Ralstonia solanacearum elicitor RipX Induces Defense Reaction by Suppressing the Mitochondrial atpA Gene in Host Plant

RipX of Ralstonia solanacearum is translocated into host cells by a type III secretion system and acts as a harpin-like protein to induce a hypersensitive response in tobacco plants. The molecular events in association with RipX-induced signaling transduction have not been fully elucidated. This work reports that transient expression of RipX induced a yellowing phenotype in Nicotiana benthamiana, coupled with activation of the defense reaction. Using yeast two-hybrid and split-luciferase complementation assays, mitochondrial ATP synthase F1 subunit α (ATPA) was identified as an interaction partner of RipX from N. benthamiana. Although a certain proportion was found in mitochondria, the YFP-ATPA fusion was able to localize to the cell membrane, cytoplasm, and nucleus. RFP-RipX fusion was found from the cell membrane and cytoplasm. Moreover, ATPA interacted with RipX at both the cell membrane and cytoplasm in vivo. Silencing of the atpA gene had no effect on the appearance of yellowing phenotype induced by RipX. However, the silenced plants improved the resistance to R. solanacearum. Moreover, qRT-PCR and promoter GUS fusion experiments revealed that the transcript levels of atpA were evidently reduced in response to expression of RipX. These data demonstrated that RipX exerts a suppressive effect on the transcription of atpA gene, to induce defense reaction in N. benthamiana.


Introduction
The soil-borne pathogen Ralstonia solanacearum is a species complex that causes bacterial wilt disease in a wide range of host plants [1]. R. solanacearum isolates possess a meta-repertoire of type-III-dependent effectors, which comprise over 110 candidates that vary in different isolates [2,3]. In aids of transcription and translocation analysis, 72 effector candidates have been identified from the reference strain GMI1000 [4,5]. The effectors RipP1, RipAA1, and RipP2 specify the avirulent property to determine host range, which causes incompatible interactions with petunia, tobacco, or Arabidopsis [6][7][8]. RipAY, RipAL, RipAW, and RipAR show a suppressive effect on plant defense reactions [9][10][11]. On the other hand, transient expression of a few effectors induced hypersensitive response (HR)-like reactions in different plant species. For example, RipAX1 induces a hypersensitive response (HR) in eggplant, relying on a zinc protease motif [12].

External Application of RipX1 and RipX3 Proteins Induces Macroscopic HR Reactions
To examine the HR-eliciting property, RipX1 and RipX3 were prokaryotically expressed in pET41a(+) fused with a GST tag. After purification, 0.2 μg/μL GST-RipX1 and GST-RipX3 proteins were infiltrated into N. benthamiana leaves. The results showed that both fusion proteins induced rapid macroscopic HR reactions 24 h post-infiltration ( Figure 2A). Similar to the transient expression performance, the transcript levels of hsr203J, hin1, and PR1a were all elevated in N. benthamiana leaves that received GST-RipX1 and GST-RipX3 treatments ( Figure 2B). To determine the minimal concentration essential for HR induction, the purified proteins were diluted to 0.1, 0.05, and 0.025 μg/μL. Infiltration of 0.1 and 0.05 μg/μL proteins induced obvious HR in N. benthamiana leaves. When a concentration of 0.025 μg/μL was applied, neither GST-RipX1 nor GST-RipX3 was able to induce macroscopic HR ( Figure 2C). Thereafter, 0.05 μg/μL was the minimal concentration of RipX fusions used to induce macroscopic HR. . The infiltration areas were indicated by circles. All the experiments were repeated three times. (C) Activation of the defense genes hsr203J, hin1, and PR1a. The transcript level of each gene in plants transformed with empty vector pHB was set to 1, and levels in other plant samples were calculated relative to that. Data represent mean and standard deviation of three technical replicates. The asterisks denote statistical significance (** p < 0.01).

External Application of RipX1 and RipX3 Proteins Induces Macroscopic HR Reactions
To examine the HR-eliciting property, RipX1 and RipX3 were prokaryotically expressed in pET41a(+) fused with a GST tag. After purification, 0.2 µg/µL GST-RipX1 and GST-RipX3 proteins were infiltrated into N. benthamiana leaves. The results showed that both fusion proteins induced rapid macroscopic HR reactions 24 h post-infiltration ( Figure 2A). Similar to the transient expression performance, the transcript levels of hsr203J, hin1, and PR1a were all elevated in N. benthamiana leaves that received GST-RipX1 and GST-RipX3 treatments ( Figure 2B). To determine the minimal concentration essential for HR induction, the purified proteins were diluted to 0.1, 0.05, and 0.025 µg/µL. Infiltration of 0.1 and 0.05 µg/µL proteins induced obvious HR in N. benthamiana leaves. When a concentration of 0.025 µg/µL was applied, neither GST-RipX1 nor GST-RipX3 was able to induce macroscopic HR ( Figure 2C). Thereafter, 0.05 µg/µL was the minimal concentration of RipX fusions used to induce macroscopic HR. A GST tag was used as a negative control. (B) Activation of the defense genes hsr203J, hin1, and PR1a examined by qRT-PCR analyses. The expression level for each gene in plants transformed with empty vector pHB was set to 1, and the level in other samples was calculated relative to that. Error bars indicate standard deviations of three individual replicates, and asterisks denote statistical significance (** p ˂ 0.01). (C) Identification of the minimum concentration of the RipX protein essential for HR induction. All experiments were repeated three times.

Isolation and Phylogenetic Analysis of RipX-Interacting Protein ATPA from N. benthamiana
In yeast two-hybrid (Y2H) assays, three positive clones were obtained from the N. benthamiana cDNA library, using RipX1 as bait. After the pGADT7 plasmids were extracted from yeast cells, one plasmid showed a positive interaction with RipX1 in the repeated Y2H assays. DNA sequencing revealed that the plasmid harbored a 2221-bp cDNA fragment that contains a 1527-bp open reading frame encoding the full-length mitochondrial atpA gene. The coding region was cloned into pGADT7 to repeat the interaction with RipX1 ( Figure 3A). A split-luciferase assay was further performed to examine the interaction of RipX1 with ATPA ( Figure 3B).
The full-length atpA gene encodes a 309-amino-acid ATPA protein (GenBank: MN365024). The amino acid sequence of this protein differed from that of the N. tabacum protein at only position 165, where arginine was changed to glutamine. The obtained N. benthamiana ATPA shared over 98% identity with the homologs from Solanum pennellii, Solanum lycopersicum, Solanum pennellii, and Solanum tuberosum. A phylogenetic tree established from 12 selected ATPA proteins revealed that the ATPA of N. benthamiana was branched into a monophyletic clade with homologs from Solanaceae plants ( Figure 3C). Due to high conservation, the phylogenetic tree did not show distinctive diversity in Solanaceae plants. However, Solanaceae plant ATPAs showed distinctive differences from Ipomoea nil, Humulus lupulus, and Arabidopsis thaliana proteins ( Figure 3C). In the genome database of N. benthamiana, there were 20 predicted cDNA transcripts showing similarity with the obtained ATPA (Table S1). This finding indicated that there were multiple copies of ATPA in N. benthamiana.

Isolation and Phylogenetic Analysis of RipX-Interacting Protein ATPA from N. benthamiana
In yeast two-hybrid (Y2H) assays, three positive clones were obtained from the N. benthamiana cDNA library, using RipX1 as bait. After the pGADT7 plasmids were extracted from yeast cells, one plasmid showed a positive interaction with RipX1 in the repeated Y2H assays. DNA sequencing revealed that the plasmid harbored a 2221-bp cDNA fragment that contains a 1527-bp open reading frame encoding the full-length mitochondrial atpA gene. The coding region was cloned into pGADT7 to repeat the interaction with RipX1 ( Figure 3A). A split-luciferase assay was further performed to examine the interaction of RipX1 with ATPA ( Figure 3B).
The full-length atpA gene encodes a 309-amino-acid ATPA protein (GenBank: MN365024). The amino acid sequence of this protein differed from that of the N. tabacum protein at only position 165, where arginine was changed to glutamine. The obtained N. benthamiana ATPA shared over 98% identity with the homologs from Solanum pennellii, Solanum lycopersicum, Solanum pennellii, and Solanum tuberosum. A phylogenetic tree established from 12 selected ATPA proteins revealed that the ATPA of N. benthamiana was branched into a monophyletic clade with homologs from Solanaceae plants ( Figure 3C). Due to high conservation, the phylogenetic tree did not show distinctive diversity in Solanaceae plants. However, Solanaceae plant ATPAs showed distinctive differences from Ipomoea nil, Humulus lupulus, and Arabidopsis thaliana proteins ( Figure 3C). In the genome database of N. benthamiana, there were 20 predicted cDNA transcripts showing similarity with the obtained ATPA (Table S1). This finding indicated that there were multiple copies of ATPA in N. benthamiana.

RipX Interacts with ATPA at the Cell Membrane and Cytoplasm
RipX1 was fused to the C-terminus of red fluorescent protein (RFP) in the pGDR vector, and ATPA was fused to the C-terminus of yellow fluorescent protein (YFP) in the pGDY vector. RFP-RipX1 was localized to the cell membrane and cytoplasm, while YFP-ATPA was localized to the cell membrane, cytoplasm, and nucleus, as well as a few mitochondrial organelles ( Figure 4A,B). RFP-RipX1 and YFP-ATPA were colocalized at the cell membrane and cytoplasm ( Figure 4C). In an effort to detect the interaction in vivo, a bimolecular fluorescence complementation (BiFC) assay was employed to confirm that RipX1 and ATPA interacted at the cell membrane and cytoplasm ( Figure  4D). In this case, the expressions of ATPA-YC and RipX1-YN were analyzed by Western blot detection ( Figure 4E). To ensure the localization of ATPA to mitochondria, the mitochondrial marker MT-rk-CD3-991 was used. Upon transformation of the MT-rk-CD3-991 construct alone, a red fluorescence signal (RFP) was observed at the cell membrane, in the nucleus, and in a large number of mitochondria ( Figure S1A). When MT-rk-CD3-991 was co-transformed with GFP-ATPA, yellow fluorescence was observed in a few mitochondria, in addition to the cell membrane and nucleus ( Figure S1B). This finding indicated that ATPA did localize to mitochondria, although a large amount of ATPA was actually located in the cell membrane and nucleus.

RipX Interacts with ATPA at the Cell Membrane and Cytoplasm
RipX1 was fused to the C-terminus of red fluorescent protein (RFP) in the pGDR vector, and ATPA was fused to the C-terminus of yellow fluorescent protein (YFP) in the pGDY vector. RFP-RipX1 was localized to the cell membrane and cytoplasm, while YFP-ATPA was localized to the cell membrane, cytoplasm, and nucleus, as well as a few mitochondrial organelles ( Figure 4A,B). RFP-RipX1 and YFP-ATPA were colocalized at the cell membrane and cytoplasm ( Figure 4C). In an effort to detect the interaction in vivo, a bimolecular fluorescence complementation (BiFC) assay was employed to confirm that RipX1 and ATPA interacted at the cell membrane and cytoplasm ( Figure 4D). In this case, the expressions of ATPA-YC and RipX1-YN were analyzed by Western blot detection ( Figure 4E). To ensure the localization of ATPA to mitochondria, the mitochondrial marker MT-rk-CD3-991 was used. Upon transformation of the MT-rk-CD3-991 construct alone, a red fluorescence signal (RFP) was observed at the cell membrane, in the nucleus, and in a large number of mitochondria ( Figure S1A). When MT-rk-CD3-991 was co-transformed with GFP-ATPA, yellow fluorescence was observed in a few mitochondria, in addition to the cell membrane and nucleus ( Figure S1B). This finding indicated that ATPA did localize to mitochondria, although a large amount of ATPA was actually located in the cell membrane and nucleus.

Silencing of atpA Gene in N. benthamiana Has No Effect on HR Induction by RipX but Improves Resistance to R. solanacearum
Tobacco-rattle-virus-induced gene silencing (VIGS) was conducted to silence atpA in N. benthamiana. In comparison with the wild-type and gfp-silenced plants, the transcript level of atpA was reduced by 76% in atpA-silenced plants, indicating that the atpA gene was successfully disrupted ( Figure 5A). In the gfp-silenced plants, transient expression of RipX1 or RipX3 induced a yellowing phenotype, and application of the RipX1 and RipX3 proteins induced macroscopic HR ( Figure S2). This indicated that VIGS had no effect on defense induction by RipX. To know any deviation in host response to RipX, GST-RipX1 or GST-RipX3 was infiltrated in atpA-silenced plants.
The results showed that external application of GST-RipX1 or GST-RipX3 caused macroscopic HR ( Figure 5B). Simultaneously, silencing of the atpA gene showed no effect on RipX-induced yellow phenotype ( Figure 5B). However, R. solanacearum caused a slowed disease development on atpA-silenced plants. In comparison with wild-type and gfp-silenced control plants, the disease index of atpA-silenced plants remained significantly low after inoculation. At 15 days post-inoculation, the disease index reached only 40%; meanwhile, wild-type and gfp-silenced plants were all dead ( Figure  5C).

Silencing of atpA Gene in N. benthamiana Has No Effect on HR Induction by RipX but Improves Resistance to R. solanacearum
Tobacco-rattle-virus-induced gene silencing (VIGS) was conducted to silence atpA in N. benthamiana. In comparison with the wild-type and gfp-silenced plants, the transcript level of atpA was reduced by 76% in atpA-silenced plants, indicating that the atpA gene was successfully disrupted ( Figure 5A). In the gfp-silenced plants, transient expression of RipX1 or RipX3 induced a yellowing phenotype, and application of the RipX1 and RipX3 proteins induced macroscopic HR ( Figure S2). This indicated that VIGS had no effect on defense induction by RipX. To know any deviation in host response to RipX, GST-RipX1 or GST-RipX3 was infiltrated in atpA-silenced plants. The results showed that external application of GST-RipX1 or GST-RipX3 caused macroscopic HR ( Figure 5B). Simultaneously, silencing of the atpA gene showed no effect on RipX-induced yellow phenotype ( Figure 5B). However, R. solanacearum caused a slowed disease development on atpA-silenced plants. In comparison with wild-type and gfp-silenced control plants, the disease index of atpA-silenced plants remained significantly low after inoculation. At 15 days post-inoculation, the disease index reached only 40%; meanwhile, wild-type and gfp-silenced plants were all dead ( Figure 5C).

Expression of RipX1 Suppresses the Transcription of atpA Gene
The knock-down of atpA gene had no effect on the host response to RipX, and meanwhile promoted plant resistance to R. solanacearum. This prompted us to know the transcription pattern of atpA in response to RipX. Moreover, qRT-PCR analysis revealed that the transcript level of atpA was reduced by 52% in N. benthamiana expressing RipX1 relative to the uninoculated control ( Figure 6A). In the N. benthamiana genome database, the contig Niben101Scf00438Ctg035 contains a 2357-bp DNA fragment upstream of the atpA gene, comprising a 789-bp partial sequence of the cox3 (cytochrome oxidase subunit 3) gene. Four putative promoter elements were predicted from the atpA promoter region ( Figure S3). To confirm the expression pattern of atpA affected by RipX, an atpA promoter GUS fusion was constructed in the pCAMBIA1381 vector. In histochemical staining analysis, promoter-driven GUS activity was reduced upon co-transformation with RipX1 ( Figure 6B). The transcription of the promoter-driven gusA gene was additionally quantified by qRT-PCR. Expression of RipX1 led to a 72% reduction in gusA transcription compared with the expression of the promoter GUS fusion alone ( Figure 6B).

Expression of RipX1 Suppresses the Transcription of atpA Gene
The knock-down of atpA gene had no effect on the host response to RipX, and meanwhile promoted plant resistance to R. solanacearum. This prompted us to know the transcription pattern of atpA in response to RipX. Moreover, qRT-PCR analysis revealed that the transcript level of atpA was reduced by 52% in N. benthamiana expressing RipX1 relative to the uninoculated control ( Figure 6A). In the N. benthamiana genome database, the contig Niben101Scf00438Ctg035 contains a 2357-bp DNA fragment upstream of the atpA gene, comprising a 789-bp partial sequence of the cox3 (cytochrome oxidase subunit 3) gene. Four putative promoter elements were predicted from the atpA promoter region ( Figure S3). To confirm the expression pattern of atpA affected by RipX, an atpA promoter GUS fusion was constructed in the pCAMBIA1381 vector. In histochemical staining analysis, promoter-driven GUS activity was reduced upon co-transformation with RipX1 ( Figure 6B). The transcription of the promoter-driven gusA gene was additionally quantified by qRT-PCR. Expression of RipX1 led to a 72% reduction in gusA transcription compared with the expression of the promoter GUS fusion alone ( Figure 6B).

Discussion
In the present study, we examined the defense reactions of N. benthamiana in response to purified RipX protein, as well as the transient expression of the ripX gene. Most importantly, the mitochondrial ATPA was identified as a RipX-interacting protein from N. benthamiana. Significant downregulation of the atpA gene was found in plants expressing ripX; meanwhile, the silencing of atpA gene improved the resistance to R. solanacearum. These results demonstrate that RipX induces defense reaction by suppressing the atpA gene, which plays a role in host susceptibility.
Harpins are heat-stable elicitors that induce HR when infiltrated into the leaves of tobacco and other plants [24]. Coupled with HR induction, a range of hormone signaling pathways are evoked to promote resistance to pathogens or insect attacks [25,26]. Notably, harpins are associated with altered mitochondrial functions [27]. HrpN treatment immediately leads to an inhibitory effect on ATP synthesis in tobacco cell cultures [27]. Heteroexpression of HrpZPss in the yeast Saccharomyces cerevisiae Y187 resulted in cell death but not in the "Petite" mutant, indicating the involvement of mitochondria in yeast cell death [28]. Here, we provide direct evidence that harpin-like RipX targets mitochondrial ATPA, to induce defense reactions in N. benthamiana plants.
Several harpins have been extensively characterized from plant-pathogenic bacteria since the first harpin HrpN was identified from Erwinia amylovora [24,29]. Harpins have been proven to form pores in liposomes of the cell membrane [24]. However, the harpins of diverse plant-pathogenic bacteria differ substantially in their primary structure, even though they carry relatively high amounts of glycine and serine residues [24]. This suggested that harpins are recognized by different host targets in the cell membrane. HrpN-interacting protein (HIPM) was cloned from Malus spp. and found to be localized to the plasma membrane [30]. N. benthamiana ATPA showed no similarity with HIPMs, which supported the conclusion that harpins target different host genes to execute their elicitor function. A recent study reported that HIPM acts as a susceptibility gene, as transgenic apple plants with disrupted HIPM expression showed a significant decrease in susceptibility to E. amylovora infection [31]. The N. benthamiana atpA gene is a potential susceptibility gene for bacterial wilt because the silencing of atpA led to increased resistance to R. solanacearum.

Discussion
In the present study, we examined the defense reactions of N. benthamiana in response to purified RipX protein, as well as the transient expression of the ripX gene. Most importantly, the mitochondrial ATPA was identified as a RipX-interacting protein from N. benthamiana. Significant downregulation of the atpA gene was found in plants expressing ripX; meanwhile, the silencing of atpA gene improved the resistance to R. solanacearum. These results demonstrate that RipX induces defense reaction by suppressing the atpA gene, which plays a role in host susceptibility.
Harpins are heat-stable elicitors that induce HR when infiltrated into the leaves of tobacco and other plants [24]. Coupled with HR induction, a range of hormone signaling pathways are evoked to promote resistance to pathogens or insect attacks [25,26]. Notably, harpins are associated with altered mitochondrial functions [27]. HrpN treatment immediately leads to an inhibitory effect on ATP synthesis in tobacco cell cultures [27]. Heteroexpression of HrpZ Pss in the yeast Saccharomyces cerevisiae Y187 resulted in cell death but not in the "Petite" mutant, indicating the involvement of mitochondria in yeast cell death [28]. Here, we provide direct evidence that harpin-like RipX targets mitochondrial ATPA, to induce defense reactions in N. benthamiana plants.
Several harpins have been extensively characterized from plant-pathogenic bacteria since the first harpin HrpN was identified from Erwinia amylovora [24,29]. Harpins have been proven to form pores in liposomes of the cell membrane [24]. However, the harpins of diverse plant-pathogenic bacteria differ substantially in their primary structure, even though they carry relatively high amounts of glycine and serine residues [24]. This suggested that harpins are recognized by different host targets in the cell membrane. HrpN-interacting protein (HIPM) was cloned from Malus spp. and found to be localized to the plasma membrane [30]. N. benthamiana ATPA showed no similarity with HIPMs, which supported the conclusion that harpins target different host genes to execute their elicitor function. A recent study reported that HIPM acts as a susceptibility gene, as transgenic apple plants with disrupted HIPM expression showed a significant decrease in susceptibility to E. amylovora infection [31]. The N. benthamiana atpA gene is a potential susceptibility gene for bacterial wilt because the silencing of atpA led to increased resistance to R. solanacearum.
Because HR induction by harpin was initially discovered from external infiltration into plant tissue, it is widely accepted that harpins are perceived by the host plant as extracellular signals. However, emerging studies have demonstrated that harpin transgenic plants show enhanced resistance to pathogen attacks [16,32]. It appears that harpins can be perceived by the host plant inside of cells, although it is uncertain whether all harpins are translocated into plant cells during pathogen infection. By using a calmodulin-dependent adenylate cyclase reporter system, small amounts of the HrpN and RipX proteins have been successfully detected in plant-host cells [5,33]. In this study, transient expression of RipX induced a weak defense reaction in N. benthamiana cells. In contrast, external infiltration of the RipX protein induced macroscopic HR reactions. This discrepancy may be caused by the different perception mechanisms of N. benthamiana plants. In fact, we started to study the phenotype of transient expression of full-length RipX1 in N benthamiana. Because transient expression did not cause macroscopic cell death, a RipX3 construct was then used to confirm the phenotype. Thereafter, RipX induces diverse signaling pathways, either inside or outside of N. benthamiana cells.
In conclusion, our work shows that the R. solanacearum harpin RipX induces the defense reaction by modulating the mitochondrial atpA gene. When translocated into host cells through a type III secretion apparatus, RipX interacts with ATPA and exerts a suppressive effect on the transcription of the atpA gene. The particular mode provides details regarding the defense signaling pathways triggered by RipX.

Plant Material and Cultivation
N. benthamiana seeds were germinated in sterilized soil mix (peat moss:perlite (2:1, v/v)), in plastic pots. After germination, the seedlings were transferred to the Pindstrup Substrate (Pindstrup Mosebrug A/S, Denmark) potting mixture. Plants were grown in a greenhouse under a 16 h/8 h light/dark photoperiod, at 25 • C.

Bacterial Strains and Plasmids
All plasmids and strains used in this study are listed in Table S2. Escherichia coli DH5α and Agrobacterium tumefaciens GV3101 were cultivated in Luria-Bertani medium, at 37 and 28 • C, respectively. R. solanacearum strain FJ1003 was grown at 28 • C in nutrient-rich broth medium [34]. Yeast strain AH109 was cultured in YPD medium (1% yeast extract, 2% peptone, and 2% glucose), at 30 • C. The 1035-bp ripX1 and 750-bp ripX3 were PCR amplified, to generate pHB:RipX1 and pHB:RipX3 for transient expression analysis. To study the subcellular localization, RipX1 was fused to RFP in pGDR, while RipX3 was fused to RFP in pGD-3G-mCherry. ATPA was fused to the C-termini of YFP and GFP in the vectors pGDY and pGDGm, respectively. To determine mitochondrial localization, the mitochondria targeting marker construct MT-rk-CD3-991 was co-transformed with GFP-ATPA. In the BiFC assays, RipX1 were cloned into 1301-YN for fusion to the N-terminal fragment of YFP. ATPA was fused with the C-terminal fragment of YFP in 1301-YC. The constructs were individually transformed into A. tumefaciens GV3101 for transient expression assays.

DNA Manipulation
DNA isolation, restriction enzyme digestion, and plasmid transformation were performed according to standard protocols [35]. The PCR primers used for molecular cloning, PCR product sizes, and detailed enzyme cleavage sites are listed in Table S3.

Histochemical Staining
Cell death and hydrogen peroxide accumulation were examined by trypan blue and 3,3'-diaminobenzidine (DAB) staining, respectively. For trypan blue staining, N. benthamiana leaves were boiled in trypan blue solution for 20 min, kept at room temperature for 8 h, and then transferred into 2.5 g/mL chloral hydrate solution for de-staining by boiling for 25 min. For DAB staining, N. benthamiana leaves were immersed in 1 mg/mL DAB solution for 8 h and then cleared in 95% ethanol.

Confocal Microscopy
Leaf disks were harvested for fluorescence signal examination by a Leica confocal laser scanning microscope (SP8, Leica, Germany). After fluorescence signals were captured, the images were saved with a resolution of 1024 × 1024 pixels. For each construct or co-transformation analysis, three different samples were examined. Experiments were repeated three times.

Quantitative Real-Time PCR (qRT-PCR)
Total RNA was extracted from N. benthamiana leaves, using the Plant RNA Kit (Omega, Shanghai, China). RNA sample examination and cDNA synthesis were performed as described previously [36]. The primers are listed in Table S3. The qRT-PCR experiments were performed with the following parameters: denaturation at 95 • C for 30 s, followed by 40 cycles of 95 • C for 5 s and 55 • C for 20 s. The expression of EF1α was used as an internal control. Relative expression levels and statistical analyses were conducted by CFX Maestro software [36]. Each experiment was repeated three times.

Examination of the HR Reaction Induced by RipX
The ripX1 and ripX3 genes were cloned into pET41a(+) for fusion with a GST tag (Table S2). After the constructs were separately transformed into BL21(DE3), GST fusion proteins were induced with 0.1 mM isopropyl-β-d-thiogalactopyranoside and purified from the supernatant, using glutathione resin [36]. The purity and concentration of protein samples were analyzed by 10% SDS-PAGE and the Bradford method. Then, 0.2 µg/µL GST-RipX1 and GST-RipX3 proteins were infiltrated into N. benthamiana leaves, to detect HR induction. The GST tag was used as a negative control. To determine the minimum concentration essential for HR induction, the proteins were infiltrated into N. benthamiana leaves at 0.1, 0.05, and 0.025 µg/µL. At 24 h post-infiltration, detection of HR induction, histochemical staining, and qRT-PCR were conducted. Each performance was repeated three times.

Y2H
The construct pGBKT7:RipX1 and cDNA library of N. benthamiana were co-transformed into the yeast host strain AH109, by LiAc-mediated transformation. The pGADT7 plasmids were extracted from the screened positive colonies, to repeat the Y2H assay. The plasmid that exhibited a positive interaction with RipX1 was identified as ATPA based on DNA sequencing. The coding region of the atpA gene was amplified and cloned into pGADT7 to generate pGADT7:ATPA. After that, pGADT7:ATPA and pGBKT7:RipX1 were then co-transformed into AH109. The transformant was examined for growth on SD/-Ade/-Leu/-Trp/-His/ supplemented with 20 µg/mL X-α-galactosidase [36]. The Y2H repeat experiment was performed three times.

Split-Luciferase Complementation Assay
The coding sequence of RipX1 was cloned into pCAMBIA-CLuc to generate cLUC-RipX1. In addition, the coding sequence of ATPA was cloned into pCAMBIA-NLuc to generate ATPA-nLUC [37]. A mixture of A. tumefaciens GV3101 carrying the cLUC-RipX1 and ATPA-nLUC constructs was infiltrated into N. benthamiana leaves. The leaves were harvested 2 days post-agroinfiltration, rubbed with 0.5 mM luciferin, and kept in the dark for 5 min, to quench the fluorescence. A cooled charge-coupled device (CCD) imaging apparatus (Roper Scientific) was used to capture luciferase images.

Virus-Induced Gene Silencing (VIGS)
A 309-bp fragment of atpA was inserted into the pTRV2 vector. TRV2:PDS was used as the control, to evaluate the silencing efficiency [38]. TRV2:gfp carrying a 358-bp fragment of the gfp gene was used as a negative control. A mixture of GV3101 cultures (1:1 v/v) containing pTRV1 and pTRV2 constructs was co-infiltrated into 20-day-old N. benthamiana leaves. When newly grown leaves were fully expanded, qRT-PCR analysis was performed, to detect the transcripts of atpA. Next, transient expression of RipX1 and RipX3 was performed. Detection of the HR reaction, histochemical staining, and qRT-PCR analysis of defense genes were conducted 2 days post-agroinfiltration. Meanwhile, 0.05 µg/µL GST-RipX1 and GST-RipX3 were infiltrated into atpA-silenced plants. Detection of the HR reaction and qRT-PCR analysis of defense genes were conducted 24 h post-infiltration. The silencing experiments were repeated four times, with four independent plants for each repeat.

Resistance of N. benthamiana to Bacterial Wilt
The soil inoculation method was used to assess the disease development in atpA-silenced N. benthamiana plants [39]. The cultured cells of R. solanacearum were prepared to a final concentration of 1 × 10 8 CFU/mL. The roots of wild-type, atpA-silenced or gfp-silenced control plants were first wounded and then were inoculated with R. solanacearum cell suspension, at 10 mL per pot. Disease severity and the percent severity index were recorded, using a previously described method [39]. This experiment was repeated four times. For every repeat, 6 wild-type, atpA-silenced, or gfp-silenced plants were assayed for disease development.

Promoter Activity Assays
The 1488-bp promoter region of atpA was cloned into pCAMBIA1381 in front of the gusA gene. The resultant pCAMBIA-PatpA construct was transformed into GV3101, for transient expression. GUS activity was studied by both histochemical staining and evaluation of gusA gene transcript levels 2 days after agroinfiltration. For histochemical staining, the infiltrated leaves were immersed in GUS staining solution (50 mM phosphate buffer (pH 7.0), 2 mM K 3 [Fe(CN) 6 ], 2 mM K 4 [Fe(CN) 6 ], and 0.1% 5-bromo-4-chloro-3-indolyl-β-d-glucuronide), at 37°C overnight. Chloroplasts were removed with 70% ethanol, to produce a transparent background. The transcript level of the gusA gene was studied by qRT-PCR analysis. The experiments were repeated three times.

Western Blot Analysis
Total protein was extracted from N. benthamiana leaf disks by using Laemmli buffer at 2 days post-inoculation [40]. The proteins were then resolved by 12% SDS-PAGE and subjected to immunoblot analysis with anti-GFP and anti-Myc.

Sequence and Data Analysis
The amino acid sequence of ATPA was used to retrieve the sequences of 11 plant homologs deposited in GenBank. Twelve ATPAs from representative plant species were subjected to phylogenetic analyses, using the neighbor-joining (NJ) method implemented in MEGA5 [41]. The evolutionary distances were computed by using the Poisson correction method and are presented as the number of amino acid substitutions per site. The confidence probability (multiplied by 100) that the interior branch length is greater than 0, as estimated by using the bootstrap test (1000 replicates), is shown next to the branches. ATPA homologs in N. benthamiana were collected from the genome database (https://solgenomics.net/organism/Nicotiana_benthamiana/genome). The promoter sequence was predicted with http://www.fruitfly.org/seq_tools/promoter.html. Except for qRT-PCR analysis, the statistical analyses were conducted with SPSS 17.0 (SPSS Inc., Chicago, IL, USA). Statistically significant differences were conducted based on Tukey's honest significant difference (HSD) test, using a one-way analysis of variance (ANOVA).
Supplementary Materials: The following are available online at http://www.mdpi.com/1422-0067/21/6/2000/s1, Figure S1: The mitochondrial targeting characteristic of ATPA, Figure S2: Examination of gfp-silenced plants in response to RipX, Figure S3: Sequence analysis of the atpA promoter region, Table S1: ATPA homologs found from Nicotiana benthamiana genome database, Table S2. Bacterial strains and plasmids used in this study, Table  S3. Primers used for molecular cloning and qRT-PCR in this study.

Conflicts of Interest:
The authors declare no conflict of interest.