Characterization and phytotoxicity of ophiobolins produced by Bipolaris setariae

Abstract

The Bipolaris setariae NY1 strain, isolated from a diseased green foxtail plant in Henan Province, China, showed strong pathogenicity towards green foxtail. In order to clarify the role of phytotoxic substances in the fungal pathogenicity, bioassay-directed isolation and bioactivity assays of secondary metabolites produced by the fungal strain were carried out. Five ophiobolins were obtained: 3-anhydro-ophiobolin A, 6-epi-ophiobolin A, 6-epi-ophiobolin B, 3-anhydro-6-epi-ophiobolin B and ophiobolin I. Bioassays on punctured and intact detached leaves of green foxtail indicated that 3-anhydro-ophiobolin A was the most phytotoxic, followed by 6-epi-ophiobolin A. The other three ophiobolins appeared to be inactive against green foxtail. The effects of 3-anhydro-ophiobolin A and 6-epi-ophiobolin A were synergistic. The symptoms on green foxtail caused by 3-anhydro-ophiobolin A or its mixture with 6-epi-ophiobolin A resembled those caused by the fungus. 3-Anhydro-ophiobolin A and 6-epi-ophiobolin A are likely the main pathogenic determinants of B. setariae. 6-epi-Ophiobolin A caused cytotoxicity against five kinds of human cancer cells: human colon adenocarcinoma cells (HCT-8), human liver cancer cells (Bel-7402), human gastric cancer cells (BGC-823), human lung adenocarcinoma cells (A549), and human ovarian adenocarcinoma cells (A2780). The results provide information for the development of herbicides and antitumor potential of the ophiobolin sesterterpenes.

1. Introduction

Green foxtail (Setaria viridis) is adapted to a wide range of habitats and environments and is widely distributed in temperate and subtropical regions between 45 °S and 55 °N (Peng & Byer, 2005). It is a serious weed in Spain, Iran, the United States, Japan, Canada and the former Soviet Union (Peng, et al, 2004), and a common weed in most parts of China. Green foxtail competes with crops for nutrients and moisture, reducing the yields of wheat, corn and soybean by up to 44%, 28%, and 29%, respectively (Defelice, 2002). In our investigation of fungal diseases of green foxtail, 218 fungal strains were isolated from naturally infected green foxtail collected from different regions of China and were screened for their herbicidal potential against green foxtail. One strain of Bipolaris setariae (NY1), showed strong pathogenicity to green foxtail plants. The leaves of the green foxtail plants inoculated with the strain were covered with oval or near-circular brown spots and blighted within 5 d (Zhao, et al., 2010). Bipolaris species are well known for their pathogenicity to gramineous plants. Some species are responsible for severe diseases of cereal crops, such as the common root rot of wheat caused by B. sorokiniana and leaf spots of maize caused by B. maydis and B. carbonum (Samuels & Sivanesan, 1989). Moreover, some of Bipolaris species are potentially able to infect humans and animals (da Cunha et al., 2012). Many studies have associated the plant pathogenicity of Bipolaris with the production of phytotoxic secondary metabolites, especially host-specific toxins. For instance, B. sorokiniana produces helminthosporal, B. maydis produces HMT toxin, and B. carbonum produces HC toxin (Yoder, 1980). Ophiobolins are also often the phytotoxins produced by Bipolaris species, and ophioblin A has been reported from B. setariae (Bhatia et al., 2016).

In recent years, with the continuous selection for herbicide-resistant weeds and an increased awareness of environmental impact, the detrimental side effects on agricultural production and the environment caused by the application of large amounts of chemical herbicides have raised widespread concern. Studies have shown that the phytotoxins produced by weed pathogenic fungi could be developed as natural herbicides and lead to design natural and safe bioherbicides to minimize the use of synthetic chemicals (Masi et al., 2018). Fungal phytotoxins have some advantages as potential natural herbicides, such as biological activity at lower concentrations, easier degradation than halogenated molecules, and greater structural complexity and diversity which makes it difficult for weeds to develop resistance. In addition, their novel structures and new modes of action provide chemical templates for the development of new herbicides ( Strobel, et al., 1991; Duke et al., 2002a, 2002b). However, little research has been done on the phytotoxins of B. setariae.

In order to further clarify the role of phytotoxic substances in the pathogenicity of B. setariae, we performed bioassay-directed isolation and bioactivity assays of secondary metabolites produced by the NY1 strain of B. setariae to elucidate their possible roles as virulence factors.

2. Materials and methods

2.1. Fungal strain

The strain NY1 was isolated from a naturally diseased green foxtail plant from a farmland of Nanyang, Henan Province in China, in 2007, and deposited in the Agricultural Culture Collection of China (accession number ACCC 37504).

2.2. Solid fermentation

To prepare the rice medium, 75 g of rice and 100 mL of distilled water were combined in a 500 mL Erlenmeyer flask and soaked overnight before autoclaving at 15 lb/in² for 30 min.

The strain was transferred from a PDA (potato dextrose agar) slant to PDA medium in Petri dishes and incubated at 26 °C for one wk to achieve profuse sporulation. Spores were rinsed with sterile distilled water and suspended to give a final concentration of 10⁶ spores/mL. Each flask of rice medium was inoculated with 5.0 mL of the spore suspension. A total of 20 flasks were prepared and incubated at 26 °C in the dark for 30 d (Zhang, Liu, Liu, Liu, & Che, 2009).

2.3. Isolation of phytotoxins

The fermented rice was extracted with an equal volume of ethyl acetate (EtOAc), and the organic solvent was evaporated to dryness under vacuum to afford a brown oil extract. The extraction and evaporation were repeated 3 times. The crude extracts were combined and mixed thoroughly with an equal weight of 100–200 mesh silica gel and then poured into a vacuum column. The mixture was fractionated by silica gel vacuum liquid chromatography using petroleum ether-EtOAc as the eluent (petroleum ether/EtOAc: 100/0, 97.5/2.5, 95/5, 92.5/7.5, 90/10, 85/15, 80/20, 75/25, 70/30, 65/35, 60/40, 55/45, 50/50, 40/60, 30/70, 20/80, 10/90, 0/100) and further eluted using EtOAc-MeOH (EtOAc/MeOH: 97.5/2.5, 95/5, 92.5/7.5, 90/10, 0/100), giving 23 fractions in total.

Each fraction was assayed on the leaves of S. viridis for phytotoxicity. The fractions that produced obvious symptoms were analyzed by thin-layer chromatography (TLC), and spots with the same color and polarity were combined. The phytotoxic fractions were further purified by silica gel column chromatography, preparative TLC or Sephadex LH-20 (GE Healthcare, Uppsala) column chromatography using CHCl₃-CH₃COCH₃, CH₃OH, CH₃COCH₃, or CHCl₃ as the eluent. Analytical and preparative TLC were performed on silica gel GF254 plates (Qingdao Haiyang Chemical Co. Ltd., Qingdao); the spots were visualized by exposure to UV light and/or by spraying first with iodine powder and 10% (v/v) H₂SO₄ in methanol, followed by heating at 110 °C for 10 min. The subfractions were assayed again as above. Fractionation was continued until the compounds included in the active fractions were pure.

2.4. Identification of phytotoxins

The main compounds were analyzed by NMR (nuclear magnetic resonance) spectroscopic methods. ¹H and ¹³C NMR spectra were recorded at 600 MHz in CDCl₃ on Bruker (Karlsruhe, Germany) spectrometers. Carbon multiplicities were determined by distortionless enhancement by polarization transfer (DEPT) spectra. DEPT and correlated spectroscopy experiments were performed using Bruker microprograms. The compounds were identified by searching the spectroscopic data and the literature.

2.5. Phytotoxicity tests of fractions

Green foxtail seeds were sown in 7.5-cm-diam plastic pots and germinated in a greenhouse at 25 ± 5 °C with 12 h of natural light and a relative humidity (RH) of 40%–60% for 3 wk until reaching the three-to-four-leaf stage; seedlings were thinned to obtain 5–7 plants per pot. A simple leaf assay was used as a rapid guide to isolate the suspected phytotoxins. The crude extract and chromatographic fractions were first dissolved in a small amount of acetone and then diluted to a final concentration of 5 mg/mL with sterile, distilled water containing 0.1% (v/v) Tween-80 (the final concentration of acetone was approximately 0.5%–1%, v/v). Two microliters of the assay solution were micropipetted on the intact leaf surface of green foxtail plants. Each solution was applied to five leaves of different plants in the same pot with 3 equidistant droplets on each leaf. The treated seedlings were held in a dew chamber at room temperature for 24 h and then moved into a greenhouse as described before. The leaves of the seedlings were observed for the appearance of symptoms after two d. Droplets of solution containing 0.1% (v/v) Tween-80 and 1% (v/v) acetone were applied to leaves as controls.

2.6. Phytotoxicity of pure compounds

The phytotoxicities of pure compounds on green foxtail were tested by using a leaf-puncture assay and leaf-intact assay. All pure compounds were diluted to concentrations of 0.02, 0.1, 0.5, and 1 mg/mL with the methods described above in section 2.5. Leaves were detached from healthy plants grown in a greenhouse and then kept in Petri dishes with two layers of absorbent paper wetted with sterile water. Large leaves were cut into segments approximately 5 cm long to be placed into Petri dishes. In the leaf-puncture assay, three 0.5 mm mild lesions were made equidistant from each other on each leaf segment with a glass capillary, and then a 5 µL droplet of the assay solution was applied to each lesion. In the leaf-intact assay, a 5 µL droplet of the assay solution was directly applied to the leaf segment, with 3 equidistant droplets per segment. Every treatment was repeated on three leaf segments. Droplets of solution containing 0.1% (v/v) Tween-80 and 1% (v/v) acetone were applied to leaves as controls. Symptoms were observed two d after droplet application. The experiment was repeated three times.

In order to determine the interaction among different compounds, 3-anhydro-ophiobolin A was mixed with equal amounts of 6-epi-ophiobolin A, 6-epi-ophiobolin B, 3-anhydro-6-epi-ophiobolin B, and ophiobolin I, respectively. The concentration was 0.5 mg/mL for each compound in the mixture solutions. The phytotoxicity of the mixtures was tested on green foxtail leaves by using the leaf-puncture assay as described in the above paragraph. Droplets of solution containing only 3-anhydro-ophiobolin A were applied to the leaves as a control, and the experiment was repeated three times.

2.7. Cytotoxicity of pure compounds

The cytotoxicities of four compounds, 6-epi-ophiobolin A, 6-epi-ophiobolin B, 3-anhydro-6-epi-ophiobolin B and ophiobolin I, were determined in vitro against five human cancer cells using the standard MTT (thiazolyl blue tetrazolium bromide) colorimetric method. The five cancer cells were human colon adenocarcinoma (HCT-8), human liver cancer cells (Bel-7402), human gastric cancer cells (BGC-823), human lung adenocarcinoma (A549), and human ovarian adenocarcinoma (A2780).

3. Results

3.1 Isolation and identification of compounds

A total of 19.7 g of crude extract was obtained from the fermented rice colonized by B. setariae NY1 strain. The results of the bioassay indicated that the phytotoxic components were mainly concentrated in Fr. 8 to Fr. 11. Fr. 8 and Fr. 9 were combined (892 mg) and further separated by silica gel column chromatography using CHCl₃-CH₃COCH₃ (10:1) as the eluent to obtain a 218 mg subfraction with phytotoxicity. The subfraction was purified by preparative TLC with CHCl₃-CH₃COCH₃ (10:1) as the eluent, and four bands were produced. The first (R_f 0.8) and second (R_f 0.7) bands were further purified by Sephadex LH-20 column chromatography with CH₃OH as the eluent. Compound 1 (13 mg) and compound 2 (11 mg) were obtained and identified as 3-anhydro-ophiobolin A and 6-epi-ophiobolin A, respectively. The other two bands were purified by Sephadex LH-20 column chromatography with CH₃COCH₃ as the eluent. Compound 3 (19 mg) and compound 4 (21 mg) were obtained and identified as 6-epi-ophiobolin B and 3-anhydro-6-epi-ophiobolin B, respectively. Fr. 10 and Fr. 11 were combined (647 mg) and further separated by silica gel column chromatography with CHCl₃-CH₃COCH₃ (10:1) as the eluent to obtain a 172 mg subfraction with phytotoxicity. The subfraction was purified by preparative TLC with CHCl₃-CH₃COCH₃ (10:1) as the eluent, and one primary band was produced. The band (R_f 0.3) was further purified by Sephadex LH-20 column chromatography with CHCl₃ as the eluent. Compound 5 (11 mg) was obtained and identified as ophiobolin I. The identification data of five ophiobolins are as follows.

3-Anhydro-ophiobolin A (Fig. 1): ¹H-NMR (600 MHz, CDCl₃) δ: 0.90 (3H, s), 1.07 (3H, d, J = 7.0 Hz), 1.43 (2H, m), 1.62 (2H, m), 1.68 (3H, s), 1.73 (3H, s), 1.82 (4H, m), 2.01 (1H, dd, J = 13.3, 3.7 Hz), 2.07 (3H, s), 2.26 (1H, m), 2.36 (1H, m), 2.68 (2H, m), 2.86 (1H, m), 3.44 (1H, d, J = 4.0 Hz), 4.62 (1H, m), 5.15 (1H, m), 6.05 (1H, m), 6.84 (1H, dd, J = 6.6, 2.4 Hz), 9.34 (1H, s); ¹³C-NMR (150 MHz, CDCl₃) δ: 16.2 (q, C-23), 17.1 (q, C-20), 18.1 (q, C-25), 22.3 (q, C-22), 25.8 (q, C-24), 29.8 (t, C-9), 30.6 (t, C-1), 35.5 (d, C-13), 41.8 (t, C-15), 42.2 (t, C-12), 42.5 (s, C-16), 46.8 (t, C-11), 49.2 (d, C-2), 49.3 (d, C-6), 53.9 (d, C-10), 72.0 (d, C-17), 96.1 (s, C-14), 126.8 (d, C-18), 130.5 (d, C-4), 135.2 (s, C-19), 141.1 (s, C-7), 155.1 (d, C-8), 177.1 (s, C-3), 192.8 (d, C-21), 206.9 (s, C-5). These data were consistent with those previously reported (Kim, et al., 1984).

Fig. 1. The structures of 6-epi-ophiobolin A and 3-anhydro-ophiobolin A. R₁=OH, R₂=αH, 6-epi-ophiobolin A; R₁=omitted, R₂=βH, 3-anhydro-ophiobolin A.

6-epi-Ophiobolin A (Fig. 1): ¹H-NMR (600 MHz, CDCl₃) δ: 9.23 (s, 1H), 7.21(t, J = 8.4 Hz, 1H), 5.15 (d, J = 8.4 Hz, 1H), 4.42 (dt, J = 5.4, 8.4 Hz, 1H), 3.25 (d, J = 10.2 Hz, 1H), 3.22 (br. s, 1H), 2.80 (d, J = 15.6 Hz, 1H), 2.50 (d, J = 15.6 Hz, 1H), 2.42 (dd, J = 9.0, 12.6 Hz, 1H), 2.35 (dt, J = 3.0, 12.6 Hz, 1H), 2.24 (dt, J = 9.0, 11.4 Hz, 1H), 2.17 (m, 1H), 2.03 (ddd, J = 8.4, 10.8, 13.8 Hz, 1H), 1.66-1.80 (m, 4H), 1.63 (dd, J = 9.0, 14.4 Hz, 2H), 1.41 (dd, J = 7.8, 12.6 Hz, 1H), 1.33 (dd, J = 12.6, 15.0 Hz, 1H), 1.73 (s, 3H), 1.69 (s, 3H), 1.36 (s, 3H), 1.14 (d, J = 7.2 Hz, 3H), 0.82 (s, 3H); ¹³C-NMR (150 MHz, CDCl₃) δ: 15.0 (q, C-23), 18.1 (q, C-25), 21.3 (q, C-20), 22.7 (q, C-22), 25.7 (q, C-24), 28.3 (t, C-9), 35.8 (t, C-1), 37.3 (d, C-13), 39.7 (t, C-15), 41.0 (s, C-16), 43.9 (t, C-11), 44.2 (t, C-12), 49.3 (d, C-6), 50.5 (d, C-2), 53.2 (d, C-10), 53.8 (t, C-4), 70.5 (d, C-17), 77.6 (s, C-3), 97.5 (s, C-14), 126.9 (d, C-18), 135.7 (s, C-19), 144.2 (s, C-7), 153.7(d, C-8), 192.7 (d, C-21), 216.3 (s, C-5). These data were consistent with those previously reported (Canales & Gray, 1988; Li,et al., 1995).

Fig. 2. The structures of 6-epi-ophiobolin B and 3-anhydro-6-epi-ophiobolin B. R₁=OH, R₂=αH, 6-epi-ophiobolin B; R₁=omitted, R₂=αH, 3-anhydro-6-epi-ophiobolin B.

6-epi-Ophiobolin B (Fig. 2): ¹H-NMR (600 MHz, CDCl₃) δ: 9.23 (s, 1H), 7.21(t, J = 8.4 Hz, 1H), 5.12 (t, J = 7.2 Hz, 1H), 3.25 (d, J = 10.2 Hz, 2H), 2.80 (d, J = 19.2 Hz, 1H), 2.53 (dd, J = 8.4, 12.0 Hz, 1H), 2.48 (d, J = 19.2 Hz, 1H), 2.39 (dt, J = 2.4, 12.6 Hz, 1H), 2.30 (m, 1H), 2.13 (m, 1H), 1.94 (m, 1H), 1.58-1.90 (m, 8H), 1.32-1.43 (m, 3H), 1.10 (m, 1H), 1.70 (s, 3H), 1.62 (s, 3H), 1.36 (s, 3H), 0.90 (s, 3H), 0.84 (d, J=7.2 Hz, 3H); ¹³C-NMR (150 MHz, CDCl₃) δ: 13.2 (q, C-23), 17.6 (q, C-22), 22.8 (q, C-25), 23.5 (q, C-20), 25.4 (d, C-13), 25.7 (q, C-24), 25.9 (t, C-17), 28.5 (t, C-9), 34.0 (s, C-16), 36.7 (t, C-15), 38.3 (t, C-1), 43.8 (t, C-12), 46.1 (t, C-11), 49.6 (d, C-6), 51.2 (d, C-2), 54.0 (t, C-4), 77.7 (s, C-3), 125.7 (d, C-18), 131.6 (s, C-19), 138.0 (d, C-10), 142.8 (s, C-14), 143.3 (s, C-7), 153.5 (d, C-8), 193.4 (d, C-21), 216.2 (s, C-5). These data were consistent with those previously reported (Sugawara et al., 1987;Li et al., 1995).

3-Anhydro-6-epi-ophiobolin B (Fig. 2): ¹H-NMR (600 MHz, CDCl₃) δ: 9.41 (s, 1H), 7.06 (d, J = 6.0 Hz, 1H), 6.11 (s, 1H), 5.09 (t, J = 6.6 Hz, 1H), 4.15 (s, 1H), 3.18 (s, 1H), 2.73 (d, J = 19.2 Hz, 1H), 2.44 (m, 1H), 2.26 (br. d J = 13.8 Hz, 1H), 2.07 (m, 1H), 1.90 (m, 1H), 1.78 (m, 1H), 1.62-1.72 (m, 3H), 1.34-1.48 (m, 4H), 1.25 (m, 1H), 0.99 (m, 1H), 2.23 (s, 3H), 1.69 (s, 3H), 1.60 (s, 3H), 0.89 (d, J = 6.6 Hz, 3H), 0.76 (s, 3H); ¹³C-NMR (150 MHz, CDCl₃) δ: 12.0 (q, C-23), 17.7 (q, C-24), 18.7 (q, C-22), 20.1 (d, C-13), 22.8 (q, C-25), 24.5 (t, C-17), 25.7 (q, C-20), 27.4 (t, C-9), 30.6 (s, C-16), 36.1 (t, C-1), 38.8 (t, C-15), 46.6 (t, C-12), 48.6 (t, C-11), 51.0 (d, C-6), 51.2 (d, C-2), 125.8 (d, C-18), 130.6 (d, C-4), 132.3 (s, C-19), 138.3 (d, C-10), 144.1 (s, C-7), 145.2 (s, C-14), 148.9 (d, C-8), 177.3 (s, C-3), 192.4 (d, C-21), 215.9 (s, C-5). These data were consistent with those previously reported (Shen et al., 1999).

Ophiobolin I (Fig. 3): ¹H-NMR (600 MHz, CDCl₃) δ: 0.99 (3H, s), 1.02 (3H, d, J = 7.0 Hz), 1.36 (3H, m), 1.58 (1H, m), 1.66 (3H, d, J = 1.1 Hz), 1.68 (1H, d, J = 6.6 Hz), 1.70 (3H, d, J = 0.8 Hz), 1.74 (2H, m), 1.80 (1H, m), 2.01 (2H, m), 2.08 (3H, s), 2.22 (1H, m), 2.53 (2H, m), 2.76 (2H, m), 3.66 (1H, d, J = 2.9 Hz), 3.92 (1H, d, J = 11.9 Hz), 4.17 (1H, d, J = 11.8 Hz), 4.58 (1H, m), 5.14 (1H, m), 5.78 (1H, d, J = 4.4 Hz), 5.95 (1H, d, J = 1.3 Hz); ¹³C-NMR (150 MHz, CDCl₃) δ: 16.2 (q, C-23), 17.5 (q, C-20), 18.1 (q, C-25), 22.4 (q, C-22), 25.8 (q, C-24), 28.1 (t, C-9), 30.7 (t, C-1), 35.2 (d, C-13), 41.6 (t, C-15), 42.1 (s, C-12), 42.2 (t, C-16), 47.2 (t, C-11), 51.1 (d, C-2), 52.7 (d, C-6), 53.5 (d, C-10), 67.4 (t, C-21), 71.9 (d, C-17), 96.3 (s, C-14), 127.1 (d, C-18), 130.2 (d, C-8), 130.9 (d, C-7), 134.6 (s, C-4), 134.7 (s, C-19), 180.7 (s, C-3), 210.2 (s, C-5). These data were consistent with those previously reported (Li et al., 1995; Sugawara et al., 1987).

Fig. 3. The structure of ophiobolin I.

3.2. Phytotoxicity of compounds on green foxtail

In the punctured detached leaves of green foxtail assay, 1 mg/mL 3-anhydro-ophiobolin A produced circular brown spots 12 h after droplet application. The lesions reached 3–4 mm diam in two d (Fig. 4). Thereafter, the lesions did not markedly vary in size, but they could darken and develop into a light brown or gray circular area limited by a dark brown border. At concentrations of 0.5 mg/mL (1.3 × 10^-3 M) and 0.1 mg/mL (2.6 × 10^-4 M), 3-anhydro-ophiobolin A produced lesions of 2–3 mm and 1–2 mm diam, respectively, two d after droplet application. The brown spots on the leaves produced by droplet application of 3-anhydro-ophiobolin A resembled those caused by the compound-producing fungus. 6-epi-Ophiobolin A (1 mg/mL) produced brown spots 1–2 mm diam two d after droplet application. The spots were much smaller than those produced by 3-anhydro-ophiobolin A. Ophiobolin I, at a concentration of 1 mg/mL, produced sporadic brown needle-like spots four d after droplet application. 6-epi-Ophiobolin B and 3-anhydro-6-epi-ophiobolin B did not produce lesions at any concentration tested (Table 1).

Fig. 4. Symptomson the green foxtail leaves treated with 1 mg/mL 3-anhydro-ophiobolin A after two d. CK is the control without ophiobolin in the solution droplets.

Table 1. Phytotoxicities of the five ophiobolins on the punctured and intact green foxtail leaves

Concentration (mg/mL)		3-Anhydro-OPH A			6-epi-OPH A			6-epi-OPH B			3-Anhydro-6-epi-OPH B			OPH I
		Punctured	Intact		Punctured	Intact		Punctured	Intact		Punctured	Intact		Punctured	Intact
1		3–4	3		1–2	<1		0	0		0	0		1	0
0.5		2–3	1–2		1	0		0	0		0	0		<1	0
0.1		1–2	<1		<1	0		0	0		0	0		0	0
0.02		<1	0		0	0		0	0		0	0		0	0

Note: 1. OPH represents ophiobolin; 2. Numerals represent diameters (in mm) of lesions two d after droplet application.

In the intact detached green foxtail leaves assay, 1 mg/mL 3-anhydro-ophiobolin A produced brown needle-like spots of 1–2 mm 12 h after droplet application; thereafter, the small spots expanded slowly and coalesced (Fig. 4). 3-Anhydro-ophiobolin A also produced brown needle-like spots at concentrations of 0.5 and 0.1 mg/mL, but the spots were smaller, especially at the concentration of 0.1 mg/mL. At a concentration of 1 mg/mL, 6-epi-ophiobolin A produced lesions less than 1 mm diam, and the other three compounds, 6-epi-ophiobolin B, 3-anhydro-6-epi-ophiobolin B and ophiobolin I, did not produce any spots on the intact detached green foxtail leaves (Table 1).

Phytotoxicity assays of ophiobolin mixtures on punctured detached green foxtail leaves indicated that the mixture of 3-anhydro-ophiobolin A and 6-epi-ophiobolin A produced larger lesions (4–5 mm diam) than when they were tested individually. However, the respective mixtures of 3-anhydro-ophiobolin A and the other three ophiobolins produced similarly sized lesions to those produced by 0.5 mg/mL 3-anhydro-ophiobolin A.

3.3. Antitumor activity of four ophiobolins

The cytotoxicity assay of four ophiobolins against five human tumor cell lines indicated that 6-epi-ophiobolin A exhibited cytotoxicity against all five cell lines, with IC₅₀ values in the range of 2.09–­­2.71 μM against four cell lines and an IC₅₀ value of 4.5 μM against human lung adenocarcinoma (A549). The other three ophiobolins were far less cytotoxic (Table 2).

Table 2. In vitro cytotoxicity of four ophiobolins against five human tumor cell lines

Ophiobolin	Cytotoxicity (IC50, μM)
	HCT-8	Bel-7402	BGC-823	A549	A2780
Ophiobolin I	>10	>10	>10	>10	>10
6-epi-Ophiobolin B	>10	>10	>10	>10	>10
6-epi-Ophiobolin A	2.7	2.71	2.36	4.5	2.09
3-Anhydro-6-epi-ophiobolin B	>10	>10	>10	>10	>10

Note: HCT-8, human colon adenocarcinoma cells; Bel-7402, human liver cancer cells; BGC-823, human gastric cancer cells; A549, human lung adenocarcinoma cells; A2780, human ovarian adenocarcinoma cells.

4. Discussion

Ophiobolins are a group of sesterterpenoids characterized by a tricyclic 5-8-5 ring system. Ophiobolin A isolated from Bipolaris sp. was the first member of the group to be isolated and characterized (Nozoe et al., 1965). Then, ophiobolin B, ophiobolin C, and ophiobolin F were isolated from B. oryzae, B. zizaniae, and B. maydis, respectively (Au et al., 2000a). Currently, nearly sixty biogenic analogs of ophiobolins have been identified. Studies on the biological actions of ophiobolins have focused on their inhibitory effects on fungi, bacteria, nematodes and their role as calmodulin antagonists (Li et al., 1995; Tsipouras et al., 1996; Au et al., 2000b). Their herbicidal activities have hardly been investigated. According to the phytotoxicity test of nine ophiobolins by using a leaf-puncture assay (Pena-Rodriguez & Chilton, 1989Kim et al., 1999 ; Evidente et al., 2006a, 2006b), ophiobolin A, 3-anhydro-ophiobolin A and 6-epi-ophiobolin A appeared to be more phytotoxic, and 3-anhydro-6-epi-ophiobolin A, ophiobolin B and ophiobolin J exhibited weak phytotoxicity. Ophiobolin E, 8-epi-ophiobolin J and ophiobolin I had very weak or no toxicity to the plants tested. Our experimental results indicated that 3-anhydro-ophiobolin A and 6-epi-ophiobolin A were strongly phytotoxic, and ophiobolin I exhibited weak phytotoxicity, whereas 6-epi-ophiobolin B and 3-anhydro-6-epi-ophiobolin B appeared to be inactive on green foxtail leaves. These results are generally consistent with previous reports. In addition, the symptoms on leaves caused by droplet application of 3-anhydro-ophiobolin A or the mixture of 3-anhydro-ophiobolin A and 6-epi-ophiobolin A resembled those caused by the compound-producing fungus B. setariae. It has been elucidated that 3-anhydro-ophiobolin A and 6-epi-ophiobolin A are the main pathogenic determinants of B. setariae. 3-Anhydro-ophiobolin A alone or together with 6-epi-ophiobolin A has certain potential for development as an herbicide to control green foxtail.

In this study, 3-anhydro-ophiobolin A showed the strongest phytotoxicity among the five ophiobolins. The addition of 6-epi-ophiobolin A markedly increased the toxicity of 3-anhydro-ophiobolin A, which indicated a synergistic effect between the two ophiobolins. In terms of phytotoxicity, there are obvious differences among the different ophiobolins, which could be attributed to their different conformations, whereas there are very few studies on the structure-activity relationships within the ophiobolin family. Evidente et al. (2006a) reported that the phytotoxicity of ophiobolin A was stronger than that of 6-epi-ophiobolin A, whereas 3-anhydro-6-epi-ophiobolin A and ophiobolin I appeared to be inactive on all tested plant species. Based on these results, the phytotoxicity of ophiobolins may be related to the hydroxy group at C-3, the stereochemistry at C-6, and the aldehyde group at C-7. Combined with the results of our study, 3-anhydro-ophiobolin A was suggested to be more active than 6-epi-ophiobolin A, whereas 6-epi-ophiobolin B, 3-anhydro-6-epi-ophiobolin B and ophiobolin I were almost inactive. Therefore, the phytotoxicity of the A-series of ophiobolins was due to the hydroxy group at C-3 and the stereochemistry at C-6, and the former was more important than the latter. Comparing the A-series and B-series of ophiobolins, the tetrahydrofuran ring formed by C-14 and C-17 should be essential for phytotoxicity. Moreover, comparing the A-series and I-series of ophiobolins, the aldehyde group at C-7 should be responsible for phytotoxicity.

Many studies have reported that ophiobolins inhibit the growth of various cancer cells to different degrees and have strong inhibitory effects on more than ten kinds of cancer cells. Some observed inhibitory effects exceeded those of current drugs. For example, 6-epi-ophiobolin A had a significantly better inhibitory effect on human lung adenocarcinoma (A549) and human colon adenocarcinoma (HCT-15) than the anticancer drug etoposide (Ahn et al., 1998), and ophiobolin K had an inhibitory effect on leukemia cells (P388) that was 7-fold greater than that of adriamycin (Wei et al., 2004). Currently, there are nearly 60 ophiobolin variants found in 23 series from A-W. The ophiobolins with antitumor activity are mainly distributed in the A-, B-, C-, G-, I-, K-, and O-series (Ahn et al., 1998; Shen et al., 1999; Wei et al., 2004; Yang et al., 2012; Zhang et al., 2012; Bladt et al., 2013; Bury et al., 2013; Sun et al., 2013; Wang et al., 2013; Morrison et al., 2014).Bhatia et al. (2016) reported that ophiobolin A isolated from B. setariae inhibited solid and haematological cancer cell proliferation. Our results showed that 6-epi-ophiobolin A had inhibitory effects not only on human lung adenocarcinoma but also on four types of other cancer cells. Although ophiobolin I had no cytotoxicity against the five cancer cells tested, it has been reported to be active against human cervical carcinoma HeLa cells and human mouth epidermal carcinoma KB cells with IC₅₀ values of 0.1 and 0.9 μg/mL, respectively (Phuwapraisirisan et al., 2007).

Through liquid or solid fermentation, generally only a few milligrams to a dozen milligrams of a certain ophiobolin can be extracted and separated per liter or per kilogram of fermented medium (Kim et al., 1999 ; Wei et al., 2004 ; Evidente et al., 2006b). In fact, in this study, in addition to the bioassay-directed isolation of secondary metabolites produced by the B. setariae NY1 strain, the main compounds suspected to be ophiobolins in fractions Fr. 7 to Fr. 12 were also isolated and identified. Among them, a total of 690 mg of pure ophiobolin I was obtained from the EtOAc extract (19.7 g) of fermented rice solid medium (175 g × 20), a yield that was quite high. Therefore, B. setariae NY1 strain could be a good candidate for improved industrial production of ophiobolin I through further optimization of fermentation conditions and genetic modification.

Currently, the herbicidal potential of metabolites isolated from fungi is commonly evaluated by the leaf-puncture assay method. However, the results obtained by using stabbed leaves might be different from those obtained by using intact leaves. In this study, both a leaf-puncture assay and an intact leaf assay were used to test the phytotoxicity of pure compounds. The results indicated that 3-anhydro-ophiobolin A produced obvious symptoms on both the stabbed leaves and intact leaves of green foxtail. This result further corroborated the strong contribution of 3-anhydro-ophiobolin A to phytopathogenicity. The rate of lesion formation and the size of the lesion on the punctured leaves appeared to be consistent and repeatable. On the intact leaves, the rate of lesion formation was slightly slower, and the size and shape of lesions varied to some extent. This may be caused by irregularities in surface structure, distribution of natural openings and other random factors on the intact leaves, which varied compound infiltration into leaf tissue. In the leaf-puncture assay, the compounds could easily infiltrate into the leaf tissue via the man-made regular wound, thus increasing the speed of lesion formation and regularity of lesion shape.

5. Conclusions

Five ophiobolins, including 3-anhydro-ophiobolin A, 6-epi-ophiobolin A, 6-epi-ophiobolin B, 3-anhydro-6-epi-ophiobolin B and ophiobolin I, were isolated from the fermented rice medium of NY1 strain of Bipolaris setariae, a pathogen of green foxtail. Bioassays determined that 3-anhydro-ophiobolin A was the most phytotoxic, followed by 6-epi-ophiobolin A, and the effects of the two ophiobolins were synergistic. 3-Anhydro-ophiobolin A and 6-epi-ophiobolin A are likely the main pathogenic determinants of B. setariae. 6-epi-Ophiobolin A caused cytotoxicity against five human cancer cells: human colon adenocarcinoma cells (HCT-8), human liver cancer cells (Bel-7402), human gastric cancer cells (BGC-823), human lung adenocarcinoma cells (A549), and human ovarian adenocarcinoma cells (A2780). The results provide information for the development of herbicides and antitumor potential of the ophiobolin sesterterpenes.

Declaration of competing interest

The authors declare that there are no conflicts of interest.

Acknowledgments

We are grateful to Prof. Yong-Sheng Che and Dr. Er-Wei Li (Institute of Microbiology, Chinese Academy of Sciences) for their help in the identification of the compounds and Prof. Ping Zhu (Institute of Materia Medica, Chinese Academy of Medical Sciences) for his help in the cytotoxicity determination of the compounds. This work was supported by the National Natural Science Foundation of China (Grant No. 31770027).

References

• Ahn, J.W. , Lee, M.K. , Choi, S.U. , Lee, C.O. , & Kim, B.S. (1998). Cytotoxic ophiobolins produced by Bipolaris sp. Journal of Microbiology and Biotechnology, 8, 406-408
• Au, T.K. , Chick, W. S. H. , & Leung, P.C. (2000a). The biology of ophiobolins, Life Sciences, 67, 733-742. https://doi.org/10.1016/s0024-3205(00)00668-8.
• Au, T.K. , Chick, W. S. H. , & Leung, P. C. (2000b). Initial kinetics of the inactivation of calmodulin by the fungal toxin ophiobolin A. International Journal of Biochemistry & Cell Biology, 32, 1173-1182. https://doi.org/10.1016/s1357-2725(00)00058-3.
• Bhatia, D.R. , Dhar, P. , Mutalik, V. , Deshmukh, S.K. , Verekar, S.A. , Desai, D.C. , Kshirsagar, R. , Thiagarajan, P. , & Agarwal, V. (2016). Anticancer activity of Ophiobolin A, isolated from the endophytic fungus Bipolaris setariae. Natural Product Research, 30, 1455-1458. http://doi.org/10.1080/14786419.2015.1062760.
• Bladt, T.T. , Dürr, C. , Knudsen, P.B. , Kildgaard, S. , Frisvad, J.C. , Gotfredsen, C.H. , Seiffert, M. , & Larsen, T.O. (2013). Bio-Activity and dereplication-based discovery of ophiobolins and other fungal secondary metabolites targeting leukemia cells. Molecules, 18, 14629-14650. https://doi.org/10.3390/molecules181214629.
• Bury, M. , Novo-Uzal, E. , Andolfi, A. , Cimini, S. , Wauthoz, N. , Heffeter, P. , Lallemand, B. , Avolio, F. , Delporte, C. , Cimmino A. , Dubois J. , Van Antwerpen P. , Zonno, M. C. , Vurro, M. , Poumay, Y. , Berger, W. , Evidente, A. , De Gara, L. , Kiss, R. Locato, V. (2013). Ophiobolin A, a sesterterpenoid fungal phytotoxin, displays higher in vitro growth-inhibitory effects in mammalian than in plant cells and displays in vivo antitumor activity International Journal of Oncology, 43, 575-585. https://doi.org/10.3892/ijo.2013.1979
• Canales, M.W. , Gray, G.R. (1988). 6-Epiophiobolin A and 3-anhydro-6-epiophiobolin A - host specific phytotoxins of Drechslera maydis (race T). Phytochemistry, 27, 1653-1663 https://doi.org/10.1016/0031-9422(88)80421-7.
• da Cunha, K.C. , Sutton, D.A. , Fothergill, A.W. , Cano, J. , Gené, J. , Madrid, H. , De Hoog, S. , Crous, P.W. , & Guarro, J. (2012). Diversity of Bipolaris species in clinical samples in the United States and their antifungal susceptibility profiles.Journal of Clinical Microbiology, 50, 4061-4066 https://doi.org/10.1128/JCM.01965-12.
• Defelice, M.S. (2002). Green Foxtail, Setaria viridis (L.) P. Beauv Weed Technology, 16, 253-257. https://doi.org/10.1614/0890-037x(2002)016[0253:gfsvlp]2.0.co;2
• Duke, S.O. , Dayan, F.E. , Rimando, A.M. , Schrader, K.K. , Aliotta, G. , Oliva, A. , & Romagni, J.G. (2002a). Chemicals from nature for weed management. Weed Science, 50, 138-151. https://doi.org/10.2307/4046356.
• Duke, S.O. , Rimando, A.M. , Baerson, S.R. , Scheffler, B.E. , Ota, E. , & Belz, R.G. (2002b). Strategies for the use of natural products for weed management., Journal of Pesticide Science, 27, 298-306. https://doi.org/10.1584/jpestics.27.298.
• Evidente, A. , Andolfi, A. , Cimmino, A. , Vurro, M. , Fracchiolla, M. , & Charudattan, R. (2006a). Herbicidal potential of ophiobolins produced by Drechslera gigantea. Journal of Agricultural and Food Chemistry, 54, 1779-1783. https://doi.org/10.1021/jf052843l.
• Evidente, A. , Andolfi, A. , Cimmino, A. , Vurro, M. , Fracchiolla, M. , & Charudattan, R. (2006b). Ophiobolin E and 8-epi-ophiobolin J produced by Drechslera gigantea, a potential mycoherbicide of weedy grasses Phytochemistry, 67, 2281-2287
• Kim, J.M. , Hyeon, S.B. , Isogai, A. , Suzuki, A. (1984). Isolation of ophiobolin A and its analogs as inhibitors to photosynthesis. Agricultural and Biological Chemistry, 48, 803-805. https://doi.org/10.1271/bbb1961.48.803.
• Kim, H.J. , Kim, J.C. , Kim, B.S. , Kim, H.G. , & Cho, K.Y. (1999). Antibiotic and phytotoxic activities of ophiobolins from Helminthosporium species. The Plant Pathology Journal, 15, 14-20
• Li, E. , Clark, A.M. , Rotella, D.P. , Hufford, C.D. (1995). Microbial metabolites of ophiobolin A and antimicrobial evaluation of ophiobolins. Journal of Natural Products., 5874-81
• Masi, M. , Nocera, P. , Reveglia, P. , Cimmino, A. , & Evidente, A. 2018 Fungal metabolites antagonists towards plant pests and human pathogens: structure-activity relationship studies. Molecules, 23, 834-834 https://doi.org/10.3390/molecules23040834.
• Morrison, R. , Gardiner, C. , Evidente, A. , Kiss, R. , & Townley, H. (2014). Incorporation of ophiobolin A into novel chemoembolization particles for cancer cell treatment. Pharmaceutical Research, 31, 2904-2917 https://doi.org/10.1007/s11095-014-1386-3.
• Nozoe, S. , Morisaki, M. , Tsuda, K. , Iitaka, Y. , Takahashi, N. , Tamura, S. , Ishibashi, K. , & Shirasaka, M. (1965). The structure of ophiobolin, a C₂₅ terpenoid having a novel skeleton.Journal of the American Chemical Society., 87, 4968-4970. https://doi.org/10.1021/ja00949a061.
• Pena-Rodriguez, L.M. , & Chilton, W.S. (1989). 3-Anhydroophiobolin A and 3-anhydro-6-epi-ophiobolin A, phytotoxic metabolites of the Johnson grass pathogen Bipolaris sorghicola. Journal of Natural Products, 52, 1170-1172. https://doi.org/10.1021/np50065a046.
• Peng, G. , & Byer, K. N. (2005). Interactions of Pyricularia setariae with herbicides for control of green foxtail (Setaria viridis)Weed Technology, 19, 589- 598. https://doi.org/10.2307/3989480.
• Peng, G. , Byer, K.N. , & Bailey, K.L. (2004). Pyricularia setariae: a potential bioherbicide agent for control of green foxtail (Setaria viridis)Weed Science, 52105114. https://doi.org/10.2307/4046799.
• Phuwapraisirisan, P. , Sawang, K. , Siripong, P. , & Tip-Pyang, S. (2007). Anhydrocochlioquinone A, a new antitumor compound from Bipolaris oryzae. Tetrahedron Letters, 48, 5193-5195. https://doi.org/10.1002/chin.200746193.
• Samuels, G.J. , Sivanesan, A. (1989). Graminicolous species of Bipolaris, Curvularia, Drechslera, Exserohilum and their teleomorphs.Mycologia, 81, 170. https://doi.org/10.2307/3759472.
• Shen, X.Y. , Krasnoff, S. B. , Lu, S. W. , Dunbar, C. D. , O'Neal, J. , Turgeon, B. G. , Yoder, O. C. , Gibson, D. M. , Hamann, M. T. ( 1999). Characterization of 6-epi-3-Anhydroophiobolin B from Cochliobolus heterostrophus. Journal of Natural Products, 62, 895-897. https://doi.org/10.1021/np980462e
• Strobel, G. , Kenfield, D. , Bunkers, G. , Sugawara, F. , Clardy, J. (1991). Phytotoxins as potential herbicides. Experientia, 47, 819-826. https://doi.org/10.1007/BF01922462.
• Sugawara, F. , Takahashi, N. , Strobel, G. , Yun, C.H. , Gray, G. , Fu, Y. , Clardy, J. (1987). Some new phytotoxic ophiobolins produced by Drechslera oryzae. Journal of Organic Chemistry, 53, 2170-2172. https://doi.org/10.1021/jo00245a008.
• Sun, W.X. , Lv, C. T. , Zhu, T. H. , Yang, X. , Wei, S.J. , Sun, J. Y. , Hong, K. , Zhu, W. M. , Huang, C. G. (2013). Ophiobolin-O reverses adriamycin resistance via cell cycle arrest and apoptosis sensitization in adriamycin-resistant human breast carcinoma (MCF-7/ADR) cells, Marine Drugs, 11, 4570-4584. https://doi.org/10.3390/md11114570.
• Tsipouras, A. , Adefarati, A. A. , Tkacz, J. S. , Frazier, E. G. , Rohrer, S. P. , Birzin, E. , Rosegay, A. , Zink, D. L. , Goetz, M. A. , Singh, S. B. , Schaeffer, J. M. (1996). Ophiobolin M and analogues, noncompetitive inhibitors of ivermectin binding with nematocidal activity. Bioorganic & Medicinal Chemistry, 4, 531-536. https://doi.org/10.1016/0968-0896(96)00036-3
• Wang, Q.X. , Bao, L. , Yang, X. L. , Liu, D. L. , Guo, H. , Dai, H.Q. , Song, F. H. , Zhang, L. X. , Guo, L. D. , Li. S. J. , Liu, H. W. (2013). Ophiobolins P-T, five new cytotoxic and antibacterial sesterterpenes from the endolichenic fungus Ulocladium sp. Fitoterapia, 90, 220-227. https://doi.org/10.1016/j.fitote.2013.08.002.
• Wei, H. , Itoh, T. , Kinoshita, M. , Nakai, Y. , Kurotaki, M. , & Kobayashi, M. (2004). Cytotoxic sesterterpenes, 6-epi-ophiobolin G and 6-epi-ophiobolin N, from marine derived fungus Emericella variecolor GF10. Tetrahedron, 60, 6015-6019. https://doi.org/10.1016/j.tet.2004.05.021.
• Yang, T.T. , Lu, Z. Y. , Meng, L. , Wei, S. J. , Hong, K. , Zhu, W.M. , Huang, C.G. (2012). The novel agent ophiobolin O induces apoptosis and cell cycle arrest of MCF-7 cells through activation of MAPK signaling pathways. Bioorganic Medicinal Chemistry Letters 22, 579-585. https://doi.org/10.1016/j.bmcl.2011.10.079.
• Yoder, O. C. (1980). Toxins in pathogenesis. Annual Review of Phytopathology, 18, 103-129. https://doi.org/10.1146/annurev.py.18.090180.000535
• Zhang, D. H. , Fukuzawa, S. , Satake, M. , Li, X. G. , Kuranaga, T. , Niitsu, A. , Yoshizawa, K. , & Tachibana, K. (2012). Ophiobolin O and 6-Epi-Ophiobolin O, two new cytotoxic sesterterpenes from the marine derived fungus Aspergillus sp. Natural Product Communications, 7, 1411–1414. https://doi.org/10.1007/s11030-012-9400-3.
• Zhang, Y. G. , Liu, S. C. , Liu, H. W. , Liu, X. Z. , & Che, Y. S. (2009). Cycloaspeptides F and G, new cyclic pentapeptides from a Cordyceps-colonizing isolate of Isaria farinosa. Journal of Natural Products, 72, 1364-1367. https://doi.org/10.1021/np900205m.
• Zhao, X. L. , Deng, H. , Niu, Y. C. (2010). Identification and pathogenicity of fungus on Setaria viridis. Mycosystema, 29, 172-177. https://doi.org/10.1360/972010-413.