Ultrafine silicon dioxide nanoparticles cause lung epithelial cells apoptosis via oxidative stress-activated PI3K/Akt-mediated mitochondria- and endoplasmic reticulum stress-dependent signaling pathways

Silicon dioxide nanoparticles (SiO2NPs) are widely applied in industry, chemical, and cosmetics. SiO2NPs is known to induce pulmonary toxicity. In this study, we investigated the molecular mechanisms of SiO2NPs on pulmonary toxicity using a lung alveolar epithelial cell (L2) model. SiO2NPs, which primary particle size was 12 nm, caused the accumulation of intracellular Si, the decrease in cell viability, and the decrease in mRNAs expression of surfactant, including surfactant protein (SP)-A, SP-B, SP-C, and SP-D. SiO2NPs induced the L2 cell apoptosis. The increases in annexin V fluorescence, caspase-3 activity, and protein expression of cleaved-poly (ADP-ribose) polymerase (PARP), cleaved-caspase-9, and cleaved-caspase-7 were observed. The SiO2NPs induced caspase-3 activity was reversed by pretreatment of caspase-3 inhibitor Z-DEVD-FMK. SiO2NPs exposure increased reactive oxygen species (ROS) production, decreased mitochondrial transmembrane potential, and decreased protein and mRNA expression of Bcl-2 in L2 cells. SiO2NPs increased protein expression of cytosolic cytochrome c and Bax, and mRNAs expression of Bid, Bak, and Bax. SiO2NPs could induce the endoplasmic reticulum (ER) stress-related signals, including the increase in CHOP, XBP-1, and phospho-eIF2α protein expressions, and the decrease in pro-caspase-12 protein expression. SiO2NPs increased phosphoinositide 3-kinase (PI3K) activity and AKT phosphorylation. Both ROS inhibitor N-acetyl-l-cysteine (NAC) and PI3K inhibitor LY294002 reversed SiO2NPs-induced signals described above. However, the LY294002 could not inhibit SiO2NPs-induced ROS generation. These findings demonstrated first time that SiO2NPs induced L2 cell apoptosis through ROS-regulated PI3K/AKT signaling and its downstream mitochondria- and ER stress-dependent signaling pathways.


SiO 2 NPs induces apoptosis in L2 alveolar epithelial cells. To investigate the harmful effects of
SiO 2 NPs in lung cells, L2 alveolar epithelial cells were used. Cells were treated with SiO 2 NPs (10-300 μg/mL) for 24 and 48 hours. Results showed that SiO 2 NPs (100 μg/mL) significantly decreased cell viability after 24 and 48 hours treatments. Moreover, the SiO 2 NPs induced cytotoxicity in L2 alveolar cells in a dose-and time-dependent manner (Fig. 1A). We next tested the mRNA expressions of surfactants. Results showed that surfactant protein (SP)-A, SP-B, SP-C and SP-D mRNA levels were significantly reduced after 48 hours treatment of SiO 2 NPs (100 μg/mL) (Fig. 1B). We also examined the intracellular Si levels to clarify whether SiO 2 NPs could enter intracellular space. Results showed that the intracellular levels of Si in L2 cells treated with SiO 2 NPs (50-300 μg/mL) were increased in a dose-dependent manner (Fig. 1C). We next investigated the effect of SiO 2 NPs on apoptosis in L2 cells treated with SiO 2 NPs (50-300 μg/mL) for 24 and 48 hours. Results showed that SiO 2 NPs markedly increased annexin-V fluorescence ( Fig. 2A) and caspase-3 activity (Fig. 2B). This increased caspase-3 activity by SiO2NPs could be reverse by caspase-3 inhibitor Z-DEVD-FMK (Fig. 2D). We also analyzed the expressions of apoptosis-related proteins in L2 cells treated with SiO 2 NPs (100 μg/mL) for 24, 36 and 48 hours. The SiO 2 NPs significantly increased cleaved-poly (ADP-ribose) polymerase (PARP), cleaved-caspase-9 and cleaved-caspase-7 protein expression (Fig. 2C). These results suggested that SiO 2 NPs was capable of inducing cytotoxicity and apoptosis in L2 cells.

SiO 2 NPs induces ROS production and mitochondria-and ER stress-related signals in L2 alveolar epithelial cells.
We next investigated the potential mechanisms of SiO 2 NPs-induced cytotoxicity in L2 cells treated with SiO 2 NPs (50-300 μg/mL) for 45 minutes to 3 hours. The ROS production was analyzed by flow cytometry. Results showed that SiO 2 NPs increased ROS production in a dose-and time-dependent manner (Fig. 3A). The SiO 2 NPs (100 μg/mL) treatments also decreased mitochondrial transmembrane potential (MMP) (Fig. 3B) and increased cytosolic cytochrome c release (Fig. 3C). In the investigation of mitochondria disruptive signals, Bax protein expression was increased, and Bcl-2 protein expression was decreased in L2 cells after SiO 2 NPs (100 μg/mL) treatment (Fig. 3D). Moreover, SiO 2 NPs also caused the increase in Bid, Bak, and Bax mRNA expressions and the decrease in Bcl-s mRNA expression in L2 cells (Fig. 3E). In the investigation of ER stress-related signals, SiO 2 NPs significantly increased the protein expression of CHOP, X-box binding protein-1 (XBP-1), and phospho-eIF2α, and reduced the protein expression of pro-caspase-12 (Fig. 3F). These results suggested that SiO 2 NPs induced cell apoptosis via ROS-, mitochondria-, and ER stress-related pathways.

Discussion
The nanosized paticles, which diameters was less than 100 nm, are easily exposure to human by various routes, such as inhalation (respiratory tract), ingestion (gastrointestinal tract), dermal (skin), and injection (blood circulation) 29 . SiO 2 NPs have applications in many industrial and medical areas. SiO 2 NPs have been found to cause adverse effects in human, such as lung fibrosis, chronic bronchitis, chronic obstructive pulmonary disease (COPD), and lung cancer [30][31][32] . In the present study, we elucidated the mechanisms of SiO 2 NPs-induced cytotoxicity in L2 alveolar cells that SiO 2 NPs-induced cell apoptosis via the ROS-activated PI3K/AKT signaling-mediated mitochondria-and ER stress-dependent signaling pathways.
The SiO 2 NPs-induced systemic toxicity is controversy. In a food additives study, rodents with oral administration of silica nanoparticles at a dose of 2500 mg/kg body weight did not cause the adverse health effects 33 . Beside, in a study of subacute inhalation toxicity test, exposure of rats with silica nanoparticles (0.407 ± 0.066 mg/m 3  www.nature.com/scientificreports www.nature.com/scientificreports/ to 5.386 ± 0.729 mg/m 3 ) for 28 days did not find the histological changes in lung tissues and the inflammatory responses in bronchoalveolar lavage fluid 34 . Yet, other studies demonstrated that oral exposure of silica nanoparticles induced liver injuries, including fatty liver, periportal liver fibrosis, and liver weight decrease 33,35,36 . van der Zande et al. 36 have shown that no obviously toxic effects after animals feed with silica nanoparticles (size: 5-200 nm) 100 to 2500 mg/kg body weight were observed after 28 days exposure; however, some adverse health effects were observed after 84 days of exposure, including the increases in serum alanine aminotransferase level, lipid droplets, and periportal liver fibrosis. A study has also demonstrated that lung tissue is a major site for 125 I labeled silica nanoparticles accumulation in mice after intravenous injection 37 . Treatment with pure silica nanoparticles (size: 50 and 100 nm) in human lung alveolar epithelial cells at the concentrations of 50 to 100 μg/ mL has been shown to induce ROS generation, DNA fragmentation, and genotoxicity 38 . In the present study, we used SiO 2 NPs (size: 12 nm) 10 to 300 μg/mL to treat normal lung epithelial cells L2 for 24 and 48 hours. SiO 2 NPs induced cytotoxicity at the concentrations of 50-300 μg/mL in a dose-dependent manner. SiO 2 NPs (50-300 μg/ mL) could also significantly increase the intracellular Si levels. These results indicated that SiO 2 NPs possessed cytotoxic effect and accumulative potential in lung cells as previous findings mentioned above.
The cell apoptosis is known to involve the extrinsic death receptor pathway and the intrinsic mitochondrial pathway. The mitochondria related apoptosis is resulted from the mitochondrial permeability transition pore opening and decrease mitochondrial transmembrane potential in inner mitochondria membrane in which cytochrome c can be released to cytosol and trigger the caspases-related apoptosis 39,40 . SiO 2 NPs has been found to induce cytotoxicity resulted from mitochondrial-related apoptosis in skin cancer A431 cells and lung cancer A549 cells 16 . Moreover, previous study has shown that ROS is involved in many phases of mitochondria-related apoptosis 41 . In the present study, we found that ROS elevation by SiO 2 NPs triggered mitochondrial damage, resulting mitochondrial transmembrane potential loss, cytochrome c release, and cleavages of PARP and caspases 3,7, and 9 in L2 cells. The antioxidant NAC effectively reversed the SiO 2 NPs-induced ROS-triggered mitochondria damage and apoptosis. These results suggest that oxidative stress-regulated mitochondria damage is an important risk factor in SiO 2 NPs-induced lung cell apoptosis.
Under ER stress conditions, ER-chaperone protein 78 kDa glucose-regulated protein (Grp78/BIP) is released from three dominant stress sensors, including inositol-requiring protein 1 (IRE1), PKR-like endoplasmic reticulum kinase (PERK), activating transcription factor-6 (ATF-6). Subsequently, the spliced form of XBP-1 is produced by IRE1 activation that triggers Grp78/BIP and CHOP expression. Activation of PERK phosphorylates eIF2α and increases ATF4 translation 42 . It had been shown that ER-stress was associated with various lung disorders, such as lung cancer, lung fibrosis, asthma, and lung injury 43 . Inhalation or intra-tracheal instillation of titanium dioxide, silver, or zinc oxide nanoparticles have been shown to induce ER stress 26,27,44 . However, the role www.nature.com/scientificreports www.nature.com/scientificreports/ of ER stress in SiO 2 NPs-induced cytotoxicity in lung epithelial cells remains unclear. In the present study, we found that ER stress-related proteins, including CHOP, XBP-1, eIF2α and caspase-12, can be upregulated or activated by SiO 2 NPs, indicating that ER stress-related signaling pathway might also contribute to SiO 2 NPs-induced cytotoxicity in L2 cells.
PI3K is a lipid kinase that involved in cell metabolism, proliferation, survival, and death 45 . AKT activation, which occurs downstream of PI3K, is known to increase ROS generation and accelerate ROS-induced www.nature.com/scientificreports www.nature.com/scientificreports/ apoptosis 19,46 . In the present study, we tested the relationship between PI3K/AKT signaling and ROS generation in SiO 2 NPs-induced lung alveolar cell damage. We found that SiO 2 NPs increased PI3K activity and AKT phosphorylation, which could be significantly reversed by antioxidant NAC and PI3K inhibitor LY294002. However, PI3K inhibitor LY294002 could not inhibit the ROS generation by SiO2NPs. Inhibition of ROS and PI3K/AKT signaling effectively protected lung alveolar cells against SiO 2 NPs-induced cytotoxicity and cell apoptosis. These results suggest that ROS-regulated PI3K/AKT signaling plays an important role in SiO 2 NPs-induced lung alveolar epithelial cell apoptosis.
In conclusion, the present study demonstrates that SiO 2 NPs is capable of inducing lung alveolar epithelial cell apoptosis. We further demonstrate that ROS-regulated PI3K/AKT-mediated mitochondria-and ER stress-dependent signaling pathways are involved in the SiO 2 NPs-induced cell apoptosis. These findings provide basic concerns of molecular mechanisms and possible therapeutic strategies in SiO 2 NPs-induced lung injury.

Materials and Methods
SiO 2 NPs. SiO 2 NPs were purchased from Sigma-Aldrich (Sigma-Aldrich, St. Louis, MO, USA). The characterization of SiO 2 NPs is 12 nm of primary particle size by transmission electron microscopy (TEM) and 99.8% of purity based on traced metal analysis. The SiO 2 NPs stock solution was modified from previous study 47 , and freshly suspended in ddH 2 O at a concentration of 5 mg/ml and then dispersed for 20 min by using a sonicator before used.
Cell culture. The cell culture was performed as described previously 48 . Rat lung epithelial derived L2 cells were purchased from ATCC (CCL-149). Cells were cultured in RPMI-1640 media supplemented with 10% fetal bovine serum and 1% penicillin-streptomycin in 75 cm 2 flask, and under a 5% CO 2 and 95% air mixture at 37 o C in a humid chamber. When growth density was reached 80%, cells were washed twice with PBS, and detached with 0.25% (w/v) trypsin-0.53 mM EDTA solution for 5 to 15 min. The aliquot of cells was added to a new flask or wells for next experiments.  RT-PCR analysis. The mRNAs expression of surfactant was analyzed by real-time quantitative RT-PCR (qPCR) as previously described 49 . Briefly, total intracellular RNA was extracted using RNeasy Mini kit (Qiagen Inc., USA), according to the instructions provided, according to the instructions provided, and was heated to 90 °C for 5 min to remove any secondary structures and then rapidly placed on ice. The samples were reverse transcribed into cDNA using the AMV RTase (reverse transcriptase enzyme, Promega Corporation, Pty. Ltd., USA) system. cDNA (2 μL) was tested with Real-time Sybr Green PCR reagent (Invitrogen, USA) with rat specific primers (as shown in Table 1). The amplification was performed using an ABI StepOnePlus sequence detection system (PE, Applied Biosystems, CA, USA). Data analysis was performed using StepOne software (Version 2.1, Applied Biosystems, CA, USA).

Caspase-3 activity analysis.
Cells were cultured at a density of 2 × 10 5 cells/well and treatment of SiO 2 NPs with or without antioxidant NAC or PI3K inhibitor LY294002 for 24 hours. Subsequently, cells were lysed and cell lysates were incubated with caspase-3/CPP32 substrate, Ac-DEVD-AMC (10 μM) (Promega Corporation,  PI3K activity assay. PI3K activity was executed according to manufacturer's protocol (Active Motif). Cells were cultured in wells with approximately 80% confluent and treated with SiO 2 NPs. After, cells were washed twice of PBS and fixed with 4% formaldehyde in PBS for 20 min at room temperature, and then formaldehyde was removed and washed with wash buffer. Blocking buffer was supplemented with samples and incubated for 1 hour at room temperature. After rinsing with PBS, all samples were incubated with primary phospho-PI3K antibody at 4 °C, overnight. Subsequently, primary antibody was removed and incubated with HRP-conjugated secondary for 1 hour at room temperature. Then, the developing solution was supplemented with each well and incubated for 15 minutes at room temperature. The phospho-PI3K absorbance of 450 nm was read on a spectrophotometer. Western blot analysis. Western blot analysis was performed as described previously 50 . Equal amount of protein samples (50 μg) were resolved on SDS-PAGE and transferred to polyvinylidine difluoride (PVDF)  Table 1. Primer sequences used for the real-time quantitative RT-PCR analysis.