Effects of Quantitative Variation in Allelochemicals in <Emphasis Type="Italic">Plantago lanceolata</Emphasis> on Development of a Generalist and a Specialist Herbivore and their Endoparasitoids

Studies in crop species show that the effect of plant allelochemicals is not necessarily restricted to herbivores, but can extend to (positive as well as negative) effects on performance at higher trophic levels, including the predators and parasitoids of herbivores. We examined how quantitative variation in allelochemicals (iridoid glycosides) in ribwort plantain, Plantago lanceolata, affects the development of a specialist and a generalist herbivore and their respective specialist and generalist endoparasitoids. Plants were grown from two selection lines that differed ca. 5-fold in the concentration of leaf iridoid glycosides. Development time of the specialist herbivore, Melitaea cinxia, and its solitary endoparasitoid, Hyposoter horticola, proceeded most rapidly when reared on the high iridoid line, whereas pupal mass in M. cinxia and adult mass in H. horticola were unaffected by plant line. Cotesia melitaearum, a gregarious endoparasitoid of M. cinxia, performed equally well on hosts feeding on the two lines of P. lanceolata. In contrast, the pupal mass of the generalist herbivore, Spodoptera exigua, and the emerging adult mass of its solitary endoparasitoid, C. marginiventris, were significantly lower when reared on the high line, whereas development time was unaffected. The results are discussed with regards to (1) differences between specialist and generalist herbivores and their natural enemies to quantitative variation in plant secondary chemistry, and (2) potentially differing selection pressures on plant defense.     

Intraspecific Variation in Plant Defense Alters Effects of Root Herbivores on Leaf Chemistry and Aboveground Herbivore Damage

Root herbivores can indirectly affect aboveground herbivores by altering the food quality of the plant. However, it is largely unknown whether plant genotypes differ in their response to root herbivores, leading to variable defensive phenotypes. In this study, we investigated whether root-feeding insect larvae (Agriotes sp. larvae, wireworms) induce different responses in Plantago lanceolata plants from lines selected for low and high levels of iridoid glycosides (IG). In the absence of wireworms, plants of the “high-IG line” contained approximately twofold higher levels of total IG and threefold higher levels of catalpol (one of the IG) in leaves than plants from the “low-IG line,” whereas both lines had similar levels of IG in roots. In response to wireworms, roots of plants from both lines showed increased concentrations of catalpol. Leaves of “low-IG line” plants increased catalpol concentrations in response to wireworms, whereas catalpol concentrations of leaves of “high-IG line” plants decreased. In contrast, glucose concentrations in roots of “low-IG” plants decreased, while they increased in “high-IG” plants after feeding by wireworms. The leaf volatile profile differed between the lines, but was not affected by root herbivores. In the field, leaf damage by herbivores was higher in wireworm-induced compared to noninduced “low-IG” plants and lower in wireworm-induced compared to noninduced “high-IG” plants, despite induction of catalpol in leaves of the “low-IG” plants and reduction in “high-IG” plants. This pattern might arise if damage is caused mainly by specialist herbivores for which catalpol may act as feeding stimulant rather than as deterrent. The present study documents for the first time that intraspecific variation in plant defense affects the outcome of plant-mediated interactions between root and shoot herbivores.     

The Effects of Arbuscular Mycorrhizal Fungi on Direct and Indirect Defense Metabolites of Plantago lanceolata L.

Arbuscular mycorrhizal fungi can strongly influence the metabolism of their host plant, but their effect on plant defense mechanisms has not yet been thoroughly investigated. We studied how the principal direct defenses (iridoid glycosides) and indirect defenses (volatile organic compounds) of Plantago lanceolata L. are affected by insect herbivory and mechanical wounding. Volatile compounds were collected and quantified from mycorrhizal and non-mycorrhizal P. lanceolata plants that underwent three different treatments: 1) insect herbivory, 2) mechanical wounding, or 3) no damage. The iridoids aucubin and catalpol were extracted and quantified from the same plants. Emission of terpenoid volatiles was significantly higher after insect herbivory than after the other treatments. However, herbivore-damaged mycorrhizal plants emitted lower amounts of sesquiterpenes, but not monoterpenes, than herbivore-damaged non-mycorrhizal plants. In contrast, mycorrhizal infection increased the emission of the green leaf volatile (Z)-3-hexenyl acetate in untreated control plants, making it comparable to emission from mechanically wounded or herbivore-damaged plants whether or not they had mycorrhizal associates. Neither mycorrhization nor treatment had any influence on the levels of iridoid glycosides. Thus, mycorrhizal infection did not have any effect on the levels of direct defense compounds measured in P. lanceolata. However, the large decline in herbivore-induced sesquiterpene emission may have important implications for the indirect defense potential of this species.     

Effects of Sequestered Iridoid Glycosides on Prey Choice of the Prairie Wolf Spider, Lycosa carolinensis

Specialist insect herbivores that sequester allelochemicals from their host plants may be unpalatable to potential predators. However, the host-plant species used may determine the degree of palatability. Spiders, including members of the family Lycosidae, are important predators of invertebrate prey. We fed buckeye caterpillars, Junonia coenia (Nymphalidae), reared on Plantago lanceolata (containing high levels of iridoid glycosides) or P. major (containing low levels of iridoid glycosides) to prairie wolf spiders, Lycosa carolinensis (Lycosidae), to determine whether the spiders found insects that sequester iridoid glycosides unpalatable. In a field experiment, spiders ate caterpillars reared on P. major significantly more often than caterpillars reared on P. lanceolata, although they attacked equal numbers of both types of prey. Spiders that bit caterpillars behind their heads or along the middle of their backs prevented caterpillars from implementing deterrent defensive strategies such as regurgitating or defecating. In a laboratory experiment, we presented spiders with P. lanceolata-reared and P. major-reared caterpillars simultaneously for eight consecutive trials. Spiders consumed P. major-reared buckeyes significantly more often than P. lanceolata-reared caterpillars. We found no evidence that the spiders learned to avoid the unpalatable prey.     

Patterns of Iridoid Glycoside Production and Induction in Plantago lanceolata and the Importance of Plant Age

Induction of allelochemicals is one way that plants efficiently deploy defenses against herbivory. In two separate experiments we investigated the time course of this inductive response and the importance of the timing of herbivory for Plantago lanceolata (Plantaginaceae). We found a localized induced response of catalpol and the ratio of catalpol to total iridoid glycosides in damaged leaves that was evident at d 6 after caterpillars of the specialist Junonia coenia were put onto the plants. On the whole plant level, we detected small, but significant changes in the iridoid glycoside metabolism of P. lanceolata on several different days following herbivory. We also found considerable change in the amounts of allelochemicals produced during P. lanceolata's ontogeny. This ontogenetic effect might help to explain some of the reasons why induction may be difficult to detect in P. lanceolata. We also investigated the importance of the timing of herbivory on P. lanceolata's inductive response, but neither herbivory after 5 wk of growth nor after 6 or 7 wk of growth induced an increase in aucubin or catalpol.     

Fate of Host-Plant Iridoid Glycosides in Lepidopteran Larvae of Nymphalidae and Arcthdae

We compared the ability of larvae of six lepidopteran species to sequester iridoid glycosides. All larvae were fed on a common host plant, Plantago lanceolata, which contains two iridoid glycosides, aucubin and catalpol. Four species of arctiids were examined: Pyrrharctia isabella, Spilosoma congrua, Spilosoma latipennis, and Spilosoma virginica. For comparison, we also examined two nymphalid species, one of which, Junonia coenia (a specialist on plants containing iridoid glycosides), was known to sequester iridoid glycosides, and the other, Vanessa cardui, was a generalist that also feeds on P. lanceolata. We found that, as expected, J. coenia larvae did contain iridoid glycosides, whereas V. cardui did not. To our surprise, Spilosoma congrua contained substantial amounts of iridoid glycosides (mean = 7.14% dry weight), whereas none of the other arctiids (S. lalipennis, S. virginica, or P. Isabella) contained detectable levels of iridoid glycosides. We found small amounts of iridoid glycosides in the frass of S. virginica but none in the frass of S. congrua or P. isabella.     

Lonicera Implexa Leaves Bearing Naturally Laid Eggs of the Specialist Herbivore Euphydryas Aurinia have Dramatically Greater Concentrations of Iridoid Glycosides than other Leaves

We tested in the field the hypothesis that the specialist butterfly Euphydryas aurinia (Lepidoptera: Nymphalidae, Melitaeinae) lays eggs on leaves of Lonicera implexa (Caprifoliaceae) plants with greater iridoid concentrations. We conducted our investigations in a Mediterranean site by analyzing leaves with and without naturally laid egg clusters. There were no significant differences in iridoid glycoside concentrations between leaves from plants that did not receive eggs and the unused leaves from plants receiving eggs, a fact that would seem to indicate that E. aurinia butterflies do not choose plants for oviposition by their iridoid content. However, the leaves of L. implexa that bore egg clusters had dramatically greater (over 15-fold) concentrations of iridoid glycosides than the directly opposite leaves on the same plant. These huge foliar concentrations of iridoids (15% leaf dry weight) may provide specialist herbivores with compounds that they either sequester for their own defense or use as a means of avoiding competition for food from generalist herbivores. Nevertheless, it may still be possible that these high concentrations are detrimental to the herbivore, even if the herbivore is a specialist feeder on the plant.     

Density-Dependent Reduction and Induction of Milkweed Cardenolides by a Sucking Insect Herbivore

The effect of aphid population size on host-plant chemical defense expression and the effect of plant defense on aphid population dynamics were investigated in a milkweed-specialist herbivore system. Density effects of the aposematic oleander aphid, Aphis nerii, on cardenolide expression were measured in two milkweed species, Asclepias curassavica and A. incarnata. These plants vary in constitutive chemical investment with high mean cardenolide concentration in A. curassavica and low to zero in A. incarnata. The second objective was to determine whether cardenolide expression in these two host plants impacts mean A. nerii colony biomass (mg) and density. Cardenolide concentration (microgram/g) of A. curassavica in both aphid-treated leaves and opposite, herbivore-free leaves decreased initially in comparison with aphid-free controls, and then increased significantly with A. nerii density. Thus, A. curassavica responds to aphid herbivory initially with density-dependent phytochemical reduction, followed by induction of cardenolides to concentrations above aphid-free controls. In addition, mean cardenolide concentration of aphid-treated leaves was significantly higher than that of opposite, herbivore-free leaves. Therefore, A. curassavica induction is strongest in herbivore-damage tissue. Conversely, A. incarnata exhibited no such chemical response to aphid herbivory. Furthermore, neither host plant responded chemically to herbivore feeding duration time (days) or to the interaction between herbivore initial density and feeding duration time. There were also no significant differences in mean colony biomass or population density of A. nerii reared on high cardenolide (A. curassavica) and low cardenolide (A. incarnata) hosts.     

Earthworms and Litter Distribution Affect Plant-Defensive Chemistry

Studies on plant-defensive chemistry have mainly focused on plants in direct interaction with aboveground and occasionally belowground herbivores and pathogens. Here we investigate whether decomposers and the spatial distribution of organic residues in soil affect plant-defensive chemistry. Litter concentrated in a patch (vs. homogeneously mixed into the soil) led to an increase in the aucubin content in shoots of Plantago lanceolata. Earthworms increased total phytosterol content of shoots, but only when the litter was mixed homogeneously into the soil. The phytosterol content increased and aphid reproduction decreased with increasing N concentration of the shoots. This study documents for the first time that earthworms and the spatial distribution of litter may change plant-defensive chemistry against herbivores.     

Effects of Elevated CO<Subscript>2</Subscript> and Herbivore Damage on Litter Quality in a Scrub Oak Ecosystem

Atmospheric CO₂ concentrations have increased dramatically over the last century and continuing increases are expected to have significant, though currently unpredictable, effects on ecosystems. One important process that may be affected by elevated CO₂ is leaf litter decomposition. We investigated the interactions among atmospheric CO₂, herbivory, and litter quality within a scrub oak community at the Kennedy Space Center, Florida. Leaf litter chemistry in 16 plots of open-top chambers was followed for 3 years; eight were exposed to ambient levels of CO₂, and eight were exposed to elevated levels of CO₂ (ambient + 350 ppmV). We focused on three dominant oak species, Quercus geminata, Quercus myrtifolia, and Quercus chapmanii. Condensed tannin concentrations in oak leaf litter were higher under elevated CO₂. Litter chemistry differed among all plant species except for condensed tannins. Phenolic concentrations were lower, whereas lignin concentrations and lignin/nitrogen ratios were higher in herbivore-damaged litter independent of CO₂ concentration. However, changes in litter chemistry from year to year were far larger than effects of CO₂ or insect damage, suggesting that these may have only minor effects on litter decomposition.     

Iridoid glycoside sequestration byThessalia leanira (Lepidoptera: Nymphalidae) feeding onCastilleja integra (Scrophulariaceae)

A small population of a polyvoltine checkerspot butterfly,Thessalia leanira fulvia (also known asChlosyne leanira ssp.fulvia), was found to useCastilleja integra as a larval food plant at a localized site (Burnt Mill) southwest of Pueblo, Colorado. Field-captured adult butterflies contained the major iridoid glycosides (catalpol and macfadienoside) of theCastilleja. The content of a third iridoid glycoside, methyl shanzhiside, was also relatively high in the collected butterflies even though most individualCastilleja plants at Burnt Mill contained little or no methyl shanzhiside. Only a few plants, restricted to a small area, did contain appreciable methyl shanzhiside. Most of the plants that lacked the ester methyl shanzhiside contained shanzhiside, the corresponding free carboxylic acid.Thessalia larvae did not normally methylate the acid to produce methyl shanzhiside. Larvae that stopped feeding at an early instar, but yet survived several weeks, did contain major amounts of methyl shanzhiside. It is suggested that only larvae that overwinter or otherwise enter diapause convert shanzhiside to methyl shanzhiside. TheCastilleja food plant also contained iridoids other than catalpol and macfadienoside, sometimes in major amounts, but these were never found in larvae, pupae, or butterflies.Paper 27 in the series Chemistry of the Scrophulariaceae. Paper 26: Stermitz, F.R., Foderaro, T.A., and Li, Y.-X., 1993.Phytochemistry 32:1151     

Iridoid glycoside content ofEuphydryas anicia (Lepidoptera: Nymphalidae) and its major hostplant,Besseya plantaginea (Scrophulariaceae), at a high plains colorado site

The checkerspot butterfly,Euphydryas anicia, utilizes mainlyBesseya plantaginea and only occasionallyCastilleja integra as a larval hostplant at Michigan Hill, a few kilometers from a site whereC. integra is used by over 90% of the butterflies. TheB. plantaginea leaves that are consumed contain 9–22% iridoid glycosides, composed mainly of catalpol and catalpol esters, while larvae from the same plants contain 6–18% iridoids, mainly catalpol and no esters. Field-collected adult butterflies contain 0.5–4.3% iridoids. Laboratory-reared adults secrete iridoids in the meconium upon eclosion and retain similar amounts. The adult and meconium iridoid content is considerably lower than in the larvae, and metabolism in the pupal stage may be occurring.This work was supported by grant CHE-8521382 from the National Science Foundation. Paper 13 in the series Chemistry of the Scrophulariaceae. Paper 12: G.N. Belofsky and F.R. Stermitz,J. Nat. Prod. 51:614–616, 1988.     

Phytotoxic Iridoid Glucosides from the Roots of Verbascum thapsus

The iridoid glucosides lateroside 1, harpagoside 2, ajugol 3, and aucubin 4 were isolated from an ethanolic extract of the roots of the weed Verbascum thapsus that exhibits antigermination activity on seeds of barley (Hordeum vulgare). Bioassays indicated that at 3 mM concentration, compounds 1, 2, and 4 showed moderate inhibition of seed germination. These compounds also reduced root length when they were assayed on pregerminated seeds at 1 mM to 0.001 mM concentration range. Of all compounds tested, aucubin 4 was the most active against root elongation. Compound 3 showed no activity in the bioassays.     

Iridoid glycoside metabolism and sequestration byPoladryas minuta (Lepidoptera: Nymphalidae) feeding onPenstemon virgatus (Scrophulariaceae)

A bivoltine checkerspot butterfly,Poladryas minuta, is aPenstemon specialist, not known to utilize any other plant genus for oviposition and larval feeding. At several intermontane plains sites of central Colorado, the butterfly utilizesPenstemon virgatus as its sole host plant. Analysis of the host plant showed it to contain three cinnamyl-type catalpol esters (scutellarioside-II, globularin, globularicisin) and catalpol. The host plant contained an average of 10% dry weight iridoids, but some variation among individual plants and leaves within plants was noted. Field-collected butterflies contained 2.1–8.7% dry weight catalpol, but no other iridoids. Adults from larvae fedP. virgatus in the lab contained 4.2–9.0% dry weight catalpol and excreted large amounts of catalpol in the meconium. No catalpol was found in the larval frass. Larvae did not consume three alternate iridoid-containing host-plant species, and most eventually died rather than feed on the alternate plants. Larvae did consume small amounts of artificial diets containing the alternate species andP. virgatus, but most went into diapause and some died. Survival was good on artificial diet containing 10% dry weight of the iridoid esters fromP. virgatus. Only catalpol was found in pupae and adults, but it was absent from the larval frass. The cinnamic-type acids expected from larval hydrolysis of the esters were not found in larval frass, pupae, or adults. These results are contrasted with those found for another checkerspot,Euphydryas anicia, which consumes a different host-plant species but was present at one of the same sites withPoladryas minuta.Paper 15 in the series Chemistry of the Scrophulariaceae. Paper 14 Boros, C.M., Stermitz, F.R., and Harris, G.H. 1990.J. Nat. Prod. 5372–80.     

Systemic,genotype-specific induction of two herbivore-deterrent iridoid glycosides in Plantago lanceolata L. in response to fungal infection by Diaporthe adunca (Rob.) Niessel

Iridoid glycosides are a group of terpenoid secondary plant compounds known to deter generalist insect herbivores. In ribwort plantain (Plantago lanceolata), the iridoid glycosides aucubin and catalpol can be induced following damage by insect herbivores. In this study, we investigated whether the same compounds can be induced following infection by the fungal pathogen Diaporthe adunca, the causal agent of a stalk disease in P. lanceolata. Significant induction of aucubin and catalpol was observed in two of the three plant genotypes used in this study following inoculation with the pathogen. In one of the genotypes, induction occurred within 6 hr after inoculation, and no decay was observed within 8 days. The highest level of induction was observed in reproductive tissues (spikes and stalks) where infection took place. In these tissues, iridoid glycoside levels in infected plants were, on average, 97% and 37% higher than the constitutive levels in the corresponding control plants, respectively. Significant induction was also observed in leaves (24%) and roots (17%). In addition to significant genotypic variation in the level of induction, we found genetic variation for the tissue-specific pattern of induction, further broadening the scope for evolutionary fine-tuning of induced responses. Recent studies have revealed a negative association between iridoid glycoside levels in P. lanceolata genotypes and the amount of growth and reproduction of D. adunca that these genotypes support. However, for the three genotypes used in the present study, differences in resistance were not related to their constitutive or induced levels of iridoid glycosides, suggesting that additional resistance mechanisms are important in this host-pathogen system. We conclude that iridoid glycosides in P. lanceolata can be induced both by arthropods and pathogenic micro-organisms. Pathogen infection could, therefore, potentially enhance resistance to generalist insect herbivores in this species.     

Iridoid glycosides in the nectar ofCatalpa speciosa are unpalatable to nectar thieves

The floral nectar ofCatalpa speciosa has a chemical mechanism that limits thievery. A bioassay employing sphingid larvae,Ceratomia catalpae, shows that catalpa iridoid glycosides are present in the floral nectar. When potential nectar thieves are fed nectar, a sucrose solution of identical concentration, or a sucrose solution plus 0.4% catalpol and 0.4% catalposide (catalpa iridoids), the thieves drink significantly more of the pure sucrose solution than either of the other two sugar sources. Those thieves that drink either the nectar or the sucrose solution plus catalpa iridoids develop behavioral abnormalities including regurgitation and loss of locomotion. The response of the potential nectar thieves to nectar or the sucrose solution plus catalpa iridoids cannot be distinguished by the amount consumed or by their behavior. The legitimate diurnal pollinators ofC. speciosa are not affected by the iridoid glycosides.     

Quantitative Effects of Cyanogenesis on an Adapted Herbivore

Plant cyanogenesis means the release of gaseous hydrogen cyanide (HCN) in response to cell damage and is considered as an effective defense against generalist herbivores. In contrast, specialists are generally believed not to be affected negatively by this trait. However, quantitative data on long-term effects of cyanogenesis on specialists are rare. In this study, we used lima bean accessions (Fabaceae: Phaseolus lunatus L.) with high quantitative variability of cyanogenic features comprising cyanogenic potential (HCNp; concentration of cyanogenic precursors) and cyanogenic capacities (HCNc; release of gaseous HCN per unit time). In feeding trials, we analyzed performance of herbivorous Mexican bean beetle (Coleoptera: Coccinellidae: Epilachna varivestis Mulsant) on selected lines characterized by high (HC-plants) and low HCNp (LC-plants). Larval and adult stages of this herbivore feed on a narrow range of legumes and prefer cyanogenic lima bean as host plant. Nevertheless, we found that performance of beetles (larval weight gain per time and body mass of adult beetles) was significantly affected by lima bean HCNp: Body weight decreased and developmental period of larvae and pupae increased on HC-plants during the first generation of beetles and then remained constant for four consecutive generations. In addition, we found continuously decreasing numbers of eggs and larval hatching as inter-generational effects on HC-plants. In contrast to HC-plants, constantly high performance was observed among four generations on LC-plants. Our results demonstrate that Mexican bean beetle, although preferentially feeding on lima bean, is quantitatively affected by the HCNp of its host plant. Effects can only be detected when considering more than one generation. Thus, cyanide-containing precursors can have negative effects even on herbivores adapted to feed on cyanogenic plants.     

Iridoid glycosides as oviposition stimulants for the buckeye butterfly,Junonia coenia (Nymphalidae)

Females ofJunonia coenia (Nymphalidae), a specialist on plants that contain iridoid glycosides, were found to use aucubin and catalpol, iridoid glycosides typical of a host plant,Plantago lanceolata (Plantaginaceae), as oviposition cues. Incorporating dried ground leaf material or pure iridoid glycosides into agar disks proved to be a very effective method of testing. In no-choice tests and choice tests, females laid more eggs on disks withP. lanceolata leaf material or iridoid glycosides, compared to agar controls. There was variation among individual females in preference for disks withP. lanceolata leaf material versus disks with iridoid glycosides. Females given a choice of three different concentrations of iridoid glycoside (0.2, 0.5, 1.0%) in the agar disks and a control laid more eggs on the disk with the highest concentration of iridoid glycoside.     

Iridoid glycosides and host-plant specificity in larvae of the buckeye butterfly,Junonia coenia (Nymphalidae)

Larvae of the buckeye,Junonia coenia (Nymphalidae) feed primarily on plants in four families: Scrophulariaceae, Plantaginaceae, Verbenaceae, and Acanthaceae. These plant families have in common the presence of a group of plant secondary compounds, the iridoid glycosides. Larvae were reared on three plant species and two artificial diets, one with and one without iridoid glycosides.Larvae grew poorly and had low survivorship on the artificial diet without iridoid glycosides, while growth and survival on the artificial diet with iridoid glycosides was comparable to that on plants. Choice tests using artificial diets with and without iridoid glycosides showed that larvae: (1) chose diets with iridoid glycosides (in the form of a crude extract or pure compound) over a diet without; (2) showed no preference between the diet with the crude extract and that with pure iridoid glycoside, and (3) preferred the artificial diet with ground leaves of the host plant,Plantago lanceolata, over the diet with pure iridoid glycosides. The artificial diet that larvae had been reared on prior to these tests had no effect on subsequent larval preferences in the choice tests.     

Iridoid glycoside sequestration by two aposematicPenstemon-feeding geometrid larvae

The iridoid glycoside catalpol was found to be sequestered by larvae ofMeris alticola feeding onPenstemon virgatus and by larvae ofNeoterpes graefiaria which utilizeP. barbatus. The strikingly similar larval patterns of these two ennomine geometrids were previously considered to be disruptive, but predator-based Mullerian mimicry is equally likely to be involved. The cryptic adult moths generally contain only small amounts of catalpol, having left most of the bitter iridoid in the pupal case and in the meconium excreted after eclosion. OneNeoterpes female did contain considerable catalpol in the abdomen, presumably in the eggs.This work was supported by grant BSR-8506064 from the National Science Foundation. Paper 11 in the series Chemistry of the Scrophulariaceae. Paper 10: F.R. Stermitz and G.H. Harris, 1987.J. Chem. Ecol. 13:1917–1925.