Pomegranate extract inhibits androgen-independent prostate cancer growth through a nuclear factor-kappaB-dependent mechanism

Abstract

Constitutive nuclear factor-kappaB (NF-kappaB) activation is observed in androgen-independent prostate cancer and represents a predictor for biochemical recurrence after radical prostatectomy. Dietary agents such as pomegranate extract (PE) have received increasing attention as potential agents to prevent the onset or progression of many malignancies, including prostate cancer. Here, we show that PE inhibited NF-kappaB and cell viability of prostate cancer cell lines in a dose-dependent fashion in vitro. Importantly, maximal PE-induced apoptosis was dependent on PE-mediated NF-kappaB blockade. In the LAPC4 xenograft model, PE delayed the emergence of LAPC4 androgen-independent xenografts in castrated mice through an inhibition of proliferation and induction of apoptosis. Moreover, the observed increase in NF-kappaB activity during the transition from androgen dependence to androgen independence in the LAPC4 xenograft model was abrogated by PE. Our study represents the first description of PE as a promising dietary agent for the prevention of the emergence of androgen independence that is driven in part by heightened NF-kappaB activity.

Figure 1

PE inhibits constitutive and TNF-α-induced NF-κB activation. A, inhibition of constitutive NF-κB-driven reporter gene expression. Cells were plated in a 24-well format in triplicate and transiently transfected with a κB-luc reporter as well as the pRL-SV40 reporter for normalization of transfection efficiency. Cells were exposed to the indicated dilutions of PE for 24 h before harvesting of protein. Mean ± SD. B, EMSA showing PE-mediated inhibition of constitutive and TNF-α-induced (10 ng/mL × 30 min) NF-κB in DU145 cells. Top, NF-κB EMSA; bottom, Oct-1 EMSA. Lanes at far right show cold competition experiments with cold wild-type (wt) and mutant (mut) probes. Cells were exposed to indicated dilutions of PE for 4 h. C, same as B but in CL1 cells. D, top, Western blotting shows effects of PE on total IκBα levels in DU145 and LAPC4 cells after exposure to TNF-α; bottom, Western blot for actin as a protein loading control.

Figure 2

Dose-dependent in vitro cytotoxicity of PE. A, DU145 cells were exposed to either PE or concentrated PJ at the indicated dilutions for 48 h and overall cell viability was measured by the methylene blue assay. Mean ± SD of four wells and were normalized to results of vehicle (PBS)-treated cells. B, photomicrographs of DU145 cells treated with the indicated dilutions of PE. Final magnification, ×200.

Figure 3

Effects of PE on apoptosis and cell cycle profile of DU145 cells. Cells were exposed to PE for 48 h and then assayed for apoptosis by Annexin V staining (A) and cell cycle profile by hypotonic propidium iodide staining (B). Representative flow cytometric results. Numeric results are depicted in table form for both Annexin and hypotonic propidium iodide staining.

Figure 4

PE-induced apoptosis is dependent on the NF-κB-inhibitory effect of PE. A, DU145 cells (0.8 × 10⁵ per well in 24-well format) were transiently transfected with PSA-P/E-luc vector (1 μg) and the pEGFP-p65 and pCMV-p50 expression vectors (0.5 μg each). Twenty-four hours after transfection, PE (or PBS control) was added at the indicated dilutions for an additional 24 h before harvesting protein for reporter gene expression. Results were normalized to that of Renilla luciferase (pRL-SV40 cotransfected at 1 ng/well). Mean ± SD of three experiments. B, DU145 cells were plated in 10 cm dishes (1.5 × 10⁶ per dish) and transfected with pEGFP-p65 and pCMV-p50 vectors (2.5 μg each). The next day, PE was added at the indicated dilutions and cells were harvested 24 h later. EGFP-positive cells were analyzed for apoptosis using a TUNEL assay as described in Materials and Methods. C, effects of PE on prostate cancer cell lines that lack constitutive NF-κB activation. Cells were exposed to the indicated dilutions of PE for 48 h and relative cell viability was determined. Mean ± SD of four wells and were normalized to results of vehicle (PBS)-treated cells.

Figure 5

Effects of PE on growth of androgen-independent mouse xenografts. After s.c. LAPC4 xenografts became palpable, mice were fed either PE or vehicle (veh) control (n = 9 for both groups) for 1 wk before castration followed by an additional 2 wk of dietary supplementation before euthanasia. A, tumor volumes in PE and vehicle-treated castrated mice. The mean tumor volume of the PE-treated group was statistically significantly smaller than that of the vehicle-treated group (P = 0.05, two-tailed Student’s t test). B, serum PSA measurements in PE- and vehicle-treated animals at the time of sacrifice. C and D, proliferation is inhibited and apoptosis is induced in tumors in PE-treated compared with vehicle-treated mice. At the time of euthanasia, tumors were formalin-fixed and paraffin-embedded and subjected to either staining for the Ki-67 proliferation marker or for an apoptosis marker with the use of a TUNEL assay. C, mean of the percentage of cells that stain for Ki-67 and are positive for TUNEL staining based on counting of three representative high-power (×400) fields. D, representative photomicrographs of Ki-67 and TUNEL staining. Final magnification, ×200.

Figure 6

PE inhibits NF-κB in androgen-independent LAPC4 xenografts. A, EMSA for NF-κB were done on nuclear protein extracted from frozen tumors taken at the time of euthanasia. Top, EMSA for NF-κB; middle, EMSA for Oct-1 as a control; bottom, Western blot for lamin B, a nuclear envelope protein, to show the integrity of our nuclear extracts and equality of protein loading. B, top, Western blot for phospho-IκBα; bottom, Western blot for actin as a protein loading control.