Harnly J, Chen P, Sun J, et al. Comparison of flow injection MS, NMR, and DNA sequencing: methods for identification and authentication of black cohosh (Actaea racemosa). Planta Med. February 2016;82(3):250-262. doi: 10.1055/s-0035-1558113.
Authentication and correct identification of botanicals in dietary supplements is critical for a number of reasons, not the least of which is potential harm from inclusion of undesirable species. The process of identification may be complicated by material processing, variable preparation methods, and the existence of very similar related species. Current methods for identification of botanicals include morphological identification, DNA sequencing, and phytochemical and metabolic fingerprinting. As black cohosh (Actaea racemosa syn. Cimicifuga racemosa, Ranunculaceae) is a popular medicinal plant, adulteration is increasingly common. This basic research study coupled two methods of phytochemical fingerprinting, flow injection mass spectrometry (FIMS) and proton nuclear magnetic resonance (1H NMR), with DNA sequencing to determine how well they could identify different species of Actaea before and after processing.
Root samples from various species of Actaea were procured from American Herbal Pharmacopoeia (AHP), Scotts Valley, California; Strategic Sourcing, Inc., Banner Elk, North Carolina; The North Carolina Arboretum Germplasm Repository (TNCAGR), Asheville, North Carolina; National Institutes of Standards and Technology, Gaithersburg, Maryland; and from Chinese commercial sources. Liquids, tablets, and capsules were bought from local stores in Maryland, and additional samples were obtained from the United States Department of Agriculture, Washington D.C. The roots and commercial samples were authenticated using DNA sequencing at AuthenTechnologies LLC; Richmond, California. DNA sequences from the nuclear ribosomal internal transcribed spacer (ITS) region and the chloroplast psbA-trnH intergenic spacer, obtained by polymerase chain reaction (PCR) and Sanger capillary sequencing, were used for authentication by comparing them to the reference sequences. To prepare samples for FIMS, root material, tablets, and contents of capsules were powdered and extracted in 70:30 methanol:water, sonicated, centrifuged, diluted 1:10 with methanol, and filtered. Liquid extracts were combined with 70:30 methanol:water and treated as described above. To prepare for NMR, samples were dissolved in dimethyl sulfoxide-d6 containing 0.47 mM 4,4-dimethyl-4-silapentane-1-sulfonic acid, vortexed, sonicated, and centrifuged.
Each sample was run on FIMS five times, with random sample order each time; most samples were run once for NMR. Raw data were exported into various computer programs for analysis. Data were normalized to unit vector length (this metric incorporates the sum of squares) and analyzed using principal component analysis (PCA, an analysis that displays data based on greatest variation) and four multivariate analysis methods, especially soft independent modeling of class analogy (SIMCA, an analysis based on modeling that fits a model to a single class, which is explained to be useful for authentication). Both methods of analysis were used to assess whether FIMS or NMR could distinguish various Actaea species. Because of the complexity of PCA plots in including all samples, the authors chose to analyze the data from only the following five American species: black cohosh, bugbane (A. cimicifuga), baneberry (A. pachypoda), American bugbane (A. podocarpa), and red baneberry (A. rubra).
In analyses of both FIMS and NMR data, samples of black cohosh generally clustered separately from other species; some analyses portray it as most similar to baneberry, suggesting a close relation, and more distant from other species. One sample of black cohosh was found to occur outside the black cohosh cluster, and one sample of baneberry clustered with black cohosh; repeated analyses starting from new raw material had the same result. Significant phytochemical variation was observed within species; in the FIMS analysis, samples from multiple sites supplied by TNCAGR clustered according to site of origin. The authors noted that the potential causes of geographically correlated variation include not only genetic differences and environmental influences on the plants, but variation in the presence of endophytic fungi, which are suspected of producing some of the secondary compounds in black cohosh. The models for FIMS and NMR were validated using bootstrap analysis, despite the authors' concern about limited sample size. It was found that FIMS had a sensitivity of 91.4% and NMR had a sensitivity of 91.7%, with specificities of 100% for both. Overall, it was concluded that both FIMS and NMR are appropriate techniques for distinguishing the five species of Actaea, with sensitivity to phytochemical composition.
For North American samples of multiple species, putative identities according to DNA sequencing were generally consistent with the AHP identification, although sequencing quality problems sometimes precluded identification. The two samples that seemed to be outliers (black cohosh and baneberry) in the FIMS and NMR analyses had DNA sequences consistent with the stated identity. This suggests that the species are biologically complex species with morphological or chemical variations greater than the genetic variations. Very few of the geographically diverse black cohosh samples from TNCAGR, which were subjected to chemical analysis, were included in DNA analyses.
Root samples procured from China were stated to belong to four different species;
nuclear ribosomal DNA (nrDNA) and chloroplast DNA (cpDNA) sequences indicated that the majority of samples, including one purported sample of A. racemosa, were Chinese cimicifuga (A. dahurica) and one was A. brachycarpa, while two contained non-Actaea spp. plant material and another had only fungal DNA amplified. All of these, as well as some North American Actaea spp. samples, were said to be "mixtures" according to ITS data, and to some extent according to psbA-trnH data; it appears that this simply means that more than one ITS sequence variant was present in some samples, while other samples may contain hybrid plant material. DNA analysis confirmed that four of the seven capsule supplements purchased locally contained black cohosh, while one contained A. brachycarpa; the other two had no detectable Actaea spp. DNA (one had no DNA and the other had rice [Oryza sativa, Poaceae] DNA or excipient DNA only). Liquid extract supplements were not analyzed with DNA sequencing since they were not expected to contain retrievable DNA. In summary, the FIMS and NMR analyses showed diverse chemical profiles that were different from any of the raw materials tested, which is presumably due to differences in processing and extraction methods or systematic geographical variations. Moreover, the DNA sequencing of roots could be used for identification of Actaea species and also for validating the discrepancies that might arise in the chemical fingerprinting methods.
In conclusion, the three methods outlined here may be useful to aid in authentication of black cohosh in commerce and identify possible adulterants. However, more experimentation and adjustment of methods would be necessary to analyze finished botanical supplements.
—Amy C. Keller, PhD
