SEARCH

SEARCH BY CITATION

Keywords:

  • Chemical defence;
  • constitutive defence;
  • generalist;
  • inducible defence;
  • specialist

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  1. Latex functions as a physical and chemical defence against herbivores that may vary in their responses to latex quantity and chemical composition. In their introduced ranges, many exotic plants experience reduced herbivore regulation, especially from specialist herbivores, which may lead to differences in defences among native and invasive populations.
  2. Here, we compared latex produced by seedlings from native and invasive populations of tallow tree (Triadica sebifera) when damaged by a native specialist or generalist caterpillar.
  3. We measured the growth of caterpillars fed leaves of plants from native and invasive populations that had latex washed from the leaves or were unwashed controls.
  4. We found that constitutive latex mass, tannins and flavonoids of plants from invasive and native populations were similar. The masses of caterpillars fed leaves from native and invasive populations were comparable regardless of leaf washing.
  5. We also found that the specialist induced more latex production than the generalist did, especially for plants from invasive populations. Tannins increased and flavonoids decreased when plants were damaged by the generalist caterpillar but neither changed when plants were damaged by the specialist caterpillar.
  6. Our results suggest divergent selection on the physical and chemical properties of latex in the introduced range. The quantity of latex produced was more sensitive to herbivore identity in the introduced range but the specificity of the latex chemical response was retained.

Introduction

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

In their introduced ranges, many exotic plants experience reduced herbivore regulation, especially from specialist herbivores (Keane & Crawley, 2002), which can lead to differences in defences among native and invasive populations (Blossey & Nötzold, 1995). As defence is costly (Zangerl & Rutledge, 1996; Morris et al., 2006), exotic plants may evolve to invest more resources in growth or reproduction (Blossey & Nötzold, 1995; Bossdorf et al., 2005), or less-costly defences more effective against generalists (Joshi & Vrieling, 2005; Doorduin & Vrieling, 2011). Inducible defences, which are only synthesised or mobilised in response to a stimulus, are thought to have lower costs than constitutive defences, which are always expressed in a plant. Therefore, invasive plants could change inducible defences in their invasive range due to differences in the frequency and types of herbivore attack in a way that is different than the pattern for constitutive defences.

The induction of specific defensive traits may be associated with a variety of herbivores, leading to distinct responses to specific herbivores (Karban & Myers, 1989). Some studies have reported that plants have different induced responses to specialist and generalist herbivores but it has been argued that specificity of induction may be more complex than simply generalist versus specialist defence induction (see review in Ali & Agrawal, 2012). Poelman et al. (2008a) found that damage by Pieris rapae (specialist within Brassicaceae) and Mamestra brassicae (generalist with host range extending beyond the Brassicaceae) induced higher indole glucosinolates in Brassica nigra than damage by Plutella xylostella (specialist within Brassicaceae). There are other examples showing that specialist and generalist herbivores elicit similar plant responses (Reymond et al., 2004; Poelman et al., 2008b). However, in the context of plant invasions, studies on defences of exotic plants produced in response to attack by specialist and generalist herbivores may be able to provide unique insights into understanding the evolution of plant defences due to the absence of specialists in the introduced range (Orians & Ward, 2010; Wang et al., 2012a).

Latex is an important defence against herbivores for some plants. When a plant is damaged, latex exudes from specialised canals and accumulates at the damaged points (Agrawal & Konno, 2009). There is evidence that plant latex affects herbivores by both mechanical and chemical mechanisms. First, latex exuded from the damaged point clots rapidly, and can glue the mouthparts or the whole body of insects (Dussourd & Eisner, 1987; Dussourd, 1995). Second, plant latex contains various secondary metabolites and proteins, many of which have been reported to serve as a defence against herbivores (Zalucki et al., 2001; Agrawal & Konno, 2009; Konno, 2011). Moreover, the concentrations of some substances in latex have been shown to be much higher than those in leaves (Sessa et al., 2000; Konno et al., 2004, 2006). There have been studies of latex and its chemicals that demonstrate that it is a potent plant defence against mandibulate herbivores (Agrawal & Konno, 2009). However, few studies to date have investigated latex chemicals and latex defensive functions in invasive plants, especially for responses to specialist and generalist herbivores.

Here we examine biogeographical variation in latex defences using Chinese tallow [Triadica sebifera (L.) Small = Sapium sebiferum (L.) Roxb.] as a model species. We also examined inducible latex variation among populations from invasive and native ranges. Previous studies reported that T. sebifera contains secondary chemicals such as tannins and flavonoids in leaves (Huo & Gao, 2004; Wang et al., 2012a). Tannins are known to defend leaves against insects by reducing digestibility (Salminen & Karonen, 2011), and may be particularly effective against specialist herbivores (Müller-Schärer et al., 2004). Flavonoids are primarily associated with defence against generalists (Müller-Schärer et al., 2004) but also are thought to have other functions such as protection from ultraviolet radiation, allelopathy or defence against pathogens (Harborne & Williams, 2000; Iwashina, 2003). Recent studies found that invasive populations of T. sebifera expressed higher flavonoids and lower tannins in leaves than did native populations (Huang et al., 2010; Wang et al., 2012a). No studies have examined variation in latex production in T. sebifera.

We addressed the following questions: (i) Do invasive and native populations differ in constitutive latex quantity and secondary chemical characteristics? (ii) Do latex quantity and secondary chemical characteristics vary in response to attack by different herbivores? (iii) Do invasive and native populations differ in latex quantity and secondary chemical characteristics after attack by different herbivores?

Materials and methods

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

Study system

Chinese tallow tree, T. sebifera, is a widely distributed, deciduous tree species in Asia. This species was introduced to Georgia in North America in the late eighteenth century for agricultural purposes, and is now a noxious invader in most of the southeastern United States (Bruce et al., 1997). Our study focused on two Lepidoptera herbivores native to China that feed on T. sebifera: Gadirtha inexacta Walker (Lepidoptera: Noctuidae) and Cnidocampa flavescens Walker (Lepidoptera: Limacodidae).

Gadirtha inexacta feeds exclusively on T. sebifera, and is considered a potential biological control agent against T. sebifera (Wang et al., 2012b). Cnidocampa flavescens is a generalist defoliator that can also seriously damage T. sebifera. All caterpillars were field-collected in 2012 and reared on potted T. sebifera (Wuhan population) in the Wuhan Botanical Garden, at the Chinese Academy of Sciences, Hubei, China (30°32′N, 114°24′E). The offspring of these collections were used for experiments in the native range. The two herbivores in this study feed tallow in similar ways. They chew leaves and create gaps. Both caterpillar species feed primarily on the underside of leaves. The neonate larvae feed on the lower leaf cuticle, producing small transparent circular patches. Feeding by late instar larvae produces large holes in the leaves. In the field, we found that they damaged tallow in the same way.

Seeds and seedlings

Seeds were collected from both native (10 populations in south China, hereafter referred to as native populations) and introduced ranges (10 populations from the southeastern United States, hereafter referred to as invasive populations), in the autumn of 2011 (Table S1). Recent molecular studies indicated that the populations in the introduced range come from at least two distinct introduction events with the original introduction to Georgia and South Carolina likely from a southern China population. Later introductions to the United States are likely from the northeast part of Triadica's range (DeWalt et al., 2011). Therefore, we consider the populations used in this experiment to be representative native and introduced populations. We removed the waxy coats of tallow seed by soaking in water with laundry detergent (10 g l−1) for 2 days, and then buried them in sand at a depth of 5–10 cm, and kept them in a refrigerator (4 °C) for 40 days (Huang et al., 2010; Wang et al., 2012a). In April 2012, seeds from each population were germinated in a glasshouse for 6 weeks and then seedlings were transplanted individually into pots (height, 16 cm; diameter, 25 cm; filled with 50% locally collected field soil and 50% sphagnum peat moss) and randomly placed in a common garden at Wuhan Botanical Garden. Each plant was enclosed in a transparent nylon mesh cage (height, 100 cm; diameter, 30 cm) to prevent herbivory. In total, there were 366 plants (126 for the latex properties experiment, 240 for the insect bioassay experiment).

Induction

To compare the inducible latex variation of introduced versus native populations with different types of feeding, we used two herbivore species to damage plants. We randomly assigned three seedlings of each population to the following treatments: (i) herbivory by G. inexacta (specialist); (ii) herbivory by C. flavescens (generalist); and (iii) unmanipulated control. There were 126 seedlings in total for measurement of latex properties (two continents × seven populations × three treatments × three replicates). Treatments were imposed in August 2012. Seedlings of this stage typically have 15–20 true leaves. For each induced seedling, we used two to five caterpillars depending on their rates of damage and allowed them to consume 20–30% of the leaf area. Caterpillars were removed 3–5 days after initiating feeding.

Latex production measurement

Seven days after removal of caterpillars from a plant, we measured latex production of leaves. We cut off the youngest, fully expanded, undamaged leaf from the leaf base. We collected latex from leaves of equivalent position for each herbivore treatment and controls for 5 s after cutting, because prior tests showed no latex was secreted after 5 s. Then we collected the latex with a pre-weighed, sterilised 1 cm filter paper disc (no. 1 Whatman International, Maidstone, Kent, U.K.) and weighed each disc to the nearest microgram.

Latex collection

We next collected latex by cutting all the intact leaves on each plant off at the leaf base. When latex was exuded from the damaged points, we used a capillary pipette (diameter, 0.5 mm) to absorb the exudation and then transferred it into a pre-weighed, sterilised centrifuge tube. All of the latex absorbed from one plant was kept in a single tube and weighed to the nearest microgram. Before high-performance liquid chromatography analysis, these tubes were stored in a freezer in −20 °C.

Chemical analyses

We assessed four tannins (gallic acid, catechin, tannic acid and ellagic acid) and five flavonoids (quercetin, isoquercetin, quercetin glycoside, kaempferitrin and kaempferol) with high-performance liquid chromatography. The latex samples of each induction treatment were dissolved in 400 µl purified water, and then diluted 10 times. The mixture was filtered through a 0.45 µm membrane. The extract was injected (100 µl) into a Dionex ultimate 3000 series high-performance liquid chromatography (Dionex, Sunnyvale, California) and compounds were separated on a ZORBAX Eclipse C18 column (4.6 × 250 mm, 5 µm; Agilent Technologies, Santa Clara, California). Tannins were eluted at a constant flow of 1.0 ml min−1 with methanol–0.1% phosphoric acid in water gradient as follows: 0–7.5 min, 30 : 70; 7.5–17 min, 55 : 45. Ultraviolet absorbance spectra were recorded at 279 nm for gallic acid, catechin and tannic acid and at 260 nm for ellagic acid. Flavonoids were eluted at a constant flow of 1.0 ml min−1 with a 100% methanol–0.4% phosphoric acid in water gradient as follows: 0–10 min, 48 : 52; 10–18.5 min, 65 : 35. Ultraviolet absorbance spectra were recorded at 254 nm. Concentrations were calculated and standardised by peak areas of criteria of known concentrations, and then reported as per gram of latex wet mass for each of the four tannins and five flavonoids (Wang et al., 2012a). We do not report the concentrations of every tannin and flavonoid here (because some were below detection levels). In addition, we measured tannins and flavonoids in leaves following latex extraction (Wang et al., 2012a).

Insect bioassays

We removed leaves from additional seedlings and placed the leaves of each plant in a plastic bag stored at 2 °C. To examine the toxicity of tallow latex, each caterpillar was fed leaves that were unmanipulated or washed to remove latex. Leaves for washing treatments were cut into three strips, and these strips were washed twice in water and blotted dry with filter paper. One fully expanded leaf (unwashed) or three washed strips of a single leaf were placed on moist filter paper in a Petri dish (inner diameter, 9 cm). A newly hatched larva was transferred to the dish and given a new leaf (or washed leaf strips) daily for 3 days. The leaves fed to each caterpillar came from a single unique plant. Petri dishes were closed and incubated in the laboratory with a LD 14 : 10 h photo-phase. After 3 days, caterpillars were weighed to the nearest microgram. Assays were replicated three times (two caterpillar species × two continents × 10 populations × two treatments × three replicates = 240 caterpillars and 240 seedlings).

Data analyses

We used anova (proc mixed, SAS 9.0 Institute Inc., Cary, North Carolina) to test the dependence of amount of latex on population origin, herbivore treatment and their interaction. We used the random variable population (origin) as the error term for population origin. We performed similar analyses to examine the dependence of concentration of individual tannins, total tannins, individual flavonoids and total flavonoids on the same predictors. For significant terms with more than two levels, we conducted post-hoc tests using partial difference tests of adjusted means to test for significance differences among treatment means. We performed another set of anovas (with post-hoc tests) to examine the dependence of specialist and generalist caterpillar masses on population origin, washing treatment and their interaction. Population (origin) was treated as a random variable and used as the error term for population origin.

We used graphical modelling to estimate the relative importance of different latex and leaf characteristics for specialist and generalist caterpillar growth using T. sebifera population averages. For strongly correlated predictors, we examined the relationships among predictors and present strongest correlation with caterpillar mass within the group of predictors. Latex variables included were volume, tannin concentration and flavonoid concentration. Leaf variables included were tannin and flavonoid concentrations. For the washed leaves treatment, we only included leaf characteristics because latex had been removed by washing.

Results

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

Latex production

The latex exudation of invasive populations was significantly greater than that of native populations (Table 1; Fig. 1). Damage by specialist G. inexacta significantly increased latex production compared to controls (Table 1; Fig. 1), and this increase was significantly larger for plants of invasive populations. However, damage by generalist C. flavescens did not induce more latex production compared to controls (Table 1; Fig. 1).

Table 1. The dependence of mass of latex exuded on continental origin of tallow tree populations, herbivory treatment and their interactions in a mixed model anova
Factord.f.FP
  1. Significant results are shown in bold.

Origin1,125.260.0406
Treatment2,10522.43< 0.0001
Origin × treatment2,244.190.0276
image

Figure 1. Effects of damage by two caterpillar species (specialist and generalist herbivore) on latex exudation of Triadica sebifera from invasive and native populations. Means with the same letter were not significantly different in post-hoc tests. Adjusted means + 1 standard error.

Download figure to PowerPoint

Tannins and flavonoids in latex

Native and invasive populations did not differ in concentrations of tannins or flavonoids (Tables 2 and 3; Figs 2 and 3). Damage by generalist C. flavescens significantly increased catechin content in latex compared to damage by specialist G. inexacta (Fig. 2a), which did not significantly induce production. Gallic acid did not depend on population origin, treatment or their interaction (Fig. 2b). The total amount of tannins increased with specialist damage. The concentrations of each flavonoid decreased with generalist damage but they did not vary with population origin or the interaction of origin and treatment (Table 3; Fig. 3).

Table 2. The dependence of concentrations of tannins in tallow latex on continental origin of tallow tree populations, herbivory treatment and their interactions in a mixed model anova
Factord.f.CatechinGallic acidTotal tannins
FPFPFP
  1. Significant results are shown in bold.

Origin1,120.310.58500.000.95570.270.6114
Treatment2,10511.81< 0.00011.070.345311.25< 0.0001
Origin × treatment2,240.480.62390.260.77050.500.6136
Table 3. The dependence of concentrations of flavonoids in tallow latex on continental origin of tallow tree populations, herbivory treatment and their interactions in a mixed model anova
Factord.f.KaempferolIsoquercetinQuercetinTotal flavonoids
FPFPFPFP
  1. Significant results are shown in bold.

Origin1,120.310.58740.000.96980.090.77320.120.7300
Treatment2,1056.390.002411.94< 0.000112.26< 0.000110.26< 0.0001
Origin × treatment2,241.450.25432.310.12111.640.21511.960.1622
image

Figure 2. Effects of damage by two caterpillar species (specialist and generalist herbivore) on latex tannins of Triadica sebifera from invasive and native populations: (a) catechin, (b) gallic acid, (c) total tannins. Adjusted means + 1 standard error.

Download figure to PowerPoint

image

Figure 3. Effects of damage by two caterpillar species (specialist and generalist herbivore) on latex flavonoids of Triadica sebifera from invasive and native populations: (a) kaempferol, (b) isoquercetin, (c) quercetin and (d) total flavonoids. Adjusted means + 1 standard error.

Download figure to PowerPoint

Insect bioassay

The washing treatment did not affect caterpillar growth (Table 4; Fig. 4). Specialist G. inexacta grew larger on leaves from invasive populations than native populations, whereas the performance of generalist C. flavescens did not vary with population origin (Table 4; Fig. 4). There were no significant interactive effects of washing and plant origin (Table 4).

Table 4. The dependence of biomass of two caterpillar species (Gadirtha inexacta and Cnidocampa flavescens) on continental origin of tallow tree populations, herbivory treatment and their interactions in a mixed model anova
FactorG. inexactaC. flavescens
d.f.FPd.f.FP
Origin1,184.330.05191,172.610.1245
Treatment1,850.060.80301,820.670.4149
Origin × treatment1,180.320.57641,180.080.7810
image

Figure 4. Effects of latex treatment (washed leaf strips vs. control) and plant origin (invasive and native populations) on the biomass of (a) specialist (Gadirtha inexacta) and (b) generalist (Cnidocampa flavescens). Adjusted means + 1 standard error.

Download figure to PowerPoint

Net changes in latex exudation and chemicals and their effect on specialists and generalists

We used graphical models to show the correlation among latex properties, correlation among foliar chemicals and correlations of each with biomass of specialist and generalist caterpillars. Solid lines represent positive correlations and dotted lines indicate negative correlations. Wider lines represent stronger correlations. We found that latex volume and flavonoid concentration were strongly negatively correlated and latex tannin was positively correlated with latex volume (Fig. 5a). The growth of generalist C. flavescens was lower when latex volume was higher but the specialist G. inexacta had a weak positive relationship with latex volume. Tannin and flavonoid concentrations in leaves were positively correlated and the specialist had a positive correlation with leaf chemicals that was stronger than that of the generalist. The growth of each herbivore was negatively correlated with leaf tannins and flavonoids and this was a stronger correlation for the generalist (Fig. 5b).

image

Figure 5. Graphical models showing the correlation among latex properties, correlation among foliar chemicals and the correlations of each with biomass of specialist and generalist caterpillars. (a) control leaves and (b) washed leaves. Solid lines represent positive correlations and dotted lines indicate negative correlations. Wider lines represent stronger correlations. Specialist-N, Noctuidae (Gadirtha inexacta); Generalist-L, Limacodidae (Cnidocampa flavescens).

Download figure to PowerPoint

Discussion

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

As a physical and chemical defence against herbivores, latex may vary in quantity and chemical composition, depending on herbivore and plant identities. In this study, we found that a specialist caterpillar induced more latex exudation than a generalist caterpillar did, and plants from invasive populations secreted more latex than those from native populations did when damaged by a specialist caterpillar. We also found latex tannin concentrations increased but flavonoid concentrations decreased when plants were damaged by a generalist but were unchanged when damaged by a specialist and that, unlike latex volume, these chemical responses did not depend on population origin.

Triadica sebifera secreted more latex when damaged by a specialist herbivore than when damaged by a generalist. Specialist herbivores have a long evolutionary history with their host plants, and so may have evolved some adaptations to plant latex. Indeed, studies have shown that specialist herbivores have physiological adaptations to plant latex (Holzinger et al., 1992; Holzinger & Wink, 1996; Labeyrie & Dobler, 2004). For instance, specialist silkworms have developed physiological and biochemical adaptations to chemicals in mulberry latex but polyphagous Lepidopteran larvae are sensitive to mulberry latex (Konno, 2011). As a response, plants may increase the volume of latex to deter specialists that are not sensitive to chemicals in latex but may be unable to counter the physical effects of large amounts of latex. In contrast, many generalist herbivores may not have adapted to chemicals in latex as specialists have and may be sensitive to even small amounts of latex. The negative correlation between latex volume and latex chemical concentrations is consistent with two defence strategies focused on physical versus chemical defence.

Invasive populations secreted more latex than native populations did when damaged by a specialist herbivore; however, invasive and native populations had similar latex responses to generalist herbivore damage. Invasive plants may evolve less costly defence strategies to cope with the lower herbivore loads in invasive ranges and reallocate more resources to growth and reproduction (Blossey & Nötzold, 1995; Cipollini et al., 2005). Because plants do not secrete latex without damage (Agrawal & Konno, 2009; Konno, 2011), latex may be less costly to maintain than other defences. Previous studies identified herbivore-derived molecules that appear to be signals of the type of herbivore damaging plants (Heil, 2009; Hilker & Meiners, 2010; Ali & Agrawal, 2012), suggesting specialist herbivores could induce distinct plant defence responses compared with generalist herbivores. In our study, a generalist herbivore induced stronger responses than a specialist herbivore, as the concentration of both tannins and flavonoids in latex changed significantly when damaged by the generalist herbivore. For tallow defence, our previous studies found that invasive populations of T. sebifera have higher tolerance to generalist herbivores and lower resistance to specialist herbivores (Huang et al., 2010), and that invasive populations have higher flavonoids and lower tannins in leaves than native populations (Wang et al., 2012a). Our study on the inducible indirect extrafloral nectar of T. sebifera also found that specialist and generalist herbivores could induce different extrafloral nectar production (Wang et al., 2013).

Plant latex contains water and a diversity of biologically active compounds (secondary metabolites) such as tannins and flavonoids, which directly defend against herbivores (Agrawal & Konno, 2009; Konno, 2011). In our study, we found latex tannin concentrations increased but flavonoid concentrations decreased when plants were damaged by a generalist but were unchanged when damaged by a specialist. Through graphical modelling (Fig. 5) we found that the combined net compound changes (flavonoids, tannins and latex exudation) from native to invasive range negatively affected the generalist but very weakly affected the specialist, indicating divergent selection on the chemical properties of latex in the introduced range where there are generalists but often lack of specialists.

Recent studies have examined secondary compounds in invasive plants, but most focused on leaf chemicals. None addressed latex exudation and chemicals, which are also important to plant defence (Agrawal & Konno, 2009). To the best of our knowledge, our study is the first to examine the amount of latex exudation and chemical composition when damaged by different herbivores and compare the induction of introduced and native plant populations. We found that invasive populations secreted more latex than native populations did when damaged by a specialist herbivore, and that there were opposite responses of latex tannins and flavonoids when damaged by generalist herbivore. Our results suggest divergent selection on the physical and chemical properties of latex in the introduced range. The quantity of latex produced was more sensitive to herbivore identity in the introduced range but the specificity of the latex chemical response was retained. Studying the expression of different defence traits in novel conditions may help to understand the ecology and evolution of plant defence and invasive mechanism.

Acknowledgements

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

We would like to thank Chen Huaixia in Hubei University for latex chemical analysis and Xuefang Yang for field assistance. This study was supported by the National Program on Key Basic Research Project (973 Program) (2012CB114104 to J.D.), a Foreign Visiting Professorship of the Chinese Academy of Sciences (O929361H02to E.S.), US-NSF (DEB 0820560 to E.S.) and Association of youth innovation promotion of the Chinese Academy of Sciences (Y329341H02 to Y.W.)

References

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

Supporting Information

  1. Top of page
  2. Abstract
  3. Introduction
  4. Materials and methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
FilenameFormatSizeDescription
een12054-sup-0001-TableS1.docWord document47KTable S1. Native and invasive populations of Triadica sebifera that were used in this study.

Please note: Neither the Editors nor Wiley Blackwell are responsible for the content or functionality of any supporting materials supplied by the authors. Any queries (other than missing material) should be directed to the corresponding author for the article.