- © 1999 American Society of Plant Physiologists
Abstract
In Arabidopsis, systemic acquired resistance against pathogens has been associated with the accumulation of salicylic acid (SA) and the expression of the pathogenesis-related proteins PR-1, PR-2, and PR-5. We report here the isolation of two nonallelic mutants impaired in the pathway leading to SA biosynthesis. These SA induction–deficient (sid) mutants do not accumulate SA after pathogen inoculation and are more susceptible to both virulent and avirulent forms of Pseudomonas syringae and Peronospora parasitica. However, sid mutants are not as susceptible to these pathogens as are transgenic plants expressing the nahG gene encoding an SA hydroxylase that degrades SA to catechol. In contrast to NahG plants, only the expression of PR-1 is strongly reduced in sid mutants, whereas PR-2 and PR-5 are still expressed after pathogen attack. Furthermore, the accumulation of the phytoalexin camalexin is normal. These results indicate that SA-independent compensation pathways that do not operate in NahG plants are active in sid mutants. One of the mutants is allelic to eds5 (for enhanced disease susceptibility), whereas the other mutant has not been described previously.
INTRODUCTION
Infection with necrotizing pathogens often leads to the subsequent systemic induction of resistance. This is referred to as systemic acquired resistance (SAR), which is characterized by a long-lasting defense response against a broad spectrum of pathogens (Kuc, 1982; Ryals et al., 1994; Sticher et al., 1997). After the initial observation that treatment of plants with salicylic acid (SA) leads to resistance (White, 1979), several lines of evidence have shown that SA is associated with SAR. The expression of a number of pathogenesis-related (PR) proteins is highly correlated with acquired resistance (Ward et al., 1991; Uknes et al., 1992). The synthesis of PR proteins can be induced by treatment with SA and also by the application of functional analogs of SA, such as 2,6-dichloroisonicotinic acid (INA; Métraux et al., 1991) or benzo(1,2,3)thiadiazole-7-carbothioic acid S-methyl ester (BTH; Görlach et al., 1996).
The accumulation of SA in plant tissues correlates with the presence of PR proteins and resistance (Malamy et al., 1990; Métraux et al., 1990; Yalpani et al., 1991). In addition, transgenic tobacco (Gaffney et al., 1993) and Arabidopsis (Delaney et al., 1994) plants that express the Pseudomonas putida nahG gene, encoding an SA hydroxylase, which degrades SA to catechol, are hypersusceptible to infection with virulent pathogens and cannot develop SAR. Experiments with NahG plants also showed that SA is involved in gene-for-gene defense responses (Delaney et al., 1994). These findings provide strong evidence for the involvement of SA in SAR.
SA also has been proposed to be the systemic signal for SAR (Malamy et al., 1990; Métraux et al., 1990). However, although SA synthesized in infected leaves can be shown to be transported to uninfected leaves in tobacco and cucumber (Shulaev et al., 1995; Mölders et al., 1996), grafting experiments with tobacco plants expressing NahG, cholera toxin, or reduced levels of phenylalanine ammonia–lyase show that other primary systemic signals are involved in the onset of SAR (Vernooij et al., 1994; Beffa et al., 1995; Pallas et al., 1996). Recently, several lines of evidence indicate that one mode of action of SA in pathogen defense is to potentiate defense reactions (Mur et al., 1996; Shirasu et al., 1997).
The dissection of the signal transduction pathway connecting SA to resistance has been studied primarily in tobacco and Arabidopsis. In tobacco, a protein has been found that binds SA in vitro with a very high affinity (Du and Klessig, 1997). Furthermore, an SA-activated mitogen-activated protein (MAP) kinase has been identified, suggesting that a MAP kinase cascade participates in SA-dependent defense signaling (Zhang and Klessig, 1997). In Arabidopsis, the signal transduction pathway downstream of SA has been characterized by analyzing the npr1, nim1, and sai1 mutants that have been isolated by several groups (for no PR gene expression [Cao et al., 1994], noninducible immunity [Delaney et al., 1995], and SA insensitive [Shah et al., 1997], respectively). PR genes are not induced in the npr1/nim1/sai1 mutants after the application of typical inducers of SAR, and they are highly susceptible to infection by pathogens. The NPR1/NIM1 gene has been cloned recently, and it encodes a protein containing ankyrin repeats (Cao et al., 1997; Ryals et al., 1997). Its postulated role as a transcriptional regulator could be substantiated by its localization in the nucleus upon treatment with SA (X. Dong, personal communication).
Other mutants affected in the signal transduction pathway linking pathogen infection with plant resistance could be positioned upstream of the synthesis or action of SA. lsd (for lesions simulating disease resistance response; Dietrich et al., 1994) and acd2 (for accelerated cell death; Greenberg et al., 1994) mutants have high SA levels when expressing lesions, which is reminiscent of a hypersensitive reaction (HR). Interestingly, in most lsd mutants, lesion formation is independent of SA; however, for lsd6 and lsd7 mutants, lesion formation depends on SA (Weymann et al., 1995). Several mutants were identified by constitutive expression of PR proteins (cpr). These mutants have constitutively high levels of SA and therefore act upstream of SA.
cpr mutants also may be affected in other disease response pathways. Whereas cpr1 seems to be specifically affected in the SA pathway (Bowling et al., 1994), cpr5 forms spontaneous lesions (Bowling et al., 1997) and shows constitutive expression of the SA-independent regulated plant defensin PDF1.2. (Penninckx et al., 1996). The cpr6 mutant has high constitutive SA and PDF1.2 levels and provides evidence for an SA-dependent expression of PR-1, PR-2, and PR-5 that is independent of NPR1/NIM1/SAI1 (Clarke et al., 1998). The dnd1 (for defense no cell death) mutant also exhibits high constitutive levels of SA, PR-1, and PR-2 and a broad spectrum of resistance (Yu et al., 1998). The dnd1 mutant, however, does not undergo an HR reaction in a gene-for-gene resistance response. Enhanced disease susceptibility (eds) mutants are compromised in resistance (Glazebrook et al., 1996; Parker et al., 1996; Rogers and Ausubel, 1997; Volko et al., 1998); however, their SA levels have not been reported. Interestingly, eds5-1 is only defective in the expression of PR-1, but not in that of PR-2 or PR-5 (Rogers and Ausubel, 1997).
In addition to being important for SAR, SA has been found to be necessary but not sufficient for the accumulation of the phytoalexin camalexin (Zhao and Last, 1996; Zhou et al., 1998). In Arabidopsis plants that express NahG, the tryptophan pathway and accumulation of camalexin are down-regulated coordinately (Zhao and Last, 1996). However, the cpr1 mutant shows no alterations in the pathway leading to the formation of tryptophan or camalexin, despite having high levels of SA (Zhao and Last, 1996). The pad4 mutant is impaired in the accumulation of SA and camalexin as well as in the expression of PR-1 after infection with virulent bacteria. Application of SA to pad4 plants restores camalexin production and PR-1 expression in the presence of the bacteria (Zhou et al., 1998).
In this study, we describe two Arabidopsis mutants that do not accumulate SA after pathogen infection. These SA induction–deficient (sid) mutants are more susceptible to P. syringae and Peronospora parasitica and show a reduced SAR response. After pathogen inoculation, only PR-1 expression was strongly reduced in sid mutants, whereas PR-2 and PR-5 were still normally expressed. Furthermore, sid mutants had normal levels of camalexin after bacterial infection. A model of pathogen response pathways in sid mutants is presented.
RESULTS
sid1 and sid2 Do Not Accumulate SA after Pathogen Inoculation
Approximately 4500 Arabidopsis plants (accession Columbia [M2]) mutagenized by using ethyl methanesulfonate were screened for altered levels of total (free and conjugated) SA after inoculation with the avirulent strain P. syringae pv tomato DC3000 carrying the avirulence gene avrRpm1. Four mutant lines that accumulate much less SA in comparison to wild-type plants were isolated; they were called sid because they were deficient in the induction of SA accumulation.
Genetic complementation tests indicated that the four mutant lines fall into two complementation groups, one group consisting of sid1 and the other of three alleles of sid2 (sid2-1, sid2-2, and sid2-3; data not shown). Therefore, two loci, SID1 and SID2, were identified by mutation. Leaf extracts from each mutant were analyzed by using HPLC to quantify free and conjugated SA after inoculation with the avirulent strain P. s. tomato DC3000 carrying avrRpt2. Figure 1 shows an ∼10- to 20-fold reduction in the amounts of free and conjugated SA in sid1 and sid2 mutants when compared with the wild type. The amounts of free SA in sid mutants were similar to that in NahG plants, whereas slightly higher amounts of conjugated SA were observed for the sid1 mutant (Figure 1). Similar results were obtained after infection with the virulent strain P. s. tomato DC3000 and with avirulent strains carrying the avrRpm1 or avrRps4 gene (data not shown).
Genetic Characterization
Measurements of the amount of SA in pathogen-inoculated F1 hybrids from crosses between wild-type plants and sid mutants indicated that both sid1 and sid2 mutants are recessive (total SA content: SID1/sid1, 2.45 ± 0.15 μg/g fresh weight, n = 3; SID2/sid2, 2.37 ± 0.378 μg/g fresh weight, n = 3; Columbia [Col-0], 3.27 ± 0.535 μg/g fresh weight, n = 3). Both sid1 and sid2 segregated in the F2 progeny as single Mendelian loci (SID1:sid1, 45:9 [0.5 < P < 1.0, χ2 test] and SID2:sid2, 43:9 [0.5 < P < 1.0, χ2 test]). To map the genes on the Arabidopsis genome, we outcrossed sid1 and sid2 to Arabidopsis accession Landsberg erecta (Ler), and F2 plants were scored for cosegregation of low SA levels with simple sequence length polymorphisms (Bell and Ecker, 1994) and cleaved amplified polymorphic sequence markers (Konieczny and Ausubel, 1993). Analysis of 48 sid1 plants selected from a Ler × sid1 F2 population mapped sid1 to the lower arm of chromosome 4, at a distance of ∼2 centimorgans from the marker nga1107. Analysis of 48 sid2 plants from a Ler × sid2 F2 population mapped sid2 between AthATPase and g17311 on the lower arm of chromosome 1.
Endogenous Levels of Free and Conjugated SA in Leaves of Arabidopsis Col-0 Plants, NahG Plants, and sid Mutants after Inoculation with P. s. tomato DC3000 Carrying avrRpt2.
Leaf material was analyzed 2 days after inoculation with a bacterial suspension of OD600 of 0.2. The values represented are the average for three treated plants (±se). The experiment was repeated twice with similar results. Conj, conjugated SA; FW, fresh weight.
sid1 and sid2 Have a Blockage in SA Biosynthesis
Accumulation of SA has been found to be triggered by various abiotic stresses that also lead to pathogen resistance (Yalpani et al., 1994). sid mutants were analyzed for their ability to accumulate SA after callus formation or after subjecting plants to abiotic stresses, such as UV-C or ozone. Ozone fumigation and callus formation gave rise to substantial amounts of SA in wild-type plants, as illustrated in Figures 2A and 2B, whereas UV-C only weakly induced SA synthesis (data not shown). However, none of these stresses could trigger accumulation of SA in sid mutants (Figures 2A and 2B). Because lack of SA accumulation could be explained by a higher rate of SA degradation, the endogenous SA content was analyzed in sid mutants after exposure to high amounts of SA. Figure 2C shows that after a soil-drench treatment with 1 mM SA, the amounts of total SA were as high in sid1 and sid2 mutants as in wild-type plants, whereas NahG plants had a very low content of SA. Thus, the low levels of SA in sid mutants do not result from enhanced degradation of SA but rather from a blockage in the steps leading to the biosynthesis of SA after pathogen infection.
Total Amounts of SA in Leaves of Arabidopsis Col-0 Plants, NahG Plants, and sid Mutants.
(A) Leaf tissue 24 hr after ozone treatment.
(B) Callus.
(C) Leaf tissue 1 day after drenching the soil with 1 mM SA was analyzed. Each histogram represents the average of two samples containing callus material or leaf material of treated plants (±se).
The experiment was repeated once for (A) and (B) or twice for (C) with similar results. FW, fresh weight.
sid1 and sid2 Have Enhanced Susceptibility to P. parasitica
SA has been shown to play an important role in resistance to the obligate biotroph P. parasitica (Mauch-Mani and Slusarenko, 1996). To investigate whether sid1 and sid2 mutants were more susceptible to virulent P. parasitica isolates, we inoculated 4-week-old plants with the P. parasitica isolate NOCO (105 conidiospores per mL), and the number of adult leaves showing conidiospores was scored 7 days later. Whereas sporulation only occurred on ∼40 to 50% of the wild-type leaves, sporulation was observed on 75 to 85% of the leaves of sid1 and sid2 mutants. In comparison, nearly 100% of the leaves of NahG plants showed sporulation under the same conditions.
When NahG plants were inoculated with isolates of P. parasitica that are avirulent on Arabidopsis accession Col-0, no HR reaction was observed, and the microorganism could complete its reproductive cycle, although growth of the hyphae was surrounded by a trailing necrosis indicating an intermediate resistance (Delaney et al., 1994). The resistance phenotype of sid mutants to the incompatible P. parasitica isolates EMWA and WELA (105 conidiospores per mL) therefore was determined. Microscopic investigation of lactophenol–trypan blue—stained leaves at 1, 3, 5, 7, and 9 days after infection showed that both sid mutants have a similarly decreased resistance to P. parasitica WELA and EMWA. The oomycete could complete its reproductive cycle, forming both oospores and conidiospores, although the growing hyphae were surrounded by a trailing necrosis similar to that in NahG plants. However, conidiospore formation was in general somewhat lower than in NahG plants. Typical reactions of wild-type plants, NahG plants, and sid mutants are presented in Figures 3A to 3D.
sid Mutants Are More Susceptible to P. s. tomato DC3000
Resistance to pathogens was investigated further using inoculation with P. syringae. After inoculations with a low titer (∼105 colony-forming units per mL) of the virulent P. s. tomato DC3000 strain, sid mutants developed enhanced disease symptoms (data not shown) and supported stronger bacterial growth in comparison to wild type, as shown in Figure 4A. Although sid1 and sid2 mutants react with an HR in <24 hr after a high-titer inoculation (∼108 colony-forming units per mL) with avirulent strains of P. s. tomato DC3000 carrying the avrRpm1, avrRpt2, or avrRps4 gene, reduced resistance could be observed after a low titer inoculation (∼105 colony-forming units per mL), as indicated by enhanced growth of P. s. tomato DC3000 carrying avrRpt2 shown in Figure 4B. However, bacteria grew to a lower titer in sid mutants compared with NahG plants, as seen best 3 days after inoculation (Figures 4A and 4B).
Resistance Response of Arabidopsis Col-0 Plants, NahG Plants, and sid Mutants to Incompatible P. parasitica Isolates.
Nine days after spray inoculation of 2-week-old plants with 105 conidiospores per mL of P. parasitica WELA ([A], [B], and [D]) or EMWA (C), respectively, leaves were stained with lactophenol–trypan blue and viewed under the microscope.
(A) Col-0 with blue staining of typical HR reactions.
(B) NahG plants with blue staining of hyphae, oospores, and necrosis following a hypha.
(C) sid1 mutant with blue staining of conidiospores.
(D) sid2 mutant with blue staining of hyphae with trailing necrosis. Bars = 25 μm.
sid1 and sid2 Mutants Have an Impaired SAR Response
SAR was tested in sid mutants by giving the lower leaves a high-titer inoculation of P. s. tomato DC3000 carrying avrRpt2. The result was that these leaves became necrotic. Five days after the primary inoculation, plants were sprayed with a spore suspension of P. parasitica isolate NOCO. Seven days after the secondary infection, the disease was evaluated on the upper leaves of the rosette by using a five-step rating of sporulation. In wild-type plants, SAR occurred, as evidenced by a strong reduction of symptoms measured on the basis of either the total number of leaves showing sporulation or the degree of sporulation (Table 1). By contrast, NahG plants showed severe disease symptoms, with almost 100% sporulation and collapse of the infected leaves. sid1 and sid2 mutants were much more susceptible to P. parasitica NOCO than were wild-type plants, but not as susceptible as their NahG counterparts.
PR Expression in sid1 and sid2 Mutants
Leaves were infected with the avirulent strain P. s. tomato DC3000 carrying avrRpt2 to examine the expression of PR genes in infected and systemic leaves. RNA gel blots of leaf samples taken 2, 3, 4, and 5 days after inoculation were probed for PR-1 (unknown function), PR-2 (β-1,3-glucanase), and PR-5 (thaumatin-like protein; Uknes et al., 1992). Figure 5 shows the expression of these PR genes in uninfected, infected, and uninfected systemic leaves of infected plants 4 days after pathogen inoculation. PR-1 expression was reduced ∼10-fold in infected and systemic leaves of sid mutants to a level similar to NahG plants. However, the expression of PR-2 and PR-5 in both infected and systemic leaves of sid mutants remained similar to that in controls in most experiments. In some experiments, a slight reduction in the expression of either PR-2 or PR-5 could be seen. Similar results were obtained from plants 2 and 3 days after inoculation with P. s. tomato DC3000 carrying avrRpt2, and 2 and 4 days after infection with P. s. tomato DC3000 (data not shown). As shown in Figure 6, PR-1 expression could be restored to the wild-type level in both sid mutants by the application of the SAR inducers INA and BTH, indicating that the signal transduction pathway downstream of SA is not affected in sid1 and sid2 mutants.
PDF1.2 Is Induced Normally in sid Mutants
Arabidopsis reacts to inoculation with the necrotizing fungus Alternaria brassicicola by developing small dark necrotic spots and inducing SA-independent defense pathways that lead to the accumulation of PDF1.2 (Penninckx et al., 1996, 1998). The effect of A. brassicicola therefore was tested in sid mutants to determine whether the PDF1.2 pathway is operative in these plants. sid1 and sid2 mutants as well as NahG and wild-type plants developed similar necrotic spots after inoculation with A. brassicicola (data not shown). In Figure 7, the results of a RNA gel blot analysis of leaves 2 days after inoculation with A. brassicicola is presented. Shown is a strong induction of PDF1.2 in wild-type and NahG plants as well as in sid1 and sid2 mutants. Therefore, the SA-independent defense pathway leading to the induction of PDF1.2 seems to operate normally in sid1 and sid2 mutants.
Growth of the Virulent Strain P. s. tomato DC3000 and of the Isogenic Strain Carrying the Avirulence Gene avrRpt2 in Leaves of Arabidopsis Col-0 Plants, NahG Plants, and sid Mutants.
(A) P. s. tomato DC3000.
(B) P. s. tomato DC3000 carrying avrRpt2.
Leaves were analyzed for bacterial density at different time points after inoculation with 0.5 × 103 bacteria per cm2. Data points represent the means of three pools derived from eight replicate samples (±se; see Methods). By using the Student-Newman-Keuls analysis performed for the 3-day values, we placed sid mutants in a separate group compared with NahG plants and Col-0 with a significance level of 0.01. The experiments were repeated twice with similar results. cfu, colony-forming unit; diamonds, Col-0; squares, NahG; triangles, sid1; circles, sid2.
SAR Analysis of Arabidopsis Col-0 Plants, NahG Plants, and sid Mutants Infected with P. parasitica Isolate NOCO after Preinoculation of Lower Leaves with P. s. tomato DC3000 Carrying avrRpt2
Expression of PR-1, PR-2, and PR-5 in Leaves of Arabidopsis Col-0 Plants, NahG Plants, and sid Mutants after High-Titer Inoculation with P. s. tomato DC3000 Carrying avrRpt2.
Four days after inoculation, infected, uninfected systemic leaves of infected plants, and leaves of control plants were harvested. Two independent samples containing RNA (10 μg) of two treated plants were analyzed together with one control sample (C, control; I, infected leaves; S, uninfected systemic leaves of infected plants). Membranes were probed with PR-1, PR-2, and PR-5, respectively, and with the 18S rDNA as internal standard. The experiment was repeated twice with similar results.
sid Mutants Accumulate Camalexin Normally
Accumulation of camalexin has been shown to be affected by the amount of SA in leaves (Zhao and Last, 1996; Zhou et al., 1998). After inoculation with P. s. tomato DC3000 carrying avrRpt2, the camalexin levels in sid1 and sid2 were higher in comparison to wild-type plants and much higher in comparison to NahG plants, as shown in Figure 8. After infection with P. s. tomato DC3000, a similar tendency was observed (data not shown). The enhancement of camalexin accumulation in sid mutants might reflect a larger proportion of the leaf area reacting to pathogen infection in sid mutants in comparison to wild-type plants. However, it is striking that sid mutants are able to accumulate high levels of camalexin independently of SA.
PR-1 Expression in Leaves of Arabidopsis Col-0 Plants, NahG Plants, and sid Mutants after the Application of SAR Inducers.
Soil was drenched with wetting agents for 2 days (C), 5 ppm INA (I) for 2 days, or 330 ppm BTH (B) for 1 day before harvesting the leaf material. Ten micrograms of total RNA was probed with PR-1 and with the 18S rDNA as internal standard. The experiment was repeated once with similar results.
sid1 Is Allelic to eds5
Similar to sid mutants, the eds5 mutant exhibits enhanced disease susceptibility toward virulent strains of P. syringae (Rogers and Ausubel, 1997; Volko et al., 1998). The eds5-1 mutant is characterized by reduced expression of PR-1, but unaltered expression of PR-2 and PR-5 and unaltered accumulation of camalexin in response to infection with virulent bacteria (Rogers and Ausubel, 1997). sid1 and eds5 map to similar positions on the Arabidopsis genome. Therefore, the amounts of SA after pathogen infection were measured in eds5-1 (Rogers and Ausubel, 1997) and eds5-2 (Volko et al., 1998). Four days after infection with P. s. tomato DC3000 carrying avrRpt2, SA had accumulated to similar levels in sid1 and eds5 mutants (sid1, 38.3 ± 2.4 ng/g fresh weight; eds5-1, 30.5 ± 1.7 ng/g fresh weight; eds5-2, 34.0 ± 3.3 ng/g fresh weight). Complementation analyses were performed between sid1 and eds5-1 and eds5-2, respectively. The F1 generations of the crosses eds5-1 × sid1 and eds5-2 × sid1 were analyzed for accumulation of SA 3 days after inoculation of 3-to-4-week-old plants with P. s. tomato DC3000 carrying avrRpt2. The amounts of SA in the F1 plants eds5-1/sid1 and eds5-2/sid1 were 403 ± 49 (n = 4) and 460 ± 110 ng/g fresh weight (n = 4), respectively. These values are similar to those of the parents (eds5-1, 394 ± 55 ng/g fresh weight [n = 3]; eds5-2, 423 ± 87 ng/g fresh weight [n = 3]; sid1, 658 ± 138 ng/g fresh weight [n = 3]; as a comparison Col-0, 5340 ± 545 ng/g fresh weight [n = 3]), indicating that sid1 belongs to the same complementation group as eds5. sid1 will therefore be referred to as eds5-3 in subsequent publications. The small differences in the amounts of SA indicate that eds5-1 is a tighter allele than is sid1.
DISCUSSION
Both sid1 and sid2 mutants have very low levels of SA after infection by virulent or avirulent bacterial and fungal pathogens. SA levels in these mutants do not increase in response to a variety of abiotic stresses, and the capacity of the mutants to degrade SA is not enhanced. Therefore, both sid mutants are very probably affected in the pathway leading to the biosynthesis of SA. sid mutants are more susceptible to local infection by a virulent strain of P. syringae as well as a virulent isolate of P. parasitica and are also more susceptible to pathogens that are normally avirulent on Arabidopsis accession Col-0. Although sid mutants exhibit a normal HR reaction <1 day after high-titer inoculation with avirulent bacteria, P. parasitica isolates that normally trigger an HR response can grow and sporulate in sid mutants. Avirulent bacteria grow to a higher titer in sid mutants than they do in wild-type plants when inoculated at low titers. SAR is reduced but not totally suppressed in sid mutants, indicating that both local and systemic resistance depend on normal SA levels.
Similar results have been observed for transgenic Arabidopsis plants that express NahG, an SA hydroxylase from P. putida that converts SA to catechol (Delaney et al., 1994). However, sid mutants exhibit less severe phenotype alterations after most pathogen infections in comparison to NahG plants. Interestingly, the different levels of resistance are reflected in the pattern of PR gene expression. Whereas expression of PR-1 was abolished to a similar extent in sid mutants and NahG plants, PR-2 and PR-5 were expressed to wild-type levels in sid mutants, but their expression was strongly reduced in NahG plants.
Expression of PDF1.2 in Arabidopsis Col-0 Plants, NahG Plants, and sid Mutants after Inoculation with A. brassicicola.
Two days after inoculation with A. brassicicola (I) or H2O (C), leaves of two plants were harvested together for each treatment. Ten micrograms of RNA was probed with PDF1.2 and with the 18S rDNA as an internal standard. The experiment was repeated once with similar results.
Accumulation of Camalexin in Leaves of Arabidopsis Col-0 Plants, NahG Plants, and sid Mutants after Inoculation with P. s. tomato DC3000 Carrying avrRpt2.
Two days after inoculation with a bacterial suspension, leaves were analyzed for accumulation of camalexin. The data represent means of three replicate samples (±se). sid mutants were placed in a separate group compared with Col-0 and NahG plants by using the Student-Newman-Keuls analysis with a significance level of 0.01. The experiment was repeated twice with similar results.
The absence of coordinate regulation between PR-1 and PR-2/PR-5 observed in sid mutants stands in contrast with other results showing that PR-1, PR-2, and PR-5 are regulated coordinately. For example, application of SA or SAR inducers, such as INA or BTH, induces the expression of PR-1, PR-2, and PR-5 coordinately (Uknes et al., 1992; Görlach et al., 1996). In addition, the npr1/nim1 mutant (Cao et al., 1994; Delaney et al., 1995; Shah et al., 1997) affects expression of PR-1, PR-2, and PR-5, whereas cpr mutants constitutively express all three PR genes (Bowling et al., 1994, 1997; Clarke et al., 1998). However, different lines of evidence hinted at the existence of an SA-independent PR-2/PR-5 expression after inoculation with P. syringae. For example, after pathogen infection, npr1 plants are deficient in PR-1 expression but still express PR-2 and PR-5 (Glazebrook et al., 1996). However, from these results, it could not be determined whether this expression of PR-2/PR-5 is SA dependent or independent. Furthermore, the pad4 mutant, which has normal expression of PR-2/PR-5, but diminished expression of PR-1 and a decreased accumulation of camalexin after infection with virulent pathogens, has been found to be affected in its ability to accumulate SA (Zhou et al., 1998). In addition, eds5-1 has been reported to be specifically affected in its expression of PR-1, but not in its expression of PR-2/PR-5 (Rogers and Ausubel, 1997).
Our findings that SA does not accumulate in eds5-1 and eds5-2 after pathogen inoculation and that sid1 is allelic to eds5 confirm the indications that inoculation with P. syringae leads to SA-independent expression of PR-2/PR-5. Why this pathway for the expression of PR-2/PR-5 is not active in NahG plants remains an open question.
Because eds5-1 does not accumulate SA after infection with pathogens, the observation that eds5-1 npr1 double mutants are more susceptible to Erysiphe orontii and have a lower expression of PR-1 than do both parental lines cannot be explained by EDS5 acting downstream of SA (Reuber et al., 1998). However, it is more likely that the eds5 mutation interrupts the pathway upstream of SA. SA might activate several proteins that act in synergy to induce the expression of PR-1. In addition, both mutants may be slightly leaky; double mutants thus could give a stronger phenotype, although being part of the same pathway (Reuber et al., 1998).
Another major difference between NahG plants and sid mutants is the different level of accumulation of camalexin. In both sid mutants, camalexin was found to accumulate to high levels in contrast to NahG plants. The low amount of camalexin in NahG plants has been interpreted to indicate that SA is essential for the normal expression of the camalexin pathway (Zhao and Last, 1996). One possible explanation for the differences in the pathogen response pathways leading to accumulation of camalexin between sid mutants and NahG plants might lie in fundamental changes of their respective metabolisms. Because SA is permanently degraded in NahG plants, the metabolic drain from the phenylpropanoid pathway may cause a downregulation of the camalexin pathway, which branches off further upstream in the phenylpropanoid pathway. However, the pad4 mutant shows reduced accumulation of camalexin that could be rescued by the application of SA when the pathogen was present (Zhou et al., 1998).
Model for the Role of sid Mutants in Pathogen Response Pathways.
CA, camalexin; JA, jasmonic acid; question mark, unknown signaling molecule; double lines crossing an arrow, blocked pathway; T bar crossed by double lines, blocked inhibitor function.
This observation seems to be in contradiction with the results reported in this study. Normal accumulation of camalexin also has been reported for eds5-1 mutants (Rogers and Ausubel, 1997). The normal accumulation of camalexin in sid1/eds5 and sid2 mutants may be explained by an SA-independent compensation pathway that is only active in sid1/eds5 and sid2 mutants. Such a pathway might be specifically activated by the presence of metabolic precursors that accumulate if enzymes leading to the synthesis of SA are affected in sid mutants. It is also possible that SID proteins have a dual regulatory function in activating the pathways that depend on SA and inactivating pathways that do not depend on SA. In the differentiation pathway of yeast, MAP kinases have been found to have such dual regulatory functions (Madhani et al., 1997). However, it also might be possible that SA regulates several pathways with different affinity.
A working model for the signal transduction pathways in sid mutants is proposed in Figure 9. sid mutants are placed upstream of SA and downstream of HR and of the branch point of the SA-independent pathways, leading either to expression of PDF1.2 or PR-2/PR-5 or to the SA-independent pathway for accumulation of camalexin. A closer genetic positioning of the eds5/sid1 and sid2 mutants in the pathogen response pathways might be obtained in epistasis experiments between different mutants affected in pathogen defense, such as the cpr1, cpr5, and cpr6 mutants (Bowling et al., 1994, 1997; Clarke et al., 1998).
It is also noteworthy that no significant difference in growth of bacteria after infection with P. syringae pv maculicula ES4326 carrying avrRpt2 was observed among eds5-1, eds5-2, and wild-type plants (Rogers and Ausubel, 1997; Volko et al., 1998), whereas the resistance was strongly diminished in the eds5 mutants after infection with the isogenic virulent strain. Because eds5/sid1 is defective in accumulation of SA after inoculation with avirulent pathogens (see above), it can be concluded that eds5 can establish a race-specific incompatible reaction in the absence of SA after inoculation with certain pathovars of P. syringae. The bacterial growth experiments presented in this study show a small but significant growth enhancement of P. s. tomato DC3000 carrying avrRpt2 when inoculated onto sid1/eds5 plants. In addition, experiments with incompatible isolates of P. parasitica showed that resistance to incompatible pathogens is decreased (see above). SA can therefore affect resistance to both virulent and avirulent pathogens, as has been shown with NahG plants.
In conclusion, the identification of sid mutants provides new information on the relationship between SA and SAR, on the induction pathways for PR proteins, and the accumulation of camalexin. The finding that eds5 and sid1 fall into the same complementation group adds evidence to the general notion that a defense network is present in plants and that different branches of the network may contribute differently to resistance against different pathogens. Characterization of the eds5/sid1 and sid2 genes will provide further information on the elements involved in the induction of SA accumulation.
METHODS
Growth Conditions for Plants, Fungi, and Bacteria
Ethyl methanesulfonate–mutagenized M2 seeds of Arabidopsis thaliana accession Columbia (Col-0; a gift from C.R. Somerville, Carnegie Institution, Stanford, CA) and ethyl methanesulfonate–mutagenized Col-0 M2 seeds homozygous for a mutation in ferulic acid 5-hydroxylase (fah1-2) (Chapple et al., 1992; Glazebrook et al., 1996; a gift from E. Rogers and F. Ausubel, Massachusetts General Hospital, Boston, MA) were used for the mutant screen. Salicylic acid (SA) induction–deficient mutants sid1 and sid2-1 were isolated from Col-0/fah1-2 M2 seeds, whereas sid2-2 and sid2-3 were isolated from Col-0 M2 seeds. For physiological analyses, plant lines of sid1 and sid2-1 backcrossed at least twice, but usually four or five times were used. Because of linkage between sid1 and fah1, sid1 still carries the fah1-2 background mutation that has been shown not to influence plant–pathogen interactions (Glazebrook et al., 1996).
Plants were grown on a pasteurized soil mix of humus/perlite (3:1) under a 16-hr-light and 8-hr-dark cycle for genetic analysis and a 12-hr-light and 12-hr-dark cycle for physiological experiments, with a night temperature of 16°C and a day temperature of 20 to 22°C (60 to 70% humidity), when not otherwise indicated.
Arabidopsis accessions Col-0, Landsberg erecta (Ler), and Wassilewskija were obtained from the Arabidopsis Biological Research Center (Columbus, OH). NahG plants were obtained from J. Ryals, Novartis Corp, Basel, Switzerland.
Peronospora parasitica isolate NOCO was transferred every week on 2-to-3-week-old Arabidopsis accession Col-0 plants by spray inoculation with a spore suspension. P. parasitica isolate WELA was cultivated in a similar manner every 10 days on Arabidopsis accession Ler, and P. parasitica isolate EMWA was cultivated every week on accession Wassilewskija (Holub et al., 1994; isolates obtained from B. Mauch-Mani, University of Fribourg, Fribourg, Switzerland). Plants inoculated with P. parasitica were kept in a 12-hr-light and 12-hr-dark cycle with 19°C day temperature and 16°C night temperature and, for the first day and the last day of the growth cycle, under 100% humidity to ensure infection and sporulation, respectively (Dangl et al., 1992).
Alternaria brassicicola was cultivated on half-strength potato dextrose agar at 20°C (Penninckx et al., 1996).
Strain DC3000 of Pseudomonas syringae pv tomato and the isogenic strains carrying the avirulence genes avrRpt2, avrRpm1, or avrRps4 (Whalen et al., 1991) were cultured at 28°C and 220 rpm in Luria-Bertani or Kings B medium (pH 7.0, 10 mg/mL protease peptone, 15 mg/mL glycerol, 1.5 mg/mL K2HPO4, and 4 mM MgSO4) containing the appropriate antibiotic.
Mutant Screen
Five to 6-week-old M2 plants were syringe-inoculated with P. s. tomato carrying avrRpm1 at a titer of ∼108 colony-forming units per mL (OD600 nm 0.2) on four to five leaves by infiltrating ∼20 to 25% of the leaf surface. Four days after inoculation, 0.05 to 0.20 g of leaf material was analyzed for total SA content.
Measurements of SA and Camalexin
For the analysis of total SA and camalexin, 5-to-6-week-old plants were inoculated on approximately one-quarter to one-fifth of the leaf surface and analyzed individually, when not described otherwise. When 3-to-4-week-old plants were analyzed, ∼30 to 50% of the leaf was inoculated and the leaves of several plants were pooled. Total SA and camalexin were determined by extracting 0.1 to 0.2 g of leaves with 90% methanol at 65°C for 4 hr in the presence of 250 ng of o-anisic acid as internal standard (Meuwly and Métraux, 1993). After evaporation of the methanol in a vortex evaporator, each sample was incubated with 2 mg of β-glucosidase (Fluka, Buchs, Switzerland) in 150 μL of 0.1 M Na acetate, pH 5.5, for 1.5 hr at 30°C. After precipitation with 0.5 mL of 5% trichloroacetic acid, the phenol compounds, including camalexin, were partitioned against cyclohexane/ethyl acetate (1:1).
For the analysis of free and conjugated SA and camalexin, 0.1 to 0.3 g of leaves (with 250 or 1000 ng of o-anisic acid) were extracted once with 2 mL of 70% methanol and once with 2 mL of 90% methanol by using a homogenizer (Polytron; Kinematica, Littau, Switzerland). After evaporation of the methanol from the combined extracts and trichloroacetic acid precipitation as described above, free phenols and camalexin were extracted into cyclohexane/ethyl acetate (1:1). The remaining aqueous phase was submitted to acid hydrolysis in the presence of 4 N HCl at 80°C for 1 hr, and the liberated phenols were extracted into cyclohexane/ethyl acetate, as described above.
For HPLC, the organic phase was evaporated, and the samples were resuspended in 85% phosphate buffer/15% acetonitrile. Chromatography was performed on a reverse phase HPLC column (ABZ+, 25 cm × 4.6 mm; Supelco, Buchs, Switzerland) (Meuwly and Métraux, 1993). The detection limit of SA was 2 ng.
RNA Gel Blot Analysis
RNA (1 g) was extracted with 3 mL of 2 M Tris-HCl, pH 8.0, 0.5 M EDTA, pH 8.0, and 20% SDS (1:2:1) (Chirjwin et al., 1979). The aqueous phase was extracted once with 1 volume of phenol–chloroform–isoamyl alcohol (50:49:1) and twice with 1 volume of chloroform and precipitated with 1 volume of 6 M LiCl overnight at 4°C. The RNA was resuspended in H2O and reprecipitated with one-tenth of a volume of 3 M NaAc, pH 5.5, and 3 volumes of ethanol. Ten micrograms of RNA was separated on a formaldehyde/agarose gel, visualized under UV light, and transferred to a Nylon membrane (Hybond-N; Amersham Pharmacia Biotech, Little Chalfont, UK). The membrane was probed with the cDNA of pathogenesis-related proteins PR-1, PR-2, PR-5 (Uknes et al., 1992), and PDF1.2 (Penninckx et al., 1996). As internal standard, the membranes were probed with a gene encoding the 18S RNA of tomato (Kiss et al., 1988).
Treatments with Chemicals, UV Light, and Ozone
Two-to-three-week-old plants were soil-drenched with 1 mM SA for SA uptake experiments. To induce PR proteins, we soil-drenched 2-to-3-week-old plants with 5 ppm of 2,6-dichloroisonicotinic acid (INA; Novartis Corp, Basel, Switzerland) or 330 mM benzo(1,2,3)thiadiazole-7-carbothioic acid S-methyl ester (BTH; Novartis Corp, Basel, Switzerland). Two-to-three-week-old plants were illuminated for 20 min with 254 nm UV-C light at a 30-cm distance. Four-week-old plants were subjected twice to 300 ppm of ozone for 6 hr at 24-hr intervals. Samples were taken after 3, 6, 12, 18, 24, and 48 hr from the beginning of the fumigation.
Callus Induction
Leaves were surface-sterilized with a 5% sodium hypochlorite and 0.01% Triton X-100 solution and incubated on callus-inducing medium (Axelos et al., 1992). Emerging calli were maintained on Murashige and Skoog medium (Sigma, Buchs, Switzerland) containing 0.2 g/L glutamine, 1 mg/L 2,4-D, and 0.3 mg/L kinetin.
Inoculation with P. syringae
The hypersensitive response (HR) was induced by inoculation with a bacterial suspension of ∼108 colony-forming units per mL in 10 mM MgCl2. To examine the growth of the bacteria, we inoculated 5-to-6-week-old plants with a bacterial suspension of ∼105 colony-forming units per mL. The bacterial titer was determined from three leaves (0.5 cm2 discs per leaf) per plant from six to eight replicates at different time points after inoculation. To obtain the mean values and standard error, we analyzed either three leaves of each plant together, and each replicate plant was taken separately, or we pooled one infected leaf disc per plant from the different replicates, resulting in three pools. Both methods gave similar results.
Inoculation with P. parasitica
To examine local resistance to P. parasitica isolate WELA or EMWA, 2-to-3-week-old plants were sprayed with a suspension of 105 conidiospores per mL and incubated as described above. Either plants were examined for sporulation or leaf samples were stained with lactophenol–trypan blue every 2 days and examined under the microscope (Koch and Slusarenko, 1990). To examine virulence or the systemic acquired resistance response, we challenge-inoculated 4-to-5-week-old plants 5 days after treatment with H2O (mock) or high-titer inoculation with P. s. tomato DC3000 carrying avrRpt2 (∼108 colony-forming units per mL) with 105 conidiospores per mL of P. parasitica isolate NOCO, and sporulation was examined 7 days later.
Inoculation with A. brassicicola
Four-to-five-week-old plants were inoculated with droplets (6 × 5 μL) of a spore suspension (106 spores per mL) on five leaves per plant. After 2 days at 100% humidity, plants were visually examined, and inoculated leaves were harvested for RNA gel blot analysis.
Acknowledgments
We gratefully acknowledge the financial support of the Swiss National Science Foundation (Grant Nos. 31-34098 and 31-55662.98 to J.-P.M.) and the EMBO and Novartis Foundations (fellowships to C.N.). We thank Drs. Antony Buchala and Ghislaine Rigoli for help with HPLC; Drs. Roger Innes, Jeff Dangl, Brigitte Mauch-Mani, and William Broekaert for providing pathogens and for helpful discussions; and Drs. Fred Ausubel and Chris Somerville for supplying us with mutagenized seed and seed of eds5-1 and eds5-2 mutants. We are grateful to Dr. Jürg Fuhrer for performing the ozone fumigations and to Drs. Antony Buchala, Thierry Genoud, Brigitte Mauch-Mani, Philippe Reymond, and Yves Poirier for critical reading of the manuscript.
- Received April 12, 1999.
- Accepted May 20, 1999.
- Published August 1, 1999.
REFERENCES
