Skip to main content
  • Loading metrics

Protein-protein interactions in the RPS4/RRS1 immune receptor complex


Plant NLR (Nucleotide-binding domain and Leucine-rich Repeat) immune receptor proteins are encoded by Resistance (R) genes and confer specific resistance to pathogen races that carry the corresponding recognized effectors. Some NLR proteins function in pairs, forming receptor complexes for the perception of specific effectors. We show here that the Arabidopsis RPS4 and RRS1 NLR proteins are both required to make an authentic immune complex. Over-expression of RPS4 in tobacco or in Arabidopsis results in constitutive defense activation; this phenotype is suppressed in the presence of RRS1. RRS1 protein co-immunoprecipitates (co-IPs) with itself in the presence or absence of RPS4, but in contrast, RPS4 does not associate with itself in the absence of RRS1. In the presence of RRS1, RPS4 associates with defense signaling regulator EDS1 solely in the nucleus, in contrast to the extra-nuclear location found in the absence of RRS1. The AvrRps4 effector does not disrupt RPS4-EDS1 association in the presence of RRS1. In the absence of RRS1, AvrRps4 interacts with EDS1, forming nucleocytoplasmic aggregates, the formation of which is disturbed by the co-expression of PAD4 but not by SAG101. These data indicate that the study of an immune receptor protein complex in the absence of all components can result in misleading inferences, and reveals an NLR complex that dynamically interacts with the immune regulators EDS1/PAD4 or EDS1/SAG101, and with effectors, during the process by which effector recognition is converted to defense activation.

Author summary

Paired NLR immune receptors have evolved in both plants and animals to enable host cells to detect intracellular pathogen effectors or ligands. Generally, one of the two NLR proteins in the two-component immune complex acts as a sensor that enables effector detection, while the helper (or executor) NLR activates an immune response, presumably via conformational changes. However, the mechanisms by which effector recognition activates NLR protein pairs to trigger downstream signal activation remain poorly understood. We suggest here that the Arabidopsis RPS4/RRS1 receptor forms hetero-oligomeric complexes in the nucleus both in the pre- and post-activation states. In contrast, some animal NLRs make oligomers solely upon detecting pathogen ligands. We propose that the RPS4/RRS1 pre-activation immune complex in the nucleus is converted to the activated state via changes in inter- and intra-molecular interactions upon effector recognition. RPS4/RRS1 associates with the EDS1/PAD4 complex, which is likely to be involved in downstream signaling, both before and after activation. Recognition of AvrRps4 or PopP2 by RPS4/RRS1 in association with EDS1/PAD4 does not disrupt this immune complex, but likely provokes conformational changes within it.


Plants and animals have evolved an effective immune system that uses both cell surface and intracellular receptors to detect pathogen invasion and then activate defense mechanisms [14]. Plant Resistance (R) genes mostly encode intra-cellular nucleotide-binding, leucine-rich repeat immune receptors (NLRs) that resemble similar receptors found in mammals (NLRs) [5, 6]. Most plant NLRs carry either a Toll, Interleukin-1 Receptor, Resistance protein (TIR) domain or a Coiled-coil (CC) domain at their N-termini [7, 8]. Plant NLRs directly or indirectly detect specific pathogen-derived “avirulence” (avr) effector proteins and activate effector-triggered immunity (ETI), which restricts the growth and spread of pathogens [9]. How plant NLR proteins activate defense upon effector recognition is poorly understood.

Plant NLRs localize to various subcellular compartments. For example, Arabidopsis Resistance to Pseudomonas maculicola 1 (RPM1), a CC-type NLR (CNL), localizes at the plasma membrane [10]. The potato Rx protein, a typical CNL protein that confers resistance to Potato Virus X, shows a nucleocytoplasmic localization, and both nuclear and cytoplasmic pools are required for full defense activation [11]. Interestingly, several plant NLRs have been shown to localize to the nucleus and directly associate with transcription-regulated proteins for disease resistance activation [1214]. For instance, suppressor of npr1-1 constitutive 1 (SNC1), which is a TIR-NLR (TNL), localizes to both cytosol and nucleus [15]. However, SNC1 function likely requires nuclear localization because of the direct interaction between SNC1 and the transcriptional co-repressor Topless-related 1 (TPR1). This interaction might indirectly regulate transcriptional reprogramming via Histone deacetylase 19 (HDA19) [15, 16]. Nuclear localization of the tobacco N and Resistance to Pseudomonas syringae 4 (RPS4) proteins is also essential for function [17, 18]. Upon effector (a viral helicase) recognition, the N protein might function in part by interactions with the transcription factor, squamosa promoter-binding protein-like 6 (SPL6) to initiate disease resistance signaling via transcriptional reprogramming [19]. Furthermore, both SNC1 and RPS4 genetically and physically interact with helix-loop-helix (bHLH) type transcription factor (TF), bHLH84 [14].

The flax (Linum usitatissimum) L6 is a typical TNL protein that directly recognizes variants of the biotrophic flax rust fungus (Melampsora lini) effector AvrL567 [20]. Transient expression of the L6 TIR domain alone is sufficient for activation of defense without effector recognition [21, 22]. RPS4 was first reported as a disease-resistance gene in Arabidopsis that specifies recognition of and response to Pseudomonas syringae effector AvrRps4 [23]. Furthermore, over-expression of full length RPS4 in tobacco induces an AvrRps4-independent Hypersensitive cell death Response (HR). Similarly, RPS4 TIR domain over-expression results in AvrRps4-independent HR induction, probably via TIR-TIR self-association [24, 25]. An interface between RPS4 and Resistance to Ralstonia solanacearum 1 (RRS1) TIR domains was revealed by X-ray crystallography [21, 25]. TIR-TIR domain interactions could play a major role in activation of cell death/resistance.

RPS4 function requires the genetically adjacent RRS1 gene, which encodes an atypical TNL with a C-terminal WRKY DNA binding domain [2628]. RPS4 and RRS1 comprise a two-component plant immune receptor complex, which recognizes AvrRps4 of P. syringae, the acetyltransferase PopP2 of Ralstonia solanacearum and an unknown effector of Colletotrichum higginsianum [26, 29, 30]. Expression of the RPS4 and RRS1 genes is regulated by a shared promoter, which indicates that both proteins are likely to be co-expressed at comparable levels in Arabidopsis. Two distinct alleles of RRS1 have been described. The RRS1-R allele recognizes AvrRps4 and PopP2, and carries a 101 amino acid C-terminal extension after the WRKY domain. In contrast, the RRS1-S allele that recognizes AvrRps4 but not PopP2 has only an 18 amino acid C-terminal extension after the WRKY domain. Furthermore, the addition of specific C-terminal extra amino acids converts RRS1-S to RRS1-R [31]. AvrRps4 interacts with, and PopP2 acetylates, the RRS1 WRKY domain, resulting in activation of the RPS4/RRS1 complex and defense induction [31, 32]. These findings suggest that RPS4/RRS1 is a two-component immune complex in which one of the two NLR proteins has an integrated domain that enables the plant to detect effectors which target that domain, consistent with the "integrated decoy" model for the evolution of two-component immune complexes [33]. Downstream signaling upon activation of RPS4/RRS1 remains poorly understood. RPS4 TIR domain-mediated HR activation can be suppressed by co-expression with the TIR domain of RRS1 [25]. However, the autoimmune phenotype of the RRS1 auto-active mutant allele, sensitive to low humidity 1 (slh1), is RPS4-dependent in Arabidopsis, as well as in N. tabacum transient assays [27, 34, 35]. Many other NLR gene pairs have been identified in both plants and animals that confer resistance to pathogens [26, 3641].

Enhanced disease susceptibility 1 (EDS1) encodes a lipase-homologous nucleo-cytoplasmic defense regulator protein essential for resistance conditioned by TNLs [42]. EDS1 is reported to associate with some TNL proteins such as RPS4, SNC1, and RPS6 (Resistance to P. syringae 6) [43]. EDS1 is functional only in conjunction with other lipase-like proteins, encoded by either phytoalexin deficient 4 (PAD4), or senescence-associated gene 101 (SAG101) [44]. One group reported that AvrRps4 and HopA1 effector proteins alter RPS4-EDS1 or RPS6-EDS1 association [43] and two groups reported that AvrRps4 directly interacts with EDS1, using in vivo co-immunoprecipitation (co-IP) and in vitro pull-down assays [43, 45]. It was also reported, using Bimolecular Fluorescence Complementation (BiFC) in N. benthamiana leaves and co-IP assays, that EDS1 forms cytoplasmic protein complexes with the TNL proteins RPS4 or RPS6, while the cognate bacterial effectors AvrRps4 and HopA1 disrupt these EDS1 complexes [43]. Other groups reported an inability to reproduce AvrRps4/EDS1 associations in co-IP and yeast two-hybrid (Y2H) experiments [46], perhaps indicating that any such interactions are indirect.

To attempt to resolve some paradoxes and inconsistencies in the literature regarding the nature of the RPS4/RRS1 complex and the roles of its components upon effector recognition, we used BiFC and co-IP to investigate the properties of immune complexes involving RPS4/RRS1 and EDS1/PAD4/SAG101. We found that RPS4 protein does not self-associate in the absence of RRS1, and that the previously reported RPS4 autoimmunity in tobacco and Arabidopsis is suppressed when co-expressed with RRS1. Likewise, although we could reproduce observations of RPS4 association with EDS1 in the cytoplasm in the absence of RRS1, RPS4/EDS1 association is nuclear localized when RRS1 protein is present. These data strongly emphasize the need to study RPS4 and RRS1 proteins together and not separately. Our findings suggest the existence of a nuclear-localized, complex that involves RPS4, RRS1, EDS1 and PAD4, in which these components remain present before and after recognition of AvrRps4 and PopP2 effectors. We infer that the RPS4/RRS1 immune complex undergoes dynamic intra- and inter-molecular protein-protein and domain-domain interactions to activate immune responses upon recognition of effector proteins.


RPS4 auto-immunity is attenuated by RRS1 and RPS4 stabilization is RRS1-dependent

The oligomerization of NLRs is often required for R protein function [8]. We previously reported that the association of RRS1 and RPS4 TIR domains and their dimerization are important for defense activation and cell death signaling [25]. However, the TIR-dimerization domain mutants of RPS4 and RRS1 still co-immunoprecipitate (co-IP), which indicates that other domains contribute to this interaction.

Transient overexpression of RPS4 alone leads to the activation of an effector-independent HR in N. tabacum leaves [17, 24] and this autoimmune phenotype is abolished in both P-loop (RPS4K242A) and TIR-TIR dimerization (RPS4SH/AA) mutants (Fig 1A) [25]. Importantly, co-expression of RRS1-3xHis-6xFLAG (HF) with RPS4-HA results in abolition of the RPS4-dependent HR in tobacco leaves (Fig 1A).

Fig 1. RPS4 auto-immunity is repressed by RRS1 and RRS1 increases RPS4 protein accumulation.

(A) Transient overexpression of RPS4-HA results in auto-immunity in tobacco leaves but not RRS1-R-HF. The P-loop mutant (RPS4K242A) and the TIR-domain dimerization mutant (RPS4SH/AA) abolish RPS4-dependent auto-immunity. Co-expression of RPS4-HA with RRS1-R-HF blocks HR induction in tobacco leaves. (B) Stunting and dwarf phenotype of Arabidopsis transgenic line stably overexpressing RPS4 is attenuated by crossing with RRS1-R transgenic Arabidopsis line. The 35S::RPS4-HS / 35S::RRS1-R-HF transgenic line was generated by crossing the line expressing the 35S::RPS4-HS with the transgenic line with the 35S::RRS1-R-HF. Images were taken with 4-week-old plants grown in short-day conditions at 22°C. Scale bar = 1.0 cm. (C) Quantification of rosette diameters at 4-week-old of the lines in (B). The leaf diameter was calculated from the plant rosette area measured in ImageJ. One-way ANOVA was used to calculate the statistical significance between genotypes, as indicated by different capital letters (P < 0.001). Bars represent mean ± SD (n = 40). (D) Fractionation of protein extracts show that RPS4 is stabilized by RRS1 in nucleus and cytoplasm. RPS4-Myc and RPS4-HA were transiently co-expressed in the presence or absence of RRS1-HF in N. benthamiana leaves. At 2 dpi, samples were harvested and then fractionated by the percoll-sucrose gradient method. Western blot analysis was performed with anti-FLAG, anti-Myc and anti-HA antibodies. Anti-PEPC was used as a cytosolic marker and anti-histone H3 was used as a nuclear marker. (E) RPS4-HA protein accumulation is increased by RRS1-HF expression. RPS4-HA and GUS-HF or RPS4-HA and RRS1-HF constructs in Agrobacterium tumefaciens were infiltrated into N. benthamiana leaves. A. tumefaciens cells were adjusted to the OD600 of 0.5 for RPS4-HA or 0.1 for RRS1-HF and GUS-HF constructs. After 2 dpi, samples were harvested and Western blots were performed using anti-FLAG and anti-HA antibodies. All the experiments were repeated three times with similar results.

Arabidopsis lines overexpressing RPS4 show constitutive defense activation giving rise to growth retardation and autoimmune phenotypes [17]. However, when such lines are crossed to RRS1-R-HF-overexpressing lines, their stunted phenotype is suppressed (Fig 1B and 1C) and constitutive PR1 protein accumulation is also abolished (S1A Fig). In the heteromeric RPS4/RRS1 complex, RPS4 activation only occurs upon interactions between an effector and the RRS1 WRKY domain [31, 32].

Accumulation of NLR proteins is tightly regulated, often by F-box proteins or HSP90 chaperones, and over-accumulation of many NLRs triggers an autoimmune phenotype, but the molecular mechanism of R protein complex regulation remains largely unknown [15, 17, 24, 47]. We investigated whether RRS1 protein could affect the accumulation of RPS4. We transiently co-expressed RPS4-Myc and RPS4-HA with or without RRS1-HF in N. benthamiana leaves. After fractionation, RPS4 protein accumulation was detected using different antibodies. Consistently, RPS4-Myc and RPS4-HA protein levels were significantly increased in the presence of RRS1 in both cytosolic and nuclear fractions (Fig 1D). To confirm this result and to check protein accumulation, we carried out Western blot analysis using HF- and HA- tagged RRS1 and RPS4 proteins, respectively. The co-expression of RRS1 and RPS4 leads to approximately 3.5 times more RPS4 protein, compared to the protein levels when co-expressed with GUS (Fig 1E). The stabilization of RPS4 protein by RRS1 was also confirmed using the RPS4/RRS1-R transgenic Arabidopsis plants (S1B Fig). Conceivably, reduced RPS4 accumulation in the absence of RRS1 could be due to reduced Agrobacterium T-DNA transfer as a consequence of the defense activation by RPS4. To test this, we evaluated GFP accumulation following transient co-expression of 35S::GFP and 35S::RPS4 in the presence or absence of 35S::RRS1-R in N. benthamiana leaves. GFP accumulation was indistinguishable in the presence or absence of RRS1 (S1C Fig).

RRS1 enables RPS4 to co-IP with itself

We investigated localization of YFP-RPS4 in the presence of GUS-HF or RRS1-R-HF. In the absence of RRS1, the YFP-RPS4 signal is mostly seen in the nucleus, with a stronger signal in the nucleolus (S2A Fig). In the presence of RRS1-R-HF, RPS4 is also mostly seen in the nucleus but not in the nucleolus. With GUS-HF or with RRS1-R-HF, a weak YFP-RPS4 signal is also visible in the cytosol (S2B Fig).

We infer that the functional RPS4/RRS1 complex is primarily in the nucleus. Based on previous results [25, 31], we tested RPS4 and RRS1 homo- and hetero-dimeric interactions using the BiFC assay in N. benthamiana. In the absence of RRS1, no signal is observed from co-expression of cCFP-RPS4 and nVenus-RPS4 (Fig 2A). However, in the presence of RRS1, a strong RPS4 BiFC signal is seen in the nucleus. (Fig 2A). Similarly, when RPS4 is co-expressed with two different epitope tags, differently tagged RPS4 molecules co-IP with each other only in the presence of RRS1 (Fig 2B). In contrast, RRS1 can self-associate in the absence of RPS4 protein (Fig 2C). We verified these results with co-IPs using different combinations of tagged RPS4 and RRS1 proteins (S3A and S3B Fig).

Fig 2. RPS4 homodimerization is dependent on RRS1.

(A) BiFC assays using nVenus- and cCFP-tagged RPS4 reveal that RPS4 self-association in the nucleus is RRS1-dependent. The nVenus-RPS4, cCFP-RPS4, and mCherry were transiently co-expressed in the presence of RRS1-HF or GUS-HF in N. benthamiana leaves. At 2 dpi, the reconstruction YFP signal is observed with confocal microscope (Leica SP5). mCherry was used as a nuclear and cytoplasmic marker. Scale bar = 10 μm. (B) Co-immunoprecipitation (co-IP) assays reveal that RPS4 self-associates only in the presence of RRS1. Agrobacterium-mediated transient co-expression of RRS1-GFP/RPS4-HF/RPS4-HA or GFP/ RPS4-HF/RPS4-HA was performed in N. benthamiana leaves. Anti-FLAG co-IPs were performed with total protein extracts and probed with anti-GFP, -FLAG, and -HA antibodies. (C) Co-IPs show that RRS1 self-associates and forms a heteromeric complex with RPS4. Transient co-expression assays of RRS1-GFP/RRS1-HF, RRS1-GFP/RPS4-HF or GFP/RRS1-HF were performed in N. benthamiana leaves. Immunoblots show the presence of proteins in total extracts (input) and after immunoprecipitation with anti-GFP beads (IP-GFP). All the experiments were repeated at least three times with similar results.

Nuclear localization of RPS4/EDS1 is enhanced by RRS1

EDS1 acts as an important regulator of TNL-mediated resistance [48]. Nuclear accumulation of EDS1 is essential for TNL-mediated resistance and transcriptional activation of defense genes during ETI [42]. It has been reported that EDS1 is recruited by and physically associates with several TNL proteins such as RPS4, RPS6, and SNC1 [43]. EDS1 was reported to interact with RPS4 and other NLRs and form complexes mainly localized to punctate spots in the cytoplasm [43]. We investigated whether RRS1 could affect the cytoplasmic association of EDS1 and RPS4 [43]. To address this question, we first used BiFC assays in N. benthamiana. In the absence of RRS1, we detected reciprocal BiFC interactions of nVenus-RPS4 and cCFP-EDS1, localized to punctate spots in the cytoplasm (Fig 3A), similar to previous reports [43, 45]; this signal sometimes appeared to be adjacent to the nucleus. We also observed nuclear localization and aggregations in the cytoplasm (Fig 3A). Importantly, co-expression of RRS1-HF with nVenus-RPS4/cCFP-EDS1 abolished the cytoplasmic signal and resulted in a nuclear-localized interaction (Fig 3A). This indicates that RRS1 is enhancing RPS4/EDS1 nuclear localization in the plant cell nucleus. Furthermore, when RPS4-HA was coexpressed with RRS1-HF and GFP-EDS1, EDS1 co-IPs with RRS1-HF, suggesting that EDS1 associates with the RPS4/RRS1 complex (Fig 3B). We transiently co-expressed RRS1-HF or RPS4-HF with GFP-EDS1 or GFP in N. benthamiana leaves, and tested for co-IP. Both RRS1-HF and RPS4-HF proteins co-IP with EDS1 (S4 Fig), so the co-IP of EDS1 with RRS1 in Fig 3B could be via direct association with RRS1, and/or indirectly via association with RPS4.

Fig 3. RRS1 promotes association of RPS4 and EDS1 in the nucleus.

(A) In the presence of RRS1, the RPS4/EDS1 are predominantly localized to the nucleus. BiFC assays with the co-expression of nVenus-RPS4/cCFP-EDS1/GUS-HF/mCherry reveal reconstruction of YFP signal in the cytoplasmic aggregations and in the nucleus (arrows). In the presence of RRS1-HF, nVenus-RPS4/cCFP-EDS1 association revealed a YFP signal in the nucleus. Scale bar = 10 μm. (B) EDS1 associates with RPS4/RRS1. Upon transient co-delivery of RPS4-HA and RRS1-HF with GFP-EDS1 or GFP in N. benthamiana leaves, samples were harvested at 2 dpi and total extracts were immunoprecipitated with anti-GFP beads. Specific protein-protein interactions were detected by immunoblotting with the indicated antibodies. All the experiments were repeated at least three times with similar results.

In the presence of RRS1, RPS4/EDS1 interactions are unaltered by AvrRps4

Both EDS1 and PAD4 are required for defense activation by the RPS4/RRS1 complex upon effector recognition [49]. Previous reports describe the disruption of RPS4/EDS1 association by the AvrRps4 and HopA1 effectors [43]. Based on this, and our findings describing the essential role of RRS1 in authentic complex formation, we investigated if RPS4/RRS1 is able to form a complex with EDS1/PAD4 in planta and if the RPS4/RRS1/EDS1/PAD4 complex is disrupted by AvrRps4 or PopP2. To address these questions, we co-expressed 35S::RRS1-HF, 35S::RPS4-HA, 35S::EDS1-V5, 35S::PAD4-HA with 35S::AvrRps4-GFP, 35S::PopP2-GFP or 35S::GFP (as a negative control) in N. benthamiana leaves. Using anti-FLAG beads to select for RRS1-HF, we efficiently pulled down RPS4, EDS1 and PAD4 (Fig 4A), suggesting that EDS1/PAD4 associates with the RPS4/RRS1 complex. However, no significant disruption of this association was observed upon co-expression with AvrRps4, compared to GFP as a negative control (Fig 4A). This indicates that AvrRps4 does not affect RPS4/EDS1 association in the presence of RRS1 and PAD4. We did not observe association between PopP2 and the RPS4/RRS1/EDS1/PAD4 complex (Fig 4A) [25].

Fig 4. AvrRps4 and PopP2 do not disrupt the EDS1/PAD4/RPS4/RRS1 complex.

(A) Anti-FLAG immunoprecipitation of RRS1-HF, RPS4, EDS1 and PAD4 in the presence and absence of AvrRps4 or PopP2. Samples were prepared from transiently co-expressed RRS1-HF, RPS4-HA, EDS1-V5 and PAD4-HA in the presence of AvrRps4-GFP, PopP2-GFP or GFP in N. benthamiana. (B) Both AvrRps4 and PopP2 associate with RPS4/RRS1/EDS1/PAD4. To confirm effector protein association with a putative RPS4/RRS1/EDS1/PAD4 complex, samples were prepared from N. benthamiana leaves transiently co-expressing RRS1-HF, RPS4-Myc, EDS1-V5 and PAD4-HA in presence of AvrRps4-GFP, PopP2-GFP or GFP. Total extracts were immunoprecipitated with anti-GFP beads followed by immunoblotting with the indicated antibodies. AvrRps4C represents processed AvrRps4 C-terminus. All the experiments were repeated at least three times with similar results.

To further test this different protein/protein interactions, we co-expressed 35S::RRS1-HF, 35S::RPS4-Myc, 35S::EDS1-V5, 35S::PAD4-HA with 35S::AvrRps4-GFP, or 35S::PopP2-GFP, or 35S::GFP in N. benthamiana and IP-ed with anti-GFP beads. Consistent with our previous observations, RPS4/EDS1 association is not disrupted by AvrRps4 in the presence of RRS1 and PAD4, while both AvrRps4 and PopP2, but not GFP, successfully pulled down all the components of the complex (Fig 4B). These findings together indicate that RPS4/RRS1 can associate with EDS1/PAD4 and this association is not disrupted by AvrRps4 or PopP2 effectors.

To further test these inferences, we used BiFC assays. We co-expressed cCFP-EDS1, nVenus-RPS4, and RRS1-R-HF with AvrRps4-mCherry or a non-functional mutant AvrRps4E187A-mCherry in N. benthamiana and observed indistinguishable YFP signals in the nucleus using the same microscope settings (S5 Fig). This suggests that AvrRps4 has no significant effect on RPS4/EDS1 association in the presence of RRS1 and these components co-localize in the nucleus.

To further verify dynamic interactions between RPS4, EDS1, and PAD4 in the presence of RRS1, we carried out multi-color BiFC analysis [50]. We co-expressed nCerulean-RPS4, cCFP-EDS1, nVenus-PAD4, and RRS1-HF with AvrRps4-mCherry or AvrRps4E187A-mCherry. A strong YFP signal indicating association between cCFP-EDS1 and nVenus-PAD4 is found in both the cytosol and nucleus (S6A and S6B Fig). The CFP signal was observed exclusively in the nucleus, indicating that the association between cCFP-EDS1 and nCerulean-RPS4 is mainly nuclear (S6A and S6B Fig), consistent with Fig 3A. These BiFC data suggest that localization and interaction of EDS1/PAD4 are not significantly affected by the RPS4/RRS1 complex. Furthermore, similar patterns of YFP or CFP signals in both cytosol and nucleus were observed in the presence of AvrRps4 or AvrRps4E187A mutant (S6A and S6B Fig), suggesting that the RPS4/RRS1 immune complex with EDS1/PAD4 localizes mainly to the nucleus in both pre- and post-activation states.

EDS1 interactions with AvrRps4 are blocked by PAD4 but not SAG101

Physical interaction between EDS1 and AvrRps4 has been reported using in vitro GST pull-down and co-IP assays [43, 45]. In contrast, our group previously reported no interaction between EDS1/AvrRps4 via yeast two-hybrid and co-IP assays [46]. Since EDS1 is a crucial immune signaling component, involved in several TNL-mediated defense responses [42, 48], we further examined EDS1/AvrRps4 interactions. First, we examined whether EDS1 associates with AvrRps4 in planta, using differentially tagged EDS1 constructs. We expressed N- or C-terminally Myc-tagged EDS1 proteins (35S::Myc-EDS1 or 35S::EDS1-Myc) with 35S::GFP or 35S::AvrRps4-GFP or 35S::PAD4-GFP in N. benthamiana leaves. Following co-IP with anti-Myc beads, both AvrRps4 and PAD4 proteins could be detected with anti-GFP (S7A and S7B Fig). EDS1 associates with AvrRps4 and AvrRps4E187A (Fig 5A). However, in reciprocal co-IP tests, the AvrRps4-GFP protein did not associate with EDS1-Myc or Myc-EDS1 proteins (S7A and S7B Fig), consistent with our previous data [46]. In these assays, we used PAD4 as a positive control, which strongly associates with EDS1 in both anti-GFP beads and anti-Myc IPs (S7A and S7B Fig). To further investigate EDS1/AvrRps4 interactions, we used BiFC assays. Co-expression of AvrRps4-cCFP with nVenus-EDS1, but not with nVenus-PAD4, gives strong nucleocytoplasmic YFP signal (S8 Fig). We also observed small aggregated foci in the cell periphery (S8 Fig). These data demonstrate that AvrRps4 can associate with the immune regulator EDS1 in planta.

Fig 5. PAD4 attenuates EDS1/AvrRps4 association.

(A) The EDS1/PAD4 complex strongly reduced EDS1/AvrRps4 co-immunoprecipitation in planta. EDS1-Myc or EDS1-Myc/PAD4-HA were transiently co-expressed with AvrRps4-GFP, AvrRps4E187A-GFP or GFP in N. benthamiana leaves. Immunoprecipitations were performed using anti-Myc agarose beads and then analyzed by immunoblot with the indicated antibodies. AvrRps4C represents processed AvrRps4 C-terminus. (B) The EDS1/SAG101 complex can associate with AvrRps4. HA-EDS1 or HA-EDS1 with SAG101-Myc were transiently co-expressed with AvrRps4-GFP or GFP in N. benthamiana leaves. Immunoprecipitation was performed using anti-HA agarose beads and then analyzed by immunoblot with the indicated antibodies. (C) BiFC analysis reveals that EDS1/AvrRps4 interaction is reduced in the presence of PAD4-HA but not of SAG101. Cytoplasmic aggregations are reduced in the presence of PAD4. BiFC assays were performed by co-expression of the indicated proteins in N. benthamiana. Images were obtained at 2 dpi. The experiment was repeated three times with similar results. Scale bar = 10 μm.

EDS1 protein forms a heterodimeric complex with the lipase-like proteins PAD4 (in both cytosol and nucleus) and SAG101 (solely in the nucleus) [44, 51]. To test whether AvrRps4 and EDS1 can still interact in the presence of PAD4 or SAG101, we investigated the effect of PAD4 and SAG101 on AvrRps4/EDS1 association. As shown in Fig 5A, we found that PAD4 inhibits EDS1 association with both AvrRps4 and AvrRps4E187A, suggesting that EDS1/PAD4 hetero-dimerization might block the EDS1 surface that interacts with AvrRps4. Since PAD4 suppresses EDS1/AvrRps4 association, we then investigated whether SAG101 could also block EDS1/AvrRps4 association. We co-expressed 35S::HA-EDS1 and 35S::AvrRps4-GFP with 35S::SAG101-Myc or 35S::GFP control. The co-IP results indicate that EDS1/AvrRps4 interaction was not significantly altered by SAG101 co-expression (Fig 5B).

We also investigated the effect of PAD4 and SAG101 on AvrRps4/EDS1 association using BiFC assays. We co-expressed AvrRps4-cCFP and nVenus-EDS1 with GUS (negative control), or PAD4-HA or SAG101-Myc. Confocal microscopy indicates a specific inhibition of AvrRps4/EDS1 cytoplasmic aggregation formation by PAD4, but not by SAG101. Weak BiFC signals are detected in the cytosol and nucleus (Fig 5C). The AvrRps4KRVY/AAAA inactive mutant and EDS1 also associate in the presence of SAG101 but not PAD4 in these BiFC assays (S9 Fig). These results suggest that PAD4 specifically inhibits EDS1/AvrRps4 association when transiently co-expressed in N. benthamiana.


Plant and animal NLRs show similar domain architectures, but do they function via similar mechanisms? In two recent studies, a paired animal immune receptor structure was investigated by cryo-electron microscopy [52, 53]. The NLR family apoptosis inhibitory proteins (NAIPs) confer pathogen perception and NLR family CARD-containing protein 4 (NLRC4) acts as an adapter to activate innate immunity via formation of an inflammasome upon bacterial ligand recognition [52, 53]. Similarly, plants carry paired NLRs. Some of these paired plant NLRs have evolved diverse ‘integrated domains’ (IDs) on one of the paired receptors. These IDs act as a sensor to detect pathogen effectors, interacting with an executor (helper) NLR to activate defense upon effector recognition [31, 33, 36, 5355]. Conceivably, activation mechanisms of both animal and plant NLRs might involve oligomerization or homo-/heteromeric complex formation [5]. The Arabidopsis dual NLR receptor complex RPS4/RRS1 confers resistance to multiple bacterial pathogens and the fungal pathogen C. higginsianum [26, 29, 30]. Both are required for defense signaling, and form hetero-oligomers [25, 27]. Several functional studies on RPS4 without RRS1 have been reported based on the autoimmune activity of RPS4 [17, 43, 45]. In this study, we provide insight into RPS4/RRS1 protein-protein interactions in the pre-activation and post-activation states.

Several studies demonstrate the autoimmunity of RPS4 using Agrobacterium-based transient overexpression assays in tobacco, or stable Arabidopsis transgenic plants overexpressing RPS4 [17, 24, 56]. It has been suggested that RPS4-mediated HR activation could be the result of homodimerization of the TIR domain, since mutations in TIRSH/AA of RPS4 TIR domain or full length RPS4 prevent HR induction (Fig 1A) [24, 25]. Our findings reveal that effector-independent RPS4 autoimmunity is strongly attenuated by RRS1, both in tobacco and in Arabidopsis transgenic plants (Fig 1A and 1B). In the dual CC-NLR receptor complex comprising RGA4 and RGA5 from rice, the RGA4 autoimmunity phenotype is suppressed by RGA5 [36] indicating that sensor NLRs might act as negative regulators in multiple paired NLR systems. In addition, RRS1 overexpression in both transient tobacco system and Arabidopsis stable transgenic plants does not trigger any HR or basal defense response (Fig 1A and S1A Fig), suggesting that the primary function of the integrated domain of RRS1 is to monitor the presence of effectors. These findings indicate that RRS1 functions prior to effector perception as a negative regulator of the immune complex [56].

Homo-multimerization of RPS4 could be sufficient for cell death initiation. However, we did not see RPS4 protein homo-multimers in co-IP and BiFC experiments in N. benthamiana. RPS4-RPS4 TIR homodimerization is required to activate RPS4-mediated HR in tobacco and the RPS4-TIR domain has a self-association surface [25]. The Arabidopsis TNL protein Recognition of Peronospora parasitica 1 (RPP1) shows self-association that involves multiple domain-domain interactions [57]. Most NLR proteins show homodimeric interactions [5], suggesting that RPS4 might also make a homodimer through TIR-TIR domain or other domain-domain interactions. Conceivably, RPS4 protein levels in our experiments are not sufficient to reveal homodimerization through co-IP or BiFC assays in the absence of RRS1 because RRS1 increases RPS4 protein accumulation. This may also indicate that a certain threshold of RPS4 protein accumulation is required for self-association, and that the act of initiating defense is associated with reduced stability of the initiating protein. RRS1 can self-associate without RPS4, and RPS4 homodimerization as a part of the RPS4/RRS1 is indistinguishable in the pre- and post-activation states. RRS1 could act as a platform that enables the correct assembly of the RPS4/RRS1 complex in the nucleus, and RPS4/RRS1 might thus form a higher-level complex comprising at least 2 RRS1 and 2 RPS4 protein molecules. RGA4/RGA5 forms a homo-/heteromeric complex in the absence of effector. However, RGA4 forms homodimers in the absence of RGA5 [36]. The behaviour of dual NLR protein complexes may vary between NLR receptors. Interestingly, effector-dependent self-association was observed in tobacco N protein upon Tobacco mosaic virus replicase recognition [58] and RPP1 upon Hyaloperonospora arabidopsis effector ATR1 recognition [56]. Other CC-type NLRs such as MLA, RPS5, and Rp1-D21 showed effector-independent self-association [5, 59, 60].

Nuclear localization and accumulation of RPS4, RRS1, and AvrRps4 are essential for an effective immune response [17, 30, 45]. Different NLR proteins are localized in various subcellular compartments in their resting states. Upon recognition of effector, some NLRs exhibit dynamic relocalization [3, 8]. It was reported that the complex between RPS4 and EDS1 mainly localizes to punctuate structures in the cytoplasm and is observed in the nucleus in steady-state or upon AvrRps4 recognition [43]. We repeated RPS4/EDS1 association assays in the presence of RRS1. Using BiFC as well as biochemical analyses, we found that in the presence of RRS1, RPS4 associates with EDS1 mainly in the nucleus (but not in the nucleolus), in both resting and activated states (Fig 3A and S6 Fig). Furthermore, based on our co-IP and BiFC data, we observed that RRS1 associates with EDS1 and RPS4, and these interactions may reflect conserved associations between TNLs and EDS1 [43].

We observed stabilization of RPS4 by RRS1 in both transient expression assays and stable transgenic Arabidopsis lines. However, using co-IP and BiFC assays, we could not find any difference in RPS4 protein accumulation between RPS4/RRS1 and RPS4/RRS1/effector combinations, suggesting that RRS1 acts as a modulator of RPS4 activity regulation via inter/intra-molecular protein-protein interactions and conformational changes.

In a previous study, it was reported that the AvrRps4 protein disrupts RPS4 association with EDS1 though interaction with EDS1, as a virulence function of AvrRps4 [43, 45] indicating that EDS1, one of the main modulator of TNL-mediate immune response, could be a target of pathogen effectors. However, these findings were obtained in the absence of RRS1, and we suggest that any such findings are misleading and do not reflect the properties or location of the authentic complex. Following these results, we investigated the disruption of RPS4/EDS1 association in the presence of RRS1 by effectors, AvrRps4 and PopP2. In both co-IP and multi-color BiFC experiments, we observed that in the presence of RRS1, AvrRps4 does not affect RPS4/EDS1 association (S5 Fig). Furthermore, RPS4, EDS1 and PAD4 continue to associate in the presence of RRS1 upon co-expression with AvrRps4 or PopP2 (Fig 4 and S6 Fig). We infer that direct interaction of AvrRps4 or PopP2 with the RRS1 WRKY domain causes RPS4/RRS1/EDS1/PAD4 complex activation, but not complex dissociation. Other host components could be, and are likely to be, associated with the RPS4/RRS1 complex in both the pre- and post- activation states.

Although we previously reported an inability to replicate this association [46], we report here that by testing more combinations of epitope tags on EDS1 and AvrRps4, we could show co-IP of these two proteins. Specifically, both N-terminally (Myc-EDS1) and C-terminally (EDS1-Myc) tagged EDS1 proteins can co-IP with AvrRps4-GFP in the IP-Myc but not IP-GFP condition (S7A and S7B Fig). We also found EDS1 and AvrRps4 associate using BiFC analysis (S8 Fig). Here our findings prove the importance of carrying out Co-IP experiments in both directions with differentially (N- or C-terminally) tagged proteins in order to avoid potential experimental artefacts. EDS1 usually makes heteromeric complexes with other lipase-like proteins, PAD4 or SAG101 in plant [44]. Importantly, we found that PAD4, but not SAG101, inhibits EDS1/AvrRps4 association, presumably via its strong affinity with EDS1 (Fig 5A and 5B). Similarly, BiFC signals observed with AvrRps4/EDS1 as cytoplasmic aggregates were specifically reduced by PAD4 co-expression but not with SAG101 (Fig 5C). Why SAG101 does not disrupt the association between AvrRps4 and EDS1, but PAD4 does, remains puzzling and requires further investigation.

Overall, our study reveals the necessity of studying proteins that are members of protein complexes in the presence of their interacting components in order to avoid misleading results. Furthermore, a significant challenge remains to address the RPS4/RRS1 conformational changes and domain/domain interactions in resting and activated states. The RPS4/RRS1 nuclear complex pre- and post-activation states are currently indistinguishable via cytology and biochemistry. Defining the dynamic changes that occur in RPS4/RRS1 upon effector recognition remains an interesting and important challenge.

Materials and methods

Plant materials and Agrobacterium-mediated transient transformation

Nicotiana benthamiana and N. tabacum plants were grown in long day conditions at 24°C [31]. Agrobacterium-mediated transient transformation assay has been described [31]. Arabidopsis plants were grown in short day conditions, at 22°C. 35S::RPS4-HA-StrepII (HS) Arabidopsis transgenic plant has been described [17]. 35S::RRS1-R-HF construct [31] was transformed into Arabidopsis Col-0 with the floral-dip method as described previously [61]. Homozygous 35S::RRS1-R-HF plants were crossed to 35S::RPS4-HS to generate double overexpression lines.

Confocal microscopy analysis

The BiFC assay is as described previously [50, 62]. BiFC constructs using the C/N-terminal fragment of cyan fluorescent protein (cCFP) and N-terminal fragment of Cerulean (nCerulean); nCerulean-RPS4, AvrRps4-cCFP. These BiFC constructs were transformed to Agrobacterium tumefaciens (strain GV3101 or Agl1). Overnight cultures of A. tumefaciens cells were collected by centrifugation at 3000 rpm for 10 min. Collected cells were resuspended in Agro-infiltration buffer (10 mM MES-KOH, pH5.7 10 mM MgCl2). A. tumefaciens cells were adjusted to the OD600 of 0.5 the constructs and were transiently co-expressed in the presence of RRS1-HF or GUS-HF or AvrRps4-mCheery or AvrRps4E187A-mCherry constructs in Agrobacterium tumefaciens were infiltrated into N. benthamiana leaves. After 2 dpi, the reconstruction signals are observed with a Leica DM6000B/TCS SP5 confocal microscope (Leica Microsystems). The free mCherry is used as a nuclear/cytoplasmic marker. The experiments were repeated at least three times with similar results.

Immuno blot and co-immunoprecipitation (co-IP) assays

Proteins were transiently expressed in 3- to 4-week-old N. benthamiana leaves and then samples were harvested at 2 dpi and ground using a mortar and pestle in liquid nitrogen. Total proteins were extracted adding cold extraction buffer [25 mM Tris-HCl, pH7.5, 150 mM NaCl, 1 mM EDTA, 10% Glycerol, 10 mM DTT, 0.2% Nonidet-40, 2% (wt/v) polyvinylpolypyrolidone, and protease inhibitor cocktail (Roche)] on ice. Samples were centrifuged at 4100 x g at 4°C for 25 min, and then the supernatant was filtered through two layers of Miracloth (Merck Millipore) for western blot analysis and co-IP. For Western blot, samples were boiled for 5 min with 3 x SDS sample loading buffer (25 mM Tris-HCl (pH 6.8), 300 mM DTT, 6% SDS, 0.3% bromophenol blue, and 30% glycerol). Proteins were separated by 6 or 10 or 12% SDS-PAGE, transferred to PVDF membrane (Bio-Rad) using Trans-Blot Turbo Transfer System (Bio-Rad). Immunoblot was performed with HRP-conjugated anti-HA (Roche), anti-GFP (Santa Cruz), anti-Myc (Santa Cruz), and anti-FLAG (Sigma). For co-IP, total proteins were re-centrifuged at 19000 x g at 4°C for 20 min and then the supernatant was transferred to 1.5 mL LoBind e-tube (Eppendorf). IP samples were mixed with 30 μL of anti-HA (Sigma), anti-FLAG M2 (Roche), anti-GFP (Chromotek) or anti-Myc (Santa cruz) beads and incubated at 4°C for 2 hr. Samples were washed six times with IP buffer (25 mM Tris-HCl, pH7.5, 150 mM NaCl, 1 mM EDTA, 10% Glycerol, 10 mM DTT, 0.2% Nonidet-40, and protease cocktail inhibitor). Following the final wash step, supernatant was removed using a syringe. The resin was mixed with 3 x SDS sample loading buffer and then boiled for 5 min prior to loading on SDS-PAGE gels. The experiments were repeated at least three times with similar results.

Total protein was extracted from Arabidopsis transgenic and Col-0 plants. For Western blot, proteins were separated by 10% SDS-PAGE or 16% Tris-Glycine mini protein gel (ThermoFisher). Immunoblot was performed with HRP-conjugated anti-HA (Roche), anti-FLAG (Sigma), and Pathogenesis-related protein 1 (PR1) antibody (Agrisera).

Nuclear fractionation

Nuclear fractionation was performed using a modified protocol described by [63]. Plant tissue was ground in nuclei isolation buffer (NIB: 10 mM MES-KOH, pH 5.4, 10 mM NaCl, 10 mM KCl, 2.5 mM EDTA, 250 mM sucrose, 0.1 mM spermine, 0.5 mM spermidine, 1 mM DTT) with protease inhibitor cocktail (Roche) using a mortar and pestle. The ground tissue in NIB was filtered with Miracloth (Merck Millipore) and 10% Triton X-100 (final concentration of 0.5%) was added. The homogenate was centrifuged at 1000 x g for 10 min. The sucrose and Percoll layers were made by 5 ml of 2.5 M sucrose and 5 ml of 60% Percoll solution with pasteur pipette to subject the gradient to centrifugation at 1000 x g for 30 min at 4°C. Nuclei were collected from the 60% Percoll layer with a pasteur pipette and then washed with 5 volumes of NIB and 0.5% Triton X-100. After washing steps, the pellet of nuclei was resuspended with 5 ml of NIB, overlaid with 5 ml of 35% Percoll solution and centrifuged at 1000 x g for 10 min at 4°C. Isolated nuclear/cytosolic fractions were evaluated by western blot analysis using specific antibodies for the nuclear protein Histone H3 or the cytosolic protein phosphoenolpyruvate carboxylase (PEPC).

Statistical analysis

Statistical analysis was carried out using the one-way analysis of variance (ANOVA).

Supporting information

S1 Fig. RPS4 protein accumulation in Arabidopsis and N. benthamiana.

(A-B) Western blot analysis of RPS4-HS, RRS1-R-HF, and RPS4-HS/RRS1-R-HF transgenic lines. Total proteins were extracted from each plant and western blot was performed with anti-PR1 (A), anti-HA, and anti-FLAG (B) antibodies. (C) Reduced RPS4 accumulation is not due to reduced T-DNA transfer. RPS4-Myc, GFP, and GUS-HF or RRS1-HF constructs in A. tumefaciens were infiltrated into N. benthamiana leaves. A. tumefaciens cells were adjusted to the OD600 of 0.5 for RPS4-Myc and GFP or 0.1 for RRS1-HF and GUS-HF constructs. After 2 dpi, samples were harvested and Western blots were performed using anti-FLAG, anti-GFP, and anti-HA antibodies. All experiments were repeated three times.


S2 Fig. RPS4 nucleolus localization is altered by RRS1 co-expression in N. benthamiana.

(A) Overexpression of N-terminally YFP-tagged RPS4 with mCherry and GUS-HF results in nucleocytoplasmic localization. YFP-RPS4 mainly localizes to the nucleolus. The experiment was repeated three times with nearly identical results. Scale bar = 10 μm. (B) When co-expressing RRS1-R-HF and mCherry with YFP-RPS4, YFP signal is mainly observed in the nucleus but not nucleolus. Images were obtained at 2 dpi. The experiment was repeated three times with nearly identical results. Scale bar = 10 μm.


S3 Fig. RPS4 self-associates only in the presence of RRS1 in co-IP assays.

(A) Agrobacterium-mediated transient co-expression of RRS1-GFP/RPS4-HF/RPS4-Myc or GFP/ RPS4-HF/RPS4-Myc was performed in N. benthamiana leaves. Anti-FLAG co-IPs were performed with total protein extracts and probed with anti-GFP, -FLAG, and -Myc antibodies. (B) Co-IPs show that RRS1 self-associates and forms a heteromeric complex with RPS4. Transient co-expression assays of RRS1-GFP/RRS1-HF, RRS1-GFP/RPS4-HF or GFP/RRS1-HF were performed in N. benthamiana leaves. Immunoblots show the presence of proteins in total extracts (input) and after immunoprecipitation with anti-FLAG beads (IP-FLAG). All experiments were repeated three times.


S4 Fig. EDS1 associates with both RPS4 and RRS1 proteins in planta.

Co-IP was performed with transiently expressed RRS1-R-HF or RPS4-HF with GFP-EDS1 or GFP in N. benthamiana leaves. After 2 dpi, samples were harvested and then immunoprecipitated with anti-GFP beads. The samples were then analyzed by immunoblotting with anti-FLAG and anti-GFP antibodies. All experiments were repeated three times.


S5 Fig. AvrRps4 does not affect RPS4/EDS1 association in the nucleus in the presence or absence of RRS1.

BiFC assays of RPS4/EDS1 association in the presence of RRS1 or both RRS1 and AvrRps4 or AvrRps4E187A. N. benthamiana leaves were co-infiltrated with nVenus-RPS4/nCFP-EDS1/RRS1-R-HF/AvrRps4E187A or nVenus-RPS4/nCFP-EDS1/RRS1-R-HF/AvrRps4-mCherry, reconstructed YFP signals (nVenus/nCFP combination) were observed at 2 dpi. In the presence of RRS1-R-HF, both cCFP-RPS4/nCFP-EDS1/AvrRps4E187A-mCherry and cCFP-RPS4/nCFP-EDS1/AvrRps4-mCherry complex provided similar nuclear YFP fluorescence. The experiment was repeated three times. Scale bar = 15 μm.


S6 Fig. AvrRps4 does not affect RPS4/EDS1/PAD4 association in the nucleus in the presence RRS1.

(A-B) Multi-color BiFC analysis between RRS1, RPS4, EDS1 and PAD4 in the presence or absence of AvrRps4. RRS1-HF, nCerulean-RPS4, cCFP-EDS1 and nVenus-PAD4 were transiently co-expressed with AvrRps4-E187A-mCherry or AvrRps4-mCherry, in N. benthamiana leaves. Co-expression of nCerulean-RPS4 and cCFP-EDS1 resulted in the reconstitution of CFP fluorescence within the nucleus. Co-expression of cCFP-EDS1 and nVenus-PAD4 reconstructed YFP fluorescence in both the nucleus and cytoplasm. No significant differences were observed in the presence of AvrRps4 or AvrRps4E187A-mCherry for both combinations. The experiment was repeated three times with similar results. Scale bar = 15 μm.


S7 Fig. EDS1 interacts with AvrRps4.

(A-B) Both N- and C-terminally Myc tagged EDS1 co-immunoprecipitate with AvrRps4 in planta. The 35S::Myc-EDS1 or the 35S::EDS1-Myc were co-infiltrated with the 35S::PAD4-GFP, 35S::AvrRps4-GFP or 35S::GFP in N. benthamiana leaves and samples were harvested at 2 dpi. Immunoprecipitations were performed using anti-GFP and anti-Myc agarose beads. Specific protein-protein interactions were detected by immunoblotting with the indicated antibodies. AvrRps4C represents processed AvrRps4C-terminus. The experiment was repeated three times with similar results.


S8 Fig. BiFC verification of the interaction between EDS1 and AvrRps4.

The AvrRps4-cCFP and nVenus EDS1 constructs were transiently co-expressed in N. benthamiana leaves. The combination of AvrRps4-cCFP with nVenus-PAD4 was used as a negative control. The functionality of nVenus-PAD4 construct was verified by co-expression with cCFP-EDS1. Red or blue fluorescence is the indicative of chloroplast auto-fluorescence. Reconstitution of yellow fluorescence protein (YFP) indicates protein-protein interactions. The experiment was repeated three times with similar results. Scale bar = 15 μm.


S9 Fig. AvrRps4KRVY/AAAA mutant and EDS1 association in BiFC assay.

BiFC reveals that interaction between of EDS1 and AvrRps4KRVY/AAAA mutant forms cytoplasmic aggregations that are reduced in the presence of PAD4-HA but not in the presence of SAG101. BiFC assays were performed by co-expression of the indicated proteins in N. benthamiana. Scale bar = 15 μm.



We would like to thank Dr. Shyam Rallapalli and Dr. Lennart Wirthmüller for critical comments, Dr. Walter Gassmann for providing cCFP-RPS4, nVenus-RPS4, cCFP-EDS1, nVenus-EDS1, nVenus-PAD4, HA-EDS1, Myc-EDS1 constructs. Dr. Jane Parker for providing Arabidopsis 35S:RPS4-HS seeds, and YFP-RPS4 construct.

Author Contributions

  1. Conceptualization: SUH VC PFS JDGJ.
  2. Data curation: SUH VC PFS.
  3. Formal analysis: SUH VC PFS.
  4. Funding acquisition: PFS JDGJ.
  5. Investigation: SUH VC PFS PD.
  6. Methodology: SUH VC PFS JDGJ.
  7. Project administration: PFS JDGJ.
  8. Resources: SUH VC PFS PD ZD YM LT JDGJ.
  9. Software: SUH VC PFS.
  10. Supervision: PFS JDGJ.
  11. Validation: SUH VC PFS JDGJ.
  12. Visualization: SUH VC PFS.
  13. Writing – original draft: SUH VC PFS JDGJ.
  14. Writing – review & editing: SUH VC PFS JDGJ.


  1. 1. Kofoed EM, Vance RE. Innate immune recognition of bacterial ligands by NAIPs determines inflammasome specificity. Nature. 2011;477(7366):592–5. PubMed Central PMCID: PMCPMC3184209. pmid:21874021
  2. 2. Tor M, Lotze MT, Holton N. Receptor-mediated signalling in plants: molecular patterns and programmes. J Exp Bot. 2009;60(13):3645–54. Epub 2009/07/25. PubMed Central PMCID: PMC2766824. pmid:19628572
  3. 3. Bonardi V, Dangl JL. How complex are intracellular immune receptor signaling complexes? Frontiers in plant science. 2012;3:237. Epub 2012/10/31. PubMed Central PMCID: PMC3478704. pmid:23109935
  4. 4. Coll NS, Epple P, Dangl JL. Programmed cell death in the plant immune system. Cell Death Differ. 2011;18(8):1247–56. Epub 2011/04/09. PubMed Central PMCID: PMC3172094. pmid:21475301
  5. 5. Maekawa T, Kufer TA, Schulze-Lefert P. NLR functions in plant and animal immune systems: so far and yet so close. Nat Immunol. 2011;12(9):817–26. Epub 2011/08/20. pmid:21852785
  6. 6. Jones JD, Vance RE, Dangl JL. Intracellular innate immune surveillance devices in plants and animals. Science. 2016;354(6316).
  7. 7. Jacob F, Vernaldi S, Maekawa T. Evolution and Conservation of Plant NLR Functions. Front Immunol. 2013;4:297. Epub 2013/10/05. PubMed Central PMCID: PMC3782705. pmid:24093022
  8. 8. Bonardi V, Cherkis K, Nishimura MT, Dangl JL. A new eye on NLR proteins: focused on clarity or diffused by complexity? Curr Opin Immunol. 2012;24(1):41–50. Epub 2012/02/07. PubMed Central PMCID: PMC3482489. pmid:22305607
  9. 9. Jones JD, Dangl JL. The plant immune system. Nature. 2006;444(7117):323–9. Epub 2006/11/17. pmid:17108957
  10. 10. Gao Z, Chung EH, Eitas TK, Dangl JL. Plant intracellular innate immune receptor Resistance to Pseudomonas syringae pv. maculicola 1 (RPM1) is activated at, and functions on, the plasma membrane. Proceedings of the National Academy of Sciences of the United States of America. 2011;108(18):7619–24. Epub 2011/04/15. PubMed Central PMCID: PMC3088580. pmid:21490299
  11. 11. Slootweg E, Roosien J, Spiridon LN, Petrescu AJ, Tameling W, Joosten M, et al. Nucleocytoplasmic distribution is required for activation of resistance by the potato NB-LRR receptor Rx1 and is balanced by its functional domains. The Plant cell. 2010;22(12):4195–215. Epub 2010/12/24. PubMed Central PMCID: PMC3027179. pmid:21177483
  12. 12. Padmanabhan MS, Dinesh-Kumar SP. All hands on deck-the role of chloroplasts, endoplasmic reticulum, and the nucleus in driving plant innate immunity. Molecular plant-microbe interactions: MPMI. 2010;23(11):1368–80. Epub 2010/10/07. pmid:20923348
  13. 13. Chang C, Yu D, Jiao J, Jing S, Schulze-Lefert P, Shen QH. Barley MLA immune receptors directly interfere with antagonistically acting transcription factors to initiate disease resistance signaling. The Plant cell. 2013;25(3):1158–73. Epub 2013/03/28. PubMed Central PMCID: PMC3634683. pmid:23532068
  14. 14. Xu F, Kapos P, Cheng YT, Li M, Zhang Y, Li X. NLR-associating transcription factor bHLH84 and its paralogs function redundantly in plant immunity. PLoS pathogens. 2014;10(8):e1004312. Epub 2014/08/22. PubMed Central PMCID: PMC4140859. pmid:25144198
  15. 15. Xu F, Cheng YT, Kapos P, Huang Y, Li X. P-loop-dependent NLR SNC1 can oligomerize and activate immunity in the nucleus. Mol Plant. 2014. Epub 2014/09/23.
  16. 16. Zhu Z, Xu F, Zhang Y, Cheng YT, Wiermer M, Li X. Arabidopsis resistance protein SNC1 activates immune responses through association with a transcriptional corepressor. Proceedings of the National Academy of Sciences of the United States of America. 2010;107(31):13960–5. Epub 2010/07/22. PubMed Central PMCID: PMC2922275. pmid:20647385
  17. 17. Wirthmueller L, Zhang Y, Jones JD, Parker JE. Nuclear accumulation of the Arabidopsis immune receptor RPS4 is necessary for triggering EDS1-dependent defense. Curr Biol. 2007;17(23):2023–9. Epub 2007/11/13. pmid:17997306
  18. 18. Burch-Smith TM, Schiff M, Caplan JL, Tsao J, Czymmek K, Dinesh-Kumar SP. A novel role for the TIR domain in association with pathogen-derived elicitors. PLoS Biol. 2007;5(3):e68. Epub 2007/02/15. PubMed Central PMCID: PMC1820829. pmid:17298188
  19. 19. Padmanabhan MS, Ma S, Burch-Smith TM, Czymmek K, Huijser P, Dinesh-Kumar SP. Novel positive regulatory role for the SPL6 transcription factor in the N TIR-NB-LRR receptor-mediated plant innate immunity. PLoS pathogens. 2013;9(3):e1003235. Epub 2013/03/22. PubMed Central PMCID: PMC3597514. pmid:23516366
  20. 20. Ravensdale M, Bernoux M, Ve T, Kobe B, Thrall PH, Ellis JG, et al. Intramolecular interaction influences binding of the Flax L5 and L6 resistance proteins to their AvrL567 ligands. PLoS pathogens. 2012;8(11):e1003004. Epub 2012/12/05. PubMed Central PMCID: PMC3510248. pmid:23209402
  21. 21. Bernoux M, Ve T, Williams S, Warren C, Hatters D, Valkov E, et al. Structural and functional analysis of a plant resistance protein TIR domain reveals interfaces for self-association, signaling, and autoregulation. Cell Host Microbe. 2011;9(3):200–11. Epub 2011/03/16. PubMed Central PMCID: PMC3142617. pmid:21402359
  22. 22. Frost D, Way H, Howles P, Luck J, Manners J, Hardham A, et al. Tobacco transgenic for the flax rust resistance gene L expresses allele-specific activation of defense responses. Molecular plant-microbe interactions: MPMI. 2004;17(2):224–32. Epub 2004/02/18. pmid:14964536
  23. 23. Gassmann W, Hinsch ME, Staskawicz BJ. The Arabidopsis RPS4 bacterial-resistance gene is a member of the TIR-NBS-LRR family of disease-resistance genes. The Plant journal: for cell and molecular biology. 1999;20(3):265–77. Epub 1999/11/26.
  24. 24. Zhang Y, Dorey S, Swiderski M, Jones JD. Expression of RPS4 in tobacco induces an AvrRps4-independent HR that requires EDS1, SGT1 and HSP90. The Plant journal: for cell and molecular biology. 2004;40(2):213–24. Epub 2004/09/28.
  25. 25. Williams SJ, Sohn KH, Wan L, Bernoux M, Sarris PF, Segonzac C, et al. Structural basis for assembly and function of a heterodimeric plant immune receptor. Science. 2014;344(6181):299–303. Epub 2014/04/20. pmid:24744375
  26. 26. Narusaka M, Shirasu K, Noutoshi Y, Kubo Y, Shiraishi T, Iwabuchi M, et al. RRS1 and RPS4 provide a dual Resistance-gene system against fungal and bacterial pathogens. The Plant journal: for cell and molecular biology. 2009;60(2):218–26. Epub 2009/06/13.
  27. 27. Sohn KH, Segonzac C, Rallapalli G, Sarris PF, Woo JY, Williams SJ, et al. The nuclear immune receptor RPS4 is required for RRS1SLH1-dependent constitutive defense activation in Arabidopsis thaliana. PLoS Genet. 2014;10(10):e1004655. Epub 2014/10/24. PubMed Central PMCID: PMC4207616. pmid:25340333
  28. 28. Narusaka M, Hatakeyama K, Shirasu K, Narusaka Y. Arabidopsis dual resistance proteins, both RPS4 and RRS1, are required for resistance to bacterial wilt in transgenic Brassica crops. Plant signaling & behavior. 2014;9. Epub 2014/05/17. PubMed Central PMCID: PMC4203570.
  29. 29. Birker D, Heidrich K, Takahara H, Narusaka M, Deslandes L, Narusaka Y, et al. A locus conferring resistance to Colletotrichum higginsianum is shared by four geographically distinct Arabidopsis accessions. The Plant journal: for cell and molecular biology. 2009;60(4):602–13. Epub 2009/08/19.
  30. 30. Deslandes L, Olivier J, Peeters N, Feng DX, Khounlotham M, Boucher C, et al. Physical interaction between RRS1-R, a protein conferring resistance to bacterial wilt, and PopP2, a type III effector targeted to the plant nucleus. Proceedings of the National Academy of Sciences of the United States of America. 2003;100(13):8024–9. Epub 2003/06/06. PubMed Central PMCID: PMC164706. pmid:12788974
  31. 31. Sarris PF, Duxbury Z, Huh SU, Ma Y, Segonzac C, Sklenar J, et al. A Plant Immune Receptor Detects Pathogen Effectors that Target WRKY Transcription Factors. Cell. 2015;161(5):1089–100. Epub 2015/05/23. pmid:26000484
  32. 32. Le Roux C, Huet G, Jauneau A, Camborde L, Tremousaygue D, Kraut A, et al. A receptor pair with an integrated decoy converts pathogen disabling of transcription factors to immunity. Cell. 2015;161(5):1074–88. pmid:26000483
  33. 33. Cesari S, Bernoux M, Moncuquet P, Kroj T, Dodds PN. A novel conserved mechanism for plant NLR protein pairs: the "integrated decoy" hypothesis. Front Plant Sci. 2014;5:606. PubMed Central PMCID: PMCPMC4246468. pmid:25506347
  34. 34. Griebel T, Maekawa T, Parker JE. Nucleotide-binding oligomerization domain-like receptor cooperativity in effector-triggered immunity. Trends Immunol. 2014. Epub 2014/10/14.
  35. 35. Noutoshi Y, Ito T, Seki M, Nakashita H, Yoshida S, Marco Y, et al. A single amino acid insertion in the WRKY domain of the Arabidopsis TIR-NBS-LRR-WRKY-type disease resistance protein SLH1 (sensitive to low humidity 1) causes activation of defense responses and hypersensitive cell death. The Plant journal: for cell and molecular biology. 2005;43(6):873–88. Epub 2005/09/09.
  36. 36. Cesari S, Kanzaki H, Fujiwara T, Bernoux M, Chalvon V, Kawano Y, et al. The NB-LRR proteins RGA4 and RGA5 interact functionally and physically to confer disease resistance. The EMBO journal. 2014;33(17):1941–59. Epub 2014/07/16. PubMed Central PMCID: PMC4195788. pmid:25024433
  37. 37. Halff EF, Diebolder CA, Versteeg M, Schouten A, Brondijk TH, Huizinga EG. Formation and structure of a NAIP5-NLRC4 inflammasome induced by direct interactions with conserved N- and C-terminal regions of flagellin. The Journal of biological chemistry. 2012;287(46):38460–72. Epub 2012/09/27. PubMed Central PMCID: PMC3493891. pmid:23012363
  38. 38. Lage SL, Longo C, Branco LM, da Costa TB, Buzzo Cde L, Bortoluci KR. Emerging Concepts about NAIP/NLRC4 Inflammasomes. Front Immunol. 2014;5:309. Epub 2014/07/30. PubMed Central PMCID: PMC4078251. pmid:25071770
  39. 39. Tenthorey JL, Kofoed EM, Daugherty MD, Malik HS, Vance RE. Molecular basis for specific recognition of bacterial ligands by NAIP/NLRC4 inflammasomes. Mol Cell. 2014;54(1):17–29. Epub 2014/03/25. PubMed Central PMCID: PMC3988258. pmid:24657167
  40. 40. Sinapidou E, Williams K, Nott L, Bahkt S, Tor M, Crute I, et al. Two TIR:NB:LRR genes are required to specify resistance to Peronospora parasitica isolate Cala2 in Arabidopsis. The Plant journal: for cell and molecular biology. 2004;38(6):898–909. Epub 2004/05/29.
  41. 41. Lightfield KL, Persson J, Trinidad NJ, Brubaker SW, Kofoed EM, Sauer JD, et al. Differential requirements for NAIP5 in activation of the NLRC4 inflammasome. Infect Immun. 2011;79(4):1606–14. Epub 2011/02/02. PubMed Central PMCID: PMC3067536. pmid:21282416
  42. 42. Garcia AV, Blanvillain-Baufume S, Huibers RP, Wiermer M, Li G, Gobbato E, et al. Balanced nuclear and cytoplasmic activities of EDS1 are required for a complete plant innate immune response. PLoS pathogens. 2010;6:e1000970. Epub 2010/07/10. PubMed Central PMCID: PMC2895645. pmid:20617163
  43. 43. Bhattacharjee S, Halane MK, Kim SH, Gassmann W. Pathogen effectors target Arabidopsis EDS1 and alter its interactions with immune regulators. Science. 2011;334(6061):1405–8. Epub 2011/12/14. pmid:22158819
  44. 44. Wagner S, Stuttmann J, Rietz S, Guerois R, Brunstein E, Bautor J, et al. Structural basis for signaling by exclusive EDS1 heteromeric complexes with SAG101 or PAD4 in plant innate immunity. Cell Host Microbe. 2013;14(6):619–30. Epub 2013/12/18. pmid:24331460
  45. 45. Heidrich K, Wirthmueller L, Tasset C, Pouzet C, Deslandes L, Parker JE. Arabidopsis EDS1 connects pathogen effector recognition to cell compartment-specific immune responses. Science. 2011;334(6061):1401–4. Epub 2011/12/14. pmid:22158818
  46. 46. Sohn KH, Hughes RK, Piquerez SJ, Jones JD, Banfield MJ. Distinct regions of the Pseudomonas syringae coiled-coil effector AvrRps4 are required for activation of immunity. Proceedings of the National Academy of Sciences of the United States of America. 2012;109(40):16371–6. Epub 2012/09/19. PubMed Central PMCID: PMC3479578. pmid:22988101
  47. 47. Huang S, Monaghan J, Zhong X, Lin L, Sun T, Dong OX, et al. HSP90s are required for NLR immune receptor accumulation in Arabidopsis. The Plant journal: for cell and molecular biology. 2014;79(3):427–39. Epub 2014/06/04.
  48. 48. Wiermer M, Feys BJ, Parker JE. Plant immunity: the EDS1 regulatory node. Curr Opin Plant Biol. 2005;8(4):383–9. Epub 2005/06/09. pmid:15939664
  49. 49. Feys BJ, Moisan LJ, Newman MA, Parker JE. Direct interaction between the Arabidopsis disease resistance signaling proteins, EDS1 and PAD4. The EMBO journal. 2001;20(19):5400–11. Epub 2001/09/28. PubMed Central PMCID: PMC125652. pmid:11574472
  50. 50. Lee LY, Fang MJ, Kuang LY, Gelvin SB. Vectors for multi-color bimolecular fluorescence complementation to investigate protein-protein interactions in living plant cells. Plant Methods. 2008;4:24. Epub 2008/10/17. PubMed Central PMCID: PMC2572157. pmid:18922163
  51. 51. Zhu S, Jeong RD, Venugopal SC, Lapchyk L, Navarre D, Kachroo A, et al. SAG101 forms a ternary complex with EDS1 and PAD4 and is required for resistance signaling against turnip crinkle virus. PLoS pathogens. 2011;7(11):e1002318. Epub 2011/11/11. PubMed Central PMCID: PMC3207898. pmid:22072959
  52. 52. Zhang L, Chen S, Ruan J, Wu J, Tong AB, Yin Q, et al. Cryo-EM structure of the activated NAIP2-NLRC4 inflammasome reveals nucleated polymerization. Science. 2015;350(6259):404–9. Epub 2015/10/10. PubMed Central PMCID: PMC4640189. pmid:26449474
  53. 53. Hu Z, Zhou Q, Zhang C, Fan S, Cheng W, Zhao Y, et al. Structural and biochemical basis for induced self-propagation of NLRC4. Science. 2015;350(6259):399–404. Epub 2015/10/10. pmid:26449475
  54. 54. Maqbool A, Saitoh H, Franceschetti M, Stevenson CE, Uemura A, Kanzaki H, et al. Structural basis of pathogen recognition by an integrated HMA domain in a plant NLR immune receptor. Elife. 2015;4. Epub 2015/08/26. PubMed Central PMCID: PMC4547098.
  55. 55. Griebel T, Maekawa T, Parker JE. NOD-like receptor cooperativity in effector-triggered immunity. Trends Immunol. 2014;35(11):562–70. Epub 2014/10/14. pmid:25308923
  56. 56. Heidrich K, Tsuda K, Blanvillain-Baufume S, Wirthmueller L, Bautor J, Parker JE. Arabidopsis TNL-WRKY domain receptor RRS1 contributes to temperature-conditioned RPS4 auto-immunity. Frontiers in plant science. 2013;4:403. Epub 2013/10/23. PubMed Central PMCID: PMC3797954. pmid:24146667
  57. 57. Schreiber KJ, Bentham A, Williams SJ, Kobe B, Staskawicz BJ. Multiple Domain Associations within the Arabidopsis Immune Receptor RPP1 Regulate the Activation of Programmed Cell Death. PLoS Pathog. 2016;12(7):e1005769. PubMed Central PMCID: PMCPMC4948778. pmid:27427964
  58. 58. Mestre P, Baulcombe DC. Elicitor-mediated oligomerization of the tobacco N disease resistance protein. The Plant cell. 2006;18(2):491–501. Epub 2006/01/03. PubMed Central PMCID: PMC1356554. pmid:16387833
  59. 59. Ade J, DeYoung BJ, Golstein C, Innes RW. Indirect activation of a plant nucleotide binding site-leucine-rich repeat protein by a bacterial protease. Proceedings of the National Academy of Sciences of the United States of America. 2007;104(7):2531–6. Epub 2007/02/06. PubMed Central PMCID: PMC1790868. pmid:17277084
  60. 60. Wang GF, Ji J, El-Kasmi F, Dangl JL, Johal G, Balint-Kurti PJ. Molecular and functional analyses of a maize autoactive NB-LRR protein identify precise structural requirements for activity. PLoS pathogens. 2015;11(2):e1004674. Epub 2015/02/27. PubMed Central PMCID: PMC4342346. pmid:25719542
  61. 61. Bent A. Arabidopsis thaliana floral dip transformation method. Methods Mol Biol. 2006;343:87–103. Epub 2006/09/22. pmid:16988336
  62. 62. Walter M, Chaban C, Schutze K, Batistic O, Weckermann K, Nake C, et al. Visualization of protein interactions in living plant cells using bimolecular fluorescence complementation. The Plant journal: for cell and molecular biology. 2004;40(3):428–38. Epub 2004/10/08.
  63. 63. Sikorskaite S, Rajamaki ML, Baniulis D, Stanys V, Valkonen JP. Protocol: Optimised methodology for isolation of nuclei from leaves of species in the Solanaceae and Rosaceae families. Plant Methods. 2013;9:31. Epub 2013/07/28. PubMed Central PMCID: PMC3728069. pmid:23886449