Black Cohosh (BC, Actaea racemosa (L.), Ranunculaceae) is a widely used herbal remedy. Wild-harvest in multi-tons leads to endangerment of the species. Efforts are made to establish cultivation. Nevertheless, different BC genotypes may exist [1], and physiography influences the rhizomes’ chemical composition [2]. This can influence efficacy of herbal products.
In this study, we chemically profile BC rhizomes from cultivation and wild-harvests. By non-targeted secondary metabolite profiling via LC/MS we explore differences between the two. We analysed methanol rhizome extracts of 42 specimen, from wild-harvests in the US (n=8), and from worldwide sources (n=26) together with clone-plants (vegetative propagation, n=8), both cultivated in Berlin. We generated metabolic data by high-resolution UPLC/ESI-QToF-MS analyses. For multivariate analysis, binning (bin-size m/z 2) was performed [3].
Generally, composition of extracts from wild plants showed quantitative differences in comparison to cultivated. This is in line with previous findings [4]. In hierarchical cluster analysis (HCA), both entities grouped separately. Especially, mass spectral bins representing triterpenes (37 bins with FC≥2/–LOG10(p)≥4 (t-tests); i.e. bins m/z 645_647 (FC=3.1/–LOG10(p)=10.3) or 661_663 (FC=2.3/–LOG10(p)=6.2) were significantly more abundant in wild. Bins indicating fragments of polyphenols in (+)-ESI, like m/z 177_179 (FC=1.2/-LOG10(p)=1.9) were not highly changed. In HCA of only cultivated plants samples clustered, too, supporting chemotype existence. Furthermore, clone-plants built a dense group, indicating homogeneity.
Growing conditions influence the compositions of extracts. Differences between i.e. wild/cultivated/chemotypes should be considered during clinical efficacy evaluation.
The authors thank Maged Sharaf, Ph.D. (AHPA) for provision of wild-harvested specimens.
[1]Motley TJ et al. Proceedings of the Global Summit on Medicinal Plants 2004; 1: 112-118
[2]Vickers A et al. Am J Plant Sci 2015;6:3272-3281
[3]Grace SC et al. BMC bioinformatics 2014;15Suppl11:S12
[4]Bittner M et al. Planta Med 2017;DOI:10.1055/s-0043-108122