Ardipusilloside inhibits survival, invasion and metastasis of human hepatocellular carcinoma cells.

ARTICLE INFO

Keywords:

Metastasis

Ardipusilloside

Metalloproteinase

Liver cancer

ABSTRACT

Ethnopharmacological relevance: Ardipusilloside I is a triterpene-saponin isolated from the Traditional Chinese Medicine Ardisia pusilla A. DC. Its effects and mechanism on invasion and metastasis of liver cancer cells are unclear.

Materials and methods: The human hepatocellular carcinoma cell line HepG2 and SMMC-7721 cells were treated with different doses of Ardipusilloside I. Cellular survival, in vitro migration and invasion were evaluated. In vivo metastatic abilities of the HCC cells were detected. We further investigated expression and phosphorylation of Mek, Erk and Ala by using Western blot. MMP2 and MMP9 activities were evaluated by gelatin zymography. E-cadherin expression, Racl and Cdc42 activities were examined by Western blot and pull-down assay.

Results: Ardipusilloside I inhibited invasion and metastasis of HCC cells both in vitro and in vivo by reducing the protein expressions of metalloproteinase (MMP)-9 and MMP2 proteins. Ardipusilloside I activated Rad that enhanced E-cadherin activity and resulted in significantly less metastasis.

Conclusion: Our findings indicate that Ardipusilloside I has the potential of inhibition of liver cancer survival, invasion and metastasis both in vitro and in vivo.

[c] 2012 Elsevier GmbH. All rights reserved.

Introduction

Hepatocellular carcinoma (HCC) is one of the leading causes for cancer-related death in Asia. Early metastasitic HCC do not respond to the cytotoxic effects of most of the conventional chemotherapeutic agents (Colombo 2002). Neither single chemotherapeutic agents nor aggressive combination chemotherapeutic regimens has led to any remarkable improvement in response rates (Greten et al. 2009). There is an urgent need for more effective agents for the clinical management of HCC. Many traditional Chinese herbs are promising drugs for cancer therapy because of both their potential as chemopreventive agents and their chemotherapeutic activities against HCC in experimental studies, e.g. Tanshinone II-A reversed malignant phenotype of human hepatocarcinoma cell line (SMMC-7721) and induced differentiation of leukemia cells (NB4, HL60 and K562), and human cervical carcinoma cell line (ME180) (Yuan et al. 1998, 2004; Yoon et al. 1999; Sung et al. 1999). Artemisinin could significantly inhibit in vivo and in vitro metastatic abilities of hepatocellular carcinoma cell lines (Weifeng et al. 2011).

Ardipusilloside I, a natural product which is isolated from Ardisia pusilla A. DC (of the family Myrsinaceae), a Chinese medicinal herb also known as Jiu Jie Long, has been found as an anticancer compound (Zhang et al. 2010). It has also been shown antiproliferative activity in both Lewis pulmonary carcinoma and hepatocarcinoma (Tao et al. 2005). Ardipusilloside I triggered NCl-H460 cells apoptosis, mediated by the regulation of Bcl-2, Bax and VEGFR, showing its anti-angiogenesis effect (Zhang et al. 2010). However, whether Ardipusilloside I exerts anti-invasion and anti-metastasis activity through these mechanisms remain unclear.

The present study was carried out to show the mechanisms of anticancer effects of Ardipusilloside I on hepatocarcinoma cells, both in vitro and in vivo. Our results provide a basis for the future development of Ardipusilloside I as an anti-HCC agent.

Materials and methods

Cell culture and material

Human hepatocellular carcinoma (HCC) cell lines HepG2 and SMMC-7721 were cultured on cell plates at 37 [degrees] C, 5% [CO.sub.2] in DMEM (GIBCO) supplemented with 10% FBS, 100 units/nil penicillin and 0.1 mg/ml streptomycin.

Extraction and isolation of Ardipusilloside I

Ardipusilloside I was extracted and isolated as previously described (Zhang et al. 2010). In brief, air-dried and powdered Jiu Jielong was extracted with 95% and 75% EtOH for 2 h, respectively. The extract was concentrated under decompression. The extracta sicca was mixed with water. The mixture was allowed to stand to separate into two layers and filtered. The filtrate was extracted by petroleum ether, nether, acetoacetate and n-butanol in turns. The extracta sicca by n-butanol was mixed with acetone to collect the precipitate. The precipitate components were separated systematically by column chromatography (CC, Silica Gel). Ardipusilloside I was isolated using a mixture of [CHCl.sub.3]/Me0H/[H.sub.2]O (65:35:10 by vol.) as a mobile phase and further purified by crystallization. Ardipusilloside I isolated by this procedure was over 99% in purity. Ardipusilloside I was dissolved in water and diluted to the desired concentrations immediately before use.

Drugs and treatment

Ardipusilloside I was dissolved in PBS (final concentration 0.2 ml/1). The solution was filtered through a 0.22 [micro]m micropore filter and stored at 4 C, and then further diluted in cell culture medium. HepG2 and SMMC-7721 cells were seeded in flasks. The Ardipusilloside I treated group was treated with different doses of Ardipusilloside I (0, 25, 50, 75, 100 and 125 [micro]M) for 48 h, respectively. Equal concentration of DMSO was added to the control group.

Cell migration and invasion assays

For migration assays, 3 x [10.sup.4] cells were plated into the upper chamber of 8 - [mu]M pore transwells and for invasion through the matrigel barrier, 3 x [10.sup.4] cells in 100 [micro]l of DMEM containing 0.2% BSA were added to the upper chamber of each well. Cells were allowed to migrate for 5 h for migration assay, for 24 or 48 h through matrigel. Migrated cells were fixed, stained and counted from six random fields and averaged. The experiments were repeated three times.

Wound healing

Confluent HepG2 cells were treated with mitomycin C (0.5 [micro]g/ml) for 4 h prior to wounding. After wounding, cells were treated with Ardipusilloside I (25, 50 and 75 [micro]M) for 24 h and 48 h. A similar assay was performed using SMMC-7721 cell clones.

In vivo orthotopic implantation in nude mice model

Male athymic BALB/c nu/nu mice were obtained from the Shanghai Institute of Materia Medica, the Chinese Academy of Sciences. Nude mice were handled using best human practices and were cared for in accordance with the NIH Animal Care and Use Committee Guidelines. Human HCC tumor models were established by orthotopic inoculation as described previously (Sun et al. 1996; Tian et al. 1999). Briefly, a left upper abdominal pararectal incision was made under anesthesia. The left lobe of the liver was exposed and a part of the liver surface mechanically injured with scissors. Then, a piece of HepG2 and SMMC-7721 tumor tissue (size, 2 mm x 2 mm x 2 mm) was fixed within the liver tissue, wounds were closed primarily and the abdominal wall was finally closed. Mice bearing orthotopic xenografts were randomized into control group C0 (0 g/kg/d), intervention groups C1 (50 mg/kg/d) and C2 (100 mg/kg/d) with stepwise increased dosage of Ardipusilloside I. Daily intragastric administration of Ardipusilloside I was given for 5 consecutive weeks 24 h after orthotopic implantation. Five weeks later, mice were sacrificed. The lung tissues were observed and the number of visible tumors in lung surface was counted. The lung tissues were made serial sections before being HE dyed and observed under a light microscope. Each experimental group contained 10 mice.

Gelatin zymography

Gelatin zymography analysis was performed. Cells were grown in SFCM for the required time period. To obtain conditioned SFCM containing MMP2 and MMP9 as a standard, cells were grown in SFCM for 24 h. The culture supernatant was collected by centrifugation. The gelatinases were separated from SFCM by using Gelatin Sepharose 4B beads shaking overnight at 4 [degrees] C. The beads were washed 3 times with Tris-buffered saline with (0.02%) Tween-20 (TBST) and suspended in 50 ml of 1 x sample buffer (0.075 g Tris, 0.2 g SDS in 10 ml water, pH 6.8) for 30 min at 37 [degrees] C. The extract was then subjected to zymography on 7.5% SDS-PAGE (sodium dodecyl sulfate polyacrylamide gel electrophoresis) co-polymerized with 0.1% gelatin. The gel was washed to remove SDS and was then incubated overnight in reaction buffer (50 mM Tris-HCI, pH 7, 4.5 mM [CaCl.sub.2], 0.2 M NaCl). After incubation, the gel was stained with 0.5% Coomassie Blue in 30% methanol and 10% glacial acetic acid. The bands were visualized by destaining the gel with 30% methanol and 10% glacial acetic acid.

Western blot analysis

Protein extraction and immunoblot analyses were performed as described below. Cells were lysed in RIPA buffer (50 mM Tris-HCl, pH 7.4, 1% (v/v) Triton X-100, 1 mM EDTA, 1 mM leupeptin, 1 mM phenylmethylsulfonyl fluoride, 10 mM NaF, 1 mM [Na.sub.3][VO.sub.4]). The lysates were centrifuged and the supernatants were collected. Cell lysate (20 [micro]g) was separated by SDS-polyacrylamide gel electrophoresis, blotted onto nitrocellulose membrane, and incubated with a primary antibody: rabbit monoclonal against Mek, p-Mek, Eric, p-Erk, Akt, p-Akt (diluted 1:1000; Cell signaling Co.), E-Cadherin, Racl, Cdc42 (diluted 1:1000; Santa Cruz Biotechnology, USA) and mouse monoclonal anti-[beta]-bubulin (diluted 1:10,000; Sigma Chemical Co.) for overnight at 4 [degrees] C. After repeated washing, the membranes were incubated with horseradish-peroxidase-conjugated anti-mouse or anti-rabbit secondary antibody (Santa Cruz Biotechnology) diluted 1:2000. The bands were visualized using the enhanced chemiluminescence (ECL) system (Amersham Pharmacia Biotech). Each experiment was performed in triplicate.

Pull down assay

Rac1 and Cdc42 activities were determined as described previously (Wang et al. 2009): GTP-hound active Racl and Cdc42 (Rac1-GTP, Cdc42-GTP) were precipitated by binding to glutathione-coupled Sepharose beads. Racl-GTP, Cdc42-GTP was detected by immunoblotting with anti-Racl and anti-Cdc42 antibody.

Statistic analysis

Data were expressed as mean [+ or -] SD. Statistical analysis was performed using the statistical software SPSS10.0. Student's test was used to analyze statistical differences between groups. p < 0.05 was considered as significant.

Results

Ardipusilloside I inhibits HCC cells survival by suppressing ERK1/2, Akt activation

To evaluate effect of Ardipusilloside I on HCC cells survival, hepatocellular carcinoma cell lines were treated with Ardipusilloside I at doses of 25, 50, 75, 100 and 125 [micro]M. MTT assay results showed that Ardipusilloside I inhibits survival of both HepG2 and SMMC-7721 cells dose-dependently (Fig. 1). ERIC and Akt are important signal transducers that are responsible for proliferation, differentiation and ECM-dependent cell motility (Zhong et al. 2006; Shih et al. 2009). To examine the relevance of Ardipusilloside I with ERK1/2 activity in liver cancer, phosphorylation of ERK1/2 and MEK, the molecular upstream of ERK1 /2, as well as Mt in lysis off cells were evaluated by western blot. The representative results of p-MEK, p-ERK1/2 and p-AKT in cell lysates are shown in Fig. 2. Ardipusillo-side I suppressed p-MEK, p-ERK1/2 and p-AICT in a dose-dependent manner.

[FIGURE 1 OMITTED]

[FIGURE 2 OMITTED]

Ardipusilloside 1 inhibited invasion and metastasis of HCC cells in vitro

Rapid invasion and metastasis result in the poor outcomes of liver cancer treatment. To investigate the effects of Ardipusilloside I on invasion and metastasis of HCC cells in vitro, cells were treated with Ardipusilloside I at doses of 0, 25, 50 and 75 [micro]M. Migrating and invasive capacities of HepG2 and SMMC-7721 cells were analyzed with wound healing assay and transwell chambers. In wound healing assay (Fig. 3C). The speed of wound healing of HepG2 cells movement was significantly lower than control cells. By 48 h after wounding, the wound of HepG2 cells was almost closed, while the cells treated with Ardipusilloside I stayed wide apart, showing Ardipusilloside I inhibited cell motility dose-dependently. Similar results were observed in SMMC-7721 cells. Meanwhile, in transwell chambers, Ardipusilloside I inhibited invasion in a dose-dependent and time-dependent manner (Fig. 3A and B). The inhibitory effects of Ardipusilloside I appeared at 50 [micro]M. At 75 [micro]M, the average inhibiting rate was 30% when compared with the control group at 48 h in HepG2 cells. Similarly, Ardipusilloside I prohibited HCC cells migration in SMMC-7721 cells. On Boyden chamber assay without matrigel, Ardipusilloside I produced a significant inhibition of the abilities of HepG2 and SMMC-7721 cells to migrate, with an average inhibiting rate of 71% when compared with the control group at 24 h (Table 1). Both in vitro invasion assay and migration assay suggested Artemisinin to have the potential to inhibit metastasis of hepatocellular carcinoma.

[FIGURE 3 OMITTED]

Table 1
Number of invasive cells of HepG2 (and SMMC-7721) cells after
treatment with Ardipusilloside I.


Time(h)     Dosages
            ([mu]M)

                     0             25             50               75

24         20 [+ or -]    18 [+ or -]    12 [+ or -]   6 [+ or -] 3(6
         7(19 [+ or -]  5(16 [+ or -]  4(11 [+ or -]      [+ or -] 4)
                    5)             6)             3)

48         37 [+ or -]    24 [+ or -]    15 [+ or -]  9 [+ or -] 4(12
         8(41 [+ or -]  7(28 [+ or -]  6(20 [+ or -]      [+ or -] 5)
                   10)             9)             5)

Ardipusilloside I inhibited HCC cells metastasis in vivo

Given the effects of Ardipusilloside I on cell motility and invasion in vitro, we further adopted an in vivo orthotopic implantation assay to examine the inhibitory effect of Ardipusilloside I on the metastatic ability of HCC cells in nude mice. In the mice treated with Ardipusilloside I in dose of 100 mg/kg/d (the C2 group), fewer tumors in the lungs were found after orthotopic inoculation (Table 2) compared with the control group (the C0 group). The tumor inhibition rates for the Cl (50 mg/kg/d) and C2 (100 mg/kg/d) groups were 51.8% and 79.6%, respectively (Fig. 6, both p <0.05 when compared with the C0 group). Both invasion assay in vitro and nude mice assay in vivo suggested Ardipusilloside I is a potential compound to inhibit metastasis of hepatocellular carcinoma.

Table 2
Number of visible tumors in lung surface of each group of
mice treated with Ardipusilloside I.

Dose group    Mice       Number of         Lung      p-Value
            (mg/kg/d)  (before/after)   metastasis

C0                  0           10/10  6 [+ or -] 3
C1                 50           10/10  4 [+ or -] 3   p<0.05
C2                100           10/10  1 [+ or -] 2   p<0.01

[FIGURE 6 OMITTED]

Ardipusilloside I inhibited cell invasion by altering activities of MMP9 and MMP2

ECM degradation is an essential step in tumor invasion and metastasis, which is mainly mediated by some matrix metalloproteinases (MMPs). To dissect the alterations of MMPs in Ardipusilloside I inhibited cell invasion, gelatinolytic activities of MMP2 and MMP9, two key members of MMP family, were examined by a zymographic assay. Fig. 4 showed that MMP2 and MMP9 activity was strongly decreased in the culture medium of HepG2 cells compared to that of control cells in a dose-dependent and time-dependent manner. These results were reproduced in SMMC-7721 cells. We, therefore, concluded that the inhibiting effect of Ardipusilloside I on metastasis of human hepatocellular carcinoma cells was at least partially mediated by the down-regulation of MMP2 and MMP9 activity, with the consequential degradation of ECM.

[FIGURE 4 OMITTED]

Ardipusilloside I increased adhesion of HCC cells by activating Rac1

Impaired cell-cell adhesion at the primary site is required for cancer cells to metastasize (Colin et al. 2009). Reduced E-cadherin function is the indicator of loss of cell-cell adhesion. Western blot results showed up-regulation of E-cadherin in HepG2 cells that treated with Ardipusilloside I at concentration of 50 [micro]M. This contributes to less detachment from the neighboring cells, which in turn resulted in inhibition of the cells migration. To further investigate the molecular signaling downstream of E-cadherin, western blot and pull-down assay were used to detect RhoGTPases Rac1 and Cdc42, expression and activity. Fig. 5 showed the Rac1 activity increased dramatically, whereas Cdc42 activity remained constant. None of their expression changed.

[FIGURE 5 OMITTED]

Discussion

Liver cancer is a common malignancy with a high mortality rate and its incidence is increasing. Patients with HCC usually have a poor prognosis; surgery is the only potential cure but the resection rate for HCC is only 10 [+ or -] 30% (Braicu et al. 2009). Postoperative metastasis and recurrence are common. The remaining patients are subjected to various forms of non-surgical therapy. HCC is relatively resistant to antineoplastic agents (Parkin et al. 1999). Recently, some herb medicines have been shown to be potent cancerprotective or anticancer agents in vivo and in vitro. The use of these herbs has attracted a great deal of attention as one of alternative cancer therapies from the viewpoint of less toxicity and cost benefit.

Ardipusilloside I is a compound which is isolated from the traditional Chinese medicine Ardisia pusilla A. DC. It has potential application to treat cancers, including glioblastoma (Tang et al. 2009; Zhang et al. 2010; Lin et al. 2008), Lewis pulmonary carcinoma and hepatocarcinoma SMMC-7721 (Tao et al. 2005). Ardipusilloside I had a good docking with VEGFR and could inhibit growth and induce apoptosis of NCl-H460 cell in a dose-dependent manner (Zhang et al. 2010). Cell cycle was significantly stopped at the G(1) phase. Ardipusilloside III, another compound isolated from Ardisia pusilla A. DC, not only markedly suppressed proliferation in a time- and dose-dependent manner, but also induced apoptosis in human glioblastoma U251MG cells. Both the intrinsic pathway of apoptosis, mediated by BAD dephosphorylation and cleavage, and the extrinsic pathway of apoptosis, mediated by caspase-8 and caspase-3 activation, were involved in ardipusilloside III-induced apoptosis (Lin et al. 2008). In the present study, we showed Ardipusilloside I reduced the survival of HCC cells dose-dependently, at least partially due to inhibition of Mek/Erk and Ala pathways. Meanwhile, Ardipusilloside I was able to effectively suppress invasion and metastasis of hepatocarcinoma by in vitro and in vivo. Previous researchers have shown that invasion and metastasis of solid tumors require the promotion of the dissolution of the surrounding tumor matrix and the basement membrane that is the action of tumor-associated proteases (Cho et al. 1997), and that matrix metalloproteinases (MMPs) play an important role in this process in carcinomas (Iwamoto et al. 2005). In this study, we showed Ardipusilloside I to reduce the activities of MMP2 and MMP9. The suppression of MMP2 and MMP9 expression and secretion would lead to less degradation of the ECM and basement membrane, thereby mobilizing growth factors that would inhibit cancer cell survival, cell migration and invasion (Farina et al. 1998). In short, our results demonstrated that Ardipusilloside I significantly inhibited the invasion of HCC cells by suppression of the activities of metalloproteinase (MMP) 2 and MMP 9.

In metastasis process of cancer cells, breaking away both from the extracellular matrix and from the cells around it is important. Cells are held together by variety of cell-to-cell adhesion molecules that expressed on the cell surfaces (Tammali et al. 2011). In general, the adhesion molecules seem to be missing or are compromised in cancer cells. E-cadherin, one subtype of Cadherins family, is the adhesion molecule found in mammalian cells which is an important factor in cell-cell adhesion. In cancer cells, E-cadherin expression is regularly impaired, results in the detachment of the cancer cells to from each other and from the matrix (Carico et al. 2010; Kondo et al. 2002). Mounting studies have proved that E-cadherin is important in cancer cells metastasis (Wendt et al. 2011). The expression of E-cadherin in HCC cells increased significantly after treated with Ardipusilloside I. Loss of cell-cell adhesion occurs as a result of reduced E-cadherin function. This leaded to lower detachment and migration of cells. Several studies have suggested that RhoGTPases, including Rac1 and Cdc42, are important in the establishment of mature epithelial cell-cell junctions. Rac1 and Cdc42 are key regulators of the actin cytoskeleton in coordinating cell migration and cell-cell adhesion (Fukata and Kaibuchi 2001; Fukata et al. 2002). They are required for cadherin-mediated cell-cell adhesion (Kitt and Nelson 2011). Cdc42 and Rac1 directly regulate the E-cadherin activity. Rac1 activity levels increased in whole cell populations at different times after initiation of cell-cell contacts, and was subsequently maintained at a high level upon E-cadherin adhesion. Our results showed that Ardipusilloside 1 treated HCC cells enhanced Rac1, not Cdc42 activity. Cell-cell contact after Ardipusilloside I may become tighter than untreated cells. As a consequence, metastasis was significantly reduced.

Taken together, we found Ardipusilloside I significantly inhibits survival, invasion and metastasis of HCC cells in vitro. The mechanism is achieved by suppression of Mek/Erk and Akt signaling pathways, downregulating the protein expressions of metalloproteinase (MMP)-2 and MMP-9. The activation of Raci by Ardipusilloside I resulted in upregulation of E-cadherin activity which subsequently increase cell-to-cell and cell-to-extracellular matrix adhesions, and leaded to significantly less metastasis. Further research is warranted to investigate whether Ardipusilloside I can be developed into a clinically effective and safe anticancer drug to treat liver cancer patients.

Abbreviations: MMP2, matrix metalloproteinase 2; MMP9, matrix metalloproteinase 9; Mek, mitogen-activated protein kinase kinase: Erk, extracellular signal-regulated kinases; Racl, Ras-related C3 botulinum toxin substrate 1; Cdec42, cell division control protein 42 homolog; ECM, extracellular matrix.

References

Braicu, C., Claudia, B., loana, B.-N., Ovidiu, B., Florin, G., 2009. Hepatocellular carcinoma: tumorigenesis and prediction markers. Gastroenterol. Res. 2, 9.

Carico, E., Atlante, M., Giarnieri, E., Raffa, S., Bucci, B., Giovagnoli, M.R., Vecchione, A., 2010. E-cadherin and alpha-catenin expression in normal, hyperplastic and neoplastic endometrium. Anticancer Res. 30, 4993-4997.

Cho, J.Y., Chung, H.C., Noh, S.H., Roh, J.K., Min, J.S., Kim, B.S., 1997. High level of urokinase-type plasminogen activator is a new prognostic marker in patients with gastric carcinoma. Cancer 79, 878-883.

Colin, D.W., Matthew, D.B., David, B., 2009. Sacks IQGAPs in cancer: a family of scaffold proteins underlying tumorigenesis. FEBS Lett. 583, 1817-1824.

Colombo, M., 2002. Malignant neoplasms of the liver. In: Schiff, E.R., Sorrel, M.F., Maddrey, W.C. (Eds.), Schiffs Disease of the Liver., 9th ed. Lippincott William & Wilkins, Philadelphia, pp. 1377-1403.

Farina, A.R., Coppa, A., Tiberio, A., Tacconelli, A., Turco, A., Colletta, G., Gulino, A., Mackay, A.R., 1998. Transforming growth factor-beta1 enhances the invasiveness of human MDA-MB-231 breast cancer cells by up-regulating urokinase activity. Intl. Cancer 75, 721-730.

Fukata, M., Kaibuchi, K., 2001. Rho-family GTPases in cadherin-mediated cell-cell adhesion. Nat. Rev. Mol. Cell. Biol. 2, 887-897.

Fukata, M., Nakagawa, M., Kuroda, S., Kaibuchi, K., 2002. Effects of Rho family GTPases on cell-cell adhesion. Methods Mol. Biol. 189, 121-128.

Greten, T.F., Korangy, F., Manns, M.P., Malek, N.P., 2009. Molecular therapy for the treatment of hepatocellular carcinoma. Br. J. Cancer 100, 19-23.

Iwamoto, J., Mizokami, Y., Takahashi, K., Nakajima, K., Ohtsubo, T., Miura, S., Narasaka, T., Takeyama, H., Omata, T., Shimokobe, IC, Ito, M., Takehara, H., Matsuoka, T., 2005. Expressions of urokinase-type plasminogen activator, its receptor and plasminogen activator inhibitor-1 in gastric cancer cells and effects of Helicobacter pylori. Scand. J. Gastroenterol. 40, 783-793.

Kitt, K.N., Nelson, W.J., 2011. Rapid suppression of activated Racl by cadherins and nectins during de novo cell-cell adhesion. PLoS One 6, e17841.

Rondo, S., Kubota, S., Shimo, T., Nishida, T., Yosimichi, G., Eguchi, T., Sugahara, T., Takigawa, M., 2002. Connective tissue growth factor increased by hypoxia may initiate angiogenesis in collaboration with matrix metalloproteinases. Carcinogenesis 23. 769-776.

Lin, H., Zhang, X., Cheng, G., Tang, H.F., Zhang, W., Zhen, Cheng, J.X., Liu, B.L., Cao, W.D., Dong, W.P., Wang, P., 2008. Apoptosis induced by ardipusilloside Ill through BAD dephosphorylation and cleavage in human glioblastoma U251 MG cells. Apoptosis 13, 247-257.

Parkin, D.M., Bray, F., Ferlay, J., 1999. Global cancer statistics. CA Cancer J. Clin. 55, 74-108.

Shih, Y.W., Shieh, J.M., Wu, P.F., Lee, Y.C., Chen, Y.Z., Chiang, T.A., 2009. Alphatomatine inactivates PI3K/Akt and ERIC signaling pathways in human lung adenocarcinoma A549 cells: effect on metastasis. Food Chem. Toxicol. 47, 1985-1995.

Sun, F.X., Tang, Z.Y., Lui, K.D., Ye, S.L., Xue, Q., Gao, D.M., Ma, Z.C., 1996. Establishment of a metastatic model of human hepatocellular carcinoma in nude mice via orthotopic implantation of histologically intact tissues. Int. J. Cancer 66. 239-243.

Sung, H.J., Choi, S.M., Yoon, Y., An, KS., 1999. Tanshinone IIA, an ingredient of Salvia miltiorrhiza BUNGE, induces apoptosis in human leukemia cell lines through the activation of caspase-3. Exp. Mol. Med. 31, 174-178.

Tammali, R., Reddy, A.B., Saxena, A., Rychahou, P.G., Evers, M.B., Qiu, S., Awasthi, S., Ramana, K.V., Srivastava, S.K., 2011. Inhibition of aldose reductase prevents colon cancer metastasis. Carcinogenesis (June) [Epub ahead of print].

Tang, H.F., Yun, J., Lin, H.W., Chen, X.L., Wang, X.J., Cheng, G., 2009. Two new triterpenoid saponins cytotoxic to human glioblastoma U251 MG cells from Ardisia pusilla. Chem. Biodivers 6, 1443-1452.

Tao, X., Wang, P., Yang, X., Yao, H., Liu, J., Cao, Y., 2005. Inhibitory effect of ardipusilloside-I on Lewis pulmonary carcinoma and hepatocarcinoma SMMC7721. Zhong Yao Cai 28, 574-577.

Tian, J., Tang, Z.Y., Ye, S.L., Liu, Y.K., Lin, Z.Y., Chen, J., Xue, Q., 1999. New human hepatocellular carcinoma (HCC) cell line with highly metastatic potential (MHCC97) and its expressions of the factors associated with metastasis. Br. J. Cancer 81, 814-821.

Yoon, Y., Kim, Y.O., Jeon, W.K., Park, H.J., Sung, H.J., 1999. Tanshinone IIA isolated from Salvia miltiorrhiza BUNGE induced apoptosis in H L60 human premyelocytic leukemia cell line. J. Ethnopharmacol. 68,121-127.

Yuan, S.L., Huang, R.M., Wang, X.J., Song, Y., Huang, G.Q., 1998. Reversing effect of Tanshinone on malignant phenotypes of human hepatocarcinoma cell line. World J. Gastroenterol. 4, 317-319.

Yuan, S.L., Wei, Y.Q., Wang, X.J., Xiao, F., Li, S.F., Zhang, J., 2004. Growth inhibition and apoptosis induction of tanshinone II-A on human hepatocellular carcinoma cells. World J. Gastroenterol. 10, 2024-2028.

Wang, Z., Jin, H., Xu, R., Mei, Q., Fan, D., 2009. Triptolide downregulates Rac1 and the JAK/STAT3 pathway and inhibits colitis-related colon cancer progression. Exp. Mol. Med. 41, 717-727.

Wendt, M.K., Taylor, M.A., Schiemann, B.J., Schiemann, W.P., 2011. Downregulation of epithelial cadherin is required to initiate metastatic outgrowth of breast cancer. Mol. Biol. Cell (May) [Epub ahead of print].

Weifeng, T., Feng, S., Xiangji, L, Changqing, S., Zhiquan, Q, Huazhong, Z., Peining, Y., Yong, Y., Mengchao, W., Xiaoqing, J., Wan-Yee, L, 2011. Artemisinin inhibits in vitro and in vivo invasion and metastasis of human hepatocellular carcinoma cells. Phytomedicine 18, 158-162.

Zhang, Y., Qu, Y., Zhang, J., Wang, X., 2010. Ardipusilloside I purified from Ardisia pusilla competitively binds VEGFR and induces apoptosis in NC1-H460 cells. Phytomedicine 17, 519-526.

Zhong, J., Gencay, M.M., Bubendorf, L., Burgess, J.K., Parson, H., Robinson, B.W., Tamm, M., Black, J.L., Roth, M., 2006. ERK1/2 and p38 MAP kinase control MMP-2, MT1-MMP, and TIMP action and affect cell migration: a comparison between mesothelioma and mesothelial cells. J. Cell. Physiol. 207, 540-552.

(1.) These two authors contributed equally to this work.

Lianqing Lou (a), *, Weiwei Ye (a), Yongxin Chen (a), Shuang Wu (a), Linzheng Jin (a), Jinke He (a), Xingfei Tao (a), Jinghong Zhu (a), Xiangyi Chen (a), Anmei Deng (b), (1), Jinhe Wang (a), (l)

(a.) Yiwu Central Hospital, 519 Nanmen Street, Yiwu, Jinhua, Zhejiang 322000. China

(b.) Department of Laboratory Diagnosis, Changhai Hospital, Second Military Medical University, Shanghai 200433, PR China

* Corresponding author.

E-mail address: [e-mail] (L. Lou).

0944-7113/$ - see front matter [c] 2012 Elsevier GmbH. All rights reserved.

doi:10.1016/j.phymed.2012.01.003