基于灰色关联度分析法的滇龙胆质量评价

Quality Evaluation of Gentiana rigescens by Grey Relational Analysis Method

摘要:

目的：测定不同来源滇龙胆样品中4种指标成分，结合灰色关联度分析（GRA）法建立其质量评价方法。方法：采用高效液相色谱法测定云南10个不同地区的50株滇龙胆根及根茎中主要药效成分马钱苷酸、獐牙菜苦苷、龙胆苦苷和当药苷的含量。基于GRA法处理样品各指标成分数据，计算各评价单元序列相对于参考序列的相对关联度作为测度，对样品进行综合评价。结果：建立滇龙胆样品的指纹图谱，并测定出4种主要成分的含量。滇龙胆样品的相对关联度r_i在0.306~0.600，其中14个样品的r_i>0.500，23个样品的r_i<0.400，表明样品质量之间存在较大差异。大理鹤庆的DH3和DH5的r_i分别为0.600和0.576，质量最优。不同地区样品r_i的平均值在0.325~0.541，鹤庆样品r_i的平均值最大，质量总体水平最高，可作为优质种源进一步研究和推广。各地区样品r_i的RSD在4.4%~11.9%，昆明寻甸样品r_i的RSD最小，样品最为稳定。结论：GRA法结合多指标成分定量的评价方法简单、全面，可用于滇龙胆的质量评价，为资源开发利用及优质种源的筛选提供了参考依据。

Abstract:

Objective: To determine the contents of four index components in Gentiana rigescens from different sources, and establish a comprehensive method to evaluate the quality of G. rigescens based on grey relational analysis (GRA). Method: The contents of loganic acid, swertiamarin, gentiopicroside and sweroside in roots and rhizomes of fifty G. rigescens samples from ten different regions of Yunnan Province were detected by high performance liquid chromatography. The relative correlation degree between reference sequence and each evaluation unit sequence was calculated to evaluate the samples synthetically based on the data of index components. Result: The fingerprints of G. rigescens samples were established, and the contents of the four main components were determined. The relative correlation degree (r_i) of all G. rigescens samples was between 0.306 and 0.600, and the r_i was above 0.500 for 14 samples, and below 0.400 for 23 samples, indicating that there were great differences in quality of samples. The r_i was 0.600 and 0.576 respectively for DH3 and DH5 from Dali Heqing, which showed optimal quality. The average values of r_i for the samples from different regions were between 0.325 and 0.541, and the average value was maximum for Heqing samples, indicating that the overall quality of Heqing samples was highest, which could be used as excellent resources for further research and extension. The RSD values of r_i for samples from different regions were between 4.4% And 11.9%. The RSD value of r_i for Kunming Xundian samples was minimum, indicating that Xundian samples had the most stable quality. Conclusion: The evaluation method with combination of GRA method and multi-ingredient quantification was simple and comprehensive to evaluate the quality of G. rigescens, and could provide references for the development and utilization of resources as well as optimal resources screening.