3-Aryl[1,2,4]triazino[4,3-α]benzimidazol-4(10H)-ones: A new class of selective A1 adenosine receptor antagonists

Article abstract of DOI:10.1021/jm001054m

Radioligand binding assays using bovine cortical membrane preparations and biochemical in vitro studies revealed that various 3-aryl[1,2,4]triazino[4,3-α]benzimidazol-4(10H)-one (ATBI) derivatives, previously reported by us as ligands of the central benzo

Full text of DOI:10.1021/jm001054m

316
J . Med. Chem. 2001, 44, 316-327
3-Ar yl[1,2,4]tr ia zin o[4,3-a ]ben zim id a zol-4(10H)-on es: A New Cla ss of Selective
A1 Ad en osin e Recep tor An ta gon ists
Federico Da Settimo,*,§ Giampaolo Primofiore,§ Sabrina Taliani,§ Anna Maria Marini,§ Concettina La Motta,§
Ettore Novellino,‡ Giovanni Greco,‡ Antonio Lavecchia,‡ Letizia Trincavelli, and Claudia Martini
Dipartimento di Scienze Farmaceutiche, Universita` di Pisa, Via Bonanno 6, 56126 Pisa, Italy, Dipartimento di Chimica
Farmaceutica e Tossicologica, Universita` di Napoli “Federico II”, Via D. Montesano, 49, 80131 Napoli, Italy, and
Dipartimento di Psichiatria, Neurobiologia, Farmacologia e Biotecnologie, Universita` di Pisa, Via Bonanno 6,
56126 Pisa, Italy
Received August 7, 2000
Radioligand binding assays using bovine cortical membrane preparations and biochemical in
vitro studies revealed that various 3-aryl[1,2,4]triazino[4,3-a]benzimidazol-4(10H)-one (ATBI)
derivatives, previously reported by us as ligands of the central benzodiazepine receptor (BzR)
(Primofiore, G.; et al. J . Med. Chem. 2000, 43, 96-102), behaved as antagonists at the A1
adenosine receptor (A1AR). Alkylation of the nitrogen at position 10 of the triazinobenzimidazole
nucleus conferred selectivity for the A1AR vs the BzR. The most potent ligand of the ATBI
series (10-methyl-3-phenyl[1,2,4]triazino[4,3-a]benzimidazol-4(10H)-one 12) displayed a Ki value
of 63 nM at the A1AR without binding appreciably to the adenosine A2A and A3 nor to the
benzodiazepine receptor. Pharmacophore-based modeling studies in which 12 was compared
against a set of well-established A1AR antagonists suggested that three hydrogen bonding sites
(HB1 acceptor, HB2 and HB3 donors) and three lipophilic pockets (L1, L2, and L3) might be
available to antagonists within the A1AR binding cleft. According to the proposed pharma-
cophore scheme, the lead compound 12 engages interactions with the HB2 site (via the N2
nitrogen) as well as with the L2 and L3 sites (through the pendant and the fused benzene
rings). The results of these studies prompted the replacement of the methyl with more lipophilic
groups at the 10-position (to fill the putative L1 lipophilic pocket) as a strategy to improve
A1AR affinity. Among the new compounds synthesized and tested, the 3,10-diphenyl[1,2,4]-
triazino[4,3-a]benzimidazol-4(10H)-one (23) was characterized by a Ki value of 18 nM which
represents a 3.5-fold gain of A1AR affinity compared with the lead 12. A rhodopsin-based model
of the bovine adenosine A1AR was built to highlight the binding mode of 23 and two well-
known A1AR antagonists (III and VII) and to guide future lead optimization projects. In our
docking simulations, 23 receives a hydrogen bond (via the N1 nitrogen) from the side chain of
Asn247 (corresponding to the HB1 and HB2 sites) and fills the L1, L2, and L3 lipophilic pockets
with the 10-phenyl, 3-phenyl, and fused benzene rings, respectively.
In tr od u ction
addressed the question of whether our ATBI derivatives
could be targeted at the A1AR. Several theoretical
studies have improved our understanding of how AR
ligands interact selectively with their binding sites,15-23
thus providing a solid background to predict the likeli-
hood for a given structure to behave as agonist or
antagonist at any of the ARs. A molecular modeling
analysis confirmed that ATBIs are superimposable on
A1AR antagonists I-VIII about pharmacophoric ele-
ments required for high A1AR potency (Figure 2).
We have recently described the synthesis and the
biological evaluation of 3-aryl[1,2,4]triazino[4,3-a]benz-
imidazol-4(10H)-ones (ATBIs, 1-17 in Table 1) as
ligands of the benzodiazepine receptor (BzR).1 While
these compounds were under investigation, it was
realized that they were structurally related to well-
known potent A1 adenosine receptor (A1AR) antagonists
(I-VIII)2-10 such as those shown in Figure 1. ARs have
been identified by pharmacological and molecular bio-
logical techniques and classified into A1, A2A, A2B, and
A3 subtypes, all belonging to the G-protein coupled
receptors superfamily.11-13 Since selective A1AR an-
tagonists have demonstrated promising therapeutic
potential for the treatment of cognitive deseases, renal
failure, Alzheimer’s disease, and cardiac failure,14 we
Given the role of the N10-H group as a hydrogen bond
donor at the BzR1 but not at the A1AR (as established
by the modeling studies), we predicted that N10-alkyl
derivatives might be A1AR selective ligands devoid of
appreciable affinity at the BzR. Expectedly, several of
our compounds were found to act as A1AR antagonists
without binding to the BzR. The above-mentioned
pharmacophore-based overlay was successfully exploited
to design novel ATBI derivatives with improved potency
at the A1AR. Finally, we built a model of the bovine
A1AR to rationalize the SARs of ATBI ligands and to
facilitate, perspectively, the design of new analogues.
* To whom all correspondence should be addressed. Tel: 39 50
500209. Fax: 39 50 40517. E-mail: fsettimo@farm.unipi.it.
§ Dipartimento di Scienze Farmaceutiche, Universita` di Pisa.
‡ Dipartimento di Chimica Farmaceutica e Tossicologica, Universita`
“Federico II” di Napoli.
Dipartimento di Psichiatria, Neurobiologia, Farmacologia e Bio-
tecnologie, Universita` di Pisa.
10.1021/jm001054m CCC: $20.00 © 2001 American Chemical Society
Published on Web 01/03/2001

3-Aryl[1,2,4]triazino[4,3-a]benzimidazol-4(10H)-ones
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 317
Ta ble 1. Affinity of ATBI Derivatives at Bovine Brain A1, A2A
and A3ARs
,
VII through the fitting of the nitrogen closest to the
pendant phenyl plus the two benzene rings.
The alignment of compounds I-VIII and 12 (see
Figure 2) suggests that, in addition to three putative
hydrogen bonding sites (HB1 acceptor, HB2, and HB3
donors), three lipophilic pockets (L1, L2, and L3) might
be available to antagonists within the A1AR binding
cleft.
Ki (nM) or % inhibitiona
Our pharmacophore model is similar to that proposed
by Dooley and co-workers,24 except for the HB3 site
which has no equivalent in the latter model and for the
alignment of some ligands.25 As an example, VIII was
oriented by Dooley et al. with the NH2 group engaging
the HB2 site and with the pendant, fused, and benzylic
benzene rings fitting into the L1, L2, and L3 regions,
respectively.
b
c
d
no.
R
R′
bA1
34%
3900 ( 280 39%
bA2A
bA3
1
2
3
4
5
6
7
8
H
H
H
H
H
H
H
H
H
H
H
H
CH3
C6H5
0%
85 ( 7.2
44%
2900 ( 190
8%
5%
C6H4-4′-Cl
C6H4-4′-OCH3
C6H4-4′-OH
30%
16%
249 ( 23
1300 ( 121
C6H3-3′,4′-(OCH3)2 1780 ( 150 35%
C6H3-3′,4′-(OH)2
470 ( 43
261 ( 24
293 ( 26
119 ( 10
0%
3500 ( 290
241 ( 23
1023 ( 95
5000 ( 470
8%
Differently from what was hypothesized to occur at
the BzR,1 the N10-H group of ATBIs does not make any
hydrogen bonds at the A1AR. Consequently, derivatives
lacking the N10-H function should retain affinity to the
A1AR and lack potency at the BzR. These results
strongly supported the screening of ATBI derivatives
as A1AR antagonists and provided the rationale to
design new compounds with improved potency and
selectivity in binding to the A1AR vs the BzR.
9
10
11
fur-2-yl
thien-2-yl
thien-3-yl
H
18 CH3
19 CH3
12 CH3
13 CH3
14 CH3
20 CH3
15 CH3
21 CH3
16 CH3
17 CH3
22 CH3
CH3
C6H5
46%
0%
63 ( 5.4
66%
25%
0%
0%
C6H4-4′-Cl
C6H4-4′-OCH3
C6H4-4′-OH
C6H3-3′,4′-(OCH3)2 24%
C6H3-3′,4′-(OH)2
fur-2-yl
thien-2-yl
thien-3-yl
C6H5
2050 ( 190 7%
206 ( 18 43%
18%
459 ( 41
32%
Biologica l Eva lu a tion of Com p ou n d s 1-22. To
better delineate the SARs of ATBI derivatives, five new
compounds (18-22), with respect to those described as
BzR ligands (1-17),1 were prepared by reacting 1-meth-
yl-2-hydrazinobenzimidazole 2926 with the appropriate
glyoxylic acids following experimental procedures simi-
lar to those previously reported to obtain the corre-
sponding N10-H analogues 1, 2, 6, 8, and 111 (Scheme
1, Tables 2 and 3). All the compounds 1-22 (Table 1)
were tested in radioligand binding assays to determine
their affinities at the bovine A1, A2A, and A3ARs. In
particular, affinities for A1 and A2AARs were determined
in competition assays of [3H]-N6-cyclohexyladenosine
([3H]CHA) and [3H]-2-[[4-(2-carboxyethyl)phenethyl]-
amino]-5′-(N-ethylcarbamoyl)adenosine ([3H]CGS 21680)
to bovine cortical (A1) and striatal (A2A) membranes,
respectively. Moreover, to determine the intrinsic activ-
ity of the most A1AR active compounds 3, 12, 23, and
25, competition studies were performed in the presence
and in the absence of 1 mM GTP using the radiolabeled
antagonist [3H]1,3-dipropyl-8-cyclopentylxantine ([3H]-
DPCPX).27 The GTP shift is an in vitro parameter often
indicative of intrinsic activity. In Table 4 the GTP shift
values of the selected ATBIs and R-PIA, included as
agonist reference compound, were reported. Affinity for
A3ARs was determined in competition assays of
813 ( 70
233 ( 21
78 ( 6.2
18 ( 2
6300 ( 610
19%
55%
23 C6H5
24 C6H5
18%
2%
fur-2-yl
315 ( 31
274 ( 25
344 ( 33
990 ( 87
446 ( 41
56%
62%
1200 ( 110
1236 ( 112
60%
25 CH2-C6H5 C6H5
26 CH2-C6H5 C6H4-4′-OH
27 CH2-C6H5 fur-2-yl
28 CH2-C6H5 thien-3-yl
DPCPX
13%
0.5 ( 0.03 337 ( 28
>10000
73 ( 5
NECA
Cl-IBMECA
14 ( 4
16 ( 3
890 ( 61
401 ( 25
0.22 ( 0.02
a
The Ki values are means ( SEM of four separate assays, each
performed in triplicate. b Displacement of specific [3H]CHA binding
in bovine cortical membranes or percentage of inhibition (I%) of
specific binding at 10 µM concentration. c Displacement of specific
[3H]CGS 21680 in bovine striatal membranes or percentage of
d
inhibition (I%) of specific binding at 10 µM concentration. Dis-
placement of specific [125I]AB-MECA binding to bovine cortical
membranes in the presence of 20 nM DPCPX or percentage of
inhibition (I%) of specific binding at 1 µM concentration. Only
compounds 3, 12, 23, and 25 were tested.
Resu lts a n d Discu ssion
Com p a r ison of ATBI Der iva t ives w it h Kn ow n
A1AR An ta gon ists. Molecular modeling studies were
performed to verify whether compound 12, selected as
a representative of the ATBI series, possessed pharma-
cophoric elements of well-known A1AR antagonists
(compounds I-VIII in Figure 1). Computational details
are given in the Experimental Section. The points used
for superimposing structures I-VIII and 12 are denoted
in Figure 1 by bold labels (atoms supposed to make
hydrogen bonds with the receptor) and dots (centroids
of aromatic rings assumed to fit into hydrophobic
regions of the binding site). Compounds I-VII were first
aligned about the three atoms in bold (N or O, N, and
H), keeping the orientation of the structural formulas
adopted in Figure 1. Compound VIII, lacking the third
hydrogen bonding atom, was overlayed on VII by
matching the corresponding N and H atoms plus the
pendant phenyl. Compound 12 was superimposed on
[
125I]-N6-(3-iodo-4-aminobenzyl)-5′-N-methylcarboxamido-
adenosine ([125I]AB-MECA) to bovine cortical mem-
branes in the presence of the A1AR selective antagonists
DPCPX (20 nM). As reported for mouse cortical mem-
branes,28 to ascertain the validity of the used method
in bovine cortical membranes, DPCPX competition
studies of 1 nM [125I]AB-MECA binding and [125I]AB-
MECA saturation isotherm in the absence and presence
of DPCPX were carried out. [125I]AB-MECA specific
binding consisted of two components, since the A1
selective antagonist DPCPX completely deleted the most
but not all of the binding with high affinity (sigmoidal,
monophasic curve had a Ki value of 0.22 nM). Then, the

318 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Da Settimo et al.
F igu r e 1. Compounds I-VIII are A1AR antagonists reported in the literature that were used, together with the ATBI derivative
12, to generate the pharmacophore model shown in Figure 2. Atoms in bold and dots served as fitting points. I-VIII are described
in the following references: I,2 II,3,4 III,5 IV,6 V,7 VI,8 VII,9 and VIII.10
Sch em e 1a
F igu r e 2. Molecular models of I-VIII and 12 (red) overlayed
about the pharmacophoric points indicated in Figure 1 through
bold labels and dots. Three putative hydrogen bonding sites
(HB1 acceptor, HB2 and HB3 donors) and three lipophilic
pockets (L1, L2, and L3) are hypothesized to exist within the
a
Reagents: (A) ethanol, gaseous HCl, reflux; (B) glacial acetic
acid, reflux.
A1AR binding cleft.
In agreement with the results of the molecular
residual binding which was not displaced by DPCPX at
concentrations as high as 20 nM (about 100-fold above
the Ki value to A1AR) represented binding to A3AR.
Scatchard analysis of [125I]AB-MECA binding, per-
formed in the presence of 20 nM DPCPX, showed an
affinity constant value of 1.02 ( 0.08 nM, and a
maximum density of binding sites of 14.7 ( 1.2 fmol/
mg of proteins according to the data reported for
A3AR in mouse brain.28 In Table 1, the affinity con-
stants values obtained in our experimental conditions
for the selected agonists 5′-N-ethylcarboxamidoadeno-
sine (NECA) and 2-chloro-N6-(3-iodobenzyl)adenosine
(Cl-IBMECA), and the antagonist DPCPX as standards,
were reported.
modeling studies, several compounds (3, 6, 8-11 and
12, 16, 17, 20-22) exhibited submicromolar affinity at
the A1AR. All the compounds, except 9, showed a
different degree of selectivity for the A1AR compared
with the A2A and A3ARs.
A crucial requirement for binding to the A1AR is an
aromatic ring at the 3-position, since 3-unsubstituted
or 3-methylated ATBIs are devoid of affinity or scarcely
active (1, 2, 18, and 19). As predicted by our pharma-
cophore model, the presence or absence of a methyl
group at the imidazole nitrogen N10 does not signifi-
cantly influence affinity to the A1AR as N10-H and N10-
CH3 derivatives are practically equipotent, with the
exception of 9 and 16. The most active compounds are

3-Aryl[1,2,4]triazino[4,3-a]benzimidazol-4(10H)-ones
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 319
Ta ble 2. Physical Properties and Spectral Data of Benzimidazol-2-ylhydrazone Derivatives 32-35 and 36-39
yield
(%)
mp
(°C)
IR
1H NMR
(ppm)
MS
m/e (%)
no.
R
R′
formulaa
(cm-1
)
32 CH3
33 CH3
34 CH3
35 CH3
H
65 175-177 C10H10N4O2 3200-2800,
3.50 (s, 3H, CH3);
7.01-7.29 (m, 4H,
Ar-H), 7.34 (s, 1H,
2-H)
2.07 (s, 3H, 2-CH3);
3.51 (s, 3H, N-CH3);
6.99-7.30 (m, 4H,
Ar-H)
3.52 (s, 3H, CH3); 6.77
(d, 2H, 3′,5′-H); 7.05-
7.35 (m, 4H, Ar-H);
7.71 (d, 2H; 2′,6′-H)
3.59 (s, 3H, CH3); 7.11-
8.23 (m, 7H, Ar-H);
12.10 (bs, 1H, COOH)
218 (11, M+); 200 (2); 174
(85); 131 (88); 44 (100)
1670, 1570,
1240, 1070,
740
b
CH3
86 200-202 C11H12N4O2 3300-2600,
232 (1, M+); 214 (1); 187 (2);
131 (3); 45 (100)
1710, 1640,
1320, 1110,
740
C6H4-4′-OH 71 212-215 C16H14N4O3 3600-2400,
310 (1, M+); 292 (28); 264
(10); 147 (40); 119 (100)
(dec)
1650, 1600,
1270, 840,
740
thien-3-yl
66 228-231 C14H12N4O2S 3050-2300,
300 (6, M+); 282 (12); 256
(28); 147 (100)
1600, 1580,
1450, 1080,
750
36 CH2C6H5 C6H5
72 189-192 C22H18N4O2 3200-2400,
5.24 (s, 2H, CH2); 7.03-
7.87 (m, 14H, ArH)
352 (15, M+-H2O); 324 (1);
222 (1); 91 (100)
(dec)
1620, 1590,
1440, 740
37 CH2C6H5 C6H4-4′-OH 50 203-205 C22H18N4O3 3400-2400,
5.19 (s, 2H, CH2); 6.75
(d, 2H, 3′,5′-H); 7.00-
7.52 (m, 9H, Ar-H);
7.68 (d, 2H, 2′,6′-H);
9.54 (s, 1H, OH,
368 (32, M+-H2O); 342 (16);
222 (20); 91 (100)
1660, 1270
730
exch. with D2O)
38 CH2C6H5 fur-2-yl
71 201-202 C20H16N4O3 1640, 1600,
5.31 (s, 2H, CH2); 6.53-
6.60 (m, 1H, 4′-H);
7.09-7.45 (m, 10H, Ar-H);
7.72 (d, 1H, 5′-H); 12.20
(bs, 1H, COOH,
360 (1, M+); 222 (7); 91 (100)
1260, 1110,
740
exch. with D2O)
39 CH2C6H5 thien-3-yl
76 143-145 C20H16N4O2S 3500-2600,
5.30 (s, 2H, CH2); 7.11-
7.54 (m, 10H, Ar-H);
7.88 (dd, 1H, Ar-H);
8.15 (dd, 1H, Ar-H);
12.20 (bs, 1H, COOH,
exch. with D2O)
376 (1, M+); 358 (48); 33
(8); 222 (11); 91 (100)
1640, 1600,
1260, 1120,
730
a
b
Elemental analyses for C, H, N, were within (0.4% of the calculated values. Literature ref 82.
those bearing an unsubstituted phenyl ring at the
3-position (3 and 12 with Ki values of 85 and 63 nM,
respectively). Any substituent on the side phenyl ring
decreases or abolishes affinity, thus suggesting that the
lipophilic site of the A1AR hosting the phenyl moiety
(the L2 region in Figure 2) is relatively small. The only
substituents tolerated on this ring were meta and para
hydroxyls (compare 6, 8, 20, and 21 vs 3 and 12). The
replacement of the side phenyl ring with a less bulky
heteroaromatic five-membered ring, such as furyl or
thienyl, gave compounds 9-11, 16, ,17, and 22, showing
an affinity only slightly lower than that of 3 and 12.
Lea d Op tim iza tion : Design , Syn th esis, a n d Test-
in g of New ATBI Der iva tives. Inspection of the
molecular alignment shown in Figure 2 suggested that
affinity of the lead compound 12 (Ki ) 63 nM) could be
improved by replacing the methyl group at the N10-
position with larger and hydrophobic substituents ca-
pable of filling the L1 lipophilic pocket. Accordingly, we
synthesized and tested a number of N10-phenyl (23, 24)
and N10-benzyl (25-28) derivatives.
zyl-2-hydrazinobenzimidazole 31 was obtained by a
synthetic procedure reported in the literature.26 We used
the same procedure to prepare the already described
2-hydrazino-1-phenylbenzimidazole 3029 from hydrazine
hydrate and 2-chloro-1-phenylbenzimidazole. This last
compound was obtained by reaction of 1-phenylbenz-
imidazol-2-one and phenylphosponic dichloride30 at 170
°C with a better yield (60%) than that reported by
Iemura (50%) by using POCl3.31 1-Phenylbenzimidazol-
2-one was prepared in quantitative yield using a con-
venient modification of the procedure of Clark and
Pessolano,32 i.e., employing triphosgene (bis-trichloro-
methyl carbonate)33 instead of phosgene for the reaction
of carbonylation of N-phenyl-1,2-phenylendiamine.
The reaction of 1-benzyl-2-hydrazinobenzimidazole 31
with the appropriate glyoxylic acids in ethanol, at reflux,
furnished the intermediate 2-(benzimidazol-2-ylhydra-
zono)-2-substituted acetic acids 36-39 (Scheme 1, Table
2), which were purified by suspension in hot methanol
as the recrystallization process always led to partial
cyclization of the products. The cyclization of these acids
was obtained using one of the following two methods
(Scheme 1, Table 3): (A) by saturating with anhydrous
hydrogen chloride a suspension of the acid in absolute
The new compounds 23, 24, and 25-28 were prepared
starting from the appropriate 2-hydrazinobenzimidazole
derivatives 30 and 31, respectively (Scheme 1). 1-Ben-

320 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Da Settimo et al.
Ta ble 3. Physical Properties and Spectral Data of Triazinobenzimidazole Derivatives 18-22, 23, 24, and 25-28
react. yield recryst
mp
(°C)
IR
(cm-1
1H NMR
(ppm)
MS
m/e (%)
no.
R
R′
procd (%)
solv
formulaa
)
18 CH3
H
B
32
MeOH
206-207 C10H8N4O
1700, 1680,
1250, 980,
750, 730
3.84 (s, 3H, CH3); 7.37- 200 (100, M+);
7.76 (m, 3H, Ar-H);
8.26 (s, 1H, 3-H); 8.34
(dd, 1H, 6-H)
2.41 (s, 3H, 3-CH3);
3.76 (s, 3H, N-CH3);
7.29-7.67 (m, 3H,
Ar-H); 8.29 (dd, 1H,
6-H)
144 (41); 118
(39)
19 CH3
CH3
B
31
MeOH
DMF
225-226 C11H10N4Ob 1670, 1570,
214 (100, M+);
186 (22); 144
(65); 118 (67)
1460, 1090,
760, 740
20 CH3
C6H4-4′-OH
A
60
67
>300
C16H12N4O2 1690, 1580,
1280, 1160,
3.84 (s, 3H, CH3); 6.85
(d, 2H, 3′,5′-H); 7.35-
7.72 (m, 3H, Ar-H);
8.05 (d, 2H, 2′,6′-H);
8.43 (dd, 1H, 6-H);
9.74 (s, 1H, OH)
292 (60, M+);
264 (25); 144
(82); 118 (100)
850, 760
21 CH3
C6H3-3′,4′-(OH)2
DMF/
H2O
282-284 C16H12N4O3 3500, 1680,
1560, 1460,
3.86 (s, 3H, CH3); 6.74- 308 (76, M+);
7.73 (m, 6H, Ar-H);
8.46 (dd, 1H, 6-H);
9.01 (bs, 2H, 3′,4′-OH)
280 (29); 144
(93); 118 (100)
1260, 750
22 CH3
23 C6H5
thien-3-yl
C6H5
A
80
40
DMF
225-226 C14H10N4OS 1690, 1580,
3.87 (s, 3H, CH3); 7.30- 282 (100, M+);
1460, 1220,
810, 750
1670, 1580,
1450, 760
7.93 (m, 6H, Ar-H);
8.47 (dd, 1H, 6-H)
7.43-7.82 (m, 11H,
Ar-H); 8.20-8.24
(m, 2H, Ar-H);
254 (25); 144
(59); 118 (90)
338 (22, M+);
206 (81); 77
(100)
EtOH
228-230 C21H14N4O
8.62 (dd, 1H, 5-H)
6.72 (m, 1H, 4′-H);
7.43-7.47 (m, 9H,
Ar-H); 7.92 (d, 1H,
5′-H); 8.61 (dd, 1H,
6-H)
5.66 (s, 2H, CH2);
7.27-8.22 (m, 13H,
Ar-H); 8.50 (dd, 1H,
6-H)
24 C6H5
fur-2-yl
31
EtOH
248-250 C19H12N4O2 1670, 1580,
328 (34, M+);
206 (34); 93
(100); 77 (80)
1420, 1140,
760
25 CH2C6H5 C6H5
A
A
70
36
DMF
DMF
180-181 C22H16N4O
1700, 1560,
1380, 1280,
1150, 750
352 (43, M+);
324 (2); 220
(19); 91 (100)
26 CH2C6H5 C6H4-4′-OH
287-289 C22H16N4O2 1660, 1600,
1560, 1280,
5.62 (s, 2H, CH2);
6.86 (d, 2H,
368 (44, M+);
340 (2); 220
(16); 91 (100)
1210, 760
3′,5′-H); 7.27-7.68
(m, 8H, Ar-H); 8.09
(d, 2H, 2′,6′-H);
8.48 (dd, 1H, 6H);
9.76 (s, 1H, OH,
exch. with D2O)
5.64 (s, 2H, CH2);
6.59-6.65 (m, 1H,
4′-H); 7.23-7.60
(m, 9H, Ar-H);
7.78 (d, 1H, 5′-H);
8.47 (dd, 1H, Ar-H)
5.64 (s, 2H, CH2);
7.24-7.63 (m, 10H,
Ar-H); 7.92 (dd,
1H, Ar-H); 8.50
(dd, 1H, 6-H)
27 CH2C6H5 fur-2-yl
A
A
25
76
DMF/
H2O
233-235 C20H14N4O2 1680, 1570,
342 (7, M+);
220 (4); 91
(100)
1310, 1020,
750
28 CH2C6H5 thien-3-yl
DMF
210-212 C20H14N4OS 1680, 1450,
358 (36, M+);
330 (1); 220
(12); 91 (100)
1140, 800,
750
a
b
Elemental analyses for C, H, N were within (0.4% of the calculated values. Literature ref 82.
ethanol at 0 °C, then refluxing the reaction mixture for
4 h; (B) by refluxing the acids for 1 h in glacial acetic
acid.
In the reaction of compound 30 with the appropriate
glyoxylic acids in refluxing ethanol, it was not possible
to isolate the intermediate hydrazono derivatives, but
the target products 23 and 24 were directly obtained
(Scheme 1, Table 3).
that a phenyl ring in place of the methyl at the N10-
position increases affinity by 2.6-3.5-fold (compare 23
and 24 vs 12 and 16), compound 23 being the most
potent of all the investigated ATBIs with a Ki value of
18 nM. In contrast, the insertion of the benzyl at the
N10-position lowered affinity by 1.2-5.7-fold (compare
25-28 vs 12, 20, 16, and 22). These data argue for a
hydrophobic interaction at the L1 site which is sensitive
to steric effects. The phenyl ring of the benzyl moiety is
oriented to a large extent out of the plane of the
The results of the radioligand binding assays per-
formed on the new ATBI compounds (Table 1) revealed

3-Aryl[1,2,4]triazino[4,3-a]benzimidazol-4(10H)-ones
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 321
F igu r e 3. Serpentine model of bovine A1AR sequence. The black lines represent the boundaries of the membrane. Filled circles
indicate the residues highly conserved among the GPCRs superfamily. The TM helices are denoted by roman numerals. The
arabic numbers indicate the position of the residues inside the TM domain. Glycosilation site on the EL-II is also shown. IL )
intracellular loop and EL ) extracellular loop.
Ta ble 4. Intrinsic Activity of ATBI Derivatives 3, 12, 23, and
25 to A1AR Expressed as GTP Shift
Ta ble 5. Geometries of the Interhelical Hydrogen Bonds in the
Model of Bovine A1 Adenosine Receptor
acceptorb
(A)
donorb
(D)
H-Ac D-H-Ad
a
a
Ki A1
Ki A1
no.
-GTP (nM)
+GTP (nM)
GTP shift
residuea
residuea
(Å)
(deg)
R-PIA
3
12
23
25
4.2 ( 0.3
141.2 ( 13
93.1 ( 7.3
24.7 ( 2.2
240 ( 19
19.9 ( 1.4
197.3 ( 16
119.2 ( 10
25.0 ( 2.3
184 ( 15
4.7
1.3
1.3
1.0
0.8
Ile15
Ser23
Cys46
O
O
O
His278
Asn 27
Ser50
Asn 27
Asn 284
Arg291
Thr141
Gln92
Tyr182
His251
Thr257
Ser277
Ser 94
NδH
2.2
2.2
2.1
2.3
2.3
2.3
2.0
2.3
2.1
2.1
1.9
2.3
2.3
2.4
170
155
147
150
153
155
177
130
151
140
167
164
139
138
Nδ2
H
OγH
Asp 55
Asp 55
Asp 104
Val137
Val138
Glu178
Tr p 247
Leu253
Ala273
Asn 284
Asn 284
Oδ1
Oδ2
Oδ2
O
Nδ1
Nδ1
H
H
Nη2H2
OγH
a
Displacement of [3H]DPCPX from bovine cortical membranes
in the absence and in the presence of 1mM GTP. Values are taken
from four experiments, expressed as means ( SEM.
O
N
ꢀ2H
Oꢀ1
O
OγH
NδH
OγH
OγH
OγH
O
molecular mainframe and probably, in such an arrange-
ment, it cannot fit properly into the L1 site.
O
Oδ1
Oδ1
At the A1AR, the selected compounds 3, 12, 23, and
25 displayed no significative GTP shift, suggesting that
they elicited an antagonist profile. In contrast, the
standard agonist R-PIA exhibited a larger GTP shift
value of 4.7 (Table 4).
Con str u ction of a Mod el of th e Bovin e A1AR a n d
Dock in g Sim u la tion s. Models of the A1AR complexed
with antagonists were developed to aid the interpreta-
tion of the SARs in the class of A1 antagonists and,
perspectively, to design new ATBI derivatives and
analogues.
A bacteriorhodopsin-based model for the canine A1AR
has been proposed by IJ zerman and co-workers.21
However, given the drawbacks of choosing bacterio-
rhodopsin as a template,34 we constructed a model of
bovine A1AR35 using as a template the frog rhodop-
sin,36,37 a membrane protein belonging to the G-protein
coupled receptors (GPCRs) superfamily.
Asn 280
Nδ2
H
a
Residues highly conserved in GPCRs are in bold. Residues
b
specifically conserved in ARs subtypes are underlined. Atom
names of the amino acids are based on IUPAC nomenclature.83
d
c H-A is the hydrogen-acceptor distance. D-H-A is the donor-
acceptor angle.
The computational procedures described in the Ex-
perimental Section resulted in a protein structure quite
stable during the molecular dynamics (MD) simulation
and deviating from the initial geometry by a root-mean-
square (rms) distance of 1.5 Å about the backbone
atoms. Figure 3 is a serpentine scheme of bovine A1AR
sequence. Examination of the receptor model reveals
two clusters of polar and aromatic side chains that form
extensive networks of interhelical contacts. Table 5
details the interhelical hydrogen bonds computed over
the minimized average structure of the protein. In this
table, residues highly conserved in GPCRs and residues

322 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Da Settimo et al.
F igu r e 4. Theoretical models of the A1AR binding site complexed with compounds III (a), VII (b), and 23 (c). Only amino acids
located within 5 Å from any atom of the bound ligand are displayed. Dotted lines highlight ligand/receptor hydrogen bonds.
specifically conserved in ARs are in bold and are under-
lined, respectively. Two hydrogen bonds received by
Asp55 (TMII) from Asn27 (TMI) and Asn284 (TMVII),
whose role in maintaining the structural organization
and functional integrity of the protein has been experi-
mentally ascertained,38-40 were initially imposed through
specific constraints and found to persist throughout the
MD simulation. The seven-helical bundle is also stabi-
lized by a network of π-stacking interactions41 involving
a number of residues highly conserved in GPCRs
(Phe243 and Trp247 in helix VI along with Tyr200 in
helix V) and peculiar of the ARs family (Phe185 and
Trp188 in helix V together with His251 in helix VI). Our
model is also in agreement with the substituted cysteine
accessibility studies of J avitch et al.,42-45 who placed
various conserved residues and residues accessible to
polar Cys probes on helices III, V, VI, and VII inside
the core of the transmembrane bundle.
Compound VII was docked into the model of A1AR
using the automated DOCK 3.5 suite of programs.46-49
This ligand was selected as representative of A1AR
antagonists primarily because, compared with the struc-
tures reported in Figure 1 and the ATBI derivative 23,
it possesses low flexibility and all the six hypothesized
pharmacophoric features (see Figure 2). Among the
“best” orientations proposed by DOCK, we chose that
with the HN-CC-N substructure of VII facing the
Asn247 side chain so as to make two hydrogen bonds.

3-Aryl[1,2,4]triazino[4,3-a]benzimidazol-4(10H)-ones
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 323
Ta ble 6. Residues of the Bovine A1 Adenosine Receptor
Binding Cleft and Their Relationship with the Pharmacophoric
Binding Sitesa
siteb
residuesc
L1
Val87 (III), Leu88 (III), Thr91 (III), Ile95 (III),
Phe185 (V), Trp188 (V), Trp247 (VI), Leu250 (VI),
His251 (VI), Leu276 (VII), Asn280 (VII)
Leu81 (III), Leu88 (III), Trp146 (IV), Val181 (V),
Tyr182 (V), Phe185 (V), Thr257 (VI), Leu258 (VI)
Leu81 (III), Ala84 (III), Val87 (III), Leu88 (III),
Ile270 (VII), Ala273 (VII), Ile274 (VII),
Leu276 (VII)
L2
L3
HB1/HB2 Asn254 (VI)
HB3
Ser277 (VII)
a
Amino acids of the binding site are those located within 5 Å
b
from any atom of the docked ligands III, VII, and 23. These sites
are illustrated in Figure 2. c A1AR specific residues are underlined.
The transmembrane domain is given in parentheses.
This type of double hydrogen bond between an Asn
residue and an adenine derivative is a motif character-
izing several ligand/protein complexes50 filed in the
Brookhaven Protein Data Bank.51
The geometry of the VII/A1AR complex was then
optimized by MD and energy minimization (EM) cycles.
Compounds III52 and 23 were initially placed into the
receptor binding site by fitting them on the docked
conformation of VII according to the pharmacophoric
alignment shown in Figure 2. The resulting complexes
were then submitted to geometry refinement protocols
based on MD and EM calculations.
F igu r e 5. Overlay of the N10-benzyl derivative 25 (red) on
the docked N10-phenyl derivative 23 (yellow). The A1AR
binding site is portrayed as a solvent accessible surface and
is clipped to display the interaction between the N10-substit-
uents and the L1 lipophilic pocket.
5 illustrates a Connolly surface of the receptor hosting
the docked compound 23 (N10-phenyl) together with the
structure of the 25 (N10-benzyl) overlayed about their
common skeleton. The excellent degree of steric comple-
mentarity between the N10-phenyl and the L1 pocket
can be appreciated in contrast with the steric conflict
between the N10-benzyl group and a part of the recep-
tor57 probably requiring energetically expensive confor-
mational changes in one or both partners of the 25/A1AR
complex. Despite the many limitations of pharmaco-
phore and docking models (e.g., the possibility of alter-
nate and multiple binding modes58,59), our model of
A1AR complexed with 23 rationalizes the SARs in the
ATBI series and will be employed to guide further lead
optimization projects.
Figure 4 illustrates the docking models developed for
compounds III, VII, and 23. Table 6 summarizes the
amino acids located within 5 Å from the bound ligands
and their relationship with the pharmacophoric binding
sites HB1, HB2, HB3, L1, L2, and L3 (Figure 2). Site-
directed mutagenesis experiments have underscored the
importance of several residues listed in Table 6 in
favoring the binding of different A1AR antagonists:
Val87, Leu88, Thr91,53 Ser277,54,55 Ile270,55 and His251.56
The Asn247 side chain corresponds to the hydrogen
bond acceptor and donor sites HB1 and HB2, while
Ser277 might represent the HB3 donor site. A hydrogen
bond with Ser277 does not seem mandatory for high
affinity since this interaction is missing in the complex
of the highly potent ligand 23. At the end of the
geometry refinement calculations on the 23/A1AR com-
plex, we found that the N1 nitrogen of 23 receives a
hydrogen bond from the Asn247 side chain (Figure 4c)
whereas it was the vicinal N2 nitrogen to be initially
considered as a pharmacophoric element in 12 (Figures
1 and 2).
The L2 site, assumed to be relatively small as inferred
from the SARs in the ATBI series, is located at the
binding site entrance of the receptor. It is likely that
the steric borders of the L2 region are atoms of the
extracellular loops II and III not included in our model
for obvious reasons. It is tempting to speculate that the
side chains of Glu178 and Tyr182 in helix V, surround-
ing the pendant phenyl moiety of the docked ligand,
might receive hydrogen bonds from the meta and para
hydroxyls which, as already stated, are the only sub-
stituents tolerated. In the previous paragraph, we
hypothesized that the lower affinity of the N10-benzyl
compared with the N10-phenyl ATBI derivatives could
be related to a hydrophobic interaction at the L1 site
strongly influenced by shape-dependent effects. Figure
Con clu sion s
We have disclosed a series of 3-aryltriazinobenzimid-
azolones (ATBIs) as a novel class of A1AR antagonists
generally selective over A2A and A3ARs. Attempts to
improve A1AR affinity in the ATBIs, guided by an
overlay of the lead compound 12 (Ki ) 63 nM) on some
well-known A1AR antagonists, yielded compound 23
characterized by a Ki of 18 nM. Finally, we built a model
of the bovine A1AR to rationalize the SARs of ATBIs
and to facilitate, perspectively, the design of new
analogues.
Exp er im en ta l Section
Ch em ist r y. Melting points were determined using
a
Reichert Ko¨fler hot-stage apparatus and are uncorrected.
Infrared spectra were obtained on a PYE/UNICAM mod. PU
9561 spectrophotometer in Nujol mulls. Nuclear magnetic
resonance spectra were recorded in DMSO-d6 solutions with
a Varian Gemini 200 or a Varian CFT-20 spectrometer using
tetramethylsilane (TMS) as an internal standard. Mass spectra
were obtained on a Hewlett-Packard 5988 A spectrometer
using a direct injection probe and an electron beam energy of
70 eV. Magnesium sulfate was always used as the drying

324 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Da Settimo et al.
agent. Evaporations were made in vacuo (rotating evaporator).
Analytical TLC was carried out on Merck 0.2 mm precoated
silica gel aluminum sheets (60 F-254). Silica gel 60 (230-400
mesh) was used for flash column chromatography. Elemental
analyses were performed by our Analytical Laboratory and
agreed with theoretical values within ( 0.4%.
Besides the commercially available starting materials,
4-hydroxyphenyl glyoxylic acid60 and 3-thienylglyoxylic acid61
were prepared in accordance with the reported methods.
3-Su bstitu ted 1-Meth yl (18-20,22) a n d 3-Su bstitu ted
1-Ben zyl[1,2,4]tr ia zin o[4,3-a ]ben zim id a zol-4(10H)-on es
(25-28). A solution of 1-methyl- (29) or 1-benzyl-2-hydrazino-
benzimidazole (31) (4 mmol) and the appropriate glyoxylic acid
(4.4 mmol) in 10 mL of ethanol was refluxed for 2 h. The
mixture was cooled, and the precipitate was collected to give
the 2-substituted 2-(benzimidazol-2-ylhydrazono)acetic acids
32-35 and 36-39 which were purified by suspension in hot
methanol (Table 2).
They can be cyclized by means of the following methods.
Meth od A. A suspension of the acid derivatives 34, 35 and
36-39 (2 mmol) in 5 mL of absolute ethanol, at 0 °C, was
saturated with anhydrous HCl and then refluxed for 4-7 h,
and the reaction was monitored by TLC analysis. The reaction
mixture was evaporated to dryness, and the solid residue was
treated with saturated sodium hydrogencarbonate aqueous
solution. The crude product obtained was purified by recrys-
tallization from the appropriate solvent (Table 3).
Meth od B. A suspension of the acids 32, 33 (2 mmol) in 20
mL of glacial acetic acid was refluxed for 1 h. The solution
obtained was evaporated to dryness, and the oily residue was
filtered on a silica gel chromatographic column (h ) 4 cm; φ )
2.5 cm; eluting system ) toluene:chloroform ) 1:1). The crude
solid product obtained was purified by recrystallization from
the appropriate solvent (Table 3).
3-Su bstitu ted 1-P h en yl[1,2,4]tr ia zin o[4,3-a ]ben zim id -
a zol-4(10H)-on es (23, 24). A solution of 30 (4 mmol) and the
appropriate glyoxylic acid (4.4 mmol) in 10 mL of absolute
ethanol was refluxed for 3-5 h (TLC analysis). After cooling,
the precipitate that formed was collected and purified by
recrystallization from the appropriate solvent (Table 3).
3-(3′,4′-Dih yd r oxyp h en yl)-1-m eth yl[1,2,4]tr ia zin o[4,3-
a ]b en zim id a zol-4(10H )-on e (21). A solution of boron tri-
bromide (6 mmol) in 1 mL of anhydrous dichloromethane was
added dropwise, at -10 °C, under stirring and in a nitrogen
atmosphere, to a suspension of the dimethoxy derivative 151
(0.9 mmol) in 8 mL of the same solvent. Stirring was continued
for 30 min at -10 °C and for 1 h at room temperature. The
reaction mixture was carefully added dropwise to 5 mL of
methanol at -10 °C. The suspension obtained was evaporated
to dryness, and the solid residue was treated with saturated
sodium hydrogen carbonate aqueous solution. The crude
product 21 was collected and purified by recrystallization
(Table 3).
Biologica l Meth od s. Ma ter ia ls. [3H]-(R)-PIA (25 Ci/mmol)
was purchased from Amersham Corp. (Little Chalfont, Buck-
inghamshire, U.K.), while [3H]CHA, [125I]AB-MECA, and
[3H]CGS 21680 were obtained from DuPont-NEN (Boston,
MA). DPCPX was purchased from RBI (Natik, MA). Adenosine
deaminase was from Sigma Chemical Co. (St. Louis, MO). All
other reagents were from standard commercial sources and
of the highest commercially available grade.
Recep tor Bin d in g Assa ys. A1 a n d A2A Recep tor Bin d -
in g. Displacement of [3H]CHA (31 Ci/mmol) from A1AR in
bovine cortical membranes and of [3H]CGS 21680 (42.1 Ci/
mmol) from A2AAR in bovine striatal membranes was per-
formed as described.62 Adenosine A1 receptor affinities with
[3H]DPCPX as radioligand were determined according to
Pirovano et al.63 Measurements with [3H]DPCPX were per-
formed in the presence and in the absence of 1 mM GTP.
A3AR Bin d in g. [125I]AB-MECA binding to A3AR in bovine
cortical membranes was performed in 50 mM Tris, 10 mM
MgCl2, and 1 mM EDTA buffer (pH 7.4) containing 0.2 mg of
proteins and 2 U/mL adenosine deaminase and 20 nM DPCPX.28
Incubations were carried out in duplicate for 90 min at 25 °C.
Nonspecific binding was determined in the presence of 50 µM
R-PIA and constituted approximately 30% of the total binding.
Binding reaction was terminated by filtration through a
Whatman GF/C filter, washing three times with 5 mL of ice-
cold buffer.
All compounds were routinely dissolved in DMSO and
diluted with assay buffer to the final concentration, where the
amount of DMSO never exceeded 2%.
At least six different concentrations spanning 3 orders of
magnitude, adjusted appropriately for the IC50 of each com-
pound, were used. IC50 values, computer-generated using a
nonlinear regression formula on a computer program (Graph-
Pad, San Diego, CA), were converted to Ki values, knowing
the Kd values of radioligands in the different tissues and using
the Cheng and Prusoff equation.64 The dissociation constant
(Kd) of [3H]CHA, [3H]CGS 21680, and [125I]AB-MECA were 1.2,
14, and 1.02 nM, respectively.
Com p u t a t ion a l P r oced u r es. A1 Ad en osin e R ecep t or
Mod el Bu ild in g. All model building, energy minimizations,
and molecular dynamics calculations were carried out using
SYBYL 6.265 and AMBER 4.166,67 modeling packages, respec-
tively. All manipulations were performed on a Silicon Graphics
R10000 workstation.
The structural model of the bovine A1AR has been derived
from electron cryomicroscopy data36 and the CR coordinate
template provided by Baldwin et al.37 This template was
generated from rhodopsin projection data in combination with
a sequence analysis of 493 GPCRs, including ARs.37 The seven
TM helical domains were identified with the aid of Kyte-
Doolittle hydrophobicity68 and Emini surface probability pa-
68
rameters. The length of the helices was determined by
considering the minimum length in the lipid bilayer, as defined
by Baldwin, with subsequent addition of residues at the
extracellular and intracellular limits based on sequence
analysis of this region. Individual TM helical segments were
built as ideal helices (using φ-ψ angles of -63.0° and -41.6°)
with side chains placed in prevalent rotamers and representa-
tive proline kink geometries. Each helix was capped with an
acetyl group and an N-methyl group at the N-terminus and
at the C-terminus, respectively. These structures were then
grouped together to form a helical bundle matching the overall
characteristics of the electron density map of rhodopsin.
The relative orientations and interactions between the
helices were adjusted based on incorporated structural infer-
ences from available experimental data, such as mutation and
ligand binding studies,69-71 cysteine scanning data,42-45 and
site-directed mutation experiments.38-40 Because earlier work
showed that polarity conserved positions cluster together in
the cores of proteins to create conserved hydrogen-bonding
interactions,72 we refined the model by applying the additional
hydrogen-bonding constraints between the conserved polar
residues Asn27, Asp55, and Asn284 in accordance with data
from site-directed mutagenesis.38-40
The helical bundle was subjected to energy-minimization
using the SANDER module of the AMBER suite of pro-
grams66,67 until the rms value of the coniugate gradient was
0.001 kcal/mol/Å. An energy penalty force constant of 5 kcal/
Å2/mol to the protein backbone atoms was applied throughout
these calculations. The resultant model was further used as
the starting point for a series of short MD simulations at 300
K, during which the positional constraints on the protein
backbone atoms were gradually released from 5 to 0.05 kcal/
Å2/mol. A 1 fs time step was used along with a nonbonded
cutoff of 8 Å, and the nonbonded pair list updated every 25 fs.
The temperature of the system was maintained at 300 K using
the Berendsen algorithm73 with a 0.2 ps coupling constant.
The structural quality of the model was also evaluated using
the PROCHECK 3.4.4 program,74 which examines the values
of the dihedral angles of the protein backbone and side chains.
Analysis of main chain torsion angles showed that 98% of the
residues are located in the most favored helical regions, the
remaining 2% occurring in the additional allowed helical
regions of the Ramachandran plot. Side chain ø1 and ø2

3-Aryl[1,2,4]triazino[4,3-a]benzimidazol-4(10H)-ones
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 325
dihedrals were determined to be in their most favorable
regions without steric conflicts.
nonbonded pairlist updated every 25 fs were used for the MD
simulations. The temperature was mantained at 300 K using
the Berendsen algorithm73 with a 0.2 ps coupling constant.
Four snapshots, extracted each 25 ps from the last 100 ps MD
simulation, were very similar in terms of rms deviation. An
average structure was calculated from the last 100 ps trajec-
tory and energy-minimized using the steepest descent and
conjugate gradient methods available within the SANDER
module of AMBER as specified above.
P h a r m a cop h or e-Ba sed Align m en t a n d Dock in g Sim u -
la tion s. The core structures of the ligands were retrieved from
the Cambridge Structural Database (CSD)75 and modified
whenever necessary using the SYBYL fragment library. The
resulting structures were fully optimized using the semi-
empirical quantum mechanics method AM176 as implemented
in the MOPAC 6.0 program77 (keywords: PREC, GNORM )
0.01, EF, MMOK). Molecular graphics, rms fit, and calcula-
tions of ring centroids were also carried out using SYBYL.
The DOCK 3.5 suite of programs46-49 was used for auto-
mated docking simulations of compound VII to the model-built
TM domain of the A1AR. DOCK identifies specific orientations
of a given ligand which are complementary to a targeted
surface area. In brief, DOCK first generates a negative image
of the ligand binding site as a set of overlapping spheres whose
centers become the potential locations for ligand atoms. To
rank each potential binding mode, a precalculated contact
scoring grid (based on distances between potential ligand and
target area atoms) and a force field scoring grid (based on steric
and electrostatic interaction energies) are generated. The
resulting output file for each screen, based on distance or force
field grids, contains the highest scoring orientations ranked
in order of their scores.
In absence of any information concerning the 3D structure
of the loops of GPCRs, these domains were not considered in
our study. On the other hand, site-directed mutagenesis
experiments and studies on chimeric A1/A2/A3ARs53-56 strongly
suggest that ligand/receptor binding occurs at the TM domain
of these receptors. A molecular surface representation of all
residues within 6 Å of the putative ligand binding pocket of
the A1AR TM helices was obtained by using the MS pro-
gram78,79 and a probe of 1.4 Å radius. The surface points and
their associated normals were adopted using the module
SPHGEN80 to fill the active site with a set of overlapping
spheres, as descriptors of the volume available to the ligand.
A cutoff distance of 4.5 Å for “good-contacts” with the receptor
was applied.46-49 The specific sphere cluster utilized was a
manually edited cluster comprising 42 individual spheres,
which satisfactorily covered the target area. The module
CHEMGRID46 was used to precompute and save in a grid file
the information necessary for force field scoring. The molecules
were docked into the receptor binding pocket using the single
mode option in DOCK, and the quality of the ligand binding
was evaluated by a force field scoring function. Each orienta-
tion of each ligand was filtered for steric fit through a
DISTMAP grid48 with polar and nonpolar contact limits of 1.3
and 1.8 Å, respectively.
Connolly solvent accessible surfaces of the A1AR binding site
(based on a probe radius of 1.4 Å) were generated using the
SYBYL/SURFACE routine run under default settings.
Ack n ow led gm en t. This work was supported by
grants from the Ministry of University and Scientific
and Technological Research (MURST).
Not e Ad d ed in P r oof: After this manuscript was sub-
mitted, a paper was published disclosing the crystal structure
of rhodopsin solved diffractometrically at 2.8 Å resolution
(Palczewski, K.; Kumasaka, T.; Hori, T.; Behnke, C. A.;
Motoshima, 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. Sci-
ence 2000, 289, 739-745). A comparison of our A1AR model
with this crystal structure revealed slight differences in the
orientations of the transmembrane helices and a satisfying
overlap about the residues of the putative adenosine binding
pocket.
Refer en ces
(1) Primofiore, G.; Da Settimo, F.; Taliani, S.; Marini, A. M.; La
Motta, C.; Novellino, E.; Greco, G.; Gesi, M.; Trincavelli, L.;
Martini, C. 3-Aryl[1,2,4]triazino[4,3-a]benzimidazol-4(10H)-ones:
Tricyclic Heteroaromatic Derivatives as a New Class of Benzo-
diazepine Receptor Ligands. J . Med. Chem. 2000, 43, 96-102.
(2) Suzuki, F.; Shimada, J .; Nonaka, H.; Ishii, A.; Shiozaki, S.;
Ichikava, S.; Ono, E. 7,8-Dihydro-8-ethyl-2-(3-noradamantyl)-
4-propyl-1H-imidazo[2,1-i]purin-5(4H)-one: A Potent and Water-
Soluble Adenosine A1 Antagonist. J . Med. Chem. 1992, 35,
3578-3581.
(3) Shimada, J .; Suzuki, F.; Nonaka, H.; Ishii, A. Polycycloalkyl-
1,3-dipropylxanthines as Potent and Selective Antagonists for
A1-Adenosine Receptor. J . Med. Chem. 1992, 35, 924-930.
(4) Suzuki, F.; Shimada, J .; Mizumoto, H.; Karasawa, A.; Kubo, K.;
Nonaka, H.; Ishii, A.; Kawakita, T. Adenosine A1 Antagonists
2. Structure-Activity Relationships on Diuretic Activities and
Protective Effects Against acute Renal Failure. J . Med. Chem.
1992, 35, 3066-3075.
(5) Bruns, R. F.; Lu, G. H.; Pugsley, T. A Characterization of the
A2 Adenosine Receptor Labeled by [3H]NECA in Rat Striatal
Membranes. Mol. Pharmacol. 1986, 29, 331-346.
(6) Hamilton, H. W.; Ortwine, D. F.; Worth, D. F.; Bristol, J . A.
Synthesis and Structure-Activity Relationships of Pyrazolo[4,3-
d]pyrimidin-7-one as Adenosine Receptor Antagonists. J . Med.
Chem. 1987, 30, 91-96.
(7) Sarges, R.; Howard, H. R.; Browne, R. G.; Lebel, L. A.; Seymour,
P. A.; Koe, B. K. 4-Amino[1,2,4]triazolo[4,3-a]quinoxalines. A
Novel Class of Potent Adenosine Receptor Antagonists and
Potential Rapid-Onset antidepressants. J . Med. Chem. 1990, 33,
2240-2254.
(8) 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.; Willams, M. Structure-Activity
Profile of a series of Novel Triazoloquinazoline Adenosine
Antagonists. J . Med. Chem. 1988, 31, 1014-1020.
The putative pharmacophore-based conformation of VII was
directly docked. Out of the 3057 orientations obtained, 75 were
within 5 kcal/mol of the best orientation based on the scoring
function. Inspection of the docked structures revealed that the
best and many of the top scoring orientations placed the HN-
CC-N substructure of VII facing the Asn247 side chain so as
to make two hydrogen bonds. From this cluster of structures,
we selected one (force field score -22.1 kcal/mol, 2 kcal/mol
above the top scoring orientation) in which the nitrogen closest
to the pendant phenyl ring made a hydrogen bond with the
Ser277 side chain.
Refinement of the ligand/receptor bound complex was
achieved by in vacuo energy minimization with the SANDER
module of AMBER 4.1 (50 000 steps; distance dependent
dielectric function of ꢀ ) 4r), applying an energy penalty force
constant of 5 kcal/mol on the protein backbone atoms. The
geometry-optimized structures were then used as the starting
point for subsequent 200 ps MD simulation, during which the
protein backbone atoms were constrained as done in the
previous step. The simulations employed the Cornell force
field,81 as implemented in the AMBER 4.1 suite of pro-
grams.66,67 The additional parameters required for the ligands
were derived by analogy to existing parameters. Partial atomic
charges for the ligands were imported from the output files of
AM1 full geometry optimizations. A time step of 1 fs and a
(9) van Galen, P. J . M.; Nissen, P.; Van Wijngaarden, I.; IJ zerman,
A. P.; Soudijn, W. 1H-Imidazo[4,5-c]quinolin-4-amines: Novel
Non-Xanthine Adenosine Antagonists. J . Med. Chem. 1991, 34,
1202-1206.
(10) Mu¨ller, C. E.; Hide, I.; Daly, J . W.; Rothenha¨usler, K.; Eger, K.
7-deaza-2-phenyladenines: Structure-Activity Relationships of
Potent A1 Selective Adenosine Receptor Antagonists. J . Med.
Chem. 1990, 33, 2822-2828.
(11) J acobson, K. A.; van Galen, P. J . M.; Williams, M. Adenosine
Receptors-Pharmacology, Structure-Activity Relationships, and
Therapeutic Potential. J . Med. Chem. 1992, 35, 407-422.
(12) Von Lubitz, D. K. J . E.; J acobson, K. A. Behavioral Effects of
Adenosine Receptor Stimulation. In Adenosine and Adenine
Nucleotides: From Molecular Biology to Integrative Physiology;
Belardinelli, L., Pelleg, A., Eds.; Kluwer: Norwell, 1995; pp 489-
498.

326 J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3
Da Settimo et al.
(13) Collis, M. G.; Hourani, S. M. O. Adenosine Receptor Subtypes.
Trends Pharmacol. Sci. 1993, 14, 360-366.
(38) Perlman, J . H.; Colson, A. O.; Wang, W.; Bence, K.; Osman R.;
Gershengorn, M. C. Interactions Between Conserved Residues
in Transmembrane Helices 1, 2, and
7 of the Tyrotropin-
(14) Poulsen, S.-A.; Quinn, R. J . Adenosine Receptors for Future
Drugs. Bioorg. Med. Chem. 1998, 6, 619-641.
Releasing Hormone Receptor. J . Biol. Chem. 1997, 272, 11937-
11942.
(15) Kim, H. O.; J i, X. D.; Melman, N.; Olah, M. E.; Stiles, G. L.;
J acobson, K. A. Structure-Activity Relationships of 1,3-Dialkyl-
xanthine Derivatives at Rat A3 Adenosine Receptors. J . Med.
Chem. 1994, 37, 3373-3382.
(16) van Galen, P. J . M.; van Bergen, A. H.; Gallo-Rodriguez, C.; Olah,
M. E.; IJ zerman, A. P.; Stiles, G. L.; J acobson, K. A. A Binding
Site Model and Structure-Activity Relationships for the Rat A3
Adenosine Receptor. Mol. Pharmacol. 1994, 45, 1101-1111.
(17) Peet, N. P.; Lentz, N. L.; Meng, E. C.; Dudley, M. W.; Ogden, A.
M. L.; Demeter, D. A.; Weintraub, H. J . R.; Bey, P. A Novel
Synthesis of Xanthines: Support for a New Binding Mode for
Xanthines with Respect to Adenosine at Adenosine Receptors.
J . Med. Chem. 1990, 33, 3127-3130.
(18) van Galen, P. J . M.; van Vlijmen, H. W. T.; IJ zerman, A. P.;
Soudijn, W. A Model for the Antagonist Binding Site on the
Adenosine A1 Receptor Based on Steric, Electrostatic, and
Hydrophobic Properties. J . Med. Chem. 1990, 33, 1708-1713.
(19) Quinn, R. J .; Dooley, M. J .; Escher, A.; Harden, F. A.; J ayasuriya,
H. A Computer Generated Model of Adenosine Receptors Ra-
tionalising binding and Selectivity of Receptor Ligands. Nucleo-
sides Nucleotides 1991, 10, 1121-1124.
(20) Dooley, M. J .; Quinn, R. J . The Three Binding Domain Model of
Adenosine Receptors: Molecular Modeling Aspects. J . Med.
Chem. 1992, 35, 211-216.
(21) IJ zerman, A. P.; van Galen, P. J . M.; J acobson, K. A. Molecular
Modeling of Adenosine Receptors. I. The Ligand Binding Site
on the A1 Receptor. Drug Des. Discovery 1992, 9, 49-67.
(22) Dudley, M. W.; Peet, N. P.; Demeter, D. A.; Weintraub, H. J . R.;
Herschel, J . R.; IJ zerman, A. P.; Nordvall, G.; van Galen, P. J .
M.; J acobson, K. A. Adenosine A1 Receptor and Ligand Molecular
Modeling. Drug Des. Discovery 1993, 28, 237-243.
(23) IJ zerman, A. P.; van der Wenden, E. M.; van Galen, P. J . M.;
J acobson, K. A. Molecular Modeling of Adenosine Receptors -
The Ligand-Binding Site on the Rat Adenosine A2a Receptor.
Eur. J . Pharmacol. Mol. Pharmacol. Sect. 1994, 268, 95-104.
(24) Dooley, M. J .; Kono, M.; Suzuki, F. Theoretical Structure-
Activity Studies of Adenosine A1 Ligands: Requirements for
Receptor Affinity. Bioorg. Med. Chem. 1996, 6, 923-934.
(25) The correspondences between Dooley’s and our receptor binding
sites are herein reported: N7 position/HB1, O6 oxygen/HB2, C8-
position/L1, N1-position/L2, and N3-position/L3 (see Figure 7
in ref 15).
(39) Sealfon, S. C.; Chi, L.; Eversole, B. J .; Rodic, V.; Zhang, D.;
Ballesteros, J . A.; Weinstein, H. Related Contribution of Specific
Helix 2 and 7 Residues to Conformational Activation of the
Serotonin 5-HT2a Receptor. J . Biol. Chem. 1995, 270, 16683-
16688.
(40) Zhou, W.; Flanagan, C. A.; Ballesteros, J .; Konvicka, K.; David-
son, J . S.; Weinstein, H.; Millar, R. P.; Sealfon, S. C. A Reciprocal
Mutation Supports Helix
2 and Helix 7 Proximity in the
Gonadotropin-releasing Hormone Receptor. Mol. Pharmacol.
1994, 45, 165-170.
(41) Burley, S. K.; Petsko, G. A. Aromatic-Aromatic Interaction: a
Mechanism of Protein Structure Stabilitation. Science 1995, 229,
23-28.
(42) J avitch, J . A.; Li, X.; Kaback, J .; Karlin, A. A Cysteine Residue
in the Third Membrane-Spanning Segment of the Human D2
Dopamine Receptor Is Exposed in the Binding-Site Crevice. Proc.
Natl. Acad. Sci. U.S.A. 1994, 91, 10355-10359.
(43) J avitch, J . A.; Fu, D. Y.; Chen, J . Y.; Karlin, A. Residues in the
Fifth Membrane-Spanning Segment of the Dopamine D2 Recep-
tor Exposed in the Binding-Site Crevice. Biochemistry 1995, 34,
16433-16439.
(44) J avitch, J . A.; Fu, D.; Liapakis, G.; Chen, J . Constitutive
Activation of the 2 Adrenergic Receptor Alters the Orientation
of its Sixth Membrane-Spanning Segment. J . Biol. Chem. 1997,
272, 18546-18549.
(45) Fu, D.; Ballesteros, J . A.; Weinstein, H.; Chen, J .; J avitch, J . A.
Residues in the Seventh Membrane-Spanning Segment of the
Dopamine D2 Receptor Accessible in the Binding-Site Crevice.
Biochemistry 1996, 35, 11278-11285.
(46) Meng, E. C.; Shoichet, B. K.; Kuntz, I. D. Automated Docking
with Grid-Based Energy Evaluation. J . Comput. Chem. 1992,
13, 505-524.
(47) Meng, E. C.; Gschwend, D. A.; Blaney, J . M.; Kuntz, I. D.
Orientational Sampling and Rigid-Body Minimization in Mo-
lecular Docking. Proteins: Struct., Funct., Genet. 1993, 17, 266-
278.
(48) Shoichet, B. K.; Bodian, D. L.; Kuntz, I. D. Molecular Modeling
Using Shape Descriptors. J . Comput. Chem. 1992, 13, 380-397.
(49) Connolly, M.; Gschwend, D. A.; Good, A. C.; Oshiro, C.; Kuntz,
I. D. DOCK 3.5, Department of Pharmaceutical Chemistry,
University of California, San Francisco, CA, 1995.
(50) The RELIBase service was used to retrieve the complexes filed
with the following PDB entry codes: 1A82; 1B8O; 1DAG; 1UOX;
1VFN; 1MUD; 1ZIN; 1DAH; 1BX4; 4HOH; 1KI5; 1AGR; 1QHI;
1GNQ; 1CLU; 1KI2; 1ULB; 1GNP; 1AZ1; 1RH3. RELIBase is a
Web-based service to retrieve ligand/receptor structures depos-
ited in the Brookhaven Protein Data Bank using substructure
queries (http://pdb.pdb.bnl.gov:8081/home.html).
(51) Bernstein, F. C.; Koetzle, T. F.; Williams, G. J . B.; Meyer, E. F.,
J r.; Brice, M. D.; Rodgers, J . R.; Kennard, O.; Shimanouchi, T.;
Tasumi, M. The Protein Data Bank: A Computer-Based Archival
File for Macromolecular Structure. J . Mol. Biol. 1977, 112, 535-
542.
(52) Compound III was selected for docking because it features,
similarly to the ATBI derivative 23, an aromatic moiety filling
the putative lipophilic L1 pocket and, similarly to compound VII,
all the six hypothesized pharmacophoric elements.
(53) Rivkees, S. A.; Barbhaiya, H.; IJ zerman, A. P. Identification of
the Adenine Binding Site of the Human A1 Adenosine Receptor.
J . Biol. Chem. 1999, 274, 3617-3621.
(54) Townsend-Nicholson, A.; Schofield, P. R. A Threonine Residue
in the 7th Transmembrane Domain of the Human A1-Adenosine
Receptor Mediates Specific Agonist Binding. J . Biol. Chem. 1994,
269, 2373-2376.
(55) Tucker, A.; Robeva, A. S.; Taylor, H. E.; Holeton, D.; Bockner,
M.; Lynch, K. R.; Linden, J . A1 Adenosine Receptors-2 Amino
acids are Responsible for Species-Differences in Ligand Recogni-
tion. J . Biol. Chem. 1994, 269, 27900-27906.
(56) Olah, M. E.; Ren, H. Z.; Ostrowski, J .; J acobson, K. A.; Stiles,
G. L. Cloning, Expression, and Characterization of the Unique
Bovine-A1 Adenosine Receptor-Studies on the Ligand Binding
Site by Site-Directed Mutagenesis. J . Biol. Chem. 1992, 267,
10764-10770.
(57) The orientation of 25 within the A1AR binding site, as depicted
in Figure 5, was obtained by a simple overlay of this ligand on
the docked geometry of 23 about their common skeleton.
(58) Mattos, C.; Ringe, D. Multiple Binding Modes. In 3D QSAR in
Drug Design. Theory Methods and Applications; Kubinyi, H., Ed.;
ESCOM: Leiden, 1993; pp 117-136.
(26) Bednyagina, N. P.; Postovskii, I. Ya. Benzazoles. 2-Hydrazino-
and 2-Azidobenzimidazoles. Zh. Obshchei Khim. 1960, 30, 1431-
1437; Chem. Abstr. 1961, 55, 1586h.
(27) de Zwart, M.; Kourounakis, A.; Kooijman, H.; Spek, A. L.; Link,
R.; von Frijtag Drabbe Ku¨nzel, J . K.; IJ zerman, A. P. 5′-N-
Substituted Carboxamidoadenosines as Agonists for Adenosine
Receptors. J . Med. Chem. 1999, 42, 1384-1392.
(28) J acobson, K. A.; Nikodijevic, O.; Shi, D.; Gallo-Rodriguez, C.;
Olah, M. E.; Stiles, G. L.; Daly, J . V. A Role for Central
A3-Adenosine Receptors Mediation of Behavioral Depressant
Effects. FEBS Lett. 1993, 336, 57-60.
(29) Tyurenkova, G. N.; Mudretsova, I. I.; Mertsalov, S. L.; Bednya-
gina, N. P. Synthesis, Structures and Tautomerism of Forma-
zans Containing
a 1-Arylbenzimidazol-2-yl Residue. Khim.
Geterotsikl. Soedin. 1980, 6, 818-821; Chem. Abstr. 1980, 93,
203744p.
(30) Robison, M. M. Chlorodehydroxylation of Nitrogen Heterocycles
with Phenylphosphonic Dichloride. J . Am. Chem. Soc. 1958, 80,
5481-5483.
(31) Iemura, R.; Kawashima, T.; Ito, K.; Tsukamoto, G. Synthesis of
2-(4-Substituted-1-piperazinyl)benzimidazoles as H1 Antihista-
minic Agents. J . Med. Chem. 1986, 29, 1178-1183.
(32) Clark, R. L.; Pessolano, A. A. Synthesis of Some Substituted
Benzimidazolones. J . Am. Chem. Soc. 1958, 80, 1657-1662.
(33) Eckert, H.; Forster, B. Triphosgene, a Crystalline Phosgene
Substitute. Angew. Chem., Int. Ed. Engl. 1987, 26, 894-895.
(34) Bikker J . A., Trumpp-Kallmeyer S., Humblet C. G-Protein
Coupled Receptors: Models, Mutagenesis, and Drug Design. J .
Med. Chem. 1998, 41, 2911-2927.
(35) The amino acid sequence of the bovine A1AR was used in these
studies to reflect the source of the binding affinity data consisting
of bovine brain A1AR membrane preparations.
(36) Unger, V. M.; Hargrave, P. A.; Baldwin J . M.; Schertler, G. F.
X. Arrangement of Rhodopsin Transmembrane Alpha Helices
Obtained by Electron Cryo-microscopy. Nature 1997, 389, 203-
206.
(37) Baldwin, J . M.; Schertler, G. F. X.; Unger, V. M. An Alpha-carbon
Template for the Transmembrane Helices in the Rhodopsin
Family of G-protein-coupled Receptors. J . Mol. Biol. 1997, 272,
144-164.
(59) Kubinyi, H. Similarity and Dissimilarity: A Medicinal Chemist’s
View. In 3D QSAR in Drug Design. Ligand-Protein Interactions
and Molecular Similarity; Kubinyi, H., Folkers, G., Martin, Y.
C., Eds.; KLUWER/ESCOM:1998; Vol. 2, pp 225-252.

3-Aryl[1,2,4]triazino[4,3-a]benzimidazol-4(10H)-ones
J ournal of Medicinal Chemistry, 2001, Vol. 44, No. 3 327
(60) Businelli, M. Metodi di Sintesi della p-Ossibenzaldeide a Partire
da Fenolo. Farmaco 1950, 5, 522-527.
(72) Zhang, D.; Weinstein, H. Polarity Conserved Positions in Trans-
membrane Domains of G-Protein Coupled Receptors and Bac-
teriorhodopsin. FEBS Lett. 1994, 337, 207-212.
(73) Berendsen, H. J . C.; Postma, J . P. M.; van Gunsteren, W. F.;
DiNola, A.; Haak, J . R. Molecular Dynamics with Coupling to
an External Bath. J . Chem. Phys. 1984, 81, 3684-3690.
(74) Laskowski, R. A.; McArthur, M. W.; Moss, D. S.; Thornton, J .
M. PROCHECK: A Program to Check the Stereochemical
Quality of Protein Structures. J . Appl. Crystallogr. 1993, 26,
283-291.
(75) Allen, F. H.; Bellard, S.; Brice, M. D.; Cartwright, B. A.;
Doubleday, A.; Higgs, H.; Hummelink, T.; Hummelink-Peters,
B. G.; Kennard, O.; Motherwell, W. D. S.; Rodgers, J . R.; Watson,
D. G. The Cambridge Crystallographic Data Centre: Computer-
Based Search, Retrieval, Analysis and Display of Information.
Acta Crystallogr. 1979, B35, 2331-2339.
(76) Dewar, M. J . S.; Zoebisch, E. G.; Healy, E. F.; Stewart, J . J . P.
AM1: a New General Purpose Mechanical Molecular Model. J .
Am. Chem. Soc. 1985, 107, 3902-3909.
(77) MOPAC (version 6.0) is available from Quantum Chemistry
Program Exchange, No. 455.
(78) Connolly, M. L. Solvent-accessible Surfaces of Proteins and
Nucleic Acids. Science 1983, 221, 709-713.
(79) Connolly, M. L. Analytical Molecular Surface Calculation. J .
Appl. Crystallogr. 1983, 16, 548-558.
(80) Kuntz, I. D.; Blaney, J . M.; Oatley, S. J .; Langridge, R.; Ferrin,
T. E. A Geometric Approach to Macromolecule-Ligand Interac-
tions. J . Mol. Biol. 1982, 161, 269-288.
(81) 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 Simula-
tion of Proteins, Nucleic Acids, and Organic Molecules. J . Am.
Chem. Soc. 1995, 117, 5179-5197.
(82) J oshi, K. C.; J ain, R.; Dandia, A.; Sharma, K. Synthesis of [1,2,4]-
Triazino[4,3-a]benzimidazol-4(10H)-ones as Antimicrobial Agents.
Indian J . Chem. 1989, 28B, 698-701.
(83) IUPAC-IUB Commission on Biochemical Nomenclature. Ab-
breviations and Symbols for the Description of the Conformation
of the Poly-Peptide Chains. J . Mol. Biol. 1970, 52, 1-17.
(61) Hatanaka, M.; Ishimaru, T. Synthetic Penicillins. Heterocyclic
Analogues of Ampicillin. Structure-Activity Relationships. J .
Med. Chem. 1973, 16, 978-984.
(62) 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.
(63) Pirovano, I. M.; IJ zerman, A. P.; van Galen, P. J . M.; Soudijn,
W. The Influence of Molecular Structure of N6-(ω-aminoalkyl)-
adenosines on Adenosine Receptor Affinity and Intrinsic Activity.
Eur. J . Pharmacol. 1989, 172, 185-193.
(64) 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.
Pharmacol. 1973, 22, 3099-3108.
(65) Sybyl Molecular Modelling System(version 6.2), Tripos Associ-
ates, St. Louis, MO.
(66) Pearlman, D. A.; Case, D. A.; Caldwell, J . W.; Ross, W. S.;
Cheatham T. E., III; Debolt, S.; Ferguson, D. M.; Seibel, G. L.;
Kollman, P. A. AMBER, a Package of Computer Programs for
Applying Molecular Mechanics, Normal-Mode Analysis, Molec-
ular Dynamics and Free Energy Calculations to Simulate the
Structural and Energetic Properties of Molecules. Comput. Phys.
Commun. 1995, 91, 1-41.
(67) Pearlman, D. A.; Case, D. A.; Caldwell, J . W.; Ross, W. S.;
Cheatham T. E., III; Ferguson, D. M.; Seibel, G.; Singh, U. C.;
Weiner, P. K.; Kollman, P. A. AMBER 4.1; Department of
Pharmaceutical Chemistry, University of California: San Fran-
cisco, CA, 1995.
(68) Kyte, J .; Doolittle, R. F. A Simple Method for Displaying the
Hydrophobic Character of a Protein. J . Mol. Biol. 1982, 157,
105-132.
(69) Baldwin, J . M. Structure and Function of Receptors Coupled to
G Proteins. Curr. Opin. Cell. Biol. 1994, 6, 180-190.
(70) Schwartz, T. W. Locating Ligand-Binding Sites in 7TM Receptors
by Protein Engineering. Curr. Opin. Biotechnol. 1994, 5, 434-
444.
(71) van Rhee, A. M.; J acobson, K. A. Molecular Architecture of G
Protein-Coupled Receptors. Drug Dev. Res. 1996, 37, 1-38.
J M001054M