Subscribe to RSS
DOI: 10.1055/a-0639-5450
An Integrated Strategy to Qualitatively Differentiate Components of Raw and Processed Viticis Fructus Based on NIR, HPLC and UPLC-MS Analysis
Publication History
received 26 April 2018
revised 20 May 2018
accepted 30 May 2018
Publication Date:
21 June 2018 (online)


Abstract
Viticis Fructus is a traditional Chinese herbal drug processed by various methods to achieve different clinical purposes. Thermal treatment potentially alters chemical composition, which may impact on effectiveness and toxicity. In order to interpret the constituent discrepancies of raw versus processed (stir-fried) Viticis Fructus, a multivariate detection method (NIR, HPLC, and UPLC-MS) based on metabonomics and chemometrics was developed. Firstly, synergy interval partial least squares and partial least squares-discriminant analysis were employed to screen the distinctive wavebands (4319 – 5459 cm−1) based on preprocessed near-infrared spectra. Then, HPLC with principal component analysis was performed to characterize the distinction. Subsequently, a total of 49 compounds were identified by UPLC-MS, among which 42 compounds were eventually characterized as having a significant change during processing via the semiquantitative volcano plot analysis. Moreover, based on the partial least squares-discriminant analysis, 16 compounds were chosen as characteristic markers that could be in close correlation with the discriminatory near-infrared wavebands. Together, all of these characterization techniques effectively discriminated raw and processed products of Viticis Fructus. In general, our work provides an integrated way of classifying Viticis Fructus, and a strategy to explore discriminatory chemical markers for other traditional Chinese herbs, thus ensuring safety and efficacy for consumers.
Key words
Viticis Fructus - Vitex trifolia - Lamiaceae - liquid chromatography-tandem mass spectrometry - near-infrared - multivariate statistical analysis - quality controlSupporting Information
- Supporting Information
The supporting information includes details of the samples and reference compounds, NIR chemometric models, pretreatment and screened wavebands, HPLC method validation, and fingerprints similarity evaluation. Moreover, RT and MS spectra for the 49 reference substances and potential discriminated compounds in the tested samples are also available as Supporting Information.