Evaluation of DNA barcoding coupled high resolution melting for discrimination of closely related species in phytopharmaceuticals

Background: Phytopharmaceuticals are increasingly popular as alternative medicines, but poorly regulated in many countries. The manufacturers of these products should be subject to strict controls regarding each product's quality and constituents. Routine testing and identification of raw materials should be performed to ensure that the raw materials used in pharmaceutical products are suitable for their intended use. Hypothesis/purpose: We have applied DNA Barcoding High Resolution Melting (Bar-HRM), an emerging method for identifying of medicinal plant species based on DNA dissociation kinetics and DNA barcoding, for the authentication of medicinal plant species. Study design: Commonly commercialized Thai medicinal plants that are widely used for medicinal purposes were used in this study. Publicly available sequences of four plastid markers were used for universal primer design. Species discrimination efficiency of the designed primers was evaluated as single and multi-locus analyses by using the primers sets. Methods: HRM analysis was performed in triplicate on each of the 26 taxa to establish the T-m for each primer set (matK, rbcLA, rbcLB, rbcLC, rpoC1, and trnL). The shapes of the melting curves were analyzed to distinguish the different plant species. Bar-HRM species identification success rates were assessed for each single-locus as well as for multi-locus combinations to establish the optimal combination of primer sets. Results: In single locus analysis the rpoC1 primer set gave the highest discrimination (53%), and in multi locus analysis this could be increased from 87% to 99% depending on the total number of regions included. Different combinations proved to be more or less effective at discrimination, depending on the genus or family examined. Conclusions: Bar-HRM has proven to be a cost-effective and reliable method for the identification of species in this study of Thai medicinal plants, and results show an identification success rate of 99% among species in the test set. (C) 2016 Elsevier GmbH. All rights reserved.