Drimane-type sesquiterpenoids and their anti-inflammatory evaluation from Pyrrhoderma noxium HNNU0524

Abstract Chemical investigation of a culture broth from the marine-derived fungus Pyrrhoderma noxium HNNU0524 yielded two new compounds including a drimane-type sesquiterpenoid named pyrrnoxin A (1) and a benzoic acid derivative, pyrrnoxin B (5), together with three related known analogues (2–4). The chemical structures of 1 and 5 were determined by detailed analysis of spectroscopic data, single-crystal X-ray crystallography, quantum mechanics-based DP4+ and ECD calculations. Compounds 2 and 3 moderately inhibited NO production of lipopolysaccharide-induced microglia cells BV2 with IC50 values of 26.6 and 60.5 µM, respectively. Graphical Abstract


Introduction
Drimane-type sesquiterpenes are a large group of natural products with unique C15 bicyclic skeletons, which are found mainly in plants (Rajia et al. 2011;Yoshinori et al. 2012), sponges (Li et al. 2009) and fungi (Du et al. 2022).These compounds have provoked much interest because of their abundant structural variants and attractive biological activities, which included antimicrobial (Intaraudom et al. 2019), anti-inflammatory (Chen et al. 2019), cytotoxic (Kwon et al. 2018) and neurotransmitter (Xu et al. 2020) bioactivities.The biosynthesis of these sesquiterpenes has been considered to involve drimenol cyclases in fungi (Huang et al. 2021).
The relative configuration of 1 was determined by a combination of NOESY analysis (Figure S3) and coupling constant values.In the NOESY spectrum of 1, the observed cross-peaks between H-5 (δ H 1.04) and both H-3 (δ H 3.18) and H-9 (δ H 2.27) indicated that they were in the same axially α-orientation and the two fused six-membered rings encompassing C-1 to C-10 preferred in chair conformations, while the decalin ring junction of 1 was trans.Additional NOESY correlations between H-6 ax (δ H 1.92) with Me-13 (δ H 0.89) and Me-15 (δ H 0.92) illustrated they were in a β-axial orientation.The coupling constants between H-6 ax and both H-5 and H-7 (J 5-6ax = 12.8 Hz, J 7-6ax = 2.4 Hz) confirmed H-6 ax was in a β-axial orientation and H-7 was in another equatorially α-orientation, which also readily indicated a formation of a 7 R,8S epoxy unit at C-7/C-8.Furthermore, the same biosynthetic pathway between 1 and inotolactone H (Zou et al. 2020) supported the assignment above.Based on this, the relative configuration of 1 was assigned as 3S*/5R*/7R*/8S*/9S*/10S*.The ECD spectrum of 1 was calculated, which showed that the ECD curve of the stereoisomer (3 R,5R,7R,8S,9S,1 0S)-1 matched well with the experimental one (Figure S4).Fortunately, crystals of 1 were also acquired after various attempts and the absolute stereo-structure of 1 was confirmed to be the same as the foregoing assignment by X-ray crystallography (Figure S5).Therefore, it was given a trivial name, pyrrnoxin A.
The molecular formula of pyrrnoxin B (5) was determined to be C 12 H 14 O 5 with six degrees of unsaturation on the basis of HRESIMS spectrum.The 1 H NMR spectrum exhibited 11 proton signals, including two para-aromatic protons at δ H 7.86 and 7.35 ppm, one oxymethine (δ H 4.04) and one doublet methyl (δ H 1.22).The 13 C NMR spectrum displayed 12 carbon signals for two carboxylic carbons (δ C 173.3, 167.7), two sets of para-aromatic methines (δ C 129.0, 128.5), two methines including one oxygenated group (δ C 71.1), one methine (δ C 40.0) and one methyl (δ C 17.3).These data suggested that 5 was a para-benzoic acid derivative.Analysis of the HMBC spectrum revealed correlations from H-2 (δ H 7.86  S2), which indicated a carboxylic acid group was substituted at C-7 and another long alkyl side chain bearing a terminal carboxylic acid group was located at C-4.Further observation of the COSY spectrum showing a cross-peak between H-2 and H-3, and a second sequential spin-spin system among H-12/H-8/H-9/H-10 (Figure S2), provided a strong evidence for the assignment as depicted (Figure S1).An initial attempt to establish the absolute configuration of C-9 by Mosher's method was restricted due to a small amount of sample.Therefore, the relative configuration of 5 was deduced using DP4+ probability analysis (Grimblat et al. 2015).The structure of 5 with two stereo-centers resulted in two possible candidate diasteroisomers (8 R*/9R*)-5 and (8 R*/9S*)-5, DP4+ probability analysis indicated that the relative configuration of 5 was inclined to be 8 R*/9R*, because the isomer (8 R*/9R*)-5 showed high probability (99.8%) relative to the other isomer (8 R*, 9S*)-5 (See Table S7).Finally, the absolute configuration of C-8 in 5 was assigned as R by comparison of its calculated and experimental ECD spectra, in which they had a similar trend (Figure S4), while the absolute configuration of C-9 was considered to be R. Thus, it was given the trivial name pyrrnoxin B.
Neuroinflammation is associated with activated microglia cells, and plays a crucial role in the pathophysiological process of various chronic brain diseases, such as Alzheimer's disease (AD) and Parkinson's disease.As an in vitro model, the measurement of NO production associated with lipopolysaccharide (LPS) induced inflammation of murine BV-2 microglial cells was performed.Prior to the bioassay, the cytotoxic effect of 1-5 against cell viability were assessed using by the MTT method, and all compounds showed no cytotoxicity with CC 50 values of >100μM.At nontoxic concentrations, compounds 2 and 3 exhibited moderate effects on reducing the LPS-induced NO production with IC 50 values of 26.6 and 60.5 μM, respectively, while the positive control L-NMMA showed antineuroinflammatory activity with a IC 50 value of 2.6 μM.
In conclusion, a new drimane-type sesquiterpenoid (1) and a new benzoic acid derivative (5), together with three known compounds (2-4) were obtained from the marine-derived fungus Pyrrhoderma noxium HNNU0524.These drimane-type sesquiterpenoid analogues were founded by fungi in the genera Pyrrhoderma for the first time.Compounds 2 and 3 exhibited moderated effects against the NO production on lipopolysaccharide (LPS)-induced microglia cells BV2.

General experimental procedures
IR spectra were recorded on a FT-IRNICOLET spectrophotometer, while UV spectra were collected on a Beckman DU 640 spectrophotometer, and NMR spectra were recorded on a Bruker AV spectrometer (400 MHz for 1 H and 100 MHz for 13 C).TMS was used as a reference.HRESIMS spectra were measured on a Waters Xevo G2QTOF spectrometer (Waters).Semi-preparative HPLC was performed on an Agilent 1260 LC series system with a diode array detector using an Agilent Eclipse XDB-C18 column (9.4 × 250 mm, 5 µm).Silica gels (Qing Dao Hai Yang Chemical Group Co.; were used for open column chromatography (CC).Pre-coated silica gel plates (Yan Tai Zi Fu Chemical Group Co.; G60, F-254) were used for thin-layer chromatography (TLC).

Biological material
The fungal strain HNNU 0524 was isolated from marine sediment collected from the South China Sea (115.35 65°E, 21.4753°N).The strain was identified as Pyrrhoderma noxium based on BLAST analysis of fungal ribosomal DNA sequence of internal transcribed spacers (GenBank accession OL774784), which exhibited 99.54% similarity with that of P. noxium strain KU194341.Strain HNNU 0524 was deposited in the Key Laboratory of Tropical Medicinal Resource Chemistry of the Ministry of Education, Hainan Normal University.

Fermentation and extraction
The fungal strain P. noxium HNNU 0524 was cultured in 125 replicate 1000 mL Erlenmeyer flasks each containing 400 mL of modified PDB liquid fermentation medium with added artificial sea salt 3%, under static incubation at 28 °C for 30 days.At the end of the fermentation period, the culture broth was separated from the mycelium by filtration.The culture filtrate was extracted with EtOAc solvent thrice.The mycelium cake was extracted under ultra-sonication in 2 L of acetone for 1 h.The acetone was removed in vacuo, and the remaining aqueous portion was extracted with an equal volume of EtOAc twice.Thus, a combined organic extract was afforded.

X-ray crystallographic analyses of 1
A white crystal of 1 was obtained in MeOH.The crystal data were recorded on an Oxford Diffraction Gemini E diffractometer with graphite-monochromated Cu Kα (λ = 1.5418Å) radiation (see Tables S2-S4).The structures were determined by direct methods using Olex2 software and refined using full-matrix least-squares difference Fourier techniques.Crystallographic information file of compound 1 with Cambridge Crystallographic Data Centre (CCDC) reference number 2189526 has been deposited at the CCDC, and can be obtained free of charge from the CCDC via https://www.ccdc.cam.ac.uk/deposit/.

Computational methods
Conformational searches of 1 and 5 for quantum calculations were performed using the MMFF94S force field by the Spartan'14 (Wave function Ind.) with low energetics from 0 to 10 kcal/mol.Geometric optimization was carried out in the gas phase at the B3LYP/6-31G(d) level.Frequency calculations were run at the same level to estimate their relative thermal free energies (ΔG) at 298.15K.The optimized conformations a the relative energy between 0 and 5 kcal/mol were selected for ECD calculations at the M06-2X/tzvp level (liquid phase) and simulated using SpecDis 1.70 (Bruhn et al. 2013).The GIAO NMR chemical shift calculations were performed at the mpw1pw91/6-31 + g(d,p) level paired with the PCM implicit solvation model.The equilibrium population of each conformer at 298.15 K was calculated from its relative free energies using Boltzmann statistics.The calculated NMR data were generated from the low-energy conformers according to the Boltzmann weighting of each conformer.

Bioassay for the inhibition of NO production
The bioassay for compounds to inhibit NO production was conducted by the method in the literature (Guo et al. 2021).BV-2 microglia cells were seeded in 96-well plates and then treated with or without the compound at different doses for 24 h.In each well, LPS (1 μg/mL) was added to induce inflammation.The NO production in the supernatant was quantified by the Griess reaction.The absorbance at 540 nm was measured in a microplate reader (Thermo Scientific).The NO concentration was calculated by a calibration curve.Experiments were performed in triplicate, and the data represent the mean ± SD of three independent experiments.