Activation of integrated stress response and disordered iron homeostasis upon combined exposure to cadmium and PCB77

Journal of Hazardous Materials xxx (xxxx) xxxx

Contents lists available at ScienceDirect

Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat

Activation of integrated stress response and disordered iron homeostasis upon combined exposure to cadmium and PCB77 Lixin Wanga, Miaomiao Zhenga, Yingxue Wanga, Lin Yuanc, Chengyong Yuc, Jiansheng Cuia,*, Shuping Zhangb,* a College of Environmental Science and Engineering, Hebei University of Science and Technology, Pollution Prevention Biotechnology Laboratory of Hebei Province, Shijiazhuang, 050018, China b Institute for Medical Engineering and Science, Massachusetts Institute of Technology, Cambridge, MA, 02139, USA c Clinical Lab, Weihai Central Hospital, Weihai, 264400, China

G R A P H I C A L A B S T R A C T

A R T I C LE I N FO

A B S T R A C T

Editor: R. Debora

Both cadmium and polychlorinated biphenyls (PCBs) can induce diverse detrimental effects on human health. Though these compounds co-exist in various environmental contexts and in the human body, studies on their joint toxicities are limited. Activation of the integrated stress response (ISR) and iron homeostasis are crucial for erythropoiesis. The impact of cadmium and PCBs on the ISR activation and iron homeostasis of erythroid progenitors is unknown. We investigated the adverse effects and mechanisms of CdCl2 and PCB77 on HEL cells, a human cell model of erythroid progenitors. We found that at high concentrations of CdCl2 and PCB77, cytotoxicity and apoptosis of HEL cells were mainly induced by CdCl2. At low concentrations of CdCl2 and PCB77, iron homeostasis inside HEL cells was disturbed by both of these two compounds. Both CdCl2 and PCB77 activated ISR to combat stress, which at high concentration was mainly induced by ROS, leading to apoptosis, and at low concentration was partly induced by disordered iron homeostasis. The patterns of ISR activation and iron homeostasis disorder were different between CdCl2 and PCB77. Their combined exposure exhibited synergetic effect on activating ISR but antagonistic effect on disturbing iron homeostasis. Our study demonstrates some previously unrecognized harmful characteristics and mechanisms of cadmium and PCB77.

Keywords: Cadmium PCB77 Joint toxicity Integrated stress response Disordered iron homeostasis

1. Introduction Cadmium is a prevalent pollutant as a result of worldwide production and consumption of cadmium compounds (Rafati Rahimzadeh

⁎

et al., 2017). Substantial concerns have been raised by the rapid increased cadmium related diseases, including cancers, renal failure and bone fragility (Satarug et al., 2010). Cadmium is also well-known to induce anemia mainly via three mechanisms: hemolysis (Kunimoto

Corresponding authors. E-mail addresses: [email protected] (J. Cui), [email protected] (S. Zhang).

https://doi.org/10.1016/j.jhazmat.2019.121833 Received 9 September 2019; Received in revised form 3 December 2019; Accepted 4 December 2019 0304-3894/ © 2019 Elsevier B.V. All rights reserved.

Please cite this article as: Lixin Wang, et al., Journal of Hazardous Materials, https://doi.org/10.1016/j.jhazmat.2019.121833

Journal of Hazardous Materials xxx (xxxx) xxxx

L. Wang, et al.

induced lipid peroxidation (Włostowski et al., 2008). Cadmium and PCBs exhibit a synergistic adverse effect on the function of the thyroid by interfering with thyroxine homeostasis (Buha et al., 2013). Nonetheless, investigation of the joint toxicity of cadmium and PCBs is still limited, including PCB77, leading to an underestimation of their real toxicities. Although cadmium is a typical anemia inducer and PCBs have been reported to be correlated to the changes of RBC and hemoglobin levels in human, no study has been performed to investigate their combined potential toxicities on RBC production. Given the coexistence of cadmium and PCB77 in seafood and human body, investigation of the potential erythroid toxicities upon exposure to them is urgent. In erythroid lineage, reactive oxygen species (ROS) can activate HRI (heme-regulated eIF2αP kinase)-eIF2αP-ATF4 (activating transcription factor 4) signaling pathway which is the key mechanism to combat diverse stresses including oxidative stress (Suragani et al., 2012). Our previous study demonstrated the activation of HRI-eIF2αP-ATF4 pathway by CdCl2 in erythroid progenitors through ROS (Wang et al., 2013a). Under stress, induction of ATF4 by HRI-eIF2αP mainly occurred at translational level rather than at transcriptional level, and subsequently, induced ATF4 activates adaptive gene expression, Chop (C/EBP homologous protein) and Gadd34 (Growth arrest and DNAdamage-inducible 34), which further activate the expression of downstream antioxidant genes. These mechanisms are coined as integrated stress response (ISR) (Chen, 2014). Whether the ISR is activated and how the ISR is modulated by combined exposure to cadmium and PCB77 is unknown. Most of the iron inside the body is used for hemoglobin synthesis in erythropoiesis, the process that produces RBCs from erythroid progenitors, and globin synthesis is strictly tied to iron availability (Ganz and Nemeth, 2012a). Thus iron homeostasis is a crucial event in erythropoiesis. Cadmium can induce iron deficiency anemia by inhibiting iron uptake in duodenum, leading to a disruption in systemic iron homeostasis and a decreased iron supply for erythropoiesis in bone marrow (Hamilton and Valberg, 1974). However, the impact of cadmium on the iron homeostasis inside erythroid cells is still unknown. PCB77 has been also reported to disrupt systemic iron homeostasis through significantly suppressing hepatic hepcidin expression, leading to the decreased iron level in spleen but increased iron level in sera (Wang et al., 2013b). Similar to cadmium, whether PCB77 can disrupt iron homeostasis inside erythroid progenitors is also unknown. Interestingly, both iron deficiency and iron overload can induce ISR. Iron deficiency, which leads to heme deficiency, activates ISR to slow down globin synthesis to avoid heme-free globin inclusion in erythroid progenitors (Zhang et al., 2018). Iron overload can promote ROS production through the Fenton reaction (Dixon and Stockwell, 2013), leading to the activation of ISR as discussed above. Thus, we here investigated the joint toxicities of cadmium-chloride (CdCl2) and PCB77 on erythroid progenitors at both high and low concentrations. Moreover, we investigated the activation and roles of ISR and iron homeostasis inside HEL cells, a human cell model of erythroid progenitors, upon combined exposure of CdCl2 and PCB77.

et al., 1985), induction of iron deficiency through competing with iron absorption in the duodenum (Hamilton and Valberg, 1974) and impaired secretion of erythropoietin (Horiguchi et al., 2010). Our previous study also indicated that cadmium significantly induced cell death and inhibited erythroid differentiation (Wang et al., 2013a). Polychlorinated biphenyls (PCBs), a class of persistent organic pollutants (POPs), have been extensively used due to their good thermal stability, chemical inertness and electrical resistivity, leading to widespread distribution in environment (Correa et al., 2010; Mlynarczuk et al., 2010). At present, PCBs still exist in various environments even with legislation and restrictions on its applications. As revealed by a survey on the indoor air in Switzerland, the most abundant coplanar (dioxin-like) PCB congeners were PCB118 followed by PCB105, PCB123 and PCB77 (Kohler et al., 2002). Another survey on ambient air, vegetation and soil samples in southern Taiwan also revealed that PCB118, PCB105, PCB77 and PCB156 dominated the profile patterns of coplanar PCBs in all three types of samples (Kuo et al., 2015). PCBs have been reported to induce adverse effects to many organs and systems, such as the liver, neurons, the immune system, the endocrine system and the reproductive system (Arisawa et al., 2005; DeRosa et al., 1998; Huisman et al., 1995; Park et al., 2010). Among the 12 coplanar congeners of PCBs, 3,3′,4,4′-tetrachlorobiphenyl (PCB77) has been extensively shown to induce different adverse effects, including anemia and arrhythmia in salmon (Olufsen and Arukwe, 2011), hepatotoxicity (Ghosh et al., 2010), nephrotoxicity (Ghosh et al., 2010; Su et al., 2015), disrupted thyroid hormone homeostasis, reproductive toxicity (Roelens et al., 2005; Desaulniers et al., 1997), and adverse effects on offspring (Cummings et al., 2005). An epidemiological investigation in the U.S.A. revealed an inverse association between serum level of PCBs and red blood cell (RBC) indices, including RBC counts, hemoglobin level and hematocrit (Serdar et al., 2014). In contrast, PCB concentrations were shown to be positively correlated with hemoglobin levels of residents near an electronic waste dismantling facility in China (Xu et al., 2015a). Although the results from these limited studies were contradictory, the studies agree that exposure to PCBs can induce adverse effect on RBCs. Whether PCBs can induce toxicity to erythroid progenitors, the precursors of RBCs, is still unknown. Cadmium is a heavy metal that widely co-exists with PCBs in river sediments (Dias-Ferreira et al., 2016; Zhu et al., 2017), edible aquatic organisms (Hoogenboom et al., 2015; Karl et al., 2016; Cano-Sancho et al., 2015; Ezemonye, 2018), and even in the urine (Dereumeaux et al., 2016), peripheral blood (Petrosino et al., 2018), cord blood and placenta (Xu et al., 2015b) of human. Together with other PCB congeners, PCB77 was reported to be co-found with cadmium in four types of seafood, including sardine, canned tuna, salmon and mussels, from five European countries (Cano-Sancho et al., 2015). Cadmium and PCBs, including PCB77, PCB105 and PCB114, were also reported to coexist in fish from three sites along the stretch of Ovia River, Southern Nigeria at concentrations above EU recommended guideline values for food safety, and both of them contributed to the toxic burden (Ezemonye, 2018). Heavy metals and PCBs, among which cadmium and PCB77 were included, were shown in the urinary samples from pregnant women who gave birth in France in 2011 (Dereumeaux et al., 2016). Studying the joint toxicities of pollutants, which represent the real environmental conditions, may assist in the determination of real toxicological effects of pollutants (Curcic et al., 2014). So far, there have been several well-designed studies on the mixture toxicology of heavy metals and persistent organic chemicals (Bandele et al., 2012; Buha et al., 2013; Zhang et al., 2016). The co-existence of cadmium and PCBs leads to combined exposure and potential joint toxic effects, such as impaired detoxification capability and disrupted thyroxine homeostasis (Buha et al., 2013; Włostowski et al., 2008). Co-treatment of PCBs was reported to suppress the expression of metallothionein, a protein to reduce toxicities from heavy metal exposure, in the kidney and liver of cadmium-treated bank voles. In contrast, cadmium decreased PCB-

2. Materials and methods 2.1. Cells and reagents Human erythroleukemia cell lines (HEL) was purchased from the Shanghai Cell Bank of Type Culture Collection. Cells were cultured following the standard protocol. Cadmium-Chloride (CdCl2xH2O, purity 99.99 %) and PCB77 (purity N 98.0 %) were purchased from Guangfu Technology Co., Ltd., Tianjin, China and J&K Scientific Ltd., Shanghai, China, respectively. CdCl2 and PCB77 were dissolved in ultrapure water and dimethyl sulphoxide (DMSO), respectively, and further dissolved in cell culture medium to create the working solution. DMSO concentration was less than 0.1 %. 2

Journal of Hazardous Materials xxx (xxxx) xxxx

L. Wang, et al.

interaction between CdCl2 and PCB77 in decreasing cell viability upon combined exposure (P > 0.05). These results indicated that the joint cytotoxicity of Cd and PCB77 was mainly induced by Cd and was not further aggravated by the co-treatment with PCB77.

2.2. Analysis on cell viability, membrane integrity, mitochondrial membrane potential, ROS production and cell apoptosis Cells were starved with 1 % FBS overnight for synchronization at a density of 4.0 × 103 cells per well in a dark 96-well plate. Then, cells were treated with different concentrations of PCB77 or CdCl2 followed by cell viability assay using Alamar Blue assay kit, ROS production was determined using 2′,7′-dichlorofluorescein diacetate (DCFH-DA) assay kit and mitochondrial membrane potential was assessed using 5,5′,6,6′tetrachloro-1,1′,3,3′-tetraethyl-benzimidazolyl carbocyanine iodide (JC-1) assay kit as previously described (Wang et al., 2018).The lactic dehydrogenase (LDH) content in culture medium was measured as a proxy for membrane integrity using an assay kit (Invitrogen, USA) following the instructions of the manufacturer. Cell apoptosis was determined by flow cytometry using FITC-Annexin V and PI according to the manufacturer instruction (BD Biosciences).

3.2. Changes of membrane permeability and mitochondrial membrane potential upon exposure to CdCl2 and PCB77 Decreased cell viability, as observed above, might correlate to injuries in the cell membrane. To detect damage to the cell membrane upon exposure to CdCl2 and PCB77, we assayed membrane permeability by measuring the release of lactate dehydrogenase (LDH). Compared to the control group, the levels of LDH in cell culture medium were significantly increased by 17.9 % and 48.1 % by treatment with CdCl2 at 15 μM and 40 μM, respectively (Fig. 2A). LDH level was not changed when cells were treated with PCB77 at all concentrations. After co-treatment with CdCl2 and PCB77, LDH levels in HEL cells were significantly increased at 15 μM and particularly 40 μM CdCl2, and were independent of the concentrations of PCB77, respectively (Fig. 2A). No significant interaction between CdCl2 and PCB77 was found in increasing LDH level upon combined exposure (P > 0.05). These results indicated that, similar to the results of decreased cell viability (Fig. 1E), the joint damage of CdCl2 and PCB77 to cell membrane mainly depended on CdCl2. Mitochondrial membrane potential (MMP) as measured by JC-1 assay was significantly decreased relative to the control group with single or combined treatment of CdCl2 and PCB77 at various concentrations, except combined treatments of 1 μM CdCl2 with 15 μM and 40 μM PCB77 (Fig. 2B). This observation suggests that mitochondrial damage can be induced at lowest concentrations even without occurrence of cell injury (Figs. 1 and 2A). Specifically, MMP was significantly decreased by 47.8 %, 47.4 % and 52.8 % in cells with combined exposure of CdCl2 at 40 μM and PCB77 at 1 μM, 15 μM and 40 μM as compared to the control group, respectively (Fig. 2B). More importantly, we observed significant synergistic effect of CdCl2 and PCB77 in decreasing MMP level upon combined exposure (P = 0.001). A previous study suggested that cell survival was decreased in rat primary neuron and glial cells 24 h after treatment with cadmium at 1.25 μM, as revealed by an MTT assay (Yang et al., 2007). The MTT assay is based on detection of succinate dehydrogenase activity, which located in the mitochondrial intima. Similarly, our results showed that mitochondrial membrane potential was significantly decreased in HEL cells with CdCl2 treatment at 1 μM (Fig. 2B). However, no impact on cell activity and cell membrane permeability was observed after CdCl2 treatment at 1 μM (Fig. 1A and 2A). Similar results were obtained after PCB77 treatment at 1 μM, consistent with a previous report that no cell viability changes were observed in mouse thymocytes exposed to PCB77 at 1 μM (Yilmaz et al., 2006).

2.3. RT-qPCR and Western blot analysis Gene expression were measured using SYBR Green qPCR master mix (Takara, Japan) on qPCR Systems (MyGO Pro, England) with primers as listed in Table 1. eIF2α was used as internal control for normalization. Protein content was measured by Western blot analysis as previously described (Wang et al., 2013a). Antibodies were anti-eIF2α (1:1000) from GeneTex, USA, anti-FTH1 (1:500), anti-S6 (1:500) and anti-phosphorylated S6 (1:500) from Affinity Biosciences, USA. Density of each band was quantified using ImageJ software (NIH). 2.4. Statistical analysis Independent t-test was used for difference analysis. Two-way ANOVA analysis was performed to determine the interaction between CdCl2 and PCB77 upon combined exposure. Data were presented as mean ± SEM and P < 0.05 was considered statistically significant. 3. Results and discussion 3.1. Cell viability upon exposure to CdCl2 and PCB77 We first investigated the cytotoxicity of CdCl2 and PCB77 in HEL cells by assessing the cell viability. Cell viability was significantly decreased by 40 % after 48 h exposure to CdCl2 at 40 μM, but not at 1 μM and 15 μM, as compared to the control group (Fig. 1A). No impact on the cell viability was induced by PCB77 at all concentrations (Fig.1B). To investigate the joint toxicity induced by these two pollutants, HEL cells were treated with CdCl2 and PCB77 at all combinations of the various concentrations. When the concentration of CdCl2 was 1 μM or 15 μM, no toxicity was induced by the co-treatment with PCB77 at all concentrations as compared to the control group (Fig. 1C–D). However, cell viability was significantly decreased by the co-treatment of CdCl2 at 40 μM and PCB77 at various concentrations (Fig. 1E). It is worth noting that there was no significant difference in cell viability between treatment with 40 μM CdCl2 and combined treatments of 40 μM CdCl2 and PCB77 at various concentrations. As expected, there was no significant

3.3. Cell apoptosis mediated by over-activated ISR upon exposure to CdCl2 and PCB77 at high concentration Upon combined exposure, interactions between pollutants depend on the number of pollutant molecules in the mixture and their toxicities (Curcic et al., 2017). In this study, we aimed to explore whether the toxic effects of cadmium on erythroid progenitors, with or without impairing cell survival, can be enhanced by combined treatment with the same molar concentration of PCB77. Since significant decreased cell viability and increased membrane permeability can be detected only upon combined exposure of the highest concentration of CdCl2 with various concentrations of PCB77, we next investigated the underlying mechanisms of the toxicities induced by the combined treatment with CdCl2 and PCB77 both at 40 μM. Previously, we identified that oxidative stress was a major mediator for the toxicity to erythroid precursors by cadmium (Wang et al., 2013a). As shown in Fig. 3A, ROS levels were significantly increased by 7.6 %, 4.5 % and 9.8 % by CdCl2, PCB77 and

Table 1 List of primers for RT-qPCR. Gene

Forward (5′-3′)

Reverse (5′-3′)

eIF2α Atf4 Chop Gadd34 Nqo1 Gstp1 Sod2

GAAGGCGTATCCGTTCTATCAAC CTCCGGGACAGATTGGATGTT CTGCCTTTCACCTTGGAGAC GAGGGACGCCCACAACTTC GAAGAGCACTGATCGTACTGGC CCCTACACCGTGGTCTATTTCC GGAAGCCATCAAACGTGACTT

AGCAACATGACGAAGAATGCTAT GGCTGCTTATTAGTCTCCTGGAC CGTTTCCTGGGGATGAGATA TTACCAGAGACAGGGGTAGGT GGATACTGAAAGTTCGCAGGG CAGGAGGCTTTGAGTGAGC CCCGTTCCTTATTGAAACCAAGC

3

Journal of Hazardous Materials xxx (xxxx) xxxx

L. Wang, et al.

Fig. 1. Cell viability of HEL cells with the treatment of CdCl2 or/and PCB77 at various concentrations. Cell viability was determined by Alamar Blue assay after treatment with CdCl2 (A), PCB77 (B), or the co-treatment of CdCl2 and PCB77 (C–E) for 48 h (n = 4). asignificantly different from the untreated cells; c- significantly different from the treatment of PCB77 at same concentration. P < 0.05 was considered statistically significant. Fig. 2. The changes of LDH and MMP levels of HEL cells upon exposure to CdCl2 and PCB77. (A) LDH concentration in cell culture medium, and (B) MMP levels in cells treated with CdCl2 and PCB77 for 48 h (n = 4). a- significantly different from the control group; b- significantly different from CdCl2-treated group; c- significantly different from PCB77-treated group. Data were presented in mean ± SE. P < 0.05 was considered statistically significant.

combined treatment of CdCl2 and PCB77 (Fig. 3A), indicating the aggravated cellular oxidative stress upon combined exposure. However, there was no significant interaction between CdCl2 and PCB77 in

inducing ROS upon combined exposure (P > 0.05). Next, we investigated the activation of ATF4-mediated ISR in HEL cells in response to oxidative stress induced by treatment of CdCl2 and 4

Journal of Hazardous Materials xxx (xxxx) xxxx

L. Wang, et al.

Fig. 3. Over-activated ISR induced cell apoptosis upon exposure to CdCl2 and PCB77 at high dose. (A) Intracellular ROS production in HEL cells, as reflected by DCF fluorescent intensity, after single or combined treatment of CdCl2 and PCB77 at 40 μM for 1.5 h (n = 3–4). (B) Relative expression levels of Atf4, Chop, Gadd34, NqO1, Sod2 and Gstp1 mRNAs after single or combined treatment of CdCl2 and PCB77 at 40 μM for 3 h (n = 3–4). (C) Representative density plots (left panel) and quantitative analysis (right panel) of cell apoptosis of HEL cells by flow cytometer analysis after single or combined treatment of CdCl2 and PCB77 at 40 μM for 48 h (n = 3–4). a- significantly different from the control group; b-significantly different from CdCl2-treated group; c-significantly different from PCB77-treated group. Data were presented in mean ± SE. P < 0.05 was considered statistically significant.

which demonstrated that induction of ATF4 downstream antioxidant genes is necessary to mitigate oxidative stress upon exposure to arsenite (Suragani et al., 2012). These results indicate that exposure to CdCl2 and PCB77 at high concentration induces extreme stress and activates ISR in HEL cells to inhibit ROS generation. However, over-activation of ISR may lead to another cell fate, mitochondrial-mediated apoptosis (Chen, 2014). Since decreased MMP was observed after treatment with CdCl2 and PCB77 (Fig. 2B), we next investigated cell apoptosis. As shown in Fig. 3C, a large number of apoptotic HEL cells were observed when treated with CdCl2 as compared to the control (4.6-fold, p < 0.05), similar to our previous finding (Wang et al., 2013a). However, no significant increase of apoptotic cells were observed when treated with PCB77. The percentage of apoptotic cells was not further increased upon combined exposure to CdCl2 and PCB77 as compared to CdCl2 treatment (Fig. 3C). All these data indicate that ISR was activated to combat oxidative stress upon exposure to CdCl2 and PCB77. However, activated ISR failed to prevent CdCl2-induced apoptosis. We did not observe increased cell membrane permeability (Fig. 2A) and apoptosis of HEL cells upon exposure to PCB77 at 40 μM. However, PCB77 exposure at 40 μM increased ROS production and activated ISR, which prevented cell apoptosis. This observation indicated that ROS production was not the only inducer for cell apoptosis upon exposure to CdCl2. A previous study reported that cadmium can activate calpain, a calcium (Ca2+)dependent cysteine protease, in HepG2 cells (Lawal et al., 2015). Activated calpain cleaves caspase 3 leading to the activation of

PCB77. Significant up-regulated expression was observed for Chop and Gadd34 mRNAs upon single exposure to CdCl2 (3.5- and 3.9-fold, respectively) and particularly combined exposure to CdCl2 and PCB77 (7.4- and 9.6-fold, respectively) (Fig. 3B, left panel). No significant change in expression level was observed for Atf4 mRNA. Induced expression levels were low upon single exposure to PCB77 for all these three genes (Fig. 3B, left panel). Consequently, CHOP and GADD34 induced three antioxidant genes: Nqo1 (NAD(P)H quinone oxidoreductase 1), Sod2 (Superoxide dismutase 2) and GstP1 (Glutathione Stranferase) (Fig. 3B, middle and right panels). Combined treatment of CdCl2 and PCB77 up-regulated expression of Nqo1, Sod2 and GstP1 to 1.6-, 1.5- and 233-fold, respectively (Fig. 3B, middle and right panels). Single treatment with CdCl2 significantly induced Nqo1 and GstP1 expression to 2- and 11.5-fold, respectively, but induced no changes in Sod2. Single treatment with PCB77 only significantly induced GstP1 expression to 18.4-fold (Fig. 3B, middle and right panels). Significant synergistic effects of CdCl2 and PCB77 were observed in inducing the expression of Chop, Gadd34 and GstP1 mRNAs upon combined exposure (P = 0.003, P < 0.001 and P < 0.001, respectively). Similar to our result, the expression of Nqo1 mRNA was significantly induced by cadmium treatment for 6 or 24 h in murine macrophages (Jin et al., 2016) and by cadmium treatment for 24 h in human astrocytoma cells (Lawal and Ellis, 2011). No induction of Nqo1 expression in PCB77treated HEL cells and no enhanced expression of Nqo1 mRNA by PCB77 treatment in CdCl2-treated HEL cells were also consistent with the result as observed in 3T3-L1 adipocytes (Baker et al., 2013). More importantly, findings from our study were consistent to a previous report 5

Journal of Hazardous Materials xxx (xxxx) xxxx

L. Wang, et al.

Fig. 4. Mild-activated ISR protected cells from apoptosis upon exposure to CdCl2 and PCB77 at low dose. (A) Relative expression levels of Atf4, Chop, Gadd34, NqO1, Sod2 and Gstp1 mRNAs after single or combined treatment with CdCl2 and PCB77 at 1 μM for 3 h (n = 3–4). (B) Representative density plots (left panel) and quantitative analysis (right panel) of cell apoptosis of HEL cells by flow cytometer analysis after single or combined treatment of CdCl2 and PCB77 at 1 μM for 48 h (n = 3–4). Data were presented in mean ± SE. P < 0.05 was considered statistically significant.

of HEL cells was observed after single or combined treatment with CdCl2 and PCB77 at low concentration (Fig. 4B).

mitochondria-dependent apoptotic pathways (Lee et al., 2006, 2007). Since cadmium (Cd2+) has the same ionic radii as Ca2+, Cd2+ can mimic Ca2+ function (Lawal et al., 2015) or disturb intracellular Ca2+ homeostasis (Lawal and Ellis, 2012). Thus, calpain activation might be involved in cell apoptosis induced by cadmium at 40 μM in addition to ROS production. PCB77 might be unable to activate calpain due to being incapable of mimicking Ca2+ function, which we observed in SMMC-7721 cells after treatment with PBDE-209 (Wang et al., 2018). Nonetheless, our results showed a synergetic effect of CdCl2 and PCB77 on inducing ISR activation. Unexpectedly, enhanced ISR by co-treatment of PCB77 and CdCl2 did not further increase of cell apoptosis as compared to CdCl2 treatment alone.

3.5. Disordered iron homeostasis and inhibited mTORC1 activity upon exposure to CdCl2 and PCB77 at low concentration Although no significant increase of membrane permeability (Fig. 2A) and cell apoptosis (Fig. 4B) occurred in HEL cells upon exposure to CdCl2 and PCB77 at low concentration, cellular processes and cell functions might be still disturbed. Therefore, we assessed cellular iron availability, as measured by the labile iron pool (LIP), and iron storage, as measured by ferritin protein level. The LIP is a pool of lowaffinity ligands, to which iron binds upon being ready for utilization. Thus, the LIP is regarded as the crossroads of iron traffic. The maturation of erythroid progenitors is associated with a decreased in LIP level, demonstrating the demand for more iron during erythroid differentiation (Prus and Fibach, 2008). Ferritin is a protein which stores cellular iron and consists of 24 polypeptide subunits of light chain (FTL) and heavy chain (FTH1) (Baraibar et al., 2008). We first determined the relative LIP level in HEL cells after treatment with CdCl2 and PCB77 at low concentration using a cell permeable iron-chelator, calcein acetoxymethyl ester (CA-AM). LIP level, as reflected by the relative fluorescence intensity of CA-AM, was significantly increased to 1.2-fold and 1.6-fold after treatment with CdCl2 and PCB77 at 1 μM, respectively, as compared to the control group (Fig. 5A). Combined treatment with CdCl2 and PCB77 also increased LIP level to 1.4-fold, a smaller increase than that after treatment with PCB77 alone, as compared to the control group. However, CdCl2 and PCB77 exhibited a significant antagonistic effect in disturbing LIP level upon combined exposure (P = 0.038), which might explain the decreased LIP level as compared to PCB77 treatment. Then, we measured FTH1 protein content to test for changes in iron storage after treatment with CdCl2 and PCB77 at 1 μM. As shown in Fig. 5B, the protein level of FTH1 was decreased in HEL cells after CdCl2 treatment but increased after PCB77 treatment, as compared to the control cells. Combined treatment with CdCl2 and PCB77 also decreased FTH1 protein level when compared to the control. Although no disordered iron homeostasis was found in HEL cells after single or combined treatments with

3.4. Prevention of cell apoptosis mediated by mild-activated ISR upon exposure to CdCl2 and PCB77 at low concentration Next we investigated whether ISR is activated to combat potential oxidative stress upon single or combined exposure to CdCl2 and PCB77 at low concentration. No obvious increase of ROS level was observed with single or combined treatment to CdCl2 and PCB77 at 1 μM (data not shown). As shown in Fig. 4A, the expression of Atf4, Chop, Gadd34, NqO1 and Sod2 mRNAs in HEL cells was slightly up-regulated by single or combined treatment with CdCl2 and PCB77 at 1 μM as compared to the control group. However, the expression of Gstp1 mRNA was significantly up-regulated to 1.5-, 14.3- and 15.7-fold by CdCl2, PCB77 and combined treatment with CdCl2 and PCB77, respectively (Fig. 4A). Nonetheless, no significant interaction between CdCl2 and PCB77 was observed in regulating Gstp1 expression upon combined exposure (P > 0.05). GSTs are a family of enzymes important for detoxification through catalyzing the conjugation of electrophilic and hydrophobic compounds, such as PCBs, and reduced glutathione. It should be noted that we did not observe ISR activation in HEL cells with single or combined treatments with CdCl2 and PCB77 at 0.01 μM (data not shown) and 0.1 μM (Supplementary Fig. 1A). Taken together, these data indicated that mild activation of ISR was induced and able to inhibit ROS generation when cells were treated by CdCl2 and PCB77 at low concentration. The expression of most ISR genes recovered to the basal levels. Consequently, no obvious apoptosis 6

Journal of Hazardous Materials xxx (xxxx) xxxx

L. Wang, et al.

Fig. 5. Disordered iron homeostasis and inhibited mTORC1 activity upon exposure to CdCl2 and PCB77 at low dose. (A) The relative fluorescence intensity of CAeAM for measuring LIP to reflect intracellular iron availability (n = 3–4), and (B) Representative blots of FTH1 protein content to reflect iron storage. Analyses were performed after single or combined exposure to CdCl2 and PCB77 at 1 μM for 48 h. (C) Phosphorylated S6 and total S6 content to reflect mTORC1 activity as measured by Western blot. Ratio of FTH1 to eIF2αP and ratio of pS6 to S6 in the control group were defined as 1. Analyses were performed after single or combined exposure to CdCl2 and PCB77 at 1 μM for 48 h. a- significantly different from the control group. Data were presented in mean ± SE. P < 0.05 was considered statistically significant.

CdCl2 and PCB77 at 0.01 μM (data not shown), significant increase of LIP levels was observed by both single and combined treatments with CdCl2 and PCB77 at 0.1 μM (Supplementary Fig. 1B), while no change was found for FTH1 protein content. LIP serves as a pool of iron which is ready for utilization and FTH1 is one of the subunits of the protein for iron storage. LIP level would be much more sensitive for fluctuation of cellular iron level than FTH1 content. All these results demonstrate that iron homeostasis is disturbed upon exposure to CdCl2 and PCB77 at low concentration. More specifically, these observations indicate that CdCl2 might inhibit iron absorption, leading to iron release into the LIP via the degradation of ferritin in HEL cells to maintain the iron supply, or enhance iron utilization of HEL cells, leading to ferritin degradation and the release of iron into the LIP to fulfill the increased demand for iron. As mentioned above, Cd2+ can mimic iron (Fe2+) and thus compete with iron uptake, which might be one of the mechanisms underlying the interference with iron homeostasis in HEL cells. Cadmium was also shown to promote exporting iron in macrophages through increasing the translation of ferroportin, the only known iron exporter in vertebrate cells (Sun et al., 2015). Macrophages, particularly those located in the liver and spleen, are the main iron supplier for erythroid progenitors in bone marrow, and therefore are important for systemic iron homeostasis (Ganz and Nemeth, 2012b). Although it warrants further investigation, induction of ferroportin translation by cadmium might not be one of the underlying mechanisms for the disordered iron hemostasis in HEL cells as observed here. Instead, PCB77 might increase iron absorption or inhibit iron utilization in HEL cells, leading to increased iron levels in both LIP and ferritin and causing the potential for iron overload. As expected, disordered iron homeostasis was attenuated upon combined exposure to CdCl2 and PCB77 as compared to single treatment of CdCl2 or PCB77. Recently, we reported that ISR activated by HRI-eIF2αP-ATF4 signaling pathway inhibits the mTORC1 (mammalian target of rapamycin complex 1) signaling pathway to coordinate hemoglobin synthesis and iron availability (Zhang et al., 2018). mTORC1 signaling pathway is an important mechanism of regulation at translational level in response to amino acid deficiency during starvation. Since activation of ISR and disordered iron homeostasis were observed above, we further investigated the activity of mTORC1 signaling pathway by measuring phosphorylation of S6. As shown in Fig. 5C, the ratio of pS6 to S6 was decreased by CdCl2, PCB77 and combined treatment of CdCl2 and PCB77 as compared to the control, respectively. These results indicated that global translation of mRNAs, mostly hemoglobin mRNAs in HEL cells, was inhibited due to disordered iron homeostasis upon exposure to CdCl2 and PCB77.

4. Conclusion In conclusion, our findings showed that ISR was activated by both cadmium and PCB77 to combat stress, as mainly induced by ROS at high concentrations of cadmium and PCB77 and partly by disordered iron homeostasis inside erythroid progenitors at low concentrations. Cadmium and PCB77 exhibited different patterns on inducing both ISR and disordered iron homeostasis. Combined exposure to cadmium and PCB77 synergistically activated ISR, but antagonistically disrupted iron homeostasis inside erythroid progenitors. Our study demonstrated the new characteristics and mechanisms of the detrimental effects of cadmium and PCB77. Authorship contributions L. W., J. C. and S. Z. designed the experiments, and wrote the paper. L. W., M. Z. and Y. W. performed and analyzed the experiment data. L. Y. and C. Y. revised the manuscript. Declaration of Competing Interest The authors declare that there is no conflict of interest. Acknowledgments This research was supported by projects of the National Natural Science Foundation of China (21407040), Program for the Top Youngaged Talents of Higher Learning Institutions of Hebei (BJ2019033). Appendix A. Supplementary data Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.jhazmat.2019.121833. References Arisawa, K., Takeda, H., Mikasa, H., 2005. Background exposure to PCDDs/PCDFs/PCBs and its potential health effects: a review of epidemiologic studies. J. Med. Investig. 52 (1,2), 10–21. Baker, N.A., et al., 2013. Resveratrol protects against polychlorinated biphenyl-mediated impairment of glucose homeostasis in adipocytes. J. Nutr. Biochem. 24 (12), 2168–2174. Bandele, O.J., et al., 2012. In vitro toxicity screening of chemical mixtures using HepG2/ C3A cells. Food Chem. Toxicol. 50 (5), 1653–1659. Baraibar, M.A., et al., 2008. Iron-mediated aggregation and a localized structural change characterize ferritin from a mutant light chain polypeptide that causes neurodegeneration. J. Biol. Chem. 283 (46), 31679–31689. Buha, A., et al., 2013. The impact of prolonged cadmium exposure and co-exposure with polychlorinated biphenyls on thyroid function in rats. Toxicol. Lett. 221 (2), 83–90.

7

Journal of Hazardous Materials xxx (xxxx) xxxx

L. Wang, et al.

induced mitochondrial-caspase 3- dependent apoptosis in human hepatoma cell line. BMC Pharmacol. Toxicol. 16, 41. Lee, W.K., Abouhamed, M., Thevenod, F., 2006. Caspase-dependent and -independent pathways for cadmium-induced apoptosis in cultured kidney proximal tubule cells. Am. J. Physiol. Renal Physiol. 291 (4), F823–32. Lee, W.K., Torchalski, B., Thevenod, F., 2007. Cadmium-induced ceramide formation triggers calpain-dependent apoptosis in cultured kidney proximal tubule cells. Am. J. Physiol., Cell Physiol. 293 (3), C839–47. Mlynarczuk, J., Wrobel, M.H., Kotwica, J., 2010. Effect of environmental pollutants on oxytocin synthesis and secretion from corpus luteum and on contractions of uterus from pregnant cows. Toxicol. Appl. Pharmacol. 247 (3), 243–249. Olufsen, M., Arukwe, A., 2011. Developmental effects related to angiogenesis and osteogenic differentiation in Salmon larvae continuously exposed to dioxin-like 3,3′,4,4′-tetrachlorobiphenyl (congener 77). Aquat. Toxicol. 105 (3), 669–680. Park, H.-Y., et al., 2010. Neurodevelopmental toxicity of prenatal polychlorinated biphenyls (PCBs) by chemical structure and activity: a birth cohort study. Environ. Health 9 (1), 51. Petrosino, V., et al., 2018. The role of heavy metals and polychlorinated biphenyls (PCBs) in the oncogenesis of head and neck tumors and thyroid diseases: a pilot study. BioMetals 31 (2), 285–295. Prus, E., Fibach, E., 2008. The labile iron pool in human erythroid cells. Br. J. Haematol. 142 (2), 301–307. Rafati Rahimzadeh, M., et al., 2017. Cadmium toxicity and treatment: an update. Caspian J. Intern. Med. 8 (3), 135–145. Roelens, S.A., et al., 2005. The dioxin-like PCB 77 but not the ortho-substituted PCB 153 interferes with chicken embryo thyroid hormone homeostasis and delays hatching. Gen. Comp. Endocrinol. 143 (1), 1–9. Satarug, S., et al., 2010. Cadmium, environmental exposure, and health outcomes. Environ. Health Perspect. 118 (2), 182–190. Serdar, B., et al., 2014. Potential effects of polychlorinated biphenyls (PCBs) and selected organochlorine pesticides (OCPs) on immune cells and blood biochemistry measures: a cross-sectional assessment of the NHANES 2003-2004 data. Environ. Health 13, 114. Su, K., et al., 2015. PCB77 inducing renal tubular cell apoptosis. Ultrastruct. Pathol. 39 (3), 192–197. Sun, L., et al., 2015. Cadmium depletes cellular iron availability through enhancing ferroportin translation via iron responsive element. Mol. Med. Rep. 11 (4), 3129–3133. Suragani, R.N.V.S., et al., 2012. Heme-regulated eIF2α kinase activated Atf4 signaling pathway in oxidative stress and erythropoiesis. Blood 119 (22), 5276–5284. Wang, L., et al., 2018. In vitro study on the joint hepatoxicity upon combined exposure of cadmium and BDE-209. Environ. Toxicol. Pharmacol. 57, 62–69. Wang, L., et al., 2013a. A protective role of heme-regulated eIF2α kinase in cadmiuminduced toxicity in erythroid cells. Food Chem. Toxicol. 62, 880–891. Wang, L., et al., 2013b. PCB-77 disturbs iron homeostasis through regulating hepcidin gene expression. Gene 532 (1), 146–151. Włostowski, T., Krasowska, A., Bonda, E., 2008. Joint effects of dietary cadmium and polychlorinated biphenyls on metallothionein induction, lipid peroxidation and histopathology in the kidneys and liver of bank voles. Ecotoxicol. Environ. Saf. 69 (3), 403–410. Xu, P., et al., 2015a. Effects of PCBs and PBDEs on thyroid hormone, lymphocyte proliferation, hematology and kidney injury markers in residents of an e-waste dismantling area in Zhejiang, China. Sci. Total Environ. 536, 215–222. Xu, X., et al., 2015b. Associations of cadmium, bisphenol A and polychlorinated biphenyl co-exposure in utero with placental gene expression and neonatal outcomes. Reprod. Toxicol. 52, 62–70. Yang, Z., et al., 2007. Cadmium-induced toxicity in rat primary mid-brain neuroglia cultures: role of oxidative stress from microglia. Toxicol. Sci. 98 (2), 488–494. Yilmaz, B., et al., 2006. Effects of PCB 52 and PCB 77 on cell viability, [Ca(2+)](i) levels and membrane fluidity in mouse thymocytes. Toxicology 217 (2-3), 184–193. Zhang, W., et al., 2016. High throughput sequencing analysis of the joint effects of BDE209-Pb on soil bacterial community structure. J. Hazard. Mater. 301, 1–7. Zhang, S., et al., 2018. HRI coordinates translation by eIF2αP and mTORC1 to mitigate ineffective erythropoiesis in mice during iron deficiency. Blood 131 (4) p. 450-450. Zhu, J., et al., 2017. Graphene oxide induced perturbation to plasma membrane and cytoskeletal meshwork sensitize Cancer cells to chemotherapeutic agents. ACS Nano 11 (3), 2637–2651.

Cano-Sancho, G., et al., 2015. Integrated risk index for seafood contaminants (IRISC): pilot study in five European countries. Environ. Res. 143, 109–115. Chen, J.J., 2014. Translational control by heme-regulated eIF2alpha kinase during erythropoiesis. Curr. Opin. Hematol. 21 (3), 172–178. Correa, P.A., et al., 2010. The effects of individual PCB congeners on the soil bacterial community structure and the abundance of biphenyl dioxygenase genes. Environ. Int. 36 (8), 901–906. Cummings, J.A., Nunez, A.A., Clemens, L.G., 2005. A cross-fostering analysis of the effects of PCB 77 on the maternal behavior of rats. Physiol. Behav. 85 (2), 83–91. Curcic, M., et al., 2014. Cadmium and decabrominated diphenyl ether mixture: in vitro evaluation of cytotoxic, prooxidative and genotoxic effects. Environ. Toxicol. Pharmacol. 38 (2), 663–671. Curcic, M., et al., 2017. Interactions between cadmium and decabrominated diphenyl ether on blood cells count in rats-Multiple factorial regression analysis. Toxicology 376, 120–125. Dereumeaux, C., et al., 2016. Biomarkers of exposure to environmental contaminants in French pregnant women from the Elfe cohort in 2011. Environ. Int. 97, 56–67. DeRosa, C., et al., 1998. Environmental exposures that affect the endocrine system: public health implications. J. Toxicol. Environ. Health Part B 1 (1), 3–26. Desaulniers, D., et al., 1997. Reproductive and thyroid hormone levels in rats following 90-Day dietary exposure to pcb 28 (2,4,4’-Trichlorobiphenyl) or pcb 77 (3,3’,4,4’Tetrachlorobiphenyl). Toxicol. Ind. Health 13 (5), 627–638. Dias-Ferreira, C., et al., 2016. Heavy metal and PCB spatial distribution pattern in sediments within an urban catchment—contribution of historical pollution sources. J. Soils Sediments 16 (11), 2594–2605. Dixon, S.J., Stockwell, B.R., 2013. The role of iron and reactive oxygen species in cell death. Nat. Chem. Biol. 10, 9. Ezemonye, I.Ta.L., 2018. Levels and human health risk assessment of persistent organics, metalloid and heavy metals in fish. Niger. J. Basic Appl. Sci. 26 (1), 55–64. Ganz, T., Nemeth, E., 2012a. Hepcidin and iron homeostasis. Biochimica et Biophysica Acta (BBA) 1823 (9), 1434–1443. Ganz, T., Nemeth, E., 2012b. Iron metabolism: interactions with normal and disordered erythropoiesis. Cold Spring Harb. Perspect. Med. 2 (5). Ghosh, S., et al., 2010. Polychlorinated biphenyls (PCB-153) and (PCB-77) absorption in human liver (HepG2) and kidney (HK2) cells in vitro: PCB levels and cell death. Environ. Int. 36 (8), 893–900. Hamilton, D.L., Valberg, L.S., 1974. Relationship between cadmium and iron absorption. Am. J. Physiol. Content 227 (5), 1033–1037. Hoogenboom, R.L.A.P., et al., 2015. Dioxins, PCBs and heavy metals in Chinese mitten crabs from Dutch rivers and lakes. Chemosphere 123, 1–8. Horiguchi, H., et al., 2010. Latest status of cadmium accumulation and its effects on kidneys, bone, and erythropoiesis in inhabitants of the formerly cadmium-polluted Jinzu River Basin in Toyama, Japan, after restoration of rice paddies. Int. Arch. Occup. Environ. Health 83 (8), 953–970. Huisman, M., et al., 1995. Perinatal exposure to polychlorinated biphenyls and dioxins and its effect on neonatal neurological development. Early Hum. Dev. 41 (2), 111–127. Jin, Y., et al., 2016. Cadmium exposure to murine macrophages decreases their inflammatory responses and increases their oxidative stress. Chemosphere 144, 168–175. Karl, H., et al., 2016. Large scale distribution of dioxins, PCBs, heavy metals, PAH-metabolites and radionuclides in cod (Gadus morhua) from the North Atlantic and its adjacent seas. Chemosphere 149, 294–303. Kohler, M., Zennegg, M., Waeber, R., 2002. Coplanar polychlorinated biphenyls (PCB) in indoor air. Environ. Sci. Technol. 36 (22), 4735–4740. Kunimoto, M., Miura, T., Kubota, K., 1985. An apparent acceleration of age-related changes of rat red blood cells by cadmium. Toxicol. Appl. Pharmacol. 77 (3), 451–457. Kuo, Y.C., et al., 2015. Ambient air concentrations of PCDD/Fs, coplanar PCBs, PBDD/Fs, and PBDEs and their impacts on vegetation and soil. Int. J. Environ. Sci. Technol. 12 (9), 2997–3008. Lawal, A.O., Ellis, E.M., 2011. Nrf2-mediated adaptive response to cadmium-induced toxicity involves protein kinase C delta in human 1321N1 astrocytoma cells. Environ. Toxicol. Pharmacol. 32 (1), 54–62. Lawal, A.O., Ellis, E.M., 2012. Phospholipase C mediates cadmium-dependent apoptosis in HEK 293 cells. Basic Clin. Pharmacol. Toxicol. 110 (6), 510–517. Lawal, A.O., Marnewick, J.L., Ellis, E.M., 2015. Heme oxygenase-1 attenuates cadmium-

8