Tauroursodeoxycholic acid alleviates pulmonary endoplasmic reticulum stress and epithelial-mesenchymal transition in bleomycin-induced lung fibrosis

BLM was purchased from HiSUN PFIZER Pharmaceuticals Co., Ltd; Zhejiang, China. TUDCA was prepared from EMD Millipore Corporation; Billerica, MA. Adult male C57BL/6 J (7 weeks-old, 21–23 g) mice were provided by Beijing Vital River Laboratory Animal Technology Co., Ltd (Beijing, China) and housed under a natural day/night cycle room with comfortable environment (temperature 20–25 °C, humidity 45–50%). Mice were supplied with enough food and water. Based on the previous studies of our laboratory and power calculation analysis, 80 mice were used in this study [11, 12]. Eighty mice were divided into 4 groups randomly. The experimental protocol was shown in Additional file 1: Figure S1. In TUDCA alone and BLM + TUDCA group, mice were intraperitoneally injected with TUDCA (250 mg/kg) once a day for 21 days. In BLM alone and BLM + TUDCA group, mice were intratracheally injected with a single dose of BLM (3.0 mg/kg, 1 mg bleomycin = 1000 IU bleomycin). Mice were intraperitoneally injected with saline and administered with saline by intratracheal injection in control group. The dose of BLM and the number of mice based on existing literature [11]. The dose of TUDCA based on existing literature [16]. Two mice were died in the BLM group. All mice were euthanized at 21 d after BLM injection. Left lungs were fixed for hematoxylin and eosin (H&E) staining and Masson trichrome staining. Right lungs were harvested and homogenized in liquid nitrogen for immunoblots, immunohistochemistry and RT-PCR [17]. This study was approved by the Association of Laboratory Animal Sciences and the Center for Laboratory Animal Sciences at Anhui Medical University (Permit Number: 13–0016). All procedures on animals followed the Guide for the Care and Use of Laboratory Animals published by the US National Institutes of Health (NIH Publication No. 85–23, revised 1996).

The left lungs of the mice were collected and fixed with 4% paraformaldehyde solution. Lung tissue sections were stained with hematoxylin eosin (H&E) and observed under a light microscope by a blinded and experienced investigator. Morphological changes were scored absent (0), mild (1), moderate (2) or severe injury (3) based on the presence of exudates, hyperemia/congestion, neutrophilic infiltrates, intraalveolar hemorrhage/debris, and cellular hyperplasia. HE staining and pathological scores were conducted based on the existing literature [11]. Mice in BLM-exposed group showed dim hair, loss of appetite and decreased activity. However, the health condition was better in control, TUDCA and BLM + TUDCA groups than these in BLM group. Collagen deposition in lung tissue was measured through Sirius Red staining. The percentage of collagen deposition area was determined using Image J Pro software. The quantification of images was performed by two investigators which were blinded to the experimental groups. Ten mice per group were used. Twelve images were randomly selected from similar lung lobe in every mouse at ×200 magnification. The hydroxyproline assay were performed in mice lungs according to the previous study [18].

The protein extraction and quantitative analysis were carried out based on the existing literature [19]. Briefly, mouse lung tissue was fully lysed in RIPA buffer and the appropriate protein concentration was determined. For immunoblots, equal amounts of protein from different treatment groups were added in sodium lauryl sulfate polyacrylamide gel and transferred to the membranes when the protein ladder was electrophoresed to the appropriate position. The membranes were incubated with different primary antibodies (α-SMA, phosphorylated-Smad2, phosphorylated-Smad3, Smad2/3, etc.) for different time. The membranes were washed in the PBST solution for three times, followed by incubating with different secondary antibodies. Antibodies against Ki-67, GRP78, E-cadherin, CHOP (C/EBP-homologous protein) and p-Smad2 were provided by Cell Signaling Technology (Beverley, MA). Antibodies against PCNA, α-SMA, p-Smad3 and Smad2/3 were provided by Abcam (Cambridge, UK). Antibody against β-actin was provided by Sigma Chemical Co., (St. Louis, MO). Antibodies against HO-1, 3-NT and secondary antibodies were provided by Santa Cruz Biotechnology Inc. (Santa Cruz, CA). Antibodies information was shown in Supplemental Table 1. Membrane signal was detected using ECL detection kit (Advansta Inc., California, USA). β-actin was used as a reference in the present study. Ten mice per group were used. Representative blots were selected and used in the figures.

Pulmonary α-SMA, p-Smad3, 3-NT and Ki67 were detected using immunohistochemistry (IHC). Briefly, lung tissues were fixed in paraformaldehyde, dehydrated and embedded. Endogenous peroxidases were blocked in 3% H₂O₂. Antigen retrieval was performed in the boiled citrate buffer for 4 min at an interval of 5 min, total 3 times. After inactivation of endogenous peroxidase and serum blocking, primary antibodies were incubated at 4℃ overnight. Then, goat anti-rabbit and goat anti-mouse IgG (1: 1000) were incubated 2 h at 37℃. Eventually, aspirate the primary antibody with a pipette, color reaction was detected with biotin-streptavidin peroxidase system [20]. Moreover, in order to suppress nonspecific staining, positive and negative controls were performed in this study. Negative control was defined as which primary antibody or second antibody were replaced with PBS. Positive control was defined as which lung sections of confirmed pulmonary fibrosis in the previous study were performed through IHC staining [11, 12]. Ten mice per group were used. Representative images were selected and used in the figures.

Total RNA extraction was referred to existing literature [21] and cDNAs were produced by reverse transcription reagent kit (Promega, Madison, USA). Real-time RT-PCR was conducted in LightCycler480 Instrument (Roche Diagnostics, Germany). Primers for real-time RT-PCR were listed as following: Tgf-β (F, TTCCGCTGCTACTGCAAGTCA; R, GGGTAGCGATCGAGTGTCCA), 18S (F, GTAACCCGTTGAACCCCATT; R, CCATCCAATCGGTAGTAGCG). Gene expression was normalized to 18S.

There was no death in Control and TUDCA groups. Bleomycin exposure induced death on the 4–7th and 14–17th days after the administration. Eight mice were died in the process of BLM-induced pulmonary fibrosis. There was no difference of healthy condition between Control and TUDCA groups. BLM exposure induced a dim hair, decreased appetite, activity and body weight. The healthy condition was better in Control and TUDCA groups than these in BLM and BLM + TUDCA groups. The average weight in four groups was as follows: Control: 25.0 g, TUDCA: 24.3 g, BLM: 19.5 g, BLM + TUDCA: 21.1 g. To investigate whether TUDCA alleviate BLM-induced pathological damage and collagen deposition of pulmonary fibrosis. H&E staining and Sirius Red staining was performed on lung tissue sections. According to H&E staining, the main histological damages in BLM-treated mice was alveolar septal destruction. Interestingly, pretreatment with TUDCA attenuated BLM-induced alveolar septal destruction in mice lungs (Fig. 1a, b). Sirius Red staining showed a large amount of collagen deposition in pulmonary interstitium of BLM-treated mice. Pretreatment with TUDCA markedly attenuated BLM-induced collagen deposition in pulmonary interstitium (Fig. 1c, d). Moreover, the levels of hydroxyproline were detected in mice lungs. The results found that BLM-exposure elevated the levels of hydroxyproline in mice lungs (Fig. 1f). Interestingly, pretreatment with TUDCA alleviated BLM-induced up-regulation of hydroxyproline (Fig. 1f).

EMT plays a crucial role in pulmonary fibrosis [12]. α-SMA is a hallmark of myofibroblasts and is generally accepted as a marker for EMT. E-cadherin, an epithelial marker, is an important cell adhesion molecule. To explore whether TUDCA alleviate pulmonary EMT in BLM-treated mouse lungs, α-SMA and E-cadherin were measured. The results indicated that the percentage of pulmonary α-SMA-positive cells was markedly elevated in the BLM-treated mice. After pretreatment with TUDCA, the percentage of pulmonary α-SMA-positive cells was reduced (Fig. 2a, b). Further analysis displayed that the expression of pulmonary α-SMA was elevated by BLM. Correspondingly, pretreatment with TUDCA downregulated α-SMA in the lungs of BLM-treated mice (Fig. 2c, d). By contrast, the expression of pulmonary E-cadherin was obviously decreased in BLM-treated mice lungs. After pretreatment with TUDCA, BLM-induced downregulation of pulmonary E-cadherin was inhibited (Fig. 2c, e).

Smad2/3-mediated EMT of alveolar epithelial cells is involved in the pathogenesis of BLM-induced lung fibrosis [22, 23]. To investigate whether TUDCA alleviate BLM-induced pulmonary TGF-β/Smad2/3 activation in mice, TGF-β/Smad2/3 signaling was detected. Pulmonary Tgf-β1 mRNA level was upregulated after BLM treatment in mice lungs. After pretreatment with TUDCA, BLM-induced upregulation of pulmonary Tgf-β1 mRNA was inhibited (Fig. 3a). As shown in IHC, pulmonary p-Smad3-positive cells were elevated by BLM treatment. After pretreatment with TUDCA, BLM-induced elevation of p-Smad3-positive cells was significantly attenuated (Fig. 3b, c). In addition, the levels of pulmonary p-Smad2 and p-Smad3 were elevated after BLM treatment. After pretreatment with TUDCA, BLM-induced phosphorylation of pulmonary Smad2 and Smad3 was inhibited (Fig. 3d–f).

In the present study, two markers of cellular proliferation were detected. Ki67 was detected by IHC. PCNA was detected by immunoblot. As shown in Fig. 4a, b, pulmonary Ki67-positive cells, were increased in BLM-treated mice. Of interest, BLM-induced cellular proliferation was alleviated in TUDCA-pretreated mice (Fig. 4a, b). The effect of TUDCA on pulmonary PCNA protein was explored. As shown in Fig. 4c, d. As expected, pulmonary PCNA level was increased in BLM-treated mice. TUDCA pretreatment obviously attenuated BLM-induced elevation of PCNA in mice lungs (Fig. 4c, d).

Increasing data have demonstrated that excess ROS (reactive oxygen species) takes part in the process of BLM-evoked lung fibrosis [24, 25]. HO-1 and 3-NT, two markers of oxidative stress, were detected in mice lungs. In the present study, we found that pulmonary HO-1 protein was increased after BLM treatment. After pretreatment with TUDCA, BLM-induced increase of HO-1 was inhibited in mice lungs (Fig. 5a, b). 3-NT was elevated in BLM-treated mice. After pretreatment with TUDCA, pulmonary 3-NT was alleviated (Fig. 5a, c). IHC showed that 3-NT-positive cells were increased in the lungs of mice after BLM treatment. After pretreatment with TUDCA, the percentage of pulmonary 3-NT-positive cells was alleviated (Fig. 5d, e).

GRP78, an ER chaperone and a marker of ER stress, was elevated in lungs of BLM-treated mice. After pretreatment with TUDCA, the expression of GRP78 was decreased (Fig. 6a, b). In addition, CHOP, a marker of UPR, was increased in BLM-treated mouse lungs. After pretreatment with TUDCA, BLM-evoked elevation of CHOP was suppressed (Fig. 6a, c).