基于谱效关系、网络药理学和分子对接技术研究藏药湿生扁蕾改善小鼠肝纤维化的有效成分及靶点预测

Active Components and Target Prediction of Tibetan Medicine Gentianopsis paludosa in Ameliorating Liver Fibrosis in Mice Based on Spectrum-Effect Relationship, Network Pharmacology and Molecular Docking

关键词:  湿生扁蕾  肝纤维化  谱效关系  有效成分  网络药理学  分子对接  

Key words:  Gentianopsis paludosa.  liver fibrosis  spectrum-effect relationship  effective components  network pharmacology  molecular docking  

摘要:

目的：基于谱效关系、网络药理学与分子对接探讨藏药湿生扁蕾抗肝纤维化的有效成分和潜在靶点预测。方法：采用高效液相色谱法（HPLC）技术构建10批湿生扁蕾药材的指纹图谱，通过聚类分析及主成分分析对不同批次样品进行质量评价。选取130只8周龄SPF级C57BL/6J雄性小鼠，分为对照组、模型组、阳性对照组、不同批次湿生扁蕾给药组S1-S10组，除对照组外，其余各组均采用腹腔注射25%CCl?橄榄油溶液造模8周，3次/周，复制小鼠肝纤维化模型；自造模第3周起，给药组每日灌胃3.094 g?kg-1生药量的不同批次湿生扁蕾提取物，连续给药6周，模型组与对照组同期灌胃等体积溶剂，末次给药后检测血清肝功能及肝纤维化标志物指标；肝组织进行病理学分析，采用Metavir系统评分，运用免疫组化法检测α-平滑肌肌动蛋白(α-smooth muscle actin, α-SMA)、I型胶原蛋白(Collagen I, Col Ⅰ)表达水平，综合评价抗肝纤维化疗效。活性成分筛选以肝纤维化组织评分、α-SMA及Collagen I蛋白表达水平为药效指标，10个共有色谱峰的峰面积为自变量，分别采用灰色关联度分析（grey relational analysis, GRA）（关联度 ＞ 0.70）和偏最小二乘回归分析（partial least-squares regression, PLSR）（VIP ＞ 1且回归系数为负）筛选与各指标相关的色谱峰，取两种方法筛选结果的交集，再合并三个指标的交集结果，最终获得的色谱峰即为湿生扁蕾抗肝纤维化有效成分；以谱效关系筛选的湿生扁蕾抗肝纤维化潜在药效成分为研究对象，通过TCMSP、PubChem等数据库收集成分靶点，同时从GeneCards数据库获取抗肝纤维化作用的疾病靶点，然后利用Venny数据库获得交集靶点，采用String数据库及Cytoscape3.9.1软件构建蛋白-蛋白相互作用网络，筛选出核心靶点；对交集靶点进行GO和KEGG富集分析，利用AutoDock Vina软件对潜在药效成分与核心靶点进行分子对接，从而初步探讨湿生扁蕾抗肝纤维化的潜在作用靶点。结果：10批湿生扁蕾药材HPLC指纹图谱相似度在0.756 ? 0.997，共标定10个共有峰，其中4个经对照品指认为异荭草苷、当药黄素、木犀草素和芹菜素，10批湿生扁蕾聚为3类；谱效关联分析显示，共有峰F1（异荭草苷）、F3（当药黄素）、F4（未指认）、F8（芹菜素）与湿生扁蕾抗肝纤维化药效密切相关。网络药理学预测发现已指认成分异荭草苷、当药黄素、芹菜素抗肝纤维化的核心靶点为TP53、AKT1、IL6、STAT3、TNF、MAPK8，主要富集于AGE-RAGE、HIF-1、IL-17及TNF等多条信号通路；分子对接结果显示，3个成分与核心靶点TP53、AKT1之间的结合性能最优（结合能均小于-7.0 kcal·mol?1）。结论：本研究通过谱效关系分析得到异荭草苷、当药黄素、芹菜素及峰4（未指认）为湿生扁蕾抗肝纤维化的潜在活性成分；结合网络药理学与分子对接，初步揭示其可能通过作用于TP53、AKT1等核心靶点，调控AGE-RAGE、HIF-1、IL-17及TNF等多条信号通路发挥抗纤维化作用。

Abstract:

Objective: To explore the effective components and potential targets of the Tibetan medicine Gentianopsis paludosa against liver fibrosis based on spectrum-effect relationship, network pharmacology, and molecular docking. Methods: High-performance liquid chromatography (HPLC) was employed to construct fingerprints of ten batches of Gentianopsis paludosa Cluster analysis and principal component analysis were used to evaluate the quality of different batches. A total of 130 eight-week-old specific pathogen-free (SPF) male C57BL/6J mice were divided into a control group, a model group, a positive control group, and ten Gentianopsis paludosa treatment groups (S1-S10). Except for the control group, all other groups received intraperitoneal injection of 25% CCl? olive oil solution three times per week for eight weeks to induce liver fibrosis. From the third week of modeling, the treatment groups were intragastrically administered 3.094 g·kg?1 (crude drug) of different batches of Gentianopsis paludosa extract daily for six consecutive weeks, while the model and control groups received an equal volume of solvent. After the final administration, serum markers of liver function and liver fibrosis were measured. Liver tissues were subjected to histopathological analysis using the Metavir scoring system. The expression levels of α-smooth muscle actin (α-SMA) and Collagen I (Col Ⅰ) were detected by immunohistochemistry to comprehensively evaluate the anti?liver?fibrosis efficacy. For active component screening, liver fibrosis pathological scores and protein expression levels of α-SMA and Col Ⅰ were taken as pharmacodynamic indices. The peak areas of ten common chromatographic peaks were used as independent variables. Grey relational analysis (GRA, with relational degree >0.70) and partial least?squares regression (PLSR, VIP >1 and negative regression coefficient) were respectively applied to screen peaks associated with each index. The intersection of the two methods for each index was taken, and then the intersections across the three indices were merged. The finally obtained chromatographic peaks were regarded as the anti?liver?fibrosis active components of Gentianopsis paludosa. Using these spectrum?effect?based potential active components as subjects, component targets were collected from TCMSP, PubChem and other databases, while disease targets for anti?liver?fibrosis were obtained from the GeneCards database. Intersection targets were identified using Venny, and a protein–protein interaction (PPI) network was constructed using the String database and Cytoscape 3.9.1 to screen core targets. GO and KEGG enrichment analyses were performed on the intersection targets. Molecular docking between the potential active components and core targets was carried out using AutoDock Vina to preliminarily explore the potential anti?liver?fibrosis targets of Gentianopsis paludosa. Results: The HPLC fingerprint similarities of the ten batches ranged from 0.756 to 0.997. Ten common peaks were identified, among which four were assigned as isoorientin, swertisin, luteolin, and apigenin by reference standards. The ten batches were clustered into three groups. Spectrum?effect correlation analysis showed that peaks1, 3, 4, and 8 were closely related to the anti?liver?fibrosis effect of Gentianopsis paludosa. Network pharmacology predicted that the core targets of the identified components (isoorientin, swertisin, and apigenin) against liver fibrosis were TP53, AKT1, IL6, STAT3, TNF, and MAPK8, which were mainly enriched in multiple signaling pathways including AGE-RAGE, HIF-1, IL-17, and TNF. Molecular docking results showed that three components exhibited the best binding affinity to the core targets TP53 and AKT1, with binding energies all less than -7.0 kcal·mol?1. Conclusion: This study demonstrates through spectrum?effect relationship analysis that isoorientin, swertisin, apigenin, and peak 4 (unidentified) are potential active components of Gentianopsis paludosa against liver fibrosis. Combined with network pharmacology and molecular docking, it is preliminarily revealed that Gentianopsis paludosa may exert its anti?fibrotic effect by acting on core targets such as TP53 and AKT1 and regulating multiple signaling pathways including AGE-RAGE, HIF-1, IL-17, and TNF.