6-(2-Furanyl)-9H-purin-2-amine derivatives as A2A adenosine antagonists

Article abstract of DOI:10.1016/j.bmcl.2005.02.031

Structure-activity relationships have been investigated through substitutions at the 9-position of the 2-amino-6-(2-furanyl) purine (5) to identify novel and selective A_2A adenosine receptor antagonists. Several potent and selective antagonists were identified. In particular, compounds 20, 25, and 26 show very high affinity with excellent selectivity.

Full text of DOI:10.1016/j.bmcl.2005.02.031

Bioorganic & Medicinal Chemistry Letters 15 (2005) 2119–2122
6-(2-Furanyl)-9H-purin-2-amine derivatives as A2A
adenosine antagonists
Eugenia Kiselgof, Deen B. Tulshian,* Leyla Arik, Hongtao Zhang and Ahmad Fawzi
Schering-Plough Research Institute, Chemical Research, 2015 Galloping Hill Road, MS 2545, Kenilworth, NJ 07033, USA
Received 10 January 2005; revised 8 February 2005; accepted 9 February 2005
Available online 19 March 2005
Abstract—Structure–activity relationships have been investigated through substitutions at the 9-position of the 2-amino-6-(2-fura-
nyl) purine (5) to identify novel and selective A2A adenosine receptor antagonists. Several potent and selective antagonists were iden-
tified. In particular, compounds 20, 25, and 26 show very high affinity with excellent selectivity.
Ó 2005 Elsevier Ltd. All rights reserved.
The purine nucleoside, adenosine, is known to act via
four major receptor subtypes, A1, A2A, A2B, and A3
which have been characterized according to their pri-
mary sequences.1 Adenosine A2A receptors are abun-
dant in the caudate–putamen, nucleus accumbens and
olfactory tubercle in several species.2 In the caudate–
putamen, adenosine A2A receptors are localized on
several neurons and have been shown to modulate the
neurotransmission of c-aminobutyric acid (GABA),
acetylcholine, and glutamate.3 These actions of the
A2A receptor contribute to the control of motor behav-
ior.4 A2A agonists inhibit locomotor activity and induce
catalepsy in rodents.5 In contrast, adenosine A2A antag-
onists prevent the motor disturbances of dopamine D2
receptor null mice.6 Recently, an A2A antagonist, KW-
6002, was found to have antiparkinsonian activity in
the parkinsonian monkey without producing hyperac-
tivity and provoking dyskinesia.7 These results suggest
that A2A antagonists have potential to be a new class
of antisymptomatic drugs for ParkinsonÕs disease.
zing the adenosine A2A receptor subtype.8 However,
SCH 58261 suffered from several drawbacks including
low selectivity, poor solubility, and pharmacokinetic
profile. Based on these observations, we undertook a
structure–activity relationship (SAR) investigation to
identify a novel A2A receptor antagonist using SCH
58261 as a template. Our plan was to replace the tricyclic
core of SCH 58261 with purine moiety such as in 1, and
investigate the SAR via substitution at N-9. In this com-
munication, we report the results of this investigation.
NH2
NH2
2
N
N
O
N
N
1
N
4
N
6
O
N
R
N
9
N
N
SCH 58261
1
Table 1. Receptor binding of phenyl alkyl analogues
NH2
In the past ten years, great efforts have been devoted to
identify potent and selective A2A adenosine antagonists.
SCH 58261, that displayed single-digit nanomolar po-
tency and modest selectivity (A2A Ki = 4.3 nM, A1/
A2A = 35) has been widely used as a tool for characteri-
N
N
O
R
N
N
Compound
R
A2A Ki (nM)
A1/A2A
SCH 58261
C6H5–
C6H5CH2–
4.3
68.5
34.9
270.5
6.8
35
3
6
7
86
7
8
C6H5CH2CH2–
C6H5CH2CH2CH2–
C6H5CH(Me)-
Keywords: Adenosine receptor; Antagonist; 2-Furanyl; 9-H Purin.
*
9
10
27
5
Corresponding author. Tel.: +1 908 740 3505; fax: +1 908 740 7164/
7152; e-mail: deen.tulshian@spcorp.com
381.0
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.02.031

2120
E. Kiselgof et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2119–2122
Table 2. Receptor binding of substituted benzyl analogues
small amount of a dialkylated derivative in 85% com-
bined yield. Bromide 3 was then subjected to Stille
coupling9 to produce compound 4, which upon depro-
tection under acidic conditions and neutralization
produced compound 5 in 70–80% yield. Compound 5,
6-(2-furanyl)-9H-purin-2-amine, was utilized to prepare
analogues with general structure 1. Alkylation of 5 with
various benzyl halides produced compounds described
in Tables 1 and 2. The benzyl halides were all commer-
cially available. Compound 6 was prepared using proce-
dure described in Scheme 1.
NH2
N
N
O
N
N
R
Compound
R
A2A Ki (nM)
A1/A2A
31
11
22.0
OMe
Compound 5 was subjected to cross-coupling reaction
using phenylboronic acid and cupric acetate10 to pro-
duce 6. All compounds gave satisfactory analytical
results.11
12
12.3
163
OMe
13
14
15
7.7
10.0
21.5
104
135
63
OMe
The results of the A2A adenosine receptor binding as-
says12 are expressed as inhibition constants (Ki, nM).
The A1/A2A describes the selectivity over A1 adenosine
receptor. Table 1 shows the SAR of compounds
where N-9 was substituted with phenyl 6 and phenyl
alkyls 7–9 derivatives. Substitution with either phenyl
(6) or phenethyl (8) moieties resulted in the significant
loss of affinity. The benzyl compound 7 shows signif-
icant increase in the selectivity with moderate decrease
in A2A adenosine receptor affinity. Introduction of a
methyl substituent at the benzylic site in compound
10 was detrimental to A2A receptor binding. Com-
pound 9 with three carbon linker exhibited binding
and selectivity very similar to SCH 58261. Since a
benzyl substituent at N-9 produced a significant in-
crease in selectivity over A1 adenosine receptor, the
SAR of a variety of substituted benzyl derivatives
was investigated in detail and results are presented
in Table 2.
Cl
16
17
20.5
4.5
98
CF3
F
F
162
F
18
19
26
58
F
F
F
8.7
185
CF3
CF3
20
21
2.8
6.3
418
106
F
The monosubstituted derivatives represented by com-
pounds 11–16 exhibited A2A adenosine receptor bind-
ing affinities in a very narrow range of 8–22 nM. The
para-substituted derivatives in general were more
selective over A1 receptor subtype. It was found that
the nature of the substituents on the benzyl was crit-
ical for the selectivity of A2A over A1 adenosine recep-
tor. The lipophilic substituents produced compounds
with significantly greater selectivity. Hence disubsti-
tuted compounds containing either two fluorine atoms
as in 17, or a combination of fluorine and trifluoro-
methyl as in 19 and 20 produced the best potency
and selectivity. Similarly, naphthalene derivatives 24
and 25 were also very potent and selective. The tri-
substituted compound 18 was significantly less potent
and selective. The phenyl sulfone derivative 26 was
the only monosubstituted compound identified to be
selective with very high affinity for A2A adenosine
receptor.
O
O
Cl
Cl
Cl
22
24
7.1
5.6
195
223
25
26
1.4
3.1
466
574
Ph
S
O
O
In summary, we have identified 6-(2-furanyl)-9H-purin-
2-amine derivatives as a novel class of compounds with
high A2A receptor antagonist affinities. Compared to
SCH 58261, several compounds in this series are more po-
tent and selective. Some of these compounds show better
solubility and pharmacokinetic profiles than SCH 58261.
Compounds presented in Tables 1 and 2 were prepared
from commercially available 2-amino-6-bromopurine 2
using general procedure described in Scheme 1. Accord-
ingly, compound 2 was protected with triphenylmethyl
to give compound 3 as a major product along with a

E. Kiselgof et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2119–2122
2121
NH2
N
NH2
N
NH2
N
N
N
N
a
b
O
Br
Br
Ph3C
N
HN
Ph3C
N
N
N
N
3
4
2
NH2
N
NH2
N
c
N
d
N
O
O
HN
R
N
N
N
5
7-26
e
NH2
N
N
O
N
N
6
Scheme 1. Reagents and conditions: (a) Ph3CCl, Et3N, CH2Cl2, rt; (b) 2-(tributylstannyl)furan, Pd(PPh3)2Cl2, CuI, THF, reflux; (c) HCl; (d) RX,
K2CO3, DMF, rt; (e) PhB(OH)2, Cu(OAc)2, pyridine, CH2Cl2.
1996, XXXVIII, 725; (b) Zocchi, C.; Ongini, E.; Ferrara,
S.; Baraldi, P. G.; Dionisotti, S. Br. J. Pharmacol. 1996,
117, 1381.
In particular, compounds 20, 25, and 26 show very high
affinity (Ki = 2.8, 1.4, 3.1 nM, respectively) with excellent
selectivity over A1 adenosine receptor.
9. Li, J. J.; Yue, W. S. Tetrahedron Lett. 1999, 40,
4507.
10. Lam, P. Y. S.; Clark, C. G.; Saubern, S.; Adams, J.;
Winters, M.; Chan, D. M. T.; Combs, A. Tetrahedron
Lett. 1998, 39, 2941.
Acknowledgments
We would like to thank Dr. William Greenlee for his
support and encouragement of this work. We would
also like to thank Zheng Tan for her help in the prepa-
ration of this manuscript.
11. For compound 20: 1H NMR (CDCl3): d 7.84 (s, 1H), 7.81
(d, J = 3.4 Hz, 1H), 7.49 (dd, J = 1.8, 0.8 Hz, 1H), 7.41–
7.37 (m, 3H), 6.65 (dd, J = 3.4, 1.8 Hz, 1H), 5.39 (s, 2H),
5.13 (br s, 2H); MS m/z (ES) 378 [M+1]+. For compound
25: 1H NMR (CDCl3): d 7.75–7.86 (m, 5H), 7.72 (dd,
J = 1.8, 0.8 Hz, 1H), 7.67 (br s, 1H), 7.49 (ddd, J = 9.9,
3.5, 3.5 Hz, 2H), 7.37 (dd, J = 8.4, 1.8 Hz, 1H), 6.63 (dd,
J = 3.4, 1.8 Hz, 1H), 5.44 (s, 2H), 5.21 (br s, 2H); MS m/z
(ES) 342 [M+1]+. For compound 26: 1H NMR (CDCl3): d
7.81–7.75 (m, 3H), 7.69 (dd, J = 1.8, 0.8 Hz, 1H), 7.65 (d,
J = 7.8 Hz, 1H), 7.52 (dd, J = 7.8, 7.8 Hz, 2H), 7.30 (ddd,
J = 14.1, 7.4, 2H), 7.20 (dd, J = 13.6, 7.4 Hz, 2H), 7.01
(d, J = 7.8 Hz, 1H), 6.61 (dd, J = 3.4, 1.8 Hz, 1H), 5.37 (s,
2H), 5.19 (br s, 2H), 4.64 (s, 2H); MS m/z (ES) 446
[M+1]+.
References and notes
1. Fredholm, B. B.; Abbracchio, M. P.; Burnstock, G.; Daly,
J. W.; Harden, T. K.; Jacobson, K. A.; Legg, P.; Williams,
M. Pharmacol. Rev. 1994, 46, 143.
2. (a) Svenningsson, P.; Le Moine, C.; Fisone, G.; Fredholm,
B. B. Prog. Neurobiol. 1999, 59, 355; (b) Schiffmann, S. N.;
Libert, F.; Vassart, G.; Dumont, J. E.; Vanderhaeghen, J.
J. Brain Res. 1990, 519, 333.
3. (a) Ochi, M.; Koga, K.; Kurokawa, M.; Kase, H.;
Nakamura, J.; Kuwana, Y. Neuroscience 2000, 100, 53;
(b) Mori, A.; Shindou, T.; Ichimura, M.; Nonka, H.;
Kase, H. J. Neurosci. 1996, 16, 605.
4. Kuwana, Y.; Shiozaki, S.; Kanda, T.; Kurokawa, M.;
Koga, K.; Ochi, M.; Ikeda, K.; Kase, H.; Jackson, M. J.;
Smith, L. A.; Pearce, R. K. B.; Jenner, P. Adv. Neurol.
1999, 80, 121.
5. (a) Shiozaki, S.; Ichikawa, S.; Nakamura, J.; Kitamura, S.;
Yamada, K.; Kuwana, Y. Psychopharmacology 1999, 147,
90; (b) Shiozaki, S.; Ichikawa, S.; Nakamura, J.; Kuwana,
Y. In Effects of Adenosine Receptors, in Adenosine Recep-
tors and ParkinsonÕs Disease; Kase, H., Richardson, P. J.,
Jenner, P., Eds.; Academic, 1999, pp 193–210.
6. Aoyama, S.; Kase, H.; Borreli, E. J. Neurosci. 2000, 20,
5848.
12. Adenosine A2A and A1 binding assays: [3H]SCH 58261
and [3H]DPCPX binding assays for adenosine A2A and A1
receptors, respectively, were performed as described
before.13 Briefly, 5 lg HEK cell membranes expressing
human adenosine A2A receptors were incubated with
different concentrations of compounds and 1 nM [3H]SCH
58261 in 200 lL assay buffer containing 2.7 mM KCl,
1.1 mM KH2PO4, 137 mM NaCl, 7.6 mM Na2HPO4,
10 mM MgCl2, 0.04% methyl cellulose, 20 lg/mL adeno-
sine deaminase, and 4% dimethyl sulfoxide. Adenosine A1
binding assays were performed on 10 lg CHO cell
membranes expressing human adenosine A1 receptors
and 1 nM [3H]DPCPX in 200 lL assay buffer. Reactions
were carried out for 60 min at room temperature (23 °C)
and were terminated by rapid filtration over GF/B filters.
Filters were washed seven times with 1 mL cold (4 °C)
distilled water, air dried, and radioactivity retained on
filters were counted in PackardÕs TopCount NXT micro-
plate scintillation counter. Compounds were tested at 10
different concentrations ranging from 0.1 nM to 3 lM.
Non-specific binding for adenosine A2A and A1 receptors
7. Kanda, T.; Jackson, M. J.; Smith, L. A.; Pearce, R. K. B.;
Nakamura, J.; Kase, H.; Kuwana, Y.; Jenner, P. Exp.
Neurol. 2000, 162, 321.
8. (a) Baraldi, P. G.; Cacciari, B.; Dionisotti, S.; Egan, J.;
Spalluto, G.; Zocchi, C. J. Labeled Compd. Radiopharm.

2122
E. Kiselgof et al. / Bioorg. Med. Chem. Lett. 15 (2005) 2119–2122
were determined in the presence of 10 lM CGS 15943 and
13. (a) Dionisotti, S.; Ongini, E.; Zocchi, C.; Kull, B.; Arslan,
G.; Fredholm, B. B. Br. J. Pharmacol. 1997, 121, 353; (b)
Klotz, K.-N.; Hessling, J.; Hegler, J.; Owman, C.; Kull,
B.; Fredholm, B. B.; Lohse, M. J. Naunyn-SchmiedebergÕs
Arch. Pharmacol. 1998, 357, 1.
10 lM NECA, respectively. Assays were performed in
duplicates and compounds were tested twice. Data were
fitted in one site competition binding model for IC50
determination using the program GraphPad Prism
(GraphPad Software, Inc., San Diego, CA) and Ki values
were calculated using Cheng and PrusoffÕs formula.14.
14. Cheng, Y.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22,
3099.