O-Aminoalkyl-O-Trimethyl-2,3-Dehydrosilybins: Synthesis and In Vitro Effects Towards Prostate Cancer Cells

As part of our ongoing silybin project, this study aims to introduce a basic nitrogen-containing group to 7-OH of 3,5,20-O-trimethyl-2,3-dehydrosilybin or 3-OH of 5,7,20-O-trimethyl-2,3-dehydrosilybin via an appropriate linker for in vitro evaluation as potential anti-prostate cancer agents. The synthetic approaches to 7-O-substituted-3,5,20-O-trimethyl-2,3-dehydrosilybins through a five-step procedure and to 3-O-substituted-5,7,20-O-trimethyl-2,3- dehydrosilybins via a four-step transformation have been developed. Thirty-two nitrogen-containing derivatives of silybin have been achieved through these synthetic methods for the evaluation of their antiproliferative activities towards both androgen-sensitive (LNCaP) and androgen-insensitive prostate cancer cell lines (PC-3 and DU145) using the WST-1 cell proliferation assay. These derivatives exhibited greater in vitro antiproliferative potency than silibinin. Among them, 11, 29, 31, 37, and 40 were identified as five optimal derivatives with IC50 values in the range of 1.40–3.06 µM, representing a 17- to 52-fold improvement in potency compared to silibinin. All these five optimal derivatives can arrest the PC-3 cell cycle in the G0/G1 phase and promote PC-3 cell apoptosis. Derivatives 11, 37, and 40 are more effective than 29 and 31 in activating PC-3 cell apoptosis.

2 of 20 g/day [15]. However, the development of silybin as anti-prostate cancer drug is hindered, at least partly, by its moderate potency with its IC50 values of 40-106 µM in prostate cancer cell models [11,13,16,17]. Structural modification of silybin can serve as a viable strategy to enhance its potency. Methylated silybins have been reported to be capable of increasing antiproliferative activities towards prostate cancer cells [18]. Additionally, 2,3-dehydrosilybin has been shown to be a significantly better anticancer agent than silybin [19].
Our previous studies on structure-activity relationships of silybin revealed that the antiproliferative potency of 2,3-dehydrosilybin in three prostate cancer cell models could be further improved through introducing a suitable alkyl group on 7-OH and 3-OH, as exemplified by 7-O-ethyl-2,3-dehydrosilybin (3) and 3-O-propyl-2,3-dehydrosilybin (4) (Figure 1) [20,21]. This encouraged us to further investigate the effects of nitrogen-containing groups on 7-OH and 3-OH of 2,3-dehydrosilybin on the biological profiles in prostate cancer cell models. Consequently, this study started with the development of general synthetic approaches to 7-O-substituted-3,5,20-O-trimethyl-2,3-dehydrosilybins and 3-O-substituted-5,7,20-O-trimethyl-2,3dehydrosilybins followed by the synthesis of thirty-two new derivatives of silybin, including six 7-O-aminoalkyl-3,5,20-O-trimethyl-2,3-dehydrosilybins and twenty-six 3-O-aminoalkyl-5,7,20-Otrimethyl-2,3-dehydrosilybins. However, it is worth noting that 2,3-dehydrosilybin A is more potent than 2,3-dehydrosilybin B in cell and whole organism models to prolong lifespan and block A aggregation [22]. It is thus imperative to confirm the respective in vitro antiproliferative potency of each enantiomer for the optimal O-aminoalkyl-O-trimethyl-2,3-dehydrosilybins before moving further for in vivo animal studies and various mechanism investigations. Additionally, the phenolic hydroxyl groups in all synthetic derivatives were converted to methoxyl groups. This conversion was expected to overcome, to some degree, pharmacokinetic limitations caused by the phenolic hydroxyl groups and to pave an avenue to selective incorporation of a basic nitrogen-containing group to the phenolic hydroxyl group either at C-7 or at C-3. The in vitro anticancer activities of these derivatives have been evaluated in three prostate cancer cell models. The design, synthesis, antiproliferative activity, and structure-activity relationships of these silybin derivatives were presented in this paper. The cell apoptosis induction and cell cycle regulation by five representative derivatives were also reported.  As illustrated in Scheme 1, synthesis of 7-O-substituted-2,3-dehydrosilybins (9)(10)(11)(12)(13)(14) started with selective benzylation (81%) of C-7 phenolic hydroxyl group of silybin according to the procedure reported by Kren et al. and us [20,23]. It is worth noting that anaerobic conditions are essential to achieving high yields for the selective benzylation. This is because the simultaneous presence of base and air led to the aerobic oxidation of silybin to 2,3-dehydrosilybin [7,21] and the 3-OH in 2,3-dehydrosilybin is readily benzylated or alkylated [7,21,23] that has been rationalized by the electrochemistry measurements and bond dissociation energy calculations [24]. 7-O-Benzyl-3,5,20-O-trimethyl-2,3-dehydrosilybin (6) was then achieved by the one-pot reaction of base-mediated oxidation of 7-O-benzylsilybin (5) followed by trimethylation of the corresponding 7-O-benzyl-2,3-dehydrosilybin. 3,5,20-O-Trimethyl-2,3-dehydrosiliybin (7) was obtained by debenzylation of aryl benzyl ether 6 using ammonium formate as the hydrogen source catalyzed by palladium on carbon. 7-O-bromopropyl-3,5,20-O-trimethyl-2,3-dehydrosilybin (8) was prepared by O-alkylating 7 with 1,3-dibromopropane mediated by potassium carbonate. 7-O-Aminoalkyl-3,5,20-O-trimethyl-2,3-dehydrosilybins (9)(10)(11)(12)(13)(14) were achieved by N-alkylation of the bromoalkyl side chain of 8 with the appropriate amine. 3-O-Substituted-2,3-dehydrosilybins (20-45) were synthesized following a four-step procedure as shown in Scheme 2. Specifically, 5,7,20-O-trimethylsilybin (15) was achieved by treating silybin (1) with dimethylsulfate in the presence of potassium carbonate under strictly anaerobic conditions. Note that small amount of 3,5,7,20-O-tetramethyl-2,3-dehydrosilybin can be formed if anaerobic conditions were not well-controlled, which would complicate the purification process and decrease the yield. Even though 5,7,20-O-tribenzylsilybin was much easier to be aerobically oxidized than that in silybin [25], oxidation of 5,7,20-O-trimethylsilybin (15) under the same conditions led to a mixture of products instead of the desired oxidation product. After several trials with different oxidation conditions, 5,7,20-O-trimethyl-2,3-dehydrosilybin (16) was eventually obtained by oxidation of 15 with sodium hydroxide and hydrogen peroxide. 10-14 Hours reaction time serves as a critical factor for the optimal yield (40-55%) of this oxidation reaction. We also found this oxidation cannot be quenched with hydrochloric acid because it selectively demethylated 5-OMe of the product. 5   The in vitro antiproliferative activities of six 7-O-substituted and twenty-six 3-O-substituted silybin derivatives were evaluated using WST-1 cell proliferation assay according to the procedure as described in the Experimental Section in both androgen-sensitive (LNCaP) and androgen-insensitive (PC-3 and DU145) human prostate cancer cell lines. Silybin was used as a positive control for comparison in the parallel experiments and the IC50 values calculated from the dose-response curves were summarized in Table  1. Clearly, 7-O-aminoalkyl-3,5,20-O-trimethyl-2,3-dehydrosilybins and 3-O-aminoalkyl-5,7,20-O-trimethyl-2,3-dehydrosilybins are more ogen-insensitive prostate cancer cell proliferation than silybin. This conclusion is supported by the following data: i) the optimal 7-O-substituted derivative (11) is 52-, 51-, and 24-fold more potent than silybin toward PC-3, DU145, and LNCaP prostate cancer cell lines; and ii) the optimal 3-O-substituted derivatives (29, 31, 37, and 40) are 26-27, 31-37, and 17-22 times more potent than silybin. Additionally, the dibutylamino group in derivatives 11, 29 and 31, the morpholino moiety in 37, and the piperidino unit in 40 are the favorable nitrogen-containing groups for the greater potency. A three-carbon linker in 11 and 29, and a five-carbon linker in 31, 37, and 40 are beneficial to the potency. [a] IC50 value is the compound concentration effective in inhibiting 50% of the cell viability measured by WST-1 cell proliferation assay after 3 days exposure. The data were presented as the mean ± SD from n = 3.
[b] Human androgen-insensitive prostate cancer cell line derived from bone metastasis of prostate tumor.
[c] Human androgen-insensitive prostate cancer cell line derived from brain metastasis of prostate tumor.
[d] Human androgen-sensitive prostate cancer cell line derived from lymph node metastasis of prostate tumor.

Cell Cycle Regulation and Cell Apoptosis.
Silybin can arrest rat (H-7 and I-8) and human prostate cancer cell (LNCaP) cycle at G1 phase [26,27], and cause G1 and G2-M PC-3 prostate cancer cell cycle arrest [28]. Five optimal derivatives, 11, 29, 31, 37, and 40, were selected for flow cytometry evaluation of their effect on PC-3 cell cycle regulation because they exhibited optimal cell proliferation inhibition on both androgen-dependent LNCaP and androgen-independent PC-3 prostate cancer cell models with ≤ 3.0 µM IC50 values. At 20 µM, all these five derivatives can cause PC-3 cell accumulation in a G0/G1 phase by increasing the cell population in this phase at 16 hours from 55.7% (control) to 66.3% (treated with 11), from 36.2% (control) to 50.3% (treated with 29), from 33.1% (control) to 35.9% (treated with 31), from 33.1% (control) to 34.6% (treated with 37), and from 33.1% (control) to 43.4% (treated with 40). Silybin was revealed by Agarwal and co-workers to activate cell apoptosis in PC-3 tumor xenografts [29]. F2N12S and SYTOX AADvanced double staining flow cytometry-based assay was used to discriminate PC-3 cells dying from apoptosis from those dying from necrosis in response to various concentrations of derivatives 11, 29, 31, 37, and 40. PC-3 cells were incubated with the test compound for 16 h. As shown in Figure 2, derivatives 11, 37, and 40 induced appreciable levels of apoptotic cell death in the androgen-insensitive PC-3 prostate cancer cell line in a dose-responsive manner after a 16-hour treatment. Specifically, 5 µM of derivatives 11, 37, and 40 could induce substantial early phase of apoptosis (26-59%) in PC-3 cells as compared with control cells; treatment with 10 µM of these three optimal derivatives led to 56-76% early apoptotic cells and 6-40% late apoptotic/necrotic cells; 20 µM of derivatives 11, 37, and 40 activated notable apoptosis as well, with 54-75% early apoptotic cells and 16-44% late apoptotic/necrotic cells. The apoptotic cell population reached maximum when PC-3 cancer cells were exposed to derivative 11, 37, and 40 at 5µM, 10 µM, and 30 µM, respectively. In contrast, derivatives 29 and 31 did not induce significant levels of apoptotic cell death (less than 10%) up to 10 µM concentration. Only 50 µM of derivatives 29 and 31 results in the maximum apoptotic cell population (71% and 95%, respectively).

General Procedures.
HRMS were obtained on an Orbitrap mass spectrometer with electrospray ionization (ESI). NMR spectra were obtained on a Bruker Fourier 300 spectrometer in CDCl3, or DMSO-d6. The chemical shifts are given in ppm referenced to the respective solvent peak, and coupling constants are reported in Hz. Anhydrous THF and dichloromethane were purified by PureSolv MD 7 Solvent Purification System from Innovative Technologies (MB-SPS-800). All other reagents and solvents 4 of 20 were purchased from commercial sources and were used without further purification. Silica gel column chromatography was performed using silica gel (32-63 µm). Preparative thin-layer chromatography (PTLC) separations were carried out on thin layer chromatography plates loaded with silica gel 60 GF254 (EMD Millipore Corporation, MA, USA). Silybin (> 98.0%) was purchased from Fisher Scientific (TCI America, Cat # 50-014-46874).

Synthesis of 7-O-benzyl-3,5,20-O-trimethyl-2,3-dehydrosilybin (6)
Potassium carbonate (3 eq.) was added to a solution of benzylsilybin (1 eq.) in DMF (0.5 M) and the reaction mixture was opened to air with stirring at room temperature (or 60 o C) for 3 hours. When most of 7-O-benzylsilybin was oxidized to 7-O-benzyl-2,3-dehydrobenzylsilybin as monitored by TLC, the reaction mixture was cooled down to room temperature. Potassium carbonate (3 eq.) followed by methyl iodide (6 eq.) were added to the reaction mixture and the reaction was allowed to proceed at room temperature overnight prior to being quenched with HCl (1 M). The subsequent mixture was diluted with water and extracted with ethyl acetate. The combined organic extracts were rinsed with brine and dried over anhydrous sodium sulfate. After filtration, the volatile components were evaporated under vacuum to give the crude product, which was purified by column chromatography over silica gel or PTLC eluting with 5% methanol in dichloromethane to

Synthesis of 5,7,20-O-trimethylsilybin (15).
A 3-neck round bottom flask was charged with silybin (2.01 g, 4.2 mmol) and potassium carbonate (3.43 gram, 25.1 mmol), which was vacuumed three times under argon prior to the addition of acetone (30.0 mL). The reaction mixture was refluxed for 15 minutes before dimethylsulfate (3.13 mL, 33.1 mmol) was added through a needle. The reaction was continued with refluxing for an additional 4 hours when the reaction was completed as monitored by TLC. After cooling down to room temperature, saturated ammonium chloride was added to quench the reaction, and the subsequent mixture was extracted with ethyl acetate for 3 times. The organic layers were combined, washed with brine twice, and dried over anhydrous sodium sulfate. Purification of the crude product through column chromatography, eluting with ethyl acetate/hexane (50/50 to 70/30, v/v), gave the product (15) as a white crystal in 80% yield. 1

Synthesis of 5,7,20-O-trimethyl-2,3-dehydrosilybin (16)
A 10-mL round flask was charged with 5,7,20-O-trimethylsiybin (150.0 mg, 0.23 mmol) in methanol (2.0 mL) and tetrahydrofuran (2.0 mL). The solution was stirred for 10 minutes at room temperature prior to being added hydrogen peroxide (0.85 mL, 30%) and sodium hydroxide aqueous solution (0.65 mL, 16%) at 0 o C. The reaction mixture was slowly warmed to room temperature and then stirred overnight before being quenched with saturated ammonium chloride. The subsequent mixture was extracted with dichloromethane for three times, and the combined extracts were dried over sodium sulfate and concentrated under vacuum. The crude product was obtained in 49% yield, which is pure enough for the next step reaction without purification. 1   A round bottom flask (10 mL) was charged with 5,7,20-O-trimethyl-2,3-dehydrosilybin (16, 83.2 mg, 0.16 mmol), potassium carbonate (352.0 mg, 2.55 mmol), and DMF (5.0 mL). The mixture was stirred for 10 minutes prior to being added of 1,3-, 1,4-, or 1,5-dibromalkane (2.56 mmol, 16 equiv.). The reaction was continued with stirring at room temperature for 24-48 hours before the reaction was quenched with water. The subsequent mixture was extracted with ethyl acetate for three times, and the combined extracts were dried over anhydrous sodium sulfate and concentrated in vacuo. The crude products was subjected to PTLC purification eluting with DCM/methanol (100/5, v/v) to yield the respective 3-O-bromoalkyl-5,7,10-O-trimethyl-2,3-dehydrosilybin. A round bottom reaction flask (10 mL) was charged with 3-O-bromoalkyl-5,7,10-O-trimethyl-2,3-dehydrosilybin (1 eq.) and potassium carbonate (10 eq.) in acetone (2.0 mL, 0.029 M). The solution was stirred for 10 minutes prior to being added the appropriate amine (16 eq.). The reaction was allowed to proceed with stirring at room temperature for 24-48 hours before being quenched with water. The subsequent mixture was extracted with ethyl acetate for three times, and the combined extracts were dried over anhydrous sodium sulfate and concentrated in vacuo. The crude product was subjected to PTLC purification eluting with DCM/methanol (100:5, v/v). Each desired nitrogen-containing compound was retrieved from PTLC silica gel by washing with dichloromethane/methanol/ammonium hydroxide (100:10:5, v/v/v).  3.14. F2N12S and SYTOX AADvanced double staining assay PC-3 cells were plated in 24-well plates at a density of 200,000 each well in 400 µL of culture medium. After 3 hours of cell attachment, the cells were then treated with each test compound at different concentration for 15 hours, while equal treatment volumes of DMSO were used as vehicle control. The cells were cultured in CO2 incubator at 37°C for 15 hours. Both attached and floating cells were collected in a centrifuge tube by centrifugation at rcf value of 450 g for 5 to 6 minutes. The collected cells were re-suspended with 500 µL HBSS to remove proteins which may affect flow signal and centrifuged again. After discarding the supernatant, the collected cells were re-suspended with 300 µL HBSS and stained with 0.3 µL of F2N12S for 3-5 minutes followed by 0.3 µL SytoxAAdvanced for an additional 5 minutes. The fluorescence intensity of the two probes was further measured in individual PC-3 cells using an Attune flow cytometer (Life Technologies) 0.5 to 1 hour after staining.

Statistical analysis:
All data are represented as the mean ± standard deviation (S.D.) for the number of experiments indicated. Other differences between treated and control groups were analyzed using the Student's t-test. A p-value < 0.05 was considered statistically significant.

Conclusions
In summary, six 7-O-aminoalkyl-3,5-20-O-trimethyl-2,3-dehydrosilybins and twenty-six 3-O-aminoalkyl-5,7,20-O-trimethyl-2,3-dehydrosilybins have been successfully synthesized through a five-step and four-step sequence, respectively. The synthetic methods can be used for the general synthesis of 7-O-substituted-3,5,20-O-trimethyl-2,3-dehydrosilybins and 3-O-substituted-5,7,20-Otrimethyl-2,3-dehydrosilybins. The antiproliferative activities of the thirty-two derivatives against three prostate cancer cell lines have been evaluated using WST-1 cell proliferation assay. All of them showed better prostate cancer cell proliferation inhibition than silybin. Derivatives 11, 29, 31, 37, and 40 were identified as the optimal derivatives with IC50 values in the range of 1.40-3.06 µM toward these three prostate cancer cell lines, a 17-to 52-fold improvement in potency as compared with silybin. All these five optimal derivatives can cause PC-3 cell accumulation in a G0/G1 phase by increasing the cell population in this phase at 16 hours. Derivatives 11, 37, and 40 show stronger ability than derivatives 29 and 31 in activating PC-3 cell apoptosis by inducing appreciable levels of apoptotic cell death at 5 µM concentration after a 16-hour treatment. In contrast, derivatives 29 and 31 did not induce significant levels of apoptotic cell death (less than 10%) up to 10 µM concentration.
Supplementary Materials: NMR-spectra ( 1 H and 13 C) of the new silybin derivatives.