Silicon Controls Bacterial Wilt Disease in Tomato Plants and Inhibits the Virulence-Related Gene Expression of Ralstonia solanacearum

Silicon (Si) has a multifunctional role in improving plant growth and enhancing plant disease resistance, but its mechanisms are not fully understood. In this study, we investigated the impacts of silicon application on the control of bacterial wilt and elucidated the molecular mechanisms using transcriptome sequencing. Compared to non-Si treatment, Si application (0.5–2 mM) significantly reduces tomato bacterial wilt index by 46.31–72.23%. However, Si does not influence the growth of R. solanacearum. Si application negatively influences R. solanacearum exopolysaccharide (EPS) synthesis and biofilm formation. Transcriptome analysis showed that Si treatment significantly downregulates the expression of virulence genes’ transcriptional regulator (xpsR), EPS synthesis-related genes (epsD and tek), and type III effectors (HrpB2, SpaO, and EscR) in R. solanacearum. In addition, Si remarkably upregulates the expression of twitch motor-related genes (pilE2, pilE, fimT, and PilX). These findings suggest that silicon-suppressed tomato wilt incidence may be due to the regulation of the virulence-related genes of R. solanacearum by Si. Our research adds new knowledge to the application of Si in the field of disease control.

Ralstonia solanacearum (R. solanacearum) is a soilborne plant pathogen that invades the plant roots from the soil and aggressively colonizes the xylem vessels, which causes wilt symptoms. R. solanacearum infection leads to lethal wilting disease in more than 200 plant species and is particularly harmful to Solanaceae plants [15,16]. Many studies report that Si significantly suppresses bacterial wilt in hydroponic, soil, and substrate culture conditions [17][18][19]. However, most of these studies only focus on the interaction between Si and plants, ignoring the effect of silicon on R. solanacearum. This pathogen is extremely vigorous and can survive for a long time in soil and water; it can infect different plant tissues through a series of virulence determinants [20]. The virulence factors include plant cell-wall-degrading enzymes, bacterial extracellular polysaccharide (EPS), type III

Silicon Application Reduced Disease Index of Bacterial Wilt in Tomato
The low Si concentrations (0.05 and 0.1 mM) did not significantly alleviate bacterial wilt in tomato plants ( Figure 1a). However, when the added Si concentration was over 0.5 mM, the disease index of bacterial wilt significantly decreased. Compared to the control (0 mM Si), applying 0.5, 1.0, and 2.0 mM of Si reduced the disease index by 46.31%, 66.67%, and 72.23%, respectively ( Figure 1b). Regression analysis showed that the disease index was significantly and negatively correlated with Si concentration (R 2 = 0.92, p < 0.01; Figure 1c).

Silicon Did Not Affect the Growth of R. solanacearum
The results show there was no significant difference in the growth of R. solanacearum among all treatments ( Figure 2). The growth curve of R. solanacearum in the liquid Luria-Bertani (LB) and minimal mineral (MM) medium was not significantly influenced by Si treatment with different concentrations, indicating that Si did not have a direct inhibition on R. solanacearum.

Silicon Regulated the Expression of 119 Genes of R. solanacearum In Vitro
The sequencing data were mapped onto the R. solanacearum strain GMI1000 reference genome [32], and 4,496 genes were detected ( Figure 3). In this study, 119 differentiallyexpressed genes (DEG) were found to be significantly influenced by Si application, accounting for 2.6% of all known genes. Among these 119 genes, 57 were upregulated, and 62 were downregulated (Tables S1 and S2).
The expression of 15 genes (gene primer was shown in Table S3) related to the virulence of R. solanacearum was analyzed by qRT-PCR, including four biological replicates under two housekeeping genes (GAPDH and thyA). The qRT-PCR data for these genes were significantly correlated with the RNA-Seq results in two housekeeping genes, namely GAPDH ( Figure S1a) and thyA ( Figure S1b), which suggested that the RNA-Seq results are reliable. DEGs were compared with the KEGG Orthology database, and the corresponding pathways were established (Table S4). A total of 37 genes (72.5%) in the KEGG pathway database were related to basal metabolism, indicating that the effect of Si on R. solanacearum was primarily focused on the basal metabolic process ( Figure S2). In addition, 15.68% of the genes were related to the development of environmental information. Si significantly upregulated the expression of the R. solanacearum DNA repair gene (ogt) and protein repair gene (degP , Table S4). Meanwhile, the KEGG BRITE functional annotation showed that 41.81% of genes were involved in enzyme functions, and 23.63% of genes were transporters (Table S5).

Silicon Did Not Affect the Growth of R. solanacearum
The results show there was no significant difference in the growth of R. sola among all treatments ( Figure 2). The growth curve of R. solanacearum in the liqui Bertani (LB) and minimal mineral (MM) medium was not significantly influenc treatment with different concentrations, indicating that Si did not have a direct in on R. solanacearum.

Silicon Did Not Affect the Growth of R. solanacearum
The results show there was no significant difference in the growth of R. solanacearum among all treatments ( Figure 2). The growth curve of R. solanacearum in the liquid Luria-Bertani (LB) and minimal mineral (MM) medium was not significantly influenced by Si treatment with different concentrations, indicating that Si did not have a direct inhibition on R. solanacearum.

Silicon Regulated the Expression of 119 Genes of R. solanacearum In Vitro
The sequencing data were mapped onto the R. solanacearum strain GMI1000 reference genome [32], and 4,496 genes were detected ( Figure 3). In this study, 119 differentially-expressed genes (DEG) were found to be significantly influenced by Si application, accounting for 2.6% of all known genes. Among these 119 genes, 57 were upregulated, and 62 were downregulated (Tables S1 and S2).  The expression of 15 genes (gene primer was shown in Table S3) related to the virulence of R. solanacearum was analyzed by qRT-PCR, including four biological replicates under two housekeeping genes (GAPDH and thyA). The qRT-PCR data for these genes were significantly correlated with the RNA-Seq results in two housekeeping genes, namely GAPDH ( Figure S1a) and thyA ( Figure S1b), which suggested that the RNA-Seq results are reliable. DEGs were compared with the KEGG Orthology database, and the corresponding pathways were established (Table S4). A total of 37 genes (72.5%) in the KEGG pathway database were related to basal metabolism, indicating that the effect of Si on R. solanacearum was primarily focused on the basal metabolic process ( Figure S2). In addition, 15.68% of the genes were related to the development of environmental information. Si significantly upregulated the expression of the R. solanacearum DNA repair gene (ogt) and protein repair gene (degP , Table S4). Meanwhile, the KEGG BRITE functional annotation showed that 41.81% of genes were involved in enzyme functions, and 23.63% of genes were transporters (Table S5).

Silicon Altered the Transcriptional Expression of xpsR, EPS, and T3SS in R. solanacearum
DEGs involved in metabolic pathways were analyzed to decipher the molecular mechanism of silicon in influencing R. solanacearum virulence. The ModA, AraH, XylF, and LivK genes involved in the ABC transporter pathway and EPS synthesis were significantly downregulated by Si (Figure 4a). These genes encoded proteins that mediated the transportation of bacterial molybdate, L-arabinose, D-xylose, and branched-chain amino acid. Moreover, IbpA, which was involved in Myo-Inositol transport, was significantly upregulated by Si. In the two-component system pathway, four DEGs (degP, DctA, xpsR, and epsD) were enriched ( Figure 4b). degP was remarkably upregulated by Si treatment. However, DctA, xpsR, and epsD were significantly downregulated by Si application. The pathway analysis results show that the FliN gene was related to bacterial chemotaxis and flagellar assembly (Figure 4c).

Silicon Altered the Transcriptional Expression of xpsR, EPS, and T3SS in R. solanacearum
DEGs involved in metabolic pathways were analyzed to decipher the molecular mechanism of silicon in influencing R. solanacearum virulence. The ModA, AraH, XylF, and LivK genes involved in the ABC transporter pathway and EPS synthesis were significantly downregulated by Si (Figure 4a). These genes encoded proteins that mediated the transportation of bacterial molybdate, L-arabinose, D-xylose, and branched-chain amino acid. Moreover, IbpA, which was involved in Myo-Inositol transport, was significantly upregulated by Si. In the two-component system pathway, four DEGs (degP, DctA, xpsR, and epsD) were enriched ( Figure 4b). degP was remarkably upregulated by Si treatment. However, DctA, xpsR, and epsD were significantly downregulated by Si application. The pathway analysis results show that the FliN gene was related to bacterial chemotaxis and flagellar assembly ( Figure 4c).
The flagellar assembly pathway is a key pathway downstream of the bacterial chemotaxis. Four genes (pilE2, pilE, PilX, and fimT) encoding type-4 fimbriae were downregulated by Si (Table S6). Twelve virulence-related genes were screened, among which four were upregulated, and eight were downregulated (Table S6). Some genes encoding known virulence traits were screened under Si treatment, such as transcription regulator (xpsR), EPS-related genes (epsD and tek), and type III effectors (HrpB2, SpaO, and EscR). The qPCR results show that the expression of the vital virulence genes, namely xpsR (2.86 folds) and epsD (3.33 folds), were significantly downregulated by Si application (Figure 5a,b). The qPCR results also show that the twitching motility-related gene (pilE) was remarkably upregulated by Si treatment (Figure 5c). Furthermore, Si application significantly reduced the expression of LivK (Figure 4d). The flagellar assembly pathway is a key pathway downstream of the bacterial chemotaxis. Four genes (pilE2, pilE, PilX, and fimT) encoding type-4 fimbriae were downregulated by Si (Table S6). Twelve virulence-related genes were screened, among which four were upregulated, and eight were downregulated (Table S6). Some genes encoding known virulence traits were screened under Si treatment, such as transcription regulator (xpsR), EPS-related genes (epsD and tek), and type III effectors (HrpB2, SpaO, and EscR). The qPCR results show that the expression of the vital virulence genes, namely xpsR (2.86 folds) and epsD (3.33 folds), were significantly downregulated by Si application (Figure 5a,b). The qPCR results also show that the twitching motility-related gene (pilE) was remarkably upregulated by Si treatment (Figure 5c). Furthermore, Si application significantly reduced the expression of LivK (Figure 4d).

Silicon Disrupted EPS and Biofilm Formation
The epsD gene, which was involved in the synthesis of EPS (Table S6), was significantly downregulated by Si application. By contrast, EPS concentration was markedly reduced in 2 mM Si treatment, which was consistent with the transcriptome results (Figure 6a). The biofilm formation was an essential mechanism for bacteria to resist external environmental stress. Our results show that bacterial biofilm synthesis in Si treatment was significantly lower than that of non-Si treatment (Figure 6b).

Silicon Disrupted EPS and Biofilm Formation
The epsD gene, which was involved in the synthesis of EPS (Table S6), was significantly downregulated by Si application. By contrast, EPS concentration was markedly reduced in 2 mM Si treatment, which was consistent with the transcriptome results (Figure 6a). The biofilm formation was an essential mechanism for bacteria to resist external environmental stress. Our results show that bacterial biofilm synthesis in Si treatment was significantly lower than that of non-Si treatment (Figure 6b). Figure 5. Effects of exogenous silicon on the relative expression of virulence-related genes of R. solanacearum. (a) xpsR (transcriptional regulator). (b) epsD (the genes coding for EPS). (c) pilE (the genes coding for twitching motility). (d) fimT (the genes coding for swimming motility). CK: non-Si treatment; Si: Si treatment. "ns" indicates not significant (p > 0.05), * and ** indicate significant difference among treatments at p ≤ 0.05 and p ≤ 0.01 in t-test.

Silicon Disrupted EPS and Biofilm Formation
The epsD gene, which was involved in the synthesis of EPS (Table S6), was significantly downregulated by Si application. By contrast, EPS concentration was markedly reduced in 2 mM Si treatment, which was consistent with the transcriptome results (Figure 6a). The biofilm formation was an essential mechanism for bacteria to resist external environmental stress. Our results show that bacterial biofilm synthesis in Si treatment was significantly lower than that of non-Si treatment (Figure 6b).

Discussion
Multiple studies have investigated the role of Si in controlling bacterial wilt [18,33,34]. Our results show that 0.5-2.0 mM of Si significantly reduced the disease index of bacterial wilt by 46.31%-72.23% (Figure 1a,b), which is consistent with previous studies. However, our study found that Si did not significantly inhibit the growth of R. solanacearum (Figure 2a,b). Previous studies demonstrated that exogenous additives (such as benzimidazole and vitamin C) could prevent pathogens from infecting the host not by

Discussion
Multiple studies have investigated the role of Si in controlling bacterial wilt [18,33,34]. Our results show that 0.5-2.0 mM of Si significantly reduced the disease index of bacterial wilt by 46.31%-72.23% (Figure 1a,b), which is consistent with previous studies. However, our study found that Si did not significantly inhibit the growth of R. solanacearum (Figure 2a,b). Previous studies demonstrated that exogenous additives (such as benzimidazole and vitamin C) could prevent pathogens from infecting the host not by directly suppressing pathogens, but by reducing the bacterial virulence, which weakens the ability of bacteria to infect the host [35,36].
Our study showed that Si could significantly downregulate EPS synthesis-related genes, namely, epsD and tek (Table S6, Figure 5b). EPS is also a key virulence factor of R. solanacearum and part of the exopolysaccharide operon, which is the primary substance used by R. solanacearum to block the xylem of plants, and the mutants of the EPS gene lose their virulence ability [37,38]. Minic et al. (2007) reported that epsD affected EPS biosynthesis in Streptococcus thermophilus, and epsD mutant did not produce EPS [39]. tek is an extracellular protein associated with EPS1 and regulated by PhcA [40].
Our studies showed that EPS production decreased in Si treatment (Figure 6a). Therefore, we suggest that Si inhibited the expression of espD and tek, thereby reducing R. solanacearum EPS synthesis. Notably, other genes, except for epsD in the eps gene cluster of R. solanacearum, are not significantly expressed in our experiment. Different genes in the same gene cluster have a different substrate specificity, so there will be differences in the expression levels in the same environmental conditions. In order to uncover the exact reason why silicon inhibits EPS synthesis of R. solanacearum, more molecular experiments need to be carried out in future research.
Biofilms are microbial defense and communication systems [41]. The results show that the formation of R. solanacearum biofilm was significantly inhibited by Si treatment (Figure 6b). Kong et al. (2018) demonstrated that a benzimidazole derivative (UM-C162) prevented the formation of Staphylococcus aureus biofilm, but this derivative had no effect on bacterial viability, and the transcriptome analysis showed that UM-C162 treatment inhibited the expression of bacterial biofilm formation and bacterial attachment-associated genes [42]. Hence, Si inhibited R. solanacearum biofilm synthesis, which might weaken bacterial virulence.
In our study, the transcriptome results show that Si treatment significantly downregulated the expression of eight genes (HrpB2, SpaO, EscR, xpsR, tek, epsD, RSc2755, and RSp1004), which were primarily related to bacterial virulence regulation (Table S6). HrpB2, SpaO, and EscR were type III effectors' proteins. T3SS were used to inject effectors proteins into plant cells and delivered collections of type III effectors proteins to weaken host defenses [43]. The HrpB2 gene belongs to the hrp gene cluster, and it is an essential component of T3SS [44]. The expression of HrpB2, SpaO, and EscR genes were downregulated by Si treatment, indicating that T3SS of R. solanacearum was inhibited, which may lead to a reduction in bacterial virulence. The xpsR gene is controlled by PhcA, which regulates EPS production [45].
Considerable evidence suggested that the LysR-type regulatory factor PhcA gene was the core of the virulence regulatory network of R. solanacearum [33,46,47]. PhcA is a LysRtype transcriptional regulator, which can regulate virulence factors, activate EPSs and cellulase, and inhibit the mobility of R. solanacearum [48].  found that the expression of R. solanacearum virulence-related genes (xpsR, tek, and epsE) was significantly inhibited in the PhcA mutant. Collectively, Si suppressed the expression of virulence-related genes, which might decrease the infection rate of R. solanacearum in tomato [45].
Twitching motility is an important bacterial behavior that allows pathogens to efficiently migrate and colonize host plants. Our study also found that the type-4 fimbriae genes, pilE2, pilE, PilX, and fimT, were significantly upregulated by Si (Table S6). The swimming motility of R. solanacearum can be affected by flagella, which are related to bacterial virulence [49]. The flagellar motor switch is composed of three proteins: FliG, FliM, and FliN, which can control flagellum's rotation [50]. Our study found that Si inhibited the expression of CheZ and FliN (Figure 4c), indicating that Si enhanced the movement of fueling flagella and promoted the twitching motility of bacteria. Twitching motility is important to pathogens in avoiding stressful environments.

Experimental Materials
The R. solanacearum strains (GMI1000) and tomato seeds (genotype HYT, bacterial wiltsusceptible tomato) were provided by Professor Guoping Wang (College of Horticulture, South China Agricultural University, Guangzhou, China). Si was applied in the form of anhydrous potassium silicate (K 2 SiO 3 , Thermo Fisher Scientific, mean weight ratio: SiO 2 :K 2 O = 2.5).

Effects of Different Si Concentrations on Wilt Incidence
Tomato seeds were surface sterilized with 10% H 2 O 2 for 10 min, followed by rinsing three times with ultrapure water. Sterilized seeds were germinated in petri dishes containing two layers of filter paper for 48 h at 30 • C. The germinated tomato seeds were transferred to plug trays containing sterilized peat soil (Klasmann, Germany), grown in an incubator (30 •  Tomato seedlings at the third-leaf stage were transplanted into the pot (5 cm × 5 cm × 8 cm) containing 1 kg of sterilized peat soil. There are six treatments, including the 0, 0.05, 0.1, 0.5, 1.0, and 2.0 mM Si, with twelve replications in this experiment. Tomato seedlings were irrigated daily with 50 mL of 1/2 concentration tomato special nutrient solution (containing 0, 0.05, 0.1, 0.5, 1.0, and 2.0 mM Si) in each pot. The non-silicon treatment was supplemented with a corresponding amount of KCl to balance the increased K element in the Si nutrient solution.
Tomato plants at the sixth-leaf stage were inoculated with pathogens. R. solanacearum strains were grown on B medium (Difco Peptone 10 g·L −1 , yeast extract 1 g·L −1 , and casamino acids 1 g·L −1 ) at 28 • C and 150 r·min −1 [51]. After 48 h, bacterial cells were collected and centrifuged three times with sterilized water. R. solanacearum suspension was resuspended in deionized water and adjusted to OD600 = 0.1 (10 8 cfu·mL −1 ) with a spectrophotometer (Shimadzu UV-2600, Kyoto, Japan). Then, 50 mL of the bacterial suspension was poured into each pot. The disease index of tomato plants was recorded from the wilting of the first leaf and continuously recorded for 20 d.

Effects of Different Silicon Concentrations on the Growth of R. solanacearum
The growth curves of R. solanacearum in LB and MM medium were investigated, according to a modification of the method described by Lowe-Power et al. (2018) [52]. There were six treatments: 0, 0.05, 0.1, 0.5, 1.0, and 2.0 mM Si. R. solanacearum strains were grown on liquid LB medium at 28 • C and 150 r·min −1 for 24 h. The cultured cells were adjusted to OD600 = 0.1 using sterile water, and 200 µL of bacterial suspension was added to 30 mL of liquid LB and MM medium. The medium was incubated at 28 • C with shaking at 150 rpm for 72 h in an incubator-shaker (HZQ-X300, Yiheng-Shanghai, China). Growth rate was assessed by measuring OD600 values every 12 h. All experiments have four replicates.

Pathogen Symptom Evaluation
The disease index was determined according to the methods described by   [33]. A disease index was recorded from the first wilting leaf of tomato plants. Grade 0: asymptomatic; Grade 1: 1 leaf half wilting; Grade 3: 2 to 3 leaves wilting; Grade 5: all leaves are wilted except for the top 1 to 2 leaves; Grade 7: all leaves are wilting; Grade 9: the leaves and plants die.

EPS Assay
EPS was extracted from R. solanacearum strain [53]. R. solanacearum strain was grown at 30 • C for 48 h in LB medium. Five milliliters of overnight cultures of R. solanacearum were adjusted to OD600 = 0.1, then R. solanacearum was added to 95 mL of sterile liquid MM medium (with and without Si medium) and cultured with shaking for 48 h at 28 • C. R. solanacearum cell suspension (10 8 CFU·mL −1 ) was centrifuged at 12,000 rpm for 10 min. The supernatant was filtered by a 0.22 µL filter membrane. The filtrate was placed in a lyophilizer overnight to freeze dry, then 1 mL of 95% ethanol was added, and the mixture was kept at 4 • C for 24 h. Then, the mixture was centrifuged at 5000 r·min −1 for 10 min to collect the precipitate, and 1 mL of water was heated to dissolve the residue and obtain a crude EPS solution. The purified EPS was diluted to 25 mL with distilled water and frozen to store for later use. EPS was determined using the Elson-Morgan method [54].

Biofilm Assay
The biofilms of R. solanacearum were measured in vitro with a minor modification of the polyvinyl chloride microtiter plate assay [55]. Briefly, R. solanacearum strain was grown at 30 • C for 48 h in LB medium. Then, 5 mL of R. solanacearum suspension (OD600 = 1.0) was added to 45 mL of MM liquid culture broth (with and without Si treatment). Two hundred microliters of culture solution were added to 96-well polystyrene microtiter plates. After the 96-well plates with culture solution were incubated at 28 • C for 48 h, the culture medium was carefully removed, and the biofilm was washed twice with 200 µL ultrapure water. The 96-well plates with bacterial membrane were dried at 60 • C for 30 min to fix the bacterial membrane. Then, 200 µL of 0.1% crystal violet was added to stain the biofilm for 30 min, and the culture solution was washed two times with 200 µL of distilled water to move the crystal violet. Then, 95% ethanol was used to adsorb the crystal violet from the biofilm. The solution was measured for absorbance at 530 nm.

Transcriptome Analysis
R. solanacearum strain was grown at 30 • C for 48 h in LB medium. Five milliliters of R. solanacearum suspension were adjusted to OD600 = 0.1 and were added into an Erlenmeyer flask containing 50 mL of MM medium with or without 2 mM of Si. Potassium chloride was used to adjust potassium differences between controls and treatments. There were six samples for each treatment. The medium was incubated at 28 • C with shaking at 150 rpm for 48 h, and bacterial cells were collected for transcriptome analysis.
RNA was extracted using the Bacterial RNA Kit (Tiangen Biotechnology, Beijing, China) according to the manufacturer's recommendations. Preliminary quantification and accurate quantification were detected using a NanoDrop 2000 spectrophotometer (Thermo Fisher Scientific, Inc., Wilmington, DE, USA) and electrophoresis concentration Agilent 2100 RNA 6000 Nano Kit ( Figure S3), respectively. Twenty microliters of RNA per treatment were taken for cDNA library construction. The construction of cDNA libraries and RNA-Seq was performed by Genedenovo Bio-Tech Co., Ltd. (Guangzhou, China).

Data Quality Check and Analysis
The high-throughput sequencing and preliminary analysis of the data were completed by Guangzhou Gideo Biotechnology Co., Ltd. The raw data obtained from Illumina HiSeqTM 2500 sequencing were converted into sequenced reads by CASAVA base calling and stored in FASTQ file format. Raw sequences with adaptors and unknown nucleotides above 5% or those that were of low quality were removed to obtain clean reads. The filtered reads of ribosomes were compared with the reference genome by TopHat2 version 2.0.14 [56]. Cufflinks version 2.2.1 [57] was used to assemble transcripts based on reference annotation-based transcripts.
Fragments per kilobase million (FPKM) value estimation was used to measure gene expression. RSEM version 1.3.3 [58] was used to count the bowtie's comparison results. The DEGseq version 1.18.0 [57] was applied to normalize the FPKM values among samples for statistical analysis. DEGs were screened based on the criteria of |log2 Ratio| ≥ 1 and Q ≤ 0.05. p-value was determined by controlling the false discovery rate. DEGs were also submitted to the GO program (http://www.blast2go.org (accessed on 22 June 2022)) for functional annotation [59]. We used KEGG (http://www.kegg.jp/ (accessed on 22 June 2022) pathway analysis to compare and enrich differential genes and determine the main biochemical metabolic pathways and signal transduction processes.

qRT-PCR
qRT-PCR was performed to validate the RNA-Seq results for 15 gene transcripts. The qPCR experiment used the SYBR Premix Ex Taq Kit (Takara, Japan), and qPCR was measured by the ABI Step One Plus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). The reaction system (total volume, 20 µL) included 10 µL of SYBR Premix Ex Taq II, 0.6 µL of PCR Forward Primer (10 µM), 0.6 µL of PCR Reversed Primer (10 µM), 2.0 µL of cDNA, and 2.8 µL of ddH 2 O. The amplification cycling program was as follows: 90 s at 95 • C, followed by 40 cycles at 95 • C for 5 s, 60 • C for 15 s, and 72 • C for 20 s. Relative quantification was used for the conversion of gene expression. Data were analyzed using the 2 −∆∆Ct method [60]. Two internal reference genes (GAPDH and thyA) were selected, and each sample was subjected to four technical replicates. The primers used for qPCR are listed in Table S3.

Statistical Analysis
All data were statistically analyzed by t-test and Duncan's method for multiple comparisons at a 5% significance level. All the statistical analyses were performed with SPSS 20.0 (IBM, Chicago, IL, USA).

Conclusions
In this study, we found that Si significantly suppressed bacterial wilt not by directly inhibiting the growth of R. solanacearum, but by influencing the expression of virulence-related genes. Si also reduced the synthesis of R. solanacearum EPS and biofilm. On the contrary, Si upregulated R. solanacearum flagellar genes that promoted bacterial twitching motility. Hence, we hypothesized that Si had a negative influence on R. solanacearum. In the presence of Si treatment, Si downregulated virulence-related genes (HrpB2, SpaO, EscR, xpsR, tek, and epsD) of R. solanacearum and inhibited R. solanacearum EPS synthesis. Meanwhile, Si promoted R. solanacearum movement-related genes (pilE2, pilE, PilX, and fimT). However, the specific mechanism by which silicon inhibits the virulence factors of R. solanacearum remains unclear. Further research needs to be conducted with R. solanacearum mutants or plant infection experiments. This study provides a new perspective to decipher the role of silicon in controlling plant diseases.

Data Availability Statement:
The data presented in this study are available on request from the corresponding author.

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