Jia-Meng
Dai‡
ab,
Bing-Chao
Yan‡
a,
Ling-Mei
Kong
a,
Kun
Hu
a,
Xing-Ren
Li
a,
Xiao-Nian
Li
a,
Han-Dong
Sun
a,
Yan
Li
a and
Pema-Tenzin
Puno
*a
aState Key Laboratory of Phytochemistry and Plant Resources in West China, and Yunnan Key Laboratory of Natural Medicinal Chemistry, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming 650201, China. E-mail: punopematenzin@mail.kib.ac.cn
bUniversity of Chinese Academy of Sciences, Beijing 100049, China
First published on 7th November 2023
Two types of novel diterpene heterodimers, isoxerophilins A (1, ent-kaurane–sempervirane) and B (2, ent-kaurane–abietane), were isolated from the roots of Isodon xerophilus. Their structures and absolute configurations were determined by spectroscopic data and single crystal X-ray diffraction analysis. A bioinspired semisynthesis of ten isoxerophilin analogues, featuring a tandem oxidative dearomatization/intermolecular Diels–Alder reaction, has been achieved, and the transformation of ferruginol into a sempervirane scaffold was accomplished by the SIBX-mediated tandem reaction. Compound 12 promotes MICA/B expression on SMMC-7721 cells.
Isodon xerophilus H. Hara, a perennial shrub, has long been used in Yunnan Province, China, for its anti-tumor, anti-inflammatory, and anti-bacterial properties. Previous studies on the aerial parts of I. xerophilus have disclosed the presence of some ent-kaurane diterpenoids, meroditerpenoids, and homodimers.13 To further explore the structural and biological diversity of diterpene dimers, we investigated the secondary metabolites in one-year-old roots of I. xerophilus for the first time, which was cultivated in Kunming Botanical Garden. Isoxerophilins A (1) and B (2) (Fig. 1) represented two classes of novel diterpene heterodimers formed from ent-kaurane and sempervirane units, and ent-kaurane and abietane units, respectively. Furthermore, a divergent bioinspired semisynthesis of ten isoxerophilin analogues was accomplished starting from natural dehydroabietic acid (3), which facilitated the discovery of a new type of dimer exhibiting tumor immunotherapy activity. Herein, we describe the structural elucidation and plausible biogenetic pathways of 1 and 2. Moreover, we present a bioinspired semisynthesis and biological evaluation of isoxerophilin analogues.
| No. | 1 | 2 | ||
|---|---|---|---|---|
| δ C | δ H (mult., J in Hz) | δ C | δ H (mult., J in Hz) | |
| 1 | 40.4 | 1.67 (overlapped) | 40.4 | 1.70 (overlapped) |
| 0.82 (overlapped) | 0.86 (overlapped) | |||
| 2 | 19.5 | 1.61 (overlapped) | 19.6 | 1.71 (overlapped) |
| 1.54 (overlapped) | 1.45 (overlapped) | |||
| 3 | 36.8 | 2.09 (overlapped) | 37.0 | 2.10 (overlapped) |
| 1.03 (overlapped) | 1.10 (overlapped) | |||
| 4 | 39.0 | 39.1 | ||
| 5 | 46.4 | 2.14 (d, 12.7) | 46.4 | 2.17 (d, 12.8) |
| 6 | 27.3 | 2.05 (overlapped) | 27.6 | 2.08 (overlapped) |
| 1.61 (overlapped) | 1.71 (overlapped) | |||
| 7 | 73.7 | 4.04 (t, 2.7) | 73.0 | 4.11 (t, 2.8) |
| 8 | 53.9 | 55.2 | ||
| 9 | 50.2 | 1.71 (overlapped) | 49.6 | 1.77 (dd, 19.2, 10.3) |
| 10 | 40.5 | 40.6 | ||
| 11 | 30.4 | 1.32 (overlapped) | 20.2 | 1.61 (overlapped) |
| 1.24 (overlapped) | 1.05 (overlapped) | |||
| 12 | 19.3 | 1.61 (overlapped) | 27.3 | 2.08 (overlapped) |
| 1.54 (overlapped) | 1.71 (overlapped) | |||
| 13 | 39.7 | 2.48 (m) | 41.0 | 2.26 (overlapped) |
| 14 | 33.8 | 2.08 (overlapped) | 34.1 | 2.25 (overlapped) |
| 1.85 (overlapped) | 1.87 (overlapped) | |||
| 15 | 228.3 | 232.8 | ||
| 16 | 56.6 | 60.8 | ||
| 17 | 29.8 | 2.57 (dd, 12.7, 3.0) | 27.4 | 2.33 (dd, 14.0, 2.7) |
| 1.67 (overlapped) | 1.54 (dd, 14.0, 2.7) | |||
| 18 | 28.3 | 1.22 (s) | 28.5 | 1.26 (s) |
| 19 | 65.1 | 3.93 (d, 10.7) | 65.4 | 3.98 (d, 10.7) |
| 3.66 (d, 10.7) | 3.75 (d, 10.7) | |||
| 20 | 18.5 | 1.00 (s) | 18.5 | 1.10 (s) |
| 1′ | 36.0 | 1.92 (m) | 36.1 | 1.80 (m) |
| 1.52 (m) | 1.29 (m) | |||
| 2′ | 19.7 | 1.68 (overlapped) | 19.0 | 1.65 (overlapped) |
| 1.42 (overlapped) | 1.41 (overlapped) | |||
| 3′ | 42.3 | 1.42 (overlapped) | 42.5 | 1.46 (m) |
| 1.28 (overlapped) | 1.25 (overlapped) | |||
| 4′ | 33.7 | 34.0 | ||
| 5′ | 52.0 | 1.61 (overlapped) | 51.7 | 1.33 (dd, 12.4, 4.3) |
| 6′ | 18.9 | 1.60 (overlapped) | 19.5 | 1.70 (overlapped) |
| 1.46 (overlapped) | 1.46 (overlapped) | |||
| 7′ | 31.5 | 2.04 (overlapped) | 30.3 | 2.30 (dd, 11.4, 6.8) |
| 1.90 (overlapped) | 2.23 (m) | |||
| 8′ | 126.6 | 139.0 | ||
| 9′ | 149.5 | 137.7 | ||
| 10′ | 38.5 | 37.9 | ||
| 11′ | 42.6 | 3.42 (t, 2.8) | 54.0 | 3.37 (s) |
| 12′ | 76.4 | 210.7 | ||
| 13′ | 209.4 | 76.2 | ||
| 14′ | 58.1 | 3.23 (s) | 46.2 | 2.86 (t, 2.9) |
| 15′ | 34.0 | 2.03 (m) | 36.0 | 1.94 (m) |
| 16′ | 18.6 | 1.40 (d, 6.8) | 19.1 | 1.19 (d, 6.9) |
| 17′ | 18.4 | 1.22 (d, 6.8) | 18.6 | 1.43 (d, 6.9) |
| 18′ | 34.0 | 0.88 (s) | 33.6 | 0.90 (s) |
| 19′ | 22.1 | 0.86 (s) | 22.1 | 0.83 (s) |
| 20′ | 20.8 | 1.18 (s) | 21.2 | 0.91 (s) |
Its planar structure was determined based on the interpretation of 2D NMR spectra, which implied that it can be divided into units A and B (Fig. 2). The key 1H–1H COSY correlations of H2-1/H2-2/H2-3, H-5/H2-6/H-7, and H-9/H2-11/H2-12/H-13/H2-14, along with the HMBC correlations of H-7 with C-9; H-9 with C-8, C-14, and C-15; H-13 with C-16; Ha-17 with C-13; H2-19 with C-3, C-4, C-5, and C-18; and H3-20 with C-1, C-5, C-9, and C-10, established a C-20 non-oxygenated ent-kaurane scaffold (unit A). Unit B was determined to be a sempervirane scaffold from the key 1H–1H COSY correlations of H2-1′/H2-2′/H2-3′, H-5′/H2-6′/H2-7′, and H3-16′/H-15′/H3-17′, along with the key HMBC correlations of H2-6′ with C-7′ and C-8′; H-11′ with C-9′, C-12′, C-13′, and C-15′; H-14′ with C-8′, C-9′, and C-13′; H3-18′ with C-3′, C-4′, and C-5′; H3-19′ with C-4′ and C-5′; and H3-20′ with C-1′, C-5′, C-9′, and C-10′. The remaining HIDs indicated that an additional ring joined units A and B together. This deduction was further confirmed by the HMBC correlations of H-11′ with C-16, and H-14′ with C-13, C-15, C-16, and C-17. Units A and B were linked via C-16/C-14′ and C-17/C-11′ in a six-membered ring, which was presumably formed through [4 + 2] cycloaddition.
The relative configuration of 1 was partially determined by the ROESY correlations of H-7/H-14b, H-14′/H-14b, H-14′/H-13, and H3-16′/H3-20′ (Fig. S2†). Finally, its structure including the absolute configuration was established as 4R,5S,7S,8R,9S,10R,13R,16R,5′S,10′S,11′S,12′S,14′S by single crystal X-ray diffraction analysis using Cu Kα radiation with a perfect Flack parameter of 0.02(14) (CCDC 2059945,†Fig. 3). Thus, 1 was identified as a novel dimeric diterpenoid formed from ent-kaurane and sempervirane units. To the best of our knowledge, semperviranoids are rare in nature, and fewer than twenty semperviranoids have been discovered so far.14
![]() | ||
| Fig. 3 X-ray crystallographic structures of 1 and 2 (displacement ellipsoids are drawn at the 30% probability level). | ||
The molecular formula of isoxerophilin B (2) was determined to be C40H60O5 based on the HRESIMS data (m/z 643.4343 [M + Na]+, calcd 643.4338), the same as that of 1. Detailed analysis of the 2D NMR spectra established the planar structure of 2 (Fig. 2), which could be divided into units A and B. Unit A was assigned to be the same ent-kaurane scaffold as that of 1, which was supported by the HMBC correlations of H-5 with C-3, C-4, C-9 and C-10; H-7 with C-9; H-9 with C-8, C-14 and C-15; H-13 with C-16 and C-17; H2-19 with C-3, C-4, C-5, and C-18; and H3-20 with C-1, C-5, C-9, and C-10, together with the 1H–1H COSY correlations of H2-1/H2-2/H2-3, H-5/H2-6/H-7, and H-9/H2-11/H2-12/H-13/H2-14 (Fig. 2). However, unit B in 2 was assigned to be an abietane scaffold by the 1H–1H COSY correlations of H2-1′/H2-2′/H2-3′, H-5′/H2-6′/H2-7′, and H3-16′/H-15′/H3-17′, along with the HMBC correlations of H-5′ with C-9′ and C-10′; H-7′ with C-8′, C-9′, and C-14′; H-11′ with C-8′, C-9′, and C-12′; H-15′ with C-12′, C-13′, and C-14′; H3-19′ with C-3′, C-4′, C-5′, and C-18′; and H3-20′ with C-1′, C-5′, C-9′, and C-10′. An additional six-membered ring in 2 different from that in 1 was formed between units A and B presumably through biogenetic [4 + 2] cycloaddition. The linkage of C-16/C-11′ and C-17/C-14′ was supported by the HMBC correlations of H-13 with C-11′ and H-11′ with C-15 and C-17 and the 1H–1H COSY correlations of H2-17/H-14′.
The relative configuration of 2 could be partially assigned by the ROESY spectrum. The ROESY correlations of H-7/H-13α, H2-19/H3-20α, and H-11′/H-14α indicated the α-orientations of H-7, HO-19, and H-11′, respectively. To establish the whole stereochemistry, single crystals of 2 were obtained and subjected to Cu Kα X-ray diffraction analysis with a sufficient Flack parameter of 0.07(6) (CCDC 2059944†), which verified the planar structure of 2 and determined its absolute configuration to be 4R,5S,7S,8R,9S,10R,13R,16R,5′S,10′S,11′R,13′R,14′R (Fig. 3).
The intriguing structures of 1 and 2, to some extent, implied a unique biosynthetic pathway. As shown in Scheme 1, heterodimer 1 was formed by oxidatively dearomatized sempervirane B and ent-kaurane C units, while 2 was formed by oxidatively dearomatized abietane A and ent-kaurane C. To the best of our knowledge, no semperviranoids have been discovered from I. xerophilus.13 Therefore, we proposed that the abietanoid, ferruginol (4), underwent oxidative dearomatization to afford intermediate A. The Diels–Alder reaction between A and C produced heterodimer 2. On the other hand, the 1,2-shift (α-ketol rearrangement) of A gave sempervirane skeleton B, which underwent a Diels–Alder reaction with C to produce heterodimer 1.
Inspired by the proposed biogenetic pathways of 1 and 2, we initiated a concise semisynthesis of isoxerophilin-like dimers for a further biological study. As shown in Scheme 2, isoxerophilin analogues 11–20 were prepared through sequential oxidative dearomatization of 4, followed by intermolecular [4 + 2] cycloaddition with ent-kauranoids 6–10.2c These samples are from the compound library of our laboratory, respectively. 8,11,13-Abietatriene (5) was prepared as previously described in the literature starting from natural dehydroabietic acid (3) through reduction (LiAlH4), tosylation (tosyl chloride in pyridine), and reduction (Zn powder-NaI in DMF).15 Treatment of 5 with phthaloyl peroxide16 afforded 4 in 65% yield in a single step, instead of Friedel–Crafts acetylation, Baeyer–Villiger oxidation and methanolysis protocols.15
With ferruginol (4) in hand, we then explored the key tandem oxidative dearomatization/intermolecular Diels–Alder reaction for the synthesis of isoxerophilin-like heterodimers. When 4 and ent-kauranoids 6–8 were treated with Pb(OAc)4 in acetic acid, respectively, only the oxidatively dearomatized acetate ketals were detected. We then changed the oxidant to SIBX (stabilized 2-iodoxybenzoic acid).17 To our delight, the tandem reaction occurred smoothly to give the desired dimers 11–16 (Scheme 2). Furthermore, oxidation of HO-15 in ent-kauranoid 9 to enone and sequential oxidative dearomatization of 4 occurred in one pot in the presence of SIBX, followed by a tandem intermolecular Diels–Alder reaction to give isoxerophilin B analogue 17. In the presence of SIBX, allylic acetate 10 (maoecrystal B) was hydrolyzed and then oxidized to enone as a dienophile for oxidative dearomatization and an intermolecular Diels–Alder tandem reaction, producing heterodimers 18–20.
The structures of 12 and 18 were confirmed by single crystal X-ray diffraction analysis (CCDC 2129366† for 12 and CCDC 2129362† for 18). The bioinspired semisynthesis provided evidence for the proposed biogenetic pathways of 1 and 2, featuring oxidative dearomatization and intermolecular [4 + 2] cycloaddition, coupled with isopropyl migration of ferruginol for 1.
NKG2D (natural killer group 2 member D) is a typical activating receptor expressed on NK cells and plays an important role in initial surveillance and elimination of cancer through interacting with NKG2D ligands represented by ULBPs and MICA/B on cancer cells.18,19 The reduction and shedding of NKG2D ligands resulting in NK-mediated immune escape have been frequently observed during tumor progression.20 Therefore, induction of NKG2D ligands on tumor cells renders cancer cells susceptible to immune destruction and is considered to be a promising immunotherapy strategy.
With ample samples of the synthesized analogues 11–20, flow cytometric analysis was undertaken to evaluate the surficial level of the NKG2D ligand MICA/B on hepatocellular carcinoma SMMC-7721 cells. As shown in Fig. 4, 12 markedly increased the cell surficial engagement of MICA/B with a mean fluorescence intensity (MFI) up to 164% at a concentration of 40 μM without dominant cytotoxicity,21 while the NKG2D ligand-inducing activity was weak in the following serial dilution. But it is of great interest to optimize the structure of 12 to improve the NKG2D ligand-inducing activity. It is worthwhile to conduct further pharmacological studies on its antitumor immune responses and mechanisms of action.
:
0–0
:
1, v/v; gradient system). Four major fractions were obtained from the silica gel column fractions, A–D. Fr. B (CHCl3–Me2CO, 8
:
2, v/v; 36.0 g), a gum, was decolorized using MCI gel and eluted with MeOH–H2O (90
:
10, v/v) to provide two fractions B1 (26.0 g) and B2 (8.0 g). Fr. B1 was subjected to RP-18 silica gel CC (MeOH–H2O, 30
:
70–100
:
0, v/v; gradient) to provide four fractions, B1-1–B1-4. Fr. B1-3 (3.0 g) was purified by repeated silica gel CC (petroleum ether–Me2CO, 10
:
1–1
:
1, v/v) to give three subfractions (B1-3-1–B1-3-3) based on TLC analysis. Fr. B1-3-2 (300.0 mg) was purified using Sephadex LH-20 with CHCl3–MeOH (1
:
1, v/v) to afford fractions B1-3-2-1–B1-3-2-3; Fr. B1-3-2-2 was subjected to semi-preparative HPLC (3 mL min−1, detector UV λmax 210 nm, CH3CN–H2O, 20
:
80, v/v) to afford compound 1 (1.0 mg) and compound 2 (6.0 mg).
ε) 203 (3.43), 248 (2.85), 333.5 (2.21) nm; ECD (MeOH) λmax (Δε) 196 (0.34), 238 (−0.26), 255 (−0.22), 308 (0.16), 336 (−0.03) nm; IR (KBr) νmax 3473, 3424, 2959, 2922, 2852, 1702, 1628, 1466, 1384, 1261, 1097, 1024, 842, 802, cm−1; 1H and 13C NMR data, see Table S1;† positive HRESIMS [M + Na]+m/z 643.4323 (calcd for C40H60O5Na, 643.4333).
ε) 203 (3.40), 239 (2.93), 318 (2.24) nm; ECD (MeOH) λmax (Δε) 209 (−0.14), 245 (0.37), 327 (−0.43) nm; IR (KBr) νmax 3440, 2926, 2854, 1717, 1630, 1384, 1169, 1053, 911, 878, 721, 668 cm−1; 1H and 13C NMR data, see Tables 1, S1 and S3;† positive HRESIMS [M + Na]+m/z 643.4343 (calcd for C40H60O5Na, 643.4338).
:
4, v/v), C40H60O5·CH4O, M = 652.92, a = 7.9127(2) Å, b = 18.4253(6) Å, c = 24.4103(7) Å, α = 90°, β = 90°, γ = 90°, V = 3558.87(18) Å3, T = 100(2) K, space group P212121, Z = 4, μ(Cu Kα) = 0.624 mm−1, 17
388 reflections measured, 6014 independent reflections (Rint = 0.1114). The final R1 values were 0.0703 (I > 2σ(I)). The final wR(F2) values were 0.1482 (I > 2σ(I)). The final R1 values were 0.1559 (all data). The final wR(F2) values were 0.1885 (all data). The goodness of fit on F2 was 1.102. Flack parameter = 0.02(14).
:
2, v/v), C40H60O5, M = 620.88, a = 10.6576(2) Å, b = 7.68660(10) Å, c = 21.0221(4) Å, α = 90°, β = 98.7830(10)°, γ = 90°, V = 1701.95(5) Å3, T = 293(2) K, space group P21, Z = 2, μ(Cu Kα) = 0.606 mm−1, 16
771 reflections measured, 5769 independent reflections (Rint = 0.0412). The final R1 values were 0.0492 (I > 2σ(I)). The final wR(F2) values were 0.1406 (I > 2σ(I)). The final R1 values were 0.0495 (all data). The final wR(F2) values were 0.1412 (all data). The goodness of fit on F2 was 1.039. Flack parameter = 0.07(6).
:
4, v/v), C40H54O8·CH4O·H2O, M = 712.89, a = 25.253(2) Å, b = 7.4272(6) Å, c = 20.3947(19) Å, α = 90°, β = 104.462(4)°, γ = 90°, V = 3704.0(6) Å3, T = 100.(2) K, space group C2, Z = 4, μ(Cu Kα) = 0.731 mm−1, 33
150 reflections measured, 7223 independent reflections (Rint = 0.0765). The final R1 values were 0.0583 (I > 2σ(I)). The final wR(F2) values were 0.1591 (I > 2σ(I)). The final R1 values were 0.0639 (all data). The final wR(F2) values were 0.1665 (all data). The goodness of fit on F2 was 1.062. Flack parameter = 0.21(8).
:
2, v/v), C40H54O7, M = 646.83, a = 11.8280(11) Å, b = 7.7673(8) Å, c = 18.926(2) Å, α = 90°, β = 93.109(6)°, γ = 90°, V = 1736.2(3) Å3, T = 100.(2) K, space group P21, Z = 2, μ(Cu Kα) = 0.663 mm−1, 23
902 reflections measured, 6476 independent reflections (Rint = 0.2120). The final R1 values were 0.1050 (I > 2σ(I)). The final wR(F2) values were 0.2852 (I > 2σ(I)). The final R1 values were 0.1867 (all data). The final wR(F2) values were 0.3370 (all data). The goodness of fit on F2 was 1.096. Flack parameter = 0.2(3).
:
1) to afford 4 (137.6 mg, 0.48 mmol, 65%) as a pale yellow oil. 4: Rf = 0.70 (pentane); 1H NMR (600 MHz, CDCl3) δH: 6.83 (s, 1H), 6.63 (s, 1H), 4.53 (s, 1H), 3.11 (p, J = 6.9 Hz, 1H), 2.81 (dddd, J = 28.2, 17.0, 9.1, 6.2 Hz, 2H), 2.17 (dt, J = 13.2, 3.5 Hz, 1H), 1.85 (ddt, J = 12.1, 7.5, 2.3 Hz, 1H), 1.80–1.54 (m, 3H), 1.46 (dq, J = 13.4, 2.9 Hz, 1H), 1.42–1.29 (m, 2H), 1.23 (t, J = 6.8 Hz, 6H), 1.17 (s, 3H), 0.94 (s, 3H), 0.91 (s, 3H); 13C NMR (150 MHz, CDCl3) δC: 150.8, 148.8, 131.5, 127.4, 126.8, 111.1, 50.5, 41.8, 39.0, 37.7, 33.6, 33.5, 29.9, 26.9, 25.0, 22.9, 22.7, 21.8, 19.5, 19.4. All analytical data agree with literature values.15
:
40) to afford the dimeric products.
Compound 11 (2.00 mg, 19%): white amorphous powder, Rf = 0.49 (silica, EtOAc–petroleum ether, 1
:
5); [α]26D = −6.3 (c = 0.100 in MeOH); 1H and 13C NMR data, see Tables S1 and S3;† HRESIMS (m/z): [M + NH4]+ calcd for C40H58O8N, 680.4157, found 680.4162.
Compound 12 (1.80 mg, 18%): colorless crystals, m.p. 276.0–281.0 °C; Rf = 0.49 (silica, EtOAc–petroleum ether, 1
:
5); [α]26D = −12.1 (c = 0.060 in MeOH); 1H and 13C NMR data, see Tables S1 and S3;† HRESIMS (m/z): [M + Na]+ calcd for C40H54O8Na, 685.3711, found 685.3716.
The 1H–1H COSY spectrum of 11 indicated the presence of two spin systems: H2-17/H-14′ and H3-16′/H-15′/H3-17′ (Fig. S1†). The linkage of 4 and 6 through C-16/C-11′ and C-17/C-14′ bonds could be determined by the HMBC correlations from H-11′ to C-13, C-15, C-16, C-17, C-8′, C-9′, and C-12′ and H-14′ to C-16, C-7′, C-8′, C-9′, C-12′, C-13′, and C-15′ (Fig. S1†). Additionally, the ROESY correlations of H-11′/H-13, H-11′/H-14b, and H3-16′/H3-20′ indicated that the configurations of C-16 and C-13′ were R and R, respectively (Fig. S2†). Finally, the structure of 11 was established as shown.
The 1H–1H COSY spectrum of 12 indicated the presence of two spin systems: H2-17/H-11′ and H3-16′/H-15′/H3-17′ (Fig. S1†). The linkage of 4 and 6 through C-16/C-14′ and C-17/C-11′ bonds could be determined by the HMBC correlations from H-11′ to C-16, C-8′, C-9′, C-12′, and C-13′; from H-14′ to C-13, C-15, C-16, C-17, C-8′, C-9′, and C-13′; and from H2-17 to C-12′ (Fig. S1†). Additionally, the ROESY correlations of H-14′/H2-14 and H3-20′/H3-16′(H3-17′) indicated that the configurations of C-16 and C-12′ were R and S, respectively (Fig. S2†). Finally, the structure of 12 was established as shown.
Compound 13 (2.00 mg, 17%): white amorphous powder, Rf = 0.51 (silica, EtOAc–petroleum ether, 1
:
5); [α]26D = −83.1 (c = 0.100 in MeOH); 1H and 13C NMR data, see Tables S1 and S3;† HRESIMS (m/z): [M + Na]+ calcd for C42H60O9Na, 731.4130, found 731.4127.
Compound 14 (1.80 mg, 19%): white amorphous powder, Rf = 0.51 (silica, EtOAc–petroleum ether, 1
:
5); [α]26D = +10.7 (c = 0.040 in MeOH); 1H and 13C NMR data, see Tables S1 and S3;† HRESIMS (m/z): [M + H]+ calcd for C42H61O9, 709.4310, found 709.4318.
The 1H–1H COSY spectrum of 13 indicated the presence of two spin systems: H2-17/H-14′ and H3-16′/H-15′/H3-17′ (Fig. S1†). The linkage of 4 and 6 through C-16/C-11′ and C-17/C-14′ bonds could be determined by the HMBC correlations from H-11′ to C-13, C-16, C-8′, C-9′, C-12′, and C-13′ and from H-14′ to C-16, C-7′, C-8′, C-9′, C-12′, C-13′ and C-15′ (Fig. S1†). Additionally, the ROESY correlations of H-11′/H-14a and H3-16′/H-7′b indicated that the configurations of C-16 and C-12′ were R and R, respectively (Fig. S2†). Finally, the structure of 13 was established as shown.
The 1H–1H COSY spectrum of 14 indicated the presence of two spin systems: H2-17/H-14′ and H3-16′/H-15′/H3-17′ (Fig. S1†). The linkage of 4 and 6 through C-16/C-11′ and C-17/C-14′ bonds could be determined by the HMBC correlations from H-11′ to C-15, C-16, C-17, C-8′, C-9′, C-12′, and C-13′ and from H-14′ to C-16, C-7′, C-8′, C-9′, C-12′, C-13′, and C-15′ (Fig. S1†). Additionally, the ROESY correlations of H-14a/H-11′ and H-15′/H-17a indicated that C-16 and C-13′ had R and S configurations, respectively (Fig. S2†). Finally, the structure of 14 was established as shown.
Compound 15 (2.00 mg, 18%): white amorphous powder, Rf = 0.51 (silica, EtOAc–petroleum ether, 1
:
5); [α]26D = −14.0 (c = 0.060 in MeOH); 1H and 13C NMR data, see Tables S2 and S3;† HRESIMS (m/z): [M + Na]+ calcd for C42H58O9Na, 729.3973, found 729.3982.
Compound 16 (2.30 mg, 20%): white amorphous powder, Rf = 0.51 (silica, EtOAc–petroleum ether, 1
:
5); [α]26D = −6.80 (c = 0.130 in MeOH); 1H and 13C NMR data, see Tables S2 and S3;† HRESIMS (m/z): [M + Na]+ calcd for C42H58O9Na, 729.3973, found 729.3972.
The 1H–1H COSY spectrum of 15 indicated the presence of two spin systems: H2-17/H-14′ and H3-16′/H-15′/H3-17′ (Fig. S1†). The linkage of 4 and 6 through C-16/C-11′ and C-17/C-14′ bonds could be determined by the HMBC correlations from H-11′ to C-13, C-15, C-16, C-17, C-8′, C-9′, C-10′, C-12′, and C-13′ and from H-14′ to C-16, C-7′, C-8′, C-9′, C-12′, C-13′, and C-15′ (Fig. S1†). Additionally, the ROESY correlations of H-11′/H-14a and H3-16′(H3-17′)/H-7′b indicated that C-16 and C-13′ had R and R configurations, respectively (Fig. S2†). Finally, the structure of 15 was established as shown.
The 1H–1H COSY spectrum of 16 indicated the presence of two spin systems: H2-17/H-14′ and H3-16′/H-15′/H3-17′ (Fig. S1†). The linkage of 4 and 6 through C-16/C-11′ and C-17/C-14′ bonds could be determined by the HMBC correlations from H-11′ to C-13, C-15, C-16, C-17, C-8′, C-9′, C-10′, C-12′, and C-13′ and from H-14′ to C-16, C-7′, C-8′, C-9′, C-12′, C-13′, and C-15′ (Fig. S1†). Additionally, the ROESY correlations of H-11′/H-14a and H3-16′(H3-17′)/H-17a(H-17b) indicated that C-16 and C-13′ had R and S configurations, respectively (Fig. S2†). Finally, the structure of 16 was established as shown.
Compound 17 (1.50 mg, 18%): white amorphous powder, Rf = 0.52 (silica, EtOAc–petroleum ether, 1
:
5); [α]24D = −2.17 (c = 0.060 in MeOH); 1H and 13C NMR data, see Tables S2 and S3;† HRESIMS (m/z): [M + Na]+ calcd for C40H60O4Na, 627.4384, found 627.4376.
The 1H–1H COSY spectrum of 17 indicated the presence of two spin systems: H2-17/H-14′ and H3-16′/H-15′/H3-17′ (Fig. S1†). The linkage of 4 and 6 through C-16/C-11′ and C-17/C-14′ bonds could be determined by the HMBC correlations from H-11′ to C-13, C-16, C-17, C-8′, C-9′, C-10′, C-12′, and C-13′ and from H-14′ to C-16, C-7′, C-8′, C-9′, C-12′, and C-13′ (Fig. S1†). Additionally, the ROESY correlations of H-11′/H-14b, H3-16′/H-7′b, and H3-17′/H-7′a indicated that C-16 and C-13′ had R and R configurations, respectively (Fig. S2†). Finally, the structure of 17 was established as shown.
:
40) to afford compounds 18 (4.00 mg, 18%), 19 (4.00 mg, 18%), and 20 (2.00 mg, 9%).
Compound 18: colorless crystals, m.p. 238.0–240.0 °C; Rf = 0.51 (silica, EtOAc–petroleum ether, 1
:
5); [α]26D = −58.2 (c = 0.075 in MeOH); 1H and 13C NMR data, see Tables S2 and S3;† HRESIMS (m/z): [M + H]+ calcd for C40H55O7, 647.3942, found 647.3944.
Compound 19: white amorphous powder, Rf = 0.51 (silica, EtOAc–petroleum ether, 1
:
5); [α]26D = −15.9 (c = 0.074 in CHCl3); 1H and 13C NMR data, see Tables S2 and S3;† HRESIMS (m/z): [M + H]+ calcd for C40H55O7, 647.3942, found 647.3948.
Compound 20: white amorphous powder, Rf = 0.51 (silica, EtOAc–petroleum ether, 1
:
5); [α]26D = −19.1 (c = 0.070 in MeOH); Tables S2 and S3;† HRESIMS (m/z): [M + Na]+ calcd for C40H54O7Na, 669.3762, found 669.3762.
The 1H–1H COSY spectrum of 18 indicated the presence of two spin systems: H2-17/H-14′ and H3-16′/H-15′/H3-17′ (Fig. S1†). The linkage of 4 and 6 through C-16/C-11′ and C-17/C-14′ bonds could be determined by the HMBC correlations from H-11′ to C-13, C-16, C-17, C-8′, C-9′, C-10′, C-12′, and C-13′ and from H-14′ to C-16, C-7′, C-8′, C-9′, C-12′, and C-15′ (Fig. S1†). Additionally, the ROESY correlations of H-11′/H-14a and H3-16′/H-20′ indicated that C-16 and C-13′ had R and R configurations, respectively (Fig. S2†). Finally, the structure of 18 was established as shown.
The 1H–1H COSY spectrum of 19 indicated the presence of two spin systems: H2-17/H-14′ and H3-16′/H-15′/H3-17′ (Fig. S1†). The linkage of 4 and 6 through C-16/C-11′ and C-17/C-14′ bonds could be determined by the HMBC correlations from H-11′ to C-13, C-15, C-16, C-17, C-9′, C-10′, and C-12′ and from H-14′ to C-16, C-7′, C-9′, C-12′, and C-15′ (Fig. S1†). Additionally, the ROESY correlations of H-11′/H-14a and H3-16′/H-17a indicated that C-16 and C-13′ had R and S configurations, respectively (Fig. S2†). Finally, the structure of 19 was established as shown.
The 1H–1H COSY spectrum of 20 indicated the presence of two spin systems: H2-17/H-11′ and H3-16′/H-15′/H3-17′ (Fig. S1†). The linkage of 4 and 6 through C-16/C-14′ and C-17/C-11′ bonds could be determined by the HMBC correlations from H-11′ to C-8′, C-9′, C-12′, and C-13′; from H-14′ to C-13, C-16, C-17, C-8′, C-9′, C-10′, and C-12′; and from H2-17 to C-12′ (Fig. S1†). Additionally, the ROESY correlations of H-14′/H-14a and H3-20′/H3-16′ indicated that C-16 and C-13′ had R and S configurations, respectively (Fig. S2†). Finally, the structure of 20 was established as shown.
Footnotes |
| † Electronic supplementary information (ESI) available. CCDC 2059945, 2059944, 2129366 and 2129362. For ESI and crystallographic data in CIF or other electronic format see DOI: https://doi.org/10.1039/d3qo01679j |
| ‡ These authors contributed equally to this work. |
| This journal is © the Partner Organisations 2024 |