2,8-Disubstituted adenosine derivatives as partial agonists for the adenosine A2A receptor

Article abstract of DOI:10.1016/S0968-0896(03)00123-8

Novel 2,8-disubstituted adenosine derivatives were synthesized in good overall yields starting from 2-iodoadenosine. Binding affinities were determined for rat adenosine A₁ and A_2A receptors and human A₃ receptors. Some compounds displayed good adenosine A_2A receptor affinities, with most of the 2-(1-hexynyl)- and 2-[(E)-1-hexenyl]-substituted derivatives having K_i values in the nanomolar range. Although the introduction of an 8-alkylamino substituents decreased the affinity for the adenosine A_2A receptor somewhat, the selectivity for this receptor compared to A₃ was improved significantly. The 8-methylamino (12) and 8-propylamino (14) derivatives of 2-(1-hexynyl)adenosine (3), showed reasonable A_2A receptor affinities with K_i values of 115 and 82 nM, respectively, and were 49- and 26-fold selective for the adenosine A_2A receptor compared to the A₃ receptor. The compounds were also evaluated for their ability to stimulate the cAMP production in CHO cells expressing the human adenosine A_2A receptor. 2-(1-Hexynyl)adenosine (3) and 2-[(E)-1-hexenyl]adenosine (4) both showed submaximal levels of produced cAMP, compared to the reference full agonist CGS 21680, and thus behaved as partial agonists. Most 8-alkylamino-substituted derivatives of 3, displayed similar cAMP production as 3, and behaved as partial agonists as well. Introduction of alkylamino groups at the 8-position of 4, showed a slight reduction of the efficacy compared to 4, and these compounds were partial agonists also.

Full text of DOI:10.1016/S0968-0896(03)00123-8

Bioorganic & Medicinal Chemistry 11 (2003) 2183–2192
2,8-Disubstituted Adenosine Derivatives as Partial Agonists
for the Adenosine A2A Receptor
Erica W. van Tilburg,* Matty Gremmen, Jacobien von Frijtag Drabbe Kunzel,
Miriam de Groote and Ad P. IJzerman
Leiden/Amsterdam Center for Drug Research, Division of Medicinal Chemistry, PO Box 9502, 2300 RA Leiden, The Netherlands
Received 30 October 2002; accepted 18 February 2003
Abstract—Novel 2,8-disubstituted adenosine derivatives were synthesized in good overall yields starting from 2-iodoadenosine.
Binding affinities were determined for rat adenosine A1 and A2A receptors and human A3 receptors. Some compounds displayed
good adenosine A2A receptor affinities, with most of the 2-(1-hexynyl)- and 2-[(E)-1-hexenyl]-substituted derivatives having Ki
values in the nanomolar range. Although the introduction of an 8-alkylamino substituents decreased the affinity for the adenosine
A
2A receptor somewhat, the selectivity for this receptor compared to A3 was improved significantly. The 8-methylamino (12) and 8-
propylamino (14) derivatives of 2-(1-hexynyl)adenosine (3), showed reasonable A2A receptor affinities with Ki values of 115 and
82 nM, respectively, and were 49- and 26-fold selective for the adenosine A2A receptor compared to the A3 receptor. The com-
pounds were also evaluated for their ability to stimulate the cAMP production in CHO cells expressing the human adenosine A2A
receptor. 2-(1-Hexynyl)adenosine (3) and 2-[(E)-1-hexenyl]adenosine (4) both showed submaximal levels of produced cAMP, com-
pared to the reference full agonist CGS 21680, and thus behaved as partial agonists. Most 8-alkylamino-substituted derivatives of 3,
displayed similar cAMP production as 3, and behaved as partial agonists as well. Introduction of alkylamino groups at the
8-position of 4, showed a slight reduction of the efficacy compared to 4, and these compounds were partial agonists also.
# 2003 Elsevier Science Ltd. All rights reserved.
Introduction
binding to dopamine D2 receptors.3,4 This raises the
possibility of using adenosine A2A receptor agonists as a
novel therapeutic approach in the treatment of psy-
chosis. Additionally, selective activation of adenosine
A2A receptors has also been shown to improve inap-
propriate and/or extensive inflammatory responses in
cells such as mast cells and neutrophils.5 However,
these desired actions of agonists for the adenosine
A2A receptor may be accompanied with undesired
effects such as cardiovascular actions, since the
receptor is ubiquitously distributed.6 The design of
partial agonists for the adenosine A1 receptor has
already been shown to be a useful tool for achieve-
ment of selectivity of action in vivo by exploitation
of the differences in receptor–effector coupling in
various tissues.7ꢀ10 Hence, partial agonists for the
adenosine A2A receptor might have potential as anti-
psychotic agents. However, whether they will be devoid
of undesired cardiovascular actions, remains to be
investigated in vivo, particularly since vascular A2A
receptors are highly expressed.11
Adenosine mediates a wide variety of effects as a result
of its activation of specific membrane-bound receptors
called P1-purinoceptors. The three subclasses of P1-puri-
noceptors are A1, A2 and A3, with A2 further sub-
divided into A2A and A2B. All adenosine receptors are
coupled to the enzyme adenylate cyclase; activation of
the A1 and A3 adenosine receptors inhibit the adenylate
cyclase, whereas activated A2A and A2B receptors stim-
ulate it. The target receptor in this study, the adenosine
A2A receptor, can be found throughout the whole body,
and A2A receptor agonists might be used to inhibit pla-
telet aggregation in thrombosis, in the diagnosis of dis-
eases in coronary arteries, in ischemia and reperfusion.1,2
Furthermore, activation of adenosine A2A receptors has
been shown to alter the binding characteristics of other
receptors. Stimulation of adenosine A2A receptors in
rat striatal membranes reduces the affinity of agonist
*Corresponding author at current address: Radionuclide Center, Vrije
Universiteit, De Boelelaan 1085c, 1081HV Amsterdam, The Nether-
lands. Tel.: +31-20-444-9707; fax.: +31-20-444-9121; e-mail: etilburg@
rnc.vu.nl
Selectivity for the adenosine A2A receptor can be
obtained by the introduction of a 2-substituent, such as
0968-0896/03/$ - see front matter # 2003 Elsevier Science Ltd. All rights reserved.
doi:10.1016/S0968-0896(03)00123-8

2184
E. W. van Tilburg et al. / Bioorg. Med. Chem. 11 (2003) 2183–2192
a 1-hexynyl or a (E)-1-hexenyl group that have been
shown to induce high affinity for the adenosine A2A
receptor compared to A1.12ꢀ14 Partial agonism for
adenosine receptor in general has been achieved by the
introduction of alkylthio-sub0stituents at the 50-posi-
tion10,15 or by removing the 2 -hydroxy group.16 How-
ever, partial agonism for the adenosine A1 receptor has
also successfully been accomplished by introducing
alkylamino substituents at the 8-position of adenosine
by our laboratory.17 8-Alkylamino substituted CPA
(N6-cyclopentyladenosine) derivatives have proven to be
adenosine A1 receptor partial agonists in vivo when
assessed for cardiovascular activity, although with
modest affinities.8,18,19
cross-coupling reaction12 by reacting 2-iodoadenosine20
(1) with 1-hexyne.12,13 In literature, classical cross-cou-
pling of terminal alkenes with compound 1 has been
shown to be disappointing, with 2-(E)-alkenyl deriva-
tives obtained in very low yields only.14 Therefore an
alternative route was used here,14 that includes the pre-
paration of a (E)-1-(borocatechol)-1-alkene complex, by
letting catecholborane react with the appropriate terminal
alkyne. These (E)-1-(borocatechol)-1-alkene complexes
were coupled successfully with 2-iodoadenosine deriva-
tives. Treatment of 1 with (E)-1-(borocatechol)-1-hexene,
gave 2-[(E)-1-hexenyl]adenosine (4) in reasonable yield.
To obtain the 2,8-disubstituted derivatives (12–23),
either unprotected 2-iodoadenosine (1) or 20,30,50-tri-O-
acetyl-2-iodoadenosine (2) was brominated at the
8-position. Using a standard bromination procedure,
for example stirring 2-iodoadenosine (1) or 20,30,50-tri-
O-acetyl-2-iodoadenosine (2) in dioxane and adding
bromine water (in phosphate buffer, 10% w/v,
pH=7),21 yielded the corresponding 8-bromo deriva-
tives, although workup often was quite laborious. With
the use of a different buffer (NaOAc buffer, 1.0 M,
pH=4), 2-iodoadenosine (1) readily dissolved (50 ꢁC)
and after addition of Br2, stirring overnight and sub-
sequent adjustment of the pH to 7, the 8-brominated
compound (5) was obtained as a white solid that could
easily be collected by filtration.22
In this study, the synthesis and biological evaluation of
a series of adenosine analogues substituted at the
2-position with a 1-hexynyl or (E)-1-hexenyl group for
A2A selectivity and at the 8-position with alkylamino
groups for potential partial agonism are described. The
compounds were tested in radioligand binding assays
for affinity. Their intrinsic activities were determined in
cAMP assays.
Chemistry
The synthesis of the 2,8-disubstituted adenosine deriva-
tives 3–5, 7–23 is illustrated in Scheme 1. 2-(1-Hex-
ynyl)adenosine (3) was prepared in good yield (85%) via
Subsequently, different amines could be introduced at
this position by stirring either 5 or 6 with the appro-
a
slightly modified traditional palladium-catalyzed
Scheme 1. (a) (i) NaOAc buffer (1.0 M, pH 4), Br2 (ii) NaHSO3, aq NaOH; (b) the appropriate amine; (c) CH3CN, Et3N, CuI, PdCl2, Ph3P and
1-hexyne; (d) CH3CN/DMF (1:1), tetrakis-(triphenylphosphine)-palladium(0), K2CO3 and (E)-1-(borocatechol)-1-hexene; (e) benzyl-amine, reflux.

E. W. van Tilburg et al. / Bioorg. Med. Chem. 11 (2003) 2183–2192
2185
priate amine overnight (less nucleophilic amines
required some heating). Under these reaction conditions
the acetyl protecting groups, when present, were readily
removed and compounds 7–11 were obtained in good
yields (70–91%).23 Finally, the 1-hexynyl and the (E)-1-
hexenyl substituents were introduced at the 2-position
as described above for compounds 3 and 4, and yielded
the desired compounds 12–21. The alternative synthesis
of compounds 12–21 by the introduction of the 2-sub-
stituent prior to the 8-substituent failed. For example,
compound 3 could indeed be brominated at the 8-posi-
tion, however, addition of bromine at the triple bond of
its 2-substituent occurred as well. Compounds 22 and 23
were obtained by treating 8-bromo-2-iodoadenosine (5)
with an excess of either 1-hexyne or benzylamine.
azolo[1,5-a][1,3,5]triazine) were used. At the time the
compounds were tested, no suitable radiolabeled
antagonists were commercially available for the adeno-
sine A3 receptor. Therefore [125I]AB-MECA (N6-(4-
amino-3-iodobenzyl)-50-methylcarboxamidoadenosine),
an A3 receptor agonist, was utilized. Displacement
experiments were performed in the absence of GTP
(guanosine 50-triphosphate).
All compounds were also tested in a functional assay.
The ability of the compounds (3–5, 7–23) to produce
cAMP by activation of human adenosine A2A receptors
expressed in CHO cells was assessed and compared to
the reference full agonist CGS 21680 (2-[4-(2-carboxy-
ethyl)phenylethylamino]-50 -N-ethylcarboxamidoadeno-
sine, 100%).
Table 1 displays radioligand-binding data for all syn-
thesized final products 1, 3–5, 7–23. From this table, it is
clear that most synthesized compounds had low affi-
nities for the adenosine A1 and A3 receptor. The affi-
nities of the compounds for the adenosine A2A receptor
were higher, depending on the 2-substituent present.
The compounds with an 8-alkylamino group had lower
adenosine A2A receptor affinities than the 8-unsub-
stituted derivatives 3 and 4. The introduction of either
an (E)-1-hexenyl or a 1-hexynyl substituent at the
2-position of 8-alkylamino adenosine derivatives led to
Biological Evaluation
All compounds were tested in radioligand binding
assays to determine their affinities for the adenosine A1
receptor in rat brain cortex, the A2A receptor in rat
striatum and the human A3 receptor as expressed in
HEK 293 cells (Table 1). For the adenosine A1 receptor,
the tritiated antagonist, [3H]-1,3-dipropyl-8-cyclo-
pentylxanthine ([3H]DPCPX), and for the adenosine A2A
receptor, the tritiated antagonist [3H]ZM 241385 (7-amino-
2-(2-furyl)-5-[2-(4-hydroxyphenyl)ethyl]amino[1,2,4]-tri-
Table 1. Affinities of 2,8-disubstituted adenosine analogues at adenosine A1, A2A and A3 receptors expressed as Ki values (ÆSEM in nM, n=3) or
percentage displacement (n=2, mean value) at 10 mM
Compd
R2
R3
Ki (nM) or% displacement at 10ꢀ5 Ma
b
c
d
A1
A2A
A3
A3/A2A
1
3
4
5
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
I
H
H
36%
64%
53%
22%
35%
35%
40%
28%
36%
32%
31%
48%
539Æ380
756Æ483
18%
30%
50%
4200Æ80
6.0Æ1.0
26Æ1.8
43%
297Æ17
17Æ4.1
0.07
2.8
2.8
—
—
—
—
—
—
49
35
26
—
—
20
9.5
21
—
—
—
—
CꢂC(CH2)3CH3
(E)CH¼CH(CH2)3CH3
I
H
Br
NHCH3
74Æ7.7
6%
I
I
I
I
43%
41%
7310Æ440
5110Æ890
8670Æ2000
43%
NHCH2CH3
NH(CH2)2CH3
NH(CH2)3CH3
NHCH2C6H5
NHCH3
NHCH2CH3
NH(CH2)2CH3
NH(CH2)3CH3
NHCH2C6H5
NHCH3
NHCH2CH3
NH(CH2)2CH3
NH(CH2)3CH3
NHCH2C6H5
CꢂC(CH2)3CH3
NHCH2C6H5
51%
3110Æ1650
I
45%
40%
CꢂC(CH2)3CH3
CꢂC(CH2)3CH3
CꢂC(CH2)3CH3
CꢂC(CH2)3CH3
CꢂC(CH2)3CH3
(E)CH¼CH(CH2)3CH3
(E)CH¼CH(CH2)3CH3
(E)CH¼CH(CH2)3CH3
(E)CH¼CH(CH2)3CH3
(E)CH¼CH(CH2)3CH3
CꢂC(CH2)3CH3
NHCH2C6H5
115Æ8.0
253Æ29
82Æ10
149Æ29
35%
663Æ106
1840Æ300
678Æ25
580Æ250
26%
5640Æ780
8830Æ1230
2160Æ120
64%
7970Æ610
13440Æ2130
17530Æ2100
14330Æ2160
47%
906Æ264
2110Æ560
50%
24%
254Æ34
29%
12%
50%
54%
aPercentages were given as the means of two independent determinations (SE<15%).
bDisplacement of [3H]DPCPX from rat cortical membranes.39
cDisplacement of [3H]ZM241385 from rat striatal membranes.40
dDisplacement of [125I]AB MECA from the human A3 receptor expressed in HEK 293 cells.37,42

2186
E. W. van Tilburg et al. / Bioorg. Med. Chem. 11 (2003) 2183–2192
an increase in affinity for the adenosine A2A receptor,
up to approximately 7- or 60-fold, respectively (Table
1).17 Remarkably, these compounds had adenosine A2A
receptor affinities in the nanomolar range, with com-
pound 14 having the highest A2A receptor affinity (Ki
value of 82 nM). This is consistent with previously
reported data on 8-substituted adenosine analogues.
The introduction of 8-substituents at adenosine decrea-
ses the affinity of the resulting compounds for the ade-
nosine receptors compared to adenosine itself.17 8-
Alkylamino adenosine derivatives had adenosine A1
and A2A receptor affinities in the mM range, without
substantial preference for either receptor, whereas
introduction of a cyclopentyl group at the N6-position
led to an increase in adenosine A1 receptor affinity up to
23-fold.18 The 2-(1-hexynyl) adenosine derivatives 12–16
generally had higher adenosine A2A receptor affinities
than the 2-[(E)-1-hexenyl]-substituted compounds 17–
21, in line with the receptor affinities of compounds 3
and 4 as well as data from literature.14 Within both
2-substituted series (12–16 and 17–21, Table 1), the
8-propylamino substituent seemed to be tolerated best
on the adenosine A2A receptor. This is in contrast with
the 2-unsubstituted derivatives, for which the 8-methyl-
amino group appeared optimal for adenosine A2A
receptor affinity.17 In general, all compounds were more
adenosine A2A receptor selective compared to the A1
receptor than when they are compared to the A3 recep-
tor. Exceptions were compounds 16 and 21, containing
the large benzylamino group at the 8-position. These
compounds displayed higher adenosine A1 receptor
affinities compared to A2A and A3, indicating that the
adenosine A1 receptor seemed best capable of accom-
modating this group. A1-hexynyl group at the 8-posi-
tion strongly decreased the A2A receptor affinity as well
(compounds 3 and 22). The adenosine A3 receptor
seemed better able to accommodate the 1-hexynyl group
(22) than the benzylamino group at the 8-position (16,
21).24 Finally, the adenosine A2A receptor affinity of
compound 16, with a 1-hexynyl group at the 2-position,
was higher than that of the 2-benzylamino-substituted
derivative (23), also in line with receptor (functional)
data on either 2-(1-hexynyl)-12 or 2-benzylamino-sub-
stituted25 derivatives. Other adenosine derivatives sub-
stituted at the 8-position have not displayed very high
receptor affinities.18 In the present study, however,
8-alkylamino-substituted adenosine derivatives were
obtained with affinities in the low micromolar (17–20)
or even nanomolar range (12–15) for the adenosine A2A
receptor. Although the adenosine A2A receptor affinity
was somewhat decreased with the introduction of the
8-alkylamino groups, the A2A receptor selectivity com-
pared to (A1 and) A3 was increased. Many 2-substituted
adenosine derivatives have been described as potent
ligands for the A2A receptor, although binding data for
the adenosine A3 receptor is often lacking.12,14,26ꢀ28
Some 2-alkynyl adenosine derivatives have been tested
on the adenosine A3 receptor.29ꢀ32 These compounds,
of 2-(1-hexynyl)adenosine (3), showed reasonable A2A
receptor affinities with Ki values of 115 and 82 nM,
respectively. Furthermore, the selectivity for the adeno-
sine A2A receptor compared to A3 was approx. 49- and
26-fold, respectively. Although the affinities of the 2-[(E)-
1-hexenyl]adenosines were somewhat lower, the selectiv-
ity for the A2A receptor was also increased. It must be
mentioned that the affinity of some (2-) substituted ade-
nosine derivatives for the human A2A receptor has been
shown to be somewhat lower than their rat A2A receptor
affinity. Examples are compounds such as CPA, CGS
21680, NECA, DMPA (N6-[2-(3,5-dimethoxyphenyl)-2-(2
-methylphenyl)ethyl]adenosine) and HENECA, which all
display lower affinity for the human adenosine A2A
receptor than for the rat A2A receptor.30,33,34 These data
suggest that our synthesized compounds might also have
somewhat lower affinities for the human adenosine A2A
receptor and therefore their selectivity compared the
(human) A3 receptor might be somewhat less.
The nanomolar adenosine A2A receptor affinities of
compounds 12–15, and the acceptable A2A receptor
affinities of compounds 17–20, indicate that the (fairly
large) size of the substituents is not the main determin-
ing factor for the affinities of these compounds. The
8-substituents may have an influence on the conforma-
tion of the ligand itself (by forcing the ribose ring into
the syn conformation). The conformation of 8-bromo-
adenosine in the crystal structure is syn, suggesting to
cause the low affinity of this compound for the adeno-
sine receptors.35ꢀ37 The bulkiness of 8-alkylamino
groups was initially thought to force the ribose ring into
the syn conformation as well. However, from the X-ray
structure of 8-(cyclopentylamino)-N6-ethyladenosine it
became apparent that the anti conformation is compat-
ible with 8-substitution.18 The difference in energy
between the syn and the anti conformation is often not
very large, and a direct steric hindrance at the receptor-
binding site seems to be a more probable explanation.
Furthermore, the electronic (hydrogen bond formation)
and lipophilic characteristics of the 8-substituents influ-
ence receptor affinity too.17
Table 2 shows the effects of the synthesized compounds
in cAMP assays. For determination of the amount of
cAMP produced via adenosine A2A receptor agonism,
all compounds were first tested at a single concentration
(approx. 100ꢃ the Ki value). Compounds 3 and 4,
without an alkylamino group at the 8-position, dis-
played partial agonism compared to the reference full
agonist CGS 21680. The intrinsic activity of compound
3 was less than that of compound 4, consistent with
data previously obtained in our laboratory.38 Com-
pound 1, with iodine at the 2-position, behaved as a full
agonist. Introduction of the 8-alkylamino groups to
compound 1, in most cases led (except compound 11) to
a decrease in intrinsic activity. Compound 9 showed a
cAMP production of only 4%, compared to CGS
21680. Within the 2-(hexynyl) series (12–16), com-
pounds 14–16 had similar intrinsic activities as the
8-unsubstituted compound 3, while 12 had a somewhat
higher intrinsic activity. They all behaved as partial
agonists compared to CGS 21680, whereas compound
0
such as HENECA[2-(1-hexynyl)-5 -deoxy-50-N-ethyl-
carboxamidoadenosine], had high affinity for the A3
receptor, and were often more selective for the adeno-
sine A3 compared to the A2A receptor. Here, the
8-methylamino (12) and 8-propylamino (14) derivatives

E. W. van Tilburg et al. / Bioorg. Med. Chem. 11 (2003) 2183–2192
2187
Table 2. Percentage of cAMP production (n=2, mean value, both
values in parentheses) via the human adenosine A2A receptor expres-
sed in CHO cells (at approx. 100 ꢃ the Ki value), compared to the
reference full agonist CGS 21680 (10 mM)
and were 49- and 26-fold selective for the adenosine A2A
receptor compared to the A3 receptor. The adenosine
A2A receptor seemed to accommodate the 8-propyl- or
8-butylamino substituents best, whereas the even larger
8-substituents [8-benzylamino or 8-(1-hexynyl)] were not
well tolerated. As for the intrinsic activity, the 8-unsub-
stituted compounds 3 and 4 displayed a reduced intrin-
sic activity compared to the reference agonist CGS
21680. In general, the introduction of 8-alkylamino
substituents did either not affect the intrinsic activity
much, or as in most cases, further reduced it, and most
of the compounds behaved as partial agonists. Most of
the 8-substituted derivatives of 2-iodoadenosine (1)
behaved as partial agonists as well, except for the very
bulky ones. Whether these novel adenosine derivatives
will be devoid of undesired cardiovascular actions,
remains to be investigated in vivo. However, these com-
pounds may be useful pharmacological tools, and may
have reduced side effects on the cardiovascular system,
while possibly retaining their useful antipsychotic prop-
erties upon adenosine A2A receptor activation.
Compd
R2
R3
% cA MP
production
hA2A
a
CGS 21680
1
3
4
5
100
126
61
82
n.d.
23
I
H
H
CꢂC(CH2)3CH3
(E) CH¼CH(CH2)3CH3
H
Br
NHCH3
I
I
7
8
9
I
I
I
NHCH2CH3
NH(CH2)2CH3
NH(CH2)3CH3
NHCH2C6H5
NHCH3
NHCH2CH3
NH(CH2)2CH3
NH(CH2)3CH3
NHCH2C6H5
NHCH3
77
4
28
125
85
134
58
72
83
54
65
73
36
73
99
10
11
12
13
14
15
16
17
18
19
20
21
22
23
I
CꢂC(CH2)3CH3
CꢂC(CH2)3CH3
CꢂC(CH2)3CH3
CꢂC(CH2)3CH3
CꢂC(CH2)3CH3
(E) CH¼CH(CH2)3CH3
(E) CH¼CH(CH2)3CH3
Experimental
Chemicals and solvents
All reagents were from standard commercial sources
and of analytic grade. [3H]DPCPX, [3H]CGS 21680 and
NHCH2CH3
(E) CH¼CH(CH2)3CH3 NH(CH2)2CH3
(E) CH¼CH(CH2)3CH3 NH(CH2)3CH3
[
125I]AB-MECA were purchased from NEN (Hoofd-
dorp, The Netherlands).
(E) CH¼CH(CH2)3CH3
CꢂC(CH2)3CH3
NHCH2C6H5
NHCH2C6H5
CꢂC(CH2)3CH3
NHCH2C6H5
95
Chromatography
aPercentages were given as the means of two independent determina-
tions (SE<20%).
Thin-layer chromatography (TLC) was carried out
using aluminum sheets (20ꢃ20 cm) with silica gel F254
from Merck. Spots were visualized under UV (254 nm).
Preparative column chromatography was performed on
silica gel (230–400 mesh ASTM).
13 behaved as a full agonist. Within the 2-(E)-alkenyl
substituted series (17–21), the introduction of methyl-,
ethyl-, and butylamino groups all decreased the intrinsic
activity compared to compound 4. The intrinsic activ-
ities of compounds 19 and 21 were similar to that of 4
and they all behaved as partial agonists as well. Most
compounds with very bulky 8-substituents (11, 16, 22
and 23) behaved as full agonists at the adenosine A2A
receptor in this assay. Actually, CGS 21680 appeared to
be a partial agonist here, since its intrinsic activity was
lower than that of compounds 1, 11 and 13.
Instruments and analyses
Elemental analyses were performed for C, H, N
(Department of analytical Chemistry, Leiden Uni-
versity, The Netherlands). 13C NMR spectra were mea-
sured at 50.1 MHz with
spectrometer equipped with a PG 200 computer oper-
a
Jeol JNM-FX 200
1
ating in the Fourier-transform mode. H NMR spectra
were measured at 200 MHz, using the above mentioned
spectrometer, or at 300 MHz, using a Bruker WM-300
spectrometer equipped with an ASPECT-2000 compu-
ter operating in the Fourier-transform mode. Chemical
shifts for 1H and 13C NMR are given in ppm (d) relative
to tetramethylsilane (TMS) as internal standard.
Conclusions
The 2,8-disubstituted adenosine derivatives described in
the present study were synthesized in good overall yields
starting from 2-iodoadenosine. Most compounds
appeared to have adenosine A2A receptor affinities in
the low micromolar or nanomolar range. Although the
affinity for the adenosine A2A receptor was decreased
somewhat with the introduction of the 8-alkylamino
substituents, the selectivity for this receptor compared
to the A3 receptor was improved significantly. The
8-methylamino (12) and 8-propylamino (14) derivatives
of 2-(1-hexynyl)adenosine (3), showed high A2A receptor
affinities with Ki values of 115 and 82 nM, respectively,
All high resolution mass spectra were measured on a
Finnigan MAT900 mass spectrometer equipped with a
direct insertion probe for EI experiments (70 eV with
resolution 1000) or on a Finnigan MAT TSQ-70 spec-
trometer equipped with an electrospray interface for
ESI experiments. Spectra were collected by constant
infusion of the analyte dissolved in 80/20 methanol/
H2O. ESI is a soft ionization technique resulting in

2188
E. W. van Tilburg et al. / Bioorg. Med. Chem. 11 (2003) 2183–2192
protonated, sodiated species in positive ionization mode
and deprotonated species in the negative ionization mode.
added bromine (83.5 mL). The mixture was stirred vig-
orously for 15 min until most of the bromine had dis-
solved. Then 2.6 mL of the decanted bromine solution
was added to 20,30,50-tri-O-acetyl-2-iodoadenosine (2,
132 mg, 0.25 mmol) in dioxane (2.6 mL) cooled in an ice/
water bath. After 20 min the icebath was removed and
the reaction mixture was stirred overnight at room
temperature. The mixture was cooled again (icebath)
and NaHSO3 (1.8 M) was added dropwise until it
became colorless. The waterlayer was extracted with
CH2Cl2 (1 ꢃ 25 mL) and EtOAc (2 ꢃ 20 mL). The
combined organic layers were dried (MgSO4), filtered
and concentrated in vacuo. The residue was purified by
column chromatography (eluent EtOAc). Yield 94.2 mg
Resolution of the compounds was achieved by reverse-
phase HPLC (Gilson HPLC system, 712 system con-
troller software. Gilson Netherlands, Meyvis en Co BV,
Bergen op Zoom, the Netherlands) using as a mobile
phase either: A: 20% CH3CN in H2O–100% CH3CN in
35min or B: 30% MeOH in H2O–100% MeOH in 40min;
an Alltima C18 5 mm (250 ꢃ 4.6 mm) column (Alltech
Nederland BV, Breda, The Netherlands) at a flow rate of
1 mL/min. The peaks were defined by measurement of UV
absorbance (254nm). Retention times are given.
1
Melting points (uncorrected) were determined in a
Buchi capillary melting point apparatus.
(0.16 mmol, 63%), Rf 0.61 (5% MeOH in CH2Cl2); H
NMR (DMSO-d6) d 7.95 (bs, 2H, NH2), 6.01–5.99 (m,
2H, H-10,20), 5.75–5.73 (m, 1H, H-30), 4.37–4.17 (m, 3H,
H-40,50), 2.10, 2.08, 1.97 (3ꢃs, 9H, 3ꢃCOCH3) ppm.
Syntheses
2-Iodoadenosine (1). This was prepared according to lit-
erature.20 Yield 80%; mp 185–187 ꢁC; Rf 0.21 (10%
MeOH in CH2Cl2). Anal. calcd for C10H12IN5O4: C,
32.18; H, 3.49; N, 16.60. Found (ꢄ0.33 CH3CO2C2H5)
C, 32.48; H, 3.10; N, 16.72.
General procedure for amination of compound 5 or 6 to
obtain the 8-alkylamino-2-iodoadenosines 7–10. To 8-
bromo-2-iodoadenosine (5) or 20,30,50-tri-O-acetyl-8-
bromo-2-iodoadenosine (6) (0.17 mmol) was added the
appropriate alkylamine (excess) and the mixture was
stirred overnight at room temperature. The reaction
mixture was concentrated in vacuo and the product was
crystallized from water.
20,30,50-tri-O-Acetyl-2-iodoadenosine (2). Acid anhydride
(0.34 mL, 3.60 mmol) was added to a suspension of
2-iodoadenosine (1, 400 mg, 1.02 mmol) and dimethyl-
aminopyridine (DMAP, 9.32 mg, 0.08 mmol) in a mix-
ture of acetonitrile (13 mL) and Et3N (0.56 mL,
4.04 mmol) at room temperature. After stirring for 1 h,
the solution became clear. Methanol (5 mL) was added
and stirring was continued for 5 min. The mixture was
concentrated in vacuo. The residue was extracted with
H2O (15 mL) and EtOAc (15 mL). The organic layer
was dried (MgSO4), concentrated and purified further
by column chromatography (eluent EtOAc). Yield
434 mg (0.84 mmol, 82%), Rf 0.70 (10% MeOH in
CH2Cl2); 1H NMR (DMSO-d6) d 8.29 (s, 1H, H-8), 7.79
(bs, 2H, NH2), 6.12 (d, 1H, J=5.15 Hz, H-10), 5.84 (t,
1H, J=5.84 Hz, H-20), 5.59 (t, 1H, J=5.14 Hz, H-30),
4.40–4.25 (m, 3H, H-40,50), 2.11 (s, 3H, COCH3), 2.04 (s,
3H, COCH3), 1.98 (s, 3H, COCH3) ppm.
2-Iodo-8-methylaminoadenosine (7). The reaction was
performed with 20,30,50-tri-O-acetyl-8-bromo-2-iodoade-
nosine (6, 100 mg, 0.17 mmol) and methylamine (16 mL,
40% w/v in water). Yield 54.6 mg (0.13 mmol, 77%), mp
162–164 ꢁC; 1H NMR (DMSO-d6) d 7.02 (d, 1H,
J=5.15 Hz, NH), 6.93 (s, 2H, NH2), 5.77 (d, 1H,
J=7.56 Hz, H-10), 5.66 (t, 1H, J=4.81 Hz, OH-20), 5.31
(d, 1H, J=6.52 Hz, OH-50), 5.18 (d, 1H, J=4.46 Hz,
OH-30), 4.56 (q, 1H, J=5.15 Hz, H-20), 4.09–4.04 (m,
1H, H-30), 3.98–3.91 (m, 1H, H-40), 3.68–3.57 (m, 2H,
H-50), 2.86 (d, 3H, J=4.46 Hz, CH3); MS m/z 423
(M+H)+. Anal. calcd for C11H15IN6O4: C, 29.00; H,
4.14; N, 18.44. Found (ꢄ 1.9H2O) C, 29.19; H, 4.21; N,
18.18.
8-Ethylamino-2-iodoadenosine (8). The reaction was
performed with 8-bromo-2-iodoadenosine (5, 90 mg,
0.19 mmol) and ethylamine (2.5 mL, 70% w/v in water).
Yield 58.3 mg (0.13 mmol, 70%), mp 205–207 ꢁC, Rf
8-Bromo-2-iodoadenosine (5). 2-Iodoadenosine (1,
2.93 g, 7.45 mmol) was dissolved in NaOAc buffer
(1.0 M, pH 4, 50 mL) at 50 ꢁC. The solution was cooled
to room temperature and bromine (0.46 mL, 8.94 mmol)
was added. After stirring overnight at room tempera-
ture, adding NaHSO3 destroyed the excess of bromine
and the pH of the solution was adjusted to 7 with
NaOH solution (5 M). The reaction mixture was kept at
4 ꢁC for 5 h and the precipitate was filtered. The white
solid was washed with water and dried. Yield 2.43 mg
1
0.15 (10% MeOH in CH2Cl2); H NMR (DMSO-d6) d
6.97 (t, 1H, J=5.15 Hz, NH), 6.89 (s, 2H, NH2), 5.79 (d,
1H, J=7.56 Hz, H-10), 5.65–5.61 (m, 1H, OH-20), 5.30
(d, 1H, J=6.18 Hz, OH-50), 5.18 (d, 1H, J=3.77 Hz,
OH-30), 4.55 (q, 1H, J=6.17 Hz, H-20), 4.08–4.04 (m,
1H, H-30), 3.94 (bs, 1H, H-40), 3.65–3.61 (m, 2H, H-50),
2.49 (m, 2H, CH2), 1.16 (t, 3H, J=7.21 Hz, CH3); MS
m/z 437 (M+H)+. Anal. calcd for C12H17IN6O4: C,
33.41; H, 4.64; N, 17.68. Found (ꢄ 1.2CH3OH) C, 33.11;
H, 4.91; N, 17.72.
1
(5.14 mmol, 69%), Rf 0.69 (10% MeOH in EtOAc); H
NMR (DMSO-d6) d 7.91 (s, 2H, NH2), 5.77 (d, 1H,
J=6.52 Hz, H-10), 5.47 (d, 1H, J=5.84 Hz, OH-20), 5.24
(d, 1H, J=4.80 Hz, OH-50), 5.03–4.84 (m, 2H, OH-30,
H-20), 4.20–4.09 (m, 1H, H-30), 3.97–3.88 (m, 1H, H-40),
3.71–3.42 (m, 2H, H-50) ppm.
2-Iodo-8-propylaminoadenosine (9). The reaction was
performed with 8-bromo-2-iodoadenosine (5, 90 mg,
0.19 mmol) and propylamine (2.5 mL, 70% w/v in
water). Yield 68.8 mg (0.15 mmol, 80%), mp 160–
162 ꢁC, Rf 0.20 (10% MeOH in CH2Cl2); 1H NMR
20,30,50-tri-O-Acetyl-8-bromo-2-iodoadenosine (6). To an
aqueous Na2HPO4 solution (10% w/v, 10 mL) was

E. W. van Tilburg et al. / Bioorg. Med. Chem. 11 (2003) 2183–2192
2189
(DMSO-d6) d 6.99 (t, 1H, J=4.80 Hz, NH), 6.87 (s, 2H,
NH2), 5.80 (d, 1H, J=8.24 Hz, H-10), 5.67–5.63 (m, 1H,
OH-20), 5.30 (d, 1H, J=6.18 Hz, OH-50), 5.18 (d, 1H,
J=3.77 Hz, OH-30), 4.56–4.51 (m, 1H, H-20), 4.09–4.03
(m, 1H, H-30), 3.96–3.94 (m, 1H, H-40), 3.62–3.59 (m,
4H, H-50, NHCH2), 1.57 (q, 2H, J=7.20 Hz, CH2CH3),
0.88 (t, 3H, J=7.50 Hz, CH3); MS m/z 451 (M+H)+.
Anal. calcd for C13H19IN6O4: C, 38.04; H, 5.49; N,
18.78. Found [ꢄ 1.8HCON(CH3)2] C, 37.95; H, 5.71; N,
18.80.
10), 5.45 (d, J=6.18 Hz, 1H, OH-20), 5.22–5.16 (m, 1H,
OH-50), 5.22–5.16 (m, 1H, OH-30), 4.52 (q, J=5.15 Hz,
1H, H-20), 4.11 (q, J=3.43 Hz, 1H, H-30), 3.93 (d,
J=3.43 Hz, 1H, H-40), 3.65–3.48 (m, 2H, H-50), 2.39 (t,
J=6.86 Hz, 2H, ꢂCCH2), 1.51–1.39 (m, 4H, CH2CH2),
0.90 (t, J=6.87 Hz, 3H, CH3); MS m/z 348 (M+H)+.
Anal. calcd for C16H21N5O4: C, 53.25; H, 5.91; N,
19.15. Found (ꢄ 0.22CH2Cl2) C, 53.31; H, 5.82; N,
19.05.
2-(1-Hexynyl)-8-methylaminoadenosine (12). The reac-
tion was performed with 2-iodo-8-methylaminoadeno-
sine (7, 50 mg, 0.12 mmol) and 1-hexyne (0.58 mmol).
The mixture was purified by column chromatography
(5% MeOH in CH2Cl2). Yield 36 mg (0.09 mmol, 79%),
mp 161–163 ꢁC; Rf 0.23 (5% MeOH in CH2Cl2); 1H
NMR (DMSO-d6) d 7.01 (d, 1H, J=4.81 Hz, NH), 6.61
(s, 2H, NH2), 5.84 (d, 1H, J=7.55 Hz, H-10), 5.78 (t,
1H, J=4.47 Hz, OH-20), 5.26 (d, 1H, J=6.87 Hz, OH-
50), 5.15 (d, 1H, J=4.12 Hz, OH-30), 4.58 (q, 1H,
J=5.83 Hz, H-20), 4.10–4.03 (m, 1H, H-30), 3.95 (bs, 1H,
H-40), 3.62 (bs, 2H, H-50), 2.87 (s, 3H, J=4.46 Hz,
NHCH3), 2.36 (t, 2H, J=6.87 Hz, ꢂCCH2), 1.53–1.39
(m, 4H, CH2CH2), 0.89 (t, 3H, J=6.87 Hz, CH3); MS
m/z 376 (M+H)+. Anal. calcd for C17H24N6O4: C,
47.69; H, 7.00; N, 19.63. Found (ꢄ 2.9H2O) C, 47.66; H,
7.21; N, 19.73.
8-Butylamino-2-iodoadenosine (10). The reaction was
performed with 8-bromo-2-iodoadenosine (5, 1.15 g,
2.44 mmol) and n-butylamine (25 mL). Some drops of
water were added to dissolve everything. Yield 0.92 g
(1.98 mmol, 81%), mp 142–144 ꢁC, Rf 0.23 (10% MeOH
1
in CH2Cl2); H NMR (DMSO-d6) d 6.94–6.86 (m, 3H,
NH, NH2), 5.79 (d, 1H, J=7.56 Hz, H-10), 5.67–5.62
(m, 1H, OH-20), 5.30–5.16 (m, 2H, OH-30,50), 4.54–4.49
(m, 1H, H-20), 4.07–4.04 (m, 1H, H-30), 3.95–3.92 (m,
1H, H-40), 3.64–3.57 (m, 2H, H-50), 2.49–2.42 (m, 2H,
NHCH2), 1.55–1.26 (m, 4H, CH2CH2CH3), 0.88 (t, 3H,
J=7.21 Hz, CH3); MS m/z 464 (M+H)+. Anal. calcd
for C14H21IN6O4: C, 36.22; H, 4.56; N, 18.10. Found C,
36.55; H, 4.24; N, 18.22.
8-Benzylamino-2-iodoadenosine (11). 8-Bromo-2-iodo-
adenosine (5, 1.28g, 2.14mmol) was dissolved in benzyl-
amine (21.4 mmol, 2.34 mL) and some drops of water
were added to dissolve everything. The mixture was
stirred overnight at 60 ꢁC and concentrated in vacuo.
The white solid was stirred in CH2Cl2, filtered and
dried. Yield 0.97 g (1.95 mmol, 91%), mp 128–130 ꢁC,
8-Ethylamino-2-(1-hexynyl)adenosine (13). The reaction
was performed with 8-ethylamino-2-iodoadenosine (8,
67 mg, 0.15 mmol) and 1-hexyne (0.73 mmol). The mix-
ture was purified by column chromatography (EtOAc–
10% MeOH in EtOAc). Yield 49 mg (0.12 mmol, 83%),
1
1
Rf 0.19 (5% MeOH in EtOAc); H NMR (DMSO-d6) d
mp 230–232 ꢁC; H NMR (DMSO-d6) d 6.97 (t, 1H,
7.65 (t, 1H, NH), 7.37–7.21 (m, 5H, phenyl), 6.91 (bs,
2H, NH2), 5.86 (d, 1H, J=7.90 Hz, H-10), 5.63 (t, 1H,
J=3.43 Hz, OH-20), 5.37 (d, 1H, J=6.52 Hz, OH-50),
5.20 (d, 1H, J=3.77 Hz, OH-30), 4.65–4.55 (m, 3H, H-
20, NHCH2), 4.12–4.07 (m, 1H, H-30), 4.07–3.97 (m, 1H,
H-40), 3.64–3.61 (m, 2H, H-50); MS m/z 498 (M+H)+.
Anal. calcd for C17H19IN6O4: C, 40.98; H, 3.84; N,
16.87. Found C, 40.59; H, 4.05; N, 16.50.
J=4.80 Hz, NH), 6.57 (s, 2H, NH2), 5.86 (d, 1H,
J=7.55 Hz, H-10), 5.74 (t, 1H, J=4.80 Hz, OH-20), 5.25
(d, 1H, J=6.52 Hz, OH-50), 5.15 (d, 1H, J=4.12 Hz,
OH-30), 4.57 (q, 1H, J=6.18 Hz, H-20), 4.08 (m, 1H, H-
30), 3.95 (m, 1H, H-40), 3.64–3.60 (m, 2H, H-50), 3.05 (m,
2H, NHCH2), 2.36 (t, 2H, J=7.21 Hz, ꢂCHCH2), 1.51–
1.40 (m, 4H, CH2CH2CH3), 1.16 (t, 3H, J=7.20 Hz,
NHCH2CH3), 0.89 (t, 3H, J=5.15 Hz, CH3); MS m/z
391 (M+H)+. Anal. calcd for C18H26N6O4: C, 54.21;
H, 6.81; N, 21.07. Found (ꢄ 0.5H2O) C, 54.10; H, 7.01;
N, 21.4.
General procedure for the introduction of an 1-hexynyl
group at derivatives 1 and 7–11 to obtain compounds 3
and 12–16. To a solution of 2-iodoadenosine (1) or the
appropriate 8-(ar)alkylamino-2-iodoadenosine (7–11)
(0.65 mmol) in dry acetonitrile (5 mL) and Et3N (5 mL)
under an atmosphere of nitrogen were added CuI
(9.3 mg, 48.8 mmol), PdCl2 (6.0 mg, 33.8 mmol), Ph3P
(19.5 mg, 74.3 mmol) and 1-hexyne (3.15 mmol, 362 mL).
The mixture was stirred overnight at room temperature
under an atmosphere of nitrogen. The mixture was fil-
tered, concentrated and purified by column chromato-
graphy.
2-(1-Hexynyl)-8-propylaminoadenosine (14). The reac-
tion was performed with 2-iodo-8-propylaminoadeno-
sine (9, 68 mg, 0.15 mmol) and 1-hexyne (0.73 mmol).
The mixture was purified by column chromatography
(EtOAc–10% MeOH in EtOAc). Yield 49 mg
(0.12 mmol, 80%), mp 184–186 ꢁC; Rf 0.60 (10% MeOH
in EtOAc); 1H NMR (DMSO-d6) d 6.98 (t, 1H,
J=5.15 Hz, NH), 6.54 (bs, 2H, NH2), 5.87 (d, 1H,
J=7.55 Hz, H-10), 5.75 (t, 1H, J=4.46 Hz, OH-20), 5.26
(d, 1H, J=6.86 Hz, OH-50), 5.15 (d, 1H, J=4.12 Hz,
OH-30), 4.55 (q, 1H, J=6.86 Hz, H-20), 4.06 (m, 1H, H-
30), 3.96 (bs, 1H, H-40), 3.61 (bs, 2H, H-50), 2.49 (m, 2H,
NHCH2), 2.36 (m, 2H, ꢂCCH2), 1.60–1.42 (m, 6H,
NHCH2CH2CH3, CH2CH2), 0.89 (t, 6H, J=7.55 Hz, 2
ꢃ CH3); MS m/z 405 (M+H)+. Anal. calcd for
C19H28N6O4: C, 52.63; H, 7.26; N, 19.38. Found
(ꢄ 1.6H2O) C, 52.70; H, 7.23; N, 19.19.
2-(1-Hexynyl)adenosine (3).12 The reaction was per-
formed with 2-iodoadenosine (1, 255 mg, 0.65 mmol)
and 1-hexyne (3.15 mmol). The mixture was purified by
column chromatography (5% MeOH in CH2Cl2). Yield
192 mg (0.55 mmol, 85%), mp 106–109 ꢁC; Rf 0.10 (10%
MeOH in CH2Cl2); 1H NMR (DMSO-d6) d 8.37 (s, 1H,
H-8), 7.41 (bs, 2H, NH2), 5.84 (d, J=6.18 Hz, 1H, H-

2190
E. W. van Tilburg et al. / Bioorg. Med. Chem. 11 (2003) 2183–2192
8-Butylamino-2-(1-hexynyl)adenosine (15). The reaction
was performed with 8-butylamino-2-iodoadenosine (10,
200 mg, 0.43 mmol) and 1-hexyne (2.08 mmol). The
mixture was purified by column chromatography (5%
MeOH in CH2Cl2). Yield 111 mg (0.26 mmol, 62%), mp
182–184 ꢁC; 1H NMR (DMSO-d6) d 6.95 (t, 1H,
J=5.49 Hz, NH), 6.55 (bs, 2H, NH2), 5.87 (d, 1H,
J=7.90 Hz, H-10), 5.77–5.72 (m, 1H, OH-20), 5.26 (d,
1H, J=6.52 Hz, OH-50), 5.15 (d, 1H, J=4.12 Hz, OH-
30), 4.54 (q, 1H, J=5.84 Hz, H-20), 4.09–4.04 (m, 1H, H-
30), 3.95–3.93 (m, 1H, H-40), 3.64–3.59 (m, 2H, H-50),
3.09 (m, 2H, NHCH2), 2.36 (m, 2H, ꢂCCH2), 1.51–1.30
(m, 8H, 2 ꢃ CH2CH2), 0.89 (t, 6H, J=7.55 Hz, 2 ꢃ
CH3); MS m/z 419 (M+H)+. Anal. calcd for
C20H30N6O4: C, 54.16; H, 7.45; N, 18.95. Found
(ꢄ 1.4H2O) C, 54.11; H, 7.72; N, 19.14.
¼CHCH2), 1.52–1.36 (m, 4H, CH2CH2), 0.94 (t, 3H,
J=6.52 Hz, CH3); MS m/z 379 (M+H)+; HPLC, Sys-
tem A: 20–100% CH3CN in H2O in 35 min, retention
time=6.15 min; System B: 30–100% CH3OH in H2O in
40 min, retention time=18.67 min.
8-Ethylamino-2-[(E)-1-hexenyl]adenosine (18). The reac-
tion was performed with 8-ethylamino-2-iodoadenosine
(8, 740 mg, 1.70 mmol) and (E)-1-(borocatechol)-1-hex-
ene (5.09 mmol). The mixture was purified by column
chromatography (10–20% MeOH in CH2Cl2).Yield
23mg (0.06 mmol, 5%), mp 128–130 ꢁC; Rf 0.61 (20%
MeOH in CH2Cl2); 1H NMR (MeOD-d4) d 6.92–6.82 (m,
1H, ¼CHCH2), 6.29 (d, 1H, J=14.04Hz, ¼CH), 6.02 (d,
1H, J=7.64 Hz, <H-10), 4.80–4.76 (m, 1H, H-20), 4.29–
4.27 (m, 1H, H-30), 4.13–4.11 (m, 1H, H-40), 3.81 (q, 2H,
J=10.89Hz, H-50), 3.42 (q, 2H, J=6.02 Hz, NHCH2),
2.24 (q, 2H, J=6.00 Hz, ¼CHCH2), 1.53–1.33 (m, 4H,
CH2CH2), 1.28 (t, 3H, J=7.72 Hz, NHCH2CH3), 0.94
(t, 3H, J=7.11 Hz, CH3); MS m/z 393 (M+H)+;
HPLC, System A: 20–100% CH3CN in H2O in 35 min,
retention time=7.42 min; System B: 30–100% CH3OH
in H2O in 40 min, retention time=20.16 min.
8-Benzylamino-2-(1-hexynyl)adenosine (16). The reac-
tion was performed with 8-benzylamino-2-iodoadeno-
sine (11, 230 mg, 0.46 mmol) and 1-hexyne (2.23 mmol).
The mixture was purified by column chromatography
(EtOAc). Yield 146 mg (0.32 mmol, 70%), mp 141–
143 ꢁC; Rf 0.26 (5% MeOH in EtOAc); 1H NMR
(DMSO-d6) d 7.64 (t, 1H, J=6.10 Hz, NH), 7.36–7.18
(m, 5H, phenyl), 6.59 (bs, 2H, NH2), 5.92 (d, 1H,
J=7.71 Hz, H-10), 5.77 (t, 1H, J=4.85 Hz, OH-20), 5.32
(d, 1H, J=6.74 Hz, OH-50), 5.18 (d, 1H, J=4.18 Hz,
OH-30), 4.64 (q, 1H, J=4.64 Hz, H-20), 4.58 (t, 2H,
J=4.65 Hz, NHCH2), 4.09 (t, 1H, J=4.09 Hz, H-30), 3.98
(d, 1H, J=1.60 Hz, H-40), 3.62 (q, 2H, J=2.58 Hz, H-50),
2.36 (t, 2H, J=6.92 Hz, ꢂCCH2), 1.55–134 (m, 4H,
CH2CH2), 0.89 (t, 3H, J=7.11Hz, CH3); MS m/z 453
(M+H)+. Anal. calcd for C23H28N6O4: C, 59.43; H, 6.71;
N, 18.65. Found (ꢄ 1.0H2O) C, 59.39; H, 6.69; N, 18.28.
2-[(E)-1-Hexenyl]-8-propylaminoadenosine (19). The
reaction was performed with 2-iodo-8-propylaminoade-
nosine (9, 400 mg, 0.89 mmol) and (E)-1-(borocatechol)-
1-hexene (2.67 mmol). The mixture was purified by col-
umn chromatography (10% MeOH in CH2Cl2). Yield
43 mg (0.11 mmol, 12%), mp 170–172 ꢁC; Rf 0.41 (10%
1
MeOH in CH2Cl2); H NMR (MeOD-d4) d 6.93–6.82
(m, 1H, ¼CHCH2), 6.29 (d, 1H, J=15.48 Hz, ¼CH),
6.04 (d, 1H, J=7.65 Hz, H-10), 4.79–4.74 (m, 1H, H-20),
4.29–4.26 (m, 1H, H-30), 4.13–4.12 (m, 1H, H-40), 3.81
(q, 2H, J=8.24 Hz, H-50), 3.37–3.32 (m, 2H, NHCH2),
2.24 (q, 2H, J=7.16 Hz,=CHCH2), 1.89–1.66 (m, 2H,
NHCH2CH2), 1.51–1.35 (m, 4H, CH2CH2), 1.02–0.91
(m, 6H, 2 ꢃ CH3); MS m/z 407 (M+H)+; HPLC, Sys-
tem A: 20–100% CH3CN in H2O in 35 min, retention
time=8.10 min; System B: 30–100% CH3OH in H2O in
40 min, retention time=20.25 min.
(E)-1-(borocatechol)-1-hexene.14 Yield 75%; Rf 0.28
(EtOAc/PE40/60 40/60=1:1).
General procedure for the introduction of an (E)-1-hex-
ene group at derivatives 1 and 7–11 to obtain compounds
3 and 17–21. To a solution of 2-iodoadenosine (1) or the
appropriate 8-(ar)alkylamino-2-iodoadenosine (7–11)
(0.82 mmol) in 20 mL of CH3CN–DMF (1:1) was added
50 mg of tetrakis(triphenylphosphine)palladium(0) and
the mixture was stirred at room temperature for 15 min.
Then 500 mg each of K2CO3 and (E)-1-(borocatechol)-
1-hexene (2.46 mmol) were added, and the suspension
was refluxed for 5 h. The mixture was filtered, con-
centrated in vacuo and purified by column chroma-
tography.
8-Butylamino-2-[(E)-1-hexenyl]adenosine (20). The reac-
tion was performed with 8-butylamino-2-iodoadenosine
(10, 380 mg, 0.82 mmol) and (E)-1-(borocatechol)-1-
hexene (2.46 mmol). The mixture was purified by col-
umn chromatography (10% MeOH in CH2Cl2). Yield
42 mg (0.10 mmol, 12%); Rf 0.44 (10% MeOH in
CH2Cl2); 1H NMR (MeOD-d4) d 6.94–6.80 (m, 1H,
¼CHCH2), 6.28 (d, 1H, J=15.42 Hz, ¼CH), 6.02 (d,
1H, J=7.62 Hz, H-10), 4.81–4.69 (m, 1H, H-20), 4.31–
4.20 (m, 1H, H-30), 4.20–4.05 (m, 1H, H-40), 3.87–3.70
(m, 2H, H-50), 3.35–3.23 (m, 2H, NHCH2), 2.29–2.05
(m, 2H, ¼CHCH2), 1.72–1.20 (m, 8H, 2 ꢃ CH2CH2),
1.00–0.83 (m, 6H, 2 ꢃ CH3); MS m/z 422 (M+H)+;
HPLC, System A: 20–100% CH3CN in H2O in 35 min,
retention time=8.53 min; System B: 30–100% CH3OH
in H2O in 40 min, retention time=20.47 min.
2-[(E)-1-Hexenyl]adenosine (4).14 2-[(E)-1-Hexenyl]-8-
methylaminoadenosine (17). The reaction was per-
formed with 2-iodo-8-methylaminoadenosine (7, 520mg,
1.23 mmol) and (E)-1-(borocatechol)-1-hexene (3.70
mmol). The mixture was purified by column chromato-
graphy (20% MeOH in CH2Cl2). Yield 42 mg
(1.11 mmol, 9%), mp 161–166 ꢁC; Rf 0.60 (20% MeOH
1
in CH2Cl2); H NMR (MeOD-d4) d 6.91–6.83 (m, 1H,
¼CHCH2), 6.29 (d, 1H, J=15.44 Hz, ¼CH), 5.98 (d,
1H, J=7.55 Hz, H-10), 4.81–4.74 (m, 1H, H-20), 4.29 (t,
1H, J=5.49 Hz, H-30), 4.13 (m, 1H, H-40), 3.82–3.79 (m,
2H, H-50), 2.97 (m, 3H, NHCH3), 2.29–2.22 (m, 2H,
8-Benzylamino-2-[(E)-1-hexenyl]adenosine (21). The
reaction was performed with 8-benzylamino-2-iodoade-
nosine (11, 410 mg, 0.82 mmol) and (E)-1-(boro-
catechol)-1-hexene (2.47 mmol). The mixture was

E. W. van Tilburg et al. / Bioorg. Med. Chem. 11 (2003) 2183–2192
2191
purified by column chromatography (10% MeOH in
CH2Cl2). Yield 41 mg (0.10 mmol, 11%), mp 153–
Briefly, assays were performed in 50/10/1 buffer [50 mM
Tris/10 mM MgCl2/1 mM ethylenediaminetetra-acetic acid
(EDTA) and 0.01% 3-([3-cholamidopropyl]-dimethyl-
ammonio)-1-propanesulfonate (CHAPS)] in glass tubes
and contained 50 mL of a HEK 293 cell membrane sus-
pension (10–30 mg), 25 mL [125I]AB MECA (final con-
centration 0.15 nM), and 25 mL of ligand. Incubations
were carried out for 1 h at 37 ꢁC and were terminated by
rapid filtration over Whatman GF/B filters, using a
Brandell cell harvester (Brandell, Gaithersburg, MD,
USA). Tubes were washed three times with 3 mL of
buffer. Radioactivity was determined in a Beckman
5500B g-counter. Nonspecific binding was determined
in the presence of 10ꢀ5 M R-PIA.
ꢁ
1
155 C; H NMR (MeOD-d4) d 7.64 (t, 1H, J=6.10 Hz,
NH), 7.36–7.18 (m, 5H, phenyl), 6.59 (bs, 2H, NH2),
6.94–6.80 (m, 1H, ¼CHCH2), 6.28 (d, 1H, J=15.42 Hz,
¼CH), 5.92 (d, 1H, J=7.71 Hz, H-10), 5.77 (t, 1H,
J=4.85 Hz, OH-20), 5.32 (d, 1H, J=6.74 Hz, OH-50),
5.18 (d, 1H, J=4.18 Hz, OH-30), 4.64 (q, 1H,
J=4.64 Hz, H-20), 4.58 (t, 2H, J=4.65 Hz, NHCH2),
4.09 (t, 1H, J=4.09 Hz, H-30), 3.98 (d, 1H, J=1.60 Hz,
H-40), 3.62 (q, 2H, J=2.58 Hz, H-50), 2.29–2.05 (m,
2H,=CHCH2), 1.72–1.20 (m, 4H, CH2CH2), 1.00–0.83
(m, 3H, CH3) ppm; MS m/z 456 (M+H)+; HPLC,
System A: 20–100% CH3CN in H2O in 35 min, reten-
tion time=9.61 min; System B: 30–100% CH3OH in
H2O in 40 min, retention time=26.10 min.
cAMP assay. A2A
2,8-di-(1-Hexynyl)adenosine (22). 8-Bromo-2-iodoade-
nosine (5, 150 mg, 0.32 mmol) was dissolved in 3 mL
CH3CN and 3 mL Et3N (dry). Then 4.5 mg CuI
(23.6 mmol), 2.9 mg PdCl2 (16.4 mmol) and 9.6 mg Ph3P
(36.4 mmol) were added. Subsequently, 1.55 mmol
(178 mL) 1-hexyne was added and the mixture was stirred
under nitrogen atmosphere at room temperature over-
night. The mixture was concentrated and purified by col-
umn chromatography (10% MeOH in CH2Cl2). Yield
96 mg (0.22 mmol, 70%), mp 146–148 ꢁC; Rf 0.48 (10%
MeOH in CH2Cl2); 1H NMR (MeOD-d4) d 7.61 (bs, 2H,
NH2), 5.89 (d, 1H, J=6.86 Hz, H-10), 5.39 (d, 1H,
J=6.18 Hz, OH-20), 5.29–5.22 (m, 1H, OH-50), 5.16 (d,
1H, J=4.12 Hz, OH-30), 4.94 (q, 1H, J=5.83 Hz, H-20),
4.15–4.13 (m, 1H, H-30), 3.94–3.92 (m, 1H, H-40), 3.72–
3.44 (m, 2H, H-50), 2.56 (t, 2H, J=6.52 Hz, ꢂCCH2),
2.44–2.34 (m, 2H, ꢂCCH2), 1.57–1.37 (m, 8H, CH2CH2),
0.93–0.86 (m, 6H, CH3) ppm; MS m/z 429 (M+H)+.
Anal. calcd for C22H29N5O4: C, 60.27; H, 7.18; N, 16.72.
Found [ꢄ 0.8HCON(CH3)2], C, 60.10; H, 7.00; N, 16.46.
CHO cells expressing the human adenosine A2A recep-
tors were grown overnight as a monolayer in 24 wells
tissue culture plates (400 mL/well; 2ꢃ105 cells/well).
cAMP generation was performed in Dulbecco’s Mod-
ified Eagles Medium (DMEM)/N-2-hydroxyethyl-
piperazin-N0-2-ethansulfonic acid (HEPES) buffer
(0.60 g HEPES/ 50 mL DMEM pH 7.4). To each well,
washed three times with DMEM/HEPES buffer
(250 mL), 100 mL DMEM/HEPES buffer, 100 mL adeno-
sine deaminase (final concentration 5 IU/mL) and
100 mL of a mixture of rolipram and cilostamide (final
concentration 50 mM each) were added. After incu-
bation for 40 min at 37 ꢁC, 100 mL of agonist-solution
was added. After 15 min at 37 ꢁC, the reaction was ter-
minated by removing the medium and adding 200 mL
0.1 M HCl. Wells were stored at ꢀ20 ꢁC until assay.
The amounts of cAMP were determined after a protocol
with cAMP binding protein17 with the following minor
modifications. As a buffer was used 150 mM K2HPO4/
10 mM EDTA/0.2% bovine serum albumin (BSA) at
pH 7.5. Samples (20 +30 mL 0.1 M HCl) were incubated
for at least 2.5 h at 0 ꢁC before filtration over Whatman
GF/B filters. Filters were additionally rinsed with 2 ꢃ
2 mL Tris–HCl buffer (pH 7.4, 4 ꢁC). Filters were coun-
ted in Packard Emulsifier Safe scintillation fluid
(3.5 mL) after 24 h of extraction.
2,8-di-Benzylaminoadenosine (23). 8-Bromo-2-iodoade-
nosine (5, 1.4 g, 2.34 mmol) was dissolved in benzyl-
amine (23.4 mmol, 2.56 mL). The mixture was heated at
140 ꢁC for 2 h and was poured into CHCl3. Awhite
precipitate was formed which was removed by filtration.
The filtrate was concentrated and purified by column
chromatography (0–10% MeOH in EtOAc). Yield
838 mg (1.76 mmol, 75%), mp 133–135 ꢁC; Rf 0.24 (5%
Data analysis
1
MeOH in EtOAc); H NMR (MeOD-d4) d 7.32–6.91
(m, 11H, 2 ꢃ phenyl, NH), 6.35–6.29 (m, 1H, NH), 6.05
(bs, 2H, NH2), 5.81 (d, 1H, J=6.86 Hz, H-10), 5.66–5.54
(m, 1H, OH-20), 5.20 (d, 1H, J=6.18 Hz, OH-50), 5.04 (d,
1H, J=4.80 Hz, OH30), 4.75–4.62 (m, 1H, H-20), 4.56–
4.38 (m, 4H, 2ꢃCH2), 4.13–4.04 (m, 1H, H-30), 3.92–3.84
(m, 1H, H-40), 3.65–3.52 (m, 2H, H-50) ppm; MS m/z 479
(M+H)+. Anal. calcd for C24H27N7O4: C, 60.37; H,
5.70; N, 20.53. Found C, 60.65; H, 5.37; N, 20.44.
Apparent Ki were computed from the displacement
curves by nonlinear regression of the competition curves
with the software package Prism (Graph Pad, San
Diego, CA, USA).
Acknowledgements
The authors thank Steve Rees (GlaxoWellcome,
Stevenage, UK) for providing CHO cells expressing
the human adenosine A2A receptor.
Radioligand binding studies
Measurements with [3H]DPCPX in the absence of GTP
were performed according to a protocol published pre-
viously.39 Adenosine A2A receptor affinities were deter-
mined according to Gao et al.40 Adenosine A3 receptor
affinities were determined essentially as described.37,41
References and Notes
1. Ralevic, V.; Burnstock, G. Pharmacol. Rev. 1998, 50, 413.
2. Jacobson, K. A.; Knutsen, L. J. S. In Handbook of

2192
E. W. van Tilburg et al. / Bioorg. Med. Chem. 11 (2003) 2183–2192
Experimental Pharmacology, Vol. 151/1: Purinergic and Pyr-
imidinergic Signalling I; Abbracchio, M. P., Williams, M., Eds:
Springer-Verlag, Berlin, Germany, 2001; p 129.
3. Fink, J. S.; Weaver, D. R.; Rivkees, S. A.; Peterfreund,
R. A.; Pollack, A. E.; Adler, E. M.; Reppert, S. M. Mol. Brain
Res. 1992, 14, 186.
22. Lin, T.-S.; Cheng, J.-C.; Ishiguro, K.; Sartorelli, A. C. J.
Med. Chem. 1985, 28, 1481.
23. Chattopadyaya, J. B.; Reese, C. B. Synthesis 1977, 725.
24. Volpini, R.; Costanzi, S.; Lambertucci, C.; Vittori, S.;
Lorenzen, A.; Klotz, K.-N.; Cristalli, G. Bioorg. Med. Chem.
Lett. 2001, 11, 1931.
4. Ferre
´
, S.; von Euler, G.; Johansson, B.; Fredholm, B. B.;
25. Francis, J. E.; Webb, R. L.; Ghai, G. R.; Hutchison, A. J.;
Moskal, M. A.; deJesus, R.; Yokoyama, R.; Rovinski, S. L.;
Contardo, N.; Dotson, R.; Barclay, B.; Stone, G. A.; Jarvis,
M. F. J. Med. Chem. 1991, 34, 2570.
26. Camaioni, E.; DiFrancesco, E.; Vittori, S.; Volpini, R.;
Cristalli, G. Bioorg. Med. Chem. 1997, 5, 2267.
27. Homma, H.; Watanabe, Y.; Abiru, T.; Murayama, T.;
Nomura, Y.; Matsuda, A. J. Med. Chem. 1992, 35, 2881.
28. Keeling, S. E.; Albinson, F. D.; Ayres, B. E.; Butchers,
P. R.; Chambers, C. L.; Cherry, P. C.; Ellis, F.; Ewan, G. B.;
Gregson, M.; Knight, J.; Mills, K.; Ravenscroft, P.; Reynolds,
L. H.; Sanjar, S.; Sheehan, M. J. Bioorg. Med. Chem. Lett.
2000, 10, 403.
29. Klotz, K.-N.; Camaioni, E.; Volpini, R.; Kachler, S.; Vit-
tori, S.; Cristalli, G. Naunyn-Schmiedeberg’s Arch. Pharmacol.
1999, 360, 103.
30. Vittori, S.; Camaioni, E.; Constanzi, S.; Volpini, R.;
Klotz, K.-N.; Cristalli, G. Nucleosides Nucleotides 1999, 18,
739.
31. Volpini, R.; Camaioni, E.; Costanzi, S.; Vittori, S.; Klotz,
K.-N.; Cristalli, G. Nucleosides Nucleotides 1999, 18, 2511.
32. Volpini, R.; Costanzi, S.; Lambertucci, C.; Taffi, S.; Vittori,
S.; Klotz, K.-N.; Cristalli, G. J. Med. Chem. 2002, 45, 3271.
33. Cristalli, G.; Camaioni, E.; DiFrancesco, E.; Eleuteri, A.;
Vittori, S.; Volpini, R. Nucleosides Nucleotides 1997, 16,
1379.
34. Gao, Z.-G. Allosteric Modulation of G Protein-coupled
Receptors’, 2000. Leiden University, Leiden, The Netherlands,
2000; pp 113.
35. Bruns, R. F. Can. J. Phys.iol. Pharmacol. 1980, 58, 673.
36. Jacobson, K. A. In Comprehensive Medicinal Chemistry,
Vol. 3: Membranes & Receptors; Emmet, J. C., Ed.; Pergamon:
Oxford, New York, 1990; p 601.
Fuxe, K. Proc. Natl. Acad. Sci. U.S.A. 1991, 88, 7238.
5. Sullivan, G. W.; Linden, J. Drug Dev. Res. 1998, 45, 103.
6. Broadley, K. J. Expert Opin. Ther. Pat. 2000, 10, 1669.
7. IJzerman, A. P.; Van der Wenden, E. M.; Von Frijtag
Drabbe Kunzel, J. K.; Mathoˆ t, R. A. A.; Danhof, M.; Borea,
P. A.; Varani, K. Naunyn-Schmiedeberg’s Arch. Pharmacol.
1994, 350, 638.
8. Lorenzen, A.; Sebastiao, A. M.; Sellink, A.; Vogt, H.;
Schwabe, U.; Ribeiro, J. A.; IJzerman, A. P. Eur. J. Pharma-
col. 1997, 334, 299.
9. Van der Graaf, P. H.; Van Schaick, E. A.; Visser, S. A. G.;
De Greef, H. J. M. M.; IJzerman, A. P.; Danhof, M. J. Phar-
macol. Exp. Ther. 1999, 290, 702.
10. Van Tilburg, E. W.; von Frijtag Drabbe Kunzel, J.;
Groote, M.; Vollinga, R. C.; Lorenzen, A.; IJzerman, A. P. J.
Med. Chem. 1999, 42, 1393.
11. Shryock, J. C.; Snowdy, S.; Baraldi, P. G.; Cacciari, B.;
Spalluto, G.; Monopoli, A.; Ongini, E.; Baker, S. P.; Belardi-
nelli, L. Circulation 1998, 98, 711.
12. Cristalli, G.; Eleuteri, A.; Vittori, S.; Volpini, R.; Lohse,
M. J.; Klotz, K.-N. J. Med. Chem. 1992, 35, 2363.
13. Matsuda, A.; Shinozaki, M.; Yamaguchi, T.; Homma, H.;
Nomoto, R.; Miyasaka, T.; Watanabe, Y.; Abiru, T. J. Med.
Chem. 1992, 35, 241.
14. Vittori, S.; Camaioni, E.; Di Francesco, E.; Volpini, R.;
Monopoli, A.; Dionisotti, S.; Ongini, E.; Cristalli, G. J. Med.
Chem. 1996, 39, 4211.
15. Van der Wenden, E. M.; Carnielli, M.; Roelen, H. C. P. F.;
Lorenzen, A.; von Frijtag Drabbe Kunzel, J. K.; IJzerman,
A . P.J. Med. Chem. 1998, 41, 102.
16. Van der Wenden, E. M.; von Frijtag von Drabbe Kunzel,
J. K.; Mathoˆ t, R. A. A.; Danhof, M.; IJzerman, A. P.; Sou-
dijn, W. J. Med. Chem. 1995, 38, 4000.
17. Van der Wenden, E. M.; Hartog-Witte, H. R.; Roelen,
H. C. P. F.; Von Frijtag Drabbe Kunzel, J. K.; Pirovano,
I. M.; Mathoˆ t, R. A. A.; Danhof, M.; Van Aerschot, A.;
Lidaks, M. J.; IJzerman, A. P.; Soudijn, W. Eur. J. Pharmacol.
Mol. Pharmacol. Sect. 1995, 290, 189.
37. van Galen, P. J. M.; Van Bergen, A. H.; Gallo-Rodriquez,
C.; Melman, N.; Olah, M. E.; IJzerman, A. P.; Stiles, G. L.;
Jacobson, K. A. Mol. Pharmacol. 1994, 45, 1101.
38. Van Tilburg, E. W.; von Frijtag Drabbe Kunzel, J.; de
Groote, M.; IJzerman, A. P. J. Med. Chem. 2002, 45, 420.
39. Pirovano, I. M.; IJzerman, A. P.; van Galen, P. J. M.;
Soudijn, W. Eur. J. Pharmacol. 1989, 172, 185.
18. Roelen, H.; Veldman, N.; Spek, A. L.; von Frijtag Drabbe
Kunzel, J.; Mathot, R. A.; IJzerman, A. P. J. Med. Chem.
1996, 39, 1463.
40. Gao, Z.-G.; IJzerman, A. P. Biochem. Pharmacol. 2000,
60, 669.
19. Van Schaick, E. A.; Tukker, H. E.; Roelen, H. C. P. F.;
IJzerman, A. P.; Danhof, M. Br. J. Pharmacol. 1998, 124, 607.
20. Robins, M. J.; Uznanski, B. Can. J. Chem. 1981, 59, 2601.
21. Ikehara, M.; Uesugi, S.; Kaneko, M. J. Chem. Soc. Chem.
Comm. 1967, 17.
41. Olah, M. E.; Gallo-Rodriquez, C.; Jacobson, K. A.; Stiles,
G. L. Mol. Pharmacol. 1994, 45, 978.
42. Liu, G.-S.; Downey, J. M.; Cohen, M. V. In Purinergic
Approaches in Experimental Therapeutics: Jacobson, K. A.,
Jarvis, M. F., Eds.; Wiley-Liss: New York, 1997; p 153.