Hepialiamides A–C: Aminated Fusaric Acid Derivatives and Related Metabolites with Anti-Inflammatory Activity from the Deep-Sea-Derived Fungus Samsoniella hepiali W7

A systematic investigation combined with a Global Natural Products Social (GNPS) molecular networking approach, was conducted on the metabolites of the deep-sea-derived fungus Samsoniella hepiali W7, leading to the isolation of three new fusaric acid derivatives, hepialiamides A–C (1–3) and one novel hybrid polyketide hepialide (4), together with 18 known miscellaneous compounds (5–22). The structures of the new compounds were elucidated through detailed spectroscopic analysis. as well as TD-DFT-based ECD calculation. All isolates were tested for anti-inflammatory activity in vitro. Under a concentration of 1 µM, compounds 8, 11, 13, 21, and 22 showed potent inhibitory activity against nitric oxide production in lipopolysaccharide (LPS)-activated BV-2 microglia cells, with inhibition rates of 34.2%, 30.7%, 32.9%, 38.6%, and 58.2%, respectively. Of particularly note is compound 22, which exhibited the most remarkable inhibitory activity, with an IC50 value of 426.2 nM.


Introduction
Secondary metabolites produced by marine-derived fungi have been proven to possess a broad spectrum of bioactivities, such as cytotoxicity, antibacterial, antioxidant, antimalarial, anti-inflammatory, and antiviral properties and have been recognized as significant chemical entities for drug discovery [1][2][3][4][5][6][7][8].However, continuous investigations on secondary metabolites from marine-derived fungi have led to a high frequency of rediscovery of known compounds, for which, dereplication becomes critical in microbial biodiscovery.Mass spectrometry-based GNPS molecular networking (https://gnps.ucsd.edu,accessed on 1 January 2016) is a strategy that has been proven to be a powerful and promising tool for dereplication.GNPS uses an untargeted metabolomics approach that powerfully processes tandem mass spectrometry (MS/MS) fragmentation data.It is a vector-based workflow that calculates cosine scores (between 0 and 1) to determine the degree of similarity between MS 2 fragments.These fragment ions (nodes) are then organized into relational networks depending on their similarity [9][10][11].GNPS has been widely used to identify of similarity between MS 2 fragments.These fragment ions (nodes) are then organized into relational networks depending on their similarity [9][10][11].GNPS has been widely used to identify known compounds and potential new analogs, greatly speeding up the dereplication and structure-based discovery of natural products [12][13][14].
Fusaric acid (FA), featuring a picolinic acid core, is a well-known mycotoxin commonly found in the metabolites of numerous Fusarium species.The biosynthesis of FA and its related analogues has attracted attention since the last century due to the intriguing pyridine moiety, which has been proven to start from an aspartic acid precursor and malonyl-CoA [15][16][17][18].However, naturally occurring amidated derivatives of FA have not been well-described yet in terms of their structural diversity and biological activity, with only two examples reported so far, including atransfusarin from the endophytic fungus Alternaria atrans MP-7 [19] and 2-(4-butylpicolinamide) acetic acid from Fusarium fujikuroi [20].Atransfusarin was documented with mild anti-fungi activity while 2-(4-butylpicolinamide) acetic acid was inactive against the tested pathogens.
As part of our ongoing exploration of structurally novel and bioactive secondary metabolites from marine microbes [21][22][23][24], we performed a systematic investigation, aided by a GNPS molecular networking approach, on the crude extract of the culture of fungus Samsoniella epialid W7, isolated from the sulfide at a depth of 3073 m in the South Atlantic.Prior to isolation, the GNPS network of the extract revealed a big molecular cluster within which none of the nodes were identified by the internal compound library (Figure 1), indicating a certain possibility of new compounds to be found among these metabolites.A further scrutiny of their MS/MS fragmentation pattern (Figure S29), followed by HRMSbased molecular formula analysis, strongly suggested that they were new analogues of fusaric amides [20].The systematic isolation, along with targeted purification, finally yielded three new fusaric acids, hepialiamides A−C (1-3), and one novel hybrid polyketide, hepialide (4), together with 18 known compounds (5-22) (Figure 2).The known compounds 8, 11, 13, 21, and 22 exhibited inhibitory activity against LPS-induced NO production.Herein, we report the fermentation, isolation, structure characterization and bioactivity of these isolates.
Naturally occurring aminated fusaric acid derivatives are uncommon in fungal metabolites, as mentioned above.Hepialiamides A-C (1-3) feature an oxidized fusaric acid core structure with a glycine or alanine coupled through an amide bond, representing the third examples of fusaric amides of fungal origin.Hepialide is a novel PKS-NRPS hybrid polyketide with an unusual isopropenylated pyrrolidone moiety that resembles the tetramic acid.
Neuroinflammation, characterized by the activation of microglia cells, is a common denominator in diverse neurological conditions, including neurodegenerative diseases, traumatic brain injury, and neuroinfectious disorders.Microglia cell activation contributes significantly to the progression of these diseases by releasing pro-inflammatory mediators, exacerbating neuronal damage and impairing synaptic plasticity [47].Hence, inhibiting glial cell activation emerges as a promising therapeutic strategy.The rich diversity and unique chemical structures of marine compounds provide a vast pool of potential drugs.Inflammatory tests were conducted on BV-2 microglia cells.Stimulation of the cells with bacterial products like lipopolysaccharide (LPS) triggers the production of NO and pro-inflammatory cytokines, key mediators in inflammation [48].Given the significance of high NO levels in inflammation, targeting inducible nitric oxide synthase (iNOS), the enzyme responsible for NO synthesis, has been suggested as an anti-inflammatory therapeutic approach.
In this LPS-induced BV-2 microglia cell model, we evaluated the inhibitory activity of these compounds on NO production at a low concentration (1 µM).The cells were preexposed to the compounds for 1 h, followed by LPS stimulation for 24 h, and subsequently, the concentration of nitrite, the primary metabolite derived from NO, was quantified.The results indicated that compounds 8, 11, 13, 21, and 22 reduced NO production by 34.2 ± 1.6%, 30.7 ± 4.8%, 32.9 ± 1.6%, 38.6 ± 2.1%, and 58.2 ± 2.6%, respectively, with respect to vehicle-treated stimulated cells (Figure 5A).Particularly noteworthy, further analysis revealed that compound 22 exhibited a remarkable dose-dependent inhibitory effect, with an IC 50 value of 426.2 nM (Figure 5B).

Fungal Identification, Fermentation, and Extract
The strain W7 was isolated from a sulfide sample (W 14.52°, S 13.59°) at a depth of 3073 m from the South Atlantic.It was identified as Samsoniella hepiali (GenBank accession number OR398925), as the ITS DNA gene sequence alignment demonstrated its similarity These compounds, especially compound 22, frequently demonstrate anti-inflammatory and immunomodulatory properties, rendering them promising candidates for the development of specific and potent inhibitors that target microglia cell activation pathways.Further research is needed for in-depth evaluation.

Fungal Identification, Fermentation, and Extract
The strain W7 was isolated from a sulfide sample (W 14.52 • , S 13.59 • ) at a depth of 3073 m from the South Atlantic.It was identified as Samsoniella hepiali (GenBank accession number OR398925), as the ITS DNA gene sequence alignment demonstrated its similarity to Samsoniella hepiali CGMCC 3.17103 (GenBank accession number NR_160318.1).The strain is preserved at the Key Laboratory of Marine Biogenetic Resources, Third Institute of Oceanography, Ministry of Natural Resources (Xiamen, China).The strain Samsoniella hepiali was cultured on a PDA plate at 25 • C for 3 days.Fresh mycelia and spores were inoculated into 250 mL Erlenmeyer flasks (×10) containing 30 mL PDB medium and incubated in a 180 rpm rotary shaker at 28 • C for 5 days.Then, the spore cultures were used to inoculate 25 × 1 L Erlenmeyer flasks containing rice medium (80 g rice and 120 mL distilled H 2 O for each flask) to perform large-scale fermentation.After 25 days, the fermented culture was extracted with EtOAc three times to provide a crude extract.The extract was redissolved in MeOH and extracted with petroleum ether (PE) three times.The MeOH solution was evaporated under reduced pressure to obtain a defatted extract (25 g).

UHPLC-Q-TOF and Molecular Networking Analysis
The crude extract was analyzed by LC-MS/MS, equipped with an Atlantis TM Premier BEH C 18 AX (2.1 × 100 mm, 1.7 µm) column.Samples were dissolved in MeOH with 1 mg/mL.A 10 µL aliquot of each sample was injected and eluted, using a gradient program with water + 0.1% formic acid (A) and CH 3 CN + 0.1% formic acid (B).The elution started with a 1% B isocratic phase for 1 min, followed by a gradient from 1% to 99% B in 10 min, maintaining 99% B for 3 min.Then, a gradient from 99% to 1% B in 1 min was applied, and 1% B was maintained for 3 min at a flow rate of 0.4 mL/min, with the temperature maintained at 40 • C. The full mass spectrometry (MS) survey scan was performed in positive electrospray ionization (ESI) mode within the range of 50 to 1000 Da.The MS/MS data were converted to mzXML files using MSConvert software (version 3.01).The converted MS/MS file was submitted to the GNPS platform for molecular networking for dereplication (https://gnps.ucsd.edu,accessed on 1 January 2016).Parameters for molecular network generation were set as follows: both precursor mass and MS/MS fragment ion tolerance were set at 0.02 Da, minimum pairs cosine score 0.6, minimum matched fragment ions 6, minimum cluster size 2, network TopK 10.The spectral networks were imported into Cytoscape 3.9.1 and visualized using the force-directed layout.

Theoretical ECD Calculation
As reported previously, conformational analysis was first performed via random searching in Sybyl-X 2.0 using the MMFF94S force field with an energy cutoff of 7.0 kcal/mol and an RMSD threshold of 0.2 Å.All conformers were consecutively optimized at PM6 and HF/6-31G(d) levels.Dominant conformers were re-optimized at the B3LYP/6-31G(d) level in the gas phase.The theoretical ECD spectra were calculated using the GIAO method at the MPW1PW91/6-31G (d, p) level in MeOH using Gaussian 09.The ECD spectrum was simulated in SpecDis (version 1.71) by overlapping Gaussian functions for each transition.

BV-2 Cell Culture and Treatment
BV-2 microglia cells were cultured in DMEM medium containing 10% fetal bovine serum and antibiotics (100 units/mL of penicillin and 100 g/mL of streptomycin) and maintained in a humidified 5% CO 2 incubator at 37 • C. For the experiment, cells were seeded overnight into 24-wells with 2 × 10 4 cells per well.The next day, cells were incubated with fresh culture medium containing the indicated concentration of compounds for an hour, followed by lipopolysaccharides (LPS) treatment (1 µg/mL).Cells treated with the vehicle (DMSO, 0.1%) served as the control.

Nitrite Quantification
The concentration of nitrite in the culture medium was determined using the Griess Reagent Kit (Therofisher, Shanghai, China).Briefly, 75 µL of cell culture supernatants was reacted with an equal volume of the Griess Reagent Kit for 30 min at room temperature, and the absorbance of the diazonium compound was obtained at a wavelength of 560 nm.Nitrite production by vehicle stimulation was designated as 100% inhibition (Ctrl) compared to LPS stimulation (Veh) for the experiment.

Conclusions
In summary, three new aminated fusaric acid derivatives, hepialiamides A-C (1-3), and one novel hybrid polyketide, hepialide (4), together with 18 known miscellaneous compounds (5-22), were isolated from the cultures of the deep-sea-derived fungus Samsoniella hepiali W7 with the aid of GNPS molecular networking.Hepialiamides A-C feature a glycine/alanine unit embedded in the fusaric acid backbone, representing the third examples of naturally occurring fusaric amides, while hepialide features an uncommon isopropenylated pyrrolidone moiety.Biologically, compound 8, 11, 13, 21, and 22 showed more than 30% inhibition against NO production induced by LPS at 1 µM, with 22 exhibiting a particularly low nanomole IC 50 , suggesting potential for developing potent anti-inflammatory drugs.

Figure 1 .
Figure 1.Molecular network of extract of the fungus Samsoniella hepiali W7.The cluster of new compounds (1-4) is expanded, and the nodes are marked with the m/z value of the parent ion.Figure 1.Molecular network of extract of the fungus Samsoniella hepiali W7.The cluster of new compounds (1-4) is expanded, and the nodes are marked with the m/z value of the parent ion.

Figure 1 .
Figure 1.Molecular network of extract of the fungus Samsoniella hepiali W7.The cluster of new compounds (1-4) is expanded, and the nodes are marked with the m/z value of the parent ion.Figure 1.Molecular network of extract of the fungus Samsoniella hepiali W7.The cluster of new compounds (1-4) is expanded, and the nodes are marked with the m/z value of the parent ion.

Figure 4 .
Figure 4. Experimental and calculated ECD spectra of 1, 3, and 4.Compound 2 was isolated as a yellow oil.The HRESI(+)MS of 2 exhibited a protonated molecular ion [M + H] + at m/z 251.0950.Analysis of the HRESI(+)MS and13 C NMR data revealed the molecular formula C 12 H 14 N 2 O 4 of 2 indicating seven DoUs.The NMR data (Table1) of 2 were closely related to those of 1, except for the presence of a methyl ketone group at (δ C 209.8, C-10) and (δ C 29.9, δ H 2.13, CH 3 -11) in 2, which was supported by the HMBC correlations of H 3 -11 to C-10 and C-9 (δ C44.6).The left of the structure was confirmed by 2D NMR correlations, as shown (Figure3).Therefore, 2 was identified as a new analogue of 1 and named hepialiamide B.Compound 3 was obtained as a yellow oil with the molecular formula of C 13 H 16 N 2 O 4 implied by its HRESI(+)MS spectrum.The 1D NMR data (Table1) of 3 were similar to those of 2 except that 3 had an extra methyl group [δ C 18.2 (C-4 )] at C-2 , which was confirmed by the COSY correlations of NH (δ H 8.74, d)/H-1 (δ H 4.46, d)/H-3 (δ H 1.42, d) and the HMBC correlations from H-1 to C-2 (δ C 173.8), C-3 (δ C 17.5), and C-7 (δ C 163.5).The absolute configuration of the chiral center C-1 in 3 was established to be S, as indicated by the agreement between the calculated ECD spectrum of 3 and the experimental data (Figure 4).Hence, compound 3 was identified as another new fusaric acid derivative, and named hepialiamide C. Compound 4 was obtained as a yellow oil.The HRESI(+)MS of 4 exhibited a sodiated molecular ion [M + Na] + at m/z 292.0800, suggesting a molecular formula (calculated for C 12 H 15 NO 6 Na, 292.0797) that requires six DoUs.The 1 H NMR (Table2) data of 4 presented the diagnostic signals of three methylene groups at δ H 2.28 (2H, t, J = 6.3 Hz, H-2 ), 2.64 (2H, q, J = 6.9 Hz, H-3 ), and 3.10 (2H, q, J = 15.0Hz, H-3), and two methyl groups at 2.08 (3H, s, H-8) and 1.81 (3H, s, H-9).In the13 C NMR spectrum, 12 carbon signals were observed and classified as two methyl (δ C 20.8, 18.5), three methylene (δ C 47.4, 36.3, 27.4), and seven non-protonated carbons (δ C 207.2, 197.0, 173.5, 171.4,1129.1, 122.3, 70.3), with the aid of HSQC data.In the 1 H-1 H COSY spectrum of 4, homonuclear vicinal coupling correlation (Figure 3) between H-7 and H-8, along with the HMBC correlations from H-7 to C-5/6, from H-8 to C-5/6, indicated an isopropenyl group.The HMBC cross-peaks observed from H-3 to C-2/4/9, H-7/8 to C-4, and the diagnostic signals from an exchangeable proton to C-2/4/5/9 permitted the construction of a five-member ring.Finally, based on the rest of the HMBC correlations from H-3 to C-1 , H-2 to C-1 /3 /4 , from H-3 to C-1 /4 , and the COSY signals between H-2 to H-3 , together with a comparable analysis of the chemical shifts of C-2/5 in 4 and those in literature [26,27], the planar structure o f 4 was assigned as shown.

Figure 5 .
Figure 5.The impact of compounds on NO release in LPS-induced microglia cells: (A) Nitrite accumulation in the culture media was quantified using the Griess method, and nitrite production inhibition was calculated compared to the only LPS treated group.Results were normalized to the LPS condition and presented as mean ± SEM (n = 4); **** p < 0.0001 vs. Ctrl; ### p < 0.0001, ## p < 0.01 vs. Veh; (B) Inhibitory curve of compound 22 against nitrite production.

Figure 5 .
Figure 5.The impact of compounds on NO release in LPS-induced microglia cells: (A) Nitrite accumulation in the culture media was quantified using the Griess method, and nitrite production inhibition was calculated compared to the only LPS treated group.Results were normalized to the LPS condition and presented as mean ± SEM (n = 4); **** p < 0.0001 vs. Ctrl; ### p < 0.0001, ## p < 0.01 vs. Veh; (B) Inhibitory curve of compound 22 against nitrite production.