Synthesis of Cucurbitacin B Derivatives as Potential Anti-Hepatocellular Carcinoma Agents

Cucurbitacin B shows potent activity against tumor cells, but its high toxicity limits its application in the clinic. A series of cucurbitacin B derivatives was synthesized and evaluated for their anti-hepatocellular carcinoma (HCC) activities against the HepG-2 cell line. These compounds were also tested for their toxicity against the L-O2 normal cell line. The compound with the most potential, 10b, exhibited potent activity against the HepG-2 cell line with an IC50 value of 0.63 μM. Moreover, compound 10b showed the highest TI value (4.71), which is a 14.7-fold improvement compared to its parent compound cucurbitacin B. A preliminary molecular mechanism study of 10b indicated that 10b could inhibit P-STAT3 to induce the activation of mitochondrial apoptotic pathways. An in vivo acute toxicity study indicated that the compound 10b has preferable safety and tolerability compared with cucurbitacin B. These findings indicate that compound 10b might be considered as a lead compound for exploring effective anti-HCC drugs.


Introduction
Cancer is a disease with high morbidity and mortality. It is the second leading cause of death, after cardiovascular disease, worldwide [1,2]. The World Health Organization has reported that there were approximately 17.5 million new cancer cases and more than 8.7 million cancer-related deaths worldwide in 2015, and cancer-related deaths are expected to increase to 13.1 million by 2030 [3][4][5]. Cancer is a type of abnormal and excessive growth of tissue, in which abnormal body cells begin to divide uncontrollably by ignoring the principles of normal cell division. Cancer cells invade or spread to other healthy parts of the body, resulting in the formation of new tumor; this process is called metastasis. Approximately 90% of cancer-linked deaths are caused by it [5][6][7]. So far, cancer treatments mostly rely on traditional chemotherapy, especially for late-stage or complex cancers. Unfortunately, side effects and acquired drug resistance limit the effectiveness of traditional chemotherapy. Side effects, such as extreme fatigue and leukopenia, are caused by traditional chemotherapy's low selectivity towards cancer cells and normal cells. These unpleasant side effects seriously reduce the life quality of cancer patients [8][9][10]. Therefore, the development of new chemotherapeutic drugs with high efficiency and low-toxicity side effects is a great challenge for medicinal chemists. Natural compounds can induce apoptosis by targeting multiple cellular signaling pathways; so, the development of new pharmacologically effective chemotherapeutic agents from natural compounds may be feasible [11][12][13].  Cucurbitacin B (1) is one of the most abundant forms of cucurbitacins, and one of the best-studied compounds of the cucurbitacins family [26][27][28][29][30]. Cucurbitacin B possesses a variety of bioactivities, such as anti-inflammatory, hepatoprotective, and anticancer activities. It has demonstrated potent anticancer activities in solid tumors in vitro and in vivo [31,32]. Cucurbitacin B is capable of inhibiting the growth of a wide spectrum of malignant human cells, such as laryngeal squamous cell carcinoma, breast cancer, pancreatic cancer, hepatocellular carcinoma (HCC), lung cancer, and melanoma cells [28,[33][34][35][36]. The mechanism underlying the anticancer effect of cucurbitacin B is not clear; however, several studies have shown that cucurbitacin B inhibits the proliferation of, and induces apoptosis in, human cancer cells via STAT3 pathway inhibition and increased intracellular reactive oxygen species (ROS) [29,[36][37][38][39][40].
At present, the main problem with cucurbitacin B is its high toxicity, which renders it unlikely to be a drug-able agent. Its low selectivity and narrow treatment window severely limit its clinical application. With cucurbitacin B as the starting substrate, we designed and synthesized a series of cucurbitacin B derivatives for the exploration of structure-activity relationships and the discovery of potential anti-HCC agents with potent anti-HCC activity and low toxicity.

Chemistry
The starting material cucurbitacin B was isolated from Cucumis melo L. The presence of free hydroxyl groups allowed us to prepare ester derivatives of cucurbitacin B for an evaluation of the influence of the ester side chain on their anti-HCC activities. It was reported that the introduction of an acyl group to a 2-hydroxyl moiety greatly reduced the anticancer activity [41]. We planned to explore the effect of selective acylation on a 16-hydroxyl group on the anti-HCC activity. However, a 2-hydroxyl group is more reactive than a 16-hydroxyl group for acylation [41]. Therefore, we first designed protection for the 2-hydroxyl group. As shown in Scheme 1, the treatment of compound 1 with t-butyldimethylsilyl chloride (TBSCl) in the presence of imidazole selectively provided the 2-TBS-protected derivative 2 without the influence of the hydroxyl group linked to C20 of the cucurbicin B. The hydroxyl group linked to C20 is a tertiary hydroxyl group that shows a higher steric hindrance effect than the C16 hydroxyl group. Therefore, the C16 hydroxyl group was selectively esterified without affecting the hydroxyl group linked to C20 of the cucurbicin B. Compound 2 was coupled with different patterns of carboxylic acids in the presence of 1-ethyl-3-(3-dimethyllaminopropyl)carbodiimide hydrochloride (EDCI) and 4-dimethylaminopyridine (DMAP), followed by removal of the TBS protection to afford the esters 3a-3q in yields of 48-86%. Direct conjuation of compound 1 with cinnamic acid provided the derivative 4 in a high yield (90%).
influence of the ester side chain on their anti-HCC activities. It was reported that the introduction of an acyl group to a 2-hydroxyl moiety greatly reduced the anticancer activity [41]. We planned to explore the effect of selective acylation on a 16-hydroxyl group on the anti-HCC activity. However, a 2-hydroxyl group is more reactive than a 16-hydroxyl group for acylation [41]. Therefore, we first designed protection for the 2-hydroxyl group. As shown in Scheme 1, the treatment of compound 1 with t-butyldimethylsilyl chloride (TBSCl) in the presence of imidazole selectively provided the 2-TBS-protected derivative 2 without the influence of the hydroxyl group linked to C20 of the cucurbicin B. The hydroxyl group linked to C20 is a tertiary hydroxyl group that shows a higher steric hindrance effect than the C16 hydroxyl group. Therefore, the C16 hydroxyl group was selectively esterified without affecting the hydroxyl group linked to C20 of the cucurbicin B. Compound 2 was coupled with different patterns of carboxylic acids in the presence of 1-ethyl-3-(3dimethyllaminopropyl)carbodiimide hydrochloride (EDCI) and 4-dimethylaminopyridine (DMAP), followed by removal of the TBS protection to afford the esters 3a-3q in yields of 48-86%. Direct conjuation of compound 1 with cinnamic acid provided the derivative 4 in a high yield (90%).
It was reported that the introduction of a phenylsulfonyl-substituted furoxan NO-releasing moiety could reduce the toxicity of brusatol [42][43][44][45][46]. Accordingly, 18 cucurbitacin B derivatives, 9a-9c and 10a-10o, with the introduction of a phenylsulfonyl-substituted furoxan NO-releasing moiety were designed (Scheme 2). Compound 5 was prepared according to a reported procedure [42]. Reaction of 5 with different diols or alchoholamines gave 6a-6o or 7a-7c, respectively [42][43][44]46]. Compounds 6a-6o were treated with succinic anhydride in the presence of DMAP to produce the corresponding acids, followed by conjuation with compound 2 and removal of the TBS group to yield 10a-10o in yields of 61-82% in two steps. The acid 8 was obtained by the reaction of 2 and succinic anhydride. The removal of Boc groups of 7a-7c gave the corresponding amines, followed by coupling with the acid 8 in the presence of 2-(7-azabenzotriazol-1-yl)-N,N,N',N'-tetramethyluronium hexafluorophosphate (HATU) and N,N-diisopropylethylamine (DIPEA) and deprotection of the TBS group to afford 9a-9c in yields of 44-73% in three steps. The removal of the TBS group of the acid 8 yielded compound 11 in a yield of 79%.
It was reported that the introduction of a phenylsulfonyl-substituted furoxan NO-releasing moiety could reduce the toxicity of brusatol [42][43][44][45][46]. Accordingly, 18 cucurbitacin B derivatives, 9a-9c and 10a-10o, with the introduction of a phenylsulfonyl-substituted furoxan NO-releasing moiety were designed (Scheme 2). Compound 5 was prepared according to a reported procedure [42]. Reaction of 5 with different diols or alchoholamines gave 6a-6o or 7a-7c, respectively [42][43][44]46]. Compounds 6a-6o were treated with succinic anhydride in the presence of DMAP to produce the corresponding acids, followed by conjuation with compound 2 and removal of the TBS group to yield 10a-10o in yields of 61-82% in two steps. The acid 8 was obtained by the reaction of 2 and succinic anhydride. The removal of Boc groups of 7a-7c gave the corresponding amines, followed by coupling with the acid 8 in the presence of 2-(7-azabenzotriazol-1-yl)-N,N,N ,N -tetramethyluronium hexafluorophosphate (HATU) and N,N-diisopropylethylamine (DIPEA) and deprotection of the TBS group to afford 9a-9c in yields

In Vitro Biological Activities against the HepG-2 Cell Line and the L-O2 Cell Line
The synthesized cucurbitacin B derivatives 3a-3q, 4, 9a-9c, 10a-10o, and 11 were evaluated for their activity against proliferation of the human hepatocellular carcinoma cell line HepG-2 and the normal hepatocellular cell line L-O2. Cucurbitacin B was also included for comparison. The therapeutic index (TI), a major pharmaceutical parameter that estimates possible future clinical development, was determined as the ratio of IC50 for L-O2 to IC50 for HepG-2. The bioactivity of each compound was evaluated by the combination of its IC50 and TI. These results are shown in Table 1 and Table 2.
The parent compound 1 exhibited potent activity against HepG-2 (IC50 = 0.06 μM), but appeared to have high toxicity (IC50 = 0.019 μM), which led to a very low therapeutic index (0.32). It was reported that the selective introduction of an acyl group to a 2-hydroxyl moiety greatly reduced the anticancer activity [41]. To further explore the influence of introducing different acyl moieties to a 16hydroxyl group on anti-HCC activity, we synthesized the 16-mono-acylated derivatives 3a-3q. For subseries of 3a-3q, all compounds showed decreased cytotoxicity against the normal hepatocellular cell line L-O2 (IC50 = 0.0422-9.59 μM) compared to the parent compound 1 (IC50 = 0.019 μM), which indicated that the introduction of an acyl moiety to a 16-hydroxyl group could reduce the toxicity to normal cells. However, their anti-HCC activities against HepG-2 cells were also greatly decreased, with IC50 values ranging from 0.22 μM to 9.62 μM. Among subseries 3a-3q, compound 3o with a 16mono-cinnamyl substitution showed the highest TI value (TI = 2.9). Therefore, we further synthesized compound 4 with a 2,16-di-cinnamyl substitution for testing its anti-HCC activity. Unfortunately, compound 4 lost inhibitory activity against HepG-2 (IC50 > 20 μM).

In Vitro Biological Activities against the HepG-2 Cell Line and the L-O2 Cell Line
The synthesized cucurbitacin B derivatives 3a-3q, 4, 9a-9c, 10a-10o, and 11 were evaluated for their activity against proliferation of the human hepatocellular carcinoma cell line HepG-2 and the normal hepatocellular cell line L-O2. Cucurbitacin B was also included for comparison. The therapeutic index (TI), a major pharmaceutical parameter that estimates possible future clinical development, was determined as the ratio of IC 50 for L-O2 to IC 50 for HepG-2. The bioactivity of each compound was evaluated by the combination of its IC 50 and TI. These results are shown in Tables 1 and 2.
The parent compound 1 exhibited potent activity against HepG-2 (IC 50 = 0.06 µM), but appeared to have high toxicity (IC 50 = 0.019 µM), which led to a very low therapeutic index (0.32). It was reported that the selective introduction of an acyl group to a 2-hydroxyl moiety greatly reduced the anticancer activity [41]. To further explore the influence of introducing different acyl moieties to a 16-hydroxyl group on anti-HCC activity, we synthesized the 16-mono-acylated derivatives 3a-3q. For subseries of 3a-3q, all compounds showed decreased cytotoxicity against the normal hepatocellular cell line L-O2 (IC 50 = 0.0422-9.59 µM) compared to the parent compound 1 (IC 50 = 0.019 µM), which indicated that the introduction of an acyl moiety to a 16-hydroxyl group could reduce the toxicity to normal cells. However, their anti-HCC activities against HepG-2 cells were also greatly decreased, with IC 50 values ranging from 0.22 µM to 9.62 µM. Among subseries 3a-3q, compound 3o with a 16-mono-cinnamyl substitution showed the highest TI value (TI = 2.9). Therefore, we further synthesized compound 4 with a 2,16-di-cinnamyl substitution for testing its anti-HCC activity. Unfortunately, compound 4 lost inhibitory activity against HepG-2 (IC 50 > 20 µM).  Considering that the introduction of a phenylsulfonyl-substituted furoxan NO-releasing moiety could reduce the toxicity of brusatol [42][43][44][45], derivatives 9a-9c were synthesized and evaluated for their anti-HCC activity. As shown in Table 2, compounds 9a-9c were less toxic than the parent compound 1 and showed improved TI values (ranging from 1.5 to 2.84) compared to compound 1 (TI = 0.32). These results prompted us to further synthesize more derivatives with a phenylsulfonyl-substituted furoxan NO-releasing moiety, i.e., compounds 10a-10o. They featured with a different linker between cucurbitacin B and the phenylsulfonyl-substituted furoxan NO-releasing moiety. Compounds 10a-10o exhibited potent or moderate activity against HepG-2 cells. Importantly, their cytotoxicity against L-O2 was significantly reduced compared to their parent compound cucurbitacin B, with IC 50 values ranging from 1.17 to 3.31 µM. The compound with the most potential, 10b, showed the highest TI value (4.71), which is a 14.7-fold improvement compared to its parent compound cucurbitacin B. Moreover, it still maintained potent activity against HepG-2 cells, with an IC 50 value of 0.63 µM. These results suggested that the introduction of a phenylsulfonyl-substituted furoxan NO-releasing moiety could reduce the toxicity against normal cells and maintain potent anti-HCC activity.
The compound with the most potential, 10b, was selected for a preliminary study on its mechanism of action.

Preliminary Study of Compound 10b's Mechanism of Action
To determine whether the cytotoxic effect of compound 10b was associated with apoptosis, a cell apoptosis assay was performed by flow cytometry. As shown in Figure 2A, compound 10b significantly induced cell apoptosis in a dose-dependent manner. To further analyze the mechanism by which compound 10b induces apoptosis, a Western blot assay was performed. STAT3 is associated with the expression of its downstream mitochondrial proteins, such as Bcl-2, Bax, and Bim, which are are critical regulators of cell apoptosis. As a result, we investigated the protein level of STAT3 and P-STAT3 using a Western blot analysis. As shown in Figure 2B, the total expression of STAT3 remained little changed. STAT3 is usually activated by phosphorylation to perform its functions. Inhibition of P-STAT3 plays an important role in cancer therapy. Our data showed that the level of P-STAT3 was clearly inhibited after treatment with compound 10b in a dose-dependent manner. Moreover, the expression of the apoptotic-related members Bax and Bim were strongly increased, while the expression of the anti-apoptotic-related member Bcl-2 was substantially diminished after treatment with compound 10b. As shown in Figure 1B, the level of caspase 3 was decreased. Accordingly, we proposed that compound 10b inhibited P-STAT3 to induce mitochondrial-pathway-mediated apoptosis.

Preliminary Study of Compound 10b's Mechanism of Action
To determine whether the cytotoxic effect of compound 10b was associated with apoptosis, a cell apoptosis assay was performed by flow cytometry. As shown in Figure 2A, compound 10b significantly induced cell apoptosis in a dose-dependent manner. To further analyze the mechanism by which compound 10b induces apoptosis, a Western blot assay was performed. STAT3 is associated with the expression of its downstream mitochondrial proteins, such as Bcl-2, Bax, and Bim, which are are critical regulators of cell apoptosis. As a result, we investigated the protein level of STAT3 and P-STAT3 using a Western blot analysis. As shown in Figure 2B, the total expression of STAT3 remained little changed. STAT3 is usually activated by phosphorylation to perform its functions. Inhibition of P-STAT3 plays an important role in cancer therapy. Our data showed that the level of P-STAT3 was clearly inhibited after treatment with compound 10b in a dose-dependent manner. Moreover, the expression of the apoptotic-related members Bax and Bim were strongly increased, while the expression of the anti-apoptotic-related member Bcl-2 was substantially diminished after treatment with compound 10b. As shown in Figure 1B, the level of caspase 3 was decreased. Accordingly, we proposed that compound 10b inhibited P-STAT3 to induce mitochondrial-pathway-mediated apoptosis.

In Vivo Acute Toxicity Study of 10b
To evaluate the toxicity of compound 10b, an acute toxicity assay was performed using Balb/C mice. After the administration of cucurbitacin B or compound 10b by intravenous injection, the clinical symptoms, deaths, and body weight were observed. As shown in Figure 3A, the percentage of death was 3/3, 3/3, 3/3, 3/3, and 3/3 after the administration of cucurbitacin B at a dose of 2 mg/kg, 5 mg/kg, 10 mg/kg, 20 mg/kg, and 50 mg/kg, respectively, while the percentage of death was 0/3, 0/3, 0/3, 0/3, and 1/3 after the administration of compound 10b at a dose of 2 mg/kg, 5 mg/kg, 10 mg/kg, 20 mg/kg, and 50 mg/kg, respectively. Moreover, as shown in Figure 3B, the body weight did not change after the administration of compound 10b at different doses as compared to the vehicle group. These results demonstrate that compound 10b shows preferable safety and tolerability as compared with cucurbitacin B.

In Vivo Acute Toxicity Study of 10b
To evaluate the toxicity of compound 10b, an acute toxicity assay was performed using Balb/C mice. After the administration of cucurbitacin B or compound 10b by intravenous injection, the clinical symptoms, deaths, and body weight were observed. As shown in Figure 3A, the percentage of death was 3/3, 3/3, 3/3, 3/3, and 3/3 after the administration of cucurbitacin B at a dose of 2 mg/kg, 5 mg/kg, 10 mg/kg, 20 mg/kg, and 50 mg/kg, respectively, while the percentage of death was 0/3, 0/3, 0/3, 0/3, and 1/3 after the administration of compound 10b at a dose of 2 mg/kg, 5 mg/kg, 10 mg/kg, 20 mg/kg, and 50 mg/kg, respectively. Moreover, as shown in Figure 3B, the body weight did not change after the administration of compound 10b at different doses as compared to the vehicle group. These results demonstrate that compound 10b shows preferable safety and tolerability as compared with cucurbitacin B.

General Methods
Unless otherwise mentioned, all reactions were carried out under a nitrogen atmosphere with dry solvents under anhydrous conditions. Reactions were monitored by thin-layer chromatography carried out on 0.25 mm Tsingdao silica gel plates (60F-254). Visualization was achieved using UV light, phosphomolybdic acid in ethanol, or potassium permanganate in water, each followed by heating. Tsingdao silica gel (60, particle size 0.040-0.063 mm) was used for flash column chromatography. NMR spectra were recorded with a 400 MHz ( 1 H: 400 MHz, 13 C: 100 MHz) spectrometer. Data are reported as follows: chemical shift, multiplicity (s = singlet, d = doublet, t = triplet, q = quartet, br. = broad, m = multiplet), coupling constants, and integration. Purity testing was done by means of analytical HPLC on a Shimadzu LD-20A system with an ODS-C18 column (4.6 × 150 mm, 5 μm) eluted at 1-1.3 mL/min with Milli-Q water and CH3CN. The purity of all tested compounds was ≥95% by HPLC (Supplementary Materials).
Compound 8 (154 mg, 0.2 mmol) and HATU (95.1 mg, 0.25 mmol) were dissolved in dry DMF (4 mL), then DIPEA (44 µL, 0.25 mmol) was added. The mixture was stirred at 30 • C for 1 h, then the corresponding amine (0.6 mmol, 3 eq.) was added. The reaction was stirred for 2 h at this temperature. Then, the reaction was quenched with saturated aqueous NH 4 Cl and extracted with EA (3 × 15 mL). The combined organic layers were washed with saturated brine, dried over Na 2 SO 4 , and concentrated to give an oily crude product, which was purified on a silica gel column (PE/EA = 3:1-1:1) to give compound 7 as a yellow solid, which was simple purified to yield a white solid. The solid was used for the next step directly. The solid was dissolved in THF (2 mL). To the mixture was added HOAc (38 µL, 0.65 mmol) and TBAF (169 mg, 0.65 mmol). The mixture was stirred at room temperature for 24 h, and then diluted with ethyl acetate (20 mL). The organic phase was washed with H 2 O (3 × 20 mL), dried over Na 2 SO 4 , and concentrated. The residue was purified by column chromatography on silica gel to obtain a white solid 9a-9c.
To a solution of compound 2 (135 mg, 0.2 mmol), EDCI (96 mg, 0.5 mmol), DMAP (1.2 mg, 0.01 mmol), and the corresponding acid (0.4 mmol, 2 eq.) in CH 2 Cl 2 (2 mL) was added Et 3 N (69.5 µL, 0.5 mmol) at 0 • C. The mixture was stirred for 8 h at room temperature. The reaction was quenched with saturated aqueous NaHCO 3 and extracted with CH 2 Cl 2 (3 × 15 mL). The combined organic layers were washed with saturated brine, dried over Na 2 SO 4 , and concentrated to give an oily crude product, which was simple purified to yield a white solid. The solid was used for the next step directly. The solid was dissolved in THF (2 mL). HOAc (38 µL, 0.65 mmol) and TBAF (169 mg, 0.65 mmol) were added to the mixture. The mixture was stirred at room temperature for 24 h, and then diluted with ethyl acetate (20 mL). The organic phase was washed with H 2 O (3 × 20 mL), dried over Na 2 SO 4 , and concentrated. The residue was purified by column chromatography on silica gel to obtain a white solid 10a-10o.

MTT Assay
HepG-2 and L-O2 cells in logarithmic growth phase were seeded into a 96-well plate at 5 × 10 3 /well/200 µL. After 12 h, compounds at different concentrations were added and incubated for 72 h at 37 • C, 5% CO 2 . Then, 20 µL MTT (5 mg/mL) was added into each well and incubated for 4 h. The supernatant was discarded, and the precipitate was dissolved with DMSO. The absorbance of the optical density (OD) at 570 nm was detected using a microplate reader and the IC 50 was calculated with GraphPad Prism 5.

Cell Apoptosis Assay
HepG-2 cells were seeded into a 24-well plate with 5 × 10 4 /well/1 mL. After 12 h, cells were treated with compound 10b at different concentrations for 48 h. The cells were collected, washed with cold PBS, and stained with Annexin-V/PI according to the manufacturer's instructions. Then, the cells were analyzed with flow cytometry.

Western Blot Assay
HepG-2 cells were treated with compound 10b for 24 h, and the cell lysates were collected. Then, the total protein was separated by SDS-PAGE and transferred to PVDF membranes. The membranes were incubated with primary antibodies (1:1000) at 4 • C overnight. After being washed with PBST (PBS + 0.5% Tween 20) 5 times, the membranes were incubated with HRP-conjugated secondary antibodies (1:5000 dilution) for 2 h at room temperature. The membranes were then washed with PBST 5 times and the protein blots were detected by ECL chemiluminescence.