Diastereoisomer-specific neurotoxicity of hexabromocyclododecane in human SH-SY5Y neuroblastoma cells

Science of the Total Environment 686 (2019) 893–902

Contents lists available at ScienceDirect

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

Diastereoisomer-specific neurotoxicity of hexabromocyclododecane in human SH-SY5Y neuroblastoma cells Xiaoli Shi a,b, Jinmiao Zha a,b, Bei Wen a,⁎, Shuzhen Zhang a,b a b

State Key Laboratory of Environmental Chemistry and Ecotoxicology, Research Center for Eco-Environmental Sciences, Chinese Academy of Sciences, Beijing 100085, China University of Chinese Academy of Sciences, Beijing 100049, China

H I G H L I G H T S

G R A P H I C A L

A B S T R A C T

• The neurotoxicity of α-, β-, and γ- HBCD in SH-SY5Y cells were investigated. • The toxic effects followed the order βHBCD N γ-HBCD N α-HBCD. • Caspase-dependent apoptosis responsible for HBCD diastereomer neurotoxicity. • ROS may be a key factor regulating the neurotoxicity of HBCD diastereoisomers.

a r t i c l e

i n f o

Article history: Received 22 March 2019 Received in revised form 29 May 2019 Accepted 1 June 2019 Available online 3 June 2019 Editor: Jay Gan Keywords: Hexabromocyclododecane diastereoisomers SH-SY5Y cells Neurotoxicity Apoptosis Oxidative stress

⁎ Corresponding author. E-mail address: [email protected] (B. Wen).

https://doi.org/10.1016/j.scitotenv.2019.06.008 0048-9697/© 2019 Published by Elsevier B.V.

a b s t r a c t Hexabromocyclododecane (HBCD) is a widely applied brominated flame retardant (BFR) and is regarded as a persistent organic pollutant. It has been found in human tissues and has the potential to cause neurological disorders. However, our understanding of HBCD neurotoxicity at the diastereoisomer level remains lacking. Here, we investigated the neurotoxicity of three HBCD diastereoisomers, i.e., α-, β-, and γ-HBCD, in SH-SY5Y human neuroblastoma cells. Results showed that the HBCD diastereoisomers decreased cell viability, increased lactate dehydrogenase (LDH) release, and impaired cytoskeleton development. Typical morphological features and apoptosis rates showed that the HBCD diastereoisomers induced SH-SY5Y cell apoptosis. The expression levels of several cell apoptosis-related genes and proteins, including Bax, caspase-3, caspase-9, cytochrome c, Bcl-2, and X-linked inhibitor of apoptosis (XIAP), as well as the cell cycle arrest, DNA damage, adenosine triphosphate (ATP) consumption, reactive oxygen species (ROS) levels, and intracellular calcium ion (Ca2+) levels, were examined. Results showed that the HBCD diastereoisomer neurotoxicity was ranked β-HBCD N γ-HBCD N α-HBCD. The cell apoptosis and caspase expression levels of the three HBCD diastereoisomers followed the same order, suggesting that caspase-dependent apoptosis may be one mechanism responsible for the structure-selective HBCD diastereoisomer neurotoxicity. The levels of intracellular Ca2+ and ROS increased significantly. The ROS levels were ordered β-HBCD N γ-HBCD N α-HBCD, whereas those of intracellular Ca2+ were γ-HBCD N β-HBCD N α-HBCD. Thus, ROS may be a key factor regulating the neurotoxicity of HBCD diastereoisomers. To the best of our knowledge, this is the first study to report on the diastereoisomer-specific toxicity of HBCD in human neural cells and on the possible mechanisms responsible for the selective neurotoxicity of HBCD diastereoisomers. © 2019 Published by Elsevier B.V.

894

X. Shi et al. / Science of the Total Environment 686 (2019) 893–902

1. Introduction Hexabromocyclododecane (HBCD) has been widely applied as a brominated flame retardant (BFR) in polystyrene insulation foam, textile coatings, and high-impact polystyrene for electronic devices, which has resulted in its ubiquitous detection (Cao et al., 2018) in different environmental matrices (Feng et al., 2012; Xiang et al., 2015), biota (Jo et al., 2017; Kim et al., 2018), and human tissues (Fujii et al., 2018; Inthavong et al., 2017). Of particular concern, various adverse effects have been demonstrated in both humans (Szabo et al., 2017) and animals (Wang et al., 2016; Wang et al., 2018). Thus, due to its environmental persistence, long distance transportation, bioaccumulation, and toxicity, HBCD was listed in Annex A (POP for elimination) under the Stockholm Convention on Persistent Organic Pollutants in May 2013. Commercially available HBCD mainly consists of three diastereoisomers, i.e., α-, β-, and γ-HBCD, which account for 1%–12%, 1%–13%, and 75%–89% of total HBCD, respectively (Ruan et al., 2018b). Enrichment and distribution of individual diastereoisomers in biota (Li et al., 2017; Ruan et al., 2018b; Sanders et al., 2013) and human tissues (Erratico et al., 2016; Ryan and Rawn, 2014) are diastereoisomer-selective. For example, γ-HBCD is the dominant diastereoisomer in intertidal surface sediment, whereas α-HBCD is the most abundant in high trophic level species from the marine food web in a contaminated Norwegian fjord (Haukas et al., 2009). In finless porpoises and Indo-Pacific humpback dolphins, α-HBCD accounts for N90% of total HBCD (Ruan et al., 2018a). The major routes of HBCD uptake in humans are dietary intake and house dust ingestion (Abdallah et al., 2008; Barghi et al., 2017; de Wit et al., 2012). Rawn et al. (2014) reported on HBCD concentrations in human fetal liver and placental tissues, which ranged from below the limit of detection (LOD, 1 ng/g) to 4500 ng/g lipid weight (lw) and b LOD to 5600 ng/g lw, respectively, with the α-, β-, and γ-HBCD constituents ranging from 55.8 ± 31.4% to 67.6 ± 32.8%, 10.5 ± 16.7% to 5.97 ± 14.3%, and 33.7 ± 32.9% to 26.4 ± 31.2%, respectively. Particular attention has also been paid to the accumulation of HBCD in body fluids, such as blood and breast milk. It has been reported that the total concentrations of HBCD in blood serum of different countries ranged from 0.46 ng/g lw (Swedish, Weiss et al., 2006) to 3.1 ng/g lw (Australian, Drage et al., 2017). Thomsen et al. (2007) reported an exceptional high blood serum HBCD levels in expandable polystyrene (EPS) plant workers from Norway, which ranged from 6 to 856 ng/g lw. Drage et al. (2017) further stated that α-HBCD is the major diastereoisomer in blood. Fujii et al. (2018) also found higher levels of α-HBCD (2.2 ng/g lw) compared to β- (0.19 ng/g lw) and γ-HBCD (0.29 ng/g lw) in human breast milk of Japan. There is compelling evidence that the toxicological risks of HBCD are diastereoisomer-dependent (Du et al., 2012; Hong et al., 2017). Zhang et al. (2018) reported that the toxicity of HBCD diastereoisomers on pak choi leaves followed the order γ-HBCD N α-HBCD N β-HBCD. Du et al. (2012) found that exposure to 0.01 mg/L γ-HBCD significantly delayed the hatching of zebrafish embryos, whereas no significant changes were observed for α- and β-HBCD treatment at the same level. In-vitro experiments have shown that the cytotoxicity of α-HBCD in Hep G2 cells is less severe than that of β-HBCD and γ-HBCD (Zhang et al., 2008). Due to its lipophilic nature, HBCD can pass through the blood brain barrier and accumulate in the brain, causing potential neurotoxicity (Szabo et al., 2017). Reffatto et al. (2018) integrated in-vivo and invitro experiments and found that apoptotic gene networks are affected in the brains of HBCD-exposed mice, as well as in N2A and NSC19 neuronal cell lines. Genskow et al. (2015) reported that HBCD can cause death in the SK-N-SH catecholaminergic cell line and reduce the growth and viability of TH+ primary cultured neurons. Although the neurotoxicity of HBCD is well documented, the diastereoisomer-specific neurotoxicity of HBCD remains unclear. In this study, human SH-SY5Y cells were chosen as an in-vitro model to compare the potential neurotoxicity of α-, β-, and γ-HBCD as their cellular characteristics are representative of those in an immature

nervous system. A series of tests, including cell apoptosis rates, expression levels of mRNA and proteins, cell cycle arrest, intracellular reactive oxygen species (ROS) level, mitochondrial superoxide level, cellular adenosine triphosphate (ATP) consumption, intracellular calcium ion (Ca2+) level, and DNA damage, were performed to investigate the toxic effect and pathway of HBCD diastereoisomers. The obtained results are expected to provide a better understanding of HBCD neurotoxicity at the diastereoisomer level. 2. Materials and methods 2.1. Chemicals The three HBCD diastereoisomers, i.e., α-, β-, and γ-HBCD, were obtained from AccuStandard (New Haven, CT, USA) and their purities were higher than 98%. Their physical and chemical properties are illustrated in Table S1 (Marvin et al., 2011). Dimethyl sulfoxide (DMSO, purity N99.9%) was obtained from Sigma-Aldrich Inc. (St. Louis, MO, USA). The MEM medium, F12 medium, and fetal bovine serum were purchased from Gibco (Life Technologies, NY, USA). The penicillinstreptomycin (100 U mL−1), phosphate buffered saline (PBS), Hank's balanced salt solution, and trypsin were obtained from Hyclone (USA). Plasticware was purchased from Corning (USA). The 2′,7′dichlorofluorescein diacetate (DCFH-DA), propidium iodide (PI), RnaseA, pluronic F-127, DAPI, phalloidin-FITC, Triton X-100, Hoechst 33342, protease inhibitor cocktail, RnaseA, and Annexin V-FITC Apoptosis Detection Kit were purchased from Sigma (USA). Fluo-3/AM, alamarBlue, TRIzol reagent, SDS-PAGE gel, and MitoSox Red were purchased from Invitrogen (USA). The CytoTox-ONE homogeneous membrane integrity assay was obtained from Promega (USA). The PrimeScript™ RT Reagent Kit, with gDNA Eraser and SYBR® Premix Ex Taq™ II, was purchased from Takara (Japan). The ATP Assay Kit, 4% formaldehyde, protease inhibitor cocktail, and RIPA lysis buffer were obtained from the Beyotime Institute (China). The BCA Protein Assay Kit was purchased from Tiangen Biotech (China). Bax (1:1000–1:10000), caspase-3 (1:200), Bcl-2 (1:1000), and γ-H2AX (1:1000) were obtained from Abcam (UK). Cytochrome c (Cyt c) (1:1000) and caspase-9 (1:1000) were obtained from Cell Signaling Technology (USA). β-actin (1:2000), Goat Anti-Mouse IgG(H + L) Dylight 800 (1:800), and Goat Anti-Rabbit IgG (H + L) Dylight 800 (1:800) were obtained from Bioworld Technology (USA). 2.2. Cell culture The SH-SY5Y human neuroblastoma cell line was obtained from the American Type Culture Collection (Manassas, VA, USA) and was maintained in a flask in an incubator in 5% CO2 at 37 °C. The culture medium contained MEM and F12 medium (1:1) supplemented with 10% fetal bovine serum and 100 U mL−1 penicillin-streptomycin. To carry out the experiments, cells were seeded in 96- or 6-well plates (flat bottomed) and allowed to adhere for 24 h at 37 °C. Cell densities were approximately in the range of 2–5 × 105 cells/well at the beginning of cell exposure. The α-, β-, and γ-HBCD diastereoisomers were dissolved in DMSO as stock solutions. Cells were treated with viable concentrations of α-, β-, and γ-HBCD (0.1% DMSO solution), respectively, for a viable time, depending on the assay performed. 0.1% DMSO was used as the control. All experiments were conducted at 37 °C. 2.3. Evaluation of cellular toxicity 2.3.1. Cell viability assay An alamarBlue assay was used to test the potential effects of the HBCD diastereoisomers on the viability of SH-SY5Y cells according to the method of Chen et al. (2016). Briefly, SH-SY5Y cells were exposed to different concentrations (0.001, 0.01, 0.1, 1.0, 2.5, 5.0 μmol/L) of the HBCD diastereoisomers. After 24 h of treatment, alamarBlue dye was

X. Shi et al. / Science of the Total Environment 686 (2019) 893–902

added. The fluorescence intensity was recorded on a microplate reader (Thermo Fisher Scientific, USA) at an excitation wavelength of 570 nm and emission wavelength of 585 nm. Six replicates were performed for each treatment. A parallel set of experiments was conducted without cells to exclude the potential interaction of the HBCD diastereoisomers with the alamarBlue dye. Results showed that there were no interactions between the HBCD diastereoisomers and dye. 2.3.2. Cell membrane integrity assay Cell membrane integrity was assessed by propidium iodide (PI) staining and lactate dehydrogenase (LDH) release in accordance with Zhu et al. (2017). The cells were treated with 0.001, 0.01, 0.1, 1.0, 2.5, or 5.0 μmol/L α-, β-, and γ-HBCD diastereoisomers, respectively, for 24 h. Thereafter, cells were stained with PI according to the manufacturer's instructions. Fluorescence was visualized on a TCS SP5 laser scanning confocal microscope (Germany) at an excitation wavelength of 535 nm and emission wavelength of 617 nm. The LDH activity in the cell culture medium was determined using a commercial kit (CytoTox-ONE homogeneous membrane integrity assay, Promega, USA). Fluorescence intensity was recorded at an excitation wavelength of 560 nm and emission wavelength of 590 nm. Six replicates were performed for each treatment. 2.3.3. Cytoskeleton assay The cell cytoskeleton was assessed by phalloidin-FITC/DAPI staining (Chen et al., 2016) according to the manufacturer's instructions (Sigma, USA). After treatment with the HBCD diastereoisomers (5.0 μmol/L) for 24 h, the cell cytoskeleton and nuclei were stained with phalloidin-FITC (green) and DAPI (blue), respectively. Fluorescence was determined using a confocal microscope. The excitation and emission wavelengths were 495 nm and 513 nm for phalloidin-FITC and 364 nm and 454 nm for DAPI, respectively. Three replicates were performed for each treatment. 2.4. Determination of cell apoptosis 2.4.1. Cell morphology and cell apoptosis analysis by flow cytometry Apoptosis was assessed by Hoechst 33342 staining and annexin-V FITC/PI double staining (Valdiglesias et al., 2013). Cells were treated with the HBCD diastereoisomers (5.0 μmol/L) for 24 h, after which the cellular morphology was observed by confocal microscopy. Hoechst 33342 staining was performed according to the kit's instructions (Sigma, USA). For the annexin-V FITC/PI double staining assay, at least 104 events were acquired with a NovoCyte D1040 flow cytometer (ACEA Biosciences Inc., USA). The cells were then treated with the HBCD diastereoisomers (1.0 and 5.0 μmol/L) for 24 h, with three replicates for each treatment. 2.4.2. Cell cycle analysis by flow cytometry After the cells were treated with different concentrations (0.001, 0.01, 0.1, 1.0, 2.5 and 5.0 μmol/L) of the HBCD diastereoisomers for 24 h, cell distribution during different phases of the cell cycle was examined by evaluating relative cellular DNA content using flow cytometry as Yu et al. (2008). DNA content was assessed from the PI signal detected by the FL2 detector in a minimum of 104 events. 2.4.3. DNA damage analysis Levels of γ-H2AX were used to assess DNA damage (Chen et al., 2016). Total proteins were extracted from cells treated with the HBCD diastereoisomers (5.0 μmol/L) for 24 h, with the levels of γ-H2AX then evaluated by Western blot analysis. 2.4.4. Apoptosis-related mRNA expression analysis by quantitative reverse transcription polymerase chain reaction (RT-qPCR) The mRNA expression of several cell apoptosis-related genes was assessed using RT-qPCR (He et al., 2008). After the cells were treated

895

with the HBCD diastereoisomers (5 μmol/L) for 24 h, total RNA was extracted from the SH-SY5Y cells using TRIzol reagent. The A260/280 ratio ranged from 1.8 to 2.0, and total RNA was electrophoresed on a 1% agarose gel to visually assess quality. DNase I-treated total RNA from each sample was converted to cDNA using the Reverse Transcript Kit following the manufacturer's instructions. The primer pair sequences and lengths of production are shown in Table S2. Primers were synthesized at AuGCT (Beijing, China). Relative gene expression was appraised with a SYBR Green qPCR master mix (Takara, Japan) on a Roche LightCycler 480II qPCR instrument (Switzerland) using β-actin as the housekeeping gene. The 2-ΔΔCT method was used for quantity calculations. 2.4.5. Apoptosis-related protein expression analysis by Western blot analysis After the cells were treated with the HBCD diastereoisomers (5.0 μmol/L) for 24 h, total protein was extracted from the cells using RIPA lysis buffer and then determined using the Lowry method (Tiangen Biotech, China). Three replicates were performed for each treatment. In brief, cells treated with the HBCD diastereoisomers (5.0 μmol/L) for 24 h were lysed with RIPA lysis buffer containing the protease inhibitor cocktail. All proteins were then boiled for 5 min in denaturing sample buffer, electrophoresed on a 4%–12% SDS polyacrylamide gel, and transferred onto a polyvinylidene difluoride membrane (Bio-Rad, Germany). The membrane was further probed with antibodies against β-actin, Bax, caspase-3, Bcl-2, γ-H2AX, Cyt c, and caspase-9 overnight at 4 °C. Secondary antibodies were added for 1.5 h, which were Goat Anti-Mouse IgG (H + L) Dylight 800 and Goat Anti-Rabbit IgG (H + L) Dylight 800. The fluorescence signal was detected by Odyssey Infrared Imaging (LI-COR, USA). 2.5. Evaluation of oxidative stress 2.5.1. Intracellular ROS accumulation analysis A nonfluorescent sensitive intracellular probe, DCFH-DA, was used for detecting ROS formation according to Gao et al. (2009). After SHSY5Y cells were treated with different concentrations of the HBCD diastereoisomers for 3 h and 24 h, the fluorescence was then measured at an emission wavelength of 525 nm using an excitation wavelength of 488 nm on a flow cytometer. Moreover, inhibitor was applied to clarify the relationship between ROS and apoptosis. The cells were preincubated with 1 mmol/L N-acetyl-L-cysteine (NAC) (Sigma Aldrich, St. Louis, MO, USA) and then treated with 5.0 μmol/L HBCD diastereoisomers. The concentration of NAC was chosen based on the results of the cytotoxicity of NAC (1.0–5.0 mmol/L) on SH-SY5Y cells. The viability of SH-SY5Y cells were determined as described above. 2.5.2. Mitochondrial superoxide accumulation analysis The MitoSOX™ Red mitochondrial superoxide indicator is a novel fluorogenic dye for highly selective detection of superoxide in the mitochondria of live cells. After cells were treated with different concentrations of the HBCD diastereoisomers for 3 h, the mitochondrial superoxide levels were analyzed on a NovoCyte 1040 flow cytometer in accordance with Chen et al. (2016). 2.6. Determination of intracellular Ca2+ concentrations To investigate whether Ca2+ signaling involves cell apoptosis, the intracellular Ca2+ dynamics in the SH-SY5Y cells were measured by flow cytometer using the Ca2+-sensitive fluorescent probe Fluo-3/AM (Yu et al., 2008). Briefly, cells were exposed to different concentrations of the HBCD diastereoisomers for 6 h and 24 h in Hank's balanced salt solution without Ca2+. Thereafter, cells were incubated with 2.5 μmol/L Fluo-3/AM in the dark at 37 °C. After 30 min, the Fluo-3 fluorescence intensity was recorded via flow cytometer (excitation at 488 nm and emission at 525 nm). Moreover, inhibitor was applied to clarify the relationship between intracellular Ca2+ concentrations and cell viabilities.

896

X. Shi et al. / Science of the Total Environment 686 (2019) 893–902

The cells were preincubated with 1 μmol/L BAPTA-AM (Sigma Aldrich, St. Louis, MO, USA) and were then treated with HBCD diastereoisomers (5.0 μmol/L). The viability of SH-SY5Y cells were determined as described above.

(A)

Cell viability

2.7. Analysis of cellular ATP levels An ATP Assay Kit was used to assess the cellular ATP levels following the instructions provided by the manufacturer (Chen et al., 2016). Briefly, cells were treated with different concentrations of the HBCD diastereoisomers for 24 h. The cells were then lysed at 4 °C using the cell lysis buffer provided by the manufacturer. Collected samples were further subjected to a bioluminescent intensity assay using a microplate reader (Thermo Fisher Scientific, USA). 2.8. Statistical analysis Statistical analyses were performed using GraphPad Prism 5. Data are shown as means ± standard error. One-way analysis of variance (ANOVA) was used to determine the differences among multiple (≥ 3) samples. P-values were calculated to quantify statistical significance, with the criterion set at P b 0.05. 3. Results

(B)

LDH release

3.1. Cytotoxicity of HBCD in SH-SY5Y cells 3.1.1. Cell viability Cell viability was determined by alamarBlue assay after exposing SH-SY5Y cells to various concentrations (0.001, 0.01, 0.1, 1.0, 2.5, and 5.0 μmol/L) of α-, β-, and γ-HBCD for 24 h. All three HBCD diastereoisomers induced dose-dependent decreases in cell viability (Fig. 1A). Significant decreases in cell viability were found at treatment concentrations higher than 0.01 μmol/L for β-HBCD, but higher than 0.1 μmol/L for α- and γ-HBCD. When the cells were exposed to a HBCD concentration of 5.0 μmol/L, the cell viabilities were 74.24 ± 4.38% (α-HBCD), 44.45 ± 4.75% (β-HBCD), and 69.00 ± 3.08% (γHBCD) of the control, suggesting that the toxic effects of the three diastereoisomers followed the order β-HBCD N γ-HBCD N α-HBCD. 3.1.2. Cell membrane integrity We used PI staining to test the effects of the HBCD diastereoisomers on cell membrane integrity. Results showed a significant increase in PIstained (positive) cells after exposure to the HBCD diastereoisomers (5.0 μmol/L) for 6 h and 24 h (Fig. S1). The toxicity of β-HBCD was higher than that of α- or γ-HBCD. The effects of the HBCD diastereoisomers on cell membrane integrity were assessed by measuring extracellular LDH release from the cytoplasm. The cells were exposed to different concentrations of the HBCD diastereoisomers (0.001–5.0 μmol/L) for 24 h. When the exposure concentrations were higher than 0.1 μmol/L, the LDH level increased significantly, indicating that the HBCD diastereomers could cause membrane breakage (Fig. 1B). The LDH levels of α-, β-, and γ-HBCD showed no significant differences among the three diastereoisomers at the 0.001–2.5 μmol/L exposure levels. However, when the exposure level increased to 5.0 μmol/L, LDH release levels in the α-, β-, and γ-HBCD-treated groups were 1.6, 1.8, and 1.7 times that of the control, respectively, indicating that β-HBCD was more toxic than α- or γ-HBCD at relatively higher concentrations. 3.1.3. Cytoskeleton The effects of the HBCD diastereoisomers on cell actin fibers and stress fiber formation were studied when the cells were treated with α-, β-, or γ-HBCD (5.0 μmol/L) for 24 h. Significant disassembly of actin fibers and reduction in stress fiber formation with reduced density of actin meshwork were observed in all treated cells (Fig. S2).

Fig. 1. HBCD diastereoisomer exposure induced cytotoxicity in SH-SY5Y cells. (A) Cell viability was determined by alamarBlue assay after exposure of SH-SY5Y cells to HBCD diastereoisomer for 24 h. (B) Extracellular LDH evaluation for HBCD-exposed SH-SY5Y cells (24 h). # P b 0.05 indicates significant difference among groups. * P b 0.05, compared with control.

3.2. SH-SY5Y cell apoptosis 3.2.1. Cell morphology and cell apoptosis Cells showed the characteristic features of apoptosis and necrosis following exposure to 5.0 μmol/L α-, β-, and γ-HBCD, including cell volume shrinkage, irregular morphological change, and apoptotic and necrosis bodies, which were not observed in the control cells (Fig. S3). Cell apoptosis and necrosis were further investigated by flow cytometry. As shown in Fig. 2A and B, the percentages of early and late apoptotic cells (Q2 + Q4), live and healthy cells (Q3 phase), and necrotic cells (Q1 phase) were evaluated for the HBCD diastereoisomer-exposed SHSY5Y cells. Under HBCD diastereoisomer treatment, concentrationdependent increases in early and late apoptotic cells (Q4 and Q2) and decreases in live cells (Q3) were found. The percentages of early and late apoptotic cells (Q2 + Q4) were much higher than those of necrotic

X. Shi et al. / Science of the Total Environment 686 (2019) 893–902

897

Fig. 2. HBCD diastereoisomer exposure induced apoptosis in SH-SY5Y cells. (A) Apoptotic rate of SH-SY5Y cells treated with HBCD diastereoisomers based on flow cytometry. Phase distribution of cells stained with Annexin V-FITC and PI. (B) Dose-response of HBCD diastereoisomers on apoptotic rate of SH-SY5Y cells. # P b 0.05 indicates significant difference among groups. * P b 0.05, compared with control.

cells (Q1), suggesting that the HBCD-induced effect was dominated by cell apoptosis. When the exposure level increased to 1.0 and 5.0 μmol/ L, the percentages of apoptotic cells (Q2 + Q4) following β-HBCD treatment were 2.0 and 4.3 times higher, respectively, than the control, and higher than those in the α-HBCD- (1.4 and 1.7 times, respectively) and γ-HBCD-treated groups (1.6 and 1.8 times, respectively).

3.2.2. Cell cycle As shown in Fig. 3A, compared with the control, HBCD diastereoisomer treatment at 0.001–0.01 μmol/L did not significantly change the percentage of cells in the sub-G1 phase of the cell cycle. However, when the exposure concentrations increased to 1.0, 2.5, and 5.0 μmol/ L, significant increases in cells in the sub-G1 phase were found (P b

898

X. Shi et al. / Science of the Total Environment 686 (2019) 893–902

(A)

Cell cycle

(B-2)

(B-1)

DNA damage

17 KDa

γ-H2AX

42 KDa

β-actin

Ctrl

α

β γ HBCD

Fig. 3. HBCD diastereoisomer exposure induced cell cycle arrest and DNA damage in SH-SY5Y cells. (A) Cell cycle distribution of SH-SY5Y cells cultured with or without HBCD diastereoisomers. (B) Western blot analysis of γ-H2AX expression in HBCD diastereoisomer-treated SH-SY5Y cells (37 °C, 24 h). (B-1) γ-H2AX protein detection using Western blot analysis (24 h). (B-2) Rate of γ-H2AX protein expression in HBCD diastereoisomer-treated SH-SY5Y cells. # P b 0.05 indicates significant difference among groups. * P b 0.05, compared with control.

0.05). At 5.0 μmol/L, the proportions of cells in the sub-G1 phase in the α-, β-, and γ-HBCD-treated groups were 3.1, 6.2, and 4.1 times higher, respectively, than that of the control. 3.2.3. DNA damage To investigate whether the three HBCD diastereoisomers induced genotoxicity in the cells, the level of phosphorylated histone H2AX was examined. Elevated levels of γ-H2AX foci were observed in the 5.0 μmol/L α-, β-, and γ-HBCD-treatment groups after 24 h. The expression levels of γ-H2AX in the α-, β-, and γ-HBCD-treated groups were 1.2, 1.6, and 1.4 times higher, respectively, than that of the control (Fig. 3B-1 and B-2). 3.2.4. Apoptosis-related mRNA expression Exposure to the HBCD diastereoisomers increased the mRNA expression levels of caspase-3, Bax, and Cyt c (Fig. 4A), but decreased the expression level of Bcl-2 and XIAP. Among the three HBCD diastereoisomers, the expression levels of caspase-3, Bax, and Cyt c were significantly higher under β-HBCD treatment than under α- or γ-HBCD treatment, whereas the expression levels of XIAP and Bcl-2 were lower following β-HBCD exposure than α- or γ-HBCD exposure. 3.2.5. Apoptosis-related protein expression Total protein mass decreased to 88.54 ± 2.17%, 65.67 ± 0.59%, and 80.75 ± 0.25% of the control following α-, β-, and γ-HBCD exposure,

respectively (Fig. 4B). The protein expression levels of apoptosisrelated proteins, including caspase-3, caspase-9, Cyt c, Bax, and Bcl-2 under 5.0 μmol/L HBCD diastereoisomer treatment are shown in Fig. 4C. The protein expression levels of caspase-3 and caspase-9 increased significantly under α-, β-, and γ-HBCD treatment, whereas Cyt c only increased significantly under β- and γ-HBCD treatment. Caspase-9 and caspase-3 expression levels followed the order β-HBCD N γ-HBCD N α-HBCD (Figs. S4-A and S4\\B), whereas those of Cyt c followed the order β-HBCD N γ-HBCD N α-HBCD (Fig. S4\\C). The ratios of Bcl-2 to Bax decreased under HBCD diastereoisomer treatment, and were 70.26 ± 0.79%, 52.38 ± 0.40%, and 63.54 ± 0.55% compared to the control following α-, β-, and γ-HBCD exposure, respectively (Fig. S4-D). 3.3. Oxidative stress Intracellular ROS generation was evaluated following exposure to different concentrations of the HBCD diastereoisomers (0.01–5.0 μmol/L) for 3 h (Fig. 5A) and 24 h (Fig. 5B). HBCD diastereoisomer exposure at 0.01–2.5 μmol/L resulted in concentration-dependent increases in intracellular ROS levels. However, when the exposure levels further increased to 5.0 μmol/L, the ROS levels decreased obviously. Comparing the different HBCD diastereoisomers indicated that the ROS levels in the treated cells did not differ significantly at the 0.01–0.1 μmol/L exposure levels for 3 h exposure. However, when the exposure levels increased to

X. Shi et al. / Science of the Total Environment 686 (2019) 893–902

(A)

(B)

899

mRNA expression

Total protein mass

(C)

Relative expression levels of proteins

17 KDa

Caspase-3

35 KDa

Caspase-9

14 KDa

Cyt C

21 KDa

Bax

26 KDa

Bcl-2

42 KDa

β-actin

Ctrl

α

β γ HBCD

Fig. 4. Analysis of mRNA and protein expression levels in HBCD diastereoisomer-treated SH-SY5Y cells. (A) RT-qPCR analysis of selected apoptotic genes in SH-SY5Y cells following HBCD diastereoisomer treatment. (B) Relative total protein mass in SH-SY5Y cells after exposure to HBCD diastereoisomers for 24 h (n = 3). (C) Western blot analysis of selected apoptotic proteins in HBCD diastereoisomer-treated SH-SY5Y cells (37 °C, 24 h).

1.0 and 2.5 μmol/L, the ROS levels in the β-HBCD treatment groups were significantly higher than those in the α- and γ-HBCD treatment groups. When the exposure levels further increased to 5.0 μmol/L, the ROS levels in the β-HBCD treated cells decreased obviously to 83.0 ± 2.0% of the control. Different exposure times showed similar trends of ROS levels. The ROS levels for 24 h treatments were a little bit lower than those of 3 h, but were not differed significantly (P N 0.05). In the presence of the ROS scavenger NAC, the cell viabilities were significantly higher than those in the absence of NAC (Fig. S5). When the exposure level was 5.0 μmol/L, cell viabilities in the α-, β-, and γHBCD-treated groups were 90.9 ± 2.0, 84.8 ± 1.3, and 87.3 ± 1.2%, respectively, of the control in the presence of NAC, while those were 74.2 ± 4.4, 44.5 ± 4.8, and 69.0 ± 3.1%, respectively of the control in the absence of NAC. These results indicated that oxidative stress contributed to the cytotoxicity of HBCD diastereoisomers. Differences in mitochondrial superoxide production were observed (0.01–5.0 μmol/L) for 3 h exposure. Exposure to the HBCD diastereoisomers at 0.1–2.5 μmol/L caused concentration-dependent increases in the mitochondrial superoxide production levels (Fig. 5C). Comparing the different HBCD diastereoisomers demonstrated that the superoxide production levels of three HBCD-treated cells did not differ significantly at the 0.01 μmol/L exposure level. However, when the exposure level increased to 2.5 μmol/L, the superoxide production levels in the β-HBCDtreated groups were significantly higher than those of the α- and γ-

HBCD-treated groups. The superoxide production levels in the mitochondria were 1.2 (α-HBCD), 3.0 (β-HBCD), and 1.5 (γ-HBCD) times higher, respectively, than those in the control (Fig. 5C). 3.4. Intracellular Ca2+ concentration Intracellular Ca2+ concentrations were determined following exposure to the HBCD diastereoisomers (0.01–5.0 μmol/L) for 6 h and 24 h using flow cytometry. Concentration-dependent increases in Fluo-3 fluorescence were observed, with the levels in the γ-HBCD-treated group higher than those in the α- and β-HBCD-treated groups (Fig. 6A and B). At exposure concentrations of 5.0 μmol/L, the intracellular Ca2 + levels in the α-, β-, and γ-HBCD-treated groups were 1.3, 1.4, and 1.9 times higher, respectively, than those of the control. No significant difference of intracellular Ca2+ level was found between different exposure times, except that the levels in the γ-HBCD-treated group for 6 h exposure was 1.4 times that of 24 h. In the presence of the Ca2+ influx inhibitor, the cytotoxic effects induced by HBCD diastereoisomers reduced significantly (Fig. S6). When the exposure level was 5.0 μmol/L, cell viabilities in the α-, β-, and γHBCD-treated groups were 96.6 ± 0. 8, 63.0 ± 2.3, and 92.3 ± 0.8% of the control in the presence of BAPTA-AM, while those were 74.2 ± 4.4, 44.5 ± 4.8, and 69.0 ± 3.1%, respectively, of the control in the absence of BAPTA-AM.

900

X. Shi et al. / Science of the Total Environment 686 (2019) 893–902

(A)

Reactive oxygen species

(B)

Reactive oxygen species

(C)

(A)

Intracellular calcium

(B)

Intracellular calcium

(C)

Cellular ATP level

Mitochondrial superoxide

Fig. 5. HBCD diastereoisomer exposure caused oxidative stress in SH-SY5Y cells. (A) Intracellular ROS production in cells after exposure to HBCD diastereoisomers for 3 h (n = 3–6). (B) Intracellular ROS production in cells after exposure to HBCD diastereoisomers for 24 h (n = 3–6). (C) Mitochondrial superoxide production in cells exposed to HBCD diastereoisomers (n = 3–6) for 3 h. # P b 0.05 indicates significant difference among groups. * P b 0.05, compared with control.

Fig. 6. Changes in the intracellular Ca2+ concentrations and ATP levels in HBCD diastereoisomer-treated SH-SY5Y cells. (A) SH-SY5Y cells treated with HBCD diastereoisomers for 6 h were incubated with Fluo-3/AM and examined for fluorescence intensity. (B) SH-SY5Y cells treated with HBCD diastereoisomers for 24 h were incubated with Fluo-3/AM and examined for fluorescence intensity. (C) Relative cellular ATP production in SH-SY5Y cells following HBCD diastereoisomers exposure for 24 h. # P b 0.05 indicates significant difference among groups. * P b 0.05, compared with control.

X. Shi et al. / Science of the Total Environment 686 (2019) 893–902

3.5. Cellular ATP level The ATP levels in the SH-SY5Y cells were tested after exposure to the HBCD diastereoisomers (0.1–5.0 μmol/L) for 24 h. Exposure did not cause significant decreases in cellular ATP levels, except for cells treated with β-HBCD at 2.5 and 5.0 μmol/L and γ-HBCD at 5.0 μmol/L, which deceased to 57.40 ± 1.86%, 28.75 ± 0.85%, and 87.91 ± 1.31%, respectively, compared with the control (Fig. 6C). 4. Discussion HBCD diastereoisomers have been detected in various environmental and biological matrices, and even in the tissues of nonoccupationally exposed humans (Feng et al., 2012; Jo et al., 2017; Kim et al., 2018; Xiang et al., 2015). Thus, considerable attention has been paid to their potential neurotoxic effects in humans. In-vivo tests have indicated that during brain development, exposure to HBCD can lead to adverse effects on memory and learning abilities in mice (Szabo et al., 2017) and dopamine-dependent behavior and hearing dysfunction in rats (Lilienthal et al., 2009). In-vitro tests have shown that HBCD is cytotoxic and affects a range of neuronal cell functions, including neurotransmission (Genskow et al., 2015) and cell death (Shi et al., 2017). To evaluate the environmental safety and health risks of HBCD, it is necessary to study the toxicities of HBCD at the level of individual isomers. In this study, SH-SY5Y human neuroblastoma cells were used to explore the toxic effects of three HBCD diastereoisomers. The cells were exposed to the HBCD diastereoisomers at a concentration range of 0.001–5.0 μmol/L (0.642–3208.5 ng/mL). Significant decreases in cell viability were found when the exposure concentrations were higher than 0.01 μmol/L (6.42 ng/mL) for β-HBCD, while higher than 0.1 μmol/L (64.2 ng/mL) for α-, and γ-HBCDs. These levels were in the range of the blood serum HBCD levels in expandable polystyrene (EPS) plant workers reported (6 to 856 ng/g lw, Thomsen et al., 2007). In the present study, all three HBCD diastereoisomers caused dosedependent inhibition of cell viability, enhancement of membrane permeability, disassembly of actin fibers, and reduction of stress fiber formation (Figs. 1 and S1-S2). These results demonstrated that HBCD exposure can cause cytotoxicity in SH-SY5Y cells, consistent with the findings of Al-Mousa and Michelangeli (2012), who reported on the toxicity of technical HBCD in SH-SY5Y cells. Moreover, this study showed that the toxicities of the HBCD diastereoisomers were diastereoisomer-selective in the order β-HBCD N γ-HBCD N α-HBCD. It is difficult to compare the diastereoisomer-selective neurotoxic order from data obtained by other researchers using similar neuro-cell lines due to the lack of relevant information. Zhang et al. (2008) showed that the cytotoxicity of HBCD diastereoisomers to liver Hep G2 cells was in the order γ-HBCD ≥ β-HBCD N α-HBCD. Huang et al. (2016) also reported that the cytotoxicity of HBCD diastereoisomers, including cell viability, ROS levels, and DNA damage, to immortalized human liver cells (L02) and hepatoma cells (HepG2), were in the order β-HBCD ≥ γHBCD N α-HBCD, which is similar to our results. Apoptosis is an autonomous cellular death model under genetic control and is featured by cell shrinkage, DNA fragmentation, membraneassociated apoptotic bodies, and specific cysteine protease activation (Lee et al., 2005). In this study, cell shrinkage and apoptotic bodies were observed by Hoechst 33342 staining (Fig. S3) and Annexin VFITC/PI double staining (Fig. 2A and B). Caspases are thought to be the central executers of apoptosis. Caspase-3 acts as a key downstream executioner in mediating neuronal apoptosis, whereas caspase-9 is primarily responsible for initiating caspase activation (Fuchs and Steller, 2011). Cyt c plays important roles in both the life and death of a cell. Once released to the cytoplasm, Cyt c acts as a co-factor in conjunction with Apaf-1, procaspase-9, and adenosine triphosphate to induce activation of caspase-9 and subsequently caspase-3 (Li et al., 1997). Bax promotes apoptosis, whereas Bcl-2 and XIAP reverse that effect (Wang et al., 2015; Fuchs and Steller, 2011). Reffatto et al. (2018)

901

showed that caspase-3 activation increases under HBCD treatment in N2A and NSC-19 cell lines. In the current study, α-, β-, and γ-HBCD treatment resulted in increased mRNA and protein expression of caspase-3 and caspase-9 (Figs. 4 and S4), as well as the significant upregulation of Cyt c expression and decrease in Bcl-2/Bax ratios. The cell apoptotic sequence of the three HBCD diastereoisomers followed the same order as the caspase expression levels, i.e., β-HBCD N γHBCD N α-HBCD. Thus, the above evidence suggests that caspasedependent apoptosis may be one of the mechanisms of the structureselective HBCD diastereoisomer neurotoxicity. Al-Mousa and Michelangeli (2012) compared the toxic effects of some brominated flame retardants (BFRs), including HBCD, tetrabromobisphenol A (TBBPA), and decabromodiphenyl ether (DBPE) in SH-SY5Y. They found that HBCD was the most potent at inducing cell death, with the LC50 being 2.7 ± 0.7 μmol/L, while those of TBBPA and DBPE being 15 ± 4 and 28 ± 7 μmol/L, respectively. They suggested that BFRs could activate the intrinsic apoptotic pathway through exaggerated temporal increases in intracellular Ca2+ levels, leading to mitochondria dysfunction, which is manifested as Cyt c release. Moreover, HBCD can generate measurable levels of ROS, implicating the role of free radical damage to cellular components, including protein, membranes, and DNA. In this study, the levels of intracellular Ca2+, mitochondrial superoxide, and ROS following exposure to the three HBCD diastereoisomers were obviously higher than those of control. The cytotoxicity induced by HBCD diastereoisomers could be attenuated by ROS and Ca2+ inhibitors, indicating that both oxidative stress and Ca2+ homeostasis contributed to the cytotoxicity of HBCD diastereoisomers. When compared of three HBCD diastereoisomers, it was found that the levels of ROS and mitochondrial superoxide in cells treated with β-HBCD were higher than those treated with α- and γ-HBCD, which is in accordance with the relatively higher cellular toxicity of β-HBCD. Among the three HBCD diastereoisomers, γ-HBCD exposure led to the greatest intracellular Ca2+ leakage, which is in discordance with the relatively higher cellular toxicity of β-HBCD. Based on the above results, one may deduce that generation of ROS following HBCD diastereoisomer exposure may be a key factor regulating their selective neurotoxicity. In summary, our study demonstrated the toxicity of HBCD diastereoisomers in SH-SY5Y cells. Caspase-dependent apoptosis appeared to be one of the main mechanisms. The three HBCD diastereoisomers exhibited different neurotoxic effects, with the toxic effects of β-HBCD found to be greater than those of γ- or α-HBCD. HBCD diastereoisomer treatment led to Ca2+ imbalance and oxidative stress, with oxidative stress possibly responsible for the HBCD diastereoisomer-selective apoptosis of the SH-SY5Y cells. Our results provide new insight into the developmental neurotoxicity of HBCD diastereoisomers.

Acknowledgements This study was financially supported by the Key Research Development Plans of Special Project for Site Soils (Projects 2018YFC1801002) and National Natural Science Foundation of China (Projects 21537005, 41877479, and 21577155). Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.scitotenv.2019.06.008.

References Abdallah, M.A.E., Harrad, S., Covaci, A., 2008. Hexabromocyclododecanes and tetrabromobisphenol-A in indoor air and dust in Birmingham, UK: implications for human exposure. Environ. Sci. Technol. 42, 6855–6861. Al-Mousa, F., Michelangeli, F., 2012. Some commonly used brominated flame retardants cause Ca2+-ATPase inhibition, beta-amyloid peptide release and apoptosis in SHSY5Y neuronal cells. PLoS One 7.

902

X. Shi et al. / Science of the Total Environment 686 (2019) 893–902

Barghi, M., Shin, E.S., Kim, J.C., Choi, S.D., Chang, Y.S., 2017. Human exposure to HBCD and TBBPA via indoor dust in Korea: estimation of external exposure and body burden. Sci. Total Environ. 593, 779–786. Cao, X., Lu, Y., Zhang, Y., Khan, K., Wang, C., Baninla, Y., 2018. An overview of hexabromocyclododecane (HBCDs) in environmental media with focus on their potential risk and management in China. Environ. Pollut. 236, 283–295. Chen, Y., Xu, M., Zhang, J., Ma, J., Gao, M., Zhang, Z.H., Xu, Y., Liu, S.J., 2016. Genome-wide DNA methylation variations upon exposure to engineered nanomaterials and their implications in nanosafety assessment. Adv. Mater. 29. de Wit, C.A., Bjorklund, J.A., Thuresson, K., 2012. Tri-decabrominated diphenyl ethers and hexabromocyclododecane in indoor air and dust from Stockholm microenvironments 2: indoor sources and human exposure. Environ. Int. 39, 141–147. Drage, D.S., Mueller, J.F., Hobson, P., Harden, F.A., Toms, L.M.L., 2017. Demographic and temporal trends of hexabromocyclododecanes (HBCDD) in an Australian population. Environ. Res. 152, 192–198. Du, M.M., Zhang, D.D., Yan, C.Z., Zhang, X., 2012. Developmental toxicity evaluation of three hexabromocyclododecane diastereoisomers on zebrafish embryos. Aquat. Toxicol. 112-113, 1–10. Erratico, C., Zheng, X.B., van den Eede, N., Tomy, G., Covaci, A., 2016. Stereoselective metabolism of α-, β-, and γ-hexabromocyclododecanes (HBCDs) by human liver microsomes and CYP3A4. Environ. Sci. Technol. 50, 8263–8273. Feng, A.H., Chen, S.J., Chen, M.Y., He, M.J., Luo, X.J., Mai, B.X., 2012. Hexabromocyclododecane (HBCD) and tetrabromobisphenol A (TBBPA) in riverine and estuarine sediments of the Pearl River Delta in southern China, with emphasis on spatial variability in diastereoisomer- and enantiomer-specific distribution of HBCD. Mar. Pollut. Bull. 64, 919–925. Fuchs, Y., Steller, H., 2011. Programmed cell death in animal development and disease. Cell 147, 742–758. Fujii, Y., Kato, Y., Masuda, N., Harada, K.H., Koizumi, A., Haraguchi, K., 2018. Contamination trends and factors affecting the transfer of hexabromocyclododecane diastereomers, tetrabromobisphenol a, and 2,4,6-tribromophenol to breast milk in Japan. Environ. Pollut. 237, 936–943. Gao, P., He, P., Wang, A.G., Xia, T., Xu, B.Y., Xu, Z.X., Niu, Q., Guo, L.J., Chen, X.M., 2009. Influence of PCB153 on oxidative DNA damage and DNA repair-related gene expression induced by PBDE-47 in human neuroblastoma cells in vitro. Toxicol. Sci. 107, 165–170. Genskow, K.R., Bradner, J.M., Hossain, M.M., Richardson, J.R., Caudle, W.M., 2015. Selective damage to dopaminergic transporters following exposure to the brominated flame retardant, HBCDD. Neurotoxicol. Teratol. 52, 162–169. Haukas, M., Hylland, K., Berge, J.A., Nygard, T., Mariussen, E., 2009. Spatial diastereomer patterns of hexabromocyclododecane (HBCD) in a Norwegian fjord. Sci. Total Environ. 407, 5907–5913. He, P., He, W.H., Wang, A.G., Xia, T., Xu, B.Y., Zhang, M., Chen, X.M., 2008. PBDE-47-induced oxidative stress, DNA damage and apoptosis in primary cultured rat hippocampal neurons. Neurotoxicology 29, 124–129. Hong, H.Z., Lv, D.M., Liu, W.X., Huang, L.M., Chen, L.Y., Shen, R., Shi, D.L., 2017. Toxicity and bioaccumulation of three hexabromocyclododecane diastereoisomers in the marine copepod Tigriopus japonicas. Aquat. Toxicol. 188, 1–9. Huang, X.M., Chen, C., Shang, Y., Zhong, Y.F., Ren, G.F., Yu, Z.Q., An, J., 2016. In vitro study on the biotransformation and cytotoxicity of three hexabromocyclododecane diastereoisomers in liver cells. Chemosphere 161, 251–258. Inthavong, C., Hommet, F., Bordet, F., Rigourd, V., Guerin, T., Dragacci, S., 2017. Simultaneous liquid chromatography-tandem mass spectrometry analysis of brominated flame retardants (tetrabromobisphenol A and hexabromocyclododecane diastereoisomers) in French breast milk. Chemosphere 186, 762–769. Jo, H., Son, M.H., Seo, S.H., Chang, Y.S., 2017. Matrix-specific distribution and diastereomeric profiles of hexabromocyclododecane (HBCD) in a multimedia environment: air, soil, sludge, sediment, and fish. Environ. Pollut. 226, 515–522. Kim, J.T., Choi, Y.J., Barghi, M., Yoon, Y.J., Kim, J.H., Kim, J.H., Chang, Y.S., 2018. Occurrence and distribution of old and new halogenated flame retardants in mosses and lichens from the South Shetland Islands, Antarctica. Environ. Pollut. 235, 302–311. Lee, H.J., Wang, C.J., Kuo, H.C., Chou, F.P., Jean, L.F., Tseng, T.H., 2005. Induction apoptosis of luteolin in human hepatoma HepG2 cells involving mitochondria translocation of Bax/Bak and activation of JNK. Toxicol. Appl. Pharmacol. 203, 124–131. Li, P., Nijhawan, D., Budihardjo, I., Srinivasula, S.M., Ahmad, M., Alnemri, E.S., Wang, X.D., 1997. Cytochrome c and dATP-dependent formation of Apaf-1/caspase-9 complex initiates an apoptotic protease cascade. Cell 91, 479–489. Li, B., Chen, H., Sun, H.W., Lan, Z.H., 2017. Distribution, isomerization and enantiomer selectivity of hexabromocyclododecane (HBCD) diastereoisomers in different tissue and subcellular fractions of earthworms. Ecotox. Environ. Safe. 139, 326–334. Lilienthal, H., van der Ven, L.T.M., Piersma, A.H., Vos, J.G., 2009. Effects of the brominated flame retardant hexabromocyclododecane (HBCD) on dopamine-dependent

behavior and brainstem auditory evoked potentials in a one-generation reproduction study in Wistar rats. Toxicol. Lett. 185, 63–72. Marvin, C.H., Tomy, G.T., Armitage, J.M., Arnot, J.A., McCarty, L., Covaci, A., Palace, V., 2011. Hexabromocyclododecane: current understanding of chemistry, environmental fate and toxicology and implications for global management. Environ. Sci. Technol. 45, 8613–8623. Rawn, D.F.K., Gaertner, D.W., Weber, D., Curran, I.H.A., Cooke, G.M., Goodyer, C.G., 2014. Hexabromocyclododecane concentrations in Canadian human fetal liver and placental tissues. Sci. Total Environ. 468-469, 622–629. Reffatto, V., Rasinger, J.D., Carroll, T.S., Ganay, T., Lundebye, A.K., Sekler, I., Hershfinkel, M., Hogstrand, C., 2018. Parallel in vivo and in vitro transcriptomics analysis reveals calcium and zinc signalling in the brain as sensitive targets of HBCD neurotoxicity. Arch. Toxicol. 92, 1189–1203. Ruan, Y.F., Lam, J.C.W., Zhang, X.H., Lam, P.K.S., 2018a. Temporal changes and Stereoisomeric compositions of 1,2,5,6,9,10-hexabromocyclododecane and 1,2-dibromo-4(1,2-dibromoethyl)cyclohexane in marine mammals from the South China Sea. Environ. Sci. Technol. 52, 2517–2526. Ruan, Y.F., Zhang, X.H., Qiu, J.W., Leung, K.M.Y., Lam, J.C.W., Lam, P.K.S., 2018b. Stereoisomer-specific trophodynamics of the chiral brominated flame retardants HBCD and TBECH in a marine food web, with implications for human exposure. Environ. Sci. Technol. 52, 8183–8193. Ryan, J.J., Rawn, D.F.K., 2014. The brominated flame retardants, PBDEs and HBCD, in Canadian human milk samples collected from 1992 to 2005; concentrations and trends. Environ. Int. 70, 1–8. Sanders, J.M., Knudsen, G.A., Birnbaum, L.S., 2013. The fate of β-hexabromocyclododecane in female C57BL/6 mice. Toxicol. Sci. 134, 251–257. Shi, D.L., Lv, D.M., Liu, W.X., Shen, R., Li, D.M., Hong, H.Z., 2017. Accumulation and developmental toxicity of hexabromocyclododecanes (HBCDs) on the marine copepod Tigriopus japonicus. Chemosphere 167, 155–162. Szabo, D.T., Pathmasiri, W., Sumner, S., Birnbaum, L.S., 2017. Serum metabolomic profiles in neonatal mice following oral brominated flame retardant exposures to hexabromocyclododecane (HBCD) alpha, gamma, and commercial mixture. Environ. Health Perspect. 125, 651–659. Thomsen, C., Molander, P., Daae, H.L., Janak, K., Froshaug, M., Liane, V.H., Thorud, S., Becher, G., Dybing, E., 2007. Occupational exposure to hexabromocyclododecane at an industrial plant. Environ. Sci. Technol. 41, 5210–5216. Valdiglesias, V., Costa, C., Kiliç, G., Costa, S., Pasaro, E., Laffon, B., Teixeira, J.P., 2013. Neuronal cytotoxicity and genotoxicity induced by zinc oxide nanoparticles. Environ. Int. 55, 92–100. Wang, Y., Xu, L., Li, D.Z., Teng, M.M., Zhang, R.K., Zhou, Z.Q., Zhu, W.T., 2015. Enantioselective bioaccumulation of hexaconazole and its toxic effects in adult zebrafish (Danio rerio). Chemosphere 138, 798–805. Wang, F.D., Zhang, H.J., Geng, N.B., Zhang, B.Q., Ren, X.Q., Chen, J.P., 2016. New insights into the cytotoxic mechanism of hexabromocyclododecane from a metabolomic approach. Environ. Sci. Technol. 50, 3145–3153. Wang, X.L., Yang, J., Li, H., Guo, S., Tariq, M., Chen, H.B., Wang, C., Liu, Y.D., 2018. Chronic toxicity of hexabromocyclododecane (HBCD) induced by oxidative stress and cell apoptosis on nematode Caenorhabditis elegans. Chemosphere 208, 31–39. Weiss, J., Wallin, E., Axmon, A., Jonsson, B.A.G., Akesson, H., Janak, K., Hagmar, L., Bergman, A., 2006. Hydroxy-PCBs, PBDEs and HBCDDs in serum from an elderly population of Swedish fishermen's wives and associations with bone density. Environ. Sci. Technol. 40, 6282–6289. Xiang, N., Chen, L., Meng, X.Z., Dai, X.H., 2015. Occurrence of hexabromocyclododecane (HBCD) in sewage sludge from Shanghai: implications for source and environmental burden. Chemosphere 118, 207–212. Yu, K., He, Y.H., Yeung, L.W.Y., Lam, P.K.S., Wu, R.S.S., Zhou, B.S., 2008. DE-71-induced apoptosis involving intracellular calcium and the Bax-mitochondria-caspase protease pathway in human neuroblastoma cells in vitro. Toxicol. Sci. 104, 341–351. Zhang, X.L., Yang, F.X., Xu, C., Liu, W.P., Wen, S., Xu, Y., 2008. Cytotoxicity evaluation of three pairs of hexabromocyclododecane (HBCD) enantiomers on Hep G2 cell. Toxicol. in Vitro 22, 1520–1527. Zhang, Y.W., Guo, Q.Q., Tan, D.F., He, Z.Y., Wang, Y.H., Liu, X.W., 2018. Effects of low-levels of three hexabromocyclododecane diastereomers on the metabolic profiles of pak choi leaves using high-throughput untargeted metabolomics approach. Environ. Pollut. 242, 1961–1969. Zhu, J.Q., Xu, M., Gao, M., Zhang, Z.H., Xu, Y., Xia, T., Liu, S.J., 2017. Graphene oxide induced perturbation to plasma membrane and cytoskeletal meshwork sensitize cancer cells to chemotherapeutic agents. ACS Nano 11, 2637–2651.