Synthesis, biological evaluation, and molecular modeling of oleuropein and its semisynthetic derivatives as cyclooxygenase inhibitors

Article abstract of DOI:10.1021/jf9033305

Oleuropein, the main phenolic compound in virgin olive oil, and several of its derivatives such as oleuropein aglycone, hydroxytyrosol, and their respective acetylated lipophilic forms were obtained by simple and environmentally friendly semisynthetic protocols. The same molecules were then tested in vitro and in vivo, comparing their intriguing anti-COX-1 and anti-COX-2 properties to those of well-known anti-inflammatory drugs such as ibuprofen and celecoxib. Finally, molecular modeling experiments displaying the most probable binding modes within the classical binding clefts of the enzymes suggest the heme moiety as a potential alternative target. 2009 American Chemical Society.

Full text of DOI:10.1021/jf9033305

J. Agric. Food Chem. 2009, 57, 11161–11167 11161
DOI:10.1021/jf9033305
Synthesis, Biological Evaluation, and Molecular Modeling of
Oleuropein and Its Semisynthetic Derivatives as
Cyclooxygenase Inhibitors
†
A
NTONIO
P
ROCOPIO,*,† STEFANO
A
LCARO,† MONICA
N
ARDI,‡ MANUELA
O
,
LIVERIO
AND
,
§,#
F
RANCESCO
O
RTUSO,† PAOLO
S
ACCHETTA
,
§,# DAMIANA
‡
P
IERAGOSTINO
G
IOVANNI INDONA
S
†
ꢀ
Dipartimento di Scienze Farmacobiologiche, Universita “Magna Græcia” di Catanzaro, Complesso Ninı
‡
ꢀ
Barbieri, 88021 Roccelletta di Borgia (CZ), Italy, Dipartimento di Chimica, Universita della Calabria,
Ponte Bucci, cubo 12C, 87036 Arcavacata di Rende (CS), Italy, §Dipartimento Scienze Biomediche,
Universita “G. d’Annunzio” Chieti-Pescara, Via dei Vestini, 66013 Chieti, Italy, and #Centro Studi
ꢀ
sull’Invecchiamento (Ce.S.I.), Fondazione “G. d’Annunzio”, Via Colle dell’Ara 66013 Chieti, Italy
Oleuropein, the main phenolic compound in virgin olive oil, and several of its derivatives such as
oleuropein aglycone, hydroxytyrosol, and their respective acetylated lipophilic forms were obtained
by simple and environmentally friendly semisynthetic protocols. The same molecules were then
tested in vitro and in vivo, comparing their intriguing anti-COX-1 and anti-COX-2 properties to those
of well-known anti-inflammatory drugs such as ibuprofen and celecoxib. Finally, molecular modeling
experiments displaying the most probable binding modes within the classical binding clefts of the
enzymes suggest the heme moiety as a potential alternative target.
KEYWORDS: Oleuropein; hydroxytyrosol; COX-1 and COX-2 inhibitor; lanthanoid triflates; Lewis acid
catalyst; molecular modeling
INTRODUCTION
neat reactions, and adopting environmentally benign solvents
(e.g., water, nonclassic solvents, etc.) (10, 11). The use of mild
Lewis acid catalysis has quickly increased in recent years (12). For
example, we proposed a series of triflate derivatives as reagents,
allowing several typical organic chemical transformations to
occur in very mild conditions (13-15).
Thus, looking for a tangible application ofthese green chemical
methodologies, we decided to apply our protocols to the chemical
manipulation of a known natural product such as oleuropein. In
this paper we report biological evaluation and molecular model-
ing results carried out with some natural and semisynthetic
oleuropein derivatives obtained by means of chemical manipula-
tion of compound 3.
Epidemiological studies have shown a relationship between the
Mediterranean diet and a lowered incidence of pathologies such
as cardiovascular diseases, cancer, and diabetes (1). Several
studies attribute these health benefits to high consumption of
virgin olive oil;rich in phenols and flavonoids, as well as in other
typical components of the Mediterranean diet. The main phenolic
compounds in virgin olive oil are secoiridoid derivatives of
2-(3,4-dihydroxyphenyl)ethanol (HTyr) and 2-(4-hydroxyphenyl)-
ethanol (Tyr) that occur as either simple phenols or esterified
with elenolic acid to form, respectively, oleuropein (Ole) and
ligstroside aglycones (Figure 1) (2). Many of the biological
activities attributed to these or to other natural phenolic deriva-
tives have anti-inflammatory components (1, 3, 4), so various
health benefits seem to overlap with those attributed to non-
steroidal anti-inflammatory drugs (5, 6). Recently, Bauchamp
et al. (7) reported that oleocanthal, the dialdehydic form of
deacetoxy-ligstroside aglycone, a well-known phenolic com-
pound contained in virgin olive oil, can inhibit cyclooxygenase
enzymes COX-1 and COX-2. Conclusions made in that paper
triggered a great interest in the scientific community. Indeed,
some very long multistep syntheses of oleocanthal were proposed
recently (7-9). The principles of green chemistry have well
emphasized the need to address increasing environmental con-
cerns by adopting more sustainable strategies in fine synthetic
processes, such as reducing the amount of chemicals, performing
MATERIALS AND METHODS
LC/MS/ESI analyses were performed on a Dionex NANO LC
Ultimate working with a C-18 column (i.d. 3 μm, 75 μm ꢀ 15 cm)
and a mobile phase consisting of the following solvents: A (5%
acetonitrile, 95% water þ 0.1% formic acid, and 0.02% TFA); B
(95% acetonitrile, 5% water þ 0.1% formic acid, and 0.02%
TFA). The gradient was as follows: 0-90 min, 70-10% A,
30-90% B; flow rate = 300 μL/min. Mass spectra were recorded
on an Applied Biosystem ibrid Q-STAR XL spectrometer work-
ing in a positive mode. An Applied Biosystem DE-STR/MALDI/
TOF spectrometer was used for MALDI/MS spectra. Samples
were dissolved in a mixture 1:1 of water/acetonitrile þ 5 mmol of
NaCl and suspended in a matrix solution of 2,5-DHB (40 mg/
mL). 1H spectra were recorded on a Bruker WM 300 instrument
on samples dissolved in CDCl3. Chemical shifts are given in parts
per million (ppm) from tetramethylsilane as the internal standard
*Author to whom correspondence should be addressed (telephone
þ39 0961 3694120; fax þ39 0961 3911270; e-mail procopio@unicz.it).
© 2009 American Chemical Society
Published on Web 11/12/2009
pubs.acs.org/JAFC

11162 J. Agric. Food Chem., Vol. 57, No. 23, 2009
Procopio et al.
Figure 1. Chemical structures of compounds 1-7.
(0.0 ppm). Coupling constants (J) are given in hertz. Reactions
were monitored by a GC-MS Agilent workstation, formed by a
GC-6890N (30 m Restek-5SIL capillary column, working in
spitless mode, 1 mL/min He as carrier gas) and by an 5973N mass
detector. MW-assisted reactions were performed in a domestic
oven Daevoo KOR-63A5 with tunneling power from 200 to 800
W. TLC was performed using silica plates 60-F264 on alumina,
commercially available from Merck. Liquid flash chromatogra-
phy was performed on a Supelco versa flash HTFP station on
silica cartridges commercially available from Supelco. A Jasco
liquid chromatography instrument consisting of a ternary pump
PU-1580, a Jasco MD-1510 diode array detector (DAD), and a
manual injection valve (100 μL) was used for HPLC purity
analysis. The mobile phase consisted of the following solvents:
A (H2O þ CF3COOH, pH3.2) andB (CH3CN). Thegradientwas
as follows: 0-5 min, 100% A; 5-10 min, 100% B; 10-15 min,
100% B; 15-20 min, 100% A; flow rate = 1 mL/min. An
Adsorbosphere 5U C-18 column (250 mm ꢀ 4.6 mm) was also
used, and all chromatograms were recorded at 254 nm. All
solvents were distilled before use according to standard methods.
All chemicals were used as commercially available. [All reported
compounds possess a purity of g95% as established by LC/MS
analysis (see Supporting Information).]
Table 1. Microwave Extraction Protocols of Oleuropein in Water
entry
time (min)
% w/w oleuropein
1
2
3
4
5
6
5
10
0.88
2.17
2.00
2.00
2.00
0.5
15
20
30
480a
a Performed in refluxing water by conventional heating.
3.17 ppm (s, 1, OH), 3.49 ppm (t, 2H, H3, JH H = 7.44 Hz), 6.66-6.41
3
2
ppm (m, 1H, HAr), 6.60-6.58 ppm (m, 2H, HAr); MS/EI, m/z = 136 (100)
[M - H2O]þ, 43 (17) [MeCO]þ.
Synthesis of Aglycons (8a-e). Oleuropein 3 (740 mg, 1,34 mmol)
was dissolved in aqueous CH3CN (12,7 mL) in the presence of 10 mol % of
Er(OTf)3 (123.5 mg, 0.20 mmol) and refluxed for 8 h at 80 °C. At the end of
the conversion, the hydrolysate was cooled, 5 mL of water was added, and
the mixture was extracted with CH2Cl2. After drying on Na2SO4, the
organic solvent was removed in vacuo and the crude product was purified
by flash chromatography (mobile phase CH2Cl2/MeOH 8:2 v/v), obtain-
ing the aglycon as a mixture of three isomers (8a-c), a hydrated form (8d),
and a methanolate form (8e), which were all characterized by HPLC,
LC/MS/ESI, and 1H NMR (total yield = 70%). 1H NMR (CDCl3) δ 1.40
Extraction MW-Assisted from Olive Leaves. Samples. The
samples used for the extraction were olive leaves from Coratina cultivar
of Olea europaea, collected, dried for 48 h at 50 °C, milled, and kept at
room temperature until use.
[d, 3H, CH3 (8c), JH
= 6.63 Hz], 1.40 [d, 3H, H10 (8b), JH
= 6.75
CH3H1
10H8
Hz], 2.13-2.28 [m, 2H, H6 (8c), H6a (8c)], 2.50-2.63 [m, 2H, H6 (8b), H6a
(8b)], 2.78-2.85 [t, 2H, H7 (8b), JH H = 5.77 Hz], 2.87 (t, 2H, H7, JH H
Conventional Extraction Procedure. One hundred grams of
dried and milled leaves and 1 L of water were refluxed at 100 °C for 8 h.
Leaves were filtered, and the solution was dried under reduced pressure.
The mixture was treated with acetone and filtered to eliminate the solid,
and the solution was evaporated under reduced pressure to obtain the
crude product. Oleuropein was separated by means of liquid chromato-
graphy on a Supelco Versa-Flash station equipped with a silica cartridge
and eluted with a mixture of CH2Cl2/MeOH 8:2 v/v as the mobile phase.
MW-Assisted Extraction Procedure. One hundred grams of
dried and milled leaves and 800 mL of water were placed in a Pyrex round-
bottom flask equipped with a jacketed coiled condenser in a domestic
microwave oven. After extraction [10 min of microwave irradiation at 800
W, shorter or longer extraction times did not improve the amounts of
oleuropein obtained (Table 1)], leaves were filtered and the solution was
dried under pressure. The mixture was filtered with acetone to eliminate
the solid residue, and the solution was evaporated under pressure to obtain
the crude product. Oleuropein was separated by liquid chromatography
on a Supelco versa-flash station equipped with a silica cartridge and
a mixture of CH2Cl2/MeOH 8:2 as mobile phase. The purity was
determined by HPLC and NANO/LC/MS/ESI. LC/MS/ESI (þ), m/z
558 (31) [M þ NH4]þ, 541 (16) [M þ H]þ, 379 (76) [M - Glu þ NH4]þ, 361
(100) [M - Glu þ H]þ
=
8
7
8
7
5.85 Hz), 3.35 [m, 1H, H4 (8b), JH H = 8.89 Hz], 3.74 [s, 3H, OCH3 (8c)],
4
5
3.77 [s, 3H, OCH3 (8b)], 4.10-4.50 [m, 6H, H5 (8c), H5 (8b), 2H8 (8c), 2H8
(8b)], 6.62[m, 2H, Haromatic (8b)], 6.75 [m, 2H, Haromatic (8c)], 7.26 [s, 2H,
Haromatic (8b, 8c)], 7.58 [s, 1H, H3(8c)], 9.52 [s, 1H, H1(8b)], 9.60 [s, 1H, H3
(8b)], 9.80 [s, 1H, H10 (8c)]. The cyclic isomer 8a is present only in trace in
the 1H NMR spectrum. LC/MS/ESI (þ), tr = 22.70 min 8a-c, m/z 379
(100) [M þ H]þ, 347 (11) [M - MeOH þ H]þ, 243 (4) [M - H]þ, 225 (12),
135 (26); tr = 27.08 min 8d (R = OH), m/z 361 (100) [M - OH þ H]þ, 137
(40); tr = 28.11 8e (R = OMe), m/z 393 (1) [M - OH]þ, 361 (100) [M -
MeOH - OH]þ, 137 (11).
Synthesis of Peracetylated Hydroxytyrosol (9). Purified
hydroxytyrosol (200 mg, 1,3 mmol) was reacted for 2 h at room temperature
with an excess of dry acetic anhydride (5 mL) under N2 atmosphere in the
presence of 2 mol % of Er(OTf)3 (0.026 mmol, 15.96 mg). At the end of the
conversion, 12 mL of MeOH was added to the mixture and stirred for 1 h.
Solvent was removed under vacuum, and the residue was solubilized in
CH2Cl2. The mixture was extracted with a satured solution of NaHCO3 until
complete elimination of the acetic acid; the organic phases collected were
dried on Na2SO4, and filtered, and the solvent was evaporated under
pressure. The crude was purified by flash chromatography on silica gel
(mobile phase CH2Cl2/MeOH 8:2) to isolate the pure product (80% yield):
1H NMR (CDCl3) 2.03 ppm (s, 3H, HAc), 2.28 ppm (s, 2 ꢀ 3H, HAc.ar.), 2.93
ppm (t, 2H, H7, JH H = 6.90 Hz), 4.27 (t, 2H, H8, JH H = 6.90 Hz),
Synthesis of Hydroxytyrosol (1). The synthesis of hydroxytyro-
sol was performed in a classical manner starting from the treatment of
3,4-dihydroxyphenylacetic acid with EtOH in the presence of 10% v/v of
H2SO4 and reduction of the ethyl ester with NaBH4, as reported in the
literature (1): 1H NMR (CDCl3) 2.53 ppm (t, 3H, H2, JH H = 7.44 Hz),
7
8
8
7
7.05-7.11 ppm (m, 3H, HAr); MS/EI, m/z = 238 (1) [M - CH2CO]þ, 220
(6) [M - CH3COOH]þ, 178 (18) [M - CH3COOH - CH2CO]þ, 136 (100)
[M - CH3COOH - 2 ꢀ CH2CO]þ, 43 (24) [MeCO]þ.
2
3

Article
Synthesis of Peracetylated Oleuropein (10). Purified oleuro-
J. Agric. Food Chem., Vol. 57, No. 23, 2009 11163
ethanol mixture and prolonged extraction times (entry 6 in
Table 1).
pein 3 (250 mg, 0,463 mmol) was allowed to react for 2 h at room
temperature with an excess of dry acetic anhydride (5 mL) under N2
atmosphere in the presence of 2 mol % of Er(OTf)3 (0.009 mmol, 6.5 mg).
At the end of the conversion, 12 mL of MeOH was added to the mixture
and stirred for 1 h. The solvent was removed in vacuo, and the residue was
solubilized in CH2Cl2. The mixture was extracted with a saturated solution
of NaHCO3 until complete elimination of the acetic acid; the organic
phases collected were dried on Na2SO4 and filtered, and the solvent was
evaporated under pressure. The crude was purified by flash chromato-
graphy on silica gel (mobile phase CH3Cl/MeOH 98:2) to isolate the pure
Examining the molecular structure ofthis secoiridoid, it isquite
evident that its aglycone form can be easily obtained by simple
acetal hydrolysis, mimicking the natural glucosidase enzyme
action (Figure 2). We successfully tested, and then patented (16),
a very gentle method to realize the selective Lewis acid catalyzed
oleuropein hydrolysis in water, applying one of the protocols we
recently published for simple acetals (13). Selective hydrolysis of
oleuropein is a challenging aim to achieve, due to the multi-
functional chemical nature of the molecule. In the practical
approach reported in Figure 2, an aqueous acetonitrile oleuropein
solution was treated with 10 mol % of a lanthanide(III) salt under
reflux until complete disappearance of the starting material,
monitored by thin layer chromatography (tlc) analysis. The
complex product characterization was performed by means of
LC-MS and 1H NMR analysis. As shown in Figure 2, the
hemiacetal aglycone product was a mix of four tautomeric forms
(aglycones 8a-d): the enolic and the dialdehydic forms, 8a and
8b, respectively, as well as the ring-closed species 8c and the
hydrated aldehydic derivative 8d. Moreover, LC-MS analysis
detected the methanolic derivative 8e, probably arising from a
reaction with the mobile phase (see the Supporting Information).
Several studies carried out with human and animal models have
shown good bioavailability of olive oil phenols (17). Nevertheless,
these active substances are effective at concentrations that are
improbable to be achieved in vivo. On the other hand, no data are
available about the bioavailability of the various aglycons,
including oleocanthal, and contrasting evidence has been re-
ported regarding their in vivo stability (18, 19).
product (65% yield): 1H NMR (CDCl3) δ 1.69 (d, 3H, H10, JH
= 7.13
10H8
Hz), 2.01-2.02-2.03 (s, 5 ꢀ 3HAc), 2.28 (s, 2 ꢀ 3HAc), 2.72 (dd, 1H, H6,
JH H = 14.58 Hz, JH H = 4.41 Hz), 2.41 (dd, 1H, H6a, JH = 14.58 Hz,
6aH6
6
6a
6
5
JH
6aH5
= 8.81 Hz), 2.91 (t, 2H, H7, JH H = 7.13 Hz), 3.72 (s, 3H,
7 8
CH3OCO), 3.96 (dd, 1H, H5, JH H = 8.81 Hz, JH H = 4.41 Hz),
5
6a
5
6
0
0
0
4.06-4.28 (m, 4H, H6 , H4 , H8), 5.03 (d, 1H, H1 , JH
= 7.80 Hz), 5.12
H20
10
0
0
(dd, 1H, H2 , JH
= 9.49 Hz, JH
= 7.80 Hz), 5.27 (t, 1H, H3 ,
H30
H10
20
20
0
JH
30
= JH
= 9.49 Hz), 5.69 (s, 1H, H1), 5.94-6.03 (m, 1H, H5 ), 7.46
H40
H10
20
(s, 1H, H3), 7.02-7.14 (m, 3H, Haromatic); LC/MS/ESI, m/z 810 [M þ
NH4]þ, 793 [M þ H]þ.
Synthesis of Peracetylated Aglycons (11a-e). Aglycon (297
mg, 0,78 mmol) was reacted for 3 h at room temperature with an excess
of dry acetic anhydride (5 mL) under N2 atmosphere in the presence of
3 mol % of Er(OTf)3 (0.023 mmol, 14.4 mg). At the end of the conversion,
12 mL of MeOH was added to the mixture and stirred for 1 h. Solvent was
removed in vacuo, and the residue was solubilized in CHCl3. The mixture
was extracted with a saturated solution of NaHCO3 until complete
elimination of the acetic acid; the organic phases collected were dried on
Na2SO4 and filtered, and the solvent was evaporated under pressure. The
crude was purified by flash chromatography on silica gel (mobile phase
CH2Cl2/MeOH 8:2) to isolate the pure product as a mixture of different
peracylated forms (11a-e) (total yield = 79%) analyzed by LC/MS/ESI,
MALDI/MS, and 1H NMR.
On the basis of these considerations, it would be useful to
conveniently produce these species as chemically more stable
derivatives that are able to be biochemically converted in vivo
into their original active forms. It was already reported that some
acetyl derivatives of phenolic compounds maintain biological
antioxidant activity compared to that of the parent compound,
probably because of extensive deacetylation of hydroxytyrosyl
acetate by carboxylesterases. Such deacetylation can take place
either within the cell, upon absorption of the acetylated molecule,
or in the extracellular space by secreted esterases (17), generating
free hydroxytyrosol, whichis the effective antioxidant compound.
Considering that the acetyl group is a ubiquitous substrate in the
biochemical processes and that the acetylating agents are very
common and manageable, we planned to prepare the acetyl
derivatives of 1, 3, and 8a-e. To obtain chemically stable forms
of oleuropein and its derivatives, we successfully tested and then
patented(20)thesimpleacetylationprotocol, reportedinFigure3,
derived by applying methods we previously developed (14, 15).
The acetylated analogues of 1, 3, and 8a-e were synthesized in
acetic anhydride in the presence of lanthanide(III) catalysts and
then characterized spectroscopically (Figures 3 and 4). All
synthesized compounds showed a high purity grade, as shown
in the corresponding HPLC chromatograms (see the Supporting
Information). To demonstrate the improved lipophilic character
of the acetylated species obtained from 1, 3, and 8a-e, that is,
triacetylated hydroxytyrosol 9 and peracetylated oleuropein 10,
their octanol-water partition coefficients were determined
using a modification of the experimental method described by
Namjesnik-Dejanovic et al. (21). All acetylated species were much
more lipophilic than their corresponding nonacetylated original
molecules. A good correlation was observed between experimen-
tal and theoretical values of log Ko/w (see the Supporting
Information) (21-23).
11a: 1H NMR (CDCl3) 9.6 ppm (m, 1H, Hal9), 9.7 ppm (m, 1H, Hal1),
2.28 ppm (s, 3H, HAc23), 2.30 ppm (s, 3H, HAc22); MALDI/MS (2,5-
DHB), m/z 484 (11) [M þ Na]þ; LC/MS/ESI (þ), tr = 30.406 min, m/z =
485 (21) [M þ Na]þ, 480 (100) [M þ NH4]þ, 463 (52) [M þ H]þ, 445 (51)
[M - OH]þ, 431 (43) [M - MeOH]þ.
11b: 1H NMR (CDCl3) 9.2 ppm (m, 1H, Hal1), 2.30 ppm (s, 3H, HAc22),
2.28 ppm (s, 3H, HAc23), 2.19 ppm (s, 3H, HAc9), 1.75-1.87 ppm (m, 1H,
H5 ꢀ 3); MALDI/MS (2,5-DHB), m/z 526 (12) [M þ Na]þ; LC/MS/ESI
(þ), tr = 29.372 min, m/z 527 (25) [M þ Na]þ, 522 (100) [M þ NH4]þ, 505
(48) [M þ H]þ, 445 (57) [M - COOMe]þ.
11c: 1H NMR (CDCl3) 660 ppm (m, 1H, OH2), 2.30 ppm (s, 3H,
HAc23), 2.28 ppm (s, 3H, HAc24), 2.20 ppm (s, 3H, HAc10), 2.07 ppm (s, 3H,
HAc1), 1.66-1.73 ppm (m, 3 H, (CH3)6); MALDI/MS (2,5-DHB), m/z 586
[M þ Na]þ; LC/MS/ESI (þ) tr = 27.587 min, m/z 587 (31) [M þ Na]þ, 582
(98) [M þ NH4]þ, 445 (100) [M - COOMe - HOAc]þ.
11d: 1H NMR (CDCl3) 3.70 ppm (s, 3 H, (OCH3)3), 2.30 ppm (s, 3H,
HAc23), 2.28 ppm (s, 3H, HAc24), 2.22 ppm (s, 3H, HAc10), 1.97 ppm(s, 3H,
HAc1); MALDI/MS (2,5-DHB), m/z 600 (30) [M þ Na]þ; LC/MS/ESI (þ)
tr = 26.8, 28.742 min, m/z 601 (15) [M þ Na]þ, 596 (100) [M þ NH4]þ, 579
(91) [M þ H]þ, 445 (73) [M - COOMe - MeOAc]þ.
11e: 1H NMR (CDCl3) 2.30 ppm (s, 3H, HAc23), 2.28 ppm (s, 3H,
HAc24), 2.19 ppm (s, 3H, HAc10), 2.02 ppm (s, 6H, HAc1 þ HAc3),
1.73-1.80 ppm (m, 3 H, (CH3)6); MALDI/MS (2,5-DHB), m/z 628 (29)
[M þ Na]þ; LC/MS/ESI (þ) tr = 33.702, 34.551, 35.083, 38.512 min, m/z
629 (15) [M þ Na]þ, 624 (86) [M þ NH4]þ, 582 (13) [M - Ac þ NH4]þ,
547 (86) [M - COOMe]þ, 505 (100) [M - COOMe - Ac]þ, 445 (38)
[M - COOMe - Ac - HOAc]þ.
RESULTS AND DISCUSSION
First, we decided to improve the oleuropein (3) extraction
protocol from olive leaves by using microwave heating and
environmentally friendly solvents. By virtue of this approach,
maximum yield was obtained in a very short extraction time,
10 min, and using plain water as the solvent. It should be noted
that the same results can be obtained with a conventional heating
system, but only by means of refluxing water/methanol or an
In the next step, the same molecules (1, 3, 8a-d, 9, 10, and
11a-d) were in vitro tested as COX-1 and COX-2 inhibitors by

11164 J. Agric. Food Chem., Vol. 57, No. 23, 2009
Procopio et al.
Figure 2. Synthesis of aglycone derivatives by acetal hydrolysis.
Figure 3. Patented acetylation protocol.
means of the evaluation of percentage inhibition in the produc-
tion of their physiological mediators in the human whole blood
(HWB) assay, respectively corresponding to thromboxane B2
(TXB2) and the prostaglandin E2 (PGE2) (24) (see the Support-
ing Information). In the preliminary in vitro screening, carried
out at three different concentrations (see the Supporting
Information), compounds 1 and 9 were shown to be the most
active inhibitors. Thus, a concentration-response investiga-
tion was performed for peracetylated hydroxytyrosol 9 only
(Table 2).
are 16 and 0.54 μM for COX-1 and COX-2, respectively.
Therefore, it is worth noting that in the case of peracetylated
hydroxytytrosol 9, the inhibition data IC50 against COX-1
and COX-2 were in the range of 3-10 and 0.1-1 μM,
respectively, quite lower than the values reported for
(-)-oleocanthal 7 (IC50 of about 25 μM for both COX-1 and
COX-2) (7).
To confirm the exciting in vitro results, we decided to screen
in vivo all synthesized compounds for their anti-inflamma-
tory activity using the well-known carrageenan-induced
paw edema (27). The synthesized compounds (1, 3, 8a-d, 9,
10, and 11a-d) and standard drugs such as ibuprofen and
celecoxib were administered orally (50 μmol/kg po), and after
As reported earlier (25), the concentration for 50% inhibi-
tion (IC50) values for ibuprofen are 5 and 223 μM for COX-1
and COX-2, respectively. The IC50 values for celecoxib (26)

Article
J. Agric. Food Chem., Vol. 57, No. 23, 2009 11165
Figure 4. Patented acetylation protocol.
Table 2. Concentration-Response Table for the Inhibition of Human Whole
Blood COX-1 and COX-2 Activities by Peracetylated Hydroxytyrosol 9
our compounds with respect to both classic (site A) and
tentative accessory (site B) clefts. According to the Glide
methodology, in which the heme iron is considered at its
maximum oxidation state (þ3) (30), the docking binding site
has been defined by means of a regular box centered onto the
chain A tyrosine 385. Its volume was about 390,000 A, widely
covering classic COX binding site and the heme moiety (see the
Supporting Information). Compounds 1, 3, 9, and 10 showed,
as appeared in the docking experiments, good affinity toward
both COX isoforms. Interestingly, all ligands were able
to be recognized, with some differences, at both sites A
and B (see the Supporting Information). On the other hand,
disagreement has been observed between experimental and
theoretical affinity data. 9 and 10 have reported a more favor-
able interaction with respect to 1 and 3 in all cases (see the
Supporting Information). Such a discrepancy can be explained
by analyzing the binding modes of our compounds at the COX
site A. Actually, only 1 and 3 deeply recognized site A,
reporting quite similar configurations with respect to the other
known NSAIDs. 9 and 10 have partially occupied binding
site A, interfering only with the catalytic gorge entrance.
Compounds 1 and 3 have reported productive contacts with
site A amino acids, and 9 and 10 did not interact with known
relevant COX-1 and -2 activity residues, excluding the cofac-
tor heme (see the Supporting Information). Moreover, their
configurations have been highly solvent exposed, suggesting
a faster bound-unbound kinetic equilibrium with respect to
1 and 3.
In conclusion, we applied some simple patented methods to
mild chemical manipulation of complex natural substances. The
entire proposed process accomplishes many green chemistry tasks
and shows that it is possible develop sustainable complex proto-
col using some nonconventional techniques. All of the considered
molecules have been shown to possess significant anticyclooxy-
genase activity, and the peracetylated hydroxytyrosol 9 seems to
be a very interesting starting point to investigate the peculiar
activity of this kind of derivatives. Particularly, in the molecular
modeling, we have considered the interaction of the cofactor as a
new interesting target to explain the different behavior of our
ligands. However, in a cellular environment, it is not easily
reachable and, consequently, the heme recognition could be
COX-1
% inhibition
COX-2
% inhibition
concn (μM)
SD
SD
30
10
3
58.22
51.54
44.81
42.26
34.51
2.06
9.81
2.74
3.46
6.06
68.20
66.37
63.98
66.54
36.71
8.06
11.19
0.39
1
0.1
0.01
0.12
30 min, all animals were injected with 0.1 mL of 1% carra-
geenan solution (prepared in normal saline) in the subplantar
aponeurosis of the left hind paw and the volume of paw was
measured by using a plethysmometer at 1 and 3 h post
carrageenan treatment. All derivatives (1, 3, 8a-d, 9, 10,
and 11a-d) showed significant activity (Table 3), and two of
them, 1 and 9, displayed more potent anti-inflamatory activity
than the standard drugs ibuprofen and celecoxib, efficaciously
reducing the increase in paw volume by 62.2 and 65.9%,
respectively (Table 3).
To rationalize the recognition mechanism of compounds 1, 3,
9, and 10 within the COX isoforms, we performed a molecular
modeling study. We focused our attention on the above-reported
molecules due to their experimentally confirmed anti-inflamma-
tory activity and because they have demonstrated chemical
stability and availability in pure form. Moreover, these com-
pounds showed antioxidant properties, suggesting a concurrent
second inhibitory mechanism of action. The Protein Data Bank
(PDB) (28) crystallographic structures 1Q4G (29) and 1PXX (30)
were considered after a preliminary pretreatment (see the Sup-
porting Information), respectively, as COX-1 and COX-2 enzyme
models. The crystallographic resolution, respectively equal to 2.0
and 2.9 A, and the chemical nature of cocrystallized ligands, two
nonsteroidal anti-inflammatory drugs (NSAIDs), were the choice
criteriaof these models within the large availability of similar ones
in the PDB.
Interestingly, the 1Q4G structure describes for the first time
a glycerol molecule interacting with the heme moiety, suggest-
ing the cofactor as a new accessory target for COX inhibition.
On the basis of this observation, we studied the recognition of

11166 J. Agric. Food Chem., Vol. 57, No. 23, 2009
Procopio et al.
Table 3. Effects of the Compounds and Standard Drugs on Carrageenan-Induced Paw Edema
compound
edema vol, 1 h
edema vol, 3 h
inibition of paw edema, 1 h (%)
inibition of paw edema, 3 h (%)
control
ibuprofen
0.82 ( 0.022
0.29 ( 0.033
0.19 ( 0.013
0.31 ( 0.008
0.42 ( 0.031
0.39 ( 0.012
0.28 ( 0.014
0.48 ( 0.023
0.34 ( 0.031
0.82 ( 0.029
0.35 ( 0.009
0.25 ( 0.003
0.38 ( 0.011
0.50 ( 0.009
0.48 ( 0.029
0.33 ( 0.019
0.51 ( 0.017
0.39 ( 0.006
64.6
76.3
62.2
48.8
52.4
65.9
41.5
58.5
57.3
69.5
53.7
39.0
41.5
59.8
37.8
52.4
celecoxib
1
3
8a-d
9
10
11a-d
considered as an unspecific COX secondary binding site. Perhaps
our compounds could interact with site B, negatively modulating
the activity of both COX isoforms.
(11) Li, C.-J.; Chen, L. Organic chemistry in water. Chem. Soc. Rev. 2006,
35, 68–82.
(12) Kobayashi, S.; Sugiura, M.; Kitagawa, H.; Lam, W. W.-L. Rare-
earth metal triflates in organic synthesis. Chem. Rev. 2002, 102,
2227–2302.
ABBREVIATIONS USED
(13) Dalpozzo, R.; De Nino, A.; Maiuolo, L.; Procopio, A.; Tagarelli, A.;
Sindona, G.; Bartoli, G. Simple and efficient chemoselective mild
deprotection of acetals and ketals using cerium(III) triflate. J. Org.
Chem. 2002, 67, 9096–9098.
(14) Procopio, A.; Dalpozzo, R.; De Nino, A.; Maiuolo, L.; Russo, B.;
Sindona, G. Erbium(III) triflate as an extremely active acylation
catalyst. Adv. Synth. Catal. 2004, 346, 1465–1470.
(15) Bartoli, G.; Dalpozzo, R.; De Nino, A.; Maiuolo, L.; Nardi, M.;
Procopio, A.; Tagarelli, A. Per-O-acetylation of sugars catalyzed by
Ce(OTf)3. Green Chem. 2004, 6, 191–192.
(16) Procopio, A.; Sindona, G.; Gaspari, M.; Costa, N.; Nardi. M.
Metodo Chimico-Catalitico per la manipolazione dell’oleuropeina
per la sintesi del suo aglicone. Italian Patent MI2007A000904
(request number of international patent for both Italian patents
reported in refs 16 and 17 is PCT/IT2008/000303 titled Chemical-
catalytic method for the peracylation of oleuropein and its products of
hydrolysis).
HTyr, hydroxytyrosol or 2-(3,4-dihydroxyphenyl)ethanol;
Tyr, tyrosol or 2-(4-hydroxyphenyl)ethanol; Ole, oleuropein;
LDL, low-density lipoprotein; TXB2, thromboxane B2; LTB4,
leukotriene B4; ROS, reactive oxygen species; LPS, lipopolysac-
charide; VHSV, hemorrhagic septicemia rhabdovirus; PGE2,
prostaglandin E2; NSAD, nonsteroidal anti-inflammatory drug.
Supporting Information Available: General experimental
procedures and spectroscopic characterization of new com-
pounds, experimental log P determination, biological assay, and
molecular modeling. This material is available free of charge via
the Internet at http://pubs.acs.org.
LITERATURE CITED
(1) Trichopoulou, A.; Costacou, T.; Bamia, C.; Trichopoulos, D.
Adherence to a Mediterranean diet and survival in a Greek popula-
tion. N. Engl. J. Med. 2003, 348, 2599–2608.
(17) Mateos, M.; Goya, L.; Bravo, L. Metabolism of the olive oil phenols
hydroxytyrosol, tyrosol, and hydroxytyrosyl acetate by human
hepatoma HepG2 cells. J. Agric. Food Chem. 2005, 53, 9897–9905.
(18) Romero, C.; Medina, E.; Vargas, J.; Brenes, M.; De Castro, A. In
vitro activity of olive oil polyphenols against Helicobacter pylori. J.
Agric. Food Chem. 2007, 55, 680–686.
(19) Gikas, E.; Papadoloulos, N.; Tsarbopoulos, A. Kinetic study of the
acidic hydrolysis of oleuropein, the major bioactive metabolite of
olive oil. J. Liq. Chromatogr. Relat. Technol. 2006, 29, 497–508.
(20) Procopio, A.; Sindona, G.; Gaspari, M.; Costa, N.; Nardi, M.
Metodo Chimico-Catalitico per la peracilazione dell’oleuropeina e
dei suoi prodotti d’idrolisi. Italian Patent MI2007A000903.
(21) Namjesnik-Dejanovic, K.; Cabaniss, S. E. Reverse-phase HPLC
method for measuring polarity distributions of natural organic
matter. Environ. Sci. Technol. 2004, 38, 1108–1114.
ꢁ
(2) Mateos, R.; Espartero, J. L.; Trujillo, M.; Rıos, J. J.; Leon-
Camacho, M.; Alcudia, F.; Cert, A. Determination of phenols,
flavones, and lignans in virgin olive oils by solid-phase extraction
and high-performance liquid chromatography with diode array
ultraviolet detection. J. Agric. Food Chem. 2001, 49, 2185–2192.
(3) Visioli, F.; Poli, A.; Galli, C. Antioxidant and other biological
activities of phenols from olives and olive oil. Med. Res. Rev. 2002,
22, 65–75.
(4) Bitler, C. M.; Viale, T. M.; Damaj, B.; Crea, R. Hydrolyzed olive
vegetation water in mice has anti-inflammatory activity. J. Nutr.
2005, 135, 1475–1479.
(5) Aruoma, O. I.; Deiana, M.; Jenner, A.; Halliwell, B.; Harparkash,
K.; Banni, S.; Corongiu, F. P.; Dessı, M. A.; Aeschbach, R. Effect of
hydroxytyrosol found in extra virgin olive oil on oxidative DNA
damage and on low-density lipoprotein oxidation. J. Agric. Food
Chem. 1998, 46, 5181–5187.
(6) de la Puerta, R.; Ruiz-Gutierrez, V.; Hoult, J. R. Inhibition of
leukocyte 5-lipoxygenase by phenolics from virgin olive oil. Biochem.
Pharmacol. 1999, 57, 445–449.
(22) www.actelion.com/uninet/www/www/_main_p.nsh/content/techno-
logie sþpropertyþexplorer.
(23) www.molinspiration.com/cgi_bin/properties.
(24) Brindeau, C.; Kargman, S.; Liu, S.; Dallob, A. L.; Ehrich, E. W.;
Rodger, I. W.; Chan, C.-C. A human whole blood assay for clinical
evaluation of biochemical efficacy of cyclooxygenase inhibitors.
Inflamm. Res. 1996, 45, 68–74.
(7) Beauchamp, G.; Keast, R.; Morel, D.; Liu, J.; Pika, J.; Han, Q.; Lee,
C.; Smith, A. B.III; Breslin, P. Ibuprofen-like activity in extra-virgin
olive oil. Nature 2005, 437, 45–46.
(8) Smith, A. B.III; Han, Q.; Breslin, P. A. S.; Beauchamp, G. K.
Synthesis and assignment of absolute configuration of (-)-oleo-
canthal: a potent, naturally occurring non-steroidal anti-inflamma-
tory and anti-oxidant agent derived from extra virgin olive oils. Org.
Lett. 2005, 7, 5075–507.
(9) Amos, B. S.III; Jeffrey, B. S.; Qiang, H. Syntheses of (-)-oleo-
canthal, a natural NSAID found in extra virgin olive oil, the (-)-
deacetoxy-oleuropein aglycone, and related analogues. J. Org.
Chem. 2007, 72, 6891–6900.
(10) Anastas, P. T.; Warner, J. C. Green Chemistry, Theory and Practice;
Oxford University Press: Oxford, U.K., 1998.
(25) Mitchell, J. A.; Akarasereenont, P.; Thiemermann, C.; Flower, R. J.;
Vane, J. R. Proc. Natl. Acad. Sci. U.S.A. 1993, 90, 11693-11697.
(26) aTacconelli, S.; Capone, M. L.; Sciulli, M. G.; Ricciotti, E.;
Patrignani, P. Curr. Med. Res. Opin. 2002, 18, 503-511.bWinter,
C. A.; Risley, E. A.; Nuss, G. W. Proc. Soc. Exp. Biol. Med. 1962,
111, 544-547. cGong, D.; Geng, C.; Jiang, L.; Cao, J.; Yoshimura,
H.; Zhong, L. Phytother. Res. 2009, 23, 646-650.
(27) Berman, H. M.; Westbrook, J.; Feng, Z.; Gilliland, G.; Bhat, T. N.;
Weissig, H.; Shindyalov, I. N.; Bourne, P. E. The Protein Data Bank.
Nucleic Acids Res. 2000, 28, 235–242.
(28) Gupta, K.; Selinsky, B. S.; Kaub, C. J.; Katz, A. K.; Loll, P. J. The
2.0 A resolution crystal structure of prostaglandin H2 synthase-1:
structural insights into an unusual peroxidase. J. Mol. Biol. 2004,
335, 503–518.

Article
J. Agric. Food Chem., Vol. 57, No. 23, 2009 11167
(29) Rowlinson, S. W.; Kiefer, J. R.; Prusakiewicz, J. J.; Pawlitz, J. L.; Kozak,
K. R.; Kalgutkar, A. S.; Stallings, W. C.; Kurumbail, R. G.; Marnett, L. J.
A novel mechanism of cyclooxygenase-2 inhibition involving interactions
with Ser-530 and Tyr-385. J. Biol. Chem. 2003, 278, 45763–45769.
(30) Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.; Klicic,
J. J.; Mainz, D. T.; Repasky, M. P.; Knoll, E. H.; Shelley, M.; Perry,
J. K.; Shaw, D. E.; Francis, P.; Shenkin, P. S. Glide: a new approach
for rapid, accurate docking and scoring. 1. Method and assessment
of docking accuracy. J. Med. Chem. 2004, 47, 1739–1749.
Received for review June 26, 2009. Revised manuscript received October
14, 2009. Accepted November 2, 2009. S.A. and F.O. are grateful to
MiUR for grant support (PRIN 2007JERJPC).