An Unusual Spiro‐Fused Indole Alkaloid‐Terrein Hybrid From Marine‐Derived Fungus Aspergillus chevalieri HK‐56

ABSTRACT

One previously undescribed alkaloid variecoterrein (1) and seven known compounds (2–8) were obtained from the solid‐state fermentation culture of the marine‐derived fungus Aspergillus chevalieri HK‐56. The planar and stereochemical structures of 1 was unambiguously elucidated by comprehensive spectroscopic analysis, including 1D and 2D nuclear magnetic resonance (NMR) spectroscopy, high‐resolution electrospray ionization mass spectrometry (HRESIMS), DP4+ probability analysis, and electronic circular dichroism (ECD) spectroscopy. Compound 1 represents a structurally unique hybrid molecule, constituting the first naturally occurring conjugate that incorporates both indole 2,5‐diketopiperazine alkaloid and terrein structural scaffolds. All isolated compounds were evaluated for their inhibitory effects on nitric oxide (NO) production in lipopolysaccharide (LPS)‐induced RAW 264.7 murine macrophage cells. Among them, compound 2 exhibited potent in vitro anti‐inflammatory activity, with an IC₅₀ value of 22.72 ± 0.22 µM, comparable to the positive control quercetin (IC₅₀ = 13.81 ± 0.35 µM).

1. Introduction

Among marine fungal taxa, the genus Aspergillus ranks among the most extensively studied filamentous fungi, renowned for its exceptional capacity to produce bioactive natural products with therapeutic potential—an attribute that endows it with substantial economic value [1, 2]. Indole diketopiperazine alkaloids constitute a class of natural products characterized by a core diketopiperazine scaffold, are predominantly isolated from fungal genera such as Aspergillus and Penicillium [3, 4]. The structural cores of these compounds are assembled via the condensation of L‐tryptophan with specific amino acids. Potential condensation partners include, such as L‐valine, L‐alanine, L‐proline, or additional L‐tryptophan molecules. Subsequent modifications introduced by diverse tailoring enzymes then give rise to a wide array of structurally distinct final products [5, 6]. Interest in indole diketopiperazines is due to their significant biological activities, such as antimicrobial, antiviral [7, 8], anticancer [9, 10], immunomodulatory [11, 12], antioxidant [13], and insecticidal activities [14, 15]. Previous studies have reported a wide spectrum of bioactivities for indole diketopiperazines. For instance, gliotoxin and its analogs have demonstrated potent immunosuppressive effects by inducing apoptosis in immune cells [16]. Echinulins and neoechinulins have shown notable anti‐inflammatory properties through the suppression of pro‐inflammatory cytokine production [17]. Notably, some indole diketopiperazines, such as brevianamide F and its derivatives, have been reported to modulate nitric oxide (NO) production in activated macrophages, highlighting their regulatory role in inflammatory responses [18]. Additionally, recent investigations have revealed that various prenylated indole diketopiperazine alkaloids exhibit significant NO inhibitory activity in lipopolysaccharide (LPS)‐induced macrophages, further supporting the anti‐inflammatory potential of this compound class [19].

NO is a crucial signaling molecule involved in various physiological and pathological processes, including immune defense, neurotransmission, and inflammation [20]. However, excessive NO production, particularly in activated macrophages, is implicated in the pathogenesis of chronic inflammatory diseases, such as rheumatoid arthritis, inflammatory bowel disease, and septic shock [21]. Therefore, the identification of natural products capable of modulating NO production represents a promising strategy for the discovery of novel anti‐inflammatory agents. Given the previously reported association between indole diketopiperazine structures and NO inhibitory activity, we selected the NO production inhibition assay in LPS‐induced RAW 264.7 macrophages as a primary screening platform to evaluate the anti‐inflammatory potential of the isolated compounds in this study [22].

In the present study, a chemical investigation of the secondary metabolites from the marine‐derived fungus Aspergillus chevalieri HK‐56 was carried out. The crude extract was monitored for alkaloid content via thin‐layer chromatography (TLC), using Dragendorff's reagent as the detection agent. This effort led to the isolation and characterization of one new alkaloid variecoterrein (1) and seven known compounds (2–8) (Figure 1). Herein, we report the isolation and structural elucidation of compound 1, as well as the inhibitory activity of compounds 1–8 on NO production in LPS‐induced RAW 264.7 murine macrophage cells.

FIGURE 1.

Chemical structures of compounds 1–8.

2. Results and Discussion

Variecoterrein (1) was separated as a yellow oil. Its molecular formula C₂₇H₂₉N₃O₅ was established by the negative HRESIMS ion peak at m/z 474.2032 ([M—H]⁻, calcd for 474.2034), indicating 15° of unsaturation. The IR spectrum suggested the presence of hydroxyl group (3200–3600 cm⁻¹), amine group (3359 cm⁻¹), and carbonyl functionalities (1673, 1635 cm⁻¹). The ¹H NMR spectrum (Table 1) exhibited signals for two hydroxy groups at δ _H 6.03 (1H, br s, 4′‐OH) and δ _H 5.82 (1H, s, 3′‐OH), three NH signals at δ _H 8.32 (1H, s, H‐11), 8.70 (1H, s, H‐14) and 11.06 (1H, s, H‐1), four aromatic protons at δ _H 6.97 (1H, dd, J = 8.0, 8.0 Hz, H‐5), 7.07 (1H, dd, J = 8.0, 8.0 Hz, H‐6), 7.26 (1H, d, J = 8.0 Hz, H‐4) and 7.41 (1H, d, J = 8.0 Hz, H‐7), four olefinic protons at δ _H 6.89 (1H, s, H‐8), 6.09 (1H, dd, J = 17.3, 10.6 Hz, H‐16), 5.07 (1H, d, J = 10.4 Hz, H₂‐17), and 5.03 (1H, d, J = 17.2 Hz, H₂‐17), three methine protons at δ _H 4.36 (1H, br s, H‐4′), 3.92 (1H, br s, H‐3′) and 2.30 (1H, m, H‐7′), two methylene protons at δ _H 2.41 (1H, dd, J = 18.4, 4.0 Hz, H₂‐6′β), 1.96 (1H, dd, J = 18.4, 11.2 Hz, H₂‐6′α), 2.12 (1H, d, J = 13.2 Hz, H₂‐8′α), and 1.51 (1H, overlapped, H₂‐8′β), and three methyl groups at δ _H 1.49 (3H, s, H₃‐18), 1.49 (3H, s, H₃‐19), and 1.01 (3H, d, J = 6.6 Hz, H₃‐9′). The ¹³C NMR (Table 1) and DEPT spectra of 1 revealed a total of 27 signals, including three carbonyl carbons, six aromatic carbons, seven quaternary carbons, five methine carbons, three methylene carbons, and three methyl carbons. Analysis of the ¹H and ¹³C data found that it was similar with those of 3 [23], with an additional architectures of D−E rings. Further observation of the ¹H‐¹H COSY spectrum of 1 showing two spin systems of H‐3′/H‐4′ and H₂‐6′/H‐7′/H₂‐8′/H₃‐9′ indicated the connectivity of C‐3′/C‐4′ and C‐6′/C‐7′/C‐8′. Moreover, the HMBC correlations from H‐3′ (δ _H 3.92) to C‐4′ (δ _C 76.0), H‐4′ to C‐1′ (δ _C 134.3)/C‐5′ (δ _C 172.0), H₂‐6′ to C‐1′/C‐5′, H₂‐8′ to C‐13/C‐1′, H₃‐9′ to C‐6′/C‐7′/C‐8′, 11‐NH to C‐9 (δ _C 125.0)/C‐12 (δ _C 56.3)/C‐13 (δ _C 165.6), in combination of the residual unsaturation and molecular weight, constructed the presence of a terrein unit composed of D−E rings, cross‐conjugated to indole 2,5‐diketopiperazine alkaloid analogue 3 [23] by the spiro carbon C‐12 (Figure 2). Thus, the planar structure of compound 1 was determined. The partial relative configuration of 1 was established on the basis of diagnostic nuclear overhauser effect (NOESY) analysis (Figure 3). The coupling constant of 11.2 Hz between H₂‐6′α and H‐7′ indicates that H₂‐6′α and H‐7′ must be on opposite sides of the six‐membered ring. Accordingly, H₂‐6′α and the methyl group at C‐9′ are assigned to be on the same side. The NOESY correlations of H‐4′ to H₂‐6′β and H₂‐6′α to H₃‐9′ indicated that 4′‐OH and H₃‐9′ were on the same side, while H‐7′ and H₂‐6′β oriented to the other side. At the same time, the broad singlets with negligible coupling constants for H‐3′ and H‐4′ indicated that the two vicinal hydroxyl groups were on opposite faces. Thus, the possible relative configurations of C‐12, C‐3′, C‐4′, and C‐7′ were assigned as (12R,3′R,4′S,7′R)‐1 and (12S,3′R,4′S,7′R)‐1. To further corroborate this assignment, we employed DP4+ probability analysis, a well‐established method for relative configurational assignment [24, 25]. Theoretical ¹H and ¹³C NMR data were calculated for eight possible diastereomers: (12S*/3′R*/4′S*/7′R*)‐1 (1a), (12S*/3′S*/4′S*/7'R*)‐1 (1b), (12S*/3′S*/4′R*/7′R*)‐1 (1c), (12S*/3′R*/4′R*/7′R*)‐1 (1d), (12S*/3′R*/4′S*/7′S*)‐1 (1e), (12S*/3′S*/4′S*/7′S*)‐1 (1f), (12S*/3′S*/4′R*/7′S*)‐1 (1g) and (12S*/3′R*/4′R*/7′S*)‐1 (1h). DP4+ analysis indicated 1g was the most probable configuration with high scores of 96% (see Supporting Information), thereby confirming the relative configuration as 12S*/3′S*/4′R*/7'S*. Finally, the absolute configuration of 1 was established by quantum‐chemical electronic circular dichroism (ECD) calculations. Theoretical ECD calculations were carried out at the B3LYP/6‐311G (d) level for the enantiomers (12S/3′S/4′R/7'S)‐1 and (12R/3′R/4′S/7′R)‐1. As a result, the calculated ECD spectrum of (12R/3′R/4′S/7′R)‐1 showed a similar trend to the experimental curve (Figure 4), which led to the conclusion that the absolute configuration of 1 is (12R,3′R,4′S,7′R)‐1. Therefore, the structure of 1 was determined and given a trivial name variecoterrein. To the best of our knowledge, compound 1 represents a new type of naturally occurring indole 2,5‐diketopiperazine‐terrein hybrid molecule by a spiro‐fused connection, expanding the structural diversity of this skeleton family and indicating a distinct bio‐synthetic pathway. The possible biosynthetic pathway of compound 1 may involve dehydration of the precursor compound to form intermediate I, followed by a Diels–Alder (DA) cycloaddition between I and terrein, and subsequent double‐bond migration (Scheme 1). Thus, the pathway is biosynthetically plausible.

TABLE 1.

¹H NMR and ¹³C NMR data for 1 (DMSO‐d ₆).

No.	δ H (J in Hz)	δ C
1	11.06 (1H, s)	—
2	—	143.9
3	—	103.4
3a	—	126.2
4	7.26 (1H, d, 8.0)	118.9
5	6.97 (1H, dd, 8.0, 8.0)	119.3
6	7.07 (1H, dd, 8.0, 8.0)	120.7
7	7.41 (1H, d, 8.0)	111.6
7a	—	135.1
8	6.89 (1H, s)	109.8
9	—	125.0
10	—	159.7
11	8.32 (1H, s)	—
12	—	56.3
13	—	165.6
14	8.70 (1H, s)	—
15	—	39.1
16	6.09 (1H, dd, 17.3, 10.6)	145.2
17	5.07 (1H, d, 10.4, H2‐17); 5.03 (1H, d, 17.2, H2‐17)	111.7
18	1.49 (3H, s)	27.6
19	1.49 (3H, s)	27.5
1'	—	134.3
2'	—	200.1
3'	3.92 (1H, br s)	81.0
4'	4.36 (1H, br s)	76.0
5'	—	172.0
6'	2.41 (1H, dd, 18.4, 4.0, H2‐6′β); 1.96 (1H, dd, 18.4, 11.2, H2‐6′α)	32.3
7'	2.30 (1H, m)	26.2
8'	2.12 (1H, d, 13.2, H2‐8′α); 1.51 (1H, overlap, H2‐8′β)	44.5
9'	1.01 (3H, d, 6.4)	21.0
3'‐OH	5.82 (1H, s)	—
4'‐OH	6.03 (1H, br s)	—

FIGURE 2.

Key ¹H‐¹H COSY and HMBC correlations of 1.

FIGURE 3.

Key NOESY correlations of 1.

FIGURE 4.

Calculated and experimental ECD spectra for 1.

SCHEME 1.

Possible biosynthetic pathway of 1.

The seven known compounds (2–8) were identified as variecolorin O (2) [7], neoechinulin A (3) [23], preechinulin (4) [26], 2‐(1,1‐dimethyl‐2‐propen‐1‐yl)‐1H‐indole‐3‐carboxaldehyde (5) [23], alantrypinone (6) [27], methyl‐3,4,5‐trimethoxyl‐2‐(2‐(nicotinamide) benzamido) benzoate (7) [28], and eurocristatine (8) [29] after thoroughly comparing their NMR and ESIMS data with literature data.

The inhibitory activity of compounds 1–8 on NO production in LPS‐induced RAW 264.7 murine macrophage cells were tested. Among them, compound 2 exhibited potent in vitro anti‐inflammatory activity, with an IC₅₀ value of 22.72 ± 0.22 µM, comparable to the positive control quercetin (IC₅₀ = 13.81 ± 0.35 µM).

3. Conclusions

In summary, one previously undescribed alkaloid variecoterrein (1) and seven known compounds (2–8) were obtained from the marine‐derived fungus A. chevalieri HK‐56. The structures of 1–8 were elucidated by extensive NMR spectroscopic data analysis, and the experimental CD spectra of compound 1 were compared with the CD spectra calculated by the time‐dependent density functional method to determine the absolute configuration of compound 1. Compound 1 represents a structurally unique hybrid molecule, constituting the first naturally occurring conjugate that incorporates both indole 2,5‐diketopiperazine alkaloid and terrein structural scaffolds. All isolated compounds were evaluated for their inhibitory effects on NO production in LPS‐induced RAW 264.7 murine macrophage cells. Among them, compound 2 exhibited potent in vitro anti‐inflammatory activity, with an IC₅₀ value of 22.72 ± 0.22 µM, comparable to the positive control quercetin (IC₅₀ = 13.81 ± 0.35 µM). These findings not only enrich the chemical diversity known from marine‐derived Aspergillus fungi but also offer promising lead compounds for anti‐inflammatory drug development.

4. Experimental Section

4.1. General Experimental Procedures

Optical rotation was measured by a General TextModular Circular Polarimeter 500 (Anton Paar, Austria), while ECD spectra were recorded on a JASCO J‐715 spectrophotometer (JASCO, Japan). NMR experiments, including 1D (¹H and ¹³C) and 2D (HSQC, HMBC, COSY, and NOESY) spectra, were performed on a JEOL 600 MHz NMR spectrometer (Japan) and a Bruker AV‐400 instrument (Bruker Corporation, Switzerland) with tetramethylsilane (TMS) as the internal standard. Chemical shifts were reported in ppm (δ) relative to residual solvent signals: DMSO‐d ₆ (δ _H 2.50; δ _C 39.5), CD₃OD‐d ₄ (δ _H 3.31; δ _C 49.0), and CDCl₃ (δ _H 7.26; δ _C 77.2). Mass spectrometric analyses were carried out on a Bruker Daltonics Apex‐Ultra 7.0 T mass spectrometer (Bruker Corporation, Billerica, MA, USA) and a Q‐TOF Ultima Global GAA076 LC‐MS system for ESIMS and HRESIMS data acquisition. Semipreparative HPLC was conducted by a Shimadzu LC‐20A prep‐HPLC system equipped with an Agilent Eclipse XDB‐C18 semipreparative column (9.4 × 250 mm, 5 µm). Column chromatography (CC) was performed with Sephadex LH‐20 (Pharmacia Co. Ltd., Sandwich, UK) and silica gel (200–300/300–400 mesh, Qingdao Marine Chemical Factory, Qingdao, China). Silica gel GF254 for TLC was purchased from Qingdao Marine Chemical Factory (China). All solvents were of analytical grade and obtained from Guangzhou, China.

4.2. Fungal Material

The fungal strain HK‐56 was obtained from the sea anemone collected in the southern sea area of Taishan City, Jiangmen, Guangdong Province, China (GPS coordinates: 113.046°E, 20.221°N). It is preserved in the Key Laboratory of Tropical Medicinal Resource Chemistry of the Ministry of Education, College of Chemistry and Chemical Engineering, Hainan Normal University, Haikou, China. Based on BLAST analysis of the ITS sequence, the strain was identified as A. chevalieri HK‐56. The corresponding sequence has been submitted to GenBank under the accession number PV307692.

4.3. Extraction, Isolation, and Purification

The fungal strain HK‐56 was cultivated on solid rice medium in 1 L Erlenmeyer flasks (150 flasks; 50 mL of rice and 50 mL of water, and 0.15 g of sea salt per Erlenmeyer flask, autoclave sterilization) at 28°C for 35 days. The fermentation was extracted three times with ethyl acetate (EtOAc), and the combined extracts were concentrated under reduced pressure to afford 160.0 g of EtOAc extract.

The organic extract (150.0 g), obtained by vacuum evaporation to dryness, was subjected to silica gel CC using a petroleum ether (PE)/ethyl acetate (EtOAc) gradient system ranging from 0:1 to 1:0, yielding eleven fractions (Fr.1 to Fr.11). Fraction Fr. 4 (25 g) was further purified under identical chromatographic conditions (PE/EtOAc, 0:1 to 1:0) to afford twelve subfractions (Frs.4‐1 to Frs.4–12). Subfraction Fr. 4–10 (2.5 g) was further purified by ODS CC using a MeOH/H₂O gradient elution (10% to 100%), affording 10 additional subfractions (Frs. 4‐10‐1 to Frs. 4‐10‐10). Subfraction Frs. 4‐10‐6 was purified by semipreparative HPLC (MeCN/H₂O, 48:52, 2 mL/min) to give 2 (2.0 mg, t _R = 7.0 min). Subfraction Frs. 4‐8 was purified by semipreparative HPLC (MeCN/H₂O, 20:80, 2 mL/min) to give 5 (3.0 mg, t _R = 29.0 min). Fraction Fr. 7 (10.0 g) was subjected to CC over ODS eluting with a MeOH/H₂O gradient (10%–100%) to yield thirty subfractions (Frs. 7–1 to Frs. 7–30). Subfraction Frs. 7–18 was purified by semipreparative HPLC (MeCN/H₂O, 35:65, 2 mL/min) to give 3 (3.0 mg, t _R = 19.0 min). Subfraction Frs. 7–13 was purified by semipreparative HPLC (MeCN/H₂O, 28:72, 2 mL/min) to give 6 (3.5 mg, t _R = 13.0 min). Subfraction Frs. 7–20 was purified by semipreparative HPLC (MeCN/H₂O, 32:68, 2 mL/min) to give 1 (4.2 mg, t _R = 25.0 min) and 7 (8.5 mg, t _R = 15.5 min). Fraction Fr. 7 (2.0 g) was purified by semipreparative HPLC (MeCN/H₂O, 43:57, 2 mL/min) to give 4 (9.6 mg, t _R = 11.0 min) and 8 (8.7 mg, t _R = 10.0 min).

4.4. Spectroscopic Data of 1

Variecoterrein (1): yellow oil; [α]_D ²⁵ + 13.9 (c 0.1, MeOH); UV (MeOH) λ _max (log ε): 225 (6.39) nm, 284 (5.73) nm, 338 (5.85) nm; ECD (MeOH) λ _max (Δε): 196 (−11.08), 208 (53.66), 220 (−108.35), 234 (40.05), 246 (10.55), 260 (80.59), 315 (−12.01), 327 (−10.69), and 343 (−16.68) nm; ¹H and ¹³C NMR data, see Table 1; IR (KBr) ν _max 3359, 2963, 2919, 1673, 1635, 1419, 1380, and 743 cm⁻¹; HRESIMS m/z 474.2032 [M—H]⁻ (calcd for C₂₇H₂₈N₃O₅, 474.2034).

4.5. TDDFT–ECD Calculations

The TDDFT–ECD calculations for compound 1 was performed by the Gaussian 16 program [30]. Conformational analysis was carried out with Spartan 14 software employing the MMFF force field within an energy window of 5.0 kcal/mol. Conformers exhibiting Boltzmann population above 5% were subsequently re‐optimized at the B3LYP/6‐31G (d) level in vacuo. TDDFT ECD calculations were executed at the B3LYP/6‐311G (d) level by the SMD solvation model for methanol, and the resulting ECD spectral curves were generated with SpecDis software [31].

4.6. Assay of Anti‐Inflammatory Activity

The anti‐inflammatory activity was assessed by measuring NO production in LPS‐induced RAW 264.7 cells (RRID: CVCL_0493) by the Griess reagent method [32]. RAW 264.7 cells were seeded into 96‐well plates at a density of 1 × 10⁵ cells per well and allowed to adhere for 12 h, after which they were treated with test samples or the positive control (dexamethasone). Following a 1 h incubation, the cells were stimulated with 1 µg/mL LPS and maintained at 37°C in a humidified atmosphere containing 5% CO₂ for 24 h. Subsequently, 50 µL of culture supernatant was mixed with 100 µL of Griess reagent, and the absorbance was measured at 540 nm by a microplate reader.

Author Contributions

WanJuan Xie, XuHua Nong, Hong Xiao, XueMing Zhou, YongMin Zhang, and GuangYing Chen designed experiments, elucidated chemical structures, wrote the article, collected fungal samples, and identified fungi by DNA. WanJuan Xie, YiLin Kuang, and YanLing Ma isolated compounds and evaluated activities.

Conflicts of Interest

The authors declare no conflicts of interest.

Supporting Information

Additional references cited with in the Supporting Information.

Supporting information

Supporting File 1: cbdv71351‐sup‐0001‐SuppMat.pdf

Data Availability Statement

The data that supports the findings of this study are available in the supplementary material of this article.