Abstract Background Vitiligo stands as a challenging skin disorder with limited treatment options available. LiuWei DiHuang formula (LDF), a renowned Traditional Chinese medicine, has exhibited promising results in treating vitiligo over an extended period. However, the precise underlying mechanism of its action remains elusive. Methods Employing a comprehensive network pharmacology approach, this study identified active compounds and their corresponding targets within LDF, while also pinpointing vitiligo-associated targets sourced from the TCMSP database, OMIM, DisGenNET, and Genecards. A network was established to illustrate the connections between active compounds and targets, alongside a protein-protein interaction network. Further analyses, encompassing Gene Ontology (GO) function and KEGG pathway enrichment, were conducted using the DAVID platform. Molecular docking simulations were performed utilizing AutoDockTools and AutoDockVina software. To validate the outcomes of the systematic pharmacological investigation, experiments were conducted using juvenile zebrafish. Results The collective effort of the network pharmacology approach yielded a compilation of 41 compounds and 192 targets. Molecular docking simulations notably revealed the lowest binding energies for CAT-quercetin and CAT-Kaempferol interactions. The utilization of juvenile zebrafish experiments highlighted a significant increase in melanocyte count following methoxsalen and LDF treatment. Notably, LDF prominently augmented the expression levels of proteins related to melanogenesis. Additionally, LDF showcased the capacity to enhance CAT and SOD levels while concurrently reducing ROS and MDA activity. In contrast to the model group, substantial increases in protein and mRNA levels of Nrf2 and HO-1 were observed in response to LDF treatment (P < 0.05). Conclusion Through a meticulous network pharmacology approach, this study successfully predicted active components and potential targets associated with LDF's application in vitiligo treatment. The therapeutic effectiveness of LDF against vitiligo is postulated to stem from its regulation of oxidative stress factors and the Nrf2/HO-1 pathway. Keywords: LDF, Vitiligo, Network pharmacology, Oxidative stress factors, Nrf2/HO-1 pathway 1. Introduction Vitiligo, a prevalent and persistent skin condition, is characterized by the gradual loss of pigmentation in both skin and mucosal tissues. This disheartening disorder affects approximately 1 % of the global population [[29]1]. The conspicuous presence of the disease on patients' skin not only takes a physical toll but also significantly disrupts their daily life and professional commitments. Consequently, individuals grappling with this condition often experience profound psychological distress. While the treatment approach for vitiligo is often tailored to each individual's specific needs, there are certain overarching strategies. For instance, in more advanced stages, the utilization of glucocorticoids emerges as a common choice, effectively minimizing the progression of depigmented areas. During the acute phase, topical glucocorticoids along with calcineurin inhibitors like tacrolimus and pimecrolimus prove beneficial. During the stable phase, therapies such as narrow-band ultraviolet phototherapy are employed. In cases where vitiligo presents a formidable challenge, techniques like skin grafting can be considered [[30]2,[31]3]. It's worth noting that despite these approaches, current vitiligo treatments are characterized by their time-intensive nature, high costs, and often suboptimal efficacy. At present, Traditional Chinese Medicine (TCM) is gradually being applied to the prevention and treatment strategy of a variety of diseases, and receiving considerable attention in treatment of vitiligo. Compared with conventional treatment regimens, the long-term application of TCM has less toxicity and minor side effects, therefore, TCM is becoming increasingly popular among patients with vitiligo [[32][4], [33][5], [34][6]]. Liuwei Dihuang formula (LDF) is a classical TCM herbal formula consisting of six herbs: Prepared Rehmannia Root (Radix Rehmanniae Praeparata), Medical Dogwood (Cornus officinalis Sieb. et Zucc.), Chinese yam (Rhizoma Dioscoreae), Oriental Waterplantain Rhizome (Alisma plantago-aquatica Linn), indian bread (Wolfiporia cocos (F.A. Wolf) Ryvarden & Gilb.), Tree Peony Bark (Moutan Cortex). It has been clinically proved that Liuwei Dihuang, a representative formula for nourishing the liver and kidney, can treat both depigmentation skin disease vitiligo and hyperpigmentation skin disease chloasma, it had significant advantages in the treatment of pigment disorders, and can improve the pathological alterations of abnormal increase or decrease of pigment [[35]7,[36]8]. The bi-directional metabolic regulation of Liuwei Dihuang formula on melanin has been verified in experimental models such as mouse B16 melanoma cells and human primary cultured normal melanocytes [[37]9,[38]10]. And our previous study showed that Liuwei Dihuang formula can promote the proliferation and melanin synthesis of human normal melanocytes cultured in vitro, and the same time the serum containing Liuwei Dihuang formula can up regulate tyrosinase activity in melanocytes and mouse models [[39][11], [40][12], [41][13]]. However, owing to the diversity of chemical components of traditional Chinese medicine, it is difficult for traditional experimental research to elaborate the mechanism of Liuwei Dihuang formula on melanin metabolism as a whole. Recently a new method to study the mechanism of action of traditional Chinese medicine has appeared, network pharmacology has been extensive used in the study of various traditional Chinese medicine compounds, providing a new research idea for the complex components and mechanism of traditional Chinese medicine. Via widely application of network pharmacology, this study screened the monomeric components of Liuwei Dihuang formula for the treatment of vitiligo with melanin metabolism disorder, established a network model of effective components for the treatment of disease, looked for the active compounds in the formula that promote the formation of melanin, and experimental validation to explore the mechanism of Liuwei Dihuang by molecular docking technology, zebrafish model and molecular biology experiment.The flow path of our study is displayed in [42]Fig. 1. Fig. 1. [43]Fig. 1 [44]Open in a new tab Technical strategy of the current study. 2. Materials and methods 2.1. Active ingredients database construction of LDF and screening Compounds of the each herb in LDF were obtained through Traditional Chinese Medicine System Pharmacology Databasetcmsp (TCMSP) ([45]http://tcmspw.com/). According to the characteristics of drug absorption, distribution, metabolism and excretion (ADME) evaluation system, in which the oral bioavailability (OB) ≥ 30 % and the drug likeness (DL) ≥ 0.18 were set to be the thresholds values for screening the active ingredients and import the active ingredients into the TCMSP for obtaining the targets of each active compounds with probability ≥0.15. And then input the targets into UniProt protein database ([46]https://www.uniprot.org), “Home sapiens” and “reviewed” were used as screening conditions to standardize the protein targets of compounds. All the action targets of the active ingredients in Liuwei Dihuang formula were sorted out, and the active ingredient targets network was constructed by using Cytosape 3.9.0 software. 2.2. Collection and screening vitiligo related targets In order to obtain the information of disease-associated target genes comprehensively, with “vitiligo” as the key word to screen the targets related to vitiligo in the following disease databases: Genecards (https://www.genecards.org)、OMIM (https://omim.org)、DisGeNET (http://www.disgenet.org). And the obtained targets were standardized through the UniProt protein database ([47]https://www.uniprot.org), which collected the genes information of targets after removing duplicates. 2.3. “Herbal-Compound-Disease-Target” network construction Via Venn 2.1（https://bioinfogp.cnb.csic.es/tools/venny/）, taking the intersection of targets of active compounds and vitiligo to obtain the action targets of Liuwei Dihuang formula for the treatment of vitiligo. The network topology of intersecting targets was analyzed by using Cytoscape 3.9.0 software, and constructed the network diagram of “Herbal-Compound-Disease-Target”. 2.4. Construction of protein-protein interaction (PPI) network Imported common targets into String database ([48]https://string-db.org) to obtain the visualization of PPI network, selected “multiple proteins”, the species were limited to Homo sapiens, and the PPI medium confidence score >0.4 was the filtering parameter recommended by String database. Download the protein TSV format file, imported it into Cytoscape 3.9.0 software for visualization, screening the core targets with cytohubba plug-in, and constructed the protein interaction network. Nodes and edges represent candidate targets and protein-protein interactions in the PPI network, respectively. In this study, the core target proteins were selected and identified by the parameter “Degree”. Degree was used to evaluate the topological importance of the nodes in the network. 2.5. KEGG and GO enrichment analyses In order to obtain key pathways of LDF treating vitiligo, the Database for Annotation, Visualization and Integrated Discovery (DAVID; version 6.8; [49]https://david.ncifcrf.gov) was used to conduct the gene ontology (GO) and the Kyoto Encyclopedia of Genes and Genomes (KEGG) Pathway Enrichment Analysis. Biologic Process (BP), Molecular Function (MF), and Cell Component (CC) are the general modules that are included in GO term enrichment. The results were visualized by website (bioinformatics.com.cn). Finally, the active component-target-pathway interaction network diagram is drawn by Cytoscape3.9.0. Results with P < 0.05 were destined for further analysis. 2.6. Molecular docking to verify the core target To screen the main effective components and targets of Liuwei Dihuang formula for promoting melanogenesis, using the small organic molecules and information on their biological activities database PubChem ([50]https://pubchem.ncbi.nlm.nih.gov/) to obtain the structure file of ligand small molecule, and construct the 3D structure of the compound with ChemOffice software, save it in *mol2 format and minimize its energy. The crystal structures were of hub target genes screened from the RSCB Protein Data Bank database (PDB) ([51]https://www.rcsb.org/). The water molecules and ligand small molecules in the downloaded protein molecules are removed by the visualization software PyMOL, AutodockTools 1.5.6 was used to open the ligand molecule, hydrogenation, gasteiger charge on the receptor protein molecules, detection of ligand root, search and definition of rotatable bond, and finally save it in pdbqt format. The software AutoDock 4.0 was used to perform molecular docking. Each compound's docked conformation was ranked into clusters according to the binding energy, and the top-ranked conformations whose position with the lowest binding energy (as the most suitable conformation) was selected to be visually analyzed by using Discovery Studio 4.5. 2.7. Main reagents and instruments MS-222 (sigma, USA); Methylcellulose (sigma, USA); RNase free water, (Takara); Fastpure cell/Tissue Total RNA Isolation Kit (Vazyme Biotech Company Ltd., Nanjing, Jiangsu, China); Strong cracking liquid Ripa (biyuntian, China); BCA protein Assay Kit (Pierce Thermo-Scientific, Rockford, IL, USA); Polyvinylidene fluoride membranes (Millipore, Bedford, MA, USA); Ethanol (sigma, USA, art. No. 270741); Isopropanol (sigma, USA, art. No. 34863); Chloroform (Shanghai McLean Biochemical Technology Co., Ltd., Article No. t819817); the MDA, SOD, CAT, and ROS content in the tissue of zebrafish was determined by detection kit (KeyGen Biotech. Co., Nanjing, China) in accordance with the instructions; Prime script RT Master Mix (perfect real Takara, rr036a); SYBR ® Premix ex taqtm Ⅱ (TII RNaseH Takara, rr820b); Primers were purchased from Shanghai Sangong Bioengineering Co., Ltd. see attached [52]Table 1 for primer sequence. Table 1. Primer sequences for real-time PCR. Gene name Primer sequence (5′–3′) Product (bp) GenBank serial number Nrf2 Forward GGTTGCCCACATTCCCAAATC 119 [53]NM_001313901.1 Reverse CAAGTGACTGAAACGTAGCCG HO-1 Forward TTCAAGCAGCTCTACCGCTC 90 [54]NM_002133.2 Reverse GAACGCAGTCTTGGCCTCTT GAPDH Forward GGACTCATGACCACAGTCCAT 109 [55]NM_000194.2 Reverse CAGGGATGATGTTCTGGAGAG [56]Open in a new tab Liuwei Dihuang formula were purchased from the pharmaceutical factory of Beijing Tongrentang science and Technology Development Co., Ltd., 120 Capsules/bottle, production batch No. 0073716. 2.8. Preparation of LDF Liuwei Dihuang formula was prepared by Beijing Tongrentang science and Technology Development Co., Ltd. In brief, 12 g Prepared Rehmannia Root, 12 g Medical Dogwood, 12 g Chinese yam, 12 g Oriental Waterplantain Rhizome, 12 g indian bread, 12 g Tree Peony Bark were decocted in 1200 ml and 900 ml of water twice for each 0.5 h. The decoction was combined, filtered and concentrated to 150 ml. After cooling, the decoction was mixed with same amount of ethanol and the mixture was kept for more than 24 h. The supernatant was filtered and the filtrate was concentrated to 10–15 ml. After 40 g of the mixture of sucrose:dextrin (3:1) was added, the mixture was granulated, dried and sieved with a 12 mesh sieve to obtain the Liuwei Dihuang formula pills. LDF was diluted with distilled water to concentrations of 20, 40, and 80 μg/ml. All solutions were stored at 4 °C until use. 2.9. Establishment of juvenile zebrafish melanin regeneration model and drug administration AB strain zebrafish were obtained from CZRC (China Zebrafish Resource Center, CZRC) ([57]http://www.zfish.cn/) and cultured in the zebrafish culture system with light: dark = 14:10 cycle at the constant temperature of 25 °C–28 °C. Embryos from natural spawning were collected in embryo medium (5 mM NaCl, 0.17 mM KCl, 0.33 mM CaCl[2], 0.33 mM MgSO[4], 0.7 mM HEPES, with distilled water, pH 7.2, supplemented with 1 % methylene blue). The male and female fish are placed in the spawning pond in the ratio of 2:3. Within 30 min after spawning, collect the embryos in egg water and put them into the incubator. The temperature is maintained at 28.5 ± 0.5 °C. Egg water was changed 2–3 times within 24 hpf (24 h post-fertilization). Then, the culture medium was replaced with LDF (20–80 μM), methoxsalen (1 μM) was used as a positive control. Zebrafish embryo medium (3000 μl) was added to each well of a six-well plate, followed by adding 10 embryos. After 1 h, replaced with LDF (0–120 μM), methoxsalen (1 μM), the survival rate of embryos was observed for the next 48 h. The dead embryos were removed from the well after 24 and 48 h and calculated survival rate of embryo. 2.10. Measure the area of melanocytes in specific parts of zebrafish head Spontaneous melanin content was measured from zebrafish larvae at 3 day post-fertilized (dpf). The zebrafish larvae were administrated with LDF (20–80 μM) and methoxsalen (1 μM) after 24hpf, and washed away after 72 h, anesthetizing zebrafish larvae in 3-Aminobenzoic acid ethyl ester methanesulfonate (MS-222) for about 5–10min, the larvae were mounted in 2 % methyl cellulose on a depression slide, and images were collected using an Olympus SZ2-ILST stereomicroscope (Tokyo, Japan). The densitometric analysis was performed using Image J software (National Institute of Health, Bethesda, MD). The quantification of pigmentation data was calculated as the percentage in comparison with the untreated control. 2.11. Determination of melanin content in juvenile zebrafish Zebrafish were washed twice with egg water after drug treatment group and experimental group. Take 10 live juveniles randomly from each group, add them into 2 ml EP tube, and suck up the liquid as much as possible. Add 200 μl Ripa lysate per tube, smash the tissue. The process needs to be operated on ice. Centrifuge at 4 °C 13,000 rpm for 10 min, and suck all the liquid from the upper layer after centrifugation. Add 600 ml of 1 mol/l NaOH to each EP tube and treat with metal bath at 100 °C for 30 min. Shake up the liquid and transfer it to 96 well plate. Each group is provided with 2 double holes, 200 μl/hole. The OD value of 490 nm was measured by microplate reader. The maximum and minimum values of each concentration group are removed and then the average value is removed. Relative melanin content = (absorbance value of drug group/absorbance value of blank concentration group) × 100 %. All experiments were repeated three times. 2.12. RNA extraction and quantitative reverse transcription polymerase chain reaction At 96 hpf, zebrafish larvae of control and LDF-treated groups were collected for RNA isolation. Total RNA was extracted from 20 zebrafish larva using Fastpure cell/Tissue Total RNA Isolation Kit (Vazyme Biotech Company Ltd., Nanjing, Jiangsu, China) following the manufacturer's protocol. cDNA was synthesized using a Primescript RT Master Mix (perfect real Takara, rr036a); SYBR ® Premix ex taqtm Ⅱ (TII RNaseH Takara, rr820b). A real-time PCR assay was performed using SYBR ® Premix ex taqtm Ⅱ (TII RNaseH Takara, rr820b); Primers were purchased from Shanghai Sangong Bioengineering Co., Ltd. The sequence of the sense and antisense primers were shown in [58]Table 1. 2.13. LDF stimulates expression of MITF, p-MITF, tyrosinase, TRP1 and TPR2 Western blotting was carried out in accordance with the relevant standard techniques. The total protein of zebrafish tissues homogenate were extract using RIPA lysis buffer. The supernatants were collected by centrifugation at 12,000×g for 30 min at 4 °C. The BCA protein Assay Kit was used to quantify the protein concentration. For WB, 40 μg protein was loaded and run on an 8 % SDS-PAGE gel. The protein bands were transfer by using Polyvinylidene fluoride membranes. 5 % BSA solution preparation with TBST was used to block the membrane for 1h at room temperature, and then incubated with primary antibodies at 4 °C overnight. The primary antibodies used in the study were as follows: Anti-MITF (Abcam, ab215845), Anti-TRP1 (Abcam, ab235447), Anti-TRP2 (Abcam, ab221144), p-MITF antibody was purchased from Affinity Biosciences (AF3027), TYR antibody (CST, 8954S), Anti-Nrf2 (CST, 12721S), and Anti–HO–1 (CST, 26416S) antibodies were purchased from Cell signaling Technology. After washing three times in Tris-buffered saline containing 0.1 % Tween-20 at pH 7.6 (TBST), the membranes were incubated with the appropriate secondary antibody (1:4000), Anti-rabbit IgG, HRP-linked Antibody #7074 and Anti-mouse IgG, HRP-linked Antibody #7076, which were purchased from Cell Signaling Technology, Inc. Quantification of the bands was performed by measuring the signal intensity using Image-Pro Plus (version 6.0) and normalizing them to the signal of the housekeeping β-Actin antibody. 2.14. Detection kit to test oxidative stress factors After administration with different concentration of LDF and methoxsalen for 72 h, 150 zebrafish larvae in each group were pooled together in centrifuge tubes and homogenized on ice-cold saline (1 g of tissue in 9 ml of normal saline). The supernatants were collected for analysis of anti oxidative enzyme activities and lipid peroxidation after centrifugation at 3500 rpm for 15 min at 4 °C. Reactive oxygen species (ROS, E004-1-1), Superoxide dismutase (SOD, A001-1-2), Malondialdehyde (MDA, A003-1-2), Catalase (CAT, A007-1-2), levels were measured by assay kits which were purchased from Nanjing Jiancheng (Nanjing, China). 2.15. Statistical analysis The results were presented as mean ± standard deviation (‘x ± s). The significance was determined by using one-way analysis of variance (ANOVA) followed by Dunnett's post-hoc test, using GraphPad Prism 6 (La Jolla, CA, USA) and SPSS 22 (IBM, Armon, NY, USA). Significant differences compared with the control were identified when the P value was less than 0.05. All the experiments were performed in triplicate. 3. Results 3.1. Screening active targets of LDF and targets vitiligo After ADME screening via TCMSP database, removing the compounds that could not find the relevant targets, 41 unduplicated compounds and 192 unduplicated targets were obtained, which included 475 compound-target relationships. The active ingredient-target network diagram was constructed by Cytoscape3.9.0. The network involves 239 nodes (6 traditional Chinese medicines, 41 compounds and 192 drug targets) and 475 edges, each edge represents the interaction between the compound and target ([59]Fig. 2A). Among five botanical drugs of Prepared Rehmannia Root, Medical Dogwood, Chinese yam, Oriental Waterplantain Rhizome, indian bread and Tree Peony Bark corresponded to 2, 13, 12, 7, 6 and 6 compounds, respectively, the general information of the active compounds of LDF is shown in [60]Table 2. We found a potential synergistic effect of these six botanical drugs in the target level without many overlaps in the compound level. Fig. 2. [61]Fig. 2 [62]Open in a new tab Constructed compound-disease-target network. A. LDF-active ingredients-target network; B. LDF-Vitiligo-common target Venn diagram; C. LDF-Vitiligo-target network. Table 2. Active compounds in LDF. Herbs number Mol ID Molecule Name OB(%) DL Prepared Rehmannia Root __________________________________________________________________ A1 MOL000359 sitosterol 36.91 0.75 A2 __________________________________________________________________ MOL000449 __________________________________________________________________ Stigmasterol __________________________________________________________________ 43.83 __________________________________________________________________ 0.76 __________________________________________________________________ Medical Dogwood __________________________________________________________________ SZY1 MOL000358 beta-sitosterol 36.91 0.75 A1 MOL000359 sitosterol 36.91 0.75 A2 MOL000449 Stigmasterol 43.83 0.76 SZY2 MOL001494 Mandenol 42 0.19 SZY3 MOL001495 Ethyl linolenate 46.1 0.2 SZY4 MOL001771 poriferast-5-en-3beta-ol 36.91 0.75 SZY5 MOL002879 Diop 43.59 0.39 SZY6 MOL002883 Ethyl oleate (NF) 32.4 0.19 SZY7 MOL005503 Cornudentanone 39.66 0.33 SZY8 MOL005530 Hydroxygenkwanin 36.47 0.27 SZY9 MOL005531 Telocinobufagin 69.99 0.79 SZY10 MOL008457 Tetrahydroalstonine 32.42 0.81 SZY11 __________________________________________________________________ MOL005481 __________________________________________________________________ 2,6,10,14,18-pentamethylicosa-2,6,10,14,18-pentaene __________________________________________________________________ 33.4 __________________________________________________________________ 0.24 __________________________________________________________________ Chinese yam __________________________________________________________________ SY1 MOL000322 Kadsurenone 54.72 0.38 A2 MOL000449 Stigmasterol 43.83 0.76 SY2 MOL000546 diosgenin 80.88 0.81 SY3 MOL000953 CLR 37.87 0.68 SY4 MOL001559 piperlonguminine 30.71 0.18 SY5 MOL001736 (−)-taxifolin 60.51 0.27 SY6 MOL005430 hancinone C 59.05 0.39 SY7 MOL005435 24-Methylcholest-5-enyl-3belta-O-glucopyranoside_qt 37.58 0.72 SY8 MOL005438 campesterol 37.58 0.71 SY9 MOL005440 Isofucosterol 43.78 0.76 SY10 MOL005458 Dioscoreside C_qt 36.38 0.87 SY11 __________________________________________________________________ MOL005465 __________________________________________________________________ AIDS180907 __________________________________________________________________ 45.33 __________________________________________________________________ 0.77 __________________________________________________________________ Oriental Waterplantain Rhizome __________________________________________________________________ A1 MOL000359 sitosterol 36.91 0.75 ZX1 MOL000831 Alisol B monoacetate 35.58 0.81 ZX2 MOL000849 16β-methoxyalisol B monoacetate 32.43 0.77 ZX3 MOL000853 alisol B 36.76 0.82 ZX4 MOL000856 alisol C monoacetate 33.06 0.83 ZX5 MOL002464 1-Monolinolein 37.18 0.3 ZX6 __________________________________________________________________ MOL000862 __________________________________________________________________ [(1S,3R)-1-[(2R)-3,3-dimethyloxiran-2-yl]-3-[(5R,8S,9S,10S,11S,14R)-11- hydroxy-4,4,8,10,14-pentamethyl-3-oxo-1,2,5,6,7,9,11,12,15,16-decahydro cyclopenta [a]phenanthren-17-yl]butyl] acetate __________________________________________________________________ 35.58 __________________________________________________________________ 0.81 __________________________________________________________________ indian bread __________________________________________________________________ FL1 MOL000273 (2R)-2-[(3S,5R,10S,13R,14R,16R,17R)-3,16-dihydroxy-4,4,10,13,14-pentame thyl-2,3,5,6,12,15,16,17-octahydro-1H-cyclopenta [a]phenanthren-17-yl]-6-methylhept-5-enoic acid 30.93 0.81 FL2 MOL000275 trametenolic acid 38.71 0.8 FL3 MOL000279 Cerevisterol 37.96 0.77 FL4 MOL000282 ergosta-7,22E-dien-3beta-ol 43.51 0.72 FL5 MOL000283 Ergosterol peroxide 40.36 0.81 FL6 __________________________________________________________________ MOL000296 __________________________________________________________________ hederagenin __________________________________________________________________ 36.91 __________________________________________________________________ 0.75 __________________________________________________________________ Tree Peony Bark MDP1 MOL000098 quercetin 46.43 0.28 MDP2 MOL000211 Mairin 55.38 0.78 A1 MOL000359 sitosterol 36.91 0.75 MDP3 MOL000422 kaempferol 41.88 0.24 MDP3 MOL000492 (+)-catechin 54.83 0.24 MDP3 MOL007374 5-[[5-(4-methoxyphenyl)-2-furyl]methylene]barbituric acid 43.44 0.3 [63]Open in a new tab A total of 650 vitiligo-related proteins were obtained from Genecards, OMIM, DisGeNET and other disease databases and 56 common targets of active components and vitiligo were screened by Venn2.1.0 software, as shown in [64]Fig. 2B. The network of vitiligo and the active ingredient of LDF, was constructed by Cytoscape3.9.0 software, showed in [65]Fig. 2C. 3.2. PPI network construction and screening key targets After screening all of the compounds and targets, we want to recognize the potential corresponding compounds and targets of LDF anti-vitiligo effect. Based on String and Cytoscape3.9.0 software, the PPI network was constructed consisting of 55 protein nodes and 555 edges, and the core targets were screened by cytohubba plug-in. The edges represent the interaction between proteins. The more connections, the greater the correlation. The size and color of the nodes represent the Degree value, the deeper the color and the larger the size of nodes, represents the larger the Degree value. The weight of the edge represents the combines core value ([66]Fig. 3). According to the properties of network topology, compounds and drug targets which connected with nodes may play a core role in the whole network. The top 10 key proteins according to the Degree value include TNF, TP53, CAT, IL6, CASP3, VEGFA, IL1B, MYC, PPARG and PTGS2, as shown in [67]Table 3. Fig. 3. [68]Fig. 3 [69]Open in a new tab Screening compoouds-disease core targets. A. LDF-Vitiligo-target PPI network; B. LDF-Vitiligo-core target network. Table 3. Core targets with top 10° values. Rank Name Degree 1 TNF 41 2 TP53 40 3 CAT 39 3 IL6 39 5 CASP3 38 6 VEGFA 37 7 IL1B 36 8 MYC 35 9 PPARG 34 9 PTGS2 34 [70]Open in a new tab 3.3. Functional enrichment analysis GO analysis of the potential therapeutic target genes was performed using the DAVID database. The x-axis indicates the number of genes enriched in that term. The redder the color, the smaller the value of P adjust (FDR); it also denotes greater credibility and greater importance. In contrast, the bluer the color, the greater is the value of P adjust. For a brief demonstration, we intercepted the top 10 terms of GO analysis and the KEGG metabolic pathway of the top 20 from small to large according to the P-value. ([71]Fig. 4A). The y-axis represents GO terms. The x-axis indicates the number of genes enriched in that term. The redder the color, the smaller the P-value; it also denotes greater credibility and greater importance. The results indicated that target genes were mostly enriched in cytokine-mediated signaling pathway, response to drug, cellular response to lipopolysaccharide in Biological Process (BP); extracellular region, macromolecular complex, mitochondrion in Cell Composition (CC) analysis; enzyme binding, protein homodimerization activity, identical protein binding, and cytokine activity in Molecular Function (MF). The result of KEGG pathway enrichment analysis indicated that target genes were significantly enriched in pathways in GE-RAGE signaling pathway, fluid shear stress and atherosclerosis, and Lipid and atherosclerosis ([72]Fig. 4B). Fig. 4. [73]Fig. 4 [74]Open in a new tab A. GO-BP, GO-CC, and GO-MF enrichment analysis plots. Only the top 10 GO-terms are displayed in the categories of biological process (GO-BP), cellular component (GO-CC) and molecular function (GO-MF); B. KEGG enrichment analysis. Only the top 20 KEGG-terms are displayed. 3.4. Molecular docking Quercetin and kaempferol are bioflavonoids. Many studies have shown that quercetin and kaempferol have good anti-inflammatory and antioxidant effects [[75][14], [76][15], [77][16], [78][17], [79][18]]. According to the results of network pharmacological analysis, kaempferol and quercetin are one of the effective components in the LDF, while the differentially expressed hub gene catalase (CAT), as a downstream crucial regulatory target, is closely related to the pathogenesis of vitiligo. Among the calculated binding energies of various compounds, the binding energies of CAT-quercetin and CAT-kaempferol are the lowest, indicating that the binding activity is better. Therefore, kaempferol and quercetin were selected for virtual molecular docking with CAT, as shown in [80]Fig. 5. Fig. 5. [81]Fig. 5 [82]Open in a new tab Molecular docking of quercetin and kaempferol with CAT. A. CAT-quercetin; B. CAT-kaempferol. 3.5. LDF promote melanogenesis in juvenile zebrafish Before the investigation of melanogenesis, a toxicity assay was performed to determine the optimal avirulence concentration of LDF to zebrafish. As shown in [83]Fig. 6A, there was no difference in survival rate of embryo when the concentration were 0, 20, 40 and 80 μg/ml of LDF, and these concentration had no significantly toxicity observed in embryos. Methoxsalen is a drug that has been historically used in the treatment of vitiligo. Our previous research found that there was no statistically significant difference in the survival rate of zebrafish embryos under 1 μM methoxsalen compared to the control group, and it could significantly increase the melanocyte area of zebrafish head. Zebrafish juvenile were administrated with 1 μmol/L methoxsalen and 20 μg/ml, 40 μg/ml, 80 μg/ml LDF respectively for 72 h. The results showed that compared with the control group, the melanocyte area of zebrafish head observably increased after treated with methoxsalen and LDF, while the concentration was 20 μg/ml, the melanocyte area showed no significant difference, and LDF promoted the melanogenesis in juvenile zebrafish, and its effect was similar to that of the methoxsalen group compared with the control group. Moreover, the relative melanogenesis after administration with different concentration of LDF and methoxsalen were increased compared with control group, the difference was statistically significant (P < 0.01; [84]Fig. 6B–D). In addition, we further refined our exploration accordingly, the juvenile zebrafish were exposured to LDF at different concentrations, the expression levels of melanogenesis-related protein: microphthalmia-associated transcription factor (MITF), phosphorylation microphthalmia-associated transcription factor, tyrosinase (TYR), tyrosinase-related protein 1 (TRP1), tyrosinase-related protein 2 (TRP2), were measured. As shown in the results, the levels of TYR and TRP1 proteins were significantly increased with the administration of the LDF (20 μg/ml, 40 μg/ml, 80 μg/ml) and methoxsalen (P < 0.05), whereas MITF, p-MITF and TRP2 expression level showed no difference with the treatment of 20 μg/ml LDF, and as up the administration dose, the expression level gradually increased ([85]Fig. 6E). The results indicated that LDF can promote melanogenesis and up-regulate the expression of melanogenesis-related genes. Fig. 6. [86]Fig. 6 [87]Fig. 6 [88]Open in a new tab LDF promote melanogenesis in juvenile zebrafish. A. Toxicity assays in embryos with drugs administration of LDF (0–120 μg/ml); B. The effect of LDF on melanogenesis with drugs administration of LDF (20–80 μg/ml and 1 μM methoxsalen) for 72 h; C-D. The effect of LDH on melanin contents; E. The effect of LDH on the protein expression of MITF, p-MITF, TYR, TRP1, and TRP2.*, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001 vs Control. 3.6. LDF regulates levels of oxidative stress factors According to the results of network pharmacological analysis, showed that the catalase (CAT), one of the differentially expressed hub genes in the LDF treatment with vitiligo. To investigate the mechanisms of LDF therapeutical effect to vitiligo, so we further detected the activities of oxidative stress indexes ROS, SOD, MDA and CAT in the tissue of juvenile zebrafish. As shown in [89]Fig. 7, compared with the control group, after administration of LDF (40 μg/ml and 80 μg/ml), the level of ROS and MDA decreased and the activities of SOD and CAT increased in juvenile zebrafish, whereas, the concentration was lower 40 μg/ml, the level of CAT, MDA have no significance. Moreover, the concentration of LDF is 80 μg/ml, the content of oxidative stress factors was similar with methoxsalen administration, which suggested that LDF can promotes melanogenesis by regulating the level of oxidative stress factors in juvenile zebrafish. Fig. 7. [90]Fig. 7 [91]Open in a new tab The effect of LDH on the activity of oxidative stress factors in juvenile zebrafish. A. SOD; B. ROS; C. CAT; D. MDA. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001vs Control. 3.7. LDF up-regulated Nrf2/HO-1 antioxidant pathway To further reveal the mechanism by which LDF regulates oxidative stress factors, we firstly measured the protein levels of Nrf2/HO-1 antioxidant pathway in juvenile zebrafish. We found that treatment with 20 μg/ml of LDF did not significantly affect the expression of Nrf2 protein, while the administration with 40 μg/ml and 80 μg/ml, the level of these two proteins increased significantly (P < 0.05; [92]Fig. 8A). Moreover, we investigated the mRNA levels of Nrf2 and HO-1 in juvenile zebrafish after LDF and methoxsalen treatment. The results showed that compared with control group, Nrf2 and HO-1 expression levels was elevated with the administration of 40 μg/ml and 80 μg/ml LDF, and further increased in methoxsalen-treated group (P < 0.05; [93]Fig. 8B). Fig. 8. [94]Fig. 8 [95]Open in a new tab The effect of LDH on Nrf2/HO-1 signal pathway in zebrafish larvae at 96 hpf. A. The protein level of Nrf2 and HO-1; B. The mRNA levels of Nrf2 and HO-1. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001vs Control. 4. Discussion Vitiligo is a common depigmentation disease of the skin caused by the disappearance or reduction of the number of melanocytes, which is seriously affecting the quality of patients’ life [[96]19,[97]20]. LDF, a well-known traditional Chinese medicine has been proved to be effective in the treatment of vitiligo due to its unique characteristics. However, the molecular mechanism of LDF on vitiligo still remained unclear, that limited its appliance and modernization. Network pharmacology is an creative method, which systematically detects the mechanism of action of traditional Chinese medicine on a variety of diseases [[98]21,[99]22]. In this work, the effects of LDF against vitiligo and its potential pharmacological mechanisms were studied based on network pharmacological analysis and experimental validation. Network analysis identified relevant potential targets (CAT). Then, the molecular docking and in vivo juvenile zebrafish model validation further confirmed that, LDF significantly increased the percentage of melanin area and melanin synthesis in juvenile zebrafish, and increased the protein levels of MITF, p-MITF, TYR, TRP1 and TRP2. At the same time, LDF inhibited levels of oxidative stress factors such as ROS and MDA, and increased the level of SOD and CAT. WB results showed LDF can active the Nrf2/HO-1 pathway. Therefore, inhibiting oxidative stress factors and increasing the synthesis of melanocytes may be the potential mechanism of LDF against vitiligo. Network pharmacological analysis applied to identify the core targets in order to elucidate the basic mechanism of LDF against vitiligo. In this part, according to the analysis of " Active ingredients -vitiligo ", quercetin, kaempferol, sitosterol are the main active components of LDF acting on vitiligo related genes. Modern studies have shown that the incidence of vitiligo is closely related to oxidative stress, and quercetin has a strong antioxidant effect. Animal experiments have shown that quercetin can increase the distribution density and number of melanosomes significantly [[100][23], [101][24], [102][25]]. H[2]O[2]can lead to the decrease of melanocyte viability, the change of cell morphology and microstructure, and the occurrence of apoptosis. Cell experiments found that quercetin can avoid oxidative damage caused by H[2]O[2] by reducing intracellular ROS, improve cell viability, and reduce the rate of apoptosis [[103]26,[104]27]. Kaempferol is a flavonoid compound with anti-inflammatory, anti-oxidation and anti-cancer effects. Tyrosinase activity was measured in vitro by the mushroom tyrosinase DOPA rate oxidation method. The results showed that the activation rate of 2 mM kaempferol on tyrosinase was as high as 110 % [[105]28]. Our molecular docking results showed that both kaempferol and quercetin could bind to CAT with good activity. The core targets of LDF active components were identified through PPI network, including CAT、TNF、TP53、IL-6, which are closely related to the process of Oxidative stress pathway. It is found that DNA base oxidative damage is closely related to the pathogenesis of vitiligo [[106]29,[107]30]. Oxidative stress can activate the unfolded protein response of keratinocytes, increase the expression of chemokine CXCL16, and induce CD8^+T cells to attack melanocytes [[108]31,[109]32]. In addition, the expression of IL-15 in keratinocytes induced by oxidative stress helps to activate CD8^+T cells, which is a new mechanism to induce autoimmunity in vitiligo [[110]33]. Therefore, antioxidant therapy may protect melanocytes and block autoimmune attack, thus contributing to the treatment of vitiligo. Researchers have studied the effect of oxidative stress on vitiligo and found that compared with the activities of MDA, SOD, GSH and other antioxidant enzymes, oxidative stress has a significant effect on the plasma activity of patients with vitiligo [[111]34,[112]35]. It can be seen that there is an imbalance between oxidation and antioxidation in the plasma of patients with vitiligo, and the imbalance between oxidation and antioxidation in the plasma of patients with advanced vitiligo is more serious than that in the stable phase [[113]36]. Our juvenile zebrafish experiments showed that LDF could significantly reduce the levels of various oxidative stress factors in a dose-dependent manner such as ROS, SOD and MDA; it can increase the level of protective factor CAT simultaneously. ROS overload significantly impaired melanogenesis [[114]37]. In depigmentation diseases such as vitiligo, unbalanced antioxidant system and uncontrolled ROS overload will damage melanocytes and reduce cell viability. Meanwhile, ROS is cleared [[115]38,[116]39]. It is reported that the activation of Nrf2/ARE antioxidant pathway is the main method for skin cells to scavenge ROS [[117]40,[118]41]. In this experiment, LDF could up regulate the protein and mRNA levels of Nrf2 and HO-1 in juvenile zebrafish, which means that LDF can activate Nrf2/HO-1 pathway. Therefore, we suggest that LDF can protect melanocytes from oxidative stress damage by activating Nrf2/HO-1 antioxidant pathway and scavenging ROS. There are many Chinese and Western medicine treatment methods for vitiligo, and they all have certain effects. TCM and Western medicine can complement each other's advantages, and maximize the clinical efficacy, which is also the inevitable trend of future medical development [[119]42]. While this study has provided valuable insights into the potential therapeutic effects of Liuwei Dihuang Formula (LDF) on vitiligo and its underlying mechanisms, several limitations should be acknowledged: In vitro and Zebrafish Model: The study predominantly relied on in vitro analyses and a zebrafish model to investigate the effects of LDF on melanogenesis and oxidative stress. Although these experimental systems offer advantages in terms of cost-effectiveness and ease of manipulation, they do not fully recapitulate the complexity of human physiology and pathophysiology. Further studies using animal models more closely related to human skin pigmentation, such as murine models or ex vivo human skin explants, would be valuable to validate the findings observed in this study. Network Pharmacology Predictions: The network pharmacology approach used in this study offers a comprehensive way to predict potential interactions between compounds and targets. However, these predictions are based on existing databases and may be subject to errors or incomplete information. Experimental validation of the interactions identified through network pharmacology, such as molecular docking studies, provides additional support but still requires further in-depth investigation. These limitations highlight the need for further research to validate and expand upon the findings presented here. Future studies incorporating more diverse experimental approaches, including clinical trials, and a deeper mechanistic understanding will contribute to a more comprehensive assessment of the therapeutic efficacy of LDF for vitiligo treatment. 5. Conclusion In a word, this study found that hub genes and some functional biological pathways that involved in the treatment of LDF on vitiligo disease based upon bioinformatics and molecular docking. Then we demonstrated that LDF increasing the synthesis of melanocytes and exerted a significant against vitiligo effect in juvenile zebrafish, the mechanism may be potent via activating the Nrf2/HO-1 pathway and inhibiting the level of oxidative stress factors. These findings provide new potential diagnostic markers and therapeutic targets of LDF. Author contribution statement Dandan Wang:conceived and designed the experiments; performed the experiments; analyzed and interpreted the data; wrote the paper. Yan Yang:performed the experiments; analyzed and interpreted the data; Gulijiayina Hengerjia:performed the experiments; analyzed and interpreted the data; Yan Deng:conceived and designed the experiments; analyzed and interpreted the data; contributed reagents, materials, analysis tools or data; wrote the paper. Funding statement We appreciate all the participants for their exceptional cooperation as well as valuable contributions. This research was supported by the National Natural Science Foundation of China (Project No. 81202704). Data availability statement Data will be made available on request. Declaration of competing interest The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper. Acknowledgements