Flavans with cytotoxic activity from the stem and root bark of Daphne giraldii

Abstract

Thirteen new flavan compounds named daphnegiravans A–M (1–13) and eight known analogues (14–21) were isolated from the stem and root bark of Daphne giraldii. Their structures were established by comprehensive analysis of NMR data and CD spectra. Five human cancer cell lines (MCF-7, Bcap37, HepG2, Hep3B and A549) were used to evaluate the antitumor activity of all the isolates and their structure–activity relationships were also discussed. Interestingly, prenylated and some methoxy flavans exhibited the highest activities against Hep3B compared with the other cell lines, especially 3 and 9–12 with IC₅₀ values ranging from 5.15 to 9.66 μM. Furthermore, flow cytometry analysis indicated that 3 and 9–11 possessing a 2,2-dimethylpyran moiety in ring B induced G2/M phase arrest in Hep3B cells, while 12 with different structural features effectively inhibited cell proliferation by evoking apoptotic cell death. The reactive oxygen species (ROS) generation might be responsible for the induction of arrest and apoptosis.

---

Introduction

According to the World Health Organization fact sheets updated in 2015, cancers are among the leading causes of morbidity and mortality worldwide regardless of developed or developing countries.¹ Although conventional chemotherapy is an effective form of cancer therapy, the toxic and adverse effects caused during therapeutic courses are inevitable and bring great suffering to patients.² Multidrug resistance can also result in treatment failure.³ Traditional Chinese medicine has historically played a significant role in preventing and curing cancer. The active components derived from them with milder side effects and improved tolerance might show the advantages over synthetic chemical drugs.^4,5 At present, a number of the chemotherapeutic agents used clinically are obtained from natural products and their derivatives.^6,7 Consequently, scientists are enthusiastically searching for novel and effective antitumor pharmaceuticals from natural resources.

The genus Daphne (Thymelaeaceae) is distributed widely around the world, and it comprises approximately 90 species, 44 of which are native to China.⁸ Previous phytochemical investigations into this genus have led to the isolation of many promising constituents. Some of them such as coumarins, flavonoids and daphnane diterpenoids have been found to possess marked cytotoxic activities,^9–12 which has made Daphne especially attractive in a search to discover lead compounds. Daphne giraldii Nitsche. is a toxic shrub which belongs to the genus Daphne.¹³ Its rhizome bark, generally called ‘ZuShima’ in Chinese, is reported to have exhibit anti-inflammatory, anti-tumor, anti-malarial and anti-fertility effects.^14,15 Flavonoids including characteristic flavans and biflavones are the main types of compounds in D. giraldii.^16,17 In our ongoing search for cytotoxic natural products from the genus Daphne, a series of flavan derivatives were isolated from the stem and root bark of D. giraldii and found to inhibit cancer cell proliferation in cell-based assays.

Here, we report the isolation and structural elucidation of thirteen new flavans daphnegiravans A–M (1–13) along with eight known ones (14–21), as well as measuring the in vitro cytotoxicities of all the isolates against five human cancer cell lines. Moreover, compounds 3 and 9–12 were also evaluated for further cell cycle distribution, apoptosis and ROS level of Hep3B cells.

Results and discussion

The 95% EtOH extract of the stem and root bark from D. giraldii was fractioned by a series of chromatographic separations, including silica gel, MCI gel and reversed-phase chromatography to yield twenty-one pure flavan compounds (1–21) (Fig. 1).

Fig. 1 Chemical structures of compounds 1–21.

Structural elucidation of the new compounds (1–13)

Compound 1 was obtained as a yellow amorphous solid. The ¹³C NMR data (Table 1) as well as the HRESIMS ion peak at m/z 309.1479 [M + H]⁺ (calcd 309.1485) suggested a molecular formula of C₂₀H₂₀O₃ for 1, with eleven indices of hydrogen deficiency. The UV spectrum exhibited maximum absorption bands at λ_max 229, 284 and 315 nm. The infrared spectrum suggested the presence of a hydroxyl group (3418 cm⁻¹) and an aromatic ring (1621 and 1461 cm⁻¹). Its ¹H NMR spectrum (Table 2) showed typical signals of a flavan skeleton at δ_H 4.89 (1H, dd, J = 10.0, 1.7 Hz, H-2), 2.05, 1.91 (each 1H, m, H-3), 2.80, 2.59 (each 1H, m, H-4) and one ABX spin system [δ_H 6.85 (1H, d, J = 8.1 Hz, H-5), 6.29 (1H, dd, J = 8.1, 2.4 Hz, H-6), 6.20 (1H, d, J = 2.4 Hz, H-8)] in ring A. Moreover, three aromatic signals at δ_H 7.10 (1H, d, J = 2.0 Hz, H-2′), 6.73 (1H, d, J = 8.1 Hz, H-5′) and 7.13 (1H, dd, J = 8.1, 2.0 Hz, H-6′) were assigned to another ABX-type coupling system in ring B by the HMBC (Fig. 2) correlations of H-2′, H-6′/C-2 and H-5′/C-1′, C-3′, C-4′. There was also a 2,2-dimethylpyran group of signals at δ_H 5.74 (1H, d, J = 9.8 Hz, H-3′′), 6.41 (1H, d, J = 9.8 Hz, H-4′′) and 1.37 (6H, s, CH₃-5′′,6′′) and linked at C-3′,4′ through the HMBC cross peaks of H-4′′/C-2′, C-3′ and C-4′. According to the literature, flavans have a half-chair conformation with the C₂-phenyl group equatorial and follow the P-/M-helicity rule of the O-heterocyclic ring.¹⁸ Their absolute configurations were confirmed by negative/positive Cotton effects (CEs) within the ¹L_b band. In our study, the circular dichroism (CD) curve of 1 showed a negative CE at Δε₂₈₄ −0.84, which established the 2S-configuration. Therefore, compound 1 was named daphnegiravan A.

Table 1 ¹³C NMR spectroscopic data for compounds 1–13

No	1^a	2^b	3^a	4^b	5^c	6^b	7^d	8^d	9^b	10^b	11^a	12^c	13^a
a Data were measured in DMSO-d6 at 100 MHz.b Data were measured in DMSO-d6 at 150 MHz.c Data were measured in methanol-d4 at 150 MHz.d Data were measured in CDCl3 at 75 MHz.
2	76.6	77.0	76.6	74.6	75.8	74.6	75.0	74.9	73.7	73.6	75.4	79.4	76.1
3	29.2	29.8	29.3	29.7	30.9	29.6	29.5	28.5	29.0	28.4	28.8	31.8	29.2
4	23.8	23.7	23.8	24.0	26.1	24.6	25.5	25.3	24.2	24.5	24.0	25.4	23.7
5	129.8	129.8	129.8	129.8	131.0	129.8	130.2	130.1	129.9	129.9	129.8	131.0	129.8
6	108.0	108.1	107.9	108.0	109.3	108.0	107.9	108.0	108.1	108.0	108.1	109.1	108.1
7	156.5	156.5	156.4	156.4	157.6	156.5	154.9	155.0	156.4	156.4	156.5	157.6	156.5
8	102.8	102.8	102.7	102.8	104.1	102.8	103.7	103.7	102.8	102.8	102.7	104.0	102.8
9	155.4	155.3	155.4	155.8	157.3	155.8	156.6	156.4	155.5	155.7	155.4	157.1	155.2
10	112.1	112.1	112.2	112.1	114.1	112.2	114.3	114.3	112.3	112.3	112.2	114.3	112.1
1′	133.9	136.5	133.7	135.5	134.2	135.7	130.7	129.6	128.6	130.9	125.9	138.1	132.7
2′	124.2	118.6	114.1	121.9	134.7	121.8	110.1	116.6	119.7	120.0	127.3	117.2	126.4
3′	120.7	127.7	145.2	139.0	145.4	139.3	143.7	140.6	140.4	142.1	142.6	129.6	121.9
4′	151.9	153.7	139.4	143.4	148.9	143.5	140.9	142.8	140.1	140.5	137.0	154.1	160.4
5′	115.6	110.9	121.4	126.7	119.4	126.2	119.8	124.8	121.6	119.1	116.5	125.9	117.4
6′	127.0	122.6	114.7	109.1	114.7	109.1	128.9	118.1	115.8	114.4	118.8	122.8	134.3
1′′				24.6	26.2	24.0			74.9	73.1
2′′	76.1	72.8	75.7	123.8	123.4	123.8	74.6	74.9	131.8	31.9	146.3	69.9
3′′	131.2	160.4	131.2	129.9	132.7	129.9	33.2	32.8	118.7	19.1	106.1	165.2
4′′	121.8	104.1	122.2	25.4	25.9	25.4	20.3	19.3	26.9	26.3		101.7
5′′	27.6	25.0	27.4	17.7	18.1	17.7	26.4	26.9	27.2	26.7		29.1
6′′	27.6	25.0	27.4				27.3	26.9				29.1
1′′′							26.9	28.4				29.2
2′′′				72.8	81.3	145.3	123.2	122.6	75.9	75.7	76.9	123.0
3′′′				159.2	49.7	107.2	131.6	132.4	131.6	130.2	129.3	133.9
4′′′				104.6	194.8		25.7	25.9	122.3	122.4	122.1	25.9
5′′′				25.0	26.7		18.1	17.9	27.4	27.5	27.6	18.0
6′′′				25.0	26.7				27.5	27.7	27.6
OCH3-3′′		50.3		50.2
CHO-3′													191.3

---

Table 2 ¹H NMR spectroscopic data for compounds 1–8

No	1^a	2^b	3^a	4^b	5^c	6^b	7^d	8^d
a Recorded in DMSO-d6 and 400 MHz.b Recorded in DMSO-d6 and 600 MHz.c Recorded in methanol-d4 and 600 MHz.d Recorded in CDCl3 and 300 MHz. o: the abbreviation for overlapped.
2	4.89 dd (10.0, 1.7)	5.12 dd (10.1, 1.8)	4.83 dd (9.8, 1.7)	5.08 dd (10.2, 1.7)	5.04 dd (10.2, 1.7)	5.09 dd (10.0, 1.6)	5.09 dd (9.9, 3.0)	5.05 m
3	2.05 m, 1.91 m	2.12 m, 1.98 m	2.03 m, 1.86 m	2.07 m, 1.86 m	2.11 m, 1.93 m	2.09 m, 1.89 m	2.09 m, 2.05 m	2.14 m, 2.11 m
4	2.80 m, 2.59 m	2.84 m, 2.60 m	2.79 m, 2.58 m	2.83 m, 2.66 m	2.88 m, 2.73 m	2.83 m, 2.66 m	2.90 m, 2.78 m	2.92 m, 2.79 m
5	6.85 d (8.1)	6.86 d (8.2)	6.84 d (8.2)	6.88 d (8.2)	6.89 d (8.2)	6.88 d (8.2)	6.93 d (8.0)	6.94 d (8.4)
6	6.29 dd (8.1, 2.4)	6.30 dd (8.2, 2.3)	6.27 dd (8.2, 2.4)	6.29 dd (8.2, 2.4)	6.33 dd (8.2, 2.3)	6.29 dd (8.2, 2.3)	6.38 dd (8.0, 2.4)	6.39 dd (8.4, 2.4)
8	6.20 d (2.4)	6.22 d (2.3)	6.18 d (2.4)	6.19 d (2.4)	6.26 d (2.3)	6.19 d (2.3)	6.36 o	6.37 o
2′	7.10 d (2.0)	7.64 d (1.7)	6.71 d (1.8)				6.97 s
5′	6.73 d (8.1)	7.56 d (8.5)
6′	7.13 dd (8.1, 2.0)	7.33 dd (8.5, 1.7)	6.57 d (1.8)	7.13 s	7.45 s	7.18 s		6.81 s
1′′				3.49 dd (15.1, 7.0), 3.41 dd (15.0, 6.5)	3.53 dd (14.7, 7.0), 3.44 o	3.50 dd (15.1, 7.0), 3.42 dd (15.0, 6.7)
2′′				5.06 o	5.10 t 7.0	5.07 o
3′′	5.74 d (9.8)		5.72 d (9.8)				1.84 t (6.6)	1.84 t (7.0)
4′′	6.41 d (9.8)	6.87 s	6.35 d (9.8)	1.62 s	1.68 s	1.62 s	2.69 t (6.8)	2.89 m, 2.72 m
5′′	1.37 s	1.55 s	1.36 s	1.68 s	1.72 s	1.68 s	1.33 s	1.36 s
6′′	1.37 s	1.55 s	1.36 s				1.37 s	1.37 s
1′′′							3.36 dd (15.6, 6.6), 3.22 dd (15.6, 5.7)	3.34 d (7.0)
2′′′						7.92 d (2.0)	4.98 t (6.6)	5.33 m
3′′′					2.76 o	6.88 d (2.0)
4′′′				6.78 s			1.67 s	1.72 s
5′′′				1.56 s	1.50 s		1.70 s	1.72 s
6′′′				1.56 s	1.50 s
OH-7	9.19 br s	9.38 br s	9.16 br s	9.17 br s		9.24 br s
OCH3-3′′		2.99 s		3.00 s

---

Fig. 2 Key HMBC correlations of compounds 1, 5, 9 and 12.

Compound 2 possessed the molecular formula C₂₁H₂₂O₄, as determined from a sodiated molecular ion in the positive HRESIMS at m/z 361.1411 [M + Na]⁺ combined with its NMR data. The ¹H and ¹³C spectroscopic data of 2 were similar to those of 1 except for the presence of an additional methoxy group, suggesting that 2 was a 3′′-OCH₃ derivative of 1, which was further supported by the downfield shift of C-3′′ (δ_C 160.4) and the HMBC long range correlations of OCH₃-3′′/C-2′′ and CH₃-5′′,6′′/C-3′′, OCH₃-3′′. Its absolute 2S configuration was determined by CD spectroscopy. Based on these examinations, compound 2 was given the name daphnegiravan B.

The molecular formula of daphnegiravan C (3) was established as C₂₀H₂₀O₄ by HRESIMS (m/z 347.1260 [M + Na]⁺) which showed that the molecule had eleven degrees of unsaturation. A comparison of the NMR spectroscopic data of 3 with those of 1 revealed that the only difference between them was the typical NMR signals of a symmetric 1′,3′,4′,5′-tetrasubstituted phenyl moiety in 3 rather than the 1′,3′,4′-trisubstituted phenyl groups in 1, as confirmed by the presence of a hydroxy group at C-3′ [δ_H 8.93 (1H, brs, OH-3′)] and two meta-coupling signals belonging to ring B [δ_H 6.71 (1H, d, J = 1.8 Hz, H-2′); 6.57, (1H, d, J = 1.8 Hz, H-6′)] in the ¹H NMR spectrum (Table 2). In addition, these findings were supported by the HMBC correlations of H-6′/C-2, C-2′, C-4′′. The stereochemistry of C-2 involved an S-configuration from the negative CE. Thus, the structure of 3 was determined assigned as daphnegiravan C.

Compound 4 was isolated as a yellow solid with the molecular formula C₂₆H₃₀O₅ (HRESIMS, m/z 445.1993 [M + Na]⁺). Comparison of the ¹H and ¹³C NMR data of 4 with those of 1 and 3 showed that it had a similar flavan skeleton, the major difference being the presence of a prenyl group [δ_H 3.49 (1H, dd, J = 15.1, 7.0 Hz, H_a-1′′); 3.41 (1H, dd, J = 15.0, 6.5 Hz, H_b-1′′); 5.06 (1H, overlapped, H-2′′); 1.62 (3H, s, CH₃-4′′) and 1.68 (3H, s, CH₃-5′′)] (Table 2) at C-2′ in 4. This was further supported by the key HMBC correlations of H-2′′/C-1′, C-2′, C-3′ and H-6′/C-2, C-2′, C-4′′′. From the negative CE at 283 nm in its CD spectrum, compound 4 was confirmed to have an S configuration, and was named daphnegiravan D.

The HRESIMS spectra of 5 and 6 exhibited quasimolecular ion peaks at m/z 409.2002 [M + H]⁺ and 373.1418 [M + Na]⁺ corresponding to molecular formulae of C₂₅H₂₈O₅ and C₂₂H₂₂O₄, respectively. Their NMR data (Tables 1 and 2) were found to be structurally similar to that of 4, whereas the 2,2-dimethyl-3-methoxylpyran group present in 4 was replaced by a 4-carbonyl-2,2-dimethyldihydropyran moiety [δ_H 2.76 (2H, overlapped, H-3′′′); 1.50 (6H, s, CH₃-5′′′, 6′′′)] in 5 and a furan group [δ_H 7.92 (1H, d, J = 2.0 Hz, H-2′′′); 6.88 (1H, d, J = 2.0 Hz, H-3′′′)] in 6. Besides, the respective HMBC cross-peaks of H-6′/C-2, C-2′, C-4′′′ (δ_C 194.8) for 5 (Fig. 2) and H-6′/C-2, C-2′, C-3′′′ (δ_C 107.2) for 6 both supported the idea that the functional groups above were fused with the C-4′ and 5′ of ring B. The absolute configurations at C-2 were deduced to be S from CD measurements (Δε₂₈₇ −0.83 of 5; Δε₂₈₂ −0.77 of 6). As a result, compounds 5 and 6 were named daphnegiravans E and F.

Compounds 7 and 8 displayed the same molecular formula of C₂₅H₃₀O₄ as inferred from the [M + Na]⁺ ion peak (m/z 417.2034 in 7; m/z 417.2037 in 8) suggestive of eleven indices of hydrogen deficiency. The ¹H and ¹³C spectroscopic data (Tables 1 and 2) of 7 and 8 showed that they shared the same 7-hydroxyflavan skeleton with a phenolic hydroxyl group, a 2,2-dimethyldihydropyran ring and a prenyl group in ring B. However, they were found to differ in terms of the substitution pattern of the aromatic ring according to 2D NMR. The HMBC spectrum of 7 displayed correlations of OH-3′/C-2′, C-3′, C-4′, H-4′′/C-4′, C-5′, C-6′ and H-1′′′/C-1′, C-5′, C-6′, which indicated that the hydroxyl, 2,2-dimethyldihydropyran and prenyl groups were linked at C-3′, C-4′,5′ and C-6′, respectively. In contrast, the position of the 2,2-dimethyldihydropyran ring of 8 changed to C-2′,3′, also the hydroxyl and prenyl groups were located at C-4′, C-5′ as deduced from the HMBC signals for H-4′′/C-1′, C-2′, C-3′ and H-1′′′/C-4′, C-5′, C-6′. The absolute configurations at C-2 were confirmed as S by comparison of the CD data with literature values (Δε₂₈₄ −0.64 of 7; Δε₂₈₆ −0.60 of 8). Based on the above analysis, compounds 7 and 8 were named daphnegiravans G and H.

Compounds 9–11 were obtained as yellow amorphous solid. Their molecular formulae were assigned as C₂₅H₂₆O₄, C₂₅H₂₈O₄ and C₂₂H₂₀O₄ by HRESIMS analysis (m/z 413.1717 [M + Na]⁺ in 9; m/z 415.1878 [M + Na]⁺ in 10; m/z 371.1259 [M + Na]⁺ in 11). The NMR data of 9–11 exhibited the signal patterns characteristic of a 7-hydroxyflavan with a 2,2-dimethylpyran moiety connected to C-4′,5′ very similar to those of 1. In addition, 9–11 had a 2,2-dimethylpyran [δ_H 5.79 (1H, d, J = 9.9 Hz, H-3′′), 6.62 (1H, d, J = 9.9 Hz, H-4′′), 1.37 (3H, s, CH₃-5′′) and 1.39 (3H, s, CH₃-6′′)], a 2,2-dimethyldihydropyran [δ_H 1.76 (1H, t, J = 6.9 Hz, H-3′′), 2.78, 2.66 (each 1H, m, H-4′′), 1.26 (3H, s, CH₃-5′′) and 1.29 (3H, s, CH₃-6′′)] and a furan [δ_H 7.94 (1H, d, J = 2.0 Hz, H-2′′) and 7.00 (1H, d, J = 2.0 Hz, H-3′′)] group, respectively (Table 3). In the HMBC spectra, correlations of H-4′′/C-1′, C-2′, C-3′ for 9 (Fig. 2), H-4′′/C-1′, C-2′, C-3′ for 10 and H-3′′/C-1′, C-2′, C-3′ for 11 indicated that the attachments of these groups were all at the C-2′,3′ in ring B. Analysis of the CD data (ESI Fig. S9.8, S10.8 and S11.7†) of 9–11 revealed that they had the same S absolute configuration at C-2. Consequently, compounds 9–11 were named daphnegiravans I–K.

Table 3 ¹H NMR spectroscopic data for compounds 9–13

No	9^a	10^a	11^b	12^c	13^b
a Recorded in DMSO-d6 and 300 MHz.b Recorded in DMSO-d6 and 400 MHz.c Recorded in methanol-d4 and 300 MHz.
2	5.12 dd (10.0, 2.0)	5.00 dd (10.0, 1.5)	5.23 dd (10.1, 1.9)	5.04 dd (9.9, 2.1)	5.00 dd (10.0, 1.8)
3	2.04 m, 1.90 m	2.08 m, 1.88 m	2.10 m, 2.00 m	2.16 m, 2.02 m	2.07 m, 1.90 m
4	2.89 m, 2.62 m	2.85 m, 2.64 m	2.90 m, 2.65 m	2.87 m, 2.65 m	2.81 m, 2.59 m
5	6.87 d (8.1)	6.87 d (8.1)	6.89 d (8.2)	6.86 d (8.1)	6.86 d (8.2)
6	6.29 dd (8.1, 2.4)	6.28 dd (8.1, 2.4)	6.31 dd (8.2, 2.2)	6.32 dd (8.1, 2.4)	6.29 dd (8.2, 2.4)
8	6.19 d (2.4)	6.18 d (2.4)	6.21 d (2.2)	6.28 d (2.4)	6.20 d (2.4)
2′				7.40 d (1.5)	7.68 d (2.2)
5′					7.02 d (8.6)
6′	6.70 s	6.67 s	7.03 s	7.08 d (1.5)	7.56 dd (8.6, 2.2)
2′′			7.94 d (2.0)
3′′	5.79 d (9.9)	1.76 t (6.9)	7.00 d (2.0)
4′′	6.62 d (9.9)	2.78 m 2.66 m		6.61 s
5′′	1.37 s	1.26 s		1.63 s
6′′	1.39 s	1.29 s		1.63 s
1′′′				3.60 d (7.4)
2′′′				5.39 m
3′′′	5.74 d (9.9)	5.65 d (9.6)	5.73 d (9.8)
4′′′	6.36 d (9.9)	6.34 d (9.6)	6.52 d (9.8)	1.73 s
5′′′	1.36 s	1.35 s	1.45 s	1.79 s
6′′′	1.38 s	1.35 s	1.45 s
OH-7	9.01 brs	9.06 brs	9.17 brs		9.19 brs
CHO-3′					10.28 s

---

The molecular formula of compound 12 was determined as C₂₅H₂₈O₄ based on HRESIMS analysis (m/z 415.1878 [M + Na]⁺, calcd 415.1880). Its ¹³C NMR spectrum (Table 1) was consistent with that of a flavan, with fifteen signals attributed to fifteen carbons of the flavan skeleton, with five signals assigned to the prenyl group and four to the 2,2-dimethylpyran ring. Apart from these characteristic signals, the absence of olefinic protons and a downfield shift of C-3′′ (δ_C 165.2) were also observed in the 1D NMR suggesting that a hydroxyl group was attached at C-3′′. Resonances for two meta-coupling protons [δ_H 7.40 (1H, d, J = 1.5 Hz, H-2′) and 7.08 (1H, d, J = 1.5 Hz, H-6′)], as well as the HMBC cross-peaks of H-4′′/C-2′, C-3′, H-1′′′/C-4′, C-5′, C-6′ implied that the positions of the 2,2-dimethylpyran ring and the prenyl group were C-3′,4′ and C-5′ respectively (Fig. 2). Accordingly, a planar structure was established for 12, and the negative CE at 286 nm (Δε −0.77) supported the 2S-configuration. Compound 12 was then named daphnegiravan L.

Compound 13, a brown solid, was deduced to have the molecular formula C₁₆H₁₄O₄, as determined by the observed ion at m/z 293.0773 [M + Na]⁺ (calcd 293.0784) in its HRESIMS. The ¹H and ¹³C NMR spectra (Tables 1 and 3) displayed the presence of an aldehyde group at δ_H 10.28 (1H, s) and the corresponding carbon signal at δ_C 191.3, with the remaining signals being a simple 7,4′-dihydroxyflavan. Moreover, the HMBC correlations of the aldehyde proton CHO-3′/C-2′, C-3′ and the aromatic proton H-2′ [δ_H 7.68 (1H, d, J = 2.2 Hz)]/CHO-3′ (δ_C 191.3) indicated that the aldehyde group was located at C-3′. Using the same methods as those used for 1–12 permitted the assignment of the absolute configuration of 13, which was named daphnegiravan M.

In addition, eight previously reported compounds were assigned as (2S)-7,4′-dihydroxy-3′-prenylflavan (14),¹⁹ (2S)-kazinol I (15),²⁰ (2S)-7,4′-dihydroxyflavane (16),²¹ (2S)-7,4′-dihydroxy-3′-methoxyflavan (17),²² (2S)-7,3′-dihydroxy-4′-methoxyflavan (18),²² (2S)-7-hydroxy-3′,4′-dimethoxyflavan (19),²³ (2S)-4′-hydroxy-7-methoxyflavan (20)²⁴ and (2S)-7,3′-dimethoxy-4′-hydroxyflavan (21).²⁴ They were all obtained from genus Dapnhe for the first time.

In vitro inhibition of cell proliferation

Extensive pharmacological studies have shown that flavans strongly inhibit cancer cell proliferation and have weak cytotoxic effects on normal cells.^25–27 Therefore, compounds 1–21 isolated from D. giraldii were examined for their cytotoxic activities on five human cancer cells, namely MCF-7, Bcap37 (breast adenocarcinoma), HepG2, Hep3B (hepatocellular carcinoma) and A549 (lung adenocarcinoma) with 5-fluorouracil (5-Fu) serving as a positive control. The MTT method was used to evaluate their activities and the IC₅₀ values are summarized in Table 4. It was clear that some of the tested compounds exhibited potential cytotoxic activities. Compounds 7, 9, 11–12 were found to moderately inhibit the proliferation of all five cancer cells (IC₅₀ < 40 μM). In addition, prenylated (1–12, 14–15) and methoxy flavans (17–19, 21) exhibited the highest cytotoxicities against Hep3B compared with the other cell lines, while 13 with an aldehyde group had a selective effect on A549. Among them, 9 and 12 displayed potent inhibition of Hep3B proliferation even when compared with the positive control with IC₅₀ values of 5.15 and 5.63 μM, respectively.

Table 4 Cytotoxic activities of compounds 1–21 against five human cancer cell lines^a

Compound	MCF-7	Bcap37	HepG2	Hep3B	A549
a Results are expressed as IC50 means ± SD in μM, 5-fluorouracil was used as the positive controls. The experiments were performed three times.
1	47.09 ± 1.24	70.42 ± 3.25	22.36 ± 0.66	19.34 ± 0.12	48.18 ± 1.11
2	>100	>100	>100	46.12 ± 1.96	>100
3	67.45 ± 2.16	>100	34.18 ± 1.11	7.97 ± 0.51	37.74 ± 0.45
4	>100	>100	89.46 ± 4.28	12.94 ± 0.93	>100
5	>100	>100	>100	92.25 ± 4.56	>100
6	>100	>100	>100	47.08 ± 1.17	>100
7	29.11 ± 0.88	27.62 ± 0.33	27.22 ± 0.51	22.23 ± 0.89	23.94 ± 0.93
8	>100	>100	51.99 ± 2.82	44.85 ± 1.62	>100
9	20.93 ± 0.32	33.13 ± 0.93	21.92 ± 0.38	5.15 ± 0.11	17.39 ± 0.23
10	88.29 ± 4.76	>100	36.99 ± 1.24	9.29 ± 0.55	45.18 ± 0.76
11	24.68 ± 0.58	37.38 ± 0.87	18.14 ± 0.50	9.66 ± 0.28	32.77 ± 0.99
12	17.78 ± 0.41	38.41 ± 0.11	20.91 ± 0.19	5.63 ± 0.23	39.39 ± 1.88
13	51.79 ± 2.51	>100	73.35 ± 1.89	31.04 ± 0.84	18.27 ± 0.62
14	52.41 ± 3.67	90.08 ± 4.95	33.54 ± 0.44	28.25 ± 0.39	56.67 ± 2.21
15	>100	>100	>100	31.34 ± 0.40	>100
16	>100	>100	>100	>100	>100
17	>100	>100	>100	49.17 ± 1.81	>100
18	>100	>100	>100	44.65 ± 1.95	>100
19	>100	>100	>100	91.46 ± 4.88	>100
20	>100	>100	>100	>100	>100
21	>100	>100	>100	60.33 ± 2.21	>100
5-Fu	42.78 ± 0.99	47.09 ± 1.44	49.71 ± 1.02	10.53 ± 0.22	34.26 ± 0.78

---

Structure–activity relationship

For prenylated flavans (1–12, 14–15), their major structural differences involved the various changes in the prenyl groups and adjacent phenolic hydroxyl in ring B, which had a great influence on their antitumor potency. By comparing two similar flavans 1 and 2, the presence of a methoxyl connected to the pyran ring reduced the cytotoxicity against all tested cancer cell lines. Compounds 4–6 had the same skeleton apart from the type of substituent groups at C-4′,5′ and the results revealed that the inhibitory effects increased in the following order: 4-carbonyl-2,2-dimethyldihydropyran, furan and 2,2-dimethyl-3-methoxylpyran groups. In the same way, the activity of the compounds possessing a 2,2-dimethylpyran was greater than that of 2,2-dimethyldihydropyran and furan substituents according to the IC₅₀ values of 9–11. A comparison of the activity data between 8 and 10 indicated that the cyclization between the prenyl and hydroxyl moieties in ring B produced a significant improvement in the reduction of cancer cell growth. Also, this explained why 1 and 7 with a pyran or dihydropyran ring produced more potent inhibition than 14 and 15, respectively.

With regard to non-prenylated flavans (16–21), they displayed the inhibitory effects on the growth of Hep3B cells associated with the substituent pattern of hydroxyl and methoxyl groups. Comparison of the cytotoxic activities between 17–19 and 16, 21 and 20 suggested that the methoxyl added in ring B could lead to a slightly increased inhibition of Hep3B proliferation. Neither compound 16 nor 20 were cytotoxic, which demonstrated that the type of substituent groups at C-7 had no influence on their activities. Moreover, the hydroxyl was more effective than the methoxyl in terms of increasing the inhibitory effect on Hep3B in comparison with 17–18 and 19.

Flow cytometry analysis

As shown in Table 4, most flavans are more sensitive to Hep3B than other cancer cell lines and compounds 3 and 9–12 markedly suppressed the proliferation of Hep3B cells compared with positive drug 5-Fu with IC₅₀ values of 5.15–9.66 μM. Since regulation of the cell cycle is important for cell growth,²⁸ the effects of 3 and 9–12 on cell cycle distribution were investigated by flow cytometry analysis after PI staining. Compounds 3 and 9–11 definitely increased the percentage of Hep3B cells at the G2/M phase in a concentration-dependent manner at 48 h (Fig. 3 and 4A). However, there was little effect on G2/M phase arrest by 12 in Hep3B, and it mainly caused a significant improvement on the sub-G1 phase cells from 2.19% to 26.71% (Fig. 4B). To our knowledge, the appearance of a sub-G1 peak is closely associated with apoptosis.²⁹ Therefore, an Annexin V-FITC/PI double staining assay was used to evaluate the apoptotic effect of 12. As illustrated in Fig. 5, the early and late apoptotic cells progressively increased from 2.74% to 66.66% and 7.14% to 23.00%, respectively, following the exposure of Hep3B cells for 48 h. Taken together, these results demonstrated that prenylated flavans (3, 9–11) inhibited cell proliferation through G2/M phase arrest, while compound 12 exhibited marked growth inhibition of Hep3B mainly due to the induction of apoptosis. In addition, 3 and 9–11 had similar flavan skeletons all with a 2,2-dimethylpyran moiety connected to C-4′,5′ in ring B which are structurally different from 12, suggesting that the type of substituent groups had an important effect on the antitumor mechanism of flavans.

Fig. 3 Columns showing the analysis of cell cycle distribution in Hep3B cancer cells incubated with compounds 3 (A), 9 (B), 10 (C) and 11 (D) for 48 h at the indicated concentrations. The DNA contents of each cell phase were detected by flow cytometry. All data are expressed as the means ± SD of three independent experiments. *p < 0.05 compared with the control group.

Fig. 4 The flow cytometry histograms of Hep3B cancer cells after treatment with compounds 9 (A) and 12 (B) for 48 h. The untreated group was used for negative control. The experiments were performed three times and the results of representative experiments are shown.

Fig. 5 Analysis of apoptosis in Hep3B cancer cells after treatment with compound 12 for 48 h at the indicated concentrations. (A) Annexin V-FITC binding and PI staining were detected by flow cytometry. (B) Columns indicating the percentages of early (Annexin V+/PI−) and late (Annexin V+/PI+) apoptotic cells. Results are expressed as the means ± SD. *p < 0.05 compared with the untreated control group.

ROS is an important regulatory factor in cellular signaling pathways close related to development and dissemination of cancer cells.³⁰ Excessive ROS generation has been reported to cause irreversible injury including DNA damage, cell cycle arrest and apoptosis.³¹ Therefore, ROS levels of Hep3B cells induced by the two most active compounds 9 and 12 were detected using the fluorescent probe DCFH-DA. As shown in Fig. 6, significant increases in the ROS formation were both observed after treatment for 48 h. These data showed that 9 and 12 were good candidates for the promotion of ROS level in cells. Moreover, intracellular ROS was considered to be associated with G2/M phase cell cycle arrest and apoptosis induced by 9 and 12, respectively.

Fig. 6 Hep3B cells were treated with compounds 9 and 12 for 48 h. The intracellular ROS levels were measured by flow cytometry. Data are expressed as means ± SD. *p < 0.05 was considered statistically significant compared with control group.

Conclusions

In conclusion, our phytochemical study described the isolation and structural identification of thirteen new flavans, along with eight known ones from D. giraldii. Their cytotoxic activities were evaluated against five kinds of human cancer cell lines. Most flavan compounds had a selective cytotoxicity towards Hep3B cells. Among them, compounds 3 and 9–12 exhibited higher cytotoxic activities than the positive control 5-Fu. The structure–activity relationships were also discussed based on MTT results. By means of cell cycle and apoptosis detection using flow cytometry, we discovered that 3 and 9–11 all with a 2,2-dimethylpyran group in ring B inhibited Hep3B cell proliferation by arresting at the G2/M phase, while 12, with different structural features, had an inhibitory effect by inducing apoptosis. Besides, compounds 9 and 12 could stimulate ROS generation which probably disrupted normal transition of G2 to M phase and lead to cancer cell apoptotic death. This investigation indicated that cytotoxic flavans from D. giraldii are promising in applications of clinical antitumor therapy. Further research into these compounds involving related proteins and cytotoxicity in normal cells is definitely needed to be conducted in our following studies.

Experimental section

General experimental procedures

Ultraviolet spectra were measured in methanol solution on an UV-1700 spectrophotometer (SHIMADZU, Japan). The IR spectra were obtained neat on a Bruker IFS 55 spectrophotometer (Bruker, Germany). Optical rotations were measured in methanol on an AUTOPOL IV automatic polarimeter (Rudolph Research Analytical, USA) at room temperature. CD spectrum was recorded on a MOS-450 spectrometer (Bio-Logic Science, France). NMR spectra were performed on Bruker ARX-300, ARX-400 and AV-600 spectrometers with chemical shifts given in ppm (δ). HRESIMS was determined on a Bruker Micro Q-TOF spectrometer. Column chromatography was carried out using silica gel (100–200 or 200–300 mesh; Qingdao Marine Chemical Inc. China), C₁₈ reversed-phase (RP) silica gel (60–80 mm; Merck, Germany), Sephadex LH-20 gel (GE Healthcare, Sweden), and MCI gel (CHP20P, 75–150 μm; Mitsubishi Chemical Industries Ltd. Japan). Precoated silica gel GF254 plates (Yantai Zifu Chemical Group Co. China) were used for TLC detection. High performance liquid chromatography was performed on a Shimadzu LC-6AD HPLC apparatus with a SPD-20A UV/VIS detector using a YMC Pack ODS-A column (250 × 10 mm, 5 μm).

Plant material

The stem and root bark of D. giraldii were purchased from Anguo herbal medicine market (HeBei, China), and authenticated by Prof. Jin-Cai Lu, Department of Traditional Chinese Materia Medica, Shenyang Pharmaceutical University, wherein a voucher specimen (DG-20120515) was deposited.

Extraction and isolation

Extraction of the stem and root bark material from D. giraldii carried out in the conventional way by soaking at the room temperature for one week. The concentrated 95% EtOH extract (986.0 g) was isolated over initial column chromatography on silica gel using CH₂Cl₂ with increasing proportions of MeOH (100 : 1 to 0 : 100) to get fractions A–D. Fraction B (280.0 g) was separated into B1–B5 by repeated silica gel eluted with PE–EtOAc (50 : 1 to 0 : 100), and B3 (65.3 g) followed by an MCI gel column (MeOH–H₂O, 40 : 60 to 100 : 0) to yield B3a–B3d. Then fraction B3b (10.6 g) was subjected sequentially on reversed-phase C₁₈ silica gel CC (MeOH–H₂O, 40 : 60 to 100 : 0) and silica gel (PE–EtOAc, 8 : 1 to 1 : 1), finally by semipreparative HPLC (MeCN–H₂O, 30 : 70) at a flow rate of 5 mL min⁻¹ to yield 16 (11.5 mg, t_R 24.5 min), 17 (29.0 mg, t_R 26.7 min) and 18 (20.6 mg, t_R 29.1 min), respectively. Fraction B3c (24.0 g) was chromatographed over ODS silica gel using MeOH–H₂O (50 : 50 to 80 : 20) as an eluent to produce five fractions (B3c-1–B3c-5), and separation of B3c-2 (2.7 g) on silica gel (PE–EtOAc, 10 : 1 to 1 : 1) obtained ten subfractions named B3c-2-1–B3c-2-10 based on TLC monitoring. Fr. B3c-2-2 (168.4 mg) was further purified by semipreparative HPLC (MeCN–H₂O, 65 : 35, 5 mL min⁻¹) to afford 8 (25.2 mg, t_R 12.3 min), 9 (7.1 mg, t_R 17.1 min) and 10 (8.9 mg, t_R 18.4 min). Fr. B3c-2-4 (94.6 mg) was applied to HPLC eluted with MeCN–H₂O (60 : 40, 5 mL min⁻¹) to obtain 7 (11.3 mg, t_R 20.7 min). Similarly, Fr. B3c-2-7 (92.1 mg) was purified through HPLC chromatography (MeCN–H₂O, 40 : 60, 5 mL min⁻¹) to give 5 (5.3 mg, t_R 15.4 min) and 12 (9.0 mg, t_R 21.9 min). Fraction B3c-3 (8.2 g) was subjected to silica gel CC (PE–EtOAc, 20 : 1 to 1 : 1) to give ten fractions (B3c-3-1–B3c-3-10). Fr. B3c-3-2 (142.6 mg) was purified by HPLC (MeCN–H₂O, 40 : 60, 3 mL min⁻¹) to yield 20 (4.6 mg, t_R 28.5 min) and 21 (4.4 mg, t_R 30.8 min). Likewise, Fr. B3c-3-3 (186.6 mg) was purified in the same way to afford 1 (35.6 mg, t_R 34.7 min), 11 (12.5 mg, t_R 37.1 min) and 6 (2.5 mg, t_R 40.5 min), respectively. The Fr. B3c-3-4 (0.7 g) was run through CC over ODS silica gel (MeOH–H₂O, 70 : 30), followed by semipreparative HPLC (MeCN–H₂O, 50 : 50, 3 mL min⁻¹) to give 3 (9.1 mg, t_R 19.4 min), 14 (46.0 mg, t_R 23.3 min), 19 (11.0 mg, t_R 34.0 min) and 13 (7.3 mg, t_R 35.6 min), respectively. Besides, compounds 2 (3.4 mg, t_R 9.7 min), 4 (2.8 mg, t_R 11.4 min) and 15 (16.0 mg, t_R 16.1 min) were discovered from Fr. B3c-3-6 by means of ODS (MeOH–H₂O, 70 : 30) and semipreparative HPLC (MeOH–H₂O, 70 : 30, 3.0 mL min⁻¹).

Daphnegiravan A (1). Yellow solid (MeOH); [α]²⁵_D −31.5 (c 0.13, MeOH); CD (MeOH) nm (Δε) 221 (−3.85), 284 (−0.84); UV (MeOH) λ_max nm (log ε) 229 (4.27), 284 (3.65), 315 (3.45); IR (KBr) ν_max 3418, 2919, 2850, 1621, 1461, 1384, 1148, 1032 cm⁻¹; ¹H NMR and ¹³C NMR spectroscopic data, see Tables 1 and 2; HRESIMS m/z 309.1479 [M + H]⁺ (calcd for C₂₀H₂₁O₃, 309.1485).

Daphnegiravan B (2). Yellow solid (MeOH); [α]²⁵_D −39.6 (c 0.05, MeOH); CD (MeOH) nm (Δε) 221 (−2.90), 283 (−0.66); UV (MeOH) λ_max nm (log ε) 229 (4.09), 274 (3.62), 284 (3.61); IR (KBr) ν_max 3424, 2919, 2850, 1618, 1447, 1384, 1146 cm⁻¹; ¹H NMR and ¹³C NMR spectroscopic data, see Tables 1 and 2; HRESIMS m/z 361.1411 [M + Na]⁺ (calcd for C₂₁H₂₂O₄Na, 361.1410).

Daphnegiravan C (3). Yellow solid (MeOH); [α]²⁵_D −40.5 (c 0.08, MeOH); CD (MeOH) nm (Δε) 224 (−3.31), 284 (−0.98); UV (MeOH) λ_max nm (log ε) 228 (4.22), 270 (3.83), 282 (3.85), 317 (3.02); IR (KBr) ν_max 3424, 2919, 2850, 1621, 1460, 1383, 1144 cm⁻¹; ¹H NMR and ¹³C NMR spectroscopic data, see Tables 1 and 2; HRESIMS m/z 347.1260 [M + Na]⁺ (calcd for C₂₀H₂₀O₄Na, 347.1254).

Daphnegiravan D (4). Yellow solid (MeOH); [α]²⁵_D −33.3 (c 0.06, MeOH); CD (MeOH) nm (Δε) 225 (−3.41), 283 (−0.87); UV (MeOH) λ_max (log ε) 223 (4.35), 284 (3.73) nm; IR (KBr) ν_max 3395, 2922, 2851, 1622, 1597, 1458, 1383, 1152, 1114 cm⁻¹; ¹H NMR and ¹³C NMR spectroscopic data, see Tables 1 and 2; HRESIMS m/z 445.1993 [M + Na]⁺ (calcd for C₂₆H₃₀O₅Na, 445.1985).

Daphnegiravan E (5). Yellow solid (MeOH); [α]²⁵_D −41.5 (c 0.08, MeOH); CD (MeOH) nm (Δε) 223 (−2.80), 287 (−0.83); UV (MeOH) λ_max (log ε) 220 (4.43), 283 (3.94) nm; IR (KBr) ν_max 3424, 2920, 2850, 1626, 1461, 1384, 1130, 1032 cm⁻¹; ¹H NMR and ¹³C NMR spectroscopic data, see Tables 1 and 2; HRESIMS m/z 409.2002 [M + H]⁺ (calcd for C₂₅H₂₉O₅, 409.2010).

Daphnegiravan F (6). Yellow solid (MeOH); [α]²⁵_D −27.5 (c 0.08, MeOH); CD (MeOH) nm (Δε) 221 (−3.55), 282 (−0.77); UV (MeOH) λ_max (log ε) 220 (4.09), 284 (3.65) nm; IR (KBr) ν_max 3417, 2920, 2850, 1621, 1596, 1459, 1383, 1152, 1115 cm⁻¹; ¹H NMR and ¹³C NMR spectroscopic data, see Tables 1 and 2; HRESIMS m/z 373.1418 [M + Na]⁺ (calcd for C₂₂H₂₂O₄Na, 373.1410).

Daphnegiravan G (7). Yellow solid (MeOH); [α]²⁵_D −29.8 (c 0.10, MeOH); CD (MeOH) nm (Δε) 222 (−4.17), 284 (−0.64); UV (MeOH) λ_max (log ε) 220 (4.12), 284 (3.64) nm; IR (KBr) ν_max 3424, 2972, 2927, 1621, 1597, 1482, 1157, 1114 cm⁻¹; ¹H NMR and ¹³C NMR spectroscopic data, see Tables 1 and 2; HRESIMS m/z 417.2034 [M + Na]⁺ (calcd for C₂₅H₃₀O₄Na, 417.2036).

Daphnegiravan H (8). Yellow solid (MeOH); [α]²⁵_D −44.8 (c 0.10, MeOH); CD (MeOH) nm (Δε) 226 (−2.58), 286 (−0.60); UV (MeOH) λ_max nm (log ε) 224 (4.04), 285 (3.51); IR (KBr) ν_max 3438, 2973, 2926, 1621, 1594, 1508, 1444, 1160, 1116 cm⁻¹; ¹H NMR and ¹³C NMR spectroscopic data, see Tables 1 and 2; HRESIMS m/z 417.2037 [M + Na]⁺ (calcd for C₂₅H₃₀O₄Na, 417.2036).

Daphnegiravan I (9). Yellow solid (MeOH); [α]²⁵_D −25.4 (c 0.09, MeOH); CD (MeOH) nm (Δε) 228 (−3.57), 302 (−0.71); UV (MeOH) λ_max nm (log ε) 218 (4.28), 304 (3.82); IR (KBr) ν_max 3462, 2969, 1721, 1621, 1508, 1395, 1145, 1026 cm⁻¹; ¹H NMR and ¹³C NMR spectroscopic data, see Tables 1 and 3; HRESIMS m/z 413.1717 [M + Na]⁺ (calcd for C₂₅H₂₆O₄Na, 413.1723).

Daphnegiravan J (10). Yellow solid (MeOH); [α]²⁵_D −44.3 (c 0.11, MeOH); CD (MeOH) nm (Δε) 227 (−4.75), 285 (−0.72); UV (MeOH) λ_max nm (log ε) 231 (4.13), 284 (3.55); IR (KBr) ν_max 3457, 2972, 1622, 1457, 1384, 1377, 1115 cm⁻¹; ¹H NMR and ¹³C NMR spectroscopic data, see Tables 1 and 3; HRESIMS m/z 415.1878 [M + Na]⁺ (calcd for C₂₅H₂₈O₄Na, 415.1880).

Daphnegiravan K (11). Yellow solid (MeOH); [α]²⁵_D −34.8 (c 0.07, MeOH); CD (MeOH) nm (Δε) 225 (−2.85), 284 (−0.61); UV (MeOH) λ_max nm (log ε) 233 (4.02), 284 (3.49), 316 (3.17); IR (KBr) ν_max 3376, 2976, 2922, 2850, 1622, 1595, 1508, 1461, 1154, 1111, 1039, 1001 cm⁻¹; ¹H NMR and ¹³C NMR spectroscopic data, see Tables 1 and 3; HRESIMS m/z 371.1259 [M + Na]⁺ (calcd for C₂₂H₂₀O₄Na, 371.1254).

Daphnegiravan L (12). Yellow solid (MeOH); [α]²⁵_D −39.1 (c 0.09, MeOH); CD (MeOH) nm (Δε) 220 (−2.77), 286 (−0.77); UV (MeOH) λ_max nm (log ε) 278 (3.77), 285 (3.63); IR (KBr) ν_max 3412, 2923, 1620, 1595, 1458, 1383, 1151, 1112 cm⁻¹; ¹H NMR and ¹³C NMR spectroscopic data, see Tables 1 and 3; HRESIMS m/z 415.1878 [M + Na]⁺ (calcd for C₂₅H₂₈O₄Na, 415.1880).

Daphnegiravan M (13). Brown solid (MeOH); [α]²⁵_D −43.2 (c 0.12, MeOH); CD (MeOH) nm (Δε) 219 (−2.48), 250 (−0.32), 286 (−0.59); UV (MeOH) λ_max nm (log ε) 213 (4.07), 256 (3.52), 283 (3.27); IR (KBr) ν_max 3422, 2920, 2850, 1657, 1623, 1460, 1384, 1278, 1151, 1113, 1002 cm⁻¹; ¹H NMR and ¹³C NMR spectroscopic data, see Tables 1 and 3; HRESIMS m/z 293.0773 [M + Na]⁺ (calcd for C₁₆H₁₄O₄Na, 293.0784).

In vitro cytotoxicity assay

Human breast adenocarcinoma (MCF-7, Bcap37), hepatocellular carcinoma (HepG2, Hep3B) and lung adenocarcinoma (A-549) cell lines were obtained from the American Type Culture Collection (USA). The cells were cultured in Dulbecco's modified eagle medium (DMEM) (Hyclone, USA) supplemented with 10% fetal bovine serum (Biological Industries, Israel) in a 37 °C, 5% CO₂ incubator. An MTT (Amersco, USA) assay was used to measure the inhibition of tumor cell growth following treatment with different compounds (1–21) in 96-well plates as previously described.³² All tests were conducted at least in triplicate.

Cell cycle assay

The cell cycle analysis was determined using the Cell Cycle and Apoptosis Analysis Kit (Beyotime, Shanghai, China) according to the manufacturer's instructions. Hep3B cells were incubated with compounds 3 and 9–12 in various concentrations. After treatment for 48 h, the cells were collected, washed with PBS, and fixed in iced 70% EtOH for 12 h at 4 °C. Then, the cells were stained with Propidium Iodide (PI) solution (50 μg mL⁻¹ RNase A and 50 μg mL⁻¹ PI in PBS) and incubated for 30 min in the dark at 37 °C before analyzing by FACSCalibur flow cytometry (Becton Dickinson, USA).

Cell apoptosis assay

The apoptotic analysis was carried out using an Annexin V-FITC Apoptosis Detection Kit (Beyotime, Shanghai, China). Hep3B cells were treated with compound 12 for 48 h. Then the cells were collected, washed with ice-cold PBS and resuspended in Annexin V-FITC/PI staining solution according to the manufacturer's instructions for 15 min in the dark. The cells were then immediately analyzed by FACSCalibur flow cytometry (Becton Dickinson, USA).

ROS production measurement

Intracellular ROS level was detected using the fluorescent probe DCFH-DA. After treatment with the tested compounds 9 and 12 for 48 h, cells were harvested and incubated with 10 μM of DCFH-DA for 30 min in the dark at 37 °C. Then, cells were washed with cold PBS and analyzed immediately using flow cytometry (Becton Dickinson, USA).

Statistical analysis

The statistical analysis was performed using SPSS 16.0 software. All the data were present as means ± SD for triplicate experiments. One-way analysis of variance (ANOVA) followed by Turkey's test was used to express the statistical differences between groups. The value of P < 0.05 was considered statistically significant.

Acknowledgements

The authors are grateful to Associate Professor Y. Peng, Mrs W. Li and Mr Y. Sha of Shenyang Pharmaceutical University for HRESIMS and NMR spectra determinations. Financial support from the Project of Innovation Team (LT2015027) of Liaoning of P. R. China is also gratefully acknowledged.

Notes and references

1. B. W. Stewart and C. P. Wild, World Cancer Report 2014, International Agency for Research on Cancer Publications, Lyon, France, 2014.
2. X. H. Tian, L. Li, J. P. Pei, R. C. Yue, X. Fang, J. P. Zhang, W. W. He, L. Shan, Y. H. Shen and W. D. Zhang, RSC Adv., 2015, 5, 75857–75862.
3. Y. Q. Liu, L. Yang and X. Tian, Curr. Bioact. Compd., 2007, 3, 37–66.
4. Y. Chen, C. Tang, Y. Wu, S. S. Mo, S. Wang, G. Z. Yang and Z. N. Mei, Org. Biomol. Chem., 2015, 13, 6773–6781.
5. X. R. Peng, X. Wang, L. Zhou, B. Hou, Z. L. Zuo and M. H. Qiu, RSC Adv., 2015, 5, 95212–95222.
6. D. J. Newman, J. Med. Chem., 2008, 51, 2589–2599.
7. D. J. Newman and G. M. Cragg, J. Nat. Prod., 2012, 75, 311–335.
8. W. C. Xua, J. G. Shen and J. Q. Jiang, Chem. Biodiversity, 2011, 8, 1215–1233.
9. S. Z. Huang, X. J. Zhang, X. Y. Li, L. M. Kong, H. Z. Jiang, Q. Y. Ma, Y. Q. Liu, J. M. Hu, Y. T. Zheng, Y. Li, J. Zhou and Y. X. Zhao, Phytochemistry, 2012, 75, 99–107.
10. R. W. Zhang, Y. L. Wang, J. X. Li, H. L. Jin, S. J. Song and C. S. Huang, J. Biol. Chem., 2014, 289, 6394–6403.
11. W. F. Zheng, X. W. Gao, C. F. Chen and R. X. Tan, Int. Immunopharmacol., 2007, 7, 117–127.
12. G. J. Finn, E. Kenealy, B. S. Creaven and D. A. Egan, Cancer Lett., 2002, 183, 61–68.
13. J. Zhao, X. J. Jin and H. J. Zhang, Chinese Wild Plant Resources, 2012, 31, 12–14.
14. Jiangsu New Medical College, Dictionary of Chinese Herbal Medicine, Shanghai People’s Publishing House, Shanghai, China, 1977, vol. 2, pp. 1739–1740.
15. P. Wang, J. P. Liu, N. Zhan, P. Y. Li and D. Lu, Special Wild Economic Animal and Plant Research, 2011, 4, 73–76.
16. J. Su, Z. J. Wu, Y. H. Shen, C. Zhang, H. L. Li, R. H. Liu and W. D. Zhang, J. Asian Nat. Prod. Res., 2008, 6, 547–550.
17. G. X. Zhou, R. W. Jiang, Y. Cheng, W. C. Ye, J. G. Shi, N. B. Gong and Y. Lu, Chem. Pharm. Bull., 2007, 9, 1287–1290.
18. D. Slade, D. Ferreira and J. P. Marais, Phytochemistry, 2005, 66, 2177–2215.
19. D. Lee, K. P. Bhat, H. H. Fong, N. R. Farnsworth, J. M. Pezzuto and A. D. Kinghorn, J. Nat. Prod., 2001, 64, 1286–1293.
20. D. Y. Lee, D. H. Kim, H. J. Lee, Y. J. Lee, K. H. Ryu, B. I. Jung, Y. S. Song and J. H. Ryu, Bioorg. Med. Chem. Lett., 2010, 20, 3764–3767.
21. Q. A. Zheng, Y. J. Zhang and C. R. Yang, J. Asian Nat. Prod. Res., 2006, 8, 571–577.
22. M. Masaoud, H. Ripperger, A. Porzel and G. Adam, Phytochemistry, 1995, 38, 745–749.
23. P. Y. Ango, D. W. Kapche, V. Kuete, B. T. Ngadjui, M. Bezabih and B. M. Abegaz, Phytochem. Lett., 2012, 5, 524–528.
24. H. Achenbach, M. Stocker and M. A. Consmnla, Phytochemistry, 1988, 27, 1835–1841.
25. W. F. Zheng, X. W. Gao, Q. Gu, C. F. Chen, Z. W. Wei and F. Shi, Int. Immunopharmacol., 2007, 7, 128–134.
26. H. Shen, Z. Y. Lou, R. H. Liu, C. Zhang, P. Fu, L. Shan and W. D. Zhang, Nat. Prod. Res., 2007, 21, 1021–1026.
27. E. C. F. Edgar, R. P. S. Sergio, W. T. T. Luis and E. M. P. Rosa, Molecules, 2015, 20, 13563–13574.
28. A. S. Zoe, D. W. Matthew and A. P. Jennifer, Trends Pharmacol. Sci., 2003, 24, 139–145.
29. K. H. Lim, V. J. Raja, T. D. Bradshaw, S. H. Lim, Y. Y. Low and T. S. Kam, J. Nat. Prod., 2015, 78, 1129–1138.
30. A. J. Sun, J. Q. Liu, S. Q. Pang, J. S. Lin and R. A. Xu, Bioorg. Med. Chem. Lett., 2016, 26, 2375–2379.
31. N. Kang, J. F. Jian, S. J. Cao, Q. Zhang, Y. W. Mao, Y. Y. Huang, Y. F. Peng, F. Qiu and X. M. Gao, Mol. Cell. Biochem., 2016, 415, 145–155.
32. L. Z. Li, S. J. Song, P. Y. Gao, F. F. Li, L. H. Wang, Q. B. Liu, X. X. Huang, D. Q. Li and Y. Sun, RSC Adv., 2015, 5, 4143–4152.

---

Footnote

† Electronic supplementary information (ESI) available: NMR, HRESIMS, UV, IR, CD spectrum and HMBC correlations of compounds 1–13, cell cycle arrest, apoptosis and ROS level flow cytometry analysis of 3 and 9–12, NMR data of known compounds 14–21. See DOI: 10.1039/c6ra08537g

---