Rd., Nangang Dist., Harbin 150090, China. E-mail addresses: [email protected] (S. Wang), [email protected] (C. Zhuang), [email protected] (J. Du), [email protected] (C. Wu), youhong@ hit.edu.cn (H. You).

et al., 2016). In recent years, many studies have attempted to confirm whether MWCNTs produce hazardous effects on hydrobios, animals, plants, and humans. For example, exposure to MWCNTs was reported to induce cellular transport, stress responses, and cell cycle regulation in human skin fibroblasts (Lee et al., 2016; Yu et al., 2016), as well as inflammatory and fibrotic reactions in the lung in a rat model (Poulsen et al., 2015). Moreover, Takagi et al. (2008) reported that MWCNTs could cause mesothelioma in p53 heterozygous mice, and Poland et al. (2008) found asbestos-like pathological changes in MWCNT-treated C57BL/6 mice. Several studies have investigated the toxicity of CNTs at the early life stage of organisms. Zebrafish specimens treated with functionalized MWCNTs showed potential reproduction toxicity (Cheng and Haitao 2014). SWCNTs at the concentration of more than 100 mg/L can significantly induce ROS and DNA damage in minnow embryos (Zhu et al., 2016). Combined functionalized SWCNTs with Agþ can cause a high mortality rate of sea urchin (Magesky and Pelletier 2015). Further studies are warranted to evaluate the potential ecotoxicity of CNTs. Perfluorooctane sulfonate (PFOS) is used in many applications such as adhesives, propellants, surfactants, retardants, lubricants, and medicines (Renzi et al., 2013). PFOS is of concern because of its

http://dx.doi.org/10.1016/j.envpol.2016.12.055 0269-7491/© 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Wang, S., et al., The presence of MWCNTs reduces developmental toxicity of PFOS in early life stage of zebrafish, Environmental Pollution (2016), http://dx.doi.org/10.1016/j.envpol.2016.12.055

2

S. Wang et al. / Environmental Pollution xxx (2016) 1e9

environmental persistence, toxicity, and non-biodegradation (Mhadhbi et al., 2012). Environmental monitoring studies showed that high concentrations of PFOS (e.g., 1790 ng/g and 2431 ng/g wet weight) existed in many aquatic species owing to bioaccumulation through the food chain (Dorneles et al., 2008; Houde et al., 2008). Therefore, studies on the toxicity of PFOS exposure in various fish species are imperative. PFOS were found to induce cell death, muscle lesions, spinal curvature, and uninflated swim bladder and to reduce hatching rates (Huang et al., 2010). In addition, apoptosis has been observed in primary cultured zebrafish hepatocytes treated with PFOS (Oakes et al., 2005; Huang et al., 2009). Moreover, Krøvel et al. (2008) found dose-dependent expression of genes, including caspase 3B, acyl-CoA oxidase, PPARa, PPARb, and PPARg, in the hepatocytes of Atlantic salmon exposed to PFOS at the concentration from 2.1 to 25.0 mg/L; the exposure also enhanced cellular stress. Contaminants in aquatic systems always exist in a complex mixture. Investigating the combined toxicity of these contaminants in an aquatic system is very important for the assessment of ecological risk (Wang et al., 2013; Alarifi et al., 2015). Song et al. (2014) investigated the combined effects of carboxylfunctionalized single-walled carbon nanotubes (cf-SWCNTs) and 17a-ethinylestradiol (EE2) in the human breast adenocarcinoma cell line (MCF-7 cells). No significant differences were found in mitochondrial activity, membrane damage, and cell apoptosis when the cells were exposed to cf-SWCNTs with and without the adsorbed EE2. However, the bioactivity of adsorbed EE2 on cfSWCNTs was significantly inhibited. Thus, EE2 can be adsorbed on cf-SWCNTs in environment-relevant settings and influence SWCNT toxicity and biological fate. Yan et al. (2016) evaluated the combined effects of MWCNTs and two functionalized MWCNTsdCOOH-MWCNTs (0.5 mg/L) and OH-MWCNTs (0.5 mg/ L)din the presence of zinc (Zn, 0.1 mg/L) on the antioxidant status and histopathological changes in Carassius auratus. The relative toxicity was in the following order: MWCNTs þ Zn < COOHMWCNTs þ Zn < OH-MWCNTs þ Zn. Zhang et al. (2014) performed a two-day co-exposure of zebrafish to MWCNTs ((400 mg/L) and sodium pentachlorophenate (PCP-Na, 22.26 mg/L and 37.1 mg/L) and found that MWCNTs could alleviate the toxicity induced by PCP-Na by changing the activity of antioxidant enzymes, including superoxide dismutase (SOD), catalase (CAT), and glutathione peroxidase (GSH-Px), of Eisenia fetida. In addition, Cd toxicity could be increased when Daphnia magna were exposed to the combination of SWCNTs and Cd, which indicated that SWCNTs induced a synergistic toxic effect on the survival of D. magna (Revel et al., 2015). Aquatic organisms are known to be easily affected by exogenous pollutants. A series of biomarkers is used to assess the impact of these pollutants. In general, the generation and scavenging of reactive oxygen species (ROS) in cells remains in a dynamic balance. When pollutants enter the body, they can produce ROS via a series of metabolic conversions. If they are not cleared promptly, the balance will be destroyed and cause cell injury or death, lipid peroxidation (LPO), and membrane damage (Nel et al., 2006). In fact, in organisms, some antioxidant enzymes such as SOD, CAT, and GSH-Px can convert O2 to H2O2 and then to H2O and O2 in order to minimize the deleterious effects of excessive ROS (Valavanidis et al., 2006). In addition to the changes in antioxidant levels, malondialdehyde (MDA), a product of LPO, also indirectly reflects the generation of ROS. Oxidative stress is thought to be the most acceptable toxic mechanism of CNTs (Srivastava et al., 2011; Saria et al., 2014). CNTs could activate specific molecular signals associated with oxidative stress, leading to the release of cytokines with the depletion of antioxidant defenses (Pacurari et al., 2008). How CNTs and PFOS

affect ROS generation and oxidative stress is not yet known. For example, carboxyl-SWCNTs significantly promoted the generation of ROS and enhanced SOD level at concentrations of 0.01e10 mg/L (Hu et al., 2015). The MWCNTs significantly increased the generation of ROS, LPO, and SOD and decreased glutathione level, as well as could induce caspase 3 activity and DNA strand breakage at a concentration of 300 mg/L, indicating that MWCNTs induced cytotoxicity and apoptosis in L929 cells via oxidative stress (Alarifi and Ali, 2015). In addition, the maternal exposure to PFOS at 2.0 mg/(kg$d) caused severe histopathological changes along with marked oxidative injuries and cell apoptosis in offspring lungs (Chen et al., 2012). Shi and Zhou (2010) found that exposure to PFOS at 1.0 mg/L significantly increased MDA production in zebrafish larvae, and nuclear factor erythroid-2 related factor 2 production was a protective mechanism against the PFOS-induced oxidative stress in zebrafish larvae. Some studies evaluated the effects of MWCNTs on the toxicity of the coexisting pollutants. However, some controversial results have been reported, exhibiting either increased or reduced toxicity. For example, Kim et al. (2014) found that both MWCNTs and SWCNTs increased the toxicity of Cu to aquatic organisms, because SWCNTs enhanced Cu accumulation in the animals. In contrast, Wang et al. (2012) found that functionalized multi-walled carbon nanotubes (fMWCNTs) could not enhance the cellular toxicity or DNA damage produced by cigarette smoke solution (CSS) on 16-HBE cells, which indicated no combined cytotoxic influence between f-MWCNTs and CSS. Thus, further research is necessary to evaluate the potential combined toxicity of MWCNTs and the coexisting pollutants in aquatic organisms. As mentioned above, both MWCNTs and PFOS have been detected in the aquatic environment. MWCNTs and PFOS coexist because of their increased release into the environment. In aqueous solution, the sulfonic group of PFOS can be ionized into anions, which can be adsorbed by positively charged MWCNTs. However, few studies investigated the combined toxicity of PFOS with MWCNTs in water. In this study, we used zebrafish embryos and larvae as test models to evaluate the developmental toxicity of PFOS and MWCNTs on parameters, including 48-h heartbeat, 72-h hatching rate, 96-h mortality, and 96-h malformation. In addition, we determined the level of ROS and MDA and the activity of SOD, CAT, and GSH-Px to understand deeply the oxidative stress as well as LPO induced by PFOS treatment in the presence of MWCNTs. This study can facilitate further analysis on the potential effects of other coexisting pollutants on aquatic organisms. 2. Materials and methods 2.1. Chemicals and preparation of stock solutions PFOS (CAS no., 2795-39-3, 98% purity) was purchased from Sigma (St. Louis, MO, USA). The stock solution (2.0  104 mg/L) of MWCNT water dispersion was purchased from the Chinese Academy of Sciences (Chengdu, China). The MWCNT properties were as follows: purity >95% wt%, OD approximately 50 nm, and length 10e20 mm. MWCNTs and PFOS were placed in the test vessels (no carrier solvent). Stock solution of PFOS (1000 mg/L) was prepared by dissolving 1000 mg PFOS in 1 L of deionized water by sonication for 30 min (Ultra sonic processor VCX-750W). All other chemicals and solvents were of analytical grade. 2.2. MWCNT characterization and PFOS analysis A series of MWCNT suspensions (20, 50, 80, 100, and 200 mg/L) was prepared for exposure by using zebrafish culture medium by using stepwise dilution. Some amount of MWCNT solution was

Please cite this article in press as: Wang, S., et al., The presence of MWCNTs reduces developmental toxicity of PFOS in early life stage of zebrafish, Environmental Pollution (2016), http://dx.doi.org/10.1016/j.envpol.2016.12.055

S. Wang et al. / Environmental Pollution xxx (2016) 1e9

dropped into a carbon-coated copper grid, air dried, and observed using transmission electron microscopy (TEM; JEOL, Tokyo, Japan). The TEM images showed that MWCNTs consisted of closed-edged structures known as nanotube tips. Each piece of MWCNTs showed a high degree of contrast between the bright central part and the dark edge, indicating the hollow character of the tube (Fig. 1).

2.3. Exposure concentration of MWCNTs and PFOS Preliminary experiments were conducted to confirm the appropriate exposure concentrations. In the preliminary experiments, embryos were exposed to different concentrations of PFOS (1.0, 2.0, 4.0, 8.0, and 16.0 mg/L). The lethal concentration of 96-h LC50 (3.502 mg/L) was obtained. We selected approximately 1/16, 1/8, 1/4, and 1/2 of the 96-h LC50 as PFOS exposure concentrations, that is, 0.2, 0.4, 0.8, and 1.6 mg/L. MWCNT exposure concentration was confirmed by exposing adult zebrafish to 10, 20, 50, 100, and 200 mg/L MWCNTs for 96 h in triplicate, and malformations such as spinal curvature, tail deformity, and uninflated swim bladder were observed at the concentration above 50 mg/L. Therefore, 50.0 mg/L was confirmed as the MWCNT exposure concentration. Accordingly, the concentrations of co-exposure of PFOS and MWCNTs were as follows: 0.2 þ 50.0, 0.4 þ 50.0, 0.8 þ 50.0, and 1.6 þ 50.0 mg/L (PFOS þ MWCNTs), as shown in Table 1. As mentioned above, the environment-relevant concentrations of PFOS and MWCNTs are under 1 mg/L level. The exposure duration could not be very long because the test model zebrafish were in the early life stage (embryos and larvae). Therefore, the experimental concentrations of MWCNTs and PFOS were higher than those of the environment-relevant concentrations. PFOS concentrations were determined using liquid chromatography-tandem mass spectrometry (LC-MS/MS). The actual concentrations of PFOS in the series of the exposure suspensions (0.2, 0.4, 0.8, and 1.6 mg/L) were confirmed to be 0.17, 0.33, 0.64, and 1.23 mg/L, respectively, and the actual concentrations of PFOS in the mixtures of the exposure suspensions (0.2 þ 50,

3

Table 1 Design of concentrations of the single- and co-exposures. Substances

Concentrations (mg/L)

Single-exposure of MWCNTs Single-exposures of PFOS Co-exposures of PFOS þ MWCNTs

50 0.2 0.2 þ 50

0.4 0.4 þ 50

0.8 0.8 þ 50

1.6 1.6 þ 50

0.4 þ 50, 0.8 þ 50, and 1.6 þ 50 mg/L) were 0.18, 0.31, 0.54, and 1.19 mg/L, respectively, indicating that the changes in concentrations were negligible over the exposure period. Owing to the marginal differences between nominal and measured concentrations of PFOS, the nominal concentrations were used in the reporting and explanation of the data for convenience in this study. 2.4. Zebrafish maintenance and toxicity test of embryos and larvae Adult zebrafish (Danio rerio) were purchased from the State Key Laboratory of Freshwater Ecology and Biotechnology, Chinese Academy of Sciences (Wuhan, China). The zebrafish were fed three times daily with freshly hatched Artemia nauplii and frozen E. fetida and maintained at 28 ± 1  C under a photoperiod of 14 h light: 10 h dark. The 4-month adult zebrafish were placed in pairs in spawning boxes overnight and cultural medium (DO, >6.3 mg/L, pH 7.2) and maintained at approximately 28  C. Spawning was induced by stimulation with light on the following morning. Fertilized eggs were collected, and the intact fertilized eggs were used for subsequent experiments. Approximately 400 normal embryos were randomly distributed into each glass vessel containing 500 mL solution of MWCNTs þ PFOS (50 þ 0.2, 50 þ 0.4, 50 þ 0.8, and 50 þ 1.6 mg/L), PFOS alone (0.2, 0.4, 0.8, and 1.6 mg/L), and MWCNT alone (50 mg/ L), respectively. Three replicates were used for each concentration, and the exposure solution was renewed daily. The samples for gene and biochemical analyses were collected and immediately stored at 80  C. 2.5. Embryos and larvae bioassays The embryos were exposed to the test solution in sterile 24-well plates, with three embryos in one well, and each test well (20 total) contained 2 mL test solution, and each control well (4 total) contained 2 mL of culture medium. Three parallel treatments were designed for each concentration. The plates were placed in an illumination incubator and covered with sealing films to prevent evaporation. The test solutions were renewed daily. The conventional toxicology endpoints of 24, 48, 72, and 96 h were observed using a stereo microscope (Olympus, Japan). Mortality rate, hatching rate, malformation rate, heart rate, and body length were recorded. For the relevant methodologies, refer to a previous study (Du et al., 2016). In all, 200 normal embryos were selected randomly and placed in a glass beaker containing 500 mL test solution. Fresh solutions were replaced with 50% of the volume every 24 h. After 96-h exposure, the embryos were stored at 80  C for biochemical analysis. 2.6. Measurement of ROS and MDA

Fig. 1. TEM image of MWCNTs suspension at the concentration of 50 mg/L.

The generation of ROS in larvae exposed to the solutions for 96 h post fertilization (hpf) was measured using dichlorofluoresceindiacetate (DCFH-DA). In all, 10 exposed larvae were washed three times with phosphate buffered solution, (PBS, pH 7.4) and then homogenized in buffer. The homogenate was centrifuged, and the

Please cite this article in press as: Wang, S., et al., The presence of MWCNTs reduces developmental toxicity of PFOS in early life stage of zebrafish, Environmental Pollution (2016), http://dx.doi.org/10.1016/j.envpol.2016.12.055

4

S. Wang et al. / Environmental Pollution xxx (2016) 1e9

supernatant was collected. Next, 150 mL of the supernatant and 10 mL of DCFH-DA stock solution (dissolved in DMSO, 0.1 mmol/L) were added to a 96-well plate and incubated at room temperature for 5 min. The plate was incubated at 37  C for 30 min in the dark. The fluorescence intensity was measured using a microplate reader (Molecular Device; M2, Union City, CA, USA) with excitation and emission at 485 and 530 nm, respectively. The total protein concentration was determined using protein assay kits (Jiancheng Biochemistry Co., Nanjing, China). The ROS concentration was expressed in arbitrary units. As a product of LPO, MDA was analyzed by referring to manufacturers’ protocols of commercial kits (Jiancheng Bioengineering Institute, Nanjing, China). MDA was condensed with thiobarbituric acid (TBA) to form a red product, which could be measured at a wavelength of 532 nm and, accordingly, the quantity of MDA was determined. 2.7. Measurement of the activities of SOD, CAT, and GSH-Px After exposure, the larvae were homogenized in buffer (pH 7.4) containing 0.01 M Tris-HCl, 0.0001 M EDTA-2Na, 0.01 M sucrose, and 0.8% NaCl by using a tissue homogenizer. The homogenate (1:10 w/v) was centrifuged at 2000 g for 15 min at 4  C. After centrifugation, the supernatant was stored at 4  C for the determination of enzyme activity. The activities of SOD, CAT, and GSH-Px were measured following the manufacturers’ protocols for commercial kits (Jiancheng Bioengineering Institute, Nanjing, China). SOD activity was measured based on the superoxide radical-dependent cytochrome C reduction method by recording the absorbance at 550 nm. The assay for CAT was based on its ability to scavenge H2O2, by measuring the absorbance of the generated stable chromophore at 405 nm. GSH-Px activity was assayed following the rate of NADPH oxidation at 340 nm by using the coupled reaction with glutathione reductase. The homogenate was collected for the analysis of SOD, CAT, and GSH-Px activities. Protein content in the samples was measured using Coomassie blue protein-binding method by using bovine serum albumin as the standard. All procedures completely complied with the manufacturer instructions. 2.8. Statistical analysis All the experiments were performed in triplicate, and data are shown as mean ± S.D. based on three separate experiments. Statistical analyses were performed using one-way analysis of variance. Dunnett's test was used to determine the treatments that were significantly different from the control. The level for statistical significance was set at p < 0.05. 3. Results 3.1. Acute toxicity of PFOS in the presence of MWCNTs to zebrafish embryos and larvae The heart rate of zebrafish embryos was measured after exposure to different concentrations of PFOS, MWCNTs, and PFOS þ MWCNTs. The heart rate decreased with increasing concentrations of PFOS in both single- and co-treatment groups (Fig. 2A). Further, the heart rates were significantly inhibited by 0.4, 0.8, and 1.6 mg/L PFOS (p < 0.05) compared with that of the control (124.3 ± 4.1 beats/min). However, heart rate increased in almost each co-treatment group unlike in the PFOS single-treatment group. The difference was statistically significant (p < 0.05) between the co-exposure (1.6 mg/L þ 50 mg/L; 109.7 ± 2.5 beats/min) and PFOS exposure (1.6 mg/L; 101.5 ± 2.2 beats/min). In addition,

single exposure to 50 mg/L MWCNTs (123.4 ± 3.4 beats/min) caused no significant difference in heart rate compared to that of the control (p > 0.05) The hatching rate of zebrafish was measured after exposure to different concentrations of PFOS and PFOS þ MWCNTs and MWCNTs at 50 mg/L. The hatching rate was not affected significantly in both PFOS treatment and co-treatment groups at low exposure concentrations (<0.4 mg/L; Fig. 2B). With an increase in PFOS concentrations, the hatching rate decreased in the PFOS treatment and co-treatment groups compared with that of the control (72.7± 1.2%). However, the hatching rate increased for coexposures at higher concentrations of PFOS unlike in the PFOS exposures alone. The difference was significant (p < 0.05) for coexposure (1.6 mg/L PFOS þ 50 mg/L MWCNTs, 63.8± 1.5%)) compared with that of the 1.6 mg/L PFOS exposure (49.4± 6.6%). The body length of zebrafish larvae gradually decreased with increasing PFOS concentrations (Fig. 2C). Compared with that of the control (3.7 ± 0.03 mm), the body length decreased significantly (p < 0.05) for the PFOS exposures at the concentrations of 0.4, 0.8, and 1.6 mg/L. Similarly, some adverse effects were observed in the co-treatment groups. However, the body length in the co-treatment groups was higher than those of the PFOS single-treatment groups at PFOS concentrations of 0.4, 0.6, 0.8, and 1.6 mg/L. In addition, MWCNT exposure (50 mg/L) could affect the body length of zebrafish larvae (3.6 ± 0.07 mm) compared with that of the control (p < 0.05). The mortality rate of zebrafish larvae generally increased with increasing PFOS concentrations following certain dose-response relationships in both the PFOS treatment and co-treatment groups (Fig. 2D). For PFOS exposures, the increase in mortality was statistically significant (p < 0.05) at concentrations of 0.4, 0.8, and 1.6 mg/L compared with that of the control (26.1± 2.5%). In the presence of MWCNTs, the mortality was lower than that in almost each PFOS single-treatment group. The difference was statistically significant (p < 0.05) at higher concentrations (0.8 and 0.6 mg/L); the mortality decreased from 38.8± 3.2% and 45.5± 3.2% to 31.6± 2.5% and 33.8± 2.0%, respectively. Malformation rate of zebrafish larvae was investigated after exposure to different concentrations of PFOS and PFOS þ MWCNTs (Fig. 2E). Further, the malformation rates of larvae in PFOS singletreatment and co-treatment groups were increased with increasing PFOS concentrations. In addition, significant differences (p < 0.05) in malformation rates were found between 0.8 mg/L PFOS exposure (61.0 ± 5.4%) and the control (17.3 ± 2.7%). At the exposure to PFOS at the concentrations of 0.4, 0.8 and 1.6 mg/L, the malformation rates of PFOS exposures were higher than those of co-exposures. The morphological changes in zebrafish larvae were investigated by exposing the embryos to PFOS and mixture of PFOS and MWCNTs (0.8 mg/L þ 50 mg/L) from 3 hpf to 96 hpf. A series of abnormalities was observed, including pericardial edema, bent spine, uninflated swim bladder, spinal curvature, and bent tail.

3.2. Effects of PFOS in the presence of MWCNTs on ROS generation and MDA level The effect of PFOS and MWCNTs on ROS production and MDA level are shown in Fig. 4. The levels of ROS and MDA were significantly increased after all PFOS exposures compared with that of the control (p < 0.05). In addition, the levels of ROS and MDA after PFOS exposures were higher than those of almost co-exposures, suggesting that MWCNTs could reduce the levels of ROS and MDA induced by PFOS in a time- and dose-dependent manner.

Please cite this article in press as: Wang, S., et al., The presence of MWCNTs reduces developmental toxicity of PFOS in early life stage of zebrafish, Environmental Pollution (2016), http://dx.doi.org/10.1016/j.envpol.2016.12.055

Fig. 2. Acute toxicity of PFOS combined with MWCNTs to zebrafish embryos and larvae. (A) heart rate of larvae at 48 hpf; (B) hatch rate of embryos at 72 h, (C) body length of larvae at 96 hpf; (D) mortality of larvae at 96 hpf; (E) malformation rate of larvae at 96 hpf. Initial normal embryos number: approximately 400; Constant MWCNTs concentration: 50 mg/L “PFOS 0” (black bar) in X axis represents MWCNTs alone; “*” indicates a statistically significant difference in comparison between PFOS-treatments and control at p < 0.05 level. “#” indicates a statistically significant difference in comparison between PFOS-treatments and co-treatments at p < 0.05 level. Values represent the means ± standard error of three replicates.

Please cite this article in press as: Wang, S., et al., The presence of MWCNTs reduces developmental toxicity of PFOS in early life stage of zebrafish, Environmental Pollution (2016), http://dx.doi.org/10.1016/j.envpol.2016.12.055

6

S. Wang et al. / Environmental Pollution xxx (2016) 1e9

Fig. 3. Malformation images of zebrafish larvae treated with PFOS combined with MWCNTs at 96 hpf. SC: spinal curvature; TD: tail deformity; UBD: uninflated swim bladder; PE: pericardial edema. Initial normal embryos number: approximately 400. PFOS and MWCNTs concentrations: 0.8 m/L and 50 mg/L, respectively.

3.3. Effects of PFOS combined with MWCNTs on CAT, SOD, and GSHPx Antioxidant defenses consisted of different enzymes, including SOD, CAT, and GSH-Px. The activities of antioxidant enzymes (SOD, CAT, and GSH-Px) in zebrafish larvae after exposure to PFOS, MWCNTs, and mixture of PFOS and MWCNTs are shown in Fig. 5. For both PFOS exposure and co-exposures (Fig. 5A), a remarkable increase in SOD activity was observed with an increase in PFOS concentrations. However, the SOD activity at each co-exposure was lower than that of MWCNTs exposure and each PFOS exposure

caused significantly higher SOD activity (p < 0.05, except for 0.2 mg/L). Changes in CAT activity are shown in Fig. 5B. CAT level increased with an increase in PFOS concentrations for single-as well as co-exposures except for 0.2 mg/L. In contrast, the CAT level of each co-exposure was significantly lower than that of MWCNTs exposure and that of each PFOS exposure (p < 0.05, except for 0.2 mg/L). Similarly, GSH-Px activity was increased with increasing concentration of PFOS for both single- and co-exposures (Fig. 5C), but GSH-Px activity of most PFOS concentrations except for 0 mg/L PFOS were higher than that of the relevant co-exposure.

Please cite this article in press as: Wang, S., et al., The presence of MWCNTs reduces developmental toxicity of PFOS in early life stage of zebrafish, Environmental Pollution (2016), http://dx.doi.org/10.1016/j.envpol.2016.12.055

S. Wang et al. / Environmental Pollution xxx (2016) 1e9

7

Fig. 4. Effects of PFOS on ROS generation and MDA level in zebrafish larvae at 96 hpf in the presence of MWCNTs. (A) ROS; (B) MDA. Constant MWCNTs concentration: 50 mg/L “0” (black bar) in X axis represents MWCNTs alone. “*” indicates a statistically significant difference in comparison between PFOS-treatments and control at p < 0.05 level. “#” indicates a statistically significant difference in comparison between PFOS-treatments and co-treatments at p < 0.05 level. Values represent the means ± standard error of three replicates.

Fig. 5. Effect of PFOS on the activities of antioxidant enzymes in zebrafish larvae at 96 hpf in the presence of MWCNTs. A (SOD), B (CAT), C (GSH-Px). Constant MWCNTs concentration: 50 mg/L “0” (black bar) in X axis represents MWCNTs alone. * Indicate a statistically significant difference in comparison between PFOS single-treatment and control at p < 0.05 level. # Indicate a statistically significant difference in comparison between single- and co-treatments at p < 0.05 level. Values represent the means ± standard error of three replicates.

4. Discussion To our knowledge, there are no reports regarding the combined toxicity effects of PFOS and MWCNTs. In the present study,

zebrafish embryos and larvae were used to assess the combined toxicity effects of PFOS and MWCNTs. MWCNTs were found to reduce the toxic effects of PFOS on zebrafish during the exposure period.

Please cite this article in press as: Wang, S., et al., The presence of MWCNTs reduces developmental toxicity of PFOS in early life stage of zebrafish, Environmental Pollution (2016), http://dx.doi.org/10.1016/j.envpol.2016.12.055

8

S. Wang et al. / Environmental Pollution xxx (2016) 1e9

PFOS is a medium toxic substance according to the World Health Organization (WHO) classification criteria for acute toxicity of compounds. The overall results of acute toxicity such as heart rate, hatching rate, mortality, and malformation are shown in Fig. 2. After PFOS exposures, a significant reduction in heart rate, delayed hatching, and notable increase in mortality and malformation were observed in a dose-dependent manner. Some of our results were consistent with those of Huang et al. (2010). After MWCNT exposure, a slight acute effect was found on the heart rate, hatching rate, mortality, and malformation compared with that in the control. However, for the co-exposures (Fig. 4), the presence of MWCNTs could reduce the toxicity of PFOS. For example, the heart rate and hatchability were increased, while mortality and malformation rates were decreased compared with those of PFOS exposures. The numbers of lesions induced by the mixture were considerably lower than those induced by PFOS alone. Further one-way ANOVA of the combined toxicity data showed a statistical interaction between PFOS and MWCNTs (Fig. 4). These findings indicate that the combined toxicity effects of PFOS with MWCNTs on zebrafish embryogenesis were induced via antagonism. This antagonistic mechanism was similar with that of the combined toxicity of MWCNTs and PCP-Na (Zhang et al., 2014). Oxidative stress can occur in biological systems when they are exposed to toxic substances; it can be defined as an imbalance between the production and depletion of ROS. Oxidative stress in fish is induced by aquatic contaminants that cause the production of excess ROS. Next, we attempted to determine the potential mechanisms for the combined toxicity of PFOS with MWCNTs. Antioxidant defense mechanisms in organisms can effectively protect against damage induced by contaminants. However, one of the important reasons for the generation of oxidative stress and disturbance of the antioxidant defense system is the release of ROS. For instance, Pulskamp et al. (2007) found that exposure to CNTs could increase ROS generation and antioxidant defense response in different types of cells. Qian et al. (2010) reported that oxidative stress is one of the important factors responsible for the developmental toxicity caused by PFOS exposure. The toxicities of PFOS and perfluorooctanoic acid (PFOA) to Escherichia coli were thought to be attributed to oxidative stress and DNA damage (Liu et al., 2016). In the present study, a significant increase in the generation of ROS and the lipid peroxidative product MDA was found (Fig. 4 B). However, their levels after co-exposures with PFOS combined with MWCNTs were lower than those after exposure to PFOS alone. In addition, the oxidative stress induced by MWCNT (50 mg/L) exposure was very mild and could not be detected. These results indicate that MWCNTs reduced the PFOS-induced ROS and LPO levels in zebrafish larvae. SOD is generally considered as a primary line of defense against tissue and cellular damage caused by ROS. CAT and GSH-Px provide a secondary line of defense by decomposing peroxide to water and molecular oxygen (Li et al., 2008). In this study, SOD activity was significantly increased with an increase in PFOS concentrations (Fig. 5). The elevated SOD activity could be considered as a direct response to the elevation of superoxide anion radicals. Moreover, CAT and GSH-Px activities, as well as LPO levels, were significantly increased after a 96-hpf exposure, suggesting that the antioxidant defense response was inadequate (Bouraoui et al., 2009). In contrast, the antioxidant enzyme levels induced by PFOS were reduced in the presence of MWCNTs at each exposure concentration of PFOS (0.2, 0.4, 0.8, and 1.6 mg/L) in a time- and dosedependent manner (Fig. 5). These results are concomitant with the findings of Petersen et al. (2009). MWCNTs are good adsorbent for many kinds of organic and inorganic substances due to the high surface area. For example, MWCNTs have high adsorption capacity for one kind of anionic

surfactant, sodium dodecyl benzene sulfonate (SDBS) in industrial wastewater (Mortazavi and Farmany 2016). In summary, the experimental results indicated that MWCNTs could alleviate the toxic effect induced by PFOS in zebrafish embryos and larvae by possibly causing the adsorption of PFOS to MWCNTs. Future studies are warranted to determine the toxicity of PFOS or MWCNTs alone and to identify the target organs of transport and metabolism when they co-exist in the environment. This might allow understanding the ecological risks associated with the co-existence of pollutants. 5. Conclusions This study aimed to investigate the developmental toxicity induced by PFOS in the presence of MWCNTs in the early life stage of zebrafish. The adverse effects of PFOS on the hatching rate (72 hpf) of zebrafish embryos and heart rate (48 hpf), body length (96 hpf), and mortality (96 hpf) and malformation (96 hpf) rates of zebrafish larvae were alleviated in a dose-dependent manner in the presence of MWCNTs. After zebrafish larvae were exposed to PFOS, the activities of SOD, CAT, and GSH-Px, as well as the levels of ROS and MDA, decreased in the presence of MWCNTs, suggesting that oxidative stress and lipid peroxidation were relieved by MWCNTs. Thus, the presence of MWCNTs might reduce the developmental toxicity induced by PFOS in the early life stage of zebrafish. Acknowledgements The Project was sponsored by the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry, China. This work was also supported by the Key Laboratory of Urban Water Resource and Environment of Harbin institute of technology, China (ES201608). References Alarifi, S., Ali, D., 2015. Mechanisms of multi-walled carbon nanotubes-induced oxidative stress and genotoxicity in mouse fibroblast cells. Int. J. Toxicol. 34 (3), 258e265. Boncel, S., Kyziol-Komosinska, J., Krzyewska, I., 2015. Interactions of carbon nanotubes with aqueous/aquatic media containing organic/inorganic contaminants and selected organisms of aquatic ecosystems-A review. Chemosphere 136, 211e221. Bouraoui, Z., Banni, M., Ghedira, J., Clerandeau, C., Narbonne, J.F., Boussetta, H., 2009. Evaluation of enzymatic biomarkers and lipid peroxidation level in Hediste diversicolor exposed to copper and benzo[a]pyrene. Ecotoxicol. Environ. Saf. 72, 1893e1898. Chen, T., Zhang, L., Yue, J.Q., 2012. Prenatal PFOS exposure induces oxidative stress and apoptosis in the lung of rat off-spring. Reprod. Toxicol. 33, 538e545. Cheng, R.W., Haitao, L., et al., 2014. Carboxylated multi-walled carbon nanotubes aggravated biochemical and subcellular damages in leaves of broad bean (Vicia faba L.) seedlings under combined stress of lead and cadmium. J. Hazard. Mat. 274, 404e412. Dorneles, P.R., Lailson-Brito, J., Azevedo, A.F., 2008. High accumulation of perfluorooctane sulfonate (PFOS) in marine tucuxi dolphins (Sotalia guianensis) from the Brazilian coast. Environ. Sci. Technol. 42, 5368e5373. Du, J., Wang, S.T., You, H., et al., 2016. Developmental toxicity and DNA damage to zebrafish induced by perfluorooctane sulfonate in the presence of ZnO nanoparticles. Environ. Toxicol. 31 (3), 360e371. Ema, M., Gamo, M., Honda, K., 2016. A review of toxicity studies of single-walled carbon nanotubes in laboratory animals. Regul. Toxicol. Pharmacol. 74, 42e63. Houde, M., Czub, G., Small, J.M., 2008. Fractionation and bioaccumulation of perfluorooctane sulfonate (PFOS) isomers in a lake Ontario food web. Environ. Sci. Technol. 42, 9397e9403. Hu, X.G., Ouyang, S.H., Mu, L., et al., 2015. Effects of graphene oxide and oxidized carbon nanotubes on the cellular division, microstructure, uptake, oxidative stress, and metabolic profiles. Environ. Sci. Technol. 10825e10833. Huang, T.S., Olsvik, P.A., Krovel, A., Tung, H.S., Torstensen, B.E., 2009. Stress induced expression of protein disulfide isomerase associated 3 (PDIA3) in Atlantic salmon (Salmo salar L.). Comp. Biochem. Physiol. B 154 (4), 435e442. Huang, H., Huang, C., Wang, L., Ye, X., Bai, C., Simonich, M.T., Tanguay, R.L., Dong, Q., 2010. Toxicity, uptake kinetics and behavior assessment in zebrafish embryos following exposure to perfluorooctane sulphonicacid (PFOS). Aquat. Toxicol. 98, 139e147. Kim, B.M., Rhee, J.S., Jeong, C.B., Seo, J.S., Park, G.S., Lee, Y.M., Lee, J.S., 2014. Heavy

Please cite this article in press as: Wang, S., et al., The presence of MWCNTs reduces developmental toxicity of PFOS in early life stage of zebrafish, Environmental Pollution (2016), http://dx.doi.org/10.1016/j.envpol.2016.12.055

S. Wang et al. / Environmental Pollution xxx (2016) 1e9 metals induce oxidative stress and trigger oxidative stress-mediated heat shock protein (hsp) modulation in the intertidal copepod Tigriopus japonicas. Comp. Biochem. Phys. C 166, 65e74. Krøvel, A.V., Søfteland, L., Torstensen, B., Olsvik, P.A., 2008. Transcriptional effects of PFOS in isolated hepatocytes from Atlantic salmon Salmo salar L. Comp. Biochem. Physiol. C 148, 14e22. Lee, W.J., Kang, H.M., Won, E.J., et al., 2016. Multi-walled carbon nanotubes (MWCNTs) lead to growth retardation, antioxidant depletion, and activation of the ERK signaling pathway but decrease copper bioavailability in the monogonont rotifer (Brachionus koreanus). Aquat. Toxicol. 172, 67e79. Li, Y.C., Bao, Y.M., Jiang, B., Wang, Z., Liu, Y.X., Zhang, C., An, L.J., 2008. Catalpol protects primary cultured astrocytes from in vitro ischemia-induced damage. Int. J. Dev. Neurosci. 26, 309e317. Liu, G.H., Zhang, S., Yang, K., Zhu, L.Z., Lin, D.H., 2016. Toxicity of perfluorooctane sulfonate and perfluorooctanoic acid to Escherichia coli: membrane disruption, oxidative stress, and DNA damage induced cell inactivation and/or death. Environ. Poll. 214, 806e815.  2015. Toxicity mechanisms of ionic silver and polymerMagesky, A., Pelletier, E., coated silver nanoparticles with interactions of functionalized carbon nanotubes on early development stages of sea urchin. Aquat. Toxicol. 167, 106e123. rez, S., Beiras, R., 2012. Ecological risk assessment of perMhadhbi, L., Rial, D., Pe fluorooctanoic acid (PFOA) and perfluorooctanesulfonic acid (PFOS) in marine environment using Isochrysis galbana, Paracentrotus lividus, Siriella armata and Psetta maxima. J. Envion. Monit. 14, 1375e1382. Mortazavi, S.S., Farmany, A., 2016. High adsorption capacity of MWCNTs for removal of anionic surfactant SDBS from aqueous solutions. J. Water Supply Res. Technol. 65 (1), 37e42. Nel, A., Xia, T., Madler, L., Li, N., 2006. Toxic potential of materials at the nanolevel. Science 311, 622e627. Oakes, K.D., Sibley, P.K., Martin, J.W., MacLean, D.D., Solomon, K.R., Mabury, S.A., Van Der Kraak, G.J., 2005. Short-term exposure of fish to perfluorooctane sulfonate: acute effects on fatty acyl-coa oxidase activity, oxidative stress, and circulating sex steroids. Environ. Toxicol. Chem. 24 (5), 1172e1181. Pacurari, M., Yin, X.J., Zhao, J., et al., 2008. Raw single-wall carbon nanotubes induce oxidative stress and activate MAPKs, AP-1, NF-kappaB, and Akt in normal and malignant human mesothelial cells. Environ. Health Perspect. 116, 1211e1217. Petersen, E.J., Pinto, R.A., Landrum, P.F., Weber, W.J., 2009. Influence of carbon nanotubes on pyrene bio-accumulation from contaminated soils by earthworms. Environ. Sci. Technol. 43, 4181e4187. Poland, C.A., Duffin, R., Kinloch, I., Maynard, A., Wallace, W.A.H., Seaton, A., Stone, V., Brown, S., MacNee, W., Donaldson, K., 2008. Carbon nanotubes introduced into the abdominal cavity of mice show asbestos-like pathogenicity in a pilot study. Nat. Nanotechnol. 3, 423e428. Poulsen, S.S., Saber, A.T., Williams, A., 2015. MWCNTs of different physicochemical properties cause similar inflammatory responses, but differences in transcriptional and histological markers of fibrosis in mouse lungs. Toxicol. Appl. Pharmacol. 284, 16e32. Pulskamp, K., Diabate, S., Krug, H.F., 2007. Carbon nanotubes show no sign of acute toxicity but induce intracellular reactive oxygen species in dependence on

9

contaminants. Toxicol. Lett. 168, 58e74. Qian, Y., Ducatman, A., Ward, R., Leonard, S., Bukowski, V., Lan Guo, N., Shi, X., Vallyathan, V., Castranova, V., 2010. Perfluorooctane sulfonate (PFOS) inducesreactive oxygen species (ROS) production in human microvascular endothelialcells: role in endothelial permeability. J. Toxicol. Environ. Health A 73, 819e836. Renzi, M., Guerranti, C., Giovani, A., Perra, G., Focardi, S.E., 2013. Perfluorinated compounds: levels, trophic web enrichments and human dietary intakes in transitional water ecosystems. Mar. Pollut. Bull. 76, 146e157. Revel, M., Fournier, M., Gagnon, C., et al., 2015. Toxic effect of single walled carbon nanotubes combined with cadmium to the crustacean Daphnia magna. Int. Lett. Nat. Sci. 42, 20e61. Saria, R., Mouchet, F., Perrault, A., et al., 2014. Short term exposure to multi-walled carbon nanotubes induce oxidative stress and DNA damage in Xenopus laevis tadpoles. Ecotoxicol. Environ. Saf. 107, 22e26. Shi, X.G., Zhou, B.S., 2010. The role of Nrf2 and MAPK pathways in PFOS-induced oxidative stress in zebrafish embryos. Toxicol. Sci. 115, 391e400. Song, M.Y., Wang, F.B., Zeng, L.Z., Yin, J.F., Wang, H.L., Jiang, G.B., 2014. Co-exposure of carboxyl-functionalized single-walled carbon nanotubes and 17a-ethinylestradiol in cultured cells: effects on bioactivity and cytotoxicity. Environ. Sci. Technol. 2014 (48), 13978e13984. Srivastava, R.K., Pant, A.B., Kashyap, M.P., 2011. Multi-walled carbon nanotubes induce oxidative stress and apoptosis in human lung cancer cell line-A549. Nanotoxicology 5 (2), 195e207. Takagi, A., Hirose, A., Nishimura, T., Fukumori, N., Ogata, A., Ohashi, N., Kitajima, S., Kanno, J., 2008. Induction of mesothelioma in p53þ/e mouse by intraperitoneal application of multi-wall carbon nanotube. J. Toxicol. Sci. 33, 105e116. Valavanidis, A., Vlahogianni, T., Dassenakis, M., Scoullos, M., 2006. Molecular biomarkers of oxidative stress in aquatic organisms in relation to toxic environmental pollutants. Ecotoxicol. Environ. Saf. 64, 178e189. Wang, X.K., Ye, B., Zhang, Y.Q., et al., 2012. Combined effects between functionalized multi-walled carbon nanotubes and cigarette smoke on human bronchial epithelial cells. Adv. Mat. Res. 486, 394e399. Wang, S.W., Huang, J., Yang, Y., et al., 2013. First report of a chinese PFOS alternative overlooked for 30 years: its toxicity, persistence, and presence in the environment. Environ. Sci. Technol. 47, 10163e10170. Yan, L.Q., Feng, M.B., Liu, J.Q., Wang, L.S., Wang, Z.Y., 2016. Antioxidant defenses and histological changes in Carassius auratus after combined exposure to zinc and three multi-walled carbon nanotubes. Ecotoxicol. Environ. Saf. 125, 61e71. Yu, J., Liu, S., Wu, B., et al., 2016. Comparison of cytotoxicity and inhibition of membrane abc transporters induced by MWCNTs with different length and functional groups. Environ. Sci. Technol. 50, 3985e3994. Zhang, L.J., Hu, C.W., Wang, W.L., Ji, F.N., Cui, Y.B., Li, M., 2014. Acute toxicity of multi-walled carbon nanotubes,sodium pentachlorophenate, and their complex on earth worm Eisenia fetida. Ecotoxicol. Environ. Saf. 103, 29e35. Zhu, B., Liu, G.L., Ling, F., Song, L.S., Wang, G.X., 2015. Development toxicity of functionalized single-walled carbon nanotubes on rare minnow embryos and larvae. Nanotoxicology 9, 579e590.

Please cite this article in press as: Wang, S., et al., The presence of MWCNTs reduces developmental toxicity of PFOS in early life stage of zebrafish, Environmental Pollution (2016), http://dx.doi.org/10.1016/j.envpol.2016.12.055