Highly Substituted Phenol Derivatives with Nitric Oxide Inhibitory Activities from the Deep-Sea-Derived Fungus Trichobotrys effuse FS524

Chemical investigation on EtOAc extract of the deep-sea-derived fungus Trichobotrys effuse FS524 resulted in the isolation of six new highly substituted phenol derivatives, trieffusols A–F (1–6), along with ten known relative analogs (7–16). Their structures with absolute configurations were extensively characterized on the basis of spectroscopic data analyses, single-crystal X-ray diffraction experiments, and electronic circular dichroism (ECD) calculations. Structurally, trieffusols A and B shared an unprecedented ploy-substituted 9-phenyl-hexahydroxanthone skeleton with an intriguing 6-6/6/6 tetracyclic fused ring system, which is often encountered as significant moieties in the pharmaceutical drugs but rarely discovered in natural products. In the screening towards their anti-inflammatory activities of 1–6, trieffusols C and D exhibited moderate inhibitory activities against nitric oxide (NO) production in LPS-induced RAW 264.7 macrophages with IC50 values ranging from 51.9 to 55.9 μM.


Introduction
Marine-derived fungi have emerged as one of the most promising strategic resources to search pharmacologically significant leads for drug discovery and have aroused widespread attention from natural product chemists, pharmacologists, as well as biosynthetic chemists, due to their structurally abundant and diverse secondary metabolites in recent years [1,2]. Over the past decades, the research articles on marine natural products (MNPs) have surged dramatically, bringing about a lot of conspicuous natural products with novel chemical scaffolds and unique biological functional arrays [3][4][5][6][7]. In terms of pharmacological research, secondary metabolites derived from marine fungi are increasingly recognized as important sources of biologically meaningful natural products [8,9]. These MNPs have exhibited a wide range of biological activities such as anti-cancer [10], fungicidal [11], pro-angiogenic [12], anti-lymphangiogenic [13], and osteoclast differentiation inhibitory activities [14]. Therefore, in-depth chemical research on the MNPs would pave the way to providing potential model structures and precursor drugs for new drug developments.

Structure Elucidation
Compound 1, a colorless crystal, was given the molecular formula as C23H26O8 determined by the HRESIMS cationic peak at m/z 431.1696 [M + H] + (calcd 431.1700), which corresponded to eleven degrees of unsaturation. The IR spectrum of 1 logically revealed the presence of hydroxyl and carbonyl functional groups through the characteristic resonance absorptions at 3443 cm −1 and 1668 cm −1 , respectively. The further inspection of its 1 H NMR spectrum (Table 1)

Structure Elucidation
Compound 1, a colorless crystal, was given the molecular formula as C 23 3 -15)]. The 13 C NMR spectrum combined with HSQC data of 1 resolved 23 carbon resonances attributable to two methyls, four methylenes, seven methines, and ten quaternary carbons containing two keto-carbonyl ones. The aforementioned aromatic ring and functionalities logically accounted for eight degrees of unsaturation, and the remaining three degrees of unsaturation necessitated that 1 should possess an additional tricyclic ring system. The chemo-logical construction of the planar structure for compound 1 featuring a tetracyclic 6-6/6/6 ring system was elucidated by the analysis of 2D NMR spectra ( Similarly, the establishment of the other cyclohexenone moiety (ring C) was confirmed by the key HMBC correlations from H2-10 to C-9, C-11, C-12, and C-16, H-11 to C-13, H2-14 to C-11 and C-13, H3-15 to C-12 as well as fragments c and d. Moreover, considering the remaining one degree of unsaturation and chemical shift of C-6 (δC 162.8) and C-9 (δC 161.8), we suspected that an oxygen atom should be connected between C-6 and C-9 with the formation of an oxygen bridge, which finally constructed the core pentasubstituted-4H-pyran skeleton (ring B). The aforementioned deduction was successfully reconfirmed by the informative HMBC correlations from H-17 to C-1, C-2, C-6, C-9, C-13, C-16, C-18, C-19, and C-23. Therefore, the planar structure of 1 was elucidated as a phenol-polyketone derivative consisting of an intriguing natural rarely-encountered 6-6/6/6 fused-ring system and given the trivial name "trieffusol A".
However, the high overlap of critical proton signals for H-4 and H-11, H2-7 and H2-14, H2-5, and H2-10, in conjunction with H3-8 and H3-15, made further construction of the relative configurations for the chiral genetic centers C-3 and C-4, as well as C-11 and C-12 in the cyclohexenone rings A and C, become a challenging issue. Moreover, all of these aforementioned carbons were far away from the central C-17 chiral genetic center, which would further give rise to two pairs of alternative diastereomeric configurations. Therefore, the assignment of the relative and absolute configurations of compound 1 through NMR and CD spectra seemed to be bleak. In order to corroborate the above structural deduction and establish absolute stereochemistry of 1, we attempted to get X-ray crystals in the methanol/water (30:1) system. Fortunately, the single crystals with good quality were obtained and subjected to an X-ray diffraction experiment with Cu Kα radiation. The crystal data ( Figure 3) not only confirmed our deduction about the planar structure of 1 but also unambiguously established its absolute configuration as 3R,4S,11S,12S,17R. Therefore, the complete structure with the absolute configuration of compound 1 was finally established and given the trivial name "trieffusol A", which possesses an unprecedented ploy-substituted 9-phenyl-hexahydroxanthone skeleton with an intriguing 6-6/6/6 tetracyclic fused ring system. Similarly, the establishment of the other cyclohexenone moiety (ring C) was confirmed by the key HMBC correlations from H 2 -10 to C-9, C-11, C-12, and C-16, H-11 to C-13, H 2 -14 to C-11 and C-13, H 3 -15 to C-12 as well as fragments c and d. Moreover, considering the remaining one degree of unsaturation and chemical shift of C-6 (δ C 162.8) and C-9 (δ C 161.8), we suspected that an oxygen atom should be connected between C-6 and C-9 with the formation of an oxygen bridge, which finally constructed the core pentasubstituted-4H-pyran skeleton (ring B). The aforementioned deduction was successfully reconfirmed by the informative HMBC correlations from H-17 to C-1, C-2, C-6, C-9, C-13, C-16, C-18, C-19, and C-23. Therefore, the planar structure of 1 was elucidated as a phenol-polyketone derivative consisting of an intriguing natural rarely-encountered 6-6/6/6 fused-ring system and given the trivial name "trieffusol A".
However, the high overlap of critical proton signals for H-4 and H-11, H 2 -7 and H 2 -14, H 2 -5, and H 2 -10, in conjunction with H 3 -8 and H 3 -15, made further construction of the relative configurations for the chiral genetic centers C-3 and C-4, as well as C-11 and C-12 in the cyclohexenone rings A and C, become a challenging issue. Moreover, all of these aforementioned carbons were far away from the central C-17 chiral genetic center, which would further give rise to two pairs of alternative diastereomeric configurations. Therefore, the assignment of the relative and absolute configurations of compound 1 through NMR and CD spectra seemed to be bleak. In order to corroborate the above structural deduction and establish absolute stereochemistry of 1, we attempted to get X-ray crystals in the methanol/water (30:1) system. Fortunately, the single crystals with good quality were obtained and subjected to an X-ray diffraction experiment with Cu Kα radiation. The crystal data ( Figure 3) not only confirmed our deduction about the planar structure of 1 but also unambiguously established its absolute configuration as 3R,4S,11S,12S,17R. Therefore, the complete structure with the absolute configuration of compound 1 was finally established and given the trivial name "trieffusol A", which possesses an unprecedented ploy-substituted 9-phenyl-hexahydroxanthone skeleton with an intriguing 6-6/6/6 tetracyclic fused ring system. Trieffusol B was also obtained as a colorless crystal and had the same molecular formula C23H26O8 as that of 1 based on the negative mode HERSIMS (m/z 429.1554 [M − H] -, calcd 429.1555), indicating the presence of eleven degrees of hydrogen deficiency. Its 1 H and 13 C NMR data closely resembled those of 1, except for chemical shift changes at C-4, C-7, C-8, C-11, C-14, and C-15. A comprehensive analysis of the 1D and 2D NMR data deduced that compounds 1 and 2 share the same planar structures, indicating that these two compounds should be a pair of diastereoisomers sharing the same ploy-substituted 9-phenyl-hexahydroxanthone skeleton with an intriguing 6-6/6/6 tetracyclic fused ring system. The aforementioned deduction was further substantiated by the X-ray single-crystallographic analysis (Figure 4), which finally clarified the absolute configuration of compound 2 as 3S,4S,11S,12R,17S. Trieffusol C was purified as a brown oil. Its molecular formula was determined as C16H20O5 based on the protonated molecule peak at m/z 293.1385 [M + H] + (calcd 293.1384) by HRESIMS, which requires seven degrees of unsaturation. The 1 H NMR data of 3 (Table 2)   A comprehensive analysis of the 1D and 2D NMR data deduced that compounds 1 and 2 share the same planar structures, indicating that these two compounds should be a pair of diastereoisomers sharing the same ploy-substituted 9-phenyl-hexahydroxanthone skeleton with an intriguing 6-6/6/6 tetracyclic fused ring system. The aforementioned deduction was further substantiated by the X-ray single-crystallographic analysis (Figure 4), which finally clarified the absolute configuration of compound 2 as 3S,4S,11S,12R,17S. Trieffusol B was also obtained as a colorless crystal and had the same molecular formula C23H26O8 as that of 1 based on the negative mode HERSIMS (m/z 429.1554 [M − H] -, calcd 429.1555), indicating the presence of eleven degrees of hydrogen deficiency. Its 1 H and 13 C NMR data closely resembled those of 1, except for chemical shift changes at C-4, C-7, C-8, C-11, C-14, and C-15. A comprehensive analysis of the 1D and 2D NMR data deduced that compounds 1 and 2 share the same planar structures, indicating that these two compounds should be a pair of diastereoisomers sharing the same ploy-substituted 9-phenyl-hexahydroxanthone skeleton with an intriguing 6-6/6/6 tetracyclic fused ring system. The aforementioned deduction was further substantiated by the X-ray single-crystallographic analysis (Figure 4), which finally clarified the absolute configuration of compound 2 as 3S,4S,11S,12R,17S.  Trieffusol C was purified as a brown oil. Its molecular formula was determined as C 16   Construction of the planar structure for 3 was accomplished by analysis of its 2D NMR data. Firstly, the presence of a para-substituted phenyl moiety was confirmed by the HMBC correlations from H-3/5 to C-1 and C-4, H-2/6 to C-1 and C-4, together with 1 H-1 H COSY correlations of H-2/6/H-3/5. Secondly, the HMBC correlations from H-10 to C-9, C-11, C-12, and C-14, H-13 to C-12 and C-15, H 2 -14 to C-9, C-10, C-12, and C-13, H 2 -15 to C-11 and C-12, H 3 -16 to C-12 and C-15, as well as 1 H-1 H COSY correlations of H-13/H 2 -14 and H 2 -15/H 3 -16 strongly indicated the existence of the trisubstituted cyclohex-2-en-1-one moiety. Finally, the connections of the two independent phenyl and cyclohex-2-en-1-one fragments through the linkage of C-1/C-7/C-8/O/C-9 were supported by the HMBC correlations from H-2/6 to C-7, H 2 -7 to C-1, C-2, and C-6, H 2 -8 to C-1 and C-9 as well as 1 H-1 H COSY fragment H 2 -7/H 2 -8. Hence, the planar structure of 3 was successfully constructed, as shown in Figure 1.
The relative configuration of 3 was assessed by the NOESY correlation, and the absence of the critical NOE correlation of H-13/H 2 -15 tentatively suggested that these two protons should orientate oppositely. The absolute stereochemistry for chiral genetic centers of C-12 and C-13 in compound 3 was determined on the basis of the comparison of experimental and the quantum mechanically calculated electric circular dichroism (ECD) data by using the time-dependent density functional theory (TDDFT) at the B3LYP/6-31+G (d,p) level in MeOH. Satisfactorily, the calculated ECD spectrum of 12R,13S-3 ( Figure 5) matched well with that of the experimental one, with a positive Cotton effect at 255 nm and a negative one at 300 nm, respectively, which unambiguously clarified the absolute configuration of 3 to be 12R,13S. The 1 H-1 H COSY correlations of H2-10/H3-11 and the HMBC correlations from H-5 to C-1, C-3, C-4, and C-6, H2-10 to C-2, C-3, and C-4, H3-11 to C-3 and C-10 led to the establishment of the pentasubstituted benzene ring. Besides, the sequential HMBC correlations from H2-8 to C-7, C-9, and C-12, H3-12 to C-7 and C-8 along with the 1 H-1 H COSY correlations of H2-8/H-7/H3-12 suggested the existence of a tetrasubstituted tetrahydro-4H-pyran-4-one scaffold. Therefore, the gross structure of trieffusol D was established undoubtedly.
As for the absolute configuration, the specific optical rotation of 4 was close to zero, which logically suggested that it might exist as a pair of enantiomers. The further chiral-phase separation via chiral HPLC yielded two optically pure enantiomers 4a and 4b, respectively. Subsequently, the theoretical ECD spectra for 4a and 4b ( Figure 6) were calculated by using the time-dependent density functional theory (TDDFT) at the B3LYP/6-31+G (d,p) level in MeOH. As a result, the calculated ECD spectra matched well with those of the experimental ones. Therefore, the absolute configurations of 4a and 4b were finally assigned as R and S, respectively.  The 1 H-1 H COSY correlations of H 2 -10/H 3 -11 and the HMBC correlations from H-5 to C-1, C-3, C-4, and C-6, H 2 -10 to C-2, C-3, and C-4, H 3 -11 to C-3 and C-10 led to the establishment of the pentasubstituted benzene ring. Besides, the sequential HMBC correlations from H 2 -8 to C-7, C-9, and C-12, H 3 -12 to C-7 and C-8 along with the 1 H-1 H COSY correlations of H 2 -8/H-7/H 3 -12 suggested the existence of a tetrasubstituted tetrahydro-4H-pyran-4-one scaffold. Therefore, the gross structure of trieffusol D was established undoubtedly.
As for the absolute configuration, the specific optical rotation of 4 was close to zero, which logically suggested that it might exist as a pair of enantiomers. The further chiral-phase separation via chiral HPLC yielded two optically pure enantiomers 4a and 4b, respectively. Subsequently, the theoretical ECD spectra for 4a and 4b ( Figure 6) were calculated by using the time-dependent density functional theory (TDDFT) at the B3LYP/6-31+G (d,p) level in MeOH. As a result, the calculated ECD spectra matched well with those of the experimental ones. Therefore, the absolute configurations of 4a and 4b were finally assigned as R and S, respectively. The 1 H-1 H COSY correlations of H2-10/H3-11 and the HMBC correlations from H-5 to C-1, C-3, C-4, and C-6, H2-10 to C-2, C-3, and C-4, H3-11 to C-3 and C-10 led to the establishment of the pentasubstituted benzene ring. Besides, the sequential HMBC correlations from H2-8 to C-7, C-9, and C-12, H3-12 to C-7 and C-8 along with the 1 H-1 H COSY correlations of H2-8/H-7/H3-12 suggested the existence of a tetrasubstituted tetrahydro-4H-pyran-4-one scaffold. Therefore, the gross structure of trieffusol D was established undoubtedly.
As for the absolute configuration, the specific optical rotation of 4 was close to zero, which logically suggested that it might exist as a pair of enantiomers. The further chiral-phase separation via chiral HPLC yielded two optically pure enantiomers 4a and 4b, respectively. Subsequently, the theoretical ECD spectra for 4a and 4b ( Figure 6) were calculated by using the time-dependent density functional theory (TDDFT) at the B3LYP/6-31+G (d,p) level in MeOH. As a result, the calculated ECD spectra matched well with those of the experimental ones. Therefore, the absolute configurations of 4a and 4b were finally assigned as R and S, respectively.   (Table 3) and 2D NMR data of 5 showed great similarity to these of 4-hydroxy-6-methoxy-5-(1 -oxobutyl)benzo[b]dihydrofuran [18]. The main difference was that the two methylenes at C-10 and C-11 positions in the known compound were oxidized to be a disubstituted double bond in 5. This conclusion was further verified by the key HMBC correlations from H-10 to C-9 and C-12, H-11 to C-9 and C-11, H 3 -12 to C-10 and C-11, together with the 1 H-1 H COSY correlations of H-10/H-11/H 3 -12. Besides, the other structural identification details, as shown in Figure 2, could also support this conclusion. Moreover, the coupling constant (J = 15.1 Hz) between H-10 and H-11 obviously illustrated the configuration of the disubstituted double bond as an E-configuration. Therefore, the structure of 5 was elucidated unambiguously, as depicted in Figure 1.  3 -11)]. The 13 C NMR data and HSQC spectrum of 6 showed 12 carbons, which were assigned to one methyl, four methylenes (two oxygenated ones), one aromatic carbon, five quaternary carbons (one oxygenated one), and one ester carbonyl functionality. The key HMBC correlations from H-2 to C-4, C-6, and C-7, H 2 -8 to C-3, C-4, C-5, and C-7 indicated the establishment of the isobenzofuran-1(3H)-one moiety. Furthermore, the attachments of the C-9 at C-1 and C-12 at C-6 were confirmed by the HMBC correlations from H 2 -9 to C-1, C-2, and C-6, H-2 to C-9, H 2 -12 to C-1, C-5, and C-6, along with the 1 H-1 H COSY correlations of H 2 -9/H 2 -10/H 3 -11. Thus, the structure of 6 was defined, as shown in

Biological Activity
Compounds 1-6 were evaluated for their inhibition effect of NO production in the lipopolysaccharide (LPS)-induced mouse macrophages. As shown in Table 4, compounds 3 and 4 exhibited the inhibitory activities with IC 50 values ranging from 51.9 to 55.9 µM, comparable to that of the positive control aminoguanidine (IC 50 : 24.8 µM). At the same time, both of them showed no cytotoxicities against macrophages, of which IC 50 values were all greater than 200 µM. Table 4. Inhibitory effects of 1-6 on the NO production and the cytotoxicity.

Fungal Material
The strain FS524 used in this work was isolated from a sediment sample, which was collected at the depth of 1428 m in the South China Sea (110 • 59 04 E, 18 • 00 47 N) in June 2017. The sequence data for this strain have been submitted to the GenBank under accession no. MN545626. By using BLAST (nucleotide sequence comparison program) to search the GenBank database, FS524 has 99.8% similarity to Trichobotrys effuse DFFSCS021 (accession no. JX156367). The strain was preserved at the Guangdong Provincial Key Laboratory of Microbial Culture Collection and Application, Guangdong Institute of Microbiology.

Fermentation and Extraction
The marine fungus T. effuse FS524 was cultured on potato dextrose agar (PDA) at 28 • C for 7 days to prepare the seed culture, and then inoculated into flasks (3 L) each containing 9 g sea salts, 250 g of rice, and 300 mL of water. After that, all flasks were incubated at 28 • C for one month and extracted repeatedly with EtOAc. After evaporation of the solvent, a dark brown solid extract (67.3 g) was obtained. The crude extract was fractionated by silica gel column chromatography (100-200 mesh) with two gradient systems of increasing polarity (petroleum ether-EtOAc, 30:1→1:1; CH 2 Cl 2 /CH 3 OH, 10:1→1:1) to furnish nine fractions (A-I).

X-ray Crytallographic Data of Compounds 1 and 2
The single-crystal X-ray diffraction data for compounds 1 and 2 were collected on an Agilent Xcalibur Nova single-crystal diffractometer using CuKα radiation at 293 and 100 K, respectively. The crystal structures were refined by full-matrix least-squares calculation (for details see X-ray crystallographic analysis, Tables S1 and S2 in the supporting information). Crystallographic data have been deposited at the Cambridge Crystallographic Data Center with the deposition number of CDCC 1974673 for 1 and CDCC 1974674 for 2, respectively. Copies of these data can be obtained free of charge via www.ccdc.cam.au.ck/conts/retrieving.html.

Quantum Chemical Calculations
Merck molecular force field (MMFF) and DFT/TD-DFT calculations were carried out with the Spartan'14 software (Wavefunction Inc., Irvine, CA, USA) and the Gaussian 09 program, respectively [24]. Conformers within the 10 kcal mol −1 energy window were generated and optimized using DFT calculations at the B3LYP/6-31+G (d,p) level. Frequency calculations were performed at the same level to confirm that each optimized conformer was true minimum and to estimate their relative thermal free energy (∆G) at 298.15 K. Conformers with the Boltzmann distribution over 5% were chosen for ECD calculations in methanol at the B3LYP/6-311+G (d,p) level. Solvent effects were taken into consideration using the self-consistent reaction field (SCRF) method with the polarizable continuum model (PCM) [25]. Details of the individual conformers are provided in the supporting information. The ECD spectrum was generated by the SpecDis program [26] using a Gaussian band shape with 0.26 eV exponential half-width from dipole-length dipolar and rotational strengths.

Nitric Oxide Inhibitory Activities Assay
Compounds 1-6 were evaluated for the inhibitory activity of nitric oxide (NO) production in lipopolysaccharide (LPS)-induced RAW 246.7 mouse macrophages [27]. The cells (180 µL) with a density of 5 × 10 5 cells/mL of media on a 96-well plate were put under 37 • C at a 5% CO 2 condition. After a 24 h preincubation, the seeded cells were treated with gradient dilutions of 1-6 with a maximum concentration of 100 µM, followed by stimulation with LPS (1 µg/mL) for 24 h. Then 50 µL cell culture supernatant solution was moved to a new plate that contained NO detection Griess A (50 µL) and Griess B (50 µL). Finally, the absorbance was measured at 540 nm. Aminoguanidine was used as a positive control and all data were obtained in triplicate. The viability of RAW264.7 cells was evaluated according to the SRB method simultaneously to exclude the interference of the cytotoxicity of 1-6. The RAW264.7 cells were purchased from the Cell Bank of the Chinese Academy of Sciences.