Novel potent and highly selective human A3 adenosine receptor antagonists belonging to the 4-amido-2-arylpyrazolo[3,4-c]quinoline series: Molecular docking analysis and pharmacological studies

Article abstract of DOI:10.1016/j.bmc.2008.10.018

The study of novel 2-arylpyrazolo[3,4-c]quinolin-4-(hetero)arylamides, designed as human (h) A₃ adenosine receptor antagonists, is reported. The new derivatives are endowed with nanomolar hA₃ receptor affinity and high selectivity versus hA₁, hA_2A and hA_2B receptors. Among the (hetero)aroyl residues introduced on the 4-amino group, the 2-furyl and 4-pyridyl rings turned out to be the most beneficial for hA₃ affinity (K_i = 3.4 and 5.0 nM, respectively). An intensive molecular docking study to a rhodopsin-based homology model of the hA₃ receptor was carried out to obtain a 'structure-based pharmacophore model' that proved to be helpful for the interpretation of the observed affinities of the new hA₃ pyrazoloquinoline antagonists.

Full text of DOI:10.1016/j.bmc.2008.10.018

Bioorganic & Medicinal Chemistry 17 (2009) 401–410
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Novel potent and highly selective human A3 adenosine receptor antagonists
belonging to the 4-amido-2-arylpyrazolo[3,4-c]quinoline series:
Molecular docking analysis and pharmacological studies
a
a
a
a
Vittoria Colotta a, , Francesca Capelli , Ombretta Lenzi , Daniela Catarzi , Flavia Varano ,
Daniela Poli a, Fabrizio Vincenzi b, Katia Varani b, Pier Andrea Borea b, Diego Dal Ben c,
Rosaria Volpini c, Gloria Cristalli c, Guido Filacchioni a
*
a Dipartimento di Scienze Farmaceutiche, Laboratorio di Progettazione, Sintesi e Studio di Eterocicli Biologicamente Attivi, Università di Firenze, Polo Scientifico, Via Ugo Schiff,
6, 50019 Sesto Fiorentino, FI, Italy
b Dipartimento di Medicina Clinica e Sperimentale, Sezione di Farmacologia, Università di Ferrara, via Fossato di Mortara 17-19, 44100 Ferrara, Italy
c Dipartimento di Scienze Chimiche, Università di Camerino, via S. Agostino 1, 62032 Camerino, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
The study of novel 2-arylpyrazolo[3,4-c]quinolin-4-(hetero)arylamides, designed as human (h) A3 aden-
osine receptor antagonists, is reported. The new derivatives are endowed with nanomolar hA3 receptor
affinity and high selectivity versus hA1, hA2A and hA2B receptors. Among the (hetero)aroyl residues intro-
duced on the 4-amino group, the 2-furyl and 4-pyridyl rings turned out to be the most beneficial for hA3
affinity (Ki = 3.4 and 5.0 nM, respectively). An intensive molecular docking study to a rhodopsin-based
homology model of the hA3 receptor was carried out to obtain a ‘structure-based pharmacophore model’
that proved to be helpful for the interpretation of the observed affinities of the new hA3 pyrazoloquino-
line antagonists.
Received 29 July 2008
Revised 7 October 2008
Accepted 8 October 2008
Available online 11 October 2008
Keywords:
G protein-coupled receptors
Adenosine receptor antagonists
Pyrazoloquinolines
Ó 2008 Elsevier Ltd. All rights reserved.
Tricyclic heteroaromatic systems
Ligand–receptor modeling studies
1. Introduction
investigated as neuroprotective agents.12–14 In our laboratory, much
effort has been directed toward the study of hA3 AR antagonists
Adenosine is a neuromodulator which exerts a host of physio-
pathological effects through activation of four receptor subtypes,
belonging to the superfamily of G protein-coupled receptors and
termed A1, A2A A2b and A3 adenosine receptors (ARs).1,2 The A3 recep-
tor subtype has a widespread distribution, both in the brain and
peripheral systems,1,2 and couples to different second messenger
pathways such as inhibition of adenylyl cyclase,2 stimulation of
phospholipase C3 and D4 and activation of KATPchannels.5 A3 AR, like
the other ARs, also couples to members of the mitogen-activated
protein kinase (MAPK) family, such as the extracellular signal-regu-
lated kinase (ERK)1/2 and p38,6 and this coupling gives this receptor
a role in cell growth, survival, death and differentiation.6,7 Interest-
ingly, A3 AR activation may afford both cell protection and cell death,
being strictly dependent on the degree of receptor activation and/or
the cell type.7,8 For A3 AR antagonists different potential therapeutic
applications have been envisaged. They could be useful tools for the
treatment of glioblastoma multiforme,9 as renal protective drugs10
and in the treatment of glaucoma.11 Recently, they have also been
belonging to different classes of heteroaromatic derivatives,14–20
and one of the most recently investigated is represented by the
4-amido-2-arylpyrazolo[3,4-c]quinoline series14,16 (Chart 1). Struc-
ture–activity relationship studies showed that the presence of acyl
substituents on the 4-amino group positively affects both A3 AR
affinity and selectivity, one of the most advantageous residues being
a benzoyl moiety (Chart 1, compound 1). Introduction of small
groups (R = 3- or 4-Me, 3- or 4-OMe) on the 2-phenyl ring of the 4-
amido derivatives, although it maintained the hA3 affinity in the
nanomolar range, reduced the selectivity toward this receptor. On
the basis of these findings, and taking 1 as lead compound, we syn-
thesized the new derivatives 2–23 (Chart 1), most of which lack sub-
stituents on the 2-phenyl ring (2–13, 18–23) while bearing various
heteroaroyl (2–15) or aroyl (18–21) moieties on the 4-amino group.
We also prepared the 4-diaroylamino derivatives 22–23 because
previous data showed that introduction of a second benzoyl residue
on the 4-amino group of compound 1 maintained both high hA3
affinity and selectivity.14
A molecular docking analysis of the new derivatives 2–23 and of
the lead 1 was performed using a homology model of human A3
receptor based on the bovine rhodopsin crystal structure as a
* Corresponding author. Tel.: +39 55 4573731; fax +39 55 4573780.
E-mail address: vittoria.colotta@unifi.it (V. Colotta).
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.10.018

402
V. Colotta et al. / Bioorg. Med. Chem. 17 (2009) 401–410
O
O
O
R4
N
N
Ph
N
N
Het
HN
HN
HN
N
N
N
N
N
R
R
R
NH2
a
b
2-21
1
R= H
2-13 R= H
N
N
14-15 R= Cl
N
R
O
O
N
R4
N
24-27
R4
O
O
N
O
R1
R1
R1
N
HN
N
N
R
N
N
R
N
N
H
24
25
26
27
N
4-Cl
3-CF3
4-CF3
22-23
16, 17 R1= H R= 3-CF3, 4-CF3
22 R1= 3-CF3
23 R1= 4-CF3
18, 19 R= H
20, 21 R= H
R1= 3-CF3, 4-CF3
R1= 3-CH3, 4-CH3
R4
R
R
H
R4
2-furyl
3-furyl
2-pyrazinyl
4-pyridyl
2
13
H
Cl
Cl
3
14
15
16
17
18
19
20
21
22
23
H
Chart 1. Previously and herein reported 2-arylpyrazolo[3,4-c]quinolin-4-amido
2-(5-methylfuryl) H
2-furyl
4
derivatives.
2-thienyl
3-thienyl
4-thiazolyl
2-pyridyl
3-pyridyl
4-pyridyl
2-pyrimidyl
4-pyrimidyl
H
H
C6H5
3-CF3
5
C6H5
4-CF3
6
C6H4-3-CF3
C6H4-4-CF3
C6H4-3-CH3
C6H4-4-CH3
C6H4-3-CF3
C6H4-4-CF3
H
H
H
H
7
H
H
H
H
H
H
template. Docking results were compared to biological data with the
aim of defining the SAR profile of the compounds and obtaining use-
ful information for the rational design of new hA3 AR antagonists.
8
9
10
11
12
2. Chemistry
The target 4-amido derivatives 2–23 were obtained starting
from the 2-arylpyrazolo[3,4-c]quinolin-4-amines 24,14,16 25,16 26
and 27 which were prepared following the pathway depicted in
Scheme 1. Briefly, a mixture of 3-ethoxalylindole21 with the suit-
able arylhydrazine hydrochloride in absolute ethanol and a few
drops of glacial acetic acid was microwave irradiated at 140 °C
for 3 min to afford the 2-arylpyrazolo[3,4-c]quinolin-4-ones
28,14,16 29,16 30 and 31. Compounds 28–31were reacted with a
mixture of phosphorus pentachloride/phosphorus oxychloride to
give the corresponding 4-chloro derivatives 32,14,16 33,16 34 and
35 which were transformed into the desired 4-amino derivatives
24–27 with ammonia. The target 2-arylpyrazoloquinolin-4-amides
2–21(Scheme 2) were obtained by reacting 24–27 with the suitable
Scheme 2. Reagents: (a) Suitable carboxylic acid, 1-hydroxybenzotriazole, NEt3,
4-(dimethylamino)pyridine, 1-(3-(dimethylamino)propyl)-3-ethylcarbodiimide hydro-
chloride, DMF; (b) ArCOCl, anhydrous pyridine, methylene chloride.
carboxylic acids in dimethylformamide and in the presence of 1-
hydroxybenzotriazole, triethylamine, and 4-(dimethylamino)pyri-
dine. Most of the used carboxylic acids were commercially
available, except for 5-methylfuran-2-carboxylic acid,22 2-pyrim-
idyl-carboxylic acid23 and 4-pyrimidyl-carboxylic acid,24 which
were synthesized following the reported procedures. Finally, the
4-diaroylamino derivatives 22 and 23 were obtained by refluxing
the 4-amino derivative 24 with an excess of 3-trifluoromethylben-
zoyl chloride and 4-trifluoromethylbenzoyl chloride, respectively,
in anhydrous methylene chloride and pyridine.
O
O
OEt
O
3. Pharmacology
N
HN
N
a
R
The 4-amido derivatives 2–23, and their newly synthesized
precursors, i.e., the 4-amino-2-arylpyrazoloquinolines 26, 27 and
the corresponding 4-oxo compounds 30 and 31 were tested for their
ability to displace [3H]8-cyclopentyl-1, 3-dipropylxanthine
(DPCPX), [3H]4-(2-[7-amino-2-(2-furyl)[1, 2, 4]triazolo[2, 3-a]
[1,3,5]triazin-5-yl-amino]ethyl)phenol (ZM 241385) and [125I]N6-
(4-amino-3-iodobenzyl)-50-(N-methylcarbamoyl) adenosine (AB-
MECA), respectively, from cloned hA1, hA2A and hA3 ARs stably
expressed in CHO cells. Compounds 2–23, 26, 27, 30, 31 were also
tested at the hA2B subtype by measuring their inhibitory effects on
NECA-stimulated cAMP levels in CHO cells stably transfected with
the hA2B AR. The antagonistic potencies of the selected derivatives
2, 4, 8–10, 14, 15, 20 and 21 were assessed by evaluating their effect
on 2-chloro-N6-(3-iodobenzyl)-50-(N-methylcarbamoyl) adenosine
(Cl-IB-MECA)-inhibited cAMP production in CHO cells, stably
expressing hA3 ARs. Finally, compounds 2, 4, 8–10, 14, 15, 20 and
21 were also tested to evaluate their affinities at the rat A3 AR, stably
expressed in HEK 293 cells. All pharmacological data are collected in
N
H
28-31
R1
N
N
N
R
R1= Cl 32-35
R1= NH2 24-27
R
c
H
24, 28, 32
25, 29, 33
26, 30, 34
27, 31, 35
4-Cl
3-CF3
4-CF3
Scheme 1. Reagents and conditions: (a) arylhydrazine hydrochloride, absolute
EtOH, AcOH, microwave irradiation; (b) PCl5/POCl3; (c) NH3(g), absolute EtOH.

V. Colotta et al. / Bioorg. Med. Chem. 17 (2009) 401–410
403
Table 1
Binding affinity at hA1, hA2A and hA3 ARs and Potencies at hA2B and hA3
O
O
O
O
HN
R4
N
N
NH2
R4
N
R4
N
N
N
N
HN
N
N
N
N
N
R
R
R
R
26, 27
Binding experiments Ki (nM) or I%
30, 31
1, 2-21, 36, 37
22, 23, 38
cAMP assays IC50 (nM) or I%
hA2Ad
hA2B
hA3
a
b
c
e
f
R4
R
hA3
rA3
hA1
1g
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
26
27
30
31
36g
37g
38g
C6H5
2-Furyl
3-Furyl
H
H
H
H
H
H
H
H
H
H
H
H
2
0.1
0%
34%
9%
500 53
45%
1%
12%
3%
26%
8%
6%
9%
6%
0%
20%
2%
10%
1%
2%
1%
9%
1%
1%
2%
6%
6%
1%
2%
1%
1%
1%
3%
1%
1%
3%
1%
7%
1%
5%
Nth
3%
5%
5%
6%
2%
2%
15%
1%
1%
1%
3%
2%
2%
3%
3%
3%
2%
Nth
21
3.4 0.3
50
7.5 0.8
50
40
60
15.5 1.3
10.5 1.7
5.0 0.6
945 82
210 18
65
11.5 1.5
24.7 2.9
52
140 15
60
415 37
6.3 0.7
11.5 1.2
24%
29%
30%
3
5
6
2-(5-Methylfuryl)
2-Thienyl
3-Thienyl
4-Tiazolyl
2-Tyridyl
3-Pyridyl
4-Pyridyl
2-Pyrimidyl
4-Pyrimidyl
2-Pyrazinyl
2-Furyl
4-Pyridyl
C6H5
C6H5
C6H4-3-CF3
C6H4-4-CF3
C6H4-3-CH3
C6H4-4-CH3
C6H4-3-CF3
C6H4-4-CF3
H
49
6
5
5
23%
27%
26%
107
76
41
9
5
4
H
6
4-Cl
4-Cl
3-CF3
4-CF3
H
H
H
H
H
11%
21%
17%
4%
83
145 12
7
5
13%
10%
23%
10%
20%
6%
4%
17%
6%
1%
2%
6%
57 4.2
629 51
0%
8
4%
5%
7%
35%
38%
6%
1%
4%
26
54
3
5
H
6%
17%
11%
4%
480 50
4.3 0.5
4.4 0.2
6.1 0.5
3-CF3
4-CF3
3-CF3
4-CF3
3-CH3
4-CH3
H
H
11%
3%
10%
2860 224
26%
0%
5%
C6H5
C6H5
C6H5
Nth
Nth
Nth
Displacement of specific [125I]AB-MECA binding to hA3CHO cells. Ki values are means SEM of four separate assays each performed in duplicate.
a
b
c
Percentage of inhibition in [125I]AB-MECA competition binding assays to rA3 HEK cells by using 1
Percentage of inhibition in [3H]DPCPX competition binding assays to hA1CHO cells by using 1
M concentration of the tested compounds; the Ki value is the mean SEM
lM concentration of the tested compounds.
l
of four separate assays each performed in duplicate.
d
Percentage of inhibition in [3H]ZM241385 competition binding assays of the tested compounds (1
l
M) to hA2ACHO cells.
e
Percentage of inhibition of the examined compounds at the 1
lM concentration on cAMP experiments in hA2BCHO cells stimulated by 200 nM NEC A (NEC A
EC50 = 150 12 nM).
f
IC50 values of the examined compounds (1 nM–l lM) are expressed as means SEM of four separate cAMP experiments in hA3CHO cells which were inhibited by 100 nM
Cl-IB-MECA (Cl-IB-MECAEC50 = 10 l nM).
g
Ref. 14.
Not tested.
h
Table 1, together with the binding results of the previously reported
2-arylpyrazoloquinolin-4-amides 1, 36–3814 included as reference
antagonists.
receptor (compounds 2, 4, 8–10, 14, 15, 20 and 21) do not show
any appreciable affinity for this receptor.
Among the probed heterocyclic substituents, the 2-furyl and
2- or 3- or 4-pyridyl rings revealed to be the most beneficial, since
compounds 2, 8–10 display the highest hA3 affinity (Ki = 3.4–
15 nM) Introduction of a 5-methyl group in the furyl moiety of
derivative 2 afforded compound 4 which still possesses a high
hA3 binding activity (Ki = 7.5 nM). Replacement of the 2-furyl ring
of 2 with other pentacyclic substituents, such as 3-furyl, 2- or
3-thienyl and 4-thiazolyl (compounds 3, 5–7), causes some reduc-
tion of the hA3 affinity (Ki = 50–60 nM). When a second nitrogen
atom was introduced in the pyridyl moiety, a drop of hA3 affinity
was observed for the 2- and 4-pyrimidyl derivatives 11 and 12
(Ki = 945 and 210 nM, respectively), while for the 2-pyrazinyl
derivative 13 a good hA3 affinity (Ki = 65 nM) was obtained. As
stated above, most of the newly synthesized derivatives lack the
R substituent on the 2-phenyl group, since in a previous study14
we found that the presence of R = Me or OMe, either in the para-
4. Results and discussion
4.1. Structure–affinity relationships
The binding data, reported in Table 1, show that replacement of
the benzoyl moiety of the lead compound 1 with various (het-
ero)aroyl rings generally maintains high hA3 AR affinity. In fact,
several of the new derivatives (compounds 2–6, 8–10, 14, 15,
20, 21) show low Ki values (Ki < 50 nM). Interestingly, all the newly
synthesized 4-amido derivatives 2–23 possess complete hA3
selectivity, being totally unable to bind hA1, hA2 A and hA2B recep-
tors, with the only exception of compound 4 which shows some
hA1 affinity. As expected on the basis of previous data on 4-ami-
dopirazolo[3,4-c]quinolines,14 all derivatives tested at the rat A3

404
V. Colotta et al. / Bioorg. Med. Chem. 17 (2009) 401–410
or meta-position, while maintaining a high hA3 affinity, reduced the
hA3 versus hA1 selectivity, as can be noted by comparing the bind-
ing data of compound 1 with those of the previously reported
derivatives 3614 (R = 3-Me) and 3714 (R = 4-Me) (Table 1). Instead,
the presence of R = 4-Cl (compounds 14 and 15), although slightly
reducing the hA3 affinity, retains high hA3 selectivity (compare
compounds 2 and 10–14 and 15, respectively).
The rationale for the synthesis of compounds 16–21 ensued
from molecular modeling findings (see Section 4.2 for details).
Introduction of the CF3 substituent on the 2-phenyl ring, both in
meta- (16) and para- (17) positions (Ki = 52 and 140 nM, respec-
tively), reduced the hA3 AR affinity, either compared to that of
based on the bovine rhodopsin crystal structure as template, with
the aim of obtaining a possible rationalization of the different bind-
ing affinities for the hA3 receptor.
As the preliminary model of hA3 receptor did not contain a
suitable cavity acting as binding site, this region was defined with
the aid of published site directed mutagenesis data, which indi-
cated, in particular, that the residues His95, Ser247, Asn250, and
His272 are important for antagonist binding.25 Thus, a manual
docking section was performed on the lead compound 1 which
was inserted in different orientations in the defined binding cavity.
Consequently, an equivalent number of protein–ligand complexes
were obtained. Each complex was then energetically minimized,
and the potential energy of the different receptor conformations
(extracted from the respective receptor–ligand complexes) was
measured. The receptor conformation with the lowest potential
energy value was selected for the docking section. For each ligand,
docking studies allowed generation of a set of possible docking
poses which were energetically minimized and re-scored.
The binding cavity is inserted between TM3 (Met86, Thr87,
Leu90, Leu91, T94, His95), EL2 (Ser165, Gln167, Phe168, Met172),
TM5 (Val178, Phe182, Ile186) TM6 (Trp243, Leu246, Ser247,
Asn250, Tyr254) and TM7 (Ile268) and has a clear Y shape.
Docking results showed that each ligand is able to interact with
the receptor binding cavity by assuming different orientations
which can be grouped into three families (Fig. 1). In the first con-
formation family (family ‘A’), the pyrazoloquinoline scaffold is
positioned in a space between TM3, TM5 and TM6, while the sub-
stituents at the 2- and 4-positions are located between TM5 and
TM6 and between TM3 and TM7, respectively. The second confor-
mation family (family ‘B’) shows ligands oriented in a symmetric
way with respect to family ‘A’. In fact, the pyrazoloquinoline
nucleus is still located in a space between TM3, TM5 and TM6,
while the 2- and 4-substituents are positioned, respectively,
between TM3 and TM7 and between TM5 and TM6. The third con-
formation family (family ‘C’) is different as the pyrazoloquinoline
scaffold is located between TM5 and TM6, while the 2- and 4- sub-
stituents are positioned between TM3 and TM7 and between TM3,
TM5 and TM6, respectively.
the lead compound
1 (R = H) or to those of compounds
3614(R = 3-Me) and 3714 (R = 4-Me). A reduction of hA3 affinity
was also obtained when the CF3 group was introduced in the meta-
(18) or para-position (19) of the benzoylamino moiety. Indeed,
compounds 18 and especially 19 show lower hA3 affinities
(Ki = 60 and 415 nM, respectively) in comparison to the lead deriv-
ative 1 (Ki = 2 nM). In contrast, introduction of the Me substituent
on the benzoylamino group of the lead 1, both in meta- (20) and
para- (21) positions, did not change much the hA3 affinity
(Ki = 6.3 nM and 11.5 nM, respectively) which remains in the low
nanomolar range.
Derivatives 22 and 23, bearing, respectively, a meta- and para-CF3
group on the two benzoyl moieties, displayed null hA3 AR affinities.
These data, if compared to the high hA3 AR affinity of the 4-dib-
enzoylamino derivative 3814 (Table 1), may be ascribed to the elec-
tron-withdrawing properties and steric hindrance of the CF3
substituent, which might cause a change in electron density on the
molecule and a distortion of the binding mode at the receptor.
The hA3 affinity values of the 4-amino- and 4-oxo-substituted
2-arylpyrazoloquinolines 26, 27 and 30, 31, synthetic precursors
of the target derivatives 16 and 17, were quite unexpected. Indeed,
on the basis of the hA3 nanomolar affinity of many previously
reported 2-arylpyrazoloquinolin-4-ones and 4-amines,14 also these
new 2-aryl derivatives, bearing a 3-CF3 (26, 30) or 4-CF3 (27, 31)
group on the 2-phenyl ring, were expected to possess at least some
affinity for this receptor subtype. On the contrary, only the 4-oxo
derivative 31 showed some capability to bind to the hA3 AR, thus
confirming that in this series of derivatives the CF3 group on the
2-phenyl ring does not have a beneficial role.
For each docking conformation family, we considered the
highest conformation score obtained for each ligand with the
two scoring panels implemented in MOE26 (London dG and Affinity
dG). In some cases, within the same conformation family, we found
the same conformation as the ‘best one’ for the same ligand, with
both scoring function panels. Hence, six sub-databases (two per
family considering the two scoring functions) were obtained. The
calculated scores were then correlated with experimental binding
affinities (pKi), and Affinity dG scores in particular were able to
give a ranking reasonably in accordance with pKi data, as can be
viewed in Fig. 2. London dG scores were unsuccessful in the same
task. Considering the average Affinity dG score for each conforma-
tion family, it can be noticed even that the ‘B’ conformations seem
the favorite among the three families. On the other hand, R square
Derivatives 2, 4, 8–10, 14, 15, 20, 21, which showed the highest
hA3 AR affinities (Ki < 30 nM), were tested in cAMP assays to
evaluate their potencies towards the hA3 receptor. Consistently
with their hA3 affinity values, the tested compounds proved to be
potent in inhibiting the NECA-inhibited cAMP accumulation in
hA3 CHO cells, thus showing an antagonistic behavior (Table 1).
4.2. Molecular modeling studies
A
molecular docking analysis of the pyrazoloquinoline
derivatives was performed on a homology model of hA3 receptor
Figure 1. Left to right, in order A, B and C docking conformation families; the three docking conformations for compound 1 are shown.

V. Colotta et al. / Bioorg. Med. Chem. 17 (2009) 401–410
405
Figure 2. Left to right, experimental pKi-docking score plots for the three conformation families. Docking scores in plots were calculated with MOE Affinity dG scoring
function for minimized docking conformations (see Section 6 for details). Family A: average docking score = ꢀ7.85; correlation R2 = 0.7412. Family B: average docking
score = ꢀ9.45; correlation R2 = 0.7187. Family C: average docking score = ꢀ8.66; correlation R2 = 0.7568.
coefficients calculated for pKi-score correlations indicate that fam-
ily ‘C’ conformation ranking is more in accordance with pKidata
compared to the other families.
Interestingly, the three resulting docking conformation families
show a similar binding motif, with the pyrazolo-pyridine portion of
the scaffold located in correspondence with the ‘Y-shaped’ binding
site core, and three aromatic rings (i.e., the fused benzo ring and
the (hetero)aryl groups at the 2-position and on the 4-carbamoyl
residue) positioned in the peripheral pockets. In addition, the 4-
arylamido carbonyl group and the N5 atom are oriented toward
Ser247 or Thr94, working as hydrogen bonding acceptors. Thus, a
pharmacophore model can be obtained which consists of three
‘peripheral’ hydrophobic/aromatic features and two possible
hydrogen bond acceptor features in proximity of Ser247 and
Thr94. Depending on the docking conformation family, an aryl
group is present in correspondence with each hydrophobic/aro-
matic feature, while at least one of the two H-bond acceptor fea-
tures is matched by the carbonyl group in the 4-arylamido
substituent and the N5 atom. Thus, at least four out of five features
of the derived pharmacophore model are matched by each docking
conformation family (Fig. 3).
These findings are consistent with previous docking stud-
ies,14,25,27,28 in particular with those we achieved on previously
synthesized 4-amidopyrazoloquinoline derivatives.14 In this previ-
ous work, the docking conformation proposed for compound 1
(namely compound 17 in Ref. 14) belongs to the ‘C’ docking confor-
mation family.
In addition, Tafi and co-workers published a pharmacophore-
based modeling study in which pharmacophore hypotheses were
used as guide points to build and refine a hA3 receptor model for
a subsequent docking study.27 In another work, Moro and col-
leagues used a combined structure-based and ligand-based drug
design strategy to identify common key structural features of pyr-
azolo-triazolo-pyrimidine derivatives as hA3 antagonists.28
Very interestingly, our pharmacophore model presents
similarities with regard to the types and 3D organizations of
features suggested in both Tafi’s and Moro’s works.
The pharmacophore map is important to describe the general
interaction scheme of the hA3 pyrazoloquinoline antagonists with
the receptor. However, the nature of the substituent either at the
4-carbamoyl group or on 2-phenyl ring is critical to differentiate
potent from weak compounds. The benzoyl group of the lead com-
pound 1 can be replaced by diverse heteroaroyl residues, such as
the 2-furoyl and 2-(5-methylfuroyl) moieties (compounds 2 and
4), and the 2- or 3- or 4-pyridoyl groups (derivatives 8–10), which af-
ford high hA3 AR affinity (Ki < 15 nM). Instead, R4 = 2- and 3-thienyl,
Figure 3. Superimposition of the three docking conformations of compound 1 (one
conf. per family) and depiction of 3D pharmacophore. Three hydrophobic/aromatic
features (green color) are located at peripheral points of Y-shaped binding site,
while two H-bond acceptor features represent the two possible locations of the
acceptors couple (the 4-arylamido carbonyl group substituent and the N5 atom) in
proximity of Thr94 or Ser247 residues.
4-thiazolyl and 2-pyrazinyl rings (compounds 5–7, 10)caused some
reduction of hA3 AR affinity (Ki = 50–65 nM) which was much more
marked in the case of the 2-and 4-pyrimidyl groups (11 and 12,
Ki = 945 and 210 nM). These results can be rationalized if we con-
sider that, in the Y-shaped binding site, the three peripheral anchor-
ing areas are basically hydrophobic, even if a particular balance of
hydrophobicity, aromaticity, and polarizability is required. In this
sense, introduction of electron-rich heterocycles would not improve
interaction with the hA3 receptor, and the same applies to aromatic
rings substituted with strong electron-withdrawing groups.
In order to confirm that this series of antagonists might assume at
least two different orientations within the hA3 receptor binding site
(Fig. 4B and C) and to validate the proposed pharmacophore model,
compounds 16–21, 36 and 37, bearing a CF3 or Me group on the 2-
phenyl moiety or on the 4-aroylamino group, were investigated.
The binding data indicate that introduction of a meta- or para-Me
substituent in the 2-phenyl ring (36, 37) or in the benzoyl pendant
(20, 21) of the lead compound 1 exerted the same effect. Indeed,
compounds 3614 and 3714 possess the same hA3 affinity of the new
derivatives 20 and 21. Quite similar results have been found for
derivatives bearing the 3-CF3 (16 and 18) and the 4-CF3 substituent
(17 and 19): derivatives 16 (Ki = 52 nM) and 18 (Ki = 60 nM) possess

406
V. Colotta et al. / Bioorg. Med. Chem. 17 (2009) 401–410
Figure 4. (A) Superimposition of ‘A’ docking conformation of compound 16 (hA3 Ki = 52 5 nM) and ‘B’ docking conformation of compound 18 (hA3 Ki = 60 8 nM); 18 is
colored in yellow; H-bond interactions of 16 with hA3 (Thr94 residue) are in lilac dash lines, while H-bond interactions of 18 with hA3 (Ser247 residue) are in orange dash
lines. (B) Schematic view of 16 ‘A’ docking conformation within hA3 binding site; H-bond interactions with Thr94 residue are indicated. (C) Schematic view of 18 ‘Bss’ docking
conformation within hA3 binding site; H-bond interactions with Ser247 residue are indicated.
the same hA3 affinity, while for compounds 17 and 19 only a three-
fold difference exists between their hA3 affinities (17: Ki = 140 nM;
19: Ki = 415 nM). These findings confirm that the 4-amido-2-arylpy-
razolo[3,4-c]quinoline derivatives can bind to the hA3 receptor with
at least two possible binding modes, belonging to the two conforma-
tion families ‘A’ and ‘B’, in which the 2-aryl or the 4-(hetero)aroyl-
amino moiety occupy the same lipophilic pocket.
showed a good druggability profile as they adhere to Lipinski’s
‘rule-of-five’.
6. Experimental
6.1. Chemistry
The significantly lower hA3 AR affinities of the CF3-substituted
compounds 16–20, with respect to those of the corresponding
CH3-substituted derivatives 20, 21, 36 and 37, could be ascribed
to the strong electron-withdrawing properties of the trifluoro-
methyl group which might cause a change in electron density on
the molecule and a distortion of the binding mode at the receptor.
Also the higher steric hindrance of the trifluoromethyl group, with
respect to that of the methyl substituent, might have a role in
reducing the anchoring to the receptor site.
Molecular modeling tools were also used for a further analysis
in order to estimate the druggability of the reported pyrazoloquin-
oline derivatives. In this sense, a set of molecular descriptors (logP,
logS, molecular weight, number of H-bond donors and acceptors,
water accessible surface area) were calculated and analyzed in
comparison with Lipinski’s ‘rule-of-five’ filter (logP < 5; molecular
weight < 500; number of H-bond donors < 5; number of H-bond
acceptors < 10) (see Section 6.2.6 for further details). Among these
calculated descriptors, logP and logS represent, respectively, the
log of the octanol/water partition coefficient and the log of the
aqueous solubility (mol/L); the water accessible surface area
parameters (water accessible surface area, ASA; hydrophobic sur-
face area, ASA_H; polar surface area, ASA_P) were also calculated
as related to the ability of molecules to get solvated.
Microwave-assisted syntheses were performed using an Initia-
tor EXP Microwave Biotage instrument (frequency of irradiation:
2.45 GHz). Silica gel plates (Merck F254) and silica gel 60 (Merck,
70-230 mesh) were used for analytical and column chromatogra-
phy, respectively. All melting points were determined on a Gal-
lenkamp melting point apparatus. Microanalyses were performed
with a Perkin-Elmer 260 elemental analyzer for C, H, N and the
results (reported in Table 2) were within 0.4% of the theoretical
values, unless stated otherwise. The IR spectra were recorded with
a Perkin-Elmer Spectrum RX I spectrometer in Nujol mulls and are
expressed in cmꢀl. The 1H NMR spectra were obtained with a Bruc-
ker Avance 400 MHz instrument. The chemical shifts are reported
in d (ppm) and are relative to the central peak of the solvent which
was DMSO-d6 or CDCl3. The following abbreviations are used:
s = singlet, d = doublet, t = triplet, m = multiplet, br = broad, and
ar = aromatic protons.
6.1.1. General procedure for the synthesis of 2-aryl-2,5-
dihydro-4H-pyrazolo[3,4-c]quinolin-4-ones (28–31)
A
suspension of suitable arylhydrazine hydrochlorides
(9.2 mmol) and ethyl 2-(3-indolyl)-2-oxoethanoate21 (4.6 mmol)
in absolute ethanol (10 mL) and glacial acetic acid (5–6 drops)
was microwave irradiated at 140 °C for 3 min (derivatives 28 and
29), 130 °C for 10 min or 6 min (derivatives 30 and 31, respec-
tively). The suspension was cooled at room temperature and the
solid collected, washed with water and re-crystallized.
The results of this analysis show that, on the whole, the reported
compounds present a good ‘druggability’ profile, and some of them
(1–13, 15, 26–31) do not violate any Lipinski parameter. In addition,
ASA_H, logS and logP values are well correlated, and the estimated
logS and logP values indicate a reasonable solubility for several of
these compounds (many of them presenting a logP value < 4).
Compound 28: yield 98%; mp > 300 °C (EtOH) (lit. mp > 300 °C).14
Compound 29: yield 50%; mp > 300 °C (AcOH) (lit. mp > 300 °C).16
Compound 30: yield 34%; mp > 300 °C (2-ethoxyethanol). 1H
NMR (DMSO-d6) 7.26 (t, 1H, ar, J = 7.6 Hz), 7.36–7.41 (m, 2H, ar),
7.87–7.97(m, 3H, ar), 8.37–8.40 (m, 2H, ar), 9.66 (s, 1H, H-1),
11.54 (br s, lH, NH);IR 3146, 1689.
5. Conclusion
This study describes a new set of 4-(hetero)aroylamino-2-aryl-
pyrazolo[3,4-c]quinoline derivatives as potent and highly selective
hA3 receptors antagonists. Molecular modeling studies showed the
ability of this class of compounds to assume different orientations
within the hA3 receptor homology model binding site, even main-
taining the interaction scheme with receptor residues. Docking
results showed an interesting correlation between docking scores
and experimental binding data, and they were employed to devel-
op a pharmacophore model which could be helpful for the inter-
pretation of the observed affinities and for the design of new hA3
pyrazoloquinoline antagonists. Several of the new derivatives
Compound 31: yield 36%; mp 279–281 °C (EtOH). 1H NMR
(DMSO-d6) 2.26 (t, 1H, ar, J = 7.6 Hz), 7.38–7.41 (m, 2H, ar), 7.97
(d, 1H, ar, J = 7.7 Hz), 8.02 (d, 2H, ar, J = 8.6 Hz), 8.25 (d, 2H, ar,
J = 8.6 Hz), 9.63 (s, 1H, H-l), 11.53 (br s, 1H, NH); IR 3150, 1669.
6.1.2. General procedure for the synthesis of 2-aryl-4-chloro-
2H-pyrazolo[3,4-c]quinoline (32–35)
A mixture of 28–31 (3.6 mmol) and phosphorus pentachloride
(1.1 mmol) in phosphorus oxychloride (20 mL) was heated at reflux
for 2–3 h. Evaporation at reduced pressure of the excess of phospho-
rus oxychloride gave an oil which after treatment with cold water

V. Colotta et al. / Bioorg. Med. Chem. 17 (2009) 401–410
407
Table 2
6.1.4.1. Method A (compounds 2, 5, 7–10, 20, 21). A mixture
of the 4-amino derivative 24 (1 mmol), the suitable carboxylic
acid (7 mmol), 1-(3-(dimethylamino)propyl)-3-ethyl-carbodiimide
hydrochloride (7 mmol), 1-hydroxybenzotriazole (7 mmol), trieth-
ylamine (15 mmol) and 4-(dimethylamino)pyridine (0.1 mmol) in
anhydrous dimethylformamide (2–3 mL) was stirred at room tem-
perature until the disappearance of the starting 4-amino derivative
(TLC monitoring, 3–85 h). The solid was filtered off and the cooled
(0 °C) mother liquor was diluted with water (about 15–20 mL). The
precipitated solid was collected and re-crystallized from the suit-
able solvent.
Combustion analysis data of the newly synthesized compounds
Compound
Formula
C
H
N
Calcd-found
Calcd-found
Calcd-found
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
26
27
30
31
C21H14N4O2
C21H14N4O2
C22H16N4O2
C21H14N4OS
C21H14N4OS
C20H13N5OS
C22H15N5O
71.18–71.30
71.18–71.45
71.73–71.58
68.09–68.24
68.09–68.19
64.68–64.41
72.32–72.15
72.32–72.55
72.32–72.65
68.84–68.60
68.84–69.05
68.84–68.99
66.09–66.23
64.87–64.51
66.66–66.90
66.66–66.73
66.66–66.80
66.66–66.46
76.17–76.31
76.17–76.39
63.58–63.70
63.58–63.33
62.20–61.91
62.20–61.98
62.01–62.22
62.01–62.31
3.98–3.71
3.98–4.15
4.38–4.12
3.81–3.69
3.81–4.02
3.53–3.86
4.14–4.41
4.14–4.33
4.14–3.89
3.85–4.08
3.85–3.98
3.85–3.44
3.53–3.24
3.37–3.02
3.50–3.28
3.50–3.87
3.50–3.76
3.50–3.19
4.79–4.43
4.79–5.00
3.00–2.86
3.00–3.28
3.37–3.60
3.37–3.01
3.06–3.28
3.06–2.94
15.81–15.65
15.81–15.58
15.21–15.44
15.12–15.35
15.12–14.95
18.86–19.08
19.17–19.38
19.17–19.01
19.17–19.24
22.94–22.75
22.94–22.65
22.94–23.11
17.52–17.37
14.41–14.29
12.96–12.72
12.96–12.68
12.96–13.06
12.96–13.15
14.81–14.55
14.81–14.51
9.27–9.41
C22H15N5O
C22H15N5O
C21H14N6O
C21H14N6O
C21H14N6O
C22H14ClN5O
C21H13ClN4O2
Compound 2: yield 90%; mp 193–194 °C (EtOH); 1H NMR
(CDCl3) 6.59–6.60 (m, 1H, furane H-4 proton), 7.41–7.66 (m, 8H,
6ar + 2 furane protons), 7.95 (d, 1H, ar, J = 7.7 Hz), 8.02 (d, 2H, ar,
J = 7.8 Hz), 8.70 (s, 1H, H-1); IR 3389, 1703.
Compound 5: yield 70%; mp 188–189 °C (2-methoxyethanol);
1H NMR (CDCl3) 7.17–7.18 (m, 1H, thiophene proton), 7.43–7.62
(m, 7H, 6 ar + 1 thiophene proton), 7.95 (d, 1H, ar, J = 7.6 Hz),
8.02 (d, 2H, ar, J = 8.0 Hz), 8.12–8.13 (m, 1H, thiophene proton),
8.71 (s, 1H, H-1).
C
24H15F3N4O
C24H15F3N4O
C24H15F3N4O
C24H15F3N4O
C24H18N4O
C24H18N4O
C32H18F6N4O2
C32H18F6N4O2
Compound 7: yield 85%; mp 215–217 °C (EtOAc/EtOH); 1H NMR
(DMSO-d6) 7.53–7.71 (m, 5H, ar), 7.92 (d, 1H, ar, J = 7.2 Hz), 8.13 (d,
2H, ar, J = 8.0 Hz), 8.26 (d, 1H, ar, J = 6.7 Hz), 8.68 (s, 1H, thiazole
proton), 9.39 (s, 1H, thiazole proton), 9.75 (s, 1H, H-1), 10.73 (s,
1H, NH); IR 3358, 1702.
9.27–9.03
C
17HnF3N4
17.06–17.29
17.06–17.14
12.76–12.50
12.76–12.92
C17HnF3N4
C17H10F3N3O
C17H10F3N3O
Compound 8: yield 82%; mp 220–222 °C (2-methoxyethanol);
1H NMR (DMSO-d6)7.53–7.71 (m, 5H, ar), 7.78–7.81 (m, 1H, ar),
7.93 (d, 1H, ar, J = 7.4 Hz), 8.13–8.30 (m, 5H, ar), 8.86 (d, 1H, pyri-
dine proton, J = 4.4 Hz), 9.75 (s, 1H, H-1), 11.33 (s, 1H, NH); IR
3324, 1711.
(50 mL) yielded a solid which was quickly collected. The 4-chloro
derivatives 32–35 were unstable, however they were pure enough
to be characterized and used without further purification.
Compound 32: yield 90%, Ref. 16.
Compound 9: yield 50%; mp 231–233°C (2-methoxyethanol); 1H
NMR (CDCl3) 7.28–7.64 (m, 7H, ar), 7.99–8.00 (d, 1H, ar, J = 7.8 Hz),
8.05 (d, 2H, ar, J = 8.0 Hz), 8.74–8.78 (m, 3H, 2 ar + H-1), 9.68 (br s,
1H, ar).
Compound 33: yield 84%, Ref. 16.
Compound 34: yield 85%; 1H NMR (DMSO-d6)7.71–7.76 (m, 2H,
ar), 7.93–7.94 (m, 2H, ar), 8.01 (d, 1H, ar, J = 7.7 Hz), 8.30 (d, 1H, ar,
J = 7.4 Hz), 8.47–8.48 (m, 2H, ar), 10.01 (s, 1H, H-1).
Compound 35: yield, 80%; 1H NMR (DMSO-d6) 7.71–7.75 (m, 2H,
ar), 8.01 (d, 1H, ar, J = 7.9 Hz), 8.04 (d, 2H, ar, J = 8.4 Hz), 8.32 (d, 1H,
ar, J = 7.9 Hz), 8.38 (d, 2H, ar, J = 8.4 Hz), 9.97 (s, 1H, H-1).
Compound 10: yield 80%; mp 246–248 °C (2-methoxyethanol);
1H NMR (CDCl3) 7.50–7.63 (m, 6H, ar), 8.00–8.06 (m, 3H, ar),
8.32–8.33 (m, 2H, ar), 8.76 (s, 1H, H-1), 8.82–8.83 (m, 2H, ar).
Compound 20: yield 40%; mp 170–172 °C (EtOH); 1H NMR
(CDCl3) 2.50 (s, 3H, Me), 7.38–7.78 (m, 8H, ar), 7.95–8.05 (m, 3H,
ar), 8.21–8.29 (m, 2H, ar), 8.77 (s, 1H, H-1).
6.1.3. General procedure for the synthesis of 2-aryl-2H-
pyrazolo[3,4-c]quinolin-4-amines (24–27)
Compound 21: yield 50%; mp 115–116 °C (cyclohexane/EtOAc);
1H NMR (CDCl3) 2.47 (s, 3H, Me), 7.44 (d, 2H, ar, J = 7.9 Hz),
7.45–7.68 (m, 6H, ar), 7.98 (d, 1H, ar, J = 7.7 Hz), 8.03 (d, 2H, ar,
J = 7.9 Hz), 8.34 (d, 2H, ar, J = 7.7 Hz), 8.74 (s, 1H, H-1).
A mixture of the suitable 4-chloro derivatives 32–35 (3 mmol)
in absolute ethanol saturated with ammonia was heated overnight
at 120 °C in a sealed tube. The solid that precipitated upon cooling
was collected, washed with water and re-crystallized.
Compound 24: yield 85%; mp 198–199 °C (EtOAc) (lit. mp 197–
198 °C).16
6.1.4.2. Method B (compounds 3, 4, 6, 11–13). The 4-amino
derivative 24 (1 mmol) was reacted with the commercially
available 3-furyl- (5 mmol), 3-thienyl- (6 mmol), 2-pyrazinyl-
(3 mmol) carboxylic acids, and with the synthesized 2-(5-meth-
ylfuryl)-2-pyrimidyl- and 4-pyrimidyl-carboxylic acids (4 mmol)
in the presence of 1-(3-(dimethylamino)propyl)-3-ethyl-carbodi-
imide hydrochloride (3 mmol), 1-hydroxybenzotriazole (3 mmol),
triethylamine (5 mmol) and 4-(dimethylamino)pyridine (0.1
mmol) in anhydrous dimethylformamide (2–3 mL). The reaction
mixture was stirred at room temperature until the disap-
pearance of the starting 4-amino derivative (TLC monitoring,
50–160 h). The insoluble solid was filtered off and the cooled
(0 °C) dimethylformamide solution diluted with water (about
10–20 mL) to yield a precipitate which was collected and re-crys-
tallized from suitable solvent. For compound 13, the reaction
mixture was filtered and the solid washed with water and
re-crystallized.
Compound 25: yield 65%; mp 244–246 °C (EtOH) (lit. mp 243–
246 °C).16
Compound 26: yield 30%; mp 237–239 °C (EtOH); 1H NMR
(DMSO-d6)7.06 (br s, 2H, NH2), 7.26 (t, 1H, ar, J = 6.9 Hz), 7.42 (t,
1H, ar, J = 6.9 Hz), 7.51 (d, 1H, ar, J = 8.1 Hz), 7.86–7.90 (m, 2H,
ar), 7.99 (d, 1H, ar, J = 7.7 Hz), 8.45 (d, 1H, ar, J = 8.2 Hz), 8.51 (s,
1H, ar), 9.67 (s, 1H, H-1); IR 3465, 3306, 1653.
Compound 27: yield 45%; mp 272–275 °C (EtOH); 1H NMR
(DMSO-d6)7.12 (br s, 2H, NH2), 7.26 (t, 1H, ar, J = 6.9 Hz), 7.41 (t,
1H, ar, J = 6.9 Hz), 7.53 (d, 1H, ar, J = 7.8 Hz), 8.00 (d, 1H, ar,
J = 7.8 Hz), 8.02 (d, 2H, ar, J = 8.4 Hz), 8.35 (d, 2H, ar, J = 8.4 Hz),
9.65 (s, 1H, H-1); IR 3460, 3295, 1648.
6.1.4. General procedures for the synthesis of 2-aryl-2H-
pyrazolo[3,4-c]quinolin-4-amides (2–21)
The title compounds 2–21 were synthesized following the same
procedure but employing different stoichiometric ratios among the
starting amino derivatives 24–27 and the reagents (Methods A–C).
Compound 3: yield 55%; mp 127–130 °C (EtOH); 1H NMR
(CDCl3) 7.03 (s, 1H, furane proton), 7.44–7.59 (m, 7H, ar), 7.96 (d,
1H, ar, J = 7.8 Hz), 8.02 (d, 2H, ar, J = 8.2 Hz), 8.46 (s, 1H, furane
proton), 8.76 (s, 1H, H-1); IR 3300–3550, 1625.

408
V. Colotta et al. / Bioorg. Med. Chem. 17 (2009) 401–410
Compound 4: yield 60%; mp 167–169 °C (EtOH); 1H NMR
6.1.5. General procedure for the synthesis of 2-phenyl-4-[bis(3-
trifluoromethylbenzoyl)]amino-pyrazolo[3,4-c]quinoline (22)
and 2-phenyl-4-[bis(4-trifluoromethylbenzoyl)]amino-
pyrazolo[3,4-c]quinoline (23)
The title compounds 22 and 23 were obtained by refluxing the
4-amino derivative 24 (2 mmol) with an excess (5.0 mmol) of 3-tri-
fluoromethylbenzoyl chloride and 4-trifluoromethylbenzoyl chlo-
ride, respectively, in anhydrous methylene chloride (5 mL) and
pyridine (5.0 mmol). The suspension was refluxed until the disap-
pearance of the starting material (TLC monitoring, 24–48 h), then
cooled at room temperature. The solid was collected by filtration,
washed with water and re-crystallized.
(CDCl3) 2.48 (s, 3H, Me), 6.21 (d, 1H, furane proton, J = 3.3 Hz),
7.45–7.61 (m, 7H, 5ar + 2 furane protons), 7.97 (d, 1H, ar,
J = 7.7 Hz), 8.02 (d, 2H, ar, J = 7.9 Hz), 8.74 (s, 1H, H-9), IR 1624.
Compound 6: yield 65%; mp 147–150 °C (EtOH); 1H NMR
(CDCl3) 7.35–7.37(m, 1H, thiophene proton), 7.44–7.62 (m, 6H,
ar), 7.85 (d, 1H, thiophene proton, J = 5.0 Hz), 7.96 (d, 1H, ar,
J = 7.7 Hz), 8.01 (d, 2H, ar, J = 8.1 Hz), 8.52 (d, 1H, thiophene proton,
J = 3.0 Hz), 8.72 (s, 1H, H-1); IR 3550, 3490, 1630.
Compound 11: yield 40%; mp 168–172 °C (2-methoxyethanol);
1H NMR (DMSO-d6)7.53–7.85 (m, 7H, ar), 8.14 (d, 2H, ar,
J = 7.9 Hz), 8.23 (d, 1H, ar, J = 8.1 Hz), 9.10 (d, 2H, pyrimidine
protons, J = 4.7 Hz), 9.74 (s, 1H, H-1), 11.32 (br s, 1H, NH); IR
3474, 3337, 1720.
Compound 22: yield 45%; mp 213–214 °C (EtOH); 1H NMR
(DMSO-d6) 7.53 (t, 1 H, ar, J = 7.1 Hz), 7.61–7.74 (m, 6H, ar), 7.82
(d, 1H, ar, J = 7.9 Hz), 7.94 (d, 2H, ar, J = 7.6 Hz), 8.03 (d, 2H, ar,
J = 7.8 Hz), 8.15–8.19 (m, 4H, ar), 8.31 (d, 1 H, ar, J = 8.0 Hz), 9.81
(s, 1H, H-9); IR 1726, 1690.
Compound 23: yield 35%; mp 252–254 °C (2-methoxyethanol/
EtOH); 1H NMR (DMSO-d6) 7.53 (t, 1H, ar, J = 7.2 Hz), 7.61–7.72
(m, 5H, ar), 7.79 (d, 4H, ar, J = 8.0 Hz), 8.03 (d, 2H, ar, J = 7.8 Hz),
8.09 (d, 4H, ar, J = 8.1 Hz), 8.31 (d, 1H, ar, J = 7.8 Hz), 9.85 (s, 1H,
H-9); IR 1730, 1700.
Compound 12: yield 45%; mp 195–198 °C (2-ethoxyethanol); 1H
NMR (DMSO-d6)7.55 (t, 1H, ar, J = 7.1 Hz), 7.64–7.71 (m, 4H, ar),
7.93–8.26 (m, 5H, ar), 9.23 (d, 1H, pyrimidine proton, J = 4.6 Hz),
8.54 (s, 1H, ar), 9.77 (s, 1H, ar), 11.26 (br s, 1H, NH); IR 3425,
3315, 1707.
Compound 13: yield 87%; mp 249–251 °C (2-ethoxyethanol); 1H
NMR (DMSO-d6)7.56 (t, 1H, ar, J = 7.3 Hz), 7.64–7.71 (m, 4H, ar),
7.95 (d, 1H, ar, J = 7.7 Hz), 8.13 (d, 2H, ar, J = 7.9 Hz), 8.26 (d, 1H,
J = 7.0 Hz), 8.93 (s, 1H, ar), 9.03 (s, 1H, ar), 9.40 (s, 1H, ar), 9.76
(s, 1H, ar), 11.12 (br s, 1H, NH);IR3395, 1716.
6.2. Molecular modeling studies
6.1.4.3. Method C (compounds 14–19). A mixture of the 4-ami-
no derivatives 24–27 (1 mmol), the suitable carboxylic acid (3 mmol),
6.2.1. Computational methodologies
All molecular modeling studies were performed on a 2 CPU (PIV
2.0–3.0 GHZ) Linux PC. Homology modeling and docking studies
were carried out using Molecular Operating Environment (MOE,
version 2007.09) suite.26 All ligand structures were optimized
using RHF/AM1 semiempirical calculations, and the software pack-
age MOPAC implemented in MOE was utilized for these
calculations.29
1-(3-(dimethylamino)propyl)-3-ethyl-carbodiimide
hydrochloride
(3 mmol), 1-hydroxybenzotriazole (3 mmol), triethylamine (5
mmol) and 4-(dimethylamino)pyridine (0.1 mmol) in anhydrous
dimethylformamide (2–3 mL) was stirred at room temperature
until the disappearance of the starting amino derivative (TLC mon-
itoring, 6–140 h). For compound 14, the reaction mixture was fil-
tered and the solid was washed with water and purified by
column chromatography (eluent CHCl3/MeOH, 9:1). For com-
pounds 15, 16–19, the reaction mixture was diluted with ice water
(about 10–20 mL) to yield a solid which was collected by filtration
and washed with water. Crude compounds 15, 16, 18, 19 were
purified by re-crystallization from suitable solvent while com-
pound 17 was purified by column chromatography (eluent
CH2Cl2/EtOAc, 9:1).
6.2.2. Homology model of the human A3 receptor
A homology model of the human A3 receptor was built using the
X-ray crystal structure of the resting state bovine rhodopsin (pdb
code: 1L9H;30 available at the RCSB Protein Data Bank, http://
www.rcsb.org) with a 2.6-Å resolution as template. The amino acid
sequences of TM helices of the human A3 receptor and bovine rho-
dopsin were aligned, and in this phase some GPCRs highly con-
served amino acid residues worked as guide, including the DRY
motif (D3.49, R3.50, Y3.51, or D107, R108, Y109, respectively)
and three Pro residues (P4.60, P6.50, P7.50 or P145, P189, P245,
respectively). The boundaries identified from the X-ray crystal
structure of bovine rhodopsin were applied for the corresponding
sequences of the TM helices of the A3 receptor. The loop domains
of the human A3 receptor were built by the loop search method
implemented in MOE. Special care had to be given to the second
extracellular (E2) loop, which folds back over TM domains. This
loop limits the dimension of the active site, and its amino acids
could be involved in direct interactions with the ligands. The pres-
ence of a conserved disulfide bridge between cysteines in TM3 and
E2 might be the driving force to this particular fold of the E2 do-
main, so this loop was modeled using a rhodopsin-like constrained
geometry around the E2-TM3 disulfide link. Because of the limited
sequence similarity between the human A3 receptors and the tem-
plate in the C-terminal domain, only a short segment of this region
was modeled. In particular, the model ends with the Ser306
residue corresponding to the Asp330 residue of the bovine rhodop-
sin template. Once the heavy atoms were modeled, all hydrogen
atoms were added, and the protein coordinates were then mini-
mized with MOE using the AMBER99 force field.31 The minimiza-
tions were performed by 1000 steps of steepest descent followed
by conjugate gradient minimization until the RMS gradient of the
Compound 14: yield 25%; mp 192–194 °C (EtOH); 1H NMR
(CDCl3) 7.53 (t, 1H, ar, J = 8.2 Hz), 7.60–7.61 (m, 4H, ar) 8.01–8.03
(m, 3H, ar), 8.43–8.44 (m, 2H, ar), 8.71 (s, 1H, H-1), 8.85 (d, 2H, pyr-
idine protons, J = 5.0 Hz); IR 3200–3500, 1625.
Compound 15: yield 15%; mp 251–252 °C (EtOH/2-methoxy-
ethanol); 1H NMR (CDCl3) 6.60–6.61 (m, 1H, furane proton),
7.47 (t, 1H, ar, J = 7.9 Hz), 7.62–7.67 (m, 6H, 4ar + 2 furane
protons), 7.96–8.01 (m, 3H, ar), 8.71 (s, 1H, H-1); IR 3300–3500,
1708.
Compound 16: yield 89%; mp 211–213 °C (2-methoxyethanol);
1H NMR (DMSO-d6) 7.56–7.67 (m, 5H, ar), 7.89–7.90 (m, 3H, ar),
8.10–8.22 (m, 3H, ar), 8.43–8.45 (m, 2H, ar), 9.89 (s, 1H, H-9),
11.10 (br s, 1H, NH); IR 3233, 1646.
Compound 17: yield 63%; mp 201–202 °C (MeOH); 1H NMR
(DMSO-d6)7.56–7.67 (m, 5H, ar), 7.90–7.92 (m, 1H, ar), 8.04 (d,
2H, ar, J = 8.4 Hz), 8.19–8.23 (m, 3H, ar), 8.33 (d, 2H, ar,
J = 8.4 Hz), 9.88 (s, 1H, H-l), 11.11 (br s, 1H, NH).
Compound 18: yield 60%; mp 188–190 °C (EtOH); 1H NMR
(CDCl3) 7.46–7.62 (m, 7H, ar), 7.81 (d, 1H, ar, J = 7.4 Hz), 7.98
(d, 1H, ar, J = 7.6 Hz), 8.05 (d, 2H, ar, J = 7.7 Hz), 8.71–8.75 (m,
3H, ar).
Compound 19: yield 74%; mp 204–207 °C (EtOH); 1H NMR
(CDCl3) 7.47–7.64 (m, 6H, ar), 7.77 (d, 2H, ar, J = 7.8 Hz), 7.98 (d,
1H, ar, J = 7.8 Hz), 8.04 (d, 2H, ar, J = 8.7 Hz), 8.61 (d, 2H, ar,
J = 7.8 Hz), 8.73 (s, 1H, H-1).
potential energy was less than 0.05 kJ molꢀ1 Åꢀ1
.

V. Colotta et al. / Bioorg. Med. Chem. 17 (2009) 401–410
409
6.2.3. TM binding site identification and refining
r2 = 0.90, ꢁ1, 200 molecules. logP (o/w): log of the octanol/water
partition coefficient (including implicit hydrogen atoms); this
property is calculated from a linear atom type model with
r2 = 0.931, RMSE = 0.393 on ꢁ1, 827 molecules; Weight: molecular
weight (including implicit hydrogen atoms) in atomic mass units
with atomic weights; a_acc: number of hydrogen bond acceptor
atoms (not counting acidic atoms but counting atoms that are both
hydrogen bond donors and acceptors such as –OH); a_don: num-
ber of hydrogen bond donor atoms (not counting basic atoms but
counting atoms that are both hydrogen bond donors and acceptors
such as -OH); ASA: water accessible surface area calculated using a
radius of 1.4 Å for the water molecule. A polyhedral representation
is used for each atom in calculating the surface area; h: number of
heavy atoms; VSA: van der Waals surface area. A polyhedral repre-
sentation is used for each atom in calculating the surface area;
ASA_H: water accessible surface area of all hydrophobic (jqij < 0.2)
atoms; ASA_P: water accessible surface area of all polar
(jqij > = 0.2) atoms. The lip_violation parameter indicates the num-
ber of violations of Lipinski’s ‘rule of five’;40 the drug-like param-
eter is the estimation of molecular druggability, and it is indicated
as ‘y’ if and only if lip_violation parameter = 0, otherwise ‘n’. The
values of the molecular descriptors and the other calculated
parameters are reported in the Supplementary data.
The human A3 receptor binding site region was established with
the aid of published site directed mutagenesis data, which indi-
cated in particular that residues His95, Ser247, Asn250, and
His272 are important for ligand binding.25 As this receptor region
did not contain a suitable cavity for antagonist docking, a manual
docking section was performed in which compound 1 was inserted
in the established binding site assuming a number of different ori-
entations. Thus, an equivalent number of protein–ligand com-
plexes was obtained. Each complex was then minimized. In this
step, the coordinates of a sphere of residues in proximity of ligand
structure (6 Å) were left unfixed, while the remaining receptor res-
idue coordinates were fixed.
The unfixed atom sets were the same for all different recep-
tor–ligand complexes. Even in this case, the minimizations were
performed by 1000 steps of steepest descent followed by conjugate
gradientminimizationuntiltheRMSgradientofthepotentialenergy
was less than 0.05 kJ mol1 Å1. The final step was the measurement of
potential energy of the different receptor conformations (extracted
from the respective receptor–ligand complexes) with the MOE
potentialenergytool. Thereceptorconformationwiththelowestpo-
tential energy value was selected for the following docking section.
6.2.4. Molecular docking of the human A3 receptor antagonists
All antagonist structures were docked into the refined TM bind-
ing site by using the MOE Dock tool. This method is divided into a
number of stages: Conformational analysis of ligands. The algo-
rithm generated conformations from a single 3D conformation by
conducting a systematic search. In this way, all combinations of
angles were created for each ligand. Placement. A collection of
poses was generated from the pool of ligand conformations using
Alpha Triangle placement method. Poses were generated by super-
position of ligand atom triplets and triplets of points in the recep-
tor binding site. The receptor site points are alpha sphere centers
which represent locations of tight packing. At each iteration a ran-
dom conformation was selected, a random triplet of ligand atoms
and a random triplet of alpha sphere centers were used to deter-
mine the pose. Scoring. Poses generated by the placement method-
ology were scored using two available methods implemented in
MOE, the London dG scoring function which estimates the free
energy of binding of the ligand from a given pose, and Affinity
dG Scoring which estimates the enthalpic contribution to the free
energy of binding. The top 30 poses for each ligand were output
in a MOE database. Each resulting ligand pose was then subjected
to MMFF9432–38 energy minimization until the RMS gradient of the
potential energy was less than 0.05 kJ molꢀ1 Åꢀ1. In this phase,
AMBER99 partial charges of receptor and MOPAC output partial
charges of ligands were conserved. The minimized docking confor-
mations were then re-scored using both London dG scoring func-
tion and Affinity dG scoring method.
6.3. Pharmacological assays
6.3.1. Human cloned A1, A2A, and A3 adenosine receptor
binding assay
All synthesized compounds were tested to evaluate their affin-
ity at human A1, A2A and A3 adenosine receptors. Displacement
experiments of [3H]DPCPX (1 nM) to hA1 CHO membranes (50
lg
of protein/assay) and at least six to eight different concentrations
of antagonists for 120 min at 25 °C in 50 mM Tris–HCl buffer pH
7.4 were performed.41 Non-specific binding was determined in
the presence of 10
l
M of CHA (<10% of the total binding). Binding
of [3H]ZM-241385 (1 nM) to hA2ACHO membranes (50
l
g of pro-
tein/assay) was performed using 50 mM Tris–HCl buffer, 10 mM
MgCl2 pH 7.4 and at least six to eight different concentrations of
antagonists studied for an incubation time of 60 min at 4 °C.42
Non-specific binding was determined in the presence of 1
ZM-241385 and was about 20% of total binding. Competition bind-
lM
ing experiments to hA3 CHO membranes (50 lg of protein/assay)
and 0.5 nM [125I]AB-MECA, 50 mM Tris–HCl buffer, 10 mM MgCl2,
1 mM EDTA, pH 7.4 and at least six to eight different concentra-
tions of examined ligands for 120 min at 4°C.43 Non-specific bind-
ing was defined as binding in the presence of 1 lM AB-MECA and
was about 20% of total binding. Bound and free radioactivity were
separated by filtering the assay mixture through Whatman GF/B
glass fiber filters using a Brandel cell harvester. The filter bound
radioactivity was counted by Scintillation Counter Packard Tri Carb
2500 TR with an efficiency of 58%.
6.2.5. Pharmacophore model
The pharmacophore model was developed from docking confor-
mations using the Pharmacophore Query Editor tool implemented
in MOE. In particular, Pharmacophore Consensus tool created a list
of suggested features from the superimposed docking conforma-
tions, based on specified consensus parameters. Suggested features
were converted to query features and were considered to develop
the pharmacophore model.
6.3.2. Rat A3 adenosine receptor binding assays
Selected compounds were tested for evaluating their affinity at
rat A3 adenosine receptors expressed in HEK 293 cells (Perkin-Elmer,
Boston, USA). Competition binding experiments to rA3 HEK mem-
branes (10 l
g of protein/assay) and 0.5 nM [125I]AB-MECA, 50 mM
Tris–HCl buffer, 10 mM MgCl2, 1 mM EDTA, pH 7.4 and at least 3–4
different concentrations of examined ligands for 120 min at
25 °C.43 Non-specific binding was defined as binding in the presence
6.2.6. Molecular descriptors and druggability evaluation
A set of molecular descriptors were calculated for compounds
1–23, 26, 27, 30, 31, 36–38 with MOE software. The descriptors,
taken from MOE software documentation, are as follows. logS:
log of the aqueous solubility (mol/L); this property is calculated
from an atom contribution linear atom type model39 with
of 1 lM AB-MECA and was about 20% of total binding. Bound and
free radioactivity were separated by filtering the assay mixture
through Whatman GF/B glass fiber filters by using a Brandel cell har-
vester. The filter bound radioactivity was counted by Scintillation
Counter Packard Tri Carb 2500 TR with an efficiency of 58%.

410
V. Colotta et al. / Bioorg. Med. Chem. 17 (2009) 401–410
5. Shneyvays, V.; Leshem, D.; Zinman, T.; Mamedova, L. K.; Jacobson, K. A.;
Shainberg, A. Am. J. Physiol. Heart Circ. Physiol. 2005, 288, H2792.
6. Schulte, G.; Fredholm, B. B. Cell Signal. 2003, 15, 813.
7. Brambilla, R.; Cattabeni, F.; Ceruti, S.; Barbieri, D.; Franceschi, C.; Kim, Y. C.;
Jacobson, K. A.; Klotz, K. N.; Lohse, M. J.; Abbracchio, M. P. Naunyn-
Schmiedeberg’s Arch. Pharmacol. 2000, 361, 225.
8. Merighi, S.; Mirandola, P.; Varani, K.; Gessi, S.; Leung, E.; Baraldi, P. G.; Tabrizi,
M. A.; Borea, P. A. Pharmacol. Ther. 2003, 100, 48.
9. Merighi, S.; Benini, A.; Mirandola, P.; Gessi, S.; Varani, K.; Leung, E.; Maclennan,
S.; Borea, P. A. Biochem. Pharmacol. 2006, 72, 19.
10. Lee, H. T.; Ota-Setlik, A.; Xu, H.; D’Agati, V. D.; Jacobson, M. A.; Emala, C. W. Am.
J. Physiol. Renal Physiol. 2003, 284, F267.
11. Yang, H.; Avila, M. Y.; Peterson-Yantorno, K.; Coca-Prados, M.; Stone, R. A.;
Jacobson, K. A.; Civan, M. M. Curr. Eye Res. 2005, 30, 747.
12. Pugliese, A. M.; Coppi, E.; Spalluto, G.; Corradetti, R.; Pedata, F. Br. J. Pharmacol.
2006, 147, 524.
13. Pugliese, A. M.; Coppi, E.; Volpini, R.; Cristalli, G.; Corradetti, R.; Jeong, L. S.;
Jacobson, K. A.; Pedata, F. Biochem. Pharmacol. 2007, 74, 768.
14. Colotta, V.; Catarzi, D.; Varano, F.; Capelli, F.; Lenzi, O.; Filacchioni, G.; Martini,
C.; Trincavelli, L.; Ciampi, O.; Pugliese, A. M.; Pedata, F.; Schiesaro, A.; Morizzo,
E.; Moro, S. J. Med. Chem. 2007, 50, 4061.
15. Colotta, V.; Catarzi, D.; Varano, F.; Cecchi, L.; Filacchioni, G.; Martini, C.;
Trincavelli, L.; Lucacchini, A. J. Med. Chem. 2000, 43, 1158.
16. Colotta, V.; Catarzi, D.; Varano, F.; Cecchi, L.; Filacchioni, G.; Martini, C.;
Trincavelli, L.; Lucacchini, A. J. Med. Chem. 2000, 43, 3118.
17. Colotta, V.; Catarzi, D.; Varano, F.; Calabri, F. R.; Lenzi, O.; Filacchioni, G.;
Martini, C.; Trincavelli, L.; Deflorian, F.; Moro, S. J. Med. Chem. 2004, 47,
3580.
18. Catarzi, D.; Colotta, V.; Varano, F.; Lenzi, O.; Filacchioni, G.; Trincavelli, L.;
Martini, C.; Montopoli, C.; Moro, S. Med. Chem. 2005, 48, 7932.
19. Lenzi, O.; Colotta, V.; Catarzi, D.; Varano, F.; Filacchioni, G.; Martini, C.;
Trincavelli, L.; Ciampi, O.; Varani, K.; Marighetti, F.; Morizzo, E.; Moro, S. J. Med.
Chem. 2006, 49, 3916.
20. Morizzo, E.; Capelli, F.; Lenzi, O.; Catarzi, D.; Varano, F.; Filacchioni, G.;
Vincenzi, F.; Varani, K.; Borea, P. Andrea; Colotta, V.; Moro, S. J. Med. Chem.
2007, 50, 6596.
21. Casnati, G.; Ricca, A. Gazz. Chim. Ital. 1963, 93, 355.
22. Robert, J. M. H.; Robert-Piessard, S.; Courant, J.; Le Baut, G.; Robert, B.;
Lang, F.; Petit, J. Y.; Grimaud, N.; Welin, L. Eur. J. Med. Chem. 1995, 30,
915.
6.3.3. Measurement of cyclic AMP levels in CHO cells
transfected with hA2B or hA3 adenosine receptors
CHO cells transfected with hAR subtypes were washed with
phosphate-buffered saline, diluted tripsine and centrifuged for
10 min at 200g. The pellet containing the CHO cells (1 ꢂ 106 cells/as-
say) was suspended in 0.5 ml of incubation mixture (mM): NaCl 15,
KCl 0.27, NaH2PO4 0.037, MgSO4 0.1, CaCl2 0.1, Hepes 0.01, MgCl2 1,
glucose 0.5, pH 7.4 at 37 °C, 2 IU/ml adenosine deaminase and4-(3-
butoxy-4-methoxybenzyl)-2-imidazolidinone (Ro 20–1724) as
phosphodiesterase inhibitor and preincubated for 10 min in a
shaking bath at 37 °C. The potency of antagonists to A2B receptors
was determined by antagonism of NECA (200 nM)-induced stimula-
tion of cyclic AMP levels. In addition, the potency of antagonists to A3
receptors was determined in the presence of forskolin 1 lM and Cl-
IB-MECA (100 nM) thatmediated inhibitionof cyclic AMP levels. The
reaction was terminated by the addition of cold 6% trichloroacetic
acid (TCA). The TCA suspension was centrifuged at 2000g for
10 min at 4 °C and the supernatant was extracted four times with
water saturated diethyl ether. The final aqueous solution was tested
for cyclic AMP levels by a competition protein binding assay. Sam-
ples of cyclic AMP standard (0–10 pmol) were added to each test
tube containing [3H] cyclic AMP and the incubation buffer (trizma
base 0.1 M, aminophylline 8.0 mM, 2-mercaptoethanol 6.0 mM, pH
7.4). The binding protein prepared from beef adrenals, was added
to the samples previously incubated at 4 °C for 150 min, and after
the addition of charcoal were centrifuged at 2000g for 10 min. The
clear supernatant was counted in a Scintillation Counter Packard
Tri Carb 2500 TR with an efficiency of 58%.44
6.3.4. Data analysis
The protein concentration was determined according to a Bio-
Rad method45 with bovine albumin as a standard reference.
Inhibitory binding constant, Ki, values were calculated from those
of IC50 according to Cheng and Prusoff equation Ki ¼ IC50=ð1þ
½Cꢃꢄ=KꢃDÞ, where [C*] is the concentration of the radioligand and
KꢃD its dissociation constant.46 A weighted non-linear least-squares
curve fitting program LIGAND47 was used for computer analysis of
inhibition experiments. EC50 and IC50 values obtained in cyclic
AMP assay were calculated by non-linear regression analysis using
the equation for a sigmoid concentration–response curve (Graph-
PAD Prism, San Diego, CA, USA).
23. Dunn, G. E.; Lawler, E. A.; Yamashita, A. B. Can. J. Chem. 1977, 55, 2478.
24. Tagawa, Y.; Yamashita, K.; Higuchi, Y.; Goto, Y. Heterocycles 2003, 60,
953.
25. Moro, S.; Spalluto, G.; Jacobson, K. A. Trends Pharmacol. Sci. 2005, 26, 44.
26. Molecular Operating Environment (MOE 2007.09); C.C.G., Inc., 1255 University
St., Suite 1600, Montreal, Quebec, Canada H3B 3X3.
27. Tafi, A.; Bernardini, C.; Botta, M.; Corelli, F.; Andreini, M.; Martinelli, A.; Ortore,
G.; Baraldi, P. G.; Fruttarolo, F.; Borea, P. A.; Tuccinardi, T. J. Med. Chem. 2006,
49, 4085.
28. Moro, S.; Braiuca, P.; Deflorian, F.; Ferrari, C.; Pastorin, G.; Cacciari, B.; Baraldi,
P. G.; Varani, K.; Borea, P. A.; Spalluto, G. J. Med. Chem. 2005, 48, 152.
29. Stewart, J. J. J. Comput. Aided Mol. Des. 1990, 4, 1.
30. Okada, T.; Fujiyoshi, Y.; Silow, M.; Navarro, J.; Landau, E. M.; Shichida, Y. Proc.
Natl. Acad. Sci. U.S.A. 2002, 99, 5982.
31. Cornell, W. D.; Cieplak, P.; Bayly, C. I.; Gould, I. R.; Merz, K. M.; Ferguson, D. M.;
Spellmeyer, D. C.; Fox, T.; Caldwell, J. W.; Kollman, P. A. Am. Chem. Soc. 1995,
117, 5179.
Acknowledgment
32. Halgren, T. A. J. Comput. Chem. 1996, 17, 490.
33. Halgren, T. A. J. Comput. Chem. 1996, 17, 520.
34. Halgren, T. A. J. Comput. Chem. 1996, 17, 553.
35. Halgren, T. A. J. Comput. Chem. 1996, 17, 587.
The work was supported by a grant from the Italian Ministry for
University and Research (MIUR, FIRB RBNE03YA3L project).
36. Halgren, T. A.; Nachbar, R. J. Comput. Chem. 1996, 17, 616.
37. Halgren, T. A. J. Comput. Chem. 1999, 20, 720.
38. Halgren, T. A. J. Comput. Chem. 1999, 20, 730.
Supplementary data
39. Hou, T. J.; Xia, K.; Zhang, W.; Xu, X. J. J. Chem. Inf. Comput. Sci. 2004, 44,
266.
40. Lipinski, C. A.; Lombardo, F.; Dominy, B. W.; Feeney, P. J. Adv. Drug Deliv. Rev.
1997, 23, 3.
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2008.10.018.
41. Borea, P. A.; Dalpiaz, A.; Varani, K.; Gessi, S.; Gilli, G. Life Sci. 1996, 59,
1373.
References and notes
42. Varani, K.; Rigamonti, D.; Sipione, S.; Camurri, A.; Borea, P. A.; Cattabeni, F.;
Abbracchio, M. P.; Cattaneo, E. FASEB J. 2001, 15, 1245.
43. Varani, K.; Cacciari, B.; Baraldi, P. G.; Dionisotti, S.; Ongini, E.; Borea, P. A. Life
Sci. 1998, 63, 81.
44. Varani, K.; Gessi, S.; Merighi, S.; Vincenzi, F.; Cattabriga, E.; Benini, A.; Klotz, K.
N.; Baraldi, P. G.; Tabrizi, M. A.; Lennan, S. M.; Leung, E.; Borea, P. A. Biochem.
Pharmacol. 2005, 70, 1601.
45. Bradford, M. M. Anal. Biochem. 1976, 72, 248.
46. Cheng, Y.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22, 3099.
47. Munson, P. J.; Rodbard, D. Anal. Biochem. 1980, 107, 220.
1. Jacobson, K. A.; Knutsen, L. J. S.. In Handbook of Experimental Pharmacology:
Purinergic and Pyrimidinergic Signalling I; Abbracchio, M. P., Williams, M., Eds.;
Springer: Berlin, 2001; Vol. 151, p 129.
2. Fredholm, B. B.; Ijzerman, A. P.; Jacobson, K. A.; Klotz, K. N.; Linden, J.
Pharmacol. Rev. 2001, 53, 527.
3. Abbracchio, M. P.; Brambilla, R.; Ceruti, S.; Kim, H. O.; von Lubitz, D. K.;
Jacobson, K. A.; Cattabeni, F. Mol. Pharmacol. 1995, 48, 1038.
4. Ali, H.; Choi, O. H.; Fraundorfer, P. F.; Yamada, K.; Gonzaga, H. M.; Beaven, M. A.
J. Pharmacol. Exp. Ther. 1996, 276, 837.