Design, synthesis and biological evaluation of caffeic acid amides as selective MMP-2 and MMP-9 inhibitors

Article abstract of DOI:10.1002/ddr.21038

Strategy, Management and Health Policy Preclinical Research A series of caffeic acid amides with extended P1' groups were synthesized and tested for their inhibitory activities on matrix metalloproteinase (MMP)-1, MMP-2, and MMP-9. Compound 3f showed considerable inhibitory activities against MMP-2, MMP-9, and best selectivity over MMP-1. Preliminary structure-activity relationship analysis and docking studies indicated that caffeic acid amides with electron-donating groups at p-position of amino phenyl group showed better inhibitory activities and selectivity than those with electron-withdrawing groups. The findings of this study would provide information for the exploitation and utilization of caffeic acid as MMP inhibitor for metastatic tumor treatment.

Full text of DOI:10.1002/ddr.21038

DRUG DEVELOPMENT RESEARCH 73 : 343–351 (2012)
Research Article
Design, Synthesis and Biological Evaluation of Caffeic
Acid Amides as Selective MMP-2 and MMP-9 Inhibitors
Zhi-Hao Shi1,3, Nian-Guang Li1,2*, Qian-Ping Shi1,2, Hao Tang1, Yu-Ping Tang1, Wei Li1,2,
Lian Yin1, Jian-Ping Yang1, and Jin-Ao Duan1
1Jiangsu Key Laboratory for High Technology Research of TCM Formulae, Nanjing University of
Chinese Medicine, Nanjing, Jiangsu 210046, China
2Department of Medicinal Chemistry, Nanjing University of Chinese Medicine, Nanjing, Jiangsu
210046, China
3Department of Organic Chemistry, China Pharmaceutical University, Nanjing, Jiangsu 211198,
China
Strategy, Management and Health Policy
Enabling
Preclinical Development Clinical Development
Technology,
Genomics,
Proteomics
Preclinical
Research
Toxicology, Formulation
Drug Delivery,
Pharmacokinetics
Phases I-III
Regulatory, Quality,
Manufacturing
Postmarketing
Phase IV
ABSTRACT
A series of caffeic acid amides with extended P1′ groups were synthesized and tested for
their inhibitory activities on matrix metalloproteinase (MMP)-1, MMP-2, and MMP-9. Compound 3f
showed considerable inhibitory activities against MMP-2, MMP-9, and best selectivity over MMP-1.
Preliminary structure–activity relationship analysis and docking studies indicated that caffeic acid amides
with electron-donating groups at p-position of amino phenyl group showed better inhibitory activities and
selectivity than those with electron-withdrawing groups. The findings of this study would provide infor-
mation for the exploitation and utilization of caffeic acid as MMP inhibitor for metastatic tumor treatment.
Drug Dev Res 73 : 343–351, 2012.
© 2012 Wiley Periodicals, Inc.
Key words: metrix metalloproteinase; caffeic acid; tumor; structure–activity relationship
INTRODUCTION
MMPs are produced and secreted as inactive
zymogens in the extracellular matrix from both tumor
cells and surrounding stromal cells that are stimulated
by the tumor. Two of these, MMP-2 (gelatinase A) and
MMP-9 (gelatinase B), appear to play a key role in these
processes [Crabbe et al., 1994; Aimes and Quigley,
1995].
Matrix metalloproteinases (MMPs) are a family
of zinc-dependent endopeptidases involved in the
breakdown of components of the extracellular matrix
that facilitates connective tissue remodeling [Nagase
and Woessner, 1999]. The latter process is important
in embryonic development, pregnancy, growth, and
wound healing. Normally, MMP activity is tightly con-
trolled by the balance between synthesis of active
enzyme and the presence of endogenous inhibitors,
e.g., the tissue inhibitor of metalloproteinases [Nagase
*Correspondence to: Nian-Guang Li, Jiangsu Key
Laboratory for High Technology Research of TCM Formulae,
and Woessner, 1999]. This balance is lost in tumor Nanjing University of Chinese Medicine, 138 Xianlin Road,
Nanjing, Jiangsu 210046, China.
E-mail: linianguang@njutcm.edu.cn
progression, where an increased expression of certain
MMPs accompanies the passage from a benign to
Received 15 July 2012; Accepted 14 August 2012
malignant phenotype that is believed to be involved in
metastatic tumor dispersion and angiogenesis [Mac-
Published online in Wiley Online Library (wileyonlinelibrary.
Dougall and Matrisian, 1995; Stetler-Stevenson, 1999]. com). DOI: 10.1002/ddr.21038
© 2012 Wiley Periodicals, Inc.

344
SHI ET AL.
In the past, some potent “broad-spectrum” MMP to selective inhibition, whereas the presence of smaller
inhibitors (MMPIs) have been tested against tumors P1′ groups generally leads to broad-spectrum inhibition.
[Ray and Stetler-Stevenson, 1996] and some entered
Considering these data, many efforts have been
clinical trials, but none was approved for human use. directed to develop a more selective second generation
One problem was the design of the clinical trials. of inhibitors against the specific MMPs believed to be
MMPIs were efficacious in animal models of early stage involved in the different pathologies [Li et al., 2009].
disease but were tested in late-stage disease in humans MMPIs may also be derived from natural resources,
[Overall and López-Otín, 2002]. Because they were e.g., herbs, plants, fruits, and other agriculture products
developed as cytostatic agents, they may be used in a have been highlighted in recent years.
prolonged therapy, and therefore their bioavailability
Caffeic acid (3) (Fig. 2), which is found in fruits,
and slow long-term toxicity are key [Hodgson, 1995]. vegetables, wine, olive oil, and coffee [Park et al., 2005],
Another problem was a lack of selectivity with com- has been shown to inhibit the activity of MMP-9 with
pounds such as CGS-27023A (1) (Ciba Geigy, Summit, IC50 values of 10–20 nM. Furthermore, caffeic acid
NJ, USA) and BMS-275291 (2) (Bristol-Myers Squibb, phenylester (4) (Fig. 2), extracted from honeybee pro-
Wallingford, CT, USA) (Fig. 1) showing a severe mus- polis [Grunberger et al., 1988] and synthesized by
culoskeletal syndrome, with fibroproliferative effects in esterification of CA [Nagaoka et al., 2002], selectively
the joint capsule of the knees [Hutchinson et al., 1998; inhibited MMP-2, and MMP-9 but not MMP-1,
Holmbeck et al., 1999; Steward, 1999]. These effects MMP-3, and MMP-7 [Chung et al., 2004].
are thought to occur via impairment of normal tissue
However, the ester groups in caffeic acid are
remodeling controlled by MMP-1 [Dahlberg et al., metabolically labile and thus limited in use [Nakawaza
2000]. For these reasons, a lack of activity with respect and Ohsawa, 1998; Graefe and Veit, 1999; Celli et al.,
to MMP-1 is considered an important factor in reduc- 2004]. Several modified caffeic acid amides have more
ing some of the side effects found for “nonselective” stable characteristics [Rajan et al., 2001].
MMPIs [Scatena, 2000].
These findings prompted us to synthesize a series
Selectivity could potentially be achieved by taking of corresponding caffeic acid amides with extended P1′
advantage of differences in size of the S1′ pocket among group for the purpose to investigate their selective
the various MMPs. From X-ray crystallographic, nuclear inhibitory on MMP-2 and MMP-9, based on their
magnetic resonance (NMR) analysis, and homology inhibitory activities on MMP-1, MMP-2, and MMP-9,
modeling, MMPs may be classified into two broad struc- and structure–activity relationship (SAR) analysis was
tural classes: those with a relatively deep S1′ pocket conducted. The results of this study may be useful to
(MMP-2, MMP-3, MMP-8, MMP-9, and MMP-13) and researchers attempting to find new potential caffeic
those with a shallow S1′ pocket (MMP-1 and MMP-7) acid derivatives as antitumor progression agents.
[Gooley et al., 1994; Lovejoy et al., 1994]. Conse-
quently, incorporation of an extended P1′ group can lead
MATERIALS AND METHODS
Synthesis
O
Reagents and solvents were purchased from com-
O
O
O
O
O
S
N
H
mercial sources and used without further purification
unless otherwise specified. Air- and moisture-sensitive
liquids and solutions were transferred via syringe or
stainless steel cannula. Organic solutions were concen-
trated by rotary evaporation below 45°C at approxi-
mately 20 mm Hg. All nonaqueous reactions were
carried out under anhydrous conditions using flame-
HS
N
HO
O
N
N
N
H
H
H
O
O
N
N
BMS-275291 (2)
CGS-27023A (1)
Fig. 1. Chemical structures of CGS-27023A (1) and BMS-275291 (2). dried glassware within an argon atmosphere in dry,
O
O
O
HO
HO
n
N
HO
HO
HO
HO
OH
R3
H
O
New designed caffeic
acid amide derivatives
Caffeic acid (CA) (3)
Caffeic acid phenethyl ester (CAPE) (4)
P1'
Fig. 2. Chemical structures of caffeic acid, caffeic acid phenethyl ester, and our further designed caffeic acid amide derivatives.
Drug Dev. Res.

CAFFEIC ACID AS MMP INHIBITOR
345
O
O
HO
HO
HO
BOP/Et3N
n
H2N
n
N
OH
+
R
R
H
DMF/CH2Cl2
HO
3
3a-3r
Fig. 3. Synthesis of compounds 3a–3r using BOP as a condensing reagent.
TABLE 1. The Structures, Reaction Time, and Yields of Compounds 3a–3r
Entry
R
n
Time
Prod. (yield)
Entry
R
n
Time
Prod. (yield)
1
2
3
4
5
6
7
8
9
H
H
H
2-OH
3-OH
0
1
2
0
0
0
0
0
0
8 h
8 h
8 h
6 h
6 h
6 h
6 h
6 h
6 h
3a (78)
3b (83)
3c (82)
3d (85)
3e (84)
3f (85)
3g (84)
3h (83)
3i (84)
10
11
12
13
14
15
16
17
18
2-CH3
3-CH3
4-CH3
2-F
3-F
4-F
2-Cl
3-Cl
4-Cl
0
0
0
0
0
0
0
0
0
6 h
6 h
6 h
7 h
7 h
7 h
7 h
7 h
7 h
3j (84)
3k (84)
3l (86)
3m (82)
3n (75)
3o (81)
3p (80)
3q (76)
3r (82)
4-OH
2-OCH3
3-OCH3
4-OCH3
freshly distilled solvents, unless otherwise noted. Yields Compounds 3d–3r were acid anilides with a single dif-
referred to chromatographical seperation, unless other- ferent substituent (OH, OCH3, CH3, F, or Cl) in o-, m-,
wise stated. Reactions were monitored by thin-layer or p-position of benzene ring (Table 1).
chromatography carried out on 0.15~0.20 mm Yantai
silica gel plates (RSGF 254, Yantai Chemical Industry
Research Institute, Yantai, Shandong Province, China)
EXPERIMENTS
using UV light as the visualizing agent. Chromatogra-
phy was performed on Qingdao silica gel (160~200
mesh, Qingdao Banke separation materials Company
Limited, Qingdao, Shandong Province, China) using
petroleum ether (60~90) and ethyl acetate as the
eluanting solvent. 1H NMR spectra were obtained
using a Bruker AV-300 (300 MHz) or Bruker AV-500
(500 MHz) (Bruker Corporation, Karlsruhe, Germany).
Chemical shifts were recorded in parts per million
downfield from tetramethylsilane. J-values were given
in Hertz (Hz). Abbreviations used were singlet (s),
doublet (d), triplet (t), quartet (q), broad (b), and
multiplet (m). Electrospray ionization tandem mass
spectrometry (ESI-MS) was recorded on a Waters
Synapt high definition mass spectrometry (HDMS)
spectrometer.
General Synthetic Method of Caffeic Acid Amides
To a stirring solution of substituted cinnamic acid
(5 mmol) dissolved in DMF (20 ml) was added triethy-
lamine (0.7 ml, 5 mmol) at 0°C. Then the amine
(5 mmol) was added followed by a solution BOP
(5 mmol) dissolved in CH2Cl2 (10 ml). The mixture was
allowed to warm to room temperature and stirred for
another 6–8 h, and then the reaction mixture was
diluted with a large amount of ethyl acetate and washed
with water and brine. The ethyl acetate layer was dried
over MgSO4, filtered and concentrated, and the crude
material was purified by column chromatography
(eluent: ethyl acetate–petroleum ether).
(E)-3-(3,4-dihydroxyphenyl)-N-phenylacrylamide
Using benzotriazole-1-yl-oxy-tris-(dimethylamino)- (3a)
phosphonium hexafluorophosphate (BOP) as a con-
Yield 78% [Fu et al., 2010]. 1H NMR (dimethyl
densing reagent, the free acid reacted with a variety of sulfoxide-d6 (DMSO-d6), 500 MHz) d: 6.53 (d,
aniline in N,N-dimethylformamide (DMF) and CH2Cl2 J = 15.5 Hz, 1H), 6.77 (d, J = 7.5 Hz, 1H), 6.91 (dd,
(Fig. 3) to give the corresponding amide products in J1 = 2 Hz, J2 = 8.0 Hz, 1H), 7.00 (d, J = 2 Hz,1H), 7.04
good yield (Table 1) [Fu et al., 2010]. 1H NMR and (m, 1H), 7.31 (m, 2H), 7.40 (d, J = 15.5 Hz, 1H), 7.68
ESI-MS spectra were consistent with the assigned (d, J = 7.5 Hz, 1H), 7.71 (m, 1H), 9.13 (s, 1H), 9.40 (s,
structures. Compounds 3a–3c were acid anilides with 1H), 10.02 (s, 1H). ESI-MS: m/z 256 [M + H]+, 278
the number of methylene were 0, 1, and 2, respectively. [M + Na]+.
Drug Dev. Res.

346
SHI ET AL.
(E)-N-benzyl-3-(3,4-dihydroxyphenyl)acrylamide
(3b)
6.86–6.93 (m, 3H), 7.02 (d, J = 2.0 Hz, 1H), 7.03–7.08
(m, 2H), 7.36 (d, J = 15.5 Hz, 1H), 8.13 (d, J = 8.0 Hz,
Yield 83% [Rajan et al., 2001]. 1H NMR (DMSO- 1H), 9.06 (s, 1H), 9.17 (s, 1H), 9.39 (s, 1H). ESI-MS:
d6, 500 MHz) d: 4.38 (d, J = 6 Hz, 2H), 6.38 (d, m/z 286 [M + H]+, 308 [M + Na]+.
J = 15.5 Hz, 1H), 6.74 (d, J = 8.0 Hz, 1H), 6.83 (dd,
J1 = 2 Hz, J2 = 8.0 Hz, 1H), 6.94 (d, J = 2 Hz,1H), 7.27
(m, 6H), 8.44 (m, 1H), 9.07 (s, 1H), 9.30 (s, 1H). ESI- (E)-3-(3,4-dihydroxyphenyl)-N-(3-methoxyphenyl)
MS: m/z 270 [M + H]+, 292 [M + Na]+.
acrylamide (3h)
Yield 83% [Fu et al., 2010]. 1H NMR (DMSO-d6,
500 MHz) d: 3.74 (s, 3H), 6.52 (d, J = 15.5 Hz, 1H),
6.62–6.64 (m, 1H), 6.78 (d, J = 8.0 Hz, 1H), 6.90 (dd,
J1 = 2.0 Hz, J2 = 8.0 Hz, 1H), 7.00 (d, J = 2.0 Hz, 1H),
7.18–7.23 (m, 2H), 7.38–7.41 (m, 2H), 9.14 (s, 1H),
9.40 (s, 1H), 10.01 (s, 1H). ESI-MS: m/z 286 [M + H]+,
308 [M + Na]+.
(E)-3-(3,4-dihydroxyphenyl)-N-phenethylacrylamide
(3c)
Yield 82% [Yang et al., 2010]. 1H NMR (DMSO-
d6, 500 MHz) d: 2.76 (m, 2H), 3.39 (m, 2H), 6.31 (d,
J = 15.5 Hz, 1H), 6.73 (d, J = 8.0 Hz, 1H), 6.82 (dd,
J1 = 2 Hz, J2 = 8.0 Hz, 1H), 6.94 (d, J = 2 Hz,1H), 7.23
(m, 4H), 7.29 (m, 2H), 8.05 (m, 1H), 9.06 (s, 1H), 9.29
(s, 1H). ESI-MS: m/z 284 [M + H]+, 306 [M + Na]+.
(E)-3-(3,4-dihydroxyphenyl)-N-(4-methoxyphenyl)
acrylamide (3i)
Yield 84% [Fu et al., 2010]. 1H NMR (DMSO-d6,
500 MHz) d: 3.73 (s, 3H), 6.16 (d, J = 15.5 Hz, 1H),
6.75 (d, J = 8.0 Hz, 1H), 6.95 (dd, J1 = 2 Hz, J2 = 8.0 Hz,
1H), 7.01 (d, J = 2.0 Hz, 1H), 7.53–7.56 (m, 3H), 7.72
(d, J = 8.0 Hz, 2H), 9.07 (s, 1H), 9.47 (s, 1H). ESI-MS:
m/z 286 [M + H]+, 308 [M + Na]+.
(E)-3-(3,4-dihydroxyphenyl)-N-(2-hydroxyphenyl)
acrylamide (3d)
Yield 85% [Hung et al., 2005]. 1H NMR (DMSO-
d6, 500 MHz) d: 6.79 (m, 2H), 6.87 (m, 1H), 6.94 (m,
2H), 7.02 (d, J = 2 Hz,1H), 7.39 (d, J = 16.0 Hz, 1H),
7.81 (d, J = 7.5 Hz, 2H), 7.97 (m, 1H), 9.09 (s, 1H), 9.41
(s, 1H), 9.46 (s, 1H), 9.93 (s, 1H). ESI-MS: m/z 272
[M + H]+, 294 [M + Na]+.
(E)-3-(3,4-dihydroxyphenyl)-N-o-tolylacrylamide
(3j)
(E)-3-(3,4-dihydroxyphenyl)-N-(3-hydroxyphenyl)
acrylamide (3e)
Yield 84% [Fu et al., 2010]. 1H NMR (DMSO-d6,
500 MHz) d: 2.24 (s, 3H), 6.67 (d, J = 15.5 Hz, 1H),
6.77 (d, J = 8.0 Hz, 1H), 6.90 (dd, J1 = 2 Hz, J2 = 8.0 Hz,
1H), 7.01 (d, J = 2 Hz, 1H), 7.07 (m, 1H), 7.17 (m, 1H),
7.22 (d, J = 8.0 Hz, 1H), 7.38 (d, J = 15.5 Hz, 1H), 7.57
(d, J = 8.0 Hz, 1H), 9.09 (s, 1H), 9.28 (s, 1H), 9.38
(s, 1H). ESI-MS: m/z 270 [M + H]+, 292 [M + Na]+.
Yield 84% [Hung et al., 2005]. 1H NMR (DMSO-
d6, 500 MHz) d: 6.44 (d, J = 7.0 Hz, 1H), 6.53 (d,
J = 15.5 Hz, 1H), 6.77 (d, J = 8.0 Hz, 1H), 6.90 (dd,
J1 = 2.0 Hz, J2 = 8.0 Hz, 1H), 7.00 (d, J = 2.0 Hz, 1H),
7.05 (m, 2H), 7.27 (s, 1H), 7.37 (d, J = 15.5 Hz,
1H), 9.13 (s, 1H), 9.33 (s, 1H), 9.39 (s, 1H), 9.88 (s, 1H).
ESI-MS: m/z 272 [M + H]+, 294 [M + Na]+.
(E)-3-(3,4-dihydroxyphenyl)-N-m-tolylacrylamide
(3k)
(E)-3-(3,4-dihydroxyphenyl)-N-(4-hydroxyphenyl)
acrylamide (3f )
Yield 84% [Fu et al., 2010]. 1H NMR (DMSO-d6,
500 MHz) d: 2.29 (s, 3H), 6.53 (d, J = 15.5 Hz, 1H),
6.77 (d, J = 8.0 Hz, 1H), 6.90 (dd, J1 = 2 Hz,
J2 = 8.0 Hz, 1H), 7.01 (d, J = 2 Hz, 1H), 7.19 (m, 1H),
7.38 (d, J = 15.5 Hz, 1H), 7.51 (m, 3H), 9.13 (s, 1H),
9.39 (s, 1H), 9.94 (s, 1H). ESI-MS: m/z 270 [M + H]+,
292 [M + Na]+.
Yield 85% [Seo et al., 2005]. 1H NMR (DMSO-
d6, 500 MHz) d: 6.48 (d, J = 16.0 Hz, 1H), 6.71 (d,
J = 8.0 Hz, 2H), 6.76 (d, J = 8.0 Hz, 1H), 6.88 (dd,
J1 = 2.0 Hz, J2 = 8.0 Hz, 1H), 6.98 (d, J = 2 Hz, 1H),
7.34 (d, J = 16.0 Hz, 1H), 7.46 (d, J = 9.0 Hz, 2H), 9.11
(s, 1H), 9.14 (s, 1H), 9.36 (s, 1H), 9.77 (s, 1H). ESI-MS:
m/z 272 [M + H]+, 294 [M + Na]+.
(E)-3-(3,4-dihydroxyphenyl)-N-p-tolylacrylamide
(3l)
(E)-3-(3,4-dihydroxyphenyl)-N-(2-methoxyphenyl)
acrylamide (3g)
Yield 86% [Fu et al., 2010]. 1H NMR (DMSO-d6,
Yield 84% [Fu et al., 2010]. 1H NMR (DMSO-d6, 500 MHz) d: 2.26 (s, 3H), 6.52 (d, J = 15.5 Hz, 1H),
500 MHz) d: 3.86 (s, 3H), 6.76 (d, J = 8.0 Hz, 1H), 6.77 (d, J = 8.0 Hz, 1H), 6.89 (dd, J1 = 2 Hz, J2 = 8.0 Hz,
Drug Dev. Res.

CAFFEIC ACID AS MMP INHIBITOR
347
1H), 6.99 (d, J = 2 Hz,1H), 7.11 (d, J = 8.0 Hz, 2H), (m, 1H), 7.37–7.43 (m, 1H), 7.47–7.55 (m, 1H), 7.90
7.36 (d, J = 15.5 Hz, 1H), 7.56 (d, J = 8.0 Hz, 2H), 9.12 (m, 1H), 9.15 (s, 1H), 9.44 (s, 1H), 10.21 (s, 1H). ESI-
(s, 1H), 9.38 (s, 1H), 9.93 (s, 1H). ESI-MS: m/z 270 MS: m/z 290 [M + H]+, 312 [M + Na]+.
[M + H]+, 292 [M + Na]+.
(E)-N-(4-chlorophenyl)-3-(3,4-dihydroxyphenyl)
acrylamide (3r)
Yield 82% [Fu et al., 2010]. 1H NMR (DMSO-d6,
(E)-3-(3,4-dihydroxyphenyl)-N-(2-fluorophenyl)
acrylamide (3m)
500 MHz) d: 6.51 (d, J = 15.5 Hz, 1H), 6.77 (d,
J = 8.0 Hz, 1H), 6.91 (dd, J1 = 1.5 Hz, J2 = 8.0 Hz, 1H),
7.01 (d, J = 1.5 Hz, 1H), 7.36–7.43 (m, 3H), 7.70–7.73
(m, 2H), 9.14 (s, 1H), 9.42 (s, 1H), 10.16 (s, 1H). ESI-
MS: m/z 290 [M + H]+, 312 [M + Na]+.
Yield 82% [Fu et al., 2010]. 1H NMR (DMSO-d6,
500 MHz) d: 6.76–6.79 (m, 2H), 6.90–6.92 (dd,
J1 = 2 Hz, J2 = 8.0 Hz, 1H), 7.01 (d, J = 2 Hz, 1H), 7.11–
7.19 (m, 2H), 7.24–7.28 (m, 1H), 7.41 (d, J = 15.5 Hz,
1H), 8.07–8.10 (m, 1H), 9.13 (s, 1H), 9.42 (s, 1H), 9.76
(s, 1H). ESI-MS: m/z 274 [M + H]+, 296 [M + Na]+.
BIOLOGICAL SCREENING
(E)-3-(3,4-dihydroxyphenyl)-N-(3-fluorophenyl)
acrylamide (3n)
MMP-1, MMP-2, and MMP-9 Inhibition Assay
The acid amides were assayed for the inhibition
against MMP-1, MMP-2, and MMP-9 in 96-well micro-
titer plates using succinylated gelatin as the substrate
[Baragi et al., 2000]. The compound and enzyme were
dissolved in sodium borate buffer (pH 8.5, 50 mM) and
incubated at 37°C for 30 min. The substrate was added
and incubated at 37°C for another 60 min. The 100%
and blank groups were also used, in which the 100%
group contained no compound and the blank group
contained only the enzyme. Then 0.03% picrylsulfonic
acid solution was added and incubated at room tem-
perature for additional 20 min. The resulting solutions
were measured under 450 nm to gain optical den-
sity 450 (OD450) values, which were then used to
calculate the inhibitory rates by [OD450(100%) -
OD450(compound)]/[OD450(100%) - OD450(blank)] ¥
100%. The IC50 values were obtained from the previous
inhibitory rates using OriginPro 7.5 software (Origin-
Lab Corporation, Hampton, MA, USA).
Yield 75% [Fu et al., 2010]. 1H NMR (DMSO-d6,
300 MHz) d: 6.50 (d, J = 15.6 Hz, 1H), 6.76 (d,
J = 8.1 Hz, 1H), 6.90 (dd, J1 = 2.1 Hz, J2 = 8.1 Hz, 1H),
6.99 (d, J = 2.1 Hz, 1H), 7.37–7.39 (m, 2H), 7.65–7.72
(m, 3H), 9.15 (s, 1H), 9.42 (s, 1H), 10.23 (s, 1H). ESI-
MS: m/z 274 [M + H]+, 296 [M + Na]+.
(E)-3-(3,4-dihydroxyphenyl)-N-(4-fluorophenyl)
acrylamide (3o)
Yield 81% [Fu et al., 2010]. 1H NMR (DMSO-d6,
300 MHz) d: 6.48 (d, J = 15.6 Hz, 1H), 6.75 (d,
J = 8.1 Hz, 1H), 6.88 (dd, J1 = 2.1 Hz, J2 = 8.1 Hz, 1H),
6.98 (d, J = 2.1 Hz, 1H), 7.11–7.16 (m, 2H), 7.39 (d,
J = 15.6 Hz, 1H), 7.67–7.72 (m, 2H), 9.13 (s, 1H), 9.40
(s, 1H), 10.08 (s, 1H). ESI-MS: m/z 274 [M + H]+, 296
[M + Na]+.
(E)-N-(2-chlorophenyl)-3-(3,4-dihydroxyphenyl)
acrylamide (3p)
Yield 80% [Fu et al., 2010]. 1H NMR (DMSO-d6,
300 MHz) d: 6.75–6.80 (m, 2H), 6.90 (dd, J1 = 1.8 Hz,
J2 = 8.1 Hz, 1H), 7.01 (d, J = 1.8 Hz, 1H), 7.13–7.18 (m,
RESULTS AND DISCUSSION
MMP-1, MMP-2, and MMP-9 inhibitory activity
Table 2 summarizes the results obtained in the
1H), 7.51 (d, J = 15.5 Hz, 1H), 7.63–7.66 (m, 1H), 7.68– assay for the inhibitory activities on MMP-1, MMP-2,
7.72 (m, 1H), 7.88–7.91 (m, 1H), 9.11 (s, 1H), 9.43 and MMP-9; caffeic acid was used as the reference
(s, 1H), 9.48 (s, 1H). ESI-MS: m/z 290 [M + H]+, 312 control, selectivity indices for MMP-2 and MMP-9 over
[M + Na]+.
MMP-1 are also reported, expressed as ratios of their
inhibitory indices (Table 2). The results showed that
these caffeic acid amides had no significant inhibition
activities against MMP-1 supported by the fact that
their IC50s were greater than 1000 nM, thus confirming
(E)-N-(3-chlorophenyl)-3-(3,4-dihydroxyphenyl)
acrylamide (3q)
Yield 76% [Fu et al., 2010]. 1H NMR (DMSO-d6, our strategy for designing MMP-2 and MMP-9 inhibi-
300 MHz) d: 6.48 (d, J = 15.6 Hz, 1H), 6.76 (d, tors (Table 2).
J = 8.1 Hz, 1H), 6.90 (dd, J1 = 2.1 Hz, J2 = 8.1 Hz, 1H),
Among the synthesized amides, the most active
6.99 (d, J = 2.1 Hz, 1H), 7.06–7.10 (m, 2H), 7.30–7.35 compound was 3f, its IC50 values on MMP-2 and
Drug Dev. Res.

348
SHI ET AL.
TABLE 2. The Inhibitory Activity of Compounds 3a–3r Toward Some of the Principal MMPs, the MMP-2 Selectivitya and the MMP-9 Selec-
tivitya (in Parentheses)
b
IC50 (nM)
Product
MMP-1
MMP-2
MMP-9
3a
3b
3c
3d
3e
3f
3g
3h
3i
3j
3k
3l
3m
3n
3o
3p
3q
3r
3,235.42 Ϯ 28.19
3,246.57 Ϯ 27.32
3,301.33 Ϯ 30.14
5,616.56 Ϯ 42.89
5,769.78 Ϯ 40.47
5,833.61 Ϯ 42.16
5,247.92 Ϯ 39.28
5,285.42 Ϯ 41.85
5,319.54 Ϯ 43.79
4,712.73 Ϯ 35.84
4,799.42 Ϯ 37.92
4,829.57 Ϯ 41.47
4,246.83 Ϯ 40.91
4,302.21 Ϯ 38.67
4,366.42 Ϯ 39.52
3,706.63 Ϯ 30.19
3,808.43 Ϯ 31.24
3,963.66 Ϯ 31.97
238.91 Ϯ 2.22
8.05 Ϯ 0.53 (402)
8.22 Ϯ 0.49 (395)
8.20 Ϯ 0.51 (403)
5.09 Ϯ 0.34 (1,103)
1,587.17 Ϯ 21.31 (4)
3.28 Ϯ 0.29 (1,779)
5.27 Ϯ 0.58 (996)
1,679.56 Ϯ 22.53 (3)
3.42 Ϯ 0.49 (1,555)
5.97 Ϯ 0.58 (789)
1,721.82 Ϯ 23.57 (3)
3.55 Ϯ 0.51 (1,360)
8.92 Ϯ 0.62 (476)
1,820.53 Ϯ 24.60 (2)
8.31 Ϯ 0.59 (525)
9.11 Ϯ 0.72 (407)
1,873.02 Ϯ 23.71 (2)
8.95 Ϯ 0.73 (443)
24.26 Ϯ 0.42 (10)
7.14 Ϯ 0.58 (453)
7.25 Ϯ 0.51 (448)
7.28 Ϯ 0.60 (453)
5.28 Ϯ 0.49 (1,064)
1,642.27 Ϯ 22.43 (4)
2.35 Ϯ 0.12 (2,482)
5.67 Ϯ 0.43 (926)
1,632.42 Ϯ 21.38 (3)
3.33 Ϯ 0.42 (1,597)
5.91 Ϯ 0.53 (797)
1,753.85 Ϯ 22.61 (3)
3.64 Ϯ 0.49 (1,327)
7.81 Ϯ 0.66 (544)
1,869.59 Ϯ 22.70 (2)
7.29 Ϯ 0.68 (599)
8.19 Ϯ 0.77 (453)
1,921.13 Ϯ 25.71 (2)
7.89 Ϯ 0.69 (502)
21.22 Ϯ 1.31 (11)
CA
aSelectivity for MMP-1 over each of the other MMPs, is expressed as the ratio of the IC50 value for MMPs over the value for MMP-1.
bIC50 values are mean of four experiments, standard deviation is given.
MMP-9 were 3.28 and 2.35 nM, respectively. Further- (3b, MMP-2 IC50 = 8.22 nM, MMP-9 IC50 = 7.25 nM),
more, compound 3f also showed the highest selectivity and phenylethylamine (3c, MMP-2 IC50 = 8.20 nM,
profile (MMP-1/MMP-2 ratio was 1,779, MMP-1/ MMP-9 IC50 = 7.28 nM) led to slightly less inhibitory
MMP-9 ratio was 2,482). The ortho isomer of 3f, com- activity on MMP-2 and MMP-9 than 3f.
pound 3d, had a good inhibitory activity on MMP-2
(IC50 = 5.09 nM) and MMP-9 (IC50 = 5.28 nM). While
Docking Studies
its meta isomer, compound 3e, had weak activity, (IC50
values for MMP-2 and MMP-9 were 1587.17 and
In the MMP inhibition tests, 3f showed the stron-
1642.27 nM, respectively). Replacing the hydroxyl gest inhibitory activity on MMP-2 and MMP-9, and the
group with electron-donating groups such as methoxy highest selectivity against MMP-1, so the caffeic acid
group or methyl group resulted in little effect on the and 3f were selected for the subsequent molecular
inhibitory activity against MMP-2, MMP-9, and selec- docking experiment with MMP-2 and MMP-9. The
tivity over MMP-1. For example, the p-MeO substi- docking studies of caffeic acid (Fig. 4) showed that
tuted compound 3i (MMP-2 IC50 = 3.42 nM, MMP-9 the carboxylic acid group in caffeic acid could occupy
IC50 = 3.33 nM) and the p-Me substituted compound the deep S1′ pocket in MMP-2 and MMP-9, and the
3l (MMP-2 IC50 = 3.55 nM, MMP-9 IC50 = 3.64 nM) carbonyl group chelated the active site zinc ion with a
proved to be slightly less active than 3f. However, distance of 2.347 Å in MMP-2 and 2.210 Å in MMP-9,
replacing the hydroxyl group with electron-withdrawing respectively. The docking studies of 3f with MMP-2 and
groups such as fluorine or chlorine resulted in MMP-9 (Fig. 5) indicated that 3f interacted well with
decreased inhibitory activity against MMP-2, MMP-9, these two MMPs. The MMP-2 docking of 3f showed
and selectivity over MMP-1. For example, the p-F that the para-hydroxy aniline group in 3f occupied well
substituted compound 3o (MMP-2 IC50 = 8.31 nM, the deep S1′ pocket of MMP-2, and the carbonyl group
MMP-9 IC50 = 7.29 nM) and the p-Cl substituted chelated the active site zinc ion with a distance of
compound 3r (MMP-2 IC50 = 8.95 nM, MMP-9 2.296 Å. The docking studies of 3f with MMP-9 showed
IC50 = 7.89 nM) exhibited an inhibitory activity about that the para-hydroxy aniline group in 3f occupied well
three times lower than 3f (MMP-2 IC50 = 3.28 nM, the deep S1′ pocket of MMP-9, and the carbonyl group
MMP-9 IC50 = 2.35 nM). In addition, replacement chelated the active site zinc ion with a distance of
of the hydroxyanil (3f) with aniline (3a, MMP-2 3.774 Å. These results indicated that compound 3f
IC50 = 8.05 nM, MMP-9 IC50 = 7.14 nM), benzylamine interacted well with MMP-2 and MMP-9 active site,
Drug Dev. Res.

CAFFEIC ACID AS MMP INHIBITOR
349
Fig. 4. Docking result of caffeic acid with MMP-2 (left) and MMP-9 (right). [Color figure can be viewed in the online issue which is available
at wileyonlinelibrary.com]
Fig. 5. Docking result of compound 3f with MMP-2 (left) and MMP-9 (right). [Color figure can be viewed in the online issue which is available
at wileyonlinelibrary.com]
especially the deep S1′ pocket and zinc ion, consistent their IC50s differ significantly. For example, among the
with MMP-2 and MMP-9 assay results.
acid anilides 3a–3c with no substitutions on the amino
phenyl group, anilides with no carbon atom between
the amide and the phenyl group showed more inhibi-
tory activities on MMP-2 and MMP-9 than those with
Discussion and SAR
Although the synthesized acid anilides exhibited one carbon atom or two carbon atoms between the
better inhibitory activity against MMP-2 and MMP-9, amide and the phenyl group. Such as in the inhibitory
Drug Dev. Res.

350
SHI ET AL.
activity on MMP-2, compound 3a with phenyl group indicated that caffeic acid anilides with electron-
showed a good potency with an IC50 was 8.05 nM, donating groups at para-position of amino phenyl group
and replacing the phenyl group with the benzyl group showed better inhibitory activities and selectivity than
and the phenylethyl group, as shown in compounds those with electron-withdrawing groups. Therefore, the
3b and 3c, resulted in slightly lower inhibitory activity findings of this study would facilitate the design of
with their IC50s were 8.22 and 8.20 nM, respectively.
chemical compounds with higher potency to serve as
Furthermore, among all the synthesized anilides, selective MMP-2 and MMP-9 inhibitors, and provide
the most active compound was 3f with one hydroxyl information for the exploitation and utilization of caffeic
group at p-position, it showed the best inhibitory acid as MMPI for metastatic tumor treatment.
potency on MMP-2 and MMP-9 with its IC50s were 3.28
and 2.35 nM, respectively. In addition, 3f was the com-
pound with the highest selectivity profile (MMP-1/
MMP-2 ratio was 1,779, MMP-1/MMP-9 ratio was
2,482). In the series of caffeic acid anilides, replacement
of the hydroxyl group of (3f) at p-position by electron-
donating groups such as methoxy group (3i) or methyl
group (3l) resulted in little effect on the inhibitory activ-
ity against MMP-2, MMP-9, and selectivity over
MMP-1. However, replacement of the hydroxyl group of
(3f) at p-positionbyelectron-withdrawing groupssuchas
fluorine (3o) or chlorine (3r) resulted in the decrease of
inhibitory activity against MMP-2, MMP-9, and selec-
tivity over MMP-1. These results suggested that com-
pounds with electron-donating groups at p-position
showed better inhibitory activities and selectivity than
those with electron-withdrawing groups.
It should be mentioned that anilides with a single
substitute at the para-positions of the benzene ring
exhibited the most powerful inhibitory activities against
MMP-2 and MMP-9, anilides with a substituent at
ortho-positions showed less potent activites, whereas
anilides with a substituent at mata-positions showed
diminished activity. From this fact, it may be reasoned
that the para-position was an essential point for the
inhibition activities against MMP-2 and MMP-9, and
substituents at the ortho-position decreased their inhi-
bition, substituents at the mata-position showed no
inhibitory activities. For example, when a hydroxyl
group was introduced at para-position to form 3f, the
inhibitory activities against MMP-2 and MMP-9 was
the most significant, and when a hydroxyl group was
introduced at ortho-position to form 3d, the activity
decreased slightly. However, when a hydroxyl group
was introduced at mata-position to form 3e, the activity
decreased significantly.
ACKNOWLEDGMENTS
This work was supported by the National Natural
Science Foundation of China (81001382, 81274058),
Program for New Century Excellent Talents by the
Ministry of Education (NCET-09–0163), Research
Fund for the Doctoral Program of Higher Education of
China (20093237120012), the Main Training Fund of
Nanjing University of Chinese Medicine (10XPY02),
333 High-Level Talents Training Project Funded by
Jiangsu Province, Six Talents Project Funded by
Jiangsu Province (2011-D-078), Program for Outstand-
ing Scientific and Technological Innovation Team of
Jiangsu Higher Education (2009), Key Research
Project in Basic Science of Jiangsu College and Univer-
sity (Nos. 06KJA36022, 07KJA36024, 10KJA360039),
Project Funded by the Priority Academic Program
Development of Jiangsu Higher Education Institutions
(ysxk-2010), and Construction Project for Jiangsu Engi-
neering Center of Innovative Drug from Blood-
Conditioning TCM Formulae.
REFERENCES
Aimes RT, Quigley JP. 1995. Matrix metalloproteinase-2 is an inter-
stitial collagenase: inhibitor-free enzyme catalyzes the cleavage of
collagen fibrils and soluble native type I collagen generating the
specific 3/4- and 1/4-length fragments. J Biol Chem 270:5872–5876.
Baragi VM, Shaw BJ, Renkiewicz RR, Kuipers PJ, Welgus HG,
Mathrubutham M, Cohen JR, Rao SK. 2000. A versatile assay for
gelatinases using succinylated gelatin. Matrix Biol 19:267–273.
Celli N, Mariani B, Dragani LK, Murzilli S, Rossi C, Rotilio D. 2004.
Development and validation of a liquid chromatographic-tandem
mass spectrometric method for the determination of caffeic acid
phenethyl ester in rat plasma and urine. J Chromatogr B 810:129–
136.
Chung TW, Moon SK, Chang YC, Ko JH, Lee YC, Cho G, Kim SH,
Kim JG, Kim CH. 2004. Novel and therapeutic effect of caffeic
acid and caffeic acid phenyl ester on hepatocarcinoma cells: com-
plete regression of hepatoma growth and metastasis by dual
mechanism. FASEB J 18:1670–1681.
CONCLUSIONS
In this study, a series of acid amides 3a–3r were
synthesized, and through measurement using a fluoro-
genic substrate assay with human recombinant MMP-1,
MMP-2, and MMP-9, compound 3f showed consider-
able inhibitory activities against MMP-2, MMP-9, and
best selectivity over MMP-1. Preliminary SAR analysis
Crabbe T, O’Connel JP, Smith BJ, Docherty AJP. 1994. Recipro-
cated matrix metalloproteinase activation: a process performed
by interstitial collagenase and progelatinase A. Biochemistry
33:14419–14425.
Dahlberg L, Billinghurst RC, Manner P, Nelson F, Webb G, Ionescu
M, Reiner A, Tanzer M, Zukor D, Chen J, et al. 2000. Selective
Drug Dev. Res.

CAFFEIC ACID AS MMP INHIBITOR
351
enhancement of collagenase-mediated cleavage of resident type II
collagen in cultured osteoarthritic cartilage and arrest with a syn-
thetic inhibitor that spares collagenase 1 (matrix metalloproteinase
1). Arthritis Rheum 43:673–682.
Nagaoka T, Banskota AH, Tezuka Y, Saiki I, Kadota S. 2002. Selec-
tive antiproliferative activity of caffeic acid phenethyl ester ana-
logues on highly liver-metastatic murine colon 26-L5 carcinoma
cell line. Bioorg Med Chem 10:3351–3359.
Fu J, Cheng K, Zhang ZM, Fang RQ, Zhu HL. 2010. Synthesis,
structure and structure-activity relationship analysis of caffeic acid
amides as potential antimicrobials. Eur J Med Chem 45:2638–
2643.
Nagase H, Woessner JF Jr. 1999. Matrix Metalloproteinases. J Biol
Chem 274:21491–21494.
Nakawaza T, Ohsawa K. 1998. Metabolism of rosmarinic acid in rats.
J Nat Prod 61:993–996.
Gooley PR, O’Connell JF, Marcy AI, Cuca GC, Salowe SP, Bush BL,
Hermes JD, Esser CK, Hagmann WK, Springer JP, et al. 1994.
The NMR structure of the inhibited catalytic domain of human
stromelysin-1. Nat Struct Biol 1:111–118.
Overall CM, López-Otín C. 2002. Strategies for MMP inhibition in
cancer: innovations for the post-trial era. Nat Rev Cancer 2:657–
672.
Graefe EU, Veit M. 1999. Urinary metabolites of flavonoids and
hydroxycinnamic acids in humans after application of a crude
extract from Equisetum arvense. Phytomedicine 6:239–246.
Park WH, Kim SH, Kim CH. 2005.
A
new matrix
metalloproteinase-9 inhibitor 3,4-dihydroxycinnamic acid (caffeic
acid) from methanol extract of Euonymus alatus: isolation and
structure determination. Toxicology 207:383–390.
Grunberger D, Banerjee R, Eisinger K, Oltz EM, Efros L, Caldwell
M, Estevez V, Nakanishi K. 1988. Preferential cytotoxicity on
tumor cells by caffeic acid phenethyl ester isolated from propolis.
Experientia 44:230–232.
Rajan P, Vedernikova I, Cos P, Berghe DV, Augustyns K, Haemers
A. 2001. Synthesis and evaluation of caffeic acid amides as anti-
oxidants. Bioorg Med Chem Lett 11:215–217.
Hodgson J. 1995. Remodeling MMPIs. Biotechnology 13:554–557.
Ray JM, Stetler-Stevenson WG. 1996. Matrix metalloproteinases and
malignant disease: recent developments. Exp Opin Invest Drugs
5:323–335.
Holmbeck K, Bianco P, Caterina J, Yamada S, Kromer M, Kuznetsov
SA, Mankani M, Robey PG, Poole AR, Pidoux I, et al. 1999.
MT1-MMP-deficient mice develop dwarfism, osteopenia, arthri-
tis, and connective tissue disease due to inadequate collagen turn-
over. Cell 99:81–92.
Scatena R. 2000. Prinomastat, a hydroxamate-based matrix metallo-
proteinase inhibitor. A novel pharmacological approach for tissue
remodelling-related diseases. Exp Opin Invest Drugs 9:2159–
2165.
Hung CC, Tsai WJ, Yang Kuo LM, Kuo YH. 2005. Evaluation of
caffeic acid amide analogues as anti-platelet aggregation and anti-
oxidative agents. Bioorg Med Chem 13:1791–1797.
Seo WD, Kim JH, Kang JE, Ryu HW, Curtis-Long MJ, Lee HS,
Yang MS, Park KH. 2005. Sulfonamide chalcone as a new class of
alpha-glucosidase inhibitors. Bioorg Med Chem Lett 15:5514–
5516.
Hutchinson JW, Tierney GM, Parsons SL, Davis TRC. 1998.
Dupuytren’s disease and frozen shoulder induced by treatment
with a matrix metalloproteinase inhibitor. J Bone Joint Surg Br
80-B:907–908.
Stetler-Stevenson WG. 1999. Matrix metalloproteinases in angiogen-
esis: a moving target for therapeutic intervention. J Clin Invest
103:1237–1241.
Li NG, Shi ZH, Tang YP, Duan JA. 2009. Selective matrix metallo
proteinase inhibitors for cancer. Curr Med Chem 16:3805–3827.
Steward WP. 1999. Marimastat (BB2516): current status of develop-
ment. Cancer Chemother Pharmacol 43:S56–S60.
Lovejoy B, Cleasby A, Hassell AM, Longley K, Luther MA, Weigl D,
McGeehan G, McElroy AB, Drewry D, Lanbert MH, et al. 1994.
Structure of the catalytic domain of fibroblast collagenase com-
plexed with an inhibitor. Science 263:375–377.
Yang J, Marriner GA, Wang XY, Bowman PD, Kerwin SM, Stavchan-
sky S. 2010. Synthesis of a series of caffeic acid phenethyl amide
(CAPA) fluorinated derivatives: comparison of cytoprotective
effects to caffeic acid phenethyl ester (CAPE). Bioorg Med Chem
18:5032–5038.
MacDougall JR, Matrisian LM. 1995. Contributions of tumor and
stromal matrix metalloproteinases to tumor progression, invasion
and metastasis. Cancer Metastasis Rev 14:351–362.
Drug Dev. Res.