Abstract
Effector proteins delivered inside plant cells are powerful weapons for bacterial pathogens, but this exposes the pathogen to potential recognition by the plant immune system. Therefore, effector acquisition must be balanced for a successful infection. Ralstonia solanacearum is an aggressive pathogen with a large repertoire of secreted effectors. One of these effectors, RipE1, is conserved in most R. solanacearum strains sequenced to date. In this work, we found that RipE1 triggers immunity in N. benthamiana, which requires the immune regulator SGT1, but not EDS1 or NRCs. Interestingly, RipE1-triggered immunity induces the accumulation of salicylic acid (SA) and the overexpression of several genes encoding phenylalanine-ammonia lyases (PALs), suggesting that the unconventional PAL-mediated pathway is responsible for the observed SA biosynthesis. Surprisingly, RipE1 recognition also induces the expression of jasmonic acid (JA)-responsive genes and JA biosynthesis, suggesting that both SA and JA may act cooperatively in response to RipE1. Finally, we found that RipE1 expression leads to the accumulation of glutathione in plant cells, which precedes the activation of immune responses. R. solanacearum encodes another effector, RipAY, which is known to inhibit immune responses by degrading cellular glutathione. Accordingly, we show that RipAY inhibits RipE1-triggered immune responses. This work shows a strategy employed by R. solanacearum to counteract the perception of its effector proteins by the plant immune system.
Introduction
Ralstonia solanacearum is considered one of the most destructive plant pathogens, and is able to cause disease in more than 250 plant species (Jiang et al., 2017; Mansfield et al., 2012). As a soil-borne bacterial pathogen, R. solanacearum enters plants through the roots, reaches the vascular system, and spreads through xylem vessels, colonizing the plant systemically (Mansfield et al., 2012). This is followed by massive bacterial replication and the disruption of the plant vascular system, leading to eventual plant wilting (Digonnet et al., 2012; Turner et al., 2009).
Most bacterial pathogens deliver proteins inside plant cells via a type-III secretion system (T3SS); such proteins are thus called type-III effectors (T3Es) (Galan et al, 2014). T3Es have been reported to mediate the suppression of basal defenses and the manipulation of plant physiological functions to support bacterial proliferation (Macho et al, 2015; Macho, 2016). Resistant plants have evolved intracellular receptors defined by the presence of nucleotide-binding sites (NBS) and leucine-rich repeat domains (LRRs), thus termed NLRs (Cui et al, 2015). Specific NLRs can detect the activities of specific T3Es, leading to the activation of immune responses, which effectively prevent pathogen proliferation (Chiang & Coaker, 2015). The outcome of these responses is named effector-triggered immunity (ETI), and, in certain cases, may cause a hypersensitive response (HR) that involves the collapse of plant cells. Hormone-mediated signaling plays an essential role in plant immunity. Salicylic acid (SA) is considered the most important hormone in plant immunity against biotrophic pathogens (Vlot et al., 2009; Burger & Chory, 2019); Jasmonic acid (JA), on the other hand, is considered the main mediator of immune responses against necrotrophic pathogens (Burger & Chory, 2019). In most cases, both hormones are considered as antagonistic, balancing the effects of each other (Burger & Chory, 2019).
In an evolutionary response to ETI, successful pathogens have acquired T3E activities to suppress this phenomenon (Jones & Dangl, 2006), although reports characterizing T3E suppression of ETI remain scarce, particularly among T3Es within the same strain. While the development of additional T3E activities is a powerful virulence strategy, it also exposes the pathogen to further events of effector recognition. Therefore, the benefits and penalties of T3E secretion need to be finely and dynamically balanced in specific hosts, to ensure the appropriate manipulation of plant functions while evading or suppressing ETI. This balance may be particularly important for R. solanacearum, which secretes a larger number of T3Es in comparison to other bacterial plant pathogens (e.g. the reference GMI1000 strain is able to secrete more than 70 T3Es) (Peeters et al, 2013).
Plants have evolved to recognize immune elicitors from R. solanacearum (Wei et al, 2018; Jayaraman et al., 2016). In terms of mechanism of T3E recognition, the most studied case in R. solanacearum is RipP2 (also known as PopP2), which is perceived in Arabidopsis by the RRS1-RPS4 NLR pair (Gassmann et al, 1999; Deslandes et al, 2002; Tasset et al, 2010; Williams et al, 2014; Le Roux et al, 2015; Sarris et al, 2015). Additionally, several R. solanacearum T3Es were shown to induce cell death in different plant species (Peeters et al, 2013; Clarke et al, 2015), although, in most cases, it is unclear whether these are due to toxic effects caused by effector overexpression or a host immune response. Some R. solanacearum T3Es have also been shown to cause a restriction of host range; such is the case for RipAA and RipP1 (also known as AvrA and PopP1, respectively), which are perceived and restrict host range in Nicotiana species (Poueymiro et al, 2009). RipP1 also triggers resistance in petunia (Lavie et al, 2002). Similarly, RipB-triggered immunity has been reported as the major cause for avirulence of R. solanacearum RS1000 in Nicotiana species (Nakano & Mukaihara, 2019), RipAX2 (also known as Rip36) have been shown to induce resistance in eggplant and its wild relative Solanum torvum (Nahar et al, 2014; Morel et al, 2018), and several T3Es from the AWR family (also known as RipA) restrict bacterial growth in Arabidopsis (Sole et al, 2012). Although the utilization of these recognition systems to generate disease-resistant crops is tantalizing, it is imperative to understand the mechanisms underlying the activation of plant immunity and their potential suppression by other T3Es within R. solanacearum.
The ripE1 gene encodes a protein secreted by the type-III secretion system in the R. solanacearum GMI1000 strain (phylotype I) (Mukaihara et al, 2010), and is conserved across R. solanacearum strains from different phylotypes (Peeters et al, 2013). Based on sequence analysis, RipE1 is homologous to other T3Es in Pseudomonas syringae (HopX) and Xanthomonas spp (XopE) (Figure S1; Peeters et al, 2013), belonging to the HopX/AvrPphB T3E family (Nimchuk et al, 2007). This family is characterized by the presence of a putative catalytic triad consisting of specific cysteine, histidine, and aspartic acid residues, which are conserved in RipE1 (Nimchuk et al, 2007; Figure S1), and is similar to several enzyme families from the transglutaminase protein superfamily, such as peptide N-glycanases, phytochelatin synthases, and cysteine proteases (Makarova et al, 1999). AvrPphB, from P. syringae pv. phaseolicola, the original member of the HopX/AvrPphB family, was identified based on its ability to activate immunity in certain bean cultivars (Mansfield et al, 1994). Divergent members from this family in other strains also trigger immunity, and this requires the putative catalytic cysteine (Nimchuk et al, 2007). Previous sequence analysis of T3Es from the HopX family also identified a conserved domain (domain A) required for HopX induction of immunity in bean and Arabidopsis, which as hypothesized to represent a host-target interaction domain or a novel nucleotide/cofactor binding domain (Nimchuk et al, 2007).
In this work, we studied the impact of RipE1 in plant cells, and found that RipE1 is recognized by the plant immune system in both N. benthamiana and Arabidopsis, leading to the activation of immune responses. We further investigate the immune components and signaling pathways associated to this effector recognition. Finally, we found that another effector in R. solanacearum GMI1000 is able to inhibit RipE1-triggered immune responses in N. benthamiana, explaining the fact that RipE1 does not seem to be an avirulence determinant in this plant species.
Results
RipE1 triggers cell death upon transient expression in Nicotiana benthamiana
In order to understand the impact of RipE1 in plant cells, we first used an Agrobacterium tumefaciens (hereafter, Agrobacterium)-mediated expression system in Nicotiana benthamiana leaves to transiently express RipE1 that is fused to a carboxyl-terminal green fluorescent protein (GFP) tag (RipE1-GFP). Two days after Agrobacterium infiltration, we noticed the collapse of infiltrated tissues expressing RipE1-GFP, but not a GFP control (Figure 1a). This tissue collapse correlated with a release of ions from plant cells (Figure 1b), indicative of cell death. Mutation of the catalytic cysteine to an alanine residue has been shown to disrupt the catalytic activity of enzymes with a catalytic triad similar to that conserved in RipE1 (Gimenez-Ibanez et al, 2014; Figure 1c). To determine if the putative catalytic activity is required for RipE1 induction of cell death, we generated an equivalent mutant in RipE1 (C172A; Figure 1c). We also generated an independent mutant with a deletion on the eight amino acids that constitute the conserved domain A (Nimchuk et al, 2007; Figure 1c). These mutations did not affect the accumulation of RipE1 (Figure 1d), but abolished the induction of tissue collapse and the ion leakage caused by RipE1 expression (Figure 1d and 1e), indicating that RipE1 requires both the catalytic cysteine and the conserved domain A for the induction of cell death in plants.
Interestingly, RipE1 was also identified in a systematic screen performed in our laboratory to identify R. solanacearum T3Es that suppress immune responses triggered by bacterial elicitors. In this screen we found that RipE1 expression suppresses the burst of reactive oxygen species (ROS) and the activation of mitogen-activated protein kinases (MAPKs) triggered upon treatment with the bacterial flagellin epitope flg22, which acts as an immune elicitor (Figure S2). RipE1 requires both the catalytic cysteine and the conserved domain A for this activity (Figure S2). However, we considered the possibility that these responses are abolished by the death of plant cells rather than an active immune suppression. Time-course experiments showed that the suppression of flg22-triggered ROS correlated with the appearance of cell death (Figure S2), making it difficult to uncouple these observations.
RipE1 activates salicylic acid-dependent immunity in N. benthamiana
The induction of cell death by pathogen effectors may reflect toxicity in plant cells or the activation of immune responses that lead to HR. Salicylic Acid (SA) plays a major role in the activation of immune responses after the perception of different types of invasion patterns (Vlot et al., 2009). To determine whether RipE1 activates immune responses, we first measured the expression of the N. benthamiana ortholog of the Arabidopsis gene PATHOGENESIS-RELATED-1 (PR1), which is a hallmark of SA-dependent immune responses (Vlot et al., 2009, Ward et al., 1991). Expression of RipE1-GFP (but not the C172A catalytic mutant) significantly enhanced the accumulation of NbPR1 transcripts (Figure 2a). The bacterial salicylate hydroxylase NahG converts SA to catechol and leads to the suppression of SA-dependent responses (Delaney et al., 1994). The expression of NahG-GFP in N. benthamiana slightly enhanced the accumulation of RipE1 fused to a carboxyl-terminal N-luciferase tag (Nluc) (Figure S3), consistent with the reported role of SA in hindering Agrobacterium-mediated transformation (Rosas-Diaz et al, 2016); despite this, NahG partially suppressed RipE1-triggered cell death, ion leakage, and NbPR1 expression (Figure 2b, c and d). Altogether, these data suggest that RipE1 induces SA-dependent immune responses in plant cells, which cause the development of HR cell death.
RipE1 enhances the expression of PAL genes and the biosynthesis of salicylic acid and jasmonic acid
The expression of RipE1 led to a dramatic increase in SA accumulation in N. benthamiana (Figure 3a), consistent with the observed overexpression of NbPR1 (Figure 2a). This reinforces the idea that RipE1 is perceived by the plant immune system and this leads to the activation of SA biosynthesis and SA-dependent immune responses. In Arabidopsis, the chloroplastic pathway mediated by isochorismate synthethase 1 (ICS1) plays a predominant role in the pathogen-induced SA biosynthesis (Wildermuth et al, Nature, 2001; Garcion et al, Plant Physiology, 2008). However, gene expression analysis showed that the expression of the N. benthamiana ortholog of the Arabidopsis ICS1, NbICS1, was significantly reduced upon RipE1 expression (Figure 3b), despite the simultaneous high NbPR1 transcript accumulation (Figure 2a). SA can also be synthesized from phenylalanine in a pathway mediated by phenylalanine ammonia lyases (PALs). In contrast with the expression of NbICS1, several genes encoding NbPALs were up-regulated upon expression of RipE1, but not the catalytic mutant version (Figure 3c-e), suggesting that this pathway may mediate the enhancement in SA biosynthesis upon perception of RipE1 activity. SA and Jasmonic Acid (JA) are considered antagonistic hormones in plant immune responses. Surprisingly, instead of a reduction of JA-associated gene expression, we found a slight increase in the accumulation of NbLOX2 transcripts in early time points upon expression of catalytically active RipE1 (Figure 3f). In Arabidopsis, LOX2 contributes to the biosynthesis of JA (Bell et al, 1995). Accordingly, we detected an increase in JA contents upon RipE1 expression (Figure S4), indicating that RipE1 perception does not inhibit JA signalling, but rather leads to an enhancement on JA biosynthesis and associated gene expression.
RipE1-triggered immunity requires SGT1, but not EDS1 or NRC proteins
The suppressor of the G2 allele of skp1 (SGT1) plays an essential role in ETI, and is required for the induction of disease resistance mediated by most NLRs (Azevedo et al., 2002; Kadota et al., 2010). Virus-induced gene silencing (VIGS) of NbSGT1 abolished RipE1-triggered cell death, ion leakage, and NbPR1 expression (Figure 4a-d), indicating that RipE1-triggered immunity requires SGT1. While most NLRs require SGT1 to function, a specific group of NLRs containing an N-terminal Toll-like interleukin-1 receptor (TIR) domain also require EDS1 (Wiermer et al, 2005; Schultink et al, 2017). N. benthamiana plants carrying a stable knockout mutation in EDS1 (Schultink et al, 2017) displayed clear RipE1-triggered cell death (Figure 4e), suggesting that RipE1-triggered immunity is not mediated by a TIR-NLR. Other NLRs contain a C-terminal coiled coil (CC) domain, and a specific subset of CC-NLRs require a network of helper NLRs termed NRC proteins (Wu et al, 2016). Interestingly, silencing of NRC proteins did not impact RipE1-triggered cell death (Figure S5), suggesting that RipE1-triggered immunity is not mediated by an NLR within the NRC network.
RipE1 activates immunity in Arabidopsis
Arabidopsis transgenic plants expressing RipE1-GFP from a 35S inducible promoter died after germination (data not shown). Therefore, we generated Arabidopsis transgenic plants expressing RipE1-GFP and RipE1C172A-GFP from an estradiol (EST)-inducible promoter. Five week-old plants expressing RipE1-GFP, but not RipE1C172A-GFP, showed reduced growth in soil upon EST treatment for 14 days (Figure 5a). To determine whether RipE1-triggered growth reduction in Arabidopsis correlates with the activation of immunity, we first monitored the expression of defence-related genes. Similar to the result observed upon expression in N. benthamiana, expression of RipE1 in Arabidopsis triggered the overexpression of AtPR1 (Figure 5b). However, in Arabidopsis, the enhanced PR1 expression correlated with an overexpression of AtICS1, but not AtPAL1, upon RipE1 expression (Figure 5b). As observed in N. benthamiana, RipE1 expression led to the overexpression of the JA marker genes AtVSP2 and AtPDF1.2 (REF; Figure 5b). This indicates that, as observed in N. benthamiana, RipE1 activates SA- and JA-dependent signalling in Arabidopsis. To determine whether the activation of defence-related genes leads to an efficient immune response in Arabidopsis, we inoculated RipE1-expressing plants by soil-drenching with R. solanacearum. As shown in the figure 5c, RipE1-expressing plants displayed weaker and delayed disease symptoms upon R. solanacearum inoculation, reflecting an enhanced disease resistance upon RipE1 expression.
RipE1-triggered immune responses are suppressed by RipAY
RipE1 expression activates immunity in Arabidopsis and N. benthamiana, although both plant species are susceptible hosts for R. solanacearum GMI1000 (or a derivative strain carrying mutations in popP1 and avrA, in the case of N. benthamiana; Poueymiro et al, 2009), which carries RipE1. Therefore, we reasoned that other T3E(s) in GMI1000 may be able to suppress RipE1-triggered immunity in the context of infection. We recently identified a R. solanacearum T3E, RipAY, which is able to suppress SA-dependent immune responses through the degradation of glutathione (Sang et al, 2016; Mukaihara et al, 2016); however, the ability of RipAY to suppress immunity triggered by other R. solanacearum T3Es remained unknown. Interestingly, the expression of RipE1 in N. benthamiana leads to an increase in glutathione accumulation in plant tissues, which precedes the onset of immune responses (Figure 6a). Considering that both RipAY and RipE1 are present in GMI1000, we sought to determine if RipAY has the ability to suppress RipE1-triggered immunity. Indeed, expression of RipAY in N. benthamiana did not affect the accumulation of RipE1 (Figure S6), but abolished the tissue collapse and ion leakage caused by RipE1 expression (Figure 6b, c, and d). Moreover, RipAY was able to suppress the overexpression of NbPR1 triggered by RipE1 (Figure 6e), indicating that RipAY suppresses RipE1-triggered immune responses. Interestingly, however, a RipAY point mutant unable to degrade glutathione (RipAYE216Q; Sang et al, 2016) did not suppress RipE1-triggered responses (Figure 6b-e), suggesting that RipAY suppresses RipE1-triggered immunity through the degradation of cellular glutathione.
Discussion
Expression of T3Es in plant cells may either induce cell death because of cell toxicity or lead to the activation of an immunity-associated HR. Over-expression of RipE1 in N. benthamiana leads to cell death that: (i) is dependent on the immune regulator SGT1; (ii) activates SA accumulation and PR1 expression; and (iii) is suppressed by the NahG and other R. solanacearum effectors, indicating that RipE1-mediated cell death is due to the activation of immunity in the host. Several T3Es within the HopX/AvrPphB family are predicted enzymes that are associated with activation of host immunity, although the association of the predicted catalytic activity with the activation of immunity seems to be differ among them. While the ability of AvrPphB and several other family members to trigger immunity requires the putative catalytic cysteine (Mansfield et al, 1994; Nimchuk et al, 2007), other members with the predicted catalytic activity, such as HopX from P. syringae pv tabaci or P. syringae pv phaseolicola race 6, do not trigger immunity in the same hosts (Stevens et al, 1998; Nimchuk et al, 2007). In the case of RipE1, the putative catalytic cysteine is required for the induction of immunity, which suggests that RipE1 is an active enzyme, and that this catalytic activity leads to perception by the host immune system. Moreover, the conserved domain A (Nimchuk et al, 2007) is also required for the activation of immunity by RipE1. In addition, we found that RipE1 is able to suppress elicitor-triggered immune responses in N. benthamiana. However, since this activity correlates with the induction of cell death, it is difficult to uncouple both observations, and further studies on the virulence activity of RipE1 will require the utilization of a host plant that is unable to recognize it.
The fact that RipE1 is recognized, and activate immune responses, in both N. benthamiana and Arabidopsis suggests at least two scenarios: it is possible that the NLR responsible for this recognition is conserved in both species; on the other hand, it is also possible that both species have independently develop NLRs that recognize RipE1. Although we did not identify the NLR involved, we determined that, at least in N. benthamiana, RipE1 recognition does not rely on EDS1 or the NRC network, pointing to a CC-NRC-independent NLR. Interestingly, although RipE1 perception leads to the accumulation of SA in both plant species, the associated gene expression patterns seem to differ. The ICS pathway plays a predominant role in the pathogen-induced SA biosynthesis in Arabidopsis (Wildermuth et al, Nature, 2001; Garcion et al, Plant Physiology, 2008). In agreement with this, the RipE1-triggered overexpression of AtPR1 in Arabidopsis correlates with an enhanced expression of AtICS1, but not AtPAL1. However, it seems that the RipE1-induced increase in SA content in N. benthamiana correlates with a reduction of NbICS1 gene expression, and an increase in the expression of several NbPAL genes. Considering that ICS1 is normally regulated at the transcriptional level upon pathogen perception (Wildemurth et al, 2001), our results suggest that the PAL pathway is more relevant than the ICS pathway for the induction of RipE1-triggered immunity in N. benthamiana, indicating that both pathways are differentially required for distinct immune responses in different plant species. Similarly, both the ICS and PAL pathways have been reported to be required for pathogen-induced SA biosynthesis in soybean (Shine et al, New Phytol, 2016). The reduction in ICS1 expression in N. benthamiana may reflect a compensatory effect between the ICS and PAL pathway. In addition to different gene expression patterns, the physiological output in both plant species may be different. Although RipE1 expression caused an inhibition of Arabidopsis growth, we did not observe any signs of cell death (data not shown), which contrasts with our observation in N. benthamiana. However, this may be caused by differences in the expression system used in both plants (Agrobacterium-mediated transient expression in N. benthamiana vs EST-induced expression in Arabidopsis stable transgenic plants).
Another surprising aspect of RipE1-triggered immunity is the fact that it leads to the simultaneous accumulation of SA and JA, and to a strong and moderate SA- and JA-triggered gene expression, respectively, in both N. benthamiana and Arabidopsis. This suggests that, in the case of RipE1-triggered immunity, SA and JA may play a cooperative role, possibly reflecting the complexity of the R. solanacearum infection process compared to other pathogens.
If RipE1 triggers immunity in N. benthamiana, why is it that a GMI1000 strain without PopP1 and AvrA (but having RipE1) can cause a successful infection in N. benthamiana without triggering immunity (Poueymiro et al, 2009)? Here, we found that other effectors within GMI1000, such as RipAY, are able to inhibit RipE1-triggered immunity. Since RipE1 perception correlates with an enhancement of cellular glutathione, and RipAY requires its gamma-glutamyl cyclotransferase activity to inhibit RipE1-triggered HR, the degradation of glutathione or other gamma-glutamyl compounds (Sang et al, 2016; Mukaihara et al, 2016; Fujiwara et al, 2016) is the most likely mechanism for this inhibition. Besides RipAY, other T3Es within GMI1000 likely contribute to the suppression of RipE1-triggered HR by targeting other immune functions (Yu et al, bioRxiv, 2019; Wang & Macho, unpublished data). This reflects bacterial adaptation: RipE1 could be important for virulence, but also triggers immunity. In this context, instead of losing RipE1, R. solanacearum has developed other effectors to suppress the induction of immunity, while keeping RipE1 virulence activity. Similarly, although transient expression of HopX from P. syringae pv tomato (Pto) triggers HR in specific Arabidopsis accessions, it does not trigger an HR in the context of Pto infection (Nimchuk et al, 2007). It is possible that, as in the case of RipE1, the immune responses triggered by HopX are masked during Pto infection (as suggested in Nimchuk et al, 2007), likely due to the suppression by other effectors within the same strain.
Materials and Methods
Plant materials and growth conditions
N. benthamiana plants were grown on soil at one plant per pot in an environmentally controlled growth room at 25 °C under a 16-h light/8-h dark photoperiod with a light-intensity of 130 mE m-2s-1. A. thaliana plants were grown under the same conditions as N. benthamiana for seeds collection. For bacteria virulence and ROS burst assays, A. thaliana plants were grown in a growth chamber controlled at 22°C with a 10 h photoperiod and a light-intensity of 100-150 mE m-2s-1. After R. solanacearum inoculation, Arabidopsis plants were transferred to a growth chamber 27°C with 75% humid under a 12-h light/12-h dark photoperiod.
Chemicals
The flg22 peptide (TRLSSGLKINSAKDDAAGLQIA) was purchased from Abclonal, USA. All other chemicals were purchased from Sigma-Aldrich unless otherwise
Plasmids, bacterial strains and cultivation conditions
R. solanacearum GMI1000 was grown on solid BG medium plates or cultivated over-night in liquid BG medium at 28°C (Morel et al., 2018). The ripE1 gene from R. solanacearum GMI1000 cloned in pDonor207 (donated by Nemo Peeters and Anne-Claire Cazale) was subcloned into pGWB505 by LR reaction (ThermoFisher, USA) to generate a fusion protein with eGFP tag at the C-terminal (Nakagawa et al., 2007). RipE1 and ripE1 mutants were inserted between BamH1 and Xho1 restriction sites on sXVE:GFPc:Bar estradiol inducible vector using enzyme digestion (Schlücking et al., 2013). These generated binary vectors were transformed into Agrobacterium tumefaciens (Agrobacterium) GV3101 for transient or stable gene expression in N. benthamiana and A. thaliana plants. Agrobacterium carrying pGWB505 vectors were grown at 29°C and 220 rpm in LB medium supplemented with rifampicin 50 mg/l, gentamycin 25 mg/l and spectinomycin 50 mg/l, while those carrying estradiol inducible vectors were grown in rifampicin 50 mg/l, gentamycin 25 mg/l and kanamycin 50 mg/l.
Site-directed mutagenesis
RipE1C172A and RipE1 ΔAD mutant variants were generated using the QuickChange Lightning Site-Directed Mutagenesis Kit (Life technologies, USA) following the manufacturer’s instructions. RipE1/pDONR207 plasmid was used as template. Primers used for the mutagenesis are listed in Table S1.
Agrobacterium-mediated gene expression in A. thaliana and N. benthamiana
Stable transgenic Arabidopsis plants with RipE1 and RipE1 mutated variants driven by estradiol inducible promoter were obtained using the floral dip method (Zhang et. al, 2006). Homozygous T3 lines were used for all the experiments. Agrobacterium-mediated transient expression in N. benthamiana was performed as described (Li, 2011). Agrobacterium carrying the resultant plasmids were suspended in infiltration buffer to a final OD600 of 0.1∼0.5 and infiltrated into the abaxial side of the leaves using the 1mL needless syringe. Leaf samples were taken at 1-3 dpi (days post infiltration) for analysis based on experimental requirements.
Protein extraction and western blots
Plant tissues were collected into 2 ml tubes with metal beads and frozen in liquid nitrogen. After grinding with a tissue lyser (Qiagen, Germany) for 1 min at 30 rpm/s, proteins were extracted using protein extraction buffer (100 mM Tris-HCl pH 8, 150 mM NaCl, 10% glycerol, 5 mM Ethylene diamine tetra acetic acid (EDTA), 2 mM Dithiothreitol (DTT), 1x Plant Protease Inhibitor cocktail, 1% NP-40, 2 mM Phenylmethylsulfonyl fluoride (PMSF), 10 mM Na2MoO4, 10 mM NaF, 2 mM Na3VO4) and incubating for 5 min. After centrifugation, the supernatants were mixed with SDS loading buffer, incubated at 70 °C for 10 min, and resolved using SDS-PAGE. Proteins were transferred to a PVDF membrane and monitored by western blot using anti-GFP (Abicode, M0802-3a) and anti-luciferase (Sigma, L0159) antibodies.
Measurement of ROS generation and MAPK activation
PAMP-triggered ROS burst and MAPK activation in plant leaves were measured as described previously (Sang et al., 2017; Segonzac et al., 2011). ROS was elicited with 50 nM flg22. MAPK activation assays were performed using 4 to 5-week-old N. benthamiana. Two days after Agrobacterium infiltration at OD600 of 0.1, the intact leaves were elicited for 15 min after vacuum infiltration of 100nM pto-flg22. Leaf discs were taken to monitor MAPK activation by western blot with Phospho-p44/42 MAPK (Erk1/2; Thr-202/Tyr-204) antibodies.
Cell death measurement
Cell death in plant leaves was quantified as previously described (Yu et al, bioRxiv, 2019) by measuring the electrolyte leakage using a conductivity meter (ThermoFisher, USA) or observing the autofluorescence using the BioRad Gel Imager (Bio-Rad, USA). Briefly, one day after Agrobacterium infiltration in N. benthamiana, one 13 mm leaf disk was immersed in 4 mL of distilled water for 1 h with gentle shaking and then transferred to a 6-well culture plate containing 4 mL distilled water in each well. The ion conductivity was then measured at different time intervals. Autofluorescence in intact N. benthamiana leaves was measured at 2.5 dpi.
RNA isolation and qRT-PCR
Total RNA was extracted using the E.Z.N.A. Plant RNA kit with DNA digestion on column (Biotek, China) according to the manufacturer’s instructions. RNA samples were quantified with a Nanodrop spectrophotometer (ThermoFisher, USA). First strand cDNA was synthesized using the iScriptTM cDNA synthesis kit (Bio-Rad). qRT-PCR was performed using the iTaqTM Universal SYBR Green Supermix (Bio-Rad) and CFX96 Real-time system (Bio-Rad) and the qPCR data was analyzed as described by (Livak & Schmittgen, 2001). Primers for qPCR of SA-JA related genes in N. benthamiana were used as described by Nakano and Mukaihara (2018). Primer sequences are listed in Table S1.
Measurements of SA and JA content in plant leaves
SA and JA content were quantified using the method described by Forcat and collaborators (2008) with the following modifications. Leaves (50 mg FW) were collected 42 hours after Agrobacterium infiltration and frozen in liquid nitrogen before grounding into fine powder with the Qiagen tissue lyser. SA and JA were extracted at 10°C for 1 h using 70% methanol extraction solvent spiked with d4-SA as internal standards. Supernatant was taken after centrifuge at 20000 rcf for 10 min and analyzed on ACQUITY UPLC I-class coupled with AB SCIEX TripleTOF 5600+. Analytical column is ACQUITY UPLC BECH C18 1.7 μm, 2.1X150 mm column. The JA concentration was calculated based on the calibration curve created by running JA standard solution. The results were analyzed by peakview1.2.
Measurements of total cellular glutathione in N. benthamiana leaves
Total cellular glutathione was measured as previously described (Sang et al, 2016). Briefly, 10 mg of N. benthamiana leaves were collected and glutathione was measured using a Glutathione Assay Kit (Beyotime, China) according to the manufacturer’s instructions.
Virus-induced gene silencing (VIGS) in N. benthamiana
VIGS in N. benthamiana plants was performed using TRV vectors as described (Senthil-Kumar & Mysore, 2014). VIGS of NbSGT1 was performed with several modifications described by Yu and collaborators (2019). Cultures of Agrobacterium carrying pTRV2:NbSGT1 plasmids or pTRV2 plasmids were mixed at 1:1 ratio and co-infiltrated into the lower leaves of 3-week-old N. benthamiana plants. The upper leaves were used for experimental assay within 7-10 days after VIGS application. Silencing of NRCs (NLR required for cell death) in N. benthamiana and subsequent expression of T3Es was performed as described by Wu and collaborators (2017).
Ralstonia solanacearum virulence assay
Four week-old A. thaliana plants, grown in Jiffy pots, were inoculated with R. solanacearum without wounding by soil drenching. An overnight-grown bacterial suspension was diluted to obtain an inoculum of 5.107 cfu/ml. Once the Jiffy pots were completely drenched, the plants were removed from the bacterial solution and placed back on a bed of potting mixture soil. The genotypes to be tested were placed in a random order in order to allow an unbiased analysis of the wilting. Daily scoring of the visible wilting on a scale ranging from 0 to 4 (or 0 to 100% leaves wilting), led to an analysis using the Kaplan-Meier survival analysis, log-rank test and hazard ratio calculation as previously described (Morel et al., 2018)
Acknowledgements
We thank Nemo Peeters and Anne-Claire Cazale for sharing unpublished biological materials, Chanhong Kim and Brian Staskawicz for sharing biological materials, Rosa Lozano-Duran for critical reading of this manuscript, and Xinyu Jian for technical and administrative assistance during this work. We thank the PSC Cell Biology, Proteomics, and Metabolomics core facilities for assistance with confocal microscopy and mass spectrometry analysis, respectively. This work was supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (grant XDB27040204), the National Natural Science Foundation of China (NSFC; grant 31571973), the Chinese 1000 Talents Program, and the Shanghai Center for Plant Stress Biology (Chinese Academy of Sciences). The authors have no conflict of interest to declare.