Abstract
A recent surge in attention devoted to the ecology of soil biota has prompted
interest in quantifying similarities and differences between interactions occurring in
above- and belowground communities. Furthermore, linkages that interconnect the dynamics
of these two spatially distinct ecosystems are increasingly documented. We use a similar
approach in the context of understanding plant defenses to herbivory, including how they are
allocated between leaves and roots (constitutive defenses), and potential cross-system linkages
(induced defenses). To explore these issues we utilized three different empirical approaches.
First, we manipulated foliar and root herbivory on tobacco (Nicotiana tabacum) and measured
changes in the secondary chemistry of above- and belowground tissues. Second, we reviewed
published studies that compared levels of secondary chemistry between leaves and roots to
determine how plants distribute putative defense chemicals across the above- and
belowground systems. Last, we used meta-analysis to quantify the impact of induced
responses across plant tissue types.
In the tobacco system, leaf-chewing insects strongly induced higher levels of secondary
metabolites in leaves but had no impact on root chemistry. Nematode root herbivores,
however, elicited changes in both leaves and roots. Virtually all secondary chemicals measured
were elevated in nematode-induced galls, whereas the impact of root herbivory on foliar
chemistry was highly variable and depended on where chemicals were produced within the
plant. Importantly, nematodes interfered with aboveground metabolites that have biosynthetic
sites located in roots (e.g., nicotine) but had the opposite effect (i.e., nematodes elevated
foliar expression) on chemicals produced in shoots (e.g., phenolics and terpenoids).
Results from our literature review suggest that, overall, constitutive defense levels are
extremely similar when comparing leaves with roots, although certain chemical classes (e.g.,
alkaloids, glucosinolates) are differentially allocated between above- and belowground parts.
Based on a meta-analysis of induced defense studies we conclude that: (1) foliar induction
generates strong responses in leaves, but much weaker responses in roots, and (2) root
induction elicits responses of equal magnitude in both leaves and roots. We discuss the
importance of this asymmetry and the paradox of cross-system induction in relation to
optimal defense theory and interactions between above- and belowground herbivory.
