Plant Molecular Biology

, Volume 70, Issue 1, pp 79–102

Early genomic responses to salicylic acid in Arabidopsis

  • Francisca Blanco
  • Paula Salinas
  • Nicolás M. Cecchini
  • Xavier Jordana
  • Paul Van Hummelen
  • María Elena Alvarez
  • Loreto Holuigue
Open AccessArticle

DOI: 10.1007/s11103-009-9458-1

Cite this article as:
Blanco, F., Salinas, P., Cecchini, N.M. et al. Plant Mol Biol (2009) 70: 79. doi:10.1007/s11103-009-9458-1

Abstract

Salicylic acid (SA) is a stress-induced hormone involved in the activation of defense genes. Here we analyzed the early genetic responses to SA of wild type and npr1-1 mutant Arabidopsis seedlings, using Complete Arabidopsis Transcriptome MicroArray (CATMAv2) chip. We identified 217 genes rapidly induced by SA (early SAIGs); 193 by a NPR1-dependent and 24 by a NPR1-independent pathway. These two groups of genes also differed in their functional classification, expression profiles and over-representation of cis-elements, supporting differential pathways for their activation. Examination of the expression patterns for selected early SAIGs from both groups indicated that their activation by SA required TGA2/5/6 subclass of transcription factors. These genes were also activated by Pseudomonas syringae pv. tomato AvrRpm1, suggesting that they might play a role in defense against bacteria. This study gives a global idea of the early response to SA in Arabidopsis seedlings, expanding our knowledge about SA function in plant defense.

Keywords

as-1 Element Glutaredoxin NPR1 Salicylic acid SA-induced genes TGA factors 

Introduction

Salicylic acid (SA) is a key hormone of stress defense responses induced by biotrophic pathogens in plants (Alvarez 2000; Durrant and Dong 2004; Loake and Grant 2007). These responses allow plants not only to survive pathogen infection, but also to acquire a long-lasting systemic resistance (SAR, systemic acquired resistance) responsible for the protection from further infections by a broad range of pathogens (Grant and Lamb 2006). Several lines of evidence give support to the idea that SA plays a key role in the establishment of SAR. Tissue levels of SA and its glucosylated conjugates (SAG) increase locally and systemically after pathogen infection (Summermatter et al. 1995). Conversely, blockade of SA accumulation severely impairs the deployment of SAR (Gaffney et al. 1993; Delaney et al. 1994; Wildermuth et al. 2001). Further supporting a direct role of SA in pathogen-induced resistance, a series of experiments showed that treatment of plant tissues with exogenous SA or its functional analogs BTH (benzothiadiazole S-methylester) and INA (2,6-dichoroisonicotinic acid), was sufficient to trigger a defense reaction resembling SAR, that protects plants against a further pathogen infection (Cao et al. 1994; Lawton et al. 1996).

The resistance to infection triggered by SA seems to be mainly due to its ability to activate defense genes. This is supported by the fact that mutants in some components of the transcriptional machinery activated by SA, such as the co-activator NPR1 (npr1 mutants) and the subclass II of TGA transcription factors (triple mutant tga2-2/tga5-1/tga6-1), are severely impaired to deploy SAR, either in response to SA/INA treatment, or to avirulent pathogens (Cao et al. 1994; Zhang et al. 2003). Interestingly, analysis of the tga2-2/tga5-1/tga6-1 triple mutant allowed the identification of TGA2, TGA5 and TGA6 as redundant factors involved in SA-mediated activation and basal repression of PR1 gene (Zhang et al. 2003). Impairment in SAR development, although less severe than that of npr1 and tga2/tga5/tga6 mutants, was also recently reported in mutants in the SA-activated WRKY 18, one of the members of the large family of WRKY transcription factors activated by pathogens (Wang et al. 2006).

Serious efforts have been done to characterize the pathogen-induced transcriptome of Arabidopsis (reviewed in (Katagiri 2004; Eulgem 2005; Glazebrook 2005). It is known that SA-induced genes (SAIGs) are an important part of this transcriptome, as revealed by the use of mutants in SA biosynthesis or signaling steps such as sid2, npr1 or nahG plants (Maleck et al. 2000; Glazebrook et al. 2003; Tao et al. 2003; Eulgem et al. 2004; Katagiri 2004; AbuQamar et al. 2006). Studies aimed at identifying and characterizing SAIGs that are important for defense reaction have focused mainly on genes that depend on the coactivator NPR1 (NPR1-dependent SAIGs) (Maleck et al. 2000; Wang et al. 2005, 2006). Some of the best-characterized NPR1-dependent SAIGs so far are the pathogenesis-related genes (PRs), a group of genes that code for proteins with antimicrobial activity (van Loon et al. 2006). PR-1, a member of this group, is the most common marker for this defense pathway (Maleck et al. 2000; van Loon et al. 2006). Furthermore, SAIGs coding for endoplasmic reticulum-localized proteins associated to the protein secretory machinery and for WRKY transcription factors involved in PR-1 activation, have been recently identified as primary targets of NPR1 involved in defense (Wang et al. 2005, 2006). Recently, two studies analyzing the response of Arabidopsis to short treatments with SA were published (Thibaud-Nissen et al. 2006; Krinke et al. 2007). One was performed in Arabidopsis cell suspensions and allowed the identification of SAIGs associated to the early activation of phosphatidylinositol 4-kinase (Krinke et al. 2007). The other study was performed in Arabidopsis plants and was oriented to identifying targets of TGA2 factor (Thibaud-Nissen et al. 2006).

Increasing evidence supports the idea that SA also plays a role in controlling the cellular redox balance at the onset of the SAR (Mou et al. 2003; Mateo et al. 2006; Holuigue et al. 2007). Treatment of Arabidopsis plants with INA produces a biphasic change of cellular redox potential (GSH/GSSG ratio): first a pro-oxidative effect and then an antioxidant effect of the SA analog (Mou et al. 2003). Associated to these changes, SA has been found to activate through a redox mechanism, NPR1, TGA1 (a member of the TGA family of bZIP transcription factors) and the SA-responsive as-1 promoter element (Garreton et al. 2002; Despres et al. 2003; Mou et al. 2003). In rice, SA has been proven essential for the protection of plants from oxidative stress caused by light, senescence, avirulent pathogens and treatment with methyl viologen (Yang et al. 2004). Consistently, in Arabidopsis, SA was required to restrict the oxidative stress and foster plant acclimation after transient exposure to high light intensity (Mateo et al. 2006). Interestingly, we and others found that SA uses a transient kinetics to activate genes coding for detoxifying or antioxidant enzymes such as glutathione S-transferases (GSTs) and glycosyltransferases (GTs) (Wagner et al. 2002; Lieberherr et al. 2003; Sappl et al. 2004; Uquillas et al. 2004; Blanco et al. 2005; Langlois-Meurinne et al. 2005). Besides activating stress defenses, SA is known to inhibit JA-and auxin-mediated responses, as part of the SA-mediated disease-resistance mechanism (Fujita et al. 2006; Ndamukong et al. 2007; Wang et al. 2007).

The aim of this study was to expand our knowledge about SA function in defense, identifying the subset of SAIGs that mediate early defense responses to SA in Arabidopsis seedlings (early SAIGs). For this purpose, we studied the expression profiles of wild type and npr1-1 mutant Arabidopsis seedlings in response to short SA treatments, with the Complete Arabidopsis Transcriptome MicroArray (CATMAv2) chip. We identified 217 early SAIGs and found that a small proportion of them (24 genes) used an NPR1-independent pathway. Analysis of the functional categories of these genes indicates that NPR1-independent SAIGs mainly code for proteins involved in stress defense responses and C-compound metabolism, while proteins involved in protein phosphorylation and response to biotic stress were over-represented in the group of NPR1-dependent SAIGs. We confirmed, for a selection of these genes, that they are also responsive to inoculation with the avirulent bacterium Pseudomonas syringae pv. tomato AvrRpm1.

In silico promoter analysis of the early SAIGs identified in this study, and expression analysis performed for a selection of these genes (coding for glutaredoxins GRXC9 and GRXS13; the ANAC102 transcription factor, the unclassified SDRLP protein, a lectin like protein LLP and the NPR1-interactor NIMIN1), indicates that SA activates the transcription of theses genes by a mechanism involving the subclass II of TGA transcription factors (TGA 2/5/6). Based on present findings and on previously published work, we propose a model to explain how a proper combination of SA-responsive promoter elements (as-1-like, TL-1 and TGA boxes), transcriptional regulators and co-regulators (TGA factors, NPR1 and other uncharacterized factors) could mediate the temporal control of early SAIGs expression.

Materials and methods

Plant growth conditions and treatments

The Arabidopsis thaliana wild-type, npr1-1 mutant (Cao et al. 1994) and tga6-1 tga2-1 tga5-1 triple mutant (Zhang et al. 2003) plants used in this study were from Columbia (Col-0) ecotype. Plants were grown in vitro in MS medium containing 15 g/l sucrose, under controlled conditions in a growth chamber (16 h light, 60 μmoles m−2 s−1, 22 ± 2°C). Experiments were performed in 15 day-old seedlings (collected at 9:00 am) carefully taken from the MS medium and placed, root side down, in a Petri dish containing the corresponding solutions. For SA treatment, seedlings (wt and npr1-1 mutant) were incubated in a 0.5 mM SA solution in MS medium, under continuous light (60 μmoles m−2 s−1) for 2.5 h in the case of microarray analysis, or for 0.5, 1, 2.5, 5, 8, 12 and 24 h for Northern blot analysis. Stock solution of 1 M SA (Sigma) was fresh-prepared in water. Control samples were incubated under the same conditions in MS medium.

To evaluate the need of de novo protein synthesis for SA-induced response, the protein synthesis inhibitor cycloheximide (CHX) was used. Seedlings were pre-incubated for 1 h in MS medium in the absence (control) or presence of 20 μg/ml CHX and then exposed to 0.5 mM SA or MS for 2.5, 8 and 24 h. Stock solution of 20 mg/ml CHX (Sigma) was fresh-prepared in water. To evaluate transcript stability, seedlings were first incubated with 0.5 mM SA (prepared in MS) for 2 h and then transferred to MS medium (controls) or to a 0.5 mM SA solution, either in the presence or absence of 0.6 mM cordycepin (Sigma). A control sample treated with MS for 2 h was included. Samples were collected at 0.5, 1, 2, 3, 5 and 8 h.; whole seedlings were immediately frozen in liquid nitrogen and stored at −70°C until RNA purification.

Microarray analysis

Total RNA was extracted from control and SA-treated seedlings of wild type and npr1-1 mutant plants (pool of 20 seedlings for replicate, two independent biological replicates for each treatment) with Trizol®, according to the manufacturer’s instructions (Invitrogen). In each case, we compared SA-treated seedlings with that from control-treated seedlings on an Arabidopsis cDNA microarray, CATMAv2. Each comparison was repeated in dye-swap leading to 4 hybridizations per comparison.

The CATMAv2 array used in this study was an updated version of CATMAv1 array that was previously described and validated (Hilson et al. 2004; Allemeersch et al. 2005). This array consisted of 23,520 features, including 18,981 unique GSTs, 768 positive/negative controls (Amersham BioSciences), and 243 blanks, printed on Type VIIstar reflective slides (Amersham BioSciences) using a Lucidea Array spotter (Amersham BioSciences). The annotation of the clone set can be accessed via the ArrayExpress database as accession number A-MEXP-10 (http://www.ebi.ac.uk/arrayexpress) or via the VIB MicroArray Facility Web site (http://www.microarrays. be). Prior to hybridization, the slides were washed in 2× saline-sodium phosphate-EDTA buffer, 0.2% SDS for 30 min at 25°C. RNA was amplified using a modified protocol of in vitro transcription (Puskas et al. 2002). Briefly, 5 mg of total RNA was reverse transcribed to double-stranded cDNA using an anchored oligo(dT) 1 T7 promoter [5′-GGCCAGTGAATTGTAATACGACTCACTATAGGGAGGCGGT24(ACG)-3′ (Eurogentec, Seraing, Belgium)]. From this cDNA, RNA was produced via T7-in vitro transcriptase until an average yield of 10–30 mg of amplified RNA. The amplified RNA (5 mg) was labeled with dCTP-Cy3 or Cy5 (Amersham BioSciences) by reverse transcription using random nonamer primers (Genset, Paris). The resulting probes were purified with Qiaquick (Qiagen) and analyzed for amplification yield and incorporation efficiency by measuring the DNA concentration at 280 nm, Cy3 incorporation at 550 nm, and Cy5 incorporation at 650 nm using a Nanodrop spectrophotometer (NanoDrop Technologies, Rockland, DE). A good target had a labeling efficiency of 1 fluorochrome every 30–80 bases. For each target, 40 pmol of incorporated Cy5 or Cy3 were mixed in 210 ml of hybridization solution containing 50% formamide, 1× hybridization buffer (Amersham BioSciences), 0.1% SDS. Each spike mix was hybridized against the reference RNA (spikes at 100 cpc) and repeated with dye swap to make up 14 hybridizations in total. Hybridization and post hybridization washing were performed at 45°C with an Automated Slide Processor (Amersham BioSciences). Post-hybridization washing was done in 1× sodium chloride/sodium citrate buffer (SSC), 0.1% SDS, followed by 0.1× SSC, 0.1% SDS and 0.1× SSC. Arrays were scanned at 532 nm and 635 nm using a Generation III scanner (Amersham BioSciences). Images were analyzed with ArrayVision (Imaging Research, St. Catharines, Canada). All protocols are available at the VIB MicroArray FacilityWeb site (http://www.microarrays.be) and at ArrayExpress under accession numbers P-MEXP-578, P-MEXP-579, P-MEXP-581, P-MEXP-582 for Cy3 labeling, Cy5 labeling, hybridization, and scanning, respectively.

Microarray data analysis

Spot intensities were measured as artifact removed total intensities, subtracted with the local background (sARVol), and filtered based on two standard deviations above background. For each gene, ratios of red (Cy-5) over green (Cy-3) intensities (I) were calculated and normalized via a Lowess Fit of the log2 ratios [log2(Icy-5/Icy-3)] over the log2 total intensity [log2(Icy-5*Icy-3)]. Genes were considered differentially expressed if the normalized ratios were statistically significant using a two-tailed t test (< 0.01) between the biological and dye-swap repeats and were higher or equal to two-fold.

Functional classification of genes was performed according to the Functional Catalogue (FunCat) version 2.1 available at http://mips.gsf.de/proj/funcatDB (Ruepp et al. 2004). Over-represented functional categories in the two lists of early SAIGs compared to the complete Arabidopsis genome were obtained according to the MIPS classification using BioMaps tool from the VirtualPlant webpage (http://virtualplant.bio.nyu.edu). Hypergeometric method and Bonferroni correction were used for the analysis with a P value cutoff of 0.05. To collect expression data of selected genes in response to stress, the stimulus analysis tool from the public GeneVestigator V3 database, which compile information of Affymetrix GeneChip experiments, was used (https://www.genevestigator.ethz.ch/; (Laule et al. 2006). Intersection analysis tool from VirtualPlant version 0.9 (http://virtualplant-prod.bio.nyu.edu/cgi-bin/virtualplant.cgi) was used to analyze the intersections between different lists of genes from microarray experiments.

Northern analysis

For Northern blot analysis, 10 μg total RNA (isolated by using Trizol®) of the different samples were separated on formaldehyde-agarose gels, as previously described (Blanco et al. 2005). Radioactive signals were either quantified by a Phosphorimager (Cyclone, Storage Phosphor screen, Packard Bioscience Company) or detected by autoradiography after 12 h exposure. Gene-specific DNA probes were obtained by PCR using cDNA from SA-treated seedlings as template and specific primers designed according to each gene sequence. Amplified probes were cloned and sequenced. Sequences of the gene specific primers are described in Table 1. Probes were labeled by PCR as described (Blanco et al. 2005).
Table 1

Primers used to obtain gene-specific probes

AGI

Forward primer

Reverse primer

AT5G24530

GGTGAGTGTTGCGCGTGAGTTCTTTAGC

GGGCGTTTGGGTCGGTATGAGCA

AT1G02450

TACAATTATTCCCTAGAGACCAT

TAATAATCACAACGCTAACAAAT

AT3G25882

GTTCGGACGGTAACGGAGGAAGAGGT

GTTCAGATCCAAACCCAAACCCTGTAACC

AT1G28480

GTGCAAGAAATTATAACATGACGACAAC

TTTACAATCACAACCACAGAGCCCCAACT

AT1G03850

GTCACCGTGGACGAAGACCG

CCAGTGGTGGTTGAGATTGG

AT4G13180

ATCGCGATTCACCTCCACCTCC

AACGGCGGCCTTTGATGC

AT1G76680

GATCTGGCATGTTGGCCGCG

CGATCTCCTTAGCAACTGCATC

AT5G03350

TTGGGAAAATGAAACACTGGTC

CATTCCGGTTACAACTTTCTGATAC

AT5G22570

TCACCGGATCCGATTTACTACGAC

AGCGGCTTTCCTTCTCCTGA

AT5G64250

ATGCGAGAGGGATTGGTCTTCT

ATGGCGTCTACTCCTACAGCAACA

AT3G11340

GACGCTCTTTGGTACTTCACTCACGA

TACGGAATTTGTTGCTTGCTTGTCTAA

AT1G37130

CCTAACGGCGGGCTCATGTCTC

TCTCGCATAATCGCCTGCGAA

AT3G53750

GCTATGTATGTCGCCATTCAAGC

CATCATATTCTGCCTTTGC(A/G)ATCC

AT2G14610

GTAGGTGCTCTTGTTCTTCCC

CACATAATTCCCACGAGGATC

AT2G29420

TGATTCCGGTTCTTGTTCATAATG

AACGCCACCAAAGTCGCCAC

AT5G63790

GGACTTTGCTCTCTTCTCC

GCACATCTCCGGCAAAG

Plant inoculation with P. syringae

For bacterial inoculation assays, 6-week old plants grown on soil under controlled condition in a growth chamber (22 ± 2°C, 10 h light, 100 μmoles m−2 s−1) were inoculated with Pseudomonas syringae pv. tomato strain DC3000 that expresses the AvrRpm1 gene (Pst/AvrRpm1). A bacterial suspension of 5 × 106 cfu/ml in 10 mM MgCl2 was infiltrated on the abaxial side of the leaves with a 1 ml syringe (Pavet et al. 2006). Mock-treated plants were infiltrated with 10 mM MgCl2. Bacteria- and mock-infiltrated leaf samples were collected at 0, 5, 21, 26 and 48 h post infiltration, immediately frozen in liquid nitrogen and stored at −70°C until RNA purification or SA quantification.

Quantification of SA

Free and conjugated SA extracted from leaf tissues was quantified by HPLC as described (Verberne et al. 2002).

Promoter analysis

Over-represented promoter elements were independently searched in the groups of 31 NPR1-independent and 197 NPR1-dependent SAIGs, using the Motif-Sampler software (http://www.esat.kuleuven.ac.be/~thijs/Work/MotifSampler.html) (Thijs et al. 2002). For each group of genes, 1 kb of the genomic sequence upstream from the inferred translational start site was downloaded from TAIR database (The Arabidopsis Information Resource, http://www.arabidopsis.org). Promoters analysis was performed essentially as described (Blanco et al. 2005), except that the motif lengths (w) were set at 16 and 20 bp. For each search, the algorithm was iterated 2,000 times; independent outputs were merged and the 10 highest scored motifs were selected and ranked according to their log-likelihood scores. Finally, to search for known plant promoter elements that matched the selected motifs, PLACE database (Database of Plant Cis-acting Regulatory DNA elements, http://www.dna.affrc.go.jp/PLACE/) was used (Higo et al. 1999).

Results

Identification of early SA-induced genes, determination of NPR1 dependence and functional classification

To analyze the early genetic response of Arabidopsis to SA, transcript profiles of seedlings incubated for 2.5 h with or without SA were compared using CATMAv2 (Complete Arabidopsis Transcriptome MicroArray) DNA chips including 18,981 unique gene specific tags (GSTs) (Lurin et al. 2004). The requirement of NPR1 for this response to SA was determined comparing transcript profiles of wild type and npr1-1 mutant seedlings. Genes differentially activated by SA were selected according to the results of a t-test analysis using a two-fold increase in expression as threshold. We selected 217 SAIGs (1.1% of total genes monitored) from wild type plants and 34 SAIGs (0.18%) from npr1-1 mutant plants (see supplementary Table S1). Genes whose steady-state mRNA levels increased in response to SA exclusively in the wild type plants, were considered as “early NPR1-dependent SAIGs” (193 genes), while genes whose mRNA levels increased after SA treatment in both wild type and npr1-1 plants were considered as “early NPR1-independent SAIGs” (24 genes). Genes that showed reduced responsiveness to SA in the npr1-1 background compared with the wild type were included in the early NPR1-independent SAIGs.

In a previous study, we identified 14 early SAIGs (Uquillas et al. 2004; Blanco et al. 2005). In the CATMAv2 chip, only 6 of these genes were represented by probes and, although 5 of them seemed to be induced by SA, the induction values were statistically significant for only 3 of these genes (At2g29420, At5g54610, At3g11280). These 3 genes were included in the list of 217 SAIGs. Therefore, for further classification and expression analysis of early SAIGs, we considered the 217 genes identified in this study and the 11 SAIGs (4 NPR1-dependent and 7 NPR1-independent) that were identified and corroborated previously (Uquillas et al. 2004; Blanco et al. 2005). The list of early NPR1-dependent SAIGs with activation levels higher than 3 (46 out of 197 genes) is shown in Table 2, whereas the list of the 31 early NPR1-independent SAIGs is shown in Table 3.
Table 2

NPR1-dependent SAIGs

AGI

Alias

SA wt

Annotation

FunCat SubCategories

References

Protein with binding function

At5g03350

LLP

191.3

Legume lectin family protein

C-compound binding

 

At1g02450

NIMIN-1

55.7

NPR1/NIM1-interacting protein 1

Disease, virulence and defense; response to biotic stimulus; transcriptional control; protein binding

Weigel et al. (2005)

At3g25882

NIMIN-2

50.9

NPR1/NIM1-interacting protein 2, encodes a kinase that physically interacts with NPR1/NIM1

Unclasssified protein

Weigel et al. (2005)

At5g54610

ANK

11.9

Ankyrin repeat family protein

Plant signaling molecules response; chemoperception and response; protein binding

Blanco et al. (2005)

At3g56710

SIB1

7.6

Sigma factor binding protein 1; interacts with Sig1R4

Protein binding; chloroplast

 

At1g73800

 

5.0

Calmodulin-binding protein

Protein binding

 

At2g46430

CNGC4

3.1

Cyclic nucleotide gated channel (CNGC4), downstream component of the signaling pathways leading to HR/resistance

Protein binding, cyclic nucleotide binding, ion channel

Gobert et al. (2006)

Protein fate (protein modification and degradation)

At3g09010

 

7.0

Protein kinase family protein

Protein modification by phosphorylation-dephosphorylation

 

At1g21250a

WAK1

5.3

Cell wall-associated kinase

Plant signaling molecules response; chemoperception and response; receptor enzyme mediated signalling

He et al. (1998); Park et al. (2001); Blanco et al. (2005)

At1g34750

 

4.3

Protein phosphatase 2C, putative

Protein modification by phosphorylation-dephosphorylation

 

At4g25110

 

4.2

Caspase family protein; similar to LOL3 (LSD ONE LIKE 3), caspase/cysteine-type endopeptidase

Protein/peptide degradation; apoptosis

Vercammen et al. (2004)

At1g51800

 

3.8

Leucine-rich repeat protein kinase, putative

Protein modification by phosphorylation-dephosphorylation

 

At1g65800

ARK2

3.6

Armadillo repeat kinesin 2. Putative receptor-like serine/threonine protein kinases that is similar to brassica self-incompatibility (S) locus

Receptor enzyme mediated signalling; protein modification by phosphorylation-dephosphorylation

Sakai et al. (2008)

At1g16260

 

3.5

Protein kinase family protein. Identical to Wall-associated receptor kinase-like 8 precursor

Protein modification by phosphorylation-dephosphorylation

 

At4g26070

MEK1

3.4

Member of MAP Kinase Kinase

Oxidative stress response; disease virulence and defense; chemoperception and response; response to biotic stimulus; protein kinase; protein binding

Xing et al. (2007)

At2g02800

APK2b

3.1

Protein kinase 2B

Protein modification by phosphorylation-dephosphorylation

Ito et al. (1997)

Transcription

At5g22570

ATWRKY 38

31.2

Member of WRKY Transcription Factor; Group III

Transcriptional control

Wang et al. (2006)

At5g01900

ATWRKY62

20.1

Member of WRKY Transcription Factor; Group III

Transcriptional control

Mao et al. (2007)

At2g40750

ATWRKY54

7.2

Member of WRKY Transcription Factor; Group III

Transcriptional control

Wang et al. (2006)

At4g23810

ATWRKY53

4.0

Member of WRKY Transcription Factor; Group III

Transcriptional control

Wang et al. (2006)

At4g31800

ATWRKY18

3.8

Member of WRKY Transcription Factor

Transcriptional control; disease, virulence and defense; chemoperception and response; response to biotic stimulus; plant signaling molecules response

Wang et al. (2006)

Perception and response to stimuli (disease resistance proteins)

At3g04210

 

3.6

Disease resistance protein (TIR-NBS class), putative

Disease, virulence and defense; receptor enzyme mediated signalling; ATP binding

 

At1g11310

ATMLO2

3.4

Mildew resistance locus O 2

Disease virulence and defense; plant defense response; cell death; protein binding

Consonni et al. (2006)

At4g26090

RPS2

3.2

Disease resistance protein RPS2 (CC-NBS-LRR class)

Disease virulence and defense; response to biotic stimulus; cell death; protein binding

Cai et al. (2006)

At5g36930

 

3.1

Disease resistance protein (TIR-NBS-LRR class), putative

Disease virulence and defense; receptor enzyme mediated signalling; ATP binding; mitochondrion

 

At3g50480

 

3.0

Homolog of RPW8 that confers broad-spectrum resistance to powdery mildew pathogens

Unclassified protein

Orgil et al. (2007)

Metabolism (C-compound and carbohydrate, nitrogen and lipid metabolism)

At3g11340

UGT76B1

52.9

UDP-glucoronosyl/UDP-glucosyl transferase family protein (Group D)

Nucleotide-sugar metabolism

 

At1g66970

 

5.8

Glycerophosphoryl diester phosphodiesterase family protein

C-3 compound metabolism, phospholipid metabolism

Borner et al. (2003)

At5g24210

LPS

4.4

Lipase class 3 family protein

Lipid, fatty acid and isoprenoid metabolism

 

At5g05460

 

4.3

Glycosyl hydrolase family 85 protein

Chloroplast

 

At2g30550

 

3.7

Lipase class 3 family protein

Lipid, fatty acid and isoprenoid metabolism; chloroplast

 

At5g37600

ATGSR1

3.1

Encodes a cytosolic glutamine synthetase

Nitrate assimilation; metabolism of glutamine-glutamate

Ishiyama et al. (2004)

Transport

At4g13510

ATAMT1

3.9

Ammonium transport 1, encodes a plasma membrane localized ammonium transporter

Ammonium transport

Loque et al. (2006)

At5g53550

YSL3

3.7

Yellow stripe like 3; member of the oligopeptide transporter (OPT) family

Response to biotic stimulus; peptide transport

Waters et al. (2006)

At1g12940

ATNRT2.5

3.0

Nitrate transporter 2.5, member of High affinity nitrate transporter family

Nitrate transport

Pourtau et al. (2006)

Unclassified

At1g14880

 

11.7

Similar to unknown protein

Unclassified protein

Rohde et al. (2004)

At3g29240

 

4.7

Similar to unknown protein

Unclassified protein

Vergnolle et al. (2005)

At3g62780

 

4.3

C2 domain-containing protein

Unclassified protein

 

At4g32870

 

4.2

Similar to unknown protein

Unclassified protein

Yu et al. (2005)

At2g39210

 

4.0

Nodulin family protein

Unclassified protein

Taji et al. (2004)

At3g26600

 

3.7

Armadillo/beta-catenin repeat family protein

Unclassified protein

Coates (2003)

At1g49750

 

3.4

Leucine-rich repeat family protein

Unclassified protein

Wong et al. (2006)

At1g52780

 

3.3

Similar to unknown protein

Unclassified protein

Dunkley et al. (2004)

At3g13950

 

3.1

Similar to unknown protein

Unclassified protein

Sato et al. (2007)

At3g48640

 

3.0

Similar to unknown protein

Unclassified protein

 

Only genes with induction ratio higher than 3 are included. Genes were classified in the more representative subcategories, as defined by MIPS (FunCat DB (Ruepp et al. 2004)) and ranked by decreasing degree of activation by SA

aGene previously reported by our group (Blanco et al. 2005) which induction ratio was quantified from Northern experiments; AGI, Arabidopsis Genome Initiative gene index; SA wt, average ratio SA/C in wt plants

Table 3

NPR1-independent SAIGs

AGI

Alias

SA wt

SA npr1

Annotation

FunCat Subcategories

References

Oxygen and radical detoxification

At1g28480

GRXC9

11.3

3.4

Glutaredoxin family protein

Oxygen and radical detoxification; electron transport

Ndamukong et al. (2007)

At1g03850

GRXS13

10.5

3.2

Glutaredoxin family protein

Oxygen and radical detoxification; electron transport; chemoperception and response; citokinin response; chloroplast

Brenner et al. (2005)

At2g47730a

AtGSTF8(GST6)

3.2

3.3

Glutathione S-transferase (phi class)

Disease, virulence and defense; oxygen and radical detoxification; metabolism of toxins/drugs; protein modification; chloroplast

Uquillas et al. (2004); Thatcher et al. (2007)

At2g29420

AtGSTU7 (GST25)

2.6

4.0

Glutathione S-transferase (tau class)

Metabolism of toxins/drugs; oxygen and radical detoxification; chemoperception and response; plant signaling molecules response

Wagner et al. (2002); Blanco et al. (2005)

Cellular sensing and response to external stimulus/disease, virulence and defense

At1g31580

ECS1

9.1

3.4

Cell wall protein linked to a locus influencing resistance to Xcc750

Disease, virulence and defense; cell wall

Aufsatz et al. (1998)

At3g01420a

α-DOX1

7.9

8.5

α-dioxygenase

Oxidative stress response; chemoperception and response; plant signaling molecules response; cell death; fatty acid metabolism

De Leon et al. (2002); Blanco et al. (2005)

At1g76680a

OPR1

7.3

8.1

12-oxophytodienoate reductase 1

Chemoperception and response; pant signaling molecules response; fatty acid derivatives mediated signal transduction; senescence; lipid metabolism; metabolism of acetic acid derivatives

He and Gan (2001); Blanco et al. (2005)

At4g14400

ACD6

6.5

2.8

Accelerated cell death 6, a novel ankyrin protein

Chemoperception and response; response to biotic stimulus; plant signaling molecules response; protein binding; cell death

Lu et al. (2003)

At4g19660

NPR4

5.4

3.3

NPR1-like protein 4

Response to biotic stimulus; protein binding; nucleus

Liu et al. (2005); Zhang et al. (2006)

At1g77120

ADH1

2.3

2.7

Alcohol dehydrogenase 1

Osmotic and salt stress response; osmosensing and response; C-compound and carbohydrate metabolism; metabolism of vitamins, cofactors, and prosthetic groups; respiration

Liu et al. (2005)

At2g24850

TAT3

2.3

2.1

Tyrosine aminotransferase 3

Stress response; chemoperception and response; plant signaling molecules response; response to wounding; aromatic aminoacid metabolism; nitrogen metabolism

Titarenko et al. (1997)

At1g07400

HSP17.8-CI

2.1

2.2

17.8 kDa class I heat shock protein

Oxidative stress response; heat shock response; temperature perception and response

Stanley et al. (2005)

Metabolism (Nucleotide-sugar metabolism, UGTs)

At1g05560a

UGT75B1 (UGT1)

5.8

5.7

UDP-Glycosyl transferase (Group L) involved in callose synthesis at the cell plate

Chemoperception and response; plant signaling molecules response; nucleotide-sugar metabolism; cell cycle; vesicle formation

Hong et al. (2001); Blanco et al. (2005)

At2g43820a

UGT74F2

4.7

5.1

UDP-Glycosyl transferase (Group L); involved in the tryptophan synthesis pathway; glycosylates SA in vitro

Nucleotide-sugar metabolism

Quiel and Bender (2003); Blanco et al. (2005)

At4g15550

UGT75D1

4.0

3.0

UDP-Glycosyl transferase (Group L)

Nucleotide-sugar metabolism; chloroplast

Jackson et al. (2001)

At4g01070

UGT72B1

3.4

2.1

UDP-Glycosyl transferase (Group E)

Degradation/modification of foreign (exogenous) compounds; chemoperception and response; nucleotide-sugar metabolism

Brazier-Hicks and Edwards (2005)

At4g34135a

UGT73B2

2.9

3.7

Flavonol 7-O-glucosyltransferase (Group D)

Nucleotide-sugar metabolism

Blanco et al. (2005); Langlois-Meurinne et al. (2005)

At4g34138a

UGT73B1

2.8

3.5

UDP-glycosyltransferase/abscisic acid glucosyltransferase (Group D)

Nucleotide-sugar metabolism

Blanco et al. (2005); Langlois-Meurinne et al. (2005)

Metabolism (secondary, carbohydrate, aminoacid and others)

At5g24530

F3′H

13.1

3.5

Oxidoreductase, 2OG-Fe(II) oxygenase family protein, similar to flavanone 3-hydroxylase

Metabolism of stilbenes, flavonoids

 

At5g51830

 

4.6

2.4

pfkB-type carbohydrate kinase family protein

C-compound and carbohydrate metabolism; acetate fermentation; phosphate metabolism

 

At3g16150

 

3.4

2.2

L-asparaginase, putative

Aminoacid metabolism; nitrogen metabolism; protein/peptide degradation

Bruneau et al. (2006)

At3g14990

 

2.8

2.5

4-methyl-5(b-hydroxyethyl)-thiazole monophosphate biosynthesis protein, thiamin biosynthesis

Biosynthesis of vitamins, cofactors, and prosthetic groups

 

At4g21990

ATAPR3

2.3

3.6

Adenosine 5’-phosphosulfate reductase-like protein; it has a domain of glutathione (GSH)-dependent reductase

Sulfur metabolism; chloroplast

Houston et al. (2005)

Transcription

At5g63790

ANAC102

5.3

4.7

Member of the plant-specific NAC-domain transcription factor family

Transcriptional control

Delessert et al. (2005)

At5g13080

ATWRKY75

3.3

2.3

Member of WRKY Transcription Factor; Group II-c

Transcriptional control; nucleus

Gechev et al. (2005); Devaiah et al. (2007)

Unclassified

At2g25510

 

13.1

4.3

Unknown protein

Unclassified protein

Sato et al. (2007)

At4g13180

SDRLP

4.9

5.2

Short-chain dehydrogenase/reductase (SDR) family protein

Mitochondrion

Catarecha et al. (2007)

At5g64250

NPD

2.8

2.6

2-nitropropane dioxygenase family/NPD family

Unclassified protein

Shahollari et al. (2007)

At5g39050

 

2.5

2.2

Transferase family protein

Unclassified protein

Tamaoki et al. (2003)

At3g25190

 

2.1

2.6

Nodulin, putative

Unclassified protein

Rampey et al. (2006)

At5g14730

 

2.0

2.8

Similar to unknown protein

Unclassified protein

Baerson et al. (2005)

All genes with induction ratio higher than 2 are included. Genes were classified in the more representative subcategories, as defined by MIPS (FunCat DB (Ruepp et al. 2004)) and ranked by decreasing degree of activation by SA

aGenes previously reported by our group (Uquillas et al. 2004; Blanco et al. 2005) which induction ratio was quantified from Northern experiments; AGI, Arabidopsis Genome Initiative gene index; SA wt, average ratio SA/C in wt plants; SA npr1, average ratio SA/C in npr1-1 mutant plants

Early NPR1-dependent and NPR1-independent SAIGs were classified in functional categories according to the Munich Information Center for Protein Sequences classification (MIPS) (Ruepp et al. 2004), using the FunCat database. Over-represented MIPS categories were identified by using the BioMaps tool from Virtual Plant database (http://virtualplant.bio.nyu.edu). Different sets of functional categories were overrepresented in NPR1-dependent and NPR1-independent SAIGs (Table 4). In the group of 197 early NPR1-dependent SAIGs, the categories “Metabolism” and “Protein Fate” (represented by subcategories “phosphate metabolism” and “protein modification by phosphorylation”) and the categories “Cell rescue, Defense and Virulence”/“Interaction with the environment”/“Cell fate” (subcategories associated to defense responses to pathogens) were overrepresented (Table 4). Accordingly, the functional groups mainly represented in this class of SAIGs were protein kinases (36 genes), disease resistance proteins (11 genes) and WRKY transcription factors (7 genes) (see Table 2 and supplementary Table S1).
Table 4

Functional categories over-represented in the groups of early NPR1-dependent and early NPR1-independent SA-induced genes, compared to the complete Arabidopsis genome

Term

Observed frequency

Expected frequency

P-value

NPR1-Dependent-SAIGs

Metabolism

35.2%

17.3%

1.61e−07

Phosphate metabolism

24.4%

6.55%

1.41e−07

Protein fate

28.0%

10.9%

3.31e−05

Protein modification

26.0%

7.61%

5.21e−05

    Modification by phosphorylation, dephosphorylation, autophosphorylation

19.8%

4.37%

2.23e−05

Cell rescue, defense and virulence

12.2%

5.01%

5.17e−03

Disease, virulence and defense

8.67%

1.61%

3.19e−06

Interaction with the environment

13.2%

5.81%

7.24e−03

Cellular sensing and response to external stimulus

13.2%

5.24%

1.36e−03

    Chemoperception and response

8.67%

2.95%

8.73e−03

    Response to biotic stimulus

6.12%

1.04%

1.48e−04

Systemic interaction with the environment

7.65%

2.66%

2.81e−02

Plant/fungal specific systemic sensing and response

7.14%

2.43%

3.69e−02

Cell fate

Cell death

3.06%

0.35%

9.95e−03

Total genes

197

28413

 

NPR1-Independent-SAIGs

Metabolism

54.8%

17.3%

1.77e−05

C-compound and carbohydrate metabolism

25.8%

5.79%

2.08e−02

Cell rescue, defense and virulence

32.2%

5.01%

1.22e−04

Detoxification

12.9%

0.93%

1.52e−02

    Oxygen and radical detoxification

12.9%

0.91%

1.42e−02

Interaction with the environment

35.4%

5.81%

5.28e−09

Cellular sensing and response to external stimulus

35.4%

5.24%

1.91e−09

    Chemoperception and response

29.0%

2.95%

1.45e−09

Systemic interaction with the environment

22.5%

2.66%

1.06e−03

    Plant/fungal specific systemic sensing and response

22.5%

2.43%

5.99e−04

Plant hormonal regulation

22.5%

2.07%

2.10e−04

Total genes

31

28413

 

The over-represented functional categories were obtained according to the MIPS classification using BioMaps tool from the VirtualPlant webpage (http://virtualplant.bio.nyu.edu). Observed frequency refers to the list of SAIGs analyzed and expected frequency refers to the whole Arabidopsis genome. Hypergeometric method and a Bonferroni correction were used for the analysis with a cutoff of 0.05

On the other hand, the over-represented categories in the 31 early NPR1-independent SAIGs were “Metabolism” (subcategory “C-compound and carbohydrate metabolism”) and “Cell Rescue, Defense and Virulence”/“Interaction with the environment” (subcategories associated to detoxification and response to chemicals and hormones) (Table 4). Consistently, 6 genes coding for UGTs and 4 genes coding for glutaredoxins and GSTs were identified in this class of SAIGs (Table 3).

To get a preliminary idea if the SAIGs identified in this study play a role in defense responses against different stressful stimuli, we analyzed the available transcriptomic expression data for these genes using GeneVestigator (http://www.genevestigator.ethz.ch/) (Laule et al. 2006). We did an independent search of the expression data for the list of early NPR1-dependent SAIGs that displayed an increase of more than 3 times in their steady-state mRNA levels in response to SA (listed in Table 2) and for the list of early NPR1-independent SAIGs (listed in Table 3). Early NPR1-independent SAIGs were more responsive than early NPR1-dependent SAIGs to a wide range of stress treatments (biotic, chemical and abiotic) that alter the redox cell homeostasis (Fig. 1). For example, early NPR1-independent SAIGs—in contrast to early NPR1-dependent SAIGs—were responsive to inoculation with Botrytis cinerea and to treatments with several chemical stressors (Fig. 1). In addition, this study showed that both groups of genes were responsive to SA but not to other hormones involved in stress (data not shown).
Fig. 1

Expression profiles of SAIGs obtained from public microarray database. The expression profiles of the NPR1-dependent SAIGs whose induction was >3-fold (see Table 2) and of all NPR1-independent SAIGs identified in this study (see Table 3) were analyzed on microarray experiments selected from the GeneVestigator database (Laule et al. 2006). Experiments with other stress treatments, either hormonal (ABA, ACC, ethylene, GA3, IAA), biotic or abiotic, in which SAIGs showed no response were not included. Expression levels (log2 ratio) are indicated by a color code shown at the bottom of the figure. The number of arrays used in each experiment is indicated at the right side. Genes with highest induction level in each group are marked with a green box (AT1G28480 for NPR1-independent and AT5G03350 for NPR1-dependent group)

With the double purpose of corroborating the results of the microarray experiment regarding the responsiveness to SA and the requirement of NPR1 coactivator observed in, and analyzing the kinetics of this process, we selected a group of representative genes according to the level of transcript accumulation by SA and to their function. Among the chosen genes were 5 early NPR1-dependent SAIGs (LLP, NIMIN1, NIMIN2, WRKY38, UGT76B1) and 5 early NPR1-independent SAIGs (GRXC9, GRXS13, F3′H, ANAC102, SDRLP) with high levels of transcript accumulation and representative of different functional categories (Tables 2, 3). OPR1 and PR-1 were used as controls for NPR1 independence and dependence, respectively (Blanco et al. 2005). The experiment involved treatment of wild type and npr1-1 mutant Arabidopsis seedlings with SA for different periods of time, and Northern blot analysis of transcripts levels (Fig. 2). As expected, all genes selected were activated by SA in wild type seedlings. We also confirmed that SA-induced activation of WRKY38, UGT76B1, NIMIN1, NIMIN2, LLP genes did not occur in the npr1-1 mutant, while activation by SA of SDRLP, ANAC102, GRXC9, GRXS13 and F3′H was detected in both genotypes (Fig. 2). Interestingly, the kinetic behavior of these genes after activation by SA was different and correlated with the degree of requirement of NPR1 for activation. Complete independence of NPR1 for SA-induced activation correlated with an earlier and transient activation by SA (SDRLP, ANAC102 and OPR1 in Fig. 2). As previously reported (Blanco et al. 2005), we found that activation of early NPR1-independent SAIGs by SA in the npr1-1 mutant had a more sustained activation (Fig. 2), which correlated with higher accumulation of SA (data not shown). Genes with partial requirement of NPR1 showed an early but more sustained activation by SA (GRXC9, GRXS13, F3′H in Fig. 2). Consistent with our microarray data (Table 3), activation of these genes was detected in the npr1-1 mutant but a functional NPR1 was required for maximum activation rate.
Fig. 2

Time-course of SA-induced expression of selected SAIGs in wild type and npr1-1 plants. RNA was isolated from 15 day-old plants either treated with 0.5 mM SA (SA) or maintained in MS medium (C) for the times (hours) indicated at the top of the figure. Ten micrograms of total RNA were loaded per lane, and the blots were hybridized with specific probes for the following genes: SDRLP (AT4G13180), ANAC102 (AT5G63790), GRXC9 (AT1G28480), GRXS13 (AT1G03850), F3′H (AT5G24530), WRKY38 (AT5G22570), UGT76B1 (AT3G11340), NIMIN-1 (AT1G02450), NIMIN-2 (AT3G25882) and LLP (AT5G03350). OPR1 (AT1G76680) and PR1 (AT2G14610) were used as controls for NPR1 independence and dependence, respectively. ACTIN 3 (AT3G53750) was used as a loading control

All genes classified as early NPR1-independent SAIGs had a peak of increase in mRNA levels around 1–2.5 h after SA treatment. In contrast, early NPR1-dependent SAIGs (WRKY38, UGT76B1, NIMIN1, NIMIN2, LLP in Fig. 2) displayed a more delayed kinetics (peak of mRNA levels after 5 h). The activation kinetics of both classes of early SAIGs was clearly distinguishable from that of PR-1 gene, a marker for “late NPR1-dependent SAIGs” (Fig. 2).

SA-mediated activation of selected SAIGs

To study the mechanism of SA-induced activation we selected six SAIGs: two completely independent of NPR1 (ANAC102 and SDRLP), two genes partially dependent on this coactivator (GRXC9 and GRXS13) and two completely dependent on NPR1 for activation (NIMIN1 and LLP).

In the first place, we determined if activation of these genes required de novo protein synthesis. Wild-type seedlings were pretreated for 1 h with the protein synthesis inhibitor cycloheximide (CHX) (or MS as a control), and then treated for 2.5, 8 or 24 h with SA (or MS as a control). After this, the steady state levels of transcripts were determined by Northern blot. Activation of SDRLP, ANAC102, GRXS13, GRXC9 and NIMIN1 genes by SA was not inhibited by CHX (Fig. 3a). Therefore, SA increases the mRNA levels of these genes by a mechanism independent of de novo protein synthesis and likely controlled by modification of pre-existing factors. Interestingly, the genes ANAC102 and GRXC9 were strongly activated by CHX in the absence of SA (Fig. 3a). This behavior was previously reported by us and others in early SAIGs of Arabidopsis and tobacco (Qin et al. 1994; Horvath et al. 1998; Uquillas et al. 2004). As judged by information from public microarray database, it seems to be a common response of a large number of early SAIGs (see response to CHX in Fig. 1). A possible explanation for this effect is that one of the mechanisms that controls transcription of these early SAIGs is basal repression by a labile protein, whose effect can be relieved by SA (see model in Fig. 6). In the case of LLP, pretreatment with CHX partially inhibited the activation by SA. This is compatible with a mixed activation mechanism. In contrast, we found that SA-mediated activation of PR-1, a marker for late genes, was completely inhibited by CHX, as previously reported (Qin et al. 1994).
Fig. 3

Analysis of SA-induction mechanism for 6 selected SAIGs. a Effect of the inhibitor of protein synthesis, cycloheximide (CHX) on SAIGs expression. 15 day-old wild type seedlings were pretreated for 1 h with CHX (six lines on the right side) or MS (control, six lines on the left sides) and then incubated for 2.5, 8 or 24 h with 0.5 mM SA or MS as a control (C), either in the presence or absence of CHX. Ten μg of total RNA were loaded per lane, and the blots were hybridized with specific probes for the genes indicated in the figure. PR1 and 18S rRNA genes were used as controls for CHX treatment and loading, respectively. b Effect of the transcriptional inhibitor cordycepin on SAIGs expression. 15 days-old wild type seedlings were pretreated with 0.5 mM SA or MS as a control (C) for 2 h. Then, seedlings were transferred to MS or 0.5 mM SA in the absence (left panel) or presence of 0.6 mM cordycepin (right panel) for the times (hours) indicated at the top of the figure. RNA was analyzed as described above. NIA2 (AT1G37130) and ACTIN 3 (AT3G53750) were used as controls of cordycepin effect and loading, respectively

To determine if the early increases of mRNA levels of selected SAIGs caused by SA result from mRNA stabilization or from transcriptional activation, we analyzed the expression of these genes in the presence of cordycepin, a transcriptional inhibitor (Gutierrez et al. 2002). Wild type seedlings were treated with SA for 2.5 h to produce the maximum transcript accumulation, and then incubated for different periods of time with SA or MS medium (control) in the presence or absence of cordycepin (0.6 mM). The mRNAs were detected by Northern blot analysis and quantified by a Phosphorimager. The effect of cordycepin was controlled by analyzing the levels of NIA2 mRNA (nitrate reductase 2), a constitutively expressed gene not responsive to SA. In the absence of cordycepin, removal of SA produced a decay of mRNA levels for all SA-responsive genes (Fig. 3b, compare C vs. SA in left panels). When transcription was inhibited by cordycepin, a similar decay of mRNA levels was detected in the absence or in the presence of SA for all genes analyzed (Fig. 3b, compare C vs. SA in right panels). To support this conclusion, we also calculated mRNA half-lives (t1/2) for all genes analyzed, in the absence and in the presence of SA, using three independent biological samples. For this, we quantified mRNA levels by Phosphorimager and the t1/2 values were calculated according to Lidder et al. (2005). Results from this analysis indicate that SA does not affect t1/2 of SAIGs mRNAs (data not shown). These results are consistent with the idea that SA does not increase mRNA levels of early SAIGs through increasing mRNA stability, and support the idea that SA induces gene activation by exerting a transcriptional control.

We then determined whether the subclass II of TGA transcription factors (TGA2, TGA5 and TGA6) was involved in the transcriptional activation or de-repression of these early genes by SA. For this purpose, we compared the response of selected genes to SA treatment in wild type and tga2-1/tga5-1/tga6-1 triple mutant Arabidopsis seedlings (Zhang et al. 2003). As shown in Fig. 4, activation of the early NPR1-independent (SDRLP, ANAC102, GRXC9 and GRXS13) and the early NPR1-dependent (LLP and NIMIN1) SAIGs was severely impaired in the tga2-1/tga5-1/tga6-1 triple mutant, indicating involvement of TGA2/5/6 in the transcriptional activation of these genes. In contrast, an increase in basal levels and impairment of SA-activation at 8, but not at 24 h post treatment, was detected for PR-1 gene in the tga2-1/tga5-1/tga6-1 triple mutant (Fig. 4). This behavior is consistent with the possibility that the TGA2/5/6 factors are involved in both the basal repression and the SA-mediated activation of PR-1 (Zhang et al. 2003; Rochon et al. 2006), but it also suggests the involvement of other factors in late activation of PR-1 by SA (Wang et al. 2005).
Fig. 4

SA-induced expression of selected SAIGs in tga6-1/tga2-1/tga5-1 plants. RNA was isolated from 15 day-old plants either treated with 0.5 mM SA (SA) or maintained in MS medium (c) for the times (hours) indicated at the top of the figure. Ten micrograms of total RNA were loaded per lane, and the blots were hybridized with specific probes for the indicated genes. ACTIN 3 was used as a loading control

SA-induced genes revealed common cis-elements in their promoters

Having found that the SAIGs selected in this study were transcriptionally activated by SA through a TGA-dependent mechanism, we looked for common putative cis-elements in the promoter sequences of both groups of SAIGs. We searched for over-represented motifs (16 and 20 bp length) separately in the 31 early NPR1-independent SAIGs (Table 3) and in the 197 early NPR1-dependent SAIGs (supplementary Table S1), with the software MotifSampler (Thijs et al. 2002). Over-represented motifs were ranked according to their log-likelihood score, which considers the conservation of the motifs and the number of copies of each motif found in the input sequences. Table 5 shows the 10 best-ranked motifs for each group of genes. To determine if the motifs found in this study correspond to defined regulatory elements already described in plant promoters, we searched for these motifs in PLACE and TRANSFAC databases (Wingender et al. 1996; Higo et al. 1999). Interestingly, the 10 best ranked 20 bp motifs found in the early NPR1-independent SAIGs, matched the as-1-like element described in the 35S promoter of the cauliflower mosaic virus (Qin et al. 1994). This is in agreement with results we described before for a smaller group of early NPR1-independent SAIGs (Blanco et al. 2005). On the other hand, for early NPR1-dependent SAIGs, the 10 best ranked 16 bp motifs found matched the core of TL-1 element (Table 5). This element was recently described as a consensus sequence over-represented in the promoter of a group of 13 NPR1-responsive ER-resident genes (Wang et al. 2005).
Table 5

In silico promoter analysis of early SAIGs

Early SAIGs are activated in the defense reaction triggered by an avirulent bacterium

Arabidopsis plants activate SA biosynthesis as part of their defense reaction to the attack of avirulent pathogens (Summermatter et al. 1995). We wanted to know whether the expression of the early SAIGs identified here increased during the activation of pathogen-induced defenses. For this purpose, we inoculated Arabidopsis plants with the avirulent pathogen Pseudomonas syringae pv. tomato DC3000 that carries the AvrRpm1 gene (Pst AvrRpm1, 5 × 106 cfu/ml); control plants were mock inoculated. RNA samples were obtained at 0, 5, 21 and 48 h post inoculation (hpi) and transcript levels of the early SAIGs were detected by Northern blot. As shown in Fig. 5a, activation of all SAIGs analyzed (SDRLP, ANAC102, GRXS13, GRXC9, NIMIN1 and LLP) was detected starting 5 h post infection (hpi). SA levels were measured by HPLC in pathogen-treated samples collected at 5, 26 and 48 hpi. Increase in SA levels was detected from 5 hpi (Fig. 5b). All genes, except GRXS13, showed a peak of activation earlier than PR-1. These results support a role for SAIGs in the defense response to pathogen infection.
Fig. 5

Expression of SAIGs and SA levels in Arabidopsis plants inoculated with Pseudomonas. Six week-old plants were inoculated with Pseudomonas syringae pv tomato DC3000 carrying the AvrRpm1 gene (Pst AvrRpm1, 5 × 106 cfu/ml) or with MgCl2 (mock). Then, leaf samples were collected at the indicated times for RNA extraction and Northern blot analysis (a), and for measurement of SA levels by HPLC (b). SAG, glycosilated-SA. Averages and standard deviations of SA and SAG levels were calculated from three independent biological replicates

Discussion

It has been clearly established that the rise of SA levels and the subsequent induction of genes triggered by this hormone (SAIGs) are essential steps for the development of a defense reaction to pathogen infection (SAR), and to other stressful environmental conditions (Durrant and Dong 2004; Loake and Grant 2007). Although important knowledge concerning SA effects has been obtained in the last years, essential elements of its action in the cellular response to stress are still unknown (Loake and Grant 2007). Looking for clues to understand SA role in defense to stress, we have focused our attention in the early responses triggered by this hormone. We previously described two pathways for rapid activation of SAIGs; one dependent on the NPR1 coactivator and other independent of it (Uquillas et al. 2004; Blanco et al. 2005; Holuigue et al. 2007).

The transcriptomic and expression analyses described in this article further support the existence of these two pathways for rapid activation of SAIGs; one independent of NPR1 used by 14% of these genes, and the other dependent of NPR1 used by the rest of them. Here we report that these two groups of genes, not only differ in their main functional categories, but also in their timing and mechanism of activation by SA. Considering this and previous published work (Niggeweg et al. 2000; Pontier et al. 2001; Uquillas et al. 2004; Blanco et al. 2005; Thurow et al. 2005; Wang et al. 2005, 2006; Butterbrodt et al. 2006; Rochon et al. 2006) we propose a model to explain the differences in the SA-mediated activation of these two classes of early SAIGs (Fig. 6).
Fig. 6

Mechanistic model for SA-mediated early activation of genes. a Temporal course of the reaction after SA addition is indicated by the color bar at the top of the figure. Windows corresponding to peaks of induction of two classes of early SAIGs are indicated. “Early NPR1-independent SAIGs” are rapidly and transiently activated by SA by reversible recruitment of a coactivator (CoA) to the TGA2/as-1-like element (T2/as-1) complex. Putative inhibitors (I) that recognize CoA and/or TGA2 could be involved. b “Early NPR1-dependent SAIGs” are activated by SA via recruitment of activated NPR1 (NPR1A) to form the transcriptionally active NPR1A-T2-TGA ternary complex. We also propose involvement of a putative transcriptional activator (TL-1-A) that recognizes the TL-1 element. as-1-like element: two TGACGTCA-like motifs spaced by 4 nucleotides; TGA box: TGACG; TL-1 box: TTCTTCTTC. Red arrows represent activation and blocked red arrows represent repression

Role of early SAIGs in the stress defense response

Several functions in the defense response to stress have been attributed to the different classes of SAIGs. An antimicrobial role has been reported for PR genes, the well known group of genes belonging to the class of late NPR1-dependent SAIGs (Maleck et al. 2000; van Loon et al. 2006). SAIGs recently identified as primary targets of NPR1, which can be classified as early NPR1-dependent SAIGs, code for proteins from the protein secretory machinery and WRKY factors (Wang et al. 2005, 2006). The identification of early SAIGs (NPR1-dependent and NPR1-independent pathways) described in this work supports the idea that SA plays additional roles in defense and acclimatory responses to stress, such as recovery of the cell redox balance, intracellular stress signaling, improvement of pathogen recognition, and promotion of metabolic changes. Based on the expression analysis of a selection of these genes in response to an avirulent bacterial pathogen, we propose that early SAIGs may contribute to generate a pathogen-induced defense reaction. Intersection of our list of 228 early SAIGs with the list of 1,187 genes activated after 8 h of treatment with BTH (Wang et al. 2006), indicates that 88 out of the 228 genes (38%) had not been previously considered as BTH-dependent SAIGs. A considerable proportion of the genes identified here (133 genes, 58%) were also detected as ozone-responsive genes (Mahalingam et al. 2005), indicating that early SAIGs also play a role in the response to ozone treatment. We will focus our discussion in some of the functions early activated by SA, which could be more relevant to the role of this hormone in defense and acclimatory responses.

The regulation of changes in the cellular redox status is one of the roles recently attributed to SA in pathogen-induced defenses and light acclimatory responses in Arabidopsis (Durrant and Dong 2004; Fobert and Despres 2005; Foyer and Noctor 2005; Mateo et al. 2006; Holuigue et al. 2007). This regulatory role seems relevant to SA function, considering that this hormone is also produced and required for defense responses to different stressful conditions associated to cellular oxidative stress, such as UV irradiation, ozone exposure and high light (Yalpani et al. 1994; Surplus et al. 1998; Rao and Davis 1999; Mateo et al. 2006; Janda et al. 2007). A pro-oxidative activity of SA recognized more than 15 years ago (Chen et al. 1993) gives support to the positive feedback interaction between SA and reactive oxygen species (ROS) associated to cell death (Shirasu et al. 1997; Overmyer et al. 2003; Dat et al. 2007). Nevertheless, evidences of an antioxidant activity of SA associated to defense and acclimation processes have appeared only in the last years (Despres et al. 2003; Mou et al. 2003; Yang et al. 2004; Mateo et al. 2006; Tada et al. 2008). The molecular basis for this antioxidant activity is still unknown.

Results reported here support an antioxidant role for SA and strengthen previous evidences showing that SA triggers rapid activation of genes with antioxidant and detoxifying activities, through an NPR1-independent pathway (Uquillas et al. 2004; Blanco et al. 2005; Holuigue et al. 2007). Interestingly, in this class of early NPR1-independent SAIGs we detected genes coding for two glutaredoxins (GRXC9 and GRXS13), two glutathione S-transferases (AtGSTU7 and AtGSTF8) and 6 UDP-glycosyl transferase (UGT72B1, UGT75B1,UGT75D1, UGT74F2, UGT73B2, UGT73B1) (Table 3).

GRXs, together with thioredoxins (TRX), are the two major enzymatic systems that catalyze reversible thiol-based reduction of target proteins (Rouhier et al. 2008). These proteins have been allegedly implicated in the SA-dependent reduction of NPR1 and TGA factors required for the transcription of defense genes (Fobert and Despres 2005; Foyer and Noctor 2005; Tada et al. 2008). The glutaredoxin GRXC9 (coded by At1g28480, also named GRX480) was recently reported to be induced by SA and to interact with TGA factors (Ndamukong et al. 2007). Ectopic expression of GRXC9 suppresses JA-responsive PDF1.2 transcription, suggesting that this enzyme could be a key piece in the SA/JA cross-talk (Ndamukong et al. 2007). Here we report that GRXC9 gene is rapidly activated by SA (peak 1–2.5 h, see Fig. 2), and by infection with an avirulent pathogen (Fig. 5). We also found that activation of GRXC9 by SA is partially dependent on NPR1 function, because considerable levels of GRXC9 mRNA were detected in the npr1.1 mutant compared to the wild type (50% at 2.5 h, Fig. 2). In contrast, Ndamukong et al. reported that activation of GRXC9 by SA is almost negligent in the npr1.1 background (Ndamukong et al. 2007). These differences could be due to differences in the experimental conditions used in both experiments; we used 15 days-old seedlings treated with 0.5 mM SA, while Ndamukong et al. used 3 week-old plants treated with 1 mM SA.

The other glutaredoxin gene (GRXS13, At1g03850), which is also included in the group of early NPR1-independent SAIGs, codes for an enzyme whose function in defense has not been studied. Here we show that SA activates GRXS13 with an early and sustained kinetics (Fig. 2). Similarly, a strong and sustained activation of this gene was also detected after pathogen infection (Fig. 5). As with GRXC9, we included GRXS13 in the list of early NPR1-independent SAIGs, due to the considerable levels of activation by SA detected in the npr1-1 mutant (Fig. 2).

We found that SA also produces a rapid activation of two GSTs and six UGTs, by an NPR1-independent pathway. These transferases catalyze the formation of S-glutathionylated and O-glycosylated conjugates of different acceptor molecules. As we previously discussed (Blanco et al. 2005; Holuigue et al. 2007), these enzymes have been implicated in the detoxification of endogenous and xenobiotic compounds, being good candidates to diminish toxic effects of metabolites (including ROS) in the stress defense reaction (Lim and Bowles 2004; Edwards et al. 2005).

In sum, a rapid induction of this group of 10 enzymes (GRXs, GSTs and UGTs) following the rise in SA levels during the pathogen-induced defense reaction could contribute to recover the cellular redox balance and to generate the reducing environment for NPR1/TGA reduction required for the activation of late SAIGs responsible for stress resistance (Despres et al. 2003; Mou et al. 2003).

This study indicates additional functions of the rapid cellular response triggered by SA. A role for SA in improving pathogen recognition and signaling for defense response is also suggested by the identification of genes coding for 11 disease resistance (R) proteins, 36 protein kinases and 28 proteins with protein binding functions among the early SAIGs. Activation of genes coding for transcription factors belonging to different families (8 WRKY, 4 NAM, 3 ERF/AP2 and 2 myb) is also part of the rapid response triggered by SA. Only two of these genes are activated by an NPR1-independent pathway, ANAC102 (NAM family) and WRKY 75. Here we show that ANAC102 is rapidly activated by SA (Fig. 2) and also by treatment with an avirulent pathogen (Fig. 5). Although ANAC102 has not been previously characterized, is highly homologous to ATAF2, previously described as a repressor of PR genes in Arabidopsis (Delessert et al. 2005). WRKY75, together with the NPR1-dependent WRKY 46, are the only WRKYs detected here as early SAIGs, that have not been previously associated to pathogen-induced defense responses (Eulgem and Somssich 2007). WRKY75 has been reported to be involved in H2O2-induced cell death and as a modulator of phosphate acquisition and root development (Gechev et al. 2005; Devaiah et al. 2007). The function of ANAC102 and WRKY75 in the defense reaction in Arabidopsis remains to be characterized.

Mechanisms for defense genes activation triggered by SA

The results of the genome-wide analysis reported here give further support to the existence of different mechanisms responsible for the rapid activation of SAIGs. Based on the study of representative members of the SAIGs identified in this work, we propose that SA-mediated gene activation involves transcriptional activation and/or de-repression. The mechanisms involved differ in their requirement for NPR1, TGA factors and SA-responsive promoter elements. Based on this and on previous mechanistic studies, we propose an explanatory model for the SA-induced activation of the two classes of early SAIGs that takes place at different times after exposure to SA (Fig. 6).

A cluster of 31 genes corresponding to the class of early NPR1-independent SAIGs, activated by SA after 0.5–2.5 h of treatment, has been identified in this (Table 3) and previous work from our group (Uquillas et al. 2004; Blanco et al. 2005). We propose that early transcriptional activation of these genes by SA is controlled by the as-1-like element (TGACGTCAnnnnTGACGTCA) that is over-represented in the group of 31 promoters (Table 5). As we previously discussed (Blanco et al. 2005), as-1 is structurally and functionally different from isolated TGA boxes (TGACG) found in PR-1 promoter (Krawczyk et al. 2002; Butterbrodt et al. 2006). as-1 element confers immediate early responsiveness to SA (Qin et al. 1994) by a mechanism involving oxidative signals (Garreton et al. 2002). We think that binding of a member of the subclass II of TGA factors (TGA2/5/6) to the as-1-like element is required for activation of these genes by SA. This is supported by our evidence that activation of SDRLP, ANAC102, GRXC9, and GRXS13 (representative early NPR1-independent SAIGs) in Arabidopsis is severely impaired in the tga2-1/tga5-1/tga6-1 triple mutant (Fig. 4). Similarly, evidence obtained in tobacco indicates that TGA2.2, member of the subclass II of TGA factors in this species, is the main component of the as-1 binding activity in nuclear extracts (Niggeweg et al. 2000) and is essential for activation of a couple of tobacco early genes (Pontier et al. 2001; Thurow et al. 2005). The question that remains open is how SA activates transcription of early NPR1-independent SAIGs mediated by TGA2/5/6-as-1 interaction. One possibility is that SA activates binding of TGA2/5/6 to the as-1 element. This possibility has been explored in tobacco with contradictory results (Jupin and Chua 1996; Stange et al. 1997; Garreton et al. 2002; Butterbrodt et al. 2006). Furthermore, it must be mentioned that neither TGA2 from Arabidopsis (Rochon et al. 2006) nor TGA2.2 from tobacco (Thurow et al. 2005) have transactivation activity. Therefore, other protein(s) must provide the transactivation domain to the TGA2/5/6-as-1 complex. In sum, as shown in Fig. 6, we propose that SA activates early NPR1-independent SAIGs by recruitment of a co-activator protein (CoA) to the TGA2-as-1 complex already formed under basal conditions. The presence of an inhibitor, which sequesters the coactivator or blocks the TGA2-as-1 complex under basal conditions has been previously proposed in tobacco (Johnson et al. 2001; Butterbrodt et al. 2006) and is consistent with the activation induced by CHX found in some of the early genes (Fig. 1 and (Uquillas et al. 2004).

Genes belonging to the class of early NPR1-dependent SAIGs have been identified in this (Table 2 and supplementary Table S1) and in previous studies oriented to identifying primary targets of NPR1 (Wang et al. 2005, 2006). Representative members of this group show a peak of induction between 5 and 12 h post SA treatment (WRKY38, UGT76B1, NIMIN1, NIMIN2, LLP in Fig. 2; BiP2 in (Wang et al. 2005)), while induction of PR-1 gene, a marker for late NPR1-dependent SAIGs, peaks after 24 h of treatment (Fig. 2).

A new SA-responsive promoter element named TL-1 element (TTCTTCTTC) was found to be over-represented in a group of early NPR1-dependent SAIGs (Wang et al. 2005). This element is functional in the Arabidopsis BiP2 promoter where SA activates binding of factors to TL-1 element by promoting its translocation to the nucleus (Wang et al. 2005). The nature of the factor(s) that recognize this TL-1 sequence is still unknown. Interestingly, we found the same TL-1 element over-represented in the group of 197 early NPR1-dependent SAIGs identified in this work (Table 5). Furthermore, expression analysis of NIMIN1 and LLP (representative early NPR1-dependent SAIGs) in response to SA in the tga2-1/tga5-1/tga6-1 triple mutant also indicates involvement of TGA2/5/6 factors in the activation of these genes by SA (Fig. 4). We propose that TGA boxes (TGACG), such as those found in the Arabidopsis PR-1 promoter, are recognized by TGA2/5/6 factors in promoters of early NPR1-dependent SAIGs (Fig. 6). In this case, requirement of NPR1 for activation of these genes by SA (Fig. 2), together with data from the literature indicating that NPR1 provides the transactivation activity to the TGA2-TGA box complex (Rochon et al. 2006), allow us to propose involvement of the NPR1-TGA2-TGA box ternary complex in the activation by SA of early NPR1-dependent SAIGs (Fig. 6). In this case, the question of how is NPR1 activated/reduced without de novo protein synthesis at the time these genes are induced remains open.

Mechanistic studies performed with PR1 gene, a marker for the group of late NPR1-dependent SAIGs (Lebel et al. 1998; Maleck et al. 2000; Rochon et al. 2006; Wang et al. 2006; Eulgem and Somssich 2007; Kesarwani et al. 2007) support the involvement of positive and negative W and TGA boxes in its SA-mediated activation, recognized by WRKY and TGA2/5/6 factors.

Taken together, our findings support the idea that SA can use different mechanisms to activate early and late SAIGs. Proper combination of SA-responsive elements (as-1-like, TGA box, W box and TL-1 element) in SAIGs promoters which are recognized by factors that act as repressors, or by complexes of factors and co-activators acting as enhancers, seems to be responsible for SA-controlled expression of these genes. Temporal control of gene expression by SA could be explained by reversible changes in the cellular redox homeostasis produced by SA. Interestingly, components of the transcriptional machinery involved in the expression of early and late SAIGs, such as NPR1, TGA factors, TGA box and as-1-like elements, seem to act as redox sensors for temporal control of gene expression by SA.

Acknowledgements

The authors are greatly indebted to Xin Li (Department of Botany, University of British Columbia, Canada) for providing the tga6-1 tga2-1 tga5-1 mutant and to Rodrigo Gutierrez (Departamento de Genética Molecular y Microbiología, Pontificia Universidad Católica de Chile) for helping with the over-representation analysis of genes categories by using the VirtualPlant webpage. This work was supported by grants from FONDECYT-CONICYT (1060494), Millennium Science Initiative (Nucleus for Plant Functional Genomics, P06-009-F) (to L.H.), ANPCyT (BID 1728/OC AR PICT 32637), CONICET, and SECyT-UNC (to M.E.A), and CONICYT/SECyT cooperation program (2005-7-186) (to L.H and M.E.A.).

Open Access

This article is distributed under the terms of the Creative Commons Attribution Noncommercial License which permits any noncommercial use, distribution, and reproduction in any medium, provided the original author(s) and source are credited.

Supplementary material

11103_2009_9458_MOESM1_ESM.xls (96 kb)
Supplementary Table S1. Genes differentially activated by SA in wild type and npr1-1 Arabidopsis seedlings. Supplementary material 1 (XLS 96 kb)

Copyright information

© The Author(s) 2009

Authors and Affiliations

  • Francisca Blanco
    • 1
  • Paula Salinas
    • 1
  • Nicolás M. Cecchini
    • 2
  • Xavier Jordana
    • 1
  • Paul Van Hummelen
    • 3
  • María Elena Alvarez
    • 2
  • Loreto Holuigue
    • 1
  1. 1.Departamento de Genética Molecular y Microbiología, Facultad de Ciencias BiológicasPontificia Universidad Católica de ChileSantiagoChile
  2. 2.CIQUIBIC-CONICET; Departamento de Química Biológica, Facultad de Ciencias QuímicasUniversidad Nacional de CórdobaCordobaArgentina
  3. 3.MicroArray Facility, Flanders Interuniversitary Institute for Biotechnology (VIB)LeuvenBelgium