Quercetin derivatives as novel antihypertensive agents: Synthesis and physiological characterization

Article abstract of DOI:10.1016/j.ejps.2015.11.021

The antihypertensive flavonol quercetin (Q1) is endowed with a cardioprotective effect against myocardial ischemic damage. Q1 inhibits angiotensin converting enzyme activity, improves vascular relaxation, and decreases oxidative stress and gene expression. However, the clinical application of this flavonol is limited by its poor bioavailability and low stability in aqueous medium. In the aim to overcome these drawbacks and preserve the cardioprotective effects of quercetin, the present study reports on the preparation of five different Q1 analogs, in which all OH groups were replaced by hydrophobic functional moieties. Q1 derivatives have been synthesized by optimizing previously reported procedures and analyzed by spectroscopic analysis. The cardiovascular properties of the obtained compounds were also investigated in order to evaluate whether chemical modification affects their biological efficacy. The interaction with β-adrenergic receptors was evaluated by molecular docking and the cardiovascular efficacy was investigated on the ex vivo Langendorff perfused rat heart. Furthermore, the bioavailability and the antihypertensive properties of the most active derivative were evaluated by in vitro studies and in vivo administration (1 month) on spontaneously hypertensive rats (SHRs), respectively. Among all studied Q1 derivatives, only the ethyl derivative reduced left ventricular pressure (at 10^{- 8} M ÷ 10^{- 6} M doses) and improved relaxation and coronary dilation. NOSs inhibition by L-NAME abolished inotropism, lusitropism and coronary effects. Chronic administration of high doses of this compound on SHR reduced systolic and diastolic pressure. Differently, the acetyl derivative induced negative inotropism and lusitropism (at 10^{- 10} M and 10^{- 8} ÷ 10^{- 6} M doses), without affecting coronary pressure. Accordingly, docking studies suggested that these compounds bind both β1/β2-adrenergic receptors. Taking into consideration all the obtained results, the replacement of OH with ethyl groups seems to improve Q1 bioavailability and stability; therefore, the ethyl derivative could represent a good candidate for clinical use in hypertension.

Full text of DOI:10.1016/j.ejps.2015.11.021

European Journal of Pharmaceutical Sciences 82 (2016) 161–170
Contents lists available at ScienceDirect
European Journal of Pharmaceutical Sciences
journal homepage: www.elsevier.com/locate/ejps
Quercetin derivatives as novel antihypertensive agents: Synthesis and
physiological characterization
a,1
a,b,1
a
c
a,
Fedora Grandea , Ortensia I. Parisi
, Roberta A. Mordocco , Carmine Rocca , Francesco Puoci ⁎,
c d d
c
e
Luca Scrivano , Anna M. Quintieri , Patrizia Cantafio , Salvatore Ferla , Andrea Brancale , Carmela Saturnino ,
c,
a,2
c,2
Maria C. Cerra ⁎, Maria S. Sinicropi , Tommaso Angelone
a
Department of Pharmacy, Health and Nutritional Sciences, University of Calabria, 87036 Arcavacata di Rende, CS, Italy
Department of Informatics, Modeling, Electronics and Systems Engineering, University of Calabria, 87036 Arcavacata di Rende, CS, Italy
Department of Biology, Ecology and E.S., University of Calabria, 87036 Arcavacata di Rende, CS, Italy
School of Pharmacy and Pharmaceutical Sciences, Cardiff University, Wales, UK
b
c
d
e
Department of Pharmaceutical and Biomedical Sciences, University of Salerno, 84084 Fisciano, SA, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 July 2015
Received in revised form 6 November 2015
Accepted 25 November 2015
Available online 2 December 2015
The antihypertensive flavonol quercetin (Q1) is endowed with a cardioprotective effect against myocardial ische-
mic damage. Q1 inhibits angiotensin converting enzyme activity, improves vascular relaxation, and decreases ox-
idative stress and gene expression. However, the clinical application of this flavonol is limited by its poor
bioavailability and low stability in aqueous medium.
In the aim to overcome these drawbacks and preserve the cardioprotective effects of quercetin, the present study
reports on the preparation of five different Q1 analogs, in which all OH groups were replaced by hydrophobic
functional moieties.
Q1 derivatives have been synthesized by optimizing previously reported procedures and analyzed by spectro-
scopic analysis. The cardiovascular properties of the obtained compounds were also investigated in order to eval-
uate whether chemical modification affects their biological efficacy. The interaction with β-adrenergic receptors
was evaluated by molecular docking and the cardiovascular efficacy was investigated on the ex vivo Langendorff
perfused rat heart. Furthermore, the bioavailability and the antihypertensive properties of the most active deriv-
ative were evaluated by in vitro studies and in vivo administration (1 month) on spontaneously hypertensive rats
Keywords:
Quercetin
Heart
Hypertension
Nitric oxide
(SHRs), respectively.
Among all studied Q1 derivatives, only the ethyl derivative reduced left ventricular pressure (at
−
8
−6
10
M ÷ 10 M doses) and improved relaxation and coronary dilation. NOSs inhibition by L-NAME abolished
inotropism, lusitropism and coronary effects. Chronic administration of high doses of this compound on SHR re-
duced systolic and diastolic pressure. Differently, the acetyl derivative induced negative inotropism and
lusitropism (at 10− M and 10 ÷ 10 M doses), without affecting coronary pressure. Accordingly, docking
studies suggested that these compounds bind both β1/β2-adrenergic receptors.
Taking into consideration all the obtained results, the replacement of OH with ethyl groups seems to improve Q1
bioavailability and stability; therefore, the ethyl derivative could represent a good candidate for clinical use in
hypertension.
10
−8
−6
©
2015 Published by Elsevier B.V.
1
. Introduction
although glycosylation can occur at any of the five hydroxyl groups,
the most common quercetin glycoside presents the sugar moiety
appended at position 3 (isoquercetin).
Quercetin (Q1) is the major representative of flavonols, a sub-
class of flavonoids (Fig. 1). This compound is often present in glyco-
sylated form, the associated sugar moiety is usually glucose and,
The major sources of Q1 are fruits such as apples, citrus, berries and
cherries, vegetables such as onions and broccoli, and beverages such as
red wine and tea (Kaur and Kapoor, 2001). It has been also found in sev-
eral medical plants, such as Ginkgo biloba, Aesculus hippocastanum and
Hypericum perforatum (Sharma and Gupta, 2010). The interest towards
this flavonol is due to its wide range of bioactivity. Q1 shows not only
antioxidant properties like all the flavonoids (Hayek et al., 1997) but
also anti-inflammatory (Ferry et al., 1996), antiviral, antibacterial,
⁎
Corresponding authors.
E-mail addresses: francesco.puoci@unical.it (F. Puoci), maria_carmela.cerra@unical.it
(
M.C. Cerra).
1
These authors have contributed equally to the manuscript.
Co-senior authors.
2
http://dx.doi.org/10.1016/j.ejps.2015.11.021
928-0987/© 2015 Published by Elsevier B.V.
0

162
F. Grande et al. / European Journal of Pharmaceutical Sciences 82 (2016) 161–170
1
using KBr disk method on a Shimadzu FT-IR, Model 8300. The H NMR
spectra (on a Bruker 300/400 MHz spectrometer) were recorded in
DMSO. Chemical shifts were reported as part per million (δ ppm)
using tetramethylsilane (TMS) as an internal standard. The assignment
2
of exchangeable protons was confirmed by the D O exchange studies
wherever required. Mass spectral data was obtained on a QTRAP Ap-
plied Biosystem SCIEX spectrometer or on JEOL-SX-102 instrument
(
JEOL, Tokyo, Japan) by fast atom bombardment (FAB positive mode).
Spectral data (IR and NMR) confirmed the structures of the synthesized
compounds and the purity was ascertained by microanalysis.
2
(
.1.2. 2-(3,4-Diacetoxyphenyl)-4-oxo-4H-chromene-3,5,7-triyl triacetate
Q2)
This compound was synthesized following experimental conditions
Fig. 1. Chemical structure of quercetin.
previously described (Picq et al., 1982).
The crude resulting was purified by column chromatography (chlo-
roform/ethyl acetate, 9:1 as solvent). The product was isolated as a
anticarcinogenic (Deschner et al., 1991; Pereira et al., 1996; Verma et al.,
988), liver-protecting, cardiovascular (Angelone et al., 2011;
Pérez-Vizcaíno et al., 2002) and anti-platelet effects (Pignatelli et al.,
000).
Inhibition of cardiac voltage-gated sodium channels (VGSCs) is anti-
arrhythmic and cardioprotective (Catalano et al., 2013; Carocci et al.,
010). Quercetin, as well as other polyphenols, possesses structural
1
1
white solid, 35% yield. Mp 190 °C (chloroform/ethyl acetate). H NMR
2
(
CDCl
3
) δ: 7.54–7.50 (m, 2H), 7.31–7.02 (m, 2H), 6.56 (d, 1H, J =
1
3
2
1
1
1
.1 Hz), 2.42–2.15 (m, 15H). C NMR (CDCl
67.74, 156.85, 154.25, 150.41, 144.37, 142.19, 127.77, 126.42, 123.92,
23.84, 113.88, 108.96, 21.17, 21.02, 20.65, 20.50. IR υmax: 1775, 1645,
178, 1149, 1127, 1107, 1074, 1021, 899 cm
3
) δ: 169.25, 167.81,
2
similarities to several antiarrhythmic VGSC inhibitors, contributing to
its cardioprotection (Wallace et al., 2006).
−
1
.
The mechanisms by which quercetin exerts these effects are not
completely understood, but it is plausible that different biochemical
processes are involved. For example, the antioxidant properties seem
to be due to the ability of Q1 to act as a scavenger of free radicals and
bind transition metals. This capability makes Q1 able to inhibit lipid per-
oxidation and protect cell membranes and nucleic acids from the free
radical-mediated molecular damages.
Several recent studies focused on the Q1 cardiovascular protective
effects. In particular, the antihypertensive effect of such a flavonol has
been extensively studied. This compound resulted able to decrease oxi-
dative stress, inhibit angiotensin converting enzyme activity, improve
endothelium vascular relaxation, and modulate cell signaling and gene
expression (Angelone et al., 2011; Larson et al., 2010; Pignatelli et al.,
2
.1.3. 2-(2,2-Diphenylbenzo[d][1,3]dioxol-5-yl)-3,5,7-trihydroxy-4H-
chromen-4-one (Q3)
This compound was prepared according to an experimental proce-
dure previously described (Lu et al., 2014).
The crude resulting oil was purified by column chromatography (pe-
troleum ether/ethyl acetate, 4:1 as solvent). The product was isolated as
a yellow solid, 7% yield. Mp 217 °C (petroleum ether/ethyl acetate). H
1
NMR (d
.38–7.23 (m, 6H); 7.09–6.98 (m, 1H); 6.42 (d, 1H); 6.15 (d, 1H).
NMR (CDCl ) δ: 175.13, 162.26, 161.13, 156.2, 148.89, 147.60, 139.76,
35.68, 129.33, 128.48, 127.51, 126.25, 125.40, 124.53, 122.95, 108.67,
07.90, 103.82, 98.84, 94.11. IR υmax: 2800, 1598, 1493, 1260, 1241,
6
-acetone) δ: 12.0 (s br, 3H, OH); 7.76 (m, 2H); 7.50 (m, 3H);
13
7
C
3
1
1
1
2
000).
However, the clinical application of Q1 is limited by its poor bioavail-
−
1
213, 1187, 1043, 1018, 697 cm
.
ability and low stability in aqueous medium, due in particular to the two
phenolic hydroxyl groups at positions 3 and 7 (Campiglia and Pezzi,
2
3
.1.4. 2-(2,2-Diphenylbenzo[d][1,3]dioxol-5-yl)-4-oxo-4H-chromene-
,5,7-triyl triacetate (Q4)
This compound was synthesized following experimental conditions
2
013).
During the last years, several studies have been carried out to en-
hance the pharmacokinetic properties and chemical stability of Q1
and numerous derivatives, showing different biological properties,
have been identified (Kim et al., 2010).
previously described (Kim et al., 2010).
The product was isolated as a white solid, 32% yield. Mp 193 °C
1
(
chloroform/ethyl acetate). H NMR (d
6
-acetone) δ: 7.53–7.31 (m, 5H,
Based on the results obtained in our previous study, which demon-
strated that quercetin is able to induce biphasic inotropic and lusitropic
effects, herein we describe the preparation of five different Q1 deriva-
tives in which OH groups have been replaced with hydrophobic func-
tional groups in the aim to enhance the lipophilic character of this
compound and, thus, improve its bioavailability.
All the synthesized molecules have been subjected to in vitro tests to
estimate their bioavailability and in vivo assays to evaluate the cardio-
vascular effects.
Ar); 7.39–7.29 (m, 5H, Ar); 7.07 (m, 3H); 6.88 (m, 2H); 2.22 (m, 9H).
C NMR (CDCl ) δ: 170.09, 169.30, 167.84, 156.81, 155.24, 154.05,
3
13
1
1
1
2
7
50.33, 149.80, 147.68, 139.53, 129.42, 128.43, 128.41, 128.40, 128.38,
28.36, 128.35, 128.33, 128.26, 126.26, 126.25, 126.23, 126.21, 126.18,
23.53, 123.04, 114.69, 113.65, 108.87, 108.74, 108.28, 21.18, 21.05,
0.66. IR υmax: 1777, 1626, 1496, 1259, 1176, 1138, 1043, 1019,
−1
54 cm
.
2.1.5. 2-(3,4-Diethoxyphenyl)-3,5,7-triethoxy-4H-chromen-4-one (Q5)
2
2
2
. Materials and methods
This compound was prepared according to an experimental proce-
dure previously described (Picq et al., 1982).
.1. Chemistry
The product was isolated as a yellow solid, 20% yield. Mp 123 °C
1
(
n-hexane–chloroform/ethyl acetate). H NMR (CDCl
3
) δ: 7.78–7.58
.1.1. General chemical information
Melting points were determined using a Veego make silicon oil bath-
(m, 2H, Ar); 7.03–6.83 (m, 2H, Ar); 6.34 (m, 1H); 4.28–3.86 (m,
1
3
3
10H); 1.92–1.09 (m, 12H), 0.89–0.75 (m, 3H). C NMR (CDCl ) δ:
type melting point apparatus and are uncorrected. Purity of the
compounds and completion of reactions were monitored by thin layer
chromatography (TLC) on silica gel plates (60 F254; Merck), visualizing
with ultraviolet light or iodine vapors. The IR spectra were recorded
178.90, 164.74, 161.97, 156.73, 151.07, 148.09, 137.85, 122.98,
122.10, 113.57, 112.17, 111.05, 98.09, 92.56, 68.56, 68.03, 64.42,
64.17, 29.68, 15.70, 15.62, 14.80, 14.69, 14.56. IR υmax: 1599, 1474,
−
1
1393, 1267, 1214, 1172, 749 cm
.

F. Grande et al. / European Journal of Pharmaceutical Sciences 82 (2016) 161–170
163
2
4
.1.6. 3,5,7-Tris(benzyloxy)-2-(3,4-bis(benzyloxy)phenyl)-4H-chromen-
-one (Q6)
This compound was prepared according to an experimental proce-
2.3.2.1. Pepsin digestion. 300 μL of a Q5 solution (1.0 mM in DMSO) were
mixed with 1.0 mL of a 0.85 N HCl solution containing 24,000 U of por-
cine pepsin per mL and 3.0 mL of a sodium cholate solution (2% w/v in
distilled water). The obtained mixture was introduced into a dialysis
bag, which was sealed on each end with clamps and immersed in a
flask containing 10 mL of a 0.85 N HCl solution (pH 1.0). The flask was
then incubated for 2 h into a shaking water bath at 37 ± 0.5 °C, simulat-
ing the human physiological conditions of temperature.
dure previously described (Bouktaib et al., 2002).
The product was isolated as a yellow solid, 15% yield. Mp 121 °C
1
(
dichloromethane). H NMR (CDCl
3
) δ: 7.68–7.50 (m, 4H), 7.40–7.08
(
m, 23H); 6.9–6.79 (m, 1H); 6.45–6.28 (m, 2H); 5.19–4.89 (m, 10H).
1
3
3
C NMR (CDCl ) δ: 178.82, 164.44, 162.02, 156.66, 156.30, 147.81,
1
1
1
1
7
45.56, 137.60, 137.44, 136.88, 136.63, 136.42, 135.78, 135.64, 128.84,
28.78, 128.73, 128.67, 128.60, 128.49, 128.33, 128.27, 128.18, 128.01,
27.86, 127.45, 127.33, 127.16, 123.95, 123.43, 122.59, 121.88, 115.26,
14.86, 113.62, 111.53, 106.20, 98.64, 92.98, 74.34, 74.22, 71.11, 70.84,
2
.3.2.2. Pancreatin digestion. At the end of the 2 h pepsin digestion, the
dialysis bag was recovered and carefully opened. 11 mg of amylase,
1
2
1 mg of esterase and 1.3 mL of a 0.8 M NaHCO
3
solution containing
−
1
0.41. IR υmax: 1586, 1255, 1194, 1166, 1123, 1001, 815 cm
.
2.60 mg porcine pancreatin/mL were added to the peptic digesta.
After the digesta and enzymes solution were mixed, the dialysis bag
was sealed again and placed into a flask with 10 mL of a phosphate buff-
er solution at pH 7.0. The flask was incubated into the shaking water
bath at 37 ± 0.5 °C for further 4 h.
In order to evaluate the in vitro bioavailability of the sample, 3 mL of
the medium were withdrawn from the flask at the time points of 2 and
2
.2. Molecular modeling and docking simulations
The molecular modeling simulations were performed on an Apple
Mac pro 2.66 GHz Quad-Core Intel Xeon, Running Ubuntu 12.04. Differ-
ent crystal structures of human β2-adrenergic receptors were examined
in order to identify the putative binding site. The β2-adrenergic receptor
bound to the partial agonist carazol (PDB ID: 2RH1) was chosen for our
docking studies (Cherezov et al., 2007).
No human crystal structure of the β1-adrenergic receptor is avail-
able, so the Meleagris gallopavo β1 receptor co-crystallized with the
partial agonist cyanopindolol has been used (PDB ID: 4BVN)
6
h, after pepsin and pancreatin digestions, respectively. The concentra-
tions of the samples were determined by UV/Vis spectroscopy and cal-
culated by using the equations obtained from the calibration curves of
Q5 standard solutions at pH 1.0 and 7.0, respectively. For this purpose,
all the prepared standard solutions were analyzed by UV–vis spectro-
2
photometer and the correlation coefficient (R ) and slope of the regres-
sion equations obtained by the method of least square were calculated
at pH 1.0 and 7.0, respectively.
The same procedure was applied to evaluate the bioavailability of
Q1, used as control for data confrontation.
(
Miller-Gallacher et al., 2014). Due to the high identity with the
human β1-adrenergic receptor (75% identity) and the conserved
amino acids residues of the binding site, no homology model prepa-
ration was required and the crystal structure was used for the
docking studies.
Each experiment was repeated three times.
Docking studies were performed using Glide docking as part of
Maestro package (Glide, Maestro version 9.5.014, Schrödinger, Inc.,
New York, NY).The two co-crystallized molecules were selected as cen-
ter of the binding site and, then, the selection was extended to the near
amino acids building a 12 Å grid. A ligands database in Maestro format,
prepared using the LigPrep function, was used as input for the docking
calculation. Ligand docking was performed using Glide SP (standard
precision) setting the default values, and no water molecules were
considered. The output solutions were visual inspected in MOE to iden-
tify the interaction between ligand and protein (Molecular Operating
Environment,http://www.chemcomp.com/MOE-Molecular_Operating_
Environment.htm, accessed Aug. 20, 2014).
2.4. In vivo studies
2.4.1. Animals
Animal care, sacrifice and experiments were in accordance with the
Italian law (D.L. 26/2014), the Guide for Care and Use of Laboratory
Animals published by the US National Institutes of Health (2011), and
the Directive 2010/63/EU of the European Parliament on the protection
of animals used for scientific research.
2.4.2. Perfusion technique
The performance of the rat heart was evaluated according to the
Langendorff technique (Angelone et al., 2015). Rats were anesthetized
with ethyl carbamate (2 g/kg rat, i.p.). Hearts were rapidly excised
and transferred in ice-cold buffered Krebs–Henseleit solution (KHS).
Aorta was immediately cannulated with a glass cannula and connected
with the Langendorff apparatus to start perfusion at a constant flow-
rate of 12 mL/min in order to avoid fluid accumulation, the apex of
the left ventricle (LV) was pierced. A water-filled latex balloon, connect-
ed to a BLPR gauge (WRI, Inc. USA), was inserted through the mitral
valve into the LV to allow isovolumic contractions and to continuously
record mechanical parameters. The balloon was progressively filled
with water up to 80 μL to obtain an initial left ventricular end diastolic
pressure of 5–8 mm Hg. Coronary pressure was recorded using another
pressure transducer located just above the aorta.
2
2
.3. Bioavailability studies
.3.1. General bioavailability methods
Hydrochloric acid, pepsin from porcine gastric mucosa, esterase
from porcine liver, α-amylase from porcine pancreas, pancreatin
from porcine pancreas, sodium cholate, sodium dihydrogen phos-
phate, disodium hydrogen phosphate and sodium hydrogen carbon-
ate were purchased by Sigma-Aldrich (Sigma Chemical Co., St. Louis,
MO, USA). All solvents were reagent-grade or HPLC-grade and pro-
vided by Carlo Erba Reagents (Milano, Italia). Dialysis tubes were
provided by Spectrum Laboratories Inc., MWCO: 12–14,000 Da,
U.S.A. Absorption spectra were recorded with a Jasco V-530 UV/Vis
spectrometer.
Left ventricular pressure (LVP) was measured by means of a latex
water-filled balloon inserted into the left ventricle via the left atrium
[
adjusted to obtain left ventricular end-diastolic pressure (LVEDP) of
2
.3.2. In vitro bioavailability studies
In vitro bioavailability studies were carried out in simulated gastric
5–7 mm Hg] and connected to a pressure transducer (BLPR gauge;
WRI, USA). The maximal values of the first derivative of LVP,
[+(LVdP/dt) max, mm Hg/s], which indicates the maximal rate of
left ventricular contraction, the maximal rate of left ventricular pres-
sure decline of LVP [(LVdP/dt) min, mm Hg/s], the rate pressure
product (RPP: HR LVP, in 104 mm Hg beats/min), used as an index
of cardiac work, and LVEDP were used to assess cardiac function.
and intestinal fluids through a slight modified dialysis tubing procedure
Chimento et al., 2013). The dialysis tubing method is based on two
(
enzymatic phases: pepsin digestion, which occurs in the first 2 h, and
pancreatin digestion, in the following 4 h. The two phases are described
below.

164
F. Grande et al. / European Journal of Pharmaceutical Sciences 82 (2016) 161–170
Scheme 1. Reagents and reaction conditions for Q1 derivatives synthesis.
Mean coronary pressure (CP, mm Hg) was calculated as the average
of values obtained during several cardiac cycles.
cardiac preparation with KHS enriched with increasing concentrations
−
10
of either Q2, or Q3, or Q4, or Q5, or Q6, or Q1 (from 10
M to
−
6
The perfusion solution consisted of a modified non recirculating KHS
10 M) for 10 min.
3. Results and discussion
3.1. Chemical
containing 113 mM NaCl, 4.7 mM KCl, 25 mM NaHCO
.8 mM CaCl , 1.2 mM KH PO , 11 mM glucose, 1.1 mM mannitol, 5 mM
Na-pyruvate. pH was adjusted to 7.4 with NaOH (1 M) and the solution
was equilibrated at 37 °C by 95% O –5% CO . All drug-containing solu-
tions were freshly prepared before the experiments.
3 4
, 1.2 mM MgSO ,
1
2
2
4
2
2
The preparation of five Q1 derivatives, in which all the OH groups
have been substituted with different hydrophobic functional groups,
has been performed optimizing previously reported procedures as
depicted in Scheme 1 (Kim et al., 2010; Lu et al., 2014; Picq et al., 1982).
2
.4.2.1. Statistics. Statistical analysis was performed by one-way ANOVA
and Newman–Keuls Multiple Comparison Test, when appropriate.
Differences were considered statistically significant at P b 0.05.
GraphPad Prism software, version 5 (GraphPad Software, San Diego,
CA) will be used for all the statistical analysis.
3.2. Molecular docking studies
2
.4.3. Drug-stimulated preparations
For each drug, preliminary experiments (data not shown) obtained
A series of molecular docking studies were performed to investigate
the putative binding mode of the reported compounds on both β1 and
β2 receptors. Quercetin (Q1) seems able to bind both receptors
(Fig. 2). In the β2-receptor Q1 forms two H-bonds with Asp113,
while in the β1 it establishes three different H-bonds (Asn129,
by the repetitive exposure of each heart to one concentration of the
single drug (10
concentration-response curves were generated by perfusing each
−
8
M) revealed the absence of desensitization. Thus,
Fig. 2. Q1 docked in the β1 (purple) and β2-adrenergic (blue) receptors.

F. Grande et al. / European Journal of Pharmaceutical Sciences 82 (2016) 161–170
165
Fig. 3. Q2 docked in the β2-adrenergic receptor (A). Q3 docked in the β1- (B) and β2-adrenergic (C) receptors. Q5 docked in the β1- (D) and β2-adrenergic (E) receptors. Superposition of
the β2 (blue) and β1 (purple) active sites (F).
Tyr207, Asn319). The possibility to bind both receptors could lead to
β1/β2-adrenergic biological effects caused by quercetin.
between the two types of β-receptors, it is likely that the reported com-
pounds can easily bind both targets balancing the different biological
effects linked with the specific β-adrenergic receptors.
Interestingly, compounds Q4 and Q6 could not be docked in any of
the receptors, probably because of their rigidity and steric hindrance.
This observation could possibly explain the total absence of activity for
Q6 and the very small activity for Q4. Compound Q2 gave good docking
results only for the β2-receptor (Fig. 3A) interacting with Tyr199 and
occupying the active site as Q1. Once again, we could not generate any
useful docking result from the β1-receptor. Q3 and Q5 docks well
with both receptors. Q3 forms H-bonds with Aps121 and Ser211 in
the β1 and with Asp113 and Ser203 in the β2 (Fig. 3B, C). Q5 does not
seem to establish any specific hydrogen bond but it occupies extensively
the active site (Fig. 3D, E).
3.3. Ex vivo studies
3.3.1. Basal conditions
After 20 min of stabilization, the following basal recordings were
measured: LVP = 85 ± 1.7 mm Hg, heart rate (HR) = 276 ± 7.3
beats/min, RPP = 3.3 ± 0.12 104 mm Hg beats/min, CP = 66 ±
1.7 mm Hg, +(LVdP/dT) max = 2768 ± 82.1 (mm Hg/s) and-(LVdP/
dT) max = 1547 ± 28.7 (mm Hg/s).
From these studies, we can say that Q1 and the other most active
compounds Q2, Q3 and Q5, could potentially interact with the β-
adrenergic receptors. Due to the similar active site geometry (Fig. 3F)
Endurance and stability of the preparation, analyzed by measur-
ing performance variables every 10 min, showed that the heart prep-
aration is stable for up to 180 min on the perfusion apparatus.
Fig. 4. Dose-dependent response curves of Q2 (10− M–10−6 M) on LVP, HR, (LVdP/dT) max, (LVdP/dT) min, RPP and CP, on Langendorff perfused rat heart preparations. Percentage
10
changes were evaluated as means ± SEM of 7 experiments. Significance of difference from control values was evaluated by one-way ANOVA; P = 0 b 0.05.

166
F. Grande et al. / European Journal of Pharmaceutical Sciences 82 (2016) 161–170
3
.3.2. Dose–response curves of Q1–Q6
The biological efficacy of Q1 and its modified derivatives was evalu-
reported in the mammalian heart, that β1- and β2-adrenoceptors rep-
resent the main intermediaries of the cardiac response to adrenergic
stimulation, each of them modulates different signal pathways resulting
in different outcomes on cardiac function. β1-adrenoceptor, the most
abundant subtype in mammalian heart (Brodde et al., 2006), induces
the activation of the stimulatory G protein (Gαs)/adenylyl cyclase
(AC)/cAMP/cAMP-dependent protein kinase A (PKA) pathway (Woo
and Xiao, 2002), while β2-adrenoceptor has been shown to regulate
an alternative signaling through the activation of inhibitory G proteins
(Gαi) (Woo and Xiao, 2002). The main signal pathway regulated by
β2-adrenoceptors through Gαi is the phosphatidylinositol-3 kinase
(PI3K) signaling cascade and the activation of the Nitric Oxide pathway
(Conti et al., 2013). Our docking studies indicated that Q5 binds both
β1- and β2-adrenergic receptors. However, since Q5 determined nega-
tive inotropism, lusitropism and vasodilation, we suggest that it mainly
interacts with β2 adrenergic receptor. In order to explore the intracellu-
lar β2-dependent mechanism of action by which Q5 determines nega-
tive inotropism and lusitropism and vasodilation, hearts were treated
with the NOSs inhibitor L-NAME in the presence of Q5 (Fig. 9). Results
indicated that L-NAME abolished Q5-dependent cardiac effects suggest-
ing the obligatory role of NO in mediating the intracellular signaling ac-
tivated by this derivative.
ated by analyzing the hemodynamic performance of the rat heart ex vivo
perfused according to Langendorff. We found that application of Q1
from 10− M to 10 M resulted in positive inotropic and lusitropic ef-
10
−6
−
10
−8
fects at low concentrations (10
and lusitropic effects at higher concentrations (10 and 10 M). Q1
induced vasodilation at low concentration (10
did not modify HR (data not shown). These effects are similar to those
previously published by Angelone et al. (2011). Q2 determined a signif-
icant negative inotropic effect revealed by the reduction of LVP, LVdP/dt
max and RPP, and a negative lusitropic effect revealed by the reduction
of LVdP/dt min. Q2 did not significantly affect HR, while it induced
vasoconstriction at all concentrations tested (Fig. 4). Contrarily, Q3
was unable to affect the rat cardiac performance, except for a small va-
to 10 M) and in negative inotropic
−
7
−6
−
10
−8
to 10 M) while it
−
7
soconstriction detected at high concentration (10 M) (Fig. 5). Q4 ex-
−
6
posure induced a decrease in LVP and LVdP/dt max at 10 M, while it
unchanged the other cardiac parameters (Fig. 6). Notably, Q5 induced
a dose-dependent significant negative inotropic effect revealed by the
reduction of LVP, LVdP/dt max and RPP, and a negative lusitropic effect
revealed by the reduction of LVdP/dt min (Fig. 4). Q5 did not change HR,
but it determined a significant reduction of the coronary pressure at
−
8
−6
high concentrations (from 10 M to 10 M) (Fig. 7). Q6 did not affect
the cardiac performance (Fig. 8).
3.5. In vitro bioavailability studies
3
.4. Cellular mechanism of action of Q5
Most of the natural compounds suffer of low in vivo bioavailability.
In the present study, we analyzed the transduction pathway activat-
Possible reasons for a reduced bioavailability are low intrinsic activity,
low degree of absorption, high metabolization rate, poor activity of me-
tabolites and/or rapid elimination (Bhattaram et al., 2002). Despite its
strong activity, several studies in the past decades related the low bio-
availability of Q1 to poor absorption and rapid metabolism (Cai et al.,
2013).
ed by Q5 since, contrary to all others quercetin derivatives here tested, it
induces a coronary vasodilation which is significant at high doses. This
coronary effect is similar to that elicited by quercetin on the ex vivo rat
heart, already shown by Angelone et al. (2011), and resembles the cor-
onary activity of flavonols (Duarte et al., 1993, 2001). It has been
Fig. 5. Dose-dependent response curves of Q3 (10− M–10−6 M) on LVP, HR, (LVdP/dT) max, (LVdP/dT) min, RPP and CP, on Langendorff perfused rat heart preparations. Percentage
10
changes were evaluated as means ± SEM of 7 experiments. Significance of difference from control values was evaluated by one-way ANOVA; P = 0 b 0.05.

F. Grande et al. / European Journal of Pharmaceutical Sciences 82 (2016) 161–170
167
Fig. 6. Dose-dependent response curves of Q4 (10− M–10−6 M) on LVP, HR, LVdP/dT max, LVdP/dT min, RPP and CP, on Langendorff perfused rat heart preparations. Percentage changes
10
were evaluated as means ± SEM of 7 experiments. Significance of difference from control values was evaluated by one-way ANOVA; P = 0 b 0.05.
In the attempt to improve Q1 bioavailability, several molecular mod-
ifications could be introduced. Considering the hydrophilic/hydrophobic
ratio of the molecule, most of the hydrophilic character of Q1 is provided
by –OH groups. In order to improve its hydrophobicity, as well as its capa-
bility to cross biological membranes, the modification of these groups in
more hydrophobic ones is an interesting strategy. Indeed, the ethylation
Fig. 7. Dose-dependent response curves of Q5 (10− M–10−6 M) on LVP, HR, LVdP/dT max, LVdP/dT min, RPP and CP, on Langendorff perfused rat heart preparations. Percentage changes
10
were evaluated as means ± SEM of 7 experiments. Significance of difference from control values was evaluated by one-way ANOVA; P = 0 b 0.05.

168
F. Grande et al. / European Journal of Pharmaceutical Sciences 82 (2016) 161–170
Fig. 8. Dose-dependent response curves of Q6 (10− M–10−6 M) on LVP, HR, LVdP/dT max, LVdP/dT min, RPP and CP, on Langendorff perfused rat heart preparations. Percentage changes
10
were evaluated as means ± SEM of 7 experiments. Significance of difference from control values was evaluated by one-way ANOVA; P = 0 b 0.05.
of hydroxyl groups (Q5) resulted in a more hydrophobic molecule than
Q1 (see Table 1 with clogP). Interestingly, the other substitutions that
increased significantly the hydrophobicity of the analogs (Q3, Q4 and
Q6), not only might lead to very poorly soluble compounds (clogP N5),
but they could also be rapidly hydrolysed in vivo.
For this reason, Q5 derivative was selected for the evaluation of the
in vitro bioavailability, according to the dialysis tubing procedure, a
fast, reliable and low cost method, which allows to evaluate the bio-
availability of different kinds of compounds. The test was conducted
on Q1 as well, used as control to compare bioavailability data.
Fig. 9. Effects of Q5 (100 nM) before and after treatment with the NOSs inhibitor L-NAME (10 μM) on LVP, LVdP/dT max, LVdP/dT min, and CP, on Langendorff perfused rat heart prepa-
rations. Percentage changes were evaluated as means ± SEM of 5 experiments. Significance of difference from control values was evaluated by t-test; P = 0 b 0.05.

F. Grande et al. / European Journal of Pharmaceutical Sciences 82 (2016) 161–170
169
Table 1
Table 3
clogP (o/w) values of quercetin (Q1) and its synthetic derivatives (Q2-Q6).
Systolic and diastolic BP values measured in WKY and SHR before and after Q5 administra-
tion. Significance of difference was evaluated by t-test.
Quercetin derivative
Weight
clogP (o/w)
Systolic BP (mm Hg)
Diastolic BP (mm Hg)
Q1 (quercetin)
302.238
512.423
466.445
592.566
442.508
752.863
2.032
2.9887
6.5850
7.3080
4.8087
12.0437
Q2
Q3
Q4
Q5
Q6
Before Q5
After Q5
Before Q5
After Q5
WKY
SHR
120 ± 2.3
197 ± 6.6
119 ± 2.5
145 ± 7.0
83 ± 3.9
75 ± 3.3
⁎
⁎
132 ± 6.3
101 ± 5.9
⁎
P b 0.05
Bioavailability is defined as the percentage of the tested molecule re-
covered in the bioaccessible fraction, after in vitro digestion, in relation
to the original non-digested sample and it can be calculated by the fol-
lowing Eq. (1):
was synthesized in the aim to overcome these limitations without af-
fecting the cardioprotective properties of this polyphenol.
The best structural combination required for cardiovascular activity
seems to be functional substitution of each OH group with an ethyl moi-
ety (Q5 derivative). Accordingly, among all the tested compounds, only
this derivative determined a significant reduction of the left ventricular
Bioaccessible content
−
8
−6
Bioavailability ð%Þ ¼
ꢀ 100
ð1Þ
pressure (LVP), from 10 M to 10 M. This compound also increased
relaxation and coronary dilation, and inotropism, lusitropism and coro-
nary effects were abolished by NOSs inhibition by L-NAME. Further-
more, the chronic administration of higher doses of such a compound
on SHR determined a significant reduction of systolic and diastolic
pressure.
Total content
The in vitro evaluation of the bioavailability of Q5 has shown an en-
hanced bioaccessibility since the first 2 h (the obtained results were re-
ported in Table 2), reaching almost 19% at the end of the experiment. On
the other hand, according to our procedure, the bioavailability of un-
modified quercetin (Q1) showed a total percentage slightly under
The acetyl derivative Q2 instead, induced a negative inotropism and
−
10
−8
−6
lusitropism significant at 10
M and from 10 to 10 M, while it
1
1%. This result suggests that improving lipophilicity by ethylation is
did not modify the coronary pressure.
important in increasing the amount of Q1 that could potentially reach
the bloodstream.
The obtained results suggest that the substitution of all OH groups
present in Q1 molecule with ethyl groups improves the bioavailability
and stability of this flavonol. Moreover, the synthesized ethyl derivative
Q5 has been shown good antihypertensive properties; therefore, it
could represent a good candidate for clinical use in the presence of hy-
pertension and, thus, warrant further preclinical and mechanistic
investigation.
Previous studies reported data about in vivo bioavailability of Q1.
Reinboth et al. (2010) reported a 4% absolute bioavailability of Q1 in
dogs, while the bioavailability of unchanged Q1 in pigs is 0.54% (Ader
et al., 2000). Also oral bioavailability evaluation in human volunteers
has been conducted in the past, showing that less than 2% of orally ad-
ministered Q1 can be found in bloodstream (Gugler et al., 1975). Com-
pared to these results, Q5 displays a greatly improved bioavailability.
Acknowledgments
3
.6. In vivo studies
This research work is financially supported under research project
PON — SPREAD BIO OIL grant no. PON01_00293.
We also acknowledge the support of the Life Science Research Net-
work Wales grant no. NRNPGSep14008, an initiative funded through
the Welsh Government's Ser Cymru program.
Due to coronary dilation, the negative inotropism and lusitropism
induced by Q5, this quercetin derivative (10 nM) was administrated
to the rat for 1 month in the drinking water in order to analyze its
anti-hypertensive properties. For this purpose, the spontaneous hy-
pertensive rat (SHR) were used as models of human hypertensive
condition, and Wistar Kyoto rats as normotensive control (n = 6
for each groups) (Harlan, Italy). Tail cuff method was used for weekly
measures of systolic and diastolic blood pressure (BP). Tail cuff was
connected to a pneumatic pulse transducer and a programmed
electro-sphygmomanometer (BP-2000 series II; blood pressure anal-
ysis system, Visitech System). Blood pressures measured are report-
ed in Table 3.
References
Ader, P., Wessmann, A., Wolffram, S., 2000. Bioavailability and metabolism of the flavonol
quercetin in the pig. Free Radic. Biol. Med. 28, 1056–1067.
Angelone, T., Caruso, A., Rochais, C., Caputo, A.M., Cerra, M.C., Dallemagne, P., Filice, E.,
Genest, D., Pasqua, T., Puoci, F., Saturnino, C., Sinicropi, M.S., El-Kashef, H., 2015.
Indenopyrazole oxime ethers: synthesis and β 1-adrenergic blocking activity. Eur.
J. Med. Chem. 92, 672–681.
Angelone, T., Pasqua, T., Di Majo, D., Quintieri, A.M., Filice, E., Amodio, N., Tota, B.,
Giammanco, M., Cerra, M.C., 2011. Distinct signalling mechanisms are involved in
the dissimilar myocardial and coronary effects elicited by quercetin and myricetin,
two red wine flavonols. Nutr. Metab. Cardiovasc. Dis. 21, 362–371.
Bhattaram, V.A., Graefe, U., Kohlert, C., Veit, M., Derendorf, H., 2002. Pharmacokinetics and
bioavailability of herbal medicinal products. Phytomedicine 9 (Suppl. 3), 1–33.
Bouktaib, M., Lebrun, S., Atmanib, A., Rolando, C., 2002. Hemisynthesis of all the O-
monomethylated analogues of quercetin including the major metabolites, through
selective protection of phenolic functions. Tetrahedron 58, 10001–10009.
Brodde, O.E., Bruck, H., Leineweber, K., 2006. Cardiac adrenoceptors: physiological and
pathophysiological relevance. J. Pharmacol. Sci. 100, 323–337.
These data suggest that the chronic administration of the Q5 deter-
mines a reduction of systolic and diastolic pressure in the presence of
hypertension.
4
. Conclusion
Poor bioavailability and low stability of quercetin could be determin-
ing factors of its beneficial effects; therefore, in the present research
study, a small series (Q2–Q6) of variously substituted Q1 derivatives
Cai, X., Fang, Z., Dou, J., Yu, A., Zhai, G., 2013. Bioavailability of quercetin: problems and
promises. Curr. Med. Chem. 20, 2572–2582.
Campiglia, P., Pezzi, V., 2013. Biological activity of 3-chloro-azetidin-2-one derivatives
having interesting antiproliferative activity on human breast cancer cell lines. Bioorg.
Med. Chem. Lett. 23, 6401–6405.
Table 2
Carocci, A., Catalano, A., Bruno, C., Lentini, G., Franchini, C., De Bellis, M., De Luca, A., Conte
Camerino, D., 2010. Synthesis and in vitro sodium channel blocking activity evalua-
tion of novel homochiral mexiletine analogs. Chirality 22 (3), 299–307.
Catalano, A., Budriesi, R., Bruno, C., Di Mola, A., Defrenza, I., Cavalluzzi, M.M., Micucci, M.,
Carocci, A., Franchini, C., 2013. Searching for new antiarrhythmic agents: evaluation
of meta-hydroxymexiletine enantiomers. Eur. J. Med. Chem. 65, 511–516.
Cherezov, V., Rosenbaum, D.M., Hanson, M.A., Rasmussen, S.G.F., Foon, S.T., Kobilka, T.S.,
Choi, H.-J., Kuhn, P., Weis, W.I., Kobilka, B.K., Stevens, R.C., 2007. High-resolution
In vitro bioavailability (%) of Q5 and Q1 after pepsin (2 h) and pancreatin (4 h) digestions.
Time point
Q5 bioavailability (%) Q1 bioavailability (%)
Pepsin digestion (2 h)
Pancreatin digestion (4 h)
Total in vitro bioavailability (6 h) 18.8 ± 1.2
8.5 ± 0.7
4.4 ± 0.9
6.3 ± 0.8
10.7 ± 1.3
10.3 ± 0.8

170
F. Grande et al. / European Journal of Pharmaceutical Sciences 82 (2016) 161–170
crystal structure of an engineered human β
Science 318, 1258–1265.
Chimento, A., Sala, M., Gomez-Monterrey, I.M., Musella, S., Bertamino, A., Caruso, A.,
Sinicropi, M.S., Sirianni, R., Puoci, F., Parisi, O.I., Campana, C., Martire, E., Novellino,
E., Saturnino, C., Conti, V., Russomanno, G., Corbi, G., Izzo, V., Vecchione, C.,
Filippelli, A., 2013. Adrenoreceptorsand nitric oxide in the cardiovascular system.
Front. Physiol. 6, 4–321.
Deschner, E.E., Ruperto, J., Wong, G., Newmark, H.L., 1991. Quercetin and rutin as inhibi-
tors of azoxymethanol-induced colonic neoplasia. Carcinogenesis 12, 1193–1196.
Duarte, J., Pérez-Palencia, R., Vargas, F., Ocete, M.A., Pérez-Vizcaino, F., Zarzuelo, A., et al.,
2
-adrenergic G protein-coupled receptor.
Miller-Gallacher, J.L., Nehmé, R., Warne, T., Edwards, P.C., Schertler, G.F.X., Leslie, A.G.W.,
Tate, C.G., 2014. The 2.1 Å resolution structure of cyanopindolol-bound β
1
-
+
adrenoceptor identifies an intramembrane Na ion that stabilises the ligand-free re-
ceptor. PLoS One 9, e92727.
Pereira, M.A., Grubbs, C.J., Barnes, L.H., Li, H., Olson, G.R., Eto, I., Juliana, M., Whitaker,
L.M., Kelloff, G.J., Steele, V.E., Lubet, R.A., 1996. Effects of the phytochemicals,
curcumin and quercetin, upon azoxymethane-induced colon cancer and 7,12-
dimethylbenz[a]anthracene-induced mammary cancer in rats. Carcinogenesis
17, 1305–1311.
Pérez-Vizcaíno, F., Ibarra, M., Cogolludo, A.L., Duarte, J., Zaragozá-Arnáez, F., Moreno, L.,
López-López, G., Tamargo, J., 2002. Endothelium-independent vasodilator effects of
the flavonoid quercetin and its methylated metabolites in rat conductance resistance
arteries. J. Pharmacol. Exp. Ther. 302, 66–72.
Picq, M., Prigent, A.F., Ngmoz, G., Andrè, A.C., Pacheco, H., 1982. Pentasubstituted querce-
tin analogues as selective inhibitors of particulate 3′:5′-cyclic-AMP phosphodiester-
ase from rat brain. J. Med. Chem. 25, 1192–1198.
Pignatelli, P., Pulcinelli, F.M., Celestini, A., Lenti, L., Ghiselli, A., Gazzaniga, P.P., Violi, F.,
2000. The flavonoids quercetin and catechin synergistically inhibit platelet function
by antagonizing the intracellular production of hydrogen peroxide. Am. J. Clin. Nutr.
72, 1150–1155.
2001. Antihypertensive effects of the flavonoid quercetin in spontaneously hyperten-
sive rats. Br. J. Pharmacol. 133, 117–124.
Duarte, J., Pérez-Vizcaíno, F., Zarzuelo, A., Jiménez, J., Tamargo, J., 1993. Vasodilator effects
of quercetin in isolated rat vascular smooth muscle. Eur. J. Pharmacol. 239, 1–7.
Ferry, D.R., Smith, A., Malkhandi, J., Fyfe, D.W., deTakats, P.G., Anderson, D., Baker, J., Kerr,
D.J., 1996. Phase I clinical trial of the flavonoid quercetin: pharmacokinetics and evi-
dence for in vivo tyrosine kinase inhibition. Clin. Cancer Res. 2, 659–668.
Gugler, R., Leschik, M., Dengler, H.J., 1975. Disposition of quercetin in man after single oral
and intravenous doses. Eur. J. Clin. Pharmacol. 9, 229–234.
Hayek, T., Fuhrman, B., Vaya, J., Rosenblat, M., Belinky, P., Coleman, R., Elis, A., Aviram, M.,
1
997. Reduced progression of atherosclerosis in apolipoprotein E-deficient mice fol-
Reinboth, M., Wolffram, S., Abraham, G., Ungemach, F.R., Cermak, R., 2010. Oral bioavail-
ability of quercetin from different quercetin glycosides in dogs. Br. J. Nutr. 104,
198–203.
Sharma, A., Gupta, H., 2010. Quercetin — a flavonoid. Chron. Young Sci. 1, 10–15.
Verma, A.K., Johnson, J.A., Gould, M.N., Tanner, M.A., 1988. Inhibition of 7,12-
dimethylbenz(a)anthracene and N-nitrosomethylurea induced mammary cancer by
dietary flavonol quercetin. Cancer Res. 48, 5754–5788.
Wallace, C.H.R., Baczko, I., Jones, L., Fercho, M., Light, P.E., 2006. Inhibition of cardiac
voltage-gated sodium channels by grape polyphenols. Br. J. Pharmacol. 149 (6),
657–665.
lowing consumption of red wine, or its polyphenols quercetin or catechin, is associ-
ated with reduced susceptibility of LDL to oxidation and aggregation. Arterioscler.
Thromb. Vasc. Biol. 17, 2744–2752.
Kaur, C., Kapoor, H.C., 2001. Antioxidants in fruits and vegetables — the millennium's
health. Int. J. Food Sci. Technol. 36, 703–725.
Kim, M.K., Park, K.S., Lee, C., Park, H.R., Choo, H., Chong, Y., 2010. Enhanced stability and
intracellular accumulation of quercetin by protection of the chemically or metaboli-
cally susceptible hydroxyl groups with a pivaloxymethyl (POM) promoiety. J. Med.
Chem. 53, 8597–8607.
Larson, A.J., Symons, J.D., Jalili, T., 2010. Quercetin: a treatment for hypertension?—a
review of efficacy and mechanisms. Pharmaceuticals 3, 237–250.
Lu, C., Huang, F., Li, Z., Ma, J., Li, H., Fang, L., 2014. Synthesis and bioactivity of quercetin
aspirinates. Bull. Kor. Chem. Soc. 35, 518–520.
Woo, A.Y., Xiao, R.P., 2002. Beta-adrenergic receptor subtype signaling in heart: from
bench to bedside. Acta Pharmacol. Sin. 33, 335–341.