A fatty acid desaturase modulates the activation of defense signaling pathways in plants

Abstract

Salicylic acid (SA) plays an important role in activating various plant defense responses, including expression of the pathogenesis-related (PR) genes and systemic acquired resistance. A critical positive regulator of the SA signaling pathway in Arabidopsis is encoded by the NPR1 gene. However, there is growing evidence that NPR1-independent pathways can also activate PR expression and disease resistance. To elucidate the components associated with NPR1-independent defense signaling, we isolated a suppressor of the npr1–5 allele, designated ssi2. The recessive ssi2 mutation confers constitutive PR gene expression, spontaneous lesion formation, and enhanced resistance to Peronospora parasitica. In contrast, a subset of defense responses regulated by the jasmonic acid (JA) signaling pathway, including expression of the defensin gene PDF1.2 and resistance to Botrytis cinerea, is impaired in ssi2 plants. With the use of a map-based approach, the SSI2 gene was cloned and shown to encode a stearoyl-ACP desaturase (S-ACP DES). S-ACP DES is an archetypical member of a family of soluble fatty acid (FA) desaturases; these enzymes play an important role in regulating the overall level of desaturated FAs in the cell. The activity of mutant S-ACP DES enzyme was reduced 10-fold, resulting in elevation of the 18:0 FA content in ssi2 plants. Because reduced S-ACP DES activity leads to the induction of certain defense responses and the inhibition of others, we propose that a FA-derived signal modulates crosstalk between different defense signaling pathways.

Disease resistance in plants is sometimes associated with the development of a hypersensitive response, in which necrotic lesions form at the sites of pathogen entry and the pathogen is restricted to these regions (1). Concurrent with the hypersensitive response, salicylic acid (SA) levels in the inoculated leaves increase, and many defense-associated genes, including the pathogenesis-related (PR) genes, are induced. Subsequent to these events, increased levels of SA and PR gene transcripts are detected in the uninoculated leaves. Furthermore, the uninoculated portions of the plant develop a long-lasting resistance to a broad spectrum of pathogens, a phenomenon known as systemic acquired resistance (SAR). Because increases in PR gene expression are tightly correlated with both hypersensitive response and SAR development, these genes are excellent markers for the activation of resistance responses.

Many studies have demonstrated that SA is an important component of the signal transduction pathway leading to SAR (2–4). Genetic analyses have further revealed that the NPR1 gene encodes a critical positive regulator of this pathway (5–8). Unlike wild-type (wt) plants, npr1 mutants fail to develop SAR or express PR genes after treatment with SA or other SAR-inducing compounds. However, inoculation with pathogen triggers PR gene expression in npr1 plants. This finding, along with studies of other Arabidopsis mutants and wt plants, suggests that an NPR1-independent pathway(s) also regulates PR gene expression and resistance to certain pathogens (7–11).

In addition to SA, the signaling molecules ethylene and jasmonic acid (JA) are involved in the induction of various defense responses. JA and/or ethylene is required for resistance to Botrytis cinerea and Alternaria brassicicola (12); they also signal induced systemic resistance triggered by Pseudomonas fluorescens (13, 14). Exogenously applied JA and/or ethylene also activates expression of the thionin (THI2.1) and defensin (PDF1.2) genes (15–17). Genetic analyses have revealed that all of these phenomena, with the exception of induced systemic resistance activation, are mediated by a pathway(s) that is independent of both SA and NPR1 (induced systemic resistance requires NPR1; 12–17).

The relationship between the SA and the JA/ethylene defense response pathways is not well understood (18). Some studies have demonstrated that these signals work synergistically to induce defense responses (19, 20). However, other evidence suggests that these pathways function antagonistically. For example, plants responding to a given pathogen usually do not activate both SA-associated and JA/ethylene-dependent defenses. Furthermore, SA and JA have been shown to antagonize the activation of each other's defense responses, and SA can inhibit JA biosynthesis (21, 22). The isolation of several Arabidopsis mutants that constitutively accumulate both SA-induced PR genes and the JA/ethylene-induced PDF1.2 gene has led to the suggestion that these pathways share signaling components that are involved in the positive and/or negative cross-regulation of their activities (10, 23, 24).

To elucidate the pathway(s) through which NPR1-independent defense responses are activated, we conducted a suppressor screen in the npr1–5 mutant background and scored for plants that constitutively expressed the PR genes. Through this process, the ssi2 mutant was identified (25). Plants carrying the recessive ssi2 mutation are severely stunted and exhibit constitutive activation of an NPR1-independent pathway, leading to spontaneous lesion formation, PR gene expression, and resistance to Peronospora parasitica. In contrast, the induction of some, but not all, JA-dependent defense responses is impaired in ssi2 plants. Cloning of the SSI2 gene revealed that it encodes a stearoyl-ACP desaturase (S-ACP DES). Consistent with this discovery, ssi2 protein exhibited reduced levels of S-ACP DES activity and the mutant plant accumulated elevated levels of its 18:0 substrate. Based on these findings, we propose a model in which a fatty acid (FA)-derived signal(s) modulates the crosstalk between different defense signaling pathways.

Materials and Methods

Plant Growth Conditions and Genetic Analysis.

Plants were grown as described (10). A ssi2/ssi2 plant derived from Arabidopsis ecotype Nössen (Nö) was crossed with a SSI2/SSI2 (wt) plant from the Columbia (Col-0) ecotype. Cleaved amplified polymorphic sequence (CAPS) (26) and simple sequence length polymorphic (SSLP) (27) marker analyses were performed on 656 F₂ progeny that, based on their morphology and PR-1 gene expression, were homozygous for the ssi2 mutation. This analysis placed ssi2 on chromosome 2, ≈0.2 cM from AthB102 on the centromeric side and 3.7 cM from GBF on the telomeric side. With the use of sequence information generated by the Arabidopsis genome project, 14 additional CAPS markers spanning this region were generated and used to further delimit the region containing ssi2.

Derived-CAPS Analysis.

A 100-bp fragment was amplified with the use of PCR primers p1 (5′-AGAGAGGGCTAGAGAGCTCCCTG-3′) and p2 (5′-AGTGTTCAACATAGTTTGATAGGTCCTAA-3′) from the chromosomal DNA of wt, mutant, and T₁ or T₂ progeny of ssi2/ssi2∷SSI2 transgenic plants. The bases italicized in p2 were present as GG in the original sequence; this modification created a DdeI site in the PCR product amplified from wt DNA. Because the ssi2 mutation alters the 3′ base flanking AA of p2, no DdeI site is present in the PCR product amplified from ssi2 DNA.

RNA Extraction and Northern Analysis.

Small-scale extraction of RNA from one or two leaves was performed in the TRIzol reagent (GIBCO/BRL, Gaithersburg, MD) following the manufacturer's instructions. Northern blot analysis and synthesis of random primed probes for PR-1, BGL2 and PR-5, PDF1.2, and THI2.1 were synthesized as described (9).

Arabidopsis Transformation.

Transformation-competent artificial chromosome (TAC), bacterial artificial chromosome, pBI121, pBin19, (28) or pVK18 (29) derived clones were moved into Agrobacterium tumefaciens strains GV3101 or MP90 by electroporation and were used to transform Arabidopsis via the floral dip method (30). Selection of transformants was carried out on media containing hygromycin or kanamycin.

Expression in Escherichia coli, in Vitro S-ACP Desaturase Assay, and Gas Chromatography–Mass Spectroscopy (GC-MS) Analysis.

The putative signal peptide region of SSI2 was predicted by aligning it with the protein sequence from castor bean S-ACP DES. cDNAs from both wt and ssi2 were amplified such that they lacked N-terminal 34 aa of the putative signal peptide, and the 35th aa was converted to a methionine. The cDNAs were isolated as a NcoI/EcoRI-linkered PCR products and cloned into pET-28a vector. Purification and determination of desaturase activity were carried out as described (31). Dimethyl disulfide adducts of fatty acid methyl esters were prepared as described (32). Methyl esters of unsaturated FA and their dimethyl disulfide derivatives were identified by MS analysis (32).

Results

Positional Cloning of ssi2.

Through codominant cleaved amplified polymorphic sequence (CAPS) (26) and SSLP (27) marker analysis, the ssi2 gene was mapped to a 41-kb region of chromosome 2 that is encompassed by the bacterial artificial chromosome clone F18O19 (Fig. 1A). To identify the SSI2 gene, npr1–5 ssi2 double-mutant plants were transformed with subclones of F18O19 that had been inserted into a binary bacterial artificial chromosome vector (33). Alternatively, these plants were transformed with overlapping clones from a transformation-competent artificial chromosome (TAC) library (34) that hybridized to a 2-kb PCR-generated probe corresponding to ORF 4 within the 41-kb region. Transformants were screened for restoration of the wt morphology and the absence of constitutive PR-1 gene expression. Only TAC clone F23 complemented the ssi2 mutation (Fig. 1 B and C). Furthermore, in 105 T₂ progeny from five independently derived F23-transformed T₁ lines, the presence or absence of the hygromycin-selectable marker correlated with the development of the SSI2 or the ssi2 phenotype, respectively (Fig. 1C).

Figure 1.

Isolation of the SSI2 gene. (A) The locations of several recombination break points identified by CAPS analysis are designated by X. Four ORFs in the 11.7-kb region are numbered and marked by arrowheads. The number of transformants obtained with B11 and F23 clones is shown in parentheses. (B) The morphological phenotype of T₂ transgenic plants complemented by the SSI2 gene in comparison with that of the ssi2 mutant. (C) Northern blot analysis showing PR-1 gene expression in the ssi2 mutant, SSI2 (Nö), and T₁ and T₂ progeny of the F23 complemented transgenic ssi2 plant. (D) Derived-CAPS analysis of same set of plants shown in C. (E) Approximately 50–60 plants of wt, ssi2 and T₂ progeny of F23 transformed ssi2 plants were spray inoculated with P. parasitica spores as described (57). Plants were sampled 7 days after inoculation and scored as susceptible if they developed 10 or more sporangiophores per cotyledon. Cotyledons of SSI2 or ssi2/ssi2∷SSI2 plants showed an average of 30–40 sporagiophores per cotyledon, and 95% of these plants were susceptible. In contrast, only 5% of ssi2 plants were susceptible, and they developed 2- to 4-fold fewer sporangiophores per cotyledon. Fungal structures and hypersensitive response-like cell death were visualized by Trypan blue staining (23). The dark-staining, round spots on SSI2 and ssi2/ssi2∷SSI2 leaves are sporangiophores, and the dark-staining specks on ssi2 are dead host cells.

Based on the complementation and recombination analyses, the SSI2-containing region of F23 was reduced to 11.7 kb. This region contains four ORFs, which were amplified by PCR and sequenced. Comparison with sequences from wt Nö plants revealed only one difference, a C-to-T transition detected in ORF2. Because this variation between wt and ssi2 sequences could not be distinguished by restriction enzyme polymorphism, a derived-CAPS marker (35) was used to confirm the identity of the ssi2 mutation. Analysis of 63 T₂ progeny from the ssi2/ssi2∷SSI2 complementing lines showed that stunted growth and constitutive PR-1 gene expression cosegregated with the ssi2-specific band pattern (Fig. 1 C and D). The presence of the SSI2 gene also correlated with a loss of ssi2-induced resistance to P. parasitica Emco5; those plants containing the hygromycin marker gene were as susceptible as the wt controls, whereas those lacking the marker were resistant (Fig. 1E). Final confirmation that the SSI2 gene was isolated came from the demonstration that both a genomic clone and a cauliflower mosaic virus 35S promoter-driven cDNA clone of ORF2 restored wt morphology to ssi2 plants (data not shown).

The ssi2 Mutant Exhibits Reduced S-ACP DES Activity.

Sequence analysis predicted that SSI2 encodes an S-ACP DES, which is a prototypical member of the soluble FA desaturase enzyme family (36). These enzymes are key regulators of FA desaturation; S-ACP DES preferentially desaturates stearoyl-ACP (18:0-ACP) between carbons 9 and 10 to yield oleoyl-ACP (18:1^Δ9-ACP). The C-to-T mutation in ssi2 changes the leucine (L) at amino acids position 146 to a phenylalanine (F). Comparison of 24 S-ACP DES proteins from various plants revealed that all, except ssi2, contain a leucine at this position (Fig. 2A). The high degree of conservation for L₁₄₆, combined with the recessive nature of the ssi2 mutation, suggested that ssi2 might have reduced and/or altered S-ACP DES activity. To assess this possibility, we first determined whether bacterially expressed SSI2 exhibited S-ACP DES activity in an in vitro assay (31). Characteristic of S-ACP DES, the specific activity of wt SSI2 was ≈800 nm/min/mg, and its substrate preference for 18-carbon versus 16-carbon chain length FAs was 88:1 (Fig. 2B). GC-MS analysis confirmed the regiospecificity as Δ⁹ (Fig. 2C). In contrast to the wt protein, ssi2 was ≈10-fold and 20-fold less active on both 18:0 and 16:0 substrates, respectively (Fig. 2B). However, its 18:16 substrate preference ratio and the Δ⁹ regiospecificity were unaltered (Fig. 2C). Together, these results indicate that the wt and mutant enzymes display S-ACP DES activity, but that of ssi2 is substantially reduced.

Figure 2.

SSI2 encodes an S-ACP DES with reduced activity. (A) A 60-aa region containing residues 121–180 was compared between S-ACP DES proteins from various plant and bacterial species. Variable amino acids are boxed; the mutated amino acids in ssi2 is marked by an asterisk. (B) Enzymatic studies were carried out with a nearly homogeneous preparation of bacterial-expressed SSI2 (Nö) and mutant proteins. Desaturase activity was determined with the use of either 18:0 or 16:0 as a substrate (31). (C) GC-MS analysis of the double-bond position in the 18:1 FA methyl ester product generated by wt (I) and mutant (II) S-ACP DES. Although the scales are different for (I) and (II), the presence of 173 (X) and 217 (Y) ions is diagnostic for the two cleavage products of the derivatized 18:1 Δ⁹ unsaturated FA formed by S-ACP DES.

The FA Composition in ssi2 Plants Is Altered.

To determine whether reduced S-ACP DES activity affects the FA composition in ssi2 plants, the levels of various 16-carbon and 18-carbon FAs were monitored by GC-MS (Table 1). Leaves of the ssi2 mutant contained considerably elevated levels of 18:0 compared with the wt and decreased levels of 16:3, 18:1, and 18:2 (Table 1). The levels of other FAs, including 18:3, were similar to or slightly reduced from those observed in wt plants. The presence of nearly wt levels of these FAs in ssi2 plants is likely because of the activity of other S-ACP DES isoforms, several of which have been identified in Arabidopsis (37).

Table 1.

Fatty acid composition of total leaf lipids from wt and ssi2

Fatty acid	wt	ssi2
16:0	19.9  ± 1.0	18.1  ± 0.7
16:1-trans	2.7  ± 0.1	2.2  ± 0.3
16:1-cis	0.1  ± 0.1	0.2  ± 0.1
16:2	0.3  ± 0.0	0.2  ± 0.0
16:3	9.9  ± 0.7	6.3  ± 0.2
18:0	1.1  ± 0.1	13.4  ± 1.7
18:1	2.7  ± 0.1	0.9  ± 0.2
18:2	18.1  ± 0.4	14.9  ± 0.6
18:3	44.8  ± 1.0	43.5  ± 2.0

All measurements were made on 22°C grown plants, and data are described as mol% ± standard error calculated for a sample size of six. 

Activation of Some JA-Inducible Defense Responses Is Impaired in ssi2 Plants.

S-ACP DES catalyzes the first step in the pathway from stearic acid (18:0) to linolenic acid (18:3), and linolenic acid is a precursor for the defense signaling molecule JA (38). Because JA is required to activate the wounding response and defenses against insect pests (39) and certain microbial pathogens (40), we monitored SSI2 gene expression after wounding, pathogen infection, or treatment with SA, JA, or ethylene. Analysis of transgenic plants expressing β-glucuronidase (GUS) driven by the SSI2 promoter revealed that this promoter is active in all tissues studied, with the highest level of expression detected in flowers (Fig. 3A). Northern analysis further indicated that SSI2 gene expression was not affected by the ssi2 or npr1–5 mutations or the presence of the NahG transgene, which encodes salicylate hydroxylase (Fig. 3B and data not shown). It also did not increase substantially over basal levels at 12, 24, or 48 h after plants were treated with SA, JA, ethylene, wounding, or infection with turnip crinkle virus (data not shown; ref. 9).

Figure 3.

Expression of the SSI2 gene. (A) Histochemical staining of GUS activity in the leaves and inflorescence of transgenic plants expressing an SSI2∷GUS reporter gene. A 1,631-bp fragment containing the SSI2 promoter was transcriptionally fused upstream of GUS in pBI121, and three independent transgenic lines were analyzed in both T₁ and T₂ generations. The control is a stained leaf from a wt plant. (B) Northern blot analysis of SSI2 (Nö), npr1–5, jar1–1, and ssi2 plants treated with water or 50 μM JA. RNA was extracted 48 h after treatment, and the blot was sequentially probed with SSI2, PDF1.2, and THI2.1. Ethidium bromide-stained rRNA served as a control for gel loading. (C) Northern blot analysis of plants inoculated with spores of A. brassicicola. Mock (M) or fungal (A) inoculations were carried out as described (12). RNA was extracted 72 h after inoculation, and PDF1.2 gene expression was monitored. (D) Northern blot analysis of SSI2, npr1–5, jar1–1, NahG, etr1–1, and ssi2 plants treated with methanol or 50 μM MeJA. The plants were placed around a beaker containing methanol or MeJA diluted in methanol and covered with plastic wrap. RNA was prepared from leaves harvested 48 h after treatment and analyzed for PDF1.2 gene expression.

The ability of ssi2 plants to activate various JA-dependent defense responses was then assessed. Although JA treatment activated PDF1.2 expression effectively in wt and npr1–5 plants, it induced only low to undetectable levels of PDF1.2 expression in ssi2 NPR1, ssi2 npr1–5, or JA-insensitive jar1–1 mutant plants (Fig. 3B). In contrast, JA-induced activation of the THI2.1 gene (15) and inhibition of root growth by JA or its derivative methyl JA (MeJA) (41) were unaffected in ssi2 plants (Fig. 3B and data not shown). Inoculation with A. brassicicola induced strong expression of PDF1.2 in wt and npr1–5 plants, but only little to no expression in ssi2 plants (Fig. 3C). Because loss of PDF1.2 inducibility could be because of antagonism by the elevated SA levels found in ssi2 mutants (25), we analyzed PDF1.2 expression in ssi2 nahG plants. The presence of the NahG transgene did not restore wt levels of PDF1.2 expression in MeJA-treated (Fig. 3D) or A. brassicicola-inoculated ssi2 NPR1 or ssi2 npr1–5 plants (data not shown). Thus, the reduction in PDF1.2 inducibility in ssi2 nahG plants is not because of elevated SA levels. Because PDF1.2 expression depends on concomitant activation of the ethylene and JA signaling pathways (17), we also attempted to determine whether ethylene signaling is altered in the ssi2 mutant. A treatment of 10 or 20 parts per million of ethylene induced PDF1.2 expression in wt plants, but not in ssi2 plants (data not shown). However, ssi2 plants were highly susceptible to infection by A. brassicicola, which is pathogenic on JA-insensitive but not ethylene-insensitive mutants (42). Based on this result, the mutation in S-ACP DES does not appear to perturb the ethylene signaling pathway.

In addition to PDF1.2 expression, resistance to B. cinerea, which is mediated by JA- and ethylene-dependent pathways, was impaired in ssi2 NPR1 and ssi2 npr1–5 plants (Fig. 4). Exogenously applied JA or MeJA failed to restore B. cinerea resistance on ssi2 or ssi2 nahG plants. Indeed, the symptoms exhibited by these plants were as severe as those displayed by jar1–1 mutants. In contrast, ethylene-insensitive etr1–1 plants displayed moderate symptoms, and wt, npr1–5, and NahG transgenic plants were resistant.

Figure 4.

Analysis of disease resistance to B. cinerea. Infections with B. cinerea were carried out by wounding the leaves by needle pricks and subsequently spot inoculating spores at the wounded site (12). The number of pricks made per leaf was based on the leaf size and ranged from three per leaf for SSI2 (Nö) to one per leaf for the ssi2 mutant. Plants were treated with either water or 50 μM JA for 48 h before and throughout the infection, and the inoculated leaves were photographed at 10 dpi.

JA Plus 18:1 Induces PDF1.2 Expression in ssi2 nahG Plants.

A likely explanation for the failure of JA to activate PDF1.2 and resistance to B. cinerea in ssi2 nahG plants is that certain JA-dependent responses require a second signal that is generated by S-ACP DES. ssi2 or ssi2 nahG plants would lack or have reduced levels of this coactivating signal. Consistent with this hypothesis, treatment of ssi2 nahG plants with a combination of JA and 18:1 activated PDF1.2 (Fig. 5). ssi2 plants failed to respond to JA plus 18:1 (which is reduced 3-fold in ssi2), probably because of antagonistic effects of the high levels of endogenous SA.

Figure 5.

Complementation of JA-dependent PDF1.2 expression in 18:1-treated ssi2 nahG plants. Oleic acid (18:1; 0.5 mM; Sigma) or water was injected into the leaves of SSI2 (Nö), ssi2, or ssi2 nahG plants followed by treatment with 50 μM JA or water. Eight to ten individual plants each of SSI2, ssi2, or ssi2 nahG were analyzed in two independent experiments. RNA was extracted 48 h after treatment, and PDF1.2 gene expression was monitored by Northern blot analysis. Ethidium bromide-stained rRNA served as a control for gel loading.

Discussion

The recessive ssi2 mutation was identified as a suppressor of the npr1–5 allele. In this paper, we describe the cloning and characterization of the SSI2 gene. Based on sequence analysis and biochemical assays, we demonstrate that SSI2 encodes S-ACP DES. This enzyme, along with other soluble FA desaturases, is a key determinant of the overall level of unsaturated FAs. Analyses of the ssi2 protein revealed that its substrate preference and regiospecificity were unaltered; however, its activity was 10- to 20-fold lower than that of the wt enzyme. Consistent with this finding, the 18:0 FA content was elevated in ssi2 plants, and the 16:3, 18:1, and 18:2 contents were reduced. The composition of 16:0, 16:1, 16:2, and 18:3 in ssi2 plants was similar to or only slightly reduced from that of wt plants, presumably because of the activity of other S-ACP DES isoforms.

Previous studies have demonstrated that Arabidopsis carrying the fab2 mutation has a stunted morphology and contains substantially elevated levels of 18:0 (43). Based on these results, it was proposed that fab2 plants contain a defect in S-ACP DES, and that the fab2 mutation causes stunted growth by increasing the saturation of membrane lipids, which reduces membrane fluidity and thereby inhibits cell expansion (44). Supporting this possibility, elevated temperatures were found to substantially correct the dwarf phenotype without lowering the 18:0 content. Because of the phenotypic similarities between the fab2 and the ssi2 mutants, we sequenced the SSI2 gene from fab2 plants and found it to be a null allele; a point mutation from G to A results in a truncated protein of 172 aa (data not shown).

Because S-ACP DES catalyzes a desaturation step that is required for JA biosynthesis, we attempted to determine whether the induction of JA-dependent defense responses is affected in ssi2 plants. Both resistance to B. cinerea and induction of PDF1.2 expression were found to be impaired. One explanation for this result is that the ssi2 mutation impairs S-ACP DES activity, which reduces JA biosynthesis and thereby blocks the signaling pathway leading to these defense responses. However, this possibility seems unlikely because the level of JA precursor 18:3 in ssi2 plants was comparable to that in wt plants. In addition, some JA-induced responses, including THI2.1 expression and root growth inhibition, were unaffected in ssi2 mutant plants. Moreover, exogenously supplied JA neither activated PDF1.2 expression nor enhanced resistance to B. cinerea.

Another explanation for the loss of PDF1.2 inducibility and B. cinerea resistance is that the elevated levels of SA found in ssi2 plants (25) inhibit the JA signaling pathway. Previous studies have demonstrated that these signals are mutually antagonistic (18, 21, 22). This hypothesis could explain why JA treatment failed to activate PDF1.2 expression or restore B. cinerea resistance. However, the presence of the NahG transgene did not restore wt levels of PDF1.2 expression in MeJA- or pathogen-treated ssi2 plants; it also did not restore resistance to B. cinerea. Furthermore, SA-mediated inhibition of JA signaling should have blocked all JA-dependent responses, rather than just a subset.

A more likely explanation for our results is that activation of certain JA-dependent responses requires a second signal that is generated by S-ACP DES. Because ssi2 mutants would lack or have depressed levels of this coactivating signal, JA treatment would be insufficient to activate PDF1.2 expression or restore resistance to B. cinerea. In contrast, activation of strictly JA-dependent responses, such as THI2.1 and root growth inhibition, would remain unimpaired. Supporting this possibility is the discovery that injecting 18:1 into the leaves of ssi2 nahG plants restores JA-inducible PDF1.2 expression. The inability of 18:1 to rescue PDF1.2 expression in ssi2 plants is likely because of the high endogenous SA levels, which could antagonize the action of JA. These results also suggest that 18:1 or an 18:1-derived signal works in conjunction with JA to induce JA-dependent defense gene expression and pathogen resistance.

In addition to lacking certain JA-induced defenses, ssi2 plants exhibit constitutive expression of several SA-associated defense responses. Because pathogen infection of wt plants generally induces the expression of either PDF1.2 or the PR genes, our results suggest that components of the FA desaturation pathway may cross-regulate the activation of these defenses. It is possible that the coactivating signal inhibits the NPR1-independent pathway; loss of this signal in ssi2 plants would allow constitutive activation of the NPR1-independent responses. Alternatively, the ratio of saturated versus unsaturated FAs or changes in their subcellular distribution might regulate crosstalk between defense signaling pathways. For example, an increase in 18:0 content might lead to activation of lipid signaling, which could then induce the PR signal transduction pathway (45). Increases in unsaturated FAs also could stimulate (46) or inhibit (47) protein phosphatase(s) activity, which might then alter protein kinase- or mitogen-activated protein kinase-regulated pathway(s), respectively. Interestingly, an Arabidopsis mutant defective in the mitogen-activated protein kinase mpk4 exhibits a phenotype similar to that of ssi2, including constitutive PR gene expression and suppressed PDF1.2 expression (48). Perhaps reduced or altered unsaturated FA levels in the ssi2 mutant relieve inhibition of phosphatase activity, which then results in inhibition of a mitogen-activated protein kinase (MPK4) pathway that negatively controls SA signaling and positively regulates JA signaling. The possibility that a decrease in S-ACP DES activity simply causes SA-mediated stress and PR gene expression is ruled out because the ssi2 phenotypes were seen in ssi2 nahG plants (25). Likewise, the possibility that stress because of high FA levels induces constitutive PR gene expression seems unlikely because fad2 mutants, which accumulate elevated levels of 18:1 (49) and fad5 and fab1 mutants, which contain high levels of 16:0 (50), do not show any of the phenotypes displayed by ssi2 plants (data not shown). Furthermore, exogenous application of 18:0 does not induce PR-1 gene expression in wt plants (data not shown).

Our results reveal intriguing parallels between the roles of FA signal signaling in mammals and plants. In mammals, FAs serve as an important energy source; they are also involved in various signal transduction pathways in different tissues, including those of the heart (51–53). In particular, altered stearoyl CoA desaturase activity has been implicated in the regulation of cell growth, differentiation, and signal transduction (53–54). Furthermore, altered activity of this enzyme is correlated with apoptosis (55) and neoplasia (56); these responses are similar to the cell death and altered defense signaling phenotypes associated with the ssi2 mutation. Thus, although the mechanism(s) through which a mutation in SSI2 affects defense response activation in plants is unclear, the discovery that a defect in FA desaturation can modulate these responses opens approaches to studying the pathways leading to disease resistance.

Abbreviations

SA

salicylic acid

PR

pathogenesis-related

SAR

systemic acquired resistance

TAC

transformation-competent artificial chromosome

GC-MS

gas chromatography–mass spectroscopy

GUS

β-glucuronidase

FA

fatty acid

S-ACP DES

stearoyl-ACP desaturase

wt

wild type

JA

jasmonic acid

MeJA

methyl JA

Nö

Arabidopsis ecotype Nössen

CAPS

cleaved amplified polymorphic sequence