SEARCH

SEARCH BY CITATION

Keywords:

  • flowering;
  • jasmonic acid;
  • jasmonic acid-isoleucine;
  • ethylene;
  • Manduca sexta;
  • Nicotiana attenuata;
  • ontogeny;
  • optimal defense theory;
  • plant-herbivore interactions

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and Methods
  7. Acknowledgments
  8. References

To defend themselves against herbivore attack, plants produce secondary metabolites, which are variously inducible and constitutively deployed, presumably to optimize their fitness benefits in light of their fitness costs. Three phytohormones, jasmonates (JA) and their active forms, the JA-isoleucine (JA-Ile) and ethylene (ET), are known to play central roles in the elicitation of induced defenses, but little is known about how this mediation changes over ontogeny. The Optimal Defense Theory (ODT) predicts changes in the costs and benefits of the different types of defenses and has been usefully extrapolated to their modes of deployment. Here we studied whether the herbivore-induced accumulation of JA, JA-Ile and ET changed over ontogeny in Nicotiana attenuata, a native tobacco in which inducible defenses are particularly well studied. Herbivore-elicited ET production changed dramatically during six developmental stages, from rosette through flowering, decreasing with the elongation of the first corollas during flower development. This decrease was largely recovered within a day after flower removal by decapitation. A similar pattern was found for the herbivore-induced accumulation of JA and JA-Ile. These results are consistent with ODT predictions and suggest that the last steps in floral development control the inducibility of at least three plant hormones, optimizing defense-growth tradeoffs.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and Methods
  7. Acknowledgments
  8. References

Plants have different strategies to defend themselves against all sorts of attackers. They possess structural barriers and chemical compounds to protect themselves from attack by a variety of herbivores and pathogens (Smith et al. 2009). These defense traits are formed regardless of the presence of herbivores and are thus called constitutive defenses. Inducible defenses, on the other hand, are produced in response to attack by herbivores (Karban and Baldwin 1997). Since the biosynthesis of defense compounds is thought to be energy and nutrient demanding (Karban and Baldwin 1997), it is not surprising that plants use highly-developed regulatory systems to balance the demands for growth and defense. This balance has been articulated by the Optimal Defense Theory (ODT), which predicts that the allocation of resources to defense is determined by the cost of production, the fitness benefit and the probability of being attacked (Mc Key 1974; Boege and Marquis 2005). In other words, tissues that contribute most to plant fitness should be best defended and these tissues will change over ontogeny.

Every plant goes through several distinct developmental stages during its life, and depending on the physiological priorities at each stage, the production of defenses may be constrained by growth early on or by reproduction at later stages. The physiological changes that accompany these developmental changes influence the production of both physical and chemical defenses (Boege and Marquis 2005; Hanley et al. 2007). Changes in physical defenses during ontogeny have been reported mainly for trees and include changes in trichome and spine density as well as leaf toughness (Kearsley and Whitham 1989; Gowda and Palo 2003; Loney et al. 2006; Traw and Feeny 2008). More studies have focused on changes in chemical defenses during ontogeny including changes in alkaloids (Ohnmeiss and Baldwin 2000; Gregianini et al. 2004; Elger et al. 2009), cyanogenic glycosides (Goodger and Woodrow 2002; Goodger et al. 2006), phenolics (Donaldson et al. 2006; Neilson et al. 2006; Elger et al. 2009), defensive proteins, enzymes such as protease inhibitors (Van Dam et al. 2001; Doan et al. 2004), terpenoids (Sinclair and Smith 1984; Langenheim et al. 1986; Boege and Marquis 2005; Barton 2007; Barton and Koricheva 2010) and plant volatiles (Hare 2010). Studies have reported both decreases (Cipollini and Redman 1999; Fritz et al. 2001; Goodger et al. 2006) and increases (Macedo and Langenheim 1989; Erwin et al. 2001; Schaffner et al. 2003; Elger et al. 2009) in the production of secondary metabolites as plants develop. As a consequence, some plants have declining defense levels as they age (Wallace and Eigenbrode 2002), while others increase (Fenner et al. 1999; Goodger et al. 2004; Boege 2005).

A recent meta-analysis performed by Barton and Koricheva (2010) concluded that no single ontogenetic pattern in plant defenses exists. They concluded that ontogenetic patterns differ among plant species, growth forms, and types of defenses. In herbaceous plants, constitutive defenses generally increase, while induced defenses decrease with age. Consistent with the predictions of the ODT is the observation that induced defense responses are best elaborated in, and perhaps limited to, relatively young plants or plant parts, a pattern well established for direct defenses (Wolfson and Murdock 1990; Thaler et al. 1996; Ohnmeiss et al. 1997; Ohnmeiss and Baldwin 2000; Van Dam et al. 2001; Matthes et al. 2008). More recent studies have reported a similar trend for indirect defenses, such as the release of herbivore-induced plant volatiles (Köllner et al. 2004; Rostas and Eggert 2008; Hare 2010) and the production of extrafloral nectar (Radhika et al. 2008). Even though ODT was originally not formulated in the context of induced defenses, several studies have extrapolated the ODT predictions to inducible defenses (see for example, Ohnmeiss and Baldwin 2000). Despite the large number of studies documenting such ontogenetic patterns in plant defense traits, the mechanisms responsible for these patterns and the exact timing of the ontogenetic changes in defenses remain unexplored.

Phytohormones function as signals that tune physiological processes to environmental stresses, such as herbivore attack. Jasmonic acid (JA) plays a key role in plant development, but JA and its derivatives such as its amino acid conjugate with isoleucine (Ile), represent also the best characterized class of signals that mediate the elicitation of defense responses to wounding and herbivore attack (Creelman and Mullet 1997; Beale and Ward 1998; Blee 1998; Devoto and Turner 2003; Farmer et al. 2003). A transient accumulation of JA occurs after wounding in nearly all plant species where it has been examined and this JA burst is amplified during herbivore attack or when oral secretions (OS) of herbivores are introduced into wounds (Baldwin et al. 2000; Schittko et al. 2000; Ziegler et al. 2001). Halitschke et al. (2001) demonstrated that fatty acid-amino acid conjugates (FACs) found in the OS of Manduca sexta were sufficient to elicit both the JA burst and JA-elicited volatiles when applied to wounded N. attenuata leaves, which allows for uniformly conducted experiments that give more precise answers to the timing of defense signaling. This OS-elicited JA burst is remarkably robust to environmental variation, and is readily detected in field-grown native plants, even with the same kinetics of waxing and waning as described in glasshouse-grown plants (Diezel et al. 2009). Defects in JA biosynthesis, perception or signaling increase a plant's susceptibility to various pathogens and insect herbivores and demonstrates that JA is a major transducer of signals that are essential for plant defense (Farmer and Ryan 1990; Howe et al. 1996; Penninckx et al. 1996; Mc Conn et al. 1997; Staswick et al. 1998; Engelberth et al. 2004).

Even though JA is thought to be the main negotiator of responses to herbivore attack in N. attenuata (Halitschke and Baldwin 2005) there is another phytohormone that plays an important role in adjusting defense responses against the specialist herbivore M. sexta: ethylene (ET). ET modulates JA-mediated defense responses, rather than eliciting defense responses on its own (for review see: von Dahl and Baldwin 2007). For example, the ET burst elicited by herbivore attack enhances the production of JA-elicited proteinase inhibitors in tomato (Solanum lycopersicum; O’Donnell et al. 1996). Specific responses tailored to the attack by certain herbivore species likely result from crosstalk among the individual signaling cascades, which includes synergistic as well as antagonistic interactions (Walling 2000; Pieterse et al. 2001). ET is known to be emitted by almost all plant taxa when attacked by arthropod herbivores of many different feeding guilds including, Thysanoptera (Wien and Roesingh 1980), Homoptera (Dillwith et al. 1991), and Lepidoptera (Schmelz et al. 2003). As for JA, the FACs found in the OS of M. sexta caterpillars are responsible for the ET burst in N. attenuata.

In addition to their function in defense, phytohormones are known to regulate nearly all developmental processes in plants, from germination through differentiation, growth and reproduction and finally senescence. Hormones mediate the adjustment of a plant's phenotype to its environment by converting external stresses/stimuli into internal responses. In addition to their role in regulating developmental processes in plants, hormones also regulate the developmental changes in defense patterns.

Here we investigated the changes in M. sexta OS-elicited ET and JA bursts during different developmental stages in N. attenuata and found that the ET burst decreased significantly with leaf and plant age. Young leaves emitted much more ET than did the same leaf at the same node at later stages in ontogeny. The reduction in ET production was related to the initiation of flower production and was largely restored by removal of the inflorescence one day prior to OS-elicitation. The same pattern was observed for OS-elicited JA and JA-Ile bursts. The reduction in OS-elicited JA, JA-Ile, and ET accumulations was due to the loss of the FAC-induced amplification of the phytohormone signals.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and Methods
  7. Acknowledgments
  8. References

Herbivore-induced ET emissions decrease with plant- and leaf-age

Herbivore attack by the specialist herbivore M. sexta or the application of its OS elicits an ET burst in N. attenuata (von Dahl and Baldwin 2007; Von Dahl et al. 2007). To determine whether the OS-induced ET burst changes during ontogeny, we measured ET emissions from WT plants at six different developmental stages ranging from rosette through flowering. On each experimental day the accumulated ET was measured from single leaves growing at different leaf nodes for 5 h after OS treatment and in unwounded leaves, with a laser photo-acoustic spectrometer. To characterize the effects of plant and leaf age, we marked the –1 leaf (one node younger than the source-sink transition leaf, which was labeled at node 0) of all plants one day prior to the first measurement in rosette stage plants and included this aging marked leaf in all subsequent measurements (Figures 1 and 2).

image

Figure 1. Oral secretions (OS)-elicited ethylene emissions decrease with leaf-age.  (A) Schematic of glasshouse-grown Nicotiana attenuata plants at different life stages: rosette stage lasts approximately 31 d; after 34 d, plants start elongating (E-1) and rapidly proceed through stages E-II and E-III as the reproductive meristem produces differentiated but unopened flowers on days 36 (E-II) to 37 (E-III). Corolla elongation occurs after about 38 d (F-I) and flowers are fully opened (F-II) on 42-d-old plants. Close-up pictures of reproductive parts that define the different stages are provided on the x-axis of part B. Plants continue to flower depending on the available resources. Inset: leaf positions of a rosette plant in which “0” is the leaf undergoing the source-sink transition and “–” values represent sink leaves located at nodes above the 0 leaf (at 120° rotations around the stem) and “+” values represent source leaves at nodes below the 0 leaf.  (B) OS-elicited ethylene emissions were quantified by laser acoustic spectroscopy with a ‘stop-flow’ protocol from single leaves after standardized puncture wounds were immediately treated with Manduca sexta oral secretions (OS). Untreated leaves at the same nodal position served as controls. The ‘–1’ leaf of all plants (yellow, outlined in bold) was marked one day before the first measurement and the ethylene emissions of this leaf node were monitored as it matured through all developmental stages of the plant. The actual nodal position of the measured leaf at the day of measurement is noted below the line graph. OS-elicited ethylene production rapidly declines at E-III, which correlates strongly with the emergence of corollas. Asterisks indicate significant differences from untreated control (n= 3) leaves occupying the same nodal positions as the treated (n= 5) leaves (Student's t-test, R-I: t6=–8.15, P < 0.001; E-I: t6=–3.12, P < 0.05; E-II: t6=–4.55, P < 0.01; E-III: t6=–2.86, P < 0.05; F-I: t6=–5.43, P < 0.01; F-II: t6=–3.25, P < 0.05 (P < 0.05 (*), P < 0.01 (**), P < 0.001 (***)).

Download figure to PowerPoint

image

Figure 2. Oral secretions (OS)-elicited ethylene emissions decrease with leaf- and plant age.  Ethylene emissions were determined by ‘stop-flow’ measurements of single leaves from different positions and plant stages (see also Figure 1 for leaf-positions) after wounding and treatment of the wounds immediately with Manduca sexta oral secretions (OS). Untreated leaves from the same nodal positions served as controls. Hatched bars represent data depicted in Figure 1 for reference (control: n= 3; wound + OS: n= 5).

Download figure to PowerPoint

Wounded and M. sexta OS treated rosette stage –1 leaves (see Figure 1A for leaf-positions) showed almost a five-fold higher ET emission rate (8.02 vs 1.73 nL/h per g fresh matter (FM)) than did leaves at node +5 on elongated/flowering stage N. attenuata plants, the leaves which had been marked on rosette stage plants as occupying node –1. The largest decrease in ET burst occurred in the transition phase to flowering. Untreated control leaf ET emissions did not change in the marked leaf over development and averaged 0.82 nL/h per g FM (Figure 1B and Table 1). When comparing different leaf positions and hence leaf-ages within one plant, a similar trend was observed. ET emissions in rosette-stage plants ranged from 8.02 nL/h per g FM in the young –1 leaves to 4.30 nL/h per g FM in the older +2 leaves after M. sexta OS treatment (Student's t-test, P < 0.05). During ontogeny, these levels of OS-induced ET emissions within one plant decreased and ranged from 3.49 in young S3 leaves to 1.73 nL/h per g FM in old +5 leaves in flowering plants (Figure 2, F-II). Untreated control plants showed a similar trend, as their constitutive ET emissions were higher in the younger than in older leaves. Interestingly this difference in ET emissions between leaves within one plant increased during ontogeny. While the constitutive ET levels in rosette stage plants were relatively low independent of the leaf age (0.85 in –1 node and 0.72 nL/h per g FM in +2 node), ET emissions in flowering plants (F-II) ranged from 1 nL/h per g FM in the oldest (+5) to 2.6 nL/h per g FM in the youngest leaf measured.

Table 1.  Ethylene emissions of all leaves testedThumbnail image of

The waning of the OS-elicited ET burst as plants start flowering is rapidly recovered by floral removal

To determine whether it is only the OS-inducible ET burst that vanishes during ontogeny or whether the decrease also applies to wound responses, we compared ET emissions of wounded-1 leaves of rosette plants and wounded S2 leaves of flowering plants after treating the standardized puncture wounds with either water (w+w) or M. sexta OS (w+OS). The –1 leaf of rosette plants and the S2 leaf of flowering plants were chosen since both leaves had about the same leaf-age on the experimental day. By doing so we could focus on the plant-age effect while minimizing the effect that leaf-age had on the ET emissions. While plants responded similarly to w+w treatment regardless of their developmental stage, the OS-elicted ET burst in rosette plants completely vanished in flowering plants (Student's t-test, rosette: t6= 4.42, P < 0.01; flowering: t8=−0.93, P= 0.38; Figure 3A).

image

Figure 3. The loss of the oral secretions (OS)-elicited ethylene (ET) burst after flowering is restored within one day by decapitating the inflorescence.  (A) OS- and wound-elicited ethylene emissions of single leaves from either rosette stage plants or flowering plants. The –1 leaf of rosette stage (n= 4) and the S2 leaf of flowering plants (n= 5) was wounded with a fabric pattern wheel and wounds were immediately treated with either water (w+w) or Manduca sexta OS (w+OS). Asterisks indicate significant differences between treatments (Student's t-test, rosette: t6= 4.42, P < 0.01 (**); flowering: t8=–0.93, P= 0.38). Note that only the OS-elicited ethylene burst, not the wound-elicited ethylene burst decreases with flowering.  (B) Ethylene emissions of single leaves from flowering plants or from flowering plants that had been decapitated one day prior. The S2 leaf of each plant (n= 5) was wounded and treated with either water (w+w) or M. sexta OS (w+OS). Letters indicate significant differences between treatments (univariate anova followed by a Fisher post-hoc test, F3,16= 2.271; P < 0.05 (*)).

Download figure to PowerPoint

To determine if the presence of flowers with elongating corollas was responsible for the disappearance of the ET burst, we decapitated flowering plants 1 d prior to w+w or w+OS treatments. Decapitation significantly increased the OS-elicited ET burst 3.5 fold, from 1 to 3.5 nL/h per g FM (anova (F3,16= 2.271; P < 0.05; Figure 3B).

The flowering-associated waning of the OS-(FAC-) inducible JA and JA-Ile bursts are also rapidly restored by inflorescence removal

In N. attenuata, FACs found in M. sexta OS are able to mimic most of the known OS-elicited changes in transcripts, proteins, and metabolites (Halitschke et al. 2003; Voelckel and Baldwin 2004; Giri et al. 2006; Gaquerel et al. 2009) and the two most abundant FACs found in M. sexta OS (N-linolenoyl-L-glutamine and N-linolenoyl-L-glutamate) are necessary and sufficient to elicit the OS-specific JA (Halitschke et al. 2003), JA-Ile (Kang et al. 2006) and ET (von Dahl et al. 2007) bursts. When applying OS or these two FACs at concentrations found in M. sexta OS (1:3 diluted) to puncture wounds (–1 node for rosette, S2 for flowering/1 d cut plants in Figure 4, and S0 for flowering/1 d cut plants in Figure 5), JA and JA-Ile bursts that were almost three-fold larger than those elicited by w+w or w+tween (the detergent for the dissolution of FACs) were detected when rosette-stage plants were treated (For JA: anova followed by Fisher's least significant difference (LSD) post-hoc test, F15,64= 20.68, P < 0.001 Figure 4; for JA-Ile: Student's t-test, t7= 6.61, P < 0.001; Figure 5). Similar to the ET burst, the JA and JA-Ile bursts decreased in flowering plants (JA-Ile: Student's t-test, t8=–0.53, P= 0.61; Figure 5), but for JA the burst was still significantly larger compared with the wound controls, albeit highly attenuated from that in rosette-stage plants (anova followed by Fisher's LSD post-hoc test, F11,48= 21.65, P < 0.01 for w+w vs w+OS; P < 0.05 for w+FAC vs w+tween; Figure 4). The OS-elicited JA burst decreased from 2 000 to 800 ng/g FM and the FAC-elicited JA-Ile burst from 110 to 60 ng/g FM as plants matured and flowered. However, within one day after removing flowers, the full OS- and FAC-elicited JA and JA-Ile bursts were restored: more than 2 000 ng/g FM in w+OS/FAC treatments compared with approximately 600 ng/g FM JA in w+w and w+tween treated de-flowered plants (anova followed by Fisher's LSD post-hoc test, F11,48= 21.65; P < 0.001) and 490 ng/g FM in w+FAC treatments compared with 150 ng/g FM JA-Ile in w+tween treated de-flowered plants were detected (Student's t-test, 1 d cut: t8= 6.29; P < 0.001).

image

Figure 4. The oral secretions (OS)-(fatty acid-amino acid conjugates (FACs)-) inducible jasmonates (JA) burst shows the same responses as the OS-elicited ethylene burst: inhibition by flowering and rapid recovery after decapitation of the inflorescence.  JA levels were measured from rosette stage, flowering plants or from flowering plants that had been decapitated one day before measurement. The S2 leaf of each plant (n= 5) was wounded with a fabric pattern wheel and wounds were immediately treated with either water (w+w) or Manduca sexta OS (w+OS) or tween (w+tween – detergent used to emulsify FACs in water solutions) or FACs (w+FAC), respectively. Letters indicate significant differences between treatments (univariate anova followed by a Fisher post-hoc test, F11,48= 21.65 P < 0.05 (*); P < 0.01 (**), P < 0.001 (***)).

Download figure to PowerPoint

image

Figure 5. Fatty acid-amino acid conjugates (FACs)-inducible bursts of jasmonates (JA)-Ile are also inhibited after flowering but are rapidly recovered 1 d after decapitation of the inflorescence.  JA-Ile levels were measured from rosette stage, flowering plants or from flowering plants that had been decapitated one day before measurement. The S0 leaf of each plant (n= 5) was wounded with a fabric pattern wheel and wounds were immediately treated with either tween (w+tween) or FACs (w+FAC), respectively. In rosette stage plants the –1 node was treated. Asterisks indicate significant differences between treatments (Student's t-test, rosette: t7= 6.61, P < 0.001 (***); flowering: t8=–0.53, P= 0.61); 1 d cut: t8= 6.29, P < 0.001).

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and Methods
  7. Acknowledgments
  8. References

Defenses against herbivores often change during the ontogeny of a plant. Since defense strategies against herbivores are regulated by phytohormones, it is not surprising that the phytohormonal response to herbivore attack also changes as plants develop. We found that OS-inducible JA and ET bursts decline with the initiation of flowering in N. attenuata plants, and that this loss in OS-inducibility can rapidly be reset within one day by removing the inflorescence of the plant (Figures 3 and 4). Ohnmeiss and Baldwin (2000) have shown a decrease in wound-elicited JA in N. sylvestris during its ontogeny. But this is the first demonstration of an apparent control of flowering over herbivore elicited ET and JA signaling. The dramatic effect of decapitation suggests a central role for auxin signaling in the expression of OS-elicited response. Auxin treatment of puncture wounds to N. sylvestris leaves is known to inhibit both the wound-induced nicotine response (Baldwin 1989) as well as the wound-induced JA burst (Baldwin et al. 1997), work that was recently confirmed with 1-naphthylacetic acid (an auxin analog) treatments in cultivated tobacco (Shi et al. 2006). Furthermore, the phytohormone, salicylic acid, which mediates defense responses against biotrophic pathogens has been shown to induce flowering in several plant species (Wildermuth et al. 2001; Loake and Grant 2007) and while we did not measure SA levels in our plants, the dose-dependent antagonistic nature of SA towards JA (Doherty et al. 1988; Péna-Cortes et al. 1993; Doares et al. 1995; Pieterse et al. 2009; Diezel et al. 2009) might explain the reduced ability of flowering plants to accumulate JA after OS-elicitation.

The flowering-induced “deafness” to OS-elicitation in N. attenuata that we observed in this study as seen so clearly in the dramatic attenuation of the OS-elicited JA-, JA-Ile- and ET-bursts might not necessarily lead to a reduced inducibility of JA-, JA-Ile- and/or ET-mediated defense responses if plants at the same time become more sensitive to these plant hormones during ontogeny leading to unchanged levels of phenotypically expressed resistance traits. However, although we did not measure these resistance traits, there are several previous results that revealed a decreased OS-inducibility of the main JA-regulated defense metabolites in Nicotiana spp. For example Van Dam et al. (2001) showed that de novo synthesis of PIs is limited to the early stages of plant development as flowering plants treated with methyl jasmonate did not significantly increase local or systemic PI activity levels. In addition, damage to older leaves elicited a much weaker systemic response in younger leaves than did damage on younger source leaves, a pattern also reported from N. tabacum (Pearce et al. 1993). Similar to PIs, wound- and jasmonate-induced whole plant nicotine contents in N. sylvestris decreased during the plant's ontogeny (Baldwin and Schmelz 1996; Ohnmeiss and Baldwin 2000). Moreover Kaur et al. (2010) showed that the high constitutive levels of caffeoylputrescine (a phenylpropanoid-polyamine conjugate) in the vegetative tissues of rosette and early elongated stages of N. attenuata clearly shifted to the reproductive tissues after flowering and capsule development. As a consequence, hardly any caffeoylputrescine could be detected in the leaves of mature plants. Heiling et al. (2010) showed that the concentrations of 17-hydroxygeranyllinalool diterpene glycosides (DTGs) were highest in young and reproductive tissue and that these DTGs effectively defend this valuable tissue against herbivores in N. attenuata. Lastly, several articles described that the removal of the flowering head increases nicotine concentration in N. tabacum and N. sylvestris leaves (Baldwin 1989; Hashimoto and Yamada 1994; Xi et al. 2005; Shi et al. 2006), suggesting a rejuvenating effect caused by the removal of a plant's inflorescence. This overall decline in inducibility in older plants or, more precisely, in plants that reach their reproductive stage is consistent with the ODT's predictions about tissue value, if reproductive growth is largely supplied by reserves acquired during rosette stage growth.

In other words, the plant's limited supply of defensive compounds should be concentrated in those regions in which their presence would be most beneficial for the plant's fitness. Mc Key (1974) described the two factors that determine the chemical-defensive needs of a plant part: the value for the plant and their vulnerability to herbivores. Usually young tissues at the tips of roots and shoots of a plant are both valuable and vulnerable organs. Since nutrients are normally transported to the sink-tissue and are thus concentrated at these sites (Fraenkel 1953; Hammond and White 2008), herbivores are likely to prefer them over older, lower nutrient-containing tissues. Thus it is not surprising that these tissues accumulate defensive compounds. Schoonhoven (1967) already noted that young tomato leaves seemed to contain some deterrent factor for M. sexta larvae but that these caterpillars do perform well on older tomato leaves. The young shoot tips of the broom, Sarothamnus scoparius, contain higher concentrations of sparteine, a sodium channel blocker, than do older leaves (Smith 1966). Clear evidence for a young leaf's higher contribution to fitness was provided by Ohnmeiss and Baldwin (2000) and Barto and Cipollini (2005) who showed that removal of young leaves led to a significantly larger decrease in overall seed mass over undamaged control plants or plants whose old leaves had been removed demonstrating an age-dependent effect of the value of leaves. Young leaves are probably more valuable to fitness than old leaves as they have a greater potential for recent and future photosynthetic rates. Since the OS-elicited declines are precisely associated with the last stages of corolla opening, the control of the OS-elicited burst may reside more specifically in the transcription factors (TFs) that have recently been shown to regulate this last step in flower development.

The floral transition is one of the most dramatic phase changes in a plant's life. This transition is coordinated by a complex genetic network and until now several transcription factors involved in this process have been identified (for reviews see Amasino 2010; Irish 2010). A recent study from Dezar et al. (2011) showed that the sunflower HD-Zip transcription factor HAHB10 is involved in the transition from vegetative to reproductive stage by inducing flower transition genes while repressing genes related to biotic stresses. When HAHB10 was constitutively expressed in Arabidopsis, these plants accumulated lower levels of JA after wounding. However, wound-induced ET levels did not differ from wild type plants. Another transcription factor that is involved in flower opening is EOBII. When EOBII transcript levels were silenced in N. attenuata by inverted-repeat expression (ir-EOBII), the flowers failed to enter anthesis, and this could be reversed by blocking the plant's ET sensitivity, suggesting an important role for ET in the plant's flowering process (Colquhoun et al. 2011). Further investigations into the roles in defense signaling of these transcription factors will likely provide valuable insights into the plant's ontogenetic switch from inducible to constitutive defense and might illuminate how the transition from OS-inducible to un-inducible ET is regulated.

The initiation of flowering is associated with changes in the relative defense requirements of different plant organs. Chemical protection of the seeds becomes more important, while protection of vegetative parts loses its importance, especially if the vegetative parts senesce during seed filling, as is the case in annual plants like N. attenuata. With the attenuation of ET and JA signaling, N. attenuata is temporally limited in its ability to activate certain defenses against herbivores. This ontogenetic constraint may result from metabolic limitations in that flowering plants prioritize the allocation of resources to seed production over other functions that are not directly involved in fitness output. While understudied in plants grown in native populations, leaf loss to herbivores during the flowering stage does not reduce lifetime seed output in glasshouse grown N. sylvestris plants (Ohnmeiss and Baldwin 2000).

In addition to the fitness-level explanations of the ODT, mechanistic-level explanations may also apply: the loss of OS inducibility during flowering might result from a switch in the function of phytohormone signaling from defense to reproduction, as the same phytohormones are used in mate choice and reproduction. For example pollination is known to trigger ET emission in flowers (Hall and Forsyth 1967) thereby inducing flower abscission and seed maturation and these pollination-induced ET emissions might be given higher priority over ET's function in defense elicitation in flowering plants. Similar constraints may apply to JA signaling. This hypothesis is highlighted in ir-coi1 N. attenuata plants that are impaired in JA-elicited direct (PI's, nicotine and caffeoylputrescine) defense responses and are also male sterile. Therefore, deciphering the role of ET and JA through N. attenuata's developmental stages is essential to an understanding of how plants optimize defense and reproductive efforts.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and Methods
  7. Acknowledgments
  8. References

Plant material and growing conditions

Wild type Niccotinia attenuata (common name: coyote tobacco) plants were from an inbred line in its 33rd generation that originated from seeds collected in Utah in 1988. Seeds were germinated on Gamborg's B5 medium (Duchefa) as described in Krügel et al. (2002). In short, seeds were sterilized and incubated in 1 mM GA3 (Roth) and 1:50 diluted liquid smoke (v/v; House of Herbs) before germination on Gamborg's B5 medium at a 26 °C/16 h 155 μm/s per m2 light: 24 °C/8 h dark cycle (Percival). Plants were grown in the glasshouse with a day/night cycle of 16 h (26 °C–28 °C)/8 h (22 °C–24 °C) under supplemental light from Master Sun-T PIA Agro 400 or Master Sun-T PIA Plus 600-W sodium lights (Philips).

Insect rearing and plant treatments

Tobacco hornworm (Manduca sexta) eggs, purchased from Carolina Biological supply and bred in an in house colony, were cultivated in climate chambers until hatching and reared on N. attenuata wild-type plants until the third to fifth instar. Oral secretions (OS) were collected on ice as described in Roda et al. (2004) and stored at –20 °C until use.

For all experiments leaves were mechanically wounded with a pattern wheel to produce three puncture rows on each side of the midvein. Fresh wounds were immediately treated with 20 μL of either water, M. sexta OS (diluted 1:3 with water), 0.02% Tween (polyoxyethylene (20) sorbitan monolaurate) or fatty-acid-amino-acid conjugates (FACs). Tween was used as detergent for the dissolution of FACs and thus served as a control for the FAC treatment. The FACs used were the two most abundant FACs in M. sexta OS, N-linolenoyl-L-glutamine and N-linolenoyl-L-glutamate (Halitschke et al. 2001) at concentrations similar to M. sexta OS (when diluted 1:3). Control plants remained untreated. For Figures 3 and 4 plants were decapitated to remove all inflorescence 1 d prior to treatment.

Phytohormone analysis

Jasmonic acid (JA) and jasmonic acid-conjugated with isoleucine (JA-Ile) were extracted from elicited leaves by adding one mL of ethyl acetate spiked with 200 ng of D2-JA and 50 ng 13C6-JA-Ile as the internal standards to each sample, which consisted of approximately 200 mg of fresh leaf material. Samples were homogenized on a FastPrep homogenizer (Thermo Electron, Waltham, MA, USA). After centrifugation at maximum speed for 15 min at 4 °C, supernatants were transferred to 2 mL Eppendorf tubes. A second extraction was performed with 1 mL ethylacetate only and evaporated to dryness on a vacuum concentrator (Eppendorf, Hamburg, Germany) at 30 °C. The residue was re-suspended in 0.5 mL of 70% methanol (v:v) and centrifuged to clarify phases. The supernatants were analyzed on a high pressure liquid chromatography (HPLC)-mass spectrometry (MS)/MS (1200L LC-MS) system (Varian, Foster City, CA, USA) as described in Diezel et al. (2009).

Ethylene emissions were measured continuously and noninvasively with a photoacoustic laser spectrometer (INVIVO) as described in von Dahl et al. (2007). Leaves were excised immediately after treatment and transferred to 250 mL cuvettes and the headspace was allowed to accumulate over a 5 h time period. During measurements, cuvettes were flushed with a flow of purified air at 130 to 150 mL/min, which had previously passed through a liquid N2 cooling trap to remove CO2 and water.

Statistical analysis

All data presented in were analyzed with Statview 5.0 (SAS Institute). Data were transformed if they did not meet the assumption of homoschedacity.

(Co-Editor: Ivan Galis)

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and Methods
  7. Acknowledgments
  8. References

We thank Dr. Tamara Krügel, Andreas Weber and Andreas Schünzel for help with plant cultivation and Andreas Schünzel for the schematic figure of developing N. attenuata plants. This work was supported by the Max Planck Society.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Results
  5. Discussion
  6. Materials and Methods
  7. Acknowledgments
  8. References