Abstract The global prevalence of obesity is a pressing health issue, increasing the medical burden and posing significant health risks to humans. The side effects and complications associated with conventional medication and surgery have spurred the search for anti-obesity drugs from plant resources. Previous studies have suggested that Artemisiae argyi Folium (Aiye) water extracts could inhibit pancreatic lipase activities, control body weight increase, and improve the plasma lipids profile. However, the exact components and mechanisms were not precisely understood. Therefore, this research aims to identify the chemical profile of Aiye and provide a comprehensive prediction of its anti-obesity mechanisms. The water extract of Aiye was subjected to LC-MS analysis, which identified 30 phenolics. The anti-obesity mechanisms of these phenolics were then predicted, employing network pharmacology and molecular docking. Among the 30 phenolics, 21 passed the drug-likeness screening and exhibited 486 anti-obesity targets. The enrichment analysis revealed that these phenolics may combat obesity through PI3K-Akt signaling and MAPK, prolactin, and cAMP signaling pathways. Eight phenolics and seven central targets were selected for molecular docking, and 45 out of 56 docking had a binding affinity of less than −5 kcal/mol. This research has indicated the potential therapy targets and signaling pathways of Aiye in combating obesity. Keywords: Artemisiae argyi Folium, obesity, LC-MS, network pharmacology, molecular docking 1. Introduction Obesity is caused by the pathological excessive deposit of body fat [[28]1]. A body mass index (BMI) higher than 30 kg/m^2 indicates the clinical sign of adult obesity [[29]2]. Nowadays, obesity has almost become a global epidemic, except in some African and Asian countries [[30]3]. The 2022 data from the Centers for Disease Control and Prevention (CDC) reported that obese adults accounted for no less than 35% of the 22 states of the United States [[31]4]. The global prevalence of overweight and obesity for both boys and girls will exceed 25% by 2030 [[32]5]. Obesity is positively correlated with various diseases, including diabetes mellitus (type II), high blood pressure, cardiovascular diseases, certain kinds of cancer, osteoarthritis, fatty liver, gallbladder diseases, sleep apnea, etc. [[33]3,[34]6,[35]7,[36]8]. The epidemic of obesity slows down economic development while increasing medical costs worldwide. Meanwhile, obesity decreases work efficiency and reduces the quality of life of the people suffering from it [[37]9]. The mechanisms of anti-obesity drugs include diminishing the appetite, inhibiting the absorption of fat, and increasing energy expenditure [[38]2]. The side effects of those anti-obesity drugs include diarrhea, steatorrhea, malnutrition, pancreatitis, gastroparesis, cholelithiasis, nephrolithiasis, severe hypertension, anxiety, depression, cognitive impairment, suicidal tendency and behavior, and congenital disabilities [[39]2,[40]7,[41]10,[42]11]. Weight-loss surgery is carried out when medication interventions are not effective in the weight control of severely obese patients, and the complications of weight-loss surgery include postoperative mortality, wound infection, ulcers, intestinal obstruction, deficiency of micronutrients, etc. [[43]8]. Due to the commonly existing side effects or complications of medications or surgery, the development of novel anti-obesity drugs from natural plant sources, which are safer and more effective, becomes a priority. Artemisiae argyi Folium (Aiye) has been used for thousands of years as an herbal medicine to treat gynecological, skin, digestive, and respiratory diseases. Pharmacological research has shown that it has anti-virus, anti-bacteria, anti-oxidative, anti-inflammatory, anti-dermatitis, gastro-protective, and anti-cancer effects, mainly due to the presence of essential oils, flavonoids, and organic acids [[44]12,[45]13,[46]14,[47]15,[48]16,[49]17,[50]18]. Pharmacological research has identified various anti-cancer components of Aiye, including jaceosidin, irigenin, luteolin, eriodictyol, apigenin, eupatilin, casticin, and artemetin [[51]19,[52]20,[53]21,[54]22,[55]23,[56]24,[57]25,[58]26,[59]27,[60]28, [61]29]. In obesity, the secretion of adipokines by adipose tissue was dysregulated, characterized by the up-regulation of proinflammation adipokines and the down-regulation of anti-inflammation adipokines [[62]30]. Furthermore, the prolactin signaling pathway participated in adipogenesis and energy balance [[63]31]. The MAPK pathway is involved in appetite regulation, thermogenesis, adipogenesis, and glucose homeostasis, which is tightly connected with obesity [[64]10]. The PI3K-Akt pathway participated in the proliferation and growth of cells, while the cAMP signaling pathway involved insulin secretion [[65]10]. All those pathways are promising for the treatment of obesity. The anti-inflammation and anti-oxidation effects of Aiye’s phenolic acids and flavonoids indicated that Aiye may combat the progress of obesity. A variety of bioactive components of Aiye possess anti-oxidation and anti-inflammation effects, including 7-hydroxycoumarin, daphnetin, nepetin, 5,6,4′-trihydroxy-7,3′-dimethoxyflavone, apigenin, kaempferol, hispidulin, jaceosidin, irigenin, and cirsimaritin [[66]32,[67]33,[68]34,[69]35,[70]36,[71]37,[72]38,[73]39,[74]40,[75]41, [76]42,[77]43,[78]44,[79]45,[80]46,[81]47,[82]48,[83]49,[84]50]. Furthermore, apigenin ameliorates insulin resistance induced by obesity, while cirsimaritin possesses anti-diabetes effects and ameliorates liver disease caused by a high-fat diet [[85]49,[86]51,[87]52]. The scopoletin attenuates the toxicity of high glucose [[88]53,[89]54]. The chlorogenic acid, neochlorogenic acid, isochlorogenic acid A, and isochlorogenic acid B of Aiye inhibit the activities of pancreatic lipase [[90]55]. The administration of water extracts derived from Aiye significantly decreased the body weight of high-fat-diet-fed mice, enhanced their plasma lipid profile, and mitigated glucose intolerance and hepatic steatosis [[91]56]. However, the exact components of Aiye’s water extracts that confer anti-obesity effects and the exact anti-obesity mechanisms remain unclear. The current limited research provided that Aiye may confer anti-obesity effects through its water-soluble phenolics. However, there was a lack of research that fully elucidated the anti-obesity mechanisms of the exact phenolics in the water extract of Aiye. The current research identified the phenolic profile of Aiye and elucidated its anti-obesity mechanisms by combining liquid chromatography-mass spectrometry (LC-MS), network pharmacology, and molecular docking. Network pharmacology has been widely applicable in revealing the therapeutic mechanisms of plant secondary metabolites on various diseases, including allergic rhinitis, inflammatory response, rheumatoid arthritis, etc. [[92]57,[93]58,[94]59]. As for anti-obesity, molecular docking technology is widely applied to evaluate the interaction between compounds and targets by visualizing the intermolecular forces and presenting the binding affinity between compounds and targets [[95]60,[96]61,[97]62]. The water-soluble bioactive components presented higher anti-oxidative capacity than nutraceutical mixtures’ lipid-soluble components [[98]63]. Therefore, in the current research, the phenolics of Aiye were extracted by water. Firstly, phenolics were extracted by water and then identified by LC-MS. Secondly, the identified phenolics were screened by Lipinski rules to ensure their bioavailability and drug-likeness. Thirdly, the targets of phenolics and obesity were collected from different databases, and the protein–protein-interaction (PPI) network of overlapped targets between phenolics and obesity was constructed. The core and central targets were selected for Gene Ontology (GO), Kyoto Encyclopedia of Genes and Genomes (KEGG) enrichment analysis, and molecular docking. GO and KEGG analysis results predicted the anti-obesity mechanisms. The binding affinity and docking details between key phenolics and central targets were achieved through molecular docking. Furthermore, the Aiye-components-targets-obesity-KEGG signaling pathways network was constructed to present the connection between them. The research results predicted the possible anti-obesity mechanisms of Aiye and revealed the possibility of developing anti-obesity drugs from Aiye. 2. Materials and Methods 2.1. LC-MS Analysis Qichun County Shenzhou Qiai Biotechnology Co., Ltd. (Qichun, China) provided the Aiye sample, cultivated in Qichun county and harvested during the Dragon Boat Festival in 2022. The fresh Aiye was sun-dried and stored in the warehouse for one year. Based on the traditional Chinese medicine theory, fresh Aiye is unsuitable for medical use. Therefore, after harvest, the sun-dried Aiye was stored for at least one year until further processing and application. A voucher specimen (ID: UIC-FS-2023-03-01) was stored at T8-508 in Food Science Laboratories in BNU-HKBU United International College. Previous studies have verified the anti-obesity effects of Aiye water extract. Therefore, the current research aims to identify the bioactive components and explore the anti-obesity mechanisms from the water extract of Aiye. A total of 100.0 g of Aiye was washed twice to remove the dust on it. After soaking in one liter of distilled water for 30 min, it was put into four liters of boiling water for 20 min. After cooling down, the decoction was filtered by cotton gauze. The filtrate was centrifuged under 10,614× g for 10 min at 25 °C. After centrifugation, the supernatant passed through the 0.22 μm microfiltration membrane. After that, the rotary evaporation machine (Shanghai Yarong Biochemistry Instrument Factory, RE-52AA, Shanghai, China) concentrated the water to decrease the volume. Finally, the concentrated filtrate was frozen at −80 °C and freeze-dried by the freeze-drying machine (Scientz-18N/A, Ningbo, China). The comprehensive chemical profile analysis of the freeze-dried water extract was performed by ultra-performance liquid chromatography–electrospray ionization–tandem mass spectrometry (UPLC-ESI-MS/MS, ExionLC™ AD, [99]https://sciex.com.cn/, accessed on 7 November 2023) [[100]64]. After the first lyophilization, the extract powder was kept in the refrigerator until the LC-MS analysis. Before the LC-MS analysis, the extract powder was lyophilized again to eliminate the moisture caused during storage and preparation. Specifically, the sample was put into a lyophilizer (Scientz-100F, Ningbo, China) for freeze-drying, and then it was ground by a grinder (MM 400, Retsch, Haan, Germany) for 1.5 min at a frequency of 30 Hz. An electronic balance (MS105DM) weighed fifty milligrams of sample powder. Subsequently, 1200 microliters of 70% methanol were mixed with the powder for sample extraction. The 70% methanol solution contained 1 mg/L of internal standard to monitor whether the components were stable during the extraction and whether the instruments were accurate during the detection. The 70% methanol solution with an internal standard was prepared in two steps. Firstly, the internal standard stock solution was prepared by dissolving 5 mg of 2-chlorophenylalanine (J&K Scientific, Beijing, China, CAS: 14091-11-3) in 5 mL of methanol. Secondly, 1 mL of the stock solution was added to 1 L of 70% methanol. The solution was stored at −20 °C before being used. Every 30 min, the mixture was vortexed for 30 s and vortexed six times. After that, the mixture was centrifuged at 16,099× g for 3 min. A 0.22 μm microfiltration membrane filtered the supernatant and was ready for UPLC-ESI-MS/MS analysis. The parameter setting of the liquid chromatography section was adapted from the method described previously [[101]65]. The column was Agilent SB-C18 (1.8 μm, 2.1 mm × 100 mm). The mobile phase A was 0.1% formic acid in ultrapure water solution, and the mobile phase B was 0.1% formic acid in acetonitrile solution. From 0.00 min to 9 min, the ratio of the mobile phase B was linearly increased from 5% to 95%. After that, the ratio of mobile phase B was kept at 95% until 10 min. From 10 to 11.10 min, the ratio of the mobile phase B decreased from 95% to 5%. The ratio of the mobile phase B was kept as 5% until 14 min. The flow rate of the mobile phase was kept as 0.35 mL/min. The column temperature was set as 40 °C. The injection volume of sample extraction was two micro-liters. The parameter setting of the mass spectrometry was adapted from the method described previously [[102]66]. The temperature of ESI was 500 °C. The spray voltage for the positive model was 5500 V, while it was −4500 V for the negative model. The gas I of ionization source, gas II of ionization source, and the curtain gas were 50, 60, and 25 psi, respectively. Multiple reaction monitoring (MRM) was used in the QQQ scanning of metabolites. The collision gas (N[2]) was set as a medium. The de-clustering potential (DP) and collision energy (CE) for each MRM precursor and product ion were optimized. The precursor and product ions were selected to monitor the relevant metabolites at the elution time. The mass spectrum data, such as the secondary spectra, retention time, and accurate mass of precursor and product ions of each metabolite, were matched to the database of Metware (Wuhan Metware Biotechnology Co., Ltd., Wuhan, China). The tolerance of the secondary spectrum was set as 20 ppm, and the retention time offset was less than 0.2 min. In this way, the chemical profile of the lyophilized water extract of Aiye was identified. 2.2. Drug-Likeness Evaluation Firstly, the canonical SMILES of the 30 components were obtained from the PubChem database (retrieved from [103]https://pubchem.ncbi.nlm.nih.gov/, accessed on 5 March 2024). Secondly, the canonical SMILES were input into the SwissADME webpage (retrieved from [104]http://www.swissadme.ch/, accessed on 5 March 2024) [[105]67]. From the SwissADME webpage, the molecular weight, the number of hydrogen bond acceptors and donors, the lipophilicity, and the bioavailability of those components were downloaded. Finally, the Lipinski rules were used to screen the drug-likeness properties of those components [[106]68,[107]69] 2.3. Prediction of Protein Targets of Components The canonical SMILES of each identified component were input into the inquiry box of both the STP (retrieved from [108]http://www.swisstargetprediction.ch/, accessed on 5 March 2024) [[109]26] and the SEA database (retrieved from [110]https://sea.bkslab.org/, accessed on 5 March 2024) [[111]70]. Homo sapiens was selected as the species for target prediction. The targets of each component from the two databases were combined, and the repeated targets were removed. The total targets of all the elements were combined for the follow-up analysis. 2.4. Collection of Obesity Targets The obesity targets were collected from five databases, which were DrugBank (retrieved from [112]https://go.drugbank.com/, accessed on 31 August 2023) [[113]71], Uniprot (retrieved from [114]https://www.uniprot.org/, accessed on 31 August 2023), OMIM (retrieved from [115]https://www.omim.org/, accessed on 1 September 2023) [[116]72], DisGeNET (retrieved from [117]https://www.disgenet.org/search, accessed on 1 September 2023) [[118]73], and GeneCards (retrieved from [119]https://www.genecards.org/, accessed on 1 September 2023) [[120]74,[121]75]. Only the protein-coding gene targets were selected for the subsequent analysis. 2.5. Targets Overlapping between Components and Obesity The targets of components and obesity were plotted by the Venny 2.1 online tool (retrieved from [122]https://bioinfogp.cnb.csic.es/tools/venny/, accessed on 5 March 2024) [[123]76]. The overlapped area indicated the potential anti-obesity targets of the identified components. 2.6. Construction of PPI Network The multiple anti-obesity targets of components were visualized by STRING 12.0 (retrieved from [124]https://string-db.org/, accessed on 5 March 2024) [[125]77]. Homo sapiens was selected for continuing analysis. The database lacked three targets: VEGFA, MDH1, and LDHB. For settings, the PPI network was functionally and physically associated. The edge between the two nodes indicated their associations with each other. All the provided sources of interaction between nodes were selected, including experiments, databases, co-expression, etc. The overlapped nodes were separated so that all could be identified. Two formats of the PPI network were exported: tab-separated values (TSVs) and portable network graphic (PNG) formats, respectively. 2.7. Construction of Interaction Networks Different interaction networks were constructed by Cytoscape 3.10.1 [[126]78] as follows. 2.8. Construction of PPI Network of Aiye Anti-Obesity Targets for the Screening of Core and Central Targets The PPI network containing 483 targets exported from STRING was imported into Cytoscape, and the three independent nodes were excluded: PASK, KLK7, and P2RX3. For the 480 nodes left, the betweenness centrality, closeness centrality, and degree centrality were calculated by CytoNCA [[127]79]. Furthermore, the maximal clique centrality (MCC), maximum neighborhood component (MNC), degree, and closeness were calculated by cytoHubba [[128]80]. 2.9. Construction of Aiye Core Anti-Obesity Targets for Enrichment Analysis The nodes whose betweenness centrality, closeness centrality, and degree centrality were all higher than the median were chosen as the core targets, and the interaction network was visualized by Cytoscape [[129]79,[130]81]. Furthermore, the GO and KEGG enrichment analysis of the core targets was performed by Metascape (retrieved from [131]https://metascape.org/gp/index.html#/main/step1, accessed on 5 March 2024) [[132]82]. 2.10. Construction of Aiye Central Targets for Component–Target Docking The top 10 targets of MCC, MNC, Degree, and Closeness were ranked by cytoHubba, and their overlapping was regarded as the central targets for subsequent docking [[133]81]. 2.11. GO and KEGG Enrichment Analysis The biological processes (BPs), cellular components (CCs), molecular functions (MFs), and KEGG pathway enrichment analysis of core targets were performed on Metascape (retrieved from [134]https://metascape.org/gp/index.html#/main/step1, accessed on 5 March 2024) [[135]82]. Homo sapiens were the organisms of enrichment analysis. The minimum overlap was 3, the p value cutoff was 0.01, and the minimum enrichment was 1.5. The enrichment analysis results were visualized by bioinformatics (retrieved from [136]https://www.bioinformatics.com.cn/, accessed on 5 March 2024). 2.12. Construction of Aiye-Components-Targets-Obesity-Signaling Pathway Network The interaction network of the Aiye-components-targets-obesity-signaling pathway was visualized by Cytoscape 3.10.1 [[137]78]. Nodes of different categories were assigned various colors and shapes. The components, targets, and pathways were all plotted in a cycle layout according to the degree centrality. The more central location indicated the higher degree centrality of the node. 2.13. Component–Target Docking The identified components, whose degree centrality was higher than the average level, were chosen as the docking ligands with receptor targets. The canonical SMILES of those components were converted into the 2-dimensional (2D) structure by ChemDraw (version 22.0.0). The 2D structure of each component was imported into Chem3D (version 22.0.0) to generate the minimized energy form of the compound and output it in PDB format. The PDB format of central targets was downloaded from the Protein Data Bank database (retrieved from [138]https://www.rcsb.org/, accessed on 5 March 2024). The water was removed from targets, and hydrogen was added to both targets and components by Autodock Tools (version 1.5.7) [[139]83,[140]84]. The results were kept as PDBQT format files. A grid box encompassing the target was established, and its binding domain was determined by utilizing AutoDock Vina (version 1.2.5) [[141]85,[142]86]. The conformation with the lowest binding affinity was kept for each component–target complex. A binding affinity less than −5 kcal/mol indicated a moderate binding affinity, while a solid binding affinity required an affinity lower than −7 kcal/mol [[143]87]. Pymol and Discovery Studio visualized the docking results with an affinity of less than −7 kcal/mol. The heat map of docking results was plotted by bioinformatics (retrieved from [144]https://www.bioinformatics.com.cn/, accessed on 13 November 2023). 3. Results 3.1. The Identification of Phenolics through UPLC-ESI-MS/MS A total of 19.34 g of freeze-dried powder were obtained from the water extract of the 100.0 g dried Aiye sample, resulting in a yield of approximately 19.34%. Before chemical profile analysis, 50 mg of a lyophilized extract was dissolved in 1200 μL of 70% methanol. There were 30 phenolics identified by UPLC-ESI-MS/MS, which belonged to four categories. There were 20 flavonoids, which accounted for 66.67% of the identified phenolics, and 15 of them were flavones, 3 were flavonols, 1 was flavanone, and 1 was isoflavone. Six phenolic acids occupied 20% of the identified phenolics, and three coumarins accounted for 10%. The last one was chromone, which comprised 3.33% of the identified phenolics. The results are presented in [145]Table 1. The secondary mass spectrum of phenolics 1–30 is provided in the [146]Supplementary Materials listed as Figures S1–S30. Table 1. The 30 phenolics in Aiye identified by LC-MS. No. Retention Time (min) Compounds CAS ID Collision Energy (eV) Classification Precursor ion (Da) Product ion (Da) Peak Area 1 2.20 1-Caffeoylquinic acid 1241-87-8 −20 Phenolic acids 353.09 191.06 69,797,865.10 2 2.30 7-Hydroxycoumarin 93-35-6 30 Coumarins 163.04 89.04 1,763,845.40 3 2.50 3,4-Dihydroxybenzoic acid 99-50-3 −18 Phenolic acids 153.02 109.03 1,662,701.81 4 3.10 Salicylic acid 69-72-7 −30 Phenolic acids 137.03 108.02 21,278,543.18 5 3.30 Schaftoside 51938-32-0 47 Flavones 565.16 529.13 7,564,970.69 6 3.40 Caffeic acid 331-39-5 −20 Phenolic acids 179.03 135.05 12,700,935.26 7 3.50 Isoschaftoside 52012-29-0 30 Flavones 565.16 409.09 11,713,194.36 8 3.50 Daphnetin 486-35-1 30 Coumarins 179.03 133.03 4,308,737.58 9 3.80 Vitexin 3681-93-4 −30 Flavones 431.10 311.05 4,787,487.98 10 3.80 Rutin 153-18-4 30 Flavonols 611.16 303.05 917,616.01 11 4.00 Cynarin 30964-13-7 30 Phenolic acids 517.13 163.04 102,685,412.20 12 4.00 Isorhoifolin 552-57-8 30 Flavones 579.17 271.06 10,644,918.47 13 4.00 Isoquercitrin 482-35-9 −30 Flavonols 463.09 300.03 617,610.61 14 4.10 Scopoletin 92-61-5 −30 Coumarins 191.04 176.01 5,348,582.99 15 4.20 Isochlorogenic acid B 14534-61-3 −30 Phenolic acids 515.12 353.09 15,763,996.82 16 5.00 Luteolin 491-70-3 −40 Flavones 285.04 151.00 59,730,853.76 17 5.20 Eriodictyol 552-58-9 −20 Flavanones 287.05 135.04 20,620,576.66 18 5.20 Nepetin 520-11-6 30 Flavones 317.07 302.04 6,988,200.11 19 5.60 5,6,4′-Trihydroxy-7,3′-dimethoxyflavone - 30 Flavones 331.08 298.04 8,576,062.16 20 5.70 Apigenin 520-36-5 30 Flavones 271.06 153.02 2,783,945.85 21 5.70 Kaempferol 520-18-3 −30 Flavonols 285.04 151.00 539,429.30 22 5.80 Hispidulin 1447-88-7 30 Flavones 301.07 286.05 3,710,550.04 23 5.80 Capillarisin 56365-38-9 30 Chromone 317.07 302.04 1,337,073.68 24 5.90 Jaceosidin 18085-97-7 30 Flavones 331.08 316.06 16,825,514.25 25 5.90 Centaureidin 17313-52-9 30 Flavones 361.09 303.05 3,058,589.54 26 6.00 Irigenin 548-76-5 40 Isoflavones 361.09 310.00 461,645.02 27 6.40 Cirsimaritin 6601-62-3 20 Flavones 315.09 254.06 5,762,441.22 28 6.70 Eupatilin 22368-21-4 30 Flavones 345.10 330.07 35,641,022.22 29 6.90 Casticin 479-91-4 30 Flavones 375.11 299.06 28,620,099.27 30 7.60 Artemetin 479-90-3 30 Flavones 389.12 331.08 8,618,699.94 [147]Open in a new tab Collision energy; negative value means negative ionization mode. 3.2. Drug-Likeness Screening of Phenolics The molecular weight, number of hydrogen bond acceptor and donor, lipophilicity, bioavailability, and polar surface area of the 30 phenolics were obtained from the webpage of SwissADME (retrieved from [148]http://www.swissadme.ch/, accessed on 5 March 2024) [[149]67,[150]68,[151]69], and 21 of them satisfied the criteria; thus, they were selected for subsequent analysis. The details are represented in [152]Table 2. Table 2. The physicochemical properties, lipophilicity, and drug-likeness of the 30 phenolics. No. Compounds Lipinski Rules Lipinski’s Violations Bioavailability Score TPSA (Å^2) MW HBA HBD MLogP <500 <10 ≤5 ≤4.15 ≤1 > 0.1 <140 1 1-Caffeoylquinic acid 354.31 9 6 −1.05 1 0.11 164.75 2 7-Hydroxycoumarin 162.14 3 1 1.04 0 0.55 50.44 3 3,4-Dihydroxybenzoic acid 154.12 4 3 0.40 0 0.56 77.76 4 Salicylic acid 138.12 3 2 0.99 0 0.85 57.53 5 Schaftoside 564.49 14 10 −3.97 3 0.17 250.97 6 Caffeic acid 180.16 4 3 0.70 0 0.56 77.76 7 Isoschaftoside 564.49 14 10 −3.97 3 0.17 250.97 8 Daphnetin 178.14 4 2 0.45 0 0.55 70.67 9 Vitexin 432.38 10 7 −2.02 2 0.55 181.05 10 Rutin 610.52 16 10 −3.89 3 0.17 269.43 11 Cynarin 516.45 12 7 −0.35 3 0.11 211.28 12 Isorhoifolin 578.52 14 8 −2.96 3 0.17 228.97 13 Isoquercitrin 464.38 12 8 −2.59 2 0.17 210.51 14 Scopoletin 192.17 4 1 0.76 0 0.55 59.67 15 Isochlorogenic acid B 516.45 12 7 −0.35 3 0.11 211.28 16 Luteolin 286.24 6 4 −0.03 0 0.55 111.13 17 Eriodictyol 288.25 6 4 0.16 0 0.55 107.22 18 Nepetin 316.26 7 4 −0.31 0 0.55 120.36 19 5,6,4’-Trihydroxy-7,3’-dimethoxyflavone 330.29 7 3 −0.07 0 0.55 109.36 20 Apigenin 270.24 5 3 0.52 0 0.55 90.90 21 Kaempferol 286.24 6 4 −0.03 0 0.55 111.13 22 Hispidulin 300.26 6 3 0.22 0 0.55 100.13 23 Capillarisin 316.26 7 3 0.37 0 0.55 109.36 24 Jaceosidin 330.29 7 3 −0.07 0 0.55 109.36 25 Centaureidin 360.31 8 3 −0.35 0 0.55 118.59 26 Irigenin 360.31 8 3 −0.35 0 0.55 118.59 27 Cirsimaritin 314.29 6 2 0.47 0 0.55 89.13 28 Eupatilin 344.32 7 2 0.17 0 0.55 98.36 29 Casticin 374.34 8 2 −0.12 0 0.55 107.59 30 Artemetin 388.37 8 1 0.11 0 0.55 96.59 [153]Open in a new tab MW, molecular weight (g/mol); HBA, hydrogen bond acceptor; HBD, hydrogen bond donor; LogP, lipophilicity; Bioavailability score, the ability of a drug or other substance to be absorbed and used by the body; TPSA, topological polar surface area. To pass the screening, a compound should simultaneously satisfy the requirements of Lipinski’s violations (≤1), bioavailability score (>0.1), and TPSA (Å^2) (<140). The italic values with strikethrough disobeyed the standards. 3.3. Prediction Targets of Phenolics The targets of the 21 phenolics were collected from STP (retrieved from [154]http://www.swisstargetprediction.ch/, accessed on 5 March 2024) [[155]88] and SEA (retrieved from [156]https://sea.bkslab.org/, accessed on 5 March 2024) [[157]70]. There were 611 protein targets identified. 3.4. Collection Targets of Obesity Eight thousand two hundred twenty-five obesity targets of Homo sapiens were identified. Two thousand eight hundred twenty-one obesity targets were collected from DisGeNET [[158]73], and 111 targets were obtained from DrugBank [[159]71]. Furthermore, there were 7506 obesity targets gained from GeneCards [[160]74,[161]75], while 89 targets were acquired from OMIM [[162]72]. 3.5. The Overlapped Targets between Phenolics and Obesity The 21 phenolics extracted from Aiye, which satisfied the Lipinski rules, possessed 611 targets. The total number of obesity targets identified was 8225. An analysis using Venny 2.1 software revealed 486 common targets between phenolics and obesity ([163]Figure 1A). The name, chemical structure, number of anti-obesity targets (degree centrality), and pharmacological properties of the 21 phenolics were summarized in [164]Table 3. The degree centrality indicated the edges of a node in a network, and the edges connected different nodes. A phenolic was connected to its anti-obesity targets by the edges. In other words, the degree centrality of a phenolic indicated its number of anti-obesity targets. The phenolics whose target number exceeded the average level were selected for target docking. The name and target number of the 21 phenolics are shown in [165]Figure 1B. Figure 1. [166]Figure 1 [167]Open in a new tab (A) The common targets of phenolics and obesity. (B) The name and target number of the 21 phenolics. (C) The PPI network of the 483 nodes. (D) The PPI network formed by the 480 nodes. Table 3. The name, chemical structure, number of anti-obesity targets (degree centrality), and pharmacological properties of the 21 phenolics. Name Chemical Structure Anti-Obesity Targets Pharmacological Properties References