CHM-1 inhibits hepatocyte growth factor-induced invasion of SK-Hep-1 human hepatocellular carcinoma cells by suppressing matrix metalloproteinase-9 expression.

(1)

CHM-1 inhibits hepatocyte growth factor-induced invasion

of SK-Hep-1 human hepatocellular carcinoma cells

by suppressing matrix metalloproteinase-9 expression

Shih-Wei Wang

a

, Shiow-Lin Pan

a

, Chieh-Yu Peng

a

, Der-Yi Huang

a

,

An-Chi Tsai

a

, Ya-Ling Chang

a

, Jih-Hwa Guh

b

, Sheng-Chu Kuo

c

,

Kuo-Hsiung Lee

d

, Che-Ming Teng

a,*

a

Pharmacological Institute, College of Medicine, National Taiwan University, No. 1, Jen-Ai Road, Sect. 1, Taipei, Taiwan

b

School of Pharmacy, College of Medicine, National Taiwan University, Taipei, Taiwan

c_{Graduate Institute of Pharmaceutical Chemistry, China Medical University, Taichung, Taiwan} d_{Natural Products Laboratory, School of Pharmacy, University of North Carolina, Chapel Hill, NC 27599, USA}

Received 5 April 2007; received in revised form 24 June 2007; accepted 2 July 2007

Abstract

Clinical observations suggest that hepatocyte growth factor (HGF) can promote invasion and metastasis in

hepatocel-lular carcinoma. In this study, we found that HGF-stimulated invasion of SK-Hep-1 cells, together with increased

expres-sion of matrix metalloproteinase (MMP)-9. CHM-1 was identiﬁed from 2-phenyl-4-quinolone derivatives to potently

inhibit HGF-induced cell invasion, proteolytic activity, and expression of MMP-9. CHM-1 signiﬁcantly inhibited tyrosine

autophosphorylation of c-Met induced by HGF. CHM-1 also suppressed HGF-induced Akt phosphorylation, and NF-jB

activation, the downstream regulators of HGF/c-Met signaling, resulting in the inhibition of MMP-9. Thus, we suggest

that CHM-1 is a potential therapeutic agent against tumor invasion.

2007 Elsevier Ireland Ltd. All rights reserved.

Keywords: HGF; MMP-9; Invasion; Hepatocellular carcinoma

1. Introduction

The lethality of most malignant tumors is the

result of local invasion and metastasis from the

pri-mary tumors to other tissues. Invasion is a

charac-teristic feature of hepatocellular carcinoma (HCC),

which frequently shows early invasion into blood

vessels as well as intrahepatic metastasis and

extra-hepatic metastasis occurs subsequently

[1]

. Thus,

the discovery and subsequent development of novel

small-molecule agents to block HCC invasion are

the goals of cancer researchers.

Hepatocyte growth factor (HGF), a pleiotropic

modulator, is produced by nonparenchymal liver

cells, and its serum levels are elevated in a variety

of liver diseases, including HCC

[2]

. The receptor

for HGF is a receptor type tyrosine kinase encoded

* _{Corresponding author. Tel./fax: +886 2 2322 1742.}

E-mail address:[email protected](C.-M. Teng).

(2)

by the c-Met proto-oncogene. c-Met is normally

expressed by epithelial cells and has been found to

be overexpressed and ampliﬁed in various human

tumor tissues

[3]

. It has been reported that

overex-pression of c-Met is detected in some cases of

HCC and that elevated levels of c-Met expression

in HCC correlate with increased incidence of liver

metastasis

[4]

. These ﬁndings suggest that the

HGF/c-Met signaling plays a pivotal role in the

invasion and metastasis of HCC cells. Activation

of c-Met by HGF can induce cell proliferation,

sur-vival, motility, invasion, and changes in

morphol-ogy.

Kinase

activation

is

achieved

through

autophosphorylation of tyrosines 1234 and 1235 in

the catalytic domain, subsequently results in

bind-ing and/or phosphorylation of adaptor proteins

including Grb2, Src, and Gab-1, which in turn, are

capable of activating downstream pathways

includ-ing PI3K/Akt, Ras/MAPK, FAK, and STAT

sig-naling

[5]

.

Cell invasion is a major component of the

com-plex multistep process of tumor metastasis. Invasion

of malignant tumor cells requires destruction of

basement membranes and proteolysis of

extracellu-lar matrix (ECM)

[6]

. Of the several families of

ECM-degrading enzymes, the most extensive are

the matrix metalloproteinases (MMPs), which are

a large family of structurally related

zinc-endopep-tidases that collectively degrade most of the

compo-nents of ECM

[7]

. Among previously reported

human MMPs, MMP-9 (gelatinase B) is thought

to be a key enzyme for degrading type IV collagen.

MMP-9 is abundantly expressed in diverse

malig-nant tumors and is postulated to play an important

role in HCC invasion and metastasis

[8]

. Therefore,

the inhibition of invasion mediated by MMP-9 may

be critical for the prevention of cancer metastasis.

The 2-phenyl-4-quinolones and related

com-pounds, a series of synthetic quinolone derivatives,

have been reported to against a broad spectrum of

human cancer cell lines

[9–11]

. However, the

anti-invasion property of the 2-phenyl-4-quinolone series

has not been demonstrated. In this study, we used

HGF to induce invasion in SK-Hep-1 cells, a highly

invasive human HCC cell line, as the screen system

in studying anti-invasive eﬀects of

2-phenyl-4-quinolone derivatives. We identiﬁed that 2

0

-ﬂuoro-6,7-methylenedioxy-2-phenyl-4-quinolone (CHM-1,

Fig. 1

) potently inhibited HGF/c-Met-mediated cell

invasion in SK-Hep-1 cells. The mechanism of

CHM-1 to inhibit HGF-induced invasion in HCC

cells was also investigated in this study.

2. Materials and methods

2.1. Materials

CHM-1 was synthesized at the Graduate Institute of

Pharmaceutical Chemistry, School of Medicine, China

Medical University (Taichung, Taiwan). Recombinant

human HGF was purchased from R&D Systems, Inc.

(Minneapolis, MN). DMEM, fetal bovine serum (FBS),

and all the other cell culture reagents were obtained from

Gibco-BRL life technologies (Grand Island, NY).

Anti-bodies to phospho-c-Met (Tyr1234/1235), phospho-Akt

(Ser473),

Akt,

phospho-ERK1/2

(Thr202/Tyr204),

ERK1/2, phospho-JNK (Thr183/Tyr185), JNK,

phos-pho-p38

(Thr180/Tyr182),

p38,

and

phospho-IjBa

(Ser32) were purchased from Cell Signaling Technologies

(Boston, MA). Antibodies to c-Met, IjBa, NF-jB/p65,

nucleolin, mouse immunoglobulin (Ig) G, and

anti-rabbit IgG were purchased from Santa Cruz Biotechnology

(Santa Cruz, CA). Antibody to GAPDH was purchased

from ABcam (Cambridge, UK). Anti-MMP-2 polyclonal

antibody,

anti-MMP-9

polyclonal

antibody,

and

GM6001 were purchased from Chemicon International

(Temecula, CA). SU11274 and other chemical agents were

obtained from Sigma Chemical Co. (St Louis, MO).

2.2. Cell culture

The human HCC cell line SK-Hep-1 was obtained

from American Type Culture Collection (ATCC,

Manas-sas, VA) and cultured in DMEM containing supplements

(10% FBS, penicillin/streptomycin, and

L

-glutamine).

Cells were maintained in humidiﬁed air containing 5%

CO

2

at 37

C. 2.3. Invasion assay

Invasion assays were performed in Transwell chambers

(Coring, Coring, NY). The upper side of the ﬁlters was

coated with Matrigel (BD Biosciences, Bedford, MA) at

a concentration of 125 lg/cm

2

. Cells were seeded

(5

· 10

4

cells/well) onto the upper chamber with

serum-Fig. 1. Structure of CHM-1 (20

(3)

free medium, then incubated in the bottom chamber with

serum-free medium containing recombinant HGF as a

source of chemoattractants. After 6 h of treatment, cells

on the upper side of the ﬁlters were mechanically

removed, and those migrated on the lower side were ﬁxed

with 4% formaldehyde, then stained with 0.5% crystal

vio-let for 10 min. Finally, invaded cells were counted at 200·

magnification in 10 different fields of each filter.

2.4. Cell viability assay

Cells were incubated in 96-well plates at a density of

10

4

cells per well, and the percentage of cell survival was

assessed using MTT colorimetric assay after drug

treatment.

2.5. Western blot analysis

Cells were lysed with lysis buﬀer and nuclear

fraction-ation was performed as described previously

[12]

. Cell

homogenates were diluted with loading buﬀer and boiled

for 5 min for detecting phosphorylation, and protein

expression. Total protein was determined and equal

amounts of protein were separated by 8–12% SDS–PAGE

and immunoblotted with speciﬁc primary antibodies.

Horseradish peroxidase-conjugated secondary antibodies

(Santa Cruz Biotechnology, Santa Cruz, CA) were used,

and the signal detected using an enhanced

chemilumines-cence detection kit (Amersham, Buckinghamshire, UK).

2.6. Gelatin zymography

The supernatant of SK-Hep-1 cells was

electrophore-sed for the analysis in 10% SDS–PAGE gels containing

gelatin (1 mg/ml). The gels were washed twice with 2.5%

Triton X-100 for 30 min to remove SDS. The gels were

incubated at 37

C in 50 mM Tris–HCl (pH 7.4),

contain-ing 10 mM CaCl

2

and 150 mM NaCl for 24 h. Following

incubation, the gels were stained with 0.25% Coomassie

Blue for 1 h, and then destained with de-staining buﬀer

until bands became clear.

2.7. MMP-2 and MMP-9 activity assay

Substrate-linked enzyme-linked immunosorbent assay

(ELISA) techniques (Amersham, Buckinghamshire, UK)

were used to quantify enzymatic activity of individual

MMPs. The samples were thawed on ice, and all reagents

needed for the assay were brought to room temperature.

The MMP-2 and MMP-9 activities were performed

according to the manufacturer’s instructions.

2.8. RT-PCR analysis

RNA was extracted from homogenized tissue with

TRIzol reagent by a standard protocol (Invitrogen,

Carls-bad, CA). Reverse transcription was performed with 5 lg

of mRNA and random primer at 65

C for 5 min and then

mixed with Moloney murine leukemia virus (M-MLV)

reverse transcriptase to react at 37

C for 1 h to obtain

cDNA. Gene ampliﬁcation was followed with reverse

transcriptase-polymerase chain reaction. Primers used in

this study were synthesized as follows, MMP-2 primers:

sense

primer,

5

0

_{-GGCCCTGTCACTCCTGAGAT-3}

0

_;

anti-sense primer, 5

0

-GGCATCCAGGTTATCGGGGA-3

0

_{, MMP-9 primers: sense primer, 5}

0

_-TGGGCTACGT

GACCTATGAC-3

0

_{; anti-sense primer, 5}

0

_-CAAAGGT

GAGAAGAGAGGGC-3

0

_{; GAPDH primers: sense}

pri-mer,

5

0

_{-TGATGACATCAAGAAGGTGGTGAAG-3}

0

_;

anti-sense primer, 5

0

_{-TCCTTGGAGGCCATGTGGGC}

CAT-3

0

_{. The PCR consisted of an initial denaturation at}

94 C for 5 min; 30 three-step cycles at 94 C for 1 min,

60 C for 1 min, and 72 C for 1 min; and a ﬁnal extension

at 72

C for 10 min. PCR products were analyzed on 1.5%

agarose gel in the presence of 1 lg/ml ethidium bromide.

The gels were photographed using a digital imaging

sys-tem (Gel DOC 2000, Bio-Rad, Hercules, CA).

2.9. Analysis of NF-jB/p65 activity

The NF-jB/p65 transcription factor ELISA kits

pur-chased from Active Motif Inc. (Carlsbad, CA) were used

for the detection of DNA binding activity of NF-jB/p65

subunit in commercial protocol.

2.10. Transfection and reporter gene assay

Reporter plasmid pNF-jB-Luc and pMMP-9-Luc

were kindly provided by Dr. J.C.-H. Cheng (National

Tai-wan University Hospital and College of Medicine, Taipei,

Taiwan). A dominant-negative IjBa mutant (IjBaM) was

a kind gift from Prof. C.-H. Lin (Taipei Medical

Univer-sity, Taipei, Taiwan). Renilla luciferase reporter vector

(phRG-TK) was purchased from Promega (Madison,

WI). SK-Hep-1 cells (2

· 10

5

) were seeded into 12-well

plates and grown overnight. Cells were transiently

trans-fected with 0.4 lg of NF-jB promoter plasmid or 2.5 lg

MMP-9 promoter plasmid using Lipofectmine 2000

(Invitrogen) according to the manufacturer’s protocol.

The phRG-TK was cotransfected with the above plasmids

as an internal control. To assay the aﬀect of

dominant-negative mutant, IjBaM (1 lg) was cotransfected with

pNF-jB-Luc or pMMP-9-Luc in this study. The

lucifer-ase activity was measured in the cellular extracts using

dual-luciferase reporter assay system (Promega).

2.11. Statistical analyses

Data are presented as the mean ± SEM for the

indi-cated number of separate experiment. Statistical analyses

of data were performed with one-way ANOVA followed

by Student’s t-test, and p-values less than 0.05 were

con-sidered signiﬁcant.

(4)

3. Results

3.1. CHM-1 inhibits HGF-induced invasion of SK-Hep-1

cells

By Transwell chamber assay, 50 ng/ml of

HGF-induced in vitro invasion of SK-Hep-1 cells. CHM-1

(0.1–10 lM) signiﬁcantly inhibited HGF-induced cell

invasion in a concentration-dependent manner (

Fig. 2

a).

Next, we determined the cytotoxicity of CHM-1 using

MTT assay. CHM-1 did not aﬀect cell viability at the

indi-cated concentrations. These results indicate that

inhibi-tory eﬀect of CHM-1 on cell invasion is independent of

cellular cytotoxicity.

3.2. CHM-1 inhibits HGF/c-Met signaling of SK-Hep-1

cells

We next investigated whether the antagonistic eﬀect of

CHM-1 on HGF-induced invasive activity could be

attributed to the inhibition of tyrosine

autophosphoryla-tion of c-Met induced by HGF. The c-Met of SK-Hep-1

cells were strongly phosphorylated in response to

stimula-tion with HGF for 15 min. CHM-1 signiﬁcantly inhibited

HGF-stimulated

c-Met

phosphorylation

(

Fig.

3 a).

SU11274 was used as a positive control for the speciﬁc

inhibition on HGF/c-Met signaling. To examine

HGF-induced downstream signaling of c-Met activation, the

phosphorylation of Akt, ERK1/2, JNK, and p38 in the

presence of CHM-1 was evaluated. CHM-1 substantially

inhibited

HGF-induced

phosphorylation

of

Akt

(

Fig. 3

b). However, CHM-1 did not suppress the

phos-phorylation of ERK1/2, JNK, or p38 induced by HGF

(

Fig. 3

c).

3.3. CHM-1 inhibits HGF-induced NF-jB activation

through inhibition of IjBa phosphorylation

Nuclear factor-jB (NF-jB) is a transcription factor

that plays an important regulator in invasion and

metastasis

[13]

. The eﬀect of CHM-1 on activation of

NF-jB was examined in HGF-treated cells. As shown

in

Fig. 4

a, HGF-induced NF-jB/p65 translocation

was signiﬁcantly inhibited by CHM-1 in a 2 h

treat-ment. Furthermore, HGF potently increased the

phos-phorylation of IjBa at the indicated times, and

treatment with CHM-1 for 2 h substantially suppressed

HGF-induced IjBa phosphorylation in SK-Hep-1 cells

(

Fig. 4

b). The NF-jB activity of SK-Hep-1 cells

trea-ted with CHM-1 for 2 h was then measured by

ELISA-based Trans-AM

TM

_{NF-jB p65 kit. As}

illus-trated in

Fig. 5

a, CHM-1-treatment suppressed

HGF-induced NF-jB activation in a concentration-dependent

manner. To further conﬁrm this result, NF-jB-binding

site-driven luciferase activity assay was performed. The

pNF-jB-Luc was transfected to measure the binding of

transcription factors to the j enhancer, providing a

direct measurement of NF-jB activation. The results

showed that HGF-induced jB-luciferase activity was

Fig. 2. Effect of CHM-1 on HGF-induced cell invasion in SK-Hep-1 cells. (a) Cells were seeded onto the upper chamber consisting of 8 lm pore-size filters coated with Matrigel, then treated without or with CHM-1 (0.1, 1, 10 lM) for 6 h in the absence or presence of HGF (50 ng/ ml) as a chemoattractant in the lower chamber. Cells that invaded the filter were counted as means ± SEM of five independent experiments. *_{p < 0.001 compared with the basal}

group; #p < 0.01, ##p < 0.001 compared the control group. (b) Cells were treated with the indicated concentrations of CHM-1 (0.1–10 lM) for 6 h in the presence of HGF (50 ng/ ml), and the cell viability was determined using MTT assay. Data are expressed as means ± SEM of four independent experiments.

(5)

inhibited by CHM-1, and the IjBa phosphorylation

inhibitor BAY 11-7082 was used as a positive control

(

Fig. 5

b).

3.4. CHM-1 inhibits HGF-activated MMP-9 expression

and enzyme activity

We then investigated the mechanism of HGF-mediated

cell invasive phenotype by looking at the involvement of

MMP-2 and MMP-9. Gelatin zymography was ﬁrstly

used to analyze the eﬀects of CHM-1 on MMP-2 and

MMP-9 activities for 6 h of treatment. As shown in

Fig. 6

a, we found that SK-Hep-1 cells constitutively

secreted high levels of 9 and low levels of

MMP-2, and the proteolytic activity of MMP-9 was dramatically

\Fig. 3. Eﬀect of CHM-1 on c-Met phosphorylation and signal transduction in SK-Hep-1 cells. Serum-starved cells were pre-treated 3 h without or with CHM-1 (1, 10 lM) or SU11274 (5 lM), then cells were stimulated with 50 ng/ml HGF for 10 min. Cells were harvested and lysed for the detection of phospho-Tyr1234/1235-c-Met and c-Met (a), phospho-Ser473-Akt and Akt (b), and phospho-Thr202/Tyr204-ERK1/2, ERK1/2, phospho-Thr183/ Tyr185-JNK, JNK, phospho-Thr180/Tyr182, and p38 (c) protein expressions by Western blot analysis. The quantitative densitom-etry of the relative level of protein was performed with Image-Pro Plus. Data are expressed as means ± SEM of ﬁve independent experiments.*_{p < 0.001 compared with the basal group;}#

p < 0.05,

##_{p < 0.01,}###_{p < 0.001 compared the control group.}

Fig. 4. Eﬀect of CHM-1 on HGF-induced NF-jB/p65 translo-cation and IjBa phosphorylation in SK-Hep-1 cells. Cells were treated without or with CHM-1 (1 lM) for the indicated times in the absence or presence of HGF (50 ng/ml). Cells were harvested and lysed for the detection of NF-jB/p65, and nucleolin (a), and phospho-Ser32-IjBa and GAPDH (b) protein expressions by Western blot analysis. The quantitative densitometry of the relative level of protein was performed with Image-Pro Plus. Data are expressed as means ± SEM of ﬁve independent experiments.

*_{p < 0.05,}**_{p < 0.01,}***_{p < 0.001 compared with the basal group;} #

(6)

activated by HGF. CHM-1 signiﬁcantly inhibited

HGF-activated MMP-9 proteolytic activity in a

concentration-dependent manner. We further conﬁrmed the inhibition

of CHM-1 on MMP-9 activity by ELISA assay, and

pan-MMP inhibitor GM6001 was used as a positive

control. We found that CHM-1 substantially inhibited

HGF-induced 9 activity, but did not aﬀect

MMP-2 activity in SK-Hep-1 cells (

Fig. 6

b). As the MMP-9

expression in CHM-1-treated cells, we demonstrated that

CHM-1 caused an inhibition of HGF-induced MMP-9

protein expression after 6 h of treatment (

Fig. 6

c), and

suppressed HGF-induced MMP-9 mRNA expression at

the earlier time point (4 h) (

Fig. 6

d). For further

conﬁrm-ing transcriptional inhibition of CHM-1 to MMP-9

expression, cells were transfected with MMP-9 promoter

containing reporter constructs and treated with CHM-1.

As shown in

Fig. 7

a, CHM-1 profoundly inhibited

HGF-induced MMP-9 promoter activity.

3.5. NF-jB activation is involved in HGF-induced MMP-9

expression

To

investigate

the

involvement

of

NF-jB

in

HGF-induced MMP-9 expression, cells were transiently

transfected with IjBaM. This IjBa mutated form contains

serine-to-alanine mutations at residues 32 and 36 and do

not undergo signal-induced phosphorylation; therefore,

cell expressing IjBaM block the NF-jB pathway

[14]

. As

shown in

Fig. 7

b, cells transfected with IjBaM almost

com-pletely abolished the HGF-induced increase in

jB-lucifer-ase activity. The HGF-induced MMP-9 promoter activity

was also signiﬁcantly attenuated by IjBaM (

Fig. 7

c). These

results suggest that NF-jB appears to serve as an upstream

signal for induction of MMP-9 expression by HGF.

4. Discussion

The high recurrence rate with intrahepatic

meta-static spread is major obstacle for improving

sur-vival of patients with HCC

[1]

. The development

of novel therapeutic agents targeting the malignant

behavior of HCC cells, especially their invasiveness,

is important to improve the prognosis of patients.

2-Phenyl-4-quinolone derivatives have been

demon-strated with potent anti-mitotic anti-tumor eﬀects

by inhibiting tubulin polymerization in a wide

variety of human cancer cells

[9–11]

. CHM-1 is a

small-molecule compound that was derived from

Fig. 5. Eﬀect of CHM-1 on HGF-induced NF-jB activation in SK-Hep-1 cells. (a) ELISA assay was performed after 2 h of incubation with CHM-1 (1, 10 lM) in HGF-treated cells. (b) Cells cotransfected with pNF-jB-Luc and phRG-TK vector were treated with CHM-1 (1, 10 lM) or BAY 11-7082 (10 lM) in the absence or presence of HGF (50 ng/ml). After 2 h of treatment, the promoter luciferase activities were detected using a lumino-meter. Data are expressed as means ± SEM of four independent experiments. *_{p < 0.001 compared with the basal group;} #_p

< 0.01,##_{p < 0.001 compared the control group.}

Fig. 6. Eﬀect of CHM-1 on HGF-increased MMP-9 expression and activity in SK-Hep-1 cells. (a) Cells were treated without or with CHM-1 (1, 10 lM) in the absence or presence of HGF (50 ng/ml). After 6 h of treatment, each conditioned medium was collected for analysis of proteolytic activities of MMP-2 and MMP-9 by gelatin zymography. (b) Cells were treated with CHM-1 (1, 10 lM) or GM6001 (10 lM) for 6 h in the absence or presence of HGF (50 ng/ml). Then, each conditioned medium was collected and subjected to ELISA assay for MMP-2 and MMP-9 activities. (c) Cells were treated with CHM-1 (1, 10 lM) or GM6001 (10 lM) for 6 h in the absence or presence of HGF (50 ng/ml). Cells were harvested and lysed for the detection of MMP-9 and GAPDH protein expressions by Western blot analysis. (d) Cells were treated without or with CHM-1 (1, 10 lM) in the absence or presence of HGF (50 ng/ml). After 4 h of treatment, the mRNA level of MMP-9 was detected by RT-PCR. Imaging was performed using quantitative densitometry with Image-Pro Plus. Data are expressed as means ± SEM of four independent experiments.*_{p < 0.05,}**_{p < 0.01 compared with the basal group;}#_{p < 0.05,}##_p

< 0.01 compared the control group.

(7)

2-phenyl-4-quinolones. In this study, we identiﬁed

CHM-1 as a potential lead base on anti-invasive

activity in HCC cells with good pharmacological

properties. CHM-1 induced a signiﬁcant

concentra-tion-dependent inhibition of HGF-activated cell

invasion, and dramatically inhibited HGF-induced

MMP-9 expression and enzyme activity in

SK-Hep-1 cells. Thus, CHM-1 is a promising

chemo-therapeutic agent worthy of further development

for treatment of human HCC.

(8)

HGF/c-Met signaling is implicated in numerous

human malignancies, including colon, gastric,

ovar-ian, lung, and liver cancer. This pathway activates a

program cell dissociation and motility coupled with

increased protease production that has been shown

to promote cellular invasion through ECM and that

closely resembles tumor metastasis in vivo

[15]

.

These suggest that strategies targeting c-Met

repre-sent an attractive novel therapeutic approach. In

the present study, we found that CHM-1 could

inhi-bit HGF-induced invasive activity through

sup-pressing

tyrosine

phosphorylation

of

c-Met.

evidence

reports

that

HGF-induced

response appears to work through both Ras/MAPK

and PI3K/Akt signaling pathways

[16,17]

. Our

results showed that CHM-1 profoundly inhibited

HGF-induced

Akt

phosphorylation,

but

not

MAPK phosphorylation. Thus, we suggest that

the PI3K/Akt pathway may play an important role

in the inhibition of CHM-1 on

HGF/c-Met-medi-ated invasion of SK-Hep-1 cells.

NF-jB regulates a variety of genes whose

prod-ucts are involved in many biological processes,

including inﬂammation, apoptosis, cell growth,

invasion, and metastasis

[18]

. In resting cells,

hete-rodimeric NF-jB complexes are located in the

cyto-plasm of most cell type by the inhibitory protein of

IjBa. NF-jB activation normally proceeds through

a pathway involving phosphorylation and

subse-quent degradation of IjBa, resulting in the

translo-cation of NF-jB from the cytoplasm to the nucleus.

In previous studies, CA and its derivative CAPE

reduced the liver metastasis through suppression

of MMP-9 gene expression by inhibiting NF-jB

activation

[19]

. Blocking NF-jB activity by

trans-fection with a mutated IjBa caused suppression of

angiogenesis, invasion, and metastasis in prostate

Fig. 7. Role of NF-jB in HGF-induced MMP-9 gene transcrip-tion in SK-Hep-1 cells. (a) Cells cotransfected with pMMP-9-Luc and phRG-TK vector were treated without or with CHM-1 (1, 10 lM) in the absence or presence of HGF (50 ng/ml). After 4 h of treatment, the promoter luciferase activities were detected using a luminometer. Data are expressed as means ± SEM of four independent experiments. *_{p < 0.001 compared with the}

basal group; #_{p < 0.01 compared the control group. (b) Cells}

cotransfected with pNF-jB-Luc and IjBaM were incubated in the absence or presence of HGF (50 ng/ml). After 2 h of incubation, the promoter luciferase activities were detected using a luminometer. The content of IjBa was determined in IjBaM-transfected cells by Western blot analysis. (c) Cells coIjBaM-transfected with pMMP-9-Luc and IjBaM were incubated in the absence or presence of HGF (50 ng/ml). After 4 h of incubation, the promoter luciferase activities were detected using a luminometer. Data represent the percentage in luciferase expression relative to that of the empty vector (EV) control in the presence of the vehicle.

(9)

cancer cells

[20]

. In this study, we demonstrated that

CHM-1 inhibited HGF-induced NF-jB activation

through inhibition of IjBa phosphorylation. These

results suggest that CHM-1 suppress the function

of NF-jB by blocking the nuclear translocation of

NF-jB. Accordingly, CHM-1 may be useful to

sup-press metastasis of liver cancer.

Many studies have revealed that growth factors

and cytokines secreted by tumor cells will induce

the production of MMPs. It has shown that elevated

serum levels of MMP-9 in HCC patients and

over-expression of MMP-9 in HCC tissues are related

to hematogenous invasion or capsular inﬁltration

of HCC cells

[8]

. On the other hand, growth factors

and cytokines can control the expression of MMP-9

by modulating the activation of transcription

fac-tors such as NF-jB and AP-1 through Ras/MAPK

and PI3K/Akt signal pathways. The NF-jB and

AP-1 elements of MMP-9 promoter are centrally

involved in the induction of MMP-9 gene associated

with the invasion of tumor cells

[21–23]

. In this

study, CHM-1 inhibited the enzymatic activity of

the MMP-9 protein secreted from SK-Hep-1 cells

via induction by HGF. We further showed that

CHM-1 inhibited HGF-induced MMP-9 protein

expression and gene transcription. Additionally,

we found that induction of MMP-9 expression by

HGF was a direct result of NF-jB activation, as

HGF did not induce MMP-9 gene expression in

IjBaM-transfected cells. We demonstrated that

suppression of NF-jB by CHM-1 down-regulated

the expression of MMP-9. Recently, Abiru et al.

showed that HGF-stimulated invasion in HCC cells

through induction of NF-jB target gene MMP-9,

and the inhibition of NF-jB activity using aspirin

and NS-398 led to suppression of HGF-induced

invasion through down-regulation of MMP-9 gene

expression

[24]

. Moreover, the PI3K/Akt pathway

plays an important role in the activation of

NF-jB, and Agarwal et al. found that PI3K/Akt/IjB

kinase pathway positively regulated NF-jB to

pro-mote metastatic gene expression in colorectal cancer

[25,26]

. Thus, we suggest that anti-invasive activity

of CHM-1 may be through the selective suppression

of MMP-9 regulated by PI3K/Akt/NF-jB signal

transduction.

In conclusion, we demonstrate that CHM-1

inhibits HGF/c-Met-mediated cell invasion via the

down-regulation of MMP-9 in SK-Hep-1 cells.

The Akt/NF-jB signaling pathway may be

coordi-nately involved in CHM-1’s anti-invasive eﬀect.

The further study of CHM-1 on the downstream

signal of HGF/c-Met activation in HCC cells is

needed to investigate in the future. Based on the

ﬁndings herein, we suggest that CHM-1 could be

eﬀective candidate for prevention of HCC cell

inva-sion associated with the HGF/c-Met system.

Acknowledgements

This work was supported by the National Science

Council of the Republic of China (NSC

94-2811-B-002-017) awarded to C.-M. Teng and in part by

grant from NIH CA17625 awarded to K.-H. Lee.

Appendix A. Supplementary data

Supplementary data associated with this article

can

be

found,

in

the

online

version,

at

doi:10.1016/j.canlet.2007.07.002

.

References

[1] M. Mitsunobu, A. Toyosaka, T. Oriyama, E. Okamoto, N. Nakao, Intrahepatic metastases in hepatocellular carci-noma: the role of the portal vein as an eﬀerent vessel, Clin. Exp. Metastasis 14 (1996) 520–529.

[2] G. Shiota, J. Okano, H. Kawasaki, T. Kawamoto, T. Nakamura, Serum hepatocyte growth factor levels in liver diseases: clinical implications, Hepatology 21 (1995) 106–112.

[3] C. Birchmeier, W. Birchmeier, E. Gherardi, G.F. Vande Woude, Met, metastasis, motility and more, Nat. Rev. Mol. Cell Biol. 4 (2003) 915–925.

[4] T. Ueki, J. Fujimoto, T. Suzuki, H. Yamamoto, E. Okam-oto, Expression of hepatocyte growth factor and its receptor c-met proto-oncogene in hepatocellular carcinoma, Hepa-tology 25 (1997) 862–866.

[5] J.G. Christensen, J. Burrows, R. Salgia, c-Met as a target for human cancer and characterization of inhibitors for thera-peutic intervention, Cancer Lett. 225 (2005) 1–26.

[6] L.A. Liotta, E.C. Kohn, The microenvironment of the tumour-host interface, Nature 411 (2001) 375–379. [7] M. Hidalgo, S.G. Eckhardt, Development of matrix

metallo-proteinase inhibitors in cancer therapy, J. Natl. Cancer Inst. 93 (2001) 178–193.

[8] A. Hayasaka, N. Suzuki, N. Fujimoto, S. Iwama, E. Fukuyama, Y. Kanda, H. Saisho, Elevated plasma levels of matrix metalloproteinase-9 (92-kd type IV collagenase/ gelatinase B) in hepatocellular carcinoma, Hepatology 24 (1996) 1058–1062.

[9] S.C. Kuo, H.Z. Lee, J.P. Juang, Y.T. Lin, T.S. Wu, J.J. Chang, D. Lednicer, K.D. Paull, C.M. Lin, E. Hamel, et al., Synthesis and cytotoxicity of 1,6,7,8-substituted 2-(40

-substi-tuted phenyl)-4-quinolones and related compounds: identi-ﬁcation as antimitotic agents interacting with tubulin, J. Med. Chem. 36 (1993) 1146–1156.

[10] L. Li, H.K. Wang, S.C. Kuo, T.S. Wu, D. Lednicer, C.M. Lin, E. Hamel, K.H. Lee, Antitumor agents. 150. 20_,30_,40_,50_{,5,6,7-substituted} _{2-phenyl-4-quinolones} _and

(10)

related compounds: their synthesis, cytotoxicity, and inhibi-tion of tubulin polymerizainhibi-tion, J. Med. Chem. 37 (1994) 1126–1135.

[11] Y.C. Chen, P.H. Lu, S.L. Pan, C.M. Teng, S.C. Kuo, T.P. Lin, Y.F. Ho, Y.C. Huang, J.H. Guh, Quinolone analogue inhibits tubulin polymerization and induces apoptosis via Cdk1-involved signaling pathways, Biochem. Pharmacol. 74 (2007) 10–19.

[12] S.W. Wang, S.L. Pan, J.H. Guh, H.L. Chen, D.M. Huang, Y.L. Chang, S.C. Kuo, F.Y. Lee, C.M. Teng, YC-1 [3-(50

-hydroxymethyl-20_{-furyl)-1-benzyl indazole] exhibits a novel}

antiproliferative eﬀect and arrests the cell cycle in G0–G1 in human hepatocellular carcinoma cells, J. Pharmacol. Exp. Ther. 312 (2005) 917–925.

[13] V.B. Andela, E.M. Schwarz, J.E. Puzas, R.J. O’Keefe, R.N. Rosier, Tumor metastasis and the reciprocal regulation of prometastatic and antimetastatic factors by nuclear factor kappaB, Cancer Res. 60 (2000) 6557–6562.

[14] K. Brown, S. Gerstberger, L. Carlson, G. Franzoso, U. Siebenlist, Control of I kappa B-alpha proteolysis by site-speciﬁc, signal-induced phosphorylation, Science 267 (1995) 1485–1488.

[15] B. Peruzzi, D.P. Bottaro, Targeting the c-Met signaling pathway in cancer, Clin. Cancer Res. 12 (2006) 3657–3660. [16] G. Hartmann, K.M. Weidner, H. Schwarz, W. Birchmeier,

The motility signal of scatter factor/hepatocyte growth factor mediated through the receptor tyrosine kinase met requires intracellular action of Ras, J. Biol. Chem. 269 (1994) 21936–21939.

[17] A. Bardelli, P. Longati, D. Gramaglia, M.C. Stella, P.M. Comoglio, Gab1 coupling to the HGF/Met receptor multi-functional docking site requires binding of Grb2 and correlates with the transforming potential, Oncogene 15 (1997) 3103–3111.

[18] F.R. Greten, M. Karin, The IKK/NF-kappaB activation pathway-a target for prevention and treatment of cancer, Cancer Lett. 206 (2004) 193–199.

[19] T.W. Chung, S.K. Moon, Y.C. Chang, J.H. Ko, Y.C. Lee, G. Cho, S.H. Kim, J.G. Kim, C.H. Kim, Novel and therapeutic effect of caffeic acid and caffeic acid phenyl ester on hepatocarcinoma cells: complete regression of hepatoma growth and metastasis by dual mechanism, FASEB J. 18 (2004) 1670–1681.

[20] S. Huang, C.A. Pettaway, H. Uehara, C.D. Bucana, I.J. Fidler, Blockade of NF-kappaB activity in human prostate cancer cells is associated with suppression of angiogenesis, invasion, and metastasis, Oncogene 20 (2001) 4188–4197. [21] R. Meyer, E.N. Hatada, H.P. Hohmann, M. Haiker, C.

Bartsch, U. Rothlisberger, H.W. Lahm, E.J. Schlaeger, A.P. van Loon, C. Scheidereit, Cloning of the DNA-binding subunit of human nuclear factor kappa B: the level of its mRNA is strongly regulated by phorbol ester or tumor necrosis factor alpha, Proc. Natl. Acad. Sci. USA 88 (1991) 966–970.

[22] T.W. Chung, Y.C. Lee, C.H. Kim, Hepatitis B viral HBx induces matrix metalloproteinase-9 gene expression through activation of ERK and PI-3K/AKT pathways: involvement of invasive potential, FASEB J. 18 (2004) 1123–1125. [23] M. Arsura, L.G. Cavin, Nuclear factor-kappaB and liver

carcinogenesis, Cancer Lett. 229 (2005) 157–169.

[24] S. Abiru, K. Nakao, T. Ichikawa, K. Migita, M. Shigeno, M. Sakamoto, H. Ishikawa, K. Hamasaki, K. Nakata, K. Eguchi, Aspirin and NS-398 inhibit hepatocyte growth factor-induced invasiveness of human hepatoma cells, Hepa-tology 35 (2002) 1117–1124.

[25] C. Beraud, W.J. Henzel, P.A. Baeuerle, Involvement of regulatory and catalytic subunits of phosphoinositide 3-kinase in NF-kappaB activation, Proc. Natl. Acad. Sci. USA 96 (1999) 429–434.

[26] A. Agarwal, K. Das, N. Lerner, S. Sathe, M. Cicek, G. Casey, N. Sizemore, The AKT/I kappa B kinase pathway promotes angiogenic/metastatic gene expression in colorectal cancer by activating nuclear factor-kappa B and beta-catenin, Oncogene 24 (2005) 1021–1031.