Plants are often attacked by multiple enemies, including pathogens and herbivores. While many plant secondary metabolites show specific effects toward either pathogens or herbivores, some can affect the performance of both these groups of natural enemies and are considered to be "generalized defense compounds". We tested whether aucubin and catalpol, two iridoid glycosides present in ribwort plantain (Plantago lanceolata), confer in vivo resistance to both the generalist insect herbivore Spodoptera exigua and the biotrophic fungal pathogen Diaporthe adunca using plants from P. lanceolata lines that had been selected for high- and low-leaf iridoid glycoside concentrations for four generations. The lines differed approximately three-fold in the levels of these compounds. Plants from the high-selection line showed enhanced resistance to both S. exigua and D. adunca, as evidenced by a smaller lesion size and a lower fungal growth rate and spore production, and a lower larval growth rate and herbivory under both choice and no-choice conditions. Gravimetric analysis revealed that the iridoid glycosides acted as feeding deterrents to S. exigua, thereby reducing its food intake rate, rather than having post-ingestive toxic effects as predicted from in vitro effects of hydrolysis products. We suggest that the bitter taste of iridoid glycosides deters feeding by S. exigua, whereas the hydrolysis products formed after tissue damage following fungal infection mediate pathogen resistance. We conclude that iridoid glycosides in P. lanceolata can serve as broad-spectrum defenses and that selection for pathogen resistance could potentially result in increased resistance to generalist insect herbivores and vice versa, resulting in diffuse rather than pairwise coevolution.

Plantago lanceolata L. (ribwort plantain) produces two costly terpenoid secondary plant compounds, the iridoid glycosides aucubin and catalpol. We performed an artificial selection experiment to investigate direct and correlated responses to selection on the constitutive level of iridoid glycosides in the leaves for four generations. Estimated realized heritabilities (±SE) were 0.23 ± 0.07 and 0.23 ± 0.04 for upward and downward selection, respectively. The response to upward selection was caused by selection for a developmental pattern characterized by the production of fewer leaves that on average contain more iridoids, and by selection for a development‐independent increase in the level of these compounds. Significant correlated responses were observed for plant growth form. Upward selection resulted in plants with larger sized, but fewer leaves, fewer side rosettes, and fewer spikes, corresponding to a previously distinguished ‘hayfield’ ecotype, whereas downward selection produced the opposite pattern, corresponding to a ‘pasture’ ecotype. This indicates that the level of iridoid glycosides is genetically correlated with morphological traits in P. lanceolata, and is part of the complex of genetically correlated traits underlying the two ecotypes. The genetic association between iridoid level and growth forms suggests that there may be constraints to the simultaneous evolution of resistance to generalist insects (by iridoid glycosides) and to larger grazers (by a high production rate of prostrate leaves and inflorescences) in open grazed habitats where the ‘pasture’ ecotype is found.

Many secondary plant compounds are involved in defense against both insect herbivores and pathogens. Two secondary plant compounds of Plantago lanceolata, the iridoid glycosides catalpol and its precursor aucubin, are well known for their deterrent effects on generalist and nonadapted specialist insect herbivores. We tested the effects of these compounds on the in-vitro growth of a specialist and generalist fungal pathogen of this host species. Two chemical forms of these iridoids were tested. The glycosides and their aglycones, the products of enzymatic conversion by specific β-glucosidase enzymes. The glycosides enhanced growth of both the specialist fungus Diaporthe adunca and the generalist fungus Fusarium moniliforme var. subglutinans. The positive effect of these glycosides on the generalist fungus is in sharp contrast with the generally negative effects of these glysosides on generalist insect herbivores. The aglycones of aucubin and catalpol reduced the growth of the specialist fungus D. adunca, but, contrary to expectation, enhanced the growth of the generalist fungus F. moniliforme var. subglutinans. Effects of aucubin on D. adunca were stronger than effects of catalpol. This was true both for the growth stimulating effects of the glycosides and for the fungitoxic effects of the aglycones. We therefore expect that the effects of these iridoids in P. lanceolata on the specialist fungus will strongly depend on the ratio between catalpol and its precursor aucubin and the chemical form (glycoside or aglycone) in which these compounds are encountered by the fungus during growth. Our results suggest that iridoid glycosides in P. lanceolata can be used as defense against both herbivores and pathogens, but that their effects are highly specific with respect to the natural enemy species that is encountered.

Abstract. Fitness costs of defense are often invoked to explain the maintenance of genetic variation in levels of chemical defense compounds in natural plant populations. We investigated fitness costs of iridoid glycosides (IGs), terpenoid compounds that strongly deter generalist insect herbivores, in ribwort plantain (Plantago lanceolata L.) using lines that had been artificially selected for high and low leaf IG concentrations for four generations. Twelve maternal half-sib families from each selection line were grown in four environments, consisting of two nutrient and two competition treatments. We tested whether: (1) in the absence of herbivores and pathogens, plants from lines selected for high IG levels have a lower fitness than plants selected for low IG levels; and (2) costs of chemical defense increase with environmental stress. Vegetative biomass did not differ between selection lines, but plants selected for high IG levels produced fewer inflorescences and had a significantly lower reproductive dry weight than plants selected for low IG levels, indicating a fitness cost of IG production. Line-by-nutrient and line-by-competition interactions were not significant for any of the fitness-related traits. Hence, there was no evidence that fitness costs increased with environmental stress. Two factors may have contributed to the absence of higher costs under environmental stress. First, IGs are carbon-based chemicals. Under nutrient limitation, the relative carbon excess may result in the production of IGs without imposing a further constraint on growth and reproduction. Second, correlated responses to selection on IG levels indicate the existence of a positive genetic association between IG level and cotyledon size. At low nutrient level, a path analysis based on family means revealed that in the presence of competitors, the negative direct effect of a high IG level on aboveground plant dry weight was partly offset by a positive direct effect of the associated larger cotyledon size. This indicates that fitness costs of defense may be modulated by environment-specific fitness effects of genetically associated traits.

No abstract