Only an early nitric oxide burst and the following wave of secondary nitric oxide generation enhanced effective defence responses of pelargonium to a necrotrophic pathogen

Authors


Author for correspondence: Jolanta Floryszak-Wieczorek Tel: +48 061 848 7181 Fax: +48 061 848 7179 Email: florysza@jay.au.poznan.pl

Summary

  • • Participation of nitric oxide (NO) in cross-talk between ivy pelargonium (Pelargonium peltatum) leaves and Botrytis cinerea was investigated using electrochemical and biochemical approaches.
  • • In response to the necrotroph, leaves initiated a near-immediate NO burst, but the specificity of its generation was dependent on the genetic makeup of the host plant.
  • • In the resistant cultivar, a strong NO burst was followed by a wave of secondary NO generation, shown by bio-imaging with DAF-2DA. The epicentre of NO synthesis was located in targeted cells, which exhibited a TUNEL-positive reaction. Soon after the challenge, an elevated concentration of hydrogen peroxide (H2O2) was correlated with a reversible inhibition of catalase (CAT), ascorbate peroxidase (APX), and suppression of ethylene synthesis. The induced NO generation initially expanded and then gradually disappeared on successive days, provoking noncell-death-associated resistance with an enhanced pool of antioxidants, which finally favoured the maintenance of homeostasis of surrounding cells.
  • • By contrast, in the susceptible pelargonium, a weak NO burst was recorded and further NO generation increased only as the disease progressed, which was accompanied by very intensive H2O2 and ethylene synthesis. The pathogen colonizing susceptible cells also acquired the ability to produce considerable amounts of NO and enhanced nitrosative and oxidative stress in host tissues.

Introduction

In the course of evolutionary changes, pathogens have developed numerous invasion strategies, to which plants have responded with a wide range of defence responses. The plant–pathogen system is very dynamic, which results in many signals originating from both the pathogen and the plant. The success of one of the adversaries, that is, the development of disease or the elimination of the pathogen, is determined by the speed and efficiency of early defence responses initiated by the plant, thus activating the further sequence of events. Frequently the effect of this interaction is host cell death, which accompanies the development of disease or is a manifestation of an effective plant defence mechanism (i.e. a hypersensitive response (HR)). From the point of view of pathogen biology, the death of host cells may be advantageous for necrotrophic parasites feeding on dead matter, while it is disadvantageous for biotrophic parasites, for which a live cell is a nutrient source. At present it is believed that a HR may also be induced by necrotrophic pathogens (Mayer et al., 2001). However, it still needs to be determined in what ways resistance is realized in the vast majority of plants devoid of resistance genes or with an unknown genotype, which are exposed to attacks by various pathogens.

Thus, it is justified to enquire whether a plant, while improving its defence strategy, initiates a different metabolic response to necrotrophic, rather than biotrophic, pathogens. Such an approach seems well founded. Results show that a necrotrophic microorganism initiates a defence response via jasmonic acid and/or ethylene, whereas a biotrophic pathogen activates the resistance pathway via salicylic acid (Thomma et al., 1998; Asai et al., 2000; McDowell & Dangl, 2000; Overmyer et al., 2003).

The death of attacked cells is usually preceded by an oxidative burst, which so far has been considered to be an important element of a successful defence strategy of the plant against biotrophic pathogens (Wojtaszek, 2000). Enhanced reactive oxygen species (ROS) generation was also found to accompany an infection caused by necrotrophs (Tiedemann, 1997; Able, 2002; Unger et al., 2005), although in that case the death of the host cells is advantageous for the pathogen (Govrin & Levine, 2000).

In the context of early defence response events, the potential interplay of nitric oxide (NO) and ROS seems to be of special interest (Bolwell, 1999; Delledonne et al., 2002; Hancock et al., 2002; Zago et al., 2006; Zaninotto et al., 2006). Obtained data indicate that NO, acting directly or indirectly as a signal, modifies poststress plant metabolism, often leading to a HR of challenged cells. Most evidence showing that NO functions as a messenger in gene-for-gene defence responses was obtained when analysing various plant–biotrophic pathogen systems (Delledonne, 2005).

It still remains to be determined what role is played by NO in the cross-talk between the plant and the necrotrophic pathogen (Arasimowicz & Floryszak-Wieczorek, 2007). Only van Baarlen et al. (2004) recorded the generation of both endogenous NO and H2O2 in contrast to the compatible interaction, that is during disease development, in the case of lily and Botrytis elliptica.

The investigations presented here were undertaken to clarify whether NO is generated and involved in the interaction between pelargonium and Botrytis cinerea. For this purpose, two low-invasive techniques were used to assay in situ NO in the tissue, namely an electrochemical method using a microelectrode and a cytochemical method with a fluorochrome, DAF-2DA. When attempting to determine how NO regulates other signals and the redox balance in defence responses of pelargonium leaves to a necrotrophic pathogen, the effect of NO donors on ethylene and H2O2 were analysed, together with the activity of catalase (CAT), ascorbate peroxidase (APX) and the antioxidant pool in host tissues.

Materials and Methods

Plant material and culture conditions

Experiments were conducted on two cultivars of ivy-leaved pelargonium (Pelargonium peltatum L.): ‘Shiva’, which is susceptible, and ‘Cascade’, which is resistant to Botrytis cinerea. Both cultivars were bred by Fisher GmbH & Co. KG and were produced by crossing wild forms (originating from North Africa) with cultivated forms. Cultivar ‘Cascade’ originates from a very old cv. ‘Ville de Paris’ (P. hederaefolium) and is a diploid, while cv. ‘Shiva’ is a tetraploid. Both cultivars propagated vegetatively are genetically stable.

Plants were placed in a growth chamber with a constant air temperature of 21 ± 2°C at a light intensity of 120 µmol m−2 s−1 (fluorescent lamps, cool-white type – TLD 36/64 Philips) with a 12 h photoperiod. Analyses were performed on plants at the 10-leaf stage.

Fungal culture

Botrytis cinerea Pers. isolate 1072 was cultured in the dark at a temperature of 23 ± 2°C, on a potato-agar medium with the addition of 2% glucose, pH 6.3. Culture was restored monthly by mycelium passage onto a fresh medium. Conidial suspension of B. cinerea was prepared from a 3 d mycelial culture in a solution of 0.1 m glucose and 0.05 m KH2PO4. The suspension was filtered through a sterile sieve and the concentration of pathogen spores was estimated using a Bürker's chamber. The concentration of 7.5 × 105B. cinerea spores per 1 ml of 0.1 m glucose and 0.05 m KH2PO4 was applied in the experiments.

Inoculation

Three drops (40 µl) of the B. cinerea conidial suspension were transferred onto the upper surface of each leaf blade placed on solidified agar. After inoculation, plant material was placed in a growth chamber, in which humidity was maintained at c. 95%.

Assessment of disease development

The area of disease spots (mm2) was measured at the site of inoculation. Next the mean area of a disease spot (mm2) was calculated in relation to the area of the transferred inoculum drop. Control means 100%.

Effect of NO on mycelial growth of B. cinerea on the medium

To a potato-agar medium with the addition of 1% glucose, we added 5 ml of solution of a NO donor (100 µm sodium nitroprusside (SNP), 100 µm S-nitroso-N-acetyl-d-penicillamine (SNAP) or 200 µm S-nitrosoglutathione (GSNO)) through a sterile filter (pore diameter = 0.2 µm). Mycelium of B. cinerea was placed centrally on the medium and the diameter (cm) of the area occupied by B. cinerea mycelium was measured 2 d and 2 wk later.

Treatment of plant material with exogenous NO and B. cinerea

To investigate the sequential effect of NO and B. cinerea on selected physiological parameters, leaf blades after a 5 h incubation (in the light, at 18 ± 2°C) with a specific NO donor or with H2O (control) were transferred onto solidified agar and point-inoculated with B. cinerea spores at a concentration of 7.5 × 105. Both control leaves and inoculated leaves were placed in a growth chamber at 18 ± 2°C at a relative humidity of 95%. For the analyses, tissue was collected at 2, 18, 24 and 48 h after inoculation, by cutting discs (diameter = 1.0 cm) with a cork borer, from the site of inoculation.

NO detection by confocal laser scanning microscopy

Nitric oxide formation was detected by using a fluorescent dye, DAF-2DA (Callbiochem, Darmstadt, Germany). Leaf sections were placed in 1 ml of buffer solution (10 mm Tris-HCl, pH 7.2). They were then incubated for 1 h at room temperature with 1 ml of DAF-2DA at a final concentration of 10 µm in loading buffer (10 mm Tris-HCl, pH 7.2), added from a 5 mm stock in dimethyl sulphoxide (DMSO). The incubation solutions were then eliminated and leaf sections were washed three times with fresh loading buffer to remove excess fluorophore. After several minutes, the sections were affixed with silicon grease to the cover slip bottom of a chamber slide, where they remained immersed in 250 µl of fresh loading buffer.

A Zeiss Axiovert 200 m inverted microscope equipped with a confocal laser scanner (Zeiss LSM 510, Germany) was used in this study and sections were excited with the 488 line of an argon laser. Dye emissions were recorded using a 505–530 nm bandpass filter and the autofluorescence of chloroplasts was captured with a 585 nm long-pass filter. Microscope, laser and photomultiplier settings were held constant during the experiment in order to obtain comparable data. Images were processed and analysed using Zeiss LSM 510 software.

TUNEL assay

The assay measures DNA fragmentation using the terminal deoxynucleotidyl transferase (TdT)-mediated deoxyuridine triphosphate (dUTP) nick-end-labelling method, in which TdT incorporates fluorescein-12-dUTP on the 3′-OH ends of fragmented DNA. An antifluorescein-alkaline phosphatase conjugate (TUNEL AP, Roche, Indianapolis, IN, USA) subsequently binds dUDP-fluorescein, and in the presence of Fast Red (Roche) shows red precipitate forms in the nuclei containing fragmented DNA. Thin slices of the surface of leaf tissue were removed with a bistoury and immediately immersed for 1 h in 4% formaldehyde in phosphate-buffered saline (PBS). After being rinsed three times in PBS, the samples were treated with liquid nitrogen. The samples were then rehydrated and incubated for 2 min in Triton X-100 solution (0.1% Triton X-100 in 0.1% (M) sodium citrate) on ice. After rinsing the samples in PBS, repeated three times, the TUNEL reaction was performed according to the manufacturer's protocol. Negative controls were conducted in the absence of the TUNEL enzyme. In positive controls, tissue was incubated with DNase I (Roche) for 10 min at 25°C, before labelling. The samples were examined under a light microscope. Experiments were repeated at least four times, with 10 slices per treatment.

NO detection and quantification by an electrochemical method

All electrochemical measurements were performed using a universal electrochemical analyser PGSTAT 30 (EcoChemie, Utrecht, the Netherlands). NO generation in leaf tissue was monitored by differential pulse amperometry with a NO-selective needle-type electrode. The electrode was prepared by electropolymerizing a thin film of polyeugenol on a cleaned Pt needle. This was done by repetitive scanning of the electrode potential between –0.2 and 0.6 V in 10 mm solution of eugenol (Fluka, Seelze, Germany) in 0.1 m NaOH (Ciszewski & Milczarek, 2003). The modified electrode was then conditioned by applying a constant potential of 0.9 V in a phosphate buffer (pH 7.4) until a stable constant background current was reached. Electrochemical monitoring of NO generation in leaf tissue upon inoculation with a pathogen was performed as follows. A healthy leaf blade was placed on an agar layer in which a Pt wire was introduced as a counter electrode. Then an AgCl-coated Ag needle was introduced into the leaf tissue close to the area of NO monitoring to serve as a quasi-reference electrode, and finally the leaf blade was gently scratched with a gold-plated needle and the NO-selective microelectrode was placed in the scratch. The electrochemical measurement was done by differential pulse voltammetry (DPV) in the range 0.5–1.0 V. In the absence of NO (i.e. before pathogen treatment), a nearly sigmoidal DPV curve was recorded. After the action of the pathogen, a peak appeared on the voltammogram at c. 0.8 V, whose height was proportional to the NO concentration. To stimulate NO release, the conidial suspension (10 µl) was placed next to the NO-monitoring electrode and it was distributed along the scratches by capillary forces.

Ethylene production assay

Leaf discs (diameter = 1.0 cm) were pretreated with NO donors and/or inoculated with B. cinerea as described earlier. After incubation, 10 leaf discs were placed in a 20 ml glass vial closed with a crimp top cap with a Teflon/silicon septum. Ethylene was extracted by the SPME (solid phase microextraction) technique with Carboxen/PDMS SPME fibre (Supelco, Bellefonte, PA, USA). The vial septum was pierced with an SPME needle and the fibre was exposed to the headspace over analysed leaves for 10 min. After that time the fibre was retracted into the SPME needle and transferred to the GC injection port, where it was exposed to desorbed volatiles adsorbed on the SPME fibre. The desorption process lasted 10 min, and after that time the fibre was used to sample the next vial. Sampling was performed at 20°C. The gas chromatograph (GC) used for analyses was a Hewlett Packard HP6890 equipped with a split/splitless injector, a flame ionization detector, and a PLOT Al2O3 KCl (50 m × 0.32 mm × 8 µm) column (19091P-M15, Agilent Technologies, Palo Alto, CA, USA). Helium was used as a carrier gas at a constant pressure of 20 psi. To facilitate sample comparison, amounts of emitted ethylene was expressed in nl per g of FW. All samples were run in triplicate.

Assays of hydrogen peroxide

Hydrogen peroxide concentration was precisely determined spectrophotometrically using the titanium (Ti4+) method (Becana et al., 1986). The analytical curve, in relation to which the H2O2 content of tissue was calculated, was prepared for H2O2 in the concentration range 5–50 µm. Absorbance of the formed hydrogen superoxide complex with the titanium reagent was measured at λ = 508 nm.

Additionally, H2O2 was assayed using the cytochemical detection method by a colour reaction with 3,3′-diaminobenzidine (DAB), according to Thordal-Christensen et al. (1997). The appearance of a red-brown colour on the leaf indicated the presence of H2O2.

Antioxidant enzyme assays

Catalase activity (EC 1.11.1.6) was assayed as described by Dhindsa et al. (1981). The consumption of H2O2 was monitored at 240 nm (ɛ = 45.2 mm−1 cm−1). Ascorbate peroxidase activity (EC 1.11.1.11) was measured according to Nakano & Asada (1981) by monitoring the rate of ascorbate oxidation at 290 nm (ɛ = 2.8 mm−1 cm−1).

Determination of ascorbic acid

For the determination of ascorbic acid (AA), tissue was homogenized in 4 ml of 5% triacetic acid (TCA). A spectrophotometric assay, as described by Mukherjee & Choudhuri (1983), was used. Absorbance was measured at ɛ = 530 nm. The concentration of AA in tissue was calculated from the analytical curve prepared for the concentration range 1–12 µg ml−1 of AA.

Total antioxidative capacity

Total antioxidative capacity was measured using the ability of antioxidants contained in the extract to reduce the ABTS•+ cation-radical according to the method described by Re et al. (1999). The initial ABTS•+ solution was prepared by dissolving 19.5 mg ABTS in 7 ml of 0.1 m potassium phosphate buffer (pH 7.4) and 3.3 mg of potassium supersulphate. After thorough mixing, the solution was left in the dark for 12 h. Immediately after the assay, the initial ABTS•+ solution was diluted with 0.1 m potassium phosphate buffer (pH 7.4) to the absorbance value of 1.0 at ɛ = 414 nm.

In order to measure the total antioxidative capacity, leaf tissue (250 mg) was homogenized in 2 ml of 5% TCA, and next centrifuged at 15 000 g for 15 min. A total of 980 µl of diluted ABTS•+ was pipetted to a cuvette and absorbance (A0) was measured at ɛ = 414 nm, followed by the addition of 20 µl of extract. After 10 s, absorbance (A1) was measured precisely again. The total antioxidative capacity dependent on ‘fast antioxidants’ (i.e. AA or glutathione) was calculated as ΔAs = A1 – A0. The analytical curve was prepared by adding to the diluted ABTS•+ successively 5 µl portions of 0.01 mm Trolox and measuring successive drops of absorbance. The final result of total antioxidative capacity was expressed in mm Trolox g−1 FW.

Statistical analysis

All experiments included three independent experiments carried out in at least three replications. For each experiment, means of the obtained values were calculated along with standard deviations. The analysis of variance was performed and the least significant differences (LSDs) between means were determined using Tukey's test at the level of significance α = 0.05.

Results

NO burst and the wave of secondary NO generation

Electrochemical detection of NO using an NO-selective electrode appeared applicable to monitor the NO burst taking place in the early phase of the pathogen–plant interaction. Figure 1 shows the shape of the DPV curve recorded in the leaf tissue before (a) and after (b–h) the inoculation with conidial suspension. In a healthy leaf, a sigmoidal curve was recorded. Immediately after the inoculation, a peak appeared on the recorded curve at a potential of c. 0.8 V. The peak current (ip) was a measure of a temporal NO concentration released, and was strictly dependent on resistance to the necrotrophic pathogen. Figure 2 compares recorded NO signal-time traces for resistant and susceptible responses of pelargonium leaves to the pathogen. The observed NO burst was c. threefold higher in the resistant response, that is, the peak current reached 12 nA after 5 min of the challenge. Then the NO signal underwent some oscillations to reach the initial concentration after c. 90 min. In contrast to the resistant reaction, in the susceptible response no such strong NO burst was found (Fig. 2b), so the current NO signal reached c. 4.0 nA after 5 min and then slowly decreased to the basal concentration.

Figure 1.

Electrochemical detection of nitric oxide (NO) emission from resistant pelargonium (Pelargonium peltatum) leaf tissue treated with a conidial suspension of Botrytis cinerea. Differential pulse voltammogram recorded in resistant leaf tissue before (a) and after pathogen treatment (b–h). Voltammograms recorded 2, 15, 35, 50, 65, 70 and 85 min after inoculation, respectively. ip, peak current.

Figure 2.

Nitric oxide (NO) burst as a result of plant–necrotroph interaction. Change of NO peak current (ip) recorded in leaf tissue as a function of time after inoculation with a fungal suspension of pelargonium (Pelargonium peltatum) cultivars: (a) the resistant ‘Cascade’; (b) the susceptible ‘Shiva’.

Additionally, bio-imaging with fluorochrome DAF2-DA made it possible to obtain a 3D image and to assess the range of NO generation in the defence response of the plant to B. cinerea. The NO-specific fluorescent dye showed that the previously recorded very early phase (within a minute range) of NO burst was followed by the circular wave of secondary NO generation, precisely organized spatially (Fig. 3e,h,k). The epicentre of NO synthesis was initially found at the site of fungal inoculation and next the zone of induced NO generation was expanding, to disappear completely after 3 or 4 d. High concentrations of emitted NO in the place of fungus contact with the host tissue most probably made it possible to generate a sufficiently strong signal for effective defence, for example, by the stimulation of death of the targeted cell, seen in Fig. 3(d,g,j) in the form of point necroses. Cells attacked by the pathogen exhibited symptoms of active death, which was found in the TUNEL-positive test illustrating the programmed DNA fragmentation (Fig. 4d–h). In successive zones of NO generation, the cells distant from the epicentre of NO burst were TUNEL-negative (Fig. 4i).

Figure 3.

Bio-imaging of nitric oxide (NO) and hydrogen peroxide (H2O2) accumulation in a resistant response of pelargonium (Pelargonium peltatum) leaves to Botrytis cinerea (a–i). Changes were analysed on three successive days after inoculation. (d, g, j) Micronecrosis; (e, h, k) amplification of wave-like NO generation (DAF-2DA-positive, green); (f, i, l) low H2O2 production (DAB-positive, purple); (a, b, c) image of control leaf tissue. Bars, 500 µm (c); 100 µm (f, i, l); 40 µm (b, e, h, k).

Figure 4.

Visualization of nitric oxide (NO) in fungal hyphae (a–c) and nuclear DNA fragmentation (d–i) in resistant pelargonium (Pelargonium peltatum) after inoculation with Botrytis cinerea. (a, b) Strong NO generation (DAF-2DA-positive mycelium of B. cinerea; green) during colonization of a susceptible leaf section by the pathogen; (c) weak NO generation in hyphae of mycelium growing on a complete medium; (d–h) TUNEL-positive nuclei (red in light microscopy, indicative of irreversible DNA fragmentation and consequent cell death in the resistant leaf after inoculation with B. cinerea; arrows indicate micro- and macrospores of the pathogen); (i) TUNEL-negative nuclei of noninoculated resistant pelargonium cells. Bars: 200 µm (a); 40 µm (b); 20 µm (c); 3 µm (d, e); 5 µm (f–i).

In susceptible leaves, no secondary wave of NO generation was observed around the site of fungal inoculation and these targeted cells were also TUNEL-negative. Along with the development of the disease (3 d) around spots of grey mould, intensively green fluorescent (DAF-2DA) spots appeared on the surface of attacked tissue, which indicated a huge overproduction of NO (Fig. 5e,h,k). It can be assumed that initially NO synthesis came from the host and next, to a large extent, from the pathogen. We observed that both hyphae and conidia generated trace amounts of NO outside the plant organism, growing on a complete medium (Fig. 4c). Only in contact with the leaf tissue of the susceptible cultivar did the emission of NO from the pathogen increase considerably (Fig. 4a,b).

Figure 5.

Bio-imaging of nitric oxide (NO) and hydrogen peroxide (H2O2) accumulation in a susceptible response of pelargonium (Pelargonium peltatum) leaves to Botrytis cinerea. Changes were analysed on three successive days after inoculation. (a–c) Image of control leaf tissue; (d, g, j) development of grey mould spots; (e, h, k) overproduction of NO (DAF-2DA-positive, green); (f, i, l) strong H2O2 accumulation (DAB-positive, purple). Bars: 500 µm (c, f, i); 40 µm (b, e, h); 60 µm (k); 1000 µm (l).

NO and ethylene

Post-infection production of ethylene was dependent on pelargonium genotype (Fig. 6). In the susceptible cultivar, a rapid increase in the amount of ethylene (seven- and 14-fold, respectively) was found on days 3 and 4 after inoculation. By contrast, in the resistant cultivar, no significant changes were recorded in the level of this signal on successive days after inoculation.

Figure 6.

Effects of Botrytis cinerea and/or an altered concentration of nitric oxide (NO) on ethylene production in resistant and susceptible pelargonium (Pelargonium peltatum) leaves. *, significantly different from the control at P < 0.05.

Administration of NO (as SNP), preceding inoculation, caused a twofold decrease in ethylene production in the susceptible pelargonium leaves (4 d after inoculation). These changes were accompanied by an inhibition of disease development (Table 1). An effective reduction of disease spots was observed under the influence of SNP (50 and 100 µm) and GSNO (200 µm). By contrast, a higher concentration of SNP (500 µm) promoted the development of grey mould on leaves considerably. At the same time, we found that exogenous NO did not have any effect on fungal growth on the medium (data not presented).

Table 1.  The effect of nitric oxide (NO) donors (sodium nitroprusside (SNP) and S-nitrosoglutathione (GSNO)) on grey-mould symptoms in pelargonium (Pelargonium peltatum) leaves 4 d after inoculation by Botrytis cinerea
TreatmentArea of disease spots in comparison to the control (%)
‘Cascade’(resistant)‘Shiva’(susceptible)
  1. Inoculated control = 100%.

Control (H2O)100100
SNP 50 µm 12 24
SNP 100 µm 60 77
SNP 500 µm177195
GSNO 200 µm 61 78

NO and H2O2

The cytochemical analyses detected the presence of H2O2 in leaves of both pelargonium cultivars inoculated with B. cinerea (Figs 3, 5). In resistant leaves with small spots, H2O2 was present only in sites of necrosis formation (Fig. l, 3). In leaves of the susceptible cultivar, red-brown colouring, indicating a considerable accumulation of H2O2, covered the whole surface on which the conidial suspension was transferred, starting from the second day after inoculation (Fig. 5f,i,l). Along with the development of disease spots in the susceptible cultivar, the range of H2O2 synthesis also increased, which was especially evident at the boundary between healthy and diseased tissue.

Effect of NO on the redox system

In order to explain the influence of NO on the redox state of the challenged pelargonium leaf, the pharmacological approach with the application of NO donors was used. In turn, the dynamics of H2O2 accumulation under the influence of B. cinerea differed markedly in both cultivars (Fig. 7a,b). An early (2 h) and periodical increase in H2O2 concentration was found in the resistant cultivar, while a late (48 h) increase was observed in the response of the susceptible pelargonium cultivar. Pretreatment with NO donors, preceding inoculation, enhanced the effect of early generation of H2O2 in the resistant ‘Cascade’. Also in leaves of the susceptible ‘Shiva’ during the first 24 h, a transient increase was observed, followed by a decrease in H2O2 accumulation after treatment with exogenous NO and B. cinerea. A transient H2O2 increase was observed as a result of pretreatment with NO, correlated with a reversible inhibition of CAT and APX activities at the same phase after infection in both cultivars (Fig. 7c–f).

Figure 7.

Effects of Botrytis cinerea and an altered concentration of nitric oxide (NO) on hydrogen peroxide (H2O2) concentration (a, b), catalase (CAT) activity (c, d) and ascorbate peroxidase (APX) (e, f) activity in resistant and susceptible pelargonium (Pelargonium peltatum) leaves. Bars represent SE of means. *, significantly different from the control at P < 0.05. Open columns, control; hatched columns, inoculated; closed columns, sodium nitroprusside (SNP)/inoculated.

A key role in the protection of cells against the destructive action of ROS, induced in response to stress, is played by antioxidant compounds. In the analysed system, NO treatment of susceptible pelargonium caused an enhanced accumulation of AA (c. fourfold), especially on days 1 and 2 after inoculation, in contrast to the infected leaves that were not treated with NO (Fig. 8a,b). In turn, in the resistant pelargonium cultivar, as a result of the tissue interaction with the necrotroph, a rapid increase in AA was found at 18 h after inoculation, and the sequential treatment of leaves with NO and B. cinerea additionally increased the pool of this antioxidant (Fig. 8a).

Figure 8.

Effects of Botrytis cinerea and an altered concentration of nitric oxide (NO) on the antioxidant pool in resistant and susceptible pelargonium (Pelargonium peltatum) leaves. (a, b) Ascorbic acid (AA) content; (c, d) ‘fast antioxidant’ pool. Bars represent SE of means. *, significantly different from the control at P < 0.05. Open columns, control; hatched columns, inoculated; closed columns, sodium nitroprusside (SNP)/inoculated.

Also only in case of NO-treated resistant tissue was a strong increase in the amounts of ‘fast antioxidants’ recorded on all the analysed days after inoculation with B. cinerea, while the most intensive, c. fourfold, increase in relation to challenged leaves occurred very early, that is, starting from 2 h after inoculation (Fig. 8c,d).

Discussion

Results of this study demonstrate that NO plays a crucial role in the initiation of a fast defence response of pelargonium leaves to the necrotrophic pathogen. The model of NO generation depending on the genotype of the plant is shown in Fig. 9. The resistance response was accompanied by a strong NO burst, which probably stimulated a HR and was the source of the secondary wave of NO generation. The observed phenomenon resembled the effect of a stone thrown into water, as successive induced rings with wider ranges had weaker signal amplitudes. As a result of this induced generation, the NO signal gradually reached distant cells and, with a decreasing intensity, affected several metabolic defence response events, for example by a reversible increase in H2O2 concentration, correlated with the inhibition of CAT and APX activity, as well as ethylene synthesis suppression. Increasing the pool of antioxidants via NO promoted the maintenance of intracellular homeostasis, which confirms the beneficial impact of NO in the resistance of pelargonium to the necrotrophic pathogen. By contrast, in susceptible plants, apart from a much weaker NO burst and a lack of the wave of induced generation, an enhanced accumulation of H2O2 was found around the developing disease spots. When B. cinerea colonized plant tissues, it destroyed intracellular compartments of the host, which contributed to the uncontrolled NO overproduction. Additionally, in contact with the plant, the pathogen itself gained the capacity to generate considerable amounts of NO, which caused a massive oxidative and nitrosative stress, and eventually NO enhanced the destruction of the tissue attacked by the necrotroph.

Figure 9.

Model of early nitric oxide (NO) burst action as a source of secondary wave-like NO generation and its effect on defence responses of pelargonium (Pelargonium peltatum) plants resistant to the necrotrophic Botrytis cinerea. HR, hypersensitive response; CAT, catalase; APX, ascorbate peroxidase; RSNO, nitrosothiols; SH, sulfhydryl groups.

In studies on NO, after the first fascination and acclamation of NO as a panacea, it needs to be asked whether, and for which organism, NO generation is advantageous in the plant–pathogen system. Most frequently this cross-talk is considered from the point of view of the plant. We do not take into consideration the fact that the pathogenic microorganism has also improved its offensive strategy evolutionarily, which may include the ability to generate NO.

So far, there are few reports on the mechanism and site of NO synthesis in pathogenic microorganisms challenging the plant. Wang & Higgins (2005) recently reported on the accumulation of NO in conidia and germinating hyphae of Colletotrichum coccodes, whose intensity and location changed depending on the developmental stage of the fungus. Also Conrath et al. (2004), using mass spectrometry (MIMS/RIMS), showed the ability of pathogenic fungi, that is, Phytium, Botrytis and Fusarium spp., induced by nitrites, to produce NO. In our experiments, the factor inducing NO generation by the necrotrophic fungus B. cinerea was the tissue of the susceptible pelargonium genotype.

In turn, van Baarlen et al. (2004) found the accumulation of NO and H2O2 to be correlated with disease development only in the compatible lily–Botrytis elliptica interaction.

A comparison of these two systems, lily–Botrytis spp. (van Baarlen et al., 2004) and pelargonium–Botrytis spp., in terms of NO generation is difficult. Mechanisms of the genetic makeup of the host and gene-for-gene dependencies in relation to necrotrophs are unknown. In the case of lily, in both the incompatible and compatible interactions, the early NO burst was not analysed. However, generally both in the lily–Botrytis elliptica compatible interaction and in our experiments on the pelargonium–B. cinerea compatible interaction, disease development was accompanied by the accumulation of NO originating from both the plant and the pathogen.

Some authors suggested that the early NO burst is a specific marker for R–avr interactions and immediately provokes the generation of H2O2 (Bennett et al., 2005; Mur et al., 2005). In soy cell suspensions, the accumulation of NO elicited by Pseudomonas syringae coincides with the generation of H2O2 (Delledonne et al., 1998), and might even precede it, which was observed in epidermal cells of tobacco, treated with a fungal elicitor, cryptogein (Foissner et al., 2000). The interaction of NO and H2O2 in the initial defence of tomato against C. coccodes via cell wall modification at sites of appressoria formation was recently noted by Wang & Higgins (2006). A huge and transient NO burst was observed in barley epidermal cells attacked by the powdery mildew fungus (Blumeria graminis) and the burst preceded their HR-associated collapse (Prats et al., 2005).

Our results show that the NO burst, although genetically conditioned, was not restricted only to the R–avr interaction. The attack of the necrotroph in a resistant pelargonium cultivar generated a sufficiently strong NO burst to induce a periodical increase in H2O2 and triggered TUNEL-positive programmed cell death (PCD) of targeted cells. The following wave of secondary NO generation was able to fine-tune resistance responses, but failed to trigger the death of surrounding cells (TUNEL-negative). Since B. cinerea can utilize dead tissue, noncell-death-associated resistance is favourable for the host.

Studies conducted on plants attacked by avirulent biotrophic microorganisms or their elicitors indicated a joint participation of NO and ROS, and especially H2O2. It may be assumed that between the aforementioned free radicals there are complex relations, dependent on quantitative ratios and physiological conditions of the plant, which determines their effectiveness in the stimulation of PCD and defence reactions, for example via intracellular signals (Foissner et al., 2000; Delledonne et al., 2001, 2002).

Reactive oxygen species alone are not always sufficient to cause PCD, and additional factors, such as NO, are required. According to Delledonne et al. (2001, 2002), NO and H2O2 may function synergistically, inducing a hypersensitive cell death. A lowering of APX activity, with the simultaneous generation of NO and H2O2 during the induction of PCD, was also observed by de Pinto et al. (2002). Moreover, studies conducted using a transgenic line of Arabidopsis thaliana with an overexpression of the thylacoidal form of APX showed that H2O2 and NO participated in inducing cell death by oxidative stress (Murgia et al., 2004). According to Mittler et al. (1999) and Bestwick et al. (2001), a reversible inhibition of both APX and CAT seems necessary in cells subjected to HRs.

Similarly, in nonR-gene resistance of pelargonium leaves, transient suppression of APX and CAT via NO was observed in both genotypes. A decrease in activity in the case of both CAT and APX may be explained by NO binding to the haem centres of these enzymes (Clark et al., 2000). In pharmacological approaches using NO donors, the duration of this inhibition is combined with the time of donor compound decomposition, and is usually limited to the first 24 h (Floryszak-Wieczorek et al., 2006). In relation to the aforementioned NO burst and the secondary wave of NO generation found in planta in the resistant pelargonium, it may be postulated that such a situation of reversible inhibition of CAT and APX via NO enables the transient amplification and gradual decay of H2O2 cooperation with other defence responses.

Botrytis cinerea, like most other necrotrophic pathogens, attacks senescent plant organs, inducing a burst of ethylene, which promotes cell death and is therefore beneficial for the pathogen (Thomma et al., 2001).

A decrease in ethylene production and the development of disease symptoms via NO in susceptible pelargonium, as well as the absence of ethylene synthesis in resistant leaves, showed the opposite role of both signals in plant resistance to B. cinerea.

Evidence of the interplay between NO and ethylene in the maturation and senescence of plant tissues suggests an antagonistic effect of both gases during these developmental stages as well (Leshem et al., 1998). On the other hand, the possibility that cross-talk between NO and ethylene plays a role also in disease suppression is intriguing, because ethylene is required for the timing of the onset of senescence, and therefore may be involved in the link between pathogenesis and senescence-dependent PCD.

Thus it can be concluded that an efficient response of the plant to the necrotroph, including also B. cinerea, should consist in the inactivation of ROS and the stimulation of the antioxidant system responsible for the control of the redox state in the cell. Such an approach is confirmed in practice, as treatment of crop plants with various antioxidants, followed by inoculation with B. cinerea, resulted in a markedly inhibited development of grey mould spots (Elad, 1992).

In the resistant pelargonium, both the concentration of AA and the pool of other ‘fast antioxidants’ strongly increased soon after the challenge, while NO enhanced the accumulation of these compounds independently of the genetic makeup of the plant.

Moreover, in animal cells, ascorbates may play a significant role in the immediate reaction of inactivation of the peroxynitrite ion (Arteel et al., 1999). Taking into consideration a high concentration of AA found in the cell, it cannot be excluded that this antioxidant participates in the mechanism of ONOO inactivation.

Based on our results, we can conclude that an early NO burst is a signal for the induction of nonspecific resistance of pelargonium to a necrotrophic pathogen, similarly to the R-avr-mediated resistance. Then the wave of secondary NO emission provokes noncell-death-associated defence, following the rule that the concentration of NO is linked to its action. A lack of such metabolic regulation via NO in the susceptible genotype, triggers the offensive strategy by B. cinerea, resulting in the overproduction of NO together with a faster colonization by the pathogen and a stronger infection.

So far the role of NO in disease resistance, as well as in human, animal and plant tissues, was mainly analysed from the viewpoint of the host organism. Little is known about NO generation by the pathogenic microorganism. However, it is suspected that NO may also be an effective weapon used by the pathogens. In endothelial cells, pathogenic microbes and viruses may induce S-nitrosylation of membrane receptors to facilitate cellular entry (Wang et al., 2006). There is evidence that NO generated by microorganisms may be responsible for the activation/deactivation of phytotoxins. In Streptomyces spp. nitric oxide, synthesized with the participation of NOS, nitrates taxtomine A, thus determining the phytotoxicity of this compound (Kers et al., 2004; Wach et al., 2005). Moreover, nitration of lipopeptide arylomycins generated by Streptomyces sp. Tü 6075 enhances the antibacterial activity (Schimana et al., 2002), which may be significant in the infestation of new ecological niches by this bacteria. NO has been shown to activate various antioxidant genes in Escherichia coli and Bacillus subtilis to protect microorganism cells against oxidative and nitrosative stress (Gusarov & Nudler, 2005).

Evidence for the presence and role of NO in the fungus is limited and it is not known to what degree resistance via NO in pelargonium is specific or universal for other plant–necrotroph systems.

Acknowledgements

This work was supported by the Polish Committee for Scientific Research (KBN, grant no. 2 P06A 016 27).

Ancillary