Determination of Polysaccharide Hydrolyzates in Chinese Herbal Medicine by UltraHigh-Performance Liquid Chromatography-Quadrupole Time-of-Flight Mass Spectrometry and Evaporative Light Scattering Detection. (13th August 2018)

• Record Type: Journal Article
• Authors: Zhong, Peipei; Chen, Weiling; Chen, Yi; Yu, Qing; Wang, Yuanxing
• Is Part Of: Analytical letters. Volume 51:Number 12(2018)
• Page Start: 1826
• Page End: 1839
• Publication Date: 2018-08-13
• Subjects: Chemometric tool -- evaporative light scattering detection -- hydrolyzate -- polysaccharide -- ultrahigh-performance liquid chromatography-quadrupole time-of-flight mass spectrometry; Chemistry, Analytic -- Periodicals; Chemistry, Analytic -- Abstracts; 543
• DOI: 10.1080/00032719.2017.1396337
• Languages: English
• ISSNs: 0003-2719

Abstract

ABSTRACT: Ultrahigh-performance liquid chromatography coupled with evaporative light scattering detector was used to optimize acid hydrolysis conditions of polysaccharides from Dendrobium candidum . The optimized conditions were the concentration of H2 SO4, 0.4 M, hydrolysis temperature, 90°C, and hydrolysis time, 1 h. Ultrahigh-performance liquid chromatography-quadrupole time-of-flight mass spectrometry was used to obtain the total ion current chromatograms and determine the hydrolyzates of polysaccharides, such as monosaccharides, oligosaccharides, and some low-molecular weight polysaccharides from D. candidum, Cordyceps sinensis, and Ganoderma atrum . The accurate mass data were analyzed by chemometric tools, including an unsupervised analytical method, principal component analysis (one-way ANOVA, p < 0.05), and five supervised analytical methods, partial least square discriminant analysis, support vector machine, naïve Bayes, decision tree, and neural network. The results showed that D. candidum polysaccharides from different locations and grades can be discriminated effectively by the principal component analysis scores. The quality predictive models showed a high value with a low prediction error calculated from both validation and training sets.