Watermelon [Citrullus lanatus (Thunb.) Matsum. and Nakai var. Zaojia 8424] and wheat (Triticum aestivum L. var. Emai 18) were used as plant materials. FON was isolated from an infected watermelon plot.

Plastic greenhouse situated at Huazhong Agricultural University (113°41′-115°05′ E, 29°58′-31°22 N) was used for these experiments from March 2015 to June 2015. The experiment consisted of four treatments: (1) watermelon monocropping (M) in which one watermelon seedling was grown alone in the pot; (2) watermelon monocropping with FON inoculation (MF); (3) watermelon/wheat intercropping (I) in which one watermelon seedling was grown with 40 wheat seedlings (the distance between the watermelon and wheat seedlings was 10 cm); and (4) watermelon/wheat intercropping with FON inoculation (IF). A total of 30 pots were utilized in each treatment and all the treatments were replicated three times.

Watermelon seeds were germinated at 30°C for 48 h. After the emergence of true leaves, transplantation of uniform plants was done in plastic pots (34 cm diameter, 24 cm height) containing 8 L of moist inoculated mixture of peat and vermiculite (v/v, 1:1). The Hoagland solution (Hoagland and Arnon, 1950) was used to fulfill the nutritional requirements of the plants. The watermelon plants of the MF and IF groups were inoculated with FON spores (2.0 × 10⁶ mL^-1; 100 mL per pot) near the roots after 7 days of transplanting.

Watermelon root samples were harvested at different intervals after FON inoculation and immediately frozen in liquid nitrogen and then stored at -80°C. The rhizosphere medium samples were taken by gently mixing the media near the roots. The media containing the samples were stored at -70°C for quantification of the FON population. Roots of watermelon samples were washed with tap water and then dipped in a container with 2 L of deionized water. Plants were placed in a phytotron at 28°C/16°C for 10 h to collect root exudates, which were subsequently filtered by 0.45 μm Millipore membranes and evaporated by rotary evaporator at 40°C before the samples were stored at -20°C. Each treatment was cultured in triplicate.

Total plants used in the experiment were divided by diseased plants to get the percentage of disease incidence (Wu et al., 2009).

To measure the dry weight of whole plants, six plants per replicate were harvested on days 15 and 25 after FON inoculation. The adhering media was removed from the roots with tap water after the collection. The plants were placed for drying in an oven for 15 min at 105°C and then placed at 70°C for 3 days. The dry weight of each plant was measured with an electric balance (MSE24P-1-CE-DA, Cubis, Sortorious, Göttingen, Germany).

The PowerSoil DNA isolation kit was used for the extraction of genomic DNA according to the manufacturer’s instructions (Omega Bio-Tek, Inc., Norcross, GA, United States). Real-time PCR (qRT-PCR) was used for the estimation of FON populations. The specific primers FON-1 and FON-2 were used for the quantification of FON (Zhang et al., 2005). Quantitative PCR (qPCR) amplifications were conducted on a QuantStudio^TM 7 Flex Real-Time PCR System using qPCR SuperMix. The reaction mixture was 10 μL containing 5 μL of 2 × Top Green qPCR SuperMix, 10 pM 0.5 μL of each primer, 1 μL of template DNA (100 ng/μL) and 3 μL of sterile water. The PCR program was run for 5 min at 94°C, followed by 40 cycles of 94°C for 30 s, 58°C for 15 s, 72°C for 15 s, and a final elongation at 72°C for 7 min. Plasmid strands for the estimation of FON were generated by cloning the target gene from FON genomic DNA. The standard curves were created by Zhou and Wu (2012a). FON population was expressed as number of copies (Wakelin et al., 2008).

The root tissues (0.3 g) were ground in liquid nitrogen and acetate buffer (0.1 M, pH 5.6) was added (2.5 mL/mg tissue), mixed, and centrifuged at 4°C for 20 min at 12 000 rpm. The supernatant was used for the measurement of SA according to the method described by Defraia et al. (2008) and Cheng et al. (2016). The estimation of PAL was done according to the method described by Edwards and Kessmann (1992). The activity of benzoic acid-2-hydroxylase (BA2H) was measured according to the method described by Pan et al. (2006).

The RNA from watermelon roots was extracted by TransZol reagent (TransGen) Biotech, Inc., according to the manufacturer’s instructions. The extracted RNA was dissolved with diethylpyrocarbonate-treated water. 1 μg of total RNA was synthesized for the cDNA template by using HiScriptII QRT Super Mix for qPCR. qRT-PCR analysis was performed after FON inoculation as described by Cheng et al. (2016). CICAC and CITUA were used as reference genes (internal control) (Kong et al., 2014). The primers were designed according to the method described for the CIPAL gene family (Dong and Shang, 2013), and the primers for specific genes of watermelon were designed from the coding DNA sequences (CDSs) (v1) in the Cucurbit Genomics Database¹ (Supplementary Table S1). These primers were utilized for amplification and data was analyzed according to the method described by Livak and Schmittgen (2001).

Fink et al. (1988) and Schraudner et al. (1992)methods were utilized for the enzymatic assays. Root tissues (0.3 g) were ground to a powder in liquid nitrogen. The homogenate was centrifuged at 4°C for 20 min at 12000 rpm after the addition of ice cold PBS buffer (50 mM, pH 7.8), 0.2 mM EDTA, 2% (w/v) PVP, and 2 mM AsA.

The seedlings of watermelon and wheat having 5 to 6 leaves were washed with distilled water and shifted into a glass container covered with a black plastic sheet to protect from light and contamination. Plants from different treatments were cultured in triplicate. Plantlets were placed in a phytotron at 28°C/16°C for 48 h to collect root exudates and filtered with 0.45 μm Millipore membranes and evaporated by rotary evaporator at 40°C. Then the samples were stored at -20°C. For producing the concentrated extract (1 g root dry weight mL^-1) the root exudates were dissolved in deionized water for the bioassay and analysis of organic components (Hao et al., 2010).

The conidial suspension was prepared by transferring a block of stock culture to potato dextrose agar (PDA) medium for 5 days at 28°C under dark conditions. Colonies were grown in Potato Dextrose Broth and incubated in a table-top concentrator at 28°C and 125 rev min^-1 for 7 days. A cheese cloth with three layers was used for the filtration of the suspension and the final concentration was adjusted to 1 × 10³ conidia/mL in sterile deionized water using hemocytometer. Different concentrations of root exudates were added on concave glass slides to check their effect on spore germination. The collected exudates from the roots of watermelon and wheat seedlings were diluted to three concentrations (0.01, 0.05, and 0.1 g mL^-1) and filtered through 0.22 μm sterile filters. Microscopic analysis was performed after 8 h for the determination of spore germination. The effect of root exudates on FON spore germination was calculated as the relative germination impact index according to the methods described by Ling et al. (2013).

To determine pathogen sporulation, Bilay’s medium (Booth, 1971) was utilized. 1 mL of root exudate from watermelon and wheat having different concentrations (0.01, 0.05, and 0.1 g mL^-1) was used for determination of pathogen sporulation. 9 mL of distilled water was added to bring the volume up to 10 mL per 50 mL Erlenmeyer flask. The spore suspension (100 μL) 2 × 10⁵ spores mL^-1 was inoculated and incubated in a shaker at 28°C at 150 rpm for 2 days. After the incubation period, the culture (100 μL) was spread onto the PDA. Plates were incubated for 4 days at 28°C. After incubation, the colonies were counted and converted in to number of condia in a liquid culture. The relative sporulation impact index was calculated according to the formula: sporulation impact index = (number of sporulation in a treatment-control)/control.

To assess the effect of root exudates of watermelon and wheat on FON mycelial growth, plugs (5 mm) containing fungus were cut from the margins and placed in the center of petri dishes containing 19 mL of PDA mixture. Then the mixture was incubated at 28°C under dark conditions with 1 mL of root exudates of watermelon and wheat at concentrations of 0.01, 0.05, and 0.1 g mL^-1, and 1 mL sterile deionized water. Diameter of colonies was measured in five different directions from each incubated plate up to 7 days. The relative growth impact index was calculated according to the method described by Ling et al. (2013).

Root exudates were filtered with a 0.22 μm filter before the estimation of phenolic acid and organic acid composition and content. HPLC-2998 system (Waters 1525, United States) was used to identify phenolic acids and organic acids from the root exudates of watermelon and wheat. The phenolic acids were determined with a Welchrom-C18 column (4.0 mm × 250 mm, 5-Micron, Agilent, United States) according to the method described by Hao et al. (2010). The mobile phase consisted of methyl alcohol (A) and 1% acetic acid solution (pH 2.59) (B) with a gradient elution. The wavelength of UV detector was adjusted to 280 nm. The column temperature was kept at 30°C. For HPLC analysis, standards such as cinnamic acid, coumaric acid, p-hydroxybenzoic acid, ferulic acid, phthalic acid, SA, and syringic acid were used. Chromatography for standard compounds was checked alone and in mixtures. The major peaks were recorded at different retention times.

The analytical conditions for organic acids were as follows: chromatographic column, Zorbax SB-Aq (4.6 mm × 250 mm, 5 μm); detector wavelength, 215 nm; column temperature, 35°C; flow velocity, 0.5 mL min^-1; and injection volume, 10 μL. Methyl alcohol (A) and 50 mmol/L KH₂PO₄-H₃PO₄ buffer solution (pH 2.45) (B) were used as mobile phases with an isocratic elution. Standard organic acids include oxalic acid, malic acid, citric acid, succinic acid, and fumaric acid for HPLC analysis.

The spores were grown on concave glass slides with a series of dilutions of p-hydroxybenzoic acid, ferulic acid, cinnamic acid, and coumaric acids for the spore germination test. The sporulation of FON was measured using Bilay’s medium (Booth, 1971) with a series of dilutions of p-hydroxybenzoic acid, ferulic acid, cinnamic acid, and coumaric acid for the sporulation test. The plugs of FON agar were cultured on PDA plates containing a series of dilutions of p-hydroxybenzoic acid, ferulic acid, cinnamic acid, and coumaric acid for the mycelial growth test. All the acid solutions were filtered (0.22 mm) and added to sterilized substrate. The control was sterilized water without acids. All the concentrations of acids were tested by using three replicates. The germination, sporulation, and growth impact index were also calculated as described earlier.

Statistical analysis was performed using SPSS statistical software (IBM SPSS 19, IBM Corporation, New York, NY, United States). Differences between two treatments were tested by independent sample t-test at P < 0.05 (Tables 1, 2). ANOVA, followed by Tukey’s test (P < 0.05) was used to analyze the differences among treatment means. A general model fitting function for analysis of dose response data was used to analyze the results of phenolic acids by the drc package of R. All the values were expressed as mean ± standard error.

The FON inoculation reduced the dry weight of watermelon plants. In a monocropping system, slow growth of watermelon plants was observed compared with the intercropping system (Supplementary Figure S1). Similarly, the dry weight of watermelon was significantly reduced in the monocropping system by FON inoculation. However, no difference was observed in each cropping system after inoculation with FON (Figure 1A).

Wheat-intercropped watermelon plants also showed less disease incidence (5.6 and 41.1%) compared with monocropped watermelon plants (16.7 and 61.1%) on days 15 and 25 after FON inoculation, respectively (Figure 1B and Supplementary Figure S2). In monocropped watermelon plants the disease symptoms were also observed earlier (on day 7) compared with wheat-intercropped watermelon plants (on day 12) after FON inoculation.

The number of FON DNA copies was higher in monocropping system compared with the intercropping system on days 15 and 25 after FON inoculation. Maximum population of FON was observed in monocropping system (12.9 × 10⁸ copies g^-1 soil) on day 25 after inoculation with FON (Figure 1C).

The germination, sporulation, and growth of FON were apparently promoted by watermelon root exudates compared with the control. A stimulatory effect of root exudates was also observed that gradually increased as the concentration increased from 0.01 to 0.1 g mL^-1. On the contrary, the germination, sporulation, and growth index of FON were apparently inhibited by the addition of wheat root exudates. These inhibitory effects were enhanced with increasing concentration of wheat root exudates (Figures 2A–D).

The root exudates were collected from watermelon and wheat and analyzed by HPLC (Supplementary Figures S3, S4). Phthalic acid, SA, and total phenolic acids present in the root exudates of wheat were three times less compared with watermelon; however, syringic acid content of root exudates of wheat was 1.8-fold compared with root exudates of watermelon (Table 1). Coumaric acid was only detected in the root exudates of wheat, whereas p-hydroxybenzoic acid, ferulic acid, and cinnamic acid were only detected in the root exudates of watermelon (Supplementary Figures S3B,C). The concentration of coumaric acid was highest (46.8%) among the phenolic acids identified in the root exudates of wheat (Table 1).

According to the retention time of the standards, four kinds of organic acids, namely oxalic acid, malic acid, citric acid, and fumaric acid, were detected in the root exudates of watermelon and wheat (Supplementary Figures S4B,C). However, only succinic acid was found in the root exudates of watermelon. The content of different organic acids was higher in the root exudates of watermelon compared with wheat. The concentration of malic acid in particular in the root exudates of watermelon was higher (55.1%) compared with that of other organic acids. The total organic acid content in the root exudates of watermelon (1.5 × 10⁴ μg⋅g^-1 root DW) was 14-fold higher compared with root exudates of wheat (1.0 × 10³ μg⋅g^-1 root DW) (Table 2).

We hypothesized that these acids might play a role in disease incidence of FON. Therefore, the effect of p-hydroxybenzoic acid, ferulic acid, cinnamic acid, and coumaric acid on FON spore germination, sporulation, and growth was determined. According to our results, p-hydroxybenzoic acid, ferulic acid, and cinnamic acid promoted spore germination, sporulation, and growth of FON. We also observed that p-hydroxybenzoic acid, ferulic acid, and cinnamic acid have a dose-dependent stimulatory effect. Ferulic acid and cinnamic acid have a greater stimulatory effect compared with p-hydroxybenzoic acid (Figure 3). At 40 μM and above, the germination, sporulation, and growth index were increased with increasing concentration of p-hydroxybenzoic acid, ferulic acid, and cinnamic acid. However, coumaric acid inhibited spore germination, sporulation, and growth of FON at a concentration above 20 μM, which showed that coumaric acid acts in a dose-dependent manner (Figures 3A–C).

Most of the phenolic acids present in the root exudates of watermelon in the monocropping system were significantly increased after FON infection (Figure 4). Similarly, phenolic acids were also found to be higher in FON-inoculated plants on day 25 compared with day 15 (Supplementary Figure S5). However, the quantity of phenolic acids secreted from watermelon roots was less in the intercropping system compared with the monocropping system on days 15 and 25 after FON inoculation. The quantity of syringic acid was dramatically reduced in the root exudates of watermelon on day 15 after FON inoculation by wheat intercropping (Figures 4A,B).

The secretion of organic acids from the roots of FON-inoculated watermelon plants was higher compared with non-inoculated watermelon plants except for citric acid secretion on day 25 after FON inoculation (Figure 5). The quantity of malic acid, citric acid, and oxalic acid secreted from the roots of watermelon was 24.0–46.9% higher in the monocropping system compared with the intercropping system on day 15 after FON inoculation (Figures 5A,B). The quantity of succinic acid was also significantly increased in the monocropping system compared with the intercropping system on day 25 after FON inoculation (Figures 5C,D).

We measured the free and conjugated SA content of watermelon after FON inoculation to determine the SA responses to FON infection. The level of SA remained constant until the 9th day in non-inoculated plants; however, both forms of SA (free and conjugated) were increased after 1 day of FON inoculation and remained higher for 9 days. Moreover, the quantity of free and conjugated SA in wheat-intercropped watermelon was higher compared with watermelon grown under the monocropping system (Figures 6A,B). Gene expression analysis showed that the transcript levels of genes in the PAL family were decreased after FON inoculation with the exception of PAL1, PAL2, PAL7, and PAL12 (Figure 6C). The relative expression of PAL1, PAL2, and PAL12 was slightly increased after FON inoculation; the relative expression of PAL7 was obviously increased and this was highest on day 1 after FON inoculation. Interestingly, the expression of the ICS gene was apparently increased by up to 2.5- to 5.9-fold and 1.7- to 3.6-fold in the intercropping system and the monocropping system, respectively, from days 1 to 5 after FON inoculation (Figure 6C). Furthermore, enzyme activities of PAL and BA2H were significantly increased in the roots of watermelon grown under the intercropping system compared with the monocropping system after FON inoculation. The increased activities of these enzymes seem involved in SA biosynthesis through the PAL pathway (Figures 6D,E).

Gene expression analysis showed that transcript abundance of NPR1, PR1, PR2, WRKY1, WRKY6, and WRKY8 genes in the roots of watermelon was induced after FON inoculation. The relative expression of NPR1 was apparently increased on day 1 after FON inoculation. This was highest on days 3 and 5 in the intercropping system and the monocropping system, respectively, after FON inoculation. The relative expression of PR1 and PR2 was increased, and the level of expression was highest on day 3 after FON inoculation in the intercropping system. The relative expression of WRKY1 was highest on day 1 after FON inoculation in the intercropping system, and this was also higher compared with the monoculture system. Increased relative expression of WRKY6 was observed on days 1, 3, 5, and 7, and the maximum level of relative expression was observed on day 1 after FON inoculation. Moreover, the relative expression of WRKY6 was higher in the intercropping system compared with the monocropping system. Enhanced transcript abundance was observed for WRKY8 in response to FON infection on day 1, and this was highest on days 3 and 1 in the monocropping system and the intercropping system, respectively (Figure 7). Furthermore, the activities of chitinase and β-1,3-glucanase were increased by 8.0 and 15.2% in the intercropping system compared with the monocropping system on days 1 and 3 after FON inoculation, respectively (Figures 8A,B).