Herbivore-induced callose deposition on the sieve plates of rice: an important mechanism for host resistance

The brown planthopper ( Nilaparvata lugens Stål, BPH) is a specialist herbivore on rice ( Oryza sativa L.) that ingests phloem sap from the plant through its stylet mouthparts. Electronic Penetration Graphs (EPGs) revealed that BPH insects spent more time wandering over plants carrying the resistance genes Bph14 and Bph15 , but less time ingesting phloem than they did on susceptible plants. They also showed that their feeding was frequently interrupted. Tests with [ 14 C]sucrose showed that insects ingested much less phloem sap from the resistant than the susceptible plants. BPH feeding up-regulated callose synthase genes and induced callose deposition in the sieve tubes at the point where the stylet was inserted. The compact callose remained intact in the resistant plants, but genes encoding β -1,3-glucanases were activated, causing unplugging of the sieve tube occlusions, in susceptible plants. Continuing ingestion led to a remarkable reduction in the susceptible plants’ sucrose contents and activation of the Ramy 3D gene, leading to starch hydrolysis and ultimately carbohydrate deprivation in the plants. Our results demonstrate that BPH feeding induces the deposition of callose on sieve plates in rice, and that this is an important defense mechanism that prevents the insects ingesting phloem sap. In response, however, the BPH can unplug sieve tube occlusions by activating β -1, 3-glucanase genes in rice plants.


INTRODUCTION
the leaf sheath of rice (Oryza sativa L.) plants, ingesting nutrients specifically from the rice phloem using its piercing mouthparts (stylet), forming a stylet sheath during the feeding process. Feeding by numerous BPHs on a single plant generally results in the susceptible plants yellowing, browning and drying. In the last decade, the BPH has frequently caused widespread destruction of rice crops and heavy losses of yields (Shi et al., 2003;Park et al., 2007).
The main methods used to control BPH pests are to apply chemical insecticides and/or develop and grow resistant varieties in an integrated pest management strategy.. However, the cost of chemical control is often very high, and the chemicals can destroy the natural balance of BPH-predators that help to keep the BPH population in check.
The misuse of chemical pesticides may also cause a resurgence of the insect. Therefore, the most economic and efficient method for controlling the BPH is to exploit the host resistance to attack (Renganayaki et al., 2002). To date, 19 BPH-resistance genes in rice have been reported, and several have been used in rice breeding programs (Yang et al., 2004;Jena et al., 2006;Zhang, 2007). Various molecular techniques, including suppression subtractive hybridization, northern blotting and cDNA array analysis, have been used to study rice responses to BPH feeding (Zhang et al., 2004;Wang et al., 2005;Yuan et al., 2005;Park et al., 2007, Wang et al., 2008. BPH feeding is thought to result in a re-organization of the gene expression profile of rice and most of the strongly-regulated genes are associated with metabolism, cell defense, cellular transport, cellular communication or signal transduction and the biogenesis of cellular components. In contrast, the expression of genes related to the flavonoid pathway, aromatic metabolism and the octadecanoid pathway are mostly unchanged or down-regulated. This indicates that BPH feeding induces plant responses associated with a JA-independent pathway and crosstalk with responses related to abiotic stress, pathogen invasion and phytohormone signaling pathways (Zhang et al., 2004;Wang et al., 2005;Yuan et al., 2005, Wang et al., 2008. Many studies have reported the effects of BPH feeding on physiological properties and metabolic changes in rice plants (Cagampang et al., 1974;Qiu et al., 2004). However, the mechanism of rice resistance to BPH attack still remains largely unknown.
The aim of this study was to further explore the interactions between the BPH insects and rice plants, in an attempt to elucidate the mechanisms involved in rice resistance to the BPH. We used a susceptible rice plant variety (TN1) as a control, and first studied the feeding behavior of the BPH on rice plants carrying the BPH-resistance genes Bph14 and Bph15 using the Electronic Penetration Graph (EPG) technique. We then examined the anatomical features of the punctured phloem cells, especially the induced callose, by observing and counting the number of sieve plates with callose deposition. Real-time PCR was performed to examine the expression of genes coding for callose synthases and degrading enzymes. To our knowledge, this is the first targeted callose analysis of rice resistance to BPH feeding. Our results suggest that the induced callose sealing in sieve tubes plays an important role in the inhibition of BPH feeding. However, the BPH can unplug the sieve tube occlusions by activating β -1, 3glucanase genes in rice plants.

EPG Monitoring the Feeding Behavior of BPH
Brown planthoppers ingest phloem sap from rice plants through their narrow piercing-sucking mouthparts, which are called stylets. During the feeding process, the stylet transiently punctures the epidermis, making the first probing, then penetrates the plant cell walls, the insect subsequently salivates into the cells, and ingests the phloem sap. In this study, the feeding activities of the BPH on different varieties of rice, with varying levels of resistance, were monitored electrically using a real-time EPG technique (Tjallingii, 2006). Five types of waveform were identified, representing different types of insect feeding behavior: non probing (Type 1), pathway (Type 2), phloem puncture (Type 3), xylem ingestion (Type 4), and phloem sap ingestion (Type 5) ( Fig. 1  Waveform Type 2 occurred when the BPH insects used their stylets to search for the target cells in plant tissues, in a series of activities including penetrating plant cells, salivating, tasting, and forming branches of the stylet sheath (Supplemental Fig. S1A).
The total duration of this waveform type on the resistant B5 plant was 114.6 min; significantly longer than that on the susceptible TN1 plant (33.3 min). This suggests that the insects spend more time searching for suitable target feeding cells in the resistant plant tissue (Table I A). The frequency of this behavior over the 8-hour recording period showed a general tendency to increase with higher levels of plant resistance (Table I B).
Waveform Type 3 occurred when the stylet penetrated the vascular bundle of the rice plant (Supplemental Fig. S1B). Overall, the total duration of Waveform Type 3 was correlated with the plants' level of resistance, but there was a significant difference between the two moderately resistant varieties (RI35 and YHY15) and the difference between TN1 and YHY15 was insignificant (Table I A). Such differences in Type 1 behavior of BPHs on these moderately resistant varieties might be attributable to the differences in resistance genes of the plants.
Waveform Type 4 represented xylem ingestion by the BPHs (Supplemental Fig.   S1C). We assumed that the xylem was not the resistant element within the plant because of the irregular duration of this behavior on the B5 (75.5 min), YHY15 (125.9 min), RI35 (123.9 min) and TN1 (58.2 min) varieties. There was no clear relationship between resistance level and total duration of this type of behavior (Table I A).
Waveform Type 5, representing phloem ingestion, gave a better indication of resistance, since it reflected the relative quantity of phloem sap ingested by the BPHs (Supplemental Fig. S1D). During the 8-hour recording period, the total duration of Type 5 behavior on the resistant variety B5 was 33 min, approximately one-tenth of that on the susceptible variety TN1 (340.2 min) ( Table I A). Moreover, the mean duration of each period of phloem sap ingestion was much shorter on the resistant varieties than on the susceptible control TN1 (Table I C

Use of [ 14 C]sucrose to Quantify Phloem Sap Ingestion by BPH
Sucrose is the main carbohydrate that is transported long distances through the phloem and ingested by the BPH. By culturing rice plants in a [ 14 C]sucrose solution, 14 C can easily be introduced into the phloem. The quantity of phloem sap ingested by the BPHs can then be estimated by monitoring the radioactivity of 14 C in the insects.
We used the ratio of radioactivity ( 14 C) in the insect to that in the plant, designated the I/P index, as an indicator of the relative quantity of phloem sap ingested by BPHs. The results showed that the I/P ratio for BPHs that had fed on resistant B5 plants was very low (0.02, compared with 0.85 for insects feeding on susceptible TN1 plants) over a 20-hour period (Fig. 2). The I/P ratios for BPHs that fed on the moderately resistant varieties YHY15 and RI35 were 0.54 and 0.28, respectively. These results strongly indicate that the BPH insects ingested less phloem sap from the resistant rice plants than they did from the susceptible control (TN1).

Callose Deposition on the Sieve Plates of BPH-infested Plants
To investigate the mechanisms that prevent BPHs from continuously ingesting phloem sap from resistant rice plants, the leaf sheaths of BPH-infested and BPH-free resistant and susceptible plants were sectioned and examined histopathologically. The sections were stained with 0.1% aniline blue and examined under a fluorescence microscope. In the BPH-free untreated rice plants, there was little or no callose deposition on the sieve plates in the leaf sheaths (Fig. 3 E and F). When the plants were infested with the BPH, more callose was deposited on the sieve plates of the target sieve tubes, where the stylets had been inserted -the sieve plates were obviously thickened and emitted strong fluorescence ( Fig. 3 A to D). Counts of the bright callose plugs revealed that callose deposition increased during the first three days of infestation in both B5 and TN1 plants, but there were more callosic sieve plates in the former (13.7 callosic sieve plates in 50 sections) than the latter (5.8). Moreover, with prolonged BPH feeding, the callose deposition decreased quickly in TN1 plants, to only 2.4 callosic sieve plates in 50 sections after four days, but a high level of callose deposition remained in the B5 plants (12.7) (Fig. 3 G

Expression of Callose Synthase and β -1, 3-Glucanase in Rice Plants
Callose deposition is a dynamic process coordinated through the activities of callose synthase and the callose-hydrolyzing enzyme β -1, 3-glucanase. To investigate the mechanisms responsible for the differential callose deposition in the resistant and susceptible rice plants, the expression of ten callose synthase-encoding genes were investigated using semi-quantitative RT-PCR. We detected transcripts of four of these genes, namely OsGSL1, OsGSL3, OsGSL5 and OsGSL7 (Supplemental Fig. S2). Three of the detected genes, OsGSL1, OsGSL3, and OSGS5, were further analyzed by real-time PCR. These genes were clearly up-regulated after the B5 and TN1 plants were treated with BPH for 6 h, reaching high levels after 12 h (2 to 4 fold of uninfested plants). The expression levels remained high in the following 72 to 96 h, generally over three fold higher than those in uninfested plants. (Fig. 3H to J). These observations suggest that the callose synthase genes were up-regulated and consequently callose synthesis was enhanced in both the resistant and susceptible plants attacked by the BPH.
The expression patterns of six β -1, 3-glucanase genes were also investigated. The patterns of four of them were found to differ between BPH-infested B5 and TN1 plants. Gns5 showed an increase both in TN1 and B5 plants, but the increase in TN1 was much higher than in B5 plants. The greatest increase in TN1 was more 9 fold relative to the uninfested control ( Fig. 4I). Gns4 was constitutively expressed in B5, but was induced in TN1 plants. Gns6 shared similar expression patterns with Gns5.
Little or no expression of either Gns2 or Gns3 was detected in the leaf sheath of the rice plants (Supplemental Fig. S2). The expression of β -1, 3-glucanase genes, such as Osg1 and Gns5, was clearly up-regulated in the susceptible rice plants, in contrast, the expression level of these genes was up-regulated much less (Gns5) or even absent at detectable level (Osg1) in resistant ones.

Decomposition of Starch in Compensation for Sugar Losses in Susceptible plants
To investigate the anatomical effects of BPH feeding on rice plants, leaf sheaths of plants representing the most resistant and most susceptible varieties (B5 and TN1, respectively) were sectioned, stained in 3% KI-1% I 2 solution, and examined under a microscope. We found abundant starch granules in the leaf sheaths of the uninfested plants, and there appeared to be more in the B5 than in the TN1 plants (Fig. 5A, upper two sections). In the BPH-infested plants, starch granules were rapidly consumed after one day in the susceptible TN1 variety, and most were exhausted after three days of infestation ( Fig. 5A, right sections). In contrast, the starch granules disappeared much more slowly in the resistant variety (B5; Fig. 5A, left sections).
The starch content of leaf sheaths was also determined. Results showed that the starch content decreased much more quickly in TN1 than in B5, under the stress of BPH infestation. However, it should be noted that there was much more starch in B5 (22.7 mg g -1 FW) than in TN1 (16.7 mg g -1 FW) plants that were not infested by BPHs ( Fig.   5B, lower panels). The sucrose content varied in a similar way to that of the starch content ( Fig. 5B, upper panels). In the BPH-infested TN1 plants, sucrose contents fell to ca. 56% and 30% of the untreated control plant levels after one day and four days, respectively ( Fig. 5B), showing that susceptible TN1 plants can be rapidly and seriously deprived of carbohydrates during infestation by these insects.
The expression of nine α -amylase and three β -amylase genes in the leaf sheaths of rice plants was investigated using semi-quantitative RT-PCR. We found that RAmy3D (Os08g0473900), one of the genes examined, was strongly induced in the BPH-infested TN1 plants, but was only weakly expressed in the B5 plants (Supplemental Fig. S2).
This indicates that RAmy3D plays a role in the response of susceptible rice plants to BPH feeding by breaking down the starch.

DISCUSSION
Enhancing host resistance is an important component of integrated pest management. However, the mechanism of rice's resistance to BPH is still uncertain. In the past, researchers considered that it might be governed by the presence of chemicals confined to the phloem (Sogawa and Pathak, 1970). Bph1 was the first BPH resistance gene identified and is associated with the production of: flavonoids, including salicylic acid; amino acids such as aspartic acid, glutamic acid, alanine, serine, leucine, asparagine and valine; and organic acids such as succinic acid, malic acid, oxalic acid and transaconitic acid (Sogawa and Pathak, 1970;Sogawa, 1976 (Zhang et al., 2004;Yuan et al., 2005;Park et al., 2007;Wang et al., 2008). Such reorganization has been well documented in other plant-herbivore systems, supporting the hypothesis that inducible defenses contribute to rice's resistance to the BPH. The data gathered in this experiment show that BPH feeding affects the expression of genes associated with the synthesis and hydrolysis of callose, and starch decomposition in rice plants, resulting in the deposition of callose plugs on the sieve plates. This prevents BPH insects from continuously ingesting phloem sap and allows normal carbohydrate levels to be maintained in resistant plants, as described below.
Phloem, the target of BPH feeding, mainly consists of sieve tubes and companion cells. The functional units of sieve tubes are series of sieve elements that have porous sieve plates at their abutting ends, allowing the phloem sap to flow continuously (Will et al., 2007). The sieve element/companion cell modules are highly sensitive to biotic and abiotic disturbance, and elaborate sealing mechanisms, such as protein plugging and callose formation, have evolved (McNairn and Currier, 1967;Will and Bel, 2006).
Callose has several functions in the normal development of plants (Chen et al., 2007), and its formation and deposition can be induced by either biotic or abiotic stress (Jacobs et al., 2003;Ueki and Citovsky, 2005). We found more callose deposits on sieve plates in both the resistant and susceptible rice plants infested by BPH than in the uninfested controls (Fig. 3). However, in the resistant B5 plants almost all the target sieve tubes showed strong fluorescence, indicating that abundant, compact callose had been deposited within them. In contrast, the callose signals were fainter in the susceptible TN1 plants, and there were no compact callose deposits in many of their sieve tubes where BPH insects had fed.
It has been reported that callose synthesis is Ca 2+ dependent (King and Zeevaart, 1974), and phloem-feeding insects seem to induce the release of Ca 2+ stored in the reticulum or the apoplasm, thereby activating callose synthesis (Arsanto, 1986;Volk and Franceschi, 2000). It has also been demonstrated that large amounts of callose on sieve plates reduce the rate of phloem translocation, and can even block it completely (McNairn and Currier, 1967 (Table I, waveform type 5). The 14 C labeling data also support the conclusions that phloem feeding was inhibited on B5 plants, and that the BPHs sucked only small amounts of sap from them (Fig. 2). Therefore, we conclude that callose deposition plays an important role in preventing BPHs from ingesting phloem sap, and thus contributes to the resistance of rice to these insects. This is the first time that callose has been implicated in the interaction between rice plants and the BPH. However, sieve tube occlusion, in response to biotic and abiotic stress, has been previously reported (Will and Bel, 2006). Cagampang et al. (1974) showed that the rate of upward sap transport in rice plants infested by BPHs was only 60% of the rate in their uninfested counterparts, and Nielson et al. (1990) found evidence that the acropetal movement of photosynthates in alfalfa was seriously disrupted by potato leafhopper feeding. In cotton, basipetal phloem translocation was completely inhibited by callose deposition in a 14 CO 2 labeling trial (McNairn and Currier, 1967). Furthermore, forisomes (spindle-like protein bodies in the sieve tubes) have been shown to inhibit aphid feeding in the broad bean, Vicia faba (Will et al., 2007).
Since callose synthase genes were up-regulated and callose deposition occurred in both the resistant and susceptible rice plants a short time after BPH feeding commenced ( Fig. 3), the insects had to overcome the physical barriers imposed by callose in order to obtain sufficient food even from susceptible varieties. In the aphid/broad bean system, aphid saliva can prevent sieve tube plugging by forisomes (Will et al., 2007), which.
have only been found in the Fabaceae. Our results, in contrast, indicate that BPH feeding induced the expression of genes encoding β -1, 3-glucanases, causing decomposition of the callose barriers in the susceptible rice plants (Fig. 4A-H). These enzymes play important roles in plant defense and development.
Beta-glucanase-encoding genes have been classified into four subfamilies, based on their structure and function. Two tandem gene clusters, Gns2-Gns3-Gns4 and  Fig. S3).
Starch is a major end-product of photosynthesis; it is produced in chloroplasts and is the main energy storage substance in cereal grains and leaf sheaths. In the chloroplast and amyloplast, starch metabolism is closely related to other metabolic processes in the cytosol, such as sucrose metabolism, glycolysis, and glyconeogenesis, In such cases, the plant only needs to provide carbohydrate for itself and, consequently, less sucrose loss and less starch decomposition will occur. Therefore, the resistant plants can survive a long period of stress from BPH, even when the phloem transportation is affected to some extent; eventually the BPH will die of starvation.
Competition for sugar, as well as other nutrients, plays an important role in the interaction between the herbivore and the plant. However, mechanisms allowing plants to resist (or tolerate) herbivore attack may differ widely between phloem-sucking insects and chewing insects. For example, in response to foliar herbivore, the allocation of sugars to roots increased in the annual Nicotiana attenuate, so that plants better tolerate herbivore (Schwachtje et al 2006).
Our understanding of the resistance mechanism can be encapsulated in the following model of interactions between the BPH and the rice plant (Fig. 6). First, the BPH acts on the plant by penetrating its tissues, ejecting saliva into its cells and sucking up phloem sap. In response to BPH feeding, the plant up-regulates expression of its callose synthase and β -1, 3-glucanase genes. Consequently, callose deposition occludes the sieve tubes, and prevents the BPH from ingesting the phloem sap.
However, β -1, 3-glucanases that decompose the deposited callose and thereby facilitate the BPH's continued feeding from the phloem are strongly induced in susceptible plants, but much more weakly induced in resistant plants. Thus, differential expression of β -1, 3-glucanases can account for between-plant differences in resistance levels.

CONCLUSION
We have demonstrated that feeding by the BPH can induce callose synthesis and deposition on the sieve plates of rice plants. Callose deposition affects phloem transportation and plays an important role in preventing the BPH from ingesting the phloem sap. Our results show not only that callose deposition is sufficient for resistant plants to defend themselves against the BPH, but also that some specific β -1, 3-glucanases are: active callose-decomposing enzymes, induced by BPH activity and responsible for the susceptibility of TN1 plants. The differential expression of these enzymes may result in different resistance levels in rice plants. Five rice (Oryza sativa L.) varieties were used in this study. B5 is an line carrying BPH resistance genes Bph14 and Bph15 from wild rice (O. officinalis) and exhibits high resistance to BPH with the severity score below 3.0 in the seedling bulk test (Huang et al., 2001). RI35 (carrying Bph14) and YHY15 (carrying Bph15) are progeny lines of B5, and are moderately resistant to BPH with the severity score about 5. TN1

Plants and Insects
and Hejiang 19 are conventional varieties susceptible to BPH and their severity scores is 9.0. All experiments were carried out on rice plants at the 3-to 5-leaf stage.
Unless otherwise stated, the brown planthopper (Nilaparvata lugens Stål, BPH) insects were 3-to 4-instar nymphae, and the insects were maintained on TN1 plants in the Genetics Institute, Wuhan University.

EPG Waveform Characterization and Quantification
To link the EPG waveforms with the feeding behavior patterns of the BPH, a microscope was coupled to EPG equipment, as follows. Special plastic slides, each with a 1-cm diameter hole in the center, were prepared and covered with stretched Parafilm.
Sucrose solution or tap water (each with a small amount of active carbon powder to trace the water flow) was dropped onto the Parafilm, to serve as an artificial food source, and mounted under a cover slip. BPH insects with a gold wire (length 3 to 5 cm, diameter 20 µ m) attached to the dorsum by conductive silver glue were then allowed to probe the food through the Parafilm. The gold wire from each insect and a copper wire (diameter 0.1 mm) immersed in the food were linked to a Giga 4 model DC-EPG amplifier (Wageningen University, the Netherlands). The EPG setup was housed in a climate-controlled room (25 ± 2°C) and shielded from electrical noise by an earthed Faraday cage. The EPG was also linked to a computer running PROBE 3.1 software (attached to the EPG equipment). The electronic signals from the different channels were converted into digital data using a DI-710 data logger (DATAQ) and transformed into waveforms displayed on the computer screen in real-time. By ralating the feeding behavior of BPH insects under the microscope with the real-time EPG waveforms displayed on the screen, we were able to categorize the waveform types (Supplemental Fig. S1).
For EPG recordings of BPH insects feeding on rice plants, adult brachypterous females (two days after the final molt) were collected at 9:00 am, and attached to a gold wire, as described above. After being starved (but provided with water) for 1h, each insect was placed on the leaf sheath of the plant to be tested and the gold wire from its dorsum was connected to the EPG. Before acquiring and processing data, WINDAQ Waveform Browser software (DATAQ) was run for 30 minutes to pre-test the activity of the insect. Data were acquired at 100 Hz sample frequency, stored on the computer's hard disk, and simultaneously displayed on a screen. The data were analyzed using ANA3.0 software (Wageningen University, the Netherlands). EPG recordings were carried out for 8 hours/insect/plant, with at least seven replicates for each variety, using fresh seedlings and insects in each case. In a single experiment, all four genotypes were examined simultaneously, one channel for each genotype, then the experiment was repeated. Data were compared using Kruskal-Wallis one way analysis of variance ranking, and Scheffe's post-hoc pairwise comparisons (P<0.05).

Isotope 14 C-labeling and Determination
For 14 C-labeling, rice seedlings at the 3-leaf stage were transferred to a vial containing a 1 mL solution of [ 14 C]sucrose from Sigma and non-labeled sucrose (4 µ Ci and 15 mg per vial, respectively). Each seedling was held in place with a sponge and pushed into the neck of the vial so that the root was immersed in the solution. Each plant, together with ten insects, was placed in a test tube (30 mm×200 mm) and covered with gauze to prevent the insects from escaping. Seedlings were allowed to take up the sucrose solution at 25±2 °C in darkness. Twenty hours later, the seedlings were removed from the tubes, their roots were discarded and their remaining parts were cut into 1 cm-long segments. Insects and plant segments were then plunged into 5 mL of 80% ethanol:water solution (v/v), boiled for 10 minutes and then centrifuged at 4000 g for 5 min at 4°C. The supernatants containing the extracted soluble fractions were collected and concentrated to 500 µ L, 100 µ L of which was used to determine their 14 C content, using a Beckman LS6500 liquid scintillation spectrometer (Beckman). The I/P index (ratio of 14 C radioactivity in the insects and plants) was used to evaluate the distribution of 14 C between the insects and plants, reflecting the proportion of soluble 14 C ingested by the insects from the phloem.

Histochemistry and Microscopy
Rice plants were each infested with ten BPHs. Leaf sheaths were collected, fixed in FAE (formaldehyde:acetic acid:70% ethanol=5:5:90, v/v/v), dehydrated, embedded in paraffin, and cut into 10-µm thick sections using a microtome. The sections were mounted on microscope slides, dewaxed, and rehydrated for staining at room

temperature.
To highlight starch and saliva sheaths, the rehydrated sections were stained in 3% (w/v) KI-1% I 2 solution for 1 min, then examined under a light microscope. For callose observations, 10-µm thick sections were mounted on glass slides (50 sections per slide).
Callose staining was performed as described by Dietrich et al. (1994) with some modifications. Rehydrated sections were stained with 0.1% (w/v) aniline blue in 0.15 M K 2 PHO 4 for 5 min, and examined under a UV epifluorescence microscope (Olympus BX51, Japan). Callose on individual sieve plates was classified as either 'faint' or 'bright': 'faint' types included clearly visible plates with a thin, green-yellow appearance, while 'bright' was used to describe all thickly callosed sieve plates with bright blue fluorescence (McNairn and Currier, 1967). The amount of callose deposition in each plant examined was evaluated by counting the number of sieve plates that had 'bright' callose. At least 400 sections were examined for each treatment.
Photographs were taken with a Coolpix 995 Digital Camera (Nikon).

Determination of Sucrose and Starch Contents
Fresh leaf sheaths (approximately 1 g) were powdered in liquid nitrogen, homogenized in 4 mL of 80% (v/v) ethanol, heated in a water bath at 80°C for 40 min, and centrifuged at 4000 g for 10 min. The supernatant fraction was collected, and the solid fraction was washed with 80% ethanol and centrifuged; this procedure was repeated twice. The supernatants collected from each sample were combined, and then active carbon was added and filtered for sucrose analysis, while the pellet was dried for starch determination.
Sucrose was measured using the anthrone-sulfuric acid method (Trevelyan and Harrison, 1952) with modifications. For colorimetric determination, 0.5 mL of the 80% ethanol extract was added to 1 mL of water and digested with 1 mL 10% (w/v) aqueous KOH in a water bath at 100 °C for 3 min. The cooled reaction mixture was placed immediately in an ice-water bath, then 4 mL of anthrone reagent (1 g anthrone dissolved in 500 mL 98% H 2 SO 4 ) was added to the cooled mixture. The mixture was again incubated in a 100°C water bath for 1 min, then placed for 90 sec in an ice-water bath. The mixture was transferred by pipette into a 1 cm diameter spectrophotometer cup, and its absorbance at 625 nm was measured using a UV-1601 spectrophotometer (Shimadzu).
The dried pellet was added to 5 mL of 80% Ca (NO 3 ) 2 (w/v), placed in a 100 °C water bath for 10 min, and then centrifuged at 4000 g for 4 min. The supernatant fraction was collected, and the solid fraction was washed with 80% Ca (NO 3 ) 2 and centrifuged; this procedure was repeated twice. All the supernatants from each sample collected were combined and added to 20 mL of water to prepare the starch solution.
One mL of starch solution was mixed with 2 mL of 80% Ca (NO 3 ) 2 and 100 µ l solution of 0.01 N I 2 -KI (1.3 g I 2 and 4.0 g KI in water, final volume 1 L). The mixture was transferred by pipette into a 1 cm diameter spectrophotometer cup, and its absorbance at 620 nm was measured using a UV-1601 spectrophotometer.

Semi-quantitative RT-PCR
Total RNA was extracted from the leaf sheaths (approximately 200 mg fresh weight) with Trizol Reagent (Invitrogen), and the remaining DNA was degraded using a Turbo DNA-free kit (Ambion). cDNA was synthesized from the total RNA (4 µ g) using a Thermoscript RT-PCR system (Invitrogen), with oligo(dT)20 primers, following the manufacturer's instructions. RT-PCR was performed using Taq DNA polymerase (MBI Fermentas) in 10 µ L reaction mixtures with the gene-specific primers listed in Supplemental Table S1, which were either directly synthesized according to previously published information (Tomoya et al., 2002;Fukao et al., 2006;Tomoya et al., 2006), or designed using the PCR primer design tool primer3 (http://frodo.wi.mit.edu) according to cDNA sequences obtained from NCBI GeneBank (http://www.ncbi.nlm.nih.gov). Actin1 control primers were used as standards for mRNA expression, all the templates for RT-PCR of different genes are from the same individual cDNA samples. The amplification program consisted of an initial denaturation step at 94°C for 5 min, followed by 28-40 cycles of 94°C for 30 sec, 58°C for 30 sec, 72°C for 1 min, and a final extension at 72°C for 5 min. This procedure was repeated at least three times.

Real-Time PCR Analysis
Genes for real-time PCR analysis were screened based on the results of semi-quantitative RT-PCR, the genes which showed obvious variation were chose for real-time PCR, whereas, the genes that could not be detected or showed no obvious variation were not chose for further study. The primers (Table II)

SUPPLEMENTAL MATERIAL
Supplemental Figure S1. Details of the EPG waveforms shown in Fig. 1.
Supplemental Figure S3. Results of RT-PCR analysis of Osg1.
Supplemental Table S1. Sequences of primers for semi-quantitative RT-PCR.     acts on a plant by penetrating its tissues, ejecting saliva into its cells and sucking up phloem sap. In response to feeding by the BPH, the plant up-regulates genes encoding callose synthases and β -1, 3-glucanases. Consequently, callose deposition occludes the sieve tubes, and prevents the BPH from ingesting the phloem sap. Then, specific β -1, 3-glucanases decompose the deposited callose in susceptible plants (but little in resistant plants), allowing the BPH to resume feeding from the phloem. Thus, differential expression of β -1, 3-glucanases accounts for the differences in their resistance levels. Arrows indicate promotion or positive modulation of the process;

ACKNOWLEDGMENTS
www.plantphysiol.org on August 16, 2017 -Published by Downloaded from Copyright © 2008 American Society of Plant Biologists. All rights reserved.