Discovery of novel A3 adenosine receptor ligands based on chromone scaffold

Article abstract of DOI:10.1016/j.bcp.2012.03.007

A project focused on the discovery of new chemical entities (NCEs) as AR ligands that incorporate a benzo-γ-pyrone [(4H)-1-benzopyran-4-one] substructure has been developed. Accordingly, two series of novel chromone carboxamides placed at positions C2 (co

Full text of DOI:10.1016/j.bcp.2012.03.007

Biochemical Pharmacology 84 (2012) 21–29
Contents lists available at SciVerse ScienceDirect
Biochemical Pharmacology
journal homepage: www.elsevier.com/locate/biochempharm
Discovery of novel A3 adenosine receptor ligands based on chromone scaffold
Alexandra Gaspar a, Joana Reis a, Sonja Kachler c, Silvia Paoletta d, Eugenio Uriarte a,b,c,d
,
Karl-Norbert Klotz c, Stefano Moro d, Fernanda Borges a,
*
a CIQUP/Departamento de Quı´mica e Bioquı´mica, Faculdade de Cieˆncias, Universidade do Porto, 4169-007 Porto, Portugal
b
´
´
Departamento de Quımica Organica, Facultad de Farmacia, Universidad de Santiago de Compostela, 15782 Santiago de Compostela, Spain
c
¨
¨
¨
Institut fu¨r Pharmakologie und Toxikologie, Universitat Wurzburg, 97078 Wurzburg, Germany
d
`
Molecular Modeling Section (MMS), Dipartimento di Scienze Farmaceutiche, Universita di Padova, via Marzolo 5, I-35131 Padova, Italy
A R T I C L E I N F O
A B S T R A C T
Article history:
A project focused on the discovery of new chemical entities (NCEs) as AR ligands that incorporate a
benzo- -pyrone [(4H)-1-benzopyran-4-one] substructure has been developed. Accordingly, two series
of novel chromone carboxamides placed at positions C2 (compounds 2–13) and C3 (compounds 15–26)
of the -pyrone ring were synthesized using chromone carboxylic acids (compounds 1 or 14) as starting
Received 27 January 2012
Accepted 9 March 2012
Available online 17 March 2012
g
g
materials. From this study and on the basis of the obtained structure–activity relationships it was
concluded that the chromone carboxamide scaffold represent a novel class of AR ligands. The most
remarkable chromones were compounds 21 and 26 that present a better affinity for A3AR (Ki = 3680 nM
and Ki = 3750 nM, respectively). Receptor-driven molecular modeling studies provide information on
the binding/selectivity data of the chromone. The data so far acquired are instrumental for future
optimization of chromone carboxamide as a selective A3AR antagonist.
Keywords:
Drug discovery
Adenosine receptor ligand
Chromone scaffold
ß 2012 Elsevier Inc. All rights reserved.
1. Introduction
cancer cell and how it differs from a normal one [1]. Accordingly, a
number of potential targets as well as the development of a new
Cancer is a very complex disease, linked with different initiating
causes, cofactors and promoters, and several types of cellular
damage. Advancing knowledge on the cellular and molecular
biology of the processes that regulate cell proliferation, cell
differentiation and cellular responses to external signals provide a
wealth of information about the biochemistry and biology of the
generation of anticancer agents must be exploited, based on the
differences between normal and cancer cells [2].
During the last decade different approaches to treating cancer
have been developed based mainly on specific targets that are
mostly expressed in tumor but not in normal cells [2]. Interest-
ingly, it was already shown that adenosine receptor (AR) levels in
various tumor cells are up regulated, a finding which may suggest
that a specific AR may serve as a biological marker and as a target
for specific ligands leading to cell growth inhibition [3]. In
particular, the human A3 AR, which is the most recently identified
Abbreviations: [3H]CCPA,
[3H](2R,3R,4S,5R)-2-[2-chloro-6-(cyclopentylamino)-
purin-9-yl]-5-(hydroxymethyl)oxolane-3,4-diol; [3H]HEMADO, [3H] 2-(1-Hexy-
nyl)-N6-methyladenosine; [3H]NECA, [3H]adenosine-50-(N-ethylcarboxamide);
AR, Adenosine receptor; Arg, arginine; Asn, asparagine; Asp, aspartic acid; BBr3,
boron tribromide; BOP, (benzotriazol-1-yloxy)tris(dimethylamino)phosphonium
hexafluorophosphate; CHO, Chinese hamster ovary cells; Cys, cysteine; DAD,
diode-array detector; DIPEA, N,N-diisopropylethylamine; DMF, dimethylformamide;
EI-MS, electron impact mass spectra; Glu, glutamic acid; HPLC, high performance
liquid chromatography; Ile, isoleucine; Ki, inhibition constant; Leu, leucine; MOE,
molecular operating environment; MOPAC, molecular orbital PACkage; NCE, new
chemical entity; NMR, nuclear magnetic resonance; PLANTS, protein-Ligand ANT
System; Phe, phenylalanine; Pro, proline; PyBOP, benzotriazol-1-yloxytripyrrolidi-
nophosphonium hexafluorophosphate; R-PIA, N-R-N6-(1-methyl-2-phenylethyl)a-
denosine; SAR, structure–affinity relationship; TLC, thin layer chromatography; TMS,
tetramethylsilane; Trp, tryptophan; Tyr, tyrosine; UV, ultraviolet; ZM241385, 4-(2-
[7-Amino-2-(2-furyl)[1,2,4]triazolo[2,3-a][1,3,5]triazin-5-ylamino]ethyl)phenol.
* Corresponding author at: CIQ/Department of Chemistry and Biochemistry,
Faculty of Sciences, University of Porto, Porto 4169-007, Portugal.
adenosine receptor, is involved in
a variety of important
physiological processes that include inflammation, cell growth
and immunosuppression [4–7].
There have been many attempts to design and develop A3 AR
agonists and antagonists, and over the past decade, the search for
ligands that show selectivity toward individual receptor subtypes
has intensified as their role in many therapeutic areas expands
[4,8,9]. Despite the intense discovery efforts the overall process
has failed to deliver selective (agonists or antagonists) drug
candidates, with exception of CF102 (1-[2-chloro-6-[[(3-iodo-
phenyl)methyl]amino]-9H-purin-9-yl]-1-deoxy-N-methyl-b-D-
ribofuran uronamide, Cl-IB-MECA) that is in clinical trials
[4,10,11]. Main problems include side effects due to the ubiquity
of the receptors or to low absorption, short half-life and toxicity of
the ligands [10]. These facts prompted an intensive research effort
Tel.: +351 220402560.
E-mail address: fborges@fc.up.pt (F. Borges).
0006-2952/$ – see front matter ß 2012 Elsevier Inc. All rights reserved.
doi:10.1016/j.bcp.2012.03.007

22
A. Gaspar et al. / Biochemical Pharmacology 84 (2012) 21–29
toward the development of novel, selective and potent AR
receptor ligands suitable for chemotherapeutic purposes.
Concurrently, despite the steady increase in R&D expendi-
tures within the pharmaceutical industry, the number of new
chemical entities (NCEs) reaching the market has actually
decreased dramatically. Therefore, privileged structures, such
as indoles, arylpiperazines, biphenyls and benzopyranes (e.g.
coumarins and chromones), are currently considered as poten-
tially successful approaches in drug discovery and have been
used successfully before in medicinal chemistry programs to
identify NCEs [12].
2. Materials and methods
2.1. Materials
Chromone-2-carboxylic and chromone-3-carboxylic acids, (ben-
zotriazol-1-yloxy)tris(dimethylamino)phosphonium hexafluoro-
phosphate (BOP), benzotriazol-1-yloxytripyrrolidinophosphonium
hexafluorophosphate (PyBOP), N,N-diisopropylethylamine (DIPEA),
dimethylformamide (DMF), boron tribromide (BBr3), aniline and its
´
derivatives were purchased from Sigma–Aldrich Quımica S.A.
(Sintra, Portugal). All other reagents and solvents were pro analysis
grade and were acquired from Merck (Lisbon, Portugal) and used
without additional purification.
Accordingly, a project focused on the discovery of NCEs as AR
ligands that incorporate a benzo-g-pyrone [(4H)-1-benzopyran-4-
one] substructure has been developed. Based on knowledge
acquired so far no information on the development of putative
adenosine ligands based on this type of scaffold have been
reported, concerning the flavonoid family. However, flavonoids are
natural secondary metabolites that possess a C6–C3–C6 skeleton.
The present chromone series possess some structural similarities
with flavones, namely the presence of A and C rings, but the B ring
is absent.
Thin-layer chromatography (TLC) was carried out on pre-coated
silica gel 60 F254 (Merck, Lisbon, Portugal) with layer thickness of
0.2 mm. For analytical control the following systems were used:
ethyl acetate/petroleum ether, ethyl acetate/methanol, chloro-
form/methanol in several proportions. The spots were visualized
under UV detection (254 and 366 nm) and iodine vapor. Normal-
phase column chromatography was performed using silica gel 60
0.2–0.5 or 0.040–0.063 mm (Merck, Lisbon, Portugal).
Therefore, the aim of the present study is the design and
synthesis of a library of novel adenosine receptor ligands based on
the chromone scaffold that was obtained through the application
of innovative synthetic strategies (Scheme 1) [13]. Lead discovery
of new AR ligands based on a chromone scaffold guided by
structure–affinity-relationships (SAR) and molecular modeling is
the aim of the present work.
The purity of the final products (>97% purity) was verified by
high-performance liquid chromatography (HPLC) equipped with a
UV detector. Chromatograms were obtained in an HPLC/DAD
system, a Jasco instrument (pumps model 880-PU and solvent
mixing model 880-30, Tokyo, Japan), equipped with a commercially
prepacked Nucleosil RP-18 analytical column (250 mm ꢀ 4.6 mm,
5
mm, Macherey-Nagel, Duren, Germany), and UV detection (Jasco
O
O
COOH
O
COOH
O
(14)
(1)
a)
a)
O
CONHR
O
CONHR
O
O
(2) R=Ph
(15) R=Ph
(3) R=C6H11
(16) R=C6H11
(4) R=C3H7
(17) R=C3H7
(5) R=(4'-OH-Ph)
(6) R=(4'-OCH3-Ph)
(7) R=(4'-CH3-Ph)
(8) R=(3'-OH-4'-OCH3-Ph)
(9) R=(3',4'-OCH3-Ph)
(10) R=(4'-Cl-Ph)
(11) R=(4'-OCF3-Ph)
(12) R=(4'-CF3-Ph)
(13) R=(4'-NO2-Ph)
(18) R=(4'-OH-Ph)
(19) R=(4'-OCH3-Ph)
(20) R=(4'-CH3-Ph)
(21) R=(3'-OH-4'-OCH3-Ph)
(22) R=(3',4'-OCH3-Ph)
(23) R=(4'-Cl-Ph)
(24) R=(4'-OCF3-Ph)
(25) R=(4'-CF3-Ph)
(26) R=(4'-NO2-Ph)
(a)
R-NH2, BOP or PyBOP, DIPEA in DCM/DMF
Scheme 1. Structure of the chromone carboxamides under study.

A. Gaspar et al. / Biochemical Pharmacology 84 (2012) 21–29
23
model 875-UV) at the maximum wavelength of 254 nm. The mobile
phase consisted of a methanol/water or acetonitrile/water (gradient
mode, room temperature) at flow rate of 1 mL/min. The
NH), 13.76 (0.7H, s, NH); MS/EI m/z (int.rel.): 296 (55), 295 (Mꢁ+
,
100), 280 (27), 252 (21), 174 (22), 173 (94), 147 (18), 132 (14), 121
(37), 92 (14), 77 (12).
a
chromatographic data was processed in a Compaq computer, fitted
with CSW 1.7 software (DataApex, Czech Republic). 1H NMR data
were acquired, at room temperature, on a Bru¨ker AMX 300
spectrometer operating at 300.13 MHz, respectively. Dimethylsulf-
2.2.5. N-(4-Methylphenyl)-4-oxo-4H-1-benzopyran-3-carboxamide
(20)
Yield: 44%; MP: 179–182 8C; 1H NMR (CDCl3): 2.39 (3H, s, CH3),
7.23–7.32 (6H, m, H(6), H(8), H(20), H(30), H(50), H(60), 7.60 (1H, ddd,
J = 8.6, 7.0, 1.6, H(7)), 8.07/8.14 (1H, dd, J = 7.8, 1.6, H(5)), 8.86/8.99
(1H, s, H(2)), 11.92/13.69 (1H, s, NH); MS/EI m/z (int.rel.): 280 (57),
279 (Mꢁ+, 99), 278 (21), 250 (10), 174 (24), 173 (100), 159 (16), 158
(23), 131 (37), 130 (44), 121 (44), 92 (11), 91 (24), 77 (12), 65 (20).
oxide-d6 was used as a solvent; chemical shifts are expressed in
d
(ppm) values relative to tetramethylsilane (TMS) as internal
reference; coupling constants (J) are given in Hz. Electron impact
massspectra (EI-MS)werecarriedout ona VGAutoSpecinstrument;
the data are reported as m/z (% of relative intensity of the most
important fragments). Melting points were obtained on a Stuart
Scientific SMP1 apparatus and are uncorrected.
2.2.6. N-(3,4-Dimethoxyphenyl)-4-oxo-4H-1-benzopyran-3-
carboxamide (22)
1
2.2. Synthesis of chromone carboxamide derivatives
Yield: 50%; MP: 248–254 8C; H NMR (CDCl3): 3.77 (3H, s, 30-
OCH3), 3.85 (3H, s, 40-OCH3), 7.01 (1H, d, J = 8.6, H(60)), 7.14 (1H, dd,
J = 8.6, 1.6, H(50)), 7.37–7.30 (3H, m, H(6), H(8), H(20)), 7.73–7.68
(1H, m, H(7)), 7.96 (1H, d, J = 7.7, H(5)), 8.86/8.83 (1H, s, H(2)),
11.82/13.58 (1H, s, NH), 13.58 (0.7 H, s, NH); MS/EI m/z (int.rel.):
326 (21), 325 (Mꢁ+, 100), 311 (10), 310 (63), 207 (60), 173 (62), 121
(17), 93 (20), 79 (15), 77 (15).
General procedure: 2-Carboxychromone (1) or 3-carboxychro-
mone (14) (0.50 g; 2.63 mmol) was dissolved in of DMF (6 mL) and
of DIPEA (0.37 mL). The solution was then cooled at 0 8C in an ice-
water bath, and a BOP (1.16 g; 2.63 mmol) or PyBOP (1.37 g;
2.63 mmol) solution in CH2Cl2 (6 mL) was added. The mixture was
stirred during 30 min. After, the phenylamine derivative was
added in equimolar amount. The temperature was gradually
increased to room temperature. The reaction was stirred for
additional 4 h. Following the workup and after extraction, the
organic phases were dried over Na2SO4. Solutions were decolorized
with activated charcoal, when necessary. The recrystallization
solvents were ethyl acetate or ethyl ether/n-hexane.
The structural elucidation of the other carboxamides was
described elsewhere [14,15].
2.3. Radioligand binding assays
2.3.1. CHO membrane preparation
All the pharmacological methods including in membrane
preparation for radioligand binding experiments followed the
procedures as described earlier [16].
2.2.1. N-(4-Methoxyphenyl)-4-oxo-4H-1-benzopyran-2-
carboxamide (6)
Membranes for radioligand binding were prepared from cells
stably transfected with the human adenosine receptor subtypes
(A1, A2A, and A3 expressed on CHO cells) in a two-step procedure. In
the first low-speed step (1000 ꢀ g for 4 min), the cell fragments
and nuclei were removed. After that, the crude membrane fraction
was sedimented from the supernatant at 100,000 ꢀ g for 30 min.
The membrane pellet was then resuspended in the specific buffer
used for the respective binding experiments, frozen in liquid
nitrogen, and stored at ꢂ80 8C. For the measurement of the
adenylyl cyclase activity in A2B receptor expressed on CHO cells,
only one step of centrifugation was used in which the homogenate
was sedimented for 30 min at 54,000 ꢀ g. The resulting crude
membrane pellet was resuspended in 50 mM Tris–HCl, pH 7.4 and
immediately used for the adenylyl cyclase assay.
Yield: 85%; MP: 214–223 8C; 1H NMR (CDCl3): 3.83 (3H, s,
OCH3), 6.94 (2H, d, J = 9.2, H(30), H(50)), 7.27 (1H, s, H(3)), 7.49 (1H,
ddd, J = 8.0; 7.2; 1.0, H(6)), 7.59–7.64 (3H, m, H(8), H(20), H(60)),
7.78 (1H, ddd, J = 8.5; 7.1; 1.6, H(7)), 8.25 (1H, dd, J = 8.0, 1.6, H(5)),
8.53 (1H, s, NH). MS/EI m/z (int.rel.): 296 (14), 295 (Mꢁ+, 100), 294
(30), 266 (15), 173 (13), 145 (10), 122 (68), 95 (17), 89 (22), 71 (10),
69 (11), 57 (15).
2.2.2. N-(4-Methylphenyl)-4-oxo-4H-1-benzopyran-2-carboxamide
(7)
Yield: 56%; MP: 233–237 8C; 1H NMR [(CD3)2SO]: 2.30 (3H, s,
CH3), 6.97 (1H, s, H(3)), 7.23 (2H, d, J = 8.2, H(30), H(50)), 7.57 (1H,
ddd, J = 7.9, 7.0, 1.0, H(6)), 7.69 (2H, d, J = 8.3, H(20), H(60)), 7.85 (1H,
dd, J = 8.5, 1.0, H(8)), 7.93 (1H, ddd, J = 8.5, 7.0, 1.5, H(7)), 8.08 (1H,
dd, J = 8.0, 1.4, H(5)), 10.68 (1H, s, NH); MS/EI m/z (int.rel.): 280
(32), 279 (Mꢁ+, 100), 278 (94), 264 (10), 262 (29), 251 (11), 250
(46), 233 (14), 158 (17), 107 (10), 106 (35), 89 (53), 79 (14), 77 (20).
2.3.2. Human cloned A1, A2A, A3 adenosine receptor binding assay
Binding of [3H]CCPA (2-chloro-N6-cyclopentyladenosine, GE
Healthcare, Freiburg, Germany) to CHO cells transfected with the
human recombinant A1 adenosine receptor was performed as
previously described [16]. Competition experiments were per-
2.2.3. N-(3,4-Dimethoxyphenyl)-4-oxo-4H-1-benzopyran-2-
carboxamide (9)
formedfor3 hat25 8Cin200
0.2 U/mL adenosine deaminase, 20
m
Lofbuffercontaining1 nM[3H]CCPA,
g of membrane protein in
Yield: 45%; MP: 196–198 8C; 1H NMR [(CD3)2SO]: 3.77/3.79 (6H,
2 s, 2ꢀ OCH3), 6.97 (1H, s, H (3)), 7.01 (1H, d, J = 8.7, H(50)), 7,40 (1H,
dd, J = 8.7; 2.4, H(60)), 7.48 (1H, d, J = 2.4, H(20)), 7.58 (1H, ddd,
J = 8.0, 6.8, 1.2, H(6)), 7,.86 (1H, d, J = 7.7, H(8)), 7.95 (1H, ddd,
J = 8.5, 7.0, 1.6, H(7)), 8.10 (1H, dd, J = 8.0, 1.5, H(5)), 10.66 (1H, s,
NH); MS/EI m/z (int.rel.): 326 (20), 325 (Mꢁ+, 100), 310 (21), 308
(15), 173 (24), 145 (22), 89 (37).
m
50 mM Tris/HCl, pH 7.4 and tested compound in different
concentrations. Nonspecificbindingwasdeterminedinthepresence
of 1 mM theophylline and amounted to <5% of total binding [16].
Binding of [3H]NECA (N-ethylcarboxamidoadenosine, GE
Healthcare, Freiburg, Germany) to CHO cells transfected with
the human recombinant A2A adenosine receptors was performed
following the conditions as described for the A1 receptor binding
[16]. In the competition experiments, samples containing a protein
2.2.4. N-(4-Methoxyphenyl)-4-oxo-4H-1-benzopyran-3-
carboxamide (19)
Yield: 55%; MP: 173–177 8C; 1H NMR (CDCl3): 3.84 (3H, s,
OCH3), 6.98 (2H, d, J = 9.0, H(30), H(50)), 7.25–7.32 (4H, m, H(6),
H(8), H(20), H(60)), 7.59 (1H, ddd, J = 8.6, 6.9, 1.7, H(7)), 8.06/8.14
(1H, dd, J = 7.8, 1.6, H(5)), 8.80/8.94 (1H, s, H(2)), 11.95/13.76 (1H, s,
amount of 50
different concentrations were incubated for 3 h at 25 8C. Nonspe-
cific binding was determined in the presence of 100 M R-PIA (R-
m
g, 30 nM [3H]NECA and tested compound in
m
N6-phenylisopropyladenosine) and represented about 50% of total
binding [16].

24
A. Gaspar et al. / Biochemical Pharmacology 84 (2012) 21–29
Binding of [3H]HEMADO (2-(1-hexynyl)-N-methyladenosine,
Tocris, Bristol, UK) to CHO cells transfected with the human
recombinant A3 adenosine receptors was carried out as previously
described [16,17].
the loop domains were constructed by the loop search method
implemented in MOE on the basis of the structure of compatible
fragments found in the Protein Data Bank. In particular, loops
were modeled first in random order. For each loop, a contact
energy function analyzed the list of candidates collected in the
segment searching stage, taking into account all atoms already
modeled and any atoms specified by the user as belonging to the
model environment. These energies were then used to make a
Boltzmann-weighted choice from the candidates, the coordinates
of which were then copied to the model. Subsequently, the side
chains were modeled using a library of rotamers generated by
systematic clustering of the Protein Data Bank data, using the
same procedure. Side chains belonging to residues whose
backbone coordinates were copied from a template and were
modeled first, followed by side chains of modeled loops. Outgaps
and their side chains were modeled last. Special caution has to be
given to EL2 because amino acids of this loop could be involved in
direct interactions with the ligands. A driving force to the peculiar
fold of the EL2 loop might be the presence of a disulfide bridge
between cysteines in TM3 and EL2. Since this covalent link is
conserved in both hA2A and hA3 receptors, the EL2 loop was
modeled using a constrained geometry around the EL2-TM3
disulfide bridge. The constraints were applied before the
construction of the homology model, in particular during the
sequence alignment, selecting the cysteine residues involved in
the disulfide bridge in hA2A to be constrained with the
corresponding cysteine residues in hA3 sequence. In particular,
Cys166 (EL2) and Cys77 (3.25) of the hA2A receptor were
constrained, respectively, with Cys166 (EL2) and Cys83 (3.25)
of the hA3 receptor. During the alignment, MOE-Align attempted
to minimize the number of constraint violations. Then, after
running the homology modeling, the presence of the conserved
disulfide bridge in the model was manually checked. After the
heavy atoms were modeled, all hydrogen atoms were added using
the Protonate 3D methodology part of the MOE suite. This
application assigned a protonation state for each chemical groups
that minimized the total free energy of the system (taking
titration into account) [26].
The competition experiments were performed for 3 h at 25 8C in
buffer solution containing 1 nM [3H]HEMADO, 20
mg membrane
protein in 50 mM Tris–HCl, 1 mM EDTA (ethylenediaminotetraa-
cetate), 10 mM MgCl2, pH 8.25 and tested compound in different
concentrations. Nonspecific binding was determined in the
presence of 100
mM R-PIA and was below 2% of total binding [17].
All incubations were done in 96 well microplates with filter
bottoms allowing for separation of bound and free ligand by
filtration. Membranes with bound ligand were washed with
icecold buffer to remove unbound ligand [16,17]. Ki values from
competition experiments were calculated with the program SCTFIT
[19] and are reported as geometric means of at least three
independent experiments with 95% confidence limits [16,17].
2.3.3. Adenylyl cyclase activity
Because of the lack of a suitable radioligand for hA2B receptor in
binding assay, the potency of antagonists at A2B receptor
(expressed on CHO cells) was determined in adenylyl cyclase
experiments instead. The procedure was carried out as described
previously with minor modifications [16]. Membranes were
incubated with about 150,000 cpm of [a-
32P]ATP (Hartmann-
Analytic, Braunschweig, Germany) for 20 min in the incubation
mixture as described [16] without EGTA and NaCl. For agonists, the
EC50 values for the stimulation of adenylyl cyclase were calculated
with the Hill equation. Hill coefficients in all experiments were
near unity. IC50 values for concentration-dependent inhibition of
NECA-stimulated adenylyl cyclase caused by antagonists were
calculated accordingly. Dissociation constants (Ki) for antagonist
were then calculated from the Cheng and Prusoff equation [18].
2.4. Molecular modeling
All modeling studies were carried out on a 20 CPU (Intel Core2
Quad CPU 2.40 GHz) Linux cluster. Homology modeling, energy
calculation, and analyses of docking poses were performed using
the Molecular Operating Environment (MOE, version 2008.10)
suite [20]. The software package MOPAC (version 7) [21],
implemented in MOE suite, was utilized for all quantum
mechanical calculations. Docking simulation was performed using
GOLD suite [21].
Protein stereochemistry evaluation was then performed by
several tools (Ramachandran plot; backbone bond lengths, angles
and dihedral plots; clash contacts report; rotamers strain energy
report) implemented in MOE suite [20].
2.4.2. Molecular docking of adenosine receptors antagonists
Ligand structures were built using MOE-builder tool, part of the
MOE suite [20], and were subjected to MMFF94x energy
minimization until the rms of conjugate gradient was
2.4.1. Homology models of hA3 AR
Based on the assumption that GPCRs share similar TM
boundaries and overall topology, a homology model of the hA3
adenosine receptor was constructed, as previously reported
[22,23], based on a template of the recently published crystal
structure of hA2A receptor (PDB code: 3EML) [24].
The numbering of the amino acids follows the arbitrary scheme
by Ballesteros and Weinstein. According to this scheme, each
amino acid identifier starts with the helix number, followed by the
position relative to a reference residue among the most conserved
amino acid in that helix. The number 50 is arbitrarily assigned to
the reference residue [25].
Firstly, the amino acid sequences of TM helices of the hA3
receptor were aligned with those of the template, guided by the
highly conserved amino acid residues, including the DRY motif
(Asp3.49, Arg3.50, and Tyr3.51) and three proline residues
(Pro4.60, Pro6.50, and Pro7.50) in the TM segments of GPCRs.
The same boundaries were applied for the TM helices of hA3
receptor as they were identified from the 3D structure for the
corresponding sequences of the template, the coordinates of which
were used to construct the seven TM helices for hA3 receptor. Then,
<0.05 kcal/mol Aꢂ1. Partial charges for the ligands were calculated
˚
using PM3/ESP methodology.
Four different programs have been used to calibrate our docking
protocols: MOE-Dock [20], GOLD [27], Glide [28], and PLANTS [29].
In particular, ZM-241385 was re-docked into the crystal structure
of the hA2A adenosine receptor (PDB code: 3EML) with different
docking algorithms and scoring functions, as already described
[22]. Then, RMSD values between predicted and crystallographic
positions of ZM-241385 were calculated for each of the docking
algorithms. The results showed that docking simulations per-
formed with GOLD gave the lowest RMSD value, the lowest mean
RMSD value and the highest number of poses with RMSD value
˚
<2.5 A.
On the basis of the best docking performance, all antagonist
structures were docked into the hypothetical TM binding site of the
hA3 AR model and that of the hA2A AR crystal structure, by using the
docking tool of the GOLD suite [27]. Searching was conducted
within
a user-specified docking sphere, using the Genetic
Algorithm protocol and the GoldScore scoring function. GOLD

A. Gaspar et al. / Biochemical Pharmacology 84 (2012) 21–29
25
performs a user-specified number of independent docking runs (25
in our specific case) and writes the resulting conformations and
their energies in a molecular database file. The resulting docked
complexes were subjected to MMFF94x energy minimization until
acid activation were organophosphoric compounds, namely
(benzotriazol-1-yloxy)tris(dimethylamino) phosphonium hexa-
fluorophosphate (BOP) and (benzotriazol-1-yloxy) tripyrrolidino-
phosphonium hexafluorophosphate (PyBOP) [31]. In all the
reactions, N,N-diisopropylethylamine (DIPEA) was used instead
of the classic triethylamine since a significant yield increase was
observed due mainly to an improvement in the purification steps
[13].
The synthetic procedure has an advantage over the Schotten–
Baumann reaction since it avoids the step of generation of an acyl
halide with reagents such as thionyl chloride or phosphorus
pentachloride, circumventing some of the drawbacks related to the
use of this type of reagents, namely the ring-opening of the
benzopyran nucleus [32]. Furthermore, phosphonium salts (BOP or
PyBOP) were selected as coupling reagents since some of the side
reactions described with the employment of carbodiimides are
avoided facilitating product purification with improvement of the
yield of the reaction [31].
the rms of conjugate gradient was <0.1 kcal/mol Aꢂ1. Charges for
˚
the ligands were imported from the MOPAC output files using
PM3/ESP methodology.
Prediction of antagonist–receptor complex stability (in terms of
corresponding pKi value) and the quantitative analysis for non-
bonded intermolecular interactions (H-bonds, transition metal,
water bridges, hydrophobic, electrostatic) were calculated and
visualized using several tools implemented in MOE suite [20].
Electrostatic and hydrophobic contributions to the binding
energy of individual amino acids have been calculated as
implemented in MOE suite [20]. In order to estimate the
electrostatic contributions, atomic charges for the ligands were
calculated using PM3/ESP methodology. Partial charges for protein
amino acids were calculated on the basis of the AMBER99 force
field.
3.2. Pharmacology
3. Results and discussion
3.2.1. Binding affinity at human A1, A2A, and A3 adenosine receptors
The affinity of the new potential antagonists for the human
adenosine receptor subtypes hA1, hA2A, hA3 (expressed in Chinese
hamster ovary (CHO) cells) was determined in radioligand
competition experiments [16–19]. In this assay, we measured
the displacement of: (i) specific [3H]CCPA binding at hA1 receptors,
(ii) specific [3H]NECA binding at hA2A and [3H]HEMADO binding
hA3 receptors. The data were expressed as Ki (dissociation
constant), which was calculated with the program SCTFIT [19],
and given as geometric means of at least three experiments,
including 95% confidence intervals. The receptor binding affinities
of the synthesized compounds (2–13 and 15–26) are reported in
Tables 1 and 2, respectively.
3.1. Chemistry
Two series of novel chromone carboxamides placed at positions
C2 (compounds 2–13) and C3 (compounds 15–26) of the g-pyrone
ring were synthesized using chromone carboxylic acids (com-
pounds 1 or 14) as starting materials (Scheme 1).
Carboxylic acids may be converted into carboxamides by
treating them with amines. However, the direct reaction does not
occur spontaneously at ambient temperature, with the necessary
elimination of water only taking place at high temperatures
[30,31]. In order to activate carboxylic acids, one can use so-called
coupling reagents that act as stand-alone reagents to generate
compounds such as acid chlorides, (mixed) anhydrides, carbonic
anhydrides or active esters [31].
3.2.2. Adenylyl cyclase activity
Because of the lack of a suitable radioligand for hA2B receptor in
binding assay, the potency of antagonists at hA2B receptor
(expressed on CHO cells) was determined in adenylyl cyclase
experiments instead. The procedure was carried out as described
previously in Klotz et al. with minor modifications [16]. In this
The synthetic strategy used in this work is depicted in Scheme
1. Briefly the synthesis of the chromone carboxamide derivatives
was based on a one-pot condensation with the activation in situ of
the carboxylic acid function using a coupling reagent under mild
reaction conditions. The coupling reagents selected for carboxylic
Table 1
Affinity (Ki, nM) of chromones 1–13 in radioligand binding assays at human A1, A2A and A3 adenosine receptors.
.
Compound
R
hA1
hA2A
hA3
Selectivity
Ki (nM)
Ki (nM)
Ki (nM)
hA1/hA3
hA2A/hA3
1
2
3
4
OH
>100,000
>100,000
>100,000
>100,000
>100,000
>100,000
>100,000
>100,000
>100,000
–
–
NH–Ph
NH–C6H11
NH–C3H7
14,200 (11,800–17,100)
38,700 (27,400–54,700)
>100,000
>7.0
>2.6
–
>7.0
>2.6
–
5
6
7
8
9
NH–(40–OH–Ph)
>100,000
>100,000
>100,000
>100,000
>100,000
28,300 (19,600–40,700)
>100,000
46,300 (38,100–56,300)
9580 (7600–12,100)
>2.2
>10
0.61
>10
NH–(40–OCH3–Ph)
NH–(40–CH3–Ph)
NH–(30–OH–40–OCH3–Ph)
NH–(30–OCH3–40–OCH3–Ph)
>100,000
15,800 (12,200–20,400)
15,400 (10,100–23,400)
27,900 (18,300–42,700)
>6.3
>6.5
>3.6
>6.3
2.3
35,700 (32,700–39,100)
>100,000
>3.6
10
11
12
13
NH–(40–Cl–Ph)
>100,000
>100,000
>100,000
>100,000
>100,000
>100,000
>100,000
>100,000
>100,000
>100,000
>100,000
>100,000
–
–
–
–
–
–
–
–
NH–(40–OCF3–Ph)
NH–(40–CF3–Ph)
NH–(40–NO2–Ph)

26
A. Gaspar et al. / Biochemical Pharmacology 84 (2012) 21–29
Table 2
Affinity (Ki, nM) of chromones 14–26 in radioligand binding assays at human A1, A2A and A3 adenosine receptors.
.
Compound
R
hA1
hA2A
hA3
Selectivity
Ki (nM)
Ki (nM)
Ki (nM)
hA1/hA3
hA2A/hA3
14
15
16
17
OH
>100,000
>100,000
>100,000
–
–
NH–Ph
NH–C6H11
NH–C3H7
>100,000
>100,000
>100,000
–
–
18,400 (16,600–20,400)
19,300 (18,400–20,200)
26,900 (18,300–39,600)
41,600 (32,400–53,500)
71,300 (66,000–77,100)
40,800 (37,000–44,900)
0.26
0.47
0.38
1.0
18
19
20
21
22
NH–(40–OH–Ph)
10,400 (8870–12,300)
18,600 (14,500–23,800)
>100,000
22,100 (20,100–24,400)
10,100 (8660–11,900)
>100,000
8,860 (7460–10,500)
6070 (4710–7830)
16,600 (15,600–17,600)
3680 (2770–4900)
6690 (5610–7980)
1.2
3.1
2.5
1.7
NH–(40–OCH3–Ph)
NH–(40–CH3–Ph)
NH–(30–OH–40–OCH3–Ph)
NH–(30–OCH3–40–OCH3–Ph)
>6.0
2.3
>6.0
1.9
8590 (7240–10,200)
11,700 (10,300–13,300)
6850 (6220–7550)
12,400 (10,200–15,000)
1.8
1.9
23
24
25
26
NH–(40–Cl–Ph)
25,600 (10,400–32,200)
>100,000
17,400 (13,700–22,200)
>100,000
16,400 (15,700–17,100)
>100000
1.6
–
1.1
–
NH–(40–OCF3–Ph)
NH–(40–CF3–Ph)
NH–(40–NO2–Ph)
>100,000
>100,000
>100000
–
–
>100,000
14,300 (10,500–19,500)
3750 (3530–3980)
>26
3.8
assay, the NECA-stimulated adenylyl cyclase activity was inhibited
with increasing concentrations of antagonist. As a result, cAMP
(cyclic adenosine monophosphate) production was inhibited in a
concentration-dependent fashion, and IC50 and Ki values, respec-
tively, were determined.
The binding assays data clearly show that chromone carboxylic
acids (compounds 1 and 14) are devoid of activity and the
corresponding carboxamides derivatives are compounds with
diverse affinity and selectivity toward human A1, A2A and A3 AR
(Tables 1 and 2).
A more cautious look on the data allows to conclude that the
affinity and/or selectivity of chromone carboxamides is influenced
by the relative position of the carbonyl function of the
carboxamide group on the benzopyran nucleus and by the type
of amine (aromatic, cyclic and aliphatic) that is located in the
nitrogen atom. In general, it can be also inferred that the presence
of electron donators or withdrawing groups on the phenyl
substituent of the carboxamide also modulate the affinity and
selectivity of chromone carboxamides for the subtypes of ARs.
From the data the following specific conclusions can be drawn:
3.3. Molecular modeling
The recently published crystal structure of the human A2A
adenosine receptor, in complex with the antagonist ZM241385
(PDB code: 3EML) [24] provides a new template to perform
homology modeling of other GPCRs and in particular of adenosine
receptors. Therefore we built up a homology model of the hA3
receptor based on the crystal structure of the hA2A receptor
(methodological details were summarized in Section 2.4.1) [22,23].
In the process of selecting a reliable docking protocol to be
employed in the following docking studies of these new
derivatives, we have evaluated the ability of different docking
softwares in reproducing the crystallographic pose of ZM241385
inside the binding cavity of human A2A receptor. As reported in
Section 2.4.2, among the four different types of programs used to
calibrate our docking protocol, the Gold programs was finally
chosen since it showed the best performance with regard to the
calculated RMSD values relative to the crystallographic pose of
ZM241385 [22].
a) For carboxamides located in position 2 of the pyran nucleus
(compounds 2–13): the presence of a phenyl (compound 2) or a
cyclohexyl substituent (compound 3) is tolerated and a fair
affinity for A3 AR is observed. However, no binding was detected
with a linear alkyl substituent (compound 4). These observa-
tions allow to infer that probably in this type of systems the
spatial volume of the substituent is an important feature.The
presence of electron withdrawing substituents in the phenyl
ring located in the nitrogen atom, such as chlorine (compound
10), trifluormethoxy (compound 11), trifluormethyl (compound
12) or nitro (compound 13) groups in para position give rise to a
lack of affinity for all the adenosine receptors subtypes.
Consequently, based on the selected docking protocol, we
performed docking simulations to identify the hypothetical
binding mode of the newly synthesized derivatives inside the
human A2A and A3 adenosine receptors.
On other hand the presence of electron donators in the
phenyl ring located in the nitrogen atom, in para position,
(compounds 5–9) result in the display of a noticeable affinity/
selectivity for the A3 AR.
3.4. Structure–affinity relationship studies
b) For carboxamides located in position 3 of the pyran nucleus
(compounds 15–26): in this chromone carboxamide isomeric
series the conclusions were not as straightforward as in the
series with a 2-substituted pyran. However, for this type of
benzopyran derivatives the results show that similar to the 2-
substituted isomers the presence of electron donating groups in
the phenyl substituent increase the affinity and selectivity for
the human A3 AR (compounds 18–22). It is to note that
compounds 18 and 19 exhibit micromolar affinity for all the
studied receptors but without a significant selectivity. With
Chromone carboxamide derivatives (2–13 and 15–26) of the
chromone carboxylic acids (compounds 1 and 14) were obtained in
a one pot reaction and with good yields (45–85%) (Scheme 1). Their
affinity to bind to human A1, A2A and A3 adenosine receptors (AR)
expressed in CHO cells (compounds 2–13 in Table
1 and
compounds 15–26 in Table 2) was determined. Due to the lack
of a useful radioligand for hA2B AR, the inhibition of NECA-
stimulated adenylyl cyclase activity by chromone carboxylic acids
and carboxamides was determined. For all the tested compounds
no measurable activity (Ki > 100,000 nM) was detected.

A. Gaspar et al. / Biochemical Pharmacology 84 (2012) 21–29
27
exception of compound 26 with a nitro function in para position,
that have high A3 AR affinity, the chromone derivatives with
electron withdrawing groups have no measurable affinity for
adenosine receptors (compound 24 and 25). Moreover, it is
important to highlight that compounds 21 and 26 present the
possesses the highest affinity at both receptors (Ki hA2A = 6850 nM,
Ki hA3 = 3684 nM). At both receptor subtypes ligand-recognition
occurred in the upper region of the TM bundle, and the chromone
nucleus was surrounded by TMs 3, 5, 6, 7.
The hypothetical binding pose of compound 21 at the hA2A AR
(Fig. 1, panel A) showed an H-bonding interaction with Asn253
(6.55) and a stabilizing interaction with Phe168 (EL2). Interest-
ingly, the important role in ligand binding of these two residues
was previously revealed by site-directed mutagenesis studies
[33,34] and by the crystallographic binding pose of ZM241385
inside the hA2A AR binding pocket [24].
Moreover, the hydroxyl group on the phenyl ring of compound
21 interacted with Tyr271 (7.36), forming a weak H-bond. Finally,
the ligand also formed hydrophobic interactions with some
residues of the binding site, including Leu85 (3.33), Trp246
(6.48), Leu249 (6.51), Tyr271 (7.36) and Ile274 (7.39).
On the other hand, the docking pose of compound 21 at the hA3
AR was located in the same region of the TM bundle as at the hA2A
AR, but the orientation of the ligand was different (Fig. 1, Panel B).
In this case, the ligand formed two H-bonds with Asn250 (6.55)
and a stabilizing interaction with Phe168 (EL2). Moreover, the
complex is stabilized by hydrophobic interactions occurring
between the ligand and some residues of the binding site, such
as Leu91 (3.33), Trp243 (6.48), Leu246 (6.51) and Leu264 (7.35).
best affinity for A3 receptor with
a Ki = 3684 nM and
Ki = 3750 nM, respectively.
A molecular modeling investigation was performed for all the
newly synthesized analogues, in order to identify their hypotheti-
cal binding modes at both the crystallographic structure of hA2A AR
and the hA3 AR model. The analysis was extended to docking
simulations and per residue electrostatic and hydrophobic con-
tributions.
The first important consideration is that almost all the new
analogues showed different possible binding poses at both the
hA2A AR and the hA3 AR. In fact, even if all ligands made contacts
mainly with residues belonging toTM2, TM3, TM6, TM7, and EL2,
they can accommodate different orientations inside the binding
pockets.
Fig. 1 shows the selected binding modes of compound 21, a
chromone with the carboxamide located in position 3 of the pyran
nucleus, obtained after docking simulations at the hA2A AR and the
hA3 AR. Among all the herein reported derivatives, compound 21
Fig. 1. Hypothetical binding modes of compound 21 obtained after docking simulations: (A) inside the hA2A AR binding site; (B) inside the hA3 AR binding site. Poses are
viewed from the membrane side facing TM6, TM7, and TM1. In panel A, the view of TM7 is omitted. Side chains of some amino acids important for ligand recognition and H-
bonding interactions are highlighted. Hydrogen atoms are not displayed.
Fig. 2. Calculated electrostatic interaction energy (in kcal/mol) between the ligand and each single amino acid involved in ligand recognition observed from the hypothetical
binding modes of compound 21 inside (A) hA2A AR and (B) hA3 AR binding sites.

28
A. Gaspar et al. / Biochemical Pharmacology 84 (2012) 21–29
Fig. 3. Calculated hydrophobic interaction scores (in arbitrary hydrophobic units) between the ligand and each single amino acid involved in ligand recognition observed from
the hypothetical binding modes of compound 21 inside (A) hA2A AR and (B) hA3 AR binding sites.
Another antagonist of this series with affinity at the hA3 AR
comparable to compound 21 is compound 26 (Ki hA3 = 3750 nM).
The binding orientation, obtained after docking simulation, of this
ligand at this receptor is very similar to the one observed for
compound 21 (data not shown). In this case, compound 26 formed
one H-bonding interaction with Asn250 (6.55) and a stabilizing
interaction with Phe168 (EL2).
that the type of substituent on the aromatic ring of the chromone
amide side chain is crucial for the optimization of affinity and
selectivity. Receptor-driven molecular modeling studies provide
information on the binding/selectivity data of the chromone. The
data so far acquired are instrumental for future optimization of
chromone carboxamide lead as a selective A3AR antagonist.
Analyzing the calculated electrostatic contribution per residue
to the whole interaction energy for both the complexes between
compound 21 and these two adenosine receptor subtypes (Fig. 2),
the main stabilizing factor was found to be related to Asn 6.55
(Asn253 in hA2A AR and Asn250 in hA3 AR), due to the H-bonding
interactions with the ligand above described. However, the
calculated electrostatic contribution of this Asn residue is much
more stabilizing for the hA3 AR complex (ꢂ16 kcal/mol for hA3 AR
complex compared to ꢂ4 kcal/mol for hA2A AR complex).
As shown in Fig. 3, the hydrophobic interaction scores patterns
showed the strongest stabilizing contribution corresponding to the
interactions of the ligand with Phe168 (EL2) at both receptors.
In conclusion, at the hA2A AR compound 21 was able to interact
with Asn253 (6.55) and Phe168 (EL2), important residues in ligand
recognition, but it did not form strong interaction with Glu169
(EL2), another residue with an important role in ligand binding as
revealed by the crystallographic binding pose of ZM241385 inside
the hA2A AR binding pocket [24]. At the hA3 AR compound 21
formed stronger, but still not optimal, interactions with the
residues of the binging site, and in particular with Asn250 (6.55).
This difference could be the reason for the higher affinity of this
compound for this adenosine receptor subtype compared to the
hA2A subtype.
Acknowledgments
This work was supported by the Foundation for Science and
Technology (FCT), Portugal (project PTDC/QUI/70359/2006 and
PTDC/QUI-QUI/113687/2009).
A. Gaspar (SFRH/BD/43531/2008) and FB (SFRH/BSAB/1090/
2010) thanks FCT grants. The work coordinated by S.M. was carried
out with financial support from the University of Padova, Italy, and
the Italian Ministry for University and Research (MIUR), Rome,
Italy. S.M. is very grateful to Chemical Computing Group Inc.
(Montreal, Quebec, Canada) for the scientific and technical
partnership.
References
[1] Erik S. Mechanisms of cancer cell invasion. Curr Opin Genetics Dev 2005;15:
87–96.
[2] Aggarwal BB, Sethi G, Baladandayuthapani V, Krishnan S, Shishodia S. Target-
ing cell signaling pathways for drug discovery: an old lock needs a new key. J
Cell Biochem 2007;102:580–92.
[3] Gessi S, Merighi S, Sacchetto V, Simioni C, Borea PA. Adenosine receptors and
cancer. BBA – Biomembr 2011;1808:1400–12.
[4] Fishman P, Bar-Yehuda S, Varani K, Gessi S, Merighi S, Borea PA. Agonists and
antagonists: molecular mechanisms and therapeutic applications. In: Borea
PE, editor. A3 adenosine receptors from cell biology to pharmacology and
therapeutics. Dordrecht Heidelberg London New York: Springer; 2010.
[5] Gessi S, Merighi S, Varani K, Leung E, Mac Lennan S, Borea PA. The A3
adenosine receptor: an enigmatic player in cell biology. Pharmacol Ther
2008;117:123–40.
[6] Fishman P, Jacobson KA, Ochaion A, Cohen S, Bar-Yehuda S. The anti-cancer
effect of A3 adenosine receptor agonists: a novel, targeted therapy. Immunol
Endocr Metabol Agents Med Chem 2007;7:298–303.
[7] Hasko G, Linden J, Cronstein B, Pacher P. Adenosine receptors: therapeutic
aspects for inflammatory and immune diseases. Nat Rev Drug Discov 2008;7:
759–70.
[8] Baraldi PG, Romagnoli R, Saponaro G, Baraldi S, Tabrizi MADP. A3 adenosine
receptor antagonists: history and future perspectives. In: Borea PA, editor. A3
adenosine receptors from cell biology to pharmacology and therapeutics.
Dordrecht Heidelberg London New York: Springer; 2010.
On the whole, the proposed binding modes reflect the ability of
compound 21 to bind both hA2A and hA3 adenosine receptor
subtypes with Ki in the low micromolar range and without a very
good selectivity profile.
4. Conclusions
Evidence was acquired to demonstrate that chromone is a valid
scaffold for the design of novel adenosine receptor ligands. The
easy synthetic accessibility and the decoration capability of
chromones make them ‘‘privileged’’ scaffolds.
[9] DeNinno MP, Masamune H, Chenard LK, DiRico KJ, Eller C, Etienne JB, et al. The
synthesis of highly potent, selective, and water-soluble agonists at the human
adenosine A3 receptor. Bioorg Med Chem Lett 2006;16:2525–7.
[10] Baraldi PG, Tabrizi MA, Gessi S, Borea PA. Adenosine receptor antagonists:
translating medicinal chemistry and pharmacology into clinical utility. Chem
Rev 2008;108:238–63.
[11] Jacobson KA. Introduction to adenosine receptors as therapeutic targets. In:
Wilson CN, Mustafa SJ, editors. Adenosine receptors in health and disease:.
Berlin Heidelberg: Springer-Verlag; 2009.
From this study and on the basis of the obtained structure–
activity relationships it was concluded that chromone carbox-
amides represent a novel class of AR ligands. The most remarkable
chromones were compounds 21 and 26 that present a better
affinity for A3 AR (Ki = 3680 nM and Ki = 3750 nM, respectively).
The results obtained so far pointed out a crucial and undisclosed
role of the presence of an amide substituent of the pyrone ring and

A. Gaspar et al. / Biochemical Pharmacology 84 (2012) 21–29
29
[12] Lars JSK. Drug discovery management, small is still beautiful: why a number of
companies get it wrong. Drug Discov Today 2011;16:476–84.
[23] Morizzo E, Federico S, Spalluto G, Human Moro S. A3 adenosine receptor as
versatile G protein-coupled receptor example to validate the receptor homol-
ogy modeling technology. Curr Pharm Des 2009;15:4069–84.
[24] Jaakola V-P, Griffith MT, Hanson MA, Cherezov V, Chien EYT, Lane JR, et al. The
2.6 Angstrom crystal structure of a human A2A adenosine receptor bound to an
antagonist. Science 2008;322:1211–7.
[25] Ballesteros JA, Weinstein H. Integrated methods for the construction of three-
dimensional models and computational probing of structure-function rela-
tions in G protein-coupled receptors. In: Stuart CS, editor. Methods in
neurosciences. Academic Press; 1995. p. 366–428.
[26] LabuteP. Protonate 3D: assignment of ionizationstates and hydrogencoordinates
to macromolecular structures. Proteins Struct Funct Bioinf 2009;75:187–205.
[27] GOLD suite, version 4.0.1; software available from Cambridge Crystallographic
Data Centre Cambridge Crystallographic Data Centre (12 Union Road Cam-
bridge CB2 1EZ UK); http://www.ccdc.cam.ac.uk.
[28] Friesner RA, Banks JL, Murphy RB, Halgren TA, Klicic JJ, Mainz DT, et al. Glide: A
new approach for rapid, accurate docking and scoring. 1. Method and assess-
ment of docking accuracy. J Med Chem 2004;47:1739–49.
[29] Korb O, Stutzle T, Exner TE. Empirical scoring functions for advanced protein-
ligand docking with PLANTS. J Chem Inf Model 2009;49:84–96.
[30] Allen CL, Williams JMJ. Metal-catalysed approaches to amide bond formation.
Chem Soc Rev 2011;40:3405–15.
[13] Borges F, Gaspar A, Garrido J, Milhazes N, Batoreu MC. Chromone derivatives
for use as antioxidants/preservatives. PT 103665, WO 2008/104925.
[14] Gaspar A, Silva T, Ya´n˜ez M, Vina D, Orallo F, Ortuso F, et al. Chromone, a
privileged scaffold for the development of monoamine oxidase inhibitors. J
Med Chem 2011;54:5165–73.
[15] Gaspar A, Teixeira F, Uriarte E, Milhazes N, Melo A, Cordeiro MNDS, et al.
Towards the discovery of a novel class of monoamine oxidase inhibitors:
structure–property–activity and docking studies on chromone amides. Chem-
MedChem 2011;6:628–32.
[16] Klotz KN, Hessling J, Hegler J, Owman C, Kull B, Fredholm BB, et al. Comparative
pharmacology of human adenosine receptor subtypes–characterization of sta-
bly transfected receptors in CHO cells. N-S Arch Pharmacol 1997;357: 1–9.
[17] Klotz KN, Falgner N, Kachler S, Lambertucci C, Vittori S, Volpini R, et al.
[3H]HEMADO–a novel tritiated agonist selective for the human adenosine
A3 receptor. Eur J Pharmacol 2007;556(1–3):14–8.
[18] Cheng Y-C, Prusoff WH. Relationship between the inhibition constant (KI) and
the concentration of inhibitor which causes 50 per cent inhibition (IC50) of an
enzymatic reaction. Biochem Pharmacol 1973;22:3099–108.
[19] De Lean A, Hancock AA, Lefkowitz RJ. Validation, statistical analysis of a
computer modeling method for quantitative analysis of radioligand binding
data for mixtures of pharmacological receptor subtypes. J Mol Pharmacol
1982;21:5–16.
[31] Han S-Y, Kim Y-A. Recent development of peptide coupling reagents in organic
synthesis. Tetrahedron 2004;60:2447–67.
[20] MOE. (Molecular Operating Environment), version 2008.10; software avail-
able from Chemical Computing Group Inc. (1010 Sherbrooke Street West, Suite
910, Montreal, Quebec, Canada H3A 2R7); http://www.chemcomp.com.
[21] Stewart JJP. MOPAC Version 7. Tokyo, Japan: Fujitsu Limited; 1993.
[22] Lenzi O, Colotta V, Catarzi D, Varano F, Poli D, Filacchioni G, et al. 2-Phenyl-
pyrazolo[4,3-d]pyrimidin-7-one as a new scaffold to obtain potent and selec-
tive human A3 adenosine receptor antagonists: new insights into the receptor-
antagonist recognition. J Med Chem 2009;52:7640–52.
[32] Ellis GP. General methods of preparing chromones. In: Chemistry of hetero-
cyclic compounds. John Wiley & Sons, Inc.; 2008. pp. 495–555.
[33] Kim J, Wess J, van Rhee AM, Scho¨neberg T, Jacobson KA. Site-directed muta-
genesis identifies residues involved in ligand recognition in the human A
adenosine receptor. J Biol Chem 1995;270:13987–97.
[34] Jaakola V-P, Lane JR, Lin JY, Katritch V, Ijzerman AP, Stevens RC. Ligand binding
and subtype selectivity of the human A2A adenosine receptor. J Biol Chem
2010;285:13032–44.