Skip to main content
PMC home page
Springer logoLink to Springer
. 2016 Mar 31;42:240–248. doi: 10.1007/s10886-016-0684-0

Two Defensive Lines in Juvenile Leaf Beetles; Esters of 3-nitropropionic Acid in the Hemolymph and Aposematic Warning

Gerhard Pauls 1, Tobias Becker 1, Peter Rahfeld 1, Rene R Gretscher 1, Christian Paetz 2, Jacques Pasteels 3, Stephan H von Reuss 1, Antje Burse 1, Wilhelm Boland 1,✉
PMCID: PMC4839037  PMID: 27033853

Abstract

Juveniles of the leaf beetles in subtribe Chrysomelina have efficient defense strategies against predators. When disturbed, they transiently expose volatile deterrents in large droplets from nine pairs of defensive glands on their back. Here, we report on an additional line of defense consisting of the non-volatile isoxazolin-5-one glucoside and its 3-nitropropanoyl ester in the larval hemolymph. Because isoxazolin-5-one derivatives were not detectable in related leaf beetle taxa, they serve as a diagnostic marker for the Chrysomelina subtribe. Conjugation of isotopically labelled 3-nitropropionic acid to isoxazolin-5-one glucoside in vivo demonstrates its function as a carrier for the 3-nitropropanoyl esters. The previous identification of characteristic glucosides as precursors of the volatile deterrents underlines the general importance of glucosides for sequestration from food plants, and the subsequent transport in the hemolymph to the defense system. The combination of repellent volatiles with non-volatile toxic compounds in the hemolymph has the potential to create synergistic effects since the odorant stimulus may help predators learn to avoid some foods. The combination of the two defense lines has the advantage, that the hemolymph toxins provide reliable and durable protection, while the repellents may vary after a host plant change.

Electronic supplementary material

The online version of this article (doi:10.1007/s10886-016-0684-0) contains supplementary material, which is available to authorized users.

Keywords: Chrysomelidae, Chemical defense, 3-nitropropionate esters, Isoxazolin-5-one glucoside, Aposematic warning

Introduction

Toxins are the most effective players on our planet when it comes to the manifold interactions in trophic networks. Reflecting successful relationships with their hosts, leaf beetles (family Chrysomelidae) of the taxon Chrysomelina have ingenious strategies that disarm plant toxins and, at the same time, produce a chemical defense against natural enemies. This chemical defense not only protects all developmental stages from larvae to adults, but also changes its composition during the life history of Chrysomelina beetles.

Adult beetles store and release defensive secretions from pronotal and elytral exocrine glands upon disturbance (Deroe and Pasteels 1982; Pasteels et al.1989). The major components are isoxazolin-5-one glucosides esterified with up to three 3-nitropropionic acid (3-NPA) moieties (Pasteels et al.2003; Sugeno and Matsuda 2002). As these compounds are originally not present in any of the beetle host plants, they are postulated to be synthesised by the insects themselves via pathways that remain to be explored (Randoux et al.1991).

While the mode of action of the isoxazolinone moiety is unknown, 3-NPA is a naturally occurring neurotoxin that, when ingested, causes poisoning in both humans and domestic livestock by irreversibly inhibiting the mitochondrial succinate dehydrogenase (E.C. 1.3.5.1), a key enzyme of the citric acid cycle (Anderson et al.2005; Beal et al.1993; Huang et al.2006). 3-Nitropropionic acid and its glucose esters have been identified in many members of the legume plant family (Fabaceae) and in certain fungi as a defense against herbivores (Chomcheon et al.2005; Francis et al.2013; Parry et al.2011). Some insect herbivores can detoxify 3-NPA (Johnson et al.2001; Majak et al.1998; Novoselov et al.2015). To date, Chrysomelina leaf beetles are the only insects in which 3-NPA and its derivatives have been described as allomones. The amounts of 3-NPA that predators of leaf beetles, for example ants or birds can tolerate before avoiding further consumption, remain to be determined.

Unlike adults, juveniles rely on volatile repellents whose molecular structure is entirely different. By displaying droplets of defensive secretions from nine pairs of everting glandular reservoirs located on their backs, the larvae have an extraordinary defense mechanism unparalleled in the insect world. The defensive chemicals in the Chrysomelina larval exudates are composed of four compound classes (Fig. 1): iridoids (cyclopentanoid monoterpenoids, e.g., chrysomelidial 2), aldehydes (salicylaldehyde 4 and benzaldehyde), esters (e.g., phenethyl esters), and the naphtoquinone juglone (Hilker and Schulz 1994; Laurent et al.2005; Matsuda and Sugawara 1980; Pasteels et al.1986). Phylogenetic analysis of Chrysomelina species revealed that the composition of their secretions reflects a step-wise scenario of host-plant adaptation (Termonia et al.2001).

Fig. 1.

Fig. 1

Open in a new tab

Compounds implicated in the volatile chemical defense of the Chrysomelina larvae. 8-hydroxygeraniol glucoside (1) and salicin (3) are precursors of chryomelidial (2) and salicylaldehyde (4) in Phaedon cochleariae and Chrysomela populi, respectively

The plant-independent biosynthesis of iridoids predated the sequestration of salicin, a plant-derived precursor from Salicaceae, used to produce the repellent salicylaldehyde (Kuhn et al.2004; Pasteels et al.1983b). Later in Chrysomelina beetle evolution, a sequestering Chrysomelina lineage -- namely, the interrupta group -- escaped the constraints of their host plant (willow) by shifting to birch. Due to the different secondary metabolites present in the two hosts, this shift resulted in modified larval exudates produced from the sequestered precursors. For example, the willow-feeding population of the species, Chrysomela lapponica, produces predominantly salicylaldehyde from sequestered salicin, whereas the birch-feeding population is able to take up a wide variety of glucosidically bound leaf alcohols. These leaf alcohols are further esterified with butyric acid, resulting in a cocktail of at least 60 esters in the defensive exudate (Geiselhardt et al.2015; Termonia et al.2001). Regardless of the different composition and origin of the defensive metabolites in the secretions of Chrysomelina larvae (de novo vs. sequestration), the synthesis of all Chrysomelina allomones includes glucoside intermediates (Discher et al.2009).

Owing to their defensive volatiles, the larvae are protected against microbial infestation (Gross et al.1998, 2002, 2008; Gross and Schmidtberg 2009; Gross et al.2008), generalist arthropod predators (Blum et al.1978; Hilker and Schulz 1994; Palokangas and Neuvonen 1992; Pasteels et al.1983a, b), and insectivorous birds (Topp 1997). These nonspecific volatile irritants, however, act as repellents rather than as toxins that target specific physiological processes (Pasteels et al.1983a), and their ecological significance has to date remained poorly understood. As toxicity is often associated with warning signals such as colors, sound, taste, or odors (Pasteels et al.1983a), we hypothesized that the volatile irritants also may be linked with toxins not yet identified in Chrysomelina larvae. Since the hemolymph provides the storage site for toxins in a wide range of insect species (Laurent et al.2005; Opitz and Muller 2009), we analyzed the inventory of secondary metabolites in the hemolymph of juvenile chrysomelids by LC-MS and NMR.

Here, we report on the identification of isoxazolin-5-one glucoside and its 6-nitropropanoate in the larval hemolymph of all tested Chrysomelina species. Previously, these compounds had been attributed exclusively to the adults. However, this finding leads to the conclusion that Chrysomelina species are protected by isoxazolinone glucosides and their 3-NPA esters throughout the beetle life history. Hence, in addition to the defensive odor released from their dorsal glands, the larvae possess toxins in their hemolymph. This association may contribute synergistically to protection against an array of vertebrate and invertebrate enemies. Further, we detected glucoside precursors for the volatile secretions in the larval hemolymph, a discovery that underlines the importance of sugar derivatives as carriers for controlled translocation processes and for preventing the insects from self-poisoning.

Methods and Materials

Insect Rearing

Chrysomela populi (L.) was collected near Dornburg, Germany on Populus maximowiczii × Populus nigra. Beetles were propagated using a cycle of 16 h L and 8 h D at 18 ± 2 °C in light and 13 ± 2 °C in darkness. Phaedon cochleariae (F.) was collected from Brassicaceae close to the city of Bayreuth (Germany) and kept as a continuous culture in the laboratory (Discher et al.2009). Larvae were reared on Brassica rapa subsp. pekinensis “Cantonner Witkrop” (Quedlinburger Saatgut, Quedlinburg, Germany) in a Snijder chamber (Snijders Scientific, Tilburg, Netherlands) in a cycle of 16 h L / 8 h D and 13 °C/11 °C ± 1 °C. The low temperature (13 °C) was necessary to reduce fungal growth on the food plant. Willow feeding C. lapponica (L.) were collected in the Altai Mountains in East Kazakhstan, near Katon-Karagai in the Katon-Karagaisky State National Nature Park (2100 m altitude). Birch-feeding C. lapponica was collected from Betula rotundifolia in the Altai Mountains in East Kazakhstan, close to Uryl, near the Burkhat Pass in the Katon-Karagaisky State National Nature Park (2130 m altitude). All other species were collected in the field, see Table S1 for details.

Preparation of Samples from the Hemolymph, Frass, and Whole Larvae Extracts

Hemolymph samples were collected as described previously (Bodemann et al.2012) in capillaries that were sealed immediately after collection and stored at −20 °C until use. Hemolymph weight was determined by measuring the weight of a filled capillary minus its dry empty weight (Mettler-Toledo XS 205, Greifensee, Switzerland). For LC-MS measurements, the hemolymph was diluted with 50 % aqueous MeOH in a ratio of 1 μl hemolymph per 100 μl solvent. Frass samples of P. cochleariae (2.5 mg), C. populi (13 mg), and C. lapponica (5 mg) were extracted with water and analyzed by LC-MS.

For crude extracts prepared from complete larvae, each larva was weighed individually using an ultra-microbalance (XS205; d = 0.01 mg; Mettler-Toledo, Greifensee, Switzerland). Individual larvae were frozen separately in liquid N2 and macerated in 500 μl MeCN using a Geno grinder. After centrifugation (10,621 rpm, 10 min, room temperature), the supernatant was subject to LC-MS analysis.

Analysis of Glucosides by LC–MS

Analyses were carried out using an Agilent HP1100 HPLC system equipped with an RP-C18 column, LiChroCART (250 × 4 mm, 5 μm;Merck KGaA, 64271, Darmstadt, Germany) connected to a Finnigan LTQ (Thermo Electron Corp, Dreieich, Germany) operated in the APCI mode (vaporizer temperature: 500 °C, capillary temperature 300 °C). Standard compounds for identification were either purchased (Sigma-Aldrich (St. Louis, MO, USA) or synthesised. Isoxazolin-5-one glucoside and its esters were synthesised according to previously described protocols (Becker et al.2013, 2015).

Samples were analyzed by injection (5 μl) and by the application of a gradient elution. The following protocol was used: 100 % solvent A (H2O + 0.1 % HCOOH) and 0 % solvent B (MeCN + 0.1 % HCOOH), linear gradient to 60 % solvent B in 35 min. Extract samples of whole larvae were analyzed by injecting a 5 μl sample and using an isocratic elution with 35 % solvent B (v/v) in H2O +0.1 % HCOOH. For identification and quantification, the formic acid adducts [M+HCOOH-H]− were used (m/z 292 for 2-(β-D-glucopyranosyl)-3-isoxazolin-5-one (5), m/z 393 for 2-[6′-(3″-nitropropanoyl)-β-D-glucopyranosyl]-3-isoxazolin-5-one (6), m/z 331for salicin (3), and m/z 377 for 8-hydroxygeraniol-8-O-β-D-glucoside (1).

Analysis of Crude Larval Hemolymph by NMR

Hemolymph from 20 larvae of P. cochleariae or C. populi was collected and taken up in 200 μl CD3OD for 1H/2D-exchange. The solution was concentrated under reduced pressure and dissolved in 500 μl CD3OD. One-dimensional 1H NMR spectra were recorded on a Bruker AV400 using water suppression (purge). Two-dimensional double quantum-filtered (dqf)-COSY spectra with phase cycling were recorded on a Bruker AV400. A total of 32 scans were acquired using a time domain of 8 k in F2 (acquisition time of 1.2 s) and 512 increment in F1. Spectra were zero-filled to 8 k × 4 k prior to Fourier transformation and phasing using the Topspin software (Bruker). Heteronuclear HSQC and HMBC spectra were recorded using Bruker AMX500 with a cryoprobe. Samples were dissolved in 100 μl CD3OD using 2 mm NMR vials. For HSQC spectra, 40 scans were acquired using a time domain of 1 k in F2 and 256 increments in F1. For HMBC spectra, 256 scans were acquired using a time domain of 4 k in F2 and 128 increments in F1. Spectra were zero-filled to 4 k × 2 k prior to Fourier transformation and phasing using the Topspin software (Bruker).

Statistical Analysis

Linear regressions were used to investigate whether the amount of 5 and 6 changed with the weight of the larvae. In order to achieve homogeneous variances and normality of the residuals, data were square root transformed. Data were analyzed with SigmaPlot 11.0.

Synthesis of Labelled [1-13C, 3-15N]-3- Nitropropionic Acid and Injection of Labelled 3-NPA into the Larval Hemolymph

Stable isotope labelled [1-13C,3-15N]-3-nitropropinoic acid was synthesized according to Baxter et al. (1992) by using K13CN instead of Na13CN, and Na15NO2 instead of NaNO2.

The mass of third-instars (15 days after hatching) was measured on an ultra-microbalance (Mettler-Toledo, Greifensee, Switzerland). Ice-chilled larvae were injected dorso-medianally in the intersegmental membrane behind the pronotum using a pulled glass capillary as a needle connected to a nanoliter-injection pump (WPI, Sarasota, FL, USA), mounted on a three-axis-micromanipulator. For labelling experiments, each larva was injected with 200 ng 3-NPA per mg body weight in 122 nl injection-buffer (Bodemann et al.2012), representing a sublethal dose determined previously in a pilot experiment. To test larval tolerance to 3-NPA, 10 P. cochleariae larvae each were injected with the following concentrations: 100, 200, 300, 400 ng/mg body weight. While 300 ng/mg body weight was tolerated (10 of 10 survived), injection of 400 ng/mg body weight was fatal (10 of 10 died).

Results

Identification of Isoxazolinone Glucosides in the Larval Hemolymph of Leaf Beetles

To detect isoxazolinone glucosides in leaf beetle hemolymph, the following model species were analyzed: P. cochleariae, representing the iridoid de novo producers; C. populi and willow-feeding C. lapponica, representing the salicin-sequestering species. The larval hemolymph samples were analyzed by HPLC-MS. APCI ionization was chosen because it is less susceptible to matrix effects (Peters and Remane 2012). Compounds were identified by comparing HPLC-MS chromatograms of the natural samples with the spectra of synthetic standards (Fig. 2, see Fig. S1 and S2 for mass spectra).

Fig. 2.

Fig. 2

Open in a new tab

Liquid chromatograms of larval hemolymph of chrysomelina leaf beetles. From top to bottom: Chrysomela populi, Phaedon cochleariae. Traces for formic acid adducts [M+HCOOH]− for glucosides are shown: isoxazoline-5-one-glucoside (m/z 292, solid line), isoxazoline-5-one-glucoside 3-NPA ester (m/z 393, dotted line), salicin (m/z 331, dashed line), 8-hydroxygeraniol glucoside (m/z 377, dashed line)

To confirm the identity of the presumed hemolymph toxins and to obtain an unbiased assessment of the whole metabolome, we employed NMR spectroscopy. Previous reports demonstrated that 1H NMR spectroscopy is well suited to analyze complex metabolome mixtures, including insect hemolymph samples (Lenz et al.2001; Phalaraksh et al.1999; Poynton et al.2011). For our analysis of the crude unfractionated C. populi and P. chochleariae hemolymph, we used the two-dimensional double quantum-filtered correlation spectroscopy (dqf-COSY), which provides outstanding sensitivity and dynamic range along with a wealth of structural information. Analysis of the dqf-COSY spectra (Fig. S3) indicated the presence of several free amino acids such as alanine, valine, leucine, isoleucine, threonine, and proline (Table S2), along with characteristic signals corresponding to two dominant β-glucosidic components. Their heteroaromatic aglycone moieties were identified as isoxazolinones based on two characteristic AX-spin systems at δH 8.453 and 5.315 or at δH 8.433 and 5.344 that both displayed a coupling constant of 3J = 3.7 Hz. The linkage of the glucose and isoxazolinone moieties was established by complementary HSQC and HMBC correlations from the anomeric hydrogen to the β-carbon. The β-configuration of the glycosidic bond was deduced from the vicinal coupling constant 3J1’,2’ = 9.2 Hz for the anomeric hydrogen. Both isoxazolinone glucosides differed in the chemical shifts of the 6′-position, indicating 6′-acylation in one of the two components. This assumption was confirmed by HMBC correlations from the 6′-methylene protons to a carbonyl moiety at δC 171.5 ppm. Furthermore, this carbonyl group displayed additional HMBC correlations to an A2M2 spin system at δH 3.02 δC 31.7 ppm and δH 4.71 δC 70.7 ppm, indicating a 3-nitropropanoate substitution. Comparison of the 1H and 13C NMR data with data of the authentic standard obtained by synthesis as previously described (Becker et al.2013, 2015) confirmed our structure assignment. Since the α- and β-anomers of compound 5 can be separated easily by LC (RP-C18 column, separation factor R = 1.74 (Becker et al.2013, 2015) by using isocratic elution with acetonitrile and water (3:97, v:v), the larval defense compound consists of the pure β-anomer (>99 %).

Both components, isoxazolinone β-glucoside and its 6′-nitropropanoate (Fig. 2, Figs. S1, S2), have been previously described in eggs and adults of the subtribe Chrysomelina (Matsuda and Sugawara 1980; Pasteels et al.1986), but the presence of these compounds in the larval hemolymph previously was unknown.

Occurrence of Isoxazolin-5-one Derivatives in Various Chrysomelidae

To determine the distribution of isoxazolin-5-one derivatives in the leaf beetle family, the larval hemolymph of Chrysomelina beetles and related subtribes Chrysolinina and Galerucinae were analyzed. In all tested species of Chrysomelina, the isoxazolinone glucoside 5 could be quantified, whereas in the two other subtribes, the glucoside was not detectable (Fig. 3, Table S3). The corresponding nitropropanoyl ester 6 was detectable in the Chrysomelina hemolymph samples of the de novo iridoid-producing species P. cochleariae, in the salicin-sequestering species C. populi, Chrysomela saliceti, Chrysomela tremulae, and willow-feeding C. lapponica and in the ester-producing C. lapponica which feeds on birch (Fig. 3). A limiting factor is the lack of the stability of the 3-NPA esters in the hemolymph, while the isoxazolinone glucoside 5 exhibits as an N-glucoside with exceptional stability. Even in acidic media, the compound is stable (Becker et al.2015). Hence, notwithstanding the production strategy of volatile deterrents in the defensive glands of juvenile Chrysomelina, all tested Chrysolinina species posess a hemolymph-based chemical defense. Leaf beetle taxa other than Chrysomelina do not possess any of the isoxazolin-5-one derivatives.

Fig. 3.

Fig. 3

Open in a new tab

Phylogeny in relation to the presence of isoxazolinone glucoside and its 6-nitropropanoate ester in the hemolymph. The phylogeny is adapted from Termonia and Pasteels (1999), Gomez-Zurita et al. (2007), and Daccordi (1994) to represent different Chrysomelidae species in relation to the presence of isoxazolinone glucoside and 6-nitropropanoate ester in the larval haemolymph (marked with grey box). The branch points marked with A show the separation into the tribe Chrysomelini (subtribe Chrysomelina) and B into the tribe Chrysolinini (subtribe Doryphorina)

Isoxazolin-5-one Derivatives are Produced During Larval Development Endogenously

Considering the high concentrations of compounds 5 and 6 in the larval hemolymph and results reported by (Randoux et al.1991), who describe the de novo production of isoxazolin-5-one derivatives in adult chrysomelids, we asked whether the larvae also are able to synthesize these derivatives endogenously during the juvenile stage. Therefore, we analyzed these compounds in different larval stages in P. cochleariae as representatives. Whole larvae extracts were analyzed by LC-MS, and the amount of isoxazolinone glucoside 5 and its 3-NPA ester 6 in nmol/mg plotted vs. fresh body weight (ranging from 2 to 15 mg) as shown in Fig. S4. We found that the concentration of compound 5 (in nmol/mg body weight) remained constant (regression analysis, P = 0.448) in larvae of different body weight, whereas the concentration of the glucoside ester 6 increased with body weight (P < 0.001), indicating that an increasing amount of total isoxazolin-5-one derivatives must be produced during larval development.

In order to find out if the toxin level in the hemolymph can be regulated by excretion via the malpighian tubules or hindgut, we analyzed the frass of chrysomeline larvae. However, we could detect neither the glucoside 5 nor the ester 6 in the frass of P. cochleariae, C. populi, or willow-feeding C. lapponica within the detection limits. Both analytes also were not detectable in defensive secretions of the larvae. This suggests that a strongly limited excretion of isoxazolinone derivatives contributes to their accumulation in the hemolymph, which also requires an internal regulation of the ratio of glucoside 5 to ester 6.

Free 3-NPA is Conjugated to Isoxazolin-5-one Glucoside

5 Given the toxicity of free 3-NPA, we tested the metabolic ability of Chrysomelina larvae to accept free 3-NPA as a substrate for the biosynthesis of the ester compound 6. For this purpose, sublethal doses (200 ng/mg larva) of stable isotope labelled 3-NPA ([15N, 1-13C] 3-NPA) were injected into the larval hemolymph of P. cochleariae, C. populi, and willow-feeding C. lapponica. HPLC-MS analyses revealed the incorporation of 3-NPA into isoxazolin-5-one glucoside 5, forming the corresponding ester 6. Diesters or triesters, as reported in adult leaf beetle secretions (Matsuda and Sugawara 1980), were absent. Furthermore, the enrichment of the isotope signals at [M + 2] in the hemolymph compared to buffer-treated control groups was determined. The values ranged between 7 and 24.5 %, indicating a tolerance to free 3-NPA (Fig. S5). The isotope enrichment for compound 6 in C. populi was 13.2 % ± 4.3 % (arithmetic mean ± standard deviation, N = 4 for each species), in the case of P. cochleariae it was 7 % ± 1.3 %, while C. lapponica showed 24.5 % ± 9.5 % enrichment. In summary, 3-NPA is esterified to 6 with differences in efficiency depending on the examined species. Compound 5 apparently serves as a carrier to attach free 3-NPA to form the non-toxic ester 6.

Detection of Salicin and 8-Hydroxygeraniol-8-O-ß-D-glucoside in the Larval Hemolymph

In addition to examining the hemolymph production of isoxazolin-5-one derivatives, we screened for de novo-produced as well as sequestered precursor glucosides of the volatile deterrents in the hemolymph (Fig. 1). Under the chosen chromatographic conditions, our target compounds showed signals at m/z = 377 [M+HCOO]− for 8-hydroxygeraniol glucoside 3 in P. cochleariae as well as m/z = 331 [M+HCOO]− for salicin 1 in C. populi. The compounds were identified by comparison of HPLC-MS chromatograms of the natural samples with the spectra of commercially available standards. To confirm our identification of previously mentioned glucosides in the hemolymph of P. cochleariae and C. populi, we reanalyzed the dqf-COSY spectra. These confirmed the presence of small amounts of 8-hydroxygeraniol glucoside 3 as well as salicin 1 (Fig. S6). In summary, we identified 8-hydroxygeraniol glucoside 3 and salicin 1 in the hemolymph of P. cochleariae and C. populi, respectively. This confirms a function of the hemolymph as a transport matrix for the isoxazoline-glucoside 5, its 3-NPA ester 6, and the deterrent precursors, produced de novo or sequestered, en route to the tissue of destination (Discher et al.2009).

Discussion

The toxicity of insects often is linked to warning signals (Pasteels et al.1983a). The adults of many Chrysomelina species, for example, have aposematic red elytra advertising the toxicity of 3-NPA esters of isoxazolinone glucosides and their break down product 3-NPA. Compared to adults, the larvae possess a strikingly different defense mechanism. When disturbed, they display large droplets that contain secretions from eighteen glands; these droplets change the larvae’s appearance dramatically. As predators often are conservative when assessing the size of their prey, this behavior alone may prevent life-threatening attacks (Cohen et al.1993). In addition to the appearance of the larvae, their odor also changes since the droplets contain volatile chemicals in high amounts, such as iridoids or salicylaldehyde. Besides their repellent effect on predators, these irritants have nonspecific toxic effects. Iridoids, for example, can bind proteins covalently that might have adverse effects upon ingestion (Kim et al.2000), whereas salicylaldehyde exhibits a non-specific cytotoxic effects to insect cell cultures (Gross et al.2002).

Our discovery of isoxazolinone-based hemolymph toxins led us to revise the view of the defense of Chrysomelina leaf beetles (Fig. 4). The 3-NPA ester 6 itself is a deterrent, as demonstrated with ants (Pasteels et al.1986; Sugeno and Matsuda 2002). Furthermore, 3-NPA is a cytotoxin that interferes with mitochondrial respiration (Huang et al.2006). Although adults possess esterase activity in their secretions, and thus, are able to cleave the esters of isoxazolin-5-one glucosides 5 to liberate 3-NPA, it is conceivable that the larvae have to be ingested to release toxic components by predator digestion.

Fig. 4.

Fig. 4

Open in a new tab

Scheme of glucoside transport in leaf beetle larvae. Glucosides implicated in the volatile defense are ingested with the food. Transport proteins mediate the uptake of glucosides into the hemolymph (1). Precursor glucosides, either sequestered from food or synthesized autogenously, are selectively transported to the defensive glands for further processing (2). Isoxazolinone glucosides are produced in the fat body and released into the hemolymph (3). Free 3-NPA can be conjugated to isoxazolin-5-one glucoside to prevent autointoxication (4)

The two described mechanisms of chemical defense, volatile and non-volatile compounds, could have synergistic effects. The odorant signal, e.g., salicylaldehyde, could be a conditioning stimulus, linking the conspicuous odor to the hemolymph toxin. This system, known as olfactory aposematism (Weldon 2013), is effective mainly for vertebrate predators, such as birds; it is how they learn to avoid certain food. As tree-living Chrysomela species share their habitat with birds, the strongly odoriferous salicylaldehyde could be especially effective against this category of predators (Topp 1997). Based on the different allomones developed by Chrysomelina beetles, this taxon represents an unrivalled case study in chemical ecology, which illuminates the concerted action of diverse defense strategies during the adaptation of herbivorous insects to a given niche in an ecosystem.

Research into these hemolymph toxins extends our understanding of the chemical defense of chrysomeline leaf beetles considerably (Fig. 3). Sequestering leaf beetle larvae have adapted to use plant-derived precursors to produce their defensive secretions, which has economic advantages but at the same time restricts host-plant affiliation. One Chrysomelina lineage, however, must have escaped the host-linked constraints (precursor uptake) by shifting host-plant families. Consequences for the changing composition of the secretions, for example, were seen in the different populations of the species C. lapponica, many of which shifted from salicin-rich willow species to salicin-poor or even salicin-devoid birch species. With the isoxazolinone-based defense, it becomes clear that the larvae are not as dependent on their volatile defense as has been previously suggested. Instead, the hemolymph toxins provide protection, independent from changes in the composition of the repellent secretions after a host-plant shift of a sequestering species such as C. lapponica.

Having detected isoaxazolinone glucoside 5 and its 3-NPA esters 6 in C. lapponica, C. populi, and P. cochleariae, we screened additional species of the Chrysomelidae family (see Fig. 3) for isoxazolinone derivatives in the larval hemolymph to obtain an estimate of the occurrence of these defensive compounds. Interestingly, although isoxazolinone glucoside 5 has been found in all analyzed members of the subtribe Chrysomelina, it has been detected neither in the larval hemolymph nor in adult secretions of species of the neighbouring subtribe Chrysolinina or in Agelastica alni, a member of the subfamily Galerucinae.

Consequently, the defense based on isoxazolinone derivatives throughout all developmental stages represents a trait unique to Chrysomelina beetles, and as such may be regarded as a chemomarker for this subtribe. For example, G. depressa, lately classified as a member of the subtribe Chrysomelina (Pasteels et al.2003), also contains isoxazolinone glucoside 5, which supports its classification into this taxon.

Considering the high concentrations of isoxazolinone glucosides 5 and 6 in the larval hemolymph, we asked whether juvenile Chrysomelina beetles derive these compounds from the eggs as a parental gift (Pasteels et al.1986) or produce them de novo during larval development, as has been suggested for adult chrysomelids (Randoux et al.1991). While an increase in ester compounds 6 during larval development was measured, the concentrations of the isoxazolinone glucoside 5 remained constant (Fig. S4). This suggests that these compounds are produced autogenously during larval development. As none of the respective host plants produce isoxazolinone derivatives, the most plausible scenario for their existence is autogenous synthesis. Furthermore, the increased overall concentration of isoxazolinone derivatives in the hemolymph can result from its lack of excretion by the malpighian tubules, as indicated by the complete absence of 5 and 6 in the larval frass of P. cochleariae, C. populi, and C. lapponica.

Neither compound is exported to the defensive system that encoloses the isoazolinone glucosides 5 and 6 efficiently in the hemolymph. The biosynthesis of the isoxazolinone derivatives most likely starts from the metabolism of amino acids (Randoux et al.1991), in particular β-alanine, which is efficiently incorporated into 6 (unpublished); however, neither the enzymatic steps of the pathway nor the regulation of the observed ratio of compound 5 to 6 have to date been resolved in the chrysomelids.

Autolysis of isoxazolinone glucoside ester 6 would lead to free 3-NPA and consequently to autointoxication. To address this possibility, stable isotope labelled 3-NPA was injected into Chrysomelina larvae. Labelled isoxazolinone glucoside ester 6 indicated the conjugation of free 3-NPA to the isoxazolinone glucoside 5. No other detoxification strategies have been found in the Chrysomelina beetles. Such strategies have been reported from other organisms, including the oxidation of 3-NPA that was reported from microbes and plants (Francis et al.2013), the conjugation to amino acids reported from Spodoptora littoralis (Novoselov et al.2015) and melanopline grasshoppers (Johnson et al.2001), or the formation of glucosides (miserotoxin) observed in grasshoppers (Johnson et al.2001). Our findings have two consequences: first, the glucoside and free 3-NPA must be considered as the biosynthetic building blocks of the ester 6; and second, isoxazolinone glucoside 5 likely serves as a storage site for the neurotoxin, which displays the physiological role of isoxazolinone glucoside 5 in Chrysomelina beetles.

The hemolymph surrounds all organs and is thus a vital transport medium between insect tissues. Intense research over the last decades has demonstrated the existence of a transport network in Chrysomelina larvae that is nonspecific in terms of import of dietary glucosides into the hemolymph (Discher et al.2009; Strauss et al.2013). While the larvae excrete glucosides that have not been utilized, they also transport genuine precursors into the defensive glands. In this study, we report for the first time the presence of actual O-β-d glucosides in the hemolymph of chrysomelid larvae, such as salicin 1 and 8-hydroxygeraniol-8-O-ß-D-glucoside 3, which support the transport model presented earlier. Our results lead us to conclude that the open circulation in the hemolymph of Chrysomelina larvae serves on the one hand as a transit site for the glucoside intermediates of the defensive secretions and on the other hand as a storage reservoir for the isoxazolinone derivatives.

Electronic supplementary material

Below is the link to the electronic supplementary material.

ESM 1 (4.3MB, docx)

(DOCX 4412 kb)

Acknowledgments

The authors express their gratitude to Dr. Maritta Kunert and Kerstin Ploss for analytical support and Emily Wheeler, Boston, for editorial assistance. We thank Dr. Grit Kunert for help with statistics. We acknowledge Angelika Berg’s help in rearing the beetles. We also thank Alija Gabdullina, Katon-Karagay State National Park, Kazakhstan, for support during sample collection of C. lapponica. This work was financially supported by the Max Planck Society.

Author contributions

G.P, S.H.vR, A.B., and W.B. designed research and analyzed the data, G.P. performed LC-MS analysis, T.B. synthesized authentic standards and isotope-labelled precursors, P.R. constructed the phylogenetic tree, R.R.B. and J.P. provided reagents and research materials, C.P. and S.H.vR. performed NMR spectroscopic analysis, G.P., S.H.vR., A.B., and W.B. wrote the paper.

References

  1. Anderson RC, Majak W, Rassmussen MA, Callaway TR, Beier RC, Nisbet DJ, Allison MJ. Toxicity and metabolism of the conjugates of 3-nitropropanol and 3-nitropropionic acid in forages poisonous to livestock. J Agric Food Chem. 2005;53:2344–2350. doi: 10.1021/jf040392j. [DOI] [PubMed] [Google Scholar]
  2. Baxter RL, Hanley AB, Chan HWS, Greenwood SL, Abbot EM, Mcfarlane IJ, Milne K. Fungal biosynthesis of 3-nitropropanoic acid. J Chem Soc Perkin Trans. 1992;1:2495–2502. doi: 10.1039/p19920002495. [DOI] [Google Scholar]
  3. Beal MF, et al. Neurochemical and histologic characterization of striatal excitotoxic lesions produced by the mitochondrial toxin 3-nitropropionic acid. J Neurosci. 1993;13:4181–4192. doi: 10.1523/JNEUROSCI.13-10-04181.1993. [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Becker T, Görls H, Pauls G, Wedekind R, Kai M, von Reuss SH, Boland W. Synthesis of isoxazolin-5-one glucosides by a cascade reaction. J Org Chem. 2013;78:12779–12783. doi: 10.1021/jo4023155. [DOI] [PubMed] [Google Scholar]
  5. Becker T, Kartikeya P, Paetz C, von Reuss SH, Boland W. Synthesis and photosensitivity of isoxazolin-5-one glycosides. Org Biomol Chem. 2015;13:4025–4030. doi: 10.1039/C5OB00244C. [DOI] [PubMed] [Google Scholar]
  6. Blum MS, Wallace JB, Duffield RM, Brand JM, Fales HM, Sokoloski EA. Chrysomelidial in the defensive secretion of the leaf beetle Gastrophysa cyanea Melsheimer. J Chem Ecol. 1978;4:47–53. doi: 10.1007/BF00988259. [DOI] [Google Scholar]
  7. Bodemann RR, et al. Precise RNAi-mediated silencing of metabolically active proteins in the defence secretions of juvenile leaf beetles. Proc Biol Sci. 2012;279:4126–4134. doi: 10.1098/rspb.2012.1342. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Chomcheon P, Wiyakrutta S, Sriubolmas N, Ngamrojanavanich N, Isarangkul D, Kittakoop P. 3-Nitropropionic acid (3-NPA), a potent antimycobacterial agent from endophytic fungi: Is 3-NPA in some plants produced by endophytes? J Nat Prod. 2005;68:1103–1105. doi: 10.1021/np050036a. [DOI] [PubMed] [Google Scholar]
  9. Cohen JE, Pimm SL, Yodzis P, Saldana J. Body sizes of animal predators and animal prey in food webs. J Anim Ecol. 1993;62:67–78. doi: 10.2307/5483. [DOI] [Google Scholar]
  10. Daccordi M (1994) Notes for phylogenetic study of Chrysomelinae, with descriptions of new taxa and a list of all the known genera (Coleoptera: Chrysomelidae, Chrysomelinae). Proceedings of the Third International Symposium on the Chrysomelidae, Beijing, 1992
  11. Deroe C, Pasteels JM. Distribution of adult defense glands in chrysomelids (Coleoptera: Chrysomelidae) and its significance in the evolution of defense mechanisms within the family. J Chem Ecol. 1982;8:67–82. doi: 10.1007/BF00984006. [DOI] [PubMed] [Google Scholar]
  12. Discher S, Burse A, Tolzin-Banasch K, Heinemann SH, Pasteels JM, Boland W. A versatile transport network for sequestering and excreting plant glycosides in leaf beetles provides an evolutionary flexible defense strategy. ChemBioChem. 2009;10:2223–2229. doi: 10.1002/cbic.200900226. [DOI] [PubMed] [Google Scholar]
  13. Francis K, Smitherman C, Nishino SF, Spain JC, Gadda G. The biochemistry of the metabolic poison propionate 3-nitronate and its conjugate acid, 3-nitropropionate. IUBMB Life. 2013;65:759–768. doi: 10.1002/iub.1195. [DOI] [PubMed] [Google Scholar]
  14. Geiselhardt S, Hilker M, Muller F, Kozlov MV, Zvereva EL. Inter- and intrapopulation variability in the composition of larval defensive secretions of willow-feeding populations of theleaf beetle Chrysomela lapponica. J Chem Ecol. 2015;41:276–286. doi: 10.1007/s10886-015-0558-x. [DOI] [PubMed] [Google Scholar]
  15. Gomez-Zurita J, Hunt T, Kopliku F, Vogler AP. Recalibrated tree of leaf beetles (Chrysomelidae) indicates independent diversification of angiosperms and their insect herbivores. PLoS ONE. 2007;2 doi: 10.1371/journal.pone.0000360. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Gross J, Schmidtberg H. Glands of leaf beetle larvae - protective structures against attacking predators and pathogens. Res Chrysomelidae. 2009;2(2):177–189. doi: 10.1163/ej.9789004169470.1-299.63. [DOI] [Google Scholar]
  17. Gross J, Muller C, Vilcinskas A, Hilker M. Antimicrobial activity of exocrine glandular secretions, hemolymph, and larval regurgitate of the mustard leaf beetle Phaedon cochleariae. J Invertebr Pathol. 1998;72:296–303. doi: 10.1006/jipa.1998.4781. [DOI] [PubMed] [Google Scholar]
  18. Gross J, Podsiadlowski L, Hilker M. Antimicrobial activity of exocrine glandular secretion of Chrysomela larvae. J Chem Ecol. 2002;28:317–331. doi: 10.1023/A:1017934124650. [DOI] [PubMed] [Google Scholar]
  19. Gross J, Schumacher K, Schmidtberg H, Vilcinskas A. Protected by fumigants: beetle perfumes in antimicrobial defense. J Chem Ecol. 2008;34:179–188. doi: 10.1007/s10886-007-9416-9. [DOI] [PubMed] [Google Scholar]
  20. Hilker M, Schulz S. Composition of larval secretion of Chrysomela lapponica (Coleoptera, Chrysomelidae) and its dependence on host plant. J Chem Ecol. 1994;20:1075–1093. doi: 10.1007/BF02059744. [DOI] [PubMed] [Google Scholar]
  21. Huang LS, et al. 3-nitropropionic acid is a suicide inhibitor of mitochondrial respiration that, upon oxidation by complex II, forms a covalent adduct with a catalytic base arginine in the active site of the enzyme. J Biol Chem. 2006;281:5965–5972. doi: 10.1074/jbc.M511270200. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Johnson DL, Majak W, Benn MH. Excretion of miserotoxin and detoxification of the aglycone by grasshoppers (Orthoptera: Acrididae) Phytochemistry. 2001;58:739–742. doi: 10.1016/S0031-9422(01)00310-7. [DOI] [PubMed] [Google Scholar]
  23. Kim DH, Kim BR, Kim JY, Jeong YC. Mechanism of covalent adduct formation of aucubin to proteins. Toxicol Lett. 2000;114:181–188. doi: 10.1016/S0378-4274(99)00295-7. [DOI] [PubMed] [Google Scholar]
  24. Kuhn J, Pettersson EM, Feld BK, Burse A, Termonia A, Pasteels JM, Boland W. Selective transport systems mediate sequestration of plant glucosides in leaf beetles: a molecular basis for adaptation and evolution. Proc Natl Acad Sci U S A. 2004;101:13808–13813. doi: 10.1073/pnas.0402576101. [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Laurent P, Braekman JC, Daloze S (2005) Insect chemical defense. In: Schulz S (ed) Chemistry of pheromones and other semiochemicals II. vol Copyright (C) 2013 American Chemical Society (ACS). All Rights Reserved. Springer Berlin Heidelberg, Berlin, p 167–229
  26. Lenz EM, Hagele BF, Wilson ID, Simpson SJ. High resolution 1H NMR spectroscopic studies of the composition of the haemolymph of crowd- and solitary-reared nymphs of the desert locust, Schistocerca gregaria. Insect Biochem Mol Biol. 2001;32:51–56. doi: 10.1016/S0965-1748(01)00078-9. [DOI] [PubMed] [Google Scholar]
  27. Majak W, Johnson DL, Benn MH. Detoxification of 3-nitropropionic acid and karakin by melanopline grasshoppers. Phytochemistry. 1998;49:419–422. doi: 10.1016/S0031-9422(97)00733-4. [DOI] [Google Scholar]
  28. Matsuda K, Sugawara F. Defensive secretion of Chrysomelid larvae Chrysomela vigintipunctata costella (Marseul), C. populi L. and Gastrolina depressa Baly (Coleoptera: Chrysomelidae) Appl Entomol Zool. 1980;15:316–320. [Google Scholar]
  29. Novoselov A, Becker T, Pauls G, von Reuss SH, Boland W. Spodoptera littoralis detoxifies neurotoxic 3-nitropropanoic acid by conjugation with amino acids. Insect Biochem Mol Biol. 2015;63:97–103. doi: 10.1016/j.ibmb.2015.05.013. [DOI] [PubMed] [Google Scholar]
  30. Opitz SEW, Muller C. Plant chemistry and insect sequestration. Chemoecology. 2009;19:117–154. doi: 10.1007/s00049-009-0018-6. [DOI] [Google Scholar]
  31. Palokangas P, Neuvonen S. Differences between species and instars of Phratora leaf beetles (Coleoptera, Chrysomelidae) in the probability of being preyed on. Ann Zool Fenn. 1992;29:273–278. [Google Scholar]
  32. Parry R, Nishino S, Spain J. Naturally-occurring nitro compounds. Nat Prod Rep. 2011;28:152–167. doi: 10.1039/C0NP00024H. [DOI] [PubMed] [Google Scholar]
  33. Pasteels JM, Gregoire JC, Rowellrahier M. The chemical ecology of defense in arthropods. Annu Rev Entomol. 1983;28:263–289. doi: 10.1146/annurev.en.28.010183.001403. [DOI] [Google Scholar]
  34. Pasteels JM, Rowell-Rahier M, Braekman JC, Dupont A. Salicin from host plant as precursor of salicyl aldehyde in defensive secretion of chrysomeline larvae. Physiol Entomol. 1983;8:307–314. doi: 10.1111/j.1365-3032.1983.tb00362.x. [DOI] [Google Scholar]
  35. Pasteels JM, Daloze D, Rowellrahier M. Chemical defense in chrysomelid eggs and neonate larvae. Physiol Entomol. 1986;11:29–37. doi: 10.1111/j.1365-3032.1986.tb00388.x. [DOI] [Google Scholar]
  36. Pasteels JM, Rowellrahier M, Braekman JC, Daloze D, Duffey S. Evolution of exocrine chemical defense in leaf beetles (Coleoptera, Chrysomelidae) Experientia. 1989;45:295–300. doi: 10.1007/BF01951815. [DOI] [Google Scholar]
  37. Pasteels JM, Termonia A, Daloze D, Windsor DM (2003) Distribution of toxins in chrysomeline leaf beetles: Possible taxonomic inferences. In: Furth DG (ed) Proceedings of the Fifth International Symposium on the Chrysomelidae, Iguacu, 2000, vol 21. Pensoft Publishers, p 261–275
  38. Peters FT, Remane D. Aspects of matrix effects in applications of liquid chromatography-mass spectrometry to forensic and clinical toxicology—a review. Anal Bioanal Chem. 2012;403:2155–2172. doi: 10.1007/s00216-012-6035-2. [DOI] [PubMed] [Google Scholar]
  39. Phalaraksh C, et al. NMR spectroscopic studies on the haemolymph of the tobacco hornworm, Manduca sexta: assignment of 1H and 13C NMR spectra. Insect Biochem Mol Biol. 1999;29:795–805. doi: 10.1016/S0965-1748(99)00053-3. [DOI] [Google Scholar]
  40. Poynton HC, et al. Metabolomics of microliter hemolymph samples enables an improved understanding of the combined metabolic and transcriptional responses of Daphnia magna to cadmium. Environ Sci Technol. 2011;45:3710–3717. doi: 10.1021/es1037222. [DOI] [PubMed] [Google Scholar]
  41. Randoux T, Braekman JC, Daloze D, Pasteels JM. De novo biosynthesis of 3-Isoxazolin-5-one and 3-Nitropropanoic acid derivatives in Chrysomela tremulae. Naturwissenschaften. 1991;78:313–314. doi: 10.1007/BF01221417. [DOI] [PubMed] [Google Scholar]
  42. Strauss AS, Peters S, Boland W, Burse A. ABC transporter functions as a pacemaker for sequestration of plant glucosides in leaf beetles. eLife. 2013;2 doi: 10.7554/eLife.01096. [DOI] [PMC free article] [PubMed] [Google Scholar]
  43. Sugeno W, Matsuda K. Adult secretions of four Japanese Chrysomelinae (Coleoptera : Chrysomelidae) Appl Entomol Zool. 2002;37:191–197. doi: 10.1303/aez.2002.191. [DOI] [Google Scholar]
  44. Termonia A, Pasteels JM. Larval chemical defence and evolution of host shifts in Chrysomela leaf beetles. Chemoecology. 1999;9:13–23. doi: 10.1007/s000490050029. [DOI] [Google Scholar]
  45. Termonia A, Hsiao TH, Pasteels JM, Milinkovitch MC. Feeding specialization and host-derived chemical defense in Chrysomeline leaf beetles did not lead to an evolutionary dead end. Proc Natl Acad Sci U S A. 2001;98:3909–3914. doi: 10.1073/pnas.061034598. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Topp W (1997) Survival in a hostile environment: evaluation of the developmental success of the oligophagous leaf beetle Chrysomela vigintipunctata (Scop). In: Dettner K, Bauer G, Voekl W (eds) Ecological studies; Vertical food web interactions: Evolutionary patterns and driving forces, vol 130. Ecological Studies. p 147–169
  47. Weldon PJ. Chemical aposematism. Chemoecology. 2013;23:201–202. doi: 10.1007/s00049-013-0140-3. [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

ESM 1 (4.3MB, docx)

(DOCX 4412 kb)

Articles from Journal of Chemical Ecology are provided here courtesy of Springer

RESOURCES