Induced resistance to Verticillium longisporum in Brassica napus by β-aminobutyric acid

Authors

  • A. Kamble,

    Corresponding author
    1. Plant Pathology and Crop Protection Division, Department of Crop Sciences, Faculty of Agriculture, Georg-August University Göttingen, Grisebachstraße 6, D-37077 Göttingen, Germany
    2. Department of Botany, University of Pune, Pune 411007, Maharashtra, India
      E-mail: kambleavinash2000@gmail.com
    Search for more papers by this author
  • B. Koopmann,

    1. Plant Pathology and Crop Protection Division, Department of Crop Sciences, Faculty of Agriculture, Georg-August University Göttingen, Grisebachstraße 6, D-37077 Göttingen, Germany
    Search for more papers by this author
  • A. von Tiedemann

    1. Plant Pathology and Crop Protection Division, Department of Crop Sciences, Faculty of Agriculture, Georg-August University Göttingen, Grisebachstraße 6, D-37077 Göttingen, Germany
    Search for more papers by this author

E-mail: kambleavinash2000@gmail.com

Abstract

A preinoculative soil drench application of 0·5 mm β-aminobutyric acid (BABA) significantly inhibited colonization of oilseed rape (Brassica napus, susceptible cultivar Falcon) by Verticillium longisporum and also prevented plant stunting caused by the pathogen. To better understand the defence responses induced by BABA, the presence of occlusions in the plant hypocotyl, levels of salicylic acid (SA) and hydrogen peroxide (H2O2), phenylalanine ammonia lyase (PAL) activity and expression of PR-1 and PDF1.2 genes were examined. Transverse sections through the hypocotyl region of BABA-treated plants showed clear vessels surrounded by phenol-storing cells, in contrast to numerous obstructed vessels in water-treated plants, in response to the pathogen. A significant increase in SA levels was observed in the hypocotyls of both water- and BABA-treated plants in response to the pathogen; however, SA levels were unrelated to disease resistance. The level of H2O2 decreased in both treatments in response to the pathogen. A significant increase in PAL activity was observed in hypocotyl tissues of BABA-treated plants. The expression patterns of PR-1 and PDF1.2 were similar in the two treatments in response to the pathogen, indicating no involvement of these genes in resistance. The results indicate a similar organ specificity of the plant hypocotyl for chemically induced internal resistance as for genotype-related resistance, two phenomena which, however, are based on contrasting cytological responses in the vascular tissues. Nonetheless, evidence is provided that the activity of the phenylpropanoid pathway plays a crucial role in both types of resistance.

Introduction

Plants have developed a wealth of defence strategies to protect themselves against the wide range of pathogens and pests that have co-evolved. Defence responses are expressed through complex signalling networks involving signalling compounds such as reactive oxygen species (ROS), salicylic acid (SA), jasmonate (JA), ethylene (ET) and abscisic acid (ABA) (Thomma et al., 1998; Mauch-Mani & Mauch, 2005). These signalling intermediates regulate the expression of pathogenesis-related (PR) genes that encode for proteins playing either a role in the structural and physiological manifestation of defence through cell wall strengthening, lignification, phytoalexin synthesis and expression of localized cell death or have direct antifungal properties.

In addition to basal resistance, plants are capable of developing an induced resistance (IR) that is a physiological state of enhanced defensive capacity elicited by specific environmental stimuli, by which plants’ innate defences are potentiated against subsequent biotic challenge. The best known systems of IR are systemic acquired resistance (SAR; Durrant & Dong, 2004) and induced systemic resistance (ISR; Pieterse et al., 1998).

In addition to these two types, treatment with a non-protein amino acid, BABA (β-aminobutyric acid), may induce resistance in a number of plant species against a wide range of pathogens (Cohen, 2002). However, the mode of action of BABA in inducing resistance remains unresolved. In some plant–pathogen systems, BABA was reported to induce resistance via the SA signalling pathway (Zimmerli et al., 2001). In some other plant–pathogen systems, BABA IR was shown to be expressed through an SA-independent pathway (Zimmerli et al., 2000). There is also some evidence of the involvement of JA in BABA IR because the inhibition of lipoxygenase (LOX) activity, which causes JA deficiency, was seen to reduce BABA IR (Hamiduzzaman et al., 2005). Finally, BABA has also been reported to induce resistance through ABA-primed callose deposition in the ArabidopsisAlternaria interaction, a resistance reaction which was expressed even in mutants defective in SA, JA or ET signalling (Ton & Mauch-Mani, 2004).

In potato plants, BABA-induced resistance against Phytophthora infestans is through early activation of the host defence genes encoding for anionic peroxidase, PR-14, proteins involved in metabolism of phytoalexins, and proteins related to jasmonic acid- and salicylic acid-dependent signalling pathways as revealed by cDNA-AFLP analysis (Li et al., 2009). Some more recent studies on mechanisms of BABA IR revealed a rapid BABA-primed encasement with callose of the primary infection structures of the pathogen and a fast accumulation of H2O2 in the penetrated epidermal host cells of lettuce against Bremia lactucae infection (Cohen et al., 2010). BABA IR in grapevine infected by downy mildew was associated with induction of phytoalexin, pterostilbene (Slaughter et al., 2008) and priming of an NADPH oxidase dependent ROS production (Dubreuil-Maurizi et al., 2010).

Recently, it has been demonstrated that BABA enhances the activation of defence responses to pathogen encounter through priming, a phenomenon that brings plants to an alarmed or primed state of defence (Conrath et al., 2006). The mechanism of priming is poorly understood. It has been hypothesized that priming leads to an increase in the level of signalling components that are involved in basal resistance (Conrath et al., 2006).

Oilseed rape (Brassica napus ssp. oleifera) is the most important oilseed crop in Europe and currently its cultivation is under threat because of a novel disease, caused by the soilborne vascular pathogen Verticillium longisporum (ex. V. dahliae var. longisporum). Control of V. longisporum is limited due to the unavailability of fungicides and resistant varieties (Eynck et al., 2009). The development of an alternative control strategy such as use of chemical resistance inducers may thus represent an option for control of verticillium disease within an integrated crop protection system.

Recently, some characteristics of the genotype-related defence responses in B. napus against V. longisporum have been partly elucidated (Eynck et al., 2009), while resistance responses during induced resistance are completely unexplored. Therefore the objective of this study was to explore the potential of induction of resistance in oilseed rape with a known chemical inducer, BABA, and to compare induced defence responses with genotype-related resistance reported previously (Eynck et al., 2009). To the authors’ knowledge, this is the first study on the effectiveness and possible mechanism of BABA induced resistance in brassicas against the vascular pathogen V. longisporum. First, the effective concentration of BABA was estimated based on the alleviating effect on disease severity in terms of stunting caused by the pathogen, followed by attempts to identify some of the components of induced resistance. Ramification of the pathogen was followed by using quantitative real-time PCR. Further parameters included the role of physical barriers in vascular tissues, endogenous H2O2 and SA levels, PAL enzyme activity and expression of defence-related genes, PR-1 and PDF1.2, as markers for the involvement of SA- and JA-mediated signalling pathways, respectively.

Materials and methods

Plant material

Seeds of B. napus cv. Falcon (winter type), which is susceptible to V. longisporum, were supplied by Norddeutsche Pflanzenzucht Hans-Georg Lembke KG (NPZ). One or two seeds were sown per pot (5 × 5 × 5 cm) in a multi-pot tray (50 × 30 × 5 cm) filled with 4 cm of a soil:sand mixture (3:1). The trays were placed in a greenhouse with a 16 h photoperiod at 20 ± 2°C. After germination, the seedlings were thinned to one plant per pot. After 3 weeks, the plants were used for further experiments.

Fungal cultures and inoculation procedure

Verticillium longisporum isolate VL43 obtained from a diseased B. napus plant was used for inoculations. Glycerol stocks of conidial suspensions (1–3 × 106 conidia mL−1 in Czapek-Dox medium supplemented with 25% glycerol) stored at −80°C were used to initiate fresh cultures. One millilitre of suspension was inoculated in 250 mL of potato dextrose broth (PDB) and incubated for 1 week at 23°C on a rotary shaker at 80 rpm. The resulting suspension was filtered through sterile gauze to remove mycelial fragments. Spore concentration was determined with a haemocytometer and diluted to 1 × 106 spores mL−1.

Three-week-old plants of B. napus were root dip inoculated with a V. longisporum spore suspension and transplanted into single pots as described previously (Eynck et al., 2009).

Chemical treatments

BABA was obtained from Sigma-Aldrich Chemical Company. Aqueous solutions of BABA were applied as soil drench. Chemical pretreatment was carried out 24 h before inoculating 3-week-old plants with the pathogen. In an initial experiment, varying concentrations of BABA (0·5, 1 and 5 mm) were used. Based on the efficacy in terms of reducing the stunting effect caused by V. longisporum infection and less phytotoxicity, 0·5 mm BABA was used in subsequent experiments. Controls were treated with water.

Disease assessment

Plants were scored 14, 21 and 28 days post-inoculation (dpi) for visible disease symptoms. The individual plant height (shoot length above the soil surface) was recorded as a parameter to assess disease severity, as this was found to be a sensitive and reliable disease parameter in pot experiments with B. napus and V. longisporum.

Measurement of fungal colonization with qPCR

Samples for qPCR analysis of fungal colonization were collected 14, 21 and 28 dpi from plant parts such as hypocotyl, middle stem and shoot. Plant tissue samples from four plants were pooled and constituted one sample in the analysis. Four such pooled samples were analysed per treatment. Fungal DNA was extracted and quantified by real-time PCR as described by Eynck et al. (2007).

Microscopy and histochemistry

The hypocotyl region of pretreated plants was collected at 28 dpi and samples were fixed in FAA (5% formaldehyde + 50% ethanol + 10% acetic acid in water) for 24 h and preserved in 70% ethanol. Transverse sections of hypocotyls (20–50 μm) were produced using a Leica VT 1000M microtome. Transverse sections of hypocotyls were stained with phloroglucinol–ethanol solution (3 g phloroglucinol in 100 mL 92% ethanol) for 0·5–2 min followed by a transfer to 20% HCl solution for about 1 min. Sections were mounted in 20% HCl and observed under a bright-field microscope (Leica DM-RB) at ×400 magnification. The percentage of vessels filled with occlusions was recorded and expressed as a percentage with respect to the total number of vessels under the field of vision (data not shown).

Endogenous levels of H2O2

Hypocotyl samples were collected from experimental plants at 7 dpi and 50 mg of tissue from a pooled sample ground in liquid nitrogen was taken and suspended in 500 μL of 50 mm HEPES buffer. Six replicates of samples pooled from five plants each were prepared for analysis. After centrifugation at 10 621 g for 6 min, 100 μL supernatant was added to a mixture containing 800 μL HEPES buffer, 1 μL H2SO4 and 100 μL potassium titanium oxalate (2·5% in 20% H2SO4). Absorbance was recorded at 410 nm (Sellers, 1980). A standard calibration curve was obtained using known amounts of H2O2.

Endogenous SA levels

Hypocotyl and leaf samples were harvested from experimental plants at 7 dpi. Pooled samples of at least five plants were merged into one sample. Six such samples were taken for each treatment. SA was extracted and analysed using a Merck-Hitachi HPLC system (Merck KGaA) as described previously (Kamble & Bhargava, 2007).

Phenylalanine ammonia-lyase (PAL) activity

One gram of plant hypocotyl tissue was harvested from experimental plants after 3, 7 and 14 dpi. Pooled samples of at least five plants were merged into one sample and three such samples were taken from each treatment. One gram of plant tissue per sample was homogenized in 3 mL of 5 mm Tris-HCl buffer (pH 8·5), containing 1·4 mm 2-mercaptoethanol. The extract was then centrifuged at 10 621 g for 15 min at 4°C. The supernatant was used as the enzyme source. Activity of PAL was determined as the rate of conversion of l-phenylalanine to trans-cinnamic acid following Dickerson et al. (1984). A 3-mL reaction mixture containing 50 mm Tris-HCl buffer, pH 8·3, 0·3 mL 1 mm l-phenylalanine, 0·9 mL distilled water and 0·3 mL enzyme extract was incubated in a waterbath at 30°C for 60 min. The reaction was terminated by adding 1 mL 2 m HCl. Finally, the reaction mixture was vortexed after addition of 3 mL toluene. After centrifugation at 106 g for 5 min, the upper toluene layer was measured at 290 nm against pure toluene as blank. A standard curve for trans-cinnamic acid at various concentrations (0–150 μm) in 50 mm Tris-HCl buffer (pH 8·3) was obtained at 290 nm. The standard curve was used to calculate the amount of trans-cinnamic acid formed in the samples. Enzyme activity was expressed as change in cinnamic acid in μm min−1 g−1 FW.

Differential expression of defence genes

Root and leaf samples were collected at 3, 7 and 14 dpi. Total RNA was isolated using TRI-Reagent (Molecular Research Center) according to the manufacturer’s instructions. RNA was checked for DNA contamination by performing PCR using total RNA as template. One microgram of total RNA was reverse transcribed to cDNA using a QuantiTect reverse transcription kit (QIAGEN) as per the manufacturer’s instructions. Two microlitres of the cDNA synthesized was amplified by PCR using specific primers designed for PR-1 and PDF1.2. 25S rRNA amplification was also carried out using 2 μL of the same cDNA pool.

Primers were designed from cDNA sequences of Brassica juncea (PR-1: GenBank accession no. DQ359128) and B. napus (PDF1.2: GenBank accession no. U59459; 25S rRNA: GenBank accession no. D10840.1) available at NCBI (http://www.ncbi.nlm.nih.gov/). Sequences of the primers were as follows: PR-1 (FP): 5′-AGTCACTAACTGTTCTCGAC-3′ and (RP): 5′-CGATTACACGTCCACATAATT-3′; PDF1.2 (FP): 5′-GAAGCACCAACAATGGTG-3′ and (RP): 5′-GTGACACAGACTTATTGAACG-3′; 25S rRNA (FP): 5′-GCCGACCCTGATCTTCTG-3′ and (RP): 5′-GATGGTTCGATTAGTCTTTCGC-3′.

Fifty micolitres of PCR reaction mixture consisted of 2 U of Taq DNA polymerase (Fermentas), 1 × Taq polymerase buffer, 0·3 mm of each dNTP and 80 pmol of each primer. PCR was carried out using a T-Gradient thermocycler (Biometra) under the following conditions: initial denaturation at 94°C for 5 min, 30 cycles of 1 min at 94°C, 1·5 min at 50°C and 2 min at 72°C, followed by a final extension step for 10 min at 72°C.

PCR products were separated on 2% agarose gels, stained with ethidium bromide and observed by a gel documentation system (Bio-Rad). The PCR products were sequenced to confirm the respective genes (data not shown). Experiments were carried out twice.

Statistical analysis

Data are represented as means ± SD. Statistical significance of differences between means was determined with Student’s t-test to compare the disease severity (Fig. 1) and amount of V. longisporum DNA (Fig. 2) in water-treated and BABA-treated plants. Significant differences of mean values ( 0·05) were marked with an asterisk (*). In addition, the differences in levels of H2O2 (Fig. 3), SA (Fig. 4a,b) and PAL activity (Fig. 5a,b) among the treatments was analysed by one-way anova with spss v. 9.0 and Fisher’s least significant test (LSD) was done at  0·05.

Figure 1.

 Effect of different concentrations of BABA pretreatments on height reduction caused by Verticillium longisporum (Vl) on Brassica napus cv. Falcon. (W+M) water-pretreated and mock-inoculated; (W+Vl) water-pretreated followed by Vl inoculation; (B-5, B-1, B-0·5 + Vl) BABA pretreatment of 5, 1 and 0·5 mm concentrations, respectively, followed by Vl inoculation. Data points represent mean heights of 20 plants ±SD at 14 (inline image) and 21 (inline image) days post-inoculation (dpi) in a single experiment. The experiment was repeated three times with similar results. Asterisks denote significant differences from the water-pretreated and mock-inoculated plants (W+M), (t-test,  0·05).

Figure 2.

 Extent and progress of Verticillium longisporum colonization in Brassica napus cv. Falcon plants pretreated with BABA (0·5 mm). Each bar represents total amounts of V. longisporum (Vl) DNA detected in hypocotyls, middle stems and shoots at 14, 21 and 28 days post-inoculation. Data points represent means of four samples each consisting of tissues pooled from four plants. Asterisks denote significant differences from the water-pretreated control (W) at given time points (t-test,  0·05) for the respective tissue. The error bars represent standard deviations (±SD) of means. The experiment was repeated twice with similar results.

Figure 3.

 Endogenous levels of H2O2 in response to Verticillium longisporum inoculation from hypocotyls of Brassica napus cv. Falcon plants pretreated with water (W) and 0·5 mm BABA at 7 days post-inoculation. W+M and B+M denote water-pretreated and BABA-pretreated mock-inoculated plants, respectively. Vl denotes inoculated plants. Data points represent means of six independent samples, each representing the plant tissues pooled from five plants. Error bars represent standard deviations (±SD) of means. Means with different letters are significantly different (LSD test;  0·05). The experiment was repeated twice with similar results.

Figure 4.

 Endogenous levels of free (inline image) and conjugated (inline image) salicylic acid (SA) in response to Verticillium longisporum inoculation of Brassica napus cv. Falcon (a) hypocotyls and (b) leaves, at 7 days post-inoculation. Data points represent means of total SA levels of six independent samples, each representing plant tissues pooled from five plants. Treatments are as described in Figure 3. Error bars represent standard deviations (±SD) of means. Means of total SA with different letters are significantly different (LSD test;  0·05). The experiment was repeated twice with similar results.

Figure 5.

 PAL activity in Brassica napus cv. Falcon (a) hypocotyls and (b) leaves, at 3, 7 and 14 days post-inoculation (dpi) with Verticillium longisporum. Data points represent means of three independent samples, each representing plant tissues pooled from five plants. Treatments are as described in Figure 3. Vertical lines represent standard deviations. Means with different letters are significantly different (LSD test;  0·05). The experiment was repeated twice with similar results.

Results

Effect of BABA pretreatments on disease symptoms and pathogen spread

A significant reduction in plant height was observed at 21 dpi with V. longisporum inoculation as compared to the non-inoculated plants (Fig. 1). Plants pretreated with 1 mm and 0·5 mm BABA 24 h prior to V. longisporum spore inoculation grew significantly higher compared to plants subjected to water pretreatment (Fig. 1). BABA pretreatment alone did not significantly change the height of the plants compared to water-treated non-inoculated plants (data not shown).

On the basis of this preliminary experiment, the concentration of BABA used in further experiments was adjusted to 0·5 mm, a concentration at which plants failed to express any significant disease symptoms (stunting) arising from V. longisporum infection.

Quantitative real-time PCR analysis using fungal DNA-specific primers and DNA isolated from the hypocotyl, middle stem and shoot of the plant, showed that a similar quantity of V. longisporum DNA was present in hypocotyl and middle stem parts of B. napus plants pretreated with BABA or water, 14 dpi. No fungal DNA was detected in the shoot samples, indicating that the pathogen ascent in the plant was only up to the middle stem until this time point (Fig. 2).

At 21 dpi, a 5-fold increase in V. longisporum DNA was observed in the water-pretreated plants, whereas in BABA-pretreated plants the amount of fungal DNA increased only about 3-fold (Fig. 2). At this time point, the pathogen had spread up to the shoot, though the amount of fungal DNA found in the shoot of BABA-pretreated plants was significantly lower than in water-pretreated plants (Fig. 2).

A significant difference in the total amount of V. longisporum DNA was also found at 28 dpi between plants receiving the different pretreatments. In water-pretreated plants the proportions of fungal DNA between 21 and 28 dpi were similar, although the fungal DNA content in the shoot increased at the expense of lower plant parts (Fig. 2). In BABA-pretreated plants, the amount of V. longisporum DNA decreased at 28 dpi compared to 21 dpi and the amount of fungal DNA in the hypocotyl region was twice that observed in the middle stem. Hence BABA pretreatment was effective in restricting both the growth and spread of the pathogen. Water-pretreated plants showed a higher proportion of fungal DNA in the middle stem compared to the hypocotyl, indicating further progression of the pathogen within the plant (Fig. 2).

Endogenous levels of H2O2

Levels of H2O2 were determined at 7 dpi in the hypocotyl of plants pretreated with BABA and compared to water-pretreated plants. About 50% reduction in H2O2 levels occurred in water-pretreated plants in response to pathogen inoculation compared to water-pretreated and mock-inoculated plants (Fig. 3). A decrease in H2O2 accumulation was also observed upon BABA pretreatment in pathogen-inoculated plants compared to the mock-inoculated plants.

Endogenous levels of free and conjugated SA in hypocotyls

As SA can exist in free or conjugated forms, levels of both forms of SA were determined in plants subjected to V. longisporum infection after BABA pretreatment and compared with water-pretreated plants. A significant increase in total SA levels was observed in the hypocotyls of inoculated, water- and BABA-pretreated plants compared to non-inoculated plants, regardless of pretreatments (Fig. 4a).

In leaves from inoculated plants, SA levels were only increased upon water pretreatment but not following BABA application (Fig. 4b). Thus, BABA pretreatments consistently did not alter SA levels without the presence of the pathogen and the pathogen was required to be present in the sampled tissue to induce SA accumulation.

PAL activity

PAL activity was determined in the hypocotyl tissues of plants pretreated with BABA and compared to water-pretreated plants at 3, 7 and 14 dpi. At 3 dpi, the level of PAL activity remained the same in all treatments. A significant rise in PAL activity was observed at 7 dpi only in plants pretreated with BABA (B+Vl), which further significantly increased by 120% at 14 dpi compared to water-pretreated, mock-inoculated plants (W+M). Increase in PAL activity, but to a moderate extent, was also observed in inoculated, water-pretreated plants (W+Vl) at 14 dpi (Fig. 5a).

In leaves, pathogen application induced a significant increase in the activity of PAL in both water- and BABA-treated plants compared to mock-inoculated plants at 14 dpi (Fig. 5b).

Microscopy and histochemical analysis

Transverse sections of hypocotyls were taken at 28 dpi to observe any histological changes arising in the vessels in response to fungal colonization. Comparisons were made between plants pretreated with BABA or water. Non-inoculated plants that were subjected to treatments with water and BABA showed xylem vessels that were clear (Fig. 6). In water-pretreated plants colonized with V. longisporum, about 30–80% of the vessels showed occlusions (Fig. 6e,h). In the hypocotyls from BABA-pretreated inoculated plants only 1–2% of vessels showed occlusions (Fig. 6f,g). In contrast, hypocotyl tissues from these plants revealed a strong accumulation of phenols in cells accompanying the vessels, indicating the formation of phenol-storing cells (Fig 6f; see arrow).

Figure 6.

 Transverse sections through the hypocotyl region of Brassica napus plants (cv. Falcon), mock-inoculated or inoculated with Verticillium longisporum (28 days post-inoculation). Plants pretreated with (a, b) water, (c, d) BABA; in both mock-inoculated treatments xylem vessels are completely lacking any physical barriers; (e, h) pretreated with water and inoculated with V. longisporum; numerous obstructed vessels occur; (f, g) pretreated with BABA and inoculated. V denotes vessels. Arrows show clear vessels surrounded by phenol-storing cells. Sections were stained with a 2% alcoholic solution of phloroglucinol-HCl. Scale bar = 100 μm in (a), (c), (e), (g) and 50 μm in (b), (d), (f), (h).

Differential expression of defence genes

At 3 dpi, no expression of PR-1 was detected in roots of plants pretreated with BABA or water, whereas a differential expression of PR-1 was detected at 7 and 14 dpi (Fig. 7). At 7 dpi, expression of PR-1 was lacking in water- and BABA-treated plants but expression was recorded in response to pathogen inoculation in both pretreatments. A similar pattern of PR-1 expression was recorded in response to pathogen inoculation at 14 dpi.

Figure 7.

 Expression of PDF1.2 and PR-1 genes in roots from Brassica napus plants mock- or root dip-inoculated with Verticillium longisporum. Total RNA was extracted 3, 7 and 14 days post-inoculation (dpi). Treatments are as described in Figure 3. The size of amplified products of PR1 and PDF1.2 was 536 bp and 279 bp, respectively. 25S rRNA was used as internal standard. The figure is representative of two independent experiments.

Similarly, at 3 dpi, no expression of PDF1.2 was detected in samples from roots of plants pretreated with BABA and water (Fig. 7). However, at 7 dpi, expression of PDF1.2 was detected in water- and BABA-treatments with or without pathogen inoculation and the expression was higher in plants after pathogen inoculation. At 14 dpi, expression of PDF1.2 was lower than at 7 dpi in all treatments. Expression of both PR-1 and PDF1.2 was not detected in leaf tissue of water- and BABA-pretreated plants with or without pathogen inoculation at 3, 7 and 14 dpi (results not shown), indicating that expression of PR-1 and PDF1.2 was locally and not systemically induced.

Discussion

Susceptibility of B. napus plants to verticillium disease was assessed by observing plant height reduction, which was conspicuous at 14 dpi. When plants were pretreated with BABA, they did not show stunting, indicating that BABA pretreatment improved the ability of B. napus to tolerate V. longisporum infection. Whether this tolerance arose from restriction of fungal colonization within the plant was investigated using real-time PCR for quantifying the amount of fungal DNA in different parts of the plant at different time points. At 14 dpi, similar quantities of V. longisporum DNA were detected in the hypocotyl and middle stem of water- and BABA-pretreated plants. No pathogen DNA was detected in the shoots. However at 21 and 28 dpi, spread of V. longisporum appeared to be restricted due to BABA-pretreatment compared to the water-pretreated plants. The amount of fungal DNA found in the shoots of BABA-pretreated plants was significantly lower than in water-pretreated plants.

The role of BABA in restricting spread of vascular pathogens has also been shown in asparagus/Fusarium oxysporum (He & Wolyn, 2005) and cotton/V. dahliae interactions (Li et al., 1996). BABA pretreatments therefore appear to induce resistance in B. napus against V. longisporum, most likely through activation of defence responses. In interactions of tomato with Fusarium oxysporum f. sp. radicis-lycopersici, benzothiadiazole (BTH)-induced resistance was expressed at the root epidermis and outer cortex level by the formation of callose-enriched wall appositions at the site of penetration of the pathogen (Benhamou & Belanger, 1998). However, in B. napus, V. longisporum appeared to penetrate the root of both water- and BABA-pretreated plants with equal efficiency, because similar amounts of fungal DNA were detected in the hypocotyls of plants having received the different pretreatments. Expression of induced resistance due to BABA appeared to be restricted to the hypocotyls, which led to suppression of further systemic hyphal growth into the upper parts of the plant. This feature of BABA-induced resistance is similar to the internal resistance detected in a resistant genotype of B. napus against V. longisporum reported previously (Eynck et al., 2009).

The nature of resistance induced by BABA pretreatments that resulted in slowing down the rate of pathogen spread into the plant shoot was further studied. Oxidative burst is an early defence response in plants and plays a role in expression of the hypersensitive reaction (HR). Programmed cell death accompanying HR is important in controlling the spread of biotrophic and necrotrophic pathogens in the host (Glazebrook, 2005). Increased generation of ROS by the pathogen has been related to the pathogenicity of necrotrophic pathogens (von Tiedemann, 1997), whereas increased ROS generation in plants has also been correlated to expression of resistance against biotrophs (Hückelhoven et al., 2001). ROS generation does not appear to play a role in the expression of virulence by biotrophic pathogens (Glazebrook, 2005), which is supposedly the type of parasitism of V. longisporum during early colonization of B. napus. The amount of H2O2 in hypocotyls of B. napus decreased significantly after V. longisporum inoculation in both water- and BABA-pretreated plants. This might be attributed to an elevated antioxidant enzyme activity of the pathogen resulting in mitigating the oxidative burst in host plants, as shown in rye/Claviceps and tobacco/Sclerotinia interactions (Cessna et al., 2000; Nathues et al., 2004), or due to increase in PAL activity in plant hypocotyls in response to pathogen invasion. Increase in PAL activity enhances the accumulation of phenylpropanoids such as phenolic acids, which are well known antioxidants capable of quenching H2O2 (Evans et al., 1997). Similar results have been obtained from rice/Magnaporthe grisea interactions, where SA was suggested to play a role as antioxidant, thereby preventing oxidative damage to plant tissue (Yang et al., 2004). However, all these examples refer to leaf pathogens, while no such findings have been reported so far for vascular pathogens.

A significant increase in endogenous SA levels was observed in B. napus hypocotyls after pathogen application to plants pretreated with water and BABA. A strong increase in SA levels in leaves of water-pretreated plants in response to the pathogen demonstrates activation of basal defence responses towards V. longisporum, although not sufficient to curtail pathogen growth, suggesting the existence of further components of resistance. Recently, Ratzinger et al. (2009) showed a significant increase in free and conjugated SA in xylem sap of B. napus infected with V. longisporum. This increase in SA was correlated with stunting and the amount of V. longisporum DNA found in hypocotyls. These results suggest the possibility of pathogen mediated diversion of the cinnamic acid pool (a common precursor for both SA and lignin biosynthesis) towards SA, at the expense of a rapid and effective lignin biosynthesis, hence weakening the plant defence response against fungal invasion. Whether in fact SA has a role as a negative regulator of resistance in B. napus responses to V. longisporum requires further investigations.

A potential role of SA in defence signalling in B. napus/V. longisporum interactions was investigated by studying the expression of PR-1, which encodes for a pathogenesis-related protein specifically induced by SA. Pathogen inoculation led to a strong induction of PR-1, as has been reported from Arabidopsis/V. longisporum interactions (Johansson et al., 2006). As PR-1 induction occurred to the same extent with or without BABA pretreatments, this protein may not be contributing to resistance. Expression of another defence-related gene, PDF1.2, was consistently induced in response to pathogen application, regardless of the pretreatments. However, expression of PDF1.2 could not be correlated to the expression of resistance either, because PDF1.2 levels were similar in inoculated plants with or without BABA pretreatment. In contrast to these results, JA and ET associated signals have been shown to be responsible for the defence responses in Arabidopsis/V. longisporum interactions (Johansson et al., 2006). Expression of PR-1 and PDF1.2 was not detected in the leaves of plants infected with V. longisporum, confirming the lack of systemic defence responses against the pathogen, as has been demonstrated in Arabidopsis/V. dahliae interactions (Veronese et al., 2003).

A possible role for BABA as inducer of resistance in B. napus against V. longisporum emerged from histochemical studies of the xylem vessels. These occlusions arise from physical blockage of the xylem vessel by the pathogen itself (Fradin & Thomma, 2006), or due to host defence responses associated with vascular plugging as a means to curtail pathogen spread (Benhamou, 1995).

Brassica napus plants colonized by V. longisporum showed the presence of a large number of occlusions in the xylem vessels. However, the occlusions did not appear to curtail spread of the pathogen. In fact, conidia have been reported to break through xylem occlusions in B. napus/V. longisporum interactions (Johansson, 2006). BABA-pretreated plants showed a striking absence of xylem occlusions. As a low amount of V. longisporum DNA was detected in middle stem and shoot regions of BABA-pretreated plants, the absence of occlusions correlated to a slower pathogen spread upward through the vascular system. A plausible explanation for this could be that occlusions are formed by the host plant only in response to a particular threshold level of fungal biomass and as BABA-pretreated plants accumulated fungal biomass below this threshold level, occlusions did not arise. In the case of BABA-pretreated plants, pathogen spread was limited without formation of occlusions in the vessel, in contrast to genotypic resistance. This suggests that occlusions are not the only way to restrict the vascular pathogen growth and spread. This is in contrast to the type of genotypic resistance in B. napus that has been attributed to higher percentage of occlusion formation in hypocotyls in resistant compared to susceptible genotypes (Eynck et al., 2009).

It is well known that the success of a plant in combating a pathogen may chiefly depend on the speed of responses and thus early defence responses by the host may be more effective in halting the invading pathogen. To study such early responses during BABA-induced resistance, activity of a defence-related enzyme, PAL, was analysed. PAL plays a significant role in the regulation of phenol biosynthesis in plants as a response to pathogen infection (Smith-Becker et al., 1998). Increase in PAL activity has been associated with incompatible interactions in several host pathosystems (Bonhoff et al., 1986).

BABA pretreatment led to an early and significant increase in PAL activity in hypocotyls in response to pathogen invasion, suggesting higher synthesis and accumulation of phenylpropanoids. Increase in PAL activity was found to correlate with large numbers of phenol-storing cells surrounding the vessels. These phenol-storing cells may play a crucial role in restricting the vascular pathogen by release of phenolics into the xylem vessel which may directly inhibit the pathogen growth. An increase in total soluble phenolics was also observed in banana roots in response to a F. oxysporum f. sp. cubense race 4 derived elicitor. The antifungal properties of phenolics are well reported. Adequate levels of chlorogenic acid account for the resistance of potato tubers against Verticillium albo-atrum (Lee & LeTourneau, 1958). In one of the studies, xylem extract obtained from a resistant tomato cultivar was found to suppress growth of F. oxysporum f. sp. lycopersici race 1 in vivo and has been considered responsible for containment of the pathogen, as a mechanism of wilt resistance in tomato (Stromberg & Corden, 1976). Similarly, genotypic resistance in B. napus was correlated with higher levels of soluble and cell wall-bound phenolics at early time points and lignin formation at the later stages (Eynck et al., 2009).

The experiments here indicated that BABA pretreatment led to induction of resistance in a susceptible B. napus cultivar by containment of the pathogen in the hypocotyl region of the plant, inhibiting shoot colonization. Absence of any occlusions in vessels during BABA IR suggests that physical localization is probably not the only mechanism contributing to curtailing the spread of this vascular pathogen. BABA primed an increase in PAL activity, and the subsequent synthesis and accumulation of phenolics are likely to be important factors exhibiting an inhibitory effect on pathogen growth, resulting in expression of resistance. In conclusion, this work provides evidence that BABA IR in B. napus against V. longisporum is through priming of the phenylpropanoid pathway and partially differs from genotype-related resistance; however, the exact role of phenylpropanoids in this defence needs to be further investigated.

Acknowledgements

This work was made possible through grants from Göttingen International funding provided by the University of Göttingen enabling A. Kamble a research placement in Göttingen. The authors thank Professor Petr Karlovsky, Göttingen, for his support in the analysis of fungal DNA by qPCR.

Ancillary