Nifuroxazide ameliorates pulmonary fibrosis by blocking myofibroblast genesis: a drug repurposing study

NIF was purchased from Xiyashiji Chemical Co., Ltd. (Chengdu, Sichuan, China). For the in vitro study, a 20 mM/50 mM NIF solution was prepared in dimethyl sulfoxide (DMSO) (Sigma, St Louis, Mo), stored at − 20 ℃ and diluted in culture medium to the required concentration. When the DMSO concentration was < 1‰, the culture medium was used as the vehicle control; when the DMSO concentration was ≥ 1‰, the same concentration of DMSO was used as the vehicle control. For in vivo studies, NIF was prepared at a 5:35:60 ratio of DMSO: polyethylene glycol 400 (PEG 400) (Sigma, St. Louis, MO, USA): normal saline and administered at a dose of 0.1 mL/10 g body weight. BLM sulphate was purchased from Chengdu Synguider Technology Co., Ltd. (Chengdu, China). Collagenase Type IV was purchased from Gibco (#17104-019; Grand Island, NY, USA). TGF-β1 was purchased from Novoprotein (#CA59; Shanghai, China). Primary antibodies against β-actin (ab8226; Abcam, Cambridge, MA, USA), GAPDH (TA-08; ZSGB-BIO, Beijing, China), α-SMA (ab5694), Collagen-I (ab88147), E-Cadherin (ab76055) and Vimentin (ab20346) were purchased from Abcam (Cambridge, MA, USA), and Smad2/3/phospho-Smad2/3 (#8685; #8828) and Stat3/phospho-Stat3^TY705 (#9139; #9145) were purchased from Cell Signaling Technology Company (MA, USA). PE-CD11b (#12-0112-82), PE-CD4 (#12-0041-82), APC-CD69 (#17-0691-82), and FITC-CD8 (#11-0081-82) were purchased from BD Biosciences (San Diego, CA, USA). PE-F4/80 (#123110), FITC-CD11b (#101206), and FITC-Gr-1 (#108406) were purchased from Biolegend (San Diego, CA, USA).

A549 (human alveolar basal epithelial cells) and NIH/3T3 (mouse embryonic fibroblasts) were purchased from ATCC (Rockville, MD, USA). HPFs (human pulmonary fibroblasts) were purchased from Science Cell (San Diego, CA, USA). The three cell types were cultured in DMEM (Gibco, Grand Island, NY, USA) supplemented with 10% or 20% heat-inactivated foetal bovine serum (FBS; HyClone, Logan, UT, USA) and 1% penicillin/streptomycin (MP Biomedical LLC) in 5% CO₂ at 37 °C.

The three-dimensional X-ray crystal structure of Stat3 (PDB ID: 6NJS) was downloaded from the Protein Data Bank [18]. Both the compound and protein were processed using the CHAR Mm force field [19]. Molecular docking was performed using the CDOCKER module in Accelrys Discovery Studio (version 3.5; Accelrys, San Diego, CA, USA). The molecular docking parameters were determined according to the standard values set by the software. After the docking study was completed, the platform was used to collect the docking score and analyse the docking modes.

The thiazolyl blue tetrazolium bromide (MTT) (Sigma, St Louis, MO) assay was used to evaluate the viability of NIF-treated cells. Cells were cultured in 96-well plates at a density of 1000–8000 cells/well and administered NIF (0–20 μM) 24 h later. Next, after coincubation for 24, 48, and 72 h, 20 μL of 5 mg/mL MTT was added to each well, and cells were further incubated at 37 °C for 2–4 h. A Spectra MAX M5 microplate spectrophotometer (Molecular Devices, Sunnyvale, CA, USA) was used to determine the absorbance at 570 nm.

Cells were cultured at a concentration of 10,000–20,000 cells/well in 24-well plates containing glass slides (14 mm × 14 mm). After 48 h of culture, the medium was discarded and replaced with serum-free DMEM for 6 h. Then, the medium was replaced with complete medium, and 5 ng/mL TGF-β1 was added. After 1 h, NIF (20 μM) was added and incubated for another 24 h. Next, the slides were fixed in 4% paraformaldehyde for 15 min at room temperature followed by washing 3 times with PBS and permeabilizing with 0.5% Triton X-100 (Sigma, St Louis, MO) for 20 min. After washing 3 times with PBS, cells were blocked in 5% BSA (BioFroxx) solution at room temperature for 30 min followed by incubation with primary antibodies (anti‐E‐cadherin (1:200; Abcam, Cambridge, MA, USA) and anti-α-SMA (1:200; Abcam, Cambridge, MA, USA)) overnight at 4℃. The slides were subsequently washed 3 times with PBS (5 min each time) and incubated with the secondary antibody (Cy3‐labelled goat anti‐mouse immunoglobulin G (IgG; Beyotime, Shanghai, China) at a 1:200 dilution (red) or FITC-488‐labelled goat anti‐rabbit immunoglobulin G (IgG; Beyotime, Shanghai, China) at a 1:200 dilution (green)) for 1 h at room temperature. The slides were then washed 3 times with PBS and stained with DAPI (Biosharp, Hefei, China) at room temperature for 10 min. After washing 3 times with PBS, an anti-fluorescence quencher (IgG; Beyotime, Shanghai, China) was added. Immunofluorescence was analysed under a fluorescence microscope (Zeiss LS880, Germany).

A549 cells were inoculated into 6-well plates at a density of 10⁵ cells/well. When the cells had reached approximately 80% confluence, the medium was discarded and replaced with serum-free medium. After incubation for 6 h at 37 ℃, the cells were scratched using 200 μL pipette tips. The cells were washed with PBS to remove cellular debris and cultured in complete medium (less than 3%) containing TGF-β1 (5 ng/mL). One hour later, NIF (20 μM) was added, and images of the scratches were acquired 0 and 24 h later using a microscope (Olympus, ix73, Japan). The wound closure rate was calculated as the ratio of the cell migration area and the original wound area.

Protein detection were performed in accordance with previous reports [20]. Cells were inoculated into Petri dishes (10 cm) and then treated with the same immunofluorescence procedure as those collected after 24 h of NIF administration. Collected tissue was frozen in liquid nitrogen and ground into a powder in a mortar. Lung tissue and cultured cells were homogenized in RIPA buffer (Beyotime, Shanghai, China) supplemented with protease and phosphatase inhibitors (Selleckchem) and quantified using the Bradford method. Then, the proteins were separated on 10% SDS–PAGE gels (Chengdu Baihe Technology Co., Ltd.) and transferred onto 0.45 μM PVDF membranes (Merck Millipore, Billerica, MA, USA). After blocking in 5% (M/V) non-fat milk for 1 h, the membranes were incubated with antibodies overnight at 4 °C. Then, the membranes were incubated with goat anti‐rabbit/mouse IgG (ZSGB-BIO Co., Beijing, China) at a 1:3000 dilution for 60 min at 37 °C. Reactive bands were identified using an enhanced chemiluminescence kit (Merck Millipore, Billerica, MA, USA). Then, the images were analysed using ImageJ software (National Institute of Health, Bethesda, MD, USA).

These methods were performed according to those reported in the literature [21] with some modifications. Cells were cultured in the same manner described above for protein extraction. After treatment with NIF (200 μM) for 2 h, the cells were collected, PBS containing protease inhibitor was added, and the cells were adjusted to the same concentration after cell counting. The cell suspension was divided into 8 equal parts and heated at 42, 44, 46, 48, 50, 52, 54 and 56 °C for 3 min, frozen in liquid nitrogen (1 min) and re-dissolved at room temperature three times. After centrifuging the samples at 13,300 r/15 min, the final protein was obtained by removing the supernatant. Then, protein images were obtained and analysed according to the method described above.

Wistar rats (male; 6–8 weeks; 180–220 g) were purchased from Vital River (Beijing, China). Rats were anaesthetized using 10% chloral hydrate and then intratracheally administered BLM (5 mg/kg). After 14 days, the rats were sacrificed, and complete lung tissue was obtained by dissection. Then, the lungs of the rats were washed with Hanks solution, minced and digested in trypsin. After incubation at 37 °C for 40 min, the cell suspension was filtered through a 70 μM screen mesh. The filtrate was centrifuged at 1500 rpm/5 min and the precipitates were collected, mainly lung fibroblasts and epithelial cells. The precipitate was resuspended in the medium and centrifuged at 800 rpm for 5 min. Take the supernatant (mainly including lung fibroblasts and a few epithelial cells), discard the precipitate (mainly including epithelial cells), centrifuge the obtained supernatant at 1500 r/5 min, and then collect the sediment containing lung fibroblasts [22, 23]. The extracted precipitate containing primary lung fibroblasts were cultured in DMEM/F-12 (GIBCO, NY, USA) medium containing 10% FBS and 1% penicillin and streptomycin and streptomycin in 5% CO₂ at 37℃. After culturing in the incubator for 40 min, discard the non-adherent cells (Lung fibroblasts adhere faster and can be separated from other cells) [24]. The adherent cells are pure lung fibroblasts and cultured in fresh culture medium. This method was also used in the subsequent passage to further purify the extracted lung fibroblasts. Lung fibroblasts were identified by morphology and immunofluorescence (Additional file 1: Fig. S1a, b) [25]. The cells were used between passages 3 and 8.

C57BL/6 mice (6–8 weeks) were purchased from Huafukang (Beijing, China). C57BL/6 mice were anaesthetized using 10% chloral hydrate and then intratracheally administered BLM, and the sham-operated group was administered with the same volume of physiological saline.

In the preventive model, mice were intratracheally administered BLM (~ 2.5 mg/kg) while the sham-operated group was administered with the same volume of physiological saline. All bronchial drip only once at day 0. On the first day of BLM bronchial drip, all BLM mice were randomly divided into 3 groups: n = 14 (Vehicle); n = 10 (NIF 25 mg/kg); n = 14 (NIF 50 mg/kg). The sham operation group (n = 10) and the vehicle group were intraperitoneally injected with dissolvent (5:35:60 ratio of DMSO: polyethylene glycol 400: normal saline) every day for 27 days. At the same time, the NIF groups received intraperitoneal injection of low-dose NIF (25 mg/kg) or high-dose NIF (50 mg/kg) every day. Mice were sacrificed on day 28.

In the therapeutic model, mice were intratracheally administered BLM (~ 1 mg/kg) while the sham-operated group was administered with the same volume of physiological saline. All bronchial drip only once at day 0. On the first day of BLM bronchial drip, the health status of mice was observed, and the previous feeding pattern was maintained until the 13th day after modelling. On the 14th day, all BLM mice were randomly divided into 3 groups. n = 9 (Vehicle), n = 8 (25 mg/kg, 50 mg/kg NIF). The sham operation group (n = 4) and the vehicle group were began intraperitoneally injected with dissolvent (5:35:60 ratio of DMSO: polyethylene glycol 400: normal saline) every day on day 14. At the same time, the NIF groups received intraperitoneal injection of low-dose NIF (25 mg/kg) or high-dose NIF (50 mg/kg) every day. Mice were sacrificed on day 28.

BALF was extracted according to previous methods reported in the literature [26] with some modifications. After 28 days of administration, the mice were anaesthetized using 10% chlorine hydrate. Anaesthetized mice were fixed on a clean foam board, and the skin around the neck was sterilized using alcohol-soaked cotton swabs. The skin of the mouse neck was cut, the neck fat was separated, and the mouse bronchi were located. An indwelling needle was inserted into the mouse bronchus and fixed using cotton thread. Then, 1 mL of normal saline was drawn continuously and slowly 3 times through the indwelling needle to the lung, and the final normal saline was collected as BALF. The recovery rate was greater than 80%.

The collected left lower lung from mice was fixed in 4% paraformaldehyde for one week, dehydrated and embedded in paraffin. The embedded tissue was cut into 3–5 μm thick sections. The tissue sections were stained separately according to the instructions of H&E staining and Masson trichrome staining (BASO, BA-40798). A Panoramic MIDI II 3DHISTECH digital pathology system was used to acquire section images. CaseViewer software was used for processing. Fibrosis was scored as previously described[27]. Fibre volume fractions were determined using ImageJ.

A hydroxyproline determination kit (#A030-2-1, Nanjing Jiancheng Institute of Biological Engineering, Nanjing, China) was used to analyse the hydroxyproline content of lung tissue samples. In brief, tissue was hydrolysed and neutralized using sodium hydroxide. Then, chloramine-T and dimethylaminobenzaldehyde were sequentially added. Finally, a spectrophotometer was used to determine the absorbance at 550 nm to evaluate hydroxyproline content in the lung tissue. Results are expressed as μg/mg wet lung tissue.

IHC staining was performed on lung tissue sections using an IHC staining kit (ZSGB-BIO Co., Beijing, China) as previously described [28]. Paraffin-embedded lung sections were stained with primary antibodies (α-SMA, Collagen I and p-Stat3 (TY705)) using a DAB Detection Kit (ZSGB-BIO Co., Beijing, China).

Lung tissue was disrupted according to a previously described method and lysed with collagenase (37 °C for 90 min) [28]. Changes in the cell ratio were assessed by flow cytometry (BD LSR II) after incubation with different antibodies. The data were analysed using FlowJo software.

The data are presented as the mean ± SD of three independent experiments. Unpaired two-tailed Student’s t tests were used to compare two groups, and one-way ANOVA was performed for multiple group comparisons followed by Tukey’s test. P-values < 0.05 were considered statistically significant; individual P-values are indicated by asterisks in the figures: *P < 0.05, **P < 0.01, ***P < 0.001; ^#P < 0.05, ^##P < 0.01 and ^###P < 0.001. Statistical analysis was performed using GraphPad Prism 6.0 (GraphPad software, San Diego, CA, USA).

Although NIF has previously been reported to be a potential inhibitor of Stat3 [15], we verified this using molecular docking. As shown in Fig. 1a, b, NIF forms intermolecular hydrogen bonds with GLU638 and MET660 of the SH2 domain, stabilizing the complex. Moreover, the benzene ring moiety forms a conjugation with TYR657 and ILE659. Together, these interactions promote the binding of NIF to Stat3. The docking results revealed that the interaction energy score between NIF and Stat3 was 22.4538, indicating that NIF stably binds to Stat3. In addition, as shown in Fig. 1c, d, Stat3 is gradually degraded with increasing temperature, but compared to DMSO, NIF slowed its degradation rate in A549 and NIH/3T3 cell lines, with the temperature required for half Stat3 degradation in the NIF group increasing by 1.8 ℃ and 2.61 ℃ in A549 and NIH/3T3 cells, respectively. It was further confirmed that NIF could binds to Stat3 in these cell lines, indicating that it is a potential inhibitor of Stat3.

First, we evaluated the effect of NIF on BLM-induced pulmonary fibrosis in mice. According to our previous research on NIF [14, 28], intraperitoneal administration (25 mg/kg or 50 mg/kg) was selected for the in vivo experiments. As shown in Fig. 2a, we administered NIF to mice beginning on the first day of bronchial instillation of BLM (~ 2.5 mg/kg) for a total of 27 days. After 28 days, BLM induced severe pulmonary fibrosis in mice, resulting in a significant increase in the lung weight coefficient. Compared to the Vehicle, NIF reduced the lung weight coefficient (Fig. 2b).

Furthermore, NIF attenuated the BLM-induced increase in hydroxyproline levels (Fig. 2c). Simultaneously, expression of the proinflammatory cytokines IL-6, IL-4, TNF-α and TGF-β1 in BALF samples was decreased in response to NIF treatment (Fig. 2d–g).

As shown by H&E and Masson staining in Fig. 3a, NIF alleviated BLM-induced structural damage and collagen proliferation in the lung and decreased the fibrosis score (Fig. 3b) and collagen volume fraction (CVF%) (Fig. 3c), indicating that NIF reduced collagen proliferation and improved pulmonary interstitial fibrosis. α-SMA and Collagen I are important proteins for characterizing fibrosis. As shown in Fig. 3d, e, and Additional file 1: Fig. S2, NIF inhibited the expression of α-SMA and Collagen I induced by BLM. Moreover, NIF inhibited EMT in vivo. As shown in Fig. 3e and Additional file 1: Fig. S2, NIF inhibited the decrease in E-Cadherin (epithelial marker) expression. This suggests that NIF inhibits BLM-induced pulmonary fibrosis and EMT in vivo. At the same time, we found that BLM induced abnormal activation of Stat3 (Fig. 3d, e; Additional file 1: Fig. S2), while this abnormal activation was decreased in response to NIF, indicating that NIF may inhibit pulmonary fibrosis by blocking abnormal Stat3 activation.

During the early stage of fibrosis, the innate immune system in the body is primarily antagonistic to stimulation, while immune cells in damaged tissues also contribute to chronic inflammation and tissue remodelling by secreting growth factors, cytokines, and chemokines and activating extracellular matrix synthesis [29, 30]. Therefore, strategies against immunosuppressive cells and cytokines may be very beneficial for tissue fibrosis during chronic inflammation. We found that BLM induced an increase in the expression of immune cells, inducing Gr-1⁺ CD11b⁺ (MDSCs), F4/80⁺ CD11b⁺ (macrophages), and CD4⁺, CD4 ⁺ CD69⁺ T lymphocytes. NIF reduced the number of these immune cells to closer to normal levels (Fig. 4a–d).

To further determine the therapeutic effect of NIF on pulmonary fibrosis, we also examined, 

the effects of delayed NIF treatment (25 mg/kg or 50 mg/kg), that is, NIF treatment beginning 14 days after injury induced by BLM (~ 1 mg/kg) for a total of 14 days (Fig. 5a). The results showed that NIF prolonged the survival rate of mice (Fig. 5b) and reduced the content of hydroxyproline (Fig. 5c). Of note, the levels of IL-2, IL-4 and other factors in the mouse serum were further reduced by NIF (Fig. 5d–h). Importantly, as shown in Fig. 6a, NIF stimulation reversed BLM-induced lung inflammation and collagen expression. Changes in the fibrosis score and CVF% were further confirmed (Fig. 6b, c). Additionally, the results of immunohistochemistry and protein analysis demonstrated that NIF reduced the expression of α-SMA and Collagen I and inhibited EMT (Fig. 6d, e; Additional file 1: Fig. S3). These results suggest that NIF promotes fibrosis resolution even when treatment is delayed.

Based on these findings, we further explored the potential mechanism of NIF and examined the effect of NIF on the activation of fibroblasts induced by TGF-β1. First, the MTT results showed that NIF inhibited the proliferation of pulmonary fibroblasts in a concentration-dependent manner (Additional file 1: Fig. S4). Next, we starved fibroblasts for 6 h, stimulated them with TGF-β1, and analysed the increase in the fibrosis phenotype at the protein level using immunofluorescence and western blot techniques. As shown in Fig. 7a–d, α-SMA and Collagen I were enhanced by stimulation with TGF-β1, consistent with previous reports[31]. Upon treatment with NIF, the increases in α-SMA and Collagen I induced by TGF-β1 were reduced (Fig. 7a–d), indicating that NIF significantly suppresses the activation of fibroblasts induced by TGF-β1. In addition, NIF decreased TGF-β1 stimulation-induced expression of p-Smad2/3 (Fig. 7c, d), further demonstrating that NIF inhibits TGF-β pathways.

To further examine the antifibrotic effects of NIF on activated lung fibroblasts, we isolated lung fibroblasts from rats treated with BLM[32]. The results showed that NIF inhibited the proliferation of activated lung fibroblasts in a concentration-dependent manner (Additional file 1: Fig. S4). Expression of α-SMA and Collagen I was also decreased in a concentration-dependent manner in response to NIF treatment (Fig. 7e, f). Furthermore, in primary activated pulmonary fibroblasts, p-Stat3 is elevated [33], and NIF decreased expression of p-Stat3 and p-Smad2/3 in a concentration-dependent manner (Fig. 7f). These results indicate that NIF may inhibit the TGF-β/Smad and Stat3 pathways and the activation of fibroblasts.

When local fibroblasts are inadequate for tissue repair and remodelling, pulmonary epithelial cells that undergo EMT are a source of myofibroblast cells that can promote the occurrence and development of pulmonary fibrosis, and TGF-β1 can induce EMT and migration of lung epithelial cells in vitro [34‐36]. Therefore, we used TGF-β1 to stimulate A549 cells, which caused them to undergo migration and EMT. As shown by MTT, NIF inhibited the proliferation of A549 cells in a time- and concentration-dependent manner (Fig. 8a). As shown in Fig. 8b, c, TGF-β1 induced the migration of A549 cells compared to the control group, but the induction of migration by TGF-β1 was inhibited after NIF administration. These results demonstrate that NIF inhibits the migration of epithelial cells.

With respect to EMT, Fig. 8d–f shows that TGF-β1 stimulation decreased the expression of E-Cadherin (epithelial marker) in A549 cells and increased Vimentin (interstitial marker) and α-SMA expression. NIF inhibited the TGF-β1-induced decrease in E-Cadherin and the increase in Vimentin and α-SMA. These results show that NIF reduces TGF-β1-induced EMT in A549 cells. We also assessed the activation of Smad2/3 in A549 cells and fibroblasts and found that NIF inhibited TGF-β1-induced expression of p-Smad2/3 (Fig. 8e, f). These results further suggest that the reduction in pulmonary fibrosis induced by NIF in vitro may be related to inhibition of abnormally activated Smad2/3.