RESUMEN
OBJECTIVE To establish the fingerprint and multi-component content determination method of Crataegus pinnatifida leaves from different producing areas， and to evaluate the quality of C. pinnatifida leaves and screen the differential markers. METHODS Seventy-eight batches of C. pinnatifida leaves were collected from Chengde of Hebei Province， Huludao of Liaoning Province， Yuncheng of Shanxi Province and Linyi of Shandong Province. High-performance liquid chromatography （HPLC） and Similarity Evaluation System for Traditional Chinese Medicine Chromatographic Fingerprints （2012 edition） were used to draw the fingerprints and conduct similarity evaluation. Grey correlation analysis， cluster analysis （CA）， principal component analysis （PCA） and orthogonal partial least squares-discriminant analysis （OPLS-DA） were performed by using SPSS 19.0， MetaboAnalyst 5.0 and SIMCA 14.1 software. The differential markers affecting the quality of C. pinnatifida leaves were screened with variable importance in the projection （VIP） value greater than 1 and the error line not exceeding the origin as the criterion. Using vitexin rhamnoside as an internal reference， the contents of chlorogenic acid， glucosylvitexin， hypericin and isoquercetin in 78 batches of C. pinnatifida leaves were determined by the same HPLC combined with quantitative analysis of multi- components by single-marker （QAMS）， and the results were compared with external standard method. RESULTS Eight common peaks were calibrated in the fingerprints for 78 batches of C. pinnatifida leaves from 4 producing areas. Five known components were identified， including chlorogenic acid （peak 1）， glucosylvitexin （peak 3）， vitexin rhamnoside （peak 4）， hypericin （peak 7） and isoquercetin （peak 8）； their similarities ranged from 0.871 to 0.998. Average relative correlations of samples from Chengde of Hebei Province， Huludao of Liaoning Province， Yuncheng of Shanxi Province and Linyi of Shandong Province were 0.538， 0.528， 0.462 and 0.435， respectively. CA and PCA showed that the samples from Chengde of Hebei Province and Huludao of Liaoning Province were roughly classified into one category， while the samples from Linyi of Shandong Province and Yuncheng of Shanxi Province were roughly classified into one category； VIP values of peak 1， 2， 3 and 5 were all greater than 1. By QAMS， the relative correction factors of chlorogenic acid， glucosylvitexin， hypericin and isoquercetin were 0.401， 0.993， 1.670 and 1.615 （RSD＜2%）. Compared with external standard method， except for isoquercetin in the two batches of samples （S39 and S41）， there was no significant difference in the content of each component in other batches of samples （the relative deviations≤ 5%）. CONCLUSIONS The established fingerprint and QAMS method are simple to operate and can be used to evaluate the quality of C. pinnatifida leaves. The sample from Chengde of Hebei Province is relatively good in quality. Chlorogenic acid （peak 1）， glucosylvitexin （peak 3）， and the corresponding components of peaks 2 and 5 may be differential markers affecting the quality of C. pinnatifida leaves.