Derivatives of the triazoloquinazoline adenosine antagonist (CGS 15943) having high potency at the human A(2B) and A3 receptor subtypes

Article abstract of DOI:10.1021/jm980094b

The adenosine antagonist 9-chloro-2-(2-furanyl)[1,2,4]triazolo[1,5- c]quinazolin-5-amine (CGS 15943) binds nonselectively to human A₁, A(2A), and A₃ receptors with high affinity. Acylated derivatives and one alkyl derivative of the 5-amino group and other modifications were prepared in an effort to enhance A(2B) or A₃ subtype potency. In general, distal modifications of the N⁵substituent were highly modulatory to potency and selectivity at adenosine receptors, as determined in radioligand binding assays at rat brain A₁ and A(2A) receptors and at recombinant human A₃ receptors. In Chinese hamster ovary cells stably transfected with human A(2B) receptor cDNA, inhibition of agonist-induced cyclic AMP production was measured. An N⁵-(2iodophenyl)acetyl derivative was highly selective for A(2A) receptors. An (R)-N⁵-α-methyl(phenylacetyl) derivative was the most potent derivative at A₃ receptors, with a K(i) value of 0.36 nM. A bulky N⁵-diphenylacetyl derivative, 13, displayed a K(i) value of 0.59 nM at human A₃ receptors and was moderately selective for that subtype. Thus, a large, nondiscriminating hydrophobic region occurs in the A₃ receptor in proximity to the N⁵-substituent. A series of straight-chain N⁵-aminoalkylacyl derivatives demonstrated that for A(2B) receptors the optimal chain length occurs with three methylene groups, i.e., the N⁵-γ-aminobutyryl derivative 27 which had a pA₂ value of 8.0 but was not selective for A(2B) receptors. At A₁, A(2A), and A₃ receptors however the optimum occurs with four methylene groups. An N⁵-pivaloyl derivative, which was less potent than 27 at A₁, A(2A), and A₃ receptors, retained moderate potency at A(2B) receptors. A molecular model of the 27-A(2B) receptor complex based on the structure of rhodopsin utilizing a 'cross-docking' procedure was developed in order to visualize the environment of the ligand binding site.

Full text of DOI:10.1021/jm980094b

J . Med. Chem. 1998, 41, 2835-2845
2835
Der iva tives of th e Tr ia zoloqu in a zolin e Ad en osin e An ta gon ist (CGS 15943)
Ha vin g High P oten cy a t th e Hu m a n A2B a n d A3 Recep tor Su btyp es
Yong-Chul Kim, Maarten de Zwart,† Louis Chang, Stefano Moro, J acobien K. von Frijtag Drabbe Ku¨nzel,†
Neli Melman, Ad P. IJ zerman,† and Kenneth A. J acobson*
Molecular Recognition Section, Laboratory of Bioorganic Chemistry, National Institute of Diabetes, Digestive and Kidney
Diseases, National Institutes of Health, Bethesda, Maryland 20892-0810, and Division of Medicinal Chemistry, Leiden/
Amsterdam Center for Drug Research, P.O. Box 9502, 2300 RA Leiden, The Netherlands
Received February 11, 1998
The adenosine antagonist 9-chloro-2-(2-furanyl)[1,2,4]triazolo[1,5-c]quinazolin-5-amine (CGS
15943) binds nonselectively to human A1, A2A, and A3 receptors with high affinity. Acylated
derivatives and one alkyl derivative of the 5-amino group and other modifications were prepared
in an effort to enhance A2B or A3 subtype potency. In general, distal modifications of the N5-
substituent were highly modulatory to potency and selectivity at adenosine receptors, as
determined in radioligand binding assays at rat brain A1 and A2A receptors and at recombinant
human A3 receptors. In Chinese hamster ovary cells stably transfected with human A2B receptor
cDNA, inhibition of agonist-induced cyclic AMP production was measured. An N5-(2-
iodophenyl)acetyl derivative was highly selective for A2A receptors. An (R)-N5-R-methyl-
(phenylacetyl) derivative was the most potent derivative at A3 receptors, with a Ki value of
0.36 nM. A bulky N5-diphenylacetyl derivative, 13, displayed a Ki value of 0.59 nM at human
A3 receptors and was moderately selective for that subtype. Thus, a large, nondiscriminating
hydrophobic region occurs in the A3 receptor in proximity to the N5-substituent. A series of
straight-chain N5-aminoalkylacyl derivatives demonstrated that for A2B receptors the optimal
chain length occurs with three methylene groups, i.e., the N5-γ-aminobutyryl derivative 27
which had a pA2 value of 8.0 but was not selective for A2B receptors. At A1, A2A, and A3 receptors
however the optimum occurs with four methylene groups. An N5-pivaloyl derivative, which
was less potent than 27 at A1, A2A, and A3 receptors, retained moderate potency at A2B receptors.
A molecular model of the 27-A2B receptor complex based on the structure of rhodopsin utilizing
a “cross-docking” procedure was developed in order to visualize the environment of the ligand
binding site.
In tr od u ction
pyridines17-19 (structurally similar to Ca2+ channel
antagonists), and triazoloquinazolines,20 such as 1a
(9-chloro-2-(2-furyl)[1,2,4]triazolo[1,5-c]quinazolin-5-
amine, CGS 15943; Chart 1). Chemical optimization of
the leads has resulted in antagonists having subnano-
molar affinity, for example, 5-(phenylacetylamino)-9-
chloro-2-(2-furyl)[1,2,4]triazolo[1,5-c]quinazoline (MRS
1220, 1b), which had a Ki value in binding of 0.65 nM
at human A3 receptors. 1,4-Dihydropyridine derivatives
2 in some cases displayed >30000-fold selectivity for
human A3 receptors (e.g., MRS 1334, 2b). Antagonist
radioligands for the A3 receptor are still lacking.
The adenosine A2B receptor21 is the poorest character-
ized subtype of adenosine receptors. Whereas for the
other subtypes selective agonists, antagonists, and
radiolabeled ligands are available, no such compounds
are known for the A2B receptor. For pharmacological
studies, both selective agonists and antagonists are
needed. An A2B selective antagonist may prove useful
in the treatment of asthma.43
Of the four known subtypes of adenosine receptors,
A1, A2A, A2B, and A3 receptors, ligand development at
the first two subtypes has been well-explored for both
selective agonists and antagonists.1-3 Radioligands
selective for A1 receptors have been known since the
early 1980s,4 and radioligands selective for A2A receptors
were first introduced in the late 1980s.5 These tools
have aided greatly in the identification of numerous
ligands for the A1 and A2A adenosine receptor subtypes.
Such selective agonists and antagonists are potentially
useful in the treatment of disorders of the CNS6-8 and
the renal system9 and also cardiovascular disorders.10
Several classes of agonists and antagonists selective
for the most recently cloned adenosine receptor, the A3
subtype, have also been reported.11 These agents have
been used pharmacologically to demonstrate both cere-
broprotective12 (chronic agonist or acute antagonist) and
cardioprotective13 (agonist) effects. A3 receptor agonists
at high concentration may also induce apoptosis.14 Four
diverse chemical classes have been identified as promis-
ing leads for A3 antagonists: flavonoids15 (phenolics
found in all vascular plants), pyridines,16 dihydro-
As an approach to find both selective antagonists for
the A2B receptor and more potent antagonists for the
A3 receptor, we have synthesized and screened a series
of triazoloquinazolines derived from the nonselective A1/
A2A receptor antagonist 1a . In this study we have
further explored the pharmacophore region in the
vicinity of the N5 position of 1a and the acyl group of
1b in binding to the human A3 receptor and have
* Address correspondence to: Dr. K. A. J acobson, Bldg 8A, Rm B1A-
19, NIH, NIDDK, LBC, Bethesda, MD 20892-0810. Tel: (301) 496-
9024. Fax: (301) 480-8422. E-mail: kajacobs@helix.nih.gov.
† Leiden/Amsterdam Center for Drug Research.
S0022-2623(98)00094-6 CCC: $15.00 © 1998 American Chemical Society
Published on Web 06/18/1998

2836 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 15
Kim et al.
Ch a r t 1. Structures of High-Affinity A3 Receptor Antagonists (Ki Values in µM).
Ta ble 1. Affinities or Antagonistic Activities of Triazoloquinazoline Derivatives in Radioligand Binding Assays at A1, A2A, and A3
Receptorsa-c
Ki or IC50 (nM)
compd
R
rA1
rA2A
hA3
rA1/hA3
rA2A/hA3
1a (CGS 15943)
1b
3
4
5
6
7
8
9
H
21 ( 3.0d
52.7 ( 11.8e
1200 ( 15
30.2 ( 6.3
217 ( 65
24.9 ( 7.7
2300 ( 590
45.4 ( 8.7
13.8 ( 4.6
43.0 ( 6.4
46.2 ( 11.9
37.5 ( 8.5
129 ( 43
3.3 ( 1.7d
10.3 ( 3.7e
200 ( 34
13.8 ( 2.4
0.65 ( 0.25
42.5 ( 6.91
14.4 ( 3.2
49.3 ( 17.9
3.56 ( 1.24
COCH2-Ph
CH2-Ph
COCH2-(4-CH3O-Ph)
COCH2-(4-NH2-3-I-Ph)
COCH2-(4-NH2-Ph)
COCH2-(2-I-Ph)
COCH2-(3-I-Ph)
COCH2-(4-I-Ph)
COCH2-(3-Cl-Ph)
(R)-COCH(CH3)(Ph)
(S)-COCH(CH3)(Ph)
COCH(Ph)2
81
28
16
4.7
28.0 ( 4.8
10.4 ( 2.2
6.97 ( 1.13
17.2 ( 4.0
9.67 ( 1.74
9.93 ( 2.00
9.43 ( 2.65
11.1 ( 1.3
20.1 ( 3.7
12.1 ( 3.4
453 ( 156
28.3 ( 10.3
59.8 ( 13.4
18.8 ( 5.3
6.65 ( 1.56
7.58 ( 0.80
19.9 ( 4.6
3.64 ( 0.34
33.7 ( 7.3
11.1 ( 1.9
29.7 ( 9.1
143 ( 13
7.58 ( 2.1
1.38 ( 0.23
1.13 ( 0.41
10.3 ( 3.7
55.9 ( 5.5
16.7 ( 2.5
28.8 ( 10.7
7.43 ( 2.72
1.15 ( 0.47
2.1
1.9
4.4
7.0
0.21
2.0
>10000
<1
,1
882 ( 242
62.9 ( 13.0
32.1 ( 11.3
0.362 ( 0.053
0.468 ( 0.111
0.586 ( 0.196
194 ( 42
23.6 ( 7.6
72.1 ( 15.6
46.3 ( 5.4
82.9 ( 2.7
6.71 ( 0.67
32.9
22.0 ( 3.1
33.8 ( 10.2
53.7 ( 31.0
1140 ( 370
1200 ( 460
874 ( 4
0.051
0.22
1.3
0.011
0.16
0.29
31
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
130
80
43
220
4.6
21
2.3
COC(CH3)(Ph)2
COCH2CH2-Ph
890 ( 131
45.2 ( 7.5
282 ( 71
1.9
1.6
COCHdCH-Ph (trans)
D-COCH(CH3)(NH-Boc)
L-COCH(CH3)(NH-Boc)
CO(CH2)2-NH-Boc
CO(CH2)3-NH-Boc
CO(CH2)4-NH-Boc
CO(CH2)5-NH-Boc
CO(CH2)6-NH-Boc
D-COCH(CH3)(NH2)
L-COCH(CH3)(NH2)
CO(CH2)2-NH2
CO(CH2)3-NH2
CO(CH2)4-NH2
CO(CH2)5-NH2
CO(CH2)6-NH2
CO(CH2)4-COOBn
CO(CH2)2-COOCH3
CO(CH2)6-COOCH3
CO(CH2)3-COOH
3.9
1.0
0.66
4.6
0.71
0.41
0.080
1.1
48.6 ( 6.7
54.7 ( 13.4
31.0 ( 2.2
45.9 ( 13.2
30.1 ( 6.8
53.8 ( 12.7
38.2 ( 2.6
193 ( 17
390 ( 127
89.8 ( 23.3
8.75 ( 2.28
6.99 ( 1.38
99.6 ( 6.7
114 ( 26
1.4
1.4
1.6
0.60
0.17
1.0
0.71
0.17
0.33
0.10
0.11
0.12
0.47
0.33
6.8
0.21
0.026
0.12
0.0088
0.017
0.020
0.048
0.16
0.37
0.52
0.13
0.014
80.8 ( 7.4
57.9 ( 20.8
213 ( 27
346 ( 77
304 ( 126
71.8 ( 7.2
46.6 ( 10.5
45 ( 4
44.7 ( 14.1
55.1 ( 8.6
59.0 ( 18.1
81.3 ( 11.0
1.3
0.79
0.55
a
b
Displacement of specific [3H]R-PIA binding in rat brain membranes, expressed as Ki ( SEM (n ) 3-5). Displacement of specific
[3H]CGS 21680 binding in rat striatal membranes, expressed as Ki ( SEM (n ) 3-6). c Displacement of specific [125I]AB-MECA binding
at human A3 receptors expressed in HEK cells, in membranes, expressed as Ki ( SEM (n ) 3-4). IC50 values for 1a (see ref 20). e Previously
d
reported Ki values were 305 and 52.0 nM for rat A1 and A2A receptors, respectively.20
defined the previously unknown structure-activity
assays at adenosine receptors are shown in Table 1. A
few derivatives, e.g., the phenylacetyl derivative 1b and
the 5-aminoacyl derivatives 20 and 27 in Table 1 and
compounds 35-46 in Table 3, had been reported by us
previously.20 The 5-benzyl derivative 3 was prepared
through an alkylation reaction using potassium hydrox-
relationship (SAR) at the human A2B receptor.
Resu lts a n d Discu ssion
Syn th esis. The structures of the triazoloquinazoline
derivatives tested for affinity in radioligand binding

Derivatives of CGS 15943
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 15 2837
Ta ble 2. Yields and Chemical Characterization of Triazoloquinazoline Derivatives
compd no.
% yield
mp (°C)
MS
formula
anal.
3
4
5
6
7
29
12
25
45
22
16
51
33
38
58
57
100
65
43
13
15
50
53
100
100
56
71
72
64
80
70
71
54
62
26
184-185
258-260
210-211
197-199
280-282
224-227
260-262 dec
248-250
207-209 dec
115-118
205-207
175-178
253-255
305-307
131-134
220
216-218
189-192
196-198
156-157
175-176
278
137-139
231-232
153-155
165-167
149-151
185-187
165
EI: 375
EI: 433
CI: 545
EI: 418
EI: 529
CI: 530
EI: 529
CI: 438
EI: 417
EI: 417
CI: 480
EI: 493
EI: 417
CI: 416
CI: 457
CI: 457
CI: 457
CI: 485
CI: 499
EI: 512
FAB: 357
FAB: 357
EI: 356
FAB: 385
CI: 399
EI: 412
EI: 503
EI: 399
EI: 455
FAB: 400
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
C
20H14N5O2Cl‚0.81H2O
22H16N5O3Cl‚0.53CH2Cl2
21H14N6O2ClI
21H15N6O2Cl‚0.64H2O
21H13N5O2ClI‚0.94(CH3)2CO
21H13N5O2ClI‚1.17(CH3)2CO
21H13N5O2ClI
21H13N5O2Cl2
22H16N5O2Cl
22H16N5O2Cl
27H18N5O2Cl
28H20N5O2Cl
22H17N5O2Cl
22H15N5O2Cl
21H21N6O4Cl
21H21N6O4Cl‚2.6CH3OH
21H21N6O4Cl
23H25N6O4Cl
24H27N6O4Cl
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
C,H,N
FAB+a
FAB+a
C,H,N
FAB+a
C,H,N
FAB+a
C,H,N
C,H,N
C,H,N
C,H,N
8
9
10
11
12
13
14
15
16
17
18
19
21
22
23
24
25
26
28
29
30
31
32
33
34
25H29N6O4Cl‚0.91H2O
16H13N6O2Cl
C16H13N6O2Cl
C16H13N6O2Cl‚CF3COOH
C
18H17N6O2Cl
C19H19N6O2Cl‚CF3COOH‚0.87CH3OH
20H21N6O2Cl
C26H22N5O4Cl‚CF3COOH
C
C
C
C
18H14N5O4Cl‚0.3EtOAc
22H22N5O4Cl‚0.37(C2H5)2O
18H14N5O4Cl‚0.56CH2Cl2
209-211
a
High-resolution mass in FAB+ mode (m/z) determined to be within acceptable limits. 24: calcd, 357.0867; found, 357.0879. 25: calcd,
357.0867; found, 357.0865. 28: calcd, 385.1180; found, 385.1168. 30: calcd, 413.1491; found, 413.1493.
ide and benzyl bromide. The 4-amino-3-iodophenyl-
acetyl derivative 5 was synthesized by iodination22
(using iodine and calcium carbonate) of the 4-amino-
phenylacetyl derivative 6 which was prepared from
catalytic reduction of the corresponding 4-nitrophenyl-
acetyl derivative. The other various acyl derivatives
were synthesized using standard acylation methods
described in the Experimental Section, except for the
4-carboxybutanoyl congener 34 which was prepared by
heating with glutaric anhydride at 80 °C without base
or coupling agents. The R- and S-enantiomeric pairs
(11 and 14; 17 and 18) were prepared from the corre-
sponding optically pure carboxylic acids. The yields and
chemical characterization of these compounds are re-
ported in Table 2.
been satisfactory for a radioligand; however upon iodi-
nation to give compound 5, the affinity at human A3
receptors decreased by 14-fold. Thus, the receptor
affinity is highly dependent on the substitution of the
phenyl ring of 1b. The affinities of ring-iodinated
isomers of 1b, 7-9, were compared. The N5-(2-iodophe-
nyl)acetyl derivative 7 was highly selective for A2A
receptors. The N5-(3-chlorophenyl)acetyl and N5-(3-
iodophenyl)acetyl derivatives 10 and 8, respectively,
differed in affinity only at the A3 receptor, such that 8
was 27-fold less potent in binding.
Substitution of the two prochiral methylene hydro-
gens of the acetyl group was carried out. The R-methyl
substitutions, resulting in the R- and S-enantiomers 11
and 12, appeared to slightly enhance affinity in both
cases. No significant stereoselectivity of binding to A1,
Affin ity a t Ad en osin e Recep tor s. Bin d in g As-
sa ys a t A1, A2A, a n d A3 Recep tor s. Ki values were
determined in radioligand binding assays at rat cortical
A1 receptors vs [3H]R-PIA ([3H]-(R)-N6-(phenyliso-
propyl)adenosine) and at rat striatal A2A receptors vs
[3H]CGS 21680 ([3H]-2-[[4-(2-carboxyethyl)phenyl]eth-
ylamino]-5′-(N-ethylcarbamoyl)adenosine).4,5 Affinity at
cloned human A3 receptors expressed in HEK-293 cells23
was determined using [125I]AB-MECA (N6-(4-amino-3-
A2A, and A3 receptors was evident at this chiral center.
The most potent derivative at A3 receptors in the
present study was the (R)-N5-R-methyl(phenylacetyl)
derivative 11 with a Ki value of 0.36 nM. Curiously,
even a phenyl substitution at the methylene group,
compound 13, resulted in extremely high affinity at
human A3 receptors, with a Ki value of 0.59 nM and
selectivities of 220- and 21-fold vs rat A1 and A2A
receptors, respectively. Thus, the binding region of the
receptor surrounding the R-carbon of the acetyl group
of 1b is not highly sterically restricted. Nevertheless,
retention of at least one methylene hydrogen of the
acetyl group was required for high affinity at the
adenosine receptors, since the R-methyldiphenyl deriva-
tive 14 was 330-fold less potent than the R-diphenyl
derivative 13 in binding to human A3 receptors. Ho-
mologation of 1b to give the phenylpropionyl derivative
15 reduced the affinity at human A3 receptors by 27-
[
125I]iodobenzyl)-5′-(N-methylcarbamoyl)adenosine).24
N5-Alkylation in the N-benzyl derivative 3 decreased
binding affinity at A1, A2A, and A3 receptors. Simple
phenyl ring substitutions were made on the structure
of the N5-phenylacetyl derivative 1b including: meth-
oxy, 4; amino, 5 and 6; and halo, 7-10. None of these
derivatives exceeded the affinity of 1b at human A3
receptors. The most potent of these derivatives was the
4-amino derivative 6 which was prepared as a potential
substrate for radioiodination. The Ki value of 6 at
human A3 receptors was 3.56 nM, which would have

2838 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 15
Kim et al.
Ta ble 3. Potency of Triazoloquinazoline Derivatives in a Functional Assay at Human A2B Receptors Stably Expressed in CHO Cells
d
compd
R1 or R3
% inhibitiona,c
IC50 (µM)b,d
1.20 ( 0.43
pA2
1a
1b
3
5
6
H
100 (100-100)
71 (73-69)
21 (20-22)
23 (18-27)
90 (85-94)
42 (39-45)
45 (55-34)
43 (50-36)
38 ( 14
47 (57-36)
83 (85-80)
100 (100-100)
100 (100-100)
79 (73-85)
75 (74-75)
8.0 ( 0.3
COCH2-Ph
CH2-Ph
COCH2(4-NH2-3-I-Ph)
COCH2(4-NH2-Ph)
COCH2-(4-I-Ph)
COCH2-(3-Cl-Ph)
COCH2CH2-Ph
COCHdCH-Ph
CO(CH2)3-NH-Boc
CO(CH2)2-NH2
CO(CH2)3-NH2
CO(CH2)4-NH2
CO(CH2)5-NH2
CO(CH2)6-NH2
CO(CH2)3-COOH
COCH3
7.4 ( 4.5
9
10
15
16
20
26
27
28
29
30
34
35
36
37
38
39
40
41
42
43
44
45
46
13.3 ( 4.0
0.27 ( 0.04
1.77 ( 1.25
8.0 ( 0.3
1.90 ( 0.13
2.65 ( 1.00
92 (94-90)
66 (75-56)
66 (73-58)
79 (86-72)
89 (92-85)
92 (94-90)
58 (69-47)
38 (51-25)
16 (24-9)
0 (0-0)
COCH2CH3
CO(CH2)2CH3
CO(CH2)3CH3
COC(CH3)3
CO-OC(CH3)3
CO-Ph
CO-(3-I-Ph)
H, R2 ) Br
CO-Ph, R2 ) Br
H
(CH2)2CH3
2.40 ( 1.70
2.41 ( 1.02
0 (0-0)
0 (0-0)
a
b
Percentage inhibition by 50 µM of each antagonist of cyclic AMP production induced by 50 µM NECA. Concentration of antagonist
inhibiting 50% of cyclic AMP production induced by 50 µM NECA. c Values are the means of duplicate measurements (values of individual
d
measurements between parentheses) or are the means of at least three experiments ( SEM. Values are given ( SEM and are the
means of at least three experiments.
fold, and the corresponding styryl derivative 16 was
even less potent at A3 receptors.
e.g., 20-22, appeared to be preferred. This suggested
that human A3 receptors contain a more highly hydro-
phobic pocket in this region than do rat A2A receptors.
Furthermore, at A1 receptors the free amine is preferred
over the Boc-amine for analogues containing three and
four methylene groups.
The effects of ester, 31-33, and carboxylic acid, 34,
groups on the N5-acyl substituent were also explored.
The affinity at A3 receptors was not significantly af-
fected by the presence of a negatively charged carboxy-
late group vs a neutral ester group.
F u n ction a l Assa y a t A2B Recep tor s. Compound
1a and selected N5-acyl-substituted analogues (Table 3)
were screened by measuring the percentage inhibition
by the analogue (50 µM) of the cyclic AMP production
induced by 50 µM 5′-(N-ethylcarboxamido)adenosine
(NECA) in a Chinese hamster ovary cell line stably
transfected with human A2B receptor cDNA. For de-
rivatives that displayed g80% inhibition, IC50 values
were determined on cyclic AMP production induced by
50 µM NECA. Finally, pA2 values were determined for
two of the most active derivatives, 1a and 27. For this
purpose, NECA concentration-response curves were
recorded in the absence and presence of three increasing
concentrations of antagonist. pA2 values were derived
from Schild plots (Figure 1), which displayed slopes of
approximately unity.
The γ-aminobutyric acid derivatives, both the Boc-
protected form 20 and the free amino form 27, were
reported in the previous study.20 The SAR of homo-
logues of these derivatives was explored. Although
nanomolar affinities at human A3 receptors were not
obtained, there was a distinct dependence of affinity
upon lengthening or shortening the methylene chain.
A minimum in the Ki values at A1, A2A, and A3 receptors
among the free amine derivatives was obtained at a
chain length of four methylene groups, corresponding
to compound 28 (Figure 1), which had a Ki value of 1.13
nM at A2A receptors. For free amine derivatives, the
order of affinity was invariably: A2A > A1 > A3 recep-
tors. Thus, the interactions responsible for this affinity
enhancement must be common to these three receptor
subtypes. For Boc-amine derivatives, however, a clear
dependence of binding affinity on chain length was not
evident at A1, A2A, and A3 adenosine receptor subtypes.
In this series of amino acid derivatives, no stereoselec-
tivity of binding to A1, A2A, and A3 receptors was evident
at the R-carbon prochiral center, as seen from the
affinities of the D- and L-Ala conjugates, 17 vs 18 and
24 vs 25. At A2A receptors the presence of the free
amino group, e.g., 27-29, in general resulted in higher
affinity, while at A3 receptors the Boc-protected form,

Derivatives of CGS 15943
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 15 2839
tives having three and four methylene groups were
nearly equipotent. At A2B receptors the Boc-protected
γ-aminobutyryl derivative 20 had a much lower activity
than the corresponding free amino derivative 27 (only
47% inhibition at 50 µM).
The γ-carboxybutyryl derivative 34 showed a high
potency (1.9 µM), but it was less active than the
γ-aminobutyryl derivative 27. Apparently, a positively
charged amino group results in a higher activity than
a negatively charged carboxylate function.
In the N5-phenylacetyl-substituted series only the
4-amino derivative 6 displayed a moderate potency (7.4
µM). A phenylacetyl group appears to be not well-
tolerated in the A2B receptor. Increase of potency by
substitution with an amino function was also observed
in propionyl, 36, vs 3-aminopropionyl, 26, and butyryl,
37, vs 4-aminobutyryl, 27, not only at the A2B receptor
(Table 3) but also at A1 and A2A receptors (refer to
binding data of 36 and 37 in ref 20).
In the series of compounds with N5-acyl chain sub-
stituents, the N5-acetyl (35) and N5-pivaloyl (39) deriva-
tives were most active. Apparently, there is room in
the A2B receptor for a bulky tert-butyl group. The N5-
Boc-substituted derivative 40, with a tert-butyl group
too, showed a similar activity. Remarkably, for maxi-
mal activity, the N5-acetyl group should either be
unsubstituted as in 35 or methyl-trisubstituted as in
39, but not methyl-monosubstituted as in the propionyl
derivative 36.
F igu r e 1. Determination of the pA2 value of compound 27 at
human A2B receptors. Data were taken from a typical experi-
ment. Dose-response curves of NECA were recorded in the
absence and presence of three concentrations of the antagonist
27: 9, no antagonist added; 2, 0.03 µM; b, 0.1 µM; ∆, 0.3 µM.
pA2 values were determined in a Schild plot (inset) with a slope
of approximately unity.
Although the N5-pivaloyl derivative 39 had a slightly
decreased activity (IC50 value 2.4 µM), it showed largely
diminished affinities for A1, A2A, and A3 receptors in the
previous study (Ki A1 ) 205 ( 20 nM; Ki A2A ) 88.8 (
20.5 nM; Ki A3 ) 244 ( 6 nM).20 The N5-Boc-substituted
derivative 40 with an IC50 value of 2.41 ( 1.02 µM
behaved similarly (Ki A1 ) 190 ( 16 nM; Ki A2A ) 92 (
8 nM; Ki A3 ) 82.5 ( 23.3 nM).20 Thus, N5-pivaloyl or
N5-Boc substitution might lead to A2B receptor selectiv-
ity.
Molecu la r Mod elin g. A rhodopsin-based molecular
model for the human A2A receptor has been published
previously.25 As an aid in interpreting the present
results at the human A2B receptor, we provide the first
description of ligand-A2B receptor interactions, using
27 as a ligand to probe the A2B receptor binding pocket.
We have recently introduced the cross-docking proce-
dure26 (see the Experimental Section) which can be
considered a further refinement in building the putative
antagonist binding site in the human A2B receptor. As
with other G protein-coupled receptor models,27 the
length of the transmembrane region is approximately
40 Å. The interhelical distance between pairs of adja-
cent helical axes is roughly 10 Å, consistent with a
common interhelical contact distance.28 The interhelical
angles, measured between the principal axes of adjacent
helices, are around -150-170° for antiparallel and 10-
25° for the parallel helices. This is typical of a 3-4 type
helix-helix contact associated with optimal interactions
between nearly parallel-aligned helices.28 Each helix
maintained almost the same position and tilting found
in the published rhodopsin 2D electron density map.29,30
F igu r e 2. Minimized energy conformations of 27, showing
two conformers (cis and trans) of the aminoalkyl side chain.
Compound 1a had an IC50 value of 1.20 µM. In the
same system a pA2 value of 8.0 ( 0.3 was measured.
This agrees with earlier findings of Alexander et al., who
have measured a value of 7.8 in a similar assay, using
another CHO cell line expressing human A2B recep-
tors.42 The γ-aminobutyryl-substituted derivative 27
displayed an elevated potency at the A2B receptor (IC50
value of 0.27 µM). However, its pA2 value was similar
to that of 1a . Concentration-response curves of NECA
in the presence and absence of 27 are shown in Figure
2. There was no gain in selectivity toward the other
receptor subtypes.
Derivatives with a 2-(5-bromofuryl) substituent (43,
44) and derivatives that lack the 5-amino function (45,
46) were weak or inactive as A2B receptor antagonists.
Diminished affinities of these analogues for the other
subtypes have been reported previously.20
Among the amino acid conjugates a minimum in the
IC50 values in A2B receptor antagonism was observed
with a chain length of three methylene groups (27),
while at A1, A2A, and A3 receptors aminoalkyl deriva-
In all our calculations, we used the γ-amino-proto-
nated form of compound 27, consistent with protonation
at physiological pH. Since 27 has high conformational

2840 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 15
Kim et al.
flexibility in the region of the γ-aminobutyryl side chain,
a complete random search conformational analysis was
performed. After sampling and minimization proce-
dures, two energetically important conformers have
been selected, one of these with the amide bond in the
cis conformation and the other one in which the amide
bond was in the trans conformation (see Figure 2).
RHF/AM1 semiempirical calculations were conducted
to obtain the optimized geometries. Only the cis con-
former may be docked in an energetically favorable
manner inside the TM bundle. The trans isomer would
force the aminobutyl side chain toward the transmem-
brane helices leading to an overall distortion of the
receptor architecture.
Moreover, in the cross-docked model TM3, TM4, TM5,
TM6, and TM7 were rotated clockwise by 15°, 10°, 5°,
10°, and 5°, respectively, about their transmembrane
axes with respect to the ligand-free receptor model. The
energy of the 27-A2B receptor complex structure was
lowered 55 kcal/mol through the cross-docking proce-
dure. Consistent with the proposal of Gouldson et al.,31
rotations and translations of the TM domains are crucial
factors in the ligand recognition process in different
GPCRs. Consequently, our approach to docking is
designed to mimic the natural domain movement within
a receptor.
One of the most important limitations in the correct
identification of the ligand binding regions of the A2B
receptor is that no site-directed mutagenesis studies are
available. Using both the A2A/A2B sequence homology
and the site-directed mutagenesis results previously
obtained in our laboratory for the human A2A receptor,25
we were able to identify a hypothetical binding site for
27 in the human A2B receptor. Figure 3 shows the 3D
structural models of the human A2B receptor after the
application of cross-docking with 27.
We identified the hypothetical binding site of 27 in
proximity to TMs 3, 5, 6, and 7, with the furan ring
pointing toward the extracellular environment. The
γ-aminobutyryl side chain is located midway between
His251 (TM6) and His280 (TM7). According to the
previously reported site-directed mutagenesis results for
A1 and A2A receptors,32,25 these two histidine residues
(conserved in the same positions of TM6 and TM7
helices for both receptors) were found crucial for the
binding of both agonists and antagonists.
the affinity values upon lengthening or shortening of
the methylene chain. A minimum in the Ki values at
A1, A2A, A2B, and A3 among the free amine derivatives
was obtained at chain lengths of three or four methylene
groups, indicating that there is a common anchor point
inside the receptor binding cavity. We speculate that
Asn254 (TM6) corresponds to this anchor point in the
human A2B receptor.
A hydrophobic region, delimited by three apolar
amino acids such as Val85 (TM3), Phe187 (TM5), and
Val191 (TM5), is also present in the binding site model.
The corresponding amino acids Leu90 (TM3) and Phe182
(TM5) were essential for binding of both agonists and
antagonists at human A2A receptors.25 The chlorophen-
yl moiety of the triazoloquinazoline structure of 27 is
located within this region. No direct interactions are
predicted between the two polar amino acids, Thr89
(TM3) and Ser92 (TM3), and the antagonist structure.
As previously reported, these two amino acids are
important only for the coordination of the agonist
derivatives at A1 and A2A receptors, respectively.25
Another unresolved question is why there are great
differences in ligand affinities between A2A and A2B
receptors. A comparison of the nature of the critical
amino acids in the TM regions involved in the recogni-
tion of the ligand at the A2A and A2B subtypes fails to
provide a clear explanation. In fact, there is a very good
correspondence between the important residues found
using site-directed mutagenesis of the human A2A
receptor and those present in the A2B receptor. We
speculate that there are two important differences
between the molecular structures of A2A and A2B recep-
tors: (i) the replacement of two amino acids, Leu267
(hA2A) with Lys269 (hA2B) and Tyr270 (hA2A) with
Asn273 (hA2B), at the top of TM7 that could be involved
in the movement of the ligand into the binding site and
(ii) the different primary and secondary structures of
the second extracellular loops that seem to be involved
in the extracellular recognition process of the ligand.33
The second extracellular loop has been shown to be
important for both agonist and antagonist binding at
the human A2A receptor and could represent one of the
strategic keys of the different selectivities between the
different receptor subtypes.
Exp er im en ta l Section
Com p u ta tion a l Meth od s. The human A2B receptor model
was built and optimized using SYBYL 6.334 and MacroModel
5.035 modeling packages, respectively, based on the approaches
described by van Rhee et al.36 and Moro et al.26 All calculations
were performed on a Silicon Graphics Indigo2 R8000 worksta-
tion. Briefly, transmembrane domains were identified with
The oxygen atom of Ser283 (TM7) is within hydrogen-
bonding distance (2.6 Å) of the carbonyl group of the
amide moiety at the 5-position of the triazoloquinazoline
structure. The amide group is surrounded by four polar
amino acids: His251 (TM6), Ser279 (TM7), His280, and
Ser283 (TM7). This region seems to be very important
for the recognition of the antagonist structures. In
particular, our side-directed mutagenesis results for
human A2A receptor suggested that this serine (Ser281)
was crucial for binding of both agonists and antago-
nists.25
Another important interaction (1.9 Å) is possible
between the amino group of Asn254 (TM6) and the
protonated nitrogen of the γ-aminobutyryl side chain.
Also this Asn, conserved among all adenosine receptor
subtypes, was found important for ligand binding. This
interaction can explain the common trend observed
between the different adenosine receptor subtypes, of
the aid of Kyte-Doolittle hydrophobicity37 and Emini surface
37
probability parameters. Transmembrane helices were built
from the sequences and minimized individually. The mini-
mized helices were then grouped together to form a helical
bundle matching the overall characteristics of the electron
density map of rhodopsin. The helical bundle was minimized
using the Amber38 all-atom force field, until the rms value of
the conjugate gradient (CG) was <0.1 kcal/mol/Å. A fixed
dielectric constant ) 4.0 was used throughout these calcula-
tions.
A model of 5-[(6-aminohexanoyl)amino]-9-chloro-2-(2-furanyl)-
[1,2,4]triazolo[1,5-c]quinazoline (27) was constructed using the
Sketch Molecule module of SYBYL. Conformational analysis
of 27 was performed on all rotatable bonds of the aminohex-
anoyl side chain, using the random search procedure of

Derivatives of CGS 15943
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 15 2841
F igu r e 3. Model of the cis conformation of 27 docked in the putative binding site of the human A2B receptor, as viewed from the
plane of the plasma membrane (top) and from the extracellular side (bottom).
SYBYL. The optimized geometries of the resulting conformers
were calculated using MOPAC software (RHF method and
AM1 Hamiltonian, keywords: PREC, GNORM)0.1, EF).
to an additional CG minimization run of 300 steps. Partial
atomic charges for the ligands were imported from the MOPAC
output files.
The fully minimized conformers were rigidly docked into the
helical bundle using graphical manipulation coupled to con-
tinuous energy monitoring (Dock module of SYBYL). When a
final position was reached, consistent with a local energy
minimum, the complexes of receptor and ligand were subjected
Recently, we have introduced a “cross-docking” procedure
to obtain energetically refined structures of ligand-P2Y1
receptor complexes.26 We applied the same technique to obtain
the structure of the 27-A2B receptor complex. This is a way
to explore possible ligand-induced rearrangements of 7TM

2842 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 15
Kim et al.
bundle by sampling 7TM conformations in the presence of the
docked ligands. Cross-docking was carried out using the Dock
module of SYBYL. Each helix was separated from the ligand-
receptor complex structure, and its relative position was
changed until a new lower-energy geometry was obtained.
These adjustments consisted of small translations and rota-
tions of the principal axis of the helix with respect to its
original position. When a new final position was reached,
consistent with the lowest-local-energy minimum, the sepa-
rated helix was merged again into the ligand-receptor com-
plex. The hydropathy profile for the new oriented helix was
checked using the Kyte-Doolittle method.37 The new complex
was subjected to an additional CG minimization run of 300
steps. This procedure was repeated for TM3, TM4, TM5, TM6,
and TM7. The manual adjustments were followed by 25 ps of
molecular dynamics (MD module of MacroModel) performed
at a constant temperature of 300 K using a time step of 0.001
ps and a dielectric constant ) 4.0. This procedure was followed
by another sequence of CG energy minimization to a gradient
threshold of < 0.1 kcal/mol/Å. Energy minimization of the
complex was performed using the AMBER all-atom (Macro-
Model) force field.
of 0.02 mmol of Boc-protected amino acyl derivatives of 1a in
1 mL of CH2Cl2 was added 50 µL of TFA, and the mixture
stirred at room temperature for 10 min. The mixture was
evaporated under reduced pressure, and white solids were
obtained from ether or CH2Cl2/MeOH as TFA salts (24-30).
9-Ch lor o-2-(2-fu r a n yl)-5-(ben zyla m in o)[1,2,4]tr ia zolo-
[1,5-c]qu in a zolin e (3). To a solution of 1a (24.0 mg, 0.08
mmol) in 2 mL of anhydrous DMSO was added powdered
potassium hydroxide (67.2 mg, 1.20 mmol) followed by benzyl
bromide (28.8 mg, 0.17 mmol). The mixture was stirred at
room temperature overnight and then diluted with 2 mL of
water. The reaction product was extracted with CHCl3 (10
mL × 3). The fractions were combined, neutralized with 1 N
HCl, washed with water (25 mL × 2), and dried over sodium
sulfate. The product was isolated from preparative silica gel
TLC (hexanes-ethyl acetate, 4:1) (29%): 1H NMR (CDCl3) δ
4.92 (2H, d, J ) 5.86 Hz, -CH2), 6.55 (1H, t, J ) 5.86 Hz,
NH), 6.61 (1H, t, J ) 1.95 Hz, H-4′), 7.28 (1H, s, H-3′), 7.36-
7.49 (4H, m, aromatic-H), 7.61-7.64 (2H, m, H-8 and H-5′),
7.70 (1H, d, J ) 9.77 Hz, H-7), 8.41 (1H, d, J ) 1.95 Hz, H-10).
9-Ch lor o-2-(2-fu r a n yl)-5-[(4-m e t h oxyp h e n yla ce t yl)-
a m in o][1,2,4]tr ia zolo[1,5-c]qu in a zolin e (4): 1H NMR (DM-
SO-d6) δ 3.74 (3H, s, -OCH3), 3.98 (2H, s, -CH2CO), 6.78-
6.80 (1H, m, H-4′), 6.93 (2H, d, J ) 8.7 Hz, aromatic-H), 7.32-
7.35 (3H, m, H-3′ and aromatic-H), 7.92 (2H, s, H-8 and H-7),
8.03 (1H, s, H-5′), 8.39 (1H, s, H-10), 11.24 (1H, s, NH).
9-Ch lor o-2-(2-fu r an yl)-5-[(4-am in o-3-iodoph en ylacetyl)-
a m in o][1,2,4]tr ia zolo[1,5-c]qu in a zolin e (5). To a solution
of 6 (10 mg, 0.024 mmol) in 1 mL of CH2Cl2/MeOH was added
solid I2 (6 mg, 0.024 mmol) followed by solid CaCO3 (3 mg,
0.028 mmol). The reaction mixture was stirred for 12 h at
Ma ter ia ls. Compound 1a , R-PIA, and 2-chloroadenosine
were purchased from Research Biochemicals International
(Natick, MA). All acylating agents were obtained from Aldrich
(St. Louis, MO).
Syn th esis. Proton nuclear magnetic resonance spectros-
copy was performed on a Varian GEMINI-300 spectrometer,
and spectra were taken in DMSO-d6 or CDCl3. Unless noted,
chemical shifts are expressed as ppm downfield from tetra-
methylsilane. Chemical-ionization (CI) mass spectrometry
was performed with a Finnigan 4600 mass spectrometer and
electron-impact (EI) mass spectrometry with a VG7070F mass
spectrometer at 6 kV. High-resolution FAB (fast atom bom-
bardment) mass spectrometry was performed with a J EOL
SX102 spectrometer using 6-kV Xe atoms. The compounds
were previously desorbed from glycerol or magic bullet matrix.
Optical rotation was measured using a Perkin-Elmer pola-
rimeter 341. Elemental analysis ((0.4% acceptable) was
performed by Atlantic Microlab Inc. (Norcross, GA) or Gal-
braith Laboratories, Inc. (Knoxville, TN). All melting points
were determined with a Unimelt capillary melting point
apparatus (Arthur H. Thomas Co., PA) and were uncorrected.
All triazoloquinazoline derivatives showed one spot on TLC
(MK6F silica, 0.25 mm, glass-backed; Whatman Inc., Clifton,
NJ ). Where needed, evaluation of purity was done on a
Hewlett-Packard 1090 HPLC system using an OD-5-60 C18
analytical column (150 mm × 4.6 mm; Separation Methods
Technologies, Inc., Newark, DE) in two different linear gradi-
ent solvent systems. One solvent system (A) was 0.1 M TEAA
(pH ) 5.0)/CH3CN, 50:50 to 10:90, in 20 min with flow rate 1
mL/min. The other (B) was H2O/MeOH, 40:60 to 10:90, in 20
min with flow rate 1 mL/min. Peaks were detected by UV
absorption using a diode array detector.
room temperature. The solution was evaporated with
a
nitrogen stream and partitioned between NaHSO3/EtOAc.
EtOAc layer was dried over anhydrous Na2SO4 and evaporated
to dryness under reduced pressure. The residue was purified
by preparative silica gel TLC (CHCl3-MeOH, 30:1) to afford
3.3 mg of 5 as a brown solid (25%):
1H NMR (CDCl3) δ 4.13 (NH2, s), 4.22 (2H, s, -CH2CO),
6.63-6.64 (1H, m, H-4′), 6.76 (1H, d, J ) 7.9 Hz, aromatic-H),
7.20 (1H, d, J ) 7.9 Hz, aromatic-H), 7.28-7.29 (2H, m, H-3′
and aromatic-H), 7.69 (1H, s, H-5′), 7.75 (1H, dd, J ) 8.7, 2.9
Hz, H-8), 7.94 (1H, d, J ) 8.7 Hz, H-7), 8.50 (1H, d, J ) 2.9
Hz, H-10), 9.08 (1H, s, NH).
9-Ch lor o-2-(2-fu r an yl)-5-[(4-am in oph en ylacetyl)am in o]-
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (6). A mixture of PtO2 (5
mg) and 9-chloro-2-(2-furanyl)-5-[(4-nitrophenylacetyl)amino]-
[1,2,4]triazolo[1,5-c]quinazoline (10 mg, 0.022 mmol), which
was prepared using method A, in 3 mL of CH2Cl2/MeOH, was
stirred under hydrogen atmosphere for 3 h at room temper-
ature. The reaction mixture was filtered through a Celite bed,
and the filtrate was concentrated and purified by preparative
silica gel TLC (Hex:CHCl3:MeOH, 1:1:0.1) to afford 4.1 mg of
6 as a brown solid (45%): 1H NMR (CDCl3) δ 3.74 (NH2, s),
4.16 (2H, s, -CH2CO), 6.61-6.62 (1H, m, H-4′), 6.73 (2H, d, J
) 7.8 Hz, aromatic-H), 7.21 (2H, d, J ) 7.8 Hz, aromatic-H),
7.24 (1H, m, H-3′), 7.66 (1H, s, H-5′), 7.72 (1H, dd, J ) 8.8,
2.9 Hz, H-8), 7.92 (1H, d, J ) 8.8 Hz, H-7), 8.45 (1H, d, J )
2.9 Hz, H-10), 9.11 (1H, bs, NH).
9-Ch lor o-2-(2-fu r a n yl)-5-[(2-iod op h en yla cetyl)a m in o]-
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (7): 1H NMR (CDCl3) δ
4.56 (2H, s, -CH2CO), 6.64-6.65 (1H, m, H-4′), 7.03-7.09 (1H,
m, aromatic-H), 7.29 (1H, m, H-3′), 7.41-7.43 (2H, m, aromatic-
H), 7.69 (1H, s, H-5′), 7.75 (1H, dd, J ) 8.8, 2.9 Hz, H-8), 7.92-
7.96 (2H, m, aromatic-H and H-7), 8.51 (1H, d, J ) 2.9 Hz,
H-10), 9.19 (1H, bs, NH).
Gen er a l P r oced u r e for t h e P r ep a r a t ion of 5-N-Acyl
Der iva tives of 1a . Meth od A (Acid Ch lor id e). To a stirred
solution of 1a (20 mg, 0.07 mmol) and anhydrous pyridine (80
µL, 1.0 mmol) in 2 mL of anhydrous CH2Cl2 was added the
desired acyl chloride (0.21 mmol), prepared from the acid and
excess thionyl chloride (unless commercially available). The
mixture was stirred at room temperature for 24 h and then
evaporated to dryness under reduced pressure. The residue
was purified by preparative silica gel TLC (CH2Cl2-MeOH,
50:1-75:1, or Hex:CHCl3:MeOH, 1:1:0.1) to afford the desired
compounds (4, 7-16, 32, 33).
Meth od B (Ca r bod iim id e). A solution of 1a (20 mg, 0.07
mmol), the desired carboxylic acid compound (0.42 mmol),
EDAC (82 mg, 0.42 mmol), DMAP (4 mg, 0.032 mmol), and
triethylamine (0.146 mL, 1.05 mmol) in 3 mL of anhydrous
DMF/CH2Cl2 (1:1 v/v) was stirred at room temperature for 48
h. The mixture was treated with the same procedure as
method A for purification of the desired compounds (17-23,
31).
9-Ch lor o-2-(2-fu r a n yl)-5-[(3-iod op h en yla cetyl)a m in o]-
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (8): 1H NMR (CDCl3) δ
4.38 (2H, s, -CH2CO), 6.63-6.65 (1H, m, H-4′), 7.10-7.19 (1H,
m, aromatic-H), 7.30-7.31 (1H, m, H-3′), 7.40 (1H, d, J ) 7.8
Hz, aromatic-H), 7.64 (1H, d, J ) 7.8 Hz, aromatic-H), 7.69
(1H, s, H-5′), 7.75 (1H, dd, J ) 8.8, 1.9 Hz, H-8), 7.82 (1H, s,
aromatic-H), 7.91 (1H, d, J ) 8.8 Hz, H-7), 8.50 (1H, d, J )
1.9 Hz, H-10), 9.10 (1H, bs, NH).
Meth od C (Dep r otection of Boc Gr ou p ). To a solution
9-Ch lor o-2-(2-fu r a n yl)-5-[(4-iod op h en yla cetyl)a m in o]-

Derivatives of CGS 15943
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 15 2843
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (9): 1H NMR (CDCl3) δ
4.37 (2H, s, -CH2CO), 6.64 (1H, s, H-4′), 7.17 (2H, d, J ) 7.8
Hz, aromatic-H), 7.28-7.29 (1H, m, H-3′), 7.69-7.76 (4H, m,
H-5′, H8 and aromatic-H), 7.90 (1H, d, J ) 8.8 Hz, H7), 8.03
(1H, s, H-5′), 8.50 (1H, s, H-10), 9.08 (1H, s, NH).
-CH2), 1.80-1.90 (2H, m, -CH2), 3.05 (2H, t, J ) 6.8 Hz,
-CH2), 3.14-3.18 (2H, m, -CH2), 4.66 (1H, bs, NH), 6.62 (1H,
m, H-4′), 7.29 (1H, m, H-3′), 7.67 (1H, s, H-5′), 7.70 (2H, dd, J
) 8.8, 1.9 Hz, H-8), 7.84 (1H, d, J ) 8.8 Hz, H-7), 8.42 (1H, d,
J ) 1.9 Hz, H-10), 9.01 (1H, bs, NH).
9-Ch lor o-2-(2-fu r an yl)-5-[(3-ch lor oph en ylacetyl)am in o]-
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (10): 1H NMR (CDCl3) δ
4.42 (2H, s, -CH2CO), 6.64-6.65 (1H, m, H-4′), 7.28-7.32 (4H,
m, aromatic-H and H-3′), 7.45 (1H, s, aromatic-H), 7.68 (1H,
s, H-5′), 7.76 (1H, dd, J ) 8.8, 2.9 Hz, H-8), 7.91 (1H, d, J )
8.8 Hz, H-7), 8.50 (1H, d, J ) 2.9 Hz, H-10), 9.10 (1H, bs, NH).
9-Ch lor o-2-(2-fu r an yl)-5-[[(2R)-ph en ylpr opion yl]am in o]-
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (11): 1H NMR (CDCl3) δ
1.69 (3H, d, J ) 6.8 Hz, -CH3), 4.54 (1H, q, J ) 6.8 Hz,
-CHCO), 6.60-6.62 (1H, m, H-4′), 7.20-7.21 (1H, m, H-3′),
7.29-7.50 (5H, m, aromatic-H), 7.65 (1H, s, H-5′), 7.72 (1H,
dd, J ) 8.8, 2.9 Hz, H-8), 7.94 (1H, d, J ) 8.8 Hz, H-7), 8.44
5-[[7-[(ter t-Bu toxyca r bon yl)a m in o]h ep ta n oyl]a m in o]-
9-ch lor o-2-(2-fu r an yl)[1,2,4]tr iazolo[1,5-c]qu in azolin e (23):
1H NMR (CDCl3) δ 1.45 (9H, s, -Boc), 1.50-1.61 (6H, m, 3 ×
-CH2), 1.86 (2H, m, J ) 6.85 Hz, -CH2), 3.03 (2H, t, J ) 7.81
Hz, -CH2), 3.13 (2H, m, -CH2), 4.59 (1H, bs, NH), 6.63 (1H,
m, J ) 1.94 Hz, H-4′), 7.30 (1H, s, H-3′), 7.67 (1H, s, H-5′),
7.72 (1H, m, H-8), 7.86 (1H, d, J ) 8.79 Hz, H-7), 8.45 (1H, d,
J ) 2.94 Hz, H-10), 9.01 (1H, s, NH).
5-[(D-Alan yl)am in o]-9-ch lor o-2-(2-fu r an yl)[1,2,4]tr iazolo-
[1,5-c]qu in a zolin e (24): 1H NMR (DMSO-d6) δ 1.14 (3H, m,
-CH3), 4.25 (1H, bs, H-R), 5.26 (2H, bs, NH2), 6.73 (1H, m,
H-4′), 7.12 (1H, m, H-3′), 7.64 (2H, m, H-8 and H-7), 7.93 (1H,
m, H-5′), 8.17 (1H, m, H-10), 11.63 (1H, bs, NH); [R]D20 ) +6.00
(c ) 0.1, CHCl3/CH3OH, 10:1). HPLC retention time: 4.11 min
(>95% purity) using solvent system A, 3.91 min (>95% purity)
using solvent system B.
20
(1H, d, J ) 2.9 Hz, H-10), 9.03 (1H, bs, NH); [R]D ) +1.04 (c
) 0.23, CHCl3).
9-Ch lor o-2-(2-fu r an yl)-5-[[(2S)-ph en ylpr opion yl]am in o]-
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (12): 1H NMR (CDCl3)
20
identical to 11, [R]D ) -2.26 (c ) 0.38, CHCl3).
5-[(L-Alan yl)am in o]-9-ch lor o-2-(2-fu r an yl)[1,2,4]tr iazolo-
[1,5-c]qu in a zolin e (25): [R]D20 ) -7.33 (c ) 0.09, CHCl3/CH3-
OH, 10:1). HPLC retention time: 4.09 min (>95% purity) using
solvent system A, 4.13 min (>95% purity) using solvent system
B.
9-Ch lor o-2-(2-fu r a n yl)-5-(d ip h en yla cetyla m in o)[1,2,4]-
tr ia zolo[1,5-c]qu in a zolin e (13): 1H NMR (CDCl3) δ 6.09 (1H,
s, -CHCO), 6.61-6.64 (1H, m, H-4′), 7.05-7.45 (11H, m, H-3′
and aromatic-H), 7.65 (1H, s, H-5′), 7.73 (1H, dd, J ) 8.8, 2.9
Hz, H-8), 7.92 (1H, d, J ) 8.8 Hz, H-7), 8.46 (1H, d, J ) 2.9
Hz, H-10), 9.26 (1H, bs, NH).
9-Ch lor o-2-(2-fu r an yl)-5-[(2,2-diph en ylpr opion yl)am in o]-
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (14): 1H NMR (CDCl3) δ
2.18 (3H, s, -CH3), 6.57-6.59 (1H, m, H-4′), 7.06 (1H, d, J )
3.9 Hz, H-3′), 7.29-7.46 (10H, m, aromatic-H), 7.63 (1H, s,
H-5′), 7.71 (1H, dd, J ) 8.8, 1.9 Hz, H-8), 8.01 (1H, d, J ) 8.8
Hz, H-7), 8.45 (1H, d, J ) 1.9 Hz, H-10), 9.33 (1H, bs, NH).
9-Ch lor o-2-(2-fu r a n yl)-5-[(3-p h en ylp r op ion yl)a m in o]-
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (15): 1H NMR (CDCl3) δ
3.16 (2H, t, J ) 7.8 Hz, -CH2), 3.43 (2H, t, J ) 7.8 Hz, -CH2),
6.64 (1H, s, H-4′), 7.22-7.25 (1H, m, H-3′), 7.30-7.33 (5H, m,
aromatic-H), 7.70 (1H, s, H-5′), 7.74 (1H, dd, J ) 8.8, 1.9 Hz,
H-8), 7.85 (1H, d, J ) 8.8 Hz, H-7), 8.49 (1H, d, J ) 1.9 Hz,
H-10), 8.98 (1H, bs, NH).
5-[(3-Am in op r op ion yl)a m in o]-9-ch lor o-2-(2-fu r a n yl)-
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (26): 1H NMR (DMSO-d6)
δ 3.13-3.15 (4H, m, 2 × -CH2), 6.80-6.81 (1H, m, H-4′), 7.36
(1H, d, J ) 2.9 Hz, H-3′), 7.83 (2H, bs, NH2), 7.90 (1H, d, J )
8.8 Hz, H-7), 7.95 (1H, dd, J ) 8.8, 1.9 Hz, H-8), 8.03 (1H, s,
H-5′), 8.42 (1H, d, J ) 1.9 Hz, H-10), 11.10 (1H, s, NH).
5-[(5-Am in op en t a n oyl)a m in o]-9-ch lor o-2-(2-fu r a n yl)-
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (28): 1H NMR (DMSO-d6)
δ 1.68 (4H, bs, 2 × -CH2), 2.73-2.76 (2H, m, -CH2), 2.85 (2H,
bs, -CH2), 6.78-6.79 (1H, m, H-4′), 7.35 (1H, d, J ) 2.9 Hz,
H-3′), 7.70 (2H, bs, NH2), 7.92 (2H, s, H-8 and H-7), 8.02 (1H,
s, H-5′), 8.40 (1H, s, H-10), 11.07 (1H, s, NH). HPLC retention
time: 6.3 min (>95% purity) using solvent system A, 8.7 min
(>95% purity) using solvent system B.
5-[(6-Am in oh exa n oyl)a m in o]-9-ch lor o-2-(2-fu r a n yl)-
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (29): 1H NMR (DMSO-d6)
δ 1.42-1.70 (6H, m, 3 × -CH2), 2.72 (2H, t, J ) 6.8 Hz, -CH2),
2.81 (2H, t, J ) 6.8 Hz, -CH2), 6.78-6.79 (1H, m, H-4′), 7.34
(1H, d, J ) 3.9, H-3′), 7.70 (2H, bs, NH2), 7.92 (2H, d, J ) 1.9
Hz, H-8 and H-7), 8.01 (1H, s, H-5′), 8.40 (1H, s, H-10), 11.04
(1H, s, NH).
5-[(7-Am in oh ep t a n oyl)a m in o]-9-ch lor o-2-(2-fu r a n yl)-
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (30): 1H NMR (DMSO-d6)
δ 1.24 (2H, m, -CH2), 1.39 (2H, m, -CH2), 1.53 (2H, m, -CH2),
1.66 (2H, m, -CH2), 2.71 (2H, m, -CH2), 2.78 (2H, m, -CH2),
6.79 (1H, s, H-4′), 7.35 (1H, s, H-3′), 7.64 (2H, bs, NH2), 7.93
(2H, s, H-7 and H-8), 8.02 (1H, s, H-5′), 8.40 (1H, s, H-10),
11.04 (1H, s, NH). HPLC retention time: 4.6 min (>95% purity)
using solvent system A, 8.5 min (>95% purity) using solvent
system B.
9-Ch lor o-2-(2-fu r a n yl)-5-(cin n a m oyla m in o)[1,2,4]t r i-
1
a zolo[1,5-c]qu in a zolin e (16): H NMR (CDCl3) δ 6.66 (1H,
m, H-4′), 7.35 (1H, d, J ) 3.9 Hz, H-3′), 7.48-7.49 (3H, m,
aromatic-H), 7.68-7.72 (3H, m, H-5′ and aromatic-H), 7.75-
7.78 (2H, m, H-8 and olefinic-H), 7.97 (1H, d, J ) 8.7 Hz, H-7),
8.02 (1H, d, J ) 16.6 Hz, olefinic-H), 8.53 (1H, d, J ) 1.9 Hz,
H-10), 9.10 (1H, bs, NH).
5-[[N-(ter t-Bu toxyca r bon yl)-D-a la n yl]a m in o]-9-ch lor o-
2-(2-fu r a n yl)[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (17) a n d th e
cor r esp on d in g L-en a n tiom er (18): 1H NMR (CDCl3) δ 1.50
(9H, s, -Boc), 1.58 (3H, d, J ) 5.8 Hz, -CH3), 4.95 (1H, m,
H-R), 5.13 (1H, bs, NH), 6.64 (1H, m, H-4′), 7.31 (1H, d, J )
2.9 Hz, H-3′), 7.67 (1H, s, H-5′), 7.75 (2H, dd, J ) 8.7, 1.9 Hz,
H-8), 7.97 (1H, d, J ) 8.7 Hz, H-7), 8.51 (1H, d, J ) 1.9 Hz,
H-10), 9.92 (1H, bs, NH).
5-[[5-(Ben zyloxyca r bon yl)p en ta n oyl]a m in o]-9-ch lor o-
2-(2-fu r a n yl)[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (31). The
title compound was synthesized using method B with monobenz-
yl adipate, which was prepared according to the literature:39
1H NMR (CDCl3) δ 1.82-1.88 (4H, m, 3 × -CH2), 2.48 (2H, t,
J ) 6.8 Hz, -CH2), 3.09 (2H, t, J ) 6.8 Hz, -CH2), 5.13 (2H,
s, -CH2), 6.62-6.63 (1H, m, H-4′), 7.28-7.36 (6H, m, H-3′ and
aromatic-H), 7.67-7.72 (2H, m, H-8 and H-5′), 7.84 (1H, d, J
) 8.8 Hz, H-7), 8.44 (1H, d, J ) 1.9 Hz, H-10), 8.97 (1H, s,
NH).
5-[[3-(Meth oxyca r bon yl)p r op ion yl]a m in o]-9-ch lor o-2-
(2-fu r a n yl)[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (32): 1H NMR
(CDCl3) δ 2.86 (2H, t, J ) 6.8 Hz, -CH2), 3.44 (2H, t, J ) 6.8
Hz, -CH2), 3.75 (3H, s, -OCH3), 6.62-6.64 (1H, m, H-4′),
7.28-7.31 (1H, m, H-3′), 7.68 (1H, s, H-5′), 7.72 (1H, d, J )
8.8, 1.9 Hz, H-8), 7.86 (1H, d, J ) 8.8 Hz, H-7), 8.46 (1H, d, J
) 1.9 Hz, H-10), 9.12 (1H, s, NH).
5-[[3-[(ter t-Bu toxyca r bon yl)a m in o]p r op ion yl]a m in o]-
9-ch lor o-2-(2-fu r an yl)[1,2,4]tr iazolo[1,5-c]qu in azolin e (19):
1H NMR (CDCl3) δ 1.44 (9H, s, -Boc), 3.32 (2H, t, J ) 5.8 Hz,
-CH2), 3.62 (2H, q, J ) 5.8 Hz, -CH2), 5.28 (1H, bs, NH),
6.64 (1H, m, H-4′), 7.31 (1H, d, J ) 3.8 Hz, H-3′), 7.68 (1H, s,
H-5′), 7.73 (2H, dd, J ) 8.8, 1.9 Hz, H-8), 7.87 (1H, d, J ) 8.8
Hz, H-7), 8.47 (1H, d, J ) 1.9 Hz, H-10), 9.06 (1H, bs, NH).
5-[[5-[(ter t-Bu toxyca r bon yl)a m in o]p en ta n oyl]a m in o]-
9-ch lor o-2-(2-fu r an yl)[1,2,4]tr iazolo[1,5-c]qu in azolin e (21):
1H NMR (CDCl3) δ 1.44 (9H, s, -Boc), 1.63-1.73 (2H, m,
-CH2), 1.81-1.92 (2H, m, -CH2), 3.10 (2H, t, J ) 6.8 Hz,
-CH2), 3.22 (2H, q, J ) 6.8 Hz, -CH2), 4.67 (1H, bs, NH),
6.63 (1H, m, H-4′), 7.30 (1H, d, J ) 3.9 Hz, H-3′), 7.68 (1H, s,
H-5′), 7.72 (2H, dd, J ) 8.8, 1.9 Hz, H-8), 7.87 (1H, d, J ) 8.8
Hz, H-7), 8.46 (1H, d, J ) 1.9 Hz, H-10), 9.01 (1H, bs, NH).
5-[[6-[(ter t-Bu t oxyca r b on yl)a m in o]h exa n oyl]a m in o]-
9-ch lor o-2-(2-fu r an yl)[1,2,4]tr iazolo[1,5-c]qu in azolin e (22):
1H NMR (CDCl3) δ 1.45 (9H, s, -Boc), 1.50-1.61 (4H, m, 2 ×
5-[[7-(Meth oxyca r bon yl)h ep ta n oyl]a m in o]-9-ch lor o-2-

2844 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 15
Kim et al.
(2-fu r a n yl)[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (33): 1H NMR
(CDCl3) δ 1.47 (2H, m, -CH2), 1.68 (2H, t, J ) 6.83 Hz, -CH2),
1.85 (2H, t, J ) 6.83 Hz, -CH2), 2.34 (2H, t, J ) 7.81 Hz,
-CH2), 3.05 (2H, t, J ) 7.81 Hz, -CH2), 3.67 (3H, s, -OCH3),
6.65 (1H, s, H-4′), 7.33 (1H, d, J ) 3.91 Hz, H-3′), 7.69 (1H, s,
H-5′), 7.76 (1H, d, J ) 6.84 Hz, H-8), 7.89 (1H, d, J ) 9.96 Hz,
H-7), 8.50 (1H, d, J ) 1.95 Hz, H-10), 8.99 (1H, s, NH).
5-[(4-Ca r boxylbu ta n oyl)a m in o]-9-ch lor o-2-(2-fu r a n yl)-
[1,2,4]tr ia zolo[1,5-c]qu in a zolin e (34).
rolipram/cilostamide (final concentration 10 µM/10 µM), fol-
lowed by 50 µL of test compound (appropriate concentration)
or buffer. After incubation for 40 min at 37 °C, 100 µL of
NECA was added (final concentration 50 µM for compound
screening and IC50 determination, appropriate concentrations
for recording dose-response curves for pA2 calculations). After
15 min, incubation at 37 °C was terminated by removing the
medium and adding 200 µL of 0.1 M HCl. Wells were stored
at -20 °C until assay.
A solution of 30 mg (0.105 mmol) of 1a and 60 mg (0.525
mmol) of glutaric anhydride in 2 mL of acetone was refluxed
at 85 °C for 2 days. After cooling, the mixture was purified
by preparative silica gel TLC (CHCl3:MeOH, 10:1) and crystal-
lization from CH2Cl2 to afford 23 mg of 34 as a white solid:
1H NMR (DMSO-d6) δ 1.87 (2H, t, J ) 6.8 Hz, -CH2), 2.35
(2H, t, J ) 6.8 Hz, -CH2), 2.73 (2H, t, J ) 6.8 Hz, -CH2),
6.77 (1H, s, H-4′), 7.33 (1H, s, H-3′), 7.89 (2H, s, H-7 and H-8),
8.00 (1H, s, H-5′), 8.37 (1H, s, H-10).
P h a r m a cology. Ra d ioliga n d Bin d in g Stu d ies. Binding
of [3H]-(R)-N6-(phenylisopropyl)adenosine ([3H]R-PIA; Amer-
sham, Chicago, IL) to A1 receptors from rat cerebral cortical
membranes and of [3H]CGS 21680 (DuPont NEN, Boston, MA)
to A2A receptors from rat striatal membranes was performed
as described previously.4,5 Adenosine deaminase (3 units/mL)
was present during the preparation of the brain membranes,
in a preincubation of 30 min at 30 °C, and during the
incubation with the radioligands.
Binding of [125I]-N6-(4-amino-3-iodobenzyl)-5′-(N-methylcar-
bamoyl)adenosine ([125I]AB-MECA; Amersham) in membranes
prepared from HEK-293 cells stably expressing the human A3
receptor (Receptor Biology, Inc., Baltimore, MD) was as
described.24 The assay medium consisted of a buffer contain-
ing 50 mM Tris, 10 mM Mg2+, and 1 mM EDTA, at pH 8.0.
The glass incubation tubes contained 100 µL of the membrane
suspension (0.3 mg of protein/mL, stored at -80 °C in the same
buffer), 50 µL of [125I]AB-MECA (final concentration 0.3 nM),
and 50 µL of a solution of the proposed antagonist. Nonspecific
binding was determined in the presence of 200 µM NECA.
All nonradioactive compounds were initially dissolved in
DMSO and diluted with buffer to the final concentration,
where the amount of DMSO never exceeded 2%.
Incubations were terminated by rapid filtration over What-
man GF/B filters, using a Brandell cell harvester (Brandell,
Gaithersburg, MD). The tubes were rinsed three times with
3 mL of buffer each.
At least five different concentrations of competitor, spanning
3 orders of magnitude adjusted appropriately for the IC50 of
each compound, were used. IC50 values, calculated with the
nonlinear regression method implemented in the InPlot pro-
gram (Graph-PAD, San Diego, CA), were converted to apparent
Ki values using the Cheng-Prusoff equation40 and Kd values
of 1.0 nM ([3H]R-PIA), 14 nM ([3H]CGS 21680), and 0.59 nM
([125I]AB-MECA at human A3 receptors). Hill coefficients of
the tested compounds were in the range of 0.8-1.1.
Cyclic AMP Assa y of An ta gon ist P oten cy a t th e Hu -
m a n A2B Recep tor : 1. Cell Cu ltu r e. CHO-A2B cells were
grown under 5% CO2/95% O2 humidified atmosphere at a
temperature of 37 °C in DMEM supplemented with Hams F12
nutrient mixture (1/1), 10% newborn calf serum, 2 mM
glutamine, and containing 50 IU/mL penicillin, 50 µg/mL
streptomycin, and 0.2 mg/mL Geneticin (G418, Boehringer
Mannheim). Cells were cultured in 10-cm diameter round
plates and subcultured when grown confluent (approximately
after 72 h). PBS/EDTA containing 0.25% trypsin was used
for detaching the cells from the plates. Experimental cultures
were grown overnight as a monolayer in 24-well tissue culture
plates (400 µL/well; 0.8 × 106 cells/well). CHO-A2B cells were
kindly provided by Dr. K.-N. Klotz (University of Wu¨rzburg,
Germany).
3. Cyclic AMP Deter m in a tion . The amounts of cyclic
AMP were determined after a protocol with cAMP binding
protein (PKA)41 with the following minor modifications. As a
buffer was used 150 mM K2HPO4/10 mM EDTA/0.2% BSA FV
at pH 7.5. Samples (20 µL) were incubated for 90 min at 0
°C. Incubates were filtered over GF/C glass microfiber filters
in a Brandel M-24 cell harvester. The filters were additionally
rinsed with 4 × 2 mL of 150 mM K2HPO4/10 mM EDTA (pH
7.5, 4 °C). Punched filters were counted in Packard emulsifier
safe scintillation fluid after 2 h of extraction.
Abbr evia tion s: AcOH, acetic acid; Boc, tert-butoxycarbon-
yl; CGS 21680, 2-[[4-(2-carboxyethyl)phenyl]ethylamino]-5′-(N-
ethylcarbamoyl)adenosine; CGS 15943, 9-chloro-2-(2-furanyl)-
[1,2,4]triazolo[1,5-c]quinazolin-5-amine; CHO cells, Chinese
hamster ovary cells; DMAP, 4-(dimethylamino)pyridine; EDAC,
3-ethyl-1-(3-(dimethylamino)propyl)carbodiimide; DMF, N,N-
dimethylformamide; DMSO, dimethyl sulfoxide; EDTA, eth-
ylenediaminetetraacetate; EI, electric ionization; CI, chemical
ionization; FAB, fast atom bombardment; HEK cells, human
embryonic kidney cells; HMPA, hexamethylphosphotriamide;
[
125I]AB-MECA, [125I]-N6-(4-amino-3-iodobenzyl)adenosine-5′-
N-methyluronamide; Ki, equilibrium inhibition constant; MS,
mass spectrum; NECA, 5′-(N-ethylcarbamoyl)adenosine; R-
PIA, (R)-N6-(phenylisopropyl)adenosine; SAR, structure-activ-
ity relationship; TFA, trifluoroacetic acid; THF, tetrahydro-
furan; TLC, thin-layer chromatography; Tris, tris(hydroxy-
methyl)aminomethane; TEAA, triethylammonium acetate.
Refer en ces
(1) Mu¨ller, C. E. A1-adenosine receptor antagonists. Exp. Opin. Ther.
Patents 1997, 7, 419-440.
(2) Baraldi, P. G.; Cacciari, B.; Spalluto, G.; Borioni, A.; Viziano,
M.; Dionisotti, S.; Ongini, E. Current developments of A2A
adenosine receptor antagonists. Curr. Med. Chem. 1995, 2, 707-
722.
(3) Mu¨ller, C. E.; Stein, B. Adenosine receptor antagonists
-
structures and potential therapeutic applications. Curr. Pharm.
Des. 1996, 2, 501-530.
(4) Schwabe, U.; Trost, T. Characterization of adenosine receptors
in rat brain by (-) [3H]N6-phenylisopropyladenosine. Naunyn-
Schmiedeberg’s Arch. Pharmacol. 1980, 313, 179-187.
(5) J arvis, M. F.; Schutz, R.; Hutchison, A. J .; Do, E.; Sills, M. A.;
Williams, M. [3H]CGS 21680, an A2 selective adenosine receptor
agonist directly labels A2 receptors in rat brain tissue. J .
Pharmacol. Exp. Ther. 1989, 251, 888-893.
(6) Fredholm, B. B. Adenosine and neuroprotection. Intl. Rev.
Neurobiol. 1997, 40, 259-280.
(7) Ongini, E.; Dionisotti, S.; Morelli, M.; Ferre, S.; Svenningsson,
P.; Fuxe, K.; Fredholm, B. B. Neuropharmacology of the adenos-
ine A2A receptors. Drug Dev. Res. 1996, 39, 450-460.
(8) Latini, S.; Pazzagli, M.; Pepeu, G.; Pedata, F. A2 adenosine
receptors - their presence and neuromodulatory role in the
central-nervous-system. Gen. Pharmacol. 1996, 27, 925-933.
(9) Mehra, A.; Cohen, G.; J ohnson, J . V.; Elkayam, U. Effect of
adenosine on the renal circulation in patients with chronic heart-
failure. J . Am. Coll. Cardiol. 1997, 29, 7834-7834.
(10) Rongen, G. A.; Floras, J . S.; Lenders, J .; Thien, T.; Smits, P.
Cardiovascular pharmacology of purines. Clin. Sci. 1997, 92, 13-
24.
(11) J acobson, K. A.; Suzuki, F. Recent developments in selective
agonists and antagonists acting at purine and pyrimidine
receptors. Drug Dev. Res. 1996, 39, 289-300.
(12) J acobson, K. A.; von Lubitz, D. K. J . E.; Daly, J . W.; Fredholm,
B. B. Adenosine receptor ligands - differences with acute versus
chronic treatment. Trends Pharmacol. Sci. 1996, 17, 108-113.
(13) Tracey, W. R.; Magee, W.; Masamune, H.; Kennedy, S. P.;
Knight, D. R.; Buchholz, R. A.; Hill, R. J . Selective adenosine
A3 receptor stimulation reduces ischemic myocardial injury in
the rabbit heart. Cardiovasc. Res. 1997, 33, 410-415.
2. Cyclic AMP Gen er a tion . Cyclic AMP generation was
performed in DMEM/HEPES buffer (DMEM containing 50
mM HEPES, pH 7.4, 37 °C). To each well, washed twice with
DMEM/HEPES buffer, were added 100 µL of adenosine
deaminase (final concentration 10 IU/mL) and 100 µL of

Derivatives of CGS 15943
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 15 2845
(14) Kohno, Y.; Sei, Y.; Koshiba, M.; Kim, H. O.; J acobson, K. A.
Induction of apoptosis in HL-60 human promyelocytic leukemia-
cells by adenosine A3 receptor agonists. Biochem. Biophys. Res.
Commun. 1996, 221, 849-849.
(15) Karton, Y.; J iang, J . L.; J i, X. D.; Melman, N.; Olah, M. E.; Stiles,
G. L.; J acobson, K. A. Synthesis and biological-activities of
flavonoid derivatives as as adenosine receptor antagonists. J .
Med. Chem. 1996, 39, 2293-2301.
(16) J acobson, M. A.; Chakravarty, P. K.; J ohnson, R. G.; Norton, R.
Novel nonxanthine selective A3 adenosine receptor antagonists.
Drug. Dev. Res. 1996, 37, 131-131.
(28) Chiota, C.; Levitt, M.; Richardson, D. Helix to helix packing in
proteins. J . Mol. Biol. 1981, 145, 215-250.
(29) Schertler, G. F.; Villa, C.; Henderson, R. Projection structure of
rhodopsin. Nature 1993, 362, 770-772.
(30) Urger, V. M.; Hargrave, P. A.; Baldwin, J . M.; Schertler, G. F.
X. Arrangement of rhodopsin transmembrane R-helices. Nature
1997, 389, 203-206.
(31) Gouldson, P. R.; Snell, C. R.; Reynolds, C. A. A new approach to
docking in the â2-adrenergic receptor that exploits the domain
structure of G-protein-coupled receptors. J . Med. Chem. 1997,
40, 3871-3886.
(17) van Rhee, A. M.; J iang, J . L.; Melman, N.; Olah, M. E.; Stiles,
G. L.; J acobson, K. A. Interaction of 1,4-dihydropyridine and
pyridine-derivatives with adenosine receptors - selectivity for
A3 receptors. J . Med. Chem. 1996, 39, 2980-2989.
(18) J iang, J . L.; Van Rhee, A. M.; Melman, N.; J i, X. D.; J acobson,
K. A. 6-phenyl-1,4-dihydropyridine derivatives as potent and
selective A3 adenosine receptor antagonists. J . Med. Chem. 1996,
39, 4667-4675.
(19) J iang, J . L.; Van Rhee, A. M.; Chang, L.; Patchornik, A.; J i, X.
D.; Evans, P.; Melman, N.; J acobson, K. A. Structure-activity-
relationships of 4-(phenylethynyl)-6-phenyl-1,4-dihydropyridines
as highly selective A3 adenosine receptor antagonists. J . Med.
Chem. 1997, 40, 2596-2608.
(20) Kim, Y. C.; J i, X. D.; J acobson, K. A. Derivatives of the
triazoloquinazoline adenosine antagonist (CGS15943) are selec-
tive for the human A3 receptor subtype. J . Med. Chem. 1996,
39, 4142-4148.
(21) J acobson, M. A.; J ohnson, R. G.; Luneau, C. J .; Salvatore, C. A.
Cloning and chromosomal localization of the human A2B ad-
enosine receptor gene (ADORA2B) and its pseudogene. Genomics
1995, 27, 374-376.
(32) 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.
(33) Kim, J .; J iang, Q.; Glashofer, M.; Yehle, S.; Wess, J .; J acobson,
K. A. Glutamate residues in the second extracellular loop of the
human A2a adenosine receptor are required for ligand recogni-
tion. Mol. Pharmacol. 1996, 49, 683-691.
(34) The program SYBYL 6.3 is available from TRIPOS Associates,
St. Louis, MO; 1993.
(35) Mohamadi, F.; Richards, N. G. J .; Guida, W. C.; Liskamp, R.;
Lipton, M.; Caufield, C.; Chang, G.; Hendrickson, T.; Still, W.
C. MacroModel-An Integrated Software System for Modeling
Organic and Bioorganic Molecules using Molecular Mechanics.
J . Comput. Chem. 1990, 11, 440-450.
(36) van Rhee, A. M.; Fischer, B.; van Galen, P. J . M.; J acobson, K.
A. Modeling the P2Y purinoceptor using rhodopsin as template.
Drug Des. Discov. 1995, 13, 133-154.
(37) Kyte, J .; Doolittle, R. F. A simple method for displaying the
hydrophobic character of a protein. J . Mol. Biol. 1982, 157, 105-
132.
(38) Weiner, S. J .; Kollman, P. A.; Nguyen, D. T.; Case, D. A. An
all-atom force field for simulation of protein and nucleic acids.
J . Comput. Chem. 1986, 7, 230-252.
(39) English, A. R.; Girard, D.; J asys, V. J .; Martingano, R. J .;
Kellogg, M. S. Orally Effective Acid Prodrugs of the â-Lactamase
Inhibitor Sulbactam. J . Med. Chem. 1990, 33, 344-347.
(40) Cheng, Y. C.; Prusoff, W. H. Relationship between the inhibition
constant (Ki) and the concentration of inhibitor which causes
50% inhibition (IC50) of an enzyme reaction. Biochem. Pharma-
col. 1973, 22, 3099-3108.
(41) van der Wenden, E. M.; Hartog-Witte, H. R.; Roelen, H. C. P.
F.; von Frijtag Drabbe Ku¨nzel, J . K.; Pirovano, I. M.; Mathoˆt,
R. A. A.; Danhof, M.; van Aerschot, A.; Lidaks, M. J .; IJ zerman,
A. P.; Soudijn, W. 8-Substituted adenosine and theophylline-7-
riboside analogues as potential partial agonists for the adenosine
A1 receptor. Eur. J . Pharmacol.-Mol. Pharmacol. Sect. 1995, 290,
189-199.
(22) Gallo-Rodriguez, C.; J i, X. D.; Melman, N.; Siegman, B. D.;
Sanders, L. H.; Orlina, J .; Fischer, B.; Pu, Q. L.; Olah, M. E.;
van Galen, P. J . M.; Stiles, G. L.; J acobson, K. A. Structure-
activity-relationships of N-6-benzyladenosine-5′-uronamides as
A3-selective adenosine agonists. J . Med. Chem. 1994, 37, 636-
646.
(23) Salvatore, C. A.; J acobson, M. A.; Taylor, H. E.; Linden, J .;
J ohnson, R. G. Molecular cloning and characterization of the
human A3 adenosine receptor. Proc. Natl. Acad. Sci. U.S.A. 1993,
90, 10365-10369.
(24) Olah, M. E.; Gallo-Rodriguez, C.; J acobson, K. A.; Stiles, G. L.
[
125I]AB-MECA,
a high affinity radioligand for the rat A3
adenosine receptor. Mol. Pharmacol. 1994, 45, 978-982.
(25) Kim, J .; Wess, J .; van Rhee, A. M.; Schoneberg, T.; J acobson, K.
A. Site-directed mutagenesis identifies residues involved in
ligand recognition in the human A2a adenosine receptor. J . Biol.
Chem. 1995, 270, 13987-13997.
(26) Moro, S.; Guo, D.; Camaioni, E.; Boyer, J . L.; Harden, K.;
J acobson, K. A. Human P2Y1 receptor: molecular modeling and
site-directed mutagenesis as a tools to identify agonist and
antagonist recognition sites. J . Med. Chem. 1998, 41, 1456-
1466.
(27) Lin, Z.; Shenker, A.; Pearlstein, R. A model of the lutropin/
choriogonadotropin receptor: insights into the structural and
functional effects of constitutively activating mutations. Protein
Eng. 1997, 10, 501-510.
(42) Alexander, S. P. H.; Cooper, J .; Shine, J .; Hill, S. J . Character-
ization of the human brain putative A2B adenosine receptor
expressed in Chinese hamster ovary (CHO.A2B.4) cells. Br. J .
Pharmacol. 1996, 119, 1286-1290.
(43) Feoktistov, I.; Biaggioni, I. Adenosine A2B receptors. Pharmacol.
Rev. 1997, 49, 381-402.
J M980094B