• estrogen;
  • tamoxifen;
  • ginseng;
  • protopanaxadiol;
  • synergy


  1. Top of page
  2. Abstract
  6. Acknowledgements


20S-protopanaxadiol (aPPD) is a major gastrointestinal metabolic product of ginsenosides. The latter share structural similarity with steroids and are the main pharmacologically active component in ginseng.


The authors investigated the interaction between aPPD and estrogen receptors (ER) in human breast adenocarcinoma MCF-7 cells through receptor binding assay, ER-induced gene expression, and cell proliferation both in vitro and in vivo.


aPPD, but not its close analog ginsenosides, competed with the [3H]-17-β estradiol (E2) for ER with IC50 at 26.3 μM. aPPD alone weakly induced luciferase reporter-gene expression controlled by an estrogen-regulated element, which was completely blocked by tamoxifen. aPPD alone, or in synergy with tamoxifen, blocked E2-induced transcriptional activation. aPPD also inhibited colony formation of endometrial cancer cells. aPPD potently inhibited estrogen-stimulated MCF-7 cell proliferation and synergistically enhanced the cytotoxicity of tamoxifen on both ER+ MCF-7 and ER− MDA-MB231 cells. Furthermore, aPPD, but not tamoxifen, inhibited Akt phosphorylation. Growth of MCF-7 xenograft tumor supplemented with E2 was completely inhibited in animals treated with aPPD, tamoxifen, or aPPD plus tamoxifen.


These results suggested that aPPD inhibits estrogen-stimulated gene expression and cell proliferation in ER-positive breast cancer cells. In addition, aPPD synergistically enhances cytotoxicity of tamoxifen in an ER-independent fashion, probably by down-regulating Akt activity. Cancer 2007. © 2007 American Cancer Society.

Clinical experimental and epidemiological studies have demonstrated that high estrogen levels play a key role in breast cancer formation and progression.1–3 The antiestrogen agent tamoxifen is widely used as a therapeutic to treat breast cancer.4–6 The mechanism of tamoxifen-induced tumor inhibition has been demonstrated to be due to both estrogen-receptor mediation and tamoxifen cytotoxicity.7 However, the development of multidrug resistance in breast cancer cells and the potential of carcinogenic effects in uterine tissue have limited the efficacy of tamoxifen.4, 8–11 Discovery of potential alternatives to tamoxifen for the treatment of breast cancer is an important objective.

Ginsenosides are steroid saponins present in ginsengs. They possess a 4 trans-ring rigid steroid skeleton with modified side chains at C20. According to the number of hydroxyl groups on the structure, most ginsenosides can be classified into either the protopanaxatriol (PPT) or the protopanaxadiol (PPD) group.12 As the ginsenosides share a backbone similar to steroidal hormones (Fig. 1a, b), some of the ginsenosides have been shown to bind to the same target proteins as steroidal hormones, including estrogen receptors.13–16 It has been reported that ginsenoside Rg1 has estrogen-like activity, stimulates breast cancer cell growth, and activates an estrogen-regulated element transcription in HeLa cells in vitro, and its activity is blocked by estrogen antagonist ICI 182780.17 On the other hand, numerous studies have demonstrated that some ginsenosides possess anticancer activities that may not relate to estrogen receptors.18–20

thumbnail image

Figure 1. (Top) Chemical structures of ginsenosides, Rg3 and Rh2, and aglycones, aPPD and aPPT. (Bottom) Chemical structure of [3H]-17-β estradiol (E2).

Download figure to PowerPoint

Recent studies have demonstrated that most ginsenosides are metabolized to their aglycones in the gastrointestinal track as forms of 20S-protopanaxadiol (aglycone PPD or aPPD) and 20S-protropanaxatriol (aPPT).21, 22 Our previous studies demonstrated that aPPD induced apoptosis in various tumor cells involving both caspase-dependent and caspase-independent pathways.23 Extremely low toxicity of aPPD on normal tissues renders it a candidate for an anticancer drug. The present study goals were to determine whether aPPD and its close analogs can interact with estrogen receptors and to determine effects of aPPD on estrogen-stimulated tumor cell proliferation.


  1. Top of page
  2. Abstract
  6. Acknowledgements

Chemicals and Reagents

17-β estradiol and tamoxifen were purchased from Sigma (Sigma-Aldrich; St. Louis, Mo). [3H]-17-β estradiol (89 Ci/mmol) was purchased from Perkin Elmer (Waltham, Mass). DNA purification Kit was bought from Qiagen USA (Germantown, Md). Lipofectamin 2000 was purchased from Invitrogen Life Technology (Carlsbad, Calif). Luciferase assay system and β-glycosidase assay kit were purchased from Promega (Madison, Wis). Fetal calf serum and DMEM were bought from Invitrogen (Burlington, Ont, Canada). All remaining chemicals were purchased from Sigma. The 20S-protopanaxandiol (aPPD) was supplied by Pegasus Pharmaceuticals Inc., (Richmond, BC, Canada). The purity of the compound was 97.5% determined by high-performance liquid chromatography (HPLC). Plasmid EREII-Luc was a gift from Dr. Peter J. Kushner (University of California at San Francisco).

Whole Cell Binding Assay

Breast adenocarcinoma cell line MCF-7 was a gift from British Columbia Cancer Agency. The cells in phenol red-free Dulbecco Modified Eagle Media (DMEM) containing 5% charcoal-stripped fetal bovine serum (FBS) were seeded in a 24-well plate (80,000 cells/well). The whole cell-binding assay was carried out as described by Lee.13 Briefly, the cells were incubated in serum-free medium for 60 minutes with 10 nM [3H]-17-β estradiol and increasing concentrations of aPPD, aPPT, Rh2, and Rg3. The same medium with 0.2% ethanol was used as vehicle control. Then, the cells were washed with ice-cold phosphate-buffered saline (PBS) 3 times. The bound and intracellular [3H]-17-β estradiol were extracted with cold absolute ethanol, and the radioactivity was measured in a β-scintillation counter (LS6500 multipurpose counter; Beckman Coulter; Fullerton, Calif). To determine the total nonspecific binding, a 200-fold excess of unlabeled 17-β estradiol was added to the above binding medium. Binding data were analyzed using a Scatchard analysis. Data presented were averages of at least 3 independent experiments with triplicate measurements for each time point.

Reporter Gene Transfection and Luciferase Assays

A reporter-gene construct, pEREII, contains an estrogen response element in front of a luciferase reporter gene. MCF-7 cells were plated in each well of 24-well plates 1 day before transfection with the reporter-gene construct by using Lipofectamin2000. The cells were subjected to transfection in triplicate with 0.1 μg of EREII-Luc and 0.2 μg of a CMV-LacZ construct as an internal control. Four hours after transfection, the medium was replaced with 1 ml of DMEM medium with 5% charcoal-stripped FBS, 10 nM E2, and/or 2.5μM or 20 μM aPPD, and/or 0.5 μM or 10 μM tamoxifen. After 24 hours, the cells were washed with Mg2+-free and Ca2+-free PBS twice and ruptured with lysis buffer. The lysate was centrifuged at 14,000 rpm for 1 minute. The supernatant (20 μL) was mixed with 80 μL luciferase assay buffer. The chemiluminescence was measured for 1 second with a microplate luminometer (Berthold; Oak Ridge, Tenn). For the β-glycosidase assay, 50 μL of supernatant was mixed with 50 μL 2× assay buffer followed by incubation at 37°C for 30 minutes. The reaction was stopped by adding 500 μL of 1 M sodium carbonate, and the absorbance of the reaction was measured at 420 nm with μQuant BioTek (BioTek Instruments; Winooski, Vt). Relative luciferase activity was calculated by normalizing the absolute luciferase activity with that of β-glycosidase activity in each well.

Cell Viability Assay

MCF-7 or MDA-MB231 cells were grown in phenol red-free DMEM supplemented with charcoal-stripped 2% fetal bovin serum (FBS) for 5 days. Five thousand cells per well were plated in 96-well plates in the above medium. The next day, the culture medium was removed by aspiration and 200 μL of fresh medium containing 10 nM estradiol (E2), and 0, 1, 2.5, 5, 10 or 15 μM of tamoxifen with or without 10 μM aPPD was added. Estrogen-receptor negative MDA-MB231 (from the American Type Culture Collection; Manassas, Va) cells were treated similarly but without E2. After 2 days of treatment, the cell proliferation was determined by MTT assay. Briefly, the medium was removed and replaced with 50 μL of 0.5 mg/mL MTT. The cells were incubated in a CO2 incubator for 4 hours, and the purple crystal was solubilized with 100 μL of lysis buffer overnight. The optical density (O.D.) of each well was measured on a spectrophotometer at 570 nm with the μQuant BioTek.

Western Blot

MCF-7 cells cultured in 6-well plates were treated with 10 μM aPPD, 5 μM tamoxifen, or a combination of both. The cells were grown for 5.5 hours and washed once before lysis buffer was added. Protein samples (40–100 μg/lane) were separated on 10% sodium dodecyl sulfate polyacrylamide electrophoresis gels and transferred onto nitrocellulose membranes (Bio-Rad Laboratories; Hercules, Calif). The transferred membranes were probed by using anti-Phospho-Akt (Ser473) antibody (Cell Signaling Technology; Danvers, Mass). The membranes were regenerated and probed with anti–β-Actin antibody (Cell Signaling Technology) for normalization. The amount of proteins was measured by densitometry with Image Station 440 (Kodak; Rochester, NY) and analyzed with Kodak 1D3.6 to quantify the levels of phosphorylated Akt/PKB.

MCF-7 Xenograft Model in SCID Mice

Female severe combined immunodeficiency (SCID) mice were obtained from the British Columbia Cancer Agency at the age of 4 weeks. One day before xenografts were implanted, a 4× 1.98 mm Silastic (an inert silicone elastomer) E2-releasing pellet was subcutaneously implanted on the near axillary area of the mouse. Mice in the control group received a pellet containing no E2. Ten million MCF-7 cells were subcutaneously inoculated in mice on the rear axillary area. All procedures were approved by the animal care committee at the University of British Columbia.

The mice were treated with aPPD and tamoxifen daily for 31 days starting from the second day post-tumor inoculation. The animals were divided into 5 groups as follows: 1) E2+ vehicle (n = 4); 2) E2+ aPPD 10 mg/kg/day (n = 6); 3) E2+ aPPD 10 mg/kg/day plus tamoxifen 1.5 mg/kg/day (n = 7); 4) E2+ tamoxifen 1.5 mg/kg/day (n = 7); 5) E2-negative control (n = 4). Tamoxifen citrate was administered intraperitoneally whereas aPPD was administered orally. The mice were observed daily, and the size of each tumor was measured every other day with electronic calipers from the 10th day post-tumor implantation. Tumor volumes were calculated as a function of (length × width × height × 3.14) ÷ 6. An average tumor size per mouse was used to calculate the group mean tumor size ± standard error of the mean (SEM).

Measurement of Sera Estradiol Level

Serum E2 level in each mouse was determined by commercial ELISA kit (Biocheck; Foster City, Calif), according to the manufacturer's instructions, and absorbance at 450 nm was determined with a μQuant BioTek.


All in vitro experiments were conducted at least twice independently, with 3–4 repeats at each data point. Student t-test was performed for statistical analysis.


  1. Top of page
  2. Abstract
  6. Acknowledgements

aPPD was a Low-Affinity but a Specific ER Binder

Whole cell binding assays were carried out to test whether Rg3, Rh2, aPPD, and aPPT could bind to estrogen receptors (ERs). ER-positive MCF-7 cells, of which estrogen-receptor–α was predominant, were incubated in the presence of 10 nmol [3H]-17-β estradiol and increasing concentrations of unlabeled aPPD or other ginsenoside as competitors. As shown in Figure 2, aPPD inhibited the binding of [3H]-17-β estradiol to its receptors with an estimated IC50 of 26.3 ± 2.1 μM. As a comparison, [3H]-17-β estradiol has the IC50 of 70 ± 4.3 nM, suggesting that the affinity of aPPD to estrogen receptors was ≈370-fold less than estrogen itself. Interestingly, Rg3, Rh2, and aPPT did not bind to E2 receptors, even though all share similar structures to aPPD.

thumbnail image

Figure 2. Whole cell binding assay of ginsenosides to the estrogen receptors. MCF-7 cells were incubated with 10 nmol of [3H]-17-β estradiol, in the absence or presence of increasing concentrations of unlabeled aPPD, aPPT, Rh2, or Rg3 for 60 minutes. aPPD but not other ginsenosides competed with [3H]-E2 for the ERs in a dose-dependent manner. The IC50 of aPPD was 26.3, whereas [3H]-17-β estradiol had an IC50 of 70 nmol.

Download figure to PowerPoint

aPPD Weakly Stimulated ER-mediated Reporter-Gene Expression

To investigate whether the binding of aPPD to ER could induce the expression of ER-regulated genes, a construct with a luciferase gene downstream to a promoter-containing ER regulatory element was subjected to transfection into ER-positive MCF-7 cells. As illustrated in Figure 3a, aPPD (2.5 μM, 5 μM, 10 μM, respectively) activated the ER regulatory element luciferase gene transcription. Compared with background expression, 2.5 μM and 5 μM aPPD increased luciferase expression by 8.48-fold (±1.72) and 10.32-fold (±1.14) (P<.05), respectively. To confirm the specificity of aPPD on ER-mediated transactivation, we examined whether tamoxifen, a specific ER antagonist, could inhibit the effect of aPPD. As shown in Figure 3a, 5 μM of tamoxifen completely blocked aPPD-induced reporter-gene expression. Therefore, the aPPD-stimulated luciferase transcription was specifically mediated by ERs, which also supports that aPPD is able to bind to ER specifically. It may worth noting that the potency of aPPD for stimulating ER-mediated gene expression was approximately 10-fold less than estrogen itself.

thumbnail image

Figure 3. Effects of aPPD on ER regulatory element luciferase reporter-gene expression. Transfected MCF-7 cells were treated with E2, aPPD, tamoxifen, and their vehicle control. (A) aPPD weakly activated ER regulatory element luciferase reporter-gene expression, which was blocked by 5 μM tamoxifen; results are expressed as -fold of control. *P<.05, **P<.01. (B) Inhibitory effects of aPPD and tamoxifen on E2-induced ER regulatory element luciferase reporter-gene expression. Results were expressed as a percentage of E2-induced ER regulatory element luciferase activity. The IC50 for tamoxifen and aPPD were 2.8 μM and 12.5 μM, respectively; **P<.01, ***P<.001 versus E2-induced activity. (C) Synergistic effects of aPPD and tamoxifen (Tam) on E2-induced ER regulatory element luciferase gene expression. The results were expressed as a percentage of E2-induced ER regulatory element luciferase activity. **P<.01 for the E2 plus Tam group versus E2 plus aPPD plus Tam group.

Download figure to PowerPoint

aPPD Inhibited E2-Stimulated Gene Expression

To investigate whether aPPD inhibits E2-induced ER regulatory element activity, E2 stimulated luciferase activity was measured in MCF-7 cells in the presence of aPPD or tamoxifen. As shown in Figure 3b, both aPPD and tamoxifen inhibited E2-induced ER regulatory element luciferase transcription in a dose-dependent manner, with an IC50 of 12.5 μM and 2.8 μM, respectively. Figure 3c showed that 5 μM aPPD alone reduced E2-stimulated ER regulatory element luciferase transcription to 82.1% ± 2.3% (P<.01) of the control. Tamoxifen combined with 5 μM aPPD synergistically inhibited E2-stimulated gene expression (P<.01). Tamoxifen alone (0.5μM and 1 μM) had no effect on E2-induced reporter gene expression (102.5% ± 2.5% and 106.4% ± 9.3%, respectively, P>.1), but the same concentrations of tamoxifen together with 5 μM aPPD reduced reporter-gene expression to 63.4% ± 1.5% and 38.8% ± 1.6% of the control level (P<.01), respectively, which were significantly more than the sum of effects of aPPD and tamoxifen alone.

aPPD Inhibited Colony Formation of Endometrial Cancer Cells

Because tamoxifen has been shown to be an antagonist in breast cancer cells but an agonist in uterine tissue to stimulate endometrial cancer cell growth,12 it is important to know whether aPPD performs in the same manner. Colony-formation assay was performed by using MCF-7 and the Ishikawa endometrial cancer cell line. The cells were grown on soft agar (0.5% agar, 10% fetal bovine serum (FBS) in 1× DMEM) with either 10 nM of E2 or 5 μM of aPPD for 14 days. Colonies of >20 cells were counted. As shown in Figure 4, aPPD significantly inhibited colony formation of both MCF-7 and Ishikawa cells whereas E2 significantly stimulated it.

thumbnail image

Figure 4. Colony-formation assay for MCF-7 cells and Ishikawa cell line endometrial cancer cells. Ten thousand cells were seeded on 0.5% gel with 5 μM aPPD, 10 nM estrogen (E2), or medium control. The colonies were counted by Day 14. The data are shown as means of at least 4 individual plates from 2 repeated experiments. (*P<.02 vs control; **P<.002 vs control).

Download figure to PowerPoint

aPPD Inhibited E2-Stimulated MCF-7 Proliferation and Synergistically Enhanced Tamoxifen Cytotoxicity

To evaluate the effects of aPPD on ER-stimulated cell growth, the proliferation assay was performed on the MCF-7 human breast cancer cells. MCF-7 cells were treated for 48 hours with 10 nM E2 plus 0 μM to 15 μM tamoxifen in the presence or absence of 10 μM aPPD. E2 increased the MCF-7 cell growth by 17.0% ± 0.08% (P<.05) compared with a vehicle-treated control. As shown in Figure 5a, 10 μM aPPD as well as tamoxifen (1μM to 5 μM) completely abrogated the E2-stimulated cell growth (P<.05). Significant cytotoxicity of tamoxifen was observed at concentrations of 10 μM and 15 μM, resulting in 25.2% and 99.8% cell death even in the presence of 10 nM E2 compared with untreated control cell cultures (Fig. 5a). The cytotoxic effect of tamoxifen was greatly increased by 10 μM aPPD in a synergistic fashion as shown in Figure 5a and c. The synergy was further analyzed by the software CompuSyn (Biosoft, Cambridge, UK), and the results (Fig. 5c) confirmed that 10 μM of aPPD combined with 5 μM, 10 μM, and 15 μM of tamoxifen were synergistic.

thumbnail image

Figure 5. Synergistic cytotoxicity on breast cancer cells by aPPD and tamoxifen. (a) MCF-7 cells were treated with 10 nM E2 and 0μM to 15 μM of tamoxifen with or without 10 μM aPPD for 48 hours. Cell viability was measured with MTT assay, and data are shown as a percentage of vehicle-treated control cells. (b) Synergistic cytotoxicity on ER-negative breast adenocarcinoma MDA-MB231 cells by 5 μM aPPD and various concentrations of tamoxifen. (c) The synergistic effect of aPPD and tamoxifen on both cells was further analyzed with the software CompuSyn. CI, Combination Index (CI<0.1: very strong synergy; 0.1<CI<0.3: strong synergy; 0.3<CI<0.7: synergy; 0.7<CI<0.85: moderate synergy; 0.85<CI<1: slight synergy). Usually a value with CI<1 and Fa (fraction affected) between 0.5 and 1 is considered synergy. The tamoxifen concentrations were indicated. Each data point was an average of at least 2 experiments with at least triplicate measurement. Statistical analysis was performed between the tamoxifen-only group and the tamoxifen plus aPPD group. *P<.05, **P<.01, ***P<.001.

Download figure to PowerPoint

Synergistic Cytotoxicity Between aPPD and Tamoxifen was ER Independent

To ask whether the synergistic cytotoxicities of aPPD and tamoxifen were ER dependent, ER-negative MDA-MB231 cells were treated with aPPD and tamoxifen at various combinations of concentrations. MDA-MB231 cells were slightly more sensitive to the cytotoxicity induced by either tamoxifen or aPPD than MCF-7 cells (Fig. 5b). But more importantly, marked synergy between aPPD and tamoxifen was also evident. As shown in Figure 5b, when the concentration of tamoxifen was 0 μM (aPPD alone), treatment with 5 μM aPPD caused ≈ 20% cell death on MDA-MB231 cells. The combination of 2.5 μM tamoxifen (cell viability was 89.3 ± 1.8% applied alone), and 5μM aPPD had significant synergy to cause >95% cell death (cell viability was 4.7 ± 0.9%). CompuSyn analysis (Fig. 5c) further showed a significant synergy for 5 μM of aPPD in combination with <10 μM tamoxifen.

aPPD Inhibited Phosphorylation of Akt

In an effort to understand the mechanism of the synergy between aPPD and tamoxifen, we examined the levels of phosphorylated Akt/PKB in MCF-7 cells treated with aPPD, tamoxifen, and a combination of aPPD and tamoxifen, because activation of Akt by phosphorylation has been reported to play a role in tamoxifen resistance in MCF-7cells.24, 25 As shown in Figure 6, levels of phosphorylated Akt were significantly reduced by aPPD (10 μM) or by aPPD (10 μM) with tamoxifen (5 μM) but not by tamoxifen alone.

thumbnail image

Figure 6. aPPD reduced levels of phosphorylated AKT in MCF-7 cells. Levels of total AKT and phosphorylated AKT in MCF-7 cells were measured, and the values were normalized with densities of actin for each sample. The data were averages of measurements from 4 independent experiments.

Download figure to PowerPoint

aPPD Inhibited ER-Dependent Tumor Growth in a Breast Cancer Animal Model

To confirm the inhibitory effect of aPPD on ER-dependent tumor growth, aPPD was orally administered to animals bearing MCF-7 tumors with a subcutaneously implanted Silastic E2 pellet (Fig. 7). Whereas the mean E2 plasma levels in animals of E2-supplemented control and treatment groups varied from 445 pg/mL to 758 pg/mL, there were no significant differences among those groups (Table 1). In addition, correlation coefficient analysis of samples from the E2+ control group indicated no correlation between the E2 level and tumor size (r = 0.15, P>.05), suggesting that the E2 concentrations in E2-supplemented animals were sufficient to support tumor growth. The tumor became measurable 1 week after implantation, and the measurement started on Day 10 postimplantation. As shown in Table 2, the tumor volume was >70-fold greater in E2-supplemented animals than that in the non–E2-supplemented control group at Day 31 postimplantation. In the non-E2 supplemented control group, a measurable transient tumor growth was observed at Day 10 postimplantation. However, the tumor regressed afterward, from an average volume of 0.42 (±0.25) mm3 to 0.13 (±0.08) mm3 by Day 29, indicating that the intrinsic E2 level was not sufficient to support the progression of xenografts. In animals with estradiol support but treated orally with aPPD, the initial tumor volume (10 days postimplantation) was 0.05 (±0.00) mm3, which was 13% of that of the non–E2-supplemented group. Tumors in the aPPD-treated animal group decreased to immeasurable size by Day 15. Tamoxifen (3 mg/kg, intraperitoneally administered) also effectively inhibited tumor growth (7 of 7 mice, P<.01.). Animals treated with the combination of 10 mg/kg aPPD plus 1.5 mg/kg tamoxifen also had a complete inhibition of tumor growth. By Day 31, there were no significant differences among any of the treatment groups and the non–E2-supplemented group in tumor sizes. It is worthwhile to notice that there was a significant difference in tumor sizes on the first day of measurement (Day 10 postimplantation) among the treatment groups. Mean tumor volume of the tamoxifen-only group is significantly larger than the aPPD group or the aPPD plus tamoxifen combined treatment group (P<.01), which may reflect a difference in potency of these treatments (Table 2). There was no toxicity in animals, even as high as 60 mg/kg/day in 1 of our preliminary experiments (data not shown), suggesting that aPPD is a very safe compound in vivo.

thumbnail image

Figure 7. Effect of aPPD on breast cancer growth in mice xenografts. Data are shown as -fold changes of volume (mean ± standard error [SEM]).

Download figure to PowerPoint

Table 1. Plasma E2 Levels in MCF-7 Tumor-bearing Mice
Group10 mg/kg aPPD3 mg/kg Tam10 mg/kg aPPD 1.5mg+1.5mg/kg TamControl E2+Control E2−
  1. Average plasma indicates E2 levels for each group of animals implanted with MCF-7 tumor cells. No statistical difference was found among the groups with E2 supplement.

  2. aPPD indicates 20S-protopanaxadiol; Tam, Tamoxifen; E2+, 17-β estradiol positive; E2−, 17-β estradiol negative; SEM, standard error of the mean.

Average of E2, pg/mL524.03757.61591.54445.043.62
Table 2. Comparison of Tumor Volumes Among Groups (mean± SEM mm3)
 Control E2+Control E2−Tam 3mg/kgaPPD10 mg/kgaPPD/Tam (10mg/1.5mg)/kg
  • Comparison of tumor indicates volume among the 5 treatment groups (Control E2+, Control E2−, 10 μg/kg aPPD, 10 mg/kg aPPD+ 1.5 mg/kg Tamoxifen, and 3 mg/kg Tamoxifen only). Results are expressed as mean volume± SEM.

  • aPPD indicates 20S-protopanaxadiol; Tam, Tamoxifen; E2+, 17-β estradiol positive; E2−, 17-β estradiol negative; SEM, standard error of the mean.

  • *

    P<.01 Tam only group compared with aPPD only and aPPD + Tam group.

  • P<.01 aPPD only group compared with aPPD + Tam group.

Day100.30± 0.120.42± 0.250.10± 0.01*0.05± 0.000.02± 0.01
Day130.75± 0.110.42± 0.250.11± 0.080.02± 0.01Immeasurable
Day150.75± 0.110.25± 0.140.03± 0.02Immeasurable


  1. Top of page
  2. Abstract
  6. Acknowledgements

Because ginseng is 1 of the most widely used medicinal herbs in the world, it is important to understand the bioactivity of its major metabolite, such as aPPD. The present study is the first to show that 1) aPPD can bind to estrogen receptors with a low affinity but not to its close structural analogs Rh2, Rg3, and aPPT; 2) aPPD weakly, but specifically, activates the estrogen receptor, as demonstrated by its ability to stimulate ER regulatory element-mediated reporter-gene expression; 3) aPPD competes with estradiol for E2 receptors and blocks estradiol's ability to stimulate ER regulatory element-mediated gene expression; 4) aPPD blocks E2-stimulated breast cancer cell proliferation both in vitro and in vivo; 5) aPPD synergistically enhances the potency of tamoxifen as an antagonist of E2-stimulated gene expression and as a cytotoxic agent (The latter is ER-independent.); 6) aPPD, but not tamoxifen, reduced levels of phosphorylated Akt at a concentration where a synergy was manifested.

The present results suggest that aPPD is a partial agonist and antagonist for estrogen receptors. Compounds with similar characters have been known as selective estrogen-receptor modulators (SERMs).9, 10 SERMs bind to estrogen receptors and have tissue-specific effects that allow them to function as estrogen agonists in some tissues and as estrogen antagonists in other tissues. These compounds include clomiphene citrate, tamoxifen, toremifene, and raloxifene.26, 27 The molecular mechanisms of SERM activity are believed to be due to conformational modifications of the estrogen-receptor protein after binding with different ligands, altering the receptor affinity to other proteins associated with its physiological activity.28 Tamoxifen has been widely used in prevention and treatment of breast cancer because of its estrogen antagonistic activity. However, tamoxifen also has an estrogen-like effect on lipids, bone, and the endometrium.29, 30 The agonistic effect on endometrium may relate to tamoxifen's reported contribution to the initiation and progression of endometrial cancer.11 Although it is still not clear whether aPPD is a true SERM at the present time, here we have shown that aPPD significantly inhibited colony formation of endometrial cancer cells and, therefore, will be unlikely to stimulate tumor growth in uterine tissue. The estrogen-related activity of aPPD remains to be investigated in cells of other tissue origins, and its conformational effect on estrogen-receptor protein after binding needs to be understood.

Another important finding from the present study is that aPPD can synergistically act with tamoxifen to either block estrogen activity on gene regulation or abolish estrogen-stimulated cell proliferation. Tamoxifen has a known cytotoxicity, possibly related to activation of caspases, elevation of intracellular calcium level, and inhibition of the insulin-like growth-factor pathway.7, 31 Tamoxifen is also known for its chemosensitizing effect when used in combination with other cytotoxic agents.32, 33 aPPD can also induce apoptosis in various cancer cells through caspase-dependent and caspase-independent pathways.23 The synergistic cytotoxicity of tamoxifen and aPPD in breast cancer cells occurred at concentrations of 5 μM and 10 μM for each compound, respectively. The synergistic cytotoxicity apparently is independent of ER, as strong synergy was also observed in ER-negative cells MDA-MB231. Conversely, the synergy between aPPD and tamoxifen was also demonstrated in their combined inhibition of estrogen-stimulated gene regulation. As shown in Figure 3c, ER regulatory element-regulated gene expression in the presence of E2 was significantly suppressed to a much greater extent when cells were treated with aPPD and tamoxifen in combination rather than with each compound alone. In this case, there was no cytotoxicity shown by cell-viability assay (data not shown). Thus, we conclude that the synergy between aPPD and tamoxifen occurred as both antiestrogenic and cytotoxic effects on the breast adenocarcinoma cells.

Even though the mechanism of the synergy between aPPD and tamoxifen remains to be further studied, our present results show that inhibition of Akt activity by aPPD is likely 1 of the reasons that cause aPPD-induced enhancement of tamoxifen cytotoxicity. As a key kinase in the cell survival pathway, Akt has been found to play a crucial role in development of resistance to tamoxifen in breast cancer cells, including MCF-7. It has been found that overexpression of phosphorylated Akt was associated with tamoxifen-resistant and ICI182,780-resistant cell lines,34 and expression of constitutively active Akt in MCF-7 cells reverses estrogen and tamoxifen responsiveness.24, 25 It has been further shown that inhibition of mTOR activity, a downstream element in the Akt pathway, restores tamoxifen response in breast cancer cells with aberrant Akt activity.35 Thus, it is reasonable to speculate that aPPD-caused inhibition in Akt activity sensitizes MCF-7 cells to tamoxifen-induced cytotoxicity, which is a possible mechanism underlying the synergy between aPPD and tamoxifen.

Our animal breast tumor model data showed that orally taken aPPD at 10mg/kg/day completely inhibited the growth of estrogen-supplemented breast cancer and was as effective as 3 mg/kg intraperitoneally administered tamoxifen. Although pharmacokinetics study is needed to verify drug concen-trations in the blood and tissues, our present results suggest that oral aPPD at this dose can reach therapeutic levels in vivo. This is in agreement with several previous studies that have demonstrated that intestinal bacterial metabolites of ginseng or ginsenosides were easily absorbed and appeared in plasma of rats or humans.22, 36–38 The synergistic effects between tamoxifen and aPPD may also contribute to the powerful inhibition in estrogen-dependent breast cancer growth in the animal model. As shown in Table 2, tumor sizes on Day 10 postimplantation in tamoxifen or aPPD groups were significantly larger (P<.01, E2 and aPPD; or P<.05, E2 and tamoxifen) than the group who received both tamoxifen and aPPD, suggesting that the combination treatment was more effective in suppressing estrogen-dependent tumor growth.

Our present study demonstrates that aPPD merits further study for its potential as a new drug for prevention and treatment of estrogen-regulated neoplasms. Its synergy with tamoxifen also deserves further investigation for potential clinical application in tamoxifen-resistant breast cancer treatment.


  1. Top of page
  2. Abstract
  6. Acknowledgements

We thank Drs. Valentine and Kushner at UCSF for the ER-related constructs and Pegasus Pharmaceuticals for the ginsenosides used in the present study. We appreciate Dr. Y. Z. Wang at British Columbia Cancer Research Centre for his assistance in mouse MCF-7 breast cancer model. We also thank Ms. Cathy Campbell for preparation of this manuscript


  1. Top of page
  2. Abstract
  6. Acknowledgements
  • 1
    Clemons M, Goss P. Estrogen and the Risk of Breast Cancer. N Engl J Med. 2001; 344: 276285.
  • 2
    Key TJ, Chen J, Wang DY, Pike MC, Boreham J. Sex hormones in women in rural China and in Britain. Br J Cancer. 1990; 62: 631636.
  • 3
    Hankinson S, Willett W, Manson J, et al. Plasma sex steroid hormone levels and risk of breast cancer in postmenopausal women. J Natl Cancer Inst. 1998; 90: 12921299.
  • 4
    Osborne CK. Tamoxifen in the treatment of breast cancer. N Engl J Med. 1998; 339: 16091618.
  • 5
    Borgna JL, Rochefort H. [Effects and mechanism of action of antiestrogens in breast cancer]. Sem Hop. 1984; 60: 703709.
  • 6
    Gail MH, Costantino JP, Bryant J, et al. Weighing the Risks and Benefits of Tamoxifen Treatment for Preventing Breast Cancer. J Natl Cancer Inst. 1999; 91: 18291846.
  • 7
    Salami S, Karami-Tehrani F. Biochemical studies of apoptosis induced by tamoxifen in estrogen receptor positive and negative breast cancer cell lines. Clin Biochem. 2003; 36: 247253.
  • 8
    Fornander T, Rutqvist LE. Adjuvant tamoxifen and second cancers. Lancet 1989; 1(8638): 616.
  • 9
    Grese TA, Dodge JA. Selective estrogen receptor modulators (SERMs). Curr Pharm Des. 1998; 4: 7192.
  • 10
    Haynes B, Dowsett M. Clinical pharmacology of selective estrogen receptor modulators. Drugs Aging. 1999; 14: 323336.
  • 11
    Jordan VC, Gottardis MM, Satyaswaroop PG. Tamoxifen-stimulated growth of human endometrial carcinoma. Ann N Y Acad Sci. 1991; 622: 439446.
  • 12
    Saga Y, Ohwada M, Kohno T, Takayashiki N, Suzuki M. High-grade endometrial stromal sarcoma after treatment with tamoxifen in a patient treated for breast cancer. Int J Gynecol Cancer. 2003; 13: 690692.
  • 13
    Lee YN, Lee HY, Chung HY, et al. In vitro induction of differentiation by ginsenosides in F9 teratocarcinoma cells. Eur J Cancer. 1996; 32: 14201428.
  • 14
    Lee YN, Lee HY, Lee YM, et al. Involvement of glucocorticoid receptor in the induction of differentiation by ginsenosides in F9 teratocarcinoma cells. J Steroid Biochem Mol Biol. 1998; 67: 105111.
  • 15
    Lee Y, Jin Y, Lim W, et al. A ginsenoside-Rh1, a component of ginseng saponin, activates estrogen receptor in human breast carcinoma MCF-7 cells. J Steroid Biochem Mol Biol. 2003; 84: 463468.
  • 16
    Gray SL, Lackey BR, Tate PL, Riley MB, Camper ND. Mycotoxins in root extracts of American and Asian ginseng bind estrogen receptors alpha and beta. Exp Biol Med (Maywood). 2004; 229: 560568.
  • 17
    Chan RY, Chen WF, Dong A, Guo D, Wong MS. Estrogen-like activity of ginsenoside Rg1 derived from Panax notoginseng. J Clin Endocrinol Metab. 2002; 87: 36913695.
  • 18
    Duda RB, Kang SS, Archer SY, Meng S, Hodin RA. American ginseng transcriptionally activates p21 mRNA in breast cancer cell lines. J Korean Med Sci. 2001; 16( suppl): S54S60.
  • 19
    Oh M, Choi YH, Choi S, et al. Anti-proliferating effects of ginsenoside Rh2 on MCF-7 human breast cancer cells. Int J Oncol. 1999; 14: 869875.
  • 20
    Chen D, Zhao Y, Bai S, Shi Z, Zhang J. [Effect of ginsenoside Rg3 on the progression of orthotopically xenotransplanted human breast cancer in nude mice and its mechanism]. Sichuan Da Xue Xue Bao Yi Xue Ban. 2003; 34: 546548.
  • 21
    Bae EA, Han MJ, Choo MK, Park SY, Kim DH. Metabolism of 20(S)- and 20(R)-ginsenoside Rg3 by human intestinal bacteria and its relation to in vitro biological activities. Biol Pharm Bull. 2002; 25: 5863.
  • 22
    Tawab MA, Bahr U, Karas M, Wurglics M, Schubert-Zsilavecz M. Degradation of ginsenosides in humans after oral administration. Drug Metab Dispos. 2003; 31(8): 10651071.
  • 23
    Jia W, Yan H, Bu X, Liu G, Zhao Y. Aglycone protopanaxadiol, a ginseng saponin inhibits P-glycoprotein and sensitizes chemotherapy drugs on multidrug resistant cancer cells [abstract]. J Clin Oncol. 2004 ASCO Meeting Proceedings (Post-Meeting Edition). 2004; 22:(Jul 15 suppl). Abstract 9663.
  • 24
    Faridi J, Wang L, Endemann G, Roth RA. Expression of constitutively active Akt-3 in MCF-7 breast cancer cells reverses the estrogen and tamoxifen responsivity of these cells in vivo. Clin Cancer Res. 2003; 9: 29332939.
  • 25
    Jordan NJ, Gee JM, Barrow D, Wakeling AE, Nicholson RI. Increased constitutive activity of PKB/Akt in tamoxifen resistant breast cancer MCF-7 cells. Breast Cancer Res Treat. 2004; 87: 167180.
  • 26
    Snyder KR, Sparano N, Malinowski JM. Raloxifene hydrochloride. Am J Health Syst Pharm. 2000; 57: 16691675; quiz 76–78.
  • 27
    Szucs J, Pavo I, Jurida N. [Clinical experience with raloxifene]. Orv Hetil. 2000; 141: 237240.
  • 28
    Hubbard RE, Pike ACW, Brzozowski AM, et al. Structural insights into the mechanisms of agonism and antagonism in oestrogen receptor isoforms. Eur J Cancer. 2000; 36( suppl 4): 1718.
  • 29
    Goldstein SR. Selective estrogen receptor modulators: a new category of compounds to extend postmenopausal women's health. Int J Fertil Womens Med. 1999; 44: 221226.
  • 30
    Kristensen B, Ejlertsen B, Dalgaard P, et al. Tamoxifen and bone metabolism in postmenopausal low-risk breast cancer patients: a randomized study. J Clin Oncol. 1994; 12: 992997.
  • 31
    Kanter-Lewensohn L, Girnita L, Girnita A, et al. Tamoxifen-induced cell death in malignant melanoma cells: possible involvement of the insulin-like growth factor-1 (IGF-1) pathway. Mol Cell Endocrinol. 2000; 165: 131137.
  • 32
    Koh J, Kubota T, Koyama T, et al. Combined antitumor activity of 7-hydroxystaurosporine (UCN-01) and tamoxifen against human breast carcinoma in vitro and in vivo. Breast Cancer. 2003; 10: 260267.
  • 33
    Ramachandran C, Khatib Z, Petkarou A, et al. Tamoxifen modulation of etoposide cytotoxicity involves inhibition of protein kinase C activity and insulin-like growth factor II expression in brain tumor cells. J Neurooncol. 2004; 67: 1928.
  • 34
    Frogne T, Jepsen JS, Larsen SS, Fog CK, Brockdorff BL, Lykkesfeldt AE. Antiestrogen-resistant human breast cancer cells require activated protein kinase B/Akt for growth. Endocr Relat Cancer. 2005; 12: 599614.
  • 35
    deGraffenried LA, Friedrichs WE, Russell DH, et al. Inhibition of mTOR activity restores tamoxifen response in breast cancer cells with aberrant Akt Activity. Clin Cancer Res. 2004; 10(23): 80598067.
  • 36
    Akao T, Kida H, Kanaoka M, Hattori M, Kobashi K. Intestinal bacterial hydrolysis is required for the appearance of compound K in rat plasma after oral administration of ginsenoside Rb1 from Panax ginseng. J Pharm Pharmacol. 1998; 50: 11551160.
  • 37
    Bae EA, Park SY, Kim DH. Constitutive beta-glucosidases hydrolyzing ginsenoside Rb1 and Rb2 from human intestinal bacteria. Biol Pharm Bull. 2000; 23: 14811485.
  • 38
    Hasegawa H, Suzuki R, Nagaoka T, Tezuka Y, Kadota S, Saiki I. Prevention of growth and metastasis of murine melanoma through enhanced natural-killer cytotoxicity by fatty acid-conjugate of protopanaxatriol. Biol Pharm Bull. 2002; 25: 861866.