Our laboratory studies the biochemical mechanisms used by bacterial pathogens to alter plant physiology during infection. Extensive genetic and phenotypic data indicate that the bacterial type three secretion (T3S) system and its protein substrates (referred to as T3S effectors) are the major virulence determinants that promote pathogen colonization in plants. The paradigm for T3S effector function has been that these proteins collectively suppress host defense responses to promote colonization and disease progression. The biological function(s) of most T3S effectors, however, is extremely limited and biochemical support for this paradigm is lacking. Thus, the goal of our research has been to elucidate T3S effector function, identify host targets, and provide fundamental knowledge of how perturbation of host signaling pathways leads to bacterial pathogenesis. To do so, we study the T3S effectors in Xanthomonas euvesicatoria (Xcv), the causal agent of leaf spot disease in tomato.
Our research focuses on the characterization of three Xcv T3S effectors: XopN, XopD, and AvrBsT. We discovered that these effectors modulate three distinct nodes of defense signal transduction, supporting the paradigm that T3S effectors encode defense suppressors. In addition, our work challenges this paradigm by demonstrating that one suppresses disease symptom development, thus illuminating the importance of tolerance promoting factors in bacterial-plant interactions. Highlights from our published research include:
I. Identifying new Xanthomonas T3S effector proteins translocated into plant cells during infection, leading to the discovery of the virulence factor – XopN.
II. Demonstrating that XopN suppresses PAMP-triggered immunity (PTI) and physically interacts with a tomato atypical receptor kinase and a 14-3-3 protein TFT1, both linked to PTI in tomato.
I. Demonstrating that XopD is a plant-specific SUMO protease that alters host transcription, promotes pathogen growth, and suppresses host defense responses at the late stages of tissue colonization by targeting the ethylene-associated transcription factor SlERF4 in tomato.
IV. Identifying new proteins that control phospholipid signaling and disease resistance responses following AvrBsT perturbation.
I. Isolation and characterization of translocated T3S effector proteins in Xcv. Initially, we thought it was critical to identify the repertoire of T3S effectors used by Xcv to manipulate tomato physiology. We developed a sensitive reporter assay and a genetic screen to isolate Xcv T3S effectors. The advantage of this screen was that: (1) only T3S effectors translocated into plant cells were isolated and (2) novel effectors unlinked to the characterized T3S pathogenicity island were discovered. We isolated 8 translocated Xcv T3S effectors. Four of these effectors are conserved in all Xanthomonas spp, indicating that they define a “core” group of proteins important for Xanthomonas pathogenesis. The others are not present in most Xanthomonas species suggesting that they might have been introduced into Xcv by horizontal gene transfer via interactions on the same host. This work contributed significantly to the discovery of bonafide Xcv T3S effectors, a proteome now predicted to contain ~35 proteins. By characterizing the “core” effectors, we discovered that one protein, XopN, contributed significantly to Xcv growth and symptom production. In light of XopN’s unique effector features, wide conservation among Xanthomonas strains, and importance in Xcv virulence, we prioritized our work on the dissection of XopN’s role in Xcv pathogenesis.
II. XopN is a suppressor of PTI and interacts with signaling proteins. The challenge of getting at effector function is the ability to uncover unique or conserved structural folds associated with known biological activity. We discovered a conserved structural fold in XopN revealing that it might function as a protein scaffold within the plant cell. XopN is predicted to have 7 anti-parallel, tandem alpha-helical repeats that stack one on top of the other to form a solenoid. Proteins with such a fold are known to assemble components of a pathway to quantitatively control signal inputs and outputs. We therefore predicted that XopN’s virulence role might be to bind components of the basal defense machinery to modulate signal output. Our phenotypic and biochemical data support such a role for XopN inside infected plant cells.
Specifically, we found that XopN action suppresses the magnitude of PTI responses in tomato and Arabidopsis. XopN reduces callose deposition and the early induction of defense gene expression in response to pathogen infection. Biochemical studies indicate that XopN is dampening basal immunity by binding to a novel tomato atypical receptor kinase (TARK1) and a pathogen-inducible 14-3-3 phospho-binding protein (TFT1). Analysis of TARK1 revealed that this putative receptor is required to restrict Xcv growth and symptom development, underlining its importance in PTI in tomato. Moreover, mutations that abolish XopN-TARK1 physical interaction also reduce XopN virulence in tomato. Taken together, these data provide the basis for a model in which XopN is altering TARK1-dependent signaling events at the plant plasma membrane to suppress basal defense during Xcv infection. This is the first report of a Xcv T3S effector disrupting the early stages of basal defense signaling.
Current work is aimed to elucidate the precise role of TARK1 in PTI signaling and how XopN physically interferes with signal integration at the plasma membrane. Our most recent work indicates that TARK1's stability is regulated during infection. We are now focusing our attention on the composition and regulation of a putative TARK1 complex at the plasma membrane during Xcv infection. We are also studying the role of TARK1 in plant development since transgenic lines with reduced TARK1 expression exhibit developmental defects that lead to organ fusion.
III. XopD is a SUMO protease that alters transcription and symptom development. Early in 2002, we discovered that XopD shares remarkably similarity with eukaryotic ubiquitin-like proteases. Biochemical characterization of XopD in vitro and in planta confirmed that the protein encodes an active cysteine protease with plant-specific SUMO specificity. This finding provided the first biochemical evidence demonstrating a functional role for a T3S effector in planta. Moreover, these studies revealed a novel mechanism used by Xanthomonas, and possibly other plant-associated microbes, to modulate plant physiology during infection.
How XopD ultimately alters plant physiology was a mystery until we looked at the late stages of Xcv infection. We made the striking observation that XopD proteolysis in tomato leaves correlates with plant tolerance (i.e., the ability of the host to cope with bacterial colonization). XopD-dependent delay of tissue degeneration correlates with reduced chlorophyll loss, reduced salicylic acid levels and changes in senescence and defense gene expression despite high pathogen titers. Thus, XopD action promotes Xcv multiplication while suppressing leaf symptom development. These findings fundamentally change the way in which we think about the factors that control plant disease symptoms. That is, disease symptoms in leaves, to a large extent, reflect cellular events orchestrated by the host not the pathogen. Our data shows that at the late stages of Xcv infection, the host alters cellular metabolism to get rid of the diseased organ. By contrast, the pathogen employs a tolerance-promoting factor, XopD, to suppress these responses. In light of this work, we now hypothesize that additional tolerance-promoting factors in bacteria play central roles in host resistance during pathogen colonization.
The localization of XopD to subnuclear foci within the plant nucleus suggested that XopD is altering host transcription during infection. Indeed, XopD modulation of host transcription during bacterial colonization correlates with tolerance to Xcv infection. Structure-function analysis revealed that XopD is also a DNA-binding protein containing two conserved EAR-motifs. These domains are found in repressors that negatively regulate defense and stress responses. Both SUMO protease activity and the EAR motifs are required for XopD-dependent modulation of defense- and senescence-associated gene transcription and for the suppression of disease symptom development. Our working model is that XopD is altering host transcription by deSUMOylating transcription factors whose SUMOylation state dynamically changes during pathogen attack. Consistent with this hypothesis, we recently discovered that XopD controls the SUMOylation state and stability of the tomato ethylene response factor SlERF4, a transcription factor that positively regulates ethylene-associated transcription at the late stages of infection.
The current goal of this project is to characterize other transcription factors that are regulated in a XopD-dependent manner and to elucidate the mechanisms of transcriptional repression in plants using XopD as a molecular probe.
IV. Molecular dissection of Xcv AvrBsT-dependent resistance in Arabidopsis. A third aspect of our research is to elucidate how plants resist bacterial infection. We exploited the natural variation that exists in Arabidopsis to identify ecotypes that are resistant and susceptible to Xanthomonas strains expressing AvrBsT. We discovered that most Arabidopsis ecotypes are susceptible to strains expressing AvrBsT. Only the Pi-0 ecotype recognizes AvrBsT action and activates a resistance response that inhibits pathogen growth. This screen hinted to us that defense against AvrBsT elicitation might not be controlled by a classic R protein, as suggested by conventional wisdom. Rather, we suspected that resistance in Pi-0 might directly reflect AvrBsT’s biological activity. Our most recent data strongly supports this model.
By positional cloning, we demonstrated that Pi-0 plants are resistant to pathogens expressing AvrBsT due to a recessive, loss of function mutation in a highly conserved carboxylesterase. We designated this gene as SOBER1 (suppressor of AvrBsT elicited resistance) because it plays a critical role in the suppression of AvrBsT-elicited changes in plant physiology during infection. SOBER1 belongs to a family of enzymes known to hydrolyze acylated proteins and lysophospholipids. Acylated proteins play key roles in the regulation of signal transduction complexes at the plasma membrane whereas lysophospholipids act as lipid second messengers at low concentrations and as membrane detergents at high concentrations.
The requirement for SOBER1 enzyme activity to suppress AvrBsT-triggered defense responses led us to hypothesize that AvrBsT action within the plant cell might generate specific lipid signal(s) that trigger stress responses that lead to defense. We also hypothesized that SOBER1 substrates might be the precursors for such lipid signals. Indeed, lipid profiling revealed that Pi-0 leaves infected with bacteria expressing AvrBsT accumulated higher levels of phosphatidic acid (PA) compared to Pi-0 leaves infected with bacteria lacking AvrBsT. PA is an important lipid second messenger associated with stress responses in plants. High PA levels correlated with low phosphatidylcholine (PC) levels in Pi-0 leaves, hinting that SOBER1 substrates might be PC and/or lysophosphatidylcholine (LysoPC). We are currently assessing the affinity of SOBER1 for these lipid species in vitro.
PA pools in cells are directly and indirectly regulated by the hydrolytic activity of phospholipase D (PLD) and phospholipase C (PLC), respectively. Lipase inhibition studies revealed that only PLD activity is required for AvrBsT-dependent defense responses in infected Pi-0 leaves. This work suggests that AvrBsT action leads to PLD-dependent hydrolysis of PC to generate high levels of PA in Pi-0 sober1 leaves. It also pinpoints PA as the key lipid signal responsible for AvrBsT-dependent defense responses in plant cells lacking SOBER1 activity. Thus, SOBER1 and PLD appear to compete for lipid substrates, highlighting the importance of both lipases in the control of phospholipid homeostasis in cells in response to plant stress.
An important lesson that emerged from this project is that highly conserved lipases fine-tune changes in lipid signaling caused by AvrBsT perturbation during infection. Resistance in Pi-0 therefore appears to be due to an imbalance in lipid homeostasis and not the evolution of a new R protein specificity. This work highlights the roles of SOBER1 and PLD in the maintenance of lipid homeostasis as well as the importance of PA as a stress signal in plant cells.
Current work is aimed at eludicating how AvrBsT biochemically alters lipid signaling and defense responses during infection. Recent work indicates that AvrBsT is an acetyltransferase that targets a novel microtubule-associated protein required for PTI and ETI.