Sesquiterpenoids and Xanthones from the Kiwifruit-Associated Fungus Bipolaris sp. and Their Anti-Pathogenic Microorganism Activity

Nine previously undescribed sesquiterpenoids, bipolarisorokins A–I (1–9); two new xanthones, bipolarithones A and B (10 and 11); two novel sativene-xanthone adducts, bipolarithones C and D (12 and 13); as well as five known compounds (14–18) were characterized from the kiwifruit-associated fungus Bipolaris sp. Their structures were elucidated by extensive spectroscopic methods, electronic circular dichroism (ECD), 13C NMR calculations, DP4+ probability analyses, and single crystal X-ray diffractions. Many compounds exhibited anti-pathogenic microorganism activity against the bacterium Pseudomonas syringae pv. actinidiae and four pathogenic microorganisms.


Introduction
Kiwifruit (Actinidia chinensis Planch., Actinidiaceae) is an emerging, healthy, and economical fruit which has become increasingly popular worldwide owing to its flavor and nutritional properties [1]. It is an excellent source of vitamin C and provides balanced nutritional components of minerals, dietary fiber, folate, and health-promoting metabolites [2,3]. China is the leading kiwifruit producer in the world, followed by Italy and New Zealand. The cultivation area and annual output reached 243,000 hm 2 and 2,500,000 tons at the end of 2020 [4]. Nevertheless, as the cultivation of kiwifruit expands rapidly, many serious diseases such as bacterial canker, soft rot, bacterial blossom blight, brown spot, and root rot are a serious and ongoing threat to kiwifruit production [5][6][7][8][9][10][11][12]. Particularly, the destructive bacterial canker disease, which is associated with an infection by P. syringae pv. actinidiae (Psa), has led to reduced kiwifruit production and huge economic losses worldwide [13,14]. Although the application of copper-based chemicals and streptomycin have played a positive role in the prevention and treatment of bacterial canker, these chemical residues are extremely threatening to human health and the ecological environment [15,16]. Additionally, chemical fungicides easily induce pathogen resistance [17,18]. Thus, it is urgent to develop safer and more effective biological pesticides.
Endophytic microorganisms reside within different tissues of the host plant without causing any disease symptoms and produce various metabolites with different activities [19,20]. Therefore, the endophytic fungi have been proved to be valuable sources of important natural products [21,22]. Some natural products from endophytic fungi play important roles in plant defense systems. Therefore, we carried out the excavation of anti-Psa active substances from metabolites of kiwifruit endophytes and harvested a number of bioactive molecules. For instance, 3-decalinoyltetramic acids and cytochalasins from the kiwifruit endophytic fungus Zopfiella sp showed anti-Psa activity [23,24], while imidazole alkaloids ether were characterized as anti-Psa agents from Fusarium tricinctum [25]. These of bioactive molecules. For instance, 3-decalinoyltetramic acids and cytochalasins from the kiwifruit endophytic fungus Zopfiella sp showed anti-Psa activity [23,24], while imidazole alkaloids ether were characterized as anti-Psa agents from Fusarium tricinctum [25]. These discoveries prompted us to search for more novel and bioactive metabolites from kiwifruit-associated fungi. In the current study, a total of eighteen compounds have been isolated from the large-scale fermentation of the kiwifruit-associated fungus Bipolaris sp. (Figure 1), which included nine new sativene or longifolene sesquiterpenoids, bipolarisorokins A-I (1-9); two new xanthones, bipolarithones A and B (10 and 11); two novel sativene-xanthone adducts, bipolarithones C and D (12 and 13); as well as five known ones (14)(15)(16)(17)(18). Their structures were established by means of spectroscopic methods, namely, ECD and 13 C NMR calculations, DP4+ probability analyses, and single crystal X-ray diffractions. All compounds were evaluated for their inhibitory activities against Psa. Additionally, their inhibitory activity against four phytopathogens (Phytophthora infestans, Alternaria solani, Rhizoctonia solani, and Fusarium oxysporum) were assessed. Here, the details of isolation, structural elucidation, and bioactivity evaluations for 1-18 are reported.

General Experimental Procedures
Melting points were obtained on an X-4 micro melting point apparatus. Optical rotations were measured with an Autopol IV polarimeter (Rudolph, Hackettstown, NJ, USA). UV spectra were obtained using a double beam spectrophotometer UH5300 (Hitachi High-Technologies, Tokyo, Japan). IR spectra were obtained by a Shimadzu IRTracer-100 spectrometer using KBr pellets. 1D and 2D NMR spectra were run on a Bruker Avance III 600 MHz spectrometer with TMS as an internal standard. Chemical

General Experimental Procedures
Melting points were obtained on an X-4 micro melting point apparatus. Optical rotations were measured with an Autopol IV polarimeter (Rudolph, Hackettstown, NJ, USA). UV spectra were obtained using a double beam spectrophotometer UH5300 (Hitachi High-Technologies, Tokyo, Japan). IR spectra were obtained by a Shimadzu IRTracer-100 spectrometer using KBr pellets. 1D and 2D NMR spectra were run on a Bruker Avance III 600 MHz spectrometer with TMS as an internal standard. Chemical shifts (δ) were expressed in ppm with references to the solvent signals. High resolution electrospray ionization mass spectra (HR-ESIMS) were recorded on a LC-MS system consisting of a Q Exactive™ Orbitrap mass spectrometer with an HRESI ion source (ThermoFisher Scientific, Bremen, Germany) used in ultra-high-resolution mode (140,000 at m/z 200) and a UPLC system (Dionex UltiMate 3000 RSLC, ThermoFisher Scientific, Bremen, Germany). Column chromatography (CC) was performed on silica gel (200-300 mesh, Qingdao Marine Chemical Ltd., Qingdao, China), RP-18 gel (20-45 µm, Fuji Silysia Chemical Ltd., Kasugai, Japan), and Sephadex LH-20 (Pharmacia Fine Chemical Co. Ltd., Uppsala, Sweden). Medium-pressure liquid chromatography (MPLC) was performed on a Büchi Sepacore System equipped with a pump manager C-615, pump modules C-605, and a fraction collector C-660 (Büchi Labortechnik AG, Flawil, Switzerland). Preparative high-performance liquid chromatography (prep-HPLC) was performed on an Agilent 1260 liquid chromatography system equipped with Zorbax SB-C18 columns (5 µm, 9.4 mm × 150 mm, or 21.2 mm × 150 mm) and a DAD detector. Chiral resolution was achieved by HPLC equipped with a Daicel AD-H column. Fractions were monitored by TLC (GF 254, Qingdao Haiyang Chemical Co. Ltd. Qingdao, China), and spots were visualized by heating silica gel plates sprayed with 10% H 2 SO 4 in EtOH.

Fermentation, Extraction, and Isolation
The fungus Bipolaris sp. was isolated from fresh and healthy stems of kiwifruit plants (Actinidia chinensis Planch., Actinidiaceae), which were collected from the Cangxi county of the Sichuan Province (GPS: N 31 • 12 , E 105 • 76 ) in July 2018. Each fungus was obtained simultaneously from at least three different healthy tissues. The fungus was identified as one species of the genus Bipolaris by observing the morphological characteristics and analysis of the internal transcribed spacer (ITS) regions. A living culture (internal number HFG-20180727-HJ32) has been deposited at the School of Pharmaceutical Sciences, South-Central University for Nationalities, China.
This fungal strain was cultured on a potato dextrose agar (PDA) medium at 24 • C for 10 days. The agar plugs were inoculated in 500 mL Erlenmeyer flasks, each containing 100 mL potato dextrose media. Flask cultures were incubated at 28 • C on a rotary shaker at 160 rpm for two days as the seed culture. Four hundred 500 mL Erlenmeyer flasks, each containing 150 mL potato dextrose broth (PDB), were individually inoculated with 25 mL of seed culture and were incubated at 25 • C on a rotary shaker at 160 rpm for 25 days.

ECD Calculations
The ECD calculations were carried out using the Gaussian 16 software package [26]. Systematic conformational analyses were performed via SYBYL-X 2.1 using the MMFF94 molecular mechanics force field calculation with 10 kcal/mol of cutoff energy [27,28]. The optimization and frequency of conformers were calculated on the B3LYP/6-31G(d) level in the Gaussian 09 program package. The ECD (TDDFT) calculations were performed on the B3LYP/6-311G(d) level of theory with an IEFPCM solvent model (MeOH). The ECD curves were simulated in SpecDis V1.71 using a Gaussian function [29]. The calculated ECD data of all conformers were Boltzmann averaged by Gibbs free energy.

NMR Calculations
All the optimized conformers in an energy window of 5 kcal/mol (with no imaginary frequency) were subjected to gauge-independent atomic orbital (GIAO) calculations of their 13 C NMR chemical shifts, using density functional theory (DFT) at the mPW1PW91/6-311+G (d,p) level with the PCM model. The calculated NMR data of these conformers were averaged according to the Boltzmann distribution theory and their relative Gibbs free energy. The 13 C NMR chemical shifts for TMS were also calculated by the same procedures and used as the reference. After the calculation, the experimental and calculated data were evaluated by the improved probability DP4 + method [30].

Antibacterial Activity Assay
The bacterium P. syringae pv. actinidiae was donated by Dr. He Yan of Northwest A&F University, China. A sample of each culture was then diluted 1000-fold in fresh Luria-Bertani (LB) (Beijing Solarbio Science & Technology. Co. Ltd., Beijing, China) and incubated with shaking (160 rpm) at 27 • C for 10 h. The resultant mid-log phase cultures were diluted to a concentration of 5 × 10 5 CFU/mL, then 160 µL was added to each well of the compound-containing plates. Subsequently, 1:1 serial dilutions with sterile PBS of each compound were performed, giving a final compound concentration range from 4 to256 µg/mL. The minimum inhibitory concentration (MIC, with an inhibition rate of ≥90%) was determined by using photometry at OD 600 nm after 24 h. Streptomycin was used as the positive control.

Anti-Phytopathogens Assay
Four phytopathogens (Phytophthora infestane, Alternaria solani, Rhizoctonia solani, and Fusarium oxysporum) were cultured in PDA with micro glass beads at 27 • C for a week, as well as shaking (160 rpm). Ninety microliters of PDA, together with a 10 µL volume of an aqueous test sample solution, was added into each well of the 96-well plate. The test solutions contained different concentrations of the sample being tested. Then, agar plugs (1 mm 3 ) with fresh phytopathogens were inoculated into each well. Subsequently, a two-fold serial dilution in the microplate wells was performed over a concentration range of 4 to 256 µg/mL. Plates were covered and incubated at 27 • C for 24 h. Finally, the minimum inhibitory concentration was determined by observing the plates, with no growth in the well taken as that value. Hygromycin B was used as the positive control.

Results and Discussion
Bipolarisorokin A (1) was isolated as colorless crystals. Its molecular formula of C 15 13 C NMR spectra, in association with the HSQC spectrum, revealed two methyls, four methenes, seven methines, and two quaternary carbons (Table 1). Of them, signals at δ C 66.9 (t, C-11), 69.6 (d, C-14), and 74.9 (d, C-15) were identified as the oxygenated methylene and methines. Two olefinic carbons at δ C 156.8 (s, C-2) and 103.5 (t, C-12) corresponded to a double bond, which suggested that 1 possessed a tricyclic system. Considering the 15 carbons in 1, as well as those isolates from the same source, compound 1 was suggested to be a tricyclic sesquiterpenoid. In the 1 H-1 H COSY spectrum, a fragment was revealed, as shown with bold lines in Figure 2 [31]. Finally, the single-crystal X-ray diffraction not only confirmed the planar structure, as elucidated above, but also established the absolute configuration of 1 (Flack parameter = −0.10(7), CCDC: 2124305; Figure 4).      Bipolarisorokin C (3) was obtained as colorless needles. Its molecular formula of C15H24O3 was determined on the basis of the HR-ESIMS data (measured at m/z 253.17971 [M+H] + , calculated for C15H25O3 + 253.17982), corresponding to four degrees of unsaturation. The 1 H NMR data (Table 1) H, s, H-15). The 1 H and 13 C NMR data, in association with the HSQC data, revealed three methyls, four methenes, five methines, and three nonprotonated carbons (Table 1). Preliminary analyses on the 1D NMR data revealed that 3 was likely to be a seco-sativene type sesquiterpenoid. Detailed analyses of the 2D NMR data indicated that the majority of the data of 3 was the same as those of helminthosporol [32], except for a hydroxy group at C-8 (t, δC 64.6) in 3, which was confirmed by the HMBC correlations from  The molecular formula of bipolarisorokin B (2) was determined to be C 15  However, signals for a methyl (δ H 0.94, d, J = 6.9 Hz, H-11; δ C 16.4, C-11) and an oxygenated quaternary carbon (δ C 73.7, C-6) in 2 was suggested to replace the oxymethylene (δ H 3.64, overlap, H-11; δ C 66.9, C-11) and the methine (δ H 1.65, m, H-6; δ C 37.6, C-6) in 1. These observations indicated that the hydroxy group at C-10 in 1 migrated to C-6 in 2. The observed 1 H− 1 H COSY cross-peak of H-10 (δ H 0.88, 3H, d, J = 6.9 Hz) and H-9 (δ H 1.57, 1H, m), and H-9/H-11, along with the HMBC correlations from H-10 to C-6, C-9, and C-11 confirmed the above deduction ( Figure 2). Furthermore, ROESY correlations of H-13/H-8, H-8/H-14, H-7/H-13, and H-1/H-9 revealed that compounds 2 and 1 shared the same relative configuration. In consideration of its biosynthetic origin, the absolute configuration of compound 2 was identified the same as that of 1.  3H, s, H-12), and the proton of an aldehyde group at δ H 10.02 (H, s, H-15). The 1 H and 13 C NMR data, in association with the HSQC data, revealed three methyls, four methenes, five methines, and three nonprotonated carbons (Table 1). Preliminary analyses on the 1D NMR data revealed that 3 was likely to be a seco-sativene type sesquiterpenoid. Detailed analyses of the 2D NMR data indicated that the majority of the data of 3 was the same as those of helminthosporol [32], except for a hydroxy group at C-8 (t, δ C 64.6) in 3, which was confirmed by the HMBC correlations from δ  (Table 2). Detailed analyses of 1D and 2D NMR data revealed the differences. At first, the loss of the aldehyde group at C-1 was revealed by the chemical shift of C-1 at δ C 124.2, along with the data from 1 H-1 H COSY and HMBC spectra as shown in Figure 2. Second, the hydroxy migrated from C-8 to C-12 (δ C 59.8, t) as identified by the HMBC correlation from δ H 4.06 (2H, m, H-12) to δ C 124.2 (d, C-1), 147.2 (s, C-2), and 47.7 (s, C-3). Third, one double bond between C-9 and C-10 was built by HMBC correlations from δ H 4.69 (2H, d, J = 5.1 Hz, H-10) to δ C 22.7 (q, C-11) and 45.2 (d, C-6). The other parts of 5 were elucidated as the same as those of 3 by a detailed analysis of 2D NMR data.
Bipolarisorokin H (8)  − 251.16527). All the spectroscopic data indicated similar patterns to those of longifolene [33]. Detailed analyses of 1D and 2D NMR data revealed the differences. Signals at δ C 67.0 (d, C-5), 70. 5 (d, C-14), and 74.9 (d, C-15) were identified as the oxygenated methines. Therefore, three hydroxyls were suggested to be placed at C-5, C-14, and C-15, respectively, which were identified by the HMBC and 1 H-1 H COSY correlations, as shown in Figure 2. Comprehensive analyses of other data suggested that the other parts of 9 were the same as those of longifolene. The relative configuration of 9 was revealed by a ROESY experiment, as shown in Bipolarithone A (10) was isolated as a yellow oil, and its molecular formula was determined to be C 17 (Table 4) of 10 were similar to those of the dechlorinated methyl ester (16) isolated in this study [34]. The major difference was that 10 exhibited a dihydrofuran ring rather than a furan ring. HMBC correlations from H-8 (H, d, J = 3.9 Hz, δ H 5.64) to C-8a (δ C 114.7, s), C-7 (δ C 170.0, s), C-9 (δ C 178.3, s), and C-10a (δ C 167.7, s), together with H-5 (H, ddd, J = 6.6, 4.4, 3.9 Hz, δ H 5.73) to C-10a, C-8a, C-6 (δ C 37.7, t), and C-2 (δ C 169.5, s), supported the above assignment. The relative configuration of 10 was identified by the analysis of its ROESY data. The ROESY correlation between H-8 and H-5 indicated that H-8 had the same orientation as H-5 (assigned as an α orientation). The calculated ECD of 10 established the configuration of 10, as shown in Figure 5. Therefore, the structure of 10 was characterized as depicted.
Bipolarithone B (11) was isolated as a yellow oil. The HRESIMS spectrum of 11 suggested a molecular formula of C 17  .09179), the same as that of 10. The planar structure of 11 was elucidated to be the same as that of 10 by the analysis of its 1D and 2D NMR data. The main difference was suggested as its stereochemistry at C-8 (δ C 79.8, d). Analyses of the 1 H NMR information showed that the coupling constants of H-8, H-5, and H-6 were significantly different from those of 11, as shown in the Table 4. Furthermore, the ROESY correlation of H-8 (δ H 5.63, 1H, d, J = 1.7 Hz)/H-5 (δ H 5.62, 1H, ddd, J = 8.4, 3.8, 1.7 Hz) was not observed in 11. These data suggested that 11 was an epimer of 10. The ECD calculation for 11 was performed, and the results of 11 matched well with the experimental ECD curve ( Figure 5). Hence, the absolute configuration of 11 can be fully assigned, as shown.
Bipolarithone C (12) was assigned a molecular formula of C 30 H 36 O 9 based on its HRESIMS data (measured at m/z 541.24310 [M+H] + , calculated for C 30 H 37 O 9 + 541.24321). The NMR data of 12 were very similar to those of bipolenin I (14) (Table 5), a novel sesquiterpenoid-xanthone adduct isolated from the fungus Bipolaris eleusines [35]. The significant differences were that there was an absence of an aldehyde group and two olefinic carbons, as well as the presence of an additional methine and carbonyl, in 12. These data suggested that the α,β-unsaturated aldehyde moiety disappeared in 12. This assignment was confirmed by the HMBC correlations of δ H 2.16 (H, m, H-2), 1.29 (H, m, H-6), 2.56 (1H, br s, H-7), 0.95 (3H, d, J = 7.2 Hz, C-12), and 1.90 (H, m, H-13) to δ C 50.6 (d, C-2) and 221.6 (s, C-1). The ROESY spectrum displayed similar patterns to those of 14. Furthermore, a cross peak between H-2 and H-14a (δ H 4.05, 1H, dd, J = 11.3, 5.1 Hz) confirmed the relative configuration of C-2, as shown. The absolute configuration of 12 was elucidated by the quantum chemistry calculations. At first, the ECD calculations were conducted on the four possible conformers (12a-d), using time-dependent density functional theory (TDDFT) at the B3LYP/6-311G (d) level in methanol with the PCM model. The overall calculated ECD spectrum of each configuration was then generated according to the Boltzmann weighting of the conformers. As a result, the calculated ECD curves of 12a and 12d matched well with the experimental data ( Figure 5). To determine its final structure, the theoretical NMR calculations and DP4+ probabilities were employed. The 13 C NMR chemical shifts of 12a and 12d were calculated at the mPW1PW91/6-311+G (d,p) level in the gas phase. According to the DP4+ probability analyses, 12a was assigned with 100% probability (see data in the Supporting Information). Structurally, compound 12 comprised of a seco-sativene sesquiterpenoid unit and a xanthone unit, whose absolute configurations were in accord with compound 6 and compound 10, respectively. Therefore, the structure of 12 was established as depicted.
Bipolarithone D (13) had the same molecular formula (C 30 Table 5) resembled those of 12, except that the resonances of C-6 (∆δ C + 1.5), H-6 a (∆δ H + 0.08), and H-6 b (∆δ H + 0.15) were shifted downfield, while the data H-5 (∆δ H − 0.08) were shifted upfield. A detailed comparison of the 1D and 2D NMR data of 13 with that of 12 indicated that the two compounds possessed the same planar structure. The main difference was the stereochemistry at C-8 .
A key ROESY correlation of H-5 /H-8 could be detected in 12 but not in 13. In addition, the coupling constants of H-8 in 13 (J = 1.8 Hz) were different from that in 12 (J = 3.9 Hz). All the data suggested that compound 13 was a C-8 epimer of 12. Finally, the absolute configuration of 13 was confirmed by ECD calculations (Figure 5).
Five known compounds were determined as bipolenins I and J (14 and 15), dechlorinated methyl ester (16), drechslerines A (17), and (+)-secolongifolene diol (18) by the comparison of their spectral data with that reported in the literature [32,34,35]. In this study, the absolute configurations of compounds 17 and 18 were confirmed by single crystal X-ray diffractions (Figure 6), which could support the absolute configurations of 1-9, 12, and 13 as depicted in the text, since they were obtained from the same source.
Finally, the absolute configuration of 13 was confirmed by ECD calculations (Figure 5).
In addition, our previous study on chemicals from B. eleusines suggested that sativene-xanthone adducts have promising inhibitory activity against plant pathogenic microorganisms [35]. Therefore, all compounds were evaluated for their inhibitory activity against four plant pathogenic microorganisms, including P. infestane, A. solani, R. solani, and F. oxysporum. As a result, many compounds showed certain inhibitory activity, as given in Table 6.
A brief structure-activity relationship analysis suggested that the aldehydecontaining sativene sesquiterpenoids were more active than the others, while the xanthones or their derivatives showed better inhibitory activities than sativene sesquiterpenoids.  Table 6. Inhibitory effects of the isolates against five plant pathogens (MIC, µg/mL) a .