Resistance to Botrytis cinerea Induced in Arabidopsis by Elicitors Is Independent of Salicylic Acid, Ethylene, or Jasmonate Signaling But Requires PHYTOALEXIN DEFICIENT3  

Abstract

Oligogalacturonides (OGs) released from plant cell walls by pathogen polygalacturonases induce a variety of host defense responses. Here we show that in Arabidopsis (Arabidopsis thaliana), OGs increase resistance to the necrotrophic fungal pathogen Botrytis cinerea independently of jasmonate (JA)-, salicylic acid (SA)-, and ethylene (ET)-mediated signaling. Microarray analysis showed that about 50% of the genes regulated by OGs, including genes encoding enzymes involved in secondary metabolism, show a similar change of expression during B. cinerea infection. In particular, expression of PHYTOALEXIN DEFICIENT3 (PAD3) is strongly up-regulated by both OGs and infection independently of SA, JA, and ET. OG treatments do not enhance resistance to B. cinerea in the pad3 mutant or in underinducer after pathogen and stress1, a mutant with severely impaired PAD3 expression in response to OGs. Similarly to OGs, the bacterial flagellin peptide elicitor flg22 also enhanced resistance to B. cinerea in a PAD3-dependent manner, independently of SA, JA, and ET. This work suggests, therefore, that elicitors released from the cell wall during pathogen infection contribute to basal resistance against fungal pathogens through a signaling pathway also activated by pathogen-associated molecular pattern molecules.

Plants need to recognize invading pathogens in a timely manner to mount appropriate defense responses. In the so-called gene-for-gene resistance, early recognition of specific pathogen strains depends on complementary pairs of dominant genes, one in the host and one in the pathogen. The outcome of this recognition is the induction of a multitude of biochemical and physiological changes, including localized programmed cell death (hypersensitive response), that restrict pathogen growth in the host tissues. A loss of or a mutation in either the plant resistance gene or in the pathogen avirulence gene leads to a compatible interaction, resulting in disease (Flor, 1971). Gene-for-gene resistance has been observed in interactions with many biotrophic pathogens, including fungi, viruses, bacteria, and nematodes (Hammond-Kosack and Jones, 1997). In contrast, many necrotrophic fungal and bacterial pathogens cause disease in a variety of plant species and resistance mediated by a single host resistance gene is uncommon. Nevertheless, plants also recognize nonspecific elicitors that activate a battery of defense responses effective against a wide range of pathogens. Some of these elicitors, referred to as pathogen-associated molecular patterns (PAMPs), are derived from essential components of the pathogen cell wall (e.g. chitin, glucan) or other macromolecular structures (e.g. the 22-amino acid flagellin peptide flg22; for review, see Nurnberger and Brunner, 2002).

Pathogen-secreted hydrolytic enzymes that degrade host cell wall polymers are also able to induce defense responses in plants. Among these, the most extensively studied are endopolygalacturonases (PGs; EC 3.2.1.15). PGs cleave the α-(1 → 4) linkages between d-GalA residues in nonmethylated homogalacturonan, a major component of pectin (De Lorenzo et al., 1997). PGs are important virulence factors for necrotrophic soft rot-causing pathogens, including the fungus Botrytis cinerea (ten Have et al., 1998). The elicitor activity of PGs in activating plant defense responses has been demonstrated in many pathosystems (De Lorenzo et al., 1997). Hahn and colleagues first showed that PGs are not directly responsible for the induction of plant defense responses, but rather cause the release from the plant cell wall of the true elicitors, namely oligogalacturonides (OGs) with a degree of polymerization between 10 and 15 (OGs; Hahn, 1981). A functional catalytic site is required for elicitor activity of a Colletotrichum lindemuthianum PG in tobacco (Nicotiana tabacum; Boudart et al., 2003), supporting the hypothesis that OGs mediate responses activated by PGs. In contrast, Poinssot and colleagues reported that enzymatic activity is not required for elicitor activity of the B. cinerea PG BcPG1 in grape (Vitis vinifera) cells (Poinssot et al., 2003), suggesting that in some biological systems, PGs themselves can be perceived and activate defense responses.

OGs elicit a variety of defense responses, including accumulation of phytoalexins (Davis et al., 1986), glucanase and chitinase (Davis and Hahlbrock, 1987; Broekaert and Pneumas, 1988), and Phe ammonia lyase (PAL; De Lorenzo et al., 1987). Exogenous treatments with OGs protect grapevine leaves against B. cinerea infection in a dose-dependent fashion (Aziz et al., 2004). Despite having been extensively studied, the role of OGs in plant defense and their mode of action are still largely unknown.

A variety of plant defense responses against microbial pathogens are regulated by the signaling molecules salicylic acid (SA), jasmonates (JAs), and ethylene (ET; for review, see Feys and Parker, 2000). These signaling pathways have been exensively studied, but a major unanswered question is how the SA, JA, and ET signaling pathways are related to the signaling pathways activated by OGs and other PAMPs. Although mutants impaired in the responses mediated by SA, JA, and ET show enhanced disease symptoms upon infection with B. cinerea (Thomma et al., 1998, 1999; Alonso et al., 2003; Ferrari et al., 2003a), treatment with OGs or infection with B. cinerea induce the expression of AtPGIP1, which encodes an Arabidopsis (Arabidopsis thaliana) inhibitor of fungal PGs, independently of these secondary signaling molecules (Ferrari et al., 2003b). Since expression of AtPGIP1 is required for full resistance to B. cinerea (Ferrari et al., 2006), we speculated that a SA-, JA-, and ET-independent defense pathway induced by OGs during fungal infection may actively contribute to plant defense. To investigate this hypothesis, we performed global transcription profiling and susceptibility assays in Arabidopsis plants treated with exogenous elicitors or inoculated with B. cinerea. Our results suggest that the expression of many defense-related genes is induced by OGs released during pathogen infection and that responses induced independently of SA, JA, and ET, and in particular the expression of genes involved in the production of antimicrobial compounds, participate to restrict the growth of B. cinerea.

RESULTS

OGs Induce Local and Systemic Resistance to B. cinerea

To assess the ability of OGs to enhance Arabidopsis resistance to fungal infection, adult plants were sprayed with OGs and inoculated with B. cinerea 24 h after the treatment. Leaves from plants treated with OGs showed significantly delayed progression of the infection, compared to control-treated plants (Supplemental Fig. S1; Fig. 1A  

Figure 1.

Induction of resistance to B. cinerea by OGs in Arabidopsis plants. A, Lesion development in Arabidopsis Col-0 plants inoculated with B. cinerea 24 h after treatment with a control solution (white circles) or with OGs (black circles). Lesion areas were measured at the indicated times. B, Lesion development in systemic leaves of wild-type plants pretreated with OGs. Lower leaves were infiltrated with OGs (black bars) or water (white bars) and upper, untreated leaves were collected after 72 h and inoculated with B. cinerea. Lesion areas were measured 48 h after inoculation. Values are means ± se of at least 12 lesions. Asterisks indicate statistically significant differences between control and OG-treated plants, according to Student's t test (*, P < 0.05; ***, P < 0.01).

). Similar results were obtained when plants were inoculated 48 or 72 h after OG treatment (data not shown). To determine whether OGs also induce systemic resistance, lower rosette leaves were injected with water or OGs and, after 72 h, upper leaves were inoculated with B. cinerea. The average area of necrotic lesions in systemic leaves of plants pretreated with OGs was significantly smaller than in control plants (Fig. 1B). These results indicate that OGs induce both local and systemic resistance against Botrytis infection in Arabidopsis plants.

Role of SA, JA, and ET in OG-Mediated Resistance

To determine the role of SA, JA, or ET in OG-induced resistance, we sprayed OGs on mutants or transgenic plants affected in the perception or transduction of these hormone signals and inoculated them with B. cinerea 24 h later. The Arabidopsis lines used were transgenic plants expressing the salicylate hydroxylase nahG gene and therefore unable to accumulate SA (Gaffney et al., 1993), plants carrying single mutations in the NONEXPRESSOR OF PATHOGENESIS-RELATED GENES1 (NPR1), ETHYLENE-INSENSITIVE2 (EIN2), or JASMONATE RESISTANT1 (JAR1) genes impaired in signaling mediated by SA, ET, or JA, respectively (Guzman and Ecker, 1990; Staswick et al., 1992; Cao et al., 1994), or a triple mutant carrying a mutation in each of these genes (npr1ein2jar1) and therefore unable to respond to any of the hormones (Clarke et al., 2000). As described previously (Thomma et al., 1999; Ferrari et al., 2003a), ein2 mutant and nahG transgenic plants are more susceptible to B. cinerea. Nevertheless, OG pretreatments of these mutants, as well as of the other mutants tested, significantly reduced lesion size (Fig. 2, A and B  

Figure 2.

Induction of resistance to B. cinerea by OGs in mutants impaired in SA, JA, or ET signaling. A, Lesion area in Arabidopsis Col-0 (WT), ein2, and nahG plants treated with a control solution (white bars) or OGs (black bars) and inoculated with B. cinerea 24 h after treatment. Lesion areas were measured 48 h after inoculation. B, Lesion area in Arabidopsis Col-0 (WT), jar1, npr1, and npr1ein2jar1 (nej) plants treated with a control solution (white bars) or OGs (black bars) and inoculated with B. cinerea 24 h after treatment. Lesion areas were measured 48 h after inoculation. C, Lesion area in Arabidopsis Col-0 (WT) or homozygous coi1 plants treated with a control solution (white bars) or OGs (black bars) and inoculated with B. cinerea 24 h after treatment. Lesion areas were measured 48 h after inoculation. Values are means ± se of at least 12 lesions. Asterisks indicate statistically significant differences between control and OG-treated plants, according to Student's t test (*, P < 0.05; ***, P < 0.01).

). To further confirm that OG-mediated resistance is JA independent, we also treated B. cinerea-susceptible coronatine-insensitive1 (coi1) homozygous mutant plants (Thomma et al., 1998) with OGs. Two days after inoculation, lesions in control coi1 plants were more than 5-fold larger than in wild-type control plants; however, OG-pretreated coi1 plants were significantly more resistant to B. cinerea, similar to the highly susceptible ein2 mutant (Fig. 2C). Taken together, these results indicate that OGs increase Arabidopsis resistance to B. cinerea through the activation of defense responses that are independent of SA, JA, and ET, but that SA, JA, and ET are also involved in defense pathways that confer resistance to B. cinerea.

Changes in Gene Expression in Arabidopsis Plants Treated with OGs or Inoculated with B. cinerea

To identify Arabidopsis genes that may be involved in OG-mediated resistance to B. cinerea, we analyzed the transcriptome of plants inoculated with the fungus and compared it to the transcription profile of 10-d-old seedlings treated with OGs for 1 or 3 h. For the analysis of infected plants, rosette leaves were inoculated with a B. cinerea spore suspension or with sterile medium and harvested after 18 or 48 h. At the early time point of infection, no macroscopic lesions were observed at the site of inoculation, though staining of fungal hyphae with trypan blue revealed that the spores had germinated and started growing on the leaf surface (data not shown). After 48 h, the infected leaves showed water-soaked lesions about 3 to 4 mm in width that are typical of soft rot disease. Total RNA from control or treated samples from two or three independent infection or elicitor-treatment experiments, respectively, was analyzed using the Affymetrix ATH1 GeneChip DNA microarray, which contains probe sets corresponding to more than 22,000 putative open reading frames. Original raw data for each experiment are available at the Nottingham Arabidopsis Stock Centre (http://affymetrix.arabidopsis.info/narrays/experimentbrowse.pl) and http://ausubellab.mgh.harvard.edu/imds (under experiment names “Botrytis cinerea infection, 18 and 48 hpi” and “Comparison of response to Flg22 and OGs elicitors”). For each probe set, mean expression fold-change, signal intensities, and P values were calculated after normalization (see Supplemental Tables S1 and S2). Only probe sets showing a statistically significant difference (P ≤ 0.01) between control and experimental treatments and a mean fold-change ≥2.0 were considered for further analysis.

As summarized in Table I  

Table I.

Number of genes showing altered expression in response to B. cinerea infection

Time	Induceda	Represseda
h
18	153	1
48	1,942	2,871

a

Number of genes reproducibly showing ≥2.0-fold induction or repression in all the analyzed replicate samples (P ≤ 0.01).

, at 48 h post inoculation (hpi), when typical water-soaked lesions were visible, more than 10 times as many genes showed significant changes in transcription than at 18 hpi. Interestingly, 588 out of 1,299 genes up-regulated by OGs after either 1 h or 3 h were also up-regulated by B. cinerea at either of the two time points analyzed (Fig. 3  

Figure 3.

Overlap between OG- and fungal infection-induced transcriptional changes. Venn diagram of the number of overlapping and nonoverlapping genes in response to OGs or B. cinerea infection. OG-up, Genes induced 2.0-fold or more (P ≤ 0.01) after 1 or 3 h of treatment with OGs; OG-down, genes repressed 2.0-fold or more (P ≤ 0.01) after 1 or 3 h of treatment with OGs; Bc-up, genes induced 2.0-fold or more (P ≤ 0.01) after 18 or 48 h of inoculation with B. cinerea; Bc-down, genes repressed 2.0-fold or more (P ≤ 0.01) after 18 or 48 h of inoculation with B. cinerea. In parentheses is indicated the total number of genes belonging to each category.

). Similarly, 316 out of 577 genes significantly repressed by OGs at either 1 h or 3 h were also down-regulated during fungal infection, indicating that about half of the genes responsive to OGs show a significant change of expression in the same direction upon B. cinerea attack (Fig. 3; a complete list of the genes coregulated by OGs and B. cinerea can be found in Supplemental Table S3).

We have also compared the transcript profile of OG-treated seedlings to the available data on the profile induced by crab shell chitin and chitin octamers (Ramonell et al., 2005), using the same threshold (expression change ≥1.5-fold). Out of 1,002 genes whose expression is induced by both crab shell chitin and octamers after 30 min of treatment, 697 were also up-regulated by OGs after either 1 or 3 h, whereas 84 genes out of 312 chitin-repressed genes were significantly down-regulated by OGs (Supplemental Table S4). In contrast, only six out of 68 genes induced specifically by octamers and 39 out of 238 genes repressed only by octamers were also induced or repressed by OGs, respectively.

To group genes up-regulated by both OGs and B. cinerea according to their predicted functions, we identified Functional Catalogue (FunCat) terms (Ruepp et al., 2004) associated with each gene using the Munich Information Center for Protein Sequences (MIPS) Arabidopsis Data Base (MAtDB; http://mips.gsf.de/proj/thal/db/index.html). For this analysis, only probes with an annotated locus identifier were used. For each expression category, frequencies of genes in a given FunCat group were compared with the frequency found for all genes represented on the array (Table II  

Table II.

Functional categories significantly overrepresented among OG- and B. cinerea-induced genes

Functional Categorya	Gene Matchesb	Totalc	P Valued
Cell rescue, defense, and virulence	35 (6.0)	567 (2.1)	1.2 × 10−6
Classification not yet clear cut	50 (8.6)	1,238 (4.6)	2.5 × 10−5
Cellular communication/signal transduction	35 (6.0)	825 (3.1)	1.7 × 10−4
Metabolism	57 (9.8)	1,722 (6.5)	1.2 × 10−3
Amino acid metabolism	13 (2.2)	237 (0.9)	2.0 × 10−3
Metabolism of the Cys-aromatic group	4 (0.7)	29 (0.1)	3.5 × 10−3
Secondary metabolism	18 (3.1)	336 (1.3)	4.9 × 10−4
Biosynthesis of phenylpropanoids	7 (1.2)	69 (0.3)	7.7 × 10−4

a

Functional categories according to MIPS FunCat database.

b

Number of genes induced ≥2.0-fold by both OGs and B. cinerea and present in the indicated category. In parentheses is indicated the percentage of total OG- and B. cinerea-induced genes that match the category.

c

Total number of Arabidopsis genes in the indicated functional category. In parentheses is indicated the percentage of genes in the Arabidopsis genome that match the category. The comparison was done to the Arabidopsis thaliana MAtDB containing 26,642 annotated genes (http://mips.gsf.de/projects/funcat).

d

P value of significance of distribution of OG- and Botrytis-induced genes in the indicated category, compared to distribution in the complete Arabidopsis dataset. Only categories significantly overrepresented (P < 0.001) are indicated in the table.

). As expected, the most represented categories included genes involved in cell rescue and defense, genes whose classification is ambiguous, genes involved in cell communication and signaling, and genes encoding proteins implicated in primary and secondary metabolism.

A majority of the genes involved in secondary metabolism that are induced by OGs and Botrytis and identified through the MAtDB database encode enzymes implicated in amino acid metabolism and phenylpropanoid biosynthesis (Table II). However, we noticed that a number of genes involved in the metabolism of Trp-derived secondary compounds, like PHYTOALEXIN DEFICIENT3 (PAD3; Zhou et al., 1999) and CYP79B2 (Hull et al., 2000), were included in the not yet clear-cut category. We therefore manually compiled a list of 66 Arabidopsis genes previously implicated in secondary metabolism (full list is available in the Supplemental Table S5). Expression of 36 of these genes appeared significantly induced or repressed by at least one treatment (Table III  

Table III.

Secondary metabolism genes that change expression in response to OGs or B. cinerea infection

Descriptiona	Transcript ID	Probe Set ID	Fold-Changeb
						OG 1 h	OG 3 h	Bc 18 h	Bc 48 h
Shikimate pathway
DHS1; 2-dehydro-3-deoxyphosphoheptonate aldolase	At4g39980	252831_at	2.7	3.1		4.2
5-Enolpyruvylshikimate-3-P synthase	At1g48860	246627_s_at		2.3		2.1
CS; chorismate synthase	At1g48850	245832_at		2.8		3.0
Anthranilate biosynthesis
ASB2; anthranilate synthase β-chain	At5g57890	247864_s_at	4.7	5.0	2.8	6.0
ASA1; anthranilate synthase component I-1 precursor	At5g05730	250738_at	4.5	5.2	3.4	6.8
Anthranilate glucosylation
UGT74F2; anthranilate glucosyltransferase	At2g43820	260567_at				6.6
Tryptophan biosynthesis
IGPS; putative indole-3-glycerol phosphate synthase	At2g04400	263807_at	3.0	6.6	2.8	5.4
PAT1; anthranilate phosphoribosyltransferase	At5g17990	250014_at		3.5	2.2	3.4
IGPS1; indole-3-glycerol phosphate synthase	At5g48220	248688_at				−3.3
IAOx biosynthesis
CYP79B2	At4g39950	252827_at	4.6	5.5	4.8	8.7
CYP79B3	At2g22330	264052_at		2.6
Glucosinolate biosynthesis (general)
UGT74B1	At1g24100	264873_at		2.5
SUR1/C-S lyase	At2g20610	263714_at		2.9
IAA metabolism
ILR1; IAA conjugate hydrolase	At3g02875	258610_at				8.7
NIT3; nitrilase	At3g44320	252677_at				4.0
Camalexin biosynthesis
PAD3/CYP71B15; camalexin biosynthesis	At3g26830	258277_at		11.5	14.6	10.1
Phenylpropanoids/1-early steps
4CL1; 4-coumarate:CoA ligase	At1g51680	256186_at	4.0
PAL1; phenylanine ammonia lyase	At2g37040	263845_at	3.6	3.5		2.8
4CL4; 4-coumarate:CoA ligase	At5g45000	248971_at	3.6
CH4/REF3; cinnamate-4-hydroxylase	At2g30490	267470_at	3.0	3.2
4CL2; 4-coumarate:CoA ligase	At3g21240	258047_at	2.7	2.7		2.8
PAL2; phenylanine ammonia lyase	At3g53260	251984_at	2.5	2.0		5.2
PAL3; phenylanine ammonia lyase	At5g04230	245690_at				−3.8
Flavonoid biosynthesis
UGT78D1; flavonol-3-rhamnosyltransferase	At1g30530	261804_at	−4.1			−2.6
UGT73C6; flavonol-7-O-glucosyltransferase	At2g36790	265200_s_at				6.3
Phenylpropanoids/2-late steps
CCR2; putative cinnamoyl-CoA reductase	At1g80820	261899_at	19.0		26.9
CCR; putative cinnamoyl-CoA reductase	At5g14700	250149_at			2.2	7.3
ELI3-2/CAD-B2/AtCAD8; cinnamyl-alcohol dehydrogenase	At4g37990	252984_at				31.5
CCOMT1; putative caffeoyl-CoA O-methyltransferase	At4g34050	253276_at		2.0
FAH1/F5H1; ferulate-5-hydroxylase 1	At4g36220	253088_at		4.8		8.0
CAD1/AtCAD9; cinnamyl-alcohol dehydrogenase	At4g39330	252943_at				−8.1
OMT1/COMT1; O-methyltransferase 1	At5g54160	248200_at		2.7		2.5
Aliphatic glucosinolate biosynthesis
UGT74C1	At2g31790	263477_at	−2.2			−6.4
CYP79F1	At1g16410	262717_s_at				−2.8
REF2/CYP83A1	At4g13770	254687_at				−5.3
Indole glucosinolate biosynthesis
SUR2/CYP83B1	At4g31500	253534_at	4.6	5.5

a

Annotation based on The Institute for Genomic Research Arabidopsis Genome Annotation Database, the MIPS Functional Categories Database (http://mips.gsf.de/projects/funcat), or the available literature (for complete list and references, see Supplemental Table S5).

b

Mean expression fold-change of probe sets is indicated only when change is significant (P ≤ 0.01) and ≥2.0, and signal intensity is ≥0.1 for at least one treatment.

). In particular, several genes encoding enzymes of the shikimate and the phenylpropanoid pathways (with the exception of flavonoid biosynthesis) and enzymes involved in the biosynthesis of Trp and of indole compounds were up-regulated by both OGs and B. cinerea, whereas most genes proposed to encode enzymes involved in the biosynthesis of aliphatic glucosinolates, like CYP79F1, REF2, and UGT74C1 (Hansen et al., 2001; Hemm et al., 2003; Gachon et al., 2005), were repressed or not significantly affected by OGs or fungal infection (Table III; Fig. 4, A and B  

Figure 4.

Expression of genes involved in secondary metabolism in response to OGs and fungal infection. The scheme summarizes the relationships between the shikimate, Trp, Phe, phenylpropanoids, flavonoids, camalexin, indole glucosinolates, and aliphatic glucosinolates biosynthetic pathways and the levels of expression of selected genes in each pathway, portrayed with MapMan software. The number of small squares next to each pathway or portion of pathway indicates how many genes present in the manually compiled list (for details, see Table III) and assigned to that pathway showed a 2.0-fold or greater change of expression (P ≤ 0.01) in response to OGs (A) at 1 h (left squares) or 3 h (right squares) or B. cinerea infection (B) at 18 h (left squares) or 48 h (right squares). Red squares represent genes showing increased expression, blue squares represent genes showing decreased expression. Color intensity indicates the extent of change, expressed as log₂ of the mean ratio between treated and control samples (see scale). Gray dots indicate that none of the genes in the pathway are significantly induced or repressed by the indicated treatment.

). Interestingly, mRNA levels of ASA1, ASB2, IGPS, CYP79B2, and PAD3, which are all involved in the biosynthesis of indolic compounds (Zhao and Last, 1996; Zook, 1998; Zhou et al., 1999; Hull et al., 2000; Glawischnig et al., 2004), increase at both 18 and 48 h after infection and by OGs after 3 h of treatment (Table III). With the exception of PAD3, which encodes the cytochrome P450 enzyme CYP71B15 that catalyzes the last step of camalexin biosynthesis (Schuhegger et al., 2006), the other genes encoding enzymes in the indole pathway may also be involved in the biosynthesis of glucosinolates or auxin. However, genes specifically required for indole-3-acetic acid (IAA) metabolism, including a nitrilase gene and ILR1, encoding an IAA-conjugate hydrolase (Bartel and Fink, 1995), were not up-regulated by OGs, although they were induced by B. cinerea (Table III; Fig. 4, A and B). Because pad3 plants are extremely susceptible to B. cinerea infection (Ferrari et al., 2003a), it appears likely that OG-mediated activation of PAD3 and other genes involved in the biosynthesis of camalexin may contribute to OG-induced resistance against B. cinerea.

PAD3 Is Expressed Independently of SA, JA, and ET and Is Required for OG-Induced Resistance against B. cinerea

Since PAD3 is required for basal resistance to B. cinerea strains sensitive to camalexin (Ferrari et al., 2003a; Kliebenstein et al., 2005) and microarray experiments showed that PAD3 expression is induced by both OGs and B. cinerea (this work), we further investigated OG-mediated induction of PAD3 expression in adult plants. A transient accumulation of PAD3 mRNA, with a peak at 3 to 6 h after treatment, was observed by quantitative reverse transcription (RT)-PCR in rosette leaves sprayed with OGs (Fig. 5A  

Figure 5.

Expression of PAD3 in response to OGs and infection. A, Adult Col-0 plants were sprayed with OGs and total RNA was extracted from rosette leaves harvested at the indicated times (hours). PAD3 expression in each sample was determined by real-time RT-PCR and normalized to the expression of UBQ5. Bars indicate average expression ± sd of two replicates, relative to the expression in untreated Col-0 plants. B, Col-0 (WT), nahG, coi1, npr1, ein2, and jar1 adult plants were inoculated with B. cinerea and total RNA was extracted from inoculated leaves at the indicated times (days post infection). PAD3 expression was determined by RNA-blot analysis. UBQ5 expression confirmed equal loading of the samples (data not shown). C, Col-0 (WT), ein2npr1jar1 (nej), and ups1 seedlings were treated with OGs and total RNA was extracted at the indicated times (hours). PAD3 expression was analyzed by real-time RT-PCR and normalized using the expression of the UBQ5 gene. Bars indicate average expression ± sd of two replicates, relative to the expression in untreated Col-0 plants. D, Col-0 (WT) and coi1 seedlings were treated with OGs and total RNA was extracted at the indicated times (hours). Expression of PAD3 and UBQ5 was analyzed by semiquantitative RT-PCR.

). Interestingly, the increase in PAD3 transcript levels after B. cinerea inoculation of leaves was not attenuated in nahG, ein2, npr1, jar1, or coi1 plants (Fig. 5B), indicating that PAD3 expression during fungal infection is also independent of SA, JA, and ET. Consistently, PAD3 mRNA levels in wild-type and npr1ein2jar1 seedlings treated with OGs for 1 and 3 h were not significantly different (Fig. 5C). OG treatments were also similarly effective in inducing PAD3 expression in single ein2, npr1, and jar1 seedlings, as determined by semiquantitative RT-PCR (data not shown). Furthermore, mRNA levels of PAD3 in coi1 seedlings treated with OGs were even higher than in wild-type seedlings (Fig. 5D). These results indicate that OGs are able to increase PAD3 expression independently of SA, JA, and ET.

Since protection against B. cinerea infection mediated by OG pretreatment is also independent of SA-, JA-, and ET-mediated signaling (Fig. 2, A–C), PAD3 may be important for OG-induced resistance to this pathogen. To test this hypothesis, we inoculated OG-treated wild-type or pad3 plants with B. cinerea. Strikingly, control and OG-treated pad3 plants showed similar lesion development, whereas, in accordance with our previous results, OG-treated wild-type plants were more resistant to infection (Fig. 6  

Figure 6.

Induction of resistance to B. cinerea by OGs in wild-type plants and in mutants impaired in camalexin production. Lesion area in Arabidopsis Col-0 (WT), pad3, and ups1 plants treated with a control solution (white bars) or OGs (black bars) and inoculated with B. cinerea 24 h after treatment. Lesion areas were measured 48 h after inoculation. Values are means ± se of at least 12 lesions. Asterisks indicate statistically significant differences between control and OG-treated plants, according to Student's t test (***, P < 0.01). The experiment was repeated three times with similar results.

). To further confirm the role of PAD3 in OG-induced resistance, we analyzed the underinducer after pathogen and stress1 (ups1) mutant, which is defective in Trp biosynthetic pathway regulation and accumulates low levels of camalexin and of PAD3 mRNA in response to fungal infection and oxidative stress (Denby et al., 2005). As shown in Figure 6, ups1 plants were not protected by OGs against B. cinerea. Consistent with our hypothesis that PAD3 expression is required for OG-induced resistance, PAD3 expression in OG-treated ups1 seedlings was strongly reduced (Fig. 5C), in contrast to the triple npr1ein2jar1 mutant. Interestingly, the bacterial flagellar peptide elicitor flg22, which is structurally unrelated to OGs, was also effective in protecting Arabidopsis wild-type plants, as well as npr1ein2jar1 triple mutant plants, against B. cinerea, and this protection was also dependent on PAD3 and UPS1 (Fig. 7, A and B  

Figure 7.

Induction of resistance to B. cinerea by flg22. A, Lesion area in Arabidopsis Col-0 (WT), ein2npr1jar1 (nej), and pad3 plants treated with a control solution (white bars) or flg22 (black bars) and inoculated with B. cinerea 24 h after treatment. B, Lesion area in Arabidopsis Col-0 (WT), pad3, and ups1 plants treated with a control solution (white bars) or flg22 (black bars) and inoculated with B. cinerea 24 h after treatment. Lesion areas were measured 48 h after inoculation. Values are means ± se of at least 12 lesions. Asterisks indicate statistically significant differences between control and flg22-treated plants, according to Student's t test (***, P < 0.01). This experiment was repeated twice with similar results.

). In one experiment we observed a slight reduction of symptoms in flg22-treated ups1 plants (data not shown), but this result could not be replicated, and in two independent experiments flg22 pretreatments failed to protect ups1 against B. cinerea infection (Fig. 7B).

Since PAD3 is required for elicitor-induced resistance against B. cinerea, we expected that OGs could increase camalexin levels in Arabidopsis adult plants. Surprisingly, however, HPLC analysis failed to reveal significant camalexin accumulation up to 24 h after treatment with the elicitor, compared to control-treated plants (data not shown). Therefore, OG-induced resistance does not seem to be directly due to an increase in the levels of camalexin in leaf tissues before pathogen inoculation.

DISCUSSION

Activation of the plant innate immune system can be mediated, as in animals, by molecular patterns conserved among different pathogen species but not present in host cells. However, in plants, some host-derived low M  _r elicitors may also be released during an infection as a consequence of pathogen activity. These endogenous elicitors, like true PAMPs, may trigger defense responses that contribute to basal resistance. OGs represent the best characterized endogenous elicitors and have been extensively studied since their identification about 25 years ago (Hahn et al., 1981). In this article we have investigated what responses activated by OGs may be involved in Arabidopsis defense against the fungal pathogen B. cinerea. Treatment with exogenous OGs enhances resistance against B. cinerea in grape (Aziz et al., 2004) and Arabidopsis leaves (this work). Since B. cinerea secretes large amounts of polygalacturonases during tissue invasion (for review, see Kars and van Kan, 2004), it is likely that OGs transiently accumulate at the interface between fungal hyphae and plant tissues. However, detection of OGs in the apoplast of infected plants is technically challenging. Recently, An and colleagues have employed matrix-assisted laser-desorption ionization-Fourier transform and matrix-assisted laser-desorption ionization-time of flight mass spectrometry to identify cell wall pectin-derived oligosaccharides generated through the breakdown of homogalacturonan pectins in B. cinerea-infected tomato (Solanum lycopersicum) fruits (An et al., 2005). This study suggests that pectic fragments with a degree of polymerization of about 16 accumulate around the lesions caused by this fungus. However, a conclusive demonstration that OGs with elicitor activity accumulate to significant levels during infection is still lacking, and their role in basal resistance to pathogens must be inferred from indirect evidence.

We have carried out a full-genome expression analysis of Arabidopsis plants treated with OGs or infected with B. cinerea to determine the extent of overlap between transcriptional responses induced by these two treatments. Previously, Navarro and colleagues, comparing microarray data obtained from flg22-treated cell cultures and seedlings, and data obtained from bacterial-inoculated rosette leaves, found a limited overlap (7%) in the compatible interaction and a more consistent overlap (34%) with an incompatible interaction (Navarro et al., 2004). Our results indicate that about half of the Arabidopsis genes affected by OG treatment display a similar behavior after fungal infection, suggesting that at least part of the responses activated by B. cinerea are mediated, directly or indirectly, by the accumulation of OGs or other elicitors able to activate the same signaling pathway, and that these responses are not suppressed by the fungus. The genes induced by both OGs and B. cinerea infection not only encode defense-related proteins, but also enzymes implicated in primary and secondary metabolism. The transcriptional activation of some of these genes during fungal infection may result in the accumulation of antimicrobial proteins or low M  _r compounds, which are able to restrict fungal growth. In the case of B. cinerea, it has been previously shown that strains sensitive to camalexin are partially restricted in planta by this phytoalexin, since pad3 and other mutants impaired in camalexin accumulation are more susceptible to infection than the parental lines (Ferrari et al., 2003a; Kliebenstein et al., 2005).Camalexin may therefore play a major role in the reduction in B. cinerea growth observed in OG-treated plants. Furthermore, OG-induced resistance and PAD3 expression are both largely independent of SA, ET, and JA. OGs (and B. cinerea) therefore appear to activate an SA-, ET-, and JA-independent signaling pathway that regulates PAD3 expression and other defense responses effective against B. cinerea. This conclusion is consistent with the previous observation that the expression of AtPGIP1, another OG-responsive gene in Botrytis-inoculated plants, is also independent of SA, ET, and JA (Ferrari et al., 2003b). Similarly, expression of two Arabidopsis chitin-inducible genes was shown to be independent of SA, ET, and JA (Zhang et al., 2002). More recently, Raacke and colleagues found that yeast (Saccharomyces cerevisiae) cells also induce resistance against B. cinerea in Arabidopsis, and that this resistance is independent of SA, JA, and ET (Raacke et al., 2006b). Interestingly, the authors observed that this resistance is also independent of PAD3; however, it is not known whether the fungal strain they used is sensitive to camalexin, and it is therefore difficult to draw any conclusion on the role of camalexin in yeast-induced resistance. Furthermore, yeast treatment, in contrast to OGs, and also to flg22, induces resistance against P. syringae through an SA-dependent mechanism (Raacke et al., 2006a), indicating that additional defense responses are activated by yeast cells, compared to chemically defined elicitors. A recent article describes the expression profile of nahG, ein2, and coi1 plants inoculated with B. cinerea, and demonstrates that the ZFAR1 gene, encoding a zinc finger protein containing ankyrin repeats, is expressed independently of SA, ET, and JA and is required for basal local resistance against this fungus (AbuQamar et al., 2006). Interestingly, ZFAR1 is also induced by OGs in our microarray experiments, and determining whether this gene is also required for OG-induced resistance could provide further insights in its role in elicitor-mediated signaling. Furthermore, the isolation of mutants impaired in OG-dependent responses will confirm the role of these elicitors in defense against pathogens.

The data presented in this manuscript and in previous publications (Ferrari et al., 2003a) indicate that PAD3-mediated resistance elicited by OGs (and other PAMPs) is one of several independent signaling pathways that contribute to basal resistance. Thus in SA, JA, and ET signaling mutants, PAMP-mediated resistance is still operating such that OG treatment will result in partial resistance as shown in Figure 2. Additional evidence that basal resistance is likely dependent on additional mechanisms, besides the PAMP-dependent activation of defense responses, is that fungal infection is able to activate multiple defense-related pathways, as indicated by the increased PR-1 and PDF1.2 expression in response to B. cinerea (Ferrari et al., 2003a). The cumulative effect of these responses results in the overall level of resistance observed in wild-type plants, and the relative contribution of each pathway appears evident only when specific mutants are assayed. Similarly, basal defense against Pseudomonas syringae requires SA (Delaney et al., 1994; Cao et al., 1997), even though flg22-mediated resistance also occurs in lines impaired in this pathway (Zipfel et al., 2004). We have also shown here that flg22, like OGs, induces resistance against B. cinerea independently of SA, ET, and JA, and that flg22-mediated resistance requires PAD3 and UPS1. This would suggest that OGs, flg22, and possibly other PAMPs act through similar or convergent signaling pathways. A comparison of our microarray data to those previously reported on chitin-treated Arabidopsis seedlings (Ramonell et al., 2005) revealed a remarkable overlap (about 70%) between genes induced by OGs and genes induced by chitin. Considering that the experimental conditions used in each analysis were variable with regards to age of seedlings and time of elicitation, this observation is quite remarkable and supports the hypothesis that a large portion of responses mediated by PAMPs and by OGs are activated through a common signaling pathway, which is largely independent of SA, ET, and JA. In contrast, transcriptional changes specifically induced by chitin octamers are quite distinct from those induced by OGs, supporting the hypothesis that a distinct signaling pathway is mediated by short fragments (Ramonell et al., 2005).

Consistent with a role of PAD3 in OG-mediated resistance, induction of PAD3 expression in response to OGs is dramatically reduced in ups1 plants, which are also not protected by elicitor treatments. In contrast to PAD3, UPS1 appears to encode a regulatory protein required for the expression of different defense genes activated by reactive oxygen species (Denby et al., 2005). OGs released during fungal infection may activate the expression of PAD3 in a UPS1-dependent manner through the activation of a localized oxidative burst. It is, however, important to note that expression of PDF1.2 and PR-1 is also partially compromised in the ups1 mutant (Denby et al., 2005). Therefore, the lack of OG-induced resistance in this genotype may due to loss of multiple defense responses beside PAD3 expression.

All genetic evidence and expression data presented in this article point to the PAD3-dependent accumulation of camalexin as a major determinant of elicitor-induced resistance against B. cinerea in Arabidopsis plants. However, no significant increase of camalexin levels could be detected in plants sprayed with OGs compared to control-treated plants. AgNO₃ treatment resulted in a significant accumulation of camalexin in wild-type adult leaves (data not shown), indicating that camalexin could be detected under our experimental conditions. Therefore, although OG-induced resistance is PAD3 dependent, it does not seem to be directly due to an increase in the levels of camalexin in leaf tissues before pathogen inoculation. This result is surprising because PAD3 catalyzes the last step in camalexin biosynthesis (Glawischnig et al., 2004). One possible explanation for this discrepancy is that elicitor treatments do not induce camalexin accumulation directly, but rather prime plants to synthesize more phytoalexin, or to do it more quickly, after pathogen infection. It was previously shown that some chemicals can increase resistance against pathogen infection by priming plant tissues to activate defense responses more efficiently after inoculation. For instance, pretreatments of parsley (Petroselinum crispum) cells with benzothiadiazole results in enhanced production of coumarin and augmented expression of genes encoding enzymes involved in phytoalexin biosynthesis after inoculation (Katz et al., 1998). Arabidopsis plants pretreated with β-amino-butyric acid show increased accumulation of callose at the site of infection (Ton and Mauch-Mani, 2004). In the latter case, camalexin is not involved in the observed enhanced resistance, since β-amino-butyric acid-treated wild-type and pad3 plants are similarly protected against necrotrophic pathogens, and, after inoculation, camalexin accumulates in primed plants to lower levels than in control plants (Ton and Mauch-Mani, 2004). Experiments are in progress to determine whether elicitors like OGs and flg22 are able to prime Arabidopsis plants to produce more camalexin in response to B. cinerea. An alternative hypothesis is that PAD3 is involved in the biosynthesis of an unknown antimicrobial compound. These compounds could either be degradation products of camalexin, or other indolic compounds whose accumulation is affected by the pad3 and ups1 mutations. To verify these hypotheses, a comprehensive analysis of secondary metabolites accumulating in wild-type and mutant plant treated with elicitors and/or infected with B. cinerea is needed. Interestingly, 16 different indolic compounds were previously identified in Arabidopsis, several of them being induced by pathogen infection (Bednarek et al., 2005). Their role in defense against different pathogens, their levels in response to elicitors, and the impact of the pad3 mutation on their accumulation are still largely unknown.

In conclusion, we have shown that OGs activate the expression of Arabidopsis responses effective against B. cinerea through a pathway that is independent of the well-characterized defense-related signaling molecules SA, JA, and ET. Among these responses, the expression of PAD3 and possibly other genes involved in the biosynthesis of secondary metabolites plays a major role in determining the enhanced resistance against B. cinerea observed in OG-treated plants. The outcome of the interaction between Arabidopsis and this fungus is mediated mainly by the levels of secondary metabolites in the host and the sensitivity of the pathogen to such compounds (Kliebenstein et al., 2005). Therefore, the activation of genes involved in secondary metabolism by cell wall fragments released at the site of infection likely represents an effective mechanism to restrict fungal growth. The dissection of the OG-activated transduction pathway and the identification of the effectors induced by this and other elicitors will provide further insights in the molecular mechanisms regulating the plant innate immune response.

MATERIALS AND METHODS

Upon request, all novel materials described in this publication will be made available in a timely manner for noncommercial research purposes, subject to the requisite permission from any third-party owners of all or parts of the material. Obtaining such permission, if necessary, will be the responsibility of the requestor.

Plant Treatments

Arabidopsis (Arabidopsis thaliana) accession Columbia-0 (Col-0) was obtained from G. Redei and A.R. Kranz (Arabidopsis Information Service, Frankfurt); ups1 seeds were a kind gift of K. Denby (University of Cape Town, South Africa). The triple npr1ein2jar1 mutant was obtained from X. Dong (Duke University, Durham, NC). Heterozygous coi1-1/COI1-1 seeds were a kind gift from J. Turner (University of East Anglia, Norwich, UK). Plants were grown on Metromix 200 medium (Scott) in a Percival AR66 growth chamber at 22°C, relative humidity of 70%, and a 12 h photoperiod with light provided by Philips Hi-Vision white fluorescent lamps at an intensity of 120 μE m⁻² s⁻¹. Plants were fertilized weekly with 0.5× Hoagland solution.

OGs with a degree of polymerization of 10 to 15 were kindly provided by G. Salvi (University of Rome “La Sapienza,” Italy). For elicitor treatments in adult plants, a solution containing 200 μg mL⁻¹ OGs or 5 μ  m flg22 and 0.01% Silwet L-77 (OSi Specialties) was uniformly sprayed on 4-week-old plants until run off (approximately 1 mL for each plant). Plants were then covered with transparent plastic sheet, placed back in the growth chamber, and the plastic cover removed after 3 to 4 h.

For expression profiling following OG treatments in seedlings, seeds were sterilized and germinated in 12-well plates (approximately 15 seeds per well) containing 1 mL per well Murashige and Skoog medium (Life Technologies; Murashige and Skoog, 1962) supplemented with 0.5% Suc and Gamborg B5 vitamins. Plates were incubated at 22°C with a 16 h photoperiod and a light intensity of 100 μE m⁻² s⁻¹. After 8 d, the medium was replaced with 1 mL of fresh medium. At 10 d, 50 μg mL⁻¹ OGs or an equivalent volume of water was added to the medium. For each biological replicate, about 15 seedlings were harvested, briefly blotted dry, and immediately frozen in liquid nitrogen. For assays of PAD3 expression in seedlings, seeds were sterilized and germinated in 24-well plates (approximately 15 seeds per well) containing 1 mL per well Murashige and Skoog medium (Life Technologies; Murashige and Skoog, 1962) supplemented with 0.5% Suc and Gamborg B5 vitamins. Plates were incubated at 22°C with a 16 h photoperiod and a light intensity of 100 μE m⁻² s⁻¹. After 10 d, the medium was replaced with 1 mL of fresh medium with or without 100 μg mL⁻¹ OGs. For each biological replicate, about 10 seedlings from each of three separate wells (fresh weight: 100–200 mg) were harvested, briefly blotted dry, pooled, and immediately frozen in liquid nitrogen. The effectiveness of the treatments was assessed by measuring hydrogen peroxide released in the medium (data not shown). For treatment of coi1 seedlings, heterozygous COI1/coi1 seeds were first germinated on agar plates containing 30 μ  m methyl jasmonate, and, after 8 d of growth, resistant seedlings were transferred to liquid Murashige and Skoog medium and treated with OGs 2 d later. As a control, wild-type seedlings were grown for 8 d on agar plates and then transferred to liquid Murashige and Skoog medium.

Inoculation with Botrytis cinerea for the microarray experiments was conducted on 4-week-old plants by placing four 5 μL droplets of a spore suspension (5 × 10⁵ conidia mL⁻¹) in 24 g L⁻¹ potato (Solanum tuberosum) dextrose broth on each rosette leaf (two fully expanded leaves per plant). Inoculated plants were covered with a transparent plastic dome to maintain high humidity and returned to the growth chamber. For each biological replicate, inoculated leaves from three different plants (corresponding to about 200 mg fresh weight) were harvested, pooled, and immediately frozen in liquid nitrogen.

Inoculation of adult plants for the pathogenicity assays was conducted on detached leaves, as previously described (Ferrari et al., 2003a). Homozygous coi1-1/coi1-1 plants were identified after fungal infection by their sterile phenotype (Feys et al., 1994). For induction of systemic resistance, OGs (200 μg mL⁻¹ in sterile distilled water) or water were infiltrated in two lower rosette leaves using a needleless syringe. After 72 h, upper, untreated fully expanded leaves were detached and inoculated with B. cinerea.

Microarray Hybridization

For the OG experiment, three biological replicates for each treatment were analyzed. For fungal infection, two replicates were analyzed. Total RNA was extracted from each sample using the Qiagen RNeasy Plant RNA Miniprep kit (Qiagen); samples were split in two before homogenization and repooled before loading on the RNA-binding column. RNA quality was assessed by determining the A  _260/280 ratio of RNA in Tris buffer and by checking the integrity of RNA on an Agilent 2100 Bioanalyzer (Agilent Technologies, www.agilent.com). Target labeling and microarray hybridizations were performed according to the protocol given in the Affymetrix GeneChip Expression Analysis Technical Manual 701025 rev 1 (for details, see Supplemental Methods S1). Arrays were scanned using an Affymetrix GeneArray 2500 scanner and Affymetrix MicroArray Suite v5.0 software. Original raw data for each experiment are available at the Nottingham Arabidopsis Stock Centre (http://affymetrix.arabidopsis.info/narrays/experimentbrowse.pl) and http://ausubellab.mgh.harvard.edu/imds (under experiment names “Botrytis cinerea infection, 18 and 48 hpi” and “Comparison of response to Flg22 and OGs elicitors”).

Data Analysis

To assess the quality of each hybridization, we used Affymetrix MicroArray Suite v5.0 analysis software for reports of background intensity, signal-to-noise ratio, scaling factor for global normalization, and ratios of intensity between 3′ and 5′ probe sets for selected genes. Further data analysis was performed with Rosetta Resolver v3.2 Gene Expression Data Analysis system (Rosetta Inpharmatics), using Affymetrix.CEL files of array feature intensities and sds as input. Determination of absolute intensity values, propagation of error and P values, and normalization for comparing arrays in the Resolver system have been described in Waring et al. (2001) and are summarized below. For each probe set, comprised of multiple perfect match (PM) and mismatch (MM) probe pairs, an intensity difference between each PM and corresponding MM was calculated. Probe pairs that differed by more than 3 sds from the mean PM-MM difference for the probe set were considered outliers and were not included in the final calculation of the mean PM-MM intensity difference. Calculation of the probability that a gene is present in the set of transcripts being analyzed was based on the intensities of negative control genes. To increase detection sensitivity, data from three biological replicates per OG treatment, and two biological replicates per B. cinerea treatment were combined. For each array, average intensities, associated intensity errors, and P values were calculated for each probe set. For calculating average intensity from replicate samples, arrays were scaled to mean intensity, intensity values were transformed for homogenous variance, nonlinear error correction was performed, and probe set average intensities computed taking into account measurement error calculations. P values were calculated and intensity transformed back to the original scale. Ratios of treated versus control intensities were computed by calculating baseline mean background and signal, calculating ratio P values, and building simple ratios. One-way error-weighted ANOVA was used to identify differentially expressed genes for each time point, using a threshold of P ≤ 0.01. Error-weighted ANOVA has two inputs, expression level and measurement error associated with the expression level, which provides additional information that yields more reliable variance estimates when the number of replicates is small. Multiple testing correction was performed using q value. Only genes for which the absolute fold-change between treated and control samples was greater than or equal to 2 were considered to be up- or down-regulated. Gene annotation and assignment to functional categories were based on The Institute for Genomic Research Arabidopsis Genome Annotation Database (http://www.tigr.org/tdb/e2k1/ath1/ath1.shtml), the MIPS Functional Categories Database (http://mips.gsf.de/projects/funcat; Ruepp et al., 2004), and, for secondary metabolism genes, the available literature (Gachon et al., 2005; Kliebenstein et al., 2005; see also Supplemental Table S5). Graphic representation of the expression of secondary metabolism genes was performed using MapMan software (Thimm et al., 2004).

RNA Analysis

Total RNA was prepared using the Trizol reagent (Life Technologies). RNA gel blots were prepared and hybridized with single-stranded radioactive probes as previously described Rogers and Ausubel (1997). Blots were washed twice with 1% SDS, 2× SSC at 65°C for 45 min, and images were taken with a Phosphorimager (Molecular Dynamics) after overnight exposure. The template used to generate the PAD3 probes was amplified by PCR from Arabidopsis Col-0 genomic DNA as previously described (Zhou et al., 1999). For quantitative RT-PCR analysis, RNA was treated with RQ1 DNase (Promega) and first-strand cDNA was synthesized using ImProm-II Reverse Transcriptase (Promega) according to the manufacturer's guide. Real-time PCR analysis was performed using an I-Cycler (Bio-Rad) according to the manufacturer's guide. Two microliters of cDNA (corresponding to 120 ng of total RNA) were amplified in 30 μL reaction mix containing IQ SYBR Green Supermix (Bio-Rad) and 0.4 mm of each primer. Primer sequences were the following: 5′-CCGGTGAATCTTGAGAGAGCC-3′ and 5′-GATCAGCTCGGTCATTCCCC-3′ (PAD3); 5′-GGAAGAAGAAGACTTACACC-3′ and 5′-AGTCCACACTTACCACAGTA-3′ (UBQ5). Relative expression of the RT-PCR products was determined using the ΔΔC_t method (Livak and Schmittgen, 2001).

NASCArrays Experiment Reference Numbers for microarray data are NASCARRAYS-409 and NASCARRAYS-167.

Supplemental Data

The following materials are available in the online version of this article.

• Supplemental Figure S1. Lesion development in control and OG-treated leaves.

• Supplemental Table S1.  B. cinerea infection microarray full dataset.

• Supplemental Table S2. OG treatment microarray full dataset.

• Supplemental Table S3. Genes coregulated by OGs and infection.

• Supplemental Table S4. Comparison between OG- and chitin-regulated genes.

• Supplemental Table S5. Secondary metabolism gene expression and references.

• Supplemental Materials and Methods S1. Microarray hybridization details.

ACKNOWLEDGMENTS

We are grateful to Jennifer Couget and Paul Grosu (Bauer Center for Genomics Research, Harvard University, Cambridge, MA) for assistance with microarray hybridization and data analysis.

Author notes

1

This work was supported by the Giovanni Armenise-Harvard Foundation, the Institute Pasteur-Fondazione Cenci Bolognetti, by Ministero dell'Università e della Ricerca Fondo per gli Investimenti della Ricerca di Base 2001 and Cofinanziamento 2002 grants awarded to G.D.L., by the European Union (grant no. 23044 “Nutra-Snacks” to S.F.), and by the National Science Foundation (grant no. DBI–0114783 to F.M.A.) and the National Institutes of Health (grant no. GM48707 to F.M.A.).

^*

Corresponding author; e-mail ; fax [phone].

The author responsible for distribution of materials integral to the findings presented in this article in accordance with the policy described in the Instructions for Authors (www.plantphysiol.org) is: Simone Ferrari ().

[W]

The online version of this article contains Web-only data.

www.plantphysiol.org/cgi/doi/10.1104/pp.107.095596