Artificial intelligence-based systematic study on the multidimensional pharmacological activity and molecular mechanism of the active ingredients of Artemisia argyi
-
摘要:
利用人工智能技术探究艾叶中的活性成分的药理活性及潜在作用机制。检索HIT、TCMSP、TCMIO数据库获取199种艾叶的活性成分。综合K最近邻、多层感知机、随机森林、支持向量机算法及基于Lipinski规则及Veber规则的算法模型对艾叶化合物成分进行毒性和口服利用度预测,发现其中14种成分满足无毒性且口服利用度较好。采用合成可行性分数(SAscore)模型对以上14种成分进行可合成性分析,并通过BRICS及RECAP算法分割分子片段。STP和PM数据库挖掘得到406个艾叶核心成分的作用靶点,利用Cytoscape筛出5个核心靶点为SRC、EGFR、PTPN11、HRAS和PDGFRB。GO与KEGG富集分析提示核心靶点涉及808项GO富集分析条目和EGFR酪氨酸激酶抑制剂抵抗、间隙连接、磷脂酶D、JAK/STAT等71条信号通路。分子对接结果显示艾叶活性化合物与SRC、EGFR、PTPN11、HRAS蛋白具有良好的结合能力。细胞实验证实了艾叶活性成分杜香醇在一定浓度范围内能够促进HUVEC细胞增殖,并可促进EGFR蛋白的表达。本研究揭示了艾叶活性成分的药理特性及其潜在的分子机制,为其药用开发奠定了坚实的科学基础。
Abstract:To investigate the pharmacological activities and potential mechanisms of action of the active components in Artemisia argyi with artificial intelligence technology, a search was conducted in the HIT, TCMSP, and TCMIO databases, obtaining 199 active components of A. argyi. A comprehensive set of algorithms, including KNN, MLP, RF, SVM, and models based on Lipinski’s and Veber’s rules, was employed to predict the toxicity and oral bioavailability of A. argyi compounds, identifying 14 components that are non-toxic and have good oral bioavailability. The synthetic accessibility score (SAscore) model was used to analyze the synthetic accessibility of the 14 components mentioned above, and molecular segments were fragmented using BRICS and RECAP algorithms. Mining of the STP and PM databases yielded 406 target proteins for the core components of A. argyi, and Cytoscape was used to screen out 5 core targets: SRC, EGFR, PTPN11, HRAS, and PDGFRB. GO and KEGG enrichment analyses indicated that the core targets were involved in 808 GO enrichment analysis entries and 71 signaling pathways, including EGFR tyrosine kinase inhibitor resistance, gap junction, phospholipase D, and JAK/STAT. Molecular docking results showed that active compounds of A. argyi have a good binding affinity with proteins SRC, EGFR, PTPN11, and HRAS. Cellular experiments have confirmed that ledol, an active component of A. argyi, can promote the proliferation of HUVEC cells within a certain concentration range and can increase the expression of EGFR protein. This study reveals the pharmacological characteristics and potential molecular mechanisms of the active components of A. argyi and lays a solid scientific foundation for its medicinal development.
-
-
Figure 4. Segmentation results of the molecular fragments in Artemisia argyi core ingredients by RECAP and BRICS algorithms. The blue module indicates that both algorithms yield identical segmentation results; the red module indicates that the segmentation results differ between the two algorithms; and the grey module indicates that segmentation could not be performed
Table 1 Toxicity prediction scores for 14 ingredients in Artemisia argyi
Compd. Molecular formula Toxicological prediction Mutagenicity Carcinogenicity Hepatotoxicity Estrogenicity Androgenicity Acute_Oral_Toxicity Bornyl acetate C12H20O2 6.27 27.96 40.48 14.21 2.09 25.42 (−)-Bornyl acetate C12H20O2 6.27 27.96 40.48 14.21 2.09 25.42 Isobornyl acetate C12H20O2 6.27 27.96 40.48 14.21 2.09 25.42 (+)-Bornyl acetate C12H20O2 6.27 27.96 40.48 14.21 2.09 25.42 [(1R,4S,7R,8R)-3,3,7,9-tetramethyl-11-oxo-4-tricyclo[5.4.0.02,8]undec-9-enyl]acetate C17H24O3 12.59 39.92 46.69 5.66 18.22 35.56 3-Methylbutylcyclohexane C11H22 8.15 39.09 48.02 7.67 0.46 14.99 Dihydroactinidiolide C11H16O2 12.87 40.79 48.40 3.28 6.36 28.68 γ-Heptalactone C7H12O2 10.43 36.91 48.84 9.88 0.79 5.54 (8αS)-5,8α-dimethyl-3-methylidene-3α,4,9,9α-tetrahydrobenzo[f][1]benzofuran-2,6-dione C15H16O3 14.29 40.72 48.88 6.49 16.75 49.69 (−)-Borneol C10H18O 4.35 28.78 49.21 44.99 13.12 33.69 Isoborneol C10H18O 4.35 28.78 49.21 44.99 13.12 33.69 Borneol C10H18O 4.35 28.78 49.21 44.99 13.12 33.69 Ledol C15H26O 15.05 33.30 49.36 35.04 6.53 28.30 1,4-Cineole C10H18O 11.42 36.57 49.53 20.06 1.16 8.76 Table 2 SAscores for 14 ingredients in Artemisia argyi
Compd. SAscore Compd. SAscore γ-Heptalactone 0.79 (−)-Borneol 0.64 1,4-Cineole 0.63 Isoborneol 0.64 Ledol 0.64 Dihydroactinidiolide 0.69 (−)-Bornyl acetate 0.66 Isobornyl acetate 0.66 3-Methylbutylcyclohexane 0.91 (+)-Bornyl acetate 0.66 Bornyl acetate 0.66 Borneol 0.64 [(1R,4S,7R,8R)-3,3,7,9-tetramethyl-11-oxo-4-tricyclo[5.4.0.02,8]undec-9-enyl]acetate 0.52 (8αS)-5,8α-dimethyl-3-methylidene-3α,4,9,9α-tetrahydrobenzo[f][1]benzofuran-2,6-dione 0.60 Table 3 Molecular docking results of core ingredients and target sites
Component Binding energy/(kJ/mol) EGFR HRAS PDGFRB PTPN11 SRC γ-Heptalactone −25.53 −28.46 −21.77 −22.19 −20.09 1,4-Cineole −26.79 −29.30 −24.70 −24.70 −24.70 Ledol −30.98 −28.88 −26.79 −30.98 −31.81 (−)-Bornyl acetate −23.86 −20.93 −21.77 −22.60 −25.95 3-Methylbutylcyclohexane −23.44 −27.63 −22.60 −24.70 −23.02 (−)-Borneol −21.35 −19.25 −26.79 −21.35 −23.86 Isoborneol −22.60 −25.95 −23.02 −20.93 −23.44 Dihydroactinidiolide −25.53 −24.70 −21.35 −25.53 −28.88 Isobornyl acetate −23.86 −25.12 −21.35 −24.28 −26.79 (+)-Bornyl acetate −23.44 −21.35 −22.19 −24.28 −25.12 (8αS)-5,8α-Dimethyl-3-methylidene-3α,4,9,9α-tetrahydrobenzo[f][1]benzofuran-2,6-dione −32.65 −32.23 −24.7 −27.63 −32.23 [(1R,4S,7R,8R)-3,3,7,9-tetramethyl-11-oxo-4-tricyclo[5.4.0.02,8]undec-9-enyl] acetate −29.30 −25.53 −24.28 −26.37 −29.72 Borneol −23.02 −20.09 −20.93 −21.77 −22.60 Bornyl acetate −23.44 −25.95 −23.86 −23.86 −23.86 -
[1] Wang JL, Guo QT, Zhang J, et al. Artemisia argyi and its application in traditional Chinese medicine agriculture[J]. Mod Horticult(现代园艺), 2024, 47(17): 80-82. [2] Erdenebileg S, Kim M, Nam Y, et al. Artemisia argyi ethanol extract ameliorates nonalcoholic steatohepatitis-induced liver fibrosis by modulating gut microbiota and hepatic signaling[J]. J Ethnopharmacol, 2024, 333: 118415.
[3] Zhang XJ. Artificial intelligence-driven pharmaceutical industry upgrade, capital rushes to AI-powered drug discovery and development[N]. Econic Inform Daily(经济参考报), 2024-06-05(003). [4] Di Stefano M, Galati S, Piazza L, et al. VenomPred 2.0: a novel in silico platform for an extended and human interpretable toxicological profiling of small molecules[J]. J Chem Inf Model, 2024, 64(7): 2275-2289. doi: 10.1021/acs.jcim.3c00692
[5] Sharma S, Pandey KM. Computational bioprospecting of phytoconstituents as potential inhibitors for peptide deformylase from Streptococcus oralis: an opportunistic pathogen[J]. Arch Biochem Biophys, 2024, 758: 110079. doi: 10.1016/j.abb.2024.110079
[6] Kerstjens A, De Winter H. Molecule auto-correction to facilitate molecular design[J]. J Comput Aided Mol Des, 2024, 38(1): 10. doi: 10.1007/s10822-024-00549-1
[7] Calvi A, Gaudin T, Miketa D, et al. Leap: molecular synthesisability scoring with intermediates[EB/OL]. 2024: 2403.13005. https://arxiv.org/abs/2403.13005v2.
[8] Skoraczyński G, Kitlas M, Miasojedow B, et al. Critical assessment of synthetic accessibility scores in computer-assisted synthesis planning[J]. J Cheminform, 2023, 15(1): 6. doi: 10.1186/s13321-023-00678-z
[9] Li JY, Liu T, Sun F, et al. Therapeutic effect and mechanism of hordenine on ovalbumin-induced allergic rhinitis in rats[J]. J China Pharm Univ(中国药科大学学报), 2025, 56(1): 80-90. [10] Xu MM, Ren L, Fan JH, et al. Berberine inhibits gastric cancer development and progression by regulating the JAK2/STAT3 pathway and downregulating IL-6[J]. Life Sci, 2022, 290: 120266. doi: 10.1016/j.lfs.2021.120266
[11] Jiang ZY, You QD. Assistance of artificial intelligence in new drug development[J]. J China Pharm Univ(中国药科大学学报), 2024, 55(3): 281-283. [12] Zhu JQ, Zhou X, Hou F, et al. DLGCN: Prediction of drug-lncRNA associations based on graph convolution network[J]. Chin J Bioinform(生物信息学), 2024, 22(2): 93-100. [13] Jiang XJ, Chen YH, Song Z, et al. Research progress in the pharmacology and mechanism of Artemisia argyi[J]. Acta Medicinae Sinica(华夏医学), 2023, 36(6): 182-188. [14] Zhang XF, Zeng YX, Zhou LP, et al. Preliminary study on optimization of extraction process and antioxidant activity of total flavonoids from Artemisia argyi leaves[J]. Ind Microbiol(工业微生物), 2024, 54(1): 134-136. [15] Zhang XY, Chen H, Wang Z, et al. Status of application of machine learning algorithm in drug toxicity prediction[J]. Chin J New Drug(中国新药杂志), 2024, 33(2): 110-117. [16] Haritha M, Sreerag M, Suresh CH. Quantifying the hydrogen-bond propensity of drugs and its relationship with Lipinski’s rule of five[J]. New J Chem, 2024, 48(11): 4896-4908. doi: 10.1039/D3NJ05476D
[17] Zeng H, Wu GZ, Zou WX, et al. Optimization of Menin inhibitors based on artificial intelligence-driven molecular factory technology[J]. J China Pharm Univ(中国药科大学学报), 2024, 55(3): 326-334. [18] Wen HQ, Sun QH, Shen WF. Targeted intelligent molecular generation framework based on fragments chemical space[J]. Ciesc J(化工学报), 2024, 75(4): 1655-1667. [19] Wang JJ, Cui WF, Dou XW, et al. Euonymus alatus delays progression of diabetic kidney disease in mice by regulating EGFR tyrosine kinase inhibitor resistance signaling pathway[J]. J S Med Univ(南方医科大学学报), 2024, 44(7): 1243-1255. [20] Cheng XD, Zhou H, Zhou Y, et al. M2 macrophage-derived exosomes inhibit apoptosis of HUVEC cell through regulating miR-221-3p expression[J]. Biomed Res Int, 2022, 2022: 1609244. doi: 10.1155/2022/1609244
[21] Xu WM, Qian JJ, Xu Q, et al. Proteins and signaling pathways response to Wenjingtongluo drug-contained serum in IHUVECs: an explorative proteomic study[J]. Cell Mol Biol (Noisy-le-grand), 2023, 69(6): 116-124. doi: 10.14715/cmb/2023.69.6.18
[22] Xue H, Dong L, Yang PS, et al. Effects of Tiaojing Cuyun prescription on endometrial receptivity and AREG/EGFR/HIF-1α signaling pathway in embryo implantation dysfunction infertility mice[J]. Chin J Inform TCM(中国中医药信息杂志), 2024, 31(10): 74-80.