Antagonists of the human adenosine A2A receptor. Part 2: Design and synthesis of 4-arylthieno[3,2-d]pyrimidine derivatives

Article abstract of DOI:10.1016/j.bmcl.2008.03.076

We describe herein the discovery and development of a series of 4-arylthieno[3,2-d]pyrimidines which are potent adenosine A_2A receptor antagonists. These novel compounds show high degrees of selectivity against the human A₁, A_2B and A₃ receptor sub-types. Moreover, a number of these compounds show promising activity in vivo, suggesting potential utility in the treatment of Parkinson's disease.

Full text of DOI:10.1016/j.bmcl.2008.03.076

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry Letters 18 (2008) 2920–2923
Antagonists of the human adenosine A2A receptor. Part 2:
Design and synthesis of 4-arylthieno[3,2-d]pyrimidine derivatives
Roger J. Gillespie,a Ian A. Cliffe,a Claire E. Dawson,a Colin T. Dourish,a Suneel Gaur,a
Paul R. Giles,a Allan M. Jordan,b,* Antony R. Knight,a Anthony Lawrence,a
Joanne Lerpiniere,a Anil Misra,a Robert M. Pratt,a Richard S. Todd,a
Rebecca Upton,a Scott M. Weissa and Douglas S. Williamsonb
aVernalis (R&D) Ltd, 613 Reading Road, Winnersh, Wokingham RG41 5UA, UK
bVernalis (R&D) Ltd, Granta Park, Cambrige CB21 6GB, UK
Received 10 January 2008; revised 26 March 2008; accepted 27 March 2008
Available online 30 March 2008
Abstract—We describe herein the discovery and development of a series of 4-arylthieno[3,2-d]pyrimidines which are potent adeno-
sine A2A receptor antagonists. These novel compounds show high degrees of selectivity against the human A1, A2B and A3 receptor
sub-types. Moreover, a number of these compounds show promising activity in vivo, suggesting potential utility in the treatment of
Parkinson’s disease.
Ó 2008 Elsevier Ltd. All rights reserved.
Adenosine receptors comprise four distinct sub-types,
designated A1, A2A, A2B and A3. In the brain, adenosine
O
S
S
N
A2A receptors are located primarily in the striatum,
playing a key role in regulating movement. There is
strong evidence that adenosine A2A receptor antagonists
may provide a novel therapy for the treatment of Par-
kinson’s disease, with a lower risk of dyskinesias.1
N
NH2
1
The biaryl ketones were prepared by direct incorpora-
tion of the keto functionality via the reaction of a 2,4-
dichlorothieno[3,2-d]pyrimidine with the appropriate
aryl aldehyde in the presence of N,N-dimethylimidazoli-
um iodide and sodium hydride.3 A significant byproduct
was often isolated from this reaction which, following
analysis, was shown to be the directly coupled biaryl
derivative 2, lacking the keto linkage, as shown in
Scheme 1.
We have previously reported that a series of thieno[3,2-
d]pyrimidine ketones, typified by 1, are potent adenosine
A2A receptor antagonists which show good selectivity
over the A1 receptor.2 Furthermore, some of these com-
pounds also demonstrate activity in vivo. Unfortunately,
the compounds with the best in vivo activity in this series
had relatively modest selectivity and it proved difficult to
optimise both properties in the same molecule.
Initial testing indicated that, like their keto counterparts,
these compounds showed good A2A receptor binding
affinity4 with some selectivity over A1. These data sug-
gested that this series warranted further investigation.
We herein report the further development of this series,
leading to a novel class of biaryl A2A antagonists which
display high potency and selectivity against other aden-
osine receptor sub-types and, moreover, promising
in vivo activity.
Initially, a small series of 4-(2-thienyl) analogues, as de-
tailed in Table 1, were prepared and evaluated.
Encouraged by these preliminary results, an investiga-
tion of SAR around the 4-aryl group was undertaken.
Since the initial results indicated that the 2-ethyl and
2-dimethylamino substituents provided the best binding
affinity and selectivity, these groups were chosen as the
Keywords: Adenosine A2A receptor; Parkinson’s disease; Thienopyr-
imidine; In vivo activity.
*
Corresponding author. Tel.: +44 (0) 1223 895 555; fax: +44 (0) 1223
895 556; e-mail: a.jordan@vernalis.com
0960-894X/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2008.03.076

R. J. Gillespie et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2920–2923
2921
Ar
N
O
Cl
N
Ar
N,N-dimethylimidazolium
iodide, ArCHO, NaH
S
S
S
N
N
+
THF, reflux
Cl
Cl
N
Cl
N
2
Major (~ 50%)
Byproduct (~ 3-8%)
Scheme 1. Formation of biaryl derivatives with representative isolated yields.
as in compounds 11, 24, 36 and 37. Although good A2A
receptor affinity is retained in compounds containing
larger alkylamino substituents, for example, compound
40, selectivity is poorer. Compound 38, containing a
bulkier tert-butyl substituent, shows a similar in vitro
binding profile to analogues containing smaller alkyl
groups, but is inactive in vivo. It is likely that the in-
creased lipophilicity and higher logD of this compound
have a detrimental effect on the pharmacokinetic profile.
Table 1. Initial 2-position SAR
S
S
N
N
R
Compound
R
A2A Ki
(nM)
A1 Ki
(nM)
HaloLMA
activity5
3
4
5
6
7
8
Cl
Et
391
69
2041
1466
1590
813
Inactive
Inactive
Inactive
Inactive
Inactive
Active
OMe
NHEt
444
154
55
A number of the most promising candidates were exam-
ined for selectivity against the adenosine A2B and A3
receptor sub-types, using methods previously described.4
These data, detailed in Table 5, demonstrate that these
compounds show excellent selectivity for A2A over both
the A2B and A3 receptors, alongside the previously ob-
served selectivity over the A1 sub-type. It is particularly
noteworthy that excellent binding affinity and selectivity
can be achieved in molecules with very low molecular
weight (e.g., compound 11, MWt 247).
NMe2
NH(CH2)2OH
1963
401
172
Table 2. 2-Ethylthieno[3,2-d]pyrimidines: 4-aryl SAR
Aryl
S
N
N
Et
The functional activity of these compounds was deter-
mined in cells by assessing Ca2+ mobilisation using a
Fluorescence Imaging Plate Reader (FLIPR).6 All the
compounds tested were found to be functional antago-
nists of the A2A receptor and exhibited no appreciable
agonist activity (EC50 ꢀ 10 lM).
Compound Aryl
A2A Ki A1 Ki HaloLMA
(nM) activity5
(nM)
4
9
2-Thienyl
2-Furyl
69
30
1466 Inactive
3205 Active
10000 n/d
10
11
12
13
14
15
16
3-Furyl
2-Thiazolyl
681
1.6
2.1
142
779
13
388 Active
413 Active
5525 Active
4938 n/da
2-(4-Methyl-thiazolyl)
2-Imidazolyl
2-Triazoyl
As illustrated in the tables, a number of compounds in
this class were active in vivo in reversing haloperidol-in-
duced hypolocomotion in mice. For example, com-
pound 37, a potent and selective A2A antagonist,
showed significant efficacy when dosed ip at 30 mg kgꢁ1
2-Pyridyl
2-(6-Methyl-pyridyl)
917 Active
1273 Inactive
12
a n/d, not determined.
or sc at 10 mg kgꢁ1
.
The compounds employed in these studies were pre-
pared as described below.7 Compound 3 was prepared
as described in Scheme 2. Cyclisation of 3-aminothioph-
ene-2-carboxylate methyl ester 43 with urea gave diol 44,
the chlorination of which yielded 2,4-dichlorothieno[3,2-
d]pyrimidine 45. The treatment of 45 with thiophene-2-
boronic acid under standard Suzuki coupling conditions
then gave regioselective access to the desired C-4 substi-
tuted regioisomer 3.8
2-substituent for most compounds in this evaluation.
The results obtained are shown in Tables 2 and 3.
It was apparent from these data that, in both series, a 2-
thiazolyl group offered a significant advantage over
other heteroaryl groups in the 4-position and provided
a number of highly potent and selective analogues, sev-
eral of which showed promising in vivo activity. Based
on these data, the 2-thiazolyl group was selected as the
4-aryl substituent for a further round of optimisation
of the C-2 substituent, as described in Table 4.
Similarly, the cyclisation of 3-aminothiophene-2-car-
boxamide 46 with propionic anhydride, followed by
chlorination with phosphorous oxychloride yielded 4-
chloro-2-ethylthieno[3,2-d]pyrimidine 47. Suzuki cou-
pling with thiophene-2-boronic acid gave 4.
These data showed that, with a 2-thiazolyl substituent in
place at the C-4 position, a wide range of C-2 substitu-
ents are tolerated and provide a series of highly potent
and selective A2A antagonists. Potency and selectivity
over A1 is particularly good where the C-2 substituent
is a small lipophilic group such as alkyl or dialkylamino
The methoxy derivative 5 and the amino derivatives 6–8
were prepared via the 2-chloro precursor 3, by heating

2922
R. J. Gillespie et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2920–2923
Table 3. 2-Aminothieno[3,2-d]pyrimidines: 4-aryl SAR
Aryl
N
S
N
NRR'
Compound
Aryl
NRR0
A2A Ki (nM)
A1 Ki (nM)
HaloLMA activity5
7
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
2-Thienyl
NMe2
NMe2
NH2
55
1326
14
1963
1255
479
Inactive
n/d
Benzo-thiophene-2-yl
2-Furyl
2-Furyl
Active
Active
Active
Active
Inactive
n/d
NHMe
NHEt
NMe2
40
58
568
375
2-Furyl
2-Furyl
60
2197
868
2-Furyl
3-Furyl
NH(CH2)2OH
NMe2
NMe2
NH2
203
748
2.3
35
3467
366
2-Thiazoyl
2-Thiazoyl
Active
Active
Active
Inactive
Inactive
Active
Inactive
n/d
675
303
2-(4-Methyl-thiazolyl)
2-(5-Methyl-thiazolyl)
2-(4,5-Dimethyl-thiazolyl)
2-Imidazolyl
2-(1-Methyl-imidazolyl)
4-Pyrazoyl
NMe2
NMe2
NMe2
NMe2
NMe2
NMe2
NMe2
NMe2
NMe2
1.3
5.7
6.6
299
14
908
254
5768
1095
5513
1588
1260
990
3079
26
2-Pyridyl
2-(5-Methyl-pyridyl)
2-Pyrazinyl
n/d
Inactive
Inactive
19
81
Table 4. 4-(Thiazol-2-yl)thieno[3,2-d]pyrimidines: 2-position SAR
N
S
S
N
R
N
Compound
R
A2A Ki (nM)
A1 Ki (nM)
HaloLMA activity5
11
24
35
36
37
38
39
40
41
42
Et
NMe2
Cl
1.6
2.3
20
388
366
615
208
259
240
675
57
Active
Active
n/d
iso-Propyl
Cyclopropyl
tert-Butyl
NH2
1.4
3.3
3.1
35
Active
Active
Inactive
Active
Active
Active
Active
NH(CH2)2OH
N-(S)-Prolinol
N-(R)-Prolinol
18
2.8
3.6
14
250
Table 5. Binding affinity for selected compounds at all four human
adenosine receptor sub-types5
was coupled with a variety of heterocycles under Stille,
Suzuki or Negishi conditions to yield the desired biaryl
systems.
Compound
A2A Ki
(nM)
A1 Ki
(nM)
A2B Ki
(nM)
A3 Ki
(nM)
The amino derivatives 17–34 were also prepared by pal-
ladium or zinc-mediated coupling employing 2,4-dichlo-
11
24
36
37
41
1.6
2.3
1.4
3.3
2.8
388
366
208
259
14
721
1560
865
1813
1759
476
910
2479
945
456
S
OH
N
O
Cl
c
a
b
S
S
S
S
OMe
NH2
43
N
N
N
OH
Cl
N
Cl
N
3
with sodium methoxide or the required amine in N-
methyl pyrrolidone, respectively (Scheme 4).
44
45
Scheme 2. Reagents and conditions: (a) urea, 200 °C, 4 h, 83%; (b)
PhPOCl2, 170 °C, 2 h, 66%; (c) Pd(OAc)2, PPh3, THF, rt, 5 min then
thiophene-2-boronic acid, satd aq NaHCO3, reflux, 4 h, 94%.
Compounds 9–16 were prepared in the manner outlined
in Scheme 5. 4-Chloro-2-ethylthieno[3,2-d]pyrimidine 47

R. J. Gillespie et al. / Bioorg. Med. Chem. Lett. 18 (2008) 2920–2923
2923
Compounds 36–42 were all prepared via methodology
similar to the preparation of 11, 24 and 25.7
S
N
Cl
CONH2
NH2
c
S
a,b
S
S
N
In summary, 4-arylthieno[3,2-d]pyrimidines, serendipi-
tously discovered as byproducts in the preparation of
a related compound series, have been shown to display
strong functional antagonism of the human A2A recep-
tor. Optimisation of this series has led to A2A antago-
nists with low nM affinity and a high degree of
selectivity over other adenosine receptor sub-types.
Moreover, a number of these compounds show promis-
ing activity in vivo suggesting that they may have poten-
tial for the treatment of Parkinson’s disease. The further
evolution of this series is reported in the following
communication.9
N
Et
N
Et
46
47
4
Scheme 3. Reagents and conditions: (a) (i) propionic anhydride,
PhCH3, NEt3, reflux, 1.5 h, 82%; (ii) NaOH, reflux, 4 h, 100%; (b)
POCl3, reflux, 2.5 h, 72%; (c) Pd(OAc)2, PPh3, THF, rt, 5 min then
thiophene-2-boronic acid, satd aq NaHCO3, reflux, 4 h, 36%.
S
N
S
N
S
N
b
a
S
S
S
N
OMe
N
NR1R2
N
Cl
5
3
6-8
References and notes
Scheme 4. Reagents and conditions: (a) sodium methoxide, reflux,
18 h, 95%; (b) NHR1R2, NMP, 90 °C, 16 h, 42–77%.
1. Xu, K.; Bastia, E.; Schwarzschild, M. Pharmacol. Ther.
2005, 105, 267.
2. Gillespie, R. J.; Adams, D. R.; Bebbington, D.; Benwell,
K.; Cliffe, I. A.; Dawson, C. E.; Dourish, C. T.; Fletcher,
A.; Gaur, S.; Giles, P. R.; Jordan, A. M.; Knight, A. R.;
Knutsen, L. J. S.; Lawrence, A.; Lerpiniere, J.; Misra, A.;
Porter, R. H. P.; Pratt, R. M.; Shepherd, R.; Upton, R.;
Ward, S. E.; Weiss, S. M.; Williamson, D. S. Bioorg. Med.
Chem. Lett. 2008, 18, 2916.
Cl
Aryl
N
a, b or c
S
S
N
N
Et
N
Et
47
9-16
3. Miyashita, A.; Obae, K.; Suzuki, Y.; Oishi, E.; Iwamoto,
K.; Higashino, T. Heterocycles 1997, 45, 2159.
Scheme 5. Reagents and conditions: (a) (compound 9) PdCl2(PPh3)2,
DMF, Ar–SnBu3, rt, 16 h, 56%; (b) (compound 10) Pd(OAc)2, PPh3,
THF, rt, 5 min then aryl boronic acid, satd aq NaHCO3, reflux, 4 h,
36–52%; (c) (compounds 11–16); (i) heterocycle, THF, n-BuLi, ꢁ78 °C,
15 min then ZnCl2, Et2O, ꢁ78 °C to rt; (ii) 47, Pd(PPh3)4, reflux, 17 h,
37–65%.
4. For details of Ki determination, see: Weiss, S. M.; Benwell,
K.; Cliffe, I. A.; Gillespie, R. J.; Knight, A. R.; Lerpiniere,
J.; Misra, A.; Pratt, R. M.; Revell, D.; Upton, R.; Dourish,
C. T. Neurology 2003, 61, S101 (all values are the mean of
at least two separate values).
5. Reversal of haloperidol-induced hypolocomotion in mice
following ip administration of 30 mg kgꢁ1 test compound.
See: Bezard, E.; Imbert, C.; Gross, C. E. Rev. Neurosci.
1998, 9, 71; Mandhane, S. N.; Chopde, C. T.; Ghosh, A. K.
Eur. J. Pharmacol. 1997, 328, 135.
6. Porter, R. H. P.; Benwell, K. R.; Lamb, H.; Malcolm, C. S.;
Allen, N. H.; Revell, D. F.; Adams, D. R.; Sheardown, M.
J. Br. J. Pharmacol. 1999, 128, 13.
R
N
R
N
Cl
d
a, b or c
S
S
S
N
N
N
NR1R2
Cl
N
Cl
45
17-34
Scheme 6. Reagents and conditions: (a) (compounds 17–22, 32 and 33)
PdCl2(PPh3)2, DMF, Ar–SnBu3, rt, 16 h, 31–67%; (b) (compound 23)
Pd(OAc)2, PPh3, THF, rt, 5 min then aryl boronic acid, satd aq
NaHCO3, reflux, 4 h, 33%; (c) (compounds 24–31 and 34); (i)
heterocycle, THF, n-BuLi, ꢁ78 °C, 15 min then ZnCl2, Et2O, ꢁ78 °C
to rt; (ii) 45, Pd(PPh3)4, reflux, 17 h, 22–87%; (d) NHR1R2, NMP,
90 °C, 16 h, 13–100%.
7. Synthetic procedures are described in more detail in:
Gillespie, R. J.; Lerpiniere, J.; Dawson, C. E.; Gaur, S.;
Pratt, R. M. PCT Int. Appl. WO2002055524, 2002.
8. The alternative 4-chloro-2-(2-thienyl)thienopyrimidine reg-
ioisomer was prepared by an unambiguous route related to
that shown in Scheme 3 for the synthesis of compound 47,
and was shown not to correspond to 3, confirming the
desired 4-aryl regiochemistry.
9. Gillespie, R. J.; Cliffe, I. A.; Dawson, C. E.; Dourish, C. T.;
Gaur, S.; Jordan, A. M.; Knight, A. R.; Lerpiniere, J.;
Misra, A.; Pratt, R. M.; Roffey, J.; Stratton, G. C.; Upton,
R.; Weiss, S. M.; Williamson, D. S. Bioorg. Med. Chem.
Lett. 2008, 18, 2924.
rothieno[3,2-d]pyrimidine 45. Coupling with the desired
heterocycle under Suzuki, Stille or Negishi conditions
was followed by the displacement of the 2-Cl with the
relevant amine, in the manner illustrated in Scheme 6.