New Polyketides from the Marine-Derived Fungus Letendraea Sp. 5XNZ4-2

Marine-derived fungi have been reported to have great potential to produce structurally unique metabolites. Our investigation on secondary metabolites from marine-derived fungi resulted in the isolation of seven new polyketides (phomopsiketones D–G (1–4) and letendronols A–C (5–7)) as well as one known xylarinol (8) in the cultural broth of Letendraea sp. Their structures and absolute configurations were elucidated using a set of spectroscopic and chemical methods, including HRESIMS, NMR, single-crystal X-ray diffraction, ECD calculation, and a modified version of Mosher’s method. Compound 2 showed weak inhibition against nitric oxide production in lipopolysaccaride-activated macrophages with an IC50 value of 86 μM.


Introduction
Metabolites from marine-derived microorganisms have become important pharmacological resources [1,2]. At least 2000 novel natural products have been identified in the past five years, nearly 50% of which were derived from marine-derived fungi [3][4][5]. In addition, the metabolic pathways of marine-derived fungi, which can be attributed to unique marine habitats, especially those that contain sea salts, may be significantly different from terrestrial ones [6]. Thus, marine-derived fungi have great potential to produce structurally unique metabolites and have attracted a considerable amount of attention.
In our ongoing research on bioactive secondary metabolites from marine-derived fungi, we isolated an endophytic fungus Letendraea sp. 5XNZ4-2 from one sea crab. We obtained two novel spiroketals (letenketals A and B) from this fungus [7]. In order to obtain more structurally novel compounds, the systematic discovery of secondary metabolites for this strain was carried out. Seven new polyketides (phomopsiketones D-G (1)(2)(3)(4) and letendronols A-C (5-7), see Figure 1) as well as one known xylarinol (8) were isolated from a Potato DFextrose Broth (1/2 PDB) cultural broth. Herein, we report the isolation, structure elucidation, and bioactivities of these compounds.
In order to establish the relative configuration of 1, the Nuclear Overhauser Effect Spectroscopy (NOESY) was measured ( Figure S15). NOESY correlations between H-2/H-10 and H-6a/H-4 suggest that both of the cyclohexanone and dihydropyran (DHP) rings were in the twist chair conformation. The cross peaks between H-4/H-2/H-10 suggested that they adopted the same orientations. The absolute configuration of 1 was established by a comparison between the experimental ECD spectrum and the theoretically calculated values of two possible stereoisomers (2R, 4S, 10R)-1 and (2S, 4R, 10S)-1 [9,10]. The experimental ECD (Figure 3) of 1 showed a positive Cotton effect at 325 nm and a negative Cotton effect at 230 nm, which matched well with the calculated value of (2R, 4S, 10R)-1, and contributed to determining the absolute configuration of 1 as (2R, 4S, 10R). Compound 1 was determined to be a 5,6-deoxy and 2-methoxyated derivative of EI-1941-1 [8] and named phomopsiketone D. In order to establish the relative configuration of 1, the Nuclear Overhauser Effect Spectroscopy (NOESY) was measured ( Figure S15). NOESY correlations between H-2/H-10 and H-6a/H-4 suggest that both of the cyclohexanone and dihydropyran (DHP) rings were in the twist chair conformation. The cross peaks between H-4/H-2/H-10 suggested that they adopted the same orientations. The absolute configuration of 1 was established by a comparison between the experimental ECD spectrum and the theoretically calculated values of two possible stereoisomers (2R, 4S, 10R)-1 and (2S, 4R, 10S)-1 [9,10]. The experimental ECD ( Figure 3) of 1 showed a positive Cotton effect at 325 nm and a negative Cotton effect at 230 nm, which matched well with the calculated value of (2R, 4S, 10R)-1, and contributed to determining the absolute configuration of 1 as (2R, 4S, 10R).  Compound 2 was obtained as a light brown oil. Its molecular formula was determined to be C18H30O5 (with 4 degrees of unsaturation) on the basis of its HRESIMS (m/z 349.1989 for [M+Na] + ) and NMR data. IR absorption at 3426 and 1678 cm -1 indicated the presence of a hydroxyl and unsaturated ketones. Its 13 C NMR and DEPT spectra ( Table 2) showed distinctive signals, including one α, β-unsaturated ketone (δC 197.9), two olefinic carbons (δC 131.5, 154.4), one ketal carbon (δC 95.6), and one oxygenated methine (δC 67.4), similar to those of 1, and indicated that compound 2 contained a similar bicyclic skeleton to that of 1. Additional signals, including three oxygenated methylenes (δC 68.4, 69.9, 71.5), two methylenes (δC 31. 5, 19.3), and one methyl (δC 14.0), were observed, while the methoxy in 1 was absent in 2. These additional signals were organized as one 2-butoxyethoxy group (CH3 (21) Figure S1). The 2-butoxyethoxy group was connected to C-2 through an ether linkage according to the HMBC correlation from H2-15 to C-2 (δC 95.6). Thus, compound 2 was elucidated as shown in Figure 1 and named phomopsiketone E. Compound 2 was obtained as a light brown oil. Its molecular formula was determined to be C 18 H 30 O 5 (with 4 degrees of unsaturation) on the basis of its HRESIMS (m/z 349.1989 for [M+Na] + ) and NMR data. IR absorption at 3426 and 1678 cm −1 indicated the presence of a hydroxyl and unsaturated ketones. Its 13 C NMR and DEPT spectra ( Table 2) showed distinctive signals, including one α, β-unsaturated ketone (δ C 197.9), two olefinic carbons (δ C 131. 5, 154.4), one ketal carbon (δ C 95.6), and one oxygenated methine (δ C 67.4), similar to those of 1, and indicated that compound 2 contained a similar bicyclic skeleton to that of 1. Additional signals, including three oxygenated methylenes (δ C 68. 4, 69.9, 71.5), two methylenes (δ C 31. 5, 19.3), and one methyl (δ C 14.0), were observed, while the methoxy in 1 was absent in 2. These additional signals were organized as one 2-butoxyethoxy group (CH 3 (21)-CH 2 (20)-CH 2 (19)-CH 2 (18)-O-CH 2 (16)-CH 2 (15)) by 1 Figure S1). The 2-butoxyethoxy group was connected to C-2 through an ether linkage according to the HMBC correlation from H 2 -15 to C-2 (δ C 95.6). Thus, compound 2 was elucidated as shown in Figure 1 and named phomopsiketone E.  The DHP and cyclohexanone rings were also in twist chair conformations according to the NOESY correlation between H-2 and H-10 as well as the H-6 axial and H-4. The ECD spectra ( Figure 3) of 1 and 2 showed similar curves (the cotton effect was positive at 325 nm (1) and 325 nm (2), and negative at 230 nm (1) and 228 nm (2)), and determined the absolute configurations of (2R, 4S, 10R) for 2.
A NOESY correlation between H-7 and H-9 was not observed. Then, the modified version of Mosher's esterification method was applied to determine the absolute configuration of 4. Esterification of 4 with (R)-and (S)-MTPA-Cl yielded the 7,10-di-S-MTPA ester (4a) and the 7,10-di-R-MTPA ester (4b), respectively. The similar 1 H NMR chemical shift differences (Δ4e-4f, Figures S6 and 3) around the C-7 and C-10 stereocenters as 3 indicated the 7S and 10R configurations for 4. Since 4 was the stereoisomer of 3, the configuration of C-9 in 4 was deduced to be R, contrary to that in 3. Thus, the absolute configuration of 4 was determined to be (7S, 9R, 10R) and it was named phomopsiketone G.
In order to establish the relative configuration, 5 was crystallized and, by use of X-ray analysis ( Figure 5), H-4 and H-9 were assigned to the same orientation, and H-7 and H-10 were assigned to the opposite orientation of H-4 and H-9.
The tri-(S/R) MTPA esters (5a/5b) were obtained by the modified version of Mosher's method together with HPLC purification. A 1 H NMR data comparison of 5a and 5b (Δ5a-5b, Figure S7) determined the absolute configurations of (4S, 7S, 9R) in 5. Finally, the configuration of 5 was unambiguously confirmed to be (4S, 7S, 9R, and 10S) by X-ray analysis combined with the modified version of Mosher's method. Compound 4 was also obtained as a yellow oil. Its molecular formula was established as C 12 H 18 O 4 , identical to that of 3, by HRESIMS (m/z 249.1101 for [M+Na] + ) and NMR data. The proton and carbon spectra were also similar to those of 3 (Tables 1 and 2) and a two-dimensional (2D) NMR analysis ( Figure S2) enabled us to deduce that it has the same planar structure as 3. Different ECD spectra of 3 and 4 ( Figure S8) were used to confirm that they were stereoisomers.
A NOESY correlation between H-7 and H-9 was not observed. Then, the modified version of Mosher's esterification method was applied to determine the absolute configuration of 4. Esterification of 4 with (R)-and (S)-MTPA-Cl yielded the 7,10-di-S-MTPA ester (4a) and the 7,10-di-R-MTPA ester (4b), respectively. The similar 1 H NMR chemical shift differences (∆δ 4e -4f , Figure S6 and Figure 3) around the C-7 and C-10 stereocenters as 3 indicated the 7S and 10R configurations for 4. Since 4 was the stereoisomer of 3, the configuration of C-9 in 4 was deduced to be R, contrary to that in 3. Thus, the absolute configuration of 4 was determined to be (7S, 9R, 10R) and it was named phomopsiketone G.
Compound 5 was crystallized as colorless needles in acetone. Its HRESIMS showed an ion peak [M+Na] + at m/z 251.1258, corresponding to the molecular formula C 12 H 20 O 4 and possessing two hydrogen atoms more than 3 and 4. A comparison of 13 C NMR data between 3 and 5 revealed that the ketone carbonyl (δ C 197.1, C-4 in 3) was replaced by one oxygenated methine (  Figure S3) confirmed that the ketone carbonyl was hydrogenated. A similar C 5 side chain (CH 3 -13/CH 2 -12/CH 2 -11/CH-10/CH-9) to that in 3 was derived from 1 H-1 H COSY correlations ( Figure S3). Key HMBC correlations from H-2 (δ H 4.02) to C-10 (δ C 80.4) and H-9 (δ H 3.99) to C-3 (δ C 137.1) ( Figure S3) indicated the presence of a dihydropyran fragment that was different from the dihydrofuran in 3. Thus, 5 was determined to be a new polyketide and named letendronol A.
In order to establish the relative configuration, 5 was crystallized and, by use of X-ray analysis ( Figure 5), H-4 and H-9 were assigned to the same orientation, and H-7 and H-10 were assigned to the opposite orientation of H-4 and H-9.
The tri-(S/R) MTPA esters (5a/5b) were obtained by the modified version of Mosher's method together with HPLC purification. A 1 H NMR data comparison of 5a and 5b (∆δ 5a -5b , Figure S7) determined the absolute configurations of (4S, 7S, 9R) in 5. Finally, the configuration of 5 was unambiguously confirmed to be (4S, 7S, 9R, and 10S) by X-ray analysis combined with the modified version of Mosher's method.
Compound 6 was contained as a colorless amorphous powder. It has the same molecular formula (C 12 H 20 O 4 ) as 5 according to its HRESIMS (m/z 251.1260 for [M+Na] + ) and 13 C NMR data. One dimensional (1D) NMR data (Tables 1 and 2) combined with a 2D NMR analysis ( Figure S3) revealed 6 Mar. Drugs 2020, 18, 18 7 of 15 to have the same planar structure as 5. Compound 6 was deduced to be a stereoisomer of 5 because of their different retention times when eluted by the same HPLC program and was named letendronol B. Compound 6 was contained as a colorless amorphous powder. It has the same molecular formula (C12H20O4) as 5 according to its HRESIMS (m/z 251.1260 for [M+Na] + ) and 13 C NMR data. One dimensional (1D) NMR data (Tables 1 and 2) combined with a 2D NMR analysis ( Figure S3) revealed 6 to have the same planar structure as 5. Compound 6 was deduced to be a stereoisomer of 5 because of their different retention times when eluted by the same HPLC program and was named letendronol B. Figure 5. The X-ray crystal structure of 5 (the thermal ellipsoid was 30%). Compounds 6 and 5 contained different a configuration at C-4, which was assigned by the modified version of Mosher's method. Treatment of 6 with (R)-or (S)-MTPA-Cl yielded the tri-(S/R) MTPA esters (6a/6b). The result of the application of the modified version of Mosher's method (Δ6a-6b, Figure S7) revealed 7S and 9R configurations, identical to 5, and a 4R configuration different from 5, in 6. H-9 and H-10 were also assigned to different orientations according to the broad singlet peaks for H-9 in 6a (δH 5.11, br s) and 6b (δH 4.78, br s) ( Figure S110), similar to those in 5a (δH 5.19, br s) and 5b (δH 4.76, br s) ( Figures S98). Then, the absolute configuration of 6 was determined to be (4R, 7S, 9R, and 10S).
The known compound 8 was determined to be xylarinol by comparison of its NMR data with the reported literature [15].  Figure S7) revealed 7S and 9R configurations, identical to 5, and a 4R configuration different from 5, in 6. H-9 and H-10 were also assigned to different orientations according to the broad singlet peaks for H-9 in 6a (δ H 5.11, br s) and 6b (δ H 4.78, br s) ( Figure S110), similar to those in 5a (δ H 5.19, br s) and 5b (δ H 4.76, br s) ( Figure S98). Then, the absolute configuration of 6 was determined to be (4R, 7S, 9R, and 10S).

General Experimental Procedures
Optical rotations were measured on Rudolph research analytical AUTOPOL I. The ultraviolet spectra were obtained from a Shimadzu UV-1800 spectrophotometer using MeOH as the solvent. Electronic circular dichroism (ECD) spectra were obtained on a JASCO J-1500 circular dichroism. The infrared (IR) spectra were acquired from a Bruker Vector 22. NMR spectra were recorded on a Bruker AVIII 500 MHz and JEOL 600 Hz, using TMS as the internal standard. HRESIMS spectra were obtained from an Agilent 6224 TOF LC/MS. Analytical HPLC was performed on an Agilent 1260 system using a C18 (Cosmosil, 5 μm, 4.6 × 250 mm) column. The column chromatography was carried out using Silica gel (200-300 mesh, Qing Dao Hai Yang Chemical Group Co., Qing Dao, China).

Fungal Material and Fermentation
The Letendraea sp. was isolated from the gut of a crab found on Zhairuoshan Island (N20.2920, E122.5), Zhoushan, Zhejiang Province, China, in August 2015. The fungus was determined to be Letendraea sp. by 26s rDNA sequence analysis (GenBank accession no. MK743951), and produced a sporulating, white-colored culture when growing on potato-dextrose agar (PDA). The strain was deposited (NO. 5XNZ4-2) at the Institute of Marine Biology of Zhejiang University. The strain was static cultured at 28 °C for 30 days in 500 mL Erlenmeyer flasks (100 × 200 mL, a total of 20 L) each containing 200 mL of 1/2 PDB liquid media (100 g of potato extract, 17 g of artificial sea salt, and 10 g of dextrose in 1 L pure water). The known compound 8 was determined to be xylarinol by comparison of its NMR data with the reported literature [15].
Compounds 1, 2, 3, and 5 were evaluated for their inhibitory activities against Lipopolysaccharide (LPS)-activated NO production in RAW264.7 [16], and all tested compounds showed no cytotoxicity to macrophage cells at the concentration of 100 µM and 50 µM by the MTT method. Compound 2 showed weak anti-inflammatory activity with an IC 50 value of 86 µM. Hydrocortisone was used as a positive control with the IC 50 value of 22.4 µM. The other compounds showed no anti-inflammatory activity with IC 50 values of >100 µM. All compounds were assayed for their cytotoxicity against the prostate cancer PC3 cell line by the MTT method [17]. Compounds 1-3, 5, 6, and 8 were also assayed for antibacterial activity against Staphylococcus aureus, Staphylococcus epidermidis, Escherichia coli, Miconia albicans and Pseudomonas aeruginosa using the Kirby−Bauer Diffusion method. Regrettably, none of these compounds showed any activities in the cytotoxicity and antibacterial tests.

General Experimental Procedures
Optical rotations were measured on Rudolph research analytical AUTOPOL I. The ultraviolet spectra were obtained from a Shimadzu UV-1800 spectrophotometer using MeOH as the solvent. Electronic circular dichroism (ECD) spectra were obtained on a JASCO J-1500 circular dichroism. The infrared (IR) spectra were acquired from a Bruker Vector 22. NMR spectra were recorded on a Bruker AVIII 500 MHz and JEOL 600 Hz, using TMS as the internal standard. HRESIMS spectra were obtained from an Agilent 6224 TOF LC/MS. Analytical HPLC was performed on an Agilent 1260 system using a C18 (Cosmosil, 5 µm, 4.6 × 250 mm) column. The column chromatography was carried out using Silica gel (200-300 mesh, Qing Dao Hai Yang Chemical Group Co., Qing Dao, China).

Fungal Material and Fermentation
The Letendraea sp. was isolated from the gut of a crab found on Zhairuoshan Island (N20.2920, E122.5), Zhoushan, Zhejiang Province, China, in August 2015. The fungus was determined to be Letendraea sp. by 26s rDNA sequence analysis (GenBank accession no. MK743951), and produced a sporulating, white-colored culture when growing on potato-dextrose agar (PDA). The strain was deposited (NO. 5XNZ4-2) at the Institute of Marine Biology of Zhejiang University. The strain was static cultured at 28 • C for 30 days in 500 mL Erlenmeyer flasks (100 × 200 mL, a total of 20 L) each containing 200 mL of 1/2 PDB liquid media (100 g of potato extract, 17 g of artificial sea salt, and 10 g of dextrose in 1 L pure water).

X-Ray Crystallographic Analysis
Compound 5 was obtained as colorless crystals from acetone-ethyl acetate (1:4) using the vapor-exchanged method [11,19,20]. X-ray single-crystal diffraction data for 5 were collected at 293 K on a Rigaku Oxford Diffraction instrument using Cu Kα (λ = 1.54184) radiation. The structures were solved by direct methods using Olex2 software [21], and the non-hydrogen atoms were refined isotropically with SHELXL-2014 [22] using a full-matrix least squares procedure based on

NO Inhibition Assay
Mouse monocyte-macrophages RAW 264.7 (ATCC TIB-71) were purchased from the Chinese Academy of Science. DMEM medium, penicillin, streptomycin, and fetal bovine serum were purchased from Invitrogen (N.Y., USA). Lipopolysaccharide (LPS), DMSO, 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-2H-tetrazolium bromide (MTT), and hydrocortisone were obtained from Sigma Co. RAW 264.7 cells were maintained in DMEM medium supplemented with penicillin, streptomycin (both 100 U/mL), and 10% heat-inactivated fetal bovine serum at 37 • C in a humidified incubator with 5% CO 2 and 95% air. RAW264.7 cells were passaged until they attained confluence. The cell concentration was adjusted to 5 × 10 5 cells/mL with 10% FBS DMEM medium, and 100 µL was seeded in each well of a 96-well plate that was maintained at 37 • C overnight. The supernatant was discarded, and the cells except for the blank control groups were treated with 100 µL of LPS DMEM (the concentration of LPS was 2 µg/mL or 5 µg/mL). After 2 h of incubation, blank and LPS model groups were treated with 100 µL of DMEM medium and the others were treated with DMEM medium containing various concentrations of test compounds for 24 h, respectively [16].
As a parameter of NO release, the nitrite concentration was measured in the supernatant of RAW 264.7 cells by the Griess reaction [16]. The inhibition rate was calculated and plotted versus test concentrations to afford the IC 50 .

Cytotoxic Assay
The cytotoxicity of all of the compounds was measured by the MTT assay against the prostate cancer PC3 cell line. Tumor cell lines were seeded in 96-well plates (4000 per well in 100 µL). After 24 h of incubation, cells were treated with gradient concentrations (100 µM, 50 µM, 25 µM, 12.5 µM, 6.25 µM, 3.125 µM) for another 72 h. Afterwards, MTT solution (5.0 mg/mL in RPMI-1640 media, Sigma, St. Louis, MO, USA) was added (20 µL/well) and then plates were incubated for another 4 h at 37 • C. The compounds were dissolved in DMSO and a cell growth inhibition assay was performed as reported previously. The growth inhibitory abilities of the compounds were calculated and expressed using the IC 50 value by the Dose-Effect Analysis software. Doxorubicin (ADR) was used as a positive control.

Antibacterial Assay
The antibacterial activities of all compounds except for 4 and 7 were evaluated with S. aureus, S. epidermidis, E. coli, M. albicans, and P. aeruginosa using the Kirby−Bauer Diffusion method. Gentamicin, vancomycin, ampicillin, amphotericin B, and gentamicin served as the positive drugs, respectively.
The stock solution of compounds 1, 2, 3, 5, and 8 was prepared in methanol with a series of concentrations (25, 5.0, 1, and 0.2 mg/mL). The stock solution of compound 6 was prepared in methanol with a series of concentrations (5.0, 1, and 0.2 mg/mL). The positive control drugs gentamicin, vancomycin, and ampicillin were dissolved in purified water at a concentration of 500 ug/mL. Amphotericin B was dissolved in DMSO at a concentration of 100 ug/mL.
S. aureus, S. epidermidis, E. coli, M. albicans, and P. aeruginosa were activated in nutrient broth at 37 • C for 6~8 h with 200 rpm. The culture broth was poured on the surface of Malt Extract Broth (MEB) Agar (20 g MEB and 20 g agar in 1 L of water) plates inside the cabinet, making sure that there was a uniform distribution of cells on the plates. The location of the negative control (methanol), the positive drugs, and different contents of tested compounds (100 µg, 20 µg, 4 µg, and 0.8 µg for 1, 2, 3, 5, and 8; 20 µg, 4 µg, and 0.8 µg for 6) was labeled on the agar side of the plate and outside the lid. The discs were arranged so as to mimic their positions outside the lid and 4 µL of the corresponding concentrations of the tested compounds and methanol was added to the discs. The contents of the positive drugs gentamicin, ampicillin, and amphotericin B were 3 ug, 1.5 ug, and 0.35 ug, respectively. The discs were transferred from the lid to the agar plate using sterile forceps. Finally, the plates were placed in an incubator agar side up at 37 • C for 12 h and the inhibition zone was observed.

Conclusions
With the aim of discovering structurally novel secondary metabolites, seven new polyketides (phomopsiketones D-G (1-4) and letendronols A-C (5-7)) were isolated from the marine endophytic fungus Letendraea sp. in 1/2 PDB medium. The structures and absolute configurations of the new compounds were sufficiently elucidated through HRESIMS, NMR, single-crystal X-ray diffraction, ECD calculation, and a modified version of Mosher's method. Compound 2 showed weak anti-inflammatory activity.
Author Contributions: Y.X. performed the experiments for the isolation, structure determination, and antimicrobial evaluation, and prepared the manuscript; H.L. performed the NO inhibition assay; Y.J. contributed to part of the structure determination; R.H. contributed to the preparation of the fermentation; W.D. performed the cytotoxic assay; T.Y. contributed to part of the isolation; P.W. and D.Z. conducted the isolation of strain 5XNZ4-2 and jointly supervised the research; J.X. supervised the research work and revised the manuscript. All authors have read and agreed to the published version of the manuscript.