Normal mouse primary fibroblasts were generated by culturing the lungs of C57BL/6 as previously described (Meng et al., 2014). The cells were cultured in Dulbecco’s modified Eagle’s medium (DMEM, Gibco, United States) supplemented with 15% fetal bovine serum (FBS, PAN, German). The cells were cultured at 37°C in 5% CO2 and 95% humidity. Unless specifically noted, all experiments were performed with cells at passage 3.

Human lung fibroblast line IMR90 was purchased from American Type Culture Collection (Manassas, VA). IMR90 were maintained in DMEM supplemented with 10% FBS, 100 units/mL penicillin, and 100 g/ml streptomycin in 5% CO2 and 95% humidity at 37°C.

Cells or dissected mouse lung tissue samples were lysed in ice-cold RIPA lysis buffer with protease inhibitors. Protein concentrations were determined using a BCA Protein Quantitative Analysis Kit (Fudebio-tech) after which protein samples were separated by 8–12% SDS-PAGE and transferred onto polyvinylidene difluoride membranes (Millipore). The membranes were then incubated at room temperature for 1 h in TBST containing 5% BSA. After blocking, the membranes were incubated with primary antibodies for 24 h at 4°C.The following primary antibodies were used: anti-Fibronectin (Abcam, ab268020); anti-Collagen I (affinity, AF7001); anti-alpha smooth muscle (Abcam, ab5694); anti-PFKFB3 (Abcam, ab181861); anti-Beta actin (proteintech, 66009-1-Ig); anti-Hexokinase 2 (proteintech, 22029-1-AP); anti-PKM2 (Proteintech, 15822-1-AP); anti-LDHA (Proteintech, 19987-1-AP); anti-LDHB (Proteintech, 14824-1-AP); and anti-PCBP3 (Abcam, ab154252). Then, the membranes were washed three times with TBST and incubated with donkey anti-rabbit IgG H&L (Abcam, ab175772) for 1 h at room temperature. The membranes were developed using the ECL method according to the manufacturer’s instructions (Millipore) and detected on a GeneGnome XRQ chemiluminescence imaging system (Syngene). ImageJ was used to calculate the relative density of proteins.

The culture medium was washed away with PBS. The cultured cells were fixed with 4% paraformaldehyde for 30 min. Then, the samples were permeabilized with 0.5% Triton X-100 in PBS for 10 min, blocked with 1% BSA in PBS for 1 h at room temperature, and incubated with primary antibodies at 4°C overnight. The primary antibodies included anti- Fibronectin (Abcam, ab268020), anti-alpha smooth muscle (Abcam, ab5694) and anti-PCBP3 (Abcam, ab154252). Then, the cells were washed three times with PBS and incubated with goat anti-rabbit IgG/Alexa Fluor 555-conjugated secondary antibodies (Biosynthesis, bs-0296GA488 and bs-0295G-AF555) for 1 h at room temperature followed by 10 min of DAPI (4’,6-diamidino-2-phenylindole dihydrochloride) staining to visualize cell nuclei visualization as previously described (Chen et al., 2021).

Total RNA was isolated from primary mouse lung fibroblasts using RNA MiniPrep Kits (Zymo Research, R2050). Reverse transcription reactions were performed with a PrimeScriptTM II 1st strand cDNA synthesis Kit (Takara, 6210A/B) according to the manufacturer’s recommendations. qPCR analysis was performed using a HiScript RT- SuperMix for qPCR kit (Vazyme, R223-01) with a CFX96 Touch Real-Time PCR Detection System. The mRNA levels of target genes were normalized to the β-actin mRNA level. Primers used for qPCR are listed in (Table 1).

Cells were seeded in six-well plates and grown until they reach 100% confluence. A “wound” was subsequently created with a sterile 100 μL pipette tip. The cells were pretreated with anlotinib (1 µM) for 3 h and then exposed to TGF-β1 (10 ng/ml) for an additional 24 h. After 24 h, the cells were fixed with 4% paraformaldehyde, and images were obtained using a fluorescence microscope. Wound area can be calculated by manually tracing the cell-free area in captured images using the ImageJ public domain software (NIH, Bethesda, MD).

Cell proliferation was determined by the CCK-8 Kit (Dojindo Laboratories) according to the manufacturer’s instructions. Briefly, 10 μL of CCK-8 solution was added to cultured cells in each well, followed by incubation at 37°C for 1 h. The OD values were measured at 450 nm using a microplate reader. EdU staining was conducted using the BeyoClick™ EdU Cell Proliferation Kit with Alexa Fluor 594 (Beyotime, Cat. No: C00788L). Cells were washed with PBS. Fresh DMEM was added, and then, 10 µM EdU was added into the medium. The cells were incubated for 2 h at 37°C/5% CO2. After the incubation, the cells were washed with PBS to remove the DMEM and the free EdU probe. The cells were then fixed in 4% paraformaldehyde at room temperature for 30 min before being stained with DAPI for 3 min. After an additional wash in PBS, the cells were observed under Nikon ECLIPSE TS100 (Japan).

Primary mouse lung fibroblasts were pretreated with anlotinib (1 µM) for 3 h and then exposed to TGF-β1 (10 ng/ml) for an additional 24 h. Then, the four types of cells were detached and transferred to a 96-well plate in fresh growth medium at a density of 10,000 cells per well for the direct 2-NBDG glucose uptake assay. The cells were rinsed twice with PBS. Glucose uptake was initiated by the addition of 100 μM 2-NBDG to each well. After 30 min, the medium was removed. The plates were then rinsed with PBS, and the fluorescence was measured at an excitation wavelength of 485 nm and an emission wavelength of 535 nm.

To measure lactate production, cells were treated as described for the glucose uptake assay. One hundred thousand cells were then plated into a 12-well plate and incubated in DMEM containing 10% FBS for 10 h. To measure the secretion of lactate, the media were removed, and the cells were incubated in FBS-free DMEM. After incubation for 1 h, the supernatant was collected to measure lactate production (Biovision). The reaction mixture was incubated for 30 min at room temperature in the dark. The lactate levels were measured at 450 nm in a microplate reader and normalized to the protein concentrations. To measure the lactate levels in mouse lung tissue, 10 mg of lung tissues was isolated and homogenized in assay buffer (Biovision). The samples were centrifuged, and the soluble fractions were measured and normalized to the protein concentrations.

The extracellular acidification rate (ECAR) was measured using the Agilent Seahorse XFp Extracellular Flux Analyzer (Seahorse Bioscience). Experiments were performed according to the manufacturer’s instructions. ECAR were measured using Seahorse XF Glycolysis Stress Test Kit (Agilent Technologies). Briefly, cells were transfected or infected as in glucose uptake assay. The transfected cells were harvested and the cell number was counted. After baseline measurements, glucose, the oxidative phosphorylation inhibitor oligomycin, and the glycolytic inhibitor 2-DG were sequentially injected into each well at the indicated time points. Data were analysed by Seahorse XFp Wave software. ECAR is reported in mpH/minute. The cells in each well were digested by trypsin digestion (Gibco, United States), and count cell numbers by cell counting chamber. The results were normalized to normalized to cell number in each well.

Cells were rinsed twice with ice-cold PBS and lysed with an equal pellet volume of RIPA-2 buffer. Protein-A Dynabeads (Invitrogen) were incubated with either mouse IgG or FLAG antibody (Abcam, ab205606). Beads coated in antibody were resuspended in NT2 buffer. Thawed and clarified lysates were added and the bead/antibody/lysate mixture was incubated at 4°C overnight rotating end-over-end. Beads were washed with cold NT2 buffer five times. Proteinase K treatment released RNAs from bound proteins and input and bound RNA was isolated with TRIzol (Invitrogen) and reverse transcribed as described above.

Cells were transfected with empty vector or Flag-PCBP3 and incubated with 100 g/ml cycloheximide for 10 min and lysed with polysome extraction buffer containing 20 mM Tris–HCl, pH 7.5, 100 mM KCl, 5 mM MgCl2 and 0.5% NP-40 as previously described (Kim et al., 2015). Cytoplasmic lysates were fractionated by ultracentrifugation through 10–50% linear sucrose gradients and divided into 12 fractions. The total RNA in each fraction was extracted and analyzed by quantitative RT-PCR analysis.

The Plasmid vector encoding PCBP3 and the empty vector were purchased from Hanbio (Shanghai, China). Primary mouse lung fibroblasts were cultured in six well plates (105 cells/well) and added with 2.5 μg of target plasmid per well. After 12 h, the transfection medium was changed to normal medium. Effects of overexpression on mRNA and protein levels were examined 36 h later. The siRNA targeting mouse PFKFB3 (PFKFB3 siRNA: 5′- CCU​CUU​GAC​CCU​GAU​AAA​UTT-3′) were synthesized by Genepharma Co. (Shanghai, China). Primary mouse lung fibroblasts were cultured in six well plates (105 cells/well) and transfected using Lipofectamine 3,000 (Invitrogen, CA) with PFKFB3 siRNA or negative control siRNA (NC siRNA) for 48 h following the manufacturer’s instructions.

All experiments were conducted in accordance with protocols approved by the Southern Medical University Institutional Animal Care and Use Committee. Female mice (C57BL/6), 6–8 weeks of age, were purchased from Southern Medical University. The mice were kept on a 12 h light-dark cycle with free access to food and water. For bleomycin administration, the mice were anesthetized with 2, 2, 2-tribromoethanol (Sigma-Aldrich), followed by intratracheal instillation of BLM (5 U/kg, i. t.) in 50 μL phosphate-buffered saline (PBS) or equally volume PBS for 21 days. The mice were administered dimethyl sulfoxide (DMSO) (control group) or anlotinib (1 mg/kg, i. p.) once daily for 21 consecutive days. Further experiments were designed to measure the effects of delayed anlotinib administration. Anlotinib treatment was initiated 1 week after exposure to bleomycin, and the mice were administered with anlotinib (1 or 2 mg/kg/day) for 2 weeks, and the mice were sacrificed at day 21. The lungs were harvested for further analyses.

At endpoint, at least 5 mice from each group were anesthetized with 2,2,2-tribromoethanol in saline, tracheotomized below the larynx, and intubated with a tracheal cannula. After the surgery, the mice were placed inside the plethysmographic chamber and the cannula was connected to the machine. Pulmonary function was measured by pulmonary function test system (BUXCO, United States). The system’s software automatically records and displays the pulmonary function parameters.

Lung collagen content was measured with a hydroxyproline (HYP) kit (Nanjing Jian Cheng Institute, Nanjing, China). The lung tissues were prepared for hydrolysis, adjusting the PH value to 6.0–6.8. Subsequently, the developing solution was added to the tissues that were incubated at 37°C for 5 min. Absorbance was read at 550 nm using a microplate reader. Data were expressed as micrograms (µg) of HYP per mg of wet lung tissue.

TGF-β1 were purchased from R&D Systems, Inc. (Minneapolis, MN, United States). Anlotinib dihydrochloride (AL3818, S8726) were purchased from Selleck (Houston, TX, United States).

The results are expressed as the means ± standard deviation (SD). Multigroup comparisons were performed using one-way ANOVA. Student’s t-test was used for comparisons between two groups. A p value of less than 0.05 was considered significant. Replicates consisted of at least three independent experiments. Analyses were performed on SPSS version 25.0 (IBM) for Windows and GraphPad Prism version 6.0 (GraphPad Software, CA).

Given that TGF-β1 is the predominant cytokine that stimulates the differentiation of lung fibroblasts into myofibroblasts and induces ECM production (Sapudom et al., 2015; Huang et al., 2020), we examined the effect of anlotinib (the chemical structure is shown in Supplementary Figure S1) on TGF-β1-induced activation of primary mouse lung fibroblasts (MLFs). The CCK-8 assay results showed that anlotinib did not cause significant cytotoxicity at doses of 1 µM (Figure 1A). To mimic the inhibitory effect of anlotinib on the progression of lung fibrosis, primary MLFs were pretreated with anlotinib (1 µM) for 3 h and then exposed to TGF-β1 (10 ng/ml) for an additional 24 h. Our results demonstrated that TGF-β1 induced the expression of fibronectin, collagen I, and α-SMA, but anlotinib reversed the expression of these fibrotic markers (Figures 1B,C). Immunofluorescence analysis of α-SMA and fibronectin showed similar results (Figures 1D,E). We also examined whether anlotinib affected the proliferation and migration of fibroblasts, which have been shown to significantly contribute to many fibrotic pathologies (Jarman et al., 2014; Huang et al., 2020). As shown by the EdU (Figures 1F,G) and CCK-8 results (Figure 1H), anlotinib treatment prevented the TGF-β1-induced proliferation of primary MLFs. Moreover, anlotinib inhibited the TGF-β1-induced migration of fibroblasts (Figures 1I,J). These results were confirmed in the human IMR90 cell line (Figures 1K, L and Supplementary Figures S2A–D). These data indicate that anlotinib can repress myofibroblast activation and the profibrogenic phenotype in vitro.

To investigate the potential antifibrotic mechanisms of anlotinib, we estimated the most likely macromolecular targets of anlotinib and obtained 100 potential targets through SwissTargetPrediction (Table 2) (Gfeller et al., 2014). A total of 7,360 lung fibrosis-related targets were obtained from the GeneCards database (Table 3) (Safran et al., 2002). To clarify the interaction between potential anlotinib targets and lung fibrosis-related targets, the intersection of the targets was mapped by drawing a Venn diagram and constructing a target network (Figure 2A). Seventy-four common targets were shared between the potential anlotinib targets and lung fibrosis-related targets (Table 4). STRING (version 11.0) was used for protein-protein interaction (PPI) analysis (Figure 2B) (Szklarczyk et al., 2019). Bioinformatics analysis data identified mitogen-activated protein kinase (MAPK) signaling pathway-related genes as the top hits among the 74 genes (Figure 2C). Given that the MAPK pathway is well recognized as a metabolic regulator and that many of these genes control metabolic processes (Figure 2D) (Ho et al., 2004; Papa et al., 2019; Hu et al., 2020; Wang F. et al., 2020), we first confirmed the presence of glycolytic alterations in lung myofibroblasts. We directly measured the levels of lactate and found that both intracellular and extracellular lactate levels in lung myofibroblasts treated with TGF-β1 were significantly increased (Figures 2E,F). Consistent with the augmented glycolysis in lung myofibroblasts, these cells also demonstrated increased glucose consumption (Figure 2G). However, anlotinib treatment decreased the production and secretion of lactate and reduced the consumption of glucose (Figures 2E–G). Accordingly, extracellular acidification rate (ECAR) analysis indicated that treatment with TGF-β1 increased glycolysis and glycolytic activity in primary MLFs, both of which were also reduced by anlotinib (Figures 2H,I). To delineate the mechanisms by which anlotinib inhibits the augmented glycolysis observed in lung myofibroblasts, we assessed the expression of key glycolytic enzymes in these cells. We found that PFKFB3 was induced by TGF-β1 in lung fibroblasts and that anlotinib significantly decreased its expression at the protein level (Figures 2J,K). PFKFB3 is not a rate-limiting glycolytic enzyme; instead, PFKFB3 catalyzes the conversion of fructose-6-phosphate to fructose-2,6-bisphosphate, which is an allosteric activator of PFK1 and a potent stimulator of glycolysis (Atsumi et al., 2002; De Bock et al., 2013). Taken together, these data suggest that anlotinib can abrogate the PFKFB3-driven increase in glycolysis, participating in myofibroblast activation.

Interestingly, the progressive upregulation of PFKFB3 during myofibroblast activation induced by TGF-β1 that was observed at the protein level was not confirmed at the mRNA level, as measured by RT-PCR (Supplementary Figures S3A, B). These results indicate that TGF-β1-induced overexpression does not require de novo transcription of PFKFB3. To further verify these findings, primary MLFs were incubated with cycloheximide to block new protein synthesis, and immunoblotting was used to measure PFKFB3 levels (Figure 3A). The half-life of PFKFB3 was not significantly altered, indicating that TGF-β1 does not influence PFKFB3 protein stability. Therefore, we postulated that PFKFB3 upregulation is modulated through posttranscriptional mechanisms in this context. To verify this hypothesis, we used the online tool catRAPID to screen for potential proteins that may interact with PFKFB3 mRNA and identified that PCBP3 (Table 5) (Agostini et al., 2013; Livi et al., 2016), a member of the PCBP family, has a high probability of directly interacting with PFKFB3 mRNA (Figure 3B) (Choi et al., 2007; Kang et al., 2012; Leidgens et al., 2013; Wang J. et al., 2020). We comparatively analyzed the expression of PCBP3 after treatment with different doses of TGF-β1 by immunoblot analysis and found that PCBP3 protein expression was increased in primary MLFs after TGF-β1 treatment (Figures 3C,D), which correlated with PFKFB3 overexpression. To better define the connection between PCBP3 function and PFKFB3, we performed RNA-protein coimmunoprecipitation (RIP) studies in primary MLFs transfected with FLAG-tagged PCBP3 (FLAG-PCBP3). An antibody targeting the FLAG protein was used to immunoprecipitate FLAG-PCBP3 and any interacting molecules from the cell lysates. Reverse transcription followed by PCR was then used to identify individual PFKFB3 mRNAs isolated with FLAG-PCBP3. We found that PFKFB3 transcripts were enriched by PCBP3 coimmunoprecipitation compared to control IgG coimmunoprecipitation (Figure 3E), demonstrating that PFKFB3 mRNA is indeed a direct target of PCBP3 in MLFs. To test the possibility that PCBP3 may influence PFKFB3 translation, we performed polysome analysis in cells transfected with FLAG-PCBP3. Cytoplasmic lysates were fractionated through sucrose gradients to separate ribosomal subunits (40S and 60S), monosomes (80S) and progressively larger polysomes. RNA was extracted from each of the 12 fractions, and the levels of PFKFB3 and β-actin mRNA were quantified by quantitative RT-PCR. While PFKFB3 mRNA levels peaked in fraction 7 in control cells, the distribution of PFKFB3 mRNA shifted rightward when PCBP3 was overexpressed, peaking in fraction 9, indicating that PFKFB3 mRNA formed, on average, larger polysomes after PCBP3 overexpression (Figure 3F). The distribution of β-actin mRNA was not affected by PCBP3 overexpression. These results indicated that overexpression of PCBP3 increases the translation of PFKFB3. Overall, these results suggest that PCBP3 improves PFKFB3 expression levels by increasing its translation rather than by influencing its protein stability.

To determine the functional impact of PCBP3-mediated regulation of PFKFB3 expression in lung fibrosis, we transfected lung fibroblasts with FLAG-PCBP3. Expression of PFKFB3 was significantly increased by PCBP3 overexpression compared to that of the empty vector control. Reliable markers of the phenotypic transformation of fibroblasts into myofibroblasts, fibronectin, collagen I and α-SMA, were markedly increased in FLAG-PCBP3-treated cells at the protein level (Figures 3G,H) compared with vector-treated cells. In turn, using small interfering RNA (siRNA) to silence PFKFB3, the FLAG-PCBP3-induced overexpression of fibronectin, collagen I and α-SMA was abolished (Figures 3G,H). These findings suggest that PCBP3 protein upregulation is an early and sustained event during fibroblast activation and that the profibrogenic effects of PCBP3 are mediated by PFKFB3 expression. Taken together, these data suggest that PCBP3 posttranscriptionally increases PFKFB3 expression by promoting its translation during myofibroblast activation.

To confirm the regulation of PCBP3 by anlotinib in vitro, we evaluated the protein expression of PCBP3 in MLFs and IMR90 cells. We found that TGF-β1 induced the expression of PCBP3 in MLFs and that anlotinib prevented PCBP3 expression by immunofluorescence analysis (Figure 4A). Western blot analysis of PCBP3 showed a similar result (Figures 4B,C) in MLFs, and these results were confirmed in the human IMR90 cell line (Figures 4D,E). Taken together, these data suggest that anlotinib can repress PCBP3 expression levels during myofibroblast activation in vitro.

To investigate the biological effects of anlotinib on pulmonary fibrosis in vivo, we established a bleomycin (BLM)-induced mouse model of pulmonary fibrosis. The mice were intraperitoneally injected with 1 mg/kg anlotinib daily after BLM administration (Figure 5A). From the first week after bleomycin instillation, the bleomycin-treated mice showed a certain reduction in activity, accompanied by slight shortness of breath. 21 days after bleomycin administration, bleomycin-treated mice showed obvious hyperventilation, accompanied by reduced activity and weight loss, but no similar symptoms were observed in the control group. A single dose of BLM (5 mg/kg) administered by intratracheal instillation successfully induced pulmonary fibrosis in C57BL/6 mice, as evidenced by a decline in pulmonary function, decreased tidal volume (TV, Figure 5B) and dynamic compliance (Cdyn, Figure 5C), and increased lung resistance (RI, Figure 5D). However, treatment with anlotinib significantly reversed bleomycin-induced pulmonary dysfunction. Moreover, we evaluated collagen deposition in the lung tissues by analyzing the hydroxyproline (HYP) content and found that anlotinib treatment reduced the amount of collagen in the lungs of bleomycin-treated mice (Figure 5E). Hematoxylin and eosin (H&E) staining indicated that anlotinib-treated mice had decreased lung inflammation and reduced lung architectural damage (Figure 5F). Accordingly, Masson’s trichrome staining showed decreased collagen deposition in anlotinib-treated mice compared with vehicle-treated mice (Figure 5F). Furthermore, attenuated fibrosis was supported by decreased protein levels of fibronectin and α-SMA by immunohistochemical (IHC) staining (Figure 5G). We also found that anlotinib treatment reduced fibronectin, collagen I and α-SMA expression by western blotting (Figures 5H,I). Taken together, these data show that anlotinib attenuates bleomycin-induced pulmonary fibrosis in vivo.

We next examined whether the levels of PCBP3 were regulated by anlotinib in vivo. We evaluated the expression of PCBP3 in lung tissues and found that the protein levels of PCBP3 were markedly increased after bleomycin instillation, while anlotinib treatment decreased PCBP3 expression (Figures 6A,B). Accordingly, IHC staining showed decreased PCBP3 protein levels in anlotinib-treated mice compared with vehicle-treated mice (Figure 6C). In addition, to confirm the regulation of PFKFB3-driven glycolysis by anlotinib in vivo, we measured the levels of lactate and the expression of PFKFB3 in the lungs of mice. We found that there were significantly higher levels of lactate in the lungs of bleomycin-treated mice than in the lungs of control mice, and anlotinib decreased lactate levels (Figure 6D). Western blot and IHC staining studies revealed that bleomycin-induced PFKFB3 expression in the lungs of mice was prevented by anlotinib (Figures 6E–G). Overall, these results suggest that anlotinib decreases PCBP3 expression and inhibits PFKFB3-driven glycolysis in fibrotic rodent lungs.

We demonstrated that anlotinib treatment could attenuate bleomycin-induced pulmonary fibrosis. In that in vivo experiment, anlotinib was administered at approximately the same time as bleomycin instillation. We further examined whether anlotinib could postpone the progression of established fibrosis. Therefore, we performed another in vivo experiment in which anlotinib was intraperitoneally injected 7 days after bleomycin instillation (Figure 7A). As interventions beginning 7 days post bleomycin were classified as therapeutic (Izbicki et al., 2002; Moeller et al., 2008), we initially treated mice with anlotinib (1 mg/kg/day or 2 mg/kg/day) beginning on day 7 after bleomycin instillation. Pulmonary function tests showed that anlotinib treatment reversed the bleomycin-induced decline in pulmonary function, with increases in TV (Figure 7B) and Cdyn (Figure 7C) and a decrease in RI (Figure 7D). HYP measurements showed that the collagen content was significantly decreased in anlotinib-treated mice compared with vehicle-treated mice (Figure 7E). H&E staining and Masson’s trichrome staining of lungs collected at day 21 showed enhanced recovery from fibrosis upon anlotinib treatment (Figure 7F). Correspondingly, IHC staining showed that anlotinib treatment reduced fibronectin and α-SMA expression in the lungs (Figure 7G). Western blot analysis also showed that anlotinib decreased the protein levels of fibronectin, collagen I and α-SMA in the lungs (Figures 7H,I). Collectively, these data clearly demonstrate that anlotinib accelerates fibrosis resolution in vivo even after the establishment of fibrosis.