SEARCH

SEARCH BY CITATION

Keywords:

  • Acyrthosiphon pisum (pea aphid);
  • elevated CO2;
  • ethylene;
  • Medicago truncatula ;
  • nitrogen (N) metabolism;
  • resistance

Summary

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  9. Supporting Information
  • The performance of herbivorous insects is greatly affected by plant nutritional quality and resistance, which are likely to be altered by rising concentrations of atmospheric CO2.
  • We previously reported that elevated CO2 enhanced biological nitrogen (N) fixation of Medicago truncatula, which could result in an increased supply of amino acids to the pea aphid (Acyrthosiphon pisum). The current study examined the N nutritional quality and aphid resistance of sickle, an ethylene-insensitive mutant of M. truncatula with supernodulation, and its wild-type control A17 under elevated CO2 in open-top field chambers.
  • Regardless of CO2 concentration, growth and amino acid content were greater and aphid resistance was lower in sickle than in A17. Elevated CO2 up-regulated N assimilation and transamination-related enzymes activities and increased phloem amino acids in both genotypes. Furthermore, elevated CO2 down-regulated expression of 1-amino-cyclopropane-carboxylic acid (ACC), sickle gene (SKL) and ethylene response transcription factors (ERF) genes in the ethylene signaling pathway of A17 when infested by aphids and decreased resistance against aphids in terms of lower activities of superoxide dismutase (SOD), peroxidase (POD), and polyphenol oxidase (PPO).
  • Our results suggest that elevated CO2 suppresses the ethylene signaling pathway in M. truncatula, which results in an increase in plant nutritional quality for aphids and a decrease in plant resistance against aphids.

Introduction

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

Global atmospheric CO2 concentrations have been increasing at an accelerating rate from 280 ppm before industrialization to 396 ppm in Feburary 2013 (Mauna Loa Observatory: NOAA-ESRL), and are anticipated to reach at least 550 ppm by the year 2050 (IPCC, 2007). Elevated CO2 is expected to enhance crop yields by increasing photosynthetic rates and water-use efficiencies, particularly in C3 crops. Observed increases in yield, however, have not always matched theoretical expectations in CO2-enrichment experiments (Ainsworth & Long, 2005; Long et al., 2005), perhaps because the theory does not include interactions between plants and herbivorous insects.

The performance of aphids and other herbivorous insects is affected by bottom-up effects of host plants in terms of nutritional status and chemical and physical defenses (Awmack & Leather, 2002). With respect to nutritional status, aphids feed exclusively on phloem (Douglas, 2003), which provides a protein : carbohydrate ratio (mainly amino acids : sugars) as low as 1 : 10 (w/w) (Nowak & Komo, 2010). Although aphids have evolved to adapt to this nutrient-poor substrate, they are still able to discriminate among host plants with low and high nitrogen (N) concentrations and tend to prefer plants with higher N concentrations (Nowak & Komo, 2010). Moreover, N-fertilized plants enhance aphid population growth because of the increased concentration of amino acids in the phloem (Honek, 1991; Petitt et al., 1994; Ponder et al., 2000). Thus, it seems that the N nutritional status of the host plant is an important determinant of aphid development and fecundity.

With respect to plant defenses, once aphid stylets penetrate the epidermis, the plant triggers a common defensive response based on reactive oxygen species (ROS) by activating superoxide dismutase (SOD) and peroxidase (POD) (Moloi & van der Westhuizen, 2006). A further line of defense involves the rapid synthesis and polymerization of phenolic compounds in the cell wall (Matern & Kneusel, 1988). During this process, polyphenol oxidase (PPO) and phenylalanine ammonia lyase (PAL) are key secondary metabolism enzymes that mediate plant resistance against aphids (He et al., 2011). Effects of elevated CO2 on crop yields should consider how elevated CO2 alters host nutrition and host defenses relative to aphids and other herbivores.

Elevated CO2 reduces the nutritional quality of some nonleguminous C3 plants by decreasing the N concentration (Ainsworth & Long, 2005; Ainsworth & Rogers, 2007), which may consequently increase the developmental time and reduce the fecundity and fitness of leaf-chewing insects (Coll & Hughes, 2008). However, N concentrations in legumes were rarely affected by elevated CO2 because of the enhancement of biological N fixation (BNF), which counteracts the adverse effect of elevated CO2 on leaf-chewing insects (Karowe, 2007; Taub & Wang, 2008; Karowe & Migliaccio, 2011). For a sap-sucking insect like the pea aphid (Acyrthosiphon pisum), the increased BNF in legumes under elevated CO2 increases available N and thereby increases aphid numbers (Guo et al., 2013). When BNF is suppressed by artificial mutation, however, BNF cannot satisfy the increased demand for N that occurs under elevated CO2, and aphid numbers do not increase (Guo et al., 2013). Thus, it appears that enhanced BNF is necessary for the positive response of the pea aphid to elevated CO2.

In legumes, BNF is regulated by several hormone signaling pathways, including the ethylene signaling pathway. The involvement of the phytohormone ethylene in nodulation was initially proposed based on studies showing that the application of exogenous ethylene or its biosynthetic precursor 1-amino-cyclopropane-carboxylic acid (ACC) suppresses nodulation, and, conversely, that application of chemical inhibitors of ethylene perception (i.e. Ag+) or biosynthesis (i.e. the amino ethoxyvinyl glycine, AVG) increases nodule numbers (Ma et al., 2002; Penmetsa et al., 2003, 2008). Once the key gene Mtskl in the ethylene-perception pathway was mutated in Medicago truncatula, the resulting ethylene-insensitive mutant, sickle, produced more nodules than the wild-type, and its nitrogenase activity was increased about two times (Penmetsa & Cook, 1997).

In addition to having a key role in the regulation of BNF, ethylene is the most important hormone involved in plant resistance against pathogens and pests. The expression of genes involved in ethylene production and ethylene signaling (ACC oxidase and ethylene-responsive elements) are up-regulated in response to aphid infestation (Moran et al., 2002; Divol et al., 2005). Ethylene is also responsible for the regulation of ROS and downstream defensive enzymes against aphids (Jung et al., 2009). The sickle mutant showed increased sensitivity to Rhizoctonia solani and other pathogens on legumes and cereals (Penmetsa et al., 2008). Thus, it is reasonable to speculate that the ethylene-insensitive mutant sickle, which produces more nodules and has enhanced BNF as well as reduced resistance to aphids relative to the wildtype, will supply more N nutrition to aphids and be less resistant to aphids.

The results of recent studies indicate that elevated CO2 fine-tunes phytohormone signaling pathways when plants encounter biotic stress, as indicated by enhanced induced defenses derived from the salicylic acid signaling pathway and reduced jasmonic acid-dependent defense (Zavala et al., 2008; Sun et al., 2011; Guo et al., 2012). Furthermore, elevated CO2 tends to increase ethylene production in healthy plants but down-regulates the expression of downstream genes in the ethylene signaling pathway when plants are attacked by Japanese beetles (Seneweera et al., 2003; Casteel et al., 2008). Although it is well established that the ethylene signaling pathway mediates plant resistance against pathogens and aphids, it is unclear whether elevated CO2, by modulating the ethylene signaling pathway, simultaneously alters ethylene-dependent defense as well plant nutrition for aphids and other herbivores.

Here, we hypothesized that elevated CO2 would decrease the responses of the ethylene signaling pathway, which would directly reduce ethylene-dependent plant defenses while indirectly enhancing N availability for aphids via an increase in nodulation, such that pea aphid abundance would be greater under elevated CO2 than under ambient CO2. To test this hypothesis, we used sickle (a supernodulating mutant of M. truncatula that is insensitive to ethylene) and the wild-type A17 to determine how elevated CO2 affects the interaction between M. truncatula and the pea aphid via the ethylene signaling pathway. The specific goals were to determine whether the supernodulating mutant sickle grows better and has higher N metabolism than the wild-type under elevated CO2; whether elevated CO2 affects the resistance of the two genotypes against the pea aphid and consequently affects pea aphid feeding behavior and abundance; and whether the ethylene signaling pathway is involved in the regulation of the plant–aphid interaction under elevated CO2.

Materials and Methods

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

Atmospheric CO2 concentration treatments

The research described in the following sections was performed in eight octagonal, open-top field chambers (OTCs; 4.2 m in diameter and 2.4 m high) at the Observation Station of the Global Change Biology Group, Institute of Zoology, Chinese Academy of Science in Xiaotangshan County, Beijing, China (40°11′N, 116°24′E). The atmospheric CO2 concentration treatments were as follows: current atmospheric CO2 concentrations (c. 390 μl l−1); and elevated CO2 concentrations (750 μl l−1, predicted concentration at the end of this century) (IPCC, 2007). Four blocks were used for CO2 treatment, and each block contained paired OTCs, one with ambient and one with elevated CO2. CO2 concentration in each OTC was monitored and adjusted with an infrared CO2 analyzer (Ventostat 8102; Telaire Company, Goleta, CA, USA) once every min to maintain relatively stable CO2 concentrations. The measured CO2 concentrations throughout the experiment (mean ± SD d–1) were 395 ± 22 ppm in the ambient CO2 chambers and 752 ± 33 ppm in the elevated CO2 chambers. The auto-control system for maintaining the CO2 concentrations, as well as specifications for the OTCs, is detailed in Chen et al. (2005). The tops of the OTCs were covered with nylon net to exclude insects. Air temperatures were measured three times per d throughout the studies and did not differ significantly between the two treatments (22.7 ± 1.9°C in OTCs with ambient CO2 vs 24.2 ± 2.0°C in OTCs with elevated CO2).

Aphids

The pea aphid Acyrthosiphon pisum Harris. was obtained from the laboratory of Dr Feng Cui (Institute of Zoology, Chinese Academy of Science). The nymphal instars from the same parthenogenetic pea aphid female were reared on Vicia faba with 14 h light (25°C) : 10 h dark (22°C) in photoclimate chambers (Safe PRX-450C, Ningbo, China).

Host plants and rhizobium inoculation

The supernodulating mutant sickle of M. truncatula Gaertn. and its wild-type background (cv Jemalong A17) were kindly provided by Professor Douglas Cook (Department of Plant Pathology and Microbiology, Texas A&M University, TX, USA). Visual assessments for delayed petal senescence and leaf abscission were conducted on glasshouse-grown plants (Penmetsa & Cook, 1997).

After seeds were chemically scarified and surface-sterilized by immersion in concentrated H2SO4 for 5 min, they were rinsed with sterilized water several times. The seeds were placed in Petri dishes containing 0.75% agar, kept in the dark at 4°C for 2 d, and then moved to 25°C for 2 d to germinate. The germinated seeds were sown on sterilized soil and inoculated 2 d later with the bacterium Sinorhizobium meliloti 1021, which was provided by Professor Xinhua Sui (Department of Microbiology, College of Biological Sciences, Chinese Agricultural University). S. meliloti had been cultured on YM medium (H2O, 1000 ml; yeast, 3 g; mannitol, 10 g; KH2PO4, 0.25 g; K2HPO4, 0.25 g; MgSO4·7H2O, 0.1 g; NaCl, 0.1 g, pH 7.0–7.2) for 3 d at 28°C to obtain an approximate cell density of 108 ml−1. At sowing, each seedling was inoculated with 0.5 ml of this suspension. After they had grown in sterilized soil for 2 wk, the M. truncatula seedlings were individually transplanted into plastic pots (35 cm diameter and 28 cm height) containing sterilized loamy field soil (organic carbon, 75 g kg−1; N, 500 mg kg−1; P, 200 mg kg−1; K, 300 mg kg−1) and placed in OTCs on 27 March 2012. Each OTC contained 40 plants.

Medicago truncatula plants were maintained in the OTCs for 75 d from seedling emergence to the end of the experiment (27 March to 7 June 2012). Pot placement was re-randomized within each OTC once every wk. No chemical fertilizers or insecticides were used. Water was added to each pot every 2 d. On 18 May 2012, after the plants had been in the OTCs for 6 wk, they were used for the three groups of assays described in the following sections.

Aphid reproduction and feeding behavior as affected by plant genotype and CO2 concentration (group 1)

Six plants of each genotype per OTC were randomly selected, and each was infested with five apterous, fourth-instar nymphs. The plants with these nymphs were individually caged (80 mesh gauze). Six other plants of each genotype per OTC were not infested but were caged and served as controls. Aphids on each plant were counted 7, 11, 15, 19, and 23 d after inoculation.

From 19 to 25 May 2012, another six plants of each genotype per OTC (96 plants in total) were randomly selected as host plant to evaluate the feeding behavior of aphids using an electrical penetration graph (EPG) technique. EPG signals have been correlated with aphid activities as well as with tissue locations of the stylet tips (Tjallingii & Hogen-Esch, 1993). Eight plants were placed in a Faraday cage to avoid noise and interference. Each plant was infested with one apterous adult pea aphid for 8 h to record feeding behavior using the EPG method. The EPG method is a powerful tool for determining, in real time, the locations and activities of the aphid stylet, including probing, salivation into sieve elements, and passive uptake of phloem sap (Walker, 2000). The feeding behavior of pea aphids on the plants was studied as described in Gao et al. (2008) with some modifications. Twenty-four biological replicates (six plants in each of four OTCs) were included for each genotype (A17 and sickle) under each CO2 concentration (96 plants in total). As noted, one apterous adult pea aphid was placed on a single trifoliate leaf, and its feeding behavior was monitored. An eight-channel amplifier simultaneously recorded eight individual aphids on separate plants (on two A17 plants and two sickle plants under ambient CO2, and on two A17 plants and two sickle plants under elevated CO2) for 8 h and this experiment was replicated for 12 times. Waveform patterns in this study were scored according to categories described by Tjallingii & Hogen-Esch (1993): nonpenetration (np); pooled pathway phase activities (C); salivary secretion into sieve elements (E1); phloem ingestion (E2); derailed styled (F); and xylem ingestion (G).

Plant growth trait and N fixation-related genes as affected by plant genotype and CO2 concentration (group 2)

On 7 June 2012, six undamaged plants were harvested for measurement of Chl content, biomass, pod number, nodule number and gene expression. Leaves and roots (0.05 g) of each plant were harvested and immediately stored in liquid nitrogen for measurement of gene expression. The quantification of gene expression is described in the following paragraphs.

Leaf Chl content was determined with a Minolta SPAD-502DL (Konica Minolta Sensing Inc., Osaka, Japan), which measures leaf transmittance at two wavelengths: red (c. 660 nm) and near-infrared (c. 940 nm). SPAD readings were taken on the fourth terminal mature trifoliate leaf from the base of the shoot. The SPAD sensor was placed randomly on leaf mesophyll tissue.

Roots of each plant were carefully removed from the soil and washed. A stereomicroscope was used to count the nodules on the entire root system of each plant. The pod numbers per plant were also determined. The shoots and roots of each plant were collected, oven-dried (65°C) for 72 h, and weighed.

Plant enzyme activity, free amino acid concentration, and gene expression as affected by plant genotype, CO2 concentration, and aphid infestation (group 3)

On 19 May 2012, two plants of each genotype per OTC were randomly selected, and each was infested with 50 apterous fourth-instar nymphs; a fine paintbrush was used to transfer the nymphs to these plants, which were individually caged (80 mesh gauze). Two other uninfested control plants of each genotype per OTC were caged in the same way. After 48 h, leaves of infested plants and leaves of the same number of uninfested control plants were harvested and immediately stored in liquid nitrogen for chemical analysis. The quantification of enzyme activities, free amino acid concentration, and ethylene signaling pathway-related gene expression are described in the following paragraphs.

The activities of key enzymes involved in N assimilation and transamination, including glutamine synthetase/glutamate synthase (GS/GOGAT), glutamate oxalate transaminase (GOT), and glutamine phenylpyruvate transaminase (GPT), were quantified (Schoenbeck et al., 2000; Andrews et al., 2004) using frozen leaf tissue (c. 0.5 g per plant). Once the tissue was ground to a fine powder, leaves from three plants of the same treatment were combined to form one sample from each OTC. The unit of replication for statistical analyses was the OTC (n = 4). An extract was obtained by grinding each leaf sample in 50 mM Tris HCl buffer (pH 7.8, 3 ml g−1 of leaf tissue) containing 1 mM MgCl2, 1 mM EDTA, 1 μM β-mercaptoethanol, and 1% (w/v) polyvinylpolypyrro-lidone. This extract was immediately frozen for later use. For assays, the thawed extract was centrifuged at 13 000 g for 10 min, and the enzymatic activities were measured in the supernatant as described by Glévarec et al. (2004) for GS, by Suzuki et al. (2001) for GOGAT, and by Asthir & Bhatia (2011) for GOT and GPT. Protein concentrations of leaves and roots were measured using BSA as a standard (Bradford, 1976).

The defensive enzymes, including lipoxygenase (LOX), proteinase inhibitors (PIs), PPO, POD, and PAL, were extracted from 0.1 g frozen leaf tissue by grinding them in a 50 mM Tris HCl buffer (pH 7.8, 3 ml g−1 of leaf tissue) containing 7% polyvinylpolypyrrolidine, 1.67 mM phenylthiourea, 0.3 M KCl, and 0.4 mM ascorbic acid. For assays, the thawed extract was centrifuged at 13 000 g for 10 min, and enzyme activity was measured in the supernatant. The activities of these defense enzymes were determined according to Guo et al. (2012).

For quantification of amino acid concentrations in phloem, phloem exudates were obtained from three trifoliate leaves per plant using the EDTA exudation technique of Douglas (1993). The amino acids in each sample were analyzed by reverse-phase high-performance liquid chromatography (HPLC) with pre-column derivatization using o-phthaldialdehyde (OPA) and 9-fluorenylmethyloxycarbonyl (FMOC). Amino acids were quantified by comparison with the AA-S-17 (PN: 5061-3331; Agilent Technologies, Palo Alto, CA, USA) reference amino acid mixture, supplemented with asparagine, glutamine, and tryptophan (Sigma-Aldrich). Standard solutions were prepared from a stock solution by diluting with 0.1 M HCl. Free amino acid concentrations of each of the five standard solutions were 250, 100, 50, 25, and 10 pmol μl−1. Before each sample was injected into the HPLC, 10 μl of the amino acid sample was mixed with 20 μl of sodium borate buffer (0.4 N, pH 10.4), 10 μl of OPA, 10 μl of FMOC, and 50 μl of water. The analysis was performed using an Agilent 1100 HPLC system (Agilent). A reverse-phase Agilent Zorbax Eclipse C18 column AAA (5 μm, 250 mm × 4.6 mm) and fluorescence detector were used for the chromatographic separation. The column was maintained at 35°C with a gradient (1 ml min−1 flow) programmed as follows: 98/2 (1 min) to 43/57 (25 min) to 0/100 (34 min) to 98/2 (42 min hold) of eluent A/eluent B. Eluent A was a 40 mM disodium phenyl phosphate buffer (pH 7.8 adjusted with sodium hydroxide). Eluent B was 45% acetonitrile, 45% methanol, and 10% water. Chemstation Plus Family for LC software was used for data acquisition and analysis. Amino acid concentrations were quantified by comparison of sample peak areas to standard curves of 20 reference amino acids (Agilent).

The expressions of three genes involved in N fixation (ENOD, nifH, and nodF) and three genes in the ethylene signaling pathway (ACC, SKL and ERF) were measured by quantitative reverse transcription polymerase chain reaction. Each combination was replicated four times for biological repeats, and each biological repeat had three technical repeats. The RNA easy Mini Kit (Qiagen) was used to isolate total RNAs from M. truncatula leaves or roots (0.05 g from samples stored at −70°C), and 1 μg of RNAs was used to generate the cDNAs. The mRNA levels of the six target genes were quantified by real-time quantitative PCR (qPCR). Specific primers for each gene were designed from the M. truncatula expressed sequence tag sequences using PRIMER5 software (Supporting Information Table S1). The PCR reactions were performed in a 20 μl total reaction volume including 10 μl of 2 × SYBRs Premix EX Taq™ (Qiagen) master mix, 5 mM of each gene-specific primer, and 1 μl of cDNA template. Reactions were carried out on the Mx 3500P detection system (Stratagene), and the parameters were as follows: 2 min at 94°C; followed by 40 cycles of 20 s at 95°C, 30 s at 56°C, and 20 s at 68°C; and finally one cycle of 30 s at 95°C, 30 s at 56°C, and 30 s at 95°C. This PCR protocol produced the melting curves, which were used to judge the specificity of PCR products. A standard curve was derived from the serial dilutions to quantify the copy numbers of target mRNAs. β-actin and pnp served as internal qPCR standards for the analysis of plant and bacterial gene expression, respectively (Vernié et al., 2008). The relative level of each target gene was standardized by comparing the copy numbers of target mRNA with β-actin or pnp (the housekeeping gene) copy numbers, which remain constant under different treatment conditions. The β-actin mRNAs of the control were examined in every PCR plate to eliminate systematic error. The fold-changes of target genes were calculated using the inline image method.

Statistical analysis

All data were checked for normality and equality of residual error variances and were appropriately transformed (log or square root) as needed to satisfy the assumptions of the ANOVA. For the amino acids, we performed a principal components analysis (PCA) on the correlations among the 20 response variables and then performed factor rotation using the Varimax method (Rasmussen et al., 2008). A split-split plot design was used to analyze the univariate responses of the growth traits, enzyme activities, and rotated factors of amino acids in plants (ANOVA, PASW, 2009). In the following ANOVA model, CO2 and block (a pair of OTCs with ambient and elevated CO2) were the main effects, M. truncatula genotype was the subplot effect, and aphid infestation was the sub-subplot effect:

  • display math

where C is the CO2 treatment (i = 2), B is the block (j = 4), G is the M. truncatula genotype (k = 2), and H is the aphid infestation treatment (l = 2). Xijklm represents the error because of the smaller scale differences between samples and variability within blocks (ANOVA, SAS Institute, Cary, NC, USA). Effects were considered significant if P < 0.05. The effect of block and the interactive effects of block and other factors were not significant (P > 0.45), and the effect of block and its interactive effects with other factors were not presented so as to simplify the presentation. Tukey's multiple range tests were used to separate means when ANOVAs were significant. For quantifying the feeding behavior of pea aphids on different M. truncatula genotypes under two CO2 concentrations, a split-plot design was also applied, with CO2 and block as the main effects and M. truncatula genotype as the subplot effect. Aphid abundance was analyzed by repeated-measures ANOVA.

Results

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

Aphid abundance

Elevated CO2 enhanced pea aphid population growth on A17 and sickle plants (Fig. 1). Regardless of CO2 concentration, aphids were more abundant on sickle than on A17 plants (Fig. 1).

image

Figure 1. Abundance of pea aphids (Acyrthosiphon pisum; number per plant) when fed on two Medicago truncatula genotypes (wild-type A17 and ethylene-insensitive sickle mutants) grown under ambient CO2 (ACO2) and elevated CO2 (ECO2). Each value represents the mean (± SE) of four replicates. Significant differences: *, < 0.05.

Download figure to PowerPoint

Aphid feeding behavior

Elevated CO2 decreased the percentage of time aphids spent salivating into sieve elements (E1 phase) on A17 plants but increased phloem sap ingestion (E2 phase) for both plant genotypes (Fig. 2). The aphids had a shorter E1 phase and a longer E2 phase on sickle than on A17 plants under ambient CO2. Under elevated CO2, however, aphid feeding was not significantly affected by plant genotype (Fig. 2).

image

Figure 2. The percentage of time pea aphids (Acyrthosiphon pisum) spent in various feeding activities on two Medicago truncatula genotypes (A17 and sickle) during an 8 h exposure to ambient CO2 (ACO2) and elevated CO2 (ECO2). ‘Nonpenetration’, stylets are outside the plant; ‘pathway’, mostly intramural probing activities between mesophyll or parenchyma cells; ‘salivation’, aphids are injecting watery saliva into the sieve element; ‘phloem ingestion’, aphids are ingesting the phloem sap; ‘xylem’, stylet penetration of tracheary elements; ‘derailed stylet’, stylets are exhibiting penetration difficulties. Values are the means (± SE) of 24 biological replicates. Different lowercase letters indicate significant differences among treatments (Tukey's multiple range test, P < 0.05).

Download figure to PowerPoint

Plant growth traits and expression of N-fixation genes

Carbon dioxide concentration and genotype had significant effects on plant Chl content, nodule numbers, biomass and pod numbers (Table S2) In the absence of pea aphids, elevated CO2 significantly increased Chl content by 18.6 and 12.6%, nodule numbers by 68.5 and 26.8%, biomass by 25.9 and 33.1%, and pod numbers by 31.6 and 21.9% for A17 and sickle plants, respectively (Fig. 3). Regardless of CO2 concentration, Chl content, nodule number, biomass, and pod number were higher for sickle than for A17 plants (Fig. 3).

image

Figure 3. Growth traits of two Medicago truncatula genotypes (A17 and sickle) grown under ambient CO2 (ACO2; open bars) and elevated CO2 (ECO2; closed bars) without pea aphid (Acyrthosiphon pisum) infestation. (a) Chl content; (b) nodule numbers per plant; (c) biomass; and (d) pod numbers. Each value represents the mean (± SE) of four replicates. Different lowercase letters indicate significant differences between ACO2 and ECO2 within the same genotype. Different uppercase letters indicate significant differences between genotypes within the same CO2 treatment as determined by Tukey's multiple range test at < 0.05.

Download figure to PowerPoint

Elevated CO2 increased the expression of ENOD, nifH, and nodF in both genotypes (Fig. 4). Regardless of CO2 concentration, the expression of these three N-fixation genes was higher in sickle than in A17 plants (Fig. 4).

image

Figure 4. Expression of genes involved in nodulation and biological N fixation (ENOD, nifH, nodF) in two Medicago truncatula genotypes (A17 and sickle) grown under ambient CO2 (ACO2; open bars) and elevated CO2 (ECO2; closed bars). Values indicate fold-change in expression based on quantitative PCR, and each value represents the mean (± SE) of four replicates. Different lowercase letters indicate significant differences between ACO2 and ECO2 within the same genotype. Different uppercase letters indicate significant differences between genotypes within the same CO2 treatment as determined by Tukey's multiple range test at < 0.05.

Download figure to PowerPoint

N assimilation and transamination enzymes of plants

Carbon dioxide concentration, genotype, aphid infestation and the interaction between CO2 and aphid infestation significantly affected the activity of GS and GOT in plant leaves (Table S3). CO2 concentration, genotype, aphid infestation, the interaction between genotype and aphid infestation, as well as the interaction among CO2, genotype and aphid infestation, significantly affected the activity of GOGAT (Table S3). CO2 concentration, genotype, aphid infestation, the interaction between CO2 and aphid infestation, as well as the interaction among CO2, genotype and aphid infestation, significantly affected the activity of GOT (Table S3). Furthermore, all factors, with the exception of the interaction between CO2 concentration and aphid infestation, significantly influenced the activity of GPT in plant leaves (Table S3).

Without aphid infestation, elevated CO2 significantly increased the activities of GOGAT and GPT in A17 and of GS in sickle (Fig. 5). After a 48 h period of aphid infestation, elevated CO2 increased the activities of all the N assimilation and transamination enzymes measured in A17 leaves and increased the activity of GS, GOGAT, and GOT in sickle leaves (Fig. 5). Regardless of aphid infestation, GS and GOGAT activities were higher in sickle than in A17 leaves at both CO2 concentrations. GOT activity was higher in sickle than in A17 leaves under elevated CO2 regardless of aphid infestation (Fig. 5).

image

Figure 5. Activities of enzymes involved in nitrogen (N) metabolism in two Medicago truncatula genotypes (A17 and sickle) grown under ambient CO2 (ACO2) and elevated CO2 (ECO2) and with (+PA) and without pea aphid (Acyrthosiphon pisum) infestation. (a) Glutamine synthetase (GS); (b) glutamate synthase (GOGAT); (c) glutamate oxalate transaminase (GOT); and (d) glutamine phenylpyruvate transaminase (GPT). Each value represents the mean (± SE) of four replicates. Different lowercase letters indicate significant differences among the combinations of aphid treatment and CO2 concentrations within the same genotype. Different uppercase letters indicate significant differences between genotypes within the same CO2 treatment and aphid treatment as determined by Tukey's multiple range test at < 0.05.

Download figure to PowerPoint

Amino acid concentration in plants

The EDTA method and HPLC technique were used to measure the relative composition of amino acids in the phloem sap of M. truncatula. Because the responses of the 20 amino acids that were measured are not independent, we then used a PCA to reduce the number of phloem amino acid response variables to a new set of composite variables (Rasmussen et al., 2008). To facilitate interpretation of principal components, we subjected the first three principal components to factor rotation with the most common form of factor rotation, varimax rotation, and we retained three rotated factors (RF1, RF2, and RF3, which accounted for 77% of the total variance) (Fig. 6). As the values of the rotated factor increase, those variables that load heavily and positively (loading ≥ + 0.5) also increase, while those variables that load heavily but negatively (loading ≤ −0.5) decrease. The standardized univariate responses of these variables are shown in Fig. 7 to facilitate the interpretation of the multivariate responses and to allow a closer inspection of those variables loading heavily onto RF1, RF2, and RF3.

image

Figure 6. The loadings for each individual amino acid of Medicago truncatula phloem sap onto the first three rotated factors (RFs). The individual amino acids loading heavily either positively (loading ≥ 0.5) or negatively (loading ≤ −0.5) are highlighted in black. These multivariate responses can be interpreted as increasing as the positively loading variables increase and decreasing as the negatively loading variables increase.

Download figure to PowerPoint

image

Figure 7. The mean response of rotated factors (a, c, e) and the standardized univariate response (b, d, f) of individual amino acids in phloem sap of two Medicago truncatula genotypes (A17 and sickle) to CO2 concentration (ambient CO2, ACO2; elevated CO2, ECO2), genotype, pea aphid (Acyrthosiphon pisum) infestation (+PA), and their interactions. Different lowercase letters indicate significant differences among the combinations of aphid treatments and CO2 concentrations within the same genotype. Different uppercase letters indicate significant differences between genotypes within the same CO2 treatment and aphid treatment as determined by Tukey's multiple range test at < 0.05. The underlined individual amino acids indicate these are essential amino acids for aphids.

Download figure to PowerPoint

Eleven amino acids (Gln, Ile, Val, Cys, Asn, Arg, Leu, Gly, Met, His, and Phe; seven essential amino acids and four nonessential amino acids for aphid) loaded heavily and positively onto RF1 (Fig. 6a). CO2 concentration, genotype, aphid infestation and the interaction between CO2 and aphid infestation significantly affected the RF1 (Table S4). Elevated CO2 without aphid infestation increased RF1 in the phloem of sickle plants. Elevated CO2 with aphid infestation increased RF1 in the phloem of both genotypes (Fig. 7a,b). Trp, Tyr, Pro, Ala, Glu, and Ser (one essential amino acid and five nonessential amino acids for aphid) loaded heavily and positively onto RF2 (Fig. 6b). Genotype, aphid infestation, the interaction between CO2 and genotype, as well as the interaction among CO2, genotype and aphid infestation significantly affected the RF2 (Table S4). Elevated CO2 without aphid infestation decreased RF2 in the phloem of both A17 and sickle plants. Elevated CO2 with aphid infestation significantly increased RF2 in the phloem of A17 plants but decreased RF2 in the phloem of sickle plants (Fig. 7c,d). Thr and His (both are essential amino acids for aphid) loaded heavily and positively onto RF3 (Fig. 6c). CO2 concentration, genotype and their interaction significantly affected RF3. Regardless of aphid infestation, elevated CO2 increased RF3 in the phloem of A17 and sickle plants (Fig. 7e,f).

Plant defensive enzymes

All factors, with the exception of the interaction between CO2 concentration and genotype, significantly influenced the activity of SOD in plant leaves (Table S3). Genotype, the interaction between CO2 and genotype, the interaction between CO2 and aphid infestation, and the interaction between genotype and aphid infestation significantly affected the activity of POD in plant leaves (Table S3). CO2 concentration, genotype, aphid infestation, the interaction between CO2 and genotype, and the interaction between genotype and aphid infestation significantly affected the activity of PPO and PAL (Table S3).

Without aphid infestation, elevated CO2 significantly increased PAL activity in A17 plants (Fig. 8). After a 48 h period of aphid infestation, elevated CO2 decreased the activities of SOD, POD, and PPO but increased the activity of PAL in A17 plants but not in sickle plants (Fig. 8). With aphid infestation but regardless of CO2 concentration, SOD and POD activities were lower in sickle than in A17 plants. With aphid infestation and ambient CO2, PPO activity was lower in sickle than in A17 plants. With aphid infestation and elevated CO2, PAL activity was lower in sickle than in A17 plants. Aphid infestation significantly increased the activities of SOD, POD, and PAL, regardless of CO2, and increased PPO activity under ambient CO2 in A17 plants but did not affect the activity of these enzymes in sickle plants (Fig. 8).

image

Figure 8. Activities of resistance enzymes of two Medicago truncatula genotypes (A17 and sickle) grown under ambient CO2 (ACO2) and elevated CO2 (ECO2) with (+PA) and without pea aphid (Acyrthosiphon pisum) infestation. (a) Superoxide dismutase (SOD); (b) peroxidase (POD); (c) polyphenol oxidase (PPO); (d) phenylalanine ammonia lyase (PAL). Each value represents the mean (± SE) of four replicates. Different lowercase letters indicate significant differences among the combinations of aphid treatment and CO2 concentrations within the same genotype. Different uppercase letters indicate significant differences between genotypes within the same CO2 treatment and aphid treatment as determined by Tukey's multiple range test at < 0.05.

Download figure to PowerPoint

Expression of ethylene signaling pathway genes

Without aphid infestation, elevated CO2 significantly down-regulated the expression of ACC, SKL, and ERF in A17 plants and down-regulated ACC in sickle plants (Fig. 9). After a 48 h period of aphid infestation, elevated CO2 down-regulated expression of ACC, SKL, and ERF in A17 plants and down-regulated expression of ACC in sickle plants (Fig. 9). Regardless of CO2 concentration, aphid infestation significantly increased the expression of ACC, SKL, and ERF in A17 plants but only that of ACC in sickle plants (Fig. 9).

image

Figure 9. Expression of genes 1-amino-cyclopropane-carboxylic acid (ACC), sickle gene (SKL) and ethylene response transcription factors (ERF) in the ethylene signaling pathway in Medicago truncatula as affected by CO2 concentration (ambient (ACO2) vs elevated (ECO2)), plant genotype (A17, sickle), and pea aphid (Acyrthosiphon pisum) infestation. Values indicate fold-change in expression based on quantitative PCR, and each value represents the mean (± SE) of four replicates. Different lowercase letters indicate significant differences among the combinations of aphid treatment and CO2 concentrations within the same genotype. Different uppercase letters indicate significant differences between genotypes within the same CO2 treatment and aphid treatment as determined by Tukey's multiple range test at < 0.05.

Download figure to PowerPoint

Discussion

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

Elevated CO2 affects herbivorous insects mainly by altering host plant nutritional quality and resistance (Awmack & Leather, 2002). Under ambient CO2, the ethylene-insensitive mutant sickle, which produces more nodules and exhibits a stronger BNF than the wild-type, grew better and was less resistant than the wild-type A17 and consequently supported higher numbers of pea aphids than A17. Although the increased growth under elevated CO2 could increase plant N demand (Daepp et al., 2000), the increased BNF in both genotypes under elevated CO2 provided sufficient N so that the plants could produce greater biomass and more pods than under ambient CO2. Our results indicate that elevated CO2 tends to suppress the ethylene signaling pathway in wild-type A17 plants, so that increased nodulation and BNF satisfy the increased N requirement for growth under elevated CO2. By decreasing the ethylene signaling pathway, however, elevated CO2 reduced plant resistance against the pea aphid. In summary, impairment of the ethylene signaling pathway by elevated CO2 has two important effects in M. truncatula: it up-regulates amino acid metabolism but reduces aphid resistance ability; and the increased plant growth and reduced resistance result in increased numbers of pea aphids per plant.

The response to elevated CO2 differs among insect feeding guilds (Robinson et al., 2012). Typically, elevated CO2 tends to prolong the development of chewing insects because it decreases the N content and increases secondary metabolites in host tissues (Coll & Hughes, 2008). By contrast, elevated CO2 has species-specific effects on phloem-sucking insects such as aphids, which obtain food from phloem sieve elements (Sun & Ge, 2011). Some aphid species exhibit increased fecundity, abundance, and survival under elevated CO2 (Pritchard et al., 2007). Although Newman proposed that aphid populations tend to be larger under elevated CO2 if host plants have higher N supplementation (Newman et al., 2003), evidence concerning how elevated CO2 affects both bottom effects on aphids (via host nutrition and resistance) has been lacking until the current study. Our previous study revealed that increases in pea aphid numbers under elevated CO2 depend on an increase in BNF and thus an increase in host plant amino acid metabolism (Guo et al., 2013). When BNF was suppressed by mutation, host plant amino acid metabolism was not increased by aphid infestation under elevated CO2. Because BNF is important in supporting the N nutrition of pea aphids, we speculated that pea aphids would be able to obtain more N from the supernodulating genotype sickle than from the wild-type A17.

As important indices of BNF, nodules were more abundant and expression of ENOD, nifH, and nodF was higher in sickle than in A17 plants. The increases in N metabolism leads to increased investment in Rubisco and other C assimilation-related enzymes (Gleadow et al., 1998), which in turn results in greater Chl content, biomass, and pod number in sickle than in A17 plants. Although elevated CO2 increased nodule number, gene expression involved in nodulation and BNF, Chl content, biomass and pod number for both genotypes, it did not change the growth advantages of sickle relative to A17. The improved growth traits of sickle had positive bottom-up effects on the aphid, as demonstrated by greater aphid abundance and feeding efficiency on sickle than on A17, regardless of CO2 concentrations.

Aphids have stylet-like mouthparts and feed mainly on phloem sap (Douglas, 2003). Most previous studies have measured aphid response to elevated CO2 as a function of whole-leaf composition rather than phloem sap composition (Robinson et al., 2012). The current study indicated that M. truncatula exhibited three patterns of individual amino acid concentrations in phloem sap in response to elevated CO2 and aphid infestation. When infested by aphids, the concentrations of amino acids loading on RF1 (seven essential amino acids and four nonessential amino acids) were higher in sickle than in A17 regardless of CO2 concentrations. However, elevated CO2 has a contrasting effect on the concentrations of amino acids loading on RF2 (mainly nonessential amino acids) in which sickle was higher than A17 plant under ambient CO2 but lower under elevated CO2. These results confirmed that sickle plants provide better N nutrition for aphids than A17 plants, and elevated CO2 tends to decrease this nutritional advantage of sickle when compared with A17. Although elevated CO2 increased the activities of N assimilation and transamination-related enzymes, and consequently increased amino acid concentration of infested plants of both genotypes, the pattern for this enhancement of amino acids differed between the two genotypes. Elevated CO2 increased individual amino acid concentrations loading onto RF1, RF2, and RF3 in A17 plants but only increased the concentration of amino acids loading onto RF1 in sickle plants. The relative improvement of N nutrition in response to elevated CO2 was less in sickle than in A17 plants because the basal N metabolism under ambient CO2 was much higher in sickle plants.

The regulation of nodule formation by ethylene signaling greatly affects the plant's ability to adapt to elevated CO2, in that enhanced BNF can satisfy the increased demand for N under elevated CO2 (Penmetsa & Cook, 1997). The current study showed that elevated CO2 down-regulated the expression of the ethylene signaling pathway genes ACC, SKL and ERF in A17 plants. Similarly, Casteel et al. (2012) found that elevated CO2 decreased the ethylene signaling pathway when attacked by Japanese beetles (Popillia japonica).This result suggests that, under elevated CO2, M. truncatula suppresses the ethylene signaling pathway so as to increase nodulation and BNF and thereby satisfy the increased demand for N. The ethylene signaling pathway in M. truncatula, however, has also been found to provide resistance against the pea aphid (Gao et al., 2008).

To access the amino acids in phloem sap, aphids must overcome a number of plant defense responses. One of the early plant responses to aphids is the release of ROS (Moloi & van der Westhuizen, 2006). Two other important defense enzymes are PPO and PAL, which are involved in the synthesis of phenolic compounds that may be absorbed by the salivary sheath of the aphid stylet. The further polymerization of phenolic compounds causes browning of cells in contact with the saliva, which is disadvantageous to aphid feeding (Jiang & Miles, 1993). In our study, aphid infestation increased the activities of SOD and POD (which are involved in ROS synthesis) and of PPO and PAL in A17 plants. In sickle plants, however, aphid infestation did not induce SOD or POD, which is consistent with a previous finding that the inhibition of ethylene synthesis or perception blocks the ROS response (de Jong et al., 2002). These results indicate that because its ethylene signaling pathway was mutated, sickle could not trigger the downstream ethylene-dependent defense in response to aphid infestation. This is consistent with our EPG finding that, relative to aphids feeding on A17, aphids feeding on sickle spend less time on salivation and more time on ingestion in phloem. Furthermore, elevated CO2 decreased the activities of SOD, POD, and PPO in infested A17 plants. As noted earlier, the decreased ethylene signaling pathway of M. truncatula under elevated CO2 affects the pea aphid in two ways, that is, by maintaining host N metablism and reducing host resistance.

Most plants are well adapted to process ‘extra’ carbon under elevated CO2, and this allows them to grow faster and larger. To satisfy the increased N demand under elevated CO2, legume plants evidently decrease their ethylene sensitivity so as to increase the formation of nodules and enhance BNF. Furthermore, considering the vital role of the ethylene signaling pathway in regulating plant resistance against aphid infestation, our results suggested that the down-regulation of the ethylene signaling pathway is accompanied by decreased plant resistance against the pea aphid. In other words, in decreasing the ethylene signaling pathway under elevated CO2 to match their N budgets and growth, plants sacrifice their resistance against aphids. Furthermore, given that the ethylene pathway was mutated in sickle plants, it is reasonable to expect that the mutated plants would lack one of the phytohormone signaling pathways that increases N metabolism in response to elevated CO2. Unexpectedly, although elevated CO2 only increased amino acids in RF1 for sickle plants infested with aphids, but increased amino acids in RF1, RF2, and RF3 for A17 plants infested with aphids, nodule numbers, expression of genes related to N fixation, growth traits, and N nutrition for pea aphids were all increased in sickle by elevated CO2. This suggests that the ethylene signaling pathway may not be the only phytohormone pathway regulating plant response to elevated CO2. The interaction between plant and aphid is coordinated by other interacting signaling phytohormones (such as jasmonic acid and salicylic acid pathways), except for ethylene (Felton & Korth, 2000). These signaling pathways are cross-talked in a complex network, which supports plants rapidly adapting to biotic and abiotic stresses by triggering an enormous regulatory mechanism. Among the three signaling pathways, jasmonic acid and ethylene signaling pathway has an antagonistic interaction with salicylic acid signaling pathway. The emerging data suggest that elevated CO2 tends to modulate these phytohormone signaling pathways, such as the jasmonic acid and salicylic acid pathways, that affect responses to insect herbivores (DeLucia et al., 2012; Sun et al., 2013; Zavala et al., 2013). Thus, additional research is needed to determine how multiple phytohormone signaling pathways are coordinately regulated by elevated CO2.

In summary, elevated CO2 increased pea aphid abundance on M. truncatula by affecting both host plant nutritional quality and resistance. Elevated CO2 decreased the ethylene-dependent resistance of wild-type M. truncatula against the pea aphid. On the other hand, the decrease in the ethylene signaling pathway increased the nodulation and BNF and thereby increased the phloem amino acids supporting aphid reproduction. The two effects of the ethylene signaling pathway would synergistically increase the fitness of pea aphids under elevated CO2. Because the supernodulating genotype sickle has higher amino acid metabolism and lower resistance, it is more suitable for the pea aphid than the A17 plant under ambient CO2, and the greater suitability of sickle in comparison to A17 is not changed by elevated CO2.

Acknowledgements

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

We thank Prof. Bruce Jaffee (University of California at Davis, CA, USA) for reviewing a draft of the manuscript. This project was supported by the ‘National Basic Research Program of China’ (973 Program) (no. 2012CB114103) and the National Nature Science Fund of China (nos. 31170390, 31000854, and 31221091).

References

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

Supporting Information

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

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

FilenameFormatSizeDescription
nph12484-sup-0001-TablesS1-S4.docWord document137K

Table S1 Primer sequences used for real-time quantitative PCR

Table S2 F- and P-values from MANOVAs for the effect of CO2 concentration and Medicago truncatula genotype on growth traits of two M. truncatula genotypes

Table S3 F- and P-values from MANOVAs for the effect of CO2 concentration, Medicago truncatula genotype, and pea aphid infestation on enzyme activities in leaves of two M. truncatula genotypes

Table S4 F- and P-values from MANOVAs for the effect of CO2 concentration, Medicago truncatula genotype, and pea aphid infestation on rotated factors of individual amino acids in phloem sap of two M. truncatula genotypes