Nucleoside-5′-monophosphates as prodrugs of adenosine A2A receptor agonists activated by ecto-5′-nucleotidase

Article abstract of DOI:10.1021/jm900538v

Prodrugs of adenosine A_2A receptor agonists were developed that are activated by ecto-5′-nucleotidase (ecto-5′-NT,CD73). Because ecto-5′-NT is upregulated in inflamed tissue, the A_2A agonists are expected to be released from their prodrug form at the sites of inflammation. 2-(Ar)alkyl-substituted AMP derivatives were synthesized and investigated. Certain 2-substituted AMP derivatives, including 2-hexylthio-AMP, 2-cyclopentylthio-AMP, 2-cyclohexylmethylthio-AMP, and 2-cyclohexylethylthio-AMP were accepted as substrates by ecto-5′-NT and readily converted to the corresponding 2-substituted adenosine derivatives. The 2-cyclohexylethylthio substitution was a good compromise between the requirements of the ecto-5′-NT and the adenosine A_2A receptor. The corresponding AMP derivative (12g) was a similarly good substrate as AMP itself, while the resulting adenosine derivative (11g) was a relatively potent A_2A agonist (radioligand binding to rat brain striatal membranes: K_i=372 nM; inhibition of anti-CD3/anti-CD28-induced IFN-γ release in mouse CD4+ cells: EC₅₀=50 nM). Compound 11g was released from 12g by incubation with CD4+ cells isolated from wild-type mice but only to a much smaller extent by cells from ecto-5′-NT knockout mice. Compound 12g will be a new lead structure for the development of more potent and selective ecto-5′-NT- activated prodrugs of selective anti-inflammatory A_2A receptor agonists.

Full text of DOI:10.1021/jm900538v

J. Med. Chem. 2009, 52, 7669–7677 7669
DOI: 10.1021/jm900538v
Nucleoside-50-monophosphates as Prodrugs of Adenosine A2A Receptor Agonists Activated
by ecto-50-Nucleotidase†
Ali El-Tayeb,‡,^ Jamshed Iqbal,‡,^ Andrea Behrenswerth,‡ Michael Romio, Marion Schneider,‡ Herbert Zimmermann,§
,‡
€
€
Jurgen Schrader, and Christa E. Muller*
‡PharmaCenter Bonn, Pharmaceutical Institute, Pharmaceutical Chemistry I, University of Bonn, An der Immenburg 4, D-53121 Bonn, Germany,
§AK Neurochemie, Biozentrum der J. W. Goethe-Universita€t, D-60438 Frankfurt am Main, Germany, and Department of Cardiovascular
Physiology, Heinrich-Heine-University Du€sseldorf, Universita€tsstrasse 1, 40225 Du€sseldorf, Germany. ^ A. El-Tayeb was on leave from the
University of Al-Azhar, Assiut, Egypt, and J. Iqbal was on leave from the Department of Pharmaceutical Sciences, COMSATS Institute of
Information Technology, Abbottabad, Pakistan
Received April 27, 2009
Prodrugs of adenosine A2A receptor agonists were developed that are activated by ecto-50-nucleotidase
(ecto-50-NT, CD73). Because ecto-50-NT is upregulated in inflamed tissue, the A2A agonists are expected
to be released from their prodrug form at the sites of inflammation. 2-(Ar)alkyl-substituted AMP
derivatives were synthesized and investigated. Certain 2-substituted AMP derivatives, including
2-hexylthio-AMP, 2-cyclopentylthio-AMP, 2-cyclohexylmethylthio-AMP, and 2-cyclohexylethylthio-
AMP were accepted as substrates by ecto-50-NT and readily converted to the corresponding
2-substituted adenosine derivatives. The 2-cyclohexylethylthio substitution was a good compromise
between the requirements of the ecto-50-NT and the adenosine A2A receptor. The corresponding
AMP derivative (12g) was a similarly good substrate as AMP itself, while the resulting adenosine
derivative (11g) was a relatively potent A2A agonist (radioligand binding to rat brain striatal
membranes: Ki=372 nM; inhibition of anti-CD3/anti-CD28-induced IFN-γ release in mouse CD4þ
cells: EC50=50 nM). Compound 11g was released from 12g by incubation with CD4þ cells isolated
from wild-type mice but only to a much smaller extent by cells from ecto-50-NT knockout
mice. Compound 12g will be a new lead structure for the development of more potent and selective
ecto-50-NT-activated prodrugs of selective anti-inflammatory A2A receptor agonists.
Introduction
found in the brain in the caudate-putamen, nucleus accum-
bens, and olfactory tubercle.5 One of the main therapeutic
potentials of adenosine A2A receptor agonists is due to their
anti-inflammatory and immunosuppressive effects.6 On the
other hand, A2A receptor agonists are potent vasodilators and
are applied as diagnostics for pharmacologic stress testing in
the heart.7 Further potential therapeutic applications of A2A
agonists are the treatment of psychosis and Huntington0s
disease.8
The main approach for discovering A2A-selective agonists
has been the modification of adenosine itself.9,10 Most of the
potent and selective adenosine derivatives are modified in the
2-position of the adenine moiety and/or in the 50-position of
the riboside as shown in Figure 1. N6,2-Disubstitution and
substitution in position 8 may also lead to potent A2A
agonists. Typical variations include the introduction of an
amide moiety at the 50-position, a tetrazole ring at the
40-position of the ribose ring, and large substituents at the
2-position of the purine moiety. Typical 2-substituents con-
tain a cyclic structure (aromatic or cycloalkyl) or a simple
lipophilic alkyl chain and a three-membered linker often
containing an electron-rich function (e.g., O, NH, triple bond)
attached to the purine ring (Figure 1). Additional features,
such as bulky substituents at N6 or alkylamino substitution at
the 8-position of the purine moiety, have been described.8-10
2-(Ar)alkylthioadenosine derivatives have also been investi-
gated as A2A receptor agonists,11,12 and some derivatives,
Adenosine (1, Figure 1) is a naturally occurring nucleoside
that plays an important role in physiological and pathophy-
siological processes through its action on G protein-coupled
adenosine receptors (P1 receptors).1 There are four subtypes
of adenosine receptors that have been cloned and character-
ized: A1, A2A, A2B, and A3. The A1 and A3 receptors are
known to couple to adenylate cyclase in an inhibitory manner
via Gi protein, whereas A2A and A2B receptor subtypes
stimulate the enzyme via Gs protein. In addition, coupling
to other second messenger systems, such as calcium or potas-
sium ion channels or phospholipase C, has been described.1,2
Recently, the first crystal structure of an adenosine receptor,
the human A2A receptor, in complex with the high affinity
antagonist 4-(2-[7-amino-2-(2-furyl)[1,2,4]triazolo[2,3-a][1,3,5]-
triazin-5-ylamino]ethyl)phenol (ZM241385), was obtained
3
˚
in 2.6 A resolution. Adenosine A2A receptors mediate
the potent immunosuppressive and hypotensive effects of
adenosine.4 They are highly expressed in cells of the immune
system, e.g., in spleen, thymus, and leukocytes, as well as on
smooth muscle cells, endothelium, and blood platelets.5
Furthermore, a high density of A2A receptor expression is
†Contribution to celebrate the 100th anniversary of the Division of
Medicinal Chemistry of the American Chemical Society.
*To whom correspondence should be addressed. Phone: þ49-228-73-
2301. Fax: þ49-228-73-2567. E-mail: christa.mueller@uni-bonn.de.
r
2009 American Chemical Society
Published on Web 07/06/2009
pubs.acs.org/jmc

7670 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
El-Tayeb et al.
Figure 1. Structures of some potent and selective adenosine A2A receptor agonists.
e.g., 2-(phenylethylthio)adenosine, were found to be potent
A2A receptor agonists.12
and ecto-50-NT is generally up-regulated at the site of inflam-
mation. It has been shown that ecto-50-NT is up-regulated on
the neointima after wire-induced injury in the carotid artery.16
In addition, ecto-50-NT is highly up-regulated in the plaque of
ApoE mouse, a well established model of arteriosclerosis.17 In
addition to hydrolysis by ecto-50-NT, AMP can also be
dephosphorylated by alkaline phosphatase. While the
Km value for the ecto-50-NT is about 14 μM, the correspond-
ing value for alkaline phosphatase is more than 30-fold higher
(441 μM).18,19 In addition, at physiological pH, the AMP-
hydrolyzing activity of alkaline phosphatase is considerably
lower than that of ecto-50-NT.20
The goal of the present study was to develop prodrugs of
adenosine A2A receptor agonists, which as such are biolo-
gically inactive and require for activity conversion by ecto-
50-NT to the corresponding drugs. The rate of activation can
be expected to be high in tissues with a high expression of the
enzyme, e.g., inflamed tissue. This might allow high drug
levels in pathological tissues and thereby would result in a
drug targeting effect, increasing effects (anti-inflammation) at
the desired site of actions while decreasing unwanted side
effects (vasodilation).
A major limitation in the use of A2A adenosine receptor
agonists when applied systemically as anti-inflammatory
agents is their potent hypotensive activity.8,13 In an attempt
to dissociate the vasodilatory from the anti-inflammatory
activity we have synthesized in this study phosphorylated
A2A adenosine receptor agonistswhichrequireecto-50-nucleo-
tidase (ecto-50-NT, CD73,a EC 3.1.3.5) to be converted into
the active drug.
Dephosphorylation of extracellular AMP to adenosine is
physiologically mediated by ecto-50-NT, which is the last step
in the enzymatic chain that catalyzes the breakdown of
extracellular ATP to adenosine.1 ecto-50-NT is attached via
a glycosylphosphatidylinositol anchor to the plasma mem-
brane, where it catalyzes the hydrolysis of nucleoside
50-monophosphates such as AMP, GMP, and UMP to the
corresponding nucleosides. The enzyme can also be cleaved
off from its anchor and retains its catalytic activity in the
soluble form. The main physiological function of ecto-50-NT
is the hydrolysis of extracellular AMP formed by the degrada-
tion of the P2 receptor agonists ATP and ADP by other ecto-
nucleotidases. Thus, the enzyme generates adenosine, which
can act on P1 (adenosine) receptors.14 ecto-50-NT is localized
on the outer surface of endothelial cells of large and small
arteries. Hypoxia enhances the expression of ecto-50-NT,15
Results and Discussion
As target structures, we selected 2-(ar)alkylthio-substituted
AMP derivatives. They allow for a broad variation of the
2-substituent in order to study structure-activity relation-
ships of this class of compounds as potential substrates of
ecto-50-NT. Furthermore, they permit study of the affinity
and selectivity of the corresponding nucleosides for adenosine
A2A receptors. Certain 2-substituted AMP derivatives have
previously been found to be agonists or antagonists at P2Y
receptor subtypes, but their potency was low (in the micro-
molar range).21a-21c Therefore, this should not limit the
feasibility of the designed prodrug concept.
a Abbreviations: CCPA, 2-chloro-N6-cyclopentyladenosine; CE, ca-
pillary electrophoresis; CD73, cluster of differentiation 73; [3H]-
CGS21680, [3H]2-(4-(2-carboxymethyl)phenyl)ethylamino-50-N-ethyl-
carboxamidoadenosine; CHO, Chinese hamster ovary; ecto-50-NT,
ecto-50-nucleotidase; ESI, electrospray ionization; DEAE, diethylami-
noethyl; IFN-γ, interferon-γ; MSX-2, 3-(3-hydroxypropyl)-7-methyl-8-
(m-methoxystyryl)-1-propargylxanthine; NECA, N-ethyl-carboxami-
doadenosine; [3H]PSB-11, [3H]8-ethyl-4-methyl-2-phenyl-(8R)-4,5,7,8-
tetrahydro-1H-imidazo[2,1-i]purin-5-one; TCR, T cell receptor; TEAB,
triethylammonium hydrogencarbonate buffer; Treg, regulatory T-cells.

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7671
Scheme 1. Synthesis of Nucleoside Monophosphatesa
a Reagents and conditions: (a) three steps: (1) H2O2, CH3COOH; (2) 5-N aq NaOH; (3) CS2, MeOH, H2O, 120 ꢀC autoclave, 5 h. (b) R-Br, H2O/
NaOH, rt, 3-7 h. (c) (OCH3)3PO, POCl3, 1,8-bis-(dimethylamino)naphthaline, 5 h, 0-4 ꢀC. (d) Triethylammonium hydrogenbicarbonate buffer
pH 7.4-7.6, rt, 1 h.
Syntheses. 2-Thioadenosine (10) was synthesized starting
from adenosine (1) according to published procedures.22
Oxidation with hydrogen peroxide in acetic acid yielded
the N1-oxide. Subsequent ring-opening using sodium hydro-
xide, followed by treatment with a mixture of carbon disul-
fide, methanol, and water (50 : 175 : 25) at 120 ꢀC in an
autoclave, gave 2-thioadenosine (10).22 2-Alkylated deriva-
tives 11a-l were obtained by reaction of 10 with alkyl or
arylalkyl halogenides in the presence of sodium methoxide in
DMF or NaOH in water11,23 (Scheme 1; for details see
Supporting Information).
The 2-substituted adenosine derivatives 11a-l were sub-
jected to phosphorylation according to the Ludwig proce-
dure24 with minor modifications (Scheme 1): The lyophilized
nucleosides 11a-l were dissolved in trimethylphosphate and
reacted with phosphorus oxychloride in the presence of 1,8-bis-
(dimethylamino)naphthaline (“proton sponge”) to yield the
reactive 50-dichlorophosphate intermediates. Hydrolysis with
triethylammonium hydrogen bicarbonate buffer yielded the
desired nucleoside monophosphates 12a-l.
lower affinity versus the antagonist radioligand than versus
the agonist radioligand.25 Selectivity versus other adenosine
receptor subtypes was assessed by performing radioligand
binding studies at rat brain cortical membranes (A1) using
[3H]CCPA as a radioligand and at membrane preparations
of human recombinant A3 receptors expressed in Chinese
hamster ovary (CHO) cells using [3H]PSB-11 as a radioli-
gand. A capillary electrophoresis-based enzyme assay14 was
used to investigate the properties of the nucleoside mono-
phosphates as substrates of ecto-50-NT. Catalytically active
recombinant soluble glutathione-S-transferase/rat ecto-50-
NT fusion protein was expressed in insect cells with the
baculovirus system and purified by affinity chromatography
using agarose-coupled glutathione as previously described.26
Furthermore, selected compounds were investigated for
their potency to inhibit interferon-γ (IFN-γ) release induced
by anti-CD3/anti-CD28 antibodies in CD4þ T-cells purified
from the spleen of C57/BL6 mice (wild-type, and ecto-50-NT
knockout, respectively) by magnetic beads.
Adenosine Receptor Affinity. All 2-alkylthio- or arylalk-
ylthio-substituted adenosine derivatives investigated showed
higher affinity at A2A receptors when determined versus the
agonist radioligand [3H]CGS21680 than versus the antagonist
radioligand [3H]MSX-2 (see Table 1). The differences were 3-
to 14-fold. This means that all compounds had a higher affinity
to the (active) agonist-labeled conformation than to the (in-
active) antagonist-labeled conformation, indicating that they
are all agonists rather than antagonists at A2A receptors.
The determined structure-activity relationships confir-
med previous observations with 2-substituted adenosine
derivatives. Small alkyl groups linked by a thioether group
to the 2-position of adenosine resulted in only weak A2A
receptor agonists with Ki values in the low micromolar range
(11a,b,c). An increase in the chain length from propyl (11a) to
hexyl (11d) increased A2A affinity by almost 3-fold (11d, Ki=
542 nM). However, 11d was even somewhat more potent at
the A1 receptor (Ki=256 nM). A cyclohexyl residue (in 11e)
instead of n-hexyl (11d) led to a 5-fold reduction in affinity.
The introduction of a methylene or, even better, an ethylene
spacer between the cyclohexyl ring and the thioether group
led to an increase in A2A affinity (2-(cyclohexylmethylthio)
adenosine, 11f, Ki=709 nM; 2-(cyclohexylethylthio)adeno-
sine, 11g, Ki=372 nM). A cyclopentyl analogue was slightly
The nucleotides 12a-l were purified by anion exchange
chromatography on Sephadex diethylaminoethyl (DEAE)
A-25 gel using a fast protein liquid chromatography (FPLC)
apparatus by applying a linear gradient (0-100%, pure
water f 0.5 M aqueous triethylammonium hydrogencarbo-
nate buffer). The neutral impurities (e.g., nucleosides) eluted
first, followed by the monophosphates. The products were
further purified by high performance liquid chromatography
(HPLC) on reverse-phase C18 material in order to remove
inorganic phosphates and buffer components.
The structures of the synthesized nucleosides and nucleo-
tides were confirmed by 1H and 13C NMR spectroscopy, in
addition to HPLC analysis coupled to electrospray ioniza-
tion mass spectrometry (LC/ESI-MS) performed in both
positive and negative mode. The nucleotides were addition-
ally investigated by 31P NMR spectroscopy.
Biological Activity. The 2-substituted (ar)alkylthioadeno-
sine derivatives were investigated in radioligand binding
studies at rat brain striatal adenosine A2A receptors using
the agonist radioligand [3H]CGS21680. Most compounds
were additionally tested versus the A2A antagonist radioli-
gand [3H]MSX-2 in order to assess whether the compounds
act as agonists or antagonists. Agonists show a markedly

7672 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
El-Tayeb et al.
Table 1. Adenosine Receptor Affinities of Nucleosides and Percent Dephosphorylation of Corresponding Nucleotides (50-Monophosphates) by ecto-
50-Nucleotidase in Comparison with the Physiological Substrate AMP
a Agonist radioligand. b Antagonist radioligand. c Affinities of the agonists will be underestimated because an antagonist radioligand was used for the
determination. d Percentage of dephosphorylation of nucleoside monophosphates was calculated with respect to the dephosphorylation of AMP, set at
100%. The peak areas of the substrate peak (nucleoside monophosphate) and the product peak (nucleoside) were measured and quantified in the same
less potent (11h, Ki=793 nM). Oxygen-containing ring ana-
logues of 11g (11k, 2-(1,4-dioxanyl) instead of cyclohexyl)
and 11l, 2-(1,3-dioxolanyl), led to a decrease in A2A affinity.
As previously reported, the introduction of a phenyl ring
increased A2A affinity, the most potent A2A agonist of the
present series being 2-(phenylethylthio)adenosine (11j, Ki
rat A2A=18.7 nM) exhibiting 10-fold selectivity vs the rat A1
receptor. The same compound had previously been investi-
gated by Volpini et al.12 at the human adenosine receptor
subtypes, where it was somewhat less potent and less selective
for the A2A receptor, indicating species differences between rat
and human adenosine receptors (see Table 1). All compounds,
except for 2-(benzylthio)adenosine (11i, Ki A3=120 nM),
showed only moderate to low affinity for the adenosine A3
receptor subtype. Affinity for the adenosine A2B receptor
subtype was not determined since Volpini et al. had shown
that A2B affinity cannot be expected for such structures.12
Hydrolysis of Nucleoside Monophosphates by Purified ecto-
50-Nucleotidase. As a next step, we investigated the properties of
the corresponding 2-substituted AMP derivatives as substrates
of rat ecto-50-NT. The percentage of dephosphorylation of
nucleoside monophosphates was calculated with respect to
the dephosphorylation of AMP, set at 100%. The enzyme
assays were performed in reaction vials, and after the reac-
tion was stopped, substrate and product were separated by
capillary electrophoresis and quantified by their UV absor-
bance at 232 nm. The results are collected in Table 1,
compounds 12a-l.
Short, unsaturated substituents, such as in 2-(allylthio)-
AMP (12b) and in 2-(propargylthio)-AMP (12c), were not
tolerated by the enzyme. Similarly, aromatic residues as in
2-(benzylthio)-AMP (12i) and in 2-(phenylethylthio)-AMP
(12j) prevented dephosphorylation of the nucleotides by
ecto-50-NT. Saturated alkyl or cycloalkyl residues linked
via a thioether bridge to the 2-position of AMP as in com-
pounds 12a (propyl, 47% dephosphorylation), 12d (hexyl,
84%), 12h (cyclopentyl, 71%), and 12e (cyclohexyl, 28%)
were tolerated by the enzyme. It is surprising that a longer
chain (hexyl) was better accepted than a small residue
(propyl), indicating that the ecto-50-NT possesses a lipophilic

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7673
binding pocket in that area. Even longer and bulkier
cyclohexylmethylthio- or cyclohexylethylthio-AMP deri-
vatives (12f, 12g) turned out to be good substrates for
ecto-50-NT. In contrast to the adenosine A2A receptor, the
enzyme tolerated oxygen-containing rings as in dio-
xanylethylthio- and dioxolanylethylthio-AMP derivatives
12k and 12l.
These results showed that the SARs of 2-substituted
adenosine derivatives at adenosine A2A receptors and those
of corresponding 2-substituted AMP derivatives as sub-
strates of ecto-50-NT were very different. The best com-
promise for both targets was the cyclohexylethylthio substi-
tuent. The respective nucleotide 12g was an excellent sub-
strate of ecto-50-NT, almost as good as AMP, while the
corresponding nucleoside 11g was still a quite potent adeno-
the phosphorylated derivative (12g) was shifted to the right by
about 2 orders of magnitude (EC50 5.70 ( 2.33 μM).
The degree of hydrolysis of the prodrug 12g was measured
at the end of the experiment. For comparison, degradation
of the prodrug by Treg cells isolated from ecto-50-NT
knockout mice was assessed under the same conditions.
The concentrations of prodrug 12g and of released nucleo-
side 11g were determined by LC/MS with electrospray
ionization (ESI). Three different starting concentrations of
prodrug 12g were investigated. A linear correlation between
the initial concentration of prodrug 12g added and the
amount of drug 11g formed during the incubation was
observed in cells from wild-type as well as from knockout
mice (see Figure 3).
From the results given in Table 2, it is evident that during
an incubation period of 18 h, 10.8-15.0% of prodrug
12g was dephosphorylated to the corresponding nucleoside
11g by CD4þ T-cells isolated from wild-type mice. In CD4þ
T-cells from knockout mice, the degree of hydrolysis was
much smaller (2.7-5.3% hydrolysis), indicating that ecto-
50-NT but not the alkaline phosphatase was mainly respon-
sible for converting the phosphate prodrug 12g to the
drug 11g.
Purified CD4þ T-cells consist of T-effector and regulatory
T-cells (Treg). Recently, it has been shown that only Treg
possesses ecto-50-NT, which is a novel marker enzyme be-
sides CD39 and FoxP3 for these cells.27 The ability of
nucleotide 12g to inhibit IFN-γ formation will depend on
the following critical factors: First, on the activity of ecto-
50-NT, in our experiments the fraction of Treg cells in
purified CD4þ T-cells was only about 20%. Furthermore,
the nucleotide was added together with stimulatory antibo-
dies and the accumulation of the nucleoside occurred linearly
over time. Consistent with this, preincubation of cells with
the nucleotide 12g considerably shifted the dose-response
curve to the left (unpublished observation). Second, the
distribution space into which the nucleoside is formed is of
importance because this space determines the steady-state
concentration of the active drug. In our cell experiments, the
ratio of cell to supernatant fluid space was about 4ꢀ105 cells
in 200 μL volume, which translates into a volume ratio of
1:2800, which in the case of Treg increases to 1:13000.
Consequently, at a given cellular ecto-50-NT activity, the
effective concentration of the nucleoside will increase faster
over time when the cell/fluid volume ratio is low. Under
pathological conditions, e.g., in an inflamed tissue with
infiltrated immune cells such as Treg cells and monocytes,
all of which express ecto-50-NT, the distribution space is the
sine A2A receptor agonist (K=372 nM) although not selective
i
versus the A1 adenosine receptor subtype. Nevertheless,
we selected these derivatives for further proof-of-principle
studies.
Interferon γ Production in T Cells. One selected nucleotide
(12g) and the corresponding nucleoside (11g) were investi-
gated in murine T cells for their inhibitory effect on T cell
receptor (TCR)-mediated IFN-γ production. To this end,
CD4þ T-cells were purified from the spleen of C57/BL6 mice
by magnetic beads and were stimulated with anti-CD3 and
anti-CD28 for 18 h. Measurement of IFN-γ production as
determined by ELISA in the supernatant was 2463 ( 600 pg/
mL (n=4). No IFN-γ could be detected in unstimulated cell.
As shown in Figure 2, cyclohexylethyadenosine (11g) potently
inhibited the IFN-γ formation. The EC50 was 50.0 ( 21.4 nM,
which is in the same range as the Ki value determined in the
receptor assay (372 ( 57 nM). The dose-response curve for
Figure 2. Effect of 2-(cyclohexylethylthio)adenosine (11g, n=2)
and of 2-(cyclohexylethylthio)-AMP (12g, n=3) on TCR-mediated
IFN-γ production in murine CD4þ T cells. Data show means ( SD.
Figure 3. Correlation between the concentration of added prodrug 12g (mM) and the produced drug 11g ( μM): the results showed that there
was a linear relationship between the concentration of the prodrug 12g and the released drug 11g in CD4þ T-cells from wild-type and ecto-
50-NT knockout mice.

7674 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
El-Tayeb et al.
Table 2. Quantitative Determination of the Nucleoside (11g) and Nucleotide (12g) Concentrations by LC-MS after Incubation with CD4þ Cells from
Wild-Type and ecto-50-NT Knockout Mice
LC/MS detection of prodrug 12g
conc. of 12g (μM) % of 12g relative to 11g
LC/MS detection of nucleoside 11g
initial concentration of nucleotide 12g (mM)
conc. of 11g (μM)
% of 11g relative to 12g
Wild-Type Mice
10
5
5.33
3.05
1.90
85.0
87.2
89.2
0.94
0.45
0.23
15.0
12.8
10.8
2.5
Knockout Mice
10
5
7.46
4.47
2.84
94.7
95.3
97.3
0.42
0.22
0.08
5.3
4.7
2.7
2.5
interstitial space only, which is relatively small. Therefore,
the curves shown in Figure 2 functionally may reflect the
other extreme: the situation in a blood perfused vessel. Here
the vascular distribution space of the nucleoside is rather
large. Because ecto-50-NT is expressed only on the vascular
endothelium and circulating immune cells,28 this would lead
to substantial dilution. Consequently, the plasma concentra-
tions of the dephosphorylated purine derivative are expected
to be lower than those of its precursor, which may result in
reduced vasodilatory side effects.
It should be noted that the prodrug can also be degraded
by alkaline phosphatase. Given the Km value of the alkaline
phosphatase to be substantially higher than that of ecto-
50-NT, it is expected that at low substrate concentrations
hydrolysis by ecto-50-NT will dominate. Little is known,
however, on the relative expression of both enzymes in
various target tissues. In support for the therapeutic concept
proposed here is that Treg cells only express negligible levels
of alkaline phosphatase (unpublished observation), which
is consistent with data given in Table 2. Another enzyme,
the membrane-bound splice variant of prostatic acid phos-
phatase (TM-PAP), has recently been shown to carry ecto-50-
nucleotidase activity.29 This enzyme is widely expressed in
mouse tissues such as prostate lobes, salivary gland, thymus,
lung, kidney, brain, spleen, and thyroid, and future experi-
ments need to define its physiological implications.29b
used without further purification. Solvents were used without
additional purification or drying unless otherwise noted. The
reactions were monitored by thin layer chromatography (TLC)
using aluminum sheets with silica gel 60 F254 (Merck). Column
chromatography was carried out with silica gel 0.060-0.200 mm,
pore diameter ca. 6 nm. Mass spectra were recorded on an API
2000 (Applied Biosystems, Darmstadt, Germany) mass spectro-
meter (turbo ion spray ion source) coupled with an HPLC system
(Agilent 1100) using a Phenomenex Luna 3 μ C18 column. 1H-,
31P-, and 13C NMR spectra were performed on a Bruker Avance
500 MHz spectrometer. DMSO-d6, MeOD-d4, or D2O were used
as solvents as indicated below. 31P NMR spectra were recorded
at rt; orthophosphoric acid (85%) was used as an external
standard. Shifts are given in ppm relative to the external standard
(31P NMR) or relative to the remaining protons of the deuterated
solvents used as internal standard(1H, 13C). Purity of the prepared
nucleosides was checked by TLC on silica gel 60 F254 (Merck)
aluminum plates, using dichloromethane:methanol (9:1) as the
mobile phase. Purityof the prepared nucleotides was confirmed by
HPLC by dissolving 1 mg/mL in H2O:MeOH=1:1, containing
2 mM ammonium acetate. A sample of 10 μL was injected into an
HPLC instrument (Agilent 1100) using a Phenomenex Luna 3 μ
C18 column. Elution was performed with a gradient of water:
methanol (containing 2 mM ammonium acetate) from 90:10 to
0:100 for 30 min at a flow rate of 250 μL/min, starting the gradient
after 10 min. UV absorption was detected from 220 to 300 nm
using a diode array detector. The purity of the nucleotide samples
was checked in a second chromatography system using capillary
electrophoresis (CE, for details see Supporting Information). The
purity of the products was generally g95%.
Conclusions
General Procedure for the Synthesis of 2-Thioadenosine Deri-
vatives. 2-Thioadenosine22 (1 mmol) was dissolved in 20 mL of
water (or water:ethanol; 1:1), and 5 mL of sodium hydroxide
(0.5 N) was added to the reaction mixture, followed by the
addition of (ar)alkyl halide (1.2 mmol) 10 min later. The reaction
mixture was stirred for 3-7 h at rt, and the completion of the
reaction was determined by TLC (CH2Cl2:MeOH=9:1). The
crude product was extracted by ethyl acetate, and the organic
phase was separated, dried over anhydrous sodium sulfate, and
evaporated to drynessunder reduced pressure. The crude product
was crystallized from ethanol and, in some cases, the product was
further purified by silica gel column chromatography (CH2Cl2:
MeOH=9:1) to afford the pure product.
2-(Cyclohexylethylthio)-AMP (12g) has been identified as
anexcellent substrateofecto-50-NT. Itis dephosphorylated by
ecto-50-NT at a similar rate as AMP to the corresponding
adenosine derivative 11g, an adenosine receptor agonist with
submicromolar affinity (Ki=372 nM) for the A2A receptor.
Immunecellssuchasmonocytes/macrophages, Treg cells, and
neutrophils express both ecto-50-NT and adenosine A2A
receptors at high density and thereby may be the preferential
target for immunosuppression by the proposed prodrug con-
cept. AMP derivative 12g will serve as a new lead compound
for the development of novel prodrugs of anti-inflammatory
adenosine A2A receptor agonists, which may be preferably
activated in inflamed tissues with high expression of ecto-
50-NT. Further studies will be aimed at improving the A2A
affinity and selectivity while keeping the ecto-50-NT substrate
properties; this includes testing of these novel compounds
under pathological conditions in the intact animal.
General Procedure for Phosphorylation. Preparation of
Triethylammonium Hydrogen Carbonate Buffer (TEAB). A 1 M
solution of TEAB was prepared by bubbling CO2 through a 1 M
triethylamine solution in water at 0-4 ꢀC for several hours
(pH approximately 7.4-7.6).30
General Procedure for the Preparation of Monophosphates by
Phosphorylation of Nucleosides (or Analogues). Lyophilized
nucleosides (1 mmol) were dissolved in 5 mL of trimethyl
Experimental Section
˚
phosphate (dried over 10 A molecular sieve). The mixture was
stirred at rt under argon and then cooled to 4 ꢀC. Dry 1,8-bis-
(dimethylamino)naphthaline (proton sponge, 0.32 g, 1.5 mmol)
All commercially available reagents were obtained from var-
ious producers (Acros, Aldrich, Fluka, Merck, and Sigma) and

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7675
was added, followed by 0.20 g (1.3 mmol) of POCl3 5 min later.
After 5 h of stirring at 0-4 ꢀC, the mixture was poured into a
cold 0.5 M aqueous TEAB solution (10 mL, pH 7.5) and stirred
at 0-4 ꢀC for several min. The solution was allowed to reach
rt upon stirring and then left standing for 1 h. Trimethyl-
phosphate was extracted with tert.-butylmethyl ether, and the
aqueous solution was lyophilized to yield glassy colorless oils.
The reactions were controlled by TLC using a freshly pre-
pared solvent system (2-propanol:NH4OH (25% of ammonia
in water):water=6:3:1). TLC plates were dried before UV
absorption was detected, and the plates were subsequently
sprayed with a phosphate reagent.30,31
31P NMR (202 MHz, D2O) δ 0.51 (s). 13C NMR (125 MHz, D2O)
δ27.71, 28.02, 29.86, 34.65, 38.61, 38.70, 66.16, 72.75, 76.49, 86.18,
88.85, 117.05, 140.15, 150.25, 156.91, 168.44. LC/ESI-MS: nega-
tive mode 488 ([M - H]-), positive mode 490 ([M þ H]þ).
2-(Phenylethylthio)adenosine-50-monophosphate (12j). 1H NMR
(500 MHz, MeOD) δ 3.05-3.08 (t, 2H, J=7.72 Hz), 3.38-3.49
(m, 2H), 3.76-3.90 (m, 2H), 4.17 (q, 1H, J=3.25 Hz), 3.35 (q, 1H,
J=4.34 Hz), 4.75 (t, 1H, J=4.75 Hz), 6.03 (d, 1H, J=5.67 Hz),
7.20-7.36 (m, 5H), 8.24 (s, 1H). 31P NMR (202 MHz, MeOD) δ
0.54 (s). 13C NMR (125 MHz, MeOD) δ 33.84, 37.52, 63.51, 72.58,
75.70, 87.72, 90.65, 123.02, 127.56-130.88, 140.77, 150.25, 157.19,
166.92. LC/ESI-MS: negative mode 482 ([M-H]-), positive mode
484 ([M þ H]þ).
Purification of Nucleotides Using Ion Exchange Chromatog-
raphy. Nucleoside 50-monophosphates were purified by ion ex-
change chromatography on an FPLC instrument (AKTA FPLC,
2-(Dioxanylethylthio)adenosine-50-monophosphate (12k). 1H
NMR (500 MHz, D2O) δ 1.39-1.4 (m, 2H), 2.12-2.15 (m,
2H), 3.02-3.05 (t, 2H, J=7.40 Hz), 3.78-3.83 (m, 2H), 3.86 (m,
1H), 3.97 (m, 2H), 4.06-4.10 (m, 2H), 4.32 (m, 1H), 4.47 (t, 1H,
J=4.57 Hz), 4.66 (m, 1H), 6.11 (d, 1H, J=5.99 Hz), 8.44 (s, 1H).
31P NMR (202 MHz, D2O) δ 2.77 (s). 13C NMR (125 MHz,
DMSO-d6) δ 27.94, 28.38, 37.27, 55.65, 69.94, 73.55, 77.15,
87.42, 89.64, 104.19, 119.20, 142.24, 153.23, 156.83, 167.67.
LC/ESI-MS: negative mode 492 ([M - H]-), positive mode 494
([M þ H]þ).
from Amersham Biosciences) with an XK 26 mm/20 cm length
column (Pharmacia) using Sephadex DEAE A-25 gel HCO3
-
-
form swelled in a 1 M solution of TEAB at 4 ꢀC. After equilibra-
tion of the column with deionized water, the crude product was
dissolved in 2 mL of aqueous triethylammonium hydrogen car-
bonate buffer. The column was washed with deionized water,
followed by a solvent gradient of water:TEAB 1 M buffer from 0
to 100% using approximately 500 mL of solvent to elute the
monophosphates. Fractions were collected and appropriate frac-
tions pooled, diluted in water, and lyophilized.
Purification of Monophosphates Using Preparative HPLC.
Lyophilized nucleoside 50-monophosphates were dissolved in
5 mL of deionized water and injected into an RP-HPLC column
(Knauer 20 mm ID, Eurospher-100 C18). The column was
eluted with a solvent gradient of 0-50% of acetonitrile in
50 mM aq NH4HCO3 buffer for 40 min at a flow rate of
5 mL/min. The UV absorption was detected at 254 nm. Frac-
tions were collected and appropriate fractions pooled, diluted
with water, and lyophilized several times to remove the
NH4HCO3 buffer, yielding the products as white powders.
Compounds 11a-f, h-j and 12a-f, h-i were synthesized
essentially as previously described; for details, see Supporting
Information.
2-(Dioxolanylethylthio)adenosine-50-monophosphate (12l). 1H
NMR (500 MHz, D2O) δ 2.12-2.16 (m, 2H), 3.04 (t, 2H, J=
7.40 Hz), 3.91-3.93 (m, 2H), 3.96-3.3.97 (m, 2H), 4.02-4.03
(m, 2H), 4.31 (m, 1H), 4.47 (t, 1H, J=4.72 Hz), 4.65 (m, 1H),
5.07 (t, 1H, J=4.57 Hz), 6.10 (d, 1H, J=5.67 Hz), 8.42 (s,
1H). 31P NMR (202 MHz, D2O) δ 2.72 (s). 13C NMR (125
MHz, DMSO-d6) δ 28.09, 35.74, 66.44, 67.58, 73.55, 77.10,
87.36, 89.68, 105.93, 119.16, 142.20, 153.22, 158.20, 167.68.
LC/ESI-MS: negative mode 478 ([M - H]-), positive mode
480 ([M þ H]þ).
Adenosine Receptor Binding Assays. Chemicals. Tris(hydro-
xymethyl)aminomethan (Tris-buffer) was obtained from Acros
Organics (Leverkusen, Germany); DMSO was from Fluka
(Switzerland), HCl was from Merck, and HAT supplement from
Gibco. [3H]CGS21680 ([3H]2-(4-(2-carboxymethyl)phenyl)ethy-
lamino)-50-N-ethylcarboxamidoadenosine, 45 Ci/mmol)32 was
purchased from PerkinElmer Life Sciences (USA). [3H]C-
CPA ([3H]2-chloro-N6-cyclopentyladenosine (48.6 Ci/mmol),33
[3H]MSX-2 ([3H]3-(3-hydroxypropyl)-7-methyl-8-(m-methoxys-
tyryl)-1-propargylxanthine, 84 Ci/mmol),25 and [3H]PSB-11
([3H]8-ethyl-4-methyl-2-phenyl-(8R)-4,5,7,8-tetrahydro-1H-imi-
dazo[2,1-i]-purin-5-one (53 Ci/mmol))34 were custom-labeled by
Amersham from suitable precursors that were synthesized as
previously described. All other chemical reagents, cell culture
materials, and adenosine receptor ligands were obtained from
Sigma.
2-(Cyclohexylethylthio)adenosine (11g). 1H NMR (500 MHz,
DMSO-d6) δ 0.87-1.60 (m, 11H), 1.61-1.74 (m, 2H), 3.0-3.14
(m, 2H), 3.50-3.64 (m, 2H), 3.90 (q, 1H, J=3.99 Hz), 4.11 (t,
1H, J=4.25 Hz), 4.59 (t, 1H, J=5.35 Hz), 4.98 (t, 1H, J=
5.51 Hz), 5.10 (m, 1H), 5.36 (m, 1H), 5.80 (d, 1H, J=5.99 Hz),
7.28 (s, 2H), 8.19 (s, 1H). 13C NMR (125 MHz, DMSO-d6)
δ25.89, 26.26, 27.94, 32.46, 36.61, 36.82, 61.75, 70.60, 73.31, 85.57,
87.42, 117.05, 138.86, 150.31, 155.63, 163.94. LC/ESI-MS: nega-
tive mode 408 ([M - H]-), positive mode 410 ([M þ H]þ).
1
2-(Dioxanylethylthio)adenosine (11k). H NMR (500 MHz,
DMSO-d6) δ 1.31-1.34 (m, 1H), 1.85-1.88 (m, 3H), 3.01-3.12
(m, 2H), 3.50-3.64 (m, 2H), 3.67-3.72 (m, 2H), 3.90 (q, 1H, J=
3.67 Hz), 3.97-4.01 (m, 2H), 4.11 (q, 1H, J=2.41 Hz), 4.59 (t,
1H, J=5.51 Hz), 4.66 (t, 1H, J=5.20 Hz), 4.98 (m, 1H), 5.26 (m,
2H), 5.79 (d, 1H, J=5.99 Hz), 7.28 (s, 2H), 8.20 (s, 1H). 13C
NMR (125 MHz, DMSO-d6) δ 25.22, 25.51, 34.69, 61.76, 66.20,
70.60, 73.31, 85.60, 87.51, 100.40, 117.10, 138.97, 150.25, 155.67,
163.58. LC/ESI-MS: negative mode 412 ([M - H]-), positive
mode 414 ([M þ H]þ).
Receptor-Radioligand Binding Studies. Rat brain cortical
membrane preparations were used as a source for A1, and rat
brain striatal membrane preparations as a source for A2A
receptors as previously described.35,36 Membrane preparations
of Chinese hamster ovary (CHO) cells expressing the human A3
receptors were used for A3 assays.33 Stock solutions of the
compounds were prepared in dimethyl sulfoxide (DMSO); the
final concentration of DMSO was 2.5%. The radioligand con-
centrations and incubation times (incubation at rt) were as
follows: [3H]CCPA, 1.0 nM (rat A1), incubation for 90 min;
[3H]CGS21680, 10.0 nM (rat A2A), incubation for 60 min; [3H]-
MSX-2, 1.0 nM (rat A2A), incubation for 30 min; [3H]PSB-11,
1.0 nM (human A3), incubation for 60 min. About 30-50 μg of
protein /vial were used in the assays. Membranes were preincu-
bated for 10-15 min with 0.12 IU/mL of adenosine deaminase
in order to remove endogenous adenosine. Binding assays were
performed essentially as previously described.35,36 Curves were
determined using 6-7 different concentrations of test com-
pounds spanning 3 orders of magnitude. At least two to three
separate experiments were performed, each in duplicate (human
receptors) or triplicate (rat receptors).
2-(Dioxolanylethylthio)adenosine (11l). 1H NMR (500 MHz,
DMSO-d6) δ 1.03-1.07 (m, 2H), 3.04-3.14 (m, 2H), 3.37-3.43
(m, 2H), 3.50-3.63 (m, 2H), 3.77-3.79 (m, 1H), 3.89-3.92 (m,
2H), 4.11 (q, 1H, J=2.41 Hz), 4.59 (t, 1H, J=5.51 Hz), 4.93 (t,
1H, J=4.88 Hz), 5.01 (m, 1H), 5.26 (m, 2H), 5.80 (d, 1H, J=5.99
Hz), 7.31 (s, 2H), 8.20 (s, 1H). 13C NMR (125 MHz, DMSO-d6)
δ 25.21, 33.58, 61.74, 64.34, 70.60, 73.37, 85.60, 87.44, 102.71,
117.09, 138.92, 150.27, 155.68, 163.55. LC/ESI-MS: negative
mode 398 ([M - H]-), positive mode 400 ([M þ H]þ).
2-(Cyclohexylethylthio)adenosine-50-monophosphate (12g).
1H NMR (500 MHz, D2O) δ 0.92-1.67 (m, 11H), 1.73-1.75
(m, 2H), 3.28-3.36 (m, 2H), 3.99 (m, 2H), 4.33 (m, 1H), 4.50
(m, 1H), 4.70 (m, 1H), 6.10 (d, 1H, J=5.67 Hz), 8.45 (s, 1H).

7676 Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23
El-Tayeb et al.
Recombinant Expression of Rat ecto-50-Nucleotidase. Cataly-
tically active recombinant soluble glutathione-S-transferase/
ecto-50-nucleotidase fusion protein was expressed in insect cells
using the baculovirus system and purified by affinity chroma-
tography using agarose-coupled reduced glutathione as descri-
bed previously.26
ecto-50-Nucleotidase Assays. Preparation of Standard Solu-
tions. Compounds were dissolved in DMSO to obtain 5.0 mM
stock solutions. These were further diluted to obtain 1 mM
solutions in assay buffer (10 mM Hepes, pH 7.4, 2 mM MgCl2,
1 mM CaCl2). The 1 mM solutions were further diluted in the
same buffer as required for the standard calibration curves and
the enzyme assays.
cells were removed with lysing buffer (ammonium chloride), and
CD4þ T lymphocytes were isolated by magnetic cell sorting with
CD4 MicroBeads (Miltenyi). After resuspending purified cells in
Panserin 413 (Pan), cells were activated by incubation in 96-well
plates coated with 0.4 μg/mL anti-CD3 mAb and 2.5 μg/mL anti-
CD28 (BD Bioscience) for 18 h. Supernatant was used for the
analysis of IFN-γ and hydrolysis of AMP derivative 12g.
Measurement of IFN-γ. IFN-γ concentrations in superna-
tants of CD4þ cell cultures were measured by ELISA according
to the manufacturer’s protocol (R&D Systems).
Quantitative Determination by LC-MS. Samples (10, 5, or
2.5 mM) were injected (10 μL) into an HPLC instrument (Agilent
1100) equipped with a Phenomenex Luna 3 μ C18 column. Elution
was performed with a gradient of water:methanol (containing
2 mM ammonium acetate) from 90:10 to 0:100 for 20 min at a
flow rate of 250 μL/min, followed by 100% methanol for 35 min.
The compounds were detected by MS/MS multireaction monitor-
ing (MS/MS-MRM) in the negative mode with an electrospray ion
source (ESI) on an API 2000 (Applied Biosystems, Darmstadt,
Germany) mass spectrometer from the masses of 408.35-276.25
for the nucleoside 11g and from the masses of 488.25-79.15 for the
nucleotide 12g. Calibration curves were obtained from the pure
compounds, and the area under the peak, for each concentration,
was quantified.
Capillary Enzymatic Reaction. CE separations were carried
out using a P/ACE MDQ system (Beckman Coulter Instru-
ments, Fullerton, CA) equipped with a DAD detection system.
The electrophoretic separations were carried out using an eCAP
fused-silica capillary [30 cm (20 cm effective length), ꢀ75 μm
internal diameter (I.D.), ꢀ375 μm outside diameter (O.D.)
obtained from Beckman Coulter]. The following conditions
were applied: T=25 ꢀC, λmax=232 nm, voltage=15 kV, running
buffer 40 mM sodium borate buffer, pH 9.1. The capillary was
washed with 0.1 M NaOH for 2 min, deionized water for 1 min,
and running buffer for 1 min before each injection. Injections
were made by applying 0.1 psi of pressure to the sample solution
for 30 s. The amount of adenosine or other nucleosides formed
was determined. The CE instrument was fully controlled
through a personal computer, which operated with the analysis
software 32 KARAT obtained from Beckman Coulter. Electro-
pherograms were evaluated using the same software.
Investigation of ecto-50-Nucleotidase Substrates by Capillary
Electrophoresis. Assays were carried out at 37 ꢀC in a final
volume of 100 μL. The reaction mixture contained 500 μM of
the nucleoside 50-monophosphate dissolved in reaction buffer
(10 mM Hepes, pH 7.4, 2 mM MgCl2, 1 mM CaCl2). The
reactions were initiated by adding 10 μL of the appropriately
diluted enzyme and were then allowed to proceed at 37 ꢀC for
20 min. The reaction was stopped by heating the mixture at
99 ꢀC for 5 min. Aliquots of 50 μL of the reaction mixture were
transferred to mini CE vials and injected into the CE instru-
ment under the conditions described above. The absorbance at
232 nm was monitored continuously, and the nucleoside con-
centrations were determined from the area under each absor-
bance peak. The experiments were repeated twice with triplicate
injections. Control experiments were performed in the absence
of substrate and in the absence of enzyme in order to take into
account the spontaneous hydrolysis of nucleoside 50-monophos-
phates under the experimental conditions, which always
amounted to less than 1%. All of the nucleosides were injected
into the capillary electrophoresis under the same assay condi-
tions at 500 μM concentrations to confirm the migration times
of the nucleoside peaks observed after enzymatic hydrolysis of
the nucleoside monophosphates. For the quantification of the
dephosphorylation products, the nucleosides, we applied the
internal normalization method. Thus, the individual peak areas
reflected directly the relative content of a corresponding analyte.
This was possible because starting compound (nucleoside
monophosphate) and product (nucleoside) exhibit the same
molar absorption coefficient; the phosphate residue does not
show any UV absorption at the wavelength used for quantifica-
tion (232 nm). This method is also completely independent of the
amount of substrate injected into the capillary, which represents
100% at zero reaction time.
Acknowledgment. This study was supported by the
Deutsche Forschungsgemeinschaft (DFG, GRK804, grant
to A.B. and C.E.M.). A.El-T. and J.I. are grateful for
a STIBΕT scholarship by the Deutscher Akademischer
Austauschdienst (DAAD). We thank Nicole Florin for per-
forming some of the radioligand binding studies and Sabine
Terhart-Krabbe and Annette Reiner for NMR spectra.
Supporting Information Available: Synthetic procedures,
1H and 13C NMR spectral data, and mass spectra of nucleosides
11a-f,h-j and nucleotides 12a-f,h-i, as well as purity deter-
mination of synthesized compounds by CE and HPLC and MS
spectra from quantitative analyses and HPLC-MS, including
MS spectra. This material is available free of charge via the
Internet at http://pubs.acs.org.
References
(1) Fredholm, B. B.; IJzerman, A. P.; Jacobson, K. A.; Klotz, K. N.;
Linden, J. International Union of Pharmacology. XXV. Nomen-
clature and classification of adenosine receptors. Pharmacol. Rev.
2001, 53, 527–552.
(2) Klotz, K. N. Adenosine receptors and their ligands. Naunyn-
Schmiedeberg’s Arch. Pharmacol. 2000, 362, 382–391.
(3) Jaakola, V. P.; Griffith, M. T.; Hanson, M. A.; Cherezov, V.;
Chien, E. Y.; Lane, J. R.; IJzerman, A. P.; Stevens, R. C. The 2.6
˚
Angstrom crystal structure of a human A2A adenosine receptor
bound to an antagonist. Science 2008, 322, 1211–1217.
(4) Hasko, G.; Linden, J.; Cronstein, B.; Pacher, P. Adenosine recep-
tors: therapeutic aspects for inflammatory and immune diseases.
Nat. Rev. Drug Discovery 2008, 7, 759–770.
(5) Fredholm, B. B.; Arslan, G.; Halldner, L.; Kull, B.; Schulte, G.;
Wasserman, W. Structure and function of adenosine receptors and
their genes. Naunyn-Schmiedeberg’s Arch. Pharmacol. 2000, 362,
364–374.
€
(6) Akkari, R.; Burbiel, J. C.; Hockemeyer, J.; Muller, C. E. Recent
progress in the development of adenosine receptor ligands as
antiinflammatory drugs. Curr. Top. Med. Chem. 2006, 6, 1375–
1399.
(7) Zhao, G.; Linke, A.; Xu, X.; Ochoa, M.; Belloni, F.; Belardinelli,
L.; Hintze, T. H. Comparative profile of vasodilation by CVT-
3146, a novel A2A receptor agonist, and adenosine in conscious
dogs. J. Pharmacol. Exp. Ther. 2003, 307, 182–189.
T-lymphocyte Isolation and Activation. C57/Bl6 male mice
(8-12 wk old) were obtained from the central animal facility of
ꢀ
(8) Cunha, R. A.; Ferre, S.; Vaugeois, J. M.; Chen, J. F. Potential
€
the Heinrich-Heine-University (Dusseldorf, Germany). They
therapeutic interest of adenosine A2A receptors in psychiatric
disorders. Curr. Pharm. Des. 2008, 14, 1512–1524.
€
(9) Yan, L.; Burbiel, J. C.; Maass, A.; Muller, C. E. Adenosine receptor
agonists: from basic medicinal chemistry to clinical development.
were fed a standard chow diet and received tap water ad libitum.
Mice were sacrificed, and the spleens were removed. Splenocytes
were passed through a 70 μm nylon cell strainer (BD Bioscience)
and collected in phosphate-buffered saline (PBS). Red blood
Expert Opin. Emerging Drugs 2003, 8, 537–576.

Article
Journal of Medicinal Chemistry, 2009, Vol. 52, No. 23 7677
3
(25) Muller, C. E.; Maurinsh, J.; Sauer, R. Binding of [ H]MSX-2 (3-(3-
€
(10) Jacobson, K. A. Adenosine receptors as therapeutic targets. Nat.
Rev. Drug Discovery. 2006, 5, 247–264.
(11) Hasan, A.; Hussain, T.; Mustafa, S. J.; Srivastava, P. C. 2-
Substituted thioadenine nucleoside and nucleotide analogues:
synthesis and receptor subtype binding affinities. Bioconjugate
Chem. 1994, 5, 364–369.
hydroxypropyl)-7-methyl-8-(m-methoxystyryl)-1-propargylxanth-
ine) to rat striatal membranes-a new, selective antagonist radi-
oligand for A(2A) adenosine receptors. Eur. J. Pharm. Sci. 2000,
10, 259–265.
€
(26) Servos, J.; Reilander, H.; Zimmermann, H. Catalytically active
(12) Volpini, R.; Costanzi, S.; Lambertucci, C.; Portino, F. R.; Taffi, S.;
Vittori, S.; Zablocki, J. A.; Klotz, K. N.; Cristalli, G. Adenosine
receptor agonists: synthesis and binding affinity of 2-(aryl)al-
kylthioadenosine derivatives. ARKIVOC 2004, 5, 301–311.
(13) Livingston, M.; Heaney, L. G.; Ennis, M. Adenosine, inflamma-
tion and asthma-a review. Inflammation Res. 2004, 53, 171–178.
soluble ecto-50-nucleotidase purified after heterologous expression
as a tool for drug screening. Drug. Dev. Res. 1998, 45, 269–276.
(27) Deaglio, S.; Dwyer, K. M.; Gao, W.; Friedman, D.; Usheva, A.;
Erat, A.; Chen, J. F.; Enjyoji, K.; Linden, J.; Oukka, M.; Kuchroo,
V. K.; Strom, T. B.; Robson, S. C. Adenosine generation catalyzed
by CD39 and CD73 expressed on regulatory T cells mediates
immune suppression. J. Exp. Med. 2007, 204, 1257–1265.
€
(14) Iqbal, J.; Jirovsky, D.; Lee, S. Y.; Zimmermann, H.; Muller, C. E.
€
€
(28) Koszalka, P.; Ozuyaman, B.; Huo, Y.; Zernecke, A.; Flogel, U.;
Capillary electrophoresis-based nanoscale assays for monitoring
ecto-50-nucleotidase activity and inhibition in preparations of
recombinant enzyme and melanoma cell membranes. Anal. Bio-
chem. 2008, 373, 129–140.
ꢀ
Braun, N.; Buchheiser, A.; Decking, U. K.; Smith, M. L.; Sevigny,
J.; Gear, A.; Weber, A. A.; Molojavyi, A.; Ding, Z.; Weber, C.; Ley,
€
K.; Zimmermann, H.; Godecke, A.; Schrader, J. Targeted disrup-
(15) Li, X.; Zhou, T.; Zhi, X.; Zhao, F.; Yin, L.; Zhou, P. Effect of
hypoxia/reoxygenation on CD73 (ecto-50-nucleotidase) in mouse
microvessel endothelial cell lines. Microvasc. Res. 2006, 72, 48–53.
tion of cd73/ecto-50-nucleotidase alters thromboregulation and
augments vascular inflammatory response. Circ. Res. 2004, 95,
814–821.
€
(16) Zernecke, A.; Bidzhekov, K.; Ozuyaman, B.; Fraemohs, L.; Liehn,
(29) (a) Zylka, M. J.; Sowa, N. A.; Taylor-Blake, B.; Twomey, M. A.;
Herrala, A.; Voikar, V.; Vihko, P. Prostatic acid phosphatase is an
ectonucleotidase and suppresses pain by generating adenosine.
Neuron 2008, 60, 111–122. (b) Quintero, I. B.; Araujo, C. L.; Pulkka,
A. E.; Wirkkala, R. S.; Herrala, A. M.; Eskelinen, E. L.; Jokitalo, E.;
€
€
E. A.; Luscher-Firzlaff, J. M.; Luscher, B.; Schrader, J.; Weber, C.
CD73/ecto-50-nucleotidase protects against vascular inflammation
and neointima formation. Circulation 2006, 113, 2120–2127.
(17) Buchheiser, A.; Pexa, A.; Emde, B.; Ding, Z.; Romio, M.; Burghoff,
€
€
S.; Viethen, C.; Lindecke, A.; Kohrer, K.; Fischer, J. W.; Schrader, J.
Hellstrom, P. A.; Tuominen, H. J.; Hirvikoski, P. P.; Vihko, P. T.
Multiple roles of CD73-derived adenosine in atherogenesis. 2009,
submitted for publication.
Prostatic acid phosphatase is not a prostate specific target. Cancer
Res. 2007, 67, 6549–6554.
€
(18) Zimmermann, H. Extracellular metabolism of ATP and other
nucleotides. Naunyn-Schmiedeberg’s Arch. Pharmacol. 2000, 362,
299–309.
(30) El-Tayeb, A.; Qi, A.; Muller, C. E. Synthesis and structure-
activity relationships of uracil nucleotide derivatives and analogs
as agonists at human P2Y2, P2Y4, and P2Y6 receptors. J. Med.
Chem. 2006, 49, 7076–7087.
€
(19) Schutz, W.; Schrader, J.; Gerlach, E. Different sites of adenosine
€
€
(31) Knoblauch, B. H. A.; Muller, C. E.; Jarlebark, L.; Lawoko, G.;
formation in the heart. Am. J. Physiol. 1981, 240, 963–970.
(20) Langer, D.; Hammer, K.; Koszalka, P.; Schrader, J.; Robson, S.;
Zimmermann, H. Distribution of ectonucleotidases in the rodent
brain revisited. Cell Tissue Res. 2008, 334, 199–217.
€
Kottke, T.; Wikstrom, M. A. Heilbronn, E. 5-Substituted UTP
derivatives as P2Y2 receptor agonists. Eur. J. Med. Chem. 1999, 34,
809–824.
(21) (a) Boyer, J. L.; Siddiqi, S.; Fischer, B.; Romero-Avila, T.; Jacobson
K. A.; Harden, T. K. Identification of potent P2Y-purinoceptor
agonists that are derivatives of adenosine 50-monophosphate.
Br. J. Pharmacol. 1996, 118, 1959–1964. (b) Hourani, S. M. O.;
Welford, L. A.; Cusack, N. J. Effects of 2-methylthioadenosine 50-
β,γ-methylenetriphosphonate and 2-ethylthioadenosine 50-monophos-
phate on human platelet activation induced by adenosine 50-diphos-
phate. Drug Dev. Res. 1996, 37, 12–23. (c) Cristalli, G; Podda, G. M.;
Costanzi, S.; Lambertucci, C.; Lecchi, A.; Vittori, S.; Volpini, R.;
Zighetti, M. L.; Cattaneo, M. Effects of 50-phosphate derivatives of
2-hexynyl adenosine and 2-phenylethynyl adenosine on responses of
human platelets mediated by P2Y receptors. J. Med. Chem. 2005, 48,
2763–2766.
(32) Jarvis, M. F.; Schulz, R.; Hutchison, A. J.; Do, U. H.; Sills, M. A.;
Williams, M. [3H]CGS 21680, a selective A2 adenosine receptor
agonist directly labels A2 receptors in rat brain. J. Pharmacol. Exp.
Ther. 1989, 251, 888–893.
(33) Klotz, K. N.; Lohse, M. J.; Schwabe, U.; Cristalli, G.; Vittori, S.;
Grifantini, M. 2-Chloro-N6-[3H]cyclopentyladenosine ([3H]
CCPA)-a high affinity agonist radioligand for A1 adenosine
receptors. Naunyn-Schmiedeberg’s Arch. Pharmacol. 1989, 340,
679–683.
3
(34) Muller, C. E.; Diekmann, M.; Thorand, M.; Ozola, V. [ H]8-Ethyl-
€
4-methyl-2-phenyl-(8R)-4,5,7,8-tetrahydro-1H-imidazo[2,1-i]-purin-
5-one ([3H]PSB-11), a novel high affinity antagonist radioligand
for human A3 adenosine receptors. Bioorg. Med. Chem. Lett. 2002,
12, 501–503.
(22) Kikugawa, K.; Suehiro, H.; Yanase, R.; Aoki, A. Platelet aggrega-
tion inhibitors. IX. Chemical transformation of adenosine into 2-
thioadenosine derivatives. Chem. Pharm. Bull. 1977, 25, 1959–
1969.
€
(35) Yan, L.; Muller, C. E. Preparation, properties, reactions, and
adenosine receptor affinities of sulfophenylxanthine nitrophenyl
esters: Toward the development of sulfonic acid prodrugs with
peroral bioavailability. J. Med. Chem. 2004, 47, 1031–1043.
(23) Kikugawa, K.; Suehiro, H.; Aoki, A. Platelet aggregation inhibi-
tors. X. S-Substituted 2-thioadenosines and their derivatives.
Chem. Pharm. Bull. 1977, 25, 2624–2627.
€
€
(36) Bulicz, J.; Bertarelli, D. C. G.; Baumert, D.; Fulle, F.; Muller, C. E.;
Heber, D. Synthesis and pharmacology of pyrido[2,3-d]pyrimidi-
nediones bearing polar substituents as adenosine receptor antago-
nists. Bioorg. Med. Chem. 2006, 14, 2837–2849.
(24) Ludwig, J. A new route to nucleoside 50-triphosphates. Acta
Biochim. Biophys. Acad. Sci. Hung. 1981, 16, 131–133.