LB 39-327 - The roots (and leaves, and seeds) of chemical defense variation: Secondary metabolite allocation across above-belowground axes and across species in 10 sympatric Psychotria (Rubiaceae)

Plant-antagonist interactions, both above- and below-ground, have been implicated in numerous aspects of the evolution and maintenance of biodiversity. Leaf, seed, and root tissues all experience selective pressure of varying degrees to defend themselves against antagonists, and in each of these tissue types, secondary metabolites are known as a widespread and functionally diverse form of defense. As leaves, seeds, and roots of a given plant species vary in fitness value and in physiological constraints on secondary metabolite investment, these tissues can be expected to vary in their secondary metabolite content. Yet it remains to be established whether variation among tissues in secondary metabolite content is characterized by tissue-level selective pressures or by species-level growth-defense tradeoffs, particularly in tropical systems where above- and below-ground antagonist pressure is generally high. To address this question, we compared the secondary metabolite content of leaf, seed, and root tissues in 10 sympatric species of Psychotria (Rubiaceae) in the moist seasonal forest of Barro Colorado Island, Panama. We collected and freeze-dried three samples per tissue type, per species, and extracted 80 mg of dried tissue using acidified ethanol. The extracts were then fully dried and weighed before being resuspended in aqueous acetonitrile for UPLC-MS/MS untargeted metabolomic analysis. We used linear regression to compare the tissue-level relationships of secondary metabolite extract masses to one another across species, finding significant (P < 0.015) positive correlations between root and leaf extract masses and between root and seed extract masses, but not between leaf and seed extract masses (P = 0.12). Next, we evaluated the relative contribution of species and tissue identity to the variance of extract mass using a permutational univariate ANOVA model. This model indicated that both tissue, species identity, and their interaction all explained a significant (P < 0.001) portion of the variance, but species identity accounted for much more of the variation (44%) than did tissue (10%) or the species-tissue interaction term (17%). Metabolomic data analysis is in progress and will allow us to probe the compound-level aspects of these summary quantitative results. Our comparisons of secondary metabolite extract masses suggest that quantitative secondary metabolite investment behaves as a species-level trait within Psychotria, with leaf, root, and seed tissues exhibiting similar quantitative means and variance despite their physiological dissimilarities.