**2. Results and Discussion**

Briarenol I (**1**) was isolated as an amorphous powder and displayed a sodiated adduct ion at *m*/*z* 649.24677 in the (+)-HRESIMS, which indicated its molecular formula was C30H42O14 (calculated for C30H42O14 + Na, 649.24668; unsaturation degrees = 10). The IR spectrum revealed absorptions for hydroxy (νmax 3524 cm–1), γ-lactone (νmax 1783 cm–1), and ester carbonyl (νmax 1736 cm–1) moieties. Resonances in the 13C NMR of **<sup>1</sup>** at <sup>δ</sup><sup>C</sup> 172.9, 172.3, 170.5, 170.0, and 170.0 (5 <sup>×</sup> C) supported the presence of a γ-lactone and four esters (Table 1). Three of the esters were identified as acetates by the presence of three methyl singlet resonances in the 1H NMR spectrum at δ<sup>H</sup> 2.34, 2.15, and 2.08 (Table 2) and the remaining ester was found to be an *n*-butyroxy group based on 1H NMR studies, including a correlation spectroscopy (COSY) experiment, which revealed seven contiguous protons (δ<sup>H</sup> 2.30, 2H, t, *J* = 7.2 Hz; 1.63, 2H, tq, *J* = 7.2 Hz; 0.95, 3H, t, *J* = 7.2 Hz). From the COSY spectrum (Figure 2), the proton sequences from H-6/H-7, H-9/H-10/H-11/H-12/H2-13/H-14, and H-11/H3-20 were established. The hydroxy proton signals at δ<sup>H</sup> 4.30 (1H, d, *J* = 12.0 Hz), 1.49 (1H, d, *J* = 4.0 Hz), and 3.49 (1H, dd, *J* = 9.6, 4.4 Hz) were found to correlate with H-3 (δ<sup>H</sup> 4.59, d, *J* = 12.0 Hz), H-12 (δ<sup>H</sup> 4.10, m), and H2-16 (δ<sup>H</sup> 4.35, 1H, dd, *J* = 13.6, 4.4 Hz; 4.04, 1H, dd, *J* = 13.6, 9.6 Hz), respectively. Thus, the hydroxy groups should be positioned at C-3, C-12, and C-16, respectively. Olefinic resonances in the 13C NMR at δ<sup>C</sup> 125.5 (CH-6) and 142.0 (C-5) indicated the presence of a trisubstituted carbon–carbon double bond. On the basis of these data and the heteronuclear multiple bond correlation (HMBC) experiment (Figure 2), the connectivity from C-1 to C-14 was established. A hydroxymethyl group at C-5 was revealed by the HMBC between C-16 oxymethylene protons to C-4, C-5, and C-6. The C-15 methyl group at C-1 was confirmed by the HMBC between H3-15/C-1, C-2, C-10, C-14, and H-10/C-15. The *n*-butyrate positioned at C-4 was confirmed from the connectivity between H-4 (δ<sup>H</sup> 6.14) and the carbonyl carbon of *n*-butyrate group (δ<sup>C</sup> 172.3). HMBC from the oxymethine protons at δ<sup>H</sup> 4.53 (H-2), 5.32 (H-9), and 4.88 (H-14) to the acetate carbonyls at δ<sup>C</sup> 172.9, 170.0, and 170.0, placed the acetoxy groups on C-2, C-9, and C-14, respectively. Thirteen of the fourteen oxygen atoms in the molecular formula of 1 could be accounted for from the presence of a γ-lactone, four esters, and three hydroxy groups. The remaining oxygen atom had to be placed between C-8 and C-17 to form a tetrasubstituted epoxide based on the 13C NMR evidences at δ<sup>C</sup> 70.8 (C-8) and 62.5 (C-17) and the 1H NMR chemical shift of a tertiary methyl at δ<sup>H</sup> 1.66 (3H, s, H3-18).


**Table 1.** 13C NMR (δ<sup>C</sup> 100 MHz, CDCl3) data for briaranes **1**–**3**.

<sup>a</sup> Multiplicity deduced by DEPT and HSQC spectra.

**Figure 2.** The COSY ( ) correlations, selective HMBC ( ), and protons with key NOESY correlations ( ) of **1**.


**Table 2.** 1H NMR (δH, 400 MHz in CDCl3) data (*J* in Hz) for briaranes **1**–**3**.

The stereochemistry of **1** was deduced from an NOESY experiment (Figure 2) and biogenetic considerations. The NOE correlations of H-10/H-11, H-10/H-12, and H-11/H-12 indicated that these protons were situated on the same face of the structure and were assigned as the α protons since the C-15 methyl is the β-substituent at C-1. The NOE correlation between H3-15 and H-14 implied that H-14 had a β-orientation. H-3 exhibited a correlation with H-10, and, as well as a lack of coupling constants were detected between H-2/H-3 and H-3/H-4, indicating the dihedral angles between H-2/H-3 and H-3/H-4 were approximately 90◦ and the 2-acetoxy, 3-hydroxy, and 4-*n*-butyroxy groups were β-, β-, and α-oriented, respectively. A correlation from H-4 to H-7, suggested that H-7 was β-oriented. The *Z*-configuration of C-5/6 double bond was confirmed based on the fact that the C-6 olefinic proton (δ<sup>H</sup> 5.53) correlated to one of the C-16 hydroxymethyl protons (δ<sup>H</sup> 4.04). H-9 was found to correlate with H-11, H3-18, and H3-20. From a consideration of molecular model, H-9 was found to be reasonably close to H-11, H3-18, and H3-20, thus, H-9 should be placed on the α face, and Me-18 was β-oriented in the γ-lactone moiety, and the 8,17-epoxy group should be α-oriented. It was found that the NMR signals of **1** were similar to those of a known briarane, briaexcavatolide P (**4**) (Figure 1) [8], except that the signals corresponding to the Me-16 vinyl methyl in **4** were replaced by signals for a hydroxymethyl group in **1**. Additionally, as briaranes **1**–**5** were isolated along with a known briarane, excavatolide B (**6**) [6,10,11] from the same target organism, *B. excavatum*, and the absolute configuration of **6** was determined by a single-crystal X-ray diffraction analysis [6,11]. Therefore, on biogenetic grounds to assume that briaranes **1**–**5** had the same absolute stereochemistry as that of **6**, tentatively, and the configurations of stereogenic carbons of **1** were determined as 1*R*,2*R*,3*S*,4*R*,7*S*,8*S*, 9*S*,10*S*,11*R*,12*S*,14*S*, and 17*R* (Supplementary Materials, Figures S1–S10).

Briarenol J (**2**) had a molecular formula C26H36O12 by its (+)-HRESIMS at *m*/*z* 563.21007 (calculated for C26H36O12 + Na, 563.20990). The IR spectrum showed bands at 3483, 1779, and 1727 cm−1, consistent with the presence of hydroxy, γ-lactone, and ester groups, respectively, in **2**. From the 13C and DEPT data (Table 2), one trisubstituted double bond was deduced from the signals of two carbons at δ<sup>C</sup> 139.3 (C-5) and 124.3 (CH-6). A methyl-containing tetrasubstituted epoxy group was confirmed from the signals of two oxygenated quaternary carbons at δ<sup>C</sup> 69.9 (C-8) and 61.8 (C-17), and from the

chemical shift of a tertiary methyl (δ<sup>H</sup> 1.66, 3H, s; δ<sup>C</sup> 10.3, CH3-18; Tables 1 and 2). Four carbonyl resonances at δ<sup>C</sup> 170.9, 170.0, 169.5, and 169.2 in the 13C spectrum confirmed the presence of a γ-lactone and three esters. All the esters were identified as acetates by the presence of three methyl singlet resonances in the 1H NMR spectrum at δ<sup>H</sup> 2.32, 2.16, and 2.14, respectively.

Coupling constants information in the COSY spectrum of **2** enabled identification of H-6/H-7, H-9/H-10/H-11/H-12/H2-13/H-14, H-11/H3-20, and H-6/H3-16 (by allylic coupling; Figure 3), these data, together with the HMBC experiment (Figure 3), the molecular framework of **2** could be established. The HMBC also indicated that the acetoxy groups should be attached at C-4, C-9, and C-14, respectively. Thus, the remaining hydroxy groups have to be positioned at C-2, C-3, and C-12, as indicated by the COSY correlations between H-2/OH-2, H-3/OH-3, and H-12/OH-12.

**Figure 3.** The COSY ( ) correlations, selective HMBC ( ), and protons with key NOESY correlations ( ) of **2**.

The stereochemistry of **2** was elucidated from the NOE interactions observed in a NOESY experiment (Figure 3) and by the vicinal 1H−1H coupling constant analysis. In the NOESY spectrum, correlations were observed between H-10 with H-3 and H-12; and H-12 correlated with H-11, indicating that these protons should be α-oriented. H-14 gave a correlation with H3-15, confirming the β-orientation for this proton. H-2 showed a correlation with H-14, and a lack of coupling constant was detected between H-2/H-3, indicating the dihedral angle between H-2/H-3 is approximately 90◦ and the 2-hydroxy group was β-oriented. H-4 exhibited correlations with H-7 and 2-hydroxy proton, confirming the β-orientations for H-4 and H-7. H-9 was found to show correlations with H-11, H3-18, and H3-20, and from molecular models, H-9 and H3-18 should be placed on the α- and β-face, respectively. The *Z*-configuration of C-5/C-6 double bond was elucidated by a correlation between H-6 and H3-16. The NMR data of **2** were found to be similar to those of a known briarane, briaexcavatin P (**5**) [9]. It was found that the 2-acetoxy substituent in **5** was replaced by a hydroxy group in **2**. By comparison of the proton and carbon chemical shifts, coupling constants, NOESY correlations, and rotation value of **2** with those of **5**, the stereochemistry of **2** was confirmed to be the same as that of **5**, and the configurations of the stereogenic centers of **2** were assigned as 1*S*,2*R*,3*R*,4*R*, 7*S*,8*S*,9*S*,10*S*,11*R*,12*S*,14*S*, and 17*R* (Supplementary Materials, Figures S11–S20).

Briarane **3** (briarenol K) was found to have a molecular formula of C26H36O11 based on its (+)-HRESIMS peak at *m*/*z* 547.21514 (calculated for C26H36O11 + Na, 547.21498). Its absorption peaks in the IR spectrum showed ester carbonyl, γ-lactone, and broad OH stretching at 1739, 1780, and 3468 cm−1, respectively. The 13C NMR spectrum indicated that three esters and a γ-lactone were present, as carbonyl resonances were observed at δ<sup>C</sup> 168.1, 170.2, 170.4, and 170.4, respectively (Table 1). The 1H NMR data also indicated that presence of three acetate methyls at δ<sup>H</sup> 2.22, 2.03, and 2.00 (each 3H × s; Table 2). It was found that the spectroscopic data of **3** were similar to those of a known briarane, briareolide B (**7**) [12]; however, by comparison of the 1H and 13C NMR chemical shifts of CH-12 oxymethine (δ<sup>H</sup> 3.72, 1H, dd, *J* = 12.4, 4.8 Hz; δ<sup>C</sup> 73.4), CH2-13 sp<sup>3</sup> methylene (δ<sup>H</sup> 1.67, 1H,

m; 2.04, 1H, m; δ<sup>C</sup> 30.2), C-11 oxygenated quaternary carbon (δ<sup>C</sup> 78.2), and Me-20 tertiary methyl (δ<sup>H</sup> 1.15, 3H, s; δ<sup>C</sup> 16.9) of **3** with those of **7** (δ<sup>H</sup> 3.56, 1H, m; δ<sup>C</sup> 73.9, CH-12; δ<sup>H</sup> 2.03, 1H, m; 2.12, 1H, m; δ<sup>C</sup> 27.6, CH2-13; δ<sup>C</sup> 74.7, C-11; δ<sup>H</sup> 1.16, 3H, s; δ<sup>C</sup> 22.5, Me-20) [12] showed that the hydroxy group at C-12 in **3** was β-oriented. The locations of the functional groups were further confirmed by other HMBC and COSY correlations (Figure 4), and hence briarenol K was assigned the structure of **3**. The NOESY spectrum exhibited a correlation from H-10 to H-12, further supporting that H-12 was α-oriented and the stereogenic centers of **3** were assigned as 1*S*,2*S*,7*S*,8*S*,9*S*,10*S*,11*S*,12*S*,14*S*, and 17*R*, by the correlations observed in a NOESY spectrum (Figure 4) and this compound was found to be the 12-epimer of briareolide B (**7**) [12] (Supplementary Materials, Figures S21–S30).

**Figure 4.** The COSY ( ) correlations, selective HMBC ( ), and protons with key NOESY correlations ( ) of **3**.

The inhibition effects of briaranes **1**–**5** on the release of inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2) protein from lipopolysaccharides (LPS)-stimulated RAW 264.7 were assessed. The results showed that briarane **4** reduced the release of iNOS and COX-2 to 35.37% and 54.61% at a concentration of 10 μM, respectively (Figure 5 and Table 3). Briarane **1** was found to be weaker than those of **4** in term of reducing the expression of iNOS and COX-2, indicating that the hydroxy group at C-16 in **1** reduced the activity.

**Figure 5.** Western blotting showed that briarane **4** downregulated the expression of iNOS and COX-2. Data were normalized to the cells treated with LPS only, and cells treated with dexamethasone (Dex; 10 μM) were used as a positive control. Data are expressed as the mean ± SEM (*n* = 2~4). \* Significantly different from cells treated with LPS (*p* < 0.05).


**Table 3.** Effects of briaranes **1**–**5** on LPS-induced pro-inflammatory iNOS and COX-2 protein expression in macrophages.

Data were normalized to those of cells treated with LPS alone, and cells treated with dexamethasone were used as a positive control. Data are expressed as the mean ± SEM (*n* = 2–4).
