Pseudomonas syringae pv. tomato DC3000 Type III Secretion Effector Polymutants Reveal an Interplay between HopAD1 and AvrPtoB

Article

Pseudomonas syringae pv. tomato DC3000 Type III Secretion Effector Polymutants Reveal an Interplay between HopAD1 and AvrPtoB Graphical Abstract

Authors Hai-Lei Wei, Suma Chakravarthy, Johannes Mathieu, ..., Bryan Swingle, Gregory B. Martin, Alan Collmer

Correspondence [email protected] (G.B.M.), [email protected] (A.C.)

In Brief Redundancy, interplay, and multiple interactions render bacterial type III secretion effectors (T3Es) challenging to understand. Wei et al. constructed a T3Edeficient strain of the plant pathogen Pseudomonas syringae and uncovered that the T3E HopAD1 activates plant immune responses while AvrPtoB suppresses HopAD1-mediated immunity, thereby revealing a complex interplay between T3Es.

Highlights d

An effectorless Pseudomonas syringae mutant was constructed to study effector interplay

d

The effector HopAD1 is sufficient to elicit cell death in Nicotiana benthamiana

d

An AvrPtoB mutant (AvrPtoBM3) suppresses HopAD1triggered host cell death

d

AvrPtoBM3 interacts with MKK2, which is required for HopAD1-triggered cell death

Wei et al., 2015, Cell Host & Microbe 17, 752–762 June 10, 2015 ª2015 Elsevier Inc. http://dx.doi.org/10.1016/j.chom.2015.05.007

Cell Host & Microbe

Article Pseudomonas syringae pv. tomato DC3000 Type III Secretion Effector Polymutants Reveal an Interplay between HopAD1 and AvrPtoB Hai-Lei Wei,1 Suma Chakravarthy,1 Johannes Mathieu,2 Tyler C. Helmann,1 Paul Stodghill,3 Bryan Swingle,1,3 Gregory B. Martin,1,2,* and Alan Collmer1,* 1School

of Integrative Plant Science, Section of Plant Pathology and Plant-Microbe Biology, Cornell University, Ithaca, NY 14853, USA Thompson Institute for Plant Research, Ithaca, NY 14853, USA 3United States Department of Agriculture-Agricultural Research Service, Ithaca, NY 14853, USA *Correspondence: [email protected] (G.B.M.), [email protected] (A.C.) http://dx.doi.org/10.1016/j.chom.2015.05.007 2Boyce

SUMMARY

The bacterial pathogen Pseudomonas syringae pv. tomato DC3000 suppresses the two-tiered plant innate immune system by injecting a complex repertoire of type III secretion effector (T3E) proteins. Beyond redundancy and interplay, individual T3Es may interact with multiple immunity-associated proteins, rendering their analysis challenging. We constructed a Pst DC3000 polymutant lacking all 36 T3Es and restored individual T3Es or their mutants to explore the interplay among T3Es. The weakly expressed T3E HopAD1 was sufficient to elicit immunity-associated cell death in Nicotiana benthamiana. HopAD1-induced cell death was suppressed partially by native AvrPtoB and completely by AvrPtoBM3, which has mutations disrupting its E3 ubiquitin ligase domain and two known domains for interacting with immunity-associated kinases. AvrPtoBM3 also gained the ability to interact with the immunity-kinase MKK2, which is required for HopAD1-dependent cell death. Thus, AvrPtoB has alternative, competing mechanisms for suppressing effector-triggered plant immunity. This approach allows the deconvolution of individual T3E activities.

INTRODUCTION Many microbial pathogens of animals and plants deploy large repertoires of effector proteins whose interplay and redundant action inside host cells enable robust subversion of immunity (Lindeberg et al., 2012; Rafiqi et al., 2012; Shames and Finlay, 2012). One approach to unraveling these complex systems is to characterize minimal subsets of effectors that can still defeat the immune system of a host. For example, the plant pathogen Pseudomonas pv. syringae pv. tomato (Pst) DC3000 injects ca. 30 effectors into plant cells via the type III secretion system (T3SS), but a subset of eight effectors is sufficient for conferring nearly full virulence in one of its hosts (Cunnac et al., 2011).

Pst DC3000 is a well-studied pathogen of tomato, Arabidopsis thaliana, and the model plant Nicotiana benthamiana (if the Pst gene for a single effector, HopQ1-1, is deleted) (Wei et al., 2007). The P. syringae type III effectors (T3Es) are designated as Hop (Hrp outer protein) or Avr (avirulence) proteins (Lindeberg et al., 2012). Successful T3E repertoires suppress pattern-triggered immunity (PTI) elicited by common bacterial factors, such as flagellin, while evading internal surveillance by plant resistance (R) proteins that activate effector-triggered immunity (ETI) (Abramovitch et al., 2006; Jones and Dangl, 2006). The ETI surveillance system is widely exploited in agriculture because introduction of a single R gene into an elite cultivar is sufficient to protect a crop if it confers detection of any pathogen effector. However, ETI-based resistance can be unstable in the field because effector redundancy allows pathogen populations to lose a betraying effector gene with little or no virulence penalty, and pathogens may acquire an effector that suppresses the ETI activity of other effectors (Abramovitch et al., 2006; Jones and Dangl, 2006). The evolution of pathosystems determined by such PTI-ETIT3E interactions has driven expansion and diversification of repertoires of pattern recognition receptors (PRRs), R proteins, and T3Es, as well as complexities in individual T3Es. For example, multiple domains in AvrPtoB have distinct roles in interactions with the host immune system, including suppression of two different PRRs and both elicitation and suppression of ETI dependent on the R proteins Pto, Fen, and Prf (Abramovitch et al., 2003; Cheng et al., 2011; Jackson et al., 1999; Rosebrock et al., 2007). The HopAB family includes AvrPtoB and VirPphA, the first T3E implicated in ETI suppression, which is now seen to be an ability of many P. syringae T3Es (Guo et al., 2009). Beyond redundancy and interplay, the activities of P. syringae T3Es are also challenging to understand because of growing evidence that an individual T3E may interact with multiple immunity-associated proteins in plants (Bogdanove and Martin, 2000; Mukhtar et al., 2011; Singh et al., 2014; Weßling et al., 2014). To facilitate study of the Pst DC3000 T3E repertoire, we previously deleted 28 well-expressed T3E genes to produce DC3000D28E and then restored avrPtoB and seven other T3E genes (chosen through a gene shuffling analysis) to the mutant, thereby revealing a minimal repertoire for growth and virulence in N. benthamiana (Cunnac et al., 2011). Importantly, although

752 Cell Host & Microbe 17, 752–762, June 10, 2015 ª2015 Elsevier Inc.

DC3000D28E appeared functionally effectorless based on several criteria, it differed from a T3SS-deficient derivative in eliciting cell death in leaves when infiltrated at very high inoculum levels (Cunnac et al., 2011). Recent work revealed that a Pst DC3000 T3SS-deficient mutant differs from the wild-type in eliciting flagellin-dependent, PTI-associated production of reactive oxygen species (ROS) in N. benthamiana leaves ca. 15 hr after inoculation, presumably because the mutant is unable to deliver PTI-suppressive T3Es (Oh et al., 2010; Wei et al., 2013). We expected that DC3000D28E would behave similarly because all known highly expressed T3E genes had been deleted. Our contrary, preliminary findings spawned this work, revealing the weakly expressed effector HopAD1 to be responsible for both the cell death and 15 hr ROS suppression responses. Further exploration of HopAD1 activity with our natural pathogen system revealed that AvrPtoB suppresses HopAD1 cell death elicitation and has evolved multiple virulence activities that interfere with each other but may enable adaptation of P. syringae T3E repertoires to different hosts. RESULTS Genes Potentially Contributing to the Residual T3SS-Dependent Phenotypes of DC3000D28E in N. benthamiana Were Identified and Deleted DC3000D28E is a polymutant deficient in the 28 T3E genes considered to be well expressed in Pst DC3000, but it retains cell death elicitation activity in N. benthamiana (Cunnac et al., 2011). Factors potentially responsible for this activity include weakly expressed T3Es whose genes were not deleted, undiscovered T3E genes, and harpins, which are extracellular T3SS components that redundantly promote T3E translocation (Kvitko et al., 2007). All known T3E genes are activated by the HrpL alternative sigma factor, and a recent ChIP-seq analysis of hrp promoters revealed PSPTO_5633, a previously uncalled ORF, to encode a T3E designated HopBM1 (Lam et al., 2014). In search of the cell death factor produced by DC3000D28E, we deleted several additional genes (Figure 1A). T3E gene cluster VIII contains six weakly expressed T3E genes and pseudogenes, which were deleted to produce DC3000D34E. hopBM1 was then deleted to produce DC3000D35E. hopAD1, previously considered functionally inactive (Chang et al., 2005), was deleted from DC3000D28E to produce DC3000D29E and from DC3000D35E to produce DC3000D36E. The genome of DC3000D36E was analyzed by Illumina sequencing as part of the study of a cryptic duplication present Figure 1. Deletion of Eight Additional T3E Genes from DC3000D28E Produces Effectorless Mutant DC3000D36E and Reveals that HopAD1 Is Responsible for the Residual Ability of DC300028E to Elicit Cell Death and to Suppress a ROS Burst and PTI in N. benthamiana (A) Overview of the Pst DC3000 T3E gene repertoire and deletions in polymutant strains. (B) ETI-like rapid, confluent plant cell death is triggered in N. benthamiana by Pst DC3000 polymutants that still carry hopAD1. Bacteria suspensions at the high inoculum level of 5 3 108 CFU/ml were infiltrated into the marked zones of N. benthamiana leaves. The leaf response was photographed 2 days later. The fraction under each image indicates the number of times that tissue death was observed relative to the number of test inoculations.

(C) ROS production is triggered in N. benthamiana by Pst DC3000 polymutants that lack hopAD1. Bacterial suspensions at 5 3 108 CFU/ml were infiltrated into leaves, and ROS production was determined 15 hr post-inoculation. Means marked with the same letter were not significantly different at the 5% confidence level on the basis of Duncan’s multiple range test; bars represent SD based on three samples from three plants. (D) PTI elicitation activity is restored to DC3000D28E derivatives by deletion of hopAD1. The fraction under each image indicates the number of times cell death was inhibited (indicating PTI elicitation by the first inoculum) relative to the number of test inoculations. All experiments were repeated three times with similar results (see also Tables S1 and S2 and Figure S1).

Cell Host & Microbe 17, 752–762, June 10, 2015 ª2015 Elsevier Inc. 753

Figure 2. Evidence that HopAD1 Is an Avirulence Determinant in N. benthamiana Based on Heterologous Expression of hopAD1 in P. syringae pv. tabaci (Pta)11528 and Deletion of hopAD1 from Pst DC3000 (A) Pta 11528 expressing HopAD1 is avirulent in N. benthamiana. Leaves were infiltrated with the indicated strains at the low level of 3 3 104 CFU/ml and photographed 8 days later. (B) Pst DC3000DhopAD1 mutants cause disease lesions in N. benthamiana. The leaves were infiltrated with the indicated strains at 3 3 104 CFU/ml using a blunt syringe and photographed 8 days after inoculation. (C) Deletion of hopAD1 from Pst DC3000 results in the same high level of growth as observed with Pst DC3000DhopQ1-1 in N. benthamiana. Bacteria were infiltrated at 3 3 104 CFU/ml and populations were measured from three 0.6-cm-diameter leaf discs 6 days after inoculation. Error bars indicate the SD of populations measured from thee leaf discs from each of three plants. Means marked with the same letter were not significantly different at the 5% confidence level on the basis of Duncan’s multiple range test. All experiments were repeated three times with similar results (see also Figures S2 and S3).

in many Pst DC3000 derivatives (Bao et al., 2014). Here, DC3000D36E was compared with the parental Pst DC3000 strain by breseq analysis (Barrick et al., 2009), which confirmed all of the predicted T3E gene deletions and identified eight spontaneous mutations, none of which would be expected to have a significant impact on virulence (Table S2). In addition, hopP1 was deleted from DC3000D34E. HopP1 has key properties both of harpins, which can elicit cell death when exogenously applied to the apoplast (Kvitko et al., 2007; Oh et al., 2007), and T3Es, which are more strongly translocated into plant cells when expressed from native promoters (Chang et al., 2005). These mutants were then assayed for loss of pathogen-related responses in N. benthamiana.

The Ability of DC3000D28E to Elicit ETI-Associated Cell Death, Suppress a ROS Burst, and Elicit Unexpectedly Weak PTI in N. benthamiana Is due Solely to HopAD1 Test bacteria were infiltrated into N. benthamiana leaves with a blunt syringe and assayed for their ability to elicit three responses (at different inoculum levels and different times post inoculation): cell death (5 3 108 CFU/ml by 2 days) (Figure 1B), ROS production (5 3 108 CFU/ml at 15 hr) (Figure 1C), and functional PTI (1 3 108 CFU/ml then challenge at 6 hr) (Figure 1D). The latter assay employs a level of DC3000D28E below the threshold for confluent cell death and is based on the observation that PTI elicited by the first inoculation blocks T3E delivery necessary for ETI cell death elicitation by challenge-inoculated wild-type Pst DC3000 (Oh et al., 2010) (Figure 1D). The T3SS-deficient DhrcQ-U mutant elicited none of the tissue collapse associated with ETI cell death, but it elicited ROS and functional PTI. DC3000D28E produced the opposite responses. The DC3000D28E phenotype was unchanged by further deletion of T3E gene cluster VIII, hopBM1, or hopP1 (Figures 1 and S1A). In contrast, deleting hopAD1 from DC3000D28E or its DVIII/ DhopBM1 derivative resulted in a loss of cell death elicitation, the appearance of ROS production, and induction of strong PTI (Figure 1). Cell death elicitation activity was restored to DC3000D29E and DC3000D36E by HopAD1 expressed from plasmid pCPP3234 (Figure S1B). Although HopAD1 strongly suppressed 15 hr ROS, it is important to note that the mechanistic basis is uncertain and could result from elicitation of incipient cell death or suppression of PTI. HopAD1 derivatives tagged with the HA epitope and the adenylate cyclase (Cya) translocation reporter were constructed and used to confirm that HopAD1 acts as a typical T3E in being secreted in culture and translocated into plant cells in a T3SSdependent manner (Figure S2). Furthermore, the absence of HopAD1 from DC3000D29E and DC3000D36E did not alter delivery of AvrPto-Cya, a model T3E for translocation studies, which further suggests that the properties of DC3000D28E we examined can be attributed solely to HopAD1 (Figure S2). The cell-death-eliciting activity of HopAD1 suggested it may be acting as an avirulence determinant in N. benthamiana. To test this notion, we expressed HopAD1 from a tac promoter and with a C-terminal Cya fusion, using pCPP3234, to replicate the approach previously used to test HopQ1-1 for avirulence activity in N. benthamiana (Wei et al., 2007). P. syringae pv. tabaci 11528 strains with hopAD1 or empty vector were infiltrated into N. benthamiana leaves at a low inoculum level (3 3 104 CFU/ml) and scored for symptom production 8 days post-inoculation. As expected, HopAD1 rendered this otherwise virulent pathovar avirulent in N. benthamiana (Figure 2A). We then deleted hopAD1 from wild-type Pst DC3000 and inoculated N. benthamiana leaves at 3 3 104 CFU/ml. Remarkably, the mutant became virulent, as indicated by symptom production and bacterial growth to population levels equivalent to that produced by a Pst DC3000DhopQ1-1 mutant or a Pst DC3000DhopQ1-1/DhopAD1 double mutant (Figures 2B and 2C). Thus, Pst DC3000 can gain virulence in N. benthamiana by loss of hopQ1-1, as previously reported (Wei et al., 2007), or by loss of hopAD1 (Figure 2). To explore the relative avirulence activity of HopAD1 and HopQ1-1, we compared the ability of functionally effectorless strains producing just these T3Es (from the tac promoter in

754 Cell Host & Microbe 17, 752–762, June 10, 2015 ª2015 Elsevier Inc.

pCPP3234) to elicit cell death in N. benthamiana at threshold levels of high inoculum. Consistent with our previous observations (Cunnac et al., 2011; Wei et al., 2007), HopAD1 cell death activity is observed only at very high inoculum levels that are 10-fold above that needed for HopQ1-1 (Figure S3). In summary, these results reveal that HopAD1 is typical of many P. syringae T3Es in having demonstrable PTI suppression and ETI elicitation abilities. Further, the interaction of Pst DC3000 with N. benthamiana has two alternative avirulence determinants.

A

B

C

Figure 3. Deletion of hopAD1 Has No Effect on the Growth or Symptom Production in N. benthamiana of DC3000D28E Derivatives Carrying T3E Gene Subsets that Include avrPtoB, and AvrPtoB Is Sufficient to Reduce Cell Death Elicitation by HopAD1 (A) Bacterial growth assay of DC3000D28E derivatives with different subsets of effectors (Table S1). White fill in the genotype grid indicates that the locus was present or restored in DC3000D28E; gray fill indicates that it was missing. Bacteria were infiltrated at the low inoculum level of 3 3 104 CFU/ml and populations in three 0.6-cm-diameter leaf discs were determined at 6 days after inoculation. Error bars indicate the SD of populations measured from three leaf discs from each of three plants. Means with the same letter were not significantly different at the 5% confidence level based on Duncan’s multiple range test. (B) Symptom production by DC3000D28E derivatives with different subsets of effectors. The indicated bacteria were inoculated as above, and leaves were photographed at 8 days post-inoculation. Asterisks indicate necrosis, plus signs indicate chlorosis, and minus signs indicate no symptoms. (C) AvrPtoB but not AvrPto suppresses cell death elicitation by HopAD1 in N. benthamiana. Leaves were infiltrated with indicated bacteria in the marked circles at the high inoculum level of 5 3 108 CFU/ml, and 2 days later they were photographed and scored for tissue collapse (unstained) or stained with trypan blue to assay cell death (stained). The fraction under each image indicates the number of times that confluent tissue collapse or trypan blue staining was observed relative to the number of test inoculations. All experiments were repeated three times with similar results (see also Figure S4).

AvrPtoB Is a Suppressor of HopAD1-Elicited Cell Death in N. benthamiana We had previously observed that DC3000D28E was able to grow substantially in N. benthamiana and produce disease symptoms following restoration of eight T3E genes: avrPtoB, hopM1, hopN1, avrE1, hopAA1-1, hopE1, hopAM1-1, and hopG1 (Cunnac et al., 2011). The virulence of this strain (CUCPB6032) was puzzling given the avirulence activity of HopAD1 described above. To further explore HopAD1 avirulence, we deleted hopAD1 from CUCPB6032 and examined bacterial growth and symptom production in N. benthamiana following a low-level inoculation. Unexpectedly, the presence of HopAD1 had no apparent effect (Figures 3A and 3B). We similarly deleted hopAD1 from a series of DC3000D28E derivatives with progressively fewer restored T3Es, including CUCPB6017, which carries just avrPtoB and hopM1. Again, no change in bacterial growth in planta was observed (Figure 3A). These observations raised the possibility that HopAD1 avirulence activity was suppressed by avrPtoB or hopM1. Given that HopM1 alone has some ability to elicit cell death in N. benthamiana (Wei et al., 2007), we predicted that AvrPtoB was more likely the suppressor of HopAD1-dependent cell death. To test this, we restored avrPtoB to its native locus in the genome of DC3000D28E. Cell death was substantially suppressed by AvrPtoB, as indicated by the lack of visible tissue collapse and trypan blue staining 2 days after inoculation of N. benthamiana leaves at the high level of 5 3 108 CFU/ml (Figure 3C). However, it is important to note that AvrPtoB suppression of HopAD1-dependent cell death is not complete, and it fails after 4 days (Figure S4A). To determine if another T3E with properties similar to AvrPtoB could also suppress HopAD1-dependent cell death, AvrPto was tested using the same procedures. AvrPto and AvrPtoB each can promote growth of DC3000D28E in N. benthamiana (Cunnac et al., 2011); they both target PRR co-receptors involved in flagellin perception, and they both trigger ETI that is dependent on the Pto decoy kinase and the Prf nucleotide-binding leucine-rich repeat protein (Martin, 2012). However, AvrPto failed to suppress HopAD1 cell death (Figure 3C), which suggests that suppression activity is not an indirect consequence of PTI suppression and is a more specific property of AvrPtoB. HopAB Family Members that Lack E3 Ubiquitin Ligase Activity More Strongly Suppress HopAD1-Dependent Cell Death The widespread HopAB family of P. syringae T3Es contains variants in three clades that have received differing levels of study (Chien et al., 2013; Jackson et al., 1999; Lin et al., 2006; Martin, 2012). We examined the relative ability of five

Cell Host & Microbe 17, 752–762, June 10, 2015 ª2015 Elsevier Inc. 755

Figure 4. AvrPtoB Homologs from Diverse P. syringae Pathovars, Expressed in DC3000D28E, Suppress HopAD1-Dependent Cell Death Elicitation in N. benthamiana, and Their Suppression Efficacy Correlates with Having an Inactive or Missing E3 Ligase Domain, Indicated by Interaction with Fen (A) Bacterial cell suspensions in MgCl2 buffer, adjusted to 5 3 108 CFU/ml, were infiltrated into leaves, and the plant response was photographed 2 days later. The fraction under each image indicates the number of times tissue collapse occurred relative to the number of test inoculations. All avrPtoB homologs (hopAB family) were expressed from the avrPto promoter in pCPP5372. The experiment was repeated three times with similar results (see also Figure S5). (B) Interaction of HopAB family members with Pto and Fen based on a yeast two-hybrid system. Blue patches indicate positive interactions.

representative HopAB family members to suppress HopAD1dependent cell death: AvrPtoB (HopAB2PtoDC3000), VirPphA (HopAB1PphISPaVe596), AvrPtoBT1 (HopAB3PtoT1), AvrPtoBB728a (HopAB1PsyB728a), and HopPmaL (HopAB3-1PmaES4326). Vector pCPP5372, which uses the avrPto promoter and generates a C-terminal HA tag, was used to express each T3E in DC3000D28E. Bacteria were inoculated at a high level (5 3 108 CFU/ml) into N. benthamiana leaves and scored for visible tissue collapse 2 days later. AvrPtoB and AvrPtoBT1 showed an incomplete ability to suppress HopAD1-dependent cell death, whereas VirPphA, AvrPtoBB728a, and HopPmaL were much more effective (Figure 4A). AvrPtoB contains N-terminal domains promoting interaction with the Pto and Fen immunity-associated kinase R proteins and a C-terminal E3 ubiquitin ligase domain that targets Fen for proteasomal degradation, thereby preventing detectable interaction in yeast two-hybrid assays (Mathieu et al., 2014; Rosebrock et al., 2007). AvrPtoBB728a has reduced E3 ubiquitin ligase activity and was unable to ubiquitinate Fen in an in vitro ubiquitination assay (Chien et al., 2013), and HopPmaL lacks the E3 ubiquitin ligase domain (Rosebrock et al., 2007). This correlation between strong HopAD1 cell death suppression and diminished E3 ubiquitin ligase activity prompted us to perform yeast two-hybrid comparisons of the interactions of these five HopAB family members with Pto and Fen. All five T3Es interacted with Pto, but interaction with Fen (and therefore apparent lack of E3 ubiquitin ligase activity) was observed only with the three HopAB proteins that had stronger HopAD1 cell death suppression activity (Figure 4B).

An AvrPtoBM3 Mutant Disrupted in Both of the Kinase Interaction Domains and in the E3 Ubiquitin Ligase Domain Strongly Suppresses HopAD1-Dependent Cell Death AvrPtoB has two kinase-interaction domains, whose residues F173 and G325 are required for interaction with Pto and Fen/ Pto, respectively (Figure 5A). The proximity of the Fen/Pto interaction site to the C-terminal E3 ubiquitin ligase domain is key to the vulnerability of these kinases to degradation (Mathieu et al., 2014). The correlations observed above suggested to us that E3 ubiquitin ligase activity interfered with suppression of HopAD1 cell death. To test that notion, we examined the ability of mutant AvrPtoBE3 (F479A, F525A, and P533A) to more completely suppress HopAD1 cell death and to interact with Fen. As expected, the mutant gained both abilities (Figures 5B and 5C). Notably, ClustalW alignment of 22 full-length HopAB family members in the P. syringae Hop database (http:// pseudomonas-syringae.org) revealed conservation in active sites F173A, G325, F479A, and P533A, but different amino acids at F525 in several, including VirPphA and AvrPtoBB728a (Figure S5). We then constructed the AvrPtoBM3 mutant (F173A, G325, F479A, F525A, and P533A), which is disrupted in all of the known functional domains in AvrPtoB. As expected, AvrPtoBM3 no longer interacts with Pto or Fen in yeast two-hybrid assays, and in repeated tests, we found it to have a stronger ability to suppress HopAD1 cell death (Figures 5B and 5C). We used two previously characterized AvrPtoB mutants to further explore structural requirements for HopAD1 cell death suppression

756 Cell Host & Microbe 17, 752–762, June 10, 2015 ª2015 Elsevier Inc.

(legend on next page)

Cell Host & Microbe 17, 752–762, June 10, 2015 ª2015 Elsevier Inc. 757

activity (Mathieu et al., 2014; Zeng et al., 2011). AvrPtoB1-307 retained suppression activity (Figure 5B), as did AvrPtoB1-387 (data not shown). We then constructed AvrPtoB1-307F173A, which also retained suppression activity (Figure 5B) but, as expected, lost interaction with Pto (Figure 5C). Thus, the ability to suppress HopAD1 cell death resides in the first 307 amino acids of AvrPtoB and is independent of all of the structure-defined functional motifs known for this model virulence protein. Expression of AvrPtoBM3 Enables Wild-Type Pst DC3000 to Become Virulent in N. benthamiana despite the Presence of hopAD1 Given our previous observation that deleting hopAD1 enabled Pst DC3000 virulence in N. benthamiana, we postulated that expression of AvrPtoBM3 in an otherwise wild-type strain could have the same effect. AvrPtoB and AvrPtoBM3 were expressed from pCPP5372 in Pst DC3000, and test bacteria were inoculated at the low level of 3 3 104 CFU/ml and then assayed for growth and symptom production. The additional copy of avrPtoB had no significant effect on the interaction; however, avrPtoBM3 promoted bacterial growth and symptom production (Figures 5D and 5E). We also asked if AvrPtoBM3 could suppress cell death elicitation by HopQ1-1 by using DC3000D29E with hopQ1-1 restored to its native locus in high-level inoculations in N. benthamiana. AvrPtoB partially and AvrPtoBM3 fully suppressed HopAD1 cell death in positive control inoculation zones while neither AvrPtoB nor AvrPtoBM3 diminished HopQ1-1 cell death in test zones (Figure S4B). These results are consistent with the apparently stronger ETI activity of HopQ1-1 (Figure S3), and they raise the question of whether the ability of AvrPtoBM3 to suppress ETI-associated cell death is limited to HopAD1. AvrPtoBM3 Suppresses the Ability of DC3000D29E Expressing AvrPto and AvrPtoB1-387 to Elicit Cell Death in N. benthamiana that Is Transgenically Producing the Tomato Pto Kinase We asked if AvrPtoBM3 could suppress the ETI-associated cell death elicited by AvrPto and AvrPtoB1-387. We performed this test using N. benthamiana plants transgenically expressing tomato 35S::Pto, which respond with ETI-associated cell death to AvrPto (Moeder et al., 2007). This cell death response is shown here by trypan blue staining 2 days after a high-level (5 3 108 CFU/ml) inoculation with DC3000D29E expressing AvrPto or AvrPtoB1-387 (Figure 6A). As expected, AvrPtoBM3 elicited no cell death in Pto+ N. benthamiana. However, AvrPtoBM3 fully

suppressed the cell death elicited by AvrPto and, remarkably, by AvrPtoB1-387 (Figure 6A). MKK2 Is Necessary for HopAD1 Cell Death Elicitation in N. benthamiana Prf and MAP kinase kinase 2 (MKK2) are known to be important for AvrPto-dependent ETI cell death (Ekengren et al., 2003; Salmeron et al., 1996). MKK2 also contributes to cell death in N. benthamiana mediated by R proteins RPS2, RPP13, Rx2, and Gpa2 that recognize effectors from multiple pathogen classes (Oh and Martin 2011), which raises the possibility of a role for this kinase in the cell death elicited by HopAD1 (cognate R protein not known). We used virus-induced gene silencing (VIGS) to silence Prf and MKK2 in N. benthamiana to assess the role of these factors in HopAD1-dependent cell death elicitation. Silencing Prf had no effect on cell death elicitation by DC3000D28E or DC3000D29E with hopQ1-1 restored to its native locus, whereas silencing MKK2 blocked cell death elicitation by DC3000D28E but had no effect with DC3000D29E with hopQ1-1 restored (Figures 6B). Identification of MKK2 as a common signaling kinase needed by AvrPto and HopAD1 suggested that MKK2 may be an interaction target for AvrPtoBM3. We accordingly used yeast two-hybrid assays to assess interactions between MKK2 and AvrPtoB derivatives. A very weak interaction was observed with AvrPtoB (seen only at late time points), but AvrPtoBE3, AvrPtoBM3, AvrPtoB1-307, and AvrPtoB1-307F173A all interacted strongly with MKK2 (Figure 5C). This interaction pattern differs notably from that observed for Pto and Fen (Figure 5C), with which AvrPtoBM3 failed to interact, and is consistent with MKK2 being the target of an additional kinase interaction domain in the first 307 amino acids of AvrPtoB (Figure 7). These observations also suggest that in yeast MKK2 may be a target of the E3 ubiquitin ligase activity of the native AvrPtoB, but interestingly, AvrPtoBM3 interactions with MKK2 without degradation are associated with more effective disruption of cell death signaling. DISCUSSION The construction and use of derivatives of Pst DC3000 expressing just two T3Es, HopAD1 and AvrPtoB, and of AvrPtoB mutants lacking all known host immune interaction motifs has uncovered an ability of AvrPtoB to interact with a conserved kinase signaling component and suppress the ETI triggered by multiple effectors. We will discuss the effectorless strain, the avirulence activity

Figure 5. AvrPtoB Mutants Show that Suppression of Cell Death Elicitation by HopAD1 Is Enhanced by Mutation of the AvrPtoB E3 Ligase Domain and Correlates with AvrPtoB Interaction with MKK2 but Not Pto or Fen, and Expression of AvrPtoBM3 in Wild-Type Pst DC3000 Promotes Bacterial Growth and Symptom Production in N. benthamiana (A) Diagram of AvrPtoB from Pst DC3000 showing the type III secretion system targeting region (T3S) and mutations ablating function of the Pto interaction domain (PID), Fen interaction domain (FID), and E3 ubiquitin ligase domain (E3). (B) Effect of mutations in AvrPtoB domains on relative ability to suppress cell death elicitation in N. benthamiana by HopAD1 in DC3000D28E. Bacteria at the high inoculum level of 5 3 108 CFU/ml were infiltrated into leaves, and the plant response was photographed 4 days later. The fraction under each image indicates the frequency of tissue collapse relative to the number of test inoculations. Similar results were found in three independent tests. (C) Interaction of AvrPtoB mutants with Pto, Fen, and MKK2 based on a yeast two-hybrid system. Blue patches indicate positive interactions. (D) Bacteria were inoculated at the low level of 3 3 104 CFU/ml in N. benthamiana, and populations were assayed 6 days later. Error bars indicate the SD of populations measured from thee leaf discs from each of three plants. Means with the same letter were not significantly different at the 5% confidence level based on Duncan’s multiple range test. (E) N. benthamiana leaves were inoculated as above and photographed 15 days later. Typical symptoms are shown, and all the experiments were repeated three times with similar results.

758 Cell Host & Microbe 17, 752–762, June 10, 2015 ª2015 Elsevier Inc.

Figure 6. Suppression of ETI by AvrPtoBM3 Action on MKK2 Is Supported by the Ability of AvrPtoBM3 to Inhibit AvrPto and AvrPtoB1-387 Cell Death Elicitation in Pto+ N. benthamiana (despite AvrPtoBM3 Lacking a Functional Pto Interaction Domain) and by the MKK2 Dependence of HopAD1 Cell Death Elicitation (A) AvrPtoBM3 suppresses cell death elicitation by avrPto+ or AvrPtoB1-387+ DC3000D29E in N. benthamiana that is transgenically producing the tomato Pto kinase. Leaves were infiltrated with indicated bacteria in the marked circles at the high inoculum level of 5 3 108 CFU/ml and stained with trypan blue at 2 days post-inoculation. The fraction under each image indicates the number of times cell death was observed relative to the number of test inoculations. (B) Cell death elicited by HopAD1 in DC300028E is compromised by silencing MKK2 but not Prf in N. benthamiana, but cell death elicited by HopQ1-1 in DC3000D29E is not suppressed by MKK2 or Prf. Bacteria at 5 3 108 CFU/ml were infiltrated into leaves, and the plant response was photographed 2 days later. The fraction under each image indicates the number of times that tissue collapse was observed relative to the number of test inoculations. Similar results were found in three independent experiments.

of HopAD1, the general phenomenon of ETI suppression, the various activities of AvrPtoB, and the evolution of T3Es with multiple domains that interfere with each other. The DC3000D36E polymutant lacks all known T3E genes, has surprisingly few spontaneous mutations in comparison with the Pst DC3000 starting parent, and is functionally effectorless. Pst DC3000 T3Es have been comprehensively sought through bio-

informatic (Lindeberg et al., 2006; Schechter et al., 2006), functional (Chang et al., 2005), and ChIP-seq screens (Lam et al., 2014), and available data suggest the repertoire of active T3E genes to be completely identified in Pst DC3000 and completely deleted in DC3000D36E. In our previously reported reassembly of a minimal functional T3E repertoire in DC3000D28E (Cunnac et al., 2011), the first effector we restored was AvrPtoB, which we now know suppresses HopAD1 ETI. HopAD1 acts as an avirulence determinant for P. syringae pv. tabaci and Pst DC3000 in N. benthamiana. The latter observation was unexpected because we had previously shown the same properties for HopQ1-1. Thus, deleting either hopQ1-1 or hopAD1 confers virulence to Pst DC3000. An intriguing puzzle is that either hopQ1-1 or hopAD1 is sufficient to trigger ETI when natively expressed in DC3000D29E or DC300028E (in the absence of other T3Es in the native repertoire) or in P. syringae pv. tabaci (in the presence of other T3Es in a heterologous repertoire), but both of these T3Es are required for functional ETI in Pst DC3000 with its full T3E repertoire. We note that the T3E repertoires of P. syringae pv. tabaci and Pst DC3000 are quite different (Baltrus et al., 2011), but the mechanism underlying this paradox awaits elucidation. The general importance of ETI suppression in enabling repertoire compatibility in the face of host ETI surveillance is an important but poorly understood phenomenon for all P. syringae pathovars. Notably, all three pathovars (out of 19 diverse P. syringae strains surveyed) that carry HopAD1 also carry a HopAB family member (Baltrus et al., 2011). Of these three, HopPmaL lacks the E3 ligase domain and HopAB3Por1_6 has the F525Y polymorphism (Figure S5B). Furthermore, many effectors can suppress ETI triggered by a test effector in heterologous systems (Guo et al., 2009). Importantly, the first evidence of an ETI suppressor functioning in its native repertoire involved the HopAB family member VirPphA: loss of virPphA ‘‘unmasked’’ a hidden avirulence activity in P. syringae pv. phaseolicola (Jackson et al., 1999). Since this pathovar lacks HopAD1 and AvrPto, VirPphA likely suppresses ETI triggered by some other T3E. Our observations support a model in which a proposed domain conferring ‘‘M3 activity’’ (functioning without the three known and mutated domains) in AvrPtoB interacts with a conserved and essential immune signaling component, MKK2 (Ekengren et al., 2003), to suppress ETI triggered by HopAD1, AvrPto, and potentially many other T3Es (Figure 7A) (Oh and Martin 2011). This model is presented in the context of host targets and AvrPtoB domain structure (Figure 7B), evolution of AvrPtoB and HopAB family members (Figure S6), and the coevolution of successive AvrPtoB domains and host targets (Figure S7). Given the central role in immunity of MKK2 relative to the more specialized and polymorphic PRRs, we propose that M3 activity was ancestral in the evolution of AvrPtoB and that subsequently acquired domains variously compete and interfere with that activity. For example, the failure of AvrPtoB1-387 to suppress cell death in Pto+ N. benthamiana is consistent with sequestration of AvrPtoB1-387 by membrane-associated Pto resulting from stronger affinity of the AvrPtoB kinase-interaction domains with a decoy kinase (Pto) than with cognate virulence target kinases (PRRs), as proposed for the evolution of decoy systems (van der Hoorn and Kamoun, 2008).

Cell Host & Microbe 17, 752–762, June 10, 2015 ª2015 Elsevier Inc. 759

Figure 7. Model of the Competing Virulence Activities of AvrPtoB and AvrPtoBM3 in the Context of Plant Immune Signaling, Relevant Activities of HopAD1 and AvrPto, and the Domain Structure of AvrPtoB (A) HopAD1 is proposed to disrupt PTI-associated kinase signaling but to be detectable by an unknown resistance protein (R1) that signals through MKK2 to ETI. AvrPtoB suppresses PTI through interaction of PID and FID domains with kinase domains of pattern recognition co-receptors (PRRs) FLS2, BAK1, and Bti9, but recognition by Fen/Pto decoy kinases leads to ETI unless blocked by E3 ligase action at the FID domain. The M3 mutant of AvrPtoB, lacking kinase-interaction and E3 ligase domains, has enhanced ability to interact with MKK2 and suppress ETI elicited by HopAD1, AvrPto, and AvrPtoB1-387. High polymorphism in the E3 ligase domain of HopAB family members suggests that these effectors may rapidly evolve between states favoring E3 ligase activity (for suppression of FID-Fen-triggered ETI) or loss of such activity (for suppression of ETI triggered by HopAD1 and other effectors that signal through MKK2) depending on the host R gene composition. (B) Depiction of the AvrPtoB activities described above in terms of AvrPtoB domains and interacting host proteins highlights the conflicting roles of the M3 activity (within gray), the PID/FID domains, and the E3 ubiquitin ligase activity. Outcomes involving different AvrPtoB variants and R proteins have been previously reported (black font) (Mathieu et al., 2014; Rosebrock et al., 2007) or shown here (blue font). Variants Ba–Bc depict the conflicting activity of the E3 ubiquitin ligase domain in suppression of intramolecular (self domain) ETI versus ETI elicited by one or more other effectors signaling through MKK2. Variants Bc–Be address the ability of AvrPtoBM3 (Be) to suppress Ptodependent ETI elicitation by AvrPtoB1-387 (Bd), which demonstrates that PID/ FID-dependent interaction with Pto (but not with PRRs) conflicts with the ability of AvrPtoB variants to suppress MKK2-dependent ETI elicitation (see also Figures S6 and S7).

The more widespread and important interference we observed involves activity of the E3 ligase domain, which is notably polymorphic in HopAB family members (Figure S5). We propose that the basis for that polymorphism is the benefit that unimpeded M3 activity may confer for ‘‘other-effector’’ ETI suppression in plants where the absence of Fen obviates ‘‘self-domain’’ ETI suppression (Figure 7B). A pivotal role of AvrPtoB E3 ubiquitin ligase polymorphism in determining P. syringae infection outcomes has experimental support. The failure of AvrPtoBB728a to suppress self-domain ETI underlies the inability of P. syringae pv. syringae B728a to cause disease in tomato VFNT Cherry, which expresses Fen resistance (Chien et al., 2013), and we have shown here that failure of AvrPtoB to suppress other-effector ETI is a determinative factor in the inability of Pst DC3000 to cause disease in N. benthamiana (Figure 5). The activities of AvrPtoB discovered in this study exemplify challenges in understanding the operation of individual T3Es and their functional domains in the face of many competing activities and host targets. Importantly, T3Es have a high degree of intrinsic disorder, which suggests the possibility of partnerinduced folding and interaction with a wide array of host proteins, and notable here is that AvrPtoB shows intrinsic disorder outside of the PID, FID, and E3 ligase domains (Marı´n and Ott, 2014). Furthermore, a comprehensive yeast two-hybrid analysis revealed the capacity of many Pst DC3000 T3Es to physically interact with multiple plant proteins (Mukhtar et al., 2011), and pertinent to this work, several Pst DC3000 T3Es interacted with multiple kinases in split-luciferase and protein array assays (Singh et al., 2014). Given the potential for promiscuous or secondary interactions when present at high levels, it is not surprising that AvrPtoB produces stronger plant responses when highly produced in planta than when delivered (at far lower levels) via the T3SS of various P. syringae strains (Abramovitch et al., 2003; de Torres et al., 2006; Lin et al., 2006). Pst DC3000D36E offers many advantages for systems analysis of the functions of AvrPtoB and other T3Es, which include (i) effector delivery via natural pathways at physiological levels; (ii) quantitative readouts for susceptibility, PTI, and ETI; (iii) efficient systems for reassembly of repertoire subsets; (iv) extensive data on natural variation in individual T3Es and whole repertoires; and (v) experimentally amenable hosts. These features of the system enabled us to exploit quantitative differences in the ability of AvrPtoB and AvrPtoBM3 to suppress ETI and, in so doing, to identify an additional activity of AvrPtoB and a potential mechanism for adaptation of the HopAB family to differences in plant immune surveillance. EXPERIMENTAL PROCEDURES Bacterial and Plant Material Bacterial strains and plasmids used in this study are listed in Table S1. Bacteria were grown with antibiotic selection, as needed, as previously described (Wei et al., 2013). N. benthamiana plants were grown in a greenhouse with 16 hr light/8 hr dark, 65% humidity, and a temperature of 24 C during daylight and 22 C at night. DNA Manipulations and Construction of Pst DC3000 Mutants DNA manipulations and analysis and bacterial mutant constructions via pK18mobsacB or pT18mobsacB constructs were done essentially as

760 Cell Host & Microbe 17, 752–762, June 10, 2015 ª2015 Elsevier Inc.

described (Wei et al., 2013). PCR products were cloned into a pENTR/SDTOPO or pENTR/D-TOPO vector and sequenced at the Cornell BioResource Center with an Applied BioSystems 3730xl DNA Analyzer. Entry clones were transferred into appropriate destination vectors, as indicated in Table S1. Each resulting construct was conjugated into P. syringae strains by triparental mating, using pRK2013.

Received: November 20, 2014 Revised: March 5, 2015 Accepted: April 17, 2015 Published: June 10, 2015

Assays for Challenge-Inoculation Cell Death and ROS Production Fresh streaks of Pst DC3000 mutants were made from isolated colonies and grown overnight on KB plates. Challenge inoculum cell death assays for functional PTI were initiated by infiltrating N. benthamiana leaves with 1 3 108 CFU/ml of the test Pst DC3000 mutant strains. After 6 hr, an overlapping inoculation of 5 3 107 CFU/ml of the cell-death-inducing strain Pst DC3000 was made. The presence or absence of confluent cell death in the overlapping region was evaluated 2 days after Pst DC3000 infiltration by directly visualizing leaf collapse or staining with trypan blue, as described (He et al., 2004; Wei et al., 2013). For ROS measurements of P. syringae strains, freshly cultured bacterial suspensions were infiltrated into 5-week-old N. benthamiana leaves at 5 3 108 CFU/ml. At 15 hr post-inoculation, leaf disks were excised and analyzed for ROS production as previously described (Wei et al., 2013).

Abramovitch, R.B., Kim, Y.J., Chen, S., Dickman, M.B., and Martin, G.B. (2003). Pseudomonas type III effector AvrPtoB induces plant disease susceptibility by inhibition of host programmed cell death. EMBO J. 22, 60–69.

Immunoblot Analysis and Cya Translocation Reporter Assays HopAD1 protein in cell and supernatant fractions was analyzed essentially as described (Oh et al., 2007). Samples were separated by electrophoresis on SDS-PAGE and analyzed by immunoblotting using primary anti-HA mouse monoclonal immunoglobulin G (IgG) antibodies and secondary anti-mouse IgG alkaline phosphatase conjugate antibodies, as described (Wei et al., 2013). Translocation assays also were performed as described (Wei et al., 2013), based on cyclic AMP (cAMP) levels determined with a Correlate-EIA cAMP immunoassay kit (Enzo). Yeast Two-Hybrid and VIGS Assays The LexA yeast two-hybrid system was used to investigate the interactions of different forms of AvrPtoB with Pto, Fen, and MKK2 (Mathieu et al., 2014). ORFs were inserted into pEG202 (bait) or pJG4-5 (prey) vectors and introduced into Saccharomyces cerevisiae strain EGY48 containing the reporter plasmid pSH18-34. Yeast cells containing both prey and bait proteins were streaked on medium containing galactose and 5-bromo-4-chloro-3-indolylb-D-galactopyranoside (X-Gal). Photographs were taken after 30–48 hr incubation of the plates at 30 C, and representative images from at least three independent experiments with similar results are shown. VIGS was performed as previously described (Oh and Martin 2011). The lack of off-targets for the MKK2-silencing construct was confirmed with the SGN VIGS Tool (Fernandez-Pozo et al., 2015). The efficacy of the identical NtMKK2 construct has been shown previously using RT-PCR and cell death suppression assays in N. benthamiana (Liu et al., 2004). SUPPLEMENTAL INFORMATION Supplemental Information includes seven figures and two tables and can be found with this article online at http://dx.doi.org/10.1016/j.chom.2015.05.007. AUTHOR CONTRIBUTIONS H.-L.W., G.B.M., and A.C. conceived the work and planned experiments; H.-L.W., S.C., J.M., and T.C.H. performed experiments; H.-L.W., S.C., P.S., B.S., G.B.M., and A.C. analyzed data; H.-L.W., G.B.M., and A.C. wrote the paper. ACKNOWLEDGMENTS We thank Diane Dunham and Paige Reeves for their assistance in the greenhouse and Noe Fernandez for assistance with the SGN VIGS Tool. Robert B. Abramovitch and Chang-Sik Oh performed preliminary experiments on MKK2. This work was supported by National Science Foundation grant IOS-1025642 (A.C. and G.B.M.) and by National Institutes of Health grant R01-GM078021 (G.B.M.). The authors declare no conflict of interest.

REFERENCES

Abramovitch, R.B., Anderson, J.C., and Martin, G.B. (2006). Bacterial elicitation and evasion of plant innate immunity. Nat. Rev. Mol. Cell Biol. 7, 601–611. Baltrus, D.A., Nishimura, M.T., Romanchuk, A., Chang, J.H., Mukhtar, M.S., Cherkis, K., Roach, J., Grant, S.R., Jones, C.D., and Dangl, J.L. (2011). Dynamic evolution of pathogenicity revealed by sequencing and comparative genomics of 19 Pseudomonas syringae isolates. PLoS Pathog. 7, e1002132. Bao, Z., Stodghill, P.V., Myers, C.R., Lam, H., Wei, H.-L., Chakravarthy, S., Kvitko, B.H., Collmer, A., Cartinhour, S.W., Schweitzer, P., and Swingle, B. (2014). Genomic plasticity enables phenotypic variation of Pseudomonas syringae pv. tomato DC3000. PLoS ONE 9, e86628. Barrick, J.E., Yu, D.S., Yoon, S.H., Jeong, H., Oh, T.K., Schneider, D., Lenski, R.E., and Kim, J.F. (2009). Genome evolution and adaptation in a long-term experiment with Escherichia coli. Nature 461, 1243–1247. Bogdanove, A.J., and Martin, G.B. (2000). AvrPto-dependent Pto-interacting proteins and AvrPto-interacting proteins in tomato. Proc. Natl. Acad. Sci. USA 97, 8836–8840. Chang, J.H., Urbach, J.M., Law, T.F., Arnold, L.W., Hu, A., Gombar, S., Grant, S.R., Ausubel, F.M., and Dangl, J.L. (2005). A high-throughput, near-saturating screen for type III effector genes from Pseudomonas syringae. Proc. Natl. Acad. Sci. USA 102, 2549–2554. Cheng, W., Munkvold, K.R., Gao, H., Mathieu, J., Schwizer, S., Wang, S., Yan, Y.B., Wang, J., Martin, G.B., and Chai, J. (2011). Structural analysis of Pseudomonas syringae AvrPtoB bound to host BAK1 reveals two similar kinase-interacting domains in a type III Effector. Cell Host Microbe 10, 616–626. Chien, C.F., Mathieu, J., Hsu, C.H., Boyle, P., Martin, G.B., and Lin, N.C. (2013). Nonhost resistance of tomato to the bean pathogen Pseudomonas syringae pv. syringae B728a is due to a defective E3 ubiquitin ligase domain in avrptobb728a. Mol. Plant Microbe Interact. 26, 387–397. Cunnac, S., Chakravarthy, S., Kvitko, B.H., Russell, A.B., Martin, G.B., and Collmer, A. (2011). Genetic disassembly and combinatorial reassembly identify a minimal functional repertoire of type III effectors in Pseudomonas syringae. Proc. Natl. Acad. Sci. USA 108, 2975–2980. de Torres, M., Mansfield, J.W., Grabov, N., Brown, I.R., Ammouneh, H., Tsiamis, G., Forsyth, A., Robatzek, S., Grant, M., and Boch, J. (2006). Pseudomonas syringae effector AvrPtoB suppresses basal defence in Arabidopsis. Plant J. 47, 368–382. Ekengren, S.K., Liu, Y., Schiff, M., Dinesh-Kumar, S.P., and Martin, G.B. (2003). Two MAPK cascades, NPR1, and TGA transcription factors play a role in Pto-mediated disease resistance in tomato. Plant J. 36, 905–917. Fernandez-Pozo, N., Rosli, H.G., Martin, G.B., and Mueller, L.A. (2015). The SGN VIGS tool: user-friendly software to design virus-induced gene silencing (VIGS) constructs for functional genomics. Mol. Plant 8, 486–488. Guo, M., Tian, F., Wamboldt, Y., and Alfano, J.R. (2009). The majority of the type III effector inventory of Pseudomonas syringae pv. tomato DC3000 can suppress plant immunity. Mol. Plant Microbe Interact. 22, 1069–1080. He, X., Anderson, J.C., del Pozo, O., Gu, Y.Q., Tang, X., and Martin, G.B. (2004). Silencing of subfamily I of protein phosphatase 2A catalytic subunits results in activation of plant defense responses and localized cell death. Plant J. 38, 563–577. Jackson, R.W., Athanassopoulos, E., Tsiamis, G., Mansfield, J.W., Sesma, A., Arnold, D.L., Gibbon, M.J., Murillo, J., Taylor, J.D., and Vivian, A. (1999). 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 96, 10875–10880.

Cell Host & Microbe 17, 752–762, June 10, 2015 ª2015 Elsevier Inc. 761

Jones, J.D., and Dangl, J.L. (2006). The plant immune system. Nature 444, 323–329. Kvitko, B.H., Ramos, A.R., Morello, J.E., Oh, H.-S., and Collmer, A. (2007). Identification of harpins in Pseudomonas syringae pv. tomato DC3000, which are functionally similar to HrpK1 in promoting translocation of type III secretion system effectors. J. Bacteriol. 189, 8059–8072. Lam, H.N., Chakravarthy, S., Wei, H.L., BuiNguyen, H., Stodghill, P.V., Collmer, A., Swingle, B.M., and Cartinhour, S.W. (2014). Global analysis of the HrpL regulon in the plant pathogen Pseudomonas syringae pv. tomato DC3000 reveals new regulon members with diverse functions. PLoS ONE 9, e106115. Lin, N.C., Abramovitch, R.B., Kim, Y.J., and Martin, G.B. (2006). Diverse AvrPtoB homologs from several Pseudomonas syringae pathovars elicit Ptodependent resistance and have similar virulence activities. Appl. Environ. Microbiol. 72, 702–712. Lindeberg, M., Cartinhour, S., Myers, C.R., Schechter, L.M., Schneider, D.J., and Collmer, A. (2006). Closing the circle on the discovery of genes encoding Hrp regulon members and type III secretion system effectors in the genomes of three model Pseudomonas syringae strains. Mol. Plant Microbe Interact. 19, 1151–1158. Lindeberg, M., Cunnac, S., and Collmer, A. (2012). Pseudomonas syringae type III effector repertoires: last words in endless arguments. Trends Microbiol. 20, 199–208. Liu, Y., Schiff, M., and Dinesh-Kumar, S.P. (2004). Involvement of MEK1 MAPKK, NTF6 MAPK, WRKY/MYB transcription factors, COI1 and CTR1 in N-mediated resistance to tobacco mosaic virus. Plant J. 38, 800–809. Marı´n, M., and Ott, T. (2014). Intrinsic disorder in plant proteins and phytopathogenic bacterial effectors. Chem. Rev. 114, 6912–6932. Martin, G.B. (2012). Suppression and activation of the plant immune system by Pseudomonas syringae effectors AvrPto and AvrPtoB. In Effectors in PlantMicrobe Interactions, F. Martin and S. Kamoun, eds. (Wiley-Blackwell), pp. 123–154. Mathieu, J., Schwizer, S., and Martin, G.B. (2014). Pto kinase binds two domains of AvrPtoB and its proximity to the effector E3 ligase determines if it evades degradation and activates plant immunity. PLoS Pathog. 10, e1004227. Moeder, W., Del Pozo, O., Navarre, D.A., Martin, G.B., and Klessig, D.F. (2007). Aconitase plays a role in regulating resistance to oxidative stress and cell death in Arabidopsis and Nicotiana benthamiana. Plant Mol. Biol. 63, 273–287. Mukhtar, M.S., Carvunis, A.R., Dreze, M., Epple, P., Steinbrenner, J., Moore, J., Tasan, M., Galli, M., Hao, T., Nishimura, M.T., et al.; European Union Effectoromics Consortium (2011). Independently evolved virulence effectors converge onto hubs in a plant immune system network. Science 333, 596–601. Oh, C.S., and Martin, G.B. (2011). Tomato 14-3-3 protein TFT7 interacts with a MAP kinase kinase to regulate immunity-associated programmed cell death mediated by diverse disease resistance proteins. J. Biol. Chem. 286, 14129– 14136.

Oh, H.-S., Kvitko, B.H., Morello, J.E., and Collmer, A. (2007). Pseudomonas syringae lytic transglycosylases coregulated with the type III secretion system contribute to the translocation of effector proteins into plant cells. J. Bacteriol. 189, 8277–8289. Oh, H.-S., Park, D.H., and Collmer, A. (2010). Components of the Pseudomonas syringae type III secretion system can suppress and may elicit plant innate immunity. Mol. Plant Microbe Interact. 23, 727–739. Rafiqi, M., Ellis, J.G., Ludowici, V.A., Hardham, A.R., and Dodds, P.N. (2012). Challenges and progress towards understanding the role of effectors in plantfungal interactions. Curr. Opin. Plant Biol. 15, 477–482. Rosebrock, T.R., Zeng, L., Brady, J.J., Abramovitch, R.B., Xiao, F., and Martin, G.B. (2007). A bacterial E3 ubiquitin ligase targets a host protein kinase to disrupt plant immunity. Nature 448, 370–374. Salmeron, J.M., Oldroyd, G.E.D., Rommens, C.M.T., Scofield, S.R., Kim, H.-S., Lavelle, D.T., Dahlbeck, D., and Staskawicz, B.J. (1996). 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 86, 123–133. Schechter, L.M., Vencato, M., Jordan, K.L., Schneider, S.E., Schneider, D.J., and Collmer, A. (2006). Multiple approaches to a complete inventory of Pseudomonas syringae pv. tomato DC3000 type III secretion system effector proteins. Mol. Plant Microbe Interact. 19, 1180–1192. Shames, S.R., and Finlay, B.B. (2012). Bacterial effector interplay: a new way to view effector function. Trends Microbiol. 20, 214–219. Singh, D.K., Calvino, M., Brauer, E.K., Fernandez-Pozo, N., Strickler, S., Yalamanchili, R.D., Suzuki, H., Aoki, K., Shibata, D., Stratmann, J.W., et al. (2014). The tomato kinome and the tomato kinase library ORFeome: novel resources for the study of kinases and signal transduction in tomato and solanaceae species. Mol. Plant Microbe Interact. 27, 7–17. van der Hoorn, R.A., and Kamoun, S. (2008). From Guard to Decoy: a new model for perception of plant pathogen effectors. Plant Cell 20, 2009–2017. Wei, C.-F., Kvitko, B.H., Shimizu, R., Crabill, E., Alfano, J.R., Lin, N.-C., Martin, G.B., Huang, H.-C., and Collmer, A. (2007). A Pseudomonas syringae pv. tomato DC3000 mutant lacking the type III effector HopQ1-1 is able to cause disease in the model plant Nicotiana benthamiana. Plant J. 51, 32–46. Wei, H.-L., Chakravarthy, S., Worley, J.N., and Collmer, A. (2013). Consequences of flagellin export through the type III secretion system of Pseudomonas syringae reveal a major difference in the innate immune systems of mammals and the model plant Nicotiana benthamiana. Cell. Microbiol. 15, 601–618. Weßling, R., Epple, P., Altmann, S., He, Y., Yang, L., Henz, S.R., McDonald, N., Wiley, K., Bader, K.C., Gla¨ßer, C., et al. (2014). Convergent targeting of a common host protein-network by pathogen effectors from three kingdoms of life. Cell Host Microbe 16, 364–375. Zeng, L., Velasquez, A.C., Munkvold, K.R., Zhang, J., and Martin, G.B. (2011). A tomato LysM receptor-like kinase promotes immunity and its kinase activity is inhibited by AvrPtoB. Plant J. 69, 92–103.

762 Cell Host & Microbe 17, 752–762, June 10, 2015 ª2015 Elsevier Inc.