Regulation of woody plant secondary metabolism by resource availability: hypothesis testing by means of meta-analysis

    loading  Checking for direct PDF access through Ovid


Our aim in this study was to determine how well phenotypic variation in foliar concentrations of carbon-based secondary compounds (CBSCs) in woody plants can be predicted on the basis of two resource-based hypotheses, i.e. the carbon-nutrient balance (CNB) and growth-differentiation balance (GDB) hypotheses. We conducted a meta-analysis of literature data with respect to responses of CBSCs, carbohydrates and nitrogen to six types of environmental manipulations (fertilization with nitrogen or phosphorus, shading, CO2 enrichment, drought stress, ozone exposure). Plant responses to nitrogen fertilization, shading and CO2 enrichment in terms of pooled CBSCs and carbohydrates were consistent with predictions made with the two hypotheses. However, among biosynthetically distinct groups of CBSCs only concentrations of phenylpropanoid-derived compounds changed as predicted; hydrolyzable tannins and terpenoids, in particular, were less responsive. Phosphorus fertilization did not affect concentrations of CBSC or primary metabolites. Plant responses to drought and ozone exposure presumably were driven by plant demands for particular types of compounds (osmolites in the case of drought and antioxidants in the case of ozone exposure) rather than by changes in resource availability. Based on the relative importance of the treatment effects, we propose a hierarchical model of carbon allocation to CBSCs. The model implies that CBSC production is determined by both resource availability and specific demand-side responses. However, these two mechanisms work at different hierarchical levels. The domain of the CNB and GDB hypotheses is at the high hierarchical levels, predicting the total amount of carbon that can be allocated to CBSCs. Predicting altered concentrations of individual CBSCs, i.e. low hierarchy levels, probably demands biosynthetically detailed models which also take into account the history of plant interactions with biotic and abiotic factors.

    loading  Loading Related Articles