High-Dose Paraquat Induces Human Bronchial 16HBE Cell Death and Aggravates Acute Lung Intoxication in Mice by Regulating Keap1/p65/Nrf2 Signal Pathway

Seven to 8 weeks male C57BL/6 mice (the total number was 20 and the weight of the mice ranged from 20 to 25 g) were purchased from Vital River Laboratory Animal Technology (Beijing, People’s Republic of China). All animal care and handling procedures were in accordance with the guidelines of the Institute for Laboratory Animal Research of the Second Affiliated Hospital of Kunming Medical University. The animal for research use was approved by the Animal Care and Use Committee of the Second Affiliated Hospital of Kunming Medical University. The animals were fed and maintained under the same conditions (temperature 23 ± 2 °C, humidity 55 ± 5%, 12:12 h light/dark cycle) in the Animal Research Center of Kunming Medical University. The mice were selected and divided into four groups and administered with 500 μM of PQ (Sigma-Aldrich, St. Louis, MO, USA) or the same volume of saline (control) by intraperitoneal injections. After 96 h, mice were sacrificed and the samples of lung tissues and peripheral blood were collected for the following studies.

The Masson Staining Kit was purchased from Shanghai Bogoo Biotechnology Co., Ltd. (China). Mice lung tissues were collected and fixed in 4% paraformaldehyde for 24–48 h and embedded in paraffin wax for long-term preservation. Wax was then cut into 5 μM thicknesses. The paraffin-embedded sections were then de-paraffinized by xylene, and a series of ethanol were used to process and dehydrate the sections. The sections were then washed with phosphate buffer solution (PBS) three times, and the tissues were then stained with reagent A, B, and C sequentially according to the manufacturer’s protocol; optical microscope was used to observe tissue morphologies.

16HBE cells were diluted into the density of 1 × 10⁴/ml and seeded into 96-well plates; cells were then cultured under the standard conditions (37 °C, 5% CO₂) for 12 h. Different doses of PQ (50 μM, 150 μM, and 500 μM) were added into the wells (each assay has three repetitions) and co-cultured with cells for 12 h, 24 h, 48 h, 72 h, and 96 h respectively. Ten microliters of CCK-8 solution was added into each well and incubated in the plate for 1–4 h in the incubator following the manufacturer’s instructions of Cell Counting Kit-8 (MedChemExpress Co., Ltd., USA). Before reading the plate, the plate was gently mixed on an orbital shaker for 1 min; the optical density (OD) values were detected by a Gemini EM microplate reader (Molecular Devices, USA) at the absorbance of 450 nm. The OD values were used to evaluate cell proliferative abilities after being treated by different doses of PQ at various time points.

After treating 16HBE cells and mice with different doses of PQ, TRIzol kit (Invitrogen, USA) was used to extract RNA from 16HBE cells or mice lung tissues following the manufacturer’s protocol. Reverse transcription PCR by iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA, USA) and real-time quantitative PCR by HiScript II Q Select RT SuperMix (Vazyme, China) were used to reverse and quantify relative GAPDH, Keap1, p65, Nrf2, IL-4, IL-6, IL-1β, and TNF-α expressions; the primers were designed and synthesized by Sangon Biotech Co., Ltd. (Shanghai, China); primers are listed in Table 1. Relative mRNA expression levels of genes were normalized by GADPH.

The samples of 16HBE cells and PQ-treated mice were collected, total protein was extracted using RIPA lysis buffer (Beyotime Biotechnology, Shanghai, China), and the protein concentration was determined with BCA Protein Assay Kit (Beyotime Biotechnology, Shanghai, China). Protein was then solubilized in 2× sample buffer; SDS-polyacrylamide gel electrophoresis was then performed to separate the targeted proteins. The proteins were then transferred to polyvinylidene difluoride (PVDF) membranes (Bio-Rad, Hercules, USA) and the membranes were incubated with 5% bovine serum albumin (BSA) for 60 min at the room temperature. The primary antibodies of anti-Keap1 (1:1000, #K2769, Sigma, USA), anti-p65 (1:1000, #P0068, Sigma, USA), anti-Nrf2 (1:500, #SAB4501984, Sigma, USA), anti-p21 (1:1000, #SAB4500065, Sigma, USA), anti-Cyclin A2 (1:1000, #C4710, Sigma, USA), anti-Cyclin D1 (1:1000, #C7464, Sigma, USA), anti-Bax (1:500, #B8429, Sigma, USA), anti-Bcl-2 (1:1000, #B3170, Sigma, USA), anti-Caspase 3 (1:500, #C8487, Sigma, USA), anti-LC-3 (1:1000, #8918, Sigma, USA), and anti-β-actin (1:1000, #A5441, Sigma, USA) were employed to be incubated with the membranes overnight at 4 °C. The anti-mouse IgG-peroxidase-conjugate secondary antibodies (Sigma, USA) were used to be incubated with the membrane for 2 h at the room temperature to combine to the primary antibodies. The bands were visualized by Enhanced Chemiluminescence Kit (Bio-Rad, USA) and ChemiDoc (Bio-Rad, USA); the expressions were quantified by ImageJ software. Each assay was repeated for at least 3 times to eliminate deviations.

CRISPR-Cas9 technology was conducted to knockout Nrf2 gene according to the protocol from Dr. Zhang F. The two sgRNAs targeting human Nrf2 were designed and constructed into pSgRNA (addgene#47108) using Bbs1 digestion (sgRNA sequences, F1:CAATTAAGGCATGGAATTCCCAT, R1: AAACCTATTCCCAGAGTCAGTCA; F2: CATTCCTCCAGGACCCTAGGATGCAGT, R2: AAACGTTCGAACTCTGAC-GGTA). In addition, the constructed plasmids and pCas9-GFP (addgene#44719) were transfected into 16HBE cells by Lipofectamine 3000 transfection reagent (#13778150, Invitrogen, USA) according to the manufacturer’s instructions. Approximately 1 × 10⁶ cells were transfected with 1 μg pCas9 plasmids, 1 μg pSgRNAs, and 0.2 μg PKG-puro plasmid with puromycin resistance gene. Puromycin (1.5 μg/ml) was then co-cultured with cells to select puromycin resistance cells. Total proteins were then extracted from the cells and Western Blot was used to verify the transfection efficiency of the plasmids.

To overexpress cellular p65 expression levels, the coding sequence (GenBank S82307.1) for the human p65 gene was obtained from the total RNA of human HEK293T cells isolated by using TRIzolreragent (Takara Bio, Kusatsu, Japan) and then transcribed using iScript cDNA Synthesis Kit (Bio-Rad, Hercules, CA, USA). The sequences were then cloned into the pEGFP-N1 gene overexpression vectors, and the pEGFP-N1-p65 overexpression vectors were transfected into 16HBE cells by Lipofectamine 3000 transfection reagent (#13778150, Invitrogen, USA) according to the manufacturer’s instructions; the pEGFP-N1 empty vectors were used as the negative control. Western Blot was used to verify that the vectors were successfully transfected into the target cells.

16HBE cells were treated with 500 μM of PQ for 0 h, 12 h, 24 h, and 48 h; L-012 dye was used to detect extracellular NADPH oxidase-derived superoxide. In brief, 16HBE cells were diluted into approximately 4–6 × 10⁴ cells/well into 96-well plates (Thermo, USA) in phenol-free DMEM medium (Sigma, USA) with L-012 at the concentration of 500 μM according to our preliminary experiments (data not shown) for 10 min and luminescence was detected by a Gemini EM microplate reader (Molecular Devices, USA) at the excitation wavelength of 488 nm and emission wavelength of 525 nm respectively. Cellular ROS levels were next measured by dihydroethidium (DHE) staining. Cells were washed with PBS twice and diluted; 10 μM of DHE (Invitrogen, USA) was selected according to our preliminary experiments (data not shown) to incubate with the cells for 30 min at 37 °C without light exposure. After incubation, cells were washed with PBS and DM500 fluorescence microscope (Leica, Germany) was employed to observe ROS productions. The fluorescence intensity was quantified and calculated by ImageJ software.

All the data collected in our experiments was showed as the mean ± standard deviation (SD), and the data was analyzed by SPSS 13.0 statistical software with one-way analysis of variance (ANOVA) for multiple groups and Student’s t test for two groups. P < 0.05 means statistical significance.

Previous study proved that PQ induces cell intoxication in a dose-dependent manner [20]. In our study, 16HBE cells were treated with 50 μM, 150 μM, and 500 μM PQ for 12 h, 24 h, 48 h, and 72 h respectively; the results showed that the inhibiting effects of PQ on cell proliferation were positively correlated with PQ concentration and treating time (Fig. 1a–b), which were in line with the previous study [9]. The Western Blot results also showed that PQ (150 μM, 48 h) significantly increases p21 and decreases cyclin A2 as well as cyclin D1 in 16HBE cells (Fig. 1c–d), which indicated that cell cycle arrest is induced by PQ treatment. After treating cells with different concentrations of PQ (50 μM, 150 μM, and 500 μM) for 2 h, 4 h, and 8 h, respectively, the FCM results showed that PQ induces cell death and apoptosis in a dose-dependent manner (Fig. 1e). Of note, PQ specifically induces cell necrosis and late apoptosis instead of early apoptosis in 16HBE cells (Fig. 1f–h). In addition, pro-apoptotic proteins caspase 3 as well as Bax are increased, and anti-apoptotic protein Bcl-2 is decreased in 16HBE cells treated with 150 μM of PQ for 48 h compared with the control group (Fig. 1i–j).

According to the previous studies, ROS generation was pivotal for PQ-induced cell intoxication and lung fibrosis [7, 21]. Given the fact that ROS generation induces cell death and apoptosis [22], we next explored the extracellular superoxide release and ROS production in 16HBE cells treated with PQ. After treating 16HBE cells with PQ (50 μM, 150 μM, and 500 μM) for 48 h, L-012 assay results showed that extracellular superoxide production was increased by PQ treatments in a dose-dependent manner (Fig. 2a–b). The DHE staining data also showed the similar results; low doses (50 μM and 150 μM) of PQ slightly increase ROS levels at 12 h and 24 h, while high dose (500 μM) of PQ significantly increases ROS levels at 12 h (Fig. 2c–h), which indicated that PQ also induces ROS production in a dose-dependent manner, and our results were in accordance with the previous study [20].

Cell autophagy has been reported to be closely related with PQ-induced cell intoxication [6]; since autophagy has dual effects on cells’ responses to environmental stress including ROS generation [23, 24], we next explored PQ’s influences on cell autophagy by electronic microscope and Western Blot. 16HBE cells were treated with different doses of PQ (50 μM and 500 μM) for 8 h; the images of electronic microscope showed that low dose of PQ (50 μM) induces autophagosome production in 16HBE cells, and the number of autophagosome dramatically decreased in 16HBE cells by treating cells with 500 μM of PQ (Fig. 3a). 16HBE cells were next treated with different doses of PQ (50 μM, 150 μM, and 500 μM) for 0 h, 12 h, 24 h, 48 h, and 96 h respectively; Western Blot results showed that 50 μM of PQ significantly increases LC3-II/LC3-I ratios and decreases p62 levels at 24 h, 48 h, and 96 h compared with the control group (Fig. 3b–d). Notably, compared with the control group, 150 μM of PQ merely increases LC3-II/LC3-I ratios at 12 h, which significantly decreases at 24 h, 48 h, and 96 h (Fig. 3e–f); p62 is also downregulated by 150 μM of PQ at 12 h (Fig. 3e, g). In addition, 500 μM of PQ has little impacts on either LC3-II/LC3-I ratios or p62 at any time point (Fig. 3h–j).

Our results showed that ROS generation plays an important role in PQ-induced cell intoxication, since Keap1/Nrf2 signal pathway is closely related with the regulation of ROS production [25], and p65 might influence Nrf2 by binding to its inhibitor Keap1 [17]. We next investigated the involvement of p65 and Keap1/Nrf2 signal pathway in 16HBE cells treated with PQ. 16HBE cells were treated with various doses of PQ (50 μM, 150 μM, and 500 μM) for 48 h; real-time qPCR results showed that 50 μM and 150 μM of PQ has little effects on Keap1 expressions compared with the control group, but Keap1 mRNA expression levels are significantly increased by treating 16HBE cells with 500 μM of PQ (Fig. 4a). Notably, 50 μM of PQ significantly increases p65 and Nrf2 mRNA expressions, which are surprisingly decreased by high doses of PQ (150 μM and 500 μM) (Fig. 4b–c). Further Western Blot results also showed that Keap1 is merely increased by treating cells with 500 μM of PQ. Besides, p65 and Nrf2 are upregulated by 50 μM of PQ, which are downregulated by treating 16HBE cells with high doses of PQ (150 μM and 500 μM) (Fig. 4d–e). The results also showed that Keap1 is not affected by p65 overexpression or knock-down (Fig. 4f–g). However, Nrf2 is significantly upregulated by p65 overexpression and downregulated by p65 knock-down compared with the control group (Fig. 4f–g). In addition, overexpressed p65 promotes Nrf2 downstream targets Nqo1 and Gclc expressions, which are reversed by synergistically knocking down Nrf2 (Fig. 4h–i). Furthermore, overexpressed p65 reverses 500 μM of PQ’s inhibiting effects on Nrf2 (Fig. 4h–i).

To explore the role of Keap1/p65/Nrf2 signal pathway in PQ-induced cell intoxication, we first overexpressed p65 or synergistically overexpressed p65 and knocked down Nrf2 in 16HBE cells (Fig. 5a–b). By treating cells with 500 μM of PQ for 12 h, 24 h, 48 h, 72 h, and 96 h, we found that cell proliferation is significantly inhibited by PQ in a time-dependent manner (Fig. 5c). Of note, overexpressed p65 alleviates PQ’s inhibiting effects on cell proliferation, which are reversed by synergistically knocking down Nrf2 (Fig. 5c–d). The FCM results showed that PQ (150 μM, 2 h) apparently increases 16HBE apoptosis ratio (Fig. 5e). Similarly, overexpressed p65 attenuates cell apoptosis, which is reversed by synergistically knocking down Nrf2 (Fig. 5e–f). In addition, compared with the control group and PQ-alone-treated group (500 μM), overexpressed p65 increases LC3-II/I ratio and decreases p62, which are also reversed by synergistically knocking down Nrf2 (Fig. 5g–i).

To investigate the involvement of Keap1/p65/Nrf2 signal pathway activation in PQ-induced cell intoxication and lung fibrosis by in vivo experiments, male C57BL/6 mice were administered with 500 μM of PQ for 96 h to establish PQ-induced lung injury mice models. We first verified that we have successfully overexpressed p65 and knocked down Nrf2 in mice models (Fig. 6a–b). Masson staining images showed that lung fibrosis is induced by high-dose PQ treatment. Overexpressed p65 alleviates PQ-induced tissue morphology destruction, which is reversed by synergistically knocking down Nrf2 (Fig. 6c). PQ-induced lung fibrosis has also been reported to be seriously aggravated by inflammatory reactions; to investigate the role of Keap1/p65/Nrf2 signal pathway in regulating inflammatory reactions, real-time qPCR was used to detect inflammatory cytokine mRNA expression levels in lung tissues and ELISA was employed to detect their expressions in mice periphery blood (Fig. 6d–e). The results showed that high dose of PQ increases IL-4, IL-6, IL-1β, and TNF-α expressions in both mice lung tissues and periphery blood (Fig. 6d–e). Similarly, overexpressed p65 decreases IL-4, IL-6, IL-1β, and TNF-α levels in mice, which are reversed by knocking down Nrf2 (Fig. 6d–e). In addition, we found that PQ increases Bax and caspase 3 decreases Bcl-2 in mice tissues. Overexpressed p65 reverses PQ’s effects on the apoptosis-associated proteins, which are abrogated by synergistically overexpressing Nrf2 (Fig. 6f–g). Furthermore, overexpressed p65 also decreases p21 and increases cyclin A2 as well as cyclin D1 in mice compared with the PQ-treated group, which are also reversed by knocking down Nrf2 (Fig. 6h–i).