干燥、粉碎10批不同产地、不同采摘期的昆仑雪菊样品,分取0.2 g,用60%(体积分数)甲醇溶液50 mL超声提取30 min,过0.45 μm滤膜,所得溶液(质量浓度为4 000 mg·L^-1)进入高效液相色谱仪,其中各组分在Agilent HC-C₁₈色谱柱上用不同体积比的0.4%(体积分数)乙酸溶液和乙腈的混合溶液进行梯度洗脱分离,在285 nm波长下检测。以中药色谱指纹图谱相似度评价系统建立指纹图谱,筛选共有组分并进行相似度评价。对绿原酸进行方法学考察,并以绿原酸为参照计算各共有组分的含量。以聚类分析和主成分分析等化学模式识别法对不同产地、不同采摘期的样品进行分类。分别采用1,1-二苯基-2-三硝基苯肼(DPPH)、2,2-联氮-二(3-乙基-苯并噻唑-6-磺酸)二铵盐(ABTS)自由基清除法测定10批样品的抗氧化活性,以灰色关联分析法建立共有组分峰面积数据和抗氧化活性的谱效关系。结果显示:10批样品指纹图谱以及它们与对照指纹图谱之间的相似度均大于0.900,共有组分有24个;两种化学模式识别法均将样品分为4类,推测分类结果和产地、采摘期有关。2种自由基清除法所得的关联度大于0.800的组分各有5个,可选作昆仑雪菊的药效控制点。

Ten batches of samples of Coreopsis tinctoria from different origins and harvesting periods were dried and crushed, and an aliquot (0.2 g) was taken, extracted with 50 mL of 60% (volume fraction) methanol solution for 30 min by ultrasound, and passed through a 0.45 μm filter membrane. The solution (mass concentration of 4 000 mg·L^-1) obtained was introduced into high performance liquid chromatograph. Separation was performed on an Agilent HC-C₁₈ column by gradient elution with mixed solutions composed of 0.4% (volume fraction) acetic acid solution and acetonitrile at different volume ratios, and targets were detected at 285 nm. The fingerprints were established by the chromatographic fingerprint similarity evaluation system of traditional Chinese medicine, and the common components were screened and used for similarity evaluation. Chlorogenic acid was investigated methodologically, and the content of each common component was calculated using chlorogenic acid as reference. The samples from different origins and harvesting periods were classified by chemical pattern recognition methods such as cluster analysis and principal component analysis. The antioxidant activities of the 10 batches of samples were determined by 1,1-diphenyl-2-trinitrophenylhydrazine (DPPH) and 2,2-benzo-bis(3-ethyl-benzothiazole-6-sulfonic acid) diammonium salt (ABTS) free radical scavenging methods, and the spectrum-effect relationships between the peak area data of common components and the antioxidant activity were established by gray correlation analysis. It was shown that the similarity between fingerprints of the 10 batches of samples and them together with control fingerprint was greater than 0.900, and 24 common components were found. Two chemical pattern recognition methods classified the samples into 4 categories, and it was assumed that the classification results were related to the origin and harvesting period. The 5 components with the correlation degree greater than 0.800 were obtained from each of the 2 free radical scavenging methods, which could be used as the pharmacodynamic control points of Coreopsis tinctoria.