目的 基于HPLC指纹图谱与多成分含量测定，并结合化学模式识别法，评价不同产地栀子的质量，优选最佳产地。方法 采用RP-HPLC切换波长法，流动相为乙腈-0.1%磷酸水溶液进行梯度洗脱，检测波长为234、440 nm，对10个不同产地的30批栀子药材进行指纹图谱及5种有效成分含量测定研究，应用聚类分析与主成分分析等化学模式识别方法对栀子质量进行评价。结果 10个不同产地30批栀子的指纹图谱相似度均在0.95以上，去乙酰车叶草酸甲酯、京尼平龙胆双糖苷、栀子苷、西红花苷I和西红花苷Ⅱ质量分数分别为0.61~4.26、1.73~12.92、51.79~82.76、5.03~12.80、0.71~2.28 mg/g；聚类分析可分为3类，第1类为产地湖南永州、湖南永安、江西丰城、江西湖口、安徽广德和福建福鼎；第2类为产地江西黎川、江苏句容和河南唐河；第3类为产地安徽六安。通过主成分分析，进一步评价不同产地间质量的差异，从而发现安徽六安、江西湖口和湖南永安等地质量较优。结论 不同产地之间栀子存在一定质量差异，同一产地不同批次间栀子质量较稳定。通过指纹图谱与含量测定相结合，应用化学模式识别方法可全面综合评价栀子质量。此方法的建立为栀子药材的质量控制及评价提供了有效的参考依据。

Objective To evaluate the quality of Gardeniae Fructus from different producing areas and optimize the best place of production by using chemical pattern recognition method based on HPLC fingerprint and multi-component content determination. Methods The RP-HPLC using acetonitrile-0.1% phosphoric acid aqueous solution as mobile phase based on the wavelength switching (234 nm and 440 nm) and gradient elution method was developed to evaluate the quality of 30 batches of Gardeniae Fructus from 10 different producing areas. The combination of chromatographic fingerprints and quality determination of five active ingredients, as well as chemometrics including hierarchical cluster analysis (HCA) and principal component analysis (PCA), were further employed for the quality assessment. Results The fingerprint similarity of 30 batches of Gardeniae Fructus from 10 different producing areas were all above 0.95. The content ranges of deacetylated methyl oxalate, genipin gentiobioside, geniposide, crocin I, and crocin Ⅱ were 0.61-4.26 mg/g, 1.73-12.92 mg/g, 51.79-82.76 mg/g, 5.03-12.80 mg/g, and 0.71-2.28 mg/g. Samples can be divided into three categories by HCA:The first category is Yongzhou in Hunan, Yong' an in Hunan, Fengcheng in Jiangxi, Hukou in Jiangxi, Guangde in Anhui, and Fuding in Fujian; The second category is Lichuan in Jiangxi, Jurong in Jiangsu and Tanghe in Henan; And the third category is Lu' an in Anhui. PCA was used to further evaluate the quality differences of the samples from different producing areas, and it was found that the medicinal materials from Lu' an in Anhui, Hukou in Jiangxi, and Yong' an in Hunan were of superior quality. Conclusion The quality of Gardeniae Fructus from different producing areas was different to a certain extent, while the quality of same batches was stable. By combination of fingerprint and content determination, the chemical pattern recognition method can be used to evaluate the quality of the Gardeniae Fructus comprehensively. The establishment of this method provided an effective reference for the quality control and evaluation of Gardeniae Fructus.

国家中药标准化项目（ZYBZH-C-JS-33）