Quality control is a basic requirement in the manufacturing of herbal medicinal products. In the past, metabolomic fingerprinting with NMR and HPLC, respectively, in combination with PCA (Principal Component Analysis) proved to be an alternative method for control of quality [1] and origin [2] and for the discrimination between species [3]. However, most projects in this context are only investigating small groups of plants, mostly one species or one genus. The intention of this research project is to develop a method which is applicable to a broader range of medicinal plants, and to evaluate the possibilities of applications to analyse also finished herbal medicinal products. In order to obtain consistent data the experimental procedure was standardised after a series of preliminary experiments. The dried and ground starting material is extracted first with dichloromethane and subsequently with methanol. The 1H NMR spectra and the HPLC/ELSD chromatograms of the plant extracts are recorded and statistically analysed by PCA. Different samples of Matricaria recutita L. and Chamaemelum nobile L. are analysed as well as Anthemis cotula L., an adulteration of chamomile. First results showed that dichloromethane extracts from M. recutita, C. nobile and A. cotula samples could be clearly classified into three groups corresponding to the species by PCA of 1H NMR spectra. So far, there was no definite discrimination into groups possible when comparing the 1H NMR spectra of methanol extracts by means of PCA using default settings. Moreover, the development of a method with HPLC/ELSD and PCA turned out to be challenging as even slight shifts in retention times led to high variability in PCA.
Acknowledgement: We thank the Federal Institute for Drugs and Medical Devices for financial support.
References: [1] Wang Y et al. (2004) Planta Med. 70: 250–255 [2] Zhao LH et al. (2005) ChemPharmBull 53: 1054–1057 [3] Choi YH et al. (2005) J. Agric. Food Chem. 53: 1237–1245
