Main

Lung fibrosis is an aggressive disease with a median survival of 3–5 years after the onset of symptoms. It involves replacement of normal lung parenchyma with fibrotic tissue accompanied by inflammation, fibroblast proliferation, and excessive collagen deposition. The scar tissue once formed is permanent, leading to irreversible decrease in oxygen-diffusion capacity of the lung. Lung fibrosis may be a secondary effect of other diseases or conditions, or appear without any known cause, in which it is called ‘idiopathic’. The cellular and molecular mechanisms underlying pathogenesis of lung fibrosis are largely unclear. To date, there is no evidence that any medications can prevent or reverse lung fibrogenesis, and new therapies are urgently needed for effective treatment.1 Pharmaceutical interventions of factors involved in fibroblast proliferation, inflammation, oxidative damage, and extracellular matrix formation have been intensively explored as potential new therapies. The transforming growth factor-β1 (TGF-β1)/Smad3 signaling pathway has a central role among various factors that regulate lung fibrosis. TGF-β1 induces differentiation of lung fibroblasts into myofibroblasts characterized by expression of α-smooth muscle actin (α-SMA) and synthesis of extracellular matrix proteins.2, 3, 4 Downstream signaling effects of TGF-β1 are mediated by Smad3,5 and Smad3 deficiency attenuates bleomycin (BLM)-induced lung fibrosis.6 Inhibition of TGF-β1/Smad3 signaling with a variety of biologics including neutralizing antibodies, antisense oligonucleotides, and short-interfering RNAs (siRNAs) has been reported to ameliorate fibrosis, validating the TGF-β1/Smad3 pathway as a therapeutic target for the treatment of lung fibrosis.7, 8, 9 However, biologics drugs are difficult to develop owing to factors such as their poor stability and bioavailability. For example, canonical double-stranded siRNAs are unstable and cause undesirable adverse effects such as deregulation of the immune system. Their therapeutic value is also limited by potential oncogenic effect of the vectors. Thus, RNAi-based therapy is already clinically applied; however, recommending this therapy to every patient remains difficult.10

All-trans retinoic acid (ATRA), a derivative of vitamin A, exhibits a number of beneficial effects such as anti-proliferative, anti-inflammatory, anti-migratory, and anti-fibrogenic activities. ATRA is currently used to induce remission in patients with acute promyelocytic leukemia and has the potential to be used to treat solid tumors.11, 12 Several studies have demonstrated the anti-fibrogenic effect of ATRA in BLM-induced lung fibrosis in mice.13, 14 However, Segel et al15 reported that ATRA does not elicit beneficial effects on BLM-induced lung fibrosis in rats. Although there is controversy in the effect of ATRA on lung fibrogenesis in animal models, the effect of ATRA on the TGF-β1/Smad3 pathway in lung fibrosis has not been reported. A recent paper showed that ATRA suppresses kidney fibrosis without suppressing the TGF-β1/Smad3 signaling in rat.16 Zhou et al17 proposed that these divergent and contradictory effects may be because of the different signaling transduction pathways involved in various tissues and/or cells. The different ATRA doses used in various studies may be another reason for the contradictory results. In the present study, we aimed to determine whether ATRA (0.4 and 1.6 mg/kg) can ameliorate BLM-induced lung fibrosis in rats and whether the mechanism involves the downregulation of the TGF-β1/Smad3 signaling pathway.

MATERIALS AND METHODS

Ethics Statement

Sprague–Dawley (SD) rats with a mean weight of 200 g were provided by the Green Leaf Experimental Animal Center (Yantai, China). All animal experiments were performed in accordance with the regulations established by the Committee on the Ethics of Animal Experiments of Binzhou Medical University. The rats were housed under a 12-h light/dark cycle and were allowed free access to food and water.

Animal Model

Pharmaceutical grade BLM was purchased from Nippon Kayaku (Tokyo, Japan). ATRA was purchased from Sigma (St Louis, MO, USA). The lung fibrosis model was established as previously described.18 A total of 40 SD rats were randomly divided into four groups (10 rats each) including the sham group, BLM-induced group (BLM group), ATRA treatment I group (ATRA I group, 0.4 mg/kg), and ATRA treatment II group (ATRA II group, 1.6 mg/kg).19 Lung fibrosis was induced by a single intratracheal instillation of 5 mg/kg BLM in 0.3 ml of saline in all the groups except the sham group. The sham group received an equal volume of saline. Beginning on day 14 after BLM treatment, rats in ATRA I and II groups received oral ATRA once daily for 14 days. On day 28, all rats were killed, and lung tissue sections were collected and immediately frozen in liquid nitrogen for further studies.

Hematoxylin and Eosin (H&E) Staining

H&E staining agents were purchased from Sigma. Lung tissues were fixed by instilling 4% paraformaldehyde through the trachea and were embedded in paraffin. Transverse sections of 4 μm thickness were stained with H&E following the manufacturer’s standard protocol. Histologic grading of lesions was performed using the Szapiel method for extent of inflammation in lung parenchyma base.20 Degrees of microscopic interstitial inflammation and fibrosis were graded on a scale of 1–4. (1), absent and appears normal (−); (2), light (+); (3), moderate (++); (4), strong (++++). In addition, total lung inflammation and fibrosis score were calculated as the sum of the two components. Three sections collected from each lung were analyzed in the experiment.

Masson’s Trichrome Staining

Collagen deposition was measured by the Masson’s trichrome method. Lung tissues were fixed by inflation with 4% paraformaldehyde overnight, dehydrated in 70% ethanol, and embedded in paraffin wax. Sections of 4 μm thickness were prepared and stained with Masson’s trichrome. Percentage of fibrosis in the lung was determined by counting the number of pixels within areas of stained collagen in digital images, using the Adobe Photoshop CS3 program as previously described.21 We can calculate the hydroxyproline (Hyp) content according to the depth of coloration using the following formula: Hyp content (μg/mg wet weight)=(measurement tube absorbance−absorbance of blank)/standard pipe of absorbance × content of standard pipe × total volume of hydrolyzate/tissue wet weight. Degrees of microscopic interstitial inflammation and fibrosis were graded on a scale of 1–4. (1), absent and appears normal (−); (2), light (+, affected area <20%); (3), moderate (++, affected area 20–40%); (4), strong (++++, affected area >50%). Three sections collected from each lung were analyzed in the experiment.

Immunohistochemistry

Anti-collagen I antibody was purchased from Santa Cruz Biotechnology (Dallas, TX, USA). Paraffin sections prepared as described above were dewaxed and immersed in 3% hydrogen peroxide solution in methanol for 10 min to block endogenous peroxidases. The slides were subsequently immersed in 10 mM citrate buffer (pH 6.0) and placed in a microwave oven for 25 min. After washing with 10 mM phosphate-buffered saline (PBS; pH 7.4), the sections were covered with serum and placed in a humidity chamber for 30 min at room temperature. Excess serum was rinsed off with 10 mM PBS, and the sections were incubated with anti-collagen I antibody (1:500) in a humidity chamber for 45 min at room temperature. The sections were then rinsed with PBS and incubated with biotinylated secondary antibody in a humidity chamber for 40 min at 37 °C. After rinsing with PBS, the slides were incubated with streptavidin–peroxidase complex reagent for 45 min at room temperature and washed again with PBS. The slides were covered with 3,3-diaminobenzidine tetrahydrochloride solution for 15 min and then examined under a microscope. The sections were then immersed in running water, counterstained with haematoxylin for 1 min, and immersed in tap water bath followed by a series of alcohol baths of increasing concentrations and xylene. The processed sections were covered with cover slips. Samples treated with primary antibody alone were assigned as negative controls.

Transmission Electron Microscopy (TEM) Observation

Lung tissues were fixed by treatment with fresh 3% glutaraldehyde at 4 °C for at least 4 h, postfixed in 1% osmium tetroxide for 1.5 h, dehydrated in gradient ethanol, infiltrated with Epon812, embedded, and cultured at 37, 45, and 60 °C for 24 h. Ultrathin sections prepared with an ultracut E ultramicrotome were stained with uranyl acetate and lead citrate and observed using a JEM-1400 TEM system from Jeol (Tokyo, Japan).22

Hyp Content

Lung specimens were washed with saline and hydrolyzed with 0.6% hydrochloric acid at 100 °C for 5 h. The hydrolysates were neutralized with sodium hydroxide and diluted with distilled water. Hyp level in the hydrolysates was colorimetrically determined by absorbance at 560 nm with p-dimethylaminobenzaldehyde and expressed as mg/g wet tissue.23

Immunofluorescence

Rabbit anti-rat α-SMA and anti-E-cadherin antibodies were purchased from Boster Bio-engineering (Wuhan city, China). Immunofluorescence staining was performed as previously described.18 Lung tissue sections were fixed in 4% paraformaldehyde, rinsed with PBS solution, incubated with 0.5% TritonX-100 for 20 min, and blocked with 10% calf serum. Tissue sections were incubated with rabbit anti-rat α-SMA or E-cadherin antibody (1:100 or 1:50) at 4 °C overnight. Tissue sections were subsequently rinsed with PBS and incubated with goat anti-rabbit IgG labeled with FITC or Cy3. Hoechst 33258 fluorochrome (Sigma) was used for nuclear staining. After washing with PBS, tissue sections were mounted in neutral glycerin and analyzed under a laser scanning confocal microscope from Leica Company (Germany). The fluorescence imaging was conducted with excitation at 488 nm and emission at 510 nm at room temperature.

Quantitative Real-time Polymerase Chain Reaction (qRT-PCR)

Total RNA was extracted from lung tissues using TRIzol reagent from Invitrogen (Carlsbad, CA, USA) according to the manufacturer’s instructions. Complementary DNA synthesis was performed using the M-MLV reverse transcriptase kit from Promega (Madison, WI, USA) following the manufacturer’s instructions. qRT-PCR was performed using a SYBR green-based PCR master mix kit from Takara Bio (Shiga, Japan) on a Rotor Gene3000 real-time PCR system from Corbett Research (Sydney, Australia). The PCR conditions were as follows: initial denaturation at 95° C for 5 min followed by 40 cycles of 60 °C for 25 s, annealing at 52 °C for 20 s, and extension at 72 °C for 20 s. Fluorescence signal was monitored at 585 nm during each extension. Glyceraldehyde 3-phosphate dehydrogenase (GAPDH) served as an internal control. Primers used in qRT-PCR are shown in Table 1.

Table 1 Primers of TGF- β 1, Smad3, ZEB1, ZEB2, HMGA2, and GAPDH
Full size table

Western Blot Analysis

Polyclonal rabbit antibodies for rat TGF-β1, Smad3, p-Smad3, ZEB1, ZEB2, Snail, Twist, and HMGA2 were purchased from Santa Cruz Biotechnology. HRP-labeled goat anti-rabbit IgG and α-tubulin antibody were from Beijing Zhong Shan-Golden Bridge Technology (Beijing, China). Total protein lysate samples containing 20 μg protein were subjected to 10% sodium dodecyl sulfate polyacrylamide gel electrophoresis, transferred onto polyvinylidene difluoride membranes, and blocked with 7% non-fat milk in Tris-buffered saline and Tween-20 (TBST; 50 mM Tris-HCl (pH 7.6), 150 mM NaCl, 0.1% Tween-20) for 1.5 h at room temperature. Membranes were washed three times with TBST buffer and incubated at 4 °C overnight with polyclonal rabbit antibodies specific for rat TGF-β1, Smad3, p-Smad3, ZEB1, ZEB2, Snail, Twist, and HMGA2. After washing with TBST, membranes were incubated with HRP-labeled goat anti-rabbit IgG (1:5000) for 1.5 h at room temperature. Membranes were then washed with TBST, incubated with ECL reagent, and exposed. Membranes were subsequently stripped and re-probed with α-tubulin antibody (1:500), which served as loading control.

Statistical Analysis

Data are expressed as the mean±s.e.m. of the indicated number of independent experiments. Student’s t-test and one-way analysis of variance were used to determine significance, with P<0.05 considered significant. The Kaplan–Meier method was used for survival analysis with a log-rank of P<0.05 to determine significance. Statistical analyses were performed using SPSS 10.0 for Macintosh. Graphs were also constructed.

RESULTS

ATRA Prevented BLM-induced Lung Fibrosis

We first examined the preventive effect of ATRA on lung fibrosis by means of H&E staining, Masson’s trichrome staining, immunohistochemistry, TEM, and Hyp content.

H&E staining results showed that tissue sections from the sham group had a continuous structure with an intact wall of bronchial mucous membrane and lung alveoli with no obvious abnormality. The alveoli in the sham group had clear hollow cavities with alveolar walls thinner than those in all other groups. The BLM group had the thickest alveolar walls among all the groups. Tissue sections from the BLM group also showed severe edema and infiltration of macrophages, neutrophils, lymphocytes, fibroblasts, and erythrocytes. In addition, the lung mesenchyme in the BLM group showed strong immunohistochemical staining for collagens and collagen fibers, indicating that the hallmark of the fibroblastic foci was distinctly present. These histological, cellular, and molecular characteristics of lung fibrosis were significantly reduced in ATRA-treated groups. Compared with the BLM group, tissue sections from ATRA-treated groups had thinner alveolar walls and showed lower degree of edema and cellular infiltration of macrophages, neutrophils, lymphocytes, fibroblasts, and erythrocytes. Lung fibrosis determined by the Szapiel method using H&E staining showed statistically significant differences between ATRA-treated groups and the BLM group (Figure 1). Masson’s trichrome staining showed that the number of fibroblasts and the collagen matrix increased significantly in the BLM group, which was accompanied by formation of fibrosis lesions with cord-like distribution. ATRA treatment significantly reduced BLM-induced alveolitis and lung fibrosis (Figure 2). Immunohistochemical analysis indicated that the BLM group had higher collagen I content compared with the sham group, and ATRA treatment significantly attenuated BLM-induced collagen I expression (Figure 3). TEM results indicated that the sham group had the lowest number of collagen fibers among all the groups. The BLM group had a large number of collagen fibers in the interstitial lung and ATRA treatment significantly reduced BLM-induced formation of collagen fibers (Figure 4a). Collagen deposition in the lung was assessed by Hyp content. Compared with the sham group, the BLM group showed significantly higher Hyp content in the lung. ATRA treatment significantly reduced BLM-induced increase in Hyp content (Figure 4b). In general, ATRA improved histological, cellular, and molecular markers of lung fibrosis in a dose-dependent manner.

Figure 1
figure 1

Grade of lung fibrosis and inflammation. (a) Sham group, (b) BLM group, (c) ATRA I group, (d) ATRA II group, and (e) score of lung fibrosis. Each bar represents the mean±s.d., n=6. Original magnification, × 400. *P<0.05.

Full size image
Figure 2
figure 2

Collagen content (blue) observed with Masson’s trichrome stain. (a) Sham group, (b) BLM group, (c) ATRA I group, (d) ATRA II group, and (e) score of collagen content. Each bar represents the mean±s.d., n=6. Original magnification, × 400. *P<0.05; **P<0.01.

Full size image
Figure 3
figure 3

ATRA suppressed collagen I (brown) expression. (a) Sham group, (b) BLM group, (c) ATRA I group, (d) ATRA II group, (e) densitometric analyses, (a’, b’, c’, d’) negative control. Each bar represents the mean±s.d., n=6. Original magnification, × 400. **P<0.01.

Full size image
Figure 4
figure 4

ATRA suppressed formation of collagen fibers and collagen deposition. (a) Collagen fibers (white arrow) observed under TEM. (A) The sham group exhibited the lowest number of collagen fibers in the interstitial lung tissue. (B) The BLM group showed the highest number of collagen fibers in the interstitial lung tissue. (C) ATRA I and (D) ATRA II groups exhibited significantly reduced collagen fibers. (b) Hyp content. Each bar represents the mean±s.d., n=6. *P<0.05, **P<0.01.

Full size image

ATRA Reduced Expression of EMT Markers

EMT is characterized by expression of the mesenchymal marker α-SMA and loss of the epithelial cell marker E-cadherin. Several transcription repressors such as Snail and Twist are involved in EMT. Immunofluorescence imaging and western blot analysis were used to analyze the expression of α-SMA, E-cadherin, Snail, and Twist. The immunofluorescence data showed that α-SMA level was significantly higher (Figure 5), whereas E-cadherin level was significantly lower (Figure 6) in the BLM group compared with the sham group. ATRA treatment markedly blocked BLM-induced changes in α-SMA and E-cadherin expression. Western blot analysis showed similar effects of BLM and ARTA on α-SMA and E-cadherin expression as those observed in immunofluorescence. In addition, ARTA downregulated BLM-induced Snail and Twist expression with kinetics similar to that observed with α-SMA (Figure 7).

Figure 5
figure 5

ATRA treatment suppressed α-SMA expression. (a) α-SMA expression (green) observed under a laser scanning confocal microscope. Nuclei (blue) were counterstained with hoechst 33342. (b) western blot analysis of α-SMA protein expression with densitometric quantification. Each bar represents the mean±s.d., n=6. *P<0.05, **P<0.01.

Full size image
Figure 6
figure 6

ATRA treatment increased E-cadherin expression. (a) E-cadherin expression (red) observed under a laser scanning confocal microscope. Nuclei (blue) were counterstained with hoechst 33342. (b) Western blot analysis of E-cadherin protein expression with densitometric quantification. Each bar represents the mean±s.d., n=6. *P<0.05, **P<0.01.

Full size image
Figure 7
figure 7

Western blot analysis of Snail and Twist protein expression with densitometric quantification. Each bar represents the mean±s.d., n=6. *P<0.05.

Full size image

ATRA Downregulated the TGF- β 1/Smad3 Signaling Pathway

The effects of ATRA on the activation of the TGF-β1/Smad3 signaling pathway, including TGF-β1, Smad3, p-Smad3, and ZEB1, ZEB2, and HMGA2 regulated by the TGF-β1/Smad3 signaling pathway were assessed using qRT-PCR and western blot analyses. The qRT-PCR analysis showed that ATRA significantly reduced mRNA expression of these proteins in a concentration-dependent manner compared with the BLM group (Figures 8 and 9). Analysis of protein expression using western blot indicated similar effects of BLM and ATRA on the TGF-β1/Smad3 signaling pathway at the protein level (Figures 10 and 11).

Figure 8
figure 8

qRT-PCR analysis of TGF-β1 (a) and Smad3 (b) mRNA expression. Each bar represents the mean±s.d., n=6. *P<0.05, **P<0.01.

Full size image
Figure 9
figure 9

Western blot analysis of TGF-β1 and Smad3 protein expression with densitometric quantification. Each bar represents the mean±s.d., n=6. *P<0.05, **P<0.01.

Full size image
Figure 10
figure 10

qRT-PCR analysis of ZEB1 (a), ZEB2 (b), and HMGA2 (c) mRNA expression. Each bar represents the mean±s.d., n=6. *P<0.05, **P<0.01.

Full size image
Figure 11
figure 11

Western blot analysis of ZEB1, ZEB2, and HMGA2 protein expression with densitometric quantification. Each bar represents the mean±s.d., n=6. *P<0.05, **P<0.01.

Full size image

DISCUSSION

In the present study, we investigated the effect of ATRA on BLM-induced lung fibrogenesis in rats and the role of TGF-β1/Smad3 signaling pathway in ATRA treatment. Our results showed that BLM significantly increased alveolar wall thickness, inflammatory cell infiltration, and collagen fiber formation. These BLM-induced changes were significantly ameliorated by ATRA treatment, and these anti-fibrotic effects of ATRA may be mediated by EMT inhibition through the downregulation of the TGF-β1/Smad3 signaling pathway.

Our data showed that ATRA ameliorated BLM-induced EMT by inhibiting α-SMA and collagen expression and by enhancing E-cadherin expression in fibroblasts. These data are in agreement with the notion that fibroblasts are the primary effectors of tissue fibrosis because fibroblasts produce collagen and other extracellular matrix proteins.24 In lung fibrogenesis, fibroblasts often differentiate into myofibroblasts, which exhibit enhanced fibrotic, contractile, and migratory activities.25 α-SMA is a marker of myofibroblasts and a primary contributor to the contractile force in myofibroblasts.26 It is believed that lung tissue fibrosis is regulated by the TGF-β1/Smad3 signaling pathway, which is activated by myofibroblast contraction.27, 28, 29, 30, 31 The TGF-β1/Smad3 signaling pathway promotes fibrotic transformation of lung tissues via activation of EMT.32, 33, 34

EMT is a process in which epithelial cells lose their epithelial phenotype, acquire fibroblast-like properties, and display reduced cell adhesion and increased motility. Reduction in expression of adhesion molecules allows the cells to detach from the epithelial layer and migrate towards the site of injury or inflammation where they exercise their profibrotic effects. EMT has attracted considerable attention from researchers in the past few years as a potential contributor to fibrotic diseases.35, 36, 37 Previous studies have demonstrated that epithelial cells undergoing EMT have decreased levels of epithelial cell markers such as E-cadherin and increased levels of myofibroblast markers such as α-SMA and mesenchymal cell marker collagen.38, 39Our data that ATRA inhibited α-SMA and collagen expression and increased the E-cadherin level support that ATRA might ameliorate lung fibrosis via suppressing EMT. To further study the effect of ATRA on EMT, we examined the expression of Snail and Twist, two most thoroughly studied transcription factors involved in EMT.40, 41 It is believed that Snail directly inhibits E-cadherin, whereas Twist suppresses the expression of E-cadherin indirectly.42 Gene overexpression studies have shown that Snail is sufficient to induce EMT, and snail deficiency attenuates TGF-β1-induced EMT in alveolar epithelial type II (ATII) cells.43 Snail is also highly expressed in hyperplasic ATII cells in lung tissues from patients with IPF, supporting its role in the pathogenesis of IPF. Twist is a well-established master transcription regulator of EMT during embryogenesis and metastasis.44 The abundant expression of Twist in alveolar epithelial cells is likely to contribute to EMT, which is an important source of fibroblasts in lung fibrosis. It has been reported that Snail and Twist are upregulated by TGF-β1 signaling. Consistent with its activation of E-cadherin expression, ATRA inhibited BLM-induced Snail and Twist expression, further supporting that ATRA ameliorated lung fibrosis via inhibiting EMT. Recent studies40, 43, 45, 46, 47 support our findings. For example, Snail and Twist are upregulated by the majority of EMT inducers and they play a crucial role in this transforming event. Snail overexpression induces EMT and renal fibrosis in vivo in a mouse model. Snail expression is increased in human lung fibrosis and is essential to the control of TGF-β1-induced EMT in alveolar epithelial cells in vitro. Twist is reportedly involved in the cadherin switch as an inducer of N-cadherin expression.

HMGA2, ZEB1, and ZEB2 are transcription factors that promote EMT and are induced by the TGF-β1/Smad3 signaling pathway during the EMT process. HMGA2 upregulates the transcription of Snail and Twist, two transcriptional repressors of adherens that are critical for intercellular adhesions.48, 49 ZEB1 and ZEB2 are E-cadherin transcriptional repressors implicated in the initial stage of EMT.50, 51, 52 ZEB1 and ZEB2 are upregulated during the pathogenesis of EMT. Xiong et al53 demonstrated that the miR-200 family is responsible for protecting tubular EMT through the targeted suppression of ZEB1 and ZEB2. Ectopic expression of HMGA2, ZEB1, and ZEB2 cause irreversible EMT characterized by severe E-cadherin suppression and α-SMA enhancement.54, 55, 56, 57, 58 Our results showed that BLM activated the TGF-β1/Smad3 signaling pathway including the expression of TGF-β1, Smad3, p-Smad3 and downstream effectors, and EMT promoters HMGA2, ZEB1, and ZEB2. ATRA blocked BLM-induced activation of the TGF-β1/Smad3 signaling pathway as shown in its inhibition of the expression of all these proteins.

In molecular terms, EMT tends to be induced by the Snail, Twist, and ZEB families of transcription factors, which may consequently be activated by the TGF-β1/Smad3 signaling pathway.59 Taken together, our study provided the first line of evidence supporting the notion that ATRA can attenuate BLM-induced lung fibrosis by suppressing EMT. The inhibitory effect of ATRA on EMT is mediated by the downregulation of the TGF-β1/Smad3 signaling pathway. Further studies are needed to fully understand the mechanisms of ATRA action and exploit ATRA as a potential therapy for lung fibrosis.