Protective effect of ebselen on bleomycin-induced lung fibrosis: analysis of the molecular mechanism of lung fibrosis mediated by oxidized diacylglycerol

Abstract

The molecular mechanisms underlying the development of pulmonary fibrosis remain unknown, and effective treatments have not yet been developed. It has been shown that oxidative stress is involved in lung fibrosis. Oxidized diacylglycerol (DAG) produced by oxidative stress is thought to play an important role in lung fibrosis. This study assessed the effect of oxidized DAG in an animal model of pulmonary fibrosis induced by aspiration of bleomycin (BLM) into the lungs. The inhibitory effect of ebselen on pulmonary fibrosis was also investigated. In lung fibrotic tissue induced by BLM, an increase in lipid peroxides and collagen accumulation was observed. Moreover, the levels of oxidized DAG, which has strong protein kinase C (PKC) activation activity, were significantly increased over time following the administration of BLM. Western blotting showed that phosphorylation of PKCα and δ isoforms was increased by BLM. Oral administration of ebselen significantly suppressed the increase in oxidized DAG induced by BLM and improved lung fibrosis. PKCα and δ phosphorylation were also significantly inhibited. The mRNA expression of α-smooth muscle actin and collagen I (marker molecules for fibrosis), as well as the production of transforming growth factor-β and tumor necrosis factor-α(a potentially important factor in the fibrotic process), were increased by BLM and significantly decreased by ebselen. The administration of BLM may induce lipid peroxidation in lung tissue, while the oxidized DAG produced by BLM may induce overactivation of PKCα and δ, resulting in the induction of lung fibrosis.

Keywords:

• Lung fibrosis
• ebselen
• oxidized diacylglycerol
• protein kinase C
• bleomycin

Introduction

Among the organs of the human body, the lung is the most susceptible to inflammation due to its direct contact with the atmosphere and exposure to bacteria, viruses, and chemical substances. Exposure of alveolar epithelial cells to proinflammatory agents induces inflammation along with the production of cytokines, such as interleukin-8 (IL-8) and transforming growth factor-β (TGF-β) [1]. Furthermore, following exposure of alveolar epithelial cells to chronic inflammatory reactions, fibroblasts produce extracellular matrices (e.g. collagen) to repair the damage, resulting in fibrosis [2]. Interstitial pneumonia is a group of diseases in which inflammation and fibrotic lesions occur in the alveolar walls of the lungs (i.e. the interstitium). There are two types of interstitial pneumonias, namely those with known causes (i.e. collagen diseases, dust pneumonia, drug-induced, radiation-induced, sarcoidosis, and hypersensitivity pneumonias) and those with unknown causes; the latter are termed idiopathic interstitial pneumonias (IIPs). Idiopathic pulmonary fibrosis (IPF) is the most common of these diseases and is associated with a poor prognosis. It is a chronic and progressive disease characterized by progressive fibrosis and irreversible cell lung formation. According to reports in Europe and the United States of America, the average survival time after diagnosis is 28–52 months [3–7]. In IIPs, repeated inflammation induces excessive repair in alveolar epithelial cells and vascular endothelial cells, leading to fibrosis and destruction of the alveolar architecture. Fibroblast growth factors, such as TGF-β [8], tumor necrosis factor-α (TNF-α) [9,10], and platelet-derived growth factor (PDGF) [11,12] are produced by macrophages and type II alveolar epithelial cells during the period of fibrosis. Although PDGF is a protein generally secreted by platelets, its production by macrophages is enhanced during lung injury. It has been reported that PDGF is involved in the proliferation of fibroblasts and smooth muscle cells, and enhances the production of other cytokines/growth factors (e.g. TGF-β) [13]. TGF-β is a protein produced in the lungs, kidneys, bone marrow, and almost all other types of cells. It is a cytokine that exhibits a variety of biological activities, such as inhibiting cell proliferation and differentiation and promoting collagen synthesis. There are five subtypes, among which TGF-β1 is involved in wound healing. Moreover, it has been reported that overproduction of TGF-β1 causes fibroblast proliferation and activation, and simultaneously promotes differentiation into myofibroblasts [14,15].

Bleomycin-induced (BLM-induced) pulmonary fibrosis models have been used to analyze the pathogenesis of lung fibrosis in IIPs. Using a BLM-induced pulmonary fibrosis model, Verma et al. reported that the administration of quercetin (an antioxidant) reduced lipid peroxides in lung tissue and improved fibrosis [9,16]. This strongly suggests that oxidative stress and lipid peroxidation reactions are involved in BLM-induced lung fibrosis. Inflammatory cells in the bronchoalveolar lavage (BAL) of patients with IPF induce oxidative stress injury by producing large amounts of superoxide radicals and hydrogen peroxide. Moreover, the levels of glutathione (an antioxidant) are decreased in the alveolar epithelium and BAL. These effects impair the antioxidant system, strongly suggesting a relationship between lung fibrosis and oxidative stress [17].

We have shown that the administration of carbon tetrachloride induces oxidative stress in the liver, resulting in the production of oxidized diacylglycerol (DAG) (a lipophilic signaling molecule). In addition, we demonstrated that oxidized DAG induces the hyperactivation of protein kinase C (PKC) signaling, leading to liver tissue injury and fibrosis [18,19]. It has also been reported that PKCδ, which is activated by oxidized DAG, promotes lung fibrosis by interacting with TGF-β [20]. This finding strongly suggested that oxidized DAG is a key molecule for lung fibrosis.

Ebselen, an agent with similar enzymatic activity to that of phospholipid hydroperoxide glutathione peroxidase (PHGPx), is an intracellular enzyme that uses glutathione to reduce phospholipid hydroperoxides. PHGPx is present in the plasma membrane, nucleus, and mitochondria; it regulates signaling by lipid peroxides and reactive oxygen species in various signaling pathways and inhibits apoptosis and necrosis [21]. In this study, the effect of oxidized DAG on pulmonary fibrosis was analyzed using an animal model of BLM-induced pulmonary fibrosis. In addition, the inhibitory effect of ebselen (a drug with reduction-elimination activity for oxidized DAG) on pulmonary fibrosis was investigated.

Materials and methods

Animals

Eight-week-old male C57BL/6J mice were provided by CREA Japan, Inc. (Tokyo, Japan). Mice were maintained in a temperature- and light-controlled environment and had ad libitum access to food and water. All animal experiments were approved by the Animal Experimentation Committee, Isehara Campus (Tokai University, Kanagawa Japan).

Treatment and experimental groups

BLM in phosphate-buffered saline (PBS) was administered (4 mg/kg body weight) into the lungs of mice. Ebselen (1.6 mg/kg) was suspended in 1% tragacanth (Sigma–Aldrich, St. Louis, MO, USA) [22]. Mice were randomly divided into four groups: (1) PBS + vehicle (tragacanth); (2) PBS + ebselen; (3) BLM + vehicle; and (4) BLM + ebselen groups. BLM-induced lung fibrosis model mice were created as described below. Mice were anesthetized with isoflurane and fixed on a surgery board. The tongue of each mouse was pulled using forceps, and BLM or PBS solutions were placed onto the oropharynx by pulmonary aspiration upon awakening. Ebselen or tragacanth gums were administered orally for 5 days per week.

Quantitation of DAG-O(O)H content

DAG-O(O)H was fractionated and analyzed using high-performance liquid chromatography (HPLC). This method was performed as previously described [18]. The lungs of mice were minced using scissors, and lipids were extracted in 2-propanol containing 1-palmitoyl-3-arachidoylglycerol hydroxide as an internal standard, 20 mM butylated hydroxytoluene and 200 mM triphenylphosphine. The extract was infused into a CAPCELL PAK C18 column (Shiseido, Tokyo, Japan) in reversed-phase HPLC using methanol as a mobile phase and corrected DAG-O(O)H fraction. The fraction was injected into a SUPELCOSIL column (Sigma–Aldrich, St. Louis, MO, USA) in normal-phase HPLC using hexane/2-propanol as a mobile phase and corrected DAG-O(O)H fraction.

Immunohistochemistry

Lung tissues were fixed overnight using Mildform 10 N (Wako, Osaka, Japan). Fixed lungs were embedded in paraffin and cut in sections (thickness: 4 μm). Sections were deparaffinized in xylene and rehydrated using a graded series of ethanol. Endogenous peroxidase activities were blocked using 0.3% hydrogen peroxide in methanol for 30 min at room temperature. For 4-hydroxyl-2-nonenal (4-HNE) immunostaining, sections were heated in a microwave for 10 min at 98 °C in 10 mM of sodium citrate (pH 6.0) for antigen retrieval and incubated in 10% normal goat serum blocking buffer for 10 min at room temperature for blocking. For α-smooth-muscle-actin (αSMA) staining, sections were incubated in 10% normal goat serum blocking buffer for 30 min at room temperature for blocking. Subsequently, sections were incubated with the primary antibodies 4-HNE (1:100; JAICA, Shizuoka, Japan), TGF-βI and αSMA (1:100; Abcam, Cambridge, MA, USA) overnight at 4 °C. After washing with PBS, the sections were incubated with secondary antibody using the Simple Stain MAX-PO MULTI (Nichirei Bioscience Inc., Tokyo, Japan) for 60 min at room temperature. The sections were visualized with 3’3-diaminobenzidine, and nuclei were counterstained with hematoxylin.

Sirius red

Section slides (thickness: 4 μm) were incubated with 0.1% Sirius Red (Wako, Osaka, Japan) dissolved in saturated picric acid for 1 h. The slides were washed with 0.5% acetic acid and water. Six images were captured from each slide. The area of Sirius Red-positive cells was calculated using the acquired image data. The positive cell area per lung tissue area and its percentage were calculated and graphically represented using the ImageJ software (National Institutes of Health, Bethesda, MD, USA). Data are presented as the mean of 4–6 individuals from each group.

Western blotting

The protein concentration of each sample was measured using a DC protein assay kit (Bio-Rad, Hercules, CA, USA). The samples were heated at 95 °C for 5 min and loaded to 10% sodium dodecyl sulfate-polyacrylamide gel electrophoresis gels. Next, separated proteins were transferred to nitrocellulose membranes (Millipore, Billerica, MA, USA). After blocking for 60 min at room temperature with 3% BSA containing 0.05% Tween 20, the membranes were incubated overnight at 4 °C with rabbit antibodies against PKCα, βΙ, βΙΙ, and δ antibodies (1:1000, Santa Cruz Biotechnology, Santa Cruz, CA, USA); phosphorylated PKCα, βΙΙ, and PKCδ antibodies (1:500; Cell Signaling Technology, Berkeley, CA, USA); phosphorylated PKCβΙ antibody (1:500; Thermo Fisher Scientific, Waltham, MA, USA); c-Raf, MEK, ERK antibodies (1:500; Cell Signaling Technology, Berkeley, CA, USA); and phosphorylated c-Raf, MEK, and ERK antibodies (1:500; Cell Signaling Technology, Berkeley, CA, USA). This was followed by incubation with peroxidase-conjugated anti-rabbit immunoglobulin G antibody (Cell Signaling Technology, Berkeley, CA, USA) at room temperature. Immune complexes were visualized using an enhanced Immobilon detection kit (Millipore, Billerica, MA, USA). Densitometric analyses were performed using the CS Analyzer ver. 3.0 software (ATTO Corp., Osaka, Japan).

Real-time reverse transcription-polymerase chain reaction

TRIzol RNA Isolation Reagent (Life Technologies, Gaithersburg, MD, USA) was used for total RNA extraction from the lungs of mice. Total RNA quantity was determined using an absorption spectrometer (NanoDrop 2000 C, Thermo Fisher Scientific, Tokyo, Japan). The High-Capacity cDNA Archive kit (Applied Biosystems, Foster City, CA, USA) was used for reverse transcription. Real-time reverse transcription-PCR was performed using the TaqMan Universal PCR Master Mix (Applied Biosystems) according to the instructions provided by the manufacturer. The levels of αSMA, collagen I, TGF-β, TNF-α, and glyceraldehyde-3-phosphate dehydrogenase were quantified using commercially available kits (TaqMan Gene Expression Assays Mm00725412_s1, Mm00801666_g1, Mm01178820_m1, Mm00443258_m1 and Mm03302249_g1 respectively; Applied Biosystems). These primer sets were designed to span one intron to allow the identification of genomic contamination. The reaction protocol consisted of the following cycles: 95 °C for 15 min, 95 °C for 15 s, and 60 °C for 1 min for 50 cycles of PCR amplification on an Opticon 2 System (Bio-Rad). All data were analyzed on an Option monitor 3 (Bio-Rad).

Results

Lung fibrosis induced by the administration of BLM

The administration of BLM induced lung fibrosis over time. Sirius Red staining revealed collagen accumulation in mouse lung tissue 3, 7, and 14 days after the administration of BLM. In the group treated with BLM, collagen accumulation was observed in the injured area compared with the control group (PBS-treated mice). Fibrosis was also increased in a time-dependent manner (Figure 1a). Quantification using Sirius Red-stained images revealed that the number of positive cells was significantly higher in the BLM group versus the control group (Figure 1b).

Figure 1. Pulmonary fibrosis induced by the administration of BLM. (a) Collagen was observed by Sirius Red staining using paraffin-embedded mouse lung sections 3, 7, and 14 days after the administration of BLM. Staining with Sirius Red was only observed around blood vessels in the control group treated with PBS. Collagen accumulation in the lung of animals was observed on days 3, 7, and 14 after the administration of BLM. (b) The graph indicates the accumulation in the lungs and represents the area of stained collagen in the section. P < 0.05 denotes a statistically significant difference. BLM: bleomycin; PBS: phosphate-buffered saline.

Increased lipid peroxidation and accumulation of oxidized DAG in fibrotic sites

Immunostaining for 4-HNE, a marker of lipid peroxidation, was performed. Accumulation of 4-HNE was observed in the lungs of mice on days 7 and 14 after the administration of BLM; the majority of positive cells were alveolar epithelial cells (Figure 2a). It has been shown that oxidized DAG is involved in neuronal cell death and liver fibrosis by over-activating PKC [18,19,23]. Therefore, it was hypothesized that oxidized DAG may be involved in the development of fibrosis in the lung. Therefore, the levels of oxidized DAG were measured through HPLC. The results showed that the oxidized DAG in the BLM group was significantly increased in a time-dependent manner (Figure 2b).

Figure 2. Accumulation of lipid peroxides in the lungs induced by the administration of BLM. (a) Immunohistochemistry was performed using paraffin-embedded mouse lung sections on days 3, 7 and 14 after the administration of BLM; staining for 4-HNE was performed. There was no staining for 4-HNE observed in the control group. Accumulation of 4-HNE was observed on days 3, 7 and 14 after the administration of BLM; 4-HNE staining was observed in lung epithelial cells. (b) Lipids in tissues were extracted from fresh frozen lung tissues using isopropyl alcohol and methanol; oxidized DAG was purified and quantitatively analyzed using HPLC. The administration of BLM resulted in a marked accumulation of oxidized DAG. P < 0.05 denotes a statistically significant difference. 4-HNE: 4-hydroxyl-2-nonenal; BLM: bleomycin; DAG: diacylglycerol; HPLC: high-performance liquid chromatography.

Inhibitory effect of ebselen on lung fibrosis

The inhibitory effect of ebselen on lung fibrosis was investigated. Lung fibrosis was observed via Sirius Red staining. The number of Sirius Red-positive cells was significantly increased in the BLM group versus the control group (Figure 3). However, the administration of ebselen significantly suppressed the lung fibrosis induced by treatment with BLM (Figure 3).

Figure 3. Effect of ebselen on BLM-induced pulmonary fibrosis. (a) Ebselen was given orally 5 days per week after the administration of BLM. Two weeks later, lungs were collected for analysis. There was no staining, other than perivascular, observed in the control and ebselen groups. Collagen accumulation was observed in the BLM + vehicle group; in contrast, collagen was suppressed in the BLM + ebselen group. (b) Observation of collagen fibers through Sirius Red staining and calculation of the area ratio of accumulated collagen were performed. BLM: bleomycin; PBS: phosphate buffered saline. P < 0.05 denotes a statistically significant difference.

Inhibition of lipid peroxidation reaction and reduction of oxidized DAG levels by ebselen

The lipid peroxidation reaction in lung tissue induced by the administration of BLM was inhibited by treatment with ebselen (Figure 4a). Also, the levels of oxidized DAG in lung tissue, which were increased by the administration of BLM, were significantly decreased by treatment with ebselen (Figure 4b).

Figure 4. Inhibitory effect of ebselen on lipid peroxide. (a) Quantitative analysis of DAG-O (O) H by 4-HNE immunohistochemistry and HPLC was performed to investigate the effect of ebselen on the accumulation of lipid peroxide. There was no accumulation of 4-HNE in the control and ebselen groups. In the BLM group, 4-HNE staining was mainly observed in the alveolar epithelium, but positive cells were also observed in pulmonary interstitium; in contrast, stainability was attenuated in the BLM + ebselen group. (b) Similarly, in the measurement of oxidized DAG using HPLC, the increase induced by the administration of BLM was significantly suppressed by treatment with ebselen. P < 0.05 denotes a statistically significant difference. BLM: bleomycin; HPLC: high-performance liquid chromatography.

Inhibition of PKCα and δ activation by ebselen

We have previously shown that oxidized DAG induces liver fibrosis by PKCα and δ activation [23]. In addition, PKCδ has been found to promote lung fibrosis by interacting with TGF-β [20]. Therefore, we analyzed the activation of PKC isoforms (PKCα, βI, βII, and δ), which are potential targets of oxidized DAG, at lesion sites in pulmonary fibrosis by BLM administration. We found that phosphorylation or activation of PKCα and δ was enhanced in the BLM group. We also found that the administration of ebselen significantly inhibited the phosphorylation of PKCα and δ (Figure 5).

Figure 5. Effect of ebselen on the phosphorylation of PKC isoforms. (a) Western blotting was performed using a anti-phosphorylated PKC antibody to investigate the effect of ebselen on the activation of PKCα, βI, βII, and δ. (b) Measurement of band intensity. Phosphorylation of PKC isoforms induced by the administration of BLM was significantly suppressed after treatment with ebselen. P < 0.05 denotes a statistically significant difference. BLM: bleomycin; PKCα: protein kinase C-alpha; PKCβI: protein kinase C-beta1; PKCβII: protein kinase C-beta2; PKCδ: protein kinase C-delta.

Inhibition of MAPK activation by ebselen

Li et al. reported that TGF-β1 was suppressed by administration of an inhibitor of Raf 1, a molecule in the MAPK pathway, in a mouse model of BLM-induced pulmonary fibrosis [24]. Therefore, we examined the MAPK cascade and found that it was activated in lung fibrosis tissue induced by BLM and that its activation was suppressed by ebselen (Figure 6).

Figure 6. Effect of ebselen on MAPK pathway activation. (a) Western blot analysis was performed with MAPK (c-Raf, MEK and ERK) phospho-specific antibodies. (b) Measurement of band intensity. Phosphorylation of MAPK was significantly increased by BLM administration and suppressed by ebselen treatment. P < 0.05 denotes a statistically significant difference. BLM: bleomycin; MAPK: mitogen-activated protein kinase; ERK: extracellular signal-regulated kinase; MEK: MAPK/ERK kinase.

Expression of fibrosis markers αSMA, collagen, TGF-β and TNF-α

The mRNA expression of αSMA and collagen I (marker molecules for fibrosis) was markedly increased by the administration of BLM. Notably, the BLM-induced increase in the mRNA expression of αSMA and collagen I was significantly suppressed by the administration of ebselen. The production of TGF-β (an important factor in the fibrotic process) and TNF-α was increased by BLM and significantly decreased by ebselen (Figure 7).

Figure 7. Inhibitory effect of ebselen on αSMA, collagen, TGF-β1, and TNF-α expression. Quantitative real-time RT-PCR was performed to investigate the effect of ebselen on the expression of αSMA, collagen, TGF-β1, and TNF-α involved in fibrosis. The administration of BLM increased the expression of αSMA, collagen, TGF-β1, and TNF-α; however, these changes were suppressed by treatment with ebselen. P < 0.05 denotes a statistically significant difference. αSMA: α-smooth muscle actin; BLM: bleomycin; RT-PCR: reverse transcription-polymerase chain reaction; TGF-β1: transforming growth factor-β1; TNF-α: tumor necrosis factor-α.

Analysis of TGF-β expression by immunohistochemistry

TGF-β showed strong staining for macrophages localized in the interstitium of alveoli that had become fibrotic after BLM administration. The increase in TGF-β expression by BLM was markedly suppressed by ebselen (Figure 8). Type II alveolar epithelial cells also showed staining but no significant changes were observed. No changes in TGF-β expression or morphological changes were observed after treatment with ebselen alone.

Figure 8. Analysis of TGF-β expression by immunohistochemistry. Immunohistochemical analysis of TGF-β expression in mouse lung. Many TGF-β positive cells were observed in BLM treated mice lungs (arrowheads), whereas few TGF-β positive cells were observed in BLM and ebselen treated mouse lungs. BLM: bleomycin; TGF-β1: transforming growth factor-β1.

Discussion

Currently, there is no effective treatment for IPF, and the disease is linked to a poor prognosis. Steroids and immunosuppressants have been used previously in the treatment of IPF, with limited effectiveness. In addition, although it has been reported that recently approved antifibrotic drugs are useful in controlling disease progression, these agents do not result in a complete cure of pulmonary fibrosis with irreversible changes [25,26]. It has been reported that inflammatory cells produce large amounts of superoxide and hydrogen peroxide in the BAL of patients with IPF, causing cell injury due to oxidative stress [17]. In addition, it has been shown that the administration of antioxidant drugs (e.g. quercetin) suppressed the increase in the plasma levels of TNF-α and decreased the levels of lipid peroxides in animal models of pulmonary fibrosis. These findings strongly suggested the involvement of oxidative stress in lung fibrosis [9].

Inflammatory tissues produce reactive oxygen species (ROS), such as hydrogen peroxide and lipid peroxide. ROS are scavenged by the antioxidant defense system in the body. PHGPx (an antioxidant defense enzyme) possesses reduction and scavenging activities for phospholipid hydroperoxide. Ebselen (an antioxidant agent) contains intramolecular organic selenium, which inhibits apoptosis, necrosis, and ferroptosis induced by lipid peroxides and reactive oxygen species due to its PHGPx-like activity [21,27,28]. It also inhibits the arachidonic acid cascade via inhibition of nicotinamide adenine dinucleotide phosphate-oxidase [29] and lipoxygenase [30]. Cardioprotective effects [31], anti-inflammatory effects [32–34], anti-carcinogenic effects [35], and inhibition of TNF-α [32] due to these actions have also been reported. In line with the recent trend for the effective utilization of therapeutic drugs by repositioning, ebselen has been reaffirmed as a useful therapeutic agent for various diseases. In a recent phase II trial conducted for the prevention of noise-induced hearing loss and mania, there were no reports of major adverse events [36,37]. It has also been demonstrated that ebselen may serve as an inhibitor of the major protease of severe acute respiratory syndrome coronavirus 2. Furthermore, it has also shown anti-inflammatory effects, such as inhibiting TNF-α and IL-1β in BAL in a murine model of bronchitis [32,38], and is expected to be a therapeutic agent against coronavirus disease [39,40]. This study revealed that ebselen inhibits lung fibrosis by scavenging the reduction of oxidized DAG produced in the lung tissue of mice with BLM-induced pulmonary fibrosis. It is expected that ebselen will be clinically utilized as a potential new therapeutic option for pulmonary fibrosis. On the other hand, antioxidants other than ebselen have been reported to inhibit experimental pulmonary fibrosis [41–43]. Many of the antioxidants that inhibit experimental lung fibrosis are known radical scavengers. These antioxidants are thought to exhibit anti-fibrotic effects in the lung by scavenging radical species upstream of lipid peroxidation reaction. Although lipid peroxidation is an important factor in promoting lung fibrosis, oxidative stress factors other than lipid peroxidation, such as ferroptosis, are thought to be involved in the induction of lung fibrosis [43]. Thus, it can be inferred that these antioxidants are more effective for a wider range of pulmonary fibrosis compared with ebselen. Relative to these other antioxidants, ebselen seems to act more specifically and strongly on lung fibrosis induced by lipid peroxides.

In this study, we used a murine lung fibrosis model in which lung fibrosis was induced by the administration of BLM (a drug causing strong oxidative stress injury). The results revealed that the levels of oxidized DAG were obviously increased by the administration of BLM. In addition, the administration of ebselen, which has lipid peroxide reduction and scavenging activities, suppressed the increase in oxidized DAG and inhibited the development of lung fibrosis. We focused on the overactivation of PKC and cytotoxic effects of oxidized DAG, and investigated the molecular mechanism involved in this process [18,23]. DAG contains an unsaturated fatty acid that is easily oxidized at the 2-position of the glycerol backbone. Thus, it was hypothesized that DAG is oxidized by ROS under oxidative stress, resulting in oxidized DAG. In fact, the levels of oxidized DAG were markedly increased in rat liver tissue following the induction of oxidative stress by carbon tetrachloride [18]. It has been shown that oxidized DAG activates the mitogen-activated protein kinase (MAPK) cascade by over activating PKC, and promotes the production of cytokines (e.g. TNFα). This results in an inflammatory pathology with neutrophilic infiltration and exacerbates the condition [18,23]. In the present study, it was also clear that the MAPK cascade is activated in lung fibrosis tissue induced by BLM. Furthermore, activation of the MAPK cascade was inhibited by ebselen. These results suggest that activation of MAPKs via the PKC signaling pathway plays a pivotal role in the induction of lung fibrosis by BLM. In addition, it has been demonstrated that the accumulation of oxidized DAG and activation of four PKC isoforms (including PKCδ) in mouse liver tissues treated with carbon tetrachloride for a long period of time increase the production of various cytokines and promote liver fibrosis [23]. In this study, we found that the administration of BLM-activated PKCα and δ, led to increased the production of TGF-β (an established initiator molecule of fibrosis). It has been reported that PKCδ exacerbates pulmonary fibrosis by interacting with TGF-β [20]. Moreover, treatment with BLM induces inflammation along with neutrophil infiltration [33], resulting in the production of TGF-β by type II alveolar epithelial cells and macrophages [33,44]. TGF-β causes multiplication and activation of fibroblasts, and simultaneously induces their differentiation into myofibroblasts, leading to organ fibrosis [14,15,45]. In this process, it is considered that myofibroblasts accumulate αSMA and shrink it in order to repair tissues damaged by inflammation. In addition, immunohistochemical analysis revealed that TGF-β is strongly expressed in macrophages in lung tissue undergoing fibrosis induced by bleomycin treatment. This suggests that alveolar macrophages play an important role in the process of bleomycin-induced pulmonary fibrosis. 4-HNE, a marker of lipid peroxidation, was stained mainly in alveolar epithelial cells, but positive images were also observed in the pulmonary interstitium, suggesting that oxidized DAG may be produced in macrophages. Nevertheless, the accumulation and replacement of collagen are thought to induce fibrosis in lung tissue. The present findings suggest that the administration of ebselen suppressed the accumulation of oxidized DAG induced by treatment with BLM. Subsequently, it suppressed the activation of PKCδ (a target of oxidized DAG), resulting in decreased production of TGF-β and suppression of lung fibrosis. These results strongly suggest that oxidized DAG is the starting point of lung fibrosis induced by BLM.

There are three main causes of lipid oxidation reactions in vivo: radical reactions, enzymatic reactions (e.g. lipoxygenase), and oxidation reactions by singlet oxygen [46,47]. The oxidation reaction of DAG (a type of lipid) is also considered to be produced by the three aforementioned causes. Therefore, radical scavengers (e.g. vitamin E, N-acetylcysteine, and coenzyme Q), lipoxygenase inhibitors, and carotenoids with singlet oxygen scavenging properties may inhibit the production of oxidized DAG and prevent the development of diseases, such as lung fibrosis. Ebselen, which was used as a reductive scavenger of oxidized DAG in this study, is thought to be more effective in combination with these drugs. Administration of ebselen at high doses is challenging owing to the presence of selenium in its enzyme-active center. Further investigation of the administration of ebselen in combination with the above-mentioned drugs is warranted to develop a protocol that can be used in clinical practice.

In this study, we report the importance of oxidized DAG as a causative factor of lung fibrosis and we examined the antifibrotic effect of ebselen. Yuan et al. have shown that ferroptosis, a form of cell death induced via lipid peroxidation reactions, is an important factor in lung fibrosis [43]. PHGPx, which has an enzymatic activity similar to that of ebselen, is an inhibitory molecule in ferroptosis. Thus, it is quite possible that ebselen may inhibit lung fibrosis by reducing and scavenging lipid peroxides other than oxidized DAG. Recently, it has also been found that lipid mediators related to the arachidonic acid cascade are involved in the regulation of lung fibrosis, and the effect of ebselen on these oxidants is also interesting. In the future, it will be necessary to explore other target molecules for ebselen besides oxidized DAG.

Conclusion

In this study, it was demonstrated that oxidized DAG and PKCδ were involved in the development of lung fibrosis induced by the administration of BLM. In addition, it was shown that ebselen inhibited the same pathway. This observation may lead to the development of new therapies for pulmonary fibrosis in the future. It has been reported that oxidative stress is involved in carcinogenesis. Hence, the findings of this study suggest that antioxidant drugs are useful in the treatment of pulmonary fibrosis, cancer, and other diseases.

Acknowledgments

We thank the staff of the Support Center for Medical Research and Education of Tokai University for technical assistance.

Disclosure statement

The authors declare no conflict of interest.

Data availability statement

Data supporting the findings of this study are available from the corresponding author (ST) upon reasonable request.

Additional information

Funding

This work was supported by the Japan Society for the Promotion of Science, a Grant-Aid for C (No.16K08721).