Fragilides U-W: New 11,20-Epoxybriaranes from the Sea Whip Gorgonian Coral Junceella fragilis.

Three new 11,20-epoxybriaranes-fragilides U-W (1-3), as well as two known metabolites, junceellonoid D (4) and junceellin (5), were obtained from the octocoral Junceella fragilis. The structures of briaranes 1-3 were elucidated by spectroscopic methods and briaranes 3 and 5 displayed inhibition effects on inducible nitric oxide synthase (iNOS) release from RAW264.7.

1. Introduction

Gorgonian corals belonging to the genus Junceella (family Ellisellidae) [1,2,3], distributed abundantly in the coral reefs of tropical Indo-Pacific Ocean have been found to produce briarane diterpenoids, natural products of a marine origin, in abundance [4]. In our further research into the natural products of Junceella fragilis (Ridley 1884) (Figure 1), which was distributed extensively in the waters of Southern Taiwan, have resulted in the isolation of three new 11,20-epoxybriaranes– fragilides U–W (1–3) along with two know compounds junceellonoid D (4) [5] and junceellin (5) [6,7,8,9,10,11,12,13,14,15] (Figure 1). An anti-inflammatory assay was employed to evaluate the activity of these isolates against the release of inducible nitric oxide synthase (iNOS) from macrophage cells RAW264.7.

Figure 1

Structures of fragilides U–W (1–3), junceellonoid D (4), junceellin (5), robustolide F (6), juncenolide M (= frajunolide S) (7), and a picture of Junceella fragilis.

2. Results and Discussion

Fragilide U (1) was isolated as an amorphous powder and displayed a sodiated pseudomolecular ion at m/z 589.22583 in the (+)-HRESIMS, which suggested that the molecular formula of 1 was C28H38O12 (calcd. for C28H38O12 + Na, 589.22555) (Ω = 10). The IR spectrum of 1 showed the presence of hydorxy (νmax 3445 cm−1), γ-lactone (νmax 1780 cm−1), and ester (νmax 1733 cm−1) groups. Analysis of the 1H, 13C NMR, and distortionless enhancement by polarization transfer (DEPT) spectra together with the molecular formula, suggested that there must be an exchangeable proton. The 13C NMR spectrum (Table 1), in combination with DEPT and heteronuclear single quantum coherence (HSQC) spectra, revealed the presence of four acetoxy groups (δC 21.7, 21.2, 20.9, 20.8, 4 × CH3; δC 170.1, 169.8, 169.4, 169.0, 4 × C), a γ-lactone moiety (δC 175.9), and a trisubstituted olefin (δC 142.0, C; 120.6, CH). Base on the 13C NMR data and numbers of unsaturation, 1 was established as a diterpenoid featuring with four rings. An exocyclic epoxy group was deduced from the signals of an oxygenated quaternary carbon and an oxymethylene at δC 62.5 and 59.1, respectively, and further supported by the chemical shifts of oxymethylene protons at δH 2.85 (1H, d, J = 4.4 Hz) and 2.98 (1H, dd, J = 4.4, 1.6 Hz). Moreover, a methyl singlet (δH 1.09, 3H, s), a methyl doublet (δH 1.15, 3H, d, J = 7.6 Hz), a vinyl methyl (δH 2.00, 3H, d, J = 1.6 Hz), three pairs of aliphatic methylene protons (δH 2.04, 1H, m; 2.62, 1H, ddd, J = 18.4, 4.0, 2.4 Hz; 1.13, 1H, m; 2.30, 1H, m; 2.11, 1H, m; 1.81, 1H, m), two aliphatic methine protons (δH 2.36, 1H, dd, J = 5.6, 1.6 Hz; 2.33, 1H, q, J = 7.6 Hz), five oxymethine protons (δH 5.91, 1H, br s; 5.65, 1H, d, J = 5.6 Hz; 5.04, 1H, d, J = 8.8 Hz; 5.03, 1H, br s; 4.71, 1H, d, J = 4.8 Hz), an olefin proton (δH 5.67, 1H, dq, J = 8.8, 1.6 Hz), four acetate methyls (δH 2.24, 2.01, 1.99, 1.98, each 3H × s), and a hydroxy proton (δH 4.85, 1H, br s) were observed in the 1H NMR spectrum (Table 2).

Table 1

13C NMR data for briaranes 1–3.

Position	1 a	2 a	3 b
1	47.0, C c	46.1, C	49.1, C
2	72.2, CH	72.1, CH	75.4, CH
3	37.5, CH2	37.6, CH2	136.4, CH
4	70.6, CH	73.4, CH	128.2, CH
5	142.0, C	144.1, C	135.9, C
6	120.6, CH	58.0, CH	128.8, CH
7	76.5, CH	79.0, CH	80.0, CH
8	80.4, C	81.9, C	80.9, C
9	67.4, CH	68.9, CH	67.1, CH
10	39.9, CH	39.8, CH	39.5, CH
11	62.5, C	62.6, C	60.8, C
12	23.6, CH2	23.5, CH2	29.4, CH2
13	24.2, CH2	24.0, CH2	25.5, CH2
14	73.2, CH	73.5, CH	78.9, CH
15	14.6, CH3	15.4, CH3	16.0, CH3
16	21.4, CH3	121.0, CH2	48.0, CH2
17	42.4, CH	43.8, CH	45.2, CH
18	6.6, CH3	7.1, CH3	6.9, CH3
19	175.9, C	175.2, C	174.8, C
20	59.1, CH2	58.2, CH2	50.4, CH2
Acetate methyls	21.7, CH3	21.9, CH3	21.7, CH3
	21.2, CH3	21.0, CH3	21.4, CH3
	20.9, CH3	21.0, CH3	21.3, CH3
	20.8, CH3
Acetate carbonyls	170.1, C	171.0, C	170.4, C
	169.8, C	170.1, C	169.7, C
	169.4, C	169.7, C	169.3, C
	169.0, C

a Spectra measured at 100 MHz in CDCl3. b Spectra measured at 150 MHz in CDCl3. c Multiplicity deduced by distortionless enhancement by polarization transfer (DEPT) and heteronuclear single quantum coherence (HSQC) spectra.

Table 2

1H NMR data (J in Hz) for briaranes 1–3.

Position	1 a	2 a	3 b
2	5.03 br s	4.96 dd (2.8, 2.8)	5.63 d (6.0)
3α/β	2.04 m; 2.62 ddd (18.4, 4.0, 2.4)	1.75 m; 2.38 m	6.04 dd (17.4, 6.0)
4	5.91 br s	4.85 m	6.42 d (17.4)
6	5.67 dq (8.8, 1.6)	5.49 br s	5.66 d (3.6)
7	5.04 d (8.8)	5.15 d (2.4)	5.62 d (3.6)
9	5.65 d (5.6)	5.52 d (6.4)	4.77 d (6.6)
10	2.36 dd (5.6, 1.6)	2.31 d (6.4)	3.32 d (6.6)
12α/β	1.13 m; 2.30 m	1.15 m; 2.30 m	1.29 m; 2.18 m
13α/β	2.11 m; 1.81 m	2.14 m; 1.78 m	2.08 m; 1.85 dddd (12.0, 12.0, 3.6, 1.8)
14	4.71 d (4.8)	4.86 d (4.4)	4.94 dd (3.6, 3.0)
15	1.09 s	1.13 s	0.87 s
16a/b	2.00 d (1.6)	5.74 s; 5.97 s	4.13 d (12.0); 4.19 d (12.0)
17	2.33 q (7.6)	2.29 q (7.6)	2.36 q (7.2)
18	1.15 d (7.6)	1.22 d (7.6)	1.18 d (7.2)
20a/b	2.85 d (4.4); 2.98 dd (4.4, 1.6)	2.83 d (4.0); 2.85 br d (4.0)	2.76 d (2.4); 3.40 dd (2.4, 2.4)
OH-8	4.85 br s	4.63 br s	3.13 d (1.2)
Acetate methyls	2.24 s	2.23 s	2.27 s
	2.01 s	2.03 s	2.25 s
	1.99 s	1.99 s	2.11 s
	1.98 s

a Spectra measured at 400 MHz in CDCl3. b Spectra measured at 600 MHz in CDCl3.

Coupling information in the correlation spectroscopy (COSY) analysis enabled the proton sequences from H-2/H2-3/H-4, H-6/H-7, H-9/H-10, H2-12/H2-13/H-14, and H-17/H3-18 (Figure 2), which was assembled with a heteronuclear multiple bond correlation (HMBC) experiment (Figure 2). The HMBC between protons and quaternary carbons, such as H-2, H-10, H-13β, H-14, H3-15/C-1; H-7, H3-16/C-5; H-9, H-10, H3-18/C-8; H-9, H-10, H2-12, H-13α, H2-20/C-11; and H-17, H3-18/C-19, permitted elucidation of the carbon skeleton of 1. A vinyl methyl at C-5 was confirmed by the HMBC between H3-16/C-4, C-5, C-6 and H-6/C-16; and further supporting by an allylic coupling between H-6 and H3-16 (J = 1.6 Hz). The methyl group on C-1 (Me-15) was substantiated by the HMBC from H3-15/C-1, C-2, C-10, C-14; and H-2, H-10, H-14/C-15. The epoxy group at C-11/20 was confirmed by the HMBC between H2-20/C-10, C-11; and H-20a/C-12; and further supporting by a long range 4J-1H–1H correlation between H-10 (δH 2.36) and H-20b (δH 2.98) (J = 1.6 Hz). A hydroxy group attaching at C-8 was to infer that an HMBC of a hydroxy proton at δH 4.85 to C-7, C-8, and C-9. Moreover, HMBC from the oxymethine protons at δH 5.03 (H-2), 5.65 (H-9), and 4.71 (H-14) to the acetate carbonyls at δC 169.0, 169.4, and 170.1, placed the acetoxy groups on C-2, C-9, and C-14, respectively. Ten of the 12 oxygen atoms in the molecular formula of 1 could be accounted for the presence of a γ-lactone, three esters, an epoxide, and a hydroxy group. Thus, the remaining two oxygen atoms had to be positioned at C-4 as an acetoxy group, as indicated by its 1H and 13C NMR chemical shifts (δH 5.91, 1H, br s; δC 70.6, CH), although no HMBC was observed from H-4 to any acetate carbonyl.

Figure 2

The correlation spectroscopy (COSY) () correlations, selective heteronuclear multiple bond correlation (HMBC) experiment (), and protons with key nuclear Overhauser effect spectroscopy (NOESY) () correlations of 1.

Based on a summary of the 13C chemical shifts of 11,20-epoxy group in naturally occurring briarane analogues, with 13C NMR data for C-11 and C-20 at δC 62–63 and 58–60 ppm, the epoxide group was β-oriented and the cyclohexane ring existed in a twist boat conformation [16]; thus, the configuration of the 11,20-epoxy group in 1 should be β-oriented and the cyclohexane ring was found to be in a twist boat conformation for the 13C chemical shifts at δC 62.5 (C-11) and 59.1 (CH2- 20). The relative stereochemistry of 1 was established by the analysis of correlations observed in a nuclear Overhauser effect spectroscopy (NOESY) experiment (Figure 2). In the NOESY spectrum of 1, NOE correlations between H-10/H-2, H-10/H-9, and H-10/OH-8, while no correlation was seen between H-10 and H3-15, suggesting that these protons H-2, H-9, and H-10, and the hydroxy group at C-8 were α-oriented; meanwhile, an NOE correlation of H3-15 with H-14 indicated that H-14 was β-oriented. The NOESY spectrum showed a correlation from H-6 to H3-16, revealing the Z geometry of C-5/6 double bond. H3-18 exhibited NOE correlations to OH-8 and H-9, suggesting the α-orientation of Me-18 at C-17. H-7 displayed NOE correlations with H-4 and H-17, which further confirmed that these three protons were in β-orientation at C-7, C-4, and C-17. Based on the above findings, the relative configurations of stereogenic carbons of 1 were elucidated as 1R*,2S*,4S*,7S*, 8R*,9S*,10S*,11S*,14S* and 17R*. However, as briaranes 1–4 were isolated along with the known chlorinated briarane, junceellin (5) [6], and the structure, including the absolute configuration of junceellin (5) was further confirmed by a single-crystal X-ray diffraction analysis [7,15]. It is reasonable, therefore, on biogenetic grounds to assume that briaranes 1–4 have the same absolute configuration as that of 5. Therefore, the configurations of the stereogenic carbons of 1 should be elucidated as 1R,2S,4S,7S,8R,9S,10S,11S,14S and 17R (Supplementary Materials, Figures S1–S8).

Our present study has also led to the isolation of a new briarane, fragilide V (2). The molecular formula of C26H35ClO11 was deduced from (+)-HRESIMS at m/z 581.17589 (calcd. for C26H3535ClO11 + Na, 581.17601). Carbonyl resonances in the 13C NMR spectrum of 2 (Table 1) at δC 175.2, 171.0, 170.1, and 169.7 revealed the presence of a γ-lactone and three esters. In the 1H NMR spectrum of 2 (Table 2), the signals for three acetate methyls were observed at δH 2.23, 2.03, and 1.99. It was found that the 1D (Table 1 and Table 2) and 2D NMR (Figure 3) data of 2 were similar to those of a known briarane, robustolide F (6) [17,18] (Figure 1), except that the signals corresponding to a hydroxy group in 2 were replaced by signals for a proton in 6. In the NOESY spectrum, one of the C-3 methylene protons (δH 2.38) showed a correlation to H-7 and not with H-2, suggesting the β-orientation of this proton by modeling study and the other was assigned as H-3α (δH 1.75). A correlation from H-4 to H-3α, suggested that H-4 was α-oriented according to modeling analysis. Therefore, the configuration of the stereogenic carbons of 2 were elucidated as 1R,2S,4R,6S,7R,8R,9S,10S,11S,14S, and 17R (Figure 3) (Supplementary Materials, Figures S9–S16).

Figure 3

The COSY () correlations, selective HMBC (), and protons with key NOESY () correlations of 2.

Briarane 3 (fragilide W) was found to have a molecular formula of C26H33ClO10 based on its (+)- HRESIMS at m/z 563.16554 (calcd. for C26H3335ClO10 + Na, 563.16545). Its absorption peaks in the IR spectrum showed ester, γ-lactone, and broad OH stretching at 1738, 1777, and 3459 cm−1, respectively. The 13C NMR spectrum indicated that three esters and a γ-lactone were present, as carbonyl resonances were observed at δC 174.8, 170.4, 169.7, 169.3 (Table 1). The 1H NMR spectrum indicated the presence of three acetate methyls (δH 2.27, 2.25, 2.11, each 3H × s) (Table 2). The 1H and 13C NMR spectra of 3 was found to be similar with those of a known briarane, juncenolide M (= frajunolide S) (7) (Figure 1), isolated from J. juncea and J. fragilis [19,20], except that the signals corresponding to the 13-acetoxy and 3(Z)-ene moieties in 7 were disappeared and replaced by a proton and an (E)-ene moieties in 3, respectively. The locations of the functional groups were confirmed by 2D-NMR correlations (Figure 4), and hence the structure of fragilide W was assigned as 3, and the configurations of the stereogenic carbons were elucidated as 1R,2S,7S,8R,9S,10S,11R, 14S, and 17R (Figure 4) (Supplementary Materials, Figures S17–S24).

Figure 4

The COSY () correlations, selective HMBC (), and protons with key NOESY () correlations of 3.

The known compound 4 was found to be identical with the known junceellonoid D, on the basis of the comparison of its physical and spectroscopic data with those of reported previously [5,21] (Supplementary Materials, Figures S25–S32).

Using an in vitro pro-inflammatory suppression assay, the activities of briaranes 1–5 on the release of iNOS and cyclooxygenase-2 (COX-2) protein from lipopolysaccharides (LPS)-stimulated RAW264.7 were assayed (Figure 5 and Table 3). The results showed that briaranes 3 and 5 reduced the release of iNOS to 28.55 and 33.72% at a concentration of 10 μM. Briarane 4 was found to be more weak in terms of reducing the expression of iNOS, indicating that the activity of briaranes 4 and 5 is largely dependent on the functional groups at C-2 and C-3. It is interesting to note that briarane 4 was found to enhance the expression of COX-2 to 130.88%, at a concentration of 10 μM.

Figure 5

Activities of briaranes 1–5 on the expression of inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2) proteins in LPS-treated murine RAW264.7 macrophage cells. Western blotting showed that briaranes 3 and 5 reduced the expression of iNOS. Data were normalized to the cells treated with LPS only, and cells treated with dexamethasone were used as a positive control. Data are expressed as the mean ± SEM (n = 3). * Significantly different from cells treated with LPS (p < 0.05).

Table 3

Effects of briaranes 1–5 on LPS-induced pro-inflammatory inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2) protein expression in macrophages at a concentration of 10 μM.

	iNOS	COX-2	β-Actin
Compound	Expression (% of LPS)
Lipopolysaccharides	100.01 ± 17.81	100.00 ± 3.04	100.00 ± 6.05
1	55.88 ± 11.42	64.73 ± 4.07	90.05 ± 7.11
2	77.58 ± 8.82	66.23 ± 6.59	99.29 ± 5.24
3	28.55 ± 5.35	86.46 ± 8.46	101.69 ± 6.46
4	67.25 ± 4.27	130.88 ± 13.94	103.20 ± 2.56
5	33.72 ± 2.57	64.15 ± 3.03	84.52 ± 7.78
Dexamethasone	12.11 ± 0.03	2.11 ± 0.44	123.86 ± 2.99

Data were normalized to those of cells treated with LPS alone, and cells treated with dexamethasone were used as a positive control (10 μM). Data are expressed as the mean ± SEM (n = 3).

3. Materials and Methods

3.1. General Experimental Procedures

Melting points were determined using a Fargo apparatus and the values were uncorrected. 1D and 2D NMR spectra were recorded on a 600 MHz Jeol NMR (model ECZ600R, Tokyo, Japan) or on a 400 MHz Jeol NMR (model ECZ400S) spectrometers using the residual CHCl3 signal (δH 7.26 ppm) and CDCl3 (δC 77.1 ppm) as internal standards for 1H and 13C NMR, respectively. ESIMS and HRESIMS were obtained from the Bruker mass spectrometer with 7 Tesla magnets (model: SolariX FTMS system, Bremen, Germany). Column chromatography, HPLC, IR, and optical rotation were performed according to our earlier research [15].

3.2. Animal Material

Specimens of J. fragilis used for this study were collected in April 2017 by self-contained underwater breathing apparatus (SCUBA) at depths of 10−15 m off the coast of South Bay, Kenting, Taiwan. The samples were stored in a −20 °C freezer until extraction. A voucher specimen was deposited in the NMMBA (voucher no.: NMMBA-TW-GC-2017-022). Identification of the species of this organism was performed by comparison as described in previous studies [1,2,3].

3.3. Extraction and Isolation

Sliced bodies (wet/dry weight = 795/313 g) of the coral specimen were prepared and extracted with a 1:1 mixture of methanol (MeOH) and dichloromethane (CH2Cl2) (1:1) to give a crude extract (19.0 g) which was partitioned between ethyl acetate (EtOAc) and H2O. The EtOAc extract (8.0 g) was then applied to a silica gel column chromatograph (C.C.) and eluted with gradients of n- hexane/acetone (stepwise from 50:1 to 1:2; volume ratio) to furnish 8 fractions (fractions: A−H). Fraction F was chromatographed on silica gel C.C. and eluted with gradients of n-hexane/EtOAc (2:1 to 1:2, stepwise) to furnish 4 fractions (fractions F1−F4). Fraction F3 was washed with a mixture of n-hexane/acetone (30:1) and the undissolved 5 (23.5 mg) was obtained. Fraction G was purified by normal-phase HPLC (NP-HPLC) using a mixture of n-hexane and EtOAc (4:1 to 1:1, stepwise) to afford 16 fractions (fractions G1−G16). Afterward, fraction G15 was separated by NP-HPLC using a mixture of CH2Cl2 and acetone (v:v = 10:1; at a flow rate = 2.0 mL/min) to yield 1 (1.7 mg). Fraction G14 was separated by NP-HPLC using a mixture of n-hexane/acetone (2:1; volume ratio) to yield 6 fractions (fractions: G14A−G14F). Fractions G14C and G14D were combined and purified by reverse-phase HPLC (RP-HPLC) using a mixture of MeOH and H2O (v:v = 65:35; at a flow rate = 4.0 mL/min) to afford 2 (0.8 mg). Fraction G9 was purified by RP-HPLC using a mixture of MeOH and H2O (v:v = 60:40; at a flow rate = 4.0 mL/min) to yield 16 fractions (fractions G9A−G9P), including compound 3 (0.8 mg, G9D). Fraction G9M was separated by RP-HPLC using a mixture of acetonitrile and H2O (v:v = 55:45; at a flow rate = 4.0 mL/min) to obtain 4 (0.8 mg).

Fragilide U (1): Amorphous powder; −12 (c 0.09, CHCl3); IR (KBr) νmax 3445, 1780, 1733 cm−1; 13C (100 MHz, CDCl3) and 1H (400 MHz, CDCl3) NMR data, see Table 1; Table 2; ESIMS: m/z 589 [M + Na]+; HRESIMS: m/z 589.22583 (calcd. for C28H38O12 + Na, 589.22555).

Fragilide V (2): Amorphous powder; −26 (c 0.04, CHCl3); IR (KBr) νmax 3444, 1779, 1738 cm−1; 13C (100 MHz, CDCl3) and 1H (400 MHz, CDCl3) NMR data, see Table 1; Table 2; ESIMS: m/z 581 [M + Na]+, 583 [M + 2 + Na]+; HRESIMS: m/z 581.17589 (calcd. for C26H3535ClO11 + Na, 581.17601).

Fragilide W (3): Amorphous powder; −21 (c 0.04, CHCl3); IR (KBr) νmax 3459, 1777, 1738 cm−1; 13C (150 MHz, CDCl3) and 1H (600 MHz, CDCl3) NMR data, see Table 1; Table 2; ESIMS: m/z 563 [M + Na]+, 565 [M + 2 + Na]+; HRESIMS: m/z 563.16554 (calcd. for C26H3335ClO10 + Na, 563.16545).

Junceellonoid D (4): Amorphous powder; −18 (c 0.04, CHCl3) (ref. [5] −44.8 (c 0.10, CHCl3, MeOH); ref. [21] −31 (c 0.05, CHCl3)); IR (ATR) νmax 3509, 1793, 1733 cm−1; 13C and 1H NMR data were found to be in full agreement with those reported previously [21]; ESIMS: m/z 521 [M + Na]+, 523 [M + 2 + Na]+; HRESIMS: m/z 521.15469 (calcd. for C24H3135ClO9 + Na, 521.15488).

Junceellin (5): Colorless crystals; mp 277–278 °C (ref. [15] mp 272–275 °C); −3 (c 1.18, CHCl3) (ref. [15] −2 (c 0.89, CHCl3)); IR (KBr) νmax 1794, 1743 cm−1; 13C and 1H NMR data were found to be in full agreement with those reported previously [6,8,9,12]; ESIMS: m/z 605 [M + Na]+, 607 [M + 2 + Na]+; HRESIMS: m/z 605.17612 (calcd. for C28H3535ClO11 + Na, 605.17601).

3.4. In Vitro Anti-Inflammatory Assay

The anti-inflammatory assay was employed to evaluate the activities of briaranes 1–5 reduce the release of iNOS and COX-2 from macrophage cells as the literature reported [22,23,24,25].

4. Conclusions

J. fragilis was proven to be a rich source to produce a wide structural diversity of briarane-type diterpenoids that possess various biomedical properties, particularly in anti-inflammatory activity [26,27]. In our continued study on J. fragilis, three previously unreported 11,20-epoxybriaranes, fragilides U–W (1–3), along with two known briaranes, junceellonoid D (4) and junceellin (5), were isolated. The exocyclic 11,20-epoxy group was proven to be a chemical marker for briarane-type natural products from the gorgonian corals belonging to the family Ellisellidae [28]. In the present study, the anti-inflammatory activity of 1–5 was assayed using inhibition of iNOS and COX-2 and the results indicated that fragilide W (3) and junceellin (5) showed the most potent suppressive effect on iNOS release.