Department of Experimental Biomedical Sciences, Medical School of Padova, Italy
Correspondence: Prof. Spiridione Garbisa, Dept. Experimental Biomedical Sciences, Medical School of Padova, viale G.Colombo 3, 35121 Padova, Italy. E-mail: garbisa{at}unipd.it
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-protease inhibitor but lower than other known natural and
synthetic elastase inhibitors. The flavanol inhibits leukocyte elastase
at concentrations of 50, 150, and 2500 times lower than that effective
on gelatinases (MMP-2 and MMP-9), thrombin, and cathepsin G,
respectively, and also blocks elastase-mediated activation of
MMP-9.
Key Words: proteinases • neutrophils • connective tissue • inhibition
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This enzyme is physiologically counterbalanced by endogenous
serine-proteinase inhibitors—1-proteinase inhibitor (1-PI; also
known as 1-antitrypsin), 2-macroglobulin, and secretory
leukoproteinase inhibitor [3
]—but any enzyme/inhibitor
imbalance may lead directly to increased lysis of extracellular matrix
macromolecules and increased risk of tissue injury in the immediate
vicinity of activated neutrophils [4
]. In particular,
1-PI deficiency is the most prevalent, potentially fatal hereditary
disease in white individuals and is an important risk factor for
pulmonary emphysema [5
]. However, elastase burden may
also be increased as a result of increased recruitment of leukocytes to
the lung, sustained by viral or bacterial pathogens encouraged by
environmental conditions or life habits. There may also be a functional
deficiency of inhibitor(s) as a result of inactivation in the lung by
oxidation (from cigarette smoke or oxygen radicals released from
inflammatory leukocytes) [6
].
As for treatment, in many cases of airway inflammation, exogenous
elastase inhibitors would be first-choice drugs, and direct 1-PI
replacement is one potential therapeutic approach currently under
investigation. Several heterocyclic inhibitors quite specific for LE
have been developed, with constant (Ki) values
in the range of
10-5–10-7 M
[7
]. Peptide chloromethyl ketones are also effective
inhibitors in animal models of emphysema and have served as a standard
of comparison for newly developed inhibitors but present side effects
that make them unsuitable for human therapeutic use [8
].
Conversely, good inhibition of metalloproteinases has been demonstrated
for noncytotoxic concentrations of some natural polyphenol compounds.
One abundant source is green tea [9
], whose polyphenols
are mainly flavanols or "catechins." In this beverage,
epigallocatechin-3-gallate (EGCG) is the most prevalent, followed by
epigallocatechin (EGC); after moderate green tea consumption, its
plasma concentration reaches 0.1–0.3 µM [10
]. Recent
in vivo experiments show that these levels are sufficient to lower
body- and liver-fat accumulation [11
] and inhibit
angiogenesis [12
] and tumor-cell invasive aggressiveness
in vitro [13
14
15
16
]. In particular, EGCG is a direct
inhibitor of MMP-2 and MMP-9, the two gelatinases most frequently
overexpressed in cancer and angiogenesis, instrumental in cutting
through basement membrane barriers [17
18
19
] and
degrading extracellular matrix proteins in human pulmonary emphysema
[20
].
This study shows that the most abundant flavanol of green tea exerts a potent inhibitory effect on human LE but also restrains thrombin activity. LE inhibition takes place at a concentration much lower than that shown on gelatinases [17 ] and urokinase-type plasminogen activator (uPA) [21 ], far from cytotoxicity, in the range of some potent synthetic LE inhibitors [7 ], and lower than that of some commonly used serine-proteinase inhibitors. The results from biochemical and biological assays strongly suggest the pharmacological use of EGCG in the prevention and containment of LE-mediated pathologies.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Substrate degradation by serine-proteinases
Leukocyte and pancreatic elastases, cathepsin G, and thrombin
were solubilized (250 mU/ml) in Hepes buffer [0.1 M Hepes, 0.5 M NaCl,
10% dimethyl sulfoxide (DMSO)] at pH 8.0 (elastase and thrombin) and
7.5 (cathepsin G). All the inhibitors were freshly prepared 5x in the
same buffers, except phenylmethylsulfonyl fluoride (PMSF; 10x), for
which 10% ethanol was added. Elastase substrate was prepared (20x) in
100% ethanol, and cathepsin substrate was prepared (20x) in 100%
DMSO. Dilutions of the inhibitor were premixed with the enzymes in
micrometer wells and maintained for 15 min at 4°C; 5 µl specific
substrate at various concentrations (10–500 µM) was then added to
100 µl final volume, and the mixture was incubated at 37°C. At
20–30-min intervals, the intensity of the color developed by digested
substrate was measured at 405 nm with a Titertek Multiskan (Flow
Laboratories, McLean, VA), and the control background was subtracted.
Logarithmic plot and double-reciprocal plot of the results allowed the
IC50s and the type and Ki of
inhibition exerted on LE by EGCG to be deduced.
Elastin-zymography and gelatin-zymography [23 ] were also used to confirm the inhibition of LE and thrombin by EGCG. Without heating, 10 mU LE and 100 mU thrombin were electrophoresed in 0.15% k-elastin- and 0.1% gelatin-containing 10% polyacrylamide, respectively. The gel was then washed twice for 15 min with 2.5% Triton X-100; cut into slices corresponding to the lanes and then put in different tanks containing the stated concentrations of EGCG; incubated for 72 h at 37°C in Hepes-buffer (as above); stained for 30 min with 30% methanol/10% acetic acid containing 0.5% Coomassie brilliant blue R-250; and destained in the same solution without dye. Clear bands represent areas of gelatinolysis on the blue background.
Elastase-mediated gelatinase activation
Serum-less HT1080 human fibrosarcoma cell-conditioned medium was
used as a source of gelatinases, being pro-MMP-9 the prevalent type. LE
and EGCG were diluted in Hepes buffer to 1 mU/µl and 25 µM,
respectively, and then 5 µl LE was mixed with 18 µl flavanol
solution at 4°C. After 15 min, 150 µl medium was added, and the
volumes adjusted to 180 µl. Following 4 h incubation at 37°C,
60 µl sodium dodecyl sulfate (SDS)-electrophoresis buffer 4x was
added, and after mixing, 1/10 of the volume (24 µl) was processed for
gelatin-zymography (6% polyacrylamide), which was developed overnight
in Tris-buffer (50 mM Tris-HCl, 200 mM NaCl, 10 mM CaCl2,
pH 7.4) and stained as above.
Neutrophil isolation and elastolytic activity
Neutrophils were isolated under endotoxin-free conditions from
buffy-coats of healthy donors, according to a single-step separation
procedure [23
], as previously described in detail
[24
]. The resulting cell population contained 96–98%
neutrophils with traces of eosinophils and mononuclear cells.
The cells were suspended in phenol red-less DME/F12 (Sigma-Aldrich, St. Louis, MO), mixed 1:1 with Hepes buffer containing µM concentration of inhibitor, directly seeded (2.5x105 cells/200 µl) onto plastic microwells and incubated at 37°C in 5% CO2 in air. After 15 min, 5 µg/ml cytochalasin B was added, followed after 5 min by 100 nM fMLP [23 ]. The elastase substrate was then added at 0.5 mM final concentration, and the absorbance was measured at 405 nm, as above, at 20–30-min intervals up to 2 h.
The inhibition of elastolytic activity was also tested in medium conditioned 6 h by the same number of neutrophils, incubated with 0.5 mM substrate ± 5 µM inhibitor. The lysis of the substrate was measured as above, and the inhibition referred as percent of control after 2 h.
Western blotting
Samples of medium conditioned by an equal number of
neutrophils ± 5 µM EGCG were electrophoresed in 10%
polyacrylamide gel in SDS and electroblotted to a Hybond-C Extra
nitrocellulose membrane (Amersham Pharmacia Biotech, Little Chalfont,
UK). The Western blotting was developed by standard procedure using
rabbit anti-human neutrophil elastase or cathepsin G antibody 1:200
(Inalco, Milano, Italy) and horseradish peroxidase-labeled anti-rabbit
immunoglobulin G (IgG) 1:1500 (Sigma Chemical Co.). Antigen detection
was achieved by incubating the membrane for 1 min at room temperature
with 0.125 ml/cm2 enhanced chemiluminescence
(ECL)-detection solution and exposing it to Hyperfilm MP (both from
Amersham Pharmacia Biotech).
Neutrophil-promoted pro-MMP-9 activation
Neutrophils (5x105 in 200 µl Dulbecco’s modified
Eagle’s medium) were seeded onto plastic microwells, activated by
cytochalasin-fMLP as above, and incubated at 37°C in 5%
CO2 in air with or without 5 µM EGCG. After 6 h, 30
µl aliquots of the conditioned medium were clarified and processed
directly for gelatin-zymography.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
 View larger version (17K): [in a new window] |
Figure 1. LE activity in the presence of flavanols. (a) Inhibition of enzymatic
activity (5 mU) by 1 µM flavanols EGCG and EGC compared with control
(•) on increasing concentration of synthetic substrate. (b)
Double-reciprocal plot demonstrating noncompetitive inhibition of LE (5
mU) by increasing concentration of EGCG (S=substrate). Examples of
triplicate experiments, analyzed at 60 min; the points represent the
mean values of three wells, with SD < 10%.
|
1-PI, whereas ovomucoid, aprotinin, and PMSF lacked any
effect (unpublished results). The inhibition by EGCG was maintained
with a constant slope throughout the 2 h of measurement
(unpublished results), and the deduced IC50 was 0.4 µM.
This concentration is approximately 50-fold lower (Fig. 2
) than that shown for MMP-2 and MMP-9 by our group
[19
] and 10,000 that shown for uPA by others
[21
]. Under the same conditions, we deduced a 30-fold
lower IC50 for 1-PI (13 nM) but much higher
values for aprotinin (0.2 mM), PMSF (0.4 mM), and ovomucoid (4 mM). The
inhibition of LE was also verified by k-elastin-zymography,
developed in the presence of increasing concentrations of EGCG;
dose-dependent inhibition of in situ degradation of
k-elastin by 50 mU LE was indeed evident within 0.1–2 µM
flavanol (unpublished results).
 View larger version (14K): [in a new window] |
Figure 2. EGCG IC50 for LE, cathepsin G, and thrombin activity on
synthetic substrates. Dose-response inhibition by the flavanol of LE,
cathepsin G, and thrombin (examples of triplicate experiment, points
average of triplicate digestions, SD<10%) compared with
gelatinase MMP-2 (redrawn from ref. [17
]) and uPA
(redrawn from ref. [21
]).
|
EGCG inhibits degradation of LE substrate by neutrophils
When freshly isolated human peripheral blood neutrophils
(2.5x105), shortly preincubated with increasing
concentrations of EGCG (0.3–9 µM), were activated with cytochalasin
B-fMLP, the degradation of elastase substrate (500 µM) was inhibited
in a dose-dependent manner (Fig. 3 a
). This was revealed by the increased absorbance at 405 nm as a
result of the release of colored digestion products. Almost 30%
inhibition was measured within the first 45 min with 9 µM EGCG, after
which the effect decreased progressively during the following 60 min at
all concentrations.
 View larger version (28K): [in a new window] |
Figure 3. Inhibition of elastase synthetic substrate degradation by human
neutrophils. (a) Freshly isolated neutrophils (2.5x105)
were incubated in microwells in the presence of 0.5 mM substrate and
increasing concentration of EGCG (0.3–9 µM). (b) Neutrophils (as
above) were incubated for 1 h in the presence of 5 µM inhibitor;
the inhibition of elastolytic activity is expressed as percent of that
exerted by EGCG. (c) The medium conditioned for 6 h by control
neutrophils (2.5x105) was incubated with 0.5 mM substrate
in the presence of 5 µM inhibitor. The inhibition is referred to as
percent of degradation in the control after 2 h. (a–c) Lysis of
the substrate was monitored by measuring the absorbance at 405 nm; the
bars represent the mean values of three wells, with SD <
10%. Examples of duplicate experiment.
|
The inhibition by other standard class-specific proteinase inhibitors
was also tested in the cell-free system, on culture medium conditioned
for 6 h by freshly isolated neutrophils and incubated up to 2 h with elastase substrate. While 5 µM EGCG reduced up to 67% of the
substrate degradation measured in the control, this was almost
completely suppressed by the same concentration of 1-PI but only
<15% and <5% by 10 mM ethylenediaminetetraacetate (EDTA) and
N-ethylmaleimide (NEM), respectively, in duplicate
experiments (Fig. 3c)
.
EGCG inhibits MMP-9 activation by purified LE and isolated
neutrophils
The possibility that EGCG may interfere with LE activation of MMPs
[2
], in particular with gelatinases MMP-2 and MMP-9, was
also analyzed. When culture medium containing mostly pro-MMP-9 was
incubated for 4 h with 0.5 mU LE before assaying in
gelatin-zymography, the conversion of the zymogen to the activated form
of MMP-9 was restrained substantially by the presence of 2.5 µM EGCG
(Fig. 4 a
) but not EGC (unpublished results). Under the same conditions, LE
alone did not produce substantial increments of the MMP-2 activated
form.
 View larger version (58K): [in a new window] |
Figure 4. EGCG inhibits activation of pro-MMP-9. Gelatin-zymography of (a) equal
volumes of MMP-9 gelatinase-containing culture medium preincubated for
4 h at 37°C with 0.5 mU LE ± 2.5 µM EGCG, and (b)
culture medium conditioned for 6 h at 37°C by equal number of
neutrophils ± 5 µM EGCG. The conversion of pro-MMP-9 (p) to the
activated form (a) is promoted by LE and neutrophils; this conversion
is restrained by EGCG. (c) Western blotting for LE in equal amount of
medium conditioned by neutrophils as above. LE is already fully
expressed after 1 h, and its secretion is not appreciably modified
by EGCG.
|
EGCG does not reduce elastase and cathepsin G secretion by
neutrophils
To verify whether EGCG restrains LE and cathepsin G secretion,
106 freshly isolated neutrophils were preincubated in
serum-less medium with and without 5 µM flavanol and activated with
cytochalasin A-fMLP as above; then the conditioned medium was analyzed
directly at 1, 2, and 4 h by Western blotting for the two
serine-proteinases. The presence of the flavanol had no measurable
effect on the level of LE (Fig. 4c)
or cathepsin G (unpublished
results) released into the culture medium, which remained constant from
1 h onward.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-PI, the endogenous
inhibitor most involved in balancing elastolytic burst in inflamed
lungs [4
, 5
]. Conversely, EGCG exerts an
inhibition superior to a variety of natural and synthetic inhibitors:
its IC50 is 1/40 that shown for the microbial elastase-like
proteinase inhibitor elastinal [26
] and between 1/50 and
1/200 that of some substituted cephalosporines, ß-lactams and
trifluoro-methyl ketones—the latter ones recently suggested for the
treatment of diseases characterized by neutrophil and LE involvement
[27
]. The inhibition is much superior to that exerted by
some standard class-specific, serine-proteinase inhibitors ovomucoid,
aprotinin, and PMSF which also exert moderate activity on LE
[28
, 29
], and is maintained by EGCG over
the 2-h period of measurement with a constant slope, suggesting a
durable EGCG effect at body temperature. EGCG is also effective on neutrophil culture. In fact, noncytotoxic concentrations of flavanol restrain the elastolytic activity of freshly isolated neutrophils in a dose-dependent manner, although less efficiently compared with the biochemical assays. At any concentration of flavanol, the inhibition of neutrophil elastolytic activity declines progressively over the first 2 h of incubation, although we verified that LE is released into the medium only during the initial phase following activation. The constant slope of the time-course inhibition registered with purified LE should rule out the possibility that this decline may be attributable to the thermo-instability of the flavanol. Most likely, the loss of EGCG potential may be attributable to flavanol oxidation by the hydrogen peroxide secreted after a lag period during a prolonged respiratory burst induced on neutrophils by cytochalasin-fMLP treatment [30 ].
Under the same conditions, aprotinin is 70% less effective in
comparison with the same concentration of flavanol; this moderate
effect is not in conflict with the complete lack of inhibition of LE in
the test tube, because it may be a result of a block of other
serine-proteinases active on the substrate. Certainly, most of the
elastolytic activity released by the neutrophils is attributable to
serine-proteinases as inferred from the complete inhibition of
substrate degradation in the presence of 1-PI and the marginal
inhibition in the presence of metallo-proteinase (1.10 phenanthroline,
EDTA) and cysteine-proteinase inhibitors (pepstatin A, NEM).
Furthermore, because of the two serine-proteinases released mostly by
neutrophils, cathepsin G is, in comparison, largely insensitive to EGCG
(see below), the prevalent serine-proteinase inhibited by µM flavanol
should be LE. Indeed, the presence of LE into the medium was proved by
specific antibodies, which revealed unmodified levels of the proteinase
in the presence of the flavanol. Conversely, the abundant MMP-9
secreted by neutrophils may be, in part, responsible for the
elastolytic metallo-proteinase activity (see below).
Regardless of the respective contribution in elastin degradation, MMP-9 and LE are extracellularly blocked by the flavanol, which thus has the potential to contain the degradative neutrophil activity in an in vivo context in the case of endogenous inhibitor failure. In fact, EGCG has already been shown to inhibit two gelatinases, MMP-2 and MMP-9 [17 18 19 ], whose activity is instrumental to endothelial and tumor cells in cutting through extracellular matrix barriers during angiogenetic and invasive/metastatic processes and pulmonary emphysema [20 ]. We showed that the IC50 of EGCG for these MMPs lies in the range of 10–30 µM [19 ], and the second-most abundant flavanol in green tea, EGC, shows much lower efficacy (1/30). Biochemical assays now reveal that EGCG exerts an even stronger inhibition of LE (50x). This inhibition occurs at concentrations approximately 25-fold and two orders of magnitude lower than the cytotoxic threshold already demonstrated on transformed and normal endothelial cells, respectively [19 ], and similar to those in the plasma of moderate green tea drinkers (0.1–0.3 µM) [10 ]. Again, EGC exerts very little inhibition.
Also, cathepsin G, stored in the azurophilic granules of neutrophils and monocytes and released upon cell stimulation or lysis, can degrade a wide array of matrix and humoral proteins, although less potently than LE [2 ], and activate MMP-9 [31 ]. While cathepsin G release is not affected by µM concentrations of EGCG, the inhibition exerted by the flavanol on its activity is markedly weaker than that on LE (1/2500) but also less pronounced than that on MMP-2 (1/50): the IC50 on cathepsin G is in the mM range, close to that on uPA [21 ], and may contribute to the residual MMP-9 activation at EGCG concentrations over one order of magnitude of the IC50 for LE (see below). The differential effectiveness of EGCG on the activity of LE and cathepsin G may prove to be useful in clinical application of the flavanol, considering the need to preserve a number of cathepsin G-mediated reactions (i.e., coagulation, immune response, and wound debridement) [2 ].
When purified LE is incubated with pro-MMP-9 in a cell-free
system, the conversion of this zymogen into activated form is
restrained substantially in the presence of EGCG. In addition, the
activation of pro-MMP-9 secreted by freshly isolated neutrophils is
somewhat limited when they are cultured briefly in the presence of the
flavanol. This reduced activation may play an important role in airway
inflammation and tumor disease; it would contribute to the
down-regulation of local proinflammatory interleukin (IL)-1ß activity
[32
], restraint of transforming growth factor-ß
(TGF-ß) induction of tumor-cell invasion and angiogenesis
[33
], preservation of the most potent inhibitor of LE
[34
], restraint of neutrophil recruitment by
chemoattractant fragments of 1-PI [35
], preservation
of the underlying elastin structure of the lung [20
,
36
], and containment of degradation of the basement
membrane molecular scaffold, a prerequisite for angiogenesis
[12
] and tumor-cell invasion.
Regarding angiogenesis, elastase has been shown to convert plasminogen into angiostatin [37 ], a potent inhibitor of angiogenesis, and elastase activity inhibition by EGCG could inhibit production of angiostatin, thus promoting inflammation. This paradox may be only apparent; in fact, angiostatin is only one of the endogenous inhibitors of angiogenesis (e.g., endostatin and thrombospondin), and inhibition of LE and inflammation reduces the expression of a number of proangiogenic factors [i.e., fibroblast growth factor (FGF)-ß, platelet-derived growth factor (PDGF), epidermal growth factor (EGF), and cytokines).
Regarding tumor-cell invasion, the inhibition of direct lytic activity of LE on collagen IV [2 ] must also be taken into account as a potential contributor to the documented reduction of tumor-invasive processes [9 , 13 ]. Conversely, the level of MMP-9 expression by neutrophils is not affected substantially by EGCG. Thus, their potential ability to traverse basement membrane barriers and their recruitment at the site of inflammation for immune-surveillance are not likely to be impaired upon treatment with flavanol concentration effective on LE (10-7 M). Indeed, neutrophil motility-proteolysis necessary to transmigrate in vitro through endothelial cell monolayers [38 ] is reduced by EGCG with an IC50 of approximately 10-5 M [39 ].
In addition, thrombin efficiently activates one of the invasion-related gelatinases, MMP-2, as registered in a microvascular endothelial cell model [40 ]. Thrombin is here shown to be inhibited by EGCG at concentrations higher than those effective on MMP-2 [17 , 19 ]; thus, degradation of basement membrane by sprouting endothelial cells during angiogenesis would be restrained primarily by direct inhibition of MMP-2. Conversely, the prevention of cardiovascular pathologies ascribed to green tea consumption [16 ] and the demonstrated antithrombotic activities of catechins [41 ] may not be attributable to direct inhibition of thrombin by EGCG, whose plasma level in green tea drinkers [10 ] reaches 1/200 the IC50 for the proteinase. Even so, the potential use of flavanols as thrombin inhibitors merits deeper investigation.
Furthermore, an IC50 of 8 x 10-8 M has been measured for EGCG on the invasive behavior of tumor cells in vitro [17 , 19 ], over 100-fold lower than that mentioned above for neutrophil transmigration. Whatever lies behind these differences, they offer the possibility of inhibiting disruption of elastic scaffold by LE and tumor-cell aggressiveness, without substantially impairing the physiological traffic of neutrophils, and should certainly facilitate future pharmacological applications of this natural inhibitor.
We conclude that EGCG is a potent inhibitor of LE and an orally available pharmacological agent that may be effective in preventive treatment of individuals exposed to risk of inflammation in general and emphysema in particular, as well as tumor invasion or neoangiogenesis. Regarding the former, the relevance of this conclusion should be confirmed by in vivo experiments; regarding the latter, the role played by LE deserves novel consideration.
Received April 1, 2001; revised August 28, 2001; accepted August 29, 2001.
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-antitrypsin deficiency Chest 117(Suppl. 1),303S 1 proteinase inhibitor complex and A (1–21) fibrinopeptide in human blood stimulated with the calcium ionophore A23187 Biochem. Pharmacol. 49,1513-1520[Medline]This article has been cited by other articles:
 |
|
 |
|
  W. Ding, L. L. Stohl, J. A. Wagner, and R. D. Granstein Calcitonin Gene-Related Peptide Biases Langerhans Cells toward Th2-Type Immunity J. Immunol., November 1, 2008; 181(9): 6020 - 6026. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 I. Dell'Aica, R. Niero, F. Piazza, A. Cabrelle, L. Sartor, C. Colalto, E. Brunetta, G. Lorusso, R. Benelli, A. Albini, et al. Hyperforin Blocks Neutrophil Activation of Matrix Metalloproteinase-9, Motility and Recruitment, and Restrains Inflammation-Triggered Angiogenesis and Lung Fibrosis J. Pharmacol. Exp. Ther., May 1, 2007; 321(2): 492 - 500. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
I. Dell'Aica, L. Sartor, P. Galletti, D. Giacomini, A. Quintavalla, F. Calabrese, C. Giacometti, E. Brunetta, F. Piazza, C. Agostini, et al. Inhibition of Leukocyte Elastase, Polymorphonuclear Chemoinvasion, and Inflammation-Triggered Pulmonary Fibrosis by a 4-Alkyliden-beta-lactam with a Galloyl Moiety J. Pharmacol. Exp. Ther., February 1, 2006; 316(2): 539 - 546. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. Dona, I. Dell'Aica, E. Pezzato, L. Sartor, F. Calabrese, M. D. Barbera, A. Donella-Deana, G. Appendino, A. Borsarini, R. Caniato, et al. Hyperforin Inhibits Cancer Invasion and Metastasis Cancer Res., September 1, 2004; 64(17): 6225 - 6232. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. Fassina, R. Vene, M. Morini, S. Minghelli, R. Benelli, D. M. Noonan, and A. Albini Mechanisms of Inhibition of Tumor Angiogenesis and Vascular Tumor Growth by Epigallocatechin-3-Gallate Clin. Cancer Res., July 15, 2004; 10(14): 4865 - 4873. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Pezzato, M. Dona, L. Sartor, I. Dell'Aica, R. Benelli, A. Albini, and S. Garbisa Proteinase-3 directly activates MMP-2 and degrades gelatin and Matrigel; differential inhibition by (-)epigallocatechin-3-gallate J. Leukoc. Biol., July 1, 2003; 74(1): 88 - 94. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Dona, I. Dell'Aica, F. Calabrese, R. Benelli, M. Morini, A. Albini, and S. Garbisa Neutrophil Restraint by Green Tea: Inhibition of Inflammation, Associated Angiogenesis, and Pulmonary Fibrosis J. Immunol., April 15, 2003; 170(8): 4335 - 4341. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
