New substituted 9-propyladenine derivatives as A2A adenosine receptor antagonists

Article abstract of DOI:10.1039/c5md00034c

A new series of 9-propyladenines bearing a phenylalkylamino group in the 2-position or a phenylalkyl chain in the N⁶-position, and further substituted with a bromine atom or a 2-furyl ring in the 8-position, were synthesized and tested at human

Full text of DOI:10.1039/c5md00034c

MedChemComm
View Article Online
View Journal | View Issue
CONCISE ARTICLE
New substituted 9-propyladenine derivatives as
A2A adenosine receptor antagonists†
Cite this: Med. Chem. Commun.,
2015, 6, 963
C. Lambertucci,a M. Buccioni,a D. Dal Ben,a S. Kachler,b G. Marucci,a A. Spinaci,a
a
b
a
*
A. Thomas, K.-N. Klotz and R. Volpini
A new series of 9-propyladenines bearing a phenylalkylamino group in the 2-position or a phenylalkyl chain
in the N6-position, and further substituted with a bromine atom or a 2-furyl ring in the 8-position, were
synthesized and tested at human adenosine receptors. The novel compounds proved to be A2A adenosine
receptor antagonists and some of them showed high A2A affinity, but moderate selectivity (18: KiA2A = 6.6
nM). Molecular modeling studies gave some explanation of the different activities of the compounds, giving
suggestions for the synthesis of new A2A adenosine receptor antagonists.
Received 24th January 2015,
Accepted 1st April 2015
DOI: 10.1039/c5md00034c
www.rsc.org/medchemcomm
last subtype to be pharmacologically characterized due to the
lack of selective ligands.7 In addition to a role in the regula-
1. Introduction
The natural nucleoside adenosine (Ado) is the basic constitu-
ent of RNA and ATP and, besides that, it interacts with spe-
cific receptors belonging to the P1 purinergic receptor family,
classified as A1, A2A, A2B, and A3 subtypes, which are coupled
to Gi (A1 and A3) and Gs (A2A and A2B) protein signalling.1
Adenosine receptors (ARs) are widely distributed in the body,
both in the central nervous system (CNS) and the periphery,
and regulate many physiological and pathological functions.2
In fact, there is strong evidence that Ado, through the interac-
tion with its receptors, has a functional role in many diseases
ranging from cardiovascular and immune disorders, renal
failure and inflammatory conditions to neurological, psychiat-
ric, and neurodegenerative disorders.3 In the last years, many
efforts have been devoted to the synthesis of potent and selec-
tive ligands of ARs, and besides that, actually only the A2A
selective agonist regadenoson has been approved by the Food
and Drug Administration (FDA) and commercialized as
Lexiscan (Astellas Pharma Company) for myocardial perfusion
imaging.4 On the other hand, the A2A selective antagonist
istradefylline (Nouriast, Kyowa Hakko Kirin Co., Ltd), a novel
antiparkinsonian agent, has been approved for manufactur-
ing and marketing at the beginning of 2013 only in Japan.5,6
Hence, the search for potent and selective ligands of ARs is
still a challenge especially for the A2B receptor, which is the
tion of mast cell secretion and intestinal functions, the activa-
tion of A2B receptors seems to increase fibroblast and endo-
thelial cell migration and to contribute to tissue protection.8,9
On the contrary, selective antagonists of the A2B receptor
could be very useful in the treatment of asthma and allergic
diseases since they prevent mast cell degranulation.10
In many papers, we have demonstrated that the introduc-
tion of diverse substituents in the 2, N6, 8, and 9-positions of
adenine led to AR antagonists endowed with different affini-
ties and selectivities for the four AR subtypes.11–16 In particu-
lar, at human receptors, the 9-ethyladenine (1; Fig. 1)
resulted in an AR antagonist endowed with μM affinity at the
A1 and A2A subtypes while it was not able to bind the A2B and
A3 receptors at concentrations up to 100 μM (1: KiA1 = 7400
nM, KiA2A = 2200 nM).10
Introduction of a bromine atom in the 8-position of 1 led
to 8-bromo-9-ethyladenine (2: Fig. 1), which showed
enhanced affinity at all receptor subtypes and resulted in an
A2A receptor antagonist endowed with moderate selectivity,
a School of Pharmacy, Medicinal Chemistry Unit, University of Camerino, Via S.
Agostino 1, 62032 Camerino, Italy. E-mail: rosaria.volpini@unicam.it
b Universität Würzburg, Institut für Pharmakologie und Toxikologie, Versbacher
Str. 9, 97078, Würzburg, Germany
† Electronic supplementary information (ESI) available: detailed description of
the synthesis and characterization of compounds 12–20, 22–31, and 33;
biological assay protocols; computational experiments. See DOI: 10.1039/
c5md00034c
Fig. 1 Known adenine derivatives as antagonists of ARs.
Med. Chem. Commun., 2015, 6, 963–970 | 963
This journal is © The Royal Society of Chemistry 2015

View Article Online
Concise Article
MedChemComm
especially versus the A1 subtype.17 The substitution of the
bromine atom of 2 with a furyl ring gave a further enhance-
ment of affinity at all ARs, leading again to an A2A antagonist
with still moderate selectivity (3, Fig. 1; KiA2A = 3.7 nM, select.
A1/A2A = 6).15 On the other hand, when a hindered phenethyl-
amino or a phenethoxy group was introduced in the 8-posi-
tion of 1, the resulted compounds maintained the A2A selec-
tivity, but were found less active than 2 and 3 at ARs.11 In the
process of investigating substitutions at other positions of
the purine core, it was found that the presence of a phen-
ethylamino group in the 2-position of 1 favored the interac-
tion of the resulted compound with all ARs IJ9-ethyl-2-
phenethylaminoadenine, 4, Fig. 1; KiA1 = 330 nM, KiA2A = 150
nM, KiA2B = 2410 nM, and KiA3 = 3200 nM) while the pres-
ence of a phenethyl chain in the N6-position of 1 gave a com-
pound endowed with enhanced affinity only at the A2B and
A3 subtypes IJ9-ethyl-N6-phenethyladenine, 5: KiA1 = 8730 nM,
KiA2A = 1340 nM, KiA2B = 8130 nM, and KiA3 = 10 800 nM).
For all these molecules, the further introduction of a bro-
mine atom in the 8-position enhanced the affinity at all ARs.
In addition, in a recent paper, we reported that the replace-
ment of the 9-ethyl group of compound 1 with a propyl chain
favors the interaction of the resulted compound with the A2B
receptors (6: KiA1 = 9400 nM, KiA2A = 9600 nM, KiA2B = 1700
nM, and KiA3 > 100 μM, Fig. 1, Table 1).18
Even in this case, the introduction of a bromine atom or a
furyl ring in the 8-position enhanced the affinity of the corre-
sponding compounds 7 and 8 at all ARs. It is worth noting
that the 8-bromo-9-propyladenine showed a slight preference
for the A2B receptor subtype (7: KiA1 = 1100 nM, KiA2A = 300
nM, KiA2B = 200 nM, and KiA3 ≥ 100 μM).18
On these bases, in the search for novel AR antagonists
and aimed at investigating the influence of the introduction
of a phenylethylamino group in the 2-position or the corre-
sponding phenylethyl chain in the N6-position combined
with a propyl group in the 9-position of adenine, the synthe-
sis of compounds 13 and 24 was undertaken (Fig. 2,
Schemes 1 and 2). Furthermore, in order to find the appro-
priate length of the alkyl chain between the phenyl and the
amino group, the ethylene chain was substituted with a
methylene or a propylene group (compounds 12, 14, 23, and
25). In addition, compounds 12–14 and 23–25 were bromi-
nated to afford compounds 15–17 and 26–28, respectively,
which were further converted to the corresponding 8-IJ2-furyl)
derivatives 18–20 and 29–31 (Fig. 2, Schemes 1 and 2) as
these modifications enhance the affinity of substituted ade-
nine derivatives at ARs.11,15,17 The new compounds 12–20,
23–31 were tested in a binding assay at human recombinant
A1, A2A, and A3 ARs and in the functional cAMP assay at
human recombinant A2B ARs. Furthermore, molecular
Table 1 Biological profile of synthesized compounds 12–20 and 23–31 at human adenosine receptors stably transfected in CHO cells
a
b
c
d
Cpd
R1
R2
X
KiA1
KiA2A
KiA2B
KiA3
6
7
8
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
H
Br
2-Furyl
H
Br
2-Furyl
H
Br
2-Furyl
H
Br
2-Furyl
H
Br
2-Furyl
H
Br
2-Furyl
H
Br
9400 (7900–11 000)
1100 (1000–1300)
19 (13.2–27.4)
7600 (5270–11 000)
358 (315–408)
138 (98.8–191)
2330 (1840–2950)
373 (289–483)
9600 (5800–16 000)
300 (260–350)
18 (8.25–39.4)
1400 (931–2120)
36 (30.7–41.4)
6.6 (4.19–10.5)
738 (609–893)
1700 (790–3800)
200 (110–380)
252 (167–379)
>30 000
1740 (1154–2630)
4610 (3910–5430)
9250 (6800–12 600)
1490 (1230–1810)
2030 (1560–2640)
>30 000
6300 (5390–7350)
8240 (6700–10 100)
>30 000
7490 (6820–8240)
1760 (1190–2620)
>30 000
>100 000
>100 000
790 (370–1700)
5840 (4840–7060)
650 (471–907)
12
15
18
13
16
19
14
17
20
23
26
29
24
27
30
25
28
31
NHCH2Ph
NHCH2Ph
NHCH2Ph
NHIJCH2)2Ph
NHIJCH2)2Ph
NHIJCH2)2Ph
NHIJCH2)3Ph
NHIJCH2)3Ph
29 (20.7–39.4)
5880 (5480–6310)
1160 (932–1450)
56 (41.7–74.7)
12 600 (11 200–14 100)
1950 (1700–2240)
266 (240–294)
29 100 (27 400–30 800)
11 100 (8010–15 400)
402 (259–625)
16 900 (16 300–17 600)
2520 (1710–3730)
811 (552–1190)
27 (21.2–34.6)
14 (11.0–17.7)
80 (61.6–105)
3640 (3440–3850)
583 (445–765)
9110 (7870–10 600)
284 (160–502)
113 (87.2–146)
22 600 (13 400–38 100)
1180 (881–1890)
80 (52.7–121)
12 000 (7270–19 900)
503 (373–680)
547 (491–610)
NHIJCH2)3Ph
230 (202–261)
H
H
H
H
H
H
H
H
H
CH2Ph
CH2Ph
CH2Ph
IJCH2)2Ph
IJCH2)2Ph
IJCH2)2Ph
IJCH2)3Ph
IJCH2)3Ph
IJCH2)3Ph
42 800 (38 900–47 100)
5620 (5060–6230)
546 (459–649)
19 200 (15 900–23 200)
1940 (1570–2410)
2440 (1760–3390)
18 600 (15 200–22 700)
2670 (2070–3460)
378 (349–408)
3190 (1890–5400)
9280 (7170–12 000)
> 30 000
1760 (1170–2640)
1870 (986–3550)
13 000 (11 900–14 200)
1020 (872–1190)
181 (137–239)
30 000 (24 200–37 400)
12 200 (8600–17 200)
959 (812–1130)
2-Furyl
a
b
Displacement of specific [3H]CCPA binding at human A1R expressed in CHO cells (n = 3−6). Displacement of specific [3H]NECA binding at
c
human A2AR expressed in CHO cells. Ki values of the inhibition of NECA-stimulated adenylyl cyclase activity in CHO cells expressing human
A2B receptors (Ki values were calculated from the corresponding –>IC50 values). Displacement of specific [3H]HEMADO binding at human
d
A3R expressed in CHO cells. Data are expressed as geometric means, with 95% confidence limits.
964 | Med. Chem. Commun., 2015, 6, 963–970
This journal is © The Royal Society of Chemistry 2015

View Article Online
MedChemComm
Concise Article
Fig. 2 Structure of designed and synthesized compounds.
Scheme
2
Synthesis of 6-phenylalkyladenine derivatives 23–31.
Reaction conditions: a) Pr-I, K2CO3, DMF, rt, 14 h; b) R-NH2, 70 °C, 12
h; c) NBS, DMF, rt, 16 h; d) IJ2-tributylstannyl)furan, IJPh3P)2PdCl2, THF,
60 °C, 18 h.
the 6-amino derivative 11.13 Reaction of 11 with the suitable
amine at 120 °C for 24 h furnished the 2-phenylalkylamino-9-
propyl adenines 12–14 with yields ranging from 45 to 88%.
Treatment of 12–14 with N-bromosuccinimide (NBS), in DMF
at room temperature, gave the 8-bromoderivatives 15–17 with
yields from 35 to 50%.
Replacement of the 8-bromine atom with a 2-furyl ring
was obtained by reaction of compounds 15–17 with
IJ2-tributylstannyl)furan using triphenylphosphine palladiumIJII)
chloride as catalyst and anhydrous THF as solvent (Scheme 1).
After 3 h at 60 °C the reactions were complete; the
8-IJ2-furyl) derivatives 18–20 were obtained in 55–70% yield.
Compounds 23–31 were synthesized by reacting commercially
available 6-chloropurine with propyliodide in anhydrous
DMF in the presence of potassium carbonate (Scheme 2). The
reaction was previously reported using propylbromide and in
that case only the N9 isomer was described. In the present
case also the N7 isomer was detected and described.19 The
reaction was conducted at room temperature and yielded the
N9, 22, and N7, 22a, isomers (64 and 28%, respectively) after
column chromatography. The two isomers were characterized
Scheme
1 Synthesis of 2-phenylalkylamino derivatives 12–20.
Reaction conditions: a) Pr-I, K2CO3, DMF, rt, 16 h; b) NH3, rt, 16 h; c)
R-NH2, 120 °C, 24 h; d) NBS, DMF, rt, 1 min; e) IJ2-tributylstannyl)-furan,
IJPh3P)2PdCl2, THF, 60 °C, 3 h.
modelling studies have been performed in order to explain
the activity of the newly synthesized adenine derivatives.
by
a comparison of NMR data with analogue purine
2. Results and discussion
2.1 Chemistry
Compounds 12–20 and 23–31 were synthesized as summa-
rized in Schemes 1, 2, and 3.
The synthesis of 2-alkylamino-9-propyl derivatives 12–20 was
realized by reaction of commercially available 2,6-dichloropurine
with propyliodide in the presence of potassium carbonate as
previously reported (Scheme 1).13 After separation of the
formed N9 and N7 isomers (10 and 10a, respectively) by chro-
matography, compound 10 was reacted with ammonia to give
Scheme
3 Synthesis of 8-bromo-6-phenylpropyladenine (28).
Reaction conditions: a) i-C5H11ONO, CH2I2, CH3CN, 85 °C, 30 min; b)
R-NH2, 70 °C, 12 h.
This journal is © The Royal Society of Chemistry 2015
Med. Chem. Commun., 2015, 6, 963–970 | 965

View Article Online
Concise Article
MedChemComm
derivatives.13,17 In fact, the N-9 alkylated adenines showed an
upfield shift of the H-8 and NH2 protons with respect to the
same protons of the N-7 isomers. Furthermore, the N-9 iso-
mers are always obtained in higher yields and show lower
polarity in the TLC run such as in compounds 22 with
respect to 22a. Reaction of 22 with the suitable amine, in the
presence of 4-IJdimethylamino)pyridine (DMAP) in anhydrous
acetonitrile at 70 °C for 12 h, furnished the 6-phenylalkyl-9-
propyladenine derivatives 23–25 with yields ranging from 58
to 92% (Scheme 2). Treatment of the latter compounds with
NBS in DMF at room temperature gave 26 and 27 with yields
of 19 and 38%, respectively, while only a few milligrams of
the impure compound 28 were obtained. Hence, a different
synthetic pathway has been chosen to obtain this compound,
as depicted in Scheme 3. To this purpose, the 8-bromo-9-
propyladenine (32)17 was treated in Sandmeyer's conditions
with isoamyl nitrite and diiodomethane at 60 °C to give the
8-bromo-6-iodo-9-propyl-adenine (33) in 54% yield. Com-
pound 33 was then reacted with 3-phenylpropylamine at 60
°C for 4 h to give 28 in 53% yield.
compound 9-propyladenine (6: KiA2B = 1700 nM) and, obvi-
ously, than the corresponding 8-bromo and 8-furyl derivatives
7 IJKiA2B = 200 nM) and 8 IJKiA2B = 250 nM). Moreover, they
were inactive when unsubstituted in the 8-position (see com-
pounds 12–14 and 24, 25: KiA2B > 30 μM), with the exception
of the 2-phenethylamino-9-propyladenine (13) which
displayed a KiA2B of 9250 nM. On the other hand, the corre-
sponding derivatives, bearing a bromine atom or a furyl ring
in the 8-position (compounds 15–20 and 26–31), were able to
activate the A2B receptor subtype but at concentration in the
μM range, so resulting in all cases less active than the
8-bromo-9-propyladenine (7: KiA2B = 200 nM) and the 8-furyl-
9-propyladenine (8: KiA2B = 250 nM). These data clearly indi-
cate that the combination of a propyl chain in the 9-position
of adenine with a phenylalkylamine in the 2-position or a
phenylalkyl chain in the N6-position does not favor the inter-
action of the corresponding derivatives with A2B receptors.
In addition, the presence of a phenylalkyl chain in the N6-
position of 6 led to the same results also at A1 and A2A AR
subtypes; in fact, compounds 23–25 were found less active
than 6 at these receptors, as well as the corresponding
8-bromo and 8-furyl derivatives 26–28 and 29–31, respectively,
compared to the 8-bromo-9-propyladenine (7: KiA1 = 1100 nM
and KiA2A = 300 nM) and the 8-furyl-9-propyladenine (8: KiA1 =
19 nM and KiA2A = 18 nM). However, the N6-benzyl-8-furyl-9-
propyladenine (29) is the most active ligand, among this
series, at the A2A receptor with Ki = 80 nM. Conversely, the
introduction of the phenylalkyl chain in the N6-position of 6
seems to favor the binding at A3 receptors (23: KiA3 = 29 100
nM; 24: KiA3 = 16 900 nM; and 25: KiA3 = 30 000 nM vs.
6: KiA3 > 100 μM). As expected, the further introduction of a
bromine atom or a furyl ring in the 8-position gave an
increase in A3 affinity; the 8-furyl derivatives 29–31 are the
most active in the series of the N6-substituted derivatives.
Also at the A3 receptor subtype compound 29 showed the bet-
ter affinity profile of this series, with a Ki slightly lower than
8 (29: KiA3 = 402 nM vs. 8: KiA3 = 790 nM). The best results
were obtained with the series of the 2-substituted derivatives.
In fact, compounds 12–20 were found in general to be more
active at A1, A2A and A3 ARs than the corresponding
2-unsubstituted derivatives 6–8. An exception is represented
by the 8-furyl derivatives 18–20 at A1 receptors (Ki values
ranging from 80 to 238 nM vs. 8: KiA1 = 19 nM) and 20 at the
The 8-bromoadenine derivatives 26–28 were used to pre-
pare the 8-furyladenine analogues 29–31 as previously
described, but for a longer time (Scheme 2).
In fact, by reaction of compounds 26–28 with
IJ2-tributylstannyl)furan,
in
the
presence
of
bisIJtriphenylphosphine) palladiumIJII) dichloride, in anhy-
drous THF at 60 °C for 18 h, the 8-furyladenine derivatives
29–31 were obtained.
2.2 Biological activity
Compounds 12–31 were tested in a radioligand binding assay
at human recombinant ARs, expressed in Chinese hamster
ovary (CHO) cells, in order to evaluate their affinity at A1, A2A
and A3 AR subtypes using as radioligands [3H]CCPA
IJ2-chloro-N6-cyclopentyl-adenosine), [3H]NECA
IJ5′-N-
,
ethylcarboxamidoadenosine), and [3H]HEMADO IJ2-hexyn-2-yl-
N6-methyladenosine), respectively.20,21 The results are
reported as Ki values nM, with 95% confidence intervals in
parentheses (Table 1). At A2B receptors, Ki values were calcu-
lated from IC50 values determined by inhibition of NECA-
stimulated adenylyl cyclase activity (Table 1).20 Previously, it
has been reported that the replacement of the ribose moiety
of adenosine nucleosides with alkyl groups led to compounds
which lose the intrinsic activity and behave as AR antago-
nists;12,22 for this reason the newly synthesized series of com-
pounds are believed to be antagonists of ARs and they have
been characterized only with the radioligand binding assay.
The 9-propyladenine (6), together with its 8-bromo and
8-furylderivatives 7 and 8, were reported as reference com-
pounds.18 Most of the newly synthesized derivatives displayed
Ki values ranging from μM to low nM concentrations at ARs
and all of them resulted in A2A antagonists without relevant
selectivity, especially versus the A1 and A3 AR subtypes. At A2B
receptors, the new substituted 9-propyladenine derivatives
12–31 showed lower affinity than the unsubstituted parent
A2A receptor subtype, which was less active than 8 (20: KiA2A =
113 nM vs. 8: KiA2A = 18 nM). It is worthwhile to note that
the 2-benzylamino-8-furyl-9-propyladenine was more active
than the 8-furyl-9-propyladenine at A2A and A3 receptors
(18: KiA2A = 6.6 nM and KiA3 = 29 nM vs. 8: KiA2A = 18 nM
and KiA3 = 790 nM) and behaved as the most active com-
pound among the two new series of 9-propyladenine deriva-
tives at the same receptor subtypes.
2.3 Molecular modelling analysis
On the basis of the biological results, we decided to perform
a molecular modelling analysis to better understand the
966 | Med. Chem. Commun., 2015, 6, 963–970
This journal is © The Royal Society of Chemistry 2015

View Article Online
MedChemComm
Concise Article
different interactions of the synthesised molecules with
A2AAR. In particular, molecular docking methods were
employed to simulate the interaction of compounds 12–20
and 23–31 with the binding site of the human A2AAR crystal
structures solved in the complex with ZM241385 antagonist
(Protein Data Bank -pdb- code: 3EML; 2.6 Å resolution23,24).
The crystal structure was re-modelled by firstly removing the
external T4L segment and secondly by performing a building
of missing receptor regions (the missing section of the extra-
cellular
2 -EL2- or intracellular 3 -IL3- domains). The
obtained A2AAR model was checked by using the Protein
Geometry Monitor application within the Molecular Operat-
ing Environment (MOE, version 2010.10) suite,25 which pro-
vides a variety of stereochemical measurements for structural
quality inspection in a given protein, such as backbone bond
lengths, angles and dihedrals, Ramachandran φ–ψ dihedral
plots, and sidechain rotamer and nonbonded contact quality.
The A2AAR structure was then used as a target for the
docking analysis of synthesised derivatives.
All ligand structures were optimized using RHF/AM1 semi-
empirical calculations and the MOPAC software package
implemented in MOE was utilized for these calculations.26
The compounds were then docked into the binding site of
the A2AAR model by using the MOE Dock tool. Docking poses
of each compound were subjected to energy minimization
and then rescored using three available methods
implemented in MOE: the London dG scoring function that
estimates the free energy of binding of the ligand from a
given pose; the Affinity dG scoring tool that estimates the
enthalpic contribution to the free energy of binding; the
dock-pKi predictor that uses the MOE scoring.svl script to esti-
mate for each ligand a pKi value, which is described by the
H-bond, transition metal interaction, and hydrophobic inter-
action energy. For each compound, the top-score docking
poses at the A2AAR model, according to at least two out of
three scoring functions, were selected for final ligand–target
interaction analysis. The docking methodology was firstly
tested by comparing the predicted binding mode of
ZM241385 within the A2AAR pocket with respect to the X-ray
data. The docking method showed good ability in reproduc-
ing the ZM241385 binding mode observed in the experimen-
tal data. In particular, the Root Mean Square Deviation
(RMSD) from the comparison of the crystal structure and the
docking conformation was calculated to be 1.45 Å.
The 2-substituted 9-propyladenine derivatives present
structural similarities with the ZM241385 compound and the
docking results show for these new molecules a general bind-
ing mode comparable to that of the A2AAR reference ligand
(observed at the 3EML crystal structure) and observable in
Fig. 3. In detail, the compound adenine scaffold is almost
vertically oriented and located between residues of the TM3
(Phe168) and TM6 (Leu249) domains, in a similar fashion to
the purine position and orientation observed from the bind-
ing mode of the nucleoside Ado within the hA2AAR crystal
structure (pdb code: 2YDO27) and analogously to other ade-
nine derivatives previously analysed at a hA2AAR model.15,18
Fig.
3
Binding modes at hA2AAR of the 2-substituted adenine
derivatives. The docking conformation of compound 18 (green) is
shown as obtained by the docking analysis (A) or as superimposed to
the original binding mode of the co-crystallized compound ZM241385
(cyan). (B) Some key residues are indicated within the figure. Polar
interactions (i.e. with N253) are indicated with red dashed lines.
The 6-amino group and the N7 nitrogen atom give a double
H-bond interaction with the TM6 Asn253 residue. The 8-posi-
tion pointing towards the receptor core is located between
residues of TM3 (Leu85, Thr88), TM5 (Met177), and TM6
(Trp246, His250, Leu249) segments, while the 9-propyl group
is located between residues of TM3 (Leu85), TM5 (Met177),
and TM6 (Trp246) domains. The oxygen atom of the furyl
ring inserted in the 8-position could provide an additional
H-bond interaction to Asn253, analogous to what is observed
for the hA2AAR–ZM241385 complex from the crystal structure
employed in this study. This data could explain the higher
affinity of the 8-furyl substituted analogues than the corre-
sponding 8-bromo or 8-unsubstituted derivatives. The 2-sub-
stituent points toward the extracellular space. Results of
binding studies report that the length of the alkyl spacer of
the 2-substituent provides higher affinity if it is among 1 and
2 carbon atoms, while a propyl spacer seems generally detri-
mental for the binding activity. The binding mode suggested
by the docking results presents the phenyl ring of the
This journal is © The Royal Society of Chemistry 2015
Med. Chem. Commun., 2015, 6, 963–970 | 967

View Article Online
Concise Article
MedChemComm
2-substituent as well inserted between EL2 and EL3 residues
only in the presence of a methyl or ethyl spacer within this
substituent, while a propyl linker makes the phenyl ring
more externally located and hence exposed to the solvent.
This could be the factor explaining the different compound
affinity based on the diverse length of the 2-group.
The 9-propyl chain is positioned in a sub-cavity given by
residues belonging to the TM3 (Leu85, Thr88), TM5
(Met177), and TM6 (Trp246, His250, Leu249) domains and
presenting mainly a hydrophobic profile. The 8-substituent
gets in proximity to residues of TM2 (Ala63), TM3 (Val84,
Leu85), TM6 (Trp246), and TM7 (Ile274, Ser277, His278).
The furyl ring inserted in this position seems to better fill
the sub-pocket given by the above cited residues with respect
to the bromine atom and a potential H-bond interaction
between the oxygen atom of the furyl ring and the hydroxyl
function of the Ser277 (TM6) sidechain is observable. In this
sense, docking results could explain why the compounds
with a 8-furyl substituent present in general a higher hA2AAR
affinity with respect to the 8-bromo analogues, considering at
least the compounds presenting a benzyl or a phenylpropyl
group in the N6-position. The N6-substituent points toward
the extracellular space and is located between residues
belonging to TM2 (Ala63, Ile66), EL2 (Leu167), and TM7
(Met270, Tyr271) residues.
Considering the N6-substituted 9-propyladenine deriva-
tives, docking results allowed to identify two families of bind-
ing modes. The first family (Fig. 4A), selected considering the
docking conformations with the highest scores from two out
of three scoring functions, presents the compound adenine
scaffold located between residues of the transmembrane
TM3 (Phe168) and TM6 (Leu249) domains. The purine moi-
ety is vertically oriented with the 8-position pointing towards
the receptor core and the 6-amino group positioned nearby
the TM2-TM3 amino acids. The N3 nitrogen atom is located
in proximity to the TM6 residues, providing H-bond interac-
tion with the polar hydrogen atoms of the Asn253 sidechain.
This substituent is hydrophobic and the interaction with
the nearby residues is nonpolar. The different length of the
spacer (methyl to propyl linker) has a significant impact on
the affinity for the A2AAR; in this sense, considering these
docking conformations, the different length of the linker
could have a modulating effect on the flexibility of the N6-
substituent and hence on its easy or difficult fitting within
the binding cavity.
The second family of binding modes contains conforma-
tions whose scores present the most similar trend compared
to the compound pKi values. These conformations are also
endowed with the highest scores according to one of the scor-
ing functions (London dG). This binding mode (Fig. 4) pre-
sents the compound adenine scaffold positioned and ori-
ented in a similar fashion than the docking conformations of
the 2-substituted derivatives, with the purine moiety located
between residues of the TM3 (Phe168) and TM6 (Leu249)
domains. This binding mode allows again the observation of
a double H-bond interaction between the 6-amino group and
the N7 nitrogen atom and the TM6 Asn253 residue, the 8-
and 9-substituents pointing toward the receptor core and
located between residues of TM3, TM5, and TM6 segments,
and the oxygen atom of the furyl ring inserted in the 8-posi-
tion providing an additional H-bond interaction to Asn253.
The N6-substituent points towards the extracellular space and
is positioned and oriented similarly to the first family of
docking conformations, even if more in proximity to the
TM6-EL3 residues. Taking together the results of the N6-
substituted derivatives, both the above described docking
conformations provide two possible binding modes that may
explain in different ways the higher activity of the 8-furyl
substituted derivatives, while the length of the N6-substituent
alkyl chain seems to modulate the affinity providing different
flexibility to the whole substituent.
Fig.
4
Binding modes at hA2AAR of the N6-substituted adenine
derivatives. Family 1 (A) and family 2 (B) docking conformations of
compound 29 are displayed. The ligand is coloured green. Some
residues are indicated within the figure. Polar interactions (i.e. with
N253) are indicated with red dashed lines.
A comparison of the docking conformations of the 2- and
N6-substituted derivatives was made by superimposition
within the MOE interface (Fig. 5). The first family of docking
968 | Med. Chem. Commun., 2015, 6, 963–970
This journal is © The Royal Society of Chemistry 2015

View Article Online
MedChemComm
Concise Article
6-amine of the 2-substituted adenines is unoccupied by an
analogue group of the N6-substituted derivatives. This data
could explain the higher affinity of the 2-substituted deriva-
tives. Hence, a possible modification of the N6-substituted
adenines by insertion of an amine group in the 2-position
seems to be suggested by the docking results. The second
family of docking conformations of the N6-substituted ade-
nines presents the purine moiety quite perfectly matching
with the same scaffold of the 2-substituted compounds, with
the 8- and 9-substituents of the two series occupying ana-
logue positions. In this sense, the different position occupied
by the phenyl-alkyl substituents could be the factor
explaining the different binding activity of the two series of
compounds.
Conclusions
In this work, two new series of 9-propyladenine derivatives
substituted at the 2-position with phenylalkylamino chains
and at the N6-position with phenylalkyl chains of different
lengths were synthesized and tested at ARs in a radioligand
binding assay (A1, A2A, and A3 AR subtypes) and evaluated for
their ability to modulate the adenylyl cyclase activity (A2B
ARs). Biological data for the new compounds showed that the
presence of a propyl chain, in the N9-position of the purine
moiety, combined with a phenylalkylamine in the 2-position
or a phenylalkyl chain in the N6-position does not improve
affinity at A2B ARs as it could be hypothesized. In fact, both
the N6- and the 2-substituted 9-propyladenine derivatives
(12–20 and 23–31, respectively) were A2A AR antagonists endo-
wed with moderate selectivity, although in some cases the
compounds showed a high A2A affinity in the low nM range.
In particular, the 2-benzylamino-8-furyl-9-propyladenine (18)
was more active than the parent 8-furyl-9-propyladenine at
A2A and A3 ARs (18: KiA2A = 6.6 nM and KiA3 = 29 nM vs. 8:
KiA2A = 18 nM and KiA3 = 790 nM). Molecular modeling stud-
ies gave some explanation of the different activities of the
compounds at the A2A AR. This study contributed to
Fig. 5 Comparison of the binding modes at hA2AAR of the 2- and N6-
substituted adenine derivatives (18, yellow, and 29, green, respectively).
Superimposition of the binding modes of the 2-substituted derivatives
with the first family of docking conformations of N6-substituted ade-
nines (A) highlights the analogue position of the phenylalkyl chains and
the matching of the 8-substituent of the first group of molecules with
the 9-substituent of the second group of compounds and vice-versa.
Superimposition of the binding modes of the 2-substituted derivatives
with the second family of docking conformations of N6-substituted
adenines (B) highlights the match of the purine scaffolds and the 8-
and 9-substituents, while the phenylalkyl chains of the two series of
molecules are differently located within the binding cavity. Red squares
indicate the matching of H-bond acceptors or donors in proximity to
the N253 residue.
strengthen the finding that the introduction of
a
phenylalkylamino chain in the 2 and 6 positions of 9-alkyl
purine derivatives favors the interaction of the resulted com-
pounds with the A2A AR subtype giving suggestions for the
synthesis of new antagonists of this receptor.
Acknowledgements
conformations of the N6-substituted derivatives presents the
purine moiety oriented in a mirrored fashion with respect to
the same scaffold of the 2-substituted compounds. This fea-
ture makes the N6-substituent of the first group of molecules
to be located and oriented similarly to the analogue group of
the 2-substituted compounds. Interestingly, the 8-substituent
of the first group of molecules is positioned in correspon-
dence to the 9-substituent of the second group of compounds
and vice-versa. The N3 atom of the N6-substituted derivatives
is similarly located at the N7 atom of the 2-substituted mole-
cules and both atoms provide H-bond interaction with the
Asn253 sidechain. The H-bond donor feature given by the
This work was supported by Fondo di Ricerca di Ateneo (Uni-
versity of Camerino) and by a grant from the Italian Ministry
for
University
and
Research
(PRIN2010-11
no
20103W4779_003).
Notes and references
1 B. B. Fredholm, A. P. IJzerman, K. A. Jacobson, J. Linden
and C. E. Muller, Pharmacol. Rev., 2011, 63, 1–34.
2 K. A. Jacobson and Z. G. Gao, Nat. Rev. Drug Discovery,
2006, 5, 247–264.
This journal is © The Royal Society of Chemistry 2015
Med. Chem. Commun., 2015, 6, 963–970 | 969

View Article Online
Concise Article
MedChemComm
3 C. N. Wilson and S. J. Mustafa, Handb. Exp. Pharmacol.,
2009, v–vi.
4 C. E. Muller and K. A. Jacobson, Biochim. Biophys. Acta,
2011, 1808, 1290–1308.
16 C. Lambertucci, M. Buccioni, B. Cacciari, D. Dal Ben, S.
Federico, K.-N. Klotz, G. Marucci, R. Volpini, G. Spalluto and
G. Cristalli, Collect. Czech. Chem. Commun., 2011, 76,
1379–1393.
5 R. Dungo and E. D. Deeks, Drugs, 2013, 73, 875–882.
6 S.-I. Uchida, K. Soshiroda, E. Okita, M. Kawai-Uchida, A.
Mori, P. Jenner and T. Kanda, Eur. J. Pharmacol., 2015, 747,
160–165.
7 R. Volpini, S. Costanzi, S. Vittori, G. Cristalli and K.-N. Klotz,
Curr. Top. Med. Chem., 2003, 3, 427–443.
17 C. Lambertucci, I. Antonini, M. Buccioni, D. Dal Ben, D. D.
Kachare, R. Volpini, K.-N. Klotz and G. Cristalli, Bioorg. Med.
Chem., 2009, 17, 2812–2822.
18 D. Dal Ben, M. Buccioni, C. Lambertucci, A. Thomas, K.-N.
Klotz, S. Federico, B. Cacciari, G. Spalluto and R. Volpini,
Eur. J. Med. Chem., 2013, 70C, 525–535.
8 C. M. Aherne, E. M. Kewley and H. K. Eltzschig, Biochim.
Biophys. Acta, 2011, 1808, 1329–1339.
19 G. D. Celik, A. Disli, Y. Oner and L. Acik, Med. Chem. Res.,
2013, 22, 1470–1479.
9 I. Feoktistov, I. Biaggioni and B. N. Cronstein, Handb. Exp.
Pharmacol., 2009, 383–397.
10 A. Mohsenin and M. R. Blackburn, Curr. Opin. Pulm. Med.,
2006, 12, 54–59.
11 E. Camaioni, S. Costanzi, S. Vittori, R. Volpini, K.-N. Klotz
and G. Cristalli, Bioorg. Med. Chem., 1998, 6, 523–533.
12 K.-N. Klotz, S. Kachler, C. Lambertucci, S. Vittori, R. Volpini
and G. Cristalli, Naunyn-Schmiedeberg's Arch. Pharmacol.,
2003, 367, 629–634.
13 C. Lambertucci, G. Cristalli, D. Dal Ben, D. D. Kachare, C.
Bolcato, K.-N. Klotz, G. Spalluto and R. Volpini, Purinergic
Signalling, 2007, 3, 339–346.
20 K.-N. Klotz, J. Hessling, J. Hegler, C. Owman, B. Kull, B. B.
Fredholm and M. J. Lohse, Naunyn-Schmiedeberg's Arch.
Pharmacol., 1998, 357, 1–9.
21 K.-N. Klotz, N. Falgner, S. Kachler, C. Lambertucci, S. Vittori,
R. Volpini and G. Cristalli, Eur. J. Pharmacol., 2007, 556, 14–18.
22 R. D. Thompson, S. Secunda, J. W. Daly and R. A. Olsson,
J. Med. Chem., 1991, 34, 2877–2882.
23 V. P. Jaakola, M. T. Griffith, M. A. Hanson, V. Cherezov, E. Y.
Chien, J. R. Lane, A. P. IJzerman and R. C. Stevens, Science,
2008, 322, 1211–1217.
24 D. Dal Ben, C. Lambertucci, G. Marucci, R. Volpini and G.
Cristalli, Curr. Top. Med. Chem., 2010, 10, 993–1018.
25 Molecular Operating Environment; C.C.G., Inc., 1255
University St., Suite 1600, Montreal, Quebec, Canada, H3B
3X3.
14 C. Lambertucci, S. Vittori, R. C. Mishra, D. Dal Ben, K.-N.
Klotz, R. Volpini and G. Cristalli, Nucleosides, Nucleotides
Nucleic Acids, 2007, 26, 1443–1446.
15 R. Volpini, D. Dal Ben, C. Lambertucci, G. Marucci, R. C.
Mishra, A. T. Ramadori, K.-N. Klotz, M. L. Trincavelli, C.
Martini and G. Cristalli, ChemMedChem, 2009, 4,
1010–1019.
26 J. J. Stewart, J. Comput.-Aided Mol. Des., 1990, 4, 1–105.
27 G. Lebon, T. Warne, P. C. Edwards, K. Bennett, C. J.
Langmead, A. G. Leslie and C. G. Tate, Nature, 2011, 474,
521–525.
970 | Med. Chem. Commun., 2015, 6, 963–970
This journal is © The Royal Society of Chemistry 2015