Synthesis and evaluation of the platelet antiaggregant properties of phenolic antioxidants structurally related to rosmarinic acid

Article abstract of DOI:10.1016/j.bioorg.2009.12.001

Polyphenols, such as rosmarinic acid, are widely distributed natural products with relevant antioxidant activity. Oxidative stress plays an important role in the pathogenesis of a number of disorders. Here, we report on the synthesis and biological effects of the polyphenolic esters hydroxytyrosyl gallate (1), hydroxytyrosyl protocatechuate (2) and hydroxytyrosyl caffeate (3), structurally related to rosmarinic acid. The three compounds showed a greater free radical scavenging activity than their precursors and also than rosmarinic acid. Esters 1 and 3 significantly reduced thrombin-evoked platelet aggregation, which is likely mediated to the attenuation of thrombin-stimulated Ca²⁺ release and entry. The three compounds reduced the ability of platelets to accumulate Ca²⁺ in the intracellular stores, probably by enhancing the Ca²⁺ leakage rate and reduced store-operated Ca²⁺ entry in these cells. These observations suggest that the structurally-simplified analogs to rosmarinic acid, compounds 1 and 3, might be the base of therapeutic strategies to prevent thrombotic complications associated to platelet hyperaggregability due to oxidative stress.

Full text of DOI:10.1016/j.bioorg.2009.12.001

Bioorganic Chemistry 38 (2010) 108–114
Contents lists available at ScienceDirect
Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg
Synthesis and evaluation of the platelet antiaggregant properties of phenolic
antioxidants structurally related to rosmarinic acid
b
Laura Chapado a, Pablo J. Linares-Palomino a, Sofía Salido a, Joaquín Altarejos a, , Juan A. Rosado ,
*
Ginés M. Salido b
a Department of Inorganic and Organic Chemistry, Faculty of Experimental Sciences, University of Jaén, 23071 Jaén, Spain
b Department of Physiology (Cell Physiology Research Group), University of Extremadura, 10071 Cáceres, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 16 November 2009
Available online 5 December 2009
Polyphenols, such as rosmarinic acid, are widely distributed natural products with relevant antioxidant
activity. Oxidative stress plays an important role in the pathogenesis of a number of disorders. Here,
we report on the synthesis and biological effects of the polyphenolic esters hydroxytyrosyl gallate (1),
hydroxytyrosyl protocatechuate (2) and hydroxytyrosyl caffeate (3), structurally related to rosmarinic
acid. The three compounds showed a greater free radical scavenging activity than their precursors and
also than rosmarinic acid. Esters 1 and 3 significantly reduced thrombin-evoked platelet aggregation,
which is likely mediated to the attenuation of thrombin-stimulated Ca2+ release and entry. The three
compounds reduced the ability of platelets to accumulate Ca2+ in the intracellular stores, probably by
enhancing the Ca2+ leakage rate and reduced store-operated Ca2+ entry in these cells. These observations
suggest that the structurally-simplified analogs to rosmarinic acid, compounds 1 and 3, might be the base
of therapeutic strategies to prevent thrombotic complications associated to platelet hyperaggregability
due to oxidative stress.
Keywords:
Antioxidant activity
Platelet aggregation
Store-operated calcium entry
Hydroxytyrosyl gallate
Hydroxytyrosyl protocatechuate
Hydroxytyrosyl caffeate
Rosmarinic acid analogs
Ó 2009 Elsevier Inc. All rights reserved.
1. Introduction
with phenolic compounds showing antioxidant properties reverses
the enhanced oxidant production, altered Ca2+ mobilization and
It is known that intracellular free Ca2+ concentration ([Ca2+]i)
controls a number of platelet functions, including aggregation,
and platelet agonists, such as thrombin, increase [Ca2+]i, which
consists of two components: release of Ca2+ from intracellular
stores and Ca2+ entry through plasma membrane channels [1]. It
has been shown that agonist-induced Ca2+ mobilization is altered
in platelets from patients suffering from oxidative stress, including
diseases like type 2 diabetes mellitus (DM), where both Ca2+ re-
lease and entry are enhanced [2] while Ca2+ extrusion by plasma
membrane Ca2+-ATPase (PMCA) is reduced [3]. Abnormal Ca2+ re-
lease, entry and extrusion in platelets from diabetic donors are re-
versed by treatment with antioxidants, which further support the
role of oxidative stress in this phenomenon [4,5]. Consistent to this,
we have recently demonstrated that in vitro platelet treatment
hyperaggregability described in platelets from diabetic patients,
supporting that natural antioxidants may be used to develop ther-
apeutic strategies to reduce cardiovascular complications in type 2
DM [6–8].
Rosmarinic acid is a naturally occurring polyphenol widely dis-
tributed in Labiatae plants, among others. It is one of the main com-
ponents responsible for the high antioxidant activity of commercial
rosemary extracts (the only spice commercially available for use as
an antioxidant in food processing in Europe and United States) [9],
and it can also be found in sage, basil, peppermint, perilla or laven-
der [10]. The biological activities and potential applications of ros-
marinic acid have been studied widely. It exhibits antioxidant [11],
anti-inflammatory [12] or anxiolytic-like [13] activities. It also has
hepatoprotective [14] and neuroprotective [15] effects, inhibits the
enzymatic thrombin activity [16], and induces apoptosis of acti-
vated T cells from rheumatoids arthritis patients [17], among other
activities. Some rosmarinic acid derivatives have been claimed for
the prevention and improvement of insulin resistance in DM [18]
and as new anti-HIV-1 agents [19]. Although the chemical synthesis
of rosmarinic acid has already been described [20], the required
amounts of this compound are still isolated from natural sources
or in vitro [21] and biotechnologically [22] produced, since the
Abbreviations: ARP, antiradical power; DIAD, diisopropyl azodicarboxylate; DM,
diabetes mellitus; DPPH, 1,1-diphenyl-2-picrylhydrazyl; PMCA, plasma membrane
Ca2+-ATPase; SERCA, sarco(endo)plasmic reticulum Ca2+-ATPase; SOCE, store-oper-
ated Ca2+ entry; TG, thapsigargin; TPP, triphenylphosphine.
* Corresponding author. Fax: +34 953 212743.
E-mail addresses: lchapado@ujaen.es (L. Chapado), plinares@ujaen.es (P.J.
Linares-Palomino), ssalido@ujaen.es (S. Salido), jaltare@ujaen.es (J. Altarejos),
jarosado@unex.es (J.A. Rosado), gsalido@unex.es (G.M. Salido).
0045-2068/$ - see front matter Ó 2009 Elsevier Inc. All rights reserved.
doi:10.1016/j.bioorg.2009.12.001

L. Chapado et al. / Bioorganic Chemistry 38 (2010) 108–114
109
stereogenic carbon at the 3,4-dihydroxyphenyllactic acid moiety
adds disadvantages to a commercial synthetic approach. For this
reason, the investigation of low-cost ways to produce rosmarinic
acid or structurally-simplified analogs with interesting biological
activities may be an attractive and profitable aim.
the solvent systems indicated. Solvent systems are reported as v/v
percent ratios. All reactions were monitored by high-performance
liquid chromatography (HPLC) (see below) and thin-layer chroma-
tography (TLC) using MERCK silica gel 60 F254 precoated aluminum
sheets (0.25 mm). Yields refer to chromatographically and spectro-
scopically pure compounds except when otherwise indicated. HPLC
analyzes were performed by analytical RP-HPLC (Waters Spheri-
sorbÒ C18 S5 ODS2, 250 mm ꢀ 4.6 mm i.d., 5
lm) and HPLC purifi-
cations by semipreparative RP-HPLC (Waters SpherisorbÒ C18 S5
ODS2, 250 mm ꢀ 22 mm i.d., 5
lm) on a Waters 600E instrument
equipped with a photodiode array detector (DAD), scan range:
190–800 nm, and operating at 30 °C. The HPLC analyzes were per-
formed by a step gradient with MeOH/AcOH 99.8:0.2 (v/v) (solvent
A) and H2O/AcOH 99.8:0.2 (v/v) (solvent B) at a flow rate of 0.7 mL/
min: linear gradient from 10% to 25% A in 15 min; linear gradient
from 25% to 50% A in 5 min; isocratic conditions of 50% A for
5 min, and other 5 min to return to the initial conditions. The total
run time excluding equilibration was 30 min. The semipreparative
HPLC purifications were performed by a step gradient with MeOH/
AcOH 99.8:0.2 (v/v) (solvent A) and H2O/AcOH 99.8:0.2 (v/v) (sol-
vent B) at a flow rate of 5 mL/min; linear gradient from 20% to
50% A in 5 min; isocratic conditions of 50% A for 5 min, and other
5 min to return to the initial conditions. The total run time exclud-
ing equilibration was 15 min. UV spectra were recorded in EtOH on
a Varian Cary 4000 UV/vis spectrophotometer. Melting points were
determined on a Barnstead Electrothermal apparatus and are
uncorrected. 1H NMR spectra were recorded on a Bruker DPX 400
spectrometer (400 MHz). Chemical shift values are reported in
parts per million (ppm, d scale) and coupling constants (J) are in
hertz (Hz). All described coupling constants are referred to a
three-bond coupling distance and further (nJ, n P 3). The splitting
patterns are abbreviated as follows: s (singlet), d (doublet), t (trip-
let), dd (double doublet). 13C NMR spectra were recorded on the
same instrument (100 MHz). Chemical shifts are also reported in
ppm and carbon substitution degrees were established by DEPT
multipulse sequence. Infrared (IR) spectra were recorder on a FT-
IR Perkin–Elmer 1760X spectrometer. High resolution mass spectra
(HRMS) were obtained on a LC–MS equipment composed of an Alli-
ance 2795 Waters HPLC chromatograph fitted to a Micromass LCT
Premier mass spectrometer, with an electrospray ionization (ESI)
source and a time-of-flight (TOF) analyzer.
In this line, here we report on the synthesis of three antioxidants
structurally related to rosmarinic acid, such as esters hydroxytyro-
syl gallate (1) [23], hydroxytyrosyl protocatechuate (2) and
hydroxytyrosyl caffeate (3) [24], from abundant polyphenolic start-
ing materials like gallic acid (4) [25], protocatechuic acid (5) [26],
caffeic acid (6) [26] and hydroxytyrosol (7) [27] (Scheme 1), and
their ability to attenuate thrombin-stimulated platelet aggregation.
In addition, we have explored the possible mechanisms involved in
the attenuation of platelet function such as their role in Ca2+ mobi-
lization. These phenolic compounds might be the base of therapeu-
tic strategies for disorders involving oxidative stress.
2. Material and methods
2.1. Materials
Hydroxytyrosol (7) was prepared by reducing 3,4-dihydroxy-
phenylacetic acid according to Capasso method [28]. Gallic acid
(4), protocatechuic acid (5), caffeic acid (6) and diisopropyl azodi-
carboxylate (DIAD) [29] were purchased from Fluka (Madrid,
Spain). Rosmarinic acid, triphenylphosphine (TPP), 1,1-diphenyl-
2-picrylhydrazyl (DPPH), 3,4-dihydroxyphenylacetic acid, apyrase
(grade VII), aspirin, thapsigargin (TG), thrombin, ionomycin (Iono)
and bovine serum albumin (BSA) were purchased from Sigma–Al-
drich (Madrid, Spain). Fura-2 acetoxymethyl ester (fura-2/AM) and
calcein-AM were from Molecular Probes (Leiden, The Netherlands).
All reagents were used without further purification. The tetrahy-
drofuran (THF) used in the syntheses of 1–3 was purchased from
Panreac (Barcelona, Spain) and distilled under argon immediately
before use from sodium/benzophenone.
2.3. Mitsunobu esterification of polyphenolic acids (4–6) with
hydroxytyrosol (7). General protocol
2.2. Chromatography and structure determination techniques
Chromatographic separations were performed on SCHARLAU sil-
ica gel (230–400 mesh) and MERCK silica gel (70–230 mesh) using
To a cooled (0 °C) solution of recently prepared 7 (see above)
[28] (1 mmol) and a given polyphenolic acid (4–6) (1.5 mmol,
Scheme 1. Synthesis of hydroxytyrosyl gallate (1), hydroxytyrosyl protocatechuate (2) and hydroxytyrosyl caffeate (3).

110
L. Chapado et al. / Bioorganic Chemistry 38 (2010) 108–114
1.5 M equiv) in dry THF (1 mL/75 mg of 7) were added TPP
(2.53 mmol, 1.5 M equiv) and DIAD (2.53 mmol, 1.5 M equiv).
The reaction mixture was stirred at room temperature for 2 days.
Then, the reaction was worked up by removal of the solvent, re-
dissolved on EtOAc and extracted with 1 N NaHCO3 (ca. 50 mL/
250 mg of 7). The organic phase was washed with brine, dried
(Na2SO4) and evaporated. The resultant ester (crude reaction)
was subjected to two successive silica gel column chromatogra-
phies, eluting with CHCl3–MeOH mixtures, and eventually purified
by semipreparative RP-HPLC.
(1H, d, J = 15.9, H-2a), 6.56 (1H, dd, J = 8.1 and 2.1, H-600), 6.69
(1H, d, J = 8.1, H-500), 6.70 (1H, d, J = 2.1, H-200), 6.76 (1H, d, J = 8.1,
H-50), 6.92 (1H, dd, J = 8.1 and 2.1, H-60), 7.02 (1H, d, J = 2.1, H-
20), 7.51 (1H, d, J = 15.9, H-3a); 13C NMR (CD3OD) d signals for
the hydroxytyrosyl moiety: 35.58 (C-2), 66.49 (C-1), 117.05 (C-
200), 116.49 (C-500), 121.24 (C-600), 130.85 (C-100), 144.91 (C-400),
146.25 (C-300), d signals for the caffeate moiety: 115.14 (C-20),
115.15 (C-2a), 116.40 (C-50), 122.78 (C-60), 127.74 (C-10), 146.78
(C-3a), 146.87 (C-30), 149.54 (C-40), 169.28 (C-1a). NMR data of this
compound (recorded in CD3OD) were in agreement with those pre-
viously reported (recorded in DMSO-d6) [24].
2.3.1. 2-(3,4-Dihydroxyphenyl)ethyl 3,4,5-trihydroxybenzoate (1)
A crude reaction (1.60 g) containing compound 1 was obtained
following the general protocol (hydroxytyrosol (7): 260 mg; gallic
2.4. Free radical scavenging activity
acid (4): 430 mg; THF: 3.5 mL; TPP: 663 mg; DIAD: 527
l
L), which
The antioxidant activity of hydroxytyrosol (7), the phenolic car-
boxylic acids 4–6, rosmarinic acid (reference compound) and the
esters 1–3 was determined by spectrophotometric measurements
according to the ability of the tested samples to scavenge the free
radical DPPH. This activity was determined using the method em-
ployed by Brand-Williams et al. [30] and modified as described be-
low. Methanolic solutions (1.2 mL) of a given compound, at
was first chromatographed on silica gel (70–230 mesh) using
CHCl3–MeOH mixtures. It was obtained crude 1 (410 mg) that
was submitted to a second column chromatography on silica gel
(230–400 mesh) using an 85:15 CHCl3–MeOH mixture. Thus, com-
pound 1 (211 mg, 41%) was obtained as a yellowish powder, and an
aliquot amount was re-purified by semipreparative HPLC as de-
scribed above (NMR purity: 95%). Mp 189–190 °C (lit. [23] 189–
191 °C). Spectral data of this compound were identical to those
previously reported [23].
different concentrations between 0.33 and 4.60 lg/mL were mixed
in a 1 cm path length disposable plastic cuvette with methanolic
solutions (2.4 mL) of DPPHÅ (ꢂ7 ꢀ 10–5 M) with an absorbance at
515 nm of 0.80 0.03 AU. The exact DPPHÅ concentration was cal-
culated from a calibration curve. Triplicate samples were shaken
and allowed to stand for 15 min in the dark at room temperature,
and the decrease of absorbance at 515 nm was measured using a
Perkin–Elmer UV/vis spectrophotometer Lambda 19 (Perkin–Elmer
Instruments, Norwalk, CT).
2.3.2. 2-(3,4-Dihydroxyphenyl)ethyl 3,4-dihydroxybenzoate (2)
A crude reaction (3.50 g) containing compound 2 was obtained
following the general protocol (hydroxytyrosol (7): 845 mg; proto-
catechuic acid (5): 1.27 g; THF: 12 mL; TPP: 2.13 g; DIAD: 1.59 mL),
which was first chromatographed on silica gel (70–230 mesh)
using CHCl3–MeOH mixtures. It was obtained crude 2 (1.63 g) that
was submitted to a second column chromatography on silica gel
(230–400 mesh) using a 95:5 CHCl3–MeOH mixture. Thus, com-
pound 2 (500 mg, 31%) was obtained as a brown-pale powder,
and an aliquot amount was re-purified by semipreparative HPLC
as described above (NMR purity: ca. 96%). Mp 153–156 °C; UV
2.5. Platelet preparation
Blood was obtained from healthy volunteers, according to the
rules of the Declaration of Helsinki, and mixed with one-sixth vol-
ume of acid/citrate dextrose anticoagulant containing (in mM): 85
(EtOH) kmax (log
e
): 264 (4.33), 290 (4.17) nm; IR (KBr) 3436 and
sodium citrate, 78 citric acid and 111
ma was then prepared by centrifugation for 5 min at 700g and
aspirin (100 M) and apyrase (40 g/mL) added. Cells were then
collected by centrifugation at 350g for 20 min and re-suspended
in HEPES-buffered saline (HBS) containing (in mM): 145 NaCl, 10
HEPES, 10 D-glucose, 5 KCl, 1 MgSO4, pH 7.45, and supplemented
with 0.1% w/v bovine serum albumin and 40 lg/mL apyrase.
D-glucose. Platelet-rich plas-
3308 (phenols), 1677 and 1613 (ArCO2R) cmꢁ1; HRMS (ESI)
(C15H14O6) m/z calculated MꢁH+ 289.0712, measured MꢁH+
289.0710; 1H NMR (CD3OD) d 2.86 (2H, t, J = 7.0, H-2), 4.35 (2H,
t, J = 7.0, H-1), 6.58 (1H, dd, J = 1.9 and 8.1, H-600), 6.69 (1H, d,
J = 8.1, H-500), 6.71 (1H, d, J = 1.9, H-200), 6.78 (1H, dd, J = 1.5 and
7.5, H-60), 7.38 (1H, d, J = 1.5, H-20), 7.39 (1H, d, J = 7.5, H-50); 13C
NMR (CD3OD) d signals for the hydroxytyrosyl moiety: 35.61 (C-
2), 66.75 (C-1), 116.40 (C-200), 117.04 (C-500), 121.29 (C-600),
130.93 (C-100), 144.87 (C-400), 146.10 (C-300), d signals for the 3,4-
dihydroxybenzoate moiety: 115.82 (C-60), 117.40 (C-20), 122.70
(C-10), 123.65 (C-50), 146.10 (C-30), 151.65 (C-40), 168.35 (C-1a).
l
l
2.6. Cell viability
Cell viability was assessed using calcein and trypan blue as de-
scribed previously [31]. For calcein loading, cells were incubated
for 30 min with 5 lM calcein-AM at 37 °C, centrifuged and the pel-
2.3.3. (E)-2-(3,4-Dihydroxyphenyl)ethyl 3-(3,4-
dihydroxyphenyl)acrilate (3)
A crude reaction (3.39 g) containing compound 3 was obtained
following the general protocol (hydroxytyrosol (7): 490 mg; caffeic
acid (6): 905 mg; THF: 9 mL; TPP: 1.25 g; DIAD: 925 lL), which
was first chromatographed on silica gel (70–230 mesh) using
CHCl3–MeOH mixtures. It was obtained crude 3 (668 mg) that
was submitted to a second column chromatography on silica gel
(230–400 mesh) using a 95:5 CHCl3–MeOH mixture. Thus, com-
pound 3 (369 mg, 47%) was obtained as a yellow-dark powder,
and an aliquot amount was re-purified by semipreparative HPLC
as described above (NMR purity: ca. 95%). Mp 118–120 °C; UV
let was re-suspended in fresh HBS. Cells were treated with the dif-
ferent inhibitors, centrifuged and re-suspended in HBS.
Fluorescence was recorded from 2 mL aliquots using a spectropho-
tometer (Varian Ltd., Madrid, Spain). Samples were excited at
494 nm and the resulting fluorescence was measured at 535 nm.
The results obtained with calcein were confirmed using the trypan
blue exclusion technique.
2.7. Platelet aggregation
Aggregation of washed platelets was monitored in a Chronolog
(Havertown, PA, USA) aggregometer at 37 °C under stirring at
1200 rpm. The percentage of aggregation or amplitude is estimated
as the percentage of the difference in light transmission between
the platelet suspension in HBS and HBS alone, and indicates the
percentage of platelets that aggregate in response to an agonist.
Resting platelets in suspension are arbitrarily considered by the
(EtOH) kmax (log e): 296 (3.96), 329 (4.49) nm; IR (KBr) 3485 and
3334 (phenols), 1672 and 1630 (ArCO2R), 976 (asym-trisubstituted
benzene and HC = CH–CO) cmꢁ1; HRMS (ESI) (C17H16O6) m/z calcu-
lated MꢁH+ 315.0869, measured MꢁH+ 315.0868; 1H NMR
(CD3OD) d 2.82 (2H, t, J = 7.1, H-2), 4.28 (2H, t, J = 7.1, H-1), 6.22

L. Chapado et al. / Bioorganic Chemistry 38 (2010) 108–114
111
Table 1
Efficient concentration (EC50
synthetic esters 1–3, their precursors 4–7 and rosmarinic acid against radical DPPH.
aggregometer as 0% aggregation and HBS is considered to be 100%
aggregation. The rate, or slope, of the aggregation is the % change of
aggregation per minute.
a
)
and antiradical power (ARP) values of the semi-
Compound
EC50
ARP
3.39
5.85
6.06
8.13
9.62
11.36
12.50
18.18
Hydroxytyrosol (7)
Caffeic acid (6)
0.30 0.02
0.17 0.01
0.17 0.02
0.12 0.01
0.10 0.01
0.09 0.01
0.08 0.01
0.06 0.01
2.8. Measurement of intracellular free calcium concentration ([Ca2+]i)
Protocatechuic acid (5)
Rosmarinic acid (reference compound)
Gallic acid (4)
Hydroxytyrosyl caffeate (3)
Hydroxytyrosyl gallate (1)
Hydroxytyrosyl protocatechuate (2)
Human platelets were loaded with fura-2 by incubation with
2 lM fura-2/AM for 45 min at 37 °C. Fluorescence was recorded
from 2 mL aliquots of magnetically stirred cellular suspension
(2 ꢀ 108 cells/mL) at 37 °C using a Cary Eclipse spectrophotometer
(Varian Ltd., Madrid, Spain) with excitation wavelengths of 340
and 380 nm, and emission at 505 nm, as previously described
[32]. Changes in [Ca2+]i were monitored using the fura-2 340/380
fluorescence ratio and calibrated as described previously.
a
EC50 expressed as moles of compound/moles of DPPHÅ SD. ARP = 1/EC50
.
worth noting that hydroxytyrosol (7), a powerful antioxidant, ap-
pears with the lowest ARP value.
2.9. Statistical analysis
3.3. Effect of the semi-synthetic antioxidants on platelet aggregation
Analysis of statistical significance was performed using one-
way analysis of variance (ANOVA) combined with the Dunnett or
Tukey test. For comparison between two groups Student’s t test
was used. p < 0.05 was considered to be significant for a difference.
Platelet treatment for 20 min at 37 °C with increasing concen-
trations (1–100
lM) of esters 1 and 3 significantly reduced the per-
centage and rate of thrombin-induced aggregation in
a
concentration-dependent manner (Table 3; p < 0.05; n = 6–10).
The effect of ester 3 cannot be attributed to a reduction in cell via-
bility, as reported in Table 2, and despite the fact that ester 1 re-
3. Results and discussion
duced cell viability by 8.5%, at the concentration of 100 lM, it
3.1. Chemical synthesis
reduces the ability of thrombin to induce aggregation by 27%,
which suggest a direct effect of 1 on platelet aggregation, rather
by inhibition due to a reduction in cell viability. In contrast, neither
ester 2 nor any of the precursors of the semi-synthetic compounds
had any significant effect on thrombin-evoked platelet aggregation
(Table 3). It is noteworthy that some of the precursors, including
caffeic acid (6), protocatechuic acid (5) and, although slightly,
hydroxytyrosol (7) reduced cell viability, while the semi-synthetic
compounds, except hydroxytyrosyl gallate (1), had a negligible ef-
fect on cell viability (Table 2).
The esterification of phenolic alcohols and phenolic acids via
acyl nucleophilic substitution under nucleophilic catalysis requires
protection of phenolic hydroxyls, since acids activated in situ (isou-
reas and mixed phosphoric anhydrides) or ex situ (chlorides, anhy-
drides, and mixed anhydrides) show poor discrimination between
hydroxyls bound to aliphatic and aromatic carbons [33]. Other
methods, such as Fischer esterification using strong protic acids
or the Lewis acid-catalyzed acylations have been described with
acceptable level of chemoselectivity, but reaction conditions in
many cases make these strategies of limited applicability [34]. In
order to find a suitable chemoselectivity in the esterification reac-
tion [35] without protecting phenolic hydroxyls, we chose the
Mitsunobu couple, which has the advantage of consider to phenolic
hydroxyls present on either substrate as inert spectators [33,36].
Thus, compounds 1–3 were prepared according to the Mitsunobu
protocol [33] with moderate yield. Compound 1 had previously
been prepared in five steps and 5% overall yield [23]. However, un-
der Mitsunobu conditions, it could be prepared in one step only
(>40% yield) and without resorting to any protecting group. The re-
moval of Mitsunobu by-products and the purification of polar com-
pounds 1–3 were achieved by two successive carefully-performed
silica gel column chromatographies. This procedure was preferred
to the previously described chromatography on Sephadex [33], on
the basis of a lower laboratory cost and similar efficiency.
3.4. Effect of the semi-synthetic antioxidants on intracellular Ca2+
mobilization
Cytosolic Ca2+ concentration, especially sustained increases due
to Ca2+ entry, has been reported to be important for agonist-in-
duced platelet aggregation [37]. Hence, we have further explored
the effect of platelet incubation with the semi-synthetic com-
pounds on Ca2+ mobilization in these cells. In a Ca2+-free medium,
treatment of platelets with the physiological agonist thrombin
(0.1 U/mL) resulted in a transient increase in [Ca2+]i due to Ca2+ re-
lease from the intracellular stores. The subsequent addition of Ca2+
(300 lM) to the external medium induced a sustained elevation in
Table 2
Cell viability upon treatmenta for 20 min at 37 °C with the semi-synthetic esters 1–3
and their precursors 4–7.
3.2. Antiradical measurements of the semi-synthetic antioxidants
Compound
Cell viability (% control)
85.7 2.2*
90.6 3.4*
92.1 5.1
91.5 0.5
95.4 1.2
94.3 2.6
95.1 0.7*
85.7 4.1*
The radical scavenging activity of the tested samples was ex-
pressed as the molar amount of antioxidant necessary to decrease
the initial DPPHÅ concentration by 50% (efficient concentra-
tion = EC50 (moles of antioxidant/moles of DPPHÅ)) and as the anti-
radical power (ARP = 1/EC50), for reasons of clarity. In Table 1,
hydroxytyrosol (7), the referred phenolic carboxylic acids 4–6, ros-
marinic acid (reference compound) and the esters 1–3 are pre-
sented in increasing order of their ARP values. The results of this
comparative study showed that esters 1–3 are more potent antiox-
idants than the starting materials used in their syntheses (4–6) and
even the natural antioxidant used as standard, rosmarinic acid. It is
Caffeic acid (6)
Protocatechuic acid (5)
Gallic acid (4)
Hydroxytyrosyl gallate (1)
Hydroxytyrosyl protocatechuate (2)
Hydroxytyrosyl caffeate (3)
Hydroxytyrosol (7)
Rosmarinic acid (reference compound)
Human platelets (108 cells/mL) were incubated with 100
lM of 1–3 and 4–7, as
well as the well known antioxidant, rosmarinic acid, for 20 min at 37 °C. Cell via-
bility was determined using calcein as described in the Experimental section.
p < 0.05 (ANOVA) compared to control.
a
*

112
L. Chapado et al. / Bioorganic Chemistry 38 (2010) 108–114
Table 3
Effect of treatmenta for 20 min at 37 °C with the semi-synthetic esters 1–3 and their precursors 4–7 in thrombin-induced platelet aggregation.
Compound
Control
Lag-time (min)
0.012 0.001
% Rate
% Aggregation
85.9 0.8
85.8 4.1
Caffeic acid (6)
1
10
100
l
l
M
M
l
0.011 0.002
0.012 0.001
0.010 0.001
85.6 4.9
87.8 9.7
78.5 2.9
84.4 0.9
85.2 1.2
76.7 7.5
M
Protocatechuic acid (5)
1
10
100
l
l
M
M
l
0.010 0.002
0.010 0.002
0.011 0.001
86.0 5.3
81.0 5.9
95.4 14.4
86.4 1.3
83.1 2.7
84.8 2.2
M
Gallic acid (4)
1
10
100
l
l
M
M
l
0.012 0.002
0.018 0.002
0.012 0.001
90.6 3.4
91.9 7.2
93.7 5.6
84.8 0.9
81.5 5.5
88.5 1.1
M
Hydroxytyrosyl gallate (1)
1
10
100
l
l
M
M
l
0.011 0.001
0.013 0.003
0.013 0.005
83.8 9.5
80.5 4.9
64.1 10.9*
76.8 7.5
78.2 8.5
63.0 13.6*
M
Hydroxytyrosyl protocatechuate (2)
1
l
M
0.008 0.001*
0.012 0.001
0.014 0.001
80.3 2.1
94.5 8.4
94.0 5.0
82.2 0.9
83.2 3.4
84.1 3.0
10
lM
100
lM
Hydroxytyrosyl caffeate (3)
1
l
M
0.007 0.001*
0.010 0.001
0.013 0.001
75.6 1.3*
76.0 1.3*
66.1 8.5*
82.2 3.5
81.6 2.4*
66.5 7.5*
10
lM
100
lM
Hydroxytyrosol (7)
1
10
100
l
l
M
M
l
0.012 0.002
0.014 0.002
0.012 0.001
84.0 3.1
88.8 7.5
91.8 5.1
83.0 2.4
85.5 2.8
85.5 1.1
M
a
Human platelets (108 cells/mL) were incubated with different concentrations (1–100
lM) of 1–3 and 4–7 for 20 min at 37 °C. Cells were then stimulated with 0.1 U/mL
thrombin in the presence of 1 mM extracellular Ca2+. Aggregation of human platelets was induced at a shear rate of 1200 rpm at 37 °C as described in the Experimental
section. Values given are presented as mean SEM of 6–10 separate determinations.
*
p < 0.05 (ANOVA) compared to control.
Fig. 1. Effect of treatment with the 1–3 on thrombin-induced Ca2+ mobilization in human platelets. Fura-2-loaded human platelets were incubated in the presence of the 1–3
(100
(final concentration 300
calibrated in terms of [Ca2+]i. Traces shown are representative of six separate experiments.
l
M), as indicated, or the vehicle (control) for 20 min at 37 °C and then stimulated with thrombin (0.1 U/mL) in a Ca2+ free medium (100
lM EGTA was added). CaCl2
l
M) was added to the medium 3 min later to initiate Ca2+ entry. Elevations in [Ca2+]i were monitored using the 340/380 nm ratio and traces were

L. Chapado et al. / Bioorganic Chemistry 38 (2010) 108–114
113
Table 4
aggregation by this agonist. All the semi-synthetic compounds
were more effective than their precursor hydroxytyrosol (7) inhib-
iting thrombin-stimulated Ca2+ mobilization.
Similarly, we have investigated the effect of compounds 1–3 on
store-operated Ca2+ entry (SOCE), an event activated by depletion of
the intracellular Ca2+ stores that has been reported as a major
mechanism for Ca2+ entry in human platelets [38,39]. In the ab-
sence of extracellular Ca2+, human platelets were treated with
Effect of treatmenta for 20 min at 37 °C with the semi-synthetic esters 1–3 and
hydroxytyrosol (7) on thrombin-evoked calcium mobilization.
Compound
Calcium release
(nM.s)
Calcium entry
(nM.s)
Thrombin
Control
2375 581
1331 400*
1161 182*
10069 742
8413 582*
9953 1005
Hydroxytyrosyl gallate (1)
Hydroxytyrosyl
protocatechuate (2)
Hydroxytyrosyl caffeate (3)
Hydroxytyrosol (7)
1 lM thapsigargin (TG), an inhibitor of the sarco(endo)plasmic
reticulum Ca2+-ATPase (SERCA) [40], plus a low concentration of
ionomycin (50 nM; required for extensive depletion of both inter-
nal Ca2+ stores in platelets) [41,42]. Platelet treatment with
TG + ionomycin in stirred cuvettes at 37 °C evoked a transient ele-
vation in [Ca2+]i due to release of Ca2+ from internal stores. Subse-
1025 119*
2141 390
8237 846*
10990 768
Thapsigargin + ionomycin
Control
Hydroxytyrosyl gallate (1)
Hydroxytyrosyl
protocatechuate (2)
Hydroxytyrosyl caffeate (3)
Hydroxytyrosol (7)
14253 956
12017 718*
10098 1630*
59582 3439
50302 3683*
47689 6840*
quent addition of Ca2+ (300
l
M) to the external medium induced
a sustained increase in [Ca2+]i indicative of SOCE (Fig. 2). Treatment
of human platelets for 20 min at 37 °C with 100 M of compounds
7081 1076*
12435 1280
31912 6398*
53717 5270
l
1–3 significantly attenuated Ca2+ release induced by TG + ionomy-
cin (Fig. 2 and Table 4), which indicates that these compounds re-
duces the ability of platelets to accumulate Ca2+ into the stores.
We have further investigated whether the reduction in TG + iono-
mycin-induced Ca2+ release by these compounds might be attrib-
uted to a reduction in the Ca2+ leakage rate from the intracellular
stores. As shown in non-stimulated cells (Table 5), the three com-
pounds enhanced significantly the rate of Ca2+ leakage from the
intracellular compartments, which indicates that the inhibition of
TG + ionomycin-evoked Ca2+ release is unlikely due to a reduction
in the permeability of the Ca2+ stores to Ca2+ and is more likely
attributed to a reduction in Ca2+ accumulation due to an increase
a
Fura-2-loaded human platelets (108 cells/mL) were incubated with 100
1–3 for 20 min at 37 °C. At the time of experiment 100 M EGTA was added. Cells
were then stimulated either with 0.1 U/mL thrombin or with thapsigargin (1 M) in
M)
lM of
l
l
combination with 50 nM ionomycin, as indicated, and 3 min later CaCl2 (300
l
was added to initiate Ca2+ entry. Changes in cytosolic Ca2+ concentration were
determined as described in the Experimental section.
*
p < 0.05 (ANOVA) compared to control.
[Ca2+]i indicative of Ca2+ entry (Fig. 1A). Treatment of human plate-
lets for 20 min at 37 °C with 100 lM of esters 1–3 significantly re-
duced thrombin-evoked Ca2+ release from the intracellular stores
and, with the exception of hydroxytyrosyl protocatechuate (2),
they also significantly reduced thrombin-induced Ca2+ entry
(Fig. 1 and Table 4; p < 0.05; n = 6). The fact that ester 2 did not sig-
nificantly reduced thrombin-stimulated Ca2+ entry might provide
an explanation to the lack of effect of this compound on platelet
in the permeability of the membrane of the Ca2+ stores for Ca2+
.
These findings also indicate that these compounds are not Ca2+ che-
lators. In addition, we have investigated the effect of these semi-
synthetic compounds on SOCE in human platelets. Treatment with
Fig. 2. Effect of treatment with the 1–3 on TG + ionomycin-evoked Ca2+ mobilization in human platelets. Fura-2-loaded human platelets were incubated in the presence of
the 1–3 (100 M), as indicated, or the vehicle (control) for 20 min at 37 °C and then stimulated with 1
M TG in combination with 50 nM ionomycin in a Ca2+ free medium
(100 M EGTA was added). CaCl2 (final concentration 300
M) was added to the medium 3 min later to initiate Ca2+ entry. Elevations in [Ca2+]i were monitored using the 340/
380 nm ratio and traces were calibrated in terms of [Ca2+]i. Traces shown are representative of six separate experiments.
l
l
l
l

114
L. Chapado et al. / Bioorganic Chemistry 38 (2010) 108–114
Table 5
[4] P.C. Redondo, I. Jardín, J.M. Hernández-Cruz, J.A. Pariente, G.M. Salido, J.A.
a
Effect of treatment for 20 min at 37 °C with the semi-synthetic esters 1–3 on Ca2+
Rosado, Biochem. Biophys. Res. Commun. 333 (2005) 794–802.
[5] I. Jardín, P.C. Redondo, G.M. Salido, J.A. Pariente, J.A. Rosado, Platelets 17 (2006)
283–288.
leakage from the intracellular stores.
Compound
Calcium leakage (nM s)
[6] A. Bouaziz, S. Salido, P.J. Linares-Palomino, A. Sánchez, J. Altarejos, A. Bartegi,
G.M. Salido, J.A. Rosado, Arch. Biochem. Biophys. 457 (2007) 235–242.
[7] N. Ben Amor, A. Bouaziz, C. Romera-Castillo, S. Salido, P.J. Linares-Palomino, A.
Bartegi, G.M. Salido, J.A. Rosado, J. Med. Chem. 50 (2007) 3937–3944.
[8] H. Zbidi, S. Salido, J. Altarejos, M. Pérez-Bonilla, A. Bartegi, J.A. Rosado, G.M.
Salido, Blood Cell. Mol. Dis. 42 (2009) 279–285.
Control
63 19
Hydroxytyrosyl gallate (1)
Hydroxytyrosyl protocatechuate (2)
Hydroxytyrosyl caffeate (3)
658 259*
588 32*
177 76*
[9] N.V. Yanishlieva-Maslarova, I.M. Heinonen, Sources of natural antioxidants:
Fura-2-loaded human platelets (108 cells/mL) were incubated with 100
lM of
a
´
vegetables, fruits, herbs, spices and teas, in: J. Pokorny, N. Yanishlieva, M.
1–3 for 20 min at 37 °C. At the time of experiment 100
lM EGTA was added and
Gordon (Eds.), Antioxidants in Food. Practical Applications, CRC Press, Boca
Raton, 2001, pp. 210–263.
cells were not stimulated for at least 3 further minutes. Changes in cytosolic Ca2+
concentration were determined as described in the Experimental section.
[10] M. Petersen, M.S. Simmonds, Phytochemistry 62 (2003) 121–125.
[11] N. Erkan, G. Ayranci, E. Ayranci, Food Chem. 110 (2008) 76–82.
[12] A. Kulhmann, C. Roehl, Pharm. Biol. 44 (2006) 401–410.
[13] P. Pereira, D. Tysca, P. Oliveira, L.F. da Silva Brum, J.N. Picada, P. Ardenghi,
Pharmacol. Res. 52 (2005) 199–203.
[14] C.F. Lima, M. Fernandes-Ferreira, C. Pereira-Wilson, Life Sci. 79 (2006) 2056–
2068.
[15] T. Iuvone, D. De Filippis, G. Esposito, A. D’Amico, A.A. Izzo, J. Pharmacol. Exp.
Ther. 317 (2006) 1143–1149.
[16] M.F. Melzig, K. Henke, Planta Med. 71 (2005) 787–789.
[17] Y.G. Hur, C.H. Suh, S. Kim, J. Won, J. Clin. Immunol. 27 (2007) 36–45.
[18] M. Morinaga, N. Osakabe, M. Natsume, US patent no: JP2006273741, Chem.
Abstr. 145 (2006) 404238.
[19] M. Dubois, F. Bailly, G. Mbemba, J.F. Mouscadet, Z. Debyser, M. Witvrouw, P.
Cotelle, J. Med. Chem. 51 (2008) 2575–2579.
[20] D.E. Bogucki, Can. J. Chem. 75 (1997) 1783–1794.
[21] O. Makri, S. Kintzios, Biotechnol. Med. Plants (2004) 19–31.
[22] D. Auriol, US patent no: EP1867229, Chem. Abstr. 148 (2007) 53217.
[23] L.M. Tillekeratne, A. Sherette, P. Grossman, L. Hupe, D. Hupe, R.A. Hudson,
Bioorg. Med. Chem. Lett. 11 (2001) 2763–2767.
[24] A.M.D. El-Mousallamy, U.W. Hawas, S.A.M. Hussein, Phytochemistry 55 (2000)
927–931.
[25] Z. Chen, Y.M. Liu, S. Yang, B.A. Song, G.F. Xu, P.S. Bhadury, L.H. Jin, D.Y. Hu, F.
Liu, W. Xue, X. Zhou, Bioorg. Med. Chem. 16 (2008) 1337–1344.
[26] C.A. Gomes, T. Girão da Cruz, J.L. Andrade, N. Milhazes, F. Borges, M.P.M.
Marques, J. Med. Chem. 46 (2003) 5395–5401.
[27] J.G. Fernández-Bolaños, O. López, J. Fernández-Bolaños, G. Rodríguez-
Gutiérrez, Curr. Org. Chem. 12 (2008) 442–463.
[28] R. Capasso, A. Evidente, S. Avolio, F. Solla, J. Agric. Food Chem. 47 (1999) 1745–
1748.
[29] L.X. Shao, M. Shi, Eur. J. Org. Chem. 2 (2004) 426–430.
[30] W. Brand-Williams, M.E. Cuvelier, C. Berset, Lebensm-Wiss. Technol. 28 (1995)
25–30.
[31] J.J. López, P.C. Redondo, G.M. Salido, J.A. Pariente, J.A. Rosado, Cell. Signal. 18
(2006) 373–381.
[32] P.C. Redondo, N. Ben Amor, G.M. Salido, A. Bartegi, J.A. Pariente, J.A. Rosado,
Cell. Signal. 17 (2005) 1011–1021.
[33] G. Appendino, A. Minassi, N. Daddario, F. Bianchi, G.C. Tron, Org. Lett. 4 (2002)
3839–3841.
*
p < 0.05 (ANOVA) compared to control.
compounds 1–3 significantly reduced TG + ionomycin-evoked
SOCE in these cells (Fig. 2 and Table 4).
4. Conclusion
We have designed and synthesized a series of semi-synthetic
antioxidants related to rosmarinic acid by simply joining well
known and abundant natural phenolic acids with hydroxytyrosol
through an ester linkage. We measured the antiradical properties,
and the effect on Ca2+ mobilization and platelet aggregation of
these compounds in human platelets. Our study shows that the
semi-synthetic compounds provide an advantage over their pre-
cursors. First of all, the antiradical activity of the semi-synthetic
compounds is greater than that of the precursors. Second, the latter
show a more marked effect on cell viability than the semi-syn-
thetic compounds. Finally, our results demonstrate that hydroxy-
tyrosyl gallate (1), and hydroxytyrosyl caffeate (3) exert an
inhibitory effect on platelet aggregation stimulated by the physio-
logical agonist thrombin, an effect that was not observed with their
precursors. The effect of these compounds on platelet aggregation
is likely mediated by attenuation of Ca2+ mobilization. This study
provides the basis for our ongoing efforts to discover new potent
antioxidant and antiaggregant compounds and to elucidate the
mechanism of the potentially aggregant activity of compounds 1
and 3.
[34] K.L. Chandra, P. Saravanan, R.K. Singh, V.K. Singh, Tetrahedron 58 (2002) 1369–
1374.
Acknowledgments
[35] M. Nahmany, A. Melman, Org. Biomol. Chem. 2 (2004) 1563–1572.
[36] O. Mitsunobu, Synthesis (1981) 1–28.
We thank Mercedes Gómez Blázquez for technical assistance.
This study was supported in part by the Spanish Ministerio de Cien-
[37] I. Jardín, N. Ben Amor, J.M. Hernández-Cruz, G.M. Salido, J.A. Rosado, Arch.
Biochem. Biophys. 465 (2007) 16–25.
[38] G.M. Salido, S.O. Sage, J.A. Rosado, Biochim. Biophys. Acta 1793 (2009) 223–
230.
cia
e Innovación (BFU2007-60104; CTQ2005-07005), Junta de
Extremadura-Fundesalud (PRI08A003) and Junta de Andalucía (FS/
FN/ARM).
[39] I. Jardín, J.J. López, G.M. Salido, J.A. Rosado, J. Biol. Chem. 283 (2008) 25296–
25304.
[40] O. Thastrup, B. Foder, O. Scharff, Biochem. Biophys. Res. Commun. 142 (1987)
654–660.
References
[41] J.J. López, C. Camello-Almaraz, J.A. Pariente, G.M. Salido, J.A. Rosado, Biochem. J.
390 (2005) 243–252.
[42] J.J. López, P.C. Redondo, G.M. Salido, J.A. Pariente, J.A. Rosado, Cell. Signal. 18
(2006) 373–381.
[1] P.C. Redondo, G.M. Salido, J.A. Pariente, J.A. Rosado, Biochem. Pharmacol. 67
(2004) 1065–1076.
[2] F.R. Saavedra, P.C. Redondo, J.M. Hernández-Cruz, G.M. Salido, J.A. Pariente, J.A.
Rosado, Arch. Biochem. Biophys. 432 (2004) 261–268.
[3] J.A. Rosado, F.R. Saavedra, P.C. Redondo, J.M. Hernández-Cruz, G.M. Salido, J.A.
Pariente, Haematologica 89 (2004) 1142–1144.