A structure-activity study for the inhibition of metalloproteinase-9 activity and gene expression by analogues of gallocatechin-3-gallate

Article abstract of DOI:10.1007/s00018-005-5422-7

Catechins are able to modulate the gelatinolytic activity of matrix metalloproteinase-9 (MMP-9) by reducing its release from macrophages. Gallocatechins decrease MMP-9 secretion by lowering MMP-9 promoter activity and mRNA levels. The effect appears to be dependent on some structural and stereochemical requirements. In this study, the relationship between chemical structure and activity was studied by testing the effect of analogues of (±)-gallocatechin-3-gallate (±)-GCG, selectively deprived of hydroxyl groups, on MMP-9 activity, transcription, and secretion. Our results indicate that (±)-GCG and (±)-catechin-3-gallate are characterized by a substitution pattern compatible with direct inhibition of MMP-9 activity. Conversely, when transcription was the target, (±)-trans-3-flavanol-3- benzoate, lacking all the hydroxyl groups, was the most effective both in lowering MMP-9 promoter activity and consequently protein secretion, and in inhibiting nuclear-factor-κB-driven transcription. Our results suggest that the structural requirements for enzyme inhibition are different from those necessary for targeting gene expression. Birkhaeuser Verlag, 2005.

Full text of DOI:10.1007/s00018-005-5422-7

Cell. Mol. Life Sci. 62 (2005) 2896–2903
1420-682X/05/232896-8
DOI 10.1007/s00018-005-5422-7
© Birkhäuser Verlag, Basel, 2005
Cellular and Molecular Life Sciences
Research Article
A structure-activity study for the inhibition of
metalloproteinase-9 activity and gene expression by
analogues of gallocatechin-3-gallate
M. Dell’Aglia, S. Bellostaa, L. Rizzib, G. V. Gallia, M. Canavesia, F. Rotab, R. Parentea, E. Bosisioa and S. Romeob,
*
a Department of Pharmacological Sciences, University of Milan, Via Balzaretti 9, 20133 Milan (Italy)
b Institute of Medicinal Chemistry, University of Milan, Viale Abruzzi 42, 20131 Milan (Italy),
Fax: + 39 02 50317565; e-mail: sergio.romeo@unimi.it
Received 14 September 2005; received after revision 6 October 2005; accepted 12 October 2005
Online First 28 November 2005
Abstract. Catechins are able to modulate the gelatino- GCG and ( )-catechin-3-gallate are characterized by a
lytic activity of matrix metalloproteinase-9 (MMP-9) by substitution pattern compatible with direct inhibition of
reducing its release from macrophages. Gallocatechins MMP-9 activity. Conversely, when transcription was the
decrease MMP-9 secretion by lowering MMP-9 promot- target, ( )-trans-3-flavanol-3-benzoate, lacking all the
er activity and mRNA levels. The effect appears to be de- hydroxyl groups, was the most effective both in lower-
pendent on some structural and stereochemical require- ing MMP-9 promoter activity and consequently protein
ments. In this study, the relationship between chemical secretion, and in inhibiting nuclear-factor-kB-driven
structure and activity was studied by testing the effect of transcription. Our results suggest that the structural
analogues of ( )-gallocatechin-3-gallate ( )-GCG, selec- requirements for enzyme inhibition are different from
tively deprived of hydroxyl groups, on MMP-9 activity, those necessary for targeting gene expression.
transcription, and secretion. Our results indicate that ( )-
Key words. Gelatinase B; matrix metalloproteinase-9; gene expression; structure-activity relationship; catechins;
gallocatechin-3-gallate; NF-kB driven-transcription
Matrix metalloproteinases (MMPs) are a family of struc- of these proteinases leads to an excessive breakdown of
turally related, zinc-containing enzymes, and more than ECM that occurs in many pathological conditions like
20 different MMPs have been described [1]. MMPs are cancer invasion and metastasis [3], cartilage destruction
secreted as proenzymes and their activity is regulated by in arthritis, atherosclerotic plaque rupture, and the devel-
a pro-domain that maintains the proteinase in a latent opment of aneurysms [4]. Among the several families of
zymogen state through a cysteine switch mechanism MMPs, the 92-kDa gelatinase B (MMP-9) is expressed
[2]. Once activated, MMPs play an important role in by virtually all activated macrophages, facilitating,
the remodeling of the extracellular matrix (ECM) under through the degradation of the basement membrane,
physiological situations, such as fetal tissue development macrophage extravasation. MMP-9 is present in atherec-
and postnatal tissue repair. Conversely, overexpression tomy materials from unstable angina [5] and abdominal
aortic aneurysm [6]. In addition, several studies revealed
that the suppression of MMP-9 reduces the invasive and
metastatic ability of tumor cells [3].
* Corresponding author.

Cell. Mol. Life Sci. Vol. 62, 2005
Research Article
2897
Green tea polyphenols have been reported to suppress interaction, since both (+)-EGCG and (–)-EGCG were
gelatin degradation mediated by MMP-2 and MMP-9 equally active on proteasome, a multicatalytic protease
under several conditions [for a recent review see ref. 7]. responsible for the degradation of most cellular proteins,
A green tea extract rich in polyphenols inhibited MMP-9 probably due to the partial symmetry of their A-C rings
activity up to 80% at 35µg/ml [8] and this effect was cor- [14]. Results obtained in our laboratory [15] showed that
related with the presence of (–)-epicatechin-3-gallate [(–)- only gallocatechin isomers with C2 (R) configuration
(ECG)] and (–)-epigallocatechin-3-gallate [(–)-EGCG)], were able to down-regulate MMP-9 promoter activity.
the most prevalent flavanols in green tea. (–)-EGCG In agreement with our findings, the antitumor-promoting
was also shown to inhibit the secretion of gelatinases activity of catechins is stereochemically dependent [16].
from the highly metastatic human fibrosarcoma cell line Other SAR studies with catechins have focused on the
HT1080 [9]. The mechanism or mechanisms that enable importance of hydroxyl groups (either the number and
catechins to down-regulate MMP-9 secretion have been the position on the flavan skeleton) in modulating their
extensively investigated: recent studies demonstrated that antioxidant and anti-scavenger activities [17, 18], or their
(–)-EGCG-mediated suppression of MMP-9 secretion anti proliferative activity in human stomach cancer cells
was correlated with decreased levels of MMP-9 mRNA, [19].
consequent to the inhibition of ERK 1/2, members of a The role of hydroxyl groups (number and position) of
MAPK family necessary for MMP-9 up-regulation [9]. catechins in determining the effects on MMP-9 activity
Similarly, (–)-EGCG blocked tumor-promoter-induced and transcription has not yet been investigated. Thus,
MMP-9 expression via the suppression of MAPK and the in the present study, analogues of ( )-GCG were syn-
activation of AP-1 in human gastric cells [10]. Several thesized to selectively deprive the flavan skeleton of
studies demonstrated the ability of (–)-EGCG to interfere phenolic hydroxyl groups (fig.1). In general, catechins
with the NF-kB pathway, thus reducing MMP-9 expres- are obtained from natural sources as a mixture of (+)
sion during inflammation and proliferation [for a review, and/or (–) isomers. The total synthesis of (–)-EGCG and
see ref. 7]. Through structure-activity relationship (SAR) of an ester analogue have been recently reported [20, 21].
studies, both the galloyl group and the planarity of the According to these methodologies, the ( )-trans stereoi-
molecules appear to be important elements in conferring somer of gallocatechins is predominantly obtained and
an inhibitory activity against gelatinases [11] and other further synthetic steps are necessary to synthesize the cis
enzymes, such as fatty acid synthase [12] and squalene stereoisomer (corresponding to (–)-EGCG). Therefore,
epoxidase [13]. On the other hand, the stereochemistry for the purpose of our study, only the ( )-trans isomers
of the C ring seems not to be relevant for the enzymatic were prepared because of the easier synthetic accessibil-
Figure 1. Structures of cat-
echin analogues.

2898
M. Dell’Agli et al.
Regulation of MMP-9 gene expression and activity by catechin analogues
ity. The synthesized compounds were tested for their of Cellular and Molecular Medicine, School of Medi-
effects on MMP-9 catalytic activity, transcription, and cine, University of California, San Diego, CA, USA).
secretion. Since MMP-9 gene expression may be regu- A luciferase reporter plasmid with three kB sites from
lated by NF-kB, the effect of the synthesized analogues the E-selectin promoter was previously described [23]
on NF- kB driven transcription was also investigated.
and kindly provided by N. Marx (Department of Internal
Medicine II-Cardiology, University of Ulm, Ulm Ger-
many). The pCMVb-galactosidase plasmid was from
Clontech (Palo Alto, CA, USA). Restriction enzymes
and luciferin were obtained from Promega (Milan, Italy).
Materials and Methods
Chemicals. All reagents and solvents for the synthesis The plasmid purification kit was purchased from Qiagen
were purchased from commercial suppliers (Sigma- (Milan, Italy). The African green monkey cell line CV-1
Aldrich-Fluka-RiedeldeHaën, Milan, Italy; Acros Organ- and human hepatoma HepG2 cells were purchased from
ics, Geel, Belgium; Lancaster Synthesis, Lancaster, UK; the American Type Culture Collection (Manassas, VA,
Iris Biotech, Marktredwitz, Germany) and used without USA). C57BL/6J female mice were purchased from
further purifications. All other chemicals of analytical Charles River (Calco, Italy). Dulbecco’s modified eagle’s
grade were purchased from Merck (Darmstadt, Ger- medium (DMEM) and bovine serum albumin (BSA)
many).
were obtained from Sigma-Aldrich (Milan, Italy).
Cell culture reagents were purchased from Invitrogen
(Life Technologies, Milan, Italy). The plasmid 2.2-Luc, Synthesis of catechin analogues. Compounds 1–10
containing the promoter fused to a luciferase reporter were prepared as outlined in figure2, and their structure
gene [22], was a kind gift of C. K. Glass (Department is given in figure l, thus the aldolic condensation between
Figure 2. Reagents and condi-
tions: (i) NaH, N,N-dimeth-
ylformamides (DMF), 0°C;
(ii) NaBH4, tetrahydrofuran
(THF)/EtOH; (iii) BH3-THF;
(iv) NaOH, H2O2; (v) O-(ben-
zotriazol-1-yl)-N,N,N',N'-
tetramethyluronium
hex-
afluorophosphate, (HBTU),
4-dimethylaminopyridine
(DMAP),4-methylmorpholine
(NMM), DMF; (vi) H2, palla-
dium on carbon, THF/MeOH;
(vii) BnBr, NaH, DMF.

Cell. Mol. Life Sci. Vol. 62, 2005
Research Article
2899
o-hydroxyacetophenones 11 and benzaldehydes 12 (step Transient transfection assay. Transfections of CV-1 and
i) gave calchones 13 [24]. 3-Flavenes 14 were obtained HepG2 cells were performed by the calcium phosphate
by reductive cyclization of calchones with NaBH4 (step co-precipitation technique [29]. Briefly, a unique co-
ii) [20]; the conversion of compound 14 into a trialkyl- precipitate containing each reporter plasmid/luciferase
borane complex (step iii), and the subsequent cleavage by plus pCMVb-gal was prepared and aliquoted in different
hydrogen peroxide gave 3-flavanols 15 (step iv), isolated wells to ensure that all samples were transfected with the
as a mixture of trans and cis diastereoisomers (trans/cis = same amount of plasmid DNA (1.2 µg of luciferase plas-
4/1). Coupling of flavanols 15 with benzoic acids 16 and mid plus 0.3 µg of b-galactosidase plasmid DNA/well).
isolation of the trans isomers afforded esters 17 (step v) After 16h at 37°C, CV-1 cells were washed with PBS
which were then deprotected by catalytic hydrogenation and incubated for 24h in medium containing the com-
(step vi), giving final products 1–10. Flavanols 15f and pounds (0.5–20µM) to be tested or the vehicle (ethanol
15g were prepared from commercially available ( )-cat- or DMSO, 0.1%) in the presence of phorbol myristate ac-
echin (step vii) by selective benzyl protection of phenolic etate 100nM. Four hours post-transfection, HepG2 cells
hydroxyl groups [25].
were incubated in medium without fetal calf serum in the
absence or presence of the compounds (10–30µM).
Cell culture. Mouse peritoneal macrophages were col-
lected by peritoneal lavage with phosphate-buffered sa- Enzyme assay. Luciferase and b-galactosidase assays
line (PBS) from C57BL/6J mice and plated as described were performed using a luminometer (Lumat 9501,
previously [26].
Berthold, Germany) and a microtiter plate reader (Bio-
CV-1 and HepG2 cells were plated in 24-well plates (105 Rad, Hercules, Calif.) respectively, as previously de-
cells/well) the day before transfection in DMEM sup- scribed [29]. Luciferase activities were normalized versus
plemented with 10% (v/v) heat-inactivated fetal bovine galactosidase activities. Results are expressed as the in-
serum.
hibition of normalized luciferase activities versus control
To generate the conditioned media, macrophages were and represent the mean SD values of triplicate samples.
incubated for 24h at 37°C with DMEM/F-12 nutrient Each experiment was repeated at least twice. Statistical
mixture (DMEM/F-12) supplemented with 0.2% BSA in analyses were performed using the Student’s t test.
the absence or presence of the compounds under study
(10µM). At the end of the incubation, the conditioned Cytotoxicity assay. Cellular toxicity caused by the com-
media were collected and the gelatinolytic capacity of pounds was assessed both by measuring cellular protein
secreted MMP-9 was evaluated by zymography. Cellular [27] and by using the 3-(4,5-dimethylthiazol-2-yl)-2,5-
protein content was measured according to Lowry et al. diphenyltetrazolium bromide) (MTT) colorimetric assay
[27]. For quantitation of zymograms, densitometric scan- [30]. CV-1 and HepG2 cells were treated with compounds
ning was performed using a system incorporating a video (up to 20µM) for 24h in DMEM supplemented with 10%
camera and a computer analysis package (NIH Image heat-inactivated FBS. The medium was removed, and
1.52 image analysis software). Each experiment was per- cells were incubated with a solution containing MTT 0.5
formed at least twice with different preparations of cells. mg/ml in PBS at 37°C for 3h. The MTT solution was
Results were normalized by cellular protein content and removed and MTT formazan was extracted with isopro-
expressed as arbitrary optical density (O.D.) units. Data panol:DMSO (9:1, v/v; 500µl/well) for 15min at 37°C.
are presented as the mean S.D. and analyzed using the Aliquots of 100µl were read on a plate reader (Bio-Rad
Dunnett test.
Laboratories) at 560nm (reference wavelength 690nm).
No sign of cytotoxicity was seen with the concentrations
SDS-PAGE zymography. MMP-9 gelatinolytic activity used for the experiments.
was evaluated as previously described [28]. Briefly, sam-
ples underwent electrophoresis on 7.5% polyacrylamide
gels containing 10% SDS and gelatin (1mg/ml). The gels Results
were then washed in 2.5% Triton X-100 (Sigma) at room
temperature and then incubated overnight at 37°C (Tris Effect of catechins on MMP-9 activity. In the first set
50mM pH 7.5 containing NaCl 150mM, CaCl 10mM, of experiments, the ability of catechins 1–10 to affect
2
ZnCl 1µM; activation buffer). At the end of the incu- directly the in vitro gelatinolytic activity of MMP-9 was
2
bation, the gels were stained with Coomassie brilliant investigated. In these experiments, (–)-EGCG was used
blue R-250 (Sigma-Aldrich). To test the direct effect of as reference compound. As shown in figure3, at 10µM,
synthesized compounds 1–10 on the activity of secreted only compounds 1 (–80%), 6 (–70%), 2 and 10 (–20%)
MMP-9, conditioned media obtained from untreated inhibited MMP-9 activity, with compound 1 being even
cells were electrophoresed as described above and the more active than the reference compound (–)-EGCG.
tested compounds added in the activation buffer.
The effect of the other catechins was negligible.

2900
M. Dell’Agli et al.
Regulation of MMP-9 gene expression and activity by catechin analogues
Table 1. Effect of catechin analogues (10 µM) on MMP-9 secretion
and promoter activity.
Percent inhibition versus control (mean SD)
Compound
MMP-9 secretion
47.6 9.9 **
45.7 10.5 **
N.S.
MMP-9 promoter activity
29.0 9.4 *
32.4 12.9 *
N.S.
(–)-EGCG
1
2
3
N.S.
40.6 6.6 **
47.7 14.2 **
37.8 5.1 **
N.S.
4
31.6 3.4 **
N.S.
5
6
N.S.
7
20.5 13.5 *
68.0 1.1 ***
N.S.
26.7 8.7 *
62.0 13.7 ***
N.S.
8
Figure 3. Effect of catechin analogues (10µM) on the activ-
ity of secreted MMP-9. (–)-EGCG (30 µM) was used as reference
compound. Data were quantified by densitometry scanning and
expressed as the mean SD of two experiments performed in dupli-
cate. ** p < 0.01; *** p < 0.001 versus control.
9
10
12.3 1.0**
11.2 5.6*
Data on MMP-9 secretion were quantified by densitometry scanning
and expressed as the mean SD of three experiments performed in
duplicate. Results on MMP-9 promoter activity were expressed as
the inhibition of normalized luciferase activities versus control and
represent the mean SD of three experiments performed in tripli-
cate. N.S., not statistically significant vs controls. * p < 0.05; ** p <
0.01; *** p < 0.001 versus control.
Effect of catechins on MMP-9 secretion and tran-
scription. We then evaluated the effect of the compounds
on MMP-9 secretion by macrophages. As shown in ta-
ble1, compounds 1, 4, 7, 8 and 10 (at 10µM) reduced
MMP-9 secretion by 10–70%, with compound 8 being
the most effective.
The inhibitory effect on MMP-9 promoter-induced tran-
To verify whether the down-regulation of MMP-9 secre- scription by compound 8 was not due to a generalized
tion was consequent to a decreased rate of MMP-9 gene suppression of gene transcription since the compound did
transcription, we tested the effect of the synthesized cat- not affect SV40 early gene promoter-driven transcription
echins on MMP-9 promoter activity. CV-1 cells were used (data not shown).
because they are easily transfectable, unlike macrophage The data in table1 show that the extent of the inhibition of
cell lines, and secrete measurable amounts of MMP-9, MMP-9 secretion by compounds 8, 4, 1, 7 and 10 mirrored
as assessed by gelatin zymography (data not shown). As that observed for promoter activity, and the two effects were
shown in table1, compound 8 was the most active (62.0% correlated (r = 0.87); compounds inactive on the promoter
inhibition); compounds 1, 3, 4, 5 and 7, showed a compa- activity did not affect MMP-9 secretion either. Only for
rable effect (inhibition in the range of 27–48%, difference compounds 5 and 3 we could not see a correlation between
not statistically significant) while compound 10 showed the effect on the promoter and the enzyme secretion.
only a negligible effect (10% inhibition). For comparison,
inhibition by (–)-EGCG was 29%. Compounds 2, 6 and Effect of catechins on NF-kB-driven-transcription.
9 characterized by the removal of one, two or three hy- To determine whether the inhibition of MMP-9 gene ex-
droxyls group from the B ring system, were inactive.
pression by compounds 8, 4 and 1 could be ascribed to a
From concentration-inhibition curves performed with reduction in activity of the NF-kB pathway, we measured
8, 4 and 1, the concentrations required to obtain 50% the effect of these compounds (at 10–30µM) on NF-kB-
inhibition were calculated. As shown in figure4, com- driven transcription. As shown in figure5, the inhibitory
pound 8 was the most potent, with an IC50 of 4.17 0.9 effect on NF-kB exerted by the compounds followed the
µM, whereas for 1 and 4, 50% inhibition was obtained at same order as for the inhibition of MMP-9 promoter ac-
20µM. The inhibition by (–)-EGCG was 55% at 20µM, tivity: 8 > 4 > 1 (72.1%, 53.6%, and 31%, respectively, at
as reported elsewhere [9]. Concentration-dependent 30µM). The inhibition by 1 was similar to that exerted by
curves of the inhibitory effect for 8 and 4 were statisti- (–)-EGCG, the reference compound (35.2% at 30µM).
cally different with respect to 1 (p = 0.0002 for 8 vs 1 and There is a direct correlation (r = 0.79) between the ef-
p = 0.004 for 4 vs 1).
fect on MMP-9 promoter activity and on NF-kB-driven

Cell. Mol. Life Sci. Vol. 62, 2005
Research Article
2901
activity), either directly or by acting on regulators of
transcription such as NF-kB [31], the MAPK family,
and AP1 [10].
The study of the structural requirements necessary for
a molecule to interact with a specific target may supply
valuable information for the molecular modeling and
design of novel inhibitors. SAR studies on catechins and
related compounds have mainly focused on the structural
requirements for radical scavenger and antioxidant ac-
tivity [32–35]. However, polyphenols may affect many
biological activities not only through their antioxidant
effect, but also by interacting with specific molecular
targets in the cell machinery. Elements that may be cru-
cial for the interaction of polyphenols with molecular
targets could be the presence of the galloyl group [11],
the number and the position of hydroxyl groups on the
flavan skeleton [17, 18] and the stereochemistry of the C
ring [15]. Synthetic analogues of (–)-EGCG have been
studied to understand the structure-function relationship
of EGCG and to identify relevant mechanisms of the che-
mopreventive action of EGCG [36–37].
Figure 4. Concentration-inhibition curves for 8, 4 and 1 on the
MMP-9 promoter activity. Results were expressed as the inhibition
of normalized luciferase activities versus control and represent the
mean SD of three experiments performed in triplicate. * p < 0.05;
** p < 0.01 versus control.
In this work we synthesized ten catechin derivatives
characterized by a selective removal of phenolic hy-
droxyl groups. New flavan-3-ols characterized by an A
ring deprived of phenolic hydroxyl groups, as well as
new ester-type catechins, were obtained. Compounds 1,
4 and 6 occur naturally in plants; in particular compound
4, robidanol-3-gallate, has been identified in the aerial
parts of Euphorbia palustris and E. stepposa [38–39].
Compounds 2, 3, 7, 8, 9 and 10 are structures never found
in plants, although compound 2 derives from afzelechin,
widely occurring in plants, and compound 9 derives from
oxykoaburagenin, a catechin previously found in Enkian-
thus nudipes [40]. Compound 5 is a novel structure and
the correspondent catechin has never been isolated.
The synthesized compounds were tested for assessing
the SAR for MMP-9 regulation both at pre- and post-
transcriptional levels.
Figure 5. Effect of compounds 1, 4 and 8 on NF-kB-driven tran-
scription. Results were expressed as the inhibition of normalized
luciferase activities versus control and represent the mean SD of
two experiments performed in triplicate. * p < 0.05; ** p < 0.01
versus control.
Our results indicate that ( )-GCG 1 and ( )-catechin-3-
gallate 6 are characterized by a substitution pattern com-
transcription, indicating that the compounds may repress patible with direct inhibition of MMP-9 activity (fig.3).
MMP-9 expression by interfering with this nuclear tran- All the other compounds prepared were substantially less
scription factor pathway.
active, thus indicating that the hydroxyl group R3 of the B
ring, which has been removed in compound 6, might not
be involved in specific interactions with the enzyme.
Some compounds reduced MMP-9 secretion and gene
expression, although at different levels. When the tar-
Discussion
Polyphenols are abundant components/micronutrients get was MMP-9 gene transcription, very surprisingly,
of the human diet that in vitro have been shown to compound 8, in which all the hydroxyl groups have been
profoundly affect ECM turnover by regulating expres- removed, was the most active with an IC50 of 4.2µM.
sion and activity of MMPs, acting at pre- and post- The effect was much less pronounced for the other cat-
transcriptional levels [7]. As shown in previous studies echins (five fold lower compared to 8). For all the tested
by ourselves and other authors, catechins modulate the compounds, gene transcription down-regulation was mir-
function of MMP-9 at various steps including catalytic rored by a decrease in the amount of protein secreted by
activation and transcription (via inhibition of promoter the macrophages, although compounds 3 and 5 did not

2902
M. Dell’Agli et al.
Regulation of MMP-9 gene expression and activity by catechin analogues
13 Abe I., Seki T., Umehara K., Miyase T., Noguchi H., Sakakibara
J. et al. (2000) Green tea polyphenols: novel and potent inhibi-
tors of squalene epoxidase. Biochem. Biophys. Res. Commun.
268: 767–771
14 Wan S. B., Chen D., Dou Q. P. and Chan T. H. (2004) Study of
the green polyphenols catechin-3-gallate (CG) and epicatechin-
3-gallate (ECG) as proteasome inhibitors. Bioorg. Med. Chem.
12: 3521–3527
15 Bellosta S., Dell’Agli M., Canavesi M., Mitro N., Monetti M.,
Crestani M. et al. (2003) Inhibition of metalloproteinase-9 ac-
tivity and gene expression by polyphenolic compounds isolated
from the bark of Tristaniopsis calobuxus (Myrtaceae). Cell.
Mol. Life Sci. 60: 1440–1448
follow this pattern. At this stage of the research we have
no explanation for this observation, although we cannot
exclude some post-transcriptional effects. Compound 8
was the most active also in modulating NF-kB driven
transcription, indicating that a very lipophylic compound
may have a preferential access at the nuclear level.
Whether this is due to better permeation through cell
membranes, or to a specific adaptability of this molecule
to interact with the molecular target, still remains to be
elucidated.
The results of the present study supply new insights into
how catechin derivatives can act at the transcriptional or
at the enzyme level, indicating that the structural features
required for the enzymatic inhibition are different from
those necessary for the down-regulation of gene expres-
sion. These data may provide new tools for designing
potent and selective agents for the modulation of MMP-9
activity and gene expression.
16 Gao H., Kuroyanagi M., Wu L., Kawahara N., Yasuno T. and
Nakamura Y. (2002) Antitumor-promoting constituents from
Dioscorea bulbifera L. in JB6 mouse epidermal cells. Biol.
Pharm. Bull. 25: 1241–1243
17 Plumb G. W., De Pascual-Teresa S., Santos-Buelga C., Chey-
nier V. and Williamson G. (1998) Antioxidant properties of
catechins and proanthocyanidins: effect of polymerisation, gal-
loylation and glycosylation. Free Radic. Res. 29: 351–358
18 Nakagawa T. and Yokozawa T. (2002) Direct scavenging of
nitric oxide and superoxide by green tea. Food Chem. Toxicol.
40: 1745–1750
19 Kinjo J., Nagao T., Tanaka T., Nonaka G., Okawa M., Nohara T.
et al. (2002) Activity-guided fractionation of green tea extract
with antiproliferative activity against human stomach cancer
cells. Biol. Pharm. Bull. 25: 1238–1240
20 Zaveri N. T. (2001) Synthesis of a 3,4,5-trimetoxybenzoyl ester
analogue of epigallocatechin-3-gallate (EGCG): a potential
route to the natural product green tea catechin, EGCG. Org.
Lett. 3: 843–846
21 Li L. H. and Chan T. H. (2001) Enantioselective synthesis of
epigallocatechin-3-gallate (EGCG), the active polyphenol com-
ponent from green tea. Org. Lett. 3: 739–741
22 Ricote M., Li A. C., Willson T. M., Kelly C. J. and Glass C. K.
(1998) The peroxisome proliferator-activated receptor-gamma
is a negative regulator of macrophage activation. Nature 391:
79–82
23 Brostjan C., Anrather J., Csizmadia V., Natarajan G. and Win-
kler H. (1997) Glucocorticoids inhibit E-selectin expression by
targeting NF-kappa B and not ATF/c-Jun. J. Immunol. 158:
3836–3844
24 Nay B., Arnaudinaud V. and Vercauteren J. (2001) Gram-scale
production and applications of optically pure C-13-labelled
(+)-catechin and (–)-epicatechin. Eur. J. Org. Chem. 12:
2379–2384
25 Tückmantel W., Kozikowski A. P. and Romanczyk L. J., Jr.
(1999) Studies in Polyphenol Chemistry and Bioactivity. 1.
Preparation of building blocks from (+)-catechin. procyanidin
formation: synthesis of the cancer cell growth inhibitor, 3-O-
galloyl-(2R,3R)-epicatechin-4b,8-[3-O-galloyl-(2R,3R)-epi-
catechin]. J. Am. Chem. Soc. 121: 12073–12081
26 Bellosta S., Canavesi M., Favari E., Cominacini L., Gaviraghi
G., Fumagalli R. et al. (2001) Lacidipine [correction of Lal-
soacidipine] modulates the secretion of matrix metalloprotei-
nase-9 by human macrophages. J. Pharmacol. Exp. Ther. 296:
736–743
1 Massova I., Kotra L. P., Fridman R. and Mobashery S. (1998)
Matrix metalloproteinases: structures, evolution and diversifica-
tion. FASEB J. 12: 1075–1095
2 Coussens L. M., Fingleton B. and Matrisian L. M. (2002) Ma-
trix metalloproteinase inhibitors and cancer: trials and tribula-
tions. Science 295: 2387–2392
3 Johansson N., Ahonen M. and Kahari V. M. (2000) Matrix
metalloproteinases in tumor invasion. Cell. Mol. Life Sci. 57:
5–15
4 Beaudeux J. L., Giral P., Bruckert E., Foglietti M. J. and Chap-
man M. J. (2004) Matrix metalloproteinases, inflammation and
atherosclerosis: therapeutic perspectives. Clin. Chem. Lab.
Med. 42: 121–131
5 Brown D. L., Hibbs M. S., Kearney M., Loushin C. and Isner J.
M. (1995) Identification of 92-kD gelatinase in human coronary
atherosclerotic lesions: association of active enzyme synthesis
with unstable angina. Circulation 91: 2125–2131
6 Newman K. M., Ogata Y., Malon A. M., Irizarry E., Gandhi R.
H., Nagase H. et al. (1994) Identification of matrix metallopro-
teinases 3 (stromelysin-1) and 9 (gelatinase B) in abdominal
aortic aneurysm. Arterioscler. Thromb. 14: 1315–1320
7 Dell’Agli M., Canavesi M., Galli G. and Bellosta S. (2005) Di-
etary polyphenols and regulation of gelatinase expression and
activity. Thromb. Haemost. 93: 751–760
8 Demeule M., Brossard M., Page M., Gingras D. and Beliveau
R. (2000) Matrix metalloproteinase inhibition by green tea cat-
echins. Biochim. Biophys. Acta 1478: 51–60
9 Maeda-Yamamoto M., Suzuki N., Sawai Y., Miyase T., Sano
M., Hashimoto-Ohta A. et al. (2003) Association of suppres-
sion of extracellular signal-regulated kinase phosphorylation
by epigallocatechin gallate with the reduction of matrix metal-
loproteinase activities in human fibrosarcoma HT1080 cells. J.
Agric. Food Chem. 51: 1858–1863
10 Kim H. S., Kim M. H., Jeong M., Hwang Y. S., Lim S. H.,
Shin B. A. et al. (2004) EGCG blocks tumor promoter-induced
MMP-9 expression via suppression of MAPK and AP-1 activa-
tion in human gastric AGS cells. Anticancer Res. 24: 747–753
11 Sartor L., Pezzato E., Dell’Aica I., Caniato R., Biggin S. and
Garbisa S. (2002) Inhibition of matrix-proteases by polyphe-
nols: chemical insights for anti-inflammatory and anti-invasion
drug design. Biochem. Pharmacol. 64: 229–237
12 Wang X., Song K. S., Guo Q. X. and Tian W. X. (2003) The
galloyl moiety of green tea catechins is the critical structural
feature to inhibit fatty-acid synthase. Biochem. Pharmacol. 66:
2039–2047
27 Lowry D. H., Rosenbrough N. J., Farr A. L. and Randall R. J.
(1951) Protein measurement with the Folin phenol reagent. J.
Biol. Chem. 193: 265–275
28 Bellosta S., Via D., Canavesi M., Pfister P., Fumagalli R.,
Paoletti R. et al. (1998) HMG-CoA reductase inhibitors reduce
MMP-9 secretion by macrophages. Arterioscler. Thromb. Vasc.
Biol. 18: 1671–1678
29 Crestani M., Stroup D. and Chiang J. Y. (1995) Hormonal regu-
lation of the cholesterol 7 alpha-hydroxylase gene (CYP7). J.
Lipid Res. 36: 2419–2432
30 Denizot F. and Lang R. (1986) Rapid colorimetric assay for cell
growth and survival: modifications to the tetrazolium dye pro-

Cell. Mol. Life Sci. Vol. 62, 2005
Research Article
2903
cedure giving improved sensitivity and reliability. J. Immunol.
Methods 89: 271–277
31 Kim C. H. and Moon S. K. (2005) Epigallocatechin-3-gallate
causes the p21/WAF1-mediated G(1)-phase arrest of cell cycle
and inhibits matrix metalloproteinase-9 expression in TNF-
alpha-induced vascular smooth muscle cells. Arch. Biochem.
Biophys. 435: 264–272
32 Moyers S. B. and Kumar N. B. (2004) Green tea polyphenols
and cancer chemoprevention: multiple mechanisms and end-
points for phase II trials. Nutr. Rev. 62: 204–211
mate (Ilex paraguariensis) and green (Camellia sinensis) teas.
J. Agric. Food Chem. 52: 3583–3589
36 Kazi A., Wang Z., Kumar N., Falsetti S. C., Chan T.-H. and Dou
Q. P. (2004) Structure-activity relationships of synthetic ana-
logs of (–)-epigallocatechin-3-gallate as proteasome inhibitors.
Anticancer Res. 24: 943–954
37 Waleh N. S., Chao W. R., Bensari A. and Zaveri N. T. (2005)
Novel D-ring analog of epigallocatechin-3-gallate inhibits tu-
mor growth and VEGF expression in breast carcinoma cells.
Anticancer Res. 25: 397–402
33 Henning S. M., Niu Y., Lee N. H., Thames G. D., Minutti
R. R., Wang H. et al. (2004) Bioavailability and antioxidant
activity of tea flavanols after consumption of green tea, black
tea, or a green tea extract supplement. Am. J. Clin. Nutr. 80:
1558–1564
34 Xu J. Z.,Yeung S.Y., Chang Q., HuangY. and Chen Z.Y. (2004)
Comparison of antioxidant activity and bioavailability of tea
epicatechins with their epimers. Br. J. Nutr. 91: 873–881
35 Chandra S. and De Mejia Gonzalez E. (2004) Polyphenolic
compounds, antioxidant capacity, and quinone reductase activ-
ity of an aqueous extract of Ardisia compressa in comparison to
38 Sotnikova O. M., Chagovets R. K. and Litvinenko V. I. (1967)
[Polyphenol compounds of Euphorbia palustris]. Farm. Zh. 22:
86–90
39 Bondarenko O. M., Chagovets R. K., Litvinenko V. I., Obolent-
seva G. V. and Syla V. I. (1971) [Euphorbia palustris and step-
posa flavonoids and their pharmacological properties]. Farm.
Zh. 26: 46–48
40 Ogawa M., Hisada S. and Inagaki I. (1972) [Constituents of
Enkianthus nudipes. II. Structure and the absolute configuration
of a new catechin compound from the leaves]. Yakuga. Zasshi
92: 1395–1399