Structure-activity relationship studies of a new series of imidazo[2,1-f]purinones as potent and selective A3 adenosine receptor antagonists

Article abstract of DOI:10.1016/j.bmc.2008.10.049

We recently described the synthesis of 1-benzyl-3-propyl-1H,8H-imidazo[2,1-f]purine-2,4-diones, new potent and selective A₃ adenosine receptor antagonists containing a xanthine core. The present work can be considered an extension of our SAR st

Full text of DOI:10.1016/j.bmc.2008.10.049

Bioorganic & Medicinal Chemistry 16 (2008) 10281–10294
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry
journal homepage: www.elsevier.com/locate/bmc
Structure–activity relationship studies of a new series of imidazo[2,1-f]
purinones as potent and selective A3 adenosine receptor antagonists
a
a
a
a
Pier Giovanni Baraldi a, , Delia Preti , Mojgan Aghazadeh Tabrizi , Romeo Romagnoli , Giulia Saponaro ,
Stefania Baraldi a, Maurizio Botta c, Cesare Bernardini c, Andrea Tafi c, Tiziano Tuccinardi d,
Adriano Martinelli d, Katia Varani b, Pier Andrea Borea b
*
a Dipartimento di Scienze Farmaceutiche, Università di Ferrara, Via Fossato di Mortara 17-19, 44100 Ferrara, Italy
b Dipartimento di Medicina Clinica e Sperimentale-Sezione di Farmacologia, Università di Ferrara, 44100 Ferrara, Italy
c Dipartimento Farmaco Chimico Tecnologico, Università degli Studi di Siena, via A. Moro, 53100 Siena, Italy
d Dipartimento di Scienze Farmaceutiche, Università di Pisa, via Bonanno 6, 56126 Pisa, Italy
a r t i c l e i n f o
a b s t r a c t
Article history:
We recently described the synthesis of 1-benzyl-3-propyl-1H,8H-imidazo[2,1-f]purine-2,4-diones, new
potent and selective A3 adenosine receptor antagonists containing a xanthine core. The present work
can be considered an extension of our SAR studies on related structures in which the effect of different
kind of substitutions at the 1-, 3- and 8-positions has been evaluated in order to improve both the
potency and the hydrophilicity of the originally synthesised molecules. The A3 binding disposition of
these compounds was also investigated through docking and 3D-QSAR studies.
Received 26 May 2008
Revised 14 October 2008
Accepted 20 October 2008
Available online 1 November 2008
Keywords:
Adenosine
Ó 2008 Elsevier Ltd. All rights reserved.
Cancer
A3 antagonists
Imidazo[2,1-f]purinones
Docking
3D-QSAR
1. Introduction
activity of cytotoxic lymphocytes, thus leading to the hypothesis
that A3 antagonists might be useful for the revelation of tumour-
Adenosine regulates a number of physiological functions through
the interaction with specific cell membrane G-protein coupled
receptors ubiquitously expressed in the body and classified as A1,
A2A, A2B and A3 (ARs).1 The four known adenosine receptors sub-
types, have been largely investigated thanks to their involvement
in important pathologies.2 The A3 AR is able to inhibit forskolin-in-
duced cAMP accumulation, to increase phosphatidylinositol-spe-
cific phospholipase C and D activity, and to elevate IP3 levels and
intracellular Ca2+ pools.3 A3 AR subtype has been subject of intensive
investigations as potential therapeutic target due to its contribution
to important pathophysiological processes such as inflammation,4
neurodegeneration,5,6 ischaemia,7–9 asthma,10,11 glaucoma12–14
and cancer.15–18 It has been suggested that adenosine inhibits tu-
mour cell growth while maintaining bone marrow cell proliferation
through the involvement of the A3 receptors.19 Studies related to a
murine model demonstrated that the activation of A3 receptors
can interfere with the tumour cell recognition and with the cytolytic
associated immunosuppression and facilitate adoptive immuno-
therapy.15 Evidence of high levels of expression of A3 adenosine
receptor subtype has been provided in several tumour cell lines20–
23 and A3 specific antagonists seem to synergistically enhance cyto-
toxic treatment and counter P-glycoprotein efflux in multi-drug
resistance.23 A recent study provides the first evidence that A3 AR
plays a role in colon tumorigenesis and, more importantly, can
potentially be used as a diagnostic marker or a therapeutic target
for colon cancer.24
Much effort has been directed toward searching for potent and
selective human A3 adenosine antagonists.25–27 Some recent
developments in this field comprise the identification of 1,2,4-triaz-
olo[1,5-i]purines,28 1,2,4-triazolo[4,3-a]quinoxalin-1-ones,29,30 thi-
azoles
and
thiadiazoles,31,32
imidazo[2,1-i]purin-5-ones,33
pyrido[2,1-f]purine-2,4-dione,34 pyrazolo[3,4-c]/[4,3-c]quinolin-4-
ones35,36 and adenosine derivatives.37–39 Generally, AR antagonists
can be structurally classified into xanthine and non-xanthine deriv-
atives, the first ones revealing, in most cases, low affinity toward A3
receptor subtype. A useful strategy employed for the identification
of adenosinereceptors antagonistsis the approachbased ontheann-
elation of xanthine derivatives.40 Compounds characterized by the
* Corresponding author. Tel.: +39 0532 291293; fax: +39 0532 291296.
E-mail addresses: baraldi@dns.unife.it, brlpgv@unife.it, baraldi@unife.it (P.G.
Baraldi).
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.10.049

10282
P. G. Baraldi et al. / Bioorg. Med. Chem. 16 (2008) 10281–10294
presence of fused rings on the xanthine nucleus have, in fact, shown
different levels of affinity and selectivity to the various AR subtypes.
During studies related to the identification of rigid analogues of
KF17837, a known A2A adenosine receptor antagonist belonging to
the class of styryl xanthines, we synthesised and evaluated a series
of 1,3-dipropyl-7-aryl/heteroaryl-1H,6H-pyrrolo[2,1-f]purine-2,4-
diones41 which did not show considerable affinity for the investi-
gated targets. Subsequently, Priego et al. underlined the fundamen-
tal role of a benzyl and a propyl moieties at the 1- and 3-positions,
respectively, of a series of 1H,3H-pyrido[2,1-f]purine-2,4-diones34
examined as A3 AR antagonists. In light of this we assumed that
the lack of affinity of our reported 1,3-dipropyl-pyrrolo[2,1-f]pur-
ine-2,4-diones might be partially due to the presence of a propyl
chain, instead of the benzyl moiety at the 1-position. We conse-
quently performed the synthesis of a series of 7-aryl/alkyl-1H,6H-
pyrrolo[2,1-f]purine-2,4-diones (see compounds 1m–p42 Table 1)
and 7-aryl/alkyl-1H,8H-imidazo[2,1-f]purine-2,4-diones (see com-
pounds 1a–l42 Table 1), evaluating in first instance the effect of the
introductionofa benzylandapropylat the1-and3-position, respec-
tively, in our previous series and in a new series of fused xanthine
compounds.42 Our attempt to introduce different kind of aryl and/
or (cyclo)alkyl moieties at the 6- and 7-positions led us to identify
very potent and selective A3 adenosine receptors antagonists among
which 1-benzyl-7-methyl-3-propyl-1H,8H-imidazo[2,1-f]purine-
2,4-dione (1a,42 Fig. 1) presents the best binding profile showing a
Ki (hA3) value from binding assay of 0.8 nM with considerable selec-
tivity versus the other adenosine receptors subtypes (selectivity ra-
tios of Ki (hA1/hA3) = 3163, Ki (hA2A/hA3) = 6250, IC50 (hA2B)/Ki
(hA3) = 2570). These compounds represent, to the best of our knowl-
edge, the most potent and selective hA3 adenosine receptor antago-
nistsbaseduponaxanthinenucleus. Inthepresentworkwedescribe
anextensionofourreportedseriesofimidazo-purinonesperforming
new substitutions alternatively at the 1-, 3- and 8-positions of the
reference compound1a (Fig. 1). Thanksto the identification of versa-
tile synthetic routes we succeeded in manipulating the imidazo[2,1-
f]purin-2,4-dione nucleus at various positions so to obtain the newly
reported derivatives 6a–c, 12, 13, 14a–f, 15a–c, 20a–b, 24a–b (Table
1) and to develop a SAR analysis combined with a molecular model-
ing study. Furthermore, on the basis of estimated CLogP values, the
hydrophilicity of some of the new synthesized molecules has been
improved in comparison with the reference compound 1a.
2. Results and discussions
2.1. Chemistry
1-Benzyl-3-propyl-imidazo[2,1-f]purine-2,4-dione derivatives
(1a–l, Table 1) and 1-benzyl-3-propyl-1H,6H-pyrrolo[2,1-f]pur-
ine-2,4-dione derivatives (1m–p) were prepared following the
general synthetic strategy previously described.42
To realize final compounds 6a–c bearing pyridyl-methyl moie-
ties at the 1-position, the 3-benzyl-8-bromo-7-(2-oxo-propyl)-1-
propyl-3,7-dihydro-purine-2,6-dione 342 has been debenzylated33
at the 3-position with anhydrous AlCl3 in toluene (Scheme 1). Sub-
sequent alkylation of 4 with 2/3/4-bromomethyl-pyridines and
K2CO3 afforded the desired 3-substituted-intermediates 5a–c
which were cyclised in a steel bomb by treatment with liquid
ammonia to the corresponding 7-methyl-3-propyl-1-pyridin-2/3/
4-ylmethyl-1H,8H-imidazo[2,1-f]purine-2,4-diones 6a–c. These
were finally converted into the related hydrochloride salts via
treatment with a saturated ethanolic solution of HCl.
1-Benzyl-3-(substituted)-7-methyl-1H,6H-pyrrolo[2,1-f]purine-
2,4-diones (12, 13 and 14a–f), were prepared as depicted in
Scheme 2. Bromination at the 8-position of 3-benzyl-3,7-dihy-
dro-purine-2,6-dione33 7 by treatment with Br2 and sodium ace-
tate in acetic acid at 110 °C for 3 h43 led to the formation of the
8-bromo-intermediate 8 with good yield. The following alkylation,
occurring in presence of equimolar amounts of K2CO3 and a-chlo-
roacetone in DMF at room temperature, permitted us to isolate
with 70% yield, the N7-2-oxo-propyl derivative 10. The structural
assignment was based on the knowledge that the distinctive reso-
nance for the proton on the unsubstituted nitrogen at the 7-posi-
tion was at about 13.5 ppm, while a broad singlet at about 11–
12 ppm is the typical H–N3/H–N1 signal.44 Carrying out the alkyl-
ation with an excess of 2 equiv of K2CO3 and 3 equiv of a-chloro-
acetone, it is possible to obtain the N1,N7-bis-(2-oxo-propyl)
derivative 9. The 3-benzyl-8-bromo-7-(2-oxo-propyl)-3,7-dihy-
dro-purine-2,6-dione 10 was then treated with K2CO3 and the
appropriate alkyl halide at room temperature, to furnish interme-
diates 11a–f. The final tricycles 12, 13 and 14a–f were obtained
by cyclisation of the corresponding intermediates 9, 10 and 11a–
f which was performed according to the same conditions above
described.
O
N
O
3
R
O
N
N
N
N
N
N
N
N
8
H
R''
O
N
R'
1
We identified intermediates 3 and 10 as suitable precursors for
the obtainment of the 8-substituted-imidazo[2,1-f]purinones 15a–
c (Scheme 3) after treatment with triphenylphosphine and diethyl
azodicarboxylate in the typical Mitsunobu reaction conditions.
The synthetic pathway illustrated in Scheme 4 permitted us to
obtain final compounds 20a–b. Refluxing 16a–b in formic acid, fol-
1a
2
Figure 1. Substitution of the 1-, 3- and 8-positions of the imidazo[2,1-f]purine-2,4-
diones.
O
O
O
O
O
N
O
O
N
N
N
N
N
N
N
N
NH
1) iii
2) iv
N
N
N
i
ii
Br
Br
Br
N
O
N
R
O
O
N
R
O
N
H
3
4
5a-c
6a-c (x HCl)
5a, 6a, 6a x HCl R = pyridin-4-ylmethyl
5b, 6b, 6bx HCl R = pyridin-3-ylmethyl
5c, 6c, 6c x HCl R = pyridin-2-ylmethyl
Scheme 1. Reagents and conditions: (i) AlCl3, toluene, 1 h, 60 °C; (ii) 2/3/4-bromomethylpyridine, K2CO3, DMF, rt; (iii) NH3, EtOH, ON, 100–120 °C; (iv) EtOH saturated with
HCl, 30’, rt.

P. G. Baraldi et al. / Bioorg. Med. Chem. 16 (2008) 10281–10294
10283
O
O
O
O
H
N
H
N
R
O
NH2
NH2
R
O
R
O
O
N
O
N
N
N
N
i
ii
Br
N
N
N
N
N
N
N
N
NH
N
R
N
R
N
R
Br
O
O
O
O
16a-b
17a-b
18a-b
12
9
ii
O
O
O
N
O
O
N
H
N
H
N
R
R
O
N
N
N
N
N
N
HN
i
iii
iv
NH
HN
Br
Br
N
O
N
R
N
R
N
O
O
7
8
19a-b
20a-b
iii
16-20a, R = propyl
16-20b, R = isobutyl
O
O
O
N
O
N
R
N
N
Scheme 4. Reagents and conditions: (i) a—HCO2H, reflux, 1 h; b—NaOH, EtOH/H2O,
reflux, 1 h; (ii) Br2, CH3CO2H, CH3CO2Na, 45 °C, 1 h; (iii)
DMF, rt, 4 h; (iv) liquid ammonia, EtOH, 120 °C, ON.
N
N
N
HN
iv
Br
a-Br-acetophenone, K2CO3,
Br
O
O
11a-f
10
lowed by treatment with NaOH in a mixture of EtOH and H2O, in-
duced the formation of 1,3-dialkyl-xanthines 17a–b which were
brominated at the 8-position (18a–b), alkylated at N7 (19a–b) with
v
v
O
O
N
a-bromoacetophenone and at last cyclised into the 1,3-dipropyl/
R
N
N
N
diisobutyl-imidazo[2,1-f]purinones 20a–b according to the syn-
thetic steps describe for preparation of compound 13. The 5,6-dia-
N
HN
NH
NH
N
O
N
O
mino-1,3-dipropyl/diisobutyl-1H-pyrimidine-2,4-dione
deriva-
tives 16a and 16b were obtained by the classical method
reported in the literature, starting from the condensation of cyano-
acetic acid and N,N0-dipropyl or N,N0-diisobutyl urea, followed by
nitrosation at the 5-position and reduction of the nitroso group.45
Scheme 5 displays the preparation of compounds 24a–b. 8-Ni-
tro-theophylline46 (21) was reduced by refluxing with hydrazine
and 10% Pd/C in methanol to furnish intermediate 22. This was
14a-f
13
11a, 14a R = CH2CH2OH
11b, 14b R = CH2CH2CH2OH
11c, 14c R = CH2CH=CH2
11d, 14d R = CH2C≡CH
alkylated with the corresponding
a-bromoacetophenones at the
N7-position to obtain derivatives 23a–b. The imidazole-ring clo-
sure was finally performed by refluxing 23a–b in acetic acid to
yield final tricycles 24a–b.
11e, 14e R = CH2CO2Et
11f, 14f R = CH2CH2N(CH3)2
Scheme 2. Reagents and conditions: (i) Br2, CH3COOH, 3 h, 110 °C; (ii) 3 equiv
a-
chloroacetone, 2 equiv K2CO3, DMF, rt; (iii) 1 equiv -chloroacetone, 1 equiv K2CO3,
a
DMF, rt; (iv) RX, K2CO3, DMF, rt; (v) NH3, EtOH, ON, 100–120 °C.
O
N
O
N
H
N
H
N
i
N
N
ii
NO2
R
NH2
N
N
O
O
iii
O
21
22
O
N
O
N
O
R
R
O
O
N
N
N
O
R
N
N
N
N
N
i
Br
N
N
N
N
R'
N
N
NH
O
NH2
O
O
N
15a-c
3, 10
23a-b
24a-b
15a R = H; R’ = 1-hydroxy-propyl-2-yl
15b R = H; R’ = CH2CH3
15c R = CH2CH2CH3; R’ = CH2CH3
23-24a, R = Ph
23-24b, R = CH3
Scheme 5. Reagents and conditions: (i) NH2NH2, 10% C/Pd, CH3OH; (ii)
a-halo-
Scheme 3. Reagents and condition: (i) R’-NH2, PPh3, DEAD, THF, 1 h, rt.
ketones, K2CO3, DMF, rt, 4–6 h; (iii) CH3COOH, reflux, 4 h.

10284
P. G. Baraldi et al. / Bioorg. Med. Chem. 16 (2008) 10281–10294
2.2. Biological evaluation and structure–affinity relationships
(compare 1a, 1i and 1g to derivatives 1n, 1m and 1p, respec-
tively) with the only exception of compound 1o which shows
4-fold higher hA3 receptor affinity than the corresponding imi-
dazo-purinone derivatives 1b. The Ki (hA3) values showed to be
strictly dependent on the nature of the substituents at the 7-po-
sition. Our preliminary study indicated that the presence of a
large aromatic and lipophilic moiety at the 7-position probably
established repulsive interactions with the receptor, so we
decided to evaluate the effect of replacing the aromatic nucleus
with various (cyclo)alkyl chains. The best results were obtained
with the introduction of a methyl group (1a, Ki (hA3) = 0.8 nM
with a surprising selectivity pattern vs the other AR subtypes).
Attempt to introduce a substituent at the 6-position led to a loss
of A3 affinity. In light of these observations, in the newly synthes-
ised series of imidazo[2,1-f]purine-2,4-dione derivatives, herein
described, we generally decided to maintain the unsubstitution
at the 6-position and the 7-methyl substitution (except for deriv-
atives 20a–b and 24a functionalized at the 7-position with a phe-
nyl ring) while investigating the effect of different kind of
All the synthesized compounds (Table 1) were evaluated in
radioligand binding assays to determine their affinities for human
A1, A2A, and A3 adenosine receptors. Potency of the compounds ver-
sus hA2B adenosine receptors were studied evaluating their capa-
bility to inhibit (100 nM) NECA stimulated cAMP production.
Basal and NECA stimulation of cAMP levels were 15 2 and
80 9 pmol cAMP/106 cells, respectively. NECA was able to stimu-
late cAMP levels in hA2BCHO cells with an EC50 value of
145 15 nM. Affinity data for A1, A2A and A3 receptors, expressed
as Ki values, and IC50 values derived from the cAMP assay carried
out for hA2B subtypes, are listed in Table 1.
With the previously reported series of compounds42 (1a–p) we
evaluated the effect of heterocycles fused on the N7–C8-positions
of the xanthine nucleus. Among the examined tricycles, the imi-
dazo[2,1-f]purine-2,4-dione derivatives were 2- to 10-fold more
active than the corresponding substituted-pyrrolo[2,1-f]purine-
2,4-dione derivatives toward the adenosine A3 receptor subtype
Table 1
Binding, functional data and selectivity ratios at hA1, hA2A, hA2B, and hA3 adenosine receptors for the reported compounds
R
R
O
N
O
N
R'
NH
R'
O
N
O
O
R'
NH
R
O
R
O
N
N
N
N
N
N
R
O
N
N
N
N
N
N
N
NH
N
R'
N
N
O
O
N
R'
N
R
1a-l [42]
1m-p [42]
6a-c,
15a-c
20a-b, 24a-b
12, 13, 14a-f
a
b
c
d
Compound
R
R’
hA1
hA2A
hA2B
hA3
1
1b
1c
1d
1e
1f
1g
1h
1i
1j
1k
1l
1m
1n
1o
H
H
H
H
H
H
CH3
CH3
H
H
H
H
H
H
H
CH3
CH3
CH2CH3
>1000
>1000
460 (424–498)
>1000
350 (299–411)
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
373 (330–422)
476 (432–525)
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
400 (323–496)
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
>1000
0.8 (0.6–0.9)
15 (9–27)
CH(CH3)2
C(CH3)3
Cyclopropyl
Cyclohexyl
CH3
CH2CH3
Ph
4-OCH3-Ph
4-Ph-Ph
4-F-Ph
Ph
CH3
CH2CH3
CH3
pyridin-4-ylmethyl
pyridin-3-ylmethyl
pyridin-2-ylmethyl
CH2Ph
CH2Ph
CH2Ph
CH2Ph
CH2Ph
CH2Ph
CH2Ph
31 (25–38)
99 (77–129)
23 (18–29)
555 (467–660)
36 (31–43)
60 (53–69)
115 (89–150)
55 (28–104)
>1000
22 (19–26)
200 (134–297)
8.0 (7.1–9.1)
3.45 (2.71–4.36)
80 (63–100)
141 (105–188)
57 (45–73)
34 (27–42)
182 (175–189)
148 (106–208)
38 (27–53)
110 (93–130)
5.13 (3.93–6.70)
15 (12–19)
20 (17–24)
98 (81–119)
>1000
1p
6a ꢀ HCl
6b ꢀ HCl
6c ꢀ HCl
12
CH2CH2CH3
CH2CH2CH3
CH2CH2CH3
CH2COCH3
H
13
14a
14b
14c
14d
14e
14f
15a
15b
15c
20a
20b
24a
24b
CH2CH2OH
CH2CH2CH2OH
CH2CH = CH2
CH2CꢁCH
CH2CO2Et
CH2CH2N(CH3)2
H
CH2Ph
1-hydroxy-propyl-2-yl
CH2CH3
CH2CH3
Ph
Ph
Ph
CH3
H
>1000
CH2CH2CH3
CH2CH2CH3
CH2CH(CH3)2
CH3
200 (163–241)
99 (77–129)
144 (136–151)
>1000
CH3
>1000
>1000
a
Displacement of specific [3H]-DPCPX binding to human A1 receptors expressed in CHO cells (Ki nM, n = 4).
b
c
Displacement of specific [3H]-ZM 241385 binding to human A2A receptors expressed in CHO cells (Ki nM, n = 4).
cAMP assay in CHO cells expressing hA2B receptors (IC50 nM, n = 4).
d
Displacement of specific [3H]-MRE3008F20 binding to human A3 receptors expressed in CHO cells (Ki nM, n = 4). Binding and functional data of compounds 1a–p from
reference 42.

P. G. Baraldi et al. / Bioorg. Med. Chem. 16 (2008) 10281–10294
10285
substituents at the 1-, 3- and 8-positions (see compounds 6a–c,
12, 13, 14a–f, 15a–c, 24b).
the calculated LogP values of compounds 6c (CLogP = 1.51 1.38)
and 14a (CLogP = 1.31 1.39). All CLogP values have been deter-
mined with the version 7.0 of the ACD/Labs software.
The substitution of the benzyl moiety at the 1-position with the
pyridin-2/3/4-yl-methyl functions resulted in a substantial loss of
affinity toward the A3 adenosine receptor subtype. The comparison
between the binding data of the reference compound 1a (Ki
(hA3) = 0.8 nM) and the ones of the three hydrochloride salts of
the 1-pyridyl isomer derivatives (6a–c ꢀ HCl) indicated that the
presence of an heteroatom in the aromatic component induces a
repulsive effect with the receptor binding site and that this effect
becomes more evident when the nitrogen is exposed: Ki values
progressively increase from the ortho- to the meta- to the para-iso-
mer (6c ꢀ HCl Ki (hA3) = 34 nM, 6b ꢀ HCl Ki (hA3) = 57 nM,
6a ꢀ HCl Ki (hA3) = 141 nM). These data induced us to suppose that
the benzyl moiety could establish lipophilic interactions with a
hydrophobic receptor pocket.
Removing the propyl chain at the 3-position of derivative 1a
determined a loss of 180-fold of affinity (13, Ki (hA3) = 148 nM),
while substitution at this position showed to be tolerated depend-
ing on the nature of the introduced function. Among these deriva-
tives we identified compounds endowed with good A3 affinity,
such as the 2-hydroxy-ethyl derivative 14a which has a Ki value
of 38 nM and complete selectivity versus all the other adenosine
receptor subtypes. The higher homologue 14b resulted less potent
(Ki (hA3) = 110 nM). Satisfying results have been achieved con-
straining with unsaturations the conformations of the propyl chain
at the 3-position of compound 1a. The allyl (14c, Ki (hA3) = 5 nM)
and the propargyl (14d, Ki (hA3) = 15 nM) derivatives resulted any-
way 6- and 18-fold, respectively, less potent than 1a in binding the
hA3 subtype.
2.3. Molecular modeling
Compound 1a, which showed the highest affinity among the
tested compounds (Ki = 0.8 nM), was docked into our previously
reported A3 receptor model (MODEL1) by means of AUTODOCK
3.0.47 MODEL 1 was built through a ligand-aided homology proce-
dure which made use of CATALYST program.48
The complex was then submitted to the optimizing protocol
fully described in Section 4.
Compound 1a was inserted between TM3, TM6 and TM7 with
the n-propyl substituent directed towards the extracellular side
of the receptor, the benzyl group towards the TM7 and the methyl
substituent towards the intracellular side between TM3 and TM5.
More in details the imidazo[2,1-f]purine-2,4-dione central scaf-
fold of compound 1a was stabilized by p–p interactions with H95
and W243, and the 2-carbonyl group formed a H bond with N250.
The 3-propyl group was inserted in a lipophilic pocket delimited by
L91, I92, F182, and W185, the 1-benzyl group interacted with F239,
L246, H272 and M276 while the 7-methyl group was stabilized by
a lipophilic interaction with I98 and M99 (see Fig. 2).
This binding mode appeared to be in agreement with the muta-
genesis studies, since the mutation of H95, W243, H272 and in par-
ticular N250 determined a reduction of the A3 antagonist affinity
and in our model all these residues seemed to interact with the
ligand.49,50
With the synthesis of compounds 12 and 14e it was possible to
evaluate the effect of the substitution of the 3-propyl chain of 1a
with a moiety containing a carbonyl function. Although both the
compounds resulted quite less potent than 1a in binding the hA3
receptor, the ethyl ester derivative 14e (Ki (hA3) = 20 nM) stands
out as one of the most potent hA3 ligand of the newly synthesised
molecules.
2.4. Automated docking analysis
All the other synthesized derivatives (compounds 1b–p, 6a–c,
12, 13, 14a–f, 15a–c, 20a–b and 24a–b) were docked by means
of AUTODOCK 3.047 in the optimized A3 receptor model obtained
as previously described.
The docking analysis revealed that all the compounds with a Ki
value lower than 100 nM showed a disposition very similar to the
one found for 1a which was characterized by: (a) an H bond be-
tween the 2-carbonyl group and N250, (b) the lipophilic interac-
tion of the 1’ substituent with F239, L246, H272 and M276, (c)
the insertion of the 3-group in the lipophilic pocket delimited by
L91, I92, F182, and W185.
To complete our SAR study of the reported tricyclic structures,
we synthesized 1,3-dipropyl- (20a), 1,3-diisobutyl- (20b) and
1,3-dimethyl- (24a–b) derivatives in order to evaluate Priego’s
hypothesis regarding the primary role of 1-benzyl-3-propyl-sub-
stitution on the affinity of the A3 receptor for our compounds.
Replacement of the 1-benzyl moiety of 1i with a propyl chain
(20a) led to a substantial maintenance of A3 receptor affinity,
inducing at the same time a significant loss of selectivity versus
adenosine hA1 subtype. Compound 20b, showing the symmetric
substitution of N1 and N3-positions with isobutyl chains, displays
a binding profile comparable to that of the dipropyl derivatives
20a as it shows a reasonable hA3 affinity (Ki hA3 = 144 nM) but
low selectivity versus the hA1 subtype if compared with the
remaining molecules of the series. According to Priego’s results,
1,3-dimethyl-derivatives 24a–b completely lose the capability to
bind the four investigated receptor subtypes in comparison with
the 1-benzyl-3-propyl-analogues 1i and 1a, respectively, confirm-
ing the fundamental contribution of the combination of the benzyl
function at the 1-position with the propyl chain at the 3-position to
the ligand-receptor binding. Since in an analogous manner, the
introduction of a (substituted)-alkyl function at the 8-position
determined a dramatically decrease of affinity of derivatives
15a–c, the free 8N–H seems to be fundamental for the ligand-
receptor interaction.
Compounds 6c (Ki (hA3) = 34 nM) and 14a (Ki (hA3) = 38 nM),
although less potent than 1a, showed a good binding profile in
association with enhanced hydrophilic properties on the basis of
estimated CLogP values. The calculated LogP value of the starting
compound 1a (CLogP = 3.0 1.38) resulted in effect higher then
Figure 2. Compound 1a docked in the A3 binding site.

10286
P. G. Baraldi et al. / Bioorg. Med. Chem. 16 (2008) 10281–10294
Compound 1l (Ki = 22 nM) possessed the 4-F-phenyl substi-
however the docking studies were not able to explain this
behaviour.
The substitution of the 4-F-phenyl with the biphenyl group re-
sulted in complete loss of A3 affinity (compound 1k,
Ki >1000 nM). As shown in Figure 3B, this substituent was too
bulky and determined the overturning of the ligand, with the loos-
ing of the H bond with N250.
tuent in 7-position. It maintained a disposition similar to that
of compound 1a, with only a slight shift of the ligand due to
the presence of the bulky 4-F-phenyl group however allowing
the formation of the H bond with N250 (see Fig. 3A). The
substitution of the Fluorine atom with the methoxy group
a
or with
a hydrogen determined a decrease in A3 affinity,
Figure 3. Compounds 1l, 1k, 1f, 6a, 14a and 15c docked into the A3 receptor model.

P. G. Baraldi et al. / Bioorg. Med. Chem. 16 (2008) 10281–10294
10287
Compound 1f (Ki = 555 nM) possessed a bulky 7-cyclohexyl
group that was not coplanar with the imidazo[2,1-f]purine-2,4-
dione central scaffold and as shown in Figure 3C determining a
large shift of the ligand with the losing of the H bond with N250.
The substitution of the 1-benzyl of 1a with pyridinylmethyl
groups (compound 6a–c, Ki = 34–141 nM) resulted in a general de-
crease of the A3 affinity and, as shown in Figure 3D that reported
the analysis of compound 6a, the docking studies showed for these
groups the absence of any favourable electrostatic interaction with
the receptor (see Fig. 3D).
A similar result was obtained also in the case of the 3-substitu-
tion: compounds 12, 14b and 14f, which possessed a 3-group able
to form H bonds, showed respect to 1a a marked decrease of affin-
ity and the docking analysis confirmed that these groups were un-
able to form any H bond with the receptor. The only exception was
represented by compound 14a, which possessed the 3-hydroxy-
ethyl substituent able to interact with H95 (see Fig. 3E).
The alkylation of the N8 nitrogen determined a great decrease
of affinity. The docking of compounds 15c, suggested that the rea-
son of this fact was not the lack of an H bond because this nitrogen
seems to be unable to form any H bond with the receptor. It seems
rather that the steric hindrance of the N substituents unable to oc-
cupy the very limited N8 region determined a completely different
rearrangement of the central scaffold, with the N8 substituent
pointing towards the extracellular side of the receptor, and the
1-benzyl substituent inserted in the lipophilic pocket delimited
by L91, I92, F182, and W185 (see Fig. 3F).
Figure 4. Plot of model 2 (four principal components): experimental/predicted pKi
is reported.
well predicting the affinity of all compounds with only the excep-
tion of 1a for which an affinity approximately 20-fold lower than
the experimental one was calculated (see Fig. 4).
2.5. 3D-QSAR analysis
One important feature of 3D-QSAR analysis is the graphic repre-
sentation of the model, usually aimed at making its interpretation
easier. In the GOLPE program,52 there are several options for dis-
playing the final model. Among these, the PLS pseudo coefficient
is very useful because it makes possible to visualise favourable
and unfavourable interactions between the probes and the studied
molecules.
Figure 5 shows compound 1o embedded in the contour maps of
the C3 probe and the sterically favoured (yellow) and unfavoured
(cyan) regions were indicated; in particular there were three prin-
cipal regions (A–C) with positive values (see Fig. 5a) in which a
favourable interaction between a substituent and the probe deter-
mined an increase in activity, whereas an unfavourable interaction
between a substituent and the probe determined a decrease in
activity. On the contrary, the negative PLS coefficients, indicated
in Figure 5b, show three main areas (A’–C’) where a favourable
interaction between a substituent and the probe determined a de-
crease in activity, whereas an unfavourable interaction between a
substituent and the probe determined an increase in activity.
The analysis of these maps suggested that the favourable re-
gions A and C, corresponding to the two lipophilic pockets, delim-
With the aim of obtaining quantitative results that could sup-
port our binding hypothesis a 3D-QSAR analysis was carried out
using the docking results as a receptor-based alignment.
The study was developed using three probes: the C3 (methyl)
probe to account for steric contacts, the N2 = (sp2 amide cation)
and O (carbonylic oxygen) probes to evaluate the ability of each
compound to be a hydrogen bond acceptor and donor respec-
tively.51 To measure the reliability of the 3D-QSAR model its corre-
lation coefficient (r2), predictive correlation coefficient (q2) and
cross-validated standard deviation of errors of prediction (SDEPCV
were taken into consideration.
)
Another 3D-QSAR model was developed using a ligand-based
alignment, that is, by aligning the central scaffold of all the 34
compounds.
As shown in Table 2 the model obtained using the ligand-based
alignment did not display good results, whereas the model ob-
tained by means of the receptor-based alignment showed good re-
sults, since the r2, q2 and SDEPCV were respectively 0.91, 0.67, and
0.42.
The plot reporting the experimental versus the calculated affin-
ity of model 2 highlighted that the 3D-QSAR model was capable of
Table 2
Statistical results of the 3D-QSAR models. In bold is reported the optimal
dimensionality.
PC
Ligand-based
3D-QSAR model
(Var = 1888)
Receptor Based
3D-QSAR model
(Var = 1674)
Model 1
Model 2
r2
q2
SDEPcv
r2
q2
SDEPcv
1
2
3
4
5
0.54
0.61
0.73
0.78
0.79
0.34
0.38
0.40
0.40
0.37
0.60
0.58
0.57
0.57
0.58
0.52
0.77
0.86
0.91
0.92
0.39
0.57
0.66
0.67
0.64
0.57
0.48
0.43
0.42
0.44
Figure 5. Positive (a) and negative (b) regions of the PLS coefficient plot obtained
with the C3 probe. Compound 14c is also displayed.

10288
P. G. Baraldi et al. / Bioorg. Med. Chem. 16 (2008) 10281–10294
ited by F239, L246, H272 and M276 and L91, I92, F182, and W185,
respectively, were important for the affinity, however the unfa-
vourable regions A’ and C’ seemed to indicate that too bulky sub-
stituents were not allowed.
of these compounds with the main residues suggested to be impor-
tant by mutagenesis studies.
Furthermore the docking poses were used as an alignment tool
for the generation of a 3D-QSAR model which was able to quanti-
tative explain the different affinity of the ligands.
The region B corresponded to the p–p interaction between the
central scaffold and H95 and the region B’ corresponded to W185,
and, as shown by the docking of compounds 15a–c (see Fig. 3F), a
substituent in this region determined a decrease of affinity.
The contour maps of the N2@ and O (see Fig. 6) were very sim-
ilar, the positive region (regions A of Fig. 6b and d) for which a
favourable interaction between a substituent and these probes
determined a decrease in activity, were positioned in the same por-
tion of space as region A of the C3 probe thus confirming that this
region should possess a lipophilic characteristic in order to im-
prove the ligand affinity. As regards the negative regions (Fig. 6a
and c), there was only one region (A’) corresponding to the electro-
static interaction of the hydroxymethyl substituent of compound
14a with H95.
4. Experimental
4.1. Chemistry
Reaction progress and product mixtures were monitored by
thin-layer chromatography (TLC) on silica gel (procoated F254
Merck plates) and visualized with aqueous potassium permanga-
nate or a methanolic solution of ninhydrin. 1H NMR were deter-
mined in CDCl3 or DMSO-d6 solutions with a Varian VXR 200
spectrometer or a Varian Mercury Plus 400 spectrometer; peak
positions are given in parts per million (d) downfield from tetra-
methylsilane as internal standard and J values are given in Hertz.
Light petroleum refers to the fractions boiling at 40–60 °C. Melting
points were determined on a Buchi–Tottoli instrument and are
uncorrected. Chromatography was performed on Merck 230–400
mesh silica gel. Organic solutions were dried over anhydrous so-
dium sulphate. Elemental analyses were performed by the micro-
analytical laboratory of Dipartimento di Chimica, University of
Ferrara, and were within 0.4% of the theoretical values for C, H
and N.
3. Conclusion
Herein we evaluated the effect of different kind of substitutions
at the 1-, 3-and 8-positions of the reference compound 1-benzyl-7-
methyl-3-propyl-1H,8H-imidazo[2,1-f]purine-2,4-dione 1a, Figure
142 Some of the new reported compounds confirmed high A3 AR
binding affinity in association with relevant selectivity versus the
remaining ARs (14c, Ki (hA3) = 5.13 nM, (hA1–hA2A–hA2B/hA3)
>195; 14d, Ki (hA3) = 15 nM, (hA1-hA2A-hA2B/hA3) >67). Com-
pounds 6c ꢀ HCl (Ki (hA3) = 34 nM) and 14a (Ki (hA3) = 38 nM),
although less potent than 1a, showed a good binding profile in
association with enhanced hydrophilic properties on the basis of
estimated CLogP values.
4.1.1. 8-Bromo-7-(2-oxo-propyl)-1-propyl-3,7-dihydro-purine-
2,6-dione (4)
A solution of 342 (0.48 mmol) in anhydrous toluene (5 mL) was
stirred under argon atmosphere at room temperature and anhy-
drous AlCl3 (2.4 mmol) was added. The reaction was heated for
1 h at 60 °C. The solvent was evaporated and the residue parti-
tioned between H2O (20 mL) and EtOAc (3ꢀ 30 mL). The combined
organic layers were dried on Na2SO4, filtered and the solvent evap-
orated to furnish a residue which was crystallized with a mixture
of Et2O/light petroleum 1:1 (20 mL). The solid was filtered, washed
with light petroleum and dried under vacuum. Yield 76%, white so-
lid; mp 244–245 °C; 1H NMR (DMSO-d6, 400 MHz) d (ppm): 0.82 (t,
3H, J = 7.4), 1.50 (m, 2 H), 2.49 (s, 3H), 3.72 (t, 2H, J = 7.5), 5.23 (s,
2H), 12 (br s, 1H).
The binding disposition of these molecules was investigated by
means of a docking approach using the mixed pharmacophoric-
molecular modeling procedure and the A3 receptor model previ-
ously reported.48 The obtained results highlighted the interaction
4.1.2. General procedure for the synthesis of 8-bromo-7-(2-oxo-
propyl)-1-propyl-3-pyridin-2/3/4-ylmethyl-3,7-dihydro-purine-
2,6-dione derivatives (5a–c)
To a solution of 4 (0.3 mmol) in anhydrous DMF (3 mL), K2CO3
was added (0.6 mmol) and the resulting suspension was stirred
at room temperature for 1 h. After cooling at 0 °C a solution of
the appropriate alkyl halide (0.33 mmol) in DMF (2 mL) was added
dropwise. The reaction proceeded under stirring at 0 °C for 5’ and
at room temperature for further 50’. The solvent was evaporated
under vacuum to obtain a residue which was suspended with
H2O, acidified with a 5% aqueous solution of HCl and extracted
with EtOAc (2ꢀ 20 mL). The aqueous phase was then basified with
NaOH 5% and extracted a second time with EtOAc (3ꢀ 20 mL). The
second organic layer was dried on Na2SO4, filtered and the solvent
was evaporated under vacuum till dryness.
4.1.3. 8-Bromo-7-(2-oxo-propyl)-1-propyl-3-pyridin-4-ylmethyl-
3,7-dihydro-purine-2,6-dione (5a)
The product was purified by column chromatography on silica
gel (EtOAc/light petroleum 7:3). Yield 60%, white solid; mp 118–
120 °C; 1H NMR (DMSO-d6, 400 MHz): d (ppm) 0.83 (t, 3H,
J = 7.2), 1.53 (m, 2H), 2.30 (s, 3H), 3.79 (m, 2H), 5.17 (s, 2H), 5.30
(s, 2H), 7.26 (dd, 2H, J = 8), 8.51 (dd, 2H, J = 8).
Figure 6. Negative (a, c) and positive (b,d) regions of the PLS coefficient plot
obtained with the N2@ (a,b) and O (c,d) probe. Compound 14c is also displayed.

P. G. Baraldi et al. / Bioorg. Med. Chem. 16 (2008) 10281–10294
10289
4.1.4. 8-Bromo-7-(2-oxo-propyl)-1-propyl-3-pyridin-3-ylmethyl-
3,7-dihydro-purine-2,6-dione (5b)
4.1.12. 8-Bromo-7-(2-oxo-propyl)-1-propyl-3-pyridin-3-ylmethyl-
3,7-dihydro-purine-2,6-dione hydrochloride (6b ꢀ HCl)
Yield 95%, white solid; 300 °C dec.; 1H NMR (DMSO-d6,
200 MHz): d (ppm) 0.85 (t, 3H, J = 7.2), 1.58 (m, 2H), 2.27 (s, 3H),
3.82 (m, 2H), 5.31 (s, 2H), 7.27 (s, 1H), 7.80 (m, 1H), 8.26 (m,
1H), 8.70–8.83 (m, 2H), 12.37 (s, 1H).
The product was purified by column chromatography on silica
gel (EtOAc/light petroleum 7:3). Yield 75%, white solid; mp
154 °C; 1H NMR (DMSO-d6, 200 MHz): d (ppm) 0.81 (t, 3H,
J = 7.6), 1.50 (m, 2H), 2.27 (s, 3H), 3.76 (m, 2H), 5.15 (s, 2H), 5.26
(s, 2H), 7.32–7.38 (m, 1H), 7.68–7.74 (m, 1H), 8.46 (m, 1H), 8.56
(m, 1H).
4.1.13. 8-Bromo-7-(2-oxo-propyl)-1-propyl-3-pyridin-2-ylmethyl-
3,7-dihydro-purine-2,6-dione hydrochloride (6c ꢀ HCl)
Yield 90%, white solid; mp >300 °C; 1H NMR (DMSO-d6,
200 MHz): d (ppm) 0.85 (t, 3H, J = 7.2), 1.55 (m, 2H), 2.26 (s, 3H),
3.87 (m, 2H), 5.35 (s, 2H), 7.27 (s, 1H), 7.45 (m, 2H), 7.95 (m,
1H), 8.56 (m, 1H), 12.26 (s, 1H).
4.1.5. 8-Bromo-7-(2-oxo-propyl)-1-propyl-3-pyridin-2-ylmethyl-
3,7-dihydro-purine-2,6-dione (5c)
The product was purified by column chromatography on silica
gel (EtOAc/light petroleum 7:3). Yield 65%, white solid; mp
160 °C; 1H NMR (DMSO-d6, 200 MHz): d (ppm) 0.80 (t, 3H,
J = 7.5), 1.48 (m, 2H), 2.27 (s, 3H), 3.76 (m, 2H), 5.22 (s, 2H),
5.28 (s, 2H), 7.22–7.31 (m, 2H), 7.71–7.79 (m, 1H), 8.42–8.45
(m, 1H).
4.1.14. 3-Benzyl-8-bromo-3,7-dihydro-purine-2,6-dione (8)
A mixture of 3-benzyl-xanthine 732 (16.5 mmol) and anhydrous
CH3CO2Na (33 mmol) in AcOH (90 mL) was heated under stirring
at 110 °C, then Br2 was added (16.5 mmol). The reaction proceeded
at the same temperature for 3 h. The solvent was evaporated and
the residue was suspended with cold water and cooled at 0 °C.
The product was filtered, washed with cold water and dried under
vacuum. Yield 80% white solid; mp 287–289 °C; 1H NMR (DMSO-
d6, 200 MHz) d (ppm): 5.06 (s, 2H), 7.19–7.36 (m, 5H), 11.28 (s,
1H), 14.38 (br s, 1H).
4.1.6. General procedure for the synthesis of the final 7-methyl-
3-propyl-1-pyridin-(2/3/4)-ylmethyl-1H,8H-imidazo[2,1-f]purine-
2,4-dione derivatives (6a–c)
The appropriate 8-bromo-7-(2-oxo-propyl)-3,7-dihydro-pur-
ine-2,6-dione derivatives 5a–c (0.4 mmol) was dissolved in EtOH
(4 mL) and liquid ammonia (2–3 mL) was added. The reaction
was heated at 100–120 °C in a steel bomb overnight. The solvent
was evaporated till dryness.
4.1.15. 3-Benzyl-8-bromo-1,7-bis-(2-oxo-propyl)-3,7-dihydro-
purine-2,6-dione (9)
4.1.7. 7-Methyl-3-propyl-1-pyridin-4-ylmethyl-1H,8H-imidazo-
[2,1-f]purine-2,4-dione (6a)
The 8-bromo-3-benzyl-xanthine 8 (3.11 mmol) was dissolved
in anhydrous DMF (4 mL) and Na2CO3 (0.66 g, 6.22 mmol) was
The product was purified by column chromatography on silica
gel (CH2Cl2/CH3OH 9.5:0.5). Yield 40%, white solid; mp 275 °C;
1H NMR (DMSO-d6, 400 MHz): d (ppm) 0.86 (t, 3H, J = 7.6), 1.56
(m, 2H), 2.26 (s, 3H), 3.84 (m, 2H), 5.18 (s, 2H), 7.26 (m, 3H),
8.48 (d, 2H, J = 6), 12.30 (br s, 1H).
added to the solution. A solution of a-chloro-acetone (9.33 mmol)
in anhydrous DMF (2 mL) was then added dropwise within 1 h. The
reaction was stirred at room temperature overnight. The solvent
was evaporated and the residue was partitioned between H2O
(30 mL) and CH2Cl2 (3 ꢀ 50 mL). The organic layer was dried on
anhydrous Na2SO4, filtered the solvent was evaporated to give a
residue which was purified by crystallization with a mixture of
CH2Cl2/light petroleum 1:2. Yield 88% white solid; mp 248–
250 °C; 1H NMR (DMSO-d6, 200 MHz) d (ppm): 2.17 (s, 3H), 2.26
(s, 3H), 4.71 (s, 2H), 5.13 (s, 2H), 5.26 (s, 2H), 7.31 (m, 5H).
4.1.8. 7-Methyl-3-propyl-1-pyridin-3-ylmethyl-1H,8H-imidazo-
[2,1-f]purine-2,4-dione (6b)
The product was purified by column chromatography on silica
gel (EtOAc/CH3OH 9.5:0.5). Yield 50%, white solid; 280 °C dec.; 1H
NMR (DMSO-d6, 200 MHz): d (ppm) 0.83 (t, 3H, J = 7.2), 1.53 (m,
2H), 2.23 (s, 3H), 3.82 (m, 2H), 5.24 (s, 2H), 7.16–7.26 (m, 3H),
7.65–7.40 (m, 1H), 8.42 (m, 1H), 12.30 (br s, 1H).
4.1.16. 3-Benzyl-8-bromo-7-(2-oxo-propyl)-3,7-dihydro-purine-
2,6-dione (10)
The product was prepared according to the procedure described
for the synthesis of 9 but reacting the 8-bromo-3-benzyl-xanthine
4.1.9. 7-Methyl-3-propyl-1-pyridin-2-ylmethyl-1H,8H-imidazo-
[2,1-f]purine-2,4-dione (6c)
(3.11 mmol) with 3.11 mmol of Na2CO3 and 3.11 mmol of
a-
The product was purified by column chromatography on silica
gel (EtOAc/CH3OH 9.5:0.5). Yield 80%, white solid; mp 285 °C; 1H
NMR (DMSO-d6, 200 MHz): d (ppm) 0.82 (t, 3H, J = 7.2), 1.53 (m,
2H), 2.24 (s, 3H), 3.81 (m, 2H), 5.16 (s, 2H), 7.21–7.34 (m, 2H),
7.69–7.74 (m, 1H), 8.44 (m, 1H), 8.58 (m, 1H), 12.30 (br s, 1H).
chloro-acetone. Yield 70% white solid; mp 207–209 °C; 1H NMR
(DMSO-d6, 200 MHz) d (ppm): 2.26 (s, 3H), 5.06 (s, 2H), 5.23 (s,
2H), 7.31 (m, 5H), 11.47 (br s, 1H).
4.1.17. General procedure for the preparation of 3-benzyl-8-
bromo-1-substituted-7-(2-oxo-propyl)-3,7-dihydro-purine-2,6-
dione derivatives (11a–f)
4.1.10. General procedure for the synthesis of the hydrochlorides
salts of compounds (6a–c)
A solution of 10 (0.40 mmol) in anhydrous DMF (2 mL) was stir-
red at room temperature and K2CO3 (0.40 mmol) was added. After
5’ a solution of the appropriate alkyl halides (0.80 mmol), in DMF
(2 mL), was slowly added dropwise. The reaction was stirred at
room temperature and monitored with TLC (2–5 h). The solvent
was evaporated to obtain a residue which was suspended with
CH2Cl2 (20 mL) and washed with H2O (2ꢀ 10 mL). The organic
phase was dried on anhydrous Na2SO4, filtered and the solvent
was evaporated till dryness.
The appropriate tricyclic derivative 6a–c as
a free base
(0.1 mmol), was suspended in a saturated ethanolic solution of
HCl. The mixture was stirred at room temperature for 30’, then
the solvent was evaporated and the residue coevaporated for three
times with Et2O. The hydrochloride salts were purified via crystal-
lization from EtOH.
4.1.11. 8-Bromo-7-(2-oxo-propyl)-1-propyl-3-pyridin-4-ylmethyl-
3,7-dihydro-purine-2,6-dione hydrochloride (6a ꢀ HCl)
Yield 90%, white solid; mp >300 °C; 1H NMR (DMSO-d6,
200 MHz): d (ppm) 0.86 (t, 3H, J = 7.2), 1.57 (m, 2H), 2.27 (s, 3H),
3.84 (m, 2H), 5.39 (s, 2H), 7.29 (s, 1H), 7.80 (d, 2H, J = 6), 8.75 (d,
2H, J = 6), 12.32 (s, 1H).
4.1.18.3-Benzyl-8-bromo-1-(2-hydroxy-ethyl)-7-(2-oxo-propyl)-
3,7-dihydro-purine-2,6-dione (11a)
The intermediate was purified via crystallization with a mix-
ture of Et2O/light petroleum 1:2. Yield 70% white solid; mp 143–

10290
P. G. Baraldi et al. / Bioorg. Med. Chem. 16 (2008) 10281–10294
144 °C; 1H NMR (DMSO-d6, 200 MHz): d (ppm) 2.29 (s, 3H), 3.49
(m, 2H), 3.92 (m, 2H), 4.77 (bt, 1H), 5.13 (s, 2H), 5.29 (s, 2H),
7.33 (m, 5H).
4.1.27. 1-Benzyl-7-methyl-3-(2-hydroxy-ethyl)-1H,8H-
imidazo[2,1-f]purine-2,4-dione (14a)
The product was purified by column chromatography on silica
gel (CH2Cl2/CH3OH 9.5:0.5). Yield 60%, white solid; mp 257–
259 °C; 1H NMR (DMSO-d6, 200 MHz): d (ppm) 2.50 (s, 3H), 3.52
(t, 2H, J = 6.6), 3.97 (t, 2H, J = 6.6), 4.9 (br s, 1H), 5.15 (s, 2H),
7.23–7.33 (m, 6H), 12.30 (br s, 1H).
4.1.19. 3-benzyl-8-bromo-1-(3-hydroxy-propyl)-7-(2-oxo-propyl)-
3,7-dihydro-purine-2,6-dione (11b)
The product was purified by column chromatography on sil-
ica gel (CH2Cl2/CH3OH 9.9:0.1). Yield 65%, colourless oil; 1H
NMR (DMSO-d6, 200 MHz): d (ppm) 1.81 (m, 2H), 2.32 (s, 3H),
3.46 (t, 2H), 4.09 (t, 2H), 5.15 (s, 2H), 5.22 (s, 2H), 7.25–7.50
(m, 5H).
4.1.28. 1-Benzyl-7-methyl-3-(2-hydroxy-propyl)-1H,8H-imidazo-
[2,1-f]purine-2,4-dione (14b)
The product was purified by column chromatography on silica
gel (CH2Cl2/CH3OH 9.5:0.5). Yield 70%, white solid; mp 259–
260 °C; 1H NMR (DMSO-d6, 200 MHz): d (ppm) 1.69 (m, 2H), 2.26
(s, 3H), 3.43 (m, 2H), 3.93 (t, 2H), 4.40 (bt, 1H), 5.16 (s, 2H),
7.24–7.33 (m, 6H), 12.20 (br s, 1H).
4.1.20. 1-Allyl-3-benzyl-8-bromo-7-(2-oxo-propyl)-3,7-dihydro-
purine-2,6-dione (11c)
The product was purified by column chromatography on silica
gel (CH2Cl2/CH3OH 9.8:0.2): Yield 80%, white solid; mp 145 °C;
1H NMR (DMSO-d6, 400 MHz): d (ppm) 2.28 (s, 3H), 4.43 (d, 2H,
J = 2.6), 5.07 (m, 2H), 5.14 (s, 2H), 5.29 (s, 2H), 5.80 (m,1H), 7.34
(m, 5H).
4.1.29. 1-Benzyl-7-methyl-3-allyl-1H,8H-imidazo[2,1-f]purine-
2,4-dione (14c)
The product was crystallized from a mixture of CH3OH/H2O 1:1.
Yield 70%, white solid; mp 258–260 °C; 1H NMR (DMSO-d6,
200 MHz): d (ppm) 2.23 (s, 3H), 4.45 (d, 2H, J = 4), 4.97 (m, 2H),
5.13 (s, 2H), 5.82 (m, 1H), 7.21–7.33 (m, 6H), 12.00 (br s, 1H).
4.1.21. 3-Benzyl-8-bromo-7-(2-oxo-propyl)-1-prop-2-ynyl-3,7-
dihydro-purine-2,6-dione (11d)
The intermediate was purified via crystallization with a mixture
of Et2O/light petroleum 1:2: Yield 77%, brown solid; mp 147 °C; 1H
NMR (DMSO-d6, 200 MHz): d (ppm) 2.16 (m, 1H), 2.34 (s, 3H), 4.70
(d, 2H, J = 2.4), 5.17 (s, 2H), 5.25 (s, 2H), 7.29–7.40 (m, 3H), 7.51–
7.56 (m, 2H).
4.1.30. 1-Benzyl-7-methyl-3-prop-2-ynyl-1H,8H-imidazo[2,1-
f]purine-2,4-dione (14d)
The product was purified by column chromatography on silica
gel (CH2Cl2/CH3OH 9.5:0.5). Yield 60%, white solid; mp 248–
249 °C; 1H NMR (DMSO-d6, 200 MHz): d (ppm) 2.22 (s, 3H), 4.57
(d, 2H, J = 2), 5.13 (s, 2H), 7.20–7.33 (m, 6H), 12.20 (br s, 1H).
4.1.22. [3-Benzyl-8-bromo-2,6-dioxo-7-(2-oxo-propyl)-2,3,6,7-
tetrahydro-purin-1-yl]-acetic acid ethyl ester (11e)
The intermediate was purified via crystallization with a mixture
of Et2O/light petroleum 1:2. Yield 62%, pale yellow solid; mp 174–
176 °C; 1H NMR (DMSO-d6, 200 MHz): d (ppm) 1.16 (t, 3H, J = 7.3),
2.28 (s, 3H), 4.10 (q, 2H, J = 7.3), 4.59 (s, 2H), 5.16 (s, 2H), 5.29 (s,
2H), 7.32 (m, 5H).
4.1.31. (1-Benzyl-7-methyl-2,4-dioxo-1,2,4,8-tetrahydro-imidazo-
[2,1-f]purin-3-yl)-acetic acid ethyl ester (14e)
The product was purified by column chromatography on silica
gel (CH2Cl2/CH3OH 9.5:0.5). Yield 55%, white solid; mp 290–
291 °C; 1H NMR (DMSO-d6, 400 MHz): d (ppm) 1.20 (t, 3H,
J = 7.6), 2.30 (s, 3H), 4.15 (q, 2H, J = 7.6), 4.60 (s, 2H), 5.15 (s, 2H),
7.20–7.40 (m, 6H), 12.00 (br s, 1H).
4.1.23. 3-Benzyl-8-bromo-1-(2-dimethylamino-ethyl)-7-(2-oxo-
propyl)-3,7-dihydro-purine-2,6-dione (11f)
The intermediate was purified via crystallization with a mixture
of Et2O/light petroleum 1:2: Yield 70%, white solid; mp 130–
131 °C; 1H NMR (DMSO-d6, 400 MHz): d (ppm) 2.16 (s, 6H), 2.30
(s, 3H), 2.40 (t, 2H, J = 7.2), 3.90 (t, 2H, J = 7.2), 5.17 (s, 2H), 5.30
(s, 2H), 7.40 (m, 5H).
4.1.32. 1-Benzyl-3-(2-dimethylamino-ethyl)-7-methyl-1H,8H-
imidazo[2,1-f]purine-2,4-dione (14f)
The product was purified by column chromatography on silica
gel (CH2Cl2/CH3OH 9:1). Yield 80%, white solid; mp 245 °C; 1H
NMR (DMSO-d6, 400 MHz): d (ppm) 2.19 (s, 6H), 2.23 (s, 3H),
2.40 (t, 2H, J = 7.3), 4.00 (t, 2H, J = 7.3), 5.20 (s, 2H), 7.20–7.40 (m,
6H), 12.00 (br s, 1H).
4.1.24. General procedure for the synthesis of the final 1-benzyl-
3-substituted-7-methyl-1H,8H-imidazo[2,1-f]purine-2,4-dione
derivatives (12, 13, 14a–f)
The appropriate 8-bromo-7-(2-oxo-propyl)-3,7-dihydro-pur-
ine-2,6-dione derivatives (9, 10, 11a–f) (0.4 mmol) was dissolved
in EtOH (4 mL) and liquid ammonia (2–3 mL) was added. The reac-
tion was heated at 100–120 °C in a steel bomb overnight. The sol-
vent was evaporated till dryness.
4.1.33. General procedure for the preparation of 8-substituted-
7-methyl-1H,8H-imidazo[2,1-f]purine-2,4-dione derivatives
(15a–c)
A mixture of 3 (0.36 mmol) or 10 (0.36 mmol), triphenylphos-
phine (0.36 mmol) and the appropriate amine (0.43 mmol) in
anhydrous THF (3 mL), was stirred at room temperature under ar-
gon atmosphere while
a solution of diethyazodicarboxylate
4.1.25. 1-Benzyl-7-methyl-3-(2-oxo-propyl)-1H,8H-imidazo-
[2,1-f]purine-2,4-dione (12)
The product was crystallized from EtOAc. Yield 85%, white solid;
mp 280–282 °C; 1H NMR (DMSO-d6, 200 MHz): d (ppm) 2.17 (s,
3H), 2.25 (s, 3H), 4.72 (s, 2H), 5.15 (s, 2H), 7.23–7.31 (m, 6H),
12.00 (br s, 1H).
(0.36 mmol) in THF was added dropwise. After 1 h at room temper-
ature, the solvent was evaporated and the residue partitioned be-
tween EtOAc (20 mL) and H2O (2ꢀ 10 mL). The organic phase
was dried on Na2SO4, filtered and the solvent was evaporated un-
der vacuum till dryness.
4.1.34. 1-Benzyl-8-(2-hydroxy-1-methyl-ethyl)-7-methyl-1H,8H-
imidazo[2,1-f]purine-2,4-dione (15a)
4.1.26. 1-Benzyl-7-methyl-1H,3H,8H-imidazo[2,1-f]purine-2,4-
dione (13)
The product was crystallized from a mixture of CH3OH/H2O 1:1.
Yield 75%, white solid; mp >300 °C; 1H NMR (DMSO-d6, 200 MHz):
d (ppm) 2.25 (s, 3H), 5.09 (s, 2H), 7.23–7.32 (m, 6H), 10.95 (br s,
1H), 13.00 (br s, 1H).
The product was purified by column chromatography on silica
gel (CH2Cl2/CH3OH 9.5:0.5). Yield 70%, white solid; mp 251–
253 °C; 1H NMR (DMSO-d6, 200 MHz): d (ppm) 1.50 (d, 3H,
J = 6.8), 2.29 (s, 3H), 3.80 (m, 1H), 3.90 (m, 1H), 4.30 (m, 1H),
4.98 (bt, 1H), 5.09 (s, 2H), 7.26–7.38 (m, 6H), 10.91 (s, 1H).

P. G. Baraldi et al. / Bioorg. Med. Chem. 16 (2008) 10281–10294
10291
4.1.35. 1-Benzyl-8-ethyl-7-methyl-1H,8H-imidazo[2,1-f]purine-
2,4-dione (15b)
nally purified by crystallisation from EtOH: pale yellow solid; 70%
yield; mp 160–161 °C; 1H NMR (CDCl3, 200 MHz): d 0.87–1.03 (m,
6H), 1.57–1.84 (m, 4H), 3.85–3.92 (m, 2H), 4.02–4.10 (m, 2H),
5.84 (s, 2H), 7.51–7.68 (m, 3H), 7.99–8.05 (m, 2H).
Il prodotto è stato The product was purified by column chroma-
tography on silica gel (CH2Cl2/CH3OH 9.9:0.1). Yield 50%, white so-
lid; mp >300 °C; 1H NMR (DMSO-d6, 200 MHz): d (ppm) 1.27 (t, 3H,
J = 7.4), 2.26 (s, 3H), 3.98 (q, 2H, J = 7.4), 5.07 (s, 2H), 7.26–7.29 (m,
6H), 10.91 (s, 1H).
4.1.44. 1,3-Diisobutyl-8-bromo-7-(2-oxo-2-phenyl-ethyl)-3,7-
dihydro-purine-2,6-dione (19b)
The product was synthesised according to the general alkylation
4.1.36. 1-Benzyl-8-ethyl-7-methyl-3-propyl-1H,8H-imidazo[2,1-f]-
purine-2,4-dione (15c)
procedure at the N7-position described for the preparation of com-
pound 10, using a-bromoacetophenone as alkylating agent, and fi-
The product was purified by column chromatography on silica
gel (CH2Cl2/CH3OH 9.9:0.1). Yield 55%, white solid; mp 190 °C;
1H NMR (DMSO-d6, 200 MHz): d (ppm) 0.84 (t, 3H, J = 7.3), 1.31
(t, 3H, J = 7.4), 1.53 (m, 2H), 2.32 (s, 3H), 3.83 (m, 2H), 4.01 (m,
2H), 5.18 (s, 2H), 7.27–7.33 (m, 6H).
nally purified by crystallisation from EtOH: pale yellow solid; 75%
yield; mp 178–181 °C; 1H NMR (CDCl3, 200 MHz): d 0.86–0.98 (m,
12H), 2.10 (m, 1H), 2.30 (m, 1H), 3.78 (d, 2H, J = 7.6), 3.93 (d, 2H,
J = 7.6), 5.84 (s, 2H), 7.51–7.68 (m, 3H), 7.99–8.04 (m, 2H).
4.1.45. 1,3-Dipropyl-7-phenyl-1H,8H-imidazo[2,1-f]purine-2,4-
dione (20a)
4.1.37. General procedure for the preparation of 1,3-dialkyl-
3,7-dihydro-purine-2,6-dione derivatives (17a–b)
The product was synthesised according to the general cyclisa-
tion procedure described for the preparation of compound 6a–c,
12, 13, 14a–f and finally purified by column chromatography on
silica gel eluting with a 3:2 mixture of light petroleum–EtOAc:
white solid; 50% yield; mp 249–250 °C; 1H NMR (DMSO-d6,
200 MHz): d 0.87 (m, 6H), 1.40–1.80 (m, 4H), 3.80–4.05 (m, 4H),
7.30–7.60 (m, 3H), 7.80 (m, 2H), 8.10 (s, 1H), 13.00 (br s, 1H).
A solution of 5,6-diamino-1,3-dialkyl-1H-pyrimidine-2,4-dione
derivatives (16a–b) (3.28 mmol) were dissolved in formic acid
(15 mL) and the resultant violet solution was refluxed for 1 h.
The excess of acid was evaporated to obtain a yellow residue which
was dissolved in a 1:1 mixture of EtOH–10% aqueous NaOH. The
reaction mixture was refluxed for 1 h then the solvents were con-
centrated to half-volume. The products were precipitated by acid-
ifying with 10% HCl, filtered and washed with cold water.
4.1.46. 1,3-Diisobutyl-7-phenyl-1H,8H-imidazo[2,1-f]purine-
2,4-dione (20b)
4.1.38. 1,3-Dipropyl-3,7-dihydro-purine-2,6-dione (17a)
Crystallization from ethanol gave the desired product as a pale
yellow solid: 95% yield; mp 205 °C; 1H NMR (CDCl3, 200 MHz): d
0.98 (t, 6H, J = 7.6), 1.67–1.88 (m, 4H), 4.03–4.16 (m, 4H), 7.81 (s,
1H), 12.98 (br s, 1H).
The product was synthesised according to the general cyclisa-
tion procedure described for the preparation of compound 6a–c,
12, 13, 14a–f and finally purified by column chromatography on
silica gel eluting with a 7:3 mixture of light petroleum–EtOAc:
white solid; 60% yield; mp 170 °C; 1H NMR (DMSO-d6, 200 MHz):
d 0.87 (m, 12H), 2.00–2.30 (m, 2H), 3.70–3.90 (m, 4H), 7.35–7.50
(m, 3H), 7.80 (m, 2H), 8.10 (s, 1H), 13.00 (br s, 1H).
4.1.39. 1,3-Diisobutyl-3,7-dihydro-purine-2,6-dione (17b)
Crystallization from ethanol gave the desired product as a pale
yellow solid: 90% yield; mp 194 °C; 1H NMR (DMSO-d6, 200 MHz):
d 0.84 (m, 12H), 1.90–2.20 (m, 2H), 3.73 (m, 4H), 7.79 (s, 1H), 13.00
(br s, 1H).
4.1.47. 1,3-Dimethyl-8-amino-3,7-dihydro-purine-2,6-dione
(22)
To a mixture of 8-nitro-theophylline (21, 1 g, 4.44 mmol) in
20 mL of methanol and hydrazine monohydrate (1 mL) was
added 10% C/Pd (200 mg) and the reaction was refluxed until
the yellow colour disappeared. 5% NaOH solution (5 mL) was
added to the reaction mixture and the suspension was filtered
through Celite. The filtrate was concentrated to half-volume
and finally acidified with acetic acid to precipitate the product
as a white solid. The precipitate was washed with water and
dried under vacuum: white solid; 80% yield; mp >300 °C; 1H
NMR (DMSO-d6, 200 MHz): d 3.17 (s, 3H), 3.33 (s, 3H), 6.52 (s,
2H), 11.17 (br s, 1H).
4.1.40. General procedure for the preparation of 1,3-dialkyl-8-
bromo-3,7-dihydro-purine-2,6-dione derivatives (18a–b)
A solution of the appropriate 1,3-dialkyl-3,7-dihydro-purine-
2,6-dione derivative (17a–b) (2.82 mmol) and sodium acetate
(2.82 mmol) in acetic acid (15 mL) was warmed at 45 °C under stir-
ring. To the mixture 0.15 mL of bromine (2.82 mmol) were added
dropwise. After 1 h at 45 °C the solvent was evaporated, the resi-
due suspended with water and extracted with EtOAc
(3ꢀ 100 mL). The organic layers were dried with anhydrous so-
dium sulphate and evaporated to dryness.
4.1.48. 1,3-Dimethyl-8-amino-7-(2-oxo-2-phenyl-ethyl)-3,7-
dihydro-purine-2,6-dione (23a)
4.1.41. 1,3-Dipropyl-8-bromo-3,7-dihydro-purine-2,6-dione(18a)
Crystallization from EtOH gave the desired product as a white
solid: 70% yield; mp 193 °C; 1H NMR (CDCl3, 200 MHz): d 0.88 (t,
6H, J = 7.6), 1.65–1.85 (m, 4H), 4.00–4.15 (m, 4H), 13.33 (br s, 1H).
The product was synthesised according to the general alkylation
procedure at the N7-position described for the preparation of com-
pound 10, using a-bromoacetphenone as alkylating agent, and
purified by crystallization from EtOH. White solid: 80% yield; mp
248–250 °C; 1H NMR (DMSO-d6, 200 MHz): d 3.10 (s, 3H), 3.36 (s,
3H), 5.60 (s, 2H), 7.00 (s, 2H), 7.70 (m, 5H).
4.1.42. 1,3-Diisobutyl-8-bromo-3,7-dihydro-purine-2,6-dione
(18b)
Crystallization from EtOH gave the desired product as a white
solid: 80% yield; mp 205–207 °C; 1H NMR (DMSO-d6, 200 MHz):
d 0.82 (m, 12H), 1.8–2.10 (m, 2H), 3.73 (m, 4H), 13.20 (br s, 1H).
4.1.49. 1,3-Dimethyl-8-amino-7-(2-oxo-propyl)-3,7-dihydro-
purine-2,6-dione (23b)
The product was synthesised according to the general alkylation
procedure at the N7-position described for the preparation of com-
pounds 5a–c. Crystallization from EtOH gave the desired product
as a white solid: 80% yield; mp 230–232 °C; 1H NMR (DMSO-d6,
200 MHz): d 2.50 (s, 3H), 3.20 (s, 3H), 3.50 (s, 3H), 5.20 (s, 2H),
7.00 (s, 2H).
4.1.43. 1,3-Dipropyl-8-bromo-7-(2-oxo-2-phenyl-ethyl)-3,7-
dihydro-purine-2,6-dione (19a)
The product was synthesised according to the general alkylation
procedure at the N7-position described for the preparation of com-
pound 10, using a-bromoacetophenone as alkylating agent, and fi-

10292
P. G. Baraldi et al. / Bioorg. Med. Chem. 16 (2008) 10281–10294
4.1.50. General procedure for the preparation of 1,3-dimethyl-
7-substituted-1H,8H-imidazo[2,1-f]purine-2,4-dione
derivatives (24a–b)
The appropriate 1,3-dimethyl-8-amino-7-(2-oxo-alkyl)-3,7-
dihydro-purine-2,6-dione derivatives 23a–b (0.3 mmol) was
heated at reflux in acetic acid (5 mL) for 4 h. The solvent was evap-
orated and the residue crystallized from the opportune solvents.
containing 1 nM [3H]-MRE 3008F20, 50 mM Tris–HCl buffer,
10 mM MgCl2, 1 mM EDTA, pH 7.4, and 100 l of diluted mem-
branes (50 g of protein/assay) and at least 6–8 different concen-
trations of examined ligands for 120 min at 4 °C. Non-specific
binding was defined as binding in the presence of 1 M MRE
3008F20 and was about 25% of total binding. Bound and free radio-
activities were separated by filtering the assay mixture through
Whatman GF/B glass fibre which were washed three times with
ice-cold buffer. Filter bound radioactivity was measured by scintil-
lation spectrometry (LS-1800 Beckman) after addition of Aquas-
sure liquid.
l
l
l
4.1.51. 1,3-Dimethyl-7-phenyl-1H,8H-imidazo[2,1-f]purine-2,4-
dione (24a)
Crystallization from DMF gave the desired product as a white
solid: 70% yield; mp 283–285 °C; 1H NMR (DMSO-d6, 200 MHz):
d 3.24 (s, 3H), 3.44 (s, 3H), 7.42 (m, 3H), 7.78 (m, 2H), 8.07 (s,
1H), 12.90 (br s, 1H).
4.2.3. Measurement of cyclic AMP levels in CHO cells
transfected with human A2B adenosine receptors
CHO cells transfected with human A2B adenosine receptors
were washed with phosphate-buffered saline, diluted trypsine
and centrifuged for 10 min at 200g. The pellet containing the
CHO cells (1 ꢀ 106 cells /assay) was suspended in 0.5 ml of incuba-
tion mixture (mM): NaCl 15, KCl 0.27, NaH2PO4 0.037, MgSO4 0.1,
CaCl2 0.1, Hepes 0.01, MgCl2 1, glucose 0.5, pH 7.4, at 37 °C, 2 IU/ml
adenosine deaminase and 4-(3-butoxy-4-methoxybenzyl)-2-imi-
dazolidinone (Ro 20–1724) as phosphodiesterase inhibitor and
preincubated for 10 min in a shaking bath at 37 °C. The potencies
of antagonists studied versus hA2B adenosine receptors were deter-
mined by antagonism of NECA (100 nM)-induced stimulation of
cyclic AMP levels. The reaction was terminated by the addition of
cold 6% trichloroacetic acid (TCA). The TCA suspension was centri-
fuged at 2000 g for 10 min at 4 °C and the supernatant was ex-
tracted four times with water saturated diethyl ether. The final
aqueous solution was tested for cyclic AMP levels by a competition
protein binding assay. Samples of cyclic AMP standard (0–10 pmol)
were added to each test tube containing the incubation buffer (tri-
zma base 0.1 M, aminophylline 8.0 mM, 2-mercaptoethanol
6.0 mM, pH 7.4) and [3H] cyclic AMP in a total volume of 0.5 ml.
The binding protein previously prepared from beef adrenals, was
added to the samples previously incubated at 4 °C for 150 min,
and after the addition of charcoal were centrifuged at 2000g for
10 min. The clear supernatant was counted in a LS-1800 Beckman
scintillation counter.
4.1.52. 1,3,7-Trimethyl-1H,8H-imidazo[2,1-f]purine-2,4-dione
(24b)
Crystallization from DMF gave the desired product as a white
solid: 60% yield; mp 247–248 °C; 1H NMR (DMSO-d6, 200 MHz):
d 2.66 (s, 3H), 3.22 (s, 3H), 3.41 (s, 3H); 7.23 (s, 1H), 12.20 (br s, 1H).
4.2. Biology experiments
4.2.1. CHO membranes preparation
The human A1, A2A, A2B and A3 receptors has been transfected in
CHO cells according with the method previously described.53 The
cells were grown adherently and maintained in Dulbecco’s modi-
fied Eagle’s medium with nutrient mixture F12 (DMEM/F12) with-
out nucleosides, containing 10% foetal calf serum, penicillin
(100 U/ml), streptomycin (100 lg/ml), L-glutamine (2 mM) and
Geneticin (G418, 0,2 mg/ml) at 37 °C in 5% CO2/95% air. Cells were
split 2 or 3 times weekly at a ratio between 1:5 and 1:20. For mem-
brane preparation the culture medium was removed and the cells
were washed with PBS and scraped off T75 flasks in ice-cold hypo-
tonic buffer (5 mM Tris–HCl, 2 mM EDTA, pH 7.4). The cell suspen-
sion was homogenized with Polytron and the homogenate was
spun for 10 min at 1000g. The supernatant was then centrifuged
for 30 min at 100,000g. The membrane pellet was suspended in
50 mM Tris–HCl buffer pH 7.4 (for A3 adenosine receptors:
50 mM Tris–HCl, 10 mM MgCl2, 1 mM EDTA) and incubated with
3 UI/ml of adenosine deaminase for 30 min at 37 °C. Then the cell
suspension was frozen at ꢂ80 °C.
4.2.4. Data analysis
The protein concentration was determined according to a Bio-
Rad method57 with bovine albumin as a standard reference. Inhib-
itory binding constants, Ki values, were calculated from those of
IC50 according to Cheng & Prusoff equation58 Ki = IC50/(1+[C*]/
4.2.2. Human cloned A1, A2A and A3 adenosine receptor binding
assay
*
All synthesized compounds have been tested for their affinity
at human A1, A2A and A3 adenosine receptors. Displacement
experiments of [3H]-DPCPX to CHO cells transfected with the hu-
man recombinant A1 adenosine receptor were performed for
120 min at 25 °C in 0.2 ml of 50 mM Tris–HCl buffer, pH 7.4,
KD*), where [C*] is the concentration of the radioligand and KD its
dissociation constant. A weighted non-linear least-squares curve
fitting program LIGAND59 was also used for computer analysis of
inhibition experiments. Potency values (IC50) obtained in cyclic
AMP assays were calculated by non-linear regression analysis
using the equation for a sigmoid concentration-response curve
(Graph PAD Prism, San Diego, CA, USA). Affinity values are ex-
pressed as geometric mean, with 95% or 99% confidence limits in
parentheses and IC50 values are expressed as the arithmetic
means SEM.
containing 1 nM [3H]-DPCPX, diluted membranes (50
lg of pro-
tein/assay) and at least 6–8 different concentrations of antago-
nists studied. Non-specific binding was determined in the
presence of 10
binding.54
l
M of CHA and this was always ꢃ10% of the total
Binding of [3H]-ZM 241385 to CHO cells transfected with the
human recombinant A2A adenosine receptors (50
lg of protein/as-
4.3. Receptor modeling
say) was performed using 0.2 ml of 50 mM Tris–HCl buffer, 10 mM
MgCl2, pH 7.4, and at least 6–8 different concentrations of antago-
nists studied for an incubation time of 60 min at 4 °C. Non-specific
Molecular mechanics (MM) and molecular dynamics (MD) cal-
culations were performed using the AMBER force field as imple-
binding was determined in the presence of 1
l
M ZM 241385 and
mented in the MacroModel software package,60 using
a
was about 20% of total binding.55
‘‘distance-dependent” dielectric constant of 4.0. Electrostatic
charges for compound 1a were calculated with RHF/AM1 semi-
empirical calculation and RESP program.61
All MM minimizations were performed with conjugate gradient
or steepest descent as minimizers and a threshold value of 0.05 kJ/
Binding of [3H]-MRE 3008F20 to CHO cells transfected with the
human recombinant A3 adenosine receptors was performed per-
formed as previously described.56 Competition experiments were
carried out in duplicate in a final volume of 250 ll in test tubes

P. G. Baraldi et al. / Bioorg. Med. Chem. 16 (2008) 10281–10294
10293
Å mol as the convergence criterion. The temperature was set at
300 K and the time step was 1.0 fs in MD simulations.
As A3 receptor starting structure was used the one reported in
our previous paper (MODEL 1),48 compound 1a was docked inside
the receptor by means of AUTODOCK 3.0.47
The regions of interest used by AUTODOCK were defined by
considering residue N250 as the central group; in particular, a grid
of 40, 44, and 40 points in the x, y, and z directions was constructed
centred on the centre of the mass of N250. A grid spacing of 0.375 Å
and a distance-dependent function of the dielectric constant were
used for the energetic map calculations.
3D-QSAR study was developed aligning the ligands on the basis
of their central scaffold. The GOLPE program52 was used to define
the 3D-QSAR models, using GRID interaction fields63 as descriptors
(see below). Given the low number of ligands (34 compounds) the
training set was composed by all the compounds and the 3D-QSAR
model was validated through the correlation coefficient (r2), pre-
dictive correlation coefficient (q2) and cross-validated standard
deviation of errors of prediction (SDEPCV), did not taking into ac-
count any external test set. Inactive compounds were also included
in the training set, to which was arbitrarily assigned a Ki value
equal to 1000 nM. The GRID program63 was used to describe the
previously superimposed molecular structure. Interaction energies
between selected probes and each molecule were calculated using
a grid spacing of 1 Å. In order to evaluate the lipophilic, H bond
acceptor and donor properties of the ligands the C3 (methyl), O
(carbonylic oxygen) and N2@ (sp2 amine cation) probes were used.
The MIFs of the training set were imported into GOLPE, variable
selection was operated by zeroing values with absolute values
smaller than 0.06 kcal/mol and removing variables with a standard
deviation below 0.1. Moreover, variables which exhibited only two
values and had a skewed distribution were also removed.
AUTODOCK TOOLS62 was used to identify the torsion angles in
the ligands, add the solvent model and assign partial atomic
charges (Gasteiger for the ligands and Kollman for the receptors).
Using the Lamarckian Genetic Algorithm, the compounds were
subjected to 100 runs of the AUTODOCK search, in which the de-
fault values of the other parameters were used.
Cluster analysis was performed on the docked results using an
RMS tolerance of 1.0 Å and the best docked structure was taken
into consideration for further analysis.
The A3–1a complex was refined using the previously reported
procedure.48 It mainly consisted of a total of 400 ps of MD in which
all the alpha carbons of the TM, and the main ligand-receptor inter-
actions were constrained during the trajectory by means of
decreasing the force constants. More in detail, an initial restraint
with a force constant of 10 Kcal/mol Å2 was applied on the alpha
carbons. This force constant decreased during the whole MD and
in the last 200 ps a value of 0.1 kcal/mol Å2 was applied. As regards
the H bond ligand-receptor interactions, a restraint of 50 kcal/
mol Å2 was applied on the H bond between N250 and the carbonyl
of the ligand. At the end of the MD simulation, three steps of min-
imization were applied on the average structure obtained during
the last 100 ps of the MD run. During these three steps a restraint
of 0.1 kcal/mol Å2 was applied on the alpha carbons, while with re-
gard to the ligand-N250H bond, in the first two steps a restraint of
respectively 25 and 10 Kcal/mol Å2 was applied and in the last one
the restraint was removed.
The smart region definition (SRD) algorithm64 was applied with
10% of the active variables as the number of seed (selected in the
PLS weights space), a critical distance cutoff of 2.5 Å, and a collaps-
ing distance cutoff of 4.0 Å. The groups were then used in the Frac-
tional Factorial Design (FFD) procedure. FFD selection was applied,
until the r2 and q2 values did not increase significantly, using the
cross-validation routine with five random sets of compounds.
Acknowledgements
We thank Prof. Karl-Norbert Klotz for cDNA encoding the hu-
man adenosine receptors and Prof. Gabriele Cruciani and Prof. Ser-
gio Clementi (Molecular Discovery and MIA srl) for the use of the
GOLPE program in their chemometric laboratory (University of
Perugia, Italy) and for having provided the GRID program.
Supplementary data
4.4. Docking studies
Supplementary data associated with this article can be found, in
the online version, at doi:10.1016/j.bmc.2008.10.049.
The ligands were built through Maestro program and were then
subjected to a conformational search of 500 steps in a water envi-
ronment (using the Generalized-Born/Surface-Area model) by
means of Macromodel. The algorithm used was the Montecarlo
method with the MMFFs forcefield and a distance-dependent
dielectric constant of 1.0. The ligands were then minimized using
the conjugated gradient method until a convergence value of
0.05 kcal/Å mol, using the same forcefield and parameters as for
the conformational search.
Then the so minimized ligands were docked in the A3 receptor
by means of AUTODOCK 3.047 using the procedure described
above.
Cluster analysis was performed on the docked results using an
RMS tolerance of 1.0 Å and the best docked structure was taken
into consideration, furthermore for most of the ligands the first
cluster corresponded also to the most populated one.
References and notes
1. Fredholm, B. B.; Arslan, G.; Halldner, L.; Kull, B.; Schulte, G.; Wasserman, W.
Naunyn-Schmied. Arch. Pharm. 2000, 362, 364–374.
2. Williams, M.; Jarvis, M. F. Biochem. Pharmacol. 2000, 59, 1173.
3. Fredholm, B. B.; Ijzerman, A. P.; Jacobson, K. A.; Klotz, K-N.; Linden, J. Pharm.
Rev. 2001, 53, 527.
4. Baharav, E.; Bar-Yehuda, S.; Madi, L.; Silberman, D.; Rath-Wolfson, L.; Halpren,
M.; Ochaion, A.; Weinberger, A.; Fishman, P. J. Rheumatol. 2005, 32, 469.
5. Jacobson, K. A.; Kim, H. O.; Siddiqui, S. M.; Olah, M. E.; Stiles, G. L.; Von Lubitz,
D. K. J. E. Drugs Future 1995, 20, 689.
6. Kohno, Y.; Ji, X. D.; Mawhorter, S. D.; Koshiba, M.; Jacobson, K. A. Blood 1996, 88,
3569.
7. Jacobson, K. A. Trends. Pharmacol. Sci. 1998, 19, 184.
8. Von Lubitz, D. K. J. E.; Lin, R. C. S.; Popik, P.; Carter, M. F.; Jacobson, K. A. Eur. J.
Pharmacol. 1994, 263, 59.
9. Rivo, J.; Zeira, E.; Galun, E.; Matot, I. Anesthesiology 2004, 101, 1153.
10. Brown, R. A.; Spina, D.; Page, C. P. Br. J. Pharmacol. 2008, 153, S446.
11. Marx, D.; Ezeamuzie, C. I.; Nieber, K.; Szelenyi, I. Drug News Perspect. 2001, 14,
89.
4.5. 3D-QSAR study
12. Okamura, T.; Kurogi, Y.; Hashimoto, K.; Sato, S.; Nishikawa, H.; Kiryu, K.;
Nagao, Y. Bioorg. Med. Chem. Lett. 2004, 14, 3775.
13. Schlotzer-Schrehardt, U.; Zenkel, M.; Decking, U.; Haubs, D.; Kruse, F. E.;
Junemann, A.; Coca-Prados, M.; Naumann, G. O. Invest. Ophthalmol. Vis. Sci.
2005, 46, 2023.
14. Avila, M. Y.; Stone, R. A.; Civan, M. M. Invest. Ophthalmol. Vis. Sci. 2002, 43, 3021.
15. Gessi, S.; Varani, K.; Merighi, S.; Cattabriga, E.; Avitabile, A.; Gavioli, R.; Fortini,
C.; Leung, E.; Mac Lennan, S.; Borea, P. A. Mol. Pharmacol. 2004, 65, 711.
16. Ohana, G.; Bar-yehuda, S.; Barer, F.; Fishman, P. J. Cell Phys. 2001, 186, 19.
17. Baraldi, P. G.; Borea, P. A. Trends Pharm. Sci. 2000, 21, 456.
Before construction of the 3D-QSAR model, the binding free en-
ergy calculated for each ligand-AT1 complex by means of the
AUTODOCK scoring function was correlated with the experimental
antagonist affinity, but the quadratic correlation (R2) showed a low
value (R2 < 0.4).
For each ligand, the best docked structure was chosen and this
receptor-based alignment was used for further studies. A second

10294
P. G. Baraldi et al. / Bioorg. Med. Chem. 16 (2008) 10281–10294
18. Merighi, S.; Mirandola, P.; Varani, K.; Gessi, S.; Leung, E.; Baraldi, P. G.; Tabrizi,
M. A.; Borea, P. A. Pharmacol. Ther. 2003, 1001, 31.
19. Fishman, P.; Bar-Yehuda, S.; Barer, F.; Ohana, G. Drug. Dev. Res. 2000, 50,
101.
20. Madi, L.; Ochaion, A.; Rath-Wolfson, L.; Bar-Yehuda, S.; Erlanger, A.; Ohana, G.;
Harish, A.; Merimski, O.; Barer, F.; Fishman, P. Clin. Cancer Res. 2004, 10, 4472.
21. Gessi, S.; Varani, K.; Merighi, S.; Morelli, A.; Ferrari, D.; Leung, E.; Baraldi, P. G.;
Spalluto, G.; Borea, P. A. Br. J. Pharmacol. 2001, 134, 116.
22. Merighi, S.; Varani, K.; Gessi, S.; Cattabriga, E.; Iannotta, V.; Uloglou, C.; Leung,
E.; Borea, P. A. Br. J. Pharmacol. 2001, 134, 1215.
23. Borea, P. A.; Baraldi, P. G.; Chen, S.F.; Leung, E. PCT Int. Appl. 2003, WO
2004000224.
24. Gessi, S.; Cattabriga, E.; Avitabile, A.; Gafa’, R.; Lanza, G.; Cavazzini, L.; Bianchi,
N.; Gambari, Roberto.; Feo, Carlo.; Liboni, A.; Gullini, S.; Leung, E.; Mac-Lennan,
S.; Borea, P. A. Clin. Canc. Res. 2004, 10, 5895.
25. Baraldi, P. G.; Tabrizi, M. A.; Fruttarolo, F.; Bovero, A.; Avitabile, B.; Preti, D.;
Romagnoli, R.; Merighi, S.; Gessi, S.; Varani, K.; Borea, P. A. Drug Dev. Res. 2003,
58, 315.
26. Baraldi, P. G.; Cacciari, B.; Romagnoli, R.; Merighi, S.; Varani, K.; Borea, P. A.;
Spalluto, G. Med. Res. Rev. 2000, 20, 103.
39. Jeong, L. S.; Lee, H. W.; Kim, H. O.; Tosh, D. K.; Pal, S.; Choi, W. J.; Gao, Z. G.;
Patel, A. R.; Williams, W.; Jacobson, K. A.; Kim, H. D. Bioorg. Med. Chem. Lett.
2008, 18, 1612.
40. Drabczyn´ ska, A.; Schumacher, B.; Müller, C. E.; Karolak-Wojciechowska, J.;
Michalak, B.; Pe˛kala, E.; Kiec´-Kononowicz, K. Eur. J. Med. Chem. 2003, 38, 397.
41. Baraldi, P. G.; Abdel, Z. N.; Fruttarolo, F.; Tabrizi, A. M.; Nuñez, M.; Spalluto, G.;
Romagnoli, R. Synthesis 2001, 5, 773.
42. Baraldi, P. G.; Preti, D.; Tabrizi, M. A.; Fruttarolo, F.; Romagnoli, R.; Zaid, N. A.;
Moorman, A. R.; Merighi, S.; Varani, K.; Borea, P. A. J. Med. Chem. 2005, 48, 4697.
43. Kramer, G. L.; Garst, J. E.; Mitchel, S. S.; Wells, J. N. Biochemistry 1977, 16, 3316.
44. Holschbach, M. H.; Fein, T.; Krummeich, C.; Lewis, R. G.; Wutz, W.; Schwabe,
U.; Unterlugauer, D.; Olsson, R. A. J. Med. Chem. 1998, 41, 555.
45. Papesch, V.; Schroeder, E. F. J. Org. Chem. 1951, 16, 1879.
46. Mosselhi, A.; Pfleidered, W. Purines. XIV. J. Heterocycl. Chem. 1993, 30, 1221.
47. Morris, G. M.; Goodsell, D. S.; Halliday, R. S.; Huey, R.; Hart, W. E.; Belew, R. K.;
Olson, A. J. J. Comp. Chem. 1998, 19, 1639.
48. Tafi, A.; Bernardini, C.; Botta, M.; Corelli, F.; Andreini, M.; Martinelli, A.; Ortore,
G.; Baraldi, P. G.; Fruttarolo, F.; Borea, P. A.; Tuccinardi, T. J. Med. Chem. 2006,
49, 4085.
49. Gao, Z. G.; Kim, S. K.; Biadatti, T.; Chen, W.; Lee, K.; Barak, D.; Kim, S. G.;
Johnson, C. R.; Jacobson, K. A. J. Med. Chem. 2002, 45, 4471.
27. Muller, C. E. Curr. Top. Med. Chem. 2003, 3, 445.
28. Okamura, T.; Kurogi, Y.; Nishikawa, H.; Hashimoto, K.; Fujiwara, H.; Nagao, Y. J.
Med. Chem. 2002, 45, 3703.
50. Gao, Z. G.; Chen, A.; Barak, D.; Kim, S. K.; Muller, C. E.; Jacobson, K. A. J. Biol.
Chem. 2002, 277, 19056.
29. Colotta, V.; Catarzi, D.; Varano, F.; Cecchi, L.; Filacchioni, G.; Martini, C.;
Trincavelli, L.; Lucacchini, A. J. Med. Chem. 2000, 43, 1158.
30. Colotta, V.; Catarzi, D.; Varano, F.; Calabri, F. R.; Lenzi, O.; Filacchioni, G.;
Martini, C.; Trincavelli, L.; Deflorian, F.; Moro, S. J. Med. Chem. 2004, 47, 3580.
31. Jung, K. Y.; Kim, S. K.; Gao, Z. G.; Gross, A. S.; Melman, N.; Jacobson, K. A.; Kim,
Y. C. Bioorg. Med. Chem. 2004, 12, 613.
32. Press, N. J.; Keller, T. H.; Tranter, P.; Beer, D.; Jones, K.; Faessler, A.; Heng, R.;
Lewis, C.; Howe, T.; Gedeck, P.; Mazzoni, L.; Fozard, J. R. Curr. Top. Med. Chem.
2004, 4, 863.
33. Muller, C. E.; Thorand, M.; Qurishi, R.; Diekmann, M.; Jacobson, K. A.; Padgett,
W. L.; Daly, J. W. J. Med. Chem. 2002, 45, 3440.
34. Priego, E. M.; von Frijtag Drabbe Kuenzel, J.; Ijzerman, A. P.; Camarasa, M. J.;
Perez-Perez, M. J. J. Med. Chem. 2002, 45, 3337.
51. Manetti, F.; Schenone, S.; Bondavalli, F.; Brullo, C.; Bruno, O.; Ranise, A.; Mosti,
L.; Menozzi, G.; Fossa, P.; Trincavelli, M. L.; Martini, C.; Martinelli, A.; Tintori, C.;
Botta, M. J. Med. Chem. 2005, 48, 7172.
52. GOLPE 4.5. Multivariate Infometric Analysis Srl., Viale dei Castagni 16, Perugia,
Italy, 1999.
53. Klotz, K. N.; Hessling, J.; Hegler, J.; Owman, C.; Kull, B.; Fredholm, B. B.; Lohse,
M. J. Naunyn-Schmied. Arch. Pharm. 1998, 357, 1.
54. Borea, P. A.; Dalpiaz, A.; Varani, K.; Gessi, S.; Gilli, G. Life Sci. 1996, 59, 1373.
55. Borea, P. A.; Dalpiaz, A.; Varani, K.; Gessi, S.; Gilli, G. Biochem. Pharmacol. 1995,
49, 461.
56. Varani, K.; Merighi, S.; Gessi, S.; Klotz, K. N.; Leung, E.; Baraldi, P. G.; Cacciari,
B.; Romagnoli, R.; Spalluto, G.; Borea, P. A. Mol. Pharmacol. 2000, 57, 968.
57. Bradford, M. M. Anal. Biochem. 1976, 72, 248.
35. Colotta, V.; Catarzi, D.; Varano, F.; Capelli, F.; Lenzi, O.; Filacchioni, G.; Martini,
C.; Trincavelli, L.; Ciampi, O.; Pugliese, A. M.; Pedata, F.; Schiesaro, A.; Morizzo,
E.; Moro, S. J. Med. Chem. 2007, 50, 4061.
36. Baraldi, P. G.; Tabrizi, M. A.; Preti, D.; Bovero, A.; Fruttarolo, F.; Romagnoli,
R.; Zaid, N. A.; Moorman, A. R.; Varani, K.; Borea, P. A. J. Med. Chem. 2005,
48, 5001.
37. Gao, Z. G.; Joshi, B. V.; Klutz, A. M.; Kim, S. K.; Lee, H. W.; Kim, H. O.; Jeong, L. S.;
Jacobson, K. A. Bioorg. Med. Chem. Lett. 2006, 16, 596.
38. Jeong, L. S.; Choe, S.; Gunaga, P.; Kim, H. O.; Lee, H. W.; Lee, S. K.; Tosh, D. K.;
Patel, A.; Palaniappan, K. K.; Gao, Z. G.; Jacobson, K. A.; Moon, H. R. J. Med. Chem.
2007, 50, 3159.
58. Cheng, Y. C.; Prusoff, W. H. Biochem. Pharmacol. 1973, 22, 3099.
59. Munson, P. J.; Rodbard, D. Anal. Biochem. 1980, 107, 220.
60. Macromodel, ver. 8.5; Schrödinger Inc.: Portland, OR, 1999.
61. Pearlman, D. A.; Case, D.; Caldwell, J. W.; Ross, W. S.; Cheatham TE, I. I. I.;
Ferguson, D. M.; Seibel, G. L.; Chandra Singh, U.; Weiner, P. K.; Kollman, P. A.
AMBER 41; University of California, San Francisco: San Francisco, 1995.
62. http://www.scripps.edu/~sanner/python/adt.
63. Cruciani, G; Goodford, P. Copyright Molecular Discovery Ltd. 2001–2003
GREATER graphical interface for GRID version, 1.1.7 GRIB, UPF/IMIM, Barcelona
(Spain). Available from: http://www.moldiscovery.com.
64. Pastor, M.; Cruciani, G.; Clementi, S. J. Med. Chem. 1997, 40, 1455.