Abstract Introduction Thyroid cancer (TC), a common endocrine malignant tumor with a higher incidence in females than in males, is a serious threat to human life and health. Although current clinical treatments can alleviate this disease, the recurrence rate remains high. Tuber Dioscoreae Bulbiferae, also called Huang-Yao-Zi (HYZ), has remarkable curative properties, few side effects and is used for the treatment of sore throat, goiter, hemoptysis, and other diseases in traditional Chinese medicine (TCM). Existing clinical studies have found that HYZ can improve the clinical symptoms of TC patients and reduce tumor volume, while in vitro studies have found that HYZ can induce the death of thyroid cancer cells. However, the mechanism of HYZ in the treatment of TC is still unclear. Methods In this study, based on network pharmacology and bioinformatics, the target and molecular mechanism of HYZ in the treatment of TC were preliminarily explored. The results suggest that the antithyroid cancer effect of HYZ may be related to the PI3K-Akt and focal adhesion pathways. Then, a TC cell model was established to further explore the detail molecular mechanisms. Results Fortunately, HYZ induced apoptosis in KMH-2 cells and regulated the expression of apoptosis-related proteins and genes. At the same time, HYZ can also significantly inhibit the migration and invasion of TC cells. Further studies showed that the pharmacological activities of HYZ were related to the regulation of the PI3K-Akt and focal adhesion pathways. Conclusion Our study provides a reference for further animal or clinical studies investigating the effectiveness and molecular mechanisms of HYZ against thyroid cancer. Keywords: Thyroid cancer, Bioinformatic analysis, Apoptosis, PI3K-Akt, Network pharmacology 1. Introduction At present, cancer is a very prominent public health problem worldwide, seriously threatening human life and health. In China, lung cancer, thyroid cancer (TC), stomach cancer and other malignant tumors account for more than 75% of all new cases [[31][1], [32][2], [33][3]]. TC is one of the most common endocrine malignant tumors, and the incidence rate continues to increase every year [[34]4]. In addition, TC is three times more common in women than in men. The pathogenesis of thyroid cancer is still unclear, but iodine intake, genetic factors, radiation exposure, and other factors, such as sex hormones, are thought to be the main causes of thyroid cancer [[35]5]. TC often presents as a painless neck mass or nodule, and the tumor can compress the esophagus and trachea [[36]6]. At present, the clinical treatment for TC is mainly surgical resection combined with postoperative drugs and local radiotherapy. Although the curative effect is good, the subsequent recurrence rate is still very high, and the adverse reactions brought by the treatment seriously increase the pain suffered by patients. In addition, the high cost of treatment increases the financial burden of patients [[37]7]. Therefore, it is necessary to develop low-cost, effective, and safe complementary or alternative antithyroid cancer drugs. Tuber Dioscoreae Bulbiferae (Huang-Yao-zi, HYZ), is the dried tuber of Dioscorea Bulbifera L., also known as Hong Yaozi, Ku Yaozi, Huang Gouzi, and so on. This medication has the effect of clearing heat, detoxifying, cooling blood and detumescence and is used for the treatment of sore throat, goiter, vomiting blood, hemoptysis and so on [[38]8,[39]9]. In recent years, with the deepening of research, scholars have found that the chemical composition of HYZ is complex, including astragalus, flavonoids, diterpenoid lactones and other components, with anti-inflammatory, antitumor, antiviral, and other pharmacological activities [[40][10], [41][11], [42][12], [43][13], [44][14], [45][15], [46][16], [47][17]]. Modern clinical application also found that HYZ and its prescription can significantly improve the symptoms of thyroid tumor and TC patients and reduce the size of tumor tissue [[48][18], [49][19], [50][20], [51][21]]. Modern studies have also shown that HYZ can inhibit the growth of human TC cells by downregulating the mRNA and protein expression of SW579 Surviving [[52]22]. However, the molecular mechanism of HYZ against TC is still unclear. Therefore, in this study, we used network pharmacology combined experimental studies to study the therapeutic effect of HYZ on TC and explore its molecular mechanism. 2. Materials and methods Ethical statement All experimental protocols were strictly followed in accordance with the International Ethical Guidelines for Biomedical Research Involving Human Subjects, and all the experimental protocols were carried out with the approval of the Ethics Committee of the Chengdu University of Traditional Chinese Medicine (No. 2019-08). 3. Reagents and materials Trans Serum EQ Fetal Bovine Serum (FBS) was purchased from TransGen Biotech (Beijing, China), and the prestained protein marker was obtained from Accurate Biotechnology (Hunan) Co., Ltd. (Hunan, China); RAPI buffer, nuclear protein extraction kit, BCA protein assay reagent, primary antibodies for c-Caspase 3, phosphorylation- (p-)-Bad, Bad and β-actin were purchased from the Boster Biological Technology Company (Wuhan, China); Annexin V-FITC/PI kit was purchased from Multisciences (Lianke) Biotechnology Corporate Limited (Hangzhou, China); CCK-8 kits were obtained from US Everbright Inc. (Suzhou, China); enhanced chemiluminescence (ECL) luminescence reagent and a cell cycle detection kit were purchased from Beijing 4A Biotech Co. (Beijing, China); primary antibodies for FAK, p-FAK, and E-CAD were purchased from Cell Signaling Technology Co. (Danvers, MA, USA); and protein kinase B (Akt), p-Akt, phosphoinositide 3-kinase (PI3K), p-PI3K and goat-anti-rabbit/rat horseradish peroxidase-conjugated (HRP) secondary antibodies were purchased from Abcam (Cambridge, MA, USA). 4. Component retrieval of HYZ The key words “Tuber Dioscoreae Bulbiferae” were searched in the TCMSP database to collect the chemical components of HYZ [[53]23]. Meanwhile, the compound library of HYZ was supplemented by literature mining. Oral bioavailability (OB) is one of the most important parameters in pharmacokinetic evaluation of drug absorption distribution, metabolism, and excretion, reflecting the speed and degree of absorption of the active ingredients in oral drugs into the body circulation, and is also an important indicator to evaluate whether oral drugs can exert efficacy. The Dl-like property (DL) value represents the similarity between the ingredients and the known chemical drugs, which has important reference value for determining whether the ingredients of traditional Chinese medicine (TCM) exert effects in the body [[54]24]. Considering the medicinal properties of the active ingredients in HYZ, we selected compounds with OB ≥ 30% and DL ≥ 0.18 as the 24 potential active ingredients of HYZ. 5. Target collection of HYZ The targets of the active components of HYZ were collected from the TCMSP database and supplemented with literature mining. In addition, the active components of HYZ were uploaded to PharmMapper ([55]http://www.lilab-ecust.cn/pharmmapper/) [[56][25], [57][26], [58][27]] and Swiss Target Prediction ([59]http://www.swisstargetprediction.ch/) [[60]28,[61]29], and the action targets of these components were obtained through reverse docking. Finally, the results of TCMSP, PharmMapper and Swiss Target Prediction were combined to delete repeated targets and construct component-target data pairs. The UniProt ([62]https://www.uniprot.org/) database was used to correct the gene standard names of the obtained targets and finally obtain the target library of the active components of HYZ. 6. Collection of TC targets The GEO database is an open public database that stores the experimental results of high-throughput sequencing of various microarray chips. With “thyroid cancer” as the key word, all data sites that study gene expression differences in thyroid tissues of cancer patients and normal people were searched in the GEO database. Finally, the gene expression profile dataset [63]GSE3678 was collected, which included 7 papillary TC tissue samples and 7 normal thyroid tissue samples. Background correction and data normalization were performed on the dataset with the help of the GEO2R online tool, and the genes with an FDR value < 0.05, |Log2FC| > 1 were screened as differentially expressed genes (DEGs). Finally, the target library of TC was obtained, and the volcano map of [64]GSE3678 was analyzed and visualized by using the online SangerBox tool ([65]http://www.sangerbox.com/). 6.1. Protein‒protein interaction network (PPI) The intersection of the HYZ target library and TC target library was processed, and the overlapping target was used as the possible target for HYZ to improve TC. The obtained overlapping genes were uploaded to the string database ([66]https://string-db.org/), and the species was restricted to humans to obtain the data pairs of interactions between these targets. The data pairs were imported into Cytoscape software to construct a protein‒protein interaction network (PPI), in which nodes represented targets and edges represented interaction relationships between targets. The network structure of PPI was analyzed by the network analyzer plug-in, and the hub gene in PPI was screened out. 6.2. Drug-component-target-disease network diagram (DCTD) The drug-component, component-target and target-disease data pairs were constructed using R language software and then imported into Cytoscape software to construct a drug-component-target-disease network diagram (DCTD). In the network diagram, nodes represent potential active ingredients and targets, and edges represent the corresponding relationship between ingredients and targets. Network structure was analyzed through the network analyzer plug-in, and network topology parameters such as degree were calculated to analyze the main active components and action targets of HYZ in improving TC and to further scientifically and reasonably clarify the material basis and action mechanism of HYZ in improving TC. 6.3. Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) analysis Gene Ontology (GO) functional annotation and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway enrichment analysis of overlapping genes were performed using the cluster Profiler toolkit of R language software. In the analysis, species were set as human sources, enrichment results with P < 0.05 were defined as having significant significance, and a relevant bar chart bubble chart was drawn. 7. Preparation of lyophilized powder from HYZ HYZ was purchased from Neautus Chinese Herbal Pieces Ltd. Co. (Chengdu, China) and identified as the whole grass of Dioscorea Bulbifera L. by Prof. Chunjie Wu (School of Pharmacy, Chengdu University of TCM). The samples of HYZ were kept in the library (No. HYZ-01). We took an appropriate amount of HYZ, added eight times the amount of distilled water to soak for 1 h, boiled it on high heat, then boiled it on low heat for 1 h. Then, we filtered the residue to take the filtrate, filtered the filtrate with filter paper, and freeze dried the filtrate with a lyophilizer to obtain the lyophilized powder of HYZ water extract, which was stored in the refrigerator at −20 °C for reserve. 7.1. Cell culture KMH-2 cells were purchased from Beina Biological Co. (Beijing, China) and cultured in DMEM containing 10% FBS and 1% penicillin at 37 °C in a humidified 5% CO2 atmosphere. The new medium was replaced every 2 days, and the cells were subcultured every 3 days. Three to five generations of cells were selected for the experiments in this study. 7.2. Cell viability detection The CCK8 kit was used to determine the effect of HYZ on cell viability. KMH-2 cells at the logarithmic growth stage were prepared into a single-cell suspension and then inoculated into 96-well plates at 1 × 10^4 cells/well. They were placed in a 37 °C constant temperature incubator for 12 h, and then different concentrations of HYZ (0–120 μg/mL) were added for 24 h, 48 h and 72 h. At the end of the experiment, 10 μL CCK-8 solution was added to each well and incubated for half an hour at 37 °C. The OD value of each well was determined at 490 nm using a microplate reader, and cell viability was calculated using the following formula. [MATH: Cellsurvivalrate(%)=(ODDrug−ODBlank)/(ODNormal−ODBlank)×100% :MATH] Wound HealingThe effect of HYZ on KMH-2 cell migration was determined by a wound healing test. KMH-2 cells (1 × 10^6) were inoculated into 6-well plates and cultured in DMEM complete medium containing 10% PBS for 24 h. When the cells were fused to 90%, serum-free medium was replaced for 12 h. Then, a 200 μL spear was used to make cross wounds at the bottom of the orifice plate, and the experiment was incubated with different concentrations of HYZ (10, 20, 40 μg/mL) for 0 h, 6 h, 12 h and 24 h. At the end of the experiment, scratch images were collected under a microscope at each time period, and the scratch area was measured and analyzed using ImageJ software. 7.2.1. Cell migration assay The effect of HYZ on KMH-2 cell migration was determined by a transwell chamber. In brief, KMH-2 cells (1 × 10^5) were resuspended in FBS-free DMEM and inoculated in the upper chamber of a Transwell chamber. The lower chamber of the transwell chamber was filled with DMEM containing 20% FBS. Meanwhile, the cells were treated with different concentrations of HYZ (10, 20, 40 μg/mL) for 24 h. At the end of the experiment, the cells were placed into crystal violet staining solution for staining, and then cotton swabs were used to gently remove the cells in the inner side of the upper ventricle membrane. The cells on the outer side of the membrane were considered to be successfully migrated cells, and migration images were collected under the microscope. 7.2.2. Cell invasion assay Transwell chambers were used to determine the effect of HYZ (10, 20, 40 μg/mL) on the invasion ability of KMH-2 cells. At 37 °C, the Transwell membrane was pretreated with 0.1% Matrigel for 30 min and then washed with PBS 3 times. The procedure was similar to that of the transwell migration experiment. Cell invasion was observed under a light microscope. 8. Apoptosis was detected by flow cytometry The Annexin V-FITC/PI apoptosis assay kit was used to determine the effect of HYZ (20, 40, 60 μg/mL) on apoptosis of KMH-2 cells. In simple terms, cells in logarithmic growth phase were inoculated into 6-well plates with 1 × 106/wells and incubated in a 37 °C constant temperature incubator for 12 h, followed by intervention with different concentrations of HYZ for 24 h. At the end of the experiment, cells were collected, washed with precooled PBS twice, and the supernatant was centrifuged. The cells were suspended in 500 μL binding buffer, mixed with 5 μL Annexin V-FITC and 5 μL PI, and incubated at room temperature for 15 min. The apoptosis rate of the cells was determined by flow cytometry (BD, New York, NY, USA). 8.1. Western blot (WB) Western blot (WB) assays were used to detect the effect of HYZ (20, 40, 60 μg/mL) intervention on protein expression in cells. In simple terms, cells in the logarithmic growth phase were inoculated into 6-well plates at 1 × 10^5 cells/well and incubated in a 37 °C constant temperature incubator for 12 h, followed by 24 h of treatment with different concentrations of HYZ. At the end of the experiment, RIPA cell lysis solution was used to extract total cell protein, and a BCA kit was used to detect the cell protein concentration. Then, an appropriate amount of 5 × protein loading buffer was added, and the protein was blown and mixed and boiled in 90 °C water baths for 10 min to desaturate the protein. Finally, sodium dodecyl sulfate‒polyacrylamide gel electrophoresis (SDS‒PAGE) was used to isolate total cell proteins and transfer them to PVDF membranes, which were placed in a blocking solution at room temperature for 1 h and then washed with TBST three times. The membrane and the corresponding primary antibody (Bad, p-Bad, PI3K, p-PI3K, Akt, p-Akt, FAK, p-FAK, E-CAD, β-actin) were incubated overnight at 4 °C and then incubated with HRP-labeled secondary antibody at room temperature for 1 h. Finally, the PVDF membrane was developed with ECL sensitizer, and the protein band images were photographed and saved. 8.2. Analysis and statistics The mean and standard deviation (SD) were used for statistical analysis. SPSS 18.0 software was used for statistical analysis. 0.05 was considered statistically significant. All experiments were repeated three times. 9. Results 9.1. HYZ compound library and target library Through database and literature retrieval, the OB and DL values of the compounds were limited, and a total of 15 potential active ingredients were obtained. Information on these compounds is shown in [67]Table 1. The targets of these 15 potential active compounds were obtained by reverse docking in the PharmMapper and SwissTargetPrediction online databases, and the repeated targets were removed. Finally, a total of 640 targets were obtained, which is the target database of 15 potential active compounds in HYZ. Table 1. Potential active compounds in Dioscorea bulbifera. ID Mol ID Name MW OB MF DL Structure C1 MOL000239 Jaranol 314.31 50.83 C[17]H[14]O[6] 0.29 Image 1 C2 MOL000358 beta-sitosterol 414.79 36.91 C[30]H[52]O 0.75 Image 2 C3 MOL000422 Kaempferol 286.25 41.88 C[15]H[10]O[6] 0.24 Image 3 C4 MOL000449 Stigmasterol 412.77 43.83 C₂₉H₄₈O 0.76 Image 4 C5 MOL000546 Diosgenin 414.69 80.88 C[27]H[42]O[3] 0.81 Image 5 C6 MOL000073 ent-Epicatechin 290.29 48.96 C[15]H[14]O[6] 0.24 Image 6 C7 MOL007939 Diosbulbin B 344.39 43.01 C[19]H[20]O[6] 0.7 Image 7 C8 MOL000096 (−)-Catechin 290.29 49.68 C[15]H[14]O[6] 0.24 Image 8 C9 MOL009772 3,5,3′-trimethoxyquercetin 374.37 37.83 C[18]H[16]O[7] 0.44 Image 9 C10 MOL009783 Diosbuibin I 506.59 37.93 C[29]H[30]O[8] 0.86 Image 10 C11 MOL009788 Diosbulbin A 376.44 39.52 C[20]H[24]O[7] 0.65 Image 11 C12 MOL009789 Diosbulbin C 362.41 65.87 C[19]H[22]O[7] 0.6 Image 12 C13 MOL009794 Diosbulbin H 418.53 55.62 C[23]H[30]O[7] 0.7 Image 13 C14 MOL000098 Quercetin 302.25 46.43 C[15]H[10]O[7] 0.28 Image 14 C15 MOL009800 Kryptogenin 430.69 35.11 C[27]H[42]O[4] 0.81 Image 15 [68]Open in a new tab MW, Molecular weight; MF, Molecular formula. 9.2. Target library for TC The gene expression of “[69]GSE3678” was analyzed, and the DEGs between thyroid tissues of TC patients and normal thyroid tissues were deleted and selected. Finally, 668 DEGs were obtained, among which 402 genes were upregulated and 266 were downregulated ([70]Fig. 1A). Finally, 50 overlapping genes were obtained by comparing these DEGs with predicted targets of potential active components of HYZ ([71]Fig. 1B). Fig. 1. [72]Fig. 1 [73]Open in a new tab Differently expressed genes screening. (A) The volcano plot of the differently expressed genes. (B) Overlapping genes of thyroid cancer (TC) and HYZ. Bottle green circle is the predictive gene in HYZ (590), the aqua circle is the DEGs in thyroid tissue of thyroid Cancer patients and normal patients (618), and the overlap in the middle is the overlapping genes between the two genes (50). 9.3. PPI of HYZ against TC The interaction between overlapping genes was analyzed with the help of the String online database and Cytoscape software, and the PPI network graph was constructed. As shown in [74]Fig. 2A, there were 38 nodes in the network and 97 interaction relationships. Through the analysis of hub genes in the network, it was found that FN1 JUN AR RUNX2 FOS BMP2 MET and other targets had a large degree, suggesting that these genes were important targets for HYZ to play anti-TC roles ([75]Fig. 2B). Fig. 2. [76]Fig. 2 [77]Open in a new tab Network analysis. (A) The PPI network of overlapping genes. (B) The 19 hub genes of higher interplay. (C) Network diagram of drug-component-target-disease network diagram (DCTD). 9.4. Analysis of the DCTD Cytoscape software was used to construct a DCTD, as shown in [78]Fig. 2C. There were 63 nodes in the network diagram, including 15 compounds and 48 protein targets. In addition, the network has 194 edges, and there is not a single correspondence between the target and compound. One target corresponds to multiple compounds, and one compound can act on multiple targets, suggesting that HYZ can play an anti-TC role by acting on multiple targets through multiple components. Further analysis of the structure and topological parameters of the network diagram showed that C1, C14, C11, C3, C6, C8, C5, C9 and other compounds had the most targets, suggesting that these compounds played a core role in the network diagram, indicating that these compounds were the main active components of HYZ against TC ([79]Table 2). Table 2. Topological parameters of compound components in the drug-component-target-disease network diagram (Top 15). Name Degree Average Shortest PathLength Betweenness Centrality Closeness Centrality Neighborhood Connectivity C1 24 2.140625 0.049463 0.467153 6.75 C14 18 2.265625 0.036134 0.441379 6.125 C11 18 2.203125 0.041581 0.453901 6.222222 C3 17 2.328125 0.026258 0.42953 6.642857 C6 16 2.296875 0.023127 0.435374 6.866667 C8 16 2.296875 0.022408 0.435374 6.866667 C5 12 2.390625 0.021567 0.418301 6.666667 C9 11 2.421875 0.013639 0.412903 7.272727 C2 10 2.453125 0.015284 0.407643 6.4 C13 10 2.484375 0.011212 0.402516 6.333333 C15 10 2.453125 0.013949 0.407643 7.5 C10 9 2.484375 0.014063 0.402516 6 C4 8 2.515625 0.008813 0.397516 7.375 C7 8 2.515625 0.010184 0.397516 6.875 C12 7 2.546875 0.008027 0.392638 6.428571 [80]Open in a new tab 9.5. GO function and KEGG signal pathway enrichment analysis R language software was used for GO function enrichment analysis of overlapping genes, and the analysis results are shown in [81]Fig. 3A and B and [82]Table 3. These overlapping genes were enriched in 591 GO functions, among which 520 were biological processes (BP), mainly including cellular response to oxidative stress; cellular response to chemical stress; hormone metabolic process; response to reactive oxygen species; cellular response to reactive oxygen species; response to oxidative stress; olefinic compound metabolic process; cyclic nucleotide catabolic process; cellular hormone metabolic process; and C21-steroid hormone metabolic process. In addition, 12 cellular components (CC) were accumulated, including platelet alpha granule lumen; basal part of cell, invadopodium; secretory granule lumen; and cytoplasmic vesicle lumen. There were 59 molecular functions (MFS), mainly including protease binding; 3′,5′-cyclic-nucleotide phosphodiesterase activity; cyclic-nucleotide phosphodiesterase activity; oxidoreductase activity, acting on the CH–OH group of donors, NAD or NADP as acceptors; oxidoreductase activity, acting on the CH–OH group of donors; monocarboxylic acid binding; bile acid binding; alditol: NADP+ 1-oxidoreductase activity; 3′,5′-cyclic-GMP phosphodiesterase activity; and monooxygenase activity. These results suggest that HYZ plays an important role in TC by regulating the oxidative stress response and metabolism. The histogram ([83]Fig. 3A) and bubble diagram ([84]Fig. 3B) of GO enrichment are shown in [85]Fig. 3. Fig. 3. [86]Fig. 3 [87]Open in a new tab GO enrichment. (A) GO functional enrichment analysis represented in bar chart. (B) GO functional enrichment analysis represented in bubble diagram. Table 3. KEGG pathway enrichment analysis. ID Description Gene Ratio Bg Ratio P value p. adjust Q value hsa04151 PI3K-Akt signaling pathway 13/47 354/8108 6.15E-08 1.08E-05 7.38E-06 hsa04510 Focal adhesion 9/47 201/8108 1.78E-06 0.000156783 0.000106897 hsa04512 ECM-receptor interaction 6/47 88/8108 1.03E-05 0.000606736 0.000413683 hsa04933 AGE-RAGE signaling pathway in diabetic complications 6/47 100/8108 2.16E-05 0.000864446 0.000589395 hsa00140 Steroid hormone biosynthesis 5/47 61/8108 2.46E-05 0.000864446 0.000589395 hsa05222 Small cell lung cancer 5/47 92/8108 0.000177551 0.005101481 0.003478283 hsa04913 Ovarian steroidogenesis 4/47 51/8108 0.000202900 0.005101481 0.003478283 hsa04010 MAPK signaling pathway 8/47 294/8108 0.000248934 0.005476545 0.003734008 hsa01521 EGFR tyrosine kinase inhibitor resistance 4/47 79/8108 0.001083577 0.021189959 0.014447699 hsa05235 PD-L1 expression and PD-1 checkpoint pathway in cancer 4/47 89/8108 0.001688007 0.028011339 0.019098640 hsa04658 Th1 and Th2 cell differentiation 4/47 92/8108 0.001907410 0.028011339 0.019098640 hsa04657 IL-17 signaling pathway 4/47 94/8108 0.002064218 0.028011339 0.019098640 hsa05171 Coronavirus disease - COVID-19 6/47 232/8108 0.002069019 0.028011339 0.019098640 hsa01522 Endocrine resistance 4/47 98/8108 0.002404233 0.028797784 0.019634852 hsa04020 Calcium signaling pathway 6/47 240/8108 0.002454357 0.028797784 0.019634852 hsa04660 T cell receptor signaling pathway 4/47 104/8108 0.002983900 0.032333384 0.022045489 hsa04928 Parathyroid hormone synthesis, secretion and action 4/47 106/8108 0.003196652 0.032333384 0.022045489 hsa04659 Th17 cell differentiation 4/47 107/8108 0.003306823 0.032333384 0.022045489 hsa05145 Toxoplasmosis 4/47 112/8108 0.003896789 0.036096573 0.024611300 [88]Open in a new tab The KEGG pathway enrichment analysis of overlapping genes revealed that these genes were enriched in 19 obvious pathways, mainly including the PI3K-Akt signaling pathway, focal adhesion, ECM-receptor interaction, AGE-RAGE signaling pathway in diabetic complications, steroid hormone biosynthesis, small cell lung cancer, ovarian steroidogenesis, MAPK signaling pathway, EGFR tyrosine kinase inhibitor resistance, PD-L1 expression and PD-1 checkpoint pathway in cancer. These results indicate that active components in HYZ may synergically play an anti-TC role in multiple signaling pathways. The histogram ([89]Fig. 4A) and bubble diagram ([90]Fig. 4B) of KEGG enrichment are shown in [91]Fig. 4. Fig. 4. [92]Fig. 4 [93]Open in a new tab KEGG enrichment. (A) Histogram of KEGG enrichment results; (B) Bubble diagram of KEGG enrichment results. 9.6. HYZ inhibits the viability of TC cells The preliminary results of network pharmacology suggested that the antithyroid cancer effect of HYZ was related to the PI3K-Akt and focal adhesion pathways. Therefore, we established an in vitro model of TC cells (KMH-2 cells) to further study the effect of HYZ on TC cells in vitro and explore the possible mechanism. As shown in [94]Fig. 5 A-C, the cell viability of KMH-2 cells was significantly inhibited after the addition of HYZ within 72 h, and the inhibition showed a significant time- and dose-dependent relationship. In subsequent experiments, we selected concentrations of 20 μg/mL, 40 μg/mL, and 80 μg/mL for 24 h as the method of HYZ administration. Fig. 5. [95]Fig. 5 [96]Open in a new tab Effects of the main constituents in HYZ on proliferation of KMH-2 cells in 24 h (A), 48 h (B), 72 h (C). 9.7. HYZ inhibits the migration of TC cells As shown in [97]Fig. 6 A & 6B, with increasing time, the scratches of each group of cells gradually healed. At 24 h, the scratches prepared by pipetting gun tips in the control group almost healed completely, but the cell scars in the positive group and the HYZ group did not heal completely, suggesting that HYZ can significantly inhibit the scratch healing ability of KMH-2 cells, and this inhibition showed a significant time and dose dependence. Similarly, Transwell assays revealed a significant reduction in the number of cells migrating in the positive and HYZ groups compared with the control group ([98]Fig. 7A). These results suggest that HYZ may inhibit the migration of KMH-2 cells. Fig. 6. [99]Fig. 6 [100]Open in a new tab HYZ inhibits the migration of KMH-2 cells. (A) The represented pictures of wound-healing assays of KMH-2 cells. (B) The statistical chart of the would-healing results. **p < 0.01 vs. control. Fig. 7. [101]Fig. 7 [102]Open in a new tab Effects of HYZ on the migration and invasion ability of KMH-2 cells using Transwell chamber. (A) The results of the migration experiment; (B) The results of the invasion experiment. **p < 0.01 vs. control. 9.8. HYZ inhibits the invasion of TC cells In addition, the effect of HYZ on the invasion ability of KMH-2 cells was determined by transwell assay. As shown in [103]Fig. 7B, compared with the control group, the number of invaded cells was significantly reduced in both the positive and HYZ groups at the end of the intervention, indicating that HYZ has a similar effect to the positive drug and can inhibit the invasion of TC cells. 9.9. HYZ can induce apoptosis of KMH-2 cells To further explore the specific mechanism of the effect of HYZ on KMH 2 cell viability, flow cytometry was used to determine whether the antitumor effect of HYZ was related to apoptosis. As shown in [104]Fig. 8 A-E, the apoptosis rate was significantly increased after HYZ intervention compared with the control group. These results indicated that the effect of HYZ on KMH-2 cell viability was related to the induction of apoptosis. Fig. 8. [105]Fig. 8 [106]Open in a new tab Apoptotic induction activities of HYZ on KMH-2 cells. Cells were treated with different concentrations of HYZ for 24 h, and the apoptotic effects of HYZ were detected by flow cytometric analysis with Annexin V-FITC and PI staining. (A)–(D) represents the flow charts of Normal, Low dose, Middle dose, and High dose of HYZ; (E) Statistical chart of the flow cytometric analysis. **p < 0.01 vs. control. 9.10. HYZ can regulate the expression of apoptosis-related proteins in KMH-2 cells Previous flow cytometry results suggested that HYZ could induce apoptosis of KMH-2 cells. Therefore, we measured the expression of apoptosis-related proteins in the cells. As shown in [107]Fig. 9A–D, HYZ intervention significantly increased the expression of Bad and inhibited its phosphorylation in KMH-2 cells. Similarly, the cleaved-Caspase 3 (C-Caspase 3) levels increased significantly after treatment with different doses of HYZ. Fig. 9. [108]Fig. 9 [109]Open in a new tab The Western blot (WB) assay was used to detect the effect of HYZ intervention on protein expression in cells. (A) Representative strip images for the WB assay. (D)–(K) is the statistical charts of C-Caspase-3, Bad, p-Bad, PI3K, p-PI3K, Akt, p-Akt, FAK, p-FAK, and E-cad, respectively. C-Caspase-3 means cleaved Caspase-3. *p < 0.05, **p < 0.01 vs. control. 9.11. HYZ inhibited the PI3K-Akt and focal adhesion pathways in KMH-2 cells The experimental results of network pharmacology suggested that the anti-TC effect of HYZ was related to the PI3K-Akt signaling pathway and focal adhesion pathway. We also found that HYZ could significantly inhibit the migration and invasion of KMH-2 cells and induce apoptosis. To further study the molecular mechanism of the antitumor effect of HYZ, the expression of related proteins in the PI3K-Akt and focal adhesion signaling pathways in KMH 2 cells was determined by WB, and the results are shown in [110]Fig. 9A and E-H. Compared with the control group, phosphorylation of PI3K and AKT was significantly inhibited after HYZ treatment, indicating that HYZ inhibited the activation of the PI3K-Akt pathway in KMH-2 cells. Focal adhesion kinase (FAK) is a multifunctional nonreceptor tyrosine kinase that mediates adhesion between cells and the extracellular matrix, as well as a key link in the focal adhesion pathway. Therefore, we examined the expression of FAK in cells and found that HYZ seemed to have no effect on the expression of proto-FAK, but p-FAK expression in cells decreased significantly after intervention ([111]Fig. 9 A & I–K). Existing studies have shown that E-CAD is a key protein in the metastasis and invasion of many tumor cells, including TC cells. Therefore, we also measured the expression of E-CAD protein in KMH-2 cells after HYZ intervention. As expected, HYZ significantly upregulated the expression of E-CAD protein ([112]Fig. 9 A & I–K). 10. Discussion TC is currently one of the most common endocrine malignant tumors in the world. The pathogenesis of TC is still unclear, and it is generally believed to be related to the levels of sodium iodine transporters and estrogen. Current clinical treatment methods have difficulty reducing the recurrence rate, while the accompanying treatment side effects and high treatment costs have greatly increased the physical burden and economic burden of patients. Therefore, it is necessary to find more efficient and safer complementary or alternative drugs. With the gradual deepening of scientific research, drugs from natural sources are increasingly sought after by researchers as candidates for antitumor drugs [[113]30]. Tuber Dioscoreae Bulbiferae (Huang Yaozi, HYZ) is a common medicine used in TCM to treat sore throat, goiter, hematemesis, hemoptysis, and other diseases. It has remarkable clinical effects, low price and few side effects. At the same time, clinical studies have reported that HYZ has a good therapeutic effect on patients with thyroid tumors and TC. In addition, modern studies have also found that HYZ has a good inhibitory effect on the growth of TC cells, but these studies are not in-depth enough to explain the molecular mechanism of HYZ in the treatment of TC. Currently, accumulating herbal medicines are being developed as candidate drugs for treating various difficult miscellaneous diseases, such as cancer [[114]31], neurodegenerative diseases [[115]32], cardiovascular diseases [[116]33], and diabetes mellitus [[117]34]. In this study, network pharmacology combined with bioinformatics methods was used to explore the molecular mechanism by which HYZ exerts its anti-TC effect. The results showed that HYZ could exert anti-TC effects through multicomponent and multitarget methods, and compounds such as C1, C14, C11, C3 and C6 might be the main active components of HYZ against TC. In addition, our results also suggest that the specific mechanism of HYZ against TC may be mainly related to regulating the PI3K-Akt pathway and focal adhesion pathway. To verify the reliability of network pharmacology and bioinformatics experiments, we designed corresponding antitumor experiments in vitro. The CCK8 assay showed that HYZ inhibited the proliferation of KMH-2 cells in a dose-dependent manner. To verify whether the effect of HYZ on cell viability was related to apoptosis induction, the apoptosis of KMH-2 cells after HYZ intervention was determined by flow cytometry. As expected, HYZ could significantly induce apoptosis of KMH-2 cells. Similarly, at the protein level, our study also found that HYZ can inhibit the phosphorylation of Bad and promote the activation of Caspase 3, the executor of apoptosis. It is well known that the PI3K-AKT pathway is closely related to cell growth, proliferation and death [[118]35,[119]36], and network pharmacology results suggest that the antitumor effect of HYZ is related to the regulation of the PI3K-AKT pathway. Therefore, we also determined the expression of PI3K and AKT in KMH-2 cells after HYZ intervention and found that HYZ could significantly inhibit the phosphorylation of PI3K and AKT, confirming that the antitumor effect of HYZ was related to the inhibition of the activation of the PI3K-AKT pathway. In addition, it has been reported that the focal adhesion pathway plays an important role in the migration and invasion of tumor cells [[120]37]. Therefore, we also studied the effect of HYZ on the migration and invasion ability of KMH-2 cells. Fortunately, different doses of HYZ significantly inhibited the migration and invasion of KMH-2 cells. FAK and E-CAD are proteins that play a key role in the metastasis and invasion of many tumor cells, as well as a key link in the focal adhesion pathway [[121]38,[122]39]. We also found that the expression of p-FAK decreased significantly after HYZ intervention, but the expression of E-CAD protein was significantly upregulated, suggesting that HYZ can inhibit the migration and invasion of thyroid cancer cells by regulating FAK phosphorylation and E-CAD expression in the focal adhesion pathway. 11. Conclusion In this study, based on network pharmacology and bioinformatics combined with antitumor experiments in vitro, we investigated the inhibitory effect and mechanism of HYZ on the growth, migration, and invasion of KMH-2 human TC cells and found that HYZ could inhibit the growth, migration, and invasion of KMH-2 cells. Importantly, these inhibitory effects are consistent with previous network pharmacology and bioinformatics predictions, that is, regulation of the PI3K-AKT pathway and focal adhesion pathway. The role of prophylactic central neck dissection is still controversial. In more destructive surgery, interventions are affected by more severe complications [[123][40], [124][41], [125][42]]. However, our present study is mainly based on in silico and in vitro experiments and is just a "first level study"; thus, in the future, more work should be devoted to evaluating the anti-TC effects of HYZ with a tumor-bearing mouse model and further clinical studies in humans. In addition, more studies should be performed to explore the possible toxicities and real active substances or marker constituents in HYZ. In conclusion, our study provides a reference for further animal or clinical studies investigating the effectiveness and molecular mechanisms of HYZ against thyroid cancer. Ethics approval and consent to participate Not applicable. Author contribution statement Ziqi Liu, Lian Zhong, Chao Chen: Conceived and designed the experiments; Performed the experiments; Contributed reagents, materials, analysis tools or data; Wrote the paper.

; Lingyu Wang: Performed the experiments; Contributed reagents, materials, analysis tools or data.

; Meiyan Li: Analyzed and interpreted the data.

Funding statement Chao Chen was supported by Sichuan Province Science and Technology Support Program {2019YFS0440}. Data availability statement Data included in article/supp. material/referenced in article. 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. Footnotes ^Appendix A Supplementary data related to this article can be found at [126]https://doi.org/10.1016/j.heliyon.2023.e18886. Appendix A. Supplementary data The following is the supplementary data related to this article: Multimedia component 1 [127]mmc1.pdf^ (209.7KB, pdf) References