PM014 attenuates radiation-induced pulmonary fibrosis via regulating NF-kB and TGF-b1/NOX4 pathways

Radiation therapy is the mainstay in the treatment of lung cancer, and lung fibrosis is a radiotherapy-related major side effect that can seriously reduce patient’s quality of life. Nevertheless, effective strategies for protecting against radiation therapy-induced fibrosis have not been developed. Hence, we investigated the radioprotective effects and the underlying mechanism of the standardized herbal extract PM014 on radiation-induced lung fibrosis. Ablative radiation dose of 75 Gy was focally delivered to the left lung of mice. We evaluated the effects of PM014 on radiation-induced lung fibrosis in vivo and in an in vitro model. Lung volume and functional changes were evaluated using the micro-CT and flexiVent system. Fibrosis-related molecules were evaluated by immunohistochemistry, western blot, and real-time PCR. A orthotopic lung tumour mouse model was established using LLC1 cells. Irradiated mice treated with PM014 showed a significant improvement in collagen deposition, normal lung volume, and functional lung parameters, and these therapeutic effects were better than those of amifostine. PM104 attenuated radiation-induced increases in NF-κB activity and inhibited radiation-induced p65 translocation, ROS production, DNA damage, and epithelial-mesenchymal transition. PM104 effectively alleviated fibrosis in an irradiated orthotopic mouse lung tumour model while not attenuating the efficacy of the radiation therapy by reduction of the tumour. Standardized herbal extract PM014 may be a potential therapeutic agent that is able to increase the efficacy of radiotherapy by alleviating radiation-induced lung fibrosis.


Results
Effect of PM014 on gross anatomical and histopathological damage. We previously identified time dose response for fibrosis after high dose focal radiotherapy to the mouse lung and as a result, considered a 6th week model suitable for studying IR-induced lung fibrosis ( Supplementary Fig. S1). To elucidate whether PM014 inhibited IR-mediated lung fibrosis, we compared changes in the surface morphology of the left lung in the control and IR groups at 6 weeks after irradiation with 75 Gy. In contrast to the control mice, the irradiated areas of the left lung clearly exhibited a local, white, ring-like injury (Fig. 1A), infiltration of inflammatory cells, and alveolar wall and bronchiolar epithelium thickening ( Fig. 1B-E). To confirm fibrosis, Masson's Trichrome staining was performed. At 6 weeks, intense blue stained collagen fibers were observed in the IR group (Fig. 1F,G). Oral administration of PM014 (200 mg/kg) on alternate days for 6 weeks after irradiation resulted in significantly less gross anatomical changes, histological damage, and fibrosis compared to the IR group. These protective effects of PM014 were better than those observed for amifostine (AMI, 100 mg/kg, i.p.).

Micro-computed tomography and lung functional analysis.
For image evaluation of IR-induced lung fibrosis, we used a non-invasive micro-computed tomography (CT), equivalent to clinical CT in humans 15 . Six weeks after irradiation, typical micro-CT manifestations of IR-induced lung injury 16 were observed in the irradiated left lung (lighter grey). However, these effects were decreased in PM014-treated mice (Fig. 1H,I). The normal lung volume in the IR group was lower than that in control mice. However, in the IR + PM014 group, the volume was significantly recovered compared to that in the IR group (P < 0.05). PM014 was more effective than AMI (100 mg/kg) in maintaining normal lung volume. Radiation-induced changes in lung function were evaluated using the flexiVent system 17 . Compared to those in the control group, both inspiratory capacity (IC) and quasistatic compliance (Cst) in the IR group were significantly decreased, reflecting a reduced total capacity and stiffness of the lung. Both tissue damping (G) and tissue elastaince (H), parameters assessing lung tissue rigidity, were significantly increased in the IR group, suggesting lung parenchymal injury, and there was an increase in Newton resistance (Rn) in the IR group, indicating airway hyper-responsiveness (Fig. 1J). However, this IR-induced respiratory distress was significantly reduced in PM014-treated mice, suggesting that PM014 has a protective effect on IR-induced deterioration in lung function. These effects of PM014 were also more prominent than those observed for AMI (100 mg/kg).

Inhibition of radiation-induced fibrosis-related molecules by PM014.
To investigate the molecules responsible for radiation-induced lung fibrosis, a cDNA microarray analysis of lung tissues after focal exposure to 75 Gy was performed 18 . Of these genes, we focused on transforming growth factor-β (TGF-β), IL-6, and Twist (Supplementary Table S1), which are key molecules involved in fibrosis progression and stimulate collagen synthesis in fibroblasts and myofibroblasts 19,20 . Immunohistochemistry and RT-PCR revealed higher TGF-β1, IL-6, and Twist expression levels in irradiated lungs than in controls, which were attenuated by PM014 ( Fig. 2A-D). We also investigated whether NF-κB is also regulated by IR and PM014 using an NF-κB promoter driving luciferase expression. Luciferase activity was increased 9.57-fold in irradiated cells compared to that in control cells (Fig. 2E); however, it was suppressed by 33.1% and 41% in PM014-treated cells (5 μg/mL and 10 μg/ www.nature.com/scientificreports/ mL PM014, respectively). Moreover, nuclear translocation of p65 was increased by IR, and this was inhibited by PM014 treatment (Fig. 2F,G). These results suggest that NF-κB is activated by IR to increase the expression of downstream pro-fibrotic effectors, resulting in fibrosis.
Inhibition of radiation-mediated epithelial-mesenchymal transition (EMT) features by PM014. During fibrosis, EMT allows epithelial cells to undergo morphological changes 21 . We hypothesized that PM014 affects the EMT-like process in epithelial cells caused by radiation-induced fibrosis. To explore this hypothesis, we first examined morphological changes in L132 human lung epithelial cells after IR exposure. Following exposure to IR, the cells underwent a morphological transformation adopting a spindle-like shape. However, PM014 treatment attenuated these IR-mediated morphological changes (Fig. 3A). Immunocytochemical analysis confirmed that IR decreased the expression of the epithelial marker E-cadherin, and increased that of the mesenchymal marker α-SMA. However, PM014 restored the altered expression of these EMT-related molecules induced by IR (Fig. 3B). Western blot and RT-PCR analysis of E-cadherin and α-SMA expression also showed a similar pattern to that observed in the immunocytochemical analysis ( Fig. 3C,D). Cell migration was significantly increased by IR, whereas in PM014-treated cells, IR-induced cell migration was greatly reduced (Fig. 3E). We also measured the invasiveness of irradiated L132 cells, which was increased 5.7-fold compared to that of control cells. However, this increase in invasiveness was attenuated by PM014 treatment (3.9-fold compared to control). These results suggest that irradiated cells undergo EMT and that PM014 attenuates the acquisition of these IR-induced EMT features.
PM014 reduces radiation-induced oxidative stress and DNA damage. We assessed oxidative stress using immunohistochemistry for 8-OHdG and NOX4 in irradiated lung tissue. 8-OHdG and NOX4 levels were increased in the IR group, and these levels were significantly decreased by PM014 treatment (Fig. 4A), suggesting that PM014 has an antioxidant capacity. Due to the potential antioxidant capacity of PM014, we hypothesized that PM014 may have ROS scavenging activity as part of its radioprotective function. To verify this, we performed a DPPH radical scavenging assay. PM014 exhibited radical scavenging activity in a relatively dose-dependent manner ( Supplementary Fig. S2). ROS formation was found to be markedly elevated in irradiated L-132 cells, whereas, in PM014-treated cells, there was a noticeable decrease in ROS formation (Fig. 4B).
To investigate the effect of PM014 on IR-induced DNA damage, we measured H2AX phosphorylation (γH2AX) status by western blot and immunocytochemistry. This IR-induced stimulation of H2AX phosphorylation was reduced by PM014 (Fig. 4C). IR treatment resulted in a significant increase in γH2AX positive nuclei (Supplementary Fig. S3). The IR-induced phosphorylation of ATM and ATR was also decreased by PM014 (Fig. 4D, Supplementary Fig. S4). Irradiated cells had a significantly longer comet tail than PM014-treated cells (Fig. 4E). These findings suggest that radiation-induced DNA damage is prevented by PM014 treatment. www.nature.com/scientificreports/

PM014 attenuates IR-induced apoptosis in alveolar epithelial cells.
To confirm the induction of apoptosis by IR, we performed an in situ terminal deoxynucleotidyl transferase dUTP nick-end labeling (TUNEL) assay using irradiated lung tissue. Radiation increased the number of apoptotic nuclei in lung tissue and this degree of apoptosis was decreased by PM104 treatment (Fig. 5A, Supplementary Fig. S5). The pro-apoptotic protein Noxa is reportedly involved in radiation-induced lung fibrosis 22 . Immunohistochemistry and western blotting showed higher expression levels of Noxa in irradiated lungs than in control lungs, and that PM014 attenuated these IR-induced increases in Noxa expression (Fig. 5B,C). Noxa levels were increased and co-localized with pro-SP-C in epithelial cells from the irradiated lung tissue, which was reduced by PM014 ( Fig. 5D, Supplementary Fig. S6). These results suggest that PM014 reduces the expression of Noxa and attenuates epithelial cell death, resulting in inhibition of fibrosis.

PM014 suppresses fibrosis in orthotopic lung tumours in combination with radiation.
To elucidate the effect of using PM014 in combination with radiotherapy of lung tumours, we established an orthotopic lung tumour model after IV injection with LLC1 cells. The left whole lung of the orthotopic tumour model was irradiated with 75 Gy. However, this 75 Gy irradiation led to fibrosis after 4 weeks, and the size of the orthotopic lung tumour was so large that the mice died. Therefore, the radiation dose was increased to 90 Gy, and a 2-week model was also chosen for the study ( Supplementary Fig. S7). The experimental schedule is shown in Fig. 6A.
In the orthotopic lung tumour model, there were 40-50 visible tumour nodules per lung. After irradiation, the number of nodules was dramatically decreased compared to that in IV only mice. Interestingly, tumour regression was greater in mice irradiated with 90 Gy and treated with PM014 (Fig. 6B,C), with the result being almost similar to that for control mice. Fibrosis was significantly increased in irradiated normal lung lesions, which was www.nature.com/scientificreports/ greatly reduced in mice treated with IR and PM014 (Fig. 6B,D). A colony formation assay was performed to confirm whether PM014 synergizes with IR. PM014 caused the lung cancer cells to become more sensitive to radiation (Fig. 6E). In this study, we treated A549 lung cancer cells cultured under sphere forming conditions with PM014 (0, 5, and 10 μg/mL) to study the effects of PM014 on lung cancer stem-like cells. We found that PM014 suppressed sphere formation in this lung cancer cell line ( Supplementary Fig. S8), suggesting that PM014 may have the potential to suppress cancer stem-like cells. However, the effect of PM014 on cancer stem cells only suggests the possibility and further research is needed. Taken together, these results suggest that PM014 treatment in combination with radiotherapy inhibits the fibrosis of normal tissue without interfering with tumour removal and is a promising therapeutic strategy for controlling IR-induced lung fibrosis.

Discussion
To study the radiation-induced damage of normal tissues adjacent to tumours that underwent radiotherapy, we previously established an in vivo mouse model that mimicked clinical SBRT 23,24 . Using this model, inflammatory responses of the irradiated lung were studied at 2 weeks after irradiation with 75 Gy 8 , and fibrosis could be clearly observed at 6 weeks. PM014 comprises seven major herbal components found in Chung-Sang-Bo-Ha-Tang, a well-known herbal mixture for treating pulmonary diseases in traditional Korean medicine 8,25 . The optimal dosing and timing of PM014 administration to treat radiation-induced lung injury have been previously assessed in our laboratory, with the optimal conditions being 200 mg/kg, 3 times a week 8 . Six weeks after irradiation, an excessive deposition of collagen was markedly attenuated in the mice treated with PM014, suggesting that PM014 has antifibrotic potential (Fig. 1f,g). These results are consistent with the reduction of collagen by PM014 in bleomycin induced pulmonary fibrosis model 26 . Immunohistochemistry to assess 8-OHdG and NOX4 levels suggested that PM014 has antioxidant properties. It is reported that expression of nox4 is increased by irradiation and blocking of nox4 is attenuated irradiation induced fibroblast differentiation 27 . Our results have shown a significant decrease in the expression of NOX4 by PM014. Thus, PM014 is an effective drug for reducing irradiated mediated fibroblast differentiation. In our previous study, we identified a new molecular mechanism underlying IR-induced fibrosis by demonstrating that Noxa-induced apoptosis plays an important role in fibrosis 22 . As shown in Fig. 5, PM014  www.nature.com/scientificreports/ attenuated IR-mediated apoptosis and Noxa upregulation in AECs. These results suggest that PM014 reduces IR-induced fibrosis by inhibiting Noxa expression and apoptosis in AECs. IR-induced apoptosis mainly occurs through DNA damage induction 28 . To investigate the role of PM014 in IR-induced DNA damage, we assessed γH2AX levels and performed a comet assay 29 . Following irradiation, the levels of γH2AX and the length of the comet tail were increased in irradiated cells, but were, importantly, decreased in PM014 treated cells, suggesting that radiation injury can be exacerbated by DNA damage of lung epithelial cells and that PM014 protects these The pro-apoptotic protein Noxa (green) was co-stained with pro-SP-C (red), a marker for type II AECs, using irradiated mouse lung tissue. Magnification × 400 (Scale bar, 50 μm).  www.nature.com/scientificreports/ cells from radiation-induced DNA damage, thereby inhibiting fibrosis. However, whether PM014 regulates DNA repair is unclear; thus, further studies are needed to address whether PM014 is related to DNA repair and fibrosis. IR is known to activate the NF-κB pathway 30 , and IL-6 and Twist are downstream molecules in the NF-κB pathway. Therefore, PM014 may play a role in suppressing the NF-κB pathway including the IL-6 and Twist genes, which are some of the major regulators of IR-mediated EMT progression. Our results showed that PM014 attenuated the activation of the NF-κB promoter and the nuclear translocationt of NF-κB induced by IR (Fig. 3E-G). These results suggest that inhibition of NF-κB activation by PM014 can lead to reduced expression of downstream molecules, thereby inhibiting IR-mediated EMT progression. Our results showed that PM014 reduced the expression of TGF-β1. The transforming growth factor β1 (TGF-β1) is reported to be an important cytokine for the process of fibrosis through the interaction between growth factors and cytokines [31][32][33][34][35][36] . After irradiation, TGF-β gene expression increases, and TGF-β strongly promotes collagen synthesis and expresses collagen and fibronectin synthesis genes in fibroblasts [37][38][39][40] . Several studies have been reported to inhibit lung damage caused by radiation using a small molecule that inhibits TGF-β 41,42 . Also, TGFβ1 Stimulates Connective Tissue Growth Factor (CTGF) Expression 43 . CTGF is known as the central mediator of tissue remodeling and fibrosis and is being studied as a target that can inhibit the fibrosis process 44 . PM014 is likely to act on various targets related to fibrosis by inhibiting TGF-β.
The results of the micro-CT and flexiVent analyses in the present study correlated with the histopathological findings, suggesting a critical role for PM014 in lung fibrosis and that PM014 may be an effective therapeutic agent for inhibiting IR-induced lung fibrosis.
In the orthotopic lung cancer mouse model, most of the tumour was removed by IR compared with the large tumour nodules observed in IV only mice; however, fibrosis was observed in the irradiated area. PM014 treatment in IR-treated mice resulted in a more effective regression of the tumour and an attenuation of fibrosis. These results show that PM014 has a beneficial effect on radiation-induced fibrosis of normal tissue in an irradiated orthotopic lung cancer model. One important factor that needs to be taken into consideration in developing a drug that inhibits the fibrosis that occurs after tumour radiotherapy is whether the drug interferes with the tumour removal by irradiation, and whether it has any effect on tumour recurrence. We hypothesise that PM014 can be effectively used in radiotherapy by attenuating radiation induced fibrosis while not attenuating the efficacy of the radiation therapy.
In conclusion, our results collectively show that PM014 may be effectively and safely used as a therapeutic drug for use in combination with radiotherapy to treat lung cancer and could also act as a radioprotector for lung tissue by attenuating pneumonitis and fibrosis.

Methods
Micro-computed tomographic analysis. Micro-computed tomography (CT) images were acquired using a volumetric CT scanner (NFRPolaris-G90MVC: NanoFocusRay, Iksan, South Korea) at 50 kVp, 180 µA, and 150 mGy (number of views, 700; frame rate, 142 ms) 8 . Images were reconstructed (image size, 1232 × 1120 pixels; number of slices, 512) by volumetric cone-beam reconstruction (Feldkamp-Davis-Kress method) in inline/off-line modes. Volumetric analysis was performed using the Image J software. In order to minimise interspecimen variations in measurement, identical level settings were used for analysis of all images.
Functional assessment of the lungs. Lung function in irradiated mice was evaluated with the flexiVent system (flexiVent; SCIREQ, Montreal, QC, Canada), which measures flow-volume relationships in the respiratory system, including forced oscillation, to discriminate between airway and lung tissue variables 17 . Evaluations were performed according to the manufacturer's instructions. Briefly, after anesthetisation, mice were connected to a computer-controlled small-animal ventilator and quasi-sinusoidally ventilated with a tidal volume of 10 mL/ kg at a frequency of 150 breaths/minute. Measurement commenced when a stable ventilation pattern without obvious spontaneous ventilator effort was observed at the ventilation pressure tracing. All perturbations were performed sequentially until three acceptable measurements (coefficient of determination > 0.95) were recorded for each subject, from which an average was calculated. www.nature.com/scientificreports/ USA) 4715, Abcam) and were developed using 3,3′-diaminobenzidine tetrachloride (DAB; Zymed Laboratories, CA, USA).

RT-PCR
Histology and immunohistochemistry evaluation. Slides were assessed according to a dual-rate semi-quantitative method by three independent pathologists, who were blinded to sample identities 46 . For histological evaluation, lung tissue sections were stained with H&E and MT staining and scored for the number of inflammation or fibrotic foci, respectively. For IHC evaluation, lung tissue sections were stained with TGF-β1 staining. Randomly selected fields of each slide were scored for area and intensity of positively stained (brown) cytoplasm and cell membrane. Intensity scores were assigned as follows: 0 = no appreciable staining (negative); 1 = barely detectable staining (weak); 2 = readily appreciable brown staining (moderate); and 3 = dark brown staining (strong positivity). The total score was calculated by adding the intensity scores from five independent views in each sample, resulting in a final score of 0 to 15. For statistical analysis, scores 3-15 and 0-2 were defined as indicating positive and negative expression, respectively.
Immunocytochemistry. Cells were cultured on coverslips coated with poly-L -lysine, fixed with 4% paraformaldehyde in PBS and permeabilised with 0.1% Triton X-100 22 . The cells were incubated with primary antibodies at 4 °C overnight, and then stained with secondary antibodies. These cells were viewed by confocal microscopy (LSM 700, Zeiss, Jena, Germany). Colony formation assay. A total of 500 H460 irradiated 2 Gy were seeded into a 60 mm dish and cultured in RPMI supplemented with 10% FBS for about 2 weeks. Then, the cells were fixed and stained with acetic acid: methanol (1:3) solution for 5 min, followed by staining with 0.5% crystal violet solution. The colonies were quantified using the NIH Image J program.
Invasion assay and wound closure assay. Invasion assays were performed using the Chemicon Cell Invasion Assay Kit (Millipore, Billerica, MA, USA) according to the manufacturer's protocol. Briefly, L132 cells (1 × 10 4 ) were plated onto a Matrigel-coated transwell invasion chamber and incubated at 37 °C for 24 h. Invading cells were then fixed with methanol and stained with haematoxylin. On average, five random fields were counted using a light microscope. For Wound closure assay, when L132 cells were confluent, a wound was created along the middle of the culture plate using a tip. Cells were then irradiated (10 Gy) with or without PM014 (10 μg/mL) for 48 h and cells images obtained using inverted microscopy.
Comet assay. Alkaline comet assays were performed using CometAssay kit (4250-050-K, Trevigen) protocol following the manufacturer's instructions. Cells were irradiated (10 Gy) with or without PM014 (10 μg/mL). After electrophoresis, the cells were stained with diluted SYBR Green and all images were captured by fluorescence microscopy. DNA damage was quantified for at least 50 randomly selected cells by measuring tail length using the NIH Image J program.
Orthotopic mice model. Mouse lung carcinoma LLC1 cells (1 × 10 6 ) in 200 µL of physiological saline were injected into the tail vein of 7-week-old male C57BL/6 mice. Two weeks later, a single dose of 90 Gy was delivered to the left whole lung using an image-guided small-animal irradiator. The mice were then randomly divided into three groups: (1) LLC1 group-i.v. injection only group; (2) LLC1 + IR group-mice were exposed to a single dose of 90 Gy delivered to the left whole lung 2 weeks after i.v. injection; (3) LLC1 + IR + PM014 group-the mice were orally administered PM014 (200 mg/kg) for 2 weeks on every other day after irradiation. On week 4, the mice were sacrificed by CO 2 asphyxiation, and lung tissues were collected for analysis.