Hit-to-lead optimization of a series of carboxamides of ethyl 2-amino-4-phenylthiazole-5-carboxylates as novel adenosine A2A receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2010.06.138

Herein we describe the discovery of a series of novel adenosine A _2A receptor antagonists. A successful hit-to-lead optimization of an HTS hit led to replacement of a metabolically labile ester moiety with a heteroaromatic group. A compound from the series, (cyclopropanecarboxylic acid [5-(5-methyl-[1,2,4]oxadiazol-3-yl)-4-phenyl-thiazol-2-yl]-amide, compound 13), was shown to be effective in reversing haloperidol-induced hypolocomotion, a model of motor dysfunction in Parkinson's Disease.

Full text of DOI:10.1016/j.bmcl.2010.06.138

Bioorganic & Medicinal Chemistry Letters 20 (2010) 5241–5244
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Hit-to-lead optimization of a series of carboxamides of ethyl
2-amino-4-phenylthiazole-5-carboxylates as novel adenosine
A2A receptor antagonists
a
a
b
b
Anette Graven Sams a, , Gitte Kobberøe Mikkelsen , Mogens Larsen , Lars Torup , Lise Tøttrup Brennum ,
*
Tenna Juul Schrøder c, Benny Bang-Andersen a
a Medicinal Chemistry Research, H. Lundbeck A/S, Ottiliavej 9, 2500 Valby, Denmark
b In vivo Neuropharmacology, H. Lundbeck A/S, Ottiliavej 9, 2500 Valby, Denmark
c Molecular Pharmacology, H. Lundbeck A/S, Ottiliavej 9, 2500 Valby, Denmark
a r t i c l e i n f o
a b s t r a c t
Article history:
Herein we describe the discovery of a series of novel adenosine A2A receptor antagonists. A successful hit-
to-lead optimization of an HTS hit led to replacement of a metabolically labile ester moiety with a het-
eroaromatic group. A compound from the series, (cyclopropanecarboxylic acid [5-(5-methyl-[1,2,4]oxa-
diazol-3-yl)-4-phenyl-thiazol-2-yl]-amide, compound 13), was shown to be effective in reversing
haloperidol-induced hypolocomotion, a model of motor dysfunction in Parkinson’s Disease.
Ó 2010 Elsevier Ltd. All rights reserved.
Received 11 June 2010
Revised 25 June 2010
Accepted 29 June 2010
Available online 24 July 2010
Keywords:
A2A
Adenosine
Antagonist
Parkinson’s Disease
Adenosine (A) is an important neuromodulator, and its action is
mediated via specific receptors, which belong to the family of G-
protein coupled receptors. Four adenosine receptors have been
cloned and characterized, A1, A2A, A2B and A3.1 The main intracellu-
lar signaling pathways involve the formation of cAMP, and signal-
ing through adenosine A2A and A2B receptors leads to activation of
adenylate cyclase, whereas signaling through adenosine A1 and
adenosine A3 receptors leads to inhibition of adenylate cyclase.2
The adenosine A2A receptor is highly expressed in the striatum, nu-
cleus accumbens, and the olfactory tubercle in humans, while low
levels are found in other brain areas. The adenosine A2A receptor is
co-expressed with dopamine D2 receptors in striatum and is in-
volved in the regulation of functional activity of dopamine D2
receptors, and heterodimerization of A2A and D2 receptor subtypes
inhibit dopamine D2 receptor function.3,4 Adenosine A2A receptors
have also been shown to modulate the release of GABA in the stri-
atum. By reducing the GABA output, adenosine A2A receptor antag-
onism helps counteract striatal dopamine depletion and restore
normal function in the basal ganglia. Therefore, A2A receptor antag-
onists may have clinical utility in the treatment of Parkinson’s Dis-
ease (PD),3 and extensive work has been done over the past
decades to discover potent and selective adenosine A2A receptor
antagonists.5–7 Thus, istradefylline (KW-6002, 1)8,9 and prelade-
nant (SCH-420814, 2)10 are currently undergoing clinical trials
for the treatment of PD.
* Corresponding author. Current address: LEO Pharma A/S, Industriparken 55,
2750 Ballerup, Denmark.
Chart 1. Structures and hA2A and hA1 binding affinities of istradefylline (KW-6002,
1), preladenant (SCH-420814, 2) and HTS hit 3.
E-mail address: anette.sams@leo-pharma.com (A.G. Sams).
0960-894X/$ - see front matter Ó 2010 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2010.06.138

5242
A. G. Sams et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5241–5244
(Chart 1). Compound 3 was the most potent among its analogues,
binding to the A2A receptor with a Ki value of 13 nM. It had a mod-
erate selectivity toward the hA1 receptor, as measured by displace-
ment of the radioligand [3H]DPCPX, in membranes from Chinese
hamster ovary (CHO) cells stably expressing hA1 receptors. For
comparison, clinically advanced istradefylline (1), currently in
phase 3 clinical trials, showed a weaker hA2A binding (Ki of
72 nM) and a Ki of 3000 nM for hA1 receptors. Also, preladenant
(2), currently in phase 2 clinical trials, displayed a Ki of 5 nM for
hA2A receptors and an inhibition of [3H]DPCPX binding of 30% at
2000 nM, indicating a Ki >1000 nM for hA1 receptors.
The efficacy of 3 as an antagonist at the A2A receptor was veri-
fied in a functional cAMP-based assay using AlphaScreenÒ technol-
ogy (PerkinElmer) with CHO cells expressing the human A2A
receptors, wherein 3 antagonised the effect of the A2A receptor ago-
nist N-ethylcarboxamidoadenosine (NECA).
The series of hits represented by 3 constitute a novel struc-
tural class of A2A receptor ligands. Previously described A2A
receptor antagonists are often classified in two major classes;
(1) xanthine like compounds, such as istradefylline (1), and (2)
the so-called non-xanthine type compounds. Members of the lat-
ter classification comprise a structurally diverse group of chem-
otypes.5–7 Interestingly, a report on the discovery of a series of
2-amino-5-benzoyl-4-(2-furyl)thiazoles as A2A receptor antago-
nists, which are also based on the 2-aminothiazole core, recently
appeared.11
Scheme 1. Reagents: (i) R2COCl, pyridine, 1,2-dichloroethane.
In our efforts to identify novel A2A receptor antagonists, a
screening campaign of a collection of approximately 200,000 com-
pounds was conducted, using a binding assay with [3H] ZM241385
as radioligand and membranes from a commercially available cell
line stably expressing rat A2A receptors (MCL-511, PerkinElmer).
Among the hits was a series of potent carboxamides of ethyl 2-
amino-4-phenylthiazole-5-carboxylate, exemplified by ethyl 2-
The carboxylic acid ester functionality of 3 made it impossible
to test the compound in vivo, as the plasma half-life is extremely
[(furan-2-carbonyl)-amino]-4-phenylthiazole-5-carboxylate
(3)
Scheme 2. Reagents and conditions: (i) LDA, EtI, THF, ꢀ78 °C, 83%; (ii) SO2Cl2, CH3Cl, 5 °C-reflux; (iii) thiourea, EtOH, reflux; (iv) I2, thiourea, neat, 100 °C, 60%; (v) Boc2O, DMF,
DMAP, 0–20 °C, 58%; (vi) n-BuLi, ClCH2OCH2CH3, THF, ꢀ78 °C, 33%; (vii) 2 M HCl/diethyl ether, 59%; (viii) EDC, DIPEA, EtNH2, HCl, DMF, 1,2-DCE, 42%; (ix) furan-2-carbonyl
chloride, NEt3, 1,2-DCE, 73%; (x) 2 M NaOH, THF, 60%.

A. G. Sams et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5241–5244
5243
short. Consequently, a hit-to-lead optimization effort was under-
taken with the aim of identifying a suitable replacement for the
carboxylic acid ester functionality. A parallel synthesis protocol
was used to rapidly explore the structure–activity relationship
(SAR) of 3, as outlined in Scheme 1. Hence, 4-phenyl-thiazol-2-
ylamines, carrying various substituents at the thiazole 5-position
(4a–i), were acylated with carboxylic acid chlorides in 1,2-dichlo-
roethane (1,2-DCE) in the presence of pyridine. The compounds
were purified by preparative HPLC with MS detection. Thus, com-
pounds 6–32 were prepared in this manner, whereas compound
5 was prepared from 3 by simple saponification, as shown in
Scheme 2. Compound 3 was obtained by acylation of commercially
available ethyl 2-amino-4-phenylthiazole-5-carboxylate with fur-
an-2-carboxylic acid chloride in the presence of pyridine. The
intermediate 4d was commercially available. The intermediates
4f–i were synthesised as described in Ref. 12. The preparation of
intermediates 4a–c,e is outlined in Scheme 2. Thus, treatment of
benzoylacetone with lithium diisopropylamide (LDA) followed by
ethyl iodide in THF at low temperature, yielded 1-phenyl-hex-
ane-1,3-dione in 83% yield; which was chlorinated with SO2Cl2
and then treated with thiourea in refluxing ethanol to afford cycli-
zation and dehydration to yield intermediate 4a, together with the
other isomer 4a0 in a 1:1 ratio. Separation of the isomers 4a and 4a0
proved extremely difficult, and 2% of 4a was obtained by prepara-
tive HPLC purification. For the preparation of intermediate 4b, 2-
amino-4-phenylthiazole was initially obtained in 60% yield from
acetophenone, by treatment with thiourea and iodine. Following
Boc-protection in 58% yield, the thiazole 5-position was deproto-
nated with n-BuLi at low temperature, and treatment with chloro-
methyl-ethyl ether (33%) followed by Boc-deprotection in 59%
yield, afforded the desired product 4b. The intermediate 4c was
prepared in 42% yield by a simple amide coupling reaction be-
tween commercially available 2-amino-4-phenyl-5-carboxylic acid
and ethylamine hydrochloride, in the presence of N-ethyl-N0-
(dimethylaminopropyl)carbodiimide hydrochloride (EDC) and
diisopropyl ethyl amine (DIPEA). The intermediate 4e was pre-
pared in 62% yield from benzoylacetonitrile, by chlorination with
SO2Cl2 followed by treatment with thiourea in refluxing ethanol.
The affinities of compounds 3–32 for hA2A and hA1 receptors are
summarized in Table 1, and expressed as Ki values (nM) or as %
inhibition of radioligand binding at 10 lM test concentration. First,
the corresponding thiazole-5-carboxylic acid analogue to 3 was
shown to be inactive (5). We then investigated the relative contri-
bution of either of the oxygen atoms of the R1 ester group of 3.
Thus, the ethyl ester was first replaced with butan-1-one as R1 sub-
stituent, which led to a slight loss of hA2A receptor affinity and a
drop in selectivity versus the hA1 receptor (6), whereas replacing
the ethyl ester by a ethyl-methylene ether as the R1 substituent
gave rise to a compound 7 which was selective for the hA1 recep-
tor, and with 10-fold less affinity for the hA2A receptor when com-
pared to 3. Hence, the carbonyl oxygen appeared to make the most
important contribution to the affinity of 3 toward the hA2A recep-
tor. Replacing the ethyl ester of 3 with ethyl amide or ethyl sulfone
at the R1 position led to a complete loss of affinity at hA2A receptors
(8, 9), and replacement by a cyano group led to a substantial loss in
hA2A receptor affinity (10). We also examined the effect of replac-
Table 1
Chemical structures and binding affinitiesa at hA2A and hA1 receptors of carboxamides of 2-amino-4-phenyl-thiazoles
N
H
N
R1
S
R2
O
Compound
R1
R2
hA2A
hA1
3
5
6
7
8
CO2Et
2-Furanyl
2-Furanyl
2-Furanyl
2-Furanyl
2-Furanyl
2-Furanyl
2-Furanyl
2-Furanyl
Methyl
13
0%
62
190
37%
25%
290
700
190
28
95
COOH
n.d.b
90
C(O)nPr
CH2OCH2CH3
CONHEt
SO2Et
CN
92
n.d.
n.d.
n.d.
780
76%
150
82
45%
260
230
310
77
300
1700
270
370
41
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
5-Methyl-[1,2,4]oxadiazole-3-yl
Cyclopropyl
Isobutyl
Phenyl
33
83
25
Benzyl
[1,2,4]Oxadiazole-3-yl
2-Furanyl
Cyclopropyl
Isobutyl
Phenyl
Benzyl
2-Furanyl
Methyl
Cyclopropyl
Isobutyl
Phenyl
Benzyl
2-Furanyl
Cyclopropyl
Phenyl
Benzyl
3,4-Dimethoxy-benzyl
410
120
340
330
710
200
110
12
48
190
57
94
77
3-Methyl-[1,2,4]oxadiazole-5-yl
27
180
340
650
210
65%
1200
1600
2-Ethyl-2H-tetrazole-5-yl
220
460
45
a
Expressed as Ki values (nM) or displacement percentage of radioligand at 10 lM test concentration were indicated.
Not determined.
b

5244
A. G. Sams et al. / Bioorg. Med. Chem. Lett. 20 (2010) 5241–5244
We early on selected compound 13 as a suitable prototype com-
1000
750
500
250
0
pound to evaluate the in vivo efficacy of this structurally novel ser-
ies of A2A receptor antagonists, and tested the compound in a
mouse haloperidol-induced hypolocomotion model of Parkinson’s
Disease. In this model, a hypolocomotive state is induced in mice
by pre-treatment with the D2 antagonist haloperidol. Compound
13 dose dependently and fully reversed the hypolocomotive state
after po administration, with an ED50 value of 7 mg/kg (Fig. 1).
For comparison, the two reference compounds istradefylline (1)
and preladenant (2) reversed the hypolocomotive state in the same
model with ED50 values of 0.13 mg/kg and 0.4 mg/kg, respectively.
To verify that the observed behavior was correlated to blockade of
A2A receptors, an in vivo binding experiment was performed. The
A2A selective ligand [3H]SCH-442416 was dosed to mice iv and
the displacement of the radioligand after po administration of
compound 13 was measured. Administration of compound 13
dose-dependently displaced [3H]SCH-442416 with an ED50 of
5.8 mg/kg (Fig. 2). This value corresponds well to the ED50 ob-
served in haloperidol-induced hypolocomotion, suggesting block-
ade of A2A receptors to be the mechanism of action of the
reversal of the haloperidol-induced hypolocomotive state.
Vehicle Vehicle
5
10
20
40
0.63mg/kg haloperidol + 13
Figure 1. Dose-response of compound 13 in a mouse haloperidol-induced hypol-
ocomotion model of Parkinson’s Disease. ED50 = 7 mg/kg
In conclusion, a hit-to-lead optimization of a screening hit 3 re-
sulted in the identification of a novel series of A2A receptor antag-
onists with in vivo efficacy. A compound 13 from the series was
selected and was shown to have effect in a model of PD, and to dis-
place a selective A2A receptor radioligand in an in vivo binding
experiment with a similar ED50 value. The optimization of the hit
3 into a lead series involved the identification of heteroaromatic
replacements for the ester functionality of 3. Thus, incorporation
of different heteroaromatics as the R1 substituent gave compounds
with affinities at the hA2A receptor in the same range as the origi-
nal hit. Interestingly, the observed SAR with respect to the R2 sub-
stituent in the series incorporating heteroaromatic R1 substituents
was not parallel to that in the original series, wherein R1 = ethyl es-
ter. As a general tendency, the compounds display low selectivity
toward the hA1 receptor, however, it was possible to identify com-
pounds with selectivity over the hA1 receptor in the same range as
istradefylline (1).
Figure 2. Dose response of displacement of [3H]SCH-442416 by compound 13
in vivo in mice. ED50 = 5.8 mg/kg
ing the ethyl ester with a range of five-membered heteroaromatic
groups as R1 substituents. Thus, potent compounds were obtained
with R1 being 5-methyl-[1,2,4]oxadiazole-3-yl or the isomeric 3-
methyl-[1,2,4]oxadiazole-5-yl, and for these R1 substituents,
R2 = cyclopropyl (13, 24), isobutyl (14), or benzyl (16) yielded com-
pounds with hA2A receptor affinities in the range of 3. Also, in the
case of 13 and 16, the selectivity toward the hA1 receptor was com-
parable to that of 3. In contrast, the removal of the methyl substi-
tuent from the [1,2,4]oxadiazole-3-yl core of these compounds, led
to a consistent drop in the hA2A receptor affinities (17–21). Using
2-ethyl-2H-tetrazole-5-yl as the R1 substituent generally led to less
potent compounds, however, a 3,4-dimethoxybenzyl group as the
R2 substituent led to a compound with just fivefold weaker hA2A
receptor affinity compared to 3, yet with an improved selectivity
versus the hA1 receptor (32). Hence, 32 show an in vitro receptor
profile similar to that of istradefylline (1).
Supplementary data
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmcl.2010.06.138.
References and notes
1. Fredholm, B.-B.; Abbracchio, M. P.; Burnstock, G.; Daly, J. W.; Harden, T. K.;
Jacobson, K. A.; Leff, P.; Williams, M. Pharmacol. Rev. 1994, 46, 143.
2. Olah, M. E.; Stiles, G. L. Pharmacol. Ther. 2000, 85, 55.
3. Xu, K.; Bastia, E.; Schwarzschild, M. Pharmacol. Ther. 2005, 105, 267.
4. Simola, N.; Morelli, M.; Pinna, A. Curr. Pharm. Des. 2008, 14, 1475.
5. Baraldi, P. G.; Tabrizi, M. A.; Gessi, S.; Borea, A. P. Chem. Rev. 2008, 108, 238.
6. Yuzlenko, O.; Kiec-Kononowicz, K. Curr. Med. Chem. 2006, 13, 3609.
7. Moro, S.; Gao, Z.-G.; Jacobson, K. A.; Spalluto, G. Med. Res. Rev. 2006, 26, 131.
8. Jenner, P. Expert Opin. Invest. Drugs 2005, 14, 729.
9. Hockemeyer, J.; Burbiel, J. C.; Müller, C. E. J. Org. Chem. 2004, 69, 3308.
10. Neustadt, B. R.; Hao, J.; Lindo, N.; Greenlee, W. J.; Stamford, A. W.; Tulshian, D.;
Ongini, E.; Hunter, J.; Monopoli, A.; Bertorelli, R.; Foster, C.; Arik, L.; Lachowicz,
J.; Ng, K.; Feng, K.-I. Bioorg. Med. Chem. Lett. 2006, 17, 1376.
11. Cole, A. G.; Stauffer, T. M.; Rokosz, L. L.; Metzger, A.; Dillard, L. W.; Zeng, W.;
Henderson, I. Bioorg. Med. Chem. Lett. 2009, 19, 378.
Interestingly, the SAR for the R2 substituent was not parallel for
the compounds with a heteroaromatic group as the R1 substituent
compared to those with ethyl ester as the R1 substituent. Thus, in
the heteroaromatic series, R2 = furan-2-yl did not yield the most
potent compounds (3 vs 11, 17, 22), instead, R2 = cyclopropyl gen-
erally gave the most potent compounds in series defined by het-
eroaromatic R1 substituents.
12. Larsen, M, Sams, A. G., Mikkelsen, G. K., Bang-Andersen, B. WO2006/032273,
2006.