Genes Controlling Expression of Defense Responses in Arabidopsis

301

Genes controlling expression of defense responses in Arabidopsis — 2001 status Jane Glazebrook

In the past two years, the focus of studies of the genes sufficient speed, these responses are generally quite effective controlling expression of defense responses in Arabidopsis in preventing disease. Consequently, there is considerable has shifted from the identification of mutants to gene isolation interest in understanding the signal transduction networks and the ordering of genes within branches of the signal that control the activation of defense responses in plants. transduction networks. It is now clear that gene-for-gene One approach to this problem is to use the powerful resistance can be mediated through at least three genetically Arabidopsis genetic system to identify plant genes that distinguishable pathways. Additional genes affecting salicylic- have crucial roles in regulating the activation of defense acid-dependent signaling have been identified, and double- responses. This article summarizes progress made using mutant analyses have begun to reveal the order in which they this system since the previous review of this topic was act. Genes required for jasmonic-acid-dependent signaling and written [1]. A model describing the circuitry of the signal for induced systemic resistance have also been identified. transduction network is presented in Figure 1.

Addresses Gene-for-gene resistance is a particularly strong form of Torrey Mesa Research Institute, 3115 Merryfield Row, San Diego, plant disease resistance. Plants carry specific resistance (R) California 92121, USA; e-mail: [email protected] genes that are able to recognize pathogens carrying corre- Current Opinion in Plant Biology 2001, 4:301–308 sponding avirulence (avr) genes. This reaction triggers a rapid defense response that generally includes the pro- 1369-5266/01/$ — see front matter grammed cell death of plant cells that are in contact with © 2001 Elsevier Science Ltd. All rights reserved. the pathogen, a phenomenon called the hypersensitive Abbreviations response (HR). Widespread difficulties in detecting direct ACD2 ACCELERATED CELL DEATH 2 interactions between R and avr proteins have led to the avr avirulence coi1 coronatine insensitive 1 hypothesis that R proteins ‘guard’ plant proteins cpr1 constitutive PR 1 (i.e. ‘guardees’) that are the targets of pathogen avr dnd1 defense no death 1 proteins, triggering the HR and other responses when dth9 detachment 9 avr–guardee interactions are detected [2]. The detection of EDS1 ENHANCED DISEASE SUSCEPTIBILITY 1 an in vivo complex containing an R protein, an avr protein, ein2 ethylene insensitive2 ET ethylene and an unidentified plant protein is consistent with this HR hypersensitive response hypothesis [3]. This article does not discuss the extensive ISR induced systemic resistance literature concerning function of R proteins, but does JA jasmonic acid describe signaling downstream from R protein function. jar1 jasmonic acid resistant 1 LRR leucine-rich repeat LZ leucine-zipper Activation of the HR triggers a systemic resistance MPK4 MITOGEN-ACTIVATED PROTEIN KINASE 4 response known as systemic acquired resistance (SAR). nahG salicylcate hydroxylase This response includes the accumulation of the signal NBS nucleotide-binding site NDR1 NON-RACE-SPECIFIC DISEASE RESISTANCE 1 molecule salicylic acid (SA) throughout the plant and the NPR1 NON-EXPRESSOR OF PR 1 consequent expression of a characteristic set of defense PAD4 PHYTOALEXIN DEFICIENT 4 genes, including PR-1. Plants expressing SAR are more PBS1 avrPphB Susceptible 1 resistant to subsequent attack by a variety of otherwise PDF1.2 PLANT DEFENSIN1.2 PR-1 PATHOGENESIS-RELATED PROTEIN-1 virulent pathogens. Many defense responses that are R resistance characteristic of SAR also contribute to local resistance that RPP7 RESISTANCE TO PERONOSPORA PARASITICA 7 is mediated by some R genes, and to the local growth RPS5 RESISTANCE TO PSEUDOMONAS SYRINGAE 5 limitation of moderately virulent pathogens. In a large RPW8 RESISTANCE TO POWDERY MILDEW 8 expression profiling experiment, at 48 hours after infec- SA salicylic acid SAR systemic acquired resistance tion, the spectra of plant genes expressed in response to a SID1 SA INDUCTION DEFICIENT 1 virulent Peronospora parasitica isolate and an isolate that sni1 suppressor of npr1-1 inducible triggers gene-for-gene resistance were similar [4••]. TIR toll-interleukin 2 receptor Responses triggered by R genes generally occur much faster than responses to virulent pathogens; this presumably Introduction explains why R-gene-mediated resistance is so effective. Plants defend themselves from attack by microbial pathogens by activating a battery of defense responses Some defense responses are activated by signal transduction after infection. Provided that they are activated with networks that require jasmonic acid (JA) and ethylene (ET) 302 Biotic interactions

Figure 1

A model describing the positions of (a) Gene-for-gene resistance (b) SA-dependent resistance (c) Dependent resistance Arabidopsis genes in signal transduction networks that control the activation of defense R gene R gene R gene Systemic or pathogen-derived signal Systemic or pathogen-derived signal responses. This figure is meant to provide a RPM1 RPW8 RPP7 DND1 RPS2 RPP2 RPP8 EDR1 CPR5 rough outline of the sites of action of various DND2 cpr6 RPS5 RPP4 RPP13 cpr6 gene products in disease resistance signaling. ACD2 RPP5 CPR1 etc. It does not account for all of the described RPP10 phenotypes of the various signaling mutants. RPP14 RPS4 PAD4 Lesion formation JA (OPR3) ET Boxes containing the word ‘resistance’ EDS1 represent the numerous effectors of COI1 EIN2 NDR1 EDS1 resistance, whose effectiveness may vary SID2 Camalexin JAR1 PBS2 PAD4 greatly according to the pathogen eliciting the EDS5 MPK4 response. There are certainly interconnections among the pathways shown in (a), (b), (c), Resistance SA MPK4 and (d), but many of these are not sufficiently PDF1.2 etc. clear to allow them to be easily drawn. Induced systemic NPR1 DTH9 (a) Three R-gene dependent pathways are (d) Resistance resistance shown, one that requires NDR1 and PBS2, a TGA factors ISR second that requires EDS1 and PAD4, and a SNI1 third for which the required genes have not PR-1 etc. been reported. (b) In SA-dependent signaling, JA CPR1, EDR1, and CPR5 tend to repress ET pathway activation, and act upstream of Resistance PAD4. The cpr6 mutation exerts a positive effect upstream of PAD4, but it is difficult to ISR1 predict the function of the wild-type gene as NPR1 cpr6 is dominant. PAD4 and EDS1 promote SA accumulation, as do SID2 and EDS5. SA also promotes expression of PAD4 and Resistance EDS1. PAD4 promotes expression of Current Opinion in Plant Biology resistance responses in an SA-independent manner. Formation of lesions promotes SA accumulation in a manner that is independent as PDF1.2. This activation is promoted by NPR1 is required downstream of JA and ET of PAD4 and EDS1, but may require EDS5. lesion formation and by cpr6. MPK4, COI1 to activate resistance. The model does not Downstream of SA, NPR1 mediates the and JAR1 are required to transduce the JA show the complicated interactions between activation of expression of genes such as signal, while EIN2 is required to transduce SA and JA signaling. For example, MPK4 PR-1, acting together with TGA transcription the ET signal. Camalexin production requires inhibits SA signaling, possibly by promoting factors in a manner that is inhibited by SNI1. both PAD4 and COI1, and some lesion-mimic JA signaling, but the nature of the DTH9 also acts downstream of SA to activate mutations cause camalexin accumulation in relationship between the promotion of JA resistance responses. (c) JA and ET act lesions. (d) ISR activation requires JA, signaling and the inhibition of SA signaling is together to activate expression of genes such followed by ET. ISR1 affects sensitivity to ET. not understood.

as signal molecules. Different pathogens are limited to dif- to require EDS1 belong to the subset of the NBS-LRR ferent extents by SA-dependent responses and by class of R proteins with an amino-terminal domain similar JA/ET-dependent responses. There appears to be con- to the Drosophila Toll and mammalian interleukin 2 siderable cross-talk between these signal transduction receptors (TIR-NBS-LRR) [6]. Although the effect of pad4 networks, with at least some SA-dependent responses mutations on R-gene function has not been studied in great limited by JA/ET-dependent responses and vice versa. detail, it seems that R genes that require EDS1 also require PAD4 [7], an idea that is consistent with the extensive phe- R-gene signal transduction notypic similarities between pad4 and eds1 mutants Mutations in PBS1, NDR1, EDS1, PAD4, and PBS2 block [8,9,10•,11•]. Similarly, R genes that require NDR1 also gene-for-gene resistance that is mediated by some require PBS2, although some R genes with an absolute R genes. The only R gene known to be affected by PBS1 requirement for NDR1 are not completely blocked by mutations is RPS5 [5]. This observation raises the intriguing mutations in PBS2 and vice versa [5]. All of this evidence is possibility that PBS1 encodes the ‘guardee’ monitored by consistent with the hypothesis of Aarts et al. [6], who sug- RPS5. As RPS5 and the corresponding avr gene, avrPphB, gested that there may be two downstream pathways have both been isolated, it should be possible to test this triggered by R genes, with the structure of the R protein hypothesis in the near future when PBS1 has been isolated. determining which downstream factors are required.

R genes known to require NDR1 belong to the leucine- Other recent results have shown that the situation cannot be zipper (LZ) subclass of nucleotide-binding site this simple. Two R genes, RPP7 and RPP8, require neither (NBS)-leucine-rich repeat (LRR) genes. R proteins known EDS1 nor NDR1 (although resistance is weakly suppressed Genes controlling expression of defense responses in Arabidopsis — 2001 status Glazebrook 303

in an eds1 ndr1 double mutant), suggesting that there is at growth of virulent pathogens, and for SAR. SAR is a resis- least a third pathway for the transduction of R-gene signals tance response that is activated throughout the plant in [12••]. RPP13-mediated resistance does not require EDS1, response to local infection by necrotizing pathogens. As PAD4, PBS2, or NDR1, is completely unaffected in an described above, EDS1 and PAD4 act upstream of SA to eds1 ndr1 double mutant, and is not blocked by nahG [13•]. promote SA accumulation. The ankyrin-repeat protein RPP7 has not been cloned, but RPP8 and RPP13 are NPR1; also known as NIM1 (NON-INDUCIBLE IMMU- LZ-NBS-LRR proteins that are similar to each other [14,15]. NITY 1) and SAI1 (SALICYLIC ACID INSENSITIVE 1) Hence, these genes are exceptions to the rule that LZ-NBS- acts downstream of SA to promote the expression of the LRR proteins require NDR1. Taken together, the data SAR-associated genes PR-1, BGL2, and PR-5. support a model including at least three signal transduction cascades acting downstream of R genes, one that is Recently, additional genes that act in the SA signaling NDR1-dependent, a second that is EDS1-dependent, and a pathway have been identified, double-mutant analyses third for which no genetic components have yet been have placed some negative regulators in the pathway, and defined. This conclusion rests on the assumption that the NPR1-interacting factors have been identified. Mutations mutant ndr1, eds1, pad4, and pbs2 alleles that were tested are in EDS5 (also known as SID1) and SID2 (also known as all null. Otherwise, the observed differences in genetic EDS16 [20]) abolish the increase in SA levels observed in requirements could reflect quantitative differences in the infected plants. In fact, SA levels in these mutants are as signaling activities of the various alleles, rather than action of low as they are in plants carrying the nahG transgene, different signaling pathways. The commonly-used ndr1, which encodes an enzyme that degrades SA to catechol. pad4, and eds1 alleles are likely null, as they are deletions or Hence, EDS5 and SID2 are suggested to function nonsense mutations that destabilize the mRNAs. The nature upstream of SA accumulation [21]. A mutation in EDS4 of the pbs2 alleles will not be known until PBS2 is isolated. causes reduced SA accumulation and reduced sensitivity to SA, so EDS4 seems to function in SA signaling [22]. What determines which pathway is used by a particular Interestingly, although eds5, sid2, eds4, eds1 and pad4 mu- R gene is not clear. For example, the R gene HRT is more tations (as well as the transgene nahG) reduce SA levels and than 90% identical to RPP8 [16], yet HRT-mediated resis- expression of PR-1, only pad4 and nahG reduce production tance is blocked by the transgene nahG [17], which of the phytoalexin camalexin [21–23]. It appears, therefore, encodes an SA-degrading enzyme, whereas resistance that low SA production does not cause the camalexin defect mediated by RPP8 or RPP13 is unaffected by nahG in these plants, and that pad4 and nahG may affect signaling [12••,13•]. Unfortunately, the effects of mutations in pathways in addition to the SA-dependent pathway. NDR1, EDS1, PAD4, or PBS2 on HRT-mediated resistance were not tested [17]. In another example, the resistance Double-mutant analyses have been used to place some of mediated by the broad-spectrum R gene RPW8 requires the regulatory factors in the SA signaling network in order. EDS1 and is blocked by nahG. The protein encoded by A probable null mutation in EDR1 enhances resistance to RPW8 includes a coiled-coil domain but otherwise bears Pseudomonas syringae and Erisyphe cichoracearum, and causes little similarity to the LZ-NBS-LRR genes [18•]. Although more rapid expression of defense-related genes, such as the pathway used by a particular R gene may be deter- PR-1, after infection [24,25••]. The resistance is blocked mined by the structural features of the R gene, factors in by nahG and by eds1, pad4, or nim1 (allelic to npr1) mu- addition to whether the R gene is of the LZ-NBS-LRR or tations, indicating that the effect of the edr1 mutation the TIR-NBS-LRR type are clearly important. requires SA-dependent signaling [25••]. Presumably, the mitogen-activated protein kinase (MAPK) kinase kinase A few of the genes that mediate R-gene signal transduction ([MAPK]KK) encoded by EDR1 is involved in the nega- have been isolated. NDR1 encodes a protein with two pre- tive regulation of SA signaling, acting upstream of pad4 dicted transmembrane domains [19]. It is possible that part and eds1 [25••]. A mutation in the MAPK gene MPK4 of the function of NDR1 is to hold R proteins close to the causes constitutively high SA accumulation, PR-1 expression membrane. EDS1 and PAD4 encode proteins with simi- and resistance to P. syringae [26•]. These phenotypes are larity to triacyl glycerol lipases [8,9]. The significance of blocked by nahG but not by the npr1 mutation [26•]. The this is not known, but it is tempting to speculate that effects of other mutations were not tested. It does not EDS1 and PAD4 are involved in the synthesis or degra- seem likely that MPK4 is a component of the EDR1 kinase dation of a signal molecule. Mutations in either gene can cascade, as the phenotypes of mpk4 are considerably more cause a defect in SA accumulation and expression of both severe than those of edr1 and are NPR1-independent. genes can be induced by SA or pathogen infection, sug- Rather, MPK4 may impact SA signaling upstream of SA by gesting that they act in the SA pathway upstream of SA, affecting the balance between SA-dependent and jasmonic possibly as part of a signal amplification loop [8,9]. acid (JA)-dependent signaling, as discussed further below [26•]. The mutations cpr1 and cpr6 cause constitutively SA-dependent signaling high SA levels, PR-1 expression, and enhanced disease SA-dependent signaling is important for some gene-for- resistance. These phenotypes are blocked by mutations in gene resistance responses, for local responses that limit the eds5 [27••], eds1 [10•], or pad4 [11•], placing the effects of 304 Biotic interactions

the cpr mutations on SA signaling upstream of PAD4. mimic mutants that have been isolated over the years may Reminiscent of the situation with MPK4, however, the prove to be useful in devising genetic screens to identify disease resistance phenotypes of cpr1 and cpr6 are only par- components of the PAD4/EDS1-independent route to SA tially blocked by mutations in npr1, demonstrating that accumulation. It would also be interesting to survey lesion- there must be SA-dependent, NPR1-independent resis- mimic mutants to see if any of them activate SA tance mechanisms [27••]. In dnd1 pad4 or dnd2 pad4 accumulation in ways that are independent of EDS5 or double mutants, the high SA and PR-1 expression levels SID2. that are characteristic of dnd plants are retained but susceptibility to P. syringae is as high as it is in pad4 plants, NPR1 is required downstream of SA to activate the expres- suggesting that there is also a PAD4-dependent, SA-inde- sion of PR-1. Recent results suggest that NPR1 may pendent component of resistance [11•]. directly participate in the transcriptional control of defense-gene expression. NPR1 concentrates in the nucleus Many mutants with constitutively high SA levels, defense- in response to SA, and nuclear localization is required for gene expression, and disease resistance also display a the activation of PR-1 expression by NPR1 [33]. NPR1 lesion-mimic phenotype. That is, they undergo sponta- also interacts with TGA-type transcription factors in yeast neous cell death in the absence of pathogen attack. Some two-hybrid assays [34–36] and promotes the binding of of the genes defined by these mutations may function as TGA2 to its binding site [36]. As yet, no role for TGA negative regulators of hypersensitive cell death, whereas factors in activating defense-gene expression has been others may merely perturb cellular metabolism in a way detected, so the significance of the NPR1–TGA-factor that leads to cell death, which then triggers the SAR signal interaction is not certain. SA-inducible PR-1 expression is transduction cascade. Indeed, some transgenes that per- restored in npr1 sni1 plants [37]. SNI1 is also a nuclear- turb cellular metabolism, including the bacterio-opsin localized protein [37]. A reasonable model to explain these proton pump [28] and antisense protoporphyrinogen oxi- results is that the activation of PR-1 expression requires dase [29], cause a lesion-mimic phenotype. The identities that repression of PR-1 must be released by SNI1, allowing of recently isolated lesion-mimic genes are consistent with NPR1 to interact with transcription factors to activate gene the idea that many lesion-mimic mutations disrupt cellular expression [37]. metabolism. ACD2 encodes a protein that is similar to red chlorophyll catabolite reductase, an enzyme responsible The existence of SA-dependent, NPR1-independent resis- for the proper catabolism of the porphyrin component of tance mechanisms was suggested by the observation that chlorophyll [30]. Presumably, loss of this enzyme in acd2 responses that are blocked by nahG are not necessarily mutants results in the accumulation of toxic catabolites blocked by npr1. This logic is now somewhat questionable that trigger the lesions. This enzyme may also play a role as gene expression and camalexin accumulation are less in coping with pathogen damage as overexpression of severely affected in the low-SA mutants eds5 and sid2 than ACD2 attenuates disease symptoms [30]. DND1 encodes a in plants containing nahG [21,38]. Nevertheless, the exis- cyclic nucleotide-gated ion channel [31]. This channel tence of an SA-dependent, NPR1-independent resistance could be involved in the regulation of ion flows that are mechanism is still supported by the observation that eds5 related to programmed cell death. Alternatively, the loss of completely blocks the enhanced disease resistance of cpr1, this function could simply perturb cellular homeostasis in cpr5, and cpr6 mutants, whereas npr1 blocks it only partially a manner that promotes cell death. (Recently, it has been [27••]. The dth9 mutation may define a component of this recognized that both dnd1 and dnd2 mutants are lesion- pathway. Plants carrying the dth9 mutation express PR mimics under certain growth conditions [11•,32].) genes normally in response to SA treatment but they fail to develop resistance, demonstrating that DTH9 is required Double-mutant experiments involving some lesion-mimic for SAR but not for the gene expression changes known to mutants have revealed interesting features of SA-depen- be characteristic of SAR [39••]. The dth9 mutant could now dent signal transduction. In cpr5 mutants, SA level and be used as a tool to identify other factors that are required PR-1 expression and disease resistance are partially for SA-dependent resistance. reduced by eds1 and pad4 mutations, and are completely blocked by the eds5 mutation [10•,11•,27••]. This finding Jasmonic acid/ethylene-dependent signaling shows that activation of SA accumulation does not The role of JA and ET in the activation of disease resis- absolutely require EDS1 and PAD4, but that EDS5 may be tance mechanisms was demonstrated by the observation required. The SA and PR-1 expression phenotypes of dnd1 that expression of the PDF1.2 gene and other genes was and dnd2 mutants are unaffected by PAD4, confirming the prevented by mutations that block JA signaling (i.e. coi1) or observations that suggested that there is a PAD4-indepen- ethylene signaling (i.e. ein2). JA and ET seem to be dent pathway to SA accumulation [11•,23]. However, pad4 required simultaneously, as PDF1.2 expression is not acti- blocks the enhanced resistance phenotypes of dnd1 and vated by either ET treatment of coi1 plants or JA treatment dnd2, suggesting that there are resistance responses whose of ein2 plants [40]. Furthermore, resistance to the fungal expression requires a function of PAD4 other than its pathogen Alternaria brassicicola is compromised in coi1 activation of SA accumulation [11•]. The numerous lesion- plants, and pretreatment with JA enhances resistance to Genes controlling expression of defense responses in Arabidopsis — 2001 status Glazebrook 305

this pathogen [41], whereas ein2 plants display enhanced former explanation is supported by the observation that susceptibility to the fungal pathogen Botrytis cinerea [42]. the activation of PDF1.2 expression is also blocked in nahG Mutations in OPR3 (12-OXOPHYTODIENOIC ACID mpk4 double mutants [26•]. Consistent with the idea that REDUCTASE 3), a gene encoding an enzyme that is JA and SA may sometimes act together rather than in required for JA synthesis, were identified as a consequence opposition to each other, some genes can be activated by of their male-sterility phenotypes [43,44]. The opr3 either JA or SA treatment [50]. mutants could be used to determine the effect of blocking JA synthesis on defense responses. The coi1 mutation is Induced systemic resistance commonly considered to reflect the loss of JA signaling, Colonization of roots by certain rhizosphere bacteria con- but COI1 encodes an F-box protein that presumably func- fers a form of disease resistance called induced systemic tions in targeting proteins for ubiquitin-dependent resistance (ISR). ISR occurs in nahG plants, so it is not an degradation [45]. It is possible, therefore, that COI1 has SA-dependent phenomenon. No significant changes in pleiotropic effects beyond its effect on JA signaling. plant gene expression have been associated with ISR. Nonetheless, progress has been made in the genetic dis- JA signaling is also involved in control of the formation of section of this phenomenon. Several years ago, it was lesions that are induced by ozone treatment. This obser- found that ISR requires NPR1. This is interesting as it vation is interesting in the context of disease resistance implies that NPR1 is involved in perception of more than signaling because ozone-induced cell death may be simi- one signal (i.e. SA and the ISR signal), and may respond lar to hypersensitive cell death. The rcd1 (radical-induced in one of two ways (by activating PR-1 expression or cell death1) mutant develops spreading lesions after ozone by activating ISR). If NPR1 acts as an adaptor molecule treatments that do not induce lesions in wild-type plants that brings the components of signaling complexes together, [46]. In this mutant, the HR lesions induced by infection this sort of dual role might be expected. ISR is blocked with a bacterial pathogen that triggers gene-for-gene by mutations that interfere with either ET or JA signaling. resistance are larger than those in wild-type plants [46]. As ethylene can induce ISR in jar1 mutants, it is thought The ozone-triggered cell death in rcd1 plants is inhibited that the requirement for JA lies upstream of the require- by JA and promoted by ethylene. Thus, JA and ET act in ment for ET in ISR signaling (for review, see [51]). opposition to each other during the formation of ozone- induced lesions, whereas they both promote PDF1.2 Some ecotypes fail to develop ISR, a trait that maps to a expression [46]. In a related study, nahG plants showed single locus dubbed ISR1 [52]. The dominant ISR1 increased tolerance to ozone, JA treatment decreased SA allele confers the ability to develop ISR [52]. levels and increased ozone tolerance, and mutants Interestingly, the ISR trait cosegregates with high basal blocked in JA signaling (i.e. jar1) or synthesis (i.e. the resistance to a virulent strain of P. syringae, suggesting fatty-acid deficient 3 [fad3] fad7 fad8 triple mutant) showed that ISR and general resistance responses are intimately decreased tolerance [47]. related [52]. ISR1 also cosegregates with sensitivity to ethylene, with the ISR1 allele correlating with increased There are many other examples of cross talk between SA sensitivity to ethylene [53]. Simultaneous induction of and JA/ET signaling. Many lesion-mimic mutants, including SAR and ISR resulted in enhanced resistance relative to cpr5, cpr6, acd2, dnd1, dnd2, and ssi1 (suppressor of SA insen- the induction of SAR alone, suggesting that ISR is an sitivity 1), display constitutively high expression of both example of a JA-dependent response that is not inhibited PR-1 and PDF1.2, suggesting that the SA and JA signaling by SA [54]. pathways may share common activating signals. Expression of the JA-dependent gene PDF1.2 is, however, Genes not yet placed into pathways strongly inhibited by SA, as demonstrated by observations Several interesting mutants have been described that including increased rose-bengal-induced PDF1.2 expres- probably affect disease resistance signaling but that are not sion in pad4 mutants [22], increased cpr6-dependent yet characterized in sufficient detail to allow them to be PDF1.2 expression in cpr6 eds5, cpr6 pad4, and cpr6 eds1 placed in particular positions in the network. Two genes, double mutants [10•,11•,27••], enhanced PDF1.2 expres- HXC2 (Hypersensitivity to Xanthomonas campestris pv. sion in sid2 plants infected with Erisyphe orontii [20], and campestris 2) and RXC5 (Resistance to Xanthomonas enhanced PDF1.2 expression in npr1 or nahG plants treated campestris 5), are required for resistance to a particular iso- with rose bengal or infected with Alternaria brassicicola late of Xanthomonas campestris, implying that they may be [48,49]. JA can also repress the expression of SA-induced part of a gene-for-gene resistance mechanism [18•]. FLS2 genes. For example, treatment with JA repressed the (FLAGELLIN SENSITIVE 2) encodes a receptor-like ozone-induced accumulation of SA and expression of PR-1 kinase that is required for responses to bacterial flagellin, [47]. The mpk4 mutation blocks the JA-inducible expres- but the effects of fls2 mutations on resistance to bacterial sion of PDF1.2 and causes the constitutive activation of pathogens have not yet been reported [55,56]. Many other SA-dependent signaling, suggesting either that the block mutants with enhanced disease resistance [57–60] or in JA signaling relieves the suppression of SA signaling or enhanced susceptibility to various pathogens [5,20,61] that the activation of SA signaling blocks JA signaling. The have also been described recently. 306 Biotic interactions

Conclusions and future prospects 7. Glazebrook J, Zook M, Mert F, Kagan I, Rogers EE, Crute IR, Holub EB, Hammerschmidt R, Ausubel FM: Phytoalexin-deficient A major area of progress in the past year has been in eluci- mutants of Arabidopsis reveal that PAD4 encodes a regulatory dating the relative sites of action of signaling components factor and that four PAD genes contribute to downy mildew resistance. Genetics 1997, 146:381-392. in signal transduction cascades. On the basis of the different 8. Falk A, Feys BJ, Frost LN, Jones JDG, Daniels MJ, Parker JE: EDS1, requirements of various R genes for other genes to medi- an essential component of R gene-mediated disease resistance ate resistance, R-gene-dependent signaling has been in Arabidopsis has homology to eukaryotic lipases. Proc Natl Acad resolved into at least three different pathways. New genes Sci USA 1999, 96:3292-3297. have been defined in the SA signaling pathway, and 9. Jirage D, Tootle TL, Reuber TL, Frost LN, Feys BJ, Parker JE, Ausubel FM, Glazebrook J: Arabidopsis thaliana PAD4 encodes a lipase-like constitutive mutants have been useful in ordering compo- gene that is important for salicylic acid signaling. Proc Natl Acad nents of the pathway. Sci USA 1999, 96:13583-13588. 10. Clarke JD, Aarts N, Feys BJ, Dong X, Parker JE: Induced disease There is still much work to be done to define genes that • resistance requires EDS1 in the Arabidopsis mutants cpr1 and cpr6 but is only partially EDS1-dependent in cpr5. Plant J 2001, are required for resistance. Although progress has been 26:409-420. made in defining the signal transduction cascades that con- The double-mutant analysis reported in this paper placed CPR1 and cpr6 trol the expression of certain defense-related genes, there upstream of EDS1 in SA-dependent signaling. is no clear correlation between gene expression patterns •11. Jirage D, Zhou N, Cooper B, Clarke JD, Dong X, Glazebrook J: Constitutive salicylic acid-dependent signaling in cpr1 and cpr6 and disease resistance (see [11•,27••] for examples). In the mutants requires PAD4. Plant J 2001, 26:395-407. future, the application of genome-wide expression profiling This double-mutant analysis placed CPR1 and cpr6 upstream of PAD4 in SA-dependent signaling. to various mutants with defects in disease resistance 12. McDowell JM, Cuzick A, Can C, Beynon J, Dangl JL, Holub EB: signaling should greatly facilitate both the elucidation of •• Downy mildew (Peronospora parasitica) resistance genes in signaling networks and the identification of genes with Arabidopsis vary in functional requirements for NDR1, EDS1, roles in the regulation and execution of disease resistance NPR1, and salicylic acid accumulation. Plant J 2000, 22:523-529. The R genes RPP7 and RPP8 were shown to function independently of responses. The work of Maleck et al. [4••] demonstrates NDR1, EDS1, and NPR1, thus defining a possible third pathway for R-gene the power of this sort of analysis. In the next year or two, function. many more reports of genome-wide expression analysis 13. Bittner-Eddy PD, Beynon JL: The Arabidopsis downy mildew • resistance gene, RPP13-Nd, functions independently of NDR1 can be expected, as well as the isolation of many of the and EDS1 and does not require the accumulation of salicylic acid. genes that correspond to the interesting mutations Mol Plant Microbe Interact 2001, 14:416-421. The R gene RPP13 was found to act independently of EDS1 and NDR1, described in this review. confirming the existence of at least three alternative pathways for R-gene function. Acknowledgements 14. McDowell JM, Dhandaydham M, Long TA, Aarts MG, Goff S, I apologize to scientists whose work I overlooked or was not able to include Holub EB, Dangl JL: Intragenic recombination and diversifying because of length limitations. selection contribute to the evolution of downy mildew resistance at the RPP8 locus of Arabidopsis. Plant Cell 1998, 10:1861-1874. 15. Bittner-Eddy PD, Crute IR, Holub EB, Beynon JL: RPP13 is a simple References and recommended reading locus in Arabidopsis thaliana for alleles that specify downy Papers of particular interest, published within the annual period of review, mildew resistance to different avirulence determinants in have been highlighted as: Personospora parasitica. Plant J 2000, 21:177-188. • of special interest 16. Cooley MB, Pathirana S, Wu HJ, Kachroo P, Klessig DF: Members of •• of outstanding interest the Arabidopsis HRT/RPP8 family of resistance genes confer resistance to both viral and oomycete pathogens. Plant Cell 2000, 1. Glazebrook J: Genes controlling expression of defense responses 12:663-676. in Arabidopsis. Curr Opin Plant Biol 1999, 2:280-286. 17. Kachroo P, Yoshioka K, Shah J, Dooner HK, Klessig DF: Resistance 2. van der Biezen E, Jones JDG: Plant disease-resistance proteins to turnip crinkle virus in Arabidopsis is regulated by two host and the gene-for-gene concept. Trends Biochem Sci 1998, genes and salicylic acid dependent but NPR1, ethylene, and 23:454-456. jasmonate independent. Plant Cell 2000, 12:677-690. 3. Leister RT, Katagiri F: A resistance gene product of the nucleotide 18. Xiao S, Ellwood S, Calis O, Patrick E, Li T, Coleman M, Turner JG: binding site-leucine rich repeats class can form a complex with • Broad-spectrum mildew resistance in Arabidopsis thaliana bacterial avirulence proteins in vivo. Plant J 2000, 22:345-354. mediated by RPW8. Science 2001, 291:118-120. 4. Maleck K, Levine A, Eulgem T, Morgan A, Schmid J, Lawton KA, The isolation of the broad-spectrum R gene RPW8 is reported. RPW8 is •• Dangl JL, Dietrich RA: The transcriptome of Arabidopsis thaliana structurally very different from known R genes, yet it requires EDS1to function. during systemic acquired resistance. Nat Genet 2000, 26:403-410. 19. Century KS, Shapiro AD, Repetti PP, Dahlbeck D, Holub E, Expression profiling was used to compare the patterns of gene expression Staskawicz BJ: NDR1, a pathogen-induced component required that are associated with various local and systemic resistance responses. for Arabidopsis disease resistance. Science 1997, 278:1963-1965. These responses were grouped on the basis of profile similarities, and induced genes were grouped on the basis of expression-pattern similarities. 20. Dewdney J, Reuber TL, Wildermuth MC, Devoto A, Cui J, Stutius LM, A promoter motif that may confer coordinate gene expression was identified. Drummond EP, Ausubel FM: Three unique mutants of Arabidopsis This work sets the precedent for the use of expression profiling for studies identify eds loci required for limiting growth of a biotrophic fungal of signaling networks. pathogen. Plant J 2000, 24:205-218. 5. Warren RF, Merritt PM, Holub E, Innes RW: Identification of three 21. Nawrath C, Metraux JP: Salicylic acid induction-deficient mutants putative signal transduction genes involved in R gene-specified of Arabidopsis express PR-2 and PR-5 and accumulate high resistance in Arabidopsis. Genetics 1999, 152:401-412. levels of camalexin. Plant Cell 1999, 11:1393-1404. 6. Aarts N, Metz M, Holub E, Staskawicz BJ, Daniels MJ, Parker JE: 22. Gupta V, Willits MG, Glazebrook J: Arabidopsis thaliana EDS4 Different requirements for EDS1 and NDR1 by disease resistance contributes to salicylic acid (SA)-dependent expression of genes define at least two R gene-mediated signaling pathways in defense responses: evidence for inhibition of jasmonic acid Arabidopsis. Proc Natl Acad Sci USA 1998, 95:10306-10311. signaling by SA. Mol Plant Microbe Interact 2000, 13:503-511. Genes controlling expression of defense responses in Arabidopsis — 2001 status Glazebrook 307

23. Zhou N, Tootle TL, Tsui F, Klessig DF, Glazebrook J: PAD4 functions required for the expression of known SAR-related genes. Thus, it appears to upstream from salicylic acid to control defense responses in define a new branch in the signal transduction network. Arabidopsis. Plant Cell 1998, 10:1021-1030. 40. Penninckx IAMA, Thomma BPHJ, Buchala A, Metraux JP, Broekaert WF: 24. Frye CA, Innes RW: An Arabidopsis mutant with enhanced Concomitant activation of jasmonate and ethylene response resistance to powdery mildew. Plant Cell 1998, 10:947-956. pathways is required for induction of a plant defensin gene in Arabidopsis. Plant Cell 1998, 10:2103-2113. 25. Frye CA, Tang D, Innes RW: Negative regulation of defense •• responses in plants by a conserved MAPKK kinase. Proc Natl 41. Thomma BPHJ, Eggermont K, Penninckx IAMA, Mauch-Mani B, Acad Sci USA 2000, 98:373-378. Vogelsang R, Cammue BPA, Broekaert WF: Separate jasmonate- EDR1 is isolated and found to encode a MAPKKK. Double-mutant analysis dependent and salicylate-dependent defense-response pathways revealed that EDR1 acts upstream of EDS1 to suppress the activation of SA in Arabidopsis are essential for resistance to distinct microbial signaling. pathogens. Proc Natl Acad Sci USA 1998, 95:15107-15111. 26. Petersen M, Brodersen P, Naested H, Andreasson E, Lindhart U, 42. Thomma BPHJ, Eggermont K, Tierens KFMJ, Broekaert WF: • Johansen B, Nielsen HB, Lacy M, Austin MJ, Parker JE et al.: Requirement of functional ethylene-insensitive 2 gene for efficient Arabidopsis MAP kinase 4 negatively regulates systemic acquired resistance of Arabidopsis to infection by Botrytis cinerea. Plant resistance. Cell 2000, 103:1111-1120. Physiol 1999, 121:1093-1101. A mutation in MPK4 was found to block JA signaling and activate SA signaling. These results demonstrate a role for a MAP kinase in defense signaling, and 43. Sanders PM, Lee PY, Biesgen C, Boone JD, Beals TP, Weiler EW, illustrate cross-talk between JA-dependent and SA-dependent signaling path- Goldberg RB: The Arabidopsis delayed dehiscence 1 gene ways. encodes an enzyme in the jasmonic acid synthesis pathway. Plant Cell 2000, 12:1041-1061. 27. Clarke JD, Volko SM, Ledford H, Ausubel FM, Dong X: Roles of •• salicylic acid, jasmonic acid, and ethylene in cpr-induced 44. Stintzi A, Browse J: The Arabidopsis male-sterile mutant, opr3, resistance in Arabidopsis. Plant Cell 2000, 12:2175-2190. lacks the 12-oxophytodienoic acid reductase required for A detailed analysis of double and triple mutants reveals that CPR1 and cpr6 jasmonate synthesis. Proc Natl Acad Sci USA 2000, act upstream of EDS5, and detects complex interactions among SA, ET, and 97:10625-10630. JA signaling that affect disease resistance. 45. Xie DX, Feys BF, James S, Nieto-Rostro M, Turner JG: COI1: an 28. Mittler R, Shulaev V, Lam E: Coordinated activation of programmed Arabidopsis gene required for jasmonate-regulated defense and cell death and defense mechanisms in transgenic tobacco plants fertility. Science 1998, 280:1091-1094. expressing a bacterial proton pump. Plant Cell 1995, 7:29-42. 46. Overmyer K, Tuominen H, Kettunen R, Betz C, Langebartels C, 29. Molina A, Volrath S, Guyer D, Maleck K, Ryals J, Ward E: Inhibition of Sandermann H, Kangasjarvi J: Ozone-sensitive Arabidopsis rcd1 protoporphyrinogen oxidase expression in Arabidopsis causes a mutant reveals opposite roles for ethylene and jasmonate lesion-mimic phenotype that induces systemic acquired signaling pathways in regulating superoxide-dependent cell resistance. Plant J 1999, 17:667-678. death. Plant Cell 2000, 12:1849-1862. 30. Mach JM, Castillo AR, Hoogstraten R, Greenberg JT: The 47. Rao MV, Lee HI, Creelman RA, Mullet JE, Davis KR: Jasmonic acid Arabidopsis accelerated cell death gene ACD2 encodes red signaling modulates ozone-induced hypersensitive cell death. chlorophyll catabolite reductase and suppresses the spread of Plant Cell 2000, 12:1633-1646. disease symptoms. Proc Natl Acad Sci USA 2001, 98:771-776. 48. Penninckx IAMA, Eggermont K, Terras FFG, Thomma BPHJ, 31. Clough SJ, Fengler KA, Yu IC, Lippok B, Smith RK, Bent AF: The De Samblancx GW, Buchala A, Metraux JP, Manners JM, Broekaert WF: Arabidopsis dnd1 ‘defense, no death’ gene encodes a mutated Pathogen-induced systemic activation of a plant defensin gene in cyclic nucleotide-gated ion channel. Proc Natl Acad Sci USA 2000, Arabidopsis follows a salicylic acid-independent pathway. Plant 97:9323-9328. Cell 1996, 8:2309-2323. 32. Yu IC, Fengler KA, Clough SJ, Bent AF: Identification of Arabidopsis 49. Clarke JD, Liu Y, Klessig DF, Dong X: Uncoupling PR gene mutants exhibiting an altered hypersensitive response in gene- expression from NPR1 and bacterial resistance: characterization for-gene resistance. Mol Plant Microbe Interact 2000, 13:277-286. of the dominant Arabidopsis cpr6-1 mutant. Plant Cell 1998, 10:557-569. 33. Kinkema M, Fan W, Dong X: Nuclear localization of NPR1 is required for activation of PR gene expression. Plant Cell 2000, 50. Schenk PM, Kazan K, Wilson I, Anderson JP, Richmond T, 12:2339-2350. Somerville SC, Manners JM: Coordinated plant defense responses in Arabidopsis revealed by microarray analysis. Proc Natl Acad Sci 34. Zhang Y, Fan W, Kinkema M, Li X, Dong X: Interaction of NPR1 with USA 2000, 97:11655-11660. basic leucine zipper protein transcription factors that bind sequences required for salicylic acid induction of the PR-1 gene. 51. Pieterse CMJ, van Pelt JA, van Wees SCM, Ton J, Leon-Kloosterziel KM, Proc Natl Acad Sci USA 1999, 96:6523-6528. Keurentjes JJB, Verhagen BWM, Knoester M, van der Sluis I, Bakker PAHM et al.: Rhizobacteria-mediated induced systemic 35. Zhou JM, Trifa Y, Silva H, Pontier D, Lam E, Shah J, Klessig DF: NPR1 resistance: triggering, signalling and expression. Eur J Plant Pathol differentially interacts with members of the TGA/OBF family of 2001, 107:51-61. transcription factors that bind an element of the PR-1 gene required for induction by salicylic acid. Mol Plant Microbe Interact 52. Ton J, Pieterse CMJ, Van Loon LC: Identification of a locus in 2000, 13:191-202. Arabidopsis controlling both the expression of rhizobacteria- mediated induced systemic resistance (ISR) and basal resistance 36. Després C, DeLong C, Glaze S, Liu E, Fobert PR: The Arabidopsis against Pseudomonas syringae pv. tomato. Mol Plant Microbe NPR1/NIM1 protein enhances the DNA binding activity of a Interact 1999, 12:911-918. subgroup of the TGA family of bZIP transcription factors. Plant Cell 2000, 12:279-290. 53. Ton J, Davison S, van Wees SCM, van Loon LC, Pieterse CMJ: The Arabidopsis ISR1 locus controlling rhizobacteria-mediated 37. Li X, Zhang Y, Clarke JD, Li Y, Dong X: Identification and cloning of induced systemic resistance is involved in ethylene signaling. a negative regulator of systemic acquired resistance, SNI1, Plant Physiol 2001, 125:652-661. through a screen for suppressors of NPR1-1. Cell 1999, 98:329-339. 54. van Wees SCM, de Swart EAM, van Pelt JA, van Loon LC, Pieterse CMJ: Enhancement of induced resistance by simultaneous activation of 38. Rogers EE, Ausubel FM: Arabidopsis enhanced disease salicylate- and jasmonate-dependent defense pathways in susceptibility mutants exhibit enhanced susceptibility to several Arabidopsis thaliana. Proc Natl Acad Sci USA 2000, 97:8711- bacterial pathogens and alterations in PR-1 gene expression. 8716. Plant Cell 1997, 9:305-316. 55. Gomez-Gomez L, Felix G, Boller T: A single locus determines 39. Mayda E, Mauch-Mani B, Vera P: Arabidopsis dth9 mutation sensitivity to bacterial flagellin in Arabidopsis thaliana. Plant J 1999, •• identifies a gene involved in regulating disease susceptibility 18:277-284. without affecting salicylic acid-dependent responses. Plant Cell 2000, 12:2119-2128. 56. Gomez-Gomez L, Boller T: FLS2: an LRR receptor-like kinase The dth9 mutant was isolated and characterized. The phenotypes of dth9 involved in the perception of the bacterial elicitor flagellin in suggest that DTH9 is required downstream of SA to activate SAR, but is not Arabidopsis. Mol Cell 2000, 5:1003-1011. 308 Biotic interactions

57. Silva H, Yoshioka K, Dooner HK, Klessig DF: Characterization of a B1-resistant (fbr) Arabidopsis mutants. Plant Cell 2000, new Arabidopsis mutant exhibiting enhanced disease resistance. 12:1811-1822. Mol Plant Microbe Interact 1999, 12:1053-1063. 60. Vogel J, Somerville S: Isolation and characterization of powdery 58. Rate DN, Cuenca JV, Bowman GR, Guttman DS, Greenberg JT: The mildew-resistant Arabidopsis mutants. Proc Natl Acad Sci USA gain-of-function Arabidopsis acd6 mutant reveals novel regulation 2000, 97:1897-1902. and function of the salicylic acid signaling pathway in controlling cell death, defenses, and cell growth. Plant Cell 1999, 11:1695-1708. 61. Greenberg JT, Silverman FP, Liang H: Uncoupling salicylic acid- dependent cell death and defense-related responses from 59. Stone JM, Heard J, Asai T, Ausubel FM: Stimulation of fungal- disease resistance in the Arabidopsis mutant acd5. Genetics mediated cell death by fumonisin B1 and selection of fumonisin 2000, 156:341-350.