In our previous study, the toxin was extracted from R. solani AG-3 TB YC-9 strain with activated carbon adsorption extraction, and the purified toxin compound was tested by Thin-Layer Chromatography (TLC) (Li et al., 2023). Then the toxin sample was prepared by the KBr compression method. KBr and 1.8 mg of toxin sample were ground into powder and made into a sheet. The infrared absorption peak was determined by the Nicolet 6700 Fourier transform spectrometer (Thermo Fisher Company, USA).

The mass charge ratio (m/z) of the toxin sample was obtained by the combined quadrupole Orbitrap mass spectrometer (Q Exactive™). Distilled water was used as a mobile phase solvent, and ESI (−) ionization mode was adopted. The toxin sample was measured by the German Elemental Verio MICRO cube element analyzer. The sample was decomposed, quantified, and transformed; finally, the percentage content of C, H, O, and N was detected. The toxin structure was analyzed by ¹H-NMR and ¹³C-NMR spectra.

The MMC with 5 μg/mL, 10 μg/mL, 20 μg/mL, 50 μg/mL, and 100 μg/mL were used for inoculation on leaves of Nicotiana tabacum (Var.: NC89), and sterile water was used as control. The leaves inoculated with MMC were cultured at 28° for 3 d and the lesion diameters were measured. Each treatment was tested three times. A total of 10 tobacco experiments were performed, and each leaf was inoculated at four acupuncture points including two sterile water acupuncture points as control and two MMC acupuncture points as the treatment group.

The MMC solution of 5 μg/mL, 10 μg/mL, 20 μg/mL, 50 μg/mL, and 100 μg/mL was injected into the leaves according to the infiltrating method (Arnon, 1949). The treated leaves were cultured in a dish soaked with three layers of sterile gauze at 28°, and the samples were collected at an interval of 8 h.

The leaf of 0.05g was dissolved with 1 mL phosphate buffer solution of 56 mmol/L and fully pestled. The extraction was centrifuged at 4° and 12000 rpm for 8 min, and 1mL of 1 mol/L light amine hydrochloride (pH 7.8) was dissolved with 0.5 mL supernatant and incubated for 1 h at 25°. Then, p-aminobenzene sulfonic acid of 1 mL with 17 mmol/L and 1 mL of 7 mmol/L 1-naphthylamine were added and cultured at 25° for 25 min. The OD₅₃₀ number of this mixture was measured. The amount of different substances in the standard solution (n) was measured as NO²⁻ standard curve, the absorbance was measured as the ordinate, and the amount of NO²⁻ was used as abscissa. According to the chemical reaction equation (μmol·g⁻¹·min⁻¹), the production rate of superoxide ions (O²⁻) was calculated.

Chemical reaction equation (μmol·g⁻¹·min⁻¹) = (C×V×1000)/(Vs×t×W)

The H₂O₂ production in leaves was conducted according to the constant volume colorimetry method (Patterson et al., 1984). Using 2 mL of precooled acetone, 0.05 g of leaf tissue was attrited evenly. The exaction was centrifuged at 4°, 12000 rpm, for 15 min. As the sample extraction solution, 1mL of supernatant was absorbed, and 1 mL of 5% titanium sulfate and 1 mL of concentrated ammonia water were added into the extraction solution. The supernatant was removed, and the precipitate was washed with acetone (three to five times) and dissolved by 4 mL H₂SO₄ (2 mol/L). The solution’s constant volume colorimetry (415 nm) was measured.

Hydrogen peroxide content in leaf tissue (μmol/g Fw) = (C×Vo)/(Fw×Vt)

The leaves were put into 7 mL of pure toxin solution with different concentrations of 5 μg/mL, 10 μg/mL, 20 μg/mL, 50 μg/mL, and 100 μg/mL and cultured at 28° in light conditions for 3 days. Then, 0.2 g of leaf tissue was cut into pieces, and the mixture of acetone [ethanol (V/V=1:1) (10 mL)] was infiltrated and incubated for 24 h in dark conditions. When the color of leaves was observed from green to white, the OD₆₄₅ and OD₆₆₃ were measured. The leaves in sterile water were used as a control, and each treatment was repeated three times.

Twenty microliter MMC at 50 μg/mL was injected into tobacco leaves (Var.: NC89). The injected leaves were incubated at 25° in dark conditions. Then leaves were washed with distilled water (two to three times), cut into 2 mm × 1 mm segments, and dissolved with 2.5% glutaraldehyde solution at 4° for 24 h. For 15 min dehydration, 70% and 80% ethanol solutions and 85%, 95%, and 100% acetone were used respectively, and the samples were embedded with EPon812 resin. The tissue was stained with 2% uranyl acetate lead citrate, and the ultrastructural changes of tobacco mesophyll cells were observed by transmission electron microscopy (TEM) (HT7700, HITACHI, Japan).

According to our previous study, the radicle elongation and lesion diameter of tobacco was inhibited by the crude toxin extraction from R. solani AG-3 TB, and the toxin extraction was analyzed by thin-layer chromatography and HPLC (Li et al., 2023). Then the compound I was collected by preparative HPLC and the single peak could be detected ( Figure 1 ).

Infrared absorption spectroscopy (IR) was used to identify the toxin compound structure. According to the IR spectrum, -C-H telescopic vibration and C-H bending vibration were observed at 2990 cm⁻¹, 2952 cm⁻¹, 2845cm⁻¹,1465 cm⁻¹, and 1439 cm⁻¹, which indicated that -CH₂- and -OCH₃ existed in the structure. The result indicated that the benzene ring structure, especially monosubstituted benzene, was identified in this structure on the basis of =C-H telescopic vibration, benzene ring skeleton C=C, and =C-H out of plane bending vibration determined at 3044 cm⁻¹, 1622 cm⁻¹, 1586 cm⁻¹, 1483 cm⁻¹, 752 cm⁻¹, 721 cm⁻¹, and 691 cm⁻¹. At ∼1120 cm⁻¹, ⁺P-C telescopic vibration existed, while at 1096 cm⁻¹, C-O-C telescopic vibration was observed; therefore, the ⁺P-Ar and dialkyl ether (ROR’) were included in this structure. According to the result of IR, monosubstituted benzene, ROR’, ⁺P-Ar, -CH₂-, and -OCH₃ were found in the toxin component ( Table 1 ; Figure 2A ).

According to mass spectrometry, the mass-to-charge ratio of the M⁺ peak of the toxin sample was 307.1 ( Figure 2B ), and the molecular weight of the phosphate cation in the compound was 307. Based on the percentage content of elements C, H, and O, the number ratio of elements C, H, and O in the compound was calculated to be 20:20:1 ( Table S1 ). Combined with the result of IR, the molecular formula of the toxin compound was consistent with C₂₀H₂₀ClOP.

The structure of the toxin compound was deduced by nuclear magnetic resonance (NMR), and ¹H-NMR results ( Figure 3A ; Table S2 ) revealed that four groups of peaks were identified and their integral ratio (from low field to high field) was 3:12:2:3, and the total number of protons was 20. According to the results of chemical shift values and COSY and HMBC experiments ( Figures 3C, D ), the structure of toxin compound had a spin system composed of 15 aromatic protons on three monosubstituted benzenes (δ 7.81 (3H, m), 7.65 (6H, m), and 7.62 (6H, m)), which were 3H₆,6H₄, and 6H₅, respectively. On the base of δ5.24 (2H, d) and 3.49 (3H, s) values, there were two groups of isolated protons, which were H₂ and H₁, among which H₂ was split into two peaks by the coupling splitting of P ion.

The results of the carbon spectrum, displacement table, and HSQC two-dimensional spectrum ( Table S3 ; Figures 3B, E ) indicated that two saturated Cs (δ 65.4(d) and 62.0(d)) were observed in the structure, which were C₂ and C₁, and they were split into two peaks by the coupling splitting of P ion. There were three kinds of 15 aromatic tertiary C according to the value of δ130.1(d), 133.8(d), and 135.5(d), which were 6C₅, 6C₄, and 3C₆, and they were isolated by the coupling splitting of P ion. One kind of three unsaturated seasons C was detected at δ 115.9(d); they were 3C₃, which were split into two peaks by the coupling splitting of P ion. According to the results of NMR, the structure of the toxin compound was consistent with (methoxymethyl)triphenylphosphonium chloride (MMC) (C₂₀H₂₀ClOP) ( Figure 3A ).

The pure compound MMC at the concentrations of 5 μg/mL, 10 μg/mL, 20 μg/mL, 50 μg/mL, and 100 μg/mL was treated on leaves, and the lesion diameter was measured to clarify its pathogenicity and virulence. The results indicated that 100 μg/mL of MMC can cause obvious necrosis around the inoculation point on leaves. The lesion diameters measured were 0.175 ± 0.120 cm, 0.263 ± 0.057 cm, 0.366 ± 0.072 cm, 0.425 ± 0.276 cm, and 0.506 ± 0.185 cm, respectively, after treatment with 5 μg/mL, 10 μg/mL, 20 μg/mL, 50 μg/mL, and 100 μg/mL of MMC ( Figure 4 ; Table 2 ).

In order to clarify the effect of MMC (5 μg/mL, 10 μg/mL, 20 μg/mL, 50 μg/mL, and 100 μg/mL) on the production of active oxygen species (AOS) on leaves, the production rate of superoxide anion O²⁻ and content of hydrogen peroxide (H₂O₂) were determined ( Figure 5A ). The results showed that the production rate of O²⁻ in leaf tissue was increased at 24 h and 48 h and decreased at 72 h after treatment with MMC. In addition, the production rate of O²⁻ declined at 72 h, of which the number of rates were 0.082 μmol/g.min and 0.104 μmol/g.min treated by 50 μg/mL and 100 μg/mL of MMC. The production rate of O²⁻ treated with different concentrations revealed that the production rate of O²⁻ treatment by 20 μg/mL, 50 μg/mL, and 100 μg/mL of MMC were 2.01, 2.25, and 3.09 folds higher compared with the control at 24 h.

The content of hydrogen peroxide (H₂O₂) in leaves indicated that H₂O₂ content treated by MMC at 48 h was higher than other treatment groups, among which H₂O₂ content was 0.298 μmol/g treated by 50 μg/mL of MMC, of which comparison with the control was 3.55 folds higher. In addition, H₂O₂ content in leaves treated for 72 h was lower than those leaves treated for 24 h or 48 h, but H₂O₂ content in leaves during this period was still significantly higher than in the control group. The H₂O₂ content produced in leaves significantly changed after treatment with MMC, of which the H₂O₂ contents in leaves were 0.163 μmol/g, 0.169 μmol/g, 0.217 μmol/g, and 0.298 μmol/g, respectively, treated with 5 μg/mL, 10 μg/mL, 20 μg/mL, and 50 μg/mL of MMC inoculation at 48 h ( Figure 5B ).

Therefore, AOS content in leaves can be affected by the pure compound MMC; especially, the higher concentration of MMC can have a serious impact and change the content of H₂O₂ and O²⁻ in leaves.

Chlorophyll is the main component for capturing light in photosynthesis. In the study, the chlorophyll content in leaves treated with different concentrations of MMC was decreased significantly ( Table 3 ). The decline ratio of chlorophyll treated by 5~20 μg/mL of MMC was not obvious, but the decline ratio of chlorophyll content treated by 50 μg/mL and 100 μg/mL of MMC were 42.35% and 56.13%, respectively. Therefore, the result revealed that chlorophyll content can be damaged by pure compound MMC identified from R. solani AG-3 TB, and the higher concentration of MMC when produced causes large damage to the chlorophyll content.

In order to clarify the toxicity of phytotoxin of R. solani AG-3 TB to the mesophyll cell structure of leaves, the structure of chloroplast and mitochondria in leaves was observed after treatment with 50 μg/mL of MMC. The result indicated that compared with control results of leaves treated with sterile water ( Figures 6A, B ), the structure of the chloroplast and mitochondrial cell began to split in those treated with 50 μg/mL of MMC at 12 h ( Figures 6C, D ). Moreover, the cytoplasmic wall was seriously separated, and the structure of the chloroplast membrane and lamellar completely disappeared in leaves treated with MMC for 48 h ( Figures 6E, F ).