Manganese (II) chloride leads to dopaminergic neurotoxicity by promoting mitophagy through BNIP3-mediated oxidative stress in SH-SY5Y cells

Background Manganese overexposure can induce neurotoxicity, lead to manganism and result in clinical manifestations similar to those of parkinsonism. However, the underlying molecular mechanism is still unclear. This study demonstrated that MnCl2 induces mitophagy and leads to neurotoxicity by promoting BNIP3-mediated reactive oxygen species (ROS) generation. Methods Human neuroblastoma SH-SY5Y cells were used throughout our experiments. Cell viability was detected by cell proliferation/toxicity test kits. Mitochondrial membrane potential was measured by flow cytometry. ROS generation was detected using a microplate reader. Protein levels were evaluated by Western blot. Transmission electron microscopy was used to evaluate mitochondrial morphology. Co-immunoprecipitation was used to verify the interaction between BNIP3 and LC3. Results MnCl2 led to loss of mitochondrial membrane potential and apoptosis of SH-SY5Y cells by enhancing expression of BNIP3 and conversion of LC3-I to LC3-II. Moreover, MnCl2 reduced expression of the mitochondrial marker protein TOMM20 and promoted interaction between BNIP3 and LC3. The results also indicated that a decrease in BNIP3 expression reduced the mitochondrial membrane potential loss, attenuated apoptosis and reduced mitochondrial autophagosome formation in SH-SY5Y cells after MnCl2 treatment. Finally, we found that manganese-induced ROS generation could be reversed by the antioxidant N-acetyl cysteine (NAC) or silencing BNIP3 expression. Conclusions BNIP3 mediates MnCl2-induced mitophagy and neurotoxicity in dopaminergic SH-SY5Y cells through ROS. Thus, BNIP3 contributes to manganese-induced neurotoxicity by functioning as a mitophagy receptor protein.

can participate in mitochondrial protein catabolism and promote the degradation of damaged proteins in mitochondria through mitochondrial quality control. In addition, after BNIP3 is activated, it can regulate the opening of mitochondrial membrane permeability transition pores to mediate lysosomal protein transfer from the cytoplasm to the mitochondrial matrix and the biological regulation of mitochondrial dysfunction [25]. BNIP3 has been found to be a target regulatory gene for manganese-induced neurotoxicity, which is associated with mitochondrial dysfunction and oxidative stress [26]. The purpose of this study was to investigate the BNIP3-dependent neurotoxic effects of MnCl 2 on human neuroblastoma SH-SY5Y cells. To this end, we used the dopaminergic cell line SH-SY5Y as a cell model and MnCl 2 as a stimulus to explore the regulation of BNIP3-mediated, MnCl 2 -induced mitophagy in dopaminergic neurons. Hopefully, this work can provide new insights into the pathophysiological mechanisms of parkinsonism.

Cell culture
SH-SY5Y human neuroblastoma cells were obtained from American Type Culture Collection (ATCC, Manassas, VA, USA). Cells were maintained in DMEM containing 10 % FBS and 1 % penicillin/streptomycin and incubated at 37 °C in 95 %/5 % CO 2 air. Cells were grown in 25 cm 2 culture flasks to 80 % confluence. The study was approved by the Ethics Committee of the First Affiliated Hospital of Guangxi Medical University. Cell viability assay SH-SY5Y cell inhibition by MnCl 2 was measured by the CCK-8 assay. Cells (1 × 10 4 cells/mL) were placed in 96-well plates and treated with 0, 200, 400, 800, or 1600 µM MnCl 2 for 12, 24 and 48 h in the incubator. After intervention with the various concentrations of MnCl 2 , 10 µL of CCK-8 solution was added to each well, and cells were incubated for 2 h at 37 °C. The absorbance was determined using a microplate reader at 450 nm (BioTek ELX800, USA).

Lentivirus vector and cell transfection
SH-SY5Y cells were transduced with LV-BNIP3-RNAi lentivirus carrying shRNA designed to knock down BNIP3 and cloned into the GV112 vector. The virus titer was 5E + 8 TU/mL. GV115-NC lentivirus with no interference sequence was used to infect SH-SY5Y cells as a negative control. The virus titer was 5E + 8 TU/mL. Cells were cultured overnight in 6-well plates, and the constructed vector was transfected into SH-SY5Y cells using 2 mL of DMEM with 5 µg/mL of polybrene. After 48 h, the medium was replaced with fresh medium, and the transduced cells were positively selected by continuous exposure to 1 µg/mL puromycin for 15 days. After selection, the knockdown efficiency of BNIP3 was determined by western blot analysis.

Mitochondrial membrane potential and apoptosis assays
Cell density was adjusted to approximately 20 × 10 4 /mL, and cells were incubated at 37 °C for 30 min in the presence of JC-1 through the use of a mitochondrial membrane potential assay kit (1×, 2 µM, Beyotime, China). Then, the cells were washed with cold JC-1 buffer 3 times for 15 min, and the fluorescence intensity of the cells was analyzed by flow cytometry (BD FACSCalibur, USA). A total of 2 × 10 5 cells were collected. To each tube, 400 µL of 1X binding buffer was added. Then, 5 µL of fluorescein isothiocyanate (FITC) Annexin V and 5 µL of PI were added. The cells were gently vortexed and incubated for 15 min in the dark. The apoptosis rate of cells was analyzed by flow cytometry (BD FACSCalibur, USA).

Coimmunoprecipitation
After the cells reached 70-80 % confluence, the medium was discarded, and the cells were washed twice with ice-cold PBS and lysed on ice for 30 min. Then, the cells were scraped off with a cell scraper and collected into a 1.5-mL centrifuge tube. The cells were centrifuged at 12 000 rpm for 30 min at 4 °C. After centrifugation, the supernatant was carefully aspirated, and the supernatant contained the total protein extract. After the total protein concentration was determined by the BCA protein assay kit, the appropriate amount of protein was incubated with anti-LC3 or anti-BNIP3. In the meantime, 5 µL of protein A agarose beads was washed with PBS twice, dissolved to 50 % concentration in PBS and stored at 4 °C. Pretreated protein A agarose beads (50 %) were added to 100 µL of the total protein extract, which had been incubated with anti-LC3 or anti-BNIP3 overnight with gentle shaking at 4 °C. After incubation, the cells were centrifuged at 12,000 rpm for 30 min at 4 °C, and the agarose beads were centrifuged to the bottom of the tube. The supernatant was removed, and the agarose beads were washed 4 times with 500 µL of ice-cold PBS. Appropriate 5× loading buffer was added to mix the agarose bead-antigen antibody complex, boiled for 5 min, and centrifuged at 12,000 rpm for 5 min at 4 °C. The supernatant was stored at -80 °C for later protein electrophoresis. The protein electrophoresis procedure was the same as that for western blot analysis.

Transmission electron microscopy
SH-SY5Y cells (1 × 10 5 /ml) were seeded in 6-well dishes 24 h before the experiment. After treatment with 400 µM MnCl 2 for 24 h, cells were collected and fixed with 3 % glutaraldehyde at 4 °C for 3 h. Then, the cells were washed three times with ice-cold PBS for 15 min and fixed with 1 % citric acid for 3 h. After fixation, dehydration, infiltration, embedding, polymerization, ultrathin sectioning and lead citrate staining were carried out, followed by gradient alcohol dehydration. After being dried, the mitochondrial autophagosome was observed by a Hitachi H7650 electron microscope (Hitachi H7650, Japan).

Statistical analysis
Data were analyzed using GraphPad Prism6 software (La Jolla, CA, USA) to perform one-way analysis of variance (ANOVA). The difference between two groups was determined using an independent-sample T test. Statistical results are expressed as the mean ± standard deviation (S.D.). P < 0.05 implies a significant difference, and P < 0.01 implies a highly significant difference.

Neurotoxic effects of MnCl 2 on human neuroblastoma SH-SY5Y cells
The purpose of this study was to investigate the neurotoxic effects of MnCl 2 on human neuroblastoma SH-SY5Y cells. After treatment with different concentrations of MnCl 2 for different time periods, the cell proliferation/toxicity test kit CCK-8 was used to determine the cell viability of SH-SY5Y cells. After 12 h, the inhibition of SH-SY5Y cells treated with different concentrations of MnCl 2 was not significantly different from that of cells treated with the control treatment. However, after 24 and 48 h, the activity of SH-SY5Y cells decreased significantly with different concentrations of MnCl 2 (Fig. 1a); the IC 50 values for 24 and 48 h were 585 ± 139 µmol/L and 345 ± 273 µmol/L, respectively. Therefore, except for special cases, subsequent experiments were carried out at 0, 200, 400, of 800 µM MnCl 2 for 24 h. In addition, to further understand the toxic effects of MnCl 2 on SH-SY5Y cells, we detected markers of early apoptosis and mitochondrial membrane potential using a JC-1 kit. The results showed that the mitochondrial membrane potential decreased in a concentration-dependent manner after treatment with 200, 400, and 800 µM MnCl 2 for 24 h (Fig. 1b). In addition, the apoptosis rate was also determined by the FITC Annexin V Apoptosis Detection Kit I. As shown in Fig. 1c, MnCl 2 induced SH-SY5Y cell apoptosis in a dose-dependent manner. The above data indicated that MnCl 2 can decrease the activity of SH-SY5Y cells and cause neurotoxicity.

MnCl 2 induced mitophagy in SH-SY5Y cells
To verify whether MnCl 2 induced mitophagy in SH-SY5Y cells, we first evaluated the expression of the mitophagy receptor protein BNIP3 and the mitochondrial marker protein TOMM20 and the transformation of LC3-I to LC3-II by western blot analysis. LC3-II is a key marker in the process of mitophagy. As shown in Fig. 2a, compared with the control treatment, treatment with 200, 400 and 800 µM MnCl 2 affected the expression of the mitophagy receptor protein BNIP3 and the conversion of LC3-I to LC3-II in a concentration-dependent manner, while the expression level of the mitochondrial marker protein TOMM20 decreased with increasing MnCl 2 concentration. Next, we blocked mitophagy by using the autophagy inhibitor 3-MA to determine whether MnCl 2 can induce mitophagy in SH-SY5Y cells. The results showed that 3-MA could reduce the expression of the mitophagy receptor protein BNIP3   26:23 and the conversion of LC3-I to LC3-II while increasing the expression of the mitochondrial marker protein TOMM20 after treatment with 400 µM MnCl 2 for 24 h (Fig. 2b). Therefore, we inferred that MnCl 2 induced mitophagy in SH-SY5Y cells. To directly and intuitively observe the formation of mitophagy bodies, we used transmission electron microscopy to evaluate mitochondrial morphological changes and mitophagic bodies in SH-SY5Y cells after treatment with 400 µM MnCl 2 for 24 h. The results showed that, compared with the control treatment, treatment with 400 µM MnCl 2 for 24 h resulted in expanded, dissolved, coagulated, or denatured mitochondria and induced the formation of mitochondrial autophagosomes with a bilayer membrane structure. Furthermore, 3-MA could alleviate MnCl 2 -induced mitochondrial autophagosome formation in SH-SY5Y cells (Fig. 2c). These data indicated that the formation of mitochondrial autophagosomes in SH-SY5Y cells was increased by MnCl 2 exposure. The above data indicated that MnCl 2 induced mitophagy in SH-SY5Y cells.

ROS involvement in MnCl 2 -induced mitophagy in SH-SY5Y cells
ROS are mainly produced in mitochondria. ROS may cause oxidative damage to mitochondria under physiological conditions, but the damage is not enough to disturb mitochondrial homeostasis. Under external stimulation, ROS generation increased, and excessive accumulation of ROS caused mitochondrial homeostasis imbalance and led to mitochondrial damage, which resulted in mitophagy. To verify whether MnCl 2 induces mitophagy in SH-SY5Y cells through ROS, we first assessed the level of ROS in SH-SY5Y cells after MnCl 2 treatment. The results showed that, compared with the control treatment, treatment with increasing MnCl 2 concentrations increased the levels of ROS in SH-SY5Y cells (Fig. 3a). In addition, we used the antioxidant N-acetyl cysteine (NAC) to determine whether MnCl 2 induced ROS generation in SH-SY5Y cells.
The results showed that NAC could reduce the production of ROS after MnCl 2 treatment (Fig. 3b). The above data indicated that MnCl 2 can promote ROS generation in SH-SY5Y cells. Therefore, we further explored whether excessive ROS production participates in MnCl 2 -mediated mitophagy in SH-SY5Y cells. The results showed that NAC can reduce the expression of the mitophagy receptor protein BNIP3, reduce the conversion of LC3-I to LC3-II and increase the expression of the mitochondrial marker protein TOMM20 (Fig. 3c). These data indicated that ROS participate in MnCl 2 -mediated mitophagy in SH-SY5Y cells. In addition, NAC also reduced the formation of mitochondrial autophagosomes (Fig. 3d). However, the autophagy inhibitor 3-MA did not reduce ROS generation (Fig. 3e). The above data further confirmed that ROS are involved in MnCl 2 -mediated mitophagy in SH-SY5Y cells.

BNIP3 mediated MnCl 2 -induced mitophagy in SH-SY5Y cells
To explore whether BNIP3 mediated mitophagy, we used BNIP3 shRNA lentiviral vectors to interfere with BNIP3 expression. The results showed that the expression of the mitophagy receptor protein BNIP3 in the negative control was not significantly changed compared with the control but was significantly decreased in BNIP3-shRNA-transfected cells compared with the negative control, suggesting that the BNIP3 shRNA lentiviral vector could significantly reduce the expression of BNIP3 in SH-SY5Y cells, and the noninterfering sequence lentiviral vector had no significant effect on the expression of BNIP3 in SH-SY5Y cells (Fig. 4a). The western blot results showed that, compared with the negative control, 400 µM MnCl 2 could increase the conversion of LC3-I to LC3-II in cells transfected with the noninterfering sequence lentiviral vector and BNIP3 shRNA for 24 h, but the increase in the former was obvious. MnCl 2 (400 µM) also decreased the expression of the mitochondrial marker protein TOMM20, but the protein decrease in the BNIP3-shRNA-transfected cells was smaller than that in the noninterfering sequence lentivirus-transfected cells (Fig. 4b), suggesting that interference with BNIP3 could reduce MnCl 2 -induced mitophagy in SH-SY5Y cells. To further demonstrate that BNIP3 mediated MnCl 2 -induced mitophagy in SH-SY5Y cells, we verified whether BNIP3 regulated mitophagy progression by binding to the autophagy marker protein LC3 through coimmunoprecipitation. The results showed that the anti-BNIP3 antibody specifically captured and pulled down anti-LC3 antibody, and the anti-LC3 antibody specifically captured and pulled down the anti-BNIP3 antibody. These results suggested that BNIP3 and LC3 in SH-SY5Y cells could interact with each other effectively. Furthermore, it was demonstrated that MnCl 2 could increase their binding, and decreased expression of BNIP3 could reduce their binding (Fig. 4c). In addition, transmission electron microscopy results showed that, after treatment with 400 µM MnCl 2 for 24 h, the number of mitochondrial autophagosomes in the BNIP3-shRNA-transfected cells was lower than that in the noninterfering sequence lentivirus-transfected cells (Fig. 4d). Taken together, these results show that BNIP3 mediates MnCl 2 -induced mitophagy in SH-SY5Y cells. for 24 h. ROS levels were measured using a microplate reader. c BNIP3, LC3 and TOMM20 protein levels were measured by western blot after treatment with 400 µM MnCl 2 with or without pretreatment with the antioxidant NAC for 24 h. β-actin served as the internal reference. Compared with control, *P < 0.05, **P < 0.01. d Mitophagy bodies and mitochondria were observed by transmission electron microscopy (TEM) in cells treated with 400 μM MnCl2 alone or co-treated with the mitophagy inhibitor NAC for 24 h. N, nucleus. White arrow, expanded, condensed or dissolved, mitochondria aggregates. Black arrow ( ), relatively normal mitochondrial morphology. Black arrow ( ), mitophagy bodies. e After co-treatment with or without the autophagy inhibitor 3-MA, ROS levels in SH-SY5Y cells were also measured using a microplate reader

Silencing BNIP3 inhibited MnCl 2 -induced ROS generation and neurotoxicity
Previous data indicated that BNIP3 knockdown could effectively reverse MnCl 2 -induced mitophagy in SH-SY5Y cells. Therefore, we further explored the role of BNIP3 in MnCl 2 -induced ROS generation and neurotoxicity in SH-SY5Y cells. Excitingly, compared with the negative control, the ROS generation in the noninterfering sequence lentivirus-transfected cells and the BNIP3-shRNA-transfected cells increased significantly after treatment with 400 µM MnCl 2 for 24 h, but the former increased significantly compared with the latter (Fig. 5a). Furthermore, the mitochondrial membrane potential of both the noninterfering sequence lentivirus-transfected cells and the BNIP3-shRNAtransfected cells was significantly lost after 400 µM MnCl 2 treatment for 24 h compared with the negative control treatment, but the mitochondrial membrane potential loss was smaller in the BNIP3-shRNA-transfected cells (Fig. 5b). The apoptosis results showed that the apoptosis rate of both the noninterfering sequence lentivirus-transfected cells and the BNIP3-shRNA-transfected cells was significantly higher after treatment with  400 µM MnCl 2 for 24 h compared with the negative control treatment, but the apoptosis rate was lower in the BNIP3-shRNA-transfected cells (Fig. 5c). These results suggest that BNIP3 shRNA could inhibit MnCl 2 -induced ROS generation and mitochondrial membrane potential loss, leading to dopaminergic neuron toxicity. Taken together, these data demonstrate that BNIP3 mediates MnCl 2 -induced mitophagy and neurotoxicity in SH-SY5Y cells through oxidative stress.

Discussion
The present study revealed for the first time that the mitophagy receptor protein BNIP3 mediated MnCl 2 -induced mitophagy in SH-SY5Y cells through ROS. This function may be one of the mechanisms responsible for MnCl 2 -induced neurotoxicity, suggesting that targeting BNIP3 is a potential strategy for the treatment of manganism and parkinsonism.
Manganese is widely distributed in the Earth's crust, and it is one of the essential trace elements in the human body. Although manganese is associated with many biological functions, excessive exposure to manganese can cause damage to many organs and lead to manganism, resulting in neurodegenerative diseases and other related diseases. However, the molecular mechanism of this pathogenesis is unclear. Mitochondria play a key role in mammalian cells, especially in mitotic cells, such as neurons. Mitochondria have essential roles in energy balance, signal transduction and cellular metabolism [27][28][29]. Therefore, mitochondrial dysfunction can cause a variety of diseases [30][31][32], especially neurodegenerative diseases. Within a certain range of damaging stimuli, mitochondrial quality control can reverse or reduce neuronal death by promoting the formation of new mitochondria and removing damaged mitochondria [33]. In the absence of appropriate mitochondrial quality control, mitochondrial dysfunction increases and induces excessive mitophagy, leading to neuronal death [34]. Until now, the cause of dopaminergic neuron death remains unclear. Aging and genetic mutations may serve as potential pathogenic factors. In recent years, the environmental factors leading to the occurrence of parkinsonism have been receiving increasing attention from scientists. As early as 1837, movement disorders caused by occupational exposure to manganese led to a preliminary understanding of manganism and manganism-related parkinsonism. Although researchers have long been concerned about the association between manganese and manganism (parkinsonism) [35,36], their shared regulatory mechanisms are still poorly understood. Therefore, further exploration of the mechanisms underlying manganese overexposure-mediated cellular damage will hopefully provide new clues for the treatment of manganism and parkinsonism. BNIP3, Bcl2 interacting protein 3, is localized on the mitochondrial outer membrane, and BNIP3 has a special LIR motif structure that can bind to autophagy marker proteins to regulate mitophagy [37]. BNIP3 has been confirmed to mediate mitophagy under hypoxic conditions and other external stimuli as a mitophagy receptor protein [38]. However, the role of BNIP3-mediated mitophagy in manganese stimulation has rarely been reported. In our study, we found that MnCl 2 could induce mitophagy in SH-SY5Y cells, thereby promoting the generation of ROS and triggering the apoptosis of SH-SY5Y cells in a dose-dependent manner. The obvious neurotoxic effect of Mn highlights the importance of clarifying its mechanism of action. Here, for the first time, we demonstrated that BNIP3 could mediate MnCl 2 -induced mitophagy through ROS in SH-SY5Y cells via serving as a mitophagy receptor protein. It has been suggested that BNIP3 promotes apoptosis by antagonizing the anti-apoptotic effect of E1B19kDa and Bcl2 through the BH3 domain [39]. This study also confirmed that BNIP3 mediated MnCl 2 -induced neurotoxicity in SH-SY5Y cells. The above data suggest that BNIP3 mediates manganese-induced mitophagy and leads to neurotoxicity through oxidative stress in dopaminergic neurons. However, there are still some limitations in the present work. For example, the specific regulatory mechanism of BNIP3-mediated mitophagy and neurotoxicity needs further study. In addition, although it is well known that MnCl 2 can lead to manganism and parkinsonism by damaging mitochondria, little is known about the use of drugs to prevent mitochondrial damage or protect mitochondria. Whether the reversal of mitochondrial damage can control the development of diseases is rarely reported and still needs further exploration. Moreover, in our study, 3-MA failed to reverse the elevated ROS level in Mn-challenged SH-SY5Y cells. There are some possible reasons for the inability of 3-MA to reduce ROS generation. For example, 3-MA was reported to have off-target effect besides inhibiting autophagy [40]. Alternatively, a more complicated mechanism exists in Mn-induced ROS generation in SH-SY5Y cells, which needs further clarification.
In conclusion, BNIP3 contributes to manganese-induced mitophagy and neurotoxicity in dopaminergic SH-SY5Y cells through modulating ROS production.