Induction of resistance to Phytophthora sojae in soyabean (Glycine max) by salicylic acid and ethylene

Authors


E-mail: cjjiang@affrc.go.jp

Abstract

This study investigated the protective effects of various phytohormones and plant activators on soyabean resistance to Phytophthora sojae, the cause of root and stem rot. It was found that application of benzothiadiazole (BTH, an activator of salicylic acid (SA) signalling) and 1-aminocyclopropane-1-carboxylic acid (ACC, a precursor of ethylene (ET) biosynthesis) markedly induced resistance to P. sojae. By contrast, gibberellin (GA) and abscisic acid (ABA) rendered soyabean seedlings more susceptible to P. sojae. Simultaneous application of ABA with BTH or ACC suppressed the protective effects of BTH and ACC, indicating that ABA acts antagonistically on the SA- and ET-signalling pathways. Neither BTH nor ACC directly inhibited growth of P. sojae. The protective effect of ACC was diminished by co-treatment with its analogue α-aminoisobutyric acid, suggesting that ET biosynthesis is required for ACC-induced soyabean resistance. Expression analysis of ET- and SA-responsive genes demonstrated the activation of ET- and SA-signalling pathways during P. sojae infection. Furthermore, ACC treatments augmented the expression of ET-responsive and pathogenesis-related genes. Taken together, the results indicate that ACC (ET)-induced soyabean resistance to P. sojae relies on transcriptional augmentation of defence-related genes.

Introduction

Plants have developed a battery of complex defence mechanisms to prevent infection by harmful attackers. Physical and chemical barriers constitute the first defence layer. When plants are overcome by certain pathogens, they recruit an inducible defence system to limit further pathogen ingression. The phytohormones salicylic acid (SA), jasmonates (JA) and ethylene (ET) are known to play key roles in regulating the induced defence response (Pieterse et al., 2009, 2012; Robert-Seilaniantz et al., 2011). SA plays a crucial role in plant defence and is generally important for the activation of defence responses to biotrophic and hemibiotrophic pathogens (Vlot et al., 2009). By contrast, JA and ET are usually associated with defence against necrotrophic pathogens (Vlot et al., 2009). The SA and JA/ET defence pathways are often mutually antagonistic. However, synergistic interactions have been reported in some pathosystems (Pieterse et al., 2009). Thus, the cross-talk between the SA and JA/ET pathways appears to be regulated differently, depending on the specific pathogen. In addition, other phytohormones have also been implicated in defence responses. Cytokinins (CK), gibberellin (GA) and brassinosteroids (BR) have been shown to positively modulate SA-mediated resistance to biotrophic pathogens, while abscisic acid (ABA) and auxins negatively modulate it (Robert-Seilaniantz et al., 2011; Pieterse et al., 2012).

At the site of pathogen invasion, plants recognize common features of microbial pathogens, e.g. flagellin, chitin, glycoproteins and lipopolysaccharides, referred to as pathogen-associated molecular patterns (PAMPs; Jones & Dangl, 2006; Hein et al., 2009). Perception of PAMPs by host-encoded pattern-recognition receptors on the cell surface initiates diverse downstream signalling events, which ultimately result in the primary immune response called PAMP-triggered immunity (PTI). Phytohormone-mediated signalling pathways are known to play important roles in PTI. For example, flagellin-triggered PTI is partially dependent on SA signalling (Tsuda et al., 2008). In flagellin-triggered PTI against virulent bacteria, ET signalling is required (Boutrot et al., 2010). Some pathogens have acquired effector proteins that actively suppress host defence responses and overcome PTI. In turn, plants have evolved resistance (R) proteins that recognize these specific pathogen effectors, resulting in secondary defence responses called effector-triggered immunity (ETI; Jones & Dangl, 2006; Hein et al., 2009).

In addition to the defence responses that are expressed locally upon primary pathogen attack, plants are capable of enhancing their level of resistance to future pathogen attack. This phenomenon is known as induced resistance, and can be triggered by a variety of biotic and abiotic stimuli. A well-known example of induced resistance is often referred to as systemic acquired resistance (SAR), in which defence responses occur in plant parts distal from infection sites following attack by a necrosis-inducing pathogen (Ryals et al., 1996). SAR requires endogenous accumulation of SA, and is associated with the transcriptional activation of SA-inducible genes encoding pathogenesis-related (PR) proteins. Exogenous application of SA can also induce SAR. Apart from pathogen-induced SAR, some strains of non-pathogenic rhizobacteria can induce systemic resistance, called induced systemic resistance (ISR). ISR requires a signalling pathway(s) dependent on JA and ET, and is not associated with transcriptional activation of the SA-responsive PR genes (Pieterse et al., 2012).

In addition to biological resistance-inducing agents, some chemicals can trigger induced resistance. Among the best characterized are 2,6-dichloroisonicotinic acid (INA), probenazole (PBZ) and benzothiadiazole (BTH) (Beckers & Conrath, 2007). These compounds induce the same spectrum of resistance as pathogen-induced SAR, with concomitant activation of SA-dependent PR genes. Upon pathogen attack, BTH-treated plants exhibited faster and stronger induction of the gene encoding phenylalanine ammonia lyase (Kohler et al., 2002). Meanwhile, a non-protein amino acid, β-aminobutyric acid (BABA), was shown to trigger a response similar to ISR, in which no major transcriptional activation of defence-related genes (including PR genes) occurred prior to pathogen infection (Zimmerli et al., 2000). Nevertheless, upon pathogen infection, plants treated with BABA showed faster and stronger transcription of the SA-inducible PR1 gene. These phenomena are called ‘priming’. In general, the previously mentioned systemic resistance responses are associated with the priming mechanism (Conrath, 2006). In contrast to constitutive activation of defence responses, priming does not require major metabolic changes in the absence of pathogen attack. Therefore, it constitutes a defence strategy imposing low fitness cost on plants, while acting against a broad spectrum of attackers (van Hulten et al., 2006).

The oomycete Phytophthora sojae causes phytophthora root and stem rot of soyabean and limits production in several regions of the world. The pathogen induces pre- and post-emergence damping off of seedlings. Annual yield losses caused by P. sojae are estimated to total $1–2 billion per year (Tyler, 2007). In Japan, the disease is particularly destructive, because ≥90% of soyabean is grown in waterlogged upland fields converted from rice paddy fields. In Japan, the disease was first observed in Hokkaido, in the northernmost part of the country, and later throughout most regions. Phytophthora sojae is a soilborne oomycete that produces free-swimming zoospores when conditions are cool and wet. The zoospores are attracted by isoflavonoids secreted from soyabean roots (Morris et al., 1998). At the root surface, they attach, encyst and extend a germ tube that penetrates into host cells. Control of root and stem rot currently relies mainly on chemical fungicides containing metalaxyl. Genetic control strategies involve single race-specific resistance encoded by Rps (resistance to P. sojae) genes, and multigenically controlled field tolerance or partial resistance (Tyler, 2007). However, the molecular mechanisms underlying soyabean resistance to P. sojae remain elusive. In the present study, the roles of various phytohormones and activators in this resistance, via effects on defence-related genes, were investigated.

Materials and methods

Plant material

Soyabean (Glycine max) cv. Enrei was used for all experiments in this study. Seeds were germinated in a commercial potting mix in 5-cm plastic pots (four seeds per pot). Seedlings were incubated in a growth chamber at 25°C, with 50% relative humidity and a 16-h photoperiod.

Pathogen culture and inoculation

Phytophthora sojae strains N1 (isolated in Niigata, Japan) and Tohoku (Th1, isolated in Iwate, Japan), which are virulent to soyabean cv. Enrei, were used in this study. Mycelial plugs from a mother plate were placed on vegetable juice (V8) agar plates and grown for a week at 26°C. For inoculation of soyabean seedlings, the surface of hypocotyls just above the ground was cut longitudinally (1-cm length) with a needle. The tip of the needle was bent perpendicularly so as to keep the cut depth approximately 1 mm. Phytophthora sojae mycelium from the V8 agar plates was homogenized by passage through an 18-gauge needle three times with a syringe, and then applied to the cut sites of soyabean hypocotyls and sealed with cellophane tape to maintain moisture. Homogenized V8 agar without mycelium was used for mock inoculation. Approximately 20 seedlings were used for each inoculation, and the percentage of surviving seedlings 6 days post-inoculation (dpi) was scored. For gene expression analysis, the inoculated parts of hypocotyls and the distal parts of the first internode just above the cotyledons (2-cm length) were sampled, immediately frozen in liquid nitrogen, and kept at −80°C until use.

Chemical treatments of soyabean seedlings

Seven-day-old soyabean seedlings were soil-drenched for 24 h with the following chemical compounds in H2O: brassinolide (BR; Wako); sodium salicylate (SA; Nacalai Tesque); indole-3-acetic acid (IAA; Sigma); kinetin (CK; Sigma); gibberellin A3 (GA3; Wako); 1-aminocyclopropane-1-carboxylic acid (ACC; Sigma); abscisic acid ((±)-cis-trans, ABA; Sigma); methyl jasmonate (JA; Wako); benzothiadiazole S-methyl ester (BTH; Wako); azelaic acid (Az; Sigma); β-aminobutyric acid (BABA; Wako); harpin (HALO Harpin; HydroGarden Ltd); and ethephon (Sigma). With the exception of BR (2 μm), SA (500 μm) and harpin (5 mg mL−1), all compounds were used at a concentration of 50 μm.

In a further experiment, 7-day-old soyabean seedlings were pretreated for 24 h with 50 μm ACC alone, or a combination of 50 μm ACC and 500 μm α-aminoisobutyric acid (AIB; ACC analogue).

RT-PCR analysis

The hypocotyl samples from five or six seedlings were ground into a fine powder in liquid N2. Total RNA was isolated using an RNeasy mini kit (QIAGEN). Reverse transcription was performed using Revertra Ace (Toyobo) and oligo(dT)23 primers. Quantitative RT-PCR (qRT-PCR) was run on a Thermal Cycler Dice TP800 system (Takara), using SYBR premix Ex Taq mixture (Takara), with cycles of 95°C for 5 s, 55°C for 20 s and 72°C for 20 s. Actin (Glyma15g05570, listed in the Phytozome soyabean genome database http://www.phytozome.net/soybean) was used as an internal standard to normalize cDNA concentrations. The expression patterns of three soyabean PR genes (PR1 [Glyma13g32510], PR4 [Glyma19g43460] and PR10 [Glyma07g37240]), PRD12 (the gene encoding an SA-inducible ATP-binding cassette transporter [Glyma03g32520] (Eichhorn et al., 2006)), ACO (the gene encoding ACC oxidase, an enzyme involved in ethylene synthesis [Glyma07g15480]), and ERF1 (a soyabean gene encoding a protein with the highest homology to the Arabidopsis thaliana transcription factor AtERF1 [Glyma20g34570, referred to as soyabean ERF1 in this study]), were examined at 0, 4, 6 and 8 h post-inoculation (hpi). The PCR primers used were: ActinF (5′-GAGCTATGAATTGCCTGATGG-3′), ActinR (5′-CGTTTCATGAATTCCAGTAGC-3′), PR1F (5′-TGATGTTGCCTACGCTCAAG-3′), PR1R (5′-AAGCAGCAACCGTATCATCC-3′), PR4F (5′-GCTTGCGGGTGACAAATAC-3′), PR4R (5′-ACACTCCCACGTCCAAATC-3′), PR10F (5′-GCCCAGGAACCATCAAGAAG-3′), PR10R (5′-CGCTGTAGCTGTATCCCAAG-3′), PRD12F (5′-CACGACCGAGAATTCCAGTGT-3′), PRD12R (5′-TCCCGTCGCTGGATTCC-3′), ACOF (5′-CATGTTTTTCGCGTTCTCCT-3′), ACOR (5′-AAGTACAGAAAGAAAGGGATGGA-3′), ERF1F (5′-GCTTAAGGAGATGAACTATGCAAA-3′) and ERF1R (5′-TTGACGCTAATTTTCCTTCTCAA-3′).

Results

Induction of soyabean resistance to P. sojae by SA and ET

Phytohormones play key signalling roles in a diverse range of disease resistance. To investigate the protective effects of phytohormones and plant activators on soyabean resistance to P. sojae, 7-day-old soyabean seedlings were soil drenched with IAA, CK, GA, ACC, ABA, BR, JA, SA, BTH, BABA, azelaic acid and harpin for 24 h, and challenged with a virulent isolate of P. sojae (N1). As shown in Figures 1 and 2, pretreatment of soyabean seedlings with SA, BTH or ACC markedly enhanced the survival rate. By contrast, pretreatment with GA or ABA rendered the seedlings more susceptible to P. sojae. Under the experimental conditions used, none of the remaining compounds tested affected the resistance of soyabean seedlings to P. sojae (Fig. 1a). Similar protective effects of BTH and ACC were observed against another virulent isolate of Psojae, Th1 (Fig. 1c). When ACC was sprayed onto the leaves instead of applied by soil drenching, no resistance to P. sojae was observed (data not shown). In order to exert systemic effects on the soyabean–P. sojae interaction, ACC may need to enter the vascular system through uptake by the roots, and be transported throughout the plant.

Figure 1.

Survival rates of soyabean plants soil-drenched for 24 h with various phytohormones and plant activators, and then inoculated with a virulent isolate of Phytophthora sojae: (a) 7 days post-inoculation (dpi) with P. sojae (N1), following treatment with IAA, GA, CK, SA, BTH, ACC, JA, ABA, BR, Az, BABA, Hrp or CON; (b) over a time course of 3–6 dpi with P. sojae (N1), following treatment with GA, ABA, ACC, BTH, SA or CON; and (c) 7 dpi with P. sojae (Th1), following treatment with BTH, ACC or CON. Representative data from three independent experiments are shown. IAA, indole-3-acetic acid (50 μm); GA, gibberellin (50 μm); CK, kinetin (50 μm); SA, salicylic acid (0·5 mm); BTH, benzothiadiazole S-methyl ester (50 μm); ACC, 1-aminocyclopropane-1-carboxylic acid (50 μm); JA, methyl jasmonic acid (50 μm); ABA, abscisic acid (50 μm); BR, brassinolide (2 μm); Az, azelaic acid (50 μm); BABA, β-aminobutyric acid (50 μm); Hrp, HALO harpin (5 mg mL−1); CON, mock treatment.

Figure 2.

Disease symptoms and effects of exogenous application of phytohormones on the resistance of soyabean plants to Phytophthora sojae. (a) Mock- (left) and P. sojae (N1)-infected seedlings. (b) Seedlings soil-drenched for 24 h with SA, BTH, ACC, ABA, GA or CON, and then inoculated with P. sojae (N1). All plants shown 7 days post-inoculation. Treatment details as in Fig. 1.

Antagonistic action of ABA on SA- and ET-induced resistance of soyabean to P. sojae

It is well known that antagonistic cross-talk between the SA and JA/ET pathways plays an important role in plant–pathogen interactions. To evaluate whether BTH- or ACC-induced resistance of soyabean plants to Psojae is regulated by other phytohormones, soyabean seedlings were treated with BTH or ACC alone and also with BTH or ACC in combination with other phytohormones. As shown in Figure 3, ABA negated the protective effects of BTH and ACC against Psojae, suggesting an antagonistic action on the SA- and ET-signalling pathways. By contrast, GA did not exhibit antagonistic activity, although when applied singly, it compromised resistance (Fig. 1). Interestingly, BTH did not compromise the effect of ACC, and vice versa. Similarly, JA did not compromise the effect of BTH. None of other phytohormones tested affected BTH- and ACC-induced resistance to P. sojae (data not shown).

Figure 3.

Suppression of BTH- and ACC-induced resistance to Phytophthora sojae by ABA. Seven-day-old soyabean seedlings were pretreated for 24 h with BTH, BTH + GA, BTH + ACC, BTH + ABA, ACC, ACC + GA, or ACC + ABA, and then inoculated with P. sojae (N1). Survival rates at 7 days post-inoculation are shown. Chemicals and concentrations as in Fig. 1. Experiments were repeated twice with similar results.

Lack of growth inhibition of P. sojae by BTH and ACC

The SA-, BTH- and ACC-induced resistance of soyabean to P. sojae may have been caused by a direct inhibitory effect on the growth of P. sojae, rather than by activation of the soyabean defence system. To investigate this possibility, the growth of P. sojae (N1) was examined on V8 plates containing ACC, SA or BTH. It was found that BTH and ACC did not inhibit the growth of P. sojae (N1) on V8 plates. By contrast, SA showed a marked inhibitory effect (Fig. 4).

Figure 4.

Effects of SA, BTH and ACC on growth of Phytophthora sojae. Mycelia were grown for 1 week on V8 agar plates containing BTH (benzothiadiazole S-methyl ester; 10, 50 and 250 μm), ACC (1-aminocyclopropane-1-carboxylic acid; 10, 50 and 250 μm) or SA (salicylic acid; 100, 500 and 2500 μm). Experiments were repeated twice with similar results.

Requirement of ethylene biosynthesis for the protective effect of ACC

It has been shown that cyanide, a by-product of ET biosynthesis, is responsible for the blast resistance of rice conferred by ACC application (Seo et al., 2011). To investigate this possibility in the soyabean–P. sojae interaction, the effect of AIB (an ACC analogue that interferes with conversion of ACC to ET) on the ACC-induced resistance of soyabean plants to P. sojae was examined. As shown in Figure 5, AIB compromised the ACC-induced resistance of soyabean to P. sojae, indicating a critical requirement for ET biosynthesis. Furthermore, pretreatment with ethephon (a compound that produces ET within plant cells without producing cyanide) increased the survival rate of soyabean plants almost as efficiently as did ACC. These findings indicate that ET, but not cyanide, is responsible for the ACC-induced resistance of soyabean plants to P. sojae.

Figure 5.

Role of ethylene biosynthesis in ACC-induced resistance to Phytophthora sojae. Seven-day-old soyabean seedlings were pretreated for 24 h with (a) 50 μm ACC alone, or a combination of 50 μm ACC (1-aminocyclopropane-1-carboxylic acid) and 500 μm α-aminoisobutyric acid (AIB), or (b) 50 μm ACC or 50 μm ethephon. Survival rates at 7 days post-inoculation are shown. Experiments were repeated twice with similar results.

Activation of the SA- and ET-signalling pathways in response to P. sojae infection

To elucidate the mechanisms of the soyabean defence response to P. sojae, an expression analysis of soyabean defence-related genes was conducted using qRT-PCR. In accordance with a previous study (Moy et al., 2004), inoculation with a virulent isolate of P. sojae (N1) induced ACO 16 times more than did mock inoculation (Fig. 6). In comparison with the control, inoculation with P. sojae (N1) also induced ERF1 13-fold, and up-regulated PDR12 58-fold. These data indicate that the ET- and SA-signalling pathways are activated in response to P. sojae infection.

Figure 6.

Induction of ethylene (ET) biosynthesis (ACO), ET signal transduction (ERF1), and salicylic acid (SA)-responsive (PDR12) genes during Phytophthora sojae infection of soyabean. Seven-day-old seedlings were inoculated with a virulent isolate of P. sojae (N1). Relative expression levels of ACO (a), ERF1 (b) and PDR12 (c) at the indicated time points (days post-inoculation, (dpi)) were determined by quantitative RT-PCR (qRT-PCR), using Actin (Glyma15g05570) as an internal control. Data are means (SD) of three independent determinations. Experiments were repeated twice with similar results.

ACC-induced modification of defence gene expression against P. sojae infection

To elucidate the mechanisms underlying ACC-induced resistance to P. sojae, the expression patterns of three soyabean PR genes was analysed: PR1 (Glyma13g32510), PR4 (Glyma19g43460) and PR10 (Glyma07g37240), following P. sojae infection of soyabean seedlings that had been either mock- or ACC-treated for 24 h (Fig. 7). All three PR genes were up-regulated in response to P. sojae infection (data not shown). In addition, a higher level of PR1 induction was detected around the infection sites of ACC-treated seedlings than around the infection sites of mock-treated seedlings (Fig. 7a), indicating a modifying effect of ACC treatment on P. sojae-induced PR1 expression. The expression levels of PR1 around the distal sites of both mock- and ACC-treated seedlings were much lower than those around the infection sites (data not shown). The expression patterns of PR4 differed from those of PR1. Prior to P. sojae infection, ACC-treated seedlings exhibited an approximately eightfold higher PR4 level than did mock-treated seedlings. Subsequent P. sojae infection resulted in approximately comparable PR4 expression levels in mock- and ACC-treated seedlings (Fig. 7b). Interestingly, following Psojae infection, the PR4 levels around the distal sites of ACC-treated seedlings were 2·8–6·2 times higher than those of mock-treated seedlings (Fig. 7d). Moreover, the PR4 levels around the distal sites of both mock- and ACC-treated seedlings were comparable to those in the infection sites (Fig. 7b,d). With respect to the expression pattern of PR10, no differences were observed between mock- and ACC-treated seedlings, in either the infection or distal sites (Fig. 7c; data not shown).

Figure 7.

Effect of ACC (1-aminocyclopropane-1-carboxylic acid) treatment on the expression of PR genes during Phytophthora sojae infection of soyabean. Seven-day-old seedlings that had been either mock- or ACC-treated for 24 h were inoculated with a virulent isolate of P. sojae. The relative expression levels of PR1 (a), PR4 (b,d) and PR10 (c) at the indicated time points (hours post-inoculation, (hpi)) around infection (a–c) and distal (d) sites were determined by qRT-PCR, using Actin as an internal control. Data are means (SD) of three independent determinations. Experiments were repeated twice with similar results.

The effect of ACC treatment on the expression patterns of ACO and ERF1 in response to P. sojae infection was investigated further. Prior to P. sojae infection, ACC-treated seedlings exhibited a 9·7-fold higher ACO level than did mock-treated seedlings. Following P. sojae infection, this higher level was retained, not only in infection sites, but also at distal sites (Fig. 8a,c). The expression pattern of ERF1 following P. sojae infection showed a response to ACC treatment similar to that of PR4. Prior to P. sojae infection, ACC-treated seedlings exhibited a 5·5-fold higher ERF1 level than did mock-treated seedlings. Following P. sojae infection, ERF1 levels in the infection sites of ACC-treated seedlings did not differ from those of mock-treated seedlings (Fig. 8b). By contrast, the ERF1 levels around the distal sites of ACC-treated seedlings were 6·4- to 15-fold higher than were those around the distal sites of mock-treated seedlings, although the expression levels around the distal sites were one order of magnitude lower than were those in the infection sites (Fig. 8b,d). Taken together, the results indicate that ACC treatment augments the induction of defence-related genes in response to Psojae infection.

Figure 8.

Effect of 1-aminocyclopropane-1-carboxylic acid (ACC) treatment on the expression of genes for ethylene (ET) biosynthesis and signal transduction during Phytophthora sojae infection of soyabean. Seven-day-old seedlings were treated as in Fig. 7. The relative expression levels of ACO (a,c) and ERF1 (b,d) at the indicated time points (hour post-inoculation, (hpi)) around the infected (a,b) and distal sites (c,d) were determined by qRT-PCR, using Actin as an internal control. Data are means (SD) of three independent determinations. Experiments were repeated twice with similar results.

Discussion

Phytophthora root and stem rot of soyabean, caused by P. sojae, is one of the most devastating diseases in soyabean-producing regions worldwide. The defence responses of soyabean to infection by this oomycete have been studied for decades. However, the molecular basis underlying these defence processes remains largely unclear. Many natural and synthetic compounds protect plants from disease by inducing resistance to pathogen infection, rather than by directly inhibiting pathogen growth (Zimmerli et al., 2000; Beckers & Conrath, 2007). The present study revealed that treatment with ACC or BTH enhanced soyabean resistance to virulent strains of P. sojae (Fig. 1). Neither of the chemicals exhibited a direct inhibitory effect on the growth of P. sojae (Fig. 4). Moreover, they showed no curative effect, because they protected soyabean against Psojae only when applied prior to inoculation (data not shown). The effect of ACC application was caused by the activation of ET signal transduction, rather than by generation of the by-product cyanide (Fig. 5). Taken together, these results indicate that ACC and BTH confer resistance to soyabean by activating host resistance mechanisms. BTH and ethephon are currently commercialized for agricultural application. The findings of this study suggest that BTH and ethephon may be used for disease control against P. sojae in the field.

The SA-signalling pathway plays a central role in plant defence in a variety of plant species, and may be required for establishment of local and systemic resistance to various biotrophic and hemibiotrophic pathogens (Vlot et al., 2009). A transcriptional cofactor, NPR1, has been shown to play a key role in the SA-signalling pathway of several plant species (Vlot et al., 2009). In soyabean, application of BTH and INA was previously demonstrated to reduce the severity of white mould disease caused by Sclerotinia sclerotiorum (Dann et al., 1998). The present study found that the SA-signalling pathway plays a positive role in resistance to P. sojae. Soyabean has two NPR1 genes, GmNPR1-2 and GmNPR1-2 (Sandhu et al., 2009), and these may be involved in BTH-induced resistance.

Ethylene is known to be an important modulator of plant defence responses to pathogens (van Loon et al., 2006). In some plant species, ET can act negatively on disease resistance. For example, disease tolerance to Xanthomonas campestris pv. vesicatoria and Fusarium oxysporum f. sp. lycopersici was enhanced in ethylene-insensitive tomato plants (Lund et al., 1998). However, many studies have demonstrated that ET plays a positive role in disease resistance. In a model legume, Medicago truncatula, the EIN2 protein (a positive regulator of ET signalling) is involved in defence against Phytophthora medicaginis (Penmetsa et al., 2008). In soyabean, mutants with reduced ET sensitivity showed compromised resistance to avirulent P. sojae isolates, but no alteration in disease severity with respect to virulent isolates (Hoffman et al., 1999). In addition, it was previously reported that ACC treatment reduced hyphal growth of Psojae in excised hypocotyls (Yoshikawa et al., 1990) and inhibited disease development in excised cotyledons (Park et al., 2002). The present study demonstrated that ET positively regulates resistance of intact soyabean seedlings to Psojae.

In general, systemic defence responses are associated with priming for faster and stronger activation of defence genes upon pathogen challenge and also direct induction of defence mechanisms (Conrath, 2006). Consistent with this, the present study revealed that the expression of PR1 was primed by ACC treatment (Fig. 7). In addition, expression of PR4 (Fig. 7), ACO (an ethylene biosynthetic gene, Fig. 8) and ERF1 (an ethylene signal transduction gene, Fig. 8) was augmented by ACC in response to P. sojae infection. ERF1 is known to play a crucial role in ET-mediated disease resistance. For example, in A. thaliana, ERF1 is induced by pathogen infection and by ethylene treatment. Overexpression of AtERF1 increases Athaliana resistance to Botrytis cinerea (Berrocal-Lobo et al., 2002). Similarly, transgenic Medicago truncatula plants that overexpress the MtERF1-1 gene in their root system show resistance to P. medicaginis and also to Rhizoctonia solani (Anderson et al., 2010). Hence, ERF1 may also be involved in soyabean resistance to P. sojae. Interestingly, the effects of ACC treatment on defence-related gene inductions appear to be selective. In the present study, up-regulation of PR1 by P. sojae infection was primed in infection sites, whereas expression of PR4 and ERF1 was augmented at distal sites. Induction of ACO was augmented at infection sites and also at distal sites. Given that ERF1 expression was affected by ACC treatment only at distal sites, it is plausible that this transcription factor plays a role in the expression of PR4 at distal sites, in response to P. sojae infection.

Effects of ET on SA- and JA-signalling pathways in defence responses are complex. In many plant–pathogen interactions, ET signalling appears to act antagonistically on SA signalling. For instance, global expression analysis of P. syringae-infected A. thaliana plants showed that many SA-responsive genes are enhanced in the ein2 mutant, indicating that ET signalling is suppressive to SA signalling (Glazebrook et al., 2003). However, in many cases, ET potentiates the effect of SA. In A. thaliana, ET not only acts synergistically with SA on PR1 induction but also bypasses NPR1 dependency of the SA-mediated antagonistic effect on JA signalling (Leon-Reyes et al., 2009). In addition, stimulation of the ET-signalling pathway prior to activation of the SA-response renders A. thaliana plants insensitive to SA-mediated suppression of JA-dependent defence responses (Leon-Reyes et al., 2010). In tobacco, ET plays an important role in SAR to Tobacco mosaic virus (TMV) (Knoester et al., 2001). In this host–pathogen combination, ET perception is required for systemic signalling in TMV-infected leaves. Similarly, in the compatible interaction of tomato and X. campestris pv. vesicatoria, suppression of disease symptom development is mediated by sequential action of JA, ET and SA (O'Donnell et al., 2003). The present study demonstrated that SA- and ET-mediated defence responses are effective in resistance to P. sojae. Moreover, co-treatment with ACC and BTH resulted in no antagonistic or cooperative effects. These data suggest a sequential action of ET and SA in the soyabean–P. sojae interaction. Alternatively, the two pathways may act in parallel during the host defence response.

As in many other plant–pathogen interactions, it was observed here that ABA or GA treatment increased soyabean susceptibility to P. sojae (Fig. 1). It was further demonstrated that the protective effect of ET application against P. sojae was suppressed by ABA, but not by GA application (Fig. 3), implying that ABA and GA act on different defence response pathways. These results are in accordance with those of a previous study, in which ABA treatment compromised the incompatible interaction of soyabean plants with an avirulent race of Psojae, whereas treatment with norflurazon (an ABA biosynthesis inhibitor) enhanced resistance to a virulent race of Psojae (McDonald & Cahill, 1999). In A. thaliana, antagonistic interactions of ABA with the SA- and ET-signalling pathways modulate disease resistance and defence gene expression (Anderson et al., 2004). Further research is required to elucidate the molecular mechanisms involved in the antagonistic action of ABA on the SA- and ET-signalling pathways.

Acknowledgements

We thank Dr Yamamoto (National Agriculture and Food Organization, Japan) for providing Psojae strain Tohoku (Th1) and Dr Kaga for technical advice. This work was supported by a Grant-in-Aid for Scientific Research (No. 22570057) from the Ministry of Education, Culture, Sports, Science and Technology, Japan and a NIAS Strategic Research Fund from the National Institute of Agrobiological Sciences, Japan.

Ancillary