Anticancer Activities of Polyynes from the Root Bark of Oplopanax horridus and Their Acetylated Derivatives

Six polyynes OH-1~6, some of which are occur naturally in acetylated form, had been isolated and identified from the root bark of Oplopanax horridus (Devil’s Club), a natural dietary supplement and medicinal plant in North America. During the evaluation of the polyynes’ potential anticancer activities, sixteen more acetylated derivatives OHR-1~16 have synthesized and their anti-proliferation activity on MCF-7, MDA-MB-231, A549, HepG2 and LO2 cells assayed to elucidate their structure-activity relationships. The results showed that OH-1 ((3S, 8S)-falcarindiol) had the most potent anticancer activity, with IC50 values of 15.3, 23.5, 7.7 and 4.7 μM on MCF-7, A549, HepG2 and MDA-MB-231 cells, respectively. For the primary structure-activity relationship, the anticancer activities of polyynes become weaker if their hydroxyl groups are acetylated, the terminal double bonds transformed into single bonds or they contain one more methylene group in the main skeleton chain.

Recent studies have revealed that four purified polyynes from O. horridus showed potential anticancer activities [16][17][18]. All these polyynes possess hydroxy groups in their structures, and two of them, which occured naturally in acetylated form had weaker anti-proliferation effects. The primary hypothesis based on the anti-proliferation investigation of only these four polyynes on the selected cancer cell lines was that acetylation of polyynes had a negative contribution to their anticancer activities. However, as the two acetylated polyynes had 18 carbon atoms in the main skeleton chain (C18 polyynes) while the others had 17 carbon atoms as the main structural chain (C17-polyynes). It seems that they could not be compared diretly together to reach that conclusion [17].
During the course of discovering interesting anticancer molecules from the title plant, six polyynes OH-1~OH-6 ( Figure 1) had been purified from hydrophobic parts of the herbal medicine extracts and identified. Some of these compounds had displayed anti-proliferation activity against certain cancer cell lines [17,19]. Among the polyynes, (3S,8S)-falcarindiol had been obtained separately from (3R, 8S)-falcarindiol, which was also reported to have anticancer activities [20]. Since two of the polyynes from O. horridus naturally occur acetylated, sixteen more acetylated derivatives OHR-1~OHR-16 ( Figure 1) have now been synthesized for the evalutation of the polyynes' potential anticancer activities. The possible mechanisms and structure-function relationships in anti-proliferative activity were also studied.

Chemistry
The known compounds were identified by comparing their physical and spectroscopic data with values reported in the literature. The structures of acetylated compounds were characterized by IR, 1 H-NMR, 13 C-NMR, and HMBC spectra. Especially, the position of the acetyl group in the compounds was elucidated and fixed by the HMBC spectra. The purity (≥96%) of the target compounds was verified by HPLC.
Six polyynes OH-1~OH-6 had been isolated and purified from the root bark of O. horridus. Acetylated polyynes OHR-1~OHR- 16 were synthesized from OH-1~OH-6 with acetic anhydride in ethyl acetate with sodium carbonate added to the stirring mixtures. After quenching the reactions, the reaction mixture was cooled to room temperature and evaporated to remove the organic solvents. The residue was diluted with water and then extracted with chloroform. The chloroform layer was evaporated to remove the organic solvents. After that, the acetylated products were subjected to chromatographic separation to obtain the derivatives OHR-1~OHR-16. Figure 1. The structures of the isolated and derivative polyynes.

Effects of the 22 Polyynes on Proliferation of Selected Cancer Cells
The polyynes isolated from O. horridus have been evaluated for antiproliferative effects on some selected cancer cell lines such as human breast cancer cell line MCF-7, non-small cell lung cancer (NSCLC) cells, human colorectal cancer cell lines HCT-116 and SW-480 [21][22][23]. As shown in Table 1, the 22 polyynes exhibited different antiproliferative effects on the human breast cancer MCF-7, human lung adenocarcinoma epithelial A549, liver hepatocellular HepG2 and human breast cancer µM-231 cell lines. At the adopted concentrations (1-300 µM), polyynes OHR-7~8 and OHR-13~16 were not observed to inhibit the cancer cell growth of any of the four cell lines. OHR-13 and OHR-14 showed some antiproliferative effects on MDA-MB-231 cancer cells at about 150 µM, but such effects were not observed in other cancer cell lines. Moreover, other polyynes showed potentially different cell growth inhibition of the four cancer cell lines.  (Table 1). OH-1 and OHR-3 showed stronger effects in that A549 cell growth was inhibited by 95.6% and 90.8%, respectively, when administrated at 60 µM (both p < 0.01). In A549 cells, OH-1 and OHR-3 showed the most potent antiproliferative effects.
In HepG2 and MM-231 cells, OH-4 and OHR-9~11 displayed moderate antiproliferative effects on the two cancer cell lines at lower concentrations than those of A549 and MCF-7 cell lines ( Table 1). The IC50 values of OH-1~3, 5~6 and OHR-3~6 were observed much lower on this cell lines than those of A549 and MCF-7 cells (Table 1). Among all the polyynes, OH-1 also showed the most potent antiproliferative effects in these two cell lines, with IC50 values of 7.7 ± 1.3 µM and 4.7 ± 1.4 µM, respectively.

Anti-Proliferative Activity and Possible Structure-Activity Relationships
Anti-proliferative tests of the six isolated polyynes OH-1~OH-6 and their derivatives OHR-1~OHR- 16 were conducted on four cancer cell lines and a normal human hepatic cell line. As shown in Table 1, the 22 compounds exhibited different extents of anti-proliferative activity on the cancer cell lines with various IC50 values.
The biological activities of the polyynes with same structure features were compared ( Figure 2). In Figure 2A, the anti-proliferation effects of all the isolated compounds OH-1~OH-6 showed that OH-1, OH-3 and OH-5 had stronger activities than OH-2, OH-4 and OH-6 on all the investigated cancer cell lines, except for OH-5 and OH-6 on A549 cells, respectively. The OH-5 and OH-6 effects on A549 cells deviated, possibly due to different mechanisms on anticancer activity of this cell line. Additionally, OH-1 and OH-2 containing 17 carbon atoms in the main chain (C17-polyynes) displayed higher activity than other polyynes consisting of 18 carbon atoms as the structure skeleton chain (C18-polyynes).
In Figure 2B, the acetylated polyynes OHR-3, OHR-9 and OHR-12 show higher activities than OHR-5, OHR-10 and OHR-16 on all the cancer cell lines. These polyynes had their acetyl groups connected with the hydroxyls in the middle of the carbon chain. Among them, OHR-3 and OHR-5 derivatizated from OH-1 and OH-2, respectively, possessed higher inhibition than other acetylated C18polyynes. In Figure 2C, the acetylated polyynes OHR-4 and OHR-11 possessed better activities than OHR-6 and OHR-11 toward all the cancer cell lines except A549 cells, respectively. OHR-13, OHR-14 and OHR-15 even exhibited no cytotoxicity on any of the selected cancer cell lines except weak activity on MDA-MB-231 cells (IC50 > 120 μM). These polyynes synthesized with the acetyl group connected with the hydroxyl near the end of the carbon chain were compared with the ones in Figure 2B to possibly conclude that the hydroxyl at the end of the carbon chain contributed more to the anticancer activity than the hydroxyl in the middle of the carbon chain. In Figure 2D, OHR-1, OHR-2, OHR-7 and OHR-8 almost showed no cytotoxicity toward most of the cancer cell lines when all the hydroxyls in these polyynes were substituted, although OHR-1 and OHR-8 had weak effects on MCF-7 and HepG2 cells with IC50 > 150 µM.
For these similar polyynes, the anti-proliferation effects data showed that the compounds with terminal ethylenic bonds exhibited higher activity compared to the polyynes with terminal single bond, without consideration of the ones (IC50 ≥ 300 μM); The C-17 polyynes possesed stronger activities than the C18-polyynes, which has one more methylene group than the C17-polyynes except the polyynes with IC50 values more than 300 μM. In addition, hydroxypolyynes had higher activity than acylated polyynes. The primary structure-activity analysis thus showed that the observed inhibitions were influenced mostly by the carbon chain length, terminal ethenyl, hydroxyl groups and acylations of polyynes.
Moreover, the compounds' effects on the human normal hepatic cells was investigated to evaluate their potential hepatotoxicity in vitro. It was found that the compounds with higher anti-proliferative activity presented more potential cytotoxic effect on normal human hepatic cells as well (Table 1), which means these compounds may have potential hepatotoxicity in vivo. Actually, the safety and efficiency of O. horridus have not been evaluated totally thus far though it has a long history of use medicinally and as a dietary supplement [21,24], so more studies are definitely needed.

Apoptosis and Cell Cycle Assays
Next, cell cycle analysis and apoptosis assays were performed on the cancer cells. Previous work had reported that OH-1~4 exhibited potential anticancer acitives on human breast cancer and colon cancers through cell arrest in G2/M phase and induction of appoptosis at both earlier and later stages [17]. In this study, the potential mechanisms of the four new compounds OHR-1, -2, -3 and -5 with the strongest anti-proliferative activity among the series of acylated polyynes were investigated. It was found that these four acylated polyynes could induce obvious apoptosis in MM-231 cells, which showed classical apoptotic morphology, chromatin condensation and apoptotic bodies ( Figure 3) with Hoechst staining. On the other hand, JC-1 dye was applied to test the Δψm, which is an important parameter of mitochondrial function used as an indicator of cell health. In apoptotic cells with low Δψm, JC-1 remained in the monomeric form, and showed only green fluorescence in cells. The results showed that OHR-1, -2, -3 and -5 at high concentration could induce MM-231 cells apoptosis by decreasing the Δψm, indicated by a change color from red to green fluorescence (Figures 4A-E) and increasing green to red fluorescence intensity ratio ( Figure 4F). Moreover, their influences on the MM-231 cell cycle were determined by PI staining and flow cytometry analysis. It was observed that OHR-1, -2, -3 and -5 could arrest MM-231 cells in G2/M phases by 14.4% ± 3.9% compared to the solvent control. The arrest rates at low conentration were 24.6% ± 3.6%, 22.6% ± 2.6%, 24.2% ± 2.9% and 33.2% ± 4.6%, repectively, while at high concentration they were 31.7% ± 3.2%, 30.4% ± 3.1%, 25.6% ± 7.5% and 38.0% ± 1.7%, repectively, as shown in Figure 5.  These four polyynes did not exhibit significant effects in the G-phase. Therefore, we regard that induction of apoptosis and cell cycle arrest by OHR-1, -2, -3 and -5 contributed to their anti-proliferative effects on MM-231 cells. Acetylated polyynes have the same mechanism as non-acetylated polyynes in the anti-proliferative activity on the cancer cells.  For this investigation, six polyynes and 16 acetylated polyynes were evaluated for their anticancer activities toward selected human cancer lines. OH-1, which has 17 carbon atoms in the main chain with no acetyl group and a terminal ethenyl group, showed the highest anticancer activities in all tested cell lines. This study also suggested that polyynes from O. horridus may possibly be active anticancer ingredients.

General Procedures
Optical rotations were recorded on a Perkin Elmer Model 341 polarimeter. UV spectrum was measured on a Beckman Coulter DU 640 spectrophotometer. IR spectra were obtained with a PerkinElmer Spectrum 100 FT-IR spectrometer with KBr pallets. The 1 H-, 13

Plant Material
The dried roots bark of O. horridus was obtained from Pacific Botanicals Co. Ltd (Chicago, IL, USA) and authenticated by one of the authors (C.-Z. Wang) in March, 2012. A voucher specimen (Lot: OHR-20120926-1) has been deposited in the Institute of Clinical Pharmacology, Central South University, Hunan Province, China.

Chemicals
HPLC-grade methanol and acetonitrile were purchased from Merck. The deionized water used for HPLC was purified by a Milli-Q purification system (Millipore, Billerica, MA, USA). Chloroform, Methanol and Ethyl acetate (analytical grade) was obtained from Beijing Chemical Reagent Plant (Beijing, China). All the liquid and solid reagents were purchased from Sigma (St. Louis, MO, USA).

Extraction and Isolation
The roots bark of O. horridus (7.2 kg) was sieved by a 20 mesh crib after pulverization. Methanol (360 L) was selected to extract the pulverized powder with infusion for 7 days × 3. Then, the methanol extract (900 g) was diffused into pure water (2 L) and extracted with ethyl acetate (EtOAc) and n-butanol to yield the corresponding fractions E (312 g) and B (320 g), respectively. The two organic solvents were used for extraction with the volume of 3 L × 3, resp., after which were saturated using pure water. The EtOAc-soluble fraction E (260 g) was separated by silica gel (100-200 mesh) column chromatography (CC) eluted with a gradient of CHCl3-MeOH (50:1 to 0:1) to give ten fractions (E1-E10). Fraction E7 (53 g

Synthesis of Acetylated Polyynes
A common acetylation method was used. Acetylated polyynes OHR-1~OHR-16 were synthesized by mixing the subfractions E7d (7.5 g), E8d (7.8 g) and E9e (7.8 g) with acetic anhydride (60 mmol) and Na2CO3 (6.36 g) in ethyl acetate (20 mL), respectively. The reaction mixture was refluxed for 2 h at 60 °C while the progress of each reaction was monitored by TLC and HPLC. After quenching the reaction, the mixture was cooled to room temperature and evaporated to remove organic solvents under diminished pressure at 50 °C. The residue was diluted with H2O (50 mL) and each one was extracted with CHCl3 (40 mL × 3). The CHCl3 layer was evaporated and the acetylated products were subjected to chromatographic separation to obtain polyyne derivatives OHR-1~OHR-16 in different yields. Repeated pre-HPLC separation afforded the pure products.

Measurement of Cell Viability
The viability of cells was measured by a colorimetric MTT assay. Briefly, 1 × 10 4 cells/well in 96well microplates were exposed to different concentration of polyynes and their acetylated derivatives respectively for 24 h incubation, following addition of 20 μL MTT solution (4 mg/mL) to each well for another 4 h incubation at 37 °C. Finally, the medium was removed and replaced with 100 μL dimethyl sulfoxide which was added to dissolve the dye crystal presented in cells. The absorbance was recorded at 570 nm using a microplate reader (1420 Multilabel counter victor 3 , Perkin-Elmer, Waltham, MA, USA). The results were expressed as ratio of absorbance between treatments and control cells (solvent vehicle set at 100%). Four replicate wells were tested per assay which was repeated three times.

Hoechst Staining
MDA-MB-231 cells were seeded in the flat-bottomed 6-well plates and treated with OHR-1, -2, -3 and -5. Cells treated with solvent vehicle served as controls. In 12 h after treatment, cells were fixed in 4% polyoxymethylene solution for 30 min and then washed with PBS. The fixed cells were incubated with Hoechst 33342 for 15 min at room temperature in the dark and then washed with PBS twice, which were finally observed under a fluorescence microscope (Convert Fluorescence Microsope Axiovert 200, HAL 100, HBO 100, Carl Zeiss MicroImaging Co., Ltd., Oberkochen, Germany) at a magnification of ×400.

Measurement of Mitochondrial Membrane Potential
To assess the mitochondrial membrane potential, Δψm, JC-1 staining was used to exhibit potential dependent accumulation in mitochondria, indicating reversibly change color from green to red fluorescence as the membrane potential increases. On the other hand, in apoptotic or unhealthy cells with low Δψm, JC-1 showed an increasing ratio of green to red fluorescence intensity. The MDA-MB-231 cells were treated with OHR-1, -2, -3 and -5 for 24 h in a humidified atmosphere (37 °C with 5% CO2). The cells were collected and responded in 0.5 mL culture medium with JC-1 dye (1 μg/mL) for 20 min at 37 °C in dark. Fluorescence was monitored by both flow cytometry (BD Biosciences, San Jose, CA, USA) and fluorescence microscope. The Δψm were indicated by the rato of green to red fluorescence intensity. The data were expressed as ratio of Δψm between treatments and control cells (solvent vehicle set at 100%).