Limonoids with diverse frameworks from the stem bark of Entandrophragma angolense and their bioactivities

Abstract

Entangolensins A–P (1–16), sixteen new limonoids with diverse frameworks, were isolated from the stem bark of Entandrophragma angolense. Their planar structures were elucidated on the basis of spectroscopic data (HRMS, 1D/2D NMR), and the absolute configurations of most isolates were determined by the electronic circular dichroism (ECD) exciton chirality method and time-dependent density functional theory (TDDFT)/ECD calculations. Entangolensin A (1) was the first example of C-9/10-seco mexicanolide reported as a natural product. These diverse carbon skeletons, associated in a clear plausible biosynthetic pathway, indicated that the biosynthetic reactions in the title plant were varied and active. Bioactivity screening indicated that compounds 6, 12, 15 and 17 showed moderate cytotoxicities against HepG2 and MCF-7 cell lines, with IC₅₀ values from 13.19 to 36.93 μM; meanwhile, 6, 11 and 17 exhibited significant NO inhibitory activities in LPS-activated RAW 264.7 macrophages, with IC₅₀ values of 1.75, 7.94 and 4.63 μM.

---

Introduction

The discovery of limonoids, characteristic secondary metabolites in plants of the Meliaceae family, with highly complex skeletons and significant bioactivities has become a hot issue in the field of natural products.¹ Meliaceae plants are mainly distributed in tropical and subtropical regions throughout the world and are used as traditional medicines and economically important trees.^2,3 In recent years, Meliaceae plants distributed in Asia have received more attention from natural products researchers, and many novel limonoids have been discovered.^4,5 However, few pieces of research have recently been concerned with the Meliaceae plants native to Africa, such as Entandrophragma congoënse.⁶ Thus, the characteristic limonoid metabolites in Meliaceae plants from Africa are worthy of being thoroughly investigated.

Endrophragma is a genus of the Meliaceae family restricted to tropical Africa,⁷ and many plants from this genus provide economically important timbers and are extensively used in folk medicine to treat various diseases including gastric ulcers, malaria, and rheumatism.^8–10 Phytochemical investigations on this genus have led to the isolation of various secondary metabolites, such as cyclic and acyclic triterpenes,^11,12 limonoids,^13–15 protolimonoids¹⁶ and steroids,¹⁷ with interesting bioactivities including antifeedant,¹⁸ anti-inflammatory¹⁰ and anti-plasmodial.⁹ E. angolense is widely distributed and employed in Western Nigeria as an antimalarial and antiulcer¹⁹ agent in ethnomedicine, and a variety of ring D-seco and B,D-seco limonoids have been reported from this plant.^15,20 The aforementioned information suggests that the limonoids of the title plant are worthy to be studied as part of our program of discovering novel limonoids with significant bioactivities from the Meliaceae family. As a result, sixteen new limonoids, entangolensins A–P (1–16) (Fig. 1), and four known limonoids (17–20) were isolated and identified from the stem bark of E. angolense collected from Ghana. Cytotoxicities against HepG2 and MCF-7 cell lines and NO inhibitory activities in LPS-activated RAW 264.7 macrophages were also tested. Herein, we described the isolation, identification, and bioactivity screening of these compounds.

Fig. 1 Chemical structures of compounds 1–16.

Results and discussion

Entangolensin A (1) was isolated as a white amorphous powder, and had a molecular formula of C₂₇H₃₂O₇ as determined from the positive HRESIMS ion at m/z 491.2037 [M + Na]⁺ (calcd for C₂₇H₃₂O₇Na, 491.2040), indicative of twelve indices of hydrogen deficiency. The strong IR absorptions at 3423 and 1714 cm⁻¹ indicated the presence of hydroxyl and carbonyl moieties, respectively. The NMR data (Table 1) of 1 revealed resonances for a β-substituted furanyl ring (δ_H 7.48, 7.43, 6.45; δ_C 143.1, 141.4, 120.3, 110.2), a methoxyl (δ_H 3.72; δ_C 52.4), a ketone carbonyl (δ_C 201.9), two ester carbonyls (δ_C 175.1, 165.7), and three unsaturated bonds (δ_H 6.37, 6.15, 5.64; δ_C 158.1, 150.9, 138.4, 132.7, 131.2, 111.2). In the HMBC spectrum, the correlations from H-17 (δ_H 5.11) to C-12 (δ_C 30.0), C-18 (δ_C 16.1), C-20 (δ_C 120.3), C-21 (δ_C 141.4), and C-22 (δ_C 110.2) suggested the presence of a δ-lactone group in the D-ring. The cross-peaks from H₂-6 (δ_H 2.66, 2.51) and OCOCH̲₃ (δ_H 3.72) to C-7 (δ_C 175.1) confirmed the characteristic C-6/7 esterified appendage of B-seco limonoids.²¹ The above evidence indicated that compound 1 was a typical B,D-seco limonoid.²²

Table 1 ¹H NMR (500 MHz) and ¹³C NMR (125 MHz) data of compounds 1–6^a

Position	1		2		3		4		5		6
	δC	δH (J in Hz)	δC	δH (J in Hz)	δC	δH (J in Hz)	δC	δH (J in Hz)	δC	δH (J in Hz)	δC	δH (J in Hz)
a Measured in CDCl3.b Signals were overlapped.
1	150.9	6.37, s	155.8	6.70, d (10.4)	77.1	3.57, dd (6.4, 4.4)	81.5	4.06, dd (4.9, 4.9)	77.6	3.53, dd (5.9, 3.9)	77.5	3.48, dd (6.5, 4.5)
2a	132.7		124.7	5.95, d (10.4)	39.7	2.88, dd (14.3, 6.4)	38.3	2.86, dd (14.3, 4.9)	39.3	2.90, dd (14.4, 5.9)	39.8	2.81, dd (14.4, 6.5)
2b						2.52, dd (14.3, 4.4)		2.70^b		2.45, dd (14.4, 3.9)		2.59, dd (14.4, 4.5)
3	201.9		203.2		212.6		212.5		212.1		212.7
4	45.5		46.5		48.3		48.5		48.2		48.1
5	50.8	2.59, dd (7.8, 4.2)	43.6	2.88, dd (8.2, 2.6)	43.4	2.72, d (10.2)	49.1	2.57^b	43.1	2.84, d (10.3)	43.1	2.91, d (10.0)
6a	31.1	2.66, dd (15.5, 7.8)	31.5	2.56, dd (16.6, 8.2)	32.6	2.64, dd (15.9, 10.2)	31.9	2.43, brd (17.2)	32.8	2.63, dd (16.4, 10.3)	33.1	2.56^b
6b		2.51, dd (15.5, 4.2)		2.44, dd (16.6, 2.6)		2.31, d (15.9)		2.54^b		2.24, d (16.4)		2.24^b
7	175.1		173.9		173.9		174.3		173.8		173.7
8	131.2		137.8		142.4		145.9		145.2		145.7
9	138.4	6.15, brs	49.2	2.33, d (3.4)	58.1	2.19, s	53.0	3.06, d (8.7)	49.9	2.22^b	49.7	2.22^b
10	71.0		43.2		44.7		49.9		44.1		44.3
11a	22.6	2.34^b	21.4	2.25, brd (14.3)	67.8	4.54, brs	72.8	4.48, ddd (8.7, 8.5, 8.0)	24.1	2.28, m	23.7	2.23, m
11b		2.34^b		1.55^b						1.65, m		1.58, m
12a	30.0	1.51, m	28.3	1.49^b	37.6	2.10, dd (14.7, 4.2)	32.5	1.87, dd (13.7, 8.5)	29.6	2.01, td (12.8, 5.2)	29.2	2.21^b
12b		1.51, m		0.98^b		1.42, d (14.7)		1.55, dd (13.7, 8.0)		1.27, dd (12.8, 4.4)		1.07^b
13	37.6		39.2		41.0		39.1		42.2		42.1
14	158.1		135.0		80.7		72.5		80.2		79.7
15a	111.2	5.64, s	66.6	4.95, s	33.7	2.94, d (18.0)	39.3	3.01, d (17.8)	33.8	2.88, d (18.2)	33.9	2.89, d (17.6)
15b						2.62, d (18.0)		2.89, d (17.8)		2.58, d (18.2)		2.57, d (17.6)
16	165.7		174.7		170.0		169.7		169.2		169.2
17	80.8	5.11, s	82.9	5.03, s	79.4	5.64, s	78.6	5.65, s	80.0	5.56, d (1.3)	78.4	5.63, s
18	16.1	1.05, s	17.6	1.02, s	16.9	1.04, s	16.3	0.85, s	14.5	0.91, s	13.9	0.90, s
19	23.4	1.31, s	23.8	1.34, s	21.8	0.98, s	17.7	1.04, s	21.9	0.97, s	21.7	0.88, s
20	120.3		120.4		120.8		120.6		162.2		135.1
21	141.4	7.48, s	141.1	7.41, s	141.0	7.45, s	141.1	7.41, s	104.0	5.78, s	168.7
22	110.2	6.45, s	109.8	6.32, s	110.0	6.40, s	110.0	6.38, s	122.3	6.24, s	148.1	7.17, s
23	143.1	7.43, s	143.4	7.40, t-like	143.1	7.40, t-like	143.1	7.41, s	168.5		102.4	5.87, s
28	25.1	1.87, s	24.5	1.12, s	26.3	1.03, s	25.2	0.98, s	25.9	1.00, s	26.9	1.08, s
29	21.5	1.05, s	23.0	1.17, s	21.5	1.20, s	21.2	1.15, s	21.6	1.19, s	21.4	1.18, s
30a	31.7	3.30, d (16.7)	22.7	2.13, s	115.4	5.29, s	116.8	5.29, s	112.4	5.20, s	112.0	5.16, s
30b		3.07, d (16.7)				5.11, s		5.07, s		4.91, s		4.90, s
OMe-7	52.4	3.72, s	52.2	3.70, s	52.4	3.73, s	52.2	3.71, s	52.2	3.72, s	52.2	3.71, s
OMe-21									57.5	3.60, s
OMe-23											56.5	3.54, s

---

The 1D NMR signals of a double bond (δ_H 6.37; δ_C 150.9, 132.7) and a carbonyl carbon (δ_C 201.9) suggested that the A-ring of 1 possessed a 1-en-3-one system, which was further confirmed by the HMBC correlations from the gem-dimethyl H₃-29 (δ_H 1.05) and H₃-28 (δ_H 1.87) to C-3 (δ_C 201.9) and C-5 (δ_C 50.8), and from H-1 (δ_H 6.37) to C-3, C-5 and C-19 (δ_C 23.4). The key HMBC correlations from H₂-30 (δ_H 3.30, 3.07) to C-1 (δ_C 150.9), C-2 (δ_C 132.7) and the olefinic carbons at δ_C 131.2 (C-8) and δ_C 158.1 (C-14) indicated that a C-2/30/8 bridge was formed between the A and C rings. The aforementioned NMR data suggested that compound 1 resembled a mexicanolide-type limonoid skeleton.²³ The remaining two characteristic olefinic protons were assigned to H-9 (δ_H 6.15) and H-15 (δ_H 5.64) by the HMBC correlations from H-9 to C-12 and C-11 (δ_C 22.6); from H-15 to C-13 (δ_C 37.6), C-16 (δ_C 165.7) and C-8; and from H-9, H-15, H-17 and H₃-18 (δ_H 1.05) to C-14, which suggested that the Δ^8,9 and Δ^14,15 double bonds were fixed and there was accompanying cleavage between C-9 (δ_C 138.4) and C-10 (δ_C 71.0). In addition, a hydroxy group was assigned at C-10 by the HMBC correlations from H-5 (δ_H 2.59) and H₃-19 (δ_H 1.31) to C-10. Thus, the planar structure of 1 was depicted in Fig. 2.

Fig. 2 Key HMBC (¹H¹³C) and ROESY (¹H¹H) correlations of compound 1.

The relative configuration of 1 was assigned using a ROESY experiment. The ROESY cross-peaks were observed between H₃-19/H-6a, H-6a/H₃-29 and H-5/H₃-28, suggesting that CH₃-19 and H-6a were co-facial, adopting an α-orientation, and OH-10 and H-5 adopted a β-orientation. Also, ROESY correlation of H-12a/H-17 suggested a β-orientation of these protons (Fig. 2). To determine its absolute configuration, the ECD spectrum of 1 was recorded in MeCN, which showed three positive Cotton effects at λ_max 237, 258 and 283 nm, and a negative Cotton effect at λ_max 207 nm. However, the lack of a proper model compound as a reference and the complex ECD curve made it difficult to determine the absolute configuration by the ECD exciton chirality method. Therefore, we calculated the ECD of 1 using time-dependent density functional theory (TDDFT)²⁴ and compared the result with the experimental ECD data. The calculated ECD spectrum for 1 matched fairly well with the experimental ECD spectrum (Fig. 3), thus allowing the assignment of the absolute configuration of 1 as depicted. Therefore, the final structure of 1 was demonstrated as shown and named entangolensin A, which was presented as the first example of a C-9/10-seco mexicanolide-type limonoid reported in nature.²⁵

Fig. 3 Calculated and experimental ECD spectra of 1 (in MeCN).

Entangolensin B (2) was afforded as a white amorphous powder. Its molecular formula of C₂₇H₃₄O₇ was established by the HRESIMS ion at m/z 471.2374 [M + H]⁺ (calcd for C₂₇H₃₅O₇, 471.2377). Its IR absorptions implied the presence of hydroxyl (3474 cm⁻¹) and carbonyl groups (1736 cm⁻¹). The ¹H NMR spectrum (Table 1) of 2 demonstrated four tertiary methyl groups (δ_H 1.34, 1.17, 1.12, 1.02), one olefinic methyl group (δ_H 2.13), one methoxyl group (δ_H 3.70), two oxymethine protons (δ_H 4.95, 5.03), a pair of AB doublets at δ_H 6.70 and 5.95, and a β-substituted furan ring (δ_H 7.41, 7.40, 6.32). The ¹³C NMR spectrum (Table 1), when combined with the HSQC experiment, indicated the existence of a ketone carbonyl (δ_C 203.2), two ester carbonyls (δ_C 173.9, 174.7), and two unsaturated bonds (δ_C 155.8, 137.8, 135.0, 124.7). The aforementioned NMR data suggested that compound 2 possessed a typical B,D-seco limonoid skeleton like thaimoluccensin A.²⁶ In the HMBC spectrum, the correlations from H-9 (δ_H 2.33) and H₃-30 (δ_H 2.13) to the olefinic carbons at δ_C 137.8 (C-8) and δ_C 135.0 (C-14) indicated the presence of a Δ^8,14 double bond. In addition, the key HMBC correlations from H-15 (δ_H 4.95) to C-8, C-14, and C-16 (δ_C 174.7) suggested an additional hydroxy group at C-15 (δ_C 66.6). Thus, the planar structure of 2 was determined.

The same relative configuration of 2 as thaimoluccensin A was suggested from the ROESY experiment. Additionally, the α-orientation of OH-15 was determined according to the similar chemical shifts of H-15 (δ_H 4.95, δ_H 5.02) and C-15 (δ_C 66.6, δ_C 65.7) to those of swietmanin F,²⁷ because no valuable ROESY peaks were observed for determining the configuration of OH-15. The ECD spectrum of 2 was dominated by a positive Cotton effect at λ_max 241 nm (Δε +7.75) and a negative Cotton effect at λ_max 190 nm (Δε −11.32) due to the transition interaction between two different chromophores of the α,β-unsaturated carbonyl and the furan ring, indicating a positive chirality for 2 ²⁸ (see Fig. S1 in the ESI†). The absolute configuration of 2 was thus assigned as shown. Therefore, the structure of 2 was determined and named entangolensin B.

Entangolensins C (3) and D (4) were a pair of isomers, possessing the same molecular formula of C₂₇H₃₄O₈, as determined from the HRESIMS [M + Na]⁺ ions (3, m/z 509.2144; 4, m/z 509.2143). Analysis of their NMR data (Table 1) and the HMBC correlations revealed that both of them possessed the same B,D-seco limonoid skeleton as methyl angolensate (18)²⁹ and shared the same planar structure. However, an additional oxymethine signal (δ_H 4.54 in 3, δ_H 4.48 in 4) was observed, suggesting that compounds 3 and 4 were oxygenated derivatives of 18. The same HMBC correlations (Fig. 4) from H-11 to C-12, C-13, C-8 and C-10 demonstrated an additional hydroxy group to be located at C-11 for both of these compounds. The relative configurations of 3 and 4 were determined using a ROESY experiment. The ROESY correlations of H-1/H₃-29 and H-1/H-15b (Fig. 4) indicated that H-1 in 3 adopted an α-orientation, like that in 18. On the contrary, H-1 was β-oriented in 4 based on the strong ROESY cross peaks of H-1/12b and H-12b/17, which forms a rare representative of a B,D-seco limonoid with a C-1/14 oxygen bridge. Moreover, the β-orientation of H-11 in 3 and α-orientation of H-11 in 4 were inferred from the key ROESY cross peaks of H-11/H-5 and H-11/H₃-18, respectively. Accordingly, the structures of 3 and 4, entangolensins C and D, were concluded as illustrated.

Fig. 4 Key HMBC (¹H¹³C) and ROESY (¹H¹H) correlations of compound 3.

Entangolensins E (5) and F (6) were also a pair of isomers, possessing the same molecular formula of C₂₈H₃₆O₉ as determined by HRESIMS. The ¹H and ¹³C NMR spectroscopic data (Table 1) of 5 and 6 were closely related to those of moluccensins O and N (20),³⁰ respectively, with the exception of the observed additional NMR signals (δ_H 3.60, δ_C 57.5 in 5; δ_H 3.54, δ_C 56.5 in 6) due to a methoxy group. A comparison of their 1D NMR data revealed the main differences at C-21 (Δδ_C-21 = +5.9 ppm) in 5 and C-23 (Δδ_C-23 = +4.6 ppm) in 6. These differences suggested that the methoxy groups were located at C-21 in 5 and C-23 in 6, which was further confirmed by the HMBC experiment. The relative configurations of 5 and 6 were identical to those of moluccensins O and N according to the similar ROESY correlations. In addition, the β-orientation of OCH₃-21 in 5 was confirmed by the ROESY correlations of H-21/H₃-18 and H-21/H-12b. However, the available evidence was insufficient to determine the configuration of OCH₃-23 in 6.^31,32 Therefore, the structures of 5 and 6 were determined, featuring a 21-methoxybutenolide and 23-methoxybutenolide ring in the C-17 side-chain, respectively. In fact, the NMR spectroscopic data of 6 were in good agreement with those reported for entangosin,¹⁵ isolated from the root bark of the same species, E. angolense. Hence, the structure of entangosin was corrected as 6.

Entangolensin G (7) was isolated as a white amorphous powder, and had a molecular formula of C₂₉H₄₀O₉ as determined from the HRESIMS ion at m/z 550.3009 [M + NH₄]⁺ (calcd for C₂₉H₄₄NO₉ 550.3011), 16 mass units greater than that of compound 5. Comparison of the ¹H and ¹³C NMR data (Table 2) with 5 suggested that 7 differed only in the replacement of the ester carbonyl (δ_C 168.5, OC̲O-23) in 5 by a methoxymethine (δ_H 5.67, δ_C 106.5, CH-23; δ_H 3.36, δ_C 53.4, OC̲H₃-23). The ROESY correlations of H-17/OCH̲₃-21, H-17/H-12a and H-21/H-12b indicated that H-17 and OCH₃-21 were oriented on the same face of the tetrahydrofuran ring and were assigned as β-oriented, while the absence of any ROESY correlation between H-23 and H-21 led to the assumption that these protons were on opposite sides of the tetrahydrofuran ring. Therefore, the structure of 7, entangolensin G, was proposed as shown.

Table 2 ¹H NMR (500 MHz) and ¹³C NMR (125 MHz) data of compounds 7–10^a and 11^b

Position	7		8		9		10		11
	δC	δH (J in Hz)	δC	δH (J in Hz)	δC	δH (J in Hz)	δC	δH (J in Hz)	δC	δH (J in Hz)
a Measured in CDCl3.b Measured in methanol-d4.c Signals were overlapped.
1	77.3	3.48, dd (5.0, 5.0)	77.3	3.48, dd (5.9, 3.5)	77.7	3.53^c	77.1	3.45, dd (6.2, 4.5)	79.1	3.63, dd (6.0, 3.7)
2a	39.6	2.84^c	39.4	2.87, dd (14.5, 5.9)	39.3	2.94, dd (14.6, 5.8)	39.5	2.85, dd (14.3, 6.2)	40.5	3.08, dd (14.5, 6.0)
2b		2.49^c		2.41, dd (14.5, 3.5)		2.38, d (14.6)		2.51, dd (14.3, 4.5)		2.43, dd (14.5, 3.7)
3	212.4		212.3		212.4		212.5		215.7
4	48.1		48.1		48.3		48.1		45.0
5	43.1	2.86^c	42.7	2.87^c	42.8	2.86^c	43.0	2.81^c	44.4	2.92, d (10.1)
6a	32.9	2.60, dd (16.2, 10.3)	32.9	2.59, dd (16.5, 10.3)	32.8	2.62, dd (16.7, 10.3)	32.8	2.56, dd (16.3, 10.3)	33.7	2.65, dd (16.8, 10.1)
6b		2.24, d (16.2)		2.24, d (16.5)		2.24, d (16.7)		2.24, d (16.3)		2.41, d (16.8)
7	173.9		173.8		173.8		173.9		175.7
8	145.7		145.3		145.0		145.0		147.0
9	49.9	2.18, d (4.7)	50.2	2.14^c	50.4	2.18^c	49.7	2.16, d (4.5)	51.5	2.32, d (4.0)
10	44.2		44.0		44.0		44.3		49.0
11a	24.0	2.25^c	24.2	2.12, m	24.3	2.16^c	23.8	2.23^c	25.0	2.26, m
11b		1.63, m		1.62, m		1.63, m		1.65, m		1.61, m
12a	29.6	1.97, dd (12.8, 5.1)	28.0	2.12, m	28.1	2.17^c	30.5	1.89, td (13.4, 5.2)	30.5	2.17, td (14.0, 5.1)
12b		1.25, td (12.8, 5.2)		1.32, dd (12.4, 4.9)		1.31, dd (11.9, 4.5)		1.37, dd (12.7, 4.8)		0.98, m
13	42.0		42.9		43.2		41.8		43.1
14	80.0		80.7		80.9		80.8		81.6
15a	33.8	2.84, d (18.0)	34.5	2.84, d (18.7)	34.7	2.89, d (18.6)	33.7	2.82, d (18.0)	34.7	3.14, d (18.3)
15b		2.53, d (18.0)		2.45, d (18.7)		2.48, d (18.6)		2.53, d (18.0)		2.64, d (18.3)
16	169.6		171.2		170.6		169.4		171.8
17	80.2	5.36, s	84.3	5.24, s	81.8	5.12, s	82.5	5.09, s	78.7	5.66, s
18	14.0	0.88, s	15.0	1.14, s	14.8	1.15, s	15.0	1.10, s	14.2	0.97, s
19	21.7	0.91, s	21.8	0.93, s	22.0	0.99, s	21.7	0.89, s	21.9	1.01, s
20	141.8		82.1		82.2		79.6		146.2
21	107.6	5.54, s	107.9	4.78, s	107.2	5.12, s	109.4	5.19, s	172.2
22a	129.3	6.06, s	80.6	4.49, d (3.5)	42.1	3.01, d (17.7)	37.0	2.72, d (16.3)	133.3	6.70, d (1.1)
22b						2.68, d (17.7)		2.43, d (16.3)
23	106.5	5.67, s	109.2	4.85, d (3.5)	174.1		174.1		172.2
28	26.4	1.02, s	25.8	0.97, s	25.3	0.96, s	26.5	1.03, s	26.1	1.01, s
29	21.5	1.17, s	21.6	1.15, s	21.7	1.18, s	21.5	1.17, s	21.9	1.21, s
30a	111.8	5.15, s	112.1	5.16, s	112.5	5.20, s	112.6	5.18, s	112.8	5.23, s
30b		4.88, s		4.85, s		4.91, s		4.87, s		5.04, s
OMe-7	52.2	3.71, s	52.1	3.69, s	52.2	3.70, s	52.3	3.72, s	52.4	3.69, s
OMe-21	54.9	3.44, s	54.6	3.39, s	56.3	3.54, s	57.2	3.56, s
OMe-23	53.4	3.36, s	56.4	3.45, s

---

Entangolensin H (8) gave a molecular formula of C₂₉H₄₂O₁₁, as established from the [M + NH₄]⁺ peak at m/z 584.3063 in the HRESIMS. The 1D NMR data (Table 2) of 8 closely resembled those of 7, except for the absence of signals for a Δ^20,22 double bond (CH-22, δ_H 6.06, δ_C 129.3; C-20, δ_C 141.8) and the presence of signals for an oxymethine (δ_H 4.49, δ_C 80.6, CH-22) and an oxygenated quaternary carbon (δ_C 82.1, C-20). These observations suggested that 8 should be a 20,22-dihydroxy derivative of 7, which was further confirmed by the HMBC correlations from H-17 (δ_H 5.24) to C-20, C-21 (δ_C 107.9) and C-22. The ROESY correlations between H-17/OCH̲₃-21, H-17/H-22 and H-22/H-23 revealed that these protons were oriented on the same face of the tetrahydrofuran ring and were assigned as β-oriented. In turn, the ROESY correlations between H-21/H₃-18 and H-21/H-12b suggested an α-orientation of these protons. Thus, structure 8 was proposed as shown and named entangolensin H.

Entangolensins I (9) and J (10) were assigned the same molecular formula C₂₈H₃₈O₁₀ as established from the same [M + NH₄]⁺ peak at m/z 552.2805 in the HRESIMS, 18 mass units more than that of 5. The NMR spectroscopic data (Table 2) of 9 and 10 revealed that they shared the same planar structure and differed from 5 only in the peaks derived from the furan ring. Compared with those of 5, the absence of signals for the Δ^20,21 double bond (CH-22, δ_H 6.24, δ_C 122.3; C-20, δ_C 162.2), additional signals for methylene [(CH₂-22, δ_H 3.01 and 2.68; δ_C 42.1) in 9; (CH₂-22, δ_H 2.72 and 2.43; δ_C 37.0) in 10] and an oxygenated quaternary carbon [(C-20, δ_C 82.2) in 9; (C-20, δ_C 79.6) in 10] indicated that 9 and 10 were derivatives of 5 with H₂O added at Δ^20,21. The HMBC correlations from H-17 to C-20, C-21 and C-22, and from H-21 to C-23 and C-22 also confirmed this inference. Thus, the planar structures of these two compounds were determined. In comparison to 9, a downfield shift of C-21 and upfield shifts of C-20 and C-22 were observed in 10, suggesting that 10 may be a C-21 epimer of 9. The ROESY cross-peak of OCH̲₃-21/H-17 was present in 9 and absent in 10, which suggested that the OCH₃-21 was β-oriented in 9 and α-oriented in 10. Therefore, the structures of 9 and 10 were established as shown in Fig. 1.

Entangolensin K (11) was assigned a molecular formula of C₂₇H₃₃NO₈, established on the basis of the HRESIMS ion at m/z 522.2095 [M + Na]⁺ (calcd for C₂₇H₃₃NO₈Na 522.2098). The ¹H and ¹³C NMR spectroscopic data (Table 2) of 11 showed close similarity to those of methyl angolensate (18), except for the resonances of the C-17 side chain. The obvious two carbonyls (δ_C 172.2, C-21; δ_C 172.2, C-23), two olefinic carbons (δ_C 146.2, C-20; δ_C 133.3, C-22) and the additional N atom in its molecular formula indicated that the most striking feature of 11 was the β-substituted furan ring being replaced by a maleimide ring,³³ which was confirmed by the HMBC correlations from a singlet proton signal at δ_H 6.70 (H-22) to C-17, C-21 and C-23. The relative configuration of 11 was identical to that of methyl angolensate according to its ROESY experiment. Therefore, the structure of 11 (Fig. 1) was finally established.

The ECD spectra of 3 and 7–11 were similar (see Fig. S2 in the ESI†). Analysis of their structural features and similar ECD spectra suggested that the identical negative Cotton effect resulted from the transition interaction between the ketone moiety (CO-3) and the terminal double bond (Δ^8,30). Therefore, we applied the ECD exciton chirality method to resolve their absolute configurations. Taking compound 7 for example, the ECD spectrum of 7 exhibited negative chirality resulting from the exciton coupling between the ketone moiety at 225 nm (Δε −18.42) and the terminal double bond at 190 nm (Δε +16.72)³⁴ (Fig. 5), which indicated the counterclockwise arrangement of two chromophores in space. Thus, the absolute configuration of 7 was assigned as shown. The same ECD Cotton effects of compounds 3 and 8–11 compared to 7 indicated that their absolute configurations were consistent with those of 7.

Fig. 5 ECD and UV spectra of 7 (in MeCN). The bold lines denote the electric transition dipole of the chromophores for 7.

Entangolensin L (12) was isolated as a white amorphous powder, and had a molecular formula of C₂₈H₃₂O₈ as determined by the HRESIMS ion at m/z 519.1985 [M + Na]⁺ (calcd for C₂₈H₃₂O₈Na, 519.1989). The ¹H NMR spectrum of 12 showed the presence of five tertiary methyl groups (δ_H 1.49, 1.47, 1.16, 1.14, 1.08), one acetate methyl group (δ_H 2.18), three oxymethine protons (δ_H 5.87, 5.45, 3.82), and a β-substituted furan ring (δ_H 7.41, 7.39 and 6.34). The aforementioned characteristic signals suggested that compound 12 possessed a gedunin-type limonoid skeleton.³⁵ The NMR spectroscopic data (Table 3) of 12 showed that its structure was closely related to that of 5-hydroxy-7-deacetoxy-7-oxogedunin,¹⁵ except for the absence of a hydroxy group at C-5 (δ_C 62.9) and the presence of an additional acetyl group at C-11 (δ_C 66.6), which was further confirmed by the HMBC correlations from H-11 (δ_H 5.87) to C-9 (δ_C 48.7), C-10 (δ_C 40.3), C-13 (δ_C 36.6), and OC̲OCH₃ (δ_C 170.2). The ROESY spectrum showed that the relative configuration of 12 was the same as that of gedunin-type compounds, with an α-oriented OAc-11 deduced from the correlation of H-11/H₃-30. The positive chirality resulting from the exciton coupling between the α,β-unsaturated carbonyl at λ_max 224 nm (Δε +5.15) and the furan ring at λ_max 208 nm (Δε −0.98) indicated the clockwise arrangement of two chromophores in space²⁸ (see Fig. S3 in the ESI†). The absolute configuration of 12 was thus assigned as shown. Therefore, the structure of 12, entangolensin L, was established as depicted.

Table 3 ¹H NMR (500 MHz) and ¹³C NMR (125 MHz) data of compounds 12–16^a

Position	12		13		14		15		16
	δC	δH (J in Hz)	δC	δH (J in Hz)	δC	δH (J in Hz)	δC	δH (J in Hz)	δC	δH (J in Hz)
a Measured in CDCl3.b Signals were overlapped.
1a	154.6	7.25, d (10.2)	33.6	1.61, m	32.8	1.54, m	156.0	7.23, d (10.3)	74.4	4.40, brs
1b				1.41, m		1.40, m
2a	127.0	5.97, d (10.2)	25.2	2.04, m	25.2	1.99, m	126.2	5.90, d (10.3)	25.0	2.23^b
2b				1.67, m		1.62, m				2.21^b
3	202.7		75.3	3.48, brs	75.5	3.43, brs	203.1		75.9	4.75, brs
4	45.3		38.4		37.0		40.7		36.4
5	55.0	2.12, dd (14.4, 3.2)	50.9	1.83, dd (14.4, 3.0)	42.7	1.81^b	47.8	2.15^b	36.3	2.38, dd (12.6, 3.3)
6a	36.9	3.01, dd (14.4, 14.4)	36.6	2.77, dd (14.4, 14.4)	23.5	1.56^b	24.8	2.01^b	22.2	1.81^b
6b		2.47, dd (14.4, 3.2)		2.32, dd (14.4, 3.0)		1.77^b		1.83, brd (15.0)		1.81^b
7	206.9		208.7		75.7	4.55, brs	75.0	4.75, s	74.9	5.16, brs
8	52.6		52.5		42.1		42.2		42.0
9	48.7	2.20, s	53.2	2.00, s	48.1	2.32, d (4.5)	43.3	2.59^b	38.7	3.09, d (10.5)
10	40.3		38.3		38.8		44.3		41.2
11a	66.6	5.87, d (7.5)	67.2	5.66, d (7.4)	67.4	5.55, ddd (9.9, 5.5, 4.5)	67.5	5.83, m	70.0	5.39, m
11b
12a	41.0	1.90, dd (15.7, 7.5)	41.3	1.80, dd (15.9, 7.4)	37.1	2.27, dd (14.3, 9.9)	40.0	2.46, dd (14.2, 9.5)	44.7	2.03^b
12b		1.76, d (15.7)		1.65^b		1.44, dd (14.3, 5.5)		1.76, dd (14.2, 3.8)		1.44, d (13.9)
13	36.6		36.5		38.0		41.8		46.0
14	64.2		64.4		69.0		70.3		158.1
15	53.9	3.82, s	53.8	3.78, s	55.2	3.55, s	60.1	3.60, s	118.9	5.33, d (1.7)
16a	166.5		167.0		167.4		76.3	4.14, brs	34.8	2.10, m
16b										2.13, m
17	78.2	5.45, s	78.4	5.42, s	78.5	5.58, s	48.4	2.58^b	57.9	1.70, m
18	20.0	1.08, s	20.1	1.08, s	17.6	1.26, s	22.8	0.96, s	20.0	1.12, s
19	21.3	1.49, s	17.9	1.24, s	18.0	1.19, s	22.1	1.40, s	16.3	1.07, s
20	119.9		120.1		120.3		120.9		37.4	2.69, m
21a	141.3	7.41, s	141.2	7.39, s	141.4	7.38, s	140.5	7.27, s	72.4	4.37, t (9.4)
21b										3.85, t (9.4)
22a	109.8	6.34, s	109.9	6.33, s	110.0	6.32, s	111.0	6.20, s	34.0	2.50, dd (17.1, 7.9)
22b										2.18^b
23	143.5	7.39, t-like	143.3	7.37, t (1.6)	143.4	7.39, s	143.6	7.40, s	176.3
28	27.1	1.16, s	27.8	0.92, s	28.2	0.84, s	27.0	1.06, s	27.9	0.80, s
29	20.8	1.14, s	21.1	0.90, s	21.9	0.83, s	21.3	1.06, s	22.3	0.94, s
30	19.8	1.47, s	19.5	1.38, s	20.2	1.28, s	21.7	1.40, s	28.3	1.17, s
1-OC̲OCH3									169.8
1-OCOC̲H3									21.1	2.08, s
3-OC̲OCH3									170.1
3-OCOC̲H3									21.6	2.01, s
7-OC̲OCH3					169.9		169.6		169.8
7-OCOC̲H3					21.7	2.09, s	21.5	2.05, s	21.2	2.02, s
11-OC̲OCH3	170.2		170.2		170.1		170.4		171.2
11-OCOC̲H3	21.6	2.18, s	21.6	2.13, s	21.3	2.09, s	21.7	2.13, s	22.2	2.00, s

---

Entangolensins M (13) and N (14) showed molecular formulas of C₂₈H₃₆O₈ and C₃₀H₄₀O₉ as determined from the HRESIMS ions at m/z 518.2745 [M + NH₄]⁺ and m/z 562.3016 [M + NH₄]⁺, respectively. The similar NMR data of 12–14 (Table 3) indicated that 13 and 14 were also gedunin-type limonoids. When compared with 12, 13 displayed no signal due to the 1-en-3-one system of the A-ring; in contrast, the presence of a hydroxyl group at C-3 (δ_C 75.3) was recognized, which was confirmed by the corresponding HMBC correlations. The α-orientation of OH-3 was assigned from the ROESY correlations of H-3/H-1a and H-1a/H₃-19. Therefore, the structure of 13 was demonstrated as shown. The NMR data of 14 were similar to those of 13, with the only difference being in the replacement of the carbonyl at δ_C 208.7 in 13 by an acetoxy group (OCOCH₃, δ_H 2.09; δ_C 169.9, 21.7) at δ_C 75.7 (C-7). In addition, the differences in the chemical shifts around C-11 suggested that 14 may be a C-11 epimer of 13. Finally, the ROESY correlations between H-11/H₃-18 and H-7/H₃-30 suggested a β-orientation of OAc-11 and an α-orientation of OAc-7. Thus, the structure of 14 was established as shown.

Entangolensin O (15) was assigned the molecular formula C₃₀H₃₈O₈, as established from the [M + H]⁺ peak at m/z 527.2636. The ¹H and ¹³C NMR data (Table 3) resembled those of known compound 14β,15β-epoxyazadirone,¹⁵ except for the presence of additional signals due to an acetyl group (δ_C 170.4, 21.7 and δ_H 2.13) at C-11 (δ_C 67.5) and a hydroxy group at C-16 (δ_C 76.3), further confirmed by the key HMBC correlations from H-11 to C-12 and OC̲OCH₃-11, and from H-16 to C-17 and C-20, respectively. The ROESY correlations of H-9/H-11, H-11/H₃-18 and H₃-18/H-16 revealed an α-orientation of these protons and a β-orientation of OH-16 and OAc-11. The absolute configuration of 15 was also determined (see Fig. S4 in the ESI†) by the ECD exciton chirality method to be the same as that of 12, due to their similar chromophores and ECD Cotton effects. Therefore, the structure of 15 was demonstrated as shown.

The molecular formula of 16 was determined to be C₃₄H₄₈O₁₀ on the basis of the HRESIMS ion at m/z 639.3136 [M + Na]⁺. The ¹H and ¹³C NMR data (Table 3) of 16 indicated that its structure was closely related to that of the known compound meliatoosenin B,³⁶ with the only difference being the presence of four acetyl groups located at C-1, C-3, C-7 and C-11, respectively. This inference was confirmed by their corresponding HMBC correlations. The relative configuration of 16 was determined from the ROESY spectrum. The ROESY correlations of H₃-18/H-20, H₃-18/H-9, H-9/H-5 and H-5/H₃-28 suggested an α-orientation of these protons and groups. Also, correlations between H-1/H₃-19, H-3/H₃-19, H-1/H-11, H-11/H₃-30, H₃-30/H-7 and H-17/H-12b suggested a β-orientation of these protons and groups. Therefore, the structure of 16 was demonstrated as shown.

Four known compounds were identified as spicatin (17),³⁷ methyl angolensate (18),²⁹ 8α-hydroxy-8,30-dihydroangolensate (19),³⁸ and moluccensin N (20)³⁰ by comparison of their spectroscopic data (MS and NMR) with reported values.

Some research papers related to the biosynthetic pathways of limonoids have been published, which focused on the conversion of multiple core frameworks.³ The previous research showed that most Meliaceae plants tended to produce limonoids with one dominant framework.^39,40 However, a series of limonoids with diverse carbon cores were discovered in our current research, which encouraged us to construct a plausible biosynthetic pathway. As described in Scheme 1, the Baeyer–Villiger oxidation of the B- and D-ring is the key transformation from the ring-intact limonoid to the B,D-seco limonoid. The electropositivity of the protonated 14,15-epoxide transfers to C-14 to afford a stable tertiary carbenium ion, which further rearranges to form 2. In this pathway, the connection of C-2/30 based on the intermediate with electropositive C-30 forms the common mexicanolide-type limonoid, and the cleavage of C-8/9 is the key step in the formation of structure 1. For another branch, the protonated 14,15-epoxide attacks the electronegative C-1 from both sides of the A-ring to produce 3 and 4, showing two configurations of the C-1/14 oxygen bridge. Further hydrolyzation and oxidation of the conjugated double bond (C20-23) of 3 could produce the diverse modified furan rings in 5–11. This plausible biosynthesis pathway suggests that the biosynthetic processes of limonoids in E. angolense might be extremely active and more novel limonoids could exist in this plant.

Scheme 1 The possible biogenetic pathway of the compounds from E. angolense.

All the isolates except 1–3 were evaluated for their cytotoxicities against HepG2 and MCF-7 cancer cell lines with doxorubicin as the positive control. The results (Table 4) revealed that compounds 6, 12, 15, and 17 showed moderate cytotoxicities. These isolates were further tested for their inhibitory effects on NO production of LPS-activated RAW 264.7 macrophages with N-monomethyl-L-arginine as the positive control (IC₅₀ = 32.55 ± 2.15 μM). The results (Table 4) revealed that compounds 6, 11 and 17 exhibited significant NO inhibitory activities with IC₅₀ values of 1.75 ± 0.09 μM, 7.94 ± 1.53 μM and 4.63 ± 1.25 μM, respectively. In addition, an MTT assay revealed no obvious cytotoxic effects (greater than 90% cell survival) in LPS-activated RAW 264.7 macrophages treated with the compounds at concentrations up to 100 μM.

Table 4 Cytotoxicity and NO inhibition activities of compounds 4–20^a,b

Compounds	Cytotoxicity		NO inhibition
	HepG2	MCF-7	RAW 264.7 macrophages
a Values were expressed as the means ± SD based on three independent experiments.b Compounds 4–5, 7–10, 13–14, 16 and 18–19 were inactive (IC50 > 50 μM).c Positive controls.
6	13.19 ± 1.33	14.06 ± 1.29	1.75 ± 0.09
11	>50	>50	7.94 ± 1.53
12	20.39 ± 1.54	17.20 ± 1.67	>50
15	21.00 ± 2.09	36.93 ± 3.12	>50
17	16.84 ± 3.52	20.28 ± 1.98	4.63 ± 1.25
20	>50	>50	11.05 ± 2.49
Doxorubicin^c	3.25 ± 0.32	1.92 ± 1.40
L-NMMA^c			32.55 ± 2.15

---

Conclusions

Entangolensins A–P (1–16), sixteen new limonoids with diverse frameworks, were obtained from the stem bark of Entandrophragma angolense collected from Ghana. They were further evaluated for their cytotoxicities and NO inhibitory activities, and certain compounds exhibited moderate cytotoxicities (13.19–36.93 μM) against HepG2 and MCF-7 cell lines and significant NO inhibitory activities (1.75–11.05 μM) in LPS-activated RAW 264.7 macrophages. These results revealed that more novel limonoids with significant bioactivities could exist in this plant, and it is worth paying more attention to the Meliaceae plants distributed in Africa.

Experimental

General experimental procedures

Optical rotations were measured on a JASCO P-1020 polarimeter (Jasco, Tokyo, Japan). ECD spectra were obtained on a JASCO J-810 spectropolarimeter (Jasco, Tokyo, Japan). UV and IR were recorded on a Shimadzu UV-2450 spectrophotometer (Shimadzu, Tokyo, Japan) and Bruker Tensor 27 spectrometer (Bruker, Karlsruhe, Germany), respectively. NMR spectra were acquired at 500 MHz (¹H) and 125 MHz (¹³C) on a Bruker Avance III NMR spectrometer (Bruker, Karlsruhe, Germany) with tetramethylsilane as the internal standard. HRESI mass spectra were acquired on an Agilent 6520B UPLC-Q-TOF mass spectrometer (Agilent Technologies, Santa Clara, CA, USA). Silica gel (Qingdao Haiyang Chemical Co., Ltd., Qingdao, People’s Republic of China), MCI gel (Mitsubishi Chemical Corp., Tokyo, Japan), and RP-C18 (40–63 μm, FuJi, Japan) were used for column chromatography. Semi-preparative HPLC was performed on a Shimadzu LC-6A instrument with a SPD-20A detector using a shim-pack RP-C₁₈ column (20 × 200 mm, i.d.). Analytical HPLC was carried out on an Agilent 1200 Series instrument with a DAD detector using a shim-pack VP-ODS column (250 × 4.6 mm, i.d.).

Plant material

The stem bark of Endrophragma angolense (Meliaceae) was collected from the Brong Ahafo Region of Ghana, in August 2013. The plant material was identified by Professor Mian Zhang of the Research Department of Pharmacognosy, China Pharmaceutical University. A voucher specimen (no. 2015-MEA) was deposited in the Department of Natural Medicinal Chemistry, China Pharmaceutical University.

Extraction and isolation

The dried and powdered stem bark of E. angolense (892 g) was exhaustively extracted with 95% ethanol (4 L) three times at room temperature. The EtOH extract was concentrated under reduced pressure to obtain a residue (114 g). Upon addition of EtOAc (3 × 300 mL), two phases were obtained, and the insoluble part was removed by filtration. The EtOAc-soluble part was suspended in MeOH : H₂O (1 : 1) and partitioned with CHCl₃ (3 × 300 mL). The CHCl₃ extract (34 g) was crystallized in CH₂Cl₂–MeOH (1 : 1), from which the most abundant component methyl angolensate (18) (12 g) was obtained. The rest of the CH₂Cl₂ part (22 g) was loaded onto a silica gel column and eluted with petroleum–acetone mixtures in a step gradient (10 : 1 to 1 : 1) to afford five fractions, A–E. Fraction C (2.2 g) was run on an ODS column using a step gradient of MeOH–H₂O (45 : 55 to 100 : 0) to afford five sub-fractions (C1–C5). Fraction C2 (231 mg) was further fractioned by middle-pressure chromatography (MeOH–H₂O, 60 : 40, v/v, 10 mL min⁻¹) to give four sub-fractions (C2a–C2d). Fraction C2b was subjected to semi-preparative HPLC (MeCN–H₂O, 50 : 50, v/v, 10 mL min⁻¹) to yield 2 (1.6 mg), 12 (6.1 mg) and 19 (3 mg). Using the same purification procedures, fraction C2c yielded 7 (6.0 mg), 15 (3.2 mg) and 1 (1.3 mg). Fraction C3 (57 mg) was subjected to semi-preparative HPLC (MeCN–H₂O, 47 : 53, v/v, 10 mL min⁻¹) to obtain 13 (16.4 mg) and 16 (8.4 mg). Fraction C4 (83 mg) was subjected to semi-preparative HPLC (MeCN–H₂O, 47 : 53, v/v, 10 mL min⁻¹) to obtain 14 (3 mg) and 17 (50 mg).

Fraction D (2.0 g) was separated by ODS (MeOH–H₂O, 30 : 70 to 100 : 0) to afford five subfractions (D1–D5). Fraction D1 (102.0 mg) was separated by semi-preparative HPLC (MeOH–H₂O, 50 : 50, v/v, 10 mL min⁻¹) to yield 3 (1.4 mg). Fraction D3 (151.0 mg) was separated by MPLC (MeOH–H₂O, 55 : 45, v/v, 10 mL min⁻¹) and further chromatographed over semi-preparative HPLC (MeCN–H₂O, 40 : 60, v/v, 10 mL min⁻¹) to obtain 10 (9 mg), 11 (3.7 mg) and 20 (10.0 mg). Using the same purification procedures, fraction D4 (301.0 mg) yielded 4 (8 mg), 5 (2.5 mg), 6 (80.0 mg) and 9 (5.4 mg). Fraction D5 (101.0 mg) was separated by semi-preparative HPLC (MeCN–H₂O, 35 : 65, v/v, 10 mL min⁻¹) and further purified by semi-preparative HPLC (MeCN–H₂O, 40 : 60, v/v, 10 mL min⁻¹) to obtain 8 (10.1 mg).

Entangolensin A (1). White, amorphous powder; [α]²⁵_D −130.8 (c 0.12, MeOH); UV (MeOH) λ_max (log ε) 278 (3.65), 251 (3.50), 212 (3.72) nm; ECD (CH₃CN, Δε) 207 (−11.13), 237 (+12.38), 258 (+6.23), 283 (+16.98) nm; IR (KBr) ν_max 3423, 2975, 2313, 1714, 1385, 1265, 1170, 1073, 1028, 876, 419 cm⁻¹; ¹H and ¹³C NMR data, see Table 1; ESIMS m/z 486.1 [M + NH₄]⁺ (100); HRESIMS m/z 491.2037 [M + Na]⁺ (calcd for C₂₇H₃₂O₇Na, 491.2040).

Entangolensin B (2). White, amorphous powder; [α]²⁵_D +33.0 (c 0.11, MeOH); UV (MeOH) λ_max (log ε) 208 (3.89) nm; ECD (CH₃CN, Δε) 190 (−11.32), 241 (+7.75) nm; IR (KBr) ν_max 3474, 2981, 1736, 1678, 1462, 1384, 1278, 1246, 1181, 1137, 1050, 876, 818, 605 cm⁻¹; ¹H and ¹³C NMR data, see Table 1; ESIMS m/z 471.2 [M + H]⁺ (100); HRESIMS m/z 471.2374 [M + H]⁺ (calcd for C₂₇H₃₅O₇, 471.2377).

Entangolensin C (3). White, amorphous powder; [α]²⁵_D −42.8 (c 0.08, MeOH); UV (MeOH) λ_max (log ε) 205 (3.76) nm; ECD (CH₃CN, Δε) 190 (+33.10), 205 (−17.08) nm; IR (KBr) ν_max 3467, 2972, 2313, 1735, 1387, 1253, 1173, 1056, 912, 875 cm⁻¹; ¹H and ¹³C NMR data, see Table 1; ESIMS m/z 504.2 [M + NH₄]⁺ (100); HRESIMS m/z 509.2144 [M + Na]⁺ (calcd for C₂₇H₃₄O₈Na, 509.2146).

Entangolensin D (4). White, amorphous powder; [α]²⁵_D −52.8 (c 0.12, MeOH); UV (MeOH) λ_max (log ε) 206 (3.49) nm; ECD (CH₃CN, Δε) 190 (+11.36) nm; IR (KBr) ν_max 3421, 2975, 1734, 1391, 1272, 1059, 1027, 876, 602, 419 cm⁻¹; ¹H and ¹³C NMR data, see Table 1; ESIMS m/z 487.1 [M + H]⁺ (100); HRESIMS m/z 509.2143 [M + Na]⁺ (calcd for C₂₇H₃₄O₈Na, 509.2146).

Entangolensin E (5). White, amorphous powder; [α]²⁵_D −16.2 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 208 (3.87) nm; ECD (CH₃CN, Δε) 201 (−12.78), 248 (+5.79) nm; IR (KBr) ν_max 3420, 2973, 1746, 1648, 1389, 1259, 1127, 1041, 965, 901 cm⁻¹; ¹H and ¹³C NMR data, see Table 1; ESIMS m/z 534.2 [M + NH₄]⁺ (100); HRESIMS m/z 534.2700 [M + NH₄]⁺ (calcd for C₂₈H₄₀NO₉, 534.2698).

Entangolensin F (6). White, amorphous powder; [α]²⁵_D −11.8 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 210 (4.18) nm; ¹H and ¹³C NMR data, see Table 1; ESIMS m/z 534.2 [M + NH₄]⁺ (100); HRESIMS m/z 517.2429 [M + H]⁺ (calcd for C₂₈H₃₇O₉, 517.2432).

Entangolensin G (7). White, amorphous powder; [α]²⁵_D −44.0 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 204 (3.47) nm; ECD (CH₃CN, Δε) 190 (+16.72), 205 (−18.42) nm; IR (KBr) ν_max 3443, 2969, 2313, 1781, 1731, 1469, 1379, 1268, 1160, 1054, 1022, 419 cm⁻¹; ¹H and ¹³C NMR data, see Table 2; ESIMS m/z 550.2 [M + NH₄]⁺ (100); HRESIMS m/z 550.3009 [M + NH₄]⁺ (calcd for C₂₉H₄₄NO₉, 550.3011).

Entangolensin H (8). White, amorphous powder; [α]²⁵_D −71.4 (c 0.06, MeOH); UV (MeOH) λ_max (log ε) 204 (3.76) nm; ECD (CH₃CN, Δε) 190 (+10.31), 205 (−15.58) nm; IR (KBr) ν_max 3444, 2972, 2377, 2350, 2312, 1715, 1643, 1386, 1269, 1074, 993 cm⁻¹; ¹H and ¹³C NMR data, see Table 2; ESIMS m/z 584.3 [M + NH₄]⁺ (100); HRESIMS m/z 584.3069 [M + NH₄]⁺ (calcd for C₂₉H₄₆NO₁₁, 584.3065).

Entangolensin I (9). White, amorphous powder; [α]²⁵_D −54.4 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 204 (3.56) nm; ECD (CH₃CN, Δε) 190 (+14.75), 205 (−22.82) nm; IR (KBr) ν_max 3445, 2975, 1796, 1734, 1457, 1390, 1269, 1198, 1139, 1106, 1055, 961 cm⁻¹; ¹H and ¹³C NMR data, see Table 2; ESIMS m/z 552.2 [M + NH₄]⁺ (100); HRESIMS m/z 552.2805 [M + NH₄]⁺ (calcd for C₂₈H₄₂NO₁₀, 552.2803).

Entangolensin J (10). White, amorphous powder; [α]²⁵_D −51.8 (c 0.12, MeOH); UV (MeOH) λ_max (log ε) 205 (3.52) nm; ECD (CH₃CN, Δε) 190 (+17.59), 204 (−20.71) nm; IR (KBr) ν_max 3447, 2972, 1794, 1734, 1457, 1390, 1270, 1177, 1132, 1053, 970, 904 cm⁻¹; ¹H and ¹³C NMR data, see Table 2; ESIMS m/z 552.2 [M + NH₄]⁺ (100); HRESIMS m/z 552.2805 [M + NH₄]⁺ (calcd for C₂₈H₄₂NO₁₀, 552.2803).

Entangolensin K (11). White, amorphous powder; [α]²⁵_D −6.1 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 224 (3.63) nm; ECD (CH₃CN, Δε) 190 (+19.63), 203 (−20.06), 220 (+9.26) nm; IR (KBr) ν_max 3482, 2975, 2313, 1723, 1456, 1388, 1244, 1175, 1138, 1039, 919, 419 cm⁻¹; ¹H and ¹³C NMR data, see Table 2; ESIMS m/z 517.2 [M + NH₄]⁺ (100); HRESIMS m/z 522.2095 [M + Na]⁺ (calcd for C₂₇H₃₃NO₈Na, 522.2098).

Entangolensin L (12). White, amorphous powder; [α]²⁵_D −41.2 (c 0.08, MeOH); UV (MeOH) λ_max (log ε) 218 (3.72) nm; ECD (CH₃CN, Δε) 208 (−0.98), 224 (+5.15) nm; IR (KBr) ν_max 3448, 1733, 1676, 1369, 1273, 1232, 1163, 1001, 877, 804, 767, 602 cm⁻¹; ¹H and ¹³C NMR data, see Table 3; ESIMS m/z 514.2 [M + NH₄]⁺ (100); HRESIMS m/z 519.1985 [M + Na]⁺ (calcd for C₂₈H₃₂O₈Na, 519.1989).

Entangolensin M (13). White, amorphous powder; [α]²⁵_D −82.4 (c 0.09, MeOH); UV (MeOH) λ_max (log ε) 206 (3.49) nm; ECD (CH₃CN, Δε) 200 (−5.05), 216 (−2.01), 232 (−3.89), 264 (−0.66), 291 (−2.51) nm; IR (KBr) ν_max 3460, 2944, 1739, 1459, 1389, 1238, 1166, 1072, 1031, 915, 876, 805, 604 cm⁻¹; ¹H and ¹³C NMR data, see Table 3; ESIMS m/z 518.1 [M + NH₄]⁺ (100); HRESIMS m/z 518.2745 [M + NH₄]⁺ (calcd for C₂₈H₄₀NO₈, 518.2748).

Entangolensin N (14). White, amorphous powder; [α]²⁵_D −12.6 (c 0.09, MeOH); UV (MeOH) λ_max (log ε) 207 (3.62) nm; ECD (CH₃CN, Δε) 202 (−4.21), 217 (−1.68), 236 (−4.63) nm; IR (KBr) ν_max 3449, 2953, 1739, 1369, 1233, 1148, 1030, 922, 876, 804, 605 cm⁻¹; ¹H and ¹³C NMR data, see Table 3; ESIMS m/z 562.1 [M + NH₄]⁺ (100); HRESIMS m/z 562.3016 [M + NH₄]⁺ (calcd for C₃₀H₄₄NO₉, 562.3011).

Entangolensin O (15). White, amorphous powder; [α]²⁵_D +31.4 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 220 (3.71) nm; ECD (CH₃CN, Δε) 210 (−0.86), 231 (+12.52) nm; IR (KBr) ν_max 3422, 2932, 2313, 1734, 1666, 1384, 1239, 1128, 1039, 618, 457, 419 cm⁻¹; ¹H and ¹³C NMR data, see Table 3; ESIMS m/z 527.1 [M + H]⁺ (100); HRESIMS m/z 527.2636 [M + H]⁺ (calcd for C₃₀H₃₉O₈, 527.2639).

Entangolensin P (16). White, amorphous powder; [α]²⁵_D −84.4 (c 0.10, MeOH); UV (MeOH) λ_max (log ε) 204 (3.51) nm; ECD (CH₃CN, Δε) 202 (−8.43) nm; IR (KBr) ν_max 3436, 2961, 1785, 1724, 1377, 1247, 1166, 1053 cm⁻¹; ¹H and ¹³C NMR data, see Table 3; ESIMS m/z 634.2 [M + NH₄]⁺ (100); HRESIMS m/z 639.3136 [M + Na]⁺ (calcd for C₃₄H₄₈O₁₀Na, 639.3140).

Determination of cytotoxic activities

The cytotoxicity assay was achieved as previously reported.⁴¹ All experiments were performed as three independent replicates, with doxorubicin as a positive control.

Determination of nitrite activity assay

The protocol for the NO production bioassay is detailed elsewhere.⁴¹ All experiments were conducted for three independent replicates, with N-monomethyl-L-arginine (L-NMMA) as a positive control.

Acknowledgements

This research work was financially supported by the National Natural Science Foundation of China (31470416), the Ph.D. Program Foundation of Ministry of Education of China (20120096130002), the Outstanding Youth Fund of the Basic Research Program of Jiangsu Province (BK20160077), and the Program for Changjiang Scholars and Innovative Research Team in University (IRT_15R63).

References

1. S. G. Liao, H. D. Chen and J. M. Yue, Chem. Rev., 2009, 109, 1092–1140.
2. Editorial Committee of Flora of China, Flora Reipublicae Popularis Sinicae, Science Press, Beijing, 1997, vol. 43.
3. Q. G. Tan and X. D. Luo, Chem. Rev., 2011, 111, 7437–7522.
4. J. Luo, J. S. Wang, J. G. Luo, X. B. Wang and L. Y. Kong, Org. Lett., 2009, 11, 2281–2284.
5. F. L. An, J. Luo, R. J. Li, J. G. Luo, X. B. Wang, M. H. Yang, L. Yang, H. Q. Yao, H. B. Sun, Y. J. Chen and L. Y. Kong, Org. Lett., 2016, 18, 1924–1927.
6. G. M. Happi, S. F. Kouam, F. M. Talontsi, M. Lamshöft, S. Zühlke, J. O. Bauer, C. Strohmann and M. Spiteller, J. Nat. Prod., 2015, 78, 604–614.
7. L. Liben and R. Dechamps, in Bulletin du Jardin Botanique de l’État à Bruxelles, National Botanic Garden of Belgium, Brussels, 1966, vol. 36, Issue 4, pp. 415–424.
8. J. Garcia, C. Morin, B. Nyasse, B. L. Sondengam, J. C. Tchouankeu and E. Tsamo, J. Nat. Prod., 1991, 54, 136–142.
9. J. Bickii, F. G. R. Tchouya, J. C. Tchouankeu and E. Tsamo, Afr. J. Tradit., Complementary Altern. Med., 2007, 4, 135–139.
10. F. S. Kouam, S. Kusari, M. Lamshöft, K. O. Tatuedom and M. Spiteller, Phytochemistry, 2012, 83, 79–86.
11. D. Ngnokam, G. Massiot, J. M. Nuzillard, J. D. Connolly, E. Tsamo and C. Morin, Phytochemistry, 1993, 34, 1603–1607.
12. D. Ngnokam, J. M. Nuzillard and C. Bliard, Bull. Chem. Soc. Ethiop., 2005, 19, 227–231.
13. J. C. Tchouankeu, E. Tsamo, B. L. Sondengam, J. D. Connolly and D. S. Rycroft, Tetrahedron Lett., 1990, 31, 4505–4508.
14. D. A. Mulholland, R. Osborne, S. L. Roberts and D. A. H. Taylor, Phytochemistry, 1994, 37, 1417–1420.
15. T. K. Nsiama, H. Okamura, T. Hamada, Y. Morimoto, M. Doe, T. Iwagawa and M. Nakatani, Phytochemistry, 2011, 72, 1854–1858.
16. G. M. Happi, S. F. Kouam, F. M. Talontsi, S. Zühlke, M. Lamshöft and M. Spiteller, Fitoterapia, 2015, 102, 35–40.
17. J. C. Tchouankeu, B. Nyasse, E. Tsamo, B. L. Sondengam and C. Morin, Phytochemistry, 1992, 31, 704–705.
18. O. Koul, W. M. Daniewski, J. S. Multani, M. Gumulka and G. Singh, J. Agric. Food Chem., 2003, 51, 7271–7275.
19. V. C. O. Njar, J. K. Adesanwo and Y. Raji, Planta Med., 1995, 61, 91–92.
20. T. K. Nsiama, H. Okamura, T. Iwagawa, M. Doe, Y. Morimoto and M. Nakatani, Heterocycles, 2008, 75, 157–164.
21. W. Kraus and W. Grimminger, Nouv. J. Chim., 1980, 4, 651–655.
22. A. M. A. Samir, D. Matsumi and N. Munehiro, Phytochemistry, 2013, 96, 312–317.
23. S. Kadota, L. Marpaung, T. Kikuchi and H. Ekimoto, Chem. Pharm. Bull., 1990, 38, 639–651.
24. F. Bracher, W. J. Eisenreich, J. Mühlbacher, M. Dreyer and G. Bringmann, J. Org. Chem., 2004, 69, 8602–8608.
25. J. D. Connolly, R. McCrindle and K. H. Overton, Tetrahedron, 1968, 24, 1489–1495.
26. W. Ravangpai, D. Sommit, T. Teerawatananond, N. Sinpranee, T. Palaga, S. Pengpreecha, N. Muangsin and K. Pudhom, Bioorg. Med. Chem. Lett., 2011, 21, 4485–4489.
27. B. D. Lin, T. Yuan, C. R. Zhang, L. Dong, B. Zhang, Y. Wu and J. M. Yue, J. Nat. Prod., 2009, 72, 2084–2090.
28. Q. Zhang, Y. T. Di, H. P. He, X. Fang, D. L. Chen, X. H. Yan, F. Zhu, T. Q. Yang, L. L. Liu and X. J. Hao, J. Nat. Prod., 2011, 74, 152–157.
29. A. Akisanya, C. W. L. Bevan, J. Hirst, T. G. Halsall and D. A. H. Taylor, J. Chem. Soc., 1960, 3827–3829.
30. J. Wu, S. X. Yang, M. Y. Li, G. Feng, J. Y. Pan, Q. Xiao, J. Sinkkonen and T. Satyanandamurty, J. Nat. Prod., 2010, 73, 644–649.
31. K. L. Ji, P. Zhang, H. B. Hu, S. Hua, S. G. Liao and Y. K. Xu, J. Nat. Prod., 2014, 77, 1764–1769.
32. R. J. Li, Y. Sun, B. Sun, X. N. Wang, S. S. Liu, J. C. Zhou, J. P. Ye, Y. Zhao, L. Liu, K. H. Lee and H. X. Lou, Phytochemistry, 2014, 105, 85–91.
33. X. B. Wang, Y. Zhang, J. S. Wang, H. Q. Yao, H. B. Sun and L. Y. Kong, Struct. Chem., 2011, 22, 1241–1248.
34. X. Fang, Y. T. Di, H. P. He, H. Y. Liu, Z. Zhang, Y. L. Ren, Z. L. Gao, S. Gao and X. J. Hao, Org. Lett., 2008, 10, 1905–1908.
35. K. Mitsui, H. Saito, R. Yamamura, H. Fukaya, Y. Hitotsuyanagi and K. Takeya, J. Nat. Prod., 2006, 69, 1310–1314.
36. Y. Zhang, C. P. Tang, C. Q. Ke, S. Yao and Y. Ye, J. Nat. Prod., 2010, 73, 664–668.
37. J. D. Connolly, W. R. Phillips, D. A. Mulholland and D. A. H. Taylor, Phytochemistry, 1981, 20, 2596–2597.
38. X. N. Wang, C. Q. Fan, S. Yin, L. S. Gan and J. M. Yue, Phytochemistry, 2008, 69, 1319–1327.
39. H. Li, Y. Li, X. B. Wang, T. Pang, L. Y. Zhang, J. Luo and L. Y. Kong, RSC Adv., 2015, 5, 40465–40474.
40. J. Luo, J. S. Wang, X. B. Wang, J. G. Luo and L. Y. Kong, J. Nat. Prod., 2010, 73, 835–843.
41. Q. M. Li, J. G. Luo, Y. M. Zhang, Z. R. Li, X. B. Wang, M. H. Yang, J. Luo, H. B. Sun, Y. J. Chen and L. Y. Kong, Chem.–Eur. J., 2015, 21, 13206–13209.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c6ra19532f

---