Food and Chemical Toxicology 47 (2009) 1985–1995

Contents lists available at ScienceDirect

Food and Chemical Toxicology

journal homepage: www.elsevier .com/locate / foodchemtox

a-Tomatine inactivates PI3K/Akt and ERK signaling pathways in humanlung adenocarcinoma A549 cells: Effect on metastasis

Yuan-Wei Shih a, Jiunn-Min Shieh b, Pei-Fen Wu c, Yi-Chieh Lee a, Yi-Zhi Chen a, Tai-An Chiang d,*

a Department of Biological Science and Technology and Graduate Institute of Biomedical Science, Chung Hwa University of Medical Technology, Tainan 717, Taiwanb Department of Pulmonary Medicine, Chi Mei Medical Center, No. 901, Chung-Hua Road, Yong-Kang, Tainan 710, Taiwanc Department of Occupational Safety and Hygiene, Tajen University, Pingtung 907, Taiwand Department of Medical Technology and Graduate Institute of Biological Science and Technology, Chung Hwa University of Medical Technology, Tainan 717, Taiwan

a r t i c l e i n f o

Article history:Received 6 January 2009Accepted 12 May 2009

Keywords:a-TomatineInvasionMigrationPI3K/AktERK

0278-6915/$ - see front matter Crown Copyright � 2doi:10.1016/j.fct.2009.05.011

Abbreviations: MMPs, matrix metalloproteinasesminogen activator; ECM, extracellular matrix; MAPkinase; ERK, extracellular signaling-regulating kinasekinase/stress-activated protein kinase; PI3K, phospphosphatase and tensin homolog deleted on chromosokappa B; AP-1, activator protein-1, IjB, Inhibitor of N

* Corresponding author. Tel.: +886 6 2674567x450E-mail address: giantful@mail.hwai.edu.tw (T.-A. C

a b s t r a c t

This study first investigates the anti-metastastic effect of a-tomatine in the human lung adenocarcinomacell line: A549. In this study, we first noted a-tomatine inhibited A549 cells invasion and migration bywound-healing assay and Boyden chamber assay. The data also showed a-tomatine could inhibit phos-phorylation of Akt and extracellular signal-regulated kinase 1 and 2 (ERK1/2), which is involved in theup-regulating matrix metalloproteinase-2 (MMP-2), matrix metalloproteinase-9 (MMP-9) or urokinase-type plasminogen activator (u-PA), whereas it did not affect phosphorylation of c-Jun N-terminal kinase(JNK) and p38. Next, a-tomatine significantly decreased the nuclear levels of nuclear factor kappa B(NF-jB), c-Fos, and c-Jun. Also, treating A549 cells with a-tomatine also leads to a dose-dependent inhi-bition on the binding abilities of NF-jB and activator protein-1 (AP-1). Further, the treatment of inhibitorsspecific for PI3K (Wortmannin) or ERK (U0126) to A549 cells could cause reduced activities of MMP-2,MMP-9, and u-PA. These results showed a-tomatine could inhibit the metastatic ability of A549 cells byreducing MMP-2, MMP-9, and u-PA activities through suppressing phosphoinositide 3-kinase/Akt(PI3K/Akt) or ERK1/2 signaling pathway and inhibition NF-jB or AP-1 binding activities. These findingsproved a-tomatine might be an anti-metastastic agent against human lung adenocarcinoma.

Crown Copyright � 2009 Published by Elsevier Ltd. All rights reserved.

1. Introduction One of the glycoalkaloids, a-tomatine, occurs naturally in toma-

Lung cancer is the major cause of malignancy-related deathsworldwide, and its incidence is rising in many countries (Greenleeet al., 2001). About 40% of lung cancers are adenocarcinomas. Ade-nocarcinomas belonging to the subgroup of the non-small cell lungcancers are the most common type in US and Asia (Shivapurkaret al., 2003). Studies have shown lung cancer cases are caused bysmoking, air pollution, environmental risk factors (for example,exposure to radiation, asbestos, heavy mental, and polycyclic aro-matic hydrocarbon) and oncogene (for example, slug gene) (Leeet al., 2005; Shih et al., 2005). Most diagnosed patients with lungadenocacinoma are in an advanced stage because of its highly met-astatic properties, and such patients are not candidates for surgicalresection.

009 Published by Elsevier Ltd. All r

; u-PA, urokinase-type plas-K, mitogen-activated protein; JNK/SAPK, c-Jun N-terminalhoinositide 3-kinase; PTEN,me 10; NF-jB, nuclear factorF-jB.; fax: +886 6 2605598.hiang).

toes (Lycopersicon esculentum). Immature green tomatoes containup to 500 mg a-tomatine/kg fresh fruit weight. The compound ispartly degraded as the tomato ripens until at maturity levels inred tomatoes are about 5 mg/kg fresh fruit weight (Friedman andLevin, 1995). Fig. 1A shows a-tomatine is constructed of an aglyconmoiety (tomatidine), and a tetrasaccharide moiety (b-lycotetrose)that contains two molecules of D-glucose and one each of D-galact-ose and D-xylose. Previous studies demonstrated a-tomatine hasexhibited anti-proliferative and apoptotic effects on the growthof cancer cells originating from the human colon and liver (Leeet al., 2004). Although it was quite clear a-tomatine may inhibitthe growth of various cancers by inducing cancer cells towardapoptosis and anti-proliferation, whereas the precise impact andrelated molecular mechanism of a-tomatine on metastasis of can-cer cells was still unclear.

Metastasis is a multistep process involving overexpression ofproteolytic enzymes, such as matrix metalloproteinases (MMPs)and u-PA. MMP-2 and MMP-9 (also known as type IV collagenasesor gelatinases) which can degrade most ECM components thatforming the basal membrane (Bernhard et al., 1994). In addition,u-PA may initiate the activation of an enzymatics cascade and con-vert the zymogen plasminogen to plasmin. The activation of these

ights reserved.

Fig. 1. Effect of a-tomatine on the viability in A549 cells. (A) Chemical structure ofglycoalkaloid a-tomatine isolated from the leaves and fruits of tomato (Lycopersiconesculentum). (B) Cells (4 � 104 cells/ml) were treated with various concentrations(0, 1, 1.5, 2, 2.5, 3, and 3.5 lM) of a-tomatine for 24 and 48 h. Cell viability wasdetermined by MTT assay. The cell viability was directly proportional to theproduction of formazan, which was measured spectrophotometrically at 563 nm.Values are expressed as mean ± SD of three independent experiments. **p < 0.01,***p < 0.001 compared with the untreated control (dose 0).

1986 Y.-W. Shih et al. / Food and Chemical Toxicology 47 (2009) 1985–1995

enzymes enable the degradation of extracellular matrix (ECM) bytumor cells, allowing their access to the vasculature, migrationand invasion into the target organ and development of tumormetastasis (Duffy and Duggan, 2004; Itoh and Nagase, 2002).

As well as MMPs and u-PA, the mitogen-activated protein ki-nases family members (MAPK) are also known to mediate metasta-sis. The MAPK serine/threonine kinase superfamily is activated bynumerous extracellular stimuli and is involved in signal transduc-tion cascades playing an important regulatory role in cell growth,differentiation, apoptosis, and metastasis (Chan-Hui and Weaver,1998). Three major mammalian MAP kinases have been described:ERK1/2 or p44/42 MAPK, c-Jun N-terminal kinase/stress-activatedprotein kinase (JNK/SAPK), and p38 MAPK. The diverse MAP kinasemembers are activated in response to different extracellular stimuliand have distinct downstream targets, thus serving different rolesin cellular responses. ERK1/2, p38 MAPK, and JNK/SAPK play a cen-tral role in regulating the expression of MMPs and u-PA (Chen et al.,2005; Kwon et al., 2008; Lee et al., 2008). In addition, PI3K/Akt sig-nal transduction pathway regulates the cell metastasis of non-smallcell lung cancer (NSCLC) and is closely associated with the develop-ment and progression of various tumors. Overexpression of PI3Kand low expression of phosphatase and tensin homolog deletedon chromosome 10 (PTEN) are closely correlated with the develop-ment, invasion and metastasis of NSCLC (Liao et al., 2006).

NF-jB is a multisubunit transcription factor, which is involvedin immune response, inflammation and malignant transformation.The active NF-jB consists of p50, p52, p65 (RelA), Rel B, and c-Rel.under normal condition, NF-jB is maintained in the cytoplasmthrough interactions with an inhibitor of NF-jB (IjB), but upondissociation, moves into the nucleus and promotes cancer cellsproliferation, angiogenesis and metastasis. AP-1 is a nuclear tran-scription, which is involved in cell proliferation, differentiation,

apoptosis and neoplastic transformation. AP-1 consists of homodi-mers and heterodimers of members from Fos (c-Fos, Fos B, Fra-1,and Fra-2) and Jun (c-Jun, Jun B, and Jun D) families (Karin andBen-Neriah 2000; Lee et al., 2007). Previous papers have showedthe MMP-2, MMP-9, and u-PA promoters are coordinately regu-lated by both NF-jB and AP-1 (Aguirre Ghiso et al., 1999; Maet al., 2004; Westermarck and Kahari, 1999). As with NF-jB andAP-1 regulates the expression of matrix metalloproteinase and u-PA, consistent with a role for this protein in regulating metastasis.Further, to establish anti-metastastic mechanism of a-tomatine,the objective of this work was to examine the inhibitory effectsand the related signaling pathways of a-tomatine on the inva-sion/migration of human lung adenocarcinoma A549 cells in vitro.

2. Materials and methods

2.1. Chemicals and reagents

a-tomatine (purity >97%), DMSO, Tris–HCl, EDTA, SDS, phenylmethylsulfonylfluoride, bovine serum albumin (BSA), gelatin, casein, plasminogen, leupeptin, Non-idet P-40, deoxycholic acid, sodium orthovanadate, wortmannin, and U0126 werepurchased from Sigma–Aldrich (St. Louis, MO); the protein assay kit was obtainedfrom Bio-Rad Labs. (Hercules, CA). Dulbecco’s phosphate buffer solution (PBS), tryp-sin-EDTA, and powdered Dulbecco’s modified Eagle’s medium (DMEM) were pur-chased from Gibco/BRL (Gaithersburg, MD). Matrigel was from BD Biosciences(Bedford, MA). Antibody against Akt, MAPK/ERK1/2, JNK/SAPK and p38 MAPK, pro-teins and phosphorylated proteins were purchased from Cell Signaling Tech. (Bev-erly, MA). PI3K (p85), NF-jB (p65), c-Fos, c-Jun, b-actin, and C23 antibodies werefrom BD Transduction Laboratories (San Diego, CA). The enhanced chemilumines-cence (ECL) kit was purchased from Amersham Life Science (Amersham, UK).

2.2. Cell culture and a-tomatine treatment

A549, a human lung adenocarcinoma cell line, was obtained from BCRC (FoodIndustry Research and Development Institute in Hsin-Chu, Taiwan). Cells were cul-tured in DMEM supplemented with 10% fetal calf serum, 2 mM L-glutamine, 100 U/ml of penicillin, 100 mg/ml streptomycin mixed antibiotics and 1 mM sodium pyru-vate. All cell cultures were maintained at 37 �C in a humidified atmosphere of 5%CO2–95% air. The culture medium was renewed every 2–3 days. Adherent cells weredetached by incubation with trypsin. For a-tomatine treatment, the stock solutionof a-tomatine was dissolved in dimethyl sulfoxide (DMSO) and sterilized by filtra-tion through 0.2 lm disc filters. Suitable amounts of stock solution (1 mg/ml inDMSO) of a-tomatine were added into the cultured medium to achieve the indi-cated concentrations (Final DMSO concentration was less than 0.2%) and then incu-bated with cells for the indicated time periods.

2.3. Analysis of cell viability (MTT assay)

To evaluate the cytotoxicity of a-tomatine, an MTT [3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyl-tetrazolium bromide] assay was performed to determine cell via-bility (Mosmann, 1983). Briefly, cells were seeded at a density of 4 � 104 cells/ml ina 24-well plate for 24 h. Then, the cells were treated with a-tomatine at variousconcentrations (0, 1, 1.5, 2, 2.5, 3, and 3.5 lM) for various periods of time (24 and48 h). Each concentration was repeated three times. After the exposure period,the medium was removed that was followed by washing the cells with PBS. Then,the medium was changed and incubated with MTT solution (5 mg/ml)/well for 4 h.The medium was removed, and formazan was solubilized in isopropanol and mea-sured spectrophotometrically at 563 nm. The percentage of viable cells was esti-mated by comparing with untreated control cells.

2.4. Analysis of MMP-2, MMP-9 and u-PA activity (zymography assay)

The activities of MMP-2 and MMP-9 were assayed by gelatin zymography as de-scribed previously (Chu et al., 2004). Briefly, conditioned media from cells culturedin the absence of serum for 24 h were collected. Samples were mixed with loadingbuffer and electrophoresed on 8% SDS–polyacrylamide gel containing 0.1% gelatin.Electrophoresis was performed at 140 V and 110 V for 3 h. Gels were then washedtwice in zymography washing buffer (2.5% Triton X-100 in double-distilled H2O) atroom temperature to remove SDS, followed by incubation at 37 �C for 12–16 h inzymography reaction buffer (40 mM Tris–HCl (pH 8.0), 10 mM CaCl2, 0.02%NaN3), stained with Coomassie blue R-250 (0.125% Comassie blue R-250, 0.1% ami-no black, 50% methanol, 10% acetic acid) for 1 h and destained with destaining solu-tion (20% methanol, 10% acetic acid, 70% double-distilled H2O). Non-staining bandsrepresenting the levels of the latent form of MMP-2 and MMP-9 were quantified bydensitometer measurement using a digital imaging analysis system.

Y.-W. Shih et al. / Food and Chemical Toxicology 47 (2009) 1985–1995 1987

Visualization of u-PA activity was performed by casein-plasminogen zymogra-phy. Briefly, 2% casein and 20 lg/ml plasminogen were added to 8% SDS–PAGE gel.Samples with a total protein of about 20 lg were then loaded onto the gels. Theu-PA activity of cells treated or untreated with a-tomatine was measured as de-scribed in the gelatin zymography section.

2.5. Wound-healing assay

For cell motility determination, A549 cells (1 � 105 cells/ml) were plated in six-well tissue culture plate and grown to 80–90% confluence. After aspirating the med-ium, the centers of the cell monolayers were scraped with a sterile micropipette tipto create a denuded zone (gap) of constant width. Subsequently, cellular debris waswashed with PBS, and A549 cells were exposed to various concentrations of a-tom-atine (0, 1, 1.5, and 2 lM). The wound closure was monitored and photographed at0, 12, 24, 36, and 48 h with an Olympus CKX-41 inverted microscope and an Olym-pus E-410 camera. To quantify migrated cells, pictures of the initial wounded mon-olayers were compared with the corresponding pictures of cells at the end of theincubation. Artificial lines fitting the cutting edges were drawn on pictures of theoriginal wounds and overlaid on the pictures of cultures following incubation. Mi-grated cells across the white lines were counted in six random fields from each trip-licate treatment, and the data were presented as mean ± SD.

2.6. Boyden chamber invasion and migration assay

The ability of A549 cells to pass through Matrigel-coated filters was measuredby the Boyden chamber invasion assay (Ochi et al., 1993). Matrigel was diluted to200 lg/ml with cold filtered distilled water and applied to the top side of the 8-lm pore polycarbonate filter. Briefly, A549 cells were treated with various con-centrations of a-tomatine. After 48 h, cells were detached by trypsin and resus-pended in serum-free medium. Medium containing 10% FBS-medium wasapplied to the lower chamber as chemoattractant, and then the cells were seededon the upper chamber at a density of 1 � 105 cells/well in 50 ll of serum-freemedium. The chamber was incubated for 8 h at 37 �C. At the end of incubation,the cells in the upper surface of the membrane were carefully removed with acotton swab and cells invading across the Matrigel to the lower surface of themembrane were fixed with methanol and stained with 5% Giemsa solution. Theinvasive cells on the lower surface of the membrane filter were counted with alight microscope. The data are presented as the average number of cells attachedto the bottom surface from randomly chosen fields. Each experiment was carriedout in triplicate.

To measure the ability of A549 cells on migration, cells were seeded into a Boy-den chamber with 8 lm pore polycarbonate filters which were not coated withMatrigel. The migration of cells was treated with various concentrations of a-tom-atine. The migration assay was measured as described in the invasion assay.

2.7. Preparation of whole-cell lysates and nuclear extracts

The cells were lysed with iced-cold RIPA buffer (1% NP-40, 50 mM Tris–base, 0.1% SDS, 0.5% deoxycholic acid, 150 mM NaCl, pH 7.5) and then the fol-lowing were added phenylmethylsulfonyl fluoride (10 mg/ml), leupeptin (17 mg/ml), and sodium orthovanadate (10 mg/ml). After vortexing for 30 min on ice,the samples were centrifuged at 12,000 � g for 10 min, and then the superna-tants were collected, denatured, and subjected to SDS–PAGE and Western blot-ting. Nuclear extracts were prepared as previously described (Hoppe-Seyleret al., 1991) and then used for NF-jB, c-Fos, c-Jun, and AP-1 detection. Each nu-clear pellet was resuspended in nuclear extract buffer (1.5 mM MgCl2, 10 mMHEPES, pH 7.9, 0.1 mM EDTA, 0.5 mM dithiothreitol, 0.5 mM phenylmethylsul-fonyl fluoride, 25% glycerol, and 420 mM NaCl). The nuclear suspension wasincubated on ice for 20 min and then centrifuged at 14,000 � g for 5 min. Thesupernatant (corresponding to the soluble nuclear fraction) was saved, andthe remaining pellet was solubilized by sonication in PBS. The protein contentwas determined with Bio-Rad protein assay reagent using bovine serum albu-min as a standard.

2.8. Western blotting assay

To analyze the migration-related proteins, Western blotting was performed asfollows. The denatured samples (50 lg purified protein) were resolved on 10–12%SDS–PAGE gels. Proteins were then transferred onto nitrocellulose membranes.Non-specific binding of the membranes was blocked with Tris-buffered saline(TBS) containing 1% (w/v) non-fat dry milk and 0.1% (v/v) Tween-20 (TBST) for morethan 2 h. Membranes were washed with TBST three times for 10 min and incubatedwith a suitable dilution of specific primary antibodies in TBST overnight at 4 �C.Subsequently, the membranes were washed with TBST and incubated with anappropriate secondary antibody (horseradish peroxidase-conjugated goat anti-mouse or antirabbit IgG) for 1 h. After washing the membrane three times for10 min in TBST, the band detection was revealed by enhanced chemiluminescenceusing ECL Western blotting detection reagents and exposed ECL hyperfilm in a UVPLuminescent image analyzer.

2.9. Analysis of NF-jB and AP-1 binding assay (electrophoretic mobility shift assay)

Cell nuclear proteins were extracted with a nuclear extract buffer and measuredby an electrophoretic mobility shift assay (EMSA) (Ma et al., 2001). Cells (1 � 105/ml) were collected in PBS buffer (pH 7.4) and centrifuged at 2000 � g for 5 min at4 �C. Cells were lysed with buffer A (10 mM HEPES, 1.5 mM MgCl2, 10 mM KCl,0.5 mM DTT, and 0.5 mM PMSF (pH 7.9) containing 5% NP-40) for 10 min on ice,and this was followed by vortexing to shear the cytoplasmic membranes. The ly-sates were centrifuged at 2000 � g for 10 min at 4 �C. The pellet containing the nu-clei was extracted with high salt buffer B (20 mM HEPES, 420 mM NaCl, 1.5 mMMgCl2, 0.5 mM DTT, 0.5 mM PMSF, 0.2 mM EDTA, and 25% glycerol) for 15 min onice. The lysates were clarified by centrifuge at 13,000 � g for 10 min at 4 �C. Thesupernatant containing the nuclear proteins was collected and frozen at �80 �C un-til use. The protein content of nuclear fractions was determined with Bio-Rad pro-tein assay. Five microgram aliquot of nuclear proteins were mixed with eitherbiotin-labeled NF-jB or AP-1 oligonucleotide probes for 15 min at room tempera-ture or with oligonucleotides containing (sense of NF-jB, 50-AGTTGAGGGGACTTTCCCAGGC-30 , antisense of NF-jB, 30-TCAACTCCCCTGAAAGGGTCCG-50; sense ofAP-1, 50-CG CTTGATGACTCAGCCGGAA-30 , antisense of AP-1, 30-GCGAACTACT-GAGTCGGCCTT). DNA probes were added to 10 ll binding reactions containing dou-ble-distilled H2O, 5 lg nuclear protein, 1 ll poly (dI-dC), 1 ll biotin-labeled double-stranded NF-jB or AP-1 oliginucleotides and 2 ll of 10-fold binding buffer into amicrocentrifuge tube and were incubated for 15 min at room temperature. Specificcompetition binding assays were performed by adding 200-fold excess of unlabeledprobe as a specific competitor. Following protein-DNA complexes formation, sam-ples were loaded on a 6% non-denaturing polyacrylamide gel in 0.5 � TBE bufferand were then transferred to positively charged nitrocellulose membranes (Mili-pore, Bedford, MA) by a transfer blotting apparatus and cross-linked in a Stratagenecross-linker. Gel shifts were visualized with streptavidin-horseradish peroxidasefollowed by chemiluminescent detection.

2.10. Statistical analysis

Data were expressed as mean ± SD of three independent experiments and ana-lyzed by Student’s t-test (Sigmaplot 2001). Significant differences were establishedat p 6 0.05.

3. Results

3.1. Cytotoxicity of a-tomatine in A549 cells

We first assayed the cytotoxicity of a-tomatine by treatingA549 cells with a-tomatine at various concentrations (0, 1, 1.5, 2,2.5, 3, and 3.5 lM) for 24 and 48 h followed by MTT assay. Asshown in Fig. 1B, a-tomatine showed a dose- and time-dependentinhibitory effect on the growth of A549 cells. Compared to 0 lM(DMSO was treated alone, data not shown), after 24 h and 48 htreatment with a-tomatine at a concentration between 0 to 2 lMwas not significantly altered, indicating that a-tomatine was nottoxic to A549 cells at these dosages. When cells were treated with2.5–3.5 lM a-tomatine for 24 and 48 h, cell viability was signifi-cantly decreased. These results demonstrated the treatment of a-tomatine with doses higher than 2 lM for 24 and 48 h resultedin dose- and time-dependent loss of cell viability in A549 cells,but doses lower than 2 lM for 24 and 48 h did not cause cytotox-icity. In the following experiments, these doses below 2 lM of a-tomatine were applied in all subsequent experiments.

3.2. a-Tomatine inhibits the activation of MMP-2, MMP-9, and u-PA inA549 cells

For the cell migration and invasion processes, pointing to theinevitable involvement of matrix-degrading proteinases, the ef-fects of a-tomatine on MMP-2, MMP-9, and u-PA activities wereinvestigated by gelatin and casein zymography. The conditionedmedia were collected, concentrated, and the inhibition of metasta-sis was measured after A549 cells were treated for 24 h by a-tom-atine. As shown in Fig. 2A, dose-dependent and markedly reducedMMP-2 and MMP-9 activities were noted in the serum-free med-ium treated with 2 lM a-tomatine for 24 h. Similarly, u-PA activitywas also inhibited in a dose-dependent manner by a-tomatinetreatment (Fig. 2B). These results suggested the anti-metastatic

Fig. 2. Effect of a-tomatine on MMP-2/MMP-9 and u-PA activities in A549 cells. Cells were treated with various concentrations (0, 1, 1.5, and 2 lM) of a-tomatine for 24 h.The conditioned media were collected, and then (A) MMP-2/MMP-9 and (B) u-PA activities were determined by gelatin zymography or casein zymography. MMP-2/MMP-9and u-PA activities were quantified by densitometric analysis. The densitometric data were expressed as mean ± SD of three independent experiments. *p < 0.05, **p < 0.01,***p < 0.001 compared with the untreated control (dose 0).

1988 Y.-W. Shih et al. / Food and Chemical Toxicology 47 (2009) 1985–1995

effect of a-tomatine was related to the inhibition of the enzymat-ically degradative processes of tumor metastasis. This study gives afirst glimpse to demonstrate a-tomatine reduced the metastasis inhuman lung adenocarcinoma cells. The activities of MMP-2, MMP-9, and u-PA have been shown to play a critical role in degrading thebasement membrane in cancer invasion and migration.

3.3. a-Tomatine inhibits the migration and invasion in A549 cells

To investigate the inhibitory effect of a-tomatine on A549 cellsmigration and invasion process, a wound-healing assay and a Boy-den chamber assay were used. In wound-healing assay, the conflu-ent monolayer was scraped with a sterile micropipette tip to createa scratch wound. After incubation with 1.5 and 2 lM of a-tomatinefor 24 and 48 h, the cells migrated to the denuded zone, and theywere counted. The results demonstrated a-tomatine dose-depen-dently suppressed A549 cell migration to the denuded zone.According to a quantitative assessment, treatment with 1.5 and2 lM of a-tomatine inhibited 60% and 69% of cell migration after24 h, respectively; and such doses of a-tomatine inhibited 45%and 55% of cell migration at 48 h, respectively. Also, the cells weretreated with various concentrations of a-tomatine for 0, 12, 24, 36,and 48 h. The results showed 2 lM of a-tomatine exhibited themost inhibiting effect on cell motility after 48 h incubation. Espe-cially, compared with the untreated cells, the level of A549 cellsnumber decreased almost 2.2-fold with the treatment of 2 lM a-tomatine for 48 h (Fig. 3A). Although the cells were not treatedby a-tomatine, the number of cells in migration property wasincreased with increasing time (Because A549 cells is under non-cytotoxic concentrations). Also, the cells were treated with non-a-tomatine for 24, 36 and 48 h compared with the untreatedcontrol (dose 0), the results was presented significantly differentand defined as constituting statistical significance. These results

revealed that a-tomatine significantly inhibited the motility ofA549 cells.

One important characteristic of metastasis is the migratory andinvasive ability of tumor cells. We used Boyden chamber assay toquantify the migratory and invasive potential of A549 cells. The re-sults showed a-tomatine induced a dose-dependent decrease inmigration with increasing concentrations of a-tomatine (Fig. 3B).At 1.5 lM, the migration was reduced to 54.9% and at 2 lM themigration was reduced to less than 43%. Subsequently, a-tomatinealso induced a dose-dependent decrease in invasion with increas-ing concentrations of a-tomatine (Fig. 3C). At 1.5 lM the invasionwas reduced to 68.1% and at 2 lM the invasion was reduced to lessthan 51%. The results demonstrated a-tomatine significantly inhib-ited the migration and invasion of A549 cells.

3.4. a-Tomatine inhibits phosphorylation of ERK and Akt

Since we have shown treatment of A549 cells with a-tomatineinhibited the cell metastasis and activities of MMP-2, MMP-9, andu-PA, the underlying mechanisms were further investigated. Sev-eral studies have indicated the transcription factors (for example,NF-kB, c-Fos, and c-Jun), JNK1/2, ERK1/2, p38 MAPK, and Akt thatare involved in activity of MMP-2, MMP-9, and u-PA on differentcell types (Chen et al., 2005; Turner et al., 2007). To assess whethera-tomatine mediates and/or inhibits phosphorylation of JNK1/2,ERK1/2, p38 MAPK, Akt, and the protein level of PI3K, we investi-gated the effect of a-tomatine on the phosphorylated status ofMAPK family members (JNK1/2, ERK1/2, and p38 MAPK) and Aktin A549 cells which were treated with various concentrations ofa-tomatine for 3 h and 2 lM of a-tomatine for various periods oftime (0, 1, 2, 3, and 6 h). Fig. 4A and B showed a-tomatine signifi-cantly inhibited the activation of ERK1 and ERK2 as shown bydecreasing the phosphorylation of ERK1 and ERK2. In contrast,

Fig. 3. Effect of a-tomatine on the migration and invasion in A549 cells. (A) In wound-healing assay, A549 cell monolayers were scraped by a sterile micropipette tip and thecells were treated with various concentrations (0, 1, 1.5, and 2 lM) of a-tomatine for 0, 12, 24, 36, and 48 h. The number of cells in the denuded zone was quantitated afterindicated times (0, 12, 24, 36, and 48 h) by inverted microscopy. White lines indicate the wound edge. Pictures only were presented 24 and 48 h. Migrated cells across thewhite lines were counted in six random fields from each treatment. (B) In Boyden chamber migration assay, cells were treated with various concentrations of fisetin for 48 h,then cell migration were measured by Boyden chamber for 6 h with polycarbonate filters (pore size, 8 lm); (C) In Boyden chamber invasion assay, cells were treated withvarious concentrations of fisetin for 48 h, then cell invasion were measured by Boyden chamber for 8 h; polycarbonate filters (pore size, 8 lm) were precoated with Matrigel.Migration and invasion ability of A549 cells were quantified by counting the number of cells that invaded to the underside of the porous polycarbonate membrane undermicroscopy. Values are expressed as mean ± SD of three independent experiments. *p < 0.05, **p < 0.01, ***p < 0.001 compared with the untreated control (dose 0); ##p < 0.01,###p < 0.001 compared with the 0-h treated time.

Y.-W. Shih et al. / Food and Chemical Toxicology 47 (2009) 1985–1995 1989

1990 Y.-W. Shih et al. / Food and Chemical Toxicology 47 (2009) 1985–1995

a-tomatine did not significantly affect phospho-JNK1/2 and phos-pho-p38 activity (Fig. 4C–F). In addition, a-tomatine inhibitedthe protein level of PI3K and phosphorylation of Akt in a dose-and time-dependent manner (Fig. 4G and H).

To further investigate whether the inhibition of a-tomatine wasmainly occurred through the inhibition of ERK1/2 or PI3K/Akt sig-naling pathway, A549 cells were pretreated with a PI3K inhibitor(Wortmannin; 5 or 10 lM) or ERK inhibitor (U0126; 10 or20 lM) for 1 h and then incubated in the present or absence ofa-tomatine (1 lM) for 24 h. Results of gelatin zymography assayhas shown that a sole treatment of Wortmannin (5 or 10 lM)and U0126 (10 or 20 lM) or a-tomatine separately reduced theexpressions of MMP-9 or MMP-2 or by 8.7%, 28.3%, 14%, 34.3%and 12.3% or 4.7%, 14.2%, 7.3%, 30% and 3.3%, respectively, andthe combination treatment could even dramatically reduced thesecretions of MMP-9 or MMP-2 by 50.7% or 25% (5 lM Wortman-nin + 1 lM a-tomatine), 63.3% or 40% (10 lM U0126 + 1 lM a-tomatine) and 79.2% or 62.5% (5 lM Wortmannin + 10 lM

Fig. 4. Dose- and time-dependent effect of a-tomatine on the phosphorylation of ERK, JNKand G), A549 cells were treated with various concentrations (0, 1, 1.5, and 2 lM) of a-tom2 lM of a-tomatine for 0, 1, 2, 3, and 6 h. Activities of ERK phosphorylation, ERK, JNK pexpression of PI3K were analyzed by Western blotting. b-Actin was used as a loading con**p < 0.01, ***p < 0.001 compared with the untreated control (dose 0).

U0126 + 1 lM a-tomatine) (Fig. 5A). Similarly, in a casein zymog-raphy assay, a sole treatment of Wortmannin (5 or 10 lM) andU0126 (10 or 20 lM) or a-tomatine reduced the expression of u-PA by 12%, 27.5%, 14.5%, 32% and 15%, respectively, and the com-bination treatment could further reduce the secretion of u-PA by67% (5 lM Wortmannin + 10 lM U0126 + 1 lM a-tomatine)(Fig. 5B). The data finding revealed the inhibition of the expres-sions of MMP-2 and MMP-9 by a-tomatine on A549 cells couldpartly occur through ERK1/2 and Akt inactivation, while the inhi-bition of the expression of u-PA could partly occur throughERK1/2 inactivation.

3.5. a-Tomatine inhibits the DNA binding activities of NF-jB, c-Fos,and c-Jun

NF-jB and AP-1 family of transcriptional factors have beenknown to translocate to the nucleus and regulate the expressionof multiple genes involved in MMPs or u-PA secretion. To clarify

, p38, Akt, and the protein expression level of PI3K. In dose-dependent assay (A, C, E,atine for 3 h. In time-dependent assay (B, D, F, and H), A549 cells were treated withhosphorylation, JNK, p38 phosphorylation, p38, Akt phosphorylation, Akt, and the

trol. Values are expressed as mean ± SD of three independent experiments. *p < 0.05,

Fig. 4. (continued)

Y.-W. Shih et al. / Food and Chemical Toxicology 47 (2009) 1985–1995 1991

the involvement of NF-jB and AP-1 proteins in the mechanism ofa-tomatine’s action, the effect of a-tomatine on the DNA bindingactivities of NF-jB and AP-1 in A549 cells was explored by EMSA.As shown in Fig. 6A, A549 cells were treated with 0–2 lM of a-tomatine for 24 h, and a-tomatine inhibited NF-jB and AP-1 tran-scriptional activities in a dose-dependent manner. Especially, thebinding activities of NF-jB and AP-1 were strongly inhibited bytreating with 2 lM a-tomatine. Further, the expressions of NF-jB, c-Fos, and c-Jun in nuclear extracts were analyzed by Westernblotting to assess the possible inhibitory effect of a-tomatine onNF-jB, c-Fos, and c-Jun. As explained in Fig. 6B, the nuclear levelsof NF-jB, c-Fos, and c-Jun were gradually diminished by a-toma-tine in a dose-dependent manner. Especially, data was shown tobe strongly inhibited by treating with 2 lM a-tomatine.

4. Discussion

Lung cancer is the most common neoplasm in humans in bothdeveloped and developing countries (Erridge et al., 2007; Gajraet al., 2003; McCracken et al., 2007). This research has confirmeda-tomatine can inhibit the invasion and migration of A549 humanadenocarcinoma cells in vitro model. We found a-tomatine cansuppress cancer cell invasion and migration possibly occursthrough inactivation of PI3K/Akt or ERK signaling pathways, exert-ing inhibitory effects on NF-jB, c-Fos, and c-Jun transcriptional fac-tors, inhibiting NF-kB and AP-1 DNA binding activities, decreasingMMP-2, MMP-9, and u-PA activities, and then having an anti-met-astatic effect. Our results strengthen the potential of a-tomatine asa new strategy for anti-cancer therapy.

Fig. 5. Effect of PI3K inhibitor (Wortmannin), ERK inhibitor (U0126), and a-tomatine on the activities of MMP-2, MMP-9, and u-PA. Cells were plated in six-welland pre-treated with Wortmannin (5 or 10 lM) or U0126 (10 or 20 lM) for 1 h andthen incubated in the presence or absence of a-tomatine (1 lM) for 24 h.Afterwards, the culture medium was subjected to gelatin and casein zymographyto analyze the activities of (A) MMP-2/MMP-9 and (B) u-PA. Determined activitiesof these proteins were subsequently quantified by densitometric analysis with thatof control being 100% as shown just below the gel data. Data represented themean ± SD of three independent experiments (*p < 0.05, **p < 0.01, ***p < 0.001).

1992 Y.-W. Shih et al. / Food and Chemical Toxicology 47 (2009) 1985–1995

a-Tomatine, a tomato glycoalkaloid, may also have beneficial ef-fects. Glycoalkaloids are reported to inactivate the Herpes simplexand Herpes zoster viruses in humans (Chataing et al., 1997), to en-hance the duration of action of anesthetics, which act by inhibitingacetylcholinesterase (McGehee et al., 2000), and to potentiate theimmune response of vaccines in mice (Rajananthanan et al.,2000). a-Tomatine may benefit cancer chemotherapy by inhibitingmultidrug resistance in human cancer cells (Lavie et al., 2001). Aspart of an effort designed to improve food safety through identifica-tion and reduction of the content of the most toxic alkaloids in plantfoods using safety evaluation. In previous study has demonstrateda-tomatine dose not appear to be toxic when consumed orally inmoderate amount, and observation that the absence of a 5, 6-dou-ble bond in the B-ring of tomatidine results in a much less toxic

molecule in mice (Friedman et al., 2000). Wilson et al. studied thepharmacology and toxicology of a-tomatine. In mice, a-tomatineappears to be non-toxic following oral consumption, presumablybecause of poor absorption from the digestive tract into the blood-stream due to formation of an insoluble complex with dietary andendogenous cholesterol which is then eliminated in the faeces(Cayen, 1971; Roddick, 1979). Nevertheless, further studies needto be done in order to investigate the anti-metastatic effect of a-tomatine in humans.

Malignant tumors invade the tissue, involving three indepen-dent processes: the degradation of the extracellular matrix(ECM), cell metastasis and proliferation. Metastasis has been foundto be accompanied by various physiological alterations involved indegrading ECM, such as the overexpression of proteolytic enzymeactivity as in MMPs or u-PA, as well as the migration and invasionof tumor cells into the bloodstream or lymphatic system to spreadto other tissues or organs (Kleiner and Stetler-Stevenson, 1999).More specifically, the ability to penetrate the basement membrane(BM) is related to an increased potential for metastasis. Basementmembranes are thin extracellular matrices underlying cells in vivo.Matrigel Basement Membrane Matrix is a solubilized basementmembrane preparation extracted from the Englebreth-HolmSwarm (EHS) mouse tumor. It major component is lamin, collagenI, entactin, heparin sulfate proteoglycan (perlecan), growth factors,and so on. A number of methods have been developed using Matri-gel Matrix to investigate the invasion of the basement membranematrix by tumor in vitro. The invasive ability of A549 cells to passthrough Matrigel-coated filters was measured by the Boydenchamber invasion assay. So, our study demonstrated treatmentwith a-tomatine at a non-cytotoxic concentration below 2 lMfor 24 h exerted an inhibitory effect in a dose- and time-dependentmanner on the migration and invasion of the highly metastaticA549 cells. In recent years, attention has been drawn to the phys-iological relevance of MMPs and u-PA markers related to the met-astatic ability and malignancy of tumor cells (Chen et al., 2005;Shih et al., 2007). Thus, many studies have shown proteinases re-lated to degradation of matrix are required for tumor cell metasta-sis and heightened production of MMPs and u-PA correlates withthe invasion, migration and angiogenesis of the tumors (Mackayet al., 1990). To further explore the exact expression of a-toma-tine-induced inhibition on migration and invasion, we performedgelatin or casein-plasminogen zymographic assays, to detect activ-ities of MMP-2, MMP-9, and u-PA. The result showed a-tomatinenoticeably downregulated the activities of MMP-2, MMP-9, andu-PA. These results suggested the anti-metastastic effect ofa-tomatine was associated with the inhibition of the enzymaticallydegradative processes of tumor metastasis. One important charac-teristic of metastasis is the migratory and invasive ability of tumorcells. Further, we used wound-healing assay and Boyden chamberassay to quantify the migratory potential of A549 cells. The resultsdemonstated a-tomatine significantly inhibited the migration andinvasion of A549 cells.

A major mechanism through which signals from extracellularstimuli are transmitted to the nucleus involves the activation of ki-nases. These kinases, serine/threonine kinases related to the mito-gen-activated proteins kinase (MAPK) superfamily, mediate signalsfrom cell membrane receptors triggered by growth factors, cyto-kines, and cell-matrix interactions. MAPKs are intricately involvedin the expression of the components involved in MMPs or u-PApromoters induction by NF-kB, AP-1, and its association with c-Fos and c-Jun. At least three subgroups of MAPK family membershave been implicated: extracellular signal-regulated kinases(ERKs), c-Jun N-terminal kinase/stress-activated protein kinase(JNK/SAPK), and p38MAPK (Robinson and Cobb, 1997). Also, thePI3K and Akt signal pathways also play a critical role in MMP-9gene regulation (Yao et al., 2001).

Fig. 6. Effects of a-tomatine on the DNA binding activities of NF-jB and AP-1 in A549 cells. Cells were treated with various concentrations (0, 1, 1.5, and 2 lM) of a-tomatinefor 24 h. Cell nuclear extracts were prepared and analyzed for (A) NF-jB and AP-1 DNA binding activities using biotin-labeled consensus NF-jB and AP-1 specificoligonucleotide, then EMSA assay were performed as described in materials and methods. Lane 1: nuclear extracts incubated with 100-fold excess unlabeled consensusoligonucleotide (comp.) to confirm the specificity of binding. Excess free probe is indicated at the bottom. (B) Nuclear extracts were also analyzed by Western blotting withanti-NF-jB (p65), c-Fos, and c-Jun antibodies. C23 was a nucleus protein loading control. Determined the protein expressions of NF-jB, c-Fos, and c-Jun were subsequentlyquantified by densitometric analysis with that of control being onefold. The densitometric results are expressed as mean ± SD of three independent experiments. *p < 0.05,**p < 0.01, ***p < 0.001 compared with the untreated control.

Y.-W. Shih et al. / Food and Chemical Toxicology 47 (2009) 1985–1995 1993

Indeed, we have demonstrated treatment of a-tomatine inhib-ited phosphorylation of ERK1/2 and Akt. In contrast, a-tomatinedid not significantly affect phospho-ERK1/2 and phospho-p38activities. The involvement of the ERK and PI3K/Akt pathwayswas further supported by using the ERK and PI3K inhibitors in

our experimental model. Treatment with an inhibitors specificfor ERK could inhibit the MMP-2, MMP-9, and u-PA secretion. Also,the PI3K/Akt signaling pathway also played a crucial role in MMP-2, MMP-9, and u-PA gene regulation, cell survival and tumor cellmetastasis (Kim et al., 2001; Rao, 2003; Westermarck and Kahari,

1994 Y.-W. Shih et al. / Food and Chemical Toxicology 47 (2009) 1985–1995

1999). Our findings suggested a-tomatine-caused decreased MMP-2 and MMP-9 activities in the culture media could possibly occurthrough suppression of phospho-ERK1/2 or phospho-Akt, whilethe reduction in u-PA activity may only be associated with a sup-pression of phospho-ERK1/2. Here, it was demonstrated a-toma-tine clearly decreased MMP-2, MMP-9, and u-PA activitiesthrough inactivating the PI3K/Akt or ERK signaling pathways, andsuch inhibitory effect on proteinases expression may contributeto the capability of a-tomatine for the inhibition of cell metastasis.

In this study, the activity of u-PA, an upstream activating en-zyme of MMP-2 and MMP-9 involved in invasion and migration,was also shown to be inhibited in a dose-dependent manner bya-tomatine treatment. The transcription of MMPs and u-PA geneis regulated by upstream regulatory sequences, including NF-jB,AP-1, and Ets-1 binding sites (Nagase and Woessner, 1999; Roth-hammer et al., 2004; Sliva, 2004; Westermarck and Kahari,1999). Therefore, this study provides insight into how a-tomatinesuppresses the ERK1/2 and Akt signaling pathways and reducesNF-kB and AP-1 transcriptional activities in A549 lung adenocarci-noma cells. Indeed, one or more of these binding sites have beenimplicated in mediating the effects of a diverse set of agents. Here,we have also found the treatment of a-tomatine to A549 cells re-sults in an inhibition of NF-kB and AP-1 DNA binding activities,which was accompanied by the inhibition of nuclear translocationof these factors. Thus, the inhibitory effect of a-tomatine on themigration and invasion of non-small lung cancer cells probably oc-curs by abating the expressions of NF-kB, c-Fos, and c-Jun, thenreducing the activities of MMP-2, MMP-9, and u-PA.

In addition, Chishma et al. (1997) indicated the tumor host-or-gan chimeric histoculture system developed in the present studywith GFP fluorescing tumor cells has significantly advanced theability to understand and treat human metastatic cancer. The re-sults provide an invaluable new tool for understanding the mostimportant steps in tumor host-organ interaction, tumor progres-sion, and metastasis. Because the metastatic colony expansiontakes place in vitro, it offers a unique opportunity for developingagents for intervention. We will carry out this histoculture exper-iment in the further.

Finally, the involvement of ERK and PI3K/Akt signaling path-ways in cell metastasis were further supported by experimentswith ERK and PI3K/Akt inhibitors, showing treatment with inhibi-tors of ERK and PI3K/Akt to A549 cells inhibited the cell metastasis.In conclusion, these results imply the therapeutic potential of a-tomatine for controlling tumor metastasis based on the observa-tion of its inhibitory effect on migration and invasion of adenocar-cinoma cancer cell line A549 cells. This study suggested a-tomatine may serve as an efficient anti-metastastic drug in cancertreatment.

Conflict of interest statement

The authors declare that there are no conflicts of interest.

Acknowledgment

This work was supported by the grant from the Center for Re-gional Industry Academia Collaboration M.O.E, Ministry of Educa-tion (96-B-17-042, 97-B-17-018).

References

Aguirre Ghiso, J.A., Alonso, D.F., Farias, E.F., Gomez, D.E., de Kier Joffe, E.B., 1999.Deregulation of the signaling pathways controlling urokinase production. Itsrelationship with the invasive phenotype. Eur. J. Biochem. 263, 295–304.

Bernhard, E.J., Gruber, S.B., Muschel, R.J., 1994. Direct evidence linking expression ofmatrix metalloproteinase 9 (92-kDa gelatinase/collagenase) to the metastatic

phenotype in transformed rat embryo cells. Proc. Natl. Acad. Sci. USA 91, 4293–4297.

Cayen, M.N., 1971. Effect of dietary tomatine on cholesterol metabolism in the rat. J.Lipid Res. 12, 482–490.

Chan-Hui, P., Weaver, R., 1998. Human mitogen-activated protein kinase kinasekinase mediates the stress-induced activation of mitogen-activated proteinkinase cascades. Biochem. J. 336, 599–609.

Chataing, B., Concepcion, J.L., de Cristancho, N.B., Usubillaga, A., 1997. Estudioclinico de la efectividad de extractos alcaloides obtenidos de los frutos delSolanum americanum Miller sobre el Herpes simplex, Herpes Zoster y Herpesgenitalis. Revista de la Facultad de Farmacia 32, 18–25.

Chen, P.N., Hsieh, Y.S., Chiou, H.L., Chu, S.C., 2005. Silibinin inhibits cell invasionthrough inactivation of both PI3K-Akt and MAPK signaling pathways. Chem.Biol. Interact. 156, 141–150.

Chishima, T., Yang, M., Miyagi, Y., Li, L., Tan, Y., Baranov, E., Shimada, H., Moossa,A.R., Penman, S., Hoffman, R.M., 1997. Governing step of metastasis visualizedin vitro. Proc. Natl. Acad. Sci. USA 94, 11573–11576.

Chu, S.C., Chiou, H.L., Chen, P.N., Yang, S.F., Hsieh, Y.S., 2004. Silibinin inhibits theinvasion of human lung cancer cells via decreased productions of urokinase-plasminogen activator and matrix metalloproteinase-2. Mol. Carcinog. 40, 143–149.

Duffy, M.J., Duggan, C., 2004. The urokinase plasminogen activator system: a richsource of tumour markers for the individualized management of patients withcancer. Clin. Biochem. 37, 541–548.

Erridge, S.C., Moller, H., Price, A., Brewster, D., 2007. International comparisons ofsurvival from lung cancer: pitfalls and warnings. Nat. Clin. Pract. Oncol. 4, 570–577.

Friedman, M., Fitch, T.E., Levin, C.E., Yokoyama, W.H., 2000. Feeding of tomatoes tohamsters reduces their plasma LDL cholesterol and triglycerides. J. Food Sci. 65,897–900.

Friedman, M., Levin, C.E., 1995. a-Tomatine content in tomato and tomato productsdetermined by HPLC with pulsed amperometric detection. J. Agric. Food Chem.43, 1507–1511.

Gajra, A., Newman, N., Gamble, G.P., Abraham, N.Z., Kohman, L.J., Graziano, S.L.,2003. Impact of tumor size on survival in stage IA non-small cell lung cancer: acase for subdividing stage IA disease. Lung Cancer 42, 51–57.

Greenlee, R.T., Hill-Harmon, M.B., Murray, T., Thun, M., 2001. Cancer statistics. CACancer J. Clin. 51, 15–36.

Hoppe-Seyler, F., Butz, K., Rittmuller, C., von Knebel Doeberitz, M., 1991. A rapidmicroscale procedure for the simultaneous preparation of cytoplasmic RNA,nuclear DNA binding proteins and enzymatically active luciferase extracts.Nucleic Acids Res. 19, 5080.

Itoh, Y., Nagase, H., 2002. Matrix metalloproteinases in cancer. Essays Biochem. 38,21–36.

Karin, M., Ben-Neriah, Y., 2000. Phosphorylation meets ubiquitination: the controlof NF-[kappa]B activity. Annu. Rev. Immunol. 18, 621–663.

Kim, D., Kim, S., Koh, H., Yoon, S.O., Chung, A.S., Cho, K.S., Chung, J., 2001. Akt/PKBpromotes cancer cell invasion via increased motility and metalloproteinaseproduction. FASEB J. 15, 1953–1962.

Kleiner, D.E., Stetler-Stevenson, W.G., 1999. Matrix metalloproteinases andmetastasis. Cancer Chemother Pharmacol. 43 (Suppl.), S42–S51.

Kwon, G.T., Cho, H.J., Chung, W.Y., Park, K.K., Moon, A., Park, J.H., 2008.Isoliquiritigenin inhibits migration and invasion of prostate cancer cells:possible mediation by decreased JNK/AP-1 signaling. J. Nutr. Biochem. [Epubahead of print].

Lavie, Y., Harel-Orbital, T., Gaeld, W., Liscovitch, M., 2001. Inhibitory effect ofsteroidal alkaloids on drug transport and multidrug resistance in human cancercells. Anticancer Res. 21, 1189–1194.

Lee, G.Y., Jang, J.S., Lee, S.Y., Jeon, H.S., Kim, K.M., Choi, J.E., Park, J.M., Chae, M.H., Lee,W.K., Kam, S., Kim, I.S., Lee, J.T., Jung, T.H., Park, J.Y., 2005. XPC polymorphismsand lung cancer risk. Int. J. Cancer 115, 807–813.

Lee, K.R., Kozukue, N., Han, J.S., Park, J.H., Chang, E.Y., Baek, E.J., Chang, J.S., Friedman,M., 2004. Glycoalkaloids and metabolites inhibit the growth of human colon(HT29) and liver (HepG2) cancer cells. J. Agric. Food Chem. 52, 2832–2839.

Lee, S.J., Park, S.S., Lee, U.S., Kim, W.J., Moon, S.K., 2008. Signaling pathway for TNF-alpha-induced MMP-9 expression: mediation through p38 MAP kinase, andinhibition by anti-cancer molecule magnolol in human urinary bladder cancer5637 cells. Int. Immunopharmacol. 8, 1821–1826.

Lee, S.O., Jeong, Y.J., Im, H.G., Kim, C.H., Chang, Y.C., Lee, I.S., 2007. Silibininsuppresses PMA-induced MMP-9 expression by blocking the AP-1 activation viaMAPK signaling pathways in MCF-7 human breast carcinoma cells. Biochem.Biophys. Res. Commun. 354, 165–171.

Liao, D.W., Wang, L., Zhang, X.G., Liu, M.Q., 2006. Expression and significance ofPTEN/PI3K signal transduction-related proteins in non-small cell lung cancer.Ai. Zheng. 25, 1238–1242.

Ma, W., Lim, W., Gee, K., Aucoin, S., Nandan, D., Kozlowski, M., Diaz-Mitoma, F.,Kumar, A., 2001. The p38 mitogen-activated kinase pathway regulates thehuman interleukin-10 promoter via the activation of Sp1 transcription factor inlipopolysaccharide-stimulated human macrophages. J. Biol. Chem. 276, 13664–13674.

Ma, Z., Shah, R.C., Chang, M.J., Benveniste, E.N., 2004. Coordination of cell signaling,chromatin remodeling, histone modifications, and regulator recruitment inhuman matrix metalloproteinase 9 gene transcription. Mol. Cell Biol. 24, 5496–5509.

Mackay, A.R., Corbitt, R.H., Hartzler, J.L., Thorgeirsson, U.P., 1990. Basementmembrane type IV collagen degradation: evidence for the involvement of a

Y.-W. Shih et al. / Food and Chemical Toxicology 47 (2009) 1985–1995 1995

proteolytic cascade independent of metalloproteinases. Cancer Res. 50, 5997–6001.

McCracken, M., Olsen, M., Chen Jr., M.S., Jemal, A., Thun, M., Cokkinides, V., Deapen,D., Ward, E., 2007. Cancer incidence, mortality, and associated risk factorsamong Asian Americans of Chinese, Filipino, Vietnamese, Korean, and Japaneseethnicities. CA Cancer J. Clin. 57, 190–205.

McGehee, D.S., Krasowski, M.D., Fung, D.L., Wilson, B., Gronert, G.A., Moss, J., 2000.Cholinesterase inhibition by potato glycoalkaloids slows mivacuriummetabolism. Anesthesiology 93, 510–519.

Mosmann, T., 1983. Rapid colorimetric assay for cellular growth and survival:application to proliferation and cytotoxicity assays. J. Immunol. Methods 65,55–63.

Nagase, H., Woessner Jr., J.F., 1999. Matrix metalloproteinases. J. Biol. Chem. 274,21491–21494.

Ochi, Y., Atsumi, S., Aoyagi, T., Umezawa, K., 1993. Inhibition of tumor cell invasionin the Boyden chamber assay by a mannosidase inhibitor, mannostatin A.Anticancer Res. 13, 1421–1424.

Rajananthanan, P., Attard, G.S., Sheikhh, N., Morrow, W.J., 2000. Novel aggregatestructure modulate lymphocyte proliferation and Th1 and Th2 cytokine profilesin ovalbumin immunized mice. Vaccine 18, 140–152.

Rao, J.S., 2003. Molecular mechanisms of glioma invasiveness: the role of proteases.Nat. Rev. Cancer 3, 489–501.

Robinson, M.J., Cobb, M.H., 1997. Mitogen-activated protein kinase pathways. Curr.Opin. Cell Biol. 9, 180–186.

Roddick, J.G., 1979. Complex formation between solanaceous steroidalglycoalkaloids and free sterols in vitro. Phytochemistry 18, 1467–1470.

Rothhammer, T., Hahne, J.C., Florin, A., Poser, I., Soncin, F., Wernert, N., Bosserhoff,A.K., 2004. The Ets-1 transcription factor is involved in the development andinvasion of malignant melanoma. Cell Mol. Life Sci. 61, 118–128.

Shih, J.Y., Tsai, M.F., Chang, T.H., Chang, Y.L., Yuan, A., Yu, C.J., Lin, S.B., Liou, G.Y., Lee,M.L., Chen, J.J., Hong, T.M., Yang, S.C., Su, J.L., Lee, Y.C., Yang, P.C., 2005.Transcription repressor slug promotes carcinoma invasion and predictsoutcome of patients with lung adenocarcinoma. Clin. Cancer Res. 11, 8070–8078.

Shih, Y.W., Chen, P.S., Wu, C.H., Jeng, Y.F., Wang, C.J., 2007. a-Chaconine-reducedmetastasis involves a PI3K/Akt signaling pathway with downregulation of NF-jB in human lung adenocarcinoma A549 cells. J. Agric. Food Chem. 55, 11035–11043.

Shivapurkar, N., Reddy, J., Chaudhary, P.M., Gazdar, A.F., 2003. Apoptosis and lungcancer: a review. J. Cell. Biochem. 88, 885–898.

Sliva, D., 2004. Signaling pathways responsible for cancer cell invasion as targets forcancer therapy. Curr. Cancer Drug Targets 4, 327–336.

Turner, N.A., Aley, P.K., Hall, K.T., Warburton, P., Galloway, S., Midgley, L., O’regan,D.J., Wood, I.C., Ball, S.G., Porter, K.E., 2007. Simvastatin inhibits TNFalpha-induced invasion of human cardiac myofibroblasts via both MMP-9-dependentand -independent mechanisms. J. Mol. Cell Cardio. 43, 168–176.

Westermarck, J., Kahari, V.M., 1999. Regulation of matrix metalloproteinaseexpression in tumor invasion. FASEB J. 13, 781–792.

Yao, J., Xiong, S., Klos, K., Nguyen, N., Grijalva, R., Li, P., Yu, D., 2001. Multiplesignaling pathways involved in activation of matrix metalloproteinase-9 (MMP-9) by heregulin-beta1 in human breast cancer cells. Oncogene 20, 8066–8074.