Fetal bovine serum (FBS), streptomycin and penicillin were obtained from HyClone. Baicalein, SB203580 and anisomycin were from Sigma. Anti-p38 MAPK, anti-phospho-p38 MAPK (Thr180/Tyr182), anti-MMP-2 and anti-MMP-9 antibodies were purchased from Cell Signaling. Anti-β-actin was purchased from Santa Cruz.

The human gastric cancer cell lines SGC7901 and MGC803 were purchased from Type Culture Collection, Chinese Academy of Sciences (Shanghai, China). The cells were cultured in RPMI-1640 (Invitrogen) supplemented with 10% FBS, 100 U/ml penicillin and 100 μg/ml streptomycin. All cells were incubated at 37°C with 5% CO₂.

Cell survival was detected using standard 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT) assay according to the manufacturer’s instructions. First, cells were seeded in 96-well culture plates (5×10⁴ cells/well). Then, cells were treated with different concentrations of baicalein. After incubating for 24 h, the cells were washed twice with PBS and incubated with 5 mg/ml MTT (Sigma) for 4 h. Finally, the optical density was measured in a microplate reader at 562 nm formazan absorbance.

The in vitro invasion and migration activity was measured according to previously described methods with minor changes (10,11). Cells were pretreated with 0, 10, 20 and 40 μM baicalein or SB203580 (20 μM) or anisomycin (25 μg/ml) for 24 h, and surviving cells were harvested and seeded to Transwell chamber (Corning) then incubated for 24 h at 37°C. Then, the cells on the upper side of inserts were completely removed by swabbing, while the cells on the bottom side of the filter were fixed, stained and measured. For invasion assay, 10 ml Matrigel (25 mg/50 ml; BD Biosciences, Bedford, MA, USA) was applied to 8-mm pore size polycarbonate membrane filters and the bottom chamber contained standard medium.

Total RNAs were prepared by using the RNeasy Mini kit (Invitrogen). cDNA was synthesized with SuperScript III Reverse Transcriptase (Invitrogen). Real-time quantitative PCR (RT-PCR) was performed using SYBR-Green II in accordance with the PrimeScript RT-PCR kit protocol (Takara). Gene-specific primer pairs used for amplification are shown in Table I. β-actin was used as an endogenous control. The analysis of the relative gene copy number data for MMP-2 and −9 was performed using the 2^−ΔΔCt method.

Following treatment with baicalein for 24 h, samples of conditioned media were collected. Then, gelatin zymography was performed as previously described (4). Briefly, appropriate volumes of the unboiled samples (adjusted by vital cell number) were separated by 0.1% gelatin-8% SDS-PAGE electrophoresis. Then, the gels were soaked in 2.5% Triton X-100 for three times at room temperature, once for 30 min, and incubated in reaction buffer (10 mM CaCl₂, 40 mM Tris-HCl and 0.01% NaN₃, pH 8.0) at 37°C for 12 h. Gels were rinsed with distilled water, stained with Coomassie brilliant blue R-250. The gelatinolytic activities were quantified and analyzed by an image analysis system (Bio-Rad Laboratories, Richmond, CA, USA).

After treating with baicalein or SB203580 or anisomycin, proteins were extracted with lysis buffer (40 mmol/l Tris-HCl, 1 mmol/l EDTA, 150 mmol/l KCl, 100 mmol/l NaVO₃, 1% Triton X-100, 1 mmol/l PMSF, pH 7.5). Then the proteins were separated by 10% SDS-polyacrylamide gel electrophoresis and transferred onto PVDF membranes. To block non-specific binding, membranes were blocked in defatted milk [5% in Tris-buffered saline with Tween-20 (TBST) buffer] at room temperature for 1 h and were then incubated with antibodies against p38 MAPK, p-p38 MAPK, MMP-2, −9 or β-actin overnight at 4°C. The membranes were then incubated with appropriate second antibodies for 1 h at room temperature. The bands were detected with an enhanced chemiluminescence kit (ECL Plus; Amersham, Freiburg, Germany) and exposed by autoradiography. The densitometric analysis was performed using ImageJ software (GE Healthcare, Buckinghamshire, UK).

Experiments were repeated three times, and the results of the studies are expressed as the means ± SD. All statistical tests and corresponding p-values were two sided. p<0.05 was considered to indicate a statistically significant difference. We performed correlation analysis by the Z-test.

The antiproliferation effects of baicalein at various concentrations (0–400 μM) on SGC7901 cells are shown in Fig. 1. At 80 μM, baicalein clearly inhibited the proliferation of SGC7901 cells, therefore we chose a concentration range of baicalein lower than this for all subsequent experiments.

Fig. 2 shows the effect of baicalein on cell migration and invasion in SGC7901 cells that were treated with 0, 10, 20 and 40 μM of baicalein for 16 h (cell migration) and 24 h (cell invasion), respectively. We showed that baicalein reduced the invasion and migration of SGC7901 cells substantially in a concentration-dependent manner. Quantification analysis indicated that the inhibition rate of migration and invasion were ~24.27, 37.13, 77.48 and 42.43, 68.81, 97.52%, respectively. Similar antimetastatic effect of baicalein was observed in MGC803 cells (data shown in Fig. 3).

SGC7901 cells were treated with 0, 10, 20 and 40 μM baicalein for 24 h and then subjected to RT-PCR. We found that baicalein significantly reduced the transcriptional levels of MMP-2 and −9 in a concentration-dependent manner (Fig. 4).

The expression and activity of MMP-2 and −9 in SGC7901 cells that were exposed to different concentrations of baicalein were examined. Cells were treated with 0, 10, 20 and 40 μM baicalein for 24 h and then subjected to western blotting. Fig. 5A and B show that baicalein significantly reduced the protein levels of MMP-2 and −9 in a concentration-dependent manner compared with the control group. Gelatin zymography was performed to assess the activity of MMP-2 and −9 in cells treated with various concentrations of baicalein. As shown by gelatinolytic activity data, baicalein inhibited the activity of MMP-2 and −9 in a concentration-dependent manner (Fig. 5C and D). The effect of baicalein on the expression of MMP-2 and −9 in MGC803 is shown in Fig. 6.

In human gastric cancer cells, activation of p38 is required for the invasion process and the mechanism is correlated with proteinases (7,12,13); thus, we investigated the effect of baicalein on the p38 signaling pathway in SGC7901 cells. Western blotting showed that baicalein reduced the phosphorylation of p38 in a concentration-dependent manner (Fig. 7A and B). Fig. 8 shows the effect of baicalein on the phosphorylation of p38 in MGC803 cells.

In order to examine whether the inhibitory effect of baicalein on cell invasion and MMP-2 and −9 expression was correlated with inhibition of the p38 signaling pathway, SGC7901 cells were pretreated with a p38 inhibitor (SB203580, 20 μM) for 30 min and then incubated in the presence or absence of baicalein for 24 h. The results showed that treatment with SB203580 and baicalein significantly reduced both cell invasion (Fig. 7C and D) as well as MMP-2 and −9 protein expression (Fig. 7E and F).

Furthermore, chemical anisomycin, a p38 activator, was employed to confirm the role of the p38 signaling pathway. Fig. 9A shows the effects of anisomycin and baicalein on the invasion of SGC7901 cells. Quantification analysis indicated that anisomycin was able to enhance the invasion ability of SGC7901 cells, and anisomycin was able to reverse the inhibitory effect of baicalein on the invasion of SGC7901 cells (Fig. 9B and C) and MMP-2 and −9 protein expression (Fig. 9D and E). These results reveal that the inhibition of both cell invasion and MMP-2 and −9 expression by baicalein occurs through the suppression of the p38 signaling pathway.