Human fibrosarcoma HT1080 (KCLB no. 10121) and murine melanoma B16F10 (KCLB no. 80008) cells were purchased from the Korean Cell Line Bank (Seoul, Korea) and maintained in RPMI-1640 or Dulbecco’s modified eagle’s medium (DMEM; Cellgro, Manassas, VA, USA) supplemented with 10% (v/v) fetal bovine serum (FBS; Cellgro), 100 U/ml penicillin and 100 µg/ml streptomycin (Cellgro) at 37°C in a humidified incubator with 5% CO₂. Human umbilical vein endothelial cells (HUVECs) were purchased from InnoPharmaScreen (Asan, Korea) at passage 2, maintained in endothelial cell growth medium-2 (EGM-2; PromoCell, Heidelberg, germany), and used for experiments at passages 3–8. For the animal experiments, female athymic nude mice were purchased from nara Biotech (Seoul, Korea). Female C57BL/6 mice and male Sprague Dawley rats were obtained from Taconic Farms Inc. (Samtako Bio Korea, Osan, Korea). All animals were housed under controlled conditions using a 12/12-h light/dark cycle at 22±1°C and 55±5% humidity under specific pathogen-free conditions. All animal experiments were approved by the Animal Care and Use Committee of the Korea Institute of Oriental Medicine (KIOM, Daejeon, Korea), with the reference numbers #13–42, #13–48 and #14–27, and were performed in accordance with the guidelines of the Animal Care and Use Committee at KIOM.

Herbs used for the preparation of MA128, including Glycyrrhizae radix, Polygoni cuspidati radix, Sophorae radix, Cnidii rhizoma, and Arctii fructus, were purchased from yeongcheon Oriental Herbal Market (yeongcheon, Korea), and all voucher specimens were deposited in the herbal bank in KIOM. The authenticity of the plant species was validated by Professor Ki Hwan Bae (Chungnam national university, Daejeon, Korea). Pure L. rhamnosus KFRI128 cultures were obtained from the Korea Food Research Institute (KFRI) and incubated in MRS medium for 24 h at 37°C as described previously (12). A total of 1,840 g MA was soaked in 18.4 l distilled water (DW) and then heat-extracted for 3 h at 115°C using an extractor (Cosmos-600 Extractor; Gyeonseo Co., Incheon, Korea). After filtration through standard testing sieves (150 µm; Retsch, Haan, Germany), the pH of decoction MA was adjusted to 8.0 using 1 M NaOH. It was then sterilized at 5 min at 121°C using an autoclave. For the fermentation of decoction MA, 10 ml L. rhamnosus (1×10⁸ CFU/ml) was added to 10 liter decoction MA and incubated at 37°C for 48 h. The fermented MA, designated as MA128, was filtered through a 60-µm nylon net filter (Millipore, Bedford, MA, USA), freeze-dried and stored in a desiccator at 4°C. A total of 376 g MA128 powder was produced and the yield was 20.44%. The freeze-dried MA128 powder was dissolved in 10% (v/v) DMSO in DW and centrifuged at 14,000 rpm for 10 min. The supernatant was filtered (0.2-µm, pore size) and kept at 4°C prior to use.

Type A gelatin from porcine skin, fibronectin from human plasma, 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyltetrazolium bromide (MTT), phorbol 12-myristate 13-acetate (PMA) and mitomycin C were obtained from Sigma Chemical Co. (St. Louis, MO, USA). Type I collagen solution (0.3%, Cellmatrix type I-A) and the growth factor-reduced Matrigel basement membrane matrix were purchased from Nitta Zerachin, Inc. (Osaka, Japan) and BD Biosciences (Bedford, MA, USA), respectively. Antibodies against p21, p27, cyclin B, cyclin D, p38, p-p38 (Thr180/Tyr182), ERK1/2, p-ERK1/2 (Thr202/Tyr204), JNK, p-JNK (Thr183/Tyr185), Akt, p-Akt (Ser473), mTOR, p-mTOR (Ser2481), IκBα, p-IκBα (Ser32/36), p-4e-BP1 (Thr37/46), MMP-9 and tubulin were purchased from Cell Signaling Technology (Danvers, MA, USA). Anti-HIF-1α antibodies were obtained from BD Biosciences.

For high-performance liquid chromatography (HPLC), HPLC-grade acetonitrile and trifluoroacetic acid (TFA) were purchased from J.T. Baker (Philipsburg, NJ, USA) and Sigma, respectively. Icariin and glycyrrhizin were obtained from the Tokyo Chemical Industry Co. (Tokyo, japan). Matrine, chlorogenic acid, nodakenin and arctiin were purchased from Faces Biochemical Co. (Wuhan, China). Arctigenin and decursin were obtained from Tocris Biosciences (Bristol, UK) and KFDA (Osong, Korea), respectively.

Five-week-old female athymic nude mice were inoculated subcutaneously with 2×10⁶ HT1080 cells/mouse in the femoral region. On day 5, the mice were divided randomly into 3 groups (n=5 per group) and treated daily with saline (control) or MA128 (75 or 150 mg/kg) in a volume of 100 µl for 12 days. Tumor volume and weight were measured as previously described (13). To induce pulmonary metastasis, C57BL/6 mice were injected intravenously with B16F10 cells (3×10⁵ cells/mouse) via the tail vein. After being divided randomly into 3 groups (n=4 per group), the mice were administered saline (control) or MA128 at 75 or 150 mg/kg. After daily treatment for 17 days, the mice were sacrificed, their lungs were fixed in Bouin’s solution (Sigma), and metastasized black colonies were counted macroscopically. To assess the safety of MA128, 5-week-old athymic nude mice and C57BL/6 mice (n=3 for each group) were treated daily with vehicle (saline) or MA128 (75 or 150 mg/kg) for 14 and 15 days, respectively. Body weight was measured daily, and the weights of the major organs, including the liver, heart, lung, spleen and kidneys, were measured at the time of sacrifice. Serological and hematological parameters were analyzed using XL 200 (erba Diagnostics, Mannheim, germany) and an ADVIA 2120i hematology system (Siemens Healthcare Diagnostics, Tarrytown, NY, USA), respectively.

To examine the effects of MA128 on cell proliferation, 5×10³ cells/well in a 96-well plate were treated with various concentrations of MA128 (50–1,000 µg/ml) for 48 h, and MTT assays were then performed as previously described (13).

To assess anchorage-independent cell growth, 5×10⁴ cells were suspended in 2 ml medium containing the indicated concentrations of MA128, 10% FBS, and 0.3% agar, and then applied over the bottom agar that had been pre-solidified with 3 ml medium containing 10% FBS and 0.6% agar. During incubation, colony formation was observed daily under a phase-contrast microscope, and images were captured.

Cell-to-ECM adhesion was measured in 96-well culture plates as reported previously with slight modifications (14). After coating the plates with 50 µl Fn (5 µg/ml) or 0.3% type I collagen solution overnight at room temperature, the wells were washed with cold PBS, blocked with medium containing 0.3% BSA (200 µl/well) and then washed again. MA128 pre-treated cells were suspended in serum-free medium at a density of 5×10⁵/ml, plated on Fn- or type I collagen-coated wells (200 µl/well), and then allowed to adhere for 1 h at 37°C. Unattached cells were removed by washing and the attached cells were stained using 0.2% crystal violet/20% methanol (w/v) solution. After staining, the cells were dissolved in 1% SDS solution, and the spectrophotometric absorbance was measured at 560 nm.

The cells were pre-incubated with mitomycin C (25 µg/ml) for 30 min to inhibit proliferation. Subsequently, they were injured by scraping across the cell monolayer. After washing to remove floating cell debris, the cells were allowed to migrate in the presence of MA128 and monitored under a phase-contrast microscope.

Migration/invasion assays were performed using Transwell chambers with polyethylene tetraphthalate membranes (6.5-mm diameter and 8-µm pore size) as described previously with slight modifications (14,15). For the invasion assays, the membranes were coated with 50 µl Matrigel (diluted in RPMI-1640, 1:8) to form an intervening invasive barrier.

Cells plated in 12-well culture plates at 80% confluence were pre-incubated with MA128 in serum-free media for 12 h and then stimulated with 5 nM PMA for an additional 24 h. Conditioned media (CM) were collected and electrophoresed using 8% SDS-PAGE containing 0.1% gelatin. After washing, the gels were incubated for 24 h and then stained and de-stained as previously described (14,15). MMP-9 was detected as a clear 92-kDa band against the blue background.

Capillary-like tube formation was assayed using HUVECs, an ECMatrix assay kit (Millipore, Temecula, CA, USA), and µ-Slide Angiogenesis ibiTreat chambers (ibidi gmbH, germany), as preivously described (16). HUVECs (5×10³ cells) were suspended in 50 µl MA128-treated HT1080 CM, seeded onto the surface of polymerized ECMatrix, and incubated for 4–8 h at 37°C in an incubator containing 5% CO₂. The cellular tube networks were observed using a phase-contrast microscope and photographed, and the relative tube area was quantified in five fields of view per well.

Dorsal aortas isolated from Sprague Dawley rats were cut transversely into 1-mm thick rings. Each aortic ring was then placed in 48-well culture plates pre-coated with 100 µl Matrigel. After sealing the rings with 40 µl Matrigel, 400 µl EGM-2 was added to the wells and the plates were incubated for 3 days to allow microvessel sprouting from the aortic rings. The EGM-2 was then replaced with control or MA128-treated CM, and microvessel outgrowths were observed and photographed for an additional day.

Whole cell lysates were extracted using M-PER Mammalian Protein extraction reagent (Thermo Scientific, Rockford, IL, USA). Equal amounts of protein were resolved using 8–15% SDS-PAGE, incubated with specific antibodies, and detected as previously described (13). The band intensities were analyzed using ImageJ software.

The cells were incubated with the indicated concentrations of MA128 for 48 h. The levels of VEGF, PDGF, and bFGF in control and MA128-treated CM were determined using a Human ELISA Development kit (Peprotech, Rocky Hill, NJ, USA) according to the manufacturer’s instructions.

The phytochemical profile of MA128 was measured using HPLC analysis on HPLC-DAD (LaChrom elite, Hitachi High-Technologies Co., Tokyo, Japan), and chromatographic separation was achieved on a Phenomenex Luna C18 column (4.6×250 mm, 5 µm). A gradient elution of 10% acetonitrile in deionized water containing 0.05% TFA (A) and 60% acetonitrile in deionized water containing 0.05% TFA (B) was performed as follows: 0–25 min with 0% B; 25–100 min with 0–45% B; 100–110 min with 15–100% B; 110–120 min with 100% B; 120–122 min with 100–0% B; and 122–140 min with 0% B. The flow rate and injection volume were 1 ml/min and 10 µl, respectively. HPLC chromatograms were obtained using UV at 190–400 nm. Standard (ST), ST-spiked sample (SP), and sample solutions were dissolved and diluted in methanol.

Data are presented as mean ± standard deviation (SDs). The statistical significance of the differences between groups was analyzed using a two-way analysis of variance (ANOVA) and the Student’s t-test with SigmaPlot 8.0 software. P<0.05 was considered to indicate significant differences.

Prior to determining the anticancer effects of MA128 in athymic nude mice, we assessed whether the repeated administration of 75 or 150 mg/kg MA128 elicited systemic toxicity. The MA128 doses used were according to the amounts administered to human adults (36.8 g/day/60 kg of body weight) and the yield of the extraction. Fourteen days of MA128 administration did not cause death or abnormal behavior. Body and organ weights, the ratios of GOT/GPT and BUN/CRE, and the hematological parameters were comparable between MA128-treated and control mice (Table I), suggesting that MA128 caused no systemic toxicity, including hepatic and renal dysfunction. To assess the inhibitory effects of MA128 on in vivo tumor growth, athymic nude mice (n=5 per group) inoculated with highly malignant HT1080 cells were treated with saline or 75 or 150 mg/kg MA128 for 12 days beginning 5 days after tumor inoculation. As shown in Fig. 1A and B, the administration of 75 or 150 mg/kg MA128 retarded tumor growth successfully and suppressed tumor volume by ~75 and 85%, respectively, compared with the saline-treated control mice on day 17. The control mice exhibited a mean tumor weight of 0.441±0.108 g, whereas the mice treated with 75 and 150 mg/kg MA128 had tumor weights of 0.139±0.025 g and 0.085±0.045 g, reflecting a decrease of 68.5 and 80.8%, respectively (Fig. 1C). In addition, there were no significant differences in body weight between the control and MA128-treated mice during the experimental period, suggesting that MA128 did not exhibit any severe toxic effects (Fig. 1D).

MTT assays were used to assess the effect of MA128 on the growth of cells treated with specified concentrations of MA128 for 48 h. As shown in Fig. 2A, cell viability was not altered by 50–250 µg/ml MA128 compared with the untreated control cells, whereas treatment with 500 and 1,000 µg/ml MA128 caused a marked decrease in cell viability and induced most cells to shrink and float (Fig. 2B). Western blotting revealed that MA128 increased the levels of the cyclin-dependent kinase inhibitors p21 and p27 in a dose-dependent manner, whereas it decreased the levels of cyclin B and D compared with the control cells (Fig. 2C). In addition, MA128 increased the levels of the pro-apoptotic proteins Bad and Bax and decreased the levels of anti-apoptotic Bcl-2 and XIAP (Fig. 2C). It was reported previously that the MAPK signaling pathways can induce cell proliferation or death depending on the stimulus and cell type (17–19). As shown in Fig. 2D, cytotoxic doses of MA128 elevated the levels of phosphorylated p38 and ERK significantly but had little effect on JNK activation. To confirm the roles of p38 and ERK activation in MA128-mediated cell death, the cells were pre-incubated with or without pharmacological inhibitors of p38 (SB203580), ERK (PD98059), or JNK (SP600125) for 1 h and then treated with MA128 for 48 h. As shown in Fig. 2E, SB203580 and PD98059 moderately protected MA128-treated cells from death by ~50%, whereas SP600125 had little effect. This result suggested that p38 and ERK activation play roles in MA128-mediated cell death.

Prior to assessing the anti-metastatic effects of MA128 in C57BL/6 mice, the safety of MA128 treatment was assessed over a 15-day period. As shown in Table II, MA128 did not induce any adverse effects, such as loss of body weight or differences in organ weights. In addition, there were no critical differences in the serological or hematological parameters between MA128-treated and control mice. After confirming safety, we evaluated whether MA128 reduced the pulmonary colonization of B16F10 cells injected intravenously into the tail vein of C57BL/6 mice. As shown in Fig. 3A and B, B16F10 cells metastasized to the lungs of control mice and formed a considerable number of black colonies (939.81±83.98). However, the incidence of metastatic colonies in mice treated with 75 and 150 mg/kg MA128 was decreased significantly by ~57.9 (395.33±75.72) and 61.1% (365.67±105.19), respectively, compared with the control mice. These results suggested that MA128 intake suppressed the pulmonary metastasis of B16F10 cells efficiently without causing systemic toxicity. To examine the anti-metastatic effects of MA128 in vitro, we monitored the cell-to-ECM adhesion, migration, and invasion of B16F10 cells in the presence or absence of MA128. Pre-treatment with MA128 inhibited the attachment of cells to FN and collagen in a dose-dependent manner (Fig. 3C). In the Transwell culture system, serum-induced migration and invasion were markedly reduced by MA128 to ~10.6–25.8 and 15.7–41.2% in the control cells, respectively (Fig. 3D).

To clarify the anti-metastatic mechanism of action of MA128, HT1080 cells and PMA were used under the conditions as previously described (20–22). At the non-toxic concentrations of 50–250 µg/ml, MA128 treatment prevented HT1080 anchorage-independent colony formation (Fig. 4A) and the adhesion to FN and collagen (Fig. 4B) in a dose-dependent manner. Results of the wound-healing assay showed that the control HT1080 cells repaired the wounded area efficiently to ~35 and 70% levels after 18 and 40 h, respectively. By contrast, MA128 treatment inhibited wound repair significantly, allowing only ~10 and 15% healing at 18 and 40 h, respectively (Fig. 4C). In addition, the ability of HT1080 cells to undergo Transwell migration and invasion was markedly increased by PMA stimulation, whereas MA128 almost completely inhibited migration and invasion under PMA-stimulated conditions (Fig. 4D), suggesting that MA128 exerted potent anti-metastatic effects. To elucidate the mechanism by which MA128 inhibited the metastatic potential, we examined whether MA128 suppressed the expression and activity of MMP-9, which is essential for facilitating tumor metastasis by degrading the surrounding ECM.

As shown in Fig. 5A, PMA strongly induced the gelatinolytic activity and secretion of MMP-9, as determined using gelatin zymography and western blotting. By contrast, treatment with MA128 markedly inhibited the PMA-induced increase in MMP-9 activity and expression in a dose-dependent manner. Since the transcription factor nF-κB is critical for MMP expression and invasion in various cancer cells (23,24), we examined the effects of MA128 on PMA-induced NF-κB activation. As shown in Fig. 5B, PMA treatment markedly increased the phosphorylation and degradation of IκBα in HT1080 control cells. However, MA128 significantly inhibited PMA-induced IκBα phosphorylation and degradation. Consequently, the p-IκBα/IκBα ratio was much lower in the MA128-treated cells compared with that in the control cells. Taken together, these results suggested that non-toxic doses of MA128 suppress the metastatic potential of HT1080 cells by reducing MMP-9 activity and inhibiting nF-κB activation.

Tumor-induced angiogenesis, the formation of new blood vessels from existing vasculature, is induced by tumor cells and is critical for sustained tumor growth and metastasis (25–27). It has been reported that supernatants harvested from various tumor cells triggered an angiogenic response. Therefore, we examined the ability of HUVECs to form network structures on Matrigel after incubation with control or MA128-treated CM derived from HT1080 cell cultures. As shown in Fig. 6A, control CM stimulated HUVECs to assemble into tube-like structures on Matrigel, whereas MA128-treated CM significantly reduced tube formation in terms of the number, length, and area of capillary-like structures in a dose-dependent manner. Using ex vivo angiogenesis assay, we assessed microvessel sprout formation in rat aortic rings, after incubation with control or MA128-treated CM from HT1080 cells. As reported previously, control CM from HT1080 cell cultures triggered microvessel sprouting around the aortic rings (16), whereas the length of the sprouting microvessels around the aortic rings incubated with MA128-treated CM was markedly shorter than that with control CM (Fig. 6B). As tumor-induced angiogenesis is affected by tumor cell-produced angiogenic factors, the levels of pro-angiogenic factors, including VEGF-α, PDGF and bFGF, were measured in the control and MA128-treated CM from HT1080 cells. As shown in Fig. 6C, ELISA revealed that MA128 treatment significantly reduced the production of angiogenic factors in a dose-dependent manner (Fig. 6C). Angiogenic factors are immediate downstream targets of HIF-1α (28). Therefore, we assessed the effect of MA128 on HIF-1α production. MA128 reduced CoCl₂-stimulated HIF-1α protein accumulation, which was accompanied via inhibition of Akt/mTOR/4E-BP-1 signaling (Fig. 6D). This result suggested that MA128 decreased tumor-induced angiogenesis by inhibiting the production of angiogenic factors via the suppression of HIF-1α accumulation.

HPLC identified eight phytochemicals in MA128: matrine, chlorogenic acid, nodakenin, arctiin, icariin, arctigenin, glycyrrhizin and decursin. To achieve the desired separation, a gradient elution of water and acetonitrile was applied. TFA was added to advance the peak shape and inhibit peak tailing, and the UV wavelengths of the eight components were adjusted based on the maximum UV spectra absorption of each component. As shown in Fig. 7, matrine, nodakenin, arctiin, icariin, arctigenin, and decursin were detected at 203 nm, and chlorogenic acid and glycyrrhizin were detected at 254 and 320 nm, respectively. Each component in MA128 was identified by comparing the retention time (t_R) and uV spectra with known data. Profiles corresponding to matrine (1, t_R 5.3 min), chlorogenic acid (2, t_R 11.3 min), nodakenin (3, t_R 52.7 min), arctiin (4, t_R 76.6 min), icariin (5, t_R 89.1 min), arctigenin (6, t_R 105.2 min), glycyrrhizin (7, t_R 109.6 min), and decursin (8, t_R 119.7 min) were identified in MA128.