The use of 1H-NMR spectroscopy and chemometrics of pattern recognition for authentication of Curcuma xanthorrhiza adulterated with Zingiber montanum

Abstract

Curcuma xanthorrhiza is widely used in food and traditional medicine products. Due to its high demand, it is potential to be substituted or mixed with other species having similar appearance, therefore, rapid and reliable analytical method is highly required. The objective of this study was to develop 1H-NMR spectroscopy and chemometrics of pattern recognition as a metabolite fingerprinting technique for authentication of C. xanthorrhiza from Zingiber montanum. The powdered rhizomes were extracted using combination of methanol-D4 and phosphate buffer pH 6.0 in deuterium oxide (1:1 v/v) containing trimethylsilyl propionic acid (TSP) for chemical shift calibration. The variables extracted from 1H-NMR spectra were used for creating chemometrics models. Chemometrics of partial least square-discriminant analysis (PLS-DA) using 7 principal components (PCs) successfully classified between authentic and adulterated samples of C. xanthorrhiza with high value of R2X (0.988), R2Y (0.998), and Q2 (0.993). Moreover, chemometrics of orthogonal projection to latent structures-discriminant analysis (OPLS-DA) using 2 PCs and 4 orthogonal components perfectly discriminated authentic and adulterated samples of C. xanthorrhiza. The model showed high R2X (0.965), R2Y (0.976) as well as Q2 (0.946) values. Validation using permutation test confirmed the validity both PLS-DA and OPLS-DA models. It suggested that combination of 1H-NMR and chemometrics method is promising for authentication of medicinal plants.