Tirbanibulin (KX-01, purity>98%, C₂₆H₂₉N₃O₃, CAS No: 897016-82-9) and the HIF-1α inhibitor Roxadustat (ROT) (purity>98%, C₁₉H₁₆N₂O₅, CAS No: 38808118-40-3) were from Shanghai Hanxiang Biomedical Company (China). The STAT3 inhibitor Stattic (purity>98%, C₈H₅NO₄S, CAS No: 19983-44-9) was from the Jiangsu Aikang Biomedicine Company (China). Antibodies specific for p-STAT3, STAT3, HIF-1α, collagen I, collagen III, and α-SMA were obtained from the Shanghai Abcam Company (China). Secondary HRP-labeled goat anti-rabbit IgG (A0208), HRP-labeled goat anti-mouse IgG (AA128), mouse monoclonal anti-actin (A030-2-1), and C0088L were from Beyotime Biotechnology (China). A Masson’s trichrome staining kit (G1340) was purchased from the Beijing Solebao Technology Company (China). A hydroxyproline (Hyp) detection kit (A030-2-1) was obtained from the Nanjing Jiancheng Bioengineering Institute (Naniing, China). In addition, a TMB-based BeyoClick™ EdU cell proliferation assay kit was purchased from Beyotime Biotechnology.

In total, 24 healthy male Sprague Dawley rats (200–220 g) were housed in a climate-controlled animal facility (22 ± 2°C, 60 ± 10% relative humidity) with free food and water access. Following a 7 days acclimatization period, a model of BLM-induced pulmonary fibrosis was established in these rats (Huang et al., 2020). Briefly, rats were anesthetized and an endotracheal infusion of BLM (6 mg/kg in 0.2 ml normal saline) was administered. Sham control rats received an equivalent volume of normal saline. On day 22 (Histological examination showed obvious fibrosis in the lung) post-BLM administration, the 16 BLM-treated pulmonary fibrosis model rats were randomly separated based on body weight into the BLM model and KX-01 treatment groups. Animals in the KX-01 group received intragastric KX-01 (1.5 mg/kg/day for 14 days; dose determined to decrease the lung coefficient in the pre-experiment), while all other animals received an equivalent volume of sodium carboxymethyl cellulose. On day 36 (In general, 3 weeks after bleomycin induced pulmonary fibrosis in rats, then the corresponding drugs were given 2 weeks, pulmonary fibrosis could be significantly relieved), animals were euthanized and lung tissue samples were collected. Lung coefficient values were calculated as follows: lung coefficient = wet lung weight/body weight × 100%. After weighing, the lung tissue samples were separated into two portions, with the right lung snap-frozen for Western blotting analysis and the left lung fixed in 4% paraformaldehyde (PFA) prior to histological analysis.

The center one-third of the left lung from each rat was fixed for 48 h in 4% PFA, dehydrated, paraffin-embedded, and cut into 4 μm-thick sections (Huang et al., 2020). The sections were then transferred onto poly-lysine-coated slides, deparaffinized using xylene, rehydrated with an ethanol gradient, and stained with hematoxylin and eosin (H&E). The degree of interstitial fibrosis in each section was then scored from 0 to 8 as follows: grade 0, normal lung; grade 1, isolated alveolar septa with subtle fibrotic changes; grade 2, fibrotic changes of alveolar septa with knot-like formation; grade 3, contiguous fibrotic walls of alveolar septa; grade 4, single fibrotic masses; grade 5, confluent fibrotic masses; grade 6, large contiguous fibrotic masses; grade 7, air bubbles; and grade 8, fibrous obliteration. Scoring was done separately in a blinded manner. Masson’s trichrome was used to measure collagen deposition, and the percentage of pulmonary fibrosis was assessed as previously described (Zhang et al., 2019).

Pulmonary α-SMA and p-STAT3 expression was assessed via immunohistochemical staining. Briefly, 4 μm-thick lung tissue sections were deparaffinized, rehydrated, and treated for 10 min with 0.01 M citric acid in a 400 W steam microwave, after which 5% H₂O₂ in methanol was applied to quench endogenous peroxidase activity for 30 min at room temperature while protected from light. Sections were then blocked with 5% normal goat serum, after which they were probed overnight with rabbit polyclonal antibodies specific for α-SMA and p-STAT3 at 4°C, followed by probing for 30 min with HRP-conjugated anti-rabbit IgG at 37°C. Samples were then visualized via light microscopy. In total, five rats per group were analyzed, with three sections per rat being assessed. The Image-Pro Plus software (Media Cybernetics) was used to analyze tissue section images.

Lung tissue samples (30–100 mg) were added to test tubes and 1 ml of hydrolysate was added and thoroughly mixed. Samples were boiled for 20 min in a water bath, after which the Hyp contents in the samples were measured using the Hyp kit, according to the manufacturer’s instructions, with absorbance measured at 550 nm.

Snap-frozen lung tissue sections or cell samples were lyzed using radioimmunoprecipitation (RIPA) buffer containing protease inhibitors, after which supernatant protein concentrations were assessed using a BCA assay. Samples of protein (50 μg) were then separated via 8–10% SDS-PAGE, and the resultant blots were probed with antibodies specific for HIF-1α, p-STAT3, p-Src, α-SMA, collagen I, collagen III, and β-actin. ImageJ software was used to quantify protein band density, and β-actin density was used as a loading control to normalize assay results.

Murine L929 lung fibroblasts from the Chinese Academy of Sciences (Beijing, China) cell bank were cultured in MEM supplemented with 10% newborn calf serum and penicillin/streptomycin in a 37°C, 5% CO₂, 95% N₂ incubator. Cells were initially plated at 1×10⁵ /ml and were passaged every 3–4 days.

L929 cells were plated at 8×10² cells/well in 96-well plates and incubated overnight at 37°C, after which the media were replaced with media containing a range of KX-01 concentrations (0, 0.01, 0.1 μM) supplemented with or without CoCl₂ (100 nM), after which cells were incubated for 72 h. In other analyses exploring the mechanistic basis for observed pulmonary fibrosis-related phenotypes, cells were cultured with CoCl₂ without or without 10 μM ROT (HIF-1α inhibitor) or 1 mM Stattic (STAT3 inhibitor) for 72 h. The BeyoClick™ 5-ethynyl-2′-deoxyuridine (EdU) Cell Proliferation Kit with TMB was used to measure cell proliferation. Absorbance values measured at 630 nm were normalized against those for untreated cells.

L929 cells were grown until 60% confluent after which they were treated with KX-01 (0.01 μM) plus CoCl₂ (100 nM) or were left untreated for 72 h. Alternatively, the mechanistic basis for pulmonary fibrosis was assessed by treating these cells with CoCl₂ (100 nM) for 72 h with or without ROT (10 μM) or Stattic (1 μM). The expression of p-STAT3, p-Src, STAT3, HIF-1α, α-SMA, collagen I, and collagen III in these cells was then assessed via Western blotting.

Statistical analysis was performed using GraphPad Prism 5.0. Lung histopathology scores were compared between groups using the Wilcoxon rank-sum test. One-way ANOVAs with Dunnett’s test were used to compare all other data between groups. Data are given as means ± standard deviation. p < 0.05 was the significance threshold for this study.

We began by evaluating fibrosis in our rat BLM-induced pulmonary fibrosis model system. As expected, sham control rats exhibited minimal alveolar inflammatory cell infiltration (Figure 1A), whereas rats in the BLM model group exhibited significant inflammatory cell infiltration, fibrosis, and alveolar tissue destruction (Figure 1A2). Rats that had been treated with KX-01 showed reduced levels of inflammatory cell infiltration and pulmonary interstitial thickening, with significantly lower pathological scores relative to the model group animals (Figures 1A3,B). We additionally measured lung coefficient values to quantify the degree of pulmonary fibrosis in these rats. During the early stages of pulmonary fibrosis, this ratio rises due to an increase in pulmonary mass attributable to cellular swelling and capillary congestion, whereas during later stages it is primarily due to collagen fiber accumulation. Relative to BLM-treated model group rats, lung coefficient values in KX-01-treated rats were significantly decreased (Figure 1C).

Next, we assessed pulmonary collagen deposition in our experimental model system via Western blotting-mediated measurement of α-SMA, collagen I, and collagen III levels (Figures 2A,B) and Masson’s trichrome staining (Figures 2C,D). We detected substantial interstitial collagen deposition in BLM model rats with Masson’s trichrome staining, whereas collagen levels were significantly reduced in lung samples from KX-01-treated rats (Figure 2D). Hyp accounts for ∼13% of collagen amino acid content and can thus be measured as a means of quantifying collagen levels within a given tissue. Consistent with our staining results, we found that Hyp levels were significantly reduced in KX-01-treated rats relative to the BLM-treated model rats (Figure 2E). Western blotting further confirmed that pulmonary α-SMA, collagen I, and collagen III expression levels were significantly elevated in BLM-treated model rats relative to sham controls, while they were significantly reduced in KX-01-treated rats relative to BLM-treated model rats (Figures 2A,B).

The impact of KX-01 on the pulmonary expression of fibrosis-related proteins.

We next analyzed pulmonary p-Src, p-STAT3, STAT3, and HIF-1α expression profiles by Western blotting. Relative to sham control rats, those in the BLM model group exhibited significant increases in HIF-1α, p-Src, and p-STAT3 protein levels, whereas all of these proteins were downregulated in KX-01-treated rats (Figures 3A–D). Additionally, overall pulmonary STAT3 expression levels were unaffected in response to BLM or KX-01 treatment (Figure 3C).

The expression of the fibroblast markers α-SMA and p-STAT3 were assessed by immunohistochemical staining in the lungs of our model rats (Figures 4A,B). These analyses revealed that the expression of α-SMA and p-STAT3 was significantly increased in BLM-treated model rats relative to the sham controls whereas KX-01 treatment reversed these increases (Figures 4C,D).

To expand upon in vivo findings, we established an in vitro model of pulmonary fibrosis by stimulating murine L929 cells with CoCl₂ (100 nM) under anoxic conditions to mimic a fibrotic environment. At 72 h post-CoCl₂ treatment, we then assessed the proliferation of these cells via an EdU incorporation assay, revealing that CoCl₂ stimulation markedly enhanced cellular proliferation, whereas KX-01 (0.01 or 0.1 μM) markedly inhibited CoCl₂-induced proliferation (Figure 5A). Therefore, subsequent in vitro assays on related proteins were studied using the 0.01 μM concentration. We found that collagen I, collagen III, and α-SMA protein levels were significantly increased in CoCl₂-treated cells, whereas KX-01 suppressed this effect (Figures 5B,C). Similarly, CoCl₂ promoted the upregulation of p-Src, p-STAT3, and HIF-1α relative to control cells while this was largely prevented by KX-01 treatment (Figures 5D–F) even to normal levels. In addition, no differences in the total STAT3 protein levels were observed in L929 cells as a function of CoCl₂ or KX-01 treatment (Figures 5F,G).

To better understand the mechanisms whereby KX-01 mediates its anti-fibrotic activity, we next treated CoCl₂-stimulated L929 cells with the HIF-1α inhibitor ROT (10 μM) in the presence or absence of KX-01 and then analyzed cellular proliferation (Figure 6A), collagen I, collagen III, α-SMA, p-STAT3, p-Src, and HIF-1α protein levels (Figures 6B–E). Relative to CoCl₂-treated cells, KX-01 or ROT single-agent treatment suppressed L929 cell proliferation and decreased HIF-1α, p-Src, and p-STAT3 protein expression. No further reductions in proliferation or protein expression were observed in cells treated with both KX-01 and ROT, thus suggesting that KX-01 ameliorates experimental pulmonary fibrosis at least in part by inhibiting the activity of HIF-1α (Figures 6D,E and Figure 8).

We next evaluated the role of STAT3 activation in the context of KX-01 anti-fibrotic activity by treating CoCl₂-stimulated L929 cells with the STAT3 inhibitor Stattic (1 μM) in the presence or absence of KX-01, after which we evaluated the proliferation (Figure 7A) and collagen I, collagen III, α-SMA, p-STAT3, p-Src, and HIF-1α protein expression in these cells (Figures 7B–E). Relative to CoCl₂-treated cells, KX-01 or Stattic single-agent treatment suppressed L929 cell proliferation and decreased HIF-1α, p-Src, and p-STAT3 protein expression. No further reductions in proliferation or protein expression were observed in cells treated with both Stattic and KX-01 plus Stattic, thus suggesting that KX-01 ameliorates experimental pulmonary fibrosis at least in part by inhibiting STAT3 activation (Figures 7D,E and Figure 8).