All the compounds (in DMSO) shown in Table I were provided by the Cooperative Research Project I of the Institute of Natural Medicine, University of Toyama, Japan. Antibody against MMP-9 or β-actin was obtained from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Mouse TNF-α was purchased from Jena Bioscience GmbH (Jena, Germany). Hirsutine was purchased from Wako Pure Chemical Industries, Ltd. (Osaka, Japan). pGL4.50 [luc2P/CMV-RE/Hygro] and pGL4.32 [luc2P/NF-κB-RE/Hygro] vector and D-Luciferin were obtained from Promega (Sunnyvale, CA, USA). Lipofectamine 2000 was purchased from Invitrogen (Carlsbad, CA, USA). Hygromycin B was obtained from Nacalai Tesque (Kyoto, Japan).

Mouse mammary carcinoma 4T1 cells were maintained in RPMI-1640 medium (Nissui, Tokyo, Japan) supplemented with 10% bovine serum. MDA-MB-231 and MDA-MB-468 human breast cancer cell lines were maintained in Dulbecco’s modified Eagle’s medium (DMEM; Nissui) supplemented with 10% bovine serum. To establish the luciferase gene expressing 4T1 cells (4T1-luc or 4T1-NFκB-luc), and 4T1 cells (5×10⁵/well) were seeded in a 6-well plate and pGL4.50 or pGL4.32 vector was transfected using Lipofectamine 2000. The cells were selected with Hygromycin B (100 μg/ml) and cloned by limiting dilution.

For the detection of luciferase activity, cells were stimulated with or without TNF-α (10 ng/ml) in the screening experiments. Briefly, for screening experiments without stimulation, 4T1-NFκB-luc cells in exponential growth were placed at a final concentration of 2×10⁴ cells/well in a 96-well plate. After 3-h incubation, the cells were treated with compounds or with the vehicle (vehicle control, 0.5% DMSO) for 24 h. Then 20 μl of luciferin (900 μg/ml) was added, and the plates were incubated for another 30 min. Luciferase activity was measured by the Glomax multi detection system (Promega). For the experiments with TNF-α (10 ng/ml) stimulation, 4T1-NFκB-luc cells (2×10⁴ cells/well) were placed in a 96-well plate for 12 h. After incubation, the cells were treated with compounds and TNF-α (10 ng/ml) for 12 h. Then luciferin was added and luciferase activity was measured.

Viability of cells was assessed using a WST-1 Cell Counting kit (Wako Pure Chemical Industries). The experimental conditions for cell viability were similar to the previous luciferase reporter assay. Twenty-four hours after treatment with compounds, 10 μl WST-1 reagent was added and incubated for another 2 h (37°C, 5% CO₂). The absorbance at 450 nm was measured using a microplate reader.

4T1-NFκB-luc cells were plated in a 24-well plate at a concentration of (1×10⁵ cells/well) and allowed to form a confluent monolayer for 24 h. The monolayer was then scratched with a sterile pipette tip (1,000 μl), washed with medium to remove floated and detached cells, and photographed (time 0 h). Cells were successively treated in medium in the presence of different concentrations of Hirsutine (12.5 and 25 μM) along with the vehicle DMSO for 24 h. Scratched areas were photographed (magnification, ×40) at 0 h and then again 24 h later to assess the degree of wound healing. The percentage of wound closure was estimated by the following equation: wound closure % = 1 − (wound area at t₂₄/wound area at t₀) × 100%, where t₂₄ is the time after wounding and t₀ is the time immediately after wounding.

The filters of a Transwell cell culture insert (8 μm pore size; Whatman Japan KK, Tokyo, Japan) were pre-coated with fibronectin (Iwaki, Tokyo, Japan, 1.25 μg/filter) on the lower surfaces. For the hapto-invasion assay, the upper surface of the filters was coated with Matrigel (Becton-Dickinson, Bedford, MA, 1 μg/filter). Cells were pre-incubated with or without Hirsutine (25 μM) for 24 h. After trypsinization, cells in 0.1% (v/v) BSA medium (1.5×10⁴) were placed in the upper chamber of Transwells. After the subsequent incubation at 37°C, the residual cells were removed from the upper surface of the membrane and the migrated cells on the lower surface of the membrane were fixed in 100% methanol and stained with hematoxylin and eosin. Migration was determined by counting the cells with a microscope at ×100 magnification. Five visual fields were chosen randomly and the average number of migrating cells in the five fields was taken for each group.

Subconfluent monolayers of 4T1-NFκB-luc cells (5×10⁴/well) pretreated for 24 h with Hirsutine (12.5 and 25 μM) were cultured for another 24 h in serum-free RPMI-1640. After incubation, cell-free supernatants were collected and mixed with sample buffer containing 2% SDS (without 2-mercaptoethalnol) and incubated at 37°C for 20 min. Comparative gelatin zymography was performed on 10% SDS-PAGE with 0.1% gelatin. Samples were electrophoresed at 10 mA for 4–5 h at 4°C. Gels were washed with buffer containing 2.5% Triton X-100 and 0.01 M Tris-HCl for 2 h at 4°C and then washed with 0.01 M Tris-HCl for 40 min at room temperature. Gels were incubated in buffer containing 0.05 M Tris-HCl, 0.5 mM CaCl₂ and 1 μM ZnCl₂ for 48 h at 37°C. After incubation, gels were stained with Coomassie Brilliant Blue for 6 h and destained with 5% acetic acid and 10% methanol. The bands were scanned by an image scanner and quantified by Image J software.

4T1-NFκB-luc cells were treated with Hirsutine for 24 h. Treated cells were collected, washed with phosphate-buffered saline (PBS), and lysed in lysis buffer [25 mM HEPES (pH 7.7), 0.3 M NaCl, 1.5 mM MgCl₂, 0.2 mM EDTA, 0.1% Triton X-100, 20 mM b-glycerophosphate, 0.1 mM sodium orthovanadate, 0.5 mM phenylemethylsulfonyl fluoride (PMSF), 1 mM dithiothreitol, 10 mg/ml aprotinin and 10 mg/ml leupeptin]. The cell lysates were separated by 10% SDS-PAGE and transferred to PVDF membranes using a glycine transfer buffer [192 mM glycine, 25 mM Tris-HCl (pH 8.8) and 20% (v/v) methanol]. After blocking with Block Ace for 4 h at room temperature, the membrane was incubated overnight with primary antibodies, and then for 60 min with secondary antibodies. Primary antibodies were used at a dilution of 1:1,000. The secondary antibodies (horseradish peroxidase-conjugated goat anti-mouse IgG) were used at a dilution of 1:2,000 and visualized with an enhanced chemiluminescence system (Amersham Biosciences).

Female BALB/c mice were purchased from Japan SLC, Inc. (Hamamatsu, Japan) and maintained in a temperature-controlled, pathogen-free room. All animals were handled according to the approved protocols and guidelines of the Animal Committee of Toyama University (A2012INM-6). For the experimental metastasis model, 4T1-luc cells were inoculated intravenously (i.v., 5×10⁵) with or without pre-treatment with Magnolol or Hirsutine (24 h, 25 μM). For lung metastasis imaging, mice were injected with D-luciferin 7 days after tumor inoculation, then the lungs were removed to measure luminescence using in vivo imaging system (IVIS Lumina II; Caliper Life Sciences, MA, USA).

All the data are expressed as the mean ± SD of at least two or three independent experiments unless otherwise stated. Statistical significance was analyzed using Student’s t-test. P<0.05 was considered significant.

In order to identify novel anti-metastatic drug candidates targeting inflammatory signals in cancer cells, we first screened our library of natural product-derived compounds using murine 4T1 breast cancer cells stably expressing NF-κB luciferase reporter (4T1-NFκB-luc cells). 4T1-NFκB-luc cells were incubated with a series of natural product-derived compounds (Table I) for 24 h and luminescence was measured to determine NF-κB activity. To discriminate whether the inhibition of NF-κB activity is related to the reduction of cell viability, all tested compounds were measured for their direct cytotoxicity on 4T1-NFκB-luc cells in parallel with the luciferase reporter assays. As summarized in Fig. 1, 4 out of 56 compounds (resveratrol, curcumin, magnolol, Hirsutine) significantly suppressed NF-κB activity in 4T1-NFκB-luc cells (>80%) with relatively little effect on cell viability (<50%). While the pharmacological effect of resveratrol (16–18), curcumin (19,20) and magnolol (21–31), in NF-κB inhibition has been previously recognized, there is almost no information of Hirsutine on its effect on NF-κB inhibition in cancer cells; therefore, we decided to further explore the potential of Hirsutine as novel anti-metastatic drug candidate targeting inflammatory signals in cancer cells.

To further determine the specificity of Hirsutine in its inhibition of NF-κB activation, 4T1-NFκB-luc cells were treated with different doses of Hirsutine to evaluate their effect on NF-κB reporter activity and cell viability. Hirsutine specifically inhibited NF-κB activation of 4T1 cells at a concentration that did not largely affect cell viability (Fig. 2A and C) and further inhibited NF-κB activation even in the presence of TNF-α stimulation in 4T1 cells (Fig. 2B and C). Collectively, the presented results strongly support that Hirsutine is a potent inhibitor of NF-κB in metastatic 4T1 breast cancer cells.

We next examined the effect of Hirsutine on the metastatic potential of 4T1 cells accordaning to the inhibition of NF-κB. As shown in Fig. 3, Hirsutine showed dose-dependent inhibition of the cell migration of 4T1 cells, as determined by the wound closure assay. In addition, pre-treatment with Hirsutine inhibited 4T1 cell haptotaxis (Fig. 4A) towards fibronectin in the Transwell chamber assay. Furthermore, Hirsutine pre-treatment showed significant inhibition of the invasion activity of 4T1 cells (Fig. 4B). In order to determine whether the inhibitory effect of Hirsutine on cellular invasion was relative to the proteolytic activity of 4T1 cells, we employed the gelatin zymography assay to examine the effect of Hirsutine treatment on the production of MMP-2 and MMP-9 in 4T1 cells. As shown in Fig. 5, Hirsutine-treated 4T1 cells showed a reduction in the activity of both MMP-2 and MMP-9 (Fig. 5A and C). We further confirmed that the cytoplasmic expression of MMP-9 was also reduced in Hirsutine-treated 4T1 cells as evaluated by western blotting (Fig. 5B). Collectively, these results clearly indicate that Hirsutine reduces the metastatic potential of 4T1 cells in vitro by regulating cellular migration or invasion in accordance with the inhibitory activity of NF-κB activation. We then tested the therapeutic potential of Hirsutine in breast cancer using an in vivo animal model. In an experimental lung metastasis model of 4T1 breast cancer, we found that pre-treatment with Hirsutine inhibited the metastatic lung colonization of 4T1 cells (Fig. 6).

To further explore the clinical relevance of our findings, we tested the effect of Hirsutine on human breast cancer cell lines MDA-MB-231 and MDA-MB-468. Similar to our observation in murine 4T1 cells, both Magnolol and Hirsutine inhibited the haptotaxis of MDA-MB-231 or MDA-MB-468 cells (Fig. 7) at a non-cytotoxic dose (data not shown) implying the efficacy of the compounds in human breast cancer cells in vitro.