Recombinant human transforming growth factor-β1 (TGF-β1) was purchased from Peprotech (United States). Antibodies against collagen Ⅰ, fibronectin, P-Smad3/Smad3, P-Akt/Akt, P-ERK/ERK, P-JNK/JNK, P-P38/P38, GAPDH, E-cadherin, vimentin, and N-cadherin were purchased from Cell Signaling Technology (United States). The β-tubulin antibody was obtained from Proteintech (China). The α-SMA antibody was obtained from Santa Cruz Biotechnology (China). Goat pAbs against Rb IgG (HRP) and Rb pAbs against Ms IgG were obtained from ImmunoWay (China). TRIzol reagent was acquired from Ambion Life Technology (China). DEPC-treated H₂O and RNase Away H₂O were purchased from Life Technologies. Fastking gDNA Dispelling RT SuperMix was purchased from Tiangen (Beijing, China). UNICON® qPCR SYBR Green Master Mix and liposomal transfection reagents were purchased from Yevsen (Shanghai, China). The Masson’s trichrome staining kit was obtained from Solarbio (China). Recombinant Mouse Platelet-derived Growth Factor-BB were purchased from MedChemExpress (China).

Human pulmonary epithelial cells (A549, ATCC) were maintained in RPMI-1640 (KeyGEN BioTECH, China) with 10% fetal bovine serum (FBS, ExCell Bio, China). Mouse embryonic fibroblasts (NIH-3T3, ATCC) were maintained in DMEM (KeyGEN BioTECH, China) with 10% FBS. NIH-3T3 cells were stably transfected with the (CAGA)₁₂-Lux reporter to generate CAGA-NIH-3T3 reporter cells. All cells were incubated with 5% CO₂ at 37°C.

Male C57BL/6 mice (6–8 weeks old, 20–25 g) were purchased from the Weitong Lihua Experimental Animal Technology Co. Ltd. (Beijing, China). The animal experimental protocol was performed in accordance with the Health Guide for Care from the National Institutes (NIH Publications No. 85–23, revised 1996). All animal care and experimental procedures complied with guidelines approved by the Institutional Animal Care and Use Committee (IACUC) of Nankai University (Permit No. SYXK 2019-0001).

A mouse pulmonary fibrosis model was established by intratracheal BLM administration as described previously (Guo et al., 2019). Briefly, mice were intratracheally administered BLM (2 U/ Kg) dissolved in physiological saline (0.9% NaCl). Forty mice were randomly divided into five groups: the NaCl group, the BLM group, the positive control group (BLM + nintedanib, 100 mg kg⁻¹), the low-dose remdesivir group (BLM + remdesivir, 10 mg kg⁻¹) and the high-dose remdesivir group (BLM + remdesivir, 20 mg kg⁻¹). The mice in the remdesivir groups were intraperitoneally injected daily with 10 mg kg⁻¹ or 20 mg kg⁻¹ remdesivir, which was suspended in 5% DMSO, 10% PEG and 2.5% polyoxyl 40-hydrogenated castor oil, while mice in the control and BLM groups received equal volumes of vehicle. Mice in the positive control group were intragastrically administered 100 mg kg⁻¹ nintedanib daily. The mice were sacrificed on the 14th day after BLM administration for subsequent analysis, including hydroxyproline content determination.

Primary pulmonary fibroblasts (PPF) were isolated from C57BL/6 J mice. Briefly, the lungs were lavaged three times with 1 ml of PBS, digested with 2.5 mg/ ml dispase II (Roche, United States), and 2.5 mg/ ml Collagenase Type 4 (Worthington, United States) for 30 min at 37°C and then centrifuged. The cell pellet was resuspended in DMEM containing 10% FBS and cultured in 5% CO₂ at 37°C in a humidified atmosphere. PPF cells at passages two to five were used for various assays.

Hydroxyproline was analyzed according to previously described methods (Li et al., 2019).

The left lung was fixed in 10% formalin, dehydrated and embedded in paraffin. Tissue sections with a thickness of 4 μm were incubated for 4 h at 60°C and stained with hematoxylin and eosin (H and E) and Masson’s trichrome. Quantification of pulmonary fibrosis was performed as described previously (Jiang et al., 2004). In brief, Masson staining slides were examined under low power and the whole lung images were captured. Images were analyzed by Image-Pro Plus version 6.0 to demarcate the entire lung area and automatically calculate the total pixel Pw of the region and then calculate the total pixel Pf of the fibrotic region (fibrosis ratio = fibrotic area total pixel Pf/total lung total pixel Pw).

After the mice were anesthetized, the trachea was exposed, and a tracheal cannula was placed and fixed. The mice were transferred to a plethysmography chamber for pulmonary function analysis using an Anires2005 system according to the manufacturer’s instructions (Beijing Biolab, Beijing, China).

Cell viability was measured using 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) as previously described (Li et al., 2019).

Luciferase reporter assays were performed as previously described (Guo et al., 2019).

Wound healing assays were performed as previously described (Li et al., 2019).

The influence of Remdesivir to NIH-3T3 cells migration was evaluated with a Transwell system. (Corning Costar, 3,422, United States) Briefly, serum-starved cells were trypsin-harvested in DMEM with 0.1% FBS. Next, 600 μL of medium containing 20% FBS was added to the lower chambers, while 200 μL NIH-3T3 cells (1 × 10⁵) were plated in the upper chambers. At the same time, NIH-3T3 cells were treated with/without PDGF-bb (10 ng/ ml) and remdesivir (12.5, 25, and 50 μM). After 24 h incubation, the cells on the bottom of the transwell membrane were fixed with 4% paraformaldehyde at 37 °C for 20 min and stained with 1% crystal violet at 37°C for 15 min, the nonmigrating cells in the upper chamber were removed with blunt-end swabs. The membranes were washed three times with PBS and photographed under a microscope.

The mRNA expression level of genes was determined by qRT-PCR using primers according to a previously described protocol (Li et al., 2020). Primer pairs of target genes used were as follows:

Mouse β-actin (NM_007393.3), 5′-AGG​CCA​ACC​GTG​AAA​AGA​TG-3′ and 5′-AGAGCA TAG​CCC​TCG​TAG​ATG​G-3′; Human GAPDH (NM_001256799), 5′-GGA​GCG​AGA​TCC​CTC​CAA​AAT-3′ and 5′-GGC​TGT​TGT​CAT​ACT​TCT​CAT​GG-3′; Mouse α-SMA (NM_001297715.1), 5′-TGGGTGAACTCCATCGCTGTA-3′and 5′-GTCGAA TGCAACAAGGAAGCC-3′; Mouse Collagen Ⅰ (NM_000088.4), 5′-AAGCCGGAGGACAACCTTTTA-3′and 5′-GCG​AAG​AGA​ATG​ACC​AGA​TCC-3′; Mouse Fibronectin (NM_001306132.1), 5′-GTGCCCGGAATACGCATGTA-3′and 5′-CTG​GTG​GAC​GGG​ATC​ATC​CT-3′; Human E-cadherin (NM_0,01317185.2), 5′-A TTT​TTC​CCT​CGA​CAC​CCG​A T-3′and 5′- TCC​CAG​GCG​TAG​ACC​AAG​A-3′; Human Vimentin (NM_003380.5), 5′-AGT​CCA​CTG​AGT​ACC​GGA​GAC-3′ and 5′-CA TTTCACGCA TCTGGCGTTC-3′; Human N-cadherin (NM_001792.5), 5′-TTTGA TGG​AGG​TCT​CCT​AAC​ACC-3′ and 5′- ACG​TTT​AAC​ACG​TTG​GAA​A TGTG-3′; Mouse P16 (NM_058197), 5′-GGC​AGG​TTC​TTG​GTC​ACT​GT-3′ and 5′-TGT​TCA​CGA​AAG​CCA​GAG​CG-3′; Mouse P21 (NM_001111099), 5′-CCTGGTGATGTCCGACCTG-3′ and 5′-CCATGAGCGCATCGCAATC-3′.

All proteins were collected from cells or lung tissues as reported previously (Ning et al., 2004). The monoclonal antibodies used in this study were: α-SMA (Affinity, BF9212), Collagen Ⅰ (Affinity, AF7001), Fibronectin (Affinity, AF5335), β-tubulin (Affinity, T0023), GAPDH (Affinity, AF7021), Phospho-Smad3 (Affinity, AF3362), Smad3 (Affinity AF6323), Smad2 (Affinity, AF6449), Phospho-Smad2 (Affinity, AF3450 Phospho-p38 (Affinity, AF4001), p38 (Affinity, BF8015), Phospho-JNK (Affinity, AF3318), JNK (Affinity, AF6318), Phospho-ERK (Affinity, AF1015), ERK (Affinity, AF0155), Phospho-AKT (Affinity, AF0832), AKT (Affinity, AF6212), Cleaved PARP (Affinity, AF7023), E-cadherin (Cell Signaling Technology, 14472S), N-cadherin (Cell Signaling Technology, 13,116), Vimentin (Cell Signaling Technology, 5,741). The monoclonal antibodies are diluted 1:1,000 with skimmed milk powder.

NIH-3T3 or A549 cells were seeded in a 24-well chamber. At the end of the treatment, the cells were fixed with a 4% fixative solution for 15 min (Solarbio), permeabilized with 0.2% Triton X-100 for 20 min, and incubated in 5% BSA for 30 min. NIH-3T3 cells were cultured overnight with α-SMA primary antibodies (Affinity, BF9212), and A549 cells were probed with E-cadherin (Cell Signaling Technology, 14472S) and vimentin antibodies (Cell Signaling Technology, 5,741) diluted in 5% BSA, the monoclonal antibodies are diluted 1:200 with skimmed milk powder. Then, the cells were incubated with TRITC-conjugated or FITC-conjugated secondary antibodies. The cells were washed with PBST, and DAPI (Beyotime Biotechnology) was used to stain the nuclei. The fluorescence was examined with a confocal microscope (Nikon, Japan).

Paraffin-embedded lung tissue was dewaxed with xylene, and the sections were heated in a microwave oven with antigen-fixing solution (0.01 M citrate buffer) for 20 min. After the sections were cooled to room temperature and blocked with an immunohistochemistry kit, the primary antibody was added and incubated at 4 °C overnight. The primary antibodies were as follows: mouse anti-α-SMA (1:200 dilution, Affinity, BF9212), mouse anti-fibronectin (1:200 dilution, Affinity, AF5335). After being washed with TBST three times, the tissue sections were incubated with the secondary antibody at room temperature for 1 h. Subsequently, the tissue sections were rehydrated through a series of ethanol solutions and were stained with hematoxylin to observe histological changes and target gene expression under an optical microscope.

NIH-3T3 cells were seeded in 6-well plates at a density of 1 × 10⁶cells/well and cultured at 37°C in a humidified 5% CO₂ atmosphere until the cell confluency reached 60–70%. After treated with remdesivir at the desired concentrations (12.5, 25, and 50 μM), NIH-3T3 cells were washed with precooled phosphate buffered saline (PBS) twice and then fixed in 70% ethanol overnight at 4°C. The fixed cells which were washed twice with pre-cooled PBS after fixation were then stained with Propidium (PI)/Rnase A and incubated at 37°C in the dark for 30 min. The DNA content of NIH-3T3 cells were assessed by using a flow cytometer (Thermo Fisher Scientific). FlowJo software (Version 7.6.1) was used to analyze the data.

The data are presented as the means ± SD and were analyzed using Prism version 7.0 software. Differences between the experimental and control groups were assessed by Student’s t tests. Significant differences among multiple groups were analyzed by one-way ANOVA. A value of p < 0.05 was considered statistically significant.

To determine the role of remdesivir in pulmonary fibrosis in vivo, C57BL/6 J mice were administered BLM (2 U), and then different doses of remdesivir (10 and 20 mg kg⁻¹) were administered for 14 days (day 1-day 14) (Figure 1A). Nintedanib (100 mg kg⁻¹) was used as a positive control. We found that the weight ratio, hydroxyproline level and percentage of fibrotic areas were significantly decreased in the remdesivir-treated group and that the inhibitory effect of remdesivir was better than that of nintedanib, while the survival rate of remdesivir was comparable to nintedanib (Figures 1B–E). Moreover, H and E and Masson’s trichrome staining were performed to assess the degree of pulmonary fibrosis. As shown in Figure 1F, BLM-induced collagen deposition was markedly reduced by remdesivir treatment. Lung function is a decisive factor in the diagnosis and treatment of pulmonary fibrosis in clinical trials. As shown in Figure 2, remdesivir significantly improved lung function in BLM-induced fibrotic mice, as seen by the increases in forced vital capacity (FVC) and dynamic compliance (Cdyn) and decreases in inspiratory resistance (Ri) and expiratory resistance (Re), and the effects were superior to those of the positive control drug nintedanib. These data revealed that remdesivir could ameliorate BLM-induced pulmonary fibrosis in mice.

The proliferation and migration of activated fibroblasts contribute to the development of lung fibrosis (Coward et al., 2010). First, we examined whether remdesivir affected the proliferation of normal fibroblasts using the MTT assay. We found that remdesivir had no obvious toxicity on normal fibroblasts, and the IC₅₀ value of remdesivir in NIH-3T3 cells was 60.32 μM (Figure 3A). We also examined the effect of remdesivir on the cell cycle and senescence of normal fibroblasts in absence of TGF-β1 stimulation and the results showed that after treated with different doses of remdesevir, the fibroblasts didn’t arrest and becoming senescent (Supplementary Figure S1). To further investigate the effect of remdesivir on cell proliferation of active fibroblasts, we co-treated NIH3T3 cells with TGF-β1 (5 ng ml⁻¹) and remdesivir (12.5, 25, and 50 μM), then used immunofluorescence and western blot to detect the expression of cell proliferation marker Ki67 and cell apoptosis maker cleaved PARP. As shown in Figures 3B–D, remdesivir could dose-dependently inhibit the proliferation and apoptosis of TGF-β1-activated fibroblasts.

Then, we used wound healing assay and transwell assay to test the effects of remdesivir on the migration of activated fibroblasts. NIH-3T3 cells were co-treated with remdesivir (12.5, 25, and 50 μM) and TGF-β1 (5 ng ml⁻¹) for 6, 12, 24, and 48 h. As shown in Figures 4A–C, remdesivir dose-dependently inhibited the migration of TGF-β1-activated fibroblasts. The transwell assay also indicated that remdesivir could dose-dependently inhibit PDGF-induced migration of fibroblasts (Figure 4D).

To further examine whether remdesivir could inhibit the activation of TGF-β1-induced fibroblasts, we treated NIH-3T3 cells and PPF cell with TGF-β1 (5 ng ml⁻¹) and remdesivir (12.5, 25, and 50 μM) to assess the gene expression levels of the typical fibroblast activation marker α-SMA and the extracellular matrix (ECM) proteins collagen I and fibronectin. As shown in Figure 5A, remdesivir dose-dependently decreased TGF-β1-induced expression of the α-SMA, collagen I and fibronectin genes. Similarly, remdesivir treatment reduced the protein expression levels of α-SMA, and fibronectin in TGF-β1-stimulated fibroblasts, including NIH3T3 cell and PPF cell (Figures 5B,C). Immunofluorescence assays were used to evaluate α-SMA expression in NIH-3T3 cells, and the results were consistent (Figures 5D,E). Accordingly, these results indicated that remdesivir suppressed TGF-β1-induced myofibroblast differentiation and reduced ECM production in myofibroblasts.

The TGF-β1 signaling pathway is critical in regulating the proliferation and activation of fibroblasts in pulmonary fibrosis (Aschner and Downey, 2016). Therefore, we examined whether remdesivir could regulate TGF-β1/Smad and non-Smad signaling in active pulmonary fibroblasts. We stably transfected the (CAGA)₁₂-luciferase reporter, which contain repeated canonical Smad3-binding element (SBE) 5′-CAGA-3′ in NIH-3T3 fibroblasts. CAGA-NIH-3T3 cells were used to examine the inhibitory effects of compounds on the TGF-β/Smad3 signaling pathway. After the administration of 5 ng ml⁻¹ TGF-β1, the fluorescence value was obviously increased in CAGA-NIH-3T3 cells, while remdesivir inhibited fluorescence in a dose-dependent manner (Figure 6A). Next we used the western blot assay to detect the effect of remdesivir on the activation of TGF-β1/Smad and non-Smad signaling in fibroblasts. We first examined the effect of TGF-β1 on the protein phosphorylation level of Smad and non-Smad under the time gradient, and the results revealed that the phosphorylation level of the protein was positively correlated with the time of adding TGF-β1 and the phosphorylation levels were more significant after 1 h of TGF-β stimulation (Supplementary Figure S2). Then we assessed the effect of remdesivir on the activation of Smad3 signaling in NIH-3T3 cells and PPF cells after 1 h treatment of TGF-β1. As shown in Figures 6B,C, remdesivir significantly reduced the phosphorylation level of Smad3 and had no significant effect on the protein levels of total Smad3. Moreover, the levels of phosphorylated P38, JNK, ERK and Akt were examined in NIH-3T3 cells by western blotting. The results revealed that remdesivir restrained the activation of TGF-β1/Smad and non-Smad signaling (Figure 6D).

Evidence suggests that lung fibrosis is a dysregulated epithelial-mesenchymal disorder in which epithelial injury is an initiating event (Selman and Pardo, 2002). The hallmark features of injured epithelial cells are the decreased expression of epithelial markers, including E-cadherin, and the increased expression of mesenchymal markers, such as N-cadherin and vimentin (King et al., 2011; Kage and Borok, 2012). First, we examined whether remdesivir affected the proliferation of A549 cells using the MTT assay. We found that remdesivir had no obvious toxicity on A549 cell, and the IC₅₀ value of remdesivir in A549 cells was 118.2 μM (Figure 7A). To explore whether remdesivir exerted antifibrotic effects by protecting against epithelial to mesenchymal transition, we used stimulated human pulmonary epithelial cells (A549 cells) with TGF-β1 and then treated the cells with remdesivir (12.5, 25, and 500 μM). After 24 h, the morphological changes and expression of epithelial/mesenchymal phenotypic markers in A549 cells were examined. The results indicated that remdesivir could effectively reverse the morphological changes in TGF-β1-induced A549 cells (Figure 7B). The qRT-PCR and western blot results indicated that remdesivir could significantly increase the expression of the epithelial marker E-cadherin and decrease the expression of the mesenchymal markers N-cadherin and vimentin in A549 cells (Figure 7C; Figure 8A). Moreover, the immunofluorescence staining results were consistent with the above results (Figures 8B,C). Collectively, these data indicate that remdesivir inhibits TGF-β1-induced phenotypic transition in alveolar epithelial cells.

To further verify whether remdesivir could reduce the activation of lung fibroblasts in vivo, we extracted protein and RNA from lung tissues and analyzed them by western blotting and qRT-PCR. The results confirmed that the protein and RNA expression levels of α-SMA and collagen I in the lung tissues of the remdesivir group were significantly lower than those in the BLM model group and better than those of the positive control group (Figures 9A,B). We also performed immunohistochemical analysis of lung tissue sections, and the expression levels of α-SMA and fibronectin in the lung tissues of remdesivir-treated mice were lower than those in the BLM model group (Figure 9C). Epithelial-mesenchymal transition markers were evaluated, and similar to our in vitro results, western blotting showed that E-cadherin expression was increased and N-cadherin expression was decreased by remdesivir treatment (Figure 9D). Then we further verified whether remdesivir could influence TGF-β1 signaling pathway in vivo, the western blotting results indicated that remdesivir significantly reduced the activation of Smad2/3 and MAPK signaling in the lung tissues of bleomycin-injured mice (Figure 10). In conclusion, remdesivir attenuated BLM-induced fibroblast activation and epithelial to mesenchymal transition in vivo.