Previous Article |
Table of Contents
| Next Article 
From the Academy: Colloquium Review
REVIEW
Unique characteristics of Xanthomonas oryzae pv. oryzae AvrXa21 and implications for plant innate immunity
Department of Plant Pathology, University of California, One Shields Avenue, Davis, CA 95616
Edited by Diter von Wettstein, Washington State University, Pullman, WA, and approved August 28, 2006
 |
Abstract |
|---|
 Top Abstract Detection of Pathogens by...PRR XA21 Represents a...AvrXa21 Activity Requires a...The Core AvrXa21 Molecule...Cell Density-Dependent...PerspectiveAcknowledgementsReferences |
|---|
This article provides a brief overview of some of the major concepts and molecular features of plant and animal innate immune systems. The rice pathogen recognition receptor, XA21, confers resistance to Xanthomonas oryzae pv. oryzae strains producing the AvrXa21 elicitor. Xa21 codes for a receptor-like kinase consisting of an extracellular leucine-rich repeat domain, a transmembrane domain, and a cytoplasmic kinase domain. We show that AvrXa21 activity requires the presence of rax (required for AvrXa21) A, raxB, and raxC genes that encode components of a type one secretion system. In contrast, an hrpC– strain deficient in type three secretion maintains AvrXa21 activity. Xanthomonas campestris pv. campestris can express AvrXa21 activity if raxST, encoding a putative sulfotransferase, and raxA are provided in trans. Expression of rax genes depends on population density and other functioning rax genes. This and other data suggest that the AvrXa21 pathogen-associated molecule is involved in quorum sensing. Together these data suggest that AvrXa21 represents a previously uncharacterized class of Gram-negative bacterial signaling molecules. These results from our studies of the XA21/AvrXa21 interaction call for some modifications in the way we think about innate immunity strategies.
XA21 | pathogen-associated molecule pattern | type I secretion | quorum sensing | rice
|
Detection of Pathogens by Plant and Animal Hosts |
|---|
|
TopAbstractDetection of Pathogens by...PRR XA21 Represents a...AvrXa21 Activity Requires a...The Core AvrXa21 Molecule...Cell Density-Dependent...PerspectiveAcknowledgementsReferences |
|---|
Evidence has accumulated that plants also detect PAMPs but, unlike animal systems, which have PRRs for cytoplasmic and extracellular perception of PAMPs, all biochemically characterized phytopathogen PAMPs are active at the cell surface (12). With notable exceptions, another general characteristic of PAMP recognition by plants is that, although it promotes expression of pathogenesis-related proteins and other characteristics of pathogen response, it does not lead to clear disease resistance or a related hypersensitive response, which includes localized plant cell death. For example, treatment of tobacco with lipopolysaccharide leads to increases in expression of genes associated with defense response but has no immediate effect on infection with virulent strains (6).
Surprisingly little is known about PAMP receptors in plants. The best characterized plant PRR for a PAMP is the Arabidopsis thaliana receptor-like kinase (RLK), FLS2 (flagellin sensing 2), which includes an extracellular LRR ligand-binding domain and an intracellular serine/threonine kinase and directly recognizes a conserved N-terminal fragment of bacterial flagellin (13, 14). Stimulation of FLS2 by flagellin activates pathogenesis-related gene expression, and pretreatment with it leads to resistance (15, 16). Recent studies have shown that the presence of FLS2 decreases host susceptibility to a pathovar of Pseudomonas syringae when the pathogen is applied to leaves by spraying but not infiltration (16, 17). This finding demonstrates a direct link between disease resistance and PAMP perception that previously had been lacking in plants. Another recently characterized plant receptor that perceives a PAMP is EFR (elongation factor Tu receptor) (18), which recognizes elf18, consisting of the first 18 aa of EF-Tu (elongation factor Tu) (19).
In addition to recognition of conserved PAMPs, plant PRRs also recognize strain-specific molecules produced by phytopathogens, termed pathogen avirulence factors (20, 21). This specific recognition generally triggers a strong defense response, often including the hypersensitive response. To explain the observation of interactions between dominant PRR genes and bacterial avr genes, H. H. Flor proposed the gene-for-gene hypothesis in which a single plant-gene product recognizes a single bacterial avr gene product and blocks disease formation (22). Twenty-one rice loci that confer genetically dominant resistance to bacterial blight disease caused by Xanthomonas oryzae pv. oryzae (Xoo) have been identified, and four of these genes have been cloned (reviewed in ref. 23).
The majority of plant PRR genes cloned to date code for cytoplasmically localized NOD family members, with a domain that contains a nucleotide-binding site (NBS) and a domain with LRRs (NBS-LRRs). All characterized NBS-LRRs that block bacterial pathogenesis recognize Avr proteins from pathogenic bacteria that are secreted through a large bacterial complex called a type three secretion system (TTSS). The TTSS is thought to function by transporting proteins, called effectors, directly into the cytoplasm of the host cell, and it is known to be an essential transport system for disease development and bacterial multiplication (24, 25). As recently reviewed in refs. 26 and 27, many Avr molecules rely on the TTSS for secretion and act as type III effectors. These molecules include AvrRpt2, AvrB, AvrRpm1, and AvrPto from P. syringae; XopD and AvrBsT from Xanthomonas campestris; and the three cloned Avr factors from Xoo, AvrXa7, AvrXa10, and AvrXa27. However, only a few type III effectors have known biochemical functions. Of those effectors that have been characterized, several are enzymes, such as proteases and phosphatases, that act on host protein substrates to interfere with, suppress, and manipulate host defense and signaling by defense hormones such as salicylic and jasmonic acid (28). A recent elegant example of this comes from the human pathogen Yersinia, in which the type III effector, YopJ, acetylates a mitogen-activated protein kinase (MAPK) kinase, blocking the site of its activation by phosphorylation (29). Others encode proteins with a nuclear localization signal and more directly modulate host transcription. For example, the AvrBs3/PthA family has a nuclear localization signal and an acidic transcription activation domain that is required for AvrBs3-dependent hypersensitive response (30). Few direct interactions have been reported between NBS-LRR PRRs and their corresponding effectors (31). Instead, the defense response often is triggered by interaction between the NBS-LRR and another plant protein, which is targeted or modified by the type III effector (32).
In addition to the cytoplasmic NBS-LRRs, two other classes of plant PRR that recognize Avr proteins have been described. These are the RLKs, composed of various putative ligand-binding extracellular domains and an intracellular kinase, and receptor-like proteins (RLPs), composed only of a membrane-anchored extracellular domain (e.g., tomato CF9) or a presumed secreted extracellular domain (e.g., rice XA21D) (33, 34). Relatively few plant RLKs and RLPs have been cloned and characterized to date, although, like NBS-LRRs, sequence analyses indicate that there are a large number of genes of these classes in plants (33). The best studied PRR RLK that confers a race-specific response is the rice XA21 protein, which recognizes Xoo strains carrying AvrXa21 activity (35). The three other dominant RLKs cloned and characterized to date are the rice Xa26 and Pi-d2 proteins and barley RPG1 (36–38). The potential ligand-binding capabilities of the extracellular domains of the RLK PRRs, and the facts that XA21 is present in microsomal fractions (39) and that Pi-d2 localizes to the cell membrane (37), suggest a simple model in which XA21 and other RLKs recognize Avr proteins in the extracellular space, directly or indirectly, as the tomato Cf RLPs do (40).
Recent discoveries, including the results described here, call for a blurring of the distinctions between PRR proteins that recognize PAMPs and those that recognize Avr proteins and other conceptual dichotomies that have been established in plant pathology. Rather, a continuum of classes seems to exist, with plants making use of a diversity of strategies for innate immunity (41). Although the AvrXa21 molecule(s) itself has not yet been identified, we show here that AvrXa21 activity (i) depends on a bacterial type one secretion system (TOSS) not on a TTSS, (ii) is regulated by a two-component regulatory system that responds to Xoo cell population density, and (iii) may be conserved in most Xanthomonas spp. Furthermore, we show that a high-pressure liquid chromatography (HPLC)-purified fraction carrying AvrXA21 activity can induce raxST gene expression at low cell density. These data suggest that AvrXA21 represents a previously uncharacterized class of signaling molecules, which is used by Xoo for quorum sensing (QS) and which does not fall into any of the previously described classes of PAMPs or Avr factors.
|
PRR XA21 Represents a Large Class of Kinases Predicted to be Involved in Innate Immunity |
|---|
|
TopAbstractDetection of Pathogens by...PRR XA21 Represents a...AvrXa21 Activity Requires a...The Core AvrXa21 Molecule...Cell Density-Dependent...PerspectiveAcknowledgementsReferences |
|---|
|
AvrXa21 Activity Requires a TOSS |
|---|
|
TopAbstractDetection of Pathogens by...PRR XA21 Represents a...AvrXa21 Activity Requires a...The Core AvrXa21 Molecule...Cell Density-Dependent...PerspectiveAcknowledgementsReferences |
|---|
|
|
The data from our previous genetic studies coupled with those from the bioassay represent the discovery of a type I secreted factor that can trigger plant innate immunity. Although there are five systems (types I to V) for protein secretion in Gram-negative bacteria (24), only the TTSS previously has been shown to secrete Avr factors, which then are detected intracellularly. We hypothesize that other members of the large complement of plant non-RD IRAK-family receptor kinases (42) will be found to detect molecules secreted by non-TTSS systems.
|
The Core AvrXa21 Molecule Is Conserved in X. campestris pv. campestris (Xcc) |
|---|
|
TopAbstractDetection of Pathogens by...PRR XA21 Represents a...AvrXa21 Activity Requires a...The Core AvrXa21 Molecule...Cell Density-Dependent...PerspectiveAcknowledgementsReferences |
|---|
We carried out the AvrXa21 activity assay with supernatant prepared from Xcc transformed with the PXO99 raxSTAB genes and compared the lesion lengths (Table 1). Pretreatment with the supernatants of the Xoo wild-type strain and the Xcc strain carrying the raxSTAB genes induces resistance against infection by raxST– Xoo (lesion lengths of 1.3 ± 0.4 cm and 1.8 ± 0.8 cm, respectively), whereas leaves pretreated with the supernatants of the Xoo raxST– and the Xcc wild-type strains have long lesions (10.1 ± 5.2 cm and 10.0 ± 4.8 cm, respectively). Thus, the supernatant of Xcc carrying the raxST, raxA, and raxB genes possesses AvrXa21 activity. Furthermore, Xcc carrying Xoo raxST and raxA genes showed full AvrXa21 activity but not a strain with raxST alone (Table 1).
|
Another element of the definition of a PAMP is that it is essential for the pathogen. The fact that the core of AvrXa21 appears to be conserved in Xcc suggests that the molecule(s) has a function that makes it selectively advantageous. In addition, although we did not observe differences in virulence in the raxSTAB and raxC knockout strains under controlled conditions, in a field study, 37 Xoo Korean strains lacking AvrXa21 activity appeared to have reduced fitness (51). These results suggest both that the core AvrXa21 molecule is conserved between Xoo and Xcc and that loss of AvrXa21 results in a fitness cost to Xoo. Therefore, AvrXa21, which is detected by the dominant, resistance-conferring XA21 PRR, shares characteristics of PAMPs and Avr proteins.
|
Cell Density-Dependent Expression of rax Genes |
|---|
|
TopAbstractDetection of Pathogens by...PRR XA21 Represents a...AvrXa21 Activity Requires a...The Core AvrXa21 Molecule...Cell Density-Dependent...PerspectiveAcknowledgementsReferences |
|---|
To test whether rax gene expression depends on cell population, rax gene expression in wild-type Xoo was measured by using real-time quantitative PCR with three primer sets specific for each of the following four rax genes, one in each operon: raxST, in raxSTAB (Fig. 3A); raxP, in raxPQ; raxR, in raxRH; and raxC (data not shown). Remarkably, the expression of all four rax genes was observed only in the wild-type strain at high cell densities, whereas levels of a ribosomal RNA were unaffected (Fig. 3A). We also measured rax gene expression in a strain constitutively expressing RaxR. Density dependence of rax gene expression in the raxR constitutive expressing strain is mostly lost (Fig. 3A). This result indicates that RaxR is involved in regulation of the density-dependent expression of the rax genes, including itself.
|
The characteristics of the Xoo rax genes suggest a QS system with features similar to those typically found in Gram-positive bacteria. For example, oligopeptides are the predominant signaling molecules for genetic competence in Bacillus substilis (58), virulence in Staphylococcus aureus (59), and the production of antimicrobial peptides, inducing bacteriocins and lantibiotics, in lactic acid bacteria (60, 61). As QS molecules, these peptides also activate their own expression and have been given the name autoinducible peptides (AIPs). AIPs are known to be secreted by ABC transporters that are similar to RaxB (44) and are sensed by two-component systems. Significantly, secretion of the putative AvrXa21 peptide by a TOSS would be a previously undescribed occurrence of an AIP in Gram-negative bacteria.
In summary, our data suggest that the AvrXa21 elicitor is a secreted peptide and hint that the biological function of the AvrXa21 elicitor is as a QS signal molecule, the production of which is regulated in a cell-density-dependent manner in the Gram-negative bacteria Xoo by a two-component system. Thus, the rice XA21 PRR effectively eavesdrops on the AvrXa21-mediated molecular conversation among Xoo, using AvrXa21 as an indicator of the presence of a pathogenic bacterial population of significant size and leading to plant responses that effectively halt infection (Fig. 1). Sequence analysis of the genomes of several phytopathogenic bacteria that are similar to Xoo reveals an abundance of the rax classes of molecular components. Xoo, Xcc, Xanthomonas axonopodis pv. citri, and Xylella fastidiosa each possess genes for 
30 two-component systems together with three to five TOSSs that, like RaxB, are predicted to transport peptides. Furthermore, it is apparent that X. fastidiosa does not use the TTSS for pathogenesis because genome analysis has revealed the absence of these components in this bacterium. Rather, X. fastidiosa contains TOSS genes similar to raxA, raxB, and raxC as well as sequences encoding as many as 10–12 peptides that are candidates for secretion (C. Dardick and P.C.R., unpublished results). These observations highlight that the interactions among phytopathogens, their hosts, and the environment are more sophisticated than are currently recognized.
Our data suggesting that the function of the AvrXa21 pathogen-associated molecule is as a QS signaling molecule may present a critical clue to answering the question, why is AvrXa21 maintained in Xoo regardless of its detection by a host protein? Based on others' results that suggest decreased fitness of rax mutants, AvrXa21 may be essential for Xoo as a signal molecule for cell–cell communication. We hypothesize that Xoo may use QS to coordinate infection and production of virulence factors. QS has global effects on bacterial growth, survival, and interactions with eukaryotes because QS-responsive genes compose up to 10% of the P. aeruginosa transcriptome (62). In that report, a large portion of QS-promoted genes code for membrane proteins and protein export apparatuses involved in secretion of virulence factors. A relationship between QS and virulence also has been described for the human pathogen Vibrio cholerae (63). Interestingly, in another case, a host can produce QS mimic molecules that interfere with the QS-regulated behaviors of the infecting bacteria (64). Given the abundance of other QS peptides and small molecules such as bacteriocins and lantibiotics in the host vicinity, it will not be surprising if these peptides are shown to play a role in the interaction of other bacteria with their hosts, as has been found for a homoserine lactone QS molecule from P. aeruginosa (57). In principle, recognition by plants of TOSS elicitors could be exploited to generate new types of specific and environmentally benign pesticides that induce plant defense responses.
|
Perspective |
|---|
|
TopAbstractDetection of Pathogens by...PRR XA21 Represents a...AvrXa21 Activity Requires a...The Core AvrXa21 Molecule...Cell Density-Dependent...PerspectiveAcknowledgementsReferences |
|---|
The fusion of PRR recognition of Avr and PAMP classes proposed in this article has implications for the mechanism of plant innate immunity. Recognition of PAMPs by the animal innate immune system has been proposed to act as guide for the adaptive immune system as to the nature of the infecting pathogen (2). There is not an obvious need for such a function in plants because each receptor can be tied to a specific response through signal transduction. Nonetheless, it is clear that the plant innate immune system recognizes both conserved and less conserved molecules. As recently articulated (67), the fact that plant responses to PAMPs often do not lead to resistance may be attributable to evolutionary selection for plant innate immune perception to require "priming" or "two-hits" so that resources are not unnecessarily wasted in responding to nonpathogenic or small populations of bacteria. In such a model, PAMP perception constitutes an early warning, and, indeed, pretreatment with lipopolysaccharide or flg22 leads to resistance to subsequent treatments with normally pathogenic bacteria (6, 16, 17). However, it may be that the necessity for the factor that primes the plant innate immune system to be a PAMP rather than an Avr molecule is simply a matter of experimental history. Because they are less useful in an agricultural context, plant pathologists have not looked for PRRs and corresponding Avr molecules that do not lead to significant resistance. However, the XA21D receptor, which lacks a transmembrane and kinase domain and confers only partial resistance, may be such a molecule (34). XA21D would likely not have been discovered without experiments showing that its gene family member, Xa21, conferred complete resistance. Conversely, as mentioned above, FLS2-dependent resistance to bacteria that express flagellin also has been observed (17). Physiologically relevant experiments in diverse genetic backgrounds may lead to further evidence that responses mediated by PRRs for both more and less conserved pathogen-associated molecules (i.e., PAMPs and Avrs) are also part of a continuum, with some functioning more often in priming and others more often leading to immediate resistance. Already among described Avr PRRs there is a diversity of strengths, with some leading to superresistance (68). Thus, in possessing receptors for more conserved molecules, it may be that plants are simply taking advantage of the conserved nature of PAMPs as targets for recognition rather than making a functional distinction in terms of responses.
Several examples support the idea that plants gain a selective advantage by making use of a variety of strategies for pathogen recognition beyond the classes of molecules that typically have been discussed (27, 32), both in terms of the molecule detected and the cognate PRRs. AvrXa21 is currently the only known elicitor that is secreted by a TOSS. Until recently, LRRs constituted the only signal recognition domain described for a cloned PRR RLK; however, a recently cloned RLK conferring rice blast resistance, Pi-d2, instead possesses an extracellular lectin-binding domain (37). Moreover, sequence analyses suggest that RLKs and receptor-like proteins with other, diverse extracellular binding domains also are involved in plant innate immunity (33, 42). Similarly, plants appear to use other, entirely different classes of molecules as PRRs as well. Indeed, the first cloned PAMP high-affinity binding site was the soybean 
-glucan elicitor binding protein (69). Localized to the extracellular side of the cell membrane, it is predicted to be tethered to the membrane via interactions with a receptor complex (70). Yet another perception strategy is represented by Xa27, which codes for a defense protein the promoter of which serves as the PRR by directly or indirectly interacting with the cognate, nuclear-localized AvrXa27 type III effector (71).
Our recent results suggest that the AvrXa21 elicitor is a TOSS-dependent secreted peptide that represents a previously undescribed family of signaling molecules for QS via a two-component system. Although QS molecules are known to be widely conserved and essential for bacterial communication, this example illustrates that such a QS molecule triggers plant innate immune responses. Our continuing work to identify the AvrXa21 molecule and understand the physical basis of its recognition may allow for identification of similar pathogen-associated molecules from plants and animal pathogens and their cognate host receptors.
|
Acknowledgements |
|---|
|
TopAbstractDetection of Pathogens by...PRR XA21 Represents a...AvrXa21 Activity Requires a...The Core AvrXa21 Molecule...Cell Density-Dependent...PerspectiveAcknowledgementsReferences |
|---|
|
Footnotes |
|---|
Abbreviations: LRR, leucine-rich repeat; PAMP, pathogen-associated molecular pattern; PRR, pathogen recognition receptor; RLK, receptor-like kinase; Xoo, Xanthomonas oryzae pv. oryzae; NBS, nucleotide-binding site; TTSS, type three secretion system; TOSS, type one secretion system; R, arginine; D, asparatate; IRAK, interleukin 1 receptor-associated kinase; Xcc, X. campestris pv. campestris; QS, quorum sensing.
*To whom correspondence should be addressed. E-mail: pcronald{at}ucdavis.edu
Author contributions: S.-W.L. and S.-W.H. designed research; S.-W.L. and S.-W.H. performed research; S.-W.L. contributed new reagents/analytical tools; S.-W.L. analyzed data; and S.-W.L., L.E.B., and P.C.R. wrote the paper.
This paper results from the Arthur M. Sackler Colloquium of the National Academy of Sciences, "From Functional Genomics of Model Organisms to Crop Plants for Global Health," held April 3–5, 2006, at The National Academy of Sciences in Washington, DC. Papers from this Colloquium will be available as a collection on the PNAS web site. The complete program is available on the NAS web site at www.nasonline.org/functional_genomics.
The authors declare no conflict of interest.
This article is a PNAS direct submission.
© 2006 by The National Academy of Sciences of the USA
|
References |
|---|
|
TopAbstractDetection of Pathogens by...PRR XA21 Represents a...AvrXa21 Activity Requires a...The Core AvrXa21 Molecule...Cell Density-Dependent...PerspectiveAcknowledgementsReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg What's this?
This article has been cited by other articles in HighWire Press-hosted journals:
|
|
 |
|
  K. Van, D. H. |
