Selective and potent adenosine A3 receptor antagonists by methoxyaryl substitution on the N-(2,6-diarylpyrimidin-4-yl)acetamide scaffold

Article abstract of DOI:10.1016/j.ejmech.2012.11.010

The influence of diverse methoxyphenyl substitution patterns on the N-(2,6-diarylpyrimidin-4-yl)acetamide scaffold is herein explored in order to modulate the A₃ adenosine receptor antagonistic profile. As a result, novel ligands exhibiting excellent potency (K_i on A₃ AR < 20 nM) and selectivity profiles (above 100-fold within the adenosine receptors family) are reported. Moreover, our joint theoretical and experimental approach allows the identification of novel pharmacophoric elements conferring A₃AR selectivity, first established by a robust computational model and thereafter characterizing the most salient features of the structure-activity and structure-selectivity relationships in this series.

Full text of DOI:10.1016/j.ejmech.2012.11.010

European Journal of Medicinal Chemistry 59 (2013) 235e242
Contents lists available at SciVerse ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Original article
Selective and potent adenosine A3 receptor antagonists by methoxyaryl
substitution on the N-(2,6-diarylpyrimidin-4-yl)acetamide scaffold
,
,
,
b
Vicente Yaziji a b, David Rodríguez d, Alberto Coelho a b, Xerardo García-Mera c, Abdelaziz El Maatougui a
,
José Brea b, María Isabel Loza b, María Isabel Cadavid b, Hugo Gutiérrez-de-Terán d , Eddy Sotelo
,
a,b,c,
**
*
a Combinatorial Chemistry Unit (COMBIOMED), Center for Research in Biological Chemistry and Molecular Materials (CIQUS), University of Santiago de Compostela, 15782, Spain
b Institute of Industrial Pharmacy, University of Santiago de Compostela, 15782, Spain
c Department of Organic Chemistry, Faculty of Pharmacy, University of Santiago de Compostela, 15782, Spain
d Public Galician Foundation of Genomic Medicine, Santiago de Compostela University Hospital (CHUS), 15706 Santiago de Compostela, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
The influence of diverse methoxyphenyl substitution patterns on the N-(2,6-diarylpyrimidin-4-yl)acet-
amide scaffold is herein explored in order to modulate the A3 adenosine receptor antagonistic profile. As
a result, novel ligands exhibiting excellent potency (Ki on A3 AR < 20 nM) and selectivity profiles (above
100-fold within the adenosine receptors family) are reported. Moreover, our joint theoretical and
experimental approach allows the identification of novel pharmacophoric elements conferring A3AR
selectivity, first established by a robust computational model and thereafter characterizing the most
salient features of the structureeactivity and structureeselectivity relationships in this series.
Ó 2012 Elsevier Masson SAS. All rights reserved.
Received 24 September 2012
Received in revised form
6 November 2012
Accepted 8 November 2012
Available online 24 November 2012
Keywords:
Adenosine receptor
3D-QSAR
Grid-Independent Descriptors (GRIND-2)
SuzukieMiyaura cross-coupling reaction
Pyrimidines
Adenosine antagonists
Library synthesis
1. Introduction
pharmacologically characterized. The A3AR is involved in a variety of
important physiological processes, including modulation of cerebral
The purine nucleoside adenosine exerts its key regulatory roles
on several tissues through activation of the four adenosine receptors
(ARs), namely A1, A2A, A2B and A3 [1]. Being members of the super-
family of G protein-coupled receptors (GPCRs), ARs are validated
targets for pharmacological intervention in several pathophysio-
logical conditions [2]. In addition, the crystallization of the human
A2AAR [3] is boosting the rational design of novel selective ligands for
the different ARs. Among the ARs, A3ARs are the latest cloned and
and cardiac ischemic damage [4,5], inflammation [6], modulation of
intraocular pressure [7], regulation of normal and tumor cell growth
[8,9] and immunosuppression [10]. However, A3-signaling in several
processes is still controversial [11,12], being perhaps the most enig-
matic among adenosine receptors. The two personalities of A3AR
often come into direct conflict, e.g., in ischemia, inflammation and
cancer, rendering this receptor as a single entity behaving in 2
different ways. Thus, the elucidation of A3AR dual behavior in several
pathophysiological conditions remains an unmet challenge.
The antagonists of the A3AR [13e15] have shown to be partic-
ularly attractive as novel potential anti-inflammatory drugs [6],
cerebro-protective agents [16], as well as for the treatment of
glaucoma [17]. Furthermore, the recent evidence of high expression
levels of A3AR in several cell lines has suggested potential appli-
cations in cancer chemotherapy [18]. The putative applications of
A3AR antagonists as drugs, as well as the growing demand for
pharmacological tools to study the dual roles of human A3AR, has
made the identification of potent and selective small molecule
ligands a topic of great interest [1,2,19]. The pursuit of non-
Abbreviations: ARs, human adenosine receptors; GPCRs,
G protein-coupled
receptors; 3D-QSAR, three-dimensional quantitative structureeactivity relation-
ships; GRIND2, GRid INdependent Descriptors version 2; MIF, molecular interaction
fields.
* Corresponding author. Combinatorial Chemistry Unit (COMBIOMED), Center for
Research in Biological Chemistry and Molecular Materials (CIQUS), University of
Santiago de Compostela, 15782, Spain. Tel.: þ34 881815732; fax: þ34 881815704.
** Corresponding author. Tel.: þ34 881815428; fax: þ34 981951473.
E-mail addresses: hugo.teran@usc.es (H. Gutiérrez-de-Terán), e.sotelo@usc.es
(E. Sotelo).
0223-5234/$ e see front matter Ó 2012 Elsevier Masson SAS. All rights reserved.
http://dx.doi.org/10.1016/j.ejmech.2012.11.010

236
V. Yaziji et al. / European Journal of Medicinal Chemistry 59 (2013) 235e242
xanthinic A3AR antagonists focused on the exploration of diverse
heterocyclic libraries (particularly tri- and bi-cyclic heteroaromatic
scaffolds and, to a lesser extent, mono-heterocyclic systems).
Whereas the systematic structural elaboration of these prototypes
has provided derivatives possessing good affinity, the selectivity
issue and the relatively poor bioavailability profiles of drug candi-
dates have remained elusive until recently [1,2,19].
We have recently described highly potent and selective A3AR
antagonists through a straightforward and convergent synthetic
pathway [20], obtaining single ringed heterocyclic cores showing
low nanomolar affinities at the A3AR and marked selectivity versus
the other ARs. These molecules were designed by the structural
redecoration of the aryl moieties at positions 2 and 6 of the N-
(pyrimidin-4-yl)-acylamide scaffold [21], a fruitful substructure that
is present in diverse ARs antagonists [2,21,22]. It was observed that
the methoxy substituents on the phenyl ring not only increased the
affinities for the hA3AR, but also remarkably enhanced the selectivity
profiles of this series (Fig. 1). This finding reinforces key observations
from other authors [22e25], and motivated a thorough exploration
of the pharmacological insights of methoxyaryl substitution patterns
on the A3AR antagonistic profile of N-(2,6-diphenylpyrimidin-4-yl)
acetamides. Building upon the development of a computational
model with satisfactory predictivity together with an efficient and
flexible synthetic methodology, we herein report the design,
synthesis and pharmacological evaluation of an expanded series
derived from the N-(2,6-diphenylpyrimidin-4-yl)acetamide scaffold.
This study, in addition to providing novel potent A3AR antagonists,
highlights several novel methoxyphenyl-derived pharmacophoric
elements conferring A3AR selectivity, as well as the most salient
features of the structureeactivity (SAR) and structureeselectivity
(SSR) relationships in this series.
Fig. 2. Combined structure- and ligand-based computational protocol for the affinity
prediction of novel compounds in A3AR.
GOLD [27] was employed with the parameters indicated in the
Experimental Section, and the observed poses were compared
with the consensus binding mode defined for this scaffold [20].
(ii) 3D-QSAR: In parallel to the docking studies, a predictive 3D-
QSAR model was built using the 68 molecules with previously
reported Ki values [20], by means of the last generation of
Grid-Independent Descriptors (GRIND-2) [28]. The software
Pentacle [29] was used for this purpose, with the parameters
detailed in the Experimental Section. Thereafter, the relative
affinities of the filtered compounds (those that achieved the
conserved binding orientation discussed below) were evalu-
ated with the 3D-QSAR model II (see Table 1), using the
“Prediction mode” included in Pentacle. The interpretation of
QSAR models developed with the GRIND-2 methodology is
based on the analysis of the correlograms, which encode the
geometrical relationships between the most relevant molec-
ular interaction fields (MIFs). Thereafter, one can easily
2. Molecular modeling
A small library consisting of ten N-[(2,6-bisaryl)pyrimidin-4-yl]-
acetamides was designed, including mono-, bi- and tri-
methoxyphenyl substitution patterns as well as 4-ethoxyphenyl,
4-trifluoromethoxyphenyl and 4-methoxystyryl residues. The
designed molecules were then evaluated in silico, using combined
Structure- and Ligand-Based computational pipeline protocol
derived from our previous work [20], which is here adapted as
summarized in Fig. 2. Briefly, the protocol includes two stages:
(i) Molecular docking of each ligand on a homology model of the
hA3AR, built with MODELLER [26] using the inactive structure
of A2AAR (PDB code 3EML) [3] as a template. The program
Fig. 1. The N-[(2,6-bisaryl)pyrimidin-4-yl]-acetamide scaffold and the effect on the selectivity profile of the (poly)methoxy substitution on the phenyl fragments. The substitutions
follow the nomenclature of an A1AR model [10].

V. Yaziji et al. / European Journal of Medicinal Chemistry 59 (2013) 235e242
237
Table 1
template. The synthetic pathway developed follows a two-step
sequence as illustrated in Scheme 1 [30]. The selected approach is
based on the highly reliable and well-established SuzukieMiyaura
cross-coupling reaction [31], exploiting the excellent commercial
availability of boronic acids (3). Acetylation of the heterocyclic core
and subsequent palladium-catalyzed arylation afforded a library of
fifteen N-(2,6-diphenylpyrimidin-4-yl)acetamides (4aeo) in yields
ranging 73e91%.
Summary of 3D-QSAR models considered in this work. The number of molecules and
latent variables considered for each model are indicated, as well as statistics
regarding their accuracy and predictivity in cross-validation. Prospective assessment
is also shown where applicable, with the number of new molecules evaluated and
the standard deviation (in pKi units) of their corresponding predictions.
Model
Cpds.
LV
R2
Q2
SDEP
SDEC
New cpds.
pred.
SDEP
pred.
Ia
62
68
76
2
2
3
0.86
0.88
0.88
0.67
0.73
0.70
0.48
0.42
0.44
0.31
0.29
0.28
6
0.37
0.73c
e
IIb
IIId
8c
e
4. Pharmacology
a
Model built with 62 diaryl-pyrimidines, see Ref. [9].
Model elaborated with both the training and tested molecules in model I.
All the novel molecules presented in this work that found a satisfactory docking
b
c
The pharmacological profile of the synthesized compounds
(4aeo) was studied in vitro at the four human adenosine receptor
subtypes, using radioligand binding assays according to the
experimental protocols described elsewhere [32]. Table 2 compiles
the pharmacological data, together with the associated computa-
tional predictions, obtained for the 10 novel compounds here
presented: 4c, 4eeh, 4jel and 4neo. For comparative purposes,
some derivatives previously described by our group [20] (4a, 4b, 4d,
4i and 4m) were evaluated again and incorporated into the series.
binding mode on a homology model of A3AR. Note that the elimination of compound
4e would diminish the value of the SDEP to 0.49 pKi units, which is close to the SDEP
of the internal validation of the model itself.
d
Partial least squares 3D-QSAR model generated with the molecules present in
model II plus the novel 8 compounds.
identify and further examine the pharmacophoric variables
that contribute the most to the partial least squares fitness,
and thus to the modulation of the A3AR affinity.
5. Results and discussion
3. Chemistry
The consensus binding mode proposed for the N-[(2,6-bisaryl)
pyrimidin-4-yl]acetamide scaffold is illustrated in Fig. 3A [9]. From
the 10 novel ligands, 8 presented this conserved binding mode,
which is defined by a double hydrogen bond of the exocyclic amido
Guided by the modeling results, we assembled a focused library
that was representative of the diverse substitution patterns on the
phenyl residues of the N(2,6-diphenylpyrimidin-4-yl)-acetamide
Scheme 1. Synthetic pathway and structure of the compounds here presented (4aeo).

238
V. Yaziji et al. / European Journal of Medicinal Chemistry 59 (2013) 235e242
Table 2
Structure and affinity data for the N-(2,6-diarylpyrimidin-4-yl)acetamides (4aeo) at the human adenosine receptors.
Comp
Ar
Ki (nM) or % at 1
m
M
pKi @ hA3
Exp
Prede
a
b
c
d
hA1
hA2A
hA2B
hA3
4af
Ph
31.2 ꢀ 4.1
255.3 ꢀ 1.3
22%
7%
12.1 ꢀ 1.3
24.1 ꢀ 1.3
7.92
7.62
7.83
7.43
4bf
2-OMe-Ph
4%
15%
(ISVY158)
4c
3-OMe-Ph
4-OMe-Ph
4-OCF3-Ph
4-OEt-Ph
4-OMe-Styryl
2,3-OMe-Ph
2,4-OMe-Ph
22.6 ꢀ 2.4
2%
8%
12%
16%
10%
3%
49.4 ꢀ 3.7
67.5 ꢀ 4.3
2.8 ꢀ 0.1
3.6 ꢀ 0.2
8.55
8.44
6.41
7.40
6.29
7.37
8.27
8.26
8.32
8.02
8.06
n.p.
4df (ISVY130)
4%
2%
4e
4f
4g
1%
2%
13%
4%
1%
2%
389.0 ꢀ 8.5
40.2 ꢀ 2.6
513.1 ꢀ 9.8
42.4 ꢀ 2.2
5.4 ꢀ 0.1
4h
375.1 ꢀ 8.3
493.7 ꢀ 5.1
7.43
8.25
4if
10%
7%
(ISVY167)
4j
2,5-OMe-Ph
4%
17%
1%
6.3 ꢀ 0.3
8.20
7.85
(ISVY350)
4k
4l
2,6-OMe-Ph
3,4-OMe-Ph
4%
10%
1%
5%
3%
2%
4%
5.2 ꢀ 0.2
e
n.p.
8.19
8.28
(ISVY345)
4mf
3,4-OCH2O-Ph
3,5-OMe-Ph
3,4,5-OMe-Ph
17.00 ꢀ 3.1
15%
14%
3345.0 ꢀ 1127
6%
1%
1%
3.3 ꢀ 0.4
55.0 ꢀ 3.1
26.1 ꢀ 2.1
8.48
7.26
7.58
8.11
8.15
8.04
4n
4o
3%
1%
a
Displacement of specific [3H]DPCPX binding in human CHO cells expressed as Ki ꢀ SEM in nM (n ¼ 3) or percentage displacement of specific binding at a concentration of
M (n ¼ 2).
1
m
b
Displacement of specific [3H]4-(2-[7-Amino-2-(2-furyl) [1,2,4]triazolo[2,3-a] [1,3,5]triazin-5-ylamino]ethyl)phenol binding in human HeLa cells expressed as Ki ꢀ SEM in
nM (n ¼ 3) or percentage displacement of specific binding at a concentration of 1 M (n ¼ 2).
m
c
Displacement of specific [3H]DPCPX binding in human HEK-293 cells expressed as Ki ꢀ SEM in nM (n ¼ 3) or percentage displacement of specific binding at a concentration
of 1
m
M (n ¼ 2).
d
Displacement of specific [3H]NECA binding in human HeLa cells expressed as Ki ꢀ SEM in nM (n ¼ 3) or percentage displacement of specific binding at a concentration of
1
m
M (n ¼ 2).
e
Ligands 4g and 4k are marked as not predicted (n.p.), since they do not present the conserved binding mode in the molecular docking stage.
f
For comparative purposes, previously described compounds (Ref. [9]) were evaluated again at 1 mM, these compounds were included in the generation of the 3D-QSAR
model, thus the calculated pKi value on A3AR is indicated in italics.
group (donating) and the N3 of the heterocyclic core (accepting)
with Asn 6.55 (note Ballesteros & Weinstein numbering [33]), plus
nanomolar range (Table 2), being particularly noteworthy some
novel, highly potent and selective ligands (e.g., compounds 4j, 4l,
4f, 4n). Moreover, the predictions of the computational model
show good agreement with the experimental results (Table 2 and
Fig. 4A). The two compounds that did not achieve the consensus
binding orientation in the docking stage (4g and 4k) were experi-
mentally confirmed as the weakest binders within the whole series.
Compound 4g, that can hardly accommodate the 4-OMe-styryl
substituents in the binding site, exhibit a reduced A3AR affinity
(pKi < 6.5). Similarly, the highly restrained conformational spec-
trum of compound 4k (due to the 2,6-dimethoxy substitution
pattern) probably precludes its binding to any AR. The predictions
of the 3D-QSAR model for the 8 new molecules (4c, 4eef, 4h, 4j, 4l
and 4neo) show an initial standard error of the prediction (SDEP)
of 0.73 pKi units. A closer examination revealed that approximately
1/3 of this error is due to the insensitivity of the model to the w100
fold decrease in affinity of 4-trifuoromethoxyphenyl derivative 4e,
as compared to the related 4-methoxyphenyl compound 4d
(ISVY130) (see Table 2). Indeed, a similar detrimental effect of the
trifluoro substitution was already observed in previous series [20],
where the 4-trifluorotolyl derivative showed a 28-fold reduction in
affinity as compared with the 4-tolyl compound (ISVY133).
a
d stacking interaction with Phe 5.29 at EL2. In addition to these
interactions with the abovementioned residues, which are totally
conserved in the ARs family, the binding of the central heterocycle
is reinforced by hydrophobic interactions with Leu 6.51 (see
Fig. 3A). The symmetric aryl substitutions are accommodated
respectively within the binding pocket defined by Leu 3.32, Ser 5.42,
Ser 6.52 and Trp 6.48 (substituent L2) and Val 2.64, Leu 3.33 and Leu
7.35 (substituent L3), the residues in italics being specific for the
A3AR. The release of a new crystal structure of the A2AAR (also in its
inactive conformation) with a 1,2,4-triazine derivative [34], an
antagonist that shows a relative chemical similarity to our scaffold,
provided us the opportunity of further assessing this binding
orientation. This comparison is illustrated in Fig. 3B, and provides
an experimental support to the binding mode discussed for our N-
[(2,6-bisaryl)pyrimidin-4-yl]-acetamide scaffold. Once defined the
binding orientation, and following the protocol in Fig. 1, the 8
ligands selected in the docking stage were evaluated in a 3D-QSAR
model. Such a model had been built with a training set of 68
molecules previously reported by us (model II in Table 1), pre-
senting very good accuracy (r2 ¼ 0.88) and predictivity (q2 ¼ 0.73)
in the leave-one-out cross-validation. The predictions for the
selected 8 compounds anticipated high affinities for all of them, in
the range of 7.43e8.26 pKi units.
In order to provide further interpretation of the structuree
activity relationships of the series, a new 3D-QSAR model was
generated now adding the novel 8 compounds to the training set of
68 compounds initially considered. The statistics of this refined
model, (model III in Table 1), remain almost unchanged as
The experimental binding affinities confirm that most of the
examined compounds display high affinities for the hA3AR, in the

V. Yaziji et al. / European Journal of Medicinal Chemistry 59 (2013) 235e242
239
ꢀ
Fig. 3. Binding mode of novel compounds in A3AR and selectivity considerations within ARs. (A) Docking pose of compound 4l (ISVY345) in A3AR: residues within 3 A of the ligand
are labeled, indicating with a square A3AR specific residues. (B) Docking solution of panel A superimposed to X-ray structure of A2AAR in complex with the 1,2,4-triazine derivative
T4E (white carbons, PDB 3UZC). Both ligands present similar chemical scaffolds, and show common anchoring points with each receptor, while the aryl substituents sit in analogous
binding subpockets. (C) Pseudo-sequence alignment the human ARs, including the binding site residues defined above, with the specific A3AR residues shadowed in gray. (D)
Docking poses of the novel compounds (4c, 4eeh, 4jel, 4neo) in A3AR, showing a superposition of the broader binding site surface of the A3AR (blue) and that of the A2AAR (orange,
PDB 3EML). (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this article.)
compared to previous models (see Table 1), thus adding confidence
on the robustness of the 3D-QSAR methodology. The most impor-
tant features identified by the model are illustrated in Fig. 4B and
can be briefly summarized as follows: (i) the optimal substitution at
the L1 site by an acetamide [TIPeTIP representing the alkyl
substitutions at L1 at the optimal distance of 3 Ǻ; TIPeN1 cross-
correlogram delimiting the optimal distance (5 Ǻ) between the
surface of L1 and the interaction with Asn 6.55] and (ii) the relative
disposition of the aromatic L2 and L3 substituents with respect to
the surface tip of L1 (DRY-TIP cross-correlogram at an optimum
distance of 13 Ǻ). These features reinforce the initial interpretation
of the interdependence of the size of L1 and L2/L3 substituents,
Fig. 4. Results of the 3D-QSAR modeling. (A) Representation of the experimental versus calculated affinities using the 3D-QSAR model II (see Table 1). Red diamonds represent
newly predicted molecules by that model. (B) Most relevant correlogram vectors in the explanatory model (model III in Table 1) are represented for compound ISVY345 (4l) in the
binding site of A3AR, where main residues are labeled. The GRID nodes are represented by dots colored in yellow (DRY probe), blue (N1), red (O) and green (TIP). The main four
vectors (N1eTIP, N1eN1, TIPeTIP and DRYeTIP) are labeled in red. (For interpretation of the references to color in this figure legend, the reader is referred to the web version of this
article.)

240
V. Yaziji et al. / European Journal of Medicinal Chemistry 59 (2013) 235e242
which is optimally achieved with the acetamide in L1 and partic-
ularly 3 and 4 methoxy-phenyl substituents, i.e., compounds 4c, 4d,
4i, 4j, 4l and the piperonyl derivative 4m.
affinity of the closest compound 4l (Ki ¼ 3.3 nM), may be explained
by the reduced volume that the piperonyl residue can explore
compared with the 3,4-OMe-Ph derivative (4l).
A comparison of the pharmacological data obtained for the
parent compound of the series (4a, Ar ¼ Ph, a potent but not
selective ligand) and its mono-methoxy derivatives (ortho 4b, meta
4c and para 4d) reveals a significant, but differentiated variation in
their pharmacological profile. On one side, the introduction of the
methoxy group at either ortho or para positions generated highly
potent and completely selective ligands (4b and 4d), with the
4-methoxyphenyl derivative (4d, Ki ¼ 3.6 nM) displaying a 7-fold
higher affinity than the 2-methoxyphenyl ligand 4b
(Ki ¼ 24.1 nM). On the other side, methoxylation at position 3
The SAR and SSR analysis for the remaining ligands can be
discussed using the 2-methoxyphenyl derivative 4b as reference.
This ligand is a relatively potent (Ki ¼ 24.1 nM) and highly selective
hA3AR antagonist, and a pairwise comparison of this compound
with the dimethoxyphenyl derivatives that retain the ortho-
methoxyphenyl (4h-4k) is especially suitable. The 2,4-
dimethoxyphenyl (4i) and 2,5-dimethoxyphenyl (4j) congeners
elicited remarkable affinity (Ki of 5.4 and 6.3 nM respectively) and
selectivity for the A3 receptor. As observed in the binding poses of
these compounds on the A3AR, depicted in Fig. 3D, the ortho-
methoxy at L2 is preferably located close to the N1 of the pyrimi-
dine ring, in a way that the para and meta methoxy groups in 4i and
4j, respectively, exploits the selective A3AR subpocket defined by
Ser 5.42 and Ser 6.52, discussed above. The 2,3-dimethoxyphenyl
derivative 4h can be compared with the monosubstituted deriva-
tives in ortho (4b) and meta (4c), since it exhibits a slightly lower
affinity for A3 as compared to 4b, while reproducing the non-
selective profile observed for 4c. Interestingly, the docking pose
of this ligand places the meta-methoxy of L2 in the most buried area
(i.e., closer to the N1 of the pyrimidine), as opposed to the selective
A3AR subpocket explored by 2,5-OMe-Ph (4j). This could explain
the lower selectivity of the ligand 4h, and also the meta mono-
substituted derivative (4c), which according to the docking simu-
lations adapts its binding pose to exploit the selective A3AR
subpocket while avoiding this region in A2AAR, a differential
binding mechanism probably occurring for the remaining ARs.
afforded
a ligand exhibiting an outstanding potency (4c,
Ki ¼ 2.8 nM), albeit reproducing the promiscuity profile of the
parent compound (4a). Three novel compounds (4e-g) were
synthesized on the basis of the optimal potency/selectivity profile
observed for the 4-methoxyphenyl derivative 4d [9], also encour-
aged by the predictions of the computational model (Table 2 and
Fig. 4A). Even if compound 4g was anticipated to hardly bind to the
A3AR model in the docking stage, we prepared and tested this styryl
derivative to provide a broader characterization of the series and to
further challenge the computational pipeline. While substitutions
at the para position significantly attenuated the potency by 10e140
fold, these compounds retained an excellent selectivity profile,
with the 4-ethoxyphenyl derivative 4f displaying a satisfactory
balance between moderate potency (Ki ¼ 40.2 nM) and high
selectivity. A complementary docking exploration in the A2AAR
crystal structure was performed to obtain deeper knowledge on the
selectivity issues, as detailed in the Experimental Section. Encour-
agingly, only compounds with sub-micromolar affinity on A2AAR
(4a, 4c, 4h) achieve the conserved binding mode depicted for the
A3AR in Fig. 3, with the only exception of the 2-OMe-Ph substituted
compound 4b, a highly selective compound that also finds this
conserved pose in A2AR. According to this finding, the enhanced
selectivity of the 4-substituted compounds (4deg, as well as 4l)
could be explained by the lack of this binding orientation due to
steric clashes with His 6.52 and Asn 5.42. Instead, A3AR presents
a smaller serine in the two equivalent positions, offering a bigger
cavity to accommodate the para substitutions in the aromatic ring
at L2. Fig. 3D shows the superposition of the binding site surface of
the A3AR (blue) and A2AAR (orange), where it can be appreciated an
increased surface to accommodate bulkier substituents, particu-
larly at meta and para positions of the L2 moiety for binding A3AR.
In an attempt to establish the optimum substitution patterns for
the possible dimethoxyphenyl (2,3-, 2,4-, 2,5-, 2,6-, 3,4- and 3,5-),
as well as trimethoxyphenyl derivatives, compounds 4h-o were
prepared and evaluated. A second methoxy group on the phenyl
rings led to the identification of very potent (low nanomolar
affinity) and completely selective ligands (e.g., compounds 4i, 4j,
4l). The first remarkable observation is the ability of the
4-methoxyphenyl substitution to accommodate a second methoxy
group in either ortho (4i) or meta (4l) without affecting the affinity
and selectivity of the reference monosubstituted compound (4d).
Even the 3,4,5-trimethoxyphenyl (4o) retains this high selectivity
profile, while experiencing a less than 10- and 5-fold drop in A3AR
affinity, as compared to the mono (4d) or dimethoxyphenyl (4l)
analogs respectively. Not surprisingly, compound 4o displays an
intermediate affinity when compared to the 3,4- (4l) and 3,5-
dimethoxyphenyl (4n) analogs. The fact that all these bi- and tri-
substituted compounds retain an excellent selectivity profile is
again in agreement with the deeper subpocket identified for these
voluminous substitutions at L2 in the A3AR, as compared to other
AR subtypes (see Fig. 3). Indeed, the limited selectivity observed for
the piperonyl analog 4m, which on the other side retains the
6. Conclusion
In summary, we have documented a comprehensive exploration
of the N-(2,6-diarylpyrymidin-4-yl)acetamide scaffold with several
combinations of methoxyphenyl residues at positions 2 and 6 of the
heterocyclic core, reporting a series of novel optimized potent and
selective A3AR antagonists. The design of the current series was
guided by our mixed structure-based/3D-QSAR computational
pipeline, which not only demonstrated a good predictive power,
but also shown its utility for interpreting the biological data here
reported concerning the four human ARs. Two new molecules are
particularly promising, 4j (ISVY350) and 4l (ISVY345), which
combine excellent affinities for the hA3AR (Ki ꢁ 7 nM) and highly
selective profiles among ARs.
7. Experimental protocols
7.1. Computational methods
7.1.1. Structure-based methods: receptor preparation and ligand
docking
Several programs and servers were employed for the generation
of the homology model of A3AR as originally reported in ref. [20].
These include (i) ClustalX for the sequence alignment with the
template A2AAR (PDB code 3EML) [3], (ii) Modeler for the generation
and selection of homology models and loop refinement procedures
[26], (iii) Molprobity [35] and PDB2PQR [36] servers for the assess-
ment of Asn/Gln/His rotamers, and side chain protonation states,
and finally (iv) tools from Schrödinger Suite for energetic structural
refinements [37]. Steps (iii) and (iv) were performed also for the
refinement of the crystal structure of A2AAR (PDB code 3EML).
Molecular dockings were performed with GOLD v4 software [27].
A sphere of 15 Ǻ was centered in the C-g of the conserved residue Ile
7.39, ensuring a proper search grid in the binding site of the receptor.
20 genetic algorithm runs were performed for each ligand,

V. Yaziji et al. / European Journal of Medicinal Chemistry 59 (2013) 235e242
241
employing ChemScore as scoring function, and allowing rotation of
the protein hydroxyl groups. For the A2AAR structure, additional side
chain rotamers at residues Glu 5.30 and His 7.29 were considered,
taking into account the induced-fit effect observed in the crystal
structure of A2AAR in complex with a relatively similar chemical
scaffold, the1,2,4-triazine derivative [34](PDB code3UZC). Fig. 3 was
prepared with PyMOL v1.2, including the structural superimposition
of A2AAR and A3AR (‘super’ command) and the calculation of Con-
7.2.1. General procedure for the synthesis of compounds 4
A mixture of the amide 2 (0.43 mmol), boronic acids 3
(1.3 mmol), Pd(PPh3)4 (0.043 mmol), and Na2CO3 (2.1 mmol) in
5 mL of a mixture of DME/H20 (3:1) in coated Kimble vials was
stirred with orbital stirring at 110 ꢃC for 12 h. The resulting mixture
was concentrated in vacuo and then ethyl acetate was added. This
solution was washed with water and NaOH 1 N to remove the acid
boronic excess. The organic layer was collected, dried over Na2SO4,
filtered and concentrated in vacuo. The residue was purified by
column chromatography or preparative method on silica gel.
ꢀ
nolly surfaces of their binding sites with the default 1.4 A probe.
7.1.2. 3D-QSAR
The software Pentacle [29] was employed for the generation of
3D-QSAR models (see Table 1) following previously published
procedures by us [20]. The model originally reported in ref. [9]
(model I) was extended with the tested molecules in that same
effort (model II), and used with predictive purposes in this work
(see Figs. 2 and 4A). Finally, a partial least squares model consid-
ering all diaryl-pyrimidines that found satisfactory binding modes
on A3AR (model III) was built in order to extract the main phar-
macophoric elements of the series (Fig. 4B). The common meth-
odology employed in these 3D-QSAR models includes four
consecutive stages, where we used default options unless specified:
(i) computation of the molecular interaction fields (MIFs) with
selected chemical probes from the GRID force field, in the present
case DRY, O, N1 and TIP probes, (ii) discretization of the MIFs with
the AMANDA algorithm [28], (iii) consistency large auto and cross
7.2.2. N-(2,6-Bis(3-methoxyphenyl)pyrimidin-4-yl)acetamide (4c)
Yield: 93%. Mp 135e136 ꢃC. 1H NMR (CDCl3)
d (ppm): 2.14 (3H, s,
CH3), 3.86 (3H, s, OCH3), 3.89 (3H, s, OCH3), 7.00e7.05 (2H, m,
Aromatics), 7.38 (2H, q, J ¼ 7.8 Hz, Aromatics), 7.78 (2H, q, J ¼ 7.8 Hz,
Aromatics), 8.02e8.09 (2H, m, Aromatics), 8.47 (1H, s, NH), 8.68
(1H, s, H5). 13C NMR (CDCl3)
d (ppm): 169.6, 165.6, 163.5, 159.8,
159.7, 158.2, 138.7, 138.4, 129.7, 129.4, 120.5, 119.8, 116.7, 116.4, 112.9,
112.6, 103.3, 55.29, 55.23, 24.7. MS m/z: 350 (MHþ) (100). HRMS
(ESI) calculated for C20H20N3O3 350.1426, found 350.1492.
7.2.3. N-(2,6-Bis(4-(trifluoromethoxy)phenyl)pyrimidin-4-yl)
acetamide (4e)
Yield: 81%. Mp 146e147 ꢃC. 1H NMR (CDCl3)
d
(ppm): 2.31 (3H, s,
CH3), 7.31e7.37 (4H, t, J ¼ 8.2 Hz, Aromatics), 8.07 (1H, s, NH), 8.26
(2H, d,
J
¼
8.7 Hz, Aromatics), 8.50e8.54 (3H, m, 2H
correlation (CLACC), which guarantees that
a
given variable
Aromatics þ 1H, H5). 13C NMR (CDCl3)
d (ppm): 169.3, 164.4, 162.6,
represents exactly the same information for every compound of the
series, in this case generated based on the biological superimpo-
sition of the ligands using the “strict” options, and (iv) two rounds
of fractional factorial design were applied for the selection of the
most relevant variables in the model.
158.2, 151.1, 151.0, 148.0, 147.5, 135.5, 135.3, 129.6, 128.9, 120.8,
120.4, 103.0, 24.8. MS m/z: 458 (MHþ) (100). HRMS (ESI) calculated
for C20H14F6N3O3 458.0861, found 458.0938.
7.2.4. N-(2,6-bis(4-ethoxyphenyl)pyrimidin-4-yl)acetamide (4f)
Yield: 76%. Mp 155e156 ꢃC. 1H NMR (CDCl3)
d (ppm): 1.44 (6H, t,
7.2. Chemistry
J ¼ 6.7 Hz, 2ꢂ CH2CH3), 2.17 (3H, s, CH3), 4.11 (4H, q, J ¼ 6.7 Hz, 2ꢂ
CH2CH3), 6.95e7.01 (4H, m, Aromatics), 8.20 (2H, d, J ¼ 8.6 Hz,
Commercially available starting materials, reagents and solvents
were purchased (SigmaeAldrich) and used without further purifi-
cation. After extraction from aqueous phases, the organic solvents
were dried over anhydrous sodium sulfate. The reactions were
monitored by thin-layer chromatography (TLC) with 2.5 mm Merck
silica gel GF 254 strips, and the purified compounds each showed
a single spot; unless stated otherwise, UV light and/or iodine vapor
were used for detection of compounds. The Suzuki cross-coupling
reactions were performed in coated Kimble vials on a PLS (6 ꢂ 4)
Organic Synthesiser with orbital stirring. Purity and identity of all
tested compounds were established by a combination of HPLC,
mass spectrometry and NMR spectra. Purification of isolated
products was carried out by column chromatography (Kieselgel
0.040e0.063 mm, E. Merck) or medium pressure liquid chroma-
tography (MPLC) on a CombiFlash Companion (Teledyne ISCO) with
Aromatics), 8.36e8.43 (4H, m, 2H Aromatics þ 1H H5 þ 1H NH). 13
C
NMR (CDCl3)d(ppm): 169.4,165.2,163.4,161.2,161.1,157.9,129.9,129.6,
129.4, 128.8, 114.4, 114.1, 101.2, 63.4, 24.7, 14.6. MS m/z: 378 (MHþ)
(100). HRMS (ESI) calculated for C22H24N3O3 378.1739, found 378.1818.
7.2.5. N-(2,6-Bis(4-methoxystyryl)pyrimidin-4-yl)acetamide (4g)
Yield: 37%. Mp 134e135 ꢃC. 1H NMR (CDCl3)
d (ppm): 2.24 (3H, s,
CH3), 3.84 (6H, s, 2ꢂ OCH3), 6.93 (4H, d, J ¼ 8.1 Hz, Aromatics), 7.01
(2H, dd, J ¼ 16.15 Hz, J ¼ 4.65 Hz, CH]CH), 7.55 (4H, d, J ¼ 8.3 Hz,
Aromatics), 7.88 (2H, dd, J ¼ 15.9 Hz, J ¼ 5.07 Hz, CH]CH), 7.96 (1H, s,
H5), 8.09 (1H, s, NH). 13C NMR (CDCl3)
d (ppm): 169.4, 164.5, 164.0,
160.4,160.3,157.3,137.5,137.4,136.3,129.0,128.9,128.6,128.5,124.9,
124.1, 114.1, 103.9, 55.2, 55.1, 24.8. MS m/z: 402 (MHþ) (100). HRMS
(ESI) calculated for C24H24N3O3 402.1739, found 402.1815 (Mþ).
RediSep pre-packed normal-phase silica gel (35e60
mm) columns
7.2.6. N-(2,6-Bis(2,3-dimethoxyphenyl)pyrimidin-4-yl)acetamide
followed by recrystallization. Melting points were determined on
a Gallenkamp melting point apparatus and are uncorrected. The
NMR spectra were recorded on Bruker AM300 and XM500 spec-
(4h)
Yield: 76%. Mp 105e106 ꢃC. 1H NMR (CDCl3)
d (ppm): 2.19 (3H, s,
CH3), 3.86e3.93 (12H, m, 4ꢂ OCH3), 7.03 (2H, d, J ¼ 6.7 Hz,
Aromatics), 7.13 (2H, t, J ¼ 7.8 Hz, Aromatics), 7.30 (1H, d, J ¼ 7.8 Hz,
Aromatics), 7.55 (1H, d, J ¼ 7.8 Hz, Aromatics), 8.31 (1H, s, NH), 8.63
trometers. Chemical shifts are given as
d values against tetrame-
thylsilane as internal standard and J values are given in Hz. Mass
spectra were obtained on a Varian MAT-711 instrument. High
resolution mass spectra were obtained on an Autospec Micromass
spectrometer. Analytical HPLC was performed on an Agilent 1100
(1H, s, H5). 13C NMR (CDCl3)
d (ppm): 169.2, 164.6, 164.4, 157.2,
153.2, 153.0, 148.0, 147.5, 133.6, 132.0, 123.8, 122.6, 122.4, 114.0,
113.5, 107.6, 61.5, 61.1, 55.9, 24.6. MS m/z: 409 (98). HRMS (EI)
calculated for C22H23N3O5 409.1638, found 409.1651.
system using an Agilent Zorbax SB-Phenyl, 2.1 mm ꢂ 150 mm, 5
mm
column with gradient elution using the mobile phases (A) H2O
containing 0.1% CF3COOH and (B) MeCN and a flow rate of
1 mL/min. The purity of all tested compounds was determined to be
>95%. Acetylation of 4-amino-2,6-dichloropyrimidine (1) was near
quantitative (97%).
7.2.7. N-(2,6-Bis(2,5-dimethoxyphenyl)pyrimidin-4-yl)acetamide
(4j, ISVY350)
Yield: 57%. Mp 176e177 ꢃC. 1H NMR (CDCl3)
d
(ppm): 2.10 (3H, s,
CH3), 3.77e3.90 (12H, m, 4ꢂ OCH3), 6.96 (4H, s, Aromatics), 7.32

242
V. Yaziji et al. / European Journal of Medicinal Chemistry 59 (2013) 235e242
(1H, s, Aromatics), 7.66 (1H, s, Aromatics), 8.72 (2H, s, 1H, H5 þ 1H,
NH). 13C NMR (CDCl3)
(ppm): 169.6, 164.2, 157.2, 153.5, 153.4,
152.3, 151.7, 128.9, 127.1, 117.1, 116.5, 116.2, 115.7, 113.4, 112.8, 108.1,
56.3, 56.1, 55.7, 24.3. MS m/z: 410 (MHþ) (100). HRMS (ESI) calcu-
lated for C22H24N3O5 409.1638, found 410.1702.
References
d
[1] B.B. Fredholm, A.P. Ijzerman, K.A. Jacobson, J. Linden, C.E. Muller, Pharmacol.
Rev. 63 (2011) 1e34.
[2] K.A. Jacobson, Z.-G. Gao, Nat. Rev. Drug Discov. 5 (2006) 247e264.
[3] V.-P. Jaakola, M.T. Griffith, M.A. Hanson, V. Cherezov, E.Y.T. Chien, J.R. Lane,
A.P. Ijzerman, R.C. Stevens, Science 322 (2008) 1211e1217.
[4] J. Linden, Trends Pharmacol. Sci. 15 (1994) 298e306.
[5] B.T. Liang, K.A.A. Jacobson, Proc. Natl. Acad. Sci. U. S. A. 9 (1998) 6995e6999.
[6] R. Akkari, J.C. Burbiel, J. Hockemeyer, C.E. Müller, Curr. Top. Med. Chem. 6
(2006) 1375e1399.
[7] C.H. Mitchell, K. Peterson-Yantorno, D.A. Carré, A.M. McGlinn, M. Coca-Prados,
R.A. Stone, M.M. Civan, Am. J. Physiol. 276 (1999) 659e666.
[8] P.G. Baraldi, M.A. Tabrizi, R. Romagnoli, F. Fruttarolo, S. Merighi, K. Varani,
S. Gessi, P.A. Borea, Curr. Med. Chem. 12 (2005) 1319e1329.
[9] S. Gessi, E. Cattabriga, A. Avitabile, R. Gafà, G. Lanza, L. Cavazzini, N. Bianchi,
R. Gambari, C. Feo, A. Liboni, S. Gullini, E. Leung, S. Mac-Lennan, P.A. Borea,
Clin. Cancer Res. 10 (2004) 5895e5901.
[10] P. Fishman, S. Bar-Yehuda, Curr. Top. Med. Chem. 3 (2003) 463e469.
[11] S. Gessi, S. Merighi, K. Varani, E. Leung, S.M. Lennan, P.A. Bore, Pharmacol.
Ther. 117 (2008) 123e140.
[12] K.A. Jacobson, A.M. Klutz, D.K. Tosh, A.A. Ivanov, D. Preti, P.G. Baraldi,
Medicinal chemistry of the A3 adenosine receptor: agonists, antagonists, and
receptor engineering, adenosine receptors in health and disease, in:
B.B. Fredholm (Ed.), Handbook of Experimental Pharmacology, vol. 193,
Springer, 2009, pp. 123e159.
[13] P.G. Baraldi, D. Preti, P.A. Borea, K. Varani, J. Med. Chem. 55 (2012) 5676e5703.
[14] P.G. Baraldi, M.A. Tabrizi, F. Fruttarolo, A. Bovero, B. Avitabile, D. Preti,
R. Romagnoli, S. Merighi, S. Gessi, K. Varani, P.A. Borea, Drug Dev. Res. 58
(2003) 315e329.
[15] F. Luan, F. Borges, M.N.D.S. Cordeiro, Curr. Top. Med. Chem. 12 (2012) 878e894.
[16] K. Nieber, A. Lewerenz, S. Hentschel, Z. Vissiennon, Bioforum25 (2002) 237e240.
[17] T.W. Stone, Adv. Exp. Med. Biol. 513 (2002) 249e280.
[18] S. Merighi, P. Mirandola, K. Varani, S. Gessi, S. Capitani, E. Leung, P.G. Baraldi,
M.A. Tabrizi, P.A. Borea, Biochem. Pharmacol. 66 (2003) 739e748.
[19] P.G. Baraldi, M.A. Tabrizi, S. Gessi, P.A. Borea, Chem. Rev. 108 (2008) 238e263.
[20] V. Yaziji, D. Rodriguez, H. Gutierrez-de-Teran, A. Coelho, O. Caamano,
X. Garcia-Mera, J. Brea, M.I. Loza, M.I. Cadavid, E. Sotelo, J. Med. Chem. 54
(2011) 457e471.
[21] L.C.W. Chang, R.F. Spanjersberg, J.K. von Kuenzel, T. Mulder-Krieger, G. van
den Hout, M.W. Beukers, J. Brussee, A.P. Ijzerman, J. Med. Chem. 47 (2004)
6529e6540.
[22] B. Cosimelli, G. Greco, M. Ehlardo, E. Novellino, F. Da Settimo, S. Taliani, C. La
Motta, M. Bellandi, T. Tuccinardi, A. Martinelli, O. Ciampi, M.L. Trincavelli,
C. Martini, J. Med. Chem. 51 (2008) 1764e1770.
[23] K.Y. Jung, S.K. Kim, Z.G. Gao, A.S. Gross, N. Melman, K.A. Jacobson, Y.C. Kim,
Bioorg. Med. Chem. 12 (2004) 613e623.
[24] O. Lenzi, V. Colotta, D. Catarzi, F. Varano, G. Filacchioni, C. Martini,
L. Trincavelli, O. Ciampi, K. Varani, F. Marighetti, E. Morizzo, S. Moro, J. Med.
Chem. 49 (2006) 3916e3925.
[25] S. Taliani, C. La Motta, L. Mugnaini, F. Simorini, S. Salerno, A.M. Marini, F. Da
Settimo, S. Cosconati, B. Cosimelli, G. Greco, V. Limongelli, L. Marinelli,
E. Novellino, O. Ciampi, S. Daniele, M.L. Trincavelli, C. Martini, J. Med. Chem. 53
(2010) 3954e3963.
[26] A. Sali, T.L. Blundell, J. Mol. Biol. 234 (1993) 779e815.
[27] M.L.Verdonk, J.C. Cole, M.J. Hartshorn, C.W. Murray,R.D. Taylor,Proteinse Struct.
Funct. Genet. 52 (2003) 609e623.
[28] A. Duran, G.C. Martinez, M. Pastor, J. Chem. Inf. Model. 48 (2008) 1813e1823.
[29] Pentacle 1.1, Molecular Discovery, Ltd, Perugia, Italy, 2009.
[30] V. Yaziji, A. Coelho, A. El Maatougui, J. Brea, M.I. Loza, X. García-Mera, E. Sotelo,
J. Comb. Chem. 11 (2009) 519e522.
[31] N. Miyara, T. Yanagi, A. Suzuki, Synth. Commun. 11 (1981) 513e519.
[32] M.P. Bosch, F. Campos, I. Niubo, G. Rosell, J.L. Diaz, J. Brea, M.I. Loza,
A. Guerrero, J. Med. Chem. 47 (2004) 4041e4053.
7.2.8. N-(2,6-Bis(2,6-dimethoxyphenyl)pyrimidin-4-yl)acetamide
(4k)
Yield: 27%. Mp 184e185 ꢃC. 1H NMR (CDCl3)
d (ppm): 1.97 (3H, s,
CH3), 3.67 (6H, s, 2ꢂ OCH3), 3.72 (6H, s, 2ꢂ OCH3), 6.57e6.59 (4H,
m, Aromatics), 7.27 (2H, s, Aromatics), 8.06 (1H, s, H5), 9.08 (1H, s,
NH). 13C NMR (CDCl3)
d (ppm): 169.5, 165.4, 163.8, 161.2, 160.7,
159.2, 138.9, 138.7, 106.2, 105.8, 103.6, 103.4, 103.1, 55.40, 55.36,
24.7. MS m/z: 410 (MHþ) (100). HRMS (ESI) calculated for
C22H24N3O5 410.1638, found 410.1717.
7.2.9. N-(2,6-Bis(3,4-dimethoxyphenyl)pyrimidin-4-yl)acetamide
(4l, ISVY345)
Yield: 85%. Mp 147e148 ꢃC. 1H NMR (CDCl3)
d (ppm): 2.26 (3H, s,
CH3), 3.96 (6H, s 2ꢂ OCH3), 4.00 (6H, s, 2ꢂ OCH3), 6.98 (2H, d,
J ¼ 8.4 Hz, Aromatics), 7.82 (1H, d, J ¼ 8.4 Hz, Aromatics), 7.86 (1H, s,
Aromatics), 8.05 (1H, s, Aromatics), 8.12 (1H, d, J ¼ 8.4 Hz,
Aromatics), 8.26 (1H, s, NH), 8.39 (1H, s, H5). 13C NMR (CDCl3)
d
(ppm): 169.4, 165.2, 163.2, 157.8, 151.3, 151.2, 148.9, 148.6, 130.1,
129.8, 121.3, 120.6, 110.7, 110.5, 109.9, 109.8, 101.6, 55.8, 55.7, 24.8.
MS m/z: 410 (MHþ) (100). HRMS (ESI) calculated for C22H24N3O5
410.1638, found 410.1710.
7.2.10. N-(2,6-Bis(3,5-dimethoxyphenyl)pyrimidin-4-yl)acetamide
(4n)
Yield: 98%. Mp 186e187 ꢃC. 1H NMR (CDCl3)
d (ppm): 2.27 (3H, s,
CH3), 3.89 (12H, s, 4ꢂ OCH3), 6.61 (2H,s, Aromatics), 7.38 (2H, s,
Aromatics), 7.67 (2H, s, Aromatics), 8.13 (1H, s, NH), 8.45 (1H, s, H5).
13C NMR (CDCl3)
d (ppm): 169.4, 165.5, 163.2, 160.9, 160.7, 158.0,
139.3, 139.1, 105.8, 105.3, 103.4, 103.2, 102.8, 55.39, 55.34, 24.7. MS
m/z: 409 (72). HRMS (EI) calculated for C22H23N3O5 409.1638, found
409.1642.
7.2.11. N-(2,6-Bis(3,4,5-trimethoxyphenyl)pyrimidin-4-yl)
acetamide (4o)
Yield: 95%. Mp 199e200 ꢃC. 1H NMR (CDCl3)
d (ppm): 2.30 (3H, s,
CH3), 3.93 (6H, s, 2ꢂ OCH3), 3.98 (12H, s, 4ꢂ OCH3), 7.48 (2H, s,
Aromatics), 7.79 (2H, s, Aromatics), 8.15 (1H, s, NH), 8.42 (1H, s, H5).
13C NMR (CDCl3)
d (ppm): 169.4, 165.2, 163.0, 158.0, 153.3, 153.0,
140.4, 132.5, 132.4, 105.1, 105.0, 104.4, 104.3, 102.4, 60.8, 56.0, 55.9,
24.8. MS m/z: 469 (100). HRMS (EI) calculated for C24H27N3O7
469.1849, found 469.1852.
Acknowledgments
[33] J.A. Ballesteros, H. Weinstein, Methods Neurosci. 25 (1995) 366e428.
[34] M. Congreve, S.P. Andrews, A.S. Dore, K. Hollenstein, E. Hurrell, C.J. Langmead,
J.S. Mason, I.W. Ng, B. Tehan, A. Zhukov, M. Weir, F.H. Marshall, J. Med. Chem.
55 (2012) 1898e1903.
[35] I.W. Davis, A. Leaver-Fay, V.B. Chen, J.N. Block, G.J. Kapral, X. Wang,
L.W. Murray, W.B. Arendall, J. Snoeyink, J.S. Richardson, D.C. Richardson,
Nucleic Acids Res. 35 (2007) W375eW383.
[36] T.J. Dolinsky, J.E. Nielsen, J.A. McCammon, N.A. Baker, Nucleic Acids Res. 32
(2004) W665eW667.
[37] Schrödinger, L.L.C., New York, NY, 2009.
This work has been developed in the frame of the Red Gallega de
Investigación y Desarrollo de Medicamentos (REGID) and was finan-
cially supported by the Ministerio de Ciencia e Innovación (MICINN
grant SAF2011-30104) and the Xunta de Galicia (09CSA016234PR).
D.R.isrecipientof apredoctoralgrantfromtheFondodeInvestigación
en Salud (ISCIII). H.G.T. and A.C. are researchers of the Isidro Parga
Pondal program (Xunta de Galicia, Spain).