Nuclear Localization of HopA1Pss61 Is Required for Effector-Triggered Immunity

Plant resistance proteins recognize cognate pathogen avirulence proteins (also named effectors) to implement the innate immune responses called effector-triggered immunity. Previously, we reported that hopA1 from Pseudomonas syringae pv. syringae strain 61 was identified as an avr gene for Arabidopsis thaliana. Using a forward genetic screen approach, we cloned a hopA1-specific TIR-NBS-LRR class disease resistance gene, RESISTANCE TO PSEUDOMONAS SYRINGAE6 (RPS6). Many resistance proteins indirectly recognize effectors, and RPS6 is thought to interact with HopA1Pss61 indirectly by surveillance of an effector target. However, the involved target protein is currently unknown. Here, we show RPS6 is the only R protein that recognizes HopA1Pss61 in Arabidopsis wild-type Col-0 accession. Both RPS6 and HopA1Pss61 are co-localized to the nucleus and cytoplasm. HopA1Pss61 is also distributed in plasma membrane and plasmodesmata. Interestingly, nuclear localization of HopA1Pss61 is required to induce cell death as NES-HopA1Pss61 suppresses the level of cell death in Nicotiana benthamiana. In addition, in planta expression of hopA1Pss61 led to defense responses, such as a dwarf morphology, a cell death response, inhibition of bacterial growth, and increased accumulation of defense marker proteins in transgenic Arabidopsis. Functional characterization of HopA1Pss61 and RPS6 will provide an important piece of the ETI puzzle.


Introduction
Plants are challenged by a wide variety of pathogens. However, they have evolved sophisticated immune systems to protect themselves from pathogen infections. [1,2]. The plant immune systems involve two different actions. One is pattern-triggered immunity (PTI), and the other is effector-triggered immunity (ETI) [3,4]. PTI is typically activated by the recognition of pathogen-associated molecular patterns (PAMPs) with pattern recognition receptors (PRRs) in the plants. This recognition process is required to promote defense signaling pathways, such as the activation of mitogen-activated protein kinase (MAPK) cascades, reactive oxygen species, ion channel opening, callose deposition, and defense-related genes [4][5][6]. However, to encounter PTI, pathogens deploy several infectious effectors. Subsequently, plants utilize a second layer of defense through the activation of resistance (R) proteins; this mode is known as ETI response [3,7].
At the ETI mode, pathogen effectors are recognized by resistance (R) proteins, which are notably found in the form of nucleotide-binding (NB) domain and leucine-rich repeat
RPS6 possesses a long 3 untranslated region (UTR) containing six exons and five introns with a length of approximately 3 kb. The long 3 UTR is a typical characteristic of nonsense-mediated mRNA decay (NMD) target, and RPS6 is required for autoimmunity in NMD-deficient mutant smg7 [28,29]. As shown in Figure 1B, rps6-5 with a T-DNA in the tenth exon of RPS6 (at position 1789 of its 3 UTR) behaves like Col-0 in response to DC3000(hopA1 Pss61 ). Consistent with these, RPS6 transcripts were detected in Col-0 and rps6-5 but not in rps6-3 and rps6-4 (Supplementary Figure S1). This result demonstrates a 1789 bp 3 UTR of RPS6 is sufficient to confer resistance to DC3000(hopA1 Pss61 ).

HopA1 Pss61
Targets to the Nucleus, Cytoplasm, Plasma Membrane, and Plasmodesmata The amino acid sequences of HopA1 Pss61 and HopA1 DC3000 are 57% identical, and diverged amino acids are distributed throughout the proteins [26]. In order to investigate the subcellular localization of the HopA1 Pss61 and HopA1 DC3000 effector proteins, we fused a GFP tag to the N-and C-terminus of each HopA1 driven by the 35S promoter. The HopA1 derivatives, GFP-HopA1 Pss61 , HopA1 Pss61 -GFP, GFP-HopA1 DC3000 , and HopA1 DC3000 -GFP, were transiently expressed in N. benthamiana leaves using agroinfiltration. GFP-HopA1 Pss61 , GFP at the N-terminus of HopA1 Pss61 , was localized to the nucleus, cytoplasm, and plasma membrane (PM) (Figure 2A,B, Supplementary Figures S2 and S3). In the nucleus, GFP-HopA1 Pss61 was detected not only in the nucleoplasm but also in the nucleolus (Figure 2A). A similar localization pattern was observed in GFP-HopA1 DC3000 ( Figure 2A). Surprisingly, HopA1 Pss61 -GFP and HopA1 DC3000 -GFP were found in the PM but not in the nucleus ( Figure 2A) and co-localized with PM localized protein PBS1-mCherry (Supplementary Figure S2) [30], suggesting C-terminally fused GFP inhibits the nuclear localization of both HopA1s. The expression of HopA1 derivatives was detected by Western blot analysis, confirming that the GFP fusion proteins were full-length (Supplementary Figure S5). In addition, HopA1 Pss61 -GFP was observed in punctate spots around the cell periphery, indicating plasmodesmata (PD) localization (Supplementary Figure S3C). PD is considered as a space that can act as a passage between cells [31,32]. To carefully investigate the HopA1 Pss61 localization, we co-expressed GFP-HopA1 Pss61 with PDLP5-RFP, a marker protein of PD [33]. As shown in Supplementary Figure S3, HopA1 Pss61 and PDLP5 were co-localized in the PD and PM in the plasmolysis condition. In addition, the co-localization of HopA1 Pss61 and PBS1 to PM was observed (Supplementary Figure S2). Taken together, we conclude that HopA1 Pss61 is distributed not only in nucleocytoplasm but also in PM and PD.

Nuclear Localization of HopA1 Pss61 Induces Cell Death Responses
The pHIR11 cosmid containing hopA1 and Pss61 type III secretion system genes, when expressed in P. fluorescens, can induce a robust hypersensitive response (HR) in tobacco [34,35]. To test whether HopA1 alone can elicit cell death, agrobacteria expressing the GFP-tagged hopA1 constructs used in Figure 2A were infiltrated into N. benthamiana leaves. Interestingly, despite the lower level of protein expression compared with other derivatives in the immunoblot (Supplementary Figure S5), only GFP-HopA1 Pss61 , which targets the nucleus, was capable of inducing an HR-like cell death four days after infiltration, whereas HopA1 Pss61 -GFP, GFP-HopA1 DC3000 , and HopA1 DC3000 -GFP failed to exhibit the cell death response ( Figure 2B). To confirm that GFP-HopA1 Pss61 -induced cell death is GFPtagging-independent, we overexpressed HopA1 Pss61 with another epitope in N. benthamiana. Indeed, similar to GFP-HopA1 Pss61 , N-terminally HA-tagged HopA1 Pss61 triggered a cell death response (Supplementary Figure S4). These results raise the possibility that effector HopA1 Pss61 , not HopA1 DC3000 , is recognized by tobacco and that the nuclear pool of GFP-HopA1 Pss61 is essential for the cell death induction.
For the confirmation of HopA1 Pss61 expression in the nucleus, a nuclear fractionation assay was performed. Proteins were extracted from N. benthamiana expressing GFP-HopA1 Pss61 and fractionated. Western blot showed HopA1 Pss61 was accumulated in both nuclear and cytoplasmic (non-nuclear) fractions (Supplementary Figure S6), consistent with the nucleocytoplasmic distribution of HopA1 Pss61 in the confocal microscopy analysis. Histone H3 and PEPC were detected only in the nucleus and cytoplasm, respectively, indicating a high degree of enrichment of the indicated compartment in our fractions.
To analyze the connection between the HopA1 Pss61 localization and function, we added the nuclear export signal (NES) that originated from the HIV-1 Rev protein to the N terminal of HopA1 Pss61 . First, we put to test the localization of HopA1 Pss61 in the presence of NES by a confocal microscopy experiment. As shown in Figure 3A, GFP-NES-HopA1 Pss61 failed to localize in the nucleus, while GFP-HopA1 Pss61 was clearly observed in the nucleus. In our repeated experiments, nuclear localization of GFP-NES-HopA1 Pss61 was barely detected in most cells, but it was found in a few cells (Supplementary Figure S7), suggesting residual localization of GFP-NES-HopA1 Pss61 in the nucleus. To optimize the effect of agrobacteria concentration on HopA1 Pss61 -and NES-HopA1 Pss61 -induced cell death response, the agrobacteria were grown to OD 600 , ranging from 0.025 to 0.2, and independently infiltrated into N. benthamiana and N. tabacum Xanthi. In comparison with GFP-HopA1 Pss61 , GFP-NES-HopA1 Pss61 showed a cell death response with a significantly reduced level at a final OD 600 of 0.2 in N. tabacum, whereas OD 600 was 0.05 in N. benthamiana (Supplementary Figure S8). As shown in Figure 3B, GFP-NES-HopA1 Pss61 showed a dramatically weakened cell death response, while GFP-HopA1 Pss61 induced strong cell death. Consistent with the visible phenotypes, GFP-NES-HopA1 Pss61 produced ion leakage intermediate between GFP-HopA1 Pss61 and negative control, GFP ( Figure 3D). Western blot analysis indicated both HopA1 Pss61 proteins were expressed with similar levels ( Figure 3C), demonstrating that the compromised cell death in GFP-NES-HopA1 Pss61 is not due to low levels of protein expression but due to the export of HopA1 Pss61 from the nucleus. Together, these results demonstrate that the nuclear pool of HopA1 Pss61 is required for RPS6 recognition and cell death induction.

RPS6 Localizes to Nucleus and Cytoplasm
To elucidate the subcellular localization of RPS6 inside the plant cell, genomic RPS6 from Col-0 was fused in frame with green fluorescent protein (GFP) at the N-terminus under control of the strong CaMV 35S promoter to generate GFP-gRPS6. GFP-gRPS6 and control EDS1-mCherry were transiently expressed in N. benthamiana leaf cells, and their localization was monitored under the confocal microscope. As shown in Figure 4A, GFP-gRPS6 represented nucleocytoplasmic distribution and co-localized with EDS1-mCherry, which is known to localize in nucleus and cytoplasm [36], reminiscent of the pattern of HopA1 localization. The expression of GFP-gRPS6 was confirmed by Western blot analysis ( Figure 4B), suggesting the localization of RPS6 was based on full-length protein expression. Since HopA1 Pss61 -triggered cell death requires nuclear localization of HopA1 Pss61 , we hypothesized that this localization is important for RPS6 activation. To test the hypothesis, we co-expressed GFP-gRPS6 and mCherry-HopA1 Pss61 in N. benthamiana leaf cells. As expected, we observed the co-localization of GFP-gRPS6 and mCherry-HopA1 Pss61 in the nucleus and cytoplasm ( Figure 4C). RPS6 recognizes HopA1 Pss61 indirectly, as HopA1 Pss61 did not interact with RPS6 in yeast-two hybrid analysis (Supplementary Figure S9). Together with Figure 3, these results suggest that the indirect recognition of HopA1 Pss61 by RPS6 in the nucleus might be necessary to activate ETI responses.

Induction of HopA1 Pss61 Triggers Defense Responses and Bacterial Growth Suppression in Arabidopsis
HopA1 Pss61 induced bacterial disease resistance and an HR response in wild-type RLD [25,26]. To gain further insight into HopA1 Pss61 -induced defense responses, we produced transgenic RLD lines that expressed hopA1 under the control of an estradiol-inducible promoter, which enabled us to activate RPS6 conditionally, and selected three independent estradiol-hopA1 homozygous lines of T3 generation (line #2, line #3, and line #5). In the absence of estradiol induction, no morphological differences were found between RLD and line #5, while line #2 and line #3 were slightly smaller than RLD. However, growth reduction and cell death-like symptoms appeared strongly in line #2 and line #3 but weakly in line #5 7 days after spraying 40 µM estradiol ( Figure 5A). Consistent with the morphological phenotypes, accumulation of HopA1 Pss61 was detected in line #2 and line #3, whereas it was below the detection limit in line #5 (Supplementary Figure S10). In addition, levels of the defense marker proteins PR1 and PR2 gradually increased and reached a higher level at 48 h after estradiol treatment in hopA1 transgenic lines ( Figure 5B). Both induction of cell death and accumulation of PR proteins correlated with the expression level of HopA1 Pss61 . The basal level of cell death symptoms and PR protein expression in the mock-treated plants is likely due to a leaky expression of HopA1 Pss61 from estradiol-inducible promoters ( Figure 5A and Figure S10) [37]. Bacterial growth assays showed that the growth of DC3000 in lines #2 and #3 was less than in RLD by a factor of 50-100 ( Figure 5C). Collectively, these data indicate that defense responses are being activated in hopA1 transgenic lines. Immunoblots were analyzed with the indicated antibodies; (C) In planta bacterial growth was measured in indicated plants after inoculation with DC3000 at 10 5 cfu/cm 2 in the absence (gray columns) or presence (black columns) of estradiol. Asterisk indicates significant differences with wild-type RLD (** p < 0.01, * p < 0.05, Student's t-test). This experiment was repeated twice with a similar result.

Discussion
Previously, we identified RPS6 encoding a TIR-NBS-LRR (TNL) protein by using a loss of resistance screen and a positional cloning approach. RPS6 specifically recognizes a bacterial effector gene hopA1 Pss61 from P. syringae pv. syringae strain 61 to trigger an immune response [25,26]. RPS6 enables genetic tools to test the range of the effector-spectrum in srfr1-mediated resistance in the RLD background. Mutations in SRFR1 increase both AvrRps4-and HopA1-triggered immunity [26,38]. In addition, SRFR1 physically interacts with some TNL proteins, such as RPS4, RPS6, SNC1, and a central immune regulator, EDS1 [39,40], demonstrating a possible role of SRFR1 as a general negative regulator in TNL R protein-mediated immunity. To date, RPS4 and RPS6 are the only Arabidopsis TIR-NBS-LRR R genes for which Pseudomonas effector genes are known, both of which are EDS1-and SRFR1-dependent. Therefore, RPS6 allows a direct comparison with RPS4 to dissect the EDS1-dependent signaling pathway in Arabidopsis.

Investigation of Genes in RPS6 Locus
Some R proteins have formed paired immune receptors for the effector recognition. For example, Arabidopsis TNL R protein RPS4 genetically and physically interacts with RRS1, another TNL R protein with a WRKY domain, and is involved in the recognition of PopP2 from Ralstonia solanacearum and AvrRps4 from P. syringae [41][42][43]. RGA4-RGA5 is another example of paired receptors to detect both Magnaporthe oryzae effector AVR1-CO39 and unrelated M. oryzae effector AVR-pia [44,45]. This evidence led us to suspect whether RPS6 requires another R protein to fully recognize HopA1 Pss61 . Hence, we first focused on genes around RPS6, including six TNL genes and At5g46460, which encodes a pentatricopeptide repeat protein ( Figure 1A). At5g46460 and RPS6 (At5g46470) possess a short intergenic region and transcribe polycistronically as a single transcript [27]. Additionally, both At5g46460 and RPS6 were included in the construct for the previous RLD accession rps6-1 complementation [26]. In Figure 1D, however, none of Col-0 T-DNA insertion lines, except rps6-3 and rps6-4, abolish HopA1 Pss61 -triggered immunity, indicating that RPS6 indeed recognizes HopA1 Pss61 in Col-0 as well as RLD, and does not require other surrounding genes for HopA1 Pss61 recognition. The current gene model in TAIR10 showed that RPS6 contains an extensive (~3 kb) 3 UTR with six exons. The long 3 UTR is a typical characteristic of nonsense-mediated mRNA decay (NMD) target [28]. RPS6 is required for autoimmunity in NMD-deficient mutant smg7 [29]. Moreover, aberrant transcripts were expressed in the 3 UTR region of RPS6 in the absence of SMN2, which encodes DEAD-Box RNA Helicase [46]. In our pathogenesis assay, rps6-5, in which the T-DNA was inserted at position 1789 of 3 UTR within exon 10, did not compromise RPS6 function (Figure 1). Consistent with this, in our previous study, we found the presence of poly-A tails within the exon 9 [26]. Together, these results raise the possibility that 1789 bp of RPS6 3 UTR region is sufficient to confer HopA1 Pss61 -triggered resistance and might be a target of NMD that possibly controls aberrant RPS6 transcripts to fine-tune plant growth and defense.

Localization and Functional Analysis of HopA1 Pss61 and RPS6
To gain a better understanding of HopA1 function, we analyzed the subcellular localization and cell death induction of HopA1 Pss61 and HopA1 DC3000 . GFP-HopA1 Pss61 (GFP fused to the N-terminus of HopA1 Pss61 ) was localized to the nucleus, cytoplasm, PM, and PD, and induced cell death. Surprisingly, HopA1 Pss61 -GFP (GFP fused to the C-terminus of HopA1 Pss61 ) was not found in the nucleus and failed to trigger cell death for unknown reasons (Figure 2A,C). The biochemical function of HopA1 Pss61 is unknown, and we cannot exclude that the C-terminally tagged GFP may cause improper protein folding or occlude a nuclear localization signal (NLS) of HopA1 Pss61 [47]. Additionally, GFP-NES-HopA1 Pss61 was excluded from the nucleus, and its mislocalization leads to the loss of HopA1 Pss61 -induced cell death (Figure 3 and Figure S8). HopA1 DC3000 did not produce cell death regardless of the position of the GFP tag (Figure 2A,C). Together, these results imply that the nuclear localization of HopA1 Pss61 is necessary to induce cell death.
RPS6 was also found in the nucleus and cytoplasm (Figure 4). In our yeast-two hybrid assay, no physical interaction was observed between RPS6 and HopA1 Pss61 (Supplementary Figure S9). Again, it is likely that a nuclear pool of RPS6 and HopA1 Pss61 may play an essential role in ETI, as was proposed for N and RPS4 [48,49]. In addition, we cannot exclude the possibility that RPS6 and HopA1 Pss61 might indirectly interact in the nucleus with the help of guardee (or decoy). The nucleolus is not only involved in the biogenesis of ribosomal RNA but is also implicated in the control of disease, regulation of cell cycle, and as a storage site [50]. Fuhrman and coworkers showed that the NOL-6 nucleolar protein in Caenorhabditis elegans suppressed innate immunity against bacterial pathogens by inhibiting the transcriptional activity of the tumor suppressor p53 [51]. Both GFP-HopA1 Pss61 and GFP-HopA1 DC3000 localized to the nucleolus (Figure 2A and Figure S3A), suggesting that both HopA1 proteins may interact with host virulence target(s) in the nucleolus to enhance bacterial virulence, whereas only HopA1 Pss61 is monitored by RPS6 to trigger ETI responses.
In Supplementary Figures S2 and S3, we found HopA1 Pss61 localized in both PM and PD. PD are known as intracellular channels in plants that offer an effective cell-tocell exchange of signal molecules [31]. PD are in charge of chloroplast metabolism, the ER to Golgi secretion system, as well as callose-deposition, a well-established defense response involved in plant innate immunity [52][53][54]. Indeed, some effectors are reported to target PD to suppress PTI responses. For example, the Fusarium graminearum effector FGL1 is known to inhibit callose-mediated immunity by releasing free fatty acids [55]. Xanthomonas campestris pv. vesicatoria effector XopJ also suppresses callose deposition [56]. Therefore, we could not exclude the possibility that HopA1 targets PD for its virulence function. As shown in the example of AvrRps4, which promotes bacterial virulence and suppresses PTI [57], HopA1 Pss61 may compromise PTI in the absence of RPS6, albeit the virulence function of HopA1 Pss61 is unknown.
The remaining open challenge is to identify the guardee (or decoy) and/or virulence target protein(s) of HopA1 Pss61 . Although HopA1 Pss61 interacts with EDS1, it has not been shown that this is what activates RPS6 [39]. This would allow us to elucidate the molecular mechanism of HopA1 Pss61 -triggered immunity or its virulence function. Functional characterization of RPS6 and comparisons with RPS4 will contribute to a closer dissection of the TNL resistance pathway, which is regulated by the positive regulator EDS1 and negative regulator SRFR1.

Disease and Bacterial Growth Curve Assay
Pseudomonas syringae pv tomato strain DC3000 expressing the empty vector pML123 or expressing shcA-hopA1 from P. syringae pv syringae strain 61 were described in previous studies [25,26]. For both disease and bacterial growth curve assay, Arabidopsis plants were grown under 11 h light/13 h dark cycle at 70% humidity and 21 • C conditions. For disease assays, 4-week-old Arabidopsis leaves were infiltrated with a bacterial suspension of 5 × 10 6 colony-forming units cfu/mL in 10 mM MgCl 2 using a 1 mL needless syringe. For in planta bacterial growth assays, bacterial suspensions of 2 × 10 5 cfu/mL were infiltrated into leaves of 4-week-old plants. After 3 days, two leaf discs (a total of 0.5 cm 2 ) were collected by cork borer (model: KA-48, size: 5) and ground in 10 mM MgCl 2 and plated in serial dilution on Pseudomonas Agar F (MB cell, Seoul, Korea) with appropriate antibiotics, all in quadruplicate at the indicated time points.

HR and Ion Leakage Assays
For HR assay, N. benthamiana or N. tabacum cv. Xanthi plant was grown under 9 h light/15 h dark cycle at 60% humidity and 24-26 • C condition for 5-6 weeks. HopA1 constructs were mobilized into the Agrobacterium tumefaciens strain C58C1 containing the virulence plasmid pCH32. After overnight culture in LB media, agrobacteria cells were pelleted and resuspended in 10 mM MgCl 2 with 100 µM acetosyringone (Sigma-Aldrich, St. Louis, MO, USA) adjusted to an OD600 of 0.2. The Agrobacterium was incubated for 2 h at room temperature and infiltrated into Nicotiana species leaves with a 1 mL needleless syringe. Silencing suppresser P19 was co-infiltrated for the Agrobacterium-mediated transient expression. Cell death phenotypes were visualized 4 to 5 days post-infiltration in N. benthamiana and 2 days post-inoculation in N. tabacum cv. Xanthi.
Ion leakage assay was performed as described [60]. Briefly, the Agrobacterium carrying relevant constructs were infiltrated into N. benthamiana leaves as described above. Six leaf discs were collected at 24 h after inoculation and washed three times for 10 min with distilled water. The leaf discs were immersed in a 12-well plate containing 4 mL of distilled water. The conductivity was measured by using Traceable (R) Conductivity/TDS Meter (VWR). The time-point for ion leakage was followed as described in Figure 3D.

Confocal Laser Scanning Microscopy
A confocal microscopy assay was performed to monitor the subcellular localization of HopA1 and RPS6. The Agrobacterium suspension was infiltrated into 4-5-week-old N. benthamiana plants by routine procedures. Two days later, plant tissues for live imaging were observed with an Olympus fluoview FV1000 or Olympus FV1000MPE. The GFP and RFP fluorescence was excited by a 488 nm laser and a 559 nm argon laser, respectively. For 4 ,6-diamidino-2-phenylindole (DAPI) staining, N. benthamiana tissues were cut into small pieces and stained in DAPI solution (Sigma-Aldrich, St. Louis, MO, USA) at a concentration of 1 µg/mL for 30 min under dark condition. DAPI signals representing for nucleus were excited by a 405 nm laser.
Protein fractionation was performed on N. benthamiana expressing GFP-HopA1 Pss61 based on Plant Nuclei Isolation/Extraction Kit (Sigma, St. Louis, MO, USA). In detail, tissue was processed in NIB buffer. Extracts were filtered and centrifuged at 3000× g for 10 min at 4 • C to produce a nuclear pellet. The supernatant containing cytoplasmic proteins was transferred to a new tube and remained on ice. The nuclear pellet was resuspended in NIBA buffer and 1.5 M sucrose, then was centrifuged at 13,000× g for 10 min at 4 • C. The white pellet containing nuclear proteins was collected and resuspended in protein extraction buffer. Further steps are similar to that in the mentioned normal Western blotting. Immunodetection was performed with anti-GFP to detect GFP-HopA1 Pss61 and with anti-Histone H3 and anti-PEPC to confirm correct nuclear and non-nuclear fractions, respectively.

Yeast Two-Hybrid
For yeast two-hybrid, hopA1 Pss61 and RPS6 (CDS) were cloned in pDEST32 and pDEST22, respectively. The pDEST32-hopA1 Pss61 and pDEST22-RPS6 constructs were transformed together into yeast strain PJ69-4A by a standard yeast transformation procedure. The transformation mixture was plated on SD media (-Trp-Leu, -Trip-Leu-His). Plates were grown at 30 • C and examined 4 days later.

Conclusions
Effector-triggered immunity (ETI) is mediated by genetic interactions between plant resistance (R) genes and pathogen avirulence (avr) genes and is highly effective in protecting plants from pathogens. Although over the past 20 years, many R genes have been identified, the mechanism of how R proteins induce resistance upon a perception of cognate effector proteins is still unclear.
Here, we investigated a bacterial effector, HopA1 Pss61 -triggered plant immune responses. Preferentially, we showed that RPS6 is the only R protein that recognizes HopA1 Pss61 in Col-0. RPS6 and HopA1 Pss61 co-localize in the nucleus and cytoplasm. Moreover, by exporting HopA1 Pss61 from the nucleus with nuclear export signal (NES), we uncovered that a nuclear pool of HopA1 Pss61 is critical for ETI responses. Additionally, we demonstrated that transgenic Arabidopsis plants expressing hopA1 Pss61 with an estradiol inducible system showed a dwarf morphology, a cell death response, bacterial growth inhibition, and increased accumulation of defense marker proteins. Together, these findings suggest that plants build up an RPS6 recognition system for HopA1 Pss61 and that the nuclear localization of HopA1 Pss61 and RPS6 is involved in ETI.
Our current research to increase the understanding of plant innate immunity in the reference plant Arabidopsis can be applied to crop plants for durable pathogen resistance, which can reduce our reliance on chemical disease control and improve agricultural safety and crop yields.