ABSTRACT
OBJECTIVE To evaluate the quality of different germplasms of Rehmannia glutinosa based on iridoid glycosides. METHODS The contents of total iridoid glycosides ，catalpol，rehmaionoside D ，rehmaionoside A ，and leonuride in 18 batches of R. glutinosa from 6 germplasms（85-5，JinJiu，BX，BJ-1，Shandong，QH-1）were determined by ultraviolet spectrophotometry and high performance liquid chromatography. After normalization of the above content determination results ，the quality of different germplasm of R. glutinosa were evaluated by multiple statistical methods such as cluster analysis ，factor comprehensive analysis and partial least squares discriminant analysis （PLS-DA）. RESULTS Among 6 germplasms of R. glutinosa ，the content of total iridoid glycosides in R. glutinosa 85-5 was the highest ，and the content of catalpol in R. glutinosa BX was the highest ；the contents of rehmannioside D and rehmannioside A in R. glutinosa JinJiu were the highest ，and the content of leonuride in R. glutinosa BX was the highest. Cluster analysis showed that R. glutinosa JinJiu were clustered into one category ，R. glutinosa BX clustered into one category ，R. glutinosa Shandong and R. glutinosa BJ-1 were clustered into one category ，and R. glutinosa QH-1 and 85-5 were clustered into one category. Through factor comprehensive analysis ，there were differences in the quality of different germplasms of R. glutinosa . The comprehensive score of R. glutinosa BX，Shandong，85-5，BJ-1，QH-1，JinJiu were 2.283 9，1.689 1，1.664 8， 1.503 3，1.469 0，1.214 6，respectively. PLS-DA showed that variable importance projection value of total iridoid glycosides ， catalpol and leonuride were all higher than 1. CONCLUSIONS The quality difference of R. glutinosa from different germplasms may be caused by total iridoid glycosides ，catalpol and leonuride.
ABSTRACT
OBJECTIVE： To establish a method for rapid determination of total phenylethanoid glycosides and total iridoid glycosides in the root of Rehmannia glutinosa. METHODS： The contents of total phenylethanoid glycosides and total iridoid glycosides in medicinal material samples were determined by UV spectrophotometry. Quantitative model of total phenylethanoid glycosides and total iridoid glycosides in medicinal samples was established by NIRS-PLS method. The optimal pretreatment spectra were multivariate scattering correction combined with first derivative method， standard normalization combined with first derivative method. The optimum spectral ranged from 6 703.35-11 065.54 cm－1 and 3 999.63-9 102.36 cm－1. The optimum principal factor number were 10 and 7. RESULTS： The content determination of total phenylethanoid glycosides and total iridoid glycosides in medicinal material samples was proved to meet the requirements by methodological experience. The internal cross validation determination coefficients of total phenylethanoid glycosides and total iridoid glycosides were 0.998 2 and 0.980 9. The correction of root mean square error was 0.032 7 and 0.186 0. The root mean square error of prediction were 0.035 5 and 0.035 1. The root mean square error of cross validation were 0.256 9 and 0.574 3. The predicted values of total phenylethanol glycosides and total iridoid glycosides were 0.268％-1.636％ and 3.424％-6.978％， respectively； the determination value of them were 0.299％-1.629％ and 3.431％-6.952％， respectively； the absolute deviations were －0.042％-0.067％ and －0.111％-0.088％， respectively；the relative deviations were －0.819％-0.076％、－2.257％-1.672％， respectively；There was no statistical significance between predicted values and measured values （P＞0.05）. CONCLUSIONS： The method is accurate and simple. The method can be used for the rapid determination of total phenylethanoid glycosides and total iridoid glycosides in different germplasms of R. glutinosa.