The dried roots of E. longifolia were provided by Prof Dr Azimahtol Hawariah Lope Pihie (National University of Malaysia, Malaysia).

The dried roots (500 g) of E. longifolia were extracted with methanol under reflux (60°C, 48 hours) and concentrated using rotary evaporator to give dark brown syrup. The methanol extract was partitioned with diethylether (Et₂O) saturated with water. The aqueous soluble portion was further partitioned with n-butanol (BuOH) and water (1:1). The BuOH extract was subjected to column chromatography on silica gel column and eluted with ethyl acetate (Et_OAc); ethanol (EtOH); water (H₂O) (100:10:1). Fractions were collected, checked by preparative Thin Layer Chromatoghraphy (TLC) using 20% methanol in chloroform as an eluent to give 6 partially purified fractions (F16–F21) of eurycomanone. Fractions contained eurycomanone were further analyzed by High Performance Liquid Chromatography (HPLC).

HepG2, HM3KO, CaOV3, Hela, CCD11114sk and Chang's liver cells were obtained from the American Type Culture Collection (ATCC, Manassas, Virginia, USA). Cells were maintained in Dulbecco's modified eagle's medium (DMEM; Invitrogen Co., Carlsbad, California, USA) supplemented with 10% heat-inactivated fetal calf serum (Invitrogen Co.), 100 U/ml penicillin and 100 mg/ml streptomycin (Flowlab, Sydney, Australia) in a humidified atmosphere of 5% CO₂ in air at 37°C. Cells were kept in the logarithmic growth phase by routine passage every 2–3 days using 0.025% trypsin-EDTA treatment.

The antiproliferative activity was evaluated by using the MTT method as previously described, with some modifications 23. Briefly, cells were seeded 24 hours prior to treatment in a 96-well plate at 5 × 10⁴ cells/well in order to obtain 70% to 80% confluent cultures. Before addition to the culture medium, the crude extract of E. longifolia was dissolved in water and eurycomanone was dissolved in DMSO (Sigma Chemical Co., St. Louis, Missouri, USA) and followed by a 2× serial dilution for 10 points ranged from 0.2 μg/ml to 100 μg/ml. The final concentration of DMSO used in the corresponding wells did not exceed 1% (v/v). This concentration of DMSO does not effects cell viability 18. Control cultures received the same concentration of solvent alone and tamoxifen (Sigma Chemical Co.) was used as a positive control. The 96-well plate was incubated for 72 hours at 37°C in a humidified atmosphere with 5% CO₂. At the end of incubation, 50 ul of MTT solution (2 mg/ml MTT in plain culture medium; Sigma Chemical Co.) was added to each well. The plate was then incubated for 4 hours. MTT solution was removed and the purple formazan crystal formed at the bottom of the wells was dissolved with 200 μl DMSO for 20 minutes. The absorbance at 570 nm was read on a spectrophotometric plate reader (Thermo Electron Co., Waltham, Massachusetts, USA). The proportion of surviving cells was calculated as (OD of drug-treated sample – OD of blank)/(OD of control – OD of blank) × 100%. Dose-response curves were constructed to obtain the IC₅₀ values. All experimental data were derived from at least 3 independent experiments.

Nuclear staining with Hoechst 33258 (Sigma Chemical Co.) was performed as described elsewhere 171824. Briefly, the floating and trypsinized-adherent of treated and non-treated cells were collected and washed with PBS. The cells were then fixed with 4% paraformaldehyde in PBS for 30 minutes at 4°C. After washing, the cells were smeared onto microscope slides followed by incubation in Hoechst 33258 at a final concentration of 30 μg/ml at room temperature for 30 minutes. Nuclear morphology was then examined under a fluorescent microscope (Imaging Source Europe GmbH, Bremen, Germany).

HepG2 cells were grown in 25 cm² culture flasks under a humidified 5% CO₂ atmosphere at 37°C. The seeding density was 5 × 10⁴ cells/ml. Flasks were pre-incubated for overnight before the solutions of drugs or control were added to each flask. Vinblastine sufate (2.5 mg/ml) was used as a positive control and negative control received only DMSO. The concentration of eurycomanone used in this experiment was 5 μg/ml. After a period of exposure (24, 48 and 72 hours), cells were harvested by trypsinization with trysin/EDTA solution (Sigma-Aldrich, T4049), collected and fixed in ice-cold 70% ethanol for at least 1 h at 4°C. Cell pellet were resuspended with 0.9 ml 1% Triton X-100 in PBS and 100 μl of 1 mg/ml RNase A (Sigma-Aldrich, R5500). The solution was incubated for 30 minutes at 37°C. Cells were stained with 100 μl of 10 μg/ml PI (Sigma-Aldrich, P4170) solution in staining buffer (1% Triton-X 100 in PBS) and protected from light. Stained cells were incubated at room temperature (RT) for 30 minutes. Samples were then put back in ice and 300 μl of staining buffer and analyzed using a Beckman-Coulter FACS Calibur-500 flow cytometry (FCM).

The number of apoptotic cell death induced by eurycomanone was measured by flow cytometry using the BD Pharmingen™ annexin V-FITC, BD 556420 kit. The treated and untreated HepG2 cells were harvested and washed with cold PBS. The cell pellet was re-suspended in 1× binding buffer at a concentration of 1 × 10⁶ cells/ml. 100 μl of cell suspension was transferred into FCM tube. 5 μl of Annexin V-FITC and 5 μl of PI were added into the cell suspension, followed by gentle vortexing. The staining samples were incubated for 15 minutes at room temperature (RT) in darkness. An additional 400 μl of 1× binding buffer was added to each tube. Samples were analysed by Beckman-Coulter FACS Calibur-500 flow cytometry within 1 hour. 100 000 events was acquired using green channel FL1 for Annexin V-FITC and the red channel FL3 for PI.

Cells were cultured in 25 cm² culture flasks (5% CO₂, 37°C) at seeding density of 5 × 10⁴ cells/ml. Plates were pre-incubated overnight before the solutions of drug were added to each well. After a period of exposure (24, 48 and 72 hours), cells were harvested by trypsinization with trypsin/EDTA solution and centrifuged at 1800 rpm. Cells were washed twice with 1× PBS and then were fixed with ice-cold 70% ethanol (≤ 1 hour, 4°C), washed twice with 1× PBS and re-suspended in blocking buffer (2% BSA) for 10 minutes, followed by another washing step. Cells pellet were re-suspended in PBS to a concentration 1 × 10⁷ cells/ml and 100 μl of cells suspension (1 × 10⁶ cells/ml) were transferred into each sample tube and 20 μl of FITC/PE conjugated antibody were added into tubes followed by gently mixed. The antibodies used in this experiment are: FITC-conjugated Bcl-2 antibody (BD Pharmingen); PE-conjugated Bax antibody (Santa Cruz Biotechnology, Inc.); PE-conjugated p53 antibody (BD Pharmingen); FITC-conjugated Cytochrome C antibody (Santa Cruz Biotechnology, Inc.). The tubes were incubated at RT for 20 – 30 minutes in the dark. The pellets were washed with 2 ml of 1× PBS and the supernatant were aspirated. Cell pellets were re-suspended in 500 μl of 1× PBS and ready to proceed for FCM analysis (Beckman-Coulter FACS Calibur-500)

All values are expressed as the mean ± S.D. Statistical analyses were evaluated by Student's t-test. Probability values p < 0.05 were considered statistically significant.

The anti-proliferative activity of E. longifolia will be assessed from its IC₅₀ on various cell lines. The aim of preliminary screening was to select the most effective suppression of cell growth against E. longifolia treatment. Because of the limited time and financial support for present study, only the cell line which showed most effective inhibition by the extract was selected for further work.

Results from the preliminary assay for cytotoxic activity of E. longifolia crude extract were evaluated by obtaining the IC₅₀ values as in figure 1a. IC₅₀ value is an effective dose required to inhibit the proliferative response by 50%. Onset of action and the percentage survival of cells have been taken into consideration in order to categories the potency of the crude extract. In the US NCI plant screening program, a crude extract is generally considered to have in vitro cytotoxic activity if the concentration that causes a 50% cell kill (in KB carcinoma cells) is less than 20 μg/ml and less than 4 μg/ml for the pure compound, following incubation between 48–72 h 25. In this study, this value was used as a rough reference point for assessing the activity of the natural agent used.

Within the four cancer cell lines used in this experiment, crude extract of E. longifolia was more effective against HepG2 cells with IC₅₀ 45 ± 0.15 μg/ml. Both HM3KO and Hela cells gave an IC₅₀ value of 60 ± 0.25 μg/ml. For the CaOV3 cells, the IC₅₀ obtained was 79 ± 0.16 μg/ml. Tamoxifen, a standard therapeutic drug used as a positive control in this experiment showed a great toxicity towards all the cancer cell lines used. Tamoxifen have been studied intensely particularly on the breast cancer and it was approved by FDA for reducing the incidence of breast cancer. As we can see from the graph in figure 1b, tamoxifen show very low IC₅₀ values with 3.7 ± 0.28 towards HepG2 cells and 2.3 ± 0.23 towards HM3KO cells. The IC₅₀ values for Hela cells and CaOV3 cells were 2.7 ± 0.18 and 2.1 ± 0.32 respectively.

However, when compared to E. longifolia crude extract, tamoxifen not only toxic against malignant cells but also toxic towards non-malignant cell, Chang's liver and CCD1114sk. Tamoxifen gave an IC₅₀ value of 2.8 ± 0.21 against CCD11114sk and 1.4 ± 0.31 towards Chang's liver cells displaying the somewhat non-selective cytotoxicity of tamoxifen. In contrast to tamoxifen, no anti-proliferative effect observed in both non-malignant cells. No IC₅₀ was recorded.

The antiproliferative assay showed a great inhibitory effect of eurycomanone towards HepG2 cells. Eurycomanone as shown in figure 2a, significantly reduced the viability of HepG2 cells by 50% inhibition at 3.8 ± 0.12 μg/ml. When compared to control drug, tamoxifen which is a standard chemotherapeutic drug, the values of IC₅₀ for both (eurycomanone and tamoxifen) were very close together. Tamoxifen inhibited HepG2 cells at IC₅₀ value of 3.7 ± 0.28 μg/ml. Eurycomanone treatment did not exhibit any cytoselectivity towards normal liver cells, Chang's Liver which gave IC₅₀ of 17 ± 0.15 μg/ml. The non-cytoselectivity effect was similar to the treatment with Tamoxifen. However, Tamoxifen is more cytotoxic towards normal liver cells which inhibit 50% cells viability at 1.4 ± 0.31 μg/ml. As compared to Tamoxifen, Eurycomanone was 12 times less toxic against normal liver cells.

Another non-malignant human liver cell, WLR-68 also used in this experiment as a negative control. Interestingly, the results showed that eurycomanone less toxic towards WLR-60 cells with IC₅₀ 20 ± 0.22 μg/ml (figure 2b). As compared to vinblastine sulfate, this compound showed a highly toxicity towards, normal liver cell WLR-68 with IC₅₀ of 4.2 ± 0.37 μg/ml was detected. This finding also supported that eurycomanone have less cytotoxic effect, which mean that the compound highly toxic toward malignant cells but not in normal cells.

The mode of killing induced by most anticancer agents is by which apoptotic cell death and DNA fragmentation is a hallmark of apoptotic cells 26. To further examined the morphological changes in responds to eurycomanone treatment, both control and eurycomanone-treated cells were stained with fluorescent dye Hoechst 33528 and visualized under a fluorescent microscope. Hoechst nuclear staining binds to the AT rich regions of double stranded DNA and exhibits enhanced fluorescence.

The Hoechst dye was able to diffuse through intact membranes of HepG2 cells and stain their DNA. Figure 3 showed the results of Hoechst staining for the eurycomanone treated-HepG2 at 24, 48 and 72 hours. The untreated HepG2 cells remained uniformly stained. HepG2 cells show that no fluorescence was detected in the nucleus, as the cells were not apoptotic and did not exhibit DNA fragmentation. At 24 hours treatment, cells began to display apoptotic morphology. Fluorescence was detected in the nuclear region of the HepG2 cells, indicating the presence of DNA fragmentation. Enucleation and apoptotic bodies can be seen as small fluorescent masses. At 48 hours, nuclear condensation and fragmentation was more apparent. The chromatin is condensed into lumps, exhibiting punctuated morphology typical of apoptotic cells, whereas at 72 hours, apoptotic events culminated in the formation of apoptotic bodies, seen as smaller intense spots. Shrinkage of cells, DNA condensations, nuclear and plasma membrane convulsion and nuclear fragmentation were all observed in treated cells. Positive control of tamoxifen-treated cells displayed similar nuclear fluorescence indicating that the data obtained can be accepted. These observations provide evidence that an apoptotic pathway is occurring with the eurycomanone treatment.

The antiproliferative activity displayed by eurycomanone may be by a selective cytotoxic manner. A wide variety of agents trigger growth arrest as well as cell death 27. Many reports indicate that cell cycle arrest leads to cell growth inhibition or apoptosis. Cell cycle arrest is a good marker for chemopreventive or anti-tumor activity of chemicals or drugs 28. However, no reports to show a checkpoint for the cell cycle arrest towards HepG2-treated eurycomanone. Therefore, further investigated the natures of cell cycle progression in HepG2 cells under eurycomanone treatment using a flow cytometer approach were done. The cell cycle distribution was examined at various times and indicated doses. Vinblastine sulfate was used as a positive control.

As shown in figure 4, cells under normal conditions (untreated) showed the expected pattern for continuing growing cells by which the highest peak in the region of G1 phase and a small low peak in G2/M phase. The number of cells count was calculated and presented in a graph in figure 5. There were no significant differences within the G1 peak for untreated cells at 24 and 48 hours with percentage cell count of 72.5% and 73.6%, respectively (figure 5a). Besides, at 72 hours of treatment duration, a bit decreased of cells population to 67.5% was observed. Maybe, it is because of the normal process for the cells to go apoptosis following increasing time of incubation. However, most of the cells accumulated at G1 phase.

Cells treated-eurycomanone showed an accumulation in G2/M phase following 48 hours of exposure with 20.8% of the cells population compared to 24 hours with 16.58% (figure 5b). At 72 hours, the significantly increased percentage of cells in G2/M phase was observed with 39.9% of cells accumulated was recorded. Along with decreasing of G2/M phase, a dramatically decreased of HepG2 cells population in G1 phase were detected with percentage of 59.9% at 24 hours to 55.4% at 48 hours and 39.9% at 72 hours. This finding suggested that eurycomanone inhibited the cell proliferation by inducing the G2/M arrest in a time-dependent manner in HepG2 cells through regulating the G2 checkpoint.

The amount of apoptotic cells was calculated based on the appearance of cells in G0. G0 phase is corresponds to apoptotic cells. Although there was an increase of apoptosis in eurycomanone-treated cells, the magnitude of change was relatively small in comparison to the growth inhibition. Distribution of the G2/M phase cells in the population increase while the percent of apoptotic cells rose from 3.5% to 10.4%. This is also evident that the G2/M phase became the dominant phase in cells treated with 5 μg/ml of eurycomanone in a time-dependent manner with a late apoptosis. These finding indicate that at the ranges of concentration studied, the antiproliferative effect of eurycomanone on HepG2 cells could be attributed primarily to the induction of G2/M arrest, with less contribution of cell division rather than DNA synthesis.

Following vinblastine sulfate treatment (figure 5c), which is a control drug, the G2/M peak dramatically decreased cell population of G1 phase (2.5%) and increased percentage of cells in G2/M phase (80.7%) at 24 hours exposure compared to control. Therefore, at 48 h of treatment, G2/M phase decreased to 49.1%, along with increased of G0 phase from 10.6% at 24 hours to 17.7%. Notably, maximal G2/M phase arrest by vinblastine sulfate administration was observed at 24 hours and at 72 hours for eurycomanone. The significant accumulation of cells clearly indicates that vinblastine sulfate efficiently arrested the cell cycle progress in HepG2 cells at the G2/M phase. This result is consistent with those in the literature 29 and testifies that the established cell cycle analysis protocol with HepG2 cell line was working properly in the present work.

To further confirm that eurycomanone induces cell apoptosis, HepG2 cells were stained with annexin V-FITC and PI, and subsequently analyzed by flow cytometry (FCM). As indicated in figure 6a, in untreated cells only 0.3% of HepG2 cells were annexin V⁺/PI^-, indicating the apoptotic cells, whereas 0.58% of the cells were double positive (late apoptotic cells) at 24 hours of incubation time. The apoptotic cells showed a bit increased after 48 hours treatment with 0.69% and late apoptotic cells with 0.95%. At 72 hours of treatment duration, only 6.4% of apoptotic cells were detected with 0.04% of late apoptotic cells. Most of the cells for all treatment duration (24, 48 and 72 hours) were still viable with percentage of 97.43%, 95.9% and 93.6%, respectively (figure 6b). This result significantly showed that in untreated HepG2 cells, cells continuously growth and an apoptosis process just appeared as a normal process in every cell lives. Results also supported the finding from the analysis of DNA contents which previously reported in this study.

After treatment with eurycomanone (figure 7a) for 24 hours, apoptotic cells increased to 32.97% along with decreased in viable cells to 65.36%. Following exposure for 48 hours, annexin V⁺ quadrant dramatically increase to 80.5% and late apoptotic cells also increased with 12.10% were detected. Viable cells decreased to 10.0%. However, at 72 hours of treatment, only 74.1% of the cells were detected in quadrant of annexin V⁺/PI^-, but the increased of cells population were observed at late apoptotic quadrant with percentage of 15.3%. Eurycomanone was showed to induce apoptosis in HepG2 cells in a time-dependent manner. After 72 hours of exposure, only 5.6% cells were alive indicating that almost all of cells undergo apoptosis (figure 7b). This data confirm that eurycomanone was capable of inducing cell death in HepG2 cells and suggested that the mechanism of cell death in HepG2-treated eurycomanone was via apoptosis process. According to Nurkhasanah et al. 2008, after exposure of Hela cells with eurycomanone, the level of apoptotic cells in annexin ⁺/PI^- quadrant was increased at 24 and 48 hours, thus indicating that eurycomanone induced apoptosis in Hela cells. Some of natural resources also reported to induce apoptosis in HepG2 cells such as solanine or Solanum nigrum Linn 30 and apigenin 31.

Vinblastine sulfate used as a positive control in this present experiment. Vinblastine sulfate showed a decreasing number of viable cells with increasing time of incubation (figure 8a). At 24 hours, percentage of cells population in double negative quadrant was 57.84% and decreased to 25.1% at 48 h and 10.3% at 72 hours. Interestingly, at 72 hours, late apoptotic cells greatly increased to 69.8% along with 17.0% of apoptotic cells compared to percentage of 46.2% at 48 hours and 40.47% at 72 hours with late apoptotic of 24.2% and 0.38%, respectively (figure 8b). This data have similarity with the results of DNA contents for cell cycle analysis which mentioned above.

To investigate the intracellular mechanism for the observed increase in apoptosis in eurycomanone-treated HepG2 cells, the expression of the pro- apoptotic Bax and anti-apoptotic Bcl-2 proteins will be studied. Also, the involvement of the tumor suppressor protein p53 will be determined along with cytochrome C release. Protein levels discerned from treated and non-treated cells will then be quantitated and compared to observe possible modulation by eurycomanone. Flow cytometry approach was used to detect the levels of proteins with specific antibody conjugated-FITC (fluorescein isothiocyanate) and PE (phycoerythrin).

The analysis of the effect of eurycomanone on the expression of Bcl-2 protein in HepG2 cells using flow cytometry shows that eurycomanone can inhibit Bcl-2 activity as the time of exposure increases. The expression of Bcl-2 protein is lower as shown in figure 9b compared to the untreated HepG2 cells (figure 9a). In untreated HepG2, a great accumulation of Bcl-2 proteins were observed. At 24 hours of treatment duration, 98.3% of Bcl-2 was detected and no significant increases for the other proteins, Bax, p53 and cytochrome C. Following treatment at 48 h and 72 hours, the Bcl-2 levels slowly decreased to 86.0% and 81.5%, respectively (figure 10a). Furthermore, the expression of Bax protein have been seen to increase but only in a small value which is 0.7% at 48 hours to 3.1% at 72 hours.

HepG2 treated-eurycomanone inhibited the expression levels of Bcl-2. On the contrary from untreated cells, at 24 hours of exposure with eurycomanone, Bcl-2 levels dramatically decreased to 14.9% and again, at 48 hours, the levels decreased to 6.9% and only 1.8% after 72 hours of treatment. The accumulation of Bax proteins were significantly occurred (figure 10b). The percentage of 79.1% of Bax proteins were calculated at 24 hours of incubation with eurycomanone. The expression levels of Bax increases in a time dependent. At 48 hours, the Bax proteins increased to 88.5% and further increased to 92.3%. These results also supported by the previous studied which shows that eurycomanone suppressed the levels of Bcl-2 proteins and increased the expression levels of Bax in Hela cells (Nurkhasanah et al. 2008).

As for in HepG2 treated-vinblastine sulfate, Bcl-2 levels also inhibit in a time dependent manner. From 27.3% at 24 hours, the expression levels of Bcl-2 decreased to 23.7% and greatly reduced to 4.1% at 74 hours. The levels of Bax proteins also dramatically increased. The expression levels of 55.8% and 89.1% were detected at 24 and 48 hours, respectively. The expression levels showed a maximal value at 72 hours with percentage level of 95.6% (figure 10c).

p53 protein was detected to increase from 24 to 72 hours of exposure, thus recommended that eurycomanone induces apoptosis dependent p53 pathway. As shown in figure 9c, eurycomanone triggered strong p53 protein expression in HepG2 cells with 78.5% of p53 proteins levels were expressed at 24 hours of exposure and at 48 hours, the expression level increase to 84.9% and 95.3% at 72 hours. Along with decreasing of p53 levels, cytochrome C also increased in a time dependent manner (figure 9d). The percentage of 74.4% was determined at 24 hours of treatment and 86.1% at 48 hours. At 72 hours, the level of cytochrome C proteins showed a maximal value of 95.5%.

Furthermore, a different pattern of HepG2-treated vinblastine sulfate was observed in the level of p53 proteins and cytochrome C. As we can see from the figure 10c, there were no significant increased in a level of p53 in HepG2 cells treated-vinblastine sulfate for all treatment durations. These results show that, vinblastine sulfate enhances apoptosis in HepG2 cells via a p53-mediated pathway. However, the expression levels of cytochrome C were increased from 76.6% at 24 hours to 85.6% and 92.8% at 48 and 72 hours, respectively.