Quantifying the roles of environmental, seasonal, and genome size variation in the expression of medicinally valuable compounds in Helichrysum kraussii Sch.Bip

Abstract

South Africa is home to a vast array of medicinal plant species. Variations in their secondary metabolomes determine their medicinal potential. Intraspecific variations in these chemical profiles are commonplace and are caused by numerous genetic and environmental factors, including epigenetic modifications, point mutations, whole genome duplication, aneuploidy, topography, soil type, climate, and biotic interactions with herbivores and pollinators. Despite the prevalence of inter- and intrapopulation variation in geno- and phenotypes, its importance in medicinal plant research is often not considered and may influence the robustness of inferences made regarding a species’ medicinal value. The present study first aimed to establish intraspecific variation in the secondary metabolite profiles, antibacterial activity, and genome sizes of Helichrysum kraussii Sch. Bip and second elucidate the effects that geographic isolation and seasonal differences have on these attributes. Plant material was collected from nine populations and two seasons (autumn and summer). Ethanolic extracts were prepared from each of the 93 samples and used in high-performance thin-layer chromatography to evaluate their secondary metabolite profiles. Statistical analyses included principal component analysis (PCA), agglomerative hierarchical clustering (AHC) with bootstrap, and k-means analysis to determine how samples could be grouped based on similarities and differences in their secondary metabolite profiles. Permutational analysis of variance (PERMANOVA) using population, season, and their interaction was performed to determine whether these factors significantly contributed to the clustering of the samples. Partial least squares discriminant analysis was used to elucidate whether samples could be divided into specific chemotypes and which potential compounds contributed most toward their distinction. From the results, it was evident that sample provenance, collection period, and their interaction significantly (p < 0.05) influenced disparities in secondary metabolite profiles. Mantel tests using geographic distance and secondary metabolite profiles revealed that the entire dataset and the autumn subset showed poor correlations, R2 = 0.185, p = 0.263 and R2 = 0.213, p = 0.208, between spatial separation and interpopulation variation in secondary metabolite profiles, respectively. The same analysis on the summer subset indicated a significant negative correlation between population location and secondary metabolite profiles with R2 = -0.396 and p = 0.04. Samples were clustered into two chemotypes with robust support (5-fold validation indicating Q2 = 0.6, and permutation analysis yielding p < 0.01) by Partial Least Squares Discriminant Analysis (PLS-DA). Kaurenoic acid (KA), a ubiquitous secondary metabolite found across numerous plant families with several reported medicinal properties, including antibacterial activity, was present in the extracts, however, concentrations differed amongst the samples and were shown to be significantly influenced by the interaction of season and population by two-way analysis of variance (p < 0.05) but only between five population and season pairs. The antibacterial activity of some samples, as selected based on their cluster assignments, was evaluated against Cutibacterium acnes. The half-maximal inhibitory concentration (IC50) values of these samples ranged between 17.10 ± 4.16 μg/mL and 391.16 ± 15.02 μg/mL, whereas KA exhibited potent (IC50 = 37.94 ± 2.73 μg/mL) antibacterial activity. However, a linear regression model revealed that a poor correlation existed between the samples’ antibacterial activity and KA concentrations, suggesting that a different compound, or set of compounds, was responsible for the observed antibacterial activity. Flow cytometry, conducted using the Mountain Sanctuary Park, Groenkloof, Faerie Glen, Hedianga Farm, and Skuilkrans Kopje populations, indicated the presence of significant interpopulation differences in relative genome size when Pisum sativum L. was used as a standard. No correlation could, however, be found between secondary metabolite profiles and genome size differences. This study concluded that season, population, and their interaction drive interpopulation variation in secondary metabolite profiles in Helichrysum kraussii, but that these factors were not sufficient predictors of variance. The PLS-DA suggested that the H. kraussii samples were better categorised into two clusters, where no pattern regarding the samples’ population and season identity could be discerned. The previous studies on intraspecific variation within medicinal plant species suggest that variation in secondary metabolomes and medicinal activity is not limited to H. kraussii, and it is therefore recommended that medicinal plant research incorporates potential inter- and intrapopulation variation into its methodologies to ensure robust inferences regarding a species’ medicinal value.