Benzo(a)pyrene-7,8-dihydrodiol-9,10-epoxide induces ferroptosis in neuroblastoma cells through redox imbalance

Abstract

As a widespread environmental pollutant, benzo(a)pyrene-7,8-diol-9,10-epoxide (BPDE)-induced neurotoxicity has received increasing attention. Studies have shown that BPDE-induced neurodegeneration is due partly to neuronal apoptosis. Unlike apoptosis, ferroptosis is a non-apoptotic form of programmed cell death, but its specific role in the neurotoxicity of BPDE remains unclear. In this work, we investigated the ferroptosis in BPDE-induced cell death in human neuroblastoma cell line SH-SY5Y using a specific pharmacological inhibitor. Lipid peroxides, malondialdehyde production, glutathione / glutathione peroxidase activity, superoxide dismutase activity, and iron content were evaluated. Consistent with previous studies, our data showed that 0.5 μM BPDE poisoning for 24 hr could induce cell apoptosis and that cell survival could be improved by using apoptosis inhibitors. But with prolonged exposure time (72 hr) or increased exposure dose (1.0 μM), we have elucidated and validated that BPDE triggered ferroptosis in human SH-SY5Y cells. We also revealed that suppression of ferroptosis by specific inhibitors, ferrostatin-1 and deferoxamine, significantly rescued the phenotypes of ferroptosis induced by BPDE. BPDE downregulated Nrf2 and its target genes related to redox regulation, GPX4 and SLC7A11, but upregulated HO-1. Our results first demonstrated that BPDE caused cytotoxic effects on cell death via apoptosis and ferroptosis. Most notably, long-term environmental exposure to BPDE becomes a concern due to ferroptosis. Redox imbalance is controlled by the Nrf2, SLC7A11, and HO-1, through which lipid peroxides and ferrous ion accumulation cause ferroptosis after BPDE treatment. These findings highlight that targeting ferroptosis could serve as an effective protective strategy for neurotoxicity of BPDE.

INTRODUCTION

Benzo(a)pyrene (BaP) is a typical representative of polycyclic aromatic hydrocarbons (PAHs), which are mainly produced during incomplete combustion of organic materials (IARC, 2010). As BaP is a ubiquitous environmental pollutant, people have to face risk of exposure through contact with contaminated air, water, and food. This includes cigarette smoke, cooked food, vehicle exhausts, and industrial emissions. In addition to its well-known carcinogenicity, BaP exhibits neurotoxicity. Because of the high lipophilicity, BaP and its metabolites can penetrate the blood-brain barrier and accumulate in various brain regions (Chepelev et al., 2015). It has been proved in animal experiments that BaP disturbs glutamate receptor subunit, diminishes long-term potentiation in rodents, and leads to poor performance in behavioral tests (Lyu et al., 2020; McCallister et al., 2008). Worryingly, a large body of epidemiological evidence also indicates that environmental exposure to BaP correlates positively with a neurobehavioral deficiency in adults and poor neurodevelopment in children (Niu et al., 2010; Perera et al., 2015; Qiu et al., 2013; Perera et al., 2006; Perera et al., 2003). Considering that neurotoxic endpoints following acute and sub-chronic BaP exposure are manifested at lower doses than cancer endpoints (Chepelev et al., 2015), more attention should be paid to the neurotoxic effects caused by exposure to BaP.

The precise molecular mechanisms of BaP-induced neurotoxicity are still elusive. Growing evidence has revealed that oxidative stress contributes to many adverse biological effects after exposure to BaP (Das et al., 2019; Lin et al., 2020). Oxidative stress results from an imbalance of prooxidants and antioxidants with increased intracellular levels of reactive oxygen species (ROS) (Jones, 2006). After entering the body, BaP induces the activation of cytochrome P450 enzyme CYP1A1 through aryl hydrocarbon receptor (AHR)/AHR-nuclear translocator (ARNT) signaling, leading to the formation of ultimate toxic metabolite benzo(a)pyrene-7,8-diol-9,10-epoxide (BPDE) (Michaelson et al., 2011; Lin et al., 2020). During the metabolic activation and detoxication of the BaP in vivo, excessive reactive oxygen species (ROS) and free radicals are produced (Yang et al., 2016). ROS accumulation in vivo causes mitochondrial dysfunction, apoptosis of neurons, and neurodegeneration; these are known as the primary mechanisms involved in the toxicity of BaP (Murawska-Ciałowicz et al., 2011; Nie et al., 2014; Barangi et al., 2020). Oxidative damage is a cause and consequence of various types of cell death (Tang et al., 2019). Beyond its implication in apoptosis, cell death may occur in other multiple forms in response to oxidative stress (Tang et al., 2019). During the past few decades, many novel forms of regulated cell death have been found, including autophagy, necroptosis, pyroptosis, and ferroptosis (Tang et al., 2019).

Ferroptosis, described in 2012 by Dixon, is likely to occur in certain human diseases, especially neurodegenerative (Han et al., 2020). It is a form of regulated cell death (RCD) characterized by iron accumulation, the process of lipid peroxidation, and subsequent plasma membrane rupture (Dixon et al., 2012). Although the molecular mechanism of ferroptosis still needs further exploration, the increase in iron accumulation, ROS production, then catalyzed the process of lipid peroxidation initiated through non-enzymatic (Fenton reactions) and enzymatic mechanisms (lipoxygenases) seem to play the central role in regulating the function of ferroptosis (Tang et al., 2019; Chen et al., 2021). Thus, the deleterious effects of lipid peroxidation in ferroptosis execution can be neutralized by lipophilic radical traps such as ferrostatin-1 (Fer-1) and liproxstatin-1. Multiple oxidative stress and antioxidant defense pathways shape the ferroptosis responses (Kuang et al., 2020). Research revealed that ferroptosis is morphologically, biochemically, and genetically distinct from apoptosis, necrosis, and autophagy. From observation of morphological features, the nuclei of cells remain intact during ferroptosis; however, the shrinking mitochondria of cells show increased membrane density and outer mitochondrial membrane rupture (Han et al., 2020). The most critical biochemical features of ferroptosis are the elevated lipid peroxidation levels and ferrous ion concentration. In addition, regulators such as Nuclear factor-erythroid 2-related factor 2 (Nrf2), glutathione peroxidase 4 (GPX4), and solute carrier family 7 membrane 11 (SLC7A11) regulate antioxidant defense or detoxification in the context of various stressors will be conducive to definitively distinguish between different RCD types (Han et al., 2020).

Despite many reports indicating that BaP and its ultimate toxic metabolite can induce oxidative stress, which plays a vital role in the process of neurodegenerative attacking. To date, there has been no study about ferroptosis involvement in BaP- or BPDE-induced neurotoxicity. Hence, the present study aimed to characterize the triggering of ferroptosis by BPDE in vitro. The SH-SY5Y human neuroblastoma cell line is a simple and commonly used in vitro model related to neurotoxicity, oxidative stress, and neurodegenerative diseases (Krishna et al., 2014). In the current study, we proved that BPDE triggers ferroptosis in SH-SY5Y cells dose- and time-dependently by using specific small molecule inhibitors. Notably, redox imbalance, especially disturbed Nrf2−SLC7A11−heme oxygenase 1 (HO-1) pathway, plays a crucial role in BPDE-induced ferroptosis. Our findings provide new insights into understanding the mechanism for neurotoxicity of BaP.

MATERIALS AND METHODS

Reagents

BPDE was purchased from the National Institute of Metrology, China. Benzyloxycarbonyl-Val-Ala-Asp-fluoromethyl ketone (Z-VAD-FMK), Necrostatin-1 (Nec-1), Ferrostatin-1 (Fer-1), and desferrioxamine (DFO) were obtained from Sigma-Aldrich. MEM/F12 medium and fetal bovine serum (FBS) were purchased from Procell Life Science & Technology, Wuhan, China. Trypsin-EDTA solution, penicillin/streptomycin mixture, and dimethyl sulfoxide (DMSO) were purchased from Solarbio Life Science, Beijing, China. The antibody against Nrf2 (Cat No. 16396-1-AP), HO-1 (Cat No. 10701-1-AP), SLC7A11 (Cat No. 26864-1-AP), β-actin (Cat No. 20536-1-AP) and goat anti‐rabbit IgG secondary antibody (Cat No. SA00001-2) were purchased from Proteintech, Wuhan, China. The antibody against GPX4 (ab125066) was purchased from Abcam.

Cell culture and treatments

Human neuroblastoma cells (SH-SY5Y cells) (Procell Life Science & Technology Co. Ltd.) were cultured in MEM/F12 medium containing 10% fetal bovine serum (FBS) and 1% penicillin/streptomycin mixture, placed in a 37°C incubator with 5% CO₂. The medium was changed every other day. When the cells were grown to 80% confluence, the cells were used for studies.

To explore the effects of different exposure doses/times of BPDE on SH-SY5Y cells, drug treatments of cells were performed as follows: (1) Cells were treated with BPDE at a final concentration of 0, 0.25, 0.50, 1.00 μmol·L⁻¹ for 24 hr. (2) Cells were treated with BPDE (0.50 μmol·L⁻¹) for 24, 48, and 72 hr. Various pharmacological inhibitors, including for apoptosis (20 μM Z-VAD-FMK), necrosis/necroptosis (20 μM Nec-1), and ferroptosis (10 μM Fer-1 and 50 μM DFO), were used to define the type of cell death by BPDE. One of the pharmacological inhibitors treated cells two hours earlier than BPDE exposure. DMSO as a solvent for BPDE and pharmacological inhibitors, its final concentration in the culture medium did not exceed 0.1% (v/v). Cells were treated only with the same concentration of DMSO as the solvent control group.

Cells viability assays

Cell viability was measured using the MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide] cell proliferation and cytotoxicity assay kit (Nanjing Jiancheng Bioengineering Institute, China). SH-SY5Y cells were suspended in 200 µL of culture medium and seeded into 96-well assay plates with a density of 50000 cells per well for 12 hr. Then cells were treated using designated conditions with three replicates in each group. After that, 50 µL/well of MTT (5 mg/mL) solution was added, and cells were incubated for another 4 hr. We removed the medium carefully and added 150 µL DMSO to dissolve the precipitate dye. The absorbance at 570 nm measured by a microplate reader (BioTek) was applied to analyze the cell viability. All experiments were repeated three times independently.

Transmission electron microscopy

After treatment, cells were collected and steeped in 2.5% glutaraldehyde at 4°C for 2 hr. Then cells were post-fixed with 1% osmium tetroxide in a 0.1 M phosphate-buffered saline (PBS, pH 7.4) at room temperature for 1.5 hr, dehydrated in a graded acetone series, and embedded in epoxy resin 618 overnight at 37°C. Ultrathin sections were made with a microtome and stained with uranyl acetate and lead citrate for 20 min. The cells were observed on a JEOL 1011 transmission electron microscope (Olympus Corporation, Tokyo, Japan).

Measurement of lipid ROS (L-ROS)

To visualize intracellular lipid ROS, cells were stained with a 5 μM C11-BODIPY^581/591 probe (Cat# D3861, Thermo Fisher Scientific) for 30 min in the dark. Subsequently, cells were washed twice with PBS and resuspended in 500 µL PBS. Fluorescence analysis was performed by flow cytometry (Beckman Coulter, CA, USA), and data were collected from at least 20000 cells. Oxidation of the polyunsaturated butadienyl portion of the dye results in a shift of the fluorescence emission peak from 590 nm to 510 nm.

Measurement of MDA, GSH, GSH-PXs activities, and SOD activities in cells

Intracellular malondialdehyde (MDA), reduced glutathione (GSH), glutathione peroxidase (GSH-PXs) activities, and superoxide dismutase (SOD) activities were measured using relevant assay kits: Cell Malondialdehyde (MDA) assay kit (Cat# A003-4), Reduced glutathione (GSH) assay kit (Cat# A006-2), Glutathione Peroxidase (GSH-PXs) assay kit (Cat# A005-1) and Superoxide Dismutase (SOD) typed assay kit (Cat# A001-3), all of which were purchased from Nanjing Jiancheng Bioengineering Institute, China. The cells were seeded (1.0 × 10⁶ cells/well) in 6-well plates, then subjected to different treatment conditions. Then, following the manufacturer’s protocols, cells were collected and ultrasonicated to obtain the supernatant used for subsequent measurement. The protein concentration was measured using a bicinchoninic acid (BCA) protein assay kit (Cat# 23225, Thermo Fisher Scientific). Each experiment was performed three times.

Iron assay

Cells were treated in designated conditions. After trypsinization, the cells were washed twice with PBS; then, the supernatant was collected by centrifugation. The iron content of SH-SY5Y cells was measured according to the manufacturer’s instructions for the iron assay kit (Cat# A039‐2, Nanjing Jincheng Bioengineering Institute, China).

Western blot

Total proteins were extracted from cells after treatment. The protein concentration was evaluated using a BCA protein assay kit (Cat# 23225, Thermo Fisher Scientific). An equivalent amount of protein (20 µg) was separated by sodium dodecyl sulfate‐polyacrylamide gel electrophoresis (SDS‐PAGE), then transferred to polyvinylidene fluoride (PVDF) membranes. Subsequently, membranes were blocked with 5% non-fat milk powder at room temperature for 1 hr, then incubated with the primary antibodies at 4°C overnight. The primary antibodies include Nrf2 (1:1000, Proteintech), HO-1 (1:1000, Proteintech), SLC7A11 (1:1000, Proteintech), β-actin (1:1000, Proteintech), and GPX4 (1:1000, Abcam). The membrane was incubated with goat anti‐rabbit IgG, horseradish peroxidase (HRP)‐labeling secondary antibody (1:4000, Proteintech) for 2 hr at room temperature. Chemiluminescence was performed to visualize all bands. The intensity of the bands was analyzed by Image J software. The β-actin antibody was used for normalization. The experiment was repeated at least three times independently.

Statistical analysis

Statistical analysis was performed using SPSS 22.0 software (SPSS Inc., Chicago, USA) and GraphPad Prism 9.2 software. All data were expressed as mean ± standard deviation (SD). One‐way analysis of variance (ANOVA) followed by the least significant difference test (LSD-t) was used to analyze data differences among groups. For all the statistical analyses, a P‐value < 0.05 was considered statistically significant.

RESULTS

Ferroptosis does occur in SH-SY5Y cells with the exposure dose of BPDE increased

BPDE dose-dependently induces cells death in SH-SY5Y

To estimate the cytotoxicity for the SH-SY5Y cells caused by BPDE, we first treated growing cells with BPDE at different final concentrations (0, 0.25, 0.50, 1.00 μmol·L⁻¹) for 24 hr. MTT assays showed that cell viability was significantly decreased with an increasing dose of BPDE. Treatment with 0.50, 1.00 μmol·L⁻¹ BPDE decreased the cell viability to 84.94 ± 2.51%, 68.44 ± 3.75% especially, compared with the control group (P < 0.001, Fig. 1A).

Fig. 1

BPDE induces ferroptosis dose-dependently in human SH-SY5Y cells. (A) SH-SY5Y cells were cultured with 0.25, 0.5, and 1.0 μmol·L⁻¹ of BPDE for 24 hr and then assessed with MTT assay; (B) The transmission electron microscopy was used to observe the mitochondria (scale bar: 200 nm), shrunken mitochondria and chromatin condensation can be seen at the arrow; (C) To define the type of cell death, Z-VAD-FMK (Z-VAD, 20 μM), necrostatin-1 (Nec-1, 20 μM), ferrostatin-1 (Fer-1, 10 μM) or desferrioxamine (DFO, 50 μM) was added to the culture two hours before different concentrations of BPDE were added, and then assessed with MTT assay; Data were derived from three independent experiments and are presented as means ± standard; ***P < 0.001, **P < 0.01, *P < 0.05, compared with the control group; ###P < 0.001, ##P < 0.01, #P < 0.05, compared with an equal dose of BPDE group.

Observation of mitochondrial ultrastructure for BPDE dosages administrated

Transmission electron microscopy was used to detect the changes in the mitochondrial structure after BPDE of different concentrations exposure. As shown in Fig. 1B, the mitochondrial structure of cells in the 0.25 μmol·L⁻¹ BPDE group was similar to that in the control group. Mitochondrial damage, including mitochondrial pyknosis and vacuolar degeneration of the membrane, happened in the 0.50 μmol·L⁻¹ BPDE group. Moreover, in the 1.00 μmol·L⁻¹ BPDE-treated groups, mitochondrial morphological changes by ferroptosis characterized by smaller mitochondria, condensation of mitochondrial membrane density, reduction or elimination of mitochondrial cristae, and rupture of mitochondrial outer membranes.

Various pharmacological inhibitors affect the cytotoxicity of BPDE dosages administrated

To appropriately determine SH-SY5Y cell fates and provide insight into the underlying mechanisms of SH-SY5Y cells, a pharmacological inhibitor was used two hours earlier before BPDE exposure. As shown in Fig. 1C, 0.50 μmol·L⁻¹ BPDE-induced cell death was almost rescued only in the presence of Z-VAD-FMK (Z-VAD) (P < 0.01) but not by necrostatin-1 (Nec-1), ferrostatin-1 (Fer-1), and deferoxamine (DFO). Interestingly, the addition of ferrostatin-1 and DFO effectively suppressed 1.00 μmol·L⁻¹ BPDE-induced cell death (P < 0.01); unwieldy Z-VAD-FMK did not work anymore. The cytotoxicity of BPDE did not respond to the Nec-1. These results excluded the involvement of necrosis/necroptosis in BPDE dosages administrated induced cell death and suggested that there was apoptosis at 0.50 μmol·L⁻¹ BPDE group and ferroptosis at 1.00 μmol·L⁻¹ BPDE group. Furthermore, the type of ferroptosis was dependent on lipid peroxidation and iron overload due to both DFO and ferrostatin-1 having resumed effects.

BPDE dosages administered induce characteristic changes of ferroptosis in SH-SY5Y cells

It is well known that lipid peroxidation, GSH depletion, and iron accumulation are critical events in ferroptosis. Using the C11-BODIPY^581/591 probe, we showed that BPDE-induced death was accompanied by lipid accumulation, particularly at 0.50 and 1.00 μmol·L⁻¹ BPDE groups (P < 0.001, Fig. 2A). For further confirmation, DFO and Fer-1, the specific inhibitors of ferroptosis, were used to explore whether ferroptosis occurred. Only at 1.00 μmol·L⁻¹ BPDE group the production of L-ROS was significantly reversed by Fer-1 or DFO (P < 0.001, Fig. 2A). MDA is a breakdown product of the oxidative degradation of cell membrane lipids and is also generally considered an indicator of the lipid peroxidation (Martínez et al., 2020). Compared with the control group, the middle and high dose of BPDE increased MDA content, which was reversed by ferroptosis inhibitors (P < 0.01, Fig. 2B). Then we detected levels of GSH, GSH-PXs activity, SOD activity, and iron in cells. BPDE increased GSH depletion and blunted the GSH-PXs activity (P < 0.01, Fig. 2C, 2D). Meanwhile, compared with the control group, SOD activity was significantly lower in the 0.50 and 1.00 μmol·L⁻¹ BPDE-treated group (P < 0.001, Fig. 2E). Additionally, after being pretreated with DFO and Fer-1, GSH and the activity of GSH-PXs, and SOD were significantly increased in the 1.00 μmol·L⁻¹ group (Fig. 2B–E). The intracellular iron content of cells treated with 0.50 μmol·L⁻¹ BPDE increased, markedly higher in the 1.00 μmol·L⁻¹ group (P < 0.001, Fig. 2F). Meanwhile, pretreatment of Fer-1 also suppressed the iron levels in cells treated with 1.00 μmol·L⁻¹ of BPDE, slightly less than DFO (Fig. 2F).

Fig. 2

Inhibition of ferroptosis attenuates BPDE-induced lipid peroxidation and iron accumulation. The ferroptosis-related changes were detected in SH-SY5Y cells after treatment with 0.25, 0.5 and 1.0 μmol·L⁻¹ of BPDE, BPDE + 10 µM Fer-1 or BPDE + 50 μM DFO. (A) The effect of DFO and Fer-1 on lipid ROS levels in BPDE-treated SH-SY5Y cells was assessed by the C11-BODIPY^581/591 probe; (B, C, D, E). The effect of DFO and Fer-1 on the level of MDA, GSH, GSH-PXs, and SOD activity was detected; (F) Content of iron in SH-SY5Y cells after BPDE and inhibitor treatment were detected; All values are presented as means ± standard of three independent experiments; ***P < 0.001, **P < 0.01, *P < 0.05, compared with the control group; ###P < 0.001, ##P < 0.01, #P < 0.05, compared with an equal dose of BPDE group.

BPDE disturbed the redox-responsive proteins, leading to the ferroptosis

To further understand the molecular events that may result in the abnormity documented above, we analyzed the protein expression profiles of GPX4, Nrf2, SLC7A11, and HO-1. Although treatment of 1.00 μmol·L⁻¹ BPDE caused significant suppression of cellular GSH-PXs activity, western blot results showed that the protein level of GPX4 was decreased only by little change (P < 0.05, Fig. 3). Nrf2, a known upstream of HO-1 and SLC7A11 in response to ROS, exhibited a dramatic decrease after 1.00 μmolL-1 BPDE exposure and resumed well after adding DFO and Fer-1 (P < 0.001, Fig. 3). Herein, we observed decreased protein expression of SLC7A11 consistent with the downregulation of Nrf2. However, in contrast to SLC7A11, the protein expression of HO-1 increased significantly in the 1.00 μmol·L⁻¹ BPDE group and was reduced by Fer-1 (Fig. 3).

Fig. 3

BPDE disturbed the expression of Nrf2, GPX4, SLC7A11, and HO-1 protein, and then DFO and Fer-1 can alleviate these changes. After 24 hr treatment, cells were harvested for protein determination through western blot analysis; All values are presented as means ± standard of three independent experiments; Representative images from three independent experiments were shown; ***P < 0.001, **P < 0.01, *P < 0.05, compared with the control group; ###P < 0.001, ##P < 0.01, #P < 0.05, compared with equal dose of BPDE group.

BPDE triggers ferroptosis in SH-SY5Y cells with the extension of exposure time

BPDE time-dependently induces ferroptosis in SH-SY5Y cells

The above study found that higher doses of BPDE (1.00 μmol·L⁻¹) induced ferroptosis in SH-SY5Y cells. Next, we checked whether prolonged exposure time of BPDE at the lower amount still triggers ferroptosis. We investigated the cell viability at three time points (24, 48, and 72 hr) following exposure to the 0.50 μmol·L⁻¹ BPDE. As shown in Fig. 4A, cell viability decreased more with further increasing the treatment time to 48 hr and 72 hr under the poisoning of BPDE (P < 0.001). By transmission electron microscopy, we observed that cells treated with 0.50 μmol·L⁻¹ BPDE for 72 hr exhibited the distinctive morphological features of smaller mitochondria with increased membrane density (Fig. 4B).

Fig. 4

BPDE induces ferroptosis time-dependently in human SH-SY5Y cells. (A) SH-SY5Y cells were cultured with 0.5 μmol·L⁻¹ of BPDE for 24, 48, and 72 hr respectively, and then assessed with MTT assay; (B) Representative cell mitochondria morphological changes are shown (scale bar: 200 nm), shrunken mitochondria and chromatin condensation can be seen at the arrow; (C) To define the type of cell death, Z-VAD-FMK (Z-VAD, 20 μM), necrostatin-1 (Nec-1, 20 μM), ferrostatin-1 (Fer-1, 10 μM) or desferrioxamine (DFO, 50 μM) was added to the culture two hours before different concentrations of BPDE were added, and then assessed with MTT assay; Data were derived from three independent experiments and are presented as means ± standard; ***P < 0.001, **P < 0.01, *P < 0.05, compared with the control group; ###P < 0.001, ##P < 0.01, #P < 0.05, compared with an equal dose of BPDE group.

For further confirmation, pharmacological inhibitors were still used to investigate the rescue effects. Consistently with the above report, apoptosis inhibitor Z-VAD-FMK was effective against 0.50 μmol·L⁻¹ BPDE-induced cell death for 24 hr, but Fer-1 started to play a healing effect for 48 hr. Significant rescue effects on BPDE-induced cell death were observed by treating cells with Fer-1 or DFO for 72 hr (P < 0.001, Fig. 4C). Collectively, BPDE may be brought cells to the final ferroptosis as the extension of poisoning time.

Characterization of intracellular oxidative stress index extending BPDE exposure time

The effects of BPDE on L-ROS production, MDA, GSH levels, the activity of SOD, GSH-PXs, and intracellular iron level in SH-SY5Y cells with the prolongation of exposure time are depicted in Fig. 5A–F, respectively. Consistent with the rising mortality of cells, both L-ROS and MDA levels significantly increased after 48 hr of incubation (P < 0.001, Fig. 5A, 5B). Simultaneously, the results showed a distinct time-dependent decrease in GSH, GSH-PXs activity, and SOD activity and an increase in iron accumulation (P < 0.001, Fig. 5C–F).

Fig. 5

The ferroptosis-related changes over time induced by BPDE in SH-SY5Y cells. The SH-SY5Y cells were treated with 0.5 μmol·L⁻¹ of BPDE for 24, 48, and 72 hr respectively; (A–C) Contents of L-ROS, MDA, and GSH were detected; (D, E) Activity of GSH-PXs and SOD were shown; (F) Content of iron was detected; Data were derived from three independent experiments and are presented as means ± standard; ***P < 0.001, **P < 0.01, *P < 0.05, compared with the control group; ###P < 0.001, ##P < 0.01, #P < 0.05, compared with an equal dose of BPDE group.

Taken together, all the indexes confirmed that the degree of oxidative stress that happened in the SH-SY5Y cells, with the administration of 0.50 μmol·L⁻¹ BPDE after 48 hr, was equal to that treated by a higher dose of BPDE. These results were validated by using ferroptosis inhibitors. DFO and Fer-1 effectively suppressed the L-ROS and MDA accumulation after 48 hr (P < 0.001, Fig. 5A, 5B).

The expression of Nrf2, GPX4, SLC7A11, and HO-1 proteins was disturbed with extension of BPDE exposure time

As shown in Fig. 6, the expression levels of Nrf2 and SLC7A11 were significantly down-regulated under the treatment of 0.50 μmol·L⁻¹ BPDE for 72 hr (P < 0.01), while the level of HO-1 was higher (P < 0.01). But there was no change in the protein level of GPX4 (P > 0.05). These data further demonstrated that 0.5 μmol·L⁻¹ of BPDE can disturb the redox imbalance, which is controlled by proteins Nrf2, SLC7A11, and HO-1, with the expression of GPX4 as an exception, then induce cell ferroptosis with the extension of exposure time.

Fig. 6

BPDE disturbed the expression of Nrf2, GPX4, SLC7A11, and HO-1 protein. After treatment, cells were harvested for protein determination through western blot analysis; All values are presented as means ± standard of three independent experiments; Representative images from three independent experiments were shown; ***P < 0.001, **P < 0.01, *P < 0.05, compared with the control group.

DISCUSSION

There is increasing awareness of the relationship between BaP exposure, neuronal death, and neurodegenerative disorder (Das et al., 2020; Liu et al., 2020a). Previous studies have shown that BaP induced toxicity via its derivative BPDE through oxidative stress and apoptosis (Mehri et al., 2020; Guo et al., 2020). However, oxidative stress is involved in various types of RCD, which will provide a new horizon to exploit other cells’ death triggered by BaP and BPDE. Our work focuses on the cell death types induced by BPDE in human SH-SY5Y cells tested by the particular pharmacological inhibitors; this is an access point of this research. Per previous studies, the current results illustrated that BPDE could induce SH-SY5Y cells apoptosis in a specific range of dosage (about 0.5 μmol·L⁻¹).

Interestingly, the damaged cells increased with the rise of dose, but not due to the increased apoptosis. Apoptosis inhibitor Z-VAD-FMK could not rescue SH-SY5Y cell’s death with the BPDE dose of 1 μmol·L⁻¹. This is credible evidence that apoptosis is not the unique mechanism of BPDE-induced nerve cell insults. Contrary to that, ferroptosis inhibitors DFO and Fer-1 both significantly reduced 1 μmol·L⁻¹ BPDE-induced cell damage, although cell vitality still cannot be reversed completely.

Ferroptosis is a newly defined form of cell death driven by iron-dependent lipid peroxidation, which differs significantly from apoptosis in the morphology and biochemistry (Dixon et al., 2012). Ferroptosis has been shown to have significant implications in several neurologic diseases, such as Alzheimer’s disease and Parkinson’s disease (Reichert et al., 2020; Weiland et al., 2019). Using neuroblastoma cell lines, researchers have found that MYCN-amplified neuroblastoma cells are sensitive to ferroptosis inducers such as erastin and sulfasalazine (SAS) (Floros et al., 2021; Lu et al., 2021). Fer-1 or mitochondrial ferritinrecently was shown to have a neuroprotective role in the SH-SY5Y (Wang et al., 2016; Kabiraj et al., 2015). More importantly, research in organotypic hippocampal slice cultures has shown that ferroptosis contributes to neuronal death (Weiland et al., 2019).

To the best of our knowledge, we first identified ferroptosis as a novel mechanism involved in the cytotoxicity of BPDE in this study. This may, to some extent, explain why BaP-induced neurotoxicity is associated with a neurodegenerative disorder. We, therefore, investigated further the characteristics of SH-SY5Y cells in morphology and biochemistry under BPDE and ferroptosis inhibitors treated. Morphologically, the most typical attributes of ferroptosis in SH-SY5Y cells under ultrastructure are mitochondria contraction, reduction or disappearance of mitochondrial cristae, increasing mitochondrial membrane density, and mitochondrial outer membrane rupture (Yan and Zhang, 2019), which were observed by TEM at 1 μmol·L⁻¹ group. Animal and cell line models have proposed that BaP and BPDE can produce excessive free radicals, enhances lipid peroxidation and MDA level, and disrupt oxidant-antioxidant balance affecting antioxidant enzyme function (Mehri et al., 2020; Guo et al., 2020). This study also proved all the characteristics that obviously changed at the dose of BPDE-induced apoptosis. It was vital to show that degree of lipid peroxidation and iron accumulation was significantly higher at the amount of 1 μmol·L⁻¹ BPDE.

Meanwhile, both the iron chelator DFO and the antioxidant Fer-1 had a good inhibition effect against adverse biochemistry changes in the 1 μmol·L⁻¹ BPDE group. In the ferroptosis process, one of the most important and studied so far is the enzyme GPX4, which can prevent the iron-dependent formation of toxic lipid peroxidation. GPX4 must use GSH as a substrate to reduce peroxides to their corresponding alcohols. Thus intracellular GSH levels are crucial to the activity of the GPX4 (Forcina and Dixon, 2019). In this study, the activity of glutathione peroxidase enzymatic was reduced with GSH depletion. As mentioned by Tang, oxidative damage is not only a cause but also a consequence of cell death (Tang et al., 2019). Based on this profile, we infer that in our cell model, oxidative damage maybe not be the cause of apoptosis, but the main reason relates to ferroptosis. On the one side, these results proved once again that a high dose of BPDE indeed caused ferroptosis in cells; on the other side, this also well explained why cells’ survival rate was seriously reduced but saved by DFO and Fer-1.

Numerous studies suggest that Nrf2 plays an essential role in developing and treating neurodegenerative diseases (Song and Long, 2020). Nrf2 is a critical regulator in the cell antioxidant defense system, also been shown to play a crucial role in mediating iron/metal metabolism, xenobiotic/drug, proteostasis, etc., thereby regulating the ferroptosis process (Dodson et al., 2019; Chen et al., 2021; Song and Long, 2020). Nrf2 modulates directly or indirectly the expression of multiple downstream genes, including GPX4, SLC7A11, HO-1, etc. (Dodson et al., 2019). SLC7A11 is composed of the system Xc⁻, a cystine/glutamine antiporter, which regulates GSH synthesis (Liu et al., 2020b). Our results showed that the protein level of Nrf2 was decreased in the BPDE-induced ferroptosis process, simultaneously reversed by DFO and Fer-1. SLC7A11 was a rationale consistent with Nrf2 and got a markable change. Downregulation of GPX4 was observed but showed an unstable recovery in our cells model. Since ferroptosis inhibitors can not completely rescue the cell’s death and oxidative damage, that seems to be a reasonable explanation for the levels of GPX4. HO-1, as an inducible enzyme, catalyzes the degradation of cellular heme to free ferrous iron, biliverdin, and carbon monoxide under stressful conditions (Loboda et al., 2016; Nitti et al., 2018). HO-1 is upregulated by several oxidative stimuli carrying out antioxidant responses; on the contrary, overexpression of HO-1 has been shown to have pro-oxidant effects (Nitti et al., 2018). However, the role of HO-1 in regulating ferroptosis is unclear; even the level of HO-1 is opposite in different models. Our data showed that BPDE caused a high expression of HO-1 and was down-regulated by Fer-1.

We demonstrate that a higher dose of BPDE could trigger ferroptosis in SH-SY5Y cells. The attendant question is what will happen in cells treated at the dose causing only apoptosis if prolonged exposure time. People always face BaP contamination in long-term and low doses in life. Therefore, studying the time effects of BPDE-induced ferroptosis is of vital importance. Using the particular inhibitors, we found that cell death poisoning of 0.5 μmol·L⁻¹ BPDE was no longer reversed by Z-VAD-FMK but reversed by Fer-1 when the exposure time was extended to 48 hr. The results represented that new ferroptosis onset increased over time as well. Morphological results also supported this result. The degree of intracellular oxidative damage occurring after 48 hr of 0.5 μmol·L⁻¹ BPDE treatment was similar to that occurring after 24 hr of 1 μmol·L⁻¹ BPDE treatment.

The results obtained from this study may have some limitations. We propose that BPDE can induce apoptosis firstly, then ferroptosis with the exposure dose and time further increased. But the underlying mechanism leading to this change in the cell death phenotype still needs to be studied further. Secondly, this study only used SH-SY5Y cells in vitro, which need to be verified with primary neurons, even animal models.

In conclusion, consistent with previous studies, in this work, we found that BPDE can induce cell apoptosis and improve cell survival by using an apoptosis inhibitor. But with prolonged exposure time and increased exposure dose, we have elucidated and validated that BPDE triggered ferroptosis in human SH-SY5Y cells. We also revealed that suppression of ferroptosis by specific inhibitors, ferrostatin-1 and DFO, significantly rescued the phenotypes of ferroptosis induced by BPDE. During the BPDE-induced ferroptosis process, redox imbalance involved in protein Nrf2, GPX4, SLC7A11, and HO-1 is the main issue. Our results suggest a novel set of mechanisms that may contribute to the neurotoxicity of BPDE. We also infer that ferroptosis inhibitors might be used as an approach for prevention. Nevertheless, the detailed mechanisms underlying how apoptosis shifts to ferroptosis remain unclear and require further study in the future.

ACKNOWLEDGMENT

This study was supported by the National Natural Science Foundation of China (NSFC 30872137); the Natural Science Foundation for Youths of Shanxi Province, China (20210302124301); the Key Program for International Cooperation Projects of Shanxi province (201703D421021); and a Fund for Shanxi “1331 Project” (2021-5-2-2-B1).

Conflict of interest

The authors declare that there is no conflict of interest.

REFERENCES

•  Barangi,  S.,  Mehri,  S.,  Moosavi,  Z.,  Hayesd,  A.W.,  Reiter,  R.J.,  Cardinali,  D.P. and  Karimi,  G. (2020): Melatonin inhibits Benzo(a)pyrene-Induced apoptosis through activation of the Mir-34a/Sirt1/autophagy pathway in mouse liver. Ecotoxicol. Environ. Saf., 196, 110556.
•  Chen,  X.,  Kang,  R.,  Kroemer,  G. and  Tang,  D. (2021): Broadening horizons: the role of ferroptosis in cancer. Nat. Rev. Clin. Oncol., 18, 280-296.
•  Chepelev,  N.L.,  Moffat,  I.D.,  Bowers,  W.J. and  Yauk,  C.L. (2015): Neurotoxicity may be an overlooked consequence of benzo[a]pyrene exposure that is relevant to human health risk assessment. Mutat. Res. Rev. Mutat. Res., 764, 64-89.
•  Das,  L.,  Patel,  B. and  Patri,  M. (2019): Adolescence benzo[a]pyrene treatment induces learning and memory impairment and anxiolytic like behavioral response altering neuronal morphology of hippocampus in adult male Wistar rats. Toxicol. Rep., 6, 1104-1113.
•  Das,  S.K.,  Aparna,  S. and  Patri,  M. (2020): Chronic waterborne exposure to benzo[a]pyrene induces locomotor dysfunction and development of neurodegenerative phenotypes in zebrafish. Neurosci. Lett., 716, 134646.
•  Dixon,  S.J.,  Lemberg,  K.M.,  Lamprecht,  M.R.,  Skouta,  R.,  Zaitsev,  E.M.,  Gleason,  C.E.,  Patel,  D.N.,  Bauer,  A.J.,  Cantley,  A.M.,  Yang,  W.S.,  Morrison,  B. 3rd and  Stockwell,  B.R. (2012): Ferroptosis: an iron-dependent form of nonapoptotic cell death. Cell, 149, 1060-1072.
•  Dodson,  M.,  Castro-Portuguez,  R. and  Zhang,  D.D. (2019): NRF2 plays a critical role in mitigating lipid peroxidation and ferroptosis. Redox Biol., 23, 101107.
•  Floros,  K.V.,  Cai,  J.,  Jacob,  S.,  Kurupi,  R.,  Fairchild,  C.K.,  Shende,  M.,  Coon,  C.M.,  Powell,  K.M.,  Belvin,  B.R.,  Hu,  B.,  Puchalapalli,  M.,  Ramamoorthy,  S.,  Swift,  K.,  Lewis,  J.P.,  Dozmorov,  M.G.,  Glod,  J.,  Koblinski,  J.E.,  Boikos,  S.A. and  Faber,  A.C. (2021): MYCN-Amplified Neuroblastoma Is Addicted to Iron and Vulnerable to Inhibition of the System Xc-/Glutathione Axis. Cancer Res., 81, 1896-1908.
•  Forcina,  G.C. and  Dixon,  S.J. (2019): GPX4 at the Crossroads of Lipid Homeostasis and Ferroptosis. Proteomics, 19, e1800311.
•  Guo,  L.,  Wei,  M.,  Li,  B.,  Yun,  Y.,  Li,  G. and  Sang,  N. (2020): The Role of Cyclooxygenases-2 in Benzo(a)pyrene-Induced Neurotoxicity of Cortical Neurons. Chem. Res. Toxicol., 33, 1364-1373.
•  Han,  C.,  Liu,  Y.,  Dai,  R.,  Ismail,  N.,  Su,  W. and  Li,  B. (2020): Ferroptosis and Its Potential Role in Human Diseases. Front. Pharmacol., 11, 239.
• IARC. (2010): ‘Some non-heterocyclic polycyclic aromatic hydrocarbons and some related exposures’, IARC monographs on the evaluation of carcinogenic risks to humans / World Health Organization, International Agency for Research on Cancer, 92, 1-853.
•  Jones,  D.P. (2006): Extracellular redox state: refining the definition of oxidative stress in aging. Rejuvenation Res., 9, 169-181.
•  Kabiraj,  P.,  Valenzuela,  C.A.,  Marin,  J.E.,  Ramirez,  D.A.,  Mendez,  L.,  Hwang,  M.S.,  Varela-Ramirez,  A.,  Fenelon,  K.,  Narayan,  M. and  Skouta,  R. (2015): The neuroprotective role of ferrostatin-1 under rotenone-induced oxidative stress in dopaminergic neuroblastoma cells. Protein J., 34, 349-358.
•  Krishna,  A.,  Biryukov,  M.,  Trefois,  C.,  Antony,  P.M.,  Hussong,  R.,  Lin,  J.,  Heinäniemi,  M.,  Glusman,  G.,  Köglsberger,  S.,  Boyd,  O.,  van den Berg,  B.H.,  Linke,  D.,  Huang,  D.,  Wang,  K.,  Hood,  L.,  Tholey,  A.,  Schneider,  R.,  Galas,  D.J.,  Balling,  R. and  May,  P. (2014): Systems genomics evaluation of the SH-SY5Y neuroblastoma cell line as a model for Parkinson’s disease. BMC Genomics, 15, 1154.
•  Kuang,  F.,  Liu,  J.,  Tang,  D. and  Kang,  R. (2020): Oxidative Damage and Antioxidant Defense in Ferroptosis. Front. Cell Dev. Biol., 8, 586578.
•  Lin,  Y.-C.,  Wu,  C.-Y.,  Hu,  C.-H.,  Pai,  T.-W.,  Chen,  Y.-R. and  Wang,  W.-D. (2020): Integrated Hypoxia Signaling and Oxidative Stress in Developmental Neurotoxicity of Benzo[a]Pyrene in Zebrafish Embryos. Antioxidants, 9, 731.
•  Liu,  D.,  Zhao,  Y.,  Qi,  Y.,  Gao,  Y.,  Tu,  D.,  Wang,  Y.,  Gao,  H.M. and  Zhou,  H. (2020a): Benzo(a)pyrene exposure induced neuronal loss, plaque deposition, and cognitive decline in APP/PS1 mice. J. Neuroinflammation, 17, 258.
•  Liu,  J.,  Xia,  X. and  Huang,  P. (2020b): xCT: A Critical Molecule That Links Cancer Metabolism to Redox Signaling. Mol. Ther., 28, 2358-2366.
•  Loboda,  A.,  Damulewicz,  M.,  Pyza,  E.,  Jozkowicz,  A. and  Dulak,  J. (2016): Role of Nrf2/HO-1 system in development, oxidative stress response and diseases: an evolutionarily conserved mechanism. Cell. Mol. Life Sci., 73, 3221-3247.
•  Lu,  Y.,  Yang,  Q.,  Su,  Y.,  Ji,  Y.,  Li,  G.,  Yang,  X.,  Xu,  L.,  Lu,  Z.,  Dong,  J.,  Wu,  Y.,  Bei,  J.X.,  Pan,  C.,  Gu,  X. and  Li,  B. (2021): MYCN mediates TFRC-dependent ferroptosis and reveals vulnerabilities in neuroblastoma. Cell Death Dis., 12, 511.
•  Lyu,  Y.,  Ren,  X.K.,  Zhang,  H.F.,  Tian,  F.J.,  Mu,  J.B. and  Zheng,  J.P. (2020): Sub-chronic administration of benzo[a]pyrene disrupts hippocampal long-term potentiation via inhibiting CaMK II/PKC/PKA-ERK-CREB signaling in rats. Environ. Toxicol., 35, 961-970.
•  Martínez,  M.A.,  Rodríguez,  J.L.,  Lopez-Torres,  B.,  Martínez,  M.,  Martínez-Larrañaga,  M.R.,  Maximiliano,  J.E.,  Anadón,  A. and  Ares,  I. (2020): Use of human neuroblastoma SH-SY5Y cells to evaluate glyphosate-induced effects on oxidative stress, neuronal development and cell death signaling pathways. Environ. Int., 135, 105414.
•  McCallister,  M.M.,  Maguire,  M.,  Ramesh,  A.,  Aimin,  Q.,  Liu,  S.,  Khoshbouei,  H.,  Aschner,  M.,  Ebner,  F.F. and  Hood,  D.B. (2008): Prenatal exposure to benzo(a)pyrene impairs later-life cortical neuronal function. Neurotoxicology, 29, 846-854.
•  Mehri,  S.,  Barangi,  S.,  Zamiri,  E. and  Karimi,  G. (2020): The protective effect of melatonin on benzo(a)pyrene-induced brain injury: role of apoptosis and autophagy pathways. Naunyn Schmiedebergs Arch. Pharmacol., 393, 2241-2251.
•  Michaelson,  J.J.,  Trump,  S.,  Rudzok,  S.,  Gräbsch,  C.,  Madureira,  D.J.,  Dautel,  F.,  Mai,  J.,  Attinger,  S.,  Schirmer,  K.,  von Bergen,  M.,  Lehmann,  I. and  Beyer,  A. (2011): Transcriptional signatures of regulatory and toxic responses to benzo-[a]-pyrene exposure. BMC Genomics, 12, 502.
•  Murawska-Ciałowicz,  E.,  Jethon,  Z.,  Magdalan,  J.,  Januszewska,  L.,  Podhorska-Okołów,  M.,  Zawadzki,  M.,  Sozański,  T. and  Dzięgiel,  P. (2011): Effects of melatonin on lipid peroxidation and antioxidative enzyme activities in the liver, kidneys and brain of rats administered with benzo(a)pyrene. Exp. Toxicol. Pathol., 63, 97-103.
•  Nie,  J.S.,  Zhang,  H.M.,  Zhao,  J.,  Liu,  H.J. and  Niu,  Q. (2014): Involvement of mitochondrial pathway in benzo[a]pyrene-induced neuron apoptosis. Hum. Exp. Toxicol., 33, 240-250.
•  Nitti,  M.,  Piras,  S.,  Brondolo,  L.,  Marinari,  U.M.,  Pronzato,  M.A. and  Furfaro,  A.L. (2018): Heme Oxygenase 1 in the Nervous System: Does It Favor Neuronal Cell Survival or Induce Neurodegeneration? Int. J. Mol. Sci., 19.
•  Niu,  Q.,  Zhang,  H.,  Li,  X. and  Li,  M. (2010): Benzo[a]pyrene-induced neurobehavioral function and neurotransmitter alterations in coke oven workers. Occup. Environ. Med., 67, 444-448.
•  Perera,  F.P.,  Rauh,  V.,  Whyatt,  R.M.,  Tsai,  W.Y.,  Tang,  D.,  Diaz,  D.,  Hoepner,  L.,  Barr,  D.,  Tu,  Y.H.,  Camann,  D. and  Kinney,  P. (2006): Effect of prenatal exposure to airborne polycyclic aromatic hydrocarbons on neurodevelopment in the first 3 years of life among inner-city children. Environ. Health Perspect., 114, 1287-1292.
•  Perera,  F.,  Phillips,  D.H.,  Wang,  Y.,  Roen,  E.,  Herbstman,  J.,  Rauh,  V.,  Wang,  S. and  Tang,  D. (2015): Prenatal exposure to polycyclic aromatic hydrocarbons/aromatics, BDNF and child development. Environ. Res., 142, 602-608.
•  Perera,  F.P.,  Rauh,  V.,  Tsai,  W.-Y.,  Kinney,  P.,  Camann,  D.,  Barr,  D.,  Bernert,  T.,  Garfinkel,  R.,  Tu,  Y.-H.,  Diaz,  D.,  Dietrich,  J. and  Whyatt,  R.M. (2003): Effects of transplacental exposure to environmental pollutants on birth outcomes in a multiethnic population. Environ. Health Perspect., 111, 201-205.
•  Qiu,  C.,  Peng,  B.,  Cheng,  S.,  Xia,  Y. and  Tu,  B. (2013): The effect of occupational exposure to benzo[a]pyrene on neurobehavioral function in coke oven workers. Am. J. Ind. Med., 56, 347-355.
•  Reichert,  C.O.,  de Freitas,  F.A.,  Sampaio-Silva,  J.,  Rokita-Rosa,  L.,  Barros,  P.L.,  Levy,  D. and  Bydlowski,  S.P. (2020): Ferroptosis Mechanisms Involved in Neurodegenerative Diseases. Int. J. Mol. Sci., 21.
•  Song,  X. and  Long,  D. (2020): Nrf2 and Ferroptosis: A New Research Direction for Neurodegenerative Diseases. Front. Neurosci., 14, 267.
•  Tang,  D.,  Kang,  R.,  Berghe,  T.V.,  Vandenabeele,  P. and  Kroemer,  G. (2019): The molecular machinery of regulated cell death. Cell Res., 29, 347-364.
•  Wang,  Y.Q.,  Chang,  S.Y.,  Wu,  Q.,  Gou,  Y.J.,  Jia,  L.,  Cui,  Y.M.,  Yu,  P.,  Shi,  Z.H.,  Wu,  W.S.,  Gao,  G. and  Chang,  Y.Z. (2016): The Protective Role of Mitochondrial Ferritin on Erastin-Induced Ferroptosis. Front. Aging Neurosci., 8, 308.
•  Weiland,  A.,  Wang,  Y.,  Wu,  W.,  Lan,  X.,  Han,  X.,  Li,  Q. and  Wang,  J. (2019): Ferroptosis and Its Role in Diverse Brain Diseases. Mol. Neurobiol., 56, 4880-4893.
•  Yan,  N. and  Zhang,  J.J. (2019): The Emerging Roles of Ferroptosis in Vascular Cognitive Impairment. Front. Neurosci., 13, 811.
•  Yang,  F.,  Yang,  H.,  Ramesh,  A.,  Goodwin,  J.S.,  Okoro,  E.U. and  Guo,  Z. (2016): Overexpression of Catalase Enhances Benzo(a)pyrene Detoxification in Endothelial Microsomes. PLoS One, 11, e0162561.