Structures and inhibitory activity against breast cancer cells of new bufadienolides from the eggs of toad Bufo bufo gargarizans

Abstract

Two new 19-norbufadienolides (1 and 2), one new 14,15-epoxy bufadienolide (3), and eight rare bufadienolide–fatty acid conjugates (4–11), together with four known ones (12–15) were isolated from the eggs of toad Bufo bufo gargarizans. Their chemical structures were elucidated by extensive spectroscopic methods in combination with X-ray diffraction analyses. Furthermore, we tested the inhibitory effect of these compounds against breast cancer cell lines MCF-7 and MDA-MB-231. Most of them showed strong cytotoxicity with IC₅₀ values less than 0.1 μM. The further mechanistic study showed that they could induce apoptosis and cycle arrest in the G₂/M phase in MCF-7 cells.

---

Introduction

Bufadienolide, characterized by the presence of a six-membered lactone ring at the C-17β position of the steroidal skeleton, is a class of interesting natural products with potential therapeutic effects.^1–3 This kind of natural molecule has been found in many animal and plant families. Animal origins of bufadienolides include fireflies Photinus spp. (Lampyridae) and toads Bufo spp. (Bufonidae), etc.^2,3 It was reported that the venom and skin of toads Bufo are rich sources of these compounds, which are used as their chemical weapons against predators.^1–4 In addition, endogenous bufadienolides with novel structures and bioactivity have also been identified in other tissues of toads.^5–7 The discovery of endogenous bufadienolides in toad eggs intrigued great interest in various research filelds.^12–15 Some related studies reported that bufadienolides in toads are more than chemical defenses,^8–15 for instance, they might be the chemical cues for intraspecific competition.¹⁴ Another study showed bufadienolides varied greatly during toad egg ontogenetic development.¹³ However, up to now, except for two earlier studies,^6,7 there are few reports about the chemical constituents of toad eggs. As a part of our continuing search for structurally interesting and bioactive compounds from genus Bufo,^16–20 eleven new bufadienolides were identified from the eggs of toad Bufo bufo gargarizans (Fig. 1), including two 19-norbufadienolides (1 and 2), one 14,15-epoxy bufadienolide (3), and eight rare bufadienolide–fatty acid conjugates (4–11), as well as four known ones, hellebrigenin (12),^20,21 bufalin (13),²² telocinobufagin (14),^22,23 and marinobufagin (15).^22,23 Then we evaluated their cytotoxic activities against human breast cancer cell lines MCF-7 and MDA-MB-231, and further investigated the underlying mechanisms.

Fig. 1 Chemical structures of 1–15.

Results and discussion

Compound 1 was obtained as a colourless block. Its molecular formula was established as C₂₃H₃₂O₆ by the pseudomolecular ion at m/z 405.2268 [M + H]⁺ (calcd for C₂₃H₃₃O₆, 405.2272) in the HRESIMS. The IR spectrum of 1 indicated the presence of hydroxyl (3360 cm⁻¹) and carbonyl (1719 cm⁻¹) groups. The maximum UV absorption at 299 nm implied the existence of 2H-pyran-2-one moiety, which is characteristic for bufadienolides.² The ¹H and ¹³C-NMR spectra revealed 1 possessed twenty three carbons, including signals for a 2H-pyran-2-one moiety [δ_H 7.43 (1H, d, J = 2.4 Hz), 8.00 (1H, dd, J = 9.4, 2.4 Hz), 6.28 (1H, d, J = 9.4 Hz); δ_C 115.4, 125.1, 149.4, 150.5, 164.8], one angular methyl [δ_H 0.73 (3H, s); δ_C 17.2], one oxygenated methine [δ_H 4.06 (1H, br. s); δ_C 68.3], and three oxygenated quaternary carbons (δ_C 75.6, 76.2 and 85.7). The above evidences indicated 1 possessed a bufadienolide skeleton with four oxygenated substitutions.^16,20

Comparison of the ¹H and ¹³C-NMR data of 1 with those of telocinobufagin (14) revealed that the signals for rings C, D, and E of the two compounds were very similar, indicating that they share the same substructure with a hydroxyl group at C-14. The main differences were the absence of one angular methyl and the downfield shift of the quaternary carbon (δ_C 75.6) in 1.

The ¹H–¹H COSY spectrum revealed the presence of the spin systems in bold face as shown in Fig. 2. In the HMBC spectrum, the correlations between H-11 (δ_H 1.51)/H-8 (δ_H 1.58) and C-10 (δ_C 75.6), and between H-3 (δ_H 4.06)/H-7 (δ_H 1.93) and C-5 (δ_C 76.2) suggested there are two hydroxyl groups located at C-5 and C-10, respectively (Fig. 2). Thus, 1 was deduced to be a 19-norbufadienolide. The relative configuration of 1 was determined by analysis of its ROESY spectrum (Fig. 3), which showed the cross-peaks between H-9 (δ_H 1.65)/H-7α (δ_H 1.23) and H-4α (δ_H 2.11), indicating that rings A and B remained cis junction as most bufadienolides. In addition, the structure of 1 was further confirmed by single crystal X-ray diffraction analysis (Fig. 4). The final refinement resulted in a small Flack parameter 0.09(10) indicating that the absolute configuration could be assigned as shown in Fig. 4. Accordingly, the structure of 1 was determined as (3β,5β,10β,14β)-3,5,10,14-tetrahydroxy-19-norbufa-20,22-dienolide.

Fig. 2 Key ¹H–¹H COSY and HMBC correlations of compounds 1, 2 and 7.

Fig. 3 Key NOE correlations of compound 1.

Fig. 4 Ortep plot of X-ray structure of 1.

Compound 2 was isolated as a colorless block. Its molecular formula was determined to be C₂₃H₃₀O₅ by its HRESIMS data at m/z 387.2167 [M + H]⁺ (calcd for C₂₃H₃₁O₅, 387.2166). The UV and IR spectra indicated the existence of a 2H-pyran-2-one moiety. Comparison of the ¹H and ¹³C NMR data (Table 1) with the known bufadienolide marinobufagin (15)^22,23 revealed 2 possessed a bufadienolide skeleton with an epoxy group at C-14 and C-15, and one hydroxyl group at C-3. Moreover, the only angular methyl (δ_C 17.0) was deduced to be located at C-13 by the HMBC correlations between H₃-18 and C-12 (δ_C 40.1)/C-14 (δ_C 75.1)/C-17 (δ_C 48.6). The above evidences suggested 2 was also a 19-norbufadienolide. In addition, the HMBC correlations between H-3 (δ_H 4.09)/H-9 (δ_H 1.41)/H-10 (δ_H 1.41) and C-5 (δ_C 74.4) suggested another hydroxyl group was connected to C-5 (Fig. 2). The ¹H and ¹³C NMR data of 2 was assigned unambiguously by extensive analysis of its 1D and 2D NMR spectra. Furthermore, the complete structure of 2 was also confirmed by single crystal X-ray diffraction analysis (Fig. 5). The small Flack parameter −0.011(14) indicating that the absolute configuration could be assigned as shown in Fig. 5. Accordingly, the structure of 2 was determined as (3β,10β,14β)-14,15-epoxy-3,5-dihydroxy-19-norbufa-20,22-dienolide.

Table 1 ¹H and ¹³C NMR data of compounds 1–3^a,b

No.	1		2		3
	δH	δC	δH	δC	δH	δC
a ^1H and ^13C NMR data were measured in CD3OD.b Overlapped signals were reported without designating multiplicity.
1α	1.71	26.2	1.40	27.4	1.44	20.2
β	1.91		1.85, m		2.34, m
2α	1.83	30.8	1.61	27.5	1.71	28.3
β	2.17		1.62		1.71
3	4.06, br. s	68.3	4.09, br. s	69.0	4.09, br. s	68.8
4α	2.11	38.7	2.15, m	37.1	2.26, m	37.9
β	1.44, m		1.55, m		1.43
5		76.2		74.4		78.5
6α	1.40, m	36.0	1.36, m	33.3	1.44	36.1
β	1.83		1.67		1.89
7α	1.23, m	24.5	0.89, m	24.0	1.04, m	24.2
β	1.93		1.65		1.90, m
8	1.58, m	41.1	1.64	40.8	2.13, m	34.0
9	1.65	41.5	1.41	40.4	1.72	43.6
10		75.6	1.41	41.9		43.9
11α	1.49	22.3	1.39, m	22.6	1.33, m	22.7
β	1.51		1.59		1.60, m
12α	1.50	41.7	1.40	40.1	1.48	40.4
β	1.80		1.66		1.70
13		49.5		46.2		46.1
14		85.7		75.1		75.9
15α	1.64, m	33.0	3.56, br. s	61.3	3.61, br. s	61.1
β	2.21, m
16α	1.75	29.9	1.90, m	33.2	1.87	33.1
β	2.17		2.37, m		2.38, m
17	2.55, dd, 9.4, 6.5	52.0	2.59, dd, 9.3, 2.1	48.6	2.58, d, 10	48.4
18	0.73, s	17.2	0.80, s	17.0	0.77, s	17.1
19					4.21, d, 11.3	65.8
					3.63, d, 11.3
20		125.1		124.5		124.5
21	7.43, d, 2.4	150.5	7.44, d, 2.4	151.8	7.44, d, 2.3	151.8
22	8.00, dd, 9.4, 2.4	149.4	7.90, dd, 9.4, 2.4	149.6	7.89, dd, 9.4, 2.3	149.6
23	6.28, d, 9.4	115.4	6.25, d, 9.4	115.4	6.26, d, 9.4	115.4
24		164.8		164.5		164.5

---

Fig. 5 Ortep plot of X-ray structure of 2.

Both 1 and 2 are composed of three cyclohexane rings (A, B and C), a five-membered ring D and a lactone ring E. In crystalline states of 1 and 2, the cyclohexane rings A, B and C exist in normal chair conformations. The five-membered ring D adopts an envelope conformation and the lactone ring E is planar. The dihedral angle between rings E and D are 103.6° and 102.2° for 1 and 2, respectively. The ring junctures for both compounds are A/B cis, B/C trans and C/D trans. Thus, 1 and 2 share similar conformation in solid state (Fig. 6). In addition, ¹H-NMR spectrum of 1 showed that the coupling constant between H(17) and H(16α) and H(16β) are 9.4 Hz and 6.5 Hz, which are consistent with the torsion angles H(17)–C(17)–C(16)–H(16α) 9.9° and H(17)–C(17)–C(16)–H(16β) 129.8°, respectively. Similarly, in compound 2, coupling constant between H(17) and H(16α) and H(16β) are 9.3 Hz and 2.1 Hz, which are consistent with the corresponding torsion angles 10.4° and 110.1°, respectively. Thus the conformation of compounds 1 and 2 in crystalline state is consistent with the solution structure in methanol, which is similar to thiocarbonylbufalin and 1β-hydroxydigitoxigenin with a rigid ring system,²⁴ but different from spiro-prorocentrimine with a soft macrolide moiety.²⁵

Fig. 6 Comparison of conformations of 1 and 2.

Compound 3 was obtained as a white powder. The molecular formula C₂₄H₃₂O₆ was established from the positive pseudomolecular ion at m/z 417.2264 [M + H]⁺ (calcd for C₂₄H₃₃O₆⁺, 417.2272) in its HRESIMS. The ¹H and ¹³C NMR spectra (Table 1) were very similar to those of the marinobufagin (15)^22,23 except for the replacement of angular methyl signal in 15 by two mutually coupled oxygenated proton signals at δ_H 4.21 (1H, d, J = 11.3 Hz) and 3.61 (1H, d, J = 11.3 Hz), both of which were correlated to an oxygenated methylene carbon at δ_C 65.8 in its HSQC spectrum. Furthermore, the oxygenated methylene was assigned to C-19 by the HMBC correlations between H₂-19 (δ_H 4.21, 3.61) and C-1 (δ_C 20.2)/C-5 (δ_C 78.5). Therefore, the structure of 3 was determined as 19-hydroxymarinobufagin.

Compound 4 was obtained as a colorless powder. The molecular formula C₃₇H₅₈O₈ was deduced by the quasi-molecular ion peak at m/z 653.4039 [M + Na]⁺ (calcd for C₃₇H₅₈O₈Na, 653.4029) in its HRESIMS. In the ¹H NMR spectrum of 4, typical signals for the 2H-pyran-2-one moiety [δ_H 6.28 (1H, d, J = 9.4 Hz), 7.44 (1H, d, J = 2.4 Hz), 8.00 (1H, dd, J = 9.4, 2.4 Hz)] could be observed. Further analysis of its 1D NMR (Tables 2 and 3) and MS data revealed that 4 contained 37 carbons, including one angular methyl (δ_H 0.73; δ_C 17.2), 22 methylenes (including one oxygenated methylene at δ_H 3.53; δ_C 63.0), seven methines (including one oxygenated methine at δ_H 5.11; δ_C 71.1) and five quaternary carbons (including three oxygenated quaternary carbons at δ_C 74.1, 75.4, 85.8 and two ester carbonyls at δ_C 164.8, 174.8). Comparison of the NMR data of 4 with those of 1 indicated the chemical shifts in A–E rings were very similar except for an obvious down-field shift of H-3 (δ_H 5.11) and C-3 (δ_C 71.1) and additional NMR signals for one carbonyl (δ_C 174.8), one hydroxymethyl (δ_H 3.53; δ_C 63.0), and thirteen methylenes between δ_C 26.0 and 36.0 in 4. These above evidences suggested 4 was an aliphatic ester at C-3 of 1. Moreover, the ¹H–¹H COSY correlation between H-13′ (δ_H 1.54) and H-14′ (δ_H 3.53) indicated the existence of a hydroxymethyl at the end of the aliphatic chain. By combination of HRESIMS data, 4 was deduced to be the 14-hydroxytetradecanoic acid ester of 1. Therefore, the structure of compound 4 was determined to be (3β,5β,10β,14β)-3,5,10,14-tetrahydroxy-19-norbufa-20,22-dienolide-3-(14-hydroxytetradecanoic acid) ester.

Table 2 ¹³C NMR data of compounds 4–11^a,b

No.	4	5	6	7	8	9	10	11
a ^13C NMR data were measured in CD3OD.b ^b–dAssignments maybe interchanged in each column.
1	26.1	19.0	19.0	19.0	26.6	20.7	74.0	24.9
2	30.8	25.6	25.6	25.6	25.4	28.1	32.0	27.2
3	71.1	71.0	70.8	70.8	72.2	71.5	70.8	71.9
4	38.7	37.9	37.9	37.9	36.3	37.3	36.6	33.7
5	75.4	73.9	73.9	73.9	74.4	76.2	74.0	30.4
6	36.0	37.2	37.2	37.2	36.1	36.6	36.0	25.0
7	24.8	24.7	24.7	24.7	25.1	25.2	24.8	22.1
8	41.0	43.1	43.1	43.1	42.0	41.7	42.0	42.8
9	41.4	40.2	40.2	40.2	40.1	39.9	42.0	36.6
10	74.1	56.0	56.0	56.0	41.5	43.4	43.9	40.4
11	22.3	23.4	23.4	23.4	22.7	22.6	22.7	22.6
12	41.6	41.5	41.5	41.5	41.7	41.9	41.5	42.2
13	49.5	49.3	49.3	49.3	49.6	49.6	49.7	49.6
14	85.8	85.6	85.6	85.6	86.0	86.1	85.8	86.2
15	32.9	32.6	32.6	32.6	33.2	32.9	33.1	33.0
16	29.9	29.7	29.7	29.7	28.6	29.7	29.8	30.3
17	52.0	51.9	51.9	51.9	51.1	52.1	52.1	52.2
18	17.2	17.1	17.1	17.1	17.1	17.2	17.2	17.3
19	—	209.6	209.6	209.6	17.2	65.9	13.2	65.4
20	125.1	124.9	124.9	124.9	125.0	124.9	124.9	125.0
21	150.5	152.3	152.3	152.3	150.5	150.5	150.5	150.5
22	149.4	149.3	149.3	149.3	149.3	149.3	149.3	149.4
23	115.4	115.6	115.6	115.6	115.4	115.4	115.5	115.4
24	164.8	164.7	164.7	164.7	164.8	164.7	164.8	164.8
1′	174.8	174.9	174.8	174.9	174.6	174.8	175.1	174.8
2′	35.5	35.5	35.5	35.5	35.6	35.5	35.8	35.7
3′	26.0	26.1	26.0	25.9	26.1	26.0	26.0	26.0
4′	30.2^b	30.2^b	30.2^b	29.8	30.1^b	30.2^b	30.1^b	30.2^b
5′	30.4^b	30.4^b	30.4^b	28.0^c	30.3^b	30.4^b	30.3^b	30.4^b
6′	30.6^b	30.6^b	30.6^b	131.0^d	30.4^b	30.6^b	30.4^b	30.4^b
7′	30.6^b	30.6^b	30.6	130.7^d	30.6^b	30.6^b	30.6^b	30.6^b
8′	30.7^b	30.7^b	30.6^b	28.1^c	30.6^b	30.7^b	30.6^b	30.6^b
9′	30.7^b	30.7^b	30.6^b	30.1^b	30.7^b	30.8^b	30.7^b	30.7^b
10′	30.8^b	30.7^b	27.0	30.9^b	30.8^b	30.8^b	30.8^b	30.7^b
11′	30.8^b	30.8^b	33.7	30.5^b	30.8^b	30.8^b	30.8^b	30.7^b
12′	27.0	27.0	63.0	26.8	27.0	27.0	27.0	30.8^b
13′	33.7	33.7		33.7	33.7	33.7	33.7	30.9^b
14′	63.0	63.0		63.0	63.0	63.0	63.0	27.0
15′								33.7
16′								63.0

---

Table 3 ¹H NMR data of compounds 4–11^a,b,c

	4	5	6	7	8	9	10	11
a ^1H and ^13C NMR data were measured in CD3OD.b Overlapped signals were reported without designating multiplicity.c ^c–eAssignments maybe interchanged in each column.
1α	1.71	1.73	1.72	1.73	1.37	1.39, m	3.90, br. s	1.44
β	1.91	2.10	2.09	2.10	1.88, m	2.32		1.85
2α	1.83	1.65	1.63	1.65	1.60	1.75	1.76	1.32
β	2.17	2.17, m	2.18	2.17, m	1.60	2.03	2.06	1.51
3	5.11, br. s	5.12, br. s	5.11, br. s	5.12, br. s	5.16, br. s	5.11, br. s	5.23, br. s	5.08, br. s
4α	1.44	1.66	1.66	1.66	1.57	1.92, m	1.73	1.54
β	1.99, m	2.36, m	2.36, m	2.36, m	2.35	2.32	2.45, m	1.72
5								1.75
6α	1.40	1.62	1.62	1.62	1.42	1.53	1.53	1.43
β	1.83	2.32, m	2.33	2.32	1.74	2.34	2.34	1.86
7α	1.10, m	1.35	1.36	1.35	1.23, m	1.33	1.33	1.32
β	1.93	1.79	1.79, m	1.79, m	1.93, m	2.00	2.00, m	1.85
8	1.58	1.98	1.98, m	1.98, m	1.65	1.74	1.75	1.70
9	1.51	1.74	1.75	1.74	1.62	1.69	1.75	2.23, m
11α	1.51	1.54	1.55	1.54	1.25, m	1.33	1.37	1.34
β	1.52	1.76	1.76	1.76	1.45	1.49, m	1.49, m	1.82
12α	1.51	1.40	1.37	1.41, m	1.30	1.42, m	1.30	1.40
β	1.80	1.55	1.55	1.55	1.54	1.57	1.59	1.55
15α	1.64	1.70	1.71	1.70	1.70	1.70	1.70	1.73
β	2.21, m	2.11	2.11	2.11	2.10, m	2.10, m	2.10	2.19, m
16α	1.75	1.75	1.74	1.75	1.75	1.72	1.73	1.74
β	2.17	2.22, m	2.20	2.22, m	2.16, m	2.20, m	2.19, m	2.27, m
17	2.55, dd, 9.4, 6.5	2.53, d, 9.3	2.53, d, 9.3	2.53, d, 9.3	2.57, br. d, 9.5	2.56, dd, 9.2, 6.5	2.56, dd, 9.2, 6.5	2.55, dd, 9.4, 6.1
18	0.73, s	0.67, s	0.67, s	0.67, s	0.71, s	0.70, s	0.70, s	0.69, s
19		10.09, s	10.10, s	10.09, s	0.96, s	4.23, d, 11.0	1.20, s	3.86, d, 11.0
						3.57, d, 11.0		3.57, d, 11.0
21	7.44, d, 2.4	7.43, d, 2.4	7.44, d, 2.4	7.43, d, 2.4	7.42, d, 2.5	7.42, d, 2.4	7.42, d, 2.4	7.42, d, 2.4
22	8.00, dd, 9.4, 2.4	7.99, dd, 9.4, 2.4	7.99, dd, 9.4, 2.4	7.99, dd, 9.4, 2.4	7.99, dd, 9.5, 2.5	7.98, dd, 9.4, 2.4	7.98, dd, 9.4, 2.4	7.98, dd, 9.4, 2.4
23	6.28, d, 9.4	6.28, d, 9.4	6.28, d, 9.4	6.28, d, 9.4	6.28, d, 9.5	6.26, d, 9.4	6.25, d, 9.4	6.27, d, 9.4
2′	2.33, m	2.33	2.34	2.32	2.34	2.33	2.34	2.33
3′	1.63	1.63	1.63	1.61	1.62	1.62	1.63	1.63
4′	1.34	1.34	1.34	1.36	1.34	1.34	1.34	1.34
5′	1.30^e	1.30^e	1.30	2.04^c	1.31^e	1.31^e	1.31^e	1.32^c
6′	1.31^e	1.30^e	1.32	5.35^d	1.31^e	1.31^e	1.31^e	1.32^c
7′	1.32^e	1.30^e	1.31^e	5.35^d	1.31^e	1.31^e	1.31^e	1.33^c
8′	1.33^e	1.30^e	1.31^e	2.04^c	1.31^e	1.31^e	1.31^e	1.33^c
9′	1.33^e	1.30^e	1.31^e	1.34^e	1.31^e	1.31^e	1.31^e	1.33^c
10′	1.33^e	1.30^e	1.36	1.36^e	1.31^e	1.31^e	1.31^e	1.32^c
11′	1.33^e	1.30^e	1.54	1.36^e	1.32^e	1.31^e	1.31^e	1.32^c
12′	1.36	1.33	3.53 (t, 6.5)	1.34	1.34	1.34	1.33	1.32^c
13′	1.54, m	1.54		1.54	1.54	1.54	1.54	1.32^c
14′	3.53, t, 6.5	3.53, t, 6.5		3.54, t, 6.5	3.54, t, 6.5	3.54, t, 6.5	3.53, t, 6.5	1.34
15′								1.55
16′								3.55, t, 6.5

---

Compound 5 was isolated as a white powder. The molecular formula of 5 was determined as C₃₈H₅₈O₈ according to the positive pseudomolecular ion peak at m/z 665.4018 [M + Na]⁺ (calcd 665.4024) in its HRESIMS. The ¹H and ¹³C NMR spectra indicated that 5 was also a bufadienolide conjugate. Comparison of the NMR data of 5 with those of hellebrigenin (12)²¹ indicated that the chemical shifts for the genin part in 5 were similar to hellebrigenin with three hydroxyl groups at C-3 (δ_C 71.0), C-5 (δ_C 73.9) and C-14 (δ_C 85.6), and an aldehyde group [δ_H 10.09 (1H, s); δ_C 209.6] at C-10. The main differences were the down field shift of H-3 [δ_H 5.11 (1H, br. s)] and C-3 (δ_C 71.0) which was confirmed by the analysis of HRESIMS data. In addition, the NMR data of the fatty acid moiety in 5 was the same as that in 4. Therefore, the structure of compound 5 was established as hellebrigenin-3-(14-hydroxytetradecanoic acid) ester.

Compound 6 was obtained as a white powder. The molecular formula, C₃₆H₅₄O₈, was established from the positive pseudomolecular ion at m/z 637.3713 [M + Na]⁺ in its HRESIMS. The ¹H and ¹³C NMR signals (Tables 2 and 3) were similar to those of 5, revealing that they had the same genin unit. However, in the ¹³C-NMR spectrum, only twelve carbon signals were observed for the side chain of compound 6, which is two less than that of compound 5. These data suggested that the side chain of compound 6 was the ester of 12-hydroxydodecanoic acid. Thus, compound 6 was determined to be hellebrigenin-3-(12-hydroxydodecanoic acid) ester.

Compound 7 was obtained as a white powder. The molecular formula, C₃₈H₅₆O₈, was established from the positive ion at m/z 663.3921 [M + Na]⁺ (calcd 663.3867) in its HRESIMS. In the ¹³C NMR spectrum, besides the signals for the genin moiety, fourteen other signals were observed, including a carbon signal at δ_C 174.8 assigned to an ester carbon, and two carbon signals at δ_C 131.0 and 130.6 assigned to unsaturated methine (–CH=CH–), and other eleven carbons appeared as methylene signals in the DEPT spectrum, among which an oxygenated methylene at δ_C 63.0 was observed. In addition, comparison of ¹H and ¹³C NMR spectra of compound 7 with those of 5 and 6 confirmed that the structure of the genin moiety in 7 was also hellebrigenin (12). These data indicated that compound 7 was a conjugated bufadienolide with an unsaturated hydroxy fatty acid side chain.

All the ¹H and ¹³C NMR signals (Tables 2 and 3) of 7 were assigned unambiguously by an extensive analysis of ¹H–¹H COSY, HSQC, HMBC and ROESY data. In the ¹H–¹H COSY spectrum (Fig. 2), spin system of H-14′ ↔ H-13′ (δ_H 1.54) ↔ H-12′ (δ_H 1.34) could be observed, demonstrating the existence of HO–C^14′H₂–C^13′H₂–C^12′H₂– fragment. The HMBC correlation between H-2′ (δ_H 2.32) and C-1′ (δ_C 174.9), as well as ¹H–¹H COSY correlations between H-3′ (δ_H 1.61) and H-2′/H-4′ (δ_H 1.34), suggested a partial structure –OC^1′O–C^2′H₂–C^3′H₂–C^4′H₂–. In addition, two methine protons at δ_H 5.35 were correlated with two pairs of methylenes at δ_H 2.04, which were in turn correlated with another two pairs of methylenes at δ_H 1.34, suggesting the presence of a partial structure –CH₂–CH₂–CH=CH–CH₂–CH₂– in the side chain. The above NMR data, together with MS data suggested 7 was an ester of 14-hydroxytetradecenoic acid, and the position of the double bond was between C-6′ and C-7′, C-7′ and C-8′, or C-8′ and C-9′. In order to determine the position of the double bond, 7 was hydrolyzed to afford a fatty acid methyl ester, which was subsequently acetylated using acetic anhydride in pyridine. Then we characterized the DMDS adduct through EI-MS analysis to identify the position of the double bond.²⁶ The EI-MS spectrum showed a molecule ion at m/z 392 (M⁺) and fragment ions at m/z 217 [(C₁₁H₂₁O₂S)⁺] and m/z 175 [(C₈H₁₅O₂S)⁺], indicating the double bond was between C-6′ and C-7′ (ESI Fig. S99†). Moreover, the chemical shifts of C-5′ and C-8′ was at δ_C 28.2 and δ_C 28.0, respectively, suggesting the Z configuration of the double bond.²⁷ Therefore, the structure of compound 7 was determined to be (6′Z)-hellebrigenin-3-(14-hydroxy-7-tetradecenoicacid) ester.

Compound 8 was obtained as a white powder. The molecular formula of 8 was established to be C₃₈H₆₀O₇ from the positive ion peak at m/z 651.4250 [M + Na]⁺ (calcd 651.4232) in its HRESIMS. The 1D NMR spectra revealed that 8 was also a bufadienolide conjugate. Comparison of the ¹H and ¹³C NMR data of 8 with those of 4 and 14 (Tables 2 and 3) indicated its genin moiety was telocinobufagin^22,23 (14) with 3,5,14-trihydroxy substitutions and the C-3 side chain was an ester of 14-hydroxytetradecanoic acid. All the ¹H and ¹³C NMR signals of 8 were assigned unambiguously by an extensive analysis of ¹H–¹H COSY, HSQC, HMBC and ROESY spectra. Accordingly, compound 8 was determined as telocinobufagin-3-(14-hydroxytetradecanoic acid) ester.

Compound 9 was isolated as a white powder. The molecular formula, C₃₈H₆₀O₈, was established from the positive pseudomolecular ion at m/z 667.4189 [M + Na]⁺ (calcd 667.4180) in its HRESIMS. The ¹H and ¹³C NMR signals for the side chain and the steroidal skeleton of 9 were similar to those of 8, except for the replacement of NMR signals δ_H 0.96 and δ_C 17.2 with signals δ_H 4.23 and 3.57 and δ_C 65.9, respectively, suggesting that the angular methyl at C-19 in 8 was changed to a hydroxymethylene in 9. Therefore, the genin of 9 was established as hellebrigenol.²⁰ All the ¹H and ¹³C NMR signals of 9 were assigned unambiguously by an extensive analysis of its ¹H–¹H COSY, HSQC and HMBC spectra. Accordingly, the structure of 9 was determined to be hellebrigenol-3-(14-hydroxytetradecanoic acid) ester.

Compound 10 was isolated as a white powder, and its molecular weight was 16 amu higher than that of 8, as revealed from the positive ion at m/z 667.4184 [M + Na]⁺ (calcd 667.4180) in its HRESIMS. Comparison of the NMR data of 10 with those of 8 revealed that 10 was a hydroxylated derivative of 8 in the genin moiety. The HMBC correlation between H₃-19 (δ_H 1.20, s) and C-1 (δ_C 74.0), as well as ¹H–¹H COSY correlation between H-1 (δ_H 3.90) and H₂-2 (δ_H 2.06) suggested the position of the hydroxyl group was at C-1. The relative configurations of 10 were determined based on analysis of ROESY spectrum. The NOE correlation between H-1 and H-11α (δ_H 1.37) suggested that H-1 was α-oriented, and accordingly the hydroxyl group at C-1 was in a β-orientation. Then the genin of 10 was identified as 1β-hydroxytelocinobufagin, which was a new bufadienolide. Thereby, the structure of 10 was established to be 1β-hydroxytelocinobufagin-3-(14-hydroxytetradecanoic acid) ester.

Compound 11 was obtained as a white powder. Its molecular formula was determined to be C₄₀H₆₄O₇ from the quasi-molecular ion peak at m/z 679.4566 [M + Na]⁺ (calcd C₄₀H₆₄O₇Na, 679.4544) in its HRESIMS. The ¹H and ¹³C NMR spectra of 11 were very similar to those of 9 except for the appearance of a methine signal [δ_H 1.75 (1H, H-5); δ_C 30.39 (C-5)] and the absence of an oxygenated quaternary carbon [δ_C 76.20 (C-5)] in 11. Other differences were showed in the ¹³C NMR and HRESIMS spectra. In ¹³C NMR spectra, more carbon signals between δ_C 30 and δ_C 31 could be observed in 11, which were correlated with the methylene protons at δ_H 1.29 in the HSQC spectrum. Due to the heavy overlapping methylene signals between δ_C 30 and δ_C 31, it's hard to clearly figure out the number of carbons. But with the aid of high resolution TOF-MS, we could conclude there were 16 carbons in the side chain of 11, which is two more carbons than that of 9. Thus, the structure of 11 was determined to be 19-hydroxybufalin-3-(16-hydroxyhexadecenoic acid).

Besides the pronounced Na⁺, K⁺-ATPase inhibitory activity,²⁸ the anticancer effects of bufadienolides have received current attentions.²⁸ All isolated compounds except for compound 2 (limited amount) were tested for in vitro inhibitory activity against human breast cancer cell lines MDA-MB-231 (estrogen insensitive) and MCF-7 (estrogen sensitive). As shown in Table 4, most of them showed stronger inhibitory activity on MCF-7 cells than that on MDA-MB-231 cells. However, the bufadienolide–fatty acid conjugates 5–6 and 9–11 exhibited equal or even more potent inhibitory activities on the more malignant cancer line MDA-MB-231. It was well noteworthy that compound 7 with an unsaturated hydroxyl fatty side chain showed the most potent inhibition on MDA-MB-231 cells with an IC₅₀ value of 0.05 ± 0.014 μM. Moreover, comparison of the activities between the genins and their corresponding fatty acid conjugates, it revealed that the conjugation with a fatty side chain can increase (1 and 4; 5, 6, 7 and 12; 8 and 14) their inhibitory activities on MDA-MB-231 cancer cells.

Table 4 Cytotoxic effects of bufadienolides on human breast cancer MCF-7 and MDA-MB-23 cells^a

Compounds	IC50 (x̄ ± SD) μM
	MCF-7	MDA-MB-231
a All data are presented as means ± standard deviation of at least three independent experiments.
1	2.706 ± 0.028	18.1 ± 1.41
3	0.180 ± 0.056	1.11 ± 0.17
4	0.229 ± 0.132	1.69 ± 0.34
5	0.16 ± 0.047	0.09 ± 0.014
6	0.129 ± 0.053	0.19 ± 0
7	0.027 ± 0.023	0.05 ± 0.014
8	0.074 ± 0.027	0.3 ± 0.13
9	0.13 ± 0.06	0.27 ± 0
10	0.315 ± 0.272	0.355 ± 0.049
11	2.633 ± 0.558	1.0 ± 0.014
12	0.032 ± 0.011	0.24 ± 0.01
13	0.006 ± 0.004	0.086 ± 0.013
14	0.081 ± 0.017	0.165 ± 0.035
15	2.394 ± 1.066	6.27 ± 0.86

---

Furthermore, we selected compound 12 as a representative bufadienolide to elucidate the underlying mechanism of action. As shown in Fig. 7A, the percentage of apoptotic MCF-7 cells including early and late apoptotic cells was increased from 2.94% to 12.85%, 31.53%, and 41.98% after treatment with compound 12 (0.02 μM) for 12 h, 24 h and 48 h, respectively. These results indicated that compound 12 induced apoptosis in a time-dependent manner. It was further supported by the changes in some apoptotic hallmarks detected by western blot. As shown in Fig. 7B & C, compound 12 triggered the cleavage of PARP, caspase-3 and caspase-9, implying that the mitochondrial apoptotic pathway was involved in compound 12-induced apoptosis. In addition, it was also observed that compound 12 caused cell cycle arrest in G₂/M phase. The percentage of cells arrested in G₂/M phase was increased from 21.58% to 50.12% and 57.59% after treatment with compound 12 (0.02 μM) for 12 h and 24 h (Fig. 8A). In addition, compound 12 could increase the levels of cyclin B1 which are the main regulatory molecules in cell cycle G₂/M transition (Fig. 8B). Activation of the protein kinase Cdc2 is required for entry into meiotic or mitotic M phase in all eukaryotic cell, and phosphorylation of Thr14 and Tyr15 inhibited the activation of Cdc2 and thus inhibited the G₂ to M phase transition. The phosphorylated Cdc2 in Tyr15 was also increased after compound 12 treatment.

Fig. 7 Compound 12 induces MCF-7 apoptosis in a time-dependent manner. (A) Flow cytometry analysis of PI/Annexin V-stained MCF-7 cells treated with compound 12 (0.02 μM) for 12 h, 24 h and 48 h. (B) The quantitative analysis of the ratio of apoptotic MCF-7 cells. **P < 0.01, one-way ANOVA, post hoc comparisons, Tukey's test. Columns, mean; error bars, S.D. (C) Western blot analysis of PARP, cleaved PARP, caspase 9, cleaved caspase 9, caspase 3 and cleaved caspase 3 in MCF-7 cells treated with compound 12 (0.02 μM) for 12 h, 24 h and 48 h. β-Actin served as a loading control.

Fig. 8 Compound 12 induces cell cycle arrest in G₂/M phase in MCF-7 cells. (A) Flow cytometry analysis of PI stained MCF-7 cells treated with compound 12 (0.02 μM) for 12 h, 24 h and 48 h. (B) The quantitative analysis of MCF-7 cell population in G₂/M phase. **P < 0.01, one-way ANOVA, post hoc comparisons, Tukey's test. Columns, mean; error bars, S.D. (C) Western blot analysis of cyclin B1, Cdc2 and phospho-Cdc2^Tyr15 in MCF-7 cells treated with compound 12 (0.02 μM) for 12 h, 24 h and 48 h. β-Actin served as a loading control.

Conclusions

In the present study we reported eleven new bufadienolides and four known ones from the toad eggs. Among them, compounds 3–11 were eight rare bufadienolide–fatty acid conjugates. Though steroid–fatty acid conjugates were found in many species,^29–32 cardiotonic steroid–fatty acid conjugates can only be found in toads,⁷ and these new fatty acid conjugates are characterized by the presence of a terminal hydroxyl group, which are different from the dicarboxylic esters and amides found in the venom or skins of toads.^2,33,34 Furthermore, the fatty acid side chain with a terminal hydroxyl group is rare in natural source and haven't been reported from other species.

All of the tested compounds showed obvious inhibitory activity in human breast cancer cell lines MCF-7 and MDA-MB-231. Interestingly, the bufadienolide–fatty acid conjugates showed potent activities with IC₅₀ in the range of 0.027–2.633 μM for MCF-7 cells and 0.05–1.69 μM for MDA-MB-231 cells. It was reported that bufadienolides would be less active when its 3-OH were modified, especially esterified by acid.^28,35 However, our results showed that bufadienolide–fatty acid conjugate with a terminal hydroxyl group showed potent or even more potent inhibitory activity, especially on the more malignant estrogen insensitive cancer cells MDA-MB-231. We further studied the underlying mechanism of those bufadienolides (compound 12 as a representative). The results showed that it could induce apoptosis and cycle arrest in G₂/M phase in MCF-7 cells.

Experimental section

General experimental procedures

Collection. Toad eggs were collected from Tianxing toad breeding farm (Jinan city, P. R. China). The protocol was approved by the Animal Care and Use Committee of the Institute of Traditional Chinese Medicine and Natural Products, Jinan University, China. Specimens (no. BBG-002) were deposited in College of Pharmacy, Jinan University. Though no wild toad egg was used, this project followed the law of the People's Republic of China on the protection of Wildlife (http://www.forestry.gov.cn/portal/jsxh/s/3477/content-885770.html).

General.
TLC. Pre-coated silica gel GF254 plates (Qingdao Marine Chemical Factory, Qing dao, China). Optical rotations were recorded in CH₃OH on a Jasco P-1020 polarimeter at room temperature. UV spectra were determined in CH₃OH on a Jasco V-550 UV/vis spectrophotometer. IR spectra were measured on a Jasco FT/IR-480 plus Fourier transform infrared spectrometer using KBr pellets. HRESIMS data were obtained on an Agilent 6210 ESI/TOF mass spectrometer. Nuclear magnetic resonance (NMR) spectra were measured on Bruker AV-300 or AV 400. Column chromatographies were performed on silica gel (200–400 mesh, Qingdao Marine Chemical Plant, Qingdao, People's Republic of China), reverse-phase C₁₈ silica gel (Merck, Darmstadt, Germany) and Sephadex LH-20 (Pharmacia). Preparative HPLC was performed on a Varian Prostar system equipped with a preparative Cosmosil C₁₈ column (5 μm, 20 × 250 mm) column. All solvents used in column chromatography and high-performance liquid chromatography (HPLC) were of analytical grade (Shanghai Chemical Plant, Shanghai, People's Republic of China) and chromatographic grade (Fisher Scientific, NJ, USA), respectively.

Extraction and isolation. The toad eggs (5 kg) were crushed to powder and were extracted by 95% ethanol under ultrasonic condition (40 min, 40 °C). The extract was concentrated under reduced pressure to provide a residue (500 g), which was subsequently partitioned by petrol ether, methylene dichloride and water. The CH₂Cl₂ solution was evaporated to give a residue (80 g), which was then subjected to silica gel (200–300 mesh), eluted with petrol ether–acetone (1 : 0, 10 : 1, 5 : 1, 3 : 1 and 1 : 1) to give 6 fractions (Fr. 1–6). The first two fractions were mainly oil and lipid, and were not further separated. Fr. 3, 4, 5 and 6 were separated on reverse-phase C₁₈ silica using methanol–water gradients (i.e., 80 : 20, 65 : 35, 50 : 50 and 0 : 1 – v/v) to give Fr. 3.1–Fr. 3.3, Fr. 4.1–Fr. 4.5, Fr. 5.1–Fr. 5.5 and Fr. 6.1–Fr. 6.5. Fr. 3.2 was purified by preparative HPLC using acetonitrile–water (70 : 30) to give 10 (4.0 mg) and 11 (2.8 mg). Fr. 3.3 was purified by preparative HPLC using acetonitrile–water (65 : 35) as eluent to give 6 (3.8 mg), 7 (3.5 mg) and 9 (4.2 mg). Fr. 4.2 were further separated by reverse-phase C₁₈ silica gel eluted with methanol–water gradients give 5 (70 mg). Compound 8 (4.7 mg) was isolated from Fr. 4.3 by preparative HPLC using acetonitrile–water (70 : 30) as eluent. Fr. 5.3 was further separated by reverse-phase C₁₈ silica gel eluted with methanol–water mixture (50 : 50 – v/v) to give 1 (5.0 mg). Fr. 5.4 and Fr. 5.5 were purified by preparative HPLC using acetonitrile–water (30 : 70) as eluent to yield 2 (3.0 mg) and 4 (4.6 mg), respectively. Compounds 12 (6.0 mg), 13 (4.2 mg), and 14 (3.9 mg) were purified using preparative HPLC with acetonitrile–water (30 : 70) as eluent from Fr. 6.2. Compound 3 (5.0 mg) was purified from Fr. 6.3 by preparative HPLC using acetonitrile–water (35 : 65) as eluent.
(3β,5β,10β,14β)-3,5,10,14-Tetrahydroxy-19-norbufa-20,22-dienolide (1). [α]²⁵_D +9.4 (c 0.10, CH₃OH); UV (CH₃OH) λ_max (log ε) 299 (3.8) nm; IR (KBr) ν_max 3360, 2936, 2854, 1719 cm⁻¹; ¹H and ¹³C NMR see Table 1; HRESIMS m/z [M + H]⁺ 405.2268 (calcd for C₂₃H₃₄O₆⁺, 405.2272).
(3β,5β)-3,5-Dihydroxy-10β-14,15-epoxy-19-norbufa-20,22-dienolide (2). [α]²⁵_D +31.4 (c 0.10, CH₃OH); UV (CH₃OH) λ_max (log ε) 298 (3.7) nm; IR (KBr) ν_max 3381, 2946, 1700 cm⁻¹; ¹H and ¹³C NMR see Table 1; HRESIMS m/z [M + H]⁺ 387.2167 (calcd for C₂₃H₃₁O₅⁺, 387.2166).
19-Hydroxymarinobufagenin (3). [α]²⁵_D −15.4 (c 0.10, CH₃OH); UV (CH₃OH) λ_max (log ε) 298 (3.6) nm; IR (KBr) ν_max 3442, 2939, 1716, 1621 cm⁻¹; ¹H and ¹³C NMR see Table 1; HRESIMS m/z [M + H]⁺ 417.2264 (calcd for C₂₄H₃₃O₆⁺, 417.2272).
(3β,5β,10β,14β)-3,5,10,14-Tetrahydroxy-19-norbufa-20,22-dienolide-3-(14-hydroxytetradecanoic acid) ester (4). [α]²⁵_D +11.0 (c 0.20, CH₃OH); UV (CH₃OH) λ_max (log ε) 299 (3.7) nm; IR (KBr) ν_max 3485, 2870, 1757, 1680 cm⁻¹; ¹H and ¹³C NMR see Tables 2 and 3; HRESIMS m/z [M + Na]⁺ 653.4039 (calcd for C₃₇H₅₈O₈Na⁺, 653.4024).
Hellebrigenin-3-(14-hydroxytetradecanoic acid) ester (5). Amorphous powder; [α]²⁵_D +14.3 (c 0.20, MeOH); IR (KBr) ν_max 3447, 2938, 1705 cm⁻¹; UV (CH₃OH) λ_max (log ε) 295 (3.5) nm; ¹H and ¹³C NMR see Tables 2 and 3; HRESIMS m/z [M + Na]⁺ 665.4018 (calcd for C₃₈H₅₈O₈Na⁺, 665.4024).
Hellebrigenin-3-(14-hydroxydodecanoic acid) ester (6). [α]²⁵_D +6.7 (c 0.10, CH₃OH); UV (CH₃OH) λ_max (log ε) 299 (3.7) nm; IR (KBr) ν_max 3430, 2927, 1717 cm⁻¹; ¹H and ¹³C NMR see Tables 2 and 3; HRESIMS m/z [M + Na]⁺ 637.3713 (calcd for C₃₆H₅₄O₈Na⁺, 637.3716).
Hellebrigenin-3-(14-hydroxy-7Z-tetradecenoic acid) ester (7). [α]²⁵_D +17.2° (c 0.20, CH₃OH); UV (CH₃OH) λ_max (log ε) 298 (3.6) nm; IR (KBr) ν_max 3509, 2930, 2853, 1718 cm⁻¹; ¹H and ¹³C NMR see Tables 2 and 3; HRESIMS m/z [M + Na]⁺ 663.3921 (calcd for C₃₈H₅₆O₈Na⁺, 663.3867).
Telocinobufagin-3-(14-hydroxytetradecanoic acid) ester (8). [α]²⁵_D +10.5 (c 0.20, MeOH); UV (CH₃OH) λ_max (log ε) 297 (3.7) nm; IR (KBr) ν_max 3448, 2927, 1719 cm⁻¹; ¹H and ¹³C NMR data, see Tables 2 and 3; HRESIMS m/z [M + Na]⁺ 651.4250 (calcd for C₃₈H₆₀O₇Na⁺, 651.4232).
19-Hydroxytelocinobufagin-3-(14-hydroxytetradecanoic acid) ester (9). [α]²⁵_D +11.6 (c 0.20), UV (CH₃OH) λ_max (log ε) 299 (3.4) nm; IR (KBr) ν_max 3422, 2925, 1716 cm⁻¹; ¹H and ¹³C NMR data, see Tables 2 and 3; HRESIMS m/z [M + Na]⁺ 667.4189 (calcd for C₃₈H₆₀O₈Na⁺, 667.41804).
1β-Hydroxytelocinobufagin-3-(14-hydroxytetradecanoic acid) ester (10). [α]²⁵_D +5.2 (c 0.10, CH₃OH); UV (CH₃OH) λ_max (log ε) 297 (3.7) nm; IR (KBr) ν_max 3492, 2927, 1696 cm⁻¹; ¹H and ¹³C NMR data, see Tables 2 and 3; HRESIMS m/z [M + Na]⁺ 667.4218 (calcd for C₃₈H₆₀O₈Na⁺, 667.4180).
Hellebrigenin-3-(16-hydroxycyclohexadecanic acid) ester (11). [α]²⁵_D −10.2 (c 0.20, CH₃OH); UV (CH₃OH) λ_max (log ε) 299 (3.6) nm; IR (KBr) ν_max 3423, 2933, 1717, 1613 cm⁻¹; ¹H and ¹³C NMR see Tables 2 and 3; HRESIMS m/z [M + Na]⁺ 679.4566 (calcd for C₄₀H₆₄O₇Na⁺, 679.4544).
(3β,5β,10β,14β)-3,5,10,14-Tetrahydroxy-19-norbufa-20,22-dienolide (1). Colorless block crystals from MeOH, C₂₃H₃₆O₈, monoclinic, C2, a = 11.8755(3), b = 10.7481(2), c = 18.0005(4) Å, β = 104.253(2), V = 2226.84(9) ų, Z = 4, d_x = 1.314 Mg cm⁻³, μ (CuKα) = 0.813 mm⁻¹, F (000) = 952, 5.07° ≤ θ ≤ 62.88°. Data collection was performed at 173 K and 2590 unique reflections were collected until θ_max = 62.88°, in which 2522 reflections were observed [F² > 4σ(F²)]. The final R = 0.0376, R_w = 0.1018, S = 1.084 and CCDC 1487649.
(3β,5β)-3,5-Dihydroxy-10β-14,15-epoxy-19-norbufa-20,22-dienolide (2). Colorless block crystals from MeOH, C₂₃H₃₂O₆, orthorhombic, P2₁2₁2₁, a = 10.7367(5), b = 11.1799(5), c = 16.8458(9) Å, V = 2022.09(17) ų, Z = 4, d_x = 1.329 Mg cm⁻³, μ (CuKα) = 0.774 mm⁻¹, F (000) = 872, 4.85° ≤ θ ≤ 62.87°. Data collection was performed at 173 K and 2932 unique reflections were collected until θ_max = 62.87°, in which 2782 reflections were observed [F² > 4σ(F²)]. The final R = 0.0633, R_w = 0.1777, S = 1.061 and CCDC 1487650.

Dimethyl disulphide (DMDS) adducts and GC-MS analysis. DMDS adducts was widely and effectively used to locate double bond position in high fatty acids.^36,37 Before DMDS adduct, methanolysis 7 (5.5 mg) was performed to yield fatty acid methyl ester (FAME) according to the method described before.³⁸ Then the FAME was purified by a micro silica gel. Acetylation was followed to protect the hydroxy group in the side chain.²⁶ After a simple extraction procedure, dimethyl disulfide (DMDS, 200 μL) and 20 μL iodine (60 mg mL⁻¹ in Et₂O) were added to the solution. After incubated at 40 °C for 48 h, the reaction was quenched using saturated sodium sulfate. The solution was extracted with hexane and concentrated by rotate evaporation. The supernatant was subjected to EI-MS analysis. In the EI-MS spectrum (ESI Fig. S97†), it showed a precursor peak at m/z 392 (M⁺), two fragment ions at m/z 217 [(C₁₁H₂₁O₂S)⁺] and m/z 175 [(C₈H₁₅O₂S)⁺].

X-ray analysis. The structures were solved by using direct methods (SHELXTL version 2014) and reined by using full-matrix least-squares treatment on F². In the structure refinements, non-hydrogen atoms were reined anisotropically. Hydrogen atoms bonded to carbons were placed at geometrically ideal positions by using the ‘ride on’ method. Hydrogen atoms bonded to oxygen were located by employing the difference Fourier method and were included in the calculation of structure factors with isotropic temperature factors. CCDC 1487649 and 1487650.

Cell culture. Human breast cancer cell lines MCF-7 (estrogen-positive) and MDA-MB-231 (estrogen-negative), obtained from American Type Culture Collection (ATCC, Manassas, VA, USA), were cultured in RPMI-1640 and DMEM medium containing 10% fetal bovine serum and 1% (v/v) penicillin–streptomycin in a humidified atmosphere with 5% CO₂ at 37 °C.

Cytotoxicity assay. The cytotoxic activities of bufadienolides were assessed by MTT assay as previously described.³⁹ Cells (5 × 10³ per wells) were added into the 96-well plates. After overnight incubation, cells were exposed to different concentrations of bufadienolides for 48 h. After that, 30 μL of 5 mg mL⁻¹ MTT (Sigma-Aldrich, St. Louis, MO, USA) solution was added to each well and further incubated for 4 h, and then the formazen crystals were solubilized with 100 μL of DMSO. Finally, absorbance of each well was determined at 595 nm by automatic microplate reader (Beckman, DTX880, USA). Cells treated with medium containing 0.2% DMSO was considered as 100% viable. The concentration required to inhibit cell growth by 50% (IC₅₀) was calculated from survival curves.

Cell cycle analysis. MCF-7 cells (3 × 10⁵ per well) in 6-well plates were treated with compound 12 (0.02 μM) for 12 h, 24 h and 48 h. After that, cells were re-suspended in 500 μL of PBS containing 10 μg mL⁻¹ PI and 0.1 mg mL⁻¹ RNase, and then incubated at 37 °C for 30 min in darkness. DNA content was measured using Guava EasyCyte 6-2L flow cytometer (Millipore, Merck KGaA, Germany). Data was analyzed quantitatively with Guava incite software (Millipore, Merck KGaA, Germany).

Apoptosis analysis. Cell apoptosis analysis was performed by Annexin V-FITC/PI staining assay kit (Biouniquer Tech, Nanjing, China) according to the manufacturer's protocol. MCF-7 cells (3 × 10⁵ per well) were treated with compound 12 (0.02 μM) for 12 h, 24 h and 48 h. After that, cells were stained with Annexin V-FITC and PI solution and then examined by Guava EasyCyte 6-2L flow cytometry (Millipore, Merck KGaA, Germany). Data was analyzed quantitatively with Guava InCyte software (Millipore, Merck KGaA, Germany).

Western blot. Cyclin B1, Cdc2, phospho-Cdc2 (Thy15), caspase-3, cleaved caspase-3, caspase-9, cleaved caspase-9, PARP, cleaved PARP and β-actin (Cell Signaling Technology, Beverly, MA, USA) were evaluated by western blot. MCF-7 cells (1 × 10⁶ per dish) were cultured with compound 12 (0.02 μM) for 0 h, 12 h, 24 h and 48 h, cells were harvested and then lysed in RIPA buffer (0.5 M DTT, 0.1 M PMSF and 20× phosphatase inhibitor) for 30 min on ice. After centrifugation at 14 000 × g at 4 °C for 15 min, supernatants were collected as total cellular proteins and stored at −80 °C. Protein concentration was determined using BCA protein assay kit (Thermo Fisher Scientific, Rockford, IL, USA). Electrophoresis and immunoblotting analysis was carried out as described previously.

Acknowledgements

The research work was financially supported by National Natural Science Foundation of China (81573315, 81102518, and 81473336), Guangdong Province Natural Science Fund (2015A030313313), Guangzhou Industry-University Collaborative Innovation Major Projects (201508030016) and Technology Program for Dr Hai-Yan Tian (201506010020).

Notes and references

1. G. P. McGlacken and I. J. S. Fairlamb, Nat. Prod. Rep., 2005, 22, 369.
2. H. M. Gao, R. Popescu, B. Kopp and Z. M. Wang, Nat. Prod. Rep., 2011, 28, 953.
3. L. Krenn and B. Kopp, Phytochemistry, 1998, 48, 1.
4. D. L. Wang, F. H. Qi, W. Tang and F. S. Wang, Chem. Biodiversity, 2011, 8, 559.
5. S. S. Lee, F. Derguini, R. C. Bruening, K. Nakanishi, E. T. Wallick, T. Akizawa, C. S. Rosenbaum and V. P. Butler Jr, Heterocycles, 1994, 39, 669.
6. T. Akizawa, T. Mukai, M. Matsukawa, M. Yoshioka, J. F. Morris and V. P. Butler Jr, Chem. Pharm. Bull., 1994, 42, 754.
7. M. Matsukawa, T. Mukai, T. Akizawa, S. Miyatake, M. Yoshioka, V. P. Butler Jr and J. F. Morris, J. Nat. Prod., 1998, 61, 1476.
8. D. Lichtstein, S. Kachalsky and J. Deutsch, Life Sci., 1986, 38, 1261.
9. D. Lichtstein, I. Gati, T. Babila, E. Haver and U. Katz, Biochim. Biophys. Acta, 1991, 1073, 65.
10. D. Lichtstein, I. Gati, E. Haver and U. Katz, Life Sci., 1992, 51, 119.
11. D. Lichtstein, I. Gati and H. J. Ovadia, Cardiovasc. Pharmacol., 1993, 22, S102.
12. V. P. Butler, J. F. Morris, T. Akizawa, M. Matsukawa, P. Keating, A. Hardart and I. Furman, Am. J. Physiol., 1996, 271, R325.
13. R. A. Hayes, M. R. Crossland, M. Hagman, R. J. Capon and R. Shine, J. Chem. Ecol., 2009, 35, 391.
14. M. R. Crossland, T. Haramura, A. A. Salim, M. Hagman, R. J. Capon and R. Shine, Proc. R. Soc. B, 2012, 279, 3436.
15. M. Hagman, R. A. Hayes, R. J. Capon and R. Shine, Funct. Ecol., 2009, 23, 126.
16. H. Y. Tian, L. Wang, X. Q. Zhang, D. M. Zhang, Y. Wang, R. W. Jiang and W. C. Ye, Steroids, 2010, 75, 884.
17. H. Y. Tian, L. Wang, X. Q. Zhang, Y. Wang, D. M. Zhang, R. W. Jiang, Z. Liu, J. S. Liu, Y. L. Li and W. C. Ye, Chem.–Eur. J., 2010, 16, 10989.
18. H. Y. Tian, P. W. Zhang, J. S. Liu, D. M. Zhang, X. Q. Zhang, R. W. Jiang and W. C. Ye, Chin. Chem. Lett., 2014, 39, 841.
19. H. Y. Tian, S. L. Luo, J. S. Liu, L. Wang, Y. Wang, D. M. Zhang, X. Q. Zhang, R. W. Jiang and W. C. Ye, J. Nat. Prod., 2013, 76, 1842.
20. P. W. Zhang, R. W. Jiang, W. C. Ye and H. Y. Tian, Zhongguo Zhongyao Zazhi, 2014, 39, 84.
21. R. Fanie, V. Heerden and R. Vleggaar, Magn. Reson. Chem., 1988, 26, 464.
22. W. F. Reynolds, A. Maxwell, B. Telang, K. Bedaisie and G. Ramcharan, Magn. Reson. Chem., 1995, 33, 412.
23. K. Shimada and T. Nambara, Chem. Pharm. Bull., 1979, 27, 1881.
24. (a) P. W. Zhang, H. J. Tang, H. Y. Tian, R. R. Zhang and R. W. Jiang, Chin. J. Struct. Chem., 2014, 33, 1123; (b) Y. H. XU, J. Xu, X. Y. Jiang, Z. H. Chen, Z. J. Xie, R. W. Jiang and F. Feng, Chin. J. Struct. Chem., 2016, 35, 1024.
25. C. K. Lu, G. H. Lee, R. Huang and H. N. Chou, Tetrahedron Lett., 2001, 42, 1713.
26. W. K. Zhang, J. K. Xu, X. Q. Zhang, X. S. Yao and W. C. Ye, Chem. Phys. Lipids, 2007, 148, 77.
27. R. Radeglia, H. Poleschner and G. Haufe, Magn. Reson. Chem., 1993, 31, 1054.
28. (a) H. J. Tang, L. J. Ruan, H. Y. Tian, G. P. Liang, W. C. Ye, E. Hughes, M. Esmann, N. U. Fedosova, T. Y. Chung, J. T. Tzen, R. W. Jiang and D. A. Middleton, Sci. Rep., 2016, 6, 29155; (b) M. Y. L. Banuls, E. Urban, M. Gelbcke, F. Dufrasne, B. Kopp, R. Kiss and M. Zehl, J. Nat. Prod., 2013, 76, 1078.
29. R. Esche, L. Muller and K. H. Engel, J. Agric. Food Chem., 2013, 61, 11636.
30. G. Janer, A. Lyssimachou, J. Bachmann, J. Oehlmann, U. Schulte-Oehlmann and C. Porte, Steroids, 2006, 71, 435–444.
31. V. Vihma and M. Tikkanen, J. Steroid Biochem. Mol. Biol., 2011, 124, 65.
32. A. H. Ludewig and F. C. Schroeder, Worm Book, 2013, 18, 1.
33. Y. Zhang, Y. K. Qiu, J. Y. Chen, K. N. Bo, Y. T. Jiang and Y. H. Pei, J. Shenyang Pharm. Univ., 2007, 24, 484.
34. G. Schmeda-Hirschmanna, C. Quispea, C. Theodulozb, P. Teixeira de Sousa Jr and C. Parizottoc, J. Ethnopharmacol., 2014, 155, 1076.
35. Y. Kamano, A. Yamashita, T. Nogawa, H. Morita, K. Takeya, H. Itokawa, T. Segawa, A. Yukita, K. Saito, M. Katsuyama and G. R. Pettit, J. Med. Chem., 2002, 45, 5440.
36. M. K. Dowd, J. Am. Oil Chem. Soc., 2012, 89, 1599.
37. W. Biedermann, E. Lücker, J. Pörschmann, S. Lachhab, U. Truyen and A. Hensel, Anal. Bioanal. Chem., 2004, 379, 1031.
38. S. Öksüz and G. Topcu, Phytochemistry, 1987, 26, 3082.
39. D. M. Zhang, Y. Wang, M. K. Tang, Y. W. Chan, H. M. Lam and W. C. Ye, Biochem. Biophys. Res. Commun., 2007, 362, 759.

---

Footnotes

† Electronic supplementary information (ESI) available: HRESIMS, IR, UV, 1D and 2D NMR spectra of 1–11. CCDC 1487649 and 1487650. For ESI and crystallographic data in CIF or other electronic format see DOI: 10.1039/c6ra18676a

‡ These authors contributed equally to this work.

---