Introduction

Animal intestinal microorganisms have been studied for decades, and the investigation mainly focused on the food digestion and absorption, immunity, pathogens resistance and health maintaining.1 Over the past few years, the authors have dedicated to research on the diversity, bioactivities and the secondary metabolites of cultivable actinobacteria from animal feces and several new microbial species and new bioactive substances have been found.2, 3, 4, 5, 6, 7 These findings showed animal intestinal and fecal microorganisms could be considered as a rich and important microbial resource for discovering new bioactive natural products.

In the course of our continuous search for potential structurally novel and bioactive natural products from fecal actinobacteria associated with animal feces, we investigated the secondary metabolites of Streptomyces violaceoruber (YIM 101131), which was isolated from Equus burchelli (zebra) feces. Separation of the fermentation broth of S. violaceoruber led to the isolation of three new compounds, a tetralone derivative, (R)-7-acetyl-3,6-dihydroxy-8-propyl-3,4-dihydronaphthalen-1(2H)-one (2), a tryptophan derivative, (S)-2-(5-(2,3-dihydroxy-3-methylbutyl)-1H-indol-3-yl) acetonitrile (3), a fatty acid amide (5Z,8Z,11 R*,12 R*)-11,12-dihydroxytetradeca-5,8-dienamide (4), as well as a known compound GTRI-02 (1)8, 9, 10 (Figure 1). The structures of these metabolites were determined on the basis of comprehensive spectroscopic analysis and Mo2(OAc)4 induced CD effects. The anti-inflammatory activity of isolated compounds 14 were evaluated by examining their ability to inhibit NO production in LPS-stimulated RAW 264.7 macrophage cells. Details of the isolation, structure elucidation and anti-inflammatory activity of the new compounds are described in the present paper.

Figure 1
figure 1

Structures of compounds 14.

Results and Discussion

Compound 2 was isolated as a white, amorphous solid. The molecular formula of 2 was determined to be C15H18O4 based on the HRESIMS at m/z 261.1218 [M-H]. The IR spectrum showed absorptions for hydroxyl (3270 cm−1), unsaturated carbonyl (1698 cm−1) and benzene ring (1582 cm−1). The 1H NMR data (Table 1) of 1 displayed signals for an aromatic proton (δH 6.59, s), an oxygenated methine proton (δH 4.09, m) and two methyl groups at δH 2.39 (3H, s) and δH 0.88 (3H, t, J=7.2 Hz). Additional signals showed six aliphatic methylene protons (δH 3.03, dd, J=15.8, 2.6 Hz; δH 2.78, dd, J=15.8, 7.2 Hz; δH 2.71, brd, J=16.0 Hz; δH 2.46, dd, J=16.0, 7.9; δH 1.39, 2H, m). The 13C NMR spectrum of 2 showed 15 carbon signals (Table 1). Except for the 13C NMR signals corresponding with the 1H NMR data, seven quaternary carbons, including two ketone carbonyls at δC 205.8, 196.8, five aromatic quaternary carbons at δC 159.1, 146.1, 142.9, 131.4, 122.1 were observed. Its 1H and 13C NMR spectroscopic data of 2 were very similar to those of the known compound GTRI-02 (1),8, 9, 10 apart from the presence of two additional methylenes. Moreover, the 1H-1H COSY experiment gave a propyl fragment in 2. These findings suggested a propyl instead of a methyl located at C-8 in 2. HMBC correlations between H-11 (δH 2.72, 2.66) and C-7 (δC 131.4), C-8 (δC 142.9) and C-8a (δC 122.1), between H-12 (δH 1.39) and C-8 (δC 142.9) confirmed the propyl was linked to C-8. The configuration of the chiral carbon C-3 was assigned as R by comparing the optical rotation value ( −10 (c 0.4, MeOH)) with that of (R)-GTRI-02 ( −16 (c 2.5, MeOH)) which had been obtained by biomimetic asymmetric synthesis and the absolute configuration was determined by CD spectroscopy.11 Consequently, 2 was identified as (R)-7-acetyl-3,6-dihydroxy-8-propyl-3,4-dihydronaphthalen-1(2H)-one.

Table 1 1H (600 MHz, DMSO-d6) and 13C (150 MHz, DMSO-d6) NMR data of compound 2

Compound 3 was obtained as a colorless, amorphous solid. The HRESIMS exhibited a prominent ion peak at m/z 257.1345 [M-H], which established the molecular formula of 3 as C15H18N2O2. 1H NMR spectrum (Table 2) of 3 indicated signals of four aromatic protons (δH 7.41, brs; δH 7.28, d, J=8.3 Hz; δH 7.28, brs; δH 7.05, d, J=8.3 Hz), one oxygenated methine (δH 3.37, m), two methylene groups (δH 4.00, 2H, s; δH 3.02, brd, J=13.6 Hz; δH 2.43, dd, J=13.6, 10.3 Hz), two singlet methyls (δH 1.14; δH 1.11). Moreover, three active protons were observed (δH 10.96, brs; δH 4.30, d, J=6.0 Hz; δH 4.24, s). 13C NMR spectrum (Table 2) of 3 presented 15 carbon signals including six carbons without directly bonded protons, five methines, two methylenes, two methyls with the aid of HSQC spectrum. These data showed 3 was a tryptophan derivative and related to 5-dimethylallylindole-3-acetonitrile.12 The major difference between these two compounds was the side chain at C-5. Two more oxygenated carbons were present and a double bond was absent in 3, which indicated the double bond in 5-dimethylallylindole-3-acetonitrile was oxygenated to vicinal diol in 3. The planar structure of 3 was further determined by HMBC correlations (Figure 2). The absolute configuration of C-9 was elucidated by Mo2(AcO)4-induced CD method.13 The CD spectrum showed a positive Cotton effect at 313 nm suggesting S configuration for C-2′ (Figure 3). Thus, 3 was identified as (S)-2-(5-(2,3-dihydroxy-3-methylbutyl)-1H-indol-3-yl) acetonitrile.

Table 2 1H (600 MHz, DMSO-d6) and 13C (150 MHz, DMSO-d6) NMR data of compound 3
Figure 2
figure 2

1H-1H COSY and key HMBC correlations for compounds 24.

Figure 3
figure 3

Mo2(AcO)4-induced CD spectrum of 3.

Compound 4 was isolated as a white, amorphous solid. Its molecular formula was determined as C14H25NO3 by HRESIMS at m/z 256.1992 [M+H]+. The 1H and 13C NMR data indicated 4 was a fatty acid amide, consisting of 14 carbons including four olefinic carbons and two oxygenated carbons. HSQC experiment distinguished the protons linked to carbons. The uninterrupted 1H-1H COSY correlations from H-2 to H3-14 determined the linkage of all the carbons. The planar structure of 4 was further confirmed by HMBC correlations from H-8 (δH 5.49) to C-6 (δC 128.9), C-7 (δC 25.9), C-10 (δC 31.5), from H-11 (δH 3.21) to C-9 (δC 128.7), C-12 (δC 75.3), C-13 (δC 25.9), from H-12 (δH 3.12) to C-10 (δC 31.5), C-11 (δC 74.2), C-14 (δC 10.7), from H-3 (δH 1.53) to C-1 (δC 174.6), C-5 (δC 129.7), from H-2 (δH 2.04) to C-1 (δC 174.6). Both the double bonds at C-5/C-6, C-8/C-9 elucidated as Z geometry based on the coupling constant of H-8 (δH 5.49, dt, J=10.8, 7.2 Hz) and the 13C NMR data of C-7 (δC 25.9).14, 15 The relative configuration of C-11 and C-12 was deduced as threo on the basis of the coupling constant 3JH-l1/H-12=6.2 Hz.16, 17, 18 Thus, the structure of 4 was established as (5Z, 8Z, 11 R*, 12 R*)-11,12-dihydroxytetradeca-5,8-dienamide.

Compounds 14 were tested for anti-inflammatory activity by examining their ability to inhibit NO production in LPS-stimulated RAW 264.7 macrophage cells. The effects of these compounds on cell viability were also determined by the MTT method, and none of these compounds exhibited cytotoxicities at 100 μm. Compared to the vehicle-treated group, compound 2 exhibited a moderate dose-dependent decrease in the NO production (P<0.01) at the concentrations of 100, 33.3 and 11.1 μm (Figure 4). While compounds 1, 3, 4 didn’t show obvious inhibition at the tested concentrations. The IC50 value of compound 2 was further determined to be 51.2 μm. Minocycline was used as a positive control with IC50 value of 25.3 μm.

Figure 4
figure 4

Effect of compound 2 on the inhibition of LPS-induced NO in Raw 264.7 cells.

Methods

General experimental procedures

Optical rotations were determined using an Anton Paar MCP200 automatic polarimeter (Graz, Austria). IR spectrum was recorded with a Bruker Tensor 27 FT–IR spectrometer (film). UV detection was carried out with a Beckman Coulter DU 730 nucleic acid/protein analyzer (Brea, CA, USA). CD spectra were acquired on a Biologic MOS-450 spectra polarimeter (Biologic Science, Claix, France). 1D and 2D NMR spectra were recorded on a Bruker AV-600 spectrometer (Faellanden, Switzerland). ESI-MS were recorded on an Agilent 1290-6420 Triple Quadrupole LC-MS spectrometer (Santa Clara, CA, USA). HRESI-MS was measured with an Agilent G6230 TOF mass spectrometer. Biological assays were analyzed using a microplate reader (BioTek Synergy H1, BioTek Instruments, Winooski, VT, USA).

Microbial material

The producing organism was isolated from fecal samples excreted by healthy adult E. burchelli living in Yunnan Wild Animal Park, Kunming, Yunnan province, China. The strain was identified as S. violaceoruber on the basis of morphological characteristics and 16 S ribosomal RNA gene sequences analysis. Phylogenetic analysis suggested that the strain shared a higher 16 S ribosomal RNA gene sequences similarity (100%) with the closely related strain S. violaceoruber (accession number AB184174). The strain (No. YIM 101131) was deposited at the Yunnan Institute of Microbiology, Yunnan University, China.

Fermentation, extraction and isolation

The strain, grown on agar plate, was prepared to inoculate 500 ml Erlenmeyer flasks each containing 100 ml of sterile seed medium composed of glucose 4.0%, malt extract 5.0%, multiple vitamin solution 1.0 ml per liter and trace element solution 1.0 ml per liter at a pH of 7.2 with no adjustment. These flasks cultures were incubated at 28 °C for 2 days on a rotary shaker set at 180 r.p.m. For large-scale fermentation, 20 ml of seed medium was used to transferred into 1000 ml Erlenmeyer flasks containing 200 ml of sterile fermentation medium composed of soybean meal 10 g, peptone 2.0 g, glucose 20 g, soluble starch 5.0 g, yeast extract 2.0 g, NaCl 4 g, K2HPO4 0.5 g, MgSO4·7H2O 0.5 g and CaCO3 2.0 g in 1 l tap water at a pH of 7.8 with no adjustment. The fermentation batches were cultured at 28 °C for 7 days on a rotary shaker set at 180 r.p.m.

The mycelium and broth filtrate (70 l) were separated by centrifugation (4000 r.p.m, 5 min). The resultant aqueous phase filtrate was extracted with EtOAc for three times. The organic partition layer was collected and evaporated in vacuo to yield 13 g of residue. The dried crude extract was further separated by chromatography on a silica gel column (CH2Cl2–MeOH, gradient 50:1-1:1) to obtain 10 fractions (Fr. 1–10). Fraction 4 was subjected to Sephadex LH-20 chromatography (MeOH) to afford six subfractions (Fr. 4.1–4.6). Fraction 4.3 was purified using semi-preparative reversed-phase HPLC (MeOH–H2O 55:45, 10 ml min−1) to yield compound 3 (5.5 mg). Fraction 8 was subjected to Sephadex LH-20 chromatography (MeOH) to produce six subfractions (Fr. 8.1–8.6). Subfraction 8.5 was applied to semi-preparative reversed-phase HPLC (MeOH-H2O 40:60, 10 ml min−1) to afford 2 (2.0 mg) and 1 (10 mg), respectively. Fraction 10 was separated by silica gel column chromatography (petroleum ether–EtOAc 1:5) to afford six subfractions (Fr. 10.1–10.6). Subfraction 10.2 was purified by semi-preparative reversed-phase HPLC (MeOH–H2O 57:43, 10 ml min−1) to yield 4 (4.2 mg).

(R)-7-acetyl-3,6-dihydroxy-8-propyl-3,4-dihydronaphthalen-1(2H)-one ( 2): white, amorphous solid; −10 (c 0.4, MeOH); UV (MeOH) λmax (log ɛ) 205 (4.40), 275 (4.22) nm; IR (film) νmax 3270, 2959, 2927, 1698, 1667, 1582, 1276, 1236, 1211, 1026 cm−1; ESI-MS m/z 263 [M+H]+, 285 [M+Na]+; HRESI-MS m/z 263.1286 [M+H]+(calcd for C15H19O4, 263.1283). For 1H and 13C NMR data see Table 1.

(S)-2-(5-(2,3-dihydroxy-3-methylbutyl)-1H-indol-3-yl) acetonitrile (3): colorless, amorphous solid; −51.8 (c 0.54, MeOH); UV (MeOH) λmax (log ɛ) 219 (4.48), 274 (3.72) nm; IR (film) νmax 3351, 2971, 2923, 2251, 1660, 1604, 1429, 1381, 1348, 1069 cm−1; Mo2(OAc)4 induced CD λmax (Δɛ) 313 (1.34); ESI-MS m/z 281 [M+Na]+; HRESI-MS m/z 281.1265 [M+Na]+ (calcd for C15H18N2O2 Na, 281.1266); For 1H and 13C NMR data see Table 2.

(5Z, 8Z, 11 R*, 12 R*)-11,12-dihydroxytetradeca-5,8-dienamide (4): white, amorphous solid; −25 (c 0.36, MeOH); UV (MeOH) λmax (log ɛ) 219 (3.90) nm; IR (film) νmax 3331, 2927, 1662, 1612, 1405 cm−1; ESI-MS m/z 256 [M+H]+, 278 [M+Na]+; HRESI-MS m/z 278.1719 [M+Na]+ (calcd for C14H25NO3Na, 278.1732); 1H NMR (600 MHz, DMSO-d6) δ: 7.23 (1H, brs, NH), 6.69 (1H, brs, NH), 5.49 (1H, dt, J=10.8, 7.2 Hz, H-8), 5.34 (3H, m, H-5, H-6, H-9), 4.40 (1H, d, J=5.9 Hz, 11-OH), 4.33 (1H, d, J=5.9 Hz, 12-OH), 3.21 (1H, m, H-11), 3.12 (1H, m, H-12), 2.75 (2H, m, H-7), 2.35 (1H, m, H-10), 2.04 (2H, t, J=7.5 Hz, H-2), 2.01 (3H, m, H-10, H-4), 1.58 (1H, m, H-13), 1.53 (2H, m, H-3), 1.25 (1H, m, H-13), 0.87 (3H, t, J=7.5 Hz, H-14); 13C NMR (150 MHz, DMSO-d6) δ: 174.6 (C-2, s), 35.1 (C-3, t), 25.6 (C-4, t), 26.8 (C-5, t), 129.7 (C-5, d), 128.9 (C-6, d), 25.9 (C-7, t), 128.3 (C-8, d), 128.7 (C-9, d), 31.5 (C-10, t), 74.2 (C-11, d), 75.3 (C-12, d), 25.9 (C-13, t), 10.7 (C-14, q). 1H NMR (600 MHz, CD3OD) δ: 5.42 (1H, brdt, J=10.8, 7.3 Hz, H-9), 5.34 (1H, brdt, J=10.8, 7.3 Hz, H-8), 5.32–5.26 (2H, m, H-5, H-6), 3.31 (1H, ddd, J=8.6, 6.2, 3.7 Hz, H-11), 3.23 (1H, m (overlapped), H-12), 2.74 (2H, m, H-7), 2.35 (1H, brddd, J=14.9, 7.3, 3.7 Hz, H-10), 2.12 (2H, t, J=7.3 Hz, H-2), 2.07 (1H, ddd (quint-like), J=14.9, 8.6, 7.3 Hz, H-10), 2.02 (2H, q, J=7.3 Hz, H-4), 1.60 (1H, m, H-13), 1.57 (2H, quint, J=7.3 Hz, H-3), 1.29 (1H, m, H-13), 0.89 (3H, t, J=7.3 Hz, H-14).

Inhibition of NO production assay

The nitrite concentration in the medium was measured using the Griess reaction as previously described.19 Compounds 1– 4 were tested for their inhibitory activity of NO production at various concentrations (100, 33.3 and 11.1 μm for dose-dependent assay; 10, 20, 40, 60, 80 and 100 μm for IC50 analysis). Minocycline was used as the positive control. Statistical analysis was performed by SPSS (SPSS Inc., Chicago, IL, USA).