Experimental insect suppression causes loss of induced, but not constitutive, resistance in Solanum carolinense.

Spatiotemporal variation in herbivory is a major driver of intraspecific variation in plant defense. Comparatively little is known, however, about how changes in herbivory regime affect the balance of constitutive and induced resistance, which are often considered alternative defensive strategies. Here, we investigated how nearly a decade of insect herbivore suppression affected constitutive and induced resistance in horsenettle (Solanum carolinense), a widespread herbaceous perennial. We allowed replicated horsenettle populations to respond to the presence or absence of herbivores by applying insecticide to all plants in half of 16 field plots. Horsenettle density rapidly increased in response to insecticide treatment, and this effect persisted for at least 4 years after the cessation of herbivore suppression. We subsequently grew half-sibling families from seeds collected during and shortly after insecticide treatment in a common garden and found strong effects of insect suppression on induced resistance. Feeding trials in field mesocosms with false Colorado potato beetles (Leptinotarsa juncta), a common specialist herbivore, revealed that multiyear herbivore suppression drove rapid attenuation of induced resistance: offspring of plants from insect-suppression plots exhibited a near-complete loss of induced resistance to beetles, whereas those from control plots incurred ~70% less damage after experimental induction. Plants from insect-suppression plots also had ~40% greater constitutive resistance compared with those from control plots, although this difference was not statistically significant. We nonetheless detected a strong trade-off between constitutive and induced resistance across families. In contrast, the constitutive expression of trypsin inhibitors (TI), an important chemical defense trait in horsenettle, was reduced by 20% in the offspring of plants from insect-suppression plots relative to those from control plots. However, TIs were induced to an equal extent whether or not insect herbivores had been historically suppressed. Although several defense and performance traits (prickle density, TI concentration, resistance against false Colorado potato beetles and flea beetles, biomass, and seed mass) varied markedly across families, no traits exhibited significant pairwise correlations. Overall, our results indicate that, whereas the divergent responses of multiple defense traits to insect suppression led to comparatively small changes in overall constitutive resistance, they significantly reduced induced resistance against false Colorado potato beetle.