If you have any problems related to the accessibility of any content (or if you want to request that a specific publication be accessible), please contact us at firstname.lastname@example.org.
Shedding Light on Chemically Mediated Tri-Trophic Interactions: A H-1-NMR Network Approach to Identify Compound Structural Features and Associated Biological Activity
AuthorRichards, Lora A.
Dyer, Lee A.
Wallace, Ian S.
Dodson, Craig D.
Jeffrey, Christopher S.
AltmetricsView Usage Statistics
Diverse mixtures of plant natural products play an important role in plant-herbivore-parasitoid interactions. In the pursuit of understanding these chemically-mediated interactions, we are often faced with the challenge of determining ecologically and biologically relevant compounds present in complex phytochemical mixtures. Using a network-based approach, we analyzed binned H-1-NMR data from 196 prepared mixtures of commonly studied secondary metabolites including alkaloids, amides, terpenes, iridoid glycosides, saponins, phenylpropanoids, flavonoids and phytosterols. The mixtures included multiple dimensions of chemical diversity, including molecular complexity, mixture complexity and differences in relative compound concentrations. This approach yielded modules of co-occurring chemical shifts that were correlated with specific compounds or common structural features shared across compounds. This approach was then applied to crude phytochemical extracts of 31 species in the phytochemically diverse tropical plant genus Piper (Piperaceae). Combining the activity of crude plant extracts in an array of bioassays with our H-1-NMR network approach, we identified a potential prenylated benzoic acid from these mixtures that exhibits antifungal properties and identified small structural differences that were potentially responsible for antifungal activity. In an intraspecific analysis of individual Piper kelleyi plants, we also found ontogenetic differences in chemistry that may affect natural plant enemies. In sum, this approach allowed us to characterize mixtures as useful network modules and to combine chemical and ecological datasets to identify biologically important compounds from crude extracts.
|Attribution 4.0 International