The aminopyridine-3,5-dicarbonitrile core for the design of new non-nucleoside-like agonists of the human adenosine A2B receptor

Article abstract of DOI:10.1016/j.ejmech.2018.02.081

A new series of amino-3,5-dicyanopyridines (3–28) as analogues of the adenosine hA_2B receptor agonist BAY60-6583 (compound 1) was synthesized. All the compounds that interact with the hA_2B adenosine receptor display EC₅₀ v

Full text of DOI:10.1016/j.ejmech.2018.02.081

European Journal of Medicinal Chemistry 150 (2018) 127e139
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
The aminopyridine-3,5-dicarbonitrile core for the design of new non-
nucleoside-like agonists of the human adenosine A2B receptor
,
Marco Betti a, Daniela Catarzi a *, Flavia Varano a, Matteo Falsini a, Katia Varani b,
Fabrizio Vincenzi b, Diego Dal Ben c, Catia Lambertucci c, Vittoria Colotta a
a
ꢀ
Dipartimento di Neuroscienze, Psicologia, Area del Farmaco e Salute del Bambino, Sezione di Farmaceutica e Nutraceutica, Universita degli Studi di
Firenze, Via Ugo Schiff, 6, 50019 Sesto Fiorentino, Italy
b
ꢀ
Dipartimento di Scienze Mediche, Sezione di Farmacologia, Universita degli Studi di Ferrara, Via Fossato di Mortara 17-19, 44121 Ferrara, Italy
c
ꢀ
Scuola di Scienze del Farmaco e dei Prodotti della Salute, Universita degli Studi di Camerino, Via S. Agostino 1, 62032 Camerino, MC, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
A new series of amino-3,5-dicyanopyridines (3e28) as analogues of the adenosine hA2B receptor agonist
BAY60-6583 (compound 1) was synthesized. All the compounds that interact with the hA2B adenosine
receptor display EC50 values in the range 9e350 nM behaving as partial agonists, with the only exception
being the 2-{[4-(4-acetamidophenyl)-6-amino-3,5-dicyanopyridin-2-yl]thio}acetamide (8) which shows
Received 5 December 2017
Received in revised form
30 January 2018
Accepted 26 February 2018
Available online 6 March 2018
a
full
agonist
profile.
Moreover,
the
2-[(1H-imidazol-2-yl)methylthio)]-6-amino-4-(4-
cyclopropylmethoxy-phenyl)pyridine-3,5-dicarbonitrile (15) turns out to be 3-fold more active than 1
although less selective. This result can be considered a real breakthrough due to the currently limited
number of non-adenosine hA2B AR agonists reported in literature. To simulate the binding mode of
nucleoside and non-nucleoside agonists at the hA2B AR, molecular docking studies were performed at
homology models of this AR subtype developed by using two crystal structures of agonist-bound A2A AR
as templates. These investigations allowed us to represent a hypothetical binding mode of hA2B receptor
agonists belonging to the amino-3,5-dicyanopyridine series and to rationalize the observed SAR.
© 2018 Elsevier Masson SAS. All rights reserved.
Keywords:
G protein-coupled receptors
Adenosine A2B receptor agonists
Aminopyridine-3,5-dicarbonitriles
Ligand-adenosine receptor modelling
studies
1. Introduction
different chemical classes has been developed [4e10], only a few
A2B AR agonists are known so far [11]. As antagonists are charac-
Adenosine (Ado) is an endogenous purine nucleoside that nor-
mally increases under pathological or stressful situations producing
its effects through activation of G proteinecoupled adenosine re-
ceptors (ARs). These latter, classified as A1, A2A, A2B, and A3, are
typically coupled to adenylate cyclase but other second messenger
systems have also been described [1,2]. Over the years, many li-
gands, agonists and antagonists, have been identified for the A1,
terized by a large structural variability, the agonist profile has been
long associated to an Ado-like structure. Starting from the 50-(N-
ethylcarboxamido)adenosine (NECA), the first Ado-derived nucle-
osidic human (h) A2B AR agonist, a slightly more potent hA2B
agonist than NECA was identified [12]. Fortunately, progress has
been made. In fact, the non-Ado-like 2-{[6-Amino-3,5-dicyano-4-
(4-(cyclopropylmethoxy)phenyl)pyridin-2-yl]thio}acetamide
(BAY60-6583, compound 1), a 2-aminopyridine-3,5-dicarbonitrile
derivative (Chart 1) discovered by Bayer Healthcare [13,14], is the
only available potent and selective hA2B AR agonist reported so far.
Its identification has invalidated the conviction that the sugar
moiety is essential for agonism at ARs, such that non-nucleoside
ligands must therefore behave as antagonists. Thus, compound 1
has been used extensively as a research tool to clarify the phar-
macological roles of A2B AR [15e28] sometimes leading to contra-
dictory results [29]. Thus, it could be a very important goal to obtain
other potent and selective hA2B AR agonists especially considering
the difficulties that have emerged in understanding of the
A
2A, and A3 ARs that, in turn, have been extensively characterized
[3]. In contrast, the A2B AR subtype is the least known. In fact, while
a large number of selective A2B AR antagonists belonging to
Abbreviations:
ABMECA,
N6-(4-aminobenzyl)eN-methylcarbox-
amidoadenosine; Ado, Adenosine; AR, adenosine receptor; CHO, Chinese hamster
ovary; DPCPX, 8-cyclopentyl-1,3-dipropylxanthine; HEPES, 4-(2-hydroxyethyl)-1-
piperazineethanesulfonic acid; NECA, 50-(N-ethylcarboxamido)adenosine; EL,
extracellular loop; MOE, molecular operating environment; RMS, root-mean-
square; TM, transmembrane.
* Corresponding author.
E-mail address: daniela.catarzi@unifi.it (D. Catarzi).
https://doi.org/10.1016/j.ejmech.2018.02.081
0223-5234/© 2018 Elsevier Masson SAS. All rights reserved.

128
M. Betti et al. / European Journal of Medicinal Chemistry 150 (2018) 127e139
Chart 1. Lead structures for the development of currently reported amino-3,5-
dicyanopyridine-based AR ligands.
Chart 2. Modification performed at R1 and R2 positions of the 2-amino-4-aryl-6-
sulfanyl-3,5-dicyanopyridine scaffold.
pharmacological properties of A2B AR agonists and the necessity to
explain some controversies concerning the A2BAR [29]. In partic-
ular, the amino-3,5-dicyanopyridine series, to which compound 1
belongs, has been demonstrated to include partial agonists with a
variable maximum agonist effect at the hA2B AR subtype [30]. More
recently, two different papers reported the partial agonist profile of
1 [27,31] and also its potential A2B AR biased agonism was hy-
pothesized [29,32].
acetamido-benzaldehyde 34 was reacted with malononitrile in a
straightforward Knoevenagel condensation in the presence of a few
drops of piperidine as catalyst to give the intermediate 36 [37]. The
latter was reacted with malononitrile in a cyclization reaction
involving thiophenol and Et3N to afford 42 [37].
To obtain the free thiols (compounds 44e50 [13,37,38]), the
corresponding 6-phenylsulfanyl derivatives 37e43 [37] were
treated with sodium sulfide in DMF at 80 ꢀC followed by 1 M HCl.
The final compounds 3e28 were obtained by reaction of the 6-
thiol-derivatives 44e50 with the suitable halides in the presence
of sodium hydrogencarbonate. These latter were all commercially
available with the exception of the 2-chloro-N-hydroxyacetamide
51 [39] which was synthesized from ethyl 2-chloroacetate with 50%
aqueous solution of hydroxylamine as reported in Scheme 2.
Moreover, the hA2B AR agonist 1 [13,21] was cyclized in absolute
ethanolic potassium hydroxide to yield the bicyclic compound 52
(Scheme 3). The forced alkaline conditions produced the conden-
sation of the 3-cyano substituent with the active methylene group
In this scenario, our research group focused attention on the
aminopyridine-3,5-dicarbonitrile series to broaden the scarcely
known structure-activity relationships (SARs) of this chemical
class. In fact, most of the non-Ado-like AR ligands belonging to this
series are included in patent documents [12,13,21] while few data
are reported in the open literature [30,33,34]. These are, however,
sufficient to underline the versatility of the amino-3,5-
dicyanopyridine scaffold for producing AR ligands with not only a
wide range of affinities but, interestingly, with different degrees of
efficacy, ranging from full to partial agonist or neutral antagonist at
the different ARs. In particular, certain 2-amino-4-aryl-6-(1H-imi-
dazol-2-yl-methylsulfanyl)-pyridine-3,5-dicarbonitriles belonging
to the LUF series [30] (2, Chart 1) displayed nanomolar affinity for
all the ARs, including the A2B AR subtype on which they showed, in
general, also considerable efficacy. Moreover, this class of com-
pounds seems to be more versatile for pharmacological studies
showing less species differences than the Ado-like AR agonists [3].
It is worth noting that in addition to compound 1, that reached
preclinical-phase investigation for treating angina pectoris, also
other amino-3,5-dicyanopyridine derivatives discovered by Bayer
Healthcare have attracted attention for their potential in heart
diseases [32,34].
Thus, taking compound 1 as lead, modifications on the amino-
3,5-dicyanopyridine core were performed at both R1 and R2 posi-
tions (compounds 3e28, Chart 2).
2. Results and discussion
2.1. Chemistry
The synthetic pathways which yielded compounds 1, 3e28, 52
and the relative intermediates are illustrated in Schemes 1e3. The
amino-3,5-dicyanopyridine derivatives 1, 3e28 [13,21] (Scheme 1)
were obtained starting from aldehydes 29e35, all commercially
available with the exception of the 4-(cyclobuthylmethoxy)benz-
aldehyde 29 [35] which was obtained by reacting 4-
hydroxybenzaldehyde with (bromomethyl)cyclobutane in reflux-
ing acetone and in the presence of potassium carbonate. By one-pot
cyclization of the suitable aldehyde 29e33, 35 with malononitrile
and thiophenol, the sulfanylphenyl intermediates 37e41, 43 were
obtained. Different cyclization alkaline adjuvants able to work in a
phase-transfer system were used, the best being DBU [36]. Mod-
erate to good yields were obtained. Differently, the para-
Scheme 1. Reagents and conditions. a) To yield compound 29: BrCH2C4H7, acetone,
anhydrous K2CO3, reflux (67%); compounds 30e35 are commercially available; b)
malononitrile, thiophenol, DBU, 10% aqueous EtOH, 55 ꢀC (18e37%); c) malononitrile,
piperidine, EtOH, 80 ꢀC (63%); d) malononitrile, thiophenol, Et3N, EtOH, reflux (44%); e)
Na2S, anhydrous DMF, 80 ꢀC; 1 M HCl, rt (72e86%); f) R2CH2X (X ¼ Cl, Br), NaHCO3,
anhydrous DMF, rt (19e80%).

M. Betti et al. / European Journal of Medicinal Chemistry 150 (2018) 127e139
129
could have some influence in determining the pharmacological
profile of this derivative as observed for series 2 [41,42]. When
position R2 was modified by introducing carboxamido bioisosteres
(compounds 9e14), best results were obtained in terms of hA2B AR
activity and selectivity. In fact, the N-methylacetamido compound 9
and its analogous N-hydroxymethylacetamido 10 are active at the
hA2B AR, compound 9 also being the most selective within the
herein reported series. In contrast, the hydroxamic acid derivative
11 loses its potency, probably due to the acidity of its R2 residue
which is scarcely tolerated by the hA2B AR. Aryl homologation of
the lead 1 yielded compound 13 which is highly potent at the hA2B
AR subtype thus confirming how both H-bond acceptor and donor
functions at this position are essential for hA2B AR-ligand interac-
Scheme 2. Reagents and conditions. a) 50% aqueous NH2OH, rt (53%).
tion. Also, the ameliorative
p-stacking contribution of the phenyl
moiety at R2 cannot be ignored leading to a great increase of
binding affinity at the A1 subtype.
Scheme 3. Reagents and conditions. a) KOH, absolute EtOH, reflux (85%).
Starting from the imidazole idea (see series 2, Chart 1), com-
pound 15 was prepared and it emerged as a very potent hA2B AR
agonist being 3-fold more active than the reference agonist 1.
Moreover, it shows good selectivity versus the other AR subtypes
and a partial agonist profile. Thus, the imidazolyl moiety at R2 was
replaced with diverse H-bond donor/acceptor-groups containing
heterocycles to yield the subset of compounds 16e24. All de-
rivatives that interact with hA2B AR behave as partial agonist as 15.
First, compound 16 bearing an imidazol-5-yl moiety at R2 was
produced, resulting less active than its isomer 15 at the hA2B AR, but
also at the other ARs. Compound 19, bearing a 3-pyrazolyl group,
shows a good hA2B AR affinity that is comparable with that of the
1,2,4-triazol-5-yl derivative 20. These data, together with the
inactivity of compound 23, confirm again the importance of the
presence of H-bond acceptor/donor functions at R2. Also tetrazolyl-
substitution (compound 21) results as detrimental for hA2B AR
activity, probably due to the excessive acidity of the NH tetrazole
moiety, as observed for the idroxamic acid derivative 11.
on the sulfanylacetamide chain [40].
2.2. Pharmacological assays
The newly synthesized derivatives 3e28, 52 and the reference
compound 1 were studied as hA2B AR agonists by evaluating their
stimulatory effect on cAMP production in Chinese Hamster Ovary
(CHO) cells, stably expressing the hA2B AR. Some selected com-
pounds (3, 12, 17, 18, 21e23, 28) were evaluated also in cAMP
production in hA2BCHO cells as hA2B AR antagonists at 1 mM con-
centration in the presence of NECA 100 nM. Moreover, their affin-
ities at hA1, hA2A, and hA3 ARs, stably transfected in CHO cells, were
measured. All pharmacological data are presented in Table 1.
2.3. Structure-activity relationships
The pharmacological data for the newly synthesized amino-3,5-
dicyanopyridine derivatives 3e28 are reported in Table 1 together
with those of the reference compound 1 [13,21]. Most of the
compounds have generally low to null hA2A and hA3 AR affinity
while the binding at the hA1 AR subtype depends strictly on both R1
and R2 substituents. All the derivatives that interact with the hA2B
AR (5, 8e10, 13e16, 19, 20, 25e27) display EC50 values from 9.5 to
347 nM behaving as partial agonists, with the only exception being
compound 8 [13] which shows a full agonist profile. Moreover,
derivative 15 that merges the typical R1 and R2 substituents of
compound 1 and series 2, respectively (Chart 1), turns out to be the
most potent hA2B receptor agonist among this series
(EC50 ¼ 9.5 nM) and is also endowed with good selectivity versus
the other ARs (25-fold vs hA1 AR, 80-fold vs hA2A AR, and 50-fold vs
hA3 AR). Some compounds (3, 12, 17, 18, 21e23 and 28), selected
among those that had no activity at the hA2B AR, were evaluated
also as hA2B AR antagonists revealing the inability to inhibit NECA-
stimulated cAMP levels.
Furthermore, selecting the best R1 and R2 substituents in terms
of hA2B AR activity, compounds 25e27 were synthesized bearing
the 4-(para-ethyloxy) group as a common feature. The potency
value of compound 26 confirms the favourable effect exerted by the
imidazolyl moiety on hA2B AR interaction making this compound as
potent as derivative 15. In contrast, 26 also shows high hA1 AR af-
finity, thus suggesting the cyclopropylmethyloxy substituent as
suitable for obtaining potent hA2B AR agonists endowed with better
selectivity versus the other ARs. The same positive effect on hA1 AR
binding is produced by the pyrazolyl group which increases the
affinity for this subtype in both derivatives 27 and 28 with respect
to the corresponding 2-sulfanylacetamido compounds 5 and 6 [13].
Finally, the bicyclic compound 52, which was originated by
intramolecular cyclization of the parent compound 1, does not bind
any of the ARs thus suggesting that this kind of molecular
complication, which makes the structure more rigid, is detrimental
for the profitable interaction with hARs and in particular with the
A
2B AR subtype.
First, to explore how slight modifications of the para-cyclo-
propylmethyloxy group of the lead compound 1 could influence
hA2B AR activity and selectivity, 2-sulfanylacetamido-derivatives
3e7 bearing different cycloalkyl- or cycloalkenyl-methyloxy sub-
stituents at R1 position were synthesized. These compounds totally
lose hA2B AR activity except for the para-ethoxy-substituted de-
rivative 5 which is as potent as 1. In contrast, the 4-(para-acet-
amidophenyl)-substituted derivative 8 is endowed with good
activity at the hA2B AR despite a low selectivity versus hA2A and hA3
AR subtypes. The latter is the only amino-3,5-dicyanopyridine in
the whole series having a full agonist profile at the hA2B AR. The
para-acetamido substituent, in addition to possessing an oxygen
atom as H-bond acceptor, also contains the NH donor group that
2.4. Molecular modelling studies
To simulate the binding mode of nucleoside and non-nucleoside
agonists at hA2B AR, molecular docking studies were performed on
homology models of hA2B AR developed using two crystal struc-
tures of agonist-bound hA2A AR [43,44] as templates (pdb code:
2YDO; 3.0 Å resolution [45] and pdb code: 3QAK; 2.7 Å resolution
[46], in complex with Ado and 6-(2,2-diphenylethylamino)-9-
[(2R,3R,4S,5S)-5-(ethylcarbamoyl)-3,4-dihydroxytetrahydrofuran-
2-yl]eN-{2-[3-(1-(pyridin-2-yl)piperidin-4-yl)ureido]ethyl}-9H-
purine-2-carboxamide, also named UK-432097, respectively). The
obtained hA2B AR homology models were checked using the

130
M. Betti et al. / European Journal of Medicinal Chemistry 150 (2018) 127e139
Table 1
Binding Affinity (Ki) at hA1, hA2A and hA3 ARs and potencies (EC50) at hA2B ARs.
R1
R2
cAMP assays
EC50 (nM)b
hA2B
Binding experimentsa
Ki (nM) or I%
Efficacy,c
hA2B
%
d
e
f
hA1
hA2A
5%
hA3
1%
3
4
CONH2
CONH2
e
5%
1%
e
1%
603 52
307 27
9%
5
CONH2
CONH2
38
3
66%
11%
345 27
323 25
1%
20%
6g
e
18%
444 39
7
CONH2
CONH2
e
2%
872 81
17%
31%
4%
8g
164 12
92%
572 51
742 67
9
CONHCH3
CONHCH2OH
CONHOH
COOCH3
94
8
63%
57%
20%
1%
14%
1%
1%
1%
1%
25%
3%
8%
31%
1%
1%
10
11
12
13
182 14
10%
e
536 49
23%
e
12.7 1.1
69%
83
7
14
15
347 31
9.5 0.9
27%
70%
66
5
1%
4%
235 24
1%
764 72
474 45
16
17
18
139 11
50%
7%
7%
1%
6%
1%
e
e
483 33
21%
14%
6%
12%
19
20
21
22
84
51
e
6
4
48%
50%
25%
17%
552 54
338 31
3%
2%
1%
1%
1%
6%
1%
1%
1%
e
140 12

M. Betti et al. / European Journal of Medicinal Chemistry 150 (2018) 127e139
131
Table 1 (continued )
R1
R2
cAMP assays
EC50 (nM)b
hA2B
Binding experimentsa
Efficacy,c
hA2B
%
Ki (nM) or I%
d
e
f
hA1
hA2A
1%
hA3
1%
23
24
e
1%
12%
e
1%
67
7
424 37
19%
25
26
CONHCH3
211 17
48%
62%
104
8
8%
834 74
11.7 1.2
8.2 0.7
221 19
85
6
27
28
332 28
34%
7%
21
44
2
4
4%
8%
18%
13%
e
e
52
e
e
1%
100%
19%
31%
3%
2%
13%
8%
1g,h
CONH2
31
3
a
Ki values are means SEM of four separate assays each performed in triplicate. Percentage of inhibition (I%) is determined at 1
EC50 values are means SEM of four separate assays each performed in triplicate.
mM concentration of the tested compounds.
b
c
Efficacy of the tested compound at 1
m
M concentration, in comparison with NECA (1
m
M ¼ 100%).
d
e
f
Displacement of specific [3H]DPCPX competition binding to hA1CHO cells.
Displacement of specific [3H]ZM241385 competition binding to hA2ACHO cells.
Displacement of specific [125I]AB-MECA competition binding to hA3CHO cells.
Reference [13].
g
h
Reference [21].
Protein Geometry Monitor application within MOE (Molecular
Operating Environment 2014.09) [47], with inspection of the
structural quality of the protein models (backbone bond lengths,
loop (EL) 2 segment and making an H-bond interaction with the N6-
amino group of Ado in the 2YDO crystal structure. This residue is
differently oriented in the 3QAK X-ray. For the non-nucleoside
agonist binding, the role of this amino acid does not appear to be
critical, according to mutagenesis studies performed at A2A AR [52].
The different arrangements of this residue can slightly change the
space available at the entrance of the binding cavity, but have a
marginal effect on the size and chemical-physical properties of the
binding site. This could explain why we observed analogous results
by comparing the obtained docking conformations at the two
cavities. The Supporting Information section contains a figure
showing the two hA2B AR model pockets with the docked com-
pound 15.
For the sake of clarity, in this section, the position numbering of
the substituents on the pyridine nucleus is defined as in compound
15. Thus, starting from the N1 position, it is assumed that the amino
and sulfanyl groups occupy positions 2 and 6, respectively. The
simulated binding mode associated to the best score, in the great
majority of cases presents the compounds oriented similarly to the
one previously described for analogue agonists at hARs (Fig. 1A)
[52,53].
In detail, the pyridine scaffold is located in the hA2B AR cavity in
correspondence to the purine moiety of Ado or UK-432097
observed from hA2A AR templates. The heterocyclic core is stabi-
lized within the cavity by interaction with Phe173 (EL2) and
Ile2767.39. Both the 3-cyano and the 2-amino groups make a polar
interaction with the amide function of Asn2546.55. Considering the
2YDO-based hA2B AR model, the 2-amino substituent gives a H-
bond interaction with Glu174, while in the case of the 3QAK-based
model such interaction is not present due to a different arrange-
ment of the glutamate. In some cases, the N1 atom makes a polar
interaction with Lys269 (EL3). The 5-cyano group on the pyridine
core gets in proximity of His2807.43 and transmembrane (TM) 2
angles and dihedrals, Ramachandran 4-
j dihedral plots, and
quality of side chain rotamer and non-bonded contact). The two
hA2B AR structures were then used as target for the docking anal-
ysis of the synthesized derivatives, whose structures were opti-
mized using RHF/AM1 semi-empirical calculations (using software
package MOPAC implemented in MOE) [48]. The docking studies
were performed by using the MOE docking tool and Gold and
Autodock software [49e51]. The MOE software analysis was made
by selecting the induced fit docking and optimization protocol
(Schematically, a preliminary docking analysis provides a set of
ligand conformations that are energy minimized, including in this
step the side chains of the receptor residues in proximity). For each
compound, the top-score docking pose at each hA2B AR model,
according to at least two out of three scoring functions, was
selected for final ligand-target interaction analysis.
We decided to employ two homology models of the hA2B AR to
consider slightly different arrangements of the binding cavities and
hence to better explore the conformational variability of the target
pocket. Homology models of the hA2B AR built on the same two
templates were used for previously reported docking studies with
non-nucleoside hAR agonists [51]; the 3QAK A2A AR crystal struc-
ture was used as target for docking studies of non-nucleoside ag-
onists of this receptor [52], as well as an X-ray structure of the same
protein very similar to the 2YDO structure employed as template in
this study (pdb code: 2YDV [45]).
As previously described [51], the binding cavities of the two
hA2B AR models are similar, considering both receptor residues
orientation and pocket volumes. The differences are due to diverse
arrangements of some residues, like a glutamate residue (Glu174,
corresponding to Glu169 in hA2A AR) located within extracellular

132
M. Betti et al. / European Journal of Medicinal Chemistry 150 (2018) 127e139
Fig. 1. Docking conformation of compound 15 at the hA2B AR receptor model (the 2YDO-based one). Global (A) and top (B) view are displayed. Key residues for ligand-target
interaction are indicated.
residues. As previously suggested in analogue studies at the hA2A
AR [52], the interaction with these amino acids could be mediated
by a bridge water molecule or by a protonated state of His2807.43 in
which its polar hydrogen could be oriented toward the 3-cyano
group. The 6-sulfanyl substituent (Fig. 1B) is located at the
entrance of the cavity getting close to residues of TM1, TM2, TM7
domains (Tyr101.35, Ser682.65, and Asn2737.36, respectively) and EL2
segment (Leu172 and Phe173). Considering the heterocycle-
containing 6-substituents, a series of potential interactions with
these residues may be observed (see compounds 15, 16, 19 and 20).
Fig. 1B shows the imidazol-2-yl moiety appended on the 6-sulfanyl
the imidazol-2-yl moiety with the polar hydrogen atom of the
backbone amino group of Phe173 and with the carbonyl group of
the Asn2737.36 side chain. Replacement of the 2-imidazolyl group
with a 4-imidazolyl substituent (compound 16) affords a decrease
of activity, probably due to the loss of the interaction with Phe173
polar hydrogen. Other modifications of this ring through the
introduction of other nitrogen-containing heterocycles (i.e. 19e21,
23) leads to a decrease or a loss of hA2B AR activity. The presence in
R2 of a phenyl ring substituted with an amide or an ester function
(compounds 13 and 14, respectively), still maintains activity at the
hA2B AR, confirming the importance of the presence of groups able
to provide H-bond interaction at this level. Compounds bringing an
amide function within the R2 substituent are generally endowed
with activity at the hA2B AR. This may be interpreted by considering
that this functional group mimics the combination of H-bond donor
group of 15. A partial
p-stacking interaction is observed with
Tyr101.35. The partial reduction of the imidazole ring of 15 (i.e.
compound 17) leads to a fall of hA2B AR activity, possibly due to the
loss of this
p-stacking interaction. Polar interactions are given by

M. Betti et al. / European Journal of Medicinal Chemistry 150 (2018) 127e139
133
and acceptor functions of the 2-imidazolyl ring and is able to pro-
3.1.1. General procedure for the synthesis of 6-substituted 2-Amino-
vide a double polar interaction with Phe173 and Asn2737.36 as
observed for compound 15. On the other hand, the activity of the
compounds is influenced by the structural and chemical properties
of the 4-substituent. In this docking arrangement, this group gets in
4-aryl-3,5-dicyanopyridine derivatives 1, 3e21, 23, 25e28 [13,21]
and 2-substituted 6-Amino-4-aryl-3,5-dicyanopyridines 22, 24
Sodium hydrogen carbonate (2.0 mmol) and the suitable
halomethyl-derivative (1.0 mmol) were consequentially added to a
solution of the sulfanyl compound 44e50 [13,37,38] (1.0 mmol) in
anhydrous DMF (1 mL). The reaction mixture was stirred at rt until
the disappearance of the starting material (TLC monitoring). At
reaction completion, water was added (25 mL) to precipitate a solid
which was collected by filtration and triturated with Et2O (5 mL).
The crude product was purified by column chromatography
(compounds 4, 7, 16, 26, 27), preparative TLC (17, 28) or recrystal-
lized (1, 3, 5, 6, 8e15, 18e25).
, , ,
proximity of Leu863.33 Thr893.36 Trp2476.48 His2516.52 and,
partially, Ser2797.42, and His2807.43. The volume of the substituent
(R1) at the para-position of the 4-phenyl ring appears a critical
feature for activity of the compounds, with a cyclopropylmethyloxy
function providing the best results at the A2B AR. Docking results
show this group located deep in the binding cavity within a narrow
hydrophobic sub-pocket that is suitable to accommodate the para-
cycloalkyl group. This observation helps to interpret the decrease or
loss of activity of compounds featuring a para-substituent with
higher hindrance than the cyclopropylmethyloxy one (i.e. com-
pound 3). Also, the presence of a smaller para-ethyloxy group
(compounds 5, 25e27) maintains the activity at the hA2B AR
although depending on the R2 substituents as observed in the case
3.1.1.1. 2-{[6-Amino-3,5-dicyano-4-(4-(cyclopropylmethoxy)phenyl)
pyridin-2-yl]thio}acetamide 1 [13,21]. Yield 40%; mp 219e220 ꢀC
(MeOH); 1H NMR (DMSO‑d6) 8.00 (br s, 2H, NH2), 7.45e7.55 (m, 3H,
Ar þ NH), 7.25 (s, 1H, NH), 7.10 (d, 2H, Ar, J ¼ 8.7 Hz), 3.91 (d, 2H,
OCH2, J ¼ 7.0 Hz), 3.88 (s, 2H, SCH2), 1.23e1.29 (m, 1H, CH),
0.58e0.62 (m, 2H, CHeq), 0.34e0.38 (m, 2H, CHax). Anal. Calcd for
of
the
corresponding
para-cyclopropylmethyloxyphenyl-
substituted derivatives 1, 9, 15, 19.
An alternative binding mode orients the compounds in a
different way with respect to the above described docking confor-
mations, and similarly to the one previously described in the
literature at this receptor [50,51,53]. This binding mode is reported
and described within the Supporting Information section.
C19H17N5O2S.
3.1.1.2. 2-{[6-Amino-3,5-dicyano-4-(4-(cyclobutylmethoxy)phenyl)
pyridin-2-yl]thio}acetamide 3. Yield 21%; mp 244e246 ꢀC (EtOAc/
cyclohexane); 1H NMR (DMSO‑d6) 7.99 (br s, 2H, NH2), 7.50e7.47
(m, 3H, Ar þ NH), 7.49 (d, 2H, Ar, J ¼ 8.6 Hz), 7.26 (s, 1H, NH), 7.10 (d,
2H, Ar, J ¼ 8.6 Hz), 4.04 (d, 2H, OCH2, J ¼ 6.7 Hz), 3.88 (s, 2H, SCH2),
2.78e2.71 (m, 1H, CH), 1.99e1.82 (m, 6H, 3CH2); IR 3664, 3361,
3252, 2214, 1635. Anal. Calcd for C20H19N5O2S.
2.5. Conclusion
The present study has led to the identification of some potent hA2B
AR ligands with a partial agonist profile, several of which are also
endowed with good selectivity towards the other AR subtypes. The 2-
[(1H-imidazol-2-yl)methylthio)]-6-amino-4-(4-cyclopropylmethoxy
-phenyl)pyridine-3,5-dicarbonitrile 15 emerged as the most inter-
esting compound being also 3-fold more active than the lead 1. This
result can be considered a real breakthrough due to the currently
limited number of non-Ado-like hA2B AR agonists reported in the
literature. The results of the docking studies at two hA2B AR homology
models allowed us to interpret the interaction features of the amino-
3,5-dicyanopyridine derivatives at this receptor, also providing useful
indications for the design of new hA2B AR ligands belonging to this
series.
3.1.1.3. 2-{[6-Amino-3,5-dicyano-4-(4-isobutoxyphenyl)pyridin-2-yl]
thio}acetamide 4. Yield 21%; column chromatography, eluting sys-
tem CH2Cl2/MeOH, 9.7:0.3; mp 233e235 ꢀC (EtOH); 1H NMR
(DMSO‑d6) 7.99 (br s, 2H, NH2), 7.49e7.47 (m, 3H, Ar þ NH), 7.25 (s,
1H, NH), 7.11 (d, 2H, Ar, J ¼ 8.8 Hz), 3.88 (s, 2H, SCH2), 3.84 (d, 2H,
OCH2, J ¼ 6.4 Hz), 2.08e2.02 (m, 1H, CH), 1.01 (d, 6H, 2CH3,
J ¼ 6.8 Hz); IR 3537, 3375, 3196, 2208, 1639. Anal. Calcd for
C19H19N5O2S.
3.1.1.4. 2-{[6-Amino-3,5-dicyano-4-(4-ethoxyphenyl)pyridin-2-yl]
thio}acetamide 5. Yield 35%; mp 235e237 ꢀC (EtOH); 1H NMR
(DMSO‑d6) 7.98 (br s, 2H, NH2), 7.50 (s, 1H, NH), 7.48 (d, 2H, Ar,
J ¼ 8.8 Hz), 7.23 (s, 1H, NH), 7.10 (d, 2H, Ar, J ¼ 8.8 Hz), 4.12 (q, 2H,
OCH2, J ¼ 7.2 Hz), 3.88 (s, 2H, SCH2), 1.36 (t, 3H, CH3, J ¼ 7.2 Hz); IR
3396, 3180, 2210, 1637. Anal. Calcd for C17H15N5O2S.
3. Experimental procedures
3.1. Chemistry
3.1.1.5. 2-{[4-(4-(Allyloxy)phenyl)-6-amino-3,5-dicyanopyridin-2-yl]
thio)}acetamide 6 [13]. Yield 67%; mp 204e206 ꢀC (EtOH); 1H NMR
(DMSO‑d6) 8.00 (s, 2H, NH2), 7.50 (m, 3H, Ar þ NH), 7.26 (s, 1H, NH),
7.14 (d, 2H, Ar, J ¼ 8.7 Hz), 5.98e6.18 (m, 1H, CH), 5.45 (dd, 1H, CH,
J ¼ 17.3, 1.4 Hz), 5.30 (dd, 1H, CH, J ¼ 10.5, 1.4 Hz), 4.67 (d, 2H, OCH2,
J ¼ 5.1 Hz), 3.89 (s, 2H, SCH2); IR 3325, 3215, 3184, 2212. Anal. Calcd
for C18H15N5O2S.
Analytical silica gel plates (Merck F254), preparative silica gel
plates (Merck F254, 2 mm) and silica gel 60 (Merck, 70e230 mesh)
were used for analytical and preparative TLC, and for column
chromatography, respectively. All melting points were determined
on a Gallenkamp melting point apparatus and are uncorrected.
Elemental analyses were performed with a Flash E1112 Thermo-
finnigan elemental analyzer for C, H, N and the results were within
0.4% of the theoretical values. All final compounds revealed purity
not less than 95%. The IR spectra were recorded with a Perkin-
Elmer Spectrum RX I spectrometer in Nujol mulls and are
expressed in cmꢁ1. NMR spectra were recorded on a Bruker Avance
400 spectrometer (400 MHz for 1H NMR and 100 MHz for 13C NMR).
3.1.1.6. 2-{[6-Amino-3,5-dicyano-4-(4-[(2-methylallyl)oxy]phenyl)
pyridin-2-yl]thio}acetamide 7. Yield 27%; column chromatography,
eluting system EtOAc/cyclohexane/MeOH, 5:5:1); mp 224e226 ꢀC
(EtOH); 1H NMR (DMSO‑d6) 7.99 (s, 2H, NH2), 7.51 (s, 1H, NH), 7.50
(d, 2H, Ar, J ¼ 8.6 Hz), 7.25 (s,1H, NH), 7.13 (d, 2H, Ar, J ¼ 8.6 Hz), 5.10
(s, 1H, CH), 4.99 (s, 1H, CH), 4.57 (s, 2H, OCH2), 3.88 (s, 2H, SCH2),
1.08 (s, 3H, CH3); IR 3481, 3334, 3234, 2212, 1681. Anal. Calcd for
The chemical shifts are reported in
d (ppm) and are relative to the
central peak of the residual nondeuterated solvent, which was
CDCl3 or DMSOd6. The following abbreviations are used: s ¼ singlet,
d ¼ doublet, t ¼ triplet, q ¼ quartet, m ¼ multiplet, br ¼ broad,
Ar ¼ aromatic protons, eq ¼ equatorial and ax ¼ axial.
C19H17N5O2S.
3.1.1.7. 2-{]4-(4-Acetamidophenyl)-6-amino-3,5-dicyanopyridin-2-
yl]thio}acetamide 8 [13]. Yield 19%; 279e281 ꢀC (MeOH); 1H NMR

134
M. Betti et al. / European Journal of Medicinal Chemistry 150 (2018) 127e139
(DMSO‑d6) 10.2 (br s, 1H, NH), 7.99 (br s, 2H, NH2), 7.74 (d, 2H, Ar,
J ¼ 7.6 Hz), 7.49e7.47 (m, 3H, Ar þ NH), 7.24 (s, 1H, NH), 3.89 (s, 2H,
SCH2), 2.09 (s, 3H, CH3); 13C NMR (DMSO‑d6): 169.38, 169.27,166.67,
160.10, 158.52, 141.70, 129.72, 128.44, 119.12, 115.82, 93.65, 86.29,
33.78, 24.57; IR 3419, 3367, 2208, 1647. Anal. Calcd for C17H14N6O2S.
1.23e1.29 (m, 1H, CH), 0.62e0.58 (m, 2H, CHeq), 0.39e0.35 (m, 2H,
CHax); 13C NMR (DMSO‑d6): 166.59, 160.74, 160.32, 158.58, 155.20,
143.02, 130.64, 126.03, 115.96, 115.91, 114.98, 93.64, 86.38, 72.75,
27.21, 10.53, 3.62; IR 3430, 3387, 2229. Anal. Calcd for C21H18N6OS.
3.1.1.15. 2-{[(1H-Imidazol-5-yl)methyl]thio}-6-amino-4-[4-(cyclo-
propylmethoxy)phenyl]pyridine-3,5-dicarbonitrile 16. Yield 32%;
column chromatography, eluting system CHCl3/MeOH, 9:1); mp
186e188 ꢀC (EtOH); 1H NMR (DMSO‑d6) 11.97 (s, 1H, NH), 8.04 (br s,
2H, NH2), 7.59 (s, 1H, Ar), 7.47 (d, 2H, Ar, J ¼ 8.5 Hz), 7.18 (s, 1H, Ar),
7.08 (d, 2H, Ar J ¼ 8.5 Hz), 4.40 (s, 2H, SCH2), 3.90 (d, 2H, OCH2,
J ¼ 6.9 Hz), 1.23e1.27 (m, 1H, CH), 0.63e0.57 (m, 2H, CHeq),
0.34e0.37 (m, 2H, CHax); 13C NMR (DMSO‑d6): 167.24, 160.67,
160.18, 158.49, 130.63, 126.16, 116.02, 114.95, 93.81, 86.05, 72.73,
10.53, 3.60; IR 3318, 3205, 3080, 2212. Anal. Calcd for C21H18N6OS.
3.1.1.8. 2-{[6-Amino-3,5-dicyano-4-(4-(cyclopropylmethoxy)phenyl)
pyridin-2-yl]thio}eN-methylacetamide
9. Yield
71%;
mp
274e276 ꢀC (Acetone); 1H NMR (DMSO‑d6) 8.20e7.90 (bm, 3H,
NH2 þ NH), 7.47 (d, 2H, Ar, J ¼ 8.7 Hz), 7.09 (d, 2H, Ar, J ¼ 8.7 Hz),
3.85e3.95 (m, 4H, SCH2 þ OCH2), 2.62 (d, 3H, CH3, J ¼ 4.6 Hz),
1.23e1.30 (m, 1H, CH), 0.57e0.61 (m, 2H, CHeq), 0.34e0.38 (m, 2H,
CHax); 13C NMR (DMSO‑d6): 167.79, 166.54, 160.76, 160.16, 158.59,
130.64, 126.04, 115.90, 115.00, 93.70, 86.24, 72.75, 33.73, 26.56,
10.53, 3.61. IR 3390, 3323, 3223, 2208. Anal. Calcd for C20H19N5O2S.
3.1.1.9. 2-{[6-Amino-3,5-dicyano-4-(4-(cyclopropylmethoxy)phenyl)
pyridin-2-yl]thio}eN-(hydroxymethyl)- acetamide 10. Yield 79%;
mp 202e204 ꢀC (MeOH); 1H NMR (DMSO‑d6) 8.69 (t, 1H, NH,
J ¼ 6.3 Hz); 7.8e8.2 (br s, 2H, NH2), 7.46e7.50 (m, 2H, Ar), 7.08e7.12
(m, 2H, Ar), 5.72 (t, 1H, OH, J ¼ 6.42 Hz), 4.53 (t, 2H, NCH2O,
J ¼ 6.00), 3.90e3.92 (m, 4H, 2CH2), 1.22e1.29 (m, 1H, CH),
0.58e0.62 (m, 2H, CHeq), 0.34e0.38 (m, 2H, CHax); IR 3394, 3319,
3224, 2208, 1639. Anal. Calcd for C20H19N5O3S.
3.1.1.16. 2-Amino-4-[4-(cyclopropylmethoxy)phenyl]-6-{[(4,5-
dihydro-1H-imidazol-2-yl)methyl]thio}pyridine-3,5-dicarbonitrile
17. Yield 37%; preparative TLC, eluent EtOAC/cyclohexane/MeOH,
6:3:1); mp 260e262 ꢀC (2-propanol/EtOAc); 1H NMR (DMSO‑d6)
7.40 (d, 2H, Ar J ¼ 8.5 Hz), 7.19 (s, 2H, NH2), 7.12 (d, 2H, Ar J ¼ 8.1 Hz),
6.44 (s, 1H, NH), 5.82 (s, 2H, SCH2), 3.92 (d, 2H, OCH2, J ¼ 7.1 Hz),
3.77 (t, 2H, CH2, J ¼ 9.3 Hz), 3.23 (t, 2H, CH2, J ¼ 10.4 Hz), 1.31e1.24
(m, 1H, CH), 0.68e0.58 (m, 2H, CHeq), 0.35e0.30 (m, 2H, CHax).
Anal. Calcd for C21H20N6OS.
3.1.1.10. 2-{[6-Amino-3,5-dicyano-4-(4-(cyclopropylmethoxy)
phenyl)pyridin-2-yl]thio}eN-hydroxyacetamide 11. Yield 60%; mp
176e178 ꢀC (EtOH); 1H NMR (DMSO‑d6) 10.58 (br s, 1H, OH), 9.06
(br s, 1H, NH), 8.01 (br s, 2H, NH2), 7.48 (d, 2H, Ar, J ¼ 8.6 Hz), 7.10 (d,
2H, Ar, J ¼ 8.6 Hz), 3.91 (d, 2H, SCH2, J ¼ 6.9 Hz), 3.81 (s, 2H, OCH2),
1.27e1.24 (m, 1H, CH), 0.58e0.62 (m, 2H, CHeq), 0.37e0.42 (m, 2H,
CHax); IR 3649, 3331, 3223, 2208. Anal. Calcd for C19H17N5O3S.
3.1.1.17. 2-{[(1H-Benzo[d]imidazol-2-yl)methyl]thio}-6-amino-4-[4-
(cyclopropylmethoxy)phenyl]pyridine-3,5-dicarbonitrile
18.
Yield 64%; mp > 300 ꢀC (MeOH); 1H NMR (DMSO‑d6) 12.3 (br s, 1H,
NH), 7.9e8.4 (br s, 2H, NH2), 7.57 (d, 1H, Ar, J ¼ 7.8 Hz), 7.46e7.48
(m, 3H, Ar), 7.15e7.20 (m, 2H, Ar), 7.09 (d, 2H, ar, J ¼ 8.7 Hz), 4.71 (s,
2H, CH2), 3.9 (d, 2H, OCH2, J ¼ 7.1 Hz), 1.22e1.30 (m, 1H, CH),
0.57e0.62 (m, 2H, CHeq), 0.33e0.37 (m, 2H, CHax); IR 3396, 3331,
2210. Anal. Calcd for C25H20N6OS.
3.1.1.11. Methyl 2-{[6-amino-3,5-dicyano-4-(4-(cyclopropylmethoxy)
phenyl)pyridin-2-yl]thio}acetate 12. Yield 41%; mp 205e207 ꢀC
(EtOH); 1H NMR (DMSO‑d6) 7.9 (br s, 2H, NH2), 7.48 (d, 2H, ar
J ¼ 8.8), 7.1 (d, 2H, ar, J ¼ 8.8), 4.20 (s, 2H, SCH2), 3.91 (d, 2H, OCH2,
J ¼ 7.0), 3.69 (s, 3H, OCH3), 1.24e1.28 (m, 1H, CH), 0.58e0.62 (m, 2H,
CHeq), 0.35e0.38 (m, 2H, CHax); IR 3446, 3342, 3224, 2210, 1739.
Anal. Calcd for C20H18N4O3S
3.1.1.18. 2-{[(1H-Pyrazol-5-yl)methyl]thio}-6-amino-4-[4-(cyclo-
propylmethoxy)phenyl]pyridine-3,5-dicarbonitrile 19. Yield 37%; mp
133e136 ꢀC (EtOH/cyclohexane); 1H NMR (DMSO‑d6) 12.7 (br s, 1H,
NH), 8.0 (br s, 2H, NH2), 7.67 (br s, 1H, Ar), 7.47 (d, 2H, Ar, J ¼ 8.3 Hz),
7.09 (d, 2H, Ar, J ¼ 8.5 Hz), 6.32 (s, 1H, Ar), 4.49 (s, 2H, SCH2), 3.91 (d,
2H, OCH2, J ¼ 6.9), 1.22e1.26 (m, 1H, CH), 0.58e0.62 (m, 2H, CHeq),
0.36e0.40 (m, 2H, CHax). Anal. Calcd for C21H18N6OS.
3.1.1.12. 3-{[(6-Amino-3,5-dicyano-4-[4-(cyclopropylmethoxy)
phenyl]pyridin-2-yl)thio]methyl}benzamide 13. Yield 47%; mp
243e245 ꢀC (EtOAc/2-methoxyethanol); 1H NMR (DMSO‑d6) 8.05
(br s, 2H, NH2), 8.00 (s, 1H, Ar), 7.96 (s, 1H, NH), 7.77 (d, 1H, Ar,
J ¼ 7.8 Hz), 7.69 (d, 1H, Ar, J ¼ 7.7 Hz), 7.47e7.38 (m, 4H, 3ar þ NH),
7.07 (d, 2H, Ar, J ¼ 8.8 Hz), 4.54 (s, 2H, SCH2), 3.90 (d, 2H, OCH2,
J ¼ 7.0), 1.23e1.28 (m, 1H, CH), 0.58e0.61 (m, 2H, CHeq), 0.33e0.37
(m, 2H, CHax); IR 3441, 3387, 3356, 3169, 2212, 1635. Anal. Calcd for
3.1.1.19. 2-{[(1H-1,2,4-Triazol-5-yl)methyl]thio}-6-amino-4-[4-
(cyclopropylmethoxy)phenyl]pyridine-3,5-dicarbonitrile
20.
Yield 72%; mp 240e242 ꢀC (EtOH/Et2O); 1H NMR (DMSO‑d6) 13.9
(br s, 1H, NH), 8.3 (br s, 1H, Ar), 8.10 (br s, 2H, NH2), 7.48 (d, 2H, Ar,
J ¼ 7.8 Hz), 7.09 (d, 2H, Ar, J ¼ 8.5 Hz), 4.59 (s, 2H, SCH2), 3.91 (d, 2H,
OCH2, J ¼ 7.1 Hz), 1.23e1.29 (m, 1H, CH), 0.57e0.62 (m, 2H, CHeq),
0.34e0.37 (m, 2H, CHax); IR 3441, 3323, 3207, 2212. Anal. Calcd for
C
25H21N5O2S.
C20H17N7OS.
3.1.1.13. Methyl
propylmethoxy)phenyl]pyridin-2-yl)thio]methyl}
3-{[(6-amino-3,5-dicyano-4-[4-(cyclo-
benzoate 14.
3.1.1.20. 2-{[(1H-Tetrazol-5-yl)methyl]thio}-6-amino-4-[4-(cyclo-
propylmethoxy)phenyl]pyridine-3,5-dicarbonitrile 21. Yield 64%; mp
225e227 ꢀC (EtOH/Et2O); 1H NMR (DMSO‑d6) 16.10 (br s, 1H, NH),
8.00 (br s, 2H, NH2), 7.47 (d, 2H, Ar, J ¼ 8.6 Hz), 7.09 (d, 2H, Ar,
J ¼ 8.6 Hz), 4.78 (s, 2H, SCH2), 3.91 (d, 2H, OCH2, J ¼ 7.0 Hz),
1.24e1.28 (m, 1H, CH), 0.58e0.61 (m, 2H, CHeq), 0.34e0.37 (m, 2H,
CHax); IR 3442, 3323, 3224, 2225, 2208. Anal. Calcd for C19H16N8OS.
Yield 58%; mp 158e160 ꢀC (MeOH); 1H NMR (DMSO‑d6) 8.10 (s, 1H,
ar), 7.90e8.35 (br s, 2H, NH2), 7.84e7.87 (m, 2H, Ar), 7.44e7.49 (m,
3H, Ar), 7.07 (d, 2H, ar, J ¼ 8.8 Hz), 4.59 (s, 2H, SCH2), 3.90 (d, 2H,
OCH2 J ¼ 7.04), 3.86 (s, 3H, OCH3), 1.23e1.27 (m, 1H, CH), 0.58e0.62
(m, 2H, CHeq), 0.34e0.37 (m, 2H, CHax); IR 3396, 3331, 3230, 2210,
1707. Anal. Calcd for C26H22N4O3S.
3.1.1.14. 2-{[(1H-Imidazol-2-yl)methyl]thio}-6-amino-4-[4-(cyclo-
propylmethoxy)phenyl]pyridine-3,5-dicarbonitrile 15. Yield 63%; mp
226e228 ꢀC (EtOH); 1H NMR (DMSO‑d6) 11.8 (s, 1H, NH), 8.07 (br s,
2H, NH2), 7.47 (d, 2H, Ar, J ¼ 8.2 Hz), 7.09 (d, 2H, Ar, J ¼ 8.2 Hz); 6.90
(br s, 2H, CH, Ar), 4.49 (s, 2H, SCH2), 3.90 (d, 2H, OCH2, J ¼ 8.0 Hz),
3.1.1.21. 2-Amino-6-{[(2-aminothiazol-4-yl)methyl]thio}-4-[4-
(cyclopropylmethoxy)phenyl]pyridine-3,5-dicarbonitrile
22.
Yield 87%; mp 238e240 ꢀC (EtOH/Et2O); 1H NMR (DMSO‑d6) 8.02
(br s, 2H, NH2), 7.46 (d, 2H, Ar, J ¼ 8.7 Hz), 7.08 (d, 2H, Ar, J ¼ 8.7 Hz),
6.98 (br s, 2H, NH2), 6.63 (s, 1H, Ar), 4.31 (s, 2H, SCH2), 3.90 (d, 2H,

M. Betti et al. / European Journal of Medicinal Chemistry 150 (2018) 127e139
135
OCH2, J ¼ 7.0 Hz), 1.23e1.29 (m, 1H, CH), 0.57e0.62 (m, 2H, CHeq),
0.34e0.37 (m, 2H, CHax); IR 3423, 3327, 3184e3170, 2216. Anal.
Calcd for C21H18N6OS2.
added, followed by the cyclobutylmethyl bromide (18.5 mmol). The
mixture was heated at reflux for 36 h. Then, the mixture was cooled
to rt and the insoluble material was filtered and washed with
acetone (3 ꢂ 20 mL). The resulting filtrates were collected and
evaporated under vacuum, affording an oily residue which was
dissolved in EtOAc (150 mL). The organic layer was washed with
25% NaOH solution (4 ꢂ 30 mL), with water (3 ꢂ 30 mL) and then
dried (Na2SO4). After evaporation of the solvent, the desired com-
pound was obtained as a viscous oil. The product was pure enough
to be used without further purification. Yield 67%; 1H NMR
(DMSO‑d6) 9.86 (s, 1H, CHO), 7.85 (d, 2H, Ar, J ¼ 8.8 Hz), 7.11 (d, 2H,
Ar, J ¼ 8.4 Hz), 4.06 (d, 2H, OCH2, J ¼ 6.8 Hz), 2.77e2.71 (m, 1H, CH),
2.08e2.07 (m, 2H, CH2), 1.90e1.82 (m, 4H, 2CH2).
3.1.1.22. 2-{[(1H-1,2,4-Triazol-1-yl)methyl]thio}-6-amino-4-[4-
(cyclopropylmethoxy)phenyl]pyridine-3,5-dicarbonitrile
23.
Yield 40%; mp 235e237 ꢀC (EtOH); 1H NMR (DMSO‑d6) 8.99 (s, 1H,
Ar), 8.75e7.78 (m, 2H, NH2), 8.04 (s, 1H, Ar), 7.47 (d, 2H, Ar,
J ¼ 8.7 Hz), 7.08 (d, 2H, Ar, J ¼ 8.8 Hz), 6.03 (s, 2H, CH2), 3.90 (d, 2H,
CH2, J ¼ 7.1 Hz), 1.32e1.24 (m, 1H, CH), 0.64e0.54 (m, 2H, CHeq),
0.38e0.32 (m, 2H CHax); IR 3311, 2212. Anal. Calcd for C20H17N7OS.
3.1.1.23. 2-Amino-4-[4-(cyclopropylmethoxy)phenyl]-6-[(pyridin-3-
ylmethyl)thio]pyridine-3,5-dicarbonitrile
24. Yield
80%;
mp
190e192 ꢀC (MeOH); 1H NMR (DMSO‑d6) 8.7 (s, 1H, Ar), 8.45 (s, 1H,
Ar), 8.11 (br s, 2H, NH2), 7.94 (d, 1H, Ar, J ¼ 5.7 Hz), 7.42 (d, 2H, Ar,
J ¼ 6.8 Hz), 7.35e7.37 (m, 1H, Ar), 7.05 (d, 2H, ar, J ¼ 6.9 Hz), 4.48 (s,
2H, SCH2), 3.86 (d, 2H, OCH2, J ¼ 7.0 Hz), 1.19e1.25 (m, 1H, CH),
0.56e0.59 (m, 2H, CH eq), 0.30e0.34 (m, 2H, CH ax); IR 3369, 2212.
Anal. Calcd for C23H19N5OS.
3.1.3. N-[4-(2,2-Dicyanovinyl)phenyl]acetamide 36 [37]
Malononitrile (17.3 mmol) and piperidine (two drops) were
added to
a
solution of commercially available 4-
acetamidobenzaldehyde 34 (17.3 mmol) in EtOH (20 mL). The
mixture was heated at reflux for 2 h, then cooled to rt affording an
orange solid which was filtered, washed with Et2O (5 mL) and pe-
troleum ether (2 mL), and recrystallized. Yield 63%; mp 234e236 ꢀC
(EtOH); 1H NMR (DMSO‑d6) 10.51 (s, 1H, NH), 8.37 (s, 1H, CH), 7.94
(d, 2H, Ar, J ¼ 8.4 Hz), 7.80 (d, 2H, Ar, J ¼ 8.8 Hz), 2.11 (s, 3H, CH3); IR
2224, 1693. Anal. Calcd for C12H9N3O.
3.1.1.24. 2-{[6-Amino-3,5-dicyano-4-(4-ethoxyphenyl)pyridin-2-yl]
thio}eN-methylacetamide 25. Yield 32%; mp 262e264 ꢀC (EtOH);
1H NMR (DMSO‑d6) 8.1e7.9 (m, 3H, NH þ NH2), 7.48 (d, 2H, Ar,
J ¼ 8.5 Hz), 7.09 (d, 2H, Ar, J ¼ 8.4 Hz), 4.12 (q, 2H, OCH2, J ¼ 7.2 Hz),
3.88 (s, 2H, SCH2), 2.61 (d, 3H, CH3, J ¼ 4.6 Hz), 1.36 (t, 3H, CH3,
J ¼ 6.8 Hz); 13C NMR (DMSO‑d6): 167.78, 166.54, 160.15, 158.57,
130.67, 126.05, 115.93, 114.94, 93.63, 86.73, 63.83, 33.71, 26.56,
15.07; IR 3394, 3223, 2210, 1637. Anal. Calcd for C18H17N5O2S.
3.1.4. General procedure for the synthesis of 2-Amino-4-aryl-6-
(phenylthio)pyridine-3,5-dicarbonitriles 37e41, 43
To a stirred solution of the suitable aldehyde 29 [35], 30e33, 35
(10 mmol), malononitrile (20 mmol) and DBU (5% mol) in 10%
aqueous EtOH (20 mL), thiophenol (10 mmol) was added after
20 min. Thus, the reaction mixture was stirred at 55 ꢀC until the
disappearance of the starting material (TLC monitoring). The sus-
pension was then cooled to rt and a solid precipitated which was
collected by filtration, washed with Et2O (5e10 mL) and then dried
in oven at 60 ꢀC overnight. The crude products were purified by
crystallization.
3.1.1.25. 2-{[(1H-Imidazol-2-yl)methyl]thio}-6-amino-4-(4-
ethoxyphenyl)pyridine-3,5-dicarbonitrile 26. Yield 41%; column
chromatography, eluting system EtOAc/cyclohexane/MeOH, 6:3:1);
mp 242e244 ꢀC (EtOH); 1H NMR (DMSO‑d6) 11.84 (s, 1H, NH), 8.1
(br s, 2H, NH2), 7.47 (d, 2H, Ar, J ¼ 8.6 Hz), 7.10e7.08 (m, 3H, Ar), 6.96
(s, 1H, Ar), 4.50 (s, 2H, SCH2), 4.10 (q, 2H, OCH2, J ¼ 6.8 Hz), 1.36 (t,
3H, CH3, J ¼ 6.8 Hz); 13C NMR (DMSO‑d6): 166.60, 160.65, 160.32,
158.56, 143.01, 130.67, 126.06, 115.94, 115.89, 114.94, 93.66, 86.40,
63.83, 27.23, 15.06; IR 3506, 3398, 3159, 2212. Anal. Calcd for
3.1.4.1. 2-Amino-4-[4-(cyclobutylmethoxy)phenyl]-6-(phenylthio)
pyridine-3,5-dicarbonitrile 37. Yield 18%; mp 180e182 ꢀC (MeOH);
1H NMR (DMSO‑d6) 7.7 (br s, 2H, NH2), 7.61e7.60 (m, 2H, Ar),
7.51e7.49 (m, 5H, Ar), 7.12 (d, 2H, Ar, J ¼ 8.6 Hz), 4.05 (d, 2H, OCH2,
J ¼ 6.5 Hz), 2.77e2.75 (m, 1H, CH), 2.11e2.09 (m, 2H, CH2),
1.93e1.89 (m, 4H, 2CH2); IR 3348, 3280, 2216. Anal. Calcd for
C
19H16N6OS.
3.1.1.26. 2-{[(1H-Pyrazol-5-yl)methyl]thio}-6-amino-4-(4-
ethoxyphenyl)pyridine-3,5-dicarbonitrile 27. Yield 29%; column
chromatography, eluting system EtOAc/cyclohexane/MeOH, 5:5:1);
mp 195e197 ꢀC (EtOH); 1H NMR (DMSO‑d6) 12.72 (s, 1H, NH), 8.05
(BR s, 2H, NH2), 7.65 (s, 1H, Ar), 7.47 (d, 2H, Ar, J ¼ 8.3 Hz), 7.08 (d,
2H, Ar, J ¼ 8.4 Hz), 6.31 (s, 1H, Ar), 4.48 (s, 2H, SCH2), 4.11 (q, 2H,
OCH2, J ¼ 6.8 Hz), 1.36 (t, 3H, CH3, J ¼ 6.8 Hz); IR 3456, 3319, 3215,
2218. Anal. Calcd for C19H16N6OS.
C24H20N4OS.
3.1.4.2. 2-Amino-4-(4-isobutoxyphenyl)-6-(phenylthio)pyridine-3,5-
dicarbonitrile 38. Yield 26%; mp 181e183 ꢀC (MeOH); 1H NMR
(DMSO‑d6) 7.74 (s, 2H, NH2), 7.61e7.59 (m, 2H, ar), 7.50e7.49 (m,
5H, Ar), 7.11 (d, 2H, Ar, J ¼ 8.8 Hz), 3.84 (d, 2H, OCH2, J ¼ 6.4 Hz),
2.09e2.02 (m, 1H, CH), 1.02 (s, 3H, CH3), 1.00 (s, 3H, CH3); IR 3302,
3232, 2214. Anal. Calcd for C23H20N4OS.
3.1.1.27. 2-{[(1H-Pyrazol-5-yl)methyl]thio}-4-(4-(allyloxy)phenyl)-6-
aminopyridine-3,5-dicarbonitrile 28. Yield 65%; preparative TLC,
eluent CH2Cl2/MeOH, 9.5:0.5); mp 129e131 ꢀC (EtOH); 1H NMR
(DMSO‑d6): 12.71 (br s, 1H, NH), 8.08 (br s, 2H, NH2), 7.59 (br s, 1H,
Ar), 7.49 (d, 2H, Ar, J ¼ 8.7 Hz), 7.12 (d, 2H, Ar, J ¼ 8.7 Hz), 6.32 (d, 1H,
Ar, J ¼ 2 Hz), 6.13e6.03 (m, 1H, CH), 5.44 (dd, 1H, CH, J ¼ 17.3,
1.6 Hz), 5.30 (d, 1H, CH, J ¼ 10.4 Hz), 4.66 (d, 2H, OCH2, J ¼ 5.2 Hz),
4.49 (s, 2H, SCH2); 13C NMR (DMSO‑d6): 160.25, 160.23, 158.49,
133.86, 130.67, 128.81, 126.41, 125.65, 118.35, 115.95, 115.18, 111.08,
109.95, 105.12, 93.71, 86.31, 68.84; IR 3335, 3179, 2210. Anal. Calcd
for C20H16N6OS.
3.1.4.3. 2-Amino-4-(4-ethoxyphenyl)-6-(phenylthio)pyridine-3,5-
dicarbonitrile 39. Yield 37%; mp 252e254 ꢀC (2-propanol); 1H NMR
(DMSO‑d6) 7.60 (s, 2H, NH2), 7.62e7.60 (m, 2H, Ar), 7.58e7.48 (m,
5H, Ar), 7.11 (d, 2H, Ar, J ¼ 8.8 Hz), 4.13 (q, 2H, OCH2, J ¼ 6.8 Hz), 1.37
(t, 3H, CH3, J ¼ 7.2 Hz); IR 3344, 3215, 2216. Anal. Calcd for
C21H16N4OS.
3.1.4.4. 4-[4-(Allyloxy)phenyl]-2-amino-6-(phenylthio)pyridine-3,5-
dicarbonitrile 40. Yield 28%; mp 234e236 ꢀC (EtOAc/cyclohexane);
1H NMR (DMSO‑d6): 7.77 (s, 2H, NH2), 7.61e7.60 (m, 2H, Ar),
7,59e7.58 (m, 3H, Ar), 7.51 (d, 2H, Ar, J ¼ 8.8 Hz), 7.14 (d, 2H, Ar,
J ¼ 8.8 Hz), 6.13e6.03 (m, 1H, CH), 5.45 (dd, 1H, CH, J ¼ 17.2, 1.6 Hz),
5.30 (d, 1H, CH, J ¼ 10.5 Hz), 4.67 (d, 2H, OCH2, J ¼ 5.6 Hz); IR 3358,
3.1.2. 4-(Cyclobutylmethoxy)benzaldehyde 29 [35]
To a solution of 4-hydroxybenzaldehyde (12.3 mmol) in anhy-
drous acetone (20 mL), potassium carbonate (18.5 mmol) was

136
M. Betti et al. / European Journal of Medicinal Chemistry 150 (2018) 127e139
3215, 2214, 1624. Anal. Calcd for C22H16N4OS.
(d, 2H, OCH2, J ¼ 5.6 Hz); IR 3476, 3319, 3209, 2212. Anal. Calcd for
16H12N4OS.
C
3.1.4.5. 2-Amino-4-{4-[(2-methylallyl)oxy]phenyl}-6-(phenylthio)
pyridine-3,5-dicarbonitrile 41. Yield 25%; mp 198e200 ꢀC (EtOH);
1H NMR (DMSO‑d6) 7.76 (s, 2H, NH2), 7.60e7.61 (m, 2H, ar),
7.60e7.59 (m, 3H, Ar), 7.55 (d, 2H, Ar, J ¼ 8.4 Hz), 7.14 (d, 2H, Ar,
J ¼ 8.4 Hz), 5.11 (d,1H, CH), 5.00 (d,1H, CH), 4.58 (s, 2H, CH2),1.81 (s,
3H, CH3); IR 3481, 3369, 2212. Anal. Calcd for C23H18N4OS.
3.1.6.5. 2-Amino-6-mercapto-4-{4-[(2-methylallyl)oxy]phenyl}pyri-
dine-3,5-dicarbonitrile 48. Yield 72%; mp 180e183 ꢀC dec. (EtOH);
1H NMR (DMSO‑d6) 12.98 (s, 1H, SH), 8.11 (br s, 2H, NH2), 7.53 (d,
2H, Ar, J ¼ 7.4 Hz), 7.08 (d, 2H, Ar, J ¼ 8.8), 5.09 (s, 1H, CH), 4.90 (s,
1H, CH), 4.58 (s, 2H, OCH2), 1.80 (s, 3H, CH3). Anal. Calcd for
C17H14N4OS.
3.1.4.6. 2-Amino-4-[4-(cyclopropylmethoxy)phenyl]-6-(phenylthio)
pyridine-3,5-dicarbonitrile 43. Yield 33%; mp 261e263 ꢀC (EtOH);
1H NMR (DMSO‑d6) 7.70 (br s, 2H, NH2), 7.60e7.62 (m, 2H, Ar),
7.48e7.51 (m, 5H, Ar), 7.10 (d, 2H, Ar, J ¼ 8.1 Hz), 3.92 (d, 2H, CH2,
J ¼ 7.8 Hz), 1.25e1.28 (m, 1H, CH), 0.58e0.62 (m, 2H, 2CHeq),
0.36e0.38 (m, 2H, 2CHax); IR 3330, 3220, 2224. Anal. Calcd for
3.1.6.6. N-[4-(2-amino-3,5-dicyano-6-mercaptopyridin-4-yl)phenyl]
acetamide 49 [37]. Yield 40%; mp 275e277 ꢀC dec. (EtOH/EtOAc);
1H NMR (DMSO‑d6) 12.94 (s, 1H, SH), 10.37 (s, 1H, NH), 8.19 (s, 2H,
NH2), 7.73 (d, 2H, Ar, J ¼ 8.0 Hz), 7.44 (d, 2H, Ar, J ¼ 8.0 Hz), 2.08 (s,
3H, CH3); IR 3317, 3215, 2214, 1539. Anal. Calcd for C15H11N5OS.
C
23H18N4OS.
3.1.6.7. 2-Amino-4-[4-(cyclopropylmethoxy)phenyl]-6-
mercaptopyridine-3,5-dicarbonitrile 50. Yield 80%; mp 168e170 ꢀC
(EtOAc); 1H NMR (DMSO‑d6) 12.9 (br s, 1H, SH), 7.8 (br s, 2H, NH2),
7.45 (d, 2H, Ar, J ¼ 8 Hz), 7.08 (d, 2H, Ar, J ¼ 8 Hz), 3.91 (d, 2H, CH2,
J ¼ 7.8 Hz), 1.23e1.27 (m, 1H, CH), 0.57e0.62 (m, 2H, CHeq),
0.33e0.37 (m, 2H, CHax); IR 3309, 3190, 2219. Anal. Calcd for
3.1.5. N-{4-[2-Amino-3,5-dicyano-6-(phenylthio)pyridin-4-yl]
phenyl}acetamide 42 [37]
A solution of compound 36 [37] (5.4 mmol), malononitrile
(5.4 mmol), thiophenol (5.4 mmol) and Et3N (0.54 mmol) in EtOH
(30 mL) was heated at reflux for 2 h. Then, after cooling to rt a
yellow solid was obtained which was filtered and washed with Et2O
(5 mL). Yield 44%; mp > 300 ꢀC (EtOH); 1H NMR (DMSO‑d6) 10.24 (s,
1H, NH), 7.76 (s, 2H, NH2), 7.74 (m, 2H, Ar), 7.61e7.60 (m, 5H, Ar),
7.50 (d, 2H, Ar, J ¼ 8.0 Hz), 2.10 (s, 3H, CH3); IR 3481, 3344, 3298,
2212, 1681. Anal. Calcd for C21H15N5OS.
C17H14N4OS.
3.1.7. 2-Chloro-N-hydroxyacetamide 51 [39]
An aqueous 50% solution of hydroxylamine (12.0 mmol) was
added to ethyl chloroacetate (12.0 mmol) in a 50 mL round
bottomed flask and stirred for 5 min at rt. Then, the mixture was
cooled at 4 ꢀC and stirred for 15 min, and finally, re-warmed to rt
and stirred for 3 h. The resulting precipitate was collected by
filtration and used for the next step without any further purifica-
tion. Yield 53%; mp 93e95 ꢀC (Lit mp 92e93 ꢀC); 1H NMR (DMSO):
7.41 (br s, 2H, OH þ NH), 3.94 (s, 2H, CH2).
3.1.6. General procedure for the synthesis of 2-Amino-4-aryl-6-
mercaptopyridine-3,5-dicarbonitriles 44e50
To a stirred solution of the suitable 6-phenylsulfanyl derivative
37e43 (10 mmol) in anhydrous DMF (1 mL) maintained at rt and
under nitrogen atmosphere, an excess of sodium sulfite (33 mmol)
was added. The reaction mixture was heated at 80 ꢀC for 2 h. Then,
1 N HCl (25 mL) was added followed by addition of 6 N HCl (5 mL) to
obtain a huge precipitate which was filtered and washed with
water (20 mL) and Et2O (5 mL). The crude derivatives were purified
by crystallization.
3.1.8. 3,6-Diamino-5-cyano-4-[4-(cyclopropylmethoxy)phenyl]
thieno [2,3-b]pyridine-2-carboxamide 52
A solution of compound 1 [13,21] (0.71 mmol) and potassium
hydroxide (1.42 mmol) in absolute EtOH (10 mL) was refluxed for
3 h. Then, iced-water (30 mL) was added and the mixture was
neutralized with 5 N HCl. The green solid was collected by filtration,
washed with water (10 mL) and with Et2O (5 mL), and then
recrystallized. Yield 85%; mp 260e262 ꢀC (Dioxane); 1H NMR
(DMSO‑d6) 7.39 (d, 2H, Ar, J ¼ 8.0 Hz), 7.27 (br s, 2H, NH2), 7.12 (d,
2H, Ar, J ¼ 8.4 Hz), 6.96 (br s, 2H, NH2), 5.70 (s, 2H, NH2), 3.91 (d, 2H,
OCH2, J ¼ 6.8 Hz), 1.30e1.23 (m, 1H, CH), 0.62e0.58 (m, 2H, CHeq),
0.38e0.34 (m, 2H, CHax); 13C NMR (DMSO‑d6): 167.23, 163.83,
160.08, 158.93, 152.76, 146.75, 130.06, 125.74, 116.43, 115.29, 114.78,
93.35, 90.92, 72.71, 10.57, 3.62; IR 3466, 3452, 3346, 3315, 3136,
2214, 1629. Anal. Calcd for C19H17N5O2S.
3.1. 6.1. 2-Amino-4-[4-(cyclobutylmethoxy)phenyl]-6-
mercaptopyridine-3,5-dicarbonitrile 44. Yield 84%; mp 256e259 ꢀC
(EtOH); 1H NMR (DMSO‑d6) 12.95 (s, 1H, SH), 7.08 (s, 2H, NH2), 7.46
(d, 2H, Ar, J ¼ 8.0 Hz), 7.09 (d, 2H, Ar, J ¼ 8.4 Hz), 4.04 (d, 2H, OCH2,
J ¼ 6 Hz), 2.87e2.73 (m, 1H, CH), 2.10e1.92 (m, 2H, CH2), 1.90e1.89
(m, 4H, 2CH2); IR 3474, 3335, 2214. Anal. Calcd for C18H16N4OS.
3.1.6.2. 2-Amino-4-(4-isobutoxyphenyl)-6-mercaptopyridine-3,5-
dicarbonitrile 45. Yield 86%; mp 210e213 ꢀC (EtOH); 1H NMR
(DMSO‑d6) 13.02 (s, 1H, SH), 7.77 (s, 2H, NH2), 7.45 (d, 2H, Ar,
J ¼ 8.8 Hz), 7.09 (d, 2H, Ar, J ¼ 8.4 Hz), 3.83 (d, 2H, CH2, J ¼ 6.4 Hz),
2.06e2.03 (m, 1H, CH), 1.02 (s, 3H, CH3), 0.99 (s, 3H, CH3). Anal.
Calcd for C17H16N4OS.
3.2. Pharmacology
3.2.1. Cell culture and membrane preparation
3.1.6.3. 2-Amino-4-(4-ethoxyphenyl)-6-mercaptopyridine-3,5-
dicarbonitrile 46 [38]. Yield 76%; mp: 290e292 ꢀC (EtOH); 1H NMR
(DMSO‑d6) 12.99 (s, 1H, SH), 8.11 (s, 2H, NH2), 7.54 (d, 2H, Ar,
J ¼ 8.8 Hz), 7.14 (d, 2H, Ar, J ¼ 8.8 Hz), 4.14 (q, 2H, CH2, J ¼ 6.8 Hz),
1.38 (t, 3H, CH3, J ¼ 6.8 Hz); IR 3336, 3117, 2212. Anal. Calcd for
CHO cells transfected with hA1, hA2A, hA2B and hA3 ARs were
grown adherently and maintained in Dulbecco's modified Eagle's
medium with nutrient mixture F12, containing 10% fetal calf serum,
penicillin (100 U/ml), streptomycin (100 mg/mL), L-glutamine
(2 mM), geneticine (G418; 0.2 mg/mL) at 37 ꢀC in 5% CO2/95% air
until the use in cAMP assays [54]. For membrane preparation the
culture medium was removed, and the cells were washed with
phosphate-buffered saline and scraped off T75 flasks in ice-cold
hypotonic buffer (5 mM Tris HCl, 1 mM EDTA, pH 7.4). The cell
suspension was homogenized with a Polytron, centrifuged for
30 min at 40000 g at 4 ꢀC and the resulting membrane pellet was
used for competition binding experiments [54].
C
15H12N4OS.
3.1.6.4. 4-[4-(Allyloxy)phenyl]-2-amino-6-mercaptopyridine-3,5-
dicarbonitrile 47 [13]. Yield 84%; mp 208e211 ꢀC (EtOAc); 1H NMR
(DMSO‑d6) 12.98 (s, 1H, SH); 7.90 (s, 2H, NH2); 7.47 (d, 2H, Ar,
J ¼ 8.6 Hz); 7.12 (d, 2H, Ar, J ¼ 8.6 Hz); 6.11e6.04 (m, 1H, CH); 5.45
(dd, 1H, CH, J ¼ 17.2, 1.2 Hz), 5.30 (dd, 1H, CH, J ¼ 10.4, 1.2 Hz), 4.66

M. Betti et al. / European Journal of Medicinal Chemistry 150 (2018) 127e139
137
3.2.2. Competition binding experiments
All synthesized compounds have been tested for their affinity to
hA1, hA2A and hA3 ARs. Competition experiments to hA1 ARs were
Data Bank (pdb code: 2YDO; 3.0 Å resolution [45] and pdb code:
3QAK; 2.7 Å resolution [46], respectively). A multiple alignment of
the hAR primary sequences was built within MOE as preliminary
step. For all hA2B AR models, the boundaries identified from the
used X-ray crystal structure of hA2A AR were then applied for the
corresponding sequences of the TM helices of the hA2B AR. The
missing loop domains were built by the loop search method
implemented in MOE. Once the heavy atoms were modelled, all
hydrogen atoms were added, and the protein coordinates were
then minimized with MOE using the AMBER99 force field [58] until
the Root Mean Square (RMS) gradient of the potential energy was
less than 0.05 kJ molꢁ1 Åꢁ1. Reliability and quality of these models
were checked using the Protein Geometry Monitor application
within MOE, which provides a variety of stereochemical measure-
ments for inspection of the structural quality in a given protein, like
carried
out
incubating
1 nM
[3H]-8-cyclopentyl-1,3-
dipropylxanthine ([3H]-DPCPX) with membrane suspension
(50 mg of protein/100 ml) and different concentrations of the
examined compounds at 25 ꢀC for 90 min in 50 mM Tris HCl, pH 7.4.
Non-specific binding was defined as binding in the presence of
1 mM DPCPX and was always <10% of the total binding [54]. Inhi-
bition experiments to hA2A ARs were performed incubating the
radioligand [3H]-ZM241385 (1 nM) with the membrane suspension
(50 mg of protein/100 ml) and different concentrations of the
examined compounds for 60 min at 4 ꢀC in 50 mM Tris HCl (pH 7.4),
10 mM MgCl2. Non-specific binding was determined in the pres-
ence of ZM241385 (1
[55]. Competition binding experiments to A3 ARs were carried out
incubating the membrane suspension (50 g of protein/100 l)
with 0.5 nM
125I]-N6-(4-aminobenzyl)eN-methylcarbox-
mM) and was about 20% of the total binding
backbone bond lengths, angles and dihedrals, Ramachandran 4-
j
m
m
dihedral plots, and quality of side chain rotamer and non-bonded
contact.
[
amidoadenosine ([125I]-ABMECA) in the presence of different con-
centration of the examined compounds for an incubation time of
120 min at 4 ꢀC in 50 mM Tris HCl (pH 7.4), 10 mM MgCl2, 1 mM
EDTA. Non-specific binding was defined as binding in the presence
3.3.2. Molecular docking analysis
All compound structures were docked into the binding site of
the hA2B AR structures using three docking tools: the Induced Fit
docking protocol of MOE, the genetic algorithm docking tool of
CCDC Gold [49], and the Lamarckian genetic algorithm of Autodock
[50,51]. The Induced Fit docking protocol of MOE is divided into
several stages: Conformational Analysis of ligands. The algorithm
generated conformations from a single 3D conformation by con-
ducting a systematic search. In this way, all combinations of angles
were created for each ligand. Placement. A collection of poses was
generated from the pool of ligand conformations using Alpha Tri-
angle placement method. Poses were generated by superposition of
ligand atom triplets and triplet points in the receptor binding site.
The receptor site points are alpha sphere centres which represent
locations of tight packing. At each iteration, a random conformation
was selected, a random triplet of ligand atoms and a random triplet
of alpha sphere centres were used to determine the pose. Scoring.
Poses generated by the placement methodology were scored using
the Alpha HB scoring function, which combines a term measuring
the geometric fit of the ligand to the binding site and a term
measuring hydrogen bonding effects. Induced Fit. The generated
docking conformations were subjected to energy minimization
within the binding site and the protein sidechains are included in
the refinement stage. In detail, the protein backbone is set as rigid
while the side chains are not set to “free to move” but are set to
“tethered”, where an atom tether is a distance restraint that re-
strains the distance not between two atoms but between an atom
and a fixed point in space. Rescoring. Complexes generated by the
Induced Fit methodology stage were scored using the Alpha HB
scoring function. Gold tool was used with default efficiency settings
through MOE interface, by selecting GoldScore as scoring function
[49]. Autodock 4.2.6 software was used with PyRx interface. La-
marckian genetic algorithm was employed for this analysis with the
following settings: 50 runs for each ligand; 2,500,000 as maximum
number of energy evaluations; 27,000 as maximum number of
generations; 0.02 as rate of gene mutation and 0.8 as rate of
crossover. The grid box was set with 50, 50, and 50 points in the x, y,
and z directions, respectively, with the default grid spacing of
0.375 Å [50,51,59].
of 1 mM ABMECA and was always <10% of the total binding [56].
Bound and free radioactivity were separated by filtering the assay
mixture through Whatman GF/B glass fiber filters using a Brandel
€
cell harvester (Brandel Instruments, Unterfohring, Germany). The
filter bound radioactivity was counted by Packard Tri Carb 2810 TR
scintillation counter (Perkin Elmer).
3.2.3. Cyclic AMP assays
CHO cells transfected with hAR subtypes were washed with
phosphate-buffered saline, detached with tripsine and centrifuged
for 10 min at 200 g. Cells were seeded in 96-well white half-area
microplate (Perkin Elmer, Boston, USA) in a stimulation buffer
composed of Hank Balanced Salt Solution, 5 mM HEPES, 0.5 mM Ro
20e1724, 0.1% BSA. The examined compounds (1 nM - 1 mM) were
tested alone and/or in the presence of NECA, 100 nM cAMP levels
were then quantified by using the AlphaScreen cAMP Detection Kit
(Perkin Elmer, Boston, USA) following the manufacturer's in-
structions [57]. At the end of the experiments, plates were read
with the Perkin Elmer EnSight Multimode Plate Reader.
3.2.4. Data analysis
The protein concentration was determined according to a Bio-
Rad method with bovine albumin as a standard reference. Inhibi-
tory binding constant (Ki) values were calculated from those of IC50
according to Cheng & Prusoff equation Ki ¼ IC50/(1þ[C*]/KD*),
where [C*] is the concentration of the radioligand and KD* its
dissociation constant [56]. Ki and IC50 values were calculated by
non-linear regression analysis using the equation for a sigmoid
concentration-response curve (Graph-PAD Prism, San Diego, CA,
U.S.A).
3.3. Molecular modelling
Homology modelling and energy minimization studies were
carried out using MOE (Molecular Operating Environment, version
2014.09) suite [47]. All ligand structures were optimized using RHF/
AM1 semiempirical calculations and the software package MOPAC
[48] implemented in MOE was used for these calculations.
4. Notes
3.3.1. Homology modelling of the hA2B AR
Homology models of the hA2B AR were built using recently
solved X-ray structures of the hA2A AR in complex with Ado and UK-
432097 as templates, both structures being retrieved from Protein
All authors materially participated in the research and article
preparation. All authors have approved the final article and declare
no competing financial interest.

138
M. Betti et al. / European Journal of Medicinal Chemistry 150 (2018) 127e139
Circulation 115 (2007) 1581e1590.
Acknowledgments
[20] S. Toldo, H. Zhong, E. Mezzaroma, B.W. Van Tassell, H. Kannan, D. Zeng,
L. Belardinelli, N.F. Voelkel, A. Abbate, GS-6201, a selective blocker of the A2B
adenosine receptor, attenuates cardiac remodeling after acute myocardial
infarction in the mouse, J. Pharmacol. Exp. Therapeut. 343 (2012) 587e595.
[21] T. Krahn, T. Kramer, U. Rosentreter, J.M. Downey, N. Solenkova, Use of
substituted 2-thio-3,5-dicyano-4-phenyl-6-aminopyridines for the treatment
of reperfusion injury and reperfusion damage, 2006, WO2006099958A1.
[22] A. Kuno, S.D. Critz, L. Cui, V. Solodushko, X.M. Yang, T. Krahn, B. Albrecht,
S. Philipp, M.V. Cohen, J.M. Downey, Protein kinase C protects preconditioned
rabbit hearts by increasing sensitivity of adenosine A2B-dependent signaling
during early reperfusion, J. Mol. Cell. Cardiol. 43 (2007) 262e271.
[23] C. Lemos, B.S. Pinheiro, R.O. Beleza, J.M. Marques, R.J. Rodrigues, R.A. Cunha,
D. Rial, A. Kofalvi, Adenosine A2B receptor activation stimulates glucose up-
take in the mouse forebrain, Purinergic Signal. 11 (2015) 561e569.
[24] H. Johnston-Cox, M. Koupenova, D. Yang, B. Corkey, N. Gokce, M.G. Farb,
N. LeBrasseur, K. Ravid, The A2B adenosine receptor modulates glucose ho-
meostasis and obesity, PLoS One 7 (2012) e40584.
[25] R.A. Figler, G. Wang, S. Srinivasan, D.Y. Jung, Z. Zhang, J.S. Pankow, K. Ravid,
B. Fredholm, C.C. Hedrick, S.S. Rich, J.K. Kim, K.F. LaNoue, J. Linden, Links be-
tween insulin resistance, adenosine A2B receptors, and inflammatory markers
in mice and humans, Diabetes 60 (2011) 669e679.
[26] S. Ryzhov, S.V. Novitskiy, R. Zaynagetdinov, A.E. Goldstein, D.P. Carbone,
I. Biaggioni, M.M. Dikov, I. Feoktistov, Host A2B adenosine receptors promote
carcinoma growth, Neoplasia 10 (2008) 987e995.
[27] Q. Wei, S. Costanzi, R. Balasubramanian, Z.G. Gao, K.A. Jacobson, A2B adenosine
receptor blockade inhibits growth of prostate cancer cells, Purinergic Signal. 9
(2013) 271e280.
[28] M.A. Zimmerman, A. Grenz, E. Tak, M. Kaplan, D. Ridyard, K.S. Brodsky,
M.S. Mandell, I. Kam, H.K. Eltzschig, Signaling through hepatocellular A2B
adenosine receptors dampens ischemia and reperfusion injury of the liver,
Proc. Natl. Acad. Sci. U. S. A 110 (2013) 12012e12017.
[29] Z.G. Gao, R. Balasubramanian, E. Kiselev, Q. Wei, K.A. Jacobson, Probing biased/
partial agonism at the G protein-coupled A2B adenosine receptor, Biochem.
Pharmacol. 90 (2014) 297e306.
[30] M.W. Beukers, L.C.W. Chang, J.K. von Frijtag Drabbe Kunzel, T. Mulder-Krieger,
R.F. Spanjersberg, J. Brussee, A.P. IJzermann, New, non-adenosine, high-po-
tency agonists for the human adenosine A2B receptor with an improved
selectivity profile compared to the reference agonist N-ethyl-
carboxamidoadenosine, J. Med. Chem. 47 (2004) 3707e3709.
The work was financially supported by the Italian Ministry for
University
and
Research
(MIUR,
PRIN
2010e2011,
20103W4779_004 project) and the University of Florence (Italy).
Appendix A. Supplementary data
Supplementary data related to this article can be found at
https://doi.org/10.1016/j.ejmech.2018.02.081.
References
[1] K.A. Jacobson, L.J.S. Knutsen, P1 and P2 purine and pyrimidine receptor li-
gands, in: M.P. Abbracchio, M. Williams (Eds.), Purinergic and Pyrimidinergic
Signalling, Handbook of Experimental Pharmacology, Berlin 151/1, 2001,
pp. 129e175.
[2] P.A. Borea, S. Gessi, S. Merighi, K. Varani, Adenosine as multi-signalling
guardian angel in human diseases: when, where and how oes it exert its
protective effects? Trends Pharmacol. Sci. 37 (2016) 419e434.
[3] C.E. Müller, K.A. Jacobson, Recent developments in adenosine receptor ligands
and their potential as novel drugs, Biochim. Biophys. Acta 1808 (2011)
1290e1308.
[4] J. Zablocki, E. Elzein, R. Kalla, A2B adenosine receptor antagonists and their
potential indications, Expert Opin. Ther. Pat. 16 (2006) 1347e1357.
[5] G. Ortore, A. Martinelli, A2B receptor ligands: past, present and future trends,
Curr. Top. Med. Chem. 10 (2010) 923e940.
[6] R.V. Kalla, J. Zablocki, Progress in the discovery of selective, high affinity A2B
adenosine receptor antagonists as clinical candidates, Purinergic Signal. 5
(2009) 21e29.
[7] A. Carotti, M.I. Cadavid, N.B. Centeno, C. Esteve, M.I. Loza, A. Martinez, R. Nieto,
E. Ravina, F. Sanz, V. Segarra, E. Sotelo, A. Stefanachi, B. Vidal, Design, syn-
thesis, and structure-activity relationships of 1-,3-,8-, and 9-substituted-9-
deazaxanthines at the human A2B adenosine receptor, J. Med. Chem. 49
(2006) 282e299.
[8] A.W. Cheung, J. Brinkman, F. Firooznia, A. Flohr, J. Grimsby, M.L. Gubler,
K. Guertin, R. Hamid, N. Marcopulos, R.D. Norcross, L. Qi, G. Ramsey, J. Tan,
Y. Wen, R. Sarabu, 4-Substituted-7-N-alkyl-N-acetyl 2-aminobenzothiazole
amides: drug-like and non-xanthine based A2B adenosine receptor antago-
nists, Bioorg. Med. Chem. Lett 20 (2010) 4140e4146.
[9] P. Eastwood, J. Gonzalez, S. Paredes, S. Fonquerna, A. Cardus, J.A. Alonso,
A. Nueda, T. Domenech, R.F. Reinoso, B. Vidal, Discovery of potent and se-
lective bicyclic A2B adenosine receptor antagonists via bioisosteric amide
replacement, Bioorg. Med. Chem. Lett 20 (2010) 1634e1637.
[10] A. El Maatougui, J. Azuaje, M. Gonzalez-Gomez, G. Miguez, A. Crespo,
C. Carbajales, L. Escalante, X. Garcia-Mera, H. Gutierrez-de-Teran, E. Sotelo,
Discovery of potent and highly selective A2B adenosine receptor antagonist
chemotypes, J. Med. Chem. 59 (2016) 1967e1983.
[31] S. Hinz, S.K. Lacher, B.F. Seibt, C.E. Müller, BAY60-6583 acts as a partial agonist
at adenosine A2B receptors, J. Pharmacol. Exp. Therapeut. 349 (2014)
427e436.
[32] J.A. Baltos, E.A. Vecchio, M.A. Harris, C.X. Qin, R.H. Ritchie, A. Christopoulos,
P.J. White, L.T. May, Capadenoson, a clinically trialed partial adenosine A1
receptor agonist, can stimulate adenosine A2B receptor biased agonism, Bio-
chem. Pharmacol. 135 (2017) 79e89.
[33] L.C.W. Chang, J.K. von Frijtag Drabbe Künzel, T. Mulder-Krieger,
R.F. Spanjersberg, S.F. Roerink, G. van den Hout, M.W. Beukers, J. Brussee,
A.P. IJzerman,
A series of ligand displaying a remarkable agonistic-
antagonistic profile at the adenosine A1 receptor, J. Med. Chem. 48 (2005)
2045e2053.
[34] D. Meibom, B. Albrecht-Kupper, N. Diedrichs, W. Hubsch, R. Kast, T. Kramer,
U. Krenz, H.G. Lerchen, J. Mittendorf, P.G. Nell, F. Sussmeier, A. Vakalopoulos,
K. Zimmermann, Neladenoson bialanate hydrochloride: a prodrug of a partial
adenosine A1 receptor agonist for the chronic treatment of heart diseases,
ChemMedChem 12 (2017) 728e737.
[11] P.G. Baraldi, M.A. Tabrizi, F. Fruttarolo, R. Romagnoli, D. Preti, Recent im-
provements in the development of A2B adenosine receptor agonists, Puri-
nergic Signal. 5 (2009) 3e19.
[12] P.G. Baraldi, D. Preti, M.A. Tabrizi, F. Fruttarolo, R. Romagnoli, M.D. Carrion,
L.C. Cara, A.R. Moorman, K. Varani, P.A. Borea, Synthesis and biological eval-
uation of novel 1-deoxy-1-[6-[((hetero)arylcarbonyl)hydrazino]- 9H-purin-9-
[35] S. Okuyama, H. Oono, M. Kawamura, Preparation of benzylidene rhodanine
derivatives for lowering blood sugar, 1995, JP07070095 A 19950314.
[36] R. Mamgain, R. Singh, D.S. Rawat, DBU-catalyzed three-component one-pot
synthesis of highly functionalized pyridines in aqueous ethanol, J. Het. Chem.
46 (2009) 69e73.
yl]-N-ethyl-b-D-ribofuranuronamide derivatives as useful templates for the
development of A2B adenosine receptor agonists, J. Med. Chem. 50 (2007)
374e380.
€
[13] U. Rosentreter, R. Henning, M. Bauser, T. Kramer, A. Vaupel, W. Hübsch, K.
€
[37] U. Rosentreter, T. Kramer, M. Shimada, W. Hübsch, N. Diedrichs, T. Krahn, K.
€
Dembowsky, O. Salcher-Schraufstatter, J.P. Stasch, T. Krahn, E. Perzborn,
Henninger,
J.P.
Stasch,
Substituted
2-thio-3,5-dicyano-4-phenyl-6-
Substituted 2-Thio-3,5-Dicyano-4-Aryl-6-Aminopyridines and the use
Thereof as Adenosine Receptor Ligands, 2001, WO/2001/025210.
[14] D. Van der Hoeven, T.C. Wan, E.T. Gizewski, L.M. Kreckler, J.E. Maas, J. Van
Orman, K. Ravid, J.A. Auchampach, Arole for the low-affinity A2B adensine
receptor in regulating superoxide generation by murine neutrophils,
J. Pharmacol. Exp. Therapeut. 338 (2011) 1004e1012.
aminopyridines and their use as adenosine receptor-selective ligands, 2003,
WO/2003/008384.
[38] A. Vakalopoulos, D. Meibom, P. Nell, F. Suessmeier, B. Albrecht-Kupper, K.
Zimmerman, J. Keldenich, D. Schneider, U. Krentz, Substituted dicyanopyr-
idines as adenosine receptor stimulators and their preparation and use for the
treatment of cardiovascular diseases, 2012, WO 20122000945 A1.
[39] L.W. Jones, L.F. Werner, Aliphatic hydroxylammonium salts and hydroxamic
acids with halogen substituents, J. Am. Chem. Soc. 39 (1917) 413e422.
[40] M.A.M. Gad-Elkareem, M.A.A. Elneairy, A.M. Taha, Reactions with 3,6-
diaminothieno[2,3-b]-pyridines: synthesis and characterization of several
new fused pyridine heterocycles, Heteroat. Chem. 18 (2007) 405e413.
[41] D. Thimm, A.C. Schiedel, F.F. Sherbiny, S. Hinz, K. Hochheiser, D.C.G. Bertarelli,
A. Maass, C.E. Müller, Ligand-specific binding and activation of the human
adenosine A2B receptor, Biochemistry 52 (2013) 726e740.
[42] D. Dal Ben, M. Buccioni, C. Lambertucci, A. Thomas, R. Volpini, Simulation and
comparative analysis of binding modes of nucleoside and non-nucleoside
agonists at the A2B adenosine receptor, In Silico Pharmacol. 1 (2013) 24.
[43] S. Costanzi, A.A. Ivanov, I.G. Tikhonova, K.A. Jacobson, Structure and function
of G protein-coupled receptors studied using sequence analysis, molecular
modeling and receptor engineering: adenosine receptors, Front. Drug Des.
Discov. 3 (2007) 63e79.
[15] I. Feoktistov, I. Biaggioni, Role of adenosine in asthma, Drug Dev. Res. 39
(1996) 333e336.
[16] C.N. Wilson, A. Nadeem, D. Spina, R. Brown, C.P. Page, S.J. Mustafa, Adenosine
receptors and asthma, Handb. Exp. Pharmacol. (2009) 329e362.
[17] R. Walaschewski, F. Begrow, E.J. Verspohl, Impact and benefit of A2B adenosine
receptor agonists for the respiratory tract: mucociliary clearance, ciliary beat
frequency, trachea muscle tonus and cytokine release, J. Pharm. Pharmacol. 65
(2013) 123e132.
[18] H. Karmouty-Quintana, H. Zhong, L. Acero, T. Weng, E. Melicoff, J.D. West,
A. Hemnes, A. Grenz, H.K. Eltzschig, T.S. Blackwell, Y. Xia, R.A. Johnston,
D. Zeng, L. Belardinelli, M.R. Blackburn, The A2B adenosine receptor modulates
pulmonary hypertension associated with interstitial lung disease, Faseb. J. 26
(2012) 2546e2557.
[19] T. Eckle, T. Krahn, A. Grenz, D. Kohler, M. Mittelbronn, C. Ledent,
M.A. Jacobson, H. Osswald, L.F. Thompson, K. Unertl, H.K. Eltzschig, Car-
dioprotection by ecto-5'-nucleotidase (CD73) and A2B adenosine receptors,

M. Betti et al. / European Journal of Medicinal Chemistry 150 (2018) 127e139
139
[44] D. Dal Ben, C. Lambertucci, G. Marucci, R. Volpini, G. Cristalli, Adenosine re-
ceptor modeling: what does the A2A crystal structure tell us? Curr. Top. Med.
Chem. 10 (2010) 993e1018.
[53] D. Rodriguez, Z.G. Gao, S.M. Moss, K.A. Jacobson, J. Carlsson, Molecular docking
screening using agonist-bound GPCR structures: probing the A2A adenosine
receptor, J. Chem. Inf. Model. 55 (2015) 550e563.
[45] G. Lebon, T. Warne, P.C. Edwards, K. Bennett, C.J. Langmead, A.G. Leslie,
C.G. Tate, Agonist-bound adenosine A2A receptor structures reveal common
features of GPCR activation, Nature 474 (2011) 521e525.
[46] F. Xu, H. Wu, V. Katritch, G.W. Han, K.A. Jacobson, Z.G. Gao, V. Cherezov,
R.C. Stevens, Structure of an agonist-bound human A2A adenosine receptor,
Science 332 (2011) 322e327.
[54] F. Vincenzi, M. Targa, R. Romagnoli, S. Merighi, S. Gessi, P.G. Baraldi, P.A. Borea,
K. Varani, TRR469, a potent A1 adenosine receptor allosteric modulator, ex-
hibits anti-nociceptive properties in acute and neuropathic pain models in
mice, Neuropharmacology 81 (2014) 6e14.
[55] K. Varani, A. Massara, F. Vincenzi, A. Tosi, M. Padovan, F. Trotta, P.A. Borea,
Normalization of A2A and A3 adenosine receptor up-regulation in rheumatoid
arthritis patients by treatment with anti-tumor necrosis factor alpha but not
methotrexate, Arthritis Rheum. 60 (2009) 2880e2891.
[47] I. Molecular Operating Environment; C.C.G., 1255 University St., Suite 1600,
Montreal, Quebec, Canada, H3B 3X3.
[48] J.J. Stewart, MOPAC: a semiempirical molecular orbital program, J. Comput.
Aided Mol. Des. 4 (1990), 1e105.
[49] G. Jones, P. Willett, R.C. Glen, A.R. Leach, R. Taylor, Development and valida-
tion of a genetic algorithm for flexible docking, J. Mol. Biol. 267 (1997)
727e748.
[50] R. Huey, G.M. Morris, A.J. Olson, D.S. Goodsell, A semiempirical free energy
force field with charge-based desolvation, J. Comput. Chem. 28 (2007)
1145e1152.
[56] K. Varani, S. Merighi, S. Gessi, K.-N. Klotz, E. Leung, P.G. Baraldi, B. Cacciari,
R. Romagnoli, G. Spalluto, P.A. Borea, [3H]MRE 3008F20: a novel antagonist
radioligand for the pharmacological and biochemical characterization of hu-
man A3 adenosine receptors, Mol. Pharmacol. 57 (2000) 968e975.
[57] A. Ravani, F. Vincenzi, A. Bortoluzzi, M. Padovan, S. Pasquini, S. Gessi,
S. Merighi, P.A. Borea, M. Govoni, K. Varani, Role and function of A2A and A3
adenosine receptors in patients with Ankylosing spondylitis, psoriatic arthritis
and rheumatoid arthritis, Int. J. Mol. Sci. 18 (2017), 697.
[51] G.M. Morris, R. Huey, W. Lindstrom, M.F. Sanner, R.K. Belew, D.S. Goodsell,
A.J. Olson, AutoDock4 and AutoDockTools4: automated docking with selective
receptor flexibility, J. Comput. Chem. 30 (2009) 2785e2791.
[52] D. Dal Ben, M. Buccioni, C. Lambertucci, G. Marucci, C. Santinelli, A. Spinaci,
A. Thomas, R. Volpini, Simulation and comparative analysis of different
binding modes of non-nucleoside agonists at the a2a adenosine receptor, Mol.
Inform 35 (2016) 403e413.
[58] W.D. Cornell, P. Cieplak, C.I. Bayly, I.R. Gould, K.M. Merz, D.M. Ferguson,
D.C. Spellmeyer, T. Fox, J.W. Caldwell, P.A. Kollman, A second generation force
field for the simulation of proteins, nucleic acids, and organic molecules, J. Am.
Chem. Soc. 117 (1995) 5179e5197.
[59] S. Dallakyan, A.J. Olson, Small-molecule library screening by docking with
PyRx, Meth. Mol. Biol. 1263 (2015) 243e250.