Tomato (Solanum lycopersicum cv. Glamour) and Nicotiana benthamiana plants were maintained in the greenhouse under a 14 h light/10 h dark photoperiod at 25±2°C. The tomato seedlings assay was used for expression analysis of the SlJAZ genes [18]. Briefly, tomato seeds were germinated and maintained on 1/2 Murashige and Skoog (MS) medium (0.3% phytagel) with Gamborg vitamins (PhytoTechnologies Laboratories, Shawnee Mission, KS, USA) and used for assay 4 days after germination at 25°C in darkness. Tomato seedlings were treated with water as a mock control or 200 pmol of COR (Sigma-Aldrich, St. Louis, MO, USA). Tomato seedlings were inoculated by a method whereby MS agar plates were flooded with bacterial suspension as described previously [18]. To perform uniform inoculation, the bacterial suspension (OD600 = 0.1) containing 0.025% Silwet L-77 (OSi Specialties Inc., Danbury, CT, USA) was dispensed into Petri dishes until the seedlings were completely submerged, and the seedlings were exposed to bacterial cells for 2–3 min with gentle agitation. After the inoculum was discarded, Petri dishes containing inoculated seedlings were incubated with a light intensity of 200 µE m-2 sec-1 and a 12 h light/12 h dark photoperiod.

Pseudomonas syringae pv. tomato DC3000 (Pst DC3000) was used as the wild-type strain for virulence assays on tomato. The COR-defective mutant Pst DB29 was previously described [19] and was kindly provided by Dr. Barbara Kunkel (Washington University, St Louis, USA). P. syringae pv. tomato T1 (Pst T1) was used as the nonhost pathogen on N. benthamiana. The strains of Pst DC3000, Pst DB29 and Pst T1 were grown at 28°C on mannitol-glutamate (MG) medium [20] containing antibiotics in the following concentrations (µg ml−1): rifampicin, 25; kanamycin, 50; and spectinomycin, 25 for 36–48 h. Prior to inoculation, bacteria were suspended in 1 ml sterile distilled H₂O.

Pathogen inoculation assays on vector control and VIGS-silenced tomato were conducted as described [8]. Briefly, bacterial suspensions (OD600 = 0.1) were prepared in sterile distilled water containing 0.025% Silwet L-77 (OSI Specialties Inc., Danbury, CT, USA) and spray-inoculated to runoff. The inoculated plants were then incubated in growth chambers at ∼100% RH for the first 24 h and at ∼70% RH for the rest of the experimental period. The inoculated plants were observed for 7 dpi for symptom development. Bacterial growth in leaves was measured by determining the internal bacterial populations. Prior to sampling, leaves were surface-sterilized with 15% H₂O₂ for 3 min to eliminate epiphytic bacteria and then washed with sterile distilled water. The leaves were then homogenized in sterile distilled water, and serial dilutions were plated onto MG medium containing antibiotics. The bacterial population at 0 dpi was estimated from leaves harvested 1 h post inoculation. Bacterial growth was evaluated in three independent experiments.

To analyze the impact of nonhost pathogen-induced HR cell death in wild-type and SlJAZs-silenced N. benthamiana plants, bacterial suspensions (OD600 = 0.05) of Pst T1 were prepared and infiltrated into leaves using a 1 ml needleless syringe. HR cell death was observed 24 hours after infiltration.

Agrobacterium tumefaciens GV2260 containing pTRV1 and pTRV2 with a gene of interest (e.g., pTRV2::GFP, pTRV2::SlJAZ2, pTRV2::SlJAZ6, and pTRV2::SlJAZ7) were grown overnight on LB medium containing antibiotics (rifampicin, 25 µg/ml; kanamycin, 50 µg/ml) at 28°C. Bacterial cells were harvested and resuspended in induction medium (10 mM MES pH 5.5; 200 µM acetosyringone) and incubated at room temperature on an orbital shaker for 5 h. The bacterial cultures containing pTRV1 and pTRV2 with the gene of interest were mixed in equal ratios (OD600 = 0.5) and infiltrated into lower leaves using a 1 ml needleless syringe [21]. The infiltrated plants were maintained in the greenhouse and used for studies 15–21 days post-infiltration.

To analyze the impact of PAMP-mediated HR in wild-type and SlJAZs-silenced plants, A. tumefaciens GV2260 containing 35S::INF1 within the T-DNA of a binary vector was grown overnight on LB medium containing antibiotics (rifampicin, 25 µg/ml; kanamycin, 50 µg/ml) at 28°C. Bacterial cells were harvested and resuspended in induction medium and incubated at room temperature in an orbital shaker for 12 h. Agrobacterium cultures (OD600 = 0.8) were infiltrated into leaves using a 1 ml needleless syringe. Four days after infiltration, HR cell death was observed.

Total RNA extraction and real-time quantitative RT-PCR (qRT-PCR) were done as described previously [11]. Total RNA was extracted using TRIzol (Sigma-Aldrich) according to manufacturer’s protocol. Total RNA was treated with Turbo DNase (Ambion, Austin, TX, USA) to eliminate genomic DNA, and 5 µg of DNase-treated RNA was reverse transcribed using Superscript III reverse transcriptase (Invitrogen, Carlsbad, CA, USA) with oligo d(T)15–20 primers (Invitrogen). The cDNA (1∶20) was then used for qRT-PCR that was performed using primers shown in Table S1 with Power SYBR Green PCR master mix (Applied Biosystems, Foster City, CA, USA) on an ABI Prism 7900 HT sequence detection system (Applied Biosystems). The N. benthamiana and tomato Actin genes were used as internal controls. To study the downregulation of transcripts in silenced plants, qRT-PCR primers that anneal outside the region targeted for silencing were used (Table S1). Average CT values calculated using Sequence Detection Systems (version 2.2.2; Applied Biosystems) from triplicate samples were used to determine the fold expression relative to controls.

The Matchmaker yeast two-hybrid system from Clontech (Takara Bio, Kyoto, Japan) was used according to manufacturer’s protocols. Briefly, full-length cDNA for tomato genes, including SlCOI1, SlJAZ1, SlJAZ2, SlJAZ3, SlJAZ4, SlJAZ5, SlJAZ6, SlJAZ7, SlJAZ8, SlJAZ9, SlJAZ10, SlJAZ11, SlJAZ12 and SlNINJA, were amplified by RT-PCR with specific primer sets (Table S1) and then were inserted into the yeast expression vectors pGBKT7 (bait) and pGADT7 (prey) using In-Fusion® HD Cloning Kit from Clontech (Takara Bio). To determine the interactions between fusion proteins, both bait and prey plasmids were co-transformed into MaV203 yeast strain (Invitrogen) carrying three GAL4-inducible reporter genes (lacZ, HIS3 and URA3). Bait-prey interactions were selected on the synthetic dropout medium lacking Leu and Trp (SC-Leu-Trp, Sigma-Aldrich). The yeast colony grown in SC-Leu-Trp was streaked on the medium lacking Leu, Trp and His (SC-His-Leu-Trp, Sigma-Aldrich) supplemented with 10 mM 3-amino-1,2,4-triazole (Sigma-Aldrich) with or without COR (20 µg/mL).

Ion leakage is used as an indirect quantitative measure for the disease-associated cell death. Ion leakage was measured as described previously [11]. Six leaf discs were punched from N. benthamiana leaves inoculated with Pst T1 and A. tumefaciens carrying 35S::INF1, and agitated in 30 ml of distilled water for 3 h; electrical conductivity was measured using an ion conductivity meter (Orion 555A, Thermo Electron Corp., Marietta, OH, USA). Plants were then autoclaved for 20 min to kill the cells and release total ions into the medium. Values relative to the whole ion content after autoclaving were used to express the percent ion leakage.

It has been reported that there are 12 members of JAZ family genes in Arabidopsis and tomato [22], [23](Fig. S1 and Table S2). In Arabidopsis, it was demonstrated that the expression of AtJAZs was upregulated during Pst DC3000 infection in a COR-dependent manner [24]. To examine the expression profiles of tomato SlJAZs during Pst DC3000 infection, tomato seedlings were treated with water or COR, or inoculated with Pst DC3000 or the COR-defective mutant Pst DB29, and qRT-PCR analysis was performed. Based on previous reports for JAZ family genes in tomato [23], we developed gene specific primer sets for 12 tomato SlJAZ genes including SlJAZ1 (Solyc07g042170), SlJAZ2 (Solyc12g009220), SlJAZ3 (Solyc03g122190), SlJAZ4 (Solyc12g049400), SlJAZ5 (Solyc03g118540), SlJAZ6 (Solyc01g005440), SlJAZ7 (Solyc11g011030), SlJAZ8 (Solyc06g068930), SlJAZ9 (Solyc08g036640), SlJAZ10 (Solyc08g036620), SlJAZ11 (Solyc08g036660) and SlJAZ12 (Solyc01g009740) (Table S1 and S2). The expression of all SlJAZ genes was induced by Pst DC3000 infection (Fig. 1). The expression of SlJAZ3, SlJAZ5, SlJAZ6, SlJAZ8 and SlJAZ12 was not induced after treatment with purified COR (Fig. 1). In contrast, SlJAZ1 and SlJAZ2 were moderately induced, and SlJAZ4, SlJAZ7, SlJAZ9, SlJAZ10 and SlJAZ11 were highly induced by purified COR (Fig. 1). It has been reported that the expression of JAZ was activated by JA-induced degradation of JAZ proteins in Arabidopsis and rice [25], [26]. We demonstrated that 12 SlJAZ genes were induced by Pst DC3000, but only seven of 12 JAZ genes were upregulated in response to COR. In Arabidopsis, most of the JAZ genes were induced within 30 min after JA treatment [26]. In this study, we examined the expression of SlJAZs at 3 hours after treatment with purified COR, suggesting that the expression of SlJAZ3, SlJAZ5, SlJAZ6, SlJAZ8 and SlJAZ12 could be transiently induced within 3 hours after treatment with purified COR.

Together with other studies in Arabidopsis, our results suggest that the expression of JAZ genes was differently regulated by environmental stimuli such as pathogen infection, wounding and herbivory [24], [27]. It has been reported that only eight of 12 JAZ genes were upregulated during Pst DC3000 infection in Arabidopsis [24]. We demonstrated that all JAZ genes are significantly induced by Pst DC3000 in tomato (Fig. 1). Although the JA-mediated signaling pathway is critical for the pathogenicity of Pst DC3000 in Arabidopsis and tomato, different sets of genes were expressed during Pst DC3000 infection between Arabidopsis and tomato [18], [28]. Therefore, it is tempting to speculate that Pst DC3000 uses different strategies to modulate disease development by regulating different JAZ genes during pathogenesis between tomato and Arabidopsis.

To investigate whether COR is the virulence factor responsible for the expression of JAZ genes in tomato, we examined the COR-defective mutant Pst DB29 and found that the induction of JAZ genes in tomato is COR-dependent (Fig. 1). Consistent with our results, it has been reported that COR is the primary virulence factor inducing the expression of JAZ genes in Arabidopsis [24]. These results clearly indicate that the COR-induced JA-signaling pathway has a critical role in the pathogenicity of Pst DC3000 in both Arabidopsis and tomato.

In previous studies, it was demonstrated that some JAZ proteins interact with COI1 to form a receptor complex for JA-Ile/COR in Arabidopsis, tomato and rice [5], [25], 29,30. To investigate whether SlJAZ proteins form a receptor complex with SlCOI1, we cloned full length ORFs for 12 SlJAZs (Fig. S2) and examined the interaction between SlJAZs and SlCOI1 using the yeast two-hybrid system. Consistent with a previous report [5], SlJAZ2 and SlJAZ6 interacted with SlCOI1 in a COR-dependent manner (Fig. 2A). In addition to SlJAZ2 and SlJAZ6, we found SlJAZ7 interacted with SlCOI1 in the presence of COR, suggesting that SlJAZ2, SlJAZ6 and SlJAZ7 could form a receptor complex for JA signaling in tomato (Fig. 2A). Unlike Arabidopsis, most SlJAZ proteins failed to interact with SlCOI1 in yeast two-hybrid system. This could be due to low expression of these SlJAZ proteins in yeast. The use of other methods for protein interactions, such as pull-down assay, may provide clues to the interactions between SlCOI1 and SlJAZ proteins.

It has been demonstrated that JAZ proteins also interact with the adaptor protein NINJA and form a repressor complex together with TOPLESS in Arabidopsis [17]. To investigate whether SlJAZ proteins form a repressor complex with SlNINJA or not, we cloned full-length ORF for SlNINJA (Fig. S2) and examined the interaction between SlJAZs and SlNINJA. SlJAZ2, SlJAZ6 and SlJAZ7 interacted with SlNINJA (Fig. 2B), suggesting that SlJAZ2, SlJAZ6 and SlJAZ7 could form a repressor complex to regulate JA-inducible genes.

Since SlCOI1 interacted with tomato JAZ proteins SlJAZ2, SlJAZ6 and SlJAZ7 (Fig. 2A), we determined the requirement of these JAZ proteins for COR-induced chlorosis using virus-induced gene silencing (VIGS). One of the hallmarks of bacterial speck disease on tomato leaves is the formation of necrotic lesions surrounded by chlorosis. COR is known to induce chlorosis in a COI1-dependent manner [14]. Since JAZ proteins are known to function as repressors, we investigated whether silencing of SlJAZs leads to hypersensitivity to COR. We treated vector control (TRV::GFP) and SlJAZs-silenced tomato and N. benthamiana plants with 200 pmol of COR and observed COR-induced chlorosis. To avoid off-target effect of VIGS, we selected gene specific region as the trigger sequence, which does not have 21 nucleotide homology to other SlJAZs. The efficiency of silencing was confirmed by the quantification of the endogenous transcript levels using qRT-PCR (Fig. S3). As shown in Fig. S4, there was no significant difference on COR-induced chlorosis between control and SlJAZs-silenced plants in tomato and N. benthamiana, indicating that SlJAZ2, SlJAZ6 and SlJAZ7 do not play an active role in COR-induced chlorosis. It is possible that these JAZ proteins have functional redundancy in COR-induced chlorosis. It was reported that Atjaz10 mutant showed hypersensitivity to JA, since alternative splice variants functioned as the stable repressors of transcription [31]. Although SlJAZ2, SlJAZ6 and SlJAZ7 have no high homology to AtJAZ10, further investigation about alternative splice variants should help us to better understand the function of SlJAZs in COR-induced chlorosis.

To determine whether SlJAZ2, SlJAZ6 and SlJAZ7 play a role in disease symptom development induced by Pst DC3000, we spray-inoculated control (TRV::GFP) and SlJAZs-silenced tomato plants with Pst DC3000. The control plants developed typical bacterial speck symptoms with necrotic lesions surrounded by chlorotic halos (Fig. 3A). However, SlJAZ2-, SlJAZ6- and SlJAZ7-silenced plants developed accelerated necrotic lesions with reduced chlorotic halos (Fig. 3A). Interestingly, no significant differences in bacterial multiplication were observed between the control and SlJAZs-silenced plants at 2 and 6 days after inoculation (Fig. 3B). In previous study, we have demonstrated that silencing of NTRC and Prx resulted in accelerated necrotic lesions during Pst DC3000 infection without increasing the bacterial growth [14]. In Arabidopsis, it has been reported that the JA-mediated signaling pathway leading to disease development is distinct from the signaling pathway regulating the bacterial multiplication [32]. Therefore, these results indicated that the disease phenotypes in SlJAZs-silenced plants might be due to hypersensitivity to COR and JAZ proteins may function as a negative regulator of disease-associated cell death without altering the pathogen multiplication.

It has been reported that the Arabidopsis Atjaz10 mutant has enhanced susceptibility to Pst DC3000 [24]. Consistent with our results, they demonstrated that AtJAZ10 affects disease development without increasing the bacterial populations [24]. Together, these results indicate that some members of JAZ proteins may function as negative regulators of disease development during Pst DC3000 infection in tomato and Arabidopsis.

In a previous report, we demonstrated that silencing of THYLAKOID FORMATION1 (THF1) in N. benthamiana and tomato produced a necrosis-like phenotype to COR application. THF1 silenced tomato and Arabidopsis thf1 mutant showed enhanced necrotic lesions with reduced chlorosis in response to Pst DC3000 [33]. As shown in Fig. 3A, SlJAZ2-, SlJAZ6 and SlJAZ7-silenced plants also developed enhanced necrotic lesions. Therefore, we speculate that JAZ protein may function as a negative regulator of disease development by modulating THF1.

To further investigate the functional role of JAZ in plant-microbe interactions, we examined nonhost pathogen- and PAMP-mediated hypersensitive response (HR) cell death in JAZs-silenced N. benthamiana plants. To determine whether JAZs affect HR cell death, we syringe-infiltrated control (TRV::GFP) and JAZs-silenced N. benthamiana plants with the nonhost bacterial pathogen P. syringae pv. tomato T1 and A. tumefaciens that can express 35S-INF1 in planta. The control and JAZ2-silenced plants showed HR cell death in response to nonhost pathogen and INF1 at 24 hpi and 3 dpi, respectively (Fig. 4). Interestingly, JAZ6- and JAZ7-silenced plants delayed HR cell death compared with control plants, suggesting that JAZ6 and JAZ7 may have a role in nonhost pathogen- and PAMP-mediated signaling pathways leading to HR cell death (Fig. 4).

It has been demonstrated that the N. benthamiana JAZ protein NbPPS3, which has a high similarity to SlJAZ5, has a role in the PAMP-mediated signaling pathway leading to HR cell death [34]. Katou et al. reported that NbPPS3 protein was phosphorylated by both SIPK and WIPK in response to PAMP. Silencing of NbPPS3 delayed PAMP-mediated HR cell death, indicating that NbPPS3 is one of the downstream components of the MAPK-mediated signaling pathway leading to HR cell death [34]. In addition, Oh et al. reported that silencing of Nicotiana attenuata JAZ gene NaJAZh, which has a high similarity to SlJAZ3, showed spontaneous cell death associated with the accumulation of salicylic acid and hydrogen peroxide in the leaves [35]. Therefore, it is likely that the COI1-JAZ complex and/or other components of JA/COR signaling pathways play a role during HR cell death.