Microsomal Prostaglandin E Synthase-1 Deficiency Exacerbates Pulmonary Fibrosis Induced by Bleomycin in Mice

Microsomal prostaglandin E2 synthase-1 (mPGES-1), an inducible enzyme that converts prostaglandin H2 (PGH2) to prostaglandin E2 (PGE2), plays an important role in a variety of diseases. So far, the role of mPGES-1 in idiopathic pulmonary fibrosis (IPF) remained unknown. The current study aimed to investigate the role of mPGES-1 in pulmonary fibrosis induced by bleomycin in mice. We found that mPGES-1 deficient (mPGES-1−/−) mice exhibited more severe fibrotic lesions with a decrease in PGE2 content in lungs after bleomycin treatment when compared with wild type (mPGES-1+/+) mice. The mPGES-1 expression levels and PGE2 content were also decreased in bleomycin-treated mPGES-1+/+ mice compared to saline-treated mPGES-1+/+ mice. Moreover, in both mPGES-1−/− and mPGES-1+/+ mice, bleomycin treatment reduced the expression levels of E prostanoid receptor 2 (EP2) and EP4 receptor in lungs, whereas had little effect on EP1 and EP3. In cultured human lung fibroblast cells (MRC-5), siRNA-mediated knockdown of mPGES-1 augmented transforming growth factor-β1 (TGF-β1)-induced α-smooth muscle actin (α-SMA) protein expression, and the increase was reversed by treatment of PGE2, selective EP2 agonist and focal adhesion kinase (FAK) inhibitor. In conclusion, these findings revealed mPGES-1 exerts an essential effect against pulmonary fibrogenesis via EP2-mediated signaling transduction, and activation of mPGES-1-PGE2-EP2-FAK signaling pathway may represent a new therapeutic strategy for treatment of IPF patients.


Introduction
Idiopathic pulmonary fibrosis (IPF) is a chronic, progressive disease primarily affecting elderly adults and characterized by the proliferation of lung fibroblasts and excessive collagen deposition [1]. Although the mechanism of IPF remains yet largely unknown, several factors such as inflammation, coagulation, oxidative stress, and epithelial mesenchymal transition (EMT) have been proposed to play important roles in the development and progression of IPF [2]. Typical histopathological changes of pulmonary fibrosis include alveolar epithelial cell injury, and differentiation of fibroblasts to myofibroblasts. An increase in α-smooth muscle actin (α-SMA) expression is the central event in the differentiation of fibroblasts to myofibroblasts [3]. Hydroxyproline is one of the main components in the collagen protein, with the content levels correlating with collagen deposition and the severity of pulmonary fibrosis [4]. While no large-scale etiological studies relating to factors such as geography, ethnicity or racial factors have been performed, it is clear that IPF cases are on the rise [5]. Thus, further research on its pathogenesis and the development of novel therapeutic strategies are required to meet this medical need.
Prostaglandins (PGs) are important internal inflammation mediators, and their abnormal production has been shown to tightly be associated with a variety of diseases. Prostaglandin E2 (PGE 2 ), a member of PG family, is synthesized when cyclooxygenase (COX) induces PGG 2 to PGH 2 [6], followed by the prostaglandin E (PGE) synthase mediated conversion of PGH 2 to PGE 2 [7]. Three types of PGE synthases controlling PGE 2 production in cells have been identified before. Two are membrane-associated, which are designated as mPGES-1 and mPGES-2, respectively. The third is cytosolic, which is designated as cPGES [7,8]. mPGES-2 is constitutively expressed in many tissues [9], whereas mPGES-1 expression is induced in response to inflammation [10,11]. So far, mPGES-1 has been demonstrated to be directly associated with many diseases such as pain, fever, tumorigenesis, atherosclerosis, reproduction and skin fibrogenesis [12][13][14][15][16][17], yet there is no current research pertaining to the function of mPGES-1 in IPF. In addition, IPF is a typical disease related with inflammation, and previous research has proved PGE 2 production was decreased in lung fibroblasts isolated from IPF patients [18], so we hypothesize that mPGES-1, one of PGE synthases, may play an essential role in IPF development and progression.
Bleomycin, an anti-tumor antibiotic isolated initially from the fungus Streptomyces verticillus [19], is extensively used in induction of pulmonary fibrosis in animal models [20]. Moreover, fibroblasts stimulated by transforming growth factor-β1 (TGF-β1) differentiate into myofibroblasts, from which extensive extracellular matrix is accumulated to form lung fibrosis [21]. In this study, we used these methods to investigate the function of mPGES-1 in pulmonary fibrosis in order to further clarify the underlying mechanisms and to search for a new target for the treatment of IPF.

mPGES-1 −/− Mice Exhibited More Severe Lung Fibrosis after Bleomycin Treatment
Histopathological evaluation of paraffin-embedded lung sections was examined to establish lung fibrosis. While no morphological changes were observed in mPGES-1 +/+ and mPGES-1 −/− treated with saline, significant fibrotic changes were noted in bleomycin-treated lung samples. The mPGES-1 +/+ mice with bleomycin displayed moderate fibrotic lesions, inflammatory cell infiltration, thickening of the interstitium, and contained moderate collagen deposition. Furthermore, the mPGES-1 −/− mice with bleomycin exhibited more severe fibrosis characterized by increased inflammatory cell infiltration, a complete loss of alveolar architecture and massive collagen deposition resulting in enhanced fibrosis ( Figure 1A).
Hydroxyproline content was quantified to reflect collagen deposition in the lungs as a means to assess the extent of lung fibrosis for each experimental group. Hydroxyproline content was assessed as μg per 30 mg tissue sample and the values in four groups of mice were as following: mPGES-1 +/+ mice with saline (37.14 ± 2.08), mPGES-1 +/+ mice with bleomycin (76.93 ± 4.81), mPGES-1 −/− mice with saline (41.81 ± 2.30) and mPGES-1 −/− mice with bleomycin (105.4 ± 11.08). A significant increase in hydroxyproline content was noted in bleomycin treated samples when compared with groups treated with saline (p < 0.001). Importantly, the hydroxyproline content of the mPGES-1 −/− mice receiving bleomycin was significantly increased when compared with the mPGES-1 +/+ mice receiving bleomycin (p < 0.05) ( Figure 1C).

Aggravation of Bleomycin-Induced Lung Fibrosis in mPGES-1 −/− Mice Was not Dependent on Smad2/3 Pathway
Western blotting analysis showed that the protein expression levels of TGF-β1 and p-Smad2/3 were significantly increased in the lungs of mPGES-1 −/− and mPGES-1 +/+ mice treated with bleomycin when compared with saline (p < 0.05). Furthermore, TGF-β1 expression levels were significantly higher in mPGES-1 −/− mice than in mPGES-1 +/+ mice with bleomycin. However, the protein expression levels of p-Smad2/3 did not differ between mPGES-1 −/− and mPGES-1 +/+ mice after bleomycin treatment ( Figure 4). These findings suggested that exacerbation of bleomycin-induced lung fibrosis in mPGES-1 −/− mice was independent on Smad2/3 pathway. 2.1.5. TGF-β1 Treatment Reduced the Expression of mPGES-1, EP2 and EP4 in MRC-5 Cells TGF-β1 was used to establish a fibrosis model in MRC-5 cell lines. In MRC-5 cells, TGF-β1 treatment significantly induced α-SMA expression, which is one of the main markers for fibrosis in lungs and other tissues or cells. Importantly, in MRC-5 cells treated with TGF-β1, with an increase in α-SMA expression, the expression levels mPGES-1, EP2 and EP4 were reduced. In contrast, the expression levels of EP1 and EP3 were not significantly affected by TGF-β1 in MRC-5 cells ( Figure 5).

Discussion
In the current study, we found that bleomycin-treatment resulted in more notable histopathological changes, higher hydroxyproline content and increased mRNA and protein expression of α-SMA in mPGES-1 −/− mice than in wild type mice. We further demonstrated beyond inhibition of PGE 2 production from mPGES-1, bleomycin also repressed the expression of PGE 2 receptors EP2 and EP4. These findings strongly suggested that inhibitoin of mPGES-1-PGE 2 -EP2/EP4 signaling axis plays an important role in bleomycin-induced IPF. Furthermore, we found that only the agonist of EP2 can inhibit TGF-β1-induced increase in α-SMA expression in MRC-5 cells with mPGES-1 knockdown.
Clearly, these findings revealed that mPGES-1-derived PGE 2 exerts a beneficial effect against pulmonary fibrosis via EP2 receptor-mediated signal transduction. mPGES-1, a terminal synthase of PGE 2 , has been reported as a key mediator in a variety of diseases. To our knowledge, little is known about the role of mPGES-1 in pulmonary fibrogenesis up to now. McCann et al reported that mPGES-1 expression is increased in bleomycin-induced skin fibrogenesis. mPGES-1 deficient mice were resistant in bleomycin-induced skin fibrogenesis. These findings suggested that mPGES-1 is required in the development of skin fibrogenesis [17]. In the present study, we found that deficiency of mPGES-1 aggravated bleomycin-induced injury and pulmonary fibrosis. Currently, the role of inflammation mediators such as PGE 2 in the pathogenesis of fibrosis is controversial, with this lack of clarity possibly attributed to mPGES-1 operating via tissue specific mechanisms.
In healthy lung tissue, microenvironment homeostasis is dependent on alveolar epithelial cell (AEC) and mesenchymal cell (MSC) crosstalk [23]. The ability of the AECs to generate PGE 2 is critical for blocking fibroblast proliferation [24][25][26]. In consistent with our observations, other studies also revealed that lung fibroblasts from IPF patients had reduced PGE 2 levels [18]. In our study, we found that bleomycin treatment reduced PGE 2 content in the lungs of wild type mice. Notably, bleomycin treatment resulted in a marked reduction in lung PGE 2 levels in mPGES-1 −/− mice. We also found that when compared to mPGES-1 +/+ mice, mPGES-1 −/− mice only exhibited a modest reduction in cellular PGE 2 levels in the lungs without bleomycin treatment. These findings may suggest that other two types of PGE 2 synthase, mPGES-2 and cPGES, play important roles in controlling PGE 2 production in the basic condition. In support, the expression levels of mPGES-2 and cPGES remained unchanged in the lungs of mPGES-1 +/+ mice and mPGES-1 −/− mice receiving belomycin or saline ( Figure 2B). Overall, these findings strongly suggested that a reduction in mPGES-1 expression, leading to the inhibition of PGE 2 synthesis, plays a unique and important role in IPF induced by bleomycin. PGE 2 activates its downstream pathways through 4 different subtypes of receptors, which are designated as EP1 (E prostanoid receptors 1), EP2, EP3 and EP4, respectively, in various tissues [27]. It had been previously reported that PGE 2 exerts anti-fibrotic actions through EP2 and EP4 pathways in vitro. The activation of these pathways increased cellular cyclic adenosine monophosphate (cAMP) level, the activation of protein kinase A (PKA), or exchange protein, leading to the repression of matrix proteins expression and blockade of the proliferation of fibroblasts [28,29]. In fibrotic lungs, the expression levels of EP receptor subtypes are altered. Similar to our findings, it had also been previously reported that EP2 expression was decreased in bleomycin-induced fibrotic mice, leading to increased fibroblast proliferation and decreased matrix proteins degradation [28]. Huang SK et al. confirmed that there was some variable PGE 2 resistance in fibroblast from usual interstitial pneumonia patients and the reason for that was partly due to decreased expression of EP2 and PKA [30]. In our experiment, we found that EP2 and EP4 expression levels was reduced in lungs from mice after bleomycin treatment, and human lung fibroblasts induced by TGF-β1. Furthermore, we observed the fibroblast to myofibroblast differentiation was inhibited by butaprost, EP2 agonist, but not by agonist of EP1, EP3 or EP4. So these findings indicated mPGES-1-derived PGE 2 exerts a beneficial effect against lung fibrosis through EP2 receptor-mediated signaling pathway, which resembled previous research [31].
TGF-β1 induced myofibroblast differentiation via adhesion-dependent pathways besides Smad pathways. FAK, one of adhesion complexes, is a cytoplasmic protein kinase, and its phosphorylation is related with myofibroblasts differentiation [22,32,33]. In our investigations, we found that butaprost could weaken FAK autophosphorylation, but had no effect on Smad2/3 phosphorylation after TGF-β1 and mPGES-1 siRNA cotreatment in MRC-5 cells, what's more, FAK inhibitor attenuated α-SMA expression co-administrated by TGF-β1 and mPGES-1 siRNA. These findings indicated that mPGES-1-PGE 2 -EP2 axis's beneficial effect against TGF-β1-induced α-SMA expression via FAK-correlated adhesion pathway. In addition, the specific mechanism between mPGES-1 and pulmonary fibrosis in patients need to be explored, which will be the key note of our work in future.

Animal Groups and Management
Mice were anesthetized with 5 mg/kg pentobarbital sodium (Yunpeng, China) injection, and their tracheas were exposed and instilled with bleomycin (5 mg/kg, Invitrogen, Carlsbad, CA, USA) or saline. After 28 days, the mice were sacrificed and perfused with phosphate buffer saline (PBS) via left ventricle to remove blood from lungs. The left lung was removed for histological examination to evaluate the fibrosis score, and the right lung was used for measurements of hydroxyproline and PGE 2 , for western blotting and real-time PCR (RT-PCR) to quantify related proteins and mRNAs. Twenty-six mice (male, 6-8 weeks old) were randomly assigned into four groups: group 1, mPGES-1 +/+ mice instilled with saline (n = 5); group 2, mPGES-1 +/+ mice instilled with bleomycin (n = 8); group 3, mPGES-1 −/− mice instilled with saline (n = 5); group 4, mPGES-1 −/− mice instilled with bleomycin (n = 8).

Cell Culture
Human fetal lung fibroblast cells (MRC-5) were obtained from Basic Research Institute of Peking Union Medical College (Beijing, China), and cultured as previously described [21]. In brief, cells were grown in minimal essential medium supplemented with 10% fetal bovine serum (FBS), essential amino acids, nonessential amino acids and antibiotics, and were maintained at 37 °C in a humidified 5% CO 2 atmosphere, and medium was changed every 3 days. Cells (3 × 10 5 cells/well) were then seeded into a 6-well cell culture plate. When cells reached 80% confluence, they were incubated in FBS-free medium for 24 h to synchronize their growth, and then were grown in FBS-containing medium and treated with TGF-β1 (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), PGE 2 , sulprostone, butaprost, CAY10580 (Cayman, Ann Arbor, MI, USA), or FAK inhibitor 14 (Santa Cruz Biotechnology, Inc.). PGE 2 was dissolved in ethanol, TGF-β1 in citrate buffer, and sulprostone, butaprost, CAY10580 and FAK inhibitor 14 in dimethyl sulfoxide (DMSO).

Knockdown of mPGES-1 Expression in Human Lung Fibroblasts by RNA Interference
We used synthetic siRNA (GenePharma, Shanghai, China) for the experiment. The siRNA sequences for knockdown of mPGES-1 were sense: 5'-GGAACGACAUGGAGACCAUTT, and antisense: 5'-AUGGUCUCCAUGUCGUUCCTT; for negative control were sense: 5'-UUCUCCGAA CGUGUCACGUTT, and antisense: 5'-ACGUGACACGUUCGGAGAATT. MRC-5 cells growing to 80% confluence were synchronized by incubation in FBS-free medium for 24 h and then transfected with 0.1 nmol siRNA-mPGES-1 or siRNA-control using lipofectamine 2000 (Invitrogen). After the transfection for 6 h, TGF-β1 or other reagents were added into the medium for defined hours to study the role of mPGES-1 in the pathogenesis of IPF.

Lung Histological Examination
Lung samples were fixed in 4% paraformaldehyde, embedded in paraffin and made sections of 4 μm thickness. Sections were stained with hematoxylin and eosin (H&E), or Masson trichrome, and assessed pulmonary fibrosis degree by Aschroft scoring method [34].

RNA Extraction and Real-Time PCR (RT-PCR)
Lung total RNA was extracted and reverse-transcribed into cDNA by using a RNA extraction kit (BioTeke, Beijing, China) and following the manufacturer's protocol (Transgen, Beijing, China). Then 1 μL of cDNA was subjected to PCR in a 25 μL final reaction volume for analyzing the expression of α-smooth muscle actin (α-SMA), fibronectin (FN) mPGES-1, EP1, EP2, EP3 and EP4. The RT-PCR cycling program consisted of a preliminary denaturation (95 °C for 7 min), followed by 35 cycles (95 °C for 30 s, 59 °C for 90 s, and 72 °C for 30 s) and a final elongation step (72 °C for 7 min). Relative gene expression was calculated using the comparative C t method [35] to assess the difference in gene expression between the gene of interest and an internal standard gene. Primer sequences and products are listed in Table 1 and the β-actin was used as an internal standard control.

Measurement of Lung Hydroxyproline Content
The hydroxyproline content of the lung for each group was quantified utilizing a Hydroxyproline Testing Kit (Jiancheng, Nanjing, China) according to the manufacturer's instructions.

Quantification of PGE 2 in Lung
Thirty micrograms of lung tissue were resolved in 1 mL homogenization buffer (0.1 M phosphate, pH 7.4, containing 1 mM EDTA and 10 μM indomethacin). The samples were homogenized with a sonicator and centrifuged at 5000 rpm speed, and then the supernatant was collected. The content of PGE 2 and LTC 4 in lungs was determined using an enzyme immunoassay (EIA) kit (Cayman) according to the manufacturer's protocols. The PGE 2 and LTC 4 content was corrected by the protein content for each sample.

Statistical Analysis
The results were presented as means ± standard error of the mean (SEM). Comparisons were performed using a one-way ANOVA followed by Student's t test, with a p value <0.05 considered statistically significant.

Conclusions
In summary, we firstly reported mPGES-1 deficiency aggravates lung fibrosis after bleomycin exposure in mice. mPGES-1-derived PGE 2 exerts beneficial effects against pulmonary fibrosis mainly via activation of EP2 receptor signaling transduciton. Activation of mPGES-1-PGE 2 -EP2 signaling pathway may represent a novel strategy for treatment of IPF patients.