Synthesis of [1,2,4]triazolo[1,5-a]pyrazines as adenosine A2A receptor antagonists

Article abstract of DOI:10.1016/j.bmcl.2005.07.052

Potent and selective antagonists of the adenosine A_2A receptor often contain a nitrogen-rich fused-ring heterocyclic core. Replacement of the core with an isomeric ring system has previously been shown to improve target affinity, selectivity, and in vivo activity. This paper describes the preparation, by a novel route, of A_2A receptor antagonists containing the [1,2,4]triazolo[1,5-a]pyrazine nucleus, which is isomeric with the [1,2,4]triazolo[1,5-c]pyrimidine core of a series of known A_2A antagonists with in vivo activity in animal models of Parkinson's disease.

Full text of DOI:10.1016/j.bmcl.2005.07.052

Bioorganic & Medicinal Chemistry Letters 15 (2005) 4809–4813
Synthesis of [1,2,4]triazolo[1,5-a]pyrazines as adenosine
A2A receptor antagonists
James E. Dowling,* Jeffrey T. Vessels, Serajul Haque, He Xi Chang, Kurt van Vloten,
Gnanasambandam Kumaravel, Thomas Engber, Xiaowei Jin, Deepali Phadke, Joy Wang,
Eman Ayyub and Russell C. Petter
Biogen Idec, Inc., 14 Cambridge Center, Cambridge, MA 02142, USA
Received 2 May 2005; revised 14 July 2005; accepted 14 July 2005
Available online 8 September 2005
Abstract—Potent and selective antagonists of the adenosine A2A receptor often contain a nitrogen-rich fused-ring heterocyclic core.
Replacement of the core with an isomeric ring system has previously been shown to improve target affinity, selectivity, and in vivo
activity. This paper describes the preparation, by a novel route, of A2A receptor antagonists containing the [1,2,4]triazolo[1,5-a]pyr-
azine nucleus, which is isomeric with the [1,2,4]triazolo[1,5-c]pyrimidine core of a series of known A2A antagonists with in vivo activ-
ity in animal models of ParkinsonÕs disease.
Ó 2005 Elsevier Ltd. All rights reserved.
Membrane-spanning G-protein coupled receptors
(GPCRs) figure prominently in the regulation of a wide
array of signal transduction pathways, and are frequent
and well-validated targets for therapeutic intervention.
Adenosine receptors, a class of GPCRs, are integral
components of signaling cascades that control a range
of physiological and cellular responses.1 Four different
adenosine receptors (A1, A2A, A2B, and A3) have been
biochemically and pharmacologically characterized
and their regulation may offer a means to control
immunological, cardiovascular, renal, or neurological
responses for therapeutic benefit.2 The identification of
small-molecule agonists and antagonists of adenosine
receptors has helped us to elucidate further the biochem-
ical roles of the individual receptor subtypes and evalu-
ate them as targets for drug discovery.3
The locomotor deficits that are characteristic of PD,
such as bradykinesias, tremor, and rigidity, have their
origins in the progressive destruction of dopamine
(DA)-producing neurons in the nigrostriatum. During
the early stages of the disease, symptomatic relief may
be provided by replenishing DA through administration
of L-dopa.7 However, the effectiveness of L-dopa is grad-
ually diminished by the continual depletion of DA
neurons, which results in the eventual manifestation of
motor and psychological side-effects.8
Due, in part, to its interaction with the D2 receptor,
antagonism of the A2A receptor is regarded as the means
to attenuate the motor fluctuations associated with de-
creased responsiveness to L-dopa.9 In an early clinical
demonstration of this effect, administration of low doses
of theophylline led to a significant improvement in
symptoms of patients with PD.10 More recently, the
A2A-selective xanthine derivative KW-6002 (1) has been
found to reverse motor disability in MPTP-lesioned pri-
mates when combined with L-dopa. KW-6002 is current-
ly under clinical evaluation as an antiparkinsonian
agent11 (see Fig. 1).
The A2A receptor is highly expressed in the nigrostria-
tum (basal ganglia) where it is co-localized with dopa-
mine D2 receptors on striatopallidal output neurons.4
Several pharmacological studies suggest that A2A recep-
tor antagonists have potential for use, in combination
with existing therapies,5 in the treatment of ParkinsonÕs
disease (PD) and may also exhibit neuroprotective
effects.6
Non-xanthine antagonists of the A2A receptor often take
the form of nitrogen-containing fused bicyclic or tricyclic
systems that lack the agonism-conferring ribose moiety
of adenosine but feature hydrophobic substituents that
*
Corresponding author. Tel.: +1 617 679 2554; fax: +1 617 679
3635; e-mail: james.dowling@biogenidec.com
0960-894X/$ - see front matter Ó 2005 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2005.07.052

4810
J. E. Dowling et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4809–4813
from 2-aryl substituted bromides (6) by employing a com-
mercially available 2-amino-3,5-dibromopyrazine (7).
O
Me
N
Et
OMe
OMe
N
KW 6002
(1)
N
Although the requisite bromides (6) can be obtained by
N-amination of 7 with mesitylenesulfonyl hydroxyl-
amine (MtsONH2), followed by cyclo-condensation
with the appropriate aldehyde,17 we desired a route that
did not rely on the use of thermally labile aminating re-
agents18 (see Fig. 5).
O
N
Et
Figure 1. Xanthine-derived A2A receptor antagonist (KW-6002).
impart selectivity.12 Compounds of this type, such as
SCH 58261 (2, A2a Ki = 2.3 nM) and ZM241385 (3,
A2a Ki = 0.3 nM) and their homologs, are among the
most selective A2A antagonists prepared to date.13
Despite their promising properties, however, the clinical
application of 2 and 3 and their derivatives has been
curtailed by their poor bioavailability (see Fig. 2).
The synthesis of fused triazoles via oxidative cyclization
of N-2-pyrazinylacetamide, obtained by treatment of 2-
aminopyrazine with AlCl3 in acetonitrile,19 has been
reported.20 In a similar fashion, treatment of a mixture
of 7 and furonitrile in toluene with AlCl3 afforded ami-
dine 8a. Oxidative cyclization of the crude preparation
with Pb(OAc)4 provided 9a in 40% isolated yield. In
the course of determining the scope and limitations of
the new method, it was observed that the conditions em-
ployed in the conversion of 7–8 were ineffective when
applied to other nitriles (Fig. 6).
As part of our ongoing interest in the design and
synthesis of adenosine receptor antagonists, we under-
took a program to develop selective A2A antagonists
for treatment of PD.14 We have recently demonstrat-
ed that the nature of the polynitrogen heterocyclic
core can dramatically influence A2A receptor potency
and selectivity, and compounds having a general
structure 4 (Fig. 3) exhibited good oral activity in
an animal model of PD.15 Other researchers in the
field have reported that the isomeric 5- and 8-ami-
no[1,2,4]triazolo[1,5-a]pyridines act as potent A2A
antagonists when functionalized at the 6- and 7-posi-
tions, respectively.16
Changing the solvent from toluene to 1,2-dichloroeth-
ane resulted in a more efficient conversion in the case
NH2
Br
NH2
N
N
N
N
R2
5
N
N
Br
Br
Based on these observations, the isomeric triazolo[1,5-
a]pyrazine nucleus (5), in which the critical hydrogen
bond donor and acceptor groups of 2, 3, and 4 are pre-
served, was pursued as a template from which to con-
struct a series of A2A antagonists.
7
6
Figure 4. Retrosynthetic approach to [1,2,4]triazolo[1,5-a]pyrazines.
NH2
Br
A retrosynthetic strategy (Fig. 4) was envisaged in which a
series of substituted derivatives (5) could be obtained
NH2
NH2
MtsONH2
R2CHO
N
N
N
N
R2
7
N
N
Br
Br
6
MtsO
NH2
N
Figure 5. Synthetic route to [1,2,4]triazolo-[1,5-a]pyrazines.
N
N
SCH 58261
(2)
N
O
O
N
N
Br
Br
ArCN
AlCl3
H
NH2
N
N
Ar
Pb(OAc)4
N
N
N
N
HO
N
N
Ar
7
ZM 241385
(3)
N
N
N
NH
8a-d
toluene
115 oC
1,2-dichloro-
ethane
115 oC
Br
Br
N
H
N
9a-d
Figure 2. Selective non-xanthine A2A receptor antagonists.
Ar
8 (% yield)
8a (70)
9 (% yield)
9a (40)
2-furanyl
2-thienyl
phenyl
NH2
N
NH2
8b (68)
9b (45)
N
N
N
N
N
N
R2
R2
8c (26)
9c (nd)
9d (46)
N
R1
R1
3-fluorophenyl
8d (30)
4
5
Figure 3. [1,2,4]Triazolo[1,5-c]pyrimidines (4) and isomeric [1,2,4]-
triazolo[1,5-a]pyrazines (5).
Figure 6. Synthesis of [1,2,4]triazolo[1,5-a]pyrazines by oxidative
amidine cyclization.

J. E. Dowling et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4809–4813
4811
of the original example (8a) and enabled the preparation
of several other derivatives (8b–d) in yields ranging from
26% to 70%.21
comparison of the receptor binding data between the
new series and classes of compounds in which the furan
ring has been found to confer A2A receptor affinity and
selectivity (Fig. 7).
The effect of AlCl3 on amidine formation was also stud-
ied and in the case of 8b it was found that one equivalent
of the Lewis acid was optimal; the use of 0.5 or two
equivalents caused a significant reduction in yield. The
resulting amidines (8a,b,d) were then subjected to oxida-
tive cyclization to afford the desired triazolo[1,5-a]pyra-
zines (9a,b,d) in yields ranging from 40% to 46%.22
Suzuki coupling of 10a, obtained by treatment of 9a
with NH3/dioxane, with a series of commercially avail-
able arylboronic acids allowed the rapid assembly of a
series of biaryls (11a–z), which were evaluated for their
ability to bind to rat A2A receptor in a radioligand
binding assay (Table 1).
We employed 2-furyl derivative 9a as a starting material
for our initial series of compounds to allow a direct
In this collection of compounds, those containing
m-substituted phenyl rings exhibited the highest A2A
activity, with Ki values for the best compounds in the
range of 100–500 nM. Bi- and heterocyclic derivatives
(11v–z) exhibited comparable potency.
NH2
NH3
N
N
N
N
Based on the promising in vitro activity of members of
our initial adducts, we chose amide 11t as a starting
point for further elaboration of this series. A collection
of amides (Table 2, 12a–j) was prepared from acid 11u
(and the corresponding p-isomer, not shown) to
9a
10a
N
1,4-dioxane
O
O
Br
NH2
ArB(OH)2, Pd(Ph3P)4
N
N
11a-z
dioxane/2M Na2CO3
100 oC
N
Ar
Table 2. Synthesis of amides 12a–j and their affinities toward rat A2A
and A1 receptors, as measured in a radioligand binding assaya
Figure 7. Synthesis of [1,2,4]triazolo[1,5-a]pyrazines 11a–z by oxida-
tive amidine cyclization.
NH2
amine, HATU
N
N
11u
(or p-isomer)
N
Table 1. A2A receptor binding affinities of Suzuki adducts 11a–z, as
determined in a radioligand binding assaya
iPrNEt, DMF
R1
N
O
Compound
Ar
A2A (rat): Ki (nM)
12a-j
11a
11b
11c
11d
11e
11f
11g
11h
11i
m-iPr-Ph
m-H2N-Ph
p-H2N-Ph
680
630
Compound R1
m or p A2A (rat) A1 (rat)
Ki (nM)
Ki (nM)
>1000
160
75
m-AcNH-Ph
m-Ac-Ph
12a
12b
12c
CONH-iPr
CONMe2
CONEt2
m
m
m
140
70
1
290
120
41
m-NO2-Ph
p-NO2-Ph
460
>1000
140
O
N
12d
12e
12f
12g
12h
m
m
m
p
680
730
240
19
>250
m-Me2N-Ph
p-Me2N-Ph
m-EtO2C-Ph
p-EtO2C-Ph
m-CN-Ph
NBn
N
>1000
370
O
N
11j
3
N
11k
11l
>1000
160
O
N
<250
1800
>250
O
11m
11n
11o
11p
11q
11r
11s
11t
11u
11v
11w
11x
11y
11z
p-CN-Ph
m-F3C-Ph
p-F3C-Ph
>1000
220
O
N
>1000
470
O
m-HOCH2-Ph
p-HOCH2-Ph
m-MeO-Ph
m-MeSO2-Ph
m-H2NCO-Ph
m-HO2C-Ph
3-Pyridyl
OH
OH
O
>1000
270
110
m
180
N
H
O
110
12i
12j
p
p
830
330
36
>1000
460
400
N
H
2-Furanyl
H
N
N
N
1-Naphthyl
8-Quinolyl
5-Pyrimidyl
420
250
>250
O
360
a A2A receptor binding measured as given in Table 1. A1 receptor
containing membranes were prepared from rat cerebral cortex.
[3H]DPCPX was used as radioligand and A1 Ki values were calcu-
lated from binding curves generated in a single experiment from the
mean of three determinations per concentration, with variation in
individual values of <15%. SCH-58261 (2) used as a control (A2A
Ki = 37 nM, A1 Ki = 390 nM).
a A2A receptor enriched membranes were prepared from rat brain
homogenates and employed in binding assays using [3H]ZM-241385
as radioligand and SCH-58261 (A2A Ki = 37 nM) as a control. Ki
values were calculated from binding curves generated in a single
experiment from the mean of three determinations per concentration,
with variation in individual values of <15%.

4812
J. E. Dowling et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4809–4813
determine the effect of further substitution on selectivity
and potency.
exhibit better potency than tertiary amides derived from
cyclic amines (15g–i).
The in vitro binding data (Table 2) indicated that simple
alkyl substitution, as with N,N-diethylamide 12c (A2A
Ki = 1 nM), could enhance the A2A affinity but concom-
itantly conferred significant potency toward the A1
receptor (A1 Ki = 40 nM), relative to monosubstituted
compounds 12a,b.
Compound 15e, which incorporates the amino-ethylphe-
nol unit from ZM241385 (3), exhibited high affinity (A2A
Ki = 1 nM) and impressive selectivity (300-fold) against
the A1 receptor.
In summary, we have described the synthesis of a series
of triazolo[1,5-a]pyrazine-derived A2A receptor antago-
nists by a novel route. The compounds obtained
through this route represent potential leads for the
development of analogs with good in vitro potency
and selectivity. Further details of our efforts to optimize
these properties through side-chain modifications will be
made in due course.
We subsequently prepared a series of compounds lack-
ing the benzene ring to determine if the amide substitu-
ents in Table 2 conferred similar in vitro activity when
directly linked to the heterocyclic core. Palladium-med-
iated carbonylation of bromide 10a was employed to
prepare ester 13 (Fig. 8). Hydrolysis of 13 to acid 14, fol-
lowed by amine coupling, afforded the desired amides
(15a–i).
References and notes
The in vitro binding data for this class of compounds of-
fer a clearer picture of the distinct SAR at the A2A and
A1 receptors (Table 3). Secondary amides having an aro-
matic group at the terminus of an alkyl chain (15b–f)
1. Jacobson, K. A.; van Galen, P. J. M.; Williams, M.
J. Med. Chem. 1992, 35, 407.
2. (a) Fredholm, B. B.; Abbraccio, M. P.; Burnstock, G.;
Daly, G. W.; Harden, T. K.; Jacobson, K. A.; Leff, P.;
Williams, M. Pharmacol. Rev. 1994, 46, 143; (b) Ralevic,
V.; Burnstock, G. Pharmacol. Rev. 1998, 50, 413; (c)
Pd(OAc)2
NH2
Muller, C. E.; Stein, B. Curr. Pharm. Des. 1996, 2, 501.
3. Poulsen, S.-A.; Quinn, R. J. Bioorg. Med. Chem. 1998, 6,
619.
4. Suzuki, F.; Shimada, J. In Adenosine Antagonists and
ParkinsonÕs Disease; Kase, H., Richardson, P. J., Jenner,
P., Eds.; Academic Press: New York, 2000; 31–49.
¨
dppp, MeOH
N
N
CO(g), 14 psi
13
N
10a
(R1 = Me)
DMSO, 80oC
52%
N
R1O2C
O
2M LiOH (aq)
MeOH/THF 70oC
67%
5. (a) Muller, C. E. Drugs Future 2000, 25, 1043; (b)
¨
NH2
N
Kurokawa, M.; Kirk, I. P.; Kirkpatrick, K. A.; Kase,
H.; Richardson, P. J. Br. J. Pharmacol. 1994, 113, 43.
6. Chen, J.-F.; Xu, K.; Petzer, J. P.; Staal, R.; Xu, Y.-H.;
Beilstein, M.; Sonsalla, P. K.; Castagnoli, K.; Castagnoli,
N.; Schwarzschild, M. A. J. Neurosci. 2001, 21, 143.
7. Jankovic, J. J. In ParkinsonÕs Disease and Movement
Disorders; Jankovic, J. J., Tolosa, E., Eds.; Lippincott
Williams & Wilkins: Philadelphia, 2002; 116–151.
8. Obeso, J. A.; Olanow, C. W.; Nutt, J. G. Trends Neurosci.
2000, 23, S2.
9. Richardson, P. J.; Kase, H.; Jenner, P. G. TIPS 1997, 18,
338.
10. Kulisevsky, J.; Barbanoj, M.; Gironell, A.; Antonijoan,
R.; Casas, M.; Pascual-Sedano, B. Clin. Neuropharmacol.
2002, 25, 25.
11. (a) Shimada, J.; Koike, N.; Nonaka, H.; Shiozaki, S.;
Yanagawa, K.; Kanda, T.; Kobayashi, H.; Ichimura, M.;
Nakamura, J.; Kase, H.; Suzuki, F. Bioorg. Med. Chem.
Lett. 1997, 7, 2349; (b) Knutsen, L. J.-S.; Weiss, S. M.
Curr. Opin. Inv. Drugs 2001, 2, 668; (c) Kanda, T.;
Jackson, M.; Smith, L. A.; Pearce, R. K. B.; Nakamura,
J.; Kase, H.; Kuwana, Y.; Jenner, P. G. Ann. Neurol. 1998,
43, 507; (d) Bara-Jimenez, W.; Sherzai, A.; Dimitrova, T.;
Favit, A.; Bibbiani, F.; Gillespie, M.; Morris, M. J.;
Mouradian, M. M.; Chase, T. N. Neurology 2003, 61, 293.
12. Cacciari, B.; Pastorin, G.; Spalluto, G. Curr. Top. Med.
Chem. 2003, 3, 403.
13. (a) Baraldi, P. G.; Cacciari, B.; Spalluto, G.; Villatoro, M. J.
P. I.; Zocchi, C.; Dionisotti, S.; Ongini, E. J. Med. Chem.
1996, 39, 1164; (b) Poucher, S. M.; Keddie, J. R.; Singh, P.;
Stoggall, S. M; Caulkett, P. W. R.; Jones, G.; Collis, M. G.
Br. J. Pharmacol. 1995, 115, 1096; (c) Ongini, E.; Dionisotti,
D.; Gessi, S.; Irenius, E.; Fredholm, B. Naunyn-Schmiede-
bergÕs Arch. Pharmacol. 1999, 359, 7.
HATU, iPr2NEt
N
R2
N
amine, DMF
N
14
(R1 = H)
O
O
15a-i
Figure 8. Synthesis of amides 15a–i by Pd-catalyzed carbonylation.
Table 3. Affinities of 15a–i toward rat A2A and A1 receptors, as
measured in radioligand binding assaysa
Compound
R2
A2A (rat)
Ki (nM)
A1 (rat)
Ki (nM)
15a
15b
HNBu
1
2
120
14
HN
Ph
N
Ph
15d
15e
15f
15g
15h
15I
27
1
160
320
Me
OH
HN
HN
28
240
Ph
>250
>250
>250
>250
>250
>250
N
N
N
O
NBn
N
N
a Refer to Tables 1 and 2 for details regarding rat membrane based
radioligand binding assays.

J. E. Dowling et al. / Bioorg. Med. Chem. Lett. 15 (2005) 4809–4813
4813
14. Vu, C. B.; Pan, D.; Peng, B.; Sha, L.; Kumaravel, G.;
Smits, G.; Jin, X.; Phadke, D.; Engber, T.; Huang, C.;
Reilly, J.; Tam, S.; Petter, R. C. Bioorg. Med. Chem. Lett.
2004, 14, 4831.
15. Vu, C. B.; Peng, B.; Kumaravel, G.; Smits, G.; Jin, X.;
Phadke, D.; Engber, T.; Huang, C.; Reilly, J.; Tam, S.;
Grant, D.; Hetu, G.; Chen, L.; Zhang, J.; Petter, R. C.
J. Med. Chem. 2004, 47, 4291.
carboxamidine (8b). A stirred solution of 3,5-dibromo-
pyrazin-2-ylamine (7, 500 mg, 1.98 mmol), thiophene-2-
carbonitrile (260 lL, 2.04 mmol), and AlCl3 (232 mg,
1.74 mmol) in dichloroethane (5 ml) was heated at
115 °C overnight. The mixture was allowed to cool to rt
and diluted with water (5 mL). After 30 min, the resulting
precipitate was collected and purified by column chroma-
tography (SiO2, THF as eluent) to afford 488 mg (68%) of
8b as a yellow solid, mp 250 °C (dec). 1H NMR (300 MHz,
16. (a) Guba, W.; Nettekoven, M.; Pulmann, B.; Riemer, C.;
¨
Schmitt, S. Bioorg. Med. Chem. Lett. 2004, 14, 3307; (b)
DMSO-d6); d 7.22 (t, J = 3.9 Hz, 1H), 7.83 (dd,
Brodbeck, B.; Pullman, B.; Schmitt, S.; Nettekoven, M.
¨
Tet. Lett. 2003, 44, 1675; (c) Nettekoven, M. Synlett 2001,
12, 1917.
J = 4.8 Hz, 1H), 7.99 (d, J = 4.1 Hz, 1H), 8.49 (s, 1H),
8.88 (br s, 1H); MS: m/z = 361 (MH+).
22. Representative procedure for the synthesis of dibromo-
[1,2,4]triazolo[1,5-a]pyrazines (9): 6,8-dibromo-2-furan-
2-yl-[1,2,4]triazolo[1,5-a]pyrazine (9a). A mixture of com-
pound 8a (47 g, 0.14 mol), Pb(OAc)4 (95% purity, 160 g,
0.34 mol), and toluene (940 ml) was heated at reflux for
2 h. The reaction mixture was allowed to cool to rt and
concentrated in vacuo. The resulting residue was purified
by flash chromatography on silica using EtOAc in hexanes
(1:5 to 1:3) as eluent to afford 9a as a yellow solid (19.2 g,
40%), mp 250 °C (dec). 1H NMR (300 MHz, CDCl3): d
6.60 (dd, 1H), 7.33 (d, 1H), 7.62 (d, 1H), 8.63 (s, 1H); MS:
m/z = 343 (MH+).
17. Yao, G.; Haque, S.; Sha, L.; Kumaravel, G.; Wang, J.;
Engber, T.; Whalley, E. T.; Conlon, P. R.; Chang, H.-X.;
Kiesman, W. F.; Petter, R. C. Bioorg. Med. Chem. Lett.
2005, 15, 511.
18. Eichenberger, T.; Balli, H. Helv. Chim. Acta 1986, 69,
1521.
19. Okamoto, T.; Torigoe, Y.; Sato, M.; Isogai, Y. Chem.
Pharm. Bull. 1968, 16, 1154.
20. Bower, J. D.; Ramage, G. R. J. Chem. Soc. 1957, 4506.
21. Representative procedure for the synthesis of amidine
derivatives (8): N-(3,5-dibromo-pyrazin-2-yl)-thiophene-2-