Abstract

Intraspecific differences in plant defence traits are often correlated with variation in transcriptional profiles and can affect the composition of herbivore communities on field-grown plants. However, most studies on transcriptional profiling of plant–herbivore interactions have been carried out under controlled conditions in the laboratory or greenhouse and only a few examine intraspecific transcriptional variation. Here, intraspecific variation in herbivore community composition and transcriptional profiles between two Brassica oleracea cultivars grown in the field is addressed. Early in the season, no differences in community composition were found for naturally occurring herbivores, whereas cultivars differed greatly in abundance, species richness, and herbivore community later in the season. Genome-wide transcriptomic analysis using an Arabidopsis thaliana oligonucleotide microarray showed clear differences for the expression levels of 26 genes between the two cultivars later in the season. Several defence-related genes showed higher levels of expression in the cultivar that harboured the lowest numbers of herbivores. Our study shows that herbivore community composition develops differentially throughout the season on the two B. oleracea cultivars grown in the field. The correlation between the differences in herbivore communities and differential expression of particular defence-related genes is discussed.

Introduction

Intraspecific variation in plant traits may influence the composition and diversity of herbivore communities on plants grown under natural conditions (Wimp et al., 2005; Whitham et al., 2006; Poelman et al., 2009). Plant traits that affect herbivores include morphological factors, such as wax layers, and defence-related secondary metabolites (Schoonhoven et al., 2005). These plant traits can be constitutively present, but plants can also alter their phenotype in response to herbivory (Kessler and Baldwin, 2002; Howe and Jander, 2008; Inbar and Gerling, 2008; Karban, 2008). Differences in the transcription of particular genes have been shown to correlate with intraspecific variation in phenotypic traits (Carroll, 2000). Studies on different populations of the same species have revealed that variation in transcription of defence-related genes is responsible for variation in secondary metabolite production (Wu et al., 2008) and herbivore resistance (Kuśnierczyk et al., 2007; Gao et al., 2008). Furthermore, the expression of genes regulating the biosynthesis of jasmonic acid, a plant hormone known to mediate plant defences, has been shown significantly to affect the composition of the herbivore community on tobacco plants (Kessler and Baldwin, 2004). However, no data are available from field studies that link intraspecific variation in plant gene expression with herbivore community composition.

Intraspecific variation in herbivore community composition can be affected by induced plant responses. In Brassica oleracea plants, for example, experimentally introducing Pieris rapae caterpillars early in the season influenced herbivore community composition later in the season (Poelman et al., 2008; EH Poelman, JJA Van Loon, NM Van Dam, LEM Vet, M Dicke, unpublished data). Depending on their feeding strategy herbivores differentially induce plant responses (Walling, 2000), which may affect the performance of the initial herbivore as well as that of subsequently colonizing species (Agrawal, 2000; Traw and Dawson, 2002). Induced plant responses not only influence the performance of subsequent herbivores feeding on the plant, but may also affect their host plant preference (De Moraes et al., 2001; Shiojiri et al., 2002; Long et al., 2007; EH Poelman, JJA Van Loon, NM Van Dam, LEM Vet, M Dicke, unpublished data).

Herbivores may be differentially affected by induced plant responses depending on their degree of host plant specialization. Induced secondary metabolites that have a negative effect on generalist herbivores may act as feeding stimulants or can be detoxified by specialists (Agrawal, 2000; Ratzka et al., 2002; Wittstock et al., 2004; Kliebenstein et al., 2005; Després et al., 2007). Specialists may even be able to accumulate certain defence-related secondary metabolites to use them for their own defence (Després et al., 2007; Kazana et al., 2007). However, it should be noted that specialists may still be susceptible to the toxic effects of secondary metabolites (Adler et al., 1995; Agrawal and Kurashige, 2003; Steppuhn et al., 2004; Ballhorn et al., 2007). Differences between generalists and specialists can also be found with regard to attraction: generalist herbivores often avoid induced plants, whereas some specialists may prefer these plants (Bolter et al., 1997; Kaplan and Denno, 2007; Long et al., 2007; Poelman et al., 2008).

Signal transduction pathways underlie induced defences in which the plant hormones jasmonic acid (JA), salicylic acid (SA), and ethylene (ET) play important roles (Kessler and Baldwin, 2002; Pieterse and Dicke, 2007). Gene expression profiling using microarrays has been profoundly useful in investigating mechanisms of induced defences (Rishi et al., 2002; Hui et al., 2003; Korth, 2003; Reymond et al., 2004; Voelckel and Baldwin, 2004; De Vos et al., 2005; Thompson and Goggin, 2006; Smith and Boyko, 2007). Most of these studies have been performed under carefully controlled environmental conditions in the greenhouse in which plants are exposed to a single attacker. In natural habitats, however, plants can be exposed to multiple herbivores simultaneously and under a variety of conditions. It is unclear whether differences in gene expression observed in the greenhouse are sustained in the field. Transcriptional responses in field-grown plants have been studied after exposure to methyl jasmonate (Schmidt and Baldwin, 2006), induction by Manduca sexta herbivory (Izaguirre et al., 2003), or Japanese beetles (Popillia japonica) (Casteel et al., 2008). None of these studies have investigated intraspecific variation in gene expression nor did they monitor the presence of naturally occurring herbivorous insects.

In the Brassicaceae, domestication has given rise to several important crops including white cabbage (B. oleracea var. capitata). As many varieties are available and plants within a variety are quite uniform, B. oleracea cultivars provide a unique opportunity to investigate intraspecific patterns of gene expression in response to herbivory. Intraspecific variation in the secondary metabolite content of four B. oleracea var. capitata cultivars (Rivera, Lennox, Christmas Drumhead, and Badger Shipper) has been shown to influence herbivore community composition in the field (Poelman et al., 2009). Two of these cultivars, Rivera and Christmas Drumhead, have also been shown to differ in transcriptional responses to herbivory by caterpillars of the Small Cabbage White Pieris rapae and the cabbage aphid Brevicoryne brassicae feeding under greenhouse conditions (Broekgaarden et al., 2007, 2008). The present study addresses the question whether differences in herbivore community composition in the field between two cultivars from the same species, i.e. the two B. oleracea cultivars Rivera and Christmas Drumhead, can be related to intraspecific variation in gene expression. To our knowledge, this is the first study that links herbivore community dynamics and whole-genome gene expression under field conditions where plants are exposed to naturally occurring herbivores.

Materials and methods

Plant growth

Seeds of the F1 hybrid white cabbage (Brassica oleracea var. capitata) cultivar Rivera and the open-pollinated cultivar Christmas Drumhead were obtained from Bejo Zaden BV (Warmenhuizen, The Netherlands) and the Centre of Genetic Resources, The Netherlands (CGN), respectively. Seeds were directly sown in peat soil cubes containing potting compost (Lentse Potgrond BV, The Netherlands) and allowed to germinate in a greenhouse compartment (22–26/18–22 oC light/dark; 40–70% relative humidity). Prior to being transplanted into the field site, trays with peat soil cubes containing 3-week-old seedlings were placed outside the greenhouse during the day for 2 weeks. Both cultivars show similar morphology and growth characteristics (see Supplementary Fig. S1 at JXB online). Christmas Drumhead is somewhat earlier in forming a head than Rivera.

Field site

In 2007, a field experiment in an agricultural field near Wageningen, The Netherlands was established. Eighteen plots (6×6 m) with a monoculture of one of the two cultivars (ten plots for Rivera and eight plots for Christmas Drumhead) were established using a randomized design. Five-week-old plants were transferred with their peat soil cubes to the field in week 19 (7 May) of 2007. Plots contained 49 plants in a square of 7×7 plants with a spacing of 75 cm between plants. A strip of 6 m sown with a grass mixture of Lolium and Poa species isolated the plots.

Collection of material

In week 23 (6 June) and week 32 (6 August), i.e. 4 weeks and 13 weeks after plants had been transferred to the field, respectively, material was collected from 18 plots (ten for Rivera and eight for Christmas Drumhead). The two time points were selected based on peaks in the herbivore abundance in 2005 (Poelman et al., 2009) and 2006 (EH Poelman, JJA Van Loon, NM Van Dam, LEM Vet, M Dicke, unpublished data). One leaf disc (diameter 2.3 cm) was harvested from a young leaf of nine separate plants in each plot, and the leaf discs were pooled to create a single sample per plot. Upon harvesting, samples were immediately flash-frozen in liquid nitrogen and stored at –80 °C. After collecting leaf discs, the same plants were completely harvested in plastic bags to monitor the presence of naturally occurring insects. Bags were stored at 4 °C until plants were monitored. All plants were monitored within 5 d. To assess whether plant biomass or the number of leaves could explain the differences in herbivore community composition between the cultivars, all plants were weighed individually and the number of leaves per plant was counted.

Herbivore biodiversity calculations and analysis

For both time points, the number of individuals per herbivore species was counted on the nine plants of a plot and herbivores were weighed on a microgram balance. These values were used to calculate per plant (i) the total herbivore abundance, (ii) the species richness, and (iii) the Shannon–Wiener diversity index. Total herbivore abundance represents the total number of individuals, whereas species richness represents the total number of herbivorous species. The Shannon–Wiener biodiversity index describes herbivore diversity by taking into account both the richness of species as well as the evenness of their distribution (Mendes et al., 2008). Both a large number of unique species and higher evenness of their abundance distribution increase the value of this index. Differences between the two cultivars for all measured parameters were statistically analysed with Mann–Whitney U tests.

Herbivore abundance, species richness, total herbivore mass, and biodiversity index were regressed onto plant weight and number of leaves in multiple linear regression analysis, with cultivar as the grouping factor.

RNA isolation, aRNA synthesis, and dye labelling

Leaf samples from two plots were pooled per cultivar and three biological replicates were analysed per cultivar. Total RNA was extracted with the TRIzol reagent (Invitrogen, Carlsbad, CA, USA) followed by a purification using the RNeasy Plant Mini kit (Qiagen, Valencia, CA, USA). Four micrograms of total RNA were linearly amplified using the Amino Allyl MessageAmp II aRNA Amplification kit (Ambion, Austin, TX, USA). Samples were labelled with Cy3 (Rivera) and Cy5 (Christmas Drumhead) monoreactive dye (Amersham, Piscataway, NY, USA). Amplified RNA (aRNA) was labelled in freshly made 0.2 M sodium carbonate buffer (pH 9.0) for 1 h at room temperature. Dye incorporation was monitored by measuring the Cy3 and Cy5 fluorescence emissions using a nanodrop ND-1000 UV-Vis Spectrophotometer (Bio-Rad, Hercules, CA, USA).

70-mer A. thaliana microarray

Microarrays containing 70-mer oligonucleotides, based on the genome of A. thaliana, were obtained from the group of David Galbraith from the University of Arizona, Tucson, AZ, USA (http://www.ag.arizona.edu/microarray). These microarrays contain 29 110 probes from the Operon Arabidopsis Genome Oligo Set Version 3.0 (Operon). This oligo set represents 26 173 protein-coding genes, 28 964 protein-coding gene transcripts, and 87 miRNAs. The majority of genes are represented by one 70-mer probe on the microarray. A 70-mer instead of a 25-mer microarray was used as the longer oligos have a higher sensitivity (Relogio et al., 2002) and non-specific binding of mismatched targets can be kept to a minimum by using long probes (Buckley, 2007). Moreover, the microarray has proved to be a good tool to study transcriptomics in B. oleracea (Lee et al., 2004; Broekgaarden et al., 2007, 2008; Fatouros et al., 2008). In addition, independent support for this approach is provided by other studies involving heterologous hybridizations (Becher et al., 2004; Buckley, 2007; Davey et al., 2009). Combined data obtained from 72 hybridizations using B. oleracea material (data from Broekgaarden et al., 2007, 2008; data presented in this manuscript; C Broekgaarden, unpublished data) show that around 90% of the oligonucleotides present on the microarray hybridized in at least two experiments. Although the two cultivars used in this study are from the same species, the overlap in hybridizing probes was not complete, and around 80% of the probes hybridized with material from each cultivar separately.

It should be realized that the transcription of genes that are specific for B. oleracea will not be detected with A. thaliana microarrays. Yet, the use of the 70-mer A. thaliana microarray provides a good tool to investigate transcriptomic changes of a large proportion of B. oleracea genes, which is a great advantage of this approach.

Microarray hybridization

Immobilization of the array elements was performed according to the manufacturer's website (see previous discussion). The arrays used all originated from the same printing batch, thus eliminating batch to batch variation. The hybridization mixture contained 100 pmol of the Cy3-labelled sample, 50 pmol of the Cy5-labelled sample, 2× SSC, 0.08% SDS, and 4.8 μl Liquid Block (Amersham) in a final volume of 80 μl. The solution was incubated at 65 °C for 5 min before applying to the microarray covered with a lifterslip (Gerhard Menzel, Braunschweig, Germany). The microarray was placed in a hybridization chamber (Genetix, New Milton, Hampshire, UK) and incubated at 50 °C. After 12 h the microarray was washed for 5 min in 2× SSC/0.5% SDS at 50 °C, followed by a 5 min wash in 0.5× SSC at room temperature, and a final 5 min wash in 0.05× SSC at room temperature. The microarray was immediately dried by centrifugation for 4 min at 200 rpm.

Hybridized microarrays were scanned with a ScanArray Express HT Scanner (PerkinElmer, Waltham, MA, USA).

Microarray analysis

Mean fluorescent intensities for Cy3 and Cy5 were determined using the ScanArray Express software (PerkinElmer). Each image was overlaid with a grid to assess the signal intensities for both dyes from each spot. Background fluorescence was subtracted and spots with adjusted intensities lower than half the background were manually raised to half the background to avoid extreme expression ratios. Spots were excluded from the analysis when: (i) showing signal intensities less than half the background for both dyes; (ii) showing aberrant shape; or (iii) located in a smear of fluorescence. To correct for hybridization efficiency differences between the cultivars, spots that have been shown to hybridize with material from one cultivar only were also removed from the data analysis. Lowess (locfit) normalization was carried out within each slide using TIGR MIDAS version 2.19 (The Institute for Genomic Research, Rockville, MD, USA) to remove any systematic dye effects, assuming that the overall gene expression for the two cultivars is approximately equal. Normalized expression ratios for each individual spot and the mean of the three replicate spots were calculated. A Student's t test on log2 transformed expression ratios was conducted for each experimental condition using TIGR MEV version 3.1. To address the issue of multiple testing and to identify the proportion of false positives among the genes identified as differentially expressed, the false discovery rate (FDR) was calculated using the Benjamini–Hochberg method (Benjamini and Hochberg, 1995). A q-value was computed for each gene with a log2 expression ratio ≥1 or ≤ –1 using the distribution of P values of all measurements. Genes that showed a significant difference in expression level (P <0.05 and FDR <0.1) and a log2 expression ratio ≥1 or ≤ –1 were considered to be higher expressed in Rivera or Christmas Drumhead, respectively. The names of A. thaliana homologues were used to identify B. oleracea genes and examined the potential function of differentially regulated genes according to gene ontology (GO) terms from The Arabidopsis Information Resource (http://www.arabidopsis.org).

Quantitative RT-PCR

Quantitative RT-PCR was used to examine gene expression of selected genes per plot by using the RNA pools of all 18 plots separately. One microgram of total RNA was treated with DNaseI (Invitrogen) according to the manufacturer's instructions. DNA-free total RNA was converted into cDNA using the iScript cDNA synthesis kit (Bio-Rad). Gene-specific primers were designed for B. oleracea genes based on sequences obtained by a BLAST search in the TIGR B. oleracea database (Lipoxygenase 2, LOX2: left 5′-CTT TGC TCA CAT ACG GTA GAA GC-3′, right 5′-CCT TTG CAT TGG GCT AGT TC-3′; Trypsin-and-protease inhibitor, TPI: left 5′-TGG TGA CAA GTA GCT GTG GTG-3′, right 5′-TCC AAG TTA TGG GCA GTG G-3′). Primers were tested for gene specificity by performing melt curve analysis and PCR products were sequenced to confirm amplification of the gene of interest. Sequence results were checked by a BLAST search in the B. oleracea as well as in the A. thaliana TIGR database. Quantitative RT-PCR analysis was done in optical 96-well plates with a MyIQ Single-Color Real-Time PCR Detection System (Bio-Rad), using SYBR Green to monitor dsDNA synthesis. Each reaction contained 10 μl 2× SYBR Green Supermix Reagent (Bio-Rad), 10 ng cDNA, and 300 nM of each gene-specific primer in a final volume of 20 μl. All qRT-PCR were performed in duplicate and average values were used in the analyses. The following PCR program was used for all PCR reactions: 3 min at 95 °C; 40 cycles of 30 s at 95 °C and 45 s at 60 °C. Threshold cycle (Ct) values were calculated using Optical System software, version 2.0 for MyIQ (Bio-Rad). Subsequently, Ct values were normalized for differences in cDNA synthesis by subtracting the Ct value of the constitutively expressed gene glyceraldehyde-3-phosphate dehydrogenase (GAPDH: left 5′-AGA GCC GCT TCC TTC AAC ATC ATT-3′, right 5′-TGG GCA CAC GGA AGG ACA TAC C-3′) from the Ct value of the gene of interest. Normalized gene expression was then calculated as 2–ΔΔCt. Differences between time points were analysed with one-way analysis of variance (ANOVA) followed by Post Hoc multiple comparison tests with Least Significant Difference (LSD).

Results

Abundance of naturally occurring herbivores

Fourteen species of herbivorous insects were found in the field (Table 1), all of which were previously reported to be associated with B. oleracea (Root, 1973; Mitchell and Richards, 1979; Poelman et al., 2009). Thirteen occurred on both cultivars and one, Autographa gamma, was only found on Christmas Drumhead. Early in the season, 4 weeks after transplanting seedlings into the field, nine herbivore species were found on Rivera and Christmas Drumhead with no differences in abundance on the two cultivars (Fig. 1). At this point in the season, Brevicoryne brassicae was the most abundant herbivore on both cultivars with about 20 individuals per plant.

Table 1.

Herbivore species found on the Brassica oleracea cultivars Rivera and Christmas Drumhead early and later in the season and their degree of specialization in relation to brassicaceous plants

Order Family Species Feeding strategy Specialization 
Lepidoptera Pieridae Pieris rapae Leaf chewing Specialist 
  Pieris brassicae Leaf chewing Specialist 
 Plutellidae Plutella xylostella Leaf chewing Specialist 
 Pyralidae Evergestris fortificalis Leaf chewing Specialist 
 Noctuidae Mamestra brassicae Leaf chewing Generalist 
  Autographa gamma Leaf chewing Generalist 
Coleoptera Chrysomelidae Phyllotreta atra Leaf chewing Specialist 
  Phyllotreta undulata Leaf chewing Specialist 
Hemiptera Aphididae Brevicoryne brassicae Phloem feeding Specialist 
  Myzus persicae Phloem feeding Generalist 
 Aleyrodidae Aleyrodes proletella Phloem feeding Specialist 
Thysanoptera Thripidae Thrips tabaci Cell content feeding Generalist 
Diptera Anthomyiidae Delia radicum Root feeding Specialist 
Order Family Species Feeding strategy Specialization 
Lepidoptera Pieridae Pieris rapae Leaf chewing Specialist 
  Pieris brassicae Leaf chewing Specialist 
 Plutellidae Plutella xylostella Leaf chewing Specialist 
 Pyralidae Evergestris fortificalis Leaf chewing Specialist 
 Noctuidae Mamestra brassicae Leaf chewing Generalist 
  Autographa gamma Leaf chewing Generalist 
Coleoptera Chrysomelidae Phyllotreta atra Leaf chewing Specialist 
  Phyllotreta undulata Leaf chewing Specialist 
Hemiptera Aphididae Brevicoryne brassicae Phloem feeding Specialist 
  Myzus persicae Phloem feeding Generalist 
 Aleyrodidae Aleyrodes proletella Phloem feeding Specialist 
Thysanoptera Thripidae Thrips tabaci Cell content feeding Generalist 
Diptera Anthomyiidae Delia radicum Root feeding Specialist 
Fig. 1.

Numbers of naturally occurring herbivores in the field on Rivera (white bars) and Christmas Drumhead (black bars) early (week 23) and later (week 32) in the season. Mean numbers of individuals per plant are given with their corresponding standard error. Brevicoryne brassicae and Aleyrodes proletella are represented separately because the numbers of these individuals were much higher than for the other species. Bars within pairs marked with one or more asterisks differ significantly (independent sample t test, *P <0.05; **P <0.01; ***P ≤0.001).

Fig. 1.

Numbers of naturally occurring herbivores in the field on Rivera (white bars) and Christmas Drumhead (black bars) early (week 23) and later (week 32) in the season. Mean numbers of individuals per plant are given with their corresponding standard error. Brevicoryne brassicae and Aleyrodes proletella are represented separately because the numbers of these individuals were much higher than for the other species. Bars within pairs marked with one or more asterisks differ significantly (independent sample t test, *P <0.05; **P <0.01; ***P ≤0.001).

Nine weeks after the first time point, the abundance and species richness of herbivores on the cultivars had changed completely and was lower on Rivera than on Christmas Drumhead (Mann–Whitney U test, abundance: U=0, P <0.001; richness: U=2, P <0.001). Rivera harboured significantly fewer Pieris rapae and Mamestra brassicae larvae than Christmas Drumhead (Mann–Whitney U test, P. rapae: U=0, P <0.001; M. brassicae: U=5, P=0.001; Fig. 1). Several larvae of A. gamma were found on Christmas Drumhead, whereas this species was absent on Rivera (Fig. 1). Furthermore, less than half as many flea beetles were found on Rivera than on Christmas Drumhead (Phyllotreta atra: U=7, P=0.002; P. undulata: U=12, P=0.012; Fig. 1). Large differences between the cultivars were found for the occurrence of cabbage aphids (B. brassicae U=0, P <0.001) and whiteflies (Aleyrodes proletella U=0, P <0.001) later in the season. Hardly any individuals of these two species were present on Rivera, whereas on Christmas Drumhead c. 30 and 70 individuals per plant were found of these two species, respectively (Fig. 1). Remarkably, a few A. proletella adults were found on Rivera, but no pupae of this species were present on this cultivar. Four times more A. proletella pupae than adults were found on Christmas Drumhead. In addition, lower numbers of the phloem-feeding herbivore Myzus persicae were found on Rivera than on Christmas Drumhead (U=0, P <0.001; Fig. 1). Three species of lepidopteran larvae were equally distributed over the two cultivars (P. brassicae: U=29, P=0.360; Plutella xylostella: U=21, P=0.101; Evergestis fortificalis: U=25, P=0.203; Fig. 1).

In general, Rivera harboured significantly lower abundance and species richness of specialist (abundance: U=0, P <0.001; richness U=7, P=0.003) as well as generalist herbivores than Christmas Drumhead (abundance: U=5, P <0.001; richness: U=1.5, P <0.001; Fig. 2A, B). The total mass of herbivores collected from Rivera was also significantly lower than the total mass of herbivores collected from Christmas Drumhead (U=4, P=0.001). This difference was mostly caused by the specialist herbivores (specialists: U=4, P=0.001; generalists: U=8, P <0.004; Fig. 2C). Due to the distribution of the herbivore species, Rivera had the highest herbivore biodiversity (Shannon–Wiener index: Rivera 1.65±0.06, Christmas Drumhead 1.08±0.13; U=8, P=0.003).

Fig. 2.

Herbivore community composition parameters later in the season (week 32) for Rivera (white bars) and Christmas Drumhead (black bars) are given for specialist and generalist herbivores. Graphs represent: (A) total number of herbivore individuals per plant (+SE); (B) total number of species per plant (+SE); (C) total mass of all herbivores per plant (+SE). Bars within pairs marked with three asterisks differ significantly (independent sample t test, P ≤0.001).

Fig. 2.

Herbivore community composition parameters later in the season (week 32) for Rivera (white bars) and Christmas Drumhead (black bars) are given for specialist and generalist herbivores. Graphs represent: (A) total number of herbivore individuals per plant (+SE); (B) total number of species per plant (+SE); (C) total mass of all herbivores per plant (+SE). Bars within pairs marked with three asterisks differ significantly (independent sample t test, P ≤0.001).

Herbivore abundance, species richness, total herbivore mass, or biodiversity were not significantly affected by plant weight or number of leaves (linear regression df=15, abundance: weight P=0.96, leaves P=0.17; richness: weight P=0.74, leaves P=1.00; mass: weight P=0.43, leaves P=0.47; Shannon–Wiener index: weight P=0.50, leaves P=0.84).

Gene expression differences between Rivera and Christmas Drumhead in the field

Arabidopsis thaliana full-genome microarrays were used to test whether differences in gene expression levels exist between the two cultivars under field conditions. Early in the season, only a small number of genes showed significantly different expression levels between the two cultivars. One and five genes showed higher levels of expression in Rivera or Christmas Drumhead, respectively, including genes mainly involved in general metabolic processes and genes of unknown function (Table 2). Later in the season, differences in expression levels between Rivera and Christmas Drumhead were more pronounced as 26 genes showed different expression levels (Table 2). The 12 genes with higher expression levels in Rivera include, among others, genes involved in defence and metabolic processes. The defence-related genes that were identified in Rivera include genes encoding lipoxygenase 2 (LOX2), lectin (At2g39310), a trypsin inhibitor, and a Bet v I allergen (Table 2). In Christmas Drumhead, 14 genes showed higher expression levels than in Rivera of which most were involved in metabolic processes (Table 2). One of the genes with a higher expression level in this cultivar is involved in defence and encodes flavin-dependent monooxygenase 1 (FMO1) (Table 2).

Table 2.

Genes with a higher level of expression in Rivera or Christmas Drumhead under field conditions early and later in the season

Probe identification AGI code Ratio P-value q-value Process 
(A) Early in the season      
    Higher expression in Rivera      
        Endonuclease/exonuclease/phosphatase At1g31500 2.35 0.004 0.082 Unknown 
    Higher expression in Christmas Drumhead      
        MMS Zwei Homologe 4 (MMZ4) At3g52560 2.22 0.008 0.091 Metabolic processes 
        Prenylcysteine alpha-carboxyl methyltransferase (STE14B) At5g08335 3.11 0.002 0.048 Metabolic processes 
        Histone H4 At5g59690 2.38 0.001 0.048 Metabolic processes 
        33 kDa secretory protein-related At5g48540 2.31 0.009 0.091 Unknown 
        Hypothetical protein At1g10800 2.37 0.008 0.091 Unknown 
(B) Later in the season      
    Higher expression in Rivera      
        Giant chloroplast 1 (GC1) At2g21280 2.29 0.023 0.096 Cell organization and biogenesis 
        Bet v I allergen At1g24020 3.49 0.01 0.092 Defence 
        Lectin At2g39310 5.16 0.007 0.092 Defence 
        Trypsin inhibitor At2g43530 2.36 0.013 0.092 Defence 
        Lipoxygenase 2 (LOX2) At3g45140 2.6 0.008 0.092 Defence 
        60S ribosomal protein L39 (RPL39A) At2g25210 4.44 0.005 0.092 Metabolic processes 
        Ubiquitin extension protein 2 (UBQ2) At2g36170 2.31 0.004 0.092 Metabolic processes 
        60S ribosomal protein L41 (RPL41C) At2g40205 2.33 0.017 0.092 Metabolic processes 
        Signal peptidase At3g15710 3.15 0.018 0.092 Metabolic processes 
        Vacuolar H+-ATPase subunit E isoform 3 At1g64200 2.15 0.016 0.092 Transport 
        Dehydroascorbate reductase At1g19550 3.22 0.022 0.096 Unknown 
        Thioredoxin-dependent peroxidase 1 (TPX1) At1g65980 2.51 0.004 0.092 Unknown 
    Higher expression in Christmas Drumhead      
        Flavin-dependent monooxygenase 1 (FMO1) At1g19250 4.33 0.02 0.095 Defence 
        Kinesin-13A At3g16630 2.23 0.012 0.092 Development 
        GDSL-motif lipase/hydrolase At1g29660 2.41 0.012 0.092 Metabolic processes 
        Prefoldin-related KE2 At3g22480 2.22 0.013 0.092 Metabolic processes 
        Tubulin beta-4 chain (TUB4) At5g44340 2.31 0.014 0.092 Metabolic processes 
        LHCA1 At3g54890 2.16 0.013 0.092 Photosynthesis 
        DC1 domain-containing protein At5g54050 2.77 0.018 0.092 Signal transduction 
        Iron superoxide dismutase (FSD1) At4g25100 2.17 0.005 0.092 Stress response 
        Transcription factor B3 At5g60140 2.56 0.012 0.092 Transcription 
        Calcium-binding EF hand At1g20760 2.51 0.018 0.092 Unknown 
        Protein kinase-related At3g03930 2.59 0.001 0.092 Unknown 
        Expressed protein At3g12320 3.21 0.022 0.096 Unknown 
        Expressed protein At4g20290 2.16 0.018 0.092 Unknown 
        Expressed protein At5g25640 3.35 0.01 0.092 Unknown 
Probe identification AGI code Ratio P-value q-value Process 
(A) Early in the season      
    Higher expression in Rivera      
        Endonuclease/exonuclease/phosphatase At1g31500 2.35 0.004 0.082 Unknown 
    Higher expression in Christmas Drumhead      
        MMS Zwei Homologe 4 (MMZ4) At3g52560 2.22 0.008 0.091 Metabolic processes 
        Prenylcysteine alpha-carboxyl methyltransferase (STE14B) At5g08335 3.11 0.002 0.048 Metabolic processes 
        Histone H4 At5g59690 2.38 0.001 0.048 Metabolic processes 
        33 kDa secretory protein-related At5g48540 2.31 0.009 0.091 Unknown 
        Hypothetical protein At1g10800 2.37 0.008 0.091 Unknown 
(B) Later in the season      
    Higher expression in Rivera      
        Giant chloroplast 1 (GC1) At2g21280 2.29 0.023 0.096 Cell organization and biogenesis 
        Bet v I allergen At1g24020 3.49 0.01 0.092 Defence 
        Lectin At2g39310 5.16 0.007 0.092 Defence 
        Trypsin inhibitor At2g43530 2.36 0.013 0.092 Defence 
        Lipoxygenase 2 (LOX2) At3g45140 2.6 0.008 0.092 Defence 
        60S ribosomal protein L39 (RPL39A) At2g25210 4.44 0.005 0.092 Metabolic processes 
        Ubiquitin extension protein 2 (UBQ2) At2g36170 2.31 0.004 0.092 Metabolic processes 
        60S ribosomal protein L41 (RPL41C) At2g40205 2.33 0.017 0.092 Metabolic processes 
        Signal peptidase At3g15710 3.15 0.018 0.092 Metabolic processes 
        Vacuolar H+-ATPase subunit E isoform 3 At1g64200 2.15 0.016 0.092 Transport 
        Dehydroascorbate reductase At1g19550 3.22 0.022 0.096 Unknown 
        Thioredoxin-dependent peroxidase 1 (TPX1) At1g65980 2.51 0.004 0.092 Unknown 
    Higher expression in Christmas Drumhead      
        Flavin-dependent monooxygenase 1 (FMO1) At1g19250 4.33 0.02 0.095 Defence 
        Kinesin-13A At3g16630 2.23 0.012 0.092 Development 
        GDSL-motif lipase/hydrolase At1g29660 2.41 0.012 0.092 Metabolic processes 
        Prefoldin-related KE2 At3g22480 2.22 0.013 0.092 Metabolic processes 
        Tubulin beta-4 chain (TUB4) At5g44340 2.31 0.014 0.092 Metabolic processes 
        LHCA1 At3g54890 2.16 0.013 0.092 Photosynthesis 
        DC1 domain-containing protein At5g54050 2.77 0.018 0.092 Signal transduction 
        Iron superoxide dismutase (FSD1) At4g25100 2.17 0.005 0.092 Stress response 
        Transcription factor B3 At5g60140 2.56 0.012 0.092 Transcription 
        Calcium-binding EF hand At1g20760 2.51 0.018 0.092 Unknown 
        Protein kinase-related At3g03930 2.59 0.001 0.092 Unknown 
        Expressed protein At3g12320 3.21 0.022 0.096 Unknown 
        Expressed protein At4g20290 2.16 0.018 0.092 Unknown 
        Expressed protein At5g25640 3.35 0.01 0.092 Unknown 

Relative differences in expression levels in Rivera compared to Christmas Drumhead were measured in field-grown plants early and later in the season. Mean expression ratios and P values (Student t test) were calculated from three biological replicates. An FDR correction was included to correct for multiple testing. AGI, Arabidopsis Genome Initiative.

Gene expression in the field compared to herbivore-induced responses in the greenhouse

The genes that showed different levels of expression in field-grown Rivera and Christmas Drumhead plants were compared to previously identified P. rapae- and B. brassicae-induced genes in plants grown in the greenhouse (Broekgaarden et al., 2007, 2008). About half (5/12) of the genes that showed a higher level of expression in Rivera compared to Christmas Drumhead in the field were previously identified as P. rapae-inducible in one or both cultivars in the greenhouse (Broekgaarden et al., 2007; Fig. 3), including the defence-related genes LOX2, trypsin inhibitor, and the gene encoding lectin (At2g39310). Only one of the 14 genes, transcription factor B3, that showed a higher expression level in Christmas Drumhead compared to Rivera in the field, were previously identified as P. rapae-inducible in one or both cultivars under greenhouse conditions (Broekgaarden et al., 2007; Fig. 3). None of the genes that showed a differential expression between the cultivars under field conditions were previously identified as B. brassicae-responsive (Broekgaarden et al., 2008).

Fig. 3.

Venn diagram representing the distribution of genes with a higher level of expression in Rivera or Christmas Drumhead in the field, compared to genes induced by P. rapae after 24, 48 and/or 72 h in the greenhouse (data from Broekgaarden et al., 2007, have also been used in this figure).

Fig. 3.

Venn diagram representing the distribution of genes with a higher level of expression in Rivera or Christmas Drumhead in the field, compared to genes induced by P. rapae after 24, 48 and/or 72 h in the greenhouse (data from Broekgaarden et al., 2007, have also been used in this figure).

Quantitative RT-PCR analysis using B. oleracea-derived primers for LOX2, a gene known to be involved in defence, confirmed the microarray result by showing a significantly higher expression in Rivera (1.30±0.09) than in Christmas Drumhead (0.52±0.13) in the field (independent sample t test, P <0.001). In order to compare gene expression levels between field- and greenhouse-grown plants, absolute gene expression levels were calculated from data obtained from control plants and P. rapae-challenged plants of Rivera and Christmas Drumhead grown in the greenhouse (Broekgaarden et al., 2007). Expression levels of LOX2 in field-grown plants were significantly higher than those in control plants grown in the greenhouse for both cultivars (one-way ANOVA, df=4, P <0.001; Fig. 4). However, expression levels in the field were significantly lower than the levels reached after 72 h of P. rapae feeding for Christmas Drumhead (P=0.048; Fig. 4).

Fig. 4.

Expression levels of LOX2 in field-grown plants and plants grown in the greenhouse (GH) that were either unchallenged (control) or challenged for 24, 48, or 72 h by P. rapae. Bars represent mean LOX2 expression levels relative to the reference gene GAPDH for Rivera (white bars) and Christmas Drumhead (black bars) with standard error bars. Bars marked with different letters are significantly different (one-way ANOVA, P <0.05).

Fig. 4.

Expression levels of LOX2 in field-grown plants and plants grown in the greenhouse (GH) that were either unchallenged (control) or challenged for 24, 48, or 72 h by P. rapae. Bars represent mean LOX2 expression levels relative to the reference gene GAPDH for Rivera (white bars) and Christmas Drumhead (black bars) with standard error bars. Bars marked with different letters are significantly different (one-way ANOVA, P <0.05).

The expression of TPI (trypsin-and-protease inhibitor), a defence-related gene that has been shown to be P. rapae- and B. brassicae-inducible in the greenhouse (Broekgaarden et al., 2007, 2008) was also monitored. In A. thaliana, disrupting the expression of TPI resulted in better performance of P. rapae and B. brassicae (Broekgaarden et al., 2008) indicating a role for this gene in induced plant defences and therefore in affecting herbivore community composition. The microarray showed that TPI expression levels were 2.61 times higher in Rivera than in Christmas Drumhead, however, due to large variation between the replicates, the difference was not significant (P=0.116; FDR=0.162). QRT-PCR analysis using B. oleracea-derived primers revealed a significantly higher level of expression in Rivera (1.56±0.20) than in Christmas Drumhead (0.93±0.36) for this gene (independent sample t test, P=0.04). For both cultivars, the expression levels of TPI were significantly higher in field-grown plants compared to control plants grown in the greenhouse for both cultivars (one-way ANOVA, df=4, P <0.001; Fig. 5A). However, TPI expression levels in field-grown plants did not reach the levels of greenhouse-grown plants challenged for 48 h or 72 h with P. rapae (Rivera 48 h: P=0.02; Rivera 72 h: P=0.003; Christmas Drumhead 48 h: P=0.01; Christmas Drumhead 72 h: P=0.02; Fig. 5A). IByn contrast, expression levels of TPI in field-grown plants were significantly higher than expression levels after B. brassicae feeding in the greenhouse (P <0.001; Fig. 5B).

Fig. 5.

Expression levels of TPI in field-grown plants and plants grown in the greenhouse that were either unchallenged (control) or challenged for 24, 48, or 72 h of P. rapae (A) or B. brassicae (B) feeding. Bars represent mean TPI expression levels relative to the reference gene GAPDH for Rivera (white bars) and Christmas Drumhead (black bars) with their corresponding standard error. Bars marked with different letters are significantly different (one-way ANOVA, P <0.05).

Fig. 5.

Expression levels of TPI in field-grown plants and plants grown in the greenhouse that were either unchallenged (control) or challenged for 24, 48, or 72 h of P. rapae (A) or B. brassicae (B) feeding. Bars represent mean TPI expression levels relative to the reference gene GAPDH for Rivera (white bars) and Christmas Drumhead (black bars) with their corresponding standard error. Bars marked with different letters are significantly different (one-way ANOVA, P <0.05).

Discussion

Rivera and Christmas Drumhead differentially affect herbivore communities throughout the season

In our field experiment, Rivera and Christmas Drumhead were exposed to naturally occurring populations of herbivorous insects and the abundance of these herbivores was monitored early and later in the season. Early in the season, i.e. 4 weeks after seedlings were planted into the field, the two B. oleracea cultivars harboured similar numbers of herbivorous insects. By contrast, later in the season, when plants were present in the field for 13 weeks, clear differences in herbivore communities were found between the cultivars (Fig. 1). These data show that intraspecific differences in herbivore communities between Rivera and Christmas Drumhead develop throughout the season. This is in agreement with the finding that genotypic differences between plants have a stronger effect on herbivore communities than environmental factors (Bangert et al., 2006; Johnson and Agrawal, 2005).

Since plant size and architecture have been shown to affect insect community composition strongly (Johnson and Agrawal, 2005), the fresh weight and the number of leaves of Rivera and Christmas Drumhead was monitored. Although plants of the two cultivars differed in these two traits, neither of these parameters were correlated with herbivore abundance, richness, and biodiversity and, therefore, are not likely to explain the observed differences in herbivore communities.

Lower numbers of P. rapae and M. brassicae larvae were found on Rivera than on Christmas Drumhead later in the season, suggesting differences in larval performance and/or oviposition preference between the cultivars. Indeed, under greenhouse conditions, P. rapae butterflies showed a higher preference for Christmas Drumhead than for Rivera (Poelman et al., 2009) and P. rapae larvae performed better when feeding on Christmas Drumhead compared to Rivera (Broekgaarden et al., 2007; Poelman et al., 2009). Larvae of M. brassicae also performed better on Christmas Drumhead than on Rivera under greenhouse conditions (Poelman et al., 2009).

Induced plant responses may not only affect the performance and host plant selection behaviour of the attacking herbivore, but also that of subsequently colonizing species (Shiojiri et al., 2002; Long et al., 2007). Initial infestations with P. rapae on Rivera negatively affected the performance of subsequently colonizing P. rapae and M. brassicae, as well as the preference of adult females from the latter species (Poelman et al., 2008).

Some remarkably large differences between Rivera and Christmas Drumhead were found for some of the herbivorous species. Several A. gamma larvae were found on Christmas Drumhead whereas this species was absent on Rivera, suggesting that butterflies of this species have a strong preference for Christmas Drumhead. This species does not completely avoid Rivera as A. gamma caterpillars were found on this cultivar in two previous years (Poelman et al., 2009; EH Poelman, JJA Van Loon, NM Van Dam, LEM Vet, M Dicke, unpublished data). Large differences in numbers of specialist whiteflies and aphids between Rivera and Christmas Drumhead were also found, ranging from (almost) zero to tens (>50) per plant, respectively. No pupae of the cabbage whitefly A. proletella have been observed on Rivera, whereas high numbers were found on Christmas Drumhead. This suggests a strong difference in host plant selection behaviour of whitefly females. Interestingly, the number of B. brassicae individuals on Rivera decreased, whereas population size of this aphid increased on Christmas Drumhead throughout the season. Both cultivars started with similar numbers of B. brassicae early in the season. In greenhouse experiments, this aphid was previously shown to be able to settle and reproduce on both cultivars (Broekgaarden et al., 2008), indicating that other factors play a role in this decrease in B. brassicae numbers under field conditions.

Herbivores not only differ in feeding strategy, but also in their degree of specialization. Specialists feed on one or a few closely related plant species, whereas generalists feed on many different plants (Schoonhoven et al., 2005). Specialists are adapted to host-plant specific characteristics such as defence compounds that are typically harmful to generalist herbivores (Ratzka et al., 2002; Wittstock et al., 2004; Kliebenstein et al., 2005; Després et al., 2007). In Brassica species, glucosinolates and their breakdown products stimulate specialists to oviposit and feed while they deter generalists (Renwick et al., 1992, 2006; Van Loon et al., 1992; Riggin-Bucci and Gould, 1996), which is in accordance with the observed higher abundance, species richness, and total herbivore mass of specialists compared to generalists in our study (Fig. 2). In addition, the specialists’ community differed more between the cultivars than that of generalists (Fig. 2), suggesting the differential induction of defence compounds between Rivera and Christmas Drumhead.

Intraspecific transcriptional variation in relation to differences in herbivore performance and behaviour

To investigate transcriptional responses that may underlie the observed differences in herbivore community composition it would be very convenient to have the ability to compare gene expression of control plants with that of plants challenged with naturally occurring herbivores. However, the treatments that would be needed to obtain herbivore-free plants in the field, for example, using nets or pesticides, have a direct effect on the study system. Using such controls introduces an unpredictable variable factor and does not contribute positively to the interpretation of the microarray results. Therefore, the ideal controls for the field situation are not available. Instead, to be able to get an impression of gene expression differences in the field, the transcriptional profiles of Rivera and Christmas Drumhead early and later in the season were compared.

Early in the season, no clear differences in gene expression levels could be detected between Rivera and Christmas Drumhead. Only a small number of genes showed differences in expression levels and none of them were related to defensive processes. Conversely, clear differences in gene expression levels between the cultivars were detected later in the season, which is correlated with the different development of herbivore communities throughout the season on the two cultivars. Although a relatively small number of genes showed differences in expression levels between the two cultivars, the genes that were differently expressed are interesting in relation to insect performance.

Later in the season, five defence-related genes showed higher levels of expression in Rivera than in Christmas Drumhead. One of these genes that probably play a central role in shaping the herbivore community is LOX2. It is likely that LOX2 in B. oleracea encodes a 13-LOX (Zheng et al., 2007), which is required for the first step in JA biosynthesis (Schaller et al., 2005; Wasternack et al., 2006). In A. thaliana, it has been shown that there is a strong correlation between the level of LOX2 expression and JA production (Spoel et al., 2003). Furthermore, RNA levels of this gene have been shown to increase in B. oleracea after JA treatment, wounding, and herbivore feeding (Broekgaarden et al., 2007; Zheng et al., 2007). Therefore, the higher expression level of LOX2 in Rivera than in Christmas Drumhead suggests that more JA accumulates in Rivera in the field. This is supported by the observation that 37% of the genes with higher expression levels in Rivera than in Christmas Drumhead are JA-responsive (Table 2). The fact that JA mediates direct defence by inducing secondary metabolites (Bruinsma et al., 2007; Van Dam et al., 2004) suggests that the absence of JA accumulation results in higher herbivore abundance and species richness. Indeed, Nicotiana attenuata plants that were artificially silenced in a 13-LOX gene harboured higher numbers of herbivores and were even attacked by a species that was never found on control plants in the field (Kessler et al., 2004). This indicates that altering JA accumulation can affect herbivore host selection and herbivore community composition (Kessler et al., 2004; Paschold et al., 2007; Halitschke et al., 2008).

The defence-related gene TPI, which encodes a trypsin-and-protease inhibitor, may also play an important role in the observed difference in herbivore community on Rivera and Christmas Drumhead. This gene is a member of the Kunitz trypsin inhibitor family that inhibits proteolytic enzymes within herbivore guts, resulting in reduced insect growth (Schuler et al., 1998; Marchetti et al., 2000). Silencing of TPI expression in A. thaliana increased P. rapae and B. brassicae performance in the greenhouse (Broekgaarden, 2008; C Broekgaarden, unpublished data). The higher expression level of TPI in Rivera compared to Christmas Drumhead is probably a result of the higher expression level of LOX2 in Rivera as TPI is JA-inducible (Broekgaarden et al., 2007).

The other three defence-related genes that showed higher levels of expression in Rivera compared to Christmas Drumhead may also affect herbivore performance and/or behaviour, resulting in differences in herbivore community composition. Lectins can function as defence proteins against herbivores (Peumans and Van Damme, 1995), Bet v 1 allergen protein is a member of the pathogenesis-related-10 family (Hoffmann-Sommergruber, 2000), and trypsin inhibitors can play a role in plant tolerance to herbivorous insects (Dunaevsky et al., 2005). However, qPCR needs to confirm the microarray results before more detailed studies can be done to determine the role of these genes in shaping herbivore communities.

Intraspecific transcriptional variation in the context of herbivore community composition

From the moment that the plants had been transplanted into the field they were exposed to all kinds of abiotic and biotic stresses such as temperature changes, rainfall, fungi, bacteria, and herbivorous insects that can all induce plant responses and, as a consequence, change gene expression. UV-B radiation, for example, has been shown to increase expression of jasmonate-signalling genes in field-grown Nicotiana longiflora (Izaguirre et al., 2003). Early in the season almost no differences in gene expression between the cultivars could be detected (Table 2A), which also holds for the composition of the herbivore community (Fig. 1). Conversely, clear differences in transcriptional profiles as well as in herbivore community composition between Rivera and Christmas Drumhead were observed later in the season (Table 2B; Fig. 1). The putative connection between gene expression and herbivore community composition is schematically shown in Fig. 6 and discussed below.

Fig. 6.

Schematic model to show the putative connections between the molecular and ecological data obtained later in the season (week 32). Differences in herbivore abundance may be related to differences in gene expression between the two B. oleracea cultivars. Based on comparisons of present field data with data from previous greenhouse studies (Broekgaarden et al., 2007, 2008) it is suggested that both cultivars induce the expression of certain defence-related genes (e.g. LOX2 and TPI), but Rivera stronger than Christmas Drumhead. (This figure is available in colour at JXB online.)

Fig. 6.

Schematic model to show the putative connections between the molecular and ecological data obtained later in the season (week 32). Differences in herbivore abundance may be related to differences in gene expression between the two B. oleracea cultivars. Based on comparisons of present field data with data from previous greenhouse studies (Broekgaarden et al., 2007, 2008) it is suggested that both cultivars induce the expression of certain defence-related genes (e.g. LOX2 and TPI), but Rivera stronger than Christmas Drumhead. (This figure is available in colour at JXB online.)

More than 50% of the genes that showed a higher level of expression in Rivera compared to Christmas Drumhead later in the season had previously been identified as P. rapae-responsive in greenhouse experiments (Broekgaarden et al., 2007). Furthermore, it has previously been shown that induction of gene expression upon P. rapae feeding lasts for at least three days (Broekgaarden et al., 2007) indicating that all kinds of changes occur in the plant upon herbivore attack beyond a few hours. These previous findings suggest that, besides other environmental factors, herbivore pressure may have a strong influence on shaping a plant's transcriptional profile in the field and thereby herbivore community composition. Indeed, initial P. rapae feeding on two B. oleracea cultivars resulted in differential regulation of gene expression upon feeding by sequential herbivores and this resulted in differential effects on performance and abundance of these herbivores on field-grown plants thereby affecting herbivore community composition (Poelman et al., 2008).

Zheng and co-workers (2007) have shown that just a single P. rapae larva can induce a fast increase in LOX2 transcript levels in B. oleracea. Our results show that the expression levels of the two defence-related genes LOX2 and TPI were higher in field-grown plants than in greenhouse-grown control plants for both cultivars and comparable to the levels in plants that were challenged for 24 h or 48 h by P. rapae under greenhouse conditions. However, the expression levels of LOX2 and TPI were not as high as those after 72 h of feeding by 10 P. rapae larvae. This shows that genes are not necessarily expressed to a maximum level, even when more than one P. rapae larva is present. The lower gene expression levels in the field may be the result of cross-talk between responses to many different signals. Different herbivores elicit very different transcriptional responses in plants (De Vos et al., 2005) that can have different effects on subsequent herbivores or pathogens (Agrawal, 2000; Heidel and Baldwin, 2004). For example, pre-infestation by P. rapae caterpillars affects the susceptibility to turnip crinkle virus in A. thaliana (De Vos et al., 2006). In N. attenuata, prior attack by sap-feeding mirids (Tupiocoris notatus) resulted in reduced performance of Manduca sexta (Voelckel and Baldwin, 2004). Thus, the induction of plant responses by herbivory affects subsequent attackers and is mediated by transcription-related changes in the plant (Kessler et al., 2004; Poelman et al., 2008). Unravelling the mechanisms underlying the dynamics of community composition is an exciting process that is now possible through a multidisciplinary approach that connects transcriptomics with metabolomics and community ecology (Kessler and Halitschke, 2007; Bruinsma and Dicke, 2008). To this end, silencing candidate genes, as identified by the microarray analysis, is needed to determine their individual contribution. Additionally, analysing segregating populations based on the two cultivars could lead to the identification of QTLs or expression QTLs (eQTLs) related to ecological communities.

Conclusions

Our results show that clear differences in herbivore community composition between two B. oleracea cultivars develop during the season. These differences could be related to differences in gene expression between the cultivars. While the herbivore populations and gene expression patterns were very similar early in the season, they became different for the two cultivars later in the season. Several defence-related genes showed higher levels of expression in the cultivar that harboured the lowest numbers of herbivores. These data provide an important step towards the analysis of the mechanisms that underlie the dynamics of ecological communities.

Supplementary data

Supplementary data can be found at JXB online.

Supplementary Fig. S1. Photographs of B. oleracea cultivars Rivera and Christmas Drumhead grown in the field.

We thank André Meijaard, Stephanie Laurent, Ciska Raaijmakers, Jeroen Jansen, Greet Steenhuis, Sylvia Lenting, and Aline Boursault for assistance with collecting material; CGN and Bejo Zaden for providing seeds of the cultivars; Unifarm for maintenance of the plants and field site; André Meijaard and Stephanie Laurent for assistance in the laboratory and data analysis. We thank an anonymous reviewer for valuable comments on a previous version. This research was supported by the Dutch Ministry of Agriculture, Nature, and Food Quality. MD was additionally supported by a VICI grant from The Netherlands Organization for Scientific Research, NWO (865.03.002).

References

Adler
LS
Schmitt
J
Bowers
MD
Genetic variation in defensive chemistry in Plantago lanceolata (Plantaginaceae) and its effect on the specialist herbivore Junonia coenia (Nymphalidae)
Oecologia
 , 
1995
, vol. 
101
 (pg. 
75
-
85
)
Agrawal
AA
Specificity of induced resistance in wild radish: causes and consequences for two specialist and two generalist caterpillars
Oikos
 , 
2000
, vol. 
89
 (pg. 
493
-
500
)
Agrawal
AA
Kurashige
NS
A role for isothiocyanates in plant resistance against the specialist herbivore Pieris rapae
Journal of Chemical Ecology
 , 
2003
, vol. 
29
 (pg. 
1403
-
1415
)
Ballhorn
DJ
Heil
M
Pietrowski
A
Lieberei
R
Quantitative effects of cyanogenesis on an adapted herbivore
Journal of Chemical Ecology
 , 
2007
, vol. 
33
 (pg. 
2195
-
2208
)
Bangert
RK
Allan
GJ
Turek
RJ
Wimp
GM
Meneses
N
Martinsen
GD
Keim
P
Whitham
TG
From genes to geography: a genetic similarity rule for arthropod community structure at multiple geographic scales
Molecular Ecology
 , 
2006
, vol. 
15
 (pg. 
4214
-
4228
)
Becher
M
Talke
IN
Kral
L
Krämer
U
Cross-species microarray transcript profiling reveals high constitutive expression of metal homeostasis genes in shoots of the zinc hyperaccumulator Arabidopsis halleri
The Plant Journal
 , 
2004
, vol. 
37
 (pg. 
251
-
268
)
Benjamini
Y
Hochberg
Y
Controlling the false discovery rate: a practical and powerful approach to multiple testing
Journal of the Royal Statistical Society: Series B
 , 
1995
, vol. 
57
 (pg. 
289
-
300
)
Bolter
CJ
Dicke
M
Van Loon
JJA
Visser
JH
Posthumus
MA
Role of volatiles from herbivore-damaged potato plants in attraction of Colorado potato beetle during herbivory and after its termination
Journal of Chemical Ecology
 , 
1997
, vol. 
23
 (pg. 
1003
-
1023
)
Broekgaarden
C
Poelman
EH
Steenhuis
G
Voorrips
RE
Dicke
M
Vosman
B
Genotypic variation in genome-wide transcription profiles induced by insect feeding: Brassica oleraceaPieris rapae interactions
BMC Genomics
 , 
2007
, vol. 
8
 pg. 
239
 
Broekgaarden
C
Poelman
EH
Steenhuis
G
Voorrips
RE
Dicke
M
Vosman
B
Responses of Brassica oleracea cultivars to infestation by the aphid Brevicoryne brassicae: an ecological and molecular approach
Plant, Cell and Environment
 , 
2008
, vol. 
31
 (pg. 
1592
-
1605
)
Bruinsma
M
Dicke
M
Schaller
A
Herbivore-induced indirect defense: from induction mechanisms to community ecology
Induced plant resistance to herbivory
 , 
2008
Berlin, Germany
Springer Verlag
(pg. 
31
-
60
)
Bruinsma
M
Van Dam
NM
Van Loon
JJA
Dicke
M
Jasmonic acid-induced changes in Brassica oleracea affect oviposition preference of two specialist herbivores
Journal of Chemical Ecology
 , 
2007
, vol. 
33
 (pg. 
655
-
668
)
Buckley
BA
Comparative environmental genomics in non-model species: using heterologous hybridization to DNA-based microarrays
Journal of Experimental Biology
 , 
2007
, vol. 
209
 (pg. 
1602
-
1606
)
Carroll
SB
Endless forms: the evolution of gene regulation and morphological diversity
Cell
 , 
2000
, vol. 
101
 (pg. 
577
-
580
)
Casteel
CL
O'Neill
BF
Zavala
JA
Bilgin
DD
Berenbaum
MR
DeLucia
EH
Transcriptional profiling reveals elevated CO2 and elevated O3 alter resistance of soybean (Glycine max) to Japanese beetles (Popillia japonica)
Plant, Cell and Environment
 , 
2008
, vol. 
31
 (pg. 
419
-
434
)
Davey
MW
Graham
NS
Vanholme
B
Swennen
R
May
ST
Keulemans
J
Heterologous oligonucleotide microarrays for transcriptomics in a non-model species; a proof-of-concept study of drought stress in Musa
BMC Genomics
 , 
2009
, vol. 
10
 pg. 
436
 
De Moraes
CM
Mescher
MC
Tumlinson
JH
Caterpillar-induced nocturnal plant volatiles repel conspecific females
Nature
 , 
2001
, vol. 
410
 (pg. 
577
-
580
)
De Vos
M
Van Oosten
VR
Van Poecke
RMP
, et al.  . 
Signal signature and transcriptome changes of Arabidopsis during pathogen and insect attack
Molecular Plant–Microbe Interactions
 , 
2005
, vol. 
18
 (pg. 
923
-
937
)
De Vos
M
Van Zaanen
W
Koornneef
A
Korzelius
JP
Van Loon
LC
Dicke
M
Pieterse
CMJ
Herbivore-induced resistance against microbial pathogens in Arabidopsis
Plant Physiology
 , 
2006
, vol. 
142
 (pg. 
352
-
363
)
Després
L
David
J-P
Gallet
C
The evolutionary ecology of insect resistance to plant chemicals
Trends in Ecology and Evolution
 , 
2007
, vol. 
22
 (pg. 
298
-
307
)
Dunaevsky
YE
Elpidina
EN
Vinokurov
KS
Belozersky
MA
Protease inhibitors in improvement of plant resistance to pathogens and insects
Molecular Biology
 , 
2005
, vol. 
39
 (pg. 
702
-
708
)
Fatouros
NE
Broekgaarden
C
Bukovinszkine'Kiss
G
Van Loon
JJA
Mumm
R
Huigens
ME
Dicke
M
Hilker
M
Male-derived butterfly anti-aphrodisiac mediates induced plant defense
Proceedings of the National Academy of Sciences, USA
 , 
2008
, vol. 
105
 (pg. 
10033
-
10038
)
Gao
L-L
Klinger
JP
Anderson
JP
Edwards
OR
Singh
KB
Characterization of pea aphid resistance in Medicago truncatula
Plant Physiology
 , 
2008
, vol. 
146
 (pg. 
996
-
1009
)
Halitschke
R
Stenberg
JA
Kessler
D
Kessler
A
Baldwin
IT
Shared signals: ‘alarm calls’ from plants increase apparency to herbivores and their enemies in nature
Ecology Letters
 , 
2008
, vol. 
11
 (pg. 
24
-
34
)
Heidel
AJ
Baldwin
IT
Microarray analysis of salicylic acid- and jasmonic acid-signalling in responses of Nicotiana attenuata to attack by insects from multiple feeding guilds
Plant, Cell and Environment
 , 
2004
, vol. 
27
 (pg. 
1362
-
1373
)
Hoffmann-Sommergruber
K
Plant allergens and pathogenesis-related proteins
International Archives of Allergy and Immunology
 , 
2000
, vol. 
122
 (pg. 
155
-
166
)
Howe
GA
Jander
G
Plant immunity to insect herbivores
Annual Review of Plant Biology
 , 
2008
, vol. 
59
 (pg. 
41
-
66
)
Hui
D
Iqbal
J
Lehmann
K
Gase
K
Saluz
HP
Baldwin
IT
Molecular interactions between the specialist herbivore Manduca sexta (Lepidoptera, Sphingidae) and its natural host Nicotiana attenuata. V. Microarray analysis and further characterization of large-scale changes in herbivore-induced mRNAs
Plant Physiology
 , 
2003
, vol. 
131
 (pg. 
1877
-
1893
)
Inbar
M
Gerling
D
Plant-mediated interactions between whiteflies, herbivores, and natural enemies
Annual Review of Entomology
 , 
2008
, vol. 
53
 (pg. 
431
-
448
)
Izaguirre
MM
Scopel
AL
Baldwin
IT
Ballaré
CL
Convergent responses to stress. Solar ultraviolet-B rediation and Manduca sexta herbivory elicit overlapping transcriptional responses in field-grown plants of Nicotiana longiflora
Plant Physiology
 , 
2003
, vol. 
132
 (pg. 
1755
-
1767
)
Johnson
MTJ
Agrawal
AA
Plant genotype and environment interact to shape a diverse arthropod community on evening primrose (Oenothera biennis)
Ecology and Systematics
 , 
2005
, vol. 
21
 (pg. 
243
-
273
)
Kaplan
I
Denno
RF
Interspecific interactions in phytophagous insects revised: a quantitative assessment of competition theory
Ecology Letters
 , 
2007
, vol. 
10
 (pg. 
977
-
994
)
Karban
R
Plant behaviour and communication
Ecology Letters
 , 
2008
, vol. 
11
 (pg. 
727
-
739
)
Kazana
E
Pope
TW
Tibbles
L
Bridges
M
Pickett
JA
Bones
AM
Powell
G
Rossiter
JT
The cabbage aphid: a walking mustard oil bomb
Proceedings of the Royal Society B: Biological Sciences
 , 
2007
, vol. 
274
 (pg. 
2271
-
2277
)
Kessler
A
Baldwin
IT
Plant response to insect herbivory: the emerging molecular analysis
Annual Review of Plant Biology
 , 
2002
, vol. 
53
 (pg. 
299
-
328
)
Kessler
A
Baldwin
IT
Herbivore-induced plant vaccination. Part I. The orchestration of plant defenses in nature and their fitness consequences in the wild tobacco Nicotiana attenuata
The Plant Journal
 , 
2004
, vol. 
28
 (pg. 
639
-
649
)
Kessler
A
Halitschke
R
Specificity and complexity: the impact of herbivore-induced plant responses on arthropod community structure
Current Opinion in Plant Biology
 , 
2007
, vol. 
10
 (pg. 
409
-
414
)
Kessler
A
Halitschke
R
Baldwin
IT
Silencing the jasmonate cascade: induced plant defenses and insect population
Science
 , 
2004
, vol. 
305
 (pg. 
665
-
668
)
Kliebenstein
DJ
Kroymann
J
Mitchell-Olds
T
The glucosinolate-myrosinase system in an ecological and evolutionary context
Current Opinion in Plant Biology
 , 
2005
, vol. 
8
 (pg. 
264
-
271
)
Korth
KL
Profiling the response of plants to herbivorous insects
Genome Biology
 , 
2003
, vol. 
4
 pg. 
221
 
Kuśnierczyk
A
Winge
P
Midelfart
H
Armbruster
WS
Rossiter
JT
Bones
AM
Transcriptional responses of Arabidopsis thaliana ecotypes with different glucosinolate profiles after attack by polyphagous Myzus persicae and oligophagous Brevicoryne brassicae
Journal of Experimental Botany
 , 
2007
, vol. 
58
 (pg. 
2537
-
2552
)
Lee
HS
Wang
JL
Tian
L
, et al.  . 
Sensitivity of 70-mer oligonucleotides and cDNAs for microarray analysis of gene expression in Arabidopsis and its related species
Plant Biotechnology Journal
 , 
2004
, vol. 
2
 (pg. 
45
-
57
)
Long
JD
Hamilton
RS
Mitchell
JL
Asymmetric competition via induced resistance: specialist herbivores indirectly suppress generalist preference and populations
Ecology
 , 
2007
, vol. 
88
 (pg. 
1232
-
1240
)
Marchetti
S
Delledonne
M
Fogher
C
Chiaba
C
Chiesa
F
Savazzini
F
Giordano
A
Soybean Kunitz C-II and PI-IV inhibitor genes confer different levels of insect resistance to tobacco and potato transgenic plants
Theoretical and Applied Genetics
 , 
2000
, vol. 
101
 (pg. 
519
-
526
)
Mendes
RS
Evangelista
LR
Thomaz
SM
Agostinho
AA
Gomes
LC
A unified index to measure ecological diversity and species rarity
Ecography
 , 
2008
, vol. 
31
 (pg. 
450
-
456
)
Mitchell
ND
Richards
AJ
Brassica oleracea L. ssp. oleracea (B. sylvestris (L.) Miller)
Journal of Ecology
 , 
1979
, vol. 
67
 (pg. 
1087
-
1096
)
Paschold
A
Halitschke
R
Baldwin
IT
Co(i)-ordinated defenses: NaCOI1 mediates herbivore-induced resistance in Nicotiana attenuata and reveals the role of herbivore movement in avoiding defenses
The Plant Journal
 , 
2007
, vol. 
51
 (pg. 
79
-
97
)
Peumans
WJ
Van Damme
EJ
Lectins as plant defense proteins
Plant Physiology
 , 
1995
, vol. 
109
 (pg. 
347
-
352
)
Pieterse
CMJ
Dicke
M
Plant interactions with microbes and insects: from molecular mechanisms to ecology
TRENDS in Plant Science
 , 
2007
, vol. 
12
 (pg. 
564
-
569
)
Poelman
EH
Broekgaarden
C
Van Loon
JJA
Dicke
M
Early-season herbivore differentially affects plant defence responses to subsequently colonizing herbivores and their abundance in the field
Molecular Ecology
 , 
2008
, vol. 
17
 (pg. 
3352
-
3365
)
Poelman
EH
Van Dam
NM
Van Loon
JJA
Vet
LEM
Dicke
M
Chemical diversity in Brassica oleracea affects biodiversity of insect herbivores
Ecology
 , 
2009
, vol. 
90
 (pg. 
1863
-
1877
)
Ratzka
A
Vogel
H
Kliebenstein
DJ
Mitchell-Olds
T
Kroymann
J
Disarming the mustard oil bomb
Proceedings of the National Academy of Sciences, USA
 , 
2002
, vol. 
99
 (pg. 
11223
-
11228
)
Relogio
A
Schwager
C
Richter
A
Ansorge
W
Valcarcel
J
Optimization of oligonucleotide-based DNA microarrays
Nucleic Acids Research
 , 
2002
, vol. 
30
 pg. 
e51
 
Renwick
JAA
Radke
CD
Sachdev-Gupta
K
Städler
E
Leaf surface chemicals stimulating oviposition by Pieris rapae (Lepidoptera: Pieridae) on cabbage
Chemoecology
 , 
1992
, vol. 
3
 (pg. 
33
-
38
)
Renwick
JAA
Haribal
M
Gouinguené
S
Städler
E
Isothiocyanates stimulating oviposition by the diamondback moth, Plutella xylostella
Journal of Chemical Ecology
 , 
2006
, vol. 
32
 (pg. 
755
-
766
)
Reymond
P
Bodenhausen
N
Van Poecke
RMP
Krishnamurthy
V
Dicke
M
Farmer
EE
A conserved transcript pattern in response to a specialist and a generalist herbivore
The Plant Cell
 , 
2004
, vol. 
16
 (pg. 
3132
-
3147
)
Riggin-Bucci
TM
Gould
F
Effect of surfactants, Bacillus thuringiensis formulations, and plant damage on oviposition by diamondback moth (Lepidoptera: Plutellidae)
Journal of Economic Entomology
 , 
1996
, vol. 
89
 (pg. 
891
-
897
)
Rishi
AS
Nelson
ND
Goyal
A
DNA microarrays: gene expression profiling in plants
Reviews in Plant Biochemistry and Biotechnology
 , 
2002
, vol. 
1
 (pg. 
81
-
100
)
Root
RB
Organization of a plant–arthropod association in simple and diverse habitats: the fauna of collards (Brassica oleracea)
Ecological Monographs
 , 
1973
, vol. 
43
 (pg. 
95
-
120
)
Schaller
F
Schaller
A
Stintzi
A
Biosynthesis and metabolism of jasmonates
Journal of Plant Growth Regulation
 , 
2005
, vol. 
23
 (pg. 
179
-
199
)
Schmidt
DD
Baldwin
IT
Transcriptional responses of Solanum nigrum to methyl jasmonate and competition: a greenhouse and field study
Functional Ecology
 , 
2006
, vol. 
20
 (pg. 
500
-
508
)
Schoonhoven
LM
van Loon
JJA
Dicke
M
Insect–plant biology
 , 
2005
2nd edn
New York
Oxford University Press
Schuler
TH
Poppy
GM
Kerry
BR
Denholm
I
Insect-resistant transgenic plants
Trends in Biotechnology
 , 
1998
, vol. 
16
 (pg. 
168
-
175
)
Shiojiri
K
Takabayashi
J
Yano
S
Takafuji
A
Oviposition preferences of herbivores are affected by tritrophic interaction webs
Ecology Letters
 , 
2002
, vol. 
5
 (pg. 
186
-
192
)
Smith
CM
Boyko
EV
The molecular bases of plant resistance and defense responses to aphid feeding: current status
Entomologia Experimentalis et Applicata
 , 
2007
, vol. 
122
 (pg. 
1
-
16
)
Spoel
SH
Koornneef
A
Claessens
SMC
, et al.  . 
NPR1 modulates cross-talk between salicylate- and jasmonate-dependent defense pathways through a novel function in the cytosol
The Plant Cell
 , 
2003
, vol. 
15
 (pg. 
760
-
770
)
Steppuhn
A
Gase
K
Krock
B
Halitschke
R
Baldwin
IT
Nicotine's defensive function in nature
PLOS Biology
 , 
2004
, vol. 
2
 (pg. 
1074
-
1080
)
Thompson
GA
Goggin
FL
Transcriptomics and functional genomics of plant defence induction by phloem-feeding insects
Journal of Experimental Botany
 , 
2006
, vol. 
57
 (pg. 
755
-
766
)
Traw
MB
Dawson
TE
Reduced performance of two specialist herbivores (Lepidoptera: Pieridae, Coleoptera: Chrysomelidae) on new leaves of damaged black mustard plants
Environmental Entomology
 , 
2002
, vol. 
31
 (pg. 
714
-
722
)
Van Dam
NM
Witjes
L
Svatos
A
Interactions between aboveground and belowground induction of glucosinolates in two wild Brassica species
New Phytologist
 , 
2004
, vol. 
161
 (pg. 
801
-
810
)
Van Loon
JJA
Blaakmeer
A
Griepink
FC
Van Beek
TA
Schoonhoven
LM
De Groot
A
Leaf surface compound from Brassica oleracea (Cruciferae) induces oviposition by Pieris brassicae (Lepidoptera: Pieridae)
Chemoecology
 , 
1992
, vol. 
3
 (pg. 
39
-
44
)
Voelckel
C
Baldwin
IT
Herbivore-induced plant vaccination. Part II. Array-studies reveal the transience of herbivore-specific transcriptional imprints and a distinct imprint from stress combinations
The Plant Journal
 , 
2004
, vol. 
38
 (pg. 
650
-
663
)
Walling
LL
The myriad plant responses to herbivores
Journal of Plant Growth Regulation
 , 
2000
, vol. 
19
 (pg. 
195
-
216
)
Wasternack
C
Stenzel
I
Hause
B
Hause
G
Kutter
C
Maucher
H
Neumerkel
J
Feussner
I
Miersch
O
The wound response in tomato: role of jasmonic acid
Journal of Plant Physiology
 , 
2006
, vol. 
163
 (pg. 
297
-
306
)
Whitham
TG
Bailey
JK
Schweitzer
JA
, et al.  . 
A framework for community and ecosystem genetics: from genes to ecosystems
Nature Reviews Genetics
 , 
2006
, vol. 
7
 (pg. 
510
-
523
)
Wimp
GM
Martinsen
GD
Floate
KD
Bangert
RK
Whitham
TG
Plant genetic determinants of arthropod community structure and diversity
Evolution
 , 
2005
, vol. 
59
 (pg. 
61
-
69
)
Wittstock
U
Agerbirk
N
Stauber
EJ
Olsen
CE
Hippler
M
Mitchell-Olds
T
Gershenzon
J
Vogel
H
Succesful herbivore attack due to metabolic diversion of a plant chemical defense
Proceedings of the National Academy of Sciences, USA
 , 
2004
, vol. 
101
 (pg. 
4859
-
4864
)
Wu
J
Hettenhausen
C
Schuman
MC
Baldwin
IT
A comparison of two Nicotiana attenuata accessions reveals large differences in signaling induced by oral secretions of the specialist herbivore Manduca sexta
Plant Physiology
 , 
2008
, vol. 
146
 (pg. 
927
-
939
)
Zheng
S-J
van Dijk
JP
Bruinsma
M
Dicke
M
Sensitivity and speed of induced defense of cabbage (Brassica oleracea L.): dynamics of BoLOX expression patterns during insect and pathogen attack
Molecular Plant–Microbe Interactions
 , 
2007
, vol. 
20
 (pg. 
1332
-
1345
)
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/2.5), which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.This paper is available online free of all access charges (see http://jxb.oxfordjournals.org/open_access.html for further details)

Comments

0 Comments