New-skeleton terpenoids are highly valued for their potential to broaden the chemical landscape, unlock novel opportunities within biological and pharmaceutical fields, and furnish synthetic chemists with fresh structural templates [
1–
6], thus having garnered widespread interest and enthusiasm. Sesterterpenoids are amongst the rarest class of terpenoids with approximately 1400 members [
3,
7,
8], of which
ca. 200 ones were isolated from terrestrial plants [
9–
12]. Prenyllabdane sesterterpenoids, existing in some plant species of the genus
Salvia [
3], are derived from a similar cyclization as labdane diterpenoids. To date, only six prenyllabdane norsesterterpenoids have been documented [
13–
15], yet no biological evaluation has been reported for these compounds. Clerodane diterpenoids are a large group of secondary metabolites found in a wide range of organisms, such as plants, microorganisms, and marine sponges [
16]. Despite the prevalence in nature, variations (rearrangement, elimination, or cyclization) in the skeletal structures of clerodanes, particularly those involving the decalin core [
17–
22], remain uncommon and underreported. These clerodane variants often exhibit diverse bioactivities, such as inflammasome activation [
17] and inhibition [
18], anti-Alzheimer's disease [
20], and modulatory activity of multidrug resistance in MCF-7 cancer cells [
22].
Croton laui Merrill & F. P. Metcalf (Euphorbiaceae), an indigenous perennial shrub in Hainan Province of China, has been utilized as a folk remedy for treating headaches, stomachaches, and diphtheria [
23]. Previous investigations have demonstrated that it is a prolific source of labdane and neoclerodane diterpenoids [
24–
30]. In connection with our ongoing efforts to identify bioactive natural products from ethnomedicinal plants [
31–
34], two new-skeleton terpenoids, including an aromatic 12,17-cyclo20-nor phenyllabdane sesterterpenoid, crolatinoid A (
1), and a 19(5→4)-abeo-3,5-cycloneoclerodane diterpenoid, crolatinoid B (
2), along with its plausible biosynthetic precursor crolatinoid C (
3) (
Fig. 1), were isolated from the bark of
Croton laui. Furthermore, the multidrug resistant reversal and antiadipogenic activities were evaluated for these terpenoids. Herein, the isolation, structural elucidation, and biological evaluation of these new-skeleton terpenoids are described.
Crolatinoid A (1), colorless gum, was assigned a molecular formula of C24H36O3 incorporating seven degrees of unsaturation (DOUs), as supported from its positive high-resolution electrospray ionization mass spectrometry [(+)-HRESIMS] ion peak at m/z 395.2555 [M + Na]+ (calcd. 395.2557). Its one-dimensional nuclear magnetic resonance (1D NMR) data (Table S1 in Supporting information), with the aid of the 1H–13C heteronuclear single quantum coherence (HSQC) and distortionless enhancement by polarization transfer-135 (DEPT-135) experiments, displayed typical signals for one 1,2,3-trisubstituted phenyl ring (δH 7.13, t, J = 7.6 Hz, 7.22, d, J = 7.6 Hz, and 7.36, d, J = 7.6 Hz; δC 125.5, 126.5, 133.9, 138.1, 140.2, and 141.1), six methyls, one oxymethine (δH 7.13, δC 81.0), one oxygenated tertiary carbon (δC 77.7), and one carbonyl group (δC 206.4). These functionalities occupied five out of the seven DOUs. The remaining two ones required the presence of two additional rings in 1.
The planar structure of
1 was constructed by the two-dimensional (2D) NMR experiments. Specifically,
1H–
13C heteronuclear multiple bond correlation (HMBC) signals (
Fig. 2A) from H
3–22 to C-7, C-8, and C-9, from H
3–23 and H
3–24 to C-3, C-4, and C-5, and from H
3–25 to C-1, C-9, and C-10, in conjunction with
1H–
1H homonuclear correlation spectroscopy (COSY) cross-peaks of H
2–1/H
2–2/H
2–3 and H-5/H
2–6/H-7, delineated a bicyclic decalin framework (A- and B-ring) adorned with four methyls (Me-22, Me-23, Me-24, and Me-25). The 7,8-vicinal diol moiety was confirmed by the chemical shift values of CH-7 and C-8. Additionally, HMBC correlations from H
2–11 to C-13 and C-17 and the COSY cross-peak of H-9/H
2–11, verified the connection of the phenyl ring with B-ring
via the C-9–C-11–C-12 sequence. The presence of an acetyl group at C-17 and a methyl at C-13 was confirmed by the HMBCs from H-16 to C-12 and C-18, H
3–19 to C-17 and C-18, and H
3–21 to C-12, C-13, and C-14. Therefore, compound
1 was established as a new-skeleton aromatic norsesterterpenoid.
The relative configuration of
1 was determined by the nuclear Overhauser effect spectroscopy (NOESY) experiment (
Fig. 2B). In the chair conformation of the decalin core, the observed NOE correlations of H
3–25 with H
3–24 and H
3–22 revealed that they were cofacial and were assigned arbitrarily as
β-oriented [
24]. Consequently, the NOE correlations of H-5 with H-7 and H-9 suggested that H-5, H-7, and H-9 were α-oriented. The absolute configuration of
1 was determined
via quantum chemical electronic circular dichroism (ECD) calculation, by using time-dependent density functional theory (TDDFT) method at the
ωB97XD/6–311G**//B3LYP/6–31G* level [
35–
38]. As shown in
Fig. 3, the experimental ECD curve matched well with the calculated one in the region of 200–400 nm, establishing its absolute configuration as 5
S,7
S, 8
S,9
R,10
S.
Crolatinoid B (
2) was obtained as colorless crystals with a negative optical rotation in acetonitrile (
−9). Its molecular formula was determined to be C
21H
20O
7 with twelve DOUs, as gleaned from the (+)-HRESIMS ion peak at
m/z 385.1279 [M + H]
+ (calcd. 385.1282) and the
13C NMR data (Table S1). The absorption bands at 1749 and 1659 cm
−1 in the infrared radiation (IR) spectrum indicated the presence of carbonyl and double bond functionalities, respectively. The 1D NMR data showcased typical resonances of a
β-substituted furan (
δH 6.41, 7.45, and 7.48;
δC 108.2, 124.2, 140.3, and 144.4) [
25], along with resonances arising from some unsaturated groups, including a terminal double bond (
δH 4.96 and 5.10;
δC 111.8 and 145.4), a keto carbonyl (
δC 206.4), two ester carbonyls (
δC 167.0 and 175.3), and an aldehyde (
δH 10.06;
δC 196.6). These functionalities occupied eight of the twelve DOUs, and the remaining four ones thus indicated the presence of a tetracyclic architecture in
2.
Further analysis of the HMBC spectra (
Fig. 4A) established the planar structure of
2. Four different spin coupling fragments of CH
2–1/CH-10, CH
2–6/CH
2–7, CH
2–11/CH-12, and CH-14/CH-15 were readily identified by interpretation of the
1H–
1H COSY spectrum. HMBC correlations from H
2–1 to C-2 (
δC 206.4), and from H-3 to C-1, C-5, and C-10 delineated a cyclopentanone moiety (A
2-ring), which was fused to the six-membered B-ring, as inferred from HMBCs from H-10 to C-8 and C-9, and from H
2–6 to C-8 and C-10. The HMBCs from H
2–17 to C-7, C-8, and C-9 anchored the location of the Δ
8(17) double bond. Additionally, the presence of a cyclopropane A
1-ring that was fused to A
2-ring was assigned by HMBC correlations from H-3, H
2–6, and from H-10 to C-4. HMBC correlations from H-3 and CH
3O-18 to C-18 indicated that a methoxycarbonyl group was attached to the C-4 quaternary carbon. By comparing the NMR data with those of methyl
exo-6-formylbicyclo[3.1.0]hexane-6-carboxylate [
39], a formyl group was assigned to C-4, which was supported by the NOESY correlation between H-19 and CH
3O-18. Furthermore, HMBC correlations from H
2–11 to C-9 and C-20 (
δC 175.3), and from H-12 to C-20 confirmed the existence of 9-spiro
γ-butyrolactone (C-ring). The attachment of the
β-substituted furan (D-ring) to C-12 was evidenced by HMBC correlations from H-12 to C-14 and C-16. Therefore, compound
2 was established as a new-skeleton diterpenoid featured by a unique 3/5/6-fused tricyclic framework.
The relative configuration of
2 was mainly established by the NOESY experiment (
Fig. 4B). The NOE correlations of H-1
α with H-19 and H-7
α with H-19 indicated that H-1
α, H-7
α, and the 19-formyl group were co-facial and arbitrarily designated as
α-oriented. Consequently, the H-1
β/H-10 and H-10/H-6
β correlations indicated that all these protons were
β-oriented. The NOE correlation between H-3 and H
2–6 suggested a
cis-fusion between A
1- and A
2-ring, and H-3 was thus assigned as
β-oriented. In addition, the NOE correlations of 11
α/H-1
β and 11
α/H-10 suggested a 9
S* configuration. However, the C-12 configuration remained undetermined due to insufficient reliable NOE data. Fortunately, this limitation was overcome by a successful single-crystal X-ray diffraction (XRD) experiment, which not only corroborated the above assignments, but also established its absolute configurations as 3
R,4
S,5
S,9
S,10
S,12
S [
Fig. 5, Flack parameter = 0.09(11)] [
40].
Crolatinoid C (
3), which took the form of colorless crystals, possesses the molecular formula C
21H
20O
7, based on analysis of its (+)-HRESIMS and the
13C NMR data. Its NMR data were highly similar to that of a known neoclerodane diterpenoid, mangelonine H [
41], indicating that they are structural analogs. The only difference between them was the presence of a C-2 keto carbonyl group (
δC 197.1) conjugated with the Δ
3(4) double bond in
3 instead of the C-2 methylene group in the latter. This assignment was verified by scrutiny of the HMBC and
1H–
1H COSY correlations (
Fig. 6). Further interpretation of the NOESY data revealed that compound
3 shared the same relative configuration as mangelonine H. Lastly, based on XRD analysis [
Fig. 5, Flack parameter = 0.01(9)] [
40], the absolute configuration of
3 was unambiguously determined to be 5
S,9
S,10
S,12
S as depicted. Crolatinoids A (
1) and B (
2) both exhibit novel skeletal structures, leading us to undertake an investigation into their biosynthetic origins. From a structural and biogenetic perspective, compound
1 likely originated from a phenyllabdane sesterterpenoid (
Scheme 1),
e.g
.,
ⅰ that was formed through enzyme-catalyzed cyclization and elimination of geranylfarnesyl diphosphate (GFDP). Subsequent oxidation would give intermediate
ⅱ, which undergoes a dehydration process to produce polyene
ⅲ. A subsequent 6
π electrocyclization connects C-12 and C-17, affording cyclohexadiene
ⅳ. Oxidation at the C-7, C-8, C-19, and C-20 positions would result in intermediate
ⅴ, featuring a vicinal diol and
β-keto carboxylic acid group. Then, this intermediate would undergo an intramolecular decarboxylation to form enol
vi, which would transform into compound
1 via enol-keto tautomerism and oxidative aromatization.
Compound
2 is characterized by an unprecedented 19(5→4)-abeo-3,5-cycloneoclerodane skeleton, which is hypothesized to be biogenetically derived from compound
3 through an oxa-di-
π-methane rearrangement (
Scheme 2) [
42]. Specifically, upon photochemical irradiation, compound
3 would initially undergo homolytic cleavage of both the Δ
3(4) double bond and the C-19 aldehyde carbonyl group, which is followed by the formation of the C-4–C-19 bond, yielding the diradical intermediate
ⅶ. Subsequent C-5–C-19 bond cleavage would generate the intermediate
ⅷ, which would be converted into compound
2 via the C-3–C-5 bond formation. To validate the biosynthetic hypothesis, a transformation from
3 to
2 was envisaged. However, the minute isolated quantity of compound
3 (1.0 mg) hindered this endeavor, necessitating additional verification through a synthetic approach.
At this stage, we moved forward to explore the biological potential of these structurally fascinating scaffolds. A significant challenge in cancer treatment arises from the frequent development of resistance in cancer cells to both chemotherapeutic agents and targeted medications. Natural products play a pivotal role as a promising resource for combating cancer multidrug resistance (MDR) [
43–
46]. In the present study, the cytotoxicity and MDR reversal activity of compound
1 were tested in paclitaxel-resistant colorectal cancer cells (HCT-15/Taxol). The findings indicated that compound
1 exhibited no cytotoxicity to these cell lines at a concentration of 20 µmol/L. Moreover, a combination therapy with compound
1 could significantly enhance the efficacy of paclitaxel in HCT-15/Taxol cell lines with a half maximal inhibitory concentration (IC
50) value of 1.27 µmol/L, demonstrating a reversal fold (RF) value of 16 (
Fig. 7).
Obesity is a wide-spreading epidemic that significantly impacts quality of life, making medical treatment increasingly crucial [
47]. In this study, compounds
1–3 were evaluated for their capacity to suppress lipid deposition in the 3T3-L1 adipocyte model [
48–
50], utilizing triglycerides (TG) levels as an indicator and oil red O (ORO) staining for visual verification (
Fig. 8, Tables S4 and S5 in Supporting information), while metformin [
51,
52] served as the positive control. The results showed that compounds
1–3 demonstrated a dose-dependent reduction in TG levels with series doses, while exhibiting low cytotoxicity (
Table 1). Significantly, compounds
1–3 all showed inhibition on the adipocyte differentiation of 3T3-L1 cells with half maximal effective concentration (EC
50) values of 4.38, 32.62, and 18.08 µmol/L, respectively, which were either superior to or comparable with the efficacy of metformin.
In conclusion, three novel terpenoids including a 12,17-cyclo20-nor phenyllabdane sesterterpenoid (1), a 19(5→4)-abeo-3,5-cycloneoclerodane diterpenoid (2), and a neoclerodane diterpenoid (3), were isolated from the bark of Croton laui in the present study. Compounds 1 and 2 exhibit two unprecedented carbon skeletons. Notably, compound 1 marks the first discovery of a 12,17-cyclostructure within the phenyllabdane sesterterpenoid class. Biological evaluation revealed that compound 1 could significantly reverse multidrug resistance in paclitaxel-resistant HCT-15 cells. Additionally, all compounds 1–3 exhibited antiadipogenic activity in 3T3-L1 adipocytes. This study not only enhances the diversity of terpenoid carbon skeletons but also offers new valuable resources for research in multidrug resistance reversal and anti-adipogenesis.
PDF
Full
65
36
关闭全屏
No. 86 Xueyuan South Road, Haidian District, Beijing
010-62662699(Operation and Technical Support)
qkjq@cast.org.cn
