All animal experiments were performed with approval from the Animal Care Committees of Chonbuk National University. Specific pathogen-free C57BL/6 mice were purchased from the Samtako Bio Korea Co., Ltd. (Osan, Korea). To confirm both the effects on tumor cells and tumor angiogenesis, C57BL/6 mice were used in the experiment based on a previous study (12). All mice were transferred to our pathogen-free animal facilities and given ad libitum access to a standard diet (PMI LabDiet) and water. Mice used for the present study were 7 weeks old. All mice were divided into 4 groups: B16F10 injected, baicalein-treated after B16F10 injection, LLC injected and baicalein treated after LLC injection.

B16F10 melanoma cells were obtained from the Korean Cell Line Bank and Lewis lung carcinoma LLC cells were obtained from Bundang CHA Medical Center. To generate the tumor models, suspended tumor cells (1ⅹ10⁶ cells in 100 µl) were subcutaneously (s.c.) injected into the dorsal flank of 7-week-old male mice. After implantation, 1.5 mg/kg baicalein was intraperitoneally (i.p.) injected at the same time every day for 2 weeks from day 7 after injection of the tumor cells. Tumor volume and tumor growth rate were measured using previously reported methods (12).

The mice were sacrificed by cervical dislocation on the indicated days. Tumor tissues and lymph nodes were fixed in 4% paraformaldehyde (PFA) for 4 h, dehydrated in 20% sucrose solution overnight, and embedded with tissue freezing medium (Leica, Wetzlar, Germany). Frozen blocks were cut into 80-µm sections. Samples were blocked with 5% donkey (or goat) serum in PBS-Tween-20 (PBST) (0.03% Triton X-100 in PBS), and then incubated for 4 h at room temperature (RT; range, 20–27°C) with the following primary antibodies: anti-CD31 (hamster, clone 2H8; Millipore Billerica, MA, USA), FITC-conjugated anti-α-SMA (mouse, clone 1A4; Sigma-Aldrich, St. Louis, MO, USA), anti-caspase-3 (rabbit polyclonal; R&D Systems, Minneapolis, MN, USA), anti-LYVE-1 (rabbit polyclonal; AngioBio, Del Mar, CA, USA), and anti-pan-cytokeratin (mouse, clone AE1/AE3; Abcam, Cambridge, MA, USA). After 3–5 washes, the samples were incubated for 2 h at RT with the following secondary antibodies: Cy3-conjugated anti-hamster IgG (127-165-160), Cy3-conjugated anti-mouse IgG (715-165-150), Cy3-conjugated anti-rabbit IgG (711-165-152) and FITC-conjugated anti-rabbit IgG (711-095-152) (Jackson ImmunoResearch, West Grove, PA, USA). Nuclei were stained with 4′,6-diamidino-2-phenylindole (DAPI). The samples were then mounted in fluorescent mounting medium (Dako, Carpinteria, CA, USA) and immunofluorescent images were acquired using a Zeiss LSM 510 confocal fluorescence microscope (Carl Zeiss, Jena, Germany).

For hematoxylin and eosin (H&E) staining, lung tissues were fixed overnight in 4% PFA. After tissue processing using standard procedures, samples were embedded in paraffin and cut into 5-µm sections followed by H&E staining.

Blood vessel density measurements in the intratumoral regions were analyzed using photographic analysis in the ImageJ software (http://rsb.info.nih.gov/ij) and the LSM Image Browser (Carl Zeiss). For blood vessel density, the CD31-positive area (per random 0.44 mm² area) was measured in the intratumoral region. The range of α-SMA-positive mural cells was calculated as a percentage of the corresponding positive fluorescence region along the CD31-positive blood vessels in random 0.44 mm² intratumoral regions.

HUVEC growth medium was purchased from Lonza Group Ltd. (Basel, Switzerland). Cells were cultured in endothelial cell growth medium (EGM-2 MV BulletKit) and used at passage 3–4 in all experiments. B16F10 and LLC cells were cultured in Dulbecco's modified Eagle's medium (DMEM) (purchased from Gibco, Grand Island, NY, USA) supplemented with 10% FBS (purchased from Atlas Biologicals, Fort Collins, CO, USA), 100 U/ml penicillin and 100 µg streptomycin (purchased from Sigma). All cells were incubated at 37°C with 5% CO₂. Baicalein (purchased from Sigma) was dissolved in dimethyl sulfoxide (DMSO) to create a stock solution at 100 mM (for in vitro use) or 3 mg/ml (for in vivo use), which was then diluted with autoclaved PBS buffer prior to use.

Cell proliferation was detected by MTT assay and cell viability was detected by Annexin V assay. The colorimetric 3–4,5-dimethylthiazol-2-yl-2,5-diphenyl tetrazolium bromide (MTT) assay was performed to quantitate the effect of baicalein on mitochondrial dysfunction. Briefly, 1ⅹ10⁵ cells/well were seeded in a 24-well microplate in a final volume of 300 µl, incubated overnight, and treated with baicalein for 24 h. Following treatment for 24 h, 30 µl of MTT (5 mg/ml in PBS) was added to the cells and incubation was carried out for 2 h. The MTT solution was then removed and replaced by 300 µl dimethyl sulfoxide (DMSO), and the plates were shaken for 10 min. Then, 200 µl of the sample were transferred to a 96-well microplate. The optical density was determined at a wavelength of 570 nm. Apoptosis was assessed by a commercial Annexin V assay (Santa Cruz Biotechnology) according to the manufacturer's protocol. Annexin V content was determined by measuring fluorescence at excitation 488 nm and emission at 525 nm using a Guava EasyCyte HT system (Millipore).

Tumor cells (B16F10, LLC) and HUVECs were grown on 6-well plates. Cultured cells were wounded by scraping with a 1,000 µl pipette tip. After wounding, the cells were gently washed twice with PBS and incubated with 2% FBS-containing medium supplemented with 100 µM baicalein for 24 h. Then, the wounded areas were observed and photographed using a microscope (Nikon Eclipse TS100; Nikon Corporation, Tokyo, Japan) at a magnification of ×10.

Tube formation was assessed by an endothelial cell tube formation assay (Corning Inc., Corning, NY, USA) according to the manufacture's protocol. Tube number was counted in 3 random fields using ImageJ software (http://rsb.info.nih.gov/ij).

B16F10 and LLC cells were homogenized in cold RIPA buffer containing a protease inhibitor cocktail 18 h after baicalein treatment. Each protein was separated by SDS-PAGE gel and transferred to nitrocellulose membranes. After blocking with 5% skim milk, the membranes were incubated with anti-caspase-3 (rabbit monoclonal) and anti-cleaved caspase-3 (rabbit) and anti-PARP (rabbit) (all from Cell Signaling Technology, Inc., Beverly, MA, USA) and anti-β-actin antibody (mouse monoclonal; Sigma-Aldrich) in blocking buffer overnight at 4°C, and then with HRP-conjugated secondary antibody for 2 h at RT. Signals were developed with enhanced chemiluminescence HRP substrate (Millipore) and detected using the Fusion FX7 acquisition system (Vilbert Lourmat, Eberhardzell, Germany).

Tumor cells (B16F10, LLC) were cultured on glass slides, fixed with cold acetone and blocked using 5% donkey serum in Tris-buffered saline with Tween-2 (TBST). The cells were incubated with anti-caspase-3 (rabbit polyclonal; R&D Systems) at 4°C. Cells were incubated with Cy3-conjugated anti-rabbit IgG (Jackson ImmunoResearch, 711-165-152). Nuclei were stained with DAPI. Then, the samples were mounted in fluorescent mounting medium (Dako) and immunofluorescent images were acquired using a Zeiss LSM510 confocal fluorescence microscope (Carl Zeiss).

Values are presented as mean ± standard deviation (SD). Significant differences between means were determined by unpaired Student t-tests or analysis of variance with one-way ANOVA followed by the Student-Newman-Keuls test. Statistical significance was set as <0.05 and indicated as *p<0.05 or **P<0.01 in the figures and legends.

We administered baicalein to mice injected with B16F10 melanoma or LLC carcinoma cells in order to examine its anticancer effects on the two solid tumor models. As a result of the baicalein treatment, tumor size and tumor volume were reduced in each model compared to the untreated group (Fig. 1A and B). The maximum effect of baicalein treatment was noted on day 11 (49.20% inhibition) for the B16F10-implanted group and day 19 (48.25% inhibition) for the LLC-implanted group (Fig. 1C). Notably, the tumor growth rate was significantly reduced in the baicalein-treated group compared to the untreated group, with the most significant decrease observed 11 days after tumor implantation (Fig. 1D and E). As shown in Fig. 1D and E, tumor growth rates in the B16F10- and LLC-injected mice decreased by 144 and 51%, respectively, upon treatment with baicalein. In vitro measurements of cell proliferation with B16F10 and LLC cells revealed dose-dependent decreases by baicalein treatment (Fig. 2). Inhibition of cell proliferation by baicalein was greater in the B16F10 cells than that in the LLC cells at a given treatment concentration. Viability of B16F10 and LLC cells was decreased upon baicalein treatment by 35 and 22%, respectively. In addition, baicalein strongly inhibited tumor cell migration (Fig. 2C and D). The Annexin V assay was performed to assess whether baicalein induces apoptosis of tumor cells. The results showed that cell viability was reduced by 55% in the B16F10 cells and 31% in the LLC cells (Fig. 3A). Western blot methods were used to assess the effect of baicalein on the expression and activation of caspase-3 (Cas3). Although Cas3 expression was not significantly changed, Cas3 activation (cleaved-cas3) through proteolytic cleavage was increased in a dose-dependent manner after baicalein treatment (Fig. 3B). In addition, Cas3 activation through cleavage was confirmed via immunofluorescence staining in both in vitro tumor cells and in vivo tumor tissue samples (Fig. 3C and D). To determine whether apoptosis by baicalein was dependent on Cas3 activation, we confirmed the expression of PARP. As a result, cleaved-PARP was also increased by baicalein treatment (Fig. 3E). Therefore, these results demonstrated that baicalein treatment induced apoptosis via Cas3 activation.

Tumor tissue was stained with CD31 (endothelial cell specific) and α-SMA (mural cell specific) immunofluorescent markers to confirm the effect of baicalein on tumor angiogenesis (Fig. 4A and B). The density of blood vessels was markedly reduced by baicalein treatment in the tumor models with B16F10 and LLC implanted cells. In mice injected with B16F10 cells, intratumoral blood vessel density was 37.56% in the untreated group and 21.24% in the baicalein-treated group. The mice injected with LLC cells showed a pattern similar to the B16F10-injected group, with intratumoral blood vessel densities of 14.88% in the untreated group and 9.51% in the baicalein-treated group. α-SMA was expressed at very low levels. These results indicate that baicalein treatment strongly inhibited vasculature formation during tumor progression. HUVECs were treated with baicalein and the effects on angiogenesis were examined in a tube formation assay and a cell migration assay (Fig. 4C and D). Baicalein strongly suppressed tube formation by almost 80% or more at concentrations above 100 µM. A cell migration assay with 100 µM baicalein significantly reduced migration of HUVECs. These results suggest that baicalein directly affects vascular endothelial cells to suppress angiogenesis.

Our results showed that baicalein inhibited tumor angiogenesis, while tumor cells are known to spread to other organs via the tumor vasculature. Therefore, we used LLC-injected mouse models to investigate the antimetastatic effect of baicalein treatment. To confirm the effect of baicalein on metastasis, we separated the inguinal lymph nodes and lungs of the LLC-implanted mice and compared the control and baicalein treatment groups. As a result, baicalein was found to delay tumor metastasis. In addition, the lymph node of the baicalein-treated group showed a similar size to the wild-type lymph node. As shown in Fig. 5A and C, the lymph nodes of the baicalein-treated mice were smaller than those of the untreated mice, in which the lymph nodes were hypertrophic. We determined the LLC cells within the lymph nodes by immunohistochemistry using the cytokeratin (LLC marker) and LYVE-1 (lymphatic marker) antibodies. These experiments also showed that baicalein reduced LLC metastasis to the lymph node (Fig. 5A and C). In addition, it was confirmed that the LLC metastasis to the lungs was reduced by baicalein treatment (Fig. 5B and D). These results showed that baicalein can attenuate tumor cell metastasis by inhibiting formation of the intratumoral vasculature.