Network Pharmacology Prediction and Molecular Docking-Based Strategy to Discover the Potential Pharmacological Mechanism of Wen-Yu-Jin against Pulmonary Fibrosis in a Mouse Model

Background Pulmonary fibrosis (PF) is a devastating lung disease, resulting in gas exchange dysfunction until death. The two drugs approved by the FDA, pirfenidone and nintedanib, have obvious side effects. Wen-yu-jin (WYJ), one of the commonly used herbs in China, can treat respiratory diseases. The potential effects and the underlying mechanism of WYJ against PF are unclear. Purpose Employing network pharmacology, molecular docking, and in vivo and in vitro experiments to explore the potential effects and underlying mechanisms of WYJ in the treatment of PF. Methods Ultra-high pressure liquid chromatography combined with linear ion trap-orbital tandem mass spectrometry (UHPLC-LTQ-orbital trap) was used to identify compounds of WYJ. We got PF-related targets and WYJ compounds-related targets from public databases and further completed critical targets exploration, network construction, and pathway analysis by network pharmacology. Molecular docking predicted binding activity of WYJ compounds and critical targets. Based on the above results, in vivo and in vitro experiments validated the potential effects and mechanisms of WYJ against PF. Results 23 major compositions of WYJ were identified based on UHPLC-LTQ-Orbitrap. According to the results of network pharmacology, STAT3, SRC, IL6, MAPK1, AKT1, EGFR, MAPK8, MAPK14, and IL1B are critical therapeutic targets. Molecular docking results showed that most of the compounds have good binding activities with critical targets. The results of in vivo and in vitro experiments showed that WYJ alleviated the process of fibrosis by targeting MAPK and STAT3 pathways. Conclusion Network pharmacology, molecular docking, and in vivo and in vitro experiments showed the potential effects and mechanisms of WYJ against PF, which provides a theoretical basis for the treatment of WYJ with PF.


Introduction
Pulmonary fibrosis (PF) characterized by progressive dyspnea with respiratory failure [1][2][3] is an irreversible lung disease [4]. PF susceptibility is closely related to age [5]. Accumulating exposures to numerous risk factors [6], such as smoking, occupational dust, drug stimulation, and bacterial and virus infection, also result in PF. Different risk factors repeatedly promote lung injury, leading to the production of profibrotic cytokines. TGF-β1 is one of the most potent profibrosis cytokines [7,8]. Profibrotic cytokines further stimulate effector cells activation and migration, leading to the deposition of extracellular matrix (ECM). ECM, a typical feature of pulmonary fibrosis, affects the gas exchange function of the lung. ey eventually lead to respiratory failure and death [9,10]. Pirfenidone and nintedanib [11], approved by FDA, are the only two effective drugs for clinical PF medical therapies. However, these drugs can only delay the disease's progression and maintain lung function but cannot cure the disease [12,13]. Moreover, these drugs can cause adverse effects [11,[13][14], resulting in treatment discontinuation and adverse gastrointestinal effects.
Multiple active ingredients and targets of Traditional Chinese Medicines (TCM) arouse the attention of pharmacologists. Previous studies showed that TCM or active ingredients from TCM have protective pulmonic [15], representing an attractive source of drug discovery for treating PF. In TCM, there are many theories about the pathogenesis of PF, such as qi stagnation and blood stasis (Qi-Zhi-Xue-Yu in Chinese) and binding of phlegm and qi (Tan-Qi-Hu-Jie in Chinese). Wen-yu-jin (WYJ, Curcuma Radix, 温郁金), derived from the steamed root of Curcuma beauty, has been used for at least 1500 years in China and has the effect of moving qi and blood circulation. Modern pharmacological studies have demonstrated that the ingredients or extracts from WYJ have antibacterial, antitumor anti-inflammatory, and antioxidant effects [16,17]. Hui's study showed potential antifibrotic effects of WYJ in liver fibrosis [18]. erefore, it is necessary to study the therapeutic effects and mechanism of WYJ in the treatment of PF using advanced methods.
In silico approaches have been widely used in the field of drug molecule design, improving efficiency and reducing cost [19,20]. Based on bioinformatics, systems biology, and pharmacology, network pharmacology is a promising tool for understanding the complex relationship between drugs and diseases [21]. is tool has been successfully used to reveal the molecular mechanism of TCM in various diseases such as cardiovascular disease [22,23], diabetes [24], and cancer [25]. Some approaches, such as molecular docking and molecular dynamics simulation (MDS), help researchers understand the pathogenesis of diseases at the molecular level and provide new ideas and theoretical guidance for the design of new drugs [20,26]. In a study, researchers organized MDS to understand the resistance mechanism of mutant BCR-ABL protein, which provided ideas for the design of new drugs [27]. Rachel through in silico operations elucidated that Ligand2 has stronger inhibitory potential for cyclin-dependent kinases (CDKs) and can develop into a new selective inhibitor of CDK2 [19].
In our study, we established network pharmacology to explore the molecular mechanism of WYJ against PF. Molecular docking was performed to predict the affinity strength between the active compounds and the critical targets and to explore the antipulmonary fibrosis potential of WYJ. Finally, the antifibrosis mechanism of WYJ was further verified via in vitro and in vivo experiments based on the predicted results. e MS analysis was conducted on both the negative and positive ion modes. Flow rates of sheath gas and auxiliary gas were set at 40 Arb and 20 Arb, respectively. e capillary voltage was set to 35.0 V. e source temperature was set to 350 and the tube lens was set to 110 V. e source voltage was set to 4 kV and 3 kV in the positive and negative ion modes, respectively.
We used "pulmonary fibrosis (PF)," "idiopathic pulmonary fibrosis (IPF)," or "interstitial lung disease (ILD)" as keywords to obtain PF-related targets from GeneCards database v5.0 [29] (https://www.genecards) and OMIM database [30] (https://omim.org/). After combining the targets obtained from different databases using the same keywords, we got the targets of PF by taking the intersection of targets with different keywords.
WYJ-related targets and PF-related targets were intersected to obtain gene symbols. Using the string database (https://www.string-db.org/) [31], the protein-protein interaction (PPI) network was constructed to explore the potential interactions of the intersected gene symbols. Cytoscape v3.6.0 visualized the PPI network. Critical nodes calculated by the median values of three topological features (degree, closeness centrality, and bitterness centrality) were identified as key targets of WYJ.
To determine the biological meaning behind key targets, pathway enrichment was analyzed by Gene Ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) database of David v6.8 (https://david.ncifcrf.gov/) [32,33]. e threshold value of confident gene enrichment was set at P < 0.05. Compound-target-pathway network explored the connection of compounds of WYJ, key targets, and important signaling pathways and was visualized using Cytoscape.

Molecular Docking.
e active compounds were molecularly docked with the core proteins based on the results of network pharmacology. e specific operations are as According to the previous literature and preliminary experimental results, WYJ-H, WYJ-M, and WYJ-L groups were given WYJ 1.2, 0.6, and 0.3 g/kg doses, respectively. e control group and the model group were treated with normal saline. Treatments were started 12 days after the establishment of the PF model, once a day, for 14 days. We sacrificed all mice on day 26 and stored lung tissue at −80°C for different experiments.

Lung/Body Weight Ratio.
We recorded the body weight (BW) and the whole lung weight (PW) after mice were sacrificed. Pulmonary indexes (PI) were calculated using the equation PI � PW/BW * 100.

Hydroxyproline Determination (HYP).
e lung tissue was processed according to the instructions of the Hydroxyproline Kit (Nanjing Jiancheng Institute of Bioengineering, Nanjing, China). HYP is the main marker of the development of PF. e level of HYP in the lung tissue was measured by the absorbance of 550 nm and the result was expressed with μg/lung.

Histopathological Examination.
Lung tissues fixed with 10% formalin were embedded in paraffin and sectioned. Sections were stained with hematoxylin and eosin (H&E) and Masson (Sinopharm Chemical Reagent Beijing Co., Ltd., Beijing, China). Inflammation and collagen deposition of the sections were observed and assessed with an optical microscope.

Culture of Human Pulmonary Fibroblast (HPF).
HPF is a normal human fibroblast cell line from the National Infrastructure of Cell Line Resource (Beijing, China). HPF cells were cultured in medium (DMEM) with 10% FBS and 1% PS and maintained at the cell incubator with 5% CO 2 and 37°C. Cells were passaged at a confluence of 80%-90% using standard trypsinization techniques. HPF cells stimulated with TGF-β1 (10 ng/mL) were compared with cells that were grown in the same conditions but without stimulation of TGF-β1.      . Membranes were blocked with 5% fatfree milk dissolved with Tris-buffered saline with Tween 20 (TBST) for 120 min at room temperature. Rinsed with TBST, these membranes were incubated with primary antibodies at 4°C overnight. On the second day, blots were rinsed with TBST and then incubated with secondary antibody for 1 h. Blots were visualized by gel imager. GAPDH was used for loading positive control and the results were analyzed using Image J. Antibodies: anti-α-SMA (ab5694, abeam, Cambridge, UK), anti-Collagen 1 (14695, Proteintech), anti-p-MAPK (ab201015, abeam), anti-MAPK (ab184699, abeam), anti-p-STAT3 (ab76315, abeam), anti-STAT3 (ab68153, abeam), anti-GAPDH (60004, Proteintech, Chicago, USA), anti-  Evidence-Based Complementary and Alternative Medicine rabbit IgG (ab205718, abeam), and anti-mouse IgG (ab6728, abeam).

Data and Statistical Analysis.
One-way ANOVA was performed using GraphPad Prism software (version 8.0.2). P values <0.05 are adopted as statistically significant.

Identified Ingredients of WYJ.
Using UHPLC-LTQ-Orbitrap to detect and identify chemical compounds of WYJ ( Figure 1), a total of 23 compounds were confirmed based on comprising with the published data. Information about the identified compounds is summarized in Tables 2 and 3.

Network and Pathway
Analysis. Based on the network pharmacology methods above, 436 WYJ-related targets were obtained from SwissTargetPrediction and TCMSP after removing duplicate items. Information about WYJ-related targets is provided in Supplementary Table S1. Combining the results from GeneCards and OMIM database, a total of 2,867 targets for PF, 1,287 targets for IPF, and 1,613 targets for ILD were collected. As shown in Figure 2(a), 846 overlapping targets were critical targets for PF. e details are shown in Supplementary Table S2. Combining 463 WYJrelated targets with 846 PF-related targets, 124 targets could be potential targets for the WYJ treatment of PF ( Figure 2(b)). Detailed information about 124 potential targets is provided in Supplementary Table S3. STRING databases were used to explore the underlying interactions of 124 potential targets. ere were 124 nodes and 1716 edges of the PPI network visualized by Cytoscape (Figure 3). According to the conditions of three topological (degree >22, bitterness centrality >0.00208, and, closeness centrality >0.5371179), 48 nodes were identified as the key targets of WYJ against PF (Supplementary Table S4). 48 key targets obtained from PPI were selected to investigate the biological processes (BP) and mechanisms of WYJ on PF treatment. Results showed that BP were involved in PF treatment (Figure 4(a)), including positive regulation of transcription from RNA polymerase II promoter, inflammatory response, and signal transduction, etc. Concerning prediction of KEGG by David 6.8, the results revealed that most of the therapeutic targets are associated with signal transduction (e.g., MAPK signaling pathway and PI3K-protein kinase B (Akt) signaling pathway), inflammation (e.g., TNF signaling pathway and HIF-1 signaling pathway), and immune response (e.g., toll-like receptor signaling pathway and Jak-STAT signaling pathway). 20 significant KEGG pathway terms are shown in Figure 4(b).

Molecular Docking.
RMSDs between the cocrystal structures and their original ligands were less than 2Å, which indicated the method for docking was reasonable (Supplementary-Molecular Docking). We found that the affinity energies of most small molecules docking with proteins were less than −5.0 kcal/mol (Supplementary-Molecular Docking). Heat map of molecular docking is shown in the Figure 6(a). e major binding interactions between the compounds and target proteins include hydrophobic interaction and hydrogen bonding. For example (Figure 6(b)), bisdemethoxycurcumin formed four hydrogen bonds with amino acid residues Glu71, Met109, Gly110, and Asp112 with distances of 1.76Å, 1.93Å, 2.64Å, and 1.72Å, respectively, and formed hydrophobic interactions with amino acid residues Lys53, Ile84, Ala111, and Leu167.

WYJ Alleviated BLM-Induced Pulmonary Fibrosis in
Mice. e BLM-induced PF model was used to verify the mechanism of WYJ on the treatment of PF. We observed significant decreases in survival rate, severe weight loss, and      e datum was presented as meaning ± SEM (n ≥ 3). * P < 0.05, * * P < 0.01 and * * * P < 0.001 versus the model group; # P < 0.05, ## P < 0.01 and ### P < 0.001 versus the control group.
increase lung/body weight ratio in the BLM-induced PF model compared with the control group. e WYJ treatment increased the survival rate, attenuated weight loss, and reduced the lung/body weight ratio (Figures 7(a)-7(c)). HYP, an important fibrosis indicator of collagen deposition, increased in the model group and decreased by the WYJ treatment (Figure 7(d)). A large amount of TGF-β1 was observed during the development of PF [34,35]. In our study, WYJ treatment downregulated the expression of TGF-β1 mRNA in lung tissue of the PF model (Figure 7(e)). As revealed by Masson staining (Figure 7(f )) and H&E staining (Figure 7(g)), WYJ treatment decreased the infiltration of inflammatory cells and relieved thrombus and structure destruction compared with the model group. e above results confirmed that WYJ exerted a therapeutic effect on PF.
Excessive deposition of ECM, such as collagen I [36], is characteristic of fibrosis. Compared with the model group, the WYJ treatment suppressed the mRNA expression of collagen I (Figure 8(a)). e differentiation of lung fibroblasts into myofibroblasts is essential for the development of PF [37,38], and α-SMA is a marker of myofibroblasts [35].
e WB results indicated that the expression of α-SMA decreased after WYJ treatment compared with the model group (Figure 8(b)), indicating that WYJ reduced the production of myofibroblasts.
According to the results of network pharmacology analysis, it is worth noting that STAT3 is the most important target in the compound-target-pathway network. Moreover, IL6, MAPK1, AKT1, EGFR, MAPK8, MAPK14, and IL1B were the key targets interacting closely with the MAPK signaling pathway. erefore, we speculated that WYJ might improve PF mainly via the STAT3 and MAPK signaling pathways. Abundant evidence has indicated that STAT3 plays a key role in the occurrence and development of fibrosis [39,40], whereas the conduction of the MAPK signaling pathway is closely related to PF [41,42]. erefore, we experimentally verified the expression of key proteins in these two pathways by WB. Compared with the control group, the phosphorylation levels of MAPK1 increased, while this trend was markedly reduced after WYJ treatment (Figure 8(c)). In addition, the same manner was observed in the expression level of phosphorylated STAT3 (Figure 8(d)). ose results revealed that the protective effect of WYJ on PF might be related to the inhibition of MAPK and STAT3 pathways.

WYJ Inhibited Fibroblast Activation in In Vitro
Experiments. In vitro experiments further verified the antipulmonary fibrosis effect of WYJ. We observed no change in cell morphology at different concentrations of WYJ using microscopy. However, CCK-8 results showed that, at a concentration of 50 μg/mL, the viability of HPF cells decreased slightly (Figures 9(a) and 9(b)). erefore, we used 0.1-20 μg/mLWYJ to observe the effect of WYJ on TGF-induced HPF cells.
Activated lung fibroblasts, the primary roles in the progress of fibrogenesis, express excessive α-SMA and collagen I. We used TGF-β1 to promote fibroblast activation [35] and evaluated the expression of α-SMA and collagen I after WYJ treatment.
e results demonstrated that HPF cells stimulated with TGF-β1 significantly increased the production of α-SMA and collagen I protein. HPF cells pretreated with WYJ markedly decreased expression of α-SMA and collagen I protein (Figures 9(c) and 9(d)). And the results also showed that WYJ prevented the activation of fibroblasts by downregulating p-MAPK1 and p-STAT3 in vitro (Figures 9(e) and 9(f )).

Discussion and Conclusions
PF is a chronic progressive tissue repair response that can lead to irreversible scarring and remodeling of the lungs [9]. As mentioned in the introduction section, the occurrence of PF involves lung injury, the production of profibrotic factors, and the deposition of ECM. Reducing the expression of profibrotic factors and the production of ECM is extremely important for alleviating PF. Pirfenidone and nintedanib, approved by the FDA, have side effects and withdrawal symptoms [11,[13][14], and there is an urgent need to develop drugs with fewer side effects. TCM has accumulated valuable information alleviating lung diseases. WYJ, from steamed roots of Curcuma beauty, was included in the 2020 edition of the Pharmacopoeia of People's Republic of China. Modern studies have shown that the ingredients of WYJ can alleviate diseases through antibacterial, antitumor, anti-inflammatory, and antioxidant effects [16,17]. Among the ingredients of WYJ, bisdemethoxycurcumin can reduce renal fibrosis through anti-inflammatory, antioxidant, and antiapoptotic effects [43,44], ar-turmerone plays an antipsoriatic effect by inhibiting cell proliferation and reducing the expression of inflammatory cytokines [45], p-Methoxycinnamic exerts anticancer properties through anti-inflammatory effects [46], and curcumin has anti-pulmonary fibrosis effect in a murine model [47,48]. Based on the above research, we studied the impact of WYJ on PF and related molecular mechanisms. It has developed many animal models to study the pathogenesis and treatment of different diseases in humans.
e BLM-induced PF mouse model has been widely accepted as a model for studying PF. is model could cause a pulmonary pathological change, which is like that of patients with lung fibrosis [49,50]. Compared with the model group, we observed through pathological slices that WYJ treatment could decrease the infiltration of inflammatory cells and relieve thrombus and structure destruction of lung tissues (Figures 7(f ) and 7(g)). Previous studies have indicated that TGF-β1, α-SMA, and ECM proteins (such as fibronectin, collagen I, and collagen III), and HYP are the main markers of the development of PF [7][8][9][10]51]. Our results showed that WYJ significantly inhibits the expression of these fibrosis markers in PF mice, which suggests that WYJ has an antifibrosis effect in the lungs. e weight change and survival curve also showed that WYJ can improve the quality of life of PF mice.
Various lung cell types (such as alveolar epithelial cells, endothelial cells, mesenchyme fibroblasts, and circulating fibroblasts) are involved in the occurrence and development of PF under pathological conditions [52]. However, differentiation of lung fibroblasts into myofibroblasts is essential for the development of PF [37,38]. As the most critical fibrosis-promoting factor, TGF-β1 can induce lung fibroblasts to differentiate into myofibroblasts [7]. erefore, we used TGF-β1 to induce HPF and observed whether WYJ can inhibit HPF proliferation and differentiation. We found that WYJ significantly reduced the expression of fibrosis markers (α-SMA, collagen I) in HPF cells induced by TGF-β1. ese results are consistent with the results of the mouse model.
Multiple targets and multiple pathways regulate the occurrence of PF. So, only targeting one target for treatment will not eliminate the pathological activation of other targets in PF. erefore, we used network pharmacology to screen out 9 critical targets of WYJ in the treatment of PF. MAPK1 (ERK2), MAPK8 (JNK1), and MAPK14 (p38-α) are essential members of the MAPKs family. MAPKs signaling cascades regulate processes, including cell cycle progression, cell migration, cell survival, and differentiation [53]. More and more evidence showed that the MAPK pathway is involved in many aspects of PF, such as the recruitment of fibroblasts and the deposition of extracellular matrix [41,42,54,55]. e production of IL-1β is achieved through a two-step process. e activation of the MAPK pathway participates in the synthesis of the inactive 31 kDa precursor (pro-IL-1β), and the inflammation induces the production of the active/mature form of IL-1β (17 kDa) [56,57]. IL-1β not only is an effective inducer of TGF-β1 [58] but also causes BLM-induced pulmonary toxicity through the caspase-1/IL-1β pathway [59]. EGFR is a transmembrane protein with intrinsic tyrosine kinase activity. Studies showed that EGFR is abnormally activated in animal models of PF and mediates the pathogenesis of pulmonary fibrosis [60][61][62]. Inhibition of EGFR reduced the phosphorylation of ERK1/2, leading to inhibition of the activation of profibrotic pathways [63]. KEGG analysis results showed that IL6, MAPK1, AKT1, EGFR, MAPK8, MAPK14, and IL1B were the key targets interacting closely with the MAPK signaling pathway. IL6, SRC, and ERGF can cause the activation of the downstream target STAT3, leading to lung fibrosis. STAT3 regulates cell growth, proliferation, differentiation, and migration [64]. Phosphorylated STAT3 has been reported to be involved in PF [39,40,65,66]. e primary receptor coupled with STAT3 is the gp130/IL-6 receptor family, including IL-6 [67,68]. In the noncanonical TGFβ signaling, EGFR and SRC also mediate the phosphorylation and activation of STAT3 [69][70][71][72]. IL6, SRC, and ERGF can cause the activation of the downstream target STAT3, leading to lung fibrosis. To sum up, we speculated that WYJ might improve PF mainly via the MAPK and STAT3 signaling pathways. We performed experimental verification of the expression of essential proteins (MAPK1 and STAT3) in the two pathways. Western blot results showed that the phosphorylation of MAPK1 and STAT3 in the lung tissues of the model group was significantly increased, which is consistent with previous studies [18,42,63,73]. WYJ alleviated PF by reducing the phosphorylation of MAPK1 and STAT3. In in vitro experiments, we also observed the same results, confirming the reliability of in vivo experiments. e molecular docking results described above are helpful for a basic understanding of the mechanism of drug action. If the affinity energy is less than −5 kcal/mol, it indicates that the target has certain binding activity with the compound. We found that the affinity energies of most small molecules docking with proteins were less than −5.0 kcal/mol. We performed further analysis of the interaction between the target proteins and the compounds. e comparative analysis showed that positive control drug nintedanib showed good binding activity, but some of the active compounds of WYJ (bisdemethoxycurcumin, 1,2-dihydrocurcumin) showed higher binding activity with the core gene target to some extent, which also provides direction for the development of new drugs and subsequent research.
In conclusion, our study identified that WYJ alleviated pulmonary fibrosis. According to the results of network pharmacology and molecular docking, the pharmacological mechanism of WYJ in the treatment of PF was further validated by experiment. e inhibitory effect of WYJ on PF may function mainly by inhibiting MAPK and STAT3 signaling pathways, which provide new insights for the treatment of WYJ at PF. e method of exploring the pharmacological mechanism of WYJ used in our study provides a novel approach to explain the pharmacological basis of other herbs. Network pharmacology has limitations. More experiments are needed to verify the reliability antifibrotic effect of WYJ.

Evidence-Based Complementary and Alternative Medicine 13
Data Availability e datasets used and analyzed during the current study are available from the corresponding author on reasonable request.

Conflicts of Interest
e authors declare that they have no conflicts of interest. Acknowledgments e manuscript is already published in the preprint https:// www.researchsquare.com/article/rs-923117/v1 [74]. No funding received.

Supplementary Materials
Supplementary