Skip to main content
Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

Unbiased Transcriptional Comparisons of Generalist and Specialist Herbivores Feeding on Progressively Defenseless Nicotiana attenuata Plants

Abstract

Background

Herbivore feeding elicits dramatic increases in defenses, most of which require jasmonate (JA) signaling, and against which specialist herbivores are thought to be better adapted than generalist herbivores. Unbiased transcriptional analyses of how neonate larvae cope with these induced plant defenses are lacking.

Methodology/Principal Findings

We created cDNA microarrays for Manduca sexta and Heliothis virescens separately, by spotting normalized midgut-specific cDNA libraries created from larvae that fed for 24 hours on MeJA-elicited wild-type (WT) Nicotiana attenuata plants. These microarrays were hybridized with labeled probes from neonates that fed for 24 hours on WT and isogenic plants progressively silenced in JA-mediated defenses (N: nicotine; N/PI: N and trypsin protease inhibitors; JA: all JA-mediated defenses). H. virescens neonates regulated 16 times more genes than did M. sexta neonates when they fed on plants silenced in JA-mediated defenses, and for both species, the greater the number of defenses silenced in the host plant (JA > N/PI > N), the greater were the number of transcripts regulated in the larvae. M. sexta larvae tended to down-regulate while H. virescens larvae up- and down-regulated transcripts from the same functional categories of genes. M. sexta larvae regulated transcripts in a diet-specific manner, while H. virescens larvae regulated a similar suite of transcripts across all diet types.

Conclusions/Significance

The observations are consistent with the expectation that specialists are better adapted than generalist herbivores to the defense responses elicited in their host plants by their feeding. While M. sexta larvae appear to be better adapted to N. attenuata's defenses, some of the elicited responses remain effective defenses against both herbivore species. The regulated genes provide novel insights into larval adaptations to N. attenuata's induced defenses, and represent potential targets for plant-mediated RNAi to falsify hypotheses about the process of adaptation.

Introduction

The co-evolution of plants and insects has primarily been driven by their interactions [1], [2], [3], [4]. Plants respond to herbivore attack with highly evolved, elegantly regulated arrays of responses. Attack triggers at least two types of inducible defense responses: those that involve the production of metabolites that directly retard the growth and development of the herbivores (direct defenses) and those that involve the production of metabolites that indirectly protect plants by attracting the herbivores' natural enemies, usually parasitoids and predators (indirect defenses) [5], [6], [7]. These inducible defense mechanisms are tightly regulated by insect elicitors, likely to curtail the costs of production in the absence of herbivory and to prevent insects from adapting to the plant's defenses.

The elicitors found in the oral secretions and regurgitants (OS) of the caterpillars enable plants to specifically recognize attack from insects; this recognition is mediated by complex signaling pathways in which jasmonic acid (JA) plays a central role [7], [8], [9]. The small quantities of OS that are transferred to leaves by the larvae during feeding are sufficient to elicit defense responses [10]. The salivary components are complex and differ among different insect species and some of these differences allow plants to tailor their defense responses to attack from different insect species [11], [12], [13]. The defense responses elicited in plants by these elicitors are also highly complex, frequently involving the production of metabolites from many different biosynthetic pathways that sometimes function synergistically to confer resistance [14], [15], [16], [17], [18]. The attacking insects, on the other hand, do not remain passive, but up-regulate detoxification systems and employ various behavioral responses to counter the plant's defense responses [18], [19], [20]. These counter defense responses are particularly well studied in specialist herbivores that have adapted to their host plant's defenses.

While generalist herbivores are often deterred by the secondary metabolites produced by their host plants, numerous studies reported that many specialists have evolved effective countermeasures [21], [22], [23]. Specialist herbivores often have well-developed specific enzymatic systems that allow them to metabolize secondary chemicals; for example, the specialist bruchid beetle metabolizes toxic non-protein amino acids and synthesizes its own amino acids [24], [25], and the larvae of the specialist Heliconius convert cyanogenic glycosides to thiols, which they use as a source of nitrogen [26]. Larvae of the lepidopteran Papilio and Helicoverpa genera metabolize furanocoumarins with the help of cytochrome-P450-dependent mono-oxygenases [27], [28], [29], [30] and Manduca sexta (Ms) larvae have developed a greater tolerance for nicotine that exceeds that of other insect species that do not regularly attack nicotine-containing plants [31], [32], [33]. Some specialists can sequester toxic secondary metabolites for their own defense, such as the larvae of the lepidopteran Uthesia ornatrix, which sequesters pyrrolizidine alkaloids to defend themselves against predators [34], [35], [36], [37].

The majority of these examples of plant induced defenses and insect counter responses come from studies that examine the responses of adapted insects to single compounds or classes of compounds. Few have examined the responses of insects in an unbiased fashion to the full complement of defenses that are elicited by herbivore attack. This stands in contrast to the many unbiased transcriptional analyses of plant responses to insect attack [38], [39], [40], [41], [42], [43].

Nicotiana attenuata, a wild tobacco plant native to the North American Great Basin Desert is regularly attacked by the larvae of specialist (Ms), as well as generalist (Heliothis virescens: Hv) lepidopteran herbivores [44]. This plant's responses to attack from Ms larvae have been extensively studied and is known to activate a well-characterized mitogen-activated protein kinase (MAPK) signaling system as soon as it perceives the elicitors, fatty acid-amino acid conjugates, in the OS of Ms larvae. This signaling system subsequently triggers JA-, SA-, and ethylene-mediated defense responses [9], [13], [45], [46] that include the accumulation of anti-feedants and secondary metabolites, such as trypsin protease inhibitors (TPIs), nicotine (N), phenolics, putrescine conjugates, diterpene glycosides, etc. [38], [42], [47], [48], [49], [50]. Microarray analysis with a custom microarray enriched in Ms-elicited N. attenuata genes revealed that there is a large overlap in the transcriptional response to attack from Ms and Hv larvae [51] despite differences in the biochemical composition of these insects' OS [11], [52], [53].

The genes up-regulated in N. attenuata by Ms and Hv herbivory include the 13-lipoxygenase, α-dioxygenase, hydroperoxide lyase, TPI, threonine deaminase, xyloglucan-endotransglycosylase and a WRKY-type transcription factor; all of these are part of N. attenuata's defense response [46], [48], [54], [55], [56], [57]. However, N. attenuata plants also elicit different transcriptional signatures in response to attack from these two herbivores. Although attack from both Ms and Hv larvae activate JA signaling, Ms larvae are more tolerant than Hv larvae to the defenses that are activated by JA signaling. It is not clear, however, whether these differences result from differences in the responses elicited in the plants by the two species, and/or differences in how these insects respond to the plant's defenses. One way to disentangle these two possibilities is to transform plants to progressively silence their defense responses and query the transcriptional responses of larvae feeding on these progressively defenseless plants in an unbiased manner.

The extensive literature on the physiological mechanisms that Ms larvae employ to detoxify N, such as the ‘multi-drug’ pump [58] and P450 enzymes [59], [60], may explain the differences in the response of Ms and Hv larvae to N. attenuata's defenses, but most of these mechanisms have not been studied in Hv. Moreover, it's not clear if these mechanisms are relevant for Ms's tolerance of dietary N as most studies use late instar larvae reared on artificial diets, as their large size greatly facilitate physiological examinations. In nature, however, it is the neonates that must adapt to the specific-defenses of the plant on which their mother oviposited them on. And this process of adaptation must be rapid, as larvae must tolerate the defenses that are elicited during their first meal. An analysis of the transcriptional responses of neonates that are activated in midgut tissues in response to their first meal are likely to reflect the mechanisms by which larvae adapt to host plant defenses [61].

The goal of this study was to compare the global changes in gene expression elicited by the feeding-elicited defenses in the host plant in two native herbivores of N. attenuata: the specialist larvae of Ms and generalist larvae of Hv. To investigate these transcriptional responses, we created cDNA microarrays for each larval species separately by spotting normalized cDNA libraries created from the midguts of neonates that had fed for 24 h on WT N. attenuata plants that had been previously elicited by MeJA treatment three days before larval feeding. These microarrays were then hybridized with labeled probes created from RNA extracted from neonates that had fed for 24 h on N. attenuata plants progressively silenced in JA-mediated defenses (N, N/PI, all JA-mediated defenses) and compared with probes from neonates that had fed for the same time on WT plants with their full complement of defenses. From the differences in expression patterns in the larvae that fed on the different diets, we draw inferences about the mechanisms by which the larvae adapt to N. attenuata's induced defenses.

Results

Experimental Overview

Both Ms and Hv larvae introduce some of the same fatty acid-amino acid conjugate elicitors from their OS into plant wounds during feeding and these elicitors induces a specific lipoxygenase (lox3), which catalyzes the oxygenation of α-linolenic acid to 13-hydroperoxides that undergoes further sequential enzymatic reactions to eventually produce JA and its metabolites. These jasmonates, in turn, activate the expression of both direct and indirect defenses. To study the transcriptional changes in Ms and Hv larvae to JA-mediated defenses elicited and present in their first meal, the larvae of both species were fed on WT plants and defenseless transgenic plants that are progressively silenced in nicotine alone (ir-pmt: N); N and TPIs (ir-pmt/pi: N/PI); and JA biosynthesis (as-lox3: JA), which are impaired in all JA-elicited direct and indirect defenses. Labeled probes prepared from larvae that fed for 24 h on WT and the different, progressively defenseless plants were hybridized to microarrays on which normalized midgut-specific cDNA libraries of Ms and Hv neonates were spotted separately to identify differentially regulated genes (up-regulated: expression ratio >1.5; down-regulated: expression ratio <−1.5) (Figure 1A).

thumbnail
Figure 1. Overview of the strategy used to isolate differentially regulated genes in larvae from a specialist (Manduca sexta: Ms) and a generalist (Heliothis virescens: Hv) lepidopteran herbivore of Nicotiana attenuata.

A) Neonates of Ms and Hv fed for 24 h on wild-type (WT), untransformed plants and plants transformed to silence: only nicotine (N) biosynthesis, by expressing an inverted-repeat putrescine N-methyl transferase (pmt) construct (ir-pmt); both N and trypsin protease inhibitor (N/PI) biosynthesis, by expressing an inverted-repeat pmt and trypsin protease inhibitor (PI) construct (ir-pmt/pi); and jasmonic acid (JA) biosynthesis, by antisense-expression of lipoxygenase 3 (as-lox3), a specific lox that supplies the JA biosynthetic cascade with fatty acid hydroperoxides. RNA was extracted from the whole larval tissue after neonates fed for 24 h, transcribed into cDNA and labeled with fluorescent dyes (WT-fed larvae: Cy5; N-, N/PI-, JA- fed larvae: Cy3) and hybridized against Manduca and Heliothis microarrays which had been spotted with midgut specific cDNA library of each species of larvae, normalized using duplex-specific nuclease in the trimmer-direct cDNA normalization kit. Differentially regulated genes (ER: Cy3/Cy5>1.5 or −1/(Cy3/Cy5) <−1.5) were identified, sequenced and analyzed for their putative function. B) The total number of genes regulated by Hv larvae (tentative unique genes: TUGs) was 16 times greater than that regulated by Ms larvae when both fed on the same defenseless plants, but the percentage of genes in contigs of various sizes (inset) was similar, because both the libraries had been normalized. Graphs show the distribution and percentage of differentially regulated TUGs by Ms and Hv larvae in contigs of different sizes. Solid bars: Ms, open bars: Hv.

https://doi.org/10.1371/journal.pone.0008735.g001

From these hybridizations, we identified 151 expressed sequence tags (ESTs) which were differentially regulated in Ms larvae that could be grouped into 108 tentative unique genes (TUGs). In contrast, Hv larvae differentially regulated 2011 ESTs that could be grouped into 1739 TUGs. Hence, when feeding on the same defenseless plants, Hv larvae regulated 16 times more transcripts than did Ms larvae (Figure 1B). Despite the large difference in the number of transcripts regulated by Ms and Hv larvae, the percentage of singletons −78.7% and 75.6% from Ms and Hv larvae, respectively – was almost the same, verifying the normalization of the libraries (Figure 1B, inset).

Overarching Patterns of Differential Transcript Accumulation

While Hv larvae regulated more transcripts than did Ms larvae, the overall pattern of regulation in both species tracked the number of defenses that were silenced in the host: the greater the number of defenses silenced in the host plant (JA>N/PI>N), the greater the number of transcripts that were differentially regulated in larvae feeding on these silenced host plants in comparison to larvae that fed on wild-type (WT) plants (Figure 2A). Moreover, the type of regulation differed between the two species: Ms larvae down-regulated 2–5 times more transcripts than they up-regulated, but in Hv larvae, an equal number of genes were up- and down-regulated. Hence, Ms larvae that fed on N-silenced plants regulated 24 transcripts of which 8 were up-regulated and 16 were down-regulated; when they fed on N/PI-silenced plants, they regulated 30 transcripts of which 5 were up-regulated and 25 were down-regulated; and on JA-silenced plants, they regulated 73 transcripts of which 22 were up-regulated and 51 were down-regulated. Hv larvae that fed on N-, N/PI- and JA-silenced plants regulated 914 transcripts (493 up-regulated and 421 down-regulated), 1006 transcripts (567 up-regulated and 439 down-regulated) and 1228 transcripts (716 up-regulated and 512 down-regulated), respectively (Figure 2A).

thumbnail
Figure 2. The number of genes regulated by both Ms and Hv larvae correlated with the number of defense traits that had been silenced in the larval host plants, but more genes were regulated in Hv than in Ms larvae.

A) Venn diagrams depicting the number and percentage of genes significantly up- (ER>1.5) or down-regulated (ER<−1.5) by Ms and Hv larvae according to which defense compound was silenced in the plants they fed on. When the generalist Hv larvae fed on JA-, N/PI-, or N-silenced N. attenuata plants 17-, 33- and 38-fold more genes were regulated compared to when the specialist, Ms larvae fed on the same plants. Values in parentheses denote the percentage of regulated genes. For both species, the number of genes and the percentage of regulated genes correlated with the number of different direct defenses that had been silenced in the plants: more genes were found to be regulated in larvae that fed on JA-silenced plants than in larvae that fed on N/PI- and N-silenced plants. B) In Ms larvae, the percentage of genes coding for enzymes, transporters and metabolism was highly regulated; in Hv larvae, the percentage of genes coding for genetic information followed by cellular processes and environmental information processing was highly regulated.

https://doi.org/10.1371/journal.pone.0008735.g002

By sequencing these differentially regulated transcripts and mapping them to biochemical pathways, we discovered that Ms larvae that fed on defenseless plants regulated more transcripts coding for enzymes (21%), followed by transporters (11%) and metabolism (21%) compared to Hv larvae that fed on these defenseless plants (9%, 4% and 19%, respectively). On the other hand, Hv larvae regulated more transcripts coding for genetic information processing (12%), followed by cellular processes (7%) and environmental information processing (6%) compared to Ms larvae (6%, 6% and 5%, respectively) (Figure 2B).

In the following sections, we discuss the differences in the regulation of transcripts in the functional categories of primary and secondary metabolism, peptidase and hydrolases, transporters, and genetic information processing observed in Ms and Hv larvae that fed on defenseless plants. Two general patterns of regulation were observed. First, while Ms larvae down-regulated, Hv larvae up- and down-regulated transcripts from the same functional categories. Second, while Ms larvae regulated transcripts in a diet-specific manner, Hv larvae regulated a more similar suite of transcripts across all diet types (Figures 3, 4, 5, 6, upper and lower panels; Tables S1 and S2).

thumbnail
Figure 3. Both Ms and Hv larvae regulated more genes related to primary and secondary metabolism when they fed on JA-silenced plants than when they fed on N/PI- or N-silenced plants.

In Ms larvae, these genes were mainly down-regulated while in Hv larvae, they were both up- and down-regulated. A) Various functional categories of genes related to primary metabolism were highly regulated in Hv larvae, while only a few were regulated in Ms larvae; however, Ms larvae that fed on JA-silenced plants regulated 1.9-fold more genes than did Hv larvae that fed on the same plants. B) In Ms larvae, most genes related to secondary metabolism were down-regulated (with percentages down-regulated in larvae fed on JA-silenced plants being the highest followed by N/PI- and N-silenced plants), while in Hv larvae, these genes were both up- and down-regulated. Ms larvae that fed on JA-silenced plants regulated 4.4-fold more genes than did Hv larvae that fed on the same plants. Open bars: up-regulated (ER>1.5); black bars: down-regulated (ER<−1.5). Inset: Venn diagrams depicting the percentage of genes regulated by larvae in response to JA-, N/PI or N-silenced diets; Ms (upper panel) and Hv (lower panel) for primary (A) and secondary (B) metabolism.

https://doi.org/10.1371/journal.pone.0008735.g003

thumbnail
Figure 4. In Ms larvae, a higher percentage of peptidases and hydrolases were down-regulated when they fed on plants silenced in JA compared to when they fed on plants silenced in N/PI, and this percentage of down-regulated enzymes was lowest in plants silenced in N; the regulation of these enzyme categories by Hv larvae did not follow a pattern organized by defense expression in the host plant.

The percentage of genes regulated for A) peptidases was 2.4-fold greater and the B) hydrolases, 5.6-fold greater in Ms larvae that fed on plants silenced for JA compared to in Hv larvae that fed on the same plants. C) Oxidoreductases, transferases and isomerases (the category referred to as ‘other enzyme’) were regulated only in Hv larvae when these fed on JA-, N/PI- or N-silenced plants. Open bars: up-regulated (ER>1.5); solid bars: down-regulated (ER<−1.5). Inset: Venn diagrams depicting the percentage of genes regulated by Ms larvae (upper panel) and Hv larvae (lower panel) for peptidases (A), hydrolases (B) and other enzymes (C) in response to feeding on variously defenseless plants.

https://doi.org/10.1371/journal.pone.0008735.g004

thumbnail
Figure 5. Only a few transporters (mainly MFS and ABC) were down-regulated by Ms larvae when the larvae fed on plants silenced for JA-, N/PI- or N; Hv larvae up- and down-regulated many transporters.

Overall, many transporter genes were down-regulated in Ms larvae that fed on JA-silenced plants, fewer were regulated in larvae that fed on N/PI- and N-silenced plants. Major facilitator superfamily transporters and lipid transporters were down-regulated in Ms larvae that fed on JA-silenced plants. Ms larvae that fed on N/PI- and N-silenced plants down-regulated their ATP-binding cassette (ABC) superfamily of transporter and ion-coupled transporters (upper panel). Few transporter genes were regulated in Hv larvae that fed on JA-, N/PI- or N-silenced plants (lower panel). Open bars: up-regulated (ER>1.5); solid bars: down-regulated (ER<−1.5).

https://doi.org/10.1371/journal.pone.0008735.g005

thumbnail
Figure 6. In Ms larvae genes for genetic information processing were regulated only when larvae fed on JA-silenced plants; Hv larvae regulated genes for genetic information processing when they fed on JA-, N/PI- and N-silenced plants.

Open bars: up-regulated (ER>1.5); solid bars: down-regulated (ER<−1.5) by Ms (upper panel) and Hv (lower panel).

https://doi.org/10.1371/journal.pone.0008735.g006

Primary Metabolism

When larvae fed on JA-silenced plants, they regulated many genes involved in primary metabolism, which included transcripts that encode for proteins involved in the metabolism of carbohydrates, lipids, proteins and amino acids, nucleotides, energy, cofactors, vitamins, and glycans. Ms and Hv larvae that fed on JA-silenced plants regulated 3 and 1.4 times more transcripts for primary metabolism than larvae that fed on N- and N/PI-silenced plants, respectively. For example, Ms larvae that fed on JA-silenced plants regulated 2 times more transcripts for carbohydrate and amino acid metabolism, and 4 times more transcripts for carbohydrate metabolism than larvae that fed on N/PI- and N-silenced plants, respectively (Figure 3A, upper panel). The regulation was more pronounced in Ms larvae which regulated 1.9 times more transcripts than did Hv larvae in response to the overall JA-mediated defenses excluding N and N/PI (Figure 3A, inset, upper and lower panels). Although more transcripts for primary metabolism were regulated in Ms larvae, they regulated only specific pathways, while Hv larvae regulated transcripts for all primary metabolic pathways. For example, none of the transcripts for energy metabolism were regulated in Ms larvae that fed on defenseless plants (Figure 3A, upper and lower panels). Regulation in Ms larvae was highly diet-specific as observed by the regulation of transcripts for purine and pyrimidine metabolism, and cofactor and vitamin metabolism, which were not regulated by larvae that fed on N-silenced plants (Figure 3A, upper panel).

Regulation of genes within a given primary metabolic pathway in Ms larvae was also highly diet-specific. For example, transcripts for galactose metabolism were regulated only by Ms larvae that fed on N-silenced plants and transcripts for starch and sucrose, cysteine and tryptophan, and folate metabolism, were regulated only by Ms larvae that fed on JA-silenced plants (Table S1). However, all of these transcripts were regulated in Hv larvae that fed on N-, N/PI- and JA-silenced plants; in addition, they regulated a few transcripts not regulated by Ms larvae. For example, they regulated transcripts for energy metabolism, metabolism of other amino acids in addition to cysteine and tryptophan, and a few other pathways of carbohydrate and lipid metabolism not regulated by Ms larvae (Figure 3A, lower panel, Table S2). Overall, these results suggest that a suite of genes involved in primary metabolism are associated with the dramatic increase in larval growth that occurs when larvae feed on plants that lack all JA-mediated defenses.

Secondary Metabolism

As expected, larvae that fed on plants silenced in JA-mediated defenses with altered levels of various secondary metabolites regulated some of the well-described detoxification enzymes, such as cytochrome P450, glutathione S-transferase and carboxylesterase, and other transcripts that code for proteins involved in the metabolism of plant secondary metabolites.

Ms larvae are known to be more tolerant to the synergistic defense of N and PI than Hv larvae [14], and this greater tolerance is reflected in the 1.2 fold greater number of transcripts regulated by Hv larvae than in Ms larvae when they fed on N/PI-silenced plants (Figure 3B, inset, upper and lower panels). However, Ms larvae regulated more of their transcripts for secondary metabolism in response to feeding on JA-silenced plants, and the response was much greater than was observed in Hv larvae. Hence, Ms larvae that fed on JA-silenced plants regulated 2–3 times more transcripts for secondary metabolism than did Ms larvae that fed on N- or N/PI-silenced plants (Figure 3B, inset, upper panel), and regulated 2 times more than Hv larvae that fed on JA-silenced plants (Figure 3B, inset, upper and lower panels). These results suggest that while Ms larvae have effectively adapted to dietary N and TPIs. N. attenuata produces other JA-regulated defenses, such as DTGs [49], which require substantial metabolic efforts on the part of the larvae to detoxify.

While the response varied considerably between Ms and Hv larvae, the two general patterns of regulation observed in genes involved in primary metabolism (up- and down-regulation, and diet-specific regulation) were also observed in the regulation of genes involved in secondary metabolism. First, while Ms larvae down-regulated genes in response to feeding on defenseless plants, Hv larvae up- and down-regulated genes from the same functional categories (Figure 3B, upper and lower panels). Second, while Ms larvae regulated transcripts in a diet-specific manner, Hv larvae regulated transcripts for similar functional categories when they fed on N-, N/PI- or JA-silenced plants. In addition, the diet-specific regulation observed in Ms larvae was also observed for transcripts of different pathways within secondary metabolism. For example, transcripts for the metabolism of terpenes (limonene and pinene) were regulated only by Ms larvae that fed on N-silenced plants, and transcripts for the metabolism of terpenoids and cytochrome P450s involved in the metabolism of xenobiotics were regulated only by Ms larvae that fed on JA-silenced plants (Table S1). In addition to these transcripts, Hv larvae also regulated transcripts for the metabolism of phenylpropanoid when they fed on N-, N/PI- or JA-silenced plants (Table S2).

Peptidases, Hydrolases, and “Other Enzymes”

The patterns of regulation in peptidases, hydrolases and ‘other enzymes’, a category which includes the oxidoreductases, transferases and isomerases is represented in Figure 4A, B, and C.

Ms larvae that fed on JA-silenced plants regulated 2.6 times more transcripts for peptidases than did Hv larvae (Figure 4A, inset, upper and lower panels). Ms larvae that fed on JA-silenced plants also regulated 3–5 times more transcripts for peptidases than did larvae that fed on N/PI- and N-silenced plants, respectively (Figure 4A, inset, upper panel). For example, for serine peptidases, Ms larvae that fed on JA-silenced plants regulated 1.5–3 times more transcripts than larvae than that fed on N/PI- and N- silenced plants, respectively. This regulation of peptidases in Ms larvae was highly diet-specific. For example, they down-regulated a few metallo-peptidases and up-regulated a few omega-peptidases which were not regulated by Ms larvae that fed on N/PI- and N-silenced plants (Figure 4A, upper panel). The regulation of peptidases appeared to be a general response in Hv larvae; they regulated a similar number of transcripts with 72–80% peptidases commonly regulated across all diet types (Figure 4A, inset, lower panel). However, Hv larvae regulated 1.6 times more peptidase transcripts than did Ms larvae when both fed on N-silenced plants, but they regulated the same number of transcripts when they fed on N/PI-silenced plants (Figure 4A, inset, upper and lower panels). Interestingly, no transcripts for omega-peptidases were regulated by Hv larvae (Figure 4A, lower panel).

Ms larvae that fed on JA-silenced plants regulated 4–8 times more transcripts for hydrolases than did larvae that fed on N/PI- and N-silenced plants (Figure 4B, inset, upper panel); and the regulation was 6 times more than in Hv larvae that fed on JA-silenced plants (Figure 4B, upper and lower panels). Ms larvae also regulated 1.9 times more of these transcripts than did Hv larvae when they fed on N/PI-silenced plants (Figure 4B, inset, upper and lower panels). Unlike Ms larvae, Hv larvae that fed on defenseless plants regulated fewer transcripts for hydrolases. The generally observed pattern of up- and down-regulation was also observed for hydrolases (Figure 4B, upper and lower panels). This pattern of regulation was also true for transcripts categorized as ‘other enzymes’ but only for Hv larvae as Ms larvae did not differentially regulate this category of transcripts (Figure 4C, upper and lower panels). However, Hv larvae regulated similar number of theses transcripts when larvae fed on N-, N/PI- and JA-silenced plants (Figure 4C, inset, lower panel).

Transporters

Ms larvae that fed on JA-silenced plants regulated 2.3–3.4 times more transcripts for transporters than larvae that fed on N- and N/PI-silenced plants (Figure 5, upper panel). However, Hv larvae regulated more transcripts for transporters when they fed on N-silenced plants (1.3 times more) than when they fed on N/PI- and JA-silenced plants (Figure 5, lower panel). Although both Ms and Hv larvae regulated the two ubiquitously occurring transporters, the ATP-binding cassette (ABC) superfamily and the major facilitator superfamily (MFS) proteins, involved in the transport of various substrates across extra- and intracellular membranes, the amount of regulation in Hv larvae was much less than in Ms larvae. Ms larvae regulated 16 times more MFS transporter transcripts than Hv larvae when they fed on JA-silenced plants, and 1.4–1.9 times more ABC transporter transcripts when they fed on N- and N/PI-silenced plants, respectively (Figure 5, upper and lower panel). Again, the general diet-specific pattern of gene regulation was also observed in the regulation of transporters. For example, while transcripts for all kinds of transporters were regulated by Hv larvae, transcripts for MFS transporters, and neurotransmitter and lipid transporters were regulated only in Ms larvae that fed on JA-silenced plants; transcripts for ABC transporters were regulated only by Ms larvae that fed on N- and NPI; and transcripts for ion-coupled transporters were regulated only by Ms larvae that fed on N/PI-silenced plants (Figure 5, upper and lower panels).

Genetic Information Processing

This category included transcripts coding for regulatory proteins that take part in regulation at the DNA, RNA and protein level which facilitate the processing of genetic information. The diet-specific regulation of transcripts for genetic information processing was prominent in Ms larvae which differentially regulated transcripts for this functional category only when they fed on JA-silenced plants. For example, transcripts for proteins involved in protein targeting and degradation, transcription factors, proteins participating in the translation machinery and proteins that bind to other proteins were regulated only by Ms larvae that fed on JA-silenced plants (Figure 6, upper panel). In contrast to the patterns observed for the regulation of genes in other functional categories, Ms larvae up-regulated most of these transcripts. Hv larvae behaved as expected and both up- and down-regulated the transcripts in this category (Figure 6, lower panel). In Hv larvae, the response to N-silenced plants and JA-silenced plants were similar, but 1.4 times more than larvae that fed on N/PI-silenced plants (Figure 6, lower panel). Ms larvae regulated 1.4 times more transcripts for protein targeting and degradation when they fed on JA-silenced plants than did Hv larvae that fed on these plants (Figure 6, upper and lower panels).

We broadly classified transcripts for maintaining cellular communication, the immune system, the growth and development of cells and larvae, as cellular processes. While transcripts for cellular processes were regulated 3–6 times more by Ms larvae that fed on JA-silenced plants than larvae that fed on N- and N/PI-silenced plants, they were equally regulated by Hv larvae independent of their ‘silenced’ diet. However, Hv larvae regulated more transcripts than did Ms larvae and regulated transcripts not regulated by Ms larvae (Figure S1, upper and lower panels). For example, they regulated transcripts for cytoskeleton-related proteins, WD-repeat protein, ankyrin and troponin (Tables S1 and S2).

Ms larvae that fed on JA-silenced plants regulated more transcripts for environmental information processing (signal perception and transduction) than did larvae that fed on N- and N/PI-silenced plants and as expected, all transcripts were down-regulated (Figure S2, upper panel). Unlike Ms larvae, Hv larvae regulated similar number of transcripts for environmental information processing across the different diets (Figure S2, lower panel). Hv larvae, in addition to regulating transcripts regulated by Ms larvae (proteins part of the ErbB, calcium and mammalian target of rapamycin (mTOR) signaling pathways) also regulated transcripts encoding proteins that are part of the two-component, mitogen-activated protein kinase, Wnt and transforming growth factor beta (TGF-beta) signaling pathways (Tables S1 and S2).

Discussion

Insect attack elicits defense responses in plants and these in turn elicit counter responses in the insects as they ingest the elicited plant material. To uncouple this cycle of response and counter response between plant and insect attackers, we fed larvae on isogenic plants that had been transformed to sequentially silence a suite of JA-mediated defenses (N, TPIs, and all JA-mediated defenses) and analyzed the transcriptional responses in neonates with an unbiased experimental protocol. We selected two insect species that commonly attack N. attenuata in nature [14], [51]: the oligophagous Ms larvae and the polyphagous Hv larvae, as examples of specialist and generalist lepidopteran herbivores, respectively. Ms adults preferentially oviposit on solanaceous plants, but have recently been reported to feed on 2 Proboscidea species, a non-solanaceous plant [62], while adult Hv females oviposit on plant species belonging to more than 14 taxonomically diverse families, including solanaceous taxa (tobacco, tomato etc.) [63]. Although N. attenuata plants elicit slightly different transcriptional signatures in response to attack from these two species, there is considerable overlap in the defense genes activated [51] and by restraining the host plant's defense response by transformation, we uncoupled the response cycle between plant and insect.

While our choice of “specialist” and “generalist” taxa for the comparison is clearly confounded by their different phylogenetic histories, the growth performances of these two species on N. attenuata have been studied and are consistent with the expectations of greater tolerance of the specialist (Ms) to the specific defense metabolites of N. attenuata than the generalist (Hv) [51]. Similar results have been reported from other, equally phylogenetically uncontrolled comparisons. For example, Ms larvae increased their mass by 1.5-fold when they fed on N-silenced plants in comparison to WT plants while another generalist larvae (Spodoptera exigua) increased their mass by 4-fold compared to WT-fed larvae [50]. The objective of the analysis was to determine if an herbivore that has evolved tolerance against the defenses of its host plant would regulate fewer genes than a non-adapted species would or more specifically if Ms would regulate a few, but specific genes and Hv would regulate more of its genes in response to the diverse array of N. attenuata's defenses. The results were consistent with this expectation. Since both generalists and specialists had been previously shown to perform better on JA-silenced plants [46], [64], we expected that the greater the number of defenses silenced in the host plant (JA>N/PI>N), the greater the number of transcripts that would be regulated in both herbivores, but the specificity of the regulation would differ, so that the specialist would display a more diet-specific response in comparison to the generalist. Again the results of the study were consistent with these expectations.

While our general expectations for the number of transcripts regulated between Hv and Ms larvae feeding on progressively defenseless plants were met, there were considerable differences (Figure 7) which can be distilled to two additional general patterns: Ms larvae largely down-regulated its genes while Hv larvae both up- and down-regulated its genes from the same functional categories, and Ms larvae regulated its genes in a diet-specific manner while Hv larvae regulated similar transcripts across all diet types. These two patterns are also consistent with general expectations for differences between generalists and specialists. The adaptations of specialists likely entail the regulation of specific genes to counter their host plant's responses and since our hybridizations compared the responses of larvae feeding on defenseless plants to larvae feeding on WT plants, the pattern of gene regulation we observed (down-regulation) is consistent with an up-regulation of detoxification responses associated with the ingestion of larger quantities of feeding-elicited defenses. Similarly, generalist herbivore species would be expected to counter the host plant's defense responses with a more generalized counter response. The details of the specific genes regulated provide insights into the mechanisms by which the neonates adjust their physiologies to diets deficient in the N, N/PI and JA-elicited responses in N. attenuata.

thumbnail
Figure 7. Summary of the pattern of transcriptional regulation in a specialist (Ms) and generalist (Hv) herbivore as they feed on N. attenuata host-plants deficient in JA-mediated defenses, defenses that are normally activated in response to herbivore attack.

Feeding by Ms and Hv larvae elicits JA-mediated defense responses. In response, the specialist herbivore (Ms) fine-tunes its transcriptome by activating a few but specific transcripts; however, the generalist herbivore (Hv) regulates a large portion of its transcriptome. Both species responded similarly to JA-mediated defenses with more genes regulated in larvae feeding on JA-silenced plants followed by N/PI- and N-silenced plants, but the effect of these defenses elicits a larger transcriptional response in Hv larvae than in Ms larvae. The thickness of the lines and arrows reflects the extent of transcriptional changes elicited in herbivores in response to the ingestion of plant material containing anti-herbivore defenses.

https://doi.org/10.1371/journal.pone.0008735.g007

Numerous mechanisms have been suggested as being responsible for the remarkable tolerance of Ms larvae to dietary N intake. Three remain viable hypotheses: rapid detoxification by P450 enzymes [59], [60], rapid excretion [58], [65], [66], [67], and intrinsic insensitivity of Manduca's central nervous system to N [68]. A fourth mechanism, modifications in N's binding to Ms nicotinic acetylcholine receptors, has recently been ruled out [66]. Although all the three mechanisms have found experimental support, their relative importance remains unclear. The N-inducible expression of a variety of P450s is also observed in Hv larvae [69]. Although less is known about the mechanisms that operate in Hv larvae, an inducible detoxification and excretion system seem to operate in Ms larvae that helps them tolerate N. However, it is not known if these mechanisms also operate in neonates, as most studies used late instar larvae. Our transcriptional analysis suggests that alkaloid excretion, as indicated by the exclusive down-regulation of members of the ABC superfamily of transporters in Ms larvae that fed on N/PI- and N-silenced plants, rather than N detoxification by P450 enzymes (which were not regulated), may be more important for the tolerance of Ms neonates to N. Hv neonates were found to regulate (both up- and down-regulate) a few transcripts for ABC transporters and P450s, but Ms larvae regulated 1.4–1.9 times more transcripts for ABC transporter than did Hv larvae when they fed on N- and N/PI-silenced plants, respectively.

N functions as an antifeedant, causing N-adapted herbivores to curtail consumption likely due to the costs of N detoxification [33], [70], [71]. PIs decrease insect growth by inhibiting gut proteases and potentially reducing the availability of essential amino acids [72]. Adaptation to the ingestion of dietary PIs has been documented in a number of insects [73], [74], [75] and involves one or a combination of the following strategies: the overproduction of existing inhibitor-sensitive digestive proteases [76], [77], [78], [79]; increased expression of inhibitor-insensitive protease isoforms [74], [77], [80], [81], [82], [83]; activation of proteases that hydrolyze plant PIs [76], [83], [84]; and compensatory feeding [74], [85]. While specialists are likely to have evolved more specific means of coping with diets rich in PIs, generalists are more likely to have evolved more generalized behavioral responses to impaired digestive function [86], [87]. This is consistent with our observation that Hv larvae regulated 72–80% of its peptidases (belonging to different classes) as a general response to feeding on plants silenced in N, N/PI and JA, while Ms larvae largely down-regulated serine peptidases when they fed on plants silenced in PIs (N/PI- and JA-silenced plants). We also observed, that Ms larvae fed on N/PI-silenced plants down-regulated a larger number of serine peptidases than did larvae that fed on N-silenced plants, a result consistent with the expectation that N complements the defensive function of TPIs in N. attenuata by restraining the compensatory feeding of herbivores [14]. Although it is difficult to conclude without the results of transcriptional changes in larvae that fed on plants silenced only for TPIs, we infer that Ms and Hv neonates employ similar strategies when they ingest host plant TPIs.

JA elicitation increases the production of some direct defenses, such as N and TPIs, against which Ms larvae have evolved tolerance. However, Ms larvae are still susceptible to other JA-mediated defenses, such as diterpene glycosides (DTGs) which accumulate to concentrations equivalent to that of starch in herbivore attacked tissues. Ms larvae that fed on plants silenced in DTG production gained 3 times more mass than larvae that fed on N- or PI-silenced plants [49], [88].

Herbivore attack to N. attenuata elicits a massive reprogramming of their transcriptome, proteome and metabolome that reflects the re-allocation of resources from primary metabolism to fuel secondary metabolism [89]. This metabolic readjustment can reduce the nutritional value of the leaf consumed by herbivores, and may function as an alternative defensive response. As JA elicitation is known to reduce the shoot quality by reducing total sugar and amino acids [90], photosynthetic proteins [91], and increasing the activity of enzymes that degrade essential amino acids in herbivores [92]. Ms larvae that fed on JA-silenced plants down-regulated MFS transporters that are thought to function in transport of sugar [93], [94] and toxic compounds [95]. A part of the rapid growth of both generalists and specialists on JA-silenced plants [46], [64] may reflect these changes in primary metabolism. The dramatic up-regulation of primary metabolism genes in both Ms and Hv larvae that fed on JA-silenced plants may reflect a similar shift in metabolic priorities from defense-detoxification to growth-related processes and may reflect the degree to which the ingestion of host plant defenses constrains the growth potential of larvae by the metabolic costs of their detoxification. This interpretation is consistent with the observation that Ms larvae that fed on JA-silenced plants down-regulated genes involved in secondary metabolism (cytochrome P450s, GSTs and COEs) while up-regulating genes involved in carbohydrate, lipid, nucleotide, cofactor and vitamin metabolism. However, the possibility that Ms and Hv larvae may eavesdrop on its host plant's JA signaling to regulate its detoxification enzymes represents an alternative explanation for the large transcriptional regulation observed in larvae that fed on JA-silenced plants. Cytochrome P450s are known to play a role in the detoxification of plant chemicals in Ms larvae [96] and generalist Helicoverpa species, in which they are elicited by the ingestion of host plant signal molecules, JA and SA [97].

The inferences reported in this study are drawn from a single pair of generalist and specialist herbivores and will need to be verified in additional comparisons, preferably ones sharing a more similar evolutionary history. However, the inferences provided by this study provide many hypotheses, which can be rigorously tested once efficient plant-mediated RNAi systems have been developed to silence the expression of the diet-induced changes in transcripts in the insects [98]. In this way, the plant-insect transcriptional duet can be uncoupled on the insect side of the equation, and would nicely complement to the uncoupling conducted in this study on the plant side of the equation. Silencing of the candidate genes that were found to be strongly regulated in this study such as MFS and ABC transporters, some of the cytochrome P450s, in combination with metabolomic studies [99] will likely provide insights into the mechanism by which neonates rapidly adapt to the ingestion of host-derived defenses, or anticipate the ingestion of these toxins.

Materials and Methods

Plant Material and Growth Conditions

We used isogenic lines of Nicotiana attenuata (Solanaceae), obtained after 22 generations of inbreeding a genotype collected from a burn in southwestern Utah, for transformations and untransformed wild type (WT) plants. We transformed WT plants with RNAi and antisense constructs to silence nicotine (N) production, N production together with trypsin protease inhibitor (TPI) and jasmonate (JA) biosynthesis. All transformants were homozygous for a single insertion and have been fully characterized in previous publications. Briefly, for plants silenced in JA biosynthesis, we used a fragment of NaLOX3, a key enzyme of JA biosynthesis, expressed in an antisense (as) orientation in the WT genotype, as characterized in [46]. These as-lox3 plants (line A300) accumulate only 35% and 50% of the JA that WT plants do after mechanical wounding and treatment with water, and mechanical wounding and the application of OS, respectively. By silencing JA signaling, the plants are impaired in all JA-mediated defenses, which include N, TPIs, diterpene glycosides (DTGs), and herbivore-induced volatile organic compounds (VOCs) that function as indirect defenses [46]. In plants silenced for N, line A03-108, in which a consensus fragment of N. attenuata's two putrescine N-methyl transferase (pmt) genes, which are regulatory enzymes of nicotine biosynthesis, was expressed in an inverted-repeat (ir) orientation, as characterized in [50]. The ir-pmt plants have drastically reduced transcripts of both PMT genes and produce no detectable quantities of N. To produce plants silenced for both N and TPIs (N/PI), homozygous T2 generation ir-pmt plants were re-transformed with ir constructs of NaTPI, as characterized in [14]. The ir-pmt/pi plants (line A04-103) are completely silenced in N and TPI production. All transgenic lines were morphologically indistinguishable from WT plants. Seeds were germinated in diluted liquid smoke solution as described in [100], and seedlings were transplanted into soil-containing pots in a glasshouse under the conditions described in [101] with 1000–1300 µmol m−2s−1 photosynthetic photon flux density supplied by 450 W Na-vapor high-intensity discharge bulbs.

Herbivores of N. attenuata

Eggs of Manduca sexta (Ms) and Heliothis virescens (Hv) from in-house reared populations were kept in a growth chamber (Snijders Scientific, http://www.snijders-tilburg.nl) at 26°C 16 h light, 24°C 8 h darkness, until the larvae hatched. Freshly hatched 20–25 Ms and 35–40 Hv neonates were placed on fully developed leaves of single rosette-stage N. attenuata plants for each genotype. After 24 h of feeding on each genotype, 15 Ms larvae and 30 Hv larvae were collected in 1.5 ml Eppendorf micro-centrifuge tubes, flash-frozen in liquid nitrogen and stored at −80°C until RNA was extracted using the TRIzol® reagent from Invitrogen (Carlsbad, CA, USA). This whole larvae RNA was used in the hybridizations of the cDNA microarrays.

For the creation of the cDNA microarrays, midgut tissues were isolated from neonates of both species that had fed for 24 h on WT plants elicited with 250 µg methyljasmonate (MeJA) in 20 µl of lanolin, three days prior to larval feeding. Three WT plants elicited with MeJA were maintained for each larval species with 20 Ms and 40 Hv neonates feeding on each plant. Approximately 40 Ms larvae and 100 Hv larvae were harvested after 24 h of feeding and dissected in ice-cold 1x PBS buffer (Applied Biosystems/Ambion, TX, USA) under a stereomicroscope (Carl Zeiss, Jena, Germany) by first de-heading the caterpillars (up to the third pair of forelegs) and excising the abdomen with a sterile scalpel. As a result of the incisions made at the anterior and posterior ends, the midguts were expelled from the center portion of the larval carcass. The dissected midguts were gently pressed to remove gut contents and free the guts from the adhesive muscle/fat tissue. Immediately following the dissections, individually isolated tissues were pooled, collected in 1.5 ml Eppendorf micro-centrifuge tubes, flash-frozen in liquid nitrogen and stored at -80°C until RNA was extracted. Total RNA was isolated from whole larvae and midgut tissues using the TRIzol® reagent from Invitrogen and used for cDNA library construction.

cDNA Library Construction

A normalized midgut cDNA library was constructed using total RNA isolated from Ms and Hv larvae that had fed on pre-elicited WT N. attenuata plants as described above. Double-stranded cDNA was synthesized using the Creator SMART cDNA library construction kit from Clontech (Mountain View, CA, USA). Following double-stranded cDNA synthesis, normalization was performed with a duplex-specific nuclease (DSN) treatment included in the Trimmer-Direct cDNA normalization kit from Evrogen (Moscow, Russia). The manufacturer's protocol was followed for both procedures except for one alteration using the cDNA library construction kit. Instead of cloning the (PCR-amplified) fragments into the provided phage vector pDNR-LIB (GenBank DQ666274), they were directionally cloned into the vector pUCLIB (2.9 kb), which was obtained by fusing the polylinker of pDNR-LIB as 0.2 kb EcoRI-HindIII-fragment to the 2.6 kb EcoRI-HindIII-fragment of pUC19 (GenBank M77789). Colony picking, plasmid DNA isolation and subsequent PCR-amplification for 5000 randomly selected clones were performed at Qiagen GmbH (Hilden, Germany).

cDNA-Array Fabrication, Hybridization, and Quantification

A total of 5000 randomly selected and PCR-amplified clones from each of the normalized Ms and Hv cDNA libraries were spotted on epoxy coated slides as described in [102]. For each cDNA, two PCR fragments, each with a 5′ -aminolink on either strand, were synthesized, and each PCR fragment was spotted four times. Hence, each gene was represented by two independent PCR fragments, which, in turn, were spotted in quadruplicate. The hybridizations were performed as described by [102] with the following modifications: 10 µg of total RNA isolated from larvae that had fed on WT and transgenic lines were reverse-transcribed without mRNA isolation. The first step of the pre-hybridization washing was performed with 2% (w/v) SDS instead of a Triton X-solution. Microarrays were scanned with an array scanner (GMS 418, MWG, Ebersberg, Germany) and spot intensities (SIs) for Cy3 and Cy5 were extracted from image files using the AIDA software, version 4.03 (Raytest, http://www.raytest.com/). Normalization and statistical analysis of each microarray were performed as described in [103]. For a simplified visualization, down-regulated transcripts with a ratio smaller than 1 were transformed by dividing ‘−1’ by the Cy3/Cy5-ratio.

EST Sequencing, Processing, and Functional Annotation

Clones that were differentially regulated in Ms and Hv larvae in response to feeding on defenseless plants, as identified by microarray studies, were selected and subjected to single-pass sequencing using a 5′ vector primer. DNA sequencing was performed by Qiagen GmbH (Hilden, Germany). Sequence information was stored in chromatograph trace files, and Phred was used to perform base-calling [104]. Flanking vector and adaptor sequences were trimmed using Cross-match (http://www.sanger.ac.uk/Software/) and Lucy [105], while low-quality bases (quality score <20) were cleaned at both sequence ends using our custom program. RepeatMasker (http://ftp.genome.washington.edu/) was used to mask repeated sequences, and the masked sequences were further screened to remove contaminating sequences from bacteria and viruses using BLASTN [106]. The cleaned expressed sequence tags (EST) were then subjected to clustering using the TIGR software TGI Clustering tool (TGICL) [107]. The clustering was performed by a modified version of NCBI's megablast. EST sequences were assigned to clusters based on identity: the clustering parameters were 98% minimum percent identity for overlaps, for a minimum overlap length of 40 nt and a maximum length of unmatched overhangs of 20 nt. The cluster names corresponded to the name of the first EST sequence assigned to the cluster. Sequences from each cluster were assembled into consensus sequences (contigs) using the CAP3 assembly program available in TGICL. High quality assembled ESTs were annotated using BLASTX through NCBI and our local BLAST server with a cut-off E-value of 1e-10. The sequences reported in this study have been deposited in GenBank under accession numbers FK816474-FK817136 (Ms) and GT054264-GT056056 (Hv). The sequences were mapped to KEGG (Kyoto Encyclopedia of Genes and Genomes) biochemical pathways according to the EC distribution in the pathway database [108] and assigned the KO IDs (KEGG orthology); sequences which did not match any KO IDs were manually searched against the Uniprot and functionally annotated. While the microarray data presented here are fully MIAME compliant, the raw data has not been deposited in a MIAME compliant database because not all spotted clones on the arrays have been sequenced (only the significantly regulated clones were sequenced). The MIAME-compliant publically available databases require that all clones spotted on cDNA arrays are sequenced. Hence we present all of the raw data, list of clone IDs with expression ratios and functional annotation, in Excel files as supplementary material (Table S1 (Ms) and S2 (Hv)).

Supporting Information

Figure S1.

Fewer genes for cellular processes were regulated in Manduca sexta than in Heliothis virescens larvae that fed on plants silenced for jasmonate (JA) signaling, N/PI, and N defenses. The highest degree of regulation for these genes was found in M. sexta larvae that fed on JA-silenced plants and H. virescens larvae that fed on N-silenced plants. Open bars: up-regulated (ER >1.5); solid bars: down-regulated (ER <−1.5) by M. sexta (upper panel) and H. virescens (lower panel).

https://doi.org/10.1371/journal.pone.0008735.s001

(0.04 MB TIF)

Figure S2.

All the genes coding for environmental information processing were down-regulated in Manduca sexta larvae, while in Heliothis virescens larvae they were both up- and down-regulated. Both species regulated a high percentage of genes when they fed on jasmonate-silenced plants. Open bars: up-regulated (ER >1.5); solid bars: down-regulated (ER <−1.5) by M. sexta (upper panel) and H. virescens (lower panel).

https://doi.org/10.1371/journal.pone.0008735.s002

(0.03 MB TIF)

Table S1.

Genes transcriptionally regulated by specialist Manduca sexta larvae in response to jasmonate-mediated defense response of Nicotiana attenuata.

https://doi.org/10.1371/journal.pone.0008735.s003

(0.09 MB XLS)

Table S2.

Genes transcriptionally regulated by generalist Heliothis virescens larvae in response to jasmonate-mediated defense response of Nicotiana attenuata.

https://doi.org/10.1371/journal.pone.0008735.s004

(0.83 MB XLS)

Acknowledgments

We thank Klaus Gase, Thomas Hahn, Wibke Kroeber, and Antje Wissgott for their help with cDNA library construction and microarray experiments, and Emily Wheeler for editorial assistance.

Author Contributions

Conceived and designed the experiments: ITB. Performed the experiments: GG OM SA. Analyzed the data: GG TG SA. Contributed reagents/materials/analysis tools: SB ITB. Wrote the paper: GG ITB.

References

  1. 1. Berenbaum MR (2001) Chemical mediation of coevolution: Phylogenetic evidence for Apiaceae and associates. Ann Missouri Bot Gard 88: 45–59.
  2. 2. Gaunt MW, Miles MA (2002) An insect molecular clock dates the origin of the insects and accords with palaeontological and biogeographic landmarks. Molecular Biology and Evolution 19: 748–761.
  3. 3. Nosil P, Crespi BJ, Sandoval CP (2002) Host-plant adaptation drives the parallel evolution of reproductive isolation. Nature 417: 440–443.
  4. 4. Becerra JX (2003) Synchronous coadaptation in an ancient case of herbivory. Proceedings of the National Academy of Sciences of the United States of America 100: 12804–12807.
  5. 5. Pare PW, Tumlinson JH (1999) Plant volatiles as a defense against insect herbivores. Plant Physiology 121: 325–331.
  6. 6. Walling LL (2000) The myriad plant responses to herbivores. Journal of Plant Growth Regulation 19: 195–216.
  7. 7. Kessler A, Baldwin IT (2002) Pant responses to insect herbivory: The emerging molecular analysis. Annual Review of Plant Biology 53: 299–328.
  8. 8. Reymond P, Farmer EE (1998) Jasmonate and salicylate as global signals for defense gene expression. Current Opinion in Plant Biology 1: 404–411.
  9. 9. Wu J, Hettenhausen C, Meldau S, Baldwin IT (2007) Herbivory rapidly activates MAPK signaling in attacked and unattacked leaf regions but not between leaves of Nicotiana attenuata. Plant Cell 19: 1096–1122.
  10. 10. Peiffer M, Felton GW (2009) Do caterpillars secrete “Oral Secretions”? Journal of Chemical Ecology 35: 326–335.
  11. 11. Alborn HT, Brennan MM, Tumlinson JH (2003) Differential activity and degradation of plant volatile elicitors in regurgitant of tobacco hornworm (Manduca sexta) larvae. Journal of Chemical Ecology 29: 1357–1372.
  12. 12. Wu J, Baldwin IT (2009) Herbivory-induced signalling in plants: perception and action. Plant, Cell & Environment 32: 1161–1174.
  13. 13. Diezel C, von Dahl CC, Gaquerel E, Baldwin IT (2009) Different lepidopteran elicitors account for cross-talk in herbivory-induced phytohormone signaling. Plant Physiology 150: 1576–1586.
  14. 14. Steppuhn A, Baldwin IT (2007) Resistance management in a native plant: nicotine prevents herbivores from compensating for plant protease inhibitors. Ecology Letters 10: 499–511.
  15. 15. Frey M, Stettner C, Paré PW, Schmelz EA, Tumlinson JH, et al. (2000) An herbivore elicitor activates the gene for indole emission in maize. Proceedings of the National Academy of Sciences of the United States of America 97: 14801–14806.
  16. 16. Ferry N, Edwards MG, Gatehouse JA, Gatehouse AMR (2004) Plant-insect interactions: molecular approaches to insect resistance. Current Opinion in Biotechnology 15: 155–161.
  17. 17. Howe GA, Jander G (2008) Plant immunity to insect herbivores. Annual Review of Plant Biology 59: 41–66.
  18. 18. Rasmann S, Agrawal AA (2009) Plant defense against herbivory: progress in identifying synergism, redundancy, and antagonism between resistance traits. Current Opinion in Plant Biology 12: 473–478.
  19. 19. Karban R, Agrawal AA (2002) Herbivore offense. Annual Review of Ecology and Systematics 33: 641–664.
  20. 20. Cornell HV, Hawkins BA (2003) Herbivore responses to plant secondary compounds: A test of phytochemical coevolution theory. American Naturalist 161: 507–522.
  21. 21. Berenbaum MR, Zangerl AR (1998) Chemical phenotype matching between a plant and its insect herbivore. Proceedings of the National Academy of Sciences of the United States of America 95: 13743–13748.
  22. 22. Kliebenstein D, Pedersen D, Barker B, Mitchell-Olds T (2002) Comparative analysis of quantitative trait loci controlling glucosinolates, myrosinase and insect resistance in Arabidapsis thaliana. Genetics 161: 325–332.
  23. 23. Ratzka A, Vogel H, Kliebenstein DJ, Mitchell-Olds T, Kroymann J (2002) Disarming the mustard oil bomb. Proceedings of the National Academy of Sciences of the United States of America 99: 11223–11228.
  24. 24. Rosenthal GA, Dahlman DL, Janzen DH (1978) L canaline detoxification a seed predators biochemical mechanism. Science 202: 528–529.
  25. 25. Rosenthal GA, Hughes CG, Janzen DH (1982) L-canavanine, a dietary nitrogen-source for the see predator Caryedes brasiliensis (Bruchidae). Science 217: 353–355.
  26. 26. Engler HS, Spencer KC, Gilbert LE (2000) Insect metabolism - Preventing cyanide release from leaves. Nature 406: 144–145.
  27. 27. Ma RL, Cohen MB, Berenbaum MR, Schuler MA (1994) Black swallowtail (Papilio polyxenes) alleles encode cytochrome P450s that selectively metabolize linear furanocoumarins. Archives of Biochemistry and Biophysics 310: 332–340.
  28. 28. Wittstock U, Gershenzon J (2002) Constitutive plant toxins and their role in defense against herbivores and pathogens. Current Opinion in Plant Biology 5: 300–307.
  29. 29. Cohen MB, Schuler MA, Berenbaum MR (1992) A host-inducible cytochrome-P-450 from a host-specific caterpillar-Molecular cloning and evolution. Proceedings of the National Academy of Sciences of the United States of America 89: 10920–10924.
  30. 30. Sasabe M, Wen Z, Berenbaum MR, Schuler MA (2004) Molecular analysis of CYP321A1, a novel cytochrome P450 involved in metabolism of plant allelochemicals (furanocoumarins) and insecticides (cypermethrin) in Helicoverpa zea. Gene 338: 163–175.
  31. 31. Glendinning JI (2002) How do herbivorous insects cope with noxious secondary plant compounds in their diet? Entomologia Experimentalis et Applicata 104: 15–25.
  32. 32. Snyder MJ, Hsu EL, Feyereisen R (1993) Induction of cytochrome P-450 activities by nicotine in the tobacco hornworm, Manduca sexta. Journal of Chemical Ecology 19: 2903–2916.
  33. 33. Parr JC, Thurston R (1972) Toxicity of nicotine in synthetic diets to larvae of the Tobacco Hornworm. Annals of the Entomological Society of America 65: 1185–1188.
  34. 34. Singer MS, Mace KC, Bernays EA (2009) Self-Medication as Adaptive Plasticity: Increased Ingestion of Plant Toxins by Parasitized Caterpillars. PLoS ONE 4: e4796.
  35. 35. Eisner T, Meinwald J (1995) Chemical ecology. Proceedings of the National Academy of Sciences of the United States of America 92:
  36. 36. Bezzerides A, Yong TH, Bezzerides J, Husseini J, Ladau J, et al. (2004) Plant-derived pyrrolizidine alkaloid protects eggs of a moth (Utetheisa ornatrix) against a parasitoid wasp (Trichogramma ostriniae). Proceedings of the National Academy of Sciences of the United States of America 101: 9029–9032.
  37. 37. Eisner T, Eisner M, Rossini C, Iyengar VK, Roach BL, et al. (2000) Chemical defense against predation in an insect egg. Proceedings of the National Academy of Sciences of the United States of America 97: 1634–1639.
  38. 38. Hui D, Iqbal J, Lehmann K, Gase K, Saluz HP, et al. (2003) 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 131: 1877–1893.
  39. 39. Mozoruk J, Hunnicutt LE, Cave RD, Hunter WB, Bausher MG (2006) Profiling transcriptional changes in Citrus sinensis (L.) Osbeck challenged by herbivory from the xylem-feeding leafhopper Homalodisca coagulata (Say) by cDNA macroarray analysis. Plant Science 170: 1068–1080.
  40. 40. Reymond P, Bodenhausen N, Van Poecke RMP, Krishnamurthy V, Dicke M, et al. (2004) A conserved transcript pattern in response to a specialist and a generalist herbivore. Plant Cell 16: 3132–3147.
  41. 41. Darwish SA, Pan L, Ide C, Bede JC (2008) Caterpillar-Specific Gene Expression in the Legume, Medicago truncatula. Plant Molecular Biology Reporter 26: 12–31.
  42. 42. Hermsmeier D, Schittko U, Baldwin IT (2001) Molecular interactions between the specialist herbivore Manduca sexta (Lepidoptera, Sphingidae) and its natural host Nicotiana attenuata. I. Large-scale changes in the accumulation of growth- and defense-related plant mRNAs. Plant Physiology 125: 683–700.
  43. 43. Vogel H, Kroymann J, Mitchell-Olds T (2007) Different transcript patterns in response to specialist and generalist herbivores in the wild Arabidopsis relative Boechera divaricarpa. PLoS ONE 2: e1081.
  44. 44. Steppuhn A, Schuman MC, Baldwin IT (2008) Silencing jasmonate signalling and jasmonate-mediated defences reveals different survival strategies between two Nicotiana attenuata accessions. Molecular Ecology 17: 3717–3732.
  45. 45. Kahl J, Siemens DH, Aerts RJ, Gaebler R, Kuehnemann F, et al. (2000) Herbivore-induced ethylene suppresses a direct defense but not a putative indirect defense against an adapted herbivore. Planta 210: 336–342.
  46. 46. Halitschke R, Baldwin IT (2003) Antisense LOX expression increases herbivore performance by decreasing defense responses and inhibiting growth-related transcriptional reorganization in Nicotiana attenuata. Plant Journal 36: 794–807.
  47. 47. Keinanen M, Oldham NJ, Baldwin IT (2001) Rapid HPLC screening of jasmonate-induced increases in tobacco alkaloids, phenolics, and diterpene glycosides in Nicotiana attenuata. Journal of Agricultural and Food Chemistry 49: 3553–3558.
  48. 48. Zavala JA, Patankar AG, Gase K, Hui D, Baldwin IT (2004) Manipulation of endogenous trypsin proteinase inhibitor production in Nicotiana attenuata demonstrates their function as antiherbivore defenses. Plant Physiology 134: 1181–1190.
  49. 49. Jassbi AR, Gase K, Hettenhausen C, Schmidt A, Baldwin IT (2008) Silencing geranylgeranyl diphosphate synthase in Nicotiana attenuata dramatically impairs resistance to tobacco hornworm. Plant Physiology 146: 974–986.
  50. 50. Steppuhn A, Gase K, Krock B, Halitschke R, Baldwin IT (2004) Nicotine's defensive function in nature. PLoS Biology 2: 1074–1080.
  51. 51. Voelckel C, Baldwin IT (2004) Generalist and specialist lepidopteran larvae elicit different transcriptional responses in Nicotiana attenuata, which correlate with larval FAC profiles. Ecology Letters 7: 770–775.
  52. 52. Pohnert G, Jung V, Haukioja E, Lempa K, Boland W (1999) New fatty acid amides from regurgitant of lepidopteran (Noctuidae, Geometridae) caterpillars. Tetrahedron 55: 11275–11280.
  53. 53. Halitschke R, Schittko U, Pohnert G, Boland W, Baldwin IT (2001) Molecular interactions between the specialist herbivore Manduca sexta (Lepidoptera, Sphingidae) and its natural host Nicotiana attenuata. III. Fatty acid-amino acid conjugates in herbivore oral secretions are necessary and sufficient for herbivore-specific plant responses. Plant Physiology 125: 711–717.
  54. 54. Gaquerel E, Weinhold A, Baldwin IT (2009) Molecular interactions between the specialist herbivore Manduca sexta (Lepidoptera, Sphigidae) and its natural host Nicotiana attenuata. VIII. An unbiased GCxGC-ToFMS analysis of the plant's elicited volatile emissions. Plant Physiology (Rockville) 149: 1408–1423.
  55. 55. Halitschke R, Ziegler J, Keinanen M, Baldwin IT (2004) Silencing of hydroperoxide lyase and allene oxide synthase reveals substrate and defense signaling crosstalk in Nicotiana attenuata. Plant Journal 40: 35–46.
  56. 56. Kang JH, Wang L, Giri A, Baldwin IT (2006) Silencing threonine deaminase and JAR4 in Nicotiana attenuata impairs jasmonic acid-isoleucine-mediated defenses against Manduca sexta. Plant Cell 18: 3303–3320.
  57. 57. Skibbe M, Qu N, Galis I, Baldwin IT (2008) Induced plant defenses in the natural environment: Nicotiana attenuata WRKY3 and WRKY6 coordinate responses to herbivory. Plant Cell 20: 1984–2000.
  58. 58. Murray CL, Quaglia M, Arnason JT, Morris CE (1994) A putative nicotine pump at the metabolic blood-brain-barrier of the tobacco hornworm. Journal of Neurobiology 25: 23–34.
  59. 59. Snyder MJ, Hsu EL, Feyereisen R (1993) Induction of cytochrome-P-450 activities by nicotine in the tobacco hornworm Manduca sexta. Journal of Chemical Ecology 19: 2903–2916.
  60. 60. Snyder MJ, Glendinning JI (1996) Causal connection between detoxification enzyme activity and consumption of a toxic plant compound. Journal of Comparative Physiology a-Sensory Neural and Behavioral Physiology 179: 255–261.
  61. 61. Krieger RI, Feeny PP, Wilkinson CF (1971) Detoxication enzymes in the guts of caterpillars: An evolutionary answer to plant defenses? Science 172: 579–581.
  62. 62. Mechaber WL, Hildebrand JG (2000) Novel, non-solanaceous hostplant record for Manduca sexta (Lepidoptera: Sphingidae) in the Southwestern United States. Annals of the Entomological Society of America 93: 447–451.
  63. 63. Waldvogel M, Gould F (1990) Variation in oviposition preference of Heliothis virescens in relation to macroevolutionary patterns of Heliothine host range. Evolution 44: 1326–1337.
  64. 64. Kessler A, Halitschke R, Baldwin IT (2004) Silencing the jasmonate cascade: Induced plant defenses and insect populations. Science 305: 665–668.
  65. 65. Maddrell SHP, Gardiner BOC (1976) Excretion of alkaloids by malpighian tubules of insects. Journal of Experimental Biology 64: 267–281.
  66. 66. Wink M, Theile V (2002) Alkaloid tolerance in Manduca sexta and phylogenetically related sphingids (Lepidoptera: Sphingidae). Chemoecology 12: 29–46.
  67. 67. Self LS, Guthrie FE, Hodgson E (1964) Adaptation of tobacco hornworms to the ingestion of nicotine. Journal of Insect Physiology 10: 907–914.
  68. 68. Morris CE (1984) Electrophysiological effects of cholinergic agents on the CNS of a nicotine-resistant insect, the tobacco hornworm (Manduca sexta). Journal of Experimental Zoology 229: 361–374.
  69. 69. Rose RL, Gould F, Levi PE, Hodgson E (1991) Differences in cytochrome-P450 activities in tobacco budworm larvae as influenced by resistance to host plant allelochemicals and induciton. Comparative Biochemistry and Physiology B-Biochemistry & Molecular Biology 99: 535–540.
  70. 70. Voelckel C, Kruegel T, Gase K, Heidrich N, van Dam NM, et al. (2001) Anti-sense expression of putrescine N-methyltransferase confirms defensive role of nicotine in Nicotiana sylvestris against Manduca sexta. Chemoecology 11: 121–126.
  71. 71. Appel HM, Martin MM (1992) Significance of metabolic load in the evolution of host specificity of Manduca sexta. Ecology 73: 216–228.
  72. 72. Birk Y (2003) Plant protease inhibitors: Significance in nutrition, plant protection, cancer prevention and genetic engineering. New York: Springer. 170 p.
  73. 73. Brunelle F, Cloutier C, Michaud D (2004) Colorado potato beetles compensate for tomato cathepsin D inhibitor expressed in transgenic potato. Archives of Insect Biochemistry and Physiology 55: 103–113.
  74. 74. Cloutier C, Jean C, Fournier M, Yelle S, Michaud D (2000) Adult Colorado potato beetles, Leptinotarsa decemlineata compensate for nutritional stress on oryzacystatin I-transgenic potato plants by hypertrophic behavior and over-production of insensitive proteases. Archives of Insect Biochemistry and Physiology 44: 69–81.
  75. 75. Jongsma MA, Bolter C (1997) The adaptation of insects to plant protease inhibitors. Journal of Insect Physiology 43: 885–895.
  76. 76. Ahn JE, Salzman RA, Braunagel SC, Koiwa H, Zhu-Salzman K (2004) Functional roles of specific bruchid protease isoforms in adaptation to a soybean protease inhibitor. Insect Molecular Biology 13: 649–657.
  77. 77. Broadway RM (1995) Are insects resistant to plant proteinase-inhibitors. Journal of Insect Physiology 41: 107–116.
  78. 78. De Leo F, Bonade-Bottino MA, Ceci LR, Gallerani R, Jouanin L (1998) Opposite effects on Spodoptera littoralis larvae of high expression level of a trypsin proteinase inhibitor in transgenic plants. Plant Physiology 118: 997–1004.
  79. 79. Broadway RM (1996) Dietary proteinase inhibitors alter complement of midgut proteases. Archives of Insect Biochemistry and Physiology 32: 39–53.
  80. 80. Jongsma MA, Bakker PL, Peters J, Bosch D, Stiekema WJ (1995) Adaptation of Spodoptera exigua larvae to plant proteinase-inhibitors by induction of gut proteinase activity insensitive to inhibition. Proceedings of the National Academy of Sciences of the United States of America 92: 8041–8045.
  81. 81. Koo YD, Ahn JE, Salzman RA, Moon J, Chi YH, et al. (2008) Functional expression of an insect cathepsin B-like counter-defence protein. Insect Molecular Biology 17: 235–245.
  82. 82. Liu YL, Salzman RA, Pankiw T, Zhu-Salzman K (2004) Transcriptional regulation in southern corn rootworm larvae challenged by soyacystatin N. Insect Biochemistry and Molecular Biology 34: 1069–1077.
  83. 83. Zhu-Salzman K, Koiwa H, Salzman RA, Shade RE, Ahn JE (2003) Cowpea bruchid Callosobruchus maculatus uses a three-component strategy to overcome a plant defensive cysteine protease inhibitor. Insect Molecular Biology 12: 135–145.
  84. 84. Giri AP, Harsulkar AM, Deshpande VV, Sainani MN, Gupta VS, et al. (1998) Chickpea defensive proteinase inhibitors can be inactivated by podborer gut proteinases. Plant Physiology 116: 393–401.
  85. 85. Winterer J, Bergelson J (2001) Diamondback moth compensatory consumption of protease inhibitor-transformed plants. Molecular Ecology 10: 1069–1074.
  86. 86. Simpson SJ, Raubenheimer D, Chambers PG (1995) The mechanisms of nutritional homeostasis. In: Chapman RF, de Boer G, editors. Regulatory Mechansims in Insect Feeding. New York: Chapman & Hall. pp. 251–277.
  87. 87. Rivard D, Cloutier C, Michaud D (2004) Colorado potato beetles show differential digestive compensatory responses to host plants expressing distinct sets of defense proteins. Archives of Insect Biochemistry and Physiology 55: 114–123.
  88. 88. Heiling S, Schuman MC, Schoettner M, Mukerjee P, Berger B, et al. (2010) Jasmonate and ppHsystemin regulate key malonylation steps in the biosynthesis of 17-hydroxygeranyllinalool diterpene glycosides, an abundant and effective direct defense against herbivores in Nicotiana attenuata. The Plant Cell. in press.
  89. 89. Schwachtje J, Baldwin IT (2008) Why does herbivore attack reconfigure primary metabolism? Plant Physiology 146: 845–851.
  90. 90. van Dam NM, Oomen WATM (2008) Root and shoot jasmonic acid applications differentially affect leaf chemistry and herbivore growth. Plant Signaling & Behavior 3: 91–98.
  91. 91. Mitra S, Baldwin IT (2008) Independently silencing two photosynthetic proteins in Nicotiana attenuata has different effects on herbivore resistance. Plant Physiology (Rockville) 148: 1128–1138.
  92. 92. Chen H, Wilkerson CG, Kuchar JA, Phinney BS, Howe GA (2005) Jasmonate-inducible plant enzymes degrade essential amino acids in the herbivore midgut. Proceedings of the National Academy of Sciences of the United States of America 102: 19237–19242.
  93. 93. Henderson PJF, Maiden MCJ (1990) Homologous sugar-transport proteins in Escherichia coli and their relatives in both prokaryotes and ecukaryotes. Philosophical Transactions of the Royal Society of London Series B-Biological Sciences 326: 391–410.
  94. 94. Maiden MCJ, Davis EO, Baldwin SA, Moore DCM, Henderson PJF (1987) Mammalian and bacterial sugar-transport proteins are homologous. Nature 325: 641–643.
  95. 95. Hayashi K, Schoonbeek HJ, De Waard MA (2002) Bcmfs1, a novel major facilitator superfamily transporter from Botrytis cinerea, provides tolerance towards the natural toxic compounds camptothecin and cercosporin and towards fungicides. Applied and Environmental Microbiology 68: 4996–5004.
  96. 96. Stevens JL, Snyder MJ, Koener JF, Feyereisen R (2000) Inducible P450s of the CYP9 family from larval Manduca sexta midgut. Insect Biochemistry and Molecular Biology 30: 559–568.
  97. 97. Li XC, Schuler MA, Berenbaum MR (2002) Jasmonate and salicylate induce expression of herbivore cytochrome P450 genes. Nature 419: 712–715.
  98. 98. Mao YB, Cai WJ, Wang JW, Hong GJ, Tao XY, et al. (2007) Silencing a cotton bollworm P450 monooxygenase gene by plant-mediated RNAi impairs larval tolerance of gossypol. Nature Biotechnology 25: 1307–1313.
  99. 99. Jansen JJ, Allwood JW, Marsden-Edwards E, van der Putten WH, Goodacre R, et al. (2009) Metabolomic analysis of the interaction between plants and herbivores. Metabolomics 5: 150–161.
  100. 100. Baldwin IT, Staszak-Kozinski L, Davidson R (1994) Up in smoke: I. Smoke-derived germination cues for postfire annual, Nicotiana attenuata Torr. Ex. Watson. Journal of Chemical Ecology 20: 2345–2371.
  101. 101. Zavala JA, Patankar AG, Gase K, Baldwin IT (2004) Constitutive and inducible trypsin proteinase inhibitor production incurs large fitness costs in Nicotiana attenuata. Proceedings of the National Academy of Sciences of the United States of America 101: 1607–1612.
  102. 102. Halitschke R, Gase K, Hui D, Schmidt DD, Baldwin IT (2003) Molecular interactions between the specialist herbivore Manduca sexta (Lepidoptera, Sphingidae) and its natural host Nicotiana attenuata. VI. Microarray analysis reveals that most herbivore-specific transcriptional changes are mediated by fatty acid-amino acid conjugates. Plant Physiology 131: 1894–1902.
  103. 103. Wang L, Allmann S, Wu J, Baldwin IT (2008) Comparisons of LIPOXYGENASE3- and JASMONATE-RESISTANT4/6-silenced plants reveal that jasmonic acid and jasmonic acid-amino acid conjugates play different roles in herbivore resistance of Nicotiana attenuata. Plant Physiology 146: 904–915.
  104. 104. Ewing B, Green P (1998) Base-calling of automated sequencer traces using phred. II. Error probabilities. Genome Research 8: 186–194.
  105. 105. Chou HH, Holmes MH (2001) DNA sequence quality trimming and vector removal. Bioinformatics 17: 1093–1104.
  106. 106. Altschul SF, Gish W, Miller W, Myers EW, Lipman DJ (1990) Basic local alignment search tool. Journal of Molecular Biology 215: 403–410.
  107. 107. Pertea G, Huang XQ, Liang F, Antonescu V, Sultana R, et al. (2003) TIGR Gene Indices clustering tools (TGICL): a software system for fast clustering of large EST datasets. Bioinformatics 19: 651–652.
  108. 108. Ogata H, Goto S, Sato K, Fujibuchi W, Bono H, et al. (1999) KEGG: Kyoto Encyclopedia of Genes and Genomes. Nucleic Acids Research 27: 29–34.