1,3-Dialkyl-8-N-substituted benzyloxycarbonylamino-9-deazaxanthines as potent adenosine receptor ligands: Design, synthesis, structure-affinity and structure-selectivity relationships

Article abstract of DOI:10.1016/j.bmc.2009.03.062

A number of 1,3-dialkyl-9-deazaxanthines (9-dAXs), bearing a variety of N-substituted benzyloxycarbonylamino substituents at position 8, were prepared and evaluated for their binding affinity to the recombinant human adenosine receptors (hARs), chiefly to

Full text of DOI:10.1016/j.bmc.2009.03.062

Bioorganic & Medicinal Chemistry 17 (2009) 3618–3629
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
1,3-Dialkyl-8-N-substituted benzyloxycarbonylamino-9-deazaxanthines
as potent adenosine receptor ligands: Design, synthesis, structure–affinity
and structure–selectivity relationships
Franco Fernández a, Olga Caamaño a, M. Isabel Nieto a, Carmen López a, Xerardo García-Mera a,
Angela Stefanachi c, Orazio Nicolotti c, M. Isabel Loza b, Jose Brea b, Cristina Esteve d,
Victor Segarra d, Bernat Vidal d, Angelo Carotti c,
*
a Departamento de Química Orgánica, Facultade de Farmacia, Universidade de Santiago de Compostela, 15782 Santiago de Compostela, Spain
b Departamento de Farmacologia, Instituto de Farmacia Industrial, Universidade de Santiago de Compostela, 15782 Santiago de Compostela, Spain
c Dipartimento Farmaco-chimico, Università degli Studi di Bari, via Orabona 4, I-70125 Bari, Italy
d Medicinal Chemistry Department, Laboratorios Almirall, S.A., E-08960 St. Just Desvern, Barcelona, Spain
a r t i c l e i n f o
a b s t r a c t
Article history:
A number of 1,3-dialkyl-9-deazaxanthines (9-dAXs), bearing a variety of N-substituted benzyloxycarbon-
ylamino substituents at position 8, were prepared and evaluated for their binding affinity to the recom-
binant human adenosine receptors (hARs), chiefly to the hA2B and hA2A AR subtypes. Several ligands
endowed with excellent binding affinity to the hA2B receptors, but low selectivity versus hA2A and hA1
were identified. Among these, 1,3-dimethyl-N-30-thienyl carbamate 15 resulted as the most potent ligand
at hA2B (Ki = 0.8 nM), with a low selectivity versus hA2A (hA2A/hA2B = 12.6) and hA1 (hA1/hA2B = 12.5) and a
higher selectivity versus hA3 (hA3/hA2B = 454). When tested in functional assays in vitro, compound 15
exhibited high antagonist activities and efficacies versus both the A2A and A2B receptor subtypes, with
pA2 values close to the corresponding pKis. A comparative analysis of structure–affinity and structure–
selectivity relationships of the similar analogues 8-N-substituted benzyloxycarbonylamino- and 8-N-
substituted phenoxyacetamido-9-dAXs suggested that their binding modes at the hA2B and hA2A ARs
may strongly differ. Computational studies help to clarify this striking difference arising from a simple,
albeit crucial, structural change, from CH2OCON to OCH2CON, in the para-position of the 8-phenyl ring.
Ó 2009 Elsevier Ltd. All rights reserved.
Received 9 February 2009
Revised 30 March 2009
Accepted 31 March 2009
Available online 5 April 2009
Keywords:
9-Deazaxanthines
Carbamates
Adenosine receptors (ARs)
Adenosine receptor antagonists
1. Introduction
for example Alzheimer’s disease, cystic fibrosis and type-II
diabetes.2
Adenosine is an endogenous purine nucleoside that mediates
several important physiological effects by activating four diverse
G protein-coupled adenosine receptors (ARs), named A1, A2A, A2B
and A3.1 The medicinal chemistry, pharmacology and potential
therapeutic applications of subtype selective AR ligands have been
extensively reviewed.2–4
In the search for new antiasthmatic drugs we have been target-
ing A2B AR antagonists based on the observation that the bron-
chodilating activity of two xanthinic drugs, that is, theophylline
and enprofylline, might be ascribed to selective, albeit small,
antagonism versus the A2B AR subtype.5,6 Since activation of A2B
ARs stimulates vasodilation, cardiac fibroblast proliferation, hepa-
tic glucose and interleukin-6 biosynthesis and Clꢀ secretion in
intestinal epithelia, it can be hypothesized that potent and selec-
tive A2B AR antagonists might be used in many other pathologies,
Several groups have designed and tested many xanthine deriv-
atives aiming at the discovery of new, potent and A2B-selective li-
gands.3,4,7–9 Good results have been initially obtained by Jacobson
and co-workers who described many xanthine derivatives, such as
XAC (1)10 and MRS1754 (2)11 (Chart 1) that were endowed with
high affinity and selectivity versus the A2B AR. Along this line of re-
search, we recently reported the design, synthesis, structure–activ-
ity (SAR) and structure–selectivity (SSR) relationships of a large
series of 9-deazaxanthines (9-dAXs) and 9-OH-9-dAXs,12–16 two
poorly exploited classes of xanthine analogues, bearing the same
8-N-substituted-phenoxyacetamido moiety proposed by Jacobson.
In contrast with the biological data reported for 9-dAXs differently
substituted at position 8,17,18 these compounds showed both high-
er affinity and selectivity versus the hA2B AR (see compounds 3,14
414 and 516 in Chart 1).
Similar results were obtained by other groups for differently
8-substituted xanthines (e.g., compound 619) and 9-dAX (e.g., com-
pound 720).
* Corresponding author. Tel.: +39 080 5442782; fax: +39 080 5442230.
E-mail address: carotti@farmchim.uniba.it (A. Carotti).
0968-0896/$ - see front matter Ó 2009 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2009.03.062

F. Fernández et al. / Bioorg. Med. Chem. 17 (2009) 3618–3629
3619
Yields in the cyclization products varied considerably depend-
ing on the starting 6-styryluracil, that is, high for the carboxy-
phenyl derivatives 11a (80%) and 11b (90%) and dramatically
reduced for 11d (18%) and the (hydroxymethyl)phenyl derivative
11c (17%), which was obtained as formate. Reduction of the car-
boxylic group of 11a, b, d with BH3ꢁTHF afforded the corresponding
9-dAX alcohols 12. Conversion of intermediates 12 to the target
compounds 13–32 was achieved by employing different methods
(Scheme 1). The first carbamate formation procedure (method A)
involved the reaction of the corresponding 9-dAX alcohol 12 with
an excess of the appropriate isocyanate in DMF at 115 °C. In the
second procedure (method B), the 9-dAX alcohol was treated
with the appropriate isocyanate in dioxane at reflux. The third
procedure (method C) involved the formation of an intermediate
imidazolide, by reaction of the alcohols 12 with N,N0-carbonyldiim-
idazole (CDI) in pyridine,24 followed by a reaction with the ade-
quate amine.25
O
H
N
Y
R1
N
OCH2COR
Z
O
N
R3
1
3
1, XAC: R = R = CH CH CH ; Z= N; Y= H; R =
2
2
3
NH(CH ) NH
2 2
2
1
3
2, MRS1754: R = R = CH CH CH ; Z= N; Y= H; R=
2
2
3
NHC H -4-CN
6
4
1
3
3, R = R = CH ; Z= CH; Y= H; R= NHC H -4-Br
3
6 4
1
3
4, R = R = CH ; Z= CH; Y= OMe; R= NHC H -4-Br
3
6 4
1
3
5, R = R = CH CH CH ; Z= COH; Y=H; R= NHC H -4-Br
2
2
3
6
4
O
N
H
N
CF3
N
N
N
N
O
6
3. Biochemical and pharmacological assays
Cl
O
S
Affinity of compounds 13–32 to human A1, A2A, A2B and A3 ARs
was evaluated as described in our previous investigations.13–16 All
active compounds fully displaced the specific binding of radioli-
gands to receptors in a concentration-dependent manner; slopes
were not significantly different from unity at the 5% level of statis-
tical significance (Fig. 1). High antagonist activity of ligand 15 was
measured by means of isolated organ assays at rat colon A2B (Fig. 2)
and rat aorta A2A AR subtypes. The compound concentration-
dependently displaced the curves of the AR agonist NECA to the
right in a parallel way without lowering their maximum, in good
agreement with a competitive antagonism.
H
N
O
S
N
N
N
O
N
O
7
Chart 1. Xanthine and 9-dAX derivatives with potent and selective antagonist
activities versus the A2B AR subtype.
To further extend the SAR study of 9-dAXs as AR ligands and
identify a new class of selective A2B AR antagonists, we designed
a series of 1,3-dialkyl-8-N-substituted benzyloxycarbonylamino-
9-dAXs (Chart 2) by replacing the phenoxyacetamido group at po-
sition 8 in compounds 3 and 4 with a benzyloxycarbonylamino
group. Synthesis, biological evaluation, and SAR and SSR studies
of this new class of AR ligands were the objectives of the present
study.
4. Results and discussion
4.1. SAR and SSR studies
Chemical structures and human AR binding affinities of the
examined ligands are reported in Table 1. Binding affinity values
are the means of two independent experiments performed with
duplicate points. SEM values were always lower than 10%. It must
be kept in mind that the main goal of our investigation was to dis-
cover new and highly potent hA2B antagonists13–16 endowed with
good selectivity versus hA2A, as measured by the selectivity index
(SI) which is the ratio Ki hA2A/Ki hA2B. Since most of the ligand
binding affinities to both the hA2B and hA2A ARs turned to be good
and rather close, resulting in very low SIs (see Table 1), the affini-
ties to hA1 and hA3 ARs were determined only for parent com-
pounds and for ligands with high hA2B affinity or relatively high
SI. In Table 1, only the percentage of radioligand displacement at
2. Chemistry
Synthesis of the target compounds 1,3-dialkyl-8-substituted-9-
dAX carbamates, (1,3-dialkyl-2,4-dioxo-1H-pyrrolo[3,2-d]pyrimi-
din-6-yl)phenyl]methylcarbamates), 13–32, performed according
to published procedures,14,21–23 is outlined in Scheme 1. First, the
condensation of uracil derivatives 8 with para-substituted benzal-
dehydes 9 afforded the appropriate 6-styryluracils 10, which were
treated with sodium dithionite in formic acid under reflux to un-
dergo a reductive cyclization to the corresponding 9-dAX deriva-
tives 11.
1 lM was reported for low active compounds or for compounds
whose low solubility prevented the determination of their Ki.
For a more immediate and efficient analysis of the SAR and SSR,
binding data were presented graphically as a plot of pKi hA2A (Y-
axis) versus pKi hA2B (X-axis) using the same scale and range for
both axes (square plot in Fig. 3); note that for the affinity of ligand
29 at hA2B an estimated pKi = 6.05 was used. On the bisector of the
plot (Y = X) compounds with equal affinities at both ARs are lo-
cated, whereas highly hA2B-selective compounds are observed far
below the bisector. The distance of their pKi values from the bisec-
tor line is a direct measure of the degree of selectivity: the greater
the distance the higher the SI values in Table 1.
O
H
R
1
N
N
O
O
O
N
R
NR
4
5
R
3
13-32
R1 = or≠ R3
R4 = or≠ R5
Alkyl
H, Alkyl, (Hetero)aryl
At a glance, binding data in Figure 3 reveal that ligands with
good affinities for both the hA2B and hA2A AR subtypes were ob-
tained. Indeed, 14 of 20 newly synthesized compounds, comprised
Chart 2. General structure and main structural variations of 1,3-dialkyl-8-N-
substituted benzyloxycarbonylamino-9-dAX AR ligands 13–32.

3620
F. Fernández et al. / Bioorg. Med. Chem. 17 (2009) 3618–3629
O
CHO
O
R1
NO2
Me
R1
NO2
a
N
N
O
N
O
N
R3
R2
R3
R2
10a (R1 = R3 = Me, R2 = CO2H)
10b (R1 = R3 = Pr, R2 = CO2H)
10c (R1 = R3 = Pr, R2 = CH2OH)
10d (R1 = Pr, R3 = Me, R2 = CO2H)
8a (R1 = R3 = Me)
8b (R1 = R3 = Pr)
8c (R1 = Pr, R3 = Me)
9a (R2 = CO2H)
9b (R2 = CH2OH)
b
O
H
O
H
R1
R1
O
N
N
c
N
N
R2
CH2OH
N
O
N
R3
R3
12a (R1 = R3 = Me)
12b (R1 = R3 =Pr)
12d (R1 = Pr, R3 = Me)
11a (R1 = R3 = Me, R2 = CO2H)
11b (R1 = R3 = Pr, R2 = CO2H)
11c (R1 = R3 = Pr, R2 = CH2OCHO)
11d (R1 = Pr, R3 = Me, R2 = CO2H)
d
O
H
N
R1
O
N
O
O
N
NR4R5
R3
13-32
Scheme 1. Reagents and conditions: (a) piperidine, dioxane,
R4NCO, dioxane, 100 °C; Method C: CDI, R4R5NH, pyridine, rt.
D
; (b) HCO2H, Na2S2O4, reflux; (c) BH3ꢁTHF 1 M, THF, 0 °C, then rt; (d) Method A: R4NCO, DMF, 115 °C; Method B:
100
75
50
25
0
125
100
75
50
25
0
-12 -11 -10
-9
-8
-7
-6
-5
-11 -10 -9
-8
-7
-6
-5
-4
-3
Log [NECA] (M)
log [Antagonist] (M)
Figure 2. Isolated organ assays for rat A2B receptors. Cumulative concentration-
response curves to NECA in the absence (j) and in the presence (N) of 15 at 10 nM
concentration. Points represent the means SEM (vertical bars) of two independent
assays (n = 2).
Figure 1. Binding competition experiments at cloned hA2B ARs. Concentration–
response curve for compound 15. Values represent the means SEM (vertical bars)
of two independent experiments.
in the cyan-highlighted area of the plot, yielded a pKi > 6.85
(Ki < 140 nM).
Noticeably, all the ligands tested for the hA1 AR subtype exhibited
a low nanomolar activity, being the 1,3-dipropyl-N-p-F-phenyl
derivative 23 the most potent ligand with a Ki = 0.4 nM. Analysis
of selectivity data in the plot, that is, the distances of pKi values
from the bisector line, revealed that no ligand exhibited a substan-
tial selectivity versus the hA2B AR. Indeed, the highest SIs were 61.6
and 50.0 for the 1,3-dimethyl-N-benzyl derivative 21 and the 1,3-
dipropyl-N-p-F-phenyl derivative 23, respectively. These discour-
aging findings prompted us to synthesize and test the 1-propyl-
3-methyl-9-dAX carbamate 32 designed on the basis of the in-
creased selectivity versus hA2B observed in 8-N-substituted phen-
oxyacetamido-9-dAXs14–16 and -xanthines,11,26 bearing larger
The 1,3-dimethyl-N-30-thienyl-carbamate 15 resulted as the
most potent ligand displaying a sub-nanomolar binding affinity
to the hA2B AR (Ki = 0.8 nM) whereas its N-20-thienyl isomer 16
exhibited the second highest affinity to the same AR subtype
(Ki = 1 nM). Ligands with the highest affinity to the hA2A AR were
the 1,3-dimethyl-N-30-thienyl carbamate 15 (Ki = 10.5 nM) and
the 1,3-dipropyl-N-30-pyrazolyl-carbamate 28 (Ki = 10 nM). The
lowest affinities to both the hA2B and hA2A ARs were displayed
by the 1,3-dipropyl-N-thienyl derivatives 24 and 25 (Ki = 501 and
602, and 3311 and 2240 nM, to hA2B and hA2A ARs, respectively).

F. Fernández et al. / Bioorg. Med. Chem. 17 (2009) 3618–3629
3621
Table 1
Chemical structures and hAR binding affinitiesa of 1,3-dialkyl-8- N-substituted benzyloxycarbonylamino-9-dAX derivatives 13–32
SIb
Ki hA1
Ki hA3
c
c
Compound
Structure
Ki hA2B
Ki hA2A
O
H
N
N
O
13
2.8
34.7
12.6
2.2
275
O
O
O
O
N
O
HN
H
N
N
N
N
O
14
9.5
0.8
10
72.4
10.5
15.1
7.6
17.8
10.0
nd
871
N
HN
F
O
N
H
N
15d
12.6
363
O
O
O
S
HN
O
N
H
N
O
16
15.1
nd
O
S
O
HN
O
H
N
N
N
N
O
17
81.3
490
6.0
nd
nd
O
O
O
N
O
O
O
N
O
HN
H
N
O
18
3.2
18.2
5.6
nd
nd
O
N
O
HN
H
N
O
N
N
19
11.5
132
11.5
nd
nd
N
O
N
H
N
N
O
N
O
20
2.1
43.7
20.4
nd
nd
O
(continued on next page)

3622
F. Fernández et al. / Bioorg. Med. Chem. 17 (2009) 3618–3629
Table 1 (continued)
SIb
Ki hA1
Ki hA3
c
c
Compound
Structure
Ki hA2B
Ki hA2A
O
N
H
N
O
O
21
16.2
1000
61.6
nd
nd
O
N
HN
O
N
H
N
N
O
22
7.1
2.0
501
602
7.8
1.7
28.8
4.1
20.4
0.4
nd
331
O
O
O
O
O
O
O
O
O
O
O
O
HN
O
N
H
N
N
N
N
N
N
O
23
24
25
26
27
118
59
71
HN
F
O
N
H
N
O
3311
2240
35.5
21.9
6.6
3.7
4.6
12.9
nd
nd
nd
nd
S
HN
O
N
H
N
O
nd
S
HN
O
N
H
N
O
nd
N
HN
N
O
N
H
N
O
nd
N
O
HN

F. Fernández et al. / Bioorg. Med. Chem. 17 (2009) 3618–3629
3623
Table 1 (continued)
SIb
Ki hA1
Ki hA3
c
c
Compound
Structure
Ki hA2B
Ki hA2A
O
N
H
N
N
O
O
28
29
30
31
5.9
10.0
447
28.2
1.7
nd
nd
nd
nd
O
N
NH
HN
O
N
O
H
N
O
N
O
N
46%e
—
nd
N
O
N
O
N
H
N
O
N
O
24.0
1.2
nd
N
Cl
O
N
O
N
H
N
O
N
O
42.7
50.1
1.2
nd
nd
N
F
O
N
O
H
N
O
N
32
13.8
15%
—
nd
nd
HN
O
a
Affinity values to the indicated human cloned ARs are expressed as Ki (nM) or as % of inhibition at a 1 lmol concentration. SEMs from two independent experiments were
always lower than 10%.
b
SI is the selectivity index defined as the ratio Ki hA2A/Ki hA2B
nd stands for not determined.
In functional assays, compound 15 exhibited the following antagonist potency: pA2 A2B (rat) = 8.66 0.30, pA2 A2A (rat) = 7.55 0.19.
An estimated Ki equal to 900 nM was used in the plot of Figure 3.
.
c
d
e
alkyl substituents at position 1 than at position 3. Ligand 32 main-
tained a good affinity to the hA2B AR (Ki = 13.8 nM) close to that
exhibited by the two corresponding symmetrically substituted
parent compounds 13 and 22 (Ki = 2.8 and 7.1 nM, respectively)
but, unfortunately, its low solubility allowed only the determina-
edly high, increase of Ki was observed for the N-30-thienyl
(compare 15 with 24) and the N-20-thienyl (compare 16 with 25)
carbamates from 0.8 to 501 and from 1 to 602 nM, respectively.
These data differ considerably from those obtained for the 8-N-
substituted phenoxyacetamido-9-dAX analogues14,15 and might
suggest a significantly diverse binding mode of the 1,3-dialkyl sub-
stituents in the two diverse classes of 8-p-substituted phenyl-9-
dAXs. The observation that 9-dAX carbamates yielded very low SI
values, arising from an unexpectedly high affinity to the hA2A AR
subtype, sharply contrasts with the high Ki hA2A/Ki hA2B SI elicited
by the 8-N-substituted phenoxyacetamido-9-dAX analogues.14,15
Furthermore, it lends additional support to the hypothesis of a dif-
ferent binding mode for the hA2B and more so for the hA2A AR
subtypes.
tion of the % of radioligand displacement at the hA2A AR at 1 lM
concentration. Eventhough the low percentage of inhibition (15%)
might be indicative of a Ki » 1000 nM, a secure assessment of the
hA2B selectivity was not possible.
A comparative analysis of affinity and selectivity data of 1,3-di-
methyl- and 1,3-dipropyl-9-dAXs pointed out some interesting
trend. With few exceptions, 1,3-dimethyl-9-dAX carbamates
exhibited higher hA2B affinities than the corresponding 1,3-dipro-
pyl analogues (see plot in Fig. 4). The most striking, and unexpect-

3624
F. Fernández et al. / Bioorg. Med. Chem. 17 (2009) 3618–3629
O
N
O
β
O
N
α
α
β
O
H
H
A
B
Chart 3. 8-N-Substituted benzyloxycarbonylamino- and 8-N-substituted phenoxy-
acetamido-9-dAX reference molecules A and B, respectively.
came out that while crystal structures related to carbamate A nor-
mally adopted extended conformations, those associated to oxy-
acetamide B assumed generally a folded rather than a linear
conformation and this was likely due to the occurrence of an intra-
molecular HB involving the phenol oxygen and the amidic hydro-
gen atoms. To obtain
a
more complete picture of the
conformational space accessible to both A and B molecules, a con-
formational analysis was run in an apolar implicit solvent (CHCl3)
which supposedly better simulates the inner binding sites and
environment of GPCR transmembranes. The variation of molecular
energy at the change of
preference for extended and folded conformations of molecules A
a and b angles (Chart 3) confirmed the
Figure 3. Affinity–selectivity plot of 9-dAX AR ligands 13–32. Blue squares and red
circles indicate 1,3-dimethyl- and 1,3-dipropyl 9-dAX derivatives, respectively.
and B, respectively (data not shown).
To complete our computational investigation, atomic charges
and molecular electrostatic potentials (MEPs)30 of molecules A
and B were also calculated (data not shown). As expected, the
two reference molecules showed different values and distributions
of partial charges which might strongly influence the HB formation
and the local and global lipophilicity. Indeed, a GRID31 analysis per-
formed using the DRY lipophilic probe and the OH polar probe
(data not shown) fully supported such a hypothesis. Taken to-
gether, our computational data suggested that the two examined
molecules might engage significantly diverse HB, polar and hydro-
phobic interactions, likely arising from their different binding con-
formations, despite the apparently high structural similarity of the
two molecular fragments linking the two phenyl groups.
The same considerations may apply to the carbamic and oxy-
acetamido 9-dAXs analyzed herein. Indeed, the two different link-
ers and, more so, the 1,3-dialkyl groups of dAX moieties, might
bind in completely diverse modes and regions to the ARs binding
sites. Consequently, SAR and SSR relationships of the two classes
of 9-dAXs resulted as being completely different.
Figure 4. hA2B pKi comparison of 1,3-dimethyl- and 1,3-dipropyl-8-N-substituted
benzyloxycarbonylamino-9-dAXs, represented in blue and red bars, respectively.
4.2. Water solubility and partition coefficients
5. Conclusions
Despite their excellent affinity and selectivity profiles, 8-N-
substituted 9-dAXs generally exhibited low water solubility and
high partition coefficients that have strongly limited their develop-
ment as in vivo bioactive agents.14–16 To verify an eventual
improvement of these key physicochemical properties in the pres-
ent series of 9-dAX carbamates, two reference compounds, that is,
15 and 22, have been assessed. Their octanol–water partition coef-
ficients calculated by the Bioloom (v.1.5)27 and the ACD/Labs (v.
11.0)28 were 2.25 and 4.57, and 2.01 and 4.46, respectively. Water
solubility, measured by the same method previously used for dif-
1,3-Dialkyl-8-N-substituted benzyloxycarbonylamino-9-dAXs,
represent a novel class of AR ligands endowed with nanomolar
affinities at the hA2B, hA1 and hA2A AR subtypes. Unfortunately,
no ligand exhibited an acceptably high selectivity versus the hA2B
AR and this strongly limits their therapeutic potential as antiasth-
matic agents.
Partition and solubility data assessed for two 1,3-dialkyl-8-N-
substituted benzyloxycarbonylamino-9-dAXs might suggest
a
ferent 9-dAX derivatives,16 were 21.5 4 and 27
3 lg/ml for
more favourable in vivo activity compared to the corresponding
8-N-substituted phenoxyacetamido analogues.16
compound 15 and 22, respectively.
1,3-Dimethyl-N-30-thienyl carbamate 15, which resulted the
most potent ligand at hA2B, with a good selectivity versus hA3
but poor selectivities versus hA2A and hA1, might be taken as a lead
for further structural modifications aimed at improving its selec-
tivity profile versus hA2B. Similarly, structural variations on 1,3-
dipropyl-N-p-F-phenyl carbamate 23, the most potent ligand at
the hA1 in the whole series, might be carried out to ameliorate
the selectivity versus this AR subtype. These important goals might
be better attained taking into account our computational findings
suggesting likely different binding modes of 1,3-dialkyl-8-N-
4.3. Molecular modeling
To help elucidate the hypothetical different binding modes of 8-
N-substituted benzyloxycarbonylamino- and 8-N-substituted
phenoxyacetamido-9-dAXs the main structural and molecular
properties of these two classes of AR ligands were evaluated by a
computational study on reference molecules A and B (Chart 3).
Using molecules A and B as a query, a substructure search of the
Cambridge Crystallographic Data Centre (CCDC)29 was made. It

F. Fernández et al. / Bioorg. Med. Chem. 17 (2009) 3618–3629
3625
substituted benzyloxycarbonylamino- and 8-N-substituted phen-
oxyacetamido-9-dAXs at the four AR subtypes.
6.1.1.4. 4-[(E)-2-(1,2,3,6-Tetrahydro-3-Methyl-5-nitro-2,6-
dioxo-1-propylpyrimidin-4-yl)-vinyl]-benzoic acid (10d). CC
(CHCl3/MeOH, 9:1), yield 65%, mp: 188–190 °C (dec). 1H NMR
(DMSO-d6): 10.50 (br s, 1H, CO2H), 7.88 (d, J = 8.3, 2H, C6H4), 7.57
(d, J = 8.3, 2H, C6H4), 7.16 (d, J = 16.2, 1H, HC@CH), 6.98 (d,
J = 16.2, 1H, HC@CH), 3.80 (t, J = 7.2, 2H, NCH2), 3.35 (s, 3H,
NCH3) 1.68–1.59 (m, 2H, CH2), 0.87 (t, 3H, J = 7.4, CH3). MS (ESI)
(m/z, %): 358 [(Mꢀ1)ꢀ, 100].
6. Experimental
6.1. Chemistry
All chemicals used were of reagent grade and were obtained
from Aldrich Chemical Co. and used without further purification.
When necessary, solvents were dried by standard techniques and
distilled. All air-sensitive reactions were carried out under argon.
Flash chromatography was performed on silica gel (Merck 60,
230–240 mesh) and analytical TLC on pre-coated silica gel plates
(Merck 60 F254, 0.25 mm). Melting points (uncorrected) were mea-
sured in a glass capillary tube on a Stuart Scientific electro thermal
apparatus SMP3. Infrared spectra were recorded in a Perkin–Elmer
1640 FTIR spectrophotometer. 1H NMR spectra were recorded in a
Bruker AMX 300 spectrometer at 300 MHz, using TMS as internal
standard (chemical shifts in d values, J in hertz). Mass spectra were
recorded on a Hewlett-Packard HP5988A or a Micromass Autospec
spectrometers or a Electrospray interface Ion Trap Mass spectrom-
eter. (1100 series LC/MSD Trap System Agilent, Palo Alto, Ca).
Microanalyses were performed in a FISONS EA 1108 Elemental
Analyser at the University of Santiago Microanalysis Service; all re-
sults shown are within 0.4% of the theoretical values.
6.1.2. General procedure for the reductive ring closure of 5-
nitro-6-styryluracils 10a–d to 11a–d
To a solution of the appropriate 5-nitro-6-styryluracil derivative
10a–d (5 mmol) in formic acid (40 mL) was slowly added sodium
dithionite (11.5 mmol) and the mixture was refluxed overnight.
The resulting suspension was cooled to room temperature and
poured into water. The precipitate was collected by filtration,
washed with water and dried under vacuum to yield the expected
compounds 11a–d.
6.1.2.1. 4-(2,3,4,5-Tetrahydro-1,3-dimethyl-2,4-dioxo-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)benzoic acid (11a). Yield 80%, mp:
>300 °C (H2O). 1H NMR (DMSO-d6): 12.93 (br s, 1H), 12.52 (br s,
1H), 8.04–7.93 (m, 4H, C6H4), 6.78 (s, 1H, 7-H), 3.33 (s, 3H,
NCH3), 3.17 (s, 3H, NCH3). MS (EI) (m/z, %): 299 (M+, 100).
Aryl isocyanates needed for the preparation of target carba-
mates 13–18, 21 and 23–26 were either commercially available
or recently obtained by standard methods, from the corresponding
amines,32 or, when amines were not commercially available, from
the appropriate carboxylic acids.33
6.1.2.2. 4-(2,3,4,5-Tetrahydro-2,4-dioxo-1,3-dipropyl-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)benzoic acid (11b). Yield 90%, mp:
340–345 °C (dec) (MeOH/Et2O). 1H NMR (DMSO-d6): 12.88 (br s,
1H), 12.42 (br s, 1H), 7.92–7.82 (m, 4H, C6H4), 6.75 (s, 1H, 7-H),
3.73 (m, 4H, 2 ꢂ NCH2), 1.56–1.43 (m, 4H, 2 ꢂ CH2), 0.78–0.73
(m, 6H, 2 ꢂ CH3). MS (CI) (m/z, %): 356 [(M+1)+, 100].
6.1.1. General procedure for the condensation between benzal-
dehydes 9 and 5-nitro-6-methyluracils 8. Preparation of 1,3-
dialkyl-6-styryluracil derivatives 10a, b, d
6.1.2.3. 4-(2,3,4,5-Tetrahydro-2,4-dioxo-1-methyl-3-propyl-1H-
pyrrolo[3,2-d]pyrimidin-6-yl)benzoic acid (11d). Yield 18%,
mp: >300 °C (MeOH/Et2O). 1H NMR (DMSO-d6): 12.92 (br s, 1H),
12.55 (br s, 1H), 8.00–7.94 (m, 4H, C6H4), 6.85 (s, 1H, 7-H), 3.83
(t, J = 7,4, 2H, NCH2), 1.61–152 (m, 2H, CH2), 0.87 t, J = 7.4, 3H,
CH3). MS (ESI) (m/z, %): 326 [(Mꢀ1)ꢀ, 100].
A solution of the suitable 1,3-dialkyl-6-methyl-5-nitropyrimi-
dine-2,4(1H,3H)-dione 8a, b, c (8 mmol), the appropriate benzalde-
hyde (8 mmol), piperidine (12 mmol) and molecular sieves of 3 Å
in dry dioxane (50 mL) was refluxed for 5–47 h, under argon atmo-
sphere. The resulting solution was concentrated under vacuum,
and the residue was dissolved in EtOAc and successively washed
with 10% HCl and brine, dried (Na2SO4), and the solvent was evap-
orated under reduce pressure. The solid residue was purified by CC
or recrystallized.
6.1.2.4. [4-(2,3,4,5-Tetrahydro-2,4-dioxo-1,3-dipropyl-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl formate (11c). CC
(hexane/EtOAc, 7:3), yield 17%, mp: 229–231 °C (EtOAc). 1H NMR
(CDCl3): 11.73 (br s, 1H, NH), 8.16 (s, 1H, CO2H), 7.88 (d, J = 8.2,
2H, C6H4), 7.44 (d, J = 8.2, 2H, C6H4), 6.27 (s, 1H, 7-H), 5.24 (s, 2H,
OCH2), 4.06–3.91 (m, 4H, 2 ꢂ NCH2), 1.82 (sext, J = 7.4, 2H, CH2),
1.67 (sext, J = 7.4, 2H, CH2), 1.02 (t, J = 7.4, 3H, CH3), 0.87 (t,
J = 7.4, 3H, CH3). MS (EI) (m/z, %): 369 (M+, 84), 285 (100).
6.1.1.1. 4-[(E)-2-(1,2,3,6-Tetrahydro-1,3-dimethyl-5-nitro-2,6-
dioxopyrimidin-4-yl)vinyl]benzoic acid (10a). Yield 70%, mp:
254–256 °C (EtOH). 1H NMR (DMSO-d6): 13.15 (br s, 1H, CO2H),
7.99 (d, J = 8.2, 2H, C6H4), 7.78 (d, J = 8.2, 2H, C6H4), 7.36 (d,
J = 16.5, 1H, HC@CH), 7.06 (d, J = 16.5, 1H, HC@CH), 3.39 (s, 3H,
NCH3), 3.26 (s, 3H, NCH3). MS (EI) (m/z, %): 331 (M+, 0.5), 267 (100).
6.1.3. General procedure for the reduction of compounds 11a,
b, d
The carboxylic acid 11 (4 mmol) was suspended in anhy-
drous THF (25 mL) and the mixture was cooled to 0 °C. 1 M
BH3ꢁTHF in THF (10 mL) was added, and the resulting suspen-
sion was stirred at rt between 6 and 24 h. MeOH was cau-
tiously added to the cooled mixture and the solvents were
evaporated under reduced pressure. The solid residue was dis-
solved in EtOAc and the solution was successively washed with
10% NaOH and H2O. The organic layer was dried (Na2SO4), fil-
tered and the solvent was evaporated to give the corresponding
hydroxymethyl derivatives of the pyrrolo[3,2-d]pyrimidine-
2,4(3H,5H)-dione, 12a, b, d.
6.1.1.2. 4-[(E)-2-(1,2,3,6-Tetrahydro-5-nitro-2,6-dioxo-1,3-
dipropylpyrimidin-4-yl)vinyl]benzoic acid (10b). CC (hexane/
EtOAc, 55:45), yield 80%, mp: 178–180 °C. 1H NMR (DMSO-d6):
11.14 (br s, 1H, CO2H), 7.99 (d, J = 8.3, 2H, C6H4), 7.78 (d, J = 8.3, 2H,
C6H4), 7.37 (d, J = 16.4, 1H, HC@CH), 7.10 (d, J = 16.4, 1H, HC@CH),
3.87–3.80 (m, 4H, 2 ꢂ NCH2), 1.67–1.55 (m, 4H, 2 ꢂ CH2), 0.91–
0.80 (m, 6H, 2 ꢂ CH3). MS (CI) (m/z, %): 388 [(M+1)+, 100].
6.1.1.3. 6-[(E)-4-(Hydroxymethyl)styryl]-5-nitro-1,3-dipropyl-
pyrimidine-2,4(1H,3H)-dione (10c). CC (hexane/EtOAc, 1:1),
yield 40%, mp: 134–136 °C. 1H NMR (CDCl3): 7.35–7.24 (m, 4H,
C6H4), 6.92 (d, J = 16.3, 1H, HC@CH), 6.55 (d, J = 16.3, 1H, HC@CH),
4.58 (s, 2H, OCH2), 3.86–3.70 (m, 4H, 2 ꢂ NCH2), 3.41 (br s, 1H, OH),
1.66–1.52 (m, 4H, 2 ꢂ CH2), 0.86–0.80 (m, 6H, 2 ꢂ CH3). MS (EI) (m/
z, %): 373 (M+, 11), 309 (100).
6.1.3.1. 6-[4-(Hydroxymethyl)phenyl]-1,3-dimethyl-1H-pyrrol-
o[3,2-d]pyrimidine-2,4(3H,5H)-dione (12a). Yield 40%, mp:
266–268 °C (MeOH/Et2O). 1H NMR (DMSO-d6): 12.34 (br s, 1H,
NH), 7.84 (d, J = 8.1, 2H, C6H4), 7.34 (d, J = 8.1, 2H, C6H4), 6.67 (s,

3626
F. Fernández et al. / Bioorg. Med. Chem. 17 (2009) 3618–3629
1H, 7-H), 5.21 (t, J = 5.6, 1H, OH), 4.48 (d, J = 5.6, 2H, OCH2), 3.39 (s,
3H, NCH3), 3.23 (s, 3H, NCH3). MS (EI) (m/z, %): 285 (M+, 100).
(m/z, %): 422 (M+, 0.2), 63 (100). Anal. Calcd for C22H19FN4O4: C,
62.55; H, 4.53; N, 13.26. Found: C, 62.88; H, 4.71; N, 13.42.
6.1.3.2. 6-[4-(Hydroxymethyl)phenyl]-1,3-dipropyl-1H-pyrrol-
o[3,2-d]pyrimidine-2,4(3H,5H)-dione (12b). Yield 90%, mp:
282–285 °C (dec) (MeOH/Et2O). 1H NMR (DMSO-d6): 12.17 (br s,
1H), 7.72 (d, J = 8.0, 2H, C6H4), 7.20 (d, J = 8.0, 2H, C6H4), 6.57 (s,
1H, 7-H), 5.08 (t, J = 5.6, 1H, OH), 4.36 (d, J = 5.6, 2H, OCH2),
3.73–3.67 (m, 4H, 2 ꢂ NCH2), 1.54–1.40 (m, 4H, 2 ꢂ CH2), 0.78–
0.67 (m, 6H, 2 ꢂ CH3). MS (CI) (m/z, %): 342 [(M+1)+, 100].
6.1.4.3. [4-(2,3,4,5-Tetrahydro-1,3-dimethyl-2,4-dioxo-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl thiophen-3-ylcarba-
mate (15). Method B. CC (CH2Cl2/MeOH, 97:3), yield 20%, mp:
279–281 °C (MeOH). 1H NMR (DMSO-d6): 12.43 (s, 1H, NH),
10.06 (s, 1H, NH), 7.92 (d, J = 8.3, 2H, C6H4), 7.45 (d, J = 8.3, 2H,
C6H4), 7.42–7.40 (m, 1H, C4H3S), 7.21–7.19 (m, 1H, C4H3S), 6.99
(d, J = 5.1, 1H, C4H3S), 6.74 (s, 1H, 7-H), 5.15 (s, 2H, OCH2), 3.41
(s, 3H, NCH3), 3.25 (s, 3H, NCH3). MS (FAB) (m/z, %): 410 (M, 4),
285 (100). Anal. Calcd for C20H18N4O4S: C, 58.53; H, 4.42; N,
13.65; S, 7.81. Found: C, 58.71; H, 4.31; N, 13.97; S, 8.08.
6.1.3.3. 6-[4-(Hydroxymethyl)phenyl]-1-methyl-3-propyl-1H-
pyrrolo[3,2-d]pyrimidine-2,4(3H,5H)-dione (12d). Yield 75%,
mp: 200–202 °C (dec) (MeOH/Et2O). 1H NMR (DMSO-d6): 12.17
(br s, 1H), 7.85 (d, J = 8.1, 2H, C6H4), 7.35 (d, J = 8.1, 2H, C6H4),
6.68 (s, 1H, 7-H), 5.20 (t, J = 5.3, 1H, OH), 4.50 (d, J = 5.3, 2H,
OCH2), 3.84 (t, J = 7.4, NCH2), 3.59 (s, 3H, NCH3), 1.48–1.38 (m,
2H, CH2), 0.86 (t, J = 7.4, 3H, CH3). MS (ESI) (m/z, %): 312 [(Mꢀ1)ꢀ,
100].
6.1.4.4. [4-(2,3,4,5-Tetrahydro-1,3-dimethyl-2,4-dioxo-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl thiophen-2-ylc-
arbamate (16). Method B. CC (CH2Cl2/MeOH, 95:5), yield 31%,
mp: 286–288 °C (MeOH). 1H NMR (DMSO-d6): 12.36 (s, 1H, NH),
10.67 (s, 1H, NH), 7.86 (d, J = 8.1, 2H, C6H4), 7.39 (d, J = 8.1, 2H,
C6H4), 6.86–6.84 (m, 1H, C4H3S), 6.75–6.71 (m, 1H, C4H3S), 6.68
(s, 1H, 7-H), 6.49–6.47 (m, 1H, C4H3S), 5.11 (s, 2H, OCH2), 3.35 (s,
3H, NCH3), 3.19 (s, 3H, NCH3). MS (FAB) (m/z, %): 410 (M, 0.7),
285 (100). Anal. Calcd for C20H18N4O4S: C, 58.53; H, 4.42; N,
13.65; S, 7.81. Found: C, 58.90; H, 4.21; N, 13.99; S, 7.59.
6.1.4. General procedures for the preparation of carbamates 13–
32 from the appropriate hydroxymethyl derivatives 12a, b d (see
Scheme 1)
Method A. The aryl isocyanate (0.42 mol) was added to a freshly
prepared solution of the hydroxymethyl derivative 12a or 12b
(0.35 mol) in dry DMF (5 mL), under Ar atmosphere and at rt,
and the mixture was thereafter heated at 115 °C until the disap-
pearance of the starting material was observed (TCL, CH2Cl2/MeOH,
5–7 h). The reaction mixture was then poured into H2O (20 mL)
and extracted with EtOAc (3 ꢂ 25 mL). The combined extracts were
washed with H2O, and dried (Na2SO4). The solvent was evaporated
under reduced pressure and the residue was recrystallized from
MeOH.
Method B. The hydroxymethyl derivative 12a, b, d (0.31 mol) in
dry dioxane (6 mL) was added to a freshly prepared solution of the
isocyanate (0.77 mol) in dry dioxane or toluene (6 mL) under Ar
atmosphere and the mixture was stirred while heating to reflux.
Progress of the reactions was monitored by TLC, and when no more
starting material was detected, the solvent was evaporated under
reduced pressure and the residue was purified by CC (CH2Cl2/
MeOH) or recrystallized from mixtures of CH2Cl2/MeOH.
Method C. A solution of the hydroxymethyl derivative 12a, b, d
(0.29 mol) in dry pyridine (1 mL) was added to the solution of CDI
(0.29 mol) in dry pyridine (0.5 mL), at 0 °C and under Ar atmo-
sphere. The reaction mixture was stirred 2 h at this temperature,
and it was thereafter allowed to reach rt and stirred for a further
2 h. The corresponding amine (0.29 mol) was then slowly added
while stirring and the precipitate formed was collected by filtration
and washed with water.
6.1.4.5. [4-(2,3,4,5-Tetrahydro-1,3-dimethyl-2,4-dioxo-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl pyridin-2-ylcar-
bamate (17). Method B. Yield 60% mp: 301–303 °C (MeOH). 1H
NMR (DMSO-d6): 12.28 (s, 1H, –NH), 10.13 (s, 1H, NH), 8.10 (d,
J = 5.2, 1H, C5H4N), 7.77 (d, J = 8.2, 2H, C6H4), 7.69–7.56 (m, 2H,
C5H4N), 7.32 (d, J = 8.2, 2H, C6H4), 6.91–6.86 (m, 1H, C5H4N), 6.59
(s, 1H, 7-H), 5.03 (s, 2H, OCH2), 3.26 (s, 3H, NCH3), 3.10 (s, 3H,
NCH3). MS (FAB) (m/z, %): 406 [(M+1)+, 3]; 154 (100). Anal. Calcd for
C21H19N5O4: C, 62.22; H, 4.72; N, 17.26. Found: C, 62.50; H, 4.75; N, 17.59.
6.1.4.6. [4-(2,3,4,5-Tetrahydro-1,3-dimethyl-2,4-dioxo-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl furan-2-ylcarbamate
(18). Method B. CC (CH2Cl2/MeOH, 98:2), yield 23%, mp: 278–
280 °C (MeOH). 1H NMR (DMSO-d6): 12.41 (s, 1H, NH), 10.29 (s,
1H, NH), 7.94 (d, J = 8.1, 2H, C6H4), 7.46 (d, J = 8.1, 2H, C6H4),
7.33–7.28 (m, 1H, C4H3O), 6.75 (s, 1H, 7-H), 6.43–6.40 (m, 1H,
C4H3O), 6.04–5.98 (m, 1H, C4H3O), 5.15 (s, 2H, CH2), 3.24 (s, 3H,
NCH3), 3.08 (s, 3H, NCH3). MS (FAB) (m/z, %): 394 [M, 0.7]; 285
(100). Anal. Calcd for C18H14N4O5: C, 59.02; H, 3.85; N, 15.29.
Found: C, 58.85; H, 4.25; N, 15.59.
6.1.4.7. [4-(2,3,4,5-Tetrahydro-1,3-dimethyl-2,4-dioxo-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl 4-phenylpiper-
azine-1-carboxylate (19). Method C. Yield 33%, mp: 254–256 °C
(EtOAc). 1H NMR (DMSO-d6): 12.41 (s, 1H, NH), 7.90 (d, J = 8.2,
2H, C6H4), 7.42 (d, J = 8.2, 2H, C6H4), 7.37–6.99 (m, 2H, C6H5),
6.93 (d, J = 8.2, 2H, C6H5), 6.81–6.78 (m, 1H, C6H5), 6.72 (s, 1H, 7-
H), 5.12 (s, 2H, OCH2), 3.13 (s, 3H, NCH3), 3.08 (s, 3H, NCH3). MS
(CI) (m/z, %): 85 (100). Anal. Calcd for C25H27N5O4: C, 65.95; H,
5.75; N, 14.79. Found: C, 65.88; H, 5.99; N, 15.05.
6.1.4.1. [4-(2,3,4,5-Tetrahydro-1,3-dimethyl-2,4-dioxo-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl phenylcarbamate
(13). Method A. Yield 40%, mp: 249–251 °C. 1H NMR (DMSO-d6):
12.50 (s, 1H, NH), 9.85 (s, 1H, NH), 8.00 (d, J = 8.2, 2H, arom.),
7.56–7.51 (m, 4H, arom.), 7.35–7.32 (m, 2H, arom.), 7.09–7.02
(m, 1H, arom.), 6.81 (s, 1H, 7-H), 5.23 (s, 2H, OCH2), 3.49 (s, 3H,
NCH3), 3.32 (s, 3H, NCH3). MS (EI) (m/z, %): 404 (M+, 8), 268
(100). Anal. Calcd for C22H20N4O4: C, 65.34; H, 4.96; N, 13.85.
Found: C, 65.01; H, 5.31; N, 14.11.
6.1.4.8. [4-(2,3,4,5-Tetrahydro-1,3-dimethyl-2,4-dioxo-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl-3,4-dihydroisoquino-
line-2(1H)-carboxylate (20). Method C. Yield 42%, mp = 225–
228 °C (EtOAc). 1H NMR (DMSO-d6): 12.14 (s, 1H, NH), 7.90 (d,
J = 7.3, 2H, C6H4), 7.42 (d, J = 7.3, 2H, C6H4), 7.16 (virtual s, 4H,
C9H10N), 6.50 (s, 1H, 7-H), 4.86 (s, 2H, OCH2), 4.60–4.55 (m, 2H,
CH2), 3.66–3.61 (m, 2H, CH2) 3.17–3.15 (m, 2H, CH2), 3.42 (s, 3H,
NCH3), 3.26 (s, 3H, NCH3). MS (FAB) (m/z, %): 445 [(M+1), 9], 444
(M, 1), 270 (100). Anal. Calcd for C25H24N4O4: C, 67.55; H, 5.44;
N, 12.60. Found: C, 67.88; H, 5.37; N, 12.85.
6.1.4.2. [4-(2,3,4,5-Tetrahydro-1,3-dimethyl-2,4-dioxo-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl-4-fluorophenylcarba-
mate (14). Method A. Yield 43%, mp 280–282 °C. 1H NMR (DMSO-
d6): 12.52 (s, 1H, NH), 9.91 (s, 1H, NH), 8.00 (d, 2H, J = 8.2, arom.),
7.56–7.51 (m, 4H, arom.), 7.35–7.32 (m, 2H, arom.), 6.83 (s, 1H, 7-
H), 5.24 (s, 2H, OCH2), 3.50 (s, 3H, NCH3), 3.33 (s, 3H, NCH3). MS (EI)

F. Fernández et al. / Bioorg. Med. Chem. 17 (2009) 3618–3629
3627
6.1.4.9. [4-(2,3,4,5-Tetrahydro-1,3-dimethyl-2,4-dioxo-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl benzylcarbamate
(21). Method A. Yield 30%, mp: 233–235 °C. 1H NMR (DMSO-d6):
12.43 (s, 1H, NH), 7.91 (d, J = 8.2, 2H, C6H4), 7.41 (d, J = 8.2, 2H,
C6H4), 7.35–7.25 (m, 5H, C6H5), 6.75 (s, 1H, 7-H), 5.07 (s, 2H,
OCH2), 4.21 (d, J = 5.9, 2H, NCH2), 3.42 (s, 3H, NCH3), 3.26 (s, 3H,
NCH3). MS (EI) (m/z, %): 418 (M+, 47), 268 (99), 63 (100). Anal.
Calcd for C23H22N4O4: C, 66.02; H, 5.30; N, 13.39. Found: C,
65.88; H, 5.67; N, 13.65.
6.1.4.15. [4-(2,3,4,5-Tetrahydro-2,4-dioxo-1,3-dipropyl-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl isoxazol-2-ylcar-
bamate (27). Method C. Yield 41%, mp 169–171 °C (EtOAc). 1H
NMR (DMSO-d6): 12.40 (s, 1H, NH), 8.30 (s, 1H, NH), 7.95 (d,
J = 8.0, 2H, C6H4), 7.60 (d, J = 8.0, 2H, C6H4), 6.76 (s, 1H, 7-H),
5.77–5.70 (m, 1H, C3H2NO), 5.55–5.42 (m, 1H, C3H2NO), 5.41 (s,
2H, OCH2), 3.86–3.75 (m, 4H, 2 ꢂ NCH2), 1.82–1.63 (m, 4H,
2 ꢂ CH2), 0.99–0.88 (m, 6H, 2 ꢂ CH3). MS (FAB) (m/z, %): 324
(100). Anal. Calcd for C23H25N5O5: 61.19; H, 5.58; N, 15.51. Found:
C, 61.59; H, 5.88; N, 15.85.
6.1.4.10. [4-(2,3,4,5-Tetrahydro-2,4-dioxo-1,3-dipropyl-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl phenylcarbamate
(22). Method C. Yield 16%, mp 226–228 °C (MeOH). 1H NMR
(DMSO-d6): 12.35 (s, 1H, NH), 9.74 (s, 1H, NH), 7.91 (dd, J = 8.4,
2.9, 2H, arom.), 7.47–7.41 (m, 3H, arom.), 7.28–7.26 (m, 2H, arom.),
6.98 (d, J = 7.4, 2H, arom.), 6.75 (s, 1H, 7-H), 5.07 (s, 2H, OCH2), 3.85
(t, J = 7.2, 2 ꢂ NCH2), 1.67 (sext, J = 7.1, 2H, CH2), 1.56 (sext, J = 7.3,
2H, CH2), 0.91 (t, J = 7.2, 3H, CH3), 0.86 (t, J = 7.3, 3H, CH3). MS (FAB)
(m/z, %): 460 [M, 26]; 324 (100). Anal. Calcd for C26H28N4O4: C,
67.81; H, 6.13; N, 12.17. Found: C, 68.18; H, 6.06; N, 12.25.
6.1.4.16. [4-(2,3,4,5-Tetrahydro-2,4-dioxo-1,3-dipropyl-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl 1H-pyrazol-3-ylcarba-
mate (28). Method C. Yield 60%, mp 211–213 °C (EtOAc). 1H NMR
(DMSO-d6): 12.39 (s, 1H, NH), 8.03–7.93 (m, 4H,
C6H4 + NH + C3H2N2), 7.49 (d, J = 7.9, 2H, C6H4), 6.78 (s, 1H, 7-H),
5.85 (s, 1H, NH), 5.51 (s, 1H, C3H2N2), 5.33 (s, 2H, OCH2), 3.89–
3.83 (m, 4H, 2 ꢂ NCH2), 1.70–1.52 (m, 4H, 2 ꢂ CH2), 0.91 (t,
J = 7.3, 3H, CH3), 0.86 (t, J = 7.3, 3H, CH3). MS (FAB) (m/z, %): 450
(M, 3). Anal. Calcd for C23H26N6O4: C, 61.32; H, 5.82; N, 18.66.
Found: C, 61.38; H, 5.72; N, 18.27.
6.1.4.11. [4-(2,3,4,5-Tetrahydro-2,4-dioxo-1,3-dipropyl-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl 4-fluorophenylcarba-
mate (23). Method A. Yield 43%, mp 259–261 °C (EtOAc). 1H NMR
(DMSO-d6): 12.35 (s, 1H, NH), 9.79 (s, 1H, NH), 8.26 (virtual s, 2H,
C6H4F), 7.90 (d, J = 8.2, 2H, C6H4), 7.84 (d, J = 7.8, 2H, C6H4), 7.08 (t,
J = 7.8, 2H, C6H4F), 6.72 (s, 1H, 7-H), 5.14 (s, 2H, OCH2), 3.85 (t,
J = 7.4 Hz, 2H, NCH2), 3.35 (t, J = 7.4 Hz, 2H, NCH2), 1.72–1.65 (m,
2H, CH2), 1.59–1.52 (m, 2H, CH2), 0.90 (t, J = 7.3, 3H, CH3), 0.83 (t,
J = 7.3, 3H, CH3). MS (FAB) (m/z, %): 478 (M, 23); 341 (100). Anal.
Calcd for C26H27FN4O4: C, 65.26; H, 5.69; N, 11.71. Found: C,
65.58; H, 5.97; N, 11.65.
6.1.4.17. [4-(2,3,4,5-Tetrahydro-2,4-dioxo-1,3-dipropyl-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl 4-phenylpiperazine-1-carbox-
ylate (29). Method C. Yield 33%, mp 240–242 °C (EtOAc). 1H NMR
(DMSO-d6): 12.36 (s, 1H, NH), 7.90 (d, J = 8.2, 2H, C6H4), 7.42 (d,
J = 8.2, 2H, C6H4), 7.24–7.19 (m, 2H, C6H5), 6.94 (d, J = 8.2, C6H5),
6.80 (t, J = 7.2, C6H5), 6.75 (s, 1H, 7-H), 5.12 (s, 2H, OCH2), 3.89–
3.84 (m, 4H, 2 ꢂ NCH2), 3.60–3.51 (m, 4H, 2 ꢂ NCH2), 3.13–3.10
(m, 4H, 2 ꢂ NCH2), 1.68 (sext, J = 7.3, 2H, CH2), 1.57 (sext, J = 7.3,
2H, CH2), 0.91 (t, J = 7.2, 3H, CH3), 0.86 (t, J = 7.2, 3H, CH3). MS
(FAB) (m/z, %): 530 [(M+1)+, 5]. Anal. Calcd for C30H35N5O4: C,
68.03; H, 6.66; N, 13.22. Found: C, 68.25; H, 6.98; N, 13.11.
6.1.4.12. [4-(2,3,4,5-Tetrahydro-2,4-dioxo-1,3-dipropyl-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl thiophen-3-ylcar-
bamate (24). Method B. Yield 75%, mp >320 °C (toluene). 1H NMR
(DMSO-d6): 12.34 (s, 1H, NH), 10.08 (s, 1H, NH), 7.89 (d, J = 8.0, 2H,
C6H4), 7.42–7.36 (m, 3H, C6H4 + C4H3S), 7.16 (s, 1H, C4H3S), 6.98 (d,
J = 4.6, 1H, C4H3S), 6.73 (s, 1H, 7-H), 5.11 (s, 2H, OCH2), 3.81 (t,
J = 6.9, 4H, 2 ꢂ NCH2), 1.70–1.47 (m, 4H, 2 ꢂ CH2), 0.86 (t, J = 7.2,
3H, CH3), 0.81 (t, J = 7.3, 3H, CH3). MS (FAB) (m/z, %): 467
[(M+1)+, 3], 177 (100). Anal. Calcd for C24H26N4O4S: C, 61.79; H,
5.62; N, 12.01, S, 6.87. Found: C, 61.92; H, 5.46; N, 12.45, S, 7.12.
6.1.4.18. [4-(2,3,4,5-Tetrahydro-2,4-dioxo-1,3-dipropyl-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl 4-(3-chlorophenyl)pipera-
zine-1-carboxylate (30). Method C. Yield 15%, mp 188–190 °C (EtOAc).
1H NMR (DMSO-d6): 12.36 (s, 1H, NH), 7.91 (d, J = 8.1, 2H, C6H4), 7.42 (d,
J = 8.1, 2H, C6H4), 7.32–7.25 (m, 1H, C6H4Cl), 6.95–6.88 (m, 2H, C6H4Cl),
6.82–6.73 (m, 2H, C6H4Cl + 7-H), 5.11 (s, 2H, OCH2), 3.88–3.81 (m, 4H,
2 ꢂ NCH2), 3.52–3.45 (m, 4H, 2 ꢂ NCH2), 3.41–3.30 (m, 4H, 2 ꢂ NCH2),
1.69–1.54 (m, 4H, 2 ꢂ CH2), 0.93–0.83 (m, 6H, 2 ꢂ CH3). MS-FAB (m/z,
%): 564 [(M+1)+, 6]. Anal. Calcd for C30H34ClN5O4: C, 63.88; H, 6.08; N,
12.42. Found: C, 63.65; H, 5.95; N, 12.64.
6.1.4.13. [4-(2,3,4,5-Tetrahydro-2,4-dioxo-1,3-dipropyl-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl thiophen-2-ylcar-
bamate (25). Method B. Yield 90%, mp 245–247 °C (toluene). 1H
NMR (DMSO-d6): 12.26 (s, 1H, NH), 10.68 (s, 1H, NH), 7.82 (d,
J = 8.1, 2H, C6H4), 7.34 (d, J = 8.1, 2H, C6H4), 8.89–6.80 (m, 1H,
C4H3S), 6.79–6.77 (m, 2H, C4H3S), 6.46 (s, 1H, 7-H), 5.05 (s, 2H,
OCH2), 3.73 (t, J = 6.9, 4H, 2 ꢂ NCH2), 1.61–1.40 (m, 4H, 2 ꢂ CH2),
0.79 (t, J = 7.2, 3H, CH3), 0.74 (t, J = 7.2, 3H, CH3). MS (FAB) (m/z,
%): 467 [(M+1)+, 3], 154 (100). Anal. Calcd for C24H26N4O4S: C,
61.79; H, 5.62; N, 12.01; S, 6.87. Found: C, 61.99; H, 5.36; N,
11.85; S, 6.77.
6.1.4.19. [4-(2,3,4,5-Tetrahydro-2,4-dioxo-1,3-dipropyl-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl 4-(4-fluorophenyl)pipera-
zine-1-carboxylate (31). Method C. Yield 24%, mp 225–227 °C
(EtOAc). 1H NMR (DMSO-d6): 12.31 (s, 1H, NH), 7.84 (d, J = 8.2,
2H, C6H4), 7.35 (d, J = 8.2, 2H, C6H4), 7.01–6.89 (m, 4H, C6H4F),
6.68 (s, 1H, 7-H), 5.04 (s, 2H, OCH2), 3.81–3.76 (m, 4H, 2 ꢂ NCH2),
3.60–3.43 (m, 4H, 2 ꢂ NCH2), 2.98–2.92 (m, 4H, 2 ꢂ NCH2), 1.61–
1.47 (m, 4H, 2 ꢂ CH2), 0.86–0.76 (m, 6H, 2 ꢂ CH3). MS (FAB) (m/z,
%): 342 (100). Anal. Calcd for C30H34FN5O4: C, 65.80; H, 6.26; N,
12.79. Found: C, 66.13; H, 6.49; N, 12.98.
6.1.4.14. [4-(2,3,4,5-Tetrahydro-2,4-dioxo-1,3-dipropyl-1H-pyr-
rolo[3,2-d]pyrimidin-6-yl)phenyl]methyl pyrazin-2-ylcar-
bamate (26). Method B. CC (CH2Cl2/MeOH, 95:5), yield 19%, mp
268–270 °C (Et2O). 1H NMR (DMSO-d6): 12.38 (s, 1H, NH), 10.65
(s, 1H, NH), 9.09 (s, 1H, C6H3N2), 8.34–8.29 (m, 2H, C6H3N2), 7.95
(d, J = 8.1, 2H, C6H4), 7.48 (d, J = 8.1, 2H, C6H4), 6.77 (s, 1H, 7-H),
5.22 (s, 2H, OCH2), 3.87–3.82 (m, 4H, 2 ꢂ NCH2), 1.71–1.50 (m,
4H, 2 ꢂ CH2), 0.92–0.82 (m, 6H, 2 ꢂ CH3). MS (FAB) (m/z, %): 463
[(M+1)+,4], 154 (100). Anal. Calcd for C24H26N6O4: C, 62.33; H,
5.67; N, 18.17. Found: C, 62.59; H, 5.78; N, 18.45.
6.1.4.20. [4-(2,3,4,5-Tetrahydro-2,4-dioxo-1-methyl-3-propyl-
1H-pyrrolo[3,2-d]pyrimidin-6-yl)phenyl]methyl phenylcarba-
mate (32). Method B. Yield 25%, mp 273–275 °C (MeOH). 1H
NMR (DMSO-d6): 12.21 (s, 1H, NH), 9.76 (s, 1H, NH), 7.90 (d,
J = 7.15, 2H, arom.), 7.48–7.40 (m, 4H, arom.), 7.28–7.24 (m, 2H,
arom.), 6.98–6.96 (m, 1H, arom.), 6.73 (s, 1H, 7-H), 5.15 (s, 2H,
OCH2), 3.85 (t, J = 7.2, 2H, NCH2), 3,4 (s, 3H, NCH3), 1.58–150 (m,
2H, CH2), 0.91 (t, J = 7.2, 3H, CH3), 0.86 (t, J = 7.3, 3H, CH3). MS
(ESI) (m/z, %): 431 [(Mꢀ1)ꢀ, 100]. Anal. Calcd for C24H24N4O4: C,
66.65; H, 5.59; N, 12.96. Found: C, 66.43; H, 5.71; N, 12.80.

3628
F. Fernández et al. / Bioorg. Med. Chem. 17 (2009) 3618–3629
6.2. Water solubility assay
6.4.1.4. Human A3 receptors. Adenosine A3 receptor competition
binding experiments were carried out in membranes from HeLa-A3
The compound was dissolved in DMSO at a concentration of
cells. Labeled with 30 nM [3H]NECA. Non-specific binding was
1 mg/ml. This solution was added in portion, (2
ll at a time) to
determined in the presence of 100 lM (R)-PIA. The reaction mix-
1 ml of a 50 mM Tris/HCl pH 7.4 buffer solution at room tempera-
ture. Typically a total of 14 additions were made so that the final
volume of DMSO was well below 5%. The appearance of the precip-
itate was detected by an absorbance increase, due to light scatter-
ture was incubated at 25 °C for 180 min.
After each incubation time samples were filtered and measured
in a microplate beta scintillation counter (Microbeta Trilux, Per-
kin–Elmer, Madrid, Spain).
ing by particulate material, in
a dedicate diode array UV
spectrometer (GE Healthcare). Increased UV absorbance was mea-
sured in the 600–820 nm range. In its simplest implementation,
the precipitation point (i.e., the upper aqueous solubility limit),
was calculate from a bilinear curve fit in a plot of the absorbance
6.4.2. Isolated organ assays
6.4.2.1. A2A receptors. These assays were performed in A2A
receptors39 from isolated aorta of 200–250 g male Wistar-Kyoto
rats as previously described.16
(y axis) versus ll of DMSO (x axis). The results reported in Section
4.2 are the means SEM of two independent assays.
6.4.2.2. A2B receptors. These assays were performed in A2B
receptors40 isolated colon of 200–250 g male Wistar-Kyoto rats
as previously described.16
6.3. Molecular modeling
Acknowledgements
Reference molecules A and B (Chart 3) were built starting from
the SENFOH and FAXYN crystal structures [CCDC].29 Geometrical
optimization and conformational analyses were performed with
MACROMODEL,34 using the force field OPLS 2001.35,36 The molecu-
lar structures were minimized using the conjugated gradient algo-
rithm setting the number of iterations equal to 1000 and the value
of convergence equal to 0.025 Å.
The Italian authors thank the MIUR (Ministero dell’Istruzione,
dell’Università e della Ricerca Scientifica, Rome-Italy) for financial
support (PRIN Project 2006). The Spanish authors thank the Health
Institute Carlos III (RD07/0067/0002) and Xunta de Galicia
(07CSA003203PR) for partial financial support. J.B. is a recipient
of a financial support from the Programa Isabel Barreto (Xunta de
Galicia).
The conformational analyses were performed setting a step
equal to 15° for both
a and b dihedral angles. 576 conformations
were obtained and single point energy calculations were per-
formed. Contour maps were generated by using an in-house script
written in MatLab.37 Charge calculations were made by means of a
quantum mechanical method with the PM3 Hamiltonian available
in MOPAC 6.0.38 The molecular electrostatic potential (MEP) was cal-
culated with VEGA-ZZ setting the point density and the probe ra-
dius equal to 2500 and 1.4 Å, respectively.30 By using standard
options, molecular interaction fields (MIFs) were calculated within
GRID31 using DRY and OH and probes.
References and notes
1. Fredholm, B. B.; IJzerman, A. P.; Jacobson, K. A.; Klotz, K. N.; Linden, J.
Pharmacol. Rev. 2001, 53, 527.
2. Jacobson, K. A.; Gao, Z.-G. Nat. Rev. Drug Disc. 2006, 5, 247.
3. Moro, S.; Gao, Z.-G.; Jacobson, K. A.; Spalluto, G. Med. Res. Rev. 2006, 26, 131.
4. Akkari, R.; Burbiel, J. C.; Hockemeyer, J.; Müller, C. E. Curr. Top. Med. Chem. 2006,
6, 1375.
5. Jacobson, K. A.; IJzerman, A. P.; Linden, J. Drug Dev. Res. 1999, 47, 45.
6. Holgate, S. T. Br. J. Pharmacol. 2005, 145, 1009.
7. Tabrizi, M. A.; Baraldi, P. G.; Preti, D.; Romagnoli, R.; Saponaro, G.; Baraldi, S.;
Moorman, A. R.; Zaid, A. N.; Varani, K.; Borea, P. A. Bioorg. Med. Chem. 2008, 16,
2419.
8. Baraldi, P. G.; Tabrizi, M. A.; Preti, D.; Bovero, A.; Romagnoli, R.; Fruttarolo, F.;
Zaid, N. A.; Moorman, A. R.; Varani, K.; Gessi, S.; Merighi, S.; Borea, P. A. J. Med.
Chem. 2004, 47, 1434.
9. Volpini, R.; Costanzi, S.; Vittori, S.; Cristalli, G.; Klotz, K. N. Curr. Top. Med. Chem.
2003, 3, 427.
10. Kim, Y. C.; Ji, X.; Melman, N.; Linden, J.; Jacobson, K. A. Drug Dev. Res. 1999, 47,
178.
11. Kim, Y. C.; Ji, X.; Melman, N.; Linden, J.; Jacobson, K. A. J. Med. Chem. 2000, 43,
1165.
6.4. Biochemistry and pharmacology
6.4.1. Radioligand binding assays
Radioligand binding competition assays were performed
in vitro using A1, A2A, A2B and A3 human receptors expressed in
transfected CHO (hA1), HeLa (hA2A and hA3) and HEK-293 (hA2B
)
cells as previously described.16 Briefly:
12. Vidal, B. J.; Esteve Trias, C.; Segarra Matamoros, V.; Ravina Rubira, E.;
Fernandez Gonzales, F.; Loza Garcia, M. I.; Sanz Carreras, F.; WO Patent 03/
000694, 2003.
13. Carotti, A.; Stefanachi, A.; Ravina, E.; Sotelo, E.; Loza, M. I.; Cadavid, M. I.;
Centeno, N. B.; Nicolotti, O. Eur. J. Med. Chem. 2004, 39, 879.
14. Carotti, A.; Cadavid, M. I.; Centeno, N. B.; Esteve, C.; Loza, M. I.; Martinez, A.;
Nieto, R.; Ravina, E.; Sanz, F.; Segarra, V.; Sotelo, E.; Stefanachi, A.; Vidal, B.
J. Med. Chem. 2006, 49, 282.
6.4.1.1. Human A1 receptors. Adenosine A1 receptor competition
binding experiments were carried out in membranes from CHO-A1
cells (Euroscreen, Gosselies, Belgium) labeled with 2 nM
[3H]DPCPX. Non-specific binding was determined in the presence
of 10 lM (R)-PIA. The reaction mixture was incubated at 25 °C
for 60 min.
15. Stefanachi, A.; Brea, J. M.; Cadavid, M. I.; Centeno, N. B.; Esteve, C.; Loza, M. I.;
Martinez, A.; Nieto, R.; Raviña, E.; Sanz, F.; Segarra, V.; Sotelo, E.; Vidal, B.;
Carotti, A. Bioorg. Med. Chem. 2008, 16, 2852.
16. Stefanachi, A.; Nicolotti, O.; Leonetti, F.; Cellamare, S.; Campagna, F.; Loza, M. I.;
Brea, J. M.; Mazza, F.; Gavuzzo, E.; Carotti, A. Bioorg. Med. Chem. 2008, 16, 9780.
17. Grahner, B.; Winiwarter, S.; Lanzner, W.; Müller, C. E. J. Med. Chem. 1994, 37,
1526.
18. Hayallah, A. M.; Sandoval-Ramirez, J.; Reith, U.; Schobert, U.; Preiss, B.;
Schumacher, B.; Daly, J. W.; Müller, C. E. J. Med. Chem. 2002, 45, 1500–1510.
19. Elzein, E.; Kalla, R.; Li, X.; Perry, T.; Gimbel, A.; Zeng, D.; Lustig, K.; Leung, K.;
Zablocki, J. J. Med. Chem. 2008, 51, 2267.
6.4.1.2. Human A2A receptors. Adenosine A2A receptor competi-
tion binding experiments were carried out in membranes from
HeLa-A2A cells labeled with 3 nM [3H]ZM241385. Non-specific
binding was determined in the presence of 50
lM NECA. The reac-
tion mixture was incubated at 25 °C for 30 min.
20. Esteve, C.; Nueda, A.; Dıaz, J. L.; Beleta, J.; Cardenas, A.; Lozoya, E.; Cadavid, M.
I.; Loza, M. I.; Ryder, H.; Vidal, B. Bioorg. Med. Chem. Lett. 2006, 51, 3642.
21. Stefanachi, A.; Leonetti, F.; Cappa, A.; Carotti, A. Tetrahedron Lett. 2003, 44,
2121.
22. Papesch, V.; Dobson, R. M. G. D. J. Org. Chem. 1963, 28, 1329.
23. Senda, S.; Suzui, A.; Honda, M.; Fujimura, H. Chem. Pharm. Bull. 1958, 6, 482.
24. Totleben, M. J.; Freeman, J. P.; Szuszkovicz, J. J. Org. Chem. 1997, 62, 7319.
25. D’Addona, D.; Bochet, C. G. Tetrahedron Lett. 2001, 42, 5227.
6.4.1.3. Human A2B receptors. Adenosine A2B receptor competi-
tion binding experiments were carried out in membranes from
HEK-293-A2B cells (Euroscreen, Gosselies, Belgium) labeled with
35 nM [3H]DPCPX. Non-specific binding was determined in the
presence of 400
lM NECA. The reaction mixture was incubated at
25 °C for 30 min.

F. Fernández et al. / Bioorg. Med. Chem. 17 (2009) 3618–3629
3629
26. Yan, L.; Bertarelli, D. C.; Hayallah, A. M.; Meyer, H.; Klotz, K. N.; Müller, C. E.
J. Med. Chem. 2006, 49, 4384.
27. BioByte Corp., Claremont, CA, USA.
28. Advanced Chemistry Development, Toronto, Canada.
29. The Cambridge Crystallographic Data Centre via http://www.ccdc.cam.ac.uk/
data_request/cif.
30. Pedretti, A.; Villa, L.; Vistoli, G. J. Mol. Graph. Model. 2002, 21, 47.
31. Goodford, P. J. J. Med. Chem. 1985, 28, 849.
34. Mohamadi, F.; Richards, N. G. J.; Guida, W. C.; Liskamp, R.; Lipton, M.; Caufield,
C.; Chang, G.; Hendrikson, T.; Still, W. C. J. Comput. Chem. 1990, 11, 440.
35. Jorgensen, W. L.; Tirado-Rives, J. J. Am. Chem. Soc. 1988, 110, 1657.
36. Jorgensen, W. L.; Maxwell, D. S.; Tirado-Rives, . J. Am. Chem. Soc. 1996, 118,
11225.
37. MATLAB The Language of Technical Computing, version 7.3; The Mathworks:
Natick, MA, USA, 2006.
38. Stewart, J. J. P. Quant. Chem. Prog. Exch. 1990, 10, 86.
39. Prentice, D. J.; Hourani, S. M. O. Br. J. Pharmacol. 1996, 118, 1509.
40. Fozard, J. R.; Baur, F.; Wolber, C. Eur. J. Pharmacol 2003, 475, 79.
32. Majer, P.; Randad, R. S. J. Org. Chem. 1994, 59, 1937.
33. Prats, M.; Gálvez, C. Heterocycles 1991, 34, 149.