Seediscussions,stats,andauthorprofilesforthispublicationat:https://www.researchgate.net/publication/26891169

ALesion-MimicSyntaxinDoubleMutantinArabidopsisRevealsNovelComplexityofPathogenDefenseSignaling

ArticleinMolecularPlant·May2008

DOI:10.1093/mp/ssn011·Source:PubMed

CITATIONS

38

READS

82

10authors,including:

Someoftheauthorsofthispublicationarealsoworkingontheserelatedprojects:

SignalsintheSea:theeffectofcopepodamidesonplanktoncommunitystructureandfunctioningView

project

MolecularmechanismofcellcycleViewproject

ZiguoZhang

UniversityofCambridge

87PUBLICATIONS3,409CITATIONS

SEEPROFILE

MatsXAndersson

UniversityofGothenburg

61PUBLICATIONS1,620CITATIONS

SEEPROFILE

CarstenPedersen

UniversityofCopenhagen

77PUBLICATIONS1,350CITATIONS

SEEPROFILE

Mari-AnneNewman

UniversityofCopenhagen

62PUBLICATIONS3,837CITATIONS

SEEPROFILE

AllcontentfollowingthispagewasuploadedbyCarstenPedersenon03December2016.

Theuserhasrequestedenhancementofthedownloadedfile.Allin-textreferencesunderlinedinblueareaddedtotheoriginaldocumentandarelinkedtopublicationsonResearchGate,lettingyouaccessandreadthemimmediately.

Molecular Plant • Volume 1 • Number 3 • Pages 510–527 • May 2008 RESEARCH ARTICLE

A Lesion-Mimic Syntaxin Double Mutant inArabidopsis Reveals Novel Complexity ofPathogen Defense Signaling

Ziguo Zhanga, Andrea Lenka,Mats X. Anderssona, Torben Gjettingb, Carsten Pedersena,Mads E. Nielsena,d,Mari-Anne Newmanc, Bi-Huei Houb, Shauna C. Somervilleb and Hans Thordal-Christensena,1

a Plant and Soil Science, Dept of Agricultural Sciences, Faculty of Life Sciences, University of Copenhagen, Thorvaldsensvej 40, DK-1871 Frederiksberg C,Denmarkb Dept of Plant Biology, Carnegie Institution, Stanford, CA 94305, USAc Plant Pathology, Dept of Plant Biology, Faculty of Life Sciences, University of Copenhagen, Thorvaldsensvej 40, DK-1871 Frederiksberg C, Denmarkd Research Laboratory, Carlsberg Research Center, Gl. Carlsbergvej 10, DK-2500 Valby, Copenhagen, Denmark

ABSTRACT The lesion-mimic Arabidopsismutant, syp121 syp122, constitutively expresses the salicylic acid (SA) signaling

pathway and has low penetration resistance to powdery mildew fungi. Genetic analyses of the lesion-mimic phenotype

have expanded our understanding of programmed cell death (PCD) in plants. Inactivation of SA signaling genes in syp121

syp122 only partially rescues the lesion-mimic phenotype, indicating that additional defenses contribute to the PCD.Whole

genome transcriptome analysis confirmed that SA-induced transcripts, as well as numerous other known pathogen-

response transcripts, are up-regulated after inactivation of the syntaxin genes. A suppressor mutant analysis of syp121

syp122 revealed that FMO1, ALD1, and PAD4 are important for lesion development. Mutant alleles of EDS1, NDR1,

RAR1, and SGT1b also partially rescued the lesion-mimic phenotype, suggesting that mutating syntaxin genes

stimulates TIR-NB-LRR and CC-NB-LRR-type resistances. The syntaxin double knockout potentiated a powdery mildew-

induced HR-like response. This required functional PAD4 but not functional SA signaling. However, SA signaling poten-

tiated the PAD4-dependent HR-like response. Analyses of quadruple mutants suggest that EDS5 and SID2 confer separate

SA-independent signaling functions, and that FMO1 and ALD1 mediate SA-independent signals that are NPR1-dependent.

These studies highlight the contribution of multiple pathways to defense and point to the complexity of their interactions.

INTRODUCTION

Plants have developed several different inducible defense

mechanisms against microbial pathogens. These defenses

are controlled by a number of signaling cascades. The best de-

scribed are the salicylic acid (SA), jasmonic acid (JA) and ethyl-

ene (ET) signaling pathways (Glazebrook, 2005). Signaling

components in each of these pathways have been discovered

primarily through mutant screens in Arabidopsis thaliana

(Durrant and Dong, 2004; Glazebrook, 2005; Beckers and

Spoel, 2006; Broekaert et al., 2006). However, the different

pathways do not function independently. The SA pathway,

for instance, can act antagonistically to the JA pathway. This

antagonism requires the SA-activated NPR1 (Spoel et al.,

2003). Disease resistance is manifested downstream of these

signaling pathways, through the expression of proteins that

counteract the invading pathogen. In addition to SA-, JA-,

and ET-regulated defenses, a major defense mechanism is

the hypersensitive necrosis response (HR). This pathogen-

induced programmed cell death (PCD) of one or more host

cells confers effective protection against biotrophic patho-

gens, which require living host tissue for their survival. The

HR is an important element of resistance (R) gene-mediated

resistance (Glazebrook, 2005).

Mutants that constitutively express defense responses,

many of which are lesion-mimic mutants with spontaneous

PCD, have been essential for unraveling defense signaling

pathways (Lorrain et al., 2003). Most of these mutations are

recessive, suggesting that their wild-type alleles encode neg-

ative regulators of defense signaling. Many of the lesion-

mimic mutants have more than one permanently activated

signaling pathway. However, the SA pathway often predomi-

nates in part due to its suppression of other pathways. There-

fore, when the SA pathway is silenced in lesion-mimic mutants,

1 To whom correspondence should be addressed. E-mail htc@life.ku.dk,

fax +45 35333460, tel +45 35333443.

ª The Author 2008. Published by the Molecular Plant Shanghai Editorial

Office in association with Oxford University Press on behalf of CSPP and

IPPE, SIBS, CAS.

doi: 10.1093/mp/ssn011, Advance Access publication 15 April 2008

up-regulation of the JA and ET pathways is often observed

(Clarke et al., 2000). The spontaneous PCD in lesion-mimic

mutants is often used as a model for HR-cell death reactions.

However, while R-gene-mediated HR appears to be stimulated

by SA (Rairdan and Delaney, 2002; Raffaele et al., 2006), the role

of SA in spontaneous PCD is ambiguous. There are examples of

both SA-dependent (Brodersen et al., 2005; Torres et al., 2005)

and SA-independent spontaneous PCD (Hunt et al., 1997).

Plants have developed a specific mechanism to block entry

of fungi that attack by penetrating through the plant epider-

mal cell wall. In Arabidopsis, we identified a series of pen

mutants that are compromised in penetration resistance to

the non-adapted barley powdery mildew fungus (Blumeria

graminis f.sp. hordei, Bgh) (Collins et al., 2003; Lipka et al.,

2005; Stein et al., 2006). Cloning of the Arabidopsis PEN1 gene

and the homologous barley ROR2 gene showed that they

encoded SYP121 syntaxins (Collins et al., 2003). SYP121 is local-

ized in the plasma membrane and was subsequently found to

be required for timely formation of the cell wall appositions,

which are important for this defense (Assaad et al., 2004).

Syntaxins are essential proteins of the SNARE machinery, con-

trolling vesicle traffic and bulk transport of cargo in cells (Lipka

et al., 2007). SYP121 is thought to influence the secretion

of components required for formation of pathogen-induced

cell wall appositions.

We have recently demonstrated that syntaxin double

mutants, of any of four alleles of syp121 and a T-DNA insertion

allele of syp122, become dwarfed and develop severe necrosis

after a period of 2–3 weeks of normal growth. These lesion-

mimic plants are affected in several different pathogen de-

fense pathways. The syntaxin double mutants lack penetration

resistance to powdery mildew fungi. At the same time, the SA,

JA, and ETsignaling pathways, as well as a pathway that poten-

tiates an HR-like defense to powdery mildew fungi, are

elevated in syp121 syp122 plants (Zhang et al., 2007). The

enhanced SA signaling is responsible for a significant part,

but not all, of the lesion-mimic phenotype. The powdery

mildew-triggered HR-like response appears to be independent

of the SA signaling pathway based on studies in which muta-

tions in SA signaling genes are introduced one by one. This re-

sponse, which may also contribute to the lesion-mimic

phenotype, occurs after attack by either Bgh or the virulent

powdery mildew fungus (Golovinomyces cichoracearum, Gc).

The response provides complete resistance to Gc (Zhang et

al., 2007). Other data implicating syntaxins in pathogen de-

fense include the demonstration that SYP121 and SYP122

are phosphorylated in response to flg22, an elicitor of basal

defenses (Nuhse et al., 2003; Benschop et al., 2007), and that

a closely related tobacco syntaxin is phosphorylated during

R-gene-mediated resistance (Heese et al., 2005). In addition,

tobacco SYP132 is involved in resistance to bacteria and secre-

tion of antimicrobial proteins (Kalde et al., 2007).

Our previous work has shown that silencing the SA pathway,

using mutations in genes for SA signal components, does not

fully rescue the necrosis and dwarfism in the syntaxin double

mutant. Here, we report a detailed survey of the defense sig-

naling in syp121 syp122 indicating that multiple defense path-

ways are active in this mutant. In addition, we show that

different pathways lead to the spontaneous cell death and

to the pathogen-triggered cell death reactions. Using global

transcript profiling, we show that many SA responses and

other defense-related transcripts are induced after mutation

of SYP121 and SYP122. We observe rescuing effects of muta-

tions in a number of well-described defense pathways, indicat-

ing that these are active in the syntaxin double mutant. We

also demonstrate that SID2 and EDS5, which are key elements

in SA biosynthesis, each have additional roles in defense sig-

naling. We provide evidence that the recently described

FMO1 and ALD1-defined pathways contribute to the pheno-

type independent of SA. However, we propose that these

pathways share the NPR1 signaling node. We present data sug-

gesting R-gene-mediated signaling contributes to the syntaxin

double mutant phenotypes and that the ET pathway may serve

to protect against necrosis development. These results empha-

size the importance of syntaxins in defense and establish the

syntaxin double mutant as a useful tool for studying defense

signaling networks.

RESULTS

Defense-Related Transcripts Are Up-Regulated as a Result

of Mutations in SYP121 and SYP122

A global transcript profiling experiment was performed to fur-

ther characterize the activated defense signaling in the syn-

taxin double mutant, syp121 syp122. Gene expression was

assayed in three biological replicates of wild-type Col-0,

syp121–1, syp122–1, and syp121–1 syp122–1 plants at

2.5 weeks of age, prior to the appearance of visible spontane-

ous lesions. Affymetrix ATH1 GeneChips, carrying 22 500 probe

sets, were used (Redman et al., 2004). Two-way ANOVA anal-

ysis identified 596 genes up-regulated (P < 0.001) in the syn-

taxin double mutant relative to wild-type, with 365 of these

being up-regulated more than two-fold. These 365 genes

and their fold change in the syp121 syp122 double mutant rel-

ative to wild-type are listed in Supplemental Table 1. One gene

(At3g30720) was also up-regulated in both of the syntaxin sin-

gle mutants, syp121–1 and syp122–1, relative to wild-type. This

gene, which encodes a unique 59 amino acid-protein of un-

known function, was up-regulated five-fold in syp121–1 and

13-fold in syp122–1. The At3g30720 transcript was the only

one to be significantly up-regulated in syp122–1. Remarkably,

this is also the only gene which was significantly constitutively

up-regulated in the pen3 mutant (Stein et al., 2006). The

syp121–1 mutation was associated with the up-regulation of

another six genes, one of which was SYP122. Interestingly,

only one gene (At4g30140) was found to be significantly

(P < 0.001) down-regulated (about three-fold) in syp121–1

syp122–1 compared to wild-type. This gene encodes a lipase/

hydrolase with a GDSL-motif.

Zhang et al. d syp121 syp122 Unravels Signaling Network | 511

We used the Meta-Analyzer tool from Genevestigator

(www.genevestigator.ethz.ch; Zimmermann et al., 2004) for

a more detailed analysis of the 365 genes up-regulated two-

fold or more. We calculated the mean of log2 of the change

in expression of these 365 genes for each of the 83 categories

in the Meta-Analyzer. For ten of the 83 categories, mean log2

values exceed 1, suggesting that the genes identified in our

experiment are also frequently up-regulated in other stress-

related experiments (Table 1). The gene expression profile

of ozone-treated plants is most similar to that of the syp121

syp122 mutant, with 328 of the 365 genes up-regulated

two-fold or more in response to ozone. This reactive oxygen

moleculehas beenreportedto induce anHR-likePCD(Overmyer

et al., 2000). The overlapping response suggests a high

similarity of the spontaneous PCD of syp121 syp122 and

ozone-induced PCD. Moreover, the ten matching categories

also include inoculation with Botrytis cinerea, Phytophthora

infestans, and Erysiphe cichoracearum (= Gc). The two

necrotrophic pathogens, B. cinerea and P. infestans, elicit

up-regulation of 241 and 273 of the 365 genes, respectively.

Even though the powdery mildew fungus, Gc, does not in-

duce cell death, it surprisingly elicits up-regulation of 285 of

the 365 genes. As expected, SA treatment was found among

the ten categories and this hormone up-regulates 232 of the

365 genes. Subsequently, we assessed the expression patterns

of the 365 genes in published Arabidopsis transcriptomics

data (Table 2). R-gene-mediated responses to Pseudomonas

syringae pv. tomato DC3000 (Pst) (Bartsch et al., 2006) of

the 365 genes were highly similar to those observed in

the syntaxin double mutant. When resistance is mediated

by RPM1 or RPS4, 223 and 200, respectively, of the 365 genes

are up-regulated. Of the 365 genes, 188 are up-regulated in

response to attack by virulent Pst (www.genevestigator.

ethz.ch). These observations suggest that at the 2.5-weeks

time-point, the syntaxin double mutant is preparing for

the initiation of PCD, which becomes visible a few days later.

Furthermore, the similarity to the response of plants inocu-

lated with avirulent Pst strains may suggest that R-gene

defenses are constitutively activated in the syntaxin double

mutant. This is supported below by the observation that

the R-gene signaling genes, PAD4, EDS1, NDR1, SGT1b,

and RAR1, are partially responsible for the spontaneous

lesion-mimic phenotype of syp121 syp122.

We also evaluated whether pathogen-associated molecular

pattern (PAMP)-induced defenses were activated in the syn-

taxin double mutant by assessing the expression of the 365

genes in plants following treatments with various PAMPs. Of

the 365 genes, 162 responded to chitin (www.genevestigator.

ethz.ch), 50 to Escherichia coli and 59 to Pst hrpA (Thilmony et

al., 2006) (see Table 2). The hrpA mutation of Pst eliminates the

secretion of effectors, which otherwise interfere with PAMP-

induced signaling. Even though these numbers are lower than

those for PCD-inducing stresses (see above), they suggest that

PAMP-like responses are also activated in the syp121 syp122

mutant.

In our previous analyses of the syntaxin double mutant, we

found that alleviating the SA signaling in syp121 syp122 by in-

troducing SA signaling mutations led to a significant increase

in PDF1.2 expression. This indicates that the JA signaling path-

way is activated in syp121 syp122, but suppressed by the antag-

onistic effect of high SA levels. Since optimal PDF1.2 expression

also requires ETsignaling (Penninckx et al., 1998), this pathway

is likely to be activated in the syntaxin double mutants as well.

The lack of expression of the JA signaling markers PDF1.2,

THI2.1 (Bohlmann et al., 1998), and VSP2 (Feys et al., 1994)

in the present transcriptomics analysis is consistent with the

antagonism between SA and JA signaling in the syp121

Table 1. Many of the 365 Genes, Up-Regulated Two-Fold or Moreas a Result of Mutation of SYP121 and SYP122, Are Also Up-Regulated in Response to Other Stresses.

Categorya Mean of log2b ‘Overlap’c Percent of 365

Ozone 2.3 328 89.9

Cyclohexamide 1.9 239 65.5

Botrytis cinerea 1.7 241 66.0

Phytophthora infestans 1.46 273 74.8

AgNO3 1.44 257 70.4

Syringolin 1.28 222 60.6

Erysiphe cichoracearum 1.26 285 78.1

NaCl 1.15 217 59.5

Nitrate low 1.07 247 67.7

SA 1.04 232 63.6

a Stress response categories selected by the Meta-Analyzer inGenevestigator (www.genevestigator.ethz.ch; Zimmermann etal., 2004).b Log2 of fold change of the 365 genes in response to theindicated stress. See text for additional information.c Number of the 365 genes that are up-regulated two-fold ormore in response to the indicated stress.

Table 2. Similarity of Selected Expression Profiles to the 365Genes, Up-Regulated Two-Fold or More as a Result of mutation ofSYP121 and SYP122.

Expression profile ‘Overlap’d

Pst AvrRPM1a 223

Pst AvrRPS4a 200

Pstb 188

Chitinb 162

ETb 101

MeJAb 87

Pst hrpAc 59

E. coli c 50

a Bartsch et al. (2006).b Genevestigator (www.genevestigator.ethz.ch).c Thilmony et al. (2006).d Number of the 365 genes that are up-regulated two-fold ormore in response to the indicated stress.

512 | Zhang et al. d syp121 syp122 Unravels Signaling Network

syp122 mutant, as noted previously (Zhang et al., 2007). Two

marker genes for ET signaling–HEL (Potter et al., 1993) and

CHI-B (Norman-Setterblad et al., 2000)–are up-regulated

6.1 and 2.2-fold, respectively. Expression of CHI-B is thought

to depend on both JA and ET signaling, which points to the

complexity of the interplay among the SA, JA, and ETsignaling

pathways in syp121 syp122. No enhanced expression of genes

involved in ET biosynthesis is observed in syp121 syp122, even

though these genes were previously reported to be induced by

ET (van der Straeten et al., 1992; Zhong and Burns, 2003). How-

ever, 101 and 87 of the genes up-regulated after treatment

with ET and MeJA, respectively (www.genevestigator.ethz.ch),

were found among the 365 genes (Table 2; Supplemental Table

1). This observation may suggest that JA and ET signaling are

active, perhaps at a low level, in syp121 syp122, despite high SA

levels.

Screen for Mutations Rescuing syp121 syp122

In order to search for novel factors that contribute to the lesion

formation of syp121 syp122, a forward genetics approach

was taken. Approximately 100 000 mutagenized syp121–1

syp122–1M2 plants were screened for individuals rescued from

the lesion-mimic phenotype. Approximately 240 candidate

mutants were retained. Improved growth and reduced necro-

sis of all the selected mutant lines were confirmed in the M3

generation. We suggest that these 240 lines carry mutant

alleles of genes required for lesion formation. These alleles

we denote suppressors of syntaxin-related death (ssd).

We previously demonstrated that knockout mutations in

EDS5, SID2, and NPR1 only partially suppressed the lesion-

mimic phenotype of the syp121 syp122 double mutant (Zhang

et al., 2007). The triple mutant, syp121–1 syp122–1 sid2-1, had

a characteristic yellow-green color, and complementation

tests with syp121–1 syp122–1 ssd mutants with a similar phe-

notype led to the identification of 15 new sid2 mutants. Also,

syp121–1 syp122–1 npr1-1 had a distinct bleaching phenotype

(see below), but none of the novel suppressor mutants resem-

bled this triple mutant in appearance. Furthermore, crossing

syp121–1 syp122–1 npr1-1 to 70 randomly selected suppressor

mutants did not reveal any npr1 mutants among the ssd

mutants. Unlike the triple mutants with sid2 and npr1,

syp121–1 syp122–1 eds5-3 mutants did not have an immedi-

ately recognizable phenotype. Crosses of syp121–1 syp122–1

eds5-3 to 16 random suppressor mutants revealed two novel

eds5 mutations. We subsequently performed inter-mutant

crosses among a subset of approximately 100 of the suppressor

mutants to place them in complementation groups. Three

groups were established consisting of ten, seven and seven

alleles, respectively. These were named ssd1-1 to ssd1-10,

ssd2-1 to ssd2-7 and ssd3-1 to ssd3-7. Representative triple

mutants of these three complementation groups are shown

in Figure 1A and in Supplemental Figure 1. The mutant alleles

of these three SSDgenes are stronger suppressors of the lesion-

mimic phenotype of syp121–1 syp122–1 than sid2-1, eds5-3,

and npr1-1 (Figure 1A; Zhang et al., 2007).

FMO1, ALD1, and PAD4 Contribute to the Lesion-Mimic

Phenotype of syp121 syp122

Initial mapping placed SSD1 on the top of chromosome 1 be-

tween markers F21M12 and ciw12. Subsequent fine-mapping,

Figure 1. SSD1 Encodes FMO1, SSD2 Encodes ALD1 and SSD3 En-codes PAD4, which All Contribute to Development of the Lesion-Mimic Phenotype of syp121 syp122.

(A) Comparison of the syp121–1 syp122–1 phenotype after SSD1,SSD2, and SSD3 knockout and after knockout of the SA-pathway.Plants grown for 6 weeks.(B) ssd1 alleles are non-functional mutations of FMO1.(C) ssd2 alleles are non-functional mutations of ALD1.(D) ssd3 alleles are non-functional mutations of PAD4.Asterisks indicate stop codons. (See more details in SupplementalFigures 1, 2, 3, and 4.)

Zhang et al. d syp121 syp122 Unravels Signaling Network | 513

using an additional 520 F2 plants, localized SSD1 to an interval

of seven genes (At1g19230 to At1g19310). One gene in this in-

terval, At1g19250, was highly up-regulated in the syp121

syp122 double mutant (22-fold; Supplemental Table 1).

At1g19250 encodes the FLAVIN-DEPENDENT MONO-OXYGEN-

ASE1 (FMO1) that recently has been implicated in defense

(Mishina and Zeier, 2006; Koch et al., 2006; Bartsch et al.,

2006; Olszak et al., 2006). We PCR amplified and sequenced

the coding region of FMO1 from the genomic DNA of the

ten ssd1 suppressor mutants. In each of them, either a prema-

ture stop codon, a base-change resulting in an amino acid sub-

stitution or a splice site mutation was identified (Figure 1B;

Supplemental Figure 2). It is discussed whether the first 45

nucleotides of exon 4 belong to the adjacent intron. Support-

ing Bartsch et al. (2006), the mutation in fmo1-10 is five base

pairs away from the potential splice site at base pair 1548

in exon 4 (Supplemental Figure 2). This is too far away to

affect splicing. This documents that the mutation in fmo1-

10 must be a premature stop-codon in exon 4 and therefore

the longer splice-version contributes to the lesion formation

and dwarfing of syp121 syp122. Subsequently, the T-DNA inser-

tion mutant allele fmo1-1 (Bartsch et al., 2006) in line

SALK_026163 was introduced into syp121–1 syp122–1. The

phenotype of the resulting triple mutant was indistinguish-

able from that of syp121 syp122 ssd1 mutants (data not

shown).

SSD2 was initially mapped to the top of chromosome 2 be-

tween markers nga1145 and ciw3 and then fine-mapped in

a population of 230 F2 plants to a 25-gene interval from

At2g13590 to At2g13850. One of these 25 genes

(At2g13810) was up-regulated in the syp121 syp122 mutant

(11-fold; Supplemental Table 1). At2g13810 encodes the

AGD2-LIKE DEFENSE RESPONSE PROTEIN1 (ALD1), which has

been shown to contribute to defense (Song et al., 2004a,

2004b). The coding region of ALD1 was PCR amplified and se-

quenced using genomic DNA from the seven syp121 syp122

ssd2 triple mutants. In all of them, alterations resulting in mo-

dification of the ALD1 protein were encountered (Figure 1C;

Supplemental Figure 3). Subsequently, the T-DNA insertion

mutant allele ald1-T2 (Song et al., 2004a) in line SALK_007673

was introduced into syp121–1 syp122–1. The phenotype of the

resulting triple mutant was indistinguishable from that of

syp121 syp122 ssd2 mutants (data not shown). An alignment

of the Arabidopsis ALD1 amino acid sequence to the most si-

milar protein for which a 3-D structure is available (i.e. aromatic

aminotransferase from Pyrococcus horikoshii Ot3 gi|14278622)

revealedthat the five amino acid substitutions found in the ssd2

mutant proteins occur in structurally important and well-

conservedaminoacids.Meanwhile, ssd2-3has aprematurestop

codon, and ssd2-4 a splice site mutation.

A map-position was not found for SSD3 on chromosomes 1,

2, 4, or 5, suggesting that it is located on chromosome 3, and

thus could not be mapped using the initial Ler syp121 syp122

line we created. To determine whether SSD3 was located on

chromosome 3 near either SYP121 or SYP122, crosses were

made between syp121–1 syp122–1 ssd3-7 and syp121–1 and

between syp121–1 syp122–1 ssd3-7 and syp122–1. While

21.8% of the F2’s from the cross to syp122–1 had the syp121

syp122 lesion-mimic phenotype (not significantly different

from a 3/16 segregation ratio), no plants with this phenotype

were recorded in the cross to syp121–1, demonstrating ssd3 is

closely linked to syp122. Subsequently, approximately 10 000

F2’s of a cross between syp121–1 syp122–1 ssd3-7 and Col-0

wild-type were analyzed. Only two F2 plants had the

syp121 syp122 phenotype. This indicated exceptionally close

linkage between SYP122 and SSD3. Near SYP122, only one

gene (At3g52430) is up-regulated in syp121 syp122 (10-fold;

Supplemental Table 1). At3g52430, which is three genes distal

from SYP122 (At3g52400), encodes PHYTOALEXIN DEFICIENT 4

(PAD4). Based on PAD4’s close functional association with EDS1

(Wiermer et al., 2005) and the ability of an eds1 mutation to

partially suppress the lesion-mimic phenotype of syp121

syp122 (Zhang et al., 2007; Figure 1A), this result was not un-

expected. We sequenced the coding region of PAD4 from ge-

nomic DNA of the seven syp121 syp122 ssd3 triple mutants.

In six of them, base changes resulting in premature stop codons

and, in one of them, a base change resulting in an amino

acid substitution were encountered (Figure 1D; Supplemental

Figure 4).

We conclude that FMO1, ALD1, and PAD4 are constitutively

active in syp121 syp122 syntaxin double mutants and that

these signaling components, and their downstream elements,

contribute to lesion formation and dwarfing. In Figure 1, the

ssd allele names are converted to fmo1, ald1, and pad4 allele

names.

SID2 and EDS5 Each Confer Activities in Addition to

SA Signaling

Numerous signaling genes are essential as positive regulators

for pathogen defense in Arabidopsis. Some of these are listed

in Supplemental Table 2. To assay which of these genes are in-

volved in development of lesions and dwarfism, mutant alleles

were crossed into syp121 syp122 (Table 3; Supplemental Figure

5). Most of these defense regulators clearly influence the de-

velopment of the lesion-mimic phenotype of syp121 syp122.

Thus, by combining the different mutations in the syntaxin

double mutant background, it is possible to study whether

the defense pathways, defined by these, genes interact.

Both SID2 and EDS5 affect SA accumulation and null

mutants contain very low SA levels after pathogen attack

(Heck et al., 2003). EDS5 encodes a MATE-type transporter re-

quired for SA signaling (Nawrath et al., 2002), and SID2 enco-

des an isochorismate synthase (Wildermuth et al., 2001).

Inactivation of either SID2 or EDS5 rescued the syp121

syp122 lesion-mimic phenotype to approximately the same

level and eliminated the constitutive SA level (Figure 2A)

and PR-1 gene expression (Zhang et al., 2007). However, the

two different triple mutant lines were easily differentiated vi-

sually. syp121–1 syp122–1 eds5-3 was relatively more necrotic

than syp121–1 syp122–1 sid2-1, which was light green with

514 | Zhang et al. d syp121 syp122 Unravels Signaling Network

Table 3. Visual Quantification of Phenotypes of syp121 syp122 Rescued by Mutation in One, Two, or Three Positive Defense RegulatoryGenes.

Size/healtha

SA- HR-like Colonies of Comments3 weeks 6 weeks Response resp. to Gc Gc

Col-0 ++++++ ++++++ – – +++++

Ler ++++++ ++++++ – – +++++

syp121–1 ++++++ ++++++ – – +++++

syp122–1 ++++++ ++++++ – – +++++

syp121–1 syp122–1 ++ + +++++ ++++++ No

eds1-2 ++++++ ++++++ – – ++++++

eds5-3 ++++++ ++++++ – – ++++++

sid2-1 ++++++ ++++++ – – ++++++

NahG ++++++ ++++++ – – ++++++

npr1-1 ++++++ ++++++ – – ++++++

jar1-1 ++++++ ++++++ – – ++++++

ein2-1 ++++++ ++++++ – – +++++

ndr1-1 ++++++ ++++++ – – ++++++

sgt1b-1 ++++++ ++++++ – – ++++

rar1-10 ++++++ ++++++ – – ++

atg7 ++++++ ++++++ – – +++++

fmo1-1 ++++++ ++++++ – – ++++++

npr1-1 ndr1-1 ++++++ ++++++ – – ++++++

rbohD ++++++ ++++++ – – +++++

syp121-NahG ++++++ ++++++ – – +++++

syp121–1 syp122–1 eds1-2 ++++++ ++++½ +++++ +++ +

syp121–1 syp122–1 eds5-3 +++++ +++ +++++ +++ No

syp121–1 syp122–1 sid2-1 +++++ +++ +++++ +++ No

syp121–1 syp122–1 NahG ++++++ ++++½ +++++ ++++ No

syp121–1 syp122–1 npr1-1 +++++ +++ – +++ No Bleaching

syp121–1 syp122–1 jar1-1 ++ + ++++++ ++++++ No

syp121–1 syp122–1 ein2-1 + ½ ++++++ ++++++ No

syp121–1 syp122–1 rbohD ++ + ++++++ ++++++ No

syp121–1 syp122–1 atg7 ++ + ++++++ ++++++ No

syp121–1 syp122–1 rar1-10 +++++ ++ ++++ ++++ No

syp121–1 syp122–1 sgt1b-1 +++++ +½ ++++ ++++ No

syp121–1 syp122–1 ndr1-1 +++++ +½ ++++++ ++++ No

syp121–1 syp122–1 fmo1-nb ++++++ ++++ ++++ ++++ No

syp121–1 syp122–1 pad4-nb ++++++ ++++ ++ ++ ++

syp121–1 syp122–1 ald1-nb ++++++ ++++ ++ +++ No

syp121–1 syp122–1 eds5-3 sid2-1 ++++++ ++++½ ++++++ +++ No

syp121–1 syp122–1 eds5-3 npr1-1 ++++++ ++++½ – ++ No No bleach

syp121–1 syp122–1 npr1-1 sid2-1 ++++++ ++++½ – ++ No No bleach

syp121–1 syp122–1 npr1-1 ndr1-1 ++++++ ++++½ – ++ No Bleaching

syp121–1 syp122–1 npr1-1 ein2-1 ++++++ ++ (dead) – ++ No Bleaching

syp121–1 syp122–1 sid2-1 ein2-1 ++++++ +++ ++++++ ++ No

syp121–1 syp122–1 npr1-1 jar1-1 ++++++ ++ – +++ No Bleaching

syp121–1 syp122–1 npr1-1 atg7 ++++++ +++ – +++ No Bleaching

syp121–1 syp122–1 sid2-1 atg7 ++++++ +++ ++++++ ++++ No

syp121–1 syp122–1 sid2-1 jar1-1 ++++++ +++ +++++ ++++ No

syp121–1 syp122–1 sid2-1 npr1-1 jar1-1 ++++++ +++++ – +++ No No bleach

syp121–1 syp122–1 sid2-1 eds5-3 jar1-1 ++++++ +++++ +++++ ++++ No

syp121–1 syp122–1 sid2-1 npr1-1 ndr1-1 ++++++ +++++½ – ++ No No bleach

Zhang et al. d syp121 syp122 Unravels Signaling Network | 515

distinct necrotic lesions surrounded by chloroses. Interestingly,

the quadruple mutant, syp121–1 syp122–1 eds5-3 sid2-1, per-

formed markedly better than either of the two triple mutants

(Figures 1A and 2B; Table 3; Supplemental Figure 5). These

observations suggest that EDS5 and SID2 have additional,

unique roles, aside from SA biosynthesis, and that all these dis-

tinct roles contribute additively to lesion development and

dwarfism in syp121 syp122.

SA-Dependent, NPR1-Independent Signaling

The triple mutants syp121–1 syp122–1 eds5-3 and syp121–1

syp122–1 sid2-1 contained very little or no SA. In contrast,

syp121–1 syp122–1 npr1-1 plants contained SA amounts ex-

ceeding those of syp121 syp122 (Figure 2A). The high SA levels

are apparently caused by the loss of the two syntaxins, which

leads to continuous SA production. In the absence of func-

tional NPR1, SA perception may be compromised and there-

fore the effects of high SA levels are diminished. However,

the triple mutant syp121–1 syp122–1 npr1-1 did develop char-

acteristic bleaching of leaf and stem tissues, which remained

alive for some additional time. To test the possibility that

the bleaching is a phenomenon restricted to the npr1-1 allele,

theNPR1mutant alleles npr1-2, npr1-3 (Cao et al., 1997),npr1-5

(Shah et al., 1997), and npr1-P4-7 (Xiao et al., 2005) were

crossed into syp121–1 syp122–1. The resulting triple mutant

lines all show the same bleaching phenotype (data not shown).

Interestingly, we also observed bleaching when npr1 single

mutants were sprayed with SA (data not shown). The bleach-

ing phenotype thus appears to be associated with high SA lev-

els. Introducing mutant alleles of SID2 and EDS5 into syp121

syp122 npr1, thereby abolishing SA biosynthesis, generated

quadruple mutants with improved performance (Figure 2B;

Table 3; Supplemental Figure 5). Importantly, these syp121

syp122 npr1 eds5 and syp121 syp122 npr1 sid2 quadruple

mutants do not have the bleaching phenotype of syp121

syp122 npr1. These observations suggest SA-dependent,

NPR1-independent processes in Arabidopsis.

ALD1 and FMO1-Mediated Signals, SA and NPR1

In order to determine how defense pathways controlled by

ALD1 and FMO1 interact with SA signaling, a number of com-

binations were made between alleles of these genes and of

EDS5, SID2, NPR1, and PAD4, all in the syp121 syp122 mutant

background. The partially rescued syp121 syp122 fmo1 and

syp121–1 syp122–1 ald1 performed even better when SA bio-

synthesis or signaling also was inactivated (i.e. in the quadru-

ple mutants including mutations in EDS5, SID2, or PAD4) (Table

3). These observations were made in F2 populations segregat-

ing for mutant alleles of the following six gene-pairs: FMO1/

ESD5, FMO1/SID2, FMO1/PAD4, ALD1/SID2, ALD1/ESD5, and

Table 3. (Continued)

Size/healtha

SA- HR-like Colonies of Comments3 weeks 6 weeks Response resp. to Gc Gc

syp121–1 syp122–1 pad4-19 eds5-3 ++++++ +++++c ++ + ++++

syp121–1 syp122–1 pad4-19 npr1-1 ++++++ +++++d – + ++++ No bleach

syp121–1 syp122–1 pad4-19 sid2-1 ++++++ ++++++e ++ + ++++

syp121–1 syp122–1 pad4-19 fmo1-10 +++++f

syp121–1 syp122–1 pad4-n ald1-n +++++g

syp121–1 syp122–1 sid2-1 fmo1-n ++++++h

syp121–1 syp122–1 sid2-1 ald1-n +++++i

syp121–1 syp122–1 npr1-1 fmo1-n ++++ j

syp121–1 syp122–1 npr1-1 ald1-n ++++k

syp121–1 syp122–1 eds5-3 fmo1-n ++++++l

syp121–1 syp122–1 eds5-3 ald1-5 +++++m

a Higher number of plusses denotes larger and more healthy plants. ‘½’ denotes half a plus.b n: more than one mutant allele tested.c Result confirmed in an F2 population segregating for eds5-3 and pad4-14.d Result confirmed in F2 populations segregating for npr1-1 and four pad4 alleles.e Result confirmed in F2 populations segregating for sid2-1 and five pad4 alleles.f Result confirmed in F2 populations segregating for combinations of four pad4 and nine fmo1 alleles.g Result obtained in F2 populations segregating for combinations of three pad4 and five ald1 alleles.h Result obtained in F2 populations segregating for combinations of sid2-1 and a novel sid2 allele, and eight fmo1 alleles.i Result obtained in F2 populations segregating for sid2-1 and four ald1 alleles.j Result obtained in F2 populations segregating for npr1-1 and three fmo1 alleles.k Result obtained in F2 populations segregating for npr1-1 and three ald1 alleles.l Result obtained in F2 populations segregating for eds5-3 and three fmo1 alleles.m Result confirmed in segregating F2 population.

516 | Zhang et al. d syp121 syp122 Unravels Signaling Network

ALD1/PAD4. Approximately 1/16 of the plants became remark-

ably larger, as exemplified in Figure 3, suggesting that ALD1

and FMO1 mediate signals that are distantly associated to

SA signaling. However, it is interesting that no notably larger

plants were found in F2 populations of crosses between syp121

syp122npr1 and syp121 syp122 fmo1or between syp121 syp122

npr1 and syp121 syp122 ald1, as exemplified in Figure 3. All

these phenotypes have been recorded using several alleles,

as indicated in Table 3. This suggests that FMO1 and ALD1

contribute to the lesion-mimic phenotype of syp121 syp122

by mechanisms that are partially parallel to SA signaling and

to PAD4-dependent signaling. Interestingly, the ALD1 and

FMO1 signals also appear to act separately from the SA-

independent signals suggested above to be conferred by

EDS5 and SID2. However, ALD1 and FMO1 seem to mediate

signals that depend on NPR1. All in all, these observations

suggest the existence of SA-independent, NPR1-dependent

signaling.

R-Gene-Mediated Signaling Pathways in syp121 syp122

In general, R-gene-dependent resistance mediated by toll-

interleukin-1receptor(TIR)-typenucleotidebinding-leucine-rich

repeat (NB-LRR) proteins depends on EDS1 and PAD4, whereas

resistance mediated by coiled-coil (CC)-type NB-LRR proteins is

NDR1-dependent (Hammond-Kosack and Parker, 2003; Wiermer

et al., 2005). The proteins RAR1 and SGT1b are important for

Figure 2. Quadruple Mutants Demonstrate SA-Independent Sig-naling Activities of SID2 and EDS5.

(A) SA content in shoots of 4-week-old plants (mean 6 SD, n = 3).(B) Plants grown for 6 weeks. (Subset of genotypes from Supple-mental Figure 5.)

Figure 3. syp121 syp122Has ALD1 and FMO1-Mediated Signals thatAre Separate from SA-Mediated Signals, but Requires NPR1.

In segregating F2 populations of approximately 100 individuals inthe syp121–1 syp122–1 genetic background, approximately 1/16 ofthe plants are performing notably better only in the sid2-13fmo1-10 and sid2-13ald1-6 crosses. Plants grown for 6 weeks. (Theseobservations are confirmed using several allele combinations, seeTable 3.)

Zhang et al. d syp121 syp122 Unravels Signaling Network | 517

the stability of the NB-LRR proteins and are thus required for

R-gene-mediated resistance in general (Holt et al., 2005).

Given the contribution of PAD4 and EDS1 to the syp121

syp122 phenotype (Figure 1; Table 3; Supplemental Figure 5;

Zhang et al., 2007), we sought to analyze whether R-gene-

mediated signals were functioning in the syp121 syp122

mutant and contributing to its lesion-mimic phenotype. Since

EDS1 and PAD4 act upstream of the SA pathway (Hammond-

Kosack and Parker, 2003; Wiermer et al., 2005), which clearly

contributes to the syp121 syp122 lesion-mimic phenotype, the

question remains whether the SA pathway activation in syp121

syp122 is the only signal dependent on EDS1/PAD4 function.

Improved performance of the syntaxin double mutant in an

eds1 or pad4 background compared to a sid2, eds5, or npr1

background indicates that these latter three SA signaling com-

ponents control only a subset of the pathways triggered by

EDS1 and PAD4 (Figure 1). Remarkably, inactivation of SID2,

EDS5, or NPR1 in the syp121–1 syp122–1 pad4-19 genetic back-

ground led to further improved performance of the resulting

quadruple mutant relative to the triple mutant (Figure 4A;

Table 3; Supplemental Figure 5). This result suggests either that

the SA signaling is independent of PAD4 or, more likely, that

SID2, EDS5, and NPR1 also control SA- and PAD4-independent

signals.

Further support for a role of R-gene-mediated signaling in

the lesion-mimic phenotype of syp121 syp122 was provided by

the partial suppression of this mutant phenotype by mutations

in NDR1, RAR1, or SGT1b (Figure 4B; Table 3). These mutations

rescued to approximately the same degree as mutations in SA

signaling genes until the plants reached the 3-week stage. In

older plants, mutations of NDR1, RAR1, or SGT1b resulted in

only slightly enhanced performance compared to syp121

syp122 (Figure 4C; Table 3; Supplemental Figure 5). To analyze

whether NDR1-mediated signaling interacts with the SA path-

way, we introduced the ndr1-1 allele into syp121–1 syp122–1

npr1-1 and syp121–1 syp122–1 sid2-1. The resulting quadruple

mutants performed significantly better than either of the tri-

ple mutants, suggesting that NDR1 activates SA-independent

defense processes (Figure 4C; Table 3). Interestingly, the SA-

related bleaching in syp121–1 syp122–1 npr1-1was maintained

in syp121–1 syp122–1 npr1-1 ndr1-1 in contrast to syp121–1

syp122–1 npr1-1 sid2-1, where no bleaching is observed. This

indicates that NDR1-mediated signaling does not influence SA

accumulation. Finally, the quintuple mutant syp121–1 syp122–1

npr1-1 sid2-1 ndr1-1 was obtained. Even though older plants of

this genotype became necrotic, the quintuple mutant grew

almost as well as wild-type Col-0.

In conclusion, these results suggest that loss of SYP121 and

SYP122 leads to the activation of similar signaling pathways as

CC-NB-LRR and TIR-NB-LRR-type R-gene-mediated pathogen

recognition. This observation is corroborated by the similari-

ties between transcript profiles of syp121 syp122 and plants

in which R-gene signaling was triggered (Table 2; Supplemen-

tal Table 1).

Figure 4. R-Gene-Mediated Type of Signals Are Active in the Syn-taxin Double Mutant.

(A) syp121 syp122 has PAD4-mediated signals that are separatefrom EDS5, SID2, and NPR1-mediated signals (6-week-old plants).(B) syp121 syp122 has NDR1, SGT1b, and RAR1-mediated signals(3-week-old plants).(C) NDR1 signal is independent of SA (6-week-old plants).A and C are subsets of genotypes from Supplemental Figure 5.

518 | Zhang et al. d syp121 syp122 Unravels Signaling Network

The autophagy component of PCD has been implicated in

R-gene-mediated resistance (Lui et al., 2005). A number of

autophagy genes (ATGs) have been identified in a variety of

organisms. In Arabidopsis, ATG7 is a non-redundant gene

(Doelling et al., 2002) and the T-DNA atg7 allele in

SALK_057605 can be used to test the role of autophagy in

the lesion-mimic phenotype of syp121 syp122. Blocking

autophagy by this method did not cause any visible phenotypic

effects (Table 3; Supplemental Figure 5). This result suggests

that autophagy does not play a role in the lesion-mimic phe-

notype, the HR-like response to the powdery mildew fungi or

the SA-hypersensitivity (see below) phenotypes of the syntaxin

double mutant. We also examined the quadruple mutants

syp121–1 syp122–1 npr1-1 atg7 and syp121–1 syp122–1 sid2-1

atg7, and were again unable to demonstrate any effect of

the atg7 mutation. These observations are supported by intro-

ductions of mutant alleles of the non-redundant genes ATG3

(SALK_031693) and ATG6 (SALK_051168; Fujiki et al., 2007) in-

to syp121–1 syp122–1. In both cases, all plants in syntaxin dou-

ble mutant homozygous families that were segregating for

the atg3 or atg6 mutations exhibited the typical syp121–1

syp122–1 phenotype (data not shown).

An oxidative burst, generated by NADPH oxidases, is an in-

dispensable element of pathogen defense mechanisms in

plants (Foyer and Noctor, 2005). In Arabidopsis, AtrbohD is

the NADPH oxidase primarily responsible for this burst (Torres

et al., 2002). We introduced the T-DNA atrbohD allele in

SALK_070610 into the syp121 syp122 background. This muta-

tion seemed to have no effect on the lesion-mimic phenotype,

the HR-like response to the powdery mildew fungi or the SA-

hypersensitivity (see below) of the syntaxin double mutant

(Table 3; Supplemental Figure 5). However, both the atrbohD

and syp121–1 syp122–1 atrhohD plants were hyper-sensitive to

high light, which confirms the presence of the atrbohD

mutant allele (data not shown).

Ethylene Signaling Antagonizes syp121 syp122 Lesion

Formation

We previously suggested that the ET and JA signaling path-

ways are activated in syp121 syp122 (Zhang et al., 2007). This

was documented by the increased expression of the PDF1.2

transcript, a marker for these two pathways (Penninckx et

al., 1998), following interruption of the SA signaling pathway

in the syp121 syp122 double mutant. To analyze the role of the

ET signaling pathway in the double syntaxin mutant pheno-

types, the mutant allele ein2-1 was crossed into syp121–1

syp122–1. The biochemical function of EIN2 is not known,

but the protein is required for general ETsignaling (Benavente

and Alonso, 2006). EIN2 appeared to suppress necrosis in the

syp121–1 syp122–1 double mutant, as suggested by the en-

hanced necrotic phenotype of syp121–1 syp122–1 ein2-1 triple

mutant (Figure 5; Table 3; Supplemental Figure 5). This ein2-1

effect was even more obvious when NPR1 was also mutated,

but not when SID2 was mutated (Table 3; Figure 5). The

quadruple mutant syp121–1 syp122–1 npr1-1 ein2-1 died after

approximately 6 weeks.

JA Signaling Impacts Lesion Formation in syp121 syp122

Plants

The jar1-1 mutant allele was used to assess the importance of

JA signaling for the syp121 syp122 phenotypes. JAR1 catalyses

conjugation of JA to amino acids (Staswick and Tiryaki, 2004).

When introduced into syp121–1 syp122–1, no phenotypic con-

sequences were observed (Figure 5; Table 3; Supplemental Fig-

ure 5). This lack of effect was confirmed by using a coi1 mutant

(Xie et al., 1998) and a mutant (SALK_017756) in the AOS gene

(Park et al., 2002). None of the plants in an F2 population of

syp121–1 syp122–1, segregating for the coi1-1 or the aos allele,

deviated from the necrotic and dwarfed appearance of the

syntaxin double mutant. During the flowering of these plants,

expected pollen-sterile individuals were detected, confirming

that these lines were homozygous for coi1-1 or aos (data not

shown).

The absence of a detectable role for JA signaling could be

due to the JA pathway being suppressed by the high SA level

found in the syntaxin double mutant (Spoel et al., 2003; Zhang

et al., 2007). To test this hypothesis, mutant combinations

blocking both SA and JA signaling or biosynthesis were crossed

into the syp121 syp122 background. However, this approach

gave what appear to be conflicting results (Figure 5; Table 3;

Figure 5. Ethylene Signal Has Anti-Necrotic Effect on syp121syp122.

Phenotypes of syp121–1 syp122–1 after introduction of ethyleneand jasmonic acid signaling mutations (6-week-old plants). (Subsetof genotypes from Supplemental Figure 5.)

Zhang et al. d syp121 syp122 Unravels Signaling Network | 519

Supplemental Figure 5). syp121–1 syp122–1 npr1-1 jar1-1

showed more necrosis than syp121–1 syp122–1 npr1-1 plants,

which is consistent with the suggestion by Rao et al. (2000) that

JA signaling protects against cell death. However, syp121–1

syp122–1 sid2-1 jar1-1 plants were indistinguishable from

syp121–1 syp122–1 sid2-1 plants. Surprisingly, the jar1-1 muta-

tion resulted in improved performance of the quintuple

mutants syp121–1 syp122–1 sid2-1 npr1-1 jar1-1 and syp121–1

syp122–1 sid2-1 eds5-3 jar1-1.

Gc-Induced HR-Like PCD

syp121 syp122 has a multi-cellular HR-like response to the two

powdery mildew fungi Bgh and Gc. Introducing mutant alleles

of gene in the SA pathway into the syntaxin double mutant

demonstrated this HR-like response to be SA-independent

(Zhang et al., 2007). The HR-like response confers complete re-

sistance to Gc. Combining any two of the three SA pathway

mutations, eds5-3, sid2-1, and npr1-1, in the syntaxin double

mutant did not break this PCD-associated resistance (Figure 6).

In order to determine which defense components contrib-

ute to this powdery mildew-induced HR-like response, a series

of triple mutants were examined macroscopically. Neither the

HR-like response nor the resistance to Gc was significantly af-

fected by inactivating the NDR1, RAR1, or SGT1b genes impli-

cated in R-gene signaling (Table 3). However, mutations in

EDS1, and, to a greater extent, in PAD4, partially restricted

the HR-like response and disease resistance. This allowed weak

colony growth occasionally to be observed on these triple

mutants (Table 3). Interestingly, introducing each of the three

SA pathway mutations, eds5-3, sid2-1, and npr1-1, into

syp121–1 syp122–1 pad4-19 had an obvious negative effect

on the HR-like response, and the quadruple mutants were sus-

ceptible to Gc during the first week after inoculation (Table 3;

Figure 6). However, at later time-points, necrosis was observed,

which arrested fungal growth (data not shown). We have no

evidence to suggest that the following genes are required for

the HR-like response and resistance to Gc resistance of the

syp121 syp122 double mutant: JAR1, EIN2, RbohD, ATG7,

FMO1, and ALD1. Furthermore, when mutations of some of

these genes were introduced into triple mutants blocked in

SA synthesis or signaling (syp121–1 syp122–1 npr1-1, syp121–1

syp122–1 eds5-3, and syp121–1 syp122–1 sid2-1), the HR-like

response was not sufficiently suppressed to permit Gc

growth (Table 3).

syp121 syp122 Mutants are Hypersensitive to SA

Wild-type plants respond to exogenous SA by induction of, for

instance, the PR-1 gene (Durrant and Dong, 2004). However,

even though SA is involved in responses leading to PCD,

wild-type plants do not necessarily develop PCD, even when

treated with high concentrations of SA. This suggests activation

of parallel signaling is required for PCD, and, as such, SA is

considered a potentiator rather than a trigger for PCD in

R-gene-mediated HR (Shirasu et al., 1997; Tenhaken and Rubel

1997).

In contrast to wild-type plants, very young syp121 syp122

mutants underwent PCD in response to exogenous SA (Fig-

ure 7). This suggests that the syntaxin double mutant is hyper-

sensitive to SA. To examine whether this SA-hypersensitivity

was due in part to the high endogenous SA level in the syn-

taxin double mutant (Figure 2A; Zhang et al., 2007), mutations

of the SID2 or EDS5 genes or the NahG gene were introduced

into the syp121–1 syp122–1 background. The SA-hypersensitiv-

ity of the syntaxin double mutant was still observed in mutants

unable to synthesize SA. However, the SA-hypersensitivity did

require an intact NPR1 protein (Figure 7). The independence of

endogenous SA and the uniqueness of this phenotype in

syp121–1 syp122–1 were supported by an analysis of other

dwarfed and/or necrotic mutants constitutively expressing

the SA signaling pathway. None of the tested lines, acd2-2,

acd11-1, cpr1-1, lsd1-1, or mpk4, all of which can be rescued

by interruption of the SA signaling pathway, exhibits such

an SA-hypersensitivity (Figure 7; Table 3).

Having shown that a number of signaling pathways, other

than the SA pathway, are active in syp121 syp122, we analyzed

whether these other defense pathways contributed to the

Figure 6. Simultaneous Mutations in PAD4 and SA-Signaling GenesDelay Powdery Mildew-Induced Multicellular HR-like Response insyp121 syp122.

Development of Gc colonies 6 d after inoculation. syp121–1syp122–1 is left out, since its severe necrosis does not allow anyGc-induced phenotype to be observed. See also Zhang et al.(2007). (Plants 3 weeks old at time of inoculation.)

520 | Zhang et al. d syp121 syp122 Unravels Signaling Network

SA-hypersensitive phenotype. Among the pathways tested,

none could be assigned to this phenotype. Furthermore, path-

ways that did not contribute to the lesion phenotype in syp121

syp122 were also not involved in the SA-hypersensitivity

(Table 3).

DISCUSSION

Activation of the different inducible defense mechanisms

against microbial pathogens can be recognized by transcrip-

tomics. Here, such analysis showed that inactivation of both

SYP121 and SYP122 led to activation of 365 genes by more

than two-fold prior to the visible appearance of cell death.

Most of these genes are also induced by ozone, by a necrotro-

phic fungus and by a cell death-inducing oomycete. Therefore,

the signals required for PCD are already present before the ap-

pearance of necroses. A strong overlap with genes induced by

HR-inducing avirulent bacteria points to the operation of R-

gene-mediated defense pathways in syp121 syp122. Previous

evidence for a role of SA signaling in the phenotype develop-

ment of this line is corroborated by the observation that more

than half of the 365 genes in syp121 syp122 also respond to SA.

However, it was surprising to discover that a large majority of

these genes, otherwise related to induction of cell death, are

also induced by a virulent isolate of the powdery mildew fun-

gus Gc, which does not elicit host cell death. A possible expla-

nation is that this fungal pathogen primes but does not fully

activate PCD, or that Gc releases efficient anti-cell death effec-

tors to the plant cell. A significant number of the 365 genes are

PAMP-induced, suggesting that inactivation of SYP121 and

SYP122 also activates PAMP-type signaling. Interestingly, only

one gene is down-regulated after mutating the syntaxin genes.

This is in contrast to the transcript analysis of the acd11 lesion-

mimic mutant, where a large set of genes involved in primary

metabolism is down-regulated (Brodersen et al., 2002). Inocula-

tions with pathogens also cause a reduction in transcript levels

of genes involved in primary metabolism, even before disease

development (e.g. Zimmerli et al., 2004; Thilmony et al., 2006;

Truman et al., 2006; Gjetting et al., 2007). This indicates that

SYP121 and SYP122 function as regulators of a signaling node

that controls defense pathways, without affecting primary me-

tabolism. The log2 of the change of the 365 genes in all 83 Gen-

evestigator categories was found to be 0.26, showing that this

set of genes are not common response genes of the many dif-

ferent kinds of stresses in this collection. This demonstrates that

the syp121 and syp122 mutations specifically elicit defense-like

gene expression changes.

The role of the JA and ET signaling pathways in the syntaxin

double mutant syp121 syp122 is less clear. A few marker genes

indicative of constitutive ET signaling are up-regulated; how-

ever, ET biosynthesis genes are not up-regulated in syp121

syp122. Yet, mutating EIN2 in syp121 syp122 suggested that

ET signaling plays a role in delaying lesion formation. ET seems

to have conflicting regulatory functions in pathogen defense

(van Loon et al., 2006). ET can apparently both promote and

inhibit PCD. It is thought to potentiate senescence-related

PCD (Oh et al., 1997), ozone-induced PCD (Overmyer et al.,

2000), and vad1 mutant PCD (Bouchez et al., 2007). On the

other hand, resistance to the necrotrophic fungus, B. cinerea,

requires functional EIN2 and application of ET reduces suscep-

tibility (Thomma et al., 1999), suggesting that ET acts antago-

nistically to PCD. Similar to the latter case, it is likely that

introducing the ein2-1 allele into syp121–1 syp122–1 interrupts

a constitutive ETsignal that otherwise serves to protect against

cell death. Interestingly, the lesion formation, delayed by ET,

appears to depend on a SID2-mediated signal. This is seen from

the fact that the ein2-1 allele does not promote lesion forma-

tion in syp121–1 syp122–1 sid2-1. However, it does not depend

on NPR1, since ein2-1 does promote lesion formation in

syp121–1 syp122–1 npr1-1 (see Figure 5). An NPR1-indepen-

dent, SID2-mediated signal in syp121–1 syp122–1 that poten-

tially can be antagonized by ET is described below.

JA can also have both anti-PCD (Overmyer et al., 2000) and

pro-PCD (Mur et al., 2006) functions. However, while we

did not observe an effect of interrupting JA signaling

in syp121–1 syp122–1, using mutations in three different

genes, there were potentially anti- and pro-PCD effects of

Figure 7. syp121 syp122 is SA-Hypersensitive, Independently of En-dogenous SA.

Necrosis stimulation after SA-treatment of syp121–1 syp122–1 andsyp121–1 syp122–1 rescued by additional mutations, and of otherlesion-mimic mutants (3-week-old plants).

Zhang et al. d syp121 syp122 Unravels Signaling Network | 521

JA signaling, depending on which SA signaling elements were

knocked out (see Figure 5). Further analyses are required in or-

der to explain these observations.

EDS1 is required for development of the syntaxin double

mutant lesion-mimic phenotype (Figure 1; Zhang et al.,

2007). Since EDS1 and PAD4 are closely associated functionally

(Wiermer et al., 2005), PAD4 would also be expected to play

a role. We were originally unable to introduce a mutation

of PAD4 into syp121–1 syp122–1 due to extremely close genetic

linkage. However, this barrier was overcome by isolation of in-

duced mutations in PAD4 after mutagenesis of syp121–1

syp122–1 and selection of genotypes with improved perfor-

mance. This confirmed the requirement of PAD4. In the same

way, mutant alleles of FMO1 and ALD1 were encountered,

demonstrating requirement of these genes for development

of the syp121 syp122 lesion-mimic phenotype. All in all, we es-

timate that PAD4, FMO1, or ALD1 mutant alleles, as well as

SID2 or EDS5 mutant alleles, each constitute 5–10% of the

240 ssd alleles recovered from the suppressor screen.

The finding that EDS1, PAD4, NDR1, RAR1, and SGT1b were

required for the lesion phenotype suggests that R-gene-

mediated types of signals are activated when both SYP121

and SYP122 are mutated. The requirement for all five genes

suggests that both CC-NB-LRR and TIR-NB-LRR types of signals

are active, but the stronger effects of the EDS1 and PAD4 muta-

tions suggest that the TIR-NB-LRR type may dominate. This was

further supported by the HR-like response to Gc being depen-

dent on EDS1 and PAD4, but not on NDR1. We have no direct

evidence to support a direct role for R-proteins in signal acti-

vation in syp121 syp122, but the simplest explanation is that

they are. Activation of NB-LRRs is not dependent on the pres-

ence of pathogens per se. In fact, NB-LRRs can be activated by

amino acid-substitutions, truncations (Zhang et al., 2003;

Howles et al., 2005; Takken et al., 2006; Weaver et al.,

2006), or simply over-expression (Stokes et al., 2002). RAR1

and SGT1b are thought to be required for the stability of

NB-LRRs (Holt et al., 2005), and the need for these genes in

the syp121 syp122 lesion-mimic phenotype development

may support that NB-LRRs themselves are involved. Therefore,

the observation that the R-gene-mediated defenses may be ac-

tivated by the syntaxin double mutant suggests that intact ves-

icle traffic is required for maintaining a stable balance

between active and inactive NB-LRR proteins in the absence

of recognizable effectors.

We found that the many active signaling pathways in

syp121 syp122 make this genotype a highly useful tool for

the study of interactions between the different pathogen de-

fense pathways. For this purpose, we combined two or more

knockout mutations in positive defense regulators in the

syp121 syp122 genetic background. The most common obser-

vation was that this improved the rescue, which, in general,

indicates that these regulators control independent signals

that make separate contributions to the lesion-mimic pheno-

type. Yet, we also encountered improved rescue by combining

presumed closely associated signaling components by what

appears tobereliefofsignalcompoundaccumulation.Anexam-

ple of this is probably what we observe when eds5-3 or sid2-1

improves the rescuing effect of npr1-1. In contrast, when no im-

proved rescue is observed after combining two mutations, this

strongly suggests that these signaling components are closely

associated.

The result of combining knockouts in EDS5 and SID2

strongly indicates that each of these two genes are involved

in mediating signals in addition to SA. Brodersen et al.

(2005) suggested that SID2 is involved in the biosynthesis of

an additional signaling molecule that might also be sensitive

to the hydroxylase activity of NahG. This is consistent with our

finding, in that expression of NahG in syp121–1 syp122–1 im-

proved the performance significantly compared to inactiva-

tion of EDS5 or SID2 (Supplemental Figure 5). Heck et al.

(2003) reach a similar conclusion regarding the effects of NahG

expression.

Abolishing SA in syp121–1 syp122–1 npr1-1 by introducing

mutant alleles of EDS5 or SID2 improved the performance of

the plants. This suggests that SA contributes to the lesion-

mimic phenotype also in the absence of functional NPR1,

and that SA in itself has NPR1-independent effects. We corre-

late this phenomenon to tissue bleaching caused by SA in the

absence of functional NPR1. It remains unclear whether this

effect is caused by SA toxicity or whether a true signaling ef-

fect of SA is involved. A similar SA-induced bleaching in npr1-1

was observed by Bowling et al. (1997) and Cao et al. (1997). SA-

dependent, NPR1-independent signaling has been suggested

several times (Nandi et al., 2003; Yoshioka et al., 2001; Durrant

and Dong, 2004; Desveaux et al., 2004; Uquillas et al., 2004;

Tsutsui et al., 2006; Yaeno et al., 2006; Zhang et al., 2003),

but further investigations are needed in order to establish

the nature of the SA-related bleaching.

The recently discovered signaling element FMO1 is sug-

gested (1) to be involved in TIR-NB-LRR resistance mediated

through EDS1 (Bartsch et al., 2006), (2) to be essential for

SA accumulation in distal but not local leaves during SAR

(Mishina and Zeier, 2006), and (3) to be required for basal de-

fense (Koch et al., 2006). The aminotransferase ALD1 has also

been reported to be required for SAR and SA accumulation in

distal leaves, as well as for resistance in local leaves. However,

mutation in ALD1 only caused a minor reduction in the SA ac-

cumulation in local leaves. The loss of resistance in the ald1-T2

T-DNA mutant line could not be rescued by treatments with

the SA analogue, BTH, suggesting that another mechanism

is involved (Song et al., 2004b). Furthermore, ALD1 is sug-

gested to function separately from PAD4 in terms of pathogen

defense and PCD development in the lesion-mimic mutant

acd6 (Song et al., 2004b).

We find it interesting that mutations in FMO1 and in ALD1

partially rescued the lesion-mimic phenotype of syp121

syp122. Firstly, this confirms once again that the syntaxin

double mutant phenotype is caused by a concurrent activation

of several defense signaling pathways. Secondly, the in-

volvement of FMO1 in TIR-NB-LRR resistance supports that

522 | Zhang et al. d syp121 syp122 Unravels Signaling Network

R-gene-mediated types of signals are activated in the syntaxin

double mutant. Furthermore, it is noteworthy that the qua-

druple mutants, where we mutated FMO1 or ALD1 on one

hand, and PAD4, EDS5, and SID2 on the other hand, all

exhibited improved rescue relative to all five triple mutants.

This suggests that both FMO1 and ALD1 may function in par-

allel to SA signaling. Regarding FMO1, this agrees with the ad-

ditive effects of FMO1 and SID2 mutations on disease

resistance (Bartsch et al., 2006) and failure of FMO1 mutations

to affect local leaf SA level after pathogen attack (Bartsch et

al., 2006; Mishina and Zeier, 2006). Regarding ALD1, it is com-

parable to the situation for the triple mutant pad4 ald1 acd6,

described by Song et al. (2004b), that performs better than

ald1 acd6 and pad4 acd6. In contrast, the fact that combina-

tions of knockouts of FMO1 and NPR1, as well as of ALD1

and NPR1, do not give an improved rescue of the syntaxin dou-

ble mutant may suggest that these two genes mediate signals

that fully depend on NPR1.

A qPCR PR-1 transcript analysis has been conducted on most

of the lines in Table 3 (data not shown). An inverse correlation

between the level of rescue and the PR-1 transcript expression

was found. While the jar1 and ein2 mutations did not reduce

the high PR-1 transcript level in syp121 syp122, rar1, sgt1b, and

ndr1 reduced it 10-fold. Meanwhile, eds5, sid2, npr1, fmo1,

and NahG reduced the level approximately a 1000-fold, thereby

approaching wild-type transcript level. The PR-1 transcript level

in the quadruple mutants, having at least one of the eds5, sid2,

or npr1 mutations, was obviously also reduced to wild-type lev-

els. For that reason, this analysis was not able to provide support

for the phenotype assay of the effect of combining mutations in

two or more positive defense regulators.

Not surprisingly, the syntaxin double mutant is powdery mil-

dew resistant. Unlike the single epidermal cell HR seen in Ara-

bidopsis in response to Bgh, this resistance is manifested as

a multi-cellular death reaction around the attacked epidermal

cell, after the initial haustorium is established (Zhang et al.,

2007). We have not yet encountered a signaling pathway that

alone is required for this resistance. Rather, it appears to be

dependent on the concerted action of a number of signaling

pathways. Mutation of PAD4 reduces the resistance to the

greatest extent. On the other hand, single or double mutations

of EDS5, SID2, and NPR1 do not affect the resistance. However,

when combined with a PAD4 mutation, they reduced resis-

tance further (Figure 6). Deciphering the complete signaling

events that lead to this resistance awaits future analyses. In

the case of the SA-hypersensitivity of the syntaxin double mu-

tant, we have found NPR1 to be strictly required. NPR1 possibly

functions like a downstream ‘receptor’ for SA, and therefore

may not be involved per se in making the plant hypersensitive.

Interestingly, both the powdery mildew resistance and the SA-

hypersensitivity have the potential to be mediated by R-gene-

type of signaling, which would agree well with the notion

above that such signaling appears active in syp121 syp122.

On the basis of well established signaling connections pre-

sented in the literature, we summarize some of our observa-

tions for the signaling network, activated by mutation of

SYP121 and SYP122, in the model shown in Figure 8. We spec-

ulate that the broad activation of defense pathways reflects an

equally broad activation of R-proteins, leaving many

unknowns to be uncovered. Thereby, syp121 syp122 may ex-

press responses that otherwise would require several inocula-

tion experiments to obtain. This underpins the usefulness of

this line for studies of the signaling network. For instance,

we find it intriguing that NPR1 is the only gene of those tested

that appears to be needed for the signaling leading to the SA-

hypersensitivity. This suggests that more active defense signal-

ing pathways can still be found in the syntaxin double mutant.

METHODS

Plants

Arabidopsis thaliana plants were grown at 20�C, 125 lE sm�2

in short day (8 h light) conditions. Genotypes are described in

Supplemental Table 2. The mutant lines jar1-1, coi1-1, ein2-1,

ndr1-1, sgt1b-1, rar1-10, acd2-2, cpr1, as well as T-DNA inser-

tion lines, were provided through NASC (http://arabidopsis.

info). NahG and npr1-1 were provided by Xinnian Dong (Duke

University, NC, USA), eds1-2 by Jane Parker (Max Planck Insti-

tute, Cologne, Germany), eds5-3 and sid2-1 by Jean-Pierre

Metraux (University of Fribourg, Switzerland), mpk4 by John

Mundy (University of Copenhagen), and npr1-N4-7 by Shu-

nyuan Xiao (University of Maryland Biotechnology Institute,

MD, USA). Triple mutants of syp121–1, syp122–1 and various

positive defense signaling mutants were selected in F3 families

derived from F2 plants showing the syp121 syp122-dwarf/

necrosis-phenotype. Homozygous triple mutant F3 plants were

identified either as partially rescued plants or using molecular

markers for the mutation in the positive defense signaling

Figure 8. Hypothetical Model for the Signaling Network, Activatedby the Knockout of SYP121 and SYP122.

Zhang et al. d syp121 syp122 Unravels Signaling Network | 523

gene. Subsequently, the mutations in positive defense signaling

were confirmed using molecular markers. The alleles eds1-2,

rar1-10, and sgt1b-1 are crossed in from Ler, and the rescuing

effects of these are confirmed in four independent F3 descents

for each (data not shown). Quadruple and quintuple mutants

were obtained in the same manner by crossing triple and qua-

druple mutants.

Expression Profiling

Col-0, syp121–1, syp122–1, and syp121–1 pen122–1 were har-

vested in three replicas 2.5 weeks after sowing, prior to the

appearance of lesions on syp121 syp122. Total RNA was

extracted using an RNeasy kit (Qiagen GmbH, Hilden,

Germany) and labeling and hybridization on the ATH1 micro-

arrays (Thibaud-Nissen et al., 2006) (one sample per chip) were

performed according to the manufacturers instructions

(www.affymetrix.com/support/technical/manual/expression_

manual.affx). Raw intensity data was normalized using the R

(R Development Core Team, 2004) Affy package (Gautier et al.,

2004) using the perfect match-only implementation of dCHIP

(Li and Wong, 2001) and qspline (Workman et al., 2002). We

applied one-way ANOVA in order to identify statistical signif-

icant differentially expressed genes. False-positive rates were

estimated by recalculating P-values with permuted sample

categories. This procedure was repeated four times, generat-

ing four sets of 22 810 permuted P-values. From this, the

chosen P-value cut-off was estimated to contain a maximum

of 15 false-positives (i.e. ;2.5%).

Mutagenesis of syp121 syp122

Approximately 30 000 Col-0 syp121–1 syp122–1 seeds were

mutagenized using 0.2% ethyl methane-sulfonate (EMS) at

room temperature for 12 h, followed by continuous rinsing

with water for 3 h. The seeds were subsequently sown on soil

for production of M2 seeds, which were harvested into ten

pools. M2 plants, partially rescued for the lesion-mimic pheno-

type of syp121 syp122, were selected in the greenhouse.

Genetic Mapping

In order to generate mapping populations for the SSD genes,

the syp121–1 and syp122–1 mutant alleles were first back-

crossed into Ler. Thereby, a line with the typical syntaxin dou-

ble mutant phenotype was obtained, which we confirmed to

be homozygous Ler at several genetic marker loci on each of

chromosomes 1, 2, 4, and 5. Both SYP121 and SYP122 are lo-

cated on chromosome 3, and no effort was made to transfer

Ler DNA to this chromosome. F2 populations were generated

for each of the SSD genes (Col-0 syp121–1 syp122–1 ssd3Ler

syp121–1 syp122–1). Initial analyses of 22 rescued F2 plants

in each population using published genetic markers (Bell

and Ecker, 1994; Lukowitz et al., 2000) suggested map-

positions. Subsequent fine-mappings were performed using

the Cereon Col/Ler marker information (www.arabidopsis.

org/cereon/index.jsp).

Salicylic Acid Measurements

Extraction and determination of total (free plus glycosylated)

SA in shoots was performed by HPLC as described by Newman

et al. (2001).

Plant Treatments

The Arabidopsis powdery mildew fungus (Golovinomyces

cichoracearum, isolate UCSC1) was propagated on pumpkin

plants. Inoculations were performed as described by Collins

et al. (2003). SA treatment was performed by spraying plants

with 250 lM SA once a day for five consecutive days. The treat-

ment was initiated when the plants were 16 d old. Photos were

taken 1 d after the last treatment.

SUPPLEMENTARY DATA

Supplementary Data are available at Molecular Plant

Online.

FUNDING

Financial support has been provided by the Danish Agricul-

tural and Veterinary Research Council; the Danish Agency

for Science, Technology and Innovation; Carl Tryggers Founda-

tion, Sweden; the Carlsberg Foundation, Denmark; the Direc-

torate for Food, Fisheries and Agri Business, Denmark; the US

National Science Foundation; and the Carnegie Institution of

Science, USA.

ACKNOWLEDGMENTS

We thank the Salk Institute Genomic Analysis Laboratory for pro-

viding the sequence-indexed Arabidopsis T-DNA insertion mutants.

We thank Dr Peter Hagedorn for initial analyses of the global tran-

script profiles, and Dr Dale Godfrey for critically reading the man-

uscript. We thank the undergraduate students Louise Flagstad,

Stine Petersen, Irene S. Rasmussen and Sabine B. Sørensen for map-

ping a mutant allele.

No conflict of interest declared.

REFERENCES

Assaad, F.F., et al. (2004). The PEN1 syntaxin defines a novel cellular

compartment upon fungal attack and is required for the timely

assembly of papillae. Mol. Biol. Cell 15, 5118–5129.

Bartsch, M., Gobbato, E., Bednarek, P., Debey, S., Schultze, J.L.,

Bautor, J., and Parker, J.E. (2006). Salicylic acid-independent EN-

HANCED DISEASE SUSCEPTIBILITY1 signaling in Arabidopsis im-

munity and cell death is regulated by the monooxygenase FMO1

and the Nudix hydrolase NUDT7. Plant Cell 18, 1038–1051.

Beckers, G.J., and Spoel, S.H. (2006). Fine-tuning plant defence sig-

nalling, salicylate versus jasmonate. Plant Biol. (Stuttg.) 8, 1–10.

Bell, C.J., and Ecker, J.R. (1994). Assignment of 30 microsatellite loci

to the linkage map of Arabidopsis. Genomics 19, 137–144.

Benavente, L.M., andAlonso, J.M. (2006). Molecular mechanisms of

ethylene signaling in Arabidopsis. Mol. Biosyst 2, 165–173.

524 | Zhang et al. d syp121 syp122 Unravels Signaling Network

Benschop, J.J., Mohammed, S., O’Flaherty, M., Heck, A.J.R.,

Slijper, M., and Menke, F.L.H. (2007). Quantitative phosphopro-

teomics of early elicitor signaling in Arabidopsis. Mol. Cell. Pro-

teom 6, 1198–1214.

Bohlmann, H., Vignutelli, A., Hilpert, B., Miersch, O.,

Wasternack, C., and Apel, K. (1998). Wounding and chemicals in-

duce expression of the Arabidopsis thaliana gene Thi2.1, encod-

ing a fungal defense thionin, via the octadecanoid pathway.

FEBS Lett 437, 281–286.

Bouchez, O., Huard, C., Lorrain, S., Roby, D., and Balague, C. (2007).

Ethylene is one of the key elements for cell death and defense

response control in the Arabidopsis lesion mimic mutant vad1.

Plant Physiol 145, 465–477.

Bowling, S.A., Clarke, J.D., Liu, Y., Klessig, D.F., and Dong, X. (1997).

The cpr5 mutant of Arabidopsis expresses both NPR1-dependent

and NPR1-independent resistance. Plant Cell 9, 1573–1584.

Brodersen, P., Malinovsky, F.G., Hematy, K., Newman, M.A., and

Mundy, J. (2005). The role of salicylic acid in the induction of cell

death in Arabidopsis acd11. Plant Physiol 138, 1037–1045.

Brodersen, P., Petersen, M., Pike, H.M., Olszak, B., Skov, S.,

Odum, N., Jorgensen, L.B., Brown, R.E., and Mundy, J. (2002).

Knockout of Arabidopsis ACCELERATED-CELL-DEATH11 encod-

ing a sphingosine transfer protein causes activation of pro-

grammed cell death and defense. Genes Dev 16, 490–502.

Broekaert, W.F., Delaure, S.L., De Bolle, M.F., and Cammue, B.P.

(2006). The role of ethylene in host-pathogen interactions.

Annu. Rev. Phytopathol 44, 393–416.

Cao, H., Glazebrook, J., Clarke, J.D., Volko, S., and Dong, X. (1997).

The Arabidopsis NPR1 gene that controls systemic acquired resis-

tance encodes a novel protein containing ankyrin repeats. Cell

88, 57–63.

Clarke, J.D., Volko, S.M., Ledford, H., Ausubel, F.M., and Dong, X.

(2000). Roles of salicylic acid, jasmonic acid, and ethylene in cpr-

induced resistance in Arabidopsis. Plant Cell 12, 2175–2190.

Collins, N.C., et al. (2003). SNARE-protein-mediated disease resis-

tance at the plant cell wall. Nature 425, 973–977.

Desveaux, D., Subramaniam, R., Despres, C., Mess, J.N.,

Levesque, C., Fobert, P.R., Dangl, J.L., and Brisson, N. (2004). A

‘Whirly’ transcription factor is required for salicylic acid-

dependent disease resistance in Arabidopsis. Dev. Cell 6,

229–240.

Doelling, J.H., Walker, J.M., Friedman, E.M., Thompson, A.R., and

Vierstra, R.D. (2002). The APG8/12-activating enzyme APG7 is re-

quired for proper nutrient recycling and senescence inArabidop-

sis thaliana. J. Biol. Chem 277, 33105–33114.

Durrant, W.E., and Dong, X. (2004). Systemic acquired resistance.

Annu. Rev. Phytopathol 42, 185–209.

Feys, B., Benedetti, C.E., Penfold, C.N., and Turner, J.G. (1994). Ara-

bidopsis mutants selected for resistance to the phytotoxin coro-

natine are male sterile, insensitive to methyl jasmonate, and

resistant to a bacterial pathogen. Plant Cell 6, 751–759.

Foyer, C.H., and Noctor, G. (2005). Redox homeostasis and antiox-

idant signaling, a metabolic interface between stress perception

and physiological responses. Plant Cell 17, 1866–1875.

Fujiki, Y., Yoshimoto, K., and Ohsumi, Y. (2007). An Arabidopsis ho-

molog of yeast ATG6/VPS30 is essential for pollen germination.

Plant Physiol 143, 1132–1139.

Gautier, L., Cope, L., Bolstad, B.M., and Irizarry, R.A. (2004). Affy-

analysis of Affymetrix GeneChip data at the probe level. Bioin-

formatics 20, 307–315.

Gjetting, T., Hagedorn, P.H., Schweizer, P., Thordal-Christensen, H.,

Carver, T.L., and Lyngkjaer, M.F. (2007). Single-cell transcript pro-

filing of barley attacked by the powdery mildew fungus. Mol.

Plant-Microbe Interact 20, 235–246.

Glazebrook, J. (2005). Contrasting mechanisms of defense against

biotrophic and necrotrophic pathogens. Annu. Rev. Phytopathol

43, 205–227.

Hammond-Kosack, K.E., and Parker, J.E. (2003). Deciphering plant–

pathogen communication, fresh perspectives for molecular resis-

tance breeding. Curr. Opin. Biotechnol 14, 177–193.

Heck, S., Grau, T., Buchala, A., Metraux, J.-P., and Nawrath, C.

(2003). Genetic evidence that expression of NahG modifies de-

fense pathways independent of salicylic acid biosynthesis in

the Arabidopsis-Pseudomonas syringae pv. tomato interaction.

Plant J 36, 342–352.

Heese, A., Ludwig, A.A., and Jones, J.D. (2005). Rapid phosphory-

lation of a syntaxin during the Avr9/Cf-9-race-specific signaling

pathway. Plant Physiol 138, 2406–2416.

Holt, B.F.,III, Belkhadir, Y., and Dangl, J.L. (2005). Antagonistic con-

trol of disease resistance protein stability in the plant immune

system. Science 309, 929–932.

Howles, P., Lawrence, G., Finnegan, J., McFadden, H., Ayliffe, M.,

Dodds, P., and Ellis, J. (2005). Autoactive alleles of the flax L6 rust

resistance gene induce non-race-specific rust resistance associ-

ated with the hypersensitive response. Mol. Plant-Microbe Inter-

act 18, 570–582.

Hunt, M.D., Delaney, T.P., Dietrich, R.A., Weymann, K.B., Dangl, J.L.,

and Ryals, J.A. (1997). Salicylate-independent lesion formation in

Arabidopsis lsd mutants. Mol. Plant-Microbe Interact 10, 531–536.

Kalde, M., Nuhse, T.S., Findlay, K., and Peck, S.C. (2007). The syn-

taxin SYP132 contributes to plant resistance against bacteria

and secretion of pathogenesis-related protein 1. Proc. Natl Acad.

Sci. U S A 104, 11850–11855.

Koch, M., Vorwerk, S., Masur, C., Sharifi-Sirchi, G., Olivieri, N., and

Schlaich, N.L. (2006). A role for a flavin-containing mono-

oxygenase in resistance against microbial pathogens in Arabi-

dopsis. Plant J 47, 629–639.

Li, C., and Wong, W.H. (2001). Model-based analysis of oligonucle-

otide arrays: expression index computation and outlier detec-

tion. Proc. Natl Acad. Sci. U S A 98, 31–36.

Lipka, V., et al. (2005). Pre- and postinvasion defenses both contrib-

ute to nonhost resistance inArabidopsis. Science 310, 1180–1183.

Lipka, V., Kwon, C., and Panstruga, R. (2007). SNARE-ware, the role

of SNARE-domain proteins in plant biology. Annu. Rev. Cell Dev.

Biol 23, 147–174.

Liu, Y., Schiff, M., Czymmek, K., Talloczy, Z., Levine, B., and Dinesh-

Kumar, S.P. (2005). Autophagy regulates programmed cell death

during the plant innate immune response. Cell 121, 567–577.

Lorrain, S., Vailleau, F., Balague, C., and Roby, D. (2003). Lesion

mimic mutants, keys for deciphering cell death and defense

pathways in plants? Trends Plant Sci 8, 263–271.

Lukowitz, W., Gillmor, C.S., and Scheible, W.R. (2000). Positional

cloning in Arabidopsis: why it feels good to have a genome ini-

tiative working for you. Plant Physiol 123, 795–805.

Zhang et al. d syp121 syp122 Unravels Signaling Network | 525

Mishina, T.E., and Zeier, J. (2006). The Arabidopsis flavin-dependent

monooxygenase FMO1 is an essential component of biologically

induced systemic acquired resistance. Plant Physiol 141,

1666–1675.

Mur, L.A., Kenton, P., Atzorn, R., Miersch, O., and Wasternack, C.

(2006). The outcomes of concentration-specific interactions be-

tween salicylate and jasmonate signaling include synergy, antag-

onism, and oxidative stress leading to cell death. Plant Physiol

140, 249–262.

Nandi, A., Kachroo, P., Fukushige, H., Hildebrand, D.F., Klessig, D.F.,

and Shah, J. (2003). Ethylene and jasmonic acid signaling affect

the NPR1-independent expression of defense genes without

impacting resistance to Pseudomonas syringae and Peronospora

parasitica in the Arabidopsis ssi1 mutant. Mol Plant-Microbe In-

teract 16, 588–599.

Nawrath, C., Heck, S., Parinthawong, N., and Metraux, J.P. (2002).

EDS5, an essential component of salicylic acid-dependent signal-

ing for disease resistance in Arabidopsis, is a member of the

MATE transporter family. Plant Cell 14, 275–286.

Newman, M.A., von Roepenack-Lahaye, E., Parr, A., Daniels, M.J.,

and Dow, J.M. (2001). Induction of hydroxycinnamoyl-tyramine

conjugates in pepper by Xanthomonas campestris, a plant de-

fense response activated by hrp gene-dependent and hrp

gene-independent mechanisms. Mol. Plant-Microbe Interact

14, 785–792.

Norman-Setterblad, C., Vidal, S., and Palva, E.T. (2000). Interacting

signal pathways control defense gene expression in Arabidopsis

in response to cell wall-degrading enzymes from Erwinia caroto-

vora. Mol. Plant-Microbe Interact 13, 430–438.

Nuhse, T.S., Boller, T., and Peck, S.C. (2003). A plasma membrane

syntaxin is phosphorylated in response to the bacterial elicitor

flagellin. J. Biol. Chem 278, 45248–45254.

Oh, S.A., Park, J.H., Lee, G.I., Paek, K.H., Park, S.K., and Nam, H.G.

(1997). Identification of three genetic loci controlling leaf senes-

cence in Arabidopsis thaliana. Plant J 12, 527–535.

Olszak, B., Malinovsky, F.G., Brodersen, P., Grell, M., Giese, H.,

Petersen, M., and Mundy, J. (2006). A putative flavin-containing

mono-oxygenase as a marker for certain defense and cell death

pathways. Plant Sci 170, 614–623.

Overmyer, K., Tuominen, H., Kettunen, R., Betz, C., Langebartels, C.,

Sandermann, H., and Kangasjarvi, J. (2000). Ozone-sensitive Ara-

bidopsis rcd1 mutant reveals opposite roles for ethylene and

jasmonate signaling pathways in regulating superoxide-

dependent cell death. Plant Cell 12, 1849–1862.

Park, J.H., Halitschke, R., Kim, H.B., Baldwin, I.T., Feldmann, K.A.,

and Feyereisen, R. (2002). A knock-out mutation in allene oxide

synthase results in male sterility and defective wound signal

transduction in Arabidopsis due to a block in jasmonic acid bio-

synthesis. Plant J 31, 1–12.

Penninckx, I.A.M.A., Thomma, B.P.H.J., Buchala, A., Metraux, J.P.,

and Broekaert,W.F. (1998). Concomitant activation of jasmonate

and ethylene response pathways is required for induction of

a plant defensin gene in Arabidopsis. Plant Cell 10, 2103–2113.

Potter, S., Uknes, S., Lawton, K., Winter, A.M., Chandler, D.,

Dimaio, J., Novitzky, R., Ward, E., and Ryals, J. (1993). Regulation

of a hevein-like gene in Arabidopsis. Mol. Plant-Microbe Interact

6, 680–685.

Raffaele, S., Rivas, S., and Roby, D. (2006). An essential role for

salicylic acid in AtMYB30-mediated control of the hypersensi-

tive cell death program in Arabidopsis. FEBS Lett 580,

3498–3504.

Rairdan, G.J., and Delaney, T.P. (2002). Role of salicylic acid and

NIM1/NPR1 in race-specific resistance in Arabidopsis. Genetics

161, 803–811.

Rao, M.V., Lee, H., Creelman, R.A., Mullet, J.E., and Davis, K.R.

(2000). Jasmonic acid signaling modulates ozone-induced hyper-

sensitive cell death. Plant Cell 12, 1633–1646.

Redman, J.C., Haas, B.J., Tanimoto, G., and Town, C.D. (2004). De-

velopment and evaluation of an Arabidopsis whole genome

Affymetrix probe array. Plant J 38, 545–561.

Shah, J., Tsui, F., and Klessig, D.F. (1997). Characterization of a sal-

icylic acid-insensitive mutant (sai1) of Arabidopsis thaliana, iden-

tified in a selective screen utilizing the SA-inducible expression

of the tms2 gene. Mol. Plant-Microbe Interact 10, 69–78.

Shirasu, K., Nakajima, H., Rajasekhar, V.K., Dixon, R.A., and Lamb, C.

(1997). Salicylic acid potentiates an agonist-dependent gain con-

trol that amplifies pathogen signals in the activation of defense

mechanisms. Plant Cell 9, 261–270.

Song, J.T., Lu, H., and Greenberg, J.T. (2004a). Divergent roles in

Arabidopsis thaliana development and defense of two homolo-

gous genes, ABERRANT GROWTH AND DEATH2 and AGD2-LIKE

DEFENSE RESPONSE PROTEIN1, encoding novel aminotrans-

ferases. Plant Cell 16, 353–366.

Song, J.T., Lu, H.,McDowell, J.M., andGreenberg, J.T. (2004b). A key

role for ALD1 in activation of local and systemic defenses in Ara-

bidopsis. Plant J 40, 200–212.

Spoel, S.H., et al. (2003). NPR1 modulates cross-talk between salic-

ylate- and jasmonate-dependent defense pathways through

a novel function in the cytosol. Plant Cell 15, 760–770.

Staswick, P.E., and Tiryaki, I. (2004). The oxylipin signal jasmonic

acid is activated by an enzyme that conjugates it to isoleucine

in Arabidopsis. Plant Cell 16, 2117–2127.

Stein, M., Dittgen, J., Sanchez-Rodriguez, C., Hou, B.H., Molina, A.,

Schulze-Lefert, P., Lipka, V., and Somerville, S. (2006). Arabidop-

sis PEN3/PDR8, an ATP binding cassette transporter, contributes

to nonhost resistance to inappropriate pathogens that enter by

direct penetration. Plant Cell 18, 731–746.

Stokes, T.L., Kunkel, B.N., and Richards, E.J. (2002). Epigenetic var-

iation in Arabidopsis disease resistance. Genes Dev 16, 171–182.

Takken, F.L.W., Albrecht, M., and Tameling, W.I.L. (2006). Resistance

proteins, molecular switches of plant defence. Curr. Opin. Plant

Biol 9, 383–390.

Tenhaken, R., and Rubel, C. (1997). Salicylic acid is needed in hyper-

sensitive cell death in soybean but does not act as a catalase in-

hibitor. Plant Physiol 115, 291–298.

Thibaud-Nissen, F., Wu, H., Richmond, T., Redman, J.C., Johnson, C.,

Green, R., Arias, J., and Town, C.D. (2006). Development of Ara-

bidopsis whole-genome microarrays and their application to the

discovery of binding sites for the TGA2 transcription factor in sal-

icylic acid-treated plants. Plant Journal 47, 152–162.

Thilmony, R., Underwood,W., andHe, S.Y. (2006). Genome-wide tran-

scriptional analysis of theArabidopsis thaliana interaction with the

plant pathogen Pseudomonas syringaepv. tomatoDC3000 and the

human pathogen Escherichia coli O157, H7. Plant J 46, 34–53.

526 | Zhang et al. d syp121 syp122 Unravels Signaling Network

Thomma, B.P.H.J., Eggermont, K., Tierens, K.F.M.J., and

Broekaert, W.F. (1999). Requirement of functional Ethylene-

Insensitive 2 gene for efficient resistance of Arabidopsis to infec-

tion by Botrytis cinerea. Plant Physiol 121, 1093–1101.

Torres, M.A., Dangl, J.L., and Jones, J.D. (2002). Arabidopsis

gp91phox homologues AtrbohD and AtrbohF are required for

accumulation of reactive oxygen intermediates in the plant de-

fense response. Proc. Natl Acad. Sci. U S A 99, 517–522.

Torres, M.A., Jones, J.D., and Dangl, J.L. (2005). Pathogen-induced,

NADPH oxidase-derived reactive oxygen intermediates suppress

spread of cell death in Arabidopsis thaliana. Nat. Genet 37,

1130–1134.

Truman, W., de Zabala, M.T., and Grant, M. (2006). Type III effectors

orchestrate a complex interplay between transcriptional net-

works to modify basal defence responses during pathogenesis

and resistance. Plant J 46, 14–33.

Tsutsui, T.,Morita-Yamamuro, C., Asada, Y.,Minami, E., Shibuya, N.,

Ikeda, A., and Yamaguchi, J. (2006). Salicylic acid and a chitin elic-

itor both control expression of the CAD1 gene involved in the

plant immunity of Arabidopsis. Biosci. Biotech. Biochem 70,

2042–2048.

Uquillas, C., Letelier, I., Blanco, F., Jordana, X., and Holuigue, L.

(2004). NPR1-independent activation of immediate early salicylic

acid-responsive genes in Arabidopsis. Mol. Plant-Microbe Inter-

act 17, 34–42.

Van Der Straeten, D., Rodrigues-Pousada, R.A., Villarroel, R.,

Hanley, S., Goodman, H.M., and Van Montagu, M. (1992). Clon-

ing, genetic mapping, and expression analysis of an Arabidopsis

thaliana gene that encodes 1-aminocyclopropane-1-carboxylate

synthase. Proc. Natl Acad. Sci. U S A 89, 9969–9973.

van Loon, L.C., Geraats, B.P., and Linthorst, H.J. (2006). Ethylene as

a modulator of disease resistance in plants. Trends Plant Sci 11,

184–191.

Weaver, L.M., Swiderski, M.R., Li, Y., and Jones, J.D.G. (2006). The

Arabidopsis thaliana TIR-NB-LRR R-protein, RPP1A: protein local-

ization and constitutive activation of defence by truncated

alleles in tobacco and Arabidopsis. Plant J 47, 829–840.

Wiermer, M., Feys, B.J., and Parker, J.E. (2005). Plant immunity, the

EDS1 regulatory node. Curr. Opin. Plant Bio 8, 383–389.

Wildermuth, M.C., Dewdney, J., Wu, G., and Ausubel, F.M. (2001).

Isochorismate synthase is required to synthesize salicylic acid for

plant defence. Nature 414, 562–565.

Workman, C., Jensen, L.J., Jarmer, H., Berka, R., Gautier, L.,

Nielsen, H.B., Saxild, H.H., Nielsen, C., Brunak, S., and

Knudsen, S. (2002). A new nonlinear normalization method for

reducing variability in DNA microarray experiments. Genome

Biology 3, research0048.1–0048.16.

Xiao, S., Calis, O., Patrick, E., Zhang, G., Charoenwattana, P.,

Muskett, P., Parker, J.E., and Turner, J.G. (2005). The atypical re-

sistance gene, RPW8, recruits components of basal defence for

powdery mildew resistance in Arabidopsis. Plant J 42, 95–110.

Xie, D.X., Feys, B.F., James, S., Nieto-Rostro, M., and Turner, J.G.

(1998). COI1, an Arabidopsis gene required for jasmonate-regu-

lated defense and fertility. Science 280, 1091–1094.

Yaeno, T., Saito, B., Katsuki, T., and Iba, K. (2006). Ozone-induced

expression of the Arabidopsis FAD7 gene requires salicylic acid,

but not NPR1 and SID2. Plant Cell Physiol 47, 355–362.

Yoshioka, K., Kachroo, P., Tsui, F., Sharma, S.B., Shah, J., and

Klessig, D.F. (2001). Environmentally sensitive, SA-dependent de-

fense responses in the cpr22 mutant of Arabidopsis. Plant J 26,

447–459.

Zhang,Y., Goritschnig, S., Dong,X., and Li, X. (2003). A gain-of-func-

tion mutation in a plant disease resistance gene leads to consti-

tutive activation of downstream signal transduction pathways in

suppressor of npr1-1, constitutive 1. Plant Cell 15, 2636–2646.

Zhang, Z., Feechan, A., Pedersen, C., Newman, M.A., Qiu, J.L.,

Olesen, K.L., and Thordal-Christensen, H. (2007). A SNARE-

protein has opposing functions in penetration resistance and de-

fence signalling pathways. Plant J 49, 302–312.

Zhong, G.Y., and Burns, J.K. (2003). Profiling ethylene-regulated

gene expression in Arabidopsis thaliana by microarray analysis.

Plant Mol. Biol 53, 117–131.

Zimmerli, L., Stein,M., Lipka, V., Schulze-Lefert, P., andSomerville, S.

(2004). Host and non-host pathogens elicit different jasmonate/

ethylene responses in Arabidopsis. Plant J 40, 633–646.

Zimmermann, P., Hirsch-Hoffmann, M., Hennig, L., and

Gruissem, W. (2004). GENEVESTIGATOR. Arabidopsis microarray

database and analysis toolbox. Plant Physiol 136, 2621–2632.

Zhang et al. d syp121 syp122 Unravels Signaling Network | 527