Briarenols W–Z: Chlorine-Containing Polyoxygenated Briaranes from Octocoral Briareum stechei (Kükenthal, 1908)

Briareum stechei is proven to be a rich source of 3,8-cyclized cembranoids (briarane) with a bicyclo[8.4.0] carbon core. In the present study, four previously unreported briaranes, briarenols W–Z (1–4), along with solenolide A (5), briarenolide M (6), briaexcavatolide F (7), and brianolide (8), were isolated and characterized through spectroscopic analysis, and the absolute configuration of 8 was corroborated by a single-crystal x-ray diffraction analysis. Briaranes 2 and 5 were found to induce significant inflammatory activity in lipopolysaccharide (LPS)-induced RAW 264.7 mouse macrophage cells by enhancing the expression of the inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2) proteins.

Briarenol W (1) was obtained as an amorphous powder. The positive mode electrospray ionization mass spectrum ((+)-ESIMS) showed a pair of peaks at m/z 461/463 ([M + Na] + /[M + 2 + Na] + ) (3:1), with a relative intensity suggestive of a chlorine atom. NMR data coupled with the [M + Na] + peak in the (+)high-resolution ESIMS ((+)-HRESIMS) at m/z 461.13377 suggested a molecular formula C 22 H 27 ClO 7 (calculated for C 22 H 27 35 ClO 7 + Na, 461.13375) that indicated nine degrees of unsaturation. The IR spectrum indicated the presence of hydroxy (ν max 3430 cm −1 ), γ-lactone (ν max 1780 cm −1 ), ester carbonyl (ν max 1733 cm −1 ), and α,β-unsaturated ketonic (ν max 1670 cm −1 ) groups. The 13 C NMR spectrum of 1 (Table 1) showed signals of 22 carbons. The multiplicity of the carbon signals was determined from the distortionless enhancement by polarization transfer (DEPT) and heteronuclear single quantum coherence (HSQC) spectra: four methyls, two methylenes (one olefin), ten methines (five bearing a heteroatom and two olefins), and six non-protonated carbons (three carbonyls, one olefin, and one bearing a heteroatom). From the 13 C and 1 H NMR spectra (Tables 1 and 2), 1 was found to possess a γ-lactone (δ C 175. 3 (Figure 2), were fit to the regiochemistry of vicinal couplings in 1. The fused tetracyclic network was established by heteronuclear multiple bond coherence (HMBC) experiments, particularly by the 2 Jand 3 J-1 H-13 C long-range correlations between protons and non-protonated carbons such as H-9, H-10, H-13, H-14, H 3 -15/C-1; H-7, H-16b/C-5; H-4, H 3 -18, OH-9/C-8; H-11, H-14, H 3 -20/C-12; and H-17, H 3 -18/C-19, thus permitted elucidation of the main carbon skeleton of 1 (Figure 2). The Me-20, Me-18, and Me-15 at C-11, C-17, and C-1 were confirmed by the HMBC correlations between H 3 -20/C-10, C-11, C-12; H 3 -18/C-8, C-17, C-19; and H 3 -15/C-1, C-2, C-10, C-14, respectively. An exocyclic double bond at C-5 was confirmed by the HMBC correlations between H 2 -16/C-4, C-5, and C-6. The hydroxy proton signal at δ H 2.84 was revealed by its 1 H-1 H COSY correlation to H-9 (δ H 4.53) and an HMBC correlation to C-9 (δ C 76.9), indicating its attachment to C-9. The acetate ester at C-2 was established by a correlation between H-2 (δ H 4.86) and the acetate carbonyl at δ C 170.6, observed in the HMBC spectrum. The methine unit at δ C 54.9 was more shielded than expected for an oxygenated C atom and correlated with the methine proton at δ H 5.52 in the HSQC spectrum. This proton showed a 3 J-correlation with H-7, in the 1 H-1 H COSY spectrum, confirming the attachment of a chlorine atom at C-6. An HMBC correlation between H-4 (δ H 4.87), an oxymethine proton, and C-8, an oxygenated non-protonated carbon resonating at δ C 82.4, suggested that the remaining oxygen atom had to be positioned between C-4 and C-8 to form an ether bridge in 1. thine proton at δH 5.52 in the HSQC spectrum. This proton showed a 3 Jcorrelation with H-7, in the 1 H-1 H COSY spectrum, confirming the attachment of a chlorine atom at C-6. An HMBC correlation between H-4 (δH 4.87), an oxymethine proton, and C-8, an oxygenated non-protonated carbon resonating at δC 82.4, suggested that the remaining oxygen atom had to be positioned between C-4 and C-8 to form an ether bridge in 1. The relative stereochemistry of 1 was established using a nuclear Overhauser effect spectroscopy (NOESY) experiment ( Figure 2) and was found to be compatible with that of 1 offered by computer modeling [19]. In naturally occurring briaranes, proton H-10 and Me-15 at C-1 are α-and β-oriented, respectively [3]. H-9, H-10, and H3-20 protons were proven to be located on the same face of the molecule. These protons, as a result of being correlated together, were assigned as α protons, as Me-15 was a β-substituent at C-1. Correlated with H-10, the H-2 proton had an α-orientation at C-2. Also, H3-18 was found to be associated with H-10, suggesting that the C-18 methyl in the γ-lactone thine proton at δH 5.52 in the HSQC spectrum. This proton showed a 3 Jcorrelation with H-7, in the 1 H-1 H COSY spectrum, confirming the attachment of a chlorine atom at C-6. An HMBC correlation between H-4 (δH 4.87), an oxymethine proton, and C-8, an oxygenated non-protonated carbon resonating at δC 82.4, suggested that the remaining oxygen atom had to be positioned between C-4 and C-8 to form an ether bridge in 1. The relative stereochemistry of 1 was established using a nuclear Overhauser effect spectroscopy (NOESY) experiment ( Figure 2) and was found to be compatible with that of 1 offered by computer modeling [19]. In naturally occurring briaranes, proton H-10 and Me-15 at C-1 are α-and β-oriented, respectively [3]. H-9, H-10, and H3-20 protons were proven to be located on the same face of the molecule. These protons, as a result of being correlated together, were assigned as α protons, as Me-15 was a β-substituent at C-1. Correlated with H-10, the H-2 proton had an α-orientation at C-2. Also, H3-18 was found to be associated with H-10, suggesting that the C-18 methyl in the γ-lactone ), heteronuclear multiple bond coherence (HMBC) ( thine proton at δH 5.52 in the HSQC spectrum. This proton showed a 3 Jcorrelation with H-7, in the 1 H-1 H COSY spectrum, confirming the attachment of a chlorine atom at C-6. An HMBC correlation between H-4 (δH 4.87), an oxymethine proton, and C-8, an oxygenated non-protonated carbon resonating at δC 82.4, suggested that the remaining oxygen atom had to be positioned between C-4 and C-8 to form an ether bridge in 1. The relative stereochemistry of 1 was established using a nuclear Overhauser effect spectroscopy (NOESY) experiment ( Figure 2) and was found to be compatible with that of 1 offered by computer modeling [19]. In naturally occurring briaranes, proton H-10 and Me-15 at C-1 are α-and β-oriented, respectively [3]. H-9, H-10, and H3-20 protons were proven to be located on the same face of the molecule. These protons, as a result of being correlated together, were assigned as α protons, as Me-15 was a β-substituent at C-1. Correlated with H-10, the H-2 proton had an α-orientation at C-2. Also, H3-18 was found to be associated with H-10, suggesting that the C-18 methyl in the γ-lactone ), and protons with nuclear Overhauser effect spectroscopy (NOESY) ( correlation with H-7, in the 1 H-1 H COSY tachment of a chlorine atom at C-6. An H 4 (δH 4.87), an oxymethine proton, and C nated carbon resonating at δC 82.4, sugg gen atom had to be positioned between bridge in 1. The relative stereochemistry of 1 w Overhauser effect spectroscopy (NOES was found to be compatible with that of ing [19]. In naturally occurring briaranes 1 are α-and β-oriented, respectively [3]. were proven to be located on the same fa tons, as a result of being correlated toge tons, as Me-15 was a β-substituent at C-1 2 proton had an α-orientation at C-2. Als ciated with H-10, suggesting that the ) correlations for 1.
The relative stereochemistry of 1 was established using a nuclear Overhauser effect spectroscopy (NOESY) experiment ( Figure 2) and was found to be compatible with that of 1 offered by computer modeling [19]. In naturally occurring briaranes, proton H-10 and Me-15 at C-1 are αand β-oriented, respectively [3]. H-9, H-10, and H 3 -20 protons were proven to be located on the same face of the molecule. These protons, as a result of being correlated together, were assigned as α protons, as Me-15 was a β-substituent at C-1. Correlated with H-10, the H-2 proton had an α-orientation at C-2. Also, H 3 -18 was found to be associated with H-10, suggesting that the C-18 methyl in the γ-lactone moiety had an α-orientation. One of the methylene protons at C-3 (δ H 3.32) exhibited a correlation with H 3 -15 and was assigned as H-3β, while the other was denoted as H-3α (δ H 1.46). The correlations observed between H-3β/H-6, H-6/H-7, and H-7/H-17 reflected the β-orientation of both protons at C-6 and C-7. The cis geometry of the C-13/14 double bond was indicated by a 10.8 Hz coupling constant between H-13 (δ H 5.95) and H-14 (δ H 6.95), and further confirmed by a NOESY correlation between these two olefinic protons. Furthermore, H-3α showed a correlation with H-4, demonstrating the S*-configuration of stereogenic center C-4. The remaining stereogenic carbon, C-8, lacked a proton but there were correlations between H-9/H-17 and H-7/H-17, indicating that C-8 was in an R*-configuration, as evidenced by modeling analysis. Based on the Mar. Drugs 2021, 19, 77 6 of 14 above findings, the configuration of the stereogenic centers of 1 was assigned as (1S*,2S*, 4S*,6S*,7R*,8R*,9S*,10S*,11S*,17R*) (Supplementary Materials, Figures S1-S11).
Briarenol X (2) had the molecular formula C 26 H 37 ClO 9 on the basis of (+)-HRESIMS at m/z 551.20198 (calculated for C 26 H 37 ClO 9 + Na, 551.20183). The IR spectrum of 2 showed bands at 3484, 1777, and 1729 cm −1 , consistent with the presence of hydroxy, γ-lactone, and ester carbonyl groups, respectively. 13  In the HMBC spectrum of 2 (Figure 3), the n-butyrate positioned at C-12 was confirmed from the long-range coupling between H-12 (δ H 4.74) with the carbonyl carbon (δ C 173.3) of the n-butyroxy group. The HMBC correlation also revealed that one acetate was attached to C-2. These data, together with the other 1 H-13 C long-range correlations, unambiguously established the molecular framework of 2. According to the above observations, metabolite 2 seemed to be very similar to solenolide A (5) [15], which was previously isolated from an octocoral Solenopodium sp. By means of 1D and 2D NMR data it was found that the n-hexanoate group at C-12 position in solenolide A (5) was replaced by an n-butyrate group in 2.  the HMBC spectrum. The methine unit at δC 54.9 was more shielded than expected for an oxygenated C atom and correlated with the methine proton at δH 5.52 in the HSQC spectrum. This proton showed a 3 Jcorrelation with H-7, in the 1 H-1 H COSY spectrum, confirming the attachment of a chlorine atom at C-6. An HMBC correlation between H-4 (δH 4.87), an oxymethine proton, and C-8, an oxygenated non-protonated carbon resonating at δC 82.4, suggested that the remaining oxygen atom had to be positioned between C-4 and C-8 to form an ether bridge in 1. the HMBC spectrum. The methine unit at δC 54.9 was more shielded than expected for an oxygenated C atom and correlated with the methine proton at δH 5.52 in the HSQC spectrum. This proton showed a 3 Jcorrelation with H-7, in the 1 H-1 H COSY spectrum, confirming the attachment of a chlorine atom at C-6. An HMBC correlation between H-4 (δH 4.87), an oxymethine proton, and C-8, an oxygenated non-protonated carbon resonating at δC 82.4, suggested that the remaining oxygen atom had to be positioned between C-4 and C-8 to form an ether bridge in 1. The relative stereochemistry of 1 was established using a nuclear Overhauser effect spectroscopy (NOESY) experiment ( Figure 2) and ), and protons with NOESY (   Mar. Drugs 2021, 19, x the HMBC spectrum. The methine unit at δC 54.9 was more s than expected for an oxygenated C atom and correlated with thine proton at δH 5.52 in the HSQC spectrum. This proton show correlation with H-7, in the 1 H-1 H COSY spectrum, confirming tachment of a chlorine atom at C-6. An HMBC correlation betw 4 (δH 4.87), an oxymethine proton, and C-8, an oxygenated non nated carbon resonating at δC 82.4, suggested that the remaini gen atom had to be positioned between C-4 and C-8 to form a bridge in 1.  13 C NMR data and degrees of unsaturation, 3 was established as a pentacyclic diterpenoid. It was found that the 1 H and 13 C NMR data of 3 resembled those of a known briarane, briarenolide M (6) (Figure 1) [16], except that the signals corresponding to the C-12 acetoxy group in 6 were replaced by signals for an n-butyroxy group in 3. Locations of the functional groups were confirmed by other HMBC and COSY correlations (Figure 4).
Briarane 4 (briarenol Z) was isolated as an amorphous powder and had the molecular formula C 26 H 35 ClO 10 on the basis of (+)-HRESIMS (see Materials and Methods section). The IR spectrum of 4 showed bands at 3450, 1777, and 1736 cm -1 , consistent with the presence of hydroxy, γ-lactone, and ester carbonyl groups. It was found that the spectroscopic data of 4 were very similar to those of a known briarane metabolite, briaexcavatolide F (7) [17]. However, a comparison of the 1 H and 13 C NMR chemical shifts of C-6 methine by signals for an n-butyroxy group in 3. Locations of the functional groups were confirmed by other HMBC and COSY correlations ( Figure  4).
The relative configuration of 3 was determined from the NOESY spectrum (Figure 4), which showed NOESY correlations among the corresponding protons similar to those of 6 [16]. The negative optical rotation value of 3 ([α] 24 D −73 (c 0.1, CHCl3)) was similar to that of 6 ([α] 25 D −58 (c 0.7, CHCl3)) [16] in direction and magnitude, suggesting that 3 and 6 had 1S*,10S*-configurations in the ring junction. Thus, briarenol Y was assigned as the structure of 3 and the configurations of the stereogenic carbons were elucidated as (1S*,2R*,3S*,4R*,7S*,8R*,9S*, 10S*,11R*,12R*,13S*,14R*,17R*) (Supplementary Materials, Figures  S23-S33). Briarane 4 (briarenol Z) was isolated as an amorphous powder and had the molecular formula C26H35ClO10 on the basis of (+)-HRESIMS (see Materials and Methods section). The IR spectrum of 4 showed bands at 3450, 1777, and 1736 cm -1 , consistent with the presence of hydroxy, γ-lactone, and ester carbonyl groups. It was found that the spectroscopic data of 4 were very similar to those of a known briarane metabolite, briaexcavatolide F (7) [17]. However, a comparison of the 1 H and 13    The relative stereochemistry of 1 was established using a nuclear Overhauser effect spectroscopy (NOESY) experiment ( Figure 2) and was found to be compatible with that of 1 offered by computer modeling [19]. In naturally occurring briaranes, proton H-10 and Me-15 at C-1 are α-and β-oriented, respectively [3]. H-9, H-10, and H3-20 protons were proven to be located on the same face of the molecule. These protons, as a result of being correlated together, were assigned as α protons, as Me-15 was a β-substituent at C-1. Correlated with H-10, the H-2 proton had an α-orientation at C-2. Also, H3-18 was found to be associated with H-10, suggesting that the C-18 methyl in the γ-lactone ), HMBC ( gen atom had to be positioned between C-4 and C-8 to form an ether bridge in 1. The relative stereochemistry of 1 was established using a nuclear Overhauser effect spectroscopy (NOESY) experiment ( Figure 2) and was found to be compatible with that of 1 offered by computer modeling [19]. In naturally occurring briaranes, proton H-10 and Me-15 at C-1 are α-and β-oriented, respectively [3]. H-9, H-10, and H3-20 protons were proven to be located on the same face of the molecule. These protons, as a result of being correlated together, were assigned as α protons, as Me-15 was a β-substituent at C-1. Correlated with H-10, the H-2 proton had an α-orientation at C-2. Also, H3-18 was found to be associated with H-10, suggesting that the C-18 methyl in the γ-lactone ), and protons with NOESY ( bridge in 1. The relative stereochemistry of 1 was established using a Overhauser effect spectroscopy (NOESY) experiment (Figure was found to be compatible with that of 1 offered by computer ing [19]. In naturally occurring briaranes, proton H-10 and Me-1 are α-and β-oriented, respectively [3]. H-9, H-10, and H3-20 were proven to be located on the same face of the molecule. Th tons, as a result of being correlated together, were assigned a tons, as Me-15 was a β-substituent at C-1. Correlated with H-10 2 proton had an α-orientation at C-2. Also, H3-18 was found to ciated with H-10, suggesting that the C-18 methyl in the γ  Known briaranes 5-8 were found to be identical with briarenolide solenolide A [15], briarenolide M [16], briaexcavatolide F [17], and brianolide [18], respectively by comparison of the spectroscopic data with those reported previously. the HMBC spectrum. The methine unit at δC 54.9 was more shielded than expected for an oxygenated C atom and correlated with the methine proton at δH 5.52 in the HSQC spectrum. This proton showed a 3 Jcorrelation with H-7, in the 1 H-1 H COSY spectrum, confirming the attachment of a chlorine atom at C-6. An HMBC correlation between H-4 (δH 4.87), an oxymethine proton, and C-8, an oxygenated non-protonated carbon resonating at δC 82.4, suggested that the remaining oxygen atom had to be positioned between C-4 and C-8 to form an ether bridge in 1. the HMBC spectrum. The methine unit at δC 54.9 was more shielded than expected for an oxygenated C atom and correlated with the methine proton at δH 5.52 in the HSQC spectrum. This proton showed a 3 Jcorrelation with H-7, in the 1 H-1 H COSY spectrum, confirming the attachment of a chlorine atom at C-6. An HMBC correlation between H-4 (δH 4.87), an oxymethine proton, and C-8, an oxygenated non-protonated carbon resonating at δC 82.4, suggested that the remaining oxygen atom had to be positioned between C-4 and C-8 to form an ether bridge in 1. The relative stereochemistry of 1 was established using a nuclear Overhauser effect spectroscopy (NOESY) experiment ( Figure 2) and ), and protons with NOESY (   Mar. Drugs 2021, 19, x the HMBC spectrum. The methine unit at δC 54.9 was more s than expected for an oxygenated C atom and correlated with thine proton at δH 5.52 in the HSQC spectrum. This proton show correlation with H-7, in the 1 H-1 H COSY spectrum, confirming tachment of a chlorine atom at C-6. An HMBC correlation betw 4 (δH 4.87), an oxymethine proton, and C-8, an oxygenated non nated carbon resonating at δC 82.4, suggested that the remaini gen atom had to be positioned between C-4 and C-8 to form a bridge in 1. Known briaranes 5-8 were found to be identical with briarenolide solenolide A [15], briarenolide M [16], briaexcavatolide F [17], and brianolide [18], respectively by comparison of the spectroscopic data with those reported previously.

Bioactivity of Isolated Briaranes
It is well documented that the microbial lipopolysaccharide (LPS) can activate toll-like receptor-4 (TLR-4), located in the mammal cell membrane surface, which triggers inflammatory responses through the activation of intracellular signal transduction and the upregulation of pro-inflammatory proteins inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2) [20]. It is well known that inhibition of the expression of pro-inflammatory proteins iNOS and COX-2 in LPS-stimulated macrophage cells can be used as for in vitro screening of antiinflammatory compounds [21][22][23]. The massive production of inflammatory mediators, nitric oxide (NO) and prostaglandin E2 (PGE2) via pro-inflammatory proteins iNOS and COX-2, respectively, plays an important pathophysiological role in inflammation. There are two COX isozymes, COX-1 (cyclooxygenase-1) and COX-2, catalyzing the prostaglandin synthesis. COX-1 is constitutively expressed in normal physiological conditions. Unlike COX-1, COX-2 is an inducible enzyme that increases following injury or inflammation [24,25]. COX-2 plays a more vital role in pathology than COX-1 under inflammatory processions.
The effects of briaranes 1-7 on the release of iNOS and COX-2 from LPS-stimulated RAW 264.7 macrophage cells were assessed (Table 3). Briaranes 2 and 5 at 10 μM enhanced the release of iNOS (142.03 and 134.11%, respectively) and COX-2 (159. 21 and 196.03%, respectively) as compared to results for the cells stimulated with LPS only. It is interesting to note that these findings seem to be contrary to results claimed to show that most briarane-type natural products from octocorals are anti-inflammatory [26]. Structure-activity relationships among these marine diterpenoids will be evaluated if enough materials are obtained.

Bioactivity of Isolated Briaranes
It is well documented that the microbial lipopolysaccharide (LPS) can activate tolllike receptor-4 (TLR-4), located in the mammal cell membrane surface, which triggers inflammatory responses through the activation of intracellular signal transduction and the upregulation of pro-inflammatory proteins inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2) [20]. It is well known that inhibition of the expression of pro-inflammatory proteins iNOS and COX-2 in LPS-stimulated macrophage cells can be used as for in vitro screening of anti-inflammatory compounds [21][22][23]. The massive production of inflammatory mediators, nitric oxide (NO) and prostaglandin E2 (PGE2) via pro-inflammatory proteins iNOS and COX-2, respectively, plays an important pathophysiological role in inflammation. There are two COX isozymes, COX-1 (cyclooxygenase-1) and COX-2, catalyzing the prostaglandin synthesis. COX-1 is constitutively expressed in normal physiological conditions. Unlike COX-1, COX-2 is an inducible enzyme that increases following injury or inflammation [24,25]. COX-2 plays a more vital role in pathology than COX-1 under inflammatory processions.
The effects of briaranes 1-7 on the release of iNOS and COX-2 from LPS-stimulated RAW 264.7 macrophage cells were assessed (Table 3). Briaranes 2 and 5 at 10 µM enhanced the release of iNOS (142.03 and 134.11%, respectively) and COX-2 (159. 21 and 196.03%, respectively) as compared to results for the cells stimulated with LPS only. It is interesting to note that these findings seem to be contrary to results claimed to show that most briarane-type natural products from octocorals are anti-inflammatory [26]. Structureactivity relationships among these marine diterpenoids will be evaluated if enough materials are obtained. 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. The β-actin of Western blotting was used for loading/internal control.

General Experimental Procedures
A digital polarimeter (model P-1010; JASCO Corp., Tokyo, Japan) was used to determine optical rotations of the samples. IR spectra were collected using a spectrophotometer (model Nicolet iS5 FT-IR; Thermo Fisher Scientific, Waltham, MA, USA). 1 H and 13 C NMR spectra were recorded on ECZ-400 or ECZ-600 spectrometers (Jeol Ltd., Tokyo, Japan) for solutions in CDCl 3 (with residual CHCl 3 (δ H 7.26 ppm) and CDCl 3 (δ C 77.0 ppm) as internal standards). For coupling constants (J), the results were given in frequency units (Hz). For positive mode ESIMS and HRESIMS, the results were obtained using a SolariX FTMS mass spectrometer (7 Tesla; Bruker, Bremen, Germany). The extracted samples were separated by column chromatography with silica gel (between 230 and 400 meshes; Merck). Thin-layer chromatography plates with silica gel coated with fluorescent indicator F 254 were employed. For visualization, the plates were charred with 10% (v/v) aqueous sulfuric acid solution, then heated at 105 • C until spots were seen. For normal-phase HPLC separation, a system containing a pump (Hitachi model L-7110; Tokyo, Japan) and an injection interface (No. 7725; Rheodyne) was employed, equipped with a semi-preparative column with dimensions of 250 × 20 mm and a 5-µm particle size (Sigma). For reverse-phase HPLC separation, a system composed of a pump (Hitachi model L-2130) and a diode-array detector (LaChrom L-2455, Hitachi) was used, equipped with a column with dimensions of 2.1 × 25 cm and a 5-µm particle size (Phenomenex).

Animal Material
Specimens of B. stechei used for this study were collected from an 80-ton culturing tank equipped with a flow-through seawater system located in the National Museum of Marine Biology and Aquarium (NMMBA) in April 2016. Identification of the species of this organism was performed by comparison, as described in previous studies [14]. Living reference specimens are maintained in the authors' marine organism culturing tanks and a voucher specimen was deposited with the NMMBA (voucher no.: NMMBA-TW-GC-2016-031), Taiwan.

Molecular Mechanics Calculations
The MM2 force field [19] in CHEM3D PRO software from CambridgeSoft Corporation (version 15.0, Cambridge, MA, USA) was used to calculate the molecular models.

In Vitro Inflammatory Assay
The inflammatory assay was employed to evaluate the activities of briaranes 1-7 related to the release of iNOS and COX-2 from macrophage cells, as reported in the literature [32].

Conclusions
Eight chlorinated briarane diterpenoids, including four new briaranes-briarenols W-Z (1-4)-as well as four known analogues-solenolide A (5), briarenolide M (6), briaexcavatolide F (7), and brianolide (8)-were identified from a cultured octocoral B. stechei, originally flourishing in Taiwanese waters where the Kuroshio current and South China Sea surface current converge to provide high biodiversity. The structures of new briaranes 1-4 were elucidated on the basis of spectroscopic analysis and the absolute configuration of 8 (brianolide) was determined by a single-crystal x-ray diffraction analysis. As briaranes 1-7 were isolated along with brianolide (8) from the same target organism, B. stechei, it is reasonable on biogenetic grounds to assume that 1-7 have the same absolute configuration as that of 8, while the protons H-10 and Me-15 at C-1 in briaranes 1-8 are αand β-oriented, respectively, and these compounds have 1S,10S-configurations in the ring junction. Briaranes 2 (briarenol X) and 5 (solenolide A) displayed enhancing effects on the production of iNOS and COX-2 at a concentration of 10 µM.