The interplay between toxin-releasing β-glucosidase and plant iridoid glycosides impairs larval development in a generalist caterpillar, Grammia incorrupta (Arctiidae)

Abstract

Herbivores with polyphagous feeding habits must cope with a diet that varies in quality. One of the most important sources of this variation in host plant suitability is plant secondary chemistry. We examined how feeding on plants containing one such group of compounds, the iridoid glycosides, might affect the growth and enzymatic activity in a polyphagous caterpillar that feeds on over 80 plant species in 50 different families. Larvae of the polyphagous arctiid, Grammia incorrupta, were reared exclusively on one of two plant species, one of which contains iridoid glycosides (Plantago lanceolata, Plantaginaceae) while the other does not (Taraxacum officinale, Asteraceae). Larval weight was measured on the two host plants, and midgut homogenates of last instar larvae were then assayed for activity and kinetic properties of β-glucosidases, using both a standard substrate, 4-nitrophenyl-β-D-glucose (NPβGlc), and the iridoid glycoside aucubin, one of the two main iridoid glycosides in P. lanceolata. Larvae feeding on P. lanceolata weighed significantly less and developed more slowly compared to larvae on T. officinale. While the larval midgut β-glucosidase activity determined with NPβGlc was significantly decreased when fed on P. lanceolata, aucubin was substantially hydrolyzed and the larval β-glucosidase activity towards both substrates correlated negatively with larval weight. Our results demonstrate that host plants containing high concentrations of iridoid glycosides have a negative impact on larval development of this generalist insect herbivore. This is most likely due to the hydrolysis of plant glycosides in the larval midgut which results in the release of toxic aglycones. Linking the reduced larval weight to the toxin-releasing action of an iridoid glycoside cleaving β-glucosidase, our results thus support the detoxification limitation hypothesis, suggesting fitness costs for the larvae feeding solely on P. lanceolata. Thus, in addition to the adaptive regulation of midgut β-glucosidase activity, host plant switching as a behavioral adaptation might be a prerequisite for generalist herbivores that allows them to circumvent the negative effects of plant secondary compounds.