Introduction. O body swayed to music, O brightening glance, How can we know the dancer from the dance?

288 Knowing the dancer from the dance: R-gene products and their interactions with other proteins from host and pathogen Zachary Nimchuk, Laurence Rohmer, Jeff H Chang and Jeffery L Dangl* Cloning of plant disease resistance genes is now commonplace in model plants. Recent attention has turned to how the proteins that they encode function biochemically to recognize their cognate Avirulence protein and to initiate the disease-resistance response. In addition, attention has turned to how the Avirulence proteins of pathogens might alter susceptible hosts for the benefit of the pathogen, and what plant proteins might be required for that process. Addresses Department of Biology and Curriculum in Genetics (JLD), Coker Hall, Room 108, CB#3280, University of North Carolina, Chapel Hill, North Carolina 27599-3280, USA *e-mail: dangl@email.unc.edu Correspondence: Jeffery L Dangl Current Opinion in Plant Biology 2001, 4:288 294 1369-5266/01/$ see front matter 2001 Elsevier Science Ltd. All rights reserved. Abbreviations Avr avirulence CC coiled coil CP coat protein HR hypersensitive response LRR leucine-rich repeat NBS nucleotide-binding site pv. pathovar PVY Potato Virus Y R resistance TCV turnip crinkle virus TIR Toll-Interleukin receptor Introduction O body swayed to music, O brightening glance, How can we know the dancer from the dance? From Among Schoolchildren by William Butler Yeats, 1928 The majority of potential pathogens are stopped before they infect the majority simply cannot start their lifecycles on most plants. When the pathogen is, in principle, capable of initiating infection, plants have at least two sets of genetically defined overlapping defense modes that they can deploy. These responses further reduce the range of plants that pathogens can infect. The first is a basal defense that limits the growth of virulent pathogens (reviewed in [1] and by Glazebrook, pp 301 308, and Dong, pp 309 314). The second is mediated by the now familiar plant disease resistance (R) genes (reviewed in [2] and by Jones, pp 281 287). R proteins determine the recognition of a specific molecule produced by pathogens. These elicitors of the resistance response are called avirulence (Avr) proteins because their recognition by the corresponding R proteins of the host results in the activation of a suite of defense responses. R-dependent responses are often, but not always, coupled with rapid programmed cell death of the host, termed the hypersensitive response (HR). If either R or avr genes are missing then recognition does not occur and disease ensues. This gene-for-gene recognition governs plant resistance against diverse classes of pathogens, including bacteria, fungi, viruses, nematodes, and insects. A set of structurally similar R proteins determines the recognition of a diversity of Avr proteins. The vast majority of R genes encode proteins containing a nucleotide-binding site (NBS) and leucine-rich repeats (LRRs). In animal systems, LRRs mediate protein protein interactions in ligand receptor models [3]. The NBS-LRR R proteins can be further subdivided into two major classes on the basis of their amino-terminal sequences, featuring either a coiled coil (CC) or a Toll-Interleukin receptor (TIR) homology domain. There are around 150 NBS-LRR genes in the Arabidopsis genome [4]. In contrast, Avr proteins show little or no homology to one another. Furthermore, most Avr proteins, with the exception of some virally encoded proteins, have no deduced or experimentally defined functions. The simplest model for R Avr protein interactions is that R proteins are receptors and Avr proteins are ligands. Intuitively, one would expect the selective pressure exerted by frequently occurring R genes to cause pathogens to jettison their avr genes. Thus, the maintenance of avr genes in pathogen populations suggests that these genes play a role in pathogen virulence on susceptible hosts, a notion supported experimentally in several cases (reviewed in [5]). This is most easily understood for the viral-encoded proteins that are recognized by host R proteins, but are also required by the virus for replication, processing or envelopment. We presume that some Avr proteins contribute to virulence of the pathogen through direct interaction with host proteins. Avr proteins could inhibit the host proteins required for establishing basal or specific defenses. On the basis of these concepts, one function of R proteins might be to guard virulence targets and to intercept incoming Avr proteins at, or close to, their cognate virulence target [6]. In this (by no means exhaustive) review, we focus on the current status of the R Avr pas de deux. In particular, we examine how this dance might reflect a cutting in by R proteins on a more sophisticated dance between Avr proteins and their actual partners of choice, their virulence targets. Circling the dance floor: right time, right place One prediction of the guard hypothesis is that the cellular location of the Avr protein will determine the location of

R-gene products and their interactions with other proteins from host and pathogen Nimchuk et al. 289 Table 1 Summary of Avirulence protein properties. Avr Protein Pathogen Homology with and/or Location in the Common domain Location of the possible virulence function* plant cell* for avirulence and matching R gene* virulence function AvrPphC Pseudomonas syringae Interferes with the plant NR NR NR pv. phaseolicola defense response [36 ] AvrPphF Pseudomonas syringae Interferes with the plant NR NR NR pv. phaseolicola defense response [36 ] AvrRpt2 Pseudomonas syringae Interferes with the plant NR NR NR pv. tomato defense response [37 ] AvrRpm1 Pseudomonas syringae Enhances growth [55] Plasma NR Plasma pv. maculicola membrane [14] membrane [15] AvrPto Pseudomonas syringae Enhances growth [31] Plasma No [30] NR pv. tomato membrane [16] AvrBst Xanthomonas campestris Ubiquitin-like protease [26 ] NR Yes [26 ] NR pv. campestris AvrXa7 Xanthomonas oryzae Transcriptional activation domain Nucleus [23 ] Yes [23 ] NR pv. oryzae required for virulence [24 ] AvrBs2 Xanthomonas campestris Agrocinopine synthase (from NR Recognition region NR pv. vesicatoria Agrobacterium tumefaciens) [51] localized in the portion with highest homology to synthase [52] AvrBs3 Xanthomonas campestris Transcriptional activation domain Nucleus [19] Yes [21] NR pv. glycinea required for virulence [19] PthA Xanthomonas citri Transcriptional activation Nucleus [53] NR NR domain required for virulence/causes cell hyperplasia [53] VirPphA Pseudomonas syringae Interferes with the plant NR NR NR pv. phaseolicola defense response [35] AvrPitA Magnaporthe grisea Protease [40 ] NR Yes NR Avr9 Cladosporium fulvum NR Exoplasmic [56] NR External plasma membrane [56] Nla Potato Virus Y Proteinase NR Yes [29] NR Coat protein Turnip Crinkle Virus coat protein NR No [54] NR This table summarizes what is known about the location, virulence and avirulence function of the effectors treated in this review, and shows how interdependent these elements are for the dual function of these proteins. *On the basis of sequence analysis and/or experimental data. On the basis of experimental data. NR = none reported. not only the R protein, but also the virulence target of the Avr protein (see Table 1). During the infection of tomato by the fungus Cladosporium fulvum, Avr proteins are secreted into the extracellular space. These Avr proteins are recognized by members of the Cf-X class of R proteins. Cf-X proteins possess a putative single-pass transmembrane domain, extracellular LRRs and a short cytoplasmic region with no known homologies. Studies of Cf-9 confirm that it is localized to the plasma membrane [7]. Hence, the cellular location of Cf-9 is consistent with both the cellular location of C. fulvum Avr proteins, and with the extracellular, noninvasive life-style of C. fulvum. Cf-9 protein was also localized to the endoplasmic reticulum [8]. Plants expressing Cf-9 carrying amino alterations in potential endoplasmic reticulum (ER)-localization sites are, however, still resistant [9]. Additional extracellular C. fulvum proteins have been identified and shown by reverse genetics to function as virulence factors. Importantly, tomato genotypes have also been identified that recognize these proteins as Avr proteins [10,11]. Taken together, these data indicate that R proteins can be localized to similar cellular locations as their putative ligands and predict the location of possible virulence targets. Many bacterial pathogens live in the intercellular spaces of their host. Yet, a growing body of data suggests that bacterial Avr proteins are translocated directly into the host cell via the Type-III secretion system. This system is evolutionarily conserved among bacterial pathogens of plants and animals (reviewed in [5,12,13]). We term the putative translocated proteins Type-III effector proteins. The characterization of several Type-III effector proteins suggests that they are localized to specific subcellular compartments in the host cell. The amino-acid sequences of several Pseudomonas syringae Type-III effector proteins contain potential amino-terminal myristoylation and palmitoylation sites, which are normally found only in eukaryotic proteins [14]. Myristoyl and palmitoyl groups can act as tethers, attaching the fatty-acid-modified proteins to membranes. We have demonstrated that myristoylation sequences in the P. syringae AvrRpm1, AvrB, and

290 Biotic interactions AvrPphB proteins are required for their localization to the plasma membrane of host cells. Furthermore, these sequences are required for bacteria expressing AvrRpm1 and AvrB to elicit a resistant response by plants expressing the cognate RPM1 protein, a plasma-membrane-associated protein [15]. Even more intriguing is that both the myristoylation and palmitoylation sites of AvrRpm1 are required for bacteria to be fully virulent on susceptible hosts [14]. This indicates that the putative virulence target of AvrRpm1 may also be associated with the plasma membrane. AvrPto was also shown to require N-myristoylation sites to localize to the host plasma membrane and trigger Pto function [16]. Pto, the cognate R protein to AvrPto, also carries putative N-myristoylation sequences, indicating that Pto may be associated with the plasma membrane. Localization of Pto to the membrane has yet to be demonstrated, however, and Pto can function without the consensus myristoylation site if driven from a strong promoter [17]. Additionally, a putative N-myristoylation site has been identified in the amino-acid sequence of RPS5, a NBS-LRR protein of Arabidopsis that recognizes the bacterial Avr protein AvrPphB [18]. Recognition of P. syringae AvrPphB has been shown to specifically require RPS5 and PBS1 [19]. PBS1 has been cloned and encodes a serine-threonine kinase that appears to be functionally distinct from Pto [20]. Interestingly, like AvrPphB and RSP5, PBS1 also contains a putative N-myristoylation site, potentially placing all three of these players at the membrane. Which one is my partner? AvrBs3 family proteins of various Xanthomonads require nuclear-localization sequences for transport to the nucleus, and for virulence and R-dependent recognition [21,22,23 ]. Thus, the corresponding R proteins might also localize to the nucleus. The exact role of these Avr proteins in the nucleus is still unclear, but this family may be involved in inducing or repressing the expression of host genes. A transcriptional activation domain of AvrXa7 is required for avirulence and virulence on rice [23 ]. When this domain is substituted with that of the unrelated herpes simplex viral protein16 (VP16), avirulence but not virulence activity is restored. This result suggests that the recognition by Xa7 mediated by the transcriptional activation domain of AvrXa7 is not sequence-specific, but rather is due to some interaction with the transcriptional machinery of the host. The AvrXa7 activation domain is, however, required for virulence, perhaps because it interacts with a host factor to modify the expression of host target genes during disease. The AvrBs3 family of proteins from Xathomonas spp. requires the presence of NLS sequences for recognition on resistant hosts. Recent work on family member AvrBs4, however, demonstrates that its NLS sequences are dispensable for recognition by Bs4 on tomato [24]. This work suggests that recognition (interception?) by R proteins can occur at different points along intracellular paths taken by translocated type-iii effectors to their cellular target. Cloning and analysis of R proteins that recognize distinct AvrBs3 family members should shed light on this puzzle. The AvrBs3 proteins may be the first known case of a family of effectors having undergone diversification aimed at a set of related host targets. Recently, several closely related members of this family (members of the AvrXa-X proteins from X. oryzae) were mutagenized in one strain [25]. Each mutation caused a different (and ultimately additive) impact on virulence. Moreover, none of the avrxa-x genes tested could complement the loss of virulence function normally encoded by other family members. Thus, their functions are non-redundant. One possible scenario is that these genes are evolving to not only evade host R functions, but also to maximize the number of interactions they can have with individual members of a host protein family that are the virulence targets. The more of these related targets that can be affected the more efficient the pathogen will become. Just holding hands or really a close dance? The most effective R protein would be one that recognizes the domains of its cognate Type-III effector that are required for virulence. In this case, effector mutations that result in the loss of or decreased resistance responses would presumably carry a fitness cost to the pathogen. Recent mutational analyses of two homologous Type-III effectors indicate that such effective R proteins might exist. The X. campestris Type-III effector AvrBsT is related to Yersinia pestis YopJ protein. The regions of homology include conserved sequences that encode a putative protease catalytic domain. Directed mutations of YopJ in one of the three conserved residues required for function in related proteases resulted in the loss of Yersinia pestis virulence on its normally susceptible host. The same mutations in avrbst resulted in the loss of recognition by the normally resistant Nicotiania benthamiana plants [26 ]. Similarly, X. campestris pathovar (pv.) vesicatoria mutants were recently isolated that can partially overcome the pepper Bs2 gene and also have reduced avrbs2 virulence activity on bs2 plants [27]. The avrbs2 gene was the first for which a requirement for bacterial fitness was demonstrated [28], giving rise to the notion that Bs2-mediated resistance would be stable. Sequence analysis of the variant avrbs2 genes isolated from strains growing on Bs2 heterozygous plants suggests that evolutionary pressure favors loss of avirulence but maintenance of virulence activities. A similar example of an effective R protein is found in potato. The Ry R protein recognizes the Potato Virus Y (PVY) Nla protein as an avirulence protein. Nla is a proteinase and mutations of the active proteinase domain compromise Ry recognition [29]. A different virus whose proteinase has the same specificity as PVY s, but is only 63% identical to the PVY proteinase, also triggers Ry function. What is not clear is whether Ry recognizes the Nla

R-gene products and their interactions with other proteins from host and pathogen Nimchuk et al. 291 active site directly or via a product of Nla s enzymatic activity. Either way, this site is required for both the virulence and avirulence functions of Nla. AvrPto from P. syringae pv. tomato (Pst) enhances the virulence of the pathogen on susceptible plants; strains expressing avrpto grow to higher population levels than those lacking avrpto [30,31]. The tomato Ser-Thr kinase R protein Pto recognizes AvrPto. A physical interaction between AvrPto and Pto has been demonstrated using the yeast two-hybrid system, and mutations that abolish the interaction in yeast abolish the induction of HR [32,33]. Therefore, this R Avr pair offers a unique system in which to examine the domains of an Avr protein required for both avirulence and virulence. AvrPto has been randomly mutagenized in two separate studies. In one, of nine non-pto-interacting AvrPto derivatives carrying point mutations, six resulted in forms of AvrPto that were incapable of conferring enhanced virulence upon Pst strains. However, the other three mutants still conferred enhanced virulence [30]. In another study, 44 amino-acid substitutions were randomly generated within the central section (amino acids 30 124) of AvrPto, which had previously been demonstrated to be sufficient for the interaction with Pto in yeast [34]. Surprisingly, 36 of these AvrPto derivatives still interacted with Pto in the yeast twohybrid system. Of these 36 derivatives, all that were tested also continued to elicit a Pto-dependent response in planta. The authors did not report whether any of these mutants were also compromised in causing enhanced virulence. Taken together, these results may indicate that the overall structure of this AvrPto domain, and not a particular set of amino-acid motifs, is required for recognition by Pto. Thus, both the recognition of specific motifs required for virulence and flexibility in recognition of Avr protein structures, may each provide effective R function. Bacterial pathogens carry suites of effectors that may be delivered into host plants. Jettisoning a defeated effector may be beneficial in overcoming resistant hosts, but might be costly to the long-term fitness of the pathogen on susceptible hosts. Recent studies indicate that some bacterial pathogens encode additional proteins that mask the presence of a particular Avr effector. For example, in Pseudomonas syringae pv. phaseolicola, the Type-III effector AvrPphF masks an unknown avr gene [35,36 ]. Nevertheless, in this evolutionary war, the hosts have developed the ability to recognize the masking effector: in another strain of P. syringae pv. phaseolicola AvrPphF is recognized. Moreover, the avirulence activity of AvrPphF can itself be masked by yet a third Type-III effector, AvrPphC, which is an avirulence determinant on soybean. Inhibition of host responses may also explain the virulence function of AvrRpt2 [37 ]. This Type-III effector strongly interferes with the RPM1-dependent HR [38] and weakly inhibits RPS4 or RPS5 function while not inhibiting HR [37 ]. It remains unclear whether these differences result from the interaction with the same or distinct plant targets. Which hand do I hold? Several lines of evidence indicate that the LRR domains of R proteins may act as determinants of specificity and ligand binding. The portions of LRR sequences corresponding to the putative solvent exposed residues are undergoing diversifying selection (see Jones, pp 281 287 and [39]). Recently, the LRR-like domain of the rice resistance protein Pita was shown to be required for interaction with Avr-Pita in the yeast two-hybrid system and by farwestern analysis. Furthermore, mutations in either Avr-Pita or Pita that abolished resistance also abolished the interactions in vitro [40 ]. This is the first demonstrated interaction between an LRR-containing R protein and its cognate Avr protein. Swapping of domains between homologous proteins further maps specificity to the LRRs (see Jones, pp 281 287). One tantalizing result from Jeff Ellis and his colleagues suggests that LRRs from the flax L proteins might determine specificity combinatorially with an amino-terminal domain. These authors first created chimeric L proteins, analyzed transgenic plants for response, and found that the LRRs determined L specificity [41]. However, they subsequently analyzed two functionally distinct L alleles that are sequence identical in the LRRs, but differ in the amino-terminal domain. Their results suggest that a subdomain of the TIR is required for specificity and is also under diversifying selection [42 ]. Thus, it is possible that the LRRs are necessary but not sufficient for the specific recognition of Avr proteins and that LRRs and aminoterminal domains have co-evolved to function in a coordinated manner. Recent genetic analysis of natural variation in RPS2 function indicates that recognition of AvrRpt2 also requires unidentified host genes [43]. The Arabidopsis Po-1 ecotype is susceptible to P. syringae expressing avrrpt2. The Po-1 RPS2 allele can, however, complement a fully susceptible rps2 mutant allele in Col-0. Conversely, Col-0 RPS2 confers full resistance when expressed in a Po-1 background. A cross between Col-0 and Po-1, and subsequent quantitative trait locus (QTL) mapping suggests that one or more loci located near RPS2 are also required for RPS2 function. Domain swaps between the RPS2 genes of Po-1 and Col-0 led to the identification of just six amino acids in the LRRs that are required for function in the different genetic backgrounds. The allele-specific nature of this response suggests that LRR regions may interact directly with the host factors required for RPS2 function. LRRs may interact with other common host components of defense. This hypothesis was initially suggested after studies of a loss-of-function allele of RPS5, an Arabidopsis R gene encoding an NBS-LRR R protein. This allele interferes with resistance pathways that are mediated by other NBS-LRR R proteins [18]. These findings suggest that the LRRs interact with a host protein that is common to several R-gene-mediated pathways. Dominant negative

292 Biotic interactions mutants have also been generated in domains other than LRRs. Mutations in the TIR or NBS gave rise to dominant negative alleles of tobacco N [44 ]. Similarly, mutations in the CC or amino-terminal hydrophobic regions of RPS2 resulted in dominant negative alleles [45 ]. It is unclear if these dominant negative alleles of N or RPS2 interfere with resistance mediated by other NBS-LRR resistant proteins. What about the chaperones? Several loci have been identified that are required for one or more R genes (see review by Glazebrook, pp 301 308 and [1]). The current challenge is to determine whether their corresponding proteins form complexes with R proteins as described in the preceding section. In tomato, Pto-mediated resistance against P. syringae expressing avrpto requires Prf, an NBS-LRR protein [46]. Use of constitutively activated Pto kinase demonstrated that Prf acts at or downstream of Pto in the signaling cascade [47]. This suggests that a ternary complex of Pto, Prf and AvrPto is necessary to activate resistance. However, physical interaction among Prf and Pto and/or AvrPto remains to be reported. Several putative interactors of AvrPto and Pto have been identified in the yeast two- and three-hybrid systems (summarized in [48]). Several, of these encode transcription factors that are associated with ethylene and defense responses. These transcription factors are phosphorylated by Pto in vitro. Current models suggest that the interaction of AvrPto with Pto activates the Pto kinase; the resistance response is mediated by the activation of these, and other, Pto partners [48]. Only Prf has, however, been demonstrated to be required for Pto-dependent resistance. The data are also consistent with a model in which Pto constitutively activates proteins that are involved in basal defense and this activation is abrogated by interaction with AvrPto. Therefore, inhibition of Pto function may result in the induction of specific resistance against avrpto-expressing pathogens. The turnip crinkle virus (TCV) coat protein (CP), in addition to having an obvious role in virulence, is also an Avr protein. TCV-CP triggers the NBS-LRR HRT gene in Arabidopsis. CP interacts in yeast two-hybrid screens with a NAC-family transcriptional activator. TCV expressing CP mutants that no longer interact with this protein are, however, virulent on HRT plants. Thus, this additional interaction may be required for the HRT-elicited resistant response [49 ]. Whether HRT interacts with the transcriptional activator or if the interaction occurs before or after HRT action is unknown. There is also evidence that at least RPS2 can interact with Type-III effectors other than its cognate Avr protein. This may support the notion that a limited set of host proteins are targets for virulence effectors and are guarded by multiple R proteins. A co-immunoprecipitation assay demonstrated that RPS2 interacted with its cognate Avr protein, AvrRpt2. Surprisingly, immune complexes between RPS2 and AvrB also formed [50 ]. Pathogens expressing avrb elicit resistant responses on plants expressing RPM1, but not RPS2. This finding suggests that perhaps more than just cognate R proteins interact with Avr proteins in planta, but that only specific R Avr protein complexes can trigger resistance. Most models that explain how R proteins specifically recognize their corresponding Avr protein have been articulated on the basis of the lines of evidence discussed above. Yet, one would expect the regions that are most variable among R proteins should be involved in the specificity of resistance responses. The region between the amino-terminal CC or TIR domain and the start of the NBS region is highly variable. We speculate that this variable region is involved in binding to additional host proteins, and these additional interactions may also determine specificity. It is this combinatorial diversity that Luck et al. [42 ] have defined as evolving in concert with the LRRs of L proteins. Conclusions The dance floor is getting crowded; a plethora of resistance and avirulence genes have now been cloned. We are now on the cusp of understanding the mechanisms by which their protein products function during the complex interplay between host and pathogen. Localization studies have shown that R proteins are ideally positioned on the dance floor to cut in and dance with the Avr proteins. Further characterization of R and Avr proteins has shown that recognition is most likely not a simple receptor ligand interaction. We may be able to use this information to predict where the true and heretofore shady partners of Avr proteins, their putative virulence targets, will be localized. Additional host factors are probably required for recognition and signaling for resistance. Parents and chaperones watch with cautious eyes in an attempt to discern the dancers from the dance. Acknowledgements We thank Andrew Bent and Pamela Ronald for providing material before publication. Work on these topics in the Dangl lab is supported by a National Science Foundation grant (IBN-9724075) and a US Department of Energy grant (DE-FG05-95ER20187). JHC is supported by an NRSA Fellowship from the National Institutes of Health. References and recommended reading Papers of particular interest, published within the annual period of review, have been highlighted as: of special interest of outstanding interest 1. Glazebrook J, Rogers EE, Ausubel FM: Use of Arabidopsis for genetic dissection of plant defense responses. Annu Rev Genet 1997, 31:547-569. 2. Hammond-Kosack KE, Jones JDG: Plant disease resistance genes. Annu Rev Plant Physiol Plant Mol Biol 1996, 48:575-607. 3. The Arabidopsis Genome Initiative: Analysis of the genome of the flowering plant Arabidopsis thaliana. Nature 2000, 408:796-815. 4. Jones DA, Jones JDG: The roles of leucine rich repeats in plant defences. Adv Bot Res Adv Plant Pathol 1996, 24:90-167. 5. Kjemtrup S, Nimchuk Z, Dangl JL: Effector proteins of phytopathogenic bacteria: bifunctional signals in virulence and host recognition. Curr Opinion Microbiol 2000, 3:73-78.

R-gene products and their interactions with other proteins from host and pathogen Nimchuk et al. 293 6. van der Biezen EA, Jones JDG: Plant disease resistance proteins and the gene-for-gene concept. Trends Biochem Sci 1998, 23:454-456. 7. Piedras P, Rivas S, Dröge S, Hillmer S, Jones JDG: Functional, c myc epitope tagged Cf-9 resistance gene products are plasmamembrane localized and glycosylated. Plant J 2000, 21:529-536. 8. Benghezai M, Wasteneys GO, Jones DA: The C-terminal dilysine motif confers endoplasmic reticulum localization to type I membrane proteins in plants. Plant Cell 2000, 12:1179-1201. 9. Van der Hoorn RAL, Van der Ploeg A, dewit PJGM, Joosten MHAJ: The C-terminal dilysine motif for targeting to the endoplasmic reticulum is not required for Cf-9 function. Mol Plant Microbe Interact 2001, 14:412-415. 10. Laugé R, Dmitriev AP, Joosten MHAJ, De Wit PGJM: Additional resistance gene(s) against Cladosporium fulvum present on the Cf-9 introgression segment are associated with strong PR protein accumulation. Mol Plant Microbe Interact 1998, 11:301-308. 11. Lauge R, Goodwin PH, de Wit PJ, Joosten MH: Specific HR-associated recognition of secreted proteins from Cladosporium fulvum occurs in both host and non-host plants. Plant J 2000, 23:735-745. 12. Gálan JE, Collmer A: Type III secretion machines: bacterial devices for protein delivery into host cells. Science 1999, 284:1322-1328. 13. Hueck CJ: Type III protein secretion systems in bacterial pathogens of animals and plants. Microbiol Mol Biol Rev 1998, 62:379-433. 14. Nimchuk Z, Marois E, Kjemtrup S, Leister RT, Katagiri F, Dangl JL: Eukaryotic fatty acylation drives plasma membrane targeting and enhances function of several Type III effector proteins from Pseudomonas syringae. Cell 2000, 101:353-363. 15. Boyes DC, Nam J, Dangl JL: The Arabidopsis thaliana RPM1 disease resistance gene product is a peripheral plasma membrane protein that is degraded coincident with the hypersensitive response. Proc Natl Acad Sci USA 1998, 95:15849-15854. 16. Shan L, Thara VK, Martin GB, Zhou J-M, Tang X: The Pseudomonas AvrPto protein is differentially recognized by tomato and tobacco and is localized to the plant plasma membrane. Plant Cell 2000, 12:2323-2337. 17. Loh Y-T, Zhou J, Martin GB: The myristylation motif of Pto is not required for disease resistance. Molec Plant Microbe Interact 1998, 11:572-576. 18. Warren RF, Henk A, Mowery P, Holub E, Innes RW: A mutation within the leucine-rich repeat domain of the Arabidopsis disease resistance gene RPS5 partially suppresses multiple bacterial and downy mildew resistance genes. Plant Cell 1998, 10:1439-1452. 19. Warren RF, Merritt PM, Holub E, Innes RW: Identification of three putative signal transduction genes involved in R gene-specified disease resistance in Arabidopsis. Genetics 1999, 152:401-412. 20. Swiderski MR, Innes RW: The Arabidopsis PBS1 resistance gene encodes a member of a novel protein kinase subfamily. Plant J 2001, 26:101-112. 21. Van den Ackerveken G, Marois E, Bonas U: Recognition of the bacterial AvrBs3 protein occurs inside the plant cell. Cell 1996, 87:1307-1316. 22. Zhu W, Yang B, Kurata N, Johnson LB, White FF: The C terminus of AvrXa10 can be replaced by the transcriptional activation domain of VP16 from the Herpes Simplex virus. Plant Cell 1999, 11:1665-1674. 23. Yang B, Zhu W, Johnson LB, White FF: The virulence factor AvrXa7 of Xanthomonas oryzae pv. oryzae is a type III secretion pathwaydependent nuclear-localized double-stranded DNA-binding protein. Proc Natl Acad Sci USA 2000, 97:9807-9812. The authors describe the roles of the different domains of the protein avrxa7 in its virulence and avirulence activity. They show that the recognition of this Avr protein by its receptor depends not only on its primary structure but also on the association of this protein with a host factor, presumably in the nucleus. The AvrXa7 protein was found to bind single-stranded DNA in vitro, further strengthening evidence for a role for this family of virulence factors in the modulation of host gene expression. 24. Ballvora A, Pierre M, Van den Ackervecken G, Schornack S, Rossier O, Ganal M, Lahaye T, Bonas U: Genetic mapping and functional analysis of the tomato Bs4 locus governing recognition of the Xanthomonas campestris pv. vesicatoria AvrBs4 protein. Mol Plant Microbe Interact 2001, 14:629-638. 25. Bai J, Choi S-H, Ponciano G, Leung H, Leach JE: Xanthomonas oryzae pv. oryzae avirulence genes contribute differentially and specifically to pathogen aggressiveness. Mol Plant Pathol 2000, 13:1322-1329. 26. Orth K, Xu Z, Mudgett MB, Bao ZQ, Palmer LE, Bliska JB, Mangel WF, Staskawicz B, Dixon JE: Disruption of signaling by Yersinia effector YopJ, a ubiquitin-like protein protease. Science 2000, 290:1594-1597. An interesting paper regarding the role of Yersinia Type-III effector YopJ in modulating host mitogen-activated protein kinase (MAPK) signaling pathways. Consensus protease sites were found to be required for inhibiting host MAPK pathways. The Xanthomonas YopJ homologue AvrBst was found to require these consensus sequences to trigger R-gene function. Does this mean that AvrBst inhibits plant MAPK pathways during disease? 27. Gassman W, Dahlbeck D, Chesnokova O, Minisavage G, Jones JB, Staskawicz BJ: Molecular evolution of virulence in natural field strains of Xanthomonas campestris pv. vesicatoria. J Bacteriol 2000, 182:7053-7059. 28. Kearney B, Staskawicz BJ: Widespread distribution and fitness contribution of Xanthomonas campestris avirulence gene avrbs2. Nature 1990, 346:385-386. 29. Mestre P, Brigneti G, Baulcombe DC: A Ry-mediated resistance response in potato requires the intact active site of the NIa proteinase from potato virus Y. Plant J 2000, 23:653-661. 30. Shan L, He P, Zhou J, Tang X: A cluster of mutations disrupt the avirulence but not the virulence functions of AvrPto. Molec Plant Microbe Interact 2000, 13:592-598. 31. Chang JH, Rathjen JP, Bernal AJ, Staskawicz BJ, Michelmore RW: avrpto enhances growth and necrosis caused by Pseudomonas syringae pv. tomato in tomato lines lacking either Pto or Prf. Molec Plant Microbe Interact 2000, 13:568-571. 32. Tang X, Frederick RD, Zhou J, Halterman DA, Jia Y, Martin GB: Physical interaction of avrpto and the Pto kinase defines a recognition event involved in plant disease resistance. Science 1996, 274:2060-2063. 33. Scofield SR, Tobias CM, Rathjen JP, Chang JH, Lavelle DT, Michelmore RW, Staskawicz BJ: Molecular basis of gene-for-gene specificity in bacterial speck disease of tomato. Science 1996, 274:2063-2065. 34. Chang JH, Tobias CM, Staskawicz B, Michelmore RW: Functional studies of the bacterial avirulence protein AvrPto by mutational analysis. Mol Plant Microbe Interact 2001, 14:451-459. 35. Jackson RW, Athanassopoulos E, Tsiamis G, Mansfield JW, Sesma A, Arnold DL, Gibbon MJ, Murillo J, Taylor JD, Vivian A: Identification of a pathogenicity island, which contains genes for virulence and avirulence, on a large native plasmid in the bean pathogen Pseudomonas syringae pathovar phaseolicola. Proc Natl Acad Sci USA 1999, 96:10875-10880. 36. Tsiamis G, Mansfield JW, Hockenhull R, Jackson RW, Sesma A, Athanassopoulos E, Bennett MA, Stevens C, Vivian A, Taylor JD et al.: Cultivar-specific avirulence and virulence functions assigned to avrpphf in Pseudomonas syringae pv. phaseolicola, the cause of bean halo-blight disease. EMBO J 2000, 19:3204-3214. The authors report that two avr genes in Pseudomonas function during virulence to mask the recognition of other avr genes by host R genes. This discovery shows that bacteria have developed mechanisms to avoid the triggering of plant defense response by some effectors. This complex web of inhibition and triggering of R-gene function predicts that virulence targets for Avr proteins will be components of R-signaling pathways. 37. Chen Z, Kloek AP, Boch J, Katagiri F, Kunkel BN: The Pseudomonas syringae avrrpt2 gene product promotes pathogenicity from inside the plant cell. Mol Plant Pathol 2000, 13:1312-1321. The authors show that avrrpt2 is a virulence factor in Pseudomonas and that virulence activity is associated with the inhibition of defense responses. Why does avrrpt2 inhibit the HR triggered by RPM1 but not by other R genes? Can this system be used to identify virulence targets for AvrRpt2? 38. Ritter C, Dangl JL: Interference between two specific pathogen recognition events mediated by distinct plant disease resistance genes. Plant Cell 1996, 8:251-257. 39. Ellis J, Dodds P, Pryor T: The generation of plant disease resistance specificities. Trends Plant Sci 2000, 5:373-379.

294 Biotic interactions 40. Jia Y, McAdams SA, Bryan GT, Hershey HP, Valent B: Direct interaction of resistance gene and avirulence gene products confers rice blast resistance. EMBO J 2000, 19:4004-4014. An exciting paper that provides evidence to show that NBS-LRR proteins directly bind Avr ligands. Another interesting observation is that fungal Avr- Pita is recognized inside the host cell. How s it getting in there? Avr-Pita appears to encode a protease, does this protease cleave a virulence target inside the host cell? 41. Ellis JG, Lawrence GJ, Luck JE, Dodds PN: Identification of regions in alleles of the flax rust resistance gene L that determine differences in gene-for-gene specificity. Plant Cell 1999, 11:495-506. 42. Luck JE, Lawrence GJ, Dodds PN, Shepherd KW, Ellis JG: Regions outside of the leucine-rich repeats of flax rust resistance proteins play a role in specificity determination. Plant Cell 2000, 12:1367-1377. Following up on earlier observations that L alleles encoding distinct specificities possess identical LRRs, the authors use recombinant alleles to demonstrate that the amino-terminal TIR domains are undergoing diversifying selection and contribute to specificity. These data allow for a model of NBS- LRR protein quaternary structure for the definition of either inter- or intra-molecular interactions. 43. Banerjee D, Zhang D, Bent A: The LRR domain can determine effective interaction between RPS2 and other host factors in Arabidopsis RPS2-mediated disease resistance. Genetics 2001, 158:439-450. 44. Dinesh-Kumar SP, Tham W-H, Baker BJ: Structure function analysis of the tobacco mosaic virus resistance gene N. Proc Natl Acad Sci USA 2000, 97:14789-14794. See annotation [45 ]. 45. Tao Y, Yuan F, Leister RT, Ausubel FM, Katagiri F: Mutational analysis of the Arabidopsis nucleotide binding site leucine-rich repeat resistance gene RPS2. Plant Cell 2000, 12:2541-2554. The authors of this paper and [44 ] present structure function analyses of two distinct subclasses of NBS-LRR R genes. Like the data described in [42 ], the findings reported in these papers suggest that R protein quaternary structure is important for regulating signaling. In addition, Tao et al. suggest that amino-terminal domains of RPS2 drive cell death signaling and interactions with a co-immunoprecipitated host protein. Overexpression of RPS2 triggered protoplast cell-death in the absence of AvrRpt2. Interestingly, although ndr1 mutants block AvrRpt2 triggered HR, overexpression of RPS2 can still causes cell death in ndr1 protoplasts. Given the results of Chen et al. [37 ], perhaps avrrpt2 inhibits RPS2 HR in the absence of NDR1. 46. Salmeron JM, Oldroyd GED, Rommens CMT, Scofield SR, Kim H-S, Lavelle DT, Dahlbeck D, Staskawicz BJ: Tomato Prf is a member of the leucine-rich repeat class of plant disease resistance genes and lies embedded within the Pto kinase gene cluster. Cell 1996, 86:123-133. 47. Rathjen JP, Chang JH, Staskawicz BJ, Michelmore RW: Constitutively active Pto induces a Prf-dependent hypersensitive response in the absence of avrpto. EMBO J 1999, 18:3232-3240. 48. Bogdanove AJ, Martin GB: AvrPto-dependent Pto-interacting proteins and AvrPto-interacting proteins in tomato. Proc Natl Acad Sci USA 2000, 97:8836-8840. 49. Ren T, Qu F, Morris TJ: HRT gene function requires interaction between a NAC protein and viral capsid protein to confer resistance to Turnip Crinkle Virus. Plant Cell 2000, 12:1917-1925. The authors report a strong correlation between loss of yeast two-hybrid interaction and R-gene-mediated resistance. Are NAC protein mutants susceptible to virus infection? 50. Leister RT, Katagiri F: A resistance gene product of the nucleotide binding site-leucine rich repeats class can form a complex with bacterial avirulence proteins in vitro. Plant J 2000, 22:345-354. The authors take a biochemical approach toward identifying RPS2 and AvrRpt2 complexes in co-immunoprecipitation assays. 51. Swords KM, Dahlbeck D, Kearney B, Roy M, Staskawicz BJ: Spontaneous and induced mutations in a single open reading frame alter both virulence and avirulence in Xanthomonas campestris pv. vesicatoria avrbs2. J Bacteriol 1996, 178:4661-4669. 52. Mudgett MB, Chesnokova O, Dahlbeck D, Clark ET, Rossier O, Bonas U, Staskawicz BJ: Molecular signals required for type III secretion and translocation of the Xanthomonas campestris AvrBs2 protein to pepper plants. Proc Natl Acad Sci USA 2000, 97:13324-13329. 53. Duan YP, Castaneda A, Zhao G, Erdos G, Gabriel DW: Expression of a single, host-specific, bacterial pathogenicity gene in plant cells elicits division, enlargement, and cell death. Mol Plant Microbe Interaction 1999, 12:556-560. 54. Zhao Y, DelGrosso L, Yigit E, Dempsey DA, Klessig DF, Wobbe KK: The amino terminus of the coat protein of Turnip crinkle virus is the AVR factor recognized by resistant Arabidopsis. Mol Plant Microbe Interact 2000, 13:1015-1018. 55. Ritter C, Dangl JL: The avrrpm1 gene of Pseudomonas syringae pv. maculicola is required for virulence on Arabidopsis. Mol Plant Microbe Interact 1995, 8:444-453. 56. Dixon MS, Golstein C, Thomas CM, van Der Biezen EA, Jones JD: Genetic complexity of pathogen perception by plants: the example of Rcr3, a tomato gene required specifically by Cf-2. Proc Natl Acad Sci USA 2000, 97:8807-8814.