Synthesis and biological activity evaluation of novel peroxo-bridged derivatives as potential anti-hepatitis B virus agents

Previous studies have demonstrated that natural steroid compounds containing a peroxide bridge exhibited potential anti-hepatitis B virus activity.

Abstract

Previous studies have demonstrated that natural steroid compounds containing a peroxide bridge exhibited potential anti-hepatitis B virus activity. To continue our research, a simple and regioselective methodology, using Eosin Y as a clean photosensitized oxidation catalyst, was developed for the synthesis of a peroxide bridge in steroids. The method that using Eosin Y as the catalyst was exposed to visible light and furbished in high yields, did not involve tedious work-up or purification, and avoided using environmentally hazardous solvents. It can be regarded as a green protocol. Moreover, a series of cholesterol and β-sitosterol derivatives containing a peroxide bridge were synthesized using this method and screened for their anti-HBV activity. Among the compounds synthesized in this research, 5α,8α-cyclicobioxygen-6-vinyl-3-oxo-cholesterone (1f, 3.13 μg ml^–1) had the most potent activity with inhibition rates of 77.45% ± 6.01% and 58.73% ± 8.64% on the secretion of HBsAg and HBeAg antigens, respectively, after 8 days. Further acute toxicity test showed that the LD₅₀ value of compound 1f was 362.46 mg kg^–1 after an intraperitoneal injection in mice. Moreover, structure–activity relationships of cholesterol and β-sitosterol derivatives were briefly discussed.

Introduction

Steroids, a widespread class of natural organic compounds occurring in animals, plants and fungi, play a highly significant role in drug discovery.1–3 These compounds have shown remarkable biological activities, including antiviral, anti-tumor, antihypertensive and anticholinergic.4–8 In previous studies, a new steroid compound (alcohol from ascidian, SC) with a peroxide bridge was discovered, which exhibited potent anti-hepatitis B virus activity.9–11 However, due to the low content of SC in Styela clava,12 it stimulated our interest in synthesizing its structural analogs. Cholesterol and β-sitosterol have the same parent nucleus as SC (Fig. 1). Therefore, they are expected to be anti-HBV drugs by building peroxide bridges.

Fig. 1. The structures of SC, cholesterol, β-sitosterol and the novel peroxo-bridged derivatives.

In our present study, a new method was established to synthesize a peroxide bridge using eosin Y as a photosensitive catalytic oxidizer, which has been patented.13 The method was operated at room temperature under visible light, which was simple and environmentally friendly compared to the conventional synthetic methods.14–16 The experimental results from the synthesis indicated that eosin Y could combine with oxygen in air and then oxidize dienes into peroxide bridges so that the dye might be regarded as a specific catalytic to diene ring in the 5,8 position of cholesterol and β-sitosterol. Therefore, we will continue to focus on its application in other steroids and optimize chemistry technology for synthesizing a peroxide bridge.

In our continuous effort, a series of novel cholesterol and β-sitosterol derivatives with peroxide bridges were synthesized using the new method. All the new derivatives were fully characterized by ¹H-NMR, ¹³C-NMR, 2D-NMR spectroscopy (HMQC, HMBC and ¹H-¹H COSY) and HRMS. Their inhibitory effects on the expression of HBsAg and HBeAg in the HepG2.2.15 cells were evaluated by MTS and ELISA assays. Moreover, the acute toxic test, as a part of the safety evaluation of peroxo-bridged derivatives, was carried out to investigate the potential toxicity by an intraperitoneal injection in ICR mice. In addition, the structure–activity relationships of these novel compounds were also briefly discussed.

Results and discussion

Chemistry

All designed derivatives were synthesized via the routes outlined in Scheme 1. Compounds 1a and 2a were prepared via acetylation reaction of cholesterol and β-sitosterol, which then underwent free radical elimination reaction with N-bromosuccinimide (NBS) to afford compounds 1b and 2b. Subsequently, products 1c and 2c were obtained after deacetylation reaction of 1b and 2b. Then 1b and 2b were further reacted with eosin Y under visible light to generate the target compounds 1d and 2d. Compounds 1e and 2e were obtained via two different routes (1): a–b–c–d and (2): a–b–d–c with different yields. Compounds 1f and 2f were eventually synthesized from 1e and 2e with chromic acid solution in acetone. All new derivatives were fully characterized with ¹H-NMR, ¹³C-NMR, 2D-NMR spectra (HMQC, HMBC and ¹H-¹H COSY) and HRMS.

Scheme 1. Synthesis of the cholesterol and β-sitosterol derivatives. Reagents and conditions: (a) toluene, acetic anhydride, pyridine, reflux, 4 h. (b) CCl₄, N-bromosuccinimide (NBS), lighted, reflux, 4 h. (c) Eosin Y, anhydrous ethanol, lighted, room temperature, 10 h. (d) Ethanol, 10% sodium hydroxide, 80 °C, 30 min. (e) Acetone, chromic acid, ice-bath, 3 h.

The molecular formula of 1f was assigned as C₂₇H₄₂O₃ on the basis of the positive high-resolution mass spectrometer (HR-MS) m/z 437.30087 [M + Na]⁺. Product 1f was further identified by HMQC, HMBC and ¹H-¹H COSY 2D-NMR spectra. Combined with the significant HMQC, HMBC and ¹H-¹H COSY correlations, CH-6 [δH 6.53 (d, 1H, J = 8.5 Hz, CH–)] was connected with δC 83.4 (C-5), and CH-7 [δH 6.27 (d, 1H, J = 8.5 Hz, CH–)] was connected with δC 80.0 (C-8), which were similar with 1b. Furthermore, there were δC 83.4 (C-5) and δC 80.0 (C-8) in the ¹³C-NMR spectrum of compound 1f and δC 141.6 (C-5) and δC 138.6 (C-8) in the ¹³C-NMR spectrum of 1b. The abovementioned analysis suggests that the peroxide bridge of compound 1f was synthesized. This inference could also be applied to compounds 1d and 1e.

Within the framework of green chemistry, solvents occupy a strategic place. Green chemistry aims to replace hazardous and/or harmful solvents with more environmentally friendly alternatives. In our experiment, eosin Y, a type of dye,17 was used as a photosensitive catalytic oxidizer.18,19 Eosin Y as a photosensitive catalytic oxidizer simplifies the reaction and avoids the use of H₂SO₄, H₂O₂ and expensive materials, such as MTO and Re₂O₇,20,21 which is regarded as a relatively green process. In addition, the possible mechanism involved in the reaction is that eosin Y activated oxygen in air, and then the structure of conjugated dienes was peroxidized to obtain corresponding products. Furthermore, the synthetic structures of the products indicate that the Eosin Y might catalyze diene specifically.

Biological activity

The cytotoxicity of all the compounds in HepG2.2.15 cells

The cytotoxicity of all the compounds was observed by a standard MTS method. After incubating with various concentrations of these synthetics, the inhibition ratios of the HepG2.2.15 cells exposed to different values of the compounds are shown in Fig. 2. There was a dose-effect relationship in all derivatives on cytotoxicity. These results were used to determine the treatment concentration of these six compounds for the subsequent experiments.

Fig. 2. Cytotoxicity effect of 1d–1f and 2d–2f on HepG2.2.15 cells line.

Inhibition of HBsAg and HBeAg production in HepG2.2.15 cells

As shown in Table 1, all derivatives with a peroxide bridge exhibited higher activity on anti-HBV than the others. Current study proved that introducing a peroxide bridge to an SC parent steroid nucleus could induce the anti-hepatitis B virus activity. Furthermore, it also proved that various 3-substituents could influence the peroxo-bridged derivatives' biological activity.

Table 1. Inhibitory effect of 1d–1f and 2d–2f on HBsAg and HBeAg secretion in HepG2.2.1 cells at 3.13 μg mL^–1.

Compound	4 d		8 d
	HBsAg (%)	HBeAg (%)	HBsAg (%)	HBeAg (%)
1d	12.96 ± 3.45	3.51 ± 2.65	13.77 ± 4.23	7.56 ± 4.78
1e	1.20 ± 4.67	1.35 ± 4.87	1.56 ± 3.12	2.55 ± 5.01
1f	72.28 ± 6.00	54.10 ± 5.06	77.45 ± 6.01	58.73 ± 8.64
2d	26.67 ± 2.98	25.64 ± 3.23	28.79 ± 4.34	25.78 ± 3.54
2e	19.01 ± 2.65	2.88 ± 5.45	20.04 ± 3.12	10.35 ± 3.56
2f	1.91 ± 4.65	3.28 ± 4.98	5.70 ± 4.13	2.36 ± 5.37
3TC	5.68 ± 3.78	4.94 ± 4.99	–0.44 ± 3.85	–9.22 ± 6.32

For example, activity of 1f was more remarkable than that of 1d and 1e, which suggested that 3-carbonyl may promote the anti-HBV activity of cholesterol derivatives with a peroxide bridge. However, compound 2d exhibited stronger activity than that of 2e and 2f, which demonstrated that 3-acetyl may play a key role in β-sitosterol derivatives. It thus appeared that diverse sterol matrices had different working groups. This phenomenon could be related to their biogenetic derivation.

Further study on antigens inhibition of compound 1f was performed by setting four concentration gradients. As shown in Fig. 3, dose- and time-effect relationship was found. After 8 days, 1f showed significant inhibitory effects on HBsAg secretion, with the highest inhibition at 77.45% ± 6.01%. Similarly, the secretion of HBeAg was significantly reduced with the highest inhibition at 58.73% ± 8.64%.

Fig. 3. The inhibition ratio of compound 1f on HBsAg and HBeAg in HepG2.2.15 cells. HepG2.2.15 cells were treated with four concentrations (0.39, 0.78, 1.56, and 3.13 μg ml^–1) of compound 1f for 4 d or 8 d. The cultural media of each treatment were collected for viral HBsAg and HBeAg via ELISA analysis. The data are expressed as the mean and the standard deviation of the mean (n = 3, p < 0.05).

Moreover, we calculated the therapeutic index of 1f to indicate its anti-hepatitis B activity (Table 2).

Table 2. Inhibitory effect on HBsAg and HBeAg in 2.2.15 cells treated with 1f.

Time	Antigen	IC50	TC50	TC50/IC50
4 d	HBeAg	1.78 ± 0.24		5.79 ± 0.52
	HBsAg	1.86 ± 0.09		5.51 ± 0.18
8 d	HBeAg	2.59 ± 0.15	10.25	3.40 ± 0.24
	HBsAg	2.95 ± 0.35		3.47 ± 0.37

Acute toxicity of compound 1fin vivo

Although compound 1f exhibited relatively low cytotoxicity in vitro, the toxicity in vivo was not clear. To evaluate the safety of compound 1f, the acute toxicity was tested in ICR mice, with doses of 56, 178, 350, 422, 750 and 1000 mg kg^–1 administered intraperitoneally (i.p.). Every dose was administered to 6 mice. Compound 1f revealed a regular dose-dependent increase in mortality (Table 3).

Table 3. Acute toxicity test of compound 1f.

Dose (mg kg–1)	Mice number start/end	Death rate (%)	LD50 (mg kg–1)	95% CIs
1000	6/6	100	362.46	233.83–561.85
750	6/6	100
422	6/4	67
350	6/3	50
178	6/0	0
56	6/0	0

The LD₅₀ value (362.46 mg kg^–1) can be worked out through the regression equation: y (Probit) = –17.114 + 8.6408 log(D), which showed the safe dose range of 1f.

(Statement: all experiments were performed in compliance with the relevant laws and institutional guidelines, and the institutional committee(s) that have approved the experiments).

Conclusions

A new method to synthesize a peroxo bridge was established using eosin Y as a photosensitive catalytic oxidizer. Compared with the conventional synthetic methods, the new method was in lower consumption and more convenient. Therefore, our synthesis of peroxide bridges was optimized by the special catalyzer. A series of novel steroid derivatives containing peroxide bridge were synthesized using this method, and their bioactivity tests showed that most derivatives had anti-HBV activities. In particular, compound 1f exhibited high inhibitory capacity against HBsAg and HBeAg secretion in HepG2.2.15 cells, whereas nearly no cytotoxicity on HepG2.2.15 cells at 3.13 μg mL^–1 was observed, showing good therapeutic index.

The structures were significantly different between steroid derivatives and 3TC. Although lamivudine was no longer recommended as first-line therapy for chronic hepatitis B, using lamivudine judiciously on the clinical and scientific research was paramount.22 Unlike the metabolism antagonism of 3TC, steroids might indicate yet another mechanism that is related with the capacity of steroid hormones to intercalate into cell membrane bilayer and alter membrane fluidity.23 This property of the steroids would affect the insertion of viral glycoproteins further inhibiting the assembly of virus particles,4 which requires further study. Therefore, these compounds were expected in collaboration with 3TC for the treatment of hepatitis B diseases. In summary, the steroid derivatives with peroxide bridges demonstrated potential anti-HBV activity, and it provided us new ideas to discover anti-HBV leading compounds.