PPD was kindly provided by Professor Ping Li of China Pharmaceutical University (Nanjing, China) with a purity >95% as confirmed by HPLC (4,15). PPD was dissolved in dimethyl sulfoxide (DMSO) (15 mM stock solution). For in vivo treatment, PPD was dissolved in PEG. Unless otherwise indicated, all chemicals were obtained from Fisher Scientific (Pittsburgh, PA, USA) or Sigma-Aldrich (St. Louis, MO, USA).

Human colorectal cancer lines (HCT116 and SW480), breast cancer cell lines (MDA-MB-468 and MDA-MB-231), prostate cancer cell lines (PC3 and DU145), osteosarcoma cell lines (MG63 and 143B) and HEK-293 cells were purchased from the American Type Culture Collection (ATCC, Manassas, VA, USA) and grown in Dulbecco’s modified Eagle’s medium (DMEM) (Invitrogen Life Technologies, Carlsbad, CA, USA) supplemented with 10% fetal bovine serum (FBS; HyClone Laboratories, Logan, UT, USA) and 50 units penicillin/streptomycin in 5% CO₂ at 37°C.

A modified MTT assay was used to examine the cell growth inhibitory effect of ginsenosides on cell proliferation as previously described (16). Cells were seeded in 96-well plates (1×10⁴ cells/well, 50–70% density). Ginsenosides were added to the cells at various concentrations and incubation was carried out for 48 h. Fifteen microliters of dye solution was added to each well and incubated for an additional 4 h. One hundred microliters/well solubilization/stop solution was added to stop the reaction and to solubilize the formazan crystals in a humidified atmosphere overnight. Absorbance at 570 nm was determined using a 96-well microplate reader.

HCT116 cells were treated with the indicated concentrations of drugs. At the endpoints, the cell culture medium was carefully removed. The wells were gently washed with phosphate-buffered saline (PBS) at room temperature. The medium was aspirated and cells were stained with 0.5% crystal violet formalin solution at room temperature for 20–30 min. After staining, the cells were washed with tap water and air-dried at room temperature (17,18).

Flow cytometry was carried out to quantitatively detect the cell cycle distribution (19). Cells were plated into 6-well plates for drug treatments. At 24, 48 and 72 h post treatment, cells were harvested, washed with PBS, fixed in cold methanol overnight at 4°C and stained with 50 ng/ml propidium iodide (PI) by incubation at 4°C for 15 min. The stained cells were analyzed by flow cytometry.

Total RNA was isolated using TRIzol reagents and used to generate cDNA templates by RT reaction with hexamer and SuperScript^® II RT (both from Invitrogen Life Technologies). The first strand cDNA products were further diluted 10-fold and used as PCR templates. Semi-quantitative RT-PCR was carried out as described (20). Briefly, PCR primers were designed using the Primer3 program to amplify the human genes of interest (product sizes 150–180 bp) as follows: GAPDH forward, 5′-CAACGAATTTGGCTACAGCA-3 and reverse, 5′-AGGGGAGATTCAGTGTGGTG-3′; PITPNA forward, 5′-CGTCCTACCCCCATGTTG-3′ and reverse, 5′-ACTGGGCAGCGTCTGTTC-3′; and AKAP8L forward, 5′-GCAGGCAGGCAAGAAGAG-3′ and reverse, 5′-TGGCCATCTCGTCCTCAT-3′. A touchdown cycling program was carried out as follows: 94°C for 2 min for 1 cycle, 92°C for 20 sec, 68°C for 30 sec, and 72°C for 12 cycles with a decrease of 18°C per cycle and then at 92°C for 20 sec, 57°C for 30 sec, and 72°C for 20 sec for 20 to 25 cycles depending on the abundance of a given gene. The specificity of PCR products was confirmed by resolving PCR products on 1.5% agarose gels. All samples were normalized with the internal control GAPDH.

The HCT116-Luc cell line, which stably expresses firefly luciferase, was generated as previously described (19,21). Animal use and care were carried out according to the protocol guidelines approved by the Institutional Animal Care and Use Committee. Athymic nude mice (female, 4–6 weeks old, ~20 g body weight, n=5/group; Harlan SD, Indianapolis, IN, USA) were used. HCT116-Luc cells were harvested and resuspended in PBS to a final density of 2×10⁷ cells/ml. Cells (1×10⁶) were injected subcutaneously into the flanks of the mice. At 1 week post injection, PPD was administered (30 mg/kg) through an i.p. injection once every 2 days for 3 week.

For whole body bioluminescence imaging, animals were anesthetized with isoflurane attached to a nose-cone mask within the Xenogen IVIS 200 imaging system after implantion for 1 week. Mice were injected (i.p.) with D-Luciferin sodium salt (Gold BioTechnology, St. Louis, MO, USA) at 100 mg/kg in 0.1 ml sterile PBS. The pseudo-images were obtained by superimposing the emitted light over the grayscale images of the animal. Quantitative analysis was carried out with Xenogen’s Living Image V2.50.1 software as described (13,19). Animals were sacrificed after 3 weeks, and tumor samples were retrieved for histologic examination.

Retrieved tumor tissues were fixed in 10% formalin and embedded in paraffin. Serial sections of the embedded specimens were stained with hematoxylin and eosin. Paraffin-embedded sections were deparaffinized and then rehydrated in a graduated manner. The deparaffinized samples were subjected to antigen retrieval and fixation. Slides were blocked and probed with an antiproliferating cell nuclear antigen (PCNA) antibody (Santa Cruz Biotechnology, Inc., Santa Cruz, CA, USA), followed by incubation with the anti-mouse IgG-biotin secondary antibody. Finally, sections were incubated with HRP-streptavidin and visualized by 3,3′-diaminobenzidine staining (22).

Promoters responsive to the following signaling pathways, including MAPK/ERK, MAPK/JNK, Wnt, Notch, cell cycle/pRb-E2F, NF-κB, Myc/Max, hypoxia (namely Elk-1/SRF, AP-1, TCF/LEF, RBP-Jκ, E2F/DP1, NF-κB and hypoxia-inducible factor-1) were cloned into our homemade pBGLuc retroviral vector. All subcloned promoter fragments were verified by DNA sequencing.

Stable HCT116-GLuc reporter lines were established using the retroviral transduction approach as previously described (14,19,21). Gaussia luciferase activity was determined using the Gaussia luciferase assay kit (New England Biolabs). Briefly, HCT116-GLuc cell lines were seeded in 24-well culture plates and treated with 0 or 10 μM PPD. After 24 h, cell culture medium was subjected to Gaussia luciferase assay. Each assay condition was conducted in triplicate.

The in vitro experiments were performed in triplicate. Data are expressed as the means ± standard error (SE). Statistical significances between vehicle treatment vs. drug-treatment were determined by one-way ANOVA and the Student’s t-test. A value of p<0.05 was considered to indicate a statistically significant result.

The effect of PPD and ginsenoside Rg3 on the proliferation of HCT116 cells was evaluated by MTT assay. The IC₅₀ of PPD was significantly lower than that of Rg3 in the HCT116 cells (Fig. 1A). At 10–30 μM, PPD exhibited strong anti-proliferation effects after 48 and 72 h of treatment (Fig. 1B). We also investigated whether PPD inhibits the viability of other human cancer cell lines. Treatment of different types of cancer cells with different dosages of PPD for 48 h significantly suppressed the cell proliferation of the tested cancer cell lines: human colon cancer (HCT116 and SW480), breast cancer (MDA-MB-468 and MDA-MB-231), prostate cancer (PC3 and DU145) and osteosarcoma cell lines (MG63 and 143B) (data not shown). The IC₅₀ value for PPD in these cancer cell lines was 4.69 μM for HCT116, 8.99 μM for SW480, 7.64 μM for MDA-MB-468, 4.49 μM for MDA-MB-231, 1.40 μM for PC3, 4.71 μM for DU145, 5.17 μM for MG63, and 8.36 μM for 143B, respectively. Crystal violet staining assay revealed that 10 μM PPD had anti-proliferation effects on the HCT116 cells (cell viability <30%) although PPD was less effective on the SW480 cells (Fig. 1C), which was consistent with the differences in their IC₅₀ values.

In order to better understand the mechanism behind PPD-mediated inhibition of cell proliferation, we analyzed the distribution of PPD-treated cancer cells in different phases of the cell cycle by flow cytometry following treatment of cells with different concentrations of PPD for 48 h. We found that PPD caused a dose-dependent cell accumulation in the G1/S phase (Fig. 2). For example, treatment of HCT116 cells with 10 μM PPD led to an 86 to 91% increase in cells in the G1+S phase (Fig. 2A vs. D), leading to fewer cells progressing to the G2 phase. These results indicate that cell cycle progression was significantly blocked in the G1/S phase when cells were treated with PPD.

We further tested the antitumor activity of PPD in a xenograft tumor model of human colon cancer. The firefly luciferase-tagged HCT116 cells were subcutaneously injected in mice to form xenograft tumors. At one week, the tumor-bearing athymic nude mice were i.p. administered PPD at 30 mg/kg once every 2 days for up to 3 weeks. The tumor growth was monitored by using whole body Xenogen imaging (Fig. 3A). PPD was shown to significantly inhibit xenograft tumor growth at 2 weeks after treatment (Fig. 3B). At the endpoint (3 months after treatment), the xenograft tumors were retrieved and were shown to be smaller in the PPD treatment group (Fig. 3C). Histologically, the PPD-treated xenograft tumors exhibited significant necrosis (Fig. 3D-a vs. -b). Immunohistochemical staining with a PCNA antibody revealed that xenograft tumor cells treated with PPD exhibited a marked decrease in cell proliferation (Fig. 3D-a vs. -b). Thus, the in vivo results suggest that PPD may be developed into an efficacious anticancer agent.

We previously found that the expression levels of AKAP8L and PITPNA were significantly altered following treatment with S2h (American ginseng extract) or ginsenoside Rg3 in HCT116 cells (13). As one of the main metabolites of S2h and the aglycon of ginsenoside Rg3, PPD was shown to affect the expression levels of AKAP8L and PITPNA in the PPD-treated HCT116 cells (Fig. 4A). Specifically, PPD treatment slightly upregulated AKAP8L expression while significantly inhibiting PITPNA expression in a time-dependent fashion. These results suggest that upregulation of AKAP8L and/or downregulation of PITPNA may play an important role in mediating the anticancer activities conferred by ginsenoside derivatives, such as PPD. However, the exact mechanism behind their roles in the anticancer action of PPD needs to be thoroughly investigated.

We sought to further investigate the mechanistic basis underlying the anticancer activity of PPD. Although the above results indicate that Rg3 targets AKAP8L and PITPNA may play an important role in the mode of anticancer action of PPD, it is known that cancer development usually hijacks multiple cellular signaling pathways (23,24). Thus, we tested the effect of PPD on 8 major cancer-related signaling pathways, including MAPK/ERK, MAPK/JNK, Wnt, Notch, cell cycle/pRb-E2F, NF-κB, Myc/Max and hypoxia. When HCT116 cells containing the pathway-speciifc reporters were treated with 10 μM PPD, we found that the relative reporter activities for NF-κB, MAPK/JNK and MAPK/ERK pathways were significantly inihibited (Fig. 4B). The other 5 pathways, noticeably Myc/Max and Wnt, were not significantly affected by PPD in HCT116 cells. Thus, these results suggest that PPD may exert its anticancer activity at least in part by targeting the ERK, JNK and/or NF-κB signaling pathways although the exact mechanism needs to be fully elucidated.