**1. Introduction**

Deep-sea hydrothermal vents are recognized as one of the most extreme and dynamic habitats on our planet [1]. These hotspot ecosystems are characterized by high temperature, high pressure, low oxygen supply, and the absence of sun light [2]. In addition, hydrothermal vent flows bring fluids with high concentrations of reduced sulfur-containing compounds and heavy metals [2]. Given this fact, microorganisms living in this specific environment are considered as a new frontier for discovery of natural products with unique structures and tremendous pharmacological activities [3].

*Aspergillus* is renowned as a prolific source of numerous fungal peptides, including lipo-, depsi-, linear-, and cyclic-peptides, which are structurally unique and demonstrated various bioactivities, such as anti-microbial, anti-fungal, anti-inflammatory, and cytotoxic activities [4,5]. Among the peptides derived from *Aspergillus* spp., unguisins are a unique cyclic heptapetide class commonly produced by *Aspergillus unguis*, and until now unguisins A–G have been reported [6,7].

Inflammation is a protective response of our body to a wide range of stimuli. This process plays a central role or is an important symptom in the pathogenesis of various chronic diseases for instance Alzheimer's disease, asthma, diabetes, and rheumatoid arthritis [8]. The inflammatory process is characterized by over secretion of nitric oxide (NO) and inflammatory cytokines such as interleukin 1 beta (IL-1β), tumor necrosis factor alpha

**Citation:** Anh, C.V.; Yoon, Y.D.; Kang, J.S.; Lee, H.-S.; Heo, C.-S.; Shin, H.J. Nitrogen-Containing Secondary Metabolites from a Deep-Sea Fungus *Aspergillus unguis* and Their Anti-Inflammatory Activity. *Mar. Drugs* **2022**, *20*, 217. https://doi.org/ 10.3390/md20030217

Academic Editors: Donatella Degl'Innocenti and Marzia Vasarri

Received: 2 March 2022 Accepted: 19 March 2022 Published: 20 March 2022

**Publisher's Note:** MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations.

**Copyright:** © 2022 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https:// creativecommons.org/licenses/by/ 4.0/).

(TNF-α), and interleukin 6 (IL-6). Therefore, reducing the production of inflammatory mediators is a key indicator for the treatment of various diseases.

As part of our study on marine-derived microorganisms isolated near hydrothermal vents, we have reported some anti-inflammatory phenazine alkaloids from a yeast-like fungus *Cystobasidium laryngis*, and nidulin-related polyketides from *A. unguis* IV17-109, which showed anti-microbial and cytotoxic activities [9,10]. Based on NMR guided isolation, we found that the 1H NMR spectra of non-polar fractions from *A. unguis* IV17-109 showed some minor interesting peaks in the olefinic region, which do not belong to unguisin peptides or nidulin-related polyketides. Further careful purification of these fractions led to the identification of two new compounds, variotin B (**1**) and coniosulfide E (**2**) (Figure 1). Anti-inflammatory activity of **1** and **2** was preliminarily evaluated and the result revealed that **1** has moderate activity. Here, we report the details of the isolation, structure identification, and anti-inflammatory nature of these compounds.

**Figure 1.** Structures of **1**–**3** isolated from *A. unguis* IV17-109, and the synthetic analogs (**4** and **5**).

#### **2. Results and Discussion**

Compound **1** was isolated as pale-yellow needles with the molecular formula of C20H27NO2 based on its HRESIMS peak at *m*/*z* 336.1938, ([M+Na]+, calculated for C20H27NO2Na, 336.1939), requiring 8 indices of hydrogen deficiency. The 1H NMR spectrum of **1** showed signals attributed to a methyl group at *δ*<sup>H</sup> 1.62 (d, *J* = 4.5, H3-21), seven methylene groups at *δ*<sup>H</sup> 2.04–3.82, and ten olefinic protons at *δ*<sup>H</sup> 5.42–7.37. The 13C NMR spectrum in combination with HSQC data revealed signals of 20 resonances belonging to a methyl at *δ*<sup>C</sup> 18.1, seven methylene carbons at *δ*<sup>C</sup> 18.1–47.0, ten olefinic carbons at *δ*<sup>C</sup> 121.7–146.9, and two carbonyl carbons at *δ*<sup>C</sup> 168.2 and 177.8. Two carbonyl and ten sp<sup>2</sup> carbons, accounting for 7 out of 8 degrees of unsaturation, indicated **1** is a monocyclic compound. The structure of a five-membered lactam ring was determined by continuous 1H-1H COSY correlations from H2-2 to H2-4, and the HMBC correlation from H2-4 to C-1. A substructure was identified as a C-16 polyunsaturated fatty acid by continuous 1H-1H COSY correlations from H-7 to H3-21, and the HMBC correlations from H-7 and H-8 to C-6 (Figure 2). The connection of the fatty acid and the lactam ring was corroborated by the HMBC correlation from H2-4 to C-6.

The geometries of Δ7,9,13,15 were deduced as *E*-form by their large coupling constants (Table 1) and the chemical shift of terminal methyl (C-21) was *δ*<sup>C</sup> 18.1, revealing the geometry of Δ<sup>19</sup> was *E*-form [11,12]. Therefore, **1** was determined as a new variotin derivative with a non-branched side chain and named variotin B [13].

**Figure 2.** Key 2D NMR data of **1** and **2**.



Compound **2** was isolated as a colorless solid and its molecular formula was determined as C22H40N2O4S, with four indices of hydrogen deficiency based on its HRESIMS peak at *m*/*z* 451.2607, ([M+Na]+, calculated for C22H40N2O4SNa, 451.2606). The 1H NMR spectrum of **2** showed signals attributed to five methyl groups at *δ*<sup>H</sup> 1.17 (s, 6H, H3-21 and H3-22), 1.61 (s, H3-23), 1.68 (s, H3-24), and 2.01 (s, H3-1); seven methylene groups at *δ*<sup>H</sup> 1.40–3.22; an oxygenated methylene group at *δ*<sup>H</sup> 3.60 and 3.64 (H-25a,b); an amide methylene at *δ*<sup>H</sup> 3.86 (m, H2-4); an amide methine at *δ*<sup>H</sup> 4.03 (m, H-7); and two olefinic protons at *δ*<sup>H</sup> 5.13 and 5.23 (m, H-11 and H-15). The 13C NMR spectrum in combination with HSQC data demonstrated signals of 22 resonances belonging to five methyls at *δ*<sup>C</sup> 16.0 (C-23), 16.2 (C-24), 22.5 (C-1), and 29.2 (2C, C-21 and C-22); seven methylenes at *δ*<sup>C</sup> 23.7–44.3; an oxygenated methylene at *δ*<sup>C</sup> 63.6; an amide methylene at *δ*<sup>C</sup> 43.6; an amide methine at *δ*<sup>C</sup> 52.3; a tertiary alcohol at *δ*<sup>C</sup> 71.4; four olefinic carbons at *δ*<sup>C</sup> 121.8–140.1; and

two carbonyl carbons at *δ*<sup>C</sup> 171.5 and 173.9. Two carbonyl and four sp<sup>2</sup> carbons accounting for all 4 degrees of unsaturation indicated **2** is an acyclic compound.

The structure of a cysteinol unit was determined by sequential 1H-1H COSY correlations of H-8a,b/H-7/H-25a,b. A partial structure of *N*-acetylglycine, which was connected to the cysteinol moiety via a peptide bond, was determined by the HMBC correlations of H-7/C-5, H2-4/C-2, and H3-1/C-2. The remaining 15 carbons were assigned as a 10-hydro-11-hydroxyfarnesyl moiety based on a detailed analysis of 1H-1H COSY and HMBC data (Figure 2), and the connection of this moiety with the cysteinol residue via a thioether bond was determined by the HMBC correlations of H-8 a,b/C-10 and H-10a,b/C-8. The geometry of Δ<sup>11</sup> was deduced as *E*-form by the strong NOESY correlations from H3-24 to H-10a,b and H-13a,b; and no-observed NOESY correlation from H3-24 to H-11. Similarly, Δ<sup>15</sup> was also determined as *E*-form (Figure 2). Consequently, the gross structure of **2** was determined as shown in Figure 1.

To determine the absolute configuration of **2**, we synthesized its analogs (**4** and **5**, a pair of enantiomers synthesized from *L*- and *D*-cysteine and farnesyl chloride, Scheme 1) from commercially available substances. By comparing the optical rotation sign of **2** [α] 20 D − 100 (*c* 0.3, MeOH) with that of **4** [α] 20 <sup>D</sup> − 110 (*c* 0.3, MeOH) and **5** [α] 20 <sup>D</sup> + 120 (*c* 0.3, MeOH), the absolute configuration of **2** was determined to be the same as that of **4** (7*R*). Thus, **2** was determined as a new derivative of sulfur-containing natural products, coniosulfides A-D [14], and named coniosulfide E.

**Scheme 1.** Synthesis of **4** and **5**.

A co-isolated known compound was identified as unguisin A (**3**) by comparing its spectroscopic data with the corresponding literature values [6].

Since some fungal peptides were reported to show anti-inflammatory activity [4], **1** and **2** were evaluated for their anti-inflammatory activity. Subsequently, **1** showed moderate anti-inflammatory activity with an IC50 value of 20.0 μM. Even though a literature review revealed many synthetic analogs of **2** demonstrated inhibitory effects on human isoprenylcysteine carboxyl methyltransferase (hIcmt) [15] or the inflammation process [16], unfortunately, **2** showed no anti-inflammatory activity at a concentration of 30.0 μM. Due to the limited amount of **2**, we were unable to check its effect on hIcmt. Therefore, further studies are needed to find the bioactivities of **2**.

To further investigate the anti-inflammatory activity of **1**, we examined the inhibitory effect of **1** on lipopolysaccharide (LPS)-induced production of inflammatory mediators, including NO, IL-6, and iNOS, in RAW 264.7 cells. The treatment of RAW 264.7 cells

with LPS led to the accumulation of nitrite and IL-6, and **1** dose-proportionally inhibited LPS-induced production of nitrite and IL-6 in LPS-stimulated RAW 264.7 cells (Figure 3A,B). To further examine whether the effect of **1** were due to its effects on the mRNA expression of cognate genes, we investigated the effect of **1** on the mRNA expression of inducible nitric oxide synthase (iNOS) and IL-6 by quantitative polymerase chain reaction (qPCR). The mRNA levels of iNOS and IL-6 were induced by LPS treatment, and this induction was suppressed by **1** in a concentration-dependent manner (Figure 3C,D). Considering the above-mentioned data, it is noticeable that **1** showed anti-inflammatory activity by suppressing the production of NO and the expression of iNOS and IL-6 with no cytotoxicity at the treated concentrations. The results revealed that fungal natural products could be an important source of leads for the development of new anti-inflammatory drugs with minimal side effects.

**Figure 3.** Inhibitory effects of **1** on LPS-induced nitrite production and IL-6 secretion in RAW 264.7 cells. RAW 264.7 cells were pretreated with **1** at the depicted concentrations (1~30 μM) for 1 h and stimulated with LPS (200 ng/mL) for 24 h. The levels of nitrite (**A**) and IL-6 (**B**) in culture supernatants were determined by Griess reaction and ELISA, respectively. The mRNA levels of IL-6 (**C**) and iNOS (**D**) were examined by qPCR. Data are represented as the mean ± SD of quadruplicate determinations. An asterisk (\*) denotes that the response is significantly different from vehicle-treated group as determined by Dunnett's multiple comparison test at *p* < 0.05. The results shown are representatives of more than two independent experiments (UN: Untreated; VH: Vehicle (0.1% DMSO)).
