In previous studies, soft corals have emerged as one of the most prolific sources of novel secondary metabolites [1]. Some of these have exhibited various biological activities, such as cytotoxic [2,3], anti-inflammatory [4,5,6], antiviral [6] and antifouling [7] activities. However, the bioactive secondary metabolites contents of wild soft coral are small, and extraction of large amounts of bioactive metabolites from wild coral is not feasible. Therefore, scientists are searching for new ways to obtain large sources of bioactive metabolites. In recent years, the farming techniques of soft corals have improved considerably. Therefore, researchers have been able to obtain greater amounts of soft corals, and thus larger amounts of bioactive metabolites, and have been engaged in examining the various pharmacological activities. During the course of our group’s (National Museum of Marine Biology & Aquarium, Taiwan) search for bioactive metabolites from cultured soft corals, several diterpenoids have been isolated from the soft corals Klyxum simplex [8,9,10,11], Sinularia flexibilis [12], Sarcophyton trocheliophorum [13], and Lobophytum crassum [14]. In continuation of our search for biologically active secondary metabolites from the cultured soft coral Sinularia leptoclados (Figure 1), we have isolated one new spatane diterpenoid, leptoclalin A (1), along with two known norcembranoid diterpenes, 5-episinuleptolide (2) and sinuleptolide (3) [15] (Chart 1). The structure of 1 was established by detailed spectroscopic analysis, including extensive examination of 2D NMR (¹H–¹H COSY, HMQC and HMBC) correlations. The cytotoxicity of compounds 1–3 against four cancer cells, DLD-1 (human colon adenocarcinoma), HCT 116 (human colorectal carcinoma), T-47D (hormone-dependent breast cancer) and K-562 (human chronic myelogenous leukemia) was studied.

The EtOAc extract of the freeze-dried specimen was fractionated by silica gel column chromatography and the eluted fractions were further separated utilizing normal phase HPLC to yield metabolites 1–3. The new compound was given the trivial name leptoclalin A (1). The known compounds were identified as 5-episinuleptolide (2) and sinuleptolide (3).

Leptoclalin A (1) was isolated as a colorless oil. Its molecular formula was determined to be C₂₀H₃₂O on the basis of HR-ESI-MS (m/z 311.2353 [M + Na]⁺), implying five degrees of unsaturation. The IR spectrum of 1 revealed the presence of a hydroxy functionality (ν_max = 3365 cm⁻¹). The ¹³C-NMR and DEPT spectroscopic data (Table 1) of 1 indicated the presence of four methyls, six methylenes, seven methines, and three quaternary carbons, implying, from the required degrees of unsaturation, a tricyclic diterpene framework. The NMR signal at δ_C 70.7 (C) showed the presence of a hydroxy group. A 1,1-disubstituted and a 1,2-disubstituted double bond were also identified from NMR signals appearing at δ_C 157.5 (C), 103.7 (CH₂), and δ_H 4.74 (1H, s) and 4.70 (1H, s), and at δ_C 139.2 (CH), 125.8 (CH), and δ_H 5.58 (2H, m), respectively. Moreover, the ¹H NMR spectroscopic data of 1 revealed evidence of four methyl groups at δ 0.83 (3H, d, J = 6.5 Hz), 1.00 (3H, s) and 1.31 (6H, s).

The planar structure and all of the ¹H and ¹³C chemical shifts of 1 were elucidated by 2D NMR experiments, in particular ¹H–¹H COSY and HMBC experiments (Figure 2). From the ¹H–¹H COSY spectrum (CDCl₃), it was possible to establish the proton sequences from H₂-2 to H-10 through H₂-3; H₂-5 to H₂-6; H-8 to H-10 through H-9; and H-13 to H₃-14. The overlapping of proton singals of H-7 and H-8 at δ_H 1.62 ppm, measured in CDCl₃, was clearly resolved by measuring the ¹H NMR spectrum in pyridine-d₅ (see Table 1) into two separate protons at δ_H1.62 and 1.67, respectively. ¹H–¹H COSY correlations (pyridine-d₅) were observed between H-6 to H-13 through H-7. Therefore, the ¹H–¹H COSY experiment allowed the building of the two partial structures of the consecutive proton spin systems indicated in bold in Figure 2. These data, together with the HMBC correlations (Figure 2) from H₂-2 to C-1, H₂-3 to C-1 and C-9, H₂-5 to C-4 and C-10, H₂-6 to C-4 and C-8, H-9 to C-7 and H-10 to C-8 established the connectivity within the 5-membered, 4-membered and 5-membered rings. Moreover, the correlations of H₂-11 to C-2 and C-9 indicated the attachment of an sp² methylene at C-1, and the methyl was attached at C-4 on the basis of the correlations of H₃-12 to C-4, C-5, C-8 and C-10. The planar structure of the side chain was elucidated mainly from the key HMBC correlations from H₃-14 to C-7, C-13, and C-15, H₂-15 to C-16 and C-17, and both methyls H₃-19 and H₃-20 to C-18 and C-17, and thus the connectivity from C-13 to C-20 was fully established. The above findings suggested that 1 has a tricyclic carbon skeleton.

A computer-modeled 3D structure of 1 (Figure 3) was generated using the molecular modeling program Chem3D Ultra version 9.0 and MM2 force-field calculations for energy minimization. The relative structure of 1, assigned by the analysis of NOE correlations, was compatible with that of 1 offered by computer modeling, in which the close contacts of the atoms in space were consistent with the NOE correlations. By the NOESY spectrum (pyridine-d₅), it was found that H-9 (δ 2.46) showed NOE interactions with H-7 (δ 1.62) and H-10 (δ 2.24); therefore, assuming the β-orientation of H-9, H-7 and H-10 should also be positioned on the β face. One of the methylene protons at C-3 (δ 1.52) exhibited NOE correlations with H-10 and was characterized as H-3β, while the other (δ 1.74) was designated H-3α. NOE correlations observed between H-3α and H₃-12 (δ 0.94) and H₃-12 and H-8 (δ 1.67) reflected the α-orientations of H₃-12 and H-8. Moreover, NOE correlations were observed between H-8 with H-13 but not with H₃-14, and between H-7 with H₃-14, indicating that H₃-14 has a β-orientation. Furthermore, the configuration of the double bond at C-16/C-17 was determined by comparison of the NMR data of 1 in CDCl₃ with those of two related synthetic compounds, (23E)-cycloart-23-ene-3β,25-diol (4) and (23E)-cycloart-23-ene-3β,25-diol (5) (Figure 4), also measured in CDCl₃ (Table 2) [16]. Comparison of the NMR data of 1 and 4 confirmed that both compounds have the same partial structure from C-15 to C-20 of 1 and from C-22 to C-27 of 4. Furthermore, the geometry at C-16 double bond is deducible with the ¹³C chemical shift of C-15 methylene (the downfield chemical shift at 38.2 ppm clearly indicates the E geometry). Thus, it was suggested that the geometry of 1 at C-16/C-17 is E. On the basis of the above analysis, the structure of 1 was established.

Finally, we used the 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl tetrazolium bromide (MTT) assay to examine the cytotoxic activities of compounds 1–3 against four cancer cells, DLD-1 (human colon adenocarcinoma), HCT 116 (human colorectal carcinoma), T-47D (hormone-dependent breast cancer), and K-562 (human chronic myelogenous leukemia). Cells were treated with different concentrations of 1–3 for 72 h. The results show that compound 1 was found to show weak cytotoxicity towards the growth of T-47D and K-562 tumor cells (the IC₅₀ values were 15.4 and 12.8 μg/mL for T-47D and K-562, respectively) (Table 3). The other tested compounds were not cytotoxic (IC₅₀ > 20 μg/mL) towards the above four cancer cell lines (Table 3).

Spatane diterpenoids were first reported in 1980 as having been obtained from the brown alga Spatoglossum schmittii from the Galapagos Islands [19]. Brown algae are now known to be a rich source of novel spatane diterpenoids [20,21,22,23,24,25,26,27]. In contrast, they are generally the minor components of marine soft coral [28]. Our investigation of the chemical constituents of the soft coral Sinularia leptoclados led to the obtainment of one new spatane diterpenoid (1). Moreover, to the best of our knowledge, metabolite 1has a spatane skeleton rarely found in soft corals.