This study was accepted by the Animal Care and Committee of Zhengzhou University. Moreover, this study was conducted in compliance with the guidelines of the National Institutes of Health for the care and the use of laboratory animals. All animals in the study groups survived throughout the experimental period.

Sprague Dawley (SD) rats (180–220 g) were purchased from Zhengzhou University. Animals were divided into four groups including control group, ozone group, control+ L-arginine group and ozone + L-NAME group and there were 20 animals or so in each group. The animals in ozone group were exposed to 2.0 ppm O₃ for 30 min and room air with a natural light cycle for consecutive 12 days and the control group was given the same treatment without ozone stress. Animals in L-arginine group were intravenously injected 60 mg/kg of L-arginine and L-NAME group were intravenously injected 15 mg/kg of L-NAME for consecutive 12 days respectively at the first day after ozone stress while animals in control group and ozone group were injected saline for consecutive 12 days as a place. The animals (4 rats) from different group were observed at different time points (Day 4, Day 8, and Day 12).

To evaluate the severity of lung injury, the right lobe lungs were obtained at Day 4, 8, 12 respectively. Lungs were filled with air and removed from the animal model. Later on it was fixed in 4% paraformaldehyde, embedded in paraffin and cut into 5 μm sections. Lung sections were immunostained with a polyclonal rabbit anti α-smooth muscle actin (α-SMA) antibody at a dilution of 1:200. As a smooth muscle marker, a polyclonal rabbit anti NOS antibody at a dilution of 1:200 and a polyclonal rabbit anti OGG1 antibody (Santa Cruz Biotech, USA) at a dilution of 1:200 at 37°C for 2 h. The slides were washed and incubated with biotinylated goat anti rabbit IgG, Cy3-labeled or FIFC-labeled lgG (Boster, Wuhan, China) for 1 h and washed again. After washing in PBS, the signal was detected with 3, 3′-diaminobenzidine (Dingguo, Beijing, China) or fluorescence microscope.

Rats were injected with an intraperitoneal injection of sodium pentobarbitone (100mg/kg). The BALF was obtained by washing the airway with 1mL of PBS into the lungs. The recovery is 80%. ELISA were performed on BALF using kits for rat-specific TNFα, TGFβ1, IL10 (Boster, Wuhan, China) and MPO ELISA Kits (CUSABIO, MD, USA), according to the manufacturer's instructions. Lung tissues were than homogenized and centrifuged at 10,000 g for 5 min and the level of OGG1 was determined using ELISA provided by CUSABIO (USA). The DNA was extracted from lung tissues and the amount of 8-hydroxy-2-deoxyguanosine (8-ohdG) was determined using ELISA.

DCFH-DA (Dichloro-dihydro-fluorescein diacetate) is a quantitative method for oxidative stress assessment. Its Probes was used to measure the ROS according to the manufacturer's protocol. The lung tissue after the proper treatment were incubated in pre-warmed PBS containing the probes in a final working concentration of 5 μM for 30 min at 37°C. Flow cytometry detected the ROS fluorescence intensity (488/518 nm).

Arginase is a manganese containing enzyme. It converts L-arginine into urea and L-ornithine. The supernatant of lung tissue homogenate was reacted with reaction solution (containing L-arginine) to detect the product of reaction, L-ornithine and the absorbance was measured at 570 nm according to an Arginase Activity Assay kit (BioVision, CA, USA). The final products of nitrates from NOS reaction with L-arginine, represented the activity of NOS, was measured at 540 nm according to a NOS Activity Assay kit (Novus Biologicals, CO, USA). The concentrations were determined by standard curves.

After protein concentrations of lung tissue homogenate were determined with a coomassie plus protein assay kit (Dingguo, Beijing, China). Then, 75 μg of proteins were separated by gel electrophoresis and then transferred onto PVDF membrane. Then membranes were blocked with 5% skim milk and incubated with rabbit polyclonal anti-Arginase I, anti-NOS and anti-OGG1. Later on membranes were washed and incubated with corresponding secondary IgG for 1 h at room temperature. Lastly, antibody–antigen complexes were detected by using ECL chemiluminescent detection system. β-actin was used as a loading control. A densitometry analysis was performed by AlphaEase software (version: 2200).

The data are presented by mean ± S.E. The statistical analyses were performed by using ANOVA for comparison between same group and the multiple treatment groups. The relationships between variables were explored by simple correlation and multiple regression analyses. P < 0.05 was considered as statistically significant.

The results of microscopic immunohistochemistry staining images of lung sections showed that the bronchial and alveolar wall are complete, no inflammatory cell infiltration, and luminal stenosis were observed in normal control group (at Day 0, Figure 1A). However, in ozone stress group, infiltration of inflammatory cells in bronchial lumen, alveolar cavity and pulmonary interstitial were obviously increased. And some bronchial epithelial cells were found to fall off and the expression of smooth muscle biomarker was found to increase with time (Figure 1A). L-arginine had obvious protective effects on airway inflammation and structure at different time points and L-NAME obviously promoted the remodeling (Figure 1A).

In order to probe the possible molecular signal mechanism of NOS signaling, several major representative regulatory mediators involved in airway injury and remodeling were measured using ELISA. The results showed that the secretion of TGF-β 1 was enhanced from Day 4 after ozone stress and kept augmented during ozone stress and L-NAME promoted the production of TGF-β 1 at Day 12 after ozone stress when compared with ozone group (Figure 1B); the secretion of TNFα increased rapidly to peak at Day 4 and gradually decreased from peak after ozone stress, and L-arginine and L-NAME had no influence on the secretion of TNFα (Figure 1B); the activities of MPO had the same tendency with TNFα after ozone stress and L-NAME inhibited the activities of MPO at Day 4 and L-arginine inhibited the activities of MPO at Day 12 after ozone stress (Figure 1B); the secretion of IL-10 increased gradually from Day 8 after ozone stress and L-arginine inhibited the production of IL-10 at Day 8 and Day 12, while L-NAME promoted the production of IL-10 at Day 4 and Day 8 (Figure 1B).

The ROS assay showed that ozone markedly enhanced the ROS production at different time points as determined using flow cytometry analysis (Figure 2A). Both L-arginine and L-NAME seem to inhibit the production of ROS (Figure 2B).

By using ELISA assay, our results showed that the contents of arginase increased with time (Day 8 and Day 12) in ozone group. L-arginine promoted the production of arginase at Day 8, but L-NAME promoted the production of Arginase at Day 4 and Day 12 after ozone stress (Figure 3A). On the other hand, the contents of NOS increased to the peak at Day 4 and decreased to the normal level from Day 8 in the ozone group. L-arginine promoted the production of NOS at Day 8 and Day 12, but L-NAME inhibited the production of NOS at Day 4 (Figure 3B). These in vivo data indicated that L-arginine promoted the ozone-inhibited NOS activity and L-NAME inhibited the production of NOS.

NOS were also assayed by western blot. Our data indicated that NOS levels initially increased at Day 4 after ozone stress and gradually decreased to normal level, but after treatment with L-arginine, NOS levels increased with time and after treatment with L-NAME, the production of NOS was inhibited from Day 4 when compared with the same time point of ozone group (Figure 3C). NOS was mainly distributed in epithelial cells and endothelial cells (Figure 3D).

8-OxoG is the most common DNA base lesions induced by ROS; so it is most commonly used as an indicator for oxidative stress (Sassa et al., 2016). With the help of ELISA assay, the 8-oxoG levels in DNA was evaluated in which specific antibodies against 8-ohdG (a nucleoside of 8-oxoG) was used. The 8-ohdg Levels were markedly increased in ozone group than control group at Day 0; But, L-arginine reduced the levels of 8-ohdG at Day 12, while L-NAME promoted the production of 8-ohdG at Day 8 and Day 12 after ozone stress (Figure 4A).

OGG1 is a constituent of the base excision repair (BER) system of DNA bases which primarily excizes the 8-oxoG base. The activities of the OGG1 were determined by using ELISA. The OGG1 level were lower in ozone group compared to control group at Day 0; But, L-arginine promoted and restored the levels of OGG1 from Day 4 and L-NAME inhibited the production of OGG1 at Day 8 and Day 12 after ozone stress (Figure 4B). Moreover, lung tissues were subjected to western blotting using anti-OGG1 antibodies. A 38 kDa protein corresponding to OGG1 were detected and decreased OGG1 were detected in ozone-induced lung injury animal models. L-arginine treatment promoted and restored the production of OGG1 at Day 8 and Day 12 and L-NAME inhibited the production of OGG1 from Day 4 after ozone stress (Figure 4C). By using immunofluorescence, we found OGG1 obviously decreased in ozone-stressed lung injury models, which was reversed by administration of L-arginine and aggravated by administration of L-NAME. Besides, OGG1 was mainly localized in epithelial cells and endothelial cells and exhibited cytoplasmic immunoreactivities (Figure 4D).

Correlated analysis results showed NOS was positively associated with OGG1, while negatively correlated to 8-ohdG contents. In multivariate linear model, the positive relationship between NOS and OGG1 level was persistent during injury process (Figures 5A,B). On the other hand, we found that there was no correlation between the expression of arginase and 8-ohdG/OGG1 (Figures 5C,D). These data indicated that NOS signaling was closely related with 8-ohdG/OGG1.

To further investigate the association between NOS and OGG1, NOS siRNA or negative control was transfected into human bronchial epithelial cells. After confirming the successful downregulation of NOS in cells by using western blot and indirect immunofluorescence (IFA), the expression of OGG1 was examined by western blot and IFA (Figure 5E). The results showed that OGG1 expression was down-regulated by NOS knockdown, suggesting that NOS signaling was closely related with OGG1.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.