Journal list menu
Herbivore-induced phenylacetonitrile is biosynthesized from de novo-synthesized l-phenylalanine in the giant knotweed, Fallopia sachalinensis
Abstract
Plants emit a series of characteristic volatile blends when damaged by insect feeding. Phenylacetonitrile is one of the volatiles from the leaves of the giant knotweed, Fallopia sachalinensis, infested by the Japanese beetle, Popillia japonica, or treated with exogenous airborne methyl jasmonate (MeJA). We examined the precursor of the nitrile and its origin in this system. l-Phenylalanine was determined to be a precursor of the nitrile in F. sachalinensis leaves, and the phenylalanine was also induced by beetle feeding and MeJA treatment. We also found that exogenous MeJA enhanced the biosynthesis of several amino acids in F. sachalinensis leaves.
1 Introduction
Plants are known to produce various low-molecular-weight secondary metabolites functioning as direct/indirect chemical defenses, such as alkaloids, flavonoids, phenylpropanoids and terpenoids, when attacked by herbivores or pathogens [1-3]. The production of these plant secondary metabolites is induced not only by elicitors derived from herbivores or pathogens [4-6] but also by jasmonic acid and its derivatives (collectively named as jasmonates) [7-13]. Jasmonates are well known to regulate plant defense responses as downstream signaling components of herbivore/pathogen elicitors as well as growth and development [6, 14-18].
Of induced plant defense substances, phenylpropanoids and alkaloids are well known to be derived from their corresponding amino acids, e.g. phenylalanine and tyrosine [10, 19-21]; for example, phenylalanine ammonia lyase (PAL), which is known to be induced by herbivory or jasmonates catalyzes phenylalanine into cinnamic acid as a precursor of phenylpropanoids that lead to diverse compounds, such as flavonoids, coumarins, lignins and other related compounds [8, 22]. In most cases, the labeled amino acids were exogenously fed to plants and their incorporation into secondary metabolites was examined to confirm whether these secondary metabolites are derived from the corresponding amino acids. However, little attention has given to the changes of the endogenous level of amino acids in plants where the biosynthesis of secondary metabolites is activated by insect herbivory, fungal infections or jasmonates, and thus the supply sources of amino-acid precursors of the induced amino-acid derived secondary metabolites need to be examined. In general, constitutive level of amino acids in plants could be maintained constantly by positive/negative feedback regulation of the end products, but plants must have a supply of amino acids to induce the production of secondary metabolites in case of attacks by herbivores and pathogens, and environmental changes. Therefore, a redundant amount of amino acids must be newly provided prior to the accumulation of secondary metabolites or a pool of amino acids must coexist where the production of secondary metabolites is induced.
Recently, we found that the leaves of the giant knotweed, Fallopia sachalinensis (Polygonaceae), infested by the Japanese beetle, Popillia japonica (Coleoptera: Scarabaeidae), emit a characteristic volatile blend of phenylacetonitrile and some terpenes [23]. Most herbivore-induced plant volatiles serve as a cue for predators and parasitoids to explore herbivores as preys or hosts [1, 24], but in some cases, these plant volatiles are beneficially used by herbivores as cues to exploit plant food sources [25, 26] as we observed in the relationship between F. sachalinensis and P. japonica. Exogenous MeJA application mimicked the induction of volatiles, suggesting that nitrile production could be regulated by jasmonate signal transduction in F. sachalinensis.
Although plant nitriles are thought to be biosynthesized from their corresponding amino acids as intermediates of cyanogenic glycosides [27-29] or to occur as decomposed products of glucosinolates [30], nitrile formation as an end product is uncommon in plants and information about nitrile biosynthesis is still limited. We predict that phenylacetonitrile from beetle-infested leaves of F. sachalinensis is synthesized from phenylalanine, and thus this system could be suitable to investigate nitrile biosynthesis in plants and the interface between amino-acid precursors and induced secondary metabolites. In this article, we investigated (1) the precursor of phenylacetonitrile found from the leaves F. sachalinensis, and (2) the supply source of the precursor, phenylalanine, in this plant using MeJA and an inhibitor of phenylalanine biosynthesis. We also report that several amino acids are induced by insect feeding or MeJA treatment in F. sachalinensis.
2 Materials and methods
2.1 Chemicals
l-Phenyl-d5-alanine-2,3,3-d3 was purchased from Cambridge Isotope Laboratories, Inc. (Andover, MA, USA). l-Phenylalanine, d-phenylalanine and dl-p-chlorophenylalanine were all from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). Glyphosate was from Kanto Chemical Co., Inc. (Tokyo, Japan). Ethyl chloroformate (ECF) and 1-fluoro-2,4-dinitrophenyl-5-l-leucinamide (l-FDLA) were from Tokyo Chemical Industry Co., Ltd. (Tokyo, Japan). Methyl jasmonate (MeJA) was from Sigma–Aldrich (St. Louis, MO, USA). All other chemicals used in this study were of high analytical grade or HPLC grade.
2.2 Plants
Roots of the giant knotweed, F. sachalinensis, were collected in Kamishinjyo, Akita City, Japan in August, 2009. The roots were kept at room temperature in commercially available culture soil. New shoots were grown from the roots and then mature green leaves of F. sachalinensis were used for this study.
2.3 Application of deuterium-labeled phenylalanine and analysis of plant volatiles
To determine whether phenylalanine is the precursor of phenylacetonitrile from the leaves of F. sachalinensis, volatiles from the leaves fed on deuterium-labeled phenylalanine were analyzed by GC–MS. A mature leaf of F. sachalinensis was excised from the base of its petiole and the petiole was placed in 1.5 ml of distilled water containing 1 mg/ml l-phenyl-d5-alanine-2,3,3-d3 in an Eppendorf tube (N = 7). The top of the tube was sealed with parafilm M (American National Can Co., Norwalk, CT, USA) to prevent water evaporation. Volatile production was induced by airborne treatment of MeJA. Each leaf was placed in a glass container (1000 ml) and a paper disk loaded with 2 μl MeJA was then introduced into the glass container without any contact with the leaf. The container was placed in a chamber at 25 °C under continuous light, and 18 h later the volatiles in the container were collected using solid phase micro extraction (SPME) fiber (65 μm Stable Flex PDMS/DVB; Supelco, Bellefonte, PA, USA) for 30 min. The SPME fiber was then injected into the GC–MS for 5 min and the volatile components were analyzed. GC/MS analysis was carried out using a Clarus 600 GC–MS (Perkin–Elmer, Shelton, CT, USA) operated at 70 eV using a DB-5MS capillary column (Agilent Technologies, Inc., Santa Clara, CA, USA, 30 m × 0.25 mm i.d., 0.25 μm film thickness) with helium carrier gas at 1.0 ml/min. Both the injector temperature and detector temperature were maintained at 250 °C. For volatile analysis, the oven temperature was programmed to change from 50 °C (3 min holding) to 290 °C at 10 °C/min and held for 3 min. The compounds were identified by comparing their GC retention times and mass spectra with our previous results [23].
2.4 Inhibition assay of phenylacetonitrile production by glyphosate and rescue assay with exogenous phenylalanine
To examine the supply source of phenylalanine in the leaves of F. sachalinensis, the effect of glyphosate, an inhibitor of phenylalanine biosynthesis [31], on phenylacetonitrile production was evaluated. A leaf of F. sachalinensis cut from a stem was placed in 1.5 ml distilled water containing 8 mg/ml glyphosate in an Eppendorf tube, and the leaf was enclosed in a glass container in the same way as described above. The leaf was incubated at 25 °C under continuous light for 6 h, and then the leaf was exposed to airborne MeJA for 18 h. Then 50 μl acetone containing 20 μg/μl n-octane as an internal standard was added to the container, and the volatiles from the F. sachalinensis leaf were collected by SPME and analyzed by GC–MS as described above. The positive control used distilled water instead of glyphosate solution. The analyses were replicated five times using different leaves for each treatment.
To evaluate the effect of exogenous phenylalanine on the inhibition of phenylacetonitrile production by glyphosate, either d- or l-phenylalanine was added to the glyphosate solution (final concentration of phenylalanine was 1 mg/ml), and then the induced volatiles were analyzed in the same procedure described above (each n = 5).
Quantification analysis of phenylacetonitrile was performed with selected ion monitoring using the ions m/z 117 for phenylacetonitrile and m/z 85 for the internal standard (n-octane), respectively. The ratio of peak area of m/z 117 to m/z 85 was calculated for the positive control, glyphosate treatment and glyphosate treatment with d-/l-phenylalanine. The resulting values of the ratio were divided by the leaf area (cm2) of each tested leaf to standardize the data. All data of the relative ratio of phenylacetonitrile production per treatment were analyzed with the Tukey–Kramer test using JMP 5.1.2 [32].
2.5 Analysis of amino acid accumulations triggered by beetle-feeding or exogenous MeJA
To confirm the induction of de novo phenylalanine biosynthesis triggered by beetle feeding or MeJA treatment, the changes of the endogenous level of amino acids in F. sachalinensis leaves infested by beetles or treated by MeJA were analyzed by GC–MS.
The leaves of F. sachalinensis either infested by the Japanese beetle or treated with exogenous airborne MeJA for 18 h at 25 °C under continuous light were homogenized in 15 ml of 0.1 M HCl. The homogenates were centrifuged (14 100×g, 5 min at room temperature), and then aliquots (1 ml) of the supernatants were passed through an Oasis MCX cartridge (500 mg/6 cc; Waters Co., Milford, MA, USA) to extract amino acids and amines according to the protocol described in Silva et al. [33]. The eluates containing amino acids and amines were evaporated and dissolved in 100 μl distilled water containing 10 μg dl-p-chlorophenylalanine as an internal standard. The solutions were mixed with 80 μl ethanol/pyridine (4:1), and then 10 μl ECF was added and mixed vigorously to derivatize amino acids and amines. After 5 min reaction, 200 μl dichloromethane was added to the mixture to extract the derivatives, and then 1 μl of the extracts was analyzed by GC–MS. For the analysis of ECF derivatives of amino acids, the oven temperature was programmed to change from 100 °C (5 min holding) to 290 °C at 10 °C/min and held for 3 min. Other conditions were the same as described above.
The accumulations of amino acids in the leaves of F. sachalinensis after MeJA treatment were examined by the time-dependent extraction of amino acids from the leaves (0, 3, 6, 12 and 24 h; each n = 3). Quantification analysis of amino acids was performed as that of the corresponding ECF derivatives with selected ion monitoring using the following ions: m/z 116 for alanine, m/z 144 for valine, m/z 158 for leucine and isoleucine, m/z 142 for proline, m/z 91 for phenylalanine and phenethylamine, m/z 81 for histidine, m/z 107 for tyrosine and tyramine, m/z 130 for tryptophan, and m/z 125 for the internal standard (dl-p-chlorophenylalanine). The amount of amino acids per leaf area (cm2) was determined by comparing the peak area ratio in the sample with those found in the calibration standard. All data were analyzed with one-way ANOVA followed by the Dunn's multiple comparison test using JMP 5.1.2 [32].
2.6 l-FDLA derivatization of amino acids and LC/MS analysis of the derivatives
To determine the absolute configuration of phenylalanine in the leaves of F. sachalinensis, the amino acids were derivatized with l-FDLA, and then the derivatives were analyzed by LC–MS. The amino acids were extracted from the MeJA-elicited F. sachalinensis leaves by the same procedure as described above (N = 5). The derivatization of the amino acids with l-FDLA was performed according to the method described in Fujii et al. [34]. Briefly, the extracted amino acid mixtures were evaporated and dissolved in 50 μl distilled water, and the solution was mixed with 20 μl of 1 M sodium bicarbonate and 100 μl of 1% l-FDLA in acetone. The mixture was incubated at 37 °C for 1 h, and then quenched with 20 μl of 1 M HCl on ice. Aliquots (5 μl) of the products were diluted with 495 μl acetonitrile and 2 μl diluted solutions were analyzed by LC–MS, a TSQ Quantum Ultra equipped with Accela 600 HPLC system (Thermo Fisher Scientific Inc., Waltham, MA, USA). l-FDLA derivatives were separated on a reversed-phase column (Inertsil ODS-4, 50 × 2.1 mm i.d., 2 μm particle size; GL Sciences Inc., Tokyo, Japan) maintained at 40 °C with solvent gradient elution using 20 mM ammonium acetate, pH 4.0 (solvent A) and methanol (solvent B) at a flow rate of 0.2 ml/min. The solvent gradient was programmed from 10% B to 50% B over 4 min followed by 50–100% B for 16 min, and then held at 100% B for 3 min. The l-FDLA derivatives of phenylalanine were detected by Q1 with selected ion monitoring at m/z 458 ([M−H]−) in negative ion mode using electro spray ionization (ESI voltage, 3000 V; capillary temperature, 330 °C; sheath gas, 50 arbitrary units; auxiliary gas, 10 arbitrary units). The authentic standards of the l-FDLA derivatives of d- and l-phenylalanine, and their mixture were also prepared and analyzed by the same procedure.
3 Results
Deuterium-labeled l-phenylalanine was incorporated into phenylacetonitrile (Fig. 1 ). Deuterated phenylacetonitrile was detected at 10.4 min and was clearly separated from natural phenylacetonitrile (tR = 10.5 min). The identification of deuterated phenylacetonitrile was confirmed based on its molecular ion at m/z 124 that derived from seven deuterium atoms, and a fragment ion observed at m/z 96 (M+−28) corresponding to the elimination of DCN from the molecular ion (Fig. 1B and C). The rate of the labeled to the natural phenylacetonitrile was 18.0 ± 6.3% (mean ± S.D., N = 7).
Glyphosate inhibited phenylacetonitrile production in MeJA-elicited leaves of F. sachalinensis (relative amount ± S.D. = 1.9 ± 1.3%, N = 5; Fig. 2 B and E) compared to the positive control treatment, water (relative amount ± S.D. = 100 ± 36.1%, N = 5; Fig. 2A and E). Glyphosate-mediated inhibition of phenylacetonitrile production was recovered considerably by the addition of l-phenylalanine (relative amount ± S.D. = 77.3 ± 28.1%, N = 5; Fig. 2C and E), while exogenous d-phenylalanine did not work as a suitable substrate for phenylacetonitrile synthesis (relative amount ± S.D. = 31.0 ± 15.3%, N = 5; Fig. 2D and E).
Both exogenous MeJA and beetle feeding induced the accumulation of phenylalanine, which was not detected in the undamaged leaves of F. sachalinensis (Fig. 3 A–D). The accumulation of phenylalanine was observed 3 h after MeJA treatment (amount ± S.D. = 0.46 ± 0.012 μg/cm2 leaf, N = 3), and the amount of phenylalanine increased time dependently during the experimental period (24 h, amount ± S.D. = 1.84 ± 0.80 μg/cm2 leaf, N = 3; Fig. 2D). Our previous results showed that phenylacetonitrile emission was detectable 6 h after MeJA treatment [23], thus the accumulation of phenylalanine triggered by MeJA was prior to nitrile synthesis.
l-Phenylalanine was dominant in MeJA-induced phenylalanine in the leaves (91.2–98.9%), even though a small amount of d-phenylalanine coexisted in the leaves (d-form/l-form + d-form = 4.9 ± 3.1%, mean ± S.D.%, N = 5; Fig. 4 ).
All amino acids except alanine increased time dependently after MeJA treatment, similarly to phenylalanine, while alanine was constantly present in the leaves of F. sachalinensis (Fig. 3D). Alanine metabolism may be independent of the jasmonate signaling pathway in this plant.
An amine, tyramine was identified as a constitutive component in the leaves of F. sachalinensis, and its amount increased gradually after MeJA treatment. Phenethylamine was not detected from the intact leaves of F. sachalinensis, and this amine was induced by MeJA treatment. The amount of phenethylamine reached a plateau at 3 h after MeJA treatment and remained almost constant during the experimental period (24 h).
4 Discussion
Our results clearly showed that phenylacetonitrile is derived from de novo synthesized l-phenylalanine in the MeJA-elicited leaves of F. sachalinensis. The inhibition of phenylacetonitrile by glyphosate suggests that the precursor of phenylacetonitrile, i.e. phenylalanine, is provided by the activated shikimate pathway and not from proteolytic degradation or a pool of phenylalanine. The time-dependent accumulation of phenylalanine in the leaves after MeJA treatment also supports the activation of de novo phenylalanine biosynthesis triggered by exogenous MeJA. Phenylalanine was not detected in the non-treated leaves of F. sachalinensis, suggesting that the redundant phenylalanine for phenylacetonitrile synthesis does not normally exist in the leaves.
The induced phenylalanine in F. sachalinensis may be derived from a new carbon source as described in tobacco where MeJA mediates the reprogramming of carbon resource allocation [35]. In our experiments, no carbon sources were provided to the leaves of F. sachalinensis, and phenylalanine is synthesized using endogenous carbon sources in the plant. Sucrose depletion was observed in MeJA-elicited Medicago truncatula [36], and thus together with this evidence of metabolic changes, exogenous MeJA could also shift the carbon partitioning to the shikimate pathway and the downstream nitrile biosynthesis in F. sachalinensis. The reallocation of carbon resulting from the decarboxylation of phenylalanine in nitrile production will be explored further.
The accumulation of phenylalanine was also observed when the leaves were infested by the Japanese beetle; thus, this phenomenon could be specific to herbivory whose information is further transmitted through jasmonate signal transduction in F. sachalinensis. The results using glyphosate and exogenous l-phenylalanine indicate that MeJA could activate both shikimate and phenylacetonitrile synthesis pathways (Fig. 5 ). The coactivation of the primary metabolic pathway (shikimate pathway) and its subsequent secondary metabolic pathway (phenylpropanoid pathway) has also been reported in Hemidesmus indicus roots and cultured tomato cells treated by microbial elicitors [37, 38].
It is reported that the desert locust, Schistocerca gregaria, synthesizes phenylacetonitrile as a pheromone from phenylalanine [39]. In microbes, phenylacetonitrile is also synthesized from phenylalanine via phenylacetoaldoxime [40]. We have not detected phenylacetaldoxime in F. sachalinensis so far, and it remains unknown whether the biosynthetic pathway of the nitrile in F. sachalinensis is the same as that of different organisms. Phenethylamine, a possible decarboxylated product of phenylalanine, has been found; however, we have not yet confirmed that the labeled phenylalanine can be incorporated into phenethylamine in F. sachalinensis, and it remains to be determined whether phenethylamine is an intermediate of nitrile biosynthesis.
Phenylacetonitrile is the only amino-acid derived nitrile detected from the leaves of MeJA-elicited F. sachalinensis in contrast to the induction of several other amino acids by MeJA. This may suggest that the initial step of phenylacetonitrile biosynthesis from phenylalanine is catalyzed by an enzyme with high substrate specificity. High substrate specificity was also observed in cytochrome P450 enzymes that convert aromatic amino acids in the first step of cyanogenic glucosides biosynthesis [41, 42]. Furthermore, the addition of d-phenylalanine, an unnatural amino acid in F. sachalinensis, did not successfully recover phenylacetonitrile production, suggesting the presence of stereoselective reaction(s) from phenylalanine to phenylacetonitrile in F. sachalinensis. Future investigations will reveal the enzymatic characteristics in nitrile production and also provide new information on the interface of primary and secondary metabolic pathways.
Our results showed that the biosynthetic pathways of several amino acids were also upregulated by jasmonates in the leaves of F. sachalinensis. Even though the fate of these induced amino acids is still unknown, some of these amino acids could be utilized for protein synthesis. Jasmonate-induced proteins were observed in F. sachalinensis (Noge et al., unpublished) and other plant species [43-45]. The biosynthesis of branched-chain amino acids (leucine, isoleucine and valine) was elicited by MeJA in Medicago cell cultures [36] and tryptophan accumulation was observed in jasmonic acid-treated barley leaves [46]. The increase of Trp was also reported in herbivore-damaged leaves of maize [47] as we observed in the leaves of F. sachalinensis infested by P. japonica. Proline is a well-known osmolyte against water stress in plants, and this amino acid is known to be accumulated by either jasmonic acid [43] or abscisic acid (ABA) [48]. ABA also activates the accumulation of branched-chain amino acids in response to water deficiency [49]. It remains to be determined whether the accumulation of proline and branched-chain amino acids in F. sachalinensis is the result of a direct or indirect effect of MeJA. The jasmonate signaling pathway is known to interact with ABA and other plant hormones, but their interactions still need to be elucidated [50]; thus, the phenomenon of amino acid accumulation in F. sachalinensis could be a good tool to examine in planta hormone crosstalk.
5 Concluding remarks
In conclusion, we determined that l-phenylalanine was a precursor of phenylacetonitrile, a herbivore-induced volatile, in F. sachalinensis leaves and that phenylalanine was de novo synthesized triggered by beetle feeding or exogenous airborne MeJA. These results indicate that MeJA, a downstream signal transduction molecule, activates both the shikimate pathway and the nitrile biosynthetic pathway.
Acknowledgements
This work was partly supported by the Asahi Glass Foundation and a Grant from the to K.N. The LC–MS, a TSQ Quantum Ultra equipped with Accela 600 HPLC system, used in this study is a commonly shared instrument at Akita Prefectural University.
