Effects of iron and boron combinations on the suppression of Fusarium wilt in banana

The effects of mineral nutrient on banana wilt disease, which are the result of a competitive relationship between host plants and pathogens, can affect the interactions of plants with microorganisms. To investigate the mineral nutrient effect, hydroponic experiments were conducted in glasshouse containing combinations of low, medium, and high iron (Fe) and boron (B) concentrations, followed by pathogen inoculation. High Fe and B treatment significantly reduced the disease index and facilitated plants growth. With increasing Fe and B concentrations, more Fe and B accumulated in plants. High Fe and B treatment dramatically reduced the Fusarium oxysporum conidial germination rate and fungal growth compared with the other two treatments, contributing to decreased numbers of the pathogen on infected plants. Furthermore, High Fe and B treatment decreased the fusaric acid production of F. oxysporum in vitro and also increased the mannitol content of the plants, which in turn decreased the phytotoxin production of infected plants and finally reduced the disease index due to the virulence factor of phytotoxin. Taken together, these results indicate that Fe and B play a multifunctional role in reducing the severity of diseases by affecting the growth of F. oxysporum and the responses between plants and pathogens.


Distribution of Fe and B in leaves and roots of banana seedlings. The Fe and B contents of tis-
sues from plants subjected to the different treatments were determined with ICP-AES. The Fe contents in both leaves and roots were dramatically enhanced with treatments of increasing Fe concentrations, especially in the H treatment (Table 3). The most significant increases in the Fe content were observed with the H treatment in both leaves and roots, with increases of 1.90 and 8.74 times in leaves and roots, respectively, over those in the L treatment. After the FOC infection, the Fe content did not change in either the LP or the MP treatment relative to the L or M treatments, respectively, but an obvious reduction was observed in the HP treatment compared with the H treatment. The B content was also significantly greater in plants given the H treatment as compared to the L or M treatment. The leaf and root B contents of plants given the H treatment were 2.67 and 2.04 times higher, respectively, than those from plants given the L treatment. When the plants were infected, the B content remained high only in plants given the HP treatment; it was almost unchanged after infection, in contrast to plants given the LP and MP treatments.

Effects of Fe and B treatments on banana seedling growth. To evaluate the effects of Fe and B
treatments on the growth of banana seedlings, dry weight (DW) was measured ( Table 4). The DW of leaves, pseudostem, and roots were significantly enhanced with a combination of increased Fe and B concentrations. The H treatment dramatically increased the biomass of all three tissues. The most significant increase in DW was observed in the H treatment, which resulted in DW increases of 1.68, 1.47, and 1.55 times in leaf, pseudostem, and root, respectively, over the L treatment. Following FOC infection, the DW of leaves, pseudostems, or roots was significantly different from that in noninfected plants.

Leaf gas exchange of banana leaves.
To compare the photosynthetic parameters of the treatment groups, the net photosynthetic rate (P n ), stomatal conductance (g s ), intercellular CO 2 concentration (C i ), and transpiration rate (T r ) were measured (Table 5). These parameters did not change under the treatments with different Fe and B combinations, but after FOC infection, reductions in some of these parameters were observed in the LP, MP, and HP treatments. The most significant decrease in the net photosynthetic rate (P n ) was observed in the LP treatment, with 2.76-, 1.67-, and 1.21-fold decreases in the LP, MP, and HP treatments compared with the L, M, and H treatments, respectively.  stems (Fig. 2). The pathogen numbers in these two parts were significantly higher than in other parts of the plant, suggesting that the pathogen spread from the root upward to the shoot. The highest pathogen numbers were observed in the root and lower stem of low concentration-treated plants, which were 7.5 and 7.8 lg (copies g −1 FW), respectively (LP). In medium concentration-treated plants, the highest pathogen numbers in the root and lower stem were 6.9 and 5.8 lg (copies g −1 FW), respectively (MP). In high concentration treated-plants, the pathogen was only detected in the root and lower stem, and the pathogen numbers in these two tissues were significantly lower than with the other two treatments: 5.3 and 3.4 lg (copies g −1 FW), or only 70% and 43% of the numbers in low concentration-treated plants (HP). The six treatments were as described in Table 2. The results shown are means ± SD of four replicates. Different small letters in the same column indicate a significant statistical difference according to the two-way analysis of variance (P < 0.05).  Table 5. Net photosynthetic rate (P n ), stomatal conductance (g s ), intercellular CO 2 concentration (C i ), and transpiration rate (Tr) of plants under different treatments. The six treatments were as described in Table 2.

Effects of
The results shown are means ± SD of four replicates. Different small letters in the same column indicate a significant statistical difference according to the two-way analysis of variance (P < 0.05). The fungal mycelium growth in agar plates containing high Fe and B was lower than in the other three treatments (Control, L, and M) after 6 days. The Fusarium colony diameter in H treatment was the smallest among the four treatments, and the other three treatments did not show a significant difference (Fig. 3B).
The FA production was the highest with the L treatment. Compared with the other two treatments (Control and H), the FA production in L and M were 1.60-fold and 1.30-fold higher, respectively. No significant difference in FA production was observed between the control and H treatments (Fig. 4).

Effects of F. oxysporum infection on the mannitol contents of banana leaves and roots. Plants
treated with different combinations of Fe and B concentrations were found to contain different mannitol content in both leaves and roots. The mannitol contents increased with increasing concentrations of Fe and B (Fig. 5A,B). In banana leaves, the mannitol contents were 1.3 and 1.22 times higher in the M and H treatments, respectively, than in the L treatment. Roots showed 1.22-and 1.62-fold increases in the M and H treatments, respectively, over the L treatment. When plants were infected, the leaf mannitol content was relatively stable, whereas in roots, the mannitol content was dramatically lower with all treatments, decreasing to 70%, 76%, and 53%, respectively, that of non-infected plants (Fig. 5A,B).
Effect of mannitol content on FA production. To illustrate the effects of mannitol on the production of FA, the sugar ingredient in Czapek Dox medium was replaced by mannitol in different ratios. The results showed that FA production decreased with increasing concentrations of mannitol and decreasing concentrations of sugar compared with the control, particularly when the mannitol concentration was 25 mg ml −1 . At this concentration, the production of FA was 55% lower than in the control treatment (Fig. 6).

FA content of banana seedlings after F. oxysporum infection. FA produced by Fusarium species was
detected in all tested tissues of infected banana plants ( Table 2). The concentrations of FA in roots, pseudostem, and leaves of plants given the LP treatment were 1.42 μ g g −1 , 1.11 μ g g −1 , and 8.64 μ g g −1 , respectively. The FA concentrations in three different organs of plants given the HP treatment were dramatically lower than in corresponding organs of plants given the LP treatment: 0.13 μ g g −1 , 0.28 μ g g −1 , and 1.39 μ g g −1 , respectively, or only 16%, 25%, and 9.3% of the LP treatment, respectively.

Discussion
Plant diseases continue to play a major limiting role in agricultural production. The control of plant diseases using classical pesticides raises serious concerns about food safety, environmental quality and pesticide resistance, which have dictated the need for alternative disease management way 15 . Nutrients are important for growth and development of plants and also microorganisms, and they are important factors in disease control 16 . All the  The six treatments were as described in Fig. 1. The results shown are means ± SD of four replicates. Different small letters indicate a significant statistical difference according to the two-way analysis of variance (P < 0.05). essential nutrients can affect disease severity 15,17 . The effects of micronutrients on reducing the severity of diseases can be attributed to their involvement in physiology and biochemistry of the plant, as many of the essential micronutrients are involved in many processes that can affect the responses of plants to pathogens 11 . Fe and B are micronutrients of plants, but their role in microbial plant pathogenesis is not well understood.
Our preliminary experiment was to determine the different B concentration on wilt disease. The results showed that different concentration of B did not influence the wilt disease index. However, high B concentration significantly promoted the growth of both banana root and shoot ( Table 1). The effect of Fe on banana wilt disease was also tested. It was observed that high Fe concentration could reduce the severity of wilt disease without growth promotion to banana seedlings (Table 1). This result was the same with Peng's result 18 . They found that through amending the soils with iron chelate could reduce the germination of chlamydospores and disease severity in banana plantlets. Therefore, we applied Fe and B combination to try to get the purpose of both growth and resistance enhancement to banana seedlings.
Through analysis of the Fe and B distribution in plants, we found that the Fe and B contents in both leaf and root were dramatically enhanced with increasing Fe and B concentrations, especially in the H treatment (Table 3). To examine different Fe and B concentration combinations on the growth of banana plants, we measured the dry weight and photosynthetic parameters (P n , g s , C i , T r ). The results showed that high applications dramatically increased the biomass of all three tissues (Table 4). However, photosynthetic parameters did not change under the treatments with different Fe and B combinations (Table 5). Following FOC infection, a strong reduction in the Fe content was observed in the HP treatment, and the B content remained at a relatively high level. Fe plays a crucial role in redox systems in cells and in various enzymes, and B is crucial for cell wall and membrane integrity 19 . Fusarium infection could disturb the water balance of infected plant, due to damage of cell membrane 20 . B promotes stability and rigidity of the cell wall structure possibly is involved in the integrity of the plasma membrane 15 . A high B content in leaves after F. oxysporum infection could help maintain the integrity of the membrane and cell wall, which was one of the important requirements of the high tolerance induced by the H treatment ( Table 2). The availability of Fe for both plants and pathogen may be quite low. Fe deficiency results in increased sensitivity to symptom development. Supplementation of plants with Fe significantly alleviated disease symptoms 21 . Similar results were obtained by Singh and Khanna 14 , who found that Fe and B deficiencies both contribute to increased lesion size, and that the low availability of these nutrients affected the accumulation of lignin. Furthermore, in the case of Dickeya dadantii, Fe deficiency caused a reduction in bacterial fitness and expression of virulence genes as well as an exacerbation of the salicylic acid-mediated defense pathway 22 . Through adding Fe-EDDHA to the soil, it was increased suppressiveness to Fusarium oxysporum f. sp. cubense 18 . Thus, the plant Fe status could influence host-pathogen relationships in different ways by affecting the pathogen's virulence as well as the host's defense 19,22 . In the present study, we observed that the conidial germination rate and fungal mycelium growth were greatly reduced by treatment with a high -Fe and -B nutrition solution (Fig. 3). Similar results were obtained by Singh and Khanna 23 , who showed that fungal conidia were very sensitive to even low amounts of Fe and B. Low amounts of Fe and B in the medium stimulated fungal growth and conidial germination, and a significant reduction in growth was observed when the nutrient concentrations were higher than 20 ppm. Reduced conidial germination and fungal mycelium growth rate would be expected to result in a decrease in the pathogen inoculum infecting the plant (Fig. 2). The reduced pathogen levels in H treatment could alleviate the injury of infected plant, which in turn maintain the relatively high photosynthetic capacity (Table 5).
FA is a nonspecific wilt-inducing toxin produced by F. oxysporum species, and the pathogenicity of this fungus is positively correlated with the FA content 24,25 . Conditions favorable for in vitro growth are also generally the most favorable for mycotoxin production 26 . Our study has shown that the toxin production was strongly reduced when the concentrations of Fe and B in the solution were 20 ppm and 2 ppm, respectively (Fig. 4).
B is not efficiently remobilized in many plant species, and sugar alcohols produced by plants, including mannitol and sorbitol, are mainly responsible for the phloem translocation of boric acid 27,28 . In our study, the mannitol contents of both leaves and roots were higher in the H treatment than in the L or M treatments (Fig. 5). The increased mannitol content was likely the reason that B was present at high levels in both leaves and roots in the H treatment (Table 3).
What's the effect of increased mannitol of plant on Fusarium? In Son's study, when mannitol supplement in medium, conidia was converted to chlamydospore and several genes are involved in this conidial modification. Fungi use mannitol to store carbon, balance redox and serve as an antioxidant 29 . Therefore, in addition to the role in B transport and increasing the resistance to both biotic and abiotic stresses 30 , mannitol could also be used as an energy source by F. oxysporum. FA plays a critical role in accelerating the development of Fusarium wilt in banana plants by acting as a phytotoxin 20 . To examine the toxin production of Fusarium on increased mannitol level, we set a series of medium containing different mannitol concentration. It was shown that both the higher concentration of mannitol and the lower concentration of sugars compared with the original concentration of culture medium probably helped decrease toxin production (Fig. 6). Although mannitol treatment could reduce the disease severity of tomato wilt 31 , the function of mannitol in the plant-pathogen interaction was not clear.
Fusarium wilt diseases are the result of the interaction between a host plant and a pathogen 6 . For F. oxysporum, entering the host and attaching to target tissues are tightly controlled by the production of virulence factors that promote the establishment of the microbe and the evasion of host defense 32 . Infection with F. oxysporum results in injury to the membrane system, which is a major effect of this disease. The injury is caused by the FA produced by the pathogen 20 . To produce disease, plant-pathogenic fungi must be able to grow on the host tissue. During root invasion and colonization, F. oxysporum is exposed to various plant defense mechanisms, such as physical barriers and antifungal compounds 33,34 . After penetration, the next step in a fungal strategy to colonize a plant species is often the secretion of toxins or plant hormonelike compounds that manipulate the plant's physiology to the benefit of the pathogen 32 . In the present study, the number of F. oxysporum in infected plants was significantly lower in the treatment with high concentrations of Fe and B (HP), as compared to the LP and MP treatments (Fig. 2). The decreased numbers of the pathogen contributed to the reduced FA production in infected plants, which resulted in a lower disease index (Fig. 2, Table 1).

Conclusion
Plants treated with combinations of a high content of Fe and B show increased resistance to F. oxysporum infection. High Fe and B contents in the plant correlated with decreased conidial germination rate, fungal growth, and FA production. The increased mannitol content may have affected the interaction between the plant and the pathogen, which in turn reduced the levels of FA produced by F. oxysporum in infected plants. Thus, it may be useful to take into account the plant Fe and B status when a need exists to control disease without compromising crop quality and yield in economically important plant species. When the plants grew to the 5 to 6-leaf stage, the seedlings were treated with Hoagland nutrient solutions containing different contents of Fe and B. Fe-EDTA and H 3 BO 3 were used as the Fe and B sources, respectively. The concentrations were classified into three levels: low, 0.5 ppm Fe + 0.05 ppm B; medium, 2 ppm Fe + 0.2 ppm B; and high, 20 ppm Fe + 2 ppm B. Each treatment group consisted of 60 plants. The placement of plants in the greenhouse was randomized to avoid edge effects.

Methods
Preparation of fungal cultures. FOC was cultured on potato dextrose agar (PDA) medium at 28 °C in darkness for 7 days. Then, 8-mm-diameter discs of fungus-containing agar were excised from the culture margins and inoculated into 500-ml Erlenmeyer flasks containing Bilay's medium 4 . The flasks were incubated for 6 days at 28 °C with rotary shaking at 180 rpm. The resulting fungal cultures were filtered through four layers of cheesecloth to remove the mycelia and then centrifuged at 8,000× g for 20 min to pellet the conidia. The conidia were resuspended in sterile water and quantified using a hemocytometer. Inoculation of plants with conidia. After the banana seedlings had grown for 14 days in solutions containing different Fe and B concentrations, they were carefully removed from the nutrient solution. The roots were submerged for 2 h in a conidial suspension containing 4 × 10 6 spores ml −1 , and the plants were then transplanted into a barrel with a nutrient solution containing 5 ml of conidial suspension to ensure infection. Control plants were treated similarly except that sterile water was used instead of the conidial suspension.
Disease severity. After  Gas exchange measurements. Net photosynthetic rate (P n ), transpiration rate (T r ), and stomatal conductance (g s ) were measured at 28 °C using a portable photosynthesis system (LI-6400; Li-Cor Biosciences, Lincoln, NE). During these measurements, the leaves were maintained at a temperature of 28 °C, a relative humidity of 50%, and a photosynthetic photon flux density of 1,000 μ mol photons m −2 s −1 . Data were recorded after the systems reached a steady-state equilibrium (approximately 10 min).
Determination of the conidial germination rate of F. oxysporum. Hoagland nutrient solutions containing various concentrations of Fe and B were prepared. The concentrations were classified into four levels: Control, 0 ppm Fe + 0 ppm B; low, 0.5 ppm Fe + 0.05 ppm B; medium, 2 ppm Fe + 0.2 ppm B; and high, 20 ppm Fe + 2 ppm B. The fungal conidial suspension was obtained from 7-day-old cultures of F. oxysporum as described by Hao et al. 37 with slight modifications, and the number of spores was adjusted to 2 × 10 6 spores per milliliter. The conidial germination rate was measured as described by 38 with some modifications. Fifty microliters of fungal conidial suspension was mixed with 0.5 ml of nutrient solution or sterile distilled water. Then 100 μ l of the mixture was then placed on separate concave glass slides. The slides were incubated for 24 h at 28 °C in darkness in a plastic container lined with moist tissue paper. The conidial germination rates were determined by microscopic observation. The percent conidial germination rate was calculated using the following formula: spore germination rate (%) = (the number of spores germinated/the number of spores observed) × 100.

Assessment of mannitol content in plant tissues.
To determine the mannitol content in plant tissues, dry powder (0.5 g) was extracted twice with 10 ml of boiling distilled water for 2 h. The resulting filtrate was collected and adjusted to a final volume of 25 ml. One milliliter of the solution containing the extract was mixed with 1 ml of sodium periodate (0.015 mol l −1 ). After 10 min, 2 ml of rhamnose (0.1%) and 4 ml of fresh Nash reagent (2 mol l −1 ammonium acetate mixed with 2 ml acetic acid and 2 ml acetyl acetone) were added to the mixture, which was then placed in a water bath at 53 °C for 15 min. A blank sample was prepared by substituting distilled water for the extract solution. The absorbance at 412 nm was measured on spectrophotometer (T6; Beijing Purkinje General Instrument Co., Ltd., Beijing, China). A standard curve was prepared using a mannitol standard. One milliliter of solution containing up to 50 μ g ml −1 of mannitol was determined by the above method, and the mannitol content of samples was calculated from a linear regression equation created from the standard curve 40 .
Extraction of FA from banana seedlings. Banana leaf, pseudostem, and root were harvested separately after the plant had been inoculated with FOC for 15 days, and then washed in tap water and wiped dry with filter paper. The tissues were weighed and homogenized in a juice extractor with MeOH/1% KH 2 PO 4 (1:1, v/v, pH 2.5). The suspension was then centrifuged at 10,000× g for 15 min. The clarified supernatants were pooled and the pH of the supernatant was adjusted to 2.5 with 2 M HCl. The acidified supernatant was sequentially extracted with 50 ml of methylene chloride. The methylene chloride extracts were pooled and evaporated to dryness at 45 °C on a rotary evaporator. The residue was redissolved in 3 ml of MeOH and stored at − 20 °C until analysis by high-performance liquid chromatography (HPLC) as described by 20 .
Scientific RepoRts | 6:38944 | DOI: 10.1038/srep38944 Specific detection of F. oxysporum by real-time polymerase chain reaction (PCR). Extraction of DNA from infected banana seedlings was performed according to 41 by grinding 100 mg of four plant tissues (root, lower stem, middle stem, upper stem) in a mortar with liquid nitrogen. The isolated DNA samples were then used as templates for PCRs. The F. oxysporum f. sp. cubense-specific primers (5′ -CAGGGGATGTATGAGGAGGCT and 5′ -GTGACAGCGTCGTCTAGTTCC) were used in a real-time PCR assay 41 . Real-time PCR amplification was performed in 25-μ l reaction mixtures containing 12.5 μ l SYBR Green PCR Master Mix (Takara, Dalian, China), 0.5 μ l ROX dye (50× ), 0.5 μ M of each primer, and 1 μ l of template DNA. The PCR program was 94 °C for 3 min, followed by 29 amplification cycles at 94 °C for 45 s, 58 °C for 45 s, and at 72 °C for 1 min. To evaluate the amplification specificity, melt-curve analysis was performed at the end of the PCR run. Standard curves were generated according to a previous report and the abundances of FOC were expressed as copy concentration as described previously 42 . Extraction of FA from pathogen culture media containing different contents of mannitol and combinations of Fe and B. To determine the effects of Fe and B on FA production, FOC was inoculated into Czapek Dox medium (40 ml in 100-ml flasks) amended with Fe-EDTA and H 3 BO 3. Four different concentrations of Fe and B culture medium were prepared. The treatments were 0 ppm Fe and 0 ppm B (Control), 0.5 ppm Fe and 0.05 ppm B (Low), 2 ppm Fe and 0.2 ppm B (Medium), and 20 ppm Fe and 2 ppm B (High). The pathogen was incubated at 28 °C on a rotary shaker (180 rpm) for about 10 days 4 . The culture was filtered with a 0.45-μ m membrane to exclude mycelia and microconidia. Subsequently, the filtrate was adjusted to pH 2.5 with 2 M HCl and extracted three times with an equal volume of methylene chloride. The organic phase (methylene chloride) was pooled and lyophilized under a vacuum. The residue was dissolved in 3 ml of methanol to obtain the crude toxin solution, and then the concentration of FA was determined by HPLC as described by 20 .
To determine the effect of mannitol on FA production, FOC was inoculated in Czapek Dox medium (40 ml in 100-ml flasks) and the sugar ingredient was replaced by mannitol in different ratios. Six treatments were performed: Control, the original sugar composition of Czapek Dox medium (30 mg ml −1 ); T 1 , 25 mg ml −1 sugar and 5 mg ml −1 mannitol in Czapek Dox medium; T 2 , 20 mg ml −1 sugar and 10 mg ml −1 mannitol in Czapek Dox medium; T 3 , 15 mg ml −1 sugar and 15 mg ml −1 mannitol in Czapek Dox medium; T 4 , 10 mg ml −1 sugar and 20 mg ml −1 mannitol in Czapek Dox medium; and T 5 , 5 mg ml −1 sugar and 25 mg ml −1 mannitol in Czapek Dox medium. FA extraction and analysis were performed as described above.
Statistical analysis. Statistical analysis was performed using the Statistix 9.0 software (Analytical Software, Tallahassee, Florida, USA). Differences between treatments were determined by the two-way analysis of variance, and P < 0.05 was taken to indicate statistical significance.