Novel adenosine A2A receptor ligands: A synthetic, functional and computational investigation of selected literature adenosine A2A receptor antagonists for extending into extracellular space

Article abstract of DOI:10.1016/j.bmcl.2013.03.070

Growing evidence has suggested a role in targeting the adenosine A _2A receptor for the treatment of Parkinson's disease. The literature compounds KW 6002 (2) and ZM 241385 (5) were used as a starting point from which a series of novel ligands targeting the adenosine A_2A receptor were synthesized and tested in a recombinant human adenosine A_2A receptor functional assay. In order to further explore these molecules, we investigated the biological effects of assorted linkers attached to different positions on selected adenosine A_2A receptor antagonists, and assessed their potential binding modes using molecular docking studies. The results suggest that linking from the phenolic oxygen of selected adenosine A_2A receptor antagonists is relatively well tolerated due to the extension towards extracellular space, and leads to the potential of attaching further functionality from this position.

Full text of DOI:10.1016/j.bmcl.2013.03.070

Bioorganic & Medicinal Chemistry Letters xxx (2013) xxx–xxx
Contents lists available at SciVerse ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Novel adenosine A2A receptor ligands: A synthetic, functional and
computational investigation of selected literature adenosine A2A
receptor antagonists for extending into extracellular space
Manuela Jörg a, Jeremy Shonberg a, Frankie S. Mak b, Neil D. Miller b, Elizabeth Yuriev a,
Peter J. Scammells a, , Ben Capuano a,
⇑
⇑
a Medicinal Chemistry, Monash Institute of Pharmaceutical Sciences, 381 Royal Parade, Parkville, Victoria 3052, Australia
b GlaxoSmithKline, GSK R&D China, Singapore Research Centre, 11 Biopolis Way, Helios Bldg #03-01/02, Singapore 138667, Singapore
a r t i c l e i n f o
a b s t r a c t
Article history:
Growing evidence has suggested a role in targeting the adenosine A2A receptor for the treatment of Par-
kinson’s disease. The literature compounds KW 6002 (2) and ZM 241385 (5) were used as a starting point
from which a series of novel ligands targeting the adenosine A2A receptor were synthesized and tested in
a recombinant human adenosine A2A receptor functional assay. In order to further explore these mole-
cules, we investigated the biological effects of assorted linkers attached to different positions on selected
adenosine A2A receptor antagonists, and assessed their potential binding modes using molecular docking
studies. The results suggest that linking from the phenolic oxygen of selected adenosine A2A receptor
antagonists is relatively well tolerated due to the extension towards extracellular space, and leads to
the potential of attaching further functionality from this position.
Received 18 January 2013
Revised 15 March 2013
Accepted 20 March 2013
Available online xxxx
Keywords:
Parkinson’s disease
Adenosine A2A antagonist
ZM 241385
Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved.
KW 6002
Congener
O
N
O
N
Adenosine initiates most of its physiological effects through
the activation of four G protein-coupled receptor (GPCR) sub-
types; A1, A2A, A2B, and A3.1,2 These four receptor subtypes play
an important role in the regulation of a number of central ner-
vous system functions, including pain, cerebral blood flow, basal
ganglia operation, respiration, and sleep.3 The receptors primarily
operate by coupling to the cyclic adenosine monophosphate
(cAMP) second-messenger system, and the adenosine A2A recep-
tor (A2AR) in particular is linked to Gs and Golf proteins. Upon
N
N
O
N
N
N
N
O
O
O
1
2
caffeine
KW 6002
aKi (nM) = 44,000(A1); 23,400(A2A
)
aKi (nM) = 841(A1); 12 (A2A
)
O
N
NH2
H
N
N
N
N
N
N
A
2AR activation, the intracellular levels of cAMP are increased.3
N
O
O
N
N
O
Interestingly, it was found that consuming just one cup of coffee
per day reduces the incidence of Parkinson’s disease by as much
as five-fold;4 postulated to result from the antagonism of the
N
HN
3
4
XAC
NH2
ST 1535
A
2AR by caffeine (1) (Fig. 1).5 Furthermore, co-administration of
aKi (nM) = 71.8(A1); 6.6 (A2A
)
bKi (nM) = 8.3(A1); 5.8(A2A
)
KW 6002 (2) (Fig. 1) with the dopamine D2 receptor agonist apo-
morphine has been shown to prevent the adverse effect of dyski-
nesias observed in monotherapy.6 Similar results were observed
in rats when using the A2AR antagonist ST 1535 (3) (Fig. 1) in
combination with levodopa.7 Therefore, it is postulated that
NH2
N
HO
N
N
O
N
N
N
H
5
caffeine (1) and other
A2AR antagonists have potential as
ZM 241385
aKi (nM) = 774(A1); 1.6 (A2A
)
⇑
Corresponding authors. Tel.: +61 3 9903 9542 (P.J.S.); tel.: +61 3 9903 9556
(B.C.).
Figure 1. Chemical structures of some important adenosine receptor antagonists
and their respective Ki values at the human A1 and A2A receptor in nM. (a) Data from
(cAMP) studies (Müller and Jacobson, 2011);11 (b) Data from (cAMP) studies
(Kecske´s et al., 2011).12
E-mail addresses: peter.scammells@monash.edu (P.J. Scammells), ben.capuano@
monash.edu (B. Capuano).
0960-894X/$ - see front matter Crown Copyright Ó 2013 Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.bmcl.2013.03.070
Please cite this article in press as: Jörg, M.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.03.070

2
M. Jörg et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
bitopic ligand
A
2AR antagonist
homobivalent ligand
homobivalent ligand
homobivalent ligand
"allosteric binding mode"
"orthosteric binding mode"
"bitopic binding mode"
bound to the receptor
allosteric interaction
A2AR antagonist
biomarker
biomarker
Figure 2. A schematic example of possible binding modes for probes incorporating an A2AR antagonist: bitopic ligand (upper), homobivalent ligand and hypothetical binding
modes (middle) and biomarker (lower).
neuroprotective agents against neuronal degeneration observed in
Parkinson’s disease.8–10
allosteric site on receptor 2 (Fig. 2, middle). Ligands of that type of-
fer potential advantages such as enhanced potency and receptor
subtype selectivity and improved pharmacokinetics compared to
a multi-drug regimen.13,16–18 Furthermore, significant research
has focused on the design of biomarkers and fluorescent ligands
targeting the adenosine receptors in an effort to further character-
ize their role in important disease states (Fig. 2).19,20 For all of these
novel concepts, the ligand must allow for extension towards the
extracellular space from the original binding mode to avoid any
significant loss of biological activity. A well-established example
of this concept is the high affinity xanthine amine congener, XAC
(4) (Fig. 1).11,21 The relatively insensitive amine terminal of XAC
A strategy in drug discovery programs with increasing popular-
ity is the concept of hybrid molecules targeting GPCRs,13–15 where-
by single molecular entities target multiple binding sites on one
receptor (bitopic ligand), or adjacent receptors (hetero- or
homobivalent ligand) simultaneously (Fig. 2). In the second case,
multiple binding modes have to be considered; both pharmaco-
phores possibly bind to an orthosteric or allosteric site or the
hetero-/homobivalent ligand could interact in a biotopic mode
across two dimers, that is one pharmacophore binds to the orthos-
teric site of receptor 1 and the second pharmacophore binds to an
O
O
N
NO
O
(ii)
O
(i)
N
N
N
N
H
N
H
OH
O
N
NH2
O
O
O
NH2
H
N
O
O
O
(i)
(ii)
N
6
7
N
NH
9
8
H2N
H2N NH
N
H2N
NH2
(iii)
15
14
16
O
N
NH2
NH2
N
NH2
NH2
O
O
O
N
(iv)
(iii)
(iv)
N
O
N
N
O
N
N
N
Cl
HO
O
O
O
N
S
N
S
N
O
O
O
11
12
18
17
10
NH2
(v)
Y
(v-vii)
N
O
N
N
N
O
X
N
O
N
(vi)
(vii)
O
H
N
O
N
5
N
X= NH, Y= OH
19 X= NH, Y= Br
20 X= O, Y= H
N
O
N
O
O
O
O
N
NH2
2
13
Scheme 2. Synthesis of ZM 241,385 (5) and analogs 19 and 20. Reagents and
conditions: (i) (S)-methylisothiosulfate hemisulfate, NaOH, water, rt, 24 h, 58%; (ii)
water, microwave, 140 °C, 1 h, 99% or water, reflux, 29 h, 80%; (iii) N-cyanodithioi-
minocarbonate, 170 °C, 1 h, 32%; (iv) m-CPBA, DCM, 22 h, 83%; (v) tyramine, CH3CN,
rt, 22 h, 33% (5); (vi) 4-bromophenethylamine, CH3CN, rt, 18 h, 32% (19); (vii) 2-
phenethyl alcohol, DBU, DME, reflux, 1 h, 20% (20).
Scheme 1. Synthesis of KW 6002 (2). Reagents and conditions: (i) acetic anhydride,
80 °C, 2 h, 83%; (ii) sodium nitrite, 50% acetic acid, 60 °C, 15 min, 86%; (iii) sodium
dithionite, NH4OH solution (12.5% (w/v)), 60 °C, 30 min, 98%; (iv) SOCl2, toluene,
75 °C, 2 h, 97%; (v) pyridine, DCM, rt, 16 h, 66%; (vi) HMDS, cat. (NH4)2SO4, CH3CN,
160 °C, microwave, 5 h, 100% followed by (vii) MeI, K2CO3, DMF, rt, 2 h, 75%.
Please cite this article in press as: Jörg, M.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.03.070

M. Jörg et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
3
NH2
(4) has been used for extending into the extracellular space with
HO
multi-functionalized molecules targeting adenosine receptors.21,22
This study has focused on identifying the ideal position to at-
tach linkers to known and novel structurally related A2AR antago-
nists in order to extend into extracellular space with minimal
penalty on binding affinity; thus allowing potential attachment
of further functionalities. The literature A2AR antagonists, KW
6002 (2) and ZM 241,385 (5) (Fig. 1), served as lead compounds
to design novel molecules, as well as, to monitor the effects of var-
ious linker types and lengths.
N
N
N
Cl
N
H
N
N
N
N
N
28
25
26
Cl
Cl
Cl
N
+
(iii)
(i)
N
N
N
H
Cl
N
Cl
NH2
N
N
N
N
(ii)
N
24
27
N
Cl
N
The synthesis of KW 6002 (2) (Scheme 1) consisted of a conver-
gent pathway from two starting points; namely the synthesis of
the diaminouracil (10) following a procedure by Hockemeyer
et al.,23 and the preparation of the acid chloride 12. Subsequent
reaction of 10 and 12 afforded the intermediate scaffold 13.23 Ring
closure was performed in a microwave reactor in the presence of
hexamethyldisilazane and catalytic ammonium sulfate following
a modified procedure by Burbiel et al.24 to obtain the xanthine mo-
tif. Extended reaction time of 5 h was required to promote reaction
completion in quantitative yield. Finally, N-methylation of the xan-
thine with iodomethane and potassium carbonate afforded KW
6002 (2) in good yield.
(iv)
NH2
N
N
N
N
Br 29
Cl
NH2
N
NH2
N
HO
N
N
N
N
(v) / (vi)
(vii)
N
N
R1-3
R1-3
N
Cl
H
R1= furan-2-yl 21
R1= furan-2-yl 30
R2= 2H-1,2,3-triazol-2-y 22
R3= 1H-1,2,3-triazol-1-y 23
R2= 2H-1,2,3-triazol-2-yl 31
R3= 1H-1,2,3-triazol-1-yl 32
Scheme 3. Synthesis of the furan-hybrid 21, the triazole-hybrids 22 and 23 and
model compound 28. Reagents and conditions: (i) MeI, K2CO3, CH3CN, rt, 24 h, 1:4
(25:26), 48% (26); (ii) NH4OH solution (32% (w/v)), CH3CN, 55 °C, 25 h, 82%; (iii)
tyramine, DIPEA, DMSO, 145 °C, 42 h, 32%; (iv) bromine, THF/MeOH/sodium acetate
buffer pH 4 1:1:1, À10 °C, 10 min then rt, 20 min, 60%; (v) 2-furanboronic acid,
Pd(PPh3)4, Cs2CO3, DME/water 10:3, 85 °C, 20 h, 51% (30); (vi) 1H-1,2,3-triazole,
Cs2CO3, DMF, 80 °C, 20 h, 32% (31) and an enriched mixture of 31 and 32; (vii)
tyramine, DIPEA, DMSO, 145 °C, 42 h, 32% (21); 28 h, 32% (22); 28 h, 10% (23).
ZM 241,385 (5) is a potent and selective A2AR antagonist25,26
and the first ligand for which crystal structures in complex with
the A2AR have been solved.3,27 ZM 241,385 (5) was synthesized
according to Scheme 2.28 The synthetic pathway has been adapted
from the original patent where the A2AR antagonist 5 was manu-
factured from aminoguanidine nitrate and 2-furonitrile.25 Our
reaction pathway commenced with commercially available 2-fur-
anhydrazide (14) and incorporates fairly inexpensive reagents.
The conditions used to afford intermediates 1529 and 1630 have
been previously published but not in context with the preparation
of ZM 241,385 (5). The final three steps were performed following
a patent procedure described by Caulkett et al.25 The overall yield
for the synthetic pathways illustrated in Scheme 2 was 5% (>97%
purity by analytical HPLC, 214 nm and 254 nm).28 Compounds 19
and 20 were synthesized using the same general synthetic path-
way. A non-nucleophilic base (1,8-diazabicycloundec-7-ene) was
employed in the final step of the synthesis of compound 20 for
reaction progression.25
may have comparable activity as A2AR antagonists due to their
structural and functional similarities. The furan-hybrid 21 differs
from ZM 241,385 (5) only by the replacement of the nitrogen atom
in position 5 with a carbon atom (diazolodiazine or purine), and
has a methyl group at position 7. The 2H-triazole-hybrid 22, on
the other hand, is very similar to the structure of ST 1535 (3) ex-
cept the alkyl chain is substituted with a tyramine moiety.
Hybrids 21–23 were synthesized via a five step pathway start-
ing from commercially available 2,6-dichloropurine (24)
(Scheme 3). The methyl group was successfully introduced under
alkaline conditions using methyl iodide in acetonitrile.32 The ratio
of the isomers 25 and 26 according to the crude 1H NMR was about
1:4 favouring the desired product 26. The isomers 25 and 26 were
separable by column chromatography. The amination of com-
pound 26 at position 6 (resulting from the nucleophilic substitu-
tion of Cl) was performed in 32% ammonia solution at room
temperature.33 The crude material was purified by suspending it
in methanol followed by filtration; resulting in an excellent yield
(82%). The reaction was also performed using 2N ammonia solu-
tion in methanol but these reaction conditions gave a complex
mixture of the desired product 27 and a by-product where the
chlorine at position 6 was substituted with a methoxy group. The
substitution of the chlorine in position 5 with tyramine (28) was
investigated as a model reaction for target compounds 21–23. In
this reaction, tyramine was attached to 27 under alkaline condi-
tions at 145 °C. Product 28 was obtained in a reasonable yield of
32%. Subsequent bromination of 27 using neat bromine at room
temperature in a solvent mixture of tetrahydrofuran/methanol/so-
dium acetate buffer (pH 4)34 afforded, after purification by column
chromatography, product 29 (60% yield). Despite optimization
(longer reaction time, heating to 50 °C and different solvent sys-
tems), 10–20% of starting material failed to convert into the desired
product 29. Attempts to effect this transformation using N-bromo-
succinimide and N-iodosuccinimide were unsuccessful. In the case
of the furan-hybrid 21, the intermediate 29 was further reacted
under Suzuki coupling conditions with 2-furanboronic acid to give
We designed and synthesized three novel hybrid compounds
21–23 based on the structures of ST 1535 (3)31 and ZM 241,385
(5)25 (Fig. 3). The hypothesis was that these novel compounds
NH2
N
NH2
HO
N
N
O
N
N
N
N
N
N
N
N
H
N
N
5
3
ST 1535
ZM 241385
hybridization
NH2
HO
N
N
N
N
N
N
N
H
N
22
2H-triazole-hybrid
NH2
NH2
HO
HO
N
N
O
N
N
N
N
N
N
N
N
H
N
N
H
N
23
21
1H-triazole-hybrid
furan-hybrid
Figure 3. Overview of the novel hybrids 21–23 derived from the parent compounds
ST 1535 (3) and ZM 241,385 (5).
Please cite this article in press as: Jörg, M.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.03.070

4
M. Jörg et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
the desired product 30 in 51% yield. The final step towards the syn-
thesis of 21 was the nucleophilic aromatic substitution of chlorine
with tyramine, which employed the model alkaline conditions at
145 °C used for the preparation of 28. The furan-hybrid 21 was iso-
lated as a white solid in 30% yield and, following column chroma-
tography and recrystallization, was obtained with purity greater
than >95% (analytical HPLC, 214 nm and 254 nm). To furnish the
triazole-hybrids 22 and 23, compound 29 was initially reacted
with 1H-1,2,3-triazole.35 This reaction was performed under alka-
line conditions in N,N-dimethylformamide at 80 °C. After 20 h, all
the starting material was converted to products 31 and 32, which
could not be separated by column chromatography. Nevertheless,
we obtained isomer 31 with >91% purity (analytical HPLC,
214 nm and 254 nm), which was subsequently added to tyramine
to obtain the desired 2H-triazole-hybrid 22 in 32% yield. Tyramine
was combined with the mixture of intermediates 31 and 32 as a
means of retrospectively characterizing compound 32 through
the structural elucidation of 23. This reaction gave the expected
mixture of compounds 22 and 23, and after column chromatogra-
phy and recrystallization, a sample of compound 23 (>95% purity
by analytical HPLC, 214 nm and 254 nm) was obtained in 10%
yield.
This study focused on the identification of the optimal position
to attach a linker for extending towards the extracellular space
with minimal penalty on binding potency. The design of such com-
pounds was aided by published crystal structures of the A2AR with
bound antagonists XAC (4) (3REY, cyan)27 and ZM 241,385 (5)
(3EML, green).3 The crystal structures show the binding mode of
the nitrogen-rich aromatic heterocycle deep into the binding cavity
(Fig. 4), thus allowing the para-substituted phenolic portion to ex-
tend towards extracellular space. In order to further explore this
binding mode, we synthesized ligands with linker attachment in
various positions, and assessed their inhibitory potency at the
ment of the linker portion, the synthesis commenced with ferulic
acid (33), with the phenol first protected as the acetate in acetic
anhydride and pyridine, to give 34. Acyl chloride formation, amide
coupling and ring closure occurred under the same conditions as
shown in Scheme 1, with the concomitant loss of the acetyl group,
to give xanthine 37. Regioselective N-methylation of 37 was
achieved with lithium bis(trimethylsilyl)amide and slow addition
of iodomethane to give 38 in good yield. The phenol group was
alkylated with ethyl 4-bromobutyrate (39), to give ether linked
functionality to the molecule. Further elaboration was accom-
plished by ethyl ester deprotection of 39 in sodium hydroxide
and methanol with acidic workup to furnish the carboxylic acid
40, followed by BOP-mediated amide coupling to give 41 and 42.
To produce ZM 241,385 analogs, we initially investigated the
phenolic oxygen of ZM 241,385 (5) for forming an ether linkage
(Scheme 5). Compound 5 was combined with ethyl 4-bromobuty-
rate under alkaline conditions, but the synthesis was insufficiently
selective and resulted in a mixture of 43–45, which were separated
chromatographically. The formation of an ether linkage via Mitsun-
obu reaction conditions also failed. A synthetic route to selectively
obtain 43 has been subsequently discovered, which entails O-
alkylation of Boc-protected tyramine with ethyl 4-bromobutyrate,
followed by removal of the protecting group to reveal the primary
amine, and finally nucleophilic substitution of the sulfone interme-
diate 18. Next, we investigated the formation of an ester linkage
under acidic conditions whereby three linkers of various lengths
were attached to ZM 241,385 (5) (Scheme 6). This protocol proved
to be productive as the acidic conditions circumvented the forma-
tion of the previously observed N-substituted by-products
(Scheme 5).
These linker types (ether and ester linkage) were also attached
to the structurally related 2H-triazole-hybrid 22, as illustrated in
Scheme 7. Unlike for ZM 241,385 (5), alkylation and acylation
afforded the desired O-linked product selectively.
A2AR.
The antagonist KW 6002 (2) shares structural similarity with
To identify A2AR specific antagonists, we utilized a recombinant
cell-based cAMP competitive immunoassay, the LANCE™ cAMP Kit
from PerkinElmer. The assay was used to compare the activity
(IC50) of the potentially novel A2AR antagonists with the literature
compounds as well as to determine the most suitable attachment
position of the linker (Table 1). The comparison of the pIC50 values
XAC (4). KW 6002 analog 41 and 42, with linkers attached through
the para-methoxy position of KW 6002 (2), were synthesized
according to Scheme 4. The key intermediate 38 was synthesized
using the same procedure as illustrated in Scheme 1 for the synthe-
sis of KW 6002 (2). However, to achieve a single point for attach-
Figure 4. (a) Binding cavity of the A2AR crystal structure (3REY, cyan) with XAC (4) bound. A hydrogen bond (black dashes) was observed between the ligand and Asn253. (b)
Binding cavity of the A2AR crystal structure (3EML, green) with ZM 241,385 (5) bound. Hydrogen bonds were observed between the ligand and Asn253, Glu169, and water
molecules. Images were generated using PyMOL software.36
Please cite this article in press as: Jörg, M.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.03.070

M. Jörg et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
5
OH
O
O
O
O
O
(ii)
(i)
O
O
O
HO
HO
Cl
O
O
O
34
35
33
(iii)
O
O
O
N
O
N
H
N
H
N
OH
(iv)
N
N
N
O
N
O
O
O
NH2
36
37
(v)
O
O
N
O
O
(vi)
O
OH
O
N
N
N
N
N
O
O
O
O
O
N
38
39
(vii)
OH
O
O
O
HN
O
n
O
N
O
N
(viii)
N
N
N
O
O
N
N
40
O
O
N
41
42
n=5
n=11
Scheme 4. Attachment of linker groups at the para-methoxy position of KW 6002 (2). Reagents and conditions: (i) Ac2O, DMAP, pyridine, 0 °C, 1 h, 79%; (ii) SOCl2, toluene,
75 °C, 2 h, 96%; (iii) 10, pyridine, DCM, rt, 16 h, 66%; (iv) HMDS, cat. (NH4)2SO4, MeCN, 160 °C, microwave, 5 h, 95%; LiHMDS, MeI, DMF, 0 °C, 2 h, 67%; (v) LiHMDS, MeI, DMF,
0 °C, 2 h, 67%; (vi) ethyl 4-bromobutyrate, K2CO3, NaI, acetone, reflux, 3 d, 57%; (vii) 1 M NaOH, MeOH, 60 °C, 2 h, 88%; (viii) methyl 6-aminohexanoate or methyl 12-
aminododecanoate, BOP, DIPEA, DCM, rt, 2 h, 31% (41), 37% (42).
O
NH2
N
NH2
N
O
N
N
O
O
(i)
O
N
N
N
O
N
O
N
H
N
N
H
N
43
46
O
NH2
N
NH
N
(ii)
O
O
N
N
O
O
i)
HO
O
5
N
N
N
N
O
5
N
O
O
N
H
N
N
H
N
47
44
NH2
N
O
(iii)
N
N
O
NH
N
O
O
N
N
O
O
N
N
H
N
48
N
H
N
45
Scheme 6. Attachment of an ester linker to ZM 241,385 (5). Reagents and
conditions: (i) acetyl chloride, TFA, DCM, rt, 44 h, 89%; (ii) butyric anhydride, TFA,
DCM, rt, 4 h, 92%; (iii) decanoyl chloride, TFA, DCM, rt, 4 h, 74%.
Scheme 5. Attachment of functionality to ZM 241,385 (5) using an ether linkage.
Reagents and conditions: (i) ethyl 4-bromobutyrate, K2CO3, acetone, 50 °C, 46 h, 5%
(43), 6% (44) and 13% (45).
O
NH2
N
O
O
(i)
N
N
N
N
O
N
N
of the synthesized A2AR antagonists (2, 5 and 19–23) showed that
the compounds based on the triazolotriazine scaffold (5, 19 and 20)
exhibited the greatest potency. Equal potency was observed be-
tween compound 5 (42 nM) and the unsubstituted phenethoxy
derivative 20 (41 nM). The substitution of the phenolic oxygen
with a bromine atom, however, lead to a four-fold decrease in
activity for compound 19, possibly due to the larger bromine atom
and its reduced capacity to act as a hydrogen bond acceptor. The
novel hybrid molecules 21–23 showed respectable pIC50 values,
but were up to two orders of magnitude less active than the liter-
ature compound ZM 241,385 (5). A preference for the furan system
over the triazole moiety was observed within the hybrid series 21–
23, whereby 21 exhibited significantly greater activity than 22 and
23 (p value < 0.05).
N
N
N
H
49
22
NH2
(ii)
N
N
N
N
O
N
H
N
50
Scheme 7. Attachment of an ester and ether linker to triazole-hybrid (22). Reagents
and conditions: (i) ethyl 4-bromobutyrate, K2CO3, NaI, acetone, reflux, 66 h, 26%; (ii)
butyric anhydride, TFA, DCM, rt, 4 h, 92%.
finding is not surprising given that literature postulates the furan
ring is essential for the activity of ZM 241,385 (5) and structurally
related compounds.37
Compound 28, which does not contain a furan ring or triazole
moiety, did not show any detectable activity at the A2AR. This
Please cite this article in press as: Jörg, M.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.03.070

6
M. Jörg et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
Table 1
Inhibitory potency of synthesized A2AR ligands and monovalent variants using the
LANCE™ cAMP assay
a
a
Entry
pIC50
IC50 (nM)
Entry
pIC50
IC50 (nM)
2
5
5.28 0.19
7.38 0.06
6.74 0.15
7.39 0.12
5.58 0.05
5.01 0.01
4.26 0.06
<4
5250
42
182
41
42
43
44
45
46
47
48
49
50
4.96 0.04
<4
7.30 0.01
4.59 0.12
<4
7.41 0.04
7.53 0.01
6.74 0.06
5.04 0.11
5.01 0.04
>10,000
>100,000
50
>10,000
>100,000
39
30
182
9120
9770
19
20
21
22
23
28
38
39
41
2630
9770
>10,000
>100,000
>10,000
4370
4.92 0.04
5.36 0.16
a
Data represent the mean SEM of two separate experiments performed in
duplicate. The pIC50 values of the A2AR antagonists were determined at the EC80 of
NECA (500 nM). The EC50 of NECA was determined as 162 nM in the same assay.
KW 6002 (2) showed lower antagonism than expected, given its
high potency binding results reported in the literature.38 Nonethe-
less, slight modification of the para-methoxy of KW 6002 (2) were
relatively well tolerated, indicated by the similar activity of com-
pounds 38 and 39 to the parent compound 2. Extending further
with longer chain linkers (compounds 41 and 42) resulted in
diminished potency in a length-dependent manner.
Compounds 43–45 showed consistent potency in support of our
hypothesis that the phenolic oxygen is pointing out of the receptor
pocket and is therefore the most optimal position for linker attach-
ment. The O-alkylated compound 43 exhibited comparable activity
to ZM 241,385 (5) whereas the N-alkylated compound 44 was vir-
tually inactive. Not surprisingly, the doubly alkylated compound
45 did not show any detectable activity.
The effect of elongation of the linker at the phenolic oxygen
(from 2 to 10 carbon atoms) was investigated with compounds
46–48. The pIC50 values were mostly maintained compared to
the parent compound 5, although a four fold decrease in activity
was observed for compound 48.
Finally, given the structural similarity of hybrid molecules 21–
23 to ZM 241,385 (5), it was expected that extension of linkers
from the phenolic oxygen would be equally well tolerated from
both scaffolds. This was confirmed with the O-linked triazole-hy-
brids 49 and 50, which maintained an activity comparable to that
of the parent compound 22, irrespective of linker type used.
To correlate the biological results of the aforementioned com-
pounds to their potential binding mode, molecular modeling was
undertaken using published crystal structures of the A2AR. The
KW 6002-based compounds, 38–42, were docked39,40 into the ther-
mostabilized A2AR, which was complexed with XAC (4) in the crys-
tal structure (3REY, cyan).27 This model was chosen due to the
structural similarities between KW 6002 (2) and XAC (4), most
notably the common N,N-diethylxanthine unit. To validate the
method, XAC (4) was redocked into the binding cavity (3REY, cyan),
producing a binding mode for the xanthine unit similar to the
crystal structure (rmsd of 1.17 Å41), as seen in Figure 5a. However,
the rmsd for the entire molecule dropped (4.07 Å) due to a different
orientation of the (2-aminoethyl)acetamido portion caused by
additional hydrogen bonding to Glu169 in the docked pose.
Using this receptor model, compounds 38–42 displayed a simi-
lar binding site orientation of the xanthine portion, as exemplified
by docked poses of 38 (Fig. 5b) and 39 (Fig. 5c). This orientation al-
lowed the extension of linkers towards extracellular space. The
length of the linker did not have a significant impact on the binding
mode of compounds 38–42.
Figure 5. Representation of the A2AR receptor crystal structure (3REY, cyan; 3EML,
green) with ligands docked in the binding cavity. Docked ligands: carbon, gray;
Nitrogen, blue; oxygen, red. (a) Overlay of docked XAC (4) with bound crystal
structure ligand (fuscia); (b) compound 38; (c) compound 39; (d) overlay of docked
ZM 241,385 (5) with bound crystal structure ligand (orange); (e) compound 48; (f)
compound 43; (g) compound 45; (h) compound 22. Images were generated using
PyMOL software.36
comparable binding mode with acceptable rmsd for the entire mol-
ecule of 1.67 Å,41 (Fig. 5d). Compounds 19 and 20 showed the same
minding mode as ZM 241,385 (5).
Next, we explored the effect on binding mode of alkylation at
various positions of ZM 241,385 (5). Linker-attachments on the ter-
minal phenolic group in compounds 43, 46–48 had minimal effect
on docked poses and confirmed the benefit of attaching alkyl
groups at this position to extend toward the extracellular space.
The length and nature of the linker attached from this position
The docking of compounds 5, 19–23 and 43–50 was based on
the crystal structure of the T4-lysozyme stabilized A2AR in complex
with ZM 241,385 (5) (3EML, green).3 To validate the method, ZM
241,385 (5) was redocked into the crystal structure and gave a
Please cite this article in press as: Jörg, M.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.03.070

M. Jörg et al. / Bioorg. Med. Chem. Lett. xxx (2013) xxx–xxx
7
O
NH2
N
Supplementary data
O
N
N
O
HO
N
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2013.03.070.
N
H
N
51
Figure 6. Proposed structure of the first triazolotriazine carboxylic acid congener
(TCAC) 51.
References
1. Fredholm, B. B.; Chen, J.-F.; Masino, S. A.; Vaugeois, J.-M. Annu. Rev. Pharmacol.
Toxicol. 2005, 45, 385.
2. Fredholm, B. B.; IJzerman, A. P.; Jacobson, K. A.; Klotz, K.-N.; Linden, J.
Pharmacol. Rev. 2001, 53, 527.
3. Jaakola, V.-P.; Griffith, M. T.; Hanson, M. A.; Cherezov, V.; Chien, E. Y. T.; Lane, J.
R.; IJzerman, A. P.; Stevens, R. C. Science 2008, 322, 1211.
4. Ross, G. W.; Abbott, R. D.; Petrovitch, H.; Morens, D. M.; Grandinetti, A.; Tung,
K.-H.; Tanner, C. M.; Masaki, K. H.; Blanchette, P. L.; Curb, J. D.; Popper, J. S.;
White, L. R. J. Am. Med. Assoc. (JAMA) 2000, 283, 2674.
5. Chen, J. F.; Xu, K.; Petzer, J. P.; Staal, R.; Beilstein, M.; Sonsalla, P. K.; Castagnoli,
K., ; Castagnoli, N., Jr; Schwarzschild, M. A. J. Neurosci. 2001, 21. RC143 (1–6).
6. Bibbiani, F.; Oh, J. D.; Petzer, J. P.; Castagnoli, N.; Chen, J. F.; Schwarzschild, M.
A.; Chase, T. N. Exp. Neurol. 2003, 184, 285.
caused little effect on the binding mode of the scaffold (illustrated
in Fig. 5e and 5f). The orientation of the alkyl group allows exten-
sion towards the extracellular space, thus minimizing interference
with key binding site interactions. The alkylation of the amino
group to yield 44 and 45 resulted in a flipped binding mode
(Fig. 5g). This finding may explain the observed reduction in antag-
onistic activity of these compounds compared to ZM 241,385 (5),
as the crucial binding interactions are lost.
7. Tronci, E.; Simola, N.; Borsini, F.; Schintu, N.; Frau, L.; Carminati, P.; Morelli, M.
Eur. J. Pharmacol. 2007, 566, 94.
The hybrid molecules 21–23 and linker-extended analogs 49–50,
all based on a purine scaffold, generally gave docked poses similar to
ZM 241,385 (5). These compounds, however, were unable to pene-
trate as deep into the binding pocket as exemplified in Figure 5h.
This phenomenon, in addition to the missing nitrogen atom at posi-
tion 5 and introduced N-methyl group at position 7, potentially ex-
plains the lower IC50 values observed for the hybrid molecules.
In conclusion, the phenolic oxygen of ZM 241,385 (5) has been
shown to be a suitable position to attach linkers. In addition, com-
pound 43 has emerged to be the ideal structure for further studies
of the A2AR. From the aforementioned biological data, compound
43 was identified as a high affinity analog of ZM 241,385 (5). Its
terminal ester group allows for facile functionalization and chem-
ical elaboration. This structural feature is in accordance with the
concept of a ‘functionalized congener’;21 a molecule that comprises
a chemical functional group (e.g. amine or carboxylic acid) that is
suitably linked to a pharmacophore. We envisage the hydrolysis
of 43 to afford compound 51 (Fig. 6), a functionalized congener,
revealing a carboxylic acid functionality that can be utilized to syn-
thesize pharmacological conjugates such as bivalent ligands, bitop-
ic ligands and biomarkers. In relation to the molecule XAC (4), a
xanthine-based versatile congener widely used to study adenosine
receptorology, compound 51 has emerged as a triazolotriazine-
based equivalent (Triazolotriazine Carboxylic Acid Congener
(TCAC)) that warrants further investigation.
8. LeWitt, P. A.; Guttman, M.; Tetrud, J. W.; Tuite, P. J.; Mori, A.; Chaikin, P.;
Sussman, N. M. Ann. Neurol. 2008, 63, 295.
9. Pinna, A. Expert Opin. Invest. Drugs 2009, 18, 1619.
10. Postuma, R. B.; Lang, A. E.; Munhoz, R. P.; Charland, K.; Pelletier, A.; Moscovich,
M.; Filla, L.; Zanatta, D.; Rios, R. S.; Altman, R.; Chuang, R.; Shah, B. Neurology
2012, 79, 651.
11. Müller, C. E.; Jacobson, K. A. Biochim. Biophys. Acta 2011, 1808, 1290.
´
12. Kecskes, A.; Tosh, D. K.; Wei, Q.; Gao, Z.-G.; Jacobson, K. A. Bioconjugate Chem.
2011, 22, 1115.
13. Shonberg, J.; Scammells, P. J.; Capuano, B. ChemMedChem 2011, 6, 963.
14. Morphy, R.; Kay, C.; Rankovic, Z. Drug Discovery Today 2004, 9, 641.
15. Valant, C.; Sexton, P. M.; Christopoulos, A. Mol. Interv. 2009, 9, 125.
16. Berque-Bestel, I.; Lezoulac’h, F.; Jockers, R. Curr. Drug Disc. Technol. 2008, 5, 312.
17. Portoghese, P. S.; Larson, D. L.; Yim, C. B.; Sayre, L. M.; Ronsisvalle, G.; Lipkowski,
A. W.; Takemori, A. E.; Rice, K. C.; Tam, S. W. J. Med. Chem. 1985, 28, 1140.
18. McRobb, F. M.; Crosby, I. T.; Yuriev, E.; Lane, J. R.; Capuano, B. J. Med. Chem.
2012, 55, 1622.
19. Vernall, A. J.; Stoddart, L. A.; Briddon, S. J.; Hill, S. J.; Kellam, B. J. Med. Chem.
2012, 55, 1771.
20. Baker, J. G.; Middleton, R.; Adams, L.; May, L. T.; Briddon, S. J.; Kellam, B.; Hill, S.
J. Br. J. Pharmacol. 2010, 159, 772.
21. Jacobson, K. A.; Ukena, D.; Padgett, W.; Kirk, K. L.; Daly, J. W. Biochem.
Pharmacol. 1987, 36, 1697.
22. Jacobson, K. A.; Daly, J. W. Nucleosides Nucleotides 1991, 10, 1029.
23. Hockemeyer, J.; Burbiel, J. C.; Müller, C. E. J. Org. Chem. 2004, 69, 3308.
24. Burbiel, J. C.; Hockemeyer, J.; Müller, C. E. J. Org. Chem. 2006, 2, 20.
25. Caulkett, P.W.R.; Jones, G.; Collis, M.G.; Poucher, S.M. EP 459702, 1991.
26. Caulkett, P. W. R.; Jones, G.; McPartlin, M.; Renshaw, N. D.; Stewart, S. K.;
Wright, B. J. Chem. Soc., Perkin Trans. 1 1995, 7, 801.
27. Doré, A. S.; Robertson, N.; Errey, J. C.; Ng, I.; Hollenstein, K.; Tehan, B.; Hurrell,
E.; Bennett, K.; Congreve, M.; Magnani, F.; Tate, C. G.; Weir, M.; Marshall, F. H.
Structure 2011, 19, 1283.
In summary, the investigation of our series of A2AR ligands,
based on the literature compounds KW 6002 (2) and ZM 241,385
(5), revealed that (i) compounds 46, 20, 5 and 43 (with IC50 values
of 39, 41, 42 and 50 nM, respectively) are equipotent to the litera-
ture compound ZM 241,385 (5) (IC50 = 42 nM) (Table 1); (ii) the
attachment of an ester or ether linker to the phenolic oxygen of
ZM 241,385 (5) was well tolerated; (iii) linkers up to 10 carbon
atoms in length were well tolerated when incorporated into ZM
241,385 (5) but less so with the KW 6002 (2) scaffold; (iv) identi-
fication of molecule 51 as a novel TCAC with scope to further ex-
plore the A2AR and associated extracellular space. The biological
results, in conjunction with molecular modeling, supported our
working hypothesis: the phenolic oxygen is a highly suitable posi-
tion for the attachment of functionalized linkers to extend towards
extracellular space from the A2AR.
28. Jörg, M.; Agostino, M.; Yuriev, E.; Mak, F.; Miller, N.; White, J.; Scammells, P.;
Capuano, B. Struct. Chem. 2012. http://dx.doi.org/10.1007/s11224-012-0151-7.
published online.
29. Srivastava, R. P.; Kumar, V. V.; Bhatia, S.; Sharma, S. Indian J. Chem., Sect. B 1995,
34B, 209.
30. Dolzhenko, A. V.; Pastorin, G.; Dolzhenko, A. V.; Chui, W. K. Tetrahedron 2009,
50, 2124.
31. Giorgio, T.; Giovann, P.; Patrizia, M.; Maria Assunta Di, C.; Grazia, G.; Fabrizio,
G.; Luca, G. WO03011864, 2003.
32. Janssen, P.A.J.; Lewi, P.J.; De Jonge, M.R.; Koymans, L.M.H.; Daeyaert, F.F.D.;
Heeres, J.; Vinkers, H.M.; Leenders, R.G.G.; Vandenput, D.A.L. WO 2005028479,
2005.
33. Foitzik, R. C.; Devine, S. M.; Hausler, N. E.; Scammells, P. J. Tetrahedron 2009, 65,
8851.
34. Maruyama, T.; Kozai, S.; Sasaki, F. Nucleosides, Nucleotides Nucleic Acids 2000,
19, 1193.
35. Bartoccini, F.; Cabri, W.; Celona, D.; Minetti, P.; Piersanti, G.; Tarzia, G. J. Org.
Chem. 2010, 75, 5398.
36. Schrödinger, L.L.C. ThePYMOL GraphicsSystem,version1.5.0.4;New York,NY, 2010.
37. Alanine, A.; Anselm, L.; Steward, L.; Thomi, S.; Vifian, W.; Groaning, M. D.
Bioorg. Med. Chem. Lett. 2004, 14, 817.
38. Chen, J. F.; Xu, K.; Petzer, J. P.; Staal, R.; Xu, Y. H.; Beilstein, M.; Sonsalla, P. K.;
Castagnoli, K.; Castagnoli, N., Jr; Schwarzschild, M. A. J. Neurosci. 2001,
21:RC143 (1–6).
39. Friesner, R. A.; Banks, J. L.; Murphy, R. B.; Halgren, T. A.; Klicic, J. J.; Mainz, D. T.;
Repasky, M. P.; Knoll, E. H.; Shelley, M.; Perry, J. K.; Shaw, D. E.; Francis, P.;
Shenkin, P. S. J. Med. Chem. 2004, 47, 1739.
40. McRobb, F. M.; Capuano, B.; Crosby, I. T.; Chalmers, D. K.; Yuriev, E. J. Chem. Inf.
Model. 2010, 50, 626.
Acknowledgments
M.J. is a recipient of the Australian Postgraduate Award Industry
(APAI). J.S. is a recipient of the Australian Postgraduate Award
(APA). Thank you to Phoebe Lei and Qianqian Wu (GSK R&D, China)
for performing the A2AR functional assay. Thank you to Dr Jason
Dang (Monash University) for helping obtain NMR spectral data,
HRMS and HPLC chromatograms.
41. Yuriev, E.; Agostino, M.; Ramsland, P. A. J. Mol. Recognit. 2011, 24, 149.
Please cite this article in press as: Jörg, M.; et al. Bioorg. Med. Chem. Lett. (2013), http://dx.doi.org/10.1016/j.bmcl.2013.03.070