Plants turn on induced defenses upon insect herbivory. In the current study, we evaluated the role of European corn borer (ECB) elicitors (molecules secreted by herbivores) that either induce/suppress defenses in Solanum lycopersicum (tomato) and Zea mays (maize), two very important crop plants that are grown for food and/or fuel throughout the world.
We used a combination of molecular, biochemical, confocal and scanning electron microscopy, caterpillar spinneret ablation/cauterization, and conventional insect bioassay methods to determine the role of ECB elicitors in modulating defenses in both tomato and maize crop plants.
Our results clearly demonstrate that the components present in the ECB saliva induce defense-related proteinase inhibitors in both tomato (PIN2) and maize (MPI). Presence of glucose oxidase in the ECB saliva induced defenses in tomato, but not in maize. However, ECB saliva induced genes present in the jasmonic acid biosynthesis pathway in both tomato and maize.
Although ECB saliva can induce defenses in both tomato and maize, our results suggest that host-specific salivary components are responsible for inducing host plant defenses. Proteomic analysis of ECB salivary elicitors and plant receptors/signaling mechanisms involved in recognizing different ECB elicitors remains to be determined.
Plants are sessile in nature and are constantly exposed to a wide range of environmental conditions, including both biotic and abiotic stresses. As a counter-measure, plants have evolved complex defense mechanisms to overcome these stresses. Upon insect herbivory, many plants rapidly activate induced defenses in response to the attack (Karban & Baldwin, 1997; Chen, 2008; Melotto et al., 2008). Induced defenses can be direct or indirect. Direct defenses can be any kind of plant characteristic that are activated upon insect infestation and can be anti-nutritional or toxic in nature (Chen, 2008). Indirect defenses include herbivore-induced plant volatiles that involve a tritrophic interaction between the plant, the insect and a predatory insect (Felton & Tumlinson, 2008). Elicitors present in the insect oral secretions (regurgitant) and/or saliva may either amplify/suppress the induced plant defenses (Turlings et al., 1990, 1993; Musser et al., 2002).
Lepidopteran insects produce two kinds of secretions from their mouth parts – oral secretions (OS) originating from the digestive system, and saliva produced by the salivary glands (Felton, 2008). Several classes of chemicals in the insect OS have been identified as key components that are involved in eliciting plant defenses. For example, OS of insects from different species contain fatty acid–amino acid conjugates (FACs) including N-17-hydroxylinolenoyl-l-glutamine (volicitin), sulfated fatty acids (caeliferins), and peptide fragments of ingested plant proteins (inceptins; Alborn et al., 1997, 2007; Schmelz et al., 2007). Caterpillar OS predominantly activate indirect plant defenses by enhancing the production of plant volatiles, thereby attracting predators of the attacking herbivore (Turlings et al., 1990, 1993; Alborn et al., 1997, 2007). However, several studies have shown that insect saliva can induce/suppress the activation of direct and indirect plant defenses (Musser et al., 2002; Delphia et al., 2006; Tian et al., 2012). For instance, glucose oxidase (GOX) from the labial glands of Helicoverpa zea (corn ear worm; CEW) suppresses induced chemical defenses and wound-induced accumulation of nicotine in tobacco (Musser et al., 2002). By contrast, H. zea GOX elicits both rapid and delayed-induced direct defenses in tomato (Tian et al., 2012). Interestingly, none of the identified elicitors from insect OS trigger defenses in tomato (Schmelz et al., 2009), suggesting that different plant species recognize herbivores by different mechanisms or cues. Taken together, these studies suggest that caterpillar OS and/or saliva can act as an important modulator in inducing/suppressing the plant indirect and direct defenses.
Ostrinia nubilalis (European corn borer; ECB) (Lepidoptera: Crambidae) is considered as the most common insect pest in the maize fields of the United States (Hutchison et al., 2010). Annual yield loss due to ECB damage in the United States accounts for > $1 billion dollars (Ostlie et al., 1997). ECB attacks all above-ground parts of maize, resulting in economic crop loss (Sunkula, 2006). Apart from maize, ECB also attacks several other crops and is considered as a highly polyphagous insect. Transgenic maize plants that express Bacillus thuringiensis (Bt) toxin have been widely used to control insect pests, including the ECB (Ostlie et al., 1997; James, 2010). However, management of ECB by conventional methods has been limited and there are concerns regarding the development of transgenic Bt resistance in field populations of ECB (Bolin et al., 1999). As an alternative crop protection strategy, we examined the elicitors present in the ECB OS and/or saliva that are responsible for inducing/suppressing plant defenses that can be used to develop novel pest management strategies.
Although there are several studies that report the induced defenses upon insect herbivory, very few studies have compared the induced defenses by an insect on two different host plants. The experimental system involved using Solanum lycopersicum (tomato) and Zea mays (maize), a dicot and monocot, as two economically important plant species. Our results demonstrate that ECB saliva contains the key elicitor(s) that is responsible for inducing defenses in host plants. However, the GOX present in the ECB saliva induced defenses only in tomato, but not in maize. Based on these observations, we suggest that ECB larvae utilize host-specific elicitor(s) in modulating plant defenses.
Materials and Methods
Plant and insect materials
Tomato (Solanum lycopersicum L.) (var. Better Boy) and maize (Zea mays L.) (var. B73) plants were grown under glasshouse conditions (16 h : 8 h, light : dark) at The Pennsylvania State University (USA). Tomato plants were grown in pro-mix potting soil and 4–5 leaf stage plants were used for all experiments. Maize plants were grown in Hagerstown loam and 4 to 5-wk-old plants in the mid-whorl (V7–V8) development stage were used for all the experiments. Ostrinia nubilalis (Hübner) (ECB) eggs were obtained from Benzon Research Inc. (Carlisle, PA, USA) and were reared on artificial diet in growth chamber with 16 h : 8 h, light : dark at 23°C as described previously (Peiffer & Felton, 2009).
Three newly molted fifth instar ECB larvae were introduced into the maize whorls or larvae were caged individually on tomato leaves using a small clip and allowed to feed for 24 h. Caterpillars with ablated or intact spinneret were introduced into the plants as already mentioned. Ablation/cauterization of ECB spinnerets was done as described previously (Peiffer & Felton, 2005). For bioassays, leaves around the ECB fed region were detached from the plants and placed in a 1 oz feeding cup containing 1% agar. Naïve ECB larvae were introduced into the feeding cups and larval weight was measured after 5 d. For the application of ECB saliva or phosphate buffered saline (PBS), plants were wounded (see below) and bioassays were performed in feeding cups as mentioned earlier. Uninfested or unwounded plants were used as the control.
Application of ECB saliva, oral secretions (OS), glucose oxidase (GOX) and glucose (Glc) to wounded plants
ECB saliva and OS were collected from fifth instar caterpillars as described previously (Tian et al., 2012). Youngest fully expanded tomato leaves were wounded (c. 3 mm diameter hole using a cut pipette tip) and 20 μl of PBS combined with saliva/fungal (Aspergillus niger) GOX (2 ng μl−1; Sigma-Aldrich)/PBS was applied to the wounded site. Similarly, maize plants were mechanically damaged using a wounding tool. Then, 20 μl of ECB saliva or GOX or d-Glucose (Glc; Sigma-Aldrich) was applied to the wounded maize plants near the whorl region. GOX was applied to the plants in an amount similar to that released by caterpillars (Peiffer & Felton, 2005). One millimolar Glc was applied to the wounded maize plants. For experiments with OS, 20 μl of ECB OS diluted with water (1 : 1) or 20 μl of water were applied to the wounded site. In all the above experiments, unwounded plants were used as the control.
Free glucose levels in tomato and maize were analyzed using a colorimetric glucose oxidase-peroxidase (GOPOD) assay. Briefly, 1 g of plant tissues were homogenized in liquid nitrogen and 3 ml of 0.1 M phosphate buffer (pH 7.0) was added to each sample. The samples were boiled for 5 min, centrifuged and 62.5 μl of the supernatant from each sample were mixed with 1 ml of the GOPOD reagent containing phosphate buffer (0.1 M, pH 7.0), GOX (20 mg ml−1), peroxidase (1 mg ml−1) and O-dianisidine (10 mg ml−1; Sigma-Aldrich). GOPOD reactions were incubated at 37°C for 30 min. Finally, 1 ml of 6 N HCl was added, and then absorbance at 540 nm was determined. Glucose standard curves were included to compare the absorbance from plant samples.
Scanning electron microscopy (SEM) images
Fifth instar ECB larval heads were removed and fixed overnight in 2.5% glutaraldehyde and 1.5% paraformaldehyde in 0.1 M phosphate buffer, pH 7.4. Larval heads were rinsed in phosphate buffer (pH 7) three times, for 15 min each. The samples were further fixed using 2% osmium tetroxide followed by three rinses in water. Larval heads were then dehydrated for 15 min each in graded ethanol series (25–100%) and dried using a Baltec CPD-030 (Techno Trade, Manchester, NH, USA) critical point dryer. Finally, the samples were coated with gold/palladium using a Baltec SCD-050 sputter coater (Bal-Tec AG, Balzers, Liechtenstein). Samples were imaged at 20 kV in a JSM 5400 scanning electron microscope (JEOL, Peabody, MA, USA), and images were captured using IMIX-PC v.10 software (Princeton Gamma Tech, Princeton, NJ, USA). All electron and confocal microscopy was done at the Penn State Microscopy and Cytometry Facility.
Confocal microscopy to detect fluorescence in ECB oral secretions
Detection of fluorescence in ECB OS was done as described previously (Peiffer & Felton, 2009). Briefly, fifth instar caterpillars were allowed to feed overnight on 0.5 g diet spiked with Alexa 488 (0.01 mg in 100 μl water; Invitrogen, Carlsbad, CA, USA). Diet treated with 100 μl water was used as the control. Larvae were then allowed to feed on fresh maize leaves until they made the first feeding damage (c. 25–45 min). Damaged leaf sections were mounted on glass slides and examined for fluorescence at an excitation of 488 nm with an Olympus FV1000 Laser Scanning Confocal Microscope (Olympus America Inc., Melville, NY, USA).
Native gel for glucose oxidase (GOX) activity
To determine GOX activity, 8% native PAGE was prepared. Saliva from 10 caterpillars of ECB and CEW combined with 20 μl of PBS were loaded. ECB OS was diluted by 50% and 20 μl sample were loaded. Proteins were stained for GOX activity as described previously (Eichenseer et al., 1999).
RNA extraction and quantitative real-time polymerase chain reaction (qRT-PCR)
RNA for quantitative real-time polymerase chain reaction (qRT-PCR) analysis was extracted by homogenizing leaves in liquid nitrogen using GenoGrinder 2000 and isolated using the TRIzol reagent (Invitrogen) as recommended by the manufacturer. One microgram of total RNA was used as a template for cDNA synthesis with High Capacity cDNA Reverse Transcription kit (Applied Biosystems, Foster City, CA, USA). Gene-specific qRT-PCR primers were designed using Primer Quest Software (Applied Biosystems; Supporting Information Table S1). Ubiquitin and actin genes were used as the reference genes for tomato and maize, respectively. All qRT-PCR reactions utilized FastStart Universal SYBR Green Master Mix (Roche Applied Science, Indianapolis, IN, USA) and were run on a 7500 Fast Real-Time PCR System (Applied Biosystems) as described previously (Peiffer et al., 2009).
For insect bioassays, glucose measurement and qRT-PCR transcript levels, analysis of variance (ANOVA) were performed using PROC GLM (SAS Institute, Cary, NC, USA). Data were checked for normality and log transformed where required. Means, when significant, were separated using the least significant difference (LSD) procedure (P <0.05).
ECB herbivory induces host plant defenses
It was previously shown that caterpillar feeding on host plants induced transcript levels of several genes involved in direct defenses (Shivaji et al., 2010; Tian et al., 2012). In this study, we examined whether the herbivory of ECB, a polyphagous insect, can induce defenses in two different host plants. To determine whether the genes related to herbivore defense responses were induced upon ECB feeding, a well established marker for insect resistance in host plants, proteinase inhibitor, Proteinase Inhibitor 2 (PIN2) and Maize Protease Inhibitor (MPI) transcript levels in tomato and maize respectively, were examined. The results suggests that ECB herbivory for 24 h significantly induced the expression of defense related proteinase inhibitors in tomato and maize when compared to the respective uninfested control plants (Fig. 1a,b). Furthermore, to examine whether these induced defenses can influence the insect performance on the host plants, insect bioassays were conducted. ECB-induced defenses on tomato and maize significantly retarded the larval growth when compared to the uninfested control plants (Fig. 1c,d), suggesting that the host plants activate induced defenses upon insect herbivory.
ECB oral secretions did not induce defenses in tomato and maize
We further examined whether ECB OS (regurgitant) can induce defenses in the host plants. ECB OS combined with water (1:1) were applied to the wounded sites of the plants. The application of ECB OS did not significantly affect the induction of PIN2 and MPI in tomato and maize, respectively (see Fig. S1a,b). To test whether ECB regurgitate during feeding we used a fluorescent dye to label the OS which allows us to visualize the OS at the site of ECB feeding. Our results clearly indicate that ECB rarely regurgitated while feeding (see Fig. S2). Only a single small fluorescence spot was detected in one of five maize leaves after being fed by ECB that had eaten the dye-spiked diet. Taken together, our data suggests that ECB OS is not the key elicitor in inducing defenses in maize. These results further support the previous reports that caterpillars rarely regurgitate during feeding on plants and caterpillar OS does not activate the induction of PIN2, as observed in tomato (Peiffer & Felton, 2009; Tian et al., 2012).
ECB saliva elicits defense responses in host plants
ECB releases saliva produced from the labial salivary glands through the spinneret (Fig. 2a,b). We can block the secretion of saliva from ECB by cauterizing or ablating the spinneret using a heated probe (Fig. 2c). The ablation of spinneret prevents ECB salivation, however, the caterpillars can feed and cause damage to the host plants. To test whether ECB saliva causes defense induction, we introduced ablated (no saliva) or unablated/intact (saliva secreted) ECBs on tomato (in cages) and maize (inside the whorls) plants, and allowed them to feed for 24 h. ECB feeding with intact spinneret significantly induced the expression of PIN2 and MPI as compared to the ECB with ablated spinnert (Fig. 3a,b). ECB uninfested plants were used as the control. In addition, we collected saliva directly from the ECB spinneret and applied it to the wounded sites of tomato and maize. Application of saliva significantly induced the expression of defense related PIN2 and MPI in tomato and maize (Fig. 3c,d), respectively, suggesting that saliva is indeed the critical component that is responsible for inducing host plant defenses.
Based on the earlier observations, we hypothesized that application of ECB saliva to the host plants can affect subsequent herbivory by inducing host plant defenses. To test this hypothesis, we applied either PBS or ECB saliva to the wounded tomato and maize plants. After 24 h, foliage near the treatment site was detached, and insect bioassays were performed using naïve ECB larvae. Caterpillar weight was significantly reduced when they were exposed to host plants that were treated with saliva compared to the plants that were treated with PBS (Fig. 3e,f). The results confirm that ECB saliva is the key component that elicits defenses in host plants.
Glucose oxidase (GOX) induce defenses in tomato, but not in maize
GOX has been identified as the most abundant protein in the saliva of CEW (Tian et al., 2012). Hence, we performed an in-gel assay to monitor the GOX activity in ECB secretions from the mouth region. Native PAGE of ECB oral secretions and saliva demonstrates that ECB saliva has GOX activity, albeit at lower levels compared to CEW saliva (Fig. 4a). We could not detect any GOX activity in ECB OS. Furthermore, application of ECB saliva and fungal GOX, which has similar substrate specificity as that of caterpillar saliva, significantly induced PIN2 expression in tomato as compared to wounded control plants treated with PBS (Fig. 4b). This data is in agreement with our previous results that caterpillar GOX can act as an insect effector in inducing defenses in tomato (Tian et al., 2012). By contrast, GOX treatment did not significantly induce MPI in maize as compared to the maize plants treated with ECB saliva. The relative expression of MPI in GOX treated maize plants was comparable to wounded maize plants treated with PBS (Fig. 4c). Because GOX oxidizes glucose (Glc) to gluconic acid and hydrogen peroxide (H2O2), we measured the levels of free Glc, the substrate for GOX, in tomato and maize. Interestingly, our data suggests tomato contains significantly higher levels of Glc compared to maize (Fig. S4a). Furthermore, exogenous application of Glc + GOX significantly induced MPI compared to application of Glc or GOX alone (Fig. S4b), indicating that the availability of Glc as the rate-limiting factor to elicit defense responses in maize. Nonetheless, the earlier results suggest that ECB saliva contains one or more host-specific salivary elicitor(s) that may be responsible for eliciting defense responses in tomato and maize.
Considerable progress has been made in identifying and characterizing effectors in bacteria, fungi and nematodes (Bent & Mackey, 2007; Panstruga & Dodds, 2009). However, very little information is available regarding insect effectors and how they interact with plant defense mechanisms. Here, using a naturally occurring host–herbivore system, we focused on identifying the effector(s) of ECB, a polyphagous insect that cause considerable economic loss to several crops (Ostlie et al., 1997; Hutchison et al., 2010). The results from our study clearly demonstrate that GOX present in ECB saliva induces defense in tomato, whereas, unidentified salivary effectors induce defenses in maize (Fig. 4). Our experiments with ECB saliva extend the work of Schmelz et al. (2009) that different plant species recognize insect elicitors by different mechanisms or cues.
Caterpillar OS have been well understood as an herbivore elicitor in inducing host defenses (Turlings et al., 1990, 1993; Alborn et al., 1997, 2007; Schmelz et al., 2007). However, our data suggest that, at least in the case of ECB, OS is not the critical component that elicits direct defenses in host plants. Quite to the contrary, it has been suggested that application of insect (Spodoptera littoralis) OS significantly induced the expression of MPI in maize (Ton et al., 2007). Unlike their observation, MPI expression was not induced in maize after ECB OS treatment. The disparity in these results may be due to several reasons, including use of different insect species, genotype differences (B73 vs Delprim), amount of OS applied and use of intact plants vs detached shoots. Furthermore, as opposed to Ton et al. (2007), the application of insect OS to young maize plants at V1 developmental stage did not significantly activate the induction of MPI as compared to the wounded maize plants treated with water (S. Ray & D. S. Luthe, unpublished data), ruling out the possibility of having any age affect of the plant in inducing defenses. It is plausible that genotype differences can be attributed for this discrepancy, since it was previously shown that American maize varieties are distinct from European maize varieties (Köllner et al., 2008). In addition, our results indicate that ECB rarely regurgitate while feeding (Fig. S2), thereby confirming that OS are not the exclusive components that elicit defenses in host plants. Taken together, these observations suggest that ECB OS do not have an important role in inducing direct defenses in their host plants.
Our experiments with ECB saliva strongly induced direct defenses in both tomato and maize (Fig. 3). Several studies have shown that feeding by chewing insects on host plants activate the jasmonic acid (JA) biosynthetic pathway genes, in addition to the genes involved in providing direct defenses (Li et al., 2003; Shivaji et al., 2010). Our data indicates that ECB saliva can also induce defense-related JA transcripts in both tomato and maize. Both Lipoxygenase (LOX) and 12-oxo-phytodienoic acid (OPR) genes, two key genes involved in JA biosynthesis/signaling, were significantly induced after ECB feeding with intact spinneret (Fig. S3). Interestingly, this is in contrast to our previous observation demonstrating that CEW saliva did not induce the early signaling genes associated with the JA pathway, even though it elicited a burst of JA in tomato (Tian et al., 2012). One possible explanation is that ECB saliva is distinct and contains multiple proteins. Consequently, the unidentified ECB effector(s) by itself or in concert with GOX can orchestrate defense induction, including the jasmonate pathway in both tomato and maize. Although GOX alone did not induce defenses in maize, ECB saliva contains elicitors that can activate direct defenses (Fig. 4c), further confirming our hypothesis that ECB saliva contains multiple effectors.
It has been suggested that H2O2 produced by insect attack and/or salivary GOX can function as a ‘signaling molecule’ that elicits defense responses in plants (Zong & Wang, 2004; Bonaventure, 2012). However, the reduced availability of the ECB GOX substrate (glucose) in maize leaves (Fig. S4a) could limit its ability to produce sufficient H2O2 to activate the downstream defense-related proteins, including MPI. Alternatively, the low level of GOX in ECB saliva or the lack of a GOX receptor or recognition mechanism in maize foliage could limit H2O2 production and the resultant downstream signaling components. In summary, the current study clearly demonstrates that ECB saliva is indeed the critical component that is responsible for inducing direct plant defenses in both tomato and maize (Figs 1, 3). In addition, our data suggest that, as described by Schmelz et al. (2009), different plant species recognize the same insect or insect salivary factor(s) by different mechanisms or cues (Fig. 4). Further experiments are required to tease apart how different plants perceive the same insect in a different manner. The proteomic analyses of ECB saliva will advance our understanding of additional effectors that are responsible for modulating plant defense pathways. Identification of novel insect elicitors will ultimately help us to engineer crops that can suppress the expression of effectors, thereby providing novel approaches to control insect pests of agricultural crops.
This work was supported by grants from USDA NIFA (2010-65105-20639 and 2011-67013-30352) awarded to G.W.F. and D.S.L.