Synthesis and biological evaluation of novel steroidal 5α,8α-endoperoxide derivatives with aliphatic side-chain as potential anticancer agents

Steroids 124 (2017) 46–53

Contents lists available at ScienceDirect

Steroids journal homepage: www.elsevier.com/locate/steroids

Synthesis and biological evaluation of novel steroidal 5α,8α-endoperoxide derivatives with aliphatic side-chain as potential anticancer agents

MARK

Ming Bua, Tingting Caoa, Hongxia Lia, Mingzhou Guoc, Burton B. Yangd, Yue Zhoua, Na Zhanga, ⁎ Chengchu Zenga, Liming Hua,b, a

Department of Biomedical Engineering, College of Life Science and Bioengineering, Beijing University of Technology, Beijing 100124, China Beijing Key Laboratory of Environmental and Viral Oncology, College of Life Science and Bioengineering, Beijing University of Technology, Beijing 100124, China c Chinese PLA General Hospital, Beijing 100853, China d Department of Laboratory Medicine and Pathobiology, University of Toronto, M4N3M5 Toronto, Canada b

A R T I C L E I N F O

A B S T R A C T

Keywords: Endoperoxide Ergosterol peroxide Peroxy bond Cytotoxic activities Photooxygenation

By inspiration of significant anti-cancer activity of our previously screened natural ergosterol peroxide (EP), a series of novel steroidal 5α,8α-endoperoxide derivatives 5a–d and 14a–f were designed, synthesized, and biologically evaluated for their in vitro anti-proliferative inhibitory and cytotoxic activity. The results revealed that most of these compounds showed moderate-to-excellent anti-proliferative effects against the tested cancer cell lines (i.e. HepG2, SK-Hep1, MDA-MB-231 and MCF-7). Among them, compound 5b and 14d exhibited preferable inhibitory activities (IC50 of 5b and 14d are 8.07 and 9.50 μM against HepG2, respectively). The structureactivity relationships indicated that incorporation the peroxidic bridge to the steroid scaffolds at C-5 and C-8 positions together with the aliphatic side-chain at the C-17 position would provide synergistic effect for the bioactivity.

1. Introduction Nowadays, natural drugs have attracted extensive attention in health promotion and disease treatment including cancer [1,2]. In addition, natural product-based drug discovery is a major route leading to developing therapeutic drugs for various diseases. Many natural health products are obtained from plants, animals, and microorganisms. Natural endoperoxides are cyclic organic compounds, with an O-O single bond as a peroxidic bridge [3,4]. They play an important role in drug synthesis as well as in medicine, and represent the central part of artemisinins, outstanding antimalarial drugs, honored with the Noble Price in Medicine 2015 [5]. Although best known as potent antimalarials, cyclic peroxides also exhibit a range of activities which encompasses antifungal, antiviral and anticancer activity [6,7]. Among natural endoperoxides, steroidal 5α,8α-endoperoxides are the important active lead compounds in drug discovery, which are well known for their 5α,8α-peroxy moiety [8]. Ergosterol peroxide (5α,8αepidioxyergosta-6,22-dien-3β-ol, EP) (Fig. 1), is a member of a class of fungal secondary metabolites of sterol 5α,8α-endoperoxide derivatives [9]. It can be isolated from many medicinal fungi, such as Sarcodon aspratus, Hericium erinaceum, Armillariella mellea, lactarius hatsudake,

hypsizigus marmoreus, et al. [10–12]. It has been reported that EP can inhibit tumor growth by anti-angiogenesis or cytotoxicity [13–15]. In our previous study, we found that EP purified from Ganoderma lucidum, induced cell death and inhibited cell migration, cell cycle progression, and colony growth of human hepatocellular carcinoma cells [16]. We further examined the mechanism associated with this effect and found that treatment with EP increased expression of Foxo3a mRNA and protein in HepG2 cells. The levels of Puma and Bax, proapoptotic proteins, were effectively enhanced. Our results suggest that ergosterol peroxide stimulated Foxo3 activity by inhibiting pAKT and cMyc and activating pro-apoptotic protein Puma and Bax to induce cancer cell death. With further clinical development, EP represents a promising new reagent that can overcome the drug-resistance of tumor cells [17]. As an important active lead compound in drug discovery, EP is well known for its 5α,8α-peroxy moiety. We have recently developed a simple and practical synthetic route to obtain novel series of sterol 5α,8α-endoperoxides from readily available natural sterols. The synthesis of endoperoxides achieved by the reaction of steroidal △5,7diene intermediates with singlet oxygen (1O2), which can be conveniently generated photochemically using eosin Y (EY) as a

⁎ Corresponding author at: Beijing Key Laboratory of Environmental and Viral Oncology, College of Life Science and Bioengineering, Beijing University of Technology, Beijing 100124, China. E-mail address: [email protected] (L. Hu).

http://dx.doi.org/10.1016/j.steroids.2017.05.013 Received 22 January 2017; Received in revised form 4 May 2017; Accepted 31 May 2017 0039-128X/ © 2017 Elsevier Inc. All rights reserved.

Steroids 124 (2017) 46–53

M. Bu et al.

Table 1 Optimization of the photooxygenation reaction conditions. Entry

Reaction systema

Temperatureb

Lightc

Oxygen sourced

Time (h)

Yield (%)e

1 2 3 4 5

EG-EY-Py EG-EY-Py EG-EY-Py EG-EY-Py EG-EYEtOH EG-EY-Py EG-EY-Py EG-EYEtOH EG-EYEtOH

Ice-water Ice-water Ice-water Ice-water Ice-water

100 W 300 W 500 W 500 W 500 W

O2 O2 O2 O2 O2

0.5 0.5 0.5 1.0 0.5

22 46 62 67 54

Ice-water bath RT RT

500 W Daylight Daylight

Air O2 Air

0.5 0.5 5

— — 7

RT

Daylight

Air

10

18

6 7 8

Fig. 1. Structure of ergosterol peroxide (EP).

photosensitive catalytic oxidizer from molecular oxygen (photooxygenation) [18–20]. Acknowledging the limited structure-activity relationship studies for EP, we preliminary designed and synthesized a series of novel steroidal 5α,8α-endoperoxide derivatives that with C-17 aliphatic side chain. Meanwhile, their biological activities (IC50 values) were compared using a MTT assay with four human cancer cell lines. We hope to get valuable information for further design of novel steroidal anticancer agents. Herein, we designed and synthesized a series of novel steroidal 5α,8α-endoperoxide derivatives that with C-17 aliphatic side chain.

9

a b c d e

bath bath bath bath bath

EG: ergosterol; EY: eosin Y; Py: pyridine; EtOH: anhydrous ethanol. RT: Room temperature. Light: Iodine-tungsten lamp 220 V (100, 300, 500 W). O2: high purity oxygen (> 99.995%). Isolation yield.

the reaction of singlet oxygen with △5,7-diene intermediates is well accepted owing to the fact that the methyl group at C-10 is β (axial) (Fig. 3, 19-CH3).

2. Result and discussion 2.1. Chemistry

2.1.2. Synthesis of 5a–d from natural steroids Using natural β-sitosterol (1a), cholesterol (1b), pregnenolone (PREG, 1c), and dehydroepiandrosterone (DHEA, 1d) as the starting materials, we performed chemical synthesis and purification as described in Scheme 2. Four new 5α,8α-endoperoxides 5a–d were synthesized. Compounds 2a–d were prepared via acetylation reaction of 1a–d, which then underwent bromination and debromination with NBS, n-Bu4NBr and n-Bu4NF to afford △5,7-diene acetates 3a–d. Subsequently, products 4a–d were obtained after deacetylation reaction of 3a–d. Finally, the target steroidal 5α,8α-endoperoxide derivatives 5a–d were obtained by optimized photooxygenation method in Table 1 (entry 3).

2.1.1. Synthesis of EP from ergosterol Using natural ergosterol (EG) as the starting material, we performed chemical synthesis and purification as described in Scheme 1. EP was purified as white crystalline needles. The photooxygenation reaction is the key step of the whole synthetic route and its reaction conditions had to be optimized in order to get high conversion. A typical experimental procedure for the optimization of the photooxygenation reaction conditions was described using EG as an example (Table 1). On the basis of the optimized reaction conditions, we synthesized EP from EG with eosin Y in pyridine, irradiated with iodine tungsten lamp and kept bubbling oxygen for 0.5 h to get EP. Eosin Y (EY) was selected as the photosensitizer, and other conditions such as the solvent, reaction time, reaction temperature and light were chosen based on the results shown in Table 1. Crystal of EP suitable for single-crystal X-ray diffraction was firstly obtained by slow crystallization from n-hexane/ethyl acetate solution at ambient temperature (Fig. 2). An interesting feature in its structure is the presence of two molecules in the crystallographic asymmetric unit. The structure confirms the α-stereochemistry of the peroxy bond at C-5 and C-8 positions. Crystal data and structure refinement details for EP was presented in Table S1. (CCDC numbers 1502457). A plausible reaction mechanism for the formation of 5α,8α-peroxy moiety from steroidal △5,7-diene intermediate is depicted in Fig. 3. First, singlet oxygen is generated from sensitization by eosin Y∗ (EY∗). Then, the clear region-selectivity of the singlet oxygen attacks to the C5 and C-8 positions of the conjugated double bond system in the [4+2] cycloaddition manner. What’s more, the α-π-facial stereo-selectivity of

2.1.3. Synthesis of 14a–f from DHEA Using readily available DHEA (1d) as the starting material, we performed chemical synthesis and purification ergosterol peroxide analogues 14a–f as described in Scheme 3. The synthesis started with 3β-(tert-butyldimethylsilyl)oxy-androst-5-ene (6), which is readily available from 1d. The side chain with natural configuration at C-17 and C-20 was introduced by the method developed by Uskokovic et al. [21]. The Wittig reaction of 6 with ethyltriphenylphosphonium bromide stereoselectively gave the Zolefin 7 with a trace of the isomeric olefin in quantitative yield. Due to the difficulty in separating the stereoisomers, the mixture of the isomers was used in the next step. The ene reaction of the thus-obtained olefins with paraformaldehyde in the presence of a catalytic amount of boron trifluoride etherate afforded the alcohol 8 stereo-specifically in 86% yield. The epimer at C-20, derived from the E-olefin, was cleanly separated by column chromatography. Catalytic hydrogenation of 8 over 5% Pt-C reduced only the △16-double bond stereoselectively to give the mono-olefin 9 in quantitative yield. Formation of the △5,7-diene moiety was best achieved by Rappoldt et al. [22], successive treatment of 9 with NBS, nBu4NBr and n-Bu4NF produced the diene 10 in 26% yield. The diene moiety is stable enough to allow further elaboration, as described in Scheme 3. Tosylate 11 was readily prepared from alcohol 10 in high yield. Then, alkylation of the intermediate 11 with different Grignard reagents to get aliphatic side chain analogues 12a–f [23]. The analogues 12a–f were further treated with n-Bu4NF to produce the intermediates 13a–f. Finally, the target steroidal 5α,8α-endoperoxide analogues 14a–f were obtained by optimized method in Table 1.

Scheme 1. Synthesis of EP from ergosterol. Reagents and conditions: O2, eosin Y, pyridine, hv, 0 °C, 0.5 h.

47

Steroids 124 (2017) 46–53

M. Bu et al.

Fig. 2. X-ray crystal structure of EP.

the positive control. The anti-proliferative potency of the tested compounds was indicated by IC50 values using the MTT assay. The results are summarized in Table 2. As shown in Table 2, some of the tested compounds showed moderate to potent anticancer activities against all the tested cells. Compound 5b and 14d were the most promising compounds amongst the tested derivatives, with IC50 values of 8.07–12.25 µM and 9.50–15.30 µM, respectively. The screening results suggested some rough structure-activity relationship considerations. First, the studies showed that the peroxidic bridge at the C-5 and C-8 position is requisite pharmacophore for these derivatives. Most of compounds, including EP, 5a, 5b and 14a–f which have peroxidic bridge at the B-ring of the sterol skeleton exhibited good activity against all the tested cells (IC50 < 20 μM). Ergosterol (EG), which has no peroxidic bridge at C-5 and C-8 positions exhibited low activity against all the tested cells (IC50 > 60 μM). Compound 5c and 5d, which have carbanyl group at the C-17 or C20 position, showed no activity against all the tested cells (IC50 > 50 μM). On the other hand, most of compounds with the aliphatic side-chain at the C-17 position exhibited potent activity against all the tested cells. The result suggested that the introduction of aliphatic side-chain to the C-17 position resulted in increased anticancer activity. A comparison of the results obtained from the endoperoxides, showed that compound 5b, 14b, 14d–f which possess saturated aliphatic side-chain at the C-17 position displayed good activities. Compared with 14a, 14e and 14f, compounds 5b, 14b and 14f with isobutyl at the terminal of the side-chain were active against some cancer cell lines. These findings indicated that incorporation the peroxidic bridge to the steroid scaffolds at C-5 and C-8 positions together with the saturated isobutyl terminally aliphatic side-chain at the C-17 position played an important role for the activity. Furthermore, it appeared that substituent changes to the C-17 position could serve as a promising launch point for further design of this type of steroidal anticancer agents. Then, EP, 5b, 14b, 14d and 14e were selected for further in vitro evaluation using a non-cancer cell line 293T (human kidney epithelial cell line) as a control. Exponentially growing 293T cells were treated with the compounds at different concentrations for 48 h. Cell-growth inhibition was analyzed by the MTT assay. As shown in Table 3, the median cytotoxic concentration (CC50) showed that most of the tested compounds displayed relative lower cytotoxicity in vitro against 293T cells. They exhibited appropriate selectivity between cancer cell line and non-cancer cell line.

Fig. 3. The plausible mechanism of photooxygenation.

2.2. Biological results and discussion The newly synthesized EP, 5a–d, 14a–f and EG were evaluated for their in vitro anti-proliferative activities against human cancer cell lines derived from various human cancer types, such as human hepatocellular cancer cell lines (HepG2, SK-Hep1) and human breast cervical cancer cell lines (MDA-MB-231, MCF-7). Cis-platinum was employed as

3. Conclusion In summary, a series of sterol 5α,8α-endoperoxide derivatives possessing various aliphatic side-chain were synthesized and evaluated for their anti-proliferative activities. Some of the synthesized compounds exhibited potent anticancer activities through inducing cancer cell apoptosis against the four tested cancer cell lines in vitro. In particular, compound 5b and 14d were the most promising derivatives, with IC50 values ranging from 8.07–12.25 µM and 9.50–15.30 µM,

Scheme 2. Synthesis of 5a–d from natural sterols. Reagents and conditions: (i) Ac2O, DCM, Pyridine, RT; (ii) NBS, NaHCO3, cyclohexane, reflux, 1 h; n-Bu4NBr, n-Bu4NF, THF, 0 °C, 4 h; (iii) NaOMe, MeOH, RT, 12 h; (iv) O2, eosin Y, pyridine, hv, 0 °C, 0.5 h.

48

Steroids 124 (2017) 46–53

M. Bu et al.

Scheme 3. Synthesis of ergosterol peroxide analogues 14a–f. Reagents and conditions: (i) TBSCl, imidazole, THF, reflux, 4 h; (ii) EtPPh3Br, t-BuOK, THF, reflux, 3 h; (iii) paraformaldehyde, BF3·Et2O, DCM, 0.5 h; (iv) H2, Pt/C, EtOH, 10 h; (v) NBS, NaHCO3, cyclohexane, reflux, 1 h; n-Bu4NBr, n-Bu4NF, THF, 0 °C, 4 h; (vi) TsCl, DMAP, DCM; 5 h; (vii) BrMg-R, CuBr·Me2S, THF, 0 °C to RT; (viii) Bu4NF, THF, reflux, 1 h; (ix) O2, eosin Y, pyridine, hv, 0 °C, 0.5 h.

based on the biological results. Future work will focus on the synthesis of additional candidate structures with different side-chain to address specific cancer cell lines.

Table 2 In vitro anti-proliferative activities of compounds. Compound

5a 5b 5c 5d 14a 14b 14c 14d 14e 14f EP EG Cis-platinum a

IC50 (μM)a HepG2

SK-Hep1

MCF-7

MDA-MB-231

4. Experiment

22.58 8.07 > 60 > 60 28.94 12.66 34.84 9.50 28.16 31.05 23.15 > 60 2.39

24.86 11.97 > 60 > 60 27.36 17.24 30.32 15.30 23.10 33.16 18.34 > 60 1.46

19.40 12.25 53.36 > 60 32.15 19.68 26.60 10.34 20.50 27.54 17.50 > 60 4.36

17.45 10.90 > 60 > 60 23.22 20.05 24.04 12.25 18.77 25.33 15.24 55.46 3.97

4.1. Chemistry All commercially available reagents were used without further purification. Melting points (uncorrected) were determined on a MP120 auto point apparatus (Hanon instruments Corp., Jinan, China). The 1H NMR and 13C NMR spectra were measured on a Bruker Avance DRX400 spectrometer with TMS and solvent signals allotted as internal standards. The chemical shifts of the 1H NMR and 13C NMR were expressed in ppm (δ). ESI mass spectra were obtained on an Esquire 6000 Mass Spectrometer. HRMS data were measured using a Bruker APEX IV Fourier transform ion cyclotron resonance mass spectrometer. X-ray diffraction data were collected on CrysAlis PRO, Agilent Technologies. Silica gel (300–400 mesh) was used for analytical and flash chromatography.

Data represent the mean values of three independent determinations.

against all the four cancer cell lines respectively. The peroxidic bridge at the C-5 and C-8 position is requisite pharmacophore for the bioactivity. Substituent changes to the C-17 position can affect potency against different kinds of cancer cell lines. In an overall view, peroxidic bridge at the B-ring of the sterol skeleton and the aliphatic side-chain at the C-17 position appeared to play a critical role for the anticancer activity. Preliminary structure-activity relation-ships were put forward

4.2. Synthesis of compounds 4.2.1. Ergosterol peroxide (EP) A solution of ergosterol (150 mg) and eosin Y (1 mg) in pyridine (20 mL) contained in a quartz tube kept in a water-cooled bath was saturated with oxygen. The solution, vigorously stirred by the gas bubbling, was irradiated with light from a 220 V 500 W iodine tungsten lamp placed at a distance of 15 cm. The oxygen was kept bubbling during the irradiation for 0.5 h. The irradiated solution was poured into ice-water and extracted with ethyl acetate. The combined organic extracts were washed with brine and dried over anhydrous Na2SO4, filtered. The filtrate was concentrated in reduced pressure to afford a crude product. Then product was purified by column chromatography on silica gel (50% EtOAc in PE, Rf = 0.51) to give EP as white needles (104 mg, 64%), m.p. 181.5–183 °C [17]. 1H NMR (400 MHz, CDCl3) δ 6.51 (d, J = 8.4 Hz, 1H, ]CHe), 6.24 (d, J = 8.4 Hz, 1H, ]CHe), 5.23 (dd, J = 7.6, 15.2 Hz, 1H), 5.12 (dd, J = 8.0, 15.2 Hz, 1H), 3.97 (tt, J = 5.04, 11.47 Hz, 1H, 3-OH), 1.00 (d, J = 6.4 Hz, 3H, –CH3), 0.91 (d, J = 6.9 Hz, 3H, –CH3), 0.89 (s, 3H, –CH3), 0.84 (d, J = 6.8 Hz, 3H, –CH3), 0.83 (s, 3H, –CH3), 0.82 (d, J = 6.8 Hz, 3H, –CH3). 13C NMR (100 MHz, CDCl3) δ 12.9, 17.6, 18.2, 19.6, 19.9, 20.6, 20.9, 23.4, 28.6, 30.1, 33.1, 34.7, 37.0, 37.0, 39.3, 39.7, 42.8, 44.6, 51.1, 51.7, 56.2,

Table 3 Cytotoxic activity data against 293T cell lines. Compound

CC50 (μM)a 293T

5a 14b 14d 14e EP Cis-platinum

48.50 62.43 59.30 67.44 56.64 23.36

a Data represent the mean values of three independent determinations.

49

Steroids 124 (2017) 46–53

M. Bu et al.

iodine tungsten lamp and kept bubbling oxygen for 0.5 h to get target compound 5d. Taken method as synthesis of EP. The product was purified by column chromatographic on silica gel (50% EtOAc in PE, Rf = 0.27) to give 5d as a white solid (107.3 mg, 64%), m.p. 166.8–168.1 °C. 1H NMR (400 MHz, CDCl3) δ 6.52 (d, J = 8.5 Hz, 1H, ]CHe), 6.33 (d, J = 8.5 Hz, 1H, ]CHe), 3.99 (s, 1H, 3-OH), 2.58–2.46 (m, 1H), 2.23–2.11 (m, 2H), 2.08–1.99 (m, 1H), 1.98–1.94 (m, 1H), 1.88 (m, 1H), 1.86–1.80 (m, 2H), 1.72 (m, 1H), 1.68–1.62 (m, 1H), 1.58–1.55 (m, 4H), 1.54–1.50 (m, 1H), 1.38–1.21 (m, 2H), 1.00 (s, 3H, –CH3), 0.91 (s, 3H, –CH3). 13C NMR (100 MHz, CDCl3) δ 218.4, 136.5, 129.7, 82.4, 79.0, 66.2, 51.4, 48.8, 47.6, 37.1, 36.7, 35.5, 34.6, 31.5, 30.0, 22.6, 18.9, 18.2, 14.9. HRMS (ESI): m/z calculated for C19H28O4 [M+H]+ 319.1909, found 319.1900.

66.4, 79.4, 82.2, 130.7, 132.3, 135.2, 135.4. MS (ESI) m/z: 451.4 [M +Na]+, 467.3 [M+K]+. HRMS (ESI) m/z: calculated for C28H44O3 [M +H]+ 429.3369, found 429.3359. 4.2.2. General procedure for the synthesis of compounds 5a–d Synthesis of compound 5d from the starting material 1d as described in Scheme 2. Then, synthesis of compound 5a–c by the same way of 5d. 4.2.2.1. 3β-Acetyloxy-5-en-androsten-17-one (2d). To a suspension of DHEA (1d) (28.8 g, 0.10 mol) in CH2Cl2 (40 mL) and pyridine (10 mL) was added Ac2O (12 mL, 0.13 mol) over 20 min. After the mixture was stirred in room temperature for 6 h, water (30 mL) was added, and then extracted with ethylacetate. The combined organic layers were washed with saturated NaHCO3 and brine. The organic extracts were dried with anhydrous MgSO4 and concentrated to get crude product 2d as a white solid (32 g). Yield: 98%, m.p. 167.6–170.1 °C. 1H NMR (CDCl3, 400 MHz) δ 5.41 (d, J = 5.1 Hz, 1H), 4.65–4.54 (m, 1H), 2.46 (dd, J = 19.3, 8.6 Hz, 1H), 2.33 (t, J = 7.9 Hz, 2H), 2.10 (dd, J = 18.8, 9.4 Hz, 2H), 2.04 (s, 3H), 1.94 (d, J = 5.7 Hz, 1H), 1.89 (d, J = 2.9 Hz, 1H), 1.87–1.81 (m, 2H), 1.71–1.61 (m, 4H), 1.56 (dd, J = 10.4, 6.4 Hz, 1H), 1.48 (dd, J = 13.3, 4.4 Hz, 1H), 1.33–1.26 (m, 2H), 1.19–1.11 (m, 1H), 1.04 (s, 3H), 0.99 (d, J = 4.6 Hz, 1H), 0.88 (s, 3H). MS (ESI) m/z 353.9 [M+Na]+, 369.7 [M+K]+.

4.2.2.5. 5α,8α-Cyclicobioxygen-6-vinyl-3β-sitosterol (5a). White solid (96.7 mg, 60%), Rf = 0.57 (50% EtOAc in PE), m.p. 138.9–140.5 °C. 1 H NMR (400 MHz, CDCl3) δ 6.51 (d, J = 8.5 Hz, 1H, ]CHe), 6.25 (d, J = 8.5 Hz, 1H, ]CHe), 3.97 (dt, J = 16.3, 5.5 Hz, 1H), 2.11 (dd, J = 13.7, 3.7 Hz, 1H), 2.04–1.78 (m, 4H), 1.74–1.43 (m, 10H), 1.43–1.29 (m, 4H), 1.23 (m, 6H), 1.01 (m, 1H), 0.91 (d, J = 6.4 Hz, 3H, –CH3), 0.88 (s, 3H, –CH3), 0.86 (s, 1H), 0.84 (s, 3H, –CH3), 0.82 (s, 3H, –CH3), 0.80 (s, 6H, 2× –CH3). 13C NMR (100 MHz, CDCl3) δ 135.4, 130.7, 82.2 (–C–O–O–), 79.5 (–C–O–O–), 66.4, 56.3, 51.5, 51.0, 45.7, 44.7, 39.4, 36.9, 36.8, 35.6, 34.7, 33.7, 30.0, 29.1, 28.3, 26.0, 23.4, 23.0, 20.6, 19.8, 19.0, 18.6, 18.2, 12.6, 11.9. HRMS (ESI): m/z calculated for C29H48O3 [M+H]+ 445.3682, found [M+H]+ 445.3670.

4.2.2.2. 3β-Acetoxyandrosta-5,7-diene-17-one (3d). A mixture of 2d (3.2 g, 0.01 mol) and cyclohexane (80 mL) was warmed to 65 °C to obtain a clear solution, then NBS (2.3 g, 0.015 mol) was added and the mixture was heated to reflux for 1 h. The mixture was cooled to room temperature and diluted with 100 mL water. After being stirred for 1 h, the precipitate was collected by filtration and washed with water. Then the solid was dissolved in CH2Cl2, the solution was washed with brine, dried with anhydrous MgSO4, and concentrated to get crude 7α-bromo3β-(acetyl-oxy)androst-5-en-17-one as a light brown solid. The crude product and 1.0 M Bu4NF in THF (10 mL, 1.5 eq) was stirred overnight. Then, the mixture was diluted with cyclohexane, washed with water, dried with anhydrous MgSO4, and concentrated to get crude product as a brown solid. The product was purified by column chromatographic on silica gel (50% EtOAc in PE, Rf = 0.62) to give 3d as a pale yellow solid (1.0 g, 33%), m.p. 167.0–169.2 °C. 1H NMR (CDCl3, 400 MHz) δ 5.6 (s, 1H), 5.57 (d, J = 2.6 Hz, 1H), 4.7 (td, J = 11.5, 5.9 Hz, 1H), 2.56–2.49 (m, 2H), 2.38 (t, J = 12.5 Hz, 1H), 2.24–2.17 (m, 2H), 2.06 (s, 3H), 1.97–1.92 (m, 2H), 1.91 (d, J = 3.7 Hz, 1H), 1.74 (d, J = 4.8 Hz, 2H), 1.70 (d, J = 5.2 Hz, 1H), 1.58 (s, 2H), 1.40–1.34 (m, 2H), 1.27 (s, 1H), 0.98 (s, 3H), 0.82 (s, 3H). MS (ESI) m/z 351.7 [M+Na]+.

4.2.2.6. 5α,8α-Cyclicobioxygen-6-vinyl-3β-cholesterol (5b). White solid (99.2 mg, 61%), Rf = 0.52 (50% EtOAc in PE), m.p. 154.8–155.4 °C. 1 H NMR (400 MHz, CDCl3) δ 6.52 (d, J = 8.5 Hz, 1H, ]CHe), 6.26 (d, J = 8.5 Hz, 1H, ]CHe), 4.08–3.93 (m, 1H), 2.13 (ddd, J = 13.8, 4.9, 1.7 Hz, 1H), 2.11–1.96 (m, 3H), 1.90–1.81 (m, 1H), 1.75–1.63 (m, 2H), 1.61–1.48 (m, 14H), 1.44–1.20 (m, 4H), 1.02 (d, J = 9.1 Hz, 1H), 0.92 (d, J = 6.5 Hz, 3H, –CH3), 0.91–0.88 (m, 6H, 2× –CH3), 0.87 (d, J = 1.8 Hz, 3H, –CH3), 0.82 (s, 3H, –CH3). 13C NMR (100 MHz, CDCl3) δ 135.4, 130.7, 82.2 (–C–O–O–), 79.5 (–C–O–O–), 66.4, 56.4, 51.5, 51.0, 44.7, 39.4, 36.9, 36.8, 35.9, 35.2, 34.7, 30.1, 28.2, 28.0, 23.8, 23.4, 22.8, 22.5, 20.6, 18.6, 18.2, 12.6. HRMS (ESI): m/z calculated for C29H48O3 [M+H]+ 417.3369, found 417.3399. 4.2.2.7. 5α,8α-Cyclicobioxygen-6-vinyl-3β-PREG (5c). White soli-de (94.4 mg, 57%), Rf = 0.29 (50% EtOAc in PE), m.p. 162.4–163.6 °C, 1 H NMR (400 MHz, CDCl3) δ 6.47 (d, J = 8.5 Hz, 1H, ]CH-), 6.27 (d, J = 8.5 Hz, 1H, ]CH-), 4.07–3.89 (m, 1H 3-OH), 2.59 (t, J = 9.1 Hz, 1H), 2.33–2.19 (m, 1H), 2.13 (s, 3H, C20–CH3), 0.89 (s, 3H,C19–CH3), 0.76 (s, 3H, C18–CH3). 13C NMR (100 MHz, CDCl3) δ 208.6, 135.8, 130.1, 82.2, 79.0, 66.3, 63.5, 51.7, 51.1, 45.6, 38.5, 37.0, 36.9, 34.7, 31.5, 30.1, 23.3, 23.1, 20.9, 18.2, 14.2. HRMS (ESI): m/z calculated for C21H30O4 [M+H]+ 347.2144, found 347.2215.

4.2.2.3. 3β-Hydroxyandrosta-5,7-diene-17-one (4d). To a suspension of intermediate 3d (1.0 g) in 10 mL of methanol was added 25% (wt) NaOMe in MeOH (5 mL). After the mixture was stirred overnight, 10 mL of water was added dropwise over 1 h. After the mixture was stirred for 2 h, the precipitate was collected by filtration, washed with MeOH-H2O (1:2), and dried at 40 °C under high vacuum to get crude product as a brown solid. The product was purified by column chromatographic on silica gel (50% EtOAc in PE, Rf = 0.38) to give 4d as a pale yellow solid (0.92 g, 96%), m.p. 156.3–157.9 °C. 1H NMR (400 MHz, CDCl3) δ 6.00 (1H, d, J = 9.8 Hz, C6-H), 5.69 (1H, d, J = 9.8 Hz, C7-H), 4.29 (1H, t, J = 7.9 Hz, C3-αH), 3.75–3.58 (1H, m, C3-OH), 2.66–2.40 (2H, m, C16-H), 1.03 (3H, s, C19–CH3), 0.95 (3H, s, C18–CH3). MS (ESI) m/z: 309.8 [M+Na]+, 325.8 [M+K]+.

4.2.3. General procedure for the synthesis of compounds 14a–f Synthesis of compound 14d from the starting material 1d as described in Scheme 3. Then, synthesis of compound 14a–c and 14e–f by the same way of 14d. 4.2.3.1. 3β-(tert-Butyldimethylsilyl)oxy-androst-5-en-17-one (6). A solution of dehydroepiandrosterone (1d). (28.8 g, 0.1 mol), TBSCl (16.6 g, 0.11 mol, 1.1 eq) and imidazole (14.9 g, 0.22 mol, 2.2 eq) in THF (200 mL) was stirred to reflux under argon for 3 h. The crude was evaporated and extracted with EtOAc. The combined extracts were washed twice with brine, dried over MgSO4 and evaporated. The crude

4.2.2.4. 5α,8α-Cyclicobioxygen-6-vinyl-3β-DHEA (5d). Finally, photoox idation of 4d (150 mg) with eosin Y (1 mg) in pyridine irradiated with

50

Steroids 124 (2017) 46–53

M. Bu et al.

the diene 10 as white solid powder (1.0 g, 35%), m.p. 155.7–158.2 °C. H NMR (400 MHz, CDCl3) δ 5.55 (s, 1H), 5.40 (s, 1H), 3.62 (d, J = 19.0 Hz, 2H), 3.40 (s, 1H), 2.35 (s, 2H), 2.08 (m, 1H), 1.86 (m, 4H), 1.73 (m, 3H), 1.59 (m, 4H), 1.46–1.36 (m, 2H), 1.29 (m, 4H), 1.12–1.04 (m, 3H), 0.94 (d, J = 2.9 Hz, 3H), 0.90 (s, 9H), 0.64 (d, J = 2.7 Hz, 3H), 0.14–0.05 (m, 6H). MS (ESI) m/z [M+H]+ 445.34.

product was purified by silica gel chromatography (25% EtOAc in PE, Rf = 0.72) to give 6 as white solid (37.4 g, 93%), m.p. 152.2–155.6 °C. 1 H NMR (400 MHz, CDCl3) δ 5.35 (d, J = 4.9 Hz, 1H), 3.57–3.38 (m, 1H), 2.46 (dd, J = 19.2, 8.8 Hz, 1H), 2.28 (m, 1H), 2.23–2.16 (m, 1H), 2.10 (dd, J = 18.7, 9.5 Hz, 2H), 2.00–1.90 (m, 1H), 1.88–1.79 (m, 2H), 1.77–1.62 (m, 4H), 1.61–1.43 (m, 4H), 1.29 (m, 2H), 1.08 (dd, J = 13.6, 3.6 Hz, 1H), 1.03 (s, 3H), 0.89 (d, J = 2.7 Hz, 12H), 0.06 (s, 6H). MS (ESI) m/z 425.9 [M+Na]+, 441.7 [M+K]+.

1

4.2.3.6. 3β-((tert-Butyldimethylsilyl)oxy)-22-tosyloxy-23,24-bisn-orchol5,7-dien (11). A solution of 10 (1.0 g, 2.2 mmol), TsCl (629 mg, 3.3 mmol, 1.5 eq) and 4-dimethylaminopridine (DMAP 670 mg, 5.5 mmol, 2.5 eq) in CH2Cl2 (60 mL) was allowed to stirred overnight at room temperature. The mixture was poured into water and extracted with CH2Cl2. The combined extracts were washed with brine, dried over MgSO4 and evaporated. The crude product was purified by silica gel flash chromatograph (20% EtOAc in PE, Rf = 0.60) to give the tosylate of 11 as white needles (1.25 g, 95%), m.p. 142.6–144.9 °C. 1H NMR (400 MHz, CDCl3) δ 7.79 (d, J = 8.1 Hz, 1H), 7.35 (d, J = 7.8 Hz, 1H), 5.54 (d, J = 5.5 Hz, 1H), 5.36 (s, 1H), 3.98 (m, 1H), 3.83 (m, 1H), 2.45 (s, 3H), 2.33 (m, 2H), 1.01 (d, J = 6.4 Hz, 3H), 1.01–0.79 (m, 12H), 0.58 (s, 3H), 0.06 (s, 6H). MS (ESI) m/z 623.4 [M+Na]+, 639.2 [M+K]+.

4.2.3.2. (17Z)-3β-((tert-Butyldimethylsilyl)oxy)pregna-5,17(20)-diene (7). A mixture of 6 (34.7 g, 0.086 mol), (ethyl)triphenyl-phosphonium bromide (96.0 g, 0.259 mol, 3.0 eq) and t-BuOK (26.0 g, 0.23 mol, 2.7 eq) in THF (200 mL) was allowed to reflux under argon for 4 h. After being cooled, the mixture was filtered and the filtrate was evaporated. The crude product was purified by column chromatographic on silica gel (20% EtOAc in PE, Rf = 0.80) to give 7 as a white solid (32.0 g, 90%), m.p. 147.4–150.1 °C. 1H NMR (400 MHz, CDCl3) δ 5.33 (d, J = 16.2 Hz, 1H), 5.15 (d, J = 7.0 Hz, 1H), 3.55–3.45 (m, 1H), 2.40 (m, 1H), 2.28 (m, 2H), 2.19 (m, 2H), 2.04 (m, 1H), 1.84 (m, 1H), 1.67 (m, 4H), 1.63–1.47 (m, 6H), 1.28–1.15 (m, 2H), 1.03 (s, 3H), 0.91 (s, 15H), 0.08 (s, 6H). MS (ESI) m/z 437.3 [M +Na]+, 453.2 [M+K]+.

4.2.3.7. 3β-((tert-butyldimethylsilyl)oxy)-23,24-bisnorchol-5,7-dien -22isobutyl (12d). A solution of 11 (250 mg, 0.42 mmol) in THF (20 mL) was stirred at 0 °C, a suspension of CuBr·Me2S (86.1 mg, 0.42 mmol, 1.0 eq) in THF was added and a solution of isobutylmagnesium bromide in THF (1 M) was added dropwise. After being stirred at room temperature for 1 h, the reaction mixture was poured into saturated aqueous NH4Cl at 0 °C and aqueous layer was extracted twice with EtOAc. The combined organic layer was washed with saturated aqueous NH4Cl, saturated aqueous NaHCO3 and brine, dried over MgSO4 and evaporated. The crude product was purified by silica gel chromatography (30% EtOAc in PE, Rf = 0.57) to give 12d (174 mg, 86%) as a white solid, m.p. 167.5–170.4 °C. 1H NMR (400 MHz, CDCl3) δ 5.57 (s, 1H), 5.40 (s, 1H), 3.62 (m, 1H), 2.35 (d, J = 12.5 Hz, 2H), 1.95–1.10 (m, 21H), 0.88–0.95 (m, 18H), 0.63 (m, 3H), 0.08 (s, 9H). MS (ESI) m/z [M+H]+ 485.37.

4.2.3.3. 3β-((tert-Butyldimethylsilyl)oxy)-22-hydroxy-23,24-bisn-orchol5,16(17)-diene (8). A suspension of 7 (6.12 g, 11.3 mmol) and paraformaldehyde (2.12 g, 56.3 mmol, 5.0 eq) in CH2Cl2 (612 mL) was treated with BF3·Et2O (0.14 mL, 1.1 mmol, 0.1 eq) and the mixture stirred under argon at room temperature for 15 min. The reaction mixture was poured into water and extracted with CH2Cl2. The combined extracts were washed with brine, dried over MgSO4 and evaporated. The crude product was purified by column chromatographic on silica gel (40% EtOAc in PE, Rf = 0.55) to give 8 as white solid (4.27 g, 85%), m.p. 156.0–158.3 °C. 1H NMR (400 MHz, CDCl3) δ 5.45 (s, 1H), 5.34 (s, 1H), 3.57 (m, 2H), 3.49 (dd, J = 13.6, 8.4 Hz, 1H), 2.41 (d, J = 6.6 Hz, 1H), 1.04 (s, 6H), 0.89 (s, 12H), 0.82 (s, 3H), 0.06 (s, 6H).MS (ESI) m/z 467.5 [M+Na]+. 4.2.3.4. 3β-((tert-Butyldimethylsilyl)oxy)-22-hydroxy-23,24-bisn-orchol5-en (9). A mixture of 8 (5.5 g, 12.3 mmol) and 5% Pt/C (2.75 g) in EtOH (200 mL) was stirred overnight under hydrogen at room temperature. The mixture was filtered through Celite and the filtrate was evaporated. The crude product was purified by column chromatographic on silica gel (40% EtOAc in PE, Rf = 0.50) to give 9 as white solid (5.0 g, 92%), m.p. 157.2–160.4 °C. 1H NMR (400 MHz, CDCl3) δ 5.32 (s, 1H), 3.64 (dd, J = 10.4, 2.7 Hz, 1H), 3.54–3.42 (m, 1H), 3.37 (m, 1H), 2.25 (d, J = 11.7 Hz, 1H), 2.21–2.13 (m, 1H), 1.99 (m, 2H), 1.81 (d, J = 12.9 Hz, 2H), 1.72 (d, J = 12.4 Hz, 1H), 1.51 (m, 9H), 1.37–1.24 (m, 2H), 1.25–1.15 (m, 2H), 1.05 (d, J = 6.5 Hz, 3H), 1.00 (s, 3H), 0.89 (s, 9H), 0.70 (s, 3H), 0.06 (s, 6H).

4.2.3.8. 3β-Hydroxy-23,24-bisnorchol-5,7-dien-22-isobutyl (13d). A solution of the 12d (174 mg, 0.36 mmol) in THF (30 mL) was added 1.0 M Bu4NF/THF (10 mL) and the mixture was heated at 70 °C for 1 h. After being cooled, the mixture was poured into water and extracted with EtOAc. The combined extracts were washed with brine, dried over MgSO4 and evaporated. The crude product was purified by silica gel flash chromatography (25% EtOAc in PE, Rf = 0.60) to give 13d as white solid (130.0 mg, 97%), m.p. 169.5–172.7 °C. 1H NMR (400 MHz, CDCl3) δ 5.57 (s, 1H), 5.39 (s, 1H), 3.64 (d, J = 11.1 Hz, 1H), 2.47 (d, J = 12.5 Hz, 1H), 2.28 (m, 1H), 1.97–1.03 (m, 20H), 0.93 (m, 6H), 0.87 (m, 9H), 0.62 (s, 3H). MS (ESI) m/z [M+H]+ 371.32.

4.2.3.5. 3β-((tert-Butyldimethylsilyl)oxy)-22-hydroxy-23,24-bisn-orchol5,7-diene (10). A mixture of 9 (3.0 g, 6.7 mmol), NBS (1.9 g, 10.7 mmol, 1.6 eq) and NaHCO3 (6 eq) in n-hexane (150 mL) was allowed to reflux for 1 h. The mixture was filtered, then the filtrate washed successively with saturated NaHCO3 and water, dried over MgSO4, and evaporated. The residue was dissolved in THF (30 mL), then n-Bu4NBr (0.1 eq) was added to the solution and the mixture was stirred under argon at 0 °C for 15 min. Then 1.0 M n-Bu4NF/THF (7 eq) was added and the whole was stirred under argon at 0 °C for 3.5 h. The reaction mixture was poured into water and extracted with AcOEt. The combined extracts were washed with brine, dried over MgSO4 and evaporated. The crude product was purified by column chromatographic on silica gel (40% EtOAc in PE, Rf = 0.63) to give

4.2.3.9. 3β-Hydroxy-5α,8α-epidioxyandrost-22-isobutyl (14d). For the synthesis of the title compound 14d, a mixture of 13d (100 mg, 0.23 mmol) and eosine (1 mg) in pyridine (10 mL) in a quartz tube kept in a water-cooled bath. Then the tube was irradiated with an iodine tungsten lamp. The oxygen was kept bubbling during the irradiation for 0.5 h. The mixture was then poured into cold water and extracted with ethyl acetate. The combined organic fractions were dried over anhydrous MgSO4 and the solvent was removed under reduced pressure. The product was purified by column chromatographic on silica gel (50% EtOAc in PE, Rf = 0.49) to give 14d as a white solid (55 mg, yield 58%), m.p. 174.3–176.0 °C. 1H NMR (400 MHz, CDCl3) δ 6.50 (d, J = 8.5 Hz, 1H, ]CH-), 6.24 (d, J = 8.5 Hz, 1H, ]CH-), 4.02–3.91 (m, 1H, −OH), 2.11 (dd,

51

Steroids 124 (2017) 46–53

M. Bu et al.

2H), 1.56–1.46 (m, 5H), 1.40–1.31 (m, 2H), 1.26 (s, 12H), 1.21–1.14 (m, 2H), 1.01 (d, J = 9.3 Hz, 1H), 0.89 (d, J = 7.7 Hz, 9H), 0.80 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 135.4, 130.7, 82.2, 79.5, 66.4, 56.3, 51.5, 51.0, 44.7, 39.4, 37.0, 36.9, 35.7, 35.2, 34.7, 31.9, 30.1, 29.4, 28.2, 26.0, 23.4, 22.7, 20.6, 18.6, 18.2, 14.2, 12.6. HRMS (ESI) m/z calculated for C28H47O3 [M+H]+ 431.3525, found 431.3521.

J = 13.8, 3.7 Hz, 1H, C3-αH), 0.87 (dd, J = 14.9, 8.0 Hz, 12H), 0.80 (s, 3H, C20–CH3). 13C NMR (100 MHz, CDCl3) δ 135.4, 130.7, 82.2, 79.5, 66.4, 56.2, 51.5, 51.0, 44.7, 39.4, 36.9, 36.8, 35.3, 35.2, 34.7, 33.3, 30.0, 28.3, 28.2, 23.4, 23.1, 22.3, 20.6, 18.6, 18.2, 12.6. HRMS (ESI) m/z calculated for C26H42O3 [M+H]+ 403.3212, found [M+H]+ 403.3215. Synthesis of endoperoxides 14a–c and 14e-f by the same way of 14d.

4.3. Biological assays 4.2.3.10. 3β-Hydroxy-5α,8α-epidioxyandrost-22-n-propyl (14a). White solid (65.4 mg, 60%), Rf = 0.47 (50% EtOAc in PE), m.p. 166.2–168.3 °C. 1H NMR (400 MHz, CDCl3) δ 6.50 (d, J = 8.5 Hz, 1H, ]CH-), 6.24 (d, J = 8.4 Hz, 1H, ]CH-), 4.02–3.89 (m, 1H, -OH), 2.16–2.06 (m, 1H), 1.98 (d, J = 5.8 Hz, 1H), 1.94 (s, 1H), 1.91 (s, 1H), 1.84 (d, J = 10.3 Hz, 1H), 1.72–1.61 (m, 2H), 1.55 (m, 1H), 1.49 (m, 3H), 1.41–1.32 (m, 4H), 1.26 (s, 2H), 1.20 (m, 4H), 1.15 (m, 2H), 1.01 (dd, J = 13.0, 8.6 Hz, 2H), 0.88 (s, 9H), 0.80 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 135.42, 130.75, 82.19, 79.47, 66.41, 56.37, 51.58, 51.07, 44.73, 39.43, 36.94, 36.90, 35.40, 35.15, 34.70, 30.06, 28.28, 28.22, 23.41, 23.11, 20.61, 18.58, 18.17, 14.16, 12.62. HRMS (ESI) m/z calculated for C25H41O3 [M+H]+ 389.3056, found 389.3060.

The effect of derivatives on cell proliferation was evaluated by MTT (3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide assay. Human hepatocellular cancer cells (HepG2, SK-Hep1) and human breast cancer cells (MCF-7, MDA-MB-231) were used in the study. The cells were cultured in DMEM/RPMI1640 supplemented with 10% FBS, 100 U/mL penicillin/streptomycin at 37 °C in an incubator containing 5% CO2. Cells (1 × 104 cells/mL) were seeded into 96-well plates and were incubated at 37 °C overnight in a humidified incubator containing 5% CO2. Cells were dosed with compounds at final concentrations ranging from 5 μM to 60 μM in each well of the plates. The cells were incubated for various periods and analyzed by MTT assay to analyze rates of cell proliferation as described. Cell survival was determined by measuring the absorbance at 490 nm using a microplate reader. A calibration curve was prepared using the SPSS to determine the IC50 of the target compounds. Cytotoxicity as IC50 for each cell line is the concentration of compound which reduced by 50% the optical density of treated cells (48 h) with respect to untreated cells using the MTT assay.

4.2.3.11. 3β-Hydroxy-5α,8α-epidioxyandrost-22-isopropyl (14b). White solid (67.5 mg, 62%), Rf = 0.38 (50% EtOAc in PE), m.p. 159.5–161.6 °C. 1H NMR (400 MHz, CDCl3) δ 6.53 (d, J = 8.5 Hz, 1H, ]CH-), 6.26 (d, J = 8.5 Hz, 1H, ]CH-), 4.02–3.94 (m, 1H, −OH), 2.13–1.13 (m, 20H), 0.97 (dd, J = 13.2, 5.8 Hz, 2H), 0.91–0.86 (m, 9H), 0.82 (m, 6H). 13C NMR (100 MHz, CDCl3) δ 135.42, 130.74, 82.20, 79.48, 66.38, 56.29, 51.55, 51.01, 45.75, 44.72, 39.39, 36.92, 36.88, 35.59, 34.68, 33.69, 30.04, 29.09, 28.28, 26.00, 23.40, 23.02, 20.63, 19.84, 19.02, 18.64, 18.18, 12.63, 11.97. HRMS (ESI) m/z calculated for C25H40O3 [M+H]+ 389.3056, found 389.3059.

Acknowledgments The authors would like to acknowledge financial support from the Chinese Natural Science Foundation Project (21272020), and Beijing Key Laboratory for Green Catalysis and Separation. Appendix A. Supplementary data

4.2.3.12. 3β-Hydroxy-5α,8α-epidioxyandrost-22-allylique (14c). White solid (60.0 mg, 55%), Rf = 0.60 (50% EtOAc in PE), m.p. 154.1–155.8 °C. 1H NMR (400 MHz, CDCl3) δ 6.50 (d, J = 8.5 Hz, 1H, ]CH-), 6.24 (d, J = 8.5 Hz, 1H, ]CH-), 5.79 (m, 1H, C24-H), 4.96 (dd, J = 27.3, 13.6 Hz, 2H, C25-CH2), 4.05–3.89 (m, 1H, –OH), 2.15–2.08 (m, 2H), 2.01–1.90 (m, 4H), 1.82 (m, 1H), 1.69 (m, 1H), 1.62 (s, 1H), 1.47 (m, 9H), 1.21–1.25 (m, 4H), 1.15–1.06 (m, 1H), 0.92 (d, J = 6.4 Hz, 3H), 0.88 (s, 3H), 0.80 (s, 3H). 13C NMR (100 MHz, CDCl3) δ 139.3, 135.4, 130.7, 114.1, 82.2, 79.4, 66.4, 56.3, 51.6, 51.0, 44.8, 39.4, 36.9, 36.9, 34.9, 34.8, 34.7, 30.4, 30.1, 28.2, 23.4, 20.6, 18.4, 18.2, 12.6. HRMS (ESI) m/z calculated for C25H41O3 [M+H]+ 387.3014, found 387.3040.

Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.steroids.2017.05.013. References [1] T. Liu, G. Wu, C. Yu, Z. Huang, W. Li, Q. Men, X. Liu, Tetrandrine induces autophagy and differentiation by activating ROS and Notch1 signaling in leukemia cells, Oncotarget 6 (2015) 7992–8006. [2] J. Wu, Y. Wu, B.B. Yang, Anticancer activity of Hemsleya amabilis extract, Life Sci. 71 (2002) 2161–2170. [3] F. Nuter, A.K.D. Dime, C. Chen, L. Bounaadja, E. Mouray, I. Florent, Y. Six, O. Buriez, A. Marinetti, A. Voituriez, Access to new endoperoxide derivatives by electrochemical oxidation of substituted 3-azabicyclo[4.1.0]hept-4-enes, Chem.Eur. J. 21 (2015) 5584–5593. [4] I. Fernandez, A. Robert, Peroxide bond strength of antimalarial drugs containing an endoperoxide cycle. Relation with biological activity, Org. Biomol. Chem. 9 (2011) 4098–4107. [5] E. Callaway, D. Cyranoski, China celebrates first Nobel, Nature 526 (2015) 174–175. [6] V.M. Dembitsky, Bioactive natural peroxides as potential therapeutic agents, Eur. J. Med. Chem. 43 (2008) 223–251. [7] M. Jung, H. Kim, K. Lee, M. Park, Naturally occurring peroxides with biological activities, Mini-Rev. Med. Chem. 3 (2003) 159–165. [8] V.M. Dembitsky, T.A. Gloriozova, V.V. Poroikov, Natural peroxy anticancer agents, Mini-Rev. Med. Chem. 7 (2007) 571–589. [9] M. Bu, B.B. Yang, L.M. Hu, Natural endoperoxides as drug lead compounds, Curr. Med. Chem. 23 (2016) 383–405. [10] R. Nowak, M. Drozd, E. Mendyk, M. Lemieszek, O. Krakowiak, W. Kisiel, W. Rzeski, K. Szewczyk, A new method for the isolation of ergosterol and peroxyergosterol as active compounds of Hygrophoropsis aurantiaca and in vitro antiproliferative activity of isolated ergosterol peroxide, Molecules 21 (2016) 946/1–10. [11] S.W. Kim, S.S. Park, T.J. Min, K.H. Yu, Antioxidant activity of ergosterol peroxide (5,8-epidioxy-5α,8-α-ergosta-6,22E-dien-3β-ol) in Armillariella mellea, Bull. Korean Chem. Soc. 20 (1999) 819–823. [12] T. Takei, M. Yoshida, M. Ohnishi-Kameyama, M. Kobori, Ergosterol peroxide, an apoptosis-induced component isolated from Sarcodon aspratus (Berk.) S. Ito, Biosci. Biotechnol. Biochem. 69 (2005) 212–215.

4.2.3.13. 3β-Hydroxy-5α,8α-epidioxyandrost-22-n-pentyl (14e). White solid (65.0 mg, 60%), Rf = 0.55 (50% EtOAc in PE), m.p. 177.3–179.4 °C. 1H NMR (400 MHz, CDCl3) δ 6.50 (d, J = 8.5 Hz, 1H, ]CH-), 6.24 (d, J = 8.5 Hz, 1H, ]CH-), 4.02–3.89 (m, 1H, –OH), 2.10 (dd, J = 13.8, 4.1 Hz, 1H), 2.00–1.78 (m, 6H), 1.72–1.58 (m, 2H), 1.50 (m, 4H), 1.36 (m, 5H), 1.29–1.14 (m, 10H), 1.05–0.97 (m, 1H), 0.89 (d, J = 8.9 Hz, 9H), 0.80–0.75 (s, 3H); 13C NMR (100 MHz, CDCl3) δ 135.4, 130.7, 82.2, 79.5, 66.4, 56.4, 51.6, 51.1, 44.7, 39.4, 36.9, 36.9, 35.7, 35.2, 34.7, 31.9, 30.1, 29.8, 28.2, 26.0, 23.4, 22.7, 20.6, 18.6, 18.2, 14.1, 12.6. HRMS (ESI): m/z calculated for C27H45O3 [M+H]+ 417.3369, found [M+H]+ 417.3361. 4.2.3.14. 3β-Hydroxy-5α,8α-epidioxyandrost-22-n-hexyl (14f). White solid (61.6 mg, 57%), Rf = 0.53 (50% EtOAc in PE), m.p. 172.6–174.3 °C. 1H NMR (400 MHz, CDCl3) δ 6.51 (d, J = 8.4 Hz, 1H, ]CH-), 6.24 (d, J = 8.4 Hz, 1H, ]CH-), 4.02–3.91 (m, 1H), 2.11 (dd, J = 13.5, 4.4 Hz, 1H), 1.96 (m, 4H), 1.84 (m, 2H), 1.72–1.59 (m,

52

Steroids 124 (2017) 46–53

M. Bu et al. [13] D.H. Kim, S.J. Jung, I.S. Chung, Y.H. Lee, D.K. Kim, S.H. Kim, B.M. Kwon, T.S. Jeong, M.H. Park, N.S. Seoung, N.I. Baek, Ergosterol peroxide from flowers of Erigeron annuus L. as an anti-atherosclerosis agent, Arch. Pharmacal. Res. 28 (2005) 541–545. [14] S. Zhao, G. Ye, G. Fu, J.X. Cheng, B.B. Yang, C. Peng, Ganoderma lucidum exerts anti-tumor effects on ovarian cancer cells and enhances their sensitivity to cisplatin, Int. J. Oncol. 38 (2011) 1319–1327. [15] R. Zhu, R. Zheng, Y. Deng, Y. Chen, S. Zhang, Ergosterol peroxide from Cordyceps cicadae ameliorates TGF-β1-induced activation of kidney fibroblasts, Phytomedicine 21 (2014) 372–378. [16] Q.P. Wu, Y.Z. Xie, Z. Deng, X.M. Li, W. Yang, C.W. Jiao, L. Fang, S.Z. Li, H.H. Pan, A.J. Yee, D.Y. Lee, C. Li, Z. Zhang, J. Guo, B.B. Yang, Ergosterol peroxide isolated from Ganoderma lucidum abolishes microRNA miR-378-mediated tumor cells on chemoresistance, PLoS One 7 (2012) e44579. [17] X.M. Li, Q.P. Wu, M. Bu, L.M. Hu, W.W. Du, C.W. Jiao, H.H. Pan, M. Sdiri, N. Wu, Y.Z. Xie, B.B. Yang, Ergosterol peroxide activates Foxo3-mediated cell death

[18] [19] [20] [21]

[22]

[23]

53

signaling by inhibiting AKT and c-Myc in human hepatocellular carcinoma cells, Oncotarget 7 (2016) 33948–33959. P.R. Ogilby, Singlet oxygen: there is indeed something new under the sun, Chem. Soc. Rev. 39 (2010) 3181–3209. D.P. Hari, B. Koenig, Synthetic applications of eosin Y in photoredox catalysis, Chem. Commun. 50 (2014) 6688–6699. W. Adam, M. Prein, π-Facial diastereoselectivity in the [4+2] cycloaddition of singlet oxygen as a mechanistic probe, Acc. Chem. Res. 29 (1996) 275–283. K. Konno, K. Ojima, T. Hayashi, H. Takayama, An alternative and efficient synthesis of 24,24-difluoro-1α,25-dihydroxyvitamin D3, Chem. Pharm. Bull. 40 (1992) 1120–1124. A.D. Batcho, D.E. Berger, S.G. Davoust, P.M. Wovkulich, M.R. Uskokovic, Stereoselective introduction of steroid side chains at C (17) and C (20), Helv. Chim. Acta 64 (2004) 1682–1687. S. Fuse, Y. Mifune, N. Tanabe, T. Takahashi, Continuous-flow synthesis of activated vitamin D3 and its analogues, Org. Biomol. Chem. 10 (2012) 5205–5211.