【作者】 肖遂；

【导师】 薛飞群；

【作者基本信息】 中国农业科学院 ， 兽药学， 2013， 博士

【摘要】 中药有效物质的辨识由于受到方法和技术限制一直是中药现代化研究中的难点问题。为了寻求一种既符合中药特点，又能用现代科学技术诠释的中药有效物质辨识的模式方法，本研究对基于谱效关系的中药有效物质辨识新模式进行探索：选择不同来源的铁苋菜作为研究对象，分别采用UPLC/UV技术和UPLC/MS技术建立化学指纹图谱，以抑菌效果作为药效评价指标，将多个样品的整体化学信息和药效数据用化学计量学和数理统计等方法建立铁苋菜化学指纹图谱与抑菌效果之间的谱效关系，找出化学指纹图谱中与药效相关的特征峰，从而推测其有效成分；之后通过半制备色谱技术分离纯化铁苋菜各组分，将其与化学指纹图谱中的峰一一对应，并分别进行药效检测，从而验证该方法辨识中药有效物质的可靠性，最后采用UPLC-Q-TOF技术对抑菌成分进行结构鉴定。主要结果如下：针对铁苋菜对细菌性腹泻的显著疗效，选择5株引发腹泻的致病菌：嗜水气单胞菌、温和气单胞菌、副溶血弧菌、鳗弧菌和金黄色葡萄球菌，以对它们的抑菌效果作为铁苋菜药效评价指标；以药效最佳为准则优化铁苋菜提取工艺；用最佳提取工艺分别获得14个不同来源铁苋菜样品粗提物，并对它们进行抑菌药效评价，结果显示药效强弱次序为S8> S3> S4> S7> S12> S11> S1> S13>S5> S2> S10> S9> S6> S14。分别获取14个不同来源铁苋菜样品的UPLC/UV指纹图谱和UPLC/MS指纹图谱，对不同来源样品的各个特征峰进行描述统计分析，结果显示铁苋菜各成分在不同来源样品中均存在差异，且显著程度不一；对不同来源样品的指纹图谱进行相似性评价和聚类分析，样品指纹图谱与相应标准指纹图谱的相似度体现在聚类分析结果上，相似度高的样品和相似度低的样品被分成不同类，且相似度高的样品抑菌效果普遍比相似度低的样品好。铁苋菜由于品种、产地、采收季节、提取方法等因素造成的样品间差异，一方面体现在化学指纹图谱各个特征峰的大小上，另一方面体现在样品药效的强弱上。从以上14批铁苋菜样品分析的指纹图谱和药效结果中分别提取代表化学信息的谱和代表药理数据的效，采用直观比较法、相关分析法和多元回归法（逐步回归、偏最小二乘法）建立谱效关系，不同方法推测出的有效成分具有较好的一致性，UPLC/UV指纹图谱中4、5、6、8、9号特征峰和UPLC/MS指纹图谱中1’、2’、6’、7’、8’、9’、10’、11’号特征峰为谱效关系推测的铁苋菜有效成分。采用半制备色谱技术纯化分离得到19个制备组分，对其进行指纹图谱归属分析，确定谱效关系推测的有效成分主要包含于组分B（4、1’）、C（5、2’）、D（6、6’）、E（7’）、F（8’）、H（8、9’）、K（10’）、L（9、11’）中；检验以上组分和其它组分的抑菌效果，证实组分C和L的抑菌效果最强，组分D、H、K的抑菌效果次之，组分B和E的抑菌效果最弱，除以上谱效关系推测的有效成分之外，组分M和Q也具有较弱抑菌效果，为铁苋菜的次要抑菌成分。通过UPLC-Q-TOF技术对铁苋菜9个抑菌成分进行结构鉴定，其中经标准品对照证实的化合物包括尼古丁酸、没食子酸、原儿茶酸、对羟基苯甲酸、槲皮素，未经标准品对照证实，仅通过Q-TOF数据推测的化合物为铁苋菜素；经Q-TOF数据分析，另外3个为首次从铁苋菜中发现的成分，推测它们的分子式分别为C13H8O8、C22H12O13、C27H22O18。综上所述，本研究通过谱效关系共推测出8个与抑菌相关的成分，其中7个被证实具有抑菌效果；在通过半制备色谱技术得到的9个抑菌成分中仅有2个次要抑菌成分未通过谱效关系辨识；经结构鉴定表明其中6个抑菌成分为已知化合物，另3个抑菌成分为铁苋菜首次报道的化合物。本研究以铁苋菜作为模式中药，充分证明了谱效关系辨识中药有效物质的可行性，为探索一种新的适合现代中药物质基础研究的模式方法提供了实验基础。更多还原

【Abstract】 Identification of efficient constituents, a core content of the modernization of traditional Chinesemedicine (TCM), is always considered as a difficult problem because of the limitation of techniques andmethods. To establish a new pattern that not only covered the theory of TCM but also interpreted theactive constituents with modern science languages, spectrum-effect relationships were applied as a newtool to investigate the identification of TMC in this study. Acalypha australis Linn. and its antibacterialeffect were chosen as the model of TCM and the pharmacodynamic evaluation criterion respectively,UPLC/UV technology and UPLC/MS technology were also applied to establish chemical fingerprints.The overall chemical information and pharmacodynamic data were acquired from the chemicalfingerprints and analysis of the inhibitory effects of various samples originated from different sources.Coupled with chemometric and mathematical statistical methods, the spectrum-effect relationships ofAcalypha australis Linn between chemical fingerprints and antibacterial effects were developed, andthen the characteristics peaks associated with pharmacodynamic effects in the fingerprints were located,and thus the active ingredients were speculated.To further certify its reliability, ingredients isolated and purified from Acalypha australis Linnusing semi-preparative chromatography were located the corresponding characteristic peaks in thechemical fingerprints and thus confirmed the putative active ingredient. Prior to the structures of theantibacterial ingredients were identified by UPLC-Q-TOF, the antibacterial experiments were performed.The main results are as follows:Since the pronounced effects of Acalypha australis Linn. on bacterial diarrhea,Aeromonashydrophila, Aeromonassobria, Vibrio parahaemolyticus, Vibrioanguillarum andStaphylococcusaureus were chosen to evaluate the antibacterial activities of Acalypha australis Linn..Crude extracts of14samples were produced using the optimum extraction process judged by theirantibacterial activity. The results from the antibacterial experiments showed that the magnitude ofantibacterial activity of the Acalypha australis Linn. samples was S8> S3> S4> S7> S12> S11> S1>S13> S5> S2> S10> S9> S6> S14.UPLC/UV fingerprint endowed with fourteen characteristic peaks and UPLC/MS fingerprintprovided with twenty ones were established respectively. UPLC/UV fingerprints and UPLC/MSfingerprints of Acalypha australis Linn. from fourteen samples were acquired to analyze similarityvalues between the reference fingerprint and the chromatograms of fourteen samples and to performhierarchical clustering analysis of samples. The results indicated that the samples could be divided intodifferent clusters according to similarity values and sample exhibiting greater similarity with referencefingerprint might produce stronger antibacterial activity.The differences, which mainly existed in the area of characteristic peaks in the fingerprints and thedegree of antibacterial effects, among the Acalypha australis Linn. samples, depended on species, origin,collection time and extraction methods. The crucial information of―spectrum‖and―effect‖weregathered from the chemical fingerprints and antibacterial effect of fourteen samples. The spectrum-effect relationships were established respectively with direct-viewing comparisons,correlational analysis method and multi-element linear regression (stepwise regression method andpartial least square method). Interestingly, the good consistency in predicting efficient constituentsanalyzed with different methods appeared.5,6,8and9on UPLC/UV fingerprint and Characteristicpeaks1’,2’,6’,7’,8’,9’,10’ and11’on UPLC/MS fingerprint were the putative efficient constituentsof Acalypha australis Linn..Nineteen fractions of Acalypha australis Linn.were separated and purified by semi-preparativeHPLC and then analyzed carefully to locate the corresponding position in the fingerprint. Eight ofthem were the efficient constituents in agreement with the results predicted by the spectrum-effectrelationship, which were belonged to B（4、1’）、C（5、2’）、D（6、6’）、E（7’）、F（8’）、H（8、9’）、K（10’）、L（9、11’）.After the antibacterial effects of these fractions prepared were performed,Fraction C, L were confirmed as the main antibacterial constituents of Acalypha australis Linn. whilefraction D, H and K were acted as the important auxiliary antibacterial constituents of Acalyphaaustralis Linn., but fraction B, E were low antibacterial effects. Either, besides the putative efficientconstituents, the role of fraction M and fraction Q in antibacterial effects played was only minorantibacterial constituents.The structures of nine antibacterial constituents of Acalypha australis Linn. were analyzed byUPLC-Q-TOF. Nicotinic acid, gallicacid, protocatechuicacid, p-hydroxybenzonic acid and quercetinwere confirmed by comparison with their standard substances. However, with the aid of data ofQ-TOF and without standard compound, one compound was inferred as copperleafelement, but themolecular formulas of the remaining three unknown constituents of Acalypha australis Linn. wediscovered for the first time were inferred as C13H8O8, C22H12O13and C27H22O18,respectively.In conclusion, eight antibacterial constituents of Acalypha australis Linn. were predicted byspectrum-effect relationship and their antibacterial effects of seven constituents have been testified. Itwas worth to notice that seven out of nine inhibitory bacterial ingredients obtained by semi-preparativechromatography were in accordance with the corresponding constituents predicted by spectrum-effectrelationship. The results from the analysis of structures demonstrated that6antibacterial constituentswere known compounds of Acalypha australis Linn. and three antibacterial constituents were unknowncompounds we discover first. Taken together, Acalypha australis Linn. was a good example as therepresentative TCM to evidence the feasibility of spectrum-effect relationship in identifying effectiveingredients from the complicated TCM. Our study would also provide the experimental basis for a newpattern that was beneficial to the recognition of efficient constituents of TCM.更多还原