AP and PMA were purchased from Sigma-Aldrich (St. Louis, MO, USA). Minimum Essential Medium-α (α-MEM), fetal bovine serum (FBS) and penicillin were obtained from Gibco-BRL (Gaithersburg, MD, USA). The Cell Counting kit (CCK)-8 assay was purchased from Dojindo Molecular Technology (Tokyo, Japan). Primary antibodies (monoclonal rabbit antibody; species reactivity, human) for β-actin, phospho-IκBα, IκBα and MMP-9 were purchased from Cell Signaling Technology, Inc. (Beverly, MA, USA). The Luciferase Assay system was from Promega (Sydney, Australia). Tris, glycine, NaCl, SDS, and other reagents were from Sigma-Aldrich. The Vybrant^® Apoptosis Assay kit #2 was from Invitrogen (Carlsbad, CA, USA).

MDA-MB-231 cells were cultured in L-15 Medium (Gibco Life Technologies, Beijing, China) with 10% FBS and maintained in a humidified atmosphere of 5% CO₂ at 37°C. The complete medium was changed every other day. The cells were treated with increasing concentrations of AP (0, 7.5, 15, 30, 60 or 120 μM) for two days prior to the cell viability assays. The anti-proliferative effect of AP on MDA-MB-231 cells was assessed using CCK-8. Briefly, following treatment, 10 μl CCK-8 solution was added to each well and incubated for 4 h. The absorbance was measured at a wavelength of 450 nm using a ELX800 absorbance microplate reader (BioTek Instruments, Inc., Winooski, VT, USA) at a wavelength of 450 nm (reference, 650 nm). The effect of AP on cell viability was expressed as a percentage of cell viability, with the vehicle-treated control cells set as 100%.

AP induction of apoptosis in MBA-MD-231 cells was determined with the Vybrant^® Apoptosis Assay kit #2. Following treatment, cells were washed twice with cold phosphate-buffered saline (PBS) and resuspended in 1X Annexin-binding buffer. Early apoptosis was detected via staining with Alexa Fluor^® 488 Annexin V and propidium iodide. Fluorescence-activated cell sorting was performed using a FACScan™ flow cytometer and data were acquired using CellQuest software, version 3.0 (BD Biosciences, Sunnyvale, CA, USA).

Transwell^® Permeable Supports (Corning Inc., Acton, MA, USA), 24-well chambers with 8-μm pore polycarbonate filters, were used as described by the manufacturer. MDA-MB-231 cells (5×10⁴) were placed in 100 μl serum-free medium in the presence or absence of AP and 600 μl complete medium with 80 nM PMA was placed into the lower wells. Following treatment, cells were fixed with 100% methanol for 20 min and stained with Trypan blue for 30 min. Non-migrating cells on the upper side of the filter were removed with cotton swabs. Migration was quantified by counting the number of cells on the lower surface of the filter.

BioCoat™ Matrigel™ Invasion Chamber (BD Biosciences), 24-well chambers with 8-μm pore polycarbonate filters, were used according to the manufacturer’s instructions. MDA-MB-231 cells (5×10⁴) were placed in 100 μl serum-free medium in the presence or absence of AP, and 600 μl complete medium with 80 nM PMA was placed in the lower wells. Following treatment, cells on the upper side of the filters were removed. Invading cells on the underside of the filter were fixed with 100% methanol for 2 min and stained with Liu’s stain for 2 min. Invasion was quantified by counting the number of cells on the lower surface of the filter.

BALB/c nu/nu mice (Harlan, Indianapolis, IN, USA) were housed in individual cages, maintained in an animal facility under controlled temperature (22–24°C) and humidity (50–60%) conditions and a 12 h light/dark cycle with free access to food and water. Cultured MDA-MB-231 cells were resuspended in PBS at a density of 5×10⁶ cells/ml (26,27). An aliquot (10 μl) of the cell suspension was slowly injected through the anterior tuberosity of the proximal tibia in the right limbs of 5- to 6-week-old female BALB/c nu/nu mice (Harlan, Indianapolis, IN, USA). The mice were randomly assigned to vehicle (0.9% NaCl, n=8) or AP (50 mg/kg body weight vehicle, n=8) groups and treated via an intraperitoneal injection every other day. After 28 days, a bioluminescence assay was performed and fluorescence intensity was quantified (Living Image v3.2, Caliper; Caliper Life Sciences, Hopkinton, MA, USA). Radiographs using the Directview Vita CR system. (Carestream Kodak, Rochester, NY, USA) of the tibiae were obtained prior to euthanasia with ketamine, administered by intraperitoneal injection (0.8 ml/100 g body weight). The product from Carestream Kodak was. Tissues were removed and fixed in 4% paraformaldehyde for 1 day at 4°C followed by decalcification in 12% EDTA. Decalcified bones were paraffin-embedded and sectioned. Samples were subjected to tartrate-resistant acid phosphatase (TRAP) staining to identify osteoclasts on the bone surface. Immunostaining for Ki67 (Dako, Carpinteria, CA, USA) and terminal deoxynucleotidyl transferase-mediated dUTP nick-end labeling (TUNEL) were performed as previously described (28,29). Ki67- and TUNEL-positive tumor cells were counted and the percentages of positive cells were calculated. This study was approved by the ethics committee of Shanghai Ninth People’s Hospital Affiliated to Shanghai Jiao Tong University School of Medicine (Shanghai, China).

RNA isolation was performed as previously described (30). Total RNA was extracted using the Qiagen RNeasy Mini kit (Qiagen, Valencia, CA, USA) following the manufacturer’s instructions. cDNA was synthesized from 1 mg of total RNA using reverse transcriptase (TaKaRa Biotechnology, Otsu, Japan). MMP-9 transcript expression levels were determined using the MiniOpticon Real-Time PCR system (Bio-Rad Laboratories, Hercules, CA, USA). qPCR was performed in a thermocycler (Biometra, T-Gradient Thermoblock, Germany) with a reaction volume of 10 μl containing 0.03 μg complementary DNA product, 2 μM forward and reverse primers and the KAPA™ SYBR^® FAST qPCR reagent (Kapa Biosystems, Wilmington, MA, USA). The primers used were as follows: Forward, 5′-GAACCAATCTCACCGACAGG-3′, and reverse, 5′-GCCACCCGAGTGTAACCATA-3′ for MMP-9; and forward, 5′-TCTGCTGGAAGGTGGACAGT-3′, and reverse, 5′-CCTCTATGCCAACACAGTGC-3′ for β-actin. Cycling conditions were as follows: 40 cycles of 95°C for 5 sec and 60°C for 34 sec. β-actin was included as a reference control. The comparative 2^−ΔΔCt method was used to calculate the relative expression of each gene (30).

The effect of AP on PMA-induced NF-κB activation was measured in MDA-MB-231 cells stably transfected with an NF-κB luciferase reporter construct (13). MDA-MB-231 cells were maintained in serum-free medium for 12 h, pretreated with AP for 1 h, followed by stimulation with PMA for 20 h. Subsequently, the cell lysis was incubated with substrate (Promega, Madison, WI, USA) at room temperature for about 2min, luciferase activity was measured using the Promega Luciferase Assay System (Promega, Madison, WI, USA). Luciferase activity was measured and normalized to the internal control. Results were obtained from three independent experiments.

Western blotting was performed as previously described (30). The vehicle- or AP-treated cells were pretreated with PMA, washed twice in PBS and lysed in ice-cold lysis buffer (50 mM Tris pH 7.5, 150 mM NaCl, 1% Nonidet P-40, 0.1% SDS, 1% sodium deoxycholate) supplemented with phenylmethanesulfonyl fluoride (Shen Neng Bo Cai Corp., Shanghai, China). Lysates were maintained on ice for 30 min followed by centrifugation at 12,000 × g for 10 min. Protein concentrations were determined using a bicinchoninic acid (BCA) assay (Thermo Scientific, Rockford, IL, USA). Equal amounts of protein were separated by 10% SDS-PAGE and electroblotted onto polyvinylidene fluoride membranes (Roche, Mannheim, Germany). The membranes were blocked with 5% (w/v) skim milk solution for 1 h and probed with primary antibodies (β-actin, 1:1,000; phospho-IκBa, 1:1,000; IκBa, 1:1,000; and MMP-9, 1:1,000) at room temperature for 4 h, followed by incubation with horseradish peroxidase-conjugated secondary antibodies (anti-human; Cell Signaling Technology, Inc.; 1:5,000) for 1 h. Antibody reactivity was visualized using an Odyssey^® Infrared Imaging system (Li-Cor, Lincoln, NE, USA).

Significant differences were determined with the Student’s t-test using SPSS v13.0 software (SPSS Inc., Chicago, IL, USA). P<0.05 was considered to indicate a statistically significant difference.

Following a 48-h culture, a CCK-8 proliferation assay revealed that AP did not affect MDA-MB-231 cell proliferation at concentrations ≤30 μM (Fig. 1A). AP significantly suppressed cell proliferation at concentrations ≥60 μM. The calculated IC₅₀ for AP is 77.87 μM (Fig. 1B). In cells treated with 10 or 30 μM AP, the observed apoptotic effects were similar to those of the vehicle control; however, the higher concentration of 60 μM AP induced apoptosis in 22% of cells (Fig. 1C and D). In order to exclude AP-mediated apoptosis, non-lethal concentrations (≤30 μM) were used in subsequent experiments.

PMA (80 nM) induced increased levels of MDA-MB-231 cell migration and invasion compared with those observed in the untreated cells; however, pretreatment with AP inhibited the PMA-induced migration and invasion in a concentration-dependent manner (Fig. 2A). Quantitative analysis confirmed AP inhibition of cell migration and invasion at concentrations as low as 10 μM (Fig. 2B and C).

To determine the effects of AP on breast cancer bone metastasis and cancer cell-induced osteolysis in vivo, a mouse xenotransplant model was used with human breast cancer cells (luciferase-labeled MDA-MB-231) (26,31). MDA-MB-231 cells were injected directly into the tibiae plateau via a percutaneous approach. After 28 days, bioluminescence was detected in the limbs of the control mice; however, the area and density of bioluminescence were reduced in the AP group compared with those in the control group (Fig. 3A), indicating that AP effectively suppressed breast cancer bone metastasis and growth in vivo. These observations were consistent with the results of the tumor volume assay (Fig. 3A). To confirm that osteolytic bone metastasis was blocked by AP, the osteolysis in the long bones of the hind legs was examined using radiography. AP significantly inhibited cancer cell-induced osteolysis (represented by radiolucency; Fig. 3A). TRAP staining (red) revealed numerous osteoclasts with intense activity in the vehicle-treated controls, however, in contrast, the number of osteoclasts was markedly reduced at the boundary in the treated mice (Fig. 3A), indicating that AP suppressed tumor-related osteolysis by inhibiting osteoclasts in vivo. All the results were confirmed using quantitative analysis (Fig. 3B). The proliferation-indicator Ki67 assay and the apoptosis-indicator TUNEL assay were also performed. Treatment of MDA-MB-231 tumor cells with AP (50 mg/kg) suppressed cellular proliferation compared with that in the control cells (Fig. 3C). The percentage of Ki67-positive cell nuclei was 7.1% in the AP-treated group and 32.4% in the vehicle-treated group (Fig. 3D). The levels of apoptosis were significantly increased in the AP-treated group of MDA-MB-231 cell-associated breast tumors compared with those of the vehicle-treated group in the TUNEL assay (Fig. 3C and D). All in vivo data were consistent with the in vitro data, demonstrating that AP inhibits MDA-MB-231 cancer cell invasion and migration and suppresses tumor-induced osteolysis, possibly via inhibited osteoclast activity.

MMP-9 mediates tumor invasion and migration. In the current study, AP reduced the levels of MMP-9 secretion into the medium compared with those observed in the control cells (Fig. 4A). Consistent with the aforementioned findings, treatment of MDA-MB-231 cells with AP reduced PMA-stimulated MMP-9 protein expression in a concentration-dependent manner (Fig. 4B). qPCR revealed that PMA-induced MMP-9 mRNA expression decreased with AP treatment, indicating that AP-mediated inhibition of MMP-9 occurs at the transcriptional level (Fig. 4C).

MMP-9 is highly inducible in response to various stimuli and the MMP-9 promoter contains a binding site for transcription factor NF-κB (12), hence it can be used to detect NF-κB signaling. Measurement of NF-κB-dependent luciferase activity in MDA-MB-231 cells revealed that PMA-induced NF-κB transcriptional activity was suppressed by AP (Fig. 4D). NF-κB is normally sequestered in the cytoplasm in an inactive form associated with NF-κ-B inhibitor α (IκBα). Upon stimulation, the NF-κB subunit is released via the phosphorylation and proteasomal degradation of IκBa and translocated to the nucleus to initiate target gene transcription (32,33). AP was observed to prevent the PMA-induced degradation of IκBa (Fig. 4E). As degradation of IκBα is primarily the result of IκBα phosphorylation (33), it was hypothesized that this effect may be due to the AP-induced inhibition of IκBα phosphorylation. In the present study, AP caused a concentration-dependent reduction in PMA-induced IκBα phosphorylation (Fig. 4E). These results indicate that the inhibitory effect of AP on NF-κB signaling occurs via the inhibition of IκBa phosphorylation, which in turn suppresses transcriptional activity. NF-κB signaling activates MMP-9 transcription; thus, these results indicate that AP attenuates MMP-9 expression by inhibiting NF-κB signaling.