Induction of systemic stress tolerance by brassinosteroid in Cucumis sativus

Authors


Author for correspondence:
Jing Quan Yu
Tel: +86 57186971120
Email: jqyu@zju.edu.cn

Summary

  • Brassinosteroids (BRs) are a new class of plant hormones that are essential for plant growth and development. Here, the involvement of BRs in plant systemic tolerance to biotic and abiotic stresses was studied.
  • The effects of 24-epibrassinolide (EBR) on plant stress tolerance were studied through the assessment of symptoms of photooxidative stress by chlorophyll fluorescence imaging pulse amplitude modulation, the analysis of gene expression using quantitative real-time PCR and the measurement of hydrogen peroxide (H2O2) production using a spectrophotometric assay or confocal laser scanning microscopy.
  • Treatment of primary leaves with EBR induced systemic tolerance to photooxidative stress in untreated upper and lower leaves. This was accompanied by the systemic accumulation of H2O2 and the systemic induction of genes associated with stress responses. Foliar treatment of EBR also enhanced root resistance to Fusarium wilt pathogen. Pharmacological study showed that EBR-induced systemic tolerance was dependent on local and systemic H2O2 accumulation. The expression of BR biosynthetic genes was repressed in EBR-treated leaves, but elevated significantly in untreated systemic leaves. Further analysis indicated that EBR-induced systemic induction of BR biosynthetic genes was mediated by systemically elevated H2O2.
  • These results strongly argue that local EBR treatment can activate the continuous production of H2O2, and the autopropagative nature of the reactive oxygen species signal, in turn, mediates EBR-induced systemic tolerance.

Introduction

As a result of their sessile life cycle, plants are constantly exposed to a variety of biotic (i.e. pathogen attack and insect herbivory) and abiotic (i.e. high or low temperature, drought and salinity) stresses. To survive these stresses, plants have evolved intricate signaling networks to adapt their cellular activities to the changing environment. At the molecular level, the perception of external stimuli and the subsequent activation of defense responses require a complex interplay of signaling cascades (Fujita et al., 2006). A large body of evidence has demonstrated that hormones, such as salicylic acid (SA), jasmonic acid (JA), ethylene (ET) and abscisic acid (ABA), are primary signals that regulate the protective responses to biotic and abiotic stresses (Lorenzo & Solano, 2005; Mauch-Mani & Mauch, 2005). The generation of reactive oxygen species (ROS) has been shown to be a key process that is shared among many biotic and abiotic stress responses (Fujita et al., 2006).

Plants can mount a systemic response to establish long-lasting protection against a broad spectrum of pathogens in tissues distant from the primary site of attack (Durrant & Dong, 2004). This mechanism is termed ‘systemic acquired resistance’ (SAR). SA is required for SAR, but is not itself the mobile signal (Vernooij et al., 1994). NPR1 is a central positive regulator of SAR. Changes in cellular redox states induced by SA accumulation trigger the reduction of NPR1 from oligomers to active monomers that translocate to the nucleus and regulate the expression of pathogenesis-related (PR) genes, such as PR-1 (Mou et al., 2003). However, despite extensive research, the identity of the mobile signal in SAR has not been firmly established. Maldonado et al. (2002) have shown that DIR1, which has sequence similarity to lipid transfer protein, is required for the development of SAR in systemic leaves, suggesting that a lipid-based molecule may be the mobile signal for SAR. More recently, Truman et al. (2007) have reported the rapid accumulation of JA in phloem exudates of leaves challenged with pathogens and that SAR can be induced by foliar JA application and abrogated in mutants impaired in JA synthesis or signaling. However, Park et al. (2007) provided evidence that methyl salicylate (MeSA), rather than SA, functions as the critical mobile signal. In addition, it has been reported that a secondary oxidative burst can be detected in systemic leaves following an initial oxidative burst in inoculated leaves, and both the primary and secondary oxidative bursts are required for the onset of SAR (Alvarez et al., 1998; Lee & Hwang, 2005). It is likely that SAR is regulated by a network of interconnecting signal transduction pathways involving SA, JA and ROS (Grant & Lamb, 2006).

In analogy with SAR, the perception of light stress produces a systemic signal that sets up an acclimation response in nonexposed parts of the plant. Such a response to abiotic stress is termed ‘systemic acquired acclimation’ (SAA) (Karpinski et al., 1999; Rossel et al., 2007). This acclimation response to high light was accompanied by the enhanced expression of APX2, which encodes a cytosolic ascorbate peroxidase. Initially, the expression of APX2 was considered to be regulated by the redox state of the photosynthetic electron transport chain (Karpinski et al., 1997). Further analyses have revealed that APX2 is induced via hydrogen peroxide (H2O2), ABA signaling and altered glutathione metabolism (Fryer et al., 2003; Ball et al., 2004; Galvez-Valdivieso et al., 2009). Recently, it has been reported that high light acclimation promotes resistance to virulent biotrophic bacteria and the SAA response crosstalks with components of the SAR signaling pathway (Mühlenbock et al., 2008). Importantly, ROS homeostasis constitutes a switch controlling coordinated light acclimation and pathogen defense responses. Thus, ROSs represent a significant point of convergence between SAR and SAA.

Brassinosteroids (BRs) are natural growth-promoting substances found at low levels in pollen, seeds and young vegetative tissues throughout the plant kingdom. Extensive studies have demonstrated that BRs are required for a range of developmental processes, including stem and root growth, floral initiation and the development of flowers and fruits (Clouse & Sasse, 1998; Sasse, 2003). BRs have been implicated in plant responses to a variety of abiotic stresses, such as high and low temperature stress, drought and salinity injury (Krishna, 2003; Kagale et al., 2007). BRs are also known to protect plants against a broad spectrum of pathogens (Nakashita et al., 2003). Genetic and biochemical studies have identified key components of the BR signaling pathway that mediates BR-regulated plant growth and development (Bishop & Koncz, 2002; Gendron & Wang, 2007). However, it is unclear whether the signaling pathway mediating BR-regulated growth and development also mediates BR-enhanced stress tolerance. We have shown recently that elevated H2O2 levels resulting from enhanced NADPH oxidase activity are observed in BR-treated plants and play a critical role in BR-induced stress tolerance (Xia et al., 2009).

Although widely distributed throughout the plants, BRs are not transported over long distances between different tissues (Symons et al., 2008). However, BRs may have an indirect role in long-distance signaling through their effects on other hormones. For example, BRs affect the long-distance transport of auxin (indole acetic acid, IAA) through modulation of the expression of PIN genes (Nakamura et al., 2004; Li et al., 2005). In addition, foliar application of brassinolide reduces the nodule number of the soybean mutant En6500 (Terakado et al., 2005, 2006). This effect on soybean nodule number is thought to be mediated by an increase in polyamine content. In addition, we have shown that BRs regulate plant stress tolerance via modulation of the ROS signal, which is also involved in the systemic stress response. Therefore, it will be of interest to examine whether BRs could themselves induce systemic tolerance in untreated leaves or tissues. In the present study, we have analyzed the expression of genes and other biochemical/molecular changes associated with photooxidative stress tolerance or disease resistance in both BR-treated leaves and untreated leaves or roots of cucumber plants. We have also investigated the role of BR-induced ROS in BR-induced systemic tolerance. Our results strongly suggest that BRs are capable of inducing systemic stress tolerance in untreated leaves and roots, and enhanced H2O2 production is critical in BR-induced systemic tolerance.

Materials and Methods

Plant materials and experimental design

Cucumber (Cucumis sativus L. cv Jinyan No. 4) was used because of its known sensitivity to photooxidative stress and susceptibility to Fusarium oxysporum (Schlechtend,:Fr) f. sp. cucumerinum (Owen) Snyder & Hansen (Fusarium wilt, FO) (Ye et al., 2004). Seeds were sterilized in 2.5% NaClO and germinated in vermiculite. After emergence, batches of eight seedlings were grown hydroponically in a plastic tank (13 l) filled with 10 l of half-strength Enshi nutrient solution (Yu & Komada, 1999). Plants were placed in a glasshouse maintained at 25°C : 18°C (day : night) with a relative humidity of 90% and a photoperiod of 16 h with a maximum photosynthetic photon flux density (PPFD) of 1000 μmol m−2 s−1.

Experiment 1  To determine the effects of BR on tolerance against photooxidative stress and the transcript level of defense-related genes, the fourth leaf from the bottom of a plant at the seven-leaf stage was pretreated with 0.2 μM 24-epibrassinolide (EBR; Sigma) or water. Twenty-four hours later, the third, fourth or fifth leaf was sprayed with 10 μM paraquat (PQ; Sigma) and stress tolerance was measured on the basis of changes in the maximal photochemical efficiency (Fv/Fm) at 6, 24 and 48 h after PQ treatment. Chlorophyll fluorescence was determined with imaging pulse amplitude modulation (PAM) (IMAG-MAXI; Heinz Walz, Effeltrich, Germany). For the measurement of Fv/Fm, plants were dark-adapted for 30 min. Minimal fluorescence (Fo) was measured during the weak measuring pulses and maximal fluorescence (Fm) was measured during a 0.8-s pulse light exposure to a PPFD of c. 4000 μmol m−2 s−1. Fv/Fm was determined with the whole leaf as the area of interest (Xia et al., 2009). The third, fourth and fifth leaves were sampled at different time points after EBR treatment for H2O2 content and gene expression analysis.

Experiment 2  To determine whether foliar application of EBR could induce tolerance against root-borne pathogens, shoots of cucumber seedlings at the two-leaf stage were sprayed with 0.2 μM EBR (10 ml per plant). After 1 d, half of the EBR pretreated plants and untreated plants were inoculated with FO pathogen by addition of a conidial suspension to the nutrient solution. The FO suspension was prepared by culturing the pathogen in a potato sucrose liquid medium at 28°C for 6 d (Yu & Komada, 1999), and was added to the nutrient solution at 104 conidia ml−1. The day for inoculation was defined as 0 d after inoculation (DAI). Therefore, there were four treatments: control, EBR, FO and EBR + FO. Each treatment had three replicates with 20 plants per replicate. The experiment was finished at 18 DAI when control plants had c. 10 leaves and FO plants showed wilting symptoms with dead plants or yellowing leaves. The percentage of wilting plants was measured at 0, 3, 6, 9, 12, 15 and 18 DAI (Ye et al., 2004). During the experiment, roots were sampled, frozen quickly in liquid nitrogen and stored at −80°C before use for gene expression and biochemical analysis.

Experiment 3  To investigate the role of H2O2 in EBR-induced systemic tolerance, primary or systemic leaves were pretreated with 100 μM diphenyleneiodonium (DPI, a NADPH flavin oxidase inhibitor) or 5 mM dimethylthiourea (DMTU, an H2O2 and OH scavenger) for 8 h, and the primary leaves were treated with 0.2 μM EBR. After 1 d, leaves and roots were sampled for the detection of H2O2, analysis of H2O2 content, gene expression and enzyme activities. Subsequently, the plants were sprayed with 10 μM PQ under the same conditions as described above. Stress tolerance was measured on the basis of changes in Fv/Fm.

Experiment 4  BR biosynthetic genes are negatively feedback regulated by BR levels. To further investigate whether BRs are transported over a long distance, we also analyzed the expression of BR biosynthetic genes in primary and systemic leaves with or without pretreatment with DPI at 100 μM and DMTU at 5 mM. In addition, the concentration effects of EBR and H2O2 on the expression of BR biosynthetic genes were analyzed.

Quantitative real-time polymerase chain reaction (qRT-PCR) analysis

Total RNA was extracted from leaves and roots using Trizol according to the supplier’s recommendation. Residual DNA was removed with a purifying column. One microgram of total RNA was reverse transcribed using 0.5 μg of oligo (dT) 12–18 (Invitrogen) and 200 units of Superscript II (Invitrogen) following the supplier’s recommendation. On the basis of expressed sequence tag (EST) sequences, the gene-specific primers were designed (Supporting Information Table S1) and used for amplification.

qRT-PCR was performed using the iCycler iQ™ Real-time PCR Detection System (Bio-Rad, Hercules, CA, USA). PCR was performed using iQ SYBR Green SuperMix (Bio-Rad). The PCR conditions consisted of denaturation at 95°C for 3 min, followed by 40 cycles of denaturation at 95°C for 30 s, annealing at 58°C for 30 s and extension at 72°C for 30 s. A dissociation curve was generated at the end of each PCR cycle to verify that a single product was amplified using the software provided with the iCycler iQ™ Real-time PCR Detection System. The identity of the PCR products was verified by single-strand sequencing using the MegaBACE 1000 DNA analysis system (Amersham Biosciences, Piscataway, NJ, USA). To minimize sample variations, mRNA expression of the target gene was normalized relative to the expression of the housekeeping gene actin. All experiments were repeated three times for cDNA prepared from two samples. The quantification of mRNA levels is based on the method of Livak & Schmittgen (2001).

Antioxidant enzyme extraction and activity assay

For the enzyme assays, 0.3 g of leaf was ground with 3 ml of ice-cold 25 mM HEPES buffer (pH 7.8) containing 0.2 mM EDTA, 2 mM ascorbic acid and 2% polyvinylpyrrolidone. The homogenates were centrifuged at 4°C for 20 min at 12 000 g and the resulting supernatants were used for the determination of enzymatic activity. Ascorbate peroxidase (APX) activity was measured by a decrease in absorbance at 290 nm according to Nakano & Asada (1981). Monodehydroascorbate reductase (MDAR) activity was measured using 1 U ascorbate oxidase, and the oxidation rate of NADH was followed at 340 nm (Hossain et al., 1984).

The content of H2O2 and malonaldehyde (MDA)

The content of H2O2 was determined according to Willekens et al. (1997) with some modifications. Leaf samples (0.3 g) were homogenized in 3 ml of cooled HClO4 (1.0 M) using a prechilled pestle and mortar. The homogenate was transferred to a 10-ml plastic tube and centrifuged at 6000 g for 5 min at 4°C. The supernatant was neutralized to pH 6.0–7.0 with 4 M KOH and centrifuged at 6000 g for 1 min at 4°C. The supernatant was loaded onto an AG1×8 prepacked column (Bio-Rad) and H2O2 was eluted with double-distilled H2O. Recovery efficiencies of H2O2 from different samples were determined by analyzing duplicate samples to which H2O2 was added during grinding at a final concentration of 50 μM. H2O2 was determined by a spectrophotometric assay. The sample (900 μl) was mixed with 900 μl of reaction buffer containing 1 mM 2,2′-azino-di(3-ethylbenzthiazoline-6-sulfonic acid) in 100 mM potassium acetate at pH 4.4. The reaction was started by the addition of 3 μl (0.5 U) horseradish peroxidase. The absorption at 412 nm was recorded when the value was stable.

The thiobarbituric acid (TBA) test, which determines MDA as an end-product of lipid peroxidation in the roots, was used to measure MDA. Samples (0.5 g) were homogenized with 10% trichloroacetic acid, followed by centrifugation at 3000 g for 10 min. An aliquot of supernatant was added to a test-tube with an equal volume of either: –TBA solution containing 20% (w/v) trichloroacetic acid and 0.01% (w/v) butylated hydroxytoluene; or +TBA solution containing the above plus 0.65% (w/v) TBA. The mixtures were heated at 95°C for 25 min. The reaction was stopped by placing the reaction tubes in an ice bath; the samples were then centrifuged at 3000 g for 10 min, and the absorption of the supernatant was read at 440, 532 and 600 nm. MDA equivalents were calculated according to the method of Hodges et al. (1999).

H2O2 detection by confocal laser scanning microscopy

Intact cucumber leaves with the petiole were incubated overnight at 30°C in a growth chamber in darkness with either 25 μM 2′,7′-dichlorofluorescein diacetate (H2DCF-DA) prepared in 10 mM Tris-HCl (pH 7.4) or Tris buffer alone. The leaves were then washed twice in the same buffer for 15 min each. After washing, 15-mm-long petiole pieces were cut using a scalpel and were glued onto the sample plate using Neg-50 (Richard-Allan Scientific, Kalamazoo, MI, USA). Using a cryostat (Microm, Waldorf, Germany), 35-μm sections were cut and transferred to glass slides where they were mounted in glycerol : phosphate-buffered saline (1 : 1 v/v). The sections were examined with a confocal laser scanning microscopy system (Leica TCS SL; Leica Microsystems, Wetzlar, Germany), using standard filters and collection modalities for DCF green fluorescence (excitation, 480 nm; emission, 530 nm). Background staining, routinely negligible, was controlled with unstained petiole sections.

Statistical analysis

Plants were arranged in three randomized blocks with three replicates per treatment. The mean values of three replicates were compared using Tukey’s test (P ≤ 0.05).

Results

Induction of systemic stress tolerance, H2O2 accumulation and gene expression by EBR

To determine whether EBR treatment induces stress tolerance in systemic leaves, we treated the fourth leaves from the bottom of the cucumber plant (called local leaves) with EBR and then subjected the fifth leaves (upper leaves), fourth leaves or third leaves (lower leaves) to PQ treatment 24 h later. As shown in Fig. 1(a), Fv/Fm, an indicator for photooxidative stress, started to decline as early as 6 h after PQ treatment, and displayed a continual decline with increased time after PQ treatment in both the local and upper leaves. However, the lower leaves showed no clear signs of photooxidative stress until 24 h after treatment, but showed c. 40% decline in Fv/Fm by 48 h after PQ treatment. Treatment of the local leaves with EBR reduced the PQ-induced decline of Fv/Fm in the local leaves and, to a lesser extent, in the untreated upper and lower leaves (Fig. 1a). Thus, BR treatment induced photooxidative stress tolerance not only in the treated local leaves, but also in untreated upper and lower systemic leaves.

Figure 1.

Enhanced tolerance to paraquat (PQ)-mediated photooxidative stress in 24-epibrassinolide (EBR)-treated local leaves and untreated systemic leaves. (a) The fourth leaf of cucumber (Cucumis sativus) seedlings at the seven-leaf stage was pretreated with 0.2 μM EBR; 24 h later the fifth, fourth or third leaf from the bottom, corresponding to the upper, local or lower leaf, was exposed to 10 μM PQ (triangles); meanwhile, the plants without EBR treatment were also challenged with PQ at the same leaf locations (circles). The maximum quantum yield of photosystem II (PSII) (Fv/Fm) was monitored at different times with imaging pulse amplitude modulation (PAM) to assess the changes in tolerance. The fifth, fourth or third leaf of the plants without any treatment served as control (squares). Data represent the means of five replicates ± SD. (b) EBR was applied to the fourth leaves of the plants as in (a). H2O2 contents of the fifth, fourth or third leaf of EBR-treated (circles) and untreated (squares) plants were assayed using a spectrophotometric method. Data represent the means of five replicates ± SD. (c) Relative mRNA abundances of stress/defense-related genes in EBR-treated local and untreated upper or lower leaves determined by real-time PCR. Samples were harvested at 24 h after EBR treatment. Control, open bars; EBR treatment, closed bars. Data represent the means of three replicates ± SD. For each column of (a), (b) and (c), the top, middle and bottom graphs represent the upper, local and lower leaves, respectively.

To determine the possible involvement of ROS in EBR-induced systemic stress tolerance, we measured the levels of H2O2 in EBR-treated cucumber plants. As reported previously, the H2O2 content was rapidly elevated in EBR-treated local leaves (within 3 h) and remained elevated for at least 24 h after BR treatment (Fig. 1b). Importantly, elevated H2O2 contents were also observed in the upper and lower untreated leaves. In the upper untreated leaves, elevated H2O2 content was first observed at 6 h and returned to control levels by 72 h after EBR treatment of the local leaves. In the lower untreated leaves, the H2O2 content increased slightly but significantly throughout the entire period of the experiment. Thus, BR treatment induced H2O2 accumulation not only in local treated leaves, but also in upper and lower untreated leaves.

We also analyzed both local and systemic induction of defense/stress-related genes in BR-treated cucumber plants. As shown in Fig. 1(c), transcripts of RBOH, WRKY6, PR-1 and cAPX were consistently upregulated in upper, local and lower leaves after EBR treatment of local leaves. The expression of GST, however, was increased in local and upper leaves, but not in lower leaves, whereas the transcripts of PAL, Fe-SOD and CAT were elevated only locally in EBR-treated leaves, but not in untreated systemic leaves (Fig. 1c). These results indicate that EBR treatment results in systemic induction of at least some genes associated with plant defense and stress responses.

Induction of systemic resistance to root disease by EBR

From the above results, EBR induced systemic defense/stress responses not only in locally treated and upper untreated leaves, but also in lower untreated leaves (Fig. 1). Therefore, we were interested in extending the analysis to determine whether EBR treatment of cucumber leaves could also induce defense/stress responses in untreated roots of the same plants. To test this, we investigated the effects of the foliar application of EBR on root resistance to Fusarium pathogen. The FO-inoculated plant began to show wilt symptoms at 3 DAI, and the percentage of Fusarium-wilted plants increased steadily to 72% at 18 DAI (Fig. 2a). However, no significant Fusarium wilt was observed until 9 DAI in EBR-treated plants, and the percentage of wilted plants at 18 DAI was only one-third of that of untreated plants. Thus, foliar EBR application induced resistance of cucumber plants to a root pathogen FO.

Figure 2.

Enhanced root tolerance to wilt disease (FO, Fusarium oxysporum f. sp. cucumerinum) after foliar treatment of 24-epibrassinolide (EBR). (a) Percentages of wilted plants as influenced by FO inoculation and EBR pretreatment. Shoots of cucumber (Cucumis sativus) seedlings at the two-leaf stage were sprayed with water or 0.2 μM EBR and, 24 h later, the roots were inoculated with FO by adding a conidial suspension to the nutrient solution (104 conidia ml−1). The percentage of wilted plants was counted until 18 d after inoculation. Each treatment had three replicates with 20 plants per replicate. Data represent the means of three replicates ± SD. (b) H2O2 contents in roots of control, EBR-pretreated, FO-inoculated and EBR-pretreated plus FO-inoculated plants. H2O2 contents were assayed using a spectrophotometric method. Data represent the means of five replicates ± SD. (c) Malonaldehyde (MDA) content was measured to assess the extent of peroxidation in roots of control, EBR-pretreated, FO-inoculated and EBR-pretreated plus FO-inoculated plants. Data represent the means of five replicates ± SD.

Foliar EBR treatment alone induced only a transient increase in H2O2 in roots at 1 DAI (Fig. 2b). H2O2 contents were also increased significantly in FO-inoculated plants at 1 DAI. The H2O2 content in FO-inoculated plants was 123% higher than that of uninoculated plants at 2 DAI and remained at a high level during the remaining period of the experiments. When the plants were pretreated with EBR, however, H2O2 contents in FO-inoculated plants were significantly lower than those of FO-inoculated plants without EBR pretreatment, although the contents were still elevated when compared with those of control plants without EBR pretreatment or FO inoculation. We also determined the effects of EBR pretreatment and FO inoculation on the content of MDA, which is a product of membrane peroxidation. EBR treatment reduced the accumulation of MDA during the first 6 d after treatment, with the MDA content in EBR-treated plants only c. 46% of that of controls at 6 d after treatment (Fig. 2c). In FO-inoculated plants, however, the MDA content increased steadily over the 9-d experimental period. Pretreatment with EBR substantially reduced the FO-induced MDA content. Indeed, the MDA contents in EBR-pretreated and FO-inoculated plants were similar to those of control plants during the first 6 d after inoculation.

To determine the mechanism of EBR-induced root resistance to FO, we analyzed the expression patterns of several genes involved in the defense/stress response. Foliar treatment with EBR induced prolonged upregulation of WRKY30, WRKY6, PR-1, PAL, GST and CAT for 2 d and a transient increase in transcripts of cAPX and POD in roots (Fig. 3). FO inoculation also induced these defense-related genes. Elevated transcripts of POD were detected at 3 d after FO inoculation. However, transcripts of other genes displayed only a transient increase, and no significant change in the transcript level of CAT was observed in the roots of FO-inoculated plants. Importantly, foliar pretreatment with EBR enhanced the induction of defense genes by FO inoculation. Elevated levels of transcripts for all tested defense genes, except cAPX, were detected earlier in EBR-pretreated and FO-inoculated plants than in plants that were inoculated with FO alone. In addition, the magnitude of induction for WRKY60, WRKY6, PR-1, POD, cAPX and CAT in FO-inoculated plants was augmented by EBR pretreatment. For example, at 2 DAI, the mRNA abundance of WRKY6 in EBR-pretreated and FO-inoculated plants was > 1.3-fold higher than that in plants with FO inoculation alone.

Figure 3.

Expression of stress/defense-related genes in cucumber (Cucumis sativus) roots as influenced by Fusarium oxysporum f. sp. cucumerinum (FO) inoculation and foliar 24-epibrassinolide (EBR) pretreatment. Relative mRNA abundances were determined by real-time PCR using root samples from control, EBR-pretreated, FO-inoculated and EBR-pretreated plus FO-inoculated plants. Data represent the means of three replicates ± SD.

Involvement of H2O2 in EBR-induced systemic tolerance

Because elevated H2O2 is produced in EBR-treated plants and has been shown previously to be involved in systemic signaling (Karpinski et al., 1999), we next examined the role of local leaf-derived H2O2 in the induction of the stress response in systemic leaves. As expected, after local EBR treatment, the accumulation of H2O2 was observed in both upper and lower untreated leaves (Fig. 4a). In addition, local EBR treatment induced the enhancement of tolerance to photooxidative stress in upper and lower untreated leaves (Fig. 4b,c). The enhanced tolerance in systemic leaves was associated with the upregulation of cAPX and MDAR transcripts and increases in their corresponding enzyme activity (Fig. 5). When local leaves were pretreated with DPI or DMTU, the systemic induction of H2O2 by EBR was abolished, and the induced expression of cAPX and MDAR, enzyme activities and tolerance to photooxidative stress in systemic leaves were inhibited. Thus, increased H2O2 accumulation in EBR-treated local leaves is important for both local and systemic stress tolerance.

Figure 4.

Systemic induction of stress tolerance mediated by reactive oxygen species produced from 24-epibrassinolide (EBR)-treated local leaves. (a) Effects of NADPH flavin oxidase inhibitor diphenyleneiodonium (DPI) or hydrogen peroxide (H2O2) scavenger dimethylthiourea (DMTU) pretreatment of EBR-treated local leaves on systemic induction of H2O2. The fourth leaf of cucumber (Cucumis sativus) seedlings was pretreated with 100 μM DPI or 5 mM DMTU; then, 8 h later, the leaves were treated with 0.2 μM EBR. H2O2 contents of EBR-treated local and untreated upper/lower leaves were assayed using a spectrophotometric method. Data represent the means of five replicates ± SD. Mean values of the same leaf location were compared by Tukey’s test. Means denoted by the same letter do not differ significantly at P ≤ 0.05. (b) Effects of DPI or DMTU pretreatment of EBR-treated local leaves on systemic induction of tolerance to photooxidative stress. Plants were treated as above. After EBR treatment, the upper, local or lower leaf was sprayed with 10 μM paraquat (PQ). After 24 h, Fv/Fm was measured using imaging pulse amplitude modulation (PAM). Fv/Fm was determined using the whole leaf as the area of interest. Data represent the means of five replicates ± SD. Mean values of the same leaf location were compared by Tukey’s test. Means denoted by the same letter do not differ significantly at ≤ 0.05. (c) Representative images of Fv/Fm indicating photooxidative injury. The false color code depicted at the bottom of the image ranges from 0.0 (black) to 1.0 (purple).

Figure 5.

Effects of diphenyleneiodonium (DPI) or dimethylthiourea (DMTU) pretreatment of 24-epibrassinolide (EBR)-treated local leaves on systemic induction of expression of antioxidant genes. The fourth leaf of cucumber (Cucumis sativus) seedlings was pretreated with 100 μM DPI or 5 mM DMTU; then, 8 h later, the leaves were treated with 0.2 μM EBR. Relative mRNA abundances of cAPX and MDAR in EBR-treated local and untreated upper/lower leaves were determined by real-time PCR. The activity of the corresponding enzyme encoded by the genes was assayed as described in the Materials and Methods section. Data represent the means of five replicates ± SD. Mean values of the same leaf location were compared by Tukey’s test. Means denoted by the same letter did not differ significantly at P ≤ 0.05. APX, ascorbate peroxidase; MDAR, monodehydroascorbate reductase.

We next determined the role of H2O2 production in untreated systemic leaves in EBR-induced systemic stress tolerance. As shown in Fig. 6(a), there was no significant increase in H2O2 accumulation in the upper or lower leaves treated with DPI after local EBR treatment. Similarly, the systemic tolerance to photooxidative stress, expressed as Fv/Fm (Fig. 6b), and the expression of cAPX and MDAR (Fig. 7) were also inhibited by DPI treatment of systemic leaves. Likewise, scavenging of H2O2 by DMTU completely decreased the H2O2 content and inhibited the protective mechanism in systemic leaves. These results suggest that the systemic accumulation of H2O2 and the induction of stress tolerance require the production of H2O2 in both EBR-treated local leaves and untreated systemic leaves.

Figure 6.

Suppression of systemic stress tolerance by inhibition of reactive oxygen species accumulation in 24-epibrassinolide (EBR)-untreated leaves. (a) Effects of diphenyleneiodonium (DPI) or dimethylthiourea (DMTU) pretreatment of EBR-untreated upper and lower leaves on systemic induction of hydrogen peroxide (H2O2). The fifth or third leaf of cucumber (Cucumis sativus) seedlings was pretreated with 100 μM DPI or 5 mM DMTU; then, 8 h later, the fourth leaves were treated with 0.2 μM EBR. H2O2 contents of EBR-treated local and untreated upper/lower leaves were assayed using a spectrophotometric method. Data represent the means of five replicates ± SD. Mean values of the same leaf location were compared by Tukey’s test. Means denoted by the same letter do not differ significantly at P ≤ 0.05. (b) Effects of DPI or DMTU pretreatment of EBR-untreated upper and lower leaves on systemic induction of tolerance to photooxidative stress. Plants were treated as above. After EBR treatment, the upper, local or lower leaves were sprayed with 10 μM paraquat (PQ). After 24 h, Fv/Fm was measured using imaging pulse amplitude modulation (PAM). Fv/Fm was determined using the whole leaf as the area of interest. Data represent the means of five replicates ± SD. Mean values of the same leaf location were compared by Tukey’s test. Means denoted by the same letter do not differ significantly at P ≤ 0.05. (c) Representative images of Fv/Fm, indicating photooxidative injury. The false color code depicted at the bottom of the image ranges from 0.0 (black) to 1.0 (purple).

Figure 7.

Effects of diphenyleneiodonium (DPI) or dimethylthiourea (DMTU) pretreatment of 24-epibrassinolide (EBR)-untreated leaves on systemic induction of expression of antioxidant genes. The fifth or third leaf of cucumber (Cucumis sativus) seedlings was pretreated with 100 μM DPI or 5 mM DMTU; then, 8 h later, the fourth leaves were treated with 0.2 μM EBR. Relative mRNA abundances of cAPX and MDAR in EBR-treated local and untreated upper/lower leaves were determined by real-time PCR. The activity of the corresponding enzyme encoded by the genes was assayed as described in the Materials and Methods section. Data represent the means of five replicates ± SD. Mean values of the same leaf location were compared by Tukey’s test. Means denoted by the same letter do not differ significantly at P ≤ 0.05. APX, ascorbate peroxidase; MDAR, monodehydroascorbate reductase.

Similarly, foliar treatment of EBR induced H2O2 accumulation and upregulation of defense genes in roots (Figs S1, S2). Furthermore, induction of H2O2 accumulation and defense genes by EBR was suppressed by pretreatment with DPI and DMTU (Figs S1, S2). These results strongly suggest that ROSs are also involved in EBR-induced systemic stress tolerance in roots.

Detection of elevated H2O2 in vascular tissues

To further analyze the systemic accumulation of H2O2, we determined the H2O2 signal in the vascular tissues following treatment of leaves with EBR. H2O2 was detected by loading the leaves with DCF fluorescent stain in the dark, and H2O2 in the cross-sections of leaf petioles was monitored by confocal laser scanning microscopy. Fig. 8 shows that local EBR treatment of the fourth leaves induced H2O2 signal around the vascular tissue of the fifth leaf petioles. When the fourth leaves were pretreated with NADPH oxidase inhibitor, DPI, or H2O2 scavenger, DMTU, EBR-induced H2O2 accumulation in the fourth leaves was abolished and no significant increases in DCF fluorescent signal were observed in the vascular tissue of the fourth or fifth leaf petioles.

Figure 8.

Images of hydrogen peroxide (H2O2) localization in the vascular tissue of petioles of systemic leaves. The fourth leaf of cucumber (Cucumis sativus) seedlings was pretreated with 100 μM diphenyleneiodonium (DPI) or 5 mM dimethylthiourea (DMTU); then, 8 h later, the leaves were treated with 0.2 μM 24-epibrassinolide (EBR). The fifth systemic leaf with the petiole was detached from the plants and loaded with 2′,7′-dichlorofluorescein diacetate (H2DCF-DA) overnight. H2O2 accumulation in the vascular tissue of the petiole segment was detected by confocal laser scanning microscopy (excitation, 480 nm; emission, 530 nm).

Systemic induction of BR biosynthetic genes by EBR

Based on a previous study (Symons & Reid, 2004), BRs are not mobile from locally treated leaves to other untreated systemic leaves, and therefore EBR-induced systemic stress tolerance in the upper and lower systemic leaves is most probably mediated by a signal(s) other than EBR itself. To test this hypothesis, we attempted to determine the possible changes in BR levels in the upper (fifth) and lower (third) leaves following EBR treatment of the fourth leaves. As direct quantification of BRs is difficult, we used an indirect approach by analyzing the expression of four BR biosynthetic genes (CPD, DWF4, DET2 and BR6ox), which are subjected to sensitive repression by BRs. As expected, transcripts for all four BR biosynthetic genes were reduced in the local fourth leaves after EBR treatment (Fig. 9). Surprisingly, the expression of the biosynthetic genes was elevated in the upper systemic leaves following EBR treatment of the fourth leaves. In the lower systemic leaves, transcripts for two (CPD and DWF4) of the four BR biosynthetic genes were also elevated, and the other two genes (DET2 and BR6ox) showed no changes. The elevated expression of BR biosynthetic genes in systemic upper and lower leaves argues against the transport of exogenously applied EBR from treated local leaves to untreated systemic leaves.

Figure 9.

Systemic induction of brassinosteroid biosynthetic genes by 24-epibrassinolide (EBR). Relative mRNA abundances of CPD, DWF4, DET2 and BR6ox in EBR-treated local and untreated upper/lower cucumber (Cucumis sativus) leaves were determined by real-time PCR. Control, open bars; EBR treatment, closed bars. Data represent the means of three replicates ± SD. Bars with an asterisk indicate a significant difference from the untreated control at P ≤ 0.05 according to Tukey’s test.

Induction of BR biosynthetic genes by H2O2

As BRs repress BR biosynthetic genes, the induction of BR biosynthetic genes in systemic leaves on local treatment of EBR is most probably mediated by a signal, such as H2O2, produced on local treatment of EBR. Alternatively, on local treatment of EBR, high levels of EBR may repress BR biosynthetic genes in local leaves, but the low levels of EBR derived from treated local leaves may induce BR biosynthetic genes in systemic leaves. To test these possibilities, we analyzed the effects of a wide range of H2O2 and EBR concentrations on the expression of BR biosynthetic genes. As shown in Fig. 10(a), very low concentrations of EBR (0.1–10 nM) had little or no effects on the transcript levels of the CPD and DET2 genes. When EBR concentrations were increased to 50–200 nM, we observed a 30–50% reduction in the transcript levels of the biosynthetic genes. Thus, even low concentrations had no stimulating effect on BR biosynthetic gene expression. When we tested the effect of H2O2 on BR biosynthetic genes, however, a different picture emerged. When applied at extremely low concentration (0.01 mM), H2O2 had little effect on the expression of these genes (Fig. 10b). However, when the concentration of H2O2 was increased to 0.05 mM, we observed a c. 70% increase in the transcript level of CPD and DET2. At higher H2O2 concentrations (0.1–50 mM), the stimulating effect of H2O2 on expression of the CPD gene was still observed, although the magnitudes were somewhat reduced. At very high concentrations of H2O2, however, the stimulating effect of H2O2 on BR biosynthetic gene expression was not observed. Thus, low concentrations of H2O2 increase BR biosynthetic gene expression.

Figure 10.

Expression of brassinosteroid biosynthetic genes in response to different concentrations of 24-epibrassinolide (EBR) or hydrogen peroxide (H2O2). Relative mRNA abundances of CPD and DET2 in cucumber (Cucumis sativus) leaves after treatment with EBR (0.1–200 nM) or H2O2 (0.01–100 mM) were determined by real-time PCR. Data represent the means of three replicates ± SD. Points with an asterisk indicate a significant difference from the untreated control at P ≤ 0.05 according to Tukey’s test.

To further analyze the role of EBR-induced H2O2 in the systemic induction of BR biosynthetic genes, we tested the effects of DPI and DMTU in EBR-treated plants. As shown in Fig. 11, transcript levels for CPD, DWF4 and DET2 were reduced in local leaves after exposure to EBR, and this decrease was almost independent of DPI and DMTU pretreatment. As described earlier, treatment of the fourth leaves with EBR led to increased expression of BR biosynthetic genes in the upper fifth leaves. This systemic induction of BR biosynthetic genes by EBR treatment, however, was largely abolished by treatment of either the local (fourth) or upper systemic (fifth) leaves with DPI or DMTU. These results suggest that the production of H2O2 in both local and systemic leaves is important for EBR-induced expression of BR biosynthetic genes in systemic leaves.

Figure 11.

Systemic induction of brassinosteroid biosynthetic genes mediated by local and systemic reactive oxygen species accumulation. The local (fourth) or systemic (fifth) cucumber (Cucumis sativus) leaf was pretreated with 100 μM diphenyleneiodonium (DPI) or 5 mM dimethylthiourea (DMTU); then, 8 h later, the fourth leaves were treated with 0.2 μM EBR. Relative mRNA abundances of CPD, DWF4 and DET2 in EBR-treated local and untreated systemic leaves were determined by real-time PCR. Data represent the means of three replicates ± SD. Means denoted by the same letter do not differ significantly at P ≤ 0.05 according to Tukey’s test.

Discussion

Over the years, a number of groups, including ours, have discovered that the application of exogenous BRs enhances plant tolerance to a wide range of biotic and abiotic stresses in important crops, including potato, maize, rice, cucumber and tomato. In this study, we demonstrated that local exposure of EBR led to systemic induction of stress tolerance and disease resistance in untreated distal tissues. EBR treatment enhanced tolerance to PQ-activated photooxidative stress not only in treated leaves, but also in upper and lower untreated leaves (Fig. 1a). Foliar treatment of EBR also activated systemic defense responses in untreated roots, resulting in the expression of defense-related genes and enhanced resistance to the root-infecting pathogen FO (Fig. 2). More recently, it has been shown that foliar application of brassinolide decreases the nodule number of soybean, which is associated with decreased leaf concentrations of spermidine (Terakado et al., 2005, 2006). Taken together, these results confirm that BRs are capable of activating systemic signaling in a variety of plants. The activity of bioactive BRs in enhancing root disease resistance might provide a new way to control plant soil diseases. FO is one of the major soil-borne pathogens causing soil sickness in commercial production of cucumber. The strong suppressive effect of EBR on the incidence of Fusarium wilt suggests that the application of BRs could potentially be developed into a new method for controlling the soil-borne pathogen.

Several lines of evidence indicate that BRs do not undergo long-distance transport between the shoot and root or between different tissues within the shoot (Symons & Reid, 2004; Symons et al., 2008). In cucumber, little radioactivity of 14C-labeled EBR moved out of the leaf during the first 3 d after application (Nishikawa et al., 1994). Through analysis of BR biosynthetic genes, we provided further evidence that EBR is not mobile in plants. Transcript levels for the BR biosynthetic genes were reduced in EBR-treated local leaves, but were significantly induced in both upper and lower untreated systemic leaves (Fig. 9). Thus, the systemic stress response in untreated leaves or roots in EBR-treated plants is likely to be mediated by a signal other than EBR. The EBR-induced systemic tolerance displayed some of the characteristics observed with SAR, which is dependent on SA (Fig. 1c). However, it has been reported that BR-mediated disease resistance is independent of SA accumulation (Nakashita et al., 2003). Likewise, we have observed that EBR-induced stress tolerance is not associated with increased SA accumulation (Xia et al., 2009). Therefore, the mechanism of EBR-induced systemic tolerance appears to be distinct from that of SAR and SAA.

Recently, we have reported that BR treatment results in increased NADPH oxidase activity and H2O2 accumulation. We have also provided strong evidence that increased H2O2 levels play an important role in the BR-induced increase in tolerance to oxidative stress and resistance to cucumber mosaic virus (Xia et al., 2009). In this study, we observed that elevated levels of H2O2 were detected not only in EBR-treated leaves, but also in untreated distant leaves and roots (Figs 1, 2). Pretreatment of EBR-treated local leaves with the NADPH flavin oxidase inhibitor DPI or H2O2 scavenger DMTU substantially inhibited systemic H2O2 accumulation, lowered the expression and activity of antioxidant enzymes, and consequently reduced tolerance to photooxidative stress in systemic leaves (Figs 4, 5). Thus, EBR-induced ROS production in local treated leaves is important for EBR-induced systemic stress tolerance. Interestingly, treatment of the systemic leaves with DPI also inhibited EBR-induced systemic stress tolerance (Fig. 6), indicating that the continuous production of ROS in tissues distant from EBR-treated leaves is also necessary for EBR-induced systemic stress tolerance.

BRs repress the expression of BR biosynthetic genes. By contrast, BR-induced H2O2 appears to positively regulate the expression of BR biosynthetic genes (Fig. 10). Systemic induction of BR biosynthetic genes by EBR treatment was largely abolished by treatment of either the local or upper systemic leaves with DPI or DMTU (Fig. 11). Furthermore, low concentrations of H2O2 increased significantly the expression of BR biosynthetic genes (Fig. 10b). The opposite effect of BRs and H2O2 on BR biosynthetic genes raises the possibility that BR-induced H2O2 production is mediated by a signaling pathway distinct from the well-established growth-regulating signaling pathway that represses BR biosynthetic genes. In EBR-treated leaves, the inhibitory effects of high levels of EBR may be greater than the stimulatory effects of EBR-induced H2O2 on BR biosynthetic genes, leading to their repression. In upper systemic leaves, however, EBR levels are unchanged because of a lack of long-distance transport, but systemic H2O2 production is induced, which would, in turn, stimulate the expression of BR biosynthetic genes.

Systemic signaling has been demonstrated to play an important role in plant responses to a variety of biotic and abiotic stimuli. The accumulation of ROS in both local and systemic tissues, including vascular tissues, has been shown to be associated with systemic signaling of plant responses to some, but not all, environmental challenges. In tomato, H2O2 was detected near cell walls of vascular bundle cells in response to wounding, and may act as a secondary messenger for the activation of defense genes in mesophyll cells (Orozco-Cárdenas et al., 2001). H2O2 accumulation has also been observed in the bundle sheath cells of Arabidopsis vascular tissues, an event that may be important for the initiation of the subsequent high light responses (Fryer et al., 2003). Likewise, we observed that EBR treatment led to H2O2 accumulation in local leaves, systemic leaves and the vascular tissue of petioles (Figs 1, 8). H2O2 is a relatively stable molecule and can migrate from the subcellular sites of synthesis to adjacent compartments and even neighboring cells (Henzler & Steudle, 2000; Bienert et al., 2006). However, several lines of evidence indicate that systemic accumulation of H2O2 does not result from long-distance transport of ROS from local treated leaves through vascular tissues. First, we have observed that treatment of not only the EBR-treated local leaves but also the upper untreated systemic leaves with the NADPH flavin oxidase inhibitor DPI suppresses systemic H2O2 accumulation (Figs 4, 6). Similar results have also been observed in wound-induced systemic signaling (Miller et al., 2009). These results indicate that systemic accumulation of H2O2 is dependent on ROS production at locations distant from the signal initiation site. Furthermore, there have been no reports showing that direct application of H2O2 is able to activate systemic stress responses, including systemic induction of stress/defense-related genes. Thus, systemic accumulation of ROS appears to result from the continuous production of ROS along the path of spread of the systemic signal(s).

In Arabidopsis, wound-induced local and systemic ROS accumulation is suppressed in the rbohD mutant, indicating that NADPH oxidase is involved in systemic ROS production (Miller et al., 2009). As H2O2 from the initiation site is important for the continuous production of H2O2 along the path, the ROS signal may itself act as a regulatory molecule capable of activating NADPH oxidase, thereby establishing a positive feedback loop allowing for the autopropagation of the ROS signal. NADPH oxidase proteins contain N-terminal calcium-binding EF hand motifs, and therefore direct activation of plant NADPH oxidases by calcium may be important for rapid stimulation of the oxidative burst (Sagi & Fluhr, 2001). Interestingly, in both plant and animal cells, ROSs are known to stimulate membrane depolarization and calcium influx. In addition, it has been shown that increased ROS accumulation and calcium influx lead to mitogen-activated protein kinase (MAPK) activation (Apel & Hirt, 2004; Mittler et al., 2004). Other studies have shown that MAPKs act downstream of H2O2 and are required for the activity of NADPH oxidase during signal amplification (Mittler et al., 2004; Zhang et al., 2006; Lin et al., 2009). Thus, ROS, calcium influx, MAPKs and NADPH oxidases may form a positive feedback loop that allows for the rapid signal propagation required for systemic signaling during plant responses to a variety of environmental stimuli.

There is also an increasing body of evidence indicating that ROSs act in coordination with plant hormones in the regulation of plant acclimatory and defense responses. For example, ABA, ET, SA and JA are important for the induction of cell death, systemic signaling and the spread of cell death during O3 stress, whereas impairment of a putative receptor of ROS, GRI, results in the suppression of hormone responses (Kangasjärvi et al., 2005; Wrzaczek et al., 2009). Likewise, recently, we have found that ABA has a critical role in some aspects of BR-regulated stress responses (J-Q. Yu et al., unpublished results). ABA, and perhaps other hormones as well, may be positive components in the ROS self-amplifying loop, thereby promoting ROS production during stress responses. Plant hormones may also promote ROS-mediated stress responses by acting in concert or synergistically with ROS in the activation of genes associated with stress and defense responses. Further studies on how ROSs functionally interact with other signal molecules, including plant hormones, will be important for a better understanding of ROS signaling, in particular, and plant stress responses in general.

Acknowledgements

We are grateful to Professor C. Foyer of Leeds University for kind advice on H2O2 analysis. This work was supported by the National Basic Research Program of China (2009CB119000), National Natural Science Foundation of China (31000905; 30671428; 30972033), Program for Promotion of Basic Research Activities for Innovative Bioscience (PROBRAIN) and Foundation for the Author of National Excellent Doctoral Dissertation of China (200766).

Ancillary

Advertisement