FAN Xue-ying , LlN Wei-peng , LlU Rui JlANG Ni-hao CAl Kun-zheng

1 College of Natural Resources and Environment/Key Laboratory of Tropical Agro-environment, Ministry of Agriculture, South China Agricultural University, Guangzhou 510642, P.R.China

2 Laboratory of Ecotoxicity and Environmental Safety, Guangdong Detection Center of Microbiology, Guangdong Institute of Microbiology, Guangzhou 510070, P.R.China

3 Tea Research Institute, Guangdong Academy of Agricultural Sciences, Guangzhou 510640, P.R.China

Abstract Bacterial wilt, caused by Ralstonia solanacearum (Rs) is a serious soil-borne disease and silicon can enhance tomato resistance against this disease. However, few studies have focused on the mechanisms of Si-mediated pathogen resistance from the rhizosphere perspective. In this study, two tomato genotypes, HYT (susceptible) and H7996 (resistant), were used to investigate the effects of silicon application on disease inhibition, root growth, and organic acid content in both roots and root exudates under R. solanacearum infection. The results showed that Si application significantly suppressed bacterial wilt in HYT, but had no effect in H7996. Silicon concentrations in roots, stems and leaves of tomato were significantly increased by Si treatment under R. solanacearum inoculation. In HYT, Si application increased root dry weight by 22.8–51.6% and leaf photosynthesis by 30.6–208.0%, and reduced the concentrations of citric acid in root exudates by 71.4% and in roots by 83.5%. However, organic acids did not influence R. solanacearum growth. Results also demonstrated that salicylic acid(SA) content in roots was significantly increased by silicon addition for H7996 and exogenous SA application could reduce bacterial wilt disease index. Collectively, these results suggest that Si-modulated phenolic compound metabolism in roots or root exudates, especially citric acid and SA, may be a potential mechanism in the amelioration of bacterial wilt disease by Si.

Keywords: silicon, root, root exudates, Solanum lycopersicum, Ralstonia solanacearum, organic acids

1. lntroduction

As the second most abundant element in soil, silicon is known to be a beneficial element which can enhance plant resistance against various abiotic (drought, salinity, metal toxicity, etc.) and biotic stresses (pathogens, insects, etc.) in many plant species (Fauteux et al. 2005; Liang et al. 2007;Debona et al. 2017). Numerous studies have found that silicon played an important role in suppressing plant disease such as powdery mildew, blast, sheath blight, anthracnose and bacterial wilt (Polanco et al. 2012; Zhang et al. 2013;Kurabachew and Wydra 2014; Liu et al. 2014; Tesfagiorgis et al. 2014). Van Bockhaven et al. (2013) summarized the role of silicon and plant activators in controlling different rice diseases and revealed that silicon can impact broadspectrum disease resistance, while other plant hormones like salicylic acid (SA), jasmonic caid (JA), ethylene (ET)do not. Many studies have shown that silicon induces plant resistance against pathogen via either physical barriers or by influencing defense-related enzyme activities,secondary metabolite secretion, cell structure, or protein and gene expression (Liang et al. 2005; Garibaldi et al.2011; Fortunato et al. 2012; He et al. 2013; Resende et al.2013; Sahebi et al. 2014; Tesfagiorgis et al. 2014). Silicon application could thicken cell wall or form silicified cells with plants which is known as physical barrier to prevent pathogen invasion (Kim et al. 2002; Cai et al. 2008; Sun et al. 2010; Zhang et al. 2013). Upon pathogen attack,silicon can increase the activities of phenolic metabolismrelated enzymes such as superoxide dismutase (SOD),peroxidase (POD), catalase (CAT), polyphenoloxidase(PPO), and phenylalanine ammonialyase (PAL) in plants(Cai et al. 2008; Silva et al. 2010; Fortunato et al. 2012;Resende et al. 2013). Moreover, the concentrations of phenolics in plant are increased after silicon addition.Studies in chili pepper found that plants augment phenolics following silicon treatments under anthracnose stress(Jayawardana et al. 2015). Under leaf scald stress, the phenylpropanoid pathway is involved in the mechanism of Si-enhanced rice resistance to leaf scald development(Araujo et al. 2016). In addition, Si can enhance phenolic metabolism and increase rice resistance against sheath blight (Zhang et al. 2013). The study of rose powdery mildew and wheat leaf blast resistance mediated by silicon also revealed similar results (Shetty et al. 2012; Silva et al.2015). Furthermore, silicon-mediated pathogen resistance in plants operates not only at the physiological level but also at the molecular level. Brunings et al. (2009) found that Si application altered differential expression of 54 unique genes under Magnaporthe oryzae-infection. Upon Ralstonia solanacearum infection, Si up-regulated 12 genes of tomato plants, including genes involved in defense and responses to stresses, which implies that the priming effect of Si may involve ET, JA and/or other signaling pathways (Ghareeb et al. 2011). Our recent studies found that silicon could influence the expression of energy metabolism-related proteins when Si was added to either M. oryzae-inoculated rice plants or R. solanacearum-inoculated tomato plants(Liu et al. 2014; Chen et al. 2015).

Root exudates, including sugar, organic acids, amino acids and soluble protein, influence plant growth and enhance plant resistance under biotic and abiotic stresses through adjusting soil physical and chemical properties and microbial community structure (Huang et al. 2014).Some reports have found that malic acid can activate zinc and phosphorus nutrients in soil, which is beneficial to wheat and Lupinus angustifolius growth (Maqsood et al.2011; Le Roux et al. 2014). Organic acids secretion can also alleviate the toxicity of heavy metals to plants (Ehsan et al. 2014; Sun et al. 2014; Amari et al. 2016; Huang et al.2016). Studies have found that root exudates can stimulate plant growth by promoting rhizobacteria colonization(Akiyama et al. 2005; Zhang et al. 2015). Moreover, the chemicals secreted by roots play positive roles in improving rhizosphere microbial community structure through either beneficial bacteria biofilm formation or chemotaxis (Tan et al. 2013). Rudrappa et al. (2008) found that root-secreted malic acid in Arabidopsis promotes FB17 biofilm formation,which can reduce the possibility of pathogen attack and increase plant defense. Meanwhile, malic acid and fumaric acid in banana root exudates increased beneficial bacteria NJN-6 colonization which may protect banana from soilborne pathogens (Yuan et al. 2015). However, evidence has showed that root exudates may increase the growth of some pathogens, such as the microconidial germination of Fusarium oxysporum (Steinkellner et al. 2008) or the colonization of R. solanacearum (Wu et al. 2015).

Tomato is known to be a non-Si accumulated plant,although it has Lsi-2 like genes (Ma and Yamaji 2015).Evidence has shown that silicon addition could significantly reduce the incidence of bacterial wilt for both susceptible and resistant tomato genotypes (Dannon and Wydra 2004).Meanwhile, silicon also could enhance the resistance of tomato against Fusarium crown rot, root rot and powdery mildew (Garibaldi et al. 2011; Huang et al. 2011). Diogo and Wydra (2007) demonstrated that Si-enhanced resistance may involve cell wall structure changes, which are considered as physical defenses as mentioned above,but the ability of Si to inhibit Pythium aphanidermatum may depend on symplastic Si (Heine et al. 2007). On the other hand, the priming effects of Si in enhancing tomato plants resistance also appear to involve either increasing defenserelated enzyme activities or up-regulating defense-related gene expression (Ghareeb et al. 2011; Kurabachew et al.2013; Wang et al. 2013).

The objective of this study is to explore the roles of organic acid of roots or root exudates in Si-mediated bacterial wilt resistance of two tomato genotypes. Disease index, root morphological traits, photosynthetic parameters, organic acid concentrations in roots and root exudates and organic acid-metabolizing enzymes activities were measured under different combinations of R. solanacearum inoculation and Si treatments. In addition, the effects of root exudates on tomato plant growth and disease resistance were alsostudied, which may help further clarify the mechanisms of Si-mediated resistance to pathogens.

2. Materials and methods

2.1. Plant materials and culture conditions

Two tomato genotypes, HYT (susceptible) and H7996(resistant) were used in this study. Tomato seeds were soaked in distilled water for 10 h and then sterilized in distilled water at (50±2)°C for 30 min. Subsequently,seeds were germinated in the dark for (48±6) h at 25°C.Seedlings were transferred to a pot (20 mm diameter and 50 mm height) with substrate (Klasmann, German) in a growth chamber, and watered daily with distilled water. The distilled water was replaced by the solution described by Dannon and Wydra (2004), first by half-concentration after one week, then full concentration one week later. After two weeks, three seedlings were transferred to a plastic pot with 600 mL full-strength solution. The experiment was performed in a growth chamber under controlled conditions:12 h light/ 12 h darkness, 26°C/23°C day/night temperature,80%/60% day/night relative humidity, and a photon flux density of 400 μmol m–2s–1.

2.2. lnoculation of bacteria

R. solanacearum strain (physiological race 1, biovar 3) from College of Horticulture, South China Agricultural University,was used for inoculation to the seedlings. Bacteria were grown on TTC medium for 48 h at 30°C. A total of 10 mL suspension (OD600nm=0.06) was poured in each plastic pot, giving a final R. solanacearum concentration of approximately 106cfu mL–1.

2.3. Experimental design

Experiment 1Si and R. solanacearum (Rs) treatments were conducted one week after seedlings transplanting. Four treatments with six replications each were prepared for both tomato genotypes: (1) CK (no silicon and no Rs inoculation); (2)Si (2.0 mmol L–1silicon application); (3) Rs (R. solanacearum inoculation); (4) Rs+Si (R. solanacearum inoculation and 2.0 mmol L–1silicon application). Silicon (2.0 mmol L–1)was added as potassium silicate (SiO2:K2O=2.5:1 wt%, Alfa Aesar, China) solution and adjusted pH to 5.5 with HCl. In the silicon-deficient treatment, potassium chloride (KCl, pH 5.5) was used to equal the potassium component of the Si treatment. Treatments for each genotype were arranged in a randomized complete block design. Roots samples were collected 7 days after treatments to determine organic acid concentration in roots or root exudates, root disease index, and root morphological traits. Plant biomass, leaf photosynthesis, silicon concentration and activity of organic acid-metabolizing enzyme were also measured at the end of this experiment.

Experiment 2Organic acids addition experiment: The objective of this experiment was to investigate the effects of selected organic acids on pathogen control and plant development parameters according to the results of experiment 1. This experiment was also conducted one week after seedlings transplanting as described above.Five treatments with five replications were prepared for both tomato genotypes: (1) CK; (2) Rs; (3) Rs+Citric acid;(4) Rs+Malic acid; (5) Rs+Salicylic acid. For each organic acid treatment, 10 mL of 5 μg mL–1solution was added into the pot and Si treatment was applied as described above for experiment 1. Tomato roots were collected 10 days after different treatments to determine organic acid concentrations, plant biomass and leaf photosynthesis.

2.4. Pathogen evaluation

Pathogen evaluation began on the first day after inoculation and using disease severity classes described by Dannon and Wydra (2004). The disease index was calculated as described by Fang et al. (1998).

2.5. Measurement of root morphological traits

Root samples were collected, washed, and then scanned to measure total root length, surface area, average diameter and volume by using WinRhizo Arabidopsis V2009c (Regent Instrument Inc., Chemin Sainte-Foy, Canada).

2.6. Determination of biomass and photosynthesis

Tomato plants were harvested and leaves, shoots and roots were then washed with distilled water twice and dried with filter paper. Subsequently, the samples were dried in an air-forced oven at 120°C at 30 min and at 80°C until a constant mass was reached.

For each treatment, net photosynthetic rate (Pn),intercellular carbon dioxide concentration (Ci), stomatic conductance (Gs) and transpiration rate (Tr) of fullyexpanded tomato leaves were measured by LI-6400XT Photosynthetic System (LI-COR, USA) at a photon flux density of 1 000 μmol m–2s–1and 500 μmol s–1air velocity.

2.7. Determination of silicon concentration

Tomato root, stem and leave samples for each treatments were collected and dried in an air-forced oven at 120°C for 30 min and 80°C for 2 days. Then, samples were ground into powder and passed through a 100-mesh sieve. Silicon concentration was measured according to the methods described by Dannon and Wydra (2004).

2.8. Determination of organic acid in root and root exudates

Collection of root exudates for organic acid analysisTomato seedlings for root exudates extraction were harvested and washed with distilled water two or three times. Then root exudates were collected in a plastic cup containing 50 mL distilled water for 5 h, which was covered with black paper.The collection was performed in a growth chamber under controlled conditions: 26°C temperature, 80% relative humidity, and a photon flux density of 400 μmol m–2s–1. After 5 h, roots were washed 5 times with distilled water, and the root exudates and washing fluid were combined to give a final total volume of 100 mL. Then 20 mL of the collection solution was concentrated to 1 mL in a 75°C water bath.The samples were filtered using nylon filters with 0.22-μm pore size. After filtration, the exudates were stored at –20°C for further analyses.

Root sample preparation for organic acid analysisTomato seedlings for root organic acid extraction were harvested and washed with distilled water two or three times. Then roots were dried with filter paper and 0.1 g of root tip was ground with liquid nitrogen. Samples were extracted with 1 mL distilled water and then centrifuged at 8 000 r min–1for 10 min. The supernatant was filtered using nylon filters with 0.22-μm pore size and stored at –20°C for further analyses.

Methods of organic acid analysisCitric acid, malic acid, fumaric acid, tartaric acid, succinic acid, and salicylic acid (SA) were isolated and identified from roots and root exudates using UPLC-MS/MS (zevo-TQD, Waters, USA).The analytical conditions of the first five organic acids (citric acid, malic acid, fumaric acid, tartaric acid and succinic acid)were as follows: chromatographic column: acquity UPLC BEH C18 (1.7 μm, 2.1 mm×50 mm); velocity flow: 0.3 mL min–1; temperature of column: 30°C; injection volume: 10 μL(root exudates), 2 μL (root); mobile phase: (A) 0.1% formic acid, and (B) acetonitrile with 0.1% formic acid, 0 min, 100%A; 2.5 min, 87.5% A+12.5% B; 3–4 min, 50% A+50% B;5–6 min, 100% A. Mass spectra were as follows: ESI–,citric acid (191.1/111.0, 87.0), malic acid (133.0/115.0, 70.9),tartaric acid (149.0/86.9, 103.0), succinic acid (117.0/73.0),and fumaric acid (115.0/70.9).

Methods of salicylic acid analysisThe analytical conditions of salicylic acid were as follows: chromatographic column: acquity UPLC BEH C18 (1.7 μm, 2.1 mm×50 mm);velocity flow: 0.2 mL min–1; temperature of column: 35°C;injection volume: 10 μL (root exudates), 2 μL (root); mobile phase: (A) 0.1% formic acid, and (B) methanol with 0.1%formic acid, 0 min, 95% A+5% B; 1 min, 90% A+10% B;1.5–4.5 min, 30% A+70% B; 5–6 min, 95% A+5% B. Mass spectrum was as follows: ESI+, salicylic acid (138.9/65.0,93.0).

2.9. Determination of organic acid-metabolizing enzyme activity in roots

Tomato seedlings were harvested and washed with distilled water two or three times. Then roots were dried with filter paper and 0.1 g root tip was ground with 900 μL phosphate buffer (pH 7.2). The samples were centrifuged at 8 000 r min–1for 10 min. Then the activities of citrate synthase, malate synthase and fumarates in the supernatant were detected by an ELISA Kit (Jianglai Biotechnology Co., Ltd., Shanghai,China).

2.10. Statistical analysis

All data were expressed as the mean±SE. One-way ANOVA was performed to test the significance of the observed differences using SPSS 20.0. Differences between parameters were evaluated using the Duncan’s method, and P≤0.05 was considered the statistically significant threshold.

3. Results

3.1. Disease development

In experiment 1, the incidence of bacterial wilt occurred 4 days after inoculation for HYT and 10 days for H7996, which illustrated the differential resistance to R. solanacearum for the two genotypes (Fig. 1). Si application had a significant role in suppressing bacterial wilt for susceptible genotype HYT, and the disease index was reduced by 25.0%.However, the disease index for resistance genotype H7996 was not influenced by Si (Fig. 1). For experiment 2, citric acid or malic acid addition did not influence R. solanacearum development, while SA significantly reduced disease index by 29.7% for HYT (Fig. 2), and organic acid or SA applications did not have any effects on pathogen control for H7996.

3.2. Silicon absorption

Silicon application significantly promotes tomato plant Si absorption regardless of R. solanacearum inoculation(Fig. 3). Si concentration was increased in Si-treated plants by 6–9 folds in roots, 1.8–2.5 folds in stems and 2.5–3.3 folds in leaves for susceptible genotype HYT. For resistant H7996, Si concentration was increased by 4–19 folds in roots, 2.1–2.4 folds in stems and 2–12 folds in leaves.Clearly, silicon accumulated to a higher degree in roots than in stems and leaves.

Fig. 1 Effects of 2 mmol L–1 Si treatment on tomato bacterial wilt resistance. Rs, Ralstonia solanacearum inoculation; Rs+Si,R. solanacearum inoculation and 2.0 mmol L–1 silicon application.

3.3. Plant growth and photosynthesis

For HYT, root surface area, average diameter and root volume were increased by 52.0, 23.0 and 47.0%,respectively after Si addition (Rs+Si) compared with Rs treatment (Table 1). However, silicon application did not affect root traits for H7996. Similarly, Si treatment increased Pn, Ci, Gsand Trin leaves by 64.0, 100.0, 31.0 and 208.0%,respectively for HYT, but had no effects for H7996.

In experiment 2, R. solanacearum infection reduced the dry weight of tomato roots, stems and leaves by 26.9–38.2%for HYT and 34.3–59.1% for H7996 (Table 2). This situation was improved by citric acid or salicylic acid treatments for HYT. Pathogen infection also significantly inhibited leaf photosynthesis for HYT, but citric acid, malic acid or salicylic acid treatments could improve photosynthesis of the plants.

Fig. 2 Disease index of Ralstonia solanacearum in organic acids addtion experiment. Rs, R. solanacearum inoculation.

Fig. 3 Effects of 2 mmol L–1 Si treatment on Si concentration in tomato roots, stems and leaves. CK, no silicon application and no Ralstonia solanacearum inoculation; Si, 2.0 mmol L–1 silicon application; Rs, R. solanacearum inoculation; Rs+Si, R. solanacearum inoculation and 2.0 mmol L–1 silicon application. Data are expressed as mean±SE (n=6). Different letters on bars show significant differences at 0.05 level of probability according to Duncan’s multiple comparison test among treatments.

3.4. Organic acid concentrations in roots and root exudates

R. solanacearum inoculation significantly increased citric acid concentrations in both root exudates (Fig. 4) and roots(Fig. 5) for HYT. The concentration of SA in tomato roots was also increased for HYT, but it decreased for H7996 after R. solanacearum inoculation. In non-inoculated tomato seedlings, Si did not influence citric acid, tartaric acid,malic acid, succinic acid or fumaric acid concentrations in tomato roots of either genotypes. Furthermore, under R. solanacearum inoculation, Si application dramatically decreased citric acid concentrations of tomato root exudates and roots by 71.4 and 83.5% for HYT, respectively. In addition, salicylic acid concentrations in tomato root exudates were also increased by 30.3% for HYT and 39.6%for H7996 after Si addition.

Fig. 4 Effects of silicon addition and Ralstonia solanacearum inoculation on organic acid concentration of tomato root exudtaes 7 days after Rs inoculation. A, citric acid. B, tartaric acid. C, malic acid. D, succinic acid. E, fumaric acid. F, salicylic acid.CK, no silicon application and no Rs inoculation; Si, 2.0 mmol L–1 silicon application; Rs, R.solanacearum inoculation; Rs+Si,R.solanacearum inoculation and 2.0 mmol L–1 silicon application. Data are expressed as mean±SE (n=6). Different letters on bars show significant differences at 0.05 level of probability according to Duncan’s multiple comparison tests among treatments.

3.5. Activity of organic acid metabolizing enzyme

The activities of citrate synthase and malate synthase in roots showed no significant difference for health plants regardless of silicon application (Fig. 6). R. solanacearum inoculation significantly reduced citrate synthase, malate synthase and fumarase activities in roots especially for resistant H7996, which were reduced by 9.1, 27.4 and 15.5%, respectively. Under R. solanacearum infection condition, Si application reduced fumarase activity by 13.3%for HYT, while increased citrate synthase, malate synthase and fumerase activity by 13.7–39.0% for H7996.

Fig. 5 Effects of silicon addition and Ralstonia solanacearum inoculation on organic acid concentration of tomato roots 7 days after Rs inoculation. A, citric acid. B, tartaric acid. C, malic acid. D, succinic acid. E. fumaric acid. F, salicylic acid. CK, no silicon application and no Rs inoculation; Si, 2.0 mmol L–1 silicon application; Rs, R. solanacearum inoculation; Rs+Si, R. solanacearum inoculation and 2.0 mmol L–1 silicon application. Data are expressed as mean±SE (n=6). Different letters on bars show significant differences at 0.05 level of probability according to Duncan’s multiple comparison tests among treatments.

4. Discussion

4.1. Silicon and bacterial wilt resistance

Our results demonstrated that Si treatments can reduce the disease index of R. solanacearum especially for susceptible genotypes (Fig.1-B). Si also improves root traits,photosynthesis and plant growth (Table 1). Dannon and Wydra (2004) conducted similar hydroponic and substrate culture experiments and found that the influence of Si on mid-resistant cultivar Kingkong 2 was more pronounced than on the susceptible cultivar L390. Combining these findings, the role of Si influencing plant resistance againstpathogens clearly differs among genotypes or varieties. In another crop, Van Bockhaven et al. (2015) found that Siinduced rice resistance against brown spot was related to the primary metabolism through increased photorespiration

Fig. 6 Effects of silicon addition and Ralstonia solanacearum inoculation on organic acid metabolize related enzyme activities of tomato roots. A, citrate synthase. B, malate synthase. C,fumarase. CK, no silicon application and no R. solanacearum inoculation; Si, 2.0 mmol L–1 silicon application; Rs,R. solanacearum inoculation; Rs+Si, R. solanacearum inoculation and 2.0 mmol L–1 silicon application. Data are expressed as mean±SE (n=4). Different letters on bars show significant differences at 0.05 level of probability according to Duncan’s multiple comparison tests among treatments.

The concentration and accumulation of Si varied with different plant species. Rice plants have Si transporter genes Lsi1 and Lsi2, which have positive influences under biotic and abiotic stresses (Ma and Yamaji 2006, 2008; Ma et al. 2007). Si accumulation in rice more than doubled after Si application and Si concentrations in shoots was higher than that in roots (Sun et al. 2010; Dallagnol et al.2011). However, unlike rice, tomato is a low Si accumulating plant, and while tomato has a similar Si transporter genes,the tomato gene product is not involved in Si accumulation(Ma and Yamaji 2015). In this study, the concentration of Si in tomato roots, stems and leaves were significantly increased after silicon application (Fig. 3), showing that tomato plants can also accumulate silicon to some degree.In addition, the distribution of Si in tomato plants is root＞stem and leaf, which is different from that of rice. Moreover, ourstudy found that under R. solanacearum inoculation, the concentration of Si was higher in susceptible HYT than that in resistant H7996 plants (Fig. 3), which is consistent with the genotype difference in pathogen inhibition by Si. Similarly,Diogo and Wydra (2007) found that the Si concentration of Kingkong 2 (mid-resistant) and Hw7998 (resistant) were higher than that of susceptible cultivar L390, and that the pathogen controlling effects of Si on Kingkong 2 and Hw7998 were more striking. Taken together with these results,we can conclude that Si-enhanced resistance againstR. solanacearum may be related to the accumulation of Si in tomato plants.

4.2. Organic acid and Si-mediated bacterial wilt resistance

Root exudates are considered as belowground defense substances which play a central role in the defense when plants are exposed to pathogen attack (Baetz and Martinoia 2014). Organic acids are part of root exudates, which attract and promote the colonization of plant-beneficial bacteria and may contribute to the growth of plants and enhance the resistance of plants attacked by pathogens (Yuan et al.2015; Zhang et al. 2015). Wu et al. (2015, 2017) found that malic acid and citric acid could attract R. solanacearum under both in vitro and hydroponic conditions, which increase the incidences of R. solanacearum. However, in our experiment 2, citric acid or malic acid application did not increase the disease index (Fig. 2), which is inconsistent with the results of Wu et al. (2015, 2017). Wu et al. (2017)also found that antagonistic bacteria SQYUV 162 werestrongly attracted to root-secreted citric acid and malic acid, which resulted in the decrease of population density of R. solanacearum. These results imply that root-secreted organic acids are treated as a carbon source by both plant-beneficial bacteria and pathogens. When plants are attacked by pathogens, their roots will secret more organic acids into the soil and attract other bacteria which may have antagonistic effects on pathogens and thereby increase the resistance of plants. Ultimately, the results of these types of experiments may depend on the balance between the beneficial and pathogenic microbes in the system.

Our results showed that citric acid concentration in roots and root exudates were significantly decreased in the Rs+Si treatment compared with the Rs treatment for HYT (Figs. 4 and 5), indicating that silicon application may influence these low molecular weight compounds under biotic stress. Otherstudies have found that chlorogenic acid and rutin levels in rose leaf extracts increased in Si-treated plants under biotrophic pathogen stress and that these two compounds could reduce powdery mildew severity (Shetty et al. 2011).Another study showed that L-histidine from yeast cell extracts could inhibit the growth of R. solanacearum, which might contribute to the activation of ethylene signaling (Seo et al. 2016). Song et al. (2016) found that silicon addition resulted in higher concentrations of soluble phenolics and lignin in leaves under bacterial blight-inoculated conditions.In addition, volatile organic compounds from bacteria had negative effects on biofilm formation and root colonization of tomato by R. solanacearum (Raza et al. 2016). Therefore,the generation and secretion of organic acids may be involved in the mechanism of silicon-enhanced resistance of tomato plants against R. solanacearum.

4.3. SA and tomato resistance against bacterial wilt

Salicylic acid is considered as a plant defense hormone,which plays an important role in enhancing plant resistance(Van Bockhaven et al. 2013). Takahashi et al. (2014)found that SA, JA and ET dependent signaling pathway may participate in Bacillus thuringiensis-induction of the belowground resistance of tomato against R. solanacearum.Another study also showed that SA and chitosan induce the tomato defense responses through strengthening cell walls and inducing defense enzymes under bacteria wilt stress (Mandal et al. 2013). These results imply that defense hormone signaling pathways may play a major role in plant resistance. In our study, the SA concentrations were significantly higher in R. solanacearum-inoculation treatments than that in the control groups for both genotypes(Fig. 5). Moreover, SA application could reduce the disease index for both genotypes (Fig. 2), showing that SA signaling pathway may be involved in the process of Si-mediated tomato resistance against bacterial wilt. Fauteux et al. (2006)demonstrated that silicon could stimulate the biosynthesis of SA, JA and ET in leaves upon powdery mildew infection.Proteomic analysis showed that tomato resistance againstR. solanacearum was associated with the upregulation of apical membrane antigen protein via SA defense pathway(Afroz et al. 2009). Moreover, another study found that the SA, ET and mitogen-activated protein kinase related pathways are involved in the resistance mechanism of resistant tomato cultivars (Chen et al. 2009). In addition to the SA pathway, some studies have shown that other defense pathways including the ET pathway, also contributed to the mechanism of Si-induced R. solanacearum resistance of tomato (Van Bockhaven et al. 2015; Vivancos et al.2015), which suggests that the effects of Si may involve a combination of two or more defense pathways.

5. Conclusion

In summary, our study indicate that Si application in tomato increases silicon accumulation especially in roots, improves plant growth and enhances plant resistance against bacterial wilt. Moreover, Si treatment influences organic acid production and secretion by reducing citric acid content and increasing SA concentration in tomato roots and root exudates under R. solanacearum inoculation. These findings suggest that root exudates are involved in Si-mediated amelioration of bacterial wilt disease in tomato.

Acknowledgements

This study was financially supported by grants from the National Natural Science Foundation of China (31370456),and the Natural Science Foundation of Guangdong Province, China (2017A030313177).