SEARCH

SEARCH BY CITATION

Keywords:

  • Arabidopsis thaliana;
  • Brassicaceae;
  • camalexin;
  • glucosinolates;
  • plant immunity;
  • tryptophan metabolism

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • A hallmark of the innate immune system of plants is the biosynthesis of low-molecular-weight compounds referred to as secondary metabolites. Tryptophan-derived branch pathways contribute to the capacity for chemical defense against microbes in Arabidopsis thaliana.
  • Here, we investigated phylogenetic patterns of this metabolic pathway in relatives of A. thaliana following inoculation with filamentous fungal pathogens that employ contrasting infection strategies.
  • The study revealed unexpected phylogenetic conservation of the pathogen-induced indole glucosinolate (IG) metabolic pathway, including a metabolic shift of IG biosynthesis to 4-methoxyindol-3-ylmethylglucosinolate and IG metabolization. By contrast, indole-3-carboxylic acid and camalexin biosyntheses are clade-specific innovations within this metabolic framework. A Capsella rubella accession was found to be devoid of any IG metabolites and to lack orthologs of two A. thaliana genes needed for 4-methoxyindol-3-ylmethylglucosinolate biosynthesis or hydrolysis. However, C. rubella was found to retain the capacity to deposit callose after treatment with the bacterial flagellin-derived epitope flg22 and pre-invasive resistance against a nonadapted powdery mildew fungus.
  • We conclude that pathogen-inducible IG metabolism in the Brassicaceae is evolutionarily ancient, while other tryptophan-derived branch pathways represent relatively recent manifestations of a plant–pathogen arms race. Moreover, at least one Brassicaceae lineage appears to have evolved IG-independent defense signaling and/or output pathway(s).

Introduction

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Plant innate immunity relies on multiple pathogen-inducible biochemical processes that are initiated upon nonself perception by immune sensors (Panstruga et al., 2009; Dodds & Rathjen, 2010). One output branch involves the microbe-triggered biosynthesis, intracellular accumulation and secretion of low-molecular-weight compounds referred to as secondary metabolites (Dixon, 2001; Bednarek & Schulze-Lefert, 2008; Bednarek & Osbourn, 2009; Bednarek et al., 2010). Remarkably, the occurrence of particular plant metabolites with functions in plant disease resistance is often restricted to narrow phylogenetic clades or even to particular plant species. This structural diversity of secondary metabolites and their demonstrated role in plant immune responses suggest that the capacity of plants to generate and maintain a diversified chemical arsenal is evolutionarily driven by parasites.

Recognition of microbe-associated molecular patterns (MAMPs) stimulates indole glucosinolate (IG) metabolism in the reference plant Arabidopsis thaliana (Bednarek et al., 2009; Clay et al., 2009; Sanchez-Vallet et al., 2010). This is critical for effective A. thaliana defense responses against a broad range of fungal and oomycete pathogens (Lipka et al., 2005; Adie et al., 2007; Bednarek et al., 2009; Maeda et al., 2009; Consonni et al., 2010; Hiruma et al., 2010; Sanchez-Vallet et al., 2010; Schlaeppi et al., 2010). Glucosinolates are β-thioglucosides produced constitutively by many plant species of the order Capparales (Rodman et al., 1998; Mithen et al., 2010). These compounds can be derived from a few amino acids, including tryptophan, the precursor of IGs. Glucosinolates have been investigated for decades as precursors of insect-deterring molecules (Blau et al., 1978). This function in plant–insect interactions is dependent on the activity of β-thioglucosidases, known as myrosinases, which in intact plant tissue are compartmentalized away from glucosinolate substrates at cellular or subcellular levels. Tissue damage caused by chewing insects initiates myrosinase-catalyzed glucosinolate hydrolysis, leading to the generation of a broad range of end-products, of which isothiocyanates are thought to possess the highest biological (toxic) activity (Bones & Rossiter, 2006). It was postulated that a similar activation principle and the same end-products are engaged in defense responses to microbial invaders in the Brassicaceae (Tierens et al., 2001). However, the recently reported function of IGs in A. thaliana MAMP-triggered immunity is completely independent from cellular destruction and involves a distinctive pathway for IG conversion (Bednarek et al., 2009; Clay et al., 2009). The latter metabolic processes take place in living plant cells and are initiated by a peroxisome-associated myrosinase, PENETRATION2 (PEN2), which accumulates at pathogen contact sites (Lipka et al., 2005; Bednarek et al., 2009). PEN2-initiated IG metabolism leads to products that are different from the metabolites generated during tissue disruption. This difference probably results from the formation of dithiocarbamate isothiocyanate-glutathione adducts, which are subsequently processed to raphanusamic acid (RA) and the respective amines, including indol-3-ylmethylamine (I3A; Fig. 1) (Bednarek et al., 2009). Genetic evidence suggests that the bioactive end-products of the PEN2 metabolic pathway are transported to the apoplast by the plasma membrane-resident ATP-binding cassette-type (ABC) transporter PEN3/ABCG36 (Stein et al., 2006).

image

Figure 1. Tryptophan-derived metabolites in Arabidopsis thaliana. (a) Biosynthesis of tryptophan-derived metabolites and their involvement in defense against fungal and oomycete pathogens. Components shown to be critical for the termination of pathogen growth are highlighted by frames. Dashed lines indicate putative steps and components. (b) Structures of metabolites discussed in the main text. I3A, indol-3-ylmethylamine; PEN, PENETRATION; RA, raphanusamic acid.

Download figure to PowerPoint

Arabidopsis thaliana, as well as other IG-accumulating Brassicaceae species, can produce four IGs that vary with respect to the substitution on the indole core: indol-3-ylmethylglucosinolate (I3G), 1-methoxyindol-3-ylglucosinolate (1MI3G), 4-methoxyindol-3-ylmethylglucosinolate (4MI3G) and 4-hydroxyindol-3-ylmethylglucosinolate, which is usually found at low concentrations (Petersen et al., 2002; Brown et al., 2003). The concentrations of the three major IGs can vary significantly between accessions of Arabidopsis spp. (Kliebenstein et al., 2001; Windsor et al., 2005). However, an IG-free accession of A. thaliana has not been identified to date. Pathogen inoculation or MAMP treatment redirects A. thaliana IG biosynthesis to 4-substituted IGs whose formation is mediated by the CYP81F2 P450 monooxygenase (Bednarek et al., 2009; Clay et al., 2009; Pfalz et al., 2009, 2011). Thus, the IG defense pathway is integrated into the framework of innate immune responses that are initiated by the perception of MAMPs by pattern recognition receptors (PRRs): the corresponding genes are transcriptionally up-regulated following specific perception of the bacterial flagellin epitope flg22 by the corresponding FLAGELLIN SENSITIVE2 (FLS2) PRR (Clay et al., 2009). CYP81F2 is indispensable for effective A. thaliana resistance responses, as indicated by the fact that cyp81F2 plants have phenotypes with enhanced susceptibility to various fungal/oomycete pathogens and a defect in callose deposition (Bednarek et al., 2009; Clay et al., 2009; Hiruma et al., 2010; Sanchez-Vallet et al., 2010). Genetic evidence suggests that, although PEN2 processes various IGs, only the hydrolysis of CYP81F2-derived substrates leads to products that are critical for resistance responses to nonadapted powdery mildew fungi or MAMP-triggered callose deposition (Bednarek et al., 2009; Clay et al., 2009).

Although pathogen inducibility is an important regulatory feature of A. thaliana glucosinolate metabolism, it is not known when this property emerged within the Brassicaceae family. Additionally, it is not known whether pathogen-triggered IG metabolism leading to amines and RA is specific to A. thaliana or whether this pathway is functionally conserved in A. thaliana relatives. Some Brassica and Sinapis spp. as well as Raphanus sativus, Wasabia japonica, Thlaspi arvense and Thellungilla halophila have been reported to accumulate indole-type sulphur-containing metabolites known as Brassicaceae phytoalexins upon pathogen challenge or abiotic stress (for a review, see Pedras & Yaya, 2010). Similar to IGs, some of those metabolites possess a 1- or 4-methoxy substitution on their indole core. Experiments in which plants were fed with labeled precursors suggested that at least some of the Brassicaceae phytoalexins could be derived from IGs via the respective isothiocyanates formed after glucosinolate hydrolysis (Monde et al., 1994; Pedras et al., 2010). This putative biosynthetic link is supported by structural overlaps of some of the phytoalexins with a presumed dithiocarbamate adduct formed during pathogen-triggered IG metabolism (Bednarek et al., 2009).

It remains unclear whether A. thaliana produces any of the phytoalexins found in tested Brassicaceae crops. The only phytoalexin reported in significant quantities in A. thaliana is camalexin (Fig. 1), which was also found to accumulate after pathogen challenge in close relatives including Camelina sativa (Browne et al., 1991), Capsella bursa-pastoris (Jimenez et al., 1997) and Arabidopsis lyrata (Zook et al., 1998), but not in any other Brassicaceae species examined for phytoalexin accumulation (Pedras & Yaya, 2010). However, these species represent only a narrow subclade of the Brassicaceae family (Clauss & Koch, 2006; Franzke et al., 2011). Moreover, in most reported cases the presence of camalexin was not investigated in a targeted manner. For these reasons, it is not clear if camalexin is restricted to a narrow phylogenetic clade, including A. thaliana, or is more ubiquitous throughout the Brassicaceae. A contribution of camalexin to A. thaliana resistance responses was confirmed by enhanced susceptibility of camalexin-deficient phytoalexin deficient3 (pad3) mutant plants to a subset of tested microbial pathogens (Thomma et al., 1999; Ferrari et al., 2003; Bednarek et al., 2009; Schlaeppi et al., 2010). Camalexin and IG biosyntheses share only the first catalytic step, the conversion of tryptophan to indole-3-acetaldoxime, which is catalyzed by two genetically redundant P450 monooxygenases, CYP79B2 and CYP79B3 (Fig. 1) (Zhao et al., 2002; Glawischnig et al., 2004). Unlike other Brassicaceae phytoalexins and IGs, camalexin was never found to be substituted with a methoxy group at position 1 or 4. In planta reported substitutions of camalexin include 1-methylation and 6-methoxylation (Browne et al., 1991; Jimenez et al., 1997). This unique substitution pattern suggests that camalexin could constitute a metabolite class that is atypical in comparison to other reported Brassicaceae phytoalexins.

Besides camalexin, A. thaliana accumulates indole-3-carboxylic acid (I3CA) and its derivatives in a pathogen-inducible manner (Fig. 1) (Hagemeier et al., 2001; Tan et al., 2004; Bednarek et al., 2005; Forcat et al., 2010; Sanchez-Vallet et al., 2010). Identified I3CA derivatives include 1-methyl-I3CA and the glucoside of 6-hydroxy-I3CA (6OGlcI3CA), while 1- or 4-methoxylated I3CA structure types have never been reported. This substitution pattern indicates a closer biosynthetic link of I3CAs with camalexin and excludes the possibility that these metabolites are derived from IGs. I3CAs are derived from tryptophan through indole-3-acetaldoxime, as indicated by their absence in the cyp79B2 cyp79B3 knockout line (Bottcher et al., 2009; Sanchez-Vallet et al., 2010). Additionally, precursor feeding experiments suggests that the biosynthesis of I3CAs may also share with camalexin the next intermediate, indole-3-acetonitrile (Fig. 1) (Bottcher et al., 2009). Although the phylogenetic conservation of I3CAs has not been systematically studied, the metyl ester of 1-methyl-I3CA was reported to accumulate in C. sativa roots infected with Rhizoctonia solani (Conn et al., 1994).

In this study, we investigated the conservation and diversification of pathogen-inducible tryptophan-derived metabolism in close and distant A. thaliana relatives by metabolic profiling. We validated the observed species-specific metabolic patterns by determining whether candidate orthologs of A. thalianaPEN2, CYP81F2, PEN3 and PAD3 were present or absent in accessible genomes of A. thaliana relatives. Our metabolic survey revealed surprising conservation of the pathogen-triggered IG metabolic pathway among the tested plant species, suggesting an ancient and important function of this metabolic branch in Brassicaceae defense responses. By contrast, I3CA and camalexin biosyntheses appear to be clade-specific innovations within the conserved framework of pathogen-inducible tryptophan metabolism and represent relatively recent manifestations of the plant–pathogen arms race.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Plant lines, growth conditions and pathogen inoculation

Seeds of Arabidopsis lyrata (L.) O’Kane & Al-Shehbaz (line Mn47), Olimarabidopsis pumila (Stephan) Al-Shehbaz, O’Kane & R.A. Price, Olimarabidopsis cabulica (Hook.f. & Thomson) Al-Shehbaz, O'Kane & R.A. Price, Capsella rubella Reut., Crucihimalaya lasiocarpa (Hook.f. & Thomson) Al-Shehbaz, O'Kane & R.A. Price, Sisymbrium irio L. and Thellungiella halophila (C.A. Mey.) O.E. Schulz were obtained from the Nottingham Arabidopsis Stock Center (stocks N9608, N3700, N4653, N22697, N6191, N22563 and N22504, respectively). Cardamine hirsuta L. and Arabis alpina L. seeds were kind gifts of Dr Miltos Tsiantis and Dr Maria Albani, respectively. Arabidopsis halleri (L.) O'Kane & Al-Shehbaz plants (accession Langelsheim-5) (Talke et al., 2006) were received from Dr Ute Kramer. This species was propagated by cloning using stem cuttings and Rhizopon AA (Hortus, NY, USA) rooting hormone mixture. Seeds of Arabidopsis thaliana (L.) Heynh. ecotypes (An-1, Bay, Cvi, Eri, Fei-0, Kas-2, Kond, Kyo-1, Ler, Nos and Sha) were a gift of Dr Matthieu Reymond. Plants were grown in growth chambers at 20–23°C with a 10-h photoperiod and a light intensity of c. 150 μmol m−2 s−1. Four- to five-week-old plants were inoculated with Blumeria graminis f. sp. hordei (isolate K1) or Erysiphe pisi (Birmingham isolate) using a settling tower. Arabidopsis halleri plants were tested 3–4 wk after cloning. For Botritis cinerea infections, isolate 2100 (also designated B. cinerea 1.29) from the Spanish Type Culture Collection was used. Plants were spray-inoculated with a spore solution at a concentration of 105 spores ml−1.

Extraction of secondary metabolites and HPLC analysis

Pathogen-inoculated and noninoculated rosette leaves were collected at the same time-point (B. g. hordei at 16 h and B. cinerea at 48 h after inoculation with spores) and frozen in liquid nitrogen. After the addition of DMSO (50 μl per 20 mg FW), the tissue was homogenized with zirconia beads (1 mm; Roth, Karlsruhe, Germany) in a Mini-Beadbeater-8 (Biospec Products, Bartlesville, OK, USA) and centrifuged for 15 min at 20 000 g. The supernatants were collected and subjected to HPLC on an Agilent (Palo Alto, CA, USA) 1100 HPLC system equipped with Diode Array (DAD) and Fluorescence (FLD) detectors. Samples were analyzed first on an Atlantis T3 C18 column (150 × 2.1 mm, 3 μm; Waters, Milford, MA, USA) with 0.1% trifluoroacetic acid as solvent A and 98% acetonitrile/0.1% trifluoroacetic acid as solvent B at a flow rate of 0.25 ml min−1 at 22°C (gradient of solvent A: 100% at 0, 100% at 2, 90% at 9, 72% at 30, 50% at 33, 20% at 40 and 100% at 41 min). If any of the peaks of interest were not separated well enough for quantification, respective samples were also analyzed under different conditions: Zorbax SB-Aq column (150 × 2.1 mm, 3.5 μm; Agilent) at a flow rate of 0.25 ml min−1 at 22°C (gradient of solvent A: 100% at 0, 100% at 2, 90% at 9, 84% at 16 and 100% at 17 min), and/or Synergi Polar-RP column (150 × 2.1 mm, 3.5 μm; Phenomenex, Torrance, CA, USA) at a flow rate of 0.25 ml min−1 at 22°C (gradient of solvent A: 100% at 0, 100% at 5, 84% at 19 and 100% at 21 min).

Camalexin was analyzed on a Zorbax Extend-C18 column (100 × 2.1 mm, 3.5μm; Agilent) with water as solvent A and 98% acetonitrile as solvent B at a flow rate of 0.3 ml min−1 at 22°C (gradient of solvent A: 96% at 0, 96% at 3, 70% at 20, 20% at 33 and 0% at 34 min). IG, I3A, 6OGlcI3CA and I3CA glucose ester (I3CAGlc) peaks were identified by comparing their retention times and spectral properties with those of compounds purified from plant tissue (Bednarek et al., 2005, 2009). RA and camalexin peaks were identified by referring to commercial (Fluka, Buchs, Switzerland) and synthetic (Bednarek et al., 2009) standards, respectively. The concentrations of the metabolites of interest were quantified on the basis of the comparison of their peak areas with those obtained during HPLC analyses of known amounts of the respective standards. 1MI3G was quantified as equivalents of I3G and I3CAGlc as equivalents of I3CA (Sigma-Aldrich, St Louis, MO, USA). The following chromatograms were used for quantifications: IGs, RA and I3CAs, UV absorbtion at 273 nm; I3A, fluorescence (ex. 275 nm; em. 350 nm); and camalexin, fluorescence (ex. 318 nm; em. 386 nm). Presented data are mean values ± SD from analyses of three samples prepared from independent plants. Each species was tested in at least three independent experiments, which gave similar results for all presented compounds.

LC-MS analysis

The LC-MS system consisted of the Ultimate 3000 series RSLC (Dionex, Sunnyvale, CA, USA) system and the Orbitrap XL mass spectrometer (Thermo Fisher Scientific, Bremen, Germany). HPLC was performed using the Acclaim C18 Column (150 × 2.1 mm, 2.2 μm; Dionex) at a constant flow rate of 300 μl min−1 using a binary solvent system: solvent A contains water with 0.1% formic acid and solvent B contains acetonitrile with 0.1% formic acid. The HPLC gradient system started with 3% B and linearly increased to 65% B in 18 min and to 95% in 21 min and was then held for 5 min, before being brought back to the initial conditions and held for 5 min for re-equilibration of the column for the next injection. All the samples were measured in positive mode using the Orbitrap analyzer and in negative mode using the ion-trap analyzer. Collision induction experiments were performed by selecting the precursor ions in the ion trap, colliding them with nitrogen gas in a higher-energy collisional dissociation (HCD) trap using 70 V collision energy and detecting the product ions in the Orbitrap analyzer at a resolution of 7500. All data analyses were performed using xcalibur (Thermo Fisher Scientific, Waltham, MA, USA) software.

Protein electrophoresis and immunodetection of PEN2

Protein extractions and western blot analysis were performed as described previously (Lipka et al., 2005). The rabbit polyclonal AtPEN2 antiserum was used for protein blots at a dilution of 1 : 3000.

Identification of PEN2, PEN3, CYP81F2 and PAD3 orthologs in Brassicaceae plants

To identify putative orthologs of PEN2, PEN3 CYP81F2 and PAD3 we used available genomic sequences of A. lyrata (http://www.phytozome.net/), Brassica rapa (http://brassicadb.org/brad/), C. rubella (http://trace.ncbi.nlm.nih.gov/Traces/sra/sra.cgi, http://www.jgi.doe.gov/sequencing/why/3066.html) and A. alpina (shared by G. Coupland & K. Nordström). The sequence reads of C. rubella were assembled using newbler (version 2.1; 454 Life Sciences, Branford, CT, USA) resulting in a genome assembly of 132 Mb (c. 22× coverage) with an estimated gene space coverage of 98% (Parra et al., 2009).

Genomic sequences of putative PEN2, PEN3,CYP81F2 and PAD3 orthologs were obtained using tblastn (Altschul et al., 1997). To predict the open reading frames (ORFs) from the genomic candidate regions, Fgenesh (http://www.softberry.com) was used with parameters optimized for A. thaliana. The predicted proteins were aligned using ClustalW and visualized using Jalview (Waterhouse et al., 2009). A phylogenetic tree (Kimura distance) was calculated using MegAlign (DNASTAR, Madison, WI, USA). Sequences of A. alpina PEN2, PEN3 and CYP81F2 orthologs have been submitted to GenBank (accession numbers JF792628JF792630).

Microscopic analysis of E. pisi inoculated leaves

Leaves were fixed and cleared, entry rates were scored, and secondary hyphae were visualized, as described previously (Lipka et al., 2005). The presented entry rates are mean values ± SD from five independent measurements. Each line was tested in three independent experiments, which gave similar results.

Flg22 treatments and callose deposition assay

Arabidopsis thaliana and C. rubella seedlings were grown in 24-well plates in liquid medium, treated with flg22, collected and used for callose visualization as described previously (Clay et al., 2009).

Results

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Metabolic diversification of UV-absorbing metabolites in the Brassicaceae

To investigate the evolutionary pattern of pathogen-inducible tryptophan metabolism, we performed an HPLC analysis of leaf extracts from select species of the Brassicaceae. We included in our survey A. thaliana relatives whose size and developmental stage 5 wk after sowing are comparable and for which corresponding genome sequencing projects are either ongoing or planned to facilitate genome-wide studies on secondary metabolite pathway diversification. We intentionally added species of the Camelineae tribe to elucidate the origin and conservation of camalexin biosynthesis. Altogether, lines representing 11 plant species were used in the pathogen-inducible metabolite survey: A. thaliana, A. lyrata, A. halleri, O. pumila, O. cabulica, C. rubella, C. lasiocarpa, C. hirsuta, S. irio, T. halophila and A. alpina (Fig. 2). These species represents lineages I and II of the Brassicaceae family (Couvreur et al., 2010; Franzke et al., 2011). In addition, for the identification of orthologs of the genes of interest, we used the genomic sequence of B. rapa, a species belonging to the same phylogenic clade as S. irio (Clauss & Koch, 2006; Couvreur et al., 2010) (Fig. 2).

image

Figure 2. Phylogenetic relationships between Brassicaceae species analyzed in this study. The relationships, with the exception of those with Thellungiella halophila, are based on the sequences of coding plastidic maturase K (matK) and nuclear chalcone synthase (Chs) sequences (Koch et al., 2001; Clauss & Koch, 2006). The phylogenetic position of T. halophila was estimated (dashed lines) based on a different study of Chs sequences (Zhao et al., 2010). The Camelinae tribe is defined according to Couvreur et al. (2010).

Download figure to PowerPoint

We found a huge diversification of UV-absorbing and fluorescent leaf compounds on comparing the HPLC chromatograms obtained from this collection of healthy plants (Supporting Information Fig. S1). However, the metabolite profiles derived from close relatives such as A. lyrata and A. halleri or for the two tested Olimarabidopsis spp. were largely overlapping. By contrast, the profiles differed markedly between A. thaliana and two other members belonging to the same genus (Fig. S1), indicating the potential for natural metabolite variation even between close relatives. The UV absorption and fluorescence spectra revealed that many of the peaks representing diversified compounds belong to phenylpropanoids, including flavonols and sinapates (not shown). This suggests that a significant part of the observed metabolic diversification between Brassicaceae plants (Fig. S1) results from variation in phenylalanine-derived metabolites. Unlike tryptophan-derived small molecules, their contribution to A. thaliana immune responses is essentially unknown.

Powdery mildew-induced IGs and their metabolic products in Brassicaceae plants

The generation of 4-substituted IGs by the P450 monooxygenase CYP81F2 represents an important step for the microbe-induced metabolic shift of IG metabolism in A. thaliana accession Col-0 leaves. This was demonstrated by an elevated accumulation of 4MI3G, which is particularly pronounced in the pen2 background (Bednarek et al., 2009; Clay et al., 2009; Sanchez-Vallet et al., 2010). To determine whether this metabolic shift is unique for Col-0, we examined the accumulation of 4MI3G after inoculation with a nonadapted pathogen of dicotyledonous Brassicaceae plants, the grass powdery mildew B. g. f. sp. hordei, in 11 A. thaliana accessions. We found that all except ecotype Bay showed increased 4MI3G concentrations upon pathogen inoculation and the fold induction in these ecotypes was even higher than in Col-0 (Fig. S2). To determine whether the metabolic shift to 4MI3G is also conserved among the Brassicaceae, we inspected the accumulation of IGs in the selected species after inoculation with B. g. hordei. We were able to detect and quantify 4MI3G in all tested species except C. rubella (Fig. 3). Elevated concentrations of this compound were found 16 h after B. g. hordei inoculation in most of the tested species, except A. lyrata, C. lasiocarpa and T. halophila (Fig. 3). We also examined the accumulation of two other IGs, I3G and 1MI3G. I3G was found in almost all tested species, although in some of them it was barely above the detection limit (Fig. 3). By contrast, 1MI3G was detected in leaf samples of only four of the tested species (Fig. 3). In addition, changes in the accumulation of I3G and 1MI3G after B. g. hordei inoculation did not reveal a phylogenetically conserved trend (Fig. 3). We note that these two metabolites, like 4MI3G, were also undetectable in C. rubella, indicating that the tested line lacks any IGs.

image

Figure 3. Accumulation of indole glucosinolates (IGs) and related metabolites in tested Brassicaceae plants 16 h after inoculation with Blumeria graminis f. sp. hordei conidiospores (closed bars) (control, open bars). Error bars indicate ± SD.

Download figure to PowerPoint

Pathogen-triggered and PEN2-dependent IG hydrolysis can be monitored in A. thaliana leaves by measuring the accumulation of I3A and RA (Fig. 1) (Bednarek et al., 2009; Sanchez-Vallet et al., 2010). In extracts from most of the analyzed species, I3A was either undetectable or was found in amounts slightly above the detection limit (Fig. 3). As I3A is a product of I3G metabolism (Fig. 1) it is not surprising that plants that do not accumulate I3G do not produce a nonnegligible amount of I3A after pathogen challenge. In accordance with this metabolic link, we observed a strong induction of I3A accumulation after B. g. hordei inoculation in A. thaliana and A. lyrata, which represent species accumulating the highest amounts of I3G (Fig. 3). However, I3A was only weakly, if at all, inducible in the two other I3G-accumulating plants, C. hirsuta and S. irio, suggesting that constitutive I3G biosynthesis and I3A inducibility are not a conserved metabolic signature of the Brassicaceae (Fig. 3). While I3A is a product derived exclusively from I3G, RA can be, at least in theory, formed during metabolism of any glucosinolate (Bednarek et al., 2009). Consequently, RA can be considered a general marker for the metabolism of different glucosinolate classes. Similar to 4MI3G, we detected RA in leaf extracts of all tested plants except C. rubella. RA accumulation was additionally responsive to B. g. hordei inoculation in most tested plants except A. lyrata and T. halophila (Fig. 3), suggesting that the pathogen-triggered glucosinolate metabolism leading to amines and RA is largely conserved among the Brassicaceae.

PEN2, PEN3 and CYP81F2 orthologs in Brassicaceae

Pathogen-inducible IG hydrolysis in leaves is mediated mainly or even exclusively by the peroxisome-associated PEN2 myrosinase in A. thaliana (Bednarek et al., 2009). This enzyme possesses a unique C-terminal extension, which was shown to be critical for its in vivo function (Lipka et al., 2005). Available genomic and expressed sequence tag (EST) sequences of putative PEN2 orthologs from A. lyrata, A. alpina and B. rapa indicate a very high level of sequence conservation of this extension among these three species and A. thaliana (Fig. 4a). An invariant stretch within the C-terminal extension comprises a 16-amino acid peptide used for the generation of a PEN2-specific antibody (Lipka et al., 2005). We identified a putative ortholog of AtBGLU27, the most closely related AtPEN2 paralog, but not of AtPEN2, in available C. rubella genomic sequences (Fig. 4b). With the PEN2 antibody, we detected on western blots signals of indistinguishable size in protein extracts from leaves of A. lyrata, A. halleri, C. hirsuta, T. halophila and A. alpina (Fig. 4c). The presence of putative PEN2 orthologs in A. alpina and T. halophila, which are phylogenetically relatively distant from A. thaliana, suggests that PEN2 emerged early in the evolution of the Brassicaceae. The apparent conservation of the epitope points to an important contribution of the corresponding peptide sequence to the function of the PEN2 myrosinase.

image

Figure 4. PENETRATION2 (PEN2) orthologs. (a) Multiple alignment of the C-terminal fragment of AtPEN2 with its putative orthologs from Arabidopsis lyrata, Brassica rapa, Arabis alpina and Capsella rubella. Green line, terminal fragment of PEN2 sequence conserved with other β-glucosidases; red line, peptide sequence used to generate the PEN2-specific antibody; blue line, predicted helical region critical for PEN2 in vivo function. (b) Phylogenetic tree of PEN2 alignments using the Kimura distance. The table shows the best blastp hit of putative PEN2 orthologs against all Arabidopsis thaliana proteins. (c) Immunodetection of putative PEN2 orthologs in leaf protein extracts from tested Brassicaceae plants using an AtPEN2 antibody that recognizes an epitope of 16 amino acids within the C-terminal extension of the enzyme (Lipka et al., 2005).

Download figure to PowerPoint

We searched for PEN3 and CYP81F2 orthologs in available genomic sequences of Brassicaceae species. We identified sequences encoding putative PEN3 ABC transporter orthologs in the tested genomes of A. lyrata, C. rubella, B. rapa and A. alpina (Fig. S3). Strikingly, our genome analysis did not reveal any evidence for CYP81F2 orthologs in C. rubella, whereas a putative ortholog of A. thaliana CYP81F1, the most closely related paralog of AtCYP81F2, was identified (Fig. S4). Thus, C. rubella appears to lack orthologs of genes needed in A. thaliana for the biosynthesis (CYP81F2) and hydrolysis (PEN2) of the pathogen-inducible IG 4MI3G.

Pathogen-triggered biosynthesis of camalexin and I3CAs

We detected a significant induction of camalexin after B. g. hordei inoculation only in the two tested Olimarabidosis spp. (Fig. 5a). As reported previously (Bednarek et al., 2009), camalexin in A. thaliana was barely inducible following inoculation with this nonadapted powdery mildew. Low concentrations of the phytoalexin were recorded in other tested Camelineae plants, but not in any of the other species (Fig. 5a). In addition, we found that both Olimarabidosis spp. and C. rubella accumulated an additional metabolite with UV and fluorescence emission spectra almost identical to those of camalexin. Detailed LC/MS analysis revealed the structure of this compound as methoxycamalexin (Figs 5a, S5). Based on the LC/MS results, we were only able to confine the position of the methoxyl group to the aromatic ring of camalexin. We assume that this compound is 6-methoxycamalexin as this is the only methoxylated derivative of camalexin reported to date in two other members of the Camelinae tribe (Browne et al., 1991; Jimenez et al., 1997).

image

Figure 5. Accumulation of camalexin and 6-methoxycamalexin in Brassicaceae plants (a) 16 h after inoculation with Blumeria graminis f. sp. hordei conidiospores (closed bars), and (b) 40 h after inoculation with Botrytis cinerea conidiospores (closed bars) (control, open bars). Error bars indicate ± SD.

Download figure to PowerPoint

As camalexin concentrations barely exceeded the detection limit in Camelineae plants inoculated with the biotrophic B. g. hordei fungus (Fig. 5a), we performed additional inoculation experiments using the necrotrophic fungus B. cinerea. This ascomycete pathogen was found to induce in A. thaliana the biosynthesis and accumulation of high amounts of camalexin (Kliebenstein et al., 2005). A strong induction of camalexin accumulation was found in A. thaliana, A. lyrata, A. halleri, O. pumila, O. cabulica and C. rubella upon B. cinerea inoculation (Fig. 5b), but not in the other tested species. This observation substantiates previous assumptions that the capacity for camalexin biosynthesis might be restricted to a set of plants belonging to the Camelineae tribe. A genome-wide analysis which revealed PAD3 orthologs in A. lyrata and C. rubella, but not in the genomes of B. rapa and A. alpina, lent support to this metabolite-based conclusion (Fig. S6). A clear accumulation of 6-methoxycamalexin was recorded only in infected Olimarabidosis spp. and C. rubella (Fig. 5b), pointing to an extremely narrow phylogenetic distribution of this metabolite.

I3CAs (Fig. 1) constitute another class of pathogen-inducible metabolites in A. thaliana whose occurrence within the Brassicaceae plant family has not been studied before. Of these metabolites, 6OGlcI3CA usually accumulates constitutively in A. thaliana. We found constitutive accumulation of this metabolite in the same species that also accumulated camalexin and, additionally, in C. hirsuta. Analysis of leaf samples from pathogen-inoculated plants revealed that 6OGlcI3CA was only moderately, if at all, induced after B. g. hordei inoculation (not shown), whereas inoculation with B. cinerea led to a large increase in 6OGlcI3CA biosynthesis (Fig. 6), but only in plants that accumulate this compound constitutively. The same plant species also accumulated I3CAGlc after B. cinerea challenge. I3CAGlc was additionally detected in C. lasiocarpa (Fig. 6), indicating that this species possesses the ability to synthesize I3CA, but not its derivatives substituted at position 6 of the indole core.

image

Figure 6. Accumulation of 6OGlcI3CA and I3CAGlc in tested Brassicaceae plants 40 h after inoculation with Botrytis cinerea conidiospores (closed bars) (control, open bars). Error bars indicate ± SD.

Download figure to PowerPoint

IG metabolism after infection with B. cinerea

It has been often postulated that the colonization of Brassicaceae plants by necrotrophic pathogens leads to the activation of a cell damage-dependent glucosinolate hydrolysis, thereby generating the same products found in interactions between plants and chewing insects (Tierens et al., 2001). However, colonization of A. thaliana by the necrotrophic fungus Plectosphaerella cucumerina triggers PEN2-mediated IG metabolism leading to the same products as those observed after challenge with biotrophic pathogens (Sanchez-Vallet et al., 2010). To obtain better insights into IG metabolism in Brassicaceae plants colonized by necrotrophic fungi, we quantified IG, I3A and RA concentrations in leaf samples at 40 h after spore inoculation with B. cinerea (Figs 7, S7). We found that this necrotrophic fungus induced qualitatively similar changes in IG metabolism as B. g. hordei, but the respective metabolic shifts were more pronounced (compare Figs 3 and 7). For example, 4MI3G accumulated to higher concentrations upon B. cinerea challenge compared with B. g. hordei inoculation. Moreover, the B. cinerea-induced biosyntheses of this compound was observed in all tested IG-producing species (Fig. 7). Similarly, changes in RA concentrations were greater and significant induction of this metabolite was seen in A. lyrata upon colonization with B. cinerea, which in this plant species was not seen upon B. g. hordei inoculation (Figs 3, 7). The B. cinerea-induced accumulation of I3G and 1MI3G revealed no apparent conserved phylogenetic pattern amongst the tested plant species, which is consistent with the distribution of these compounds after inoculation with B. g. hordei (Figs 3, S7). Taken together, these findings indicate that necrotrophic pathogens induce in Brassicaceae plants the same changes in IG metabolism as obligate biotrophs, including a shift of IG biosynthesis to 4-substituted derivatives and IG conversion to products different from those observed after host tissue destruction.

image

Figure 7. Accumulation of indole glucosinolates (IGs) and related metabolites in Brassicaceae plants 40 h after inoculation with Botrytis cinerea conidiospores (closed bars) (control, open bars). Error bars indicate ± SD.

Download figure to PowerPoint

Effect of IG deficiency on C. rubella innate immunity

The complete lack of IGs in the tested C. rubella line raised the question of whether this accession fails to induce MAMP-induced defense responses. In A. thaliana, treatment of seedlings with the bacterial MAMP flagellin (flg22) induces extracellular callose deposition and this response is completely dependent on CYP81F2/PEN2-derived IG metabolism products (Clay et al., 2009). Similarly, pre-invasive resistance responses to the nonadapted powdery mildew fungus E. pisi, which colonizes (Pisum sativum) pea plants in nature, are partially dependent on these IG metabolism products (Lipka et al., 2005; Bednarek et al., 2009). We found that the entry rates of E. pisi into leaf epidermal cells were slightly higher on C. rubella than on A. thaliana, but this difference was not statistically significant (Fig. 8a). Moreover, C. rubella seedlings were able to deposit callose after treatment with flg22 (Fig. 8b). This indicates that the tested C. rubella accession employs an IG-independent signaling and/or execution pathway(s) for these extracellular defense outputs.

image

Figure 8. Comparison of defense responses of Arabidopsis thaliana and Capsella rubella. (a) Frequency of invasive growth of Erysiphe pisi sporelings 72 h after inoculation with the fungal conidiospores. (b) Callose deposition 24 h after flg22 treatment in cotyledons, visualized by aniline blue staining and subsequent fluorescence microscopy. Error bars indicate ± SD.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Here, we examined phylogenetic patterns of constitutive and pathogen-inducible IGs in the Brassicaceae. Our results clearly point to a conservation of pathogen-inducible IG metabolism. This conservation of metabolic changes includes a shift of IG biosynthesis to 4MI3G and IG hydrolysis, probably mediated by PEN2-like enzymes, to products including RA (Figs 1, 3, 4, 7, S2). The predominance of the substitution of IGs at position 4 is especially striking in the context of the variation of constitutive IG concentrations reported not only between Brassicaceae species, but also between particular accessions within the same species (Kliebenstein et al., 2001; Windsor et al., 2005). The physiological significance of this substitution remains to be resolved, as the respective biologically active end products and their function(s) in plant resistance responses are still unknown (Bednarek et al., 2009; Clay et al., 2009). The pathogen-induced accumulation of RA was observed in all tested IG-producing plants (Figs 3, 7). This provides for the first time evidence that the pathway for IG metabolism in living cells is evolutionarily ancient and has been largely retained among the Brassicaceae. Finally, our data demonstrate that this tissue damage-independent pathway can be triggered by pathogens employing a biotrophic or a necrotrophic mode of plant colonization (Figs 3, 7).

The detection of putative PEN2 orthologs in at least five of the tested A. thaliana relatives (Fig. 4) suggests that in these Brassicaceae plants pathogen-triggered IG metabolism is also initiated by PEN2-like myrosinases. We were unable to detect any signals with the PEN2-specific antibody in protein extracts from O. lasiocarpa, O. cabulica, C. lasiocarpa and S. irio (Fig. 4), but this does not allow us to conclude that PEN2 orthologs are not present in these species. The epitope recognized by the PEN2 antibody is conserved among A. thaliana, A. lyrata, A. alpina and B. rapa (Fig. 4a), but the possibility of amino acid exchanges cannot be excluded in other species. The PEN2 steady-state concentrations in A. thaliana and A. lyrata leaves differed dramatically (Fig. 4c), yet the epitope recognized by the PEN2 antibody in the putative A. lyrata PEN2 ortholog is conserved (Fig. 4). This finding points to high variation in the abundance of PEN2 orthologs between particular Brassicaceae species, which could contribute to the lack of PEN2 immunosignals in some of the tested A. thaliana relatives.

The observed phylogenetic conservation of pathogen-inducible IG metabolism in the Brassicaceae is important to extend the previously proposed metabolic link between the biosynthetic pathways for IGs and some of the Brassicaceae phytoalexins (Monde et al., 1994; Bednarek et al., 2009). The conserved shift of IG biosynthesis to 4MI3G predicts that, if such a metabolic link exists, then 4-substituted compounds should be overrepresented among Brassicaceae phytoalexins. However, only seven out of 44 reported ‘noncamalexin’-type Brassicaceae phytoalexins are substituted at position 4 (Pedras & Yaya, 2010). However, these data are based mainly on phytoalexins from very closely related Brassicaceae crop plants (Pedras & Yaya, 2010). For this reason, there is a need to reassess phytoalexin diversity among members of the Brassicaceae plant family on the basis of phylogenetic relatedness.

The only exception to the conserved changes in IG metabolism is the tested C. rubella line, which is naturally depleted of any IGs. It is not clear at present whether the lack of IG biosynthesis is limited to this C. rubella accession or is characteristic for this species or even a higher taxonomic rank. Indeed, the tested accessions of a closely related species, C. bursa-pastoris, were found to be either devoid of or containing low concentrations of glucosinolates (Ockendon & Buczacki, 1979; Griffiths et al., 2001; Chen et al., 2007). The IG biosynthetic pathway in A. thaliana is interconnected with the biosynthesis of indole-3-acetic acid (Bak et al., 2001; Zhao et al., 2002). Loss-of-function mutations in genes encoding a subset of enzymes involved in IG biosynthesis lead to severe growth defects as a consequence of auxin overproduction (Bak et al., 2001; Mikkelsen et al., 2004; Morant et al., 2010). For this reason, the loss of the IG biosynthetic pathway is unlikely to be caused by the disappearance of a single gene, as this would additionally require significant changes in the regulation of tryptophan metabolism and auxin biosynthesis.

Given the known functions of IGs in A. thaliana innate immunity (Bednarek et al., 2009; Clay et al., 2009), it is noteworthy that the tested C. rubella plants retain the capacity to deposit callose after flg22 treatment and efficiently restrict the entry of a nonadapted powdery mildew fungus (Fig. 8). Pre-invasive resistance to nonadapted powdery mildews in A. thaliana is mediated by at least two parallel secretory pathways: one utilizes CYP81F2/PEN2-dependent IG metabolism and the PEN3 ABC transporter for extrusion of the respective end products (Stein et al., 2006). The second pathway requires a set of soluble N-ethylmaleimide sensitive factor attachment protein receptor (SNARE) domain-containing proteins, including PEN1 syntaxin (Collins et al., 2003; Kwon et al., 2008). It is possible that the loss of the biosynthetic capacity for IGs in C. rubella was compensated by a novel pre-invasive resistance mechanism. Unlike the parallel pre-invasive defense pathways restricting entry of powdery mildews into plant cells, the end product(s) of CYP81F2/PEN2-dependent IG metabolism are thought to be indispensable for flg22-inducible extracellular callose deposition in A. thaliana (Clay et al., 2009). If this is the case, then either this absolute requirement in A. thaliana is not conserved among the Brassicaceae or C. rubella has invented an alternative signaling pathway to induce callose deposition upon MAMP perception.

Available C. rubella genomic sequences (corresponding to the line tested in this study) lack candidates for PEN2 orthologs (Fig. 4), but reveal a putative PEN3 ortholog (Fig. S3). The PEN3 ABC transporter in A. thaliana has, in addition to its role in innate immune responses, additional physiological functions (Kim et al., 2007; Strader & Bartel, 2009; Ruzicka et al., 2010), which could explain the conservation of PEN3 in C. rubella. It would be interesting to determine whether the presumed C. rubella ortholog still contributes to the observed pre-invasive disease resistance (Fig. 8a) by secreting other types of biologically active molecules. Unlike PEN3, the reported biochemical function of CYP81F2 is thought to be confined to IG biosynthesis only (Pfalz et al., 2009). Our analysis did not reveal CYP81F2 orthologs in C. rubella, whereas a putative ortholog of A. thaliana CYP81F1 was identified (Fig. S4). The gene product of AtCYP81F1 is thought to possess a similar biochemical activity to CYP81F2. The metabolite phenotype of cyp81F1 plants suggests a minor contribution of this P450 monooxygenase to IG biosynthesis in A. thaliana (Pfalz et al., 2011). Thus, the apparent lack of both CYP81F2 and PEN2 orthologs in the C. rubella genome strongly corroborates the metabolic profiling data obtained from C. rubella leaves. Although it is tempting to speculate that the loss of PEN2 and CYP81F2 orthologs in C. rubella causally contributes to the inability of this plant to synthesize and metabolize IGs, it remains possible that the presumed gene losses occurred subsequent to other unknown primary genomic lesions disrupting IG biosynthesis. Future access to a fully annotated C. rubella genome is needed to better understand how this species could sacrifice its capacity to synthesize IGs without recognizable trade-offs in immunity or auxin-associated growth defects (Bak et al., 2001: Mikkelsen et al., 2004; Morant et al., 2010).

In contrast to IGs and related downstream products, other known pathogen-inducible and tryptophan-derived A. thaliana metabolites seem to be restricted to particular phylogenetic lineages within the Brassicaceae. All tested species of the Camelineae tribe as well as C. lasiocarpa and C. hirsuta accumulated I3CAs (Fig. 6), while camalexin accumulation was confined to Camelineae only (Fig. 5). This phylogenetic restriction of the camalexin biosynthetic pathway to the tested members of the Camelinae tribe was additionally supported by the lack of PAD3 orthologs in B. rapa and A. alpina genomes (Fig. S6). This could suggest that the I3CA pathway is evolutionarily ancient while camalexin biosynthesis evolved later from the first intermediate in I3CA biosynthesis (Fig. 1) (Bottcher et al., 2009). All I3CA-producing species, except C. lasiocarpa, possess the ability to accumulate 6-substituted I3CA, but only three closely related Camelineae species were found to produce 6-methoxycamalexin. Low amounts of sugar conjugates of hydroxycamalexin were detected before in silver nitrate-challenged and Phytophthora infestans-inoculated A. thaliana leaves (Bottcher et al., 2009). This suggests that all Camelineae may have the ability to produce trace amounts of 6-methoxy or hydroxycamalexin, possibly by the I3CA substituting enzyme(s), but only in a subclade of this tribe enzyme(s) with camalexin substrate specificity evolved.

Our work suggests that IGs and their metabolism products, including some of the Brassicaceae phytoalexins, constitute the most ancient group of Brassicaceae tryptophan-derived metabolites involved in inducible plant immune responses. The integration of tryptophan metabolism into plant disease resistance mechanisms certainly required the coordination of the biosynthesis of this amino acid with other defense responses. The resulting embedment of the core pathway for tryptophan biosynthesis and metabolism into the network of MAMP-controlled processes (Clay et al., 2009) probably facilitated the evolution of metabolic sub-branches leading to novel products that appeared as a result of a plant–pathogen arms race. I3CAs, camalexin and 6-methoxycamalexin, suggest the existence of many other undiscovered pathogen-inducible and tryptophan-derived metabolites that may be present only in particular phylogenetic lineages within the Brassicaceae plant family. It is likely that such new defense-related metabolites, combined with novel resistance mechanisms, allowed some Brassicaceae plant species to lose their IG biosynthetic capacity without weakening their overall chemical defense arsenal. Future insights into the genome sequences of Brassicaceae plants will be critical to reconstruct the evolution of pathogen-inducible tryptophan metabolic pathways and their functions in innate immunity at a genome-wide scale.

Acknowledgements

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

We thank George Coupland and Karl Nordström for making the unpublished A. alpina genome sequence available for our analysis. We would like to acknowledge Christian Latza and Heidi Höck for their excellent technical assistance. This work was supported by the Bundesministerium für Bildung und Forschung (Plant-KBBE grant: BALANCE). P.B. was additionally supported by an EMBO Installation Grant.

References

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • Adie BAT, Perez-Perez J, Perez-Perez MM, Godoy M, Sanchez-Serrano JJ, Schmelz EA, Solano R. 2007. ABA is an essential signal for plant resistance to pathogens affecting JA biosynthesis and the activation of defenses in Arabidopsis. Plant Cell 19: 16651681.
  • Altschul SF, Madden TL, Schäffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Research 25: 33893402.
  • Bak S, Tax FE, Feldmann KA, Galbraith DW, Feyereisen R. 2001. CYP83B1, a cytochrome P450 at the metabolic branch point in auxin and indole glucosinolate biosynthesis in Arabidopsis. Plant Cell 13: 101111.
  • Bednarek P, Kwon C, Schulze-Lefert P. 2010. Not a peripheral issue: secretion in plant–microbe interactions. Current Opinion in Plant Biology 13: 378387.
  • Bednarek P, Osbourn A. 2009. Plant–microbe interactions: chemical diversity in plant defense. Science 324: 746748.
  • Bednarek P, Pislewska-Bednarek M, Svatos A, Schneider B, Doubsky J, Mansurova M, Humphry M, Consonni C, Panstruga R, Sanchez-Vallet A et al. 2009. A glucosinolate metabolism pathway in living plant cells mediates broad-spectrum antifungal defense. Science 323: 101106.
  • Bednarek P, Schneider B, Svatos A, Oldham NJ, Hahlbrock K. 2005. Structural complexity, differential response to infection, and tissue specificity of indolic and phenylpropanoid secondary metabolism in Arabidopsis roots. Plant Physiology 138: 10581070.
  • Bednarek P, Schulze-Lefert P. 2008. Role of plant secondary metabolites at the host–pathogen interface. In: Parker J, ed. Molecular aspects of plant disease resistance. Chichester, UK: Wiley-Blackwell, 220260.
  • Blau PA, Feeny P, Contardo L, Robson DS. 1978. Allylglucosinolate and herbivorous caterpillars: a contrast in toxicity and tolerance. Science 200: 12961298.
  • Bones AM, Rossiter JT. 2006. The enzymic and chemically induced decomposition of glucosinolates. Phytochemistry 67: 10531067.
  • Bottcher C, Westphal L, Schmotz C, Prade E, Scheel D, Glawischnig E. 2009. The multifunctional enzyme CYP71B15 (PHYTOALEXIN DEFICIENT3) converts cysteine-indole-3-acetonitrile to camalexin in the indole-3-acetonitrile metabolic network of Arabidopsis thaliana. Plant Cell 21: 18301845.
  • Brown PD, Tokuhisa JG, Reichelt M, Gershenzon J. 2003. Variation of glucosinolate accumulation among different organs and developmental stages of Arabidopsis thaliana. Phytochemistry 62: 471481.
  • Browne LM, Conn KL, Ayer WA, Tewari JP. 1991. The camalexins – new phytoalexins produced in the leaves of Camelina sativa (Cruciferae). Tetrahedron 47: 39093914.
  • Chen HF, Wang H, Li ZY. 2007. Production and genetic analysis of partial hybrids in intertribal crosses between Brassica species (B. rapa, B. napus) and Capsella bursa-pastoris. Plant Cell Reports 26: 17911800.
  • Clauss MJ, Koch MA. 2006. Poorly known relatives of Arabidopsis thaliana. Trends in Plant Science 11: 449459.
  • Clay NK, Adio AM, Denoux C, Jander G, Ausubel FM. 2009. Glucosinolate metabolites required for an Arabidopsis innate immune response. Science 323: 95101.
  • Collins NC, Thordal-Christensen H, Lipka V, Bau S, Kombrink E, Qiu JL, Huckelhoven R, Stein M, Freialdenhoven A, Somerville SC et al. 2003. SNARE-protein-mediated disease resistance at the plant cell wall. Nature 425: 973977.
  • Conn KL, Browne LM, Tewari JP, Ayer WA. 1994. Resistance to Rhizoctonia solani and presence of antimicrobial compounds in Camelina sativa roots. Journal of Plant Biochememistry and Biotechnology 3: 125130.
  • Consonni C, Bednarek P, Humphry M, Francocci F, Ferrari S, Harzen A, Ver Loren van Themaat E, Panstruga R. 2010. Tryptophan-derived metabolites are required for antifungal defense in the Arabidopsis mlo2 mutant. Plant Physiology 152: 15441561.
  • Couvreur TLP, Franzke A, Al-Shehbaz IA, Bakker FT, Koch MA, Mummenhoff K. 2010. Molecular phylogenetics, temporal diversification, and principles of evolution in the mustard family (Brassicaceae). Molecular Biology and Evolution 27: 5571.
  • Dixon RA. 2001. Natural products and plant disease resistance. Nature 411: 843847.
  • Dodds PN, Rathjen JP. 2010. Plant immunity: towards an integrated view of plant–pathogen interactions. Nature Reviews Genetics 11: 539548.
  • Ferrari S, Plotnikova JM, De Lorenzo G, Ausubel FM. 2003. Arabidopsis local resistance to Botrytis cinerea involves salicylic acid and camalexin and requires EDS4 and PAD2, but not SID2, EDS5 or PAD4. Plant Journal 35: 193205.
  • Forcat S, Bennett M, Grant M, Mansfield JW. 2010. Rapid linkage of indole carboxylic acid to the plant cell wall identified as a component of basal defence in Arabidopsis against hrp mutant bacteria. Phytochemistry 71: 870876.
  • Franzke A, Lysak MA, Al-Shehbaz IA, Koch MA, Mummenhoff K. 2011. Cabbage family affairs: the evolutionary history of Brassicaceae. Trends in Plant Science 16: 108116.
  • Glawischnig E, Hansen BG, Olsen CE, Halkier BA. 2004. Camalexin is synthesized from indole-3-acetaldoxime, a key branching point between primary and secondary metabolism in Arabidopsis. Proceedings of the National Academy of Sciences, USA 101: 82458250.
  • Griffiths DW, Deighton N, Birch ANE, Patrian B, Baur R, Städler E. 2001. Identification of glucosinolates on the leaf surface of plants from the Cruciferae and other closely related species. Phytochemistry 57: 693700.
  • Hagemeier J, Schneider B, Oldham NJ, Hahlbrock K. 2001. Accumulation of soluble and wall-bound indolic metabolites in Arabidopsis thaliana leaves infected with virulent or avirulent Pseudomonas syringae pathovar tomato strains. Proceedings of the National Academy of Sciences, USA 98: 753758.
  • Hiruma K, Onozawa-Komori M, Takahashi F, Asakura M, Bednarek P, Okuno T, Schulze-Lefert P, Takano Y. 2010. Entry mode-dependent function of an indole glucosinolate pathway in Arabidopsis for nonhost resistance against anthracnose pathogens. Plant Cell 22: 24292443.
  • Jimenez LD, Ayer WA, Tewari JP. 1997. Phytoalexins produced in the leaves of Capsella bursa-pastoris (shepherd’s purse). Phytoprotection 78: 99103.
  • Kim DY, Bovet L, Maeshima M, Martinoia E, Lee Y. 2007. The ABC transporter AtPDR8 is a cadmium extrusion pump conferring heavy metal resistance. Plant Journal 50: 207218.
  • Kliebenstein DJ, Kroymann J, Brown P, Figuth A, Pedersen D, Gershenzon J, Mitchell-Olds T. 2001. Genetic control of natural variation in Arabidopsis glucosinolate accumulation. Plant Physiology 126: 811825.
  • Kliebenstein DJ, Rowe HC, Denby KJ. 2005. Secondary metabolites influence Arabidopsis/Botrytis interactions: variation in host production and pathogen sensitivity. Plant Journal 44: 2536.
  • Koch M, Haubold B, Mitchell-Olds T. 2001. Molecular systematics of the Brassicaceae: evidence from coding plastidic matK and nuclear Chs sequences. American Journal of Botany 88: 534544.
  • Kwon C, Neu C, Pajonk S, Yun HS, Lipka U, Humphry M, Bau S, Straus M, Kwaaitaal M, Rampelt H et al. 2008. Co-option of a default secretory pathway for plant immune responses. Nature 451: 835840.
  • Lipka V, Dittgen J, Bednarek P, Bhat R, Wiermer M, Stein M, Landtag J, Brandt W, Rosahl S, Scheel D et al. 2005. Pre- and postinvasion defenses both contribute to nonhost resistance in Arabidopsis. Science 310: 11801183.
  • Maeda K, Houjyou Y, Komatsu T, Hori H, Kodaira T, Ishikawa A. 2009. AGB1 and PMR5 contribute to PEN2-mediated preinvasion resistance to Magnaporthe oryzae in Arabidopsis thaliana. Molecular Plant-Microbe Interactions 22: 13311340.
  • Mikkelsen MD, Naur P, Halkier BA. 2004. Arabidopsis mutants in the C-S lyase of glucosinolate biosynthesis establish a critical role for indole-3-acetaldoxime in auxin homeostasis. Plant Journal 37: 770777.
  • Mithen R, Bennett R, Marquez J. 2010. Glucosinolate biochemical diversity and innovation in the Brassicales. Phytochemistry 71: 20742086.
  • Monde K, Takasugi M, Ohnishi T. 1994. Biosynthesis of Cruciferous phytoalexins. Journal of the American Chemical Society 116: 66506657.
  • Morant M, Ekstrom C, Ulvskov P, Kristensen C, Rudemo M, Olsen CE, Hansen J, Jorgensen K, Jorgensen B, Moller BL et al. 2010. Metabolomic, transcriptional, hormonal, and signaling cross-talk in superroot2. Molecular Plant 3: 192211.
  • Ockendon JG, Buczacki ST. 1979. Indole glucosinolate incidence and clubroot susceptibility of three cruciferous weeds. Transactions of the British Mycological Society 72: 156157.
  • Panstruga R, Parker JE, Schulze-Lefert P. 2009. SnapShot: plant immune response pathways. Cell 136: 978U976.
  • Parra G, Bradnam K, Ning ZM, Keane T, Korf I. 2009. Assessing the gene space in draft genomes. Nucleic Acids Research 37: 289297.
  • Pedras MSC, Yaya EE. 2010. Phytoalexins from Brassicaceae: news from the front. Phytochemistry 71: 11911197.
  • Pedras MSC, Yaya EE, Hossain S. 2010. Unveiling the phytoalexin biosynthetic puzzle in salt cress: unprecedented incorporation of glucobrassicin into wasalexins A and B. Organic & Biomolecular Chemistry 8: 51505158.
  • Petersen BL, Chen SX, Hansen CH, Olsen CE, Halkier BA. 2002. Composition and content of glucosinolates in developing Arabidopsis thaliana. Planta 214: 562571.
  • Pfalz M, Mikkelsen MD, Bednarek P, Olsen CE, Halkier BA, Kroymann J. 2011. Metabolic engineering in Nicotiana benthamiana reveals key enzyme functions in Arabidopsis indole glucosinolate modification. Plant Cell 23: 716729.
  • Pfalz M, Vogel H, Kroymann J. 2009. The gene controlling the Indole Glucosinolate Modifier1 quantitative trait locus alters indole glucosinolate structures and aphid resistance in Arabidopsis. Plant Cell 21: 985999.
  • Rodman JE, Soltis PS, Soltis DE, Sytsma KJ, Karol KG. 1998. Parallel evolution of glucosinolate biosynthesis inferred from congruent nuclear and plastid gene phylogenies. American Journal of Botany 85: 9971006.
  • Ruzicka K, Strader LC, Bailly A, Yang HB, Blakeslee J, Langowski L, Nejedla E, Fujita H, Itoh H, Syono K et al. 2010. Arabidopsis PIS1 encodes the ABCG37 transporter of auxinic compounds including the auxin precursor indole-3-butyric acid. Proceedings of the National Academy of Sciences, USA 107: 1074910753.
  • Sanchez-Vallet A, Ramos B, Bednarek P, López G, Pislewska-Bednarek M, Schulze-Lefert P, Molina A. 2010. Tryptophan-derived secondary metabolites in Arabidopsis thaliana confer nonhost resistance to necrotrophic Plectosphaerella cucumerina fungi. Plant Journal 63: 115127.
  • Schlaeppi K, Abou-Mansour E, Buchala A, Mauch F. 2010. Disease resistance of Arabidopsis to Phytophthora brassicae is established by the sequential action of indole glucosinolates and camalexin. Plant Journal 62: 840851.
  • Stein M, Dittgen J, Sanchez-Rodriguez C, Hou BH, Molina A, Schulze-Lefert P, Lipka V, Somerville S. 2006. Arabidopsis PEN3/PDR8, an ATP binding cassette transporter, contributes to nonhost resistance to inappropriate pathogens that enter by direct penetration. Plant Cell 18: 731746.
  • Strader LC, Bartel B. 2009. The Arabidopsis PLEIOTROPIC DRUG RESISTANCE8/ABCG36 ATP binding cassette transporter modulates sensitivity to the auxin precursor indole-3-butyric acid. Plant Cell 21: 19922007.
  • Talke IN, Hanikenne M, Kramer U. 2006. Zinc-dependent global transcriptional control, transcriptional deregulation, and higher gene copy number for genes in metal homeostasis of the hyperaccumulator Arabidopsis halleri. Plant Physiology 142: 148167.
  • Tan JW, Bednarek P, Liu HK, Schneider B, Svatos A, Hahlbrock K. 2004. Universally occurring phenylpropanoid and species-specific indolic metabolites in infected and uninfected Arabidopsis thaliana roots and leaves. Phytochemistry 65: 691699.
  • Thomma BPHJ, Nelissen I, Eggermont K, Broekaert WF. 1999. Deficiency in phytoalexin production causes enhanced susceptibility of Arabidopsis thaliana to the fungus Alternaria brassicicola. Plant Journal 19: 163171.
  • Tierens KFMJ, Thomma BPHJ, Brouwer M, Schmidt J, Kistner K, Porzel A, Mauch-Mani B, Cammue BPA, Broekaert WF. 2001. Study of the role of antimicrobial glucosinolate-derived isothiocyanates in resistance of Arabidopsis to microbial pathogens. Plant Physiology 125: 16881699.
  • Waterhouse AM, Procter JB, Martin DMA, Clamp M, Barton GJ. 2009. Jalview Version 2–a multiple sequence alignment editor and analysis workbench. Bioinformatics 25: 11891191.
  • Windsor AJ, Reichelt M, Figuth A, Svatos A, Kroymann J, Kliebenstein DJ, Gershenzon J, Mitchell-Olds T. 2005. Geographic and evolutionary diversification of glucosinolates among near relatives of Arabidopsis thaliana (Brassicaceae). Phytochemistry 66: 13211333.
  • Zhao B, Liu L, Tan D, Wang J. 2010. Analysis of phylogenetic relationships of Brassicaceae species based on Chs sequences. Biochemical Systematics and Ecology 38: 731739.
  • Zhao YD, Hull AK, Gupta NR, Goss KA, Alonso J, Ecker JR, Normanly J, Chory J, Celenza JL. 2002. Trp-dependent auxin biosynthesis in Arabidopsis: involvement of cytochrome P450s CYP79B2 and CYP79B3. Genes & Development 16: 31003112.
  • Zook M, Leege L, Jacobson D, Hammerschmidt R. 1998. Camalexin accumulation in Arabis lyrata. Phytochemistry 49: 22872289.

Supporting Information

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information

Fig. S1 Comparison of UV-absorbance chromatograms obtained during HPLC analyses of samples extracted from leaves of Brassicaceae plants.

Fig. S2 Accumulation of 4-methoxyindol-3-ylmethylglucosinolate (4MI3G) in the indicated A. thaliana ecotypes 16 h after inoculation with Blumeria graminis f. sp. hordei conidiospores.

Fig. S3 PENETRATION3 (PEN3) orthologs.

Fig. S4 CYP81F2 orthologs.

Fig. S5 Collision-induced dissociation mass spectrum of protonated methoxycamalexin m/z 231 ion.

Fig. S6 PHYTOALEXIN DEFICIENT3 (PAD3) orthologs.

Fig. S7 Accumulation of indol-3-ylmethylglucosinolate (I3G) and 1-methoxyindol-3-ylglucosinolate (1MI3G) in Brassicaceae plants 40 h after inoculation with Botrytis cinerea conidiospores.

Please note: Wiley-Blackwell are not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing material) should be directed to the New Phytologist Central Office.

FilenameFormatSizeDescription
NPH_3824_sm_FigS1-S7.pdf1538KSupporting info item