Received: 1 June 2021 | Revised: 27 September 2021 | Accepted: 27 September 2021 DOI: 10.1111/mpp.13153 ORIGINAL ARTICLE Different epitopes of Ralstonia solanacearum effector RipAW are recognized by two Nicotiana species and trigger immune responses Yang Niu1 | Shouyang Fu1 | Gong Chen1 | Huijuan Wang1 | Yisa Wang1 | JinXue Hu1 | Xin Jin1 | Mancang Zhang1 | Mingxia Lu1 | Yizhe He1 | Dongdong Wang2 | Yue Chen1 | Yong Zhang3,4 | Núria S. Coll5 | Marc Valls5,6 | Cuizhu Zhao1 | Qin Chen2 | Haibin Lu1,7 1 State Key Laboratory of Crop Stress Biology for Arid Areas, College of Agronomy, Northwest A&F University, Yangling, China 2 Shaanxi Key State Laboratory of Crop Heterosis, Northwest A&F University, Yangling, China 3 College of Food Science and Engineering, Northwest A&F University, Yangling, China 4 College of Resources and Environment, Southwest University, Chongqing, China 5 Interdisciplinary Research Center for Agriculture Green Development in Yangtze River Basin, Southeast University, Chongqing, China 6 Centre for Research in Agricultural Genomics, CSIC-­IRTA-­UAB-­UB, Bellaterra, Catalonia, Spain 7 Department of Genetics, University of Barcelona, Barcelona, Spain Correspondence Haibin Lu, State Key Laboratory of Crop Abstract Stress Biology for Arid Areas, College of Diverse pathogen effectors convergently target conserved components in plant Agronomy, Northwest A&F University, Yangling, Shaanxi 712100, China. immunity guarded by intracellular nucleotide-­binding domain leucine-­rich repeat Email: luhaibin011@hotmail.com receptors (NLRs) and activate effector-­triggered immunity (ETI), often causing cell Cuizhu Zhao, State Key Laboratory of Crop death. Little is known of the differences underlying ETI in different plants triggered Stress Biology for Arid Areas, College of Agronomy, Northwest A&F University, by the same effector. In this study, we demonstrated that effector RipAW triggers Yangling, China. ETI on Nicotiana benthamiana and Nicotiana tabacum. Both the first 107 amino acids Email: zhaocuizhu2002@163.com (N1-­107) and RipAW E3-­ligase activity are required but not sufficient for triggering Qin Chen, College of Food Science and Engineering, Northwest A&F University, ETI on N. benthamiana. However, on N. tabacum, the N1-­107 fragment is essential Yangling, China. and sufficient for inducing cell death. The first 60 amino acids of the protein are not Email: chenpeter2289@nwsuaf.edu.cn essential for RipAW-­triggered cell death on either N. benthamiana or N. tabacum. Funding information Furthermore, simultaneous mutation of both R75 and R78 disrupts RipAW-­triggered Shaanxi the Nature Science Basic Research Plan, Grant/Award Number: 2020JM-­160; ETI on N. tabacum, but not on N. benthamiana. In addition, N. tabacum recognizes more the Fundamental Research Fund for the RipAW orthologs than N. benthamiana. These data showcase the commonalities and Central Universities of China, Grant/Award Number: Z109021706; the Programme of specificities of RipAW-­activated ETI in two evolutionally related species, suggesting Introducing Talents of Innovative Discipline Nicotiana species have acquired different abilities to perceive RipAW and activate to Universities (Project 111) from the State administration of Foreign Experts Affairs, plant defences during plant–­pathogen co-­evolution. Grant/Award Number: B18042; External Science and Technology Cooperation Yang Niu, Shouyang Fu, Gong Chen, and Huijuan Wang contributed equally. This is an open access article under the terms of the Creat​ive Commo​ns Attri​butio​n-­NonCo​mmerc​ial-­NoDerivs License, which permits use and distribution in any medium, provided the original work is properly cited, the use is non-­commercial and no modifications or adaptations are made. © 2021 The Authors. Molecular Plant Pathology published by British Society for Plant Pathology and John Wiley & Sons Ltd. 188 |  wileyonlinelibrary.com/journal/mpp Mol Plant Pathol. 2022;23:188–203. NIU et al. | 189 Program of Ningxia Academy of Agriculture KEYWORDS and Forestry Sciences, Grant/Award Number: DW-­X-­2018012; the Start-­up cell death, E3 ligase, effector, effector-­triggered immunity, Nicotiana, Ralstonia solanacearum, Funds of Northwest A&F University, RipAW Grant/Award Number: Z111021601; National Natural Science Foundation of China, Grant/Award Number: 31901573; the Special Research project from Shaanxi province, Grant/Award Number: F2020221001; the Key Research and Development plan of Ningxia Province, Grant/Award Number: 2021BEF02033 1 | I NTRO D U C TI O N and suppressing plant immune responses (Fujiwara et al., 2016; Mukaihara et al., 2016; Sang et al., 2018), and RipAC interrupts Through evolution, plants have acquired two layers of immunity that MAPK-­mediated phosphorylation of SGT1 to suppress ETI (Yu et al., enable them to perceive pathogens and induce plant defences. In 2020). Some effectors, such as Brg11, RipAK, and RipI, hijack the the first layer, pattern recognition receptors (PRRs) localized on the metabolism to promote R. solanacearum growth in planta (Wang plasma membrane directly detect conserved microbe-­associated et al., 2021; Wu et al., 2019; Xian et al., 2020). molecular patterns (MAMPs), resulting in activation of pattern-­ Protein modification by ubiquitin is one of the most important triggered immunity (PTI) (Wan et al., 2019). In turn, pathogens use posttranslational modifications in eukaryotes. It modulates the sta- effector proteins, which are delivered into the plant cell cytoplasm, bility, activity, and sublocalization of target proteins (Kerscher et al., to perturb physiological processes and suppress PTI in order to 2006; Zheng & Shabek, 2017). Ubiquitination is accomplished by thrive inside plant hosts. Plants have evolved intracellular resistance the sequential function of three enzymes: a ubiquitin-­activating en- (R) proteins to directly or indirectly perceive effectors, triggering zyme (E1), a ubiquitin-­conjugating enzyme (E2), and a ubiquitin ligase effector-­triggered immunity (ETI) (Jones & Dangl, 2006). ETI is often (E3). E3 ligases catalyse the transfer of ubiquitin from an E2 to ly- accompanied with a hypersensitive response (HR), a local cell death sine residues of the target substrates. E3 ligases have been catego- at the pathogen entry site (Balint-­Kurti, 2019). Most R proteins are rized into the REALLY NEW INTERESTING GENE (RING) family and nucleotide-­binding domain leucine-­rich repeat receptors (NLRs), HOMOLOGUES TO THE E6AP CARBOXYL TERMINUS (HECT) fam- whose activation and downstream signalling mechanisms are being ily. RING-­t ype E3 ligases directly transfer ubiquitin from E2 to their intensively studied (Jubic et al., 2019; Zhou & Zhang, 2020). substrates, whereas the HECT-­t ype E3 ligases form a thioester-­linked Ralstonia solanacearum is a soilborne pathogen that causes de- intermediate with ubiquitin before ubiquitin attaches to the sub- structive vascular bacterial wilt disease on plants. The pathogen strate (Zheng & Shabek, 2017). Albeit ubiquitination does not exist in infects more than 200 plant species, including potato, tomato, to- prokaryotes, effectors from many pathogens mimic E3 ligases to pro- bacco, and many ornamental plants (Genin & Denny, 2012). R. so- mote their virulence on hosts (Spallek et al., 2009). Examples include lanacearum is considered one of the most important plant bacterial Pseudomonas syringae AvrPtoB, XopL from Xanthomonas campestris, pathogens worldwide due to its long persistence in the soil, un- and RipG7 effectors from R. solanacearum. AvrPtoB belongs to the usually large host range, broad geographical distribution, and un- RING E3 ligase family and suppresses Fen-­mediated ETI on tomato available resistant crop varieties (Mansfield et al., 2012). The main and PRRs-­mediated PTI in Arabidopsis by ubiquitinating and degrad- virulence determinants of R. solanacearum are the type III secretion ing Fen and PRRs through the 26S proteasome (Abramovitch et al., system (T3SS) and its translocated effector proteins. The genome 2006; Gimenez-­Ibanez et al., 2009; Gohre et al., 2008; Janjusevic of R. solanacearum is estimated to encode about 70 T3SS effectors et al., 2006). XopL acts as a RING-­finger E3 ligase, whose E3 activ- (Genin & Denny, 2012). To date very few R. solanacearum effectors ity is required to eliminate stromules and induce plastid clustering have been functionally characterized in planta (Landry et al., 2020). (Erickson et al., 2018; Singer et al., 2013). RipG7 also harbours E3 For example, AWR5 acts as an inhibitor of the target of rapamycin ligase activity on still unknown substrates and plays a crucial role pathway and enhances plant sensitivity to R. solanacearum (Popa in pathogenicity on Arabidopsis, tomato, and Medicago truncatula et al., 2016). Simultaneously knocking out seven RipG effector genes (Angot et al., 2006). Recently, a conserved cysteine-­dependent E3 decreases R. solanacearum virulence on tomato and Arabidopsis ligase domain has been identified in animal and plant bacterial effec- (Angot et al., 2006). The effector PopP2 decreases the DNA affinity tors such as SspH2 and SspH1 from Salmonella, IpaH9.8 and IpaH3 of defence-­promoting WRKY transcription factors by acetylation, from Shigella, and NopM from Rhizobium. They are named as the repressing plant defence in Arabidopsis. The RPS4/RRS1-­R receptor NEW E3 LIGASE (NEL) family because of their difference in both pair mimics PopP2 substrates to capture PopP2 and activate plant structure and mechanism from all known E3 ligases (Quezada et al., defence (Le Roux et al., 2015; Sarris et al., 2015). Effector RipAY de- 2009; Singer et al., 2008; Xin et al., 2012; Zhu et al., 2008). RipAW grades glutathione in plant cells through its γ-­glutamyl cyclotrans- from R. solanacearum RS1000 was reported to be a new member ferase (GGCT) activity, altering the intracellular redox environment of the NEL family and to suppress PTI in Nicotiana benthamiana 190 | NIU et al. (Nakano et al., 2017). It has also been demonstrated that another effectors RipE1 and RipAW have been shown to subvert the flg22-­ R. solanacearum effector, RipV2, contains a NEL domain and plays a triggered reactive oxygen species (ROS) burst, callose deposition, and critical role in R. solanacearum infection on potato (Chen et al., 2021). defence-­related gene expression on N. benthamiana (Jeon et al., 2020; Species of Nicotiana are not only hosts for some R. solanacearum Nakano & Mukaihara, 2019a; Nakano et al., 2017). Other effectors, strains but also a very useful model to test gene function by the such as RipB, RipE1, PopP1, and AvrA, are recognized by Nicotiana Agrobacterium-­mediated transient expression assay. This transient ex- spp., leading to activation of ETI and cell death (Jeon et al., 2020; pression system is widely used to study R. solanacearum effector func- Nakano & Mukaihara, 2019b; Poueymiro et al., 2009; Sang et al., tion in hosts (Bally et al., 2018; Sabbagh et al., 2019). R. solanacearum 2020; Sun et al., 2020; Zheng et al., 2019; Zhuo et al., 2020). Some NIU et al. | 191 F I G U R E 1 RipAW triggers cell death on Nicotiana benthamiana. Five-­week-­old Nicotiana spp. were leaf-­inoculated with Ralstonia solanacearum strains or agrobacteria containing the effector constructs. (a, b) R. solanacearum strains GMI1000, CFBP2957, and CIP301 trigger cell death in N. benthamiana (a) and Nicotiana tabacum (b). Leaves inoculated with R. solanacearum strains were digitally photographed at 4 days postinoculation (dpi). (c) RipAW, but not RipAU, RipAF1, and RipAZ1, is able to trigger cell death in N. benthamiana. The photographs of leaves expressing those effectors were taken with a camera at 3 dpi. (d) Expression of RipAW enhances accumulation of NbPR1. The total RNA was extracted from the agroinfiltrated N. benthamiana leaves with RipAW and green fluorescent protein (GFP) control at 24 hours postinoculation (hpi). The expression of NbPR-­1 was determined by reverse transcription quantitative PCR and normalized to NbActin. (e) Expression of RipAW limits R. solanacearum JY1 growth. RipAW and control GFP were agroinfiltrated into N. benthamiana leaves. After 24 h, R. solanacearum JY1 strain solution (105 cfu) was infiltrated into the same leaves. Then the JY strain growth in leaves expressing RipAW was measured at the indicated times. Significant differences were evaluated with Student's t test (**p < 0.01). (f) NbSGT1 is required for RipAW-­triggered cell death on N. benthamiana. The first two true leaves of 2-­week-­old N. benthamiana plants were infiltrated with agrobacteria carrying TRV1, TRV2::NbSGT1, TRV2::GFP, and TRV2::NbPDS. Then NbSGT1-­silencing plants were agroinfiltrated with RipAW and investigated for RipAW-­triggered cell death. The picture of RipAW-­triggered cell death on N. benthamiana was taken at 5 dpi. All experiments were repeated three times with similar results. Strain names have been shortened in the image as follows: GMI1000 (GM), CFBP2957 (CF), CIP301 (CP), IPO1609 (IP), UW551 (UW), NCPPB3987 (NC), and UY031 (UY). H represents the number of leaves showing cell death, T represents the total number of R. solanacearum inoculated leaves R. solanacearum effectors have acquired the ability to repress ETI. For Landry et al., 2020; Sang et al., 2020), we preferred to explore example, RipAK targets catalases and inhibits their activities to abro- new cell death-­t riggering effectors. We guessed that the effectors gate R. solanacearum-­triggered HR on Nicotiana tabacum ‘Xanthi’ (Sun might be conserved among cell death-­t riggering strains but absent et al., 2017) and RipAY degrades cellular glutathione to inhibit RipE1-­ from those strains unable to cause this response. Because the triggered cell death on N. benthamiana (Sang et al., 2020). genomes of CIP301 and NCPPB3987 are not available on public Although several R. solanacearum effectors have been shown to databases, we searched which effectors were present in GMI1000 trigger cell death in different Nicotiana species (Jeon et al., 2020; and CFBP2957 but absent in UW551 and UY031. Using the Ralsto Landry et al., 2020), the molecular mechanism underlying this T3E tool (www.ralst​o-­T3E.org) (Sabbagh et al., 2019), we identi- phenomenon has not been deeply investigated. In this study we fied only seven effectors that met this criterion: RipAW, RipAZ1, describe that RipAW, a R. solanacearum GMI1000 effector, uses dif- RipAU, RipAF1, RipA4, RipS4, and RipJ. To determine whether ferent ways to trigger cell death in N. benthamiana and N. tabacum, these effectors triggered cell death, RipAW, RipAZ1, RipAU, and suggesting that different strategies have evolved in a single plant RipAF1 from the strain GMI1000 and green fluorescent protein genus to detect the same pathogen effector. (GFP) as a control were transiently expressed in N. benthamiana by agroinfiltration. We found that RipAF1, RipAZ1, RipAU, or GFP did not trigger any visible cell death, but RipAW induced a very strong 2 | R E S U LT S tissue collapse and ion leakage on N. benthamiana (Figures 1c and S1). This observation suggested that RipAW might contribute to 2.1 | Cell death on N. benthamiana caused by the GMI1000 and CFBP2957-­induced cell death in Nicotiana spp. effector RipAW is NbSGT1-­dependent To test whether cell death triggered by RipAW results from the activation of plant immune responses, we checked the expression R. solanacearum GMI1000 (phylotype I) triggers a very strong of PATHOGENESIS-­RELATED 1 (PR1) on N. benthamiana, used as a HR on N. benthamiana and N. tabacum, while the distantly re- marker gene of salicylic acid-­mediated plant immunity (Ding & Ding, lated strain UW551 (phylotype II) does not trigger this response 2020). RipAW expression significantly increased the expression (Poueymiro et al., 2009). To find out whether other phylotype II level of NbPR1 compared with that of GFP (Figure 1d). To determine strains fail to trigger cell death in both Nicotiana species, bacte- whether or not the immune response triggered by RipAW affects rial solutions of strains CIP301 and CFBP2957 (phylotype IIA) and the proliferation of R. solanacearum in plants, we measured R. sola- UY031, UW551, IPO1609, and NCPPB3987 (phylotype IIB) as well nacearum JY1 growth on RipAW-­expressing N. benthamiana leaves. as the control GMI1000 were directly infiltrated into the leaves Compared with GFP control, the expression of RipAW significantly of Nicotiana spp. Strains GMI1000, CIP301, and CFBP2957, but restricted JY1 growth (Figure 1e). The expression of GFP and RipAW not those belonging to phylotype IIB (UY031, UW551, IPO1609, was detected by western blot (Figure S2). This suggests that RipAW and NCPPB3987) induced strong cell death in N. benthamiana and can be recognized by N. benthamiana, activating plant immunity and N. tabacum at 3 days postinoculation (dpi; Figure 1a,b). However, cell death, which is a typical feature of ETI (Jones & Dangl, 2006). cell death trigged by CIP301 in both plants was weaker than that SGT1 plays a critical role in ETI signalling by regulating the stability caused by GMI1000 and CFBP2957 (Figure 1a,b). A functional of NLR proteins (Azevedo et al., 2002; Holt et al., 2005). To examine T3SS is required for cell death triggered by the strain GMI1000 on whether or not RipAW-­triggered cell death is dependent on NbSGT1, Nicotiana species (Poueymiro et al., 2009), implying that the ob- we investigated RipAW-­triggered cell death in N. benthamiana after served cell death in Figure 1a could be attributed to the activity of knocking down NbSGT1 by virus-­induced gene silencing (VIGS). We an effector. Instead of analysing the natural variation of all known found that the cell death triggered by RipAW was specifically abol- cell death-­t riggering effectors among strains (Jeon et al., 2020; ished in NbSGT1-­silenced plants, but not in control plants (Figures 1f 192 | NIU et al. and S3). This suggests that RipAW might be perceived by (an) NLR strain RS1000 (RipAWRS) induced leaf chlorosis on N. benthamiana receptor(s) eliciting ETI in N. benthamiana. instead of cell death, and that this phenotype was dependent on its E3 ligase activity (Nakano et al., 2017). We decided to test whether RipAWGM-­triggered cell death was caused by a loss of its ligase activ- 2.2 | E3 ligase activity is required, but not sufficient ity. Recombinant glutathione S-­transferase (GST)-­tagged RipAWGM for RipAW-­triggered cell death in N. benthamiana purified from Escherichia coli was directly incubated with human E1, human UbcH5B (E2), and the substrate HA-­Ubiquitin in vitro, and the RipAW from the strain GMI1000 (RipAWGM) induced strong cell reaction mixture was subjected to western blot. A poly-­ubiquitin lad- death (Figure 1), while a previous report showed that RipAW from der was detected with anti-­HA antibody (Figure S4) indicating that NIU et al. | 193 F I G U R E 2 E3 ligase activity is essential for RipAW-­triggered cell death in Nicotiana benthamiana, but not in Nicotiana tabacum. RipAW and its derivatives were transiently expressed in leaves of 5-­week-­old Nicotiana spp. by agroinfiltration. (a) Schematic illustration of RipAW and its derivatives RipAW (ΔNEL), RipAW (ΔN1-­107), and RipAW (C177A). (b) Loss of E3 ligase activity abolishes RipAW-­triggered cell death in N. benthamiana. Cell death phenotype on N. benthamiana was photographed at 4 days postinoculation (dpi) with a digital camera. (c) Cell death in (b) was quantified by investigating electrolyte leakage. The N. benthamiana leaves expressing RipAW and its derivatives were harvested at 4 dpi and the conductivity was measured by a conductimeter. Electrolyte leakage was manifested as the percentage of sample conductivityunboiling/conductivityboiling (the following assay is similar). Different letters above the columns indicate significant differences between conditions (one-­way analysis of variance, Tukey's test, p < 0.01). (d) E3 ligase activity is not required for RipAW-­triggered cell death in N. tabacum. Photographs of cell death in N. tabacum induced by RipAW and its mutants were taken at 4 dpi. (e) Cell death in (d) was quantified by measuring electrolyte leakage. The agroinfiltrated leaves were collected at 4 dpi and their electrolyte leakage was measured with a conductimeter. Asterisks indicate significant differences between RipAW mutants and RipAW (FL) (Student's t test, ****p < 0.001). (f) Detection of RipAW and its derivatives in N. benthamiana with western blot using anti-­GFP antibody. Leaves expressing RipAW and its mutants were harvested and protein was extracted around 24 h after agroinfiltration. H is the number of leaves with cell death and T is the total number of infiltrated leaves. All experiments were repeated three times with the similar results RipAWGM is a functional E3 ligase. Importantly, the poly-­ubiquitin death in N. tabacum. In line with this, the E3 ligase mutant RipAW ladder was completely abrogated when cysteine 177 was substituted (C177A) still kept the ability to induce cell death (Figure 2d,e). These by an alanine or a serine (Figure S4), suggesting that this is the key results indicate that in contrast to N. benthamiana, the RipAW catalytic residue for RipAW E3 ligase activity, consistent with the 107 N-­terminal amino acids fragment, but not E3 ligase activity, is NEL family features (Nakano et al., 2017; Quezada et al., 2009). required for RipAW-­triggered cell death in N. tabacum, suggesting Subsequently, to understand whether the E3 ligase domain is suf- that N. tabacum and N. benthamiana employ different strategies to ficient for RipAW-­triggered cell death, we split the full length RipAW initiate the cell death programme on RipAW recognition. into an N-­terminal fragment spanning 1–­107 residues [RipAW (ΔNEL)] and an NEL-­domain containing fragment [RipAW (ΔN1-­107)] (Figure 2a). Neither RipAW (ΔNEL) nor RipAW (ΔN1-­107) caused leaf tissue col- 2.4 | The first 60 amino acids of lapse (Figure 2b). However, electrolyte leakage assay showed RipAW RipAW are not required for the activation of cell (ΔN1-­107) induced higher conductivity than RipAW (ΔNEL) (Figure 2c). death on Nicotiana spp. Western blot analysis with anti-­GFP antibody confirmed the expression of RipAW and its variants in N. benthamiana (Figure 2f). These results The N1-­107 fragment is required for RipAW-­triggered cell death in both indicate that both the N-­terminus and the NEL domain are required N. benthamiana and N. tabacum (Figure 2). To further check the effect for RipAW-­triggered cell death in N. benthamiana. To explore the role of the N-­terminus on the function of RipAW on Nicotiana spp., we of the E3 ligase activity in RipAW-­triggered cell death in plants, we an- generated a series of truncated RipAW versions by deleting 30 amino alysed cell death in N. benthamiana leaves expressing RipAW (C177A). acids from its N-­terminus and investigated cell death by transiently In contrast to clear tissue collapse caused by RipAW, the C177A muta- expressing them in N. benthamiana (Figure 3a). Leaves infiltrated with tion completely abolished cell death symptoms (Figure 2b). We further RipAW (ΔN1-­60) displayed strong cell death phenotypes, while those quantified cell death by measuring ion leakage and found that the con- infiltrated with RipAW (ΔN1-­30) and RipAW (ΔN1-­90) looked healthy ductivity caused by RipAW (C177A) was comparable to the GFP control (Figure 3b). Consistent with the leaf symptoms, ion leakage assays and clearly lower than that caused by RipAW (Figure 2c). These results showed that leaf conductivity triggered by RipAW (ΔN1-­60) was indis- indicate that the E3 ligase activity of RipAW is essential for cell death in cernable from that of RipAW, which was clearly higher than RipAW N. benthamiana. (ΔN1-­30) and RipAW (ΔN1-­90) (Figure 3c). Similar to N. benthamiana, deletion of these first 60 amino acids did not affect RipAW-­triggered cell death in N. tabacum, but lack of the first 90 amino acids blocked 2.3 | The 107 N-­terminal amino acid residues, not cell death appearance (Figure 3d). Western blot analysis indicated its E3 ligase activity, is required for RipAW-­triggered that RipAW (ΔN1-­30) was not expressed in Nicotiana spp., which may cell death on N. tabacum underpin the failure of RipAW (ΔN1-­30) to elicit cell death (Figure 3e). Interestingly, all truncated proteins were detected as a single band, To test whether RipAW and its variants also trigger cell death in N. ta- while the full-­length RipAW showed two bands (Figure 3e). bacum, we agroinfiltrated them and investigated cell death display in N. tabacum. When the full length of RipAW was expressed, a very clear cell death phenotype appeared compared to the GFP control, 2.5 | RipAW orthologs from different strains which is consistent with the cell death-­inducing ability of the strain exhibit differential abilities to trigger cell death on GMI1000 in N. tabacum (Figure 1b). Unexpectedly, RipAW (ΔNEL), Nicotiana species but not RipAW (ΔN1-­107), retained the ability to induce cell death in N. tabacum (Figure 2d,e), implying that the first 107 amino acids frag- RipAW is present in the genomes of 89 of the 155 sequenced R. sola- ment (N1-­107) of RipAW are required and sufficient for eliciting cell nacearum strains (Sabbagh et al., 2019). To explore if RipAWs from 194 | NIU et al. different R. solanacearum strains have acquired the ability to subvert (Figure S5). This raised the possibility that this N-­terminal peptide may recognition by the immune surveillance system of N. benthamiana, we function in RipAWGM-­induced cell death on N. benthamiana. Because synthesized RipAW orthologs from representative strains in the four charged amino acids normally play very important roles in maintain- phylotypes: RS1000 (RS, phylotype I), CFBP2957 (CF, phylotype IIA), ing protein configurations and in protein–­protein interactions, we Po82 (Po, phylotype IIB), and CMR15 (CM, phylotype III) and PSI07 generated two RipAWGM double point mutants, RipAW (H62D/ (PS, phylotype IV) (Sabbagh et al., 2019), and expressed these RipAW H64D) and RipAW (R75A/R78A), by substituting two histidine and orthologs (RipAWRS, RipAWCM, RipAWPS, RipAWCF, and RipAWPo) by two arginine residues with aspartate and alanine, respectively, in this agroinfiltration on Nicotiana spp. We found that RipAWRS induced region. These mutated versions of RipAW were agroinfiltrated on clear cell death in leaves of N. benthamiana, while other RipAWs did N. benthamiana and cell death was investigated. While the H62D/ not trigger this response (Figure 4a). Western blot analysis showed H64D mutations did not affect cell death caused by RipAW, the that RipAWCM and RipAWPS were undetectable, therefore they expression of RipAW (R75A/R78A) resulted in patchy yellowing of were excluded from subsequent assays (Figure 4c). Consistent with the leaves, suggesting that the R75A and R78A mutations attenuate the observed macroscopic cell death, leaves expressing RipAWCF RipAW-­triggered cell death in N. benthamiana (Figure 6a). However, and RipAWPo showed significantly lower conductivities than those quantification of cell death by ion leakage did not show significant expressing RipAWGM and RipAWRS (Figure 4b). This indicates that differences between RipAW and RipAW (R75A/R78A), although the RipAWCF and RipAWPo might escape perception by N. benthamiana. conductivity value triggered by RipAW (R75A/R78A) was slightly In contrast, when inoculated on N. tabacum leaves, RipAWPo caused lower (Figure 6b). Both mutated variants of RipAW showed similar a strong cell death phenotype (Figure 4d). This reinforces the idea protein levels to the wild-­t ype protein (Figure 6e). These results in- that these two Nicotiana species have evolved different surveillance dicate that the arginine R75/R78 and histidine H62/H64 residues of systems to recognize members of the RipAW family. RipAW have a negligible role in cell death elicited by this effector in N. benthamiana. Considering that the N-­terminal fragment (1–­107) of RipAW is suf- 2.6 | N-­terminus of RipAW orthologs show ficient for triggering cell death in N. tabacum, we speculate that the differential abilities to induce cell death in N. tabacum R75A/R78A mutation might alter RipAW-­triggered cell death in this species. To test this idea, the RipAW (R75A/R78A) and RipAW (H62D/ The N1-­107 region is responsible for cell death induced by RipAWGM in H64D) were expressed in N. tabacum leaves. Similar to what we ob- N. tabacum (Figure 2). Therefore we speculated that natural variation on served in N. benthamiana, the H62D/H64D mutation did not alter the N-­termini of allelic RipAWs might enable the RipAW N-­terminus to RipAW-­triggered cell death in N. tabacum. Interestingly, the R75A/ overcome the N. tabacum recognition system. Because RipAWCM and R78A mutation completely suppressed RipAW-­elicited cell death RipAWPS could not be detected by western blot (Figure 4), we investigated (Figure 6c,d). These results suggest that in contrast to N. benthamiana, the role in cell death of RipAWGM, RipAWRS, RipAWCF, and RipAWPo R75 and R78 play an important role in RipAWGM-­triggered cell death in N-­termini. We aligned the protein sequences of RipAW orthologs N. tabacum. These data again imply that N. tabacum and N. benthami- (Figure S5), identified and cloned their N-­termini, and expressed them ana use different strategies to detect RipAW and initiate ETI. in N. tabacum by agroinfiltration. Both cell death symptoms and ion leakage assays showed that RipAW (N1-­107)GM, RipAW (N1-­107)RS, and RipAW(N1-­90)CF elicited strong cell death and ion leakage in N. tabacum 2.8 | Loss of function RipAW does not leaves while RipAW (N1-­90)Po did not cause any cell death in N. tabacum change the virulence/avirulence of R. solanacearum on (Figure 5a,b). These data suggest that although N. tabacum possesses a solanaceous plants very powerful surveillance system to detect the divergent N-­termini of RipAW orthologs and activate cell death, some R. solanacearum strains To determine the contribution of RipAW in GMI1000-­triggered cell have acquired supposedly undetectable variations at RipAW N-­termini death, we generated a mutant strain in which the RipAW coding se- as a strategy to overcome this surveillance system during the arms race quence was completely deleted from its genome (ΔripAW), and in- between R. solanacearum and its plant hosts. vestigated the effect of this mutation on cell death in Nicotiana spp. The ΔripAW strain was able to trigger cell death in both N. benthami- ana and N. tabacum to similar levels as the wild-­t ype strain GMI1000 2.7 | N-­terminal arginines R75 and R78 are required (Figure S6a). Consistent with cell death symptoms, no difference for RipAWGM-­triggered cell death in N. tabacum, in conductivity was detected between the wild type strain and the but not in N. benthamiana ΔripAW strain (Figure S6b), suggesting that GMI1000-­triggered cell death was mediated by additional effectors besides RipAW. This To explore how RipAWCF and RipAWPo escape N. benthamina de- result is consistent with previous findings demonstrating that RipB, tection, we analysed their protein sequences using Cluster Omega. AvrA, PopP1, and RipE1 effectors trigger cell death in N. benthami- Strikingly, in comparison to RipAWGM and RipAWRS, RipAWCF ana or N. tabacum (Jeon et al., 2020; Nakano & Mukaihara, 2019a, and RipAWPo lacked a 17 amino acid fragment after position 61 2019b; Poueymiro et al., 2009; Sang et al., 2020). NIU et al. | 195 F I G U R E 3 The first 60 N-­terminal amino acids are not required for RipAW-­triggered cell death on Nicotiana spp. RipAW N-­terminal deletion mutants were agroinfiltrated into leaves of 5-­week-­old N. benthamiana and N. tabacum. (a) Schematic illustration of RipAW N-­terminal mutants. (b) Deletion of the first 60 amino acids does not affect RipAW-­triggered cell death on N. benthamiana. Leaves agroinfiltrated with RipAW mutants were digitally photographed at 4 days postinoculation (dpi). (c) Cell death triggered by the truncated RipAW variants in (b) was quantified by measuring ion leakage. N. benthamiana leaves expressing the truncated RipAW variants were collected at 4 dpi and immersed into double-­deionized water. Then their ion leakage was measured with a conductimeter. Different letters above the columns indicate significant differences between conditions (one-­way analysis of variance, Tukey's test, p < 0.01). (d) The first 60 amino acids are not essential for RipAW-­triggered N. tabacum cell death. N. tabacum leaves agroinfitrated with the truncated RipAW mutants were photographed at 5 dpi. (e) Detection of RipAW and its mutant variants in N. benthamiana with western blot using anti-­FLAG antibody. The leaves expressing RipAW and its mutants were harvested and protein was extracted around 24 h after agroinfiltration. H is the number of leaves displaying cell death and T is the total number of infiltrated leaves. All experiments were repeated three times with similar results Because R. solanacearum GMI1000 also causes brown rot disease plants (Wang et al., 2019). Similar to the wild-­t ype strain, the mu- on potato, we used an in vitro potato infection system to investi- tant strain caused severe wilting symptoms on potato (Figure S6c,d). gate whether this deletion affected the pathogen's fitness on potato In line with this, R. solanacearum wild-­t ype strain and ΔripAW strain 196 | NIU et al. F I G U R E 4 RipAW orthologs exhibit differential abilities to trigger cell death on Nicotiana spp. RipAW members from different strains were transiently expressed on leaves of 5-­week-­old Nicotiana spp. by agroinifiltration. (a) RipAWRS orthologs triggered strong cell death on N. benthamiana. Leaves expressing RipAW orthologs were photographed at 4 days postinoculation (dpi). (b) Cell death in (a) was quantified by measuring electrolyte leakage. The conductivities of N. benthamiana leaves agroinfiltrated with RipAW orthologs were measured with a conductimeter at 4 dpi. Different letters above columns indicate significant differences between conditions (one-­way analysis of variance, Tukey's test, p < 0.01). (c) The expression of RipAW orthologs in N. benthamiana was detected by western blot using anti-­FLAG antibody. Leaves expressing RipAW orthologs were harvested and protein was extracted at 24 h after agroinfiltration. (d) RipAWRS and RipAWPo triggered N. tabacum cell death. Leaves agroinfiltrated with RipAW family members were photographed at 5 dpi. H is the number of leaves with cell death and T is the total number of infiltrated leaves. The experiments were repeated three times with similar results NIU et al. | 197 3 | D I S CU S S I O N Effectors are the most important virulence determinants of gram-­negative bacterial pathogens infecting plants. Occasionally, effectors betray the pathogen and activate ETI, resulting in failure of pathogen invasion. Except for a few examples, it remains largely unknown how different effectors together with different host targets induce ETI and cell death. Here, we characterized how RipAWGM from the bacterial wilt pathogen R. solanacearum triggered cell death in Nicotiana species (N. benthamiana and N. tabacum) in different ways. We found that the recognition of RipAW by N. benthamiana up-­regulates defence-­related gene NbPR1 expression and activates cell death and plant immunity. Thus, the cell death triggered by RipAW is dependent on NbSGT1. We also showed that RipAWRS elicits very strong cell death in N. benthamiana and N. tabacum (Figure 4). However, Nakano and colleagues reported that RipAWRS induced leaf chlorosis, but not cell death in N. benthamiana (Nakano et al., 2017). This discrepancy could be attributed to the different experimental conditions used for plant growth (23°C and 70% humidity in our assay, 25°C and 50% humidity in Nakano's assay) and/or the Agrobacterium inoculum used for leaf infiltration (OD600 = 1 in our assay and OD600 = 0.5 in Nakano's assay). The RipAWGM protein appears as two isoforms when expressed in N. benthamiana (Figures 3e and 6e). This could suggest that RipAWGM may undergo posttranslational modification(s) on its entry to the host cells. However, these two isoforms do not always have similar intensities (Figures S2 and 6e). Because the RipAW-­FLAG size (471 amino acids) overlaps with that of the large subunit of RuBisCO, the observed two RipAW-­FLAG bands might be artificial because the RuBisCO large subunit split one band into two, a phenomenon known as a “hamburger band”. An alternative explanation is that RipAW undergoes proteolytic cleavage in planta. More experiments (such as immunoprecipitation assays) need to be performed to verify the nature of the two bands observed. Deletion of RipAW did not have any noticeable effect on F I G U R E 5 RipAW (N1-­90)CF and RipAW (N1-­107)RS induce strong GMI1000-­triggered HR on Nicotiana species (Figure S6a,b). This cell death on Nicotiana tabacum. (a) Expression of RipAW (N1-­90)CF indicates that besides RipAW, other effectors also contribute to and RipAW (N1-­107)RS caused strong cell death. Leaves transiently the cell death phenotype observed on Nicotiana species on infec- expressing the N-­terminus of RipAW orthologs were photographed at 4 days postinoculation (dpi). (b) Cell death in (a) was quantified tion. Consistent with this, several effectors, such as AvrA, PopP1, by testing electrolyte leakage. Leaves agroinfiltrated with the RipAB, RipB, RipE1, and RipI, have been sequentially shown to trig- N-­terminus of RipAW orthologs were collected at 4 dpi and ger cell death in Nicotiana species (Landry et al., 2020; Poueymiro immersed in double-­deionized water. Ion leakage was measured et al., 2014). Interestingly, some R. solanacearum effectors sup- with a conductimeter. Different letters above the columns indicate press cell death triggered by other effectors in Nicotiana. For ex- significant differences between conditions (one-­way analysis of ample, RipAY and RipAC block RipE1-­induced Nicotiana leaf cell variance, Tukey's test, p < 0.01). H is the number of dead leaves and T is the total number of infiltrated leaves. The experiments were death by degrading glutathione and inhibiting SGT1-­mediated repeated three times with similar results MAPK activation (Sang et al., 2020; Yu et al., 2020). RipAK inhib- its R. solanacearum-­triggered cell death on N. tabacum (Sun et al., reached similar growth levels in potato stems at 4 dpi, regardless of 2017). It will be interesting to know the inhibition spectrum of the inoculation method used (with or without wounding the roots) these effectors on the known R. solanacearum effector-­triggered (Figure S6c,d). These results indicate that RipAW deficiency does not cell death. In potato, RipAW deficiency did not decrease GMI1000 seem to affect the proliferation of R. solanacearum GMI1000 in po- pathogenicity (Figure S6c,d), similar to what has been previously tato plants. shown in pepper and tomato (Nakano et al., 2017). Notably, 198 | NIU et al. F I G U R E 6 Arginine in positions 75 and 78 of RipAW is essential for cell death in Nicotiana tabacum. The indicated RipAW versions were agroinfiltrated into leaves of 5-­week-­old Nicotiana spp. (a) Substitution of both R75/R78 for alanine and H62/H64 for aspartate did not affect tissue collapse elicited by RipAW in N. benthamiana. Cell death symptoms on leaves expressing RipAW (R75/78A) and RipAW (H62/64D) were photographed at 4 days postinoculation (dpi). (b) Cell death induced by RipAW (R75/78A) and RipAW (H62/64D) on N. benthamiana was quantified by electrolyte leakage. Leaves expressing RipAW (R75/78A) and RipAW (H62/64D) were collected at 4 dpi and immersed in double-­deionized water. The conductivity of samples was measured by a conductimeter. Different letters above columns indicate the significant difference (one-­way analysis of variance, Tukey's test, p < 0.01). (c) Simultaneous mutation on R75 and R78 abolished cell death triggered by RipAW on N. tabacum. Cell death symptoms caused by RipAW (R75/78A) were photographed and the leaves showing cell death were counted at 4 dpi. (d) Cell death in (c) was quantified by electrolyte leakage assay. N. tabacum leaves agroinfiltrated with RipAW and its point mutants were harvested at 4 dpi and put into double-­deionized water. Electrolyte leakage was detected by a conductimeter. Asterisks indicate statistically significant differences between RipAW and its variants (Student's t test, ****p < 0.001). (e) Expression of RipAW and its mutants was detected by western blot with anti-­FLAG antibody in N. benthamiana. Leaves expressing RipAW point mutants were harvested and protein was extracted 24 h after agroinfiltration. H is the number of leaves with cell death and T is the total number of infiltrated leaves. The experiments were repeated three times with the similar results overexpressing RipAW in Arabidopsis showed enhanced plant functional redundancy with some of the 70 R. solanacearum effec- sensitivity to P. syringae DC3000, suggesting that RipAW inhibits tors (Sabbagh et al., 2019). In agreement with this, lack of individual plant immunity (Nakano et al., 2017). This implies that RipAW has effectors in P. syringae does not change its virulence in terms of NIU et al. | 199 bacterial growth in leaves and bacterial symptoms (Kvitko et al., confers the ability of JY1 to grow on the spectinomycin-­amended me- 2009). In addition, restoration of most individual effectors to a dium. All strains used in this study were grown at 28°C on solid or DC3000 mutant strain lacking 28 effectors does not significantly liquid rich B medium (1% Bacto peptone, 0.1% yeast extract, and 0.1% enhance its virulence on plants (Cunnac et al., 2011). Similarly, casamino acids; Becton Dickinson) with the appropriate antibiotics. disruption of single R. solanacearum effectors showed a very mar- ginal effect on R. solanacearum virulence on plants (Cunnac et al., 2004; Sole et al., 2012). In addition, successful natural infection by 4.2 | Plasmid constructions R. solanacearum is a very complex process that has been divided into four steps: root colonization, vertical movements from roots All primers used in this study are listed in Table S1. Genomic DNA to shoots, circular vascular bundle invasion, and radial apoplast of R. solanacearum GMI1000 was used as the template for PCR spread in the cortex (Planas-­Marques et al., 2020). Because wilting amplification of RipAW and its derivatives, which were cloned symptoms and large bacterial multiplication occur at late stages of into pCambia 6604-­GFP (BamHI and SalI/XhoI) (Bai et al., 2016) or R. solanacearum infection, we cannot exclude the possibility that pCambia1300-­FLAG (KpnI/SacI and BstBI) with a NovoRes plus one-­ RipAW plays a role in root colonization or bacterial movements in- step cloning kit (Novoprotein Scientific Inc.). The RipAW orthologs side the sensitive plant. from strains RS1000, CFBP2957, PSI07, and Po82 were synthesized In this study, we proved the E3 ligase activity of RipAWGM is es- and cloned into pUC57 by the Gene Create company in Wuhan, sential, but not sufficient for triggering cell death on N. benthamiana China. These constructs were then directly digested with KpnI and (Figure 2). On the contrary, the first 107 amino acids of RipAW are BstBI and subcloned into pCambia 1300-­FLAG with KpnI and BstBI. sufficient for cell death induction in N. tabacum, but not in N. ben- All plasmids made in this study are listed in Table S2. thamiana (Figure 2). Furthermore, mutation R72A/R78A blocks RipAW-­triggered cell death in N. tabacum, but not in N. benthamiana (Figure 6). Together, these results indicate that N. tabacum and N. ben- 4.3 | Construction of ΔRipAW strain thamiana employ different surveillance systems for sensing the effec- tor RipAW and activating cell death and ETI. It will be very interesting A 539 bp fragment at the upstream of RipAW start codon and a to further explore how N. benthamiana and N. tabacum differentially 682 bp fragment downstream of the RipAW stop codon were ampli- perceive RipAW and activate ETI. Furthermore, although RipAWPo fied from genomic DNA of R. solanacearum GMI1000 by chimeric and RipAWCF escaped surveillance by the N. benthamiana immune PCR and were cloned into pURF80 vector. The resulting construct system, they were still detected by N. tabacum. This suggests that was introduced into R. solanacearum GMI1000 by natural transfor- the recognition spectrum of N. tabacum on RipAW might be broader mation (Valls et al., 2006). The colonies grown on B medium con- than that of N. benthamiana. In the future, identifying the NLR pro- taining 50 µg/ml kanamycin were selected and cultured in liquid B teins responsible for detecting RipAWs will help in deciphering the medium without antibiotics overnight. Cell cultures were spread on co-­evolution strategies of NLR proteins and their cognate effectors solid B medium containing 5% sucrose. Positive ΔRipAW candidates during the long-­lasting battle between host and pathogens. This will were confirmed by PCR. provide new insights into the molecular mechanism underlying rec- ognition of allelic effectors by NLRs in different plant species during plant–­microbe co-­evolution and valuable information for breeding 4.4 | Ubiquitination assay highly resistant crops based on NLR-­protein mediated plant immunity. R. solanacearum GMI1000 RipAW was cloned into pGEX6p-­1 vec- tor with BamHI/XhoI. PCR-­based site-­directed mutagenesis was 4 | E X PE R I M E NTA L PRO C E D U R E S used to mutate the cysteine 177 of RipAW into an alanine or a ser- ine. The obtained pGEX-­RipAW, pGEX-­RipAW (C177A) and pGEX-­ 4.1 | Plant materials and microbial strains RipAW (C177S) were transformed into E. coli BL21. The GST-­t agged RipAW, RipAW (C177A), and RipAW (C177S) were purified from the Nicotiana spp. plants were sown in soil and grown in controlled IPTG-­induced cell cultures by glutathione resin beads (Sigma) and chambers under 23°C and short-­day conditions with 70% humidity. detected by western blot with anti-­GST antibody (Sigma). Potato explants of cv. Desirée were grown in liquid Murashige & Two micrograms of purified GST-­t agged RipAW and its deriva- Skoog medium with 20% sucrose (MS2) at 22–­23°C, 70% humidity, tives were incubated with 0.5 μg human E1, 2 µg UbcH5B E2, and and long-­day conditions. 2 µg HA-­ubiquitin (Boston Biochem) in 40 µl of reaction buffer R. solanacearum strains GMI1000, CIP301, CFBP2957, UW551, (25 mM Tris.HCl pH 7.5, 50 mM NaCl, 5 mM ATP, 10 mM MgCl2, IPO1905, NCPPB3987, and UY031 are described in Lu et al. (2018). 0.1 mM dithiothreitol) for 2 h at 37°C. The reaction was stopped Wild-­t ype R. solanacearum JY1 strain was isolated from wilting to- by addition of 5× SDS-­loading buffer. The reaction mixtures were bacco plants (Yunyan 87) at Jiangyong county, Hunan province, China. separated by SDS-­PAGE and detected with anti-­GST antibody and The wild-­t ype JY1 strain was transformed with a pHM1 plasmid which anti-­HA antibody (Sigma). 200 | NIU et al. 4.5 | Agrobacterium-­mediated transient Five-­ to 6-­week-­old N. benthamiana plants were directly infiltrated expression assay with the bacterial solution. The inoculated N. benthamiana plants were put back into the infection chamber at 26°C, 75% humidity, Gene transient expression in Nicotiana species was performed and a 16 h light and 8 h dark photoperiod. Cell death symptoms as previously reported (Grosse-­Holz et al., 2018). Agrobacterium were recorded at 3 dpi with a digital camera and quantified with tumefaciens GV3101 carrying RipAW and its derivatives were electrolyte leakage (described above). cultured in liquid LB medium at 28°C overnight, then collected by For potato explants infection we followed the protocol as de- centrifugation and suspended with the inoculation buffer (10 mM scribed in Wang et al. (2019). Briefly, the roots of 2-­ to 3-­week-­old MES pH 5.0, 10 mM MgCl2, 400 nM acetosyringone). Five-­week-­old potato plants grown in liquid MS2 medium were cut with scissors N. benthamiana and N. tabacum were infiltrated with Agrobacterium and plants were subsequently transferred into glass jars containing at an optical density of 1 (OD600 = 1) and covered with a transparent 40 ml of a solution of R. solanacearum (OD600 = 0.01) and kept in the plastic membrane for 2 days and kept in a chamber under 23°C and infection chamber under 26°C, 75% humidity, and 16 h light and 8 h long-­day conditions with 70% humidity for gene expression. dark photoperiod conditions. Photographs were taken with a digital camera and bacterial loads in the aerial part of the infected plants were measured when plants started wilting (Wang et al., 2019). 4.6 | Cell death and ion leakage assay To measure R. solanacearum growth in N. benthamiana, we fol- lowed a R. solanacearum infection protocol (Sang et al., 2020) with To visualize cell death triggered by RipAW and its mutants on minor modifications. Briefly, leaves were first infiltrated with agro- leaves, photographs of the leaves of N. benthamiana and N. tabacum bacteria containing RipAW (OD600 = 0.01), which is resistant to inoculated with agrobacteria were taken with a digital camera at kanamycin. Around 20 h after agroinfiltration, a 105 cfu/ml solu- 4–­5 dpi. For electrolyte leakage measurements, 15 leaf discs per tion of R. solanacearum JY1 with resistance to spectinomycin was sample were harvested, put into three tubes containing 5 ml of inoculated by infiltration into the leaves expressing RipAW. R. so- double-­deionized water (five leaf discs/tuber), and subjected to lanacearum JY1 was isolated from wilting tobacco plants (Yunyan vacuum for 20 min (Szczesny et al., 2010). Then all samples were 87) at Jiangyong county, Hunan province, China and is also patho- left at room temperature for 2 h and the ion leakage caused by genic on N. benthamiana (unpublished data). Twelve leaf discs were RipAW and its derivatives was measured with a conductimeter. collected, weighed, and homogenized in sterile double-­deionized To determine total conductivity of the entire sample, conductivity water. Bacterial cfu were counted after plating on B medium with was measured again after boiling the solution for 20 min (Szczesny spectinomycin. et al., 2010). Q uantification of cell death triggered by RipAW and its mutants was displayed as the percentage of Conductivity unboiling / Conductivity boiling. 4.9 | Gene expression assay Total RNA from leaves of N. benthamiana at the indicated times 4.7 | Virus-­induced gene silencing assay were extracted with TRIzol reagent (Invitrogen) and the first-­ strand cDNA was synthesized with an EvoM-­MLV RT kit (Accurate An equal volume mixture of Agrobacterium (OD600 = 1) containing Biotechnology). Five microlitres of the 20-­fold diluted cDNA and TRV2::NbSGT1, TRV2::GFP, and TRV2::NbPDS with Agrobacterium primers (Table S1) was directly added into the SYBR Green Mix (OD600 = 1) carrying TRV1 was infiltrated into the first two true (Gene Star Co.). The prepared quantitative PCR (qPCR) solutions leaves of 2-­week-­old N. benthamiana, which were then covered were run on a iQ7 real-­time cycler (Life Technologies). Gene ex- in plastic film for 2 days to keep high humidity. After that, the pression was quantified and normalized to NbActin using the 2−ΔΔCt inoculated plants continued growing under 23°C and short-­day method. conditions with 70% humidity for another 3 weeks. When plants inoculated with TRV2::NbPDS turned completely white (Senthil-­ Kumar et al., 2007), TRV2::GFP and TRV2::NbSGT1 gene-­silenced 4.10 | Statistical analysis plants were agroinfiltrated with RipAW and its derivatives. After 3–­4 dpi cell death phenotypes were investigated. Statistical analyses were performed using Graphpad Prism v. 7.0 software. All statistical tests are indicated in the respective figure legends. 4.8 | R. solanacearum infection assay AC K N OW L E D G E M E N T S R. solanacearum strains were grown at 28°C in liquid rich B medium We are grateful to Dr Yu Du (Northwest A&F University) for the overnight. Bacteria were collected by centrifugation and washed with constructs TRV1, TRV2::GFP, TRV2::NbSGT1, and TRV2::NbPDS, and sterile double-­deionized water and adjusted to a final OD600 = 0.01. to Dr Changgen Xie for pCambia6604-­G FP vector and Dr Wu Chen NIU et al. | 201 (College of Plant Protection, Hunan Agricultural University) for in Pseudomonas syringae. Proceedings of the National Academy of Sciences USA, 108, 2975–­2980. R. solanacearum JY1. We also thank the Crop Biology Innovation Cunnac, S., Occhialini, A., Barberis, P., Boucher, C. & Genin, S. (2004) Platform in the Agronomy College and Public Platform in the State Inventory and functional analysis of the large Hrp regulon in Key Laboratory of Crop Stress Biology for Arid Areas at Northwest Ralstonia solanacearum: identification of novel effector proteins A&F University, China. This work was funded by the Nature translocated to plant host cells through the type III secretion sys- Science Basic Research Plan of Shaanxi Province (2020JM-­160), the tem. Molecular Microbiology, 53, 115–­128. Ding, P. & Ding, Y. (2020) Stories of salicylic acid: a plant defense hor- Special Research project from Shaanxi province (F2020221001), mone. Trends in Plant Science, 25, 549–­565. the Program of Introducing Talents of Innovative Discipline to Erickson, J.L., Adlung, N., Lampe, C., Bonas, U. & Schattat, M.H. (2018) Universities (Project 111) from the State administration of Foreign The Xanthomonas effector XopL uncovers the role of microtubules Experts Affairs (B18042) and the External Science and Technology in stromule extension and dynamics in Nicotiana benthamiana. The Plant Journal, 93, 856–­870. Cooperation Program of Ningxia Academy of Agriculture and Fujiwara, S., Kawazoe, T., Ohnishi, K., Kitagawa, T., Popa, C., Valls, M. et al. Forestry Sciences (DW-­X-­2018012), the Start-­up Funds of (2016) RipAY, a plant pathogen effector protein, exhibits robust Northwest A&F University (Z111021601), the Fundamental γ-­γ-­glutamyl cyclotransferase activity when stimulated by eukary- Research Fund for the Central Universities of China (Z109021706), otic thioredoxins. Journal of Biological Chemistry, 291, 6813–­6830. Genin, S. & Denny, T.P. (2012) Pathogenomics of the Ralstonia sola- the Key Research and Development plan of Ningxia Province nacearum species complex. Annual Review of Phytopathology, 50, (2021BEF02033), and the National Natural Science Foundation of 67–­89. China for C.Z. (31901573). Gimenez-­Ibanez, S., Hann, D.R., Ntoukakis, V., Petutschnig, E., Lipka, V. & Rathjen, J.P. (2009) AvrPtoB targets the LysM receptor kinase CERK1 to promote bacterial virulence on plants. Current Biology, DATA AVA I L A B I L I T Y S TAT E M E N T 19, 423–­429. All data supporting the findings of this study are available from the Göhre, V., Spallek, T., Häweker, H., Mersmann, S., Mentzel, T., Boller, T. corresponding author upon reasonable request. et al. (2008) Plant pattern-­recognition receptor FLS2 is directed for degradation by the bacterial ubiquitin ligase AvrPtoB. Current Biology, 18, 1824–­1832. ORCID Grosse-­Holz, F., Madeira, L., Zahid, M.A., Songer, M., Kourelis, J., Yong Zhang https://orcid.org/0000-0003-1820-0927 Fesenko, M. et al. (2018) Three unrelated protease inhibitors en- Marc Valls https://orcid.org/0000-0003-2312-0091 hance accumulation of pharmaceutical recombinant proteins in Haibin Lu https://orcid.org/0000-0002-0613-4697 Nicotiana benthamiana. Plant Biotechnology Journal, 16, 1797–­1810. Holt, B.F., Belkhadir, Y. & Dangl, J.L. (2005) Antagonistic control of disease resistance protein stability in the plant immune system. REFERENCES Science, 309, 929–­932. Abramovitch, R.B., Janjusevic, R., Stebbins, C.E. & Martin, G.B. (2006) Janjusevic, R., Abramovitch, R.B., Martin, G.B. & Stebbins, C.E. (2006) A Type III effector AvrPtoB requires intrinsic E3 ubiquitin ligase ac- bacterial inhibitor of host programmed cell death defenses is an E3 tivity to suppress plant cell death and immunity. Proceedings of the ubiquitin ligase. Science, 311, 222–­226. National Academy of Sciences USA, 103, 2851–­2856. Jeon, H., Kim, W., Kim, B., Lee, S., Jayaraman, J., Jung, G. et al. (2020) Angot, A., Peeters, N., Lechner, E., Vailleau, F., Baud, C., Gentzbittel, L. Ralstonia solanacearum type III effectors with predicted nuclear lo- et al. (2006) Ralstonia solanacearum requires F-­box-­like domain-­ calization signal localize to various cell compartments and modulate containing type III effectors to promote disease on several host immune responses in Nicotiana spp. The Plant Pathology Journal, 36, plants. Proceedings of the National Academy of Sciences USA, 103, 43–­53. 14620–­14625. Jones, J.D.G. & Dangl, J.L. (2006) The plant immune system. Nature, 444, Azevedo, C., Sadanandom, A., Kitagawa, K., Freialdenhoven, A., Shirasu, 323–­329. K. & Schulze-­Lefert, P. (2002) The RAR1 interactor SGT1, an Jubic, L.M., Saile, S., Furzer, O.J., El Kasmi, F. & Dangl, J.L. (2019) Help ­essential component of R gene-­triggered disease resistance. wanted: helper NLRs and plant immune responses. Current Opinion Science, 295, 2073–­2076. in Plant Biology, 50, 82–­94. Bai, Y., Zhu, W., Hu, X., Sun, C., Li, Y., Wang, D. et al. (2016) Genome-­wide Kerscher, O., Felberbaum, R. & Hochstrasser, M. (2006) Modification of analysis of the bZIP gene family identifies two ABI5-­like bZIP tran- proteins by ubiquitin and ubiquitin-­like proteins. Annual Review of scription factors, BrABI5a and BrABI5b, as positive modulators of Cell and Developmental Biology, 22, 159–­180. ABA signalling in Chinese cabbage. PLoS One, 11, e0158966. Kvitko, B.H., Park, D.H., Velasquez, A.C., Wei, C.F., Russell, A.B., Martin, Balint-­Kurti, P. (2019) The plant hypersensitive response: concepts, con- G.B. et al. (2009) Deletions in the repertoire of Pseudomonas sy- trol and consequences. Molecular Plant Pathology, 20, 1163–­1178. ringae pv. tomato DC3000 type III secretion effector genes reveal Bally, J., Jung, H., Mortimer, C., Naim, F., Philips, J.G., Hellens, R. et al. functional overlap among effectors. PLoS Pathogens, 5, e1000388. (2018) The rise and rise of Nicotiana benthamiana: a plant for all Landry, D., Gonzalez-­Fuente, M., Deslandes, L. & Peeters, N. (2020) The reasons. Annual Review of Phytopathology, 56, 405–­426. large, diverse, and robust arsenal of Ralstonia solanacearum type III Chen, D., Zhou, D., Wang, Y., Wang, B., He, Q., Song, B. et al. (2021) effectors and their in planta functions. Molecular Plant Pathology, Ralstonia solanacearum type III effector RipV2 encoding a novel 21, 1377–­1388. E3 ubiquitin ligase (NEL) is requried for full virulence by suppress- Le Roux, C., Huet, G., Jauneau, A., Camborde, L., Trémousaygue, D., ing plant PAMP-­triggered immunity. Biochemical and Biophysical Kraut, A. et al. (2015) A receptor pair with an integrated decoy con- Research Communication, 550, 120–­126. verts pathogen disabling of transcription factors to immunity. Cell, Cunnac, S., Chakravarthy, S., Kvitko, B.H., Russell, A.B., Martin, G.B. & 161, 1074–­1088. Collmer, A. (2011) Genetic disassembly and combinatorial reas- Lu, H., Lema, A.S., Planas-­Marques, M., Alonso-­Diaz, A., Valls, M. & sembly identify a minimal functional repertoire of type III effectors Coll, N.S. (2018) Type III secretion-­dependent and -­independent 202 | NIU et al. phenotypes caused by Ralstonia solanacearum in Arabidopsis roots. Singer, A.U., Schulze, S., Skarina, T., Xu, X., Cui, H., Eschen-­Lippold, L. Molecular Plant-­Microbe Interactions, 31, 175–­184. et al. (2013) A pathogen type III effector with a novel E3 ubiquitin Mansfield, J., Genin, S., Magori, S., Citovsky, V., Sriariyanum, M., Ronald, ligase architecture. PLoS Pathogens, 9, e1003121. P. et al. (2012) Top 10 plant pathogenic bacteria in molecular plant Sole, M., Popa, C., Mith, O., Sohn, K.H., Jones, J.D., Deslandes, L. et al. pathology. Molecular Plant Pathology, 13, 614–­629. (2012) The awr gene family encodes a novel class of Ralstonia sola- Mukaihara, T., Hatanaka, T., Nakano, M. & Oda, K. (2016) Ralstonia sola- nacearum type III effectors displaying virulence and avirulence activi- nacearum type III effector RipAY is a glutathione-­degrading enzyme ties. Molecular Plant-­Microbe Interactions, 25, 941–­953. that is activated by plant cytosolic thioredoxins and suppresses Spallek, T., Robatzek, S. & Gohre, V. (2009) How microbes utilize host plant immunity. mBio, 7, e00359–­e 416. ubiquitination. Cellular Microbiology, 11, 1425–­1434. Nakano, M. & Mukaihara, T. (2019a) Comprehensive identification of Sun, T., Wu, W., Wu, H., Rou, W., Zhou, Y., Zhuo, T. et al. (2020) Ralstonia PTI suppressors in type III effector repertoire reveals that Ralstonia solanacearum elicitor RipX induces defense reaction by suppressing solanacearum activates jasmonate signaling at two different steps. the mitochondrial atpA gene in host plant. International Journal of International Journal of Molecular Sciences, 20, 5992. Molecular Sciences, 21, 2000. Nakano, M. & Mukaihara, T. (2019b) The type III effector RipB from Sun, Y., Li, P., Deng, M., Shen, D., Dai, G., Yao, N. et al. (2017) The Ralstonia solanacearum RS1000 acts as a major avirulence factor in Ralstonia solanacearum effector RipAK suppresses plant hypersen- Nicotiana benthamiana and other Nicotiana species. Molecular Plant sitive response by inhibiting the activity of host catalases. Cellular Pathology, 20, 1237–­1251. Microbiology, 19, e12736. Nakano, M., Oda, K. & Mukaihara, T. (2017) Ralstonia solanacearum Szczesny, R., Buttner, D., Escolar, L., Schulze, S., Seiferth, A. & Bonas, U. novel E3 ubiquitin ligase (NEL) effectors RipAW and RipAR sup- (2010) Suppression of the AvrBs1-­specific hypersensitive response press pattern-­t riggered immunity in plants. Microbiology, 163, by the YopJ effector homolog AvrBsT from Xanthomonas depends 992–­1002. on a SNF1-­related kinase. New Phytologist, 187, 1058–­1074. Planas-­Marquès, M., Kressin, J.P., Kashyap, A., Panthee, D.R., Louws, F.J., Valls, M., Genin, S. & Boucher, C. (2006) Integrated regulation of the type Coll, N.S. et al. (2020) Four bottlenecks restrict colonization and in- III secretion system and other virulence determinants in Ralstonia vasion by the pathogen Ralstonia solanacearum in resistant tomato. solanacearum. PLoS Pathogens, 2, e82. Journal of Experimental Botany, 71, 2157–­2171. Wan, W.L., Frohlich, K., Pruitt, R.N., Nurnberger, T. & Zhang, L. (2019) Popa, C., Li, L., Gil, S., Tatjer, L., Hashii, K., Tabuchi, M. et al. (2016) The Plant cell surface immune receptor complex signaling. Current effector AWR5 from the plant pathogen Ralstonia solanacearum Opinion in Plant Biology, 50, 18–­28. is an inhibitor of the TOR signalling pathway. Scientific Reports, Wang, H., Hu, J., Lu, Y., Zhang, M., Qin, N., Zhang, R. et al. (2019) A quick 6, 27058. and efficient hydroponic potato infection method for evaluat- Poueymiro, M., Cazale, A.C., Francois, J.M., Parrou, J.L., Peeters, N. ing potato resistance and Ralstonia solanacearum virulence. Plant & Genin, S. (2014) A Ralstonia solanacearum type III effector di- Methods, 15, 145. rects the production of the plant signal metabolite trehalose-­6-­ Wang, Y., Zhao, A., Morcillo, R.J.L., Yu, G., Xue, H., Rufian, J.S. et al. phosphate. mBio, 5, e02065-­14. (2021) A bacterial effector protein uncovers a plant metabolic path- Poueymiro, M., Cunnac, S., Barberis, P., Deslandes, L., Peeters, N., Cazale-­ way involved in tolerance to bacterial wilt disease. Molecular Plant, Noel, A.C. et al. (2009) Two type III secretion system effectors from 14, 1281–­1296. Ralstonia solanacearum GMI1000 determine host-­range specificity Wu, D., von Roepenack-­L ahaye, E., Buntru, M., de Lange, O., Schandry, on tobacco. Molecular Plant-­Microbe Interactions, 22, 538–­550. N., Pérez-­Q uintero, A.L. et al. (2019) A plant pathogen type III ef- Quezada, C.M., Hicks, S.W., Galan, J.E. & Stebbins, C.E. (2009) A family fector protein subverts translational regulation to boost host poly- of Salmonella virulence factors functions as a distinct class of auto- amine levels. Cell Host & Microbe, 26, 638–­6 49 e635. regulated E3 ubiquitin ligases. Proceedings of the National Academy Xian, L., Yu, G., Wei, Y., Rufian, J.S., Li, Y., Zhuang, H. et al. (2020) A bacte- of Sciences USA, 106, 4864–­4869. rial effector protein hijacks plant metabolism to support pathogen Sabbagh, C.R.R., Carrere, S., Lonjon, F., Vailleau, F., Macho, A.P., Genin, nutrition. Cell Host & Microbe, 28, 548–­557 e547. S. et al. (2019) Pangenomic type III effector database of the plant Xin, D.-­W., Liao, S., Xie, Z.-­P., Hann, D.R., Steinle, L., Boller, T. et al. (2012) pathogenic Ralstonia spp. Peer J, 7, e7346. Functional analysis of NopM, a novel E3 ubiquitin ligase (NEL) do- Sang, Y., Wang, Y., Ni, H., Cazale, A.C., She, Y.M., Peeters, N. et al. (2018) main effector of Rhizobium sp. strain NGR234. PLoS Pathogens, 8, The Ralstonia solanacearum type III effector RipAY targets plant e1002707. redox regulators to suppress immune responses. Molecular Plant Yu, G., Xian, L., Xue, H., Yu, W., Rufian, J.S., Sang, Y. et al. (2020) A Pathology, 19, 129–­142. bacterial effector protein prevents MAPK-­m ediated phosphor- Sang, Y., Yu, W., Zhuang, H., Wei, Y., Derevnina, L., Yu, G. et al. (2020) ylation of SGT1 to suppress plant immunity. PLoS Pathogens, 16, Intra-­strain elicitation and suppression of plant immunity by e1008933. Ralstonia solanacearum type-­III effectors in Nicotiana benthamiana. Zheng, N. & Shabek, N. (2017) Ubiquitin ligases: structure, function, and Plant Communications, 1, 100025. regulation. Annual Review of Biochemistry, 86, 129–­157. Sarris, P., Duxbury, Z., Huh, S., Ma, Y., Segonzac, C., Sklenar, J. et al. Zheng, X., Li, X., Wang, B., Cheng, D., Li, Y., Li, W. et al. (2019) A sys- (2015) A plant immune receptor detects pathogen effectors that tematic screen of conserved Ralstonia solanacearum effectors target WRKY transcription factors. Cell, 161, 1089–­1100. reveals the role of RipAB, a nuclear-­localized effector that sup- Senthil-­Kumar, M., Hema, R., Anand, A., Kang, L., Udayakumar, M. & presses immune responses in potato. Molecular Plant Pathology, 20, Mysore, K.S. (2007) A systematic study to determine the extent of 547–­561. gene silencing in Nicotiana benthamiana and other Solanaceae spe- Zhou, J.M. & Zhang, Y. (2020) Plant immunity: danger perception and cies when heterologous gene sequences are used for virus-­induced signaling. Cell, 181, 971–­989. gene silencing. New Phytologist, 176, 782–­791. Zhu, Y., Li, H., Hu, L., Wang, J., Zhou, Y., Pang, Z. et al. (2008) Structure Singer, A.U., Rohde, J.R., Lam, R., Skarina, T., Kagan, O., DiLeo, R. et al. of a Shigella effector reveals a new class of ubiquitin ligases. Nature (2008) Structure of the Shigella T3SS effector IpaH defines a new Structural & Molecular Biology, 15, 1302–­1308. class of E3 ubiquitin ligases. Nature Structural & Molecular Biology, Zhuo, T., Wang, X., Chen, Z., Cui, H., Zeng, Y., Chen, Y. et al. (2020) The 15, 1293–­1301. Ralstonia solanacearum effector RipI induces a defence reaction by NIU et al. | 203 interacting with the bHLH93 transcription factor in Nicotiana ben- thamiana. Molecular Plant Pathology, 21, 999–­1004. How to cite this article: Niu, Y., Fu, S., Chen, G., Wang, H., Wang, Y., Hu, J., et al (2022) Different epitopes of Ralstonia S U P P O R T I N G I N FO R M AT I O N solanacearum effector RipAW are recognized by two Nicotiana Additional supporting information may be found in the online ver- species and trigger immune responses. Molecular Plant sion of the article at the publisher’s website. Pathology, 23, 188–­203. https://doi.org/10.1111/mpp.13153