天然产物研究与开发 ›› 2020, Vol. 32 ›› Issue (2): 305-316.doi: 10.16333/j.1001-6880.2019.2.015

• 开发研究 • 上一篇    下一篇

基于HPLC和NIRS建立快速检测盾叶薯蓣中3种皂苷含量的方法

陶晓赛1,陈志红1,2,谢彩侠1,2*,张娟1,2*,龚海燕1,2,刘庆普1,2,雷敬卫1,2   

  1. 1河南中医药大学药学院;2河南省中药质量控制与评价工程技术研究中心,郑州 450046
  • 出版日期:2020-02-29 发布日期:2020-04-23
  • 基金资助:

    国家重点研发计划(2017YFC1700705);河南省高等学校重点科研项目(20A360016)

Establishing a method for rapid detection of three saponins in Dioscorea zingiberensis C.H.Wright based on HPLC and NIRS

TAO Xiao-sai1,CHEN Zhi-hong1,2,XIE Cai-xia1,2*,ZHANG Juan1,2*,GONG Hai-yan1,2,LIU Qing-pu1,2,LEI Jing-wei1,2   

  1. 1School of Pharmacy,Henan University of Chinese Medicine;2Henan Province Traditional Chinese Medicine Quality Control and Evaluation Engineering Technology Research Center,Zhengzhou 450046,China
  • Online:2020-02-29 Published:2020-04-23

摘要: 建立一种快速检测盾叶薯蓣中三角叶薯蓣皂苷、盾叶新苷和薯蓣皂苷含量的方法。本研究以全国8个产地的盾叶薯蓣药材为研究对象,首先,利用HPLC-ELSD建立同时测定盾叶薯蓣中三角叶薯蓣皂苷、盾叶新苷及薯蓣皂苷含量的方法,并对不同产地的盾叶薯蓣药材进行三种皂苷的含量测定;其次,扫描盾叶薯蓣药材样品的近红外光谱,分别将盾叶薯蓣药材校正集样品的三种皂苷含量作为参考值,结合其近红外光谱图,以内部交叉验证决定系数(R2)、校正均方根偏差(RMSEC) 、预测均方根偏差(RMSEP)及预测性能指数(PI)作为评价所建定量检测模型性能的指标,利用TQ8.0分析软件结合偏最小二乘法(PLS),通过光谱预处理方法筛选、建模波段及主成分数的确定分别建立盾叶薯蓣药材中三种皂苷含量的快速检测模型;最后,分别利用验证集样品对所建三种皂苷检测模型的预测准确性进行检验。盾叶薯蓣样品中三角叶薯蓣皂苷、盾叶新苷和薯蓣皂苷含量测定方法经考察符合定量分析的要求;盾叶薯蓣药材中三角叶薯蓣皂苷定量检测模型的R2为0.981 17、RMSEC为0.086 3、RMSEP为0.063 8、PI为90.5;盾叶新苷定量检测模型的R2为0.982 64、RMSEC为0.042 0、RMSEP为0.027 4、PI为91.1;薯蓣皂苷定量检测模型的R2为0.943 64、RMSEC为0.009 90、RMSEP为0.005 41、PI为85.8;经统计学检验,三个模型对三种皂苷的预测值与实测值之间无显著性差异。该方法可以相对快速、准确测定盾叶薯蓣中三角叶薯蓣皂苷、盾叶新苷及薯蓣皂苷的含量,为盾叶薯蓣药材质量的快速评价提供依据。

关键词: HPLC, NIRS, 盾叶薯蓣, 偏最小二乘法, 三角叶薯蓣皂苷, 盾叶新苷, 薯蓣皂苷

Abstract: To establish a rapid method for the determination of deltonin,zingiberensis newsaponin and dioscin in Dioscorea zingiberensis C.H.Wright.The research objects were the medicinal materials of D. zingiberensis from 8 producing areas in China.Firstly,using HPLC-ELSD established the method of simultaneously determining the contents of deltonin ,zingiberensis newsaponin and dioscin,which could be used to determine the contents of three saponins in D. zingiberensis from different habitats .Secondly,the samples were scanned for obtainning their near infrared spectrums,the contents of three saponins in the calibration set samples of D. zingiberensis were taken as reference values respectively,combined with their near-infrared spectrums,the internal cross-validation determination coefficient (R2),corrected root mean square deviation (RMSEC),predicted root mean square deviation (RMSEP) and predicted performance index (PI) were used as indicators to evaluate the performance of the quantitative test model.TQ 8.0 analysis software were combined with partial least squares (PLS) to establish a rapid detection model for the content of three saponins in D. zingiberensis through screening the spectrum pretreatment method,determining the modeling band and principal component number.Finally,the validation set samples was used to test the prediction accuracy of the three saponin detection models respectively,the content determination methods of deltonin ,zingiberensis newsaponin and dioscin in D. zingiberensis samples met the requirements of quantitative analysis after investigation.The R2, RMSEC,RMSEP and PI of the quantitative detection model for deltonin in D. zingiberensis were 0.981 17,0086 3,0.063 8 and 90.5 respectively.The R2,RMSEC,RMSEP and PI of the quantitative detection model of zingiberensis newsaponin were 0.982 64,0.042 0,0.027 4 and 91.1 respectively.The R2 of dioscin quantitative detection model was 0.943 64,RMSEC was 0.009 90,RMSEP was 0.005 41,PI was 85.8;Statistical tests showed that there was no significant difference between the predicted and measured values of the three saponins in the three models.The method can be used to determine the contents of deltonin ,zingiberensis newsaponin and dioscin in Dioscorea zingiberensisC.H.Wright relatively quickly and accurately,which provide the basis for rapid quality evaluation of D. zingiberensis.

Key words: HPLC, NIRS, Dioscorea zingiberensis C.H.Wright, PLS, deltonin, zingiberensis newsaponin, dioscin

中图分类号: 

R917