Antibodies were purchased as follows: anti-rabbit cyclooxygenase-2 (Cox-2), survivin, Bcl-2, Bax, β-actin (Santa Cruz Biotechnology, Santa Cruz, CA, USA), cleaved-polyADP ribose polymerase (PARP), Ki-67 and p65 antibody (Cell Signaling Technology, Danvers, MA, USA). Coix seed emulsion (Kanglaite^® injection) was obtained from Zhejiang Kanglaite Pharmaceutical Co., Ltd. (Hangzhou, China). Gemcitabine-HCl was purchased from Eli Lilly and Company (Indianapolis, IN, USA). All other chemical reagents were purchased from Sigma-Aldrich (St. Louis, MO, USA).

The human PC cell lines PANC-1, BxPC-3 and AspC-1, and the human embryonic kidney cell line HEK-293T were obtained from the Type Cell Collection of the Chinese Academy of Sciences (Shanghai, China). PANC-1, BxPC-3 and AspC-1 cells were cultured in RPMI-1640 medium, and HEK-293T in Dulbecco's modified Eagle's medium (DMEM) (Gibco, Grand Island, NY, USA) in a 5% CO₂ and 95% air humidified atmosphere at 37°C. All media were supplemented with 10% fetal bovine serum (Gibco) and 100 U/ml of penicillin-streptomycin.

Male nude BALB/c mice (6 week of age) were obtained from the Zhejiang Chinese Medical University Experimental Animal Center (Hangzhou, China), and housed under pathogen-free conditions and were provided with standard laboratory food and water. All animal-handling protocols were approved by the Experimental Animal Use and Management Committee of Zhejiang Chinese Medical University. To induce tumor growth, the right flank of each mouse was subcutaneously injected with 5×10⁶ PC BxPC-3 cells suspended in 0.1 ml phosphate-buffered saline (PBS). When subcutaneous tumors developed to ~30 mm³ in size (7 days after inoculation), the mice (n=24) were randomly divided into four groups: a control group (0.9% saline, i.p. daily), a coix seed emulsion alone group (12.5 ml/kg, i.p. daily), a gemcitabine alone group (50 mg/kg, i.p. 3/week) and a coix seed emulsion and gemcitabine combination group (coix seed emulsion 12.5 ml/kg, i.p. daily; gemcitabine 50 mg/kg, i.p. 3/week). Body weights and tumor volumes were measured every three days. Tumor volumes were calculated with the formula: Tumor volume (mm³) = maximal length × maximal width²/2. After 24 days of treatment, all mice were overdosed with anesthesia, and tumors were harvested and weighed. The rate of inhibition of tumor growth was calculated using the formula: [(Mean tumor weight of the control group - mean tumor weight of the treatment group)/mean tumor weight of the control group] × 100%. Tumors tissue were fixed in buffered formalin for further analyses.

Cells were seeded in 96-well microplates at a density of 4×10⁴ cells/ml and cultured overnight; they were exposed to various concentrations of coix seed emulsion and gemcitabine, alone or in combination for the desired time. Control cells received only dimethyl sulfoxide (DMSO). Cell viability was assessed using the 3-(4,5-dimethylthiazol-2yl)-5- (3-carboxymethoxypheny l)-(4-sulfophenyl)-2H-tetrazolium (MTS) assay (Promega, Madison, WI, USA) according to the manufacturer's instructions. Cell viability is directly proportional to the absorbance at 490 nm of a formazan product that is bio-reduced from MTS in living cells as previously described (23). To determine the synergetic effect between coix seed emulsion and gemcitabine, the cells were exposed to drugs in a fixed ratio and the combination index (CI) was calculated by CalcuSyn software (Biosoft, Cambridge, UK). All experiments were carried out in hexaplicate, and were independently repeated at least twice.

Apoptosis of cells after the treatment of coix seed emulsion and gemcitabine, alone or in combination, was evaluated with the use of the Caspase-3/CPP32 Fluorometric Assay kit (BioVision Research Products, Mountain View, CA, USA; according to the manufacturer's instructions), as previously described (23). Cellular lysates (50 μl) were extracted and protein concentration was measured using BCA protein assay reagents (Pierce, Rockford, IL, USA). Cleavage of DEVD-AFC, a substrate of caspase-3, was quantified using a fluorescence microplate reader, with 400 nm excitation and 505 nm emission filters.

Cells were seeded in 100-mm tissue culture dishes and grown to 80% confluency. Medium was then replaced with fresh medium containing selected drugs at the various concentrations. The control group was treated with the same volume of DMSO. After 48 h, all cells were washed twice with ice-cold PBS, collected in RIPA lysis buffer (20 mM Tris-HCl, 150 mM NaCl, 1% NP-40, 5 mM EDTA, 1 mM Na₃VO₄, pH 7.5) supplemented with a protease inhibitor mixture (Sigma-Aldrich, St. Louis, MO, USA), and incubated on ice for 20 min. Afterwards, the cell lysate was centrifuged at 14,000 rpm for 10 min, and the pellet was diluted in SDS-sample buffer. Protein concentration was determined using BCA protein reagents (Pierce). The nuclear and cytoplasmic proteins were extracted with the Nuclear and Cytoplasmic Protein Extraction kit (Pierce; according to the manufacturer's instructions). The cell lysate (30 μg) was subjected to electrophoresis on a 4–12% sodium dodecyl sulfate (SDS)-polyacrylamide gel, and then transferred onto a polyvinylidene difluoride (PVDF) membrane (Bio-Rad, Hercules, CA, USA). Blots were probed with appropriate primary antibodies and IRDye^® 800CW secondary antibody (LI-COR Biotechnology, Lincoln, NE, USA). Bands were visualized by Odyssey infrared imaging system (LI-COR Biotechnology) and quantified by densitometry analysis, using ImageJ software (NIH, Bethesda, MA, USA).

HEK-293T cells (at 70% confluency) were transfected with the NF-κB-EGFP construct (GeneChem, Shanghai, China), which links an enhanced GFP reporter to an NF-κB response element as previously described (16). To prepare DNA for transfection, 10 μl Lipofectamine 2000 and 2 μg DNA were diluted with 0.2 ml of Opti-MEM (both from Invitrogen, Carlsbad, CA, USA), according to the manufacturer's instructions. After incubation for 24 h, the transfection rate was tested (data not shown), and HEK-293T cells for which the transfection rate was >90% were treated with the desired concentration of coix seed emulsion alone, with gemcitabine alone, with the coix seed emulsion and gemcitabine combination or with DMSO. After an additional 48 h, the expression of EGFP in each group was captured on a fluorescence microscope system (Olympus, Tokyo, Japan); the cells (1×10⁶ cells/ml) were then washed twice with cold PBS, half of them were re-suspended in fixing buffer and analyzed by flow cytometry (Millipore, Billerica, MA, USA) within 1 h. The other half of the cells were lysed and the protein concentration was quantified. Data are reported as the average of three independent experiments.

The electrophoretic mobility shift assay (EMSA) was used to assess NF-κB activation as previously described (24). Briefly, nuclear extracts prepared from cells treated with nuclear and cytoplasmic extraction reagents (Pierce) were incubated with biotin-labeled double-stranded NF-κB oligonucleotide (Beyotime, Beijing, China) for 30 min at 37°C, and EMSA was performed, following the instructions in the LightShift Chemiluminescent EMSA kit (Pierce).

Formalin-fixed, paraffin-embedded tumor tissues were sectioned (4-μm) and incubated with antibodies against Ki-67 and P65, at 1:100 dilutions. All stained slides were examined by an independent observer; and the tissue was assessed for protein expression in neoplastic areas at a magnification of ×40. Percentage of Ki-67 and P65 expression was recorded for each area (cytosolic or nuclear), and then averaged for each mouse. A total of five fields were examined in six samples from each of the treatment groups.

All data were represented as mean ± standard deviation (SD) for at least three independent experiments; representative examples are shown. Statistical analysis of multiple group comparisons was performed by one-way analysis of variance (ANOVA). Comparisons between two groups were analyzed using Student's t-tests. A P-value of <0.05 was considered to indicate a statistically significant result.

To investigate the cytotoxicity of coix seed emulsion on PC cell lines in vitro, we exposed BxPC-3, PANC-1 and AsPC-1 cells for 72 h to coix seed emulsion at increasing concentrations (0–10.0 mg/ml), and examined the cell viability by the MTS assay. As shown in Fig. 1A, coix seed emulsion exhibited dose-dependent cytotoxicity in the BxPC-3, PANC-1 and AsPC-1 cells with an IC₅₀ value of 1.50, 1.75 and 1.80 mg/ml, respectively.

To further study whether emulsion treatment decreases cell viability through apoptosis, we measured caspase-3 activity in all three cell lines and found that, compared to untreated cells, caspase-3 activity decreased from 1.84 to 4.33-fold in the treated PC cells, in a dose-dependent manner (Fig. 1B). To clarify the molecular mechanism of coix seed emulsion-induced apoptosis, we determined apoptosis-related protein expression in the BxPC-3, Panc-1 and AsPC-1 cells, and reported that expression of the anti-apoptotic protein Bcl-2 was downregulated, whereas that of the pro-apoptotic protein Bax was upregulated, also in a dose-dependent manner; these findings indicate that the apoptosis-inducing effect of coix seed emulsion can be partly attributed to an altered Bax/Bcl-2 protein ratio, which is a critical determinant of apoptosis (21). The caspase-3 downstream substrate cleaved-PARP was also increased in the emulsion-treated PC cells. Our results confirm previous studies on the use of coix seed emulsion with breast (25) and liver cancer cells (15). Even though coix seed emulsion does not inhibit the viability of cultured cancer cells as much as do other small molecular inhibitors, it can alter apoptosis signaling in PC cells.

Gemcitabine is the only first-line drug for patients with advanced PC; however it only provides a modest improvement in survival (26), in part due to the development of drug resistance. NF-κB regulation is involved in the formation of such resistance (27,28). Since coix seed emulsion inhibits NF-κB activity in MDA-MB-231 breast cancer cells, we next assessed whether coix seed emulsion enhances the cytotoxicity of gemcitabine in PC cells. We found that pretreatment of coix seed emulsion was more efficacious in sensitizing the cells to gemcitabine compared to co-treatment of the two drugs or whether the emulsion was washed away before adding drugs (data not shown). To confirm synergism between the emulsion and gemcitabine, we pretreated BxPC-3, PANC-1 and AsPC-1 cells with coix seed emulsion alone (0–5.0 mg/ml) for 48 h, following by co-treatment with varying concentrations of gemcitabine (0–0.75 μg/ml) for another 24 h. Our results showed that pretreatment with coix seed emulsion effectively decreased the IC₅₀ value of gemcitabine from 0.18–0.39 to 0.066–0.099 μg/ml (Fig. 2A). We then determined CI values for the combination treatment group. CI is a quantitative measure of the degree of drug interaction; CI <1 indicates synergism, CI >1 indicates antagonism and CI=1 indicates an additive effect. As shown in Fig. 2B, BxPC-3, PANC-1 and AsPC-1 cells pretreated with coix seed emulsion showed synergistic loss of cell viability, when the pretreatment was combined with gemcitabine (CI=0.333, 0.765 and 0.282, respectively).

To confirm whether the enhanced loss of cell viability resulting from pretreatment with coix seed emulsion is mediated by apoptosis, the cells were pretreated (4.0 mg/ml) with emulsion, followed by gemcitabine. This significantly elicited gemcitabine-induced caspase-3 activity from 1.89- to 2.43-fold, to 7.52- to 8.61-fold (P<0.05) in all PC cell lines used (Fig. 3A), suggesting that the loss of viable cells by coix seed emulsion and gemcitabine was due to the induction of the cell death pathway. In agreement with these results, we also found significant upregulation of Bax and downregulation of Bcl-2 protein expression in the combination treatment group (Fig. 3B), and that the combination treatment significantly suppressed other pro-survival molecules such as survivin and Cox-2. Our results indicate that coix seed emulsion and gemcitabine indeed synergized the cytotoxic effect of gemcitabine on PC cells.

Since Bcl-2, survivin and Cox-2 proteins are downstream effectors regulated by NF-κB (29,30), we next assessed the possible mechanistic role of NF-κB in our experimental system. Employing an NF-κB-dependent enhanced green fluorescent protein (EGFP) reporter assay, we directly measured the effects of the drugs on a gene promoter element that has a consensus NF-κB response element in HEK-293T cells. Gemcitabine (0.18 μg/ml) treatment alone stimulated the reporter activity; meanwhile coix seed emulsion (4.0 mg/ml) effectively diminished gemcitabine-induced or non-induced reporter activity (Fig. 4A and B). Activation of NF-κB is associated with translocation to the nucleus of the p65/Rel-A subunit of this transcription factor, as well as with the NF-κB DNA-binding activity (31). PANC-1 cells were exposed to coix seed emulsion (4.0 mg/ml) for 72 h, followed by 3 h of gemcitabine (0.50 μg/ml) treatment; the nuclear and cytoplasmic fractions of the cell lysate were subjected to electrophoresis and immunoblotting by an anti-p65 antibody. Meanwhile, an electrophoretic mobility shift assay was used to assess the NF-κB DNA-binding activity in the nuclear extracts. Fig. 4C shows that gemcitabine stimulated the translocation of p65 from the cytoplasm to the nucleus, consistent with activation of the NF-κB transcription factor; this effect was markedly attenuated by treating the cells with coix seed emulsion. These observations were further supported by the finding that coix seed emulsion sequestered the basal level of NF-κB DNA-binding activity in unstimulated PC cells, and gemcitabine induced NF-κB DNA-binding activity (Fig. 4D). Our results indicate that the chemosensitizing effect of coix seed emulsion is, in part, due to inactivation of NF-κB and its downstream genes.

Based on the in vitro finding that pretreatment of coix seed emulsion sensitizes gemcitabine-induced cytotoxicity in PC cells, we evaluated the therapeutic advantage of using coix seed emulsion and gemcitabine in a xenograft nude mouse model implanted with PANC-1 cells (see experimental protocol in Fig. 5A). Relative to the untreated control group, administration of coix seed emulsion by intraperitoneal injection caused a 30% reduction in tumor weight (P<0.05), while gemcitabine treatment alone caused a 50% reduction in tumor volume and weight (P<0.05) (Fig. 5B and C). However, the combination of coix seed emulsion and gemcitabine demonstrated a 68% reduction in tumor volume and weight, a result of which was superior to treatment with coix seed emulsion alone (P<0.05) or gemcitabine alone (P<0.05) (Fig. 5B and C). Moreover, coix seed emulsion partly rescued the severe weight loss induced in the mice by gemcitabine (Fig. 5D). We additionally investigated whether the coix seed emulsion alone, or in combination with gemcitabine, downregulated NF-κB in the tumor tissue. Immunohistochemistry revealed significant downregulation of p65 expression in groups treated with coix seed emulsion alone (P<0.05) and with the combination (P<0.05), compared with the control; only modest alterations were noted in the gemcitabine alone group (Fig. 5E). These findings paralleled the study on a significant reduction in Ki-67 staining in tumors derived from the mice treated with the combination of the two drugs (P<0.05), indicating diminished cellular viability in the tumors.

Taken together, our in vivo results are consistent with our in vitro findings and collectively provide convincing evidence in support of the superior antitumor efficacy of the combination of coix seed emulsion and gemcitabine.