1,2,4-Triazolo[1,5-a]quinoxaline as a versatile tool for the design of selective human A3 adenosine receptor antagonists: Synthesis, biological evaluation, and molecular modeling studies of 2-(hetero)aryl- and 2-carboxy-substitued derivatives

Article abstract of DOI:10.1021/jm0504149

A number of 4-oxo-substituted 1,2,4-triazolo[1,5-a]quinoxaline derivatives bearing at position-2 the claimed (hetero)aryl moiety (compounds 1-15) but also a carboxylate group (16-28, 32-36) or a hydrogen atom (29-31) were designed as human A₃ (

Full text of DOI:10.1021/jm0504149

7932
J. Med. Chem. 2005, 48, 7932-7945
1,2,4-Triazolo[1,5-a]quinoxaline as a Versatile Tool for the Design of Selective
Human A3 Adenosine Receptor Antagonists: Synthesis, Biological Evaluation,
and Molecular Modeling Studies of 2-(Hetero)aryl- and 2-Carboxy-Substitued
Derivatives
Daniela Catarzi,*,† Vittoria Colotta,† Flavia Varano,† Ombretta Lenzi,† Guido Filacchioni,† Letizia Trincavelli,‡
Claudia Martini,‡ Christian Montopoli,§ and Stefano Moro§
Dipartimento di Scienze Farmaceutiche, Universita` degli Studi di Firenze, Polo Scientifico, Via U. Schiff, 6-50019 Sesto
Fiorentino, Firenze, Italy, Dipartimento di Psichiatria, Neurobiologia, Farmacologia e Biotecnologie, Universita` degli Studi di
Pisa, Via Bonanno, 6-50126 Pisa, Italy, Molecular Modeling Section, Dipartimento di Scienze Farmaceutiche, Universita` degli
Studi di Padova, via Marzolo 5, 35131 Padova, Italy
Received May 2, 2005
A number of 4-oxo-substituted 1,2,4-triazolo[1,5-a]quinoxaline derivatives bearing at position-2
the claimed (hetero)aryl moiety (compounds 1-15) but also a carboxylate group (16-28, 32-
36) or a hydrogen atom (29-31) were designed as human A3 (hA3) adenosine receptor (AR)
antagonists. This study produced some interesting compounds and among them the 2-(4-
methoxyphenyl)-1,2,4-triazolo[1,5-a]quinoxalin-4-one (8), which can be considered one of the
most potent and selective hA3 adenosine receptor antagonists reported till now. Moreover, as
a new finding, replacement of the classical 2-(hetero)aryl moiety with a 2-carboxylate function
(compounds 16-28 and 32-36) maintained good hA3 AR binding activity but, most importantly
and interestingly, produced a large increase in hA3 versus hA1 selectivity. A receptor-based
SAR analysis provided new interesting insights about the steric and electrostatic requirements
that are important for the anchoring of these derivatives at the hA3 receptor recognition site,
thus highlighting the versatility of the triazoloquinoxaline scaffold for obtaining potent and
selective hA3 AR antagonists.
Introduction
are also useful in the treatment of cancer15,16 and
glaucoma.17 Other applications of both A1 and A3 AR
antagonists, such as prevention and treatment of cere-
bral and cardiac ischemic injuries, have been pro-
posed.14,18-20 Accordingly, taking into account the po-
tential of A3 adenosine receptor antagonists in the
development of prospective therapeutic agents,10,11 in
the past few years many attempts in our laboratory
have been made to explore structural requirements of
different classes of tricyclic heteroaromatic compounds
that could be important for the binding at each receptor
subtype.21-31
Adenosine, a regulatory autacoid that is generated as
a result of cellular injury or stress, interacts with at
least four specific G-protein-coupled receptor subtypes
(GPCRs) classified as A1, A2A, A2B, and A3.1,2 Interaction
of adenosine with its receptors mediates signal trans-
duction pathways, which utilize different effector sys-
tems including adenylate cyclase (A1, A2A, A2B, and A3
receptors), potassium and calcium channels (A1 recep-
tor), and phospholipase C (A1, A2B, and A3 receptors)
and D (A3 receptor).3-7
Among the four adenosine receptors (ARs), the A1 and
In recent papers, we reported the synthesis and
binding activities at bovine A1 (bA1) and A2A (bA2A) and
human cloned A1 (hA1), A2A (hA2A), and A3 (hA3) ARs of
some 4-amino-2-aryl-1,2,4-triazolo[1,5-a]quinoxaline de-
rivatives31 bearing different substituents on the 4-amino
group (Chart 1, Series A). Structure-activity relation-
ship studies (SAR) of these derivatives showed that, in
general, the presence of substituents on the 4-amino
group was important to increase A1 or A3 potency or
both. In particular, the nature of the substituent on the
4-amino group was essential to modulate the A1 and A3
receptor affinity. In fact, the presence of an acyl sub-
stituent on the 4-amino group afforded potent A3 AR
antagonists such as the 8-chloro-2-(4-methoxyphenyl)-
1,2,4-triazolo[1,5-a]quinoxaline-4-acetylamine,31 which
can be considered one of the most potent and hA3 versus
hA1 selective AR antagonists reported till now.
A
2A subtypes have been well pharmacologically charac-
terized, while both A2B and A3 receptors are the most
recently identified and, for this reason, are still under
study to better understand their physiological functions.
In fact, adenosine plays an important role in many
diseases of the central nervous system (CNS), as well
as in peripheral organs and tissues.8,9 Thus, the use of
AR antagonists as potential therapeutic agents has been
proposed.10,11 In particular, while A1 AR antagonists
have been developed as antihypertensives, potassium-
saving diuretics,12 and cognitive enhancers8,11-13 for
geriatric therapy and for the treatment of CNS disorders
such as Alzheimer’s disease,8 A3 AR antagonists are
sought as potential antiinflammatory agents,14 but they
* To whom correspondence should be addressed. Tel: +39 55
4573722. Fax: +39 55 4573780. E-mail: daniela.catarzi@unifi.it.
† Universita´ degli Studi di Firenze.
Therefore, to evaluate the importance of the presence
of a CdO proton acceptor at the level of position-4 of
‡ Universita` degli Studi di Pisa.
§ Universita` degli Studi di Padova.
10.1021/jm0504149 CCC: $30.25 © 2005 American Chemical Society
Published on Web 11/11/2005

Selective Human A3 Adenosine Receptor Antagonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25 7933
Chart 1. Previously and Currently Reported 1,2,4-Triazolo[1,5-a]quinoxaline Derivatives
Scheme 1a
the triazoloquinoxaline ring system, in the present work
we studied a series of 2-(hetero)aryl-1,2,4-triazolo[1,5-
a]quinoxaline derivatives (compounds 1-15, Chart 1)
bearing a 4-oxo function replacing the 4-amino group
of previously reported Series A.23,31 Moreover, with the
aim of shedding light on the SAR of this triazoloqui-
noxaline class also a series of 2-carboxy-substituted
compounds (16-28, 32-36), together with some tri-
azoloquinoxaline derivatives lacking a substituent at
position-2 (compounds 29-31) were designed. Both
these series can be considered the 4-oxo analogues of
previously described 4-amino derivatives29 (series B).
The modifications carried out on the triazoloquinoxaline
scaffold could be useful to further investigate position-2
of this tricyclic system. In fact, the diverse groups
introduced at this position are endowed with different
lipophilic and electronic properties and, consequently,
different capabilities to engage interactions with the
receptor site when compared to the 2-(hetero)aryl
groups. To our knowledge, an aryl moiety is always
present at position-2 in most of the structurally related
tricyclic AR antagonists reported in the litera-
ture10-12,14,32-34 with the exception of only a few ex-
amples.11
To elucidate our binding results, we decided to
theoretically depict the putative transmembrane (TM)
binding motif of the herein reported 1,2,4-triazolo[1,5-
a]quinoxalin-4-one derivatives on hA3 receptor. SARs
have been explained by analyzing the three-dimensional
structure of the antagonist-receptor models obtained
by molecular docking simulation.
Chemistry
The triazoloquinoxaline derivatives herein reported
were synthesized as illustrated in Schemes 1-5. Com-
pounds 26, 27,35,38 29,29 30,38 and 3238 (Tables 1 and 2)
were prepared following synthetic pathways previously
reported.
Scheme 1 shows the synthesis of 4,5-dihydro-2-(4-
a Reagents: (a) para-anisoylchloride, pyridine; (b) POCl3; (c)
NH3 (gas), anhydrous dioxane; (d) ClCOCOOEt, anhydrous tolu-
ene; (e) iron powder, glacial acetic acid.
methoxyphenyl)-8-methyl-1,2,4-triazolo[1,5-a]quinoxa-
lin-4-one 12, obtained following the procedures that
yielded the already described compounds 1-11.35,36
Thus, by reacting the known 5-methyl-2-nitrophenyl-
hydrazine35,37 with para-anisoyl chloride, we prepared
the hydrazide 37. The latter was transformed into the
corresponding chloroimine 38, which, when treated with
ammonia, gave the amidrazone 39. Reaction of com-
pound 39 with ethyloxalyl chloride directly yielded the
ethyl 1-(5-methyl-2-nitrophenyl)-3-(4-methoxyphenyl)-
1,2,4-triazole-5-carboxylate 40, which, after reduction
and contemporary cyclization, afforded the tricyclic
derivative 12.
By a method previously reported for obtaining com-
pound 1435 from 7, reaction of compounds 1 and 11 with

7934 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25
Catarzi et al.
Scheme 2a
Scheme 4a
a Reagents: (a) CH3I, NaH, DMF.
Scheme 3a
a Reagents: (a) CH3I, NaH, DMF; (b) ClCOOEt, K2CO3, acetone.
Scheme 5a
a Reagents: (a) NH3 (gas), EtOH; (b) SOCl2; (c) HOCH(CH3)2
or C6H5CH2OH, anhydrous tetrahydrofuran; (d) C6H5CH2NH2,
1-(3-dimethylaminopropyl)-3-ethylcarbodiimide hydrochloride, 1-hy-
droxybenzotriazole.
and 23.39 Thus, the diazonium salt of the commercially
available 5-methyl-2-nitroaniline was reacted with ethyl
2-chloro-3-oxobutanoate to yield the N2-chloroacetate 41,
which was transformed with ammonia into the corre-
sponding N2-oxamidrazonate 42. Reaction of 42 with
ethyloxalyl chloride afforded the N3-ethoxalyl derivative
43, which was cyclized to the ethyl triazole-3,5-dicar-
boxylate 44 by heating over its melting point. By
reduction of 44 with iron in glacial acetic acid, the
tricyclic derivative 18 was obtained. Compounds 19-
21 (Scheme 3) were obtained starting from the corre-
sponding benzo-unsubstituted derivatives 17 following
reported pathways.40,41
a Reagents: (a) NaNO2/HCl, CH3COCHClCOOEt; (b) NH3 (gas),
anhydrous dioxane; (c) ClCOCOOEt, anhydrous toluene, diethyl
ether; (d) 180 °C; (e) iron powder, glacial acetic acid.
methyl iodide produced the 5-N-methyl-substituted
derivatives 13 and 15 (Scheme 2).
Scheme 3 depicts the synthesis of the ethyl 4,5-
dihydro-4-oxo-1,2,4-triazolo[1,5-a]quinoxaline-2-carbox-
ylate 18, which was prepared following reported proce-
dures described to obtain derivatives 16,35 17,35,38 22,
The synthetic procedures that yielded compounds 24,
25, 28, and 31 are illustrated in Scheme 4. By reacting
compounds 16 and 17 with methyl iodide, we obtained
compounds 24 and 25, respectively, while reaction of
17 with ethyl chlorocarbonate in alkaline condition

Selective Human A3 Adenosine Receptor Antagonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25 7935
yielded the corresponding 5-N-ethoxycarbonyl-deriva-
tive 28. N-Methylation of the 2-unsubstituted compound
29 under the same conditions used to obtain 24 and 25
yielded 31.
Transformations of the 2-carboxylate function of 17
(derivatives 32-36) are reported in Scheme 5. By
treating 1735,38 with ammonia, we obtained the corre-
sponding 2-carboxamide compound 33 in good yield.
Reaction of 3238 with thionyl chloride gave the key
intermediate 4,5-dihydro-4-oxo-1,2,4-triazolo[1,5-a]qui-
noxaline-2-carbonyl chloride (not isolated), which was
treated with different reactants and conditions to give
the 2-carboxamide derivatives 34-36.
In general, all the 4-oxo derivatives, 1-15, are in-
active or active in the micromolar range at the bA2A
receptor. For this reason, the 4-amino group seems to
be essential for the interaction of these derivatives with
the bA2A subtype.26 Moreover, while most of the com-
pounds are active at the hA3 receptor in the low
micromolar range, the bA1 binding activities have, in
general, values higher than 1 µM. Among the 8-chloro-
substituted series (derivatives 1-6), all the compounds
follow this trend for the bA1 binding activity with the
exception of the 2-thienyl derivative 4, which is inactive.
These data suggest that the nature of the substituent
at position-2 of the triazoloquinoxaline ring system
influences just a little the bA1 receptor-ligand interac-
tion. In contrast, the presence of either the 2-furyl or
2-thienyl moiety at the same position positively affects
the bA2A binding affinity: in fact, 3 and 4 are two among
the only three compounds herein reported that are
endowed with a micromolar bA2A binding activity.
Moreover, the nature of the 2-substituent seems to be
essential in affecting markedly the activity at the hA3
receptor subtype; in fact, while the 2-(4-methoxyphenyl)
derivative 2 is 6-fold more active than the 2-phenyl-
substituted 1, introduction of the 2-furyl group at the
2-position (compound 3) produces a 30-fold ameliorated
hA3 binding affinity. A possible explanation for this
trend is that the oxygen of the furane ring engages a
hydrogen bond with a specific proton donor site on the
receptor; this hydrogen bond is less effective in the
corresponding 2-(3-furyl) derivative 5 showing a Ki value
comparable to that of the 2-phenyl-substituted deriva-
tive 1.
Biochemistry
Compounds 1-36 were tested for their ability to
displace [3H]N6-cyclohexyladenosine ([3H]CHA) from A1
AR in bovine cerebral cortical membranes, [3H]2-[4-(2-
carboxyethyl)phenethyl]amino-5′-(N-ethylcarbamoyl)ad-
enosine ([3H]CGS 21680) from A2A AR in bovine striatal
membranes, and [125I]N6-(4-amino-3-iodobenzyl)-5′-(N-
methylcarbamoyl)adenosine ([125I]AB-MECA) from cloned
hA3 receptor stably expressed in CHO cells. In fact, due
to the high species differences in A3 primary amino acid
sequence,42-44 we tested our A3 AR ligands on cloned
hA3 receptors.
Subsequently, some selected compounds (2, 3, 5, 8,
9, 10, 12, 17, 20, 28, 35, and 36) showing high hA3 AR
affinity (Ki < 100 nM) were tested for their ability to
displace [3H]CHA from cloned hA1 AR to establish their
A3 vs A1 selectivity within the same species. Moreover,
the most hA3 versus hA1 selective compounds (3, 8, 17,
20, 28, and 35) were also tested for their ability to
displace [3H]-5′-(N-ethylcarboxamido)adenosine ([3H]-
NECA) from cloned hA2A ARs stably expressed in CHO
cells. The binding results of 1-36, together with those
of compounds 1A, 2A, and 1B as comparison, are shown
in Tables 1-4. Moreover, the binding data of theophyl-
line and 1,3-dipropyl-8-cyclopentylxanthine (DPCPX),
included as antagonist reference compounds, are also
reported.
It is difficult to explain the low hA3 binding activity
of the 2-(2-thienyl) derivative 4 when compared to the
corresponding 2-(2-furyl) compound 3. In fact, the
binding result seems to indicate that the sulfur atom
of the 2-(2-thienyl) moiety, in contrast to the oxygen of
the 2-(2-furyl) group, is not capable of engaging a
hydrogen bond receptor-ligand interaction. The hA3
affinity of the 2-(3-thienyl) derivative 6 is in accordance
with that of compound 5.
Elimination of the 8-chloro atom, in general, positively
affects the bA1 affinity (compare compounds 1, 3, and 4
to 7, 9, and 10, respectively) the only exception being
the 2-(4-methoxyphenyl) derivative 8, which is com-
pletely inactive. In contrast, this modification does not
influence the bA2A binding activity, while it has diverse
effects on the hA3 binding depending on the nature of
the 2-substituent. In particular, when a phenyl or
2-furyl group is present at position 2, a 3-fold decrease
of the hA3 binding affinity is observed (compare com-
pounds 1 and 3 to 7 and 9, respectively), while an
ameliorated interaction is obtained when the 2-sub-
stituent is represented by a 4-methoxy-phenyl or a
2-thienyl group (compare derivatives 2 and 4 to 8 and
10, respectively). It is worth noting that elimination of
the 8-chloro substituent gives rise to the most potent
hA3 receptor antagonists herein reported, that is, com-
pounds 8 and 10.
Results and Discussion
The binding results reported in Tables 1-4 indicate
that we have produced some new potent and selective
A3 receptor antagonists belonging to the 1,2,4-triazolo-
[1,5-a]quinoxaline series. In particular, when the 8-chlo-
ro substitution pattern is held constant as in the lead
compound CGS 15943 (9-chloro-2-furyl-1,2,4-triazolo-
[1,5-c]quinazolin-5-amine),45 replacement of the 4-amino
group of compounds 1A and 2A23,31 with a 4-oxo function
(Table 1) abolishes the bA2A binding, while it reduces
to different degrees the bA1 and the hA3 affinities
depending on the nature of the 2-substituent (compare
compounds 1A and 2A to 1 and 2, respectively).
In the 4-amino series (compounds 1A and 2A), intro-
duction of the 4-methoxy substituent on the 2-phenyl
ring has little effect on both bA1 (3-fold decrease) and
bA2A binding activity (3-fold increase), while it positively
influences (30-fold increase) the hA3 affinity (compare
compound 2A to 1A).31 To a minor extent, the same
applies in the 4-oxo series: while this modification has
little or no effect on both bA1 and bA2A binding activity,
the hA3 affinity increases about 6-fold (compare 2 to 1).
Replacement of the 8-chlorine atom of 2 with an
8-methyl group produced a 10-fold increase of hA3 AR
affinity. In fact, the 2-(4-methoxyphenyl)-8-methyl-
substituted derivative 12 is one of the most active
compounds at the hA3 subtype among the 1,2,4-triazolo-

7936 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25
Catarzi et al.
Table 1. Binding Activity of 2-(Hetero)aryl-1,2,4-triazolo[1,5-a]quinoxalin-4-one Derivatives at Bovine A1 and A2A and Human A3
ARs
a The Ki values are means ( SEM of four separate assays, each performed in triplicate. b Displacement of specific [3H]CHA binding in
bovine brain membranes or percentage of inhibition (I%) of specific binding at 20 µM concentration except where stated otherwise.
c Displacement of specific [3H]CGS 21680 binding from bovine striatal membranes or percentage of inhibition (I%) of specific binding at
20 µM concentration except where stated otherwise. d Displacement of specific [125I]AB-MECA binding at human A3 receptors expressed
in CHO cells or percentage of inhibition (I%) of specific binding at 1 µM concentration. e Reference 31. f Percentage of inhibition (I%) of
specific binding at 34 µM concentration.
[1,5-a]quinoxaline series. In contrast, this modification
maintained both bA1 and bA2A binding activity un-
changed.
ero)aryl moiety of compounds 1-15 (Table 1) with an
ethyl carboxylate group (Table 2). The improved hA3
binding activity of the 2-(2-furyl) derivatives 3 and 9
compared to the corresponding 2-phenyl ones 1 and 7,
respectively, has suggested this modification: in fact,
an ethyl carboxylate substituent could be capable of
giving the hydrogen bond interaction with a receptor
site hypothesized above for the 2-furyl moiety. Thus,
replacing the 2-phenyl group of 1 with an ethyl car-
boxylate group yields compound 16, which is, without
any doubt, less active at the hA3 receptor than 1 (Table
2). Nevertheless, it has to be noted that while the
2-carboxyethyl-4-imino derivative 1B29 is completely
When the 2-phenyl substituent on the triazoloqui-
noxaline nucleus is kept constant, introduction of a
methyl group at position 5 (derivatives 13-15) does not
influence the bA2A binding activity, while it reduces both
the bA1 and hA3 affinities to different extents; the only
exception is represented by compound 15, which is
equiactive to the corresponding des-methyl derivative
11 at the bA1 receptor subtype.
In pursuing our studies on the 1,2,4-triazolo[1,5-a]-
quinoxaline ring system, we have replaced the 2-(het-

Selective Human A3 Adenosine Receptor Antagonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25 7937
Table 2. Binding Activity of Ethyl 2-Carboxy-Substituted Trizoloquinoxaline Derivatives at Bovine A1 and A2A and Human A3 ARs
Ki (nM)a or I%
b
c
d
compd
R2
R5
R7
R8
bA1
bA2A
hA3
1Be
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
5 ( 0.7
20%
4900 ( 396
7%f
15%
557 ( 38
70.8 ( 4.5
108 ( 8.9
0%
25 ( 11
673 ( 58
340 (26
354 ( 19
60%
820 ( 54
0%
20%
28.2
COOEt
COOEt
COOEt
COOEt
COOEt
COOEt
COOEt
COOEt
COOEt
COOEt
COOEt
COOEt
COOEt
H
H
H
H
H
Cl
H
26%
3.6%
6.5%
7.4%
32.7%
8.8%
15.2%
9%
H
H
CH3
41%
H
H
NO2
58%g
64%
H
H
NH2
H
H
NHCOCH3
14170 ( 902
48%
H
Cl
NH2
H
H
H
Cl
H
H
H
H
Cl
H
Cl
H
43%
8%
CH3
CH3
C2H5
OH
COOEt
H
0%
H
6.4%
13.3%f
6.9%f
14%
9.2%
20.5%f
2%f
H
H
H
4%
H
0%
23%
18.8%
53%
0%
H
H
CH3
H
29%
20%
H
H
11%
8.9%
a The Ki values are means ( SEM of four separate assays, each performed in triplicate. b Displacement of specific [3H]CHA binding in
bovine brain membranes or percentage of inhibition (I%) of specific binding at 20 µM concentration except where stated otherwise.
c Displacement of specific [3H]CGS 21680 binding from bovine striatal membranes or percentage of inhibition (I%) of specific binding at
20 µM concentration except where stated otherwise. d Displacement of specific [125I]AB-MECA binding at human A3 receptors expressed
in CHO cells or percentage of inhibition (I%) of specific binding at 1 µM concentration. e Reference 29. f Percentage of inhibition (I%) of
specific binding at 34 µM concentration. g Percentage of inhibition (I%) of specific binding at 10 µM concentration.
inactive at the hA3 receptor, the corresponding 4-oxo
compound 16 showed a binding activity in the micro-
molar range. Moreover, the latter shows no affinity for
either bA1 or bA2A AR; this is the trend of all the ethyl
2-carboxylate derivatives 16-28, which are inactive or
very poorly active (compound 21) at both these receptor
subtypes. In contrast to derivative 16, some of the
herein reported compounds have a good hA3 binding
activity, in general depending on the nature of the
substituent on the benzo-fused moiety. Elimination of
the chlorine atom at position 8 increases the hA3 activity
(compare 16 to 17), while movement from position 8 to
7 maintains the Ki value unchanged (compare compound
16 to 22), thus suggesting that the presence of an
electron-withdrawing substituent is not profitable for
anchoring at the hA3 receptor site. This hypothesis may
find confirmation in the total absence of hA3 activity of
compound 19, which bears an 8-nitro group. In contrast,
introduction at position 8 of electron-donating substit-
uents does not affect (methyl group) or 3-fold increases
(amino group) the hA3 binding affinity (compare com-
pound 17 to 18 and 20). It is possible that the 8-amino
group plays a role in the receptor-ligand interaction
by engaging a hydrogen bond with a receptor site. This
hypothesis can be supported by the 14-fold reduction
in potency that can be observed when the amino group
is moved from position 8 to 7 (compound 23). Introduc-
tion of an acetyl group on the 8-amino moiety (compound
21), though increasing the 8-NH acidity, strongly de-
creases the hA3 binding activity; this result could
suggest the existence of a receptor pocket with a limited
steric tolerance.
As observed above for the 2-aryl-substituted series
(Table 1), alkylation of the NH at position 5 causes a
dramatic reduction in hA3 potency (compounds 24-26),
which however falls into the micromolar range for
compound 25. Also introduction of the hydrophilic
hydroxy group at the same position (compound 27)
completely abolishes the hA3 receptor affinity. In con-
trast, introduction of a 5-ethyl carboxylate group (com-
pound 28) is profitable for hA3 receptor-ligand inter-
action: in fact, compound 28 is one of the most active
hA3 receptor antagonists among the 2-carboxy-substi-
tuted derivatives, 16-28 and 32-36.
The importance of the presence of a (hetero)aryl or
an ethyl carboxylate group at position 2 for the interac-
tion of these 1,2,4-triazolo[1,5-a]quinoxaline derivatives
with all three ARs is suggested by the total inactivity
of compounds 29 and 30, which are devoid of the
2-substituent.
Further results regarding investigation of the 2-posi-
tion of the triazoloquinoxaline moiety are reported in
Table 3. Transformation of the ethyl ester (compound
17) into the corresponding carboxylic acid (compound
32) or carboxamide (compound 33) produces a total
collapse of the hA3 binding affinity, while trans-esteri-
fication to the isopropyl or benzyl ester (derivatives 34
and 35) yields, respectively, a 6-fold reduction or a
preservation of potency. A good hA3 affinity is also
observed for the benzylamide 36, which can be consid-
ered a bioisoster of the ester 35. These data may
indicate that the pocket that binds the substituent at
position 2 has precise steric requirements, the isopropyl

7938 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25
Catarzi et al.
pound.31 It is particularly evident (Table 4) that all the
compounds tested on hA2A receptors (3, 8, 17, 20, 28,
and 35) were completely inactive at this AR subtype and
thus highly hA3 versus hA2A selective. In contrast, the
hA1 binding results indicate that most of the 2-(hetero)-
aryl compounds (2, 5, 10, and 12) are nonselective or
scarcely selective hA3 versus hA1 receptor antagonists.
The only exceptions are represented by the 8-chloro-2-
furyl-derivative 3 and the 2-(4-methoxyphenyl) com-
pound 8. In particular, the latter is inactive at both the
hA1 and hA2A AR, and consequently, it possesses a high
hA3 versus hA1 and hA2A selectivity. Thus, the 2-(4-
methoxyphenyl)-1,2,4-triazolo[1,5-a]quinoxalin-4-one (8)
can be considered one of the most potent selective AR
antagonists reported till now.
Moreover, it has to be noted that all the tested
2-carboxy-substituted compounds, 17, 20, 28, and 35,
are highly hA3 versus hA1 selective with the exception
of the 2-benzyl-carboxamide 36. This result is particu-
larly important considering that introduction of a 2-car-
boxylate function on the tricyclic heteroaromatic system
of similar size and shape is an original approach to
obtain AR antagonists.29
In conclusion, the SARs of the 2-heteroaryl-triazolo-
quinoxalin-4-one derivatives studied here are in ac-
cordance with those of previously reported series.26-28,30
In particular, the importance of a methoxy group in the
para position on the 2-phenyl ring for increasing potency
and, in some cases, selectivity toward the hA3 AR was
confirmed.26,27,31 Most important and intriguing, re-
placement of the 2-(hetero)aryl moiety with a 2-car-
boxylate function produces a large increase of hA3
versus hA1 selectivity by maintaining a good hA3 AR
binding activity. Moreover, it has to be noted that the
triazoloquinoxaline framework was utilized for the
synthesis of a large number of previously reported Gly/
N-methyl-D-aspartate (NMDA) or R-amino-3-hydroxy-
5-methyl-4-isoxazolepropionic acid (AMPA) receptor
antagonists.39-41 Thus, we can consider this tricyclic
ring system as a potential multitarget drug scaffold.
Table 3. Binding Activity of 2-Carboxy-Substituted
Triazoloquinoxaline Derivatives at Bovine A1 and A2A and
Human A3 ARs
Ki (nM)a or I%
b
c
d
compd
R2
bA1
bA2A
hA3
17
32
33
34
35
36
COOEt
COOH
26%
10%e
3.6%
0%e
17%
2.9%
47%
27%
70.8 ( 5.8
6.4%
CONH2
15%
32%
COOCH(CH3)2
COOCH2C6H5
CONHCH2C6H5
0%
3220 ( 250
578 ( 71
420 ( 23
51.6 ( 3.9
82 ( 6.3
a The Ki values are means ( SEM of four separate assays, each
performed in triplicate. b Displacement of specific [3H]CHA binding
in bovine brain membranes or percentage of inhibition (I%) of
specific binding at 20 µM concentration. c Displacement of specific
[3H]CGS 21680 binding from bovine striatal membranes or
percentage of inhibition (I%) of specific binding at 20 µM concen-
tration. d Displacement of specific [125I]AB-MECA binding at hu-
man A3 receptors expressed in CHO cells or percentage of
inhibition (I%) of specific binding at 1 µM concentration. e Per-
centage of inhibition (I%) of specific binding at 34 µM concentra-
tion.
Table 4. Binding Activity of Some Selected
Triazoloquinoxaline Derivatives at Human A1 and A2A ARs
hA3 vs hA1
hA3 vs hA2A
selectivity
ratio
Ki (nM)a or selectivity Ki (nM)a or
b
c
compd
I% hA1
ratio
I% hA2A
2Ad
2
36 ( 2.8
3.29 ( 0.22
2740 ( 128
124.7 ( 9.8
32%
12.4
0.13
456.7
1.6
.4700
58
21%
.3400
.1700
>4700
3
33%
49%
5
8
9
1926 ( 13
2 ( 0.1
2.86 ( 0.17
5%
10
12
17
20
28
35
36
1
1.1
.150
.400
.350
.200
2.5
0.07
0.0025
0%
22%
9%
.150
.400
.350
.200
33%
Structure-activity relationships of these triazoloqui-
noxaline derivatives were explained by analyzing the
three-dimensional structure of the antagonist-receptor
models obtained by molecular docking simulation.
12%
33%
27%
205 ( 16
theophylline 6200 ( 530
DPCPX
7900 ( 630
260 ( 18
0.092
0.2
3.2 ( 0.2
We recently proposed, using a computational ap-
proach, the putative TM binding motif of 1,2,4-triazolo-
[4,3-a]quinoxaline analogues30 and 8-chloro-substituted
1,2,4-triazolo[1,5-a]quinoxaline family31 on hA3 receptor.
As previously reported, the recognition of classic hA3
AR antagonists seems to occur in the upper region of
the TM helical bundle. In particular, we described that
TM domains 3, 5, 6, and 7 seem to be crucial for the
recognition of both agonists and antagonists.46,47 As
found for other classes of hA3 receptor antagonists, both
quinoxaline families30,31 fit nicely inside the TM region
of our proposed hA3 receptor model.30 To complete the
validation of our receptor model, molecular modeling
studies have been carried out for all the compounds
reported in Tables 1-3. Consistent with our previously
published docking studies on 4-amino-1,2,4-triazolo[1,5-
a]quinoxaline derivatives, the binding of the corre-
sponding 1,2,4-triazolo[1,5-a]quinoxalin-4-one moiety
seems to occur in the same region of the helical bundle.
A reasonable steric and electrostatic complementarity
a The Ki values are means ( SEM of four separate assays, each
performed in triplicate. b Displacement of specific [3H]CHA binding
in cloned hA1 receptors expressed in CHO cells or percentage of
inhibition (I%) of specific binding at 10 µM concentration. c Dis-
placement of specific [3H]NECA binding in cloned hA2A receptors
expressed in CHO cells or percentage of inhibition (I%) of specific
binding at 10 µM concentration. d Reference 31.
ester 34 being 6-fold less active than the corresponding
benzylic one 35.
Introduction of hindering substituents at position 2
(compounds 35 and 36) also produces an increase in the
bA1 binding activity, while other modifications yield
derivatives 32-34, which are completely inactive. All
the compounds reported in Table 3 do not bind at the
bA2A receptor.
To evaluate the A3 versus A1 and A2A selectivity
among the same species, we tested some selected
compounds on both hA1 and hA2A AR, together with the
corresponding 4-amino-derivative 2A as reference com-

Selective Human A3 Adenosine Receptor Antagonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25 7939
Figure 1. Binding site of the new triazoloquinoxaline derivatives in the human A3 receptor: derivatives 3 (upper left), 8 (upper
right), 10 (bottom left), and 17 (bottom right) docked into the ligand binding crevice of the human A3 receptor viewed from the
membrane side facing TM helices 5 and 6. TM hydrogen atoms are not displayed. Receptor domains and TM amino acids are
labeled only for derivative 3 (upper left).
has been found among all analyzed derivatives and the
hypothetical binding cavity on hA3 receptor model.
However, based on our model, derivatives 3, 8 10, and
17 present the best fitting into the hA3 binding cleft
(Figure 1). It is particularly interesting that also in the
new 1,2,4-triazolo[1,5-a]quinoxaline-4-one series at least
four stabilizing hydrogen-bonding interactions can be
observed in all energetically stable docked conforma-
tions. Thr94 (TM3), His95 (TM3), Ser170 (EL2), and
Asn250 (TM6) seem to characterize the ligand recogni-
tion region in the receptor cavity, as shown in Figure
1. In particular, the novel replacement of the 4-amino
with the 4-oxo moiety seems to be generally accepted
even if it is crucial for differentiating both potency and
affinity among this series of new A3 AR antagonists.
Analyzing our structure-activity data in detail, we find
that the 4-oxo position is surrounded by two polar amino
acids, Thr94 (TM3) and His95 (TM3). It is assumed that
these two polar amino acids give stabilizing interactions
with the 4-oxo moiety of all new synthesized compounds
through hydrogen bonding.
scaffold slightly shifts its position inside the TM cleft
to optimize both steric and electrostatic complementa-
rity. As previously described,30,31 the 2-aryl and 2-het-
eroaryl substituents are always positioned in a rela-
tively small cleft between TM2 and TM7. In our model,
the furan ring of derivative 3 sits nicely in this receptor
cavity forming a stabilizing interaction between the
oxygen atom of the furan ring and the NH2 of Gln167
(EL2).
However, the corresponding analogue with the p-
methoxy group at the 2-aryl position (derivative 2 in
Table 1) has slightly reduced activity because it loses
its steric complementarity with the TM receptor cavity.
Accordingly, the removal of the chlorine atom from
position 8 (derivative 8 in Table 1) restores the good
antagonist-receptor complementarity. Moreover, con-
sidering the 2-aryl family, an additional hydrogen-
bonding interaction seems occur between the 4-methoxy
group of the 2-aryl substituent and the hydroxy group
of Ser165 positioned on the EL2 domain, as previously
reported.30,31 This supplementary interaction could
explain the increase of affinity experimentally observed
when the para position of the 2-aryl substituent is
replaced with a methoxy group, when other unfavorable
steric or electrostatic interactions or both do not occur
(compare, for example, derivatives 1A and 2A, 1 and 2,
and 7 and 8 in Table 1).
Considering the other general structural binding
requirements, the triazoloquinoxaline nucleus should be
most favorably oriented perpendicular to the plane of
the lipid bilayer with the 2-aryl (or 2-heteroaryl) sub-
stituent in proximity to TM2 and TM7 and the 8-chlo-
robenzene moiety close to TM5 and TM6. Depending on
the presence and, consequently, chemical nature of the
substituents in position 7 and 8, the triazoloquinoxaline
Depending on the nature and on the position of the
substituents (R2, R5, and R8 in Table 1), the balancing

7940 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25
Catarzi et al.
of steric and electrostatic control can also influence the
activity of the 2-heteroaryl derivatives. The fine struc-
tural control imposed by the receptor binding cleft can
be observed when, in the presence of a chlorine atom
at position-8, the 2-furan moiety was replaced with a
phenyl or thienyl substituent: the stabilizing inter-
action with Gln167 disappears or drastically reduces its
importance. In the case of the 2-(2-furyl)-derivatives 3
and 9, we identified a favorable dipole-dipole inter-
action between the furan ring and the amide group of
the Gln167 side chain. This stabilizing interaction is
missing in both 2-(2-thienyl) (compounds 4 and 10) and
2-phenyl (derivatives 1 and 7) modified antagonists.
This similar chemical control can operate also when we
consider the 2-(3-furyl)-substituted compound. In fact,
depending on the link position of the furan ring to the
triazoloquinoxaline moiety (position 2 or 3), the total
dipole moment of the molecule can change in both
intensity and orientation.
Another interesting observation is that 2-carboxy-
substituited triazoloquinoxalines are still tolerated by
the TM receptor cavity, even if we always detected a
reduction of their activities compared with the corre-
sponding 2-furan or 2-(4-methoxyphenyl) derivatives
(see Tables 2 and 3). As shown in Figure 1, following
our docking simulations, the ethyl ester moiety roughly
maintains the 2-aryl (or 2-heteroaryl) substituent posi-
tion inside the TM recognition cavity. The most notable
structural differences among these 2-carboxy-substitu-
ited antagonists with respect to all the other 2-aryl or
2-heteroaryl derivatives is the loss of substituent pla-
narity at position 2. This fact inevitably introduces an
unfavorable entropic control.
Moreover, also in this case depending on the nature
and on the position of the substituents (R2, R5, R7, and
R8 in Tables 2 and 3), the balancing of steric and
electrostatic control can profoundly influence the activ-
ity of these new 2-carboxy-substituited derivatives.
Fine-tuning of the observed activity differences among
the new 2-carboxy-substituited antagonists seems to be
effected by the presence of a substituent at position-8.
Indeed, the activity of all 8-chlororinated, or vice versa
of all 8-deschlorinated, antagonists is dependent on the
chemical nature of the substituent at position 2. In fact,
the triazoloquinoxaline scaffold can slightly shift its
position inside the TM cleft to optimize both steric and
electrostatic complementarity, depending on both the
presence and the chemical nature of the substituents
at positions 2 and 8. In particular, the case of the
8-chloro-substituted derivatives has been analyzed. In
the 8-chloro-2-(hetero)aryl family, a stabilizing dipolar
interaction between the chlorine substituent and the
hydroxy group of Ser170 (EL2) occurs. To preserve this
interaction, the 8-chloro derivatives have to shift their
position inside the receptor cleft, reducing the stabilizing
effect of the dipole interaction between the Gln167 (EL2)
and the substituent at position 2. However, this supple-
mentary interaction makes up for this destabilizing
effect and explains the increase of affinity observed in
the 8-chloro-2-(hetero)aryl derivatives 1 and 3 with
respect to the corresponding deschloro compounds 7 and
9. In contrast, in the 2-carboxy-substituted series re-
moval of the 8-chlorine atom produces an increase of
binding affinitiy (compare 16 to 17). The loss of 2-sub-
Figure 2. Triazoloquinoxaline derivatives 7, 1, and 17 and
their corresponding N-methyl derivatives, 14, 13, and 25.
Binding activity at hA3 receptor is also shown.
stituent planarity and the corresponding increase of its
steric hindrance does not guarantee that the ethyl ester
preserves the stabilizing interaction with Gln167 (EL2).
In fact, the best docked 8-chloro compound 16 shifts its
position significantly inside the binding cleft, thus
reducing its steric and electrostatic complementarity.
As illustrated in Table 3, the chlorine atom at position
8 can be substituted only with other relatively small
polar substituents (compare derivatives 16-21, Table
3) without total loss of their binding activities. In effect,
looking inside the receptor cleft, position 8 is accom-
modated in a tiny cavity near TM4 and TM5 (see Figure
1).
Interestingly, derivative 28 (R5 ) COOEt, Table 3)
presents an increased A3 receptor affinity compared
with derivatives 25 (R5 ) CH3, Table 3) or 17 (R5 ) H,
Table 3). In this case, our docking studies have shown
two additional stabilizing interactions among the car-
boxylate moiety at position-5 and the two polar amino
acids, Thr94 (TM3) and His95 (TM3).
Finally, a very peculiar effect was observed when
methylation of the N-5 position occurred. As described
in Tables 1 and 2, alkylation of the NH at position 5
causes a dramatic reduction in hA3 potency as sum-
marized in Figure 2. This peculiar effect is very am-
biguous, and two reasonable explanations can be pro-
posed. The first hypothesis is a severe steric hindrance
control in proximity of the NH at position 5. In fact,
Ile186 (TM5) becomes closer to the van der Waals
contact distance with the methyl group of all 5-N-
methylated derivatives. The second intriguing hypoth-
esis is related to a possible “tautomeric” selection
brought about by the receptor binding site. All 4-oxo
compounds can formally exit in two different tau-
tomers: the keto and the enol forms. Considering their
chemical stability, in water solution all keto forms are
8-10 kcal/mol more stable with respect to the corre-
sponding enol forms (data not shown; see Experimental
Section for details). However, as shown by docking
simulations, enol forms fit inside the TM binding cavity
equally well or better than keto forms (see Figure 3,

Selective Human A3 Adenosine Receptor Antagonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25 7941
Figure 3. The left panel shows triazoloquinoxaline derivative 1 and its possible enolic form docked into the ligand binding
crevice of the human A3 receptor viewed from the membrane side facing TM helices 5 and 6. The right panel shows the best
docking conformation of the enolic form of triazoloquinoxaline derivative 1 compared to CGS 15943 docked into the ligand binding
crevice of the human A3 receptor.
mined on a Gallenkamp melting point apparatus. Microanaly-
ses were performed with a Perkin-Elmer 260 elemental
analyzer for C, H, and N, and the results were within (0.4%
of the theoretical values except where stated otherwise (Sup-
porting Information). The IR spectra were recorded with a
Perkin-Elmer 1420 spectrometer in Nujol mulls and are
expressed in cm-1. The 1H NMR spectra were obtained with a
Varian Gemini 200 instrument at 200 MHz. The chemical
shifts are reported in δ (ppm) and are relative to the central
peak of the solvent. All the exchangeable protons were
confirmed by addition of D2O. The following abbreviations are
used: s ) singlet, d ) doublet, dd ) double doublet, t ) triplet,
q ) quartet, m ) multiplet, br ) broad, and ar ) aromatic
protons.
N1-(5-Methyl-2-nitrophenyl)-N2-(4-methoxybenzoyl)-
hydrazide (37). A solution of para-anisoyl chloride (7.77
mmol) in anhydrous pyridine (3 mL) was added dropwise at
room temperature to a solution of 5-methyl-2-nitrophenylhy-
drazine35,37 (5.98 mmol) in anhydrous pyridine (18 mL). The
reaction mixture was heated at reflux for 1 h. Distillation of
the solvent under reduced pressure gave an orange solid, which
was treated with 2 M HCl (8 mL), collected under vacuum,
and washed with water. Yield 89%; mp 182-184°C (ethanol);
H1 NMR (DMSO-d6) δ 2.28 (s, 3H, CH3), 3.83 (s, 3H, OCH3),
6.68 (d, 1H, ar, J ) 8.79 Hz), 6.93 (s, 1H, ar), 7.06 (d, 2H, ar,
J ) 8.79 Hz), 7.94 (d, 2H, J ) 8.79 Hz), 8.02 (d, 1H, ar, J )
8.79 Hz), 9.41 (s, 1H, NH), 10.63 (s, 1H, NH); IR 3340, 3250,
1645 cm-1. Anal. (C15H15N3O4) C, H, N.
N1-(5-Methyl-2-nitrophenyl)-N2-[R-chloro-(4-methoxy-
benzyliden)]hydrazine (38). A suspension of 37 (4.31 mmol)
in POCl3 (12 mL) was heated at reflux for 1 h. The excess
POCl3 was removed under reduced pressure, and the red
residue was treated with cyclohexane (30 mL). Evaporation
of the solvent under reduced pressure gave a mull, which was
treated with cyclohexane/ethyl acetate 1:1. The orange solid
obtained was collected by filtration and recrystallized. Yield
60%; mp 125-126°C (cyclohexane/ethyl acetate); H1 NMR
(DMSO-d6) δ 2.42 (s, 3H, CH3), 3.84 (s, 3H, OCH3), 6.85 (d,
1H, ar, J ) 8.79 Hz), 7.06 (d, 2H, ar, J ) 8.79 Hz), 7.71 (s, 1H,
derivative 1 and its enol form). The best docked enol
form binding arrangement is superimposable on the
keto form conformation, but a stronger interaction
among the 4-enol moiety and the two polar amino acids,
Thr94 (TM3) and His95 (TM3), occurrs. The enol form
of the 4-oxo family is chemically equivalent to the
4-amino derivatives, such as CGS 15943 or 2A. In fact,
as reported in Figure 3, the 4-OH group of the docked
enol form maintains the 4-NH2 substituent position and
also forms similar interactions inside the TM recogni-
tion cavity. Obviously the 5-N-methylated derivatives
do not engage in a tautomeric equilibrium. Further
investigations are in progress in our laboratories to
support this tautomer selection.
Conclusions
The present study has pointed out that the 1,2,4-
triazolo[1,5-a]quinoxaline moiety is a versatile tool for
the design of selective hA3 AR antagonists. The classical
SAR analysis, supported by molecular docking simula-
tion, confirmed previous findings and also provides new
interesting insights about the steric and electrostatic
requirements that are important for the optimal an-
choring of these triazoloquinoxaline derivatives at the
hA3 receptor recognition site. In fact, the importance of
the 2-(hetero)aryl group was confirmed, and as a new
finding, the 2-carboxylate function has emerged as a
possible alternative to the classical 2-(hetero)aryl moiety
for obtaining potent tricyclic hA3 AR antagonists of
similar size and shape. For this reason, further modi-
fications of these derivatives are planned.
Experimental Section
Chemistry. Silica gel plates (Merck F254) and silica gel
60 (Merck; 70-230 mesh) were used for analytical and column
chromatography, respectively. All melting points were deter-

7942 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25
Catarzi et al.
ar) 7.94 (d, 2H, J ) 8.79 Hz), 8.06 (d, 1H, ar, J ) 8.79 Hz),
11.16 (s, 1H, NH); IR 3290, 1610 cm-1. Anal. (C15H14ClN3O3)
C, H, N.
Hz), 7.56 (s, 1H, ar), 8.10 (d, 1H, ar, J ) 8.35), 11.17 (s, 1H,
NH); IR 3260, 1720 cm-1. Anal. (C11H12ClN3O4) C, H, N.
Ethyl N1-(5-Methyl-2-nitrophenyl)-N2-oxamidrazonate
(42). Ammonia was bubbled until saturation into a stirred
solution of 41 (1.40 mmol) in anhydrous dioxane (18 mL). The
mixture was stirred at room temperature until the disappear-
ance of the starting material and then was diluted with water
(30 mL) to yield a red solid, which was collected and washed
with water. Yield 80%; mp 153-155 °C (cyclohexane); 1H NMR
(DMSO-d6) δ 1.28 (t, 3H, CH3, J ) 6.90 Hz), 2.33 (s, 3H, CH3),
4.28 (q, 2H, CH2, J ) 6.90 Hz), 6.62 (s, 2H, NH2), 6.70 (d, 1H,
ar, J ) 8.40 Hz), 7.51 (s, 1H, ar), 7.98 (d, 1H, ar, J ) 8.40 Hz),
10.07 (s, 1H, NH); IR 3470, 3380, 3300, 1700 cm-1. Anal.
(C11H14N4O4) C, H, N.
Ethyl N1-(5-Methyl-2-nitrophenyl)-N3-ethoxalyl-N2-
oxamidrazonate (43). A solution of ethyloxalyl chloride (2.02
mmol) in anhydrous diethyl ether (2 mL) was slowly added to
a suspension of the amidrazonate 42 (1.01 mmol) in anhydrous
diethyl ether (2 mL). The reaction mixture was diluted with
anhydrous toluene (4 mL) and then heated at reflux for 45
min. Upon cooling a solid precipitated, which was collected
and washed with petroleum ether. Yield 73%; mp 150-152
°C (toluene); 1H NMR (DMSO-d6) δ 1.21-1.33 (m, 6H, 2CH3),
2.39 (s, 3H, CH3), 4.19-4.37 (m, 4H, 2CH2), 6.92 (d, 1H, ar, J
) 8.20 Hz), 7.97 (s, 1H, ar), 7.98 (d, 1H, ar, J ) 8.20 Hz), 10.90
(s, 1H, NH); 13.14 (s, 1H, NH); IR 3370, 3310, 1700 cm-1. Anal.
(C15H18N4O7) C, H, N.
N1-(5-Methyl-2-nitrophenyl)-N2-[R-amino-(4-methoxy-
benzylidene)]hydrazine (39). Ammonia was bubbled until
saturation into a stirred solution of 38 (1.69 mmol) in
anhydrous dioxane (20 mL). The mixture was stirred at room
temperature for 1 h and then was diluted with water (30 mL)
to yield a dark red solid, which was collected and washed with
1
water. Yield 81%; mp 130-132 °C (ethanol/water); H NMR
(DMSO-d6) δ 2.32 (s, 3H, CH3), 3.80 (s, 3H, OCH3), 6.52-6.65
(m, 3H, NH2 + ar), 6.99 (d, 2H, ar, J ) 8.79 Hz), 7.48 (s, 1H,
ar), 7.85 (d, 2H, ar, J ) 8.79 Hz), 7.94 (d, 1H ar, J ) 8.79 Hz),
10.27 (s, 1H, NH); IR 3440, 3360, 3300 cm-1. Anal. (C15H16N4O3)
C, H, N.
Ethyl 1-(5-Methyl-2-nitrophenyl)-3-(4-methoxyphenyl)-
1,2,4-triazolo-5-carboxylate (40). The amidrazonate 39 (0.94
mmol) was slowly added in small portions to a solution of
ethyloxalyl chloride (2.97 mmol) in anhydrous toluene (20 mL)
at 80 °C. The reaction mixture was heated at reflux for 3 h.
Distillation of the solvent under reduced pressure gave an oil,
which was treated with a little ethyl acetate; a yellow solid
precipitates, which was collected and washed with diethyl
ether. Yield 48%; mp 160-162 °C (toluene); 1H NMR (DMSO-
d6) δ 1.15 (t, 3H, CH3, J ) 7.11 Hz), 2.49 (s, 3H, CH3), 3.81 (s,
3H, CH3), 4.24 (q, 2H, CH2, J ) 7.11 Hz), 7.06 (d, 2H, ar, J )
8.79 Hz), 7.68 (d, 1H, ar, J ) 8.42 Hz), 7.76 (s, 1H, ar), 7.98
(d, 2H, ar, J ) 8.79 Hz), 8.23 (d, 1H, ar, J ) 8.42 Hz); IR 1730,
1620 cm-1. Anal. (C19H18N4O5) C, H, N.
4,5-Dihydro-8-methyl-2-(4-methoxyphenyl)-1,2,4-tria-
zolo[1,5-a]quinoxalin-4-one (12). Iron powder (1.0 g) was
added to a suspension of 40 (1.0 mmol) in glacial acetic acid
(6 mL). The mixture was heated at 90 °C for 10 min.
Evaporation at reduced pressure of the solvent yielded a
residue, which was suspended in water (50 mL) and then
bleached with 6 N HCl. The solid was collected by filtration,
washed with water, dried, and extracted in soxlhet with
acetone (250 mL). Evaporation of the solvent under reduced
pressure yielded a solid, which was worked up with diethyl
ether (10 mL) and collected by filtration. Yield 70%; mp > 300
°C (acetone); 1H NMR (DMSO-d6) δ 2.07 (s, 3H, CH3), 3.84 (s,
3H, OCH3), 7.12 (d, 2H, ar, J ) 8.79 Hz), 7.34 (s, 2H, ar), 7.95
(s, 1H, ar), 8.15 (d, 2H, J ) 8.42 Hz), 12.30 (s, 1H, NH); IR
3160, 1680 cm-1. Anal. (C17H14N4O2) C, H, N.
Diethyl 1-(5-Methyl-2-nitrophenyl)-1,2,4-triazolo-3,5-
dicarboxylate (44). The finely powdered ethoxalyl derivative
43 (0.68 mmol) was heated at 180 °C for 45 min. The cooled
mass was dissolved in chloroform (20 mL), and the organic
solvent washed with a diluted (0.5 N) solution of potassium
hydroxide (15 mL × 3) and then with water (15 mL). Evapora-
tion under reduced pressure of the dried (anhydrous sodium
sulfate) organic solvent afforded a residue, which was treated
with a little cyclohexane and collected by filtration. Yield 53%;
1
mp 134-136 °C (cyclohexane); H NMR (DMSO-d6) δ 1.14 (t,
3H, CH3, J ) 7.00 Hz), 1.31 (t, 3H, CH3, J ) 7.00 Hz), 2.47(s,
3H, CH3), 4.22 (q, 2H, CH2, J ) 7.00 Hz), 4.38 (q, 2H, CH2,
J
) 7.00 Hz), 7.72-7.74 (m, 2H, ar), 8.27 (d, 1H, ar, J ) 9.60
Hz); IR 1750 cm-1. Anal. (C15H16N4O6) C, H, N.
Ethyl 4,5-Dihydro-8-methyl-4-oxo-1,2,4-triazolo[1,5-a]-
quinoxaline-2-carboxylate (18). Iron powder (0.32 g) was
added to a solution of 44 (0.32 mmol) in glacial acetic acid (2
mL). The mixture was heated at 90 °C for 1 h. Evaporation
under reduced pressure of the solvent yielded a residue, which
was treated with water (150 mL), and the resulting suspension
was bleached with 6 N HCl. The solid was collected by
filtration, dried, and extracted in soxlhet with acetone (250
mL). Evaporation of the solvent at small volume (about 5 mL)
under reduced pressure yielded a solid, which was collected
General Procedure To Prepare 8-Chloro-4,5-dihydro-
5-methyl-2-phenyl-1,2,4-triazolo[1,5-a]quinoxalin-4-
one (13) and 4,5-Dihydro-5,8-dimethyl-2-phenyl-1,2,4-
triazolo[1,5-a]quinoxalin-4-one (15). Methyl iodide (1.5
mmol) was added to a suspension of 1 or 1135 (1.0 mmol) and
sodium hydride (60% dispersion in mineral oil, 2.0 mmol) in
anhydrous dimethylformamide (15 mL). The reaction mixture
was stirred at room temperature for 2 h and then quenched
with ice (30 g). The resulting solid was collected and washed
with water.
1
by filtration. Yield 58%; mp 270-272 °C (acetone); H NMR
(DMSO-d6) δ 1.36 (t, 3H, CH3, J ) 7.10 Hz), 2.42 (s, 3H, CH3),
4.40 (q, 2H, CH2, J ) 7.10 Hz), 7.35-7.42 (m, 2H, ar), 7.93 (s,
1H, ar), 12.40 (s, 1H, NH); IR 3300, 1745 cm-1. Anal.
(C13H12N4O3) C, H, N.
1
13: yield 96%; mp > 300 (acetic acid); H NMR (DMSO-d6)
δ 3.70 (s, 3H, CH3), 7.57-7.76 (m, 5H, ar), 8.21-8.25 (m, 3H,
ar); IR 1680 cm-1. Anal. (C16H11ClN4O) C, H, N.
General Procedure To Prepare Ethyl 8-Chloro-4,5-
dihydro-5-methyl-4-oxo-1,2,4-triazolo[1,5-a]quinoxaline-
2-carboxylate (24) and Ethyl 4,5-Dihydro-5-methyl-4-oxo-
1,2,4-triazolo[1,5-a]quinoxaline-2-carboxylate (25). The
title compounds were synthesized from 1635 and 17,38 respec-
tively, according to the procedure reported to obtain derivatives
13 and 15.
24: yield 85%; mp 278-280 °C (acetic acid); 1H NMR
(DMSO-d6) δ 1.38 (t, 3H, CH3, J ) 6.96 Hz), 3.69 (s, 3H, CH3),
4.44 (q, 2H, CH2, J ) 6.96 Hz), 7.71-7.83 (m, 2H, ar), 8.20 (d,
1H, ar, J ) 2.20 Hz); IR 1750, 1685 cm-1. Anal. (C13H11ClN4O3)
C, H, N.
25: yield 53%; mp 252-254 °C (acetone); 1H NMR (DMSO-
d6) δ 1.36 (t, 3H, CH3, J ) 6.96 Hz), 3.69 (s, 3H, CH3), 4.44 (q,
2H, CH2, J ) 6.96 Hz), 7.48 (t, 1H, ar, J ) 7.69 Hz), 7.63-
7.78 (m, 2H, ar), 8.22 (d, 1H, ar, J ) 8.06 Hz); IR 1740, 1680
cm-1. Anal. (C13H12N4O3) C, H, N.
15: yield 86%; mp 274-276 °C (acetic acid); 1H NMR (CDCl3)
δ 2.56 (s, 3H, CH3), 3.84 (s, 3H, N-CH3), 7.38-7.39 (m, 2H,
ar), 7.50-7.54 (m, 3H, ar), 8.21 (s, 1H, ar), 8.39-8.44 (m, 2H,
ar); IR 1675 cm-1. Anal. (C17H14N4O) C, H, N.
Ethyl N1-(5-Methyl-2-nitrophenyl)hydrazono-N2-chlo-
roacetate (41). A solution of 36% HCl (1.5 mL) was added to
solution of 5-methyl-2-nitroaniline (2.63 mmol) in methanol
(4 mL). A solution of natrium nitrite (2.89 mmol) in water (1
mL) was added dropwise to the cooled (0-5 °C) mixture in an
overall time of 15 min. The resulting natrium chloride was
eliminated by quick filtration under vacuum, and ethyl
2-chloro-3-oxobutanoate (2.89 mmol) was added to the clear
solution. The reaction mixture was kept at room temperature
for 2 h. The yellow precipitate was collected by filtration and
1
washed with water. Yield 70%; mp 152-154 °C (ethanol); H
NMR (DMSO-d6) δ 1.30 (t, 3H, CH3, J ) 7.00 Hz), 2.42 (s, 3H,
CH3), 4.33 (q, 2H, CH2, J ) 7.00 Hz), 7.02 (d, 1H, ar, J ) 8.35

Selective Human A3 Adenosine Receptor Antagonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25 7943
Diethyl 4,5-Dihydro-4-oxo-1,2,4-triazolo[1,5-a]quinox-
aline-2,5-dicarboxylate (28). A suspension of 1738 (1.15
mmol) and potassium carbonate (2.30 mmol) in acetone (50
mL) was stirred at room temperature for 10 min, and then an
excess of ethyl chlorocarbonate (2.30 mmol) was added. The
reaction mixture was heated at reflux for 8 h. After elimination
by filtration of the resulting solid, the clear solution was
evaporated at small volume under reduced pressure; a solid
precipitated, which was collected by filtration. Yield 95%; mp
135-137 °C (ethanol); 1H NMR (DMSO-d6) δ 1.31-1.41 (m,
6H, 2CH3), 4.36-4.48 (m, 4H, 2CH2), 7.86-7.95 (m, 2H, ar),
8.15 (d, 1H, ar, J ) 7.69 Hz), 8.48 (d, 1H, ar, J ) 7.69 Hz); IR
1800, 1750, 1690 cm-1. Anal. (C15H14N4O5) C, H, N.
ar, J ) 8.00 Hz), 9.58 (t, 1H, NH, J ) 5.86 Hz), 12.4 (br s, 1H,
NH); IR 3400, 1690 cm-1. Anal. (C17H13N5O2) C, H, N.
Biochemistry. Bovine A1 and A2A Receptor Binding.
Displacement of [3H]CHA from A1 ARs in bovine cerebral
cortical membranes and [3H]CGS 21680 from A2A ARs in
bovine striatal membranes was performed as described in ref
48.
Human A1, A2A, and A3 Receptor Binding. Binding
experiments at hA1 and hA3 ARs, stably expressed in CHO
cells, were performed as previously described,28 using [3H]CHA
and [125I]AB-MECA, respectively, as radioligands. Displace-
ment of [3H]NECA from hA2A ARs, stably expressed in CHO
cells, was performed as reported in ref 49.
8-Chloro-4,5-dihydro-5-methyl-1,2,4-triazolo[1,5-a]qui-
noxalin-4-one (31). The title compound was synthesized from
2929 (1.0 mmol) according to the procedure described above to
obtain derivatives 13 and 15. Yield 53%; mp 243-245 °C
(acetic acid); 1H NMR (DMSO-d6) δ 3.67 (s, 3H, CH3), 7.67 (dd,
1H, ar, J ) 8.97, 2.20 Hz), 7.76 (d, 1H, ar, J ) 8.97 Hz), 8.16
(d, 1H, ar, J ) 2.20 Hz), 8.66 (s, 1H, H-2); IR 1690 cm-1. Anal.
(C10H7ClN4O) C, H, N.
4,5-Dihydro-4-oxo-1,2,4-triazolo[1,5-a]quinoxaline-2-
carboxamide (33). A suspension of 1738(0.77 mmol) in
absolute ethanol (15 mL) saturated with ammonia was heated
in a sealed tube at 120 °C for 15 h. The resulting solid was
collected by filtration and washed with water. Yield 68%; mp
> 300 °C (2-methoxyethanol); 1H NMR (DMSO-d6) δ 7.36-
7.54 (m, 3H, ar), 7.86 (s, 1H, NH), 8.10 (d, 1H, ar, J ) 7.05
Hz), 8.27 (s, 1H, NH), 12.3 (br s, 1H, NH); IR 3570, 3400, 3320,
1700, 1680 cm-1. Anal. (C10H7N5O2) C, H, N.
The concentration of the tested compounds that produced
50% inhibition of specific [3H]CHA, [3H]CGS 21680, [125I]AB-
MECA, or [3H]NECA binding (IC50) was calculated using a
nonlinear regression method implemented in the InPlot pro-
gram (Graph-Pad, San Diego, CA) with five concentrations of
displacer, each performed in triplicate. Inhibition constants
(Ki) were calculated according to the Cheng-Prusoff equa-
tion.50 The dissociation constants (Kd) of [3H]CHA and [3H]-
CGS 21680 in cortical and striatal bovine brain membranes
were 1.2 and 14 nM, respectively. The Kd values of [3H]CHA,
[3H]NECA, and [125I]AB-MECA in hA1, hA2A, and hA3 ARs in
CHO cell membranes were 1.9, 30, and 1.4 nM, respectively.
Computational Methodologies. All molecular modeling
studies were carried out on a 6 CPU (PIV 2.0-3.0 GHZ) linux
cluster running under openMosix architecture.51
Homology modeling, energy calculation, and docking studies
were performed using the Molecular Operating Environment
(MOE, version 2004.03) suite.52
Isopropyl 4,5-Dihydro-4-oxo-1,2,4-triazolo[1,5-a]qui-
noxaline-2-carboxylate (34). A suspension of 328 (0.96
mmol) in thionyl chloride (12 mL) was heated at reflux for 4
h. The excess of thionyl chloride was removed under reduced
pressure, and the residue was treated with cyclohexane (30
mL). Evaporation of the solvent under reduced pressure gave
a solid, which was rapidly reacted with 2-propanol (0.261
mmol). The reaction mixture was heated at reflux for 1 h and
then was stirred at room temperature for 15 h. Evaporation
of the solvent under reduced pressure gave a solid, which was
treated with a little diethyl ether and collected by filtration.
The ground-state geometries of all tautomeric docked struc-
tures were fully optimized without geometry constraints using
RHF/AM1 semiempirical calculations. Vibrational frequency
analysis was used to characterize the minima stationary points
(zero imaginary frequencies). Enthalpies of formation in water
were calculated using the SM5.4/A solvation model imple-
mented in Spartan O2 software.53 The software package
Spartan O2 was utilized for all quantum mechanical calcula-
tions.
Homology Model of the hA3 AR. Based on the assumption
that GPCRs share similar TM boundaries and overall topol-
ogy,54 a homology model of the hA3 receptor was constructed.
First, the amino acid sequences of TM helices of the A3 receptor
were aligned with those of bovine rhodopsin, guided by the
highly conserved amino acid residues, including the DRY motif
(D3.49, R3.50, and Y3.51) and three proline residues (P4.60,
P6.50, and P7.50) in the TM segments of GPCRs. The same
boundaries were applied for the TM helices of the A3 receptor
as they were identified from the X-ray crystal structure for
the corresponding sequences of bovine rhodopsin, the CR
coordinates of which were used to construct the seven TM
helices for the human A3 receptor. The loop domains of the
hA3 receptor were constructed by the loop search method
implemented in MOE. In particular, loops are modeled first
in random order. For each loop, a contact energy function
analyzes 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 are then used to make a Boltz-
mann-weighted choice from the candidates, the coordinates of
which are then copied to the model. Any missing side chain
atoms are modeled using the same procedure. Side chains
belonging to residues whose backbone coordinates were copied
from a template are modeled first, followed by side chains of
modeled loops. Outgaps and their side chains are modeled last.
Special caution has to be given to the second extracellular (E2)
loop, which has been described in bovine rhodopsin as folding
back over transmembrane helices54 and, therefore, limiting the
size of the active site. Hence, amino acids of this loop could be
involved in direct interactions with the ligands. A driving force
to this peculiar fold of the E2 loop might be the presence of a
disulfide bridge between cysteines in TM3 and E2. Since this
covalent link is conserved in all receptors modeled in the
1
Yield 70%; mp 294-296 °C (acetone); H NMR (DMSO-d6) δ
1.36 (t, 6H, 2CH3, J ) 6.00 Hz), 5.20-5.27 (m, 1H, CH), 7.34-
7.54 (m, 3H, ar), 8.11 (d, 1H, ar, J ) 8.05 Hz), 12.45 (s, 1H,
NH); IR 3300, 1750, 1680 cm-1. Anal. (C13H12N4O3) C, H, N.
Benzyl 4,5-Dihydro-4-oxo-1,2,4-triazolo[1,5-a]quinoxa-
line-2-carboxylate (35). A suspension of 3238 (0.96 mmol) in
thionyl chloride (12 mL) was heated at reflux for 4 h. The
excess of thionyl chloride was removed under reduced pressure,
and the residue was treated with cyclohexane (30 mL).
Evaporation of the solvent under reduced pressure gave a solid,
which was suspended in anhydrous tetrahydrofuran (60 mL).
Then, benzyl alcohol (13 mL) was added to the resulting
suspension, and the reaction mixture was heated at reflux for
8 h. Evaporation of the solvent at small volume under reduced
pressure gave a solid, which was collected by filtration and
washed with diethyl ether. Yield 50%; mp 253-255 °C (etha-
1
nol); H NMR (DMSO-d6) δ 5.46 (s, 2H, CH2), 7.39-7.55 (m,
8H, ar), 8.10 (d, 1H, ar, J ) 8.42 Hz), 12.47 (s, 1H, NH); IR
1750, 1680 cm-1. Anal. (C17H12N4O3) C, H, N.
4,5-Dihydro-4-oxo-1,2,4-triazolo[1,5-a]quinoxaline-2-
benzylcarboxamide (36). Equimolar amounts of N-(3-di-
methylamino)propyl-N′-ethylcarbodiimide hydrochloride and
1-hydroxybenzotriazole (1.50 mmol), followed by an excess of
benzylamine (1.94 mmol), were added to a solution of 3238 (1.3
mmol) in anhydrous dimethylformamide (25 mL). The reaction
mixture was heated at reflux for 24 h. The resulting solid was
eliminated by filtration, and the clear solution was evaporated
under reduced pressure. When the residue was treated with
a little ethyl acetate/methanol 1:1, a solid precipitated, which
was collected by filtration and washed with diethyl ether. Yield
32%; mp 291-293 °C (ethanol); 1H NMR (DMSO-d6) δ 4.47
(d, 2H, CH2, J ) 6.20 Hz), 7.26-7.53 (m, 8H, ar), 8.10 (d, 1H,

7944 Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25
Catarzi et al.
(9) Impagnatiello, F.; Bastia; E. Ongini; E. Monopoli, A. Adenosine
Receptors in Neurological Disorders. Emerging Ther. Targets
2000, 4, 635-663.
(10) Muller, C. E.; Stein, B. Adenosine receptor antagonists: struc-
tures and potential therapeutic applications. Curr. Pharm. Des.
1996, 2, 501-530.
(11) Hess, S. Advances in Adenosine Receptor Antagonist Research.
Exp. Opin. Ther. Patents 2001, 11, 1533-1561.
(12) Mu¨ller, C. E. A1-Adenosine receptor antagonists. Exp. Opin.
Ther. Patents 1997, 7, 419-440.
current study, the E2 loop was modeled using a rhodopsin-
like constrained geometry around the E2-TM3 disulfide
bridge. After the heavy atoms were modeled, all hydrogen
atoms were added, and the protein coordinates were then
minimized with MOE using the AMBER94 force field.55 The
minimizations were carried out by the 1000 steps of steepest
descent followed by conjugate gradient minimization until the
rms gradient of the potential energy was less than 0.1 kcal
mol-1 Å-1
.
(13) Poulsen, S.-A.; Quinn, R. J. Adenosine receptors: new op-
portunities for future drugs. Bioorg. Med. Chem. 1998, 6, 619-
641.
(14) Mu¨ller, C. E. Medicinal Chemistry of adenosine A3 receptor
ligands. Curr. Top. Med. Chem. 2003, 3, 445-462.
(15) Fishman, P.; Madi, L.; Bar-Yehuda, S.; Barer, F.; Del Valle, L.;
Khalili, K. Evidence for involvement of Wnt signaling pathway
in IB-MECA mediated suppression of melanoma cells. Oncogene
2002, 21, 4060-4064.
(16) Merighi, S.; Prisco, M.; Varani, K.; Gessi, S.; Leung, E.; Baraldi,
P. G.; Tabrizi, M. A.; Borea, P. A. A glance at adenosine
receptors: novel target for antitumor terapy. Pharmacol. Ther.
2003, 100, 31-48.
(17) Okamura, T.; Kurogi, Y.; Hashimota, K.; Sato, S.; Nishikawa,
H.; Kiryu, K.; Nagao, Y. Structure-activity relationships of
adenosine A3 receptor ligands: new potential terapy for the
treatment of glaucoma. Bioorg. Med. Chem. Lett. 2004, 14,
3775-3779.
(18) Liang, B. T.; Stewart, D.; Jacobson, K. A. Adenosine A1 and A3
receptors: distinct cardioprotection. Drug Dev. Res. 2001, 52,
366-378.
(19) von Lubitz, D. K. J. E.; Lin, R. C. S.; Popik, P.; Carter, M. F.;
Jacobson, K. A. Adenosine A3 receptor stimulation and cerebral
ischemia. Eur. J. Pharmacol. 1994, 263, 59-67.
(20) Dunwiddie, T. V.; Masino, S. A. The role and regulation of
adenosine in the central nervous system. Ann. Rev. Neurosci.
2001, 24, 31-35.
(21) Colotta, V.; Cecchi, L.; Catarzi, D.; Melani, F.; Filacchioni, G.;
Martini, C.; Tacchi, P.; Lucacchini, A. 1-(3-Aminophenyl)-3-
methyl[1]benzopyrano[2,3-c]pyrazol-4-one: a new selective A2
adenosine receptor antagonist. Pharm. Pharmacol. Lett. 1992,
2, 74-76.
Molecular Docking of the hA3 AR antagonists. All
antagonist structures were docked into the hypothetical TM
binding site by using the DOCK docking program, part of the
MOE suite. Searching is conducted within a user-specified 3D
docking box, using the Tabu` Search protocol56 and the MMFF94
force field.57-63 MOE-Dock performs a user-specified number
of independent docking runs (50 in our specific case) and writes
the resulting conformations and their energies in a molecular
database file. The resulting docked complexes were subjected
to MMFF94 energy minimization until the rms of conjugate
gradient was <0.1 kcal mol-1 Å-1. Charges for the ligands were
imported from the Spartan output files.
The interaction energy values were calculated as follows:
∆Ebinding ) Ecomplex - (Eligand + Ereceptor). These energies are not
rigorous thermodynamic quantities but can only be used to
compare the relative stabilities of the complexes. Conse-
quently, these interaction energy values cannot be used to
calculate binding affinities since changes in entropy and
solvation effects are not taken into account.
Acknowledgment. This work was supported by a
grant from the Ministero della Universita` e della Ricerca
Scientifica e Tecnologica (MIUR), Rome, Italy. We thank
Dr. Karl-Norbert Klotz of the University of Wu¨zburg,
Germany, for providing cloned hA1, hA2A, and hA3
receptors expressed in CHO cells. The molecular model-
ing work coordinated by S. Moro was carried out with
financial support from Associazione Italiana per la
Ricerca sul Cancro (AIRC), Milan, and the MIUR. S.
Moro is also grateful to Chemical Computing Group for
their scientific and technical partnership.
(22) Colotta, V.; Cecchi, L.; Catarzi, D.; Melani, F.; Filacchioni, G.;
Martini, C.; Tacchi, P.; Lucacchini, A. Novel adenosine receptor
ligands: 1,3-disubstituted[1]benzopyrano[2,3-c]pyrazol-4-ones.
Synthesis and structure-activity relationships. Recept. Channels
1993, 1, 111-119.
(23) Colotta, V.; Cecchi, L.; Catarzi, D.; Filacchioni, G.; Martini, C.;
Tacchi, P.; Lucacchini, A. Synthesis of some tricyclic heteroaro-
matic systems and their A1 and A2a adenosine binding activity.
Eur. J. Med. Chem. 1995, 30, 133-139.
Supporting Information Available: Elemental analyses
of the newly synthesized compounds. This material is available
free of charge via the Internet at http://pubs.acs.org.
(24) Catarzi, D.; Cecchi, L.; Colotta, V.; Filacchioni, G.; Martini, C.;
Tacchi, P.; Lucacchini, A. Tricyclic heteroaromatic systems.
Synthesis and A1 and A2a adenosine binding activities of
some 1-aryl-1,4-dihydro-3-methyl[1]benzopyrano[2,3-c]pyrazol-
4-ones, 1-aryl-4,9-dihydro-3-methyl-1H-pyrazolo[3,4-b]quinolin-
4-ones, and 1-aryl-1H-imidazo[4,5-b]quinoxalines. J. Med. Chem.
1995, 38, 1330-1336.
(25) Colotta, V.; Catarzi, D.; Varano, F.; Cecchi, L.; Filacchioni, G.;
Martini, C.; Trincavelli, L.; Lucacchini, A. 4-Amino-6-benzyl-
amino-1,2-dihydro-2-phenyl-1,2,4-triazolo[4,3a]quinoxalin-1-one:
a new A2A adenosine receptor antagonist with high selectivity
versus A1 receptors. Arch. Pharm. Pharm. Med. Chem. 1999,
332, 39-41.
(26) Colotta, V.; Catarzi, D.; Varano, F.; Cecchi, L.; Filacchioni, G.;
Martini, C.; Trincavelli, L.; Lucacchini, A. Synthesis and struc-
ture-activity relationships of a new sets of 2-arylpyrazolo[3,4-
c]quinoline derivatives as adenosine receptor antagonists. J.
Med. Chem. 2000, 43, 3118-3124.
(27) Colotta, V.; Catarzi, D.; Varano, F.; Cecchi, L.; Filacchioni, G.;
Martini, C.; Trincavelli, L.; Lucacchini, A. 1,2,4-Triazolo[4,3-a]-
quinoxalin-1-one: a versatile tool for the synthesis of potent and
selective adenosine receptor antagonists. J. Med. Chem. 2000,
43, 1158-1164.
References
(1) Fredholm, B. B.; Ijzerman, A. P.; Jacobson, K. A.; Klotz, K. N.;
Linden, J. International union of Pharmacology XXV. Nomen-
clature and classification of adenosine receptors. Pharmacol. Rev.
2001, 53, 527-552.
(2) Jacobson, K. A.; Knutsen, L. J. S. P1 and P2 purine and
pyrimidine receptor ligands. In Purinergic and Pyrimidinergic
Signalling; Abbracchio, M. P., Williams, M., Eds; Handbook of
Experimental Pharmacology, Vol. 151/1; Springer: Berlin, 2001;
pp 129-175.
(3) Olah, M.; Stiles, G. L. The role of receptor structure in determin-
ing adenosine receptor activity. Pharmacol. Ther. 2000, 85, 55-
75.
(4) Fredholm, B. B.; Arslan, G.; Halldner, L.; Kull, B.; Schulte, G.;
Wasserman, W. Structure and function of adenosine receptors
and their genes. Naunyn-Schmiedeberg’s Arch. Pharmacol. 2000,
362, 364-374.
(5) Ralevich, V.; Burnstock, G. Receptors for purines and pyrim-
idines. Pharmacol. Rev. 1998, 50, 413-492.
(6) Abbracchio, M. P.; Brambilla, R.; Kim, H. O.; von Lubitz, D. K.
J. E.; Jacobson, K. A.; Cattabeni, F. G-protein-dependent activa-
tion of phospholipase-C by adenosine A3 receptor in rat brain.
Mol. Pharmacol. 1995, 48, 1083-1045.
(7) Ali, H.; Choi, O. H.; Fraundorfer, P. F.; Yamada, K.; Gonzaga,
H. M. S.; Beaven, M. A. Sustained activation of phospholipase-D
via adenosine A3 receptors is associated with enhancement of
antigen-ionophore-induced and Ca2+-ionophore-induced secretion
in a rat mast-cell line. J. Pharmacol. Exp. Ther. 1996, 276, 837-
845.
(28) Colotta, V.; Catarzi, D.; Varano, Filacchioni, G.; Martini, C.;
Trincavelli, L.; Lucacchini, A. Synthesis and structure-activity
relationships of a new set of 1,2,4-Triazolo[4,3-a]quinoxalin-1-
one derivatives as adenosine receptor antagonists. Bioorg. Med.
Chem. 2003, 11, 3541-3550.
(29) Catarzi, D.; Colotta, V.; Varano, F.; Filacchioni, G.; Martini, C.;
Trincavelli, L.; Lucacchini, A. 1,2,4-Triazolo[1,5-a]quinoxaline
derivatives: synthesis and biological evaluation as adenosine
receptor antagonists. Il Farmaco 2004, 59, 71-81.
(8) Ribeiro, J. A.; Sebastiao, A. M.; de Mendica, A. Adenosine
receptors in the nervous system: pathophysiological implica-
tions. Prog. Neurobiol. 2003, 68, 377-392.

Selective Human A3 Adenosine Receptor Antagonists
Journal of Medicinal Chemistry, 2005, Vol. 48, No. 25 7945
(30) Colotta, V., Catarzi, D.; Varano, F.; Calabri, F. R.; Lenzi, O.;
Filacchioni, G.; Martini, C.; Trincavelli, L.; Deflorian, F.; Moro,
S. 1,2,4-Triazolo[4,3-a]quinoxalin-1-one moiety as an attractive
scaffold to develop new potent and selective human A3 adenosine
receptor antagonists: synthesis, pharmacological and ligand-
receptor modeling studies. J. Med. Chem. 2004, 47, 3580-3590.
(31) Catarzi, D.; Colotta, V.; Varano, F.; Calabri, F. R.; Lenzi, O.;
Filacchioni, G.; Martini, C.; Trincavelli, L.; Tralli, A.; Monopoli,
C.; Moro, S. 2-Aryl-8-chloro-1,2,4-triazolo[1,5-a]quinoxalin-4-
amines as highly potent A1 and A3 adenosine receptor antago-
nists. Bioorg. Med. Chem. 2005, 13, 705-715.
profile of a novel triazoloquinazoline adenosine antagonists. J.
Med. Chem. 1988, 31, 1014-1020.
(46) Moro, S.; Deflorian, F.; Spalluto, G.; Pastorin, G.; Cacciari, B.;
Soo-Kyung, K.; Jacobson, J. A. Demystifying the three-dimen-
sional structure of G protein-coupled receptors (GPCRs) with
the aid of molecular modeling. Chem. Commun. 2003, 2949-
2956.
(47) Moro, S.; Spalluto, G.; Jacobson, J. A. Techniques: recent
developments in computer-aided engineering of GPCR ligands
using the human adenosine A3 receptor as an example. Trends
Pharmacol. Sci. 2005, 26, 44-51.
(48) Colotta, V.; Catarzi, D.; Varano, F.; Melani, F.; Filacchioni, G.;
Cecchi, L.; Trincavelli, L.; Martini, C.; Lucacchini, A. Synthesis
and A1 and A2A adenosine binding activity of some pyrano[2,3-
c]pyrazol-4-ones. Farmaco 1998, 53, 189-196.
(32) Jacobson, K. A.; Tchilibon, S.; Bhalchandra, V. J.; Zhan-Guo,
G. A3 Adenosine receptors. Annu. Rep. Med. Chem. 2003, 38,
121-130.
(33) Baraldi, P. G.; Tabrizi, M. A.; Fruttarolo, F.; Bovero, A.;
Avitabile, B.; Preti, D.; Romagnoli, R.; Meriggi, S.; Gessi, S.;
Varani, K.; Borea, P. A. Recent developments in the field of A3
adenosine receptor antagonists. Drug. Dev. Res. 2003, 58, 315-
329.
(34) Baraldi, P. G.; Cacciari, B.; Romagnoli, R.; Meriggi, S.; Varani,
K.; Borea, P. A.; Spalluto, G. A3 Adenosine Receptor Ligands:
History and Perspectives. Med. Res. Rev. 2000, 20, 103-128.
(35) Catarzi, D.; Cecchi, L.; Colotta, V.; Melani, F.; Filacchioni, G.;
Martini, C.; Giusti, L.; Lucacchini, A. Tricyclic heteroaromatic
systems. 1,2,4-triazolo[1,5-a]quinoxalines: synthesis and ben-
zodiazepine receptor activity. Il Farmaco 1993, 48, 1065-1078.
(36) Catarzi, D.; Cecchi, L.; Colotta, V.; Melani, F.; Filacchioni, G.;
Martini, C.; Giusti, L.; Lucacchini, A. Structure-activity rela-
tionships of 1,2,4-triazolo[1,5-a]quinoxalines and their 1-deaza
analogues imidazo[1,2-a]quinoxalines at the benzodiazepine
receptor. J. Med. Chem. 1994, 37, 2846-2850.
(37) Mangini, A.; Colonna, M. Ricerche sui nitroderivati aromatici.-
Nota XVI. 3,4-Dinitro-toluolo: reagibilita` e configurazione nucle-
are. Gazz. Chim. Ital. 1938, 68, 708-718.
(38) Catarzi, D.; Cecchi, L.; Colotta, V.; Filacchioni, G.; Melani, F.
Tricyclic Heteroaromatic Systems. 1,2,4-Triazolo[1,5-a]quinoxa-
line. J. Heterocycl. Chem. 1992, 29, 1161-1163.
(39) Catarzi, D.; Colotta, V.; Varano, F.; Cecchi, L.; Filacchioni, G.;
Galli, A.; Costagli, C. 4,5-Dihydro-1,2,4-triazolo[1,5-a]quinoxalin-
4-ones: Excitatory amino acid antagonists with combined Gly-
cine/NMDA and AMPA receptor affinity. J. Med. Chem. 1999,
42, 2478-2484.
(40) Catarzi, D.; Colotta, V.; Varano, F.; Filacchioni, G.; Galli, A.;
Costagli, C.; Carla`, V. Synthesis, ionotropic glutamate receptor
binding affinity and structure-activity relationships of a new
set of 4,5-dihydro-8-heteroaryl-4-oxo-1,2,4-triazolo[1,5-a]quinox-
aline-2-carboxylates analogues of TQX-173. J. Med. Chem. 2001,
44, 3157-3165.
(49) Klotz, K.-N.; Hessling, J.; Hegler, J.; Owman, C.; Kull, B.;
Fredholm, B. B.; Lohes, M. J. Comparative pharmacology of
human adenosine receptor subtypes-characterization of stably
transfected receptors in CHO cells. Naunyn-Schiedeberg’s Arch.
Pharmacol. 1998, 357, 1-9.
(50) Cheng, Y. C.; Prusoff, W. H. Relation between the inhibition
constant Ki and the concentration of inhibitor which causes fifty
percent inhibition (IC50) of an enzyme reaction. Biochem. Phar-
macol. 1973, 22, 3099-3108.
(51) OpenMosix, http://www.openMosix.org.
(52) Molecular Operating Environment (MOE 2003.02): Chemical
Computing Group, Inc: Montreal, Quebec, Canada, 2003.
(53) Spartan O2; Wavefunction Inc.: Irvine, CA, 2002.
(54) Palczewski, K.; Kumasaka, T.; Hori, T.; Behnke, C. A.; Mo-
toshima, H.; Fox, B. A.; Le Trong, I.; Teller, D. C.; Okada, T.;
Stenkamp, R. E.; Yamamoto, M.; Miyano, M. Crystal structure
of rhodopsin: a G protein-coupled receptor. Science 2000, 289,
739-745.
(55) 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. A second generation force field for the simulation
of proteins, nucleic acids and organic molecules. J. Am. Chem.
Soc. 1995, 117, 5179-5196.
(56) Baxter, C. A.; Murray, C. W.; Clark, D. E.; Westhead, D. R.;
Eldridge, M. D. Flexible docking using tabu search and an
empirical estimate of binding affinity. Proteins: Struct., Funct.,
Genet. 1998, 33, 367-382.
(57) Halgren, T. A. Merck Molecular Force Field. I. Basis, form, scope,
parametrization, and performance of MMFF94. J. Comput.
Chem. 1996, 17, 490-519.
(58) Halgren, T. A. Merck Molecular Force Field. II. MMFF94 van
der Waals and electrostatic parameters for intermolecular
interactions. J. Comput. Chem. 1996, 17, 520-552.
(59) Halgren, T. A. Merck Molecular Force Field. III. Molecular
geometries and vibrational frequencies for MMFF94. J. Comput.
Chem. 1996, 17, 553-586.
(60) Halgren, T. A. Merck Molecular Force Field. IV. Conformational
energies and geometries for MMFF94. J. Comput. Chem. 1996,
17, 587-615.
(41) Catarzi, D.; Colotta, V.; Varano, F.; Calabri, F. R.; Filacchioni,
G.; Galli, A.; Costagli, C.; Carla`, V. Synthesis and biological
evaluation of analogues of 7-chloro-4,5-dihydro-4-oxo-8-(1,2,4-
triazol-4-yl)-1,2,4-triazolo[1,5-a]quinoxaline-2-carboxylic acid (TQX-
173) as novel selective AMPA receptor antagonists. J. Med.
Chem. 2004, 47, 262-272.
(42) Gao, Z.-G.; Blaustein J. B.; Gross, A. S.; Melman, N.; Jacobson,
K. A. N6-Substituted adenosine derivatives: selectivity, efficacy
and species differences at A3 adenosine receptors. Biochem.
Pharmacol. 2003, 65, 1675-1684.
(43) Salvatore, C. A.; Jacobson, M. A.; Taylor, H. E.; Linden, J.;
Johnson, R. G. Molecular cloning and characterization of the
human A3 adenosine receptor. Proc. Natl. Acad. Sci. U.S.A. 1993,
90, 10365-10369.
(44) Klotz, K.-N. Adenosine receptors and their ligands. Naunyn-
Schmiedeberg’s Arch. Pharmacol. 2000, 362, 392-401.
(45) Francis, J. E.; Cash, W. D.; Psychoyos, S.; Ghai, G.; Wenk, P.;
Friedmann, R. C.; Atkins, C.; Warren, V.; Furness, P.; Hyun, J.
L.; Stone, G. A.; Desai, M.; Williams, M. Structure-activity
(61) Halgren, T. A.; Nachbar R. Merck Molecular Force Field. V.
Extension of MMFF94 using experimental data, additional
computational data, and empirical rules J. Comput. Chem. 1996,
17, 616-641.
(62) Halgren, T. A. MMFF VI. MMFF94s Option for energy minimi-
zation studies. J. Comput Chem. 1999, 20, 720-729.
(63) Halgren, T. A. MMFF VII. Characterization of MMFF94,
MMFF94s, and other widely available force fields for conforma-
tional energies and for intermolecular-interaction energies and
geometries. J. Comput. Chem. 1999, 20, 730-748.
JM0504149