HPTLC and DNA Metabarcoding Analysis of 53 Commercial Echinacea Dietary Supplement Products

Reviewed: Raclariu AC, Ţebrencu CE, Ichim MC, Ciupercǎ OT, Brysting AK, De Boer H. What's in the box? Authentication of Echinacea herbal products using DNA metabarcoding and HPTLC. Phytomedicine. 2018;44:32-38.

Keywords: DNA metabarcoding, echinacea, Echinacea angustifolia, Echinacea pallida, Echinacea purpurea, HPTLC, quality control

Research from Germany in the 1980s and 1990s suggested that confusion among the medicinally used Echinacea angustifolia (Asteraceae), E. pallida, and E. purpurea, and substitution with the roots of Parthenium integrifolium (Asteraceae) was quite common in these decades.^1,2 While investigations into the authenticity of commercial echinacea products have been scarce in the past years, to our knowledge, there have not been any recent reports of adulteration issues with P. integrifolium.

In order to evaluate the composition of echinacea products, 53 commercial samples from Romania (n = 26) or other European countries (n = 15), China (n = 1), the United States (n = 7), or of unknown origin (N = 4), were analyzed by high-performance thin-layer chromatography (HPTLC) and DNA metabarcoding. The latter is a term generally used for the analysis of all DNA in a sample using a next-generation sequencing instrument. The nuclear ribosomal internal transcribed spacer (ITS) regions 1 and 2 were used as barcodes. For the HPTLC analysis, conditions specified in the European Pharmacopoeia 8 were applied.^3-6

Products were obtained in form of teas (n = 17), tablets (n = 16), capsules (n = 13), and liquid extracts (n = 7). The results indicate that a vast majority of the samples contain echinacea, although a few products contain material from unlabeled Echinacea species. In four products (7.5%), no phenolic compounds could be detected by HPTLC, and three additional products (5.7%) had very low amounts. Two glycerin extracts could not be evaluated due to the low solubility of glycerin and the inherent problems to obtain reproducible results with these extracts with HPTLC. Suitable DNA for metabarcoding was obtained for 38 products, of which 34 (89.5%) contained echinacea DNA. It is not clear if the remaining four products were devoid of echinacea, or if the DNA was too fragmented to allow detection.

Of the single-ingredient echinacea products (n = 27), 24 contained echinacea DNA. Not surprisingly, a majority were found to contain DNA from other plant sources, most likely from environmental contaminants or accidental contamination during the harvesting and processing stages. The most frequently detected contaminant species included burdock (Arctium lappa, Asteraceae), greater celandine (Chelidonium majus, Papaveraceae), horseweed (Erigeron canadensis, Asteraceae), lemon balm (Melissa officinalis, Lamiaceae), Mongolian thyme (Thymus mongolicus, Lamiaceae), or grass (Poaceae) species such as Echinochloa colona.

Comment: The study suggests that most of the products analyzed contained one or several Echinacea species. While discrepancies from the expected HPTLC fingerprint may be explained by the low stability of phenolic compounds in liquids (in particular those containing substantial amounts of water), most of the products of doubtful quality were sold as herbal tea, tablet, or capsule. Looking at these questionable fingerprints (especially those lanes where only very weak or no bands were detected), it is difficult to explain the pattern solely by differences in the manufacturing processes, or by the presence of appropriately labeled additional herbal ingredients.

All of the 53 samples yielded DNA, but only 72% (38 of 53 samples) provided DNA of sufficient quality to allow genus or species identification, while no identifiable genetic marker was obtained for the other 15 products. In addition to echinacea DNA, a large number of additional species were apparently detected. The authors reported a 43% ingredient fidelity (a term to describe the number of species detected compared to those indicated on the label) based on the DNA metabarcoding results. It is crucial to put such statements in context of the research, as this percentage can be easily misinterpreted by non-experts in the field as the number of products that are acceptable. As we have stated repeatedly in previous issues of this newsletter, herbal ingredients normally contain DNA from other plant materials (e.g., pollen, inadvertent contamination during harvest, drying, processing, etc.). For this reason, pharmacopeial monographs allow for a specific amount (usually up to 2%) of foreign organic matter. As such, the value of DNA metabarcoding results listing dozens of contaminants, which are likely present at levels well below 2%, is questionable.

References

1. Bauer R, Khan IA, Wagner H. 1987. Echinacea: Nachweis einer Verfälschung von Echinacea purpurea (L.) Moench. mit Parthenium integrifolium L. Dtsch Apoth Ztg. 127: 1325–1328.
2. Bauer R. Chemistry, pharmacology and clinical applications of echinacea products. In: Mazza G, Oomah BD (eds). Herbs, Botanicals and Teas. Boca Raton, FL: CRC Press; 1998:45-73.
3. Echinacea angustifoliae radix. European Pharmacopoeia (EP 8.0). Strasbourg, France: European Directorate for the Quality of Medicines and Health Care; 2012:1327-1329
4. Echinaceae pallidae radix. European Pharmacopoeia (EP 8.0). Strasbourg, France: European Directorate for the Quality of Medicines and Health Care; 2012:1345-1346.
5. Echinaceae purpureae herba. European Pharmacopoeia (EP 8.0). Strasbourg, France: European Directorate for the Quality of Medicines and Health Care; 2012:1357-1359.
6. Echinaceae purpureae radix. European Pharmacopoeia (EP 8.0). Strasbourg, France: European Directorate for the Quality of Medicines and Health Care; 2012:1359-1361.