Open Access Article
Gia Hung Phan‡
ab,
Hao-Chun Hu‡c,
Fang-Rong Chang
c,
Zhi-Hong Wende,
Jih-Jung Chenf,
Hsu-Ming Chung
g,
Yu-Chi Tsai*h and
Ping-Jyun Sung
*bcdij
aGraduate Institute of Marine Biology, National Dong Hwa University, Pingtung 944401, Taiwan
bNational Museum of Marine Biology and Aquarium, Pingtung 944401, Taiwan. E-mail: pjsung@nmmba.gov.tw
cGraduate Institute of Natural Products, Kaohsiung Medical University, Kaohsiung 807378, Taiwan
dDepartment of Marine Biotechnology and Resources, National Sun Yat-sen University, Kaohsiung 804201, Taiwan
eInstitute of BioPharmaceutical Sciences, National Sun Yat-sen University, Kaohsiung 804201, Taiwan
fDepartment of Pharmacy, School of Pharmaceutical Sciences, National Yang Ming Chiao Tung University, Taipei 112304, Taiwan
gDepartment of Applied Chemistry, National Pingtung University, Pingtung 900391, Taiwan
hSchool of Chinese Medicine, College of Chinese Medicine, China Medical University, Taichung 404328, Taiwan. E-mail: yuchi0713@gmail.com
iChinese Medicine Research and Development Center, China Medical University Hospital, Taichung 404394, Taiwan
jPhD Program in Pharmaceutical Biotechnology, Fu Jen Catholic University, New Taipei City 242062, Taiwan
First published on 30th September 2022
Three norsesquiterpenoids, pathyspirolactones A (1) and B (2), and napalilactone (3), featuring a γ-spirolactone moiety, were isolated from the cultured octocoral Paralemnalia thyrsoides. The structures of 1–3 were determined by analyzing spectroscopic data, DP4+ computation, specific optical rotation, and X-ray diffraction. In addition, we explored the absolute configurations of pathyspirolactone A (1) and its conformation of the cyclohexane ring to resolve the stereochemical confusion of those of norsesquiterpenoid compounds. Furthermore, we proved that pathyspirolactone B (2) was the first bromine-containing norsesquiterpenoid reported from octocorals.
| Position | 1 | 2 | 3 | epi-pathylactone | ||||
|---|---|---|---|---|---|---|---|---|
| δHa (J in Hz) | δCb, Multc | δHa (J in Hz) | δCb, Multc | δHa (J in Hz) | δCb, Multc | δHh (J in Hz) | δCh | |
| a Spectra recorded at 400 MHz in CDCl3.b Spectra recorded at 100 MHz in CDCl3.c Multiplicity deduced by DEPT and HSQC spectra and indicated by usual symbols.d Signals overlapped.e Signals overlapped.f Signals overlapped.g Signals overlapped.h Data were reported by Coelho and Diaz.16 These data were recorded at 500 MHz for 1H and 125 MHz for 13C in CDCl3.i The coupling pattern and coupling constants were assigned as “two similar coupling constants (J = 4.2, 2.9 Hz)” in the content text of the ref. 16. | ||||||||
| 1β(axial) | 3.70 dd (11.6, 4.8) | 73.4, CH | ||||||
| 1β(equiv.) | 4.27 mi | 65.0 | ||||||
| 1α(axial) | 4.59 dd (12.0, 4.4) | 58.2, CH | 4.36 dd (10.4, 4.4) | 63.8, CH | ||||
| 2α(axial) | 1.63 m | 30.4, CH2 | ||||||
| 2β(equiv.) | 1.81 m | |||||||
| 2α(equiv.)/β(axial) | 2.23 m; 2.08 ddd (14.4, 12.0, 4.0) | 29.8, CH2 | 2.12 m; 1.94 m | 28.9, CH2 | 1.78–2.0 | 27.0 | ||
| 3α(equiv.) | 1.61 m | 28.4, CH2 | ||||||
| 3β (axial) | 1.41 m | |||||||
| 3α(axial)/β(equiv.) | 1.89 m; 1.45 dddd (14.0, 4.0, 4.0, 4.0) | 28.1, CH2 | 1.83 m; 1.50 m | 27.0, CH2 | 1.44–1.79 | 26.0 | ||
| 4 | 2.17 m | 34.3, CH | 2.56 md | 33.2, CH | 2.53 m | 33.2, CH | 2.38 | 33.7 |
| 5 | 46.7, C | 46.4,f C | 46.2, C | 45.3 | ||||
| 6 | 91.8, C | 91.1, C | 91.5, C | 92.0 | ||||
| 7 | 177.9, C | 175.8, C | 175.9, C | 177.0 | ||||
| 8 | 2.74 ddd (17.2, 12.4, 8.0); 2.47 ddd (17.2, 11.2, 4.2) | 29.6, CH2 | 2.74 me; 2.53 md | 29.7, CH2 | 2.68 mg; 2.46 m | 29.4, CH2 | 2.48–2.67 | 29.0 |
| 9 | 2.37 m; 2.19 m | 26.3, CH2 | 2.48 m; 2.41 m | 26.2, CH2 | 2.39 d (10.0); 2.37 dd (10.0, 2.0) | 24.8, CH2 | 2.30 | 24.0 |
| 10 | 2.47 s | 48.9, CH2 | 2.89 d (16.0); 2.76 d (16.0e) | 46.4f, CH2 | 2.84 d (16.0); 2.70 d (16.0g) | 46.8, CH2 | 2.80–2.73 (two doublet, J = 15.1) | 47.0 |
| 11 | 208.4, C | 207.9, C | 208.0, C | 207.0 | ||||
| 12 | 2.20 s | 32.3, CH3 | 2.16 s | 33.0, CH3 | 2.16 s | 32.9, CH3 | 2.09 s | 32.2 |
| 13 | 0.99 d (7.2) | 16.6, CH3 | 1.03 d (7.6) | 15.7, CH3 | 1.01 d (7.6) | 15.7, CH3 | 0.96 d (7.7) | 15.9 |
| 14 | 0.91 s | 15.0, CH3 | 1.23 s | 18.6, CH3 | 1.18 s | 18.0, CH3 | 1.12, s | 17.2 |
From the 1H–1H COSY spectrum (Fig. 2), the data suggested that it was possible to differentiate between the separate spin systems of H-1/H2-2/H2-3/H-4/H3-13 and H2-8/H2-9. Together with the key HMBC correlations of 1, such as H-4, H2-9, H2-10, H3-13, H3-14/C-5; H2-2, H-8, H-9, H2-10, H3-14/C-6; H2-8, H2-9/C-7; and H2-10, H3-12/C-11, these data confirmed the main carbon skeleton of 1. The HMBC correlations from H3-12/C-10, C-11; H3-13/C-3, C-4, C-5; and H3-14/C-4, C-5, C-6, C-10 indicated that CH3-12, CH3-13, and CH3-14 were placed at C-11, C-4, and C-5, respectively (Fig. 2).
The relative stereochemistry of 1 was deduced mainly from the NOE interactions in the NOESY experiment, Chem3D Model, and vicinal 1H–1H coupling constant analysis. In accordance with convention, when analyzing the stereochemistry of 1, the tertiary C-14 methyl group at C-5 was assigned to the β-face, anchoring stereochemical analysis. In the NOESY experiment (Fig. 3), the correlations of H3-14 with H-1 and H2-9 showed that these protons were positioned on the same face of the molecule, and therefore they were assigned as β-protons. One of the methylene protons at C-3 (δH 1.41) exhibited a correlation with H3-14, leading to its assignment as H-3β, while the other was denoted as H-3α (δH 1.61). The correlation between H-3β and H3-13 reflected the β-orientation of the CH3-13 group at C-4. The configuration at the cyclohexane ring in 1 is worthy of comment. H-1 was found to exhibit correlations with H3-14, as well as coupling between H-1 and H2-2 (J = 11.6, 4.8 Hz), indicating that both H-1 and C-14 methyl at C-5 should be oriented at β-axial positions. Therefore, based on the above findings, the configurations of the stereogenic carbons of 1 were determined as (1S*, 4S*, 5R*, 6R*).
![]() | ||
| Fig. 3 Stereo-view of 1 (generated by computer modeling) and calculated distances (Å) between selected protons with key NOESY correlations. | ||
It is very interesting to note that the structure of 1 as we presented herein had been reported and named epi-pathylactone A, a synthetic product by Coelho and Diaz in 2002.16 Although these two compounds possessed the same relative configurations (1S*, 4S*, 5R*, 6S*) (Fig. 3 and 4), the different chemical shifts and coupling constants of H-1 in 1 (δH 3.70, 1H, dd, J = 11.6, 4.8 Hz) and epi-pathylactone A (δH 4.27, 1H, J = 4.2, 2.9 Hz)16 demonstrated that H-1 had significantly different dihedral angles with H-2α and H-2β in both compounds, respectively. Furthermore, the 1H-NMR spectra of compound 1 at temperatures of 0, 25, and 50 °C were also measured to discover the 1H chemical shift changes in those spectra. However, the critical proton NMR signal on C-1 (δH 3.70) in the different temperature experiments was quite similar (Fig. S27–S29†). Therefore, compound 1 was proposed as a new conformational isomer and did not undergo conformational interconversion after heat.
Based on the Newman projection analysis, H-1 in 1 expressed anti with H-2α and gauch with H-2β; H-1 in epi-pathylactone A showed gauch with the methylene protons on C-2 (Fig. 5), indicating H-1 should be in a β-axial position in 1 and β-equatorial in epi-pathylactone A. In addition, the conformational variation in the cyclohexane ring led to the structural difference in those compounds, resulting in the absolute configuration of 1 being hard to discuss by the reference comparison.
For solving the absolute stereochemistry issue of compound 1, the DP4+ analysis was selective for double-checking the configuration of position C-1, which was the most confusing location for the type of compounds. The structures of 1-1S*, 4S*, 5R*, 6R* (diastereomer 1) and 1-1R*, 4S*, 5R*, 6R* (diastereomer 2) were computed GIAO-NMR data by Gaussian 09, and the calculated results were analysis by DP4+ (Fig. S11†). The DP4+ analysis results of 1-1S*, 4S*, 5R*, 6R* displayed the match ratio 99.94%, 99.98%, and 100.00% in sDP4+ (all data), uDP4+ (all data), and DP4+ (all data), respectively (Table S1†). Furthermore, the possible configurations of 1-1S, 4S, 5R, 6R and 1-1R, 4R, 5S, 6S were input into spartan’16 and Gaussian 09 software for calculating conformational search, structure optimization, and specific optical rotation (SOR) value. As a result, the calculated SOR value of 1-1S, 4S, 5R, 6R (66) was consistent with the experiment result of 1 (positive) (Table S2†). This is the first to clarify the absolute configuration of these norsesquiterpenoid -type compounds by in silico method and expounded on the differences in 1D-NMR data of the cyclohexane ring conformational variants.
Pathyspirolactone B (2) was isolated as an amorphous powder. ESIMS showed a pair of ion peaks at m/z 339/341 ([M + Na]+/[M + 2 + Na]+) (1
:
1), with the relative intensity indicative of a bromide substituent. NMR data coupled with the [M + Na]+ peak in HRESIMS at m/z 339.05646 suggested a molecular formula C14H21BrO3 (calcd for C14H21BrO3 + Na, 339.05663), which indicated four degrees of unsaturation. IR absorptions at νmax 1777 and 1712 cm−1 suggested the presence of γ-lactone and ketonic groups. From the 1H and 13C NMR spectroscopic data, in combination with the DEPT spectrum (Table 1), two carbonyl resonances at δC 207.9 and 175.7 confirmed the presence of ketonic and γ-lactone moieties. So, from the NMR data, two degrees of unsaturation were accounted for, and 2 must be a bicyclic compound.
Due to electronegativity effects, the methine unit at δC 58.2 was more shielded than expected for oxygenated carbon and was correlated to the methine proton at δH 4.59 in the HSQC spectrum, proving the attachment of a bromide atom at C-1. These data, together with the 1H–1H COSY and HMBC correlations, established the molecular framework of 2 (Fig. 6). The NMR data of 2 were found to be close to those of a known halogenated norsesquiterpenoid, napalilactone (3),17 which was also isolated in this study (Fig. 1), except for replacing a chlorine atom at C-1 in 3 (δH 4.36, 1H, dd, J = 10.4, 4.4 Hz/δC 63.8, CH-1) (Fig. S25 and S26†) with a bromine atom in 2 (δH 4.59, 1H, dd, J = 12.0, 4.4 Hz 4.0 Hz/δC 58.2, CH-1). To the best of our knowledge, compound 2 is the first bromine-containing norsesquiterpenoid reported from octocorals.
The relative configuration of 2 was elucidated from the interactions observed in a NOESY experiment. Furthermore, it was found to be compatible with that of 2 offered by computer modeling (Fig. 7) and that obtained from vicinal proton coupling constant analysis. In the NOESY spectrum of 2, H-1 showed a correlation with one proton of CH2-10 (δH 2.76) and a large coupling constant with H-2β (J = 12.0 Hz), indicating an α-axial orientation of H-1. The methyl proton H3-14 exhibited correlations with H3-13 and H2-9 but without H2-10, revealing the β-orientations of Me-13, Me-14, and C-9 methylene at C-4, C-5, and C-7, respectively. Moreover, 2-1R*, 4S*, 5R*, 6R* (diastereomer 1) and 2-1S*, 4S*, 5R*, 6R* (diastereomer 2) were further submitted into Gaussian 09 for computed GIAO-NMR data for DP4+ analysis. The analysis results of 2-1R*, 4S*, 5R*, 6R* exhibited a 100% matched ratio with the experimental data of 2 in sDP4+ (all data), uDP4+ (all data), and DP4+ (all data) (Table S1†). Consequently, the relative configuration of 2 was elucidated to be 1R*, 4S*, 5R*, 6R*.
![]() | ||
| Fig. 7 Stereo-view of 2 (generated by computer modeling) and calculated distances (Å) between selected protons with key NOESY correlations. | ||
Furthermore, the SOR was used for determining the absolute configuration of 2. The calculated SOR of 2-1R, 4S, 5R, 6R and 2-1S, 4R, 5S, 6S exhibited a positive (12) and a negative (−12) value, respectively (Table S1†). The experiment SOR data of 2 (positive) was matched with the 2-1R, 4S, 5R, 6R. However, as the similar issue of 1, the absolute configuration of C-1 cast doubt on 2.
Moreover, the absolute configuration of napalilactone (3) was fully established by a single-crystal X-ray diffraction analysis with the Flack parameter x = −0.04(4).18,19 The computer-generated ORTEP diagram (Fig. 8) showed the absolute configuration of stereogenic centers of 3 were 1R, 4S, 5R, 6R. Based on the principles of biogenetics and the above DP4+ analysis, pathyspirolactone B (2) can be verified as the same absolute configuration as 3.
Based on the past reports, Paralemnalia spp. showed a promising anti-inflammatory effect and cytotoxic activity.7,8,10,14 Therefore, in in vitro anti-inflammatory activity tests, upregulation of pro-inflammatory inducible nitric oxide synthase (iNOS) and cyclooxygenase-2 (COX-2) protein expression in LPS-stimulated RAW 264.7 macrophage cells were evaluated using immunoblot analysis. At a concentration of 10 μM, norsesquiterpenoids 1 and 2 were found to be inactive to reduce the level of iNOS and COX-2 in relation to control cells stimulated with LPS only. Using trypan blue staining to measure the cytotoxic effects of the compounds, it was observed that 1 and 2 did not induce cytotoxicity in RAW 264.7 macrophage cells.
:
1) to give an extract II (29.5 g). Then extracts I and II were partitioned with EtOAc and H2O to obtain the EtOAc-soluble layers A (8.9 g) and B (1.8 g), respectively. The EtOAc layers A and B were then combined, placed in a silica column, and eluted by hexanes/EtOAc (pure hexanes to pure EtOAc, stepwise) to yield ten fractions A–J. Fraction G (121.3 mg) was further separated by normal-phase HPLC on Galaksil® EF-SiO2 column with a mixture of n-hexane and acetone (65
:
35) at a rate of 5 mL min−1 to give four subfractions (G1–4). G2 (63.2 mg) was subjected to the normal-phase HPLC system with an isocratic solvent system of n-hexane and ethyl acetate (80
:
20, 5 mL min−1) to yield 12 fractions (G2A–L). G2L (48.6 mg) was then separated by the normal-phase HPLC system using a mixture of n-hexane and acetone (80
:
20, 5 mL min−1) to obtain 15 subfractions (G2L1–15). G2L10 (10.9 mg) was further purified by reverse-phase HPLC on Ascentis® C18 column with an isocratic solvent system of MeOH and H2O (55
:
45, 5 mL min−1) to obtain compound 1 (0.2 mg). Fraction D (710.1 mg) was subjected to the normal-phase HPLC system with a mixture of n-hexane and ethyl acetate (85
:
15, 5 mL min−1) to give nine subfractions (D1–9). D7 (18.2 mg) was further purified by the reverse-phase HPLC using a mixture solvent system of MeOH
:
H2O (50
:
50, 5 mL min−1) to obtain compounds 2 (0.3 mg) and 3 (11.6 mg).
483 reflections were collected, of which 4100 were independent. The structure was solved by direct methods and refined by a full-matrix least-squares on the F2 procedure.23 The refined structural model converged to a final R1 = 0.0460; wR2 = 0.1060 for 3494 observed reflections [I > 2σ(I)] and 172 variable parameters; and the absolute configuration was determined from the Flack parameter x = −0.04(4).18,19 Crystallographic data for the structure of napalilactone (3) were submitted to the Cambridge Crystallographic Data Center (CCDC) with supplementary publication number CCDC 2190441 (data can be obtained from the CCDC website at https://www.ccdc.cam.ac.uk/conts/retrieving.html).
Footnotes |
| † Electronic supplementary information (ESI) available: HRESI-MS, 1D and 2D-NMR spectra of 1 and 2; ESI-MS and 1D-NMR spectra of 3; experimental and calculated SOR values of 1 and 2; X-ray crystallo-graphic data of 3; DP4+ analysis of 1 and 2. CCDC 2190441. For ESI and crystallographic data in CIF or other electronic format see https://doi.org/10.1039/d2ra05015c |
| ‡ These authors have contributed equally to this work. |
| This journal is © The Royal Society of Chemistry 2022 |