Diadenosine 5',5'-(Boranated)polyphosphonate analogues as selective nucleotide pyrophosphatase/phosphodiesterase inhibitors ⊥

Article abstract of DOI:10.1021/jm100597c

Nucleotide pyrophosphatase/phosphodiesterases (NPPs) hydrolyze extracellular nucleotides and dinucleotides and thus control purinergic signaling. Enhanced NPP activity is implicated in health disorders such as osteoarthritis and cancer. We designed novel

Full text of DOI:10.1021/jm100597c

J. Med. Chem. 2010, 53, 8485–8497 8485
DOI: 10.1021/jm100597c
Diadenosine 50,500-(Boranated)polyphosphonate Analogues as Selective Nucleotide
Pyrophosphatase/Phosphodiesterase Inhibitors^
,‡
§
§
‡
‡
Shay Eliahu,†, Joanna Lecka, Georg Reiser, Michael Haas, Franc-ois Bigonnesse, Sebastien A. Levesque,
ꢀ
ꢀ
Julie Pelletier, Jean Sevigny,*,‡ and Bilha Fischer*,†
‡
ꢀ
†Department of Chemistry, Bar-Ilan University, Ramat-Gan 52900, Israel, ‡Centre de recherche en Rhumatologie et Immunologie,
Centre Hospitalier Universitaire de Queꢀbec ( pavillon CHUL), Queꢀbec, QC, Canada, and Deꢀpartement de microbiologie-infectiologie et
§
d’immunologie, Faculteꢀ de Meꢀdecine, Universiteꢀ Laval, Queꢀbec, QC, Canada, and Institute for Neurobiochemistry, Faculty of Medicine,
Otto von Guericke University, Leipziger Strasse 44, D-39120 Magdeburg, Germany. These authors contributed equally to this work.
Received May 17, 2010
Nucleotide pyrophosphatase/phosphodiesterases (NPPs) hydrolyze extracellular nucleotides and
dinucleotides and thus control purinergic signaling. Enhanced NPP activity is implicated in health disorders
such as osteoarthritis and cancer. We designed novel diadenosine polyphosphonate derivatives as potential
NPP inhibitors. Analogues 1-4 bear a phosphonate and/or boranophosphate group and/or a 20-H atom
instead of a 20-OH group. In comparison to ATP, analogues 1-4 were barely hydrolyzed by human
NTPDase1, -2, -3, and -8 (<5% hydrolysis) and NPP1 and -3 (e13%) and were not hydrolyzed by ecto-
50-nucleotidase, unlike AMP. These derivatives did not affect NTPDase activity, and analogues 1 and 2
did not inhibit ecto-50-nucleotidase. All analogues blocked ∼80% of the NPP2-dependent hydrolysis of
pnp-TMP, a specific NPP substrate, and inhibited the catabolism of pnp-TMP (Ki and IC50 both found
to be between 10 and 60 μM), Ap5A, and ATP by NPP1. The activity of NPP3 was inhibited to a lesser
extent by the new analogues, with compounds 1 and 4 being the most effective in that respect. The
analogues dramatically reduced the level of hydrolysis of pnp-TMP at the cell surface of both
osteocarcinoma and colon cancer cells. Importantly, analogues 1-4 exhibited significantly reduced
agonistic activity toward human P2Y1,11 receptors (except for analogue 1) and no activity with human
P2Y2 receptor. Our data provide strong evidence that analogue 2 is the first specific NPP inhibitor to be
described.
Introduction
activity, AMP, is further hydrolyzed by ecto-50-nucleotidase
(CD73), giving adenosine, which is the natural ligand of P1
receptors.6,7
Nucleoside triphosphate diphosphohydrolase-1, -2, -3, and -8
[NTPDase1, -2, -3, and -8, respectively (EC 3.6.1.5)] and nucleo-
tide pyrophosphatase/phosphodiesterase-1 and -3 [NPP1 and -3,
respectively (EC 3.1.3.1 and EC 3.6.1.9, respectively)] are the
dominant ectonucleotidases that terminate nucleotide signal-
ing through the hydrolysis of nucleotide agonists of the P2X
and P2Y receptors.1-3
NTPDases catalyze the hydrolysis of the terminal phos-
phate of nucleoside triphosphates (e.g., ATP and UTP) and
diphosphates (ADP and UDP) at different rates. NTPDase1,
-2, -3, and -8 are plasma membrane-bound with an extra-
cellular activesite. NTPDase1 (CD39/ATPDase/ectoapyrase/
ecto-ADPase) hydrolyzes ATP and ADP equally well,4 while
NTPDase2 (ecto-ATPase/CD39L1) is a preferential triphospho-
nucleosidase.5 Both NTPDase3 (CD39L3/HB6) and NTPDase8
are functional intermediates between NTPDase1 and
NTPDase2.3 NTPDase4-7 are mainly associated with intra-
cellular organelles and are therefore not expected to signifi-
cantly affect P2 receptor activation. The product of NTPDase
E-NPPa family members are conserved eukaryotic enzymes
that, as for NTPDases, exist as membrane glycoproteins, with
an extracellular active site. Three members of the E-NPP family
hydrolyze nucleotides, namely, NPP1 (PC-1), NPP2 (PD-IR
autotaxin), and NPP3 (gp130RB13-6/B10/PD-Iβ). NPP1 and
NPP3 are closely related, being ∼50% identical, and are 39
and 31% identical to NPP2, respectively.8 NPP1-3 have broad
substrate specificity. They hydrolyze nucleotides, dinucleotides,
and nucleotide sugars, e.g., ATP, ADP, NADþ, ADP-ribose,
and diadenosine polyphosphates, and normally release AMP
and the remaining part of the molecule as the main hydrolysis
products. Both purine and pyrimidine nucleotides may serve as
substrates.9,10 Thymidine 50-monophosphate p-nitrophenyl es-
ter (pnp-TMP) is often used as a synthetic substrate for NPP
activity assays.11
a Abbreviations: E-NPP, ecto-nucleotide pyrophosphatase/phospho-
diesterase; E-NTPDase, ecto-nucleoside triphosphate diphosphohydro-
lase; pnp-TMP, thymidine 50-monophosphate p-nitrophenyl ester; P2R,
P2 receptor; BPi, boranophosphate; CDI, carbonyldimidazole; ESI, elec-
tron spray ionization; MPLC, medium-pressure liquid chromatography; RT,
room temperature; TEAA, triethylammonium acetate; HRMS-MALDI,
high-resolution mass spectrometry matrix-assisted laser desorption ioniza-
tion; FBS, fetal bovine serum; [Ca2þ]i, intracellular Ca2þ concentration; SD,
standard deviation.
^ Patent pending.
*To whom correspondence should be addressed. B.F.: Department
of Chemistry, Bar-Ilan University, Ramat-Gan 52900, Israel; fax, 972-3-
6353907; telephone, 972-3-5318303; e-mail, bfischer@mail.biu.ac.il. J.S.:
Centre de Recherche en Rhumatologie et Immunologie, 2705 Laurier
ꢀ
Blvd., Room T1-39, Quebec, QC, Canada G1V 4G2; fax, (418) 654-2765;
telephone, (418) 654-2772; e-mail, Jean.Sevigny@crchul.ulaval.ca.
r
2010 American Chemical Society
Published on Web 11/22/2010
pubs.acs.org/jmc

8486 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 24
Eliahu et al.
Figure 1. Dinucleotide analogues synthesized and investigated in this study.
NPP2 has a much lower ATPase activity than NPP1 and
NPP3 and therefore may not play an important role in the
regulation of P2 receptor activation. Nevertheless, NPP1,
NPP2, and NPP3 are the major if not only enzymes that have
the ability to hydrolyze extracellular diadenosine polyphos-
phates in vertebrate tissues12 and may therefore play a crucial
role in the regulation of the functions played by these mole-
cules. Other cloned NPPs (NPP4-7) do not hydrolyze
nucleotides.9,13
NPP1-3 have been implicated in various biological pro-
cesses. NPP1 plays an important role in the regulation of bone
mineralization.14 NPP1-deficient humans exhibit pathological
calcification.15 Although osteoblasts express NPP1-3, only
NPP1 is enriched in membrane-limited vesicles (matrix vesicles),
where it plays an important role in the control of miner-
alization.16 A loss of functional NPP1 is associated with
hypermineralization abnormalities such as osteoarthritis and
ossification of the posterior longitudinal ligament of the
spine.17 NPP1 has also been reported to affect insulin signal-
ing via the inhibition of insulin receptor tyrosine kinase
activity.12,18 In addition, NPPs are involved in cell differentia-
tion and motility.19,20 NPP3 expression is associated with
carcinogenesis and metastasis of cancer cells and has been
proposed as a tumor marker.21 As the level of NPP expression
is increased in membranes of aged rat brains (NPP1) and in
the brain cortex of Alzheimer’s disease patients (NPP2), the
inhibition of these enzymes has been proposed as a novel
alternative for targeting neurodegenerative diseases.22
There is currently a lack of specific NPP inhibitors. Due
to this state of affairs, the therapeutic potential of NPP
inhibition for the treatment of health disorders such as
chondrocalcinosis15 or cancer23,24 remains virtually unex-
plored. Indeed, NPP inhibitors have scarcely been reported.
Thus, suramin wasreportedtoreducethe level of hydrolysis of
p-Nph-50-TMP by NPP by ca. 36% at 250 μM.25 It is
noteworthy that suramin and its derivatives antagonize most
P2receptors andalso efficiently inhibit NTPDasesand cannot
therefore be considered as specific NPP inhibitors.26
Recently, [3-(tert-butyldimethylsilyloxy)phenyl]-1,3,3-oxa-
diazole-2(3H)-thione was reported as an NPP1 inhibitor
(Ki = 100 μM).27 Likewise, biscoumarin derivatives were
identified as pure noncompetitive inhibitors of snake venom
and human NPP1 enzymes, with Ki and IC50 values as low as
50 and 164 μM, respectively, for human NPP1.28
analogues 1-4 as potential NPP inhibitors (Figure 1). These
analogues were evaluated for their protein selectivity as either
agonists of P2Y1,2,11 receptors or substrates for the major
ectonucleotidases. Furthermore, their inhibitory activity and
NPP subtypeselectivity wereevaluatedbycomparisonoftheir
effects on the other main ectonucleotidases, in the presence of
pnp-TMP, Ap5A, or ATP as the substrate. In addition,
dinucleotide analogues 1-4 were evaluated as inhibitors of
cell surface NPP activity in two cancer cell lines. With this
approach, we have been able to identify a most selective NPP
inhibitor and establish important structure-activity relation-
ships for such inhibitors.
Results
Synthesis of Dinucleoside Polyphosphonate Analogues 1-4.
Dinucleoside polyphosphates are conventionally prepared
via the activation of the 50-terminal phosphate of a nucleo-
tide, thus forming a phosphoryl donor (P-donor), followed
by reaction with a nonactivated nucleoside 50-phosphate, or
phosphonate analogue [phosphoryl acceptor (P-acceptor)].
A common method for activation of phosphates/phospho-
nates uses carbonyldiimidazole (CDI) to form phosphoroi-
midazolides. The latter may be generated in situ or isolated
prior to the reaction with the corresponding nucleotides.29
We had previously synthesized diadenosine (γ-borano)penta-
phosphate (5) using an inorganic boranophosphate salt
(BPi)30 as a P-acceptor and two nucleoside phosphoroimi-
dazolides (NDP-Im) as P-donors.1 Here, we used the same
synthetic method to prepare the corresponding phosphonate
analogues, 1 and 2 (Schemes 1 and 2).
Diadenosine R,β-δ,ε-dimethylene-pentaphosphonate, 1, and
di-20-deoxyadenosine R,β-δ,ε-dimethylene-pentaphosphonate, 2,
were prepared as described above from the R,β-methylene-
ADP building blocks, 9, and R,β-methylene-20-deoxy-ADP,
13, respectively. R,β-Methylene-ADP derivatives were syn-
thesized as previously reported.31 Specifically, adenosine
analogue 6, which is 20- and 30-OH-protected, was activated
with tosyl chloride to form analogue 7.31 The activated nucleo-
side 7 was then coupled with a tris(tetra-n-butylammonium)-
methylene diphosphonate salt to form analogue 8, followed
by removal of the protecting group, which provided pro-
duct 9 (Scheme 1).31 The related scaffold, R,β-methylene-20-
deoxy-ADP, 13, was prepared from 20-deoxyadenosine. A
selective tosylation at the 50-OH position of 11 was con-
ducted at 0 ꢀC to form product 12,32 which was then coupled
with a tris(tetra-n-butylammonium)methylene diphosphonate
salt to yield product 13 (Scheme 2).32
The potential of NPP inhibitors as efficient therapeutic
agents, on one hand, and the lack of structure-activity relation-
ship and NPP selectivity data, on the other hand, encour-
aged us to design and synthesize a series of novel dinucleotide

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 24 8487
Scheme 1a
a Reaction conditions: (a) CH2Cl2, DMAP, TsCl, RT, 12 h, 67%; (b) tetrakis(n-butylammonium)methylenediphosphonate, dry DMF, RT, 38 h,
63%; (c) (1) 18% HCl, pH 2.3, RT, 3 h, (2) 23% NH4OH, pH 9, RT, 35 min, 63%; (d) CDI, DMF, RT, 12 h; (e) BPi, 15, MgCl2, RT, 23 h. 1 and 3 obtained
in 10 and 20% yields, respectively.
Nucleotides 9 and 13 were activated with CDI in situ to
form P-donors 10 and 14, which were then treated with BPi,
15, as a P-acceptor. MgCl2 was added as an activator to
overcome the low nucleophilicity of BPi as a P-acceptor.33
Products 1 and 2 were obtained in 10 and 21% overall yields,
respectively, after LC separation.
Dinucleoside poly(borano)phosphonate products 1 and 2 are
probably formed because of the preorganization of a P-acceptor
(BPi) and two P-donors (nucleoside phosphorimidazolides)
coordinated with one Mg2þ ion (Figure 2). Specifically, the
Mg2þ ion probably stablizes the folded structure, which
involve two molecules of R,β-methylene-ADP-imidazolide
(Im) or R,β-methylene-20-deoxy-ADP-Im and one BPi ion.
This structure provides the correct orientation and proxim-
ity for a nucleophilic attack of both nucleoside phosphor-
oimidazolide reactants by BPi in an octahedral complex
(Figure 2A, paths a and b). Although we expected to obtain
analogues 1 and 2 as exclusive products, byproducts 3 and 4
were also isolated with 20 and 28% yields, respectively. The
formation of 3 and 4 products is driven by R,β-methylene-ADP,
9, and R,β-methylene-20-deoxy-ADP, 13, which remained in the
reaction mixture because of incomplete reaction with CDI.
Thus, the activated forms, R,β-methylene-ADP-Im, 10, and
R,β-methylene-20-deoxy-ADP-Im, 14, become P-donors,
whereas R,β-methylene-ADP, 9, and R,β-methylene-20-
deoxy-ADP, 13, rather than BPi, function as P-acceptors
(Figure 2B). Furthermore, because the phosphonate is as-
sumed to be a better nucleophile than BPi, 15, byproducts
3 and 4 are obtained (path c) and not adenosine-R,β-CH2-
γ-boranotriphosphate, 16.
The identity and purity of products 1-4 were established
by 1H and 31P NMR, ESI or MALDI negative mass spectrom-
etry, and HPLC in two solvent systems. 31P NMR spectra of
products 1 and 2 showed a typical Pγ signal as a multiplet at
∼80 ppm in addition to two phosphonate signals at 20 and 10
ppm. 1H NMR spectra showed borane hydrogen atoms
as a very broad signal at ∼0.3 ppm and a typical triplet at
2.3 ppm of the bridging methylene group.

8488 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 24
Eliahu et al.
Scheme 2a
a Reaction conditions: (a) pyridine, DMAP, TsCl, 0 ꢀC, 3 h, 62%; (b) tetrakis(n-butylammonium)methylenediphosphonate, dry DMF, RT, 38 h,
62%; (c) CDI, DMF, RT, 12 h; (d) BPi, 15, MgCl2, RT, 23 h. 2 and 4 obtained in 21 and 28% overall yields in steps c and d, respectively.
Influence of Analogues 1-4 on Ectonucleotidase Activity.
(i) Effect of Analogues 1-4 on Recombinant Ectonucleotidases.
The experiments were conducted with protein extracts of COS-
7 cells transfected with an expression vector encoding one of
each ectonucleotidase, or, in a few experiments, with intact cell
lines exhibiting NPP activity. In addition, medium from NPP2-
transfected cells was assayed for activity to test the secreted
form of this enzyme. The protein extracts and media of non-
transfected COS-7 cells exhibited a negligible level of NTPDase
and NPP activity, thus allowing the analysis of each ectonu-
cleotidase in its native membrane-bound form.34
None of the analogues (1-4) was metabolized by human
NTPDases or ecto-50-nucleotidase (Table 1). These ana-
logues were modestly hydrolyzed by NPPs (Table 1). While
compounds 1 and 3 were hydrolyzed by NPP1 and -3
at ∼7.5-13% of the level observed for ATP hydrolysis,
analogues 2 and 4 were more resistant to hydrolysis
(Table 1). The hydrolysis of ATP by NTPDase1 and -8 was
not affected by any of the analogues (1-4) when used at the
same concentration as the substrate (100 μM). NTPDase2
and -3 were modestly inhibited (10-30%) by these analogues
(Figure 3). While analogues 3 and 4 inhibited ecto-50-nucleo-
tidase activity by 90 and 80%, respectively, analogues 1 and 2
did not affect the latter enzymatic activity (Figure 3).
The effect of analogues 1-4 on NPP activities was tested
using synthetic (pnp-TMP) and natural substrates (Ap5A
and ATP). The level of hydrolysis of pnp-TMP by NPP1 was
decreased by more than 90% by all compounds tested
(Figure 4A). In the presence of analogues 1-4, the hydrolysis
of pnp-TMP by NPP2 was similarly blocked at ∼95%
(Figure 4A, inset) and at 60-70% by the secreted form of
NPP2 (data not shown). The activity of NPP1 with Ap5A as
the substrate was reduced by 60-80% by analogues 1, 2, and
4 and by 20% by analogue 3. When using ATP as the
substrate, NPP1 was inhibited by ∼70-80% in the presence
of analogues 2 and 3 and by more than 90% by analogues
1 and 4 (Figure 4B,C). The presence of analogues 1-4
reduced the level of hydrolysis of pnp-TMP by NPP3
by ∼30% (Figure 4A) and the level of hydrolysis of Ap5A
by ∼10-60% (Figure 4B). The inhibition of NPP3 activity
using ATP as the substrate was more pronounced (∼90%) in
the presence of analogues 1 and 4, and around 65% with
analogues 2 and 3 (Figure 4C).
As analogues 1-4 significantly reduced the activity of
human NPP1, we have tested IC50 values as well as some kinetic
parameters (Ki and Ki0) of the inhibition with pnp-TMP as the
substrate. IC50 values were similar for all tested analogues,
on the order of 10-60 μM, while inhibition constants (Ki) were
in the range of 10-50 μM, always smaller than (enzyme-
substrate)-inhibitor dissociation constants (Ki0) that were
in the range of 30-150 μM. The lowest Ki value was observed
for analogue 2 (Table 2). Kinetic analysis of Ki and Ki0 using
the Dixon and Cornish-Bowden methods showed a mixed-
type, predominantly competitive, inhibition of NPP1 by all
tested analogues (data not shown).
(ii) Effect of Analogues 1-4 on NPP Activity at the Surface
of Two Cell Lines. NPP1 is critical in regulating miner-
alization by generating inorganic pyrophosphate, a potent
inhibitor of hydroxyapatite crystal growth. On the other
hand, NPP3 is associated with carcinogenesis. Human os-
teoblastic SaOS-2 cells (HTB-85) are used to investigate the
activity of NPP1,16 as well as HT29, a human colon cancer
cell line.35 HTB-85 and HT29 catabolized pnp-TMP, indi-
cating the presence of NPPs at their surface. As for the
enzymes obtained from cell extracts, NPP activity exhibi-
ted by both cell lines was blocked by ∼90 and ∼80% by
analogues 1 and 2 and by analogues 3 and 4, respectively
(Figure 5).

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 24 8489
Figure 2. Proposed structures for nucleotide-BPi Mg2þ complexes leading to products 1 and 2 (A) and products 3 and 4 (B).
Table 1. Hydrolysis of Analogues 1-4 by Human Ectonucleotidasesa
relative activity (% ( SD of ATP, AMP, or Ap5A hydrolysis)
human ectonucleotidase
1
2
3
4
NTPDase1
NTPDase2
NTPDase3
NTPDase8
ecto-50-nucleotidase
NPP1
0.5 ( 0.02
NDb
NDb
1.5 ( 0.1
0.1 ( 0.01
5.3 ( 0.2
NDb
1.1 ( 0.04
NDb
0.2 ( 0.01
NDb
NDb
1.8 ( 0.1
NDb
NDb
NDb
NDb
NDb
NDb
NDb
7.5 ( 0.3
12 ( 0.6
NDb
13 ( 0.6
11 ( 0.6
2.3 ( 0.1
5.5 ( 0.3
NPP3
5.5 ( 0.2
a Dinucleotide analogues 1-4 (each at 100 μM) were incubated with the indicated ectonucleotidases. The activity with 100 μM ATP (for NTPDases)
or AMP (for the ecto-50-nucleotidase) was set as 100%: 1270 ( 35, 928 ( 55, 202 ( 37, 129 ( 11, and 357 ( 10 nmol of Pi min-1 (mg of protein)-1 for
NTPDase1, -2, -3, and -8 and ecto-50-nucleotidase, respectively; 100% of the activity with 100 μM Ap5A as the substrate was 71 ( 5 and 98 ( 9 nmol of
nucleotide min-1 (mg of protein)-1 for NPP1 and NPP3, respectively (n = 3). b No hydrolysis detected.
Activity of Analogues 1-4 on the P2Y1, P2Y2, and P2Y11
Receptors. GFP (green fluorescent protein) constructs of
human P2Y1 and P2Y11 receptors were expressed in 1321N1
astrocytoma cells, which lack endogenous expression of both
P2X and P2Y receptors. These cells were then incubated with
various concentrations of analogues 1-4. The Ca2þ response
to derivatives 1-4 was compared with that due to ATP
(Figure 6A,B).
ATP (EC50 = 0.15 μM) and 30-fold more potent than Ap4A
derivative 3 [EC50 = 9 μM (Table 3)]. The 20-deoxy ana-
logues, 2 and 4, exhibited comparably weak activities, with
EC50 values of g30 μM for the P2Y1 receptor. No clear
plateau was reached up to 100 μM for analogue 2.
Derivative 1 also exhibited the highest P2Y11 receptor
agonist potency (EC50 = 13 μM) among the diadenosine
polyphosphate analogues. The maximal response reached
∼80% of that obtained with the standard agonist ATP
(EC50 =3.3 μM), but with a 4-fold lower potency. Analogue 3
was found to be a very weak agonist of the P2Y11 receptor,
With the exception of derivative 1, the other analogues
(2-4) were weak agonists of the P2Y1 receptor (Table 3).
Compound 1 was 2-fold less potent than the standard agonist

8490 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 24
Eliahu et al.
Figure 3. Effects of analogues 1-4 on NTPDase and ecto-50-nucleotidase activity. Either ATP or AMP was used as a substrate in the presence
of analogue 1 (A), 2 (B), 3 (C), or 4 (D). Both substrate and analogues 1-4 were used at 100 μM. The 100% activity was set with the nucleotide
substrate alone (ATP for NTPDases or AMP for the ecto-50-nucleotidase): 1270 ( 35, 928 ( 55, 202 ( 37, 129 ( 11, and 357 ( 10 nmol of
Pi min-1 (mg of protein)-1 for NTPDase1, -2, -3, and -8 and ecto-50-nucleotidase, respectively. Data are presented as the mean ( SD of three
experiments conducted in triplicate.
with an EC50 of g40 μM, whose maximal response corre-
sponded to only 15% of that of ATP. The 20-deoxy analogues
2 and 4 were both inactive at concentrations of e50 μM.
Analogues 1-4 were completely inactive toward the P2Y2
receptor at concentrations of e25 μM.
inhibitors could serve as promising therapeutic agents for the
treatment of osteoarthritis39 or cancer.24
Nucleotide scaffolds suffer from inherent limitations as
therapeutic agents as they interact with numerous proteins1
and are metabolically unstable.40 Therefore, here, we selected a
dinucleoside polyphosphate scaffold for the development of
NPP inhibitors, as this scaffold offers better stability and
selectivity than nucleotides.1 To prevent any activity toward
the P2Y1 receptor, we conserved the adenine ring without a
methylthio substitution at the C-2 position, as the latter
enhances potency toward the P2Y1 receptor.41 Because
NPP1 hydrolyzes the PR-Pβ or Pδ-Pε phosphodiester bond
in ApnAs, as observed in our previous study with analogue 5
(Figure 1),1 we substituted the oxygen atoms bridging the
PR-Pβ and Pδ-Pε bonds with methylene groups to obtain a
hydrolysis-resistant scaffold, as in analogues 3 and 4. In
addition, because substitution of the nonbridging oxygen
atom with a BH3 group on the central phosphate of Ap3A
conferred resistance to hydrolysis by NPP1 and NPP3,1 we
have introduced a central boranophosphate moiety as in ana-
logue 1. Moreover, because 20-deoxyadenosine 50-(R-thio)tri-
phosphate was shown to be a potent inhibitor of NPPs,42 we
also synthesized 20-deoxy analogues such as 2.
Discussion
Design of NPP Inhibitors. NPP1-3 have a nucleotide
pyrophosphatase activity and metabolize NTP directly to
NMP and PPi but may also produce NDP and Pi.36 These
enzymes can be discriminated from other ectonucleotidases
by their ability to hydrolyze the diadenosine 50,500-polyphos-
phate analogues, ApnA, to AMP and adenosine nucleoside
50-(n - 1) phosphate.37 The regulation of the dinucleotide
levels by NPPs may be one of the dominant functions exerted
by these enzymes.36 Indeed, the physiological importance of
dinucleotides is becoming increasingly evident.38
The field of NPP enzymology is still in its infancy. There-
fore, NPP specific inhibitors that do not affect other ecto-
nucleotidases, such as NTPDases and 50-ectonucleotidase and
do not trigger or interfere with P2 receptor activation, would be
extremely valuable. Furthermore, potent and selective NPP

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 24 8491
Figure 4. Analogues 1-4 inhibit NPP activities. The activity of human NPP1 and NPP3 with either pnp-TMP (A), Ap5A (B), or ATP (C) as
the substrate is shown. NPP2 (membrane-bound form) activity with pnp-TMP as the substrate is presented in the inset of panel A. Substrates
and analogues 1-4 were studied at a concentration of 100 μM. In the control (ctrl), the substrate was tested in the absence of analogues and was
set to 100% of activity. The percentage of residual activity is presented at the top of each bar. Data are presented as the mean ( SD of three to six
experiments conducted in triplicate.
Effect of Analogues 1-4 on Ectonucleotidase Activities. All
derivatives (1-4) strongly inhibited the metabolism of both
synthetic (pnp-TMP) and natural substrates (Ap5A and ATP)
by NPP1 (Figure 4). Additionally, compounds 1and 4 inhibited

8492 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 24
Eliahu et al.
Table 2. IC50, Ki, and Ki0 Analysis of the Inhibition of NPP1 by
Analogues 1-4a
inhibitor
IC50 ( SD (μM)
Ki ( SD (μM)
Ki0 ( SD (μM)
1
2
3
4
63 ( 4.4
13 ( 0.7
33 ( 1.7
50 ( 2.0
20 ( 1.4
9 ( 0.5
13 ( 0.7
51 ( 2.5
31 ( 1.5
145 ( 6.9
147 ( 7.2
80 ( 3.8
a For the determination of Ki and Ki0, pnp-TMP substrate and
inhibitors 1-4 were used in the concentration range of 2.5 ꢀ 10-5 to
1 ꢀ 10-4 M. For the determination of IC50, the pnp-TMP concentration
was 5 ꢀ 10-5 M and the concentrations of inhibitors were in the range of
5 ꢀ 10-7 to 2 ꢀ 10-4 M. All experiments were performed three times in
triplicate.
Figure 6. Relative concentration-response plots for dinucleotides
1-4 via (A) the P2Y11 receptor and (B) the P2Y1 receptor. Data
were obtained from 1321N1 cells stably expressing the P2Y11GFP
receptor (A) or P2Y1GFP receptor (B), triggering the ligand-induced
change in [Ca2þ]i. Cells were preincubated with 2 μM fura-2 AM for
30 min, and the change in fluorescence (4F340/F380) was monitored.
Figure 5. Analogues 1-4 inhibit NPP activity at the surface of
HTB-85 and HT29 cells. Substrate, pnp-TMP, and analogues 1-4
were used at the concentration of 100 μM. In the control, the substrate
was tested in the absence of analogues and was set to 100% activity.
The percentage of residual activity is presented at the top of each
bar. Data are presented as the mean ( SD of three experiments
performed in triplicate.
is not important for a NPP1 inhibitor directed against Ap5A
hydrolysis, possibly because recognition of Ap5A does not
involve a 20-OH group.
NPP3-mediated hydrolysis of ATP is sensitive to inhibi-
tion by the dinucleotide analogues [analogues 1-4 (Figure 4)].
Apparently, the patterns of recognition of Ap5A and ATP by
NPP3 are different than those for NPP1. Therefore, NPP3 is
not affected by analogues 1-4 as much as NPP1.
Finally, NPP2 nucleotidase activity for both the membrane-
bound forms (Figure 4A) and the secreted forms (data not
shown) was highly affected by analogues 1-4. It is noteworthy
that in addition to its nucleotidase activity, NPP2 prefers lyso-
phospholipids as substrates. Because the hydrolysis of lysopho-
spholipids and nucleotides is performed by the same catalytic
site,12,43 one can speculate that analogues 1-4 might also inhibit
the hydrolysis of lysophospholipids by NPP2, and potentially
also by NPP4-7. This intriguing possibility will require
further investigation.
As for R,β-methylene-ADP, a known ecto-50-nucleotidase
inhibitor,44 the methylene groups between R,β and γ,δ or δ,
ε phosphates conferred strong inhibitory activity to com-
pounds 3 and 4 toward ecto-50-nucleotidase. In contrast,
compounds 1 and 2 had no effect on ecto-50-nucleotidase
activity, thus further emphasizing the specificity of the latter
analogues as inhibitors of NPPs.
the hydrolysis by NPP3 of pnp-TMP, Ap5A, and ATP by 30,
50, and >90%, respectively. The kinetic parameters (Table 2)
further indicate that analogues 1-4 are inhibitors of NPP1.
Interestingly, analogues 1-4 were not hydrolyzed by
NTPDases and also did not affect hydrolysis of ATP by
NTPDase1 and -8. Likewise, NTPDase2 and -3 activities
were reduced by e30% by these derivatives (Figure 3). In
addition, analogues 1 and 2 exhibited no inhibitory effect
toward ecto-50-nucleotidase.
From the data given above, the following molecular recogni-
tion requirements emerge for NPP1 and -3, for NTPDase1, -2,
-3, and -8, and for ecto-50-nucleotidase. Dinucleotides, hav-
ing either a tetra- or pentaphosphate linker, do not, or barely,
affect NTPDase activity (analogues 1-4). Indeed, such dinu-
cleotides are not recognized as substrates or inhibitors by these
enzymes. Furthermore, dinucleotides, having a pentaphosphate
linker, were not recognized by ecto-50-nucleotidase (analogues
1 and 2). Thus, an Ap5A scaffold is especially suitable for
designing NPP-selective inhibitors.
Analogues 1 and 2 having a pentaphosphate linker inhibited
Ap5A hydrolysis by NPP1 better than analogues 3and 4bearing
a tetraphosphate chain (Figure 4B), yet analogue 1 inhibited the
hydrolysis of ATP by NPP1 better than analogue 2 (Figure 4C),
implying that 1 competes with ATP because it has a 20-OH
group, unlike analogue 2. Namely, recognition of ATP by
NPP1 probably involves the 20-OH group. This requirement
Activity of Analogues 1-4 at the P2 Receptors. Adenine
nucleotide analogues with a methylene group substituting
for a bridging oxygen atom and with the replacement of a

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 24 8493
Conclusion
Table 3. EC50 Values for [Ca2þ]i Elevation by Ligands 1-4, Mediated
by the P2Y1,2,11 Receptors
This study shows that analogues 1-4 are moderate but
effective inhibitors of NPP1 activity in either cell extracts
or intact cells. Analogues 1 and 4 strongly blocked the activity
of both NPP1 and -3, yet analogue 1, found here to be an
effective NPP1 inhibitor, exhibited a low activity toward the
P2Y1 and P2Y11 receptors. On the other hand, analogue
2 did not significantly block NPP3 activity, had no activity
on NTPDase1, -2, -3, and-8, aswellasecto-50-nucleotidase, and
virtually no activity toward the P2Y1, P2Y2, and P2Y11
receptors, and is therefore the most specific inhibitor
of NPP1. Analogue 2 represents a future tool for explor-
ing therapeutic applications such as the development of
drugs for the management of cancer, metastasis, and osteoar-
thritis.
reduction in
affinity vs
ATP
receptor
subtype
nucleotide
analogue
EC50 (μM),
[Ca2þ]i elevation
P2Y1
4
30
g30
200
g200
60
2
3
9
0.3
1
2
ATPa
1-4
0.15
not active up to 25
1
P2Y2
-
μM
not active up to 50
P2Y11
4
2
-
-
μM
not active up to 50
μM
3
g40 (∼15% of ATP)
-
4
1
13
3.3
ATPa
1
Experimental Section
a ATP was selected as the common reference agonist at both P2Y1 and
P2Y11 receptors, although ADP is the preferred endogenous P2Y1
receptor agonist.
General. All commercial reagents were used without further
purification, unless otherwise noted. All air- and moisture-
sensitive reactions were conducted in flame-dried, argon-flushed,
two-neck flasks sealed with rubber septa, and the reagents were
introduced with a syringe. Progress of reactions was monitored by
TLC using precoated Merck silica gel plates (60F-253). Reac-
tants and products were visualized using UV light (Isco, UA-5).
Compounds were characterized by NMR using a Bruker AC-
200, DPX-300, or DMX-600 spectrometer. 1H NMR spectra were
recorded at 200, 300, or 600 MHz. Nucleotides were characterized
also by 31P NMR in D2O, using 85% H3PO4 as an external
reference on Bruker AC-200 and DMX-600 spectrometers.
High-resolution mass spectra were recorded on an AutoSpec-E
FISION VG mass spectrometer by chemical ionization. Nucleo-
tides were analyzed using electron spray ionization (ESI) on a
Q-TOF microinstrument (Waters). Primary purification of the
nucleotides was achieved on a LC (Isco UA-6) system using a
column of Sephadex DEAE-A25, swollen in 1 M NaHCO3 at
3 ꢀC for 24 h. The resin was washed with deionized water before
use. LC separation was monitored by UV detection at 280 nm.
Final purification of the dinucleotides was achieved on a
HPLC (Merck-Hitachi) system using a semipreparative reverse-
phase column [Gemini 5u C-18 110A, 250 mm ꢀ 10 mm, 5 μm
(Phenomenex, Torrance, CA)]. The purity of the dinucleotides
was evaluated on an analytical reverse-phase HPLC column system
[Gemini 5u C-18 110A, 150 mm ꢀ 3.60 mm, 5 μm (Phenomenex)]
in two-solvent systems with either solvent systems I and II or
solvent system III. Solvent system I consisted of (A) 100 mM
triethylammonium acetate (TEAA) (pH 7) and (B) MeOH.
Solvent system II consisted of (A) 100 mM TEAA (pH 7) and
(B) CH3CN. Solvent system III consisted of (A) 0.01 M KH2PO3
(pH 3.5) and (B) CH3CN. The details of the solvent system
gradients used for the separation of each product are provided
below. The products, obtained as triethylammonium salts, were
generally g95% pure. All reactants for moisture-sensitive reac-
tions were dried overnight in a vacuum oven. 20,30-O-Methoxy-
methylidene adenosine, 6,50 R,β-methylene-ADP, 9, and
20-deoxy-R,β-methyleme-ADP, 13,31 were prepared, as previously
described. R,β-Methylene-ADP and 20-deoxy-R,β-methylene-ADP
were separated using a MPLC system [Biotage (Kungsgatan,
Uppsala, Sweden)] using an RP-C18 (12þM) column and the
following gradient scheme: 3 column volumes of a 100:0 A/B
mixture (A being 100 mM TEAA and B being MeOH), 9 column
volumes of an A/B gradient from 100:0 to 60:30, and then
5 column volumes of a 60:30 A/B mixture, at a flow rate of
12 mL/min.
nonbridging oxygen in PR with a BH3 group have proven to
be weak agonists of P2Y1,4,6 receptors.25 Likewise, replacing
the PR-Pβ bond bridging oxygen in the potent P2Y1 receptor
agonist (2-MeS-ADP) with a dihalomethylene group (e.g., CCl2
or CF2) resulted in reductions in potency of 390- and 1200-
fold.45 In the present work, therefore, we replaced both
PR-Pβ and Pδ-Pε bond bridging oxygen atoms with methyl-
ene groups to prevent activity of the dinucleotide analogues
toward the P2Y receptors.
Analogue 5 has been previously shown to be a highly
potent P2Y1 receptor agonist (EC50 = 63 nM, vs 100 nM for
2-MeS-ADP).1 Upon replacing both PR-Pβ and Pδ-Pε bond
bridging oxygen atoms in 5 with methylene groups, yielding
analogue 1, we observed decreased activity toward the P2Y1
receptor. The boranophosphate modification in 1, which in
the case of ApnA resulted in significant activity via the P2Y1
receptor,1 increased the potency of 1 toward the P2Y1
receptor, as compared to 3. This result might indicate an
improved binding to the P2Y1 receptor because of the presence
of the relatively lipophilic borane moiety.46 Analogue 3
was ∼60-fold less potent than ATP, while Ap4A itself had
a potency similar to that of ATP,47 indicating that replacing
the bridging oxygen atom with a methylene group reduces
P2Y1 receptor agonist potency. Analogue 2 was >200-fold
less potent than ATP, indicating the importance of the 20-hydroxyl
group for molecular recognition by the P2Y1 receptor.
Among natural diadenosine polyphosphates, only Ap4A
may be considered as an agonist of the P2Y11 receptor, which
is normally activated by ATP derivatives.48-49 In this study,
Ap4A derivatives 3 and 4 were poor P2Y11 receptor agonists
or completely inactive, probably resulting from the replace-
ment of the bridging oxygen with a methylene group in the
polyphosphate chain, as observed for the P2Y1 receptor. The
most potent P2Y11 receptor agonist among analogues 1-4 was
compound 1 (Figure 6A and Table 3). Although a borano-
phosphate modification increased the potency of analogue 1
as a P2Y11 receptor agonist versus the other dinucleotide
derivatives tested here, it had a lower potency than ATP
(EC50 = 13 μM, vs 3.3 μM for ATP). The 20-deoxy-related
Ap5A scaffold, 2, was inactive up to 50 μM, thus indicating
the essential role of the 20-hydroxyl group for activity at the
P2Y11 receptor, as noted above for the P2Y1 receptor.
Typical Procedure for the Preparation of Diadenosine 50,500-
(Boranated)polyphosphonate Derivatives (1-4). R,β-Methylene-
ADP(Bu3NH)þ2 and 20-deoxy-R,β-methylene-ADP(Bu3NH)þ
were prepared by applying the corresponding ADP analogues
2

8494 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 24
Eliahu et al.
through a column of activated Dowex 50WX-8 (200 mesh, Hþ
form). The eluate was collected in an ice-cooled flask contain-
ing tributylamine (2 equiv) and EtOH. The resulting solution
was freeze-dried to yield R,β-methylene-ADP(Bu3NH)þ2, 9, and
20-deoxy-R,β-methylene ADP(Bu3NH)þ2, 13, as a viscous oil. Bis-
(tributylammonium) R,β-methylene-ADP salt (100 mg, 0.16 mmol)
was dissolved in dry DMF (2 mL), and CDI (180 mg, 1.11 mmol,
5 equiv) was added. The resulting solution was stirred at room
temperature (RT) for 12 h. BPi(Bu3NHþ)2, 16 (130 mg, 0.27 mmol,
1.7 equiv), in dry DMF (1.5 mL), and MgCl2 (120 mg, 1.28 mmol,
8 equiv) were added. The resulting solution was stirred at RT for
23 h. The semisolid obtained after evaporation of the solvent
was chromatographed at RT on a Sephadex DEAE-A25 column,
which was swelled in 1 M NaHCO3 prior to column preparation.
The separation was monitored by UV detection (λ = 280 nm). A
buffer gradient of water (1 L) to 0.7 M NH4HCO3 (1 L) was
applied. The relevant fractions were pooled and freeze-dried to
yield a white solid. Final purification was achieved on a semi-
preparative C18 HPLC column. Product 1 was obtained in 10%
yield (16 mg). Product 3 was obtained in 20% yield (29 mg).51
The spectral data for 3 are consistent with the literature.51
Product 2 was obtained in 21% yield (15 mg). Product 4 was
obtained in 28% yield (20 mg) after LC separation.
(2:8:11 NH4OH/H2O/2-propanol) Rf = 0.22. The following
purity data were obtained on an analytical column: tR = 3.39 min
(100% purity) using solvent system I with an A/B gradient from
80:20 to 60:30 over 10 min at a flow rate of 1 mL/min; tR = 1.51 min
(98.5% purity) using solvent system III with an A/B gradient from
85:15 to 50:50 over 10 min at a flow rate of 1 mL/min.
Plasmids. Except for human NPP2, the plasmids used in this
study have all been described in published reports: human
NTPDase1 (GenBank accession no. U87967),52 human NTPDase2
(GenBank accession no. NM_203368),53 human NTPDase3
(GenBank accession no. AF033830),54 human NTPDase8
(GenBank accession no. AY330313),55 human NPP1 (GenBank
accession no. NM_006208),11 and human NPP3 (GenBank acces-
sion no. NM_005021).56 Human NPP2 (XXU13871, unpublished,
submitted to NCBI by J. A. Malone) was subcloned here in
expression vector pcDNA3.1.
Cell Culture and Transfection. HTB-85 and HT29 cell lines
were grown in 10 cm plates and then transferred into 1 cm plates
for activity assays. Cells were incubated in 1 cm plates at 37 ꢀC in
R-MEM medium or in Dulbecco’s modified Eagle’s medium/F-12
nutrient mixture (DMEM/F-12) in the presence of 10% fetal
bovine serum (FBS). After reaching full confluence, cells were
used in intact cell assays (see below).
Ectonucleotidases were produced by transient transfection of
COS-7 cells in 10 cm plates using Lipofectamine (Invitrogen,
Burlington, ON), as previously described.6 Briefly, 80-90%
confluent cells were incubated for 5 h at 37 ꢀC in DMEM/F-12 in
the absence of fetal bovine serum (FBS) with 6 μg of plasmid
DNA and 23 μL of Lipofectamine reagent. The reaction was
stopped by the addition of an equal volume of DMEM/F-12
containing 20% FBS, and the cells were harvested 33-72 h later.
The conditioned medium of NPP2-transfected cells was also
collected.
GFP (green fluorescent protein) constructs of the human
P2Y1 receptor and P2Y11 receptor were expressed in 1321N1
astrocytoma cells, which lack endogenous expression of both
P2X and P2Y receptors. The respective cDNA of the given
receptor gene was cloned into the pEGFPN1 vector, and after
transfection using the FuGENE 6 Transfection Reagent (Roche
Molecular Biochemicals, Mannheim, Germany), cells were
selected with 0.5 mg/mL G318 (Merck Chemicals, Darmstadt,
Germany) and grown in DMEM supplemented with 10% fetal
calf serum, 100 units/mL penicillin, and 100 units/mL strepto-
mycin at 37 ꢀC and 5% CO2. The functional expression of the
receptor was confirmed by GFP fluorescence and the change in
intracellular Ca2þ concentration ([Ca2þ]i) upon incubation with
the respective standard receptor agonists.
Preparation of Protein Fractions. For the preparation of protein
extracts enriched with membrane proteins, transfected cells were
washed three times with Tris-saline buffer at 4 ꢀC, collected by being
scraped in harvesting buffer [95 mM NaCl, 0.1 mM phenylmetha-
nesulfonyl fluoride (PMSF), and 35 mM Tris (pH 7.5)], and washed
twice by 300g centrifugation for 10 min at 3 ꢀC. Cells were
resuspended in the harvesting buffer supplemented with 10 mg/mL
aprotinin and sonicated. Nucleus and cellular debris were discarded
by centrifugation at 300g for 10 min at 3 ꢀC, and the supernatant
(crude protein extract) was aliquoted and stored at -80 ꢀC until it
was used for activity assays. The secreted form of NPP2 was
prepared from the conditioned media of transfected cells that were
frozen and stored at -80 ꢀC until they were tested for activity.
Protein concentrations were estimated by the Bradford microplate
assay using bovine serum albumin (BSA) as the standard.57
Enzymatic Activity Assays. (i) NTPDases and Ecto-50-nucleo-
tidase. Activity was measured as described previously3 in 0.2 mL
of incubation medium Tris-Ringer buffer [120 mM NaCl, 5 mM
KCl, 2.5 mM CaCl2, 1.2 mM MgSO3, 25 mM NaHCO3, 5 mM
glucose, and 80 mM Tris (pH 7.3)] at 37 ꢀC with or with-
out analogues 1-4 (final concentration of 100 μM), and
with or without 100 μM ATP (for NTPDases) or 100 μM
AMP (for ecto-50-nucleotidase) as a substrate. The analogues
Purification and Characterization of Diadenosine 50,500-P1,P5,r,-
β-methylene-δ,ε-methylene-pentaphosphate-γ-borano (1). Product 1
was purified by HPLC on a semipreparative reverse-phase column,
using solvent system I, with an A/B gradient from 95:5 to 75:25 over
1
15 min at a flow rate of 3 mL/min: tR = 12.88 min; H NMR
(D2O, 300 MHz) δ 8.31 (s, H-8, 2H), 8.08 (s, H-2, 2H), 5.99 (d, J =
6.00 Hz, H-10, 2H), (H20 signals hidden by the water signal at
3.78 ppm), 3.53 (m, H-30, 2H), 3.29 (m, H-30, 2H), 3.13 (m, H-50,
3H), 2.31 (t, J = 20.71 Hz, CH2, 3H), 0.50 (m, BH3, 3H); 31P NMR
(D2O, 81 MHz) δ 78.38 (m,Pγ-BH3, 1P), 20.55 (d, J = 7.93 Hz,
PR, 2P), 10.77 (m, Pβ, 2P); HRMS MALDI (negative) m/z
C22H27D5N10O20P5 calcd 916.0796, found 916.0791 [MD5]; TLC
(2:8:11 NH4OH/H2O/2-propanol) Rf = 0.17. The following purity
data were obtained on an analytical column: tR = 1.81 min (99.97%
purity) using solvent system II with an A/B gradient from 80:20 to
70:30 over 10 min at a flow rate of 1 mL/min; tR = 8.00 min (98.31%
purity) using solvent system III with an A/B gradient from 100:0 to
70:30 over 10 min at a flow rate of 1 mL/min.
Purification and Characterization of Di-20-deoxyadenosine 50,500-
P1,P5,r,β-methylene-δ,ε-methylene-pentaphosphate-γ-borano
(2). Product 2 was purified by HPLC, on a semipreparative
reverse-phase column, using solvent system I, with an A/B
gradient from 95:5 to 70:30 over 20 min at a flow rate of
5 mL/min: tR = 15.35 min; 1H NMR (D2O, 300 MHz) δ 8.33
(s, H-8, 1H), 8.33 (s, H-8, 1H), 8.06 (s, H-2, 1H), 8.05 (s, H-2, 1H),
6.32 (t, J = 6.90 Hz, H-10, 2H) (H20 and H30 signals hidden by
the water signal at 3.78 ppm), 3.19 (m, H-30, 2H), 3.07 (m, H-50, 3H),
2.38 (t, J = 20.70 Hz, CH2, 3H), 0.50 (m, BH3, 3H); 31P NMR
(D2O, 81 MHz) δ 76.00 (m, Pγ-BH3, 1P), 17.82 (s, PR, 2P), 9.13
(d, J = 32.08 Hz, Pβ, 2P); MS ESI m/z 899 (M-Naþ); TLC
(2:8:11 NH4OH/H2O/2-propanol) Rf = 0.12. The following
purity data were obtained on an analytical column: tR = 2.18 min
(93.1% purity) using solvent system I with an A/B gradient from
70:30 to 50:50 over 10 min at a flow rate of 1 mL/min; tR = 1.37 min
(99.5% purity) using solvent system III with an A/B gradient
from 85:15 to 50:50 over 10 min at a flow rate of 1 mL/min.
Purification and Characterization of Di-20-deoxyadenosine 50,500-
P1,P5,r,β-methylene-γ,δ-methylene-tetraphosphate (4). Product
4 was purified by HPLC, with a semipreparative reverse-phase
column, using solvent system I, with an A/B gradient from 95:5
to 75:25 over 20 min at a flow rate of 5 mL/min: tR = 16.26 min;
1H NMR (D2O, 300 MHz) δ 8.33 (s, H-8, 2H), 8.06 (s, H-2, 2H),
6.32 (t, J = 6.30 Hz, H-10, 2H) (H20 and H30 signals hidden by
the water signal at 3.78 ppm), 3.18 (m, H-30, 2H), 3.06 (m, H-50, 3H),
2.31 (t, J = 2.31 Hz, CH2, 3H); 31P NMR (D2O 81 MHz) δ 18.02
(s, PR, 2P), 8.15 (s, Pβ, 2P); HRMS MALDI (negative) m/z
C22H31N10O15P3 calcd 799.0920, found 799.0915 [M3-]; TLC

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 24 8495
were added without ATP or AMP when tested as a potential
substrate, and with ATP or AMP when tested for their effect on
nucleotide hydrolysis. Either NTPDase or ecto-50-nucleotidase
protein extracts were added to the incubation mixture and pre-
incubated at 37 ꢀC for 3 min. The reaction was initiated by the
addition of substrate (ATP, AMP, or 1-4) and stopped after 15 min
with 50 μL of malachite green reagent. The amount of released
inorganic phosphate (Pi) was measured at 630 nm using the
method of Baykov et al.58
(SPSS Inc., Chicago, IL) using the ratio of the fluorescence
intensities with 340 and 380 nm excitation wavelengths.59,63
Acknowledgment. This work was supported by grants from
the Canadian Institutes of Health Research (CIHR) to J.S., who
was also a recipient of a new investigator award from the CIHR
and of a Junior 2 Scholarship from the FRSQ, and a grant from
Deutsche Forschungsgemeinschaft (DFG) to G.R. (Grant
Re 563/15). We thank Dr. Richard Poulin (Scientific Proof-
reading and Writing Service, CHUQ Research Center) for
editing this paper.
(ii) NPPs. Evaluation of the effect of analogues 1-4 on human
NPP1-3 activity was conducted with either pnp-TMP, ATP, or
Ap5A as the substrate.11 The reactions were conducted at 37 ꢀC
in 0.2 mL of the following incubation mixture: 1 mM CaCl2,
130 mM NaCl, 5 mM KCl, and 50 mM Tris (pH 8.5) with or
without analogues 1-4 and/or substrates. Substrates and ana-
logues 1-4 were all used at a final concentration of 100 μM.
Recombinant human NPP1, -2, or -3 cell lysates, as well as
soluble proteins containing the secreted form of NPP2, were
added to the incubation mixture and preincubated at 37 ꢀC for
3 min. The reaction was initiated by addition of the substrate.
For pnp-TMP hydrolysis, the production of p-nitrophenol was
measured at 405 nm, 15 min after the initiation of the reaction.
For Ap5A and ATP, we stopped the reaction after 30 min by
transferring an aliquot of 0.1 mL from the reaction mixture to
0.125 mL of ice-cold 1 M perchloric acid. The samples were
centrifuged for 5 min at 13000g. Supernatants were neutralized
with 1 M KOH (3 ꢀC) and centrifuged for 5 min at 13000g. An
aliquot of 20 μL was separated by reverse-phase HPLC to
evaluate the nucleotide content of each reaction sample
(see below). The type of inhibition, IC50 (50 μM substrate with
the inhibitors at concentrations of 0, 0.5, 5, 25, 50, 75, 100, and
200 μM), Ki, and Ki0 (substrate concentrations of 25, 50, 75, and
100 μM and inhibitor concentrations of 0, 25, 50, 75, and
100 μM) were calculated by plotting the data of three indepen-
dent experiments using pnp-TMP as the substrate according
to Dixon and Cornish-Bowden kinetics.
References
ꢀ
ꢀ
(1) Nahum, V.; Tulapurkar, M.; Levesque, S. A.; Sevigny, J.; Reiser,
G.; Fischer, B. Diadenosine and diuridine Poly(borano)phosphate
analogs: Synthesis, chemical and enzymatic stability, and activity
at P2Y1 and P2Y2 receptors. J. Med. Chem. 2006, 49, 1980–1990.
ꢀ
(2) Shirley, D. G.; Vekaria, R.; Sevigny, J. Ectonucleotidases in the
kidney. Purinergic Signalling 2009, 5, 501–511.
ꢀ
(3) Kukulski, F.; Levesque, S. A.; Lavoie, E. G.; Lecka, J.; Bigonnesse,
F.; Knowles, A. F.; Robson, S. C.; Kirley, T. L.; Sevigny, J.
ꢀ
Comparative hydrolysis of P2 receptor agonists by NTPDases 1,
2, 3, and 8. Purinergic Signalling 2005, 1, 193–204.
ꢀ
(4) Sevigny, J.; Levesque, F. P.; Grondin, G.; Beaudoin, A. R. Purification
of the blood vessel ATP diphosphohydrolase, identification and loca-
lization by immunological techniques. Biochim. Biophys. Acta 1997,
1334, 73–88.
(5) Heine, P.; Braun, N.; Heilbronn, A.; Zimmermann, H. Functional
characterization of rat ecto-ATPase and ecto-ATP diphosphohy-
drolase after heterologous expression in CHO cells. Eur. J. Bio-
chem. 1999, 262, 102–107.
(6) Colgan, S. P.; Eltzschig, H. K.; Eckle, T.; Thompson, L. F.
Physiological roles for ecto-50-nucleotidase (CD73). Purinergic
Signalling 2006, 2, 351–360.
(7) Resta, R.; Hooker, S. W.; Hansen, K. R.; Laurent, A. B.; Park,
J. L.; Blackburn, M. R.; Knudsen, T. B.; Thompson, L. F. Murine
ecto-50-nucleotidase (CD73): cDNA cloning and tissue distribu-
tion. Gene 1993, 133, 171–177.
(iii) Separation and Quantification of Nucleotides and Dinu-
cleotides by HPLC. An aliquot of 20 μL of the reaction products
(described above) was used for nucleotide analysis by HPLC
using a 15 cm ꢀ 3.6 mm, 3 μm SUPELCOSIL LC-18-T column
(Supelco, Bellefonte, PA). ATP, Ap5A, analogues 1-4, and
their hydrolysis products were separated with a mobile phase
consisting of 25 mM TBA, 5 mM EDTA, 100 mM KH2PO3/
K2HPO3 (pH 7.0), and 2% methanol (v/v), at a flow rate of
1 mL/min for 30 min. Separated nucleotides were detected by
UV absorption at 260 and 253 nm, identified, and quantified by
the comparison of the retention times with those of the appro-
priate standards.
(8) Deissler, H.; Genc, B.; Doerfler, W. Restriction endonuclease
BsoFI is sensitive to the 50-methylation of deoxycytidines in its
recognition sequence. Nucleic Acids Res. 1995, 23, 4227–4228.
(9) Bollen, M.; Gijsbers, R.; Ceulemans, H.; Stalmans, W.; Stefan, C.
Nucleotide pyrophosphatases/phosphodiesterases on the move.
Crit. Rev. Biochem. Mol. Biol. 2000, 35, 393–432.
(10) Cimpean, A.; Stefan, C.; Gijsbers, R.; Stalmans, W.; Bollen, M.
Substrate-specifying determinants of the nucleotide pyrophospha-
tases/phosphodiesterases NPP1 and NPP2. Biochem. J. 2004, 381,
71–77.
(11) Belli, S. I.; Goding, J. W. Biochemical characterization of human
PC-1, an enzyme possessing alkaline phosphodiesterase I and
nucleotide pyrophosphatase activities. Eur. J. Biochem. 1994,
226, 433–443.
(iv) Activity Assays with Intact HTB-85 and HT29 Cell Lines.
For intact cells, activity assays at the cell surface were conducted
in 0.25 mL of the incubation mixture containing 135 mM NaCl
in 24-well plates. Reaction was initiated by the addition of pnp-
TMP, yielding a final concentration of 100 μM. After 20 min,
0.2 mL of the reaction mixture was transferred to a 96-well
plate and the production of p-nitrophenol was measured at
310 nm.
Calcium Measurements. 1321N1 astrocytoma cells were trans-
fected with the respective plasmid for P2YR-GFP expression,
namely pEGFPN1 expression vector plasmids encoding the cDNA
for human P2Y1 or P2Y11 receptors59 and the P2Y2 receptor.60,61
Cells plated on coverslips (22 mm in diameter) and grown to approx-
imately 80% density were incubated with 2 μM fura-2 AM and
0.02% pluronic acid in HEPES-buffered saline [135 mM NaCl,
5.3 mM KCl, 1.8 mM CaCl2, 1 mM MgCl2, 25 mM glucose,
and 20 mM HEPES/Tris (pH 7.3)] for 30 min at 37 ꢀC. Cells
were superfused (1 mL/min, 37 ꢀC) with different concentrations
of dinucleotides 1-4, and the change in [Ca2þ]i was monitored
from the respective emission intensity at 510 nm after alter-
nating between 340 and 380 nm as the excitation wavelengths.62
Concentration-response data were analyzed with SigmaPlot
(12) Gijsbers, R.; Aoki, J.; Arai, H.; Bollen, M. The hydrolysis of
lysophospholipids and nucleotides by autotaxin (NPP2) involves
a single catalytic site. FEBS Lett. 2003, 538, 60–64.
(13) Sakagami, H.; Aoki, J.; Natori, Y.; Nishikawa, K.; Kakehi, Y.;
Natori, Y.; Arai, H. Biochemical and molecular characterization of
a novel choline-specific glycerophosphodiester phosphodiesterase
belonging to the nucleotide pyrophosphatase/phosphodiesterase
family. J. Biol. Chem. 2005, 280, 23084–23093.
(14) Okawa, A.; Nakamura, I.; Goto, S.; Moriya, H.; Nakamura, Y.;
Ikegawa, S. Mutation in Npps in a mouse model of ossification of
the posterior longitudinal ligament of the spine. Nat. Genet. 1998,
19, 271–273.
(15) Johnson, K.; Terkeltaub, R. Inorganic pyrophosphate (PPi) in
pathologic calcification of articular cartilage. Front. Biosci. 2005,
10, 988–997.
(16) Vaingankar, S. M.; Fitzpatrick, T. A.; Johnson, K.; Goding, J. W.;
Maurice, M.; Terkeltaub, R. Subcellular targeting and function
of osteoblast nucleotide pyrophosphatase phosphodiesterase 1.
Am. J. Physiol. 2004, 286, C1177–C1187.
(17) Damek-Poprawa, M.; Golub, E.; Otis, L.; Harrison, G.; Phillips,
C.; Boesze-Battaglia, K. Chondrocytes utilize a cholesterol-depen-
dent lipid translocator to externalize phosphatidylserine. Biochem-
istry 2006, 45, 3325–3336.
(18) Maddux, B. A.; Goldfine, I. D. Membrane glycoprotein PC-1
inhibition of insulin receptor function occurs via direct interaction
with the receptor R-subunit. Diabetes 2000, 49, 13–19.

8496 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 24
Eliahu et al.
(19) Fuerstenau, C. R.; Trentin, D. D. S.; Barreto-Chaves, M. L. M.;
Sarkis, J. J. F. Ecto-nucleotide pyrophosphatase/phosphodiester-
ase as part of a multiple system for nucleotide hydrolysis by
platelets from rats: Kinetic characterization and biochemical prop-
erties. Platelets 2006, 17, 84–91.
(20) Grobben, B.; Anciaux, K.; Roymans, D.; Stefan, C.; Bollen, M.;
Esmans, E. L.; Slegers, H. An ecto-nucleotide pyrophosphatase is
one of the main enzymes involved in the extracellular metabolism
of ATP in rat C6 glioma. J. Neurochem. 1999, 72, 826–834.
(21) Yano, Y.; Hayashi, Y.; Sano, K.; Shinmaru, H.; Kuroda, Y.;
Yokozaki, H.; Yoon, S.; Kasuga, M. Expression and localization
of ecto-nucleotide pyrophosphatase/phosphodiesterase I-3 (E-NPP3/
CD203c/PD-Iβ/B10/gp130RB13-6) in human colon carcinoma. Int.
J. Mol. Med. 2003, 12, 763–766.
endogenous agonists of the purinergic system. Br. J. Pharmacol.
2009, 157, 1142–1153.
(39) Tenenbaum, J.; Muniz, O.; Schumacher, H. R.; Good, A. E.;
Howell, D. S. Comparison of phosphohydrolase activities from
articular cartilage in calcium pyrophosphate deposition disease
and primary osteoarthritis. Arthritis Rheum. 1981, 24, 492–500.
(40) Sellers, L. A.; Simon, J.; Lundahl, T. S.; Cousens, D. J.;
Humphrey, P. P. A.; Barnard, E. A. Adenosine nucleotides
acting at the human P2Y1 receptor stimulate mitogen-activated
protein kinases and induce apoptosis. J. Biol. Chem. 2001, 276,
16379–16390.
ꢀ
(41) Eliahu, S. E.; Camden, J.; Lecka, J.; Weisman, G. A.; Sevigny, J.;
ꢀ
Gelinas, S.; Fischer, B. Identification of hydrolytically stable and
selective P2Y1 receptor agonists. Eur. J. Med. Chem. 2009, 44,
1525–1536.
(22) Savaskan, N. E.; Rocha, L.; Kotter, M. R.; Baer, A.; Lubec, G.;
van Meeteren, L. A.; Kishi, Y.; Aoki, J.; Moolenaar, W. H.; Nitsch,
R.; Braeuer, A. U. Autotaxin (NPP-2) in the brain: Cell type-
specific expression and regulation during development and after
neurotrauma. Cell. Mol. Life Sci. 2007, 64, 230–243.
(42) Wojcik, M.; Cieslak, M.; Stec, W. J.; Goding, J. W.; Koziolkiewicz,
M. Nucleotide pyrophosphatase/phosphodiesterase 1 is responsi-
ble for degradation of antisense phosphorothioate oligonucleo-
tides. Oligonucleotides 2007, 17, 134–145.
(43) Koh, E.; Clair, T.; Woodhouse, E. C.; Schiffmann, E.; Liotta,
L.; Stracke, M. Site-directed mutations in the tumor-associated
cytokine, autotaxin, eliminate nucleotide phosphodiesterase,
lysophospholipase D, and motogenic activities. Cancer Res.
2003, 63, 2042–2045.
(23) Maldonado, P. A.; Correa, M. d. C.; Vargas Becker, L.; Flores, C.;
Moretto, M. B.; Morsch, V.; Schetinger, M. R. C. Ectonucleotide
pyrophosphatase/phosphodiesterase (E-NPP) and adenosine deam-
inase (ADA) activities in patients with uterine cervix neoplasia. Clin.
Biochem. 2008, 41, 400–406.
(44) Bar, H. P.; Simonson, L. P. Inhibition of extracellular and purified
50-nucleotidase from rat heart. Recent Adv. Stud. Card. Struct.
Metab. 1975, 10, 583–590.
(24) Stefan, C.; Jansen, S.; Bollen, M. NPP-type ectophosphodies-
terases: Unity in diversity. Trends Biochem. Sci. 2005, 30, 542–550.
(25) Ruecker, B.; Almeida, M. E.; Libermann, T. A.; Zerbini, L. F.;
Wink, M. R.; Sarkis, J. J. F. Biochemical characterization of ecto-
nucleotide pyrophosphatase/phosphodiesterase (E-NPP, E.C.
3.1.4.1) from rat heart left ventricle. Mol. Cell. Biochem. 2007,
306, 247–254.
(45) Eliahu, S.; Martin-Gil, A.; Perez de Lara, M. J.; Pintor, J.; Camden,
ꢀ
J.; Weisman, G. A.; Lecka, J.; Sevigny, J.; Fischer, B. 2-MeS-β,
γ-CCl2-ATP is a potent agent for reducing intraocular pressure.
J. Med. Chem. 2010, 53, 3305–3319.
ꢀ
(26) Munkonda, M. N.; Kauffenstein, G.; Kukulski, F.; Levesque,
S. A.; Legendre, C.; Pelletier, J.; Lavoie, E. G.; Lecka, J.; Sevigny,
(46) Shaw, B. R.; Sergueev, D.; He, K.; Porter, K.; Summers, J.;
Sergueeva, Z.; Rait, V. Boranophosphate backbone: A mimic of
phosphodiesters, phosphorothioates, and methylphosphonates.
Methods Enzymol. 2000, 313, 226–257.
(47) Shaver, S. R.; Rideout, J. L.; Pendergast, W.; Douglass, J. G.;
Brown, E. G.; Boyer, J. L.; Patel, R. I.; Redick, C. C.; Jones, A. C.;
Picher, M.; Yerxa, B. R. Structure-activity relationships of dinu-
cleotides: Potent and selective agonists of P2Y receptors. Puriner-
gic Signalling 2005, 1, 183–191.
(48) Communi, D.; Janssens, R.; Robaye, B.; Zeelis, N.; Boeynaems,
J.-M. Role of P2Y11 receptors in hematopoiesis. Drug Dev. Res.
2001, 52, 156–163.
(49) Patel, K.; Barnes, A.; Camacho, J.; Paterson, C.; Boughtflower, R.;
Cousens, D.; Marshall, F. Activity of diadenosine polyphosphates
at P2Y receptors stably expressed in 1321N1 cells. Eur. J. Pharma-
col. 2001, 430, 203–210.
ꢀ
J. Inhibition of human and mouse plasma membrane bound
NTPDases by P2 receptor antagonists. Biochem. Pharmacol.
2007, 74, 1524–1534.
(27) Khan, K. M.; Fatima, N.; Rasheed, M.; Jalil, S.; Ambreen, N.;
Perveen, S.; Choudhary, M. I. 1,3,4-Oxadiazole-2(3H)-thione and
its analogues: A new class of non-competitive nucleotide pyrophos-
phatases/phosphodiesterases 1 inhibitors. Bioorg. Med. Chem.
2009, 17, 7816–7822.
(28) Choudhary, M. I.; Fatima, N.; Khan, K. M.; Jalil, S.; Iqbal, S.;
Atta-ur-Rahman New biscoumarin derivatives: Cytotoxicity and
enzyme inhibitory activities. Bioorg. Med. Chem. 2006, 14, 8066–
8072.
(29) Zatorski, A.; Goldstein, B. M.; Colby, T. D.; Jones, J. P.; Pankie-
wicz, K. W. Potent inhibitors of human inosine monophosphate
dehydrogenase type II. Fluorine-substituted analogues of thiazole-
4-carboxamide adenine dinucleotide. J. Med. Chem. 1995, 38,
1098–1105.
(30) Nahum, V.; Fischer, B. Boranophosphate salts as an excellent
mimic of phosphate salts: Preparation, characterization, and prop-
erties. Eur. J. Inorg. Chem. 2004, 4124–4131.
(50) Nahum, V.; Zuendorf, G.; Levesque, S. A.; Beaudoin, A. R.;
Reiser, G.; Fischer, B. Adenosine 50-O-(1-boranotriphosphate)
derivatives as novel P2Y1 receptor agonists. J. Med. Chem. 2002,
45, 5384–5396.
(51) Pankiewicz, K. W.; Lesiak, K.; Watanabe, K. A. Efficient synthesis
of methylenebis(phosphonate) analogs of P1,P2-disubstituted pyro-
phosphates of biological interest. A novel plausible mechanism.
J. Am. Chem. Soc. 1997, 119, 3691–3695.
(31) Davisson, V. J.; Davis, D. R.; Dixit, V. M.; Poulter, C. D. Synthesis
of nucleotide 50-diphosphates from 50-O-tosyl nucleosides. J. Org.
Chem. 1987, 52, 1794–1801.
ꢀ
(52) Kaczmarek, E.; Koziak, K.; Sevigny, J.; Siegel, J. B.; Anrather, J.;
(32) Liang, F.; Jain, N.; Hutchens, T.; Shock, D. D.; Beard, W. A.;
Wilson, S. H.; Chiarelli, M. P.; Cho, B. P. R,β-Methylene-
20-deoxynucleoside 50-Triphosphates as noncleavable substrates
for DNA polymerases: Isolation, characterization, and stability
studies of novel 20-deoxycyclonucleosides, 3,50-cyclo-dG, and
2,50-cyclo-dT. J. Med. Chem. 2008, 51, 6460–6470.
Beaudoin, A. R.; Bach, F. H.; Robson, S. C. Identification and
characterization of CD39/vascular ATP diphosphohydrolase.
J. Biol. Chem. 1996, 271, 33116–33122.
(53) Knowles, A. F.; Chiang, W.-C. Enzymatic and transcriptional
regulation of human ecto-ATPase/E-NTPDase 2. Arch. Biochem.
Biophys. 2003, 418, 217–227.
(33) Hoard, D. E.; Ott, D. G. Conversion of mono- and oligodeoxyr-
ibonucleotides to 50-triphosphates. J. Am. Chem. Soc. 1965, 87,
1785–1788.
(34) Levesque, S. A.; Lavoie, E. G.; Lecka, J.; Bigonnesse, F.; Sevigny,
J. Specificity of the ecto-ATPase inhibitor ARL 67156 on human
and mouse ectonucleotidases. Br. J. Pharmacol. 2007, 152, 141–
150.
(35) Baricault, L.; Denariaz, G.; Houri, J.-J.; Bouley, C.; Sapin, C.;
Trugnan, G. Use of HT-29, a cultured human colon cancer cell line,
to study the effect of fermented milks on colon cancer cell growth
and differentiation. Carcinogenesis 1995, 16, 245–252.
(36) Stefan, C.; Jansen, S.; Bollen, M. Modulation of purinergic signal-
ing by NPP-type ectophosphodiesterases. Purinergic Signalling
2006, 2, 361–370.
(37) Rotlian, P.; Asensio, A. C.; Ramos, A.; Rodriguez-Ferrer, C. R.;
Oaknin, S. Ectoenzymatic hydrolysis of the signalling nucleotides
diadenosine polyphosphates. Recent Res. Dev. Biochem. 2002, 3,
191–209.
(54) Smith, T. M.; Kirley, T. L. Cloning, sequencing, and expression
of a human brain ecto-apyrase related to both the ecto-ATPases
and CD39 ecto-apyrases. Biochim. Biophys. Acta 1998, 1386,
65–78.
ꢀ
ꢀ
ꢀ
(55) Fausther, M.; Lecka, J.; Kukulski, F.; Levesque, S. A.; Pelletier, J.;
ꢀ
Zimmermann, H.; Dranoff, J. A.; Sevigny, J. Cloning, purification,
and identification of theliver canalicular ecto-ATPase as NTPDase8.
Am. J. Physiol. 2007, 292, G785–G795.
(56) Jin-Hua, P.; Goding, J. W.; Nakamura, H.; Sano, K. Molecular
cloning and chromosomal localization of PD-Iβ (PDNP3), a new
member of the human phosphodiesterase I genes. Genomics 1997,
45, 412–415.
(57) Bradford, M. M. A rapid and sensitive method for the quantitation of
microgram quantities of protein utilizing the principle of protein-
dye binding. Anal. Biochem. 1976, 72, 248–254.
(58) Baikov, A. A.; Kasho, V. N.; Avaeva, S. M. Inorganic pyrophos-
phatase as a label in heterogeneous enzyme immunoassay.
Anal. Biochem. 1988, 171, 271–276.
(38) Jankowski, V.; van der Giet, M.; Mischak, H.; Morgan, M.;
Zidek, W.; Jankowski, J. Dinucleoside polyphosphates: Strong
(59) Ecke, D.; Tulapurkar, M. E.; Nahum, V.; Fischer, B.; Reiser, G.
Oppositediastereoselective activationofP2Y1 andP2Y11 nucleotide

Article
Journal of Medicinal Chemistry, 2010, Vol. 53, No. 24 8497
receptors by adenosine 50-O-(R-boranotriphosphate) analogues.
(62) Ubl, J. J.; Vohringer, C.; Reiser, G. Co-existence of two types of
[Ca2þ]i-inducing protease-activated receptors (PAR-1 and PAR-2)
in rat astrocytes and C6 glioma cells. Neuroscience 1998, 86,
597–609.
(63) Ecke, D.; Hanck, T.; Tulapurkar, M. E.; Schaefer, R.; Kassack,
M.; Stricker, R.; Reiser, G. Hetero-oligomerization of the P2Y11
receptor with the P2Y1 receptor controls the internalization and
ligand selectivity of the P2Y11 receptor. Biochem. J. 2008, 409,
107–116.
Br. J. Pharmacol. 2006, 149, 416–423.
(60) Ginsburg-Shmuel, T.; Haas, M.; Schumann, M.; Reiser, G.; Kalid,
O.; Stern, N.; Fischer, B. 5-OMe-UDP is a potent and selective
P2Y6-receptor agonist. J. Med. Chem. 2010, 53, 1673–1685.
(61) Tulapurkar, M. E.; Laubinger, W.; Nahum, V.; Fischer, B.; Reiser,
G. Subtype specific internalization of P2Y1 and P2Y2 receptors
induced by novel adenosine 50-O-(1-boranotriphosphate) deriva-
tives. Br. J. Pharmacol. 2004, 142, 869–878.