Substituted thieno[2,3-d]pyrimidines as adenosine A2A receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2013.02.078

A novel series of benzyl substituted thieno[2,3-d]pyrimidines were identified as potent A_2A receptor antagonists. Several five- and six-membered heterocyclic replacements for the optimized methylfuran were explored. Select compounds effectively reverse catalepsy in mice when dosed orally.

Full text of DOI:10.1016/j.bmcl.2013.02.078

Bioorganic & Medicinal Chemistry Letters 23 (2013) 2688–2691
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Substituted thieno[2,3-d]pyrimidines as adenosine A2A receptor
antagonists
⇑
Brian C. Shook , Devraj Chakravarty, J. Kent Barbay, Aihua Wang, Kristi Leonard, Vernon Alford,
Mark T. Powell, Stefanie Rassnick, Robert H. Scannevin, Karen Carroll, Nathaniel Wallace, Jeffrey Crooke,
Mark Ault, Lisa Lampron, Lori Westover, Kenneth Rhodes, Paul F. Jackson
Janssen Research & Development, LLC, Welsh and McKean Roads, PO Box 776, Spring House, PA 19477, United States
a r t i c l e i n f o
a b s t r a c t
Article history:
A novel series of benzyl substituted thieno[2,3-d]pyrimidines were identified as potent A2A receptor
antagonists. Several five- and six-membered heterocyclic replacements for the optimized methylfuran
were explored. Select compounds effectively reverse catalepsy in mice when dosed orally.
Ó 2013 Elsevier Ltd. All rights reserved.
Received 15 October 2012
Revised 11 February 2013
Accepted 19 February 2013
Available online 27 February 2013
Keywords:
Adenosine
A2A antagonists
Parkinson’s disease
Catalepsy
Parkinson’s disease (PD) is a progressive neurological disease
that is characterized by loss of dopamine (DA) producing neurons
in the ventral midbrain resulting in loss of DA input to the striatum.1
The loss of DA input leads to dysregulation of striatal function and
the classic motor symptoms of PD. The motor impairment is
typically treated with DA agonists or DA replacement therapy with
L-DOPA, the precursor to DA, being the gold standard treatment for
PD.
memory, and it is known that cognitive deficiencies increase as
the disease progresses.5a,8 Astellas’ dual A2A/A1 antagonist, ASP-
5854, significantly outperformed the selective A2A antagonist istra-
defylline (KW-6002) in several animal models of cognition.9
We recently reported a series of aminomethyl substituted thie-
no[2,3-d]pyrimidines that were potent adenosine A2A receptor
antagonists having excellent in vivo activity in the haloperidol-in-
duced catalepsy model in mice.10 We now report a series of aryl
Adenosine is a neuromodulator that coordinates responses to DA
and other neurotransmitters that play important roles for motor
function, mood, and memory.2 Adenosine receptors are comprised
of four distinct sub-types designated A1, A2A, A2B, and A3.3 Both
NH2
CN
R1-CN
N
tBuOK (0.2 eq)
N
R1
S
S
dioxane, 100°C
(25-95%)
NH2
NBS, DMF
(50-84%)
A2A and A1 receptors are highly expressed in the brain, particularly
1
2
striatum, while A2B and A3 receptors are more widely distributed in
tissues throughout the body.4 Selective A2A antagonists have been a
much sought after target for the symptomatic relief of PD,5 with
some reports suggesting that they may also slow disease progres-
sion because of their neuroprotective potential.6 To date, several
selective A2A antagonists have advanced into clinical development
including preladenant7 which is currently being evaluated in sev-
eral phase III trials.5
NH2
N
NH2
N
Path 2
Pd(dppf)Cl2
Br
S
R2
R1
N
R1
S
N
5
R2
3
B(OH)2
(Method C)
(77-87%)
R2B(OH)2 (Method A)
or
Me3SOI,
tBuOK,
(42-54%)
Path 1
R2ZnX (Method B)
(65-90%)
DMSO
More recently, it has been suggested that dual A2A/A1 antago-
nists may offer improved benefit to PD patients because A1 antag-
onists have shown efficacy in animal models of learning and
NH2
N
NH2
N
R2
R2
R1
N
R1
S
S
N
6
4
⇑
Corresponding author. Tel.: +1 215 628 7047.
E-mail address: bshook@its.jnj.com (B.C. Shook).
Scheme 1. Synthesis of thieno[2,3-d]pyrimidines via Methods A–C.
0960-894X/$ - see front matter Ó 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.02.078

B. C. Shook et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2688–2691
2689
NH2
N
malononitrile and reacted with elemental sulfur to give the substi-
tuted amino nitriles 8. The amino nitrile 8 was reacted under basic
condition with a variety of aryl and heteroaryl nitriles to provide
the target compounds 9, similarly to that outlined in Scheme 1.
Our previous work showed that 2-pyrimidyl-5-methyl furan
was an ideal group for obtaining good in vitro and in vivo po-
tency.10 Compound 10 was prepared and had modest functional
antagonist activity (109 nM) against human A2A receptors (Table
1).11a,12 We wanted to explore substitution in the 6-position with
various aryl and heteroaryl substituents. Simple phenyl substitu-
tion (11) gave a slight increase in A2A activity, but withdrawing
and donating groups on the phenyl caused decreases in activity.
Moving the aromatic group out afforded the benzyl substituted
compound 12 which was ꢀsixfold more potent for A2A compared
to the corresponding phenyl analog 11. Extending the phenyl ring
by another methylene gave the phenethyl analog 13 that was
much less potent than 12. The benzyl group seemed to provide
the ideal length needed to achieve A2A potency. Like the phenyl
analogs, substitution in the 3- or 4-position of the benzyl group
had a detrimental effect on A2A activity. However, we found that
ortho substituents on the benzyl group gave compounds 14–16
that were quite potent for A2A, and the halogen substituted com-
pounds 15 and 16 were also potent A1 antagonists. The methyl
substituted analog 17 resulted in ꢀfourfold decrease in A2A po-
tency compared to 12. Replacing the methyl on the furan with
ethyl (18), i-propyl (19), cyclopropyl (20), or cyano (21) resulted
in compounds which had decreased activity compared to 12. How-
ever, the gem-difluoromethyl compounds 22 and 23 maintained
good A2A activity. The methyl furan analogs 12 and 14–16 provided
a baseline in which we could evaluate the viability of other five-
membered heterocycles.
NC
CN
R1-CN
R3
R2
R3
R2
CN
R3
O
Et3N
Sulfur
(27-52%)
tBuOK (0.2 eq)
R2
H
dioxane, 100 °C
(48-88%)
R1
S
S
N
NH2
7
8
9
R3 = H or Me; (Method D)
Scheme 2. Synthesis of thieno[2,3-d]pyrimidines via Method D.
and benzyl substituted thieno[2,3-d]pyrimidines that are potent
adenosine A2A antagonists having varying degrees of selectivity
versus adenosine A1 receptors.
The compounds were synthesized in various ways depending on
the substitution of the target compound (Schemes 1 and 2). The
commercially available amino nitrile 1 was reacted with various
aryl and heteroaryl nitriles by heating with a catalytic amount of
t-BuOK in dioxane (Scheme 1).10,11 The resulting aminopyrimidine
2 was reacted with N-bromosuccinimide (NBS) to give the corre-
sponding bromide 3. The bromide was then reacted under Suzuki
(Method A) or Negishi (Method B) conditions to give the target
compounds 4. The bromide 3 was also reacted, following Method
C, to provide the olefinic compounds 5. Reacting 5 with the trimeth-
ylsulfoxonium ylide afforded the corresponding cyclopropane com-
pounds 6.
Alternatively, target compounds 9 were prepared according to
Method D (Scheme 2). The aldehyde 7 was condensed with
Table 1
In vitro activity for Human A2A and A1 functional assays; substituted furan derivatives
NH2
N
R1
Table 2
S
N
R2
In vitro activity for Human A2A and A1 functional assays; five-membered heterocycles
Compound R1
R2
A2A Ki
(nM)
A1 Ki
(nM)
Method
NH2
N
R1
O
O
O
O
10
11
12
H
109
67.3
11.1
493
520
298
—
A
B
S
N
R2
Ph
Bn
Compound
R1
R2
A2A Ki (nM)
A1 Ki (nM)
Method
D
O
O
24
Bn
683
2954
13
14
15
16
203
11.0
2.9
2784
122
B
B
D
D
25
26
27
Bn
Bn
Bn
4.7
9.0
310
D
D
B
N
O
O
O
O
O
106
N
OMe
Cl
N
16.1
1491
26.1
<16
S
N
28
29
30
13.2
6.9
186
49.6
28.6
B
3.7
S
N
OMe
Cl
F
D
D
S
N
17
40.1
439
D
O
O
9.7
18
19
Bn
Bn
49.0
168
223
283
B
S
N
F
D
31
32
33
100
31.0
107
1416
>594
>610
D
C
C
S
N
S
O
O
20
21
Bn
Bn
263
304
861
D
D
CN
F
N
S
3163
O
O
22
23
Bn
25.2
13.1
2998
516
B
B
N
F
34
35
Bn
Bn
57.0
539
263
D
D
S
N
F
N
F
2332
OMe

2690
B. C. Shook et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2688–2691
Table 3
Table 4
In vitro activity for Human A2A and A1 functional assays; substituted aryls
In vitro activity for Human A2A and A1 functional assays; six-membered heterocycles
NH2
NH2
N
N
R1
R1
R2
R2
S
S
N
N
Compound R1
R2
A2A Ki
(nM)
A1 Ki
(nM)
Method
Compound
R1
R2
A2A Ki (nM)
594
A1 Ki (nM)
3713
Method
N
N
50
51
Bn
Bn
B
B
36
37
38
39
40
H
500
—
—
A
B
B
B
N
N
215
1418
Ph
Bn
207
237
151
—
52
53
Bn
Bn
67.3
2.4
812
B
29.8
>1000
11.4
91.0
D
N
N
CN
CN
54
55
56
Bn
Bn
Bn
17.1
26.0
173
618
D
D
D
Bn
222
N
N
N
CN
Cl
>1000
>1000
41
42
43
6.2
1.4
7.3
129
24.8
<16
B
OMe
Cl
CN
CN
CN
D
D
CN
57
Bn
>168
1159
D
N
F
44
45
46
340
63.4
222
1772
>1000
>610
D
C
C
CN
Table 5
Mouse catalepsy results of select A2A antagonists
Compound Mouse catalepsy, single oral
dose
Mouse catalepsy ED50 (mg/kg,
p.o.)
CN
F
12
14
25
26
27
40
Active at 10 mg/kg
Active at 10 mg/kg
Not active at 3 mg/kg
Not active at 3 mg/kg
Not active at 10 mg/kg
Active at 3 mg/kg
2.2
0.8
—
—
—
47
48
49
Bn
Bn
Bn
42.0
1016
118
123
B
B
B
CF3
>1000
266
<1.0
OMe
resulted in decreased activity. Replacement of the cyano with fluo-
rine (47) moderately decreased activity, but trifluoromethyl (48)
and methoxy (49) resulted in a substantial decrease of activity.
We then looked at six-membered heterocycles and their possi-
ble utility in replacing the furan group (Table 4). The 2-, 3-, and 4-
pyridyl (50–52) substituted compounds were less potent than the
corresponding phenyl 38 and 3-cyanophenyl 40 compounds. Inter-
estingly the pyrazine 53 was very potent for A2A and the pyrimi-
dine 54 maintained good activity. The 3-cyano-2-pyridine 55 was
more than 20-fold more potent than the 2-pyridine 50. However
the chloride 56 and 5-cyano-3-pyridine 57 analogs were signifi-
cantly less potent.
Several of the compounds were tested for oral activity in the
haloperidol-induced catalepsy model in mice (Table 5).13 This is
a common animal model used in Parkinson’s programs because it
mimics the muscular stiffness seen in patients, has fairly high
throughput, and requires little compound. Compounds were typi-
cally tested as a single oral dose of 3 or 10 mg/kg in mice and eval-
uated for the ability to reverse cataleptic activity. A positive or
‘active’ result corresponds to at least 50% reversal of cataleptic
activity compared to vehicle. In general, catalepsy was measured
1 h after drug administration.14 Compounds 12 and 14 were active
at 10 mg/kg and 40 was active at 3 mg/kg, but 25–27 were not ac-
tive despite having similar in vitro potency for A2A. We previously
reported that methylfuran and 3-cyanophenyl substituents were
ideal for achieving in vivo activity for aminomethyl substituted thi-
eno[2,3-d]pyrimidines, and is further supported with the benzyl
The tetrahydrofuran derivative 24 also resulted in decreased
activity (Table 2). Replacing the furan with the 2- or 5-substituted
oxazoles 25 and 26, respectively, gave compounds that were po-
tent A2A antagonists. Likewise, the 4-methylthiazoles 27–30
showed excellent A2A activity while 29 and 30 also had increased
A1 activity which is consistent to what was seen with 15 and 16.
Substitution at the benzyl position (31–33) had decreased activity
analogous to the decrease with 17. The 2-methylthiazole 34 was
ꢀthreefold less potent than the corresponding 4-methylthiazole
27, and the imidazole 35 had significantly lower potency at both
A2A and A1.
We wanted to explore how aryl groups would influence A2A
activity of thienopyrimidine system (Table 3). The phenyl analog
36 had ꢀfivefold less activity compared to the furan derivative
10, which is consistent with what we had seen previously. Prepa-
ration of the phenyl (37), benzyl (38), and phenethyl (39) deriva-
tives revealed the same trend in activity as seen with the
methylfuran analogs 11–13, and supported that the benzyl group
provided the ideal length to achieve A2A activity. From our previous
work we found that the 3-cyanophenyl was an acceptable replace-
ment for the furan moiety to achieve good in vitro and in vivo
activity. The benzyl substituted compounds 40–43 were potent
A2A antagonists, and the halogenated compounds 42 and 43 had
increased potency for A1, analogous to that seen with the furan
and 4-methylthiazoles analogs. Substituting the benzyl position

B. C. Shook et al. / Bioorg. Med. Chem. Lett. 23 (2013) 2688–2691
2691
Table 6
Rat PK and tissue distribution of 12
Mouse
IV/oral dose (mg/kg)
CL (mL/min/kg)
Oral t1/2 (h)
Oral plasma Cmax (nM)
F (%)
Brain/plasma 1 h
12
Rat
12
5/10
69
0.3
Oral t1/2 (h)
0.7
5009
5
F (%)
83
—
IV/oral dose (mg/kg)
2/10
CL (mL/min/kg)
65
Oral plasma Cmax (nM)
2978
Brain/plasma 1 h
1.2
2. (a) Muller, C. E.; Scior, T. Pharm. Acta Helv. 1993, 68, 77; (b) DeNinno, M. P.
Annu. Rep. Med. Chem. 1998, 33, 111; (c) Pires, V. A.; Pamplona, F. A.; Pandolfon,
P.; Fernandes, D.; Prediger, R. D. S.; Takahashi, R. N. Behav. Pharmacol. 2009, 20,
134; (d) Davidson, D. A.; Weidley, E. Life Sci. 1976, 18, 1279.
3. Fredholm, B. B.; Abbraccio, M. P.; Burnstock, G.; Daly, J. W.; Harden, T. K.;
Jacobson, K. A.; Leff, P.; Williams, M. Pharmacol. Rev. 1994, 46, 143.
4. Dunwiddie, T. V.; Masino, S. A. Annu. Rev. Neurosci. 2001, 24, 31.
5. (a) Shook, B. C.; Jackson, P. F. ACS Chem. Neurosci. 2011, 555; (b) Shah, U.;
Hodgson, R. Curr. Opin. Drug Discov. Devel. 2010, 13, 466; (c) Szabo, N.; Kincses,
Z. T.; Vecsei, L. Expert Opin. Drug Metab. Toxicol. 2011, 7, 441.
6. Ikeda, K.; Kurokawa, M.; Aoyama, S.; Kuwana, Y. J. Neurochem. 2002, 80, 262.
7. Hodgson, R. A.; Bertorelli, R.; Varty, G. B.; Lachowicz, J. E.; Forlani, A.; Fredduzzi,
S.; Cohen-Williams, M. E.; Higgins, G. A.; Impagnatiello, F.; Nicolussi, E.; Parra,
L. E.; Foster, C.; Zhai, Y.; Neustadt, B. R.; Stamford, A. W.; Parker, E. M.; Reggiani,
A.; Hunter, J. C. J. Pharmacol. Exp. Ther. 2009, 330, 294.
8. (a) Shook, B. C.; Rassnick, S.; Wallace, N.; Crooke, J.; Ault, M.; Chakravarty, D.;
Barbay, J. K.; Wang, A.; Leonard, K.; Alford, V.; Powell, M. T.; Scannevin, R. H.;
Carroll, K.; Lampron, L.; Westover, L.; Russell, R.; Branum, S.; Wells, K.; Damon,
S.; Youells, S.; Li, X.; Beauchamp, D. A.; Rhodes, K.; Jackson, P. F. J. Med. Chem.
2012, 55, 1402; (b) Shook, B. C.; Rassnick, S.; Osborne, M. C.; Davis, S.;
Westover, L.; Boulet, J.; Hall, D.; Rupert, K. C.; Heintzelman, G. R.; Hansen, K.;
Chakravarty, D.; Bullington, J. L.; Russell, R.; Branum, S.; Wells, K. M.; Damon,
S.; Youells, S.; Li, X.; Beauchamp, D. A.; Palmer, D.; Reyes, M.; Demarest, K.;
Tang, Y.-T.; Rhodes, K.; Jackson, P. F. J. Med. Chem. 2010, 53, 8104.
9. Mihara, T.; Mihara, K.; Yarimizu, J.; Mitani, Y.; Matsuda, R.; Yamamoto, H.; Aoki,
S.; Akahane, A.; Iwashita, A.; Matsuoka, N. J. Pharmacol. Exp. Ther. 2007, 323,
708.
10. Shook, B. C.; Chakravarty, D.; Barbay, J. K.; Wang, W.; Leonard, K.; Alford, V.;
Powell, M.; Beauchamp, D. A.; Rassnick, S.; Scannevin, R.; Carroll, K.; Wallace,
N.; Crooke, J.; Ault, M.; Lampron, L.; Westover, L.; Rhodes, K.; Jackson, P. F. Med.
Chem. Commun. 2011, 2, 950.
11. (a) J. K. Barbay, D. Chakravarty, B. C. Shook, Wang, A. WO 2010045006.; (b)
Seijas, J. A.; Vazquez-Tato, M. P.; Martinez, M. M. Tetrahedron Lett. 2000, 41,
2215.
12. Ki’s are reported as an average of at least two assay runs. For the functional
assay protocol please refer to Ref. 11a.
13. (a) Zarrindast, M. R.; Modabber, M.; Sabetkasai, M. Pyschopharamcology 1993,
113, 257; (b) Schwarzschild, M. A.; Agnati, L.; Fuxe, K.; Chen, J. F.; Morelli, M.
Trends Neurosci. 2006, 29, 647.
substituted compounds in Table 5. Dose–response studies were
carried out with active compounds and revealed that 12, 14, and
40 were all quite potent in reversing catalepsy in mice. We did
not investigate 25–27 further with pharmacokinetic (PK) or tissue
distribution studies, but in general 25–27 had moderate metabolic
stability and similar protein binding properties compared to the
three active compounds.
Compound 12 showed good plasma levels (Cmax = 5.0 lM) in
mice but suffered from a very short half-life and poor bioavailablity
(Table 6). A duration of action in the mouse catalepsy model
showed that the compound was only able to reverse catalepsy at
the 0.5 and 1 h time points, but not at 2 or 4 h, confirming the re-
sults obtained in the PK study. Further characterization of 12 in a
rat PK/tissue distribution study showed that the compound had
good oral exposure (Cmax ꢀ 3
lM) and good bioavailability, but suf-
fered from high clearance and a short half-life (Table 6). The com-
pound also had a brain to plasma ratio of 1.2 at 1 h post-dosing.
Compounds 14 and 40 had similar liabilities in both mice and rats
including high clearance, short half-life, and short duration of ac-
tion in mouse catalepsy.
In summary, we have shown that benzyl substituted thieno[2,3-
d]pyrimidines are potent A2A antagonists having good in vivo effi-
cacy in the mouse catalepsy model, but in general these com-
pounds suffer from a short half-life, limiting their use as a
potential therapeutic for PD. The benzyl thieno[2,3-d]pyrimidines
had good in vitro potency when substituted with methylfuran
and 3-cyanophenyl as previously reported, but several new 5-
and six-membered heterocycles (25–27 and 53–54) have been
identified that maintain in vitro potency.
14. A 300 mg/kg dose of L-DOPA was used as the positive control in the mouse
catalepsy studies.
References and notes
1. Lozano, A. M.; Lang, A. E.; Hutchison, W. D.; Dostrovsky, J. O. Curr. Opin.
Neurobiol. 1998, 8, 783.